From 6e4abc40fc125b1dcc2792eacac17606a4d86043 Mon Sep 17 00:00:00 2001
From: James Courtier-Dutton <James@superbug.co.uk>
Date: Sat, 26 Mar 2005 19:35:29 +0100
Subject: [ALSA] Adds Capture to P16V chip.

EMU10K1/EMU10K2 driver
One can select which capture source, but one cannot yet set volumes.

Signed-off-by: James Courtier-Dutton <James@superbug.co.uk>
---
 include/sound/emu10k1.h | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/sound/emu10k1.h b/include/sound/emu10k1.h
index 43b6786abae..8221df88053 100644
--- a/include/sound/emu10k1.h
+++ b/include/sound/emu10k1.h
@@ -83,7 +83,8 @@
 #define IPR			0x08		/* Global interrupt pending register		*/
 						/* Clear pending interrupts by writing a 1 to	*/
 						/* the relevant bits and zero to the other bits	*/
-
+#define IPR_P16V		0x80000000	/* Bit set when the CA0151 P16V chip wishes
+						   to interrupt */
 #define IPR_GPIOMSG		0x20000000	/* GPIO message interrupt (RE'd, still not sure 
 						   which INTE bits enable it)			*/
 
@@ -1109,7 +1110,9 @@ struct _snd_emu10k1 {
 
 	emu10k1_voice_t voices[NUM_G];
 	emu10k1_voice_t p16v_voices[4];
+	emu10k1_voice_t p16v_capture_voice;
 	int p16v_device_offset;
+	u32 p16v_capture_source;
 	emu10k1_pcm_mixer_t pcm_mixer[32];
 	emu10k1_pcm_mixer_t efx_pcm_mixer[NUM_EFX_PLAYBACK];
 	snd_kcontrol_t *ctl_send_routing;
-- 
cgit v1.2.3-70-g09d2


From 2b637da5a1bb3c128ecdadea6aee693f6ff3b786 Mon Sep 17 00:00:00 2001
From: Lee Revell <rlrevell@joe-job.com>
Date: Wed, 30 Mar 2005 13:51:18 +0200
Subject: [ALSA] clean up card features

EMU10K1/EMU10K2 driver
This patch converts the emu10k1 driver to use the card capabilities
structure for some more things.

Not extensively tested but seems to work.

Signed-off-by: Lee Revell <rlrevell@joe-job.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 include/sound/emu10k1.h          |  9 +++----
 sound/pci/emu10k1/emu10k1_main.c | 29 +++++++++------------
 sound/pci/emu10k1/emufx.c        | 56 +++++++++++++++++++++-------------------
 sound/pci/emu10k1/emumixer.c     | 10 +++----
 sound/pci/emu10k1/emuproc.c      |  2 +-
 5 files changed, 51 insertions(+), 55 deletions(-)

(limited to 'include')

diff --git a/include/sound/emu10k1.h b/include/sound/emu10k1.h
index 8221df88053..b1e8ee8e0fa 100644
--- a/include/sound/emu10k1.h
+++ b/include/sound/emu10k1.h
@@ -1046,6 +1046,7 @@ typedef struct {
 	unsigned char ca0108_chip;  /* Audigy 2 Value */
 	unsigned char ca0151_chip;  /* P16V */
 	unsigned char spk71;        /* Has 7.1 speakers */
+	unsigned char sblive51;	    /* SBLive! 5.1 - extout 0x11 -> center, 0x12 -> lfe */
 	unsigned char spdif_bug;    /* Has Spdif phasing bug */
 	unsigned char ac97_chip;    /* Has an AC97 chip */
 	unsigned char ecard;        /* APS EEPROM */
@@ -1057,11 +1058,8 @@ struct _snd_emu10k1 {
 	int irq;
 
 	unsigned long port;			/* I/O port number */
-	unsigned int APS: 1,			/* APS flag */
-	    no_ac97: 1,				/* no AC'97 */
-	    tos_link: 1,			/* tos link detected */
-	    rear_ac97: 1,			/* rear channels are on AC'97 */
-	    spk71:1;				/* 7.1 configuration (Audigy 2 ZS) */
+	unsigned int tos_link: 1,		/* tos link detected */
+	    rear_ac97: 1;			/* rear channels are on AC'97 */
 	const emu_chip_details_t *card_capabilities;	/* Contains profile of card capabilities */
 	unsigned int audigy;			/* is Audigy? */
 	unsigned int revision;			/* chip revision */
@@ -1456,7 +1454,6 @@ int snd_emu10k1_fx8010_unregister_irq_handler(emu10k1_t *emu,
 #endif
 
 typedef struct {
-	unsigned int card;			/* card type */
 	unsigned int internal_tram_size;	/* in samples */
 	unsigned int external_tram_size;	/* in samples */
 	char fxbus_names[16][32];		/* names of FXBUSes */
diff --git a/sound/pci/emu10k1/emu10k1_main.c b/sound/pci/emu10k1/emu10k1_main.c
index ff220fc3142..38be0f1b0e7 100644
--- a/sound/pci/emu10k1/emu10k1_main.c
+++ b/sound/pci/emu10k1/emu10k1_main.c
@@ -170,7 +170,7 @@ static int __devinit snd_emu10k1_init(emu10k1_t * emu, int enable_ir)
 			SPCS_GENERATIONSTATUS | 0x00001200 |
 			0x00000000 | SPCS_EMPHASIS_NONE | SPCS_COPYRIGHT);
 
-	if (emu->audigy && emu->revision == 4) { /* audigy2 */
+	if (emu->card_capabilities->ca0151_chip) { /* audigy2 */
 		/* Hacks for Alice3 to work independent of haP16V driver */
 		u32 tmp;
 
@@ -600,7 +600,7 @@ static int snd_emu10k1_free(emu10k1_t *emu)
 	if (emu->port)
 		pci_release_regions(emu->pci);
 	pci_disable_device(emu->pci);
-	if (emu->audigy && emu->revision == 4) /* P16V */	
+	if (emu->card_capabilities->ca0151_chip) /* P16V */	
 		snd_p16v_free(emu);
 	kfree(emu);
 	return 0;
@@ -612,8 +612,6 @@ static int snd_emu10k1_dev_free(snd_device_t *device)
 	return snd_emu10k1_free(emu);
 }
 
-/* vendor, device, subsystem, emu10k1_chip, emu10k2_chip, ca0102_chip, ca0108_chip, ca0151_chip, spk71, spdif_bug, ac97_chip, ecard, driver, name */
-
 static emu_chip_details_t emu_chip_details[] = {
 	/* Audigy 2 Value AC3 out does not work yet. Need to find out how to turn off interpolators.*/
 	{.vendor = 0x1102, .device = 0x0008, .subsystem = 0x10011102,
@@ -688,15 +686,22 @@ static emu_chip_details_t emu_chip_details[] = {
 	{.vendor = 0x1102, .device = 0x0002, .subsystem = 0x80641102,
 	 .driver = "EMU10K1", .name = "SB Live 5.1", 
 	 .emu10k1_chip = 1,
-	 .ac97_chip = 1} ,
+	 .ac97_chip = 1,
+	 .sblive51 = 1} ,
 	{.vendor = 0x1102, .device = 0x0002, .subsystem = 0x80401102,
 	 .driver = "EMU10K1", .name = "SBLive! Platinum [CT4760P]", 
 	 .emu10k1_chip = 1,
 	 .ac97_chip = 1} ,
+	{.vendor = 0x1102, .device = 0x0002, .subsystem = 0x80271102,
+	 .driver = "EMU10K1", .name = "SBLive! Value [CT4832]", 
+	 .emu10k1_chip = 1,
+	 .ac97_chip = 1,
+	 .sblive51 = 1} ,
 	{.vendor = 0x1102, .device = 0x0002,
 	 .driver = "EMU10K1", .name = "SB Live [Unknown]", 
 	 .emu10k1_chip = 1,
-	 .ac97_chip = 1} ,
+	 .ac97_chip = 1,
+	 .sblive51 = 1} ,
 	{ } /* terminator */
 };
 
@@ -747,7 +752,6 @@ int __devinit snd_emu10k1_create(snd_card_t * card,
 	emu->revision = revision;
 	pci_read_config_dword(pci, PCI_SUBSYSTEM_VENDOR_ID, &emu->serial);
 	pci_read_config_word(pci, PCI_SUBSYSTEM_ID, &emu->model);
-	emu->card_type = EMU10K1_CARD_CREATIVE;
 	snd_printdd("vendor=0x%x, device=0x%x, subsystem_vendor_id=0x%x, subsystem_id=0x%x\n",pci->vendor, pci->device, emu->serial, emu->model);
 
 	for (c = emu_chip_details; c->vendor; c++) {
@@ -825,15 +829,6 @@ int __devinit snd_emu10k1_create(snd_card_t * card,
 
 	pci_set_master(pci);
 
-	if (c->ecard) {
-		emu->card_type = EMU10K1_CARD_EMUAPS;
-		emu->APS = 1;
-	}
-	if (! c->ac97_chip)
-		emu->no_ac97 = 1;
-	
-	emu->spk71 = c->spk71;
-	
 	emu->fx8010.fxbus_mask = 0x303f;
 	if (extin_mask == 0)
 		extin_mask = 0x3fcf;
@@ -842,7 +837,7 @@ int __devinit snd_emu10k1_create(snd_card_t * card,
 	emu->fx8010.extin_mask = extin_mask;
 	emu->fx8010.extout_mask = extout_mask;
 
-	if (emu->APS) {
+	if (emu->card_capabilities->ecard) {
 		if ((err = snd_emu10k1_ecard_init(emu)) < 0) {
 			snd_emu10k1_free(emu);
 			return err;
diff --git a/sound/pci/emu10k1/emufx.c b/sound/pci/emu10k1/emufx.c
index b9fa2e887fe..0529fb28112 100644
--- a/sound/pci/emu10k1/emufx.c
+++ b/sound/pci/emu10k1/emufx.c
@@ -1077,7 +1077,7 @@ static int __devinit _snd_emu10k1_audigy_init_efx(emu10k1_t *emu)
 	gpr += 2;
 	
 	/* PCM Side Playback (independent from stereo mix) */
-	if (emu->spk71) {
+	if (emu->card_capabilities->spk71) {
 		A_OP(icode, &ptr, iMAC0, A_GPR(playback+6), A_C_00000000, A_GPR(gpr), A_FXBUS(FXBUS_PCM_LEFT_SIDE));
 		A_OP(icode, &ptr, iMAC0, A_GPR(playback+7), A_C_00000000, A_GPR(gpr+1), A_FXBUS(FXBUS_PCM_RIGHT_SIDE));
 		snd_emu10k1_init_stereo_control(&controls[nctl++], "PCM Side Playback Volume", gpr, 100);
@@ -1145,14 +1145,14 @@ A_OP(icode, &ptr, iMAC0, A_GPR(var), A_GPR(var), A_GPR(vol), A_EXTIN(input))
 	A_ADD_VOLUME_IN(stereo_mix, gpr, A_EXTIN_SPDIF_CD_L);
 	A_ADD_VOLUME_IN(stereo_mix+1, gpr+1, A_EXTIN_SPDIF_CD_R);
 	snd_emu10k1_init_stereo_control(&controls[nctl++],
-					emu->no_ac97 ? "CD Playback Volume" : "Audigy CD Playback Volume",
+					emu->card_capabilities->ac97_chip ? "Audigy CD Playback Volume" : "CD Playback Volume",
 					gpr, 0);
 	gpr += 2;
 	/* Audigy CD Capture Volume */
 	A_ADD_VOLUME_IN(capture, gpr, A_EXTIN_SPDIF_CD_L);
 	A_ADD_VOLUME_IN(capture+1, gpr+1, A_EXTIN_SPDIF_CD_R);
 	snd_emu10k1_init_stereo_control(&controls[nctl++],
-					emu->no_ac97 ? "CD Capture Volume" : "Audigy CD Capture Volume",
+					emu->card_capabilities->ac97_chip ? "Audigy CD Capture Volume" : "CD Capture Volume",
 					gpr, 0);
 	gpr += 2;
 
@@ -1171,14 +1171,14 @@ A_OP(icode, &ptr, iMAC0, A_GPR(var), A_GPR(var), A_GPR(vol), A_EXTIN(input))
 	A_ADD_VOLUME_IN(stereo_mix, gpr, A_EXTIN_LINE2_L);
 	A_ADD_VOLUME_IN(stereo_mix+1, gpr+1, A_EXTIN_LINE2_R);
 	snd_emu10k1_init_stereo_control(&controls[nctl++],
-					emu->no_ac97 ? "Line Playback Volume" : "Line2 Playback Volume",
+					emu->card_capabilities->ac97_chip ? "Line2 Playback Volume" : "Line Playback Volume",
 					gpr, 0);
 	gpr += 2;
 	/* Line2 Capture Volume */
 	A_ADD_VOLUME_IN(capture, gpr, A_EXTIN_LINE2_L);
 	A_ADD_VOLUME_IN(capture+1, gpr+1, A_EXTIN_LINE2_R);
 	snd_emu10k1_init_stereo_control(&controls[nctl++],
-					emu->no_ac97 ? "Line Capture Volume" : "Line2 Capture Volume",
+					emu->card_capabilities->ac97_chip ? "Line2 Capture Volume" : "Line Capture Volume",
 					gpr, 0);
 	gpr += 2;
         
@@ -1197,14 +1197,14 @@ A_OP(icode, &ptr, iMAC0, A_GPR(var), A_GPR(var), A_GPR(vol), A_EXTIN(input))
 	A_ADD_VOLUME_IN(stereo_mix, gpr, A_EXTIN_AUX2_L);
 	A_ADD_VOLUME_IN(stereo_mix+1, gpr+1, A_EXTIN_AUX2_R);
 	snd_emu10k1_init_stereo_control(&controls[nctl++],
-					emu->no_ac97 ? "Aux Playback Volume" : "Aux2 Playback Volume",
+					emu->card_capabilities->ac97_chip ? "Aux2 Playback Volume" : "Aux Playback Volume",
 					gpr, 0);
 	gpr += 2;
 	/* Aux2 Capture Volume */
 	A_ADD_VOLUME_IN(capture, gpr, A_EXTIN_AUX2_L);
 	A_ADD_VOLUME_IN(capture+1, gpr+1, A_EXTIN_AUX2_R);
 	snd_emu10k1_init_stereo_control(&controls[nctl++],
-					emu->no_ac97 ? "Aux Capture Volume" : "Aux2 Capture Volume",
+					emu->card_capabilities->ac97_chip ? "Aux2 Capture Volume" : "Aux Capture Volume",
 					gpr, 0);
 	gpr += 2;
 	
@@ -1232,7 +1232,7 @@ A_OP(icode, &ptr, iMAC0, A_GPR(var), A_GPR(var), A_GPR(vol), A_EXTIN(input))
 	snd_emu10k1_init_mono_control(&controls[nctl++], "LFE Playback Volume", gpr, 0);
 	gpr++;
 	
-	if (emu->spk71) {
+	if (emu->card_capabilities->spk71) {
 		/* Stereo Mix Side Playback */
 		A_OP(icode, &ptr, iMAC0, A_GPR(playback+6), A_GPR(playback+6), A_GPR(gpr), A_GPR(stereo_mix));
 		A_OP(icode, &ptr, iMAC0, A_GPR(playback+7), A_GPR(playback+7), A_GPR(gpr+1), A_GPR(stereo_mix+1));
@@ -1266,7 +1266,7 @@ A_OP(icode, &ptr, iMAC0, A_GPR(var), A_GPR(var), A_GPR(vol), A_EXTIN(input))
 	A_OP(icode, &ptr, iACC3, A_GPR(playback + SND_EMU10K1_PLAYBACK_CHANNELS + 3), A_GPR(playback + 3), A_C_00000000, A_C_00000000); /* rear right */
 	A_OP(icode, &ptr, iACC3, A_GPR(playback + SND_EMU10K1_PLAYBACK_CHANNELS + 4), A_GPR(playback + 4), A_C_00000000, A_C_00000000); /* center */
 	A_OP(icode, &ptr, iACC3, A_GPR(playback + SND_EMU10K1_PLAYBACK_CHANNELS + 5), A_GPR(playback + 5), A_C_00000000, A_C_00000000); /* LFE */
-	if (emu->spk71) {
+	if (emu->card_capabilities->spk71) {
 		A_OP(icode, &ptr, iACC3, A_GPR(playback + SND_EMU10K1_PLAYBACK_CHANNELS + 6), A_GPR(playback + 6), A_C_00000000, A_C_00000000); /* side left */
 		A_OP(icode, &ptr, iACC3, A_GPR(playback + SND_EMU10K1_PLAYBACK_CHANNELS + 7), A_GPR(playback + 7), A_C_00000000, A_C_00000000); /* side right */
 	}
@@ -1359,7 +1359,7 @@ A_OP(icode, &ptr, iMAC0, A_GPR(var), A_GPR(var), A_GPR(vol), A_EXTIN(input))
 	A_PUT_STEREO_OUTPUT(A_EXTOUT_AREAR_L, A_EXTOUT_AREAR_R, playback+2 + SND_EMU10K1_PLAYBACK_CHANNELS);
 	A_PUT_OUTPUT(A_EXTOUT_ACENTER, playback+4 + SND_EMU10K1_PLAYBACK_CHANNELS);
 	A_PUT_OUTPUT(A_EXTOUT_ALFE, playback+5 + SND_EMU10K1_PLAYBACK_CHANNELS);
-	if (emu->spk71)
+	if (emu->card_capabilities->spk71)
 		A_PUT_STEREO_OUTPUT(A_EXTOUT_ASIDE_L, A_EXTOUT_ASIDE_R, playback+6 + SND_EMU10K1_PLAYBACK_CHANNELS);
 
 	/* headphone */
@@ -1982,22 +1982,27 @@ static int __devinit _snd_emu10k1_init_efx(emu10k1_t *emu)
 		OP(icode, &ptr, iACC3, EXTOUT(EXTOUT_MIC_CAP), GPR(capture + 2), C_00000000, C_00000000);
 
 	/* EFX capture - capture the 16 EXTINS */
-	OP(icode, &ptr, iACC3, FXBUS2(14), C_00000000, C_00000000, EXTIN(0));
-	OP(icode, &ptr, iACC3, FXBUS2(15), C_00000000, C_00000000, EXTIN(1));
-	OP(icode, &ptr, iACC3, FXBUS2(0), C_00000000, C_00000000, EXTIN(2));
-	OP(icode, &ptr, iACC3, FXBUS2(3), C_00000000, C_00000000, EXTIN(3));
-	/* Dont connect anything to FXBUS2 1 and 2.  These are shared with 
-	 * Center/LFE on the SBLive 5.1.  The kX driver only changes the 
-	 * routing when it detects an SBLive 5.1.
-	 *
-	 * Since only 14 of the 16 EXTINs are used, this is not a big problem.  
-	 * We route AC97L and R to FX capture 14 and 15, SPDIF CD in to FX capture 
-	 * 0 and 3, then the rest of the EXTINs to the corresponding FX capture 
-	 * channel.
-	 */
-	for (z = 4; z < 14; z++) {
-		OP(icode, &ptr, iACC3, FXBUS2(z), C_00000000, C_00000000, EXTIN(z));
+	if (emu->card_capabilities->sblive51) {
+		/* On the Live! 5.1, FXBUS2(1) and FXBUS(2) are shared with EXTOUT_ACENTER
+		 * and EXTOUT_ALFE, so we can't connect inputs to them for multitrack recording.
+		 *
+		 * Since only 14 of the 16 EXTINs are used, this is not a big problem.  
+		 * We route AC97L and R to FX capture 14 and 15, SPDIF CD in to FX capture 
+		 * 0 and 3, then the rest of the EXTINs to the corresponding FX capture 
+		 * channel.  Multitrack recorders will still see the center/lfe output signal 
+		 * on the second and third channels.
+		 */
+		OP(icode, &ptr, iACC3, FXBUS2(14), C_00000000, C_00000000, EXTIN(0));
+		OP(icode, &ptr, iACC3, FXBUS2(15), C_00000000, C_00000000, EXTIN(1));
+		OP(icode, &ptr, iACC3, FXBUS2(0), C_00000000, C_00000000, EXTIN(2));
+		OP(icode, &ptr, iACC3, FXBUS2(3), C_00000000, C_00000000, EXTIN(3));
+		for (z = 4; z < 14; z++)
+			OP(icode, &ptr, iACC3, FXBUS2(z), C_00000000, C_00000000, EXTIN(z));
+	} else {
+		for (z = 0; z < 16; z++)
+			OP(icode, &ptr, iACC3, FXBUS2(z), C_00000000, C_00000000, EXTIN(z));
 	}
+	    
 
 	if (gpr > tmp) {
 		snd_BUG();
@@ -2128,7 +2133,6 @@ static int snd_emu10k1_fx8010_info(emu10k1_t *emu, emu10k1_fx8010_info_t *info)
 	int res;
 
 	memset(info, 0, sizeof(info));
-	info->card = emu->card_type;
 	info->internal_tram_size = emu->fx8010.itram_size;
 	info->external_tram_size = emu->fx8010.etram_pages.bytes / 2;
 	fxbus = fxbuses;
diff --git a/sound/pci/emu10k1/emumixer.c b/sound/pci/emu10k1/emumixer.c
index 044663d31aa..d0b296587cc 100644
--- a/sound/pci/emu10k1/emumixer.c
+++ b/sound/pci/emu10k1/emumixer.c
@@ -791,7 +791,7 @@ int __devinit snd_emu10k1_mixer(emu10k1_t *emu)
 		NULL
 	};
 
-	if (!emu->no_ac97) {
+	if (emu->card_capabilities->ac97_chip) {
 		ac97_bus_t *pbus;
 		ac97_template_t ac97;
 		static ac97_bus_ops_t ops = {
@@ -833,7 +833,7 @@ int __devinit snd_emu10k1_mixer(emu10k1_t *emu)
 		for (; *c; c++)
 			remove_ctl(card, *c);
 	} else {
-		if (emu->APS)
+		if (emu->card_capabilities->ecard)
 			strcpy(emu->card->mixername, "EMU APS");
 		else if (emu->audigy)
 			strcpy(emu->card->mixername, "SB Audigy");
@@ -918,7 +918,7 @@ int __devinit snd_emu10k1_mixer(emu10k1_t *emu)
 		mix->attn[0] = 0xffff;
 	}
 	
-	if (! emu->APS) { /* FIXME: APS has these controls? */
+	if (! emu->card_capabilities->ecard) { /* FIXME: APS has these controls? */
 		/* sb live! and audigy */
 		if ((kctl = snd_ctl_new1(&snd_emu10k1_spdif_mask_control, emu)) == NULL)
 			return -ENOMEM;
@@ -939,14 +939,14 @@ int __devinit snd_emu10k1_mixer(emu10k1_t *emu)
 			return -ENOMEM;
 		if ((err = snd_ctl_add(card, kctl)))
 			return err;
-	} else if (! emu->APS) {
+	} else if (! emu->card_capabilities->ecard) {
 		/* sb live! */
 		if ((kctl = snd_ctl_new1(&snd_emu10k1_shared_spdif, emu)) == NULL)
 			return -ENOMEM;
 		if ((err = snd_ctl_add(card, kctl)))
 			return err;
 	}
-	if (emu->audigy && emu->revision == 4) { /* P16V */
+	if (emu->card_capabilities->ca0151_chip) { /* P16V */
 		if ((err = snd_p16v_mixer(emu)))
 			return err;
 	}
diff --git a/sound/pci/emu10k1/emuproc.c b/sound/pci/emu10k1/emuproc.c
index d990d5eb45a..732d17d1307 100644
--- a/sound/pci/emu10k1/emuproc.c
+++ b/sound/pci/emu10k1/emuproc.c
@@ -182,7 +182,7 @@ static void snd_emu10k1_proc_read(snd_info_entry_t *entry,
 	
 	snd_iprintf(buffer, "EMU10K1\n\n");
 	snd_iprintf(buffer, "Card                  : %s\n",
-		    emu->audigy ? "Audigy" : (emu->APS ? "EMU APS" : "Creative"));
+		    emu->audigy ? "Audigy" : (emu->card_capabilities->ecard ? "EMU APS" : "Creative"));
 	snd_iprintf(buffer, "Internal TRAM (words) : 0x%x\n", emu->fx8010.itram_size);
 	snd_iprintf(buffer, "External TRAM (words) : 0x%x\n", (int)emu->fx8010.etram_pages.bytes / 2);
 	snd_iprintf(buffer, "\n");
-- 
cgit v1.2.3-70-g09d2


From aec72e0a4be407fb69fbee812cf0028d62e75152 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Wed, 30 Mar 2005 14:22:25 +0200
Subject: [ALSA] Use old default id strings for compatibility

EMU10K1/EMU10K2 driver
Use expliciitly the old default id strings for backward compatibility.
This will make 'alsactl restore' working again.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 include/sound/emu10k1.h          |  5 +++--
 sound/pci/emu10k1/emu10k1_main.c | 18 ++++++++++++++++++
 2 files changed, 21 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/sound/emu10k1.h b/include/sound/emu10k1.h
index b1e8ee8e0fa..6647919768b 100644
--- a/include/sound/emu10k1.h
+++ b/include/sound/emu10k1.h
@@ -1050,8 +1050,9 @@ typedef struct {
 	unsigned char spdif_bug;    /* Has Spdif phasing bug */
 	unsigned char ac97_chip;    /* Has an AC97 chip */
 	unsigned char ecard;        /* APS EEPROM */
-	char * driver;
-	char * name;
+	const char *driver;
+	const char *name;
+	const char *id;		/* for backward compatibility - can be NULL if not needed */
 } emu_chip_details_t;
 
 struct _snd_emu10k1 {
diff --git a/sound/pci/emu10k1/emu10k1_main.c b/sound/pci/emu10k1/emu10k1_main.c
index 38be0f1b0e7..a2fa5012c84 100644
--- a/sound/pci/emu10k1/emu10k1_main.c
+++ b/sound/pci/emu10k1/emu10k1_main.c
@@ -616,15 +616,18 @@ static emu_chip_details_t emu_chip_details[] = {
 	/* Audigy 2 Value AC3 out does not work yet. Need to find out how to turn off interpolators.*/
 	{.vendor = 0x1102, .device = 0x0008, .subsystem = 0x10011102,
 	 .driver = "Audigy2", .name = "Audigy 2 Value [SB0400]", 
+	 .id = "Audigy2",
 	 .emu10k2_chip = 1,
 	 .ca0108_chip = 1,
 	 .spk71 = 1} ,
 	{.vendor = 0x1102, .device = 0x0008, 
 	 .driver = "Audigy2", .name = "Audigy 2 Value [Unknown]", 
+	 .id = "Audigy2",
 	 .emu10k2_chip = 1,
 	 .ca0108_chip = 1} ,
 	{.vendor = 0x1102, .device = 0x0004, .subsystem = 0x20071102,
 	 .driver = "Audigy2", .name = "Audigy 4 PRO [SB0380]", 
+	 .id = "Audigy2",
 	 .emu10k2_chip = 1,
 	 .ca0102_chip = 1,
 	 .ca0151_chip = 1,
@@ -633,6 +636,7 @@ static emu_chip_details_t emu_chip_details[] = {
 	 .ac97_chip = 1} ,
 	{.vendor = 0x1102, .device = 0x0004, .subsystem = 0x20021102,
 	 .driver = "Audigy2", .name = "Audigy 2 ZS [SB0350]", 
+	 .id = "Audigy2",
 	 .emu10k2_chip = 1,
 	 .ca0102_chip = 1,
 	 .ca0151_chip = 1,
@@ -641,6 +645,7 @@ static emu_chip_details_t emu_chip_details[] = {
 	 .ac97_chip = 1} ,
 	{.vendor = 0x1102, .device = 0x0004, .subsystem = 0x20011102,
 	 .driver = "Audigy2", .name = "Audigy 2 ZS [2001]", 
+	 .id = "Audigy2",
 	 .emu10k2_chip = 1,
 	 .ca0102_chip = 1,
 	 .ca0151_chip = 1,
@@ -649,6 +654,7 @@ static emu_chip_details_t emu_chip_details[] = {
 	 .ac97_chip = 1} ,
 	{.vendor = 0x1102, .device = 0x0004, .subsystem = 0x10071102,
 	 .driver = "Audigy2", .name = "Audigy 2 [SB0240]", 
+	 .id = "Audigy2",
 	 .emu10k2_chip = 1,
 	 .ca0102_chip = 1,
 	 .ca0151_chip = 1,
@@ -657,12 +663,14 @@ static emu_chip_details_t emu_chip_details[] = {
 	 .ac97_chip = 1} ,
 	{.vendor = 0x1102, .device = 0x0004, .subsystem = 0x10051102,
 	 .driver = "Audigy2", .name = "Audigy 2 EX [1005]", 
+	 .id = "Audigy2",
 	 .emu10k2_chip = 1,
 	 .ca0102_chip = 1,
 	 .ca0151_chip = 1,
 	 .spdif_bug = 1} ,
 	{.vendor = 0x1102, .device = 0x0004, .subsystem = 0x10021102,
 	 .driver = "Audigy2", .name = "Audigy 2 Platinum [SB0240P]", 
+	 .id = "Audigy2",
 	 .emu10k2_chip = 1,
 	 .ca0102_chip = 1,
 	 .ca0151_chip = 1,
@@ -671,34 +679,41 @@ static emu_chip_details_t emu_chip_details[] = {
 	 .ac97_chip = 1} ,
 	{.vendor = 0x1102, .device = 0x0004, .subsystem = 0x10020052,
 	 .driver = "Audigy", .name = "Audigy 1 ES [SB0160]", 
+	 .id = "Audigy",
 	 .emu10k2_chip = 1,
 	 .ca0102_chip = 1,
 	 .spdif_bug = 1} ,
 	{.vendor = 0x1102, .device = 0x0004,
 	 .driver = "Audigy", .name = "Audigy 1 or 2 [Unknown]", 
+	 .id = "Audigy",
 	 .emu10k2_chip = 1,
 	 .ca0102_chip = 1,
 	 .spdif_bug = 1} ,
 	{.vendor = 0x1102, .device = 0x0002, .subsystem = 0x40011102,
 	 .driver = "EMU10K1", .name = "E-mu APS [4001]", 
+	 .id = "APS",
 	 .emu10k1_chip = 1,
 	 .ecard = 1} ,
 	{.vendor = 0x1102, .device = 0x0002, .subsystem = 0x80641102,
 	 .driver = "EMU10K1", .name = "SB Live 5.1", 
+	 .id = "Live",
 	 .emu10k1_chip = 1,
 	 .ac97_chip = 1,
 	 .sblive51 = 1} ,
 	{.vendor = 0x1102, .device = 0x0002, .subsystem = 0x80401102,
 	 .driver = "EMU10K1", .name = "SBLive! Platinum [CT4760P]", 
+	 .id = "Live",
 	 .emu10k1_chip = 1,
 	 .ac97_chip = 1} ,
 	{.vendor = 0x1102, .device = 0x0002, .subsystem = 0x80271102,
 	 .driver = "EMU10K1", .name = "SBLive! Value [CT4832]", 
+	 .id = "Live",
 	 .emu10k1_chip = 1,
 	 .ac97_chip = 1,
 	 .sblive51 = 1} ,
 	{.vendor = 0x1102, .device = 0x0002,
 	 .driver = "EMU10K1", .name = "SB Live [Unknown]", 
+	 .id = "Live",
 	 .emu10k1_chip = 1,
 	 .ac97_chip = 1,
 	 .sblive51 = 1} ,
@@ -772,6 +787,9 @@ int __devinit snd_emu10k1_create(snd_card_t * card,
 	else
 		snd_printdd("Sound card name=%s, vendor=0x%x, device=0x%x, subsystem=0x%x\n", c->name, pci->vendor, pci->device, emu->serial);
 	
+	if (!*card->id && c->id)
+		strlcpy(card->id, c->id, sizeof(card->id));
+
 	is_audigy = emu->audigy = c->emu10k2_chip;
 
 	/* set the DMA transfer mask */
-- 
cgit v1.2.3-70-g09d2


From 024ac44c701d43f5e2d34bd6a35b2813a36e6010 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Sun, 29 May 2005 02:26:31 -0500
Subject: Input: This patch implements compat_ioctl for joydev.

       I've tested it with a Logitech WingMan Rumblepad on an x86-64
       machine, and on an ia32 machine to make sure I didn't break
       anything.

Signed-off-by: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Vojtech Pavlik <vojtech@suse.cz>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/joydev.c   | 116 +++++++++++++++++++++++++++++++++++++----------
 include/linux/joystick.h |  33 ++++++++++----
 2 files changed, 116 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/drivers/input/joydev.c b/drivers/input/joydev.c
index 627d343dfba..816a585a0e6 100644
--- a/drivers/input/joydev.c
+++ b/drivers/input/joydev.c
@@ -285,48 +285,33 @@ static unsigned int joydev_poll(struct file *file, poll_table *wait)
 		(POLLIN | POLLRDNORM) : 0) | (list->joydev->exist ? 0 : (POLLHUP | POLLERR));
 }
 
-static int joydev_ioctl(struct inode *inode, struct file *file, unsigned int cmd, unsigned long arg)
+static int joydev_ioctl_common(struct joydev *joydev, unsigned int cmd, void __user *argp)
 {
-	struct joydev_list *list = file->private_data;
-	struct joydev *joydev = list->joydev;
 	struct input_dev *dev = joydev->handle.dev;
-	void __user *argp = (void __user *)arg;
 	int i, j;
 
-	if (!joydev->exist) return -ENODEV;
-
 	switch (cmd) {
 
 		case JS_SET_CAL:
 			return copy_from_user(&joydev->glue.JS_CORR, argp,
-				sizeof(struct JS_DATA_TYPE)) ? -EFAULT : 0;
+				sizeof(joydev->glue.JS_CORR)) ? -EFAULT : 0;
 		case JS_GET_CAL:
 			return copy_to_user(argp, &joydev->glue.JS_CORR,
-				sizeof(struct JS_DATA_TYPE)) ? -EFAULT : 0;
+				sizeof(joydev->glue.JS_CORR)) ? -EFAULT : 0;
 		case JS_SET_TIMEOUT:
-			return get_user(joydev->glue.JS_TIMEOUT, (int __user *) arg);
+			return get_user(joydev->glue.JS_TIMEOUT, (s32 __user *) argp);
 		case JS_GET_TIMEOUT:
-			return put_user(joydev->glue.JS_TIMEOUT, (int __user *) arg);
-		case JS_SET_TIMELIMIT:
-			return get_user(joydev->glue.JS_TIMELIMIT, (long __user *) arg);
-		case JS_GET_TIMELIMIT:
-			return put_user(joydev->glue.JS_TIMELIMIT, (long __user *) arg);
-		case JS_SET_ALL:
-			return copy_from_user(&joydev->glue, argp,
-						sizeof(struct JS_DATA_SAVE_TYPE)) ? -EFAULT : 0;
-		case JS_GET_ALL:
-			return copy_to_user(argp, &joydev->glue,
-						sizeof(struct JS_DATA_SAVE_TYPE)) ? -EFAULT : 0;
+			return put_user(joydev->glue.JS_TIMEOUT, (s32 __user *) argp);
 
 		case JSIOCGVERSION:
-			return put_user(JS_VERSION, (__u32 __user *) arg);
+			return put_user(JS_VERSION, (__u32 __user *) argp);
 		case JSIOCGAXES:
-			return put_user(joydev->nabs, (__u8 __user *) arg);
+			return put_user(joydev->nabs, (__u8 __user *) argp);
 		case JSIOCGBUTTONS:
-			return put_user(joydev->nkey, (__u8 __user *) arg);
+			return put_user(joydev->nkey, (__u8 __user *) argp);
 		case JSIOCSCORR:
 			if (copy_from_user(joydev->corr, argp,
-				      sizeof(struct js_corr) * joydev->nabs))
+				      sizeof(joydev->corr[0]) * joydev->nabs))
 			    return -EFAULT;
 			for (i = 0; i < joydev->nabs; i++) {
 				j = joydev->abspam[i];
@@ -335,7 +320,7 @@ static int joydev_ioctl(struct inode *inode, struct file *file, unsigned int cmd
 			return 0;
 		case JSIOCGCORR:
 			return copy_to_user(argp, joydev->corr,
-						sizeof(struct js_corr) * joydev->nabs) ? -EFAULT : 0;
+						sizeof(joydev->corr[0]) * joydev->nabs) ? -EFAULT : 0;
 		case JSIOCSAXMAP:
 			if (copy_from_user(joydev->abspam, argp, sizeof(__u8) * (ABS_MAX + 1)))
 				return -EFAULT;
@@ -371,6 +356,84 @@ static int joydev_ioctl(struct inode *inode, struct file *file, unsigned int cmd
 	return -EINVAL;
 }
 
+#ifdef CONFIG_COMPAT
+static long joydev_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	struct joydev_list *list = file->private_data;
+	struct joydev *joydev = list->joydev;
+	void __user *argp = (void __user *)arg;
+	s32 tmp32;
+	struct JS_DATA_SAVE_TYPE_32 ds32;
+	int err;
+
+	if (!joydev->exist) return -ENODEV;
+	switch(cmd) {
+	case JS_SET_TIMELIMIT:
+		err = get_user(tmp32, (s32 __user *) arg);
+		if (err == 0)
+			joydev->glue.JS_TIMELIMIT = tmp32;
+		break;
+	case JS_GET_TIMELIMIT:
+		tmp32 = joydev->glue.JS_TIMELIMIT;
+		err = put_user(tmp32, (s32 __user *) arg);
+		break;
+
+	case JS_SET_ALL:
+		err = copy_from_user(&ds32, argp,
+				     sizeof(ds32)) ? -EFAULT : 0;
+		if (err == 0) {
+			joydev->glue.JS_TIMEOUT    = ds32.JS_TIMEOUT;
+			joydev->glue.BUSY          = ds32.BUSY;
+			joydev->glue.JS_EXPIRETIME = ds32.JS_EXPIRETIME;
+			joydev->glue.JS_TIMELIMIT  = ds32.JS_TIMELIMIT;
+			joydev->glue.JS_SAVE       = ds32.JS_SAVE;
+			joydev->glue.JS_CORR       = ds32.JS_CORR;
+		}
+		break;
+
+	case JS_GET_ALL:
+		ds32.JS_TIMEOUT    = joydev->glue.JS_TIMEOUT;
+		ds32.BUSY          = joydev->glue.BUSY;
+		ds32.JS_EXPIRETIME = joydev->glue.JS_EXPIRETIME;
+		ds32.JS_TIMELIMIT  = joydev->glue.JS_TIMELIMIT;
+		ds32.JS_SAVE       = joydev->glue.JS_SAVE;
+		ds32.JS_CORR       = joydev->glue.JS_CORR;
+
+		err = copy_to_user(argp, &ds32,
+					  sizeof(ds32)) ? -EFAULT : 0;
+		break;
+
+	default:
+		err = joydev_ioctl_common(joydev, cmd, argp);
+	}
+	return err;
+}
+#endif /* CONFIG_COMPAT */
+
+static int joydev_ioctl(struct inode *inode, struct file *file, unsigned int cmd, unsigned long arg)
+{
+	struct joydev_list *list = file->private_data;
+	struct joydev *joydev = list->joydev;
+	void __user *argp = (void __user *)arg;
+
+	if (!joydev->exist) return -ENODEV;
+
+	switch(cmd) {
+		case JS_SET_TIMELIMIT:
+			return get_user(joydev->glue.JS_TIMELIMIT, (long __user *) arg);
+		case JS_GET_TIMELIMIT:
+			return put_user(joydev->glue.JS_TIMELIMIT, (long __user *) arg);
+		case JS_SET_ALL:
+			return copy_from_user(&joydev->glue, argp,
+						sizeof(joydev->glue)) ? -EFAULT : 0;
+		case JS_GET_ALL:
+			return copy_to_user(argp, &joydev->glue,
+						sizeof(joydev->glue)) ? -EFAULT : 0;
+		default:
+			return joydev_ioctl_common(joydev, cmd, argp);
+	}
+}
+
 static struct file_operations joydev_fops = {
 	.owner =	THIS_MODULE,
 	.read =		joydev_read,
@@ -379,6 +442,9 @@ static struct file_operations joydev_fops = {
 	.open =		joydev_open,
 	.release =	joydev_release,
 	.ioctl =	joydev_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl =	joydev_compat_ioctl,
+#endif
 	.fasync =	joydev_fasync,
 };
 
diff --git a/include/linux/joystick.h b/include/linux/joystick.h
index b7e0ab622cd..06b9af77eb7 100644
--- a/include/linux/joystick.h
+++ b/include/linux/joystick.h
@@ -111,18 +111,35 @@ struct js_corr {
 #define JS_SET_ALL		8
 
 struct JS_DATA_TYPE {
-	int buttons;
-	int x;
-	int y;
+	__s32 buttons;
+	__s32 x;
+	__s32 y;
 };
 
-struct JS_DATA_SAVE_TYPE {
-	int JS_TIMEOUT;
-	int BUSY;
-	long JS_EXPIRETIME;
-	long JS_TIMELIMIT;
+struct JS_DATA_SAVE_TYPE_32 {
+	__s32 JS_TIMEOUT;
+	__s32 BUSY;
+	__s32 JS_EXPIRETIME;
+	__s32 JS_TIMELIMIT;
 	struct JS_DATA_TYPE JS_SAVE;
 	struct JS_DATA_TYPE JS_CORR;
 };
 
+struct JS_DATA_SAVE_TYPE_64 {
+	__s32 JS_TIMEOUT;
+	__s32 BUSY;
+	__s64 JS_EXPIRETIME;
+	__s64 JS_TIMELIMIT;
+	struct JS_DATA_TYPE JS_SAVE;
+	struct JS_DATA_TYPE JS_CORR;
+};
+
+#if BITS_PER_LONG == 64
+#define JS_DATA_SAVE_TYPE JS_DATA_SAVE_TYPE_64
+#elif BITS_PER_LONG == 32
+#define JS_DATA_SAVE_TYPE JS_DATA_SAVE_TYPE_32
+#else
+#error Unexpected BITS_PER_LONG
+#endif
+
 #endif /* _LINUX_JOYSTICK_H */
-- 
cgit v1.2.3-70-g09d2


From 0fbf87caf70acec0c435233fbc39c7bd0aca3ca6 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor_core@ameritech.net>
Date: Sun, 29 May 2005 02:29:25 -0500
Subject: Input: add semaphore and user count to input_dev structure;       
 serialize open and close calls and ensure that device's        open and close
 methods are only called when first user        opens it or last user closes
 it.

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/input.c | 33 ++++++++++++++++++++++++++++-----
 include/linux/input.h |  4 ++++
 2 files changed, 32 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/drivers/input/input.c b/drivers/input/input.c
index 3385dd03abf..1885f369e3e 100644
--- a/drivers/input/input.c
+++ b/drivers/input/input.c
@@ -219,10 +219,24 @@ void input_release_device(struct input_handle *handle)
 
 int input_open_device(struct input_handle *handle)
 {
+	struct input_dev *dev = handle->dev;
+	int err;
+
+	err = down_interruptible(&dev->sem);
+	if (err)
+		return err;
+
 	handle->open++;
-	if (handle->dev->open)
-		return handle->dev->open(handle->dev);
-	return 0;
+
+	if (!dev->users++ && dev->open)
+		err = dev->open(dev);
+
+	if (err)
+		handle->open--;
+
+	up(&dev->sem);
+
+	return err;
 }
 
 int input_flush_device(struct input_handle* handle, struct file* file)
@@ -235,10 +249,17 @@ int input_flush_device(struct input_handle* handle, struct file* file)
 
 void input_close_device(struct input_handle *handle)
 {
+	struct input_dev *dev = handle->dev;
+
 	input_release_device(handle);
-	if (handle->dev->close)
-		handle->dev->close(handle->dev);
+
+	down(&dev->sem);
+
+	if (!--dev->users && dev->close)
+		dev->close(dev);
 	handle->open--;
+
+	up(&dev->sem);
 }
 
 static void input_link_handle(struct input_handle *handle)
@@ -415,6 +436,8 @@ void input_register_device(struct input_dev *dev)
 
 	set_bit(EV_SYN, dev->evbit);
 
+	init_MUTEX(&dev->sem);
+
 	/*
 	 * If delay and period are pre-set by the driver, then autorepeating
 	 * is handled by the driver itself and we don't do it in input.c.
diff --git a/include/linux/input.h b/include/linux/input.h
index 72731d7d189..43e8ecec602 100644
--- a/include/linux/input.h
+++ b/include/linux/input.h
@@ -859,6 +859,10 @@ struct input_dev {
 	int (*erase_effect)(struct input_dev *dev, int effect_id);
 
 	struct input_handle *grab;
+
+	struct semaphore sem;	/* serializes open and close operations */
+	unsigned int users;
+
 	struct device *dev;
 
 	struct list_head	h_list;
-- 
cgit v1.2.3-70-g09d2


From bdaed50292bea3e2b20c68c2ffe9dbde7c0d6910 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Thu, 7 Apr 2005 15:48:42 +0200
Subject: [ALSA] Check revision for the proper detection of audigy 2

EMU10K1/EMU10K2 driver
Check ther revision to detect non-listed audigy 2 boards.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 include/sound/emu10k1.h          |  1 +
 sound/pci/emu10k1/emu10k1_main.c | 18 ++++++++++++++----
 2 files changed, 15 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/sound/emu10k1.h b/include/sound/emu10k1.h
index 6647919768b..f5babd3f845 100644
--- a/include/sound/emu10k1.h
+++ b/include/sound/emu10k1.h
@@ -1040,6 +1040,7 @@ typedef struct {
 	u32 vendor;
 	u32 device;
 	u32 subsystem;
+	unsigned char revision;
 	unsigned char emu10k1_chip; /* Original SB Live. Not SB Live 24bit. */
 	unsigned char emu10k2_chip; /* Audigy 1 or Audigy 2. */
 	unsigned char ca0102_chip;  /* Audigy 1 or Audigy 2. Not SB Audigy 2 Value. */
diff --git a/sound/pci/emu10k1/emu10k1_main.c b/sound/pci/emu10k1/emu10k1_main.c
index a3a1a10fb0c..c6d53b45925 100644
--- a/sound/pci/emu10k1/emu10k1_main.c
+++ b/sound/pci/emu10k1/emu10k1_main.c
@@ -679,6 +679,14 @@ static emu_chip_details_t emu_chip_details[] = {
 	 .spk71 = 1,
 	 .spdif_bug = 1,
 	 .ac97_chip = 1} ,
+	{.vendor = 0x1102, .device = 0x0004, .revision = 0x04,
+	 .driver = "Audigy2", .name = "Audigy 2 [Unknown]",
+	 .id = "Audigy2",
+	 .emu10k2_chip = 1,
+	 .ca0102_chip = 1,
+	 .ca0151_chip = 1,
+	 .spdif_bug = 1,
+	 .ac97_chip = 1} ,
 	{.vendor = 0x1102, .device = 0x0004, .subsystem = 0x10020052,
 	 .driver = "Audigy", .name = "Audigy 1 ES [SB0160]", 
 	 .id = "Audigy",
@@ -693,11 +701,10 @@ static emu_chip_details_t emu_chip_details[] = {
 	 .ca0102_chip = 1,
 	 .ac97_chip = 1} ,
 	{.vendor = 0x1102, .device = 0x0004,
-	 .driver = "Audigy", .name = "Audigy 1 or 2 [Unknown]", 
+	 .driver = "Audigy", .name = "Audigy 1 [Unknown]", 
 	 .id = "Audigy",
 	 .emu10k2_chip = 1,
 	 .ca0102_chip = 1,
-	 .spdif_bug = 1,
 	 .ac97_chip = 1} ,
 	{.vendor = 0x1102, .device = 0x0002, .subsystem = 0x40011102,
 	 .driver = "EMU10K1", .name = "E-mu APS [4001]", 
@@ -781,8 +788,11 @@ int __devinit snd_emu10k1_create(snd_card_t * card,
 
 	for (c = emu_chip_details; c->vendor; c++) {
 		if (c->vendor == pci->vendor && c->device == pci->device) {
-			if (c->subsystem == emu->serial) break;
-			if (c->subsystem == 0) break;
+			if (c->subsystem && c->subsystem != emu->serial)
+				continue;
+			if (c->revision && c->revision != emu->revision)
+				continue;
+			break;
 		}
 	}
 	if (c->vendor == 0) {
-- 
cgit v1.2.3-70-g09d2


From 001f758990d685e7023008763795f1970ef56614 Mon Sep 17 00:00:00 2001
From: James Courtier-Dutton <James@superbug.co.uk>
Date: Sat, 9 Apr 2005 23:38:25 +0200
Subject: [ALSA] Improve SPDIF playback via the P16V/CA0151 chip.

EMU10K1/EMU10K2 driver
Although we can set 44100 as the output rate, the SPDIF can do it, but the Analog output cannot.
The SPDIF has the bug, whereby the Left channel arrives one sample late, so although we don't do any resampling,
it is not good for AC3 non-audio output.

Signed-off-by: James Courtier-Dutton <James@superbug.co.uk>
---
 include/sound/emu10k1.h      |  1 +
 sound/pci/emu10k1/emumixer.c |  2 ++
 sound/pci/emu10k1/emuproc.c  | 21 +++++++++++++++++++++
 sound/pci/emu10k1/p16v.c     | 16 +++++++---------
 4 files changed, 31 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/sound/emu10k1.h b/include/sound/emu10k1.h
index f5babd3f845..61a3f418f30 100644
--- a/include/sound/emu10k1.h
+++ b/include/sound/emu10k1.h
@@ -747,6 +747,7 @@
 						/* Assumes sample lock				*/
 
 /* These three bitfields apply to CDSRCS, GPSRCS, and (except as noted) ZVSRCS.			*/
+#define SRCS_SPDIFVALID		0x04000000	/* SPDIF stream valid				*/
 #define SRCS_SPDIFLOCKED	0x02000000	/* SPDIF stream locked				*/
 #define SRCS_RATELOCKED		0x01000000	/* Sample rate locked				*/
 #define SRCS_ESTSAMPLERATE	0x0007ffff	/* Do not modify this field.			*/
diff --git a/sound/pci/emu10k1/emumixer.c b/sound/pci/emu10k1/emumixer.c
index d0b296587cc..b544c258266 100644
--- a/sound/pci/emu10k1/emumixer.c
+++ b/sound/pci/emu10k1/emumixer.c
@@ -935,10 +935,12 @@ int __devinit snd_emu10k1_mixer(emu10k1_t *emu)
 			return -ENOMEM;
 		if ((err = snd_ctl_add(card, kctl)))
 			return err;
+#if 0
 		if ((kctl = snd_ctl_new1(&snd_audigy_spdif_output_rate, emu)) == NULL)
 			return -ENOMEM;
 		if ((err = snd_ctl_add(card, kctl)))
 			return err;
+#endif
 	} else if (! emu->card_capabilities->ecard) {
 		/* sb live! */
 		if ((kctl = snd_ctl_new1(&snd_emu10k1_shared_spdif, emu)) == NULL)
diff --git a/sound/pci/emu10k1/emuproc.c b/sound/pci/emu10k1/emuproc.c
index 187a4e60a5f..356fb710425 100644
--- a/sound/pci/emu10k1/emuproc.c
+++ b/sound/pci/emu10k1/emuproc.c
@@ -30,6 +30,7 @@
 #include <linux/init.h>
 #include <sound/core.h>
 #include <sound/emu10k1.h>
+#include "p16v.h"
 
 static void snd_emu10k1_proc_spdif_status(emu10k1_t * emu,
 					  snd_info_buffer_t * buffer,
@@ -62,6 +63,7 @@ static void snd_emu10k1_proc_spdif_status(emu10k1_t * emu,
 
 		if (rate_reg > 0) {
 			rate = snd_emu10k1_ptr_read(emu, rate_reg, 0);
+			snd_iprintf(buffer, "S/PDIF Valid          : %s\n", rate & SRCS_SPDIFVALID ? "on" : "off");
 			snd_iprintf(buffer, "S/PDIF Locked         : %s\n", rate & SRCS_SPDIFLOCKED ? "on" : "off");
 			snd_iprintf(buffer, "Rate Locked           : %s\n", rate & SRCS_RATELOCKED ? "on" : "off");
 			/* From ((Rate * 48000 ) / 262144); */
@@ -244,6 +246,21 @@ static void snd_emu10k1_proc_spdif_read(snd_info_entry_t *entry,
 #endif
 }
 
+static void snd_emu10k1_proc_rates_read(snd_info_entry_t *entry, 
+				  snd_info_buffer_t * buffer)
+{
+	static int samplerate[8] = { 44100, 48000, 96000, 192000, 4, 5, 6, 7 };
+	emu10k1_t *emu = entry->private_data;
+	unsigned int val, tmp, n;
+	val = snd_emu10k1_ptr20_read(emu, CAPTURE_RATE_STATUS, 0);
+	tmp = (val >> 16) & 0x8;
+	for (n=0;n<4;n++) {
+		tmp = val >> (16 + (n*4));
+		if (tmp & 0x8) snd_iprintf(buffer, "Channel %d: Rate=%d\n", n, samplerate[tmp & 0x7]);
+		else snd_iprintf(buffer, "Channel %d: No input\n", n);
+	}
+}
+
 static void snd_emu10k1_proc_acode_read(snd_info_entry_t *entry, 
 				        snd_info_buffer_t * buffer)
 {
@@ -540,6 +557,10 @@ int __devinit snd_emu10k1_proc_init(emu10k1_t * emu)
 		if (! snd_card_proc_new(emu->card, "spdif-in", &entry))
 			snd_info_set_text_ops(entry, emu, 2048, snd_emu10k1_proc_spdif_read);
 	}
+	if (emu->card_capabilities->ca0151_chip) {
+		if (! snd_card_proc_new(emu->card, "capture-rates", &entry))
+			snd_info_set_text_ops(entry, emu, 2048, snd_emu10k1_proc_rates_read);
+	}
 
 	if (! snd_card_proc_new(emu->card, "voices", &entry))
 		snd_info_set_text_ops(entry, emu, 2048, snd_emu10k1_proc_voices_read);
diff --git a/sound/pci/emu10k1/p16v.c b/sound/pci/emu10k1/p16v.c
index dd6ce9927e1..8dd87838fb2 100644
--- a/sound/pci/emu10k1/p16v.c
+++ b/sound/pci/emu10k1/p16v.c
@@ -119,8 +119,8 @@ static snd_pcm_hardware_t snd_p16v_playback_hw = {
 				 SNDRV_PCM_INFO_BLOCK_TRANSFER |
 				 SNDRV_PCM_INFO_MMAP_VALID),
 	.formats =		SNDRV_PCM_FMTBIT_S32_LE, /* Only supports 24-bit samples padded to 32 bits. */
-	.rates =		SNDRV_PCM_RATE_192000 | SNDRV_PCM_RATE_96000 | SNDRV_PCM_RATE_48000 ,
-	.rate_min =		48000,
+	.rates =		SNDRV_PCM_RATE_192000 | SNDRV_PCM_RATE_96000 | SNDRV_PCM_RATE_48000 | SNDRV_PCM_RATE_44100, 
+	.rate_min =		44100,
 	.rate_max =		192000,
 	.channels_min =		8, 
 	.channels_max =		8,
@@ -324,19 +324,17 @@ static int snd_p16v_pcm_prepare_playback(snd_pcm_substream_t *substream)
 	tmp = snd_emu10k1_ptr_read(emu, A_SPDIF_SAMPLERATE, channel);
         switch (runtime->rate) {
 	case 44100:
-	  snd_emu10k1_ptr_write(emu, A_SPDIF_SAMPLERATE, channel, (tmp & ~0xe000) | 0x8000); /* FIXME: This will change the capture rate as well! */
-	  break;
-	case 48000:
-	  snd_emu10k1_ptr_write(emu, A_SPDIF_SAMPLERATE, channel, (tmp & ~0xe000) | 0x0000); /* FIXME: This will change the capture rate as well! */
+	  snd_emu10k1_ptr_write(emu, A_SPDIF_SAMPLERATE, channel, (tmp & ~0xe0e0) | 0x8080);
 	  break;
 	case 96000:
-	  snd_emu10k1_ptr_write(emu, A_SPDIF_SAMPLERATE, channel, (tmp & ~0xe000) | 0x4000); /* FIXME: This will change the capture rate as well! */
+	  snd_emu10k1_ptr_write(emu, A_SPDIF_SAMPLERATE, channel, (tmp & ~0xe0e0) | 0x4040);
 	  break;
 	case 192000:
-	  snd_emu10k1_ptr_write(emu, A_SPDIF_SAMPLERATE, channel, (tmp & ~0xe000) | 0x2000); /* FIXME: This will change the capture rate as well! */
+	  snd_emu10k1_ptr_write(emu, A_SPDIF_SAMPLERATE, channel, (tmp & ~0xe0e0) | 0x2020);
 	  break;
+	case 48000:
 	default:
-	  snd_emu10k1_ptr_write(emu, A_SPDIF_SAMPLERATE, channel, 0x0000); /* FIXME: This will change the capture rate as well! */
+	  snd_emu10k1_ptr_write(emu, A_SPDIF_SAMPLERATE, channel, (tmp & ~0xe0e0) | 0x0000);
 	  break;
 	}
 	/* FIXME: Check emu->buffer.size before actually writing to it. */
-- 
cgit v1.2.3-70-g09d2


From 267cdf4036ed9e8565a7d909fdf854b9c7e1c5ff Mon Sep 17 00:00:00 2001
From: Jaroslav Kysela <perex@suse.cz>
Date: Wed, 13 Apr 2005 13:25:30 +0200
Subject: [ALSA] replace SNDRV_PCM_HW_PARAMS_RUNTIME ->
 SNDRV_PCM_HW_PARAMS_NORESAMPLE

ALSA Core


Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/asound.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/sound/asound.h b/include/sound/asound.h
index a4d149f3454..716227eed3e 100644
--- a/include/sound/asound.h
+++ b/include/sound/asound.h
@@ -344,7 +344,7 @@ enum sndrv_pcm_hw_param {
 	SNDRV_PCM_HW_PARAM_LAST_INTERVAL = SNDRV_PCM_HW_PARAM_TICK_TIME
 };
 
-#define SNDRV_PCM_HW_PARAMS_RUNTIME		(1<<0)
+#define SNDRV_PCM_HW_PARAMS_NORESAMPLE		(1<<0)	/* avoid rate resampling */
 
 struct sndrv_interval {
 	unsigned int min, max;
-- 
cgit v1.2.3-70-g09d2


From eb8caf30f4c059ddfdfa32b6034549622953db6f Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Wed, 13 Apr 2005 14:32:57 +0200
Subject: [ALSA] Improve the shared-jack handling on ac97

AC97 Codec
The handling of shared surround/clfe output jacks with line/mic-in
on some AC97 codecs is improved.

Instead of 'Line-In As Surround' or 'Mic As Center/LFE' switch, two
new enum controls are introduced: 'Channel Mode' and 'Surround Jack Mode'.
The formar changes the current output mode among 2, 4 and 6-channels.
The latter controls whether the jacks are shared or independent.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 include/sound/ac97_codec.h  |   4 +
 sound/pci/ac97/ac97_patch.c | 426 ++++++++++++++++++++++++--------------------
 2 files changed, 240 insertions(+), 190 deletions(-)

(limited to 'include')

diff --git a/include/sound/ac97_codec.h b/include/sound/ac97_codec.h
index 2433e279e07..996eeab683b 100644
--- a/include/sound/ac97_codec.h
+++ b/include/sound/ac97_codec.h
@@ -437,6 +437,7 @@ struct snd_ac97_build_ops {
 	void (*suspend) (ac97_t *ac97);
 	void (*resume) (ac97_t *ac97);
 #endif
+	void (*update_jacks) (ac97_t *ac97);	/* for jack-sharing */
 };
 
 struct _snd_ac97_bus_ops {
@@ -516,6 +517,9 @@ struct _snd_ac97 {
 		} ad18xx;
 		unsigned int dev_flags;		/* device specific */
 	} spec;
+	/* jack-sharing info */
+	unsigned char indep_surround;
+	unsigned char channel_mode;
 };
 
 /* conditions */
diff --git a/sound/pci/ac97/ac97_patch.c b/sound/pci/ac97/ac97_patch.c
index 473840d431b..7f16c306165 100644
--- a/sound/pci/ac97/ac97_patch.c
+++ b/sound/pci/ac97/ac97_patch.c
@@ -64,6 +64,116 @@ static int ac97_update_bits_page(ac97_t *ac97, unsigned short reg, unsigned shor
 	return ret;
 }
 
+/*
+ * shared line-in/mic controls
+ */
+static int ac97_enum_text_info(snd_kcontrol_t *kcontrol, snd_ctl_elem_info_t *uinfo,
+			       const char **texts, unsigned int nums)
+{
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED;
+	uinfo->count = 1;
+	uinfo->value.enumerated.items = nums;
+	if (uinfo->value.enumerated.item > nums - 1)
+		uinfo->value.enumerated.item = nums - 1;
+	strcpy(uinfo->value.enumerated.name, texts[uinfo->value.enumerated.item]);
+	return 0;
+}
+
+static int ac97_surround_jack_mode_info(snd_kcontrol_t *kcontrol, snd_ctl_elem_info_t *uinfo)
+{
+	static const char *texts[] = { "Shared", "Independent" };
+	return ac97_enum_text_info(kcontrol, uinfo, texts, 2);
+}
+
+static int ac97_surround_jack_mode_get(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucontrol)
+{
+	ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
+
+	ucontrol->value.enumerated.item[0] = ac97->indep_surround;
+	return 0;
+}
+
+static int ac97_surround_jack_mode_put(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucontrol)
+{
+	ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
+	unsigned char indep = !!ucontrol->value.enumerated.item[0];
+
+	if (indep != ac97->indep_surround) {
+		ac97->indep_surround = indep;
+		if (ac97->build_ops->update_jacks)
+			ac97->build_ops->update_jacks(ac97);
+		return 1;
+	}
+	return 0;
+}
+
+static int ac97_channel_mode_info(snd_kcontrol_t *kcontrol, snd_ctl_elem_info_t *uinfo)
+{
+	static const char *texts[] = { "2ch", "4ch", "6ch" };
+	if (kcontrol->private_value)
+		return ac97_enum_text_info(kcontrol, uinfo, texts, 2); /* 4ch only */
+	return ac97_enum_text_info(kcontrol, uinfo, texts, 3);
+}
+
+static int ac97_channel_mode_get(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucontrol)
+{
+	ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
+
+	ucontrol->value.enumerated.item[0] = ac97->channel_mode;
+	return 0;
+}
+
+static int ac97_channel_mode_put(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucontrol)
+{
+	ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
+	unsigned char mode = ucontrol->value.enumerated.item[0];
+
+	if (mode != ac97->channel_mode) {
+		ac97->channel_mode = mode;
+		if (ac97->build_ops->update_jacks)
+			ac97->build_ops->update_jacks(ac97);
+		return 1;
+	}
+	return 0;
+}
+
+#define AC97_SURROUND_JACK_MODE_CTL \
+	{ \
+		.iface	= SNDRV_CTL_ELEM_IFACE_MIXER, \
+		.name	= "Surround Jack Mode", \
+		.info = ac97_surround_jack_mode_info, \
+		.get = ac97_surround_jack_mode_get, \
+		.put = ac97_surround_jack_mode_put, \
+	}
+#define AC97_CHANNEL_MODE_CTL \
+	{ \
+		.iface	= SNDRV_CTL_ELEM_IFACE_MIXER, \
+		.name	= "Channel Mode", \
+		.info = ac97_channel_mode_info, \
+		.get = ac97_channel_mode_get, \
+		.put = ac97_channel_mode_put, \
+	}
+#define AC97_CHANNEL_MODE_4CH_CTL \
+	{ \
+		.iface	= SNDRV_CTL_ELEM_IFACE_MIXER, \
+		.name	= "Channel Mode", \
+		.info = ac97_channel_mode_info, \
+		.get = ac97_channel_mode_get, \
+		.put = ac97_channel_mode_put, \
+		.private_value = 1, \
+	}
+
+static inline int is_shared_linein(ac97_t *ac97)
+{
+	return ! ac97->indep_surround && ac97->channel_mode >= 1;
+}
+
+static inline int is_shared_micin(ac97_t *ac97)
+{
+	return ! ac97->indep_surround && ac97->channel_mode >= 2;
+}
+
+
 /* The following snd_ac97_ymf753_... items added by David Shust (dshust@shustring.com) */
 
 /* It is possible to indicate to the Yamaha YMF753 the type of speakers being used. */
@@ -1390,6 +1500,16 @@ static int snd_ac97_ad1888_downmix_put(snd_kcontrol_t *kcontrol, snd_ctl_elem_va
 				    AC97_AD198X_DMIX0 | AC97_AD198X_DMIX1, val);
 }
 
+static void ad1888_update_jacks(ac97_t *ac97)
+{
+	/* shared Line-In */
+	snd_ac97_update_bits(ac97, AC97_AD_MISC, 1 << 12,
+			     is_shared_linein(ac97) ? 0 : 1 << 12);
+	/* shared Mic */
+	snd_ac97_update_bits(ac97, AC97_AD_MISC, 1 << 11,
+			     is_shared_micin(ac97) ? 0 : 1 << 13);
+}
+
 static const snd_kcontrol_new_t snd_ac97_ad1888_controls[] = {
 	{
 		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
@@ -1406,8 +1526,13 @@ static const snd_kcontrol_new_t snd_ac97_ad1888_controls[] = {
 		.get = snd_ac97_ad1888_downmix_get,
 		.put = snd_ac97_ad1888_downmix_put
 	},
+#if 0
 	AC97_SINGLE("Surround Jack as Input", AC97_AD_MISC, 12, 1, 0),
 	AC97_SINGLE("Center/LFE Jack as Input", AC97_AD_MISC, 11, 1, 0),
+#else
+	AC97_SURROUND_JACK_MODE_CTL,
+	AC97_CHANNEL_MODE_CTL,
+#endif
 };
 
 static int patch_ad1888_specific(ac97_t *ac97)
@@ -1422,8 +1547,9 @@ static struct snd_ac97_build_ops patch_ad1888_build_ops = {
 	.build_post_spdif = patch_ad198x_post_spdif,
 	.build_specific = patch_ad1888_specific,
 #ifdef CONFIG_PM
-	.resume = ad18xx_resume
+	.resume = ad18xx_resume,
 #endif
+	.update_jacks = ad1888_update_jacks,
 };
 
 int patch_ad1888(ac97_t * ac97)
@@ -1521,31 +1647,25 @@ int patch_ad1985(ac97_t * ac97)
 /*
  * realtek ALC65x/850 codecs
  */
-static int snd_ac97_alc650_mic_get(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t * ucontrol)
-{
-        ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
-        ucontrol->value.integer.value[0] = (ac97->regs[AC97_ALC650_MULTICH] >> 10) & 1;
-        return 0;
-}
-
-static int snd_ac97_alc650_mic_put(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t * ucontrol)
+static void alc650_update_jacks(ac97_t *ac97)
 {
-        ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
-	int change, val;
-	val = !!(snd_ac97_read(ac97, AC97_ALC650_MULTICH) & (1 << 10));
-	change = (ucontrol->value.integer.value[0] != val);
-	if (change) {
-		/* disable/enable vref */
-		snd_ac97_update_bits(ac97, AC97_ALC650_CLOCK, 1 << 12,
-				     ucontrol->value.integer.value[0] ? (1 << 12) : 0);
-		/* turn on/off center-on-mic */
-		snd_ac97_update_bits(ac97, AC97_ALC650_MULTICH, 1 << 10,
-				     ucontrol->value.integer.value[0] ? (1 << 10) : 0);
-		/* GPIO0 high for mic */
-		snd_ac97_update_bits(ac97, AC97_ALC650_GPIO_STATUS, 0x100,
-				     ucontrol->value.integer.value[0] ? 0 : 0x100);
-        }
-        return change;
+	int shared;
+	
+	/* shared Line-In */
+	shared = is_shared_linein(ac97);
+	snd_ac97_update_bits(ac97, AC97_ALC650_MULTICH, 1 << 9,
+			     shared ? (1 << 9) : 0);
+	/* update shared Mic */
+	shared = is_shared_micin(ac97);
+	/* disable/enable vref */
+	snd_ac97_update_bits(ac97, AC97_ALC650_CLOCK, 1 << 12,
+			     shared ? (1 << 12) : 0);
+	/* turn on/off center-on-mic */
+	snd_ac97_update_bits(ac97, AC97_ALC650_MULTICH, 1 << 10,
+			     shared ? (1 << 10) : 0);
+	/* GPIO0 high for mic */
+	snd_ac97_update_bits(ac97, AC97_ALC650_GPIO_STATUS, 0x100,
+			     shared ? 0 : 0x100);
 }
 
 static const snd_kcontrol_new_t snd_ac97_controls_alc650[] = {
@@ -1558,8 +1678,8 @@ static const snd_kcontrol_new_t snd_ac97_controls_alc650[] = {
 	/* 6: Independent Master Volume Right */
 	/* 7: Independent Master Volume Left */
 	/* 8: reserved */
-	AC97_SINGLE("Line-In As Surround", AC97_ALC650_MULTICH, 9, 1, 0),
-	/* 10: mic, see below */
+	/* 9: Line-In/Surround share */
+	/* 10: Mic/CLFE share */
 	/* 11-13: in IEC958 controls */
 	AC97_SINGLE("Swap Surround Slot", AC97_ALC650_MULTICH, 14, 1, 0),
 #if 0 /* always set in patch_alc650 */
@@ -1570,14 +1690,8 @@ static const snd_kcontrol_new_t snd_ac97_controls_alc650[] = {
 	AC97_SINGLE("Center/LFE DAC Switch", AC97_ALC650_LFE_DAC_VOL, 15, 1, 1),
 	AC97_DOUBLE("Center/LFE DAC Volume", AC97_ALC650_LFE_DAC_VOL, 8, 0, 31, 1),
 #endif
-	{
-		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
-		.name = "Mic As Center/LFE",
-		.info = snd_ac97_info_volsw,
-		.get = snd_ac97_alc650_mic_get,
-		.put = snd_ac97_alc650_mic_put,
-		.private_value = AC97_SINGLE_VALUE(0, 0, 1, 0) /* only mask needed */
-	},
+	AC97_SURROUND_JACK_MODE_CTL,
+	AC97_CHANNEL_MODE_CTL,
 };
 
 static const snd_kcontrol_new_t snd_ac97_spdif_controls_alc650[] = {
@@ -1601,7 +1715,8 @@ static int patch_alc650_specific(ac97_t * ac97)
 }
 
 static struct snd_ac97_build_ops patch_alc650_ops = {
-	.build_specific	= patch_alc650_specific
+	.build_specific	= patch_alc650_specific,
+	.update_jacks = alc650_update_jacks
 };
 
 int patch_alc650(ac97_t * ac97)
@@ -1659,37 +1774,27 @@ int patch_alc650(ac97_t * ac97)
 	return 0;
 }
 
-static int snd_ac97_alc655_mic_get(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t * ucontrol)
-{
-        ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
-        ucontrol->value.integer.value[0] = (ac97->regs[AC97_ALC650_MULTICH] >> 10) & 1;
-        return 0;
-}
-
-static int snd_ac97_alc655_mic_put(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t * ucontrol)
+static void alc655_update_jacks(ac97_t *ac97)
 {
-        ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
-
+	int shared;
+	
+	/* shared Line-In */
+	shared = is_shared_linein(ac97);
+	ac97_update_bits_page(ac97, AC97_ALC650_MULTICH, 1 << 9,
+			      shared ? (1 << 9) : 0, 0);
+	/* update shared mic */
+	shared = is_shared_micin(ac97);
 	/* misc control; vrefout disable */
 	snd_ac97_update_bits(ac97, AC97_ALC650_CLOCK, 1 << 12,
-			     ucontrol->value.integer.value[0] ? (1 << 12) : 0);
-	return ac97_update_bits_page(ac97, AC97_ALC650_MULTICH, 1 << 10,
-				     ucontrol->value.integer.value[0] ? (1 << 10) : 0,
-				     0);
+			     shared ? (1 << 12) : 0);
+	ac97_update_bits_page(ac97, AC97_ALC650_MULTICH, 1 << 10,
+			      shared ? (1 << 10) : 0, 0);
 }
 
-
 static const snd_kcontrol_new_t snd_ac97_controls_alc655[] = {
 	AC97_PAGE_SINGLE("Duplicate Front", AC97_ALC650_MULTICH, 0, 1, 0, 0),
-	AC97_PAGE_SINGLE("Line-In As Surround", AC97_ALC650_MULTICH, 9, 1, 0, 0),
-	{
-		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
-		.name = "Mic As Center/LFE",
-		.info = snd_ac97_info_volsw,
-		.get = snd_ac97_alc655_mic_get,
-		.put = snd_ac97_alc655_mic_put,
-		.private_value = AC97_SINGLE_VALUE(0, 0, 1, 0) /* only mask needed */
-	},
+	AC97_SURROUND_JACK_MODE_CTL,
+	AC97_CHANNEL_MODE_CTL,
 };
 
 static int alc655_iec958_route_info(snd_kcontrol_t *kcontrol, snd_ctl_elem_info_t *uinfo)
@@ -1759,7 +1864,8 @@ static int patch_alc655_specific(ac97_t * ac97)
 }
 
 static struct snd_ac97_build_ops patch_alc655_ops = {
-	.build_specific	= patch_alc655_specific
+	.build_specific	= patch_alc655_specific,
+	.update_jacks = alc655_update_jacks
 };
 
 int patch_alc655(ac97_t * ac97)
@@ -1798,63 +1904,32 @@ int patch_alc655(ac97_t * ac97)
 #define AC97_ALC850_JACK_SELECT	0x76
 #define AC97_ALC850_MISC1	0x7a
 
-static int ac97_alc850_surround_get(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t * ucontrol)
+static void alc850_update_jacks(ac97_t *ac97)
 {
-        ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
-        ucontrol->value.integer.value[0] = ((ac97->regs[AC97_ALC850_JACK_SELECT] >> 12) & 7) == 2;
-        return 0;
-}
-
-static int ac97_alc850_surround_put(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t * ucontrol)
-{
-        ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
-
+	int shared;
+	
+	/* shared Line-In */
+	shared = is_shared_linein(ac97);
 	/* SURR 1kOhm (bit4), Amp (bit5) */
 	snd_ac97_update_bits(ac97, AC97_ALC850_MISC1, (1<<4)|(1<<5),
-			     ucontrol->value.integer.value[0] ? (1<<5) : (1<<4));
+			     shared ? (1<<5) : (1<<4));
 	/* LINE-IN = 0, SURROUND = 2 */
-	return snd_ac97_update_bits(ac97, AC97_ALC850_JACK_SELECT, 7 << 12,
-				    ucontrol->value.integer.value[0] ? (2<<12) : (0<<12));
-}
-
-static int ac97_alc850_mic_get(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t * ucontrol)
-{
-        ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
-        ucontrol->value.integer.value[0] = ((ac97->regs[AC97_ALC850_JACK_SELECT] >> 4) & 7) == 2;
-        return 0;
-}
-
-static int ac97_alc850_mic_put(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t * ucontrol)
-{
-        ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
-
+	snd_ac97_update_bits(ac97, AC97_ALC850_JACK_SELECT, 7 << 12,
+			     shared ? (2<<12) : (0<<12));
+	/* update shared mic */
+	shared = is_shared_micin(ac97);
 	/* Vref disable (bit12), 1kOhm (bit13) */
 	snd_ac97_update_bits(ac97, AC97_ALC850_MISC1, (1<<12)|(1<<13),
-			     ucontrol->value.integer.value[0] ? (1<<12) : (1<<13));
+			     shared ? (1<<12) : (1<<13));
 	/* MIC-IN = 1, CENTER-LFE = 2 */
-	return snd_ac97_update_bits(ac97, AC97_ALC850_JACK_SELECT, 7 << 4,
-				    ucontrol->value.integer.value[0] ? (2<<4) : (1<<4));
+	snd_ac97_update_bits(ac97, AC97_ALC850_JACK_SELECT, 7 << 4,
+			     shared ? (2<<4) : (1<<4));
 }
 
 static const snd_kcontrol_new_t snd_ac97_controls_alc850[] = {
 	AC97_PAGE_SINGLE("Duplicate Front", AC97_ALC650_MULTICH, 0, 1, 0, 0),
-	{
-		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
-		.name = "Line-In As Surround",
-		.info = snd_ac97_info_volsw,
-		.get = ac97_alc850_surround_get,
-		.put = ac97_alc850_surround_put,
-		.private_value = AC97_SINGLE_VALUE(0, 0, 1, 0) /* only mask needed */
-	},
-	{
-		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
-		.name = "Mic As Center/LFE",
-		.info = snd_ac97_info_volsw,
-		.get = ac97_alc850_mic_get,
-		.put = ac97_alc850_mic_put,
-		.private_value = AC97_SINGLE_VALUE(0, 0, 1, 0) /* only mask needed */
-	},
-
+	AC97_SURROUND_JACK_MODE_CTL,
+	AC97_CHANNEL_MODE_CTL,
 };
 
 static int patch_alc850_specific(ac97_t *ac97)
@@ -1871,7 +1946,8 @@ static int patch_alc850_specific(ac97_t *ac97)
 }
 
 static struct snd_ac97_build_ops patch_alc850_ops = {
-	.build_specific	= patch_alc850_specific
+	.build_specific	= patch_alc850_specific,
+	.update_jacks = alc850_update_jacks
 };
 
 int patch_alc850(ac97_t *ac97)
@@ -1911,9 +1987,17 @@ int patch_alc850(ac97_t *ac97)
 /*
  * C-Media CM97xx codecs
  */
+static void cm9738_update_jacks(ac97_t *ac97)
+{
+	/* shared Line-In */
+	snd_ac97_update_bits(ac97, AC97_CM9738_VENDOR_CTRL, 1 << 10,
+			     is_shared_linein(ac97) ? (1 << 10) : 0);
+}
+
 static const snd_kcontrol_new_t snd_ac97_cm9738_controls[] = {
-	AC97_SINGLE("Line-In As Surround", AC97_CM9738_VENDOR_CTRL, 10, 1, 0),
 	AC97_SINGLE("Duplicate Front", AC97_CM9738_VENDOR_CTRL, 13, 1, 0),
+	AC97_SURROUND_JACK_MODE_CTL,
+	AC97_CHANNEL_MODE_4CH_CTL,
 };
 
 static int patch_cm9738_specific(ac97_t * ac97)
@@ -1922,7 +2006,8 @@ static int patch_cm9738_specific(ac97_t * ac97)
 }
 
 static struct snd_ac97_build_ops patch_cm9738_ops = {
-	.build_specific	= patch_cm9738_specific
+	.build_specific	= patch_cm9738_specific,
+	.update_jacks = cm9738_update_jacks
 };
 
 int patch_cm9738(ac97_t * ac97)
@@ -1986,34 +2071,19 @@ static const snd_kcontrol_new_t snd_ac97_cm9739_controls_spdif[] = {
 	/* BIT 8: SPD32 - 32bit SPDIF - not supported yet */
 };
 
-static int snd_ac97_cm9739_center_mic_get(snd_kcontrol_t * kcontrol, snd_ctl_elem_value_t * ucontrol)
-{
-	ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
-	if (ac97->regs[AC97_CM9739_MULTI_CHAN] & 0x1000)
-		ucontrol->value.integer.value[0] = 1;
-	else
-		ucontrol->value.integer.value[0] = 0;
-	return 0;
-}
-
-static int snd_ac97_cm9739_center_mic_put(snd_kcontrol_t * kcontrol, snd_ctl_elem_value_t * ucontrol)
+static void cm9739_update_jacks(ac97_t *ac97)
 {
-	ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
-	return snd_ac97_update_bits(ac97, AC97_CM9739_MULTI_CHAN, 0x3000,
-				    ucontrol->value.integer.value[0] ? 
-				    0x1000 : 0x2000);
+	/* shared Line-In */
+	snd_ac97_update_bits(ac97, AC97_CM9739_MULTI_CHAN, 1 << 10,
+			     is_shared_linein(ac97) ? (1 << 10) : 0);
+	/* shared Mic */
+	snd_ac97_update_bits(ac97, AC97_CM9739_MULTI_CHAN, 0x3000,
+			     is_shared_micin(ac97) ? 0x1000 : 0x2000);
 }
 
 static const snd_kcontrol_new_t snd_ac97_cm9739_controls[] = {
-	AC97_SINGLE("Line-In As Surround", AC97_CM9739_MULTI_CHAN, 10, 1, 0),
-	{
-		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
-		.name = "Mic As Center/LFE",
-		.info = snd_ac97_info_volsw,
-		.get = snd_ac97_cm9739_center_mic_get,
-		.put = snd_ac97_cm9739_center_mic_put,
-		.private_value = AC97_SINGLE_VALUE(0, 0, 1, 0) /* only mask needed */
-	},
+	AC97_SURROUND_JACK_MODE_CTL,
+	AC97_CHANNEL_MODE_CTL,
 };
 
 static int patch_cm9739_specific(ac97_t * ac97)
@@ -2028,7 +2098,8 @@ static int patch_cm9739_post_spdif(ac97_t * ac97)
 
 static struct snd_ac97_build_ops patch_cm9739_ops = {
 	.build_specific	= patch_cm9739_specific,
-	.build_post_spdif = patch_cm9739_post_spdif
+	.build_post_spdif = patch_cm9739_post_spdif,
+	.update_jacks = cm9739_update_jacks
 };
 
 int patch_cm9739(ac97_t * ac97)
@@ -2090,67 +2161,28 @@ int patch_cm9739(ac97_t * ac97)
 #define AC97_CM9761_FUNC	0x66
 #define AC97_CM9761_SPDIF_CTRL	0x6c
 
-static int snd_ac97_cm9761_linein_rear_get(snd_kcontrol_t * kcontrol, snd_ctl_elem_value_t * ucontrol)
+static void cm9761_update_jacks(ac97_t *ac97)
 {
-	ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
-	if (ac97->regs[AC97_CM9761_MULTI_CHAN] & 0x0400)
-		ucontrol->value.integer.value[0] = 1;
-	else
-		ucontrol->value.integer.value[0] = 0;
-	return 0;
-}
-
-static int snd_ac97_cm9761_linein_rear_put(snd_kcontrol_t * kcontrol, snd_ctl_elem_value_t * ucontrol)
-{
-	ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
-	unsigned short vals[2][2] = {
+	unsigned short surr_vals[2][2] = {
 		{ 0x0008, 0x0400 }, /* off, on */
 		{ 0x0000, 0x0408 }, /* off, on (9761-82 rev.B) */
 	};
-	return snd_ac97_update_bits(ac97, AC97_CM9761_MULTI_CHAN, 0x0408,
-				    vals[ac97->spec.dev_flags][!!ucontrol->value.integer.value[0]]);
-}
-
-static int snd_ac97_cm9761_center_mic_get(snd_kcontrol_t * kcontrol, snd_ctl_elem_value_t * ucontrol)
-{
-	ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
-	if (ac97->regs[AC97_CM9761_MULTI_CHAN] & 0x1000)
-		ucontrol->value.integer.value[0] = 1;
-	else
-		ucontrol->value.integer.value[0] = 0;
-	if (ac97->spec.dev_flags) /* 9761-82 rev.B */
-		ucontrol->value.integer.value[0] = !ucontrol->value.integer.value[0];
-	return 0;
-}
-
-static int snd_ac97_cm9761_center_mic_put(snd_kcontrol_t * kcontrol, snd_ctl_elem_value_t * ucontrol)
-{
-	ac97_t *ac97 = snd_kcontrol_chip(kcontrol);
-	unsigned short vals[2][2] = {
+	unsigned short clfe_vals[2][2] = {
 		{ 0x2000, 0x1880 }, /* off, on */
 		{ 0x1000, 0x2880 }, /* off, on (9761-82 rev.B) */
 	};
-	return snd_ac97_update_bits(ac97, AC97_CM9761_MULTI_CHAN, 0x3880,
-				    vals[ac97->spec.dev_flags][!!ucontrol->value.integer.value[0]]);
+
+	/* shared Line-In */
+	snd_ac97_update_bits(ac97, AC97_CM9761_MULTI_CHAN, 0x0408,
+			     surr_vals[ac97->spec.dev_flags][is_shared_linein(ac97)]);
+	/* shared Mic */
+	snd_ac97_update_bits(ac97, AC97_CM9761_MULTI_CHAN, 0x3880,
+			     clfe_vals[ac97->spec.dev_flags][is_shared_micin(ac97)]);
 }
 
 static const snd_kcontrol_new_t snd_ac97_cm9761_controls[] = {
-	{
-		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
-		.name = "Line-In As Surround",
-		.info = snd_ac97_info_volsw,
-		.get = snd_ac97_cm9761_linein_rear_get,
-		.put = snd_ac97_cm9761_linein_rear_put,
-		.private_value = AC97_SINGLE_VALUE(0, 0, 1, 0) /* only mask needed */
-	},
-	{
-		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
-		.name = "Mic As Center/LFE",
-		.info = snd_ac97_info_volsw,
-		.get = snd_ac97_cm9761_center_mic_get,
-		.put = snd_ac97_cm9761_center_mic_put,
-		.private_value = AC97_SINGLE_VALUE(0, 0, 1, 0) /* only mask needed */
-	},
+	AC97_SURROUND_JACK_MODE_CTL,
+	AC97_CHANNEL_MODE_CTL,
 };
 
 static int cm9761_spdif_out_source_info(snd_kcontrol_t *kcontrol, snd_ctl_elem_info_t *uinfo)
@@ -2224,7 +2256,8 @@ static int patch_cm9761_specific(ac97_t * ac97)
 
 static struct snd_ac97_build_ops patch_cm9761_ops = {
 	.build_specific	= patch_cm9761_specific,
-	.build_post_spdif = patch_cm9761_post_spdif
+	.build_post_spdif = patch_cm9761_post_spdif,
+	.update_jacks = cm9761_update_jacks
 };
 
 int patch_cm9761(ac97_t *ac97)
@@ -2370,9 +2403,21 @@ int patch_vt1616(ac97_t * ac97)
 	return 0;
 }
 
+/*
+ */
+static void it2646_update_jacks(ac97_t *ac97)
+{
+	/* shared Line-In */
+	snd_ac97_update_bits(ac97, 0x76, 1 << 9,
+			     is_shared_linein(ac97) ? (1<<9) : 0);
+	/* shared Mic */
+	snd_ac97_update_bits(ac97, 0x76, 1 << 10,
+			     is_shared_micin(ac97) ? (1<<10) : 0);
+}
+
 static const snd_kcontrol_new_t snd_ac97_controls_it2646[] = {
-	AC97_SINGLE("Line-In As Surround", 0x76, 9, 1, 0),
-	AC97_SINGLE("Mic As Center/LFE", 0x76, 10, 1, 0),
+	AC97_SURROUND_JACK_MODE_CTL,
+	AC97_CHANNEL_MODE_CTL,
 };
 
 static const snd_kcontrol_new_t snd_ac97_spdif_controls_it2646[] = {
@@ -2392,7 +2437,8 @@ static int patch_it2646_specific(ac97_t * ac97)
 }
 
 static struct snd_ac97_build_ops patch_it2646_ops = {
-	.build_specific	= patch_it2646_specific
+	.build_specific	= patch_it2646_specific,
+	.update_jacks = it2646_update_jacks
 };
 
 int patch_it2646(ac97_t * ac97)
-- 
cgit v1.2.3-70-g09d2


From 07cf374169699d78721668b4e4bd02097c971f75 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Thu, 14 Apr 2005 16:21:03 +0200
Subject: [ALSA] Increase timer protocol number

ALSA Core
Increase the timer protocl number (to distinguish the fix for TREAD ioctls).

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 include/sound/asound.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/sound/asound.h b/include/sound/asound.h
index 716227eed3e..26db585a181 100644
--- a/include/sound/asound.h
+++ b/include/sound/asound.h
@@ -559,7 +559,7 @@ enum {
  *  Timer section - /dev/snd/timer
  */
 
-#define SNDRV_TIMER_VERSION		SNDRV_PROTOCOL_VERSION(2, 0, 2)
+#define SNDRV_TIMER_VERSION		SNDRV_PROTOCOL_VERSION(2, 0, 3)
 
 enum sndrv_timer_class {
 	SNDRV_TIMER_CLASS_NONE = -1,
-- 
cgit v1.2.3-70-g09d2


From ade2916109dc53350298f1ccfb8ab03432c590b4 Mon Sep 17 00:00:00 2001
From: Jaroslav Kysela <perex@suse.cz>
Date: Wed, 27 Apr 2005 16:09:21 +0200
Subject: ALSA CVS update ALSA Version 1.0.9rc3

Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/version.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/sound/version.h b/include/sound/version.h
index 98b4230778e..f5959de329c 100644
--- a/include/sound/version.h
+++ b/include/sound/version.h
@@ -1,3 +1,3 @@
 /* include/version.h.  Generated by configure.  */
-#define CONFIG_SND_VERSION "1.0.9rc2"
+#define CONFIG_SND_VERSION "1.0.9rc3"
 #define CONFIG_SND_DATE "  (Thu Mar 24 10:33:39 2005 UTC)"
-- 
cgit v1.2.3-70-g09d2


From b259b10c420a59a2fdbcf5a3498253ebcbdffa1e Mon Sep 17 00:00:00 2001
From: Clemens Ladisch <clemens@ladisch.de>
Date: Fri, 29 Apr 2005 16:29:28 +0200
Subject: [ALSA] usb-audio - add Extigy/Audigy 2 NX remote control support

ALSA Core,USB generic driver
Add an hwdep interface that supports reading remote control data from
Sound Blaster Extigy and Audigy 2 NX devices.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
---
 include/sound/asound.h |   3 +-
 sound/usb/usbmixer.c   | 159 +++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 161 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/sound/asound.h b/include/sound/asound.h
index 26db585a181..4321e92a7f8 100644
--- a/include/sound/asound.h
+++ b/include/sound/asound.h
@@ -113,9 +113,10 @@ enum sndrv_hwdep_iface {
 	SNDRV_HWDEP_IFACE_BLUETOOTH,	/* Bluetooth audio */
 	SNDRV_HWDEP_IFACE_USX2Y_PCM,	/* Tascam US122, US224 & US428 rawusb pcm */
 	SNDRV_HWDEP_IFACE_PCXHR,	/* Digigram PCXHR */
+	SNDRV_HWDEP_IFACE_SB_RC,	/* SB Extigy/Audigy2NX remote control */
 
 	/* Don't forget to change the following: */
-	SNDRV_HWDEP_IFACE_LAST = SNDRV_HWDEP_IFACE_PCXHR
+	SNDRV_HWDEP_IFACE_LAST = SNDRV_HWDEP_IFACE_SB_RC
 };
 
 struct sndrv_hwdep_info {
diff --git a/sound/usb/usbmixer.c b/sound/usb/usbmixer.c
index dd045ea6fb0..7ea42d43d7f 100644
--- a/sound/usb/usbmixer.c
+++ b/sound/usb/usbmixer.c
@@ -35,6 +35,7 @@
 #include <linux/usb.h>
 #include <sound/core.h>
 #include <sound/control.h>
+#include <sound/hwdep.h>
 
 #include "usbaudio.h"
 
@@ -57,6 +58,19 @@ struct usb_mixer_interface {
 	unsigned int ignore_ctl_error;
 	struct urb *urb;
 	usb_mixer_elem_info_t **id_elems; /* array[256], indexed by unit id */
+
+	/* Sound Blaster remote control stuff */
+	enum {
+		RC_NONE,
+		RC_EXTIGY,
+		RC_AUDIGY2NX,
+	} rc_type;
+	unsigned long rc_hwdep_open;
+	u32 rc_code;
+	wait_queue_head_t rc_waitq;
+	struct urb *rc_urb;
+	struct usb_ctrlrequest *rc_setup_packet;
+	u8 rc_buffer[6];
 };
 
 
@@ -1536,6 +1550,9 @@ static void snd_usb_mixer_free(struct usb_mixer_interface *mixer)
 		kfree(mixer->urb->transfer_buffer);
 		usb_free_urb(mixer->urb);
 	}
+	if (mixer->rc_urb)
+		usb_free_urb(mixer->rc_urb);
+	kfree(mixer->rc_setup_packet);
 	kfree(mixer);
 }
 
@@ -1604,6 +1621,17 @@ static void snd_usb_mixer_notify_id(struct usb_mixer_interface *mixer,
 			       info->elem_id);
 }
 
+static void snd_usb_mixer_memory_change(struct usb_mixer_interface *mixer,
+					int unitid)
+{
+	/* SB remote control */
+	if (mixer->rc_type != RC_NONE && unitid == 0) {
+		/* read control code from device memory */
+		mixer->rc_urb->dev = mixer->chip->dev;
+		usb_submit_urb(mixer->rc_urb, GFP_ATOMIC);
+	}
+}
+
 static void snd_usb_mixer_status_complete(struct urb *urb, struct pt_regs *regs)
 {
 	struct usb_mixer_interface *mixer = urb->context;
@@ -1620,6 +1648,8 @@ static void snd_usb_mixer_status_complete(struct urb *urb, struct pt_regs *regs)
 				continue;
 			if (!(buf[0] & 0x40))
 				snd_usb_mixer_notify_id(mixer, buf[1]);
+			else
+				snd_usb_mixer_memory_change(mixer, buf[1]);
 		}
 	}
 	if (urb->status != -ENOENT && urb->status != -ECONNRESET) {
@@ -1664,6 +1694,126 @@ static int snd_usb_mixer_status_create(struct usb_mixer_interface *mixer)
 	return 0;
 }
 
+static void snd_usb_soundblaster_remote_complete(struct urb *urb,
+						 struct pt_regs *regs)
+{
+	struct usb_mixer_interface *mixer = urb->context;
+	/*
+	 * format of remote control data:
+	 * Extigy:	xx 00
+	 * Audigy 2 NX:	06 80 xx 00 00 00
+	 */
+	int offset = mixer->rc_type == RC_EXTIGY ? 0 : 2;
+	u32 code;
+
+	if (urb->status < 0 || urb->actual_length <= offset)
+		return;
+	code = mixer->rc_buffer[offset];
+	/* the Mute button actually changes the mixer control */
+	if (code == 13)
+		snd_usb_mixer_notify_id(mixer, 18);
+	mixer->rc_code = code;
+	wmb();
+	wake_up(&mixer->rc_waitq);
+}
+
+static int snd_usb_sbrc_hwdep_open(snd_hwdep_t *hw, struct file *file)
+{
+	struct usb_mixer_interface *mixer = hw->private_data;
+
+	if (test_and_set_bit(0, &mixer->rc_hwdep_open))
+		return -EBUSY;
+	return 0;
+}
+
+static int snd_usb_sbrc_hwdep_release(snd_hwdep_t *hw, struct file *file)
+{
+	struct usb_mixer_interface *mixer = hw->private_data;
+
+	clear_bit(0, &mixer->rc_hwdep_open);
+	smp_mb__after_clear_bit();
+	return 0;
+}
+
+static long snd_usb_sbrc_hwdep_read(snd_hwdep_t *hw, char __user *buf,
+				     long count, loff_t *offset)
+{
+	struct usb_mixer_interface *mixer = hw->private_data;
+	int err;
+	u32 rc_code;
+
+	if (count != 1)
+		return -EINVAL;
+	err = wait_event_interruptible(mixer->rc_waitq,
+				       (rc_code = xchg(&mixer->rc_code, 0)) != 0);
+	if (err == 0) {
+		err = put_user(rc_code, buf);
+	}
+	return err < 0 ? err : count;
+}
+
+static unsigned int snd_usb_sbrc_hwdep_poll(snd_hwdep_t *hw, struct file *file,
+					    poll_table *wait)
+{
+	struct usb_mixer_interface *mixer = hw->private_data;
+
+	poll_wait(file, &mixer->rc_waitq, wait);
+	return mixer->rc_code ? POLLIN | POLLRDNORM : 0;
+}
+
+static int snd_usb_soundblaster_remote_init(struct usb_mixer_interface *mixer)
+{
+	snd_hwdep_t *hwdep;
+	int err, len;
+
+	switch (le16_to_cpu(mixer->chip->dev->descriptor.idProduct)) {
+	case 0x3000:
+		mixer->rc_type = RC_EXTIGY;
+		len = 2;
+		break;
+	case 0x3020:
+		mixer->rc_type = RC_AUDIGY2NX;
+		len = 6;
+		break;
+	default:
+		return 0;
+	}
+
+	init_waitqueue_head(&mixer->rc_waitq);
+	err = snd_hwdep_new(mixer->chip->card, "SB remote control", 0, &hwdep);
+	if (err < 0)
+		return err;
+	snprintf(hwdep->name, sizeof(hwdep->name),
+		 "%s remote control", mixer->chip->card->shortname);
+	hwdep->iface = SNDRV_HWDEP_IFACE_SB_RC;
+	hwdep->private_data = mixer;
+	hwdep->ops.read = snd_usb_sbrc_hwdep_read;
+	hwdep->ops.open = snd_usb_sbrc_hwdep_open;
+	hwdep->ops.release = snd_usb_sbrc_hwdep_release;
+	hwdep->ops.poll = snd_usb_sbrc_hwdep_poll;
+
+	mixer->rc_urb = usb_alloc_urb(0, GFP_KERNEL);
+	if (!mixer->rc_urb)
+		return -ENOMEM;
+	mixer->rc_setup_packet = kmalloc(sizeof(*mixer->rc_setup_packet), GFP_KERNEL);
+	if (!mixer->rc_setup_packet) {
+		usb_free_urb(mixer->rc_urb);
+		mixer->rc_urb = NULL;
+		return -ENOMEM;
+	}
+	mixer->rc_setup_packet->bRequestType =
+		USB_DIR_IN | USB_TYPE_CLASS | USB_RECIP_INTERFACE;
+	mixer->rc_setup_packet->bRequest = GET_MEM;
+	mixer->rc_setup_packet->wValue = cpu_to_le16(0);
+	mixer->rc_setup_packet->wIndex = cpu_to_le16(0);
+	mixer->rc_setup_packet->wLength = cpu_to_le16(len);
+	usb_fill_control_urb(mixer->rc_urb, mixer->chip->dev,
+			     usb_rcvctrlpipe(mixer->chip->dev, 0),
+			     (u8*)mixer->rc_setup_packet, mixer->rc_buffer, len,
+			     snd_usb_soundblaster_remote_complete, mixer);
+	return 0;
+}
+
 int snd_usb_create_mixer(snd_usb_audio_t *chip, int ctrlif)
 {
 	static snd_device_ops_t dev_ops = {
@@ -1694,6 +1844,13 @@ int snd_usb_create_mixer(snd_usb_audio_t *chip, int ctrlif)
 		return err;
 	}
 
+	if (le16_to_cpu(chip->dev->descriptor.idVendor) == 0x041e) {
+		if ((err = snd_usb_soundblaster_remote_init(mixer)) < 0) {
+			snd_usb_mixer_free(mixer);
+			return err;
+		}
+	}
+
 	err = snd_device_new(chip->card, SNDRV_DEV_LOWLEVEL, mixer, &dev_ops);
 	if (err < 0) {
 		snd_usb_mixer_free(mixer);
@@ -1710,4 +1867,6 @@ void snd_usb_mixer_disconnect(struct list_head *p)
 	mixer = list_entry(p, struct usb_mixer_interface, list);
 	if (mixer->urb)
 		usb_kill_urb(mixer->urb);
+	if (mixer->rc_urb)
+		usb_kill_urb(mixer->rc_urb);
 }
-- 
cgit v1.2.3-70-g09d2


From 14c7e472aa979eecc15255eec5cec2763649c599 Mon Sep 17 00:00:00 2001
From: James Courtier-Dutton <James@superbug.co.uk>
Date: Wed, 4 May 2005 16:53:53 +0200
Subject: [ALSA] Update A_SAMPLE_RATE register details.

EMU10K1/EMU10K2 driver


Signed-off-by: James Courtier-Dutton <James@superbug.co.uk>
---
 include/sound/emu10k1.h | 20 ++++++++++++++++++--
 1 file changed, 18 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/sound/emu10k1.h b/include/sound/emu10k1.h
index 61a3f418f30..23dabbceb4b 100644
--- a/include/sound/emu10k1.h
+++ b/include/sound/emu10k1.h
@@ -805,10 +805,26 @@
 #define A_FXWC2			0x75		/* Selects 0x9f-0x80 for FX recording           */
 
 #define A_SPDIF_SAMPLERATE	0x76		/* Set the sample rate of SPDIF output		*/
-#define A_SPDIF_RATE_MASK	0x000000c0
+#define A_SAMPLE_RATE		0x76		/* Various sample rate settings. */
+#define A_SAMPLE_RATE_NOT_USED  0x0ffc111e	/* Bits that are not used and cannot be set. 	*/
+#define A_SAMPLE_RATE_UNKNOWN	0xf0030001	/* Bits that can be set, but have unknown use. 	*/
+#define A_SPDIF_RATE_MASK	0x000000e0	/* Any other values for rates, just use 48000	*/
 #define A_SPDIF_48000		0x00000000
-#define A_SPDIF_44100		0x00000080
+#define A_SPDIF_192000		0x00000020
 #define A_SPDIF_96000		0x00000040
+#define A_SPDIF_44100		0x00000080
+
+#define A_I2S_CAPTURE_RATE_MASK	0x00000e00	/* This sets the capture PCM rate, but it is    */
+#define A_I2S_CAPTURE_48000	0x00000000	/* unclear if this sets the ADC rate as well.	*/
+#define A_I2S_CAPTURE_192000	0x00000200
+#define A_I2S_CAPTURE_96000	0x00000400
+#define A_I2S_CAPTURE_44100	0x00000800
+
+#define A_PCM_RATE_MASK		0x0000e000	/* This sets the playback PCM rate on the P16V	*/
+#define A_PCM_48000		0x00000000
+#define A_PCM_192000		0x00002000
+#define A_PCM_96000		0x00004000
+#define A_PCM_44100		0x00008000
 
 /* 0x77,0x78,0x79 "something i2s-related" - default to 0x01080000 on my audigy 2 ZS --rlrevell	*/
 /* 0x7a, 0x7b - lookup tables */
-- 
cgit v1.2.3-70-g09d2


From f927c8fc648420ad8edd7e4699b4ba510c2e9c6b Mon Sep 17 00:00:00 2001
From: James Courtier-Dutton <James@superbug.co.uk>
Date: Sat, 7 May 2005 15:34:13 +0200
Subject: [ALSA] Implement different capture sources.

EMU10K1/EMU10K2 driver
e.g. When HD Capture source is set to SPDIF,
     setting HD Capture channel to 0 captures from CDROM digital input.
     setting HD Capture channel to 1 captures from SPDIF in.

Signed-off-by: James Courtier-Dutton <James@superbug.co.uk>
---
 include/sound/emu10k1.h  |  1 +
 sound/pci/emu10k1/p16v.c | 62 +++++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 62 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/sound/emu10k1.h b/include/sound/emu10k1.h
index 23dabbceb4b..c50b91958ff 100644
--- a/include/sound/emu10k1.h
+++ b/include/sound/emu10k1.h
@@ -1130,6 +1130,7 @@ struct _snd_emu10k1 {
 	emu10k1_voice_t p16v_capture_voice;
 	int p16v_device_offset;
 	u32 p16v_capture_source;
+	u32 p16v_capture_channel;
 	emu10k1_pcm_mixer_t pcm_mixer[32];
 	emu10k1_pcm_mixer_t efx_pcm_mixer[NUM_EFX_PLAYBACK];
 	snd_kcontrol_t *ctl_send_routing;
diff --git a/sound/pci/emu10k1/p16v.c b/sound/pci/emu10k1/p16v.c
index 13f7e62ee56..93dff4c6b23 100644
--- a/sound/pci/emu10k1/p16v.c
+++ b/sound/pci/emu10k1/p16v.c
@@ -41,7 +41,13 @@
  *    Integrated with snd-emu10k1 driver.
  *  0.22
  *    Removed #if 0 ... #endif
- *
+ *  0.23
+ *    Implement different capture rates.
+ *  0.24
+ *    Implement different capture source channels.
+ *    e.g. When HD Capture source is set to SPDIF,
+ *    setting HD Capture channel to 0 captures from CDROM digital input.
+ *    setting HD Capture channel to 1 captures from SPDIF in.
  *
  *  BUGS:
  *    Some stability problems when unloading the snd-p16v kernel module.
@@ -933,6 +939,56 @@ static snd_kcontrol_new_t snd_p16v_capture_source __devinitdata =
 	.get =		snd_p16v_capture_source_get,
 	.put =		snd_p16v_capture_source_put
 };
+
+static int snd_p16v_capture_channel_info(snd_kcontrol_t *kcontrol, snd_ctl_elem_info_t * uinfo)
+{
+	static char *texts[4] = { "0", "1", "2", "3",  };
+
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED;
+	uinfo->count = 1;
+	uinfo->value.enumerated.items = 4;
+	if (uinfo->value.enumerated.item > 3)
+                uinfo->value.enumerated.item = 3;
+	strcpy(uinfo->value.enumerated.name, texts[uinfo->value.enumerated.item]);
+	return 0;
+}
+
+static int snd_p16v_capture_channel_get(snd_kcontrol_t * kcontrol,
+					snd_ctl_elem_value_t * ucontrol)
+{
+	emu10k1_t *emu = snd_kcontrol_chip(kcontrol);
+
+	ucontrol->value.enumerated.item[0] = emu->p16v_capture_channel;
+	return 0;
+}
+
+static int snd_p16v_capture_channel_put(snd_kcontrol_t * kcontrol,
+					snd_ctl_elem_value_t * ucontrol)
+{
+	emu10k1_t *emu = snd_kcontrol_chip(kcontrol);
+	unsigned int val;
+	int change = 0;
+	u32 tmp;
+
+	val = ucontrol->value.enumerated.item[0] ;
+	change = (emu->p16v_capture_channel != val);
+	if (change) {
+		emu->p16v_capture_channel = val;
+		tmp = snd_emu10k1_ptr20_read(emu, CAPTURE_P16V_SOURCE, 0) & 0xfffc;
+		snd_emu10k1_ptr20_write(emu, CAPTURE_P16V_SOURCE, 0, tmp | val);
+	}
+        return change;
+}
+
+static snd_kcontrol_new_t snd_p16v_capture_channel __devinitdata =
+{
+	.iface =	SNDRV_CTL_ELEM_IFACE_MIXER,
+	.name =		"HD Capture channel",
+	.info =		snd_p16v_capture_channel_info,
+	.get =		snd_p16v_capture_channel_get,
+	.put =		snd_p16v_capture_channel_put
+};
+
 int snd_p16v_mixer(emu10k1_t *emu)
 {
         int err;
@@ -974,6 +1030,10 @@ int snd_p16v_mixer(emu10k1_t *emu)
                 return -ENOMEM;
         if ((err = snd_ctl_add(card, kctl)))
                 return err;
+        if ((kctl = snd_ctl_new1(&snd_p16v_capture_channel, emu)) == NULL)
+                return -ENOMEM;
+        if ((err = snd_ctl_add(card, kctl)))
+                return err;
         return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 8c50b37c04a026ab6641ecb7eaf0fd479798e8b8 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Sun, 15 May 2005 15:43:54 +0200
Subject: [ALSA] Change some timer ioctls due to confliction

Timer Midlevel,ALSA Core
Change values of some timer ioctls to avoid confliction with FIO* ioctls.
The protocol version is increased to indicate this change.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 include/sound/asound.h    | 11 ++++++-----
 sound/core/timer.c        | 11 +++++++++++
 sound/core/timer_compat.c |  4 ++++
 3 files changed, 21 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/sound/asound.h b/include/sound/asound.h
index 4321e92a7f8..9974f83cca4 100644
--- a/include/sound/asound.h
+++ b/include/sound/asound.h
@@ -560,7 +560,7 @@ enum {
  *  Timer section - /dev/snd/timer
  */
 
-#define SNDRV_TIMER_VERSION		SNDRV_PROTOCOL_VERSION(2, 0, 3)
+#define SNDRV_TIMER_VERSION		SNDRV_PROTOCOL_VERSION(2, 0, 4)
 
 enum sndrv_timer_class {
 	SNDRV_TIMER_CLASS_NONE = -1,
@@ -673,10 +673,11 @@ enum {
 	SNDRV_TIMER_IOCTL_INFO = _IOR('T', 0x11, struct sndrv_timer_info),
 	SNDRV_TIMER_IOCTL_PARAMS = _IOW('T', 0x12, struct sndrv_timer_params),
 	SNDRV_TIMER_IOCTL_STATUS = _IOR('T', 0x14, struct sndrv_timer_status),
-	SNDRV_TIMER_IOCTL_START = _IO('T', 0x20),
-	SNDRV_TIMER_IOCTL_STOP = _IO('T', 0x21),
-	SNDRV_TIMER_IOCTL_CONTINUE = _IO('T', 0x22),
-	SNDRV_TIMER_IOCTL_PAUSE = _IO('T', 0x23),
+	/* The following four ioctls are changed since 1.0.9 due to confliction */
+	SNDRV_TIMER_IOCTL_START = _IO('T', 0xa0),
+	SNDRV_TIMER_IOCTL_STOP = _IO('T', 0xa1),
+	SNDRV_TIMER_IOCTL_CONTINUE = _IO('T', 0xa2),
+	SNDRV_TIMER_IOCTL_PAUSE = _IO('T', 0xa3),
 };
 
 struct sndrv_timer_read {
diff --git a/sound/core/timer.c b/sound/core/timer.c
index 48aebdf6550..305e39d7409 100644
--- a/sound/core/timer.c
+++ b/sound/core/timer.c
@@ -1690,6 +1690,13 @@ static int snd_timer_user_pause(struct file *file)
 	return (err = snd_timer_continue(tu->timeri)) < 0 ? err : 0;
 }
 
+enum {
+	SNDRV_TIMER_IOCTL_START_OLD = _IO('T', 0x20),
+	SNDRV_TIMER_IOCTL_STOP_OLD = _IO('T', 0x21),
+	SNDRV_TIMER_IOCTL_CONTINUE_OLD = _IO('T', 0x22),
+	SNDRV_TIMER_IOCTL_PAUSE_OLD = _IO('T', 0x23),
+};
+
 static long snd_timer_user_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
 	snd_timer_user_t *tu;
@@ -1734,12 +1741,16 @@ static long snd_timer_user_ioctl(struct file *file, unsigned int cmd, unsigned l
 	case SNDRV_TIMER_IOCTL_STATUS:
 		return snd_timer_user_status(file, argp);
 	case SNDRV_TIMER_IOCTL_START:
+	case SNDRV_TIMER_IOCTL_START_OLD:
 		return snd_timer_user_start(file);
 	case SNDRV_TIMER_IOCTL_STOP:
+	case SNDRV_TIMER_IOCTL_STOP_OLD:
 		return snd_timer_user_stop(file);
 	case SNDRV_TIMER_IOCTL_CONTINUE:
+	case SNDRV_TIMER_IOCTL_CONTINUE_OLD:
 		return snd_timer_user_continue(file);
 	case SNDRV_TIMER_IOCTL_PAUSE:
+	case SNDRV_TIMER_IOCTL_PAUSE_OLD:
 		return snd_timer_user_pause(file);
 	}
 	return -ENOTTY;
diff --git a/sound/core/timer_compat.c b/sound/core/timer_compat.c
index 420817d10b7..3de552dfe80 100644
--- a/sound/core/timer_compat.c
+++ b/sound/core/timer_compat.c
@@ -106,9 +106,13 @@ static long snd_timer_user_ioctl_compat(struct file *file, unsigned int cmd, uns
 	case SNDRV_TIMER_IOCTL_SELECT:
 	case SNDRV_TIMER_IOCTL_PARAMS:
 	case SNDRV_TIMER_IOCTL_START:
+	case SNDRV_TIMER_IOCTL_START_OLD:
 	case SNDRV_TIMER_IOCTL_STOP:
+	case SNDRV_TIMER_IOCTL_STOP_OLD:
 	case SNDRV_TIMER_IOCTL_CONTINUE:
+	case SNDRV_TIMER_IOCTL_CONTINUE_OLD:
 	case SNDRV_TIMER_IOCTL_PAUSE:
+	case SNDRV_TIMER_IOCTL_PAUSE_OLD:
 	case SNDRV_TIMER_IOCTL_NEXT_DEVICE:
 		return snd_timer_user_ioctl(file, cmd, (unsigned long)argp);
 	case SNDRV_TIMER_IOCTL_INFO32:
-- 
cgit v1.2.3-70-g09d2


From 9502dcad6c1138a3ce2bae23ccd4be44c718d2a9 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Wed, 18 May 2005 16:25:46 +0200
Subject: [ALSA] Export missing snd_pcm_format_*()

PCM Midlevel
Export snd_pcm_format_size().  This function is used by some out-of-kernel
drivers.

Make snd_pcm_format_cpu_endian() macro for optimization.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 include/sound/pcm.h   | 14 ++++++++++++++
 sound/core/pcm.c      |  1 +
 sound/core/pcm_misc.c | 16 ----------------
 3 files changed, 15 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/sound/pcm.h b/include/sound/pcm.h
index 53fc04d75ba..50a6ee1aeab 100644
--- a/include/sound/pcm.h
+++ b/include/sound/pcm.h
@@ -922,8 +922,22 @@ int snd_pcm_format_unsigned(snd_pcm_format_t format);
 int snd_pcm_format_linear(snd_pcm_format_t format);
 int snd_pcm_format_little_endian(snd_pcm_format_t format);
 int snd_pcm_format_big_endian(snd_pcm_format_t format);
+/**
+ * snd_pcm_format_cpu_endian - Check the PCM format is CPU-endian
+ * @format: the format to check
+ *
+ * Returns 1 if the given PCM format is CPU-endian, 0 if
+ * opposite, or a negative error code if endian not specified.
+ */
+/* int snd_pcm_format_cpu_endian(snd_pcm_format_t format); */
+#ifdef SNDRV_LITTLE_ENDIAN
+#define snd_pcm_format_cpu_endian	snd_pcm_format_little_endian
+#else
+#define snd_pcm_format_cpu_endian	snd_pcm_format_big_endian
+#endif
 int snd_pcm_format_width(snd_pcm_format_t format);			/* in bits */
 int snd_pcm_format_physical_width(snd_pcm_format_t format);		/* in bits */
+ssize_t snd_pcm_format_size(snd_pcm_format_t format, size_t samples);
 const unsigned char *snd_pcm_format_silence_64(snd_pcm_format_t format);
 int snd_pcm_format_set_silence(snd_pcm_format_t format, void *buf, unsigned int frames);
 snd_pcm_format_t snd_pcm_build_linear_format(int width, int unsignd, int big_endian);
diff --git a/sound/core/pcm.c b/sound/core/pcm.c
index a2757fcec1f..d57f4ec3b08 100644
--- a/sound/core/pcm.c
+++ b/sound/core/pcm.c
@@ -1069,6 +1069,7 @@ EXPORT_SYMBOL(snd_pcm_format_little_endian);
 EXPORT_SYMBOL(snd_pcm_format_big_endian);
 EXPORT_SYMBOL(snd_pcm_format_width);
 EXPORT_SYMBOL(snd_pcm_format_physical_width);
+EXPORT_SYMBOL(snd_pcm_format_size);
 EXPORT_SYMBOL(snd_pcm_format_silence_64);
 EXPORT_SYMBOL(snd_pcm_format_set_silence);
 EXPORT_SYMBOL(snd_pcm_build_linear_format);
diff --git a/sound/core/pcm_misc.c b/sound/core/pcm_misc.c
index 422b8db1415..1453743e4da 100644
--- a/sound/core/pcm_misc.c
+++ b/sound/core/pcm_misc.c
@@ -269,22 +269,6 @@ int snd_pcm_format_big_endian(snd_pcm_format_t format)
 	return !val;
 }
 
-/**
- * snd_pcm_format_cpu_endian - Check the PCM format is CPU-endian
- * @format: the format to check
- *
- * Returns 1 if the given PCM format is CPU-endian, 0 if
- * opposite, or a negative error code if endian not specified.
- */
-int snd_pcm_format_cpu_endian(snd_pcm_format_t format)
-{
-#ifdef SNDRV_LITTLE_ENDIAN
-	return snd_pcm_format_little_endian(format);
-#else
-	return snd_pcm_format_big_endian(format);
-#endif
-}
-
 /**
  * snd_pcm_format_width - return the bit-width of the format
  * @format: the format to check
-- 
cgit v1.2.3-70-g09d2


From 123992f728785e05f385d23893bd5ec69871aeb4 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Wed, 18 May 2005 18:02:04 +0200
Subject: [ALSA] sound/core/: possible cleanups

PCM Midlevel,ALSA Core,Timer Midlevel,ALSA sequencer,Virtual Midi
This patch contains the following possible cleanups:
- make needlessly global code static
- #if 0 the following unused global functions
- remove the following unneeded EXPORT_SYMBOL's

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 include/sound/pcm.h             | 18 --------------
 include/sound/seq_midi_event.h  |  2 --
 include/sound/seq_virmidi.h     |  1 -
 include/sound/timer.h           |  2 --
 sound/core/pcm.c                |  1 -
 sound/core/pcm_lib.c            | 52 +++++++++++++++++++++++------------------
 sound/core/pcm_native.c         |  4 ++--
 sound/core/seq/seq_midi_event.c |  6 +++--
 sound/core/seq/seq_queue.c      |  3 ++-
 sound/core/seq/seq_queue.h      |  1 -
 sound/core/seq/seq_timer.c      |  3 ++-
 sound/core/seq/seq_timer.h      |  2 --
 sound/core/seq/seq_virmidi.c    |  4 ++--
 sound/core/sound.c              |  1 -
 sound/core/timer.c              |  8 +------
 15 files changed, 42 insertions(+), 66 deletions(-)

(limited to 'include')

diff --git a/include/sound/pcm.h b/include/sound/pcm.h
index 50a6ee1aeab..d935417575b 100644
--- a/include/sound/pcm.h
+++ b/include/sound/pcm.h
@@ -848,23 +848,6 @@ int snd_interval_ratnum(snd_interval_t *i,
 
 void _snd_pcm_hw_params_any(snd_pcm_hw_params_t *params);
 void _snd_pcm_hw_param_setempty(snd_pcm_hw_params_t *params, snd_pcm_hw_param_t var);
-int snd_pcm_hw_param_min(snd_pcm_substream_t *substream, 
-			 snd_pcm_hw_params_t *params,
-			 snd_pcm_hw_param_t var,
-			 unsigned int val, int *dir);
-int snd_pcm_hw_param_max(snd_pcm_substream_t *substream, 
-			 snd_pcm_hw_params_t *params,
-			 snd_pcm_hw_param_t var,
-			 unsigned int val, int *dir);
-int snd_pcm_hw_param_setinteger(snd_pcm_substream_t *substream, 
-				snd_pcm_hw_params_t *params,
-				snd_pcm_hw_param_t var);
-int snd_pcm_hw_param_first(snd_pcm_substream_t *substream, 
-			   snd_pcm_hw_params_t *params,
-			   snd_pcm_hw_param_t var, int *dir);
-int snd_pcm_hw_param_last(snd_pcm_substream_t *substream, 
-			  snd_pcm_hw_params_t *params,
-			  snd_pcm_hw_param_t var, int *dir);
 int snd_pcm_hw_param_near(snd_pcm_substream_t *substream, 
 			  snd_pcm_hw_params_t *params,
 			  snd_pcm_hw_param_t var, 
@@ -876,7 +859,6 @@ int snd_pcm_hw_param_set(snd_pcm_substream_t *pcm,
 int snd_pcm_hw_params_choose(snd_pcm_substream_t *substream, snd_pcm_hw_params_t *params);
 
 int snd_pcm_hw_refine(snd_pcm_substream_t *substream, snd_pcm_hw_params_t *params);
-int snd_pcm_hw_params(snd_pcm_substream_t *substream, snd_pcm_hw_params_t *params);
 
 int snd_pcm_hw_constraints_init(snd_pcm_substream_t *substream);
 int snd_pcm_hw_constraints_complete(snd_pcm_substream_t *substream);
diff --git a/include/sound/seq_midi_event.h b/include/sound/seq_midi_event.h
index 4357cac0750..8857e2bd31a 100644
--- a/include/sound/seq_midi_event.h
+++ b/include/sound/seq_midi_event.h
@@ -41,9 +41,7 @@ struct snd_midi_event_t {
 };
 
 int snd_midi_event_new(int bufsize, snd_midi_event_t **rdev);
-int snd_midi_event_resize_buffer(snd_midi_event_t *dev, int bufsize);
 void snd_midi_event_free(snd_midi_event_t *dev);
-void snd_midi_event_init(snd_midi_event_t *dev);
 void snd_midi_event_reset_encode(snd_midi_event_t *dev);
 void snd_midi_event_reset_decode(snd_midi_event_t *dev);
 void snd_midi_event_no_status(snd_midi_event_t *dev, int on);
diff --git a/include/sound/seq_virmidi.h b/include/sound/seq_virmidi.h
index cf4e2388103..1ad27e859af 100644
--- a/include/sound/seq_virmidi.h
+++ b/include/sound/seq_virmidi.h
@@ -79,6 +79,5 @@ struct _snd_virmidi_dev {
 #define SNDRV_VIRMIDI_SEQ_DISPATCH	2
 
 int snd_virmidi_new(snd_card_t *card, int device, snd_rawmidi_t **rrmidi);
-int snd_virmidi_receive(snd_rawmidi_t *rmidi, snd_seq_event_t *ev);
 
 #endif /* __SOUND_SEQ_VIRMIDI */
diff --git a/include/sound/timer.h b/include/sound/timer.h
index 57fde990606..1898511a0f3 100644
--- a/include/sound/timer.h
+++ b/include/sound/timer.h
@@ -152,6 +152,4 @@ extern int snd_timer_pause(snd_timer_instance_t * timeri);
 
 extern void snd_timer_interrupt(snd_timer_t * timer, unsigned long ticks_left);
 
-extern unsigned int snd_timer_system_resolution(void);
-
 #endif /* __SOUND_TIMER_H */
diff --git a/sound/core/pcm.c b/sound/core/pcm.c
index d57f4ec3b08..9f4c9209b27 100644
--- a/sound/core/pcm.c
+++ b/sound/core/pcm.c
@@ -1049,7 +1049,6 @@ EXPORT_SYMBOL(snd_pcm_release_substream);
 EXPORT_SYMBOL(snd_pcm_format_name);
   /* pcm_native.c */
 EXPORT_SYMBOL(snd_pcm_link_rwlock);
-EXPORT_SYMBOL(snd_pcm_start);
 #ifdef CONFIG_PM
 EXPORT_SYMBOL(snd_pcm_suspend);
 EXPORT_SYMBOL(snd_pcm_suspend_all);
diff --git a/sound/core/pcm_lib.c b/sound/core/pcm_lib.c
index 151fd99ca2c..7ce8b2164f6 100644
--- a/sound/core/pcm_lib.c
+++ b/sound/core/pcm_lib.c
@@ -1143,7 +1143,8 @@ int snd_pcm_hw_constraint_pow2(snd_pcm_runtime_t *runtime,
 #define INT_MIN ((int)((unsigned int)INT_MAX+1))
 #endif
 
-void _snd_pcm_hw_param_any(snd_pcm_hw_params_t *params, snd_pcm_hw_param_t var)
+static void _snd_pcm_hw_param_any(snd_pcm_hw_params_t *params,
+				  snd_pcm_hw_param_t var)
 {
 	if (hw_is_mask(var)) {
 		snd_mask_any(hw_param_mask(params, var));
@@ -1163,12 +1164,14 @@ void _snd_pcm_hw_param_any(snd_pcm_hw_params_t *params, snd_pcm_hw_param_t var)
 /**
  * snd_pcm_hw_param_any
  */
+#if 0
 int snd_pcm_hw_param_any(snd_pcm_t *pcm, snd_pcm_hw_params_t *params,
 			 snd_pcm_hw_param_t var)
 {
 	_snd_pcm_hw_param_any(params, var);
 	return snd_pcm_hw_refine(pcm, params);
 }
+#endif  /*  0  */
 
 void _snd_pcm_hw_params_any(snd_pcm_hw_params_t *params)
 {
@@ -1186,11 +1189,13 @@ void _snd_pcm_hw_params_any(snd_pcm_hw_params_t *params)
  *
  * Fill PARAMS with full configuration space boundaries
  */
+#if 0
 int snd_pcm_hw_params_any(snd_pcm_t *pcm, snd_pcm_hw_params_t *params)
 {
 	_snd_pcm_hw_params_any(params);
 	return snd_pcm_hw_refine(pcm, params);
 }
+#endif  /*  0  */
 
 /**
  * snd_pcm_hw_param_value
@@ -1198,8 +1203,8 @@ int snd_pcm_hw_params_any(snd_pcm_t *pcm, snd_pcm_hw_params_t *params)
  * Return the value for field PAR if it's fixed in configuration space 
  *  defined by PARAMS. Return -EINVAL otherwise
  */
-int snd_pcm_hw_param_value(const snd_pcm_hw_params_t *params,
-			   snd_pcm_hw_param_t var, int *dir)
+static int snd_pcm_hw_param_value(const snd_pcm_hw_params_t *params,
+				  snd_pcm_hw_param_t var, int *dir)
 {
 	if (hw_is_mask(var)) {
 		const snd_mask_t *mask = hw_param_mask_c(params, var);
@@ -1303,6 +1308,7 @@ int _snd_pcm_hw_param_setinteger(snd_pcm_hw_params_t *params,
  * non integer values. Reduce configuration space accordingly.
  * Return -EINVAL if the configuration space is empty
  */
+#if 0
 int snd_pcm_hw_param_setinteger(snd_pcm_t *pcm, 
 				snd_pcm_hw_params_t *params,
 				snd_pcm_hw_param_t var)
@@ -1317,9 +1323,10 @@ int snd_pcm_hw_param_setinteger(snd_pcm_t *pcm,
 	}
 	return 0;
 }
+#endif  /*  0  */
 
-int _snd_pcm_hw_param_first(snd_pcm_hw_params_t *params,
-			    snd_pcm_hw_param_t var)
+static int _snd_pcm_hw_param_first(snd_pcm_hw_params_t *params,
+				   snd_pcm_hw_param_t var)
 {
 	int changed;
 	if (hw_is_mask(var))
@@ -1345,9 +1352,9 @@ int _snd_pcm_hw_param_first(snd_pcm_hw_params_t *params,
  * values > minimum. Reduce configuration space accordingly.
  * Return the minimum.
  */
-int snd_pcm_hw_param_first(snd_pcm_t *pcm, 
-			   snd_pcm_hw_params_t *params, 
-			   snd_pcm_hw_param_t var, int *dir)
+static int snd_pcm_hw_param_first(snd_pcm_t *pcm, 
+				  snd_pcm_hw_params_t *params, 
+				  snd_pcm_hw_param_t var, int *dir)
 {
 	int changed = _snd_pcm_hw_param_first(params, var);
 	if (changed < 0)
@@ -1359,8 +1366,8 @@ int snd_pcm_hw_param_first(snd_pcm_t *pcm,
 	return snd_pcm_hw_param_value(params, var, dir);
 }
 
-int _snd_pcm_hw_param_last(snd_pcm_hw_params_t *params,
-			   snd_pcm_hw_param_t var)
+static int _snd_pcm_hw_param_last(snd_pcm_hw_params_t *params,
+				  snd_pcm_hw_param_t var)
 {
 	int changed;
 	if (hw_is_mask(var))
@@ -1386,9 +1393,9 @@ int _snd_pcm_hw_param_last(snd_pcm_hw_params_t *params,
  * values < maximum. Reduce configuration space accordingly.
  * Return the maximum.
  */
-int snd_pcm_hw_param_last(snd_pcm_t *pcm, 
-			  snd_pcm_hw_params_t *params,
-			  snd_pcm_hw_param_t var, int *dir)
+static int snd_pcm_hw_param_last(snd_pcm_t *pcm, 
+				 snd_pcm_hw_params_t *params,
+				 snd_pcm_hw_param_t var, int *dir)
 {
 	int changed = _snd_pcm_hw_param_last(params, var);
 	if (changed < 0)
@@ -1437,8 +1444,9 @@ int _snd_pcm_hw_param_min(snd_pcm_hw_params_t *params,
  * values < VAL. Reduce configuration space accordingly.
  * Return new minimum or -EINVAL if the configuration space is empty
  */
-int snd_pcm_hw_param_min(snd_pcm_t *pcm, snd_pcm_hw_params_t *params,
-			 snd_pcm_hw_param_t var, unsigned int val, int *dir)
+static int snd_pcm_hw_param_min(snd_pcm_t *pcm, snd_pcm_hw_params_t *params,
+				snd_pcm_hw_param_t var, unsigned int val,
+				int *dir)
 {
 	int changed = _snd_pcm_hw_param_min(params, var, val, dir ? *dir : 0);
 	if (changed < 0)
@@ -1451,8 +1459,9 @@ int snd_pcm_hw_param_min(snd_pcm_t *pcm, snd_pcm_hw_params_t *params,
 	return snd_pcm_hw_param_value_min(params, var, dir);
 }
 
-int _snd_pcm_hw_param_max(snd_pcm_hw_params_t *params,
-			   snd_pcm_hw_param_t var, unsigned int val, int dir)
+static int _snd_pcm_hw_param_max(snd_pcm_hw_params_t *params,
+				 snd_pcm_hw_param_t var, unsigned int val,
+				 int dir)
 {
 	int changed;
 	int open = 0;
@@ -1490,8 +1499,9 @@ int _snd_pcm_hw_param_max(snd_pcm_hw_params_t *params,
  *  values >= VAL + 1. Reduce configuration space accordingly.
  *  Return new maximum or -EINVAL if the configuration space is empty
  */
-int snd_pcm_hw_param_max(snd_pcm_t *pcm, snd_pcm_hw_params_t *params,
-			  snd_pcm_hw_param_t var, unsigned int val, int *dir)
+static int snd_pcm_hw_param_max(snd_pcm_t *pcm, snd_pcm_hw_params_t *params,
+				snd_pcm_hw_param_t var, unsigned int val,
+				int *dir)
 {
 	int changed = _snd_pcm_hw_param_max(params, var, val, dir ? *dir : 0);
 	if (changed < 0)
@@ -2564,9 +2574,6 @@ snd_pcm_sframes_t snd_pcm_lib_readv(snd_pcm_substream_t *substream,
 EXPORT_SYMBOL(snd_interval_refine);
 EXPORT_SYMBOL(snd_interval_list);
 EXPORT_SYMBOL(snd_interval_ratnum);
-EXPORT_SYMBOL(snd_interval_muldivk);
-EXPORT_SYMBOL(snd_interval_mulkdiv);
-EXPORT_SYMBOL(snd_interval_div);
 EXPORT_SYMBOL(_snd_pcm_hw_params_any);
 EXPORT_SYMBOL(_snd_pcm_hw_param_min);
 EXPORT_SYMBOL(_snd_pcm_hw_param_set);
@@ -2580,7 +2587,6 @@ EXPORT_SYMBOL(snd_pcm_hw_param_last);
 EXPORT_SYMBOL(snd_pcm_hw_param_near);
 EXPORT_SYMBOL(snd_pcm_hw_param_set);
 EXPORT_SYMBOL(snd_pcm_hw_refine);
-EXPORT_SYMBOL(snd_pcm_hw_params);
 EXPORT_SYMBOL(snd_pcm_hw_constraints_init);
 EXPORT_SYMBOL(snd_pcm_hw_constraints_complete);
 EXPORT_SYMBOL(snd_pcm_hw_constraint_list);
diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c
index cad9bbde998..4e582415a08 100644
--- a/sound/core/pcm_native.c
+++ b/sound/core/pcm_native.c
@@ -337,8 +337,8 @@ out:
 	return err;
 }
 
-int snd_pcm_hw_params(snd_pcm_substream_t *substream,
-		      snd_pcm_hw_params_t *params)
+static int snd_pcm_hw_params(snd_pcm_substream_t *substream,
+			     snd_pcm_hw_params_t *params)
 {
 	snd_pcm_runtime_t *runtime;
 	int err;
diff --git a/sound/core/seq/seq_midi_event.c b/sound/core/seq/seq_midi_event.c
index 21e569062bc..df1e2bb3974 100644
--- a/sound/core/seq/seq_midi_event.c
+++ b/sound/core/seq/seq_midi_event.c
@@ -171,11 +171,13 @@ void snd_midi_event_reset_decode(snd_midi_event_t *dev)
 	spin_unlock_irqrestore(&dev->lock, flags);
 }
 
+#if 0
 void snd_midi_event_init(snd_midi_event_t *dev)
 {
 	snd_midi_event_reset_encode(dev);
 	snd_midi_event_reset_decode(dev);
 }
+#endif  /*  0  */
 
 void snd_midi_event_no_status(snd_midi_event_t *dev, int on)
 {
@@ -185,6 +187,7 @@ void snd_midi_event_no_status(snd_midi_event_t *dev, int on)
 /*
  * resize buffer
  */
+#if 0
 int snd_midi_event_resize_buffer(snd_midi_event_t *dev, int bufsize)
 {
 	unsigned char *new_buf, *old_buf;
@@ -204,6 +207,7 @@ int snd_midi_event_resize_buffer(snd_midi_event_t *dev, int bufsize)
 	kfree(old_buf);
 	return 0;
 }
+#endif  /*  0  */
 
 /*
  *  read bytes and encode to sequencer event if finished
@@ -517,8 +521,6 @@ static int extra_decode_xrpn(snd_midi_event_t *dev, unsigned char *buf, int coun
  
 EXPORT_SYMBOL(snd_midi_event_new);
 EXPORT_SYMBOL(snd_midi_event_free);
-EXPORT_SYMBOL(snd_midi_event_resize_buffer);
-EXPORT_SYMBOL(snd_midi_event_init);
 EXPORT_SYMBOL(snd_midi_event_reset_encode);
 EXPORT_SYMBOL(snd_midi_event_reset_decode);
 EXPORT_SYMBOL(snd_midi_event_no_status);
diff --git a/sound/core/seq/seq_queue.c b/sound/core/seq/seq_queue.c
index 3afc7cc0c9a..98de2e711fd 100644
--- a/sound/core/seq/seq_queue.c
+++ b/sound/core/seq/seq_queue.c
@@ -672,7 +672,8 @@ static void queue_broadcast_event(queue_t *q, snd_seq_event_t *ev, int atomic, i
  * process a received queue-control event.
  * this function is exported for seq_sync.c.
  */
-void snd_seq_queue_process_event(queue_t *q, snd_seq_event_t *ev, int atomic, int hop)
+static void snd_seq_queue_process_event(queue_t *q, snd_seq_event_t *ev,
+					int atomic, int hop)
 {
 	switch (ev->type) {
 	case SNDRV_SEQ_EVENT_START:
diff --git a/sound/core/seq/seq_queue.h b/sound/core/seq/seq_queue.h
index b1bf5519fb3..ea3c54216ea 100644
--- a/sound/core/seq/seq_queue.h
+++ b/sound/core/seq/seq_queue.h
@@ -111,7 +111,6 @@ int snd_seq_queue_use(int queueid, int client, int use);
 int snd_seq_queue_is_used(int queueid, int client);
 
 int snd_seq_control_queue(snd_seq_event_t *ev, int atomic, int hop);
-void snd_seq_queue_process_event(queue_t *q, snd_seq_event_t *ev, int atomic, int hop);
 
 /*
  * 64bit division - for sync stuff..
diff --git a/sound/core/seq/seq_timer.c b/sound/core/seq/seq_timer.c
index 753f1c0863c..a7f76fc9528 100644
--- a/sound/core/seq/seq_timer.c
+++ b/sound/core/seq/seq_timer.c
@@ -36,7 +36,8 @@ extern int seq_default_timer_resolution;
 
 #define SKEW_BASE	0x10000	/* 16bit shift */
 
-void snd_seq_timer_set_tick_resolution(seq_timer_tick_t *tick, int tempo, int ppq, int nticks)
+static void snd_seq_timer_set_tick_resolution(seq_timer_tick_t *tick,
+					      int tempo, int ppq, int nticks)
 {
 	if (tempo < 1000000)
 		tick->resolution = (tempo * 1000) / ppq;
diff --git a/sound/core/seq/seq_timer.h b/sound/core/seq/seq_timer.h
index 4c0872df893..287ed68591d 100644
--- a/sound/core/seq/seq_timer.h
+++ b/sound/core/seq/seq_timer.h
@@ -64,8 +64,6 @@ extern seq_timer_t *snd_seq_timer_new(void);
 /* delete timer (destructor) */
 extern void snd_seq_timer_delete(seq_timer_t **tmr);
 
-void snd_seq_timer_set_tick_resolution(seq_timer_tick_t *tick, int tempo, int ppq, int nticks);
-
 /* */
 static inline void snd_seq_timer_update_tick(seq_timer_tick_t *tick, unsigned long resolution)
 {
diff --git a/sound/core/seq/seq_virmidi.c b/sound/core/seq/seq_virmidi.c
index 58c56a198d2..a66484b5cf0 100644
--- a/sound/core/seq/seq_virmidi.c
+++ b/sound/core/seq/seq_virmidi.c
@@ -110,7 +110,7 @@ static int snd_virmidi_dev_receive_event(snd_virmidi_dev_t *rdev, snd_seq_event_
  * handler of a remote port which is attached to the virmidi via
  * SNDRV_VIRMIDI_SEQ_ATTACH.
  */
-/* exported */
+#if 0
 int snd_virmidi_receive(snd_rawmidi_t *rmidi, snd_seq_event_t *ev)
 {
 	snd_virmidi_dev_t *rdev;
@@ -118,6 +118,7 @@ int snd_virmidi_receive(snd_rawmidi_t *rmidi, snd_seq_event_t *ev)
 	rdev = rmidi->private_data;
 	return snd_virmidi_dev_receive_event(rdev, ev);
 }
+#endif  /*  0  */
 
 /*
  * event handler of virmidi port
@@ -548,4 +549,3 @@ module_init(alsa_virmidi_init)
 module_exit(alsa_virmidi_exit)
 
 EXPORT_SYMBOL(snd_virmidi_new);
-EXPORT_SYMBOL(snd_virmidi_receive);
diff --git a/sound/core/sound.c b/sound/core/sound.c
index 88e052079f8..fa92e660ec2 100644
--- a/sound/core/sound.c
+++ b/sound/core/sound.c
@@ -431,7 +431,6 @@ EXPORT_SYMBOL(snd_card_pci_resume);
 EXPORT_SYMBOL(snd_device_new);
 EXPORT_SYMBOL(snd_device_register);
 EXPORT_SYMBOL(snd_device_free);
-EXPORT_SYMBOL(snd_device_free_all);
   /* isadma.c */
 #ifdef CONFIG_ISA
 EXPORT_SYMBOL(snd_dma_program);
diff --git a/sound/core/timer.c b/sound/core/timer.c
index cb011a1d4c3..d67a5e91a10 100644
--- a/sound/core/timer.c
+++ b/sound/core/timer.c
@@ -845,7 +845,7 @@ int snd_timer_dev_register(snd_device_t *dev)
 	return 0;
 }
 
-int snd_timer_unregister(snd_timer_t *timer)
+static int snd_timer_unregister(snd_timer_t *timer)
 {
 	struct list_head *p, *n;
 	snd_timer_instance_t *ti;
@@ -946,11 +946,6 @@ struct snd_timer_system_private {
 	unsigned long correction;
 };
 
-unsigned int snd_timer_system_resolution(void)
-{
-	return 1000000000L / HZ;
-}
-
 static void snd_timer_s_function(unsigned long data)
 {
 	snd_timer_t *timer = (snd_timer_t *)data;
@@ -1938,4 +1933,3 @@ EXPORT_SYMBOL(snd_timer_global_free);
 EXPORT_SYMBOL(snd_timer_global_register);
 EXPORT_SYMBOL(snd_timer_global_unregister);
 EXPORT_SYMBOL(snd_timer_interrupt);
-EXPORT_SYMBOL(snd_timer_system_resolution);
-- 
cgit v1.2.3-70-g09d2


From 209ac85d76e4edf05779b4bd5c2a92b059e9ab4d Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Mon, 23 May 2005 10:29:53 +0200
Subject: [ALSA] sound/isa/: cleanups

GUS Library
This patch contains the following possible cleanups:
- make needlesly global code static
- #if 0 the following unused global functions:
  - gus/gus_volume.c: snd_gf1_gvol_to_lvol_raw
  - gus/gus_volume.c: snd_gf1_calc_ramp_rate
  - gus/gus_volume.c: snd_gf1_compute_vibrato
  - gus/gus_volume.c: snd_gf1_compute_pitchbend
  - gus/gus_volume.c: snd_gf1_compute_freq
  - gus/gus_io.c: snd_gf1_i_adlib_write
  - gus/gus_io.c: snd_gf1_i_write_addr
  - gus/gus_io.c: snd_gf1_pokew
  - gus/gus_io.c: snd_gf1_peekw
  - gus/gus_io.c: snd_gf1_dram_setmem
  - gus/gus_io.c: snd_gf1_print_global_registers
  - gus/gus_io.c: snd_gf1_print_setup_registers
  - gus/gus_io.c: snd_gf1_peek_print_block
  - gus/gus_io.c: snd_gf1_print_setup_registers
  - gus/gus_io.c: snd_gf1_peek_print_block
- #if 0 the following unused global variable:
  - gus/gus_tables.h: snd_gf1_scale_table
- remove the following unneeded EXPORT_SYMBOL's:
  - gus/gus_main.c: snd_gf1_i_write16
  - gus/gus_main.c: snd_gf1_start
  - gus/gus_main.c: snd_gf1_stop

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 include/sound/gus.h        | 23 -----------------------
 sound/isa/gus/gus_io.c     | 14 +++++++++++---
 sound/isa/gus/gus_main.c   |  3 ---
 sound/isa/gus/gus_mem.c    | 12 ++++++------
 sound/isa/gus/gus_reset.c  |  3 ++-
 sound/isa/gus/gus_synth.c  |  3 ++-
 sound/isa/gus/gus_tables.h |  4 ++++
 sound/isa/gus/gus_volume.c |  8 ++++++++
 8 files changed, 33 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/include/sound/gus.h b/include/sound/gus.h
index 8b6287a6fff..b4b461ca173 100644
--- a/include/sound/gus.h
+++ b/include/sound/gus.h
@@ -526,9 +526,6 @@ extern void snd_gf1_adlib_write(snd_gus_card_t * gus, unsigned char reg, unsigne
 extern void snd_gf1_dram_addr(snd_gus_card_t * gus, unsigned int addr);
 extern void snd_gf1_poke(snd_gus_card_t * gus, unsigned int addr, unsigned char data);
 extern unsigned char snd_gf1_peek(snd_gus_card_t * gus, unsigned int addr);
-extern void snd_gf1_pokew(snd_gus_card_t * gus, unsigned int addr, unsigned short data);
-extern unsigned short snd_gf1_peekw(snd_gus_card_t * gus, unsigned int addr);
-extern void snd_gf1_dram_setmem(snd_gus_card_t * gus, unsigned int addr, unsigned short value, unsigned int count);
 extern void snd_gf1_write_addr(snd_gus_card_t * gus, unsigned char reg, unsigned int addr, short w_16bit);
 extern unsigned int snd_gf1_read_addr(snd_gus_card_t * gus, unsigned char reg, short w_16bit);
 extern void snd_gf1_i_ctrl_stop(snd_gus_card_t * gus, unsigned char reg);
@@ -544,9 +541,6 @@ extern inline unsigned short snd_gf1_i_read16(snd_gus_card_t * gus, unsigned cha
 {
 	return snd_gf1_i_look16(gus, reg | 0x80);
 }
-extern void snd_gf1_i_adlib_write(snd_gus_card_t * gus, unsigned char reg, unsigned char data);
-extern void snd_gf1_i_write_addr(snd_gus_card_t * gus, unsigned char reg, unsigned int addr, short w_16bit);
-extern unsigned int snd_gf1_i_read_addr(snd_gus_card_t * gus, unsigned char reg, short w_16bit);
 
 extern void snd_gf1_select_active_voices(snd_gus_card_t * gus);
 
@@ -580,10 +574,6 @@ extern void snd_gf1_lfo_command(snd_gus_card_t * gus, int voice, unsigned char *
 
 void snd_gf1_mem_lock(snd_gf1_mem_t * alloc, int xup);
 int snd_gf1_mem_xfree(snd_gf1_mem_t * alloc, snd_gf1_mem_block_t * block);
-snd_gf1_mem_block_t *snd_gf1_mem_look(snd_gf1_mem_t * alloc,
-				      unsigned int address);
-snd_gf1_mem_block_t *snd_gf1_mem_share(snd_gf1_mem_t * alloc,
-				       unsigned int *share_id);
 snd_gf1_mem_block_t *snd_gf1_mem_alloc(snd_gf1_mem_t * alloc, int owner,
 				       char *name, int size, int w_16,
 				       int align, unsigned int *share_id);
@@ -608,23 +598,13 @@ int snd_gf1_dma_transfer_block(snd_gus_card_t * gus,
 /* gus_volume.c */
 
 unsigned short snd_gf1_lvol_to_gvol_raw(unsigned int vol);
-unsigned int snd_gf1_gvol_to_lvol_raw(unsigned short gf1_vol);
-unsigned int snd_gf1_calc_ramp_rate(snd_gus_card_t * gus,
-				    unsigned short start,
-				    unsigned short end,
-				    unsigned int us);
 unsigned short snd_gf1_translate_freq(snd_gus_card_t * gus, unsigned int freq2);
-unsigned short snd_gf1_compute_pitchbend(unsigned short pitchbend, unsigned short sens);
-unsigned short snd_gf1_compute_freq(unsigned int freq,
-				    unsigned int rate,
-				    unsigned short mix_rate);
 
 /* gus_reset.c */
 
 void snd_gf1_set_default_handlers(snd_gus_card_t * gus, unsigned int what);
 void snd_gf1_smart_stop_voice(snd_gus_card_t * gus, unsigned short voice);
 void snd_gf1_stop_voice(snd_gus_card_t * gus, unsigned short voice);
-void snd_gf1_clear_voices(snd_gus_card_t * gus, unsigned short v_min, unsigned short v_max);
 void snd_gf1_stop_voices(snd_gus_card_t * gus, unsigned short v_min, unsigned short v_max);
 snd_gus_voice_t *snd_gf1_alloc_voice(snd_gus_card_t * gus, int type, int client, int port);
 void snd_gf1_free_voice(snd_gus_card_t * gus, snd_gus_voice_t *voice);
@@ -641,9 +621,6 @@ int snd_gf1_pcm_new(snd_gus_card_t * gus, int pcm_dev, int control_index, snd_pc
 
 #ifdef CONFIG_SND_DEBUG
 extern void snd_gf1_print_voice_registers(snd_gus_card_t * gus);
-extern void snd_gf1_print_global_registers(snd_gus_card_t * gus);
-extern void snd_gf1_print_setup_registers(snd_gus_card_t * gus);
-extern void snd_gf1_peek_print_block(snd_gus_card_t * gus, unsigned int addr, int count, int w_16bit);
 #endif
 
 /* gus.c */
diff --git a/sound/isa/gus/gus_io.c b/sound/isa/gus/gus_io.c
index f0570f2bf75..337b0e2a8a3 100644
--- a/sound/isa/gus/gus_io.c
+++ b/sound/isa/gus/gus_io.c
@@ -244,6 +244,8 @@ unsigned short snd_gf1_i_look16(snd_gus_card_t * gus, unsigned char reg)
 	return res;
 }
 
+#if 0
+
 void snd_gf1_i_adlib_write(snd_gus_card_t * gus,
 		           unsigned char reg,
 		           unsigned char data)
@@ -265,6 +267,8 @@ void snd_gf1_i_write_addr(snd_gus_card_t * gus, unsigned char reg,
 	spin_unlock_irqrestore(&gus->reg_lock, flags);
 }
 
+#endif  /*  0  */
+
 unsigned int snd_gf1_i_read_addr(snd_gus_card_t * gus,
 				 unsigned char reg, short w_16bit)
 {
@@ -329,6 +333,8 @@ unsigned char snd_gf1_peek(snd_gus_card_t * gus, unsigned int addr)
 	return res;
 }
 
+#if 0
+
 void snd_gf1_pokew(snd_gus_card_t * gus, unsigned int addr, unsigned short data)
 {
 	unsigned long flags;
@@ -405,9 +411,7 @@ void snd_gf1_dram_setmem(snd_gus_card_t * gus, unsigned int addr,
 	spin_unlock_irqrestore(&gus->reg_lock, flags);
 }
 
-/*
-
- */
+#endif  /*  0  */
 
 void snd_gf1_select_active_voices(snd_gus_card_t * gus)
 {
@@ -469,6 +473,8 @@ void snd_gf1_print_voice_registers(snd_gus_card_t * gus)
 		printk(" -%i- GF1  pan                    = 0x%x\n", voice, snd_gf1_i_read8(gus, 0x0c));
 }
 
+#if 0
+
 void snd_gf1_print_global_registers(snd_gus_card_t * gus)
 {
 	unsigned char global_mode = 0x00;
@@ -528,4 +534,6 @@ void snd_gf1_peek_print_block(snd_gus_card_t * gus, unsigned int addr, int count
 	}
 }
 
+#endif  /*  0  */
+
 #endif
diff --git a/sound/isa/gus/gus_main.c b/sound/isa/gus/gus_main.c
index 73f81c14f76..94bbd344be5 100644
--- a/sound/isa/gus/gus_main.c
+++ b/sound/isa/gus/gus_main.c
@@ -459,7 +459,6 @@ EXPORT_SYMBOL(snd_gf1_write16);
 EXPORT_SYMBOL(snd_gf1_look16);
 EXPORT_SYMBOL(snd_gf1_i_write8);
 EXPORT_SYMBOL(snd_gf1_i_look8);
-EXPORT_SYMBOL(snd_gf1_i_write16);
 EXPORT_SYMBOL(snd_gf1_i_look16);
 EXPORT_SYMBOL(snd_gf1_dram_addr);
 EXPORT_SYMBOL(snd_gf1_write_addr);
@@ -470,8 +469,6 @@ EXPORT_SYMBOL(snd_gf1_alloc_voice);
 EXPORT_SYMBOL(snd_gf1_free_voice);
 EXPORT_SYMBOL(snd_gf1_ctrl_stop);
 EXPORT_SYMBOL(snd_gf1_stop_voice);
-EXPORT_SYMBOL(snd_gf1_start);
-EXPORT_SYMBOL(snd_gf1_stop);
   /* gus_mixer.c */
 EXPORT_SYMBOL(snd_gf1_new_mixer);
   /* gus_pcm.c */
diff --git a/sound/isa/gus/gus_mem.c b/sound/isa/gus/gus_mem.c
index bfc2b91001d..609838e8ef6 100644
--- a/sound/isa/gus/gus_mem.c
+++ b/sound/isa/gus/gus_mem.c
@@ -39,8 +39,8 @@ void snd_gf1_mem_lock(snd_gf1_mem_t * alloc, int xup)
 	}
 }
 
-snd_gf1_mem_block_t *snd_gf1_mem_xalloc(snd_gf1_mem_t * alloc,
-				        snd_gf1_mem_block_t * block)
+static snd_gf1_mem_block_t *snd_gf1_mem_xalloc(snd_gf1_mem_t * alloc,
+					       snd_gf1_mem_block_t * block)
 {
 	snd_gf1_mem_block_t *pblock, *nblock;
 
@@ -105,8 +105,8 @@ int snd_gf1_mem_xfree(snd_gf1_mem_t * alloc, snd_gf1_mem_block_t * block)
 	return 0;
 }
 
-snd_gf1_mem_block_t *snd_gf1_mem_look(snd_gf1_mem_t * alloc,
-				      unsigned int address)
+static snd_gf1_mem_block_t *snd_gf1_mem_look(snd_gf1_mem_t * alloc,
+					     unsigned int address)
 {
 	snd_gf1_mem_block_t *block;
 
@@ -118,8 +118,8 @@ snd_gf1_mem_block_t *snd_gf1_mem_look(snd_gf1_mem_t * alloc,
 	return NULL;
 }
 
-snd_gf1_mem_block_t *snd_gf1_mem_share(snd_gf1_mem_t * alloc,
-				       unsigned int *share_id)
+static snd_gf1_mem_block_t *snd_gf1_mem_share(snd_gf1_mem_t * alloc,
+					      unsigned int *share_id)
 {
 	snd_gf1_mem_block_t *block;
 
diff --git a/sound/isa/gus/gus_reset.c b/sound/isa/gus/gus_reset.c
index b4e66f6a10a..ef687abc707 100644
--- a/sound/isa/gus/gus_reset.c
+++ b/sound/isa/gus/gus_reset.c
@@ -161,7 +161,8 @@ void snd_gf1_stop_voice(snd_gus_card_t * gus, unsigned short voice)
 #endif
 }
 
-void snd_gf1_clear_voices(snd_gus_card_t * gus, unsigned short v_min, unsigned short v_max)
+static void snd_gf1_clear_voices(snd_gus_card_t * gus, unsigned short v_min,
+				 unsigned short v_max)
 {
 	unsigned long flags;
 	unsigned int daddr;
diff --git a/sound/isa/gus/gus_synth.c b/sound/isa/gus/gus_synth.c
index 66552e6013a..f51c386ee19 100644
--- a/sound/isa/gus/gus_synth.c
+++ b/sound/isa/gus/gus_synth.c
@@ -99,7 +99,8 @@ static void snd_gus_synth_free_private_instruments(snd_gus_port_t *p, int client
 	snd_seq_instr_list_free_cond(p->gus->gf1.ilist, &ifree, client, 0);
 }
  
-int snd_gus_synth_event_input(snd_seq_event_t *ev, int direct, void *private_data, int atomic, int hop)
+static int snd_gus_synth_event_input(snd_seq_event_t *ev, int direct,
+				     void *private_data, int atomic, int hop)
 {
 	snd_gus_port_t * p = (snd_gus_port_t *) private_data;
 	
diff --git a/sound/isa/gus/gus_tables.h b/sound/isa/gus/gus_tables.h
index ed8e9d85ad3..4adf098d326 100644
--- a/sound/isa/gus/gus_tables.h
+++ b/sound/isa/gus/gus_tables.h
@@ -23,6 +23,8 @@
 
 #ifdef __GUS_TABLES_ALLOC__
 
+#if 0
+
 unsigned int snd_gf1_scale_table[SNDRV_GF1_SCALE_TABLE_SIZE] =
 {
       8372,      8870,      9397,      9956,     10548,     11175,
@@ -49,6 +51,8 @@ unsigned int snd_gf1_scale_table[SNDRV_GF1_SCALE_TABLE_SIZE] =
   12123977,  12844906
 };
 
+#endif  /*  0  */
+
 unsigned short snd_gf1_atten_table[SNDRV_GF1_ATTEN_TABLE_SIZE] = {
   4095 /* 0   */,1789 /* 1   */,1533 /* 2   */,1383 /* 3   */,1277 /* 4   */,
   1195 /* 5   */,1127 /* 6   */,1070 /* 7   */,1021 /* 8   */,978  /* 9   */,
diff --git a/sound/isa/gus/gus_volume.c b/sound/isa/gus/gus_volume.c
index b72bcfb2861..3d36f6c8ee6 100644
--- a/sound/isa/gus/gus_volume.c
+++ b/sound/isa/gus/gus_volume.c
@@ -55,6 +55,8 @@ unsigned short snd_gf1_lvol_to_gvol_raw(unsigned int vol)
 	return (e << 8) | m;
 }
 
+#if 0
+
 unsigned int snd_gf1_gvol_to_lvol_raw(unsigned short gf1_vol)
 {
 	unsigned int rvol;
@@ -108,6 +110,8 @@ unsigned int snd_gf1_calc_ramp_rate(snd_gus_card_t * gus,
 	return (range << 6) | (increment & 0x3f);
 }
 
+#endif  /*  0  */
+
 unsigned short snd_gf1_translate_freq(snd_gus_card_t * gus, unsigned int freq16)
 {
 	freq16 >>= 3;
@@ -120,6 +124,8 @@ unsigned short snd_gf1_translate_freq(snd_gus_card_t * gus, unsigned int freq16)
 	return ((freq16 << 9) + (gus->gf1.playback_freq >> 1)) / gus->gf1.playback_freq;
 }
 
+#if 0
+
 short snd_gf1_compute_vibrato(short cents, unsigned short fc_register)
 {
 	static short vibrato_table[] =
@@ -208,3 +214,5 @@ unsigned short snd_gf1_compute_freq(unsigned int freq,
 	}
 	return (unsigned short) fc;
 }
+
+#endif  /*  0  */
-- 
cgit v1.2.3-70-g09d2


From bbc0274e9bb2e3f1d724d445a2bd32566b9b66f7 Mon Sep 17 00:00:00 2001
From: Jaroslav Kysela <perex@suse.cz>
Date: Sun, 29 May 2005 10:32:48 +0200
Subject: [ALSA] version 1.0.9

---
 include/sound/version.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/sound/version.h b/include/sound/version.h
index f5959de329c..46acfa8c998 100644
--- a/include/sound/version.h
+++ b/include/sound/version.h
@@ -1,3 +1,3 @@
 /* include/version.h.  Generated by configure.  */
-#define CONFIG_SND_VERSION "1.0.9rc3"
-#define CONFIG_SND_DATE "  (Thu Mar 24 10:33:39 2005 UTC)"
+#define CONFIG_SND_VERSION "1.0.9"
+#define CONFIG_SND_DATE " (Sun May 29 07:31:02 2005 UTC)"
-- 
cgit v1.2.3-70-g09d2


From c611763d048990de5cdf848d97af6392f8fa7430 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor_core@ameritech.net>
Date: Wed, 1 Jun 2005 02:39:51 -0500
Subject: Input: add ps2_drain() to libps2 to allow reading and discarding     
   given number of bytes from device. Change ps2_command to        allow using
 0 as command ID and actually pass it to the        device instead of working
 as a drain.

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/mouse/alps.c   |  3 +--
 drivers/input/serio/libps2.c | 46 ++++++++++++++++++++++++++++++++++++--------
 include/linux/libps2.h       |  1 +
 3 files changed, 40 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/drivers/input/mouse/alps.c b/drivers/input/mouse/alps.c
index 2679a165d39..ffdc8231319 100644
--- a/drivers/input/mouse/alps.c
+++ b/drivers/input/mouse/alps.c
@@ -270,7 +270,6 @@ static struct alps_model_info *alps_get_model(struct psmouse *psmouse, int *vers
 static int alps_passthrough_mode(struct psmouse *psmouse, int enable)
 {
 	struct ps2dev *ps2dev = &psmouse->ps2dev;
-	unsigned char param[3];
 	int cmd = enable ? PSMOUSE_CMD_SETSCALE21 : PSMOUSE_CMD_SETSCALE11;
 
 	if (ps2_command(ps2dev, NULL, cmd) ||
@@ -280,7 +279,7 @@ static int alps_passthrough_mode(struct psmouse *psmouse, int enable)
 		return -1;
 
 	/* we may get 3 more bytes, just ignore them */
-	ps2_command(ps2dev, param, 0x0300);
+	ps2_drain(ps2dev, 3, 100);
 
 	return 0;
 }
diff --git a/drivers/input/serio/libps2.c b/drivers/input/serio/libps2.c
index c978657068c..92b92ee0379 100644
--- a/drivers/input/serio/libps2.c
+++ b/drivers/input/serio/libps2.c
@@ -29,6 +29,7 @@ MODULE_LICENSE("GPL");
 
 EXPORT_SYMBOL(ps2_init);
 EXPORT_SYMBOL(ps2_sendbyte);
+EXPORT_SYMBOL(ps2_drain);
 EXPORT_SYMBOL(ps2_command);
 EXPORT_SYMBOL(ps2_schedule_command);
 EXPORT_SYMBOL(ps2_handle_ack);
@@ -45,11 +46,11 @@ struct ps2work {
 
 
 /*
- * ps2_sendbyte() sends a byte to the mouse, and waits for acknowledge.
- * It doesn't handle retransmission, though it could - because when there would
- * be need for retransmissions, the mouse has to be replaced anyway.
+ * ps2_sendbyte() sends a byte to the device and waits for acknowledge.
+ * It doesn't handle retransmission, though it could - because if there
+ * is a need for retransmissions device has to be replaced anyway.
  *
- * ps2_sendbyte() can only be called from a process context
+ * ps2_sendbyte() can only be called from a process context.
  */
 
 int ps2_sendbyte(struct ps2dev *ps2dev, unsigned char byte, int timeout)
@@ -71,6 +72,31 @@ int ps2_sendbyte(struct ps2dev *ps2dev, unsigned char byte, int timeout)
 	return -ps2dev->nak;
 }
 
+/*
+ * ps2_drain() waits for device to transmit requested number of bytes
+ * and discards them.
+ */
+
+void ps2_drain(struct ps2dev *ps2dev, int maxbytes, int timeout)
+{
+	if (maxbytes > sizeof(ps2dev->cmdbuf)) {
+		WARN_ON(1);
+		maxbytes = sizeof(ps2dev->cmdbuf);
+	}
+
+	down(&ps2dev->cmd_sem);
+
+	serio_pause_rx(ps2dev->serio);
+	ps2dev->flags = PS2_FLAG_CMD;
+	ps2dev->cmdcnt = maxbytes;
+	serio_continue_rx(ps2dev->serio);
+
+	wait_event_timeout(ps2dev->wait,
+			   !(ps2dev->flags & PS2_FLAG_CMD),
+			   msecs_to_jiffies(timeout));
+	up(&ps2dev->cmd_sem);
+}
+
 /*
  * ps2_command() sends a command and its parameters to the mouse,
  * then waits for the response and puts it in the param array.
@@ -86,6 +112,11 @@ int ps2_command(struct ps2dev *ps2dev, unsigned char *param, int command)
 	int rc = -1;
 	int i;
 
+	if (receive > sizeof(ps2dev->cmdbuf)) {
+		WARN_ON(1);
+		return -1;
+	}
+
 	down(&ps2dev->cmd_sem);
 
 	serio_pause_rx(ps2dev->serio);
@@ -101,10 +132,9 @@ int ps2_command(struct ps2dev *ps2dev, unsigned char *param, int command)
 	 * ACKing the reset command, and so it can take a long
 	 * time before the ACK arrrives.
 	 */
-	if (command & 0xff)
-		if (ps2_sendbyte(ps2dev, command & 0xff,
-			command == PS2_CMD_RESET_BAT ? 1000 : 200))
-			goto out;
+	if (ps2_sendbyte(ps2dev, command & 0xff,
+			 command == PS2_CMD_RESET_BAT ? 1000 : 200))
+		goto out;
 
 	for (i = 0; i < send; i++)
 		if (ps2_sendbyte(ps2dev, param[i], 200))
diff --git a/include/linux/libps2.h b/include/linux/libps2.h
index 923bdbc6d9e..a710bddda4e 100644
--- a/include/linux/libps2.h
+++ b/include/linux/libps2.h
@@ -41,6 +41,7 @@ struct ps2dev {
 
 void ps2_init(struct ps2dev *ps2dev, struct serio *serio);
 int ps2_sendbyte(struct ps2dev *ps2dev, unsigned char byte, int timeout);
+void ps2_drain(struct ps2dev *ps2dev, int maxbytes, int timeout);
 int ps2_command(struct ps2dev *ps2dev, unsigned char *param, int command);
 int ps2_schedule_command(struct ps2dev *ps2dev, unsigned char *param, int command);
 int ps2_handle_ack(struct ps2dev *ps2dev, unsigned char data);
-- 
cgit v1.2.3-70-g09d2


From dbf4ccd6043e58ed32fbf253fb3f0a9991e4c13a Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor_core@ameritech.net>
Date: Wed, 1 Jun 2005 02:40:01 -0500
Subject: Input: psmouse - export protocol as a sysfs per-device attribute     
   to allow easy switching at run-time.

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/mouse/psmouse-base.c | 291 +++++++++++++++++++++++++++++++------
 drivers/input/mouse/psmouse.h      |   1 +
 drivers/input/serio/serio.c        |  18 ++-
 include/linux/serio.h              |   6 +
 4 files changed, 266 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/drivers/input/mouse/psmouse-base.c b/drivers/input/mouse/psmouse-base.c
index 259e6b70544..19785a6c5ab 100644
--- a/drivers/input/mouse/psmouse-base.c
+++ b/drivers/input/mouse/psmouse-base.c
@@ -32,15 +32,14 @@ MODULE_AUTHOR("Vojtech Pavlik <vojtech@suse.cz>");
 MODULE_DESCRIPTION(DRIVER_DESC);
 MODULE_LICENSE("GPL");
 
-static unsigned int psmouse_max_proto = -1U;
+static unsigned int psmouse_max_proto = PSMOUSE_AUTO;
 static int psmouse_set_maxproto(const char *val, struct kernel_param *kp);
 static int psmouse_get_maxproto(char *buffer, struct kernel_param *kp);
-static char *psmouse_proto_abbrev[] = { NULL, "bare", NULL, NULL, NULL, "imps", "exps", NULL, NULL, "lifebook" };
 #define param_check_proto_abbrev(name, p)	__param_check(name, p, unsigned int)
 #define param_set_proto_abbrev			psmouse_set_maxproto
 #define param_get_proto_abbrev			psmouse_get_maxproto
 module_param_named(proto, psmouse_max_proto, proto_abbrev, 0644);
-MODULE_PARM_DESC(proto, "Highest protocol extension to probe (bare, imps, exps, lifebook, any). Useful for KVM switches.");
+MODULE_PARM_DESC(proto, "Highest protocol extension to probe (bare, imps, exps, any). Useful for KVM switches.");
 
 static unsigned int psmouse_resolution = 200;
 module_param_named(resolution, psmouse_resolution, uint, 0644);
@@ -58,6 +57,7 @@ static unsigned int psmouse_resetafter;
 module_param_named(resetafter, psmouse_resetafter, uint, 0644);
 MODULE_PARM_DESC(resetafter, "Reset device after so many bad packets (0 = never).");
 
+PSMOUSE_DEFINE_ATTR(protocol);
 PSMOUSE_DEFINE_ATTR(rate);
 PSMOUSE_DEFINE_ATTR(resolution);
 PSMOUSE_DEFINE_ATTR(resetafter);
@@ -77,7 +77,14 @@ __obsolete_setup("psmouse_rate=");
  */
 static DECLARE_MUTEX(psmouse_sem);
 
-static char *psmouse_protocols[] = { "None", "PS/2", "PS2++", "ThinkPS/2", "GenPS/2", "ImPS/2", "ImExPS/2", "SynPS/2", "AlpsPS/2", "LBPS/2" };
+struct psmouse_protocol {
+	enum psmouse_type type;
+	char *name;
+	char *alias;
+	int maxproto;
+	int (*detect)(struct psmouse *, int);
+	int (*init)(struct psmouse *);
+};
 
 /*
  * psmouse_process_byte() analyzes the PS/2 data stream and reports
@@ -417,12 +424,15 @@ static int thinking_detect(struct psmouse *psmouse, int set_properties)
  */
 static int ps2bare_detect(struct psmouse *psmouse, int set_properties)
 {
-	if (!psmouse->vendor) psmouse->vendor = "Generic";
-	if (!psmouse->name) psmouse->name = "Mouse";
+	if (set_properties) {
+		if (!psmouse->vendor) psmouse->vendor = "Generic";
+		if (!psmouse->name) psmouse->name = "Mouse";
+	}
 
 	return 0;
 }
 
+
 /*
  * psmouse_extensions() probes for any extensions to the basic PS/2 protocol
  * the mouse may have.
@@ -437,9 +447,7 @@ static int psmouse_extensions(struct psmouse *psmouse,
  * We always check for lifebook because it does not disturb mouse
  * (it only checks DMI information).
  */
-	if (lifebook_detect(psmouse, set_properties) == 0 ||
-	    max_proto == PSMOUSE_LIFEBOOK) {
-
+	if (lifebook_detect(psmouse, set_properties) == 0) {
 		if (max_proto > PSMOUSE_IMEX) {
 			if (!set_properties || lifebook_init(psmouse) == 0)
 				return PSMOUSE_LIFEBOOK;
@@ -529,6 +537,103 @@ static int psmouse_extensions(struct psmouse *psmouse,
 	return PSMOUSE_PS2;
 }
 
+static struct psmouse_protocol psmouse_protocols[] = {
+	{
+		.type		= PSMOUSE_PS2,
+		.name		= "PS/2",
+		.alias		= "bare",
+		.maxproto	= 1,
+		.detect		= ps2bare_detect,
+	},
+	{
+		.type		= PSMOUSE_PS2PP,
+		.name		= "PS2++",
+		.alias		= "logitech",
+		.detect		= ps2pp_init,
+	},
+	{
+		.type		= PSMOUSE_THINKPS,
+		.name		= "ThinkPS/2",
+		.alias		= "thinkps",
+		.detect		= thinking_detect,
+	},
+	{
+		.type		= PSMOUSE_GENPS,
+		.name		= "GenPS/2",
+		.alias		= "genius",
+		.detect		= genius_detect,
+	},
+	{
+		.type		= PSMOUSE_IMPS,
+		.name		= "ImPS/2",
+		.alias		= "imps",
+		.maxproto	= 1,
+		.detect		= intellimouse_detect,
+	},
+	{
+		.type		= PSMOUSE_IMEX,
+		.name		= "ImExPS/2",
+		.alias		= "exps",
+		.maxproto	= 1,
+		.detect		= im_explorer_detect,
+	},
+	{
+		.type		= PSMOUSE_SYNAPTICS,
+		.name		= "SynPS/2",
+		.alias		= "synaptics",
+		.detect		= synaptics_detect,
+		.init		= synaptics_init,
+	},
+	{
+		.type		= PSMOUSE_ALPS,
+		.name		= "AlpsPS/2",
+		.alias		= "alps",
+		.detect		= alps_detect,
+		.init		= alps_init,
+	},
+	{
+		.type		= PSMOUSE_LIFEBOOK,
+		.name		= "LBPS/2",
+		.alias		= "lifebook",
+		.init		= lifebook_init,
+	},
+	{
+		.type		= PSMOUSE_AUTO,
+		.name		= "auto",
+		.alias		= "any",
+		.maxproto	= 1,
+	},
+};
+
+static struct psmouse_protocol *psmouse_protocol_by_type(enum psmouse_type type)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(psmouse_protocols); i++)
+		if (psmouse_protocols[i].type == type)
+			return &psmouse_protocols[i];
+
+	WARN_ON(1);
+	return &psmouse_protocols[0];
+}
+
+static struct psmouse_protocol *psmouse_protocol_by_name(const char *name, size_t len)
+{
+	struct psmouse_protocol *p;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(psmouse_protocols); i++) {
+		p = &psmouse_protocols[i];
+
+		if ((strlen(p->name) == len && !strncmp(p->name, name, len)) ||
+		    (strlen(p->alias) == len && !strncmp(p->alias, name, len)))
+			return &psmouse_protocols[i];
+	}
+
+	return NULL;
+}
+
+
 /*
  * psmouse_probe() probes for a PS/2 mouse.
  */
@@ -680,6 +785,7 @@ static void psmouse_disconnect(struct serio *serio)
 
 	psmouse = serio_get_drvdata(serio);
 
+	device_remove_file(&serio->dev, &psmouse_attr_protocol);
 	device_remove_file(&serio->dev, &psmouse_attr_rate);
 	device_remove_file(&serio->dev, &psmouse_attr_resolution);
 	device_remove_file(&serio->dev, &psmouse_attr_resetafter);
@@ -712,6 +818,49 @@ static void psmouse_disconnect(struct serio *serio)
 	up(&psmouse_sem);
 }
 
+static int psmouse_switch_protocol(struct psmouse *psmouse, struct psmouse_protocol *proto)
+{
+	memset(&psmouse->dev, 0, sizeof(struct input_dev));
+
+	init_input_dev(&psmouse->dev);
+
+	psmouse->dev.private = psmouse;
+	psmouse->dev.dev = &psmouse->ps2dev.serio->dev;
+
+	psmouse->dev.evbit[0] = BIT(EV_KEY) | BIT(EV_REL);
+	psmouse->dev.keybit[LONG(BTN_MOUSE)] = BIT(BTN_LEFT) | BIT(BTN_MIDDLE) | BIT(BTN_RIGHT);
+	psmouse->dev.relbit[0] = BIT(REL_X) | BIT(REL_Y);
+
+	psmouse->set_rate = psmouse_set_rate;
+	psmouse->set_resolution = psmouse_set_resolution;
+	psmouse->protocol_handler = psmouse_process_byte;
+	psmouse->pktsize = 3;
+
+	if (proto && (proto->detect || proto->init)) {
+		if (proto->detect && proto->detect(psmouse, 1) < 0)
+			return -1;
+
+		if (proto->init && proto->init(psmouse) < 0)
+			return -1;
+
+		psmouse->type = proto->type;
+	}
+	else
+		psmouse->type = psmouse_extensions(psmouse, psmouse_max_proto, 1);
+
+	sprintf(psmouse->devname, "%s %s %s",
+		psmouse_protocol_by_type(psmouse->type)->name, psmouse->vendor, psmouse->name);
+
+	psmouse->dev.name = psmouse->devname;
+	psmouse->dev.phys = psmouse->phys;
+	psmouse->dev.id.bustype = BUS_I8042;
+	psmouse->dev.id.vendor = 0x0002;
+	psmouse->dev.id.product = psmouse->type;
+	psmouse->dev.id.version = psmouse->model;
+
+	return 0;
+}
+
 /*
  * psmouse_connect() is a callback from the serio module when
  * an unhandled serio port is found.
@@ -739,11 +888,7 @@ static int psmouse_connect(struct serio *serio, struct serio_driver *drv)
 
 	ps2_init(&psmouse->ps2dev, serio);
 	sprintf(psmouse->phys, "%s/input0", serio->phys);
-	psmouse->dev.evbit[0] = BIT(EV_KEY) | BIT(EV_REL);
-	psmouse->dev.keybit[LONG(BTN_MOUSE)] = BIT(BTN_LEFT) | BIT(BTN_MIDDLE) | BIT(BTN_RIGHT);
-	psmouse->dev.relbit[0] = BIT(REL_X) | BIT(REL_Y);
-	psmouse->dev.private = psmouse;
-	psmouse->dev.dev = &serio->dev;
+
 	psmouse_set_state(psmouse, PSMOUSE_INITIALIZING);
 
 	serio_set_drvdata(serio, psmouse);
@@ -767,25 +912,10 @@ static int psmouse_connect(struct serio *serio, struct serio_driver *drv)
 	psmouse->resolution = psmouse_resolution;
 	psmouse->resetafter = psmouse_resetafter;
 	psmouse->smartscroll = psmouse_smartscroll;
-	psmouse->set_rate = psmouse_set_rate;
-	psmouse->set_resolution = psmouse_set_resolution;
-	psmouse->protocol_handler = psmouse_process_byte;
-	psmouse->pktsize = 3;
-
-	psmouse->type = psmouse_extensions(psmouse, psmouse_max_proto, 1);
-
-	sprintf(psmouse->devname, "%s %s %s",
-		psmouse_protocols[psmouse->type], psmouse->vendor, psmouse->name);
 
-	psmouse->dev.name = psmouse->devname;
-	psmouse->dev.phys = psmouse->phys;
-	psmouse->dev.id.bustype = BUS_I8042;
-	psmouse->dev.id.vendor = 0x0002;
-	psmouse->dev.id.product = psmouse->type;
-	psmouse->dev.id.version = psmouse->model;
+	psmouse_switch_protocol(psmouse, NULL);
 
 	input_register_device(&psmouse->dev);
-
 	printk(KERN_INFO "input: %s on %s\n", psmouse->devname, serio->phys);
 
 	psmouse_set_state(psmouse, PSMOUSE_CMD_MODE);
@@ -795,6 +925,7 @@ static int psmouse_connect(struct serio *serio, struct serio_driver *drv)
 	if (parent && parent->pt_activate)
 		parent->pt_activate(parent);
 
+	device_create_file(&serio->dev, &psmouse_attr_protocol);
 	device_create_file(&serio->dev, &psmouse_attr_rate);
 	device_create_file(&serio->dev, &psmouse_attr_resolution);
 	device_create_file(&serio->dev, &psmouse_attr_resetafter);
@@ -946,11 +1077,14 @@ ssize_t psmouse_attr_set_helper(struct device *dev, const char *buf, size_t coun
 		parent = serio_get_drvdata(serio->parent);
 		psmouse_deactivate(parent);
 	}
+
 	psmouse_deactivate(psmouse);
 
 	retval = handler(psmouse, buf, count);
 
-	psmouse_activate(psmouse);
+	if (retval != -ENODEV)
+		psmouse_activate(psmouse);
+
 	if (parent)
 		psmouse_activate(parent);
 
@@ -961,6 +1095,75 @@ ssize_t psmouse_attr_set_helper(struct device *dev, const char *buf, size_t coun
 	return retval;
 }
 
+static ssize_t psmouse_attr_show_protocol(struct psmouse *psmouse, char *buf)
+{
+	return sprintf(buf, "%s\n", psmouse_protocol_by_type(psmouse->type)->name);
+}
+
+static ssize_t psmouse_attr_set_protocol(struct psmouse *psmouse, const char *buf, size_t count)
+{
+	struct serio *serio = psmouse->ps2dev.serio;
+	struct psmouse *parent = NULL;
+	struct psmouse_protocol *proto;
+	int retry = 0;
+
+	if (!(proto = psmouse_protocol_by_name(buf, count)))
+		return -EINVAL;
+
+	if (psmouse->type == proto->type)
+		return count;
+
+	while (serio->child) {
+		if (++retry > 3) {
+			printk(KERN_WARNING "psmouse: failed to destroy child port, protocol change aborted.\n");
+			return -EIO;
+		}
+
+		up(&psmouse_sem);
+		serio_unpin_driver(serio);
+		serio_unregister_child_port(serio);
+		serio_pin_driver_uninterruptible(serio);
+		down(&psmouse_sem);
+
+		if (serio->drv != &psmouse_drv)
+			return -ENODEV;
+
+		if (psmouse->type == proto->type)
+			return count; /* switched by other thread */
+	}
+
+	if (serio->parent && serio->id.type == SERIO_PS_PSTHRU) {
+		parent = serio_get_drvdata(serio->parent);
+		if (parent->pt_deactivate)
+			parent->pt_deactivate(parent);
+	}
+
+	if (psmouse->disconnect)
+		psmouse->disconnect(psmouse);
+
+	psmouse_set_state(psmouse, PSMOUSE_IGNORE);
+	input_unregister_device(&psmouse->dev);
+
+	psmouse_set_state(psmouse, PSMOUSE_INITIALIZING);
+
+	if (psmouse_switch_protocol(psmouse, proto) < 0) {
+		psmouse_reset(psmouse);
+		/* default to PSMOUSE_PS2 */
+		psmouse_switch_protocol(psmouse, &psmouse_protocols[0]);
+	}
+
+	psmouse_initialize(psmouse);
+	psmouse_set_state(psmouse, PSMOUSE_CMD_MODE);
+
+	input_register_device(&psmouse->dev);
+	printk(KERN_INFO "input: %s on %s\n", psmouse->devname, serio->phys);
+
+	if (parent && parent->pt_activate)
+		parent->pt_activate(parent);
+
+	return count;
+}
+
 static ssize_t psmouse_attr_show_rate(struct psmouse *psmouse, char *buf)
 {
 	return sprintf(buf, "%d\n", psmouse->rate);
@@ -1017,34 +1220,26 @@ static ssize_t psmouse_attr_set_resetafter(struct psmouse *psmouse, const char *
 
 static int psmouse_set_maxproto(const char *val, struct kernel_param *kp)
 {
-	int i;
+	struct psmouse_protocol *proto;
 
 	if (!val)
 		return -EINVAL;
 
-	if (!strncmp(val, "any", 3)) {
-		*((unsigned int *)kp->arg) = -1U;
-		return 0;
-	}
+	proto = psmouse_protocol_by_name(val, strlen(val));
 
-	for (i = 0; i < ARRAY_SIZE(psmouse_proto_abbrev); i++) {
-		if (!psmouse_proto_abbrev[i])
-			continue;
+	if (!proto || !proto->maxproto)
+		return -EINVAL;
 
-		if (!strncmp(val, psmouse_proto_abbrev[i], strlen(psmouse_proto_abbrev[i]))) {
-			*((unsigned int *)kp->arg) = i;
-			return 0;
-		}
-	}
+	*((unsigned int *)kp->arg) = proto->type;
 
-	return -EINVAL;					\
+	return 0;					\
 }
 
 static int psmouse_get_maxproto(char *buffer, struct kernel_param *kp)
 {
-	return sprintf(buffer, "%s\n",
-			psmouse_max_proto < ARRAY_SIZE(psmouse_proto_abbrev) ?
-				psmouse_proto_abbrev[psmouse_max_proto] : "any");
+	int type = *((unsigned int *)kp->arg);
+
+	return sprintf(buffer, "%s\n", psmouse_protocol_by_type(type)->name);
 }
 
 static int __init psmouse_init(void)
diff --git a/drivers/input/mouse/psmouse.h b/drivers/input/mouse/psmouse.h
index 4848be627a6..dc8e9ae07f3 100644
--- a/drivers/input/mouse/psmouse.h
+++ b/drivers/input/mouse/psmouse.h
@@ -78,6 +78,7 @@ enum psmouse_type {
 	PSMOUSE_SYNAPTICS,
 	PSMOUSE_ALPS,
 	PSMOUSE_LIFEBOOK,
+	PSMOUSE_AUTO		/* This one should always be last */
 };
 
 int psmouse_sliced_command(struct psmouse *psmouse, unsigned char command);
diff --git a/drivers/input/serio/serio.c b/drivers/input/serio/serio.c
index b82815a0b65..615bf62ad46 100644
--- a/drivers/input/serio/serio.c
+++ b/drivers/input/serio/serio.c
@@ -42,6 +42,7 @@ MODULE_LICENSE("GPL");
 EXPORT_SYMBOL(serio_interrupt);
 EXPORT_SYMBOL(__serio_register_port);
 EXPORT_SYMBOL(serio_unregister_port);
+EXPORT_SYMBOL(serio_unregister_child_port);
 EXPORT_SYMBOL(__serio_unregister_port_delayed);
 EXPORT_SYMBOL(__serio_register_driver);
 EXPORT_SYMBOL(serio_unregister_driver);
@@ -179,12 +180,12 @@ static void serio_queue_event(void *object, struct module *owner,
 	spin_lock_irqsave(&serio_event_lock, flags);
 
 	/*
- 	 * Scan event list for the other events for the same serio port,
+	 * Scan event list for the other events for the same serio port,
 	 * starting with the most recent one. If event is the same we
 	 * do not need add new one. If event is of different type we
 	 * need to add this event and should not look further because
 	 * we need to preseve sequence of distinct events.
- 	 */
+	 */
 	list_for_each_entry_reverse(event, &serio_event_list, node) {
 		if (event->object == object) {
 			if (event->type == event_type)
@@ -653,6 +654,19 @@ void serio_unregister_port(struct serio *serio)
 	up(&serio_sem);
 }
 
+/*
+ * Safely unregisters child port if one is present.
+ */
+void serio_unregister_child_port(struct serio *serio)
+{
+	down(&serio_sem);
+	if (serio->child) {
+		serio_disconnect_port(serio->child);
+		serio_destroy_port(serio->child);
+	}
+	up(&serio_sem);
+}
+
 /*
  * Submits register request to kseriod for subsequent execution.
  * Can be used when it is not obvious whether the serio_sem is
diff --git a/include/linux/serio.h b/include/linux/serio.h
index a2d3b9ae06f..aa4d6493a03 100644
--- a/include/linux/serio.h
+++ b/include/linux/serio.h
@@ -83,6 +83,7 @@ static inline void serio_register_port(struct serio *serio)
 }
 
 void serio_unregister_port(struct serio *serio);
+void serio_unregister_child_port(struct serio *serio);
 void __serio_unregister_port_delayed(struct serio *serio, struct module *owner);
 static inline void serio_unregister_port_delayed(struct serio *serio)
 {
@@ -153,6 +154,11 @@ static inline int serio_pin_driver(struct serio *serio)
 	return down_interruptible(&serio->drv_sem);
 }
 
+static inline void serio_pin_driver_uninterruptible(struct serio *serio)
+{
+	down(&serio->drv_sem);
+}
+
 static inline void serio_unpin_driver(struct serio *serio)
 {
 	up(&serio->drv_sem);
-- 
cgit v1.2.3-70-g09d2


From 986a80d5c154808cc78170584670324a22fd8219 Mon Sep 17 00:00:00 2001
From: Jesper Juhl <juhl-lkml@dif.dk>
Date: Thu, 16 Jun 2005 15:14:00 -0700
Subject: [PATCH] avoid signed vs unsigned comparison in efi_range_is_wc()

warning when building with gcc -W :

 include/linux/efi.h: In function `efi_range_is_wc':
 include/linux/efi.h:320: warning: comparison between signed and unsigned

It looks to me like a significantly large 'len' passed in could cause the
loop to never end. Isn't it safer to make 'i' an unsigned long as well?
Like this little patch below (which of course also kills the warning) :

Signed-off-by: Jesper Juhl <juhl-lkml@dif.dk>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 include/linux/efi.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/efi.h b/include/linux/efi.h
index 047e7222df7..73781ec165b 100644
--- a/include/linux/efi.h
+++ b/include/linux/efi.h
@@ -315,7 +315,7 @@ extern struct efi_memory_map memmap;
  */
 static inline int efi_range_is_wc(unsigned long start, unsigned long len)
 {
-	int i;
+	unsigned long i;
 
 	for (i = 0; i < len; i += (1UL << EFI_PAGE_SHIFT)) {
 		unsigned long paddr = __pa(start + i);
-- 
cgit v1.2.3-70-g09d2


From b8a9b66fbee09d0cc71c272b5c1d1f3320afbbf0 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 20 Jun 2005 11:31:09 +0100
Subject: [PATCH] ARM: Add common CACHE_COLOUR macro

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/copypage-v6.c    | 6 ++----
 include/asm-arm/cacheflush.h | 3 +++
 2 files changed, 5 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mm/copypage-v6.c b/arch/arm/mm/copypage-v6.c
index a8c00236bd3..27d041574ea 100644
--- a/arch/arm/mm/copypage-v6.c
+++ b/arch/arm/mm/copypage-v6.c
@@ -30,8 +30,6 @@
 
 static DEFINE_SPINLOCK(v6_lock);
 
-#define DCACHE_COLOUR(vaddr) ((vaddr & (SHMLBA - 1)) >> PAGE_SHIFT)
-
 /*
  * Copy the user page.  No aliasing to deal with so we can just
  * attack the kernel's existing mapping of these pages.
@@ -55,7 +53,7 @@ void v6_clear_user_page_nonaliasing(void *kaddr, unsigned long vaddr)
  */
 void v6_copy_user_page_aliasing(void *kto, const void *kfrom, unsigned long vaddr)
 {
-	unsigned int offset = DCACHE_COLOUR(vaddr);
+	unsigned int offset = CACHE_COLOUR(vaddr);
 	unsigned long from, to;
 
 	/*
@@ -95,7 +93,7 @@ void v6_copy_user_page_aliasing(void *kto, const void *kfrom, unsigned long vadd
  */
 void v6_clear_user_page_aliasing(void *kaddr, unsigned long vaddr)
 {
-	unsigned int offset = DCACHE_COLOUR(vaddr);
+	unsigned int offset = CACHE_COLOUR(vaddr);
 	unsigned long to = to_address + (offset << PAGE_SHIFT);
 
 	/*
diff --git a/include/asm-arm/cacheflush.h b/include/asm-arm/cacheflush.h
index 09ffeed507c..035cdcff43d 100644
--- a/include/asm-arm/cacheflush.h
+++ b/include/asm-arm/cacheflush.h
@@ -16,6 +16,9 @@
 
 #include <asm/mman.h>
 #include <asm/glue.h>
+#include <asm/shmparam.h>
+
+#define CACHE_COLOUR(vaddr)	((vaddr & (SHMLBA - 1)) >> PAGE_SHIFT)
 
 /*
  *	Cache Model
-- 
cgit v1.2.3-70-g09d2


From 09f0551d20ddf6d22c333adcc59f2b1148734273 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 20 Jun 2005 18:44:37 +0100
Subject: [PATCH] ARM: Add iomap support for ARM

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig      |  4 ----
 arch/arm/mm/ioremap.c | 47 +++++++++++++++++++++++++++++++++++++++++++++++
 include/asm-arm/io.h  | 27 +++++++++++++++++++++++++++
 3 files changed, 74 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 475950c8a83..1a1773f8139 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -67,10 +67,6 @@ config GENERIC_BUST_SPINLOCK
 config GENERIC_ISA_DMA
 	bool
 
-config GENERIC_IOMAP
-	bool
-	default y
-
 config FIQ
 	bool
 
diff --git a/arch/arm/mm/ioremap.c b/arch/arm/mm/ioremap.c
index 00bb8fd37a5..7110e54182b 100644
--- a/arch/arm/mm/ioremap.c
+++ b/arch/arm/mm/ioremap.c
@@ -170,3 +170,50 @@ void __iounmap(void __iomem *addr)
 	vfree((void *) (PAGE_MASK & (unsigned long) addr));
 }
 EXPORT_SYMBOL(__iounmap);
+
+#ifdef __io
+void __iomem *ioport_map(unsigned long port, unsigned int nr)
+{
+	return __io(port);
+}
+EXPORT_SYMBOL(ioport_map);
+
+void ioport_unmap(void __iomem *addr)
+{
+}
+EXPORT_SYMBOL(ioport_unmap);
+#endif
+
+#ifdef CONFIG_PCI
+#include <linux/pci.h>
+#include <linux/ioport.h>
+
+void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen)
+{
+	unsigned long start = pci_resource_start(dev, bar);
+	unsigned long len   = pci_resource_len(dev, bar);
+	unsigned long flags = pci_resource_flags(dev, bar);
+
+	if (!len || !start)
+		return NULL;
+	if (maxlen && len > maxlen)
+		len = maxlen;
+	if (flags & IORESOURCE_IO)
+		return ioport_map(start, len);
+	if (flags & IORESOURCE_MEM) {
+		if (flags & IORESOURCE_CACHEABLE)
+			return ioremap(start, len);
+		return ioremap_nocache(start, len);
+	}
+	return NULL;
+}
+EXPORT_SYMBOL(pci_iomap);
+
+void pci_iounmap(struct pci_dev *dev, void __iomem *addr)
+{
+	if ((unsigned long)addr >= VMALLOC_START &&
+	    (unsigned long)addr < VMALLOC_END)
+		iounmap(addr);
+}
+EXPORT_SYMBOL(pci_iounmap);
+#endif
diff --git a/include/asm-arm/io.h b/include/asm-arm/io.h
index 658ffa384fd..08a46302d26 100644
--- a/include/asm-arm/io.h
+++ b/include/asm-arm/io.h
@@ -272,6 +272,33 @@ extern void __iounmap(void __iomem *addr);
 #define iounmap(cookie)			__arch_iounmap(cookie)
 #endif
 
+/*
+ * io{read,write}{8,16,32} macros
+ */
+#define ioread8(p)	({ unsigned int __v = __raw_readb(p); __v; })
+#define ioread16(p)	({ unsigned int __v = le16_to_cpu(__raw_readw(p)); __v; })
+#define ioread32(p)	({ unsigned int __v = le32_to_cpu(__raw_readl(p)); __v; })
+
+#define iowrite8(v,p)	__raw_writeb(v, p)
+#define iowrite16(v,p)	__raw_writew(cpu_to_le16(v), p)
+#define iowrite32(v,p)	__raw_writel(cpu_to_le32(v), p)
+
+#define ioread8_rep(p,d,c)	__raw_readsb(p,d,c)
+#define ioread16_rep(p,d,c)	__raw_readsw(p,d,c)
+#define ioread32_rep(p,d,c)	__raw_readsl(p,d,c)
+
+#define iowrite8_rep(p,s,c)	__raw_writesb(p,s,c)
+#define iowrite16_rep(p,s,c)	__raw_writesw(p,s,c)
+#define iowrite32_rep(p,s,c)	__raw_writesl(p,s,c)
+
+extern void __iomem *ioport_map(unsigned long port, unsigned int nr);
+extern void ioport_unmap(void __iomem *addr);
+
+struct pci_dev;
+
+extern void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen);
+extern void pci_iounmap(struct pci_dev *dev, void __iomem *addr);
+
 /*
  * can the hardware map this into one segment or not, given no other
  * constraints.
-- 
cgit v1.2.3-70-g09d2


From 038c5b602524b33447008492e932cdd0a1e806c9 Mon Sep 17 00:00:00 2001
From: Bellido Nicolas <nb-ml@be.rmk.(none)>
Date: Mon, 20 Jun 2005 18:51:05 +0100
Subject: [PATCH] ARM: 2686/2: AAEC-2000 Core support

Patch from Bellido Nicolas

Core support for AAEC-2000 based platforms.
This is an updated version of the previous patch, and takes
into account Russell's comments.
AAED-2000 default configuration will follow as soon
as some problems with the bootloader are sorted out...

Signed-off-by: Nicolas Bellido
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig                            |   7 ++
 arch/arm/Makefile                           |   1 +
 arch/arm/mach-aaec2000/Kconfig              |  11 ++
 arch/arm/mach-aaec2000/Makefile             |   9 ++
 arch/arm/mach-aaec2000/aaed2000.c           |  48 +++++++++
 arch/arm/mach-aaec2000/core.c               | 157 ++++++++++++++++++++++++++++
 arch/arm/mach-aaec2000/core.h               |  16 +++
 arch/arm/mm/Kconfig                         |   2 +-
 include/asm-arm/arch-aaec2000/aaec2000.h    | 151 ++++++++++++++++++++++++++
 include/asm-arm/arch-aaec2000/debug-macro.S |  36 +++++++
 include/asm-arm/arch-aaec2000/dma.h         |  17 +++
 include/asm-arm/arch-aaec2000/entry-macro.S |  33 ++++++
 include/asm-arm/arch-aaec2000/hardware.h    |  49 +++++++++
 include/asm-arm/arch-aaec2000/io.h          |  19 ++++
 include/asm-arm/arch-aaec2000/irqs.h        |  46 ++++++++
 include/asm-arm/arch-aaec2000/memory.h      |  73 +++++++++++++
 include/asm-arm/arch-aaec2000/param.h       |  15 +++
 include/asm-arm/arch-aaec2000/system.h      |  24 +++++
 include/asm-arm/arch-aaec2000/timex.h       |  18 ++++
 include/asm-arm/arch-aaec2000/uncompress.h  |  47 +++++++++
 include/asm-arm/arch-aaec2000/vmalloc.h     |  16 +++
 21 files changed, 794 insertions(+), 1 deletion(-)
 create mode 100644 arch/arm/mach-aaec2000/Kconfig
 create mode 100644 arch/arm/mach-aaec2000/Makefile
 create mode 100644 arch/arm/mach-aaec2000/aaed2000.c
 create mode 100644 arch/arm/mach-aaec2000/core.c
 create mode 100644 arch/arm/mach-aaec2000/core.h
 create mode 100644 include/asm-arm/arch-aaec2000/aaec2000.h
 create mode 100644 include/asm-arm/arch-aaec2000/debug-macro.S
 create mode 100644 include/asm-arm/arch-aaec2000/dma.h
 create mode 100644 include/asm-arm/arch-aaec2000/entry-macro.S
 create mode 100644 include/asm-arm/arch-aaec2000/hardware.h
 create mode 100644 include/asm-arm/arch-aaec2000/io.h
 create mode 100644 include/asm-arm/arch-aaec2000/irqs.h
 create mode 100644 include/asm-arm/arch-aaec2000/memory.h
 create mode 100644 include/asm-arm/arch-aaec2000/param.h
 create mode 100644 include/asm-arm/arch-aaec2000/system.h
 create mode 100644 include/asm-arm/arch-aaec2000/timex.h
 create mode 100644 include/asm-arm/arch-aaec2000/uncompress.h
 create mode 100644 include/asm-arm/arch-aaec2000/vmalloc.h

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 1a1773f8139..efdb12d7356 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -198,6 +198,11 @@ config ARCH_H720X
 	help
 	  This enables support for systems based on the Hynix HMS720x
 
+config ARCH_AAEC2000
+	bool "Agilent AAEC-2000 based"
+	help
+	  This enables support for systems based on the Agilent AAEC-2000
+
 endchoice
 
 source "arch/arm/mach-clps711x/Kconfig"
@@ -230,6 +235,8 @@ source "arch/arm/mach-h720x/Kconfig"
 
 source "arch/arm/mach-versatile/Kconfig"
 
+source "arch/arm/mach-aaec2000/Kconfig"
+
 # Definitions to make life easier
 config ARCH_ACORN
 	bool
diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index 2277e3d179c..8330495e244 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -97,6 +97,7 @@ textaddr-$(CONFIG_ARCH_FORTUNET)   := 0xc0008000
  machine-$(CONFIG_ARCH_VERSATILE)  := versatile
  machine-$(CONFIG_ARCH_IMX)	   := imx
  machine-$(CONFIG_ARCH_H720X)	   := h720x
+ machine-$(CONFIG_ARCH_AAEC2000)   := aaec2000
 
 ifeq ($(CONFIG_ARCH_EBSA110),y)
 # This is what happens if you forget the IOCS16 line.
diff --git a/arch/arm/mach-aaec2000/Kconfig b/arch/arm/mach-aaec2000/Kconfig
new file mode 100644
index 00000000000..5e4bef93754
--- /dev/null
+++ b/arch/arm/mach-aaec2000/Kconfig
@@ -0,0 +1,11 @@
+if ARCH_AAEC2000
+
+menu "Agilent AAEC-2000 Implementations"
+
+config MACH_AAED2000
+	bool "Agilent AAED-2000 Development Platform"
+	select CPU_ARM920T
+
+endmenu
+
+endif
diff --git a/arch/arm/mach-aaec2000/Makefile b/arch/arm/mach-aaec2000/Makefile
new file mode 100644
index 00000000000..20ec83896c3
--- /dev/null
+++ b/arch/arm/mach-aaec2000/Makefile
@@ -0,0 +1,9 @@
+#
+# Makefile for the linux kernel.
+#
+
+# Common support (must be linked before board specific support)
+obj-y += core.o
+
+# Specific board support
+obj-$(CONFIG_MACH_AAED2000) += aaed2000.o
diff --git a/arch/arm/mach-aaec2000/aaed2000.c b/arch/arm/mach-aaec2000/aaed2000.c
new file mode 100644
index 00000000000..5417ca3f462
--- /dev/null
+++ b/arch/arm/mach-aaec2000/aaed2000.c
@@ -0,0 +1,48 @@
+/*
+ *  linux/arch/arm/mach-aaec2000/aaed2000.c
+ *
+ *  Support for the Agilent AAED-2000 Development Platform.
+ *
+ *  Copyright (c) 2005 Nicolas Bellido Y Ortega
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ *
+ */
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/device.h>
+#include <linux/major.h>
+#include <linux/interrupt.h>
+
+#include <asm/setup.h>
+#include <asm/memory.h>
+#include <asm/mach-types.h>
+#include <asm/hardware.h>
+#include <asm/irq.h>
+
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+#include <asm/mach/irq.h>
+
+#include "core.h"
+
+static void __init aaed2000_init_irq(void)
+{
+	aaec2000_init_irq();
+}
+
+static void __init aaed2000_map_io(void)
+{
+	aaec2000_map_io();
+}
+
+MACHINE_START(AAED2000, "Agilent AAED-2000 Development Platform")
+	MAINTAINER("Nicolas Bellido Y Ortega")
+	BOOT_MEM(0xf0000000, PIO_BASE, VIO_BASE)
+	MAPIO(aaed2000_map_io)
+	INITIRQ(aaed2000_init_irq)
+	.timer		= &aaec2000_timer,
+MACHINE_END
diff --git a/arch/arm/mach-aaec2000/core.c b/arch/arm/mach-aaec2000/core.c
new file mode 100644
index 00000000000..fc145b3768f
--- /dev/null
+++ b/arch/arm/mach-aaec2000/core.c
@@ -0,0 +1,157 @@
+/*
+ *  linux/arch/arm/mach-aaec2000/core.c
+ *
+ *  Code common to all AAEC-2000 machines
+ *
+ *  Copyright (c) 2005 Nicolas Bellido Y Ortega
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/list.h>
+#include <linux/errno.h>
+#include <linux/interrupt.h>
+#include <linux/timex.h>
+#include <linux/signal.h>
+
+#include <asm/hardware.h>
+#include <asm/irq.h>
+
+#include <asm/mach/irq.h>
+#include <asm/mach/time.h>
+#include <asm/mach/map.h>
+
+/*
+ * Common I/O mapping:
+ *
+ * Static virtual address mappings are as follow:
+ *
+ * 0xf8000000-0xf8001ffff: Devices connected to APB bus
+ * 0xf8002000-0xf8003ffff: Devices connected to AHB bus
+ *
+ * Below 0xe8000000 is reserved for vm allocation.
+ *
+ * The machine specific code must provide the extra mapping beside the
+ * default mapping provided here.
+ */
+static struct map_desc standard_io_desc[] __initdata = {
+ /* virtual         physical       length           type */
+  { VIO_APB_BASE,   PIO_APB_BASE,  IO_APB_LENGTH,   MT_DEVICE },
+  { VIO_AHB_BASE,   PIO_AHB_BASE,  IO_AHB_LENGTH,   MT_DEVICE }
+};
+
+void __init aaec2000_map_io(void)
+{
+	iotable_init(standard_io_desc, ARRAY_SIZE(standard_io_desc));
+}
+
+/*
+ * Interrupt handling routines
+ */
+static void aaec2000_int_ack(unsigned int irq)
+{
+	IRQ_INTSR = 1 << irq;
+}
+
+static void aaec2000_int_mask(unsigned int irq)
+{
+	IRQ_INTENC |= (1 << irq);
+}
+
+static void aaec2000_int_unmask(unsigned int irq)
+{
+	IRQ_INTENS |= (1 << irq);
+}
+
+static struct irqchip aaec2000_irq_chip = {
+	.ack	= aaec2000_int_ack,
+	.mask	= aaec2000_int_mask,
+	.unmask	= aaec2000_int_unmask,
+};
+
+void __init aaec2000_init_irq(void)
+{
+	unsigned int i;
+
+	for (i = 0; i < NR_IRQS; i++) {
+		set_irq_handler(i, do_level_IRQ);
+		set_irq_chip(i, &aaec2000_irq_chip);
+		set_irq_flags(i, IRQF_VALID);
+	}
+
+	/* Disable all interrupts */
+	IRQ_INTENC = 0xffffffff;
+
+	/* Clear any pending interrupts */
+	IRQ_INTSR = IRQ_INTSR;
+}
+
+/*
+ * Time keeping
+ */
+/* IRQs are disabled before entering here from do_gettimeofday() */
+static unsigned long aaec2000_gettimeoffset(void)
+{
+	unsigned long ticks_to_match, elapsed, usec;
+
+	/* Get ticks before next timer match */
+	ticks_to_match = TIMER1_LOAD - TIMER1_VAL;
+
+	/* We need elapsed ticks since last match */
+	elapsed = LATCH - ticks_to_match;
+
+	/* Now, convert them to usec */
+	usec = (unsigned long)(elapsed * (tick_nsec / 1000))/LATCH;
+
+	return usec;
+}
+
+/* We enter here with IRQs enabled */
+static irqreturn_t
+aaec2000_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+{
+	/* TODO: Check timer accuracy */
+	write_seqlock(&xtime_lock);
+
+	timer_tick(regs);
+	TIMER1_CLEAR = 1;
+
+	write_sequnlock(&xtime_lock);
+
+	return IRQ_HANDLED;
+}
+
+static struct irqaction aaec2000_timer_irq = {
+	.name		= "AAEC-2000 Timer Tick",
+	.flags		= SA_INTERRUPT,
+	.handler	= aaec2000_timer_interrupt
+};
+
+static void __init aaec2000_timer_init(void)
+{
+	/* Disable timer 1 */
+	TIMER1_CTRL = 0;
+
+	/* We have somehow to generate a 100Hz clock.
+	 * We then use the 508KHz timer in periodic mode.
+	 */
+	TIMER1_LOAD = LATCH;
+	TIMER1_CLEAR = 1; /* Clear interrupt */
+
+	setup_irq(INT_TMR1_OFL, &aaec2000_timer_irq);
+
+	TIMER1_CTRL = TIMER_CTRL_ENABLE |
+	                TIMER_CTRL_PERIODIC |
+	                TIMER_CTRL_CLKSEL_508K;
+}
+
+struct sys_timer aaec2000_timer = {
+	.init		= aaec2000_timer_init,
+	.offset		= aaec2000_gettimeoffset,
+};
+
diff --git a/arch/arm/mach-aaec2000/core.h b/arch/arm/mach-aaec2000/core.h
new file mode 100644
index 00000000000..91893d848c1
--- /dev/null
+++ b/arch/arm/mach-aaec2000/core.h
@@ -0,0 +1,16 @@
+/*
+ *  linux/arch/arm/mach-aaec2000/core.h
+ *
+ *  Copyright (c) 2005 Nicolas Bellido Y Ortega
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ *
+ */
+
+struct sys_timer;
+
+extern struct sys_timer aaec2000_timer;
+extern void __init aaec2000_map_io(void);
+extern void __init aaec2000_init_irq(void);
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index 3fefb43c67f..95606b4a3ba 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -62,7 +62,7 @@ config CPU_ARM720T
 # ARM920T
 config CPU_ARM920T
 	bool "Support ARM920T processor" if !ARCH_S3C2410
-	depends on ARCH_INTEGRATOR || ARCH_S3C2410 || ARCH_IMX
+	depends on ARCH_INTEGRATOR || ARCH_S3C2410 || ARCH_IMX || ARCH_AAEC2000
 	default y if ARCH_S3C2410
 	select CPU_32v4
 	select CPU_ABRT_EV4T
diff --git a/include/asm-arm/arch-aaec2000/aaec2000.h b/include/asm-arm/arch-aaec2000/aaec2000.h
new file mode 100644
index 00000000000..0e9b7e18af0
--- /dev/null
+++ b/include/asm-arm/arch-aaec2000/aaec2000.h
@@ -0,0 +1,151 @@
+/*
+ *  linux/include/asm-arm/arch-aaec2000/aaec2000.h
+ *
+ *  AAEC-2000 registers definition
+ *
+ *  Copyright (c) 2005 Nicolas Bellido Y Ortega
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#ifndef __ASM_ARCH_AAEC2000_H
+#define __ASM_ARCH_AAEC2000_H
+
+#ifndef __ASM_ARCH_HARDWARE_H
+#error You must include hardware.h not this file
+#endif /* __ASM_ARCH_HARDWARE_H */
+
+/* Interrupt controller */
+#define IRQ_BASE	__REG(0x80000500)
+#define IRQ_INTSR	__REG(0x80000500)	/* Int Status Register */
+#define IRQ_INTRSR	__REG(0x80000504)	/* Int Raw (unmasked) Status */
+#define IRQ_INTENS	__REG(0x80000508)	/* Int Enable Set */
+#define IRQ_INTENC	__REG(0x8000050c)	/* Int Enable Clear */
+
+/* UART 1 */
+#define UART1_BASE	__REG(0x80000600)
+#define UART1_DR	__REG(0x80000600) /* Data/FIFO Register */
+#define UART1_LCR	__REG(0x80000604) /* Link Control Register */
+#define UART1_BRCR	__REG(0x80000608) /* Baud Rate Control Register */
+#define UART1_CR	__REG(0x8000060c) /* Control Register */
+#define UART1_SR	__REG(0x80000610) /* Status Register */
+#define UART1_INT	__REG(0x80000614) /* Interrupt Status Register */
+#define UART1_INTM	__REG(0x80000618) /* Interrupt Mask Register */
+#define UART1_INTRES	__REG(0x8000061c) /* Int Result (masked status) Register */
+
+/* UART 2 */
+#define UART2_BASE	__REG(0x80000700)
+#define UART2_DR	__REG(0x80000700) /* Data/FIFO Register */
+#define UART2_LCR	__REG(0x80000704) /* Link Control Register */
+#define UART2_BRCR	__REG(0x80000708) /* Baud Rate Control Register */
+#define UART2_CR	__REG(0x8000070c) /* Control Register */
+#define UART2_SR	__REG(0x80000710) /* Status Register */
+#define UART2_INT	__REG(0x80000714) /* Interrupt Status Register */
+#define UART2_INTM	__REG(0x80000718) /* Interrupt Mask Register */
+#define UART2_INTRES	__REG(0x8000071c) /* Int Result (masked status) Register */
+
+/* UART 3 */
+#define UART3_BASE	__REG(0x80000800)
+#define UART3_DR	__REG(0x80000800) /* Data/FIFO Register */
+#define UART3_LCR	__REG(0x80000804) /* Link Control Register */
+#define UART3_BRCR	__REG(0x80000808) /* Baud Rate Control Register */
+#define UART3_CR	__REG(0x8000080c) /* Control Register */
+#define UART3_SR	__REG(0x80000810) /* Status Register */
+#define UART3_INT	__REG(0x80000814) /* Interrupt Status Register */
+#define UART3_INTM	__REG(0x80000818) /* Interrupt Mask Register */
+#define UART3_INTRES	__REG(0x8000081c) /* Int Result (masked status) Register */
+
+/* These are used in some places */
+#define _UART1_BASE __PREG(UART1_BASE)
+#define _UART2_BASE __PREG(UART2_BASE)
+#define _UART3_BASE __PREG(UART3_BASE)
+
+/* UART Registers Offsets */
+#define UART_DR		0x00
+#define UART_LCR	0x04
+#define UART_BRCR	0x08
+#define UART_CR		0x0c
+#define UART_SR		0x10
+#define UART_INT	0x14
+#define UART_INTM	0x18
+#define UART_INTRES	0x1c
+
+/* UART_LCR Bitmask */
+#define UART_LCR_BRK	(1 << 0) /* Send Break */
+#define UART_LCR_PEN	(1 << 1) /* Parity Enable */
+#define UART_LCR_EP	(1 << 2) /* Even/Odd Parity */
+#define UART_LCR_S2	(1 << 3) /* One/Two Stop bits */
+#define UART_LCR_FIFO	(1 << 4) /* FIFO Enable */
+#define UART_LCR_WL5	(0 << 5) /* Word Length - 5 bits */
+#define UART_LCR_WL6	(1 << 5) /* Word Length - 6 bits */
+#define UART_LCR_WL7	(1 << 6) /* Word Length - 7 bits */
+#define UART_LCR_WL8	(1 << 7) /* Word Length - 8 bits */
+
+/* UART_CR Bitmask */
+#define UART_CR_EN	(1 << 0) /* UART Enable */
+#define UART_CR_SIR	(1 << 1) /* IrDA SIR Enable */
+#define UART_CR_SIRLP	(1 << 2) /* Low Power IrDA Enable */
+#define UART_CR_RXP	(1 << 3) /* Receive Pin Polarity */
+#define UART_CR_TXP	(1 << 4) /* Transmit Pin Polarity */
+#define UART_CR_MXP	(1 << 5) /* Modem Pin Polarity */
+#define UART_CR_LOOP	(1 << 6) /* Loopback Mode */
+
+/* UART_SR Bitmask */
+#define UART_SR_CTS	(1 << 0) /* Clear To Send Status */
+#define UART_SR_DSR	(1 << 1) /* Data Set Ready Status */
+#define UART_SR_DCD	(1 << 2) /* Data Carrier Detect Status */
+#define UART_SR_TxBSY	(1 << 3) /* Transmitter Busy Status */
+#define UART_SR_RxFE	(1 << 4) /* Receive FIFO Empty Status */
+#define UART_SR_TxFF	(1 << 5) /* Transmit FIFO Full Status */
+#define UART_SR_RxFF	(1 << 6) /* Receive FIFO Full Status */
+#define UART_SR_TxFE	(1 << 7) /* Transmit FIFO Empty Status */
+
+/* UART_INT Bitmask */
+#define UART_INT_RIS	(1 << 0) /* Rx Interrupt */
+#define UART_INT_TIS	(1 << 1) /* Tx Interrupt */
+#define UART_INT_MIS	(1 << 2) /* Modem Interrupt */
+#define UART_INT_RTIS	(1 << 3) /* Receive Timeout Interrupt */
+
+/* Timer 1 */
+#define TIMER1_BASE	__REG(0x80000c00)
+#define TIMER1_LOAD	__REG(0x80000c00)	/* Timer 1 Load Register */
+#define TIMER1_VAL	__REG(0x80000c04)	/* Timer 1 Value Register */
+#define TIMER1_CTRL	__REG(0x80000c08)	/* Timer 1 Control Register */
+#define TIMER1_CLEAR	__REG(0x80000c0c)	/* Timer 1 Clear Register */
+
+/* Timer 2 */
+#define TIMER2_BASE	__REG(0x80000d00)
+#define TIMER2_LOAD	__REG(0x80000d00)	/* Timer 2 Load Register */
+#define TIMER2_VAL	__REG(0x80000d04)	/* Timer 2 Value Register */
+#define TIMER2_CTRL	__REG(0x80000d08)	/* Timer 2 Control Register */
+#define TIMER2_CLEAR	__REG(0x80000d0c)	/* Timer 2 Clear Register */
+
+/* Timer 3 */
+#define TIMER3_BASE	__REG(0x80000e00)
+#define TIMER3_LOAD	__REG(0x80000e00)	/* Timer 3 Load Register */
+#define TIMER3_VAL	__REG(0x80000e04)	/* Timer 3 Value Register */
+#define TIMER3_CTRL	__REG(0x80000e08)	/* Timer 3 Control Register */
+#define TIMER3_CLEAR	__REG(0x80000e0c)	/* Timer 3 Clear Register */
+
+/* Timer Control register bits */
+#define TIMER_CTRL_ENABLE	(1 << 7) /* Enable (Start° Timer */
+#define TIMER_CTRL_PERIODIC	(1 << 6) /* Periodic Running Mode */
+#define TIMER_CTRL_FREE_RUNNING (0 << 6) /* Normal Running Mode */
+#define TIMER_CTRL_CLKSEL_508K	(1 << 3) /* 508KHz Clock select (Timer 1, 2) */
+#define TIMER_CTRL_CLKSEL_2K	(0 << 3) /* 2KHz Clock Select (Timer 1, 2)*/
+
+/* Power and State Control */
+#define POWER_BASE	__REG(0x80000400)
+#define POWER_PWRSR	__REG(0x80000400) /* Power Status Register */
+#define POWER_PWRCNT	__REG(0x80000404) /* Power/Clock control */
+#define POWER_HALT	__REG(0x80000408) /* Power Idle Mode */
+#define POWER_STDBY	__REG(0x8000040c) /* Power Standby Mode */
+#define POWER_BLEOI	__REG(0x80000410) /* Battery Low End of Interrupt */
+#define POWER_MCEOI	__REG(0x80000414) /* Media Changed EoI */
+#define POWER_TEOI	__REG(0x80000418) /* Tick EoI */
+#define POWER_STFCLR	__REG(0x8000041c) /* NbFlg, RSTFlg, PFFlg, CLDFlg Clear */
+#define POWER_CLKSET	__REG(0x80000420) /* Clock Speed Control */
+
+#endif /* __ARM_ARCH_AAEC2000_H */
diff --git a/include/asm-arm/arch-aaec2000/debug-macro.S b/include/asm-arm/arch-aaec2000/debug-macro.S
new file mode 100644
index 00000000000..e4f1fa539a7
--- /dev/null
+++ b/include/asm-arm/arch-aaec2000/debug-macro.S
@@ -0,0 +1,36 @@
+/*  linux/include/asm-arm/arch-aaec2000/debug-macro.S
+ *
+ *  Debugging macro include header
+ *
+ *  Copyright (c) 2005 Nicolas Bellido Y Ortega
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+		.macro	addruart,rx
+		mrc	p15, 0, \rx, c1, c0
+		tst	\rx, #1			@ MMU enabled?
+		moveq	\rx, #0x80000000		@ physical
+		movne	\rx, #io_p2v(0x80000000)	@ virtual
+		orr	\rx, \rx, #0x00000800
+		.endm
+
+		.macro	senduart,rd,rx
+		str	\rd, [\rx, #0]
+		.endm
+
+		.macro	busyuart,rd,rx
+1002:		ldr	\rd, [\rx, #0x10]
+		tst	\rd, #(1 << 7)
+		beq	1002b
+		.endm
+
+		.macro	waituart,rd,rx
+#if 0
+1001:		ldr	\rd, [\rx, #0x10]
+		tst	\rd, #(1 << 5)
+		beq	1001b
+#endif
+		.endm
diff --git a/include/asm-arm/arch-aaec2000/dma.h b/include/asm-arm/arch-aaec2000/dma.h
new file mode 100644
index 00000000000..28c890b4a1d
--- /dev/null
+++ b/include/asm-arm/arch-aaec2000/dma.h
@@ -0,0 +1,17 @@
+/*
+ *  linux/include/asm-arm/arch-aaec2000/dma.h
+ *
+ *  Copyright (c) 2005 Nicolas Bellido Y Ortega
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#ifndef __ASM_ARCH_DMA_H
+#define __ASM_ARCH_DMA_H
+
+#define MAX_DMA_ADDRESS         0xffffffff
+#define MAX_DMA_CHANNELS        0
+
+#endif
diff --git a/include/asm-arm/arch-aaec2000/entry-macro.S b/include/asm-arm/arch-aaec2000/entry-macro.S
new file mode 100644
index 00000000000..df31313ab07
--- /dev/null
+++ b/include/asm-arm/arch-aaec2000/entry-macro.S
@@ -0,0 +1,33 @@
+/*
+ *  linux/include/asm-arm/arch-aaec2000/entry-macro.S
+ *
+ *  Low-level IRQ helper for aaec-2000 based platforms
+ *
+ *  Copyright (c) 2005 Nicolas Bellido Y Ortega
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ *
+ */
+
+		.macro	disable_fiq
+		.endm
+
+		.macro	get_irqnr_and_base, irqnr, irqstat, base, tmp
+		mov	r4, #0xf8000000
+		add	r4, r4, #0x00000500
+		mov	\base, r4
+		ldr	\irqstat, [\base, #0]
+		cmp	\irqstat, #0
+		bne	1001f
+		ldr	\irqnr, =NR_IRQS+1
+		b       1003f
+1001:		mov	\irqnr, #0
+1002:		ands	\tmp, \irqstat, #1
+		mov	\irqstat, \irqstat, LSR #1
+		add	\irqnr, \irqnr, #1
+		beq	1002b
+		sub	\irqnr, \irqnr, #1
+1003:
+		.endm
diff --git a/include/asm-arm/arch-aaec2000/hardware.h b/include/asm-arm/arch-aaec2000/hardware.h
new file mode 100644
index 00000000000..4c37219e030
--- /dev/null
+++ b/include/asm-arm/arch-aaec2000/hardware.h
@@ -0,0 +1,49 @@
+/*
+ *  linux/include/asm-arm/arch-aaec2000/hardware.h
+ *
+ *  Copyright (c) 2005 Nicolas Bellido Y Ortega
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#ifndef __ASM_ARCH_HARDWARE_H
+#define __ASM_ARCH_HARDWARE_H
+
+#include <linux/config.h>
+
+/* The kernel is loaded at physical address 0xf8000000.
+ * We map the IO space a bit after
+ */
+#define PIO_APB_BASE	0x80000000
+#define VIO_APB_BASE	0xf8000000
+#define IO_APB_LENGTH	0x2000
+#define PIO_AHB_BASE	0x80002000
+#define VIO_AHB_BASE	0xf8002000
+#define IO_AHB_LENGTH	0x2000
+
+#define VIO_BASE    VIO_APB_BASE
+#define PIO_BASE    PIO_APB_BASE
+
+#define io_p2v(x) ( (x) - PIO_BASE + VIO_BASE )
+#define io_v2p(x) ( (x) + PIO_BASE - VIO_BASE )
+
+#ifndef __ASSEMBLY__
+
+#include <asm/types.h>
+
+/* FIXME: Is it needed to optimize this a la pxa ?? */
+#define __REG(x)    (*((volatile u32 *)io_p2v(x)))
+#define __PREG(x)   (io_v2p((u32)&(x)))
+
+#else /* __ASSEMBLY__ */
+
+#define __REG(x)    io_p2v(x)
+#define __PREG(x)   io_v2p(x)
+
+#endif
+
+#include "aaec2000.h"
+
+#endif /* __ASM_ARCH_HARDWARE_H */
diff --git a/include/asm-arm/arch-aaec2000/io.h b/include/asm-arm/arch-aaec2000/io.h
new file mode 100644
index 00000000000..c58a8d10425
--- /dev/null
+++ b/include/asm-arm/arch-aaec2000/io.h
@@ -0,0 +1,19 @@
+/*
+ *  linux/include/asm-arm/arch-aaec2000/io.h
+ *
+ *  Copied from asm/arch/sa1100/io.h
+ */
+#ifndef __ASM_ARM_ARCH_IO_H
+#define __ASM_ARM_ARCH_IO_H
+
+#define IO_SPACE_LIMIT 0xffffffff
+
+/*
+ * We don't actually have real ISA nor PCI buses, but there is so many
+ * drivers out there that might just work if we fake them...
+ */
+#define __io(a)			((void __iomem *)(a))
+#define __mem_pci(a)		(a)
+#define __mem_isa(a)		(a)
+
+#endif
diff --git a/include/asm-arm/arch-aaec2000/irqs.h b/include/asm-arm/arch-aaec2000/irqs.h
new file mode 100644
index 00000000000..de252220e80
--- /dev/null
+++ b/include/asm-arm/arch-aaec2000/irqs.h
@@ -0,0 +1,46 @@
+/*
+ *  linux/include/asm-arm/arch-aaec2000/irqs.h
+ *
+ *  Copyright (c) 2005 Nicolas Bellido Y Ortega
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#ifndef __ASM_ARCH_IRQS_H
+#define __ASM_ARCH_IRQS_H
+
+
+#define INT_GPIOF0_FIQ	0  /* External GPIO Port F O Fast Interrupt Input */
+#define INT_BL_FIQ	1  /* Battery Low Fast Interrupt */
+#define INT_WE_FIQ	2  /* Watchdog Expired Fast Interrupt */
+#define INT_MV_FIQ	3  /* Media Changed Interrupt */
+#define INT_SC		4  /* Sound Codec Interrupt */
+#define INT_GPIO1	5  /* GPIO Port F Configurable Int 1 */
+#define INT_GPIO2	6  /* GPIO Port F Configurable Int 2 */
+#define INT_GPIO3	7  /* GPIO Port F Configurable Int 3 */
+#define INT_TMR1_OFL	8  /* Timer 1 Overflow Interrupt */
+#define INT_TMR2_OFL	9  /* Timer 2 Overflow Interrupt */
+#define INT_RTC_CM	10 /* RTC Compare Match Interrupt */
+#define INT_TICK	11 /* 64Hz Tick Interrupt */
+#define INT_UART1	12 /* UART1 Interrupt */
+#define INT_UART2	13 /* UART2 & Modem State Changed Interrupt */
+#define INT_LCD		14 /* LCD Interrupt */
+#define INT_SSI		15 /* SSI End of Transfer Interrupt */
+#define INT_UART3	16 /* UART3 Interrupt */
+#define INT_SCI		17 /* SCI Interrupt */
+#define INT_AAC		18 /* Advanced Audio Codec Interrupt */
+#define INT_MMC		19 /* MMC Interrupt */
+#define INT_USB		20 /* USB Interrupt */
+#define INT_DMA		21 /* DMA Interrupt */
+#define INT_TMR3_UOFL	22 /* Timer 3 Underflow Interrupt */
+#define INT_GPIO4	23 /* GPIO Port F Configurable Int 4 */
+#define INT_GPIO5	24 /* GPIO Port F Configurable Int 4 */
+#define INT_GPIO6	25 /* GPIO Port F Configurable Int 4 */
+#define INT_GPIO7	26 /* GPIO Port F Configurable Int 4 */
+#define INT_BMI		27 /* BMI Interrupt */
+
+#define NR_IRQS		(INT_BMI + 1)
+
+#endif /* __ASM_ARCH_IRQS_H */
diff --git a/include/asm-arm/arch-aaec2000/memory.h b/include/asm-arm/arch-aaec2000/memory.h
new file mode 100644
index 00000000000..681b6a6171a
--- /dev/null
+++ b/include/asm-arm/arch-aaec2000/memory.h
@@ -0,0 +1,73 @@
+/*
+ *  linux/include/asm-arm/arch-aaec2000/memory.h
+ *
+ *  Copyright (c) 2005 Nicolas Bellido Y Ortega
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#ifndef __ASM_ARCH_MEMORY_H
+#define __ASM_ARCH_MEMORY_H
+
+#include <linux/config.h>
+
+#define PHYS_OFFSET	(0xf0000000UL)
+
+#define __virt_to_bus(x)	__virt_to_phys(x)
+#define __bus_to_virt(x)	__phys_to_virt(x)
+
+#ifdef CONFIG_DISCONTIGMEM
+
+/*
+ * The nodes are the followings:
+ *
+ *   node 0: 0xf000.0000 - 0xf3ff.ffff
+ *   node 1: 0xf400.0000 - 0xf7ff.ffff
+ *   node 2: 0xf800.0000 - 0xfbff.ffff
+ *   node 3: 0xfc00.0000 - 0xffff.ffff
+ */
+
+/*
+ * Given a kernel address, find the home node of the underlying memory.
+ */
+#define KVADDR_TO_NID(addr) \
+	(((unsigned long)(addr) - PAGE_OFFSET) >> NODE_MAX_MEM_SHIFT)
+
+/*
+ * Given a page frame number, convert it to a node id.
+ */
+#define PFN_TO_NID(pfn) \
+	(((pfn) - PHYS_PFN_OFFSET) >> (NODE_MAX_MEM_SHIFT - PAGE_SHIFT))
+
+/*
+ * Given a kaddr, ADDR_TO_MAPBASE finds the owning node of the memory
+ * and return the mem_map of that node.
+ */
+#define ADDR_TO_MAPBASE(kaddr)  NODE_MEM_MAP(KVADDR_TO_NID(kaddr))
+
+/*
+ * Given a page frame number, find the owning node of the memory
+ * and return the mem_map of that node.
+ */
+#define PFN_TO_MAPBASE(pfn)     NODE_MEM_MAP(PFN_TO_NID(pfn))
+
+/*
+ *  Given a kaddr, LOCAL_MEM_MAP finds the owning node of the memory
+ *  and returns the index corresponding to the appropriate page in the
+ *  node's mem_map.
+ */
+#define LOCAL_MAP_NR(addr) \
+        (((unsigned long)(addr) & (NODE_MAX_MEM_SIZE - 1)) >> PAGE_SHIFT)
+
+#define NODE_MAX_MEM_SHIFT	26
+#define NODE_MAX_MEM_SIZE	(1 << NODE_MAX_MEM_SHIFT)
+
+#else
+
+#define PFN_TO_NID(addr)	(0)
+
+#endif /* CONFIG_DISCONTIGMEM */
+
+#endif /* __ASM_ARCH_MEMORY_H */
diff --git a/include/asm-arm/arch-aaec2000/param.h b/include/asm-arm/arch-aaec2000/param.h
new file mode 100644
index 00000000000..139936c2faf
--- /dev/null
+++ b/include/asm-arm/arch-aaec2000/param.h
@@ -0,0 +1,15 @@
+/*
+ *  linux/include/asm-arm/arch-aaec2000/param.h
+ *
+ *  Copyright (c) 2005 Nicolas Bellido Y Ortega
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#ifndef __ASM_ARCH_PARAM_H
+#define __ASM_ARCH_PARAM_H
+
+#endif /* __ASM_ARCH_PARAM_H */
+
diff --git a/include/asm-arm/arch-aaec2000/system.h b/include/asm-arm/arch-aaec2000/system.h
new file mode 100644
index 00000000000..08de97b407a
--- /dev/null
+++ b/include/asm-arm/arch-aaec2000/system.h
@@ -0,0 +1,24 @@
+/*
+ *  linux/include/asm-arm/arch-aaed2000/system.h
+ *
+ *  Copyright (c) 2005 Nicolas Bellido Y Ortega
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#ifndef __ASM_ARCH_SYSTEM_H
+#define __ASM_ARCH_SYSTEM_H
+
+static inline void arch_idle(void)
+{
+	cpu_do_idle();
+}
+
+static inline void arch_reset(char mode)
+{
+	cpu_reset(0);
+}
+
+#endif /* __ASM_ARCH_SYSTEM_H */
diff --git a/include/asm-arm/arch-aaec2000/timex.h b/include/asm-arm/arch-aaec2000/timex.h
new file mode 100644
index 00000000000..f5708b38fb7
--- /dev/null
+++ b/include/asm-arm/arch-aaec2000/timex.h
@@ -0,0 +1,18 @@
+/*
+ *  linux/include/asm-arm/arch-aaec2000/timex.h
+ *
+ *  AAEC-2000 Architecture timex specification
+ *
+ *  Copyright (c) 2005 Nicolas Bellido Y Ortega
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#ifndef __ASM_ARCH_TIMEX_H
+#define __ASM_ARCH_TIMEX_H
+
+#define CLOCK_TICK_RATE		508000
+
+#endif /* __ASM_ARCH_TIMEX_H */
diff --git a/include/asm-arm/arch-aaec2000/uncompress.h b/include/asm-arm/arch-aaec2000/uncompress.h
new file mode 100644
index 00000000000..fff0c94b75c
--- /dev/null
+++ b/include/asm-arm/arch-aaec2000/uncompress.h
@@ -0,0 +1,47 @@
+/*
+ *  linux/include/asm-arm/arch-aaec2000/uncompress.h
+ *
+ *  Copyright (c) 2005 Nicolas Bellido Y Ortega
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#ifndef __ASM_ARCH_UNCOMPRESS_H
+#define __ASM_ARCH_UNCOMPRESS_H
+
+#include "hardware.h"
+
+#define UART(x)         (*(volatile unsigned long *)(serial_port + (x)))
+
+static void putstr( const char *s )
+{
+	unsigned long serial_port;
+        do {
+		serial_port = _UART3_BASE;
+		if (UART(UART_CR) & UART_CR_EN) break;
+		serial_port = _UART1_BASE;
+		if (UART(UART_CR) & UART_CR_EN) break;
+		serial_port = _UART2_BASE;
+		if (UART(UART_CR) & UART_CR_EN) break;
+		return;
+	} while (0);
+
+	for (; *s; s++) {
+		/* wait for space in the UART's transmitter */
+		while ((UART(UART_SR) & UART_SR_TxFF));
+		/* send the character out. */
+		UART(UART_DR) = *s;
+		/* if a LF, also do CR... */
+		if (*s == 10) {
+			while ((UART(UART_SR) & UART_SR_TxFF));
+			UART(UART_DR) = 13;
+		}
+	}
+}
+
+#define arch_decomp_setup()
+#define arch_decomp_wdog()
+
+#endif /* __ASM_ARCH_UNCOMPRESS_H */
diff --git a/include/asm-arm/arch-aaec2000/vmalloc.h b/include/asm-arm/arch-aaec2000/vmalloc.h
new file mode 100644
index 00000000000..ecb991e2e4f
--- /dev/null
+++ b/include/asm-arm/arch-aaec2000/vmalloc.h
@@ -0,0 +1,16 @@
+/*
+ *  linux/include/asm-arm/arch-aaec2000/vmalloc.h
+ *
+ *  Copyright (c) 2005 Nicolas Bellido Y Ortega
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#ifndef __ASM_ARCH_VMALLOC_H
+#define __ASM_ARCH_VMALLOC_H
+
+#define VMALLOC_END		(PAGE_OFFSET + 0x10000000)
+
+#endif /* __ASM_ARCH_VMALLOC_H */
-- 
cgit v1.2.3-70-g09d2


From c0da085ad2e6b1419b8a7439538f7f15eb5c4777 Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@com.rmk.(none)>
Date: Mon, 20 Jun 2005 18:51:06 +0100
Subject: [PATCH] ARM: 2693/1: Add PCI support for Versatile/PB

Patch from Catalin Marinas

This patch adds PCI support for the Versatile PB926 platform.

Signed-off-by: Colin King
Signed-off-by: Catalin Marinas
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig                          |   2 +-
 arch/arm/mach-versatile/Makefile          |   1 +
 arch/arm/mach-versatile/core.c            |  14 +-
 arch/arm/mach-versatile/pci.c             | 360 ++++++++++++++++++++++++++++++
 include/asm-arm/arch-versatile/hardware.h |  27 ++-
 include/asm-arm/arch-versatile/io.h       |   2 +-
 include/asm-arm/arch-versatile/platform.h |  17 +-
 7 files changed, 404 insertions(+), 19 deletions(-)
 create mode 100644 arch/arm/mach-versatile/pci.c

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index efdb12d7356..ee8a9ad7bbd 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -280,7 +280,7 @@ config ISA_DMA_API
 	default y
 
 config PCI
-	bool "PCI support" if ARCH_INTEGRATOR_AP
+	bool "PCI support" if ARCH_INTEGRATOR_AP || ARCH_VERSATILE_PB
 	help
 	  Find out whether you have a PCI motherboard. PCI is the name of a
 	  bus system, i.e. the way the CPU talks to the other stuff inside
diff --git a/arch/arm/mach-versatile/Makefile b/arch/arm/mach-versatile/Makefile
index 5d608837757..ba81e70ed81 100644
--- a/arch/arm/mach-versatile/Makefile
+++ b/arch/arm/mach-versatile/Makefile
@@ -5,3 +5,4 @@
 obj-y					:= core.o clock.o
 obj-$(CONFIG_ARCH_VERSATILE_PB)		+= versatile_pb.o
 obj-$(CONFIG_MACH_VERSATILE_AB)		+= versatile_ab.o
+obj-$(CONFIG_PCI)			+= pci.o
diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c
index 302c2a7b9b6..6a7cbea5e09 100644
--- a/arch/arm/mach-versatile/core.c
+++ b/arch/arm/mach-versatile/core.c
@@ -196,11 +196,15 @@ static struct map_desc versatile_io_desc[] __initdata = {
 #ifdef CONFIG_DEBUG_LL
  { IO_ADDRESS(VERSATILE_UART0_BASE), VERSATILE_UART0_BASE, SZ_4K,      MT_DEVICE },
 #endif
-#ifdef FIXME
- { PCI_MEMORY_VADDR,		     PHYS_PCI_MEM_BASE,    SZ_16M,     MT_DEVICE },
- { PCI_CONFIG_VADDR,		     PHYS_PCI_CONFIG_BASE, SZ_16M,     MT_DEVICE },
- { PCI_V3_VADDR,		     PHYS_PCI_V3_BASE,     SZ_512K,    MT_DEVICE },
- { PCI_IO_VADDR,		     PHYS_PCI_IO_BASE,     SZ_64K,     MT_DEVICE },
+#ifdef CONFIG_PCI
+ { IO_ADDRESS(VERSATILE_PCI_CORE_BASE), VERSATILE_PCI_CORE_BASE, SZ_4K, MT_DEVICE },
+ { VERSATILE_PCI_VIRT_BASE,          VERSATILE_PCI_BASE,   VERSATILE_PCI_BASE_SIZE, MT_DEVICE },
+ { VERSATILE_PCI_CFG_VIRT_BASE,      VERSATILE_PCI_CFG_BASE, VERSATILE_PCI_CFG_BASE_SIZE, MT_DEVICE },
+#if 0
+ { VERSATILE_PCI_VIRT_MEM_BASE0,     VERSATILE_PCI_MEM_BASE0, SZ_16M,  MT_DEVICE },
+ { VERSATILE_PCI_VIRT_MEM_BASE1,     VERSATILE_PCI_MEM_BASE1, SZ_16M,  MT_DEVICE },
+ { VERSATILE_PCI_VIRT_MEM_BASE2,     VERSATILE_PCI_MEM_BASE2, SZ_16M,  MT_DEVICE },
+#endif
 #endif
 };
 
diff --git a/arch/arm/mach-versatile/pci.c b/arch/arm/mach-versatile/pci.c
new file mode 100644
index 00000000000..d1565e851f0
--- /dev/null
+++ b/arch/arm/mach-versatile/pci.c
@@ -0,0 +1,360 @@
+/*
+ *  linux/arch/arm/mach-versatile/pci.c
+ *
+ * (C) Copyright Koninklijke Philips Electronics NV 2004. All rights reserved.
+ * You can redistribute and/or modify this software under the terms of version 2
+ * of the GNU General Public License as published by the Free Software Foundation.
+ * THIS SOFTWARE IS PROVIDED "AS IS" WITHOUT ANY WARRANTY; WITHOUT EVEN THE IMPLIED
+ * WARRANTY OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ * Koninklijke Philips Electronics nor its subsidiaries is obligated to provide any support for this software.
+ *
+ * ARM Versatile PCI driver.
+ *
+ * 14/04/2005 Initial version, colin.king@philips.com
+ *
+ */
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/ptrace.h>
+#include <linux/slab.h>
+#include <linux/ioport.h>
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/init.h>
+
+#include <asm/hardware.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/system.h>
+#include <asm/mach/pci.h>
+#include <asm/mach-types.h>
+
+/*
+ * these spaces are mapped using the following base registers:
+ *
+ * Usage Local Bus Memory         Base/Map registers used
+ *
+ * Mem   50000000 - 5FFFFFFF      LB_BASE0/LB_MAP0,  non prefetch
+ * Mem   60000000 - 6FFFFFFF      LB_BASE1/LB_MAP1,  prefetch
+ * IO    44000000 - 4FFFFFFF      LB_BASE2/LB_MAP2,  IO
+ * Cfg   42000000 - 42FFFFFF	  PCI config
+ *
+ */
+#define SYS_PCICTL			IO_ADDRESS(VERSATILE_SYS_PCICTL)
+#define PCI_IMAP0			IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x0)
+#define PCI_IMAP1			IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x4)
+#define PCI_IMAP2			IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x8)
+#define PCI_SMAP0			IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x10)
+#define PCI_SMAP1			IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x14)
+#define PCI_SMAP2			IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x18)
+#define PCI_SELFID			IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0xc)
+
+#define DEVICE_ID_OFFSET		0x00
+#define CSR_OFFSET			0x04
+#define CLASS_ID_OFFSET			0x08
+
+#define VP_PCI_DEVICE_ID		0x030010ee
+#define VP_PCI_CLASS_ID			0x0b400000
+
+static unsigned long pci_slot_ignore = 0;
+
+static int __init versatile_pci_slot_ignore(char *str)
+{
+	int retval;
+	int slot;
+
+	while ((retval = get_option(&str,&slot))) {
+		if ((slot < 0) || (slot > 31)) {
+			printk("Illegal slot value: %d\n",slot);
+		} else {
+			pci_slot_ignore |= (1 << slot);
+		}
+	}
+	return 1;
+}
+
+__setup("pci_slot_ignore=", versatile_pci_slot_ignore);
+
+
+static unsigned long __pci_addr(struct pci_bus *bus,
+				unsigned int devfn, int offset)
+{
+	unsigned int busnr = bus->number;
+
+	/*
+	 * Trap out illegal values
+	 */
+	if (offset > 255)
+		BUG();
+	if (busnr > 255)
+		BUG();
+	if (devfn > 255)
+		BUG();
+
+	return (VERSATILE_PCI_CFG_VIRT_BASE | (busnr << 16) |
+		(PCI_SLOT(devfn) << 11) | (PCI_FUNC(devfn) << 8) | offset);
+}
+
+static int versatile_read_config(struct pci_bus *bus, unsigned int devfn, int where,
+				 int size, u32 *val)
+{
+	unsigned long addr = __pci_addr(bus, devfn, where);
+	u32 v;
+	int slot = PCI_SLOT(devfn);
+
+	if (pci_slot_ignore & (1 << slot)) {
+		/* Ignore this slot */
+		switch (size) {
+		case 1:
+			v = 0xff;
+			break;
+		case 2:
+			v = 0xffff;
+			break;
+		default:
+			v = 0xffffffff;
+		}
+	} else {
+		switch (size) {
+		case 1:
+			addr &= ~3;
+			v = __raw_readb(addr);
+			break;
+
+		case 2:
+			v = __raw_readl(addr & ~3);
+			if (addr & 2) v >>= 16;
+ 			v &= 0xffff;
+			break;
+
+		default:
+			addr &= ~3;
+			v = __raw_readl(addr);
+			break;
+		}
+	}
+
+	*val = v;
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int versatile_write_config(struct pci_bus *bus, unsigned int devfn, int where,
+				  int size, u32 val)
+{
+	unsigned long addr = __pci_addr(bus, devfn, where);
+	int slot = PCI_SLOT(devfn);
+
+	if (pci_slot_ignore & (1 << slot)) {
+		return PCIBIOS_SUCCESSFUL;
+	}
+
+	switch (size) {
+	case 1:
+		__raw_writeb((u8)val, addr);
+		break;
+
+	case 2:
+		__raw_writew((u16)val, addr);
+		break;
+
+	case 4:
+		__raw_writel(val, addr);
+		break;
+	}
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static struct pci_ops pci_versatile_ops = {
+	.read	= versatile_read_config,
+	.write	= versatile_write_config,
+};
+
+static struct resource io_mem = {
+	.name	= "PCI I/O space",
+	.start	= VERSATILE_PCI_MEM_BASE0,
+	.end	= VERSATILE_PCI_MEM_BASE0+VERSATILE_PCI_MEM_BASE0_SIZE-1,
+	.flags	= IORESOURCE_IO,
+};
+
+static struct resource non_mem = {
+	.name	= "PCI non-prefetchable",
+	.start	= VERSATILE_PCI_MEM_BASE1,
+	.end	= VERSATILE_PCI_MEM_BASE1+VERSATILE_PCI_MEM_BASE1_SIZE-1,
+	.flags	= IORESOURCE_MEM,
+};
+
+static struct resource pre_mem = {
+	.name	= "PCI prefetchable",
+	.start	= VERSATILE_PCI_MEM_BASE2,
+	.end	= VERSATILE_PCI_MEM_BASE2+VERSATILE_PCI_MEM_BASE2_SIZE-1,
+	.flags	= IORESOURCE_MEM | IORESOURCE_PREFETCH,
+};
+
+static int __init pci_versatile_setup_resources(struct resource **resource)
+{
+	int ret = 0;
+
+	ret = request_resource(&iomem_resource, &io_mem);
+	if (ret) {
+		printk(KERN_ERR "PCI: unable to allocate I/O "
+		       "memory region (%d)\n", ret);
+		goto out;
+	}
+	ret = request_resource(&iomem_resource, &non_mem);
+	if (ret) {
+		printk(KERN_ERR "PCI: unable to allocate non-prefetchable "
+		       "memory region (%d)\n", ret);
+		goto release_io_mem;
+	}
+	ret = request_resource(&iomem_resource, &pre_mem);
+	if (ret) {
+		printk(KERN_ERR "PCI: unable to allocate prefetchable "
+		       "memory region (%d)\n", ret);
+		goto release_non_mem;
+	}
+
+	/*
+	 * bus->resource[0] is the IO resource for this bus
+	 * bus->resource[1] is the mem resource for this bus
+	 * bus->resource[2] is the prefetch mem resource for this bus
+	 */
+	resource[0] = &io_mem;
+	resource[1] = &non_mem;
+	resource[2] = &pre_mem;
+
+	goto out;
+
+ release_non_mem:
+	release_resource(&non_mem);
+ release_io_mem:
+	release_resource(&io_mem);
+ out:
+	return ret;
+}
+
+int __init pci_versatile_setup(int nr, struct pci_sys_data *sys)
+{
+	int ret = 0;
+        int i;
+        int myslot = -1;
+	unsigned long val;
+
+	if (nr == 0) {
+		sys->mem_offset = 0;
+		ret = pci_versatile_setup_resources(sys->resource);
+		if (ret < 0) {
+			printk("pci_versatile_setup: resources... oops?\n");
+			goto out;
+		}
+	} else {
+		printk("pci_versatile_setup: resources... nr == 0??\n");
+		goto out;
+	}
+
+	__raw_writel(VERSATILE_PCI_MEM_BASE0 >> 28,PCI_IMAP0);
+	__raw_writel(VERSATILE_PCI_MEM_BASE1 >> 28,PCI_IMAP1);
+	__raw_writel(VERSATILE_PCI_MEM_BASE2 >> 28,PCI_IMAP2);
+
+	__raw_writel(1, SYS_PCICTL);
+
+	val = __raw_readl(SYS_PCICTL);
+	if (!(val & 1)) {
+		printk("Not plugged into PCI backplane!\n");
+		ret = -EIO;
+		goto out;
+	}
+
+	/*
+	 *  We need to discover the PCI core first to configure itself
+	 *  before the main PCI probing is performed
+	 */
+	for (i=0; i<32; i++) {
+		if ((__raw_readl(VERSATILE_PCI_VIRT_BASE+(i<<11)+DEVICE_ID_OFFSET) == VP_PCI_DEVICE_ID) &&
+		    (__raw_readl(VERSATILE_PCI_VIRT_BASE+(i<<11)+CLASS_ID_OFFSET) == VP_PCI_CLASS_ID)) {
+			myslot = i;
+
+			__raw_writel(myslot, PCI_SELFID);
+			val = __raw_readl(VERSATILE_PCI_CFG_VIRT_BASE+(myslot<<11)+CSR_OFFSET);
+			val |= (1<<2);
+			__raw_writel(val, VERSATILE_PCI_CFG_VIRT_BASE+(myslot<<11)+CSR_OFFSET);
+			break;
+		}
+	}
+
+	if (myslot == -1) {
+		printk("Cannot find PCI core!\n");
+		ret = -EIO;
+	} else {
+		printk("PCI core found (slot %d)\n",myslot);
+		/* Do not to map Versatile FPGA PCI device
+		   into memory space as we are short of
+		   mappable memory */
+		pci_slot_ignore |= (1 << myslot);
+		ret = 1;
+	}
+
+ out:
+	return ret;
+}
+
+
+struct pci_bus *pci_versatile_scan_bus(int nr, struct pci_sys_data *sys)
+{
+	return pci_scan_bus(sys->busnr, &pci_versatile_ops, sys);
+}
+
+/*
+ * V3_LB_BASE? - local bus address
+ * V3_LB_MAP?  - pci bus address
+ */
+void __init pci_versatile_preinit(void)
+{
+}
+
+void __init pci_versatile_postinit(void)
+{
+}
+
+
+/*
+ * map the specified device/slot/pin to an IRQ.   Different backplanes may need to modify this.
+ */
+static int __init versatile_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
+{
+	int irq;
+	int devslot = PCI_SLOT(dev->devfn);
+
+	/* slot,  pin,  irq
+	    24	  1	27
+	    25    1	28	untested
+	    26	  1	29
+	    27    1	30	untested
+	*/
+
+	irq = 27 + ((slot + pin + 2) % 3);	/* Fudged */
+
+	printk("map irq: slot %d, pin %d, devslot %d, irq: %d\n",slot,pin,devslot,irq);
+
+	return irq;
+}
+
+static struct hw_pci versatile_pci __initdata = {
+	.swizzle		= NULL,
+	.map_irq		= versatile_map_irq,
+	.nr_controllers		= 1,
+	.setup			= pci_versatile_setup,
+	.scan			= pci_versatile_scan_bus,
+	.preinit		= pci_versatile_preinit,
+	.postinit		= pci_versatile_postinit,
+};
+
+static int __init versatile_pci_init(void)
+{
+	pci_common_init(&versatile_pci);
+	return 0;
+}
+
+subsys_initcall(versatile_pci_init);
diff --git a/include/asm-arm/arch-versatile/hardware.h b/include/asm-arm/arch-versatile/hardware.h
index d5fb4a251e7..41c1bee342a 100644
--- a/include/asm-arm/arch-versatile/hardware.h
+++ b/include/asm-arm/arch-versatile/hardware.h
@@ -25,19 +25,26 @@
 #include <asm/sizes.h>
 #include <asm/arch/platform.h>
 
-// FIXME = PCI settings need to be fixed!!!!!
-
 /*
- * Similar to above, but for PCI addresses (memory, IO, Config and the
- * V3 chip itself).  WARNING: this has to mirror definitions in platform.h
+ * PCI space virtual addresses
  */
-#define PCI_MEMORY_VADDR       0xe8000000
-#define PCI_CONFIG_VADDR       0xec000000
-#define PCI_V3_VADDR           0xed000000
-#define PCI_IO_VADDR           0xee000000
+#define VERSATILE_PCI_VIRT_BASE		0xe8000000
+#define VERSATILE_PCI_CFG_VIRT_BASE	0xe9000000
+
+#if 0
+#define VERSATILE_PCI_VIRT_MEM_BASE0	0xf4000000
+#define VERSATILE_PCI_VIRT_MEM_BASE1	0xf5000000
+#define VERSATILE_PCI_VIRT_MEM_BASE2	0xf6000000
+
+#define PCIO_BASE			VERSATILE_PCI_VIRT_MEM_BASE0
+#define PCIMEM_BASE			VERSATILE_PCI_VIRT_MEM_BASE1
+#endif
+
+/* CIK guesswork */
+#define PCIBIOS_MIN_IO			0x44000000
+#define PCIBIOS_MIN_MEM			0x50000000
 
-#define PCIO_BASE		PCI_IO_VADDR
-#define PCIMEM_BASE		PCI_MEMORY_VADDR
+#define pcibios_assign_all_busses()     1
 
 /* macro to get at IO space when running virtually */
 #define IO_ADDRESS(x)		(((x) & 0x0fffffff) + (((x) >> 4) & 0x0f000000) + 0xf0000000)
diff --git a/include/asm-arm/arch-versatile/io.h b/include/asm-arm/arch-versatile/io.h
index dbb7158788f..9f895bf6149 100644
--- a/include/asm-arm/arch-versatile/io.h
+++ b/include/asm-arm/arch-versatile/io.h
@@ -20,7 +20,7 @@
 #ifndef __ASM_ARM_ARCH_IO_H
 #define __ASM_ARM_ARCH_IO_H
 
-#define IO_SPACE_LIMIT 0xffff
+#define IO_SPACE_LIMIT 0xffffffff
 
 #define __io(a)			((void __iomem *)(a))
 #define __mem_pci(a)		(a)
diff --git a/include/asm-arm/arch-versatile/platform.h b/include/asm-arm/arch-versatile/platform.h
index a71093e44c5..cbdd9fb9633 100644
--- a/include/asm-arm/arch-versatile/platform.h
+++ b/include/asm-arm/arch-versatile/platform.h
@@ -76,7 +76,7 @@
 #define VERSATILE_SYS_NVFLAGSSET_OFFSET       0x38
 #define VERSATILE_SYS_NVFLAGSCLR_OFFSET       0x3C
 #define VERSATILE_SYS_RESETCTL_OFFSET         0x40
-#define VERSATILE_SYS_PICCTL_OFFSET           0x44
+#define VERSATILE_SYS_PCICTL_OFFSET           0x44
 #define VERSATILE_SYS_MCI_OFFSET              0x48
 #define VERSATILE_SYS_FLASH_OFFSET            0x4C
 #define VERSATILE_SYS_CLCD_OFFSET             0x50
@@ -114,7 +114,7 @@
 #define VERSATILE_SYS_NVFLAGSSET              (VERSATILE_SYS_BASE + VERSATILE_SYS_NVFLAGSSET_OFFSET)
 #define VERSATILE_SYS_NVFLAGSCLR              (VERSATILE_SYS_BASE + VERSATILE_SYS_NVFLAGSCLR_OFFSET)
 #define VERSATILE_SYS_RESETCTL                (VERSATILE_SYS_BASE + VERSATILE_SYS_RESETCTL_OFFSET)
-#define VERSATILE_SYS_PICCTL                  (VERSATILE_SYS_BASE + VERSATILE_SYS_PICCTL_OFFSET)
+#define VERSATILE_SYS_PCICTL                  (VERSATILE_SYS_BASE + VERSATILE_SYS_PCICTL_OFFSET)
 #define VERSATILE_SYS_MCI                     (VERSATILE_SYS_BASE + VERSATILE_SYS_MCI_OFFSET)
 #define VERSATILE_SYS_FLASH                   (VERSATILE_SYS_BASE + VERSATILE_SYS_FLASH_OFFSET)
 #define VERSATILE_SYS_CLCD                    (VERSATILE_SYS_BASE + VERSATILE_SYS_CLCD_OFFSET)
@@ -225,7 +225,20 @@
 #define VERSATILE_SSMC_BASE            0x20000000	/* SSMC */
 #define VERSATILE_IB2_BASE             0x24000000	/* IB2 module */
 #define VERSATILE_MBX_BASE             0x40000000	/* MBX */
+
+/* PCI space */
 #define VERSATILE_PCI_BASE             0x41000000	/* PCI Interface */
+#define VERSATILE_PCI_CFG_BASE	       0x42000000
+#define VERSATILE_PCI_MEM_BASE0        0x44000000
+#define VERSATILE_PCI_MEM_BASE1        0x50000000
+#define VERSATILE_PCI_MEM_BASE2        0x60000000
+/* Sizes of above maps */
+#define VERSATILE_PCI_BASE_SIZE	       0x01000000
+#define VERSATILE_PCI_CFG_BASE_SIZE    0x02000000
+#define VERSATILE_PCI_MEM_BASE0_SIZE   0x0c000000	/* 32Mb */
+#define VERSATILE_PCI_MEM_BASE1_SIZE   0x10000000	/* 256Mb */
+#define VERSATILE_PCI_MEM_BASE2_SIZE   0x10000000	/* 256Mb */
+
 #define VERSATILE_SDRAM67_BASE         0x70000000	/* SDRAM banks 6 and 7 */
 #define VERSATILE_LT_BASE              0x80000000	/* Logic Tile expansion */
 
-- 
cgit v1.2.3-70-g09d2


From e4fe19819ef32950541503042f32e71b67edffc7 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@org.rmk.(none)>
Date: Mon, 20 Jun 2005 18:51:07 +0100
Subject: [PATCH] ARM: 2701/1: free up ixp2000 timer 4 for the watchdog

Patch from Lennert Buytenhek

The IXP2000 has four timers, but if we're on an A-step IXP2800, timer
2 and 3 don't work.  We need two timers for timekeeping (one for the
timer interrupt and one for tracking missed jiffies), so on early
IXP2800s we have no other choice but to use timer 1 and 4 for that,
but on all other IXP2000s we'd rather leave timer 4 free since that's
the only timer we can use for the watchdog.
So, on buggy IXP2000s (i.e. the A-step IXP2800) we use timer 4 for
tracking missed jiffies, and on all all non-buggy IXP2000s (i.e.
everything but the A-step IXP2800) we use timer 2.
On a pre-production IXP2800, this patch should print these messages
on boot:
	Enabling IXP2800 erratum #25 workaround
	Unable to use IXP2000 watchdog due to IXP2800 erratum #25
On any non-buggy IXP2800 (as well as on IXP2400s) you shouldn't see
anything at all, and the watchdog should be usable again.

Signed-off-by: Lennert Buytenhek
Signed-off-by: Deepak Saxena
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-ixp2000/core.c                | 34 ++++++++++++++++++++++-------
 drivers/char/watchdog/ixp2000_wdt.c         |  7 +++++-
 include/asm-arm/arch-ixp2000/ixp2000-regs.h |  1 +
 3 files changed, 33 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-ixp2000/core.c b/arch/arm/mach-ixp2000/core.c
index 4f3c3d5c781..fc0555596d6 100644
--- a/arch/arm/mach-ixp2000/core.c
+++ b/arch/arm/mach-ixp2000/core.c
@@ -162,12 +162,13 @@ void __init ixp2000_map_io(void)
 static unsigned ticks_per_jiffy;
 static unsigned ticks_per_usec;
 static unsigned next_jiffy_time;
+static volatile unsigned long *missing_jiffy_timer_csr;
 
 unsigned long ixp2000_gettimeoffset (void)
 {
  	unsigned long offset;
 
-	offset = next_jiffy_time - *IXP2000_T4_CSR;
+	offset = next_jiffy_time - *missing_jiffy_timer_csr;
 
 	return offset / ticks_per_usec;
 }
@@ -179,7 +180,7 @@ static int ixp2000_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 	/* clear timer 1 */
 	ixp2000_reg_write(IXP2000_T1_CLR, 1);
 	
-	while ((next_jiffy_time - *IXP2000_T4_CSR) > ticks_per_jiffy) {
+	while ((next_jiffy_time - *missing_jiffy_timer_csr) > ticks_per_jiffy) {
 		timer_tick(regs);
 		next_jiffy_time -= ticks_per_jiffy;
 	}
@@ -197,20 +198,37 @@ static struct irqaction ixp2000_timer_irq = {
 
 void __init ixp2000_init_time(unsigned long tick_rate)
 {
-	ixp2000_reg_write(IXP2000_T1_CLR, 0);
-	ixp2000_reg_write(IXP2000_T4_CLR, 0);
-
 	ticks_per_jiffy = (tick_rate + HZ/2) / HZ;
 	ticks_per_usec = tick_rate / 1000000;
 
+	/*
+	 * We use timer 1 as our timer interrupt.
+	 */
+	ixp2000_reg_write(IXP2000_T1_CLR, 0);
 	ixp2000_reg_write(IXP2000_T1_CLD, ticks_per_jiffy - 1);
 	ixp2000_reg_write(IXP2000_T1_CTL, (1 << 7));
 
 	/*
-	 * We use T4 as a monotonic counter to track missed jiffies
+	 * We use a second timer as a monotonic counter for tracking
+	 * missed jiffies.  The IXP2000 has four timers, but if we're
+	 * on an A-step IXP2800, timer 2 and 3 don't work, so on those
+	 * chips we use timer 4.  Timer 4 is the only timer that can
+	 * be used for the watchdog, so we use timer 2 if we're on a
+	 * non-buggy chip.
 	 */
-	ixp2000_reg_write(IXP2000_T4_CLD, -1);
-	ixp2000_reg_write(IXP2000_T4_CTL, (1 << 7));
+	if ((*IXP2000_PRODUCT_ID & 0x001ffef0) == 0x00000000) {
+		printk(KERN_INFO "Enabling IXP2800 erratum #25 workaround\n");
+
+		ixp2000_reg_write(IXP2000_T4_CLR, 0);
+		ixp2000_reg_write(IXP2000_T4_CLD, -1);
+		ixp2000_reg_write(IXP2000_T4_CTL, (1 << 7));
+		missing_jiffy_timer_csr = IXP2000_T4_CSR;
+	} else {
+		ixp2000_reg_write(IXP2000_T2_CLR, 0);
+		ixp2000_reg_write(IXP2000_T2_CLD, -1);
+		ixp2000_reg_write(IXP2000_T2_CTL, (1 << 7));
+		missing_jiffy_timer_csr = IXP2000_T2_CSR;
+	}
  	next_jiffy_time = 0xffffffff;
 
 	/* register for interrupt */
diff --git a/drivers/char/watchdog/ixp2000_wdt.c b/drivers/char/watchdog/ixp2000_wdt.c
index ab659d37b4d..4e98c215e5b 100644
--- a/drivers/char/watchdog/ixp2000_wdt.c
+++ b/drivers/char/watchdog/ixp2000_wdt.c
@@ -192,7 +192,12 @@ static struct miscdevice ixp2000_wdt_miscdev =
 
 static int __init ixp2000_wdt_init(void)
 {
-	wdt_tick_rate = (*IXP2000_T1_CLD * HZ)/ 256;;
+	if ((*IXP2000_PRODUCT_ID & 0x001ffef0) == 0x00000000) {
+		printk(KERN_INFO "Unable to use IXP2000 watchdog due to IXP2800 erratum #25.\n");
+		return -EIO;
+	}
+
+	wdt_tick_rate = (*IXP2000_T1_CLD * HZ) / 256;
 
 	return misc_register(&ixp2000_wdt_miscdev);
 }
diff --git a/include/asm-arm/arch-ixp2000/ixp2000-regs.h b/include/asm-arm/arch-ixp2000/ixp2000-regs.h
index 6c56708d0ff..a1d9e181b10 100644
--- a/include/asm-arm/arch-ixp2000/ixp2000-regs.h
+++ b/include/asm-arm/arch-ixp2000/ixp2000-regs.h
@@ -363,6 +363,7 @@
 #define IXP2000_MIN_REV_MASK	      	0x0000000F
 #define IXP2000_PROD_ID_MASK		0xFFFFFFFF
 
+#define IXP2000_PRODUCT_ID		GLOBAL_REG(0x00)
 #define IXP2000_MISC_CONTROL		GLOBAL_REG(0x04)
 #define IXP2000_MSF_CLK_CNTRL  		GLOBAL_REG(0x08)
 #define IXP2000_RESET0      		GLOBAL_REG(0x0c)
-- 
cgit v1.2.3-70-g09d2


From 3f7a87d2fa9b42f7aade43914f060df68cc89cc7 Mon Sep 17 00:00:00 2001
From: Frank Filz <ffilzlnx@us.ibm.com>
Date: Mon, 20 Jun 2005 13:14:57 -0700
Subject: [SCTP] sctp_connectx() API support

Implements sctp_connectx() as defined in the SCTP sockets API draft by
tunneling the request through a setsockopt().

Signed-off-by: Frank Filz <ffilzlnx@us.ibm.com>
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/command.h   |   8 +-
 include/net/sctp/constants.h |   7 -
 include/net/sctp/sctp.h      |  17 ++
 include/net/sctp/sm.h        |   8 +-
 include/net/sctp/structs.h   |  41 ++++-
 include/net/sctp/user.h      |   3 +
 net/sctp/associola.c         | 151 ++++++++++++----
 net/sctp/endpointola.c       |   1 -
 net/sctp/input.c             |   2 +-
 net/sctp/outqueue.c          |  11 +-
 net/sctp/sm_make_chunk.c     |  20 ++-
 net/sctp/sm_sideeffect.c     | 105 ++++++++---
 net/sctp/sm_statefuns.c      | 148 ++++++++++------
 net/sctp/sm_statetable.c     |   6 +-
 net/sctp/socket.c            | 405 ++++++++++++++++++++++++++++++-------------
 net/sctp/transport.c         |   4 +-
 16 files changed, 677 insertions(+), 260 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/command.h b/include/net/sctp/command.h
index ebc5282e6d5..dc107ffad48 100644
--- a/include/net/sctp/command.h
+++ b/include/net/sctp/command.h
@@ -65,9 +65,11 @@ typedef enum {
 	SCTP_CMD_TIMER_START,	/* Start a timer.  */
 	SCTP_CMD_TIMER_RESTART,	/* Restart a timer. */
 	SCTP_CMD_TIMER_STOP,	/* Stop a timer. */
-	SCTP_CMD_COUNTER_RESET, /* Reset a counter. */
-	SCTP_CMD_COUNTER_INC,   /* Increment a counter. */
+	SCTP_CMD_INIT_CHOOSE_TRANSPORT, /* Choose transport for an INIT. */
+	SCTP_CMD_INIT_COUNTER_RESET, /* Reset init counter. */
+	SCTP_CMD_INIT_COUNTER_INC,   /* Increment init counter. */
 	SCTP_CMD_INIT_RESTART,  /* High level, do init timer work. */
+	SCTP_CMD_COOKIEECHO_RESTART,  /* High level, do cookie-echo timer work. */
 	SCTP_CMD_INIT_FAILED,   /* High level, do init failure work. */
 	SCTP_CMD_REPORT_DUP,	/* Report a duplicate TSN.  */
 	SCTP_CMD_STRIKE,	/* Mark a strike against a transport.  */
@@ -118,7 +120,6 @@ typedef union {
 	int error;
 	sctp_state_t state;
 	sctp_event_timeout_t to;
-	sctp_counter_t counter;
 	void *ptr;
 	struct sctp_chunk *chunk;
 	struct sctp_association *asoc;
@@ -165,7 +166,6 @@ SCTP_ARG_CONSTRUCTOR(U16,	__u16, u16)
 SCTP_ARG_CONSTRUCTOR(U8,	__u8, u8)
 SCTP_ARG_CONSTRUCTOR(ERROR,     int, error)
 SCTP_ARG_CONSTRUCTOR(STATE,	sctp_state_t, state)
-SCTP_ARG_CONSTRUCTOR(COUNTER,	sctp_counter_t, counter)
 SCTP_ARG_CONSTRUCTOR(TO,	sctp_event_timeout_t, to)
 SCTP_ARG_CONSTRUCTOR(PTR,	void *, ptr)
 SCTP_ARG_CONSTRUCTOR(CHUNK,	struct sctp_chunk *, chunk)
diff --git a/include/net/sctp/constants.h b/include/net/sctp/constants.h
index 2b76c0f4bab..4868c7f7749 100644
--- a/include/net/sctp/constants.h
+++ b/include/net/sctp/constants.h
@@ -263,13 +263,6 @@ enum { SCTP_MIN_PMTU = 576 };
 enum { SCTP_MAX_DUP_TSNS = 16 };
 enum { SCTP_MAX_GABS = 16 };
 
-typedef enum {
-	SCTP_COUNTER_INIT_ERROR,
-} sctp_counter_t;
-
-/* How many counters does an association need? */
-#define SCTP_NUMBER_COUNTERS	5
-
 /* Here we define the default timers.  */
 
 /* cookie timer def = ? seconds */
diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h
index 960abfa48d6..ef2738159ab 100644
--- a/include/net/sctp/sctp.h
+++ b/include/net/sctp/sctp.h
@@ -223,6 +223,22 @@ DECLARE_SNMP_STAT(struct sctp_mib, sctp_statistics);
 extern int sctp_debug_flag;
 #define SCTP_DEBUG_PRINTK(whatever...) \
 	((void) (sctp_debug_flag && printk(KERN_DEBUG whatever)))
+#define SCTP_DEBUG_PRINTK_IPADDR(lead, trail, leadparm, saddr, otherparms...) \
+	if (sctp_debug_flag) { \
+		if (saddr->sa.sa_family == AF_INET6) { \
+			printk(KERN_DEBUG \
+			       lead "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x" trail, \
+			       leadparm, \
+			       NIP6(saddr->v6.sin6_addr), \
+			       otherparms); \
+		} else { \
+			printk(KERN_DEBUG \
+			       lead "%u.%u.%u.%u" trail, \
+			       leadparm, \
+			       NIPQUAD(saddr->v4.sin_addr.s_addr), \
+			       otherparms); \
+		} \
+	}
 #define SCTP_ENABLE_DEBUG { sctp_debug_flag = 1; }
 #define SCTP_DISABLE_DEBUG { sctp_debug_flag = 0; }
 
@@ -236,6 +252,7 @@ extern int sctp_debug_flag;
 #else	/* SCTP_DEBUG */
 
 #define SCTP_DEBUG_PRINTK(whatever...)
+#define SCTP_DEBUG_PRINTK_IPADDR(whatever...)
 #define SCTP_ENABLE_DEBUG
 #define SCTP_DISABLE_DEBUG
 #define SCTP_ASSERT(expr, str, func)
diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h
index f4fcee10470..a53e08a45e3 100644
--- a/include/net/sctp/sm.h
+++ b/include/net/sctp/sm.h
@@ -116,7 +116,8 @@ sctp_state_fn_t sctp_sf_eat_data_fast_4_4;
 sctp_state_fn_t sctp_sf_eat_sack_6_2;
 sctp_state_fn_t sctp_sf_tabort_8_4_8;
 sctp_state_fn_t sctp_sf_operr_notify;
-sctp_state_fn_t sctp_sf_t1_timer_expire;
+sctp_state_fn_t sctp_sf_t1_init_timer_expire;
+sctp_state_fn_t sctp_sf_t1_cookie_timer_expire;
 sctp_state_fn_t sctp_sf_t2_timer_expire;
 sctp_state_fn_t sctp_sf_t4_timer_expire;
 sctp_state_fn_t sctp_sf_t5_timer_expire;
@@ -258,7 +259,10 @@ struct sctp_chunk *sctp_make_fwdtsn(const struct sctp_association *asoc,
 void sctp_chunk_assign_tsn(struct sctp_chunk *);
 void sctp_chunk_assign_ssn(struct sctp_chunk *);
 
-void sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, __u16 error);
+sctp_disposition_t  sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands,
+					   __u16 error,
+					   const struct sctp_association *asoc,
+					   struct sctp_transport *transport);
 
 /* Prototypes for statetable processing. */
 
diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index 6c24d9cd3d6..dfad4d3c581 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -867,10 +867,13 @@ struct sctp_transport {
 	 */
 	unsigned long last_time_ecne_reduced;
 
-	/* active      : The current active state of this destination,
-	 *	       :  i.e. DOWN, UP, etc.
+	/* The number of times INIT has been sent on this transport. */
+	int init_sent_count;
+
+	/* state       : The current state of this destination,
+	 *             : i.e. SCTP_ACTIVE, SCTP_INACTIVE, SCTP_UNKOWN.
 	 */
-	int active;
+	int state;
 
 	/* hb_allowed  : The current heartbeat state of this destination,
 	 *	       :  i.e. ALLOW-HB, NO-HEARTBEAT, etc.
@@ -1222,9 +1225,6 @@ struct sctp_endpoint {
 
 	/* sendbuf acct. policy.	*/
 	__u32 sndbuf_policy;
-
-	/* Name for debugging output... */
-	char *debug_name;
 };
 
 /* Recover the outter endpoint structure. */
@@ -1314,11 +1314,23 @@ struct sctp_association {
 		 *	       : association. Normally this information is
 		 *	       : hashed or keyed for quick lookup and access
 		 *	       : of the TCB.
+		 *	       : The list is also initialized with the list
+		 *	       : of addresses passed with the sctp_connectx()
+		 *	       : call.
 		 *
 		 * It is a list of SCTP_transport's.
 		 */
 		struct list_head transport_addr_list;
 
+		/* transport_count
+		 *
+		 * Peer        : A count of the number of peer addresses
+		 * Transport   : in the Peer Transport Address List.
+		 * Address     :
+		 * Count       :
+		 */
+		__u16 transport_count;
+
 		/* port
 		 *   The transport layer port number.
 		 */
@@ -1486,6 +1498,9 @@ struct sctp_association {
 	/* Transport to which SHUTDOWN chunk was last sent.  */
 	struct sctp_transport *shutdown_last_sent_to;
 
+	/* Transport to which INIT chunk was last sent.  */
+	struct sctp_transport *init_last_sent_to;
+
 	/* Next TSN    : The next TSN number to be assigned to a new
 	 *	       : DATA chunk.  This is sent in the INIT or INIT
 	 *	       : ACK chunk to the peer and incremented each
@@ -1549,8 +1564,11 @@ struct sctp_association {
 	/* The message size at which SCTP fragmentation will occur. */
 	__u32 frag_point;
 
-	/* Currently only one counter is used to count INIT errors. */
-	int counters[SCTP_NUMBER_COUNTERS];
+	/* Counter used to count INIT errors. */
+	int init_err_counter;
+
+	/* Count the number of INIT cycles (for doubling timeout). */
+	int init_cycle;
 
 	/* Default send parameters. */
 	__u16 default_stream;
@@ -1708,6 +1726,8 @@ void sctp_association_free(struct sctp_association *);
 void sctp_association_put(struct sctp_association *);
 void sctp_association_hold(struct sctp_association *);
 
+struct sctp_transport *sctp_assoc_choose_init_transport(
+	struct sctp_association *);
 struct sctp_transport *sctp_assoc_choose_shutdown_transport(
 	struct sctp_association *);
 void sctp_assoc_update_retran_path(struct sctp_association *);
@@ -1717,9 +1737,12 @@ int sctp_assoc_lookup_laddr(struct sctp_association *asoc,
 			    const union sctp_addr *laddr);
 struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *,
 				     const union sctp_addr *address,
-				     const int gfp);
+				     const int gfp,
+				     const int peer_state);
 void sctp_assoc_del_peer(struct sctp_association *asoc,
 			 const union sctp_addr *addr);
+void sctp_assoc_rm_peer(struct sctp_association *asoc,
+			 struct sctp_transport *peer);
 void sctp_assoc_control_transport(struct sctp_association *,
 				  struct sctp_transport *,
 				  sctp_transport_cmd_t, sctp_sn_error_t);
diff --git a/include/net/sctp/user.h b/include/net/sctp/user.h
index 2758e8ce4f2..f6328aeddcc 100644
--- a/include/net/sctp/user.h
+++ b/include/net/sctp/user.h
@@ -111,6 +111,8 @@ enum sctp_optname {
 #define SCTP_GET_LOCAL_ADDRS_NUM	SCTP_GET_LOCAL_ADDRS_NUM
 	SCTP_GET_LOCAL_ADDRS, 	/* Get all local addresss. */
 #define SCTP_GET_LOCAL_ADDRS	SCTP_GET_LOCAL_ADDRS
+	SCTP_SOCKOPT_CONNECTX, /* CONNECTX requests. */
+#define SCTP_SOCKOPT_CONNECTX	SCTP_SOCKOPT_CONNECTX
 };
 
 /*
@@ -527,6 +529,7 @@ struct sctp_paddrinfo {
 enum sctp_spinfo_state {
 	SCTP_INACTIVE,
 	SCTP_ACTIVE,
+	SCTP_UNKNOWN = 0xffff  /* Value used for transport state unknown */
 };
 
 /*
diff --git a/net/sctp/associola.c b/net/sctp/associola.c
index 663843d97a9..7ae6aa772da 100644
--- a/net/sctp/associola.c
+++ b/net/sctp/associola.c
@@ -191,10 +191,6 @@ static struct sctp_association *sctp_association_init(struct sctp_association *a
 	asoc->last_cwr_tsn = asoc->ctsn_ack_point;
 	asoc->unack_data = 0;
 
-	SCTP_DEBUG_PRINTK("myctsnap for %s INIT as 0x%x.\n",
-			  asoc->ep->debug_name,
-			  asoc->ctsn_ack_point);
-
 	/* ADDIP Section 4.1 Asconf Chunk Procedures
 	 *
 	 * When an endpoint has an ASCONF signaled change to be sent to the
@@ -211,6 +207,7 @@ static struct sctp_association *sctp_association_init(struct sctp_association *a
 
 	/* Make an empty list of remote transport addresses.  */
 	INIT_LIST_HEAD(&asoc->peer.transport_addr_list);
+	asoc->peer.transport_count = 0;
 
 	/* RFC 2960 5.1 Normal Establishment of an Association
 	 *
@@ -288,6 +285,7 @@ struct sctp_association *sctp_association_new(const struct sctp_endpoint *ep,
 
 	asoc->base.malloced = 1;
 	SCTP_DBG_OBJCNT_INC(assoc);
+	SCTP_DEBUG_PRINTK("Created asoc %p\n", asoc);
 
 	return asoc;
 
@@ -356,6 +354,8 @@ void sctp_association_free(struct sctp_association *asoc)
 		sctp_transport_free(transport);
 	}
 
+	asoc->peer.transport_count = 0;
+
 	/* Free any cached ASCONF_ACK chunk. */
 	if (asoc->addip_last_asconf_ack)
 		sctp_chunk_free(asoc->addip_last_asconf_ack);
@@ -400,7 +400,7 @@ void sctp_assoc_set_primary(struct sctp_association *asoc,
 	/* If the primary path is changing, assume that the
 	 * user wants to use this new path.
 	 */
-	if (transport->active)
+	if (transport->state != SCTP_INACTIVE)
 		asoc->peer.active_path = transport;
 
 	/*
@@ -428,10 +428,58 @@ void sctp_assoc_set_primary(struct sctp_association *asoc,
 	transport->cacc.next_tsn_at_change = asoc->next_tsn;
 }
 
+/* Remove a transport from an association.  */
+void sctp_assoc_rm_peer(struct sctp_association *asoc,
+			struct sctp_transport *peer)
+{
+	struct list_head	*pos;
+	struct sctp_transport	*transport;
+
+	SCTP_DEBUG_PRINTK_IPADDR("sctp_assoc_rm_peer:association %p addr: ",
+				 " port: %d\n",
+				 asoc,
+				 (&peer->ipaddr),
+				 peer->ipaddr.v4.sin_port);
+
+	/* If we are to remove the current retran_path, update it
+	 * to the next peer before removing this peer from the list.
+	 */
+	if (asoc->peer.retran_path == peer)
+		sctp_assoc_update_retran_path(asoc);
+
+	/* Remove this peer from the list. */
+	list_del(&peer->transports);
+
+	/* Get the first transport of asoc. */
+	pos = asoc->peer.transport_addr_list.next;
+	transport = list_entry(pos, struct sctp_transport, transports);
+
+	/* Update any entries that match the peer to be deleted. */
+	if (asoc->peer.primary_path == peer)
+		sctp_assoc_set_primary(asoc, transport);
+	if (asoc->peer.active_path == peer)
+		asoc->peer.active_path = transport;
+	if (asoc->peer.last_data_from == peer)
+		asoc->peer.last_data_from = transport;
+
+	/* If we remove the transport an INIT was last sent to, set it to
+	 * NULL. Combined with the update of the retran path above, this
+	 * will cause the next INIT to be sent to the next available
+	 * transport, maintaining the cycle.
+	 */
+	if (asoc->init_last_sent_to == peer)
+		asoc->init_last_sent_to = NULL;
+
+	asoc->peer.transport_count--;
+
+	sctp_transport_free(peer);
+}
+
 /* Add a transport address to an association.  */
 struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc,
 					   const union sctp_addr *addr,
-					   int gfp)
+					   const int gfp,
+					   const int peer_state)
 {
 	struct sctp_transport *peer;
 	struct sctp_sock *sp;
@@ -442,14 +490,25 @@ struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc,
 	/* AF_INET and AF_INET6 share common port field. */
 	port = addr->v4.sin_port;
 
+	SCTP_DEBUG_PRINTK_IPADDR("sctp_assoc_add_peer:association %p addr: ",
+				 " port: %d state:%s\n",
+				 asoc,
+				 addr,
+				 addr->v4.sin_port,
+				 peer_state == SCTP_UNKNOWN?"UNKNOWN":"ACTIVE");
+
 	/* Set the port if it has not been set yet.  */
 	if (0 == asoc->peer.port)
 		asoc->peer.port = port;
 
 	/* Check to see if this is a duplicate. */
 	peer = sctp_assoc_lookup_paddr(asoc, addr);
-	if (peer)
+	if (peer) {
+		if (peer_state == SCTP_ACTIVE &&
+		    peer->state == SCTP_UNKNOWN)
+		     peer->state = SCTP_ACTIVE;
 		return peer;
+	}
 
 	peer = sctp_transport_new(addr, gfp);
 	if (!peer)
@@ -516,8 +575,12 @@ struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc,
 	/* Set the transport's RTO.initial value */
 	peer->rto = asoc->rto_initial;
 
+	/* Set the peer's active state. */
+	peer->state = peer_state;
+
 	/* Attach the remote transport to our asoc.  */
 	list_add_tail(&peer->transports, &asoc->peer.transport_addr_list);
+	asoc->peer.transport_count++;
 
 	/* If we do not yet have a primary path, set one.  */
 	if (!asoc->peer.primary_path) {
@@ -525,8 +588,9 @@ struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc,
 		asoc->peer.retran_path = peer;
 	}
 
-	if (asoc->peer.active_path == asoc->peer.retran_path)
+	if (asoc->peer.active_path == asoc->peer.retran_path) {
 		asoc->peer.retran_path = peer;
+	}
 
 	return peer;
 }
@@ -537,37 +601,16 @@ void sctp_assoc_del_peer(struct sctp_association *asoc,
 {
 	struct list_head	*pos;
 	struct list_head	*temp;
-	struct sctp_transport	*peer = NULL;
 	struct sctp_transport	*transport;
 
 	list_for_each_safe(pos, temp, &asoc->peer.transport_addr_list) {
 		transport = list_entry(pos, struct sctp_transport, transports);
 		if (sctp_cmp_addr_exact(addr, &transport->ipaddr)) {
-			peer = transport;
-			list_del(pos);
+			/* Do book keeping for removing the peer and free it. */
+			sctp_assoc_rm_peer(asoc, transport);
 			break;
 		}
 	}
-
-	/* The address we want delete is not in the association. */
-	if (!peer)
-		return;
-
-	/* Get the first transport of asoc. */ 
-	pos = asoc->peer.transport_addr_list.next;
-	transport = list_entry(pos, struct sctp_transport, transports);
-
-	/* Update any entries that match the peer to be deleted. */  
-	if (asoc->peer.primary_path == peer)
-		sctp_assoc_set_primary(asoc, transport);
-	if (asoc->peer.active_path == peer)
-		asoc->peer.active_path = transport;
-	if (asoc->peer.retran_path == peer)
-		asoc->peer.retran_path = transport;
-	if (asoc->peer.last_data_from == peer)
-		asoc->peer.last_data_from = transport;
-
-	sctp_transport_free(peer);
 }
 
 /* Lookup a transport by address. */
@@ -608,12 +651,12 @@ void sctp_assoc_control_transport(struct sctp_association *asoc,
 	/* Record the transition on the transport.  */
 	switch (command) {
 	case SCTP_TRANSPORT_UP:
-		transport->active = SCTP_ACTIVE;
+		transport->state = SCTP_ACTIVE;
 		spc_state = SCTP_ADDR_AVAILABLE;
 		break;
 
 	case SCTP_TRANSPORT_DOWN:
-		transport->active = SCTP_INACTIVE;
+		transport->state = SCTP_INACTIVE;
 		spc_state = SCTP_ADDR_UNREACHABLE;
 		break;
 
@@ -643,7 +686,7 @@ void sctp_assoc_control_transport(struct sctp_association *asoc,
 	list_for_each(pos, &asoc->peer.transport_addr_list) {
 		t = list_entry(pos, struct sctp_transport, transports);
 
-		if (!t->active)
+		if (t->state == SCTP_INACTIVE)
 			continue;
 		if (!first || t->last_time_heard > first->last_time_heard) {
 			second = first;
@@ -663,7 +706,7 @@ void sctp_assoc_control_transport(struct sctp_association *asoc,
 	 * [If the primary is active but not most recent, bump the most
 	 * recently used transport.]
 	 */
-	if (asoc->peer.primary_path->active &&
+	if (asoc->peer.primary_path->state != SCTP_INACTIVE &&
 	    first != asoc->peer.primary_path) {
 		second = first;
 		first = asoc->peer.primary_path;
@@ -958,7 +1001,7 @@ void sctp_assoc_update(struct sctp_association *asoc,
 					   transports);
 			if (!sctp_assoc_lookup_paddr(asoc, &trans->ipaddr))
 				sctp_assoc_add_peer(asoc, &trans->ipaddr,
-						    GFP_ATOMIC);
+						    GFP_ATOMIC, SCTP_ACTIVE);
 		}
 
 		asoc->ctsn_ack_point = asoc->next_tsn - 1;
@@ -998,7 +1041,7 @@ void sctp_assoc_update_retran_path(struct sctp_association *asoc)
 
 		/* Try to find an active transport. */
 
-		if (t->active) {
+		if (t->state != SCTP_INACTIVE) {
 			break;
 		} else {
 			/* Keep track of the next transport in case
@@ -1019,6 +1062,40 @@ void sctp_assoc_update_retran_path(struct sctp_association *asoc)
 	}
 
 	asoc->peer.retran_path = t;
+
+	SCTP_DEBUG_PRINTK_IPADDR("sctp_assoc_update_retran_path:association"
+				 " %p addr: ",
+				 " port: %d\n",
+				 asoc,
+				 (&t->ipaddr),
+				 t->ipaddr.v4.sin_port);
+}
+
+/* Choose the transport for sending a INIT packet.  */
+struct sctp_transport *sctp_assoc_choose_init_transport(
+	struct sctp_association *asoc)
+{
+	struct sctp_transport *t;
+
+	/* Use the retran path. If the last INIT was sent over the
+	 * retran path, update the retran path and use it.
+	 */
+	if (!asoc->init_last_sent_to) {
+		t = asoc->peer.active_path;
+	} else {
+		if (asoc->init_last_sent_to == asoc->peer.retran_path)
+			sctp_assoc_update_retran_path(asoc);
+		t = asoc->peer.retran_path;
+	}
+
+	SCTP_DEBUG_PRINTK_IPADDR("sctp_assoc_update_retran_path:association"
+				 " %p addr: ",
+				 " port: %d\n",
+				 asoc,
+				 (&t->ipaddr),
+				 t->ipaddr.v4.sin_port);
+
+	return t;
 }
 
 /* Choose the transport for sending a SHUTDOWN packet.  */
diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c
index 334f61773e6..2ec0320fac3 100644
--- a/net/sctp/endpointola.c
+++ b/net/sctp/endpointola.c
@@ -134,7 +134,6 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep,
 	ep->last_key = ep->current_key = 0;
 	ep->key_changed_at = jiffies;
 
-	ep->debug_name = "unnamedEndpoint";
 	return ep;
 }
 
diff --git a/net/sctp/input.c b/net/sctp/input.c
index fffc880a646..339f7acfdb6 100644
--- a/net/sctp/input.c
+++ b/net/sctp/input.c
@@ -353,7 +353,7 @@ void sctp_icmp_proto_unreachable(struct sock *sk,
 
 	sctp_do_sm(SCTP_EVENT_T_OTHER,
 		   SCTP_ST_OTHER(SCTP_EVENT_ICMP_PROTO_UNREACH),
-		   asoc->state, asoc->ep, asoc, NULL,
+		   asoc->state, asoc->ep, asoc, t,
 		   GFP_ATOMIC);
 
 }
diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
index 1b2d4adc4dd..4eb81a1407b 100644
--- a/net/sctp/outqueue.c
+++ b/net/sctp/outqueue.c
@@ -682,9 +682,9 @@ int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
 
 		if (!new_transport) {
 			new_transport = asoc->peer.active_path;
-		} else if (!new_transport->active) {
-			/* If the chunk is Heartbeat or Heartbeat Ack, 
-			 * send it to chunk->transport, even if it's 
+		} else if (new_transport->state == SCTP_INACTIVE) {
+			/* If the chunk is Heartbeat or Heartbeat Ack,
+			 * send it to chunk->transport, even if it's
 			 * inactive.
 			 *
 			 * 3.3.6 Heartbeat Acknowledgement:
@@ -840,7 +840,8 @@ int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
 			 * Otherwise, we want to use the active path.
 			 */
 			new_transport = chunk->transport;
-			if (!new_transport || !new_transport->active)
+			if (!new_transport ||
+			    new_transport->state == SCTP_INACTIVE)
 				new_transport = asoc->peer.active_path;
 
 			/* Change packets if necessary.  */
@@ -1454,7 +1455,7 @@ static void sctp_check_transmitted(struct sctp_outq *q,
 			/* Mark the destination transport address as
 			 * active if it is not so marked.
 			 */
-			if (!transport->active) {
+			if (transport->state == SCTP_INACTIVE) {
 				sctp_assoc_control_transport(
 					transport->asoc,
 					transport,
diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index 33ac8bf47b0..5baed9bb7de 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -1830,7 +1830,7 @@ int sctp_process_init(struct sctp_association *asoc, sctp_cid_t cid,
 	 * be a a better choice than any of the embedded addresses.
 	 */
 	if (peer_addr)
-		if(!sctp_assoc_add_peer(asoc, peer_addr, gfp))
+		if(!sctp_assoc_add_peer(asoc, peer_addr, gfp, SCTP_ACTIVE))
 			goto nomem;
 
 	/* Process the initialization parameters.  */
@@ -1841,6 +1841,14 @@ int sctp_process_init(struct sctp_association *asoc, sctp_cid_t cid,
                         goto clean_up;
 	}
 
+	/* Walk list of transports, removing transports in the UNKNOWN state. */
+	list_for_each_safe(pos, temp, &asoc->peer.transport_addr_list) {
+		transport = list_entry(pos, struct sctp_transport, transports);
+		if (transport->state == SCTP_UNKNOWN) {
+			sctp_assoc_rm_peer(asoc, transport);
+		}
+	}
+
 	/* The fixed INIT headers are always in network byte
 	 * order.
 	 */
@@ -1906,7 +1914,8 @@ int sctp_process_init(struct sctp_association *asoc, sctp_cid_t cid,
 	 * stream sequence number shall be set to 0.
 	 */
 
-	/* Allocate storage for the negotiated streams if it is not a temporary 	 * association.
+	/* Allocate storage for the negotiated streams if it is not a temporary
+ 	 * association.
 	 */
 	if (!asoc->temp) {
 		int assoc_id;
@@ -1952,6 +1961,9 @@ clean_up:
 		list_del_init(pos);
 		sctp_transport_free(transport);
 	}
+
+	asoc->peer.transport_count = 0;
+
 nomem:
 	return 0;
 }
@@ -1995,7 +2007,7 @@ static int sctp_process_param(struct sctp_association *asoc,
 		af->from_addr_param(&addr, param.addr, asoc->peer.port, 0);
 		scope = sctp_scope(peer_addr);
 		if (sctp_in_scope(&addr, scope))
-			if (!sctp_assoc_add_peer(asoc, &addr, gfp))
+			if (!sctp_assoc_add_peer(asoc, &addr, gfp, SCTP_ACTIVE))
 				return 0;
 		break;
 
@@ -2396,7 +2408,7 @@ static __u16 sctp_process_asconf_param(struct sctp_association *asoc,
 	 	 * Due to Resource Shortage'.
 	 	 */
 
-		peer = sctp_assoc_add_peer(asoc, &addr, GFP_ATOMIC);
+		peer = sctp_assoc_add_peer(asoc, &addr, GFP_ATOMIC, SCTP_ACTIVE);
 		if (!peer)
 			return SCTP_ERROR_RSRC_LOW;
 
diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c
index f65fa441952..778639db125 100644
--- a/net/sctp/sm_sideeffect.c
+++ b/net/sctp/sm_sideeffect.c
@@ -414,11 +414,13 @@ static void sctp_do_8_2_transport_strike(struct sctp_association *asoc,
 	 */
 	asoc->overall_error_count++;
 
-	if (transport->active &&
+	if (transport->state != SCTP_INACTIVE &&
 	    (transport->error_count++ >= transport->max_retrans)) {
-		SCTP_DEBUG_PRINTK("transport_strike: transport "
-				  "IP:%d.%d.%d.%d failed.\n",
-				  NIPQUAD(transport->ipaddr.v4.sin_addr));
+		SCTP_DEBUG_PRINTK_IPADDR("transport_strike:association %p",
+					 " transport IP: port:%d failed.\n",
+					 asoc,
+					 (&transport->ipaddr),
+					 transport->ipaddr.v4.sin_port);
 		sctp_assoc_control_transport(asoc, transport,
 					     SCTP_TRANSPORT_DOWN,
 					     SCTP_FAILED_THRESHOLD);
@@ -593,7 +595,7 @@ static void sctp_cmd_transport_on(sctp_cmd_seq_t *cmds,
 	/* Mark the destination transport address as active if it is not so
 	 * marked.
 	 */
-	if (!t->active)
+	if (t->state == SCTP_INACTIVE)
 		sctp_assoc_control_transport(asoc, t, SCTP_TRANSPORT_UP,
 					     SCTP_HEARTBEAT_SUCCESS);
 
@@ -665,8 +667,11 @@ static void sctp_cmd_new_state(sctp_cmd_seq_t *cmds,
 
 	asoc->state = state;
 
+	SCTP_DEBUG_PRINTK("sctp_cmd_new_state: asoc %p[%s]\n",
+			  asoc, sctp_state_tbl[state]);
+
 	if (sctp_style(sk, TCP)) {
-		/* Change the sk->sk_state of a TCP-style socket that has 
+		/* Change the sk->sk_state of a TCP-style socket that has
 		 * sucessfully completed a connect() call.
 		 */
 		if (sctp_state(asoc, ESTABLISHED) && sctp_sstate(sk, CLOSED))
@@ -678,6 +683,16 @@ static void sctp_cmd_new_state(sctp_cmd_seq_t *cmds,
 			sk->sk_shutdown |= RCV_SHUTDOWN;
 	}
 
+	if (sctp_state(asoc, COOKIE_WAIT)) {
+		/* Reset init timeouts since they may have been
+		 * increased due to timer expirations.
+		 */
+		asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT] =
+			asoc->ep->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT];
+		asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE] =
+			asoc->ep->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE];
+	}
+
 	if (sctp_state(asoc, ESTABLISHED) ||
 	    sctp_state(asoc, CLOSED) ||
 	    sctp_state(asoc, SHUTDOWN_RECEIVED)) {
@@ -1120,10 +1135,10 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
 			 * to be executed only during failed attempts of
 			 * association establishment.
 			 */
-			if ((asoc->peer.retran_path != 
-			     asoc->peer.primary_path) && 
-			    (asoc->counters[SCTP_COUNTER_INIT_ERROR] > 0)) {
-				sctp_add_cmd_sf(commands, 
+			if ((asoc->peer.retran_path !=
+			     asoc->peer.primary_path) &&
+			    (asoc->init_err_counter > 0)) {
+				sctp_add_cmd_sf(commands,
 				                SCTP_CMD_FORCE_PRIM_RETRAN,
 						SCTP_NULL());
 			}
@@ -1237,18 +1252,67 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
 				sctp_association_put(asoc);
 			break;
 
+		case SCTP_CMD_INIT_CHOOSE_TRANSPORT:
+			chunk = cmd->obj.ptr;
+			t = sctp_assoc_choose_init_transport(asoc);
+			asoc->init_last_sent_to = t;
+			chunk->transport = t;
+			t->init_sent_count++;
+			break;
+
 		case SCTP_CMD_INIT_RESTART:
 			/* Do the needed accounting and updates
 			 * associated with restarting an initialization
-			 * timer.
+			 * timer. Only multiply the timeout by two if
+			 * all transports have been tried at the current
+			 * timeout.
+			 */
+			t = asoc->init_last_sent_to;
+			asoc->init_err_counter++;
+
+			if (t->init_sent_count > (asoc->init_cycle + 1)) {
+				asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT] *= 2;
+				if (asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT] >
+				    asoc->max_init_timeo) {
+					asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT] =
+						asoc->max_init_timeo;
+				}
+				asoc->init_cycle++;
+				SCTP_DEBUG_PRINTK(
+					"T1 INIT Timeout adjustment"
+					" init_err_counter: %d"
+					" cycle: %d"
+					" timeout: %d\n",
+					asoc->init_err_counter,
+					asoc->init_cycle,
+					asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT]);
+			}
+
+			sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART,
+					SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT));
+			break;
+
+		case SCTP_CMD_COOKIEECHO_RESTART:
+			/* Do the needed accounting and updates
+			 * associated with restarting an initialization
+			 * timer. Only multiply the timeout by two if
+			 * all transports have been tried at the current
+			 * timeout.
 			 */
-			asoc->counters[SCTP_COUNTER_INIT_ERROR]++;
-			asoc->timeouts[cmd->obj.to] *= 2;
-			if (asoc->timeouts[cmd->obj.to] >
+			asoc->init_err_counter++;
+
+			asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE] *= 2;
+			if (asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE] >
 			    asoc->max_init_timeo) {
-				asoc->timeouts[cmd->obj.to] =
+				asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE] =
 					asoc->max_init_timeo;
 			}
+			SCTP_DEBUG_PRINTK(
+				"T1 COOKIE Timeout adjustment"
+				" init_err_counter: %d"
+				" timeout: %d\n",
+				asoc->init_err_counter,
+				asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE]);
 
 			/* If we've sent any data bundled with
 			 * COOKIE-ECHO we need to resend.
@@ -1261,7 +1325,7 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
 
 			sctp_add_cmd_sf(commands,
 					SCTP_CMD_TIMER_RESTART,
-					SCTP_TO(cmd->obj.to));
+					SCTP_TO(SCTP_EVENT_TIMEOUT_T1_COOKIE));
 			break;
 
 		case SCTP_CMD_INIT_FAILED:
@@ -1273,12 +1337,13 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
 					      subtype, chunk, cmd->obj.u32);
 			break;
 
-		case SCTP_CMD_COUNTER_INC:
-			asoc->counters[cmd->obj.counter]++;
+		case SCTP_CMD_INIT_COUNTER_INC:
+			asoc->init_err_counter++;
 			break;
 
-		case SCTP_CMD_COUNTER_RESET:
-			asoc->counters[cmd->obj.counter] = 0;
+		case SCTP_CMD_INIT_COUNTER_RESET:
+			asoc->init_err_counter = 0;
+			asoc->init_cycle = 0;
 			break;
 
 		case SCTP_CMD_REPORT_DUP:
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 8e01b8f09ac..058189684c7 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -533,6 +533,9 @@ sctp_disposition_t sctp_sf_do_5_1C_ack(const struct sctp_endpoint *ep,
 	sctp_add_cmd_sf(commands, SCTP_CMD_PEER_INIT,
 			SCTP_PEER_INIT(initchunk));
 
+	/* Reset init error count upon receipt of INIT-ACK.  */
+	sctp_add_cmd_sf(commands, SCTP_CMD_INIT_COUNTER_RESET, SCTP_NULL());
+
 	/* 5.1 C) "A" shall stop the T1-init timer and leave
 	 * COOKIE-WAIT state.  "A" shall then ... start the T1-cookie
 	 * timer, and enter the COOKIE-ECHOED state.
@@ -775,8 +778,7 @@ sctp_disposition_t sctp_sf_do_5_1E_ca(const struct sctp_endpoint *ep,
 	 * from the COOKIE-ECHOED state to the COOKIE-WAIT
 	 * state is performed.
 	 */
-	sctp_add_cmd_sf(commands, SCTP_CMD_COUNTER_RESET,
-	                SCTP_COUNTER(SCTP_COUNTER_INIT_ERROR));
+	sctp_add_cmd_sf(commands, SCTP_CMD_INIT_COUNTER_RESET, SCTP_NULL());
 
 	/* RFC 2960 5.1 Normal Establishment of an Association
 	 *
@@ -1019,10 +1021,22 @@ sctp_disposition_t sctp_sf_backbeat_8_3(const struct sctp_endpoint *ep,
 	link = sctp_assoc_lookup_paddr(asoc, &from_addr);
 
 	/* This should never happen, but lets log it if so.  */
-	if (!link) {
-		printk(KERN_WARNING
-		       "%s: Could not find address %d.%d.%d.%d\n",
-		       __FUNCTION__, NIPQUAD(from_addr.v4.sin_addr));
+	if (unlikely(!link)) {
+		if (from_addr.sa.sa_family == AF_INET6) {
+			printk(KERN_WARNING
+			       "%s association %p could not find address "
+			       "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n",
+			       __FUNCTION__,
+			       asoc,
+			       NIP6(from_addr.v6.sin6_addr));
+		} else {
+			printk(KERN_WARNING
+			       "%s association %p could not find address "
+			       "%u.%u.%u.%u\n",
+			       __FUNCTION__,
+			       asoc,
+			       NIPQUAD(from_addr.v4.sin_addr.s_addr));
+		}
 		return SCTP_DISPOSITION_DISCARD;
 	}
 
@@ -2095,9 +2109,7 @@ static sctp_disposition_t sctp_sf_do_5_2_6_stale(const struct sctp_endpoint *ep,
 	sctp_errhdr_t *err;
 	struct sctp_chunk *reply;
 	struct sctp_bind_addr *bp;
-	int attempts;
-
-	attempts = asoc->counters[SCTP_COUNTER_INIT_ERROR] + 1;
+	int attempts = asoc->init_err_counter + 1;
 
 	if (attempts >= asoc->max_init_attempts) {
 		sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED,
@@ -2157,8 +2169,7 @@ static sctp_disposition_t sctp_sf_do_5_2_6_stale(const struct sctp_endpoint *ep,
 	/* Cast away the const modifier, as we want to just
 	 * rerun it through as a sideffect.
 	 */
-	sctp_add_cmd_sf(commands, SCTP_CMD_COUNTER_INC,
-			SCTP_COUNTER(SCTP_COUNTER_INIT_ERROR));
+	sctp_add_cmd_sf(commands, SCTP_CMD_INIT_COUNTER_INC, SCTP_NULL());
 
 	sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP,
 			SCTP_TO(SCTP_EVENT_TIMEOUT_T1_COOKIE));
@@ -2281,8 +2292,7 @@ sctp_disposition_t sctp_sf_cookie_wait_abort(const struct sctp_endpoint *ep,
 	if (len >= sizeof(struct sctp_chunkhdr) + sizeof(struct sctp_errhdr))
 		error = ((sctp_errhdr_t *)chunk->skb->data)->cause;
 
- 	sctp_stop_t1_and_abort(commands, error);
-	return SCTP_DISPOSITION_ABORT;
+	return sctp_stop_t1_and_abort(commands, error, asoc, chunk->transport);
 }
 
 /*
@@ -2294,8 +2304,8 @@ sctp_disposition_t sctp_sf_cookie_wait_icmp_abort(const struct sctp_endpoint *ep
 					void *arg,
 					sctp_cmd_seq_t *commands)
 {
-	sctp_stop_t1_and_abort(commands, SCTP_ERROR_NO_ERROR);
- 	return SCTP_DISPOSITION_ABORT;
+	return sctp_stop_t1_and_abort(commands, SCTP_ERROR_NO_ERROR, asoc,
+				      (struct sctp_transport *)arg);
 }
 
 /*
@@ -2318,8 +2328,12 @@ sctp_disposition_t sctp_sf_cookie_echoed_abort(const struct sctp_endpoint *ep,
  *
  * This is common code called by several sctp_sf_*_abort() functions above.
  */
-void sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, __u16 error)
+sctp_disposition_t  sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands,
+					   __u16 error,
+					   const struct sctp_association *asoc,
+					   struct sctp_transport *transport)
 {
+	SCTP_DEBUG_PRINTK("ABORT received (INIT).\n");
 	sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE,
 			SCTP_STATE(SCTP_STATE_CLOSED));
 	SCTP_INC_STATS(SCTP_MIB_ABORTEDS);
@@ -2328,6 +2342,7 @@ void sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, __u16 error)
 	/* CMD_INIT_FAILED will DELETE_TCB. */
 	sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED,
 			SCTP_U32(error));
+	return SCTP_DISPOSITION_ABORT;
 }
 
 /*
@@ -3805,6 +3820,10 @@ sctp_disposition_t sctp_sf_do_prm_asoc(const struct sctp_endpoint *ep,
 	sctp_add_cmd_sf(commands, SCTP_CMD_NEW_ASOC,
 			SCTP_ASOC((struct sctp_association *) asoc));
 
+	/* Choose transport for INIT. */
+	sctp_add_cmd_sf(commands, SCTP_CMD_INIT_CHOOSE_TRANSPORT,
+			SCTP_CHUNK(repl));
+
 	/* After sending the INIT, "A" starts the T1-init timer and
 	 * enters the COOKIE-WAIT state.
 	 */
@@ -4589,7 +4608,7 @@ sctp_disposition_t sctp_sf_do_6_2_sack(const struct sctp_endpoint *ep,
 }
 
 /*
- * sctp_sf_t1_timer_expire
+ * sctp_sf_t1_init_timer_expire
  *
  * Section: 4 Note: 2
  * Verification Tag:
@@ -4603,7 +4622,59 @@ sctp_disposition_t sctp_sf_do_6_2_sack(const struct sctp_endpoint *ep,
  *     endpoint MUST abort the initialization process and report the
  *     error to SCTP user.
  *
- *   3) If the T1-cookie timer expires, the endpoint MUST retransmit
+ * Outputs
+ * (timers, events)
+ *
+ */
+sctp_disposition_t sctp_sf_t1_init_timer_expire(const struct sctp_endpoint *ep,
+					   const struct sctp_association *asoc,
+					   const sctp_subtype_t type,
+					   void *arg,
+					   sctp_cmd_seq_t *commands)
+{
+	struct sctp_chunk *repl = NULL;
+	struct sctp_bind_addr *bp;
+	int attempts = asoc->init_err_counter + 1;
+
+	SCTP_DEBUG_PRINTK("Timer T1 expired (INIT).\n");
+
+	if (attempts < asoc->max_init_attempts) {
+		bp = (struct sctp_bind_addr *) &asoc->base.bind_addr;
+		repl = sctp_make_init(asoc, bp, GFP_ATOMIC, 0);
+		if (!repl)
+			return SCTP_DISPOSITION_NOMEM;
+
+		/* Choose transport for INIT. */
+		sctp_add_cmd_sf(commands, SCTP_CMD_INIT_CHOOSE_TRANSPORT,
+				SCTP_CHUNK(repl));
+
+		/* Issue a sideeffect to do the needed accounting. */
+		sctp_add_cmd_sf(commands, SCTP_CMD_INIT_RESTART,
+				SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT));
+
+		sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl));
+	} else {
+		SCTP_DEBUG_PRINTK("Giving up on INIT, attempts: %d"
+				  " max_init_attempts: %d\n",
+				  attempts, asoc->max_init_attempts);
+		sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED,
+				SCTP_U32(SCTP_ERROR_NO_ERROR));
+		return SCTP_DISPOSITION_DELETE_TCB;
+	}
+
+	return SCTP_DISPOSITION_CONSUME;
+}
+
+/*
+ * sctp_sf_t1_cookie_timer_expire
+ *
+ * Section: 4 Note: 2
+ * Verification Tag:
+ * Inputs
+ * (endpoint, asoc)
+ *
+ *  RFC 2960 Section 4 Notes
+ *  3) If the T1-cookie timer expires, the endpoint MUST retransmit
  *     COOKIE ECHO and re-start the T1-cookie timer without changing
  *     state.  This MUST be repeated up to 'Max.Init.Retransmits' times.
  *     After that, the endpoint MUST abort the initialization process and
@@ -4613,46 +4684,26 @@ sctp_disposition_t sctp_sf_do_6_2_sack(const struct sctp_endpoint *ep,
  * (timers, events)
  *
  */
-sctp_disposition_t sctp_sf_t1_timer_expire(const struct sctp_endpoint *ep,
+sctp_disposition_t sctp_sf_t1_cookie_timer_expire(const struct sctp_endpoint *ep,
 					   const struct sctp_association *asoc,
 					   const sctp_subtype_t type,
 					   void *arg,
 					   sctp_cmd_seq_t *commands)
 {
-	struct sctp_chunk *repl;
-	struct sctp_bind_addr *bp;
-	sctp_event_timeout_t timer = (sctp_event_timeout_t) arg;
-	int timeout;
-	int attempts;
-
-	timeout = asoc->timeouts[timer];
-	attempts = asoc->counters[SCTP_COUNTER_INIT_ERROR] + 1;
-	repl = NULL;
+	struct sctp_chunk *repl = NULL;
+	int attempts = asoc->init_err_counter + 1;
 
-	SCTP_DEBUG_PRINTK("Timer T1 expired.\n");
+	SCTP_DEBUG_PRINTK("Timer T1 expired (COOKIE-ECHO).\n");
 
 	if (attempts < asoc->max_init_attempts) {
-		switch (timer) {
-		case SCTP_EVENT_TIMEOUT_T1_INIT:
-			bp = (struct sctp_bind_addr *) &asoc->base.bind_addr;
-			repl = sctp_make_init(asoc, bp, GFP_ATOMIC, 0);
-			break;
-
-		case SCTP_EVENT_TIMEOUT_T1_COOKIE:
-			repl = sctp_make_cookie_echo(asoc, NULL);
-			break;
-
-		default:
-			BUG();
-			break;
-		};
-
+		repl = sctp_make_cookie_echo(asoc, NULL);
 		if (!repl)
-			goto nomem;
+			return SCTP_DISPOSITION_NOMEM;
 
 		/* Issue a sideeffect to do the needed accounting. */
-		sctp_add_cmd_sf(commands, SCTP_CMD_INIT_RESTART,
-				SCTP_TO(timer));
+		sctp_add_cmd_sf(commands, SCTP_CMD_COOKIEECHO_RESTART,
+				SCTP_TO(SCTP_EVENT_TIMEOUT_T1_COOKIE));
+
 		sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl));
 	} else {
 		sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED,
@@ -4661,9 +4712,6 @@ sctp_disposition_t sctp_sf_t1_timer_expire(const struct sctp_endpoint *ep,
 	}
 
 	return SCTP_DISPOSITION_CONSUME;
-
-nomem:
-	return SCTP_DISPOSITION_NOMEM;
 }
 
 /* RFC2960 9.2 If the timer expires, the endpoint must re-send the SHUTDOWN
diff --git a/net/sctp/sm_statetable.c b/net/sctp/sm_statetable.c
index 8967846f69e..75ef1040876 100644
--- a/net/sctp/sm_statetable.c
+++ b/net/sctp/sm_statetable.c
@@ -783,7 +783,8 @@ static const sctp_sm_table_entry_t other_event_table[SCTP_NUM_OTHER_TYPES][SCTP_
 	/* SCTP_STATE_COOKIE_WAIT */ \
 	{.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \
 	/* SCTP_STATE_COOKIE_ECHOED */ \
-	{.fn = sctp_sf_t1_timer_expire, .name = "sctp_sf_t1_timer_expire"}, \
+	{.fn = sctp_sf_t1_cookie_timer_expire, \
+	 .name = "sctp_sf_t1_cookie_timer_expire"}, \
 	/* SCTP_STATE_ESTABLISHED */ \
 	{.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \
 	/* SCTP_STATE_SHUTDOWN_PENDING */ \
@@ -802,7 +803,8 @@ static const sctp_sm_table_entry_t other_event_table[SCTP_NUM_OTHER_TYPES][SCTP_
 	/* SCTP_STATE_CLOSED */ \
 	{.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \
 	/* SCTP_STATE_COOKIE_WAIT */ \
-	{.fn = sctp_sf_t1_timer_expire, .name = "sctp_sf_t1_timer_expire"}, \
+	{.fn = sctp_sf_t1_init_timer_expire, \
+	 .name = "sctp_sf_t1_init_timer_expire"}, \
 	/* SCTP_STATE_COOKIE_ECHOED */ \
 	{.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \
 	/* SCTP_STATE_ESTABLISHED */ \
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index e6926cb1942..aad55dc3792 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -262,18 +262,18 @@ static struct sctp_transport *sctp_addr_id2transport(struct sock *sk,
  *             sockaddr_in6 [RFC 2553]),
  *   addr_len - the size of the address structure.
  */
-SCTP_STATIC int sctp_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len)
+SCTP_STATIC int sctp_bind(struct sock *sk, struct sockaddr *addr, int addr_len)
 {
 	int retval = 0;
 
 	sctp_lock_sock(sk);
 
-	SCTP_DEBUG_PRINTK("sctp_bind(sk: %p, uaddr: %p, addr_len: %d)\n",
-			  sk, uaddr, addr_len);
+	SCTP_DEBUG_PRINTK("sctp_bind(sk: %p, addr: %p, addr_len: %d)\n",
+			  sk, addr, addr_len);
 
 	/* Disallow binding twice. */
 	if (!sctp_sk(sk)->ep->base.bind_addr.port)
-		retval = sctp_do_bind(sk, (union sctp_addr *)uaddr,
+		retval = sctp_do_bind(sk, (union sctp_addr *)addr,
 				      addr_len);
 	else
 		retval = -EINVAL;
@@ -318,23 +318,27 @@ SCTP_STATIC int sctp_do_bind(struct sock *sk, union sctp_addr *addr, int len)
 	unsigned short snum;
 	int ret = 0;
 
-	SCTP_DEBUG_PRINTK("sctp_do_bind(sk: %p, newaddr: %p, len: %d)\n",
-			  sk, addr, len);
-
 	/* Common sockaddr verification. */
 	af = sctp_sockaddr_af(sp, addr, len);
-	if (!af)
+	if (!af) {
+		SCTP_DEBUG_PRINTK("sctp_do_bind(sk: %p, newaddr: %p, len: %d) EINVAL\n",
+				  sk, addr, len);
 		return -EINVAL;
+	}
+
+	snum = ntohs(addr->v4.sin_port);
+
+	SCTP_DEBUG_PRINTK_IPADDR("sctp_do_bind(sk: %p, new addr: ",
+				 ", port: %d, new port: %d, len: %d)\n",
+				 sk,
+				 addr,
+				 bp->port, snum,
+				 len);
 
 	/* PF specific bind() address verification. */
 	if (!sp->pf->bind_verify(sp, addr))
 		return -EADDRNOTAVAIL;
 
-	snum= ntohs(addr->v4.sin_port);
-
-	SCTP_DEBUG_PRINTK("sctp_do_bind: port: %d, new port: %d\n",
-			  bp->port, snum);
-
 	/* We must either be unbound, or bind to the same port.  */
 	if (bp->port && (snum != bp->port)) {
 		SCTP_DEBUG_PRINTK("sctp_do_bind:"
@@ -816,7 +820,8 @@ out:
  *
  * Basically do nothing but copying the addresses from user to kernel
  * land and invoking either sctp_bindx_add() or sctp_bindx_rem() on the sk.
- * This is used for tunneling the sctp_bindx() request through sctp_setsockopt() * from userspace.
+ * This is used for tunneling the sctp_bindx() request through sctp_setsockopt()
+ * from userspace.
  *
  * We don't use copy_from_user() for optimization: we first do the
  * sanity checks (buffer size -fast- and access check-healthy
@@ -913,6 +918,243 @@ out:
 	return err;
 }
 
+/* __sctp_connect(struct sock* sk, struct sockaddr *kaddrs, int addrs_size)
+ *
+ * Common routine for handling connect() and sctp_connectx().
+ * Connect will come in with just a single address.
+ */
+static int __sctp_connect(struct sock* sk,
+			  struct sockaddr *kaddrs,
+			  int addrs_size)
+{
+	struct sctp_sock *sp;
+	struct sctp_endpoint *ep;
+	struct sctp_association *asoc = NULL;
+	struct sctp_association *asoc2;
+	struct sctp_transport *transport;
+	union sctp_addr to;
+	struct sctp_af *af;
+	sctp_scope_t scope;
+	long timeo;
+	int err = 0;
+	int addrcnt = 0;
+	int walk_size = 0;
+	struct sockaddr *sa_addr;
+	void *addr_buf;
+
+	sp = sctp_sk(sk);
+	ep = sp->ep;
+
+	/* connect() cannot be done on a socket that is already in ESTABLISHED
+	 * state - UDP-style peeled off socket or a TCP-style socket that
+	 * is already connected.
+	 * It cannot be done even on a TCP-style listening socket.
+	 */
+	if (sctp_sstate(sk, ESTABLISHED) ||
+	    (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING))) {
+		err = -EISCONN;
+		goto out_free;
+	}
+
+	/* Walk through the addrs buffer and count the number of addresses. */
+	addr_buf = kaddrs;
+	while (walk_size < addrs_size) {
+		sa_addr = (struct sockaddr *)addr_buf;
+		af = sctp_get_af_specific(sa_addr->sa_family);
+
+		/* If the address family is not supported or if this address
+		 * causes the address buffer to overflow return EINVAL.
+		 */
+		if (!af || (walk_size + af->sockaddr_len) > addrs_size) {
+			err = -EINVAL;
+			goto out_free;
+		}
+
+		err = sctp_verify_addr(sk, (union sctp_addr *)sa_addr,
+				       af->sockaddr_len);
+		if (err)
+			goto out_free;
+
+		memcpy(&to, sa_addr, af->sockaddr_len);
+		to.v4.sin_port = ntohs(to.v4.sin_port);
+
+		/* Check if there already is a matching association on the
+		 * endpoint (other than the one created here).
+		 */
+		asoc2 = sctp_endpoint_lookup_assoc(ep, &to, &transport);
+		if (asoc2 && asoc2 != asoc) {
+			if (asoc2->state >= SCTP_STATE_ESTABLISHED)
+				err = -EISCONN;
+			else
+				err = -EALREADY;
+			goto out_free;
+		}
+
+		/* If we could not find a matching association on the endpoint,
+		 * make sure that there is no peeled-off association matching
+		 * the peer address even on another socket.
+		 */
+		if (sctp_endpoint_is_peeled_off(ep, &to)) {
+			err = -EADDRNOTAVAIL;
+			goto out_free;
+		}
+
+		if (!asoc) {
+			/* If a bind() or sctp_bindx() is not called prior to
+			 * an sctp_connectx() call, the system picks an
+			 * ephemeral port and will choose an address set
+			 * equivalent to binding with a wildcard address.
+			 */
+			if (!ep->base.bind_addr.port) {
+				if (sctp_autobind(sk)) {
+					err = -EAGAIN;
+					goto out_free;
+				}
+			}
+
+			scope = sctp_scope(&to);
+			asoc = sctp_association_new(ep, sk, scope, GFP_KERNEL);
+			if (!asoc) {
+				err = -ENOMEM;
+				goto out_free;
+			}
+		}
+
+		/* Prime the peer's transport structures.  */
+		transport = sctp_assoc_add_peer(asoc, &to, GFP_KERNEL,
+						SCTP_UNKNOWN);
+		if (!transport) {
+			err = -ENOMEM;
+			goto out_free;
+		}
+
+		addrcnt++;
+		addr_buf += af->sockaddr_len;
+		walk_size += af->sockaddr_len;
+	}
+
+	err = sctp_assoc_set_bind_addr_from_ep(asoc, GFP_KERNEL);
+	if (err < 0) {
+		goto out_free;
+	}
+
+	err = sctp_primitive_ASSOCIATE(asoc, NULL);
+	if (err < 0) {
+		goto out_free;
+	}
+
+	/* Initialize sk's dport and daddr for getpeername() */
+	inet_sk(sk)->dport = htons(asoc->peer.port);
+	af = sctp_get_af_specific(to.sa.sa_family);
+	af->to_sk_daddr(&to, sk);
+
+	timeo = sock_sndtimeo(sk, sk->sk_socket->file->f_flags & O_NONBLOCK);
+	err = sctp_wait_for_connect(asoc, &timeo);
+
+	/* Don't free association on exit. */
+	asoc = NULL;
+
+out_free:
+
+	SCTP_DEBUG_PRINTK("About to exit __sctp_connect() free asoc: %p"
+		          " kaddrs: %p err: %d\n",
+	                  asoc, kaddrs, err);
+	if (asoc)
+		sctp_association_free(asoc);
+	return err;
+}
+
+/* Helper for tunneling sctp_connectx() requests through sctp_setsockopt()
+ *
+ * API 8.9
+ * int sctp_connectx(int sd, struct sockaddr *addrs, int addrcnt);
+ *
+ * If sd is an IPv4 socket, the addresses passed must be IPv4 addresses.
+ * If the sd is an IPv6 socket, the addresses passed can either be IPv4
+ * or IPv6 addresses.
+ *
+ * A single address may be specified as INADDR_ANY or IN6ADDR_ANY, see
+ * Section 3.1.2 for this usage.
+ *
+ * addrs is a pointer to an array of one or more socket addresses. Each
+ * address is contained in its appropriate structure (i.e. struct
+ * sockaddr_in or struct sockaddr_in6) the family of the address type
+ * must be used to distengish the address length (note that this
+ * representation is termed a "packed array" of addresses). The caller
+ * specifies the number of addresses in the array with addrcnt.
+ *
+ * On success, sctp_connectx() returns 0. On failure, sctp_connectx() returns
+ * -1, and sets errno to the appropriate error code.
+ *
+ * For SCTP, the port given in each socket address must be the same, or
+ * sctp_connectx() will fail, setting errno to EINVAL.
+ *
+ * An application can use sctp_connectx to initiate an association with
+ * an endpoint that is multi-homed.  Much like sctp_bindx() this call
+ * allows a caller to specify multiple addresses at which a peer can be
+ * reached.  The way the SCTP stack uses the list of addresses to set up
+ * the association is implementation dependant.  This function only
+ * specifies that the stack will try to make use of all the addresses in
+ * the list when needed.
+ *
+ * Note that the list of addresses passed in is only used for setting up
+ * the association.  It does not necessarily equal the set of addresses
+ * the peer uses for the resulting association.  If the caller wants to
+ * find out the set of peer addresses, it must use sctp_getpaddrs() to
+ * retrieve them after the association has been set up.
+ *
+ * Basically do nothing but copying the addresses from user to kernel
+ * land and invoking either sctp_connectx(). This is used for tunneling
+ * the sctp_connectx() request through sctp_setsockopt() from userspace.
+ *
+ * We don't use copy_from_user() for optimization: we first do the
+ * sanity checks (buffer size -fast- and access check-healthy
+ * pointer); if all of those succeed, then we can alloc the memory
+ * (expensive operation) needed to copy the data to kernel. Then we do
+ * the copying without checking the user space area
+ * (__copy_from_user()).
+ *
+ * On exit there is no need to do sockfd_put(), sys_setsockopt() does
+ * it.
+ *
+ * sk        The sk of the socket
+ * addrs     The pointer to the addresses in user land
+ * addrssize Size of the addrs buffer
+ *
+ * Returns 0 if ok, <0 errno code on error.
+ */
+SCTP_STATIC int sctp_setsockopt_connectx(struct sock* sk,
+				      struct sockaddr __user *addrs,
+				      int addrs_size)
+{
+	int err = 0;
+	struct sockaddr *kaddrs;
+
+	SCTP_DEBUG_PRINTK("%s - sk %p addrs %p addrs_size %d\n",
+			  __FUNCTION__, sk, addrs, addrs_size);
+
+	if (unlikely(addrs_size <= 0))
+		return -EINVAL;
+
+	/* Check the user passed a healthy pointer.  */
+	if (unlikely(!access_ok(VERIFY_READ, addrs, addrs_size)))
+		return -EFAULT;
+
+	/* Alloc space for the address array in kernel memory.  */
+	kaddrs = (struct sockaddr *)kmalloc(addrs_size, GFP_KERNEL);
+	if (unlikely(!kaddrs))
+		return -ENOMEM;
+
+	if (__copy_from_user(kaddrs, addrs, addrs_size)) {
+		err = -EFAULT;
+	} else {
+		err = __sctp_connect(sk, kaddrs, addrs_size);
+	}
+
+	kfree(kaddrs);
+	return err;
+}
+
 /* API 3.1.4 close() - UDP Style Syntax
  * Applications use close() to perform graceful shutdown (as described in
  * Section 10.1 of [SCTP]) on ALL the associations currently represented
@@ -1095,7 +1337,7 @@ SCTP_STATIC int sctp_sendmsg(struct kiocb *iocb, struct sock *sk,
 	sp = sctp_sk(sk);
 	ep = sp->ep;
 
-	SCTP_DEBUG_PRINTK("Using endpoint: %s.\n", ep->debug_name);
+	SCTP_DEBUG_PRINTK("Using endpoint: %p.\n", ep);
 
 	/* We cannot send a message over a TCP-style listening socket. */
 	if (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING)) {
@@ -1306,7 +1548,7 @@ SCTP_STATIC int sctp_sendmsg(struct kiocb *iocb, struct sock *sk,
 		}
 
 		/* Prime the peer's transport structures.  */
-		transport = sctp_assoc_add_peer(asoc, &to, GFP_KERNEL);
+		transport = sctp_assoc_add_peer(asoc, &to, GFP_KERNEL, SCTP_UNKNOWN);
 		if (!transport) {
 			err = -ENOMEM;
 			goto out_free;
@@ -2208,6 +2450,12 @@ SCTP_STATIC int sctp_setsockopt(struct sock *sk, int level, int optname,
 					       optlen, SCTP_BINDX_REM_ADDR);
 		break;
 
+	case SCTP_SOCKOPT_CONNECTX:
+		/* 'optlen' is the size of the addresses buffer. */
+		retval = sctp_setsockopt_connectx(sk, (struct sockaddr __user *)optval,
+					       optlen);
+		break;
+
 	case SCTP_DISABLE_FRAGMENTS:
 		retval = sctp_setsockopt_disable_fragments(sk, optval, optlen);
 		break;
@@ -2283,112 +2531,29 @@ out_nounlock:
  *
  * len: the size of the address.
  */
-SCTP_STATIC int sctp_connect(struct sock *sk, struct sockaddr *uaddr,
+SCTP_STATIC int sctp_connect(struct sock *sk, struct sockaddr *addr,
 			     int addr_len)
 {
-	struct sctp_sock *sp;
-	struct sctp_endpoint *ep;
-	struct sctp_association *asoc;
-	struct sctp_transport *transport;
-	union sctp_addr to;
-	struct sctp_af *af;
-	sctp_scope_t scope;
-	long timeo;
 	int err = 0;
+	struct sctp_af *af;
 
 	sctp_lock_sock(sk);
 
-	SCTP_DEBUG_PRINTK("%s - sk: %p, sockaddr: %p, addr_len: %d)\n",
-			  __FUNCTION__, sk, uaddr, addr_len);
-
-	sp = sctp_sk(sk);
-	ep = sp->ep;
-
-	/* connect() cannot be done on a socket that is already in ESTABLISHED
-	 * state - UDP-style peeled off socket or a TCP-style socket that
-	 * is already connected.
-	 * It cannot be done even on a TCP-style listening socket.
-	 */
-	if (sctp_sstate(sk, ESTABLISHED) ||
-	    (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING))) {
-		err = -EISCONN;
-		goto out_unlock;
-	}
-
-	err = sctp_verify_addr(sk, (union sctp_addr *)uaddr, addr_len);
-	if (err)
-		goto out_unlock;
-
-	if (addr_len > sizeof(to))
-		addr_len = sizeof(to);
-	memcpy(&to, uaddr, addr_len);
-	to.v4.sin_port = ntohs(to.v4.sin_port);
-
-	asoc = sctp_endpoint_lookup_assoc(ep, &to, &transport);
-	if (asoc) {
-		if (asoc->state >= SCTP_STATE_ESTABLISHED)
-			err = -EISCONN;
-		else
-			err = -EALREADY;
-		goto out_unlock;
-	}
-
-	/* If we could not find a matching association on the endpoint,
-	 * make sure that there is no peeled-off association matching the
-	 * peer address even on another socket.
-	 */
-	if (sctp_endpoint_is_peeled_off(ep, &to)) {
-		err = -EADDRNOTAVAIL;
-		goto out_unlock;
-	}
-
-	/* If a bind() or sctp_bindx() is not called prior to a connect()
-	 * call, the system picks an ephemeral port and will choose an address
-	 * set equivalent to binding with a wildcard address.
-	 */
-	if (!ep->base.bind_addr.port) {
-		if (sctp_autobind(sk)) {
-			err = -EAGAIN;
-			goto out_unlock;
-		}
-	}
-
-	scope = sctp_scope(&to);
-	asoc = sctp_association_new(ep, sk, scope, GFP_KERNEL);
-	if (!asoc) {
-		err = -ENOMEM;
-		goto out_unlock;
-  	}
+	SCTP_DEBUG_PRINTK("%s - sk: %p, sockaddr: %p, addr_len: %d\n",
+			  __FUNCTION__, sk, addr, addr_len);
 
-	/* Prime the peer's transport structures.  */
-	transport = sctp_assoc_add_peer(asoc, &to, GFP_KERNEL);
-	if (!transport) {
-		sctp_association_free(asoc);
-		goto out_unlock;
-	}
-	err = sctp_assoc_set_bind_addr_from_ep(asoc, GFP_KERNEL);
-	if (err < 0) {
-		sctp_association_free(asoc);
-		goto out_unlock;
-	}
-
-	err = sctp_primitive_ASSOCIATE(asoc, NULL);
-	if (err < 0) {
-		sctp_association_free(asoc);
-		goto out_unlock;
+	/* Validate addr_len before calling common connect/connectx routine. */
+	af = sctp_get_af_specific(addr->sa_family);
+	if (!af || addr_len < af->sockaddr_len) {
+		err = -EINVAL;
+	} else {
+		/* Pass correct addr len to common routine (so it knows there
+		 * is only one address being passed.
+		 */
+		err = __sctp_connect(sk, addr, af->sockaddr_len);
 	}
 
-	/* Initialize sk's dport and daddr for getpeername() */
-	inet_sk(sk)->dport = htons(asoc->peer.port);
-	af = sctp_get_af_specific(to.sa.sa_family);
-	af->to_sk_daddr(&to, sk);
-
-	timeo = sock_sndtimeo(sk, sk->sk_socket->file->f_flags & O_NONBLOCK);
-	err = sctp_wait_for_connect(asoc, &timeo);
-
-out_unlock:
 	sctp_release_sock(sk);
-
 	return err;
 }
 
@@ -2677,12 +2842,15 @@ static int sctp_getsockopt_sctp_status(struct sock *sk, int len,
 	/* Map ipv4 address into v4-mapped-on-v6 address.  */
 	sctp_get_pf_specific(sk->sk_family)->addr_v4map(sctp_sk(sk),
 		(union sctp_addr *)&status.sstat_primary.spinfo_address);
-	status.sstat_primary.spinfo_state = transport->active;
+	status.sstat_primary.spinfo_state = transport->state;
 	status.sstat_primary.spinfo_cwnd = transport->cwnd;
 	status.sstat_primary.spinfo_srtt = transport->srtt;
 	status.sstat_primary.spinfo_rto = jiffies_to_msecs(transport->rto);
 	status.sstat_primary.spinfo_mtu = transport->pmtu;
 
+	if (status.sstat_primary.spinfo_state == SCTP_UNKNOWN)
+		status.sstat_primary.spinfo_state = SCTP_ACTIVE;
+
 	if (put_user(len, optlen)) {
 		retval = -EFAULT;
 		goto out;
@@ -2733,12 +2901,15 @@ static int sctp_getsockopt_peer_addr_info(struct sock *sk, int len,
 		return -EINVAL;
 
 	pinfo.spinfo_assoc_id = sctp_assoc2id(transport->asoc);
-	pinfo.spinfo_state = transport->active;
+	pinfo.spinfo_state = transport->state;
 	pinfo.spinfo_cwnd = transport->cwnd;
 	pinfo.spinfo_srtt = transport->srtt;
 	pinfo.spinfo_rto = jiffies_to_msecs(transport->rto);
 	pinfo.spinfo_mtu = transport->pmtu;
 
+	if (pinfo.spinfo_state == SCTP_UNKNOWN)
+		pinfo.spinfo_state = SCTP_ACTIVE;
+
 	if (put_user(len, optlen)) {
 		retval = -EFAULT;
 		goto out;
@@ -3591,7 +3762,8 @@ SCTP_STATIC int sctp_getsockopt(struct sock *sk, int level, int optname,
 	int retval = 0;
 	int len;
 
-	SCTP_DEBUG_PRINTK("sctp_getsockopt(sk: %p, ...)\n", sk);
+	SCTP_DEBUG_PRINTK("sctp_getsockopt(sk: %p... optname: %d)\n",
+			  sk, optname);
 
 	/* I can hardly begin to describe how wrong this is.  This is
 	 * so broken as to be worse than useless.  The API draft
@@ -4596,8 +4768,7 @@ out:
 	return err;
 
 do_error:
-	if (asoc->counters[SCTP_COUNTER_INIT_ERROR] + 1 >=
-					 	asoc->max_init_attempts)
+	if (asoc->init_err_counter + 1 >= asoc->max_init_attempts)
 		err = -ETIMEDOUT;
 	else
 		err = -ECONNREFUSED;
diff --git a/net/sctp/transport.c b/net/sctp/transport.c
index f30882e1e96..0ec0fde6e6c 100644
--- a/net/sctp/transport.c
+++ b/net/sctp/transport.c
@@ -83,7 +83,9 @@ static struct sctp_transport *sctp_transport_init(struct sctp_transport *peer,
 	peer->last_time_used = jiffies;
 	peer->last_time_ecne_reduced = jiffies;
 
-	peer->active = SCTP_ACTIVE;
+	peer->init_sent_count = 0;
+
+	peer->state = SCTP_ACTIVE;
 	peer->hb_allowed = 0;
 
 	/* Initialize the default path max_retrans.  */
-- 
cgit v1.2.3-70-g09d2


From 72cb6962a91f2af9eef69a06198e1949c10259ae Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 20 Jun 2005 13:18:08 -0700
Subject: [IPSEC]: Add xfrm_init_state

This patch adds xfrm_init_state which is simply a wrapper that calls
xfrm_get_type and subsequently x->type->init_state.  It also gets rid
of the unused args argument.

Abstracting it out allows us to add common initialisation code, e.g.,
to set family-specific flags.

The add_time setting in xfrm_user.c was deleted because it's already
set by xfrm_state_alloc.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: James Morris <jmorris@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h      |  3 ++-
 net/ipv4/ah4.c          |  2 +-
 net/ipv4/esp4.c         |  2 +-
 net/ipv4/ipcomp.c       | 11 +++--------
 net/ipv4/xfrm4_tunnel.c |  2 +-
 net/ipv6/ah6.c          |  2 +-
 net/ipv6/esp6.c         |  2 +-
 net/ipv6/ipcomp6.c      |  9 ++-------
 net/ipv6/xfrm6_tunnel.c |  2 +-
 net/key/af_key.c        | 12 +++---------
 net/xfrm/xfrm_policy.c  |  1 -
 net/xfrm/xfrm_state.c   | 21 +++++++++++++++++++++
 net/xfrm/xfrm_user.c    |  9 +--------
 13 files changed, 38 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 0e65e02b7a1..77bfdde440f 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -225,7 +225,7 @@ struct xfrm_type
 	struct module		*owner;
 	__u8			proto;
 
-	int			(*init_state)(struct xfrm_state *x, void *args);
+	int			(*init_state)(struct xfrm_state *x);
 	void			(*destructor)(struct xfrm_state *);
 	int			(*input)(struct xfrm_state *, struct xfrm_decap_state *, struct sk_buff *skb);
 	int			(*post_input)(struct xfrm_state *, struct xfrm_decap_state *, struct sk_buff *skb);
@@ -839,6 +839,7 @@ extern int xfrm_replay_check(struct xfrm_state *x, u32 seq);
 extern void xfrm_replay_advance(struct xfrm_state *x, u32 seq);
 extern int xfrm_state_check(struct xfrm_state *x, struct sk_buff *skb);
 extern int xfrm_state_mtu(struct xfrm_state *x, int mtu);
+extern int xfrm_init_state(struct xfrm_state *x);
 extern int xfrm4_rcv(struct sk_buff *skb);
 extern int xfrm4_output(struct sk_buff *skb);
 extern int xfrm4_tunnel_register(struct xfrm_tunnel *handler);
diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c
index 0e98f2235b6..514c85b2631 100644
--- a/net/ipv4/ah4.c
+++ b/net/ipv4/ah4.c
@@ -200,7 +200,7 @@ static void ah4_err(struct sk_buff *skb, u32 info)
 	xfrm_state_put(x);
 }
 
-static int ah_init_state(struct xfrm_state *x, void *args)
+static int ah_init_state(struct xfrm_state *x)
 {
 	struct ah_data *ahp = NULL;
 	struct xfrm_algo_desc *aalg_desc;
diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index eae84cc39d3..ba57446d5d1 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -362,7 +362,7 @@ static void esp_destroy(struct xfrm_state *x)
 	kfree(esp);
 }
 
-static int esp_init_state(struct xfrm_state *x, void *args)
+static int esp_init_state(struct xfrm_state *x)
 {
 	struct esp_data *esp = NULL;
 
diff --git a/net/ipv4/ipcomp.c b/net/ipv4/ipcomp.c
index 1a23c5263b9..2065944fd9e 100644
--- a/net/ipv4/ipcomp.c
+++ b/net/ipv4/ipcomp.c
@@ -236,15 +236,10 @@ static struct xfrm_state *ipcomp_tunnel_create(struct xfrm_state *x)
 	t->props.mode = 1;
 	t->props.saddr.a4 = x->props.saddr.a4;
 	t->props.flags = x->props.flags;
-	
-	t->type = xfrm_get_type(IPPROTO_IPIP, t->props.family);
-	if (t->type == NULL)
-		goto error;
-		
-	if (t->type->init_state(t, NULL))
+
+	if (xfrm_init_state(t))
 		goto error;
 
-	t->km.state = XFRM_STATE_VALID;
 	atomic_set(&t->tunnel_users, 1);
 out:
 	return t;
@@ -422,7 +417,7 @@ static void ipcomp_destroy(struct xfrm_state *x)
 	kfree(ipcd);
 }
 
-static int ipcomp_init_state(struct xfrm_state *x, void *args)
+static int ipcomp_init_state(struct xfrm_state *x)
 {
 	int err;
 	struct ipcomp_data *ipcd;
diff --git a/net/ipv4/xfrm4_tunnel.c b/net/ipv4/xfrm4_tunnel.c
index 413191f585f..e1fe360ed27 100644
--- a/net/ipv4/xfrm4_tunnel.c
+++ b/net/ipv4/xfrm4_tunnel.c
@@ -84,7 +84,7 @@ static void ipip_err(struct sk_buff *skb, u32 info)
 		handler->err_handler(skb, &arg);
 }
 
-static int ipip_init_state(struct xfrm_state *x, void *args)
+static int ipip_init_state(struct xfrm_state *x)
 {
 	if (!x->props.mode)
 		return -EINVAL;
diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c
index e3ecf626cbf..986fdfdccbc 100644
--- a/net/ipv6/ah6.c
+++ b/net/ipv6/ah6.c
@@ -339,7 +339,7 @@ static void ah6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	xfrm_state_put(x);
 }
 
-static int ah6_init_state(struct xfrm_state *x, void *args)
+static int ah6_init_state(struct xfrm_state *x)
 {
 	struct ah_data *ahp = NULL;
 	struct xfrm_algo_desc *aalg_desc;
diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
index be7095d6bab..324db62515a 100644
--- a/net/ipv6/esp6.c
+++ b/net/ipv6/esp6.c
@@ -296,7 +296,7 @@ static void esp6_destroy(struct xfrm_state *x)
 	kfree(esp);
 }
 
-static int esp6_init_state(struct xfrm_state *x, void *args)
+static int esp6_init_state(struct xfrm_state *x)
 {
 	struct esp_data *esp = NULL;
 
diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c
index 6cde5310cd7..423feb46ccc 100644
--- a/net/ipv6/ipcomp6.c
+++ b/net/ipv6/ipcomp6.c
@@ -234,14 +234,9 @@ static struct xfrm_state *ipcomp6_tunnel_create(struct xfrm_state *x)
 	t->props.mode = 1;
 	memcpy(t->props.saddr.a6, x->props.saddr.a6, sizeof(struct in6_addr));
 
-	t->type = xfrm_get_type(IPPROTO_IPV6, t->props.family);
-	if (t->type == NULL)
+	if (xfrm_init_state(t))
 		goto error;
 
-	if (t->type->init_state(t, NULL))
-		goto error;
-
-	t->km.state = XFRM_STATE_VALID;
 	atomic_set(&t->tunnel_users, 1);
 
 out:
@@ -420,7 +415,7 @@ static void ipcomp6_destroy(struct xfrm_state *x)
 	xfrm6_tunnel_free_spi((xfrm_address_t *)&x->props.saddr);
 }
 
-static int ipcomp6_init_state(struct xfrm_state *x, void *args)
+static int ipcomp6_init_state(struct xfrm_state *x)
 {
 	int err;
 	struct ipcomp_data *ipcd;
diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c
index ffcadd68b95..60c26c87277 100644
--- a/net/ipv6/xfrm6_tunnel.c
+++ b/net/ipv6/xfrm6_tunnel.c
@@ -466,7 +466,7 @@ static void xfrm6_tunnel_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	return;
 }
 
-static int xfrm6_tunnel_init_state(struct xfrm_state *x, void *args)
+static int xfrm6_tunnel_init_state(struct xfrm_state *x)
 {
 	if (!x->props.mode)
 		return -EINVAL;
diff --git a/net/key/af_key.c b/net/key/af_key.c
index 98b72f2024f..652dd09ccd3 100644
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
@@ -1096,17 +1096,11 @@ static struct xfrm_state * pfkey_msg2xfrm_state(struct sadb_msg *hdr,
 		}
 	}
 
-	x->type = xfrm_get_type(proto, x->props.family);
-	if (x->type == NULL) {
-		err = -ENOPROTOOPT;
-		goto out;
-	}
-	if (x->type->init_state(x, NULL)) {
-		err = -EINVAL;
+	err = xfrm_init_state(x);
+	if (err)
 		goto out;
-	}
+
 	x->km.seq = hdr->sadb_msg_seq;
-	x->km.state = XFRM_STATE_VALID;
 	return x;
 
 out:
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 0a4260719a1..d65ed8684fc 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -118,7 +118,6 @@ retry:
 	xfrm_policy_put_afinfo(afinfo);
 	return type;
 }
-EXPORT_SYMBOL(xfrm_get_type);
 
 int xfrm_dst_lookup(struct xfrm_dst **dst, struct flowi *fl, 
 		    unsigned short family)
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 2537f26f097..1845b73d69f 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -1055,6 +1055,27 @@ int xfrm_state_mtu(struct xfrm_state *x, int mtu)
 }
 
 EXPORT_SYMBOL(xfrm_state_mtu);
+
+int xfrm_init_state(struct xfrm_state *x)
+{
+	int err;
+
+	err = -ENOENT;
+	x->type = xfrm_get_type(x->id.proto, x->props.family);
+	if (x->type == NULL)
+		goto error;
+
+	err = x->type->init_state(x);
+	if (err)
+		goto error;
+
+	x->km.state = XFRM_STATE_VALID;
+
+error:
+	return err;
+}
+
+EXPORT_SYMBOL(xfrm_init_state);
  
 void __init xfrm_state_init(void)
 {
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index 5ce8558eac9..ecade4893a1 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -249,17 +249,10 @@ static struct xfrm_state *xfrm_state_construct(struct xfrm_usersa_info *p,
 	if ((err = attach_encap_tmpl(&x->encap, xfrma[XFRMA_ENCAP-1])))
 		goto error;
 
-	err = -ENOENT;
-	x->type = xfrm_get_type(x->id.proto, x->props.family);
-	if (x->type == NULL)
-		goto error;
-
-	err = x->type->init_state(x, NULL);
+	err = xfrm_init_state(x);
 	if (err)
 		goto error;
 
-	x->curlft.add_time = (unsigned long) xtime.tv_sec;
-	x->km.state = XFRM_STATE_VALID;
 	x->km.seq = p->seq;
 
 	return x;
-- 
cgit v1.2.3-70-g09d2


From d094cd83c06e06e01d8edb540555f3f64e4081c2 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 20 Jun 2005 13:19:41 -0700
Subject: [IPSEC]: Add xfrm_state_afinfo->init_flags

This patch adds the xfrm_state_afinfo->init_flags hook which allows
each address family to perform any common initialisation that does
not require a corresponding destructor call.

It will be used subsequently to set the XFRM_STATE_NOPMTUDISC flag
in IPv4.

It also fixes up the error codes returned by xfrm_init_state.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: James Morris <jmorris@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h    |  1 +
 net/xfrm/xfrm_state.c | 20 ++++++++++++++++++--
 2 files changed, 19 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 77bfdde440f..029522a4ced 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -204,6 +204,7 @@ struct xfrm_state_afinfo {
 	rwlock_t		lock;
 	struct list_head	*state_bydst;
 	struct list_head	*state_byspi;
+	int			(*init_flags)(struct xfrm_state *x);
 	void			(*init_tempsel)(struct xfrm_state *x, struct flowi *fl,
 						struct xfrm_tmpl *tmpl,
 						xfrm_address_t *daddr, xfrm_address_t *saddr);
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 1845b73d69f..9d206c282cf 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -1058,10 +1058,26 @@ EXPORT_SYMBOL(xfrm_state_mtu);
 
 int xfrm_init_state(struct xfrm_state *x)
 {
+	struct xfrm_state_afinfo *afinfo;
+	int family = x->props.family;
 	int err;
 
-	err = -ENOENT;
-	x->type = xfrm_get_type(x->id.proto, x->props.family);
+	err = -EAFNOSUPPORT;
+	afinfo = xfrm_state_get_afinfo(family);
+	if (!afinfo)
+		goto error;
+
+	err = 0;
+	if (afinfo->init_flags)
+		err = afinfo->init_flags(x);
+
+	xfrm_state_put_afinfo(afinfo);
+
+	if (err)
+		goto error;
+
+	err = -EPROTONOSUPPORT;
+	x->type = xfrm_get_type(x->id.proto, family);
 	if (x->type == NULL)
 		goto error;
 
-- 
cgit v1.2.3-70-g09d2


From dd87147eed934eaff92869f3d158697c7239d1d2 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 20 Jun 2005 13:21:43 -0700
Subject: [IPSEC]: Add XFRM_STATE_NOPMTUDISC flag

This patch adds the flag XFRM_STATE_NOPMTUDISC for xfrm states.  It is
similar to the nopmtudisc on IPIP/GRE tunnels.  It only has an effect
on IPv4 tunnel mode states.  For these states, it will ensure that the
DF flag is always cleared.

This is primarily useful to work around ICMP blackholes.

In future this flag could also allow a larger MTU to be set within the
tunnel just like IPIP/GRE tunnels.  This could be useful for short haul
tunnels where temporary fragmentation outside the tunnel is desired over
smaller fragments inside the tunnel.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: James Morris <jmorris@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pfkeyv2.h | 1 +
 include/linux/xfrm.h    | 1 +
 net/ipv4/xfrm4_output.c | 8 ++++++--
 net/ipv4/xfrm4_state.c  | 9 +++++++++
 net/key/af_key.c        | 4 ++++
 5 files changed, 21 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/pfkeyv2.h b/include/linux/pfkeyv2.h
index e6b51922024..724066778af 100644
--- a/include/linux/pfkeyv2.h
+++ b/include/linux/pfkeyv2.h
@@ -245,6 +245,7 @@ struct sadb_x_nat_t_port {
 
 /* Security Association flags */
 #define SADB_SAFLAGS_PFS	1
+#define SADB_SAFLAGS_NOPMTUDISC	0x20000000
 #define SADB_SAFLAGS_DECAP_DSCP	0x40000000
 #define SADB_SAFLAGS_NOECN	0x80000000
 
diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index d68391a9b9f..f0d423300d8 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -196,6 +196,7 @@ struct xfrm_usersa_info {
 	__u8				flags;
 #define XFRM_STATE_NOECN	1
 #define XFRM_STATE_DECAP_DSCP	2
+#define XFRM_STATE_NOPMTUDISC	4
 };
 
 struct xfrm_usersa_id {
diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c
index af2392ae576..66620a95942 100644
--- a/net/ipv4/xfrm4_output.c
+++ b/net/ipv4/xfrm4_output.c
@@ -33,6 +33,7 @@ static void xfrm4_encap(struct sk_buff *skb)
 	struct dst_entry *dst = skb->dst;
 	struct xfrm_state *x = dst->xfrm;
 	struct iphdr *iph, *top_iph;
+	int flags;
 
 	iph = skb->nh.iph;
 	skb->h.ipiph = iph;
@@ -51,10 +52,13 @@ static void xfrm4_encap(struct sk_buff *skb)
 
 	/* DS disclosed */
 	top_iph->tos = INET_ECN_encapsulate(iph->tos, iph->tos);
-	if (x->props.flags & XFRM_STATE_NOECN)
+
+	flags = x->props.flags;
+	if (flags & XFRM_STATE_NOECN)
 		IP_ECN_clear(top_iph);
 
-	top_iph->frag_off = iph->frag_off & htons(IP_DF);
+	top_iph->frag_off = (flags & XFRM_STATE_NOPMTUDISC) ?
+		0 : (iph->frag_off & htons(IP_DF));
 	if (!top_iph->frag_off)
 		__ip_select_ident(top_iph, dst, 0);
 
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index 223a2e83853..050611d7a96 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -7,12 +7,20 @@
  *
  */
 
+#include <net/ip.h>
 #include <net/xfrm.h>
 #include <linux/pfkeyv2.h>
 #include <linux/ipsec.h>
 
 static struct xfrm_state_afinfo xfrm4_state_afinfo;
 
+static int xfrm4_init_flags(struct xfrm_state *x)
+{
+	if (ipv4_config.no_pmtu_disc)
+		x->props.flags |= XFRM_STATE_NOPMTUDISC;
+	return 0;
+}
+
 static void
 __xfrm4_init_tempsel(struct xfrm_state *x, struct flowi *fl,
 		     struct xfrm_tmpl *tmpl,
@@ -109,6 +117,7 @@ __xfrm4_find_acq(u8 mode, u32 reqid, u8 proto,
 static struct xfrm_state_afinfo xfrm4_state_afinfo = {
 	.family			= AF_INET,
 	.lock			= RW_LOCK_UNLOCKED,
+	.init_flags		= xfrm4_init_flags,
 	.init_tempsel		= __xfrm4_init_tempsel,
 	.state_lookup		= __xfrm4_state_lookup,
 	.find_acq		= __xfrm4_find_acq,
diff --git a/net/key/af_key.c b/net/key/af_key.c
index 652dd09ccd3..4879743b945 100644
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
@@ -690,6 +690,8 @@ static struct sk_buff * pfkey_xfrm_state2msg(struct xfrm_state *x, int add_keys,
 		sa->sadb_sa_flags |= SADB_SAFLAGS_NOECN;
 	if (x->props.flags & XFRM_STATE_DECAP_DSCP)
 		sa->sadb_sa_flags |= SADB_SAFLAGS_DECAP_DSCP;
+	if (x->props.flags & XFRM_STATE_NOPMTUDISC)
+		sa->sadb_sa_flags |= SADB_SAFLAGS_NOPMTUDISC;
 
 	/* hard time */
 	if (hsc & 2) {
@@ -974,6 +976,8 @@ static struct xfrm_state * pfkey_msg2xfrm_state(struct sadb_msg *hdr,
 		x->props.flags |= XFRM_STATE_NOECN;
 	if (sa->sadb_sa_flags & SADB_SAFLAGS_DECAP_DSCP)
 		x->props.flags |= XFRM_STATE_DECAP_DSCP;
+	if (sa->sadb_sa_flags & SADB_SAFLAGS_NOPMTUDISC)
+		x->props.flags |= XFRM_STATE_NOPMTUDISC;
 
 	lifetime = (struct sadb_lifetime*) ext_hdrs[SADB_EXT_LIFETIME_HARD-1];
 	if (lifetime != NULL) {
-- 
cgit v1.2.3-70-g09d2


From f852640e74f71e6dd38146e1149ec1fe6da2fb07 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Mon, 20 Jun 2005 13:31:11 -0700
Subject: [AX25]: endian-annotate ax25_type_trans()

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Acked-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ax25.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/ax25.h b/include/net/ax25.h
index 9e6368a5454..828a3a93dda 100644
--- a/include/net/ax25.h
+++ b/include/net/ax25.h
@@ -220,7 +220,7 @@ static __inline__ void ax25_cb_put(ax25_cb *ax25)
 	}
 }
 
-static inline unsigned short ax25_type_trans(struct sk_buff *skb, struct net_device *dev)
+static inline __be16 ax25_type_trans(struct sk_buff *skb, struct net_device *dev)
 {
 	skb->dev      = dev;
 	skb->pkt_type = PACKET_HOST;
-- 
cgit v1.2.3-70-g09d2


From f6e276ee67c0ac9efafd24bc6f7a84aa359656df Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Mon, 20 Jun 2005 13:32:05 -0700
Subject: [ATALK]: endian annotations

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/atalk.h | 26 +++++++++++++-------------
 net/appletalk/aarp.c  |  2 +-
 net/appletalk/ddp.c   |  2 +-
 3 files changed, 15 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/atalk.h b/include/linux/atalk.h
index 31d3fc25ccb..09a1451c115 100644
--- a/include/linux/atalk.h
+++ b/include/linux/atalk.h
@@ -20,7 +20,7 @@
 #define SIOCATALKDIFADDR       (SIOCPROTOPRIVATE + 0)
 
 struct atalk_addr {
-	__u16	s_net;
+	__be16	s_net;
 	__u8	s_node;
 };
 
@@ -33,8 +33,8 @@ struct sockaddr_at {
 
 struct atalk_netrange {
 	__u8	nr_phase;
-	__u16	nr_firstnet;
-	__u16	nr_lastnet;
+	__be16	nr_firstnet;
+	__be16	nr_lastnet;
 };
 
 #ifdef __KERNEL__
@@ -70,8 +70,8 @@ struct atalk_iface {
 struct atalk_sock {
 	/* struct sock has to be the first member of atalk_sock */
 	struct sock	sk;
-	unsigned short	dest_net;
-	unsigned short	src_net;
+	__be16		dest_net;
+	__be16		src_net;
 	unsigned char	dest_node;
 	unsigned char	src_node;
 	unsigned char	dest_port;
@@ -95,9 +95,9 @@ struct ddpehdr {
 		deh_hops:4,
 		deh_len:10;
 #endif
-	__u16	deh_sum;
-	__u16	deh_dnet;
-	__u16	deh_snet;
+	__be16	deh_sum;
+	__be16	deh_dnet;
+	__be16	deh_snet;
 	__u8	deh_dnode;
 	__u8	deh_snode;
 	__u8	deh_dport;
@@ -142,24 +142,24 @@ struct ddpshdr {
 
 /* AppleTalk AARP headers */
 struct elapaarp {
-	__u16	hw_type;
+	__be16	hw_type;
 #define AARP_HW_TYPE_ETHERNET		1
 #define AARP_HW_TYPE_TOKENRING		2
-	__u16	pa_type;
+	__be16	pa_type;
 	__u8	hw_len;
 	__u8	pa_len;
 #define AARP_PA_ALEN			4
-	__u16	function;
+	__be16	function;
 #define AARP_REQUEST			1
 #define AARP_REPLY			2
 #define AARP_PROBE			3
 	__u8	hw_src[ETH_ALEN]	__attribute__ ((packed));
 	__u8	pa_src_zero		__attribute__ ((packed));
-	__u16	pa_src_net		__attribute__ ((packed));
+	__be16	pa_src_net		__attribute__ ((packed));
 	__u8	pa_src_node		__attribute__ ((packed));
 	__u8	hw_dst[ETH_ALEN]	__attribute__ ((packed));
 	__u8	pa_dst_zero		__attribute__ ((packed));
-	__u16	pa_dst_net		__attribute__ ((packed));
+	__be16	pa_dst_net		__attribute__ ((packed));
 	__u8	pa_dst_node		__attribute__ ((packed));	
 };
 
diff --git a/net/appletalk/aarp.c b/net/appletalk/aarp.c
index 54640c01b50..10d04046102 100644
--- a/net/appletalk/aarp.c
+++ b/net/appletalk/aarp.c
@@ -565,7 +565,7 @@ int aarp_send_ddp(struct net_device *dev, struct sk_buff *skb,
 			 *	numbers	we just happen to need. Now put the
 			 *	length in the lower two.
 			 */
-			*((__u16 *)skb->data) = htons(skb->len);
+			*((__be16 *)skb->data) = htons(skb->len);
 			ft = 1;
 		}
 		/*
diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c
index 876dbac7106..192b529f86a 100644
--- a/net/appletalk/ddp.c
+++ b/net/appletalk/ddp.c
@@ -401,7 +401,7 @@ out_err:
 }
 
 /* Find a match for a specific network:node pair */
-static struct atalk_iface *atalk_find_interface(int net, int node)
+static struct atalk_iface *atalk_find_interface(__be16 net, int node)
 {
 	struct atalk_iface *iface;
 
-- 
cgit v1.2.3-70-g09d2


From 246955fe4c38bd706ae30e37c64892c94213775d Mon Sep 17 00:00:00 2001
From: Robert Olsson <Robert.Olsson@data.slu.se>
Date: Mon, 20 Jun 2005 13:36:39 -0700
Subject: [NETLINK]: fib_lookup() via netlink

Below is a more generic patch to do fib_lookup via netlink. For others
we should say that we discussed this as a way to verify route selection.
It's also possible there are others uses for this.

In short the fist half of struct fib_result_nl is filled in by caller
and netlink call fills in the other half and returns it.

In case anyone is interested there is a corresponding user app to compare
the full routing table this was used to test implementation of the LC-trie.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netlink.h |  1 +
 include/net/ip_fib.h    | 14 +++++++++++++
 net/ipv4/fib_frontend.c | 55 +++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 70 insertions(+)

(limited to 'include')

diff --git a/include/linux/netlink.h b/include/linux/netlink.h
index e38407a23d0..561d4dc7583 100644
--- a/include/linux/netlink.h
+++ b/include/linux/netlink.h
@@ -14,6 +14,7 @@
 #define NETLINK_SELINUX		7	/* SELinux event notifications */
 #define NETLINK_ARPD		8
 #define NETLINK_AUDIT		9	/* auditing */
+#define NETLINK_FIB_LOOKUP	10	
 #define NETLINK_ROUTE6		11	/* af_inet6 route comm channel */
 #define NETLINK_IP6_FW		13
 #define NETLINK_DNRTMSG		14	/* DECnet routing messages */
diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index e5a5f6b62f8..a4208a336ac 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -109,6 +109,20 @@ struct fib_result {
 #endif
 };
 
+struct fib_result_nl {
+	u32		fl_addr;   /* To be looked up*/ 
+	u32		fl_fwmark; 
+	unsigned char	fl_tos;
+	unsigned char   fl_scope;
+	unsigned char   tb_id_in;
+
+	unsigned char   tb_id;      /* Results */
+	unsigned char	prefixlen;
+	unsigned char	nh_sel;
+	unsigned char	type;
+	unsigned char	scope;
+	int             err;      
+};
 
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
 
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 563e7d61270..cd8e45ab958 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -516,6 +516,60 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa)
 #undef BRD1_OK
 }
 
+static void nl_fib_lookup(struct fib_result_nl *frn, struct fib_table *tb )
+{
+	
+	struct fib_result       res;
+	struct flowi            fl = { .nl_u = { .ip4_u = { .daddr = frn->fl_addr, 
+							    .fwmark = frn->fl_fwmark,
+							    .tos = frn->fl_tos,
+							    .scope = frn->fl_scope } } };
+	if (tb) {
+		local_bh_disable();
+
+		frn->tb_id = tb->tb_id;
+		frn->err = tb->tb_lookup(tb, &fl, &res);
+
+		if (!frn->err) {
+			frn->prefixlen = res.prefixlen;
+			frn->nh_sel = res.nh_sel;
+			frn->type = res.type;
+			frn->scope = res.scope;
+		}
+		local_bh_enable();
+	}
+}
+
+static void nl_fib_input(struct sock *sk, int len)
+{
+	struct sk_buff *skb = NULL;
+        struct nlmsghdr *nlh = NULL;
+	struct fib_result_nl *frn;
+	int err;
+	u32 pid;     
+	struct fib_table *tb;
+	
+	skb = skb_recv_datagram(sk, 0, 0, &err);
+	nlh = (struct nlmsghdr *)skb->data;
+	
+	frn = (struct fib_result_nl *) NLMSG_DATA(nlh);
+	tb = fib_get_table(frn->tb_id_in);
+
+	nl_fib_lookup(frn, tb);
+	
+	pid = nlh->nlmsg_pid;           /*pid of sending process */
+	NETLINK_CB(skb).groups = 0;     /* not in mcast group */
+	NETLINK_CB(skb).pid = 0;         /* from kernel */
+	NETLINK_CB(skb).dst_pid = pid;
+	NETLINK_CB(skb).dst_groups = 0;  /* unicast */
+	netlink_unicast(sk, skb, pid, MSG_DONTWAIT);
+}    
+
+static void nl_fib_lookup_init(void)
+{
+      netlink_kernel_create(NETLINK_FIB_LOOKUP, nl_fib_input);
+}
+
 static void fib_disable_ip(struct net_device *dev, int force)
 {
 	if (fib_sync_down(0, dev, force))
@@ -604,6 +658,7 @@ void __init ip_fib_init(void)
 
 	register_netdevice_notifier(&fib_netdev_notifier);
 	register_inetaddr_notifier(&fib_inetaddr_notifier);
+	nl_fib_lookup_init();
 }
 
 EXPORT_SYMBOL(inet_addr_type);
-- 
cgit v1.2.3-70-g09d2


From e3a15db2415579d5136b9ba9b52fe27c66da8780 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor_core@ameritech.net>
Date: Tue, 26 Apr 2005 02:31:08 -0500
Subject: [PATCH] sysfs_{create|remove}_link should take const char *

sysfs: make sysfs_{create|remove}_link to take const char * name.

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 fs/sysfs/symlink.c    |  8 ++++----
 include/linux/sysfs.h | 10 +++++-----
 2 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/fs/sysfs/symlink.c b/fs/sysfs/symlink.c
index dfdf7017435..fae57c83a72 100644
--- a/fs/sysfs/symlink.c
+++ b/fs/sysfs/symlink.c
@@ -43,7 +43,7 @@ static void fill_object_path(struct kobject * kobj, char * buffer, int length)
 	}
 }
 
-static int sysfs_add_link(struct dentry * parent, char * name, struct kobject * target)
+static int sysfs_add_link(struct dentry * parent, const char * name, struct kobject * target)
 {
 	struct sysfs_dirent * parent_sd = parent->d_fsdata;
 	struct sysfs_symlink * sl;
@@ -79,7 +79,7 @@ exit1:
  *	@target:	object we're pointing to.
  *	@name:		name of the symlink.
  */
-int sysfs_create_link(struct kobject * kobj, struct kobject * target, char * name)
+int sysfs_create_link(struct kobject * kobj, struct kobject * target, const char * name)
 {
 	struct dentry * dentry = kobj->dentry;
 	int error = 0;
@@ -99,13 +99,13 @@ int sysfs_create_link(struct kobject * kobj, struct kobject * target, char * nam
  *	@name:	name of the symlink to remove.
  */
 
-void sysfs_remove_link(struct kobject * kobj, char * name)
+void sysfs_remove_link(struct kobject * kobj, const char * name)
 {
 	sysfs_hash_and_remove(kobj->dentry,name);
 }
 
 static int sysfs_get_target_path(struct kobject * kobj, struct kobject * target,
-				   char *path)
+				 char *path)
 {
 	char * s;
 	int depth, size;
diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
index 38b58b30814..931b5aaaf38 100644
--- a/include/linux/sysfs.h
+++ b/include/linux/sysfs.h
@@ -105,11 +105,11 @@ sysfs_chmod_file(struct kobject *kobj, struct attribute *attr, mode_t mode);
 extern void
 sysfs_remove_file(struct kobject *, const struct attribute *);
 
-extern int 
-sysfs_create_link(struct kobject * kobj, struct kobject * target, char * name);
+extern int
+sysfs_create_link(struct kobject * kobj, struct kobject * target, const char * name);
 
 extern void
-sysfs_remove_link(struct kobject *, char * name);
+sysfs_remove_link(struct kobject *, const char * name);
 
 int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr);
 int sysfs_remove_bin_file(struct kobject * kobj, struct bin_attribute * attr);
@@ -153,12 +153,12 @@ static inline void sysfs_remove_file(struct kobject * k, const struct attribute
 	;
 }
 
-static inline int sysfs_create_link(struct kobject * k, struct kobject * t, char * n)
+static inline int sysfs_create_link(struct kobject * k, struct kobject * t, const char * n)
 {
 	return 0;
 }
 
-static inline void sysfs_remove_link(struct kobject * k, char * name)
+static inline void sysfs_remove_link(struct kobject * k, const char * name)
 {
 	;
 }
-- 
cgit v1.2.3-70-g09d2


From f3b4f3c6dec04c6c8261fe22645f07b39976595a Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor_core@ameritech.net>
Date: Tue, 26 Apr 2005 02:32:00 -0500
Subject: [PATCH] Make kobject's name be const char *

kobject: make kobject's name const char * since users should not
	 attempt to change it (except by calling kobject_rename).

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h | 6 +++---
 lib/kobject.c           | 2 +-
 lib/kobject_uevent.c    | 4 ++--
 3 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 765d660d3be..76dc67245c0 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -33,7 +33,7 @@
 extern u64 hotplug_seqnum;
 
 struct kobject {
-	char			* k_name;
+	const char		* k_name;
 	char			name[KOBJ_NAME_LEN];
 	struct kref		kref;
 	struct list_head	entry;
@@ -46,7 +46,7 @@ struct kobject {
 extern int kobject_set_name(struct kobject *, const char *, ...)
 	__attribute__((format(printf,2,3)));
 
-static inline char * kobject_name(struct kobject * kobj)
+static inline const char * kobject_name(const struct kobject * kobj)
 {
 	return kobj->k_name;
 }
@@ -57,7 +57,7 @@ extern void kobject_cleanup(struct kobject *);
 extern int kobject_add(struct kobject *);
 extern void kobject_del(struct kobject *);
 
-extern int kobject_rename(struct kobject *, char *new_name);
+extern int kobject_rename(struct kobject *, const char *new_name);
 
 extern int kobject_register(struct kobject *);
 extern void kobject_unregister(struct kobject *);
diff --git a/lib/kobject.c b/lib/kobject.c
index 94048826624..dd0917dd9fa 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -279,7 +279,7 @@ EXPORT_SYMBOL(kobject_set_name);
  *	@new_name: object's new name
  */
 
-int kobject_rename(struct kobject * kobj, char *new_name)
+int kobject_rename(struct kobject * kobj, const char *new_name)
 {
 	int error = 0;
 
diff --git a/lib/kobject_uevent.c b/lib/kobject_uevent.c
index 1ede5aa3337..8e49d21057e 100644
--- a/lib/kobject_uevent.c
+++ b/lib/kobject_uevent.c
@@ -197,7 +197,7 @@ void kobject_hotplug(struct kobject *kobj, enum kobject_action action)
 	int i = 0;
 	int retval;
 	char *kobj_path = NULL;
-	char *name = NULL;
+	const char *name = NULL;
 	char *action_string;
 	u64 seq;
 	struct kobject *top_kobj = kobj;
@@ -249,7 +249,7 @@ void kobject_hotplug(struct kobject *kobj, enum kobject_action action)
 		name = kobject_name(&kset->kobj);
 
 	argv [0] = hotplug_path;
-	argv [1] = name;
+	argv [1] = (char *)name; /* won't be changed but 'const' has to go */
 	argv [2] = NULL;
 
 	/* minimal command environment */
-- 
cgit v1.2.3-70-g09d2


From 419cab3fc69588ebe35b845cc3a584ae172463de Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor_core@ameritech.net>
Date: Tue, 26 Apr 2005 02:32:54 -0500
Subject: [PATCH] kset_hotplug_ops->name shoudl return const char *

kobject: change name() method in kset_hotplug_ops return const char *
	 since users shoudl not try to modify returned data.

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/class.c    | 2 +-
 drivers/base/core.c     | 2 +-
 include/linux/kobject.h | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/drivers/base/class.c b/drivers/base/class.c
index d2a2f8f2b4e..ff33ebbf1dc 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -262,7 +262,7 @@ static int class_hotplug_filter(struct kset *kset, struct kobject *kobj)
 	return 0;
 }
 
-static char *class_hotplug_name(struct kset *kset, struct kobject *kobj)
+static const char *class_hotplug_name(struct kset *kset, struct kobject *kobj)
 {
 	struct class_device *class_dev = to_class_dev(kobj);
 
diff --git a/drivers/base/core.c b/drivers/base/core.c
index fbc223486f8..32f3f09a42e 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -102,7 +102,7 @@ static int dev_hotplug_filter(struct kset *kset, struct kobject *kobj)
 	return 0;
 }
 
-static char *dev_hotplug_name(struct kset *kset, struct kobject *kobj)
+static const char *dev_hotplug_name(struct kset *kset, struct kobject *kobj)
 {
 	struct device *dev = to_dev(kobj);
 
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 76dc67245c0..3b22304f12f 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -94,7 +94,7 @@ struct kobj_type {
  */
 struct kset_hotplug_ops {
 	int (*filter)(struct kset *kset, struct kobject *kobj);
-	char *(*name)(struct kset *kset, struct kobject *kobj);
+	const char *(*name)(struct kset *kset, struct kobject *kobj);
 	int (*hotplug)(struct kset *kset, struct kobject *kobj, char **envp,
 			int num_envp, char *buffer, int buffer_size);
 };
-- 
cgit v1.2.3-70-g09d2


From 8d790d74085833ba2a3e84b5bcd683be4981c29a Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor_core@ameritech.net>
Date: Tue, 26 Apr 2005 02:34:05 -0500
Subject: [PATCH] make driver's name be const char *

Driver core:
  change driver's, bus's, class's and platform device's names
  to be const char * so one can use
            const char *drv_name = "asdfg";
  when initializing structures.
  Also kill couple of whitespaces.

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/devices.c |  2 +-
 include/linux/device.h     | 12 ++++++------
 2 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/drivers/usb/core/devices.c b/drivers/usb/core/devices.c
index ef0b35731ff..83e815d3cd5 100644
--- a/drivers/usb/core/devices.c
+++ b/drivers/usb/core/devices.c
@@ -239,7 +239,7 @@ static char *usb_dump_interface_descriptor(char *start, char *end,
 	int setno)
 {
 	const struct usb_interface_descriptor *desc = &intfc->altsetting[setno].desc;
-	char *driver_name = "";
+	const char *driver_name = "";
 
 	if (start > end)
 		return start;
diff --git a/include/linux/device.h b/include/linux/device.h
index df94c0de53f..fa9e6ca08f5 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -47,7 +47,7 @@ struct class_device;
 struct class_simple;
 
 struct bus_type {
-	char			* name;
+	const char		* name;
 
 	struct subsystem	subsys;
 	struct kset		drivers;
@@ -98,17 +98,17 @@ extern int bus_create_file(struct bus_type *, struct bus_attribute *);
 extern void bus_remove_file(struct bus_type *, struct bus_attribute *);
 
 struct device_driver {
-	char			* name;
+	const char		* name;
 	struct bus_type		* bus;
 
 	struct completion	unloaded;
 	struct kobject		kobj;
 	struct list_head	devices;
 
-	struct module 		* owner;
+	struct module		* owner;
 
 	int	(*probe)	(struct device * dev);
-	int 	(*remove)	(struct device * dev);
+	int	(*remove)	(struct device * dev);
 	void	(*shutdown)	(struct device * dev);
 	int	(*suspend)	(struct device * dev, pm_message_t state, u32 level);
 	int	(*resume)	(struct device * dev, u32 level);
@@ -142,7 +142,7 @@ extern void driver_remove_file(struct device_driver *, struct driver_attribute *
  * device classes
  */
 struct class {
-	char			* name;
+	const char		* name;
 
 	struct subsystem	subsys;
 	struct list_head	children;
@@ -366,7 +366,7 @@ extern struct device *device_find(const char *name, struct bus_type *bus);
 /* drivers/base/platform.c */
 
 struct platform_device {
-	char		* name;
+	const char	* name;
 	u32		id;
 	struct device	dev;
 	u32		num_resources;
-- 
cgit v1.2.3-70-g09d2


From d48593bf208e0d046c35fb0707ae5b23fef8c4ff Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor_core@ameritech.net>
Date: Fri, 29 Apr 2005 00:58:46 -0500
Subject: [PATCH] Make attributes names const char *

sysfs: make attributes and attribute_group's names const char *

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/infiniband/core/sysfs.c | 120 ++++++++++++++++------------------------
 drivers/pci/pci-sysfs.c         |  13 +++--
 include/linux/sysfs.h           |   4 +-
 3 files changed, 58 insertions(+), 79 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c
index 3a413f72ff6..5febd6d8b88 100644
--- a/drivers/infiniband/core/sysfs.c
+++ b/drivers/infiniband/core/sysfs.c
@@ -40,9 +40,7 @@ struct ib_port {
 	struct kobject         kobj;
 	struct ib_device      *ibdev;
 	struct attribute_group gid_group;
-	struct attribute     **gid_attr;
 	struct attribute_group pkey_group;
-	struct attribute     **pkey_attr;
 	u8                     port_num;
 };
 
@@ -60,8 +58,9 @@ struct port_attribute port_attr_##_name = __ATTR(_name, _mode, _show, _store)
 struct port_attribute port_attr_##_name = __ATTR_RO(_name)
 
 struct port_table_attribute {
-	struct port_attribute attr;
-	int                   index;
+	struct port_attribute	attr;
+	char			name[8];
+	int			index;
 };
 
 static ssize_t port_attr_show(struct kobject *kobj,
@@ -398,17 +397,16 @@ static void ib_port_release(struct kobject *kobj)
 	struct attribute *a;
 	int i;
 
-	for (i = 0; (a = p->gid_attr[i]); ++i) {
-		kfree(a->name);
+	for (i = 0; (a = p->gid_group.attrs[i]); ++i)
 		kfree(a);
-	}
 
-	for (i = 0; (a = p->pkey_attr[i]); ++i) {
-		kfree(a->name);
+	kfree(p->gid_group.attrs);
+
+	for (i = 0; (a = p->pkey_group.attrs[i]); ++i)
 		kfree(a);
-	}
 
-	kfree(p->gid_attr);
+	kfree(p->pkey_group.attrs);
+
 	kfree(p);
 }
 
@@ -449,58 +447,45 @@ static int ib_device_hotplug(struct class_device *cdev, char **envp,
 	return 0;
 }
 
-static int alloc_group(struct attribute ***attr,
-		       ssize_t (*show)(struct ib_port *,
-				       struct port_attribute *, char *buf),
-		       int len)
+static struct attribute **
+alloc_group_attrs(ssize_t (*show)(struct ib_port *,
+				  struct port_attribute *, char *buf),
+		  int len)
 {
-	struct port_table_attribute ***tab_attr =
-		(struct port_table_attribute ***) attr;
+	struct attribute **tab_attr;
+	struct port_table_attribute *element;
 	int i;
-	int ret;
-
-	*tab_attr = kmalloc((1 + len) * sizeof *tab_attr, GFP_KERNEL);
-	if (!*tab_attr)
-		return -ENOMEM;
 
-	memset(*tab_attr, 0, (1 + len) * sizeof *tab_attr);
+	tab_attr = kcalloc(1 + len, sizeof(struct attribute *), GFP_KERNEL);
+	if (!tab_attr)
+		return NULL;
 
-	for (i = 0; i < len; ++i) {
-		(*tab_attr)[i] = kmalloc(sizeof *(*tab_attr)[i], GFP_KERNEL);
-		if (!(*tab_attr)[i]) {
-			ret = -ENOMEM;
+	for (i = 0; i < len; i++) {
+		element = kcalloc(1, sizeof(struct port_table_attribute),
+				  GFP_KERNEL);
+		if (!element)
 			goto err;
-		}
-		memset((*tab_attr)[i], 0, sizeof *(*tab_attr)[i]);
-		(*tab_attr)[i]->attr.attr.name = kmalloc(8, GFP_KERNEL);
-		if (!(*tab_attr)[i]->attr.attr.name) {
-			ret = -ENOMEM;
-			goto err;
-		}
 
-		if (snprintf((*tab_attr)[i]->attr.attr.name, 8, "%d", i) >= 8) {
-			ret = -ENOMEM;
+		if (snprintf(element->name, sizeof(element->name),
+			     "%d", i) >= sizeof(element->name))
 			goto err;
-		}
 
-		(*tab_attr)[i]->attr.attr.mode  = S_IRUGO;
-		(*tab_attr)[i]->attr.attr.owner = THIS_MODULE;
-		(*tab_attr)[i]->attr.show       = show;
-		(*tab_attr)[i]->index           = i;
-	}
-
-	return 0;
+		element->attr.attr.name  = element->name;
+		element->attr.attr.mode  = S_IRUGO;
+		element->attr.attr.owner = THIS_MODULE;
+		element->attr.show       = show;
+		element->index		 = i;
 
-err:
-	for (i = 0; i < len; ++i) {
-		if ((*tab_attr)[i])
-			kfree((*tab_attr)[i]->attr.attr.name);
-		kfree((*tab_attr)[i]);
+		tab_attr[i] = &element->attr.attr;
 	}
 
-	kfree(*tab_attr);
+	return tab_attr;
 
-	return ret;
+err:
+	while (--i >= 0)
+		kfree(tab_attr[i]);
+	kfree(tab_attr);
+	return NULL;
 }
 
 static int add_port(struct ib_device *device, int port_num)
@@ -541,23 +526,20 @@ static int add_port(struct ib_device *device, int port_num)
 	if (ret)
 		goto err_put;
 
-	ret = alloc_group(&p->gid_attr, show_port_gid, attr.gid_tbl_len);
-	if (ret)
-		goto err_remove_pma;
-
 	p->gid_group.name  = "gids";
-	p->gid_group.attrs = p->gid_attr;
+	p->gid_group.attrs = alloc_group_attrs(show_port_gid, attr.gid_tbl_len);
+	if (!p->gid_group.attrs)
+		goto err_remove_pma;
 
 	ret = sysfs_create_group(&p->kobj, &p->gid_group);
 	if (ret)
 		goto err_free_gid;
 
-	ret = alloc_group(&p->pkey_attr, show_port_pkey, attr.pkey_tbl_len);
-	if (ret)
-		goto err_remove_gid;
-
 	p->pkey_group.name  = "pkeys";
-	p->pkey_group.attrs = p->pkey_attr;
+	p->pkey_group.attrs = alloc_group_attrs(show_port_pkey,
+						attr.pkey_tbl_len);
+	if (!p->pkey_group.attrs)
+		goto err_remove_gid;
 
 	ret = sysfs_create_group(&p->kobj, &p->pkey_group);
 	if (ret)
@@ -568,23 +550,19 @@ static int add_port(struct ib_device *device, int port_num)
 	return 0;
 
 err_free_pkey:
-	for (i = 0; i < attr.pkey_tbl_len; ++i) {
-		kfree(p->pkey_attr[i]->name);
-		kfree(p->pkey_attr[i]);
-	}
+	for (i = 0; i < attr.pkey_tbl_len; ++i)
+		kfree(p->pkey_group.attrs[i]);
 
-	kfree(p->pkey_attr);
+	kfree(p->pkey_group.attrs);
 
 err_remove_gid:
 	sysfs_remove_group(&p->kobj, &p->gid_group);
 
 err_free_gid:
-	for (i = 0; i < attr.gid_tbl_len; ++i) {
-		kfree(p->gid_attr[i]->name);
-		kfree(p->gid_attr[i]);
-	}
+	for (i = 0; i < attr.gid_tbl_len; ++i)
+		kfree(p->gid_group.attrs[i]);
 
-	kfree(p->gid_attr);
+	kfree(p->gid_group.attrs);
 
 err_remove_pma:
 	sysfs_remove_group(&p->kobj, &pma_group);
diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index 6ca0061137a..e8aad151175 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -339,16 +339,17 @@ pci_create_resource_files(struct pci_dev *pdev)
 		if (!pci_resource_len(pdev, i))
 			continue;
 
-		res_attr = kmalloc(sizeof(*res_attr) + 10, GFP_ATOMIC);
+		/* allocate attribute structure, piggyback attribute name */
+		res_attr = kcalloc(1, sizeof(*res_attr) + 10, GFP_ATOMIC);
 		if (res_attr) {
-			memset(res_attr, 0, sizeof(*res_attr) + 10);
+			char *res_attr_name = (char *)(res_attr + 1);
+
 			pdev->res_attr[i] = res_attr;
-			/* Allocated above after the res_attr struct */
-			res_attr->attr.name = (char *)(res_attr + 1);
-			sprintf(res_attr->attr.name, "resource%d", i);
-			res_attr->size = pci_resource_len(pdev, i);
+			sprintf(res_attr_name, "resource%d", i);
+			res_attr->attr.name = res_attr_name;
 			res_attr->attr.mode = S_IRUSR | S_IWUSR;
 			res_attr->attr.owner = THIS_MODULE;
+			res_attr->size = pci_resource_len(pdev, i);
 			res_attr->mmap = pci_mmap_resource;
 			res_attr->private = &pdev->resource[i];
 			sysfs_create_bin_file(&pdev->dev.kobj, res_attr);
diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
index 931b5aaaf38..d9cd2d31d37 100644
--- a/include/linux/sysfs.h
+++ b/include/linux/sysfs.h
@@ -16,13 +16,13 @@ struct kobject;
 struct module;
 
 struct attribute {
-	char			* name;
+	const char		* name;
 	struct module 		* owner;
 	mode_t			mode;
 };
 
 struct attribute_group {
-	char			* name;
+	const char		* name;
 	struct attribute	** attrs;
 };
 
-- 
cgit v1.2.3-70-g09d2


From e9ba6365fd4f0d9e7d022c883bd044fbaa48257f Mon Sep 17 00:00:00 2001
From: "gregkh@suse.de" <gregkh@suse.de>
Date: Tue, 15 Mar 2005 11:54:21 -0800
Subject: [PATCH] CLASS: move a "simple" class logic into the class core.

One step on improving the class api so that it can not be used incorrectly.
This also fixes the module owner issue with the dev files that happened when
the devt logic moved to the class core.

Based on a patch originally written by Kay Sievers <kay.sievers@vrfy.org>

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/class.c   | 145 +++++++++++++++++++++++++++++++++++++++++++++----
 include/linux/device.h |   9 +++
 2 files changed, 143 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/drivers/base/class.c b/drivers/base/class.c
index 344b8cd7390..a3b006b6f2b 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -16,6 +16,7 @@
 #include <linux/init.h>
 #include <linux/string.h>
 #include <linux/kdev_t.h>
+#include <linux/err.h>
 #include "base.h"
 
 #define to_class_attr(_attr) container_of(_attr, struct class_attribute, attr)
@@ -162,6 +163,51 @@ void class_unregister(struct class * cls)
 	subsystem_unregister(&cls->subsys);
 }
 
+static void class_create_release(struct class *cls)
+{
+	kfree(cls);
+}
+
+static void class_device_create_release(struct class_device *class_dev)
+{
+	kfree(class_dev);
+}
+
+struct class *class_create(struct module *owner, char *name)
+{
+	struct class *cls;
+	int retval;
+
+	cls = kmalloc(sizeof(struct class), GFP_KERNEL);
+	if (!cls) {
+		retval = -ENOMEM;
+		goto error;
+	}
+	memset(cls, 0x00, sizeof(struct class));
+
+	cls->name = name;
+	cls->owner = owner;
+	cls->class_release = class_create_release;
+	cls->release = class_device_create_release;
+
+	retval = class_register(cls);
+	if (retval)
+		goto error;
+
+	return cls;
+
+error:
+	kfree(cls);
+	return ERR_PTR(retval);
+}
+
+void class_destroy(struct class *cls)
+{
+	if ((cls == NULL) || (IS_ERR(cls)))
+		return;
+
+	class_unregister(cls);
+}
 
 /* Class Device Stuff */
 
@@ -375,7 +421,6 @@ static ssize_t show_dev(struct class_device *class_dev, char *buf)
 {
 	return print_dev_t(buf, class_dev->devt);
 }
-static CLASS_DEVICE_ATTR(dev, S_IRUGO, show_dev, NULL);
 
 void class_device_initialize(struct class_device *class_dev)
 {
@@ -412,7 +457,31 @@ int class_device_add(struct class_device *class_dev)
 	if ((error = kobject_add(&class_dev->kobj)))
 		goto register_done;
 
-	/* now take care of our own registration */
+	/* add the needed attributes to this device */
+	if (MAJOR(class_dev->devt)) {
+		struct class_device_attribute *attr;
+		attr = kmalloc(sizeof(*attr), GFP_KERNEL);
+		if (!attr) {
+			error = -ENOMEM;
+			kobject_del(&class_dev->kobj);
+			goto register_done;
+		}
+		memset(attr, sizeof(*attr), 0x00);
+		attr->attr.name = "dev";
+		attr->attr.mode = S_IRUGO;
+		attr->attr.owner = parent->owner;
+		attr->show = show_dev;
+		attr->store = NULL;
+		class_device_create_file(class_dev, attr);
+		class_dev->devt_attr = attr;
+	}
+
+	class_device_add_attrs(class_dev);
+	if (class_dev->dev)
+		sysfs_create_link(&class_dev->kobj,
+				  &class_dev->dev->kobj, "device");
+
+	/* notify any interfaces this device is now here */
 	if (parent) {
 		down(&parent->sem);
 		list_add_tail(&class_dev->node, &parent->children);
@@ -421,16 +490,8 @@ int class_device_add(struct class_device *class_dev)
 				class_intf->add(class_dev);
 		up(&parent->sem);
 	}
-
-	if (MAJOR(class_dev->devt))
-		class_device_create_file(class_dev, &class_device_attr_dev);
-
-	class_device_add_attrs(class_dev);
-	if (class_dev->dev)
-		sysfs_create_link(&class_dev->kobj,
-				  &class_dev->dev->kobj, "device");
-
 	kobject_hotplug(&class_dev->kobj, KOBJ_ADD);
+
  register_done:
 	if (error && parent)
 		class_put(parent);
@@ -444,6 +505,41 @@ int class_device_register(struct class_device *class_dev)
 	return class_device_add(class_dev);
 }
 
+struct class_device *class_device_create(struct class *cls, dev_t devt,
+					 struct device *device, char *fmt, ...)
+{
+	va_list args;
+	struct class_device *class_dev = NULL;
+	int retval = -ENODEV;
+
+	if (cls == NULL || IS_ERR(cls))
+		goto error;
+
+	class_dev = kmalloc(sizeof(struct class_device), GFP_KERNEL);
+	if (!class_dev) {
+		retval = -ENOMEM;
+		goto error;
+	}
+	memset(class_dev, 0x00, sizeof(struct class_device));
+
+	class_dev->devt = devt;
+	class_dev->dev = device;
+	class_dev->class = cls;
+
+	va_start(args, fmt);
+	vsnprintf(class_dev->class_id, BUS_ID_SIZE, fmt, args);
+	va_end(args);
+	retval = class_device_register(class_dev);
+	if (retval)
+		goto error;
+
+	return class_dev;
+
+error:
+	kfree(class_dev);
+	return ERR_PTR(retval);
+}
+
 void class_device_del(struct class_device *class_dev)
 {
 	struct class * parent = class_dev->class;
@@ -460,6 +556,11 @@ void class_device_del(struct class_device *class_dev)
 
 	if (class_dev->dev)
 		sysfs_remove_link(&class_dev->kobj, "device");
+	if (class_dev->devt_attr) {
+		class_device_remove_file(class_dev, class_dev->devt_attr);
+		kfree(class_dev->devt_attr);
+		class_dev->devt_attr = NULL;
+	}
 	class_device_remove_attrs(class_dev);
 
 	kobject_hotplug(&class_dev->kobj, KOBJ_REMOVE);
@@ -477,6 +578,24 @@ void class_device_unregister(struct class_device *class_dev)
 	class_device_put(class_dev);
 }
 
+void class_device_destroy(struct class *cls, dev_t devt)
+{
+	struct class_device *class_dev = NULL;
+	struct class_device *class_dev_tmp;
+
+	down(&cls->sem);
+	list_for_each_entry(class_dev_tmp, &cls->children, node) {
+		if (class_dev_tmp->devt == devt) {
+			class_dev = class_dev_tmp;
+			break;
+		}
+	}
+	up(&cls->sem);
+
+	if (class_dev)
+		class_device_unregister(class_dev);
+}
+
 int class_device_rename(struct class_device *class_dev, char *new_name)
 {
 	int error = 0;
@@ -576,6 +695,8 @@ EXPORT_SYMBOL_GPL(class_register);
 EXPORT_SYMBOL_GPL(class_unregister);
 EXPORT_SYMBOL_GPL(class_get);
 EXPORT_SYMBOL_GPL(class_put);
+EXPORT_SYMBOL_GPL(class_create);
+EXPORT_SYMBOL_GPL(class_destroy);
 
 EXPORT_SYMBOL_GPL(class_device_register);
 EXPORT_SYMBOL_GPL(class_device_unregister);
@@ -584,6 +705,8 @@ EXPORT_SYMBOL_GPL(class_device_add);
 EXPORT_SYMBOL_GPL(class_device_del);
 EXPORT_SYMBOL_GPL(class_device_get);
 EXPORT_SYMBOL_GPL(class_device_put);
+EXPORT_SYMBOL_GPL(class_device_create);
+EXPORT_SYMBOL_GPL(class_device_destroy);
 EXPORT_SYMBOL_GPL(class_device_create_file);
 EXPORT_SYMBOL_GPL(class_device_remove_file);
 EXPORT_SYMBOL_GPL(class_device_create_bin_file);
diff --git a/include/linux/device.h b/include/linux/device.h
index fa9e6ca08f5..73250d01c01 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -143,6 +143,7 @@ extern void driver_remove_file(struct device_driver *, struct driver_attribute *
  */
 struct class {
 	const char		* name;
+	struct module		* owner;
 
 	struct subsystem	subsys;
 	struct list_head	children;
@@ -185,6 +186,7 @@ struct class_device {
 	struct kobject		kobj;
 	struct class		* class;	/* required */
 	dev_t			devt;		/* dev_t, creates the sysfs "dev" */
+	struct class_device_attribute *devt_attr;
 	struct device		* dev;		/* not necessary, but nice to have */
 	void			* class_data;	/* class-specific data */
 
@@ -245,6 +247,13 @@ struct class_interface {
 extern int class_interface_register(struct class_interface *);
 extern void class_interface_unregister(struct class_interface *);
 
+extern struct class *class_create(struct module *owner, char *name);
+extern void class_destroy(struct class *cls);
+extern struct class_device *class_device_create(struct class *cls, dev_t devt,
+						struct device *device, char *fmt, ...)
+					__attribute__((format(printf,4,5)));
+extern void class_device_destroy(struct class *cls, dev_t devt);
+
 /* interface for class simple stuff */
 extern struct class_simple *class_simple_create(struct module *owner, char *name);
 extern void class_simple_destroy(struct class_simple *cs);
-- 
cgit v1.2.3-70-g09d2


From 1235686f6e67cf30c460eb77d90a6cb4be57b92f Mon Sep 17 00:00:00 2001
From: "gregkh@suse.de" <gregkh@suse.de>
Date: Tue, 15 Mar 2005 14:26:30 -0800
Subject: [PATCH] INPUT: move to use the new class code, instead of
 class_simple

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/input/evdev.c    |  9 +++++----
 drivers/input/input.c    | 10 +++++-----
 drivers/input/joydev.c   |  8 ++++----
 drivers/input/mousedev.c | 16 +++++++++-------
 drivers/input/tsdev.c    |  9 +++++----
 include/linux/input.h    |  2 +-
 6 files changed, 29 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/drivers/input/evdev.c b/drivers/input/evdev.c
index 17552a29978..556264b4342 100644
--- a/drivers/input/evdev.c
+++ b/drivers/input/evdev.c
@@ -431,9 +431,9 @@ static struct input_handle *evdev_connect(struct input_handler *handler, struct
 
 	devfs_mk_cdev(MKDEV(INPUT_MAJOR, EVDEV_MINOR_BASE + minor),
 			S_IFCHR|S_IRUGO|S_IWUSR, "input/event%d", minor);
-	class_simple_device_add(input_class,
-				MKDEV(INPUT_MAJOR, EVDEV_MINOR_BASE + minor),
-				dev->dev, "event%d", minor);
+	class_device_create(input_class,
+			MKDEV(INPUT_MAJOR, EVDEV_MINOR_BASE + minor),
+			dev->dev, "event%d", minor);
 
 	return &evdev->handle;
 }
@@ -443,7 +443,8 @@ static void evdev_disconnect(struct input_handle *handle)
 	struct evdev *evdev = handle->private;
 	struct evdev_list *list;
 
-	class_simple_device_remove(MKDEV(INPUT_MAJOR, EVDEV_MINOR_BASE + evdev->minor));
+	class_device_destroy(input_class,
+			MKDEV(INPUT_MAJOR, EVDEV_MINOR_BASE + evdev->minor));
 	devfs_remove("input/event%d", evdev->minor);
 	evdev->exist = 0;
 
diff --git a/drivers/input/input.c b/drivers/input/input.c
index 3385dd03abf..83c77c990dd 100644
--- a/drivers/input/input.c
+++ b/drivers/input/input.c
@@ -702,13 +702,13 @@ static int __init input_proc_init(void)
 static inline int input_proc_init(void) { return 0; }
 #endif
 
-struct class_simple *input_class;
+struct class *input_class;
 
 static int __init input_init(void)
 {
 	int retval = -ENOMEM;
 
-	input_class = class_simple_create(THIS_MODULE, "input");
+	input_class = class_create(THIS_MODULE, "input");
 	if (IS_ERR(input_class))
 		return PTR_ERR(input_class);
 	input_proc_init();
@@ -718,7 +718,7 @@ static int __init input_init(void)
 		remove_proc_entry("devices", proc_bus_input_dir);
 		remove_proc_entry("handlers", proc_bus_input_dir);
 		remove_proc_entry("input", proc_bus);
-		class_simple_destroy(input_class);
+		class_destroy(input_class);
 		return retval;
 	}
 
@@ -728,7 +728,7 @@ static int __init input_init(void)
 		remove_proc_entry("handlers", proc_bus_input_dir);
 		remove_proc_entry("input", proc_bus);
 		unregister_chrdev(INPUT_MAJOR, "input");
-		class_simple_destroy(input_class);
+		class_destroy(input_class);
 	}
 	return retval;
 }
@@ -741,7 +741,7 @@ static void __exit input_exit(void)
 
 	devfs_remove("input");
 	unregister_chrdev(INPUT_MAJOR, "input");
-	class_simple_destroy(input_class);
+	class_destroy(input_class);
 }
 
 subsys_initcall(input_init);
diff --git a/drivers/input/joydev.c b/drivers/input/joydev.c
index 627d343dfba..39775fc380c 100644
--- a/drivers/input/joydev.c
+++ b/drivers/input/joydev.c
@@ -452,9 +452,9 @@ static struct input_handle *joydev_connect(struct input_handler *handler, struct
 
 	devfs_mk_cdev(MKDEV(INPUT_MAJOR, JOYDEV_MINOR_BASE + minor),
 			S_IFCHR|S_IRUGO|S_IWUSR, "input/js%d", minor);
-	class_simple_device_add(input_class,
-				MKDEV(INPUT_MAJOR, JOYDEV_MINOR_BASE + minor),
-				dev->dev, "js%d", minor);
+	class_device_create(input_class,
+			MKDEV(INPUT_MAJOR, JOYDEV_MINOR_BASE + minor),
+			dev->dev, "js%d", minor);
 
 	return &joydev->handle;
 }
@@ -464,7 +464,7 @@ static void joydev_disconnect(struct input_handle *handle)
 	struct joydev *joydev = handle->private;
 	struct joydev_list *list;
 
-	class_simple_device_remove(MKDEV(INPUT_MAJOR, JOYDEV_MINOR_BASE + joydev->minor));
+	class_device_destroy(input_class, MKDEV(INPUT_MAJOR, JOYDEV_MINOR_BASE + joydev->minor));
 	devfs_remove("input/js%d", joydev->minor);
 	joydev->exist = 0;
 
diff --git a/drivers/input/mousedev.c b/drivers/input/mousedev.c
index 96fb9870834..062848ac7e6 100644
--- a/drivers/input/mousedev.c
+++ b/drivers/input/mousedev.c
@@ -647,9 +647,9 @@ static struct input_handle *mousedev_connect(struct input_handler *handler, stru
 
 	devfs_mk_cdev(MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + minor),
 			S_IFCHR|S_IRUGO|S_IWUSR, "input/mouse%d", minor);
-	class_simple_device_add(input_class,
-				MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + minor),
-				dev->dev, "mouse%d", minor);
+	class_device_create(input_class,
+			MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + minor),
+			dev->dev, "mouse%d", minor);
 
 	return &mousedev->handle;
 }
@@ -659,7 +659,8 @@ static void mousedev_disconnect(struct input_handle *handle)
 	struct mousedev *mousedev = handle->private;
 	struct mousedev_list *list;
 
-	class_simple_device_remove(MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + mousedev->minor));
+	class_device_destroy(input_class,
+			MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + mousedev->minor));
 	devfs_remove("input/mouse%d", mousedev->minor);
 	mousedev->exist = 0;
 
@@ -735,8 +736,8 @@ static int __init mousedev_init(void)
 
 	devfs_mk_cdev(MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + MOUSEDEV_MIX),
 			S_IFCHR|S_IRUGO|S_IWUSR, "input/mice");
-	class_simple_device_add(input_class, MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + MOUSEDEV_MIX),
-				NULL, "mice");
+	class_device_create(input_class,
+			MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + MOUSEDEV_MIX), NULL, "mice");
 
 #ifdef CONFIG_INPUT_MOUSEDEV_PSAUX
 	if (!(psaux_registered = !misc_register(&psaux_mouse)))
@@ -755,7 +756,8 @@ static void __exit mousedev_exit(void)
 		misc_deregister(&psaux_mouse);
 #endif
 	devfs_remove("input/mice");
-	class_simple_device_remove(MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + MOUSEDEV_MIX));
+	class_device_destroy(input_class,
+			MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + MOUSEDEV_MIX));
 	input_unregister_handler(&mousedev_handler);
 }
 
diff --git a/drivers/input/tsdev.c b/drivers/input/tsdev.c
index d0afba85720..50c63a15515 100644
--- a/drivers/input/tsdev.c
+++ b/drivers/input/tsdev.c
@@ -414,9 +414,9 @@ static struct input_handle *tsdev_connect(struct input_handler *handler,
 			S_IFCHR|S_IRUGO|S_IWUSR, "input/ts%d", minor);
 	devfs_mk_cdev(MKDEV(INPUT_MAJOR, TSDEV_MINOR_BASE + minor + TSDEV_MINORS/2),
 			S_IFCHR|S_IRUGO|S_IWUSR, "input/tsraw%d", minor);
-	class_simple_device_add(input_class,
-				MKDEV(INPUT_MAJOR, TSDEV_MINOR_BASE + minor),
-				dev->dev, "ts%d", minor);
+	class_device_create(input_class,
+			MKDEV(INPUT_MAJOR, TSDEV_MINOR_BASE + minor),
+			dev->dev, "ts%d", minor);
 
 	return &tsdev->handle;
 }
@@ -426,7 +426,8 @@ static void tsdev_disconnect(struct input_handle *handle)
 	struct tsdev *tsdev = handle->private;
 	struct tsdev_list *list;
 
-	class_simple_device_remove(MKDEV(INPUT_MAJOR, TSDEV_MINOR_BASE + tsdev->minor));
+	class_device_destroy(input_class,
+			MKDEV(INPUT_MAJOR, TSDEV_MINOR_BASE + tsdev->minor));
 	devfs_remove("input/ts%d", tsdev->minor);
 	devfs_remove("input/tsraw%d", tsdev->minor);
 	tsdev->exist = 0;
diff --git a/include/linux/input.h b/include/linux/input.h
index 72731d7d189..9d9598ed760 100644
--- a/include/linux/input.h
+++ b/include/linux/input.h
@@ -1015,7 +1015,7 @@ static inline void input_set_abs_params(struct input_dev *dev, int axis, int min
 	dev->absbit[LONG(axis)] |= BIT(axis);
 }
 
-extern struct class_simple *input_class;
+extern struct class *input_class;
 
 #endif
 #endif
-- 
cgit v1.2.3-70-g09d2


From 8561b10f6e7ef0a085709ffc844f74130a067abe Mon Sep 17 00:00:00 2001
From: "gregkh@suse.de" <gregkh@suse.de>
Date: Tue, 15 Mar 2005 15:10:13 -0800
Subject: [PATCH] USB: move the usb hcd code to use the new class code.

This moves a kref into the main hcd structure, which detaches it from
the class device structure.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hcd.c      | 61 ++++++++++++++++++++-------------------------
 drivers/usb/host/ehci-dbg.c | 10 ++++----
 drivers/usb/host/ohci-dbg.c | 10 ++++----
 include/linux/usb.h         |  5 ++--
 4 files changed, 39 insertions(+), 47 deletions(-)

(limited to 'include')

diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index 266e9e06a9f..d041782e0c8 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -651,50 +651,45 @@ static int usb_rh_urb_dequeue (struct usb_hcd *hcd, struct urb *urb)
 /*-------------------------------------------------------------------------*/
 
 /* exported only within usbcore */
-struct usb_bus *usb_bus_get (struct usb_bus *bus)
+struct usb_bus *usb_bus_get(struct usb_bus *bus)
 {
-	struct class_device *tmp;
+	if (bus)
+		kref_get(&bus->kref);
+	return bus;
+}
 
-	if (!bus)
-		return NULL;
+static void usb_host_release(struct kref *kref)
+{
+	struct usb_bus *bus = container_of(kref, struct usb_bus, kref);
 
-	tmp = class_device_get(&bus->class_dev);
-	if (tmp)        
-		return to_usb_bus(tmp);
-	else
-		return NULL;
+	if (bus->release)
+		bus->release(bus);
 }
 
 /* exported only within usbcore */
-void usb_bus_put (struct usb_bus *bus)
+void usb_bus_put(struct usb_bus *bus)
 {
 	if (bus)
-		class_device_put(&bus->class_dev);
+		kref_put(&bus->kref, usb_host_release);
 }
 
 /*-------------------------------------------------------------------------*/
 
-static void usb_host_release(struct class_device *class_dev)
-{
-	struct usb_bus *bus = to_usb_bus(class_dev);
-
-	if (bus->release)
-		bus->release(bus);
-}
-
-static struct class usb_host_class = {
-	.name		= "usb_host",
-	.release	= &usb_host_release,
-};
+static struct class *usb_host_class;
 
 int usb_host_init(void)
 {
-	return class_register(&usb_host_class);
+	int retval = 0;
+
+	usb_host_class = class_create(THIS_MODULE, "usb_host");
+	if (IS_ERR(usb_host_class))
+		retval = PTR_ERR(usb_host_class);
+	return retval;
 }
 
 void usb_host_cleanup(void)
 {
-	class_unregister(&usb_host_class);
+	class_destroy(usb_host_class);
 }
 
 /**
@@ -719,8 +714,7 @@ static void usb_bus_init (struct usb_bus *bus)
 
 	INIT_LIST_HEAD (&bus->bus_list);
 
-	class_device_initialize(&bus->class_dev);
-	bus->class_dev.class = &usb_host_class;
+	kref_init(&bus->kref);
 }
 
 /**
@@ -761,7 +755,6 @@ struct usb_bus *usb_alloc_bus (struct usb_operations *op)
 static int usb_register_bus(struct usb_bus *bus)
 {
 	int busnum;
-	int retval;
 
 	down (&usb_bus_list_lock);
 	busnum = find_next_zero_bit (busmap.busmap, USB_MAXBUS, 1);
@@ -774,15 +767,15 @@ static int usb_register_bus(struct usb_bus *bus)
 		return -E2BIG;
 	}
 
-	snprintf(bus->class_dev.class_id, BUS_ID_SIZE, "usb%d", busnum);
-	bus->class_dev.dev = bus->controller;
-	retval = class_device_add(&bus->class_dev);
-	if (retval) {
+	bus->class_dev = class_device_create(usb_host_class, MKDEV(0,0), bus->controller, "usb%d", busnum);
+	if (IS_ERR(bus->class_dev)) {
 		clear_bit(busnum, busmap.busmap);
 		up(&usb_bus_list_lock);
-		return retval;
+		return PTR_ERR(bus->class_dev);
 	}
 
+	class_set_devdata(bus->class_dev, bus);
+
 	/* Add it to the local list of buses */
 	list_add (&bus->bus_list, &usb_bus_list);
 	up (&usb_bus_list_lock);
@@ -820,7 +813,7 @@ static void usb_deregister_bus (struct usb_bus *bus)
 
 	clear_bit (bus->busnum, busmap.busmap);
 
-	class_device_del(&bus->class_dev);
+	class_device_unregister(bus->class_dev);
 }
 
 /**
diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.c
index 9b347d76538..2ff11d53567 100644
--- a/drivers/usb/host/ehci-dbg.c
+++ b/drivers/usb/host/ehci-dbg.c
@@ -450,7 +450,7 @@ show_async (struct class_device *class_dev, char *buf)
 
 	*buf = 0;
 
-	bus = to_usb_bus(class_dev);
+	bus = class_get_devdata(class_dev);
 	hcd = bus->hcpriv;
 	ehci = hcd_to_ehci (hcd);
 	next = buf;
@@ -496,7 +496,7 @@ show_periodic (struct class_device *class_dev, char *buf)
 		return 0;
 	seen_count = 0;
 
-	bus = to_usb_bus(class_dev);
+	bus = class_get_devdata(class_dev);
 	hcd = bus->hcpriv;
 	ehci = hcd_to_ehci (hcd);
 	next = buf;
@@ -633,7 +633,7 @@ show_registers (struct class_device *class_dev, char *buf)
 	static char		fmt [] = "%*s\n";
 	static char		label [] = "";
 
-	bus = to_usb_bus(class_dev);
+	bus = class_get_devdata(class_dev);
 	hcd = bus->hcpriv;
 	ehci = hcd_to_ehci (hcd);
 	next = buf;
@@ -735,7 +735,7 @@ static CLASS_DEVICE_ATTR (registers, S_IRUGO, show_registers, NULL);
 
 static inline void create_debug_files (struct ehci_hcd *ehci)
 {
-	struct class_device *cldev = &ehci_to_hcd(ehci)->self.class_dev;
+	struct class_device *cldev = ehci_to_hcd(ehci)->self.class_dev;
 
 	class_device_create_file(cldev, &class_device_attr_async);
 	class_device_create_file(cldev, &class_device_attr_periodic);
@@ -744,7 +744,7 @@ static inline void create_debug_files (struct ehci_hcd *ehci)
 
 static inline void remove_debug_files (struct ehci_hcd *ehci)
 {
-	struct class_device *cldev = &ehci_to_hcd(ehci)->self.class_dev;
+	struct class_device *cldev = ehci_to_hcd(ehci)->self.class_dev;
 
 	class_device_remove_file(cldev, &class_device_attr_async);
 	class_device_remove_file(cldev, &class_device_attr_periodic);
diff --git a/drivers/usb/host/ohci-dbg.c b/drivers/usb/host/ohci-dbg.c
index 62f53a21380..c58408c95c3 100644
--- a/drivers/usb/host/ohci-dbg.c
+++ b/drivers/usb/host/ohci-dbg.c
@@ -481,7 +481,7 @@ show_async (struct class_device *class_dev, char *buf)
 	size_t			temp;
 	unsigned long		flags;
 
-	bus = to_usb_bus(class_dev);
+	bus = class_get_devdata(class_dev);
 	hcd = bus->hcpriv;
 	ohci = hcd_to_ohci(hcd);
 
@@ -514,7 +514,7 @@ show_periodic (struct class_device *class_dev, char *buf)
 		return 0;
 	seen_count = 0;
 
-	bus = to_usb_bus(class_dev);
+	bus = class_get_devdata(class_dev);
 	hcd = bus->hcpriv;
 	ohci = hcd_to_ohci(hcd);
 	next = buf;
@@ -611,7 +611,7 @@ show_registers (struct class_device *class_dev, char *buf)
 	char			*next;
 	u32			rdata;
 
-	bus = to_usb_bus(class_dev);
+	bus = class_get_devdata(class_dev);
 	hcd = bus->hcpriv;
 	ohci = hcd_to_ohci(hcd);
 	regs = ohci->regs;
@@ -684,7 +684,7 @@ static CLASS_DEVICE_ATTR (registers, S_IRUGO, show_registers, NULL);
 
 static inline void create_debug_files (struct ohci_hcd *ohci)
 {
-	struct class_device *cldev = &ohci_to_hcd(ohci)->self.class_dev;
+	struct class_device *cldev = ohci_to_hcd(ohci)->self.class_dev;
 
 	class_device_create_file(cldev, &class_device_attr_async);
 	class_device_create_file(cldev, &class_device_attr_periodic);
@@ -694,7 +694,7 @@ static inline void create_debug_files (struct ohci_hcd *ohci)
 
 static inline void remove_debug_files (struct ohci_hcd *ohci)
 {
-	struct class_device *cldev = &ohci_to_hcd(ohci)->self.class_dev;
+	struct class_device *cldev = ohci_to_hcd(ohci)->self.class_dev;
 
 	class_device_remove_file(cldev, &class_device_attr_async);
 	class_device_remove_file(cldev, &class_device_attr_periodic);
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 2d1ac505853..3d508bf0840 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -287,15 +287,14 @@ struct usb_bus {
 
 	struct dentry *usbfs_dentry;	/* usbfs dentry entry for the bus */
 
-	struct class_device class_dev;	/* class device for this bus */
+	struct class_device *class_dev;	/* class device for this bus */
+	struct kref kref;		/* handles reference counting this bus */
 	void (*release)(struct usb_bus *bus);	/* function to destroy this bus's memory */
 #if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE)
 	struct mon_bus *mon_bus;	/* non-null when associated */
 	int monitored;			/* non-zero when monitored */
 #endif
 };
-#define	to_usb_bus(d) container_of(d, struct usb_bus, class_dev)
-
 
 /* -------------------------------------------------------------------------- */
 
-- 
cgit v1.2.3-70-g09d2


From cd987d38cc59053e0bab8150ffaca33b109067f3 Mon Sep 17 00:00:00 2001
From: "gregkh@suse.de" <gregkh@suse.de>
Date: Wed, 23 Mar 2005 11:12:38 -0800
Subject: [PATCH] class: remove class_simple code, as no one in the tree is
 using it anymore.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/Makefile       |   2 +-
 drivers/base/class_simple.c | 199 --------------------------------------------
 include/linux/device.h      |  10 ---
 3 files changed, 1 insertion(+), 210 deletions(-)
 delete mode 100644 drivers/base/class_simple.c

(limited to 'include')

diff --git a/drivers/base/Makefile b/drivers/base/Makefile
index a47928a2e57..771f6c80e82 100644
--- a/drivers/base/Makefile
+++ b/drivers/base/Makefile
@@ -1,7 +1,7 @@
 # Makefile for the Linux device tree
 
 obj-y			:= core.o sys.o bus.o \
-			   driver.o class.o class_simple.o platform.o \
+			   driver.o class.o platform.o \
 			   cpu.o firmware.o init.o map.o dmapool.o \
 			   attribute_container.o transport_class.o
 obj-y			+= power/
diff --git a/drivers/base/class_simple.c b/drivers/base/class_simple.c
deleted file mode 100644
index 27699eb20a3..00000000000
--- a/drivers/base/class_simple.c
+++ /dev/null
@@ -1,199 +0,0 @@
-/*
- * class_simple.c - a "simple" interface for classes for simple char devices.
- *
- * Copyright (c) 2003-2004 Greg Kroah-Hartman <greg@kroah.com>
- * Copyright (c) 2003-2004 IBM Corp.
- *
- * This file is released under the GPLv2
- *
- */
-
-#include <linux/config.h>
-#include <linux/device.h>
-#include <linux/err.h>
-
-struct class_simple {
-	struct class class;
-};
-#define to_class_simple(d) container_of(d, struct class_simple, class)
-
-struct simple_dev {
-	struct list_head node;
-	struct class_device class_dev;
-};
-#define to_simple_dev(d) container_of(d, struct simple_dev, class_dev)
-
-static LIST_HEAD(simple_dev_list);
-static DEFINE_SPINLOCK(simple_dev_list_lock);
-
-static void release_simple_dev(struct class_device *class_dev)
-{
-	struct simple_dev *s_dev = to_simple_dev(class_dev);
-	kfree(s_dev);
-}
-
-static void class_simple_release(struct class *class)
-{
-	struct class_simple *cs = to_class_simple(class);
-	kfree(cs);
-}
-
-/**
- * class_simple_create - create a struct class_simple structure
- * @owner: pointer to the module that is to "own" this struct class_simple
- * @name: pointer to a string for the name of this class.
- *
- * This is used to create a struct class_simple pointer that can then be used
- * in calls to class_simple_device_add().  This is used when you do not wish to
- * create a full blown class support for a type of char devices.
- *
- * Note, the pointer created here is to be destroyed when finished by making a
- * call to class_simple_destroy().
- */
-struct class_simple *class_simple_create(struct module *owner, char *name)
-{
-	struct class_simple *cs;
-	int retval;
-
-	cs = kmalloc(sizeof(*cs), GFP_KERNEL);
-	if (!cs) {
-		retval = -ENOMEM;
-		goto error;
-	}
-	memset(cs, 0x00, sizeof(*cs));
-
-	cs->class.name = name;
-	cs->class.class_release = class_simple_release;
-	cs->class.release = release_simple_dev;
-
-	retval = class_register(&cs->class);
-	if (retval)
-		goto error;
-
-	return cs;
-
-error:
-	kfree(cs);
-	return ERR_PTR(retval);
-}
-EXPORT_SYMBOL(class_simple_create);
-
-/**
- * class_simple_destroy - destroys a struct class_simple structure
- * @cs: pointer to the struct class_simple that is to be destroyed
- *
- * Note, the pointer to be destroyed must have been created with a call to
- * class_simple_create().
- */
-void class_simple_destroy(struct class_simple *cs)
-{
-	if ((cs == NULL) || (IS_ERR(cs)))
-		return;
-
-	class_unregister(&cs->class);
-}
-EXPORT_SYMBOL(class_simple_destroy);
-
-/**
- * class_simple_device_add - adds a class device to sysfs for a character driver
- * @cs: pointer to the struct class_simple that this device should be registered to.
- * @dev: the dev_t for the device to be added.
- * @device: a pointer to a struct device that is assiociated with this class device.
- * @fmt: string for the class device's name
- *
- * This function can be used by simple char device classes that do not
- * implement their own class device registration.  A struct class_device will
- * be created in sysfs, registered to the specified class.  A "dev" file will
- * be created, showing the dev_t for the device.  The pointer to the struct
- * class_device will be returned from the call.  Any further sysfs files that
- * might be required can be created using this pointer.
- * Note: the struct class_simple passed to this function must have previously been
- * created with a call to class_simple_create().
- */
-struct class_device *class_simple_device_add(struct class_simple *cs, dev_t dev, struct device *device, const char *fmt, ...)
-{
-	va_list args;
-	struct simple_dev *s_dev = NULL;
-	int retval;
-
-	if ((cs == NULL) || (IS_ERR(cs))) {
-		retval = -ENODEV;
-		goto error;
-	}
-
-	s_dev = kmalloc(sizeof(*s_dev), GFP_KERNEL);
-	if (!s_dev) {
-		retval = -ENOMEM;
-		goto error;
-	}
-	memset(s_dev, 0x00, sizeof(*s_dev));
-
-	s_dev->class_dev.devt = dev;
-	s_dev->class_dev.dev = device;
-	s_dev->class_dev.class = &cs->class;
-
-	va_start(args, fmt);
-	vsnprintf(s_dev->class_dev.class_id, BUS_ID_SIZE, fmt, args);
-	va_end(args);
-	retval = class_device_register(&s_dev->class_dev);
-	if (retval)
-		goto error;
-
-	spin_lock(&simple_dev_list_lock);
-	list_add(&s_dev->node, &simple_dev_list);
-	spin_unlock(&simple_dev_list_lock);
-
-	return &s_dev->class_dev;
-
-error:
-	kfree(s_dev);
-	return ERR_PTR(retval);
-}
-EXPORT_SYMBOL(class_simple_device_add);
-
-/**
- * class_simple_set_hotplug - set the hotplug callback in the embedded struct class
- * @cs: pointer to the struct class_simple to hold the pointer
- * @hotplug: function pointer to the hotplug function
- *
- * Implement and set a hotplug function to add environment variables specific to this
- * class on the hotplug event.
- */
-int class_simple_set_hotplug(struct class_simple *cs,
-	int (*hotplug)(struct class_device *dev, char **envp, int num_envp, char *buffer, int buffer_size))
-{
-	if ((cs == NULL) || (IS_ERR(cs)))
-		return -ENODEV;
-	cs->class.hotplug = hotplug;
-	return 0;
-}
-EXPORT_SYMBOL(class_simple_set_hotplug);
-
-/**
- * class_simple_device_remove - removes a class device that was created with class_simple_device_add()
- * @dev: the dev_t of the device that was previously registered.
- *
- * This call unregisters and cleans up a class device that was created with a
- * call to class_device_simple_add()
- */
-void class_simple_device_remove(dev_t dev)
-{
-	struct simple_dev *s_dev = NULL;
-	int found = 0;
-
-	spin_lock(&simple_dev_list_lock);
-	list_for_each_entry(s_dev, &simple_dev_list, node) {
-		if (s_dev->class_dev.devt == dev) {
-			found = 1;
-			break;
-		}
-	}
-	if (found) {
-		list_del(&s_dev->node);
-		spin_unlock(&simple_dev_list_lock);
-		class_device_unregister(&s_dev->class_dev);
-	} else {
-		spin_unlock(&simple_dev_list_lock);
-	}
-}
-EXPORT_SYMBOL(class_simple_device_remove);
diff --git a/include/linux/device.h b/include/linux/device.h
index 73250d01c01..68a95358013 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -44,7 +44,6 @@ struct device;
 struct device_driver;
 struct class;
 struct class_device;
-struct class_simple;
 
 struct bus_type {
 	const char		* name;
@@ -254,15 +253,6 @@ extern struct class_device *class_device_create(struct class *cls, dev_t devt,
 					__attribute__((format(printf,4,5)));
 extern void class_device_destroy(struct class *cls, dev_t devt);
 
-/* interface for class simple stuff */
-extern struct class_simple *class_simple_create(struct module *owner, char *name);
-extern void class_simple_destroy(struct class_simple *cs);
-extern struct class_device *class_simple_device_add(struct class_simple *cs, dev_t dev, struct device *device, const char *fmt, ...)
-	__attribute__((format(printf,4,5)));
-extern int class_simple_set_hotplug(struct class_simple *, 
-	int (*hotplug)(struct class_device *dev, char **envp, int num_envp, char *buffer, int buffer_size));
-extern void class_simple_device_remove(dev_t dev);
-
 
 struct device {
 	struct list_head node;		/* node in sibling list */
-- 
cgit v1.2.3-70-g09d2


From af70316af182f4716cc5eec7e0d27fc731d164bd Mon Sep 17 00:00:00 2001
From: "mochel@digitalimplant.org" <mochel@digitalimplant.org>
Date: Mon, 21 Mar 2005 10:41:04 -0800
Subject: [PATCH] Add a semaphore to struct device to synchronize calls to its
 driver.

This adds a per-device semaphore that is taken before every call from the core to a
driver method. This prevents e.g. simultaneous calls to the ->suspend() or ->resume()
and ->probe() or ->release(), potentially saving a whole lot of headaches.

It also moves us a step closer to removing the bus rwsem, since it protects the fields
in struct device that are modified by the core.

Signed-off-by: Patrick Mochel <mochel@digitalimplant.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/bus.c           | 14 +++++++++++---
 drivers/base/core.c          |  1 +
 drivers/base/power/resume.c  |  8 ++++++--
 drivers/base/power/suspend.c |  3 ++-
 include/linux/device.h       |  4 ++++
 5 files changed, 24 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 80ce88de56f..aa27f76d28c 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -283,18 +283,22 @@ void device_bind_driver(struct device * dev)
  */
 int driver_probe_device(struct device_driver * drv, struct device * dev)
 {
+	int error = 0;
+
 	if (drv->bus->match && !drv->bus->match(dev, drv))
 		return -ENODEV;
 
+	down(&dev->sem);
 	dev->driver = drv;
 	if (drv->probe) {
-		int error = drv->probe(dev);
+		error = drv->probe(dev);
 		if (error) {
 			dev->driver = NULL;
+			up(&dev->sem);
 			return error;
 		}
 	}
-
+	up(&dev->sem);
 	device_bind_driver(dev);
 	return 0;
 }
@@ -385,7 +389,10 @@ void driver_attach(struct device_driver * drv)
 
 void device_release_driver(struct device * dev)
 {
-	struct device_driver * drv = dev->driver;
+	struct device_driver * drv;
+
+	down(&dev->sem);
+	drv = dev->driver;
 	if (drv) {
 		sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj));
 		sysfs_remove_link(&dev->kobj, "driver");
@@ -394,6 +401,7 @@ void device_release_driver(struct device * dev)
 			drv->remove(dev);
 		dev->driver = NULL;
 	}
+	up(&dev->sem);
 }
 
 
diff --git a/drivers/base/core.c b/drivers/base/core.c
index a293a788abd..93440824b80 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -212,6 +212,7 @@ void device_initialize(struct device *dev)
 	INIT_LIST_HEAD(&dev->driver_list);
 	INIT_LIST_HEAD(&dev->bus_list);
 	INIT_LIST_HEAD(&dev->dma_pools);
+	init_MUTEX(&dev->sem);
 }
 
 /**
diff --git a/drivers/base/power/resume.c b/drivers/base/power/resume.c
index 26468971ef5..bdd96b03b88 100644
--- a/drivers/base/power/resume.c
+++ b/drivers/base/power/resume.c
@@ -22,6 +22,9 @@ extern int sysdev_resume(void);
 
 int resume_device(struct device * dev)
 {
+	int error = 0;
+
+	down(&dev->sem);
 	if (dev->power.pm_parent
 			&& dev->power.pm_parent->power.power_state) {
 		dev_err(dev, "PM: resume from %d, parent %s still %d\n",
@@ -31,9 +34,10 @@ int resume_device(struct device * dev)
 	}
 	if (dev->bus && dev->bus->resume) {
 		dev_dbg(dev,"resuming\n");
-		return dev->bus->resume(dev);
+		error = dev->bus->resume(dev);
 	}
-	return 0;
+	up(&dev->sem);
+	return error;
 }
 
 
diff --git a/drivers/base/power/suspend.c b/drivers/base/power/suspend.c
index 0ec44ef840b..807e13fb205 100644
--- a/drivers/base/power/suspend.c
+++ b/drivers/base/power/suspend.c
@@ -39,6 +39,7 @@ int suspend_device(struct device * dev, pm_message_t state)
 {
 	int error = 0;
 
+	down(&dev->sem);
 	if (dev->power.power_state) {
 		dev_dbg(dev, "PM: suspend %d-->%d\n",
 			dev->power.power_state, state);
@@ -58,7 +59,7 @@ int suspend_device(struct device * dev, pm_message_t state)
 		dev_dbg(dev, "suspending\n");
 		error = dev->bus->suspend(dev, state);
 	}
-
+	up(&dev->sem);
 	return error;
 }
 
diff --git a/include/linux/device.h b/include/linux/device.h
index 68a95358013..195b327f3e1 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -264,6 +264,10 @@ struct device {
 	struct kobject kobj;
 	char	bus_id[BUS_ID_SIZE];	/* position on parent bus */
 
+	struct semaphore	sem;	/* semaphore to synchronize calls to
+					 * its driver.
+					 */
+
 	struct bus_type	* bus;		/* type of bus device is on */
 	struct device_driver *driver;	/* which driver has allocated this
 					   device */
-- 
cgit v1.2.3-70-g09d2


From fae3cd00255e3e51ffd59fedb1bdb91ec96be395 Mon Sep 17 00:00:00 2001
From: "mochel@digitalimplant.org" <mochel@digitalimplant.org>
Date: Mon, 21 Mar 2005 10:59:56 -0800
Subject: [PATCH] Add driver_for_each_device().

Now there's an iterator for accessing each device bound to a driver.

Signed-off-by: Patrick Mochel <mochel@digitalimplant.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Index: linux-2.6.12-rc2/drivers/base/driver.c
===================================================================
---
 drivers/base/driver.c  | 35 +++++++++++++++++++++++++++++++++++
 include/linux/device.h |  3 +++
 2 files changed, 38 insertions(+)

(limited to 'include')

diff --git a/drivers/base/driver.c b/drivers/base/driver.c
index 3b269f7e521..484fed1985a 100644
--- a/drivers/base/driver.c
+++ b/drivers/base/driver.c
@@ -18,6 +18,41 @@
 #define to_dev(node) container_of(node, struct device, driver_list)
 #define to_drv(obj) container_of(obj, struct device_driver, kobj)
 
+
+/**
+ *	driver_for_each_device - Iterator for devices bound to a driver.
+ *	@drv:	Driver we're iterating.
+ *	@data:	Data to pass to the callback.
+ *	@fn:	Function to call for each device.
+ *
+ *	Take the bus's rwsem and iterate over the @drv's list of devices,
+ *	calling @fn for each one.
+ */
+
+int driver_for_each_device(struct device_driver * drv, struct device * start, 
+			   void * data, int (*fn)(struct device *, void *))
+{
+	struct list_head * head;
+	struct device * dev;
+	int error = 0;
+
+	down_read(&drv->bus->subsys.rwsem);
+	head = &drv->devices;
+	dev = list_prepare_entry(start, head, driver_list);
+	list_for_each_entry_continue(dev, head, driver_list) {
+		get_device(dev);
+		error = fn(dev, data);
+		put_device(dev);
+		if (error)
+			break;
+	}
+	up_read(&drv->bus->subsys.rwsem);
+	return error;
+}
+
+EXPORT_SYMBOL(driver_for_each_device);
+
+
 /**
  *	driver_create_file - create sysfs file for driver.
  *	@drv:	driver.
diff --git a/include/linux/device.h b/include/linux/device.h
index 195b327f3e1..10f5aa20e04 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -136,6 +136,9 @@ struct driver_attribute driver_attr_##_name = __ATTR(_name,_mode,_show,_store)
 extern int driver_create_file(struct device_driver *, struct driver_attribute *);
 extern void driver_remove_file(struct device_driver *, struct driver_attribute *);
 
+extern int driver_for_each_device(struct device_driver * drv, struct device * start,
+				  void * data, int (*fn)(struct device *, void *));
+
 
 /*
  * device classes
-- 
cgit v1.2.3-70-g09d2


From 9a19fea43616066561e221359596ce532e631395 Mon Sep 17 00:00:00 2001
From: "mochel@digitalimplant.org" <mochel@digitalimplant.org>
Date: Mon, 21 Mar 2005 11:45:16 -0800
Subject: [PATCH] Add initial implementation of klist helpers.

This klist interface provides a couple of structures that wrap around
struct list_head to provide explicit list "head" (struct klist) and
list "node" (struct klist_node) objects. For struct klist, a spinlock
is included that protects access to the actual list itself. struct
klist_node provides a pointer to the klist that owns it and a kref
reference count that indicates the number of current users of that node
in the list.

The entire point is to provide an interface for iterating over a list
that is safe and allows for modification of the list during the
iteration (e.g. insertion and removal), including modification of the
current node on the list.

It works using a 3rd object type - struct klist_iter - that is declared
and initialized before an iteration. klist_next() is used to acquire the
next element in the list. It returns NULL if there are no more items.
This klist interface provides a couple of structures that wrap around
struct list_head to provide explicit list "head" (struct klist) and
list "node" (struct klist_node) objects. For struct klist, a spinlock
is included that protects access to the actual list itself. struct
klist_node provides a pointer to the klist that owns it and a kref
reference count that indicates the number of current users of that node
in the list.

The entire point is to provide an interface for iterating over a list
that is safe and allows for modification of the list during the
iteration (e.g. insertion and removal), including modification of the
current node on the list.

It works using a 3rd object type - struct klist_iter - that is declared
and initialized before an iteration. klist_next() is used to acquire the
next element in the list. It returns NULL if there are no more items.
Internally, that routine takes the klist's lock, decrements the reference
count of the previous klist_node and increments the count of the next
klist_node. It then drops the lock and returns.

There are primitives for adding and removing nodes to/from a klist.
When deleting, klist_del() will simply decrement the reference count.
Only when the count goes to 0 is the node removed from the list.
klist_remove() will try to delete the node from the list and block
until it is actually removed. This is useful for objects (like devices)
that have been removed from the system and must be freed (but must wait
until all accessors have finished).

Internally, that routine takes the klist's lock, decrements the reference
count of the previous klist_node and increments the count of the next
klist_node. It then drops the lock and returns.

There are primitives for adding and removing nodes to/from a klist.
When deleting, klist_del() will simply decrement the reference count.
Only when the count goes to 0 is the node removed from the list.
klist_remove() will try to delete the node from the list and block
until it is actually removed. This is useful for objects (like devices)
that have been removed from the system and must be freed (but must wait
until all accessors have finished).

Signed-off-by: Patrick Mochel <mochel@digitalimplant.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

diff -Nru a/include/linux/klist.h b/include/linux/klist.h
---
 include/linux/klist.h |  53 +++++++++++
 lib/Makefile          |   7 +-
 lib/klist.c           | 248 ++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 305 insertions(+), 3 deletions(-)
 create mode 100644 include/linux/klist.h
 create mode 100644 lib/klist.c

(limited to 'include')

diff --git a/include/linux/klist.h b/include/linux/klist.h
new file mode 100644
index 00000000000..fb52f9d9d61
--- /dev/null
+++ b/include/linux/klist.h
@@ -0,0 +1,53 @@
+/*
+ *	klist.h - Some generic list helpers, extending struct list_head a bit.
+ *
+ *	Implementations are found in lib/klist.c
+ *
+ *
+ *	Copyright (C) 2005 Patrick Mochel
+ *
+ *	This file is rleased under the GPL v2.
+ */
+
+#include <linux/spinlock.h>
+#include <linux/completion.h>
+#include <linux/kref.h>
+#include <linux/list.h>
+
+
+struct klist {
+	spinlock_t		k_lock;
+	struct list_head	k_list;
+};
+
+
+extern void klist_init(struct klist * k);
+
+
+struct klist_node {
+	struct klist		* n_klist;
+	struct list_head	n_node;
+	struct kref		n_ref;
+	struct completion	n_removed;
+};
+
+extern void klist_add_tail(struct klist * k, struct klist_node * n);
+extern void klist_add_head(struct klist * k, struct klist_node * n);
+
+extern void klist_del(struct klist_node * n);
+extern void klist_remove(struct klist_node * n);
+
+
+struct klist_iter {
+	struct klist		* i_klist;
+	struct list_head	* i_head;
+	struct klist_node	* i_cur;
+};
+
+
+extern void klist_iter_init(struct klist * k, struct klist_iter * i);
+extern void klist_iter_init_node(struct klist * k, struct klist_iter * i, 
+				 struct klist_node * n);
+extern void klist_iter_exit(struct klist_iter * i);
+extern struct klist_node * klist_next(struct klist_iter * i);
+
diff --git a/lib/Makefile b/lib/Makefile
index 7c70db79c0e..9eccea9429a 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -4,9 +4,10 @@
 
 lib-y := errno.o ctype.o string.o vsprintf.o cmdline.o \
 	 bust_spinlocks.o rbtree.o radix-tree.o dump_stack.o \
-	 kobject.o kref.o idr.o div64.o int_sqrt.o \
-	 bitmap.o extable.o kobject_uevent.o prio_tree.o sha1.o \
-	 halfmd4.o
+	 idr.o div64.o int_sqrt.o bitmap.o extable.o prio_tree.o \
+	 sha1.o halfmd4.o
+
+lib-y	+= kobject.o kref.o kobject_uevent.o klist.o
 
 obj-y += sort.o parser.o
 
diff --git a/lib/klist.c b/lib/klist.c
new file mode 100644
index 00000000000..6f760424648
--- /dev/null
+++ b/lib/klist.c
@@ -0,0 +1,248 @@
+/*
+ *	klist.c - Routines for manipulating klists.
+ *
+ *
+ *	This klist interface provides a couple of structures that wrap around 
+ *	struct list_head to provide explicit list "head" (struct klist) and 
+ *	list "node" (struct klist_node) objects. For struct klist, a spinlock
+ *	is included that protects access to the actual list itself. struct 
+ *	klist_node provides a pointer to the klist that owns it and a kref
+ *	reference count that indicates the number of current users of that node
+ *	in the list.
+ *
+ *	The entire point is to provide an interface for iterating over a list
+ *	that is safe and allows for modification of the list during the
+ *	iteration (e.g. insertion and removal), including modification of the
+ *	current node on the list.
+ *
+ *	It works using a 3rd object type - struct klist_iter - that is declared
+ *	and initialized before an iteration. klist_next() is used to acquire the
+ *	next element in the list. It returns NULL if there are no more items.
+ *	Internally, that routine takes the klist's lock, decrements the reference
+ *	count of the previous klist_node and increments the count of the next
+ *	klist_node. It then drops the lock and returns.
+ *
+ *	There are primitives for adding and removing nodes to/from a klist. 
+ *	When deleting, klist_del() will simply decrement the reference count. 
+ *	Only when the count goes to 0 is the node removed from the list. 
+ *	klist_remove() will try to delete the node from the list and block
+ *	until it is actually removed. This is useful for objects (like devices)
+ *	that have been removed from the system and must be freed (but must wait
+ *	until all accessors have finished).
+ *
+ *	Copyright (C) 2005 Patrick Mochel
+ *
+ *	This file is released under the GPL v2.
+ */
+
+#include <linux/klist.h>
+#include <linux/module.h>
+
+
+/**
+ *	klist_init - Initialize a klist structure. 
+ *	@k:	The klist we're initializing.
+ */
+
+void klist_init(struct klist * k)
+{
+	INIT_LIST_HEAD(&k->k_list);
+	spin_lock_init(&k->k_lock);
+}
+
+EXPORT_SYMBOL_GPL(klist_init);
+
+
+static void add_head(struct klist * k, struct klist_node * n)
+{
+	spin_lock(&k->k_lock);
+	list_add(&n->n_node, &k->k_list);
+	spin_unlock(&k->k_lock);
+}
+
+static void add_tail(struct klist * k, struct klist_node * n)
+{
+	spin_lock(&k->k_lock);
+	list_add_tail(&n->n_node, &k->k_list);
+	spin_unlock(&k->k_lock);
+}
+
+
+static void klist_node_init(struct klist * k, struct klist_node * n)
+{
+	INIT_LIST_HEAD(&n->n_node);
+	init_completion(&n->n_removed);
+	kref_init(&n->n_ref);
+	n->n_klist = k;
+}
+
+
+/**
+ *	klist_add_head - Initialize a klist_node and add it to front.
+ *	@k:	klist it's going on.
+ *	@n:	node we're adding.
+ */
+
+void klist_add_head(struct klist * k, struct klist_node * n)
+{
+	klist_node_init(k, n);
+	add_head(k, n);
+}
+
+EXPORT_SYMBOL_GPL(klist_add_head);
+
+
+/**
+ *	klist_add_tail - Initialize a klist_node and add it to back.
+ *	@k:	klist it's going on.
+ *	@n:	node we're adding.
+ */
+
+void klist_add_tail(struct klist * k, struct klist_node * n)
+{
+	klist_node_init(k, n);
+	add_tail(k, n);
+}
+
+EXPORT_SYMBOL_GPL(klist_add_tail);
+
+
+static void klist_release(struct kref * kref)
+{
+	struct klist_node * n = container_of(kref, struct klist_node, n_ref);
+	list_del(&n->n_node);
+	complete(&n->n_removed);
+}
+
+static int klist_dec_and_del(struct klist_node * n)
+{
+	return kref_put(&n->n_ref, klist_release);
+}
+
+
+/**
+ *	klist_del - Decrement the reference count of node and try to remove.
+ *	@n:	node we're deleting.
+ */
+
+void klist_del(struct klist_node * n)
+{
+	struct klist * k = n->n_klist;
+
+	spin_lock(&k->k_lock);
+	klist_dec_and_del(n);
+	spin_unlock(&k->k_lock);
+}
+
+EXPORT_SYMBOL_GPL(klist_del);
+
+
+/**
+ *	klist_remove - Decrement the refcount of node and wait for it to go away.
+ *	@n:	node we're removing.
+ */
+
+void klist_remove(struct klist_node * n)
+{
+	spin_lock(&n->n_klist->k_lock);
+	klist_dec_and_del(n);
+	spin_unlock(&n->n_klist->k_lock);
+	wait_for_completion(&n->n_removed);
+}
+
+EXPORT_SYMBOL_GPL(klist_remove);
+
+
+/**
+ *	klist_iter_init_node - Initialize a klist_iter structure.
+ *	@k:	klist we're iterating.
+ *	@i:	klist_iter we're filling.
+ *	@n:	node to start with.
+ *
+ *	Similar to klist_iter_init(), but starts the action off with @n, 
+ *	instead of with the list head.
+ */
+
+void klist_iter_init_node(struct klist * k, struct klist_iter * i, struct klist_node * n)
+{
+	i->i_klist = k;
+	i->i_head = &k->k_list;
+	i->i_cur = n;
+}
+
+EXPORT_SYMBOL_GPL(klist_iter_init_node);
+
+
+/**
+ *	klist_iter_init - Iniitalize a klist_iter structure.
+ *	@k:	klist we're iterating.
+ *	@i:	klist_iter structure we're filling.
+ *
+ *	Similar to klist_iter_init_node(), but start with the list head.
+ */
+
+void klist_iter_init(struct klist * k, struct klist_iter * i)
+{
+	klist_iter_init_node(k, i, NULL);
+}
+
+EXPORT_SYMBOL_GPL(klist_iter_init);
+
+
+/**
+ *	klist_iter_exit - Finish a list iteration.
+ *	@i:	Iterator structure.
+ *
+ *	Must be called when done iterating over list, as it decrements the 
+ *	refcount of the current node. Necessary in case iteration exited before
+ *	the end of the list was reached, and always good form.
+ */
+
+void klist_iter_exit(struct klist_iter * i)
+{
+	if (i->i_cur) {
+		klist_del(i->i_cur);
+		i->i_cur = NULL;
+	}
+}
+
+EXPORT_SYMBOL_GPL(klist_iter_exit);
+
+
+static struct klist_node * to_klist_node(struct list_head * n)
+{
+	return container_of(n, struct klist_node, n_node);
+}
+
+
+/**
+ *	klist_next - Ante up next node in list.
+ *	@i:	Iterator structure.
+ *
+ *	First grab list lock. Decrement the reference count of the previous
+ *	node, if there was one. Grab the next node, increment its reference 
+ *	count, drop the lock, and return that next node.
+ */
+
+struct klist_node * klist_next(struct klist_iter * i)
+{
+	struct list_head * next;
+	struct klist_node * knode = NULL;
+
+	spin_lock(&i->i_klist->k_lock);
+	if (i->i_cur) {
+		next = i->i_cur->n_node.next;
+		klist_dec_and_del(i->i_cur);
+	} else
+		next = i->i_head->next;
+
+	if (next != i->i_head) {
+		knode = to_klist_node(next);
+		kref_get(&knode->n_ref);
+	}
+	i->i_cur = knode;
+	spin_unlock(&i->i_klist->k_lock);
+	return knode;
+}
+
+EXPORT_SYMBOL_GPL(klist_next);
-- 
cgit v1.2.3-70-g09d2


From 465c7a3a3a5aabcedd2e43612cac5a12f23da19a Mon Sep 17 00:00:00 2001
From: "mochel@digitalimplant.org" <mochel@digitalimplant.org>
Date: Mon, 21 Mar 2005 11:49:14 -0800
Subject: [PATCH] Add a klist to struct bus_type for its devices.

- Use it for bus_for_each_dev().
- Use the klist spinlock instead of the bus rwsem.

Signed-off-by: Patrick Mochel <mochel@digitalimplant.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/bus.c     | 54 ++++++++++++++++++++++----------------------------
 include/linux/device.h |  3 +++
 2 files changed, 27 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 1b05c1399e3..c8da37230b3 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -134,28 +134,6 @@ static struct kobj_type ktype_bus = {
 
 decl_subsys(bus, &ktype_bus, NULL);
 
-static int __bus_for_each_dev(struct bus_type *bus, struct device *start,
-			      void *data, int (*fn)(struct device *, void *))
-{
-	struct list_head *head;
-	struct device *dev;
-	int error = 0;
-
-	if (!(bus = get_bus(bus)))
-		return -EINVAL;
-
-	head = &bus->devices.list;
-	dev = list_prepare_entry(start, head, bus_list);
-	list_for_each_entry_continue(dev, head, bus_list) {
-		get_device(dev);
-		error = fn(dev, data);
-		put_device(dev);
-		if (error)
-			break;
-	}
-	put_bus(bus);
-	return error;
-}
 
 static int __bus_for_each_drv(struct bus_type *bus, struct device_driver *start,
 			      void * data, int (*fn)(struct device_driver *, void *))
@@ -180,6 +158,13 @@ static int __bus_for_each_drv(struct bus_type *bus, struct device_driver *start,
 	return error;
 }
 
+
+static struct device * next_device(struct klist_iter * i)
+{
+	struct klist_node * n = klist_next(i);
+	return n ? container_of(n, struct device, knode_bus) : NULL;
+}
+
 /**
  *	bus_for_each_dev - device iterator.
  *	@bus:	bus type.
@@ -203,12 +188,19 @@ static int __bus_for_each_drv(struct bus_type *bus, struct device_driver *start,
 int bus_for_each_dev(struct bus_type * bus, struct device * start,
 		     void * data, int (*fn)(struct device *, void *))
 {
-	int ret;
+	struct klist_iter i;
+	struct device * dev;
+	int error = 0;
 
-	down_read(&bus->subsys.rwsem);
-	ret = __bus_for_each_dev(bus, start, data, fn);
-	up_read(&bus->subsys.rwsem);
-	return ret;
+	if (!bus)
+		return -EINVAL;
+
+	klist_iter_init_node(&bus->klist_devices, &i,
+			     (start ? &start->knode_bus : NULL));
+	while ((dev = next_device(&i)) && !error)
+		error = fn(dev, data);
+	klist_iter_exit(&i);
+	return error;
 }
 
 /**
@@ -293,6 +285,7 @@ int bus_add_device(struct device * dev)
 		list_add_tail(&dev->bus_list, &dev->bus->devices.list);
 		device_attach(dev);
 		up_write(&dev->bus->subsys.rwsem);
+		klist_add_tail(&bus->klist_devices, &dev->knode_bus);
 		device_add_attrs(bus, dev);
 		sysfs_create_link(&bus->devices.kobj, &dev->kobj, dev->bus_id);
 		sysfs_create_link(&dev->kobj, &dev->bus->subsys.kset.kobj, "bus");
@@ -315,6 +308,7 @@ void bus_remove_device(struct device * dev)
 		sysfs_remove_link(&dev->kobj, "bus");
 		sysfs_remove_link(&dev->bus->devices.kobj, dev->bus_id);
 		device_remove_attrs(dev->bus, dev);
+		klist_remove(&dev->knode_bus);
 		down_write(&dev->bus->subsys.rwsem);
 		pr_debug("bus %s: remove device %s\n", dev->bus->name, dev->bus_id);
 		device_release_driver(dev);
@@ -439,9 +433,7 @@ int bus_rescan_devices(struct bus_type * bus)
 {
 	int count = 0;
 
-	down_write(&bus->subsys.rwsem);
-	__bus_for_each_dev(bus, NULL, &count, bus_rescan_devices_helper);
-	up_write(&bus->subsys.rwsem);
+	bus_for_each_dev(bus, NULL, &count, bus_rescan_devices_helper);
 
 	return count;
 }
@@ -542,6 +534,8 @@ int bus_register(struct bus_type * bus)
 	retval = kset_register(&bus->drivers);
 	if (retval)
 		goto bus_drivers_fail;
+
+	klist_init(&bus->klist_devices);
 	bus_add_attrs(bus);
 
 	pr_debug("bus type '%s' registered\n", bus->name);
diff --git a/include/linux/device.h b/include/linux/device.h
index 10f5aa20e04..e36953cf7f1 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -14,6 +14,7 @@
 #include <linux/config.h>
 #include <linux/ioport.h>
 #include <linux/kobject.h>
+#include <linux/klist.h>
 #include <linux/list.h>
 #include <linux/types.h>
 #include <linux/module.h>
@@ -51,6 +52,7 @@ struct bus_type {
 	struct subsystem	subsys;
 	struct kset		drivers;
 	struct kset		devices;
+	struct klist		klist_devices;
 
 	struct bus_attribute	* bus_attrs;
 	struct device_attribute	* dev_attrs;
@@ -262,6 +264,7 @@ struct device {
 	struct list_head bus_list;	/* node in bus's list */
 	struct list_head driver_list;
 	struct list_head children;
+	struct klist_node	knode_bus;
 	struct device 	* parent;
 
 	struct kobject kobj;
-- 
cgit v1.2.3-70-g09d2


From 38fdac3cdce276554b4484a41f8ec2daf81cb2ff Mon Sep 17 00:00:00 2001
From: "mochel@digitalimplant.org" <mochel@digitalimplant.org>
Date: Mon, 21 Mar 2005 12:00:18 -0800
Subject: [PATCH] Add a klist to struct bus_type for its drivers.

- Use it in bus_for_each_drv().
- Use the klist spinlock instead of the bus rwsem.

Signed-off-by: Patrick Mochel <mochel@digitalimplant.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/bus.c     | 52 ++++++++++++++++++++++----------------------------
 include/linux/device.h |  2 ++
 2 files changed, 25 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index c8da37230b3..736dc1f5a31 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -135,30 +135,6 @@ static struct kobj_type ktype_bus = {
 decl_subsys(bus, &ktype_bus, NULL);
 
 
-static int __bus_for_each_drv(struct bus_type *bus, struct device_driver *start,
-			      void * data, int (*fn)(struct device_driver *, void *))
-{
-	struct list_head *head;
-	struct device_driver *drv;
-	int error = 0;
-
-	if (!(bus = get_bus(bus)))
-		return -EINVAL;
-
-	head = &bus->drivers.list;
-	drv = list_prepare_entry(start, head, kobj.entry);
-	list_for_each_entry_continue(drv, head, kobj.entry) {
-		get_driver(drv);
-		error = fn(drv, data);
-		put_driver(drv);
-		if (error)
-			break;
-	}
-	put_bus(bus);
-	return error;
-}
-
-
 static struct device * next_device(struct klist_iter * i)
 {
 	struct klist_node * n = klist_next(i);
@@ -203,6 +179,14 @@ int bus_for_each_dev(struct bus_type * bus, struct device * start,
 	return error;
 }
 
+
+
+static struct device_driver * next_driver(struct klist_iter * i)
+{
+	struct klist_node * n = klist_next(i);
+	return n ? container_of(n, struct device_driver, knode_bus) : NULL;
+}
+
 /**
  *	bus_for_each_drv - driver iterator
  *	@bus:	bus we're dealing with.
@@ -226,12 +210,19 @@ int bus_for_each_dev(struct bus_type * bus, struct device * start,
 int bus_for_each_drv(struct bus_type * bus, struct device_driver * start,
 		     void * data, int (*fn)(struct device_driver *, void *))
 {
-	int ret;
+	struct klist_iter i;
+	struct device_driver * drv;
+	int error = 0;
 
-	down_read(&bus->subsys.rwsem);
-	ret = __bus_for_each_drv(bus, start, data, fn);
-	up_read(&bus->subsys.rwsem);
-	return ret;
+	if (!bus)
+		return -EINVAL;
+
+	klist_iter_init_node(&bus->klist_drivers, &i,
+			     start ? &start->knode_bus : NULL);
+	while ((drv = next_driver(&i)) && !error)
+		error = fn(drv, data);
+	klist_iter_exit(&i);
+	return error;
 }
 
 static int device_add_attrs(struct bus_type * bus, struct device * dev)
@@ -376,6 +367,7 @@ int bus_add_driver(struct device_driver * drv)
 		down_write(&bus->subsys.rwsem);
 		driver_attach(drv);
 		up_write(&bus->subsys.rwsem);
+		klist_add_tail(&bus->klist_drivers, &drv->knode_bus);
 		module_add_driver(drv->owner, drv);
 
 		driver_add_attrs(bus, drv);
@@ -397,6 +389,7 @@ void bus_remove_driver(struct device_driver * drv)
 {
 	if (drv->bus) {
 		driver_remove_attrs(drv->bus, drv);
+		klist_remove(&drv->knode_bus);
 		down_write(&drv->bus->subsys.rwsem);
 		pr_debug("bus %s: remove driver %s\n", drv->bus->name, drv->name);
 		driver_detach(drv);
@@ -536,6 +529,7 @@ int bus_register(struct bus_type * bus)
 		goto bus_drivers_fail;
 
 	klist_init(&bus->klist_devices);
+	klist_init(&bus->klist_drivers);
 	bus_add_attrs(bus);
 
 	pr_debug("bus type '%s' registered\n", bus->name);
diff --git a/include/linux/device.h b/include/linux/device.h
index e36953cf7f1..ea9ab33dfe7 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -53,6 +53,7 @@ struct bus_type {
 	struct kset		drivers;
 	struct kset		devices;
 	struct klist		klist_devices;
+	struct klist		klist_drivers;
 
 	struct bus_attribute	* bus_attrs;
 	struct device_attribute	* dev_attrs;
@@ -105,6 +106,7 @@ struct device_driver {
 	struct completion	unloaded;
 	struct kobject		kobj;
 	struct list_head	devices;
+	struct klist_node	knode_bus;
 
 	struct module		* owner;
 
-- 
cgit v1.2.3-70-g09d2


From 94e7b1c5ff2055571703e38b059afffe17658432 Mon Sep 17 00:00:00 2001
From: "mochel@digitalimplant.org" <mochel@digitalimplant.org>
Date: Mon, 21 Mar 2005 12:25:36 -0800
Subject: [PATCH] Add a klist to struct device_driver for the devices bound to
 it.

- Use it in driver_for_each_device() instead of the regular list_head and stop using
  the bus's rwsem for protection.
- Use driver_for_each_device() in driver_detach() so we don't deadlock on the
  bus's rwsem.
- Remove ->devices.
- Move klist access and sysfs link access out from under device's semaphore, since
  they're synchronized through other means.

Signed-off-by: Patrick Mochel <mochel@digitalimplant.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    |  2 +-
 drivers/base/dd.c      | 34 +++++++++++++++++++---------------
 drivers/base/driver.c  | 27 +++++++++++++++------------
 include/linux/device.h |  3 ++-
 4 files changed, 37 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 93440824b80..bc5bec61a01 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -209,8 +209,8 @@ void device_initialize(struct device *dev)
 	kobject_init(&dev->kobj);
 	INIT_LIST_HEAD(&dev->node);
 	INIT_LIST_HEAD(&dev->children);
-	INIT_LIST_HEAD(&dev->driver_list);
 	INIT_LIST_HEAD(&dev->bus_list);
+	INIT_LIST_HEAD(&dev->driver_list);
 	INIT_LIST_HEAD(&dev->dma_pools);
 	init_MUTEX(&dev->sem);
 }
diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index b709b1e0cb2..47cbb564123 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -40,7 +40,7 @@ void device_bind_driver(struct device * dev)
 {
 	pr_debug("bound device '%s' to driver '%s'\n",
 		 dev->bus_id, dev->driver->name);
-	list_add_tail(&dev->driver_list, &dev->driver->devices);
+	klist_add_tail(&dev->driver->klist_devices, &dev->knode_driver);
 	sysfs_create_link(&dev->driver->kobj, &dev->kobj,
 			  kobject_name(&dev->kobj));
 	sysfs_create_link(&dev->kobj, &dev->driver->kobj, "driver");
@@ -164,31 +164,35 @@ void driver_attach(struct device_driver * drv)
  */
 void device_release_driver(struct device * dev)
 {
-	struct device_driver * drv;
+	struct device_driver * drv = dev->driver;
+
+	if (!drv)
+		return;
+
+	sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj));
+	sysfs_remove_link(&dev->kobj, "driver");
+	klist_remove(&dev->knode_driver);
 
 	down(&dev->sem);
-	drv = dev->driver;
-	if (drv) {
-		sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj));
-		sysfs_remove_link(&dev->kobj, "driver");
-		list_del_init(&dev->driver_list);
-		if (drv->remove)
-			drv->remove(dev);
-		dev->driver = NULL;
-	}
+	if (drv->remove)
+		drv->remove(dev);
+	dev->driver = NULL;
 	up(&dev->sem);
 }
 
+static int __remove_driver(struct device * dev, void * unused)
+{
+	device_release_driver(dev);
+	return 0;
+}
+
 /**
  * driver_detach - detach driver from all devices it controls.
  * @drv: driver.
  */
 void driver_detach(struct device_driver * drv)
 {
-	while (!list_empty(&drv->devices)) {
-		struct device * dev = container_of(drv->devices.next, struct device, driver_list);
-		device_release_driver(dev);
-	}
+	driver_for_each_device(drv, NULL, NULL, __remove_driver);
 }
 
 
diff --git a/drivers/base/driver.c b/drivers/base/driver.c
index 484fed1985a..34bd38aa7eb 100644
--- a/drivers/base/driver.c
+++ b/drivers/base/driver.c
@@ -19,6 +19,12 @@
 #define to_drv(obj) container_of(obj, struct device_driver, kobj)
 
 
+static struct device * next_device(struct klist_iter * i)
+{
+	struct klist_node * n = klist_next(i);
+	return n ? container_of(n, struct device, knode_driver) : NULL;
+}
+
 /**
  *	driver_for_each_device - Iterator for devices bound to a driver.
  *	@drv:	Driver we're iterating.
@@ -32,21 +38,18 @@
 int driver_for_each_device(struct device_driver * drv, struct device * start, 
 			   void * data, int (*fn)(struct device *, void *))
 {
-	struct list_head * head;
+	struct klist_iter i;
 	struct device * dev;
 	int error = 0;
 
-	down_read(&drv->bus->subsys.rwsem);
-	head = &drv->devices;
-	dev = list_prepare_entry(start, head, driver_list);
-	list_for_each_entry_continue(dev, head, driver_list) {
-		get_device(dev);
+	if (!drv)
+		return -EINVAL;
+
+	klist_iter_init_node(&drv->klist_devices, &i,
+			     start ? &start->knode_driver : NULL);
+	while ((dev = next_device(&i)) && !error)
 		error = fn(dev, data);
-		put_device(dev);
-		if (error)
-			break;
-	}
-	up_read(&drv->bus->subsys.rwsem);
+	klist_iter_exit(&i);
 	return error;
 }
 
@@ -120,7 +123,7 @@ void put_driver(struct device_driver * drv)
  */
 int driver_register(struct device_driver * drv)
 {
-	INIT_LIST_HEAD(&drv->devices);
+	klist_init(&drv->klist_devices);
 	init_completion(&drv->unloaded);
 	return bus_add_driver(drv);
 }
diff --git a/include/linux/device.h b/include/linux/device.h
index ea9ab33dfe7..96c71b59fde 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -105,7 +105,7 @@ struct device_driver {
 
 	struct completion	unloaded;
 	struct kobject		kobj;
-	struct list_head	devices;
+	struct klist		klist_devices;
 	struct klist_node	knode_bus;
 
 	struct module		* owner;
@@ -266,6 +266,7 @@ struct device {
 	struct list_head bus_list;	/* node in bus's list */
 	struct list_head driver_list;
 	struct list_head children;
+	struct klist_node	knode_driver;
 	struct klist_node	knode_bus;
 	struct device 	* parent;
 
-- 
cgit v1.2.3-70-g09d2


From cb85b6f1cc811ecb9ed4b950206d8941ba710e68 Mon Sep 17 00:00:00 2001
From: "mochel@digitalimplant.org" <mochel@digitalimplant.org>
Date: Thu, 24 Mar 2005 10:48:35 -0800
Subject: [PATCH] Remove the unused device_find().

Signed-off-by: Patrick Mochel <mochel@digitalimplant.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 19 -------------------
 include/linux/device.h |  1 -
 2 files changed, 20 deletions(-)

(limited to 'include')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index bc5bec61a01..0eb1d424bd1 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -401,24 +401,6 @@ int device_for_each_child(struct device * dev, void * data,
 	return error;
 }
 
-/**
- *	device_find - locate device on a bus by name.
- *	@name:	name of the device.
- *	@bus:	bus to scan for the device.
- *
- *	Call kset_find_obj() to iterate over list of devices on
- *	a bus to find device by name. Return device if found.
- *
- *	Note that kset_find_obj increments device's reference count.
- */
-struct device *device_find(const char *name, struct bus_type *bus)
-{
-	struct kobject *k = kset_find_obj(&bus->devices, name);
-	if (k)
-		return to_dev(k);
-	return NULL;
-}
-
 int __init devices_init(void)
 {
 	return subsystem_register(&devices_subsys);
@@ -434,7 +416,6 @@ EXPORT_SYMBOL_GPL(device_del);
 EXPORT_SYMBOL_GPL(device_unregister);
 EXPORT_SYMBOL_GPL(get_device);
 EXPORT_SYMBOL_GPL(put_device);
-EXPORT_SYMBOL_GPL(device_find);
 
 EXPORT_SYMBOL_GPL(device_create_file);
 EXPORT_SYMBOL_GPL(device_remove_file);
diff --git a/include/linux/device.h b/include/linux/device.h
index 96c71b59fde..f1c38d869ac 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -372,7 +372,6 @@ extern int (*platform_notify_remove)(struct device * dev);
  */
 extern struct device * get_device(struct device * dev);
 extern void put_device(struct device * dev);
-extern struct device *device_find(const char *name, struct bus_type *bus);
 
 
 /* drivers/base/platform.c */
-- 
cgit v1.2.3-70-g09d2


From 8b0c250be489dcbf1a3a33bb4ec4c7f33735a365 Mon Sep 17 00:00:00 2001
From: "mochel@digitalimplant.org" <mochel@digitalimplant.org>
Date: Thu, 24 Mar 2005 12:58:57 -0800
Subject: [PATCH] add klist_node_attached() to determine if a node is on a list
 or not.

Signed-off-by: Patrick Mochel <mochel@digitalimplant.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

diff -Nru a/include/linux/klist.h b/include/linux/klist.h
---
 include/linux/klist.h |  2 ++
 lib/klist.c           | 16 ++++++++++++++++
 2 files changed, 18 insertions(+)

(limited to 'include')

diff --git a/include/linux/klist.h b/include/linux/klist.h
index fb52f9d9d61..eebf5e5696e 100644
--- a/include/linux/klist.h
+++ b/include/linux/klist.h
@@ -37,6 +37,8 @@ extern void klist_add_head(struct klist * k, struct klist_node * n);
 extern void klist_del(struct klist_node * n);
 extern void klist_remove(struct klist_node * n);
 
+extern int klist_node_attached(struct klist_node * n);
+
 
 struct klist_iter {
 	struct klist		* i_klist;
diff --git a/lib/klist.c b/lib/klist.c
index 6f760424648..02177d72dc8 100644
--- a/lib/klist.c
+++ b/lib/klist.c
@@ -112,6 +112,7 @@ static void klist_release(struct kref * kref)
 	struct klist_node * n = container_of(kref, struct klist_node, n_ref);
 	list_del(&n->n_node);
 	complete(&n->n_removed);
+	n->n_klist = NULL;
 }
 
 static int klist_dec_and_del(struct klist_node * n)
@@ -153,6 +154,19 @@ void klist_remove(struct klist_node * n)
 EXPORT_SYMBOL_GPL(klist_remove);
 
 
+/**
+ *	klist_node_attached - Say whether a node is bound to a list or not.
+ *	@n:	Node that we're testing.
+ */
+
+int klist_node_attached(struct klist_node * n)
+{
+	return (n->n_klist != NULL);
+}
+
+EXPORT_SYMBOL_GPL(klist_node_attached);
+
+
 /**
  *	klist_iter_init_node - Initialize a klist_iter structure.
  *	@k:	klist we're iterating.
@@ -246,3 +260,5 @@ struct klist_node * klist_next(struct klist_iter * i)
 }
 
 EXPORT_SYMBOL_GPL(klist_next);
+
+
-- 
cgit v1.2.3-70-g09d2


From 7dc35cdf69149a7f2b5216ada9bafe359746ac1c Mon Sep 17 00:00:00 2001
From: "mochel@digitalimplant.org" <mochel@digitalimplant.org>
Date: Thu, 24 Mar 2005 13:03:35 -0800
Subject: [PATCH] Remove struct device::bus_list.

Signed-off-by: Patrick Mochel <mochel@digitalimplant.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 1 -
 include/linux/device.h | 1 -
 2 files changed, 2 deletions(-)

(limited to 'include')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 0eb1d424bd1..f258a21ed78 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -209,7 +209,6 @@ void device_initialize(struct device *dev)
 	kobject_init(&dev->kobj);
 	INIT_LIST_HEAD(&dev->node);
 	INIT_LIST_HEAD(&dev->children);
-	INIT_LIST_HEAD(&dev->bus_list);
 	INIT_LIST_HEAD(&dev->driver_list);
 	INIT_LIST_HEAD(&dev->dma_pools);
 	init_MUTEX(&dev->sem);
diff --git a/include/linux/device.h b/include/linux/device.h
index f1c38d869ac..13f65853e58 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -263,7 +263,6 @@ extern void class_device_destroy(struct class *cls, dev_t devt);
 
 struct device {
 	struct list_head node;		/* node in sibling list */
-	struct list_head bus_list;	/* node in bus's list */
 	struct list_head driver_list;
 	struct list_head children;
 	struct klist_node	knode_driver;
-- 
cgit v1.2.3-70-g09d2


From 63c4f204ffc8219696bda88eb20c9873d007a2fc Mon Sep 17 00:00:00 2001
From: "mochel@digitalimplant.org" <mochel@digitalimplant.org>
Date: Thu, 24 Mar 2005 13:08:05 -0800
Subject: [PATCH] Remove struct device::driver_list.

Signed-off-by: Patrick Mochel <mochel@digitalimplant.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 1 -
 include/linux/device.h | 1 -
 2 files changed, 2 deletions(-)

(limited to 'include')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index f258a21ed78..bfa4268289e 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -209,7 +209,6 @@ void device_initialize(struct device *dev)
 	kobject_init(&dev->kobj);
 	INIT_LIST_HEAD(&dev->node);
 	INIT_LIST_HEAD(&dev->children);
-	INIT_LIST_HEAD(&dev->driver_list);
 	INIT_LIST_HEAD(&dev->dma_pools);
 	init_MUTEX(&dev->sem);
 }
diff --git a/include/linux/device.h b/include/linux/device.h
index 13f65853e58..d2434934d09 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -263,7 +263,6 @@ extern void class_device_destroy(struct class *cls, dev_t devt);
 
 struct device {
 	struct list_head node;		/* node in sibling list */
-	struct list_head driver_list;
 	struct list_head children;
 	struct klist_node	knode_driver;
 	struct klist_node	knode_bus;
-- 
cgit v1.2.3-70-g09d2


From 36239577cfb6b9a7c111209536b54200b0252ebf Mon Sep 17 00:00:00 2001
From: "mochel@digitalimplant.org" <mochel@digitalimplant.org>
Date: Thu, 24 Mar 2005 19:08:30 -0800
Subject: [PATCH] Use a klist for device child lists.

- Use klist iterator in device_for_each_child(), making it safe to use for
  removing devices.
- Remove unused list_to_dev() function.
- Kills all usage of devices_subsys.rwsem.

Signed-off-by: Patrick Mochel <mochel@digitalimplant.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 30 +++++++++++++++---------------
 include/linux/device.h | 10 ++--------
 2 files changed, 17 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index bfa4268289e..1d8c7790b55 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -207,8 +207,7 @@ void device_initialize(struct device *dev)
 {
 	kobj_set_kset_s(dev, devices_subsys);
 	kobject_init(&dev->kobj);
-	INIT_LIST_HEAD(&dev->node);
-	INIT_LIST_HEAD(&dev->children);
+	klist_init(&dev->klist_children);
 	INIT_LIST_HEAD(&dev->dma_pools);
 	init_MUTEX(&dev->sem);
 }
@@ -249,10 +248,8 @@ int device_add(struct device *dev)
 		goto PMError;
 	if ((error = bus_add_device(dev)))
 		goto BusError;
-	down_write(&devices_subsys.rwsem);
 	if (parent)
-		list_add_tail(&dev->node, &parent->children);
-	up_write(&devices_subsys.rwsem);
+		klist_add_tail(&parent->klist_children, &dev->knode_parent);
 
 	/* notify platform of device entry */
 	if (platform_notify)
@@ -335,10 +332,8 @@ void device_del(struct device * dev)
 {
 	struct device * parent = dev->parent;
 
-	down_write(&devices_subsys.rwsem);
 	if (parent)
-		list_del_init(&dev->node);
-	up_write(&devices_subsys.rwsem);
+		klist_remove(&dev->knode_parent);
 
 	/* Notify the platform of the removal, in case they
 	 * need to do anything...
@@ -372,6 +367,12 @@ void device_unregister(struct device * dev)
 }
 
 
+static struct device * next_device(struct klist_iter * i)
+{
+	struct klist_node * n = klist_next(i);
+	return n ? container_of(n, struct device, knode_parent) : NULL;
+}
+
 /**
  *	device_for_each_child - device child iterator.
  *	@dev:	parent struct device.
@@ -384,18 +385,17 @@ void device_unregister(struct device * dev)
  *	We check the return of @fn each time. If it returns anything
  *	other than 0, we break out and return that value.
  */
-int device_for_each_child(struct device * dev, void * data,
+int device_for_each_child(struct device * parent, void * data,
 		     int (*fn)(struct device *, void *))
 {
+	struct klist_iter i;
 	struct device * child;
 	int error = 0;
 
-	down_read(&devices_subsys.rwsem);
-	list_for_each_entry(child, &dev->children, node) {
-		if((error = fn(child, data)))
-			break;
-	}
-	up_read(&devices_subsys.rwsem);
+	klist_iter_init(&parent->klist_children, &i);
+	while ((child = next_device(&i)) && !error)
+		error = fn(child, data);
+	klist_iter_exit(&i);
 	return error;
 }
 
diff --git a/include/linux/device.h b/include/linux/device.h
index d2434934d09..43249260cd1 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -262,8 +262,8 @@ extern void class_device_destroy(struct class *cls, dev_t devt);
 
 
 struct device {
-	struct list_head node;		/* node in sibling list */
-	struct list_head children;
+	struct klist		klist_children;
+	struct klist_node	knode_parent;		/* node in sibling list */
 	struct klist_node	knode_driver;
 	struct klist_node	knode_bus;
 	struct device 	* parent;
@@ -298,12 +298,6 @@ struct device {
 	void	(*release)(struct device * dev);
 };
 
-static inline struct device *
-list_to_dev(struct list_head *node)
-{
-	return list_entry(node, struct device, node);
-}
-
 static inline void *
 dev_get_drvdata (struct device *dev)
 {
-- 
cgit v1.2.3-70-g09d2


From 0d3e5a2e39b6ba2974e9e7c2a429018c45de8e76 Mon Sep 17 00:00:00 2001
From: Patrick Mochel <mochel@digitalimplant.org>
Date: Tue, 5 Apr 2005 23:46:33 -0700
Subject: [PATCH] Driver Core: fix bk-driver-core kills ppc64

There's no check to see if the device is already bound to a driver, which
could do bad things.  The first thing to go wrong is that it will try to match
a driver with a device already bound to one.  In some cases (it appears with
USB with drivers/usb/core/usb.c::usb_match_id()), some drivers will match a
device based on the class type, so it would be common (especially for HID
devices) to match a device that is already bound.

The fun comes when ->probe() is called, it fails, then
driver_probe_device() does this:

	dev->driver = NULL;

Later on, that pointer could be be dereferenced without checking and cause
hell to break loose.

This problem could be nasty. It's very hardware dependent, since some
devices could have a different set of matching qualifiers than others.

Now, I don't quite see exactly where/how you were getting that crash.
You're dereferencing bad memory, but I'm not sure which pointer was bad
and where it came from, but it could have come from a couple of different
places.

The patch below will hopefully fix it all up for you. It's against
2.6.12-rc2-mm1, and does the following:

- Move logic to driver_probe_device() and comments uncommon returns:
  1 - If device is bound
  0 - If device not bound, and no error
  error - If there was an error.

- Move locking to caller of that function, since we want to lock a
  device for the entire time we're trying to bind it to a driver (to
  prevent against a driver being loaded at the same time).

- Update __device_attach() and __driver_attach() to do that locking.

- Check if device is already bound in __driver_attach()

- Update the converse device_release_driver() so it locks the device
  around all of the operations.

- Mark driver_probe_device() as static and remove export. It's an
  internal function, it should stay that way, and there are no other
  callers. If there is ever a need to export it, we can audit it as
  necessary.

Signed-off-by: Andrew Morton <akpm@osdl.org>
---
 drivers/base/dd.c      | 142 ++++++++++++++++++++++++++-----------------------
 include/linux/device.h |   1 -
 2 files changed, 74 insertions(+), 69 deletions(-)

(limited to 'include')

diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index dd2a8a79c12..8510918109e 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -35,6 +35,8 @@
  *	nor take the bus's rwsem. Please verify those are accounted
  *	for before calling this. (It is ok to call with no other effort
  *	from a driver's probe() method.)
+ *
+ *	This function must be called with @dev->sem held.
  */
 void device_bind_driver(struct device * dev)
 {
@@ -57,54 +59,56 @@ void device_bind_driver(struct device * dev)
  *	because we don't know the format of the ID structures, nor what
  *	is to be considered a match and what is not.
  *
- *	If we find a match, we call @drv->probe(@dev) if it exists, and
- *	call device_bind_driver() above.
+ *
+ *	This function returns 1 if a match is found, an error if one
+ *	occurs (that is not -ENODEV or -ENXIO), and 0 otherwise.
+ *
+ *	This function must be called with @dev->sem held.
  */
-int driver_probe_device(struct device_driver * drv, struct device * dev)
+static int driver_probe_device(struct device_driver * drv, struct device * dev)
 {
-	int error = 0;
+	int ret = 0;
 
 	if (drv->bus->match && !drv->bus->match(dev, drv))
-		return -ENODEV;
+		goto Done;
 
-	down(&dev->sem);
+	pr_debug("%s: Matched Device %s with Driver %s\n",
+		 drv->bus->name, dev->bus_id, drv->name);
 	dev->driver = drv;
 	if (drv->probe) {
-		error = drv->probe(dev);
-		if (error) {
+		ret = drv->probe(dev);
+		if (ret) {
 			dev->driver = NULL;
-			up(&dev->sem);
-			return error;
+			goto ProbeFailed;
 		}
 	}
-	up(&dev->sem);
 	device_bind_driver(dev);
-	return 0;
+	ret = 1;
+	pr_debug("%s: Bound Device %s to Driver %s\n",
+		 drv->bus->name, dev->bus_id, drv->name);
+	goto Done;
+
+ ProbeFailed:
+	if (ret == -ENODEV || ret == -ENXIO) {
+		/* Driver matched, but didn't support device
+		 * or device not found.
+		 * Not an error; keep going.
+		 */
+		ret = 0;
+	} else {
+		/* driver matched but the probe failed */
+		printk(KERN_WARNING
+		       "%s: probe of %s failed with error %d\n",
+		       drv->name, dev->bus_id, ret);
+	}
+ Done:
+	return ret;
 }
 
 static int __device_attach(struct device_driver * drv, void * data)
 {
 	struct device * dev = data;
-	int error;
-
-	error = driver_probe_device(drv, dev);
-	if (error) {
-		if ((error == -ENODEV) || (error == -ENXIO)) {
-			/* Driver matched, but didn't support device
-			 * or device not found.
-			 * Not an error; keep going.
-			 */
-			error = 0;
-		} else {
-			/* driver matched but the probe failed */
-			printk(KERN_WARNING
-			       "%s: probe of %s failed with error %d\n",
-			       drv->name, dev->bus_id, error);
-		}
-		return error;
-	}
-	/* stop looking, this device is attached */
-	return 1;
+	return driver_probe_device(drv, dev);
 }
 
 /**
@@ -114,37 +118,43 @@ static int __device_attach(struct device_driver * drv, void * data)
  *	Walk the list of drivers that the bus has and call
  *	driver_probe_device() for each pair. If a compatible
  *	pair is found, break out and return.
+ *
+ *	Returns 1 if the device was bound to a driver; 0 otherwise.
  */
 int device_attach(struct device * dev)
 {
+	int ret = 0;
+
+	down(&dev->sem);
 	if (dev->driver) {
 		device_bind_driver(dev);
-		return 1;
-	}
-
-	return bus_for_each_drv(dev->bus, NULL, dev, __device_attach);
+		ret = 1;
+	} else
+		ret = bus_for_each_drv(dev->bus, NULL, dev, __device_attach);
+	up(&dev->sem);
+	return ret;
 }
 
 static int __driver_attach(struct device * dev, void * data)
 {
 	struct device_driver * drv = data;
-	int error = 0;
-
-	if (!dev->driver) {
-		error = driver_probe_device(drv, dev);
-		if (error) {
-			if (error != -ENODEV) {
-				/* driver matched but the probe failed */
-				printk(KERN_WARNING
-				       "%s: probe of %s failed with error %d\n",
-				       drv->name, dev->bus_id, error);
-			} else
-				error = 0;
-			return error;
-		}
-		/* stop looking, this driver is attached */
-		return 1;
-	}
+
+	/*
+	 * Lock device and try to bind to it. We drop the error
+	 * here and always return 0, because we need to keep trying
+	 * to bind to devices and some drivers will return an error
+	 * simply if it didn't support the device.
+	 *
+	 * driver_probe_device() will spit a warning if there
+	 * is an error.
+	 */
+
+	down(&dev->sem);
+	if (!dev->driver)
+		driver_probe_device(drv, dev);
+	up(&dev->sem);
+
+
 	return 0;
 }
 
@@ -156,9 +166,6 @@ static int __driver_attach(struct device * dev, void * data)
  *	match the driver with each one.  If driver_probe_device()
  *	returns 0 and the @dev->driver is set, we've found a
  *	compatible pair.
- *
- *	Note that we ignore the -ENODEV error from driver_probe_device(),
- *	since it's perfectly valid for a driver not to bind to any devices.
  */
 void driver_attach(struct device_driver * drv)
 {
@@ -176,19 +183,19 @@ void driver_attach(struct device_driver * drv)
  */
 void device_release_driver(struct device * dev)
 {
-	struct device_driver * drv = dev->driver;
-
-	if (!drv)
-		return;
-
-	sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj));
-	sysfs_remove_link(&dev->kobj, "driver");
-	klist_del(&dev->knode_driver);
+	struct device_driver * drv;
 
 	down(&dev->sem);
-	if (drv->remove)
-		drv->remove(dev);
-	dev->driver = NULL;
+	if (dev->driver) {
+		drv = dev->driver;
+		sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj));
+		sysfs_remove_link(&dev->kobj, "driver");
+		klist_del(&dev->knode_driver);
+
+		if (drv->remove)
+			drv->remove(dev);
+		dev->driver = NULL;
+	}
 	up(&dev->sem);
 }
 
@@ -208,7 +215,6 @@ void driver_detach(struct device_driver * drv)
 }
 
 
-EXPORT_SYMBOL_GPL(driver_probe_device);
 EXPORT_SYMBOL_GPL(device_bind_driver);
 EXPORT_SYMBOL_GPL(device_release_driver);
 EXPORT_SYMBOL_GPL(device_attach);
diff --git a/include/linux/device.h b/include/linux/device.h
index 43249260cd1..91aac349b9a 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -325,7 +325,6 @@ extern int device_for_each_child(struct device *, void *,
  * Manual binding of a device to driver. See drivers/base/bus.c
  * for information on use.
  */
-extern int  driver_probe_device(struct device_driver * drv, struct device * dev);
 extern void device_bind_driver(struct device * dev);
 extern void device_release_driver(struct device * dev);
 extern int  device_attach(struct device * dev);
-- 
cgit v1.2.3-70-g09d2


From 4b45099b75832434c5113b9aed1499f8a69d13d5 Mon Sep 17 00:00:00 2001
From: Keiichiro Tokunaga <tokunaga.keiich@jp.fujitsu.com>
Date: Sun, 8 May 2005 21:28:53 +0900
Subject: [PATCH] Driver core: unregister_node() for hotplug use

This adds a generic function 'unregister_node()'.
It is used to remove objects of a node going away
for hotplug.  All the devices on the node must be
unregistered before calling this function.

Signed-off-by: Keiichiro Tokunaga <tokunaga.keiich@jp.fujitsu.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

diff -puN drivers/base/node.c~numa_hp_base drivers/base/node.c
---
 drivers/base/node.c  | 20 ++++++++++++++++++--
 include/linux/node.h |  1 +
 2 files changed, 19 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/base/node.c b/drivers/base/node.c
index 583d57ec49a..5d4517ccc42 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -136,7 +136,7 @@ static SYSDEV_ATTR(distance, S_IRUGO, node_read_distance, NULL);
  *
  * Initialize and register the node device.
  */
-int __init register_node(struct node *node, int num, struct node *parent)
+int register_node(struct node *node, int num, struct node *parent)
 {
 	int error;
 
@@ -153,8 +153,24 @@ int __init register_node(struct node *node, int num, struct node *parent)
 	return error;
 }
 
+/**
+ * unregister_node - unregister a node device
+ * @node: node going away
+ *
+ * Unregisters a node device @node.  All the devices on the node must be
+ * unregistered before calling this function.
+ */
+void unregister_node(struct node *node)
+{
+	sysdev_remove_file(&node->sysdev, &attr_cpumap);
+	sysdev_remove_file(&node->sysdev, &attr_meminfo);
+	sysdev_remove_file(&node->sysdev, &attr_numastat);
+	sysdev_remove_file(&node->sysdev, &attr_distance);
+
+	sysdev_unregister(&node->sysdev);
+}
 
-int __init register_node_type(void)
+static int __init register_node_type(void)
 {
 	return sysdev_class_register(&node_class);
 }
diff --git a/include/linux/node.h b/include/linux/node.h
index 6e0a697e594..254dc3de650 100644
--- a/include/linux/node.h
+++ b/include/linux/node.h
@@ -27,6 +27,7 @@ struct node {
 };
 
 extern int register_node(struct node *, int, struct node *);
+extern void unregister_node(struct node *node);
 
 #define to_node(sys_device) container_of(sys_device, struct node, sysdev)
 
-- 
cgit v1.2.3-70-g09d2


From acaefc25d21f850e47ecc5098d1e0bc442c526be Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Wed, 18 May 2005 14:40:59 +0200
Subject: [PATCH] libfs: add simple attribute files

Based on the discussion about spufs attributes, this is my suggestion
for a more generic attribute file support that can be used by both
debugfs and spufs.

Simple attribute files behave similarly to sequential files from
a kernel programmers perspective in that a standard set of file
operations is provided and only an open operation needs to
be written that registers file specific get() and set() functions.

These operations are defined as

void foo_set(void *data, u64 val); and
u64 foo_get(void *data);

where data is the inode->u.generic_ip pointer of the file and the
operations just need to make send of that pointer. The infrastructure
makes sure this works correctly with concurrent access and partial
read calls.

A macro named DEFINE_SIMPLE_ATTRIBUTE is provided to further simplify
using the attributes.

This patch already contains the changes for debugfs to use attributes
for its internal file operations.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 fs/debugfs/file.c  |  67 ++++++++++++++++-------------------
 fs/libfs.c         | 100 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/fs.h |  46 ++++++++++++++++++++++++
 3 files changed, 175 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c
index 548556ff250..efc97d9b786 100644
--- a/fs/debugfs/file.c
+++ b/fs/debugfs/file.c
@@ -45,44 +45,15 @@ struct file_operations debugfs_file_operations = {
 	.open =		default_open,
 };
 
-#define simple_type(type, format, temptype, strtolfn)				\
-static ssize_t read_file_##type(struct file *file, char __user *user_buf,	\
-				size_t count, loff_t *ppos)			\
-{										\
-	char buf[32];								\
-	type *val = file->private_data;						\
-										\
-	snprintf(buf, sizeof(buf), format "\n", *val);				\
-	return simple_read_from_buffer(user_buf, count, ppos, buf, strlen(buf));\
-}										\
-static ssize_t write_file_##type(struct file *file, const char __user *user_buf,\
-				 size_t count, loff_t *ppos)			\
-{										\
-	char *endp;								\
-	char buf[32];								\
-	int buf_size;								\
-	type *val = file->private_data;						\
-	temptype tmp;								\
-										\
-	memset(buf, 0x00, sizeof(buf));						\
-	buf_size = min(count, (sizeof(buf)-1));					\
-	if (copy_from_user(buf, user_buf, buf_size))				\
-		return -EFAULT;							\
-										\
-	tmp = strtolfn(buf, &endp, 0);						\
-	if ((endp == buf) || ((type)tmp != tmp))				\
-		return -EINVAL;							\
-	*val = tmp;								\
-	return count;								\
-}										\
-static struct file_operations fops_##type = {					\
-	.read =		read_file_##type,					\
-	.write =	write_file_##type,					\
-	.open =		default_open,						\
-};
-simple_type(u8, "%c", unsigned long, simple_strtoul);
-simple_type(u16, "%hi", unsigned long, simple_strtoul);
-simple_type(u32, "%i", unsigned long, simple_strtoul);
+static void debugfs_u8_set(void *data, u64 val)
+{
+	*(u8 *)data = val;
+}
+static u64 debugfs_u8_get(void *data)
+{
+	return *(u8 *)data;
+}
+DEFINE_SIMPLE_ATTRIBUTE(fops_u8, debugfs_u8_get, debugfs_u8_set, "%llu\n");
 
 /**
  * debugfs_create_u8 - create a file in the debugfs filesystem that is used to read and write a unsigned 8 bit value.
@@ -116,6 +87,16 @@ struct dentry *debugfs_create_u8(const char *name, mode_t mode,
 }
 EXPORT_SYMBOL_GPL(debugfs_create_u8);
 
+static void debugfs_u16_set(void *data, u64 val)
+{
+	*(u16 *)data = val;
+}
+static u64 debugfs_u16_get(void *data)
+{
+	return *(u16 *)data;
+}
+DEFINE_SIMPLE_ATTRIBUTE(fops_u16, debugfs_u16_get, debugfs_u16_set, "%llu\n");
+
 /**
  * debugfs_create_u16 - create a file in the debugfs filesystem that is used to read and write a unsigned 8 bit value.
  *
@@ -148,6 +129,16 @@ struct dentry *debugfs_create_u16(const char *name, mode_t mode,
 }
 EXPORT_SYMBOL_GPL(debugfs_create_u16);
 
+static void debugfs_u32_set(void *data, u64 val)
+{
+	*(u32 *)data = val;
+}
+static u64 debugfs_u32_get(void *data)
+{
+	return *(u32 *)data;
+}
+DEFINE_SIMPLE_ATTRIBUTE(fops_u32, debugfs_u32_get, debugfs_u32_set, "%llu\n");
+
 /**
  * debugfs_create_u32 - create a file in the debugfs filesystem that is used to read and write a unsigned 8 bit value.
  *
diff --git a/fs/libfs.c b/fs/libfs.c
index f90b2959592..5025563e737 100644
--- a/fs/libfs.c
+++ b/fs/libfs.c
@@ -519,6 +519,102 @@ int simple_transaction_release(struct inode *inode, struct file *file)
 	return 0;
 }
 
+/* Simple attribute files */
+
+struct simple_attr {
+	u64 (*get)(void *);
+	void (*set)(void *, u64);
+	char get_buf[24];	/* enough to store a u64 and "\n\0" */
+	char set_buf[24];
+	void *data;
+	const char *fmt;	/* format for read operation */
+	struct semaphore sem;	/* protects access to these buffers */
+};
+
+/* simple_attr_open is called by an actual attribute open file operation
+ * to set the attribute specific access operations. */
+int simple_attr_open(struct inode *inode, struct file *file,
+		     u64 (*get)(void *), void (*set)(void *, u64),
+		     const char *fmt)
+{
+	struct simple_attr *attr;
+
+	attr = kmalloc(sizeof(*attr), GFP_KERNEL);
+	if (!attr)
+		return -ENOMEM;
+
+	attr->get = get;
+	attr->set = set;
+	attr->data = inode->u.generic_ip;
+	attr->fmt = fmt;
+	init_MUTEX(&attr->sem);
+
+	file->private_data = attr;
+
+	return nonseekable_open(inode, file);
+}
+
+int simple_attr_close(struct inode *inode, struct file *file)
+{
+	kfree(file->private_data);
+	return 0;
+}
+
+/* read from the buffer that is filled with the get function */
+ssize_t simple_attr_read(struct file *file, char __user *buf,
+			 size_t len, loff_t *ppos)
+{
+	struct simple_attr *attr;
+	size_t size;
+	ssize_t ret;
+
+	attr = file->private_data;
+
+	if (!attr->get)
+		return -EACCES;
+
+	down(&attr->sem);
+	if (*ppos) /* continued read */
+		size = strlen(attr->get_buf);
+	else	  /* first read */
+		size = scnprintf(attr->get_buf, sizeof(attr->get_buf),
+				 attr->fmt,
+				 (unsigned long long)attr->get(attr->data));
+
+	ret = simple_read_from_buffer(buf, len, ppos, attr->get_buf, size);
+	up(&attr->sem);
+	return ret;
+}
+
+/* interpret the buffer as a number to call the set function with */
+ssize_t simple_attr_write(struct file *file, const char __user *buf,
+			  size_t len, loff_t *ppos)
+{
+	struct simple_attr *attr;
+	u64 val;
+	size_t size;
+	ssize_t ret;
+
+	attr = file->private_data;
+
+	if (!attr->set)
+		return -EACCES;
+
+	down(&attr->sem);
+	ret = -EFAULT;
+	size = min(sizeof(attr->set_buf) - 1, len);
+	if (copy_from_user(attr->set_buf, buf, size))
+		goto out;
+
+	ret = len; /* claim we got the whole input */
+	attr->set_buf[size] = '\0';
+	val = simple_strtol(attr->set_buf, NULL, 0);
+	attr->set(attr->data, val);
+out:
+	up(&attr->sem);
+	return ret;
+}
+
 EXPORT_SYMBOL(dcache_dir_close);
 EXPORT_SYMBOL(dcache_dir_lseek);
 EXPORT_SYMBOL(dcache_dir_open);
@@ -547,3 +643,7 @@ EXPORT_SYMBOL(simple_read_from_buffer);
 EXPORT_SYMBOL(simple_transaction_get);
 EXPORT_SYMBOL(simple_transaction_read);
 EXPORT_SYMBOL(simple_transaction_release);
+EXPORT_SYMBOL_GPL(simple_attr_open);
+EXPORT_SYMBOL_GPL(simple_attr_close);
+EXPORT_SYMBOL_GPL(simple_attr_read);
+EXPORT_SYMBOL_GPL(simple_attr_write);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 0180102dace..9b8b696d4f1 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1657,6 +1657,52 @@ static inline void simple_transaction_set(struct file *file, size_t n)
 	ar->size = n;
 }
 
+/*
+ * simple attribute files
+ *
+ * These attributes behave similar to those in sysfs:
+ *
+ * Writing to an attribute immediately sets a value, an open file can be
+ * written to multiple times.
+ *
+ * Reading from an attribute creates a buffer from the value that might get
+ * read with multiple read calls. When the attribute has been read
+ * completely, no further read calls are possible until the file is opened
+ * again.
+ *
+ * All attributes contain a text representation of a numeric value
+ * that are accessed with the get() and set() functions.
+ */
+#define DEFINE_SIMPLE_ATTRIBUTE(__fops, __get, __set, __fmt)		\
+static int __fops ## _open(struct inode *inode, struct file *file)	\
+{									\
+	__simple_attr_check_format(__fmt, 0ull);			\
+	return simple_attr_open(inode, file, __get, __set, __fmt);	\
+}									\
+static struct file_operations __fops = {				\
+	.owner	 = THIS_MODULE,						\
+	.open	 = __fops ## _open,					\
+	.release = simple_attr_close,					\
+	.read	 = simple_attr_read,					\
+	.write	 = simple_attr_write,					\
+};
+
+static inline void __attribute__((format(printf, 1, 2)))
+__simple_attr_check_format(const char *fmt, ...)
+{
+	/* don't do anything, just let the compiler check the arguments; */
+}
+
+int simple_attr_open(struct inode *inode, struct file *file,
+		     u64 (*get)(void *), void (*set)(void *, u64),
+		     const char *fmt);
+int simple_attr_close(struct inode *inode, struct file *file);
+ssize_t simple_attr_read(struct file *file, char __user *buf,
+			 size_t len, loff_t *ppos);
+ssize_t simple_attr_write(struct file *file, const char __user *buf,
+			  size_t len, loff_t *ppos);
+
+
 #ifdef CONFIG_SECURITY
 static inline char *alloc_secdata(void)
 {
-- 
cgit v1.2.3-70-g09d2


From 54b6f35c99974e99e64c05c2895718355123c55f Mon Sep 17 00:00:00 2001
From: Yani Ioannou <yani.ioannou@gmail.com>
Date: Tue, 17 May 2005 06:39:34 -0400
Subject: [PATCH] Driver core: change device_attribute callbacks

This patch adds the device_attribute paramerter to the
device_attribute store and show sysfs callback functions, and passes a
reference to the attribute when the callbacks are called.

Signed-off-by: Yani Ioannou <yani.ioannou@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 4 ++--
 include/linux/device.h | 6 ++++--
 2 files changed, 6 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 1d8c7790b55..86d79755fbf 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -39,7 +39,7 @@ dev_attr_show(struct kobject * kobj, struct attribute * attr, char * buf)
 	ssize_t ret = -EIO;
 
 	if (dev_attr->show)
-		ret = dev_attr->show(dev, buf);
+		ret = dev_attr->show(dev, dev_attr, buf);
 	return ret;
 }
 
@@ -52,7 +52,7 @@ dev_attr_store(struct kobject * kobj, struct attribute * attr,
 	ssize_t ret = -EIO;
 
 	if (dev_attr->store)
-		ret = dev_attr->store(dev, buf, count);
+		ret = dev_attr->store(dev, dev_attr, buf, count);
 	return ret;
 }
 
diff --git a/include/linux/device.h b/include/linux/device.h
index 91aac349b9a..7b781a72b29 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -335,8 +335,10 @@ extern void driver_attach(struct device_driver * drv);
 
 struct device_attribute {
 	struct attribute	attr;
-	ssize_t (*show)(struct device * dev, char * buf);
-	ssize_t (*store)(struct device * dev, const char * buf, size_t count);
+	ssize_t (*show)(struct device *dev, struct device_attribute *attr,
+			char *buf);
+	ssize_t (*store)(struct device *dev, struct device_attribute *attr,
+			 const char *buf, size_t count);
 };
 
 #define DEVICE_ATTR(_name,_mode,_show,_store) \
-- 
cgit v1.2.3-70-g09d2


From f2d03e1b3f00f1c5971463ab0101bef0c521ad3b Mon Sep 17 00:00:00 2001
From: Yani Ioannou <yani.ioannou@gmail.com>
Date: Tue, 17 May 2005 06:44:59 -0400
Subject: [PATCH] Driver Core: include: update device attribute callbacks

Signed-off-by: Yani Ioannou <yani.ioannou@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/asm-ppc/ocp.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-ppc/ocp.h b/include/asm-ppc/ocp.h
index b98db3cdae8..c726f184519 100644
--- a/include/asm-ppc/ocp.h
+++ b/include/asm-ppc/ocp.h
@@ -189,7 +189,7 @@ extern void ocp_for_each_device(void(*callback)(struct ocp_device *, void *arg),
 /* Sysfs support */
 #define OCP_SYSFS_ADDTL(type, format, name, field)			\
 static ssize_t								\
-show_##name##_##field(struct device *dev, char *buf)			\
+show_##name##_##field(struct device *dev, struct device_attribute *attr, char *buf)			\
 {									\
 	struct ocp_device *odev = to_ocp_dev(dev);			\
 	type *add = odev->def->additions;				\
-- 
cgit v1.2.3-70-g09d2


From 0a3e7eeabc9f76b7496488aad2d11626ff6a2a4f Mon Sep 17 00:00:00 2001
From: Yani Ioannou <yani.ioannou@gmail.com>
Date: Tue, 17 May 2005 22:59:05 -0400
Subject: [PATCH] I2C: add i2c sensor_device_attribute and macros

This patch creates a new header with a potential standard i2c sensor
attribute type (which simply includes an int representing the sensor
number/index) and the associated macros, SENSOR_DEVICE_ATTR to define
a static attribute and to_sensor_dev_attr to get a
sensor_device_attribute reference from an embedded device_attribute
reference.

Signed-off-by: Yani Ioannou <yani.ioannou@gmail.com>
---
 include/linux/i2c-sysfs.h | 36 ++++++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)
 create mode 100644 include/linux/i2c-sysfs.h

(limited to 'include')

diff --git a/include/linux/i2c-sysfs.h b/include/linux/i2c-sysfs.h
new file mode 100644
index 00000000000..d7bf6ce1167
--- /dev/null
+++ b/include/linux/i2c-sysfs.h
@@ -0,0 +1,36 @@
+/*
+ *  i2c-sysfs.h - i2c chip driver sysfs defines
+ *
+ *  Copyright (C) 2005 Yani Ioannou <yani.ioannou@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+#ifndef _LINUX_I2C_SYSFS_H
+#define _LINUX_I2C_SYSFS_H
+
+struct sensor_device_attribute{
+	struct device_attribute dev_attr;
+	int index;
+};
+#define to_sensor_dev_attr(_dev_attr) \
+	container_of(_dev_attr, struct sensor_device_attribute, dev_attr)
+
+#define SENSOR_DEVICE_ATTR(_name,_mode,_show,_store,_index)	\
+struct sensor_device_attribute sensor_dev_attr_##_name = {	\
+	.dev_attr =	__ATTR(_name,_mode,_show,_store),	\
+	.index =	_index,					\
+}
+
+#endif /* _LINUX_I2C_SYSFS_H */
-- 
cgit v1.2.3-70-g09d2


From 988d186de5b6966a71a8cc52e6cb4895fd2f7799 Mon Sep 17 00:00:00 2001
From: Maneesh Soni <maneesh@in.ibm.com>
Date: Tue, 31 May 2005 10:39:14 +0530
Subject: [PATCH] sysfs-iattr: add sysfs_setattr

o This adds ->i_op->setattr VFS method for sysfs inodes. The changed
  attribues are saved in the persistent sysfs_dirent structure as a pointer
  to struct iattr. The struct iattr is allocated only for those sysfs_dirent's
  for which default attributes are getting changed. Thanks to Jon Smirl for
  this suggestion.

Signed-off-by: Maneesh Soni <maneesh@in.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 fs/sysfs/dir.c        |  1 +
 fs/sysfs/inode.c      | 65 +++++++++++++++++++++++++++++++++++++++++++++++++++
 fs/sysfs/sysfs.h      |  2 ++
 include/linux/sysfs.h |  1 +
 4 files changed, 69 insertions(+)

(limited to 'include')

diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c
index 5cf3270014c..37d7a6875d8 100644
--- a/fs/sysfs/dir.c
+++ b/fs/sysfs/dir.c
@@ -233,6 +233,7 @@ static struct dentry * sysfs_lookup(struct inode *dir, struct dentry *dentry,
 
 struct inode_operations sysfs_dir_inode_operations = {
 	.lookup		= sysfs_lookup,
+	.setattr	= sysfs_setattr,
 };
 
 static void remove_dir(struct dentry * d)
diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c
index aff7b2dfa8e..400c90be568 100644
--- a/fs/sysfs/inode.c
+++ b/fs/sysfs/inode.c
@@ -26,6 +26,71 @@ static struct backing_dev_info sysfs_backing_dev_info = {
 	.capabilities	= BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK,
 };
 
+static struct inode_operations sysfs_inode_operations ={
+	.setattr	= sysfs_setattr,
+};
+
+int sysfs_setattr(struct dentry * dentry, struct iattr * iattr)
+{
+	struct inode * inode = dentry->d_inode;
+	struct sysfs_dirent * sd = dentry->d_fsdata;
+	struct iattr * sd_iattr;
+	unsigned int ia_valid = iattr->ia_valid;
+	int error;
+
+	if (!sd)
+		return -EINVAL;
+
+	sd_iattr = sd->s_iattr;
+
+	error = inode_change_ok(inode, iattr);
+	if (error)
+		return error;
+
+	error = inode_setattr(inode, iattr);
+	if (error)
+		return error;
+
+	if (!sd_iattr) {
+		/* setting attributes for the first time, allocate now */
+		sd_iattr = kmalloc(sizeof(struct iattr), GFP_KERNEL);
+		if (!sd_iattr)
+			return -ENOMEM;
+		/* assign default attributes */
+		memset(sd_iattr, 0, sizeof(struct iattr));
+		sd_iattr->ia_mode = sd->s_mode;
+		sd_iattr->ia_uid = 0;
+		sd_iattr->ia_gid = 0;
+		sd_iattr->ia_atime = sd_iattr->ia_mtime = sd_iattr->ia_ctime = CURRENT_TIME;
+		sd->s_iattr = sd_iattr;
+	}
+
+	/* attributes were changed atleast once in past */
+
+	if (ia_valid & ATTR_UID)
+		sd_iattr->ia_uid = iattr->ia_uid;
+	if (ia_valid & ATTR_GID)
+		sd_iattr->ia_gid = iattr->ia_gid;
+	if (ia_valid & ATTR_ATIME)
+		sd_iattr->ia_atime = timespec_trunc(iattr->ia_atime,
+						inode->i_sb->s_time_gran);
+	if (ia_valid & ATTR_MTIME)
+		sd_iattr->ia_mtime = timespec_trunc(iattr->ia_mtime,
+						inode->i_sb->s_time_gran);
+	if (ia_valid & ATTR_CTIME)
+		sd_iattr->ia_ctime = timespec_trunc(iattr->ia_ctime,
+						inode->i_sb->s_time_gran);
+	if (ia_valid & ATTR_MODE) {
+		umode_t mode = iattr->ia_mode;
+
+		if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID))
+			mode &= ~S_ISGID;
+		sd_iattr->ia_mode = mode;
+	}
+
+	return error;
+}
+
 struct inode * sysfs_new_inode(mode_t mode)
 {
 	struct inode * inode = new_inode(sysfs_sb);
diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h
index a8a24a0c0b3..e8e9f0c2573 100644
--- a/fs/sysfs/sysfs.h
+++ b/fs/sysfs/sysfs.h
@@ -17,6 +17,7 @@ extern void sysfs_remove_subdir(struct dentry *);
 
 extern const unsigned char * sysfs_get_name(struct sysfs_dirent *sd);
 extern void sysfs_drop_dentry(struct sysfs_dirent *sd, struct dentry *parent);
+extern int sysfs_setattr(struct dentry *dentry, struct iattr *iattr);
 
 extern struct rw_semaphore sysfs_rename_sem;
 extern struct super_block * sysfs_sb;
@@ -75,6 +76,7 @@ static inline void release_sysfs_dirent(struct sysfs_dirent * sd)
 		kobject_put(sl->target_kobj);
 		kfree(sl);
 	}
+	kfree(sd->s_iattr);
 	kmem_cache_free(sysfs_dir_cachep, sd);
 }
 
diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
index d9cd2d31d37..392da5a6dac 100644
--- a/include/linux/sysfs.h
+++ b/include/linux/sysfs.h
@@ -73,6 +73,7 @@ struct sysfs_dirent {
 	int			s_type;
 	umode_t			s_mode;
 	struct dentry		* s_dentry;
+	struct iattr		* s_iattr;
 };
 
 #define SYSFS_ROOT		0x0001
-- 
cgit v1.2.3-70-g09d2


From 18b504e25fd617bee8830d2cdcaff7fb7b5931bb Mon Sep 17 00:00:00 2001
From: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Date: Tue, 21 Jun 2005 12:38:48 -0700
Subject: [NETLINK]: netlink_callback structure needs 5 args not 4

net/ipv4/tcp_diag.c uses up to ->args[4]

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netlink.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/netlink.h b/include/linux/netlink.h
index 561d4dc7583..3029cad63a0 100644
--- a/include/linux/netlink.h
+++ b/include/linux/netlink.h
@@ -147,7 +147,7 @@ struct netlink_callback
 	int		(*dump)(struct sk_buff * skb, struct netlink_callback *cb);
 	int		(*done)(struct netlink_callback *cb);
 	int		family;
-	long		args[4];
+	long		args[5];
 };
 
 struct netlink_notify
-- 
cgit v1.2.3-70-g09d2


From 0d51aa80a9b1db43920c0770c3bb842dd823c005 Mon Sep 17 00:00:00 2001
From: Jamal Hadi Salim <hadi@cyberus.ca>
Date: Tue, 21 Jun 2005 13:51:04 -0700
Subject: [IPV6]: V6 route events reported with wrong netlink PID and seq
 number

Essentially netlink at the moment always reports a pid and sequence of 0
always for v6 route activities.
To understand the repurcassions of this look at:
http://lists.quagga.net/pipermail/quagga-dev/2005-June/003507.html

While fixing this, i took the liberty to resolve the outstanding issue
of IPV6 routes inserted via ioctls to have the correct pids as well.

This patch tries to behave as close as possible to the v4 routes i.e
maintains whatever PID the socket issuing the command owns as opposed to
the process. That made the patch a little bulky.

I have tested against both netlink derived utility to add/del routes as
well as ioctl derived one. The Quagga folks have tested against quagga.
This fixes the problem and so far hasnt been detected to introduce any
new issues.

Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h   |  9 ++++--
 include/net/ip6_route.h |  9 ++++--
 net/ipv6/addrconf.c     | 14 ++++-----
 net/ipv6/anycast.c      |  4 +--
 net/ipv6/ip6_fib.c      | 19 ++++++------
 net/ipv6/ndisc.c        |  4 +--
 net/ipv6/route.c        | 78 ++++++++++++++++++++++++++-----------------------
 7 files changed, 74 insertions(+), 63 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 31990451819..a66e9de16a6 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -167,14 +167,17 @@ extern int			fib6_walk_continue(struct fib6_walker_t *w);
 extern int			fib6_add(struct fib6_node *root,
 					 struct rt6_info *rt,
 					 struct nlmsghdr *nlh,
-					 void *rtattr);
+					 void *rtattr,
+					 struct netlink_skb_parms *req);
 
 extern int			fib6_del(struct rt6_info *rt,
 					 struct nlmsghdr *nlh,
-					 void *rtattr);
+					 void *rtattr,
+					 struct netlink_skb_parms *req);
 
 extern void			inet6_rt_notify(int event, struct rt6_info *rt,
-						struct nlmsghdr *nlh);
+						struct nlmsghdr *nlh,
+						struct netlink_skb_parms *req);
 
 extern void			fib6_run_gc(unsigned long dummy);
 
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index d5d1dd10cdb..f920706d526 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -41,13 +41,16 @@ extern int			ipv6_route_ioctl(unsigned int cmd, void __user *arg);
 
 extern int			ip6_route_add(struct in6_rtmsg *rtmsg,
 					      struct nlmsghdr *,
-					      void *rtattr);
+					      void *rtattr,
+					      struct netlink_skb_parms *req);
 extern int			ip6_ins_rt(struct rt6_info *,
 					   struct nlmsghdr *,
-					   void *rtattr);
+					   void *rtattr,
+					   struct netlink_skb_parms *req);
 extern int			ip6_del_rt(struct rt6_info *,
 					   struct nlmsghdr *,
-					   void *rtattr);
+					   void *rtattr,
+					   struct netlink_skb_parms *req);
 
 extern int			ip6_rt_addr_add(struct in6_addr *addr,
 						struct net_device *dev,
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 47a30c3188e..14f5c53235f 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -695,7 +695,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp)
 
 		if (rt && ((rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0)) {
 			if (onlink == 0) {
-				ip6_del_rt(rt, NULL, NULL);
+				ip6_del_rt(rt, NULL, NULL, NULL);
 				rt = NULL;
 			} else if (!(rt->rt6i_flags & RTF_EXPIRES)) {
 				rt->rt6i_expires = expires;
@@ -1340,7 +1340,7 @@ addrconf_prefix_route(struct in6_addr *pfx, int plen, struct net_device *dev,
 	if (dev->type == ARPHRD_SIT && (dev->flags&IFF_POINTOPOINT))
 		rtmsg.rtmsg_flags |= RTF_NONEXTHOP;
 
-	ip6_route_add(&rtmsg, NULL, NULL);
+	ip6_route_add(&rtmsg, NULL, NULL, NULL);
 }
 
 /* Create "default" multicast route to the interface */
@@ -1357,7 +1357,7 @@ static void addrconf_add_mroute(struct net_device *dev)
 	rtmsg.rtmsg_ifindex = dev->ifindex;
 	rtmsg.rtmsg_flags = RTF_UP;
 	rtmsg.rtmsg_type = RTMSG_NEWROUTE;
-	ip6_route_add(&rtmsg, NULL, NULL);
+	ip6_route_add(&rtmsg, NULL, NULL, NULL);
 }
 
 static void sit_route_add(struct net_device *dev)
@@ -1374,7 +1374,7 @@ static void sit_route_add(struct net_device *dev)
 	rtmsg.rtmsg_flags	= RTF_UP|RTF_NONEXTHOP;
 	rtmsg.rtmsg_ifindex	= dev->ifindex;
 
-	ip6_route_add(&rtmsg, NULL, NULL);
+	ip6_route_add(&rtmsg, NULL, NULL, NULL);
 }
 
 static void addrconf_add_lroute(struct net_device *dev)
@@ -1467,7 +1467,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len)
 		if (rt && ((rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0)) {
 			if (rt->rt6i_flags&RTF_EXPIRES) {
 				if (valid_lft == 0) {
-					ip6_del_rt(rt, NULL, NULL);
+					ip6_del_rt(rt, NULL, NULL, NULL);
 					rt = NULL;
 				} else {
 					rt->rt6i_expires = rt_expires;
@@ -3094,7 +3094,7 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
 	switch (event) {
 	case RTM_NEWADDR:
 		dst_hold(&ifp->rt->u.dst);
-		if (ip6_ins_rt(ifp->rt, NULL, NULL))
+		if (ip6_ins_rt(ifp->rt, NULL, NULL, NULL))
 			dst_release(&ifp->rt->u.dst);
 		if (ifp->idev->cnf.forwarding)
 			addrconf_join_anycast(ifp);
@@ -3104,7 +3104,7 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
 			addrconf_leave_anycast(ifp);
 		addrconf_leave_solict(ifp->idev, &ifp->addr);
 		dst_hold(&ifp->rt->u.dst);
-		if (ip6_del_rt(ifp->rt, NULL, NULL))
+		if (ip6_del_rt(ifp->rt, NULL, NULL, NULL))
 			dst_free(&ifp->rt->u.dst);
 		else
 			dst_release(&ifp->rt->u.dst);
diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c
index 5d22ca3cca2..6b729404723 100644
--- a/net/ipv6/anycast.c
+++ b/net/ipv6/anycast.c
@@ -337,7 +337,7 @@ int ipv6_dev_ac_inc(struct net_device *dev, struct in6_addr *addr)
 	write_unlock_bh(&idev->lock);
 
 	dst_hold(&rt->u.dst);
-	if (ip6_ins_rt(rt, NULL, NULL))
+	if (ip6_ins_rt(rt, NULL, NULL, NULL))
 		dst_release(&rt->u.dst);
 
 	addrconf_join_solict(dev, &aca->aca_addr);
@@ -380,7 +380,7 @@ int __ipv6_dev_ac_dec(struct inet6_dev *idev, struct in6_addr *addr)
 	addrconf_leave_solict(idev, &aca->aca_addr);
 
 	dst_hold(&aca->aca_rt->u.dst);
-	if (ip6_del_rt(aca->aca_rt, NULL, NULL))
+	if (ip6_del_rt(aca->aca_rt, NULL, NULL, NULL))
 		dst_free(&aca->aca_rt->u.dst);
 	else
 		dst_release(&aca->aca_rt->u.dst);
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 405740b75ab..1b354aa9793 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -394,7 +394,7 @@ insert_above:
  */
 
 static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt,
-    struct nlmsghdr *nlh)
+		struct nlmsghdr *nlh,  struct netlink_skb_parms *req)
 {
 	struct rt6_info *iter = NULL;
 	struct rt6_info **ins;
@@ -449,7 +449,7 @@ out:
 	*ins = rt;
 	rt->rt6i_node = fn;
 	atomic_inc(&rt->rt6i_ref);
-	inet6_rt_notify(RTM_NEWROUTE, rt, nlh);
+	inet6_rt_notify(RTM_NEWROUTE, rt, nlh, req);
 	rt6_stats.fib_rt_entries++;
 
 	if ((fn->fn_flags & RTN_RTINFO) == 0) {
@@ -479,7 +479,8 @@ void fib6_force_start_gc(void)
  *	with source addr info in sub-trees
  */
 
-int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
+int fib6_add(struct fib6_node *root, struct rt6_info *rt, 
+		struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
 {
 	struct fib6_node *fn;
 	int err = -ENOMEM;
@@ -552,7 +553,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nlmsghdr *nlh,
 	}
 #endif
 
-	err = fib6_add_rt2node(fn, rt, nlh);
+	err = fib6_add_rt2node(fn, rt, nlh, req);
 
 	if (err == 0) {
 		fib6_start_gc(rt);
@@ -859,7 +860,7 @@ static struct fib6_node * fib6_repair_tree(struct fib6_node *fn)
 }
 
 static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
-    struct nlmsghdr *nlh, void *_rtattr)
+    struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
 {
 	struct fib6_walker_t *w;
 	struct rt6_info *rt = *rtp;
@@ -915,11 +916,11 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
 		if (atomic_read(&rt->rt6i_ref) != 1) BUG();
 	}
 
-	inet6_rt_notify(RTM_DELROUTE, rt, nlh);
+	inet6_rt_notify(RTM_DELROUTE, rt, nlh, req);
 	rt6_release(rt);
 }
 
-int fib6_del(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
+int fib6_del(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
 {
 	struct fib6_node *fn = rt->rt6i_node;
 	struct rt6_info **rtp;
@@ -944,7 +945,7 @@ int fib6_del(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
 
 	for (rtp = &fn->leaf; *rtp; rtp = &(*rtp)->u.next) {
 		if (*rtp == rt) {
-			fib6_del_route(fn, rtp, nlh, _rtattr);
+			fib6_del_route(fn, rtp, nlh, _rtattr, req);
 			return 0;
 		}
 	}
@@ -1073,7 +1074,7 @@ static int fib6_clean_node(struct fib6_walker_t *w)
 		res = c->func(rt, c->arg);
 		if (res < 0) {
 			w->leaf = rt;
-			res = fib6_del(rt, NULL, NULL);
+			res = fib6_del(rt, NULL, NULL, NULL);
 			if (res) {
 #if RT6_DEBUG >= 2
 				printk(KERN_DEBUG "fib6_clean_node: del failed: rt=%p@%p err=%d\n", rt, rt->rt6i_node, res);
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 7c291f4e9ed..7ae72d4c9bd 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -955,7 +955,7 @@ static void ndisc_recv_na(struct sk_buff *skb)
 			struct rt6_info *rt;
 			rt = rt6_get_dflt_router(saddr, dev);
 			if (rt)
-				ip6_del_rt(rt, NULL, NULL);
+				ip6_del_rt(rt, NULL, NULL, NULL);
 		}
 
 out:
@@ -1096,7 +1096,7 @@ static void ndisc_router_discovery(struct sk_buff *skb)
 
 	if (rt && lifetime == 0) {
 		neigh_clone(neigh);
-		ip6_del_rt(rt, NULL, NULL);
+		ip6_del_rt(rt, NULL, NULL, NULL);
 		rt = NULL;
 	}
 
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 1f5b226c357..878789b3122 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -384,12 +384,13 @@ struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr,
    be destroyed.
  */
 
-int ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
+int ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh,
+		void *_rtattr, struct netlink_skb_parms *req)
 {
 	int err;
 
 	write_lock_bh(&rt6_lock);
-	err = fib6_add(&ip6_routing_table, rt, nlh, _rtattr);
+	err = fib6_add(&ip6_routing_table, rt, nlh, _rtattr, req);
 	write_unlock_bh(&rt6_lock);
 
 	return err;
@@ -400,7 +401,7 @@ int ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
  */
 
 static struct rt6_info *rt6_cow(struct rt6_info *ort, struct in6_addr *daddr,
-				struct in6_addr *saddr)
+				struct in6_addr *saddr, struct netlink_skb_parms *req)
 {
 	int err;
 	struct rt6_info *rt;
@@ -432,7 +433,7 @@ static struct rt6_info *rt6_cow(struct rt6_info *ort, struct in6_addr *daddr,
 
 		dst_hold(&rt->u.dst);
 
-		err = ip6_ins_rt(rt, NULL, NULL);
+		err = ip6_ins_rt(rt, NULL, NULL, req);
 		if (err == 0)
 			return rt;
 
@@ -491,7 +492,8 @@ restart:
 		read_unlock_bh(&rt6_lock);
 
 		nrt = rt6_cow(rt, &skb->nh.ipv6h->daddr,
-			      &skb->nh.ipv6h->saddr);
+			      &skb->nh.ipv6h->saddr,
+			      &NETLINK_CB(skb));
 
 		dst_release(&rt->u.dst);
 		rt = nrt;
@@ -551,7 +553,7 @@ restart:
 		dst_hold(&rt->u.dst);
 		read_unlock_bh(&rt6_lock);
 
-		nrt = rt6_cow(rt, &fl->fl6_dst, &fl->fl6_src);
+		nrt = rt6_cow(rt, &fl->fl6_dst, &fl->fl6_src, NULL);
 
 		dst_release(&rt->u.dst);
 		rt = nrt;
@@ -598,7 +600,7 @@ static struct dst_entry *ip6_negative_advice(struct dst_entry *dst)
 
 	if (rt) {
 		if (rt->rt6i_flags & RTF_CACHE)
-			ip6_del_rt(rt, NULL, NULL);
+			ip6_del_rt(rt, NULL, NULL, NULL);
 		else
 			dst_release(dst);
 	}
@@ -787,7 +789,8 @@ int ipv6_get_hoplimit(struct net_device *dev)
  *
  */
 
-int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, void *_rtattr)
+int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, 
+		void *_rtattr, struct netlink_skb_parms *req)
 {
 	int err;
 	struct rtmsg *r;
@@ -974,7 +977,7 @@ install_route:
 		rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&rt->u.dst));
 	rt->u.dst.dev = dev;
 	rt->rt6i_idev = idev;
-	return ip6_ins_rt(rt, nlh, _rtattr);
+	return ip6_ins_rt(rt, nlh, _rtattr, req);
 
 out:
 	if (dev)
@@ -986,7 +989,7 @@ out:
 	return err;
 }
 
-int ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
+int ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
 {
 	int err;
 
@@ -994,7 +997,7 @@ int ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
 
 	rt6_reset_dflt_pointer(NULL);
 
-	err = fib6_del(rt, nlh, _rtattr);
+	err = fib6_del(rt, nlh, _rtattr, req);
 	dst_release(&rt->u.dst);
 
 	write_unlock_bh(&rt6_lock);
@@ -1002,7 +1005,7 @@ int ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr)
 	return err;
 }
 
-static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, void *_rtattr)
+static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
 {
 	struct fib6_node *fn;
 	struct rt6_info *rt;
@@ -1029,7 +1032,7 @@ static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, void *_r
 			dst_hold(&rt->u.dst);
 			read_unlock_bh(&rt6_lock);
 
-			return ip6_del_rt(rt, nlh, _rtattr);
+			return ip6_del_rt(rt, nlh, _rtattr, req);
 		}
 	}
 	read_unlock_bh(&rt6_lock);
@@ -1136,11 +1139,11 @@ source_ok:
 	nrt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(neigh->dev);
 	nrt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&nrt->u.dst));
 
-	if (ip6_ins_rt(nrt, NULL, NULL))
+	if (ip6_ins_rt(nrt, NULL, NULL, NULL))
 		goto out;
 
 	if (rt->rt6i_flags&RTF_CACHE) {
-		ip6_del_rt(rt, NULL, NULL);
+		ip6_del_rt(rt, NULL, NULL, NULL);
 		return;
 	}
 
@@ -1204,7 +1207,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
 	   2. It is gatewayed route or NONEXTHOP route. Action: clone it.
 	 */
 	if (!rt->rt6i_nexthop && !(rt->rt6i_flags & RTF_NONEXTHOP)) {
-		nrt = rt6_cow(rt, daddr, saddr);
+		nrt = rt6_cow(rt, daddr, saddr, NULL);
 		if (!nrt->u.dst.error) {
 			nrt->u.dst.metrics[RTAX_MTU-1] = pmtu;
 			if (allfrag)
@@ -1232,7 +1235,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
 		nrt->u.dst.metrics[RTAX_MTU-1] = pmtu;
 		if (allfrag)
 			nrt->u.dst.metrics[RTAX_FEATURES-1] |= RTAX_FEATURE_ALLFRAG;
-		ip6_ins_rt(nrt, NULL, NULL);
+		ip6_ins_rt(nrt, NULL, NULL, NULL);
 	}
 
 out:
@@ -1305,7 +1308,7 @@ struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr,
 
 	rtmsg.rtmsg_ifindex = dev->ifindex;
 
-	ip6_route_add(&rtmsg, NULL, NULL);
+	ip6_route_add(&rtmsg, NULL, NULL, NULL);
 	return rt6_get_dflt_router(gwaddr, dev);
 }
 
@@ -1323,7 +1326,7 @@ restart:
 
 			read_unlock_bh(&rt6_lock);
 
-			ip6_del_rt(rt, NULL, NULL);
+			ip6_del_rt(rt, NULL, NULL, NULL);
 
 			goto restart;
 		}
@@ -1349,10 +1352,10 @@ int ipv6_route_ioctl(unsigned int cmd, void __user *arg)
 		rtnl_lock();
 		switch (cmd) {
 		case SIOCADDRT:
-			err = ip6_route_add(&rtmsg, NULL, NULL);
+			err = ip6_route_add(&rtmsg, NULL, NULL, NULL);
 			break;
 		case SIOCDELRT:
-			err = ip6_route_del(&rtmsg, NULL, NULL);
+			err = ip6_route_del(&rtmsg, NULL, NULL, NULL);
 			break;
 		default:
 			err = -EINVAL;
@@ -1546,7 +1549,7 @@ int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 
 	if (inet6_rtm_to_rtmsg(r, arg, &rtmsg))
 		return -EINVAL;
-	return ip6_route_del(&rtmsg, nlh, arg);
+	return ip6_route_del(&rtmsg, nlh, arg, &NETLINK_CB(skb));
 }
 
 int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
@@ -1556,7 +1559,7 @@ int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 
 	if (inet6_rtm_to_rtmsg(r, arg, &rtmsg))
 		return -EINVAL;
-	return ip6_route_add(&rtmsg, nlh, arg);
+	return ip6_route_add(&rtmsg, nlh, arg, &NETLINK_CB(skb));
 }
 
 struct rt6_rtnl_dump_arg
@@ -1566,12 +1569,9 @@ struct rt6_rtnl_dump_arg
 };
 
 static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
-			 struct in6_addr *dst,
-			 struct in6_addr *src,
-			 int iif,
-			 int type, u32 pid, u32 seq,
-			 struct nlmsghdr *in_nlh, int prefix,
-			 unsigned int flags)
+			 struct in6_addr *dst, struct in6_addr *src,
+			 int iif, int type, u32 pid, u32 seq,
+			 int prefix, unsigned int flags)
 {
 	struct rtmsg *rtm;
 	struct nlmsghdr  *nlh;
@@ -1585,10 +1585,6 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
 		}
 	}
 
-	if (!pid && in_nlh) {
-		pid = in_nlh->nlmsg_pid;
-	}
-
 	nlh = NLMSG_NEW(skb, pid, seq, type, sizeof(*rtm), flags);
 	rtm = NLMSG_DATA(nlh);
 	rtm->rtm_family = AF_INET6;
@@ -1675,7 +1671,7 @@ static int rt6_dump_route(struct rt6_info *rt, void *p_arg)
 
 	return rt6_fill_node(arg->skb, rt, NULL, NULL, 0, RTM_NEWROUTE,
 		     NETLINK_CB(arg->cb->skb).pid, arg->cb->nlh->nlmsg_seq,
-		     NULL, prefix, NLM_F_MULTI);
+		     prefix, NLM_F_MULTI);
 }
 
 static int fib6_dump_node(struct fib6_walker_t *w)
@@ -1823,7 +1819,7 @@ int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg)
 			    &fl.fl6_dst, &fl.fl6_src,
 			    iif,
 			    RTM_NEWROUTE, NETLINK_CB(in_skb).pid,
-			    nlh->nlmsg_seq, nlh, 0, 0);
+			    nlh->nlmsg_seq, 0, 0);
 	if (err < 0) {
 		err = -EMSGSIZE;
 		goto out_free;
@@ -1839,17 +1835,25 @@ out_free:
 	goto out;	
 }
 
-void inet6_rt_notify(int event, struct rt6_info *rt, struct nlmsghdr *nlh)
+void inet6_rt_notify(int event, struct rt6_info *rt, struct nlmsghdr *nlh, 
+			struct netlink_skb_parms *req)
 {
 	struct sk_buff *skb;
 	int size = NLMSG_SPACE(sizeof(struct rtmsg)+256);
+	u32 pid = current->pid;
+	u32 seq = 0;
 
+	if (req)
+		pid = req->pid;
+	if (nlh)
+		seq = nlh->nlmsg_seq;
+	
 	skb = alloc_skb(size, gfp_any());
 	if (!skb) {
 		netlink_set_err(rtnl, 0, RTMGRP_IPV6_ROUTE, ENOBUFS);
 		return;
 	}
-	if (rt6_fill_node(skb, rt, NULL, NULL, 0, event, 0, 0, nlh, 0, 0) < 0) {
+	if (rt6_fill_node(skb, rt, NULL, NULL, 0, event, pid, seq, 0, 0) < 0) {
 		kfree_skb(skb);
 		netlink_set_err(rtnl, 0, RTMGRP_IPV6_ROUTE, EINVAL);
 		return;
-- 
cgit v1.2.3-70-g09d2


From e45b1be8bcb3643808975a426fa3e201a2588e87 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 21 Jun 2005 14:01:30 -0700
Subject: [NETFILTER]: Kill lockhelp.h

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4/ip_conntrack_core.h |   3 +-
 include/linux/netfilter_ipv4/ip_nat.h            |   3 +-
 include/linux/netfilter_ipv4/listhelp.h          |   1 -
 include/linux/netfilter_ipv4/lockhelp.h          | 129 -----------------------
 net/ipv4/netfilter/arp_tables.c                  |   1 -
 net/ipv4/netfilter/ip_conntrack_amanda.c         |   7 +-
 net/ipv4/netfilter/ip_conntrack_core.c           |  84 +++++++--------
 net/ipv4/netfilter/ip_conntrack_ftp.c            |   7 +-
 net/ipv4/netfilter/ip_conntrack_irc.c            |   7 +-
 net/ipv4/netfilter/ip_conntrack_proto_sctp.c     |  23 ++--
 net/ipv4/netfilter/ip_conntrack_proto_tcp.c      |  27 +++--
 net/ipv4/netfilter/ip_conntrack_standalone.c     |  22 ++--
 net/ipv4/netfilter/ip_nat_core.c                 |  32 +++---
 net/ipv4/netfilter/ip_nat_helper.c               |  10 +-
 net/ipv4/netfilter/ip_nat_rule.c                 |   4 +-
 net/ipv4/netfilter/ip_nat_standalone.c           |   5 +-
 net/ipv4/netfilter/ip_tables.c                   |   1 -
 net/ipv4/netfilter/ipt_CLUSTERIP.c               |  49 ++++-----
 net/ipv4/netfilter/ipt_MASQUERADE.c              |  10 +-
 net/ipv4/netfilter/ipt_ULOG.c                    |  15 ++-
 net/ipv4/netfilter/ipt_hashlimit.c               |  17 ++-
 net/ipv4/netfilter/ipt_helper.c                  |   4 +-
 net/ipv6/netfilter/ip6_tables.c                  |   1 -
 23 files changed, 160 insertions(+), 302 deletions(-)
 delete mode 100644 include/linux/netfilter_ipv4/lockhelp.h

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4/ip_conntrack_core.h b/include/linux/netfilter_ipv4/ip_conntrack_core.h
index d84be02cb4f..694aec9b478 100644
--- a/include/linux/netfilter_ipv4/ip_conntrack_core.h
+++ b/include/linux/netfilter_ipv4/ip_conntrack_core.h
@@ -1,7 +1,6 @@
 #ifndef _IP_CONNTRACK_CORE_H
 #define _IP_CONNTRACK_CORE_H
 #include <linux/netfilter.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
 
 /* This header is used to share core functionality between the
    standalone connection tracking module, and the compatibility layer's use
@@ -47,6 +46,6 @@ static inline int ip_conntrack_confirm(struct sk_buff **pskb)
 
 extern struct list_head *ip_conntrack_hash;
 extern struct list_head ip_conntrack_expect_list;
-DECLARE_RWLOCK_EXTERN(ip_conntrack_lock);
+extern rwlock_t ip_conntrack_lock;
 #endif /* _IP_CONNTRACK_CORE_H */
 
diff --git a/include/linux/netfilter_ipv4/ip_nat.h b/include/linux/netfilter_ipv4/ip_nat.h
index 2b72b86176f..e201ec6e990 100644
--- a/include/linux/netfilter_ipv4/ip_nat.h
+++ b/include/linux/netfilter_ipv4/ip_nat.h
@@ -50,10 +50,9 @@ struct ip_nat_multi_range_compat
 
 #ifdef __KERNEL__
 #include <linux/list.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
 
 /* Protects NAT hash tables, and NAT-private part of conntracks. */
-DECLARE_RWLOCK_EXTERN(ip_nat_lock);
+extern rwlock_t ip_nat_lock;
 
 /* The structure embedded in the conntrack structure. */
 struct ip_nat_info
diff --git a/include/linux/netfilter_ipv4/listhelp.h b/include/linux/netfilter_ipv4/listhelp.h
index f2ae7c5e57b..360429f4873 100644
--- a/include/linux/netfilter_ipv4/listhelp.h
+++ b/include/linux/netfilter_ipv4/listhelp.h
@@ -2,7 +2,6 @@
 #define _LISTHELP_H
 #include <linux/config.h>
 #include <linux/list.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
 
 /* Header to do more comprehensive job than linux/list.h; assume list
    is first entry in structure. */
diff --git a/include/linux/netfilter_ipv4/lockhelp.h b/include/linux/netfilter_ipv4/lockhelp.h
deleted file mode 100644
index a3288633ab4..00000000000
--- a/include/linux/netfilter_ipv4/lockhelp.h
+++ /dev/null
@@ -1,129 +0,0 @@
-#ifndef _LOCKHELP_H
-#define _LOCKHELP_H
-#include <linux/config.h>
-
-#include <linux/spinlock.h>
-#include <asm/atomic.h>
-#include <linux/interrupt.h>
-#include <linux/smp.h>
-
-/* Header to do help in lock debugging. */
-
-#ifdef CONFIG_NETFILTER_DEBUG
-struct spinlock_debug
-{
-	spinlock_t l;
-	atomic_t locked_by;
-};
-
-struct rwlock_debug
-{
-	rwlock_t l;
-	long read_locked_map;
-	long write_locked_map;
-};
-
-#define DECLARE_LOCK(l) 						\
-struct spinlock_debug l = { SPIN_LOCK_UNLOCKED, ATOMIC_INIT(-1) }
-#define DECLARE_LOCK_EXTERN(l) 			\
-extern struct spinlock_debug l
-#define DECLARE_RWLOCK(l)				\
-struct rwlock_debug l = { RW_LOCK_UNLOCKED, 0, 0 }
-#define DECLARE_RWLOCK_EXTERN(l)		\
-extern struct rwlock_debug l
-
-#define MUST_BE_LOCKED(l)						\
-do { if (atomic_read(&(l)->locked_by) != smp_processor_id())		\
-	printk("ASSERT %s:%u %s unlocked\n", __FILE__, __LINE__, #l);	\
-} while(0)
-
-#define MUST_BE_UNLOCKED(l)						\
-do { if (atomic_read(&(l)->locked_by) == smp_processor_id())		\
-	printk("ASSERT %s:%u %s locked\n", __FILE__, __LINE__, #l);	\
-} while(0)
-
-/* Write locked OK as well. */
-#define MUST_BE_READ_LOCKED(l)						    \
-do { if (!((l)->read_locked_map & (1UL << smp_processor_id()))		    \
-	 && !((l)->write_locked_map & (1UL << smp_processor_id())))	    \
-	printk("ASSERT %s:%u %s not readlocked\n", __FILE__, __LINE__, #l); \
-} while(0)
-
-#define MUST_BE_WRITE_LOCKED(l)						     \
-do { if (!((l)->write_locked_map & (1UL << smp_processor_id())))	     \
-	printk("ASSERT %s:%u %s not writelocked\n", __FILE__, __LINE__, #l); \
-} while(0)
-
-#define MUST_BE_READ_WRITE_UNLOCKED(l)					  \
-do { if ((l)->read_locked_map & (1UL << smp_processor_id()))		  \
-	printk("ASSERT %s:%u %s readlocked\n", __FILE__, __LINE__, #l);	  \
- else if ((l)->write_locked_map & (1UL << smp_processor_id()))		  \
-	 printk("ASSERT %s:%u %s writelocked\n", __FILE__, __LINE__, #l); \
-} while(0)
-
-#define LOCK_BH(lk)						\
-do {								\
-	MUST_BE_UNLOCKED(lk);					\
-	spin_lock_bh(&(lk)->l);					\
-	atomic_set(&(lk)->locked_by, smp_processor_id());	\
-} while(0)
-
-#define UNLOCK_BH(lk)				\
-do {						\
-	MUST_BE_LOCKED(lk);			\
-	atomic_set(&(lk)->locked_by, -1);	\
-	spin_unlock_bh(&(lk)->l);		\
-} while(0)
-
-#define READ_LOCK(lk) 						\
-do {								\
-	MUST_BE_READ_WRITE_UNLOCKED(lk);			\
-	read_lock_bh(&(lk)->l);					\
-	set_bit(smp_processor_id(), &(lk)->read_locked_map);	\
-} while(0)
-
-#define WRITE_LOCK(lk)							  \
-do {									  \
-	MUST_BE_READ_WRITE_UNLOCKED(lk);				  \
-	write_lock_bh(&(lk)->l);					  \
-	set_bit(smp_processor_id(), &(lk)->write_locked_map);		  \
-} while(0)
-
-#define READ_UNLOCK(lk)							\
-do {									\
-	if (!((lk)->read_locked_map & (1UL << smp_processor_id())))	\
-		printk("ASSERT: %s:%u %s not readlocked\n", 		\
-		       __FILE__, __LINE__, #lk);			\
-	clear_bit(smp_processor_id(), &(lk)->read_locked_map);		\
-	read_unlock_bh(&(lk)->l);					\
-} while(0)
-
-#define WRITE_UNLOCK(lk)					\
-do {								\
-	MUST_BE_WRITE_LOCKED(lk);				\
-	clear_bit(smp_processor_id(), &(lk)->write_locked_map);	\
-	write_unlock_bh(&(lk)->l);				\
-} while(0)
-
-#else
-#define DECLARE_LOCK(l) spinlock_t l = SPIN_LOCK_UNLOCKED
-#define DECLARE_LOCK_EXTERN(l) extern spinlock_t l
-#define DECLARE_RWLOCK(l) rwlock_t l = RW_LOCK_UNLOCKED
-#define DECLARE_RWLOCK_EXTERN(l) extern rwlock_t l
-
-#define MUST_BE_LOCKED(l)
-#define MUST_BE_UNLOCKED(l)
-#define MUST_BE_READ_LOCKED(l)
-#define MUST_BE_WRITE_LOCKED(l)
-#define MUST_BE_READ_WRITE_UNLOCKED(l)
-
-#define LOCK_BH(l) spin_lock_bh(l)
-#define UNLOCK_BH(l) spin_unlock_bh(l)
-
-#define READ_LOCK(l) read_lock_bh(l)
-#define WRITE_LOCK(l) write_lock_bh(l)
-#define READ_UNLOCK(l) read_unlock_bh(l)
-#define WRITE_UNLOCK(l) write_unlock_bh(l)
-#endif /*CONFIG_NETFILTER_DEBUG*/
-
-#endif /* _LOCKHELP_H */
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index df79f5ed6a0..fa163425668 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -60,7 +60,6 @@ static DECLARE_MUTEX(arpt_mutex);
 
 #define ASSERT_READ_LOCK(x) ARP_NF_ASSERT(down_trylock(&arpt_mutex) != 0)
 #define ASSERT_WRITE_LOCK(x) ARP_NF_ASSERT(down_trylock(&arpt_mutex) != 0)
-#include <linux/netfilter_ipv4/lockhelp.h>
 #include <linux/netfilter_ipv4/listhelp.h>
 
 struct arpt_table_info {
diff --git a/net/ipv4/netfilter/ip_conntrack_amanda.c b/net/ipv4/netfilter/ip_conntrack_amanda.c
index 3dbddd06260..a78a320eee0 100644
--- a/net/ipv4/netfilter/ip_conntrack_amanda.c
+++ b/net/ipv4/netfilter/ip_conntrack_amanda.c
@@ -26,7 +26,6 @@
 #include <net/checksum.h>
 #include <net/udp.h>
 
-#include <linux/netfilter_ipv4/lockhelp.h>
 #include <linux/netfilter_ipv4/ip_conntrack_helper.h>
 #include <linux/netfilter_ipv4/ip_conntrack_amanda.h>
 
@@ -42,7 +41,7 @@ static char *conns[] = { "DATA ", "MESG ", "INDEX " };
 
 /* This is slow, but it's simple. --RR */
 static char amanda_buffer[65536];
-static DECLARE_LOCK(amanda_buffer_lock);
+static DEFINE_SPINLOCK(amanda_buffer_lock);
 
 unsigned int (*ip_nat_amanda_hook)(struct sk_buff **pskb,
 				   enum ip_conntrack_info ctinfo,
@@ -76,7 +75,7 @@ static int help(struct sk_buff **pskb,
 		return NF_ACCEPT;
 	}
 
-	LOCK_BH(&amanda_buffer_lock);
+	spin_lock_bh(&amanda_buffer_lock);
 	skb_copy_bits(*pskb, dataoff, amanda_buffer, (*pskb)->len - dataoff);
 	data = amanda_buffer;
 	data_limit = amanda_buffer + (*pskb)->len - dataoff;
@@ -134,7 +133,7 @@ static int help(struct sk_buff **pskb,
 	}
 
 out:
-	UNLOCK_BH(&amanda_buffer_lock);
+	spin_unlock_bh(&amanda_buffer_lock);
 	return ret;
 }
 
diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c
index 09e82462297..a7377a331ad 100644
--- a/net/ipv4/netfilter/ip_conntrack_core.c
+++ b/net/ipv4/netfilter/ip_conntrack_core.c
@@ -38,10 +38,10 @@
 #include <linux/percpu.h>
 #include <linux/moduleparam.h>
 
-/* This rwlock protects the main hash table, protocol/helper/expected
+/* ip_conntrack_lock protects the main hash table, protocol/helper/expected
    registrations, conntrack timers*/
-#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_conntrack_lock)
-#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_conntrack_lock)
+#define ASSERT_READ_LOCK(x)
+#define ASSERT_WRITE_LOCK(x)
 
 #include <linux/netfilter_ipv4/ip_conntrack.h>
 #include <linux/netfilter_ipv4/ip_conntrack_protocol.h>
@@ -57,7 +57,7 @@
 #define DEBUGP(format, args...)
 #endif
 
-DECLARE_RWLOCK(ip_conntrack_lock);
+DEFINE_RWLOCK(ip_conntrack_lock);
 
 /* ip_conntrack_standalone needs this */
 atomic_t ip_conntrack_count = ATOMIC_INIT(0);
@@ -147,7 +147,7 @@ static void destroy_expect(struct ip_conntrack_expect *exp)
 
 static void unlink_expect(struct ip_conntrack_expect *exp)
 {
-	MUST_BE_WRITE_LOCKED(&ip_conntrack_lock);
+	ASSERT_WRITE_LOCK(&ip_conntrack_lock);
 	list_del(&exp->list);
 	/* Logically in destroy_expect, but we hold the lock here. */
 	exp->master->expecting--;
@@ -157,9 +157,9 @@ static void expectation_timed_out(unsigned long ul_expect)
 {
 	struct ip_conntrack_expect *exp = (void *)ul_expect;
 
-	WRITE_LOCK(&ip_conntrack_lock);
+	write_lock_bh(&ip_conntrack_lock);
 	unlink_expect(exp);
-	WRITE_UNLOCK(&ip_conntrack_lock);
+	write_unlock_bh(&ip_conntrack_lock);
 	destroy_expect(exp);
 }
 
@@ -209,7 +209,7 @@ clean_from_lists(struct ip_conntrack *ct)
 	unsigned int ho, hr;
 	
 	DEBUGP("clean_from_lists(%p)\n", ct);
-	MUST_BE_WRITE_LOCKED(&ip_conntrack_lock);
+	ASSERT_WRITE_LOCK(&ip_conntrack_lock);
 
 	ho = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
 	hr = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple);
@@ -240,7 +240,7 @@ destroy_conntrack(struct nf_conntrack *nfct)
 	if (ip_conntrack_destroyed)
 		ip_conntrack_destroyed(ct);
 
-	WRITE_LOCK(&ip_conntrack_lock);
+	write_lock_bh(&ip_conntrack_lock);
 	/* Expectations will have been removed in clean_from_lists,
 	 * except TFTP can create an expectation on the first packet,
 	 * before connection is in the list, so we need to clean here,
@@ -254,7 +254,7 @@ destroy_conntrack(struct nf_conntrack *nfct)
 	}
 
 	CONNTRACK_STAT_INC(delete);
-	WRITE_UNLOCK(&ip_conntrack_lock);
+	write_unlock_bh(&ip_conntrack_lock);
 
 	if (ct->master)
 		ip_conntrack_put(ct->master);
@@ -268,12 +268,12 @@ static void death_by_timeout(unsigned long ul_conntrack)
 {
 	struct ip_conntrack *ct = (void *)ul_conntrack;
 
-	WRITE_LOCK(&ip_conntrack_lock);
+	write_lock_bh(&ip_conntrack_lock);
 	/* Inside lock so preempt is disabled on module removal path.
 	 * Otherwise we can get spurious warnings. */
 	CONNTRACK_STAT_INC(delete_list);
 	clean_from_lists(ct);
-	WRITE_UNLOCK(&ip_conntrack_lock);
+	write_unlock_bh(&ip_conntrack_lock);
 	ip_conntrack_put(ct);
 }
 
@@ -282,7 +282,7 @@ conntrack_tuple_cmp(const struct ip_conntrack_tuple_hash *i,
 		    const struct ip_conntrack_tuple *tuple,
 		    const struct ip_conntrack *ignored_conntrack)
 {
-	MUST_BE_READ_LOCKED(&ip_conntrack_lock);
+	ASSERT_READ_LOCK(&ip_conntrack_lock);
 	return tuplehash_to_ctrack(i) != ignored_conntrack
 		&& ip_ct_tuple_equal(tuple, &i->tuple);
 }
@@ -294,7 +294,7 @@ __ip_conntrack_find(const struct ip_conntrack_tuple *tuple,
 	struct ip_conntrack_tuple_hash *h;
 	unsigned int hash = hash_conntrack(tuple);
 
-	MUST_BE_READ_LOCKED(&ip_conntrack_lock);
+	ASSERT_READ_LOCK(&ip_conntrack_lock);
 	list_for_each_entry(h, &ip_conntrack_hash[hash], list) {
 		if (conntrack_tuple_cmp(h, tuple, ignored_conntrack)) {
 			CONNTRACK_STAT_INC(found);
@@ -313,11 +313,11 @@ ip_conntrack_find_get(const struct ip_conntrack_tuple *tuple,
 {
 	struct ip_conntrack_tuple_hash *h;
 
-	READ_LOCK(&ip_conntrack_lock);
+	read_lock_bh(&ip_conntrack_lock);
 	h = __ip_conntrack_find(tuple, ignored_conntrack);
 	if (h)
 		atomic_inc(&tuplehash_to_ctrack(h)->ct_general.use);
-	READ_UNLOCK(&ip_conntrack_lock);
+	read_unlock_bh(&ip_conntrack_lock);
 
 	return h;
 }
@@ -352,7 +352,7 @@ __ip_conntrack_confirm(struct sk_buff **pskb)
 	IP_NF_ASSERT(!is_confirmed(ct));
 	DEBUGP("Confirming conntrack %p\n", ct);
 
-	WRITE_LOCK(&ip_conntrack_lock);
+	write_lock_bh(&ip_conntrack_lock);
 
 	/* See if there's one in the list already, including reverse:
            NAT could have grabbed it without realizing, since we're
@@ -380,12 +380,12 @@ __ip_conntrack_confirm(struct sk_buff **pskb)
 		atomic_inc(&ct->ct_general.use);
 		set_bit(IPS_CONFIRMED_BIT, &ct->status);
 		CONNTRACK_STAT_INC(insert);
-		WRITE_UNLOCK(&ip_conntrack_lock);
+		write_unlock_bh(&ip_conntrack_lock);
 		return NF_ACCEPT;
 	}
 
 	CONNTRACK_STAT_INC(insert_failed);
-	WRITE_UNLOCK(&ip_conntrack_lock);
+	write_unlock_bh(&ip_conntrack_lock);
 
 	return NF_DROP;
 }
@@ -398,9 +398,9 @@ ip_conntrack_tuple_taken(const struct ip_conntrack_tuple *tuple,
 {
 	struct ip_conntrack_tuple_hash *h;
 
-	READ_LOCK(&ip_conntrack_lock);
+	read_lock_bh(&ip_conntrack_lock);
 	h = __ip_conntrack_find(tuple, ignored_conntrack);
-	READ_UNLOCK(&ip_conntrack_lock);
+	read_unlock_bh(&ip_conntrack_lock);
 
 	return h != NULL;
 }
@@ -419,13 +419,13 @@ static int early_drop(struct list_head *chain)
 	struct ip_conntrack *ct = NULL;
 	int dropped = 0;
 
-	READ_LOCK(&ip_conntrack_lock);
+	read_lock_bh(&ip_conntrack_lock);
 	h = LIST_FIND_B(chain, unreplied, struct ip_conntrack_tuple_hash *);
 	if (h) {
 		ct = tuplehash_to_ctrack(h);
 		atomic_inc(&ct->ct_general.use);
 	}
-	READ_UNLOCK(&ip_conntrack_lock);
+	read_unlock_bh(&ip_conntrack_lock);
 
 	if (!ct)
 		return dropped;
@@ -508,7 +508,7 @@ init_conntrack(const struct ip_conntrack_tuple *tuple,
 	conntrack->timeout.data = (unsigned long)conntrack;
 	conntrack->timeout.function = death_by_timeout;
 
-	WRITE_LOCK(&ip_conntrack_lock);
+	write_lock_bh(&ip_conntrack_lock);
 	exp = find_expectation(tuple);
 
 	if (exp) {
@@ -532,7 +532,7 @@ init_conntrack(const struct ip_conntrack_tuple *tuple,
 	list_add(&conntrack->tuplehash[IP_CT_DIR_ORIGINAL].list, &unconfirmed);
 
 	atomic_inc(&ip_conntrack_count);
-	WRITE_UNLOCK(&ip_conntrack_lock);
+	write_unlock_bh(&ip_conntrack_lock);
 
 	if (exp) {
 		if (exp->expectfn)
@@ -723,17 +723,17 @@ void ip_conntrack_unexpect_related(struct ip_conntrack_expect *exp)
 {
 	struct ip_conntrack_expect *i;
 
-	WRITE_LOCK(&ip_conntrack_lock);
+	write_lock_bh(&ip_conntrack_lock);
 	/* choose the the oldest expectation to evict */
 	list_for_each_entry_reverse(i, &ip_conntrack_expect_list, list) {
 		if (expect_matches(i, exp) && del_timer(&i->timeout)) {
 			unlink_expect(i);
-			WRITE_UNLOCK(&ip_conntrack_lock);
+			write_unlock_bh(&ip_conntrack_lock);
 			destroy_expect(i);
 			return;
 		}
 	}
-	WRITE_UNLOCK(&ip_conntrack_lock);
+	write_unlock_bh(&ip_conntrack_lock);
 }
 
 struct ip_conntrack_expect *ip_conntrack_expect_alloc(void)
@@ -808,7 +808,7 @@ int ip_conntrack_expect_related(struct ip_conntrack_expect *expect)
 	DEBUGP("tuple: "); DUMP_TUPLE(&expect->tuple);
 	DEBUGP("mask:  "); DUMP_TUPLE(&expect->mask);
 
-	WRITE_LOCK(&ip_conntrack_lock);
+	write_lock_bh(&ip_conntrack_lock);
 	list_for_each_entry(i, &ip_conntrack_expect_list, list) {
 		if (expect_matches(i, expect)) {
 			/* Refresh timer: if it's dying, ignore.. */
@@ -832,7 +832,7 @@ int ip_conntrack_expect_related(struct ip_conntrack_expect *expect)
 	ip_conntrack_expect_insert(expect);
 	ret = 0;
 out:
-	WRITE_UNLOCK(&ip_conntrack_lock);
+	write_unlock_bh(&ip_conntrack_lock);
  	return ret;
 }
 
@@ -841,7 +841,7 @@ out:
 void ip_conntrack_alter_reply(struct ip_conntrack *conntrack,
 			      const struct ip_conntrack_tuple *newreply)
 {
-	WRITE_LOCK(&ip_conntrack_lock);
+	write_lock_bh(&ip_conntrack_lock);
 	/* Should be unconfirmed, so not in hash table yet */
 	IP_NF_ASSERT(!is_confirmed(conntrack));
 
@@ -851,15 +851,15 @@ void ip_conntrack_alter_reply(struct ip_conntrack *conntrack,
 	conntrack->tuplehash[IP_CT_DIR_REPLY].tuple = *newreply;
 	if (!conntrack->master && conntrack->expecting == 0)
 		conntrack->helper = ip_ct_find_helper(newreply);
-	WRITE_UNLOCK(&ip_conntrack_lock);
+	write_unlock_bh(&ip_conntrack_lock);
 }
 
 int ip_conntrack_helper_register(struct ip_conntrack_helper *me)
 {
 	BUG_ON(me->timeout == 0);
-	WRITE_LOCK(&ip_conntrack_lock);
+	write_lock_bh(&ip_conntrack_lock);
 	list_prepend(&helpers, me);
-	WRITE_UNLOCK(&ip_conntrack_lock);
+	write_unlock_bh(&ip_conntrack_lock);
 
 	return 0;
 }
@@ -878,7 +878,7 @@ void ip_conntrack_helper_unregister(struct ip_conntrack_helper *me)
 	struct ip_conntrack_expect *exp, *tmp;
 
 	/* Need write lock here, to delete helper. */
-	WRITE_LOCK(&ip_conntrack_lock);
+	write_lock_bh(&ip_conntrack_lock);
 	LIST_DELETE(&helpers, me);
 
 	/* Get rid of expectations */
@@ -893,7 +893,7 @@ void ip_conntrack_helper_unregister(struct ip_conntrack_helper *me)
 	for (i = 0; i < ip_conntrack_htable_size; i++)
 		LIST_FIND_W(&ip_conntrack_hash[i], unhelp,
 			    struct ip_conntrack_tuple_hash *, me);
-	WRITE_UNLOCK(&ip_conntrack_lock);
+	write_unlock_bh(&ip_conntrack_lock);
 
 	/* Someone could be still looking at the helper in a bh. */
 	synchronize_net();
@@ -925,14 +925,14 @@ void ip_ct_refresh_acct(struct ip_conntrack *ct,
 		ct->timeout.expires = extra_jiffies;
 		ct_add_counters(ct, ctinfo, skb);
 	} else {
-		WRITE_LOCK(&ip_conntrack_lock);
+		write_lock_bh(&ip_conntrack_lock);
 		/* Need del_timer for race avoidance (may already be dying). */
 		if (del_timer(&ct->timeout)) {
 			ct->timeout.expires = jiffies + extra_jiffies;
 			add_timer(&ct->timeout);
 		}
 		ct_add_counters(ct, ctinfo, skb);
-		WRITE_UNLOCK(&ip_conntrack_lock);
+		write_unlock_bh(&ip_conntrack_lock);
 	}
 }
 
@@ -997,7 +997,7 @@ get_next_corpse(int (*iter)(struct ip_conntrack *i, void *data),
 {
 	struct ip_conntrack_tuple_hash *h = NULL;
 
-	WRITE_LOCK(&ip_conntrack_lock);
+	write_lock_bh(&ip_conntrack_lock);
 	for (; *bucket < ip_conntrack_htable_size; (*bucket)++) {
 		h = LIST_FIND_W(&ip_conntrack_hash[*bucket], do_iter,
 				struct ip_conntrack_tuple_hash *, iter, data);
@@ -1009,7 +1009,7 @@ get_next_corpse(int (*iter)(struct ip_conntrack *i, void *data),
 				struct ip_conntrack_tuple_hash *, iter, data);
 	if (h)
 		atomic_inc(&tuplehash_to_ctrack(h)->ct_general.use);
-	WRITE_UNLOCK(&ip_conntrack_lock);
+	write_unlock_bh(&ip_conntrack_lock);
 
 	return h;
 }
@@ -1201,14 +1201,14 @@ int __init ip_conntrack_init(void)
 	}
 
 	/* Don't NEED lock here, but good form anyway. */
-	WRITE_LOCK(&ip_conntrack_lock);
+	write_lock_bh(&ip_conntrack_lock);
 	for (i = 0; i < MAX_IP_CT_PROTO; i++)
 		ip_ct_protos[i] = &ip_conntrack_generic_protocol;
 	/* Sew in builtin protocols. */
 	ip_ct_protos[IPPROTO_TCP] = &ip_conntrack_protocol_tcp;
 	ip_ct_protos[IPPROTO_UDP] = &ip_conntrack_protocol_udp;
 	ip_ct_protos[IPPROTO_ICMP] = &ip_conntrack_protocol_icmp;
-	WRITE_UNLOCK(&ip_conntrack_lock);
+	write_unlock_bh(&ip_conntrack_lock);
 
 	for (i = 0; i < ip_conntrack_htable_size; i++)
 		INIT_LIST_HEAD(&ip_conntrack_hash[i]);
diff --git a/net/ipv4/netfilter/ip_conntrack_ftp.c b/net/ipv4/netfilter/ip_conntrack_ftp.c
index dd86503aa78..fea6dd2a00b 100644
--- a/net/ipv4/netfilter/ip_conntrack_ftp.c
+++ b/net/ipv4/netfilter/ip_conntrack_ftp.c
@@ -16,7 +16,6 @@
 #include <net/checksum.h>
 #include <net/tcp.h>
 
-#include <linux/netfilter_ipv4/lockhelp.h>
 #include <linux/netfilter_ipv4/ip_conntrack_helper.h>
 #include <linux/netfilter_ipv4/ip_conntrack_ftp.h>
 #include <linux/moduleparam.h>
@@ -28,7 +27,7 @@ MODULE_DESCRIPTION("ftp connection tracking helper");
 /* This is slow, but it's simple. --RR */
 static char ftp_buffer[65536];
 
-static DECLARE_LOCK(ip_ftp_lock);
+static DEFINE_SPINLOCK(ip_ftp_lock);
 
 #define MAX_PORTS 8
 static int ports[MAX_PORTS];
@@ -319,7 +318,7 @@ static int help(struct sk_buff **pskb,
 	}
 	datalen = (*pskb)->len - dataoff;
 
-	LOCK_BH(&ip_ftp_lock);
+	spin_lock_bh(&ip_ftp_lock);
 	fb_ptr = skb_header_pointer(*pskb, dataoff,
 				    (*pskb)->len - dataoff, ftp_buffer);
 	BUG_ON(fb_ptr == NULL);
@@ -442,7 +441,7 @@ out_update_nl:
 	if (ends_in_nl)
 		update_nl_seq(seq, ct_ftp_info,dir);
  out:
-	UNLOCK_BH(&ip_ftp_lock);
+	spin_unlock_bh(&ip_ftp_lock);
 	return ret;
 }
 
diff --git a/net/ipv4/netfilter/ip_conntrack_irc.c b/net/ipv4/netfilter/ip_conntrack_irc.c
index 33cc7348b6e..cd98772cc33 100644
--- a/net/ipv4/netfilter/ip_conntrack_irc.c
+++ b/net/ipv4/netfilter/ip_conntrack_irc.c
@@ -29,7 +29,6 @@
 #include <net/checksum.h>
 #include <net/tcp.h>
 
-#include <linux/netfilter_ipv4/lockhelp.h>
 #include <linux/netfilter_ipv4/ip_conntrack_helper.h>
 #include <linux/netfilter_ipv4/ip_conntrack_irc.h>
 #include <linux/moduleparam.h>
@@ -41,7 +40,7 @@ static int max_dcc_channels = 8;
 static unsigned int dcc_timeout = 300;
 /* This is slow, but it's simple. --RR */
 static char irc_buffer[65536];
-static DECLARE_LOCK(irc_buffer_lock);
+static DEFINE_SPINLOCK(irc_buffer_lock);
 
 unsigned int (*ip_nat_irc_hook)(struct sk_buff **pskb,
 				enum ip_conntrack_info ctinfo,
@@ -141,7 +140,7 @@ static int help(struct sk_buff **pskb,
 	if (dataoff >= (*pskb)->len)
 		return NF_ACCEPT;
 
-	LOCK_BH(&irc_buffer_lock);
+	spin_lock_bh(&irc_buffer_lock);
 	ib_ptr = skb_header_pointer(*pskb, dataoff,
 				    (*pskb)->len - dataoff, irc_buffer);
 	BUG_ON(ib_ptr == NULL);
@@ -237,7 +236,7 @@ static int help(struct sk_buff **pskb,
 	} /* while data < ... */
 
  out:
-	UNLOCK_BH(&irc_buffer_lock);
+	spin_unlock_bh(&irc_buffer_lock);
 	return ret;
 }
 
diff --git a/net/ipv4/netfilter/ip_conntrack_proto_sctp.c b/net/ipv4/netfilter/ip_conntrack_proto_sctp.c
index ff8c34a860f..31d75390bf1 100644
--- a/net/ipv4/netfilter/ip_conntrack_proto_sctp.c
+++ b/net/ipv4/netfilter/ip_conntrack_proto_sctp.c
@@ -26,7 +26,6 @@
 
 #include <linux/netfilter_ipv4/ip_conntrack.h>
 #include <linux/netfilter_ipv4/ip_conntrack_protocol.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
 
 #if 0
 #define DEBUGP(format, ...) printk(format, ## __VA_ARGS__)
@@ -35,7 +34,7 @@
 #endif
 
 /* Protects conntrack->proto.sctp */
-static DECLARE_RWLOCK(sctp_lock);
+static DEFINE_RWLOCK(sctp_lock);
 
 /* FIXME: Examine ipfilter's timeouts and conntrack transitions more
    closely.  They're more complex. --RR 
@@ -199,9 +198,9 @@ static int sctp_print_conntrack(struct seq_file *s,
 	DEBUGP(__FUNCTION__);
 	DEBUGP("\n");
 
-	READ_LOCK(&sctp_lock);
+	read_lock_bh(&sctp_lock);
 	state = conntrack->proto.sctp.state;
-	READ_UNLOCK(&sctp_lock);
+	read_unlock_bh(&sctp_lock);
 
 	return seq_printf(s, "%s ", sctp_conntrack_names[state]);
 }
@@ -343,13 +342,13 @@ static int sctp_packet(struct ip_conntrack *conntrack,
 
 	oldsctpstate = newconntrack = SCTP_CONNTRACK_MAX;
 	for_each_sctp_chunk (skb, sch, _sch, offset, count) {
-		WRITE_LOCK(&sctp_lock);
+		write_lock_bh(&sctp_lock);
 
 		/* Special cases of Verification tag check (Sec 8.5.1) */
 		if (sch->type == SCTP_CID_INIT) {
 			/* Sec 8.5.1 (A) */
 			if (sh->vtag != 0) {
-				WRITE_UNLOCK(&sctp_lock);
+				write_unlock_bh(&sctp_lock);
 				return -1;
 			}
 		} else if (sch->type == SCTP_CID_ABORT) {
@@ -357,7 +356,7 @@ static int sctp_packet(struct ip_conntrack *conntrack,
 			if (!(sh->vtag == conntrack->proto.sctp.vtag[CTINFO2DIR(ctinfo)])
 				&& !(sh->vtag == conntrack->proto.sctp.vtag
 							[1 - CTINFO2DIR(ctinfo)])) {
-				WRITE_UNLOCK(&sctp_lock);
+				write_unlock_bh(&sctp_lock);
 				return -1;
 			}
 		} else if (sch->type == SCTP_CID_SHUTDOWN_COMPLETE) {
@@ -366,13 +365,13 @@ static int sctp_packet(struct ip_conntrack *conntrack,
 				&& !(sh->vtag == conntrack->proto.sctp.vtag
 							[1 - CTINFO2DIR(ctinfo)] 
 					&& (sch->flags & 1))) {
-				WRITE_UNLOCK(&sctp_lock);
+				write_unlock_bh(&sctp_lock);
 				return -1;
 			}
 		} else if (sch->type == SCTP_CID_COOKIE_ECHO) {
 			/* Sec 8.5.1 (D) */
 			if (!(sh->vtag == conntrack->proto.sctp.vtag[CTINFO2DIR(ctinfo)])) {
-				WRITE_UNLOCK(&sctp_lock);
+				write_unlock_bh(&sctp_lock);
 				return -1;
 			}
 		}
@@ -384,7 +383,7 @@ static int sctp_packet(struct ip_conntrack *conntrack,
 		if (newconntrack == SCTP_CONNTRACK_MAX) {
 			DEBUGP("ip_conntrack_sctp: Invalid dir=%i ctype=%u conntrack=%u\n",
 			       CTINFO2DIR(ctinfo), sch->type, oldsctpstate);
-			WRITE_UNLOCK(&sctp_lock);
+			write_unlock_bh(&sctp_lock);
 			return -1;
 		}
 
@@ -396,7 +395,7 @@ static int sctp_packet(struct ip_conntrack *conntrack,
 			ih = skb_header_pointer(skb, offset + sizeof(sctp_chunkhdr_t),
 			                        sizeof(_inithdr), &_inithdr);
 			if (ih == NULL) {
-					WRITE_UNLOCK(&sctp_lock);
+					write_unlock_bh(&sctp_lock);
 					return -1;
 			}
 			DEBUGP("Setting vtag %x for dir %d\n", 
@@ -405,7 +404,7 @@ static int sctp_packet(struct ip_conntrack *conntrack,
 		}
 
 		conntrack->proto.sctp.state = newconntrack;
-		WRITE_UNLOCK(&sctp_lock);
+		write_unlock_bh(&sctp_lock);
 	}
 
 	ip_ct_refresh_acct(conntrack, ctinfo, skb, *sctp_timeouts[newconntrack]);
diff --git a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c
index 721ddbf522b..809dfed766d 100644
--- a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c
+++ b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c
@@ -36,7 +36,6 @@
 #include <linux/netfilter_ipv4.h>
 #include <linux/netfilter_ipv4/ip_conntrack.h>
 #include <linux/netfilter_ipv4/ip_conntrack_protocol.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
 
 #if 0
 #define DEBUGP printk
@@ -46,7 +45,7 @@
 #endif
 
 /* Protects conntrack->proto.tcp */
-static DECLARE_RWLOCK(tcp_lock);
+static DEFINE_RWLOCK(tcp_lock);
 
 /* "Be conservative in what you do, 
     be liberal in what you accept from others." 
@@ -330,9 +329,9 @@ static int tcp_print_conntrack(struct seq_file *s,
 {
 	enum tcp_conntrack state;
 
-	READ_LOCK(&tcp_lock);
+	read_lock_bh(&tcp_lock);
 	state = conntrack->proto.tcp.state;
-	READ_UNLOCK(&tcp_lock);
+	read_unlock_bh(&tcp_lock);
 
 	return seq_printf(s, "%s ", tcp_conntrack_names[state]);
 }
@@ -738,14 +737,14 @@ void ip_conntrack_tcp_update(struct sk_buff *skb,
 
 	end = segment_seq_plus_len(ntohl(tcph->seq), skb->len, iph, tcph);
 	
-	WRITE_LOCK(&tcp_lock);
+	write_lock_bh(&tcp_lock);
 	/*
 	 * We have to worry for the ack in the reply packet only...
 	 */
 	if (after(end, conntrack->proto.tcp.seen[dir].td_end))
 		conntrack->proto.tcp.seen[dir].td_end = end;
 	conntrack->proto.tcp.last_end = end;
-	WRITE_UNLOCK(&tcp_lock);
+	write_unlock_bh(&tcp_lock);
 	DEBUGP("tcp_update: sender end=%u maxend=%u maxwin=%u scale=%i "
 	       "receiver end=%u maxend=%u maxwin=%u scale=%i\n",
 		sender->td_end, sender->td_maxend, sender->td_maxwin,
@@ -857,7 +856,7 @@ static int tcp_packet(struct ip_conntrack *conntrack,
 				sizeof(_tcph), &_tcph);
 	BUG_ON(th == NULL);
 	
-	WRITE_LOCK(&tcp_lock);
+	write_lock_bh(&tcp_lock);
 	old_state = conntrack->proto.tcp.state;
 	dir = CTINFO2DIR(ctinfo);
 	index = get_conntrack_index(th);
@@ -879,7 +878,7 @@ static int tcp_packet(struct ip_conntrack *conntrack,
 			 * that the client cannot but retransmit its SYN and 
 			 * thus initiate a clean new session.
 			 */
-		    	WRITE_UNLOCK(&tcp_lock);
+		    	write_unlock_bh(&tcp_lock);
 			if (LOG_INVALID(IPPROTO_TCP))
 				nf_log_packet(PF_INET, 0, skb, NULL, NULL, 
 					  "ip_ct_tcp: killing out of sync session ");
@@ -894,7 +893,7 @@ static int tcp_packet(struct ip_conntrack *conntrack,
 		conntrack->proto.tcp.last_end = 
 		    segment_seq_plus_len(ntohl(th->seq), skb->len, iph, th);
 		
-		WRITE_UNLOCK(&tcp_lock);
+		write_unlock_bh(&tcp_lock);
 		if (LOG_INVALID(IPPROTO_TCP))
 			nf_log_packet(PF_INET, 0, skb, NULL, NULL, 
 				  "ip_ct_tcp: invalid packet ignored ");
@@ -904,7 +903,7 @@ static int tcp_packet(struct ip_conntrack *conntrack,
 		DEBUGP("ip_ct_tcp: Invalid dir=%i index=%u ostate=%u\n",
 		       dir, get_conntrack_index(th),
 		       old_state);
-		WRITE_UNLOCK(&tcp_lock);
+		write_unlock_bh(&tcp_lock);
 		if (LOG_INVALID(IPPROTO_TCP))
 			nf_log_packet(PF_INET, 0, skb, NULL, NULL, 
 				  "ip_ct_tcp: invalid state ");
@@ -918,13 +917,13 @@ static int tcp_packet(struct ip_conntrack *conntrack,
 		    	     conntrack->proto.tcp.seen[dir].td_end)) {	
 		    	/* Attempt to reopen a closed connection.
 		    	* Delete this connection and look up again. */
-		    	WRITE_UNLOCK(&tcp_lock);
+		    	write_unlock_bh(&tcp_lock);
 		    	if (del_timer(&conntrack->timeout))
 		    		conntrack->timeout.function((unsigned long)
 		    					    conntrack);
 		    	return -NF_REPEAT;
 		} else {
-			WRITE_UNLOCK(&tcp_lock);
+			write_unlock_bh(&tcp_lock);
 			if (LOG_INVALID(IPPROTO_TCP))
 				nf_log_packet(PF_INET, 0, skb, NULL, NULL,
 				              "ip_ct_tcp: invalid SYN");
@@ -949,7 +948,7 @@ static int tcp_packet(struct ip_conntrack *conntrack,
 
 	if (!tcp_in_window(&conntrack->proto.tcp, dir, index, 
 			   skb, iph, th)) {
-		WRITE_UNLOCK(&tcp_lock);
+		write_unlock_bh(&tcp_lock);
 		return -NF_ACCEPT;
 	}
     in_window:
@@ -972,7 +971,7 @@ static int tcp_packet(struct ip_conntrack *conntrack,
 	timeout = conntrack->proto.tcp.retrans >= ip_ct_tcp_max_retrans
 		  && *tcp_timeouts[new_state] > ip_ct_tcp_timeout_max_retrans
 		  ? ip_ct_tcp_timeout_max_retrans : *tcp_timeouts[new_state];
-	WRITE_UNLOCK(&tcp_lock);
+	write_unlock_bh(&tcp_lock);
 
 	if (!test_bit(IPS_SEEN_REPLY_BIT, &conntrack->status)) {
 		/* If only reply is a RST, we can consider ourselves not to
diff --git a/net/ipv4/netfilter/ip_conntrack_standalone.c b/net/ipv4/netfilter/ip_conntrack_standalone.c
index bc59f7b3980..42dc9510287 100644
--- a/net/ipv4/netfilter/ip_conntrack_standalone.c
+++ b/net/ipv4/netfilter/ip_conntrack_standalone.c
@@ -28,8 +28,8 @@
 #include <net/checksum.h>
 #include <net/ip.h>
 
-#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_conntrack_lock)
-#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_conntrack_lock)
+#define ASSERT_READ_LOCK(x)
+#define ASSERT_WRITE_LOCK(x)
 
 #include <linux/netfilter_ipv4/ip_conntrack.h>
 #include <linux/netfilter_ipv4/ip_conntrack_protocol.h>
@@ -119,7 +119,7 @@ static struct list_head *ct_get_idx(struct seq_file *seq, loff_t pos)
 
 static void *ct_seq_start(struct seq_file *seq, loff_t *pos)
 {
-	READ_LOCK(&ip_conntrack_lock);
+	read_lock_bh(&ip_conntrack_lock);
 	return ct_get_idx(seq, *pos);
 }
 
@@ -131,7 +131,7 @@ static void *ct_seq_next(struct seq_file *s, void *v, loff_t *pos)
   
 static void ct_seq_stop(struct seq_file *s, void *v)
 {
-	READ_UNLOCK(&ip_conntrack_lock);
+	read_unlock_bh(&ip_conntrack_lock);
 }
  
 static int ct_seq_show(struct seq_file *s, void *v)
@@ -140,7 +140,7 @@ static int ct_seq_show(struct seq_file *s, void *v)
 	const struct ip_conntrack *conntrack = tuplehash_to_ctrack(hash);
 	struct ip_conntrack_protocol *proto;
 
-	MUST_BE_READ_LOCKED(&ip_conntrack_lock);
+	ASSERT_READ_LOCK(&ip_conntrack_lock);
 	IP_NF_ASSERT(conntrack);
 
 	/* we only want to print DIR_ORIGINAL */
@@ -239,7 +239,7 @@ static void *exp_seq_start(struct seq_file *s, loff_t *pos)
 
 	/* strange seq_file api calls stop even if we fail,
 	 * thus we need to grab lock since stop unlocks */
-	READ_LOCK(&ip_conntrack_lock);
+	read_lock_bh(&ip_conntrack_lock);
 
 	if (list_empty(e))
 		return NULL;
@@ -267,7 +267,7 @@ static void *exp_seq_next(struct seq_file *s, void *v, loff_t *pos)
 
 static void exp_seq_stop(struct seq_file *s, void *v)
 {
-	READ_UNLOCK(&ip_conntrack_lock);
+	read_unlock_bh(&ip_conntrack_lock);
 }
 
 static int exp_seq_show(struct seq_file *s, void *v)
@@ -921,22 +921,22 @@ int ip_conntrack_protocol_register(struct ip_conntrack_protocol *proto)
 {
 	int ret = 0;
 
-	WRITE_LOCK(&ip_conntrack_lock);
+	write_lock_bh(&ip_conntrack_lock);
 	if (ip_ct_protos[proto->proto] != &ip_conntrack_generic_protocol) {
 		ret = -EBUSY;
 		goto out;
 	}
 	ip_ct_protos[proto->proto] = proto;
  out:
-	WRITE_UNLOCK(&ip_conntrack_lock);
+	write_unlock_bh(&ip_conntrack_lock);
 	return ret;
 }
 
 void ip_conntrack_protocol_unregister(struct ip_conntrack_protocol *proto)
 {
-	WRITE_LOCK(&ip_conntrack_lock);
+	write_lock_bh(&ip_conntrack_lock);
 	ip_ct_protos[proto->proto] = &ip_conntrack_generic_protocol;
-	WRITE_UNLOCK(&ip_conntrack_lock);
+	write_unlock_bh(&ip_conntrack_lock);
 	
 	/* Somebody could be still looking at the proto in bh. */
 	synchronize_net();
diff --git a/net/ipv4/netfilter/ip_nat_core.c b/net/ipv4/netfilter/ip_nat_core.c
index 9fc6f93af0d..739b6dde1c8 100644
--- a/net/ipv4/netfilter/ip_nat_core.c
+++ b/net/ipv4/netfilter/ip_nat_core.c
@@ -22,8 +22,8 @@
 #include <linux/udp.h>
 #include <linux/jhash.h>
 
-#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_nat_lock)
-#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_nat_lock)
+#define ASSERT_READ_LOCK(x)
+#define ASSERT_WRITE_LOCK(x)
 
 #include <linux/netfilter_ipv4/ip_conntrack.h>
 #include <linux/netfilter_ipv4/ip_conntrack_core.h>
@@ -41,7 +41,7 @@
 #define DEBUGP(format, args...)
 #endif
 
-DECLARE_RWLOCK(ip_nat_lock);
+DEFINE_RWLOCK(ip_nat_lock);
 
 /* Calculated at init based on memory size */
 static unsigned int ip_nat_htable_size;
@@ -65,9 +65,9 @@ static void ip_nat_cleanup_conntrack(struct ip_conntrack *conn)
 	if (!(conn->status & IPS_NAT_DONE_MASK))
 		return;
 
-	WRITE_LOCK(&ip_nat_lock);
+	write_lock_bh(&ip_nat_lock);
 	list_del(&conn->nat.info.bysource);
-	WRITE_UNLOCK(&ip_nat_lock);
+	write_unlock_bh(&ip_nat_lock);
 }
 
 /* We do checksum mangling, so if they were wrong before they're still
@@ -142,7 +142,7 @@ find_appropriate_src(const struct ip_conntrack_tuple *tuple,
 	unsigned int h = hash_by_src(tuple);
 	struct ip_conntrack *ct;
 
-	READ_LOCK(&ip_nat_lock);
+	read_lock_bh(&ip_nat_lock);
 	list_for_each_entry(ct, &bysource[h], nat.info.bysource) {
 		if (same_src(ct, tuple)) {
 			/* Copy source part from reply tuple. */
@@ -151,12 +151,12 @@ find_appropriate_src(const struct ip_conntrack_tuple *tuple,
 			result->dst = tuple->dst;
 
 			if (in_range(result, range)) {
-				READ_UNLOCK(&ip_nat_lock);
+				read_unlock_bh(&ip_nat_lock);
 				return 1;
 			}
 		}
 	}
-	READ_UNLOCK(&ip_nat_lock);
+	read_unlock_bh(&ip_nat_lock);
 	return 0;
 }
 
@@ -297,9 +297,9 @@ ip_nat_setup_info(struct ip_conntrack *conntrack,
 		unsigned int srchash
 			= hash_by_src(&conntrack->tuplehash[IP_CT_DIR_ORIGINAL]
 				      .tuple);
-		WRITE_LOCK(&ip_nat_lock);
+		write_lock_bh(&ip_nat_lock);
 		list_add(&info->bysource, &bysource[srchash]);
-		WRITE_UNLOCK(&ip_nat_lock);
+		write_unlock_bh(&ip_nat_lock);
 	}
 
 	/* It's done. */
@@ -474,23 +474,23 @@ int ip_nat_protocol_register(struct ip_nat_protocol *proto)
 {
 	int ret = 0;
 
-	WRITE_LOCK(&ip_nat_lock);
+	write_lock_bh(&ip_nat_lock);
 	if (ip_nat_protos[proto->protonum] != &ip_nat_unknown_protocol) {
 		ret = -EBUSY;
 		goto out;
 	}
 	ip_nat_protos[proto->protonum] = proto;
  out:
-	WRITE_UNLOCK(&ip_nat_lock);
+	write_unlock_bh(&ip_nat_lock);
 	return ret;
 }
 
 /* Noone stores the protocol anywhere; simply delete it. */
 void ip_nat_protocol_unregister(struct ip_nat_protocol *proto)
 {
-	WRITE_LOCK(&ip_nat_lock);
+	write_lock_bh(&ip_nat_lock);
 	ip_nat_protos[proto->protonum] = &ip_nat_unknown_protocol;
-	WRITE_UNLOCK(&ip_nat_lock);
+	write_unlock_bh(&ip_nat_lock);
 
 	/* Someone could be still looking at the proto in a bh. */
 	synchronize_net();
@@ -509,13 +509,13 @@ int __init ip_nat_init(void)
 		return -ENOMEM;
 
 	/* Sew in builtin protocols. */
-	WRITE_LOCK(&ip_nat_lock);
+	write_lock_bh(&ip_nat_lock);
 	for (i = 0; i < MAX_IP_NAT_PROTO; i++)
 		ip_nat_protos[i] = &ip_nat_unknown_protocol;
 	ip_nat_protos[IPPROTO_TCP] = &ip_nat_protocol_tcp;
 	ip_nat_protos[IPPROTO_UDP] = &ip_nat_protocol_udp;
 	ip_nat_protos[IPPROTO_ICMP] = &ip_nat_protocol_icmp;
-	WRITE_UNLOCK(&ip_nat_lock);
+	write_unlock_bh(&ip_nat_lock);
 
 	for (i = 0; i < ip_nat_htable_size; i++) {
 		INIT_LIST_HEAD(&bysource[i]);
diff --git a/net/ipv4/netfilter/ip_nat_helper.c b/net/ipv4/netfilter/ip_nat_helper.c
index 1637b96d8c0..9cd51f180dc 100644
--- a/net/ipv4/netfilter/ip_nat_helper.c
+++ b/net/ipv4/netfilter/ip_nat_helper.c
@@ -28,8 +28,8 @@
 #include <net/tcp.h>
 #include <net/udp.h>
 
-#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_nat_lock)
-#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_nat_lock)
+#define ASSERT_READ_LOCK(x)
+#define ASSERT_WRITE_LOCK(x)
 
 #include <linux/netfilter_ipv4/ip_conntrack.h>
 #include <linux/netfilter_ipv4/ip_conntrack_helper.h>
@@ -47,7 +47,7 @@
 #define DUMP_OFFSET(x)
 #endif
 
-static DECLARE_LOCK(ip_nat_seqofs_lock);
+static DEFINE_SPINLOCK(ip_nat_seqofs_lock);
 
 /* Setup TCP sequence correction given this change at this sequence */
 static inline void 
@@ -70,7 +70,7 @@ adjust_tcp_sequence(u32 seq,
 	DEBUGP("ip_nat_resize_packet: Seq_offset before: ");
 	DUMP_OFFSET(this_way);
 
-	LOCK_BH(&ip_nat_seqofs_lock);
+	spin_lock_bh(&ip_nat_seqofs_lock);
 
 	/* SYN adjust. If it's uninitialized, or this is after last
 	 * correction, record it: we don't handle more than one
@@ -82,7 +82,7 @@ adjust_tcp_sequence(u32 seq,
 		    this_way->offset_before = this_way->offset_after;
 		    this_way->offset_after += sizediff;
 	}
-	UNLOCK_BH(&ip_nat_seqofs_lock);
+	spin_unlock_bh(&ip_nat_seqofs_lock);
 
 	DEBUGP("ip_nat_resize_packet: Seq_offset after: ");
 	DUMP_OFFSET(this_way);
diff --git a/net/ipv4/netfilter/ip_nat_rule.c b/net/ipv4/netfilter/ip_nat_rule.c
index 581f097f5a2..60d70fa41a1 100644
--- a/net/ipv4/netfilter/ip_nat_rule.c
+++ b/net/ipv4/netfilter/ip_nat_rule.c
@@ -19,8 +19,8 @@
 #include <net/route.h>
 #include <linux/bitops.h>
 
-#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_nat_lock)
-#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_nat_lock)
+#define ASSERT_READ_LOCK(x)
+#define ASSERT_WRITE_LOCK(x)
 
 #include <linux/netfilter_ipv4/ip_tables.h>
 #include <linux/netfilter_ipv4/ip_nat.h>
diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c
index 79f56f662b3..bc59d0d6e89 100644
--- a/net/ipv4/netfilter/ip_nat_standalone.c
+++ b/net/ipv4/netfilter/ip_nat_standalone.c
@@ -31,8 +31,8 @@
 #include <net/checksum.h>
 #include <linux/spinlock.h>
 
-#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_nat_lock)
-#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_nat_lock)
+#define ASSERT_READ_LOCK(x)
+#define ASSERT_WRITE_LOCK(x)
 
 #include <linux/netfilter_ipv4/ip_nat.h>
 #include <linux/netfilter_ipv4/ip_nat_rule.h>
@@ -373,7 +373,6 @@ static int init_or_cleanup(int init)
  cleanup_rule_init:
 	ip_nat_rule_cleanup();
  cleanup_nothing:
-	MUST_BE_READ_WRITE_UNLOCKED(&ip_nat_lock);
 	return ret;
 }
 
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index 8a54f92b849..c88dfcd38c5 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -67,7 +67,6 @@ static DECLARE_MUTEX(ipt_mutex);
 /* Must have mutex */
 #define ASSERT_READ_LOCK(x) IP_NF_ASSERT(down_trylock(&ipt_mutex) != 0)
 #define ASSERT_WRITE_LOCK(x) IP_NF_ASSERT(down_trylock(&ipt_mutex) != 0)
-#include <linux/netfilter_ipv4/lockhelp.h>
 #include <linux/netfilter_ipv4/listhelp.h>
 
 #if 0
diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c
index 0f12e3a3dc7..dc4362b57cf 100644
--- a/net/ipv4/netfilter/ipt_CLUSTERIP.c
+++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c
@@ -29,7 +29,6 @@
 #include <linux/netfilter_ipv4/ip_tables.h>
 #include <linux/netfilter_ipv4/ipt_CLUSTERIP.h>
 #include <linux/netfilter_ipv4/ip_conntrack.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
 
 #define CLUSTERIP_VERSION "0.6"
 
@@ -41,6 +40,8 @@
 #define DEBUGP
 #endif
 
+#define ASSERT_READ_LOCK(x)
+
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>");
 MODULE_DESCRIPTION("iptables target for CLUSTERIP");
@@ -67,7 +68,7 @@ static LIST_HEAD(clusterip_configs);
 
 /* clusterip_lock protects the clusterip_configs list _AND_ the configurable
  * data within all structurses (num_local_nodes, local_nodes[]) */
-static DECLARE_RWLOCK(clusterip_lock);
+static DEFINE_RWLOCK(clusterip_lock);
 
 #ifdef CONFIG_PROC_FS
 static struct file_operations clusterip_proc_fops;
@@ -82,9 +83,9 @@ clusterip_config_get(struct clusterip_config *c) {
 static inline void
 clusterip_config_put(struct clusterip_config *c) {
 	if (atomic_dec_and_test(&c->refcount)) {
-		WRITE_LOCK(&clusterip_lock);
+		write_lock_bh(&clusterip_lock);
 		list_del(&c->list);
-		WRITE_UNLOCK(&clusterip_lock);
+		write_unlock_bh(&clusterip_lock);
 		dev_mc_delete(c->dev, c->clustermac, ETH_ALEN, 0);
 		dev_put(c->dev);
 		kfree(c);
@@ -97,7 +98,7 @@ __clusterip_config_find(u_int32_t clusterip)
 {
 	struct list_head *pos;
 
-	MUST_BE_READ_LOCKED(&clusterip_lock);
+	ASSERT_READ_LOCK(&clusterip_lock);
 	list_for_each(pos, &clusterip_configs) {
 		struct clusterip_config *c = list_entry(pos, 
 					struct clusterip_config, list);
@@ -114,14 +115,14 @@ clusterip_config_find_get(u_int32_t clusterip)
 {
 	struct clusterip_config *c;
 
-	READ_LOCK(&clusterip_lock);
+	read_lock_bh(&clusterip_lock);
 	c = __clusterip_config_find(clusterip);
 	if (!c) {
-		READ_UNLOCK(&clusterip_lock);
+		read_unlock_bh(&clusterip_lock);
 		return NULL;
 	}
 	atomic_inc(&c->refcount);
-	READ_UNLOCK(&clusterip_lock);
+	read_unlock_bh(&clusterip_lock);
 
 	return c;
 }
@@ -160,9 +161,9 @@ clusterip_config_init(struct ipt_clusterip_tgt_info *i, u_int32_t ip,
 	c->pde->data = c;
 #endif
 
-	WRITE_LOCK(&clusterip_lock);
+	write_lock_bh(&clusterip_lock);
 	list_add(&c->list, &clusterip_configs);
-	WRITE_UNLOCK(&clusterip_lock);
+	write_unlock_bh(&clusterip_lock);
 
 	return c;
 }
@@ -172,25 +173,25 @@ clusterip_add_node(struct clusterip_config *c, u_int16_t nodenum)
 {
 	int i;
 
-	WRITE_LOCK(&clusterip_lock);
+	write_lock_bh(&clusterip_lock);
 
 	if (c->num_local_nodes >= CLUSTERIP_MAX_NODES
 	    || nodenum > CLUSTERIP_MAX_NODES) {
-		WRITE_UNLOCK(&clusterip_lock);
+		write_unlock_bh(&clusterip_lock);
 		return 1;
 	}
 
 	/* check if we alrady have this number in our array */
 	for (i = 0; i < c->num_local_nodes; i++) {
 		if (c->local_nodes[i] == nodenum) {
-			WRITE_UNLOCK(&clusterip_lock);
+			write_unlock_bh(&clusterip_lock);
 			return 1;
 		}
 	}
 
 	c->local_nodes[c->num_local_nodes++] = nodenum;
 
-	WRITE_UNLOCK(&clusterip_lock);
+	write_unlock_bh(&clusterip_lock);
 	return 0;
 }
 
@@ -199,10 +200,10 @@ clusterip_del_node(struct clusterip_config *c, u_int16_t nodenum)
 {
 	int i;
 
-	WRITE_LOCK(&clusterip_lock);
+	write_lock_bh(&clusterip_lock);
 
 	if (c->num_local_nodes <= 1 || nodenum > CLUSTERIP_MAX_NODES) {
-		WRITE_UNLOCK(&clusterip_lock);
+		write_unlock_bh(&clusterip_lock);
 		return 1;
 	}
 		
@@ -211,12 +212,12 @@ clusterip_del_node(struct clusterip_config *c, u_int16_t nodenum)
 			int size = sizeof(u_int16_t)*(c->num_local_nodes-(i+1));
 			memmove(&c->local_nodes[i], &c->local_nodes[i+1], size);
 			c->num_local_nodes--;
-			WRITE_UNLOCK(&clusterip_lock);
+			write_unlock_bh(&clusterip_lock);
 			return 0;
 		}
 	}
 
-	WRITE_UNLOCK(&clusterip_lock);
+	write_unlock_bh(&clusterip_lock);
 	return 1;
 }
 
@@ -286,21 +287,21 @@ clusterip_responsible(struct clusterip_config *config, u_int32_t hash)
 {
 	int i;
 
-	READ_LOCK(&clusterip_lock);
+	read_lock_bh(&clusterip_lock);
 
 	if (config->num_local_nodes == 0) {
-		READ_UNLOCK(&clusterip_lock);
+		read_unlock_bh(&clusterip_lock);
 		return 0;
 	}
 
 	for (i = 0; i < config->num_local_nodes; i++) {
 		if (config->local_nodes[i] == hash) {
-			READ_UNLOCK(&clusterip_lock);
+			read_unlock_bh(&clusterip_lock);
 			return 1;
 		}
 	}
 
-	READ_UNLOCK(&clusterip_lock);
+	read_unlock_bh(&clusterip_lock);
 
 	return 0;
 }
@@ -578,7 +579,7 @@ static void *clusterip_seq_start(struct seq_file *s, loff_t *pos)
 	struct clusterip_config *c = pde->data;
 	unsigned int *nodeidx;
 
-	READ_LOCK(&clusterip_lock);
+	read_lock_bh(&clusterip_lock);
 	if (*pos >= c->num_local_nodes)
 		return NULL;
 
@@ -608,7 +609,7 @@ static void clusterip_seq_stop(struct seq_file *s, void *v)
 {
 	kfree(v);
 
-	READ_UNLOCK(&clusterip_lock);
+	read_unlock_bh(&clusterip_lock);
 }
 
 static int clusterip_seq_show(struct seq_file *s, void *v)
diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c
index 57e9f6cf1c3..91e74502c3d 100644
--- a/net/ipv4/netfilter/ipt_MASQUERADE.c
+++ b/net/ipv4/netfilter/ipt_MASQUERADE.c
@@ -33,7 +33,7 @@ MODULE_DESCRIPTION("iptables MASQUERADE target module");
 #endif
 
 /* Lock protects masq region inside conntrack */
-static DECLARE_RWLOCK(masq_lock);
+static DEFINE_RWLOCK(masq_lock);
 
 /* FIXME: Multiple targets. --RR */
 static int
@@ -103,9 +103,9 @@ masquerade_target(struct sk_buff **pskb,
 		return NF_DROP;
 	}
 
-	WRITE_LOCK(&masq_lock);
+	write_lock_bh(&masq_lock);
 	ct->nat.masq_index = out->ifindex;
-	WRITE_UNLOCK(&masq_lock);
+	write_unlock_bh(&masq_lock);
 
 	/* Transfer from original range. */
 	newrange = ((struct ip_nat_range)
@@ -122,9 +122,9 @@ device_cmp(struct ip_conntrack *i, void *ifindex)
 {
 	int ret;
 
-	READ_LOCK(&masq_lock);
+	read_lock_bh(&masq_lock);
 	ret = (i->nat.masq_index == (int)(long)ifindex);
-	READ_UNLOCK(&masq_lock);
+	read_unlock_bh(&masq_lock);
 
 	return ret;
 }
diff --git a/net/ipv4/netfilter/ipt_ULOG.c b/net/ipv4/netfilter/ipt_ULOG.c
index 6f2cefbe16c..52a0076302a 100644
--- a/net/ipv4/netfilter/ipt_ULOG.c
+++ b/net/ipv4/netfilter/ipt_ULOG.c
@@ -56,7 +56,6 @@
 #include <linux/netfilter.h>
 #include <linux/netfilter_ipv4/ip_tables.h>
 #include <linux/netfilter_ipv4/ipt_ULOG.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
 #include <net/sock.h>
 #include <linux/bitops.h>
 
@@ -99,8 +98,8 @@ typedef struct {
 
 static ulog_buff_t ulog_buffers[ULOG_MAXNLGROUPS];	/* array of buffers */
 
-static struct sock *nflognl;	/* our socket */
-static DECLARE_LOCK(ulog_lock);	/* spinlock */
+static struct sock *nflognl;		/* our socket */
+static DEFINE_SPINLOCK(ulog_lock);	/* spinlock */
 
 /* send one ulog_buff_t to userspace */
 static void ulog_send(unsigned int nlgroupnum)
@@ -135,9 +134,9 @@ static void ulog_timer(unsigned long data)
 
 	/* lock to protect against somebody modifying our structure
 	 * from ipt_ulog_target at the same time */
-	LOCK_BH(&ulog_lock);
+	spin_lock_bh(&ulog_lock);
 	ulog_send(data);
-	UNLOCK_BH(&ulog_lock);
+	spin_unlock_bh(&ulog_lock);
 }
 
 static struct sk_buff *ulog_alloc_skb(unsigned int size)
@@ -193,7 +192,7 @@ static void ipt_ulog_packet(unsigned int hooknum,
 
 	ub = &ulog_buffers[groupnum];
 	
-	LOCK_BH(&ulog_lock);
+	spin_lock_bh(&ulog_lock);
 
 	if (!ub->skb) {
 		if (!(ub->skb = ulog_alloc_skb(size)))
@@ -278,7 +277,7 @@ static void ipt_ulog_packet(unsigned int hooknum,
 		ulog_send(groupnum);
 	}
 
-	UNLOCK_BH(&ulog_lock);
+	spin_unlock_bh(&ulog_lock);
 
 	return;
 
@@ -288,7 +287,7 @@ nlmsg_failure:
 alloc_failure:
 	PRINTR("ipt_ULOG: Error building netlink message\n");
 
-	UNLOCK_BH(&ulog_lock);
+	spin_unlock_bh(&ulog_lock);
 }
 
 static unsigned int ipt_ulog_target(struct sk_buff **pskb,
diff --git a/net/ipv4/netfilter/ipt_hashlimit.c b/net/ipv4/netfilter/ipt_hashlimit.c
index f1937190cd7..564b49bfebc 100644
--- a/net/ipv4/netfilter/ipt_hashlimit.c
+++ b/net/ipv4/netfilter/ipt_hashlimit.c
@@ -37,7 +37,6 @@
 
 #include <linux/netfilter_ipv4/ip_tables.h>
 #include <linux/netfilter_ipv4/ipt_hashlimit.h>
-#include <linux/netfilter_ipv4/lockhelp.h>
 
 /* FIXME: this is just for IP_NF_ASSERRT */
 #include <linux/netfilter_ipv4/ip_conntrack.h>
@@ -92,7 +91,7 @@ struct ipt_hashlimit_htable {
 	struct hlist_head hash[0];	/* hashtable itself */
 };
 
-static DECLARE_LOCK(hashlimit_lock);	/* protects htables list */
+static DEFINE_SPINLOCK(hashlimit_lock);	/* protects htables list */
 static DECLARE_MUTEX(hlimit_mutex);	/* additional checkentry protection */
 static HLIST_HEAD(hashlimit_htables);
 static kmem_cache_t *hashlimit_cachep;
@@ -233,9 +232,9 @@ static int htable_create(struct ipt_hashlimit_info *minfo)
 	hinfo->timer.function = htable_gc;
 	add_timer(&hinfo->timer);
 
-	LOCK_BH(&hashlimit_lock);
+	spin_lock_bh(&hashlimit_lock);
 	hlist_add_head(&hinfo->node, &hashlimit_htables);
-	UNLOCK_BH(&hashlimit_lock);
+	spin_unlock_bh(&hashlimit_lock);
 
 	return 0;
 }
@@ -301,15 +300,15 @@ static struct ipt_hashlimit_htable *htable_find_get(char *name)
 	struct ipt_hashlimit_htable *hinfo;
 	struct hlist_node *pos;
 
-	LOCK_BH(&hashlimit_lock);
+	spin_lock_bh(&hashlimit_lock);
 	hlist_for_each_entry(hinfo, pos, &hashlimit_htables, node) {
 		if (!strcmp(name, hinfo->pde->name)) {
 			atomic_inc(&hinfo->use);
-			UNLOCK_BH(&hashlimit_lock);
+			spin_unlock_bh(&hashlimit_lock);
 			return hinfo;
 		}
 	}
-	UNLOCK_BH(&hashlimit_lock);
+	spin_unlock_bh(&hashlimit_lock);
 
 	return NULL;
 }
@@ -317,9 +316,9 @@ static struct ipt_hashlimit_htable *htable_find_get(char *name)
 static void htable_put(struct ipt_hashlimit_htable *hinfo)
 {
 	if (atomic_dec_and_test(&hinfo->use)) {
-		LOCK_BH(&hashlimit_lock);
+		spin_lock_bh(&hashlimit_lock);
 		hlist_del(&hinfo->node);
-		UNLOCK_BH(&hashlimit_lock);
+		spin_unlock_bh(&hashlimit_lock);
 		htable_destroy(hinfo);
 	}
 }
diff --git a/net/ipv4/netfilter/ipt_helper.c b/net/ipv4/netfilter/ipt_helper.c
index 33fdf364d3d..3e7dd014de4 100644
--- a/net/ipv4/netfilter/ipt_helper.c
+++ b/net/ipv4/netfilter/ipt_helper.c
@@ -53,7 +53,7 @@ match(const struct sk_buff *skb,
 		return ret;
 	}
 
-	READ_LOCK(&ip_conntrack_lock);
+	read_lock_bh(&ip_conntrack_lock);
 	if (!ct->master->helper) {
 		DEBUGP("ipt_helper: master ct %p has no helper\n", 
 			exp->expectant);
@@ -69,7 +69,7 @@ match(const struct sk_buff *skb,
 		ret ^= !strncmp(ct->master->helper->name, info->name, 
 		                strlen(ct->master->helper->name));
 out_unlock:
-	READ_UNLOCK(&ip_conntrack_lock);
+	read_unlock_bh(&ip_conntrack_lock);
 	return ret;
 }
 
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index c735276fdd5..73034511c8d 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -71,7 +71,6 @@ static DECLARE_MUTEX(ip6t_mutex);
 /* Must have mutex */
 #define ASSERT_READ_LOCK(x) IP_NF_ASSERT(down_trylock(&ip6t_mutex) != 0)
 #define ASSERT_WRITE_LOCK(x) IP_NF_ASSERT(down_trylock(&ip6t_mutex) != 0)
-#include <linux/netfilter_ipv4/lockhelp.h>
 #include <linux/netfilter_ipv4/listhelp.h>
 
 #if 0
-- 
cgit v1.2.3-70-g09d2


From 18b8afc771102b1b6af97962808291a7d27f52af Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 21 Jun 2005 14:01:57 -0700
Subject: [NETFILTER]: Kill nf_debug

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4.h         |   6 --
 include/linux/skbuff.h                 |  13 ----
 net/bridge/br_forward.c                |   3 -
 net/bridge/br_input.c                  |   4 -
 net/bridge/br_netfilter.c              |  38 ---------
 net/core/netfilter.c                   | 138 ---------------------------------
 net/core/skbuff.c                      |   6 --
 net/ipv4/ip_input.c                    |   4 -
 net/ipv4/ip_output.c                   |  11 ---
 net/ipv4/ipvs/ip_vs_xmit.c             |   1 -
 net/ipv4/netfilter/ip_conntrack_core.c |   9 ---
 net/ipv4/netfilter/ip_nat_helper.c     |   3 -
 net/ipv6/ip6_output.c                  |   3 -
 13 files changed, 239 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4.h b/include/linux/netfilter_ipv4.h
index 9e5750079e0..3ebc36afae1 100644
--- a/include/linux/netfilter_ipv4.h
+++ b/include/linux/netfilter_ipv4.h
@@ -75,12 +75,6 @@ enum nf_ip_hook_priorities {
 #define SO_ORIGINAL_DST 80
 
 #ifdef __KERNEL__
-#ifdef CONFIG_NETFILTER_DEBUG
-void nf_debug_ip_local_deliver(struct sk_buff *skb);
-void nf_debug_ip_loopback_xmit(struct sk_buff *newskb);
-void nf_debug_ip_finish_output2(struct sk_buff *skb);
-#endif /*CONFIG_NETFILTER_DEBUG*/
-
 extern int ip_route_me_harder(struct sk_buff **pskb);
 
 /* Call this before modifying an existing IP packet: ensures it is
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index cc04f5cd228..d7c839a2184 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -193,7 +193,6 @@ struct skb_shared_info {
  *	@nfcache: Cache info
  *	@nfct: Associated connection, if any
  *	@nfctinfo: Relationship of this skb to the connection
- *	@nf_debug: Netfilter debugging
  *	@nf_bridge: Saved data about a bridged frame - see br_netfilter.c
  *      @private: Data which is private to the HIPPI implementation
  *	@tc_index: Traffic control index
@@ -264,9 +263,6 @@ struct sk_buff {
 	__u32			nfcache;
 	__u32			nfctinfo;
 	struct nf_conntrack	*nfct;
-#ifdef CONFIG_NETFILTER_DEBUG
-        unsigned int		nf_debug;
-#endif
 #ifdef CONFIG_BRIDGE_NETFILTER
 	struct nf_bridge_info	*nf_bridge;
 #endif
@@ -1219,15 +1215,6 @@ static inline void nf_reset(struct sk_buff *skb)
 {
 	nf_conntrack_put(skb->nfct);
 	skb->nfct = NULL;
-#ifdef CONFIG_NETFILTER_DEBUG
-	skb->nf_debug = 0;
-#endif
-}
-static inline void nf_reset_debug(struct sk_buff *skb)
-{
-#ifdef CONFIG_NETFILTER_DEBUG
-	skb->nf_debug = 0;
-#endif
 }
 
 #ifdef CONFIG_BRIDGE_NETFILTER
diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index ef9f2095f96..069253f830c 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -57,9 +57,6 @@ int br_forward_finish(struct sk_buff *skb)
 static void __br_deliver(const struct net_bridge_port *to, struct sk_buff *skb)
 {
 	skb->dev = to->dev;
-#ifdef CONFIG_NETFILTER_DEBUG
-	skb->nf_debug = 0;
-#endif
 	NF_HOOK(PF_BRIDGE, NF_BR_LOCAL_OUT, skb, NULL, skb->dev,
 			br_forward_finish);
 }
diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c
index 8f5f2e73099..9a45e6279c5 100644
--- a/net/bridge/br_input.c
+++ b/net/bridge/br_input.c
@@ -23,11 +23,7 @@ const unsigned char bridge_ula[6] = { 0x01, 0x80, 0xc2, 0x00, 0x00, 0x00 };
 
 static int br_pass_frame_up_finish(struct sk_buff *skb)
 {
-#ifdef CONFIG_NETFILTER_DEBUG
-	skb->nf_debug = 0;
-#endif
 	netif_receive_skb(skb);
-
 	return 0;
 }
 
diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c
index be03d3ad264..03ae4edddac 100644
--- a/net/bridge/br_netfilter.c
+++ b/net/bridge/br_netfilter.c
@@ -102,10 +102,6 @@ static int br_nf_pre_routing_finish_ipv6(struct sk_buff *skb)
 {
 	struct nf_bridge_info *nf_bridge = skb->nf_bridge;
 
-#ifdef CONFIG_NETFILTER_DEBUG
-	skb->nf_debug ^= (1 << NF_BR_PRE_ROUTING);
-#endif
-
 	if (nf_bridge->mask & BRNF_PKT_TYPE) {
 		skb->pkt_type = PACKET_OTHERHOST;
 		nf_bridge->mask ^= BRNF_PKT_TYPE;
@@ -182,10 +178,6 @@ static void __br_dnat_complain(void)
  * --Bart, 20021007 (updated) */
 static int br_nf_pre_routing_finish_bridge(struct sk_buff *skb)
 {
-#ifdef CONFIG_NETFILTER_DEBUG
-	skb->nf_debug |= (1 << NF_BR_PRE_ROUTING) | (1 << NF_BR_FORWARD);
-#endif
-
 	if (skb->pkt_type == PACKET_OTHERHOST) {
 		skb->pkt_type = PACKET_HOST;
 		skb->nf_bridge->mask |= BRNF_PKT_TYPE;
@@ -207,10 +199,6 @@ static int br_nf_pre_routing_finish(struct sk_buff *skb)
 	struct iphdr *iph = skb->nh.iph;
 	struct nf_bridge_info *nf_bridge = skb->nf_bridge;
 
-#ifdef CONFIG_NETFILTER_DEBUG
-	skb->nf_debug ^= (1 << NF_BR_PRE_ROUTING);
-#endif
-
 	if (nf_bridge->mask & BRNF_PKT_TYPE) {
 		skb->pkt_type = PACKET_OTHERHOST;
 		nf_bridge->mask ^= BRNF_PKT_TYPE;
@@ -382,9 +370,6 @@ static unsigned int br_nf_pre_routing_ipv6(unsigned int hook,
 	if (hdr->nexthdr == NEXTHDR_HOP && check_hbh_len(skb))
 			goto inhdr_error;
 
-#ifdef CONFIG_NETFILTER_DEBUG
-	skb->nf_debug ^= (1 << NF_IP6_PRE_ROUTING);
-#endif
 	if ((nf_bridge = nf_bridge_alloc(skb)) == NULL)
 		return NF_DROP;
 	setup_pre_routing(skb);
@@ -468,9 +453,6 @@ static unsigned int br_nf_pre_routing(unsigned int hook, struct sk_buff **pskb,
 			skb->ip_summed = CHECKSUM_NONE;
 	}
 
-#ifdef CONFIG_NETFILTER_DEBUG
-	skb->nf_debug ^= (1 << NF_IP_PRE_ROUTING);
-#endif
 	if ((nf_bridge = nf_bridge_alloc(skb)) == NULL)
 		return NF_DROP;
 	setup_pre_routing(skb);
@@ -517,10 +499,6 @@ static int br_nf_forward_finish(struct sk_buff *skb)
 	struct net_device *in;
 	struct vlan_ethhdr *hdr = vlan_eth_hdr(skb);
 
-#ifdef CONFIG_NETFILTER_DEBUG
-	skb->nf_debug ^= (1 << NF_BR_FORWARD);
-#endif
-
 	if (skb->protocol != __constant_htons(ETH_P_ARP) && !IS_VLAN_ARP) {
 		in = nf_bridge->physindev;
 		if (nf_bridge->mask & BRNF_PKT_TYPE) {
@@ -566,9 +544,6 @@ static unsigned int br_nf_forward_ip(unsigned int hook, struct sk_buff **pskb,
 		(*pskb)->nh.raw += VLAN_HLEN;
 	}
 
-#ifdef CONFIG_NETFILTER_DEBUG
-	skb->nf_debug ^= (1 << NF_BR_FORWARD);
-#endif
 	nf_bridge = skb->nf_bridge;
 	if (skb->pkt_type == PACKET_OTHERHOST) {
 		skb->pkt_type = PACKET_HOST;
@@ -605,10 +580,6 @@ static unsigned int br_nf_forward_arp(unsigned int hook, struct sk_buff **pskb,
 		(*pskb)->nh.raw += VLAN_HLEN;
 	}
 
-#ifdef CONFIG_NETFILTER_DEBUG
-	skb->nf_debug ^= (1 << NF_BR_FORWARD);
-#endif
-
 	if (skb->nh.arph->ar_pln != 4) {
 		if (IS_VLAN_ARP) {
 			skb_push(*pskb, VLAN_HLEN);
@@ -627,9 +598,6 @@ static unsigned int br_nf_forward_arp(unsigned int hook, struct sk_buff **pskb,
 /* PF_BRIDGE/LOCAL_OUT ***********************************************/
 static int br_nf_local_out_finish(struct sk_buff *skb)
 {
-#ifdef CONFIG_NETFILTER_DEBUG
-	skb->nf_debug &= ~(1 << NF_BR_LOCAL_OUT);
-#endif
 	if (skb->protocol == __constant_htons(ETH_P_8021Q)) {
 		skb_push(skb, VLAN_HLEN);
 		skb->nh.raw -= VLAN_HLEN;
@@ -731,10 +699,6 @@ static unsigned int br_nf_local_out(unsigned int hook, struct sk_buff **pskb,
 			       realoutdev, br_nf_local_out_finish,
 			       NF_IP_PRI_BRIDGE_SABOTAGE_FORWARD + 1);
 	} else {
-#ifdef CONFIG_NETFILTER_DEBUG
-		skb->nf_debug ^= (1 << NF_IP_LOCAL_OUT);
-#endif
-
 		NF_HOOK_THRESH(pf, NF_IP_LOCAL_OUT, skb, realindev,
 			       realoutdev, br_nf_local_out_finish,
 			       NF_IP_PRI_BRIDGE_SABOTAGE_LOCAL_OUT + 1);
@@ -779,8 +743,6 @@ static unsigned int br_nf_post_routing(unsigned int hook, struct sk_buff **pskb,
 		printk(KERN_CRIT "br_netfilter: skb->dst == NULL.");
 		goto print_error;
 	}
-
-	skb->nf_debug ^= (1 << NF_IP_POST_ROUTING);
 #endif
 
 	/* We assume any code from br_dev_queue_push_xmit onwards doesn't care
diff --git a/net/core/netfilter.c b/net/core/netfilter.c
index 22a8f127c4a..076c156d5ed 100644
--- a/net/core/netfilter.c
+++ b/net/core/netfilter.c
@@ -141,136 +141,6 @@ void nf_unregister_sockopt(struct nf_sockopt_ops *reg)
 	up(&nf_sockopt_mutex);
 }
 
-#ifdef CONFIG_NETFILTER_DEBUG
-#include <net/ip.h>
-#include <net/tcp.h>
-#include <linux/netfilter_ipv4.h>
-
-static void debug_print_hooks_ip(unsigned int nf_debug)
-{
-	if (nf_debug & (1 << NF_IP_PRE_ROUTING)) {
-		printk("PRE_ROUTING ");
-		nf_debug ^= (1 << NF_IP_PRE_ROUTING);
-	}
-	if (nf_debug & (1 << NF_IP_LOCAL_IN)) {
-		printk("LOCAL_IN ");
-		nf_debug ^= (1 << NF_IP_LOCAL_IN);
-	}
-	if (nf_debug & (1 << NF_IP_FORWARD)) {
-		printk("FORWARD ");
-		nf_debug ^= (1 << NF_IP_FORWARD);
-	}
-	if (nf_debug & (1 << NF_IP_LOCAL_OUT)) {
-		printk("LOCAL_OUT ");
-		nf_debug ^= (1 << NF_IP_LOCAL_OUT);
-	}
-	if (nf_debug & (1 << NF_IP_POST_ROUTING)) {
-		printk("POST_ROUTING ");
-		nf_debug ^= (1 << NF_IP_POST_ROUTING);
-	}
-	if (nf_debug)
-		printk("Crap bits: 0x%04X", nf_debug);
-	printk("\n");
-}
-
-static void nf_dump_skb(int pf, struct sk_buff *skb)
-{
-	printk("skb: pf=%i %s dev=%s len=%u\n", 
-	       pf,
-	       skb->sk ? "(owned)" : "(unowned)",
-	       skb->dev ? skb->dev->name : "(no dev)",
-	       skb->len);
-	switch (pf) {
-	case PF_INET: {
-		const struct iphdr *ip = skb->nh.iph;
-		__u32 *opt = (__u32 *) (ip + 1);
-		int opti;
-		__u16 src_port = 0, dst_port = 0;
-
-		if (ip->protocol == IPPROTO_TCP
-		    || ip->protocol == IPPROTO_UDP) {
-			struct tcphdr *tcp=(struct tcphdr *)((__u32 *)ip+ip->ihl);
-			src_port = ntohs(tcp->source);
-			dst_port = ntohs(tcp->dest);
-		}
-	
-		printk("PROTO=%d %u.%u.%u.%u:%hu %u.%u.%u.%u:%hu"
-		       " L=%hu S=0x%2.2hX I=%hu F=0x%4.4hX T=%hu",
-		       ip->protocol, NIPQUAD(ip->saddr),
-		       src_port, NIPQUAD(ip->daddr),
-		       dst_port,
-		       ntohs(ip->tot_len), ip->tos, ntohs(ip->id),
-		       ntohs(ip->frag_off), ip->ttl);
-
-		for (opti = 0; opti < (ip->ihl - sizeof(struct iphdr) / 4); opti++)
-			printk(" O=0x%8.8X", *opt++);
-		printk("\n");
-	}
-	}
-}
-
-void nf_debug_ip_local_deliver(struct sk_buff *skb)
-{
-	/* If it's a loopback packet, it must have come through
-	 * NF_IP_LOCAL_OUT, NF_IP_RAW_INPUT, NF_IP_PRE_ROUTING and
-	 * NF_IP_LOCAL_IN.  Otherwise, must have gone through
-	 * NF_IP_RAW_INPUT and NF_IP_PRE_ROUTING.  */
-	if (!skb->dev) {
-		printk("ip_local_deliver: skb->dev is NULL.\n");
-	} else {
-		if (skb->nf_debug != ((1<<NF_IP_PRE_ROUTING)
-				      | (1<<NF_IP_LOCAL_IN))) {
-			printk("ip_local_deliver: bad skb: ");
-			debug_print_hooks_ip(skb->nf_debug);
-			nf_dump_skb(PF_INET, skb);
-		}
-	}
-}
-
-void nf_debug_ip_loopback_xmit(struct sk_buff *newskb)
-{
-	if (newskb->nf_debug != ((1 << NF_IP_LOCAL_OUT)
-				 | (1 << NF_IP_POST_ROUTING))) {
-		printk("ip_dev_loopback_xmit: bad owned skb = %p: ", 
-		       newskb);
-		debug_print_hooks_ip(newskb->nf_debug);
-		nf_dump_skb(PF_INET, newskb);
-	}
-}
-
-void nf_debug_ip_finish_output2(struct sk_buff *skb)
-{
-	/* If it's owned, it must have gone through the
-	 * NF_IP_LOCAL_OUT and NF_IP_POST_ROUTING.
-	 * Otherwise, must have gone through
-	 * NF_IP_PRE_ROUTING, NF_IP_FORWARD and NF_IP_POST_ROUTING.
-	 */
-	if (skb->sk) {
-		if (skb->nf_debug != ((1 << NF_IP_LOCAL_OUT)
-				      | (1 << NF_IP_POST_ROUTING))) {
-			printk("ip_finish_output: bad owned skb = %p: ", skb);
-			debug_print_hooks_ip(skb->nf_debug);
-			nf_dump_skb(PF_INET, skb);
-		}
-	} else {
-		if (skb->nf_debug != ((1 << NF_IP_PRE_ROUTING)
-				      | (1 << NF_IP_FORWARD)
-				      | (1 << NF_IP_POST_ROUTING))) {
-			/* Fragments, entunnelled packets, TCP RSTs
-                           generated by ipt_REJECT will have no
-                           owners, but still may be local */
-			if (skb->nf_debug != ((1 << NF_IP_LOCAL_OUT)
-					      | (1 << NF_IP_POST_ROUTING))){
-				printk("ip_finish_output:"
-				       " bad unowned skb = %p: ",skb);
-				debug_print_hooks_ip(skb->nf_debug);
-				nf_dump_skb(PF_INET, skb);
-			}
-		}
-	}
-}
-#endif /*CONFIG_NETFILTER_DEBUG*/
-
 /* Call get/setsockopt() */
 static int nf_sockopt(struct sock *sk, int pf, int val, 
 		      char __user *opt, int *len, int get)
@@ -488,14 +358,6 @@ int nf_hook_slow(int pf, unsigned int hook, struct sk_buff **pskb,
 	/* We may already have this, but read-locks nest anyway */
 	rcu_read_lock();
 
-#ifdef CONFIG_NETFILTER_DEBUG
-	if (unlikely((*pskb)->nf_debug & (1 << hook))) {
-		printk("nf_hook: hook %i already set.\n", hook);
-		nf_dump_skb(pf, *pskb);
-	}
-	(*pskb)->nf_debug |= (1 << hook);
-#endif
-
 	elem = &nf_hooks[pf][hook];
 next_hook:
 	verdict = nf_iterate(&nf_hooks[pf][hook], pskb, hook, indev,
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index f65b3de590a..6d68c03bc05 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -365,9 +365,6 @@ struct sk_buff *skb_clone(struct sk_buff *skb, int gfp_mask)
 	C(nfct);
 	nf_conntrack_get(skb->nfct);
 	C(nfctinfo);
-#ifdef CONFIG_NETFILTER_DEBUG
-	C(nf_debug);
-#endif
 #ifdef CONFIG_BRIDGE_NETFILTER
 	C(nf_bridge);
 	nf_bridge_get(skb->nf_bridge);
@@ -432,9 +429,6 @@ static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old)
 	new->nfct	= old->nfct;
 	nf_conntrack_get(old->nfct);
 	new->nfctinfo	= old->nfctinfo;
-#ifdef CONFIG_NETFILTER_DEBUG
-	new->nf_debug	= old->nf_debug;
-#endif
 #ifdef CONFIG_BRIDGE_NETFILTER
 	new->nf_bridge	= old->nf_bridge;
 	nf_bridge_get(old->nf_bridge);
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index 4e47a2658c7..2b7485e6504 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -200,10 +200,6 @@ static inline int ip_local_deliver_finish(struct sk_buff *skb)
 {
 	int ihl = skb->nh.iph->ihl*4;
 
-#ifdef CONFIG_NETFILTER_DEBUG
-	nf_debug_ip_local_deliver(skb);
-#endif /*CONFIG_NETFILTER_DEBUG*/
-
 	__skb_pull(skb, ihl);
 
 	/* Free reference early: we don't need it any more, and it may
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 760dc8238d6..ee07aec215a 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -107,10 +107,6 @@ static int ip_dev_loopback_xmit(struct sk_buff *newskb)
 	newskb->pkt_type = PACKET_LOOPBACK;
 	newskb->ip_summed = CHECKSUM_UNNECESSARY;
 	BUG_TRAP(newskb->dst);
-
-#ifdef CONFIG_NETFILTER_DEBUG
-	nf_debug_ip_loopback_xmit(newskb);
-#endif
 	nf_reset(newskb);
 	netif_rx(newskb);
 	return 0;
@@ -192,10 +188,6 @@ static inline int ip_finish_output2(struct sk_buff *skb)
 		skb = skb2;
 	}
 
-#ifdef CONFIG_NETFILTER_DEBUG
-	nf_debug_ip_finish_output2(skb);
-#endif /*CONFIG_NETFILTER_DEBUG*/
-
 	nf_reset(skb);
 
 	if (hh) {
@@ -415,9 +407,6 @@ static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from)
 	to->nf_bridge = from->nf_bridge;
 	nf_bridge_get(to->nf_bridge);
 #endif
-#ifdef CONFIG_NETFILTER_DEBUG
-	to->nf_debug = from->nf_debug;
-#endif
 #endif
 }
 
diff --git a/net/ipv4/ipvs/ip_vs_xmit.c b/net/ipv4/ipvs/ip_vs_xmit.c
index de21da00057..a8512a3fd08 100644
--- a/net/ipv4/ipvs/ip_vs_xmit.c
+++ b/net/ipv4/ipvs/ip_vs_xmit.c
@@ -127,7 +127,6 @@ ip_vs_dst_reset(struct ip_vs_dest *dest)
 
 #define IP_VS_XMIT(skb, rt)				\
 do {							\
-	nf_reset_debug(skb);				\
 	(skb)->nfcache |= NFC_IPVS_PROPERTY;		\
 	(skb)->ip_summed = CHECKSUM_NONE;		\
 	NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, (skb), NULL,	\
diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c
index a7377a331ad..ffba0ad3c88 100644
--- a/net/ipv4/netfilter/ip_conntrack_core.c
+++ b/net/ipv4/netfilter/ip_conntrack_core.c
@@ -940,10 +940,6 @@ void ip_ct_refresh_acct(struct ip_conntrack *ct,
 struct sk_buff *
 ip_ct_gather_frags(struct sk_buff *skb, u_int32_t user)
 {
-#ifdef CONFIG_NETFILTER_DEBUG
-	unsigned int olddebug = skb->nf_debug;
-#endif
-
 	skb_orphan(skb);
 
 	local_bh_disable(); 
@@ -953,12 +949,7 @@ ip_ct_gather_frags(struct sk_buff *skb, u_int32_t user)
 	if (skb) {
 		ip_send_check(skb->nh.iph);
 		skb->nfcache |= NFC_ALTERED;
-#ifdef CONFIG_NETFILTER_DEBUG
-		/* Packet path as if nothing had happened. */
-		skb->nf_debug = olddebug;
-#endif
 	}
-
 	return skb;
 }
 
diff --git a/net/ipv4/netfilter/ip_nat_helper.c b/net/ipv4/netfilter/ip_nat_helper.c
index 9cd51f180dc..158f34f32c0 100644
--- a/net/ipv4/netfilter/ip_nat_helper.c
+++ b/net/ipv4/netfilter/ip_nat_helper.c
@@ -142,9 +142,6 @@ static int enlarge_skb(struct sk_buff **pskb, unsigned int extra)
 	/* Transfer socket to new skb. */
 	if ((*pskb)->sk)
 		skb_set_owner_w(nskb, (*pskb)->sk);
-#ifdef CONFIG_NETFILTER_DEBUG
-	nskb->nf_debug = (*pskb)->nf_debug;
-#endif
 	kfree_skb(*pskb);
 	*pskb = nskb;
 	return 1;
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index b78a5358680..06e7cdaeedc 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -484,9 +484,6 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from)
 	to->nf_bridge = from->nf_bridge;
 	nf_bridge_get(to->nf_bridge);
 #endif
-#ifdef CONFIG_NETFILTER_DEBUG
-	to->nf_debug = from->nf_debug;
-#endif
 #endif
 }
 
-- 
cgit v1.2.3-70-g09d2


From 7049e6800f40046c384c522a990669024d5f5836 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 21 Jun 2005 16:20:28 -0700
Subject: [SPARC64]: Add prefetch support.

The implementation is optimal for UltraSPARC-III and later.
It will work, however suboptimally, on UltraSPARC-II and
be treated as a NOP on UltraSPARC-I.

It is not worth code patching this thing as the highest cost
is the code space, and code patching cannot eliminate that.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/asm-sparc64/processor.h | 34 ++++++++++++++++++++++++++++++++++
 1 file changed, 34 insertions(+)

(limited to 'include')

diff --git a/include/asm-sparc64/processor.h b/include/asm-sparc64/processor.h
index bc1445b904e..d0bee241356 100644
--- a/include/asm-sparc64/processor.h
+++ b/include/asm-sparc64/processor.h
@@ -192,6 +192,40 @@ extern unsigned long get_wchan(struct task_struct *task);
 
 #define cpu_relax()	barrier()
 
+/* Prefetch support.  This is tuned for UltraSPARC-III and later.
+ * UltraSPARC-I will treat these as nops, and UltraSPARC-II has
+ * a shallower prefetch queue than later chips.
+ */
+#define ARCH_HAS_PREFETCH
+#define ARCH_HAS_PREFETCHW
+#define ARCH_HAS_SPINLOCK_PREFETCH
+
+static inline void prefetch(const void *x)
+{
+	/* We do not use the read prefetch mnemonic because that
+	 * prefetches into the prefetch-cache which only is accessible
+	 * by floating point operations in UltraSPARC-III and later.
+	 * By contrast, "#one_write" prefetches into the L2 cache
+	 * in shared state.
+	 */
+	__asm__ __volatile__("prefetch [%0], #one_write"
+			     : /* no outputs */
+			     : "r" (x));
+}
+
+static inline void prefetchw(const void *x)
+{
+	/* The most optimal prefetch to use for writes is
+	 * "#n_writes".  This brings the cacheline into the
+	 * L2 cache in "owned" state.
+	 */
+	__asm__ __volatile__("prefetch [%0], #n_writes"
+			     : /* no outputs */
+			     : "r" (x));
+}
+
+#define spin_lock_prefetch(x)	prefetchw(x)
+
 #endif /* !(__ASSEMBLY__) */
 
 #endif /* !(__ASM_SPARC64_PROCESSOR_H) */
-- 
cgit v1.2.3-70-g09d2


From 9a558cb4ec43ccbe82e2cf899ced37f29ac017df Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Tue, 21 Jun 2005 17:14:28 -0700
Subject: [PATCH] arm: irqs_disabled() type fix

kernel/sched.c: In function `__might_sleep':
kernel/sched.c:5461: warning: int format, long unsigned int arg (arg 3)

We expect irqs_disabled() to return an int (poor man's bool).

Acked-by: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-arm/system.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-arm/system.h b/include/asm-arm/system.h
index 8405eb6558e..39dd7008013 100644
--- a/include/asm-arm/system.h
+++ b/include/asm-arm/system.h
@@ -308,7 +308,7 @@ do {									\
 ({					\
 	unsigned long flags;		\
 	local_save_flags(flags);	\
-	flags & PSR_I_BIT;		\
+	(int)(flags & PSR_I_BIT);	\
 })
 
 #ifdef CONFIG_SMP
-- 
cgit v1.2.3-70-g09d2


From 84929801e14d968caeb84795bfbb88f04283fbd9 Mon Sep 17 00:00:00 2001
From: Suresh Siddha <suresh.b.siddha@intel.com>
Date: Tue, 21 Jun 2005 17:14:32 -0700
Subject: [PATCH] x86_64: TASK_SIZE fixes for compatibility mode processes

Appended patch will setup compatibility mode TASK_SIZE properly.  This will
fix atleast three known bugs that can be encountered while running
compatibility mode apps.

a) A malicious 32bit app can have an elf section at 0xffffe000.  During
   exec of this app, we will have a memory leak as insert_vm_struct() is
   not checking for return value in syscall32_setup_pages() and thus not
   freeing the vma allocated for the vsyscall page.  And instead of exec
   failing (as it has addresses > TASK_SIZE), we were allowing it to
   succeed previously.

b) With a 32bit app, hugetlb_get_unmapped_area/arch_get_unmapped_area
   may return addresses beyond 32bits, ultimately causing corruption
   because of wrap-around and resulting in SEGFAULT, instead of returning
   ENOMEM.

c) 32bit app doing this below mmap will now fail.

  mmap((void *)(0xFFFFE000UL), 0x10000UL, PROT_READ|PROT_WRITE,
	MAP_FIXED|MAP_PRIVATE|MAP_ANON, 0, 0);

Signed-off-by: Zou Nan hai <nanhai.zou@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/ia32/ia32_binfmt.c  |  5 +----
 arch/x86_64/kernel/process.c    |  4 ++--
 arch/x86_64/kernel/ptrace.c     | 17 ++++++++++-------
 arch/x86_64/kernel/sys_x86_64.c | 14 ++++----------
 arch/x86_64/mm/fault.c          |  2 +-
 include/asm-x86_64/a.out.h      |  2 +-
 include/asm-x86_64/processor.h  | 11 ++++++-----
 7 files changed, 25 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/ia32/ia32_binfmt.c b/arch/x86_64/ia32/ia32_binfmt.c
index 99b522052d1..c8131f342cf 100644
--- a/arch/x86_64/ia32/ia32_binfmt.c
+++ b/arch/x86_64/ia32/ia32_binfmt.c
@@ -46,7 +46,7 @@ struct elf_phdr;
 
 #define IA32_EMULATOR 1
 
-#define ELF_ET_DYN_BASE		(TASK_UNMAPPED_32 + 0x1000000)
+#define ELF_ET_DYN_BASE		(TASK_UNMAPPED_BASE + 0x1000000)
 
 #undef ELF_ARCH
 #define ELF_ARCH EM_386
@@ -307,9 +307,6 @@ MODULE_AUTHOR("Eric Youngdale, Andi Kleen");
 
 #define elf_addr_t __u32
 
-#undef TASK_SIZE
-#define TASK_SIZE 0xffffffff
-
 static void elf32_init(struct pt_regs *);
 
 #define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1
diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 761b6d35e33..dce8bab4306 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -656,7 +656,7 @@ long do_arch_prctl(struct task_struct *task, int code, unsigned long addr)
 
 	switch (code) { 
 	case ARCH_SET_GS:
-		if (addr >= TASK_SIZE) 
+		if (addr >= TASK_SIZE_OF(task))
 			return -EPERM; 
 		cpu = get_cpu();
 		/* handle small bases via the GDT because that's faster to 
@@ -682,7 +682,7 @@ long do_arch_prctl(struct task_struct *task, int code, unsigned long addr)
 	case ARCH_SET_FS:
 		/* Not strictly needed for fs, but do it for symmetry
 		   with gs */
-		if (addr >= TASK_SIZE)
+		if (addr >= TASK_SIZE_OF(task))
 			return -EPERM; 
 		cpu = get_cpu();
 		/* handle small bases via the GDT because that's faster to 
diff --git a/arch/x86_64/kernel/ptrace.c b/arch/x86_64/kernel/ptrace.c
index 525f6a128a2..bbf64b59a21 100644
--- a/arch/x86_64/kernel/ptrace.c
+++ b/arch/x86_64/kernel/ptrace.c
@@ -257,12 +257,12 @@ static int putreg(struct task_struct *child,
 			value &= 0xffff;
 			return 0;
 		case offsetof(struct user_regs_struct,fs_base):
-			if (value >= TASK_SIZE)
+			if (value >= TASK_SIZE_OF(child))
 				return -EIO;
 			child->thread.fs = value;
 			return 0;
 		case offsetof(struct user_regs_struct,gs_base):
-			if (value >= TASK_SIZE)
+			if (value >= TASK_SIZE_OF(child))
 				return -EIO;
 			child->thread.gs = value;
 			return 0;
@@ -279,7 +279,7 @@ static int putreg(struct task_struct *child,
 			break;
 		case offsetof(struct user_regs_struct, rip):
 			/* Check if the new RIP address is canonical */
-			if (value >= TASK_SIZE)
+			if (value >= TASK_SIZE_OF(child))
 				return -EIO;
 			break;
 	}
@@ -419,6 +419,8 @@ asmlinkage long sys_ptrace(long request, long pid, unsigned long addr, long data
 		break;
 
 	case PTRACE_POKEUSR: /* write the word at location addr in the USER area */
+	{
+		int dsize = test_tsk_thread_flag(child, TIF_IA32) ? 3 : 7;
 		ret = -EIO;
 		if ((addr & 7) ||
 		    addr > sizeof(struct user) - 7)
@@ -430,22 +432,22 @@ asmlinkage long sys_ptrace(long request, long pid, unsigned long addr, long data
 			break;
 		/* Disallows to set a breakpoint into the vsyscall */
 		case offsetof(struct user, u_debugreg[0]):
-			if (data >= TASK_SIZE-7) break;
+			if (data >= TASK_SIZE_OF(child) - dsize) break;
 			child->thread.debugreg0 = data;
 			ret = 0;
 			break;
 		case offsetof(struct user, u_debugreg[1]):
-			if (data >= TASK_SIZE-7) break;
+			if (data >= TASK_SIZE_OF(child) - dsize) break;
 			child->thread.debugreg1 = data;
 			ret = 0;
 			break;
 		case offsetof(struct user, u_debugreg[2]):
-			if (data >= TASK_SIZE-7) break;
+			if (data >= TASK_SIZE_OF(child) - dsize) break;
 			child->thread.debugreg2 = data;
 			ret = 0;
 			break;
 		case offsetof(struct user, u_debugreg[3]):
-			if (data >= TASK_SIZE-7) break;
+			if (data >= TASK_SIZE_OF(child) - dsize) break;
 			child->thread.debugreg3 = data;
 			ret = 0;
 			break;
@@ -469,6 +471,7 @@ asmlinkage long sys_ptrace(long request, long pid, unsigned long addr, long data
 		  break;
 		}
 		break;
+	}
 	case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */
 	case PTRACE_CONT:    /* restart after signal. */
 
diff --git a/arch/x86_64/kernel/sys_x86_64.c b/arch/x86_64/kernel/sys_x86_64.c
index dbebd5ccba6..d9798dd433f 100644
--- a/arch/x86_64/kernel/sys_x86_64.c
+++ b/arch/x86_64/kernel/sys_x86_64.c
@@ -68,13 +68,7 @@ out:
 static void find_start_end(unsigned long flags, unsigned long *begin,
 			   unsigned long *end)
 {
-#ifdef CONFIG_IA32_EMULATION
-	if (test_thread_flag(TIF_IA32)) { 
-		*begin = TASK_UNMAPPED_32;
-		*end = IA32_PAGE_OFFSET; 
-	} else 
-#endif
-	if (flags & MAP_32BIT) { 
+	if (!test_thread_flag(TIF_IA32) && (flags & MAP_32BIT)) {
 		/* This is usually used needed to map code in small
 		   model, so it needs to be in the first 31bit. Limit
 		   it to that.  This means we need to move the
@@ -84,10 +78,10 @@ static void find_start_end(unsigned long flags, unsigned long *begin,
 		   of playground for now. -AK */ 
 		*begin = 0x40000000; 
 		*end = 0x80000000;		
-	} else { 
-		*begin = TASK_UNMAPPED_64; 
+	} else {
+		*begin = TASK_UNMAPPED_BASE;
 		*end = TASK_SIZE; 
-		}
+	}
 } 
 
 unsigned long
diff --git a/arch/x86_64/mm/fault.c b/arch/x86_64/mm/fault.c
index 5d6b2114cc9..57d3ab15a5c 100644
--- a/arch/x86_64/mm/fault.c
+++ b/arch/x86_64/mm/fault.c
@@ -350,7 +350,7 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long error_code)
 	 * (error_code & 4) == 0, and that the fault was not a
 	 * protection error (error_code & 1) == 0.
 	 */
-	if (unlikely(address >= TASK_SIZE)) {
+	if (unlikely(address >= TASK_SIZE64)) {
 		if (!(error_code & 5) &&
 		      ((address >= VMALLOC_START && address < VMALLOC_END) ||
 		       (address >= MODULES_VADDR && address < MODULES_END))) {
diff --git a/include/asm-x86_64/a.out.h b/include/asm-x86_64/a.out.h
index 5952914f412..7255cde0653 100644
--- a/include/asm-x86_64/a.out.h
+++ b/include/asm-x86_64/a.out.h
@@ -21,7 +21,7 @@ struct exec
 
 #ifdef __KERNEL__
 #include <linux/thread_info.h>
-#define STACK_TOP (test_thread_flag(TIF_IA32) ? IA32_PAGE_OFFSET : TASK_SIZE)
+#define STACK_TOP TASK_SIZE
 #endif
 
 #endif /* __A_OUT_GNU_H__ */
diff --git a/include/asm-x86_64/processor.h b/include/asm-x86_64/processor.h
index d641b19f6da..8b55f139968 100644
--- a/include/asm-x86_64/processor.h
+++ b/include/asm-x86_64/processor.h
@@ -160,16 +160,17 @@ static inline void clear_in_cr4 (unsigned long mask)
 /*
  * User space process size. 47bits minus one guard page.
  */
-#define TASK_SIZE	(0x800000000000UL - 4096)
+#define TASK_SIZE64	(0x800000000000UL - 4096)
 
 /* This decides where the kernel will search for a free chunk of vm
  * space during mmap's.
  */
 #define IA32_PAGE_OFFSET ((current->personality & ADDR_LIMIT_3GB) ? 0xc0000000 : 0xFFFFe000)
-#define TASK_UNMAPPED_32 PAGE_ALIGN(IA32_PAGE_OFFSET/3)
-#define TASK_UNMAPPED_64 PAGE_ALIGN(TASK_SIZE/3) 
-#define TASK_UNMAPPED_BASE	\
-	(test_thread_flag(TIF_IA32) ? TASK_UNMAPPED_32 : TASK_UNMAPPED_64)  
+
+#define TASK_SIZE 		(test_thread_flag(TIF_IA32) ? IA32_PAGE_OFFSET : TASK_SIZE64)
+#define TASK_SIZE_OF(child) 	((test_tsk_thread_flag(child, TIF_IA32)) ? IA32_PAGE_OFFSET : TASK_SIZE64)
+
+#define TASK_UNMAPPED_BASE	PAGE_ALIGN(TASK_SIZE/3)
 
 /*
  * Size of io_bitmap.
-- 
cgit v1.2.3-70-g09d2


From 39c715b71740c4a78ba4769fb54826929bac03cb Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Tue, 21 Jun 2005 17:14:34 -0700
Subject: [PATCH] smp_processor_id() cleanup

This patch implements a number of smp_processor_id() cleanup ideas that
Arjan van de Ven and I came up with.

The previous __smp_processor_id/_smp_processor_id/smp_processor_id API
spaghetti was hard to follow both on the implementational and on the
usage side.

Some of the complexity arose from picking wrong names, some of the
complexity comes from the fact that not all architectures defined
__smp_processor_id.

In the new code, there are two externally visible symbols:

 - smp_processor_id(): debug variant.

 - raw_smp_processor_id(): nondebug variant. Replaces all existing
   uses of _smp_processor_id() and __smp_processor_id(). Defined
   by every SMP architecture in include/asm-*/smp.h.

There is one new internal symbol, dependent on DEBUG_PREEMPT:

 - debug_smp_processor_id(): internal debug variant, mapped to
                             smp_processor_id().

Also, i moved debug_smp_processor_id() from lib/kernel_lock.c into a new
lib/smp_processor_id.c file.  All related comments got updated and/or
clarified.

I have build/boot tested the following 8 .config combinations on x86:

 {SMP,UP} x {PREEMPT,!PREEMPT} x {DEBUG_PREEMPT,!DEBUG_PREEMPT}

I have also build/boot tested x64 on UP/PREEMPT/DEBUG_PREEMPT.  (Other
architectures are untested, but should work just fine.)

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/traps.c          |  2 +-
 arch/i386/lib/delay.c             |  2 +-
 arch/ppc/lib/locks.c              |  4 +--
 arch/ppc64/kernel/idle.c          |  2 +-
 arch/sh/lib/delay.c               |  2 +-
 arch/sparc64/lib/delay.c          |  2 +-
 arch/x86_64/lib/delay.c           |  2 +-
 drivers/acpi/processor_idle.c     |  2 +-
 drivers/input/gameport/gameport.c |  2 +-
 drivers/oprofile/buffer_sync.c    |  4 +--
 fs/xfs/linux-2.6/xfs_linux.h      |  6 ++---
 include/asm-alpha/smp.h           |  2 +-
 include/asm-arm/smp.h             |  2 +-
 include/asm-i386/smp.h            |  2 +-
 include/asm-ia64/smp.h            |  2 +-
 include/asm-m32r/smp.h            |  2 +-
 include/asm-mips/smp.h            |  2 +-
 include/asm-parisc/smp.h          |  2 +-
 include/asm-ppc/smp.h             |  2 +-
 include/asm-ppc64/smp.h           |  2 +-
 include/asm-s390/smp.h            |  2 +-
 include/asm-sh/smp.h              |  2 +-
 include/asm-sparc/smp.h           |  2 +-
 include/asm-sparc64/smp.h         |  2 +-
 include/asm-um/smp.h              |  3 ++-
 include/asm-x86_64/smp.h          |  2 +-
 include/linux/mmzone.h            |  2 +-
 include/linux/smp.h               | 40 ++++++++++++----------------
 include/net/route.h               |  2 +-
 include/net/snmp.h                | 14 +++++-----
 kernel/module.c                   |  2 +-
 kernel/power/smp.c                |  4 +--
 kernel/sched.c                    |  4 +--
 kernel/stop_machine.c             |  4 +--
 lib/Makefile                      |  1 +
 lib/kernel_lock.c                 | 55 ---------------------------------------
 lib/smp_processor_id.c            | 55 +++++++++++++++++++++++++++++++++++++++
 37 files changed, 119 insertions(+), 125 deletions(-)
 create mode 100644 lib/smp_processor_id.c

(limited to 'include')

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 00c63419c06..83c579e82a8 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -306,7 +306,7 @@ void die(const char * str, struct pt_regs * regs, long err)
 	};
 	static int die_counter;
 
-	if (die.lock_owner != _smp_processor_id()) {
+	if (die.lock_owner != raw_smp_processor_id()) {
 		console_verbose();
 		spin_lock_irq(&die.lock);
 		die.lock_owner = smp_processor_id();
diff --git a/arch/i386/lib/delay.c b/arch/i386/lib/delay.c
index 080639f262b..eb0cdfe9280 100644
--- a/arch/i386/lib/delay.c
+++ b/arch/i386/lib/delay.c
@@ -34,7 +34,7 @@ inline void __const_udelay(unsigned long xloops)
 	xloops *= 4;
 	__asm__("mull %0"
 		:"=d" (xloops), "=&a" (d0)
-		:"1" (xloops),"0" (cpu_data[_smp_processor_id()].loops_per_jiffy * (HZ/4)));
+		:"1" (xloops),"0" (cpu_data[raw_smp_processor_id()].loops_per_jiffy * (HZ/4)));
         __delay(++xloops);
 }
 
diff --git a/arch/ppc/lib/locks.c b/arch/ppc/lib/locks.c
index 694163d696d..c450dc4b766 100644
--- a/arch/ppc/lib/locks.c
+++ b/arch/ppc/lib/locks.c
@@ -130,7 +130,7 @@ void _raw_read_lock(rwlock_t *rw)
 		while (!read_can_lock(rw)) {
 			if (--stuck == 0) {
 				printk("_read_lock(%p) CPU#%d lock %d\n",
-				       rw, _smp_processor_id(), rw->lock);
+				       rw, raw_smp_processor_id(), rw->lock);
 				stuck = INIT_STUCK;
 			}
 		}
@@ -158,7 +158,7 @@ void _raw_write_lock(rwlock_t *rw)
 		while (!write_can_lock(rw)) {
 			if (--stuck == 0) {
 				printk("write_lock(%p) CPU#%d lock %d)\n",
-				       rw, _smp_processor_id(), rw->lock);
+				       rw, raw_smp_processor_id(), rw->lock);
 				stuck = INIT_STUCK;
 			}
 		}
diff --git a/arch/ppc64/kernel/idle.c b/arch/ppc64/kernel/idle.c
index f24ce2b8720..ff8a7db142d 100644
--- a/arch/ppc64/kernel/idle.c
+++ b/arch/ppc64/kernel/idle.c
@@ -292,7 +292,7 @@ static int native_idle(void)
 		if (need_resched())
 			schedule();
 
-		if (cpu_is_offline(_smp_processor_id()) &&
+		if (cpu_is_offline(raw_smp_processor_id()) &&
 		    system_state == SYSTEM_RUNNING)
 			cpu_die();
 	}
diff --git a/arch/sh/lib/delay.c b/arch/sh/lib/delay.c
index 50b36037d86..351714694d6 100644
--- a/arch/sh/lib/delay.c
+++ b/arch/sh/lib/delay.c
@@ -24,7 +24,7 @@ inline void __const_udelay(unsigned long xloops)
 	__asm__("dmulu.l	%0, %2\n\t"
 		"sts	mach, %0"
 		: "=r" (xloops)
-		: "0" (xloops), "r" (cpu_data[_smp_processor_id()].loops_per_jiffy)
+		: "0" (xloops), "r" (cpu_data[raw_smp_processor_id()].loops_per_jiffy)
 		: "macl", "mach");
 	__delay(xloops * HZ);
 }
diff --git a/arch/sparc64/lib/delay.c b/arch/sparc64/lib/delay.c
index f6b4c784d53..e8808727617 100644
--- a/arch/sparc64/lib/delay.c
+++ b/arch/sparc64/lib/delay.c
@@ -31,7 +31,7 @@ void __const_udelay(unsigned long n)
 {
 	n *= 4;
 
-	n *= (cpu_data(_smp_processor_id()).udelay_val * (HZ/4));
+	n *= (cpu_data(raw_smp_processor_id()).udelay_val * (HZ/4));
 	n >>= 32;
 
 	__delay(n + 1);
diff --git a/arch/x86_64/lib/delay.c b/arch/x86_64/lib/delay.c
index 6e2d66472eb..aed61a668a1 100644
--- a/arch/x86_64/lib/delay.c
+++ b/arch/x86_64/lib/delay.c
@@ -34,7 +34,7 @@ void __delay(unsigned long loops)
 
 inline void __const_udelay(unsigned long xloops)
 {
-	__delay(((xloops * cpu_data[_smp_processor_id()].loops_per_jiffy) >> 32) * HZ);
+	__delay(((xloops * cpu_data[raw_smp_processor_id()].loops_per_jiffy) >> 32) * HZ);
 }
 
 void __udelay(unsigned long usecs)
diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index ff64d333e95..c9d671cf785 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -171,7 +171,7 @@ static void acpi_processor_idle (void)
 	int			sleep_ticks = 0;
 	u32			t1, t2 = 0;
 
-	pr = processors[_smp_processor_id()];
+	pr = processors[raw_smp_processor_id()];
 	if (!pr)
 		return;
 
diff --git a/drivers/input/gameport/gameport.c b/drivers/input/gameport/gameport.c
index 9b8ff396e6f..e152d0fa0cd 100644
--- a/drivers/input/gameport/gameport.c
+++ b/drivers/input/gameport/gameport.c
@@ -134,7 +134,7 @@ static int gameport_measure_speed(struct gameport *gameport)
 	}
 
 	gameport_close(gameport);
-	return (cpu_data[_smp_processor_id()].loops_per_jiffy * (unsigned long)HZ / (1000 / 50)) / (tx < 1 ? 1 : tx);
+	return (cpu_data[raw_smp_processor_id()].loops_per_jiffy * (unsigned long)HZ / (1000 / 50)) / (tx < 1 ? 1 : tx);
 
 #else
 
diff --git a/drivers/oprofile/buffer_sync.c b/drivers/oprofile/buffer_sync.c
index 55720dc6ec4..745a1418363 100644
--- a/drivers/oprofile/buffer_sync.c
+++ b/drivers/oprofile/buffer_sync.c
@@ -62,7 +62,7 @@ static int task_exit_notify(struct notifier_block * self, unsigned long val, voi
 	/* To avoid latency problems, we only process the current CPU,
 	 * hoping that most samples for the task are on this CPU
 	 */
-	sync_buffer(_smp_processor_id());
+	sync_buffer(raw_smp_processor_id());
   	return 0;
 }
 
@@ -86,7 +86,7 @@ static int munmap_notify(struct notifier_block * self, unsigned long val, void *
 		/* To avoid latency problems, we only process the current CPU,
 		 * hoping that most samples for the task are on this CPU
 		 */
-		sync_buffer(_smp_processor_id());
+		sync_buffer(raw_smp_processor_id());
 		return 0;
 	}
 
diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h
index 71bb41019a1..7d7c8788ea7 100644
--- a/fs/xfs/linux-2.6/xfs_linux.h
+++ b/fs/xfs/linux-2.6/xfs_linux.h
@@ -145,10 +145,10 @@ static inline void set_buffer_unwritten_io(struct buffer_head *bh)
 #define xfs_inherit_nosymlinks	xfs_params.inherit_nosym.val
 #define xfs_rotorstep		xfs_params.rotorstep.val
 
-#ifndef __smp_processor_id
-#define __smp_processor_id()	smp_processor_id()
+#ifndef raw_smp_processor_id
+#define raw_smp_processor_id()	smp_processor_id()
 #endif
-#define current_cpu()		__smp_processor_id()
+#define current_cpu()		raw_smp_processor_id()
 #define current_pid()		(current->pid)
 #define current_fsuid(cred)	(current->fsuid)
 #define current_fsgid(cred)	(current->fsgid)
diff --git a/include/asm-alpha/smp.h b/include/asm-alpha/smp.h
index cbc173ae45a..9950706abdf 100644
--- a/include/asm-alpha/smp.h
+++ b/include/asm-alpha/smp.h
@@ -43,7 +43,7 @@ extern struct cpuinfo_alpha cpu_data[NR_CPUS];
 #define PROC_CHANGE_PENALTY     20
 
 #define hard_smp_processor_id()	__hard_smp_processor_id()
-#define smp_processor_id()	(current_thread_info()->cpu)
+#define raw_smp_processor_id()	(current_thread_info()->cpu)
 
 extern cpumask_t cpu_present_mask;
 extern cpumask_t cpu_online_map;
diff --git a/include/asm-arm/smp.h b/include/asm-arm/smp.h
index bd44f894690..6c6c60adbba 100644
--- a/include/asm-arm/smp.h
+++ b/include/asm-arm/smp.h
@@ -21,7 +21,7 @@
 # error "<asm-arm/smp.h> included in non-SMP build"
 #endif
 
-#define smp_processor_id()	(current_thread_info()->cpu)
+#define raw_smp_processor_id() (current_thread_info()->cpu)
 
 extern cpumask_t cpu_present_mask;
 #define cpu_possible_map cpu_present_mask
diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h
index e03a206dfa3..55ef31f66bb 100644
--- a/include/asm-i386/smp.h
+++ b/include/asm-i386/smp.h
@@ -51,7 +51,7 @@ extern u8 x86_cpu_to_apicid[];
  * from the initial startup. We map APIC_BASE very early in page_setup(),
  * so this is correct in the x86 case.
  */
-#define __smp_processor_id() (current_thread_info()->cpu)
+#define raw_smp_processor_id() (current_thread_info()->cpu)
 
 extern cpumask_t cpu_callout_map;
 extern cpumask_t cpu_callin_map;
diff --git a/include/asm-ia64/smp.h b/include/asm-ia64/smp.h
index 3ba1a061e4a..a3914352c99 100644
--- a/include/asm-ia64/smp.h
+++ b/include/asm-ia64/smp.h
@@ -46,7 +46,7 @@ ia64_get_lid (void)
 #define SMP_IRQ_REDIRECTION	(1 << 0)
 #define SMP_IPI_REDIRECTION	(1 << 1)
 
-#define smp_processor_id()	(current_thread_info()->cpu)
+#define raw_smp_processor_id() (current_thread_info()->cpu)
 
 extern struct smp_boot_data {
 	int cpu_count;
diff --git a/include/asm-m32r/smp.h b/include/asm-m32r/smp.h
index 8cd4d0da4be..b9a20cdad65 100644
--- a/include/asm-m32r/smp.h
+++ b/include/asm-m32r/smp.h
@@ -66,7 +66,7 @@ extern volatile int cpu_2_physid[NR_CPUS];
 #define physid_to_cpu(physid)	physid_2_cpu[physid]
 #define cpu_to_physid(cpu_id)	cpu_2_physid[cpu_id]
 
-#define smp_processor_id()	(current_thread_info()->cpu)
+#define raw_smp_processor_id()	(current_thread_info()->cpu)
 
 extern cpumask_t cpu_callout_map;
 #define cpu_possible_map cpu_callout_map
diff --git a/include/asm-mips/smp.h b/include/asm-mips/smp.h
index 8ba370ecfd4..5618f1e12f4 100644
--- a/include/asm-mips/smp.h
+++ b/include/asm-mips/smp.h
@@ -21,7 +21,7 @@
 #include <linux/cpumask.h>
 #include <asm/atomic.h>
 
-#define smp_processor_id()	(current_thread_info()->cpu)
+#define raw_smp_processor_id() (current_thread_info()->cpu)
 
 /* Map from cpu id to sequential logical cpu number.  This will only
    not be idempotent when cpus failed to come on-line.  */
diff --git a/include/asm-parisc/smp.h b/include/asm-parisc/smp.h
index fde77ac3546..9413f67a540 100644
--- a/include/asm-parisc/smp.h
+++ b/include/asm-parisc/smp.h
@@ -51,7 +51,7 @@ extern void smp_send_reschedule(int cpu);
 
 extern unsigned long cpu_present_mask;
 
-#define smp_processor_id()	(current_thread_info()->cpu)
+#define raw_smp_processor_id()	(current_thread_info()->cpu)
 
 #endif /* CONFIG_SMP */
 
diff --git a/include/asm-ppc/smp.h b/include/asm-ppc/smp.h
index ebfb614f55f..17530c232c7 100644
--- a/include/asm-ppc/smp.h
+++ b/include/asm-ppc/smp.h
@@ -44,7 +44,7 @@ extern void smp_message_recv(int, struct pt_regs *);
 #define NO_PROC_ID		0xFF            /* No processor magic marker */
 #define PROC_CHANGE_PENALTY	20
 
-#define smp_processor_id() (current_thread_info()->cpu)
+#define raw_smp_processor_id()	(current_thread_info()->cpu)
 
 extern int __cpu_up(unsigned int cpu);
 
diff --git a/include/asm-ppc64/smp.h b/include/asm-ppc64/smp.h
index c8646fa999c..8115ecb8fee 100644
--- a/include/asm-ppc64/smp.h
+++ b/include/asm-ppc64/smp.h
@@ -45,7 +45,7 @@ void generic_cpu_die(unsigned int cpu);
 void generic_mach_cpu_die(void);
 #endif
 
-#define __smp_processor_id() (get_paca()->paca_index)
+#define raw_smp_processor_id()	(get_paca()->paca_index)
 #define hard_smp_processor_id() (get_paca()->hw_cpu_id)
 
 extern cpumask_t cpu_sibling_map[NR_CPUS];
diff --git a/include/asm-s390/smp.h b/include/asm-s390/smp.h
index 9473786387a..dd50e57a928 100644
--- a/include/asm-s390/smp.h
+++ b/include/asm-s390/smp.h
@@ -47,7 +47,7 @@ extern int smp_call_function_on(void (*func) (void *info), void *info,
  
 #define PROC_CHANGE_PENALTY	20		/* Schedule penalty */
 
-#define smp_processor_id() (S390_lowcore.cpu_data.cpu_nr)
+#define raw_smp_processor_id()	(S390_lowcore.cpu_data.cpu_nr)
 
 extern int smp_get_cpu(cpumask_t cpu_map);
 extern void smp_put_cpu(int cpu);
diff --git a/include/asm-sh/smp.h b/include/asm-sh/smp.h
index 38b54469d7d..f19a8b3b69a 100644
--- a/include/asm-sh/smp.h
+++ b/include/asm-sh/smp.h
@@ -25,7 +25,7 @@ extern cpumask_t cpu_possible_map;
 
 #define cpu_online(cpu)		cpu_isset(cpu, cpu_online_map)
 
-#define smp_processor_id()	(current_thread_info()->cpu)
+#define raw_smp_processor_id()	(current_thread_info()->cpu)
 
 /* I've no idea what the real meaning of this is */
 #define PROC_CHANGE_PENALTY	20
diff --git a/include/asm-sparc/smp.h b/include/asm-sparc/smp.h
index f986c0d0922..4f96d8333a1 100644
--- a/include/asm-sparc/smp.h
+++ b/include/asm-sparc/smp.h
@@ -148,7 +148,7 @@ extern __inline__ int hard_smp_processor_id(void)
 }
 #endif
 
-#define smp_processor_id()	(current_thread_info()->cpu)
+#define raw_smp_processor_id()		(current_thread_info()->cpu)
 
 #define prof_multiplier(__cpu)		cpu_data(__cpu).multiplier
 #define prof_counter(__cpu)		cpu_data(__cpu).counter
diff --git a/include/asm-sparc64/smp.h b/include/asm-sparc64/smp.h
index 5e3e06d908f..110a2de8912 100644
--- a/include/asm-sparc64/smp.h
+++ b/include/asm-sparc64/smp.h
@@ -64,7 +64,7 @@ static __inline__ int hard_smp_processor_id(void)
 	}
 }
 
-#define smp_processor_id() (current_thread_info()->cpu)
+#define raw_smp_processor_id() (current_thread_info()->cpu)
 
 #endif /* !(__ASSEMBLY__) */
 
diff --git a/include/asm-um/smp.h b/include/asm-um/smp.h
index 4412d5d9c26..d879eba2b52 100644
--- a/include/asm-um/smp.h
+++ b/include/asm-um/smp.h
@@ -8,7 +8,8 @@
 #include "asm/current.h"
 #include "linux/cpumask.h"
 
-#define smp_processor_id() (current_thread->cpu)
+#define raw_smp_processor_id() (current_thread->cpu)
+
 #define cpu_logical_map(n) (n)
 #define cpu_number_map(n) (n)
 #define PROC_CHANGE_PENALTY	15 /* Pick a number, any number */
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index 96844fecbde..a7425aa5a3b 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -68,7 +68,7 @@ static inline int num_booting_cpus(void)
 	return cpus_weight(cpu_callout_map);
 }
 
-#define __smp_processor_id() read_pda(cpunumber)
+#define raw_smp_processor_id() read_pda(cpunumber)
 
 extern __inline int hard_smp_processor_id(void)
 {
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index e530c6c092f..beacd931b60 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -381,7 +381,7 @@ int lowmem_reserve_ratio_sysctl_handler(struct ctl_table *, int, struct file *,
 
 #include <linux/topology.h>
 /* Returns the number of the current Node. */
-#define numa_node_id()		(cpu_to_node(_smp_processor_id()))
+#define numa_node_id()		(cpu_to_node(raw_smp_processor_id()))
 
 #ifndef CONFIG_DISCONTIGMEM
 
diff --git a/include/linux/smp.h b/include/linux/smp.h
index dcf1db3b35d..9dfa3ee769a 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -92,10 +92,7 @@ void smp_prepare_boot_cpu(void);
 /*
  *	These macros fold the SMP functionality into a single CPU system
  */
-
-#if !defined(__smp_processor_id) || !defined(CONFIG_PREEMPT)
-# define smp_processor_id()			0
-#endif
+#define raw_smp_processor_id()			0
 #define hard_smp_processor_id()			0
 #define smp_call_function(func,info,retry,wait)	({ 0; })
 #define on_each_cpu(func,info,retry,wait)	({ func(info); 0; })
@@ -106,30 +103,25 @@ static inline void smp_send_reschedule(int cpu) { }
 #endif /* !SMP */
 
 /*
- * DEBUG_PREEMPT support: check whether smp_processor_id() is being
- * used in a preemption-safe way.
+ * smp_processor_id(): get the current CPU ID.
  *
- * An architecture has to enable this debugging code explicitly.
- * It can do so by renaming the smp_processor_id() macro to
- * __smp_processor_id().  This should only be done after some minimal
- * testing, because usually there are a number of false positives
- * that an architecture will trigger.
+ * if DEBUG_PREEMPT is enabled the we check whether it is
+ * used in a preemption-safe way. (smp_processor_id() is safe
+ * if it's used in a preemption-off critical section, or in
+ * a thread that is bound to the current CPU.)
  *
- * To fix a false positive (i.e. smp_processor_id() use that the
- * debugging code reports but which use for some reason is legal),
- * change the smp_processor_id() reference to _smp_processor_id(),
- * which is the nondebug variant.  NOTE: don't use this to hack around
- * real bugs.
+ * NOTE: raw_smp_processor_id() is for internal use only
+ * (smp_processor_id() is the preferred variant), but in rare
+ * instances it might also be used to turn off false positives
+ * (i.e. smp_processor_id() use that the debugging code reports but
+ * which use for some reason is legal). Don't use this to hack around
+ * the warning message, as your code might not work under PREEMPT.
  */
-#ifdef __smp_processor_id
-# if defined(CONFIG_PREEMPT) && defined(CONFIG_DEBUG_PREEMPT)
-   extern unsigned int smp_processor_id(void);
-# else
-#  define smp_processor_id() __smp_processor_id()
-# endif
-# define _smp_processor_id() __smp_processor_id()
+#ifdef CONFIG_DEBUG_PREEMPT
+  extern unsigned int debug_smp_processor_id(void);
+# define smp_processor_id() debug_smp_processor_id()
 #else
-# define _smp_processor_id() smp_processor_id()
+# define smp_processor_id() raw_smp_processor_id()
 #endif
 
 #define get_cpu()		({ preempt_disable(); smp_processor_id(); })
diff --git a/include/net/route.h b/include/net/route.h
index d34ca8fc675..c3cd069a9ac 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -107,7 +107,7 @@ struct rt_cache_stat
 
 extern struct rt_cache_stat *rt_cache_stat;
 #define RT_CACHE_STAT_INC(field)					  \
-		(per_cpu_ptr(rt_cache_stat, _smp_processor_id())->field++)
+		(per_cpu_ptr(rt_cache_stat, raw_smp_processor_id())->field++)
 
 extern struct ip_rt_acct *ip_rt_acct;
 
diff --git a/include/net/snmp.h b/include/net/snmp.h
index a15ab256276..a36bed8ea21 100644
--- a/include/net/snmp.h
+++ b/include/net/snmp.h
@@ -128,18 +128,18 @@ struct linux_mib {
 #define SNMP_STAT_USRPTR(name)	(name[1])
 
 #define SNMP_INC_STATS_BH(mib, field) 	\
-	(per_cpu_ptr(mib[0], _smp_processor_id())->mibs[field]++)
+	(per_cpu_ptr(mib[0], raw_smp_processor_id())->mibs[field]++)
 #define SNMP_INC_STATS_OFFSET_BH(mib, field, offset)	\
-	(per_cpu_ptr(mib[0], _smp_processor_id())->mibs[field + (offset)]++)
+	(per_cpu_ptr(mib[0], raw_smp_processor_id())->mibs[field + (offset)]++)
 #define SNMP_INC_STATS_USER(mib, field) \
-	(per_cpu_ptr(mib[1], _smp_processor_id())->mibs[field]++)
+	(per_cpu_ptr(mib[1], raw_smp_processor_id())->mibs[field]++)
 #define SNMP_INC_STATS(mib, field) 	\
-	(per_cpu_ptr(mib[!in_softirq()], _smp_processor_id())->mibs[field]++)
+	(per_cpu_ptr(mib[!in_softirq()], raw_smp_processor_id())->mibs[field]++)
 #define SNMP_DEC_STATS(mib, field) 	\
-	(per_cpu_ptr(mib[!in_softirq()], _smp_processor_id())->mibs[field]--)
+	(per_cpu_ptr(mib[!in_softirq()], raw_smp_processor_id())->mibs[field]--)
 #define SNMP_ADD_STATS_BH(mib, field, addend) 	\
-	(per_cpu_ptr(mib[0], _smp_processor_id())->mibs[field] += addend)
+	(per_cpu_ptr(mib[0], raw_smp_processor_id())->mibs[field] += addend)
 #define SNMP_ADD_STATS_USER(mib, field, addend) 	\
-	(per_cpu_ptr(mib[1], _smp_processor_id())->mibs[field] += addend)
+	(per_cpu_ptr(mib[1], raw_smp_processor_id())->mibs[field] += addend)
 
 #endif
diff --git a/kernel/module.c b/kernel/module.c
index 83b3d376708..a566745dde6 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -379,7 +379,7 @@ static void module_unload_init(struct module *mod)
 	for (i = 0; i < NR_CPUS; i++)
 		local_set(&mod->ref[i].count, 0);
 	/* Hold reference count during initialization. */
-	local_set(&mod->ref[_smp_processor_id()].count, 1);
+	local_set(&mod->ref[raw_smp_processor_id()].count, 1);
 	/* Backwards compatibility macros put refcount during init. */
 	mod->waiter = current;
 }
diff --git a/kernel/power/smp.c b/kernel/power/smp.c
index cba3584b80f..457c2302ed4 100644
--- a/kernel/power/smp.c
+++ b/kernel/power/smp.c
@@ -48,11 +48,11 @@ void disable_nonboot_cpus(void)
 {
 	oldmask = current->cpus_allowed;
 	set_cpus_allowed(current, cpumask_of_cpu(0));
-	printk("Freezing CPUs (at %d)", _smp_processor_id());
+	printk("Freezing CPUs (at %d)", raw_smp_processor_id());
 	current->state = TASK_INTERRUPTIBLE;
 	schedule_timeout(HZ);
 	printk("...");
-	BUG_ON(_smp_processor_id() != 0);
+	BUG_ON(raw_smp_processor_id() != 0);
 
 	/* FIXME: for this to work, all the CPUs must be running
 	 * "idle" thread (or we deadlock). Is that guaranteed? */
diff --git a/kernel/sched.c b/kernel/sched.c
index f12a0c8a7d9..deca041fc36 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -3814,7 +3814,7 @@ EXPORT_SYMBOL(yield);
  */
 void __sched io_schedule(void)
 {
-	struct runqueue *rq = &per_cpu(runqueues, _smp_processor_id());
+	struct runqueue *rq = &per_cpu(runqueues, raw_smp_processor_id());
 
 	atomic_inc(&rq->nr_iowait);
 	schedule();
@@ -3825,7 +3825,7 @@ EXPORT_SYMBOL(io_schedule);
 
 long __sched io_schedule_timeout(long timeout)
 {
-	struct runqueue *rq = &per_cpu(runqueues, _smp_processor_id());
+	struct runqueue *rq = &per_cpu(runqueues, raw_smp_processor_id());
 	long ret;
 
 	atomic_inc(&rq->nr_iowait);
diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
index 6116b25aa7c..84a9d18aa8d 100644
--- a/kernel/stop_machine.c
+++ b/kernel/stop_machine.c
@@ -100,7 +100,7 @@ static int stop_machine(void)
 	stopmachine_state = STOPMACHINE_WAIT;
 
 	for_each_online_cpu(i) {
-		if (i == _smp_processor_id())
+		if (i == raw_smp_processor_id())
 			continue;
 		ret = kernel_thread(stopmachine, (void *)(long)i,CLONE_KERNEL);
 		if (ret < 0)
@@ -182,7 +182,7 @@ struct task_struct *__stop_machine_run(int (*fn)(void *), void *data,
 
 	/* If they don't care which CPU fn runs on, bind to any online one. */
 	if (cpu == NR_CPUS)
-		cpu = _smp_processor_id();
+		cpu = raw_smp_processor_id();
 
 	p = kthread_create(do_stop, &smdata, "kstopmachine");
 	if (!IS_ERR(p)) {
diff --git a/lib/Makefile b/lib/Makefile
index 9eccea9429a..5f10cb89840 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -20,6 +20,7 @@ lib-$(CONFIG_RWSEM_GENERIC_SPINLOCK) += rwsem-spinlock.o
 lib-$(CONFIG_RWSEM_XCHGADD_ALGORITHM) += rwsem.o
 lib-$(CONFIG_GENERIC_FIND_NEXT_BIT) += find_next_bit.o
 obj-$(CONFIG_LOCK_KERNEL) += kernel_lock.o
+obj-$(CONFIG_DEBUG_PREEMPT) += smp_processor_id.o
 
 ifneq ($(CONFIG_HAVE_DEC_LOCK),y) 
   lib-y += dec_and_lock.o
diff --git a/lib/kernel_lock.c b/lib/kernel_lock.c
index 99b0ae3d51d..bd2bc5d887b 100644
--- a/lib/kernel_lock.c
+++ b/lib/kernel_lock.c
@@ -9,61 +9,6 @@
 #include <linux/module.h>
 #include <linux/kallsyms.h>
 
-#if defined(CONFIG_PREEMPT) && defined(__smp_processor_id) && \
-		defined(CONFIG_DEBUG_PREEMPT)
-
-/*
- * Debugging check.
- */
-unsigned int smp_processor_id(void)
-{
-	unsigned long preempt_count = preempt_count();
-	int this_cpu = __smp_processor_id();
-	cpumask_t this_mask;
-
-	if (likely(preempt_count))
-		goto out;
-
-	if (irqs_disabled())
-		goto out;
-
-	/*
-	 * Kernel threads bound to a single CPU can safely use
-	 * smp_processor_id():
-	 */
-	this_mask = cpumask_of_cpu(this_cpu);
-
-	if (cpus_equal(current->cpus_allowed, this_mask))
-		goto out;
-
-	/*
-	 * It is valid to assume CPU-locality during early bootup:
-	 */
-	if (system_state != SYSTEM_RUNNING)
-		goto out;
-
-	/*
-	 * Avoid recursion:
-	 */
-	preempt_disable();
-
-	if (!printk_ratelimit())
-		goto out_enable;
-
-	printk(KERN_ERR "BUG: using smp_processor_id() in preemptible [%08x] code: %s/%d\n", preempt_count(), current->comm, current->pid);
-	print_symbol("caller is %s\n", (long)__builtin_return_address(0));
-	dump_stack();
-
-out_enable:
-	preempt_enable_no_resched();
-out:
-	return this_cpu;
-}
-
-EXPORT_SYMBOL(smp_processor_id);
-
-#endif /* PREEMPT && __smp_processor_id && DEBUG_PREEMPT */
-
 #ifdef CONFIG_PREEMPT_BKL
 /*
  * The 'big kernel semaphore'
diff --git a/lib/smp_processor_id.c b/lib/smp_processor_id.c
new file mode 100644
index 00000000000..42c08ef828c
--- /dev/null
+++ b/lib/smp_processor_id.c
@@ -0,0 +1,55 @@
+/*
+ * lib/smp_processor_id.c
+ *
+ * DEBUG_PREEMPT variant of smp_processor_id().
+ */
+#include <linux/module.h>
+#include <linux/kallsyms.h>
+
+unsigned int debug_smp_processor_id(void)
+{
+	unsigned long preempt_count = preempt_count();
+	int this_cpu = raw_smp_processor_id();
+	cpumask_t this_mask;
+
+	if (likely(preempt_count))
+		goto out;
+
+	if (irqs_disabled())
+		goto out;
+
+	/*
+	 * Kernel threads bound to a single CPU can safely use
+	 * smp_processor_id():
+	 */
+	this_mask = cpumask_of_cpu(this_cpu);
+
+	if (cpus_equal(current->cpus_allowed, this_mask))
+		goto out;
+
+	/*
+	 * It is valid to assume CPU-locality during early bootup:
+	 */
+	if (system_state != SYSTEM_RUNNING)
+		goto out;
+
+	/*
+	 * Avoid recursion:
+	 */
+	preempt_disable();
+
+	if (!printk_ratelimit())
+		goto out_enable;
+
+	printk(KERN_ERR "BUG: using smp_processor_id() in preemptible [%08x] code: %s/%d\n", preempt_count(), current->comm, current->pid);
+	print_symbol("caller is %s\n", (long)__builtin_return_address(0));
+	dump_stack();
+
+out_enable:
+	preempt_enable_no_resched();
+out:
+	return this_cpu;
+}
+
+EXPORT_SYMBOL(debug_smp_processor_id);
+
-- 
cgit v1.2.3-70-g09d2


From 753ee728964e5afb80c17659cc6c3a6fd0a42fe0 Mon Sep 17 00:00:00 2001
From: Martin Hicks <mort@sgi.com>
Date: Tue, 21 Jun 2005 17:14:41 -0700
Subject: [PATCH] VM: early zone reclaim

This is the core of the (much simplified) early reclaim.  The goal of this
patch is to reclaim some easily-freed pages from a zone before falling back
onto another zone.

One of the major uses of this is NUMA machines.  With the default allocator
behavior the allocator would look for memory in another zone, which might be
off-node, before trying to reclaim from the current zone.

This adds a zone tuneable to enable early zone reclaim.  It is selected on a
per-zone basis and is turned on/off via syscall.

Adding some extra throttling on the reclaim was also required (patch
4/4).  Without the machine would grind to a crawl when doing a "make -j"
kernel build.  Even with this patch the System Time is higher on
average, but it seems tolerable.  Here are some numbers for kernbench
runs on a 2-node, 4cpu, 8Gig RAM Altix in the "make -j" run:

			wall  user   sys   %cpu  ctx sw.  sleeps
			----  ----   ---   ----   ------  ------
No patch		1009  1384   847   258   298170   504402
w/patch, no reclaim     880   1376   667   288   254064   396745
w/patch & reclaim       1079  1385   926   252   291625   548873

These numbers are the average of 2 runs of 3 "make -j" runs done right
after system boot.  Run-to-run variability for "make -j" is huge, so
these numbers aren't terribly useful except to seee that with reclaim
the benchmark still finishes in a reasonable amount of time.

I also looked at the NUMA hit/miss stats for the "make -j" runs and the
reclaim doesn't make any difference when the machine is thrashing away.

Doing a "make -j8" on a single node that is filled with page cache pages
takes 700 seconds with reclaim turned on and 735 seconds without reclaim
(due to remote memory accesses).

The simple zone_reclaim syscall program is at
http://www.bork.org/~mort/sgi/zone_reclaim.c

Signed-off-by: Martin Hicks <mort@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/syscall_table.S |  2 +-
 arch/ia64/kernel/entry.S         |  2 +-
 include/asm-i386/unistd.h        |  2 +-
 include/asm-ia64/unistd.h        |  1 +
 include/linux/mmzone.h           |  6 ++++
 include/linux/swap.h             |  1 +
 kernel/sys_ni.c                  |  1 +
 mm/page_alloc.c                  | 33 +++++++++++++++++----
 mm/vmscan.c                      | 64 ++++++++++++++++++++++++++++++++++++++++
 9 files changed, 104 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S
index 6cd1ed311f0..d408afaf649 100644
--- a/arch/i386/kernel/syscall_table.S
+++ b/arch/i386/kernel/syscall_table.S
@@ -251,7 +251,7 @@ ENTRY(sys_call_table)
 	.long sys_io_submit
 	.long sys_io_cancel
 	.long sys_fadvise64	/* 250 */
-	.long sys_ni_syscall
+	.long sys_set_zone_reclaim
 	.long sys_exit_group
 	.long sys_lookup_dcookie
 	.long sys_epoll_create
diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index d99316c9be2..b1d5d3d5276 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -1579,7 +1579,7 @@ sys_call_table:
 	data8 sys_keyctl
 	data8 sys_ni_syscall
 	data8 sys_ni_syscall			// 1275
-	data8 sys_ni_syscall
+	data8 sys_set_zone_reclaim
 	data8 sys_ni_syscall
 	data8 sys_ni_syscall
 	data8 sys_ni_syscall
diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h
index 61bcc1b1e3f..176413fb9ae 100644
--- a/include/asm-i386/unistd.h
+++ b/include/asm-i386/unistd.h
@@ -256,7 +256,7 @@
 #define __NR_io_submit		248
 #define __NR_io_cancel		249
 #define __NR_fadvise64		250
-
+#define __NR_set_zone_reclaim	251
 #define __NR_exit_group		252
 #define __NR_lookup_dcookie	253
 #define __NR_epoll_create	254
diff --git a/include/asm-ia64/unistd.h b/include/asm-ia64/unistd.h
index 33e26c557c5..f7f43ec2483 100644
--- a/include/asm-ia64/unistd.h
+++ b/include/asm-ia64/unistd.h
@@ -263,6 +263,7 @@
 #define __NR_add_key			1271
 #define __NR_request_key		1272
 #define __NR_keyctl			1273
+#define __NR_set_zone_reclaim		1276
 
 #ifdef __KERNEL__
 
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index beacd931b60..dfc2452ccb1 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -144,6 +144,12 @@ struct zone {
 	unsigned long		pages_scanned;	   /* since last reclaim */
 	int			all_unreclaimable; /* All pages pinned */
 
+	/*
+	 * Does the allocator try to reclaim pages from the zone as soon
+	 * as it fails a watermark_ok() in __alloc_pages?
+	 */
+	int			reclaim_pages;
+
 	/*
 	 * prev_priority holds the scanning priority for this zone.  It is
 	 * defined as the scanning priority at which we achieved our reclaim
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 3bbc41be9bd..0d21e682d99 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -173,6 +173,7 @@ extern void swap_setup(void);
 
 /* linux/mm/vmscan.c */
 extern int try_to_free_pages(struct zone **, unsigned int, unsigned int);
+extern int zone_reclaim(struct zone *, unsigned int, unsigned int);
 extern int shrink_all_memory(int);
 extern int vm_swappiness;
 
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 0dda70ed1f9..6f15bea7d1a 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -77,6 +77,7 @@ cond_syscall(sys_request_key);
 cond_syscall(sys_keyctl);
 cond_syscall(compat_sys_keyctl);
 cond_syscall(compat_sys_socketcall);
+cond_syscall(sys_set_zone_reclaim);
 
 /* arch-specific weak syscall entries */
 cond_syscall(sys_pciconfig_read);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 40169f0b7e9..3c0f69ded6b 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -724,6 +724,14 @@ int zone_watermark_ok(struct zone *z, int order, unsigned long mark,
 	return 1;
 }
 
+static inline int
+should_reclaim_zone(struct zone *z, unsigned int gfp_mask)
+{
+	if (!z->reclaim_pages)
+		return 0;
+	return 1;
+}
+
 /*
  * This is the 'heart' of the zoned buddy allocator.
  */
@@ -760,17 +768,32 @@ __alloc_pages(unsigned int __nocast gfp_mask, unsigned int order,
 
 	classzone_idx = zone_idx(zones[0]);
 
- restart:
+restart:
 	/* Go through the zonelist once, looking for a zone with enough free */
 	for (i = 0; (z = zones[i]) != NULL; i++) {
-
-		if (!zone_watermark_ok(z, order, z->pages_low,
-				       classzone_idx, 0, 0))
-			continue;
+		int do_reclaim = should_reclaim_zone(z, gfp_mask);
 
 		if (!cpuset_zone_allowed(z))
 			continue;
 
+		/*
+		 * If the zone is to attempt early page reclaim then this loop
+		 * will try to reclaim pages and check the watermark a second
+		 * time before giving up and falling back to the next zone.
+		 */
+zone_reclaim_retry:
+		if (!zone_watermark_ok(z, order, z->pages_low,
+				       classzone_idx, 0, 0)) {
+			if (!do_reclaim)
+				continue;
+			else {
+				zone_reclaim(z, gfp_mask, order);
+				/* Only try reclaim once */
+				do_reclaim = 0;
+				goto zone_reclaim_retry;
+			}
+		}
+
 		page = buffered_rmqueue(z, order, gfp_mask);
 		if (page)
 			goto got_pg;
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 6379ddbffd9..7da846960d8 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1323,3 +1323,67 @@ static int __init kswapd_init(void)
 }
 
 module_init(kswapd_init)
+
+
+/*
+ * Try to free up some pages from this zone through reclaim.
+ */
+int zone_reclaim(struct zone *zone, unsigned int gfp_mask, unsigned int order)
+{
+	struct scan_control sc;
+	int nr_pages = 1 << order;
+	int total_reclaimed = 0;
+
+	/* The reclaim may sleep, so don't do it if sleep isn't allowed */
+	if (!(gfp_mask & __GFP_WAIT))
+		return 0;
+	if (zone->all_unreclaimable)
+		return 0;
+
+	sc.gfp_mask = gfp_mask;
+	sc.may_writepage = 0;
+	sc.may_swap = 0;
+	sc.nr_mapped = read_page_state(nr_mapped);
+	sc.nr_scanned = 0;
+	sc.nr_reclaimed = 0;
+	/* scan at the highest priority */
+	sc.priority = 0;
+
+	if (nr_pages > SWAP_CLUSTER_MAX)
+		sc.swap_cluster_max = nr_pages;
+	else
+		sc.swap_cluster_max = SWAP_CLUSTER_MAX;
+
+	shrink_zone(zone, &sc);
+	total_reclaimed = sc.nr_reclaimed;
+
+	return total_reclaimed;
+}
+
+asmlinkage long sys_set_zone_reclaim(unsigned int node, unsigned int zone,
+				     unsigned int state)
+{
+	struct zone *z;
+	int i;
+
+	if (node >= MAX_NUMNODES || !node_online(node))
+		return -EINVAL;
+
+	/* This will break if we ever add more zones */
+	if (!(zone & (1<<ZONE_DMA|1<<ZONE_NORMAL|1<<ZONE_HIGHMEM)))
+		return -EINVAL;
+
+	for (i = 0; i < MAX_NR_ZONES; i++) {
+		if (!(zone & 1<<i))
+			continue;
+
+		z = &NODE_DATA(node)->node_zones[i];
+
+		if (state)
+			z->reclaim_pages = 1;
+		else
+			z->reclaim_pages = 0;
+	}
+
+	return 0;
+}
-- 
cgit v1.2.3-70-g09d2


From 0c35bbadc59f5ed105c34471143eceb4c0dd9c95 Mon Sep 17 00:00:00 2001
From: Martin Hicks <mort@sgi.com>
Date: Tue, 21 Jun 2005 17:14:42 -0700
Subject: [PATCH] VM: add __GFP_NORECLAIM

When using the early zone reclaim, it was noticed that allocating new pages
that should be spread across the whole system caused eviction of local pages.

This adds a new GFP flag to prevent early reclaim from happening during
certain allocation attempts.  The example that is implemented here is for page
cache pages.  We want page cache pages to be spread across the whole system,
and we don't want page cache pages to evict other pages to get local memory.

Signed-off-by:  Martin Hicks <mort@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/gfp.h     | 3 ++-
 include/linux/pagemap.h | 4 ++--
 mm/page_alloc.c         | 2 ++
 3 files changed, 6 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index af7407e8cfc..208535fd483 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -39,6 +39,7 @@ struct vm_area_struct;
 #define __GFP_COMP	0x4000u	/* Add compound page metadata */
 #define __GFP_ZERO	0x8000u	/* Return zeroed page on success */
 #define __GFP_NOMEMALLOC 0x10000u /* Don't use emergency reserves */
+#define __GFP_NORECLAIM  0x20000u /* No realy zone reclaim during allocation */
 
 #define __GFP_BITS_SHIFT 20	/* Room for 20 __GFP_FOO bits */
 #define __GFP_BITS_MASK ((1 << __GFP_BITS_SHIFT) - 1)
@@ -47,7 +48,7 @@ struct vm_area_struct;
 #define GFP_LEVEL_MASK (__GFP_WAIT|__GFP_HIGH|__GFP_IO|__GFP_FS| \
 			__GFP_COLD|__GFP_NOWARN|__GFP_REPEAT| \
 			__GFP_NOFAIL|__GFP_NORETRY|__GFP_NO_GROW|__GFP_COMP| \
-			__GFP_NOMEMALLOC)
+			__GFP_NOMEMALLOC|__GFP_NORECLAIM)
 
 #define GFP_ATOMIC	(__GFP_HIGH)
 #define GFP_NOIO	(__GFP_WAIT)
diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index 0422031161b..d9a25647a29 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -52,12 +52,12 @@ void release_pages(struct page **pages, int nr, int cold);
 
 static inline struct page *page_cache_alloc(struct address_space *x)
 {
-	return alloc_pages(mapping_gfp_mask(x), 0);
+	return alloc_pages(mapping_gfp_mask(x)|__GFP_NORECLAIM, 0);
 }
 
 static inline struct page *page_cache_alloc_cold(struct address_space *x)
 {
-	return alloc_pages(mapping_gfp_mask(x)|__GFP_COLD, 0);
+	return alloc_pages(mapping_gfp_mask(x)|__GFP_COLD|__GFP_NORECLAIM, 0);
 }
 
 typedef int filler_t(void *, struct page *);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 3c0f69ded6b..a9da20bc2ed 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -729,6 +729,8 @@ should_reclaim_zone(struct zone *z, unsigned int gfp_mask)
 {
 	if (!z->reclaim_pages)
 		return 0;
+	if (gfp_mask & __GFP_NORECLAIM)
+		return 0;
 	return 1;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 1e7e5a9048b30c57ba1ddaa6cdf59b21b65cde99 Mon Sep 17 00:00:00 2001
From: Martin Hicks <mort@sgi.com>
Date: Tue, 21 Jun 2005 17:14:43 -0700
Subject: [PATCH] VM: rate limit early reclaim

When early zone reclaim is turned on the LRU is scanned more frequently when a
zone is low on memory.  This limits when the zone reclaim can be called by
skipping the scan if another thread (either via kswapd or sync reclaim) is
already reclaiming from the zone.

Signed-off-by: Martin Hicks <mort@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mmzone.h |  2 ++
 mm/page_alloc.c        |  1 +
 mm/vmscan.c            | 10 ++++++++++
 3 files changed, 13 insertions(+)

(limited to 'include')

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index dfc2452ccb1..18fed8b6794 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -149,6 +149,8 @@ struct zone {
 	 * as it fails a watermark_ok() in __alloc_pages?
 	 */
 	int			reclaim_pages;
+	/* A count of how many reclaimers are scanning this zone */
+	atomic_t		reclaim_in_progress;
 
 	/*
 	 * prev_priority holds the scanning priority for this zone.  It is
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index a9da20bc2ed..2019c1b1925 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1738,6 +1738,7 @@ static void __init free_area_init_core(struct pglist_data *pgdat,
 		zone->nr_scan_inactive = 0;
 		zone->nr_active = 0;
 		zone->nr_inactive = 0;
+		atomic_set(&zone->reclaim_in_progress, -1);
 		if (!size)
 			continue;
 
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 7da846960d8..24da725a30f 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -900,7 +900,9 @@ shrink_caches(struct zone **zones, struct scan_control *sc)
 		if (zone->all_unreclaimable && sc->priority != DEF_PRIORITY)
 			continue;	/* Let kswapd poll it */
 
+		atomic_inc(&zone->reclaim_in_progress);
 		shrink_zone(zone, sc);
+		atomic_dec(&zone->reclaim_in_progress);
 	}
 }
  
@@ -1111,7 +1113,9 @@ scan:
 			sc.nr_reclaimed = 0;
 			sc.priority = priority;
 			sc.swap_cluster_max = nr_pages? nr_pages : SWAP_CLUSTER_MAX;
+			atomic_inc(&zone->reclaim_in_progress);
 			shrink_zone(zone, &sc);
+			atomic_dec(&zone->reclaim_in_progress);
 			reclaim_state->reclaimed_slab = 0;
 			nr_slab = shrink_slab(sc.nr_scanned, GFP_KERNEL,
 						lru_pages);
@@ -1354,9 +1358,15 @@ int zone_reclaim(struct zone *zone, unsigned int gfp_mask, unsigned int order)
 	else
 		sc.swap_cluster_max = SWAP_CLUSTER_MAX;
 
+	/* Don't reclaim the zone if there are other reclaimers active */
+	if (!atomic_inc_and_test(&zone->reclaim_in_progress))
+		goto out;
+
 	shrink_zone(zone, &sc);
 	total_reclaimed = sc.nr_reclaimed;
 
+ out:
+	atomic_dec(&zone->reclaim_in_progress);
 	return total_reclaimed;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 63551ae0feaaa23807ebea60de1901564bbef32e Mon Sep 17 00:00:00 2001
From: David Gibson <david@gibson.dropbear.id.au>
Date: Tue, 21 Jun 2005 17:14:44 -0700
Subject: [PATCH] Hugepage consolidation

A lot of the code in arch/*/mm/hugetlbpage.c is quite similar.  This patch
attempts to consolidate a lot of the code across the arch's, putting the
combined version in mm/hugetlb.c.  There are a couple of uglyish hacks in
order to covert all the hugepage archs, but the result is a very large
reduction in the total amount of code.  It also means things like hugepage
lazy allocation could be implemented in one place, instead of six.

Tested, at least a little, on ppc64, i386 and x86_64.

Notes:
	- this patch changes the meaning of set_huge_pte() to be more
	  analagous to set_pte()
	- does SH4 need s special huge_ptep_get_and_clear()??

Acked-by: William Lee Irwin <wli@holomorphy.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mm/hugetlbpage.c    | 170 ++----------------------------------
 arch/ia64/mm/hugetlbpage.c    | 158 +---------------------------------
 arch/ppc64/mm/hugetlbpage.c   | 180 +-------------------------------------
 arch/sh/mm/hugetlbpage.c      | 196 ++++++------------------------------------
 arch/sh64/mm/hugetlbpage.c    |  18 +++-
 arch/sparc64/mm/hugetlbpage.c | 195 ++++++-----------------------------------
 include/asm-i386/page.h       |   1 +
 include/asm-i386/pgtable.h    |   2 +-
 include/asm-ia64/pgtable.h    |   1 +
 include/asm-sh/page.h         |   1 +
 include/asm-sh/pgtable.h      |   1 +
 include/asm-sh64/page.h       |   1 +
 include/asm-sh64/pgtable.h    |   2 +
 include/asm-sparc64/page.h    |   2 +
 include/asm-sparc64/pgtable.h |   1 +
 include/asm-x86_64/page.h     |   1 +
 include/asm-x86_64/pgtable.h  |   3 +-
 include/linux/hugetlb.h       |  40 +++++++--
 mm/hugetlb.c                  | 177 +++++++++++++++++++++++++++++++++++++-
 19 files changed, 300 insertions(+), 850 deletions(-)

(limited to 'include')

diff --git a/arch/i386/mm/hugetlbpage.c b/arch/i386/mm/hugetlbpage.c
index 171fc925e1e..5aa06001a4b 100644
--- a/arch/i386/mm/hugetlbpage.c
+++ b/arch/i386/mm/hugetlbpage.c
@@ -18,7 +18,7 @@
 #include <asm/tlb.h>
 #include <asm/tlbflush.h>
 
-static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
 {
 	pgd_t *pgd;
 	pud_t *pud;
@@ -30,7 +30,7 @@ static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
 	return (pte_t *) pmd;
 }
 
-static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 {
 	pgd_t *pgd;
 	pud_t *pud;
@@ -42,21 +42,6 @@ static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 	return (pte_t *) pmd;
 }
 
-static void set_huge_pte(struct mm_struct *mm, struct vm_area_struct *vma, struct page *page, pte_t * page_table, int write_access)
-{
-	pte_t entry;
-
-	add_mm_counter(mm, rss, HPAGE_SIZE / PAGE_SIZE);
-	if (write_access) {
-		entry =
-		    pte_mkwrite(pte_mkdirty(mk_pte(page, vma->vm_page_prot)));
-	} else
-		entry = pte_wrprotect(mk_pte(page, vma->vm_page_prot));
-	entry = pte_mkyoung(entry);
-	mk_pte_huge(entry);
-	set_pte(page_table, entry);
-}
-
 /*
  * This function checks for proper alignment of input addr and len parameters.
  */
@@ -69,77 +54,6 @@ int is_aligned_hugepage_range(unsigned long addr, unsigned long len)
 	return 0;
 }
 
-int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
-			struct vm_area_struct *vma)
-{
-	pte_t *src_pte, *dst_pte, entry;
-	struct page *ptepage;
-	unsigned long addr = vma->vm_start;
-	unsigned long end = vma->vm_end;
-
-	while (addr < end) {
-		dst_pte = huge_pte_alloc(dst, addr);
-		if (!dst_pte)
-			goto nomem;
-		src_pte = huge_pte_offset(src, addr);
-		entry = *src_pte;
-		ptepage = pte_page(entry);
-		get_page(ptepage);
-		set_pte(dst_pte, entry);
-		add_mm_counter(dst, rss, HPAGE_SIZE / PAGE_SIZE);
-		addr += HPAGE_SIZE;
-	}
-	return 0;
-
-nomem:
-	return -ENOMEM;
-}
-
-int
-follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
-		    struct page **pages, struct vm_area_struct **vmas,
-		    unsigned long *position, int *length, int i)
-{
-	unsigned long vpfn, vaddr = *position;
-	int remainder = *length;
-
-	WARN_ON(!is_vm_hugetlb_page(vma));
-
-	vpfn = vaddr/PAGE_SIZE;
-	while (vaddr < vma->vm_end && remainder) {
-
-		if (pages) {
-			pte_t *pte;
-			struct page *page;
-
-			pte = huge_pte_offset(mm, vaddr);
-
-			/* hugetlb should be locked, and hence, prefaulted */
-			WARN_ON(!pte || pte_none(*pte));
-
-			page = &pte_page(*pte)[vpfn % (HPAGE_SIZE/PAGE_SIZE)];
-
-			WARN_ON(!PageCompound(page));
-
-			get_page(page);
-			pages[i] = page;
-		}
-
-		if (vmas)
-			vmas[i] = vma;
-
-		vaddr += PAGE_SIZE;
-		++vpfn;
-		--remainder;
-		++i;
-	}
-
-	*length = remainder;
-	*position = vaddr;
-
-	return i;
-}
-
 #if 0	/* This is just for testing */
 struct page *
 follow_huge_addr(struct mm_struct *mm, unsigned long address, int write)
@@ -204,83 +118,15 @@ follow_huge_pmd(struct mm_struct *mm, unsigned long address,
 }
 #endif
 
-void unmap_hugepage_range(struct vm_area_struct *vma,
-		unsigned long start, unsigned long end)
+void hugetlb_clean_stale_pgtable(pte_t *pte)
 {
-	struct mm_struct *mm = vma->vm_mm;
-	unsigned long address;
-	pte_t pte, *ptep;
+	pmd_t *pmd = (pmd_t *) pte;
 	struct page *page;
 
-	BUG_ON(start & (HPAGE_SIZE - 1));
-	BUG_ON(end & (HPAGE_SIZE - 1));
-
-	for (address = start; address < end; address += HPAGE_SIZE) {
-		ptep = huge_pte_offset(mm, address);
-		if (!ptep)
-			continue;
-		pte = ptep_get_and_clear(mm, address, ptep);
-		if (pte_none(pte))
-			continue;
-		page = pte_page(pte);
-		put_page(page);
-	}
-	add_mm_counter(mm ,rss, -((end - start) >> PAGE_SHIFT));
-	flush_tlb_range(vma, start, end);
-}
-
-int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma)
-{
-	struct mm_struct *mm = current->mm;
-	unsigned long addr;
-	int ret = 0;
-
-	BUG_ON(vma->vm_start & ~HPAGE_MASK);
-	BUG_ON(vma->vm_end & ~HPAGE_MASK);
-
-	spin_lock(&mm->page_table_lock);
-	for (addr = vma->vm_start; addr < vma->vm_end; addr += HPAGE_SIZE) {
-		unsigned long idx;
-		pte_t *pte = huge_pte_alloc(mm, addr);
-		struct page *page;
-
-		if (!pte) {
-			ret = -ENOMEM;
-			goto out;
-		}
-
-		if (!pte_none(*pte))
-			continue;
-
-		idx = ((addr - vma->vm_start) >> HPAGE_SHIFT)
-			+ (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT));
-		page = find_get_page(mapping, idx);
-		if (!page) {
-			/* charge the fs quota first */
-			if (hugetlb_get_quota(mapping)) {
-				ret = -ENOMEM;
-				goto out;
-			}
-			page = alloc_huge_page();
-			if (!page) {
-				hugetlb_put_quota(mapping);
-				ret = -ENOMEM;
-				goto out;
-			}
-			ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC);
-			if (! ret) {
-				unlock_page(page);
-			} else {
-				hugetlb_put_quota(mapping);
-				free_huge_page(page);
-				goto out;
-			}
-		}
-		set_huge_pte(mm, vma, page, pte, vma->vm_flags & VM_WRITE);
-	}
-out:
-	spin_unlock(&mm->page_table_lock);
-	return ret;
+	page = pmd_page(*pmd);
+	pmd_clear(pmd);
+	dec_page_state(nr_page_table_pages);
+	page_cache_release(page);
 }
 
 /* x86_64 also uses this file */
diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c
index df08ae7634b..e0a776a3044 100644
--- a/arch/ia64/mm/hugetlbpage.c
+++ b/arch/ia64/mm/hugetlbpage.c
@@ -24,7 +24,7 @@
 
 unsigned int hpage_shift=HPAGE_SHIFT_DEFAULT;
 
-static pte_t *
+pte_t *
 huge_pte_alloc (struct mm_struct *mm, unsigned long addr)
 {
 	unsigned long taddr = htlbpage_to_page(addr);
@@ -43,7 +43,7 @@ huge_pte_alloc (struct mm_struct *mm, unsigned long addr)
 	return pte;
 }
 
-static pte_t *
+pte_t *
 huge_pte_offset (struct mm_struct *mm, unsigned long addr)
 {
 	unsigned long taddr = htlbpage_to_page(addr);
@@ -67,23 +67,6 @@ huge_pte_offset (struct mm_struct *mm, unsigned long addr)
 
 #define mk_pte_huge(entry) { pte_val(entry) |= _PAGE_P; }
 
-static void
-set_huge_pte (struct mm_struct *mm, struct vm_area_struct *vma,
-	      struct page *page, pte_t * page_table, int write_access)
-{
-	pte_t entry;
-
-	add_mm_counter(mm, rss, HPAGE_SIZE / PAGE_SIZE);
-	if (write_access) {
-		entry =
-		    pte_mkwrite(pte_mkdirty(mk_pte(page, vma->vm_page_prot)));
-	} else
-		entry = pte_wrprotect(mk_pte(page, vma->vm_page_prot));
-	entry = pte_mkyoung(entry);
-	mk_pte_huge(entry);
-	set_pte(page_table, entry);
-	return;
-}
 /*
  * This function checks for proper alignment of input addr and len parameters.
  */
@@ -99,68 +82,6 @@ int is_aligned_hugepage_range(unsigned long addr, unsigned long len)
 	return 0;
 }
 
-int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
-			struct vm_area_struct *vma)
-{
-	pte_t *src_pte, *dst_pte, entry;
-	struct page *ptepage;
-	unsigned long addr = vma->vm_start;
-	unsigned long end = vma->vm_end;
-
-	while (addr < end) {
-		dst_pte = huge_pte_alloc(dst, addr);
-		if (!dst_pte)
-			goto nomem;
-		src_pte = huge_pte_offset(src, addr);
-		entry = *src_pte;
-		ptepage = pte_page(entry);
-		get_page(ptepage);
-		set_pte(dst_pte, entry);
-		add_mm_counter(dst, rss, HPAGE_SIZE / PAGE_SIZE);
-		addr += HPAGE_SIZE;
-	}
-	return 0;
-nomem:
-	return -ENOMEM;
-}
-
-int
-follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
-		    struct page **pages, struct vm_area_struct **vmas,
-		    unsigned long *st, int *length, int i)
-{
-	pte_t *ptep, pte;
-	unsigned long start = *st;
-	unsigned long pstart;
-	int len = *length;
-	struct page *page;
-
-	do {
-		pstart = start & HPAGE_MASK;
-		ptep = huge_pte_offset(mm, start);
-		pte = *ptep;
-
-back1:
-		page = pte_page(pte);
-		if (pages) {
-			page += ((start & ~HPAGE_MASK) >> PAGE_SHIFT);
-			get_page(page);
-			pages[i] = page;
-		}
-		if (vmas)
-			vmas[i] = vma;
-		i++;
-		len--;
-		start += PAGE_SIZE;
-		if (((start & HPAGE_MASK) == pstart) && len &&
-				(start < vma->vm_end))
-			goto back1;
-	} while (len && start < vma->vm_end);
-	*length = len;
-	*st = start;
-	return i;
-}
-
 struct page *follow_huge_addr(struct mm_struct *mm, unsigned long addr, int write)
 {
 	struct page *page;
@@ -212,81 +133,6 @@ void hugetlb_free_pgd_range(struct mmu_gather **tlb,
 	free_pgd_range(tlb, addr, end, floor, ceiling);
 }
 
-void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, unsigned long end)
-{
-	struct mm_struct *mm = vma->vm_mm;
-	unsigned long address;
-	pte_t *pte;
-	struct page *page;
-
-	BUG_ON(start & (HPAGE_SIZE - 1));
-	BUG_ON(end & (HPAGE_SIZE - 1));
-
-	for (address = start; address < end; address += HPAGE_SIZE) {
-		pte = huge_pte_offset(mm, address);
-		if (pte_none(*pte))
-			continue;
-		page = pte_page(*pte);
-		put_page(page);
-		pte_clear(mm, address, pte);
-	}
-	add_mm_counter(mm, rss, - ((end - start) >> PAGE_SHIFT));
-	flush_tlb_range(vma, start, end);
-}
-
-int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma)
-{
-	struct mm_struct *mm = current->mm;
-	unsigned long addr;
-	int ret = 0;
-
-	BUG_ON(vma->vm_start & ~HPAGE_MASK);
-	BUG_ON(vma->vm_end & ~HPAGE_MASK);
-
-	spin_lock(&mm->page_table_lock);
-	for (addr = vma->vm_start; addr < vma->vm_end; addr += HPAGE_SIZE) {
-		unsigned long idx;
-		pte_t *pte = huge_pte_alloc(mm, addr);
-		struct page *page;
-
-		if (!pte) {
-			ret = -ENOMEM;
-			goto out;
-		}
-		if (!pte_none(*pte))
-			continue;
-
-		idx = ((addr - vma->vm_start) >> HPAGE_SHIFT)
-			+ (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT));
-		page = find_get_page(mapping, idx);
-		if (!page) {
-			/* charge the fs quota first */
-			if (hugetlb_get_quota(mapping)) {
-				ret = -ENOMEM;
-				goto out;
-			}
-			page = alloc_huge_page();
-			if (!page) {
-				hugetlb_put_quota(mapping);
-				ret = -ENOMEM;
-				goto out;
-			}
-			ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC);
-			if (! ret) {
-				unlock_page(page);
-			} else {
-				hugetlb_put_quota(mapping);
-				page_cache_release(page);
-				goto out;
-			}
-		}
-		set_huge_pte(mm, vma, page, pte, vma->vm_flags & VM_WRITE);
-	}
-out:
-	spin_unlock(&mm->page_table_lock);
-	return ret;
-}
-
 unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr, unsigned long len,
 		unsigned long pgoff, unsigned long flags)
 {
diff --git a/arch/ppc64/mm/hugetlbpage.c b/arch/ppc64/mm/hugetlbpage.c
index d3bf86a5c1a..b4ab766f598 100644
--- a/arch/ppc64/mm/hugetlbpage.c
+++ b/arch/ppc64/mm/hugetlbpage.c
@@ -121,7 +121,7 @@ static pte_t *hugepte_alloc(struct mm_struct *mm, pud_t *dir, unsigned long addr
 	return hugepte_offset(dir, addr);
 }
 
-static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 {
 	pud_t *pud;
 
@@ -134,7 +134,7 @@ static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 	return hugepte_offset(pud, addr);
 }
 
-static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
 {
 	pud_t *pud;
 
@@ -147,25 +147,6 @@ static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
 	return hugepte_alloc(mm, pud, addr);
 }
 
-static void set_huge_pte(struct mm_struct *mm, struct vm_area_struct *vma,
-			 unsigned long addr, struct page *page,
-			 pte_t *ptep, int write_access)
-{
-	pte_t entry;
-
-	add_mm_counter(mm, rss, HPAGE_SIZE / PAGE_SIZE);
-	if (write_access) {
-		entry =
-		    pte_mkwrite(pte_mkdirty(mk_pte(page, vma->vm_page_prot)));
-	} else {
-		entry = pte_wrprotect(mk_pte(page, vma->vm_page_prot));
-	}
-	entry = pte_mkyoung(entry);
-	entry = pte_mkhuge(entry);
-
-	set_pte_at(mm, addr, ptep, entry);
-}
-
 /*
  * This function checks for proper alignment of input addr and len parameters.
  */
@@ -259,80 +240,6 @@ int prepare_hugepage_range(unsigned long addr, unsigned long len)
 	return -EINVAL;
 }
 
-int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
-			struct vm_area_struct *vma)
-{
-	pte_t *src_pte, *dst_pte, entry;
-	struct page *ptepage;
-	unsigned long addr = vma->vm_start;
-	unsigned long end = vma->vm_end;
-	int err = -ENOMEM;
-
-	while (addr < end) {
-		dst_pte = huge_pte_alloc(dst, addr);
-		if (!dst_pte)
-			goto out;
-
-		src_pte = huge_pte_offset(src, addr);
-		entry = *src_pte;
-		
-		ptepage = pte_page(entry);
-		get_page(ptepage);
-		add_mm_counter(dst, rss, HPAGE_SIZE / PAGE_SIZE);
-		set_pte_at(dst, addr, dst_pte, entry);
-
-		addr += HPAGE_SIZE;
-	}
-
-	err = 0;
- out:
-	return err;
-}
-
-int
-follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
-		    struct page **pages, struct vm_area_struct **vmas,
-		    unsigned long *position, int *length, int i)
-{
-	unsigned long vpfn, vaddr = *position;
-	int remainder = *length;
-
-	WARN_ON(!is_vm_hugetlb_page(vma));
-
-	vpfn = vaddr/PAGE_SIZE;
-	while (vaddr < vma->vm_end && remainder) {
-		if (pages) {
-			pte_t *pte;
-			struct page *page;
-
-			pte = huge_pte_offset(mm, vaddr);
-
-			/* hugetlb should be locked, and hence, prefaulted */
-			WARN_ON(!pte || pte_none(*pte));
-
-			page = &pte_page(*pte)[vpfn % (HPAGE_SIZE/PAGE_SIZE)];
-
-			WARN_ON(!PageCompound(page));
-
-			get_page(page);
-			pages[i] = page;
-		}
-
-		if (vmas)
-			vmas[i] = vma;
-
-		vaddr += PAGE_SIZE;
-		++vpfn;
-		--remainder;
-		++i;
-	}
-
-	*length = remainder;
-	*position = vaddr;
-
-	return i;
-}
-
 struct page *
 follow_huge_addr(struct mm_struct *mm, unsigned long address, int write)
 {
@@ -363,89 +270,6 @@ follow_huge_pmd(struct mm_struct *mm, unsigned long address,
 	return NULL;
 }
 
-void unmap_hugepage_range(struct vm_area_struct *vma,
-			  unsigned long start, unsigned long end)
-{
-	struct mm_struct *mm = vma->vm_mm;
-	unsigned long addr;
-	pte_t *ptep;
-	struct page *page;
-
-	WARN_ON(!is_vm_hugetlb_page(vma));
-	BUG_ON((start % HPAGE_SIZE) != 0);
-	BUG_ON((end % HPAGE_SIZE) != 0);
-
-	for (addr = start; addr < end; addr += HPAGE_SIZE) {
-		pte_t pte;
-
-		ptep = huge_pte_offset(mm, addr);
-		if (!ptep || pte_none(*ptep))
-			continue;
-
-		pte = *ptep;
-		page = pte_page(pte);
-		pte_clear(mm, addr, ptep);
-
-		put_page(page);
-	}
-	add_mm_counter(mm, rss, -((end - start) >> PAGE_SHIFT));
-	flush_tlb_pending();
-}
-
-int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma)
-{
-	struct mm_struct *mm = current->mm;
-	unsigned long addr;
-	int ret = 0;
-
-	WARN_ON(!is_vm_hugetlb_page(vma));
-	BUG_ON((vma->vm_start % HPAGE_SIZE) != 0);
-	BUG_ON((vma->vm_end % HPAGE_SIZE) != 0);
-
-	spin_lock(&mm->page_table_lock);
-	for (addr = vma->vm_start; addr < vma->vm_end; addr += HPAGE_SIZE) {
-		unsigned long idx;
-		pte_t *pte = huge_pte_alloc(mm, addr);
-		struct page *page;
-
-		if (!pte) {
-			ret = -ENOMEM;
-			goto out;
-		}
-		if (! pte_none(*pte))
-			continue;
-
-		idx = ((addr - vma->vm_start) >> HPAGE_SHIFT)
-			+ (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT));
-		page = find_get_page(mapping, idx);
-		if (!page) {
-			/* charge the fs quota first */
-			if (hugetlb_get_quota(mapping)) {
-				ret = -ENOMEM;
-				goto out;
-			}
-			page = alloc_huge_page();
-			if (!page) {
-				hugetlb_put_quota(mapping);
-				ret = -ENOMEM;
-				goto out;
-			}
-			ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC);
-			if (! ret) {
-				unlock_page(page);
-			} else {
-				hugetlb_put_quota(mapping);
-				free_huge_page(page);
-				goto out;
-			}
-		}
-		set_huge_pte(mm, vma, addr, page, pte, vma->vm_flags & VM_WRITE);
-	}
-out:
-	spin_unlock(&mm->page_table_lock);
-	return ret;
-}
-
 /* Because we have an exclusive hugepage region which lies within the
  * normal user address space, we have to take special measures to make
  * non-huge mmap()s evade the hugepage reserved regions. */
diff --git a/arch/sh/mm/hugetlbpage.c b/arch/sh/mm/hugetlbpage.c
index 1f897bab231..95bb1a6c606 100644
--- a/arch/sh/mm/hugetlbpage.c
+++ b/arch/sh/mm/hugetlbpage.c
@@ -24,7 +24,7 @@
 #include <asm/tlbflush.h>
 #include <asm/cacheflush.h>
 
-static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
 {
 	pgd_t *pgd;
 	pmd_t *pmd;
@@ -39,7 +39,7 @@ static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
 	return pte;
 }
 
-static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 {
 	pgd_t *pgd;
 	pmd_t *pmd;
@@ -56,28 +56,34 @@ static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 
 #define mk_pte_huge(entry) do { pte_val(entry) |= _PAGE_SZHUGE; } while (0)
 
-static void set_huge_pte(struct mm_struct *mm, struct vm_area_struct *vma,
-			 struct page *page, pte_t * page_table, int write_access)
+void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
+		     pte_t *ptep, pte_t entry)
 {
-	unsigned long i;
-	pte_t entry;
+	int i;
 
-	add_mm_counter(mm, rss, HPAGE_SIZE / PAGE_SIZE);
+	for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
+		set_pte_at(mm, addr, ptep, entry);
+		ptep++;
+		addr += PAGE_SIZE;
+		pte_val(entry) += PAGE_SIZE;
+	}
+}
 
-	if (write_access)
-		entry = pte_mkwrite(pte_mkdirty(mk_pte(page,
-						       vma->vm_page_prot)));
-	else
-		entry = pte_wrprotect(mk_pte(page, vma->vm_page_prot));
-	entry = pte_mkyoung(entry);
-	mk_pte_huge(entry);
+pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
+			      pte_t *ptep)
+{
+	pte_t entry;
+	int i;
 
-	for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
-		set_pte(page_table, entry);
-		page_table++;
+	entry = *ptep;
 
-		pte_val(entry) += PAGE_SIZE;
+	for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
+		pte_clear(mm, addr, ptep);
+		addr += PAGE_SIZE;
+		ptep++;
 	}
+
+	return entry;
 }
 
 /*
@@ -92,79 +98,6 @@ int is_aligned_hugepage_range(unsigned long addr, unsigned long len)
 	return 0;
 }
 
-int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
-			    struct vm_area_struct *vma)
-{
-	pte_t *src_pte, *dst_pte, entry;
-	struct page *ptepage;
-	unsigned long addr = vma->vm_start;
-	unsigned long end = vma->vm_end;
-	int i;
-
-	while (addr < end) {
-		dst_pte = huge_pte_alloc(dst, addr);
-		if (!dst_pte)
-			goto nomem;
-		src_pte = huge_pte_offset(src, addr);
-		BUG_ON(!src_pte || pte_none(*src_pte));
-		entry = *src_pte;
-		ptepage = pte_page(entry);
-		get_page(ptepage);
-		for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
-			set_pte(dst_pte, entry);
-			pte_val(entry) += PAGE_SIZE;
-			dst_pte++;
-		}
-		add_mm_counter(dst, rss, HPAGE_SIZE / PAGE_SIZE);
-		addr += HPAGE_SIZE;
-	}
-	return 0;
-
-nomem:
-	return -ENOMEM;
-}
-
-int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
-			struct page **pages, struct vm_area_struct **vmas,
-			unsigned long *position, int *length, int i)
-{
-	unsigned long vaddr = *position;
-	int remainder = *length;
-
-	WARN_ON(!is_vm_hugetlb_page(vma));
-
-	while (vaddr < vma->vm_end && remainder) {
-		if (pages) {
-			pte_t *pte;
-			struct page *page;
-
-			pte = huge_pte_offset(mm, vaddr);
-
-			/* hugetlb should be locked, and hence, prefaulted */
-			BUG_ON(!pte || pte_none(*pte));
-
-			page = pte_page(*pte);
-
-			WARN_ON(!PageCompound(page));
-
-			get_page(page);
-			pages[i] = page;
-		}
-
-		if (vmas)
-			vmas[i] = vma;
-
-		vaddr += PAGE_SIZE;
-		--remainder;
-		++i;
-	}
-
-	*length = remainder;
-	*position = vaddr;
-
-	return i;
-}
-
 struct page *follow_huge_addr(struct mm_struct *mm,
 			      unsigned long address, int write)
 {
@@ -181,84 +114,3 @@ struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address,
 {
 	return NULL;
 }
-
-void unmap_hugepage_range(struct vm_area_struct *vma,
-			  unsigned long start, unsigned long end)
-{
-	struct mm_struct *mm = vma->vm_mm;
-	unsigned long address;
-	pte_t *pte;
-	struct page *page;
-	int i;
-
-	BUG_ON(start & (HPAGE_SIZE - 1));
-	BUG_ON(end & (HPAGE_SIZE - 1));
-
-	for (address = start; address < end; address += HPAGE_SIZE) {
-		pte = huge_pte_offset(mm, address);
-		BUG_ON(!pte);
-		if (pte_none(*pte))
-			continue;
-		page = pte_page(*pte);
-		put_page(page);
-		for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
-			pte_clear(mm, address+(i*PAGE_SIZE), pte);
-			pte++;
-		}
-	}
-	add_mm_counter(mm, rss, -((end - start) >> PAGE_SHIFT));
-	flush_tlb_range(vma, start, end);
-}
-
-int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma)
-{
-	struct mm_struct *mm = current->mm;
-	unsigned long addr;
-	int ret = 0;
-
-	BUG_ON(vma->vm_start & ~HPAGE_MASK);
-	BUG_ON(vma->vm_end & ~HPAGE_MASK);
-
-	spin_lock(&mm->page_table_lock);
-	for (addr = vma->vm_start; addr < vma->vm_end; addr += HPAGE_SIZE) {
-		unsigned long idx;
-		pte_t *pte = huge_pte_alloc(mm, addr);
-		struct page *page;
-
-		if (!pte) {
-			ret = -ENOMEM;
-			goto out;
-		}
-		if (!pte_none(*pte))
-			continue;
-
-		idx = ((addr - vma->vm_start) >> HPAGE_SHIFT)
-			+ (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT));
-		page = find_get_page(mapping, idx);
-		if (!page) {
-			/* charge the fs quota first */
-			if (hugetlb_get_quota(mapping)) {
-				ret = -ENOMEM;
-				goto out;
-			}
-			page = alloc_huge_page();
-			if (!page) {
-				hugetlb_put_quota(mapping);
-				ret = -ENOMEM;
-				goto out;
-			}
-			ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC);
-			if (! ret) {
-				unlock_page(page);
-			} else {
-				hugetlb_put_quota(mapping);
-				free_huge_page(page);
-				goto out;
-			}
-		}
-		set_huge_pte(mm, vma, page, pte, vma->vm_flags & VM_WRITE);
-	}
-out:
-	spin_unlock(&mm->page_table_lock);
-	return ret;
-}
diff --git a/arch/sh64/mm/hugetlbpage.c b/arch/sh64/mm/hugetlbpage.c
index bcad2aefa4e..dcd9c8a8baf 100644
--- a/arch/sh64/mm/hugetlbpage.c
+++ b/arch/sh64/mm/hugetlbpage.c
@@ -24,7 +24,7 @@
 #include <asm/tlbflush.h>
 #include <asm/cacheflush.h>
 
-static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
 {
 	pgd_t *pgd;
 	pmd_t *pmd;
@@ -39,7 +39,7 @@ static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
 	return pte;
 }
 
-static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 {
 	pgd_t *pgd;
 	pmd_t *pmd;
@@ -80,6 +80,20 @@ static void set_huge_pte(struct mm_struct *mm, struct vm_area_struct *vma,
 	}
 }
 
+pte_t huge_ptep_get_and_clear(pte_t *ptep)
+{
+	pte_t entry;
+
+	entry = *ptep;
+
+	for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
+		pte_clear(pte);
+		pte++;
+	}
+
+	return entry;
+}
+
 /*
  * This function checks for proper alignment of input addr and len parameters.
  */
diff --git a/arch/sparc64/mm/hugetlbpage.c b/arch/sparc64/mm/hugetlbpage.c
index 5a1f831b2de..625cbb336a2 100644
--- a/arch/sparc64/mm/hugetlbpage.c
+++ b/arch/sparc64/mm/hugetlbpage.c
@@ -22,7 +22,7 @@
 #include <asm/cacheflush.h>
 #include <asm/mmu_context.h>
 
-static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
 {
 	pgd_t *pgd;
 	pud_t *pud;
@@ -41,7 +41,7 @@ static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
 	return pte;
 }
 
-static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 {
 	pgd_t *pgd;
 	pud_t *pud;
@@ -62,30 +62,34 @@ static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 
 #define mk_pte_huge(entry) do { pte_val(entry) |= _PAGE_SZHUGE; } while (0)
 
-static void set_huge_pte(struct mm_struct *mm, struct vm_area_struct *vma,
-			 unsigned long addr,
-			 struct page *page, pte_t * page_table, int write_access)
+void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
+		     pte_t *ptep, pte_t entry)
 {
-	unsigned long i;
-	pte_t entry;
+	int i;
+
+	for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
+		set_pte_at(mm, addr, ptep, entry);
+		ptep++;
+		addr += PAGE_SIZE;
+		pte_val(entry) += PAGE_SIZE;
+	}
+}
 
-	add_mm_counter(mm, rss, HPAGE_SIZE / PAGE_SIZE);
+pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
+			      pte_t *ptep)
+{
+	pte_t entry;
+	int i;
 
-	if (write_access)
-		entry = pte_mkwrite(pte_mkdirty(mk_pte(page,
-						       vma->vm_page_prot)));
-	else
-		entry = pte_wrprotect(mk_pte(page, vma->vm_page_prot));
-	entry = pte_mkyoung(entry);
-	mk_pte_huge(entry);
+	entry = *ptep;
 
 	for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
-		set_pte_at(mm, addr, page_table, entry);
-		page_table++;
+		pte_clear(mm, addr, ptep);
 		addr += PAGE_SIZE;
-
-		pte_val(entry) += PAGE_SIZE;
+		ptep++;
 	}
+
+	return entry;
 }
 
 /*
@@ -100,79 +104,6 @@ int is_aligned_hugepage_range(unsigned long addr, unsigned long len)
 	return 0;
 }
 
-int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
-			    struct vm_area_struct *vma)
-{
-	pte_t *src_pte, *dst_pte, entry;
-	struct page *ptepage;
-	unsigned long addr = vma->vm_start;
-	unsigned long end = vma->vm_end;
-	int i;
-
-	while (addr < end) {
-		dst_pte = huge_pte_alloc(dst, addr);
-		if (!dst_pte)
-			goto nomem;
-		src_pte = huge_pte_offset(src, addr);
-		BUG_ON(!src_pte || pte_none(*src_pte));
-		entry = *src_pte;
-		ptepage = pte_page(entry);
-		get_page(ptepage);
-		for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
-			set_pte_at(dst, addr, dst_pte, entry);
-			pte_val(entry) += PAGE_SIZE;
-			dst_pte++;
-			addr += PAGE_SIZE;
-		}
-		add_mm_counter(dst, rss, HPAGE_SIZE / PAGE_SIZE);
-	}
-	return 0;
-
-nomem:
-	return -ENOMEM;
-}
-
-int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
-			struct page **pages, struct vm_area_struct **vmas,
-			unsigned long *position, int *length, int i)
-{
-	unsigned long vaddr = *position;
-	int remainder = *length;
-
-	WARN_ON(!is_vm_hugetlb_page(vma));
-
-	while (vaddr < vma->vm_end && remainder) {
-		if (pages) {
-			pte_t *pte;
-			struct page *page;
-
-			pte = huge_pte_offset(mm, vaddr);
-
-			/* hugetlb should be locked, and hence, prefaulted */
-			BUG_ON(!pte || pte_none(*pte));
-
-			page = pte_page(*pte);
-
-			WARN_ON(!PageCompound(page));
-
-			get_page(page);
-			pages[i] = page;
-		}
-
-		if (vmas)
-			vmas[i] = vma;
-
-		vaddr += PAGE_SIZE;
-		--remainder;
-		++i;
-	}
-
-	*length = remainder;
-	*position = vaddr;
-
-	return i;
-}
-
 struct page *follow_huge_addr(struct mm_struct *mm,
 			      unsigned long address, int write)
 {
@@ -190,34 +121,6 @@ struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address,
 	return NULL;
 }
 
-void unmap_hugepage_range(struct vm_area_struct *vma,
-			  unsigned long start, unsigned long end)
-{
-	struct mm_struct *mm = vma->vm_mm;
-	unsigned long address;
-	pte_t *pte;
-	struct page *page;
-	int i;
-
-	BUG_ON(start & (HPAGE_SIZE - 1));
-	BUG_ON(end & (HPAGE_SIZE - 1));
-
-	for (address = start; address < end; address += HPAGE_SIZE) {
-		pte = huge_pte_offset(mm, address);
-		BUG_ON(!pte);
-		if (pte_none(*pte))
-			continue;
-		page = pte_page(*pte);
-		put_page(page);
-		for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
-			pte_clear(mm, address+(i*PAGE_SIZE), pte);
-			pte++;
-		}
-	}
-	add_mm_counter(mm, rss, -((end - start) >> PAGE_SHIFT));
-	flush_tlb_range(vma, start, end);
-}
-
 static void context_reload(void *__data)
 {
 	struct mm_struct *mm = __data;
@@ -226,12 +129,8 @@ static void context_reload(void *__data)
 		load_secondary_context(mm);
 }
 
-int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma)
+void hugetlb_prefault_arch_hook(struct mm_struct *mm)
 {
-	struct mm_struct *mm = current->mm;
-	unsigned long addr;
-	int ret = 0;
-
 	/* On UltraSPARC-III+ and later, configure the second half of
 	 * the Data-TLB for huge pages.
 	 */
@@ -261,50 +160,4 @@ int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma)
 		}
 		spin_unlock(&ctx_alloc_lock);
 	}
-
-	BUG_ON(vma->vm_start & ~HPAGE_MASK);
-	BUG_ON(vma->vm_end & ~HPAGE_MASK);
-
-	spin_lock(&mm->page_table_lock);
-	for (addr = vma->vm_start; addr < vma->vm_end; addr += HPAGE_SIZE) {
-		unsigned long idx;
-		pte_t *pte = huge_pte_alloc(mm, addr);
-		struct page *page;
-
-		if (!pte) {
-			ret = -ENOMEM;
-			goto out;
-		}
-		if (!pte_none(*pte))
-			continue;
-
-		idx = ((addr - vma->vm_start) >> HPAGE_SHIFT)
-			+ (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT));
-		page = find_get_page(mapping, idx);
-		if (!page) {
-			/* charge the fs quota first */
-			if (hugetlb_get_quota(mapping)) {
-				ret = -ENOMEM;
-				goto out;
-			}
-			page = alloc_huge_page();
-			if (!page) {
-				hugetlb_put_quota(mapping);
-				ret = -ENOMEM;
-				goto out;
-			}
-			ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC);
-			if (! ret) {
-				unlock_page(page);
-			} else {
-				hugetlb_put_quota(mapping);
-				free_huge_page(page);
-				goto out;
-			}
-		}
-		set_huge_pte(mm, vma, addr, page, pte, vma->vm_flags & VM_WRITE);
-	}
-out:
-	spin_unlock(&mm->page_table_lock);
-	return ret;
 }
diff --git a/include/asm-i386/page.h b/include/asm-i386/page.h
index ed13969fa2d..41400d342d4 100644
--- a/include/asm-i386/page.h
+++ b/include/asm-i386/page.h
@@ -68,6 +68,7 @@ typedef struct { unsigned long pgprot; } pgprot_t;
 #define HPAGE_MASK	(~(HPAGE_SIZE - 1))
 #define HUGETLB_PAGE_ORDER	(HPAGE_SHIFT - PAGE_SHIFT)
 #define HAVE_ARCH_HUGETLB_UNMAPPED_AREA
+#define ARCH_HAS_HUGETLB_CLEAN_STALE_PGTABLE
 #endif
 
 #define pgd_val(x)	((x).pgd)
diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h
index 8d60c2b4b00..e9efe148fdf 100644
--- a/include/asm-i386/pgtable.h
+++ b/include/asm-i386/pgtable.h
@@ -236,6 +236,7 @@ static inline pte_t pte_mkexec(pte_t pte)	{ (pte).pte_low |= _PAGE_USER; return
 static inline pte_t pte_mkdirty(pte_t pte)	{ (pte).pte_low |= _PAGE_DIRTY; return pte; }
 static inline pte_t pte_mkyoung(pte_t pte)	{ (pte).pte_low |= _PAGE_ACCESSED; return pte; }
 static inline pte_t pte_mkwrite(pte_t pte)	{ (pte).pte_low |= _PAGE_RW; return pte; }
+static inline pte_t pte_mkhuge(pte_t pte)	{ (pte).pte_low |= _PAGE_PRESENT | _PAGE_PSE; return pte; }
 
 #ifdef CONFIG_X86_PAE
 # include <asm/pgtable-3level.h>
@@ -275,7 +276,6 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr,
  */
 
 #define mk_pte(page, pgprot)	pfn_pte(page_to_pfn(page), (pgprot))
-#define mk_pte_huge(entry) ((entry).pte_low |= _PAGE_PRESENT | _PAGE_PSE)
 
 static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 {
diff --git a/include/asm-ia64/pgtable.h b/include/asm-ia64/pgtable.h
index fcc9c3344ab..48586e08f43 100644
--- a/include/asm-ia64/pgtable.h
+++ b/include/asm-ia64/pgtable.h
@@ -283,6 +283,7 @@ ia64_phys_addr_valid (unsigned long addr)
 #define pte_mkyoung(pte)	(__pte(pte_val(pte) | _PAGE_A))
 #define pte_mkclean(pte)	(__pte(pte_val(pte) & ~_PAGE_D))
 #define pte_mkdirty(pte)	(__pte(pte_val(pte) | _PAGE_D))
+#define pte_mkhuge(pte)		(__pte(pte_val(pte) | _PAGE_P))
 
 /*
  * Macro to a page protection value as "uncacheable".  Note that "protection" is really a
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 4c6d129e7d9..180467be8e7 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -31,6 +31,7 @@
 #define HPAGE_SIZE		(1UL << HPAGE_SHIFT)
 #define HPAGE_MASK		(~(HPAGE_SIZE-1))
 #define HUGETLB_PAGE_ORDER	(HPAGE_SHIFT-PAGE_SHIFT)
+#define ARCH_HAS_SETCLEAR_HUGE_PTE
 #endif
 
 #ifdef __KERNEL__
diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h
index cd847a47a9a..ecb909572d3 100644
--- a/include/asm-sh/pgtable.h
+++ b/include/asm-sh/pgtable.h
@@ -196,6 +196,7 @@ static inline pte_t pte_mkexec(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _
 static inline pte_t pte_mkdirty(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_DIRTY)); return pte; }
 static inline pte_t pte_mkyoung(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_ACCESSED)); return pte; }
 static inline pte_t pte_mkwrite(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_RW)); return pte; }
+static inline pte_t pte_mkhuge(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_SZHUGE)); return pte; }
 
 /*
  * Macro and implementation to make a page protection as uncachable.
diff --git a/include/asm-sh64/page.h b/include/asm-sh64/page.h
index e1f7f5a4121..d6167f1c0e9 100644
--- a/include/asm-sh64/page.h
+++ b/include/asm-sh64/page.h
@@ -41,6 +41,7 @@
 #define HPAGE_SIZE		(1UL << HPAGE_SHIFT)
 #define HPAGE_MASK		(~(HPAGE_SIZE-1))
 #define HUGETLB_PAGE_ORDER	(HPAGE_SHIFT-PAGE_SHIFT)
+#define ARCH_HAS_SETCLEAR_HUGE_PTE
 #endif
 
 #ifdef __KERNEL__
diff --git a/include/asm-sh64/pgtable.h b/include/asm-sh64/pgtable.h
index 525e1523ef5..78ac6be2d9e 100644
--- a/include/asm-sh64/pgtable.h
+++ b/include/asm-sh64/pgtable.h
@@ -430,6 +430,8 @@ extern inline pte_t pte_mkwrite(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) |
 extern inline pte_t pte_mkexec(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_EXECUTE)); return pte; }
 extern inline pte_t pte_mkdirty(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_DIRTY)); return pte; }
 extern inline pte_t pte_mkyoung(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_ACCESSED)); return pte; }
+extern inline pte_t pte_mkhuge(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_SZHUGE)); return pte; }
+
 
 /*
  * Conversion functions: convert a page and protection to a page entry.
diff --git a/include/asm-sparc64/page.h b/include/asm-sparc64/page.h
index 219ea043a14..b87dbbd64bc 100644
--- a/include/asm-sparc64/page.h
+++ b/include/asm-sparc64/page.h
@@ -95,6 +95,8 @@ typedef unsigned long pgprot_t;
 #define HPAGE_SIZE		(_AC(1,UL) << HPAGE_SHIFT)
 #define HPAGE_MASK		(~(HPAGE_SIZE - 1UL))
 #define HUGETLB_PAGE_ORDER	(HPAGE_SHIFT - PAGE_SHIFT)
+#define ARCH_HAS_SETCLEAR_HUGE_PTE
+#define ARCH_HAS_HUGETLB_PREFAULT_HOOK
 #endif
 
 #define TASK_UNMAPPED_BASE	(test_thread_flag(TIF_32BIT) ? \
diff --git a/include/asm-sparc64/pgtable.h b/include/asm-sparc64/pgtable.h
index ae2cd5b09a7..1ae00c5087f 100644
--- a/include/asm-sparc64/pgtable.h
+++ b/include/asm-sparc64/pgtable.h
@@ -286,6 +286,7 @@ static inline pte_t pte_modify(pte_t orig_pte, pgprot_t new_prot)
 #define pte_mkyoung(pte)	(__pte(pte_val(pte) | _PAGE_ACCESSED | _PAGE_R))
 #define pte_mkwrite(pte)	(__pte(pte_val(pte) | _PAGE_WRITE))
 #define pte_mkdirty(pte)	(__pte(pte_val(pte) | _PAGE_MODIFIED | _PAGE_W))
+#define pte_mkhuge(pte)		(__pte(pte_val(pte) | _PAGE_SZHUGE))
 
 /* to find an entry in a page-table-directory. */
 #define pgd_index(address)	(((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD - 1))
diff --git a/include/asm-x86_64/page.h b/include/asm-x86_64/page.h
index f43048035a0..9ce338c3a71 100644
--- a/include/asm-x86_64/page.h
+++ b/include/asm-x86_64/page.h
@@ -28,6 +28,7 @@
 #define HPAGE_SIZE	((1UL) << HPAGE_SHIFT)
 #define HPAGE_MASK	(~(HPAGE_SIZE - 1))
 #define HUGETLB_PAGE_ORDER	(HPAGE_SHIFT - PAGE_SHIFT)
+#define ARCH_HAS_HUGETLB_CLEAN_STALE_PGTABLE
 
 #ifdef __KERNEL__
 #ifndef __ASSEMBLY__
diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h
index db2a0efbf57..4eec176c3c3 100644
--- a/include/asm-x86_64/pgtable.h
+++ b/include/asm-x86_64/pgtable.h
@@ -253,6 +253,7 @@ extern inline int pte_young(pte_t pte)		{ return pte_val(pte) & _PAGE_ACCESSED;
 extern inline int pte_write(pte_t pte)		{ return pte_val(pte) & _PAGE_RW; }
 static inline int pte_file(pte_t pte)		{ return pte_val(pte) & _PAGE_FILE; }
 
+#define __LARGE_PTE (_PAGE_PSE|_PAGE_PRESENT)
 extern inline pte_t pte_rdprotect(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_USER)); return pte; }
 extern inline pte_t pte_exprotect(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_USER)); return pte; }
 extern inline pte_t pte_mkclean(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_DIRTY)); return pte; }
@@ -263,6 +264,7 @@ extern inline pte_t pte_mkexec(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _
 extern inline pte_t pte_mkdirty(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_DIRTY)); return pte; }
 extern inline pte_t pte_mkyoung(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_ACCESSED)); return pte; }
 extern inline pte_t pte_mkwrite(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_RW)); return pte; }
+extern inline pte_t pte_mkhuge(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | __LARGE_PTE)); return pte; }
 
 struct vm_area_struct;
 
@@ -290,7 +292,6 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr,
  */
 #define pgprot_noncached(prot)	(__pgprot(pgprot_val(prot) | _PAGE_PCD | _PAGE_PWT))
 
-#define __LARGE_PTE (_PAGE_PSE|_PAGE_PRESENT) 
 static inline int pmd_large(pmd_t pte) { 
 	return (pmd_val(pte) & __LARGE_PTE) == __LARGE_PTE; 
 } 	
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index 6af1ae4a821..f529d144281 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -4,6 +4,7 @@
 #ifdef CONFIG_HUGETLB_PAGE
 
 #include <linux/mempolicy.h>
+#include <asm/tlbflush.h>
 
 struct ctl_table;
 
@@ -22,12 +23,6 @@ int hugetlb_report_meminfo(char *);
 int hugetlb_report_node_meminfo(int, char *);
 int is_hugepage_mem_enough(size_t);
 unsigned long hugetlb_total_pages(void);
-struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address,
-			      int write);
-struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address,
-				pmd_t *pmd, int write);
-int is_aligned_hugepage_range(unsigned long addr, unsigned long len);
-int pmd_huge(pmd_t pmd);
 struct page *alloc_huge_page(void);
 void free_huge_page(struct page *);
 
@@ -35,6 +30,17 @@ extern unsigned long max_huge_pages;
 extern const unsigned long hugetlb_zero, hugetlb_infinity;
 extern int sysctl_hugetlb_shm_group;
 
+/* arch callbacks */
+
+pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr);
+pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr);
+struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address,
+			      int write);
+struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address,
+				pmd_t *pmd, int write);
+int is_aligned_hugepage_range(unsigned long addr, unsigned long len);
+int pmd_huge(pmd_t pmd);
+
 #ifndef ARCH_HAS_HUGEPAGE_ONLY_RANGE
 #define is_hugepage_only_range(mm, addr, len)	0
 #define hugetlb_free_pgd_range(tlb, addr, end, floor, ceiling) \
@@ -48,6 +54,28 @@ extern int sysctl_hugetlb_shm_group;
 int prepare_hugepage_range(unsigned long addr, unsigned long len);
 #endif
 
+#ifndef ARCH_HAS_SETCLEAR_HUGE_PTE
+#define set_huge_pte_at(mm, addr, ptep, pte)	set_pte_at(mm, addr, ptep, pte)
+#define huge_ptep_get_and_clear(mm, addr, ptep) ptep_get_and_clear(mm, addr, ptep)
+#else
+void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
+		     pte_t *ptep, pte_t pte);
+pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
+			      pte_t *ptep);
+#endif
+
+#ifndef ARCH_HAS_HUGETLB_PREFAULT_HOOK
+#define hugetlb_prefault_arch_hook(mm)		do { } while (0)
+#else
+void hugetlb_prefault_arch_hook(struct mm_struct *mm);
+#endif
+
+#ifndef ARCH_HAS_HUGETLB_CLEAN_STALE_PGTABLE
+#define hugetlb_clean_stale_pgtable(pte)	BUG()
+#else
+void hugetlb_clean_stale_pgtable(pte_t *pte);
+#endif
+
 #else /* !CONFIG_HUGETLB_PAGE */
 
 static inline int is_vm_hugetlb_page(struct vm_area_struct *vma)
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 4eb5ae3fbe1..fbd1111ea11 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -7,10 +7,14 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/mm.h>
-#include <linux/hugetlb.h>
 #include <linux/sysctl.h>
 #include <linux/highmem.h>
 #include <linux/nodemask.h>
+#include <linux/pagemap.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+
+#include <linux/hugetlb.h>
 
 const unsigned long hugetlb_zero = 0, hugetlb_infinity = ~0UL;
 static unsigned long nr_huge_pages, free_huge_pages;
@@ -249,6 +253,72 @@ struct vm_operations_struct hugetlb_vm_ops = {
 	.nopage = hugetlb_nopage,
 };
 
+static pte_t make_huge_pte(struct vm_area_struct *vma, struct page *page)
+{
+	pte_t entry;
+
+	if (vma->vm_flags & VM_WRITE) {
+		entry =
+		    pte_mkwrite(pte_mkdirty(mk_pte(page, vma->vm_page_prot)));
+	} else {
+		entry = pte_wrprotect(mk_pte(page, vma->vm_page_prot));
+	}
+	entry = pte_mkyoung(entry);
+	entry = pte_mkhuge(entry);
+
+	return entry;
+}
+
+int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
+			    struct vm_area_struct *vma)
+{
+	pte_t *src_pte, *dst_pte, entry;
+	struct page *ptepage;
+	unsigned long addr = vma->vm_start;
+	unsigned long end = vma->vm_end;
+
+	while (addr < end) {
+		dst_pte = huge_pte_alloc(dst, addr);
+		if (!dst_pte)
+			goto nomem;
+		src_pte = huge_pte_offset(src, addr);
+		BUG_ON(!src_pte || pte_none(*src_pte)); /* prefaulted */
+		entry = *src_pte;
+		ptepage = pte_page(entry);
+		get_page(ptepage);
+		add_mm_counter(dst, rss, HPAGE_SIZE / PAGE_SIZE);
+		set_huge_pte_at(dst, addr, dst_pte, entry);
+		addr += HPAGE_SIZE;
+	}
+	return 0;
+
+nomem:
+	return -ENOMEM;
+}
+
+void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start,
+			  unsigned long end)
+{
+	struct mm_struct *mm = vma->vm_mm;
+	unsigned long address;
+	pte_t pte;
+	struct page *page;
+
+	WARN_ON(!is_vm_hugetlb_page(vma));
+	BUG_ON(start & ~HPAGE_MASK);
+	BUG_ON(end & ~HPAGE_MASK);
+
+	for (address = start; address < end; address += HPAGE_SIZE) {
+		pte = huge_ptep_get_and_clear(mm, address, huge_pte_offset(mm, address));
+		if (pte_none(pte))
+			continue;
+		page = pte_page(pte);
+		put_page(page);
+	}
+	add_mm_counter(mm, rss,  -((end - start) >> PAGE_SHIFT));
+	flush_tlb_range(vma, start, end);
+}
+
 void zap_hugepage_range(struct vm_area_struct *vma,
 			unsigned long start, unsigned long length)
 {
@@ -258,3 +328,108 @@ void zap_hugepage_range(struct vm_area_struct *vma,
 	unmap_hugepage_range(vma, start, start + length);
 	spin_unlock(&mm->page_table_lock);
 }
+
+int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma)
+{
+	struct mm_struct *mm = current->mm;
+	unsigned long addr;
+	int ret = 0;
+
+	WARN_ON(!is_vm_hugetlb_page(vma));
+	BUG_ON(vma->vm_start & ~HPAGE_MASK);
+	BUG_ON(vma->vm_end & ~HPAGE_MASK);
+
+	hugetlb_prefault_arch_hook(mm);
+
+	spin_lock(&mm->page_table_lock);
+	for (addr = vma->vm_start; addr < vma->vm_end; addr += HPAGE_SIZE) {
+		unsigned long idx;
+		pte_t *pte = huge_pte_alloc(mm, addr);
+		struct page *page;
+
+		if (!pte) {
+			ret = -ENOMEM;
+			goto out;
+		}
+		if (! pte_none(*pte))
+			hugetlb_clean_stale_pgtable(pte);
+
+		idx = ((addr - vma->vm_start) >> HPAGE_SHIFT)
+			+ (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT));
+		page = find_get_page(mapping, idx);
+		if (!page) {
+			/* charge the fs quota first */
+			if (hugetlb_get_quota(mapping)) {
+				ret = -ENOMEM;
+				goto out;
+			}
+			page = alloc_huge_page();
+			if (!page) {
+				hugetlb_put_quota(mapping);
+				ret = -ENOMEM;
+				goto out;
+			}
+			ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC);
+			if (! ret) {
+				unlock_page(page);
+			} else {
+				hugetlb_put_quota(mapping);
+				free_huge_page(page);
+				goto out;
+			}
+		}
+		add_mm_counter(mm, rss, HPAGE_SIZE / PAGE_SIZE);
+		set_huge_pte_at(mm, addr, pte, make_huge_pte(vma, page));
+	}
+out:
+	spin_unlock(&mm->page_table_lock);
+	return ret;
+}
+
+int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
+			struct page **pages, struct vm_area_struct **vmas,
+			unsigned long *position, int *length, int i)
+{
+	unsigned long vpfn, vaddr = *position;
+	int remainder = *length;
+
+	BUG_ON(!is_vm_hugetlb_page(vma));
+
+	vpfn = vaddr/PAGE_SIZE;
+	while (vaddr < vma->vm_end && remainder) {
+
+		if (pages) {
+			pte_t *pte;
+			struct page *page;
+
+			/* Some archs (sparc64, sh*) have multiple
+			 * pte_ts to each hugepage.  We have to make
+			 * sure we get the first, for the page
+			 * indexing below to work. */
+			pte = huge_pte_offset(mm, vaddr & HPAGE_MASK);
+
+			/* hugetlb should be locked, and hence, prefaulted */
+			WARN_ON(!pte || pte_none(*pte));
+
+			page = &pte_page(*pte)[vpfn % (HPAGE_SIZE/PAGE_SIZE)];
+
+			WARN_ON(!PageCompound(page));
+
+			get_page(page);
+			pages[i] = page;
+		}
+
+		if (vmas)
+			vmas[i] = vma;
+
+		vaddr += PAGE_SIZE;
+		++vpfn;
+		--remainder;
+		++i;
+	}
+
+	*length = remainder;
+	*position = vaddr;
+
+	return i;
+}
-- 
cgit v1.2.3-70-g09d2


From e7c8d5c9955a4d2e88e36b640563f5d6d5aba48a Mon Sep 17 00:00:00 2001
From: Christoph Lameter <christoph@lameter.com>
Date: Tue, 21 Jun 2005 17:14:47 -0700
Subject: [PATCH] node local per-cpu-pages

This patch modifies the way pagesets in struct zone are managed.

Each zone has a per-cpu array of pagesets.  So any particular CPU has some
memory in each zone structure which belongs to itself.  Even if that CPU is
not local to that zone.

So the patch relocates the pagesets for each cpu to the node that is nearest
to the cpu instead of allocating the pagesets in the (possibly remote) target
zone.  This means that the operations to manage pages on remote zone can be
done with information available locally.

We play a macro trick so that non-NUMA pmachines avoid the additional
pointer chase on the page allocator fastpath.

AIM7 benchmark on a 32 CPU SGI Altix

w/o patches:
Tasks    jobs/min  jti  jobs/min/task      real       cpu
    1      484.68  100       484.6769     12.01      1.97   Fri Mar 25 11:01:42 2005
  100    27140.46   89       271.4046     21.44    148.71   Fri Mar 25 11:02:04 2005
  200    30792.02   82       153.9601     37.80    296.72   Fri Mar 25 11:02:42 2005
  300    32209.27   81       107.3642     54.21    451.34   Fri Mar 25 11:03:37 2005
  400    34962.83   78        87.4071     66.59    588.97   Fri Mar 25 11:04:44 2005
  500    31676.92   75        63.3538     91.87    742.71   Fri Mar 25 11:06:16 2005
  600    36032.69   73        60.0545     96.91    885.44   Fri Mar 25 11:07:54 2005
  700    35540.43   77        50.7720    114.63   1024.28   Fri Mar 25 11:09:49 2005
  800    33906.70   74        42.3834    137.32   1181.65   Fri Mar 25 11:12:06 2005
  900    34120.67   73        37.9119    153.51   1325.26   Fri Mar 25 11:14:41 2005
 1000    34802.37   74        34.8024    167.23   1465.26   Fri Mar 25 11:17:28 2005

with slab API changes and pageset patch:

Tasks    jobs/min  jti  jobs/min/task      real       cpu
    1      485.00  100       485.0000     12.00      1.96   Fri Mar 25 11:46:18 2005
  100    28000.96   89       280.0096     20.79    150.45   Fri Mar 25 11:46:39 2005
  200    32285.80   79       161.4290     36.05    293.37   Fri Mar 25 11:47:16 2005
  300    40424.15   84       134.7472     43.19    438.42   Fri Mar 25 11:47:59 2005
  400    39155.01   79        97.8875     59.46    590.05   Fri Mar 25 11:48:59 2005
  500    37881.25   82        75.7625     76.82    730.19   Fri Mar 25 11:50:16 2005
  600    39083.14   78        65.1386     89.35    872.79   Fri Mar 25 11:51:46 2005
  700    38627.83   77        55.1826    105.47   1022.46   Fri Mar 25 11:53:32 2005
  800    39631.94   78        49.5399    117.48   1169.94   Fri Mar 25 11:55:30 2005
  900    36903.70   79        41.0041    141.94   1310.78   Fri Mar 25 11:57:53 2005
 1000    36201.23   77        36.2012    160.77   1458.31   Fri Mar 25 12:00:34 2005

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Shobhit Dayal <shobhit@calsoftinc.com>
Signed-off-by: Shai Fultheim <Shai@Scalex86.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/base/node.c    |   2 +-
 include/linux/mm.h     |   6 ++
 include/linux/mmzone.h |  11 ++-
 init/main.c            |   1 +
 mm/mempolicy.c         |   2 +-
 mm/page_alloc.c        | 211 +++++++++++++++++++++++++++++++++++++++++--------
 6 files changed, 195 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/drivers/base/node.c b/drivers/base/node.c
index 5d4517ccc42..904b27caf69 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -87,7 +87,7 @@ static ssize_t node_read_numastat(struct sys_device * dev, char * buf)
 	for (i = 0; i < MAX_NR_ZONES; i++) {
 		struct zone *z = &pg->node_zones[i];
 		for (cpu = 0; cpu < NR_CPUS; cpu++) {
-			struct per_cpu_pageset *ps = &z->pageset[cpu];
+			struct per_cpu_pageset *ps = zone_pcp(z,cpu);
 			numa_hit += ps->numa_hit;
 			numa_miss += ps->numa_miss;
 			numa_foreign += ps->numa_foreign;
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 17518fe0b31..1813b162b0a 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -691,6 +691,12 @@ extern void show_mem(void);
 extern void si_meminfo(struct sysinfo * val);
 extern void si_meminfo_node(struct sysinfo *val, int nid);
 
+#ifdef CONFIG_NUMA
+extern void setup_per_cpu_pageset(void);
+#else
+static inline void setup_per_cpu_pageset(void) {}
+#endif
+
 /* prio_tree.c */
 void vma_prio_tree_add(struct vm_area_struct *, struct vm_area_struct *old);
 void vma_prio_tree_insert(struct vm_area_struct *, struct prio_tree_root *);
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 18fed8b6794..4733d35d822 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -63,6 +63,12 @@ struct per_cpu_pageset {
 #endif
 } ____cacheline_aligned_in_smp;
 
+#ifdef CONFIG_NUMA
+#define zone_pcp(__z, __cpu) ((__z)->pageset[(__cpu)])
+#else
+#define zone_pcp(__z, __cpu) (&(__z)->pageset[(__cpu)])
+#endif
+
 #define ZONE_DMA		0
 #define ZONE_NORMAL		1
 #define ZONE_HIGHMEM		2
@@ -122,8 +128,11 @@ struct zone {
 	 */
 	unsigned long		lowmem_reserve[MAX_NR_ZONES];
 
+#ifdef CONFIG_NUMA
+	struct per_cpu_pageset	*pageset[NR_CPUS];
+#else
 	struct per_cpu_pageset	pageset[NR_CPUS];
-
+#endif
 	/*
 	 * free areas of different sizes
 	 */
diff --git a/init/main.c b/init/main.c
index 40bf367ffdf..d324801729b 100644
--- a/init/main.c
+++ b/init/main.c
@@ -490,6 +490,7 @@ asmlinkage void __init start_kernel(void)
 	vfs_caches_init_early();
 	mem_init();
 	kmem_cache_init();
+	setup_per_cpu_pageset();
 	numa_policy_init();
 	if (late_time_init)
 		late_time_init();
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 08c41da429c..39252c732db 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -721,7 +721,7 @@ static struct page *alloc_page_interleave(unsigned int __nocast gfp, unsigned or
 	zl = NODE_DATA(nid)->node_zonelists + (gfp & GFP_ZONEMASK);
 	page = __alloc_pages(gfp, order, zl);
 	if (page && page_zone(page) == zl->zones[0]) {
-		zl->zones[0]->pageset[get_cpu()].interleave_hit++;
+		zone_pcp(zl->zones[0],get_cpu())->interleave_hit++;
 		put_cpu();
 	}
 	return page;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 2019c1b1925..95cbd30a67b 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -71,6 +71,11 @@ EXPORT_SYMBOL(nr_swap_pages);
 struct zone *zone_table[1 << (ZONES_SHIFT + NODES_SHIFT)];
 EXPORT_SYMBOL(zone_table);
 
+#ifdef CONFIG_NUMA
+static struct per_cpu_pageset
+	pageset_table[MAX_NR_ZONES*MAX_NUMNODES*NR_CPUS] __initdata;
+#endif
+
 static char *zone_names[MAX_NR_ZONES] = { "DMA", "Normal", "HighMem" };
 int min_free_kbytes = 1024;
 
@@ -520,7 +525,7 @@ static void __drain_pages(unsigned int cpu)
 	for_each_zone(zone) {
 		struct per_cpu_pageset *pset;
 
-		pset = &zone->pageset[cpu];
+		pset = zone_pcp(zone, cpu);
 		for (i = 0; i < ARRAY_SIZE(pset->pcp); i++) {
 			struct per_cpu_pages *pcp;
 
@@ -583,12 +588,12 @@ static void zone_statistics(struct zonelist *zonelist, struct zone *z)
 
 	local_irq_save(flags);
 	cpu = smp_processor_id();
-	p = &z->pageset[cpu];
+	p = zone_pcp(z,cpu);
 	if (pg == orig) {
-		z->pageset[cpu].numa_hit++;
+		p->numa_hit++;
 	} else {
 		p->numa_miss++;
-		zonelist->zones[0]->pageset[cpu].numa_foreign++;
+		zone_pcp(zonelist->zones[0], cpu)->numa_foreign++;
 	}
 	if (pg == NODE_DATA(numa_node_id()))
 		p->local_node++;
@@ -615,7 +620,7 @@ static void fastcall free_hot_cold_page(struct page *page, int cold)
 	if (PageAnon(page))
 		page->mapping = NULL;
 	free_pages_check(__FUNCTION__, page);
-	pcp = &zone->pageset[get_cpu()].pcp[cold];
+	pcp = &zone_pcp(zone, get_cpu())->pcp[cold];
 	local_irq_save(flags);
 	if (pcp->count >= pcp->high)
 		pcp->count -= free_pages_bulk(zone, pcp->batch, &pcp->list, 0);
@@ -659,7 +664,7 @@ buffered_rmqueue(struct zone *zone, int order, unsigned int __nocast gfp_flags)
 	if (order == 0) {
 		struct per_cpu_pages *pcp;
 
-		pcp = &zone->pageset[get_cpu()].pcp[cold];
+		pcp = &zone_pcp(zone, get_cpu())->pcp[cold];
 		local_irq_save(flags);
 		if (pcp->count <= pcp->low)
 			pcp->count += rmqueue_bulk(zone, 0,
@@ -1262,7 +1267,7 @@ void show_free_areas(void)
 			if (!cpu_possible(cpu))
 				continue;
 
-			pageset = zone->pageset + cpu;
+			pageset = zone_pcp(zone, cpu);
 
 			for (temperature = 0; temperature < 2; temperature++)
 				printk("cpu %d %s: low %d, high %d, batch %d\n",
@@ -1645,6 +1650,157 @@ void zone_init_free_lists(struct pglist_data *pgdat, struct zone *zone,
 	memmap_init_zone((size), (nid), (zone), (start_pfn))
 #endif
 
+static int __devinit zone_batchsize(struct zone *zone)
+{
+	int batch;
+
+	/*
+	 * The per-cpu-pages pools are set to around 1000th of the
+	 * size of the zone.  But no more than 1/4 of a meg - there's
+	 * no point in going beyond the size of L2 cache.
+	 *
+	 * OK, so we don't know how big the cache is.  So guess.
+	 */
+	batch = zone->present_pages / 1024;
+	if (batch * PAGE_SIZE > 256 * 1024)
+		batch = (256 * 1024) / PAGE_SIZE;
+	batch /= 4;		/* We effectively *= 4 below */
+	if (batch < 1)
+		batch = 1;
+
+	/*
+	 * Clamp the batch to a 2^n - 1 value. Having a power
+	 * of 2 value was found to be more likely to have
+	 * suboptimal cache aliasing properties in some cases.
+	 *
+	 * For example if 2 tasks are alternately allocating
+	 * batches of pages, one task can end up with a lot
+	 * of pages of one half of the possible page colors
+	 * and the other with pages of the other colors.
+	 */
+	batch = (1 << fls(batch + batch/2)) - 1;
+	return batch;
+}
+
+#ifdef CONFIG_NUMA
+/*
+ * Dynamicaly allocate memory for the
+ * per cpu pageset array in struct zone.
+ */
+static int __devinit process_zones(int cpu)
+{
+	struct zone *zone, *dzone;
+	int i;
+
+	for_each_zone(zone) {
+		struct per_cpu_pageset *npageset = NULL;
+
+		npageset = kmalloc_node(sizeof(struct per_cpu_pageset),
+					 GFP_KERNEL, cpu_to_node(cpu));
+		if (!npageset) {
+			zone->pageset[cpu] = NULL;
+			goto bad;
+		}
+
+		if (zone->pageset[cpu]) {
+			memcpy(npageset, zone->pageset[cpu],
+					sizeof(struct per_cpu_pageset));
+
+			/* Relocate lists */
+			for (i = 0; i < 2; i++) {
+				INIT_LIST_HEAD(&npageset->pcp[i].list);
+				list_splice(&zone->pageset[cpu]->pcp[i].list,
+					&npageset->pcp[i].list);
+			}
+ 		} else {
+			struct per_cpu_pages *pcp;
+			unsigned long batch;
+
+			batch = zone_batchsize(zone);
+
+			pcp = &npageset->pcp[0];		/* hot */
+			pcp->count = 0;
+			pcp->low = 2 * batch;
+			pcp->high = 6 * batch;
+			pcp->batch = 1 * batch;
+			INIT_LIST_HEAD(&pcp->list);
+
+			pcp = &npageset->pcp[1];		/* cold*/
+			pcp->count = 0;
+			pcp->low = 0;
+			pcp->high = 2 * batch;
+			pcp->batch = 1 * batch;
+			INIT_LIST_HEAD(&pcp->list);
+		}
+		zone->pageset[cpu] = npageset;
+	}
+
+	return 0;
+bad:
+	for_each_zone(dzone) {
+		if (dzone == zone)
+			break;
+		kfree(dzone->pageset[cpu]);
+		dzone->pageset[cpu] = NULL;
+	}
+	return -ENOMEM;
+}
+
+static inline void free_zone_pagesets(int cpu)
+{
+#ifdef CONFIG_NUMA
+	struct zone *zone;
+
+	for_each_zone(zone) {
+		struct per_cpu_pageset *pset = zone_pcp(zone, cpu);
+
+		zone_pcp(zone, cpu) = NULL;
+		kfree(pset);
+	}
+#endif
+}
+
+static int __devinit pageset_cpuup_callback(struct notifier_block *nfb,
+		unsigned long action,
+		void *hcpu)
+{
+	int cpu = (long)hcpu;
+	int ret = NOTIFY_OK;
+
+	switch (action) {
+		case CPU_UP_PREPARE:
+			if (process_zones(cpu))
+				ret = NOTIFY_BAD;
+			break;
+#ifdef CONFIG_HOTPLUG_CPU
+		case CPU_DEAD:
+			free_zone_pagesets(cpu);
+			break;
+#endif
+		default:
+			break;
+	}
+	return ret;
+}
+
+static struct notifier_block pageset_notifier =
+	{ &pageset_cpuup_callback, NULL, 0 };
+
+void __init setup_per_cpu_pageset()
+{
+	int err;
+
+	/* Initialize per_cpu_pageset for cpu 0.
+	 * A cpuup callback will do this for every cpu
+	 * as it comes online
+	 */
+	err = process_zones(smp_processor_id());
+	BUG_ON(err);
+	register_cpu_notifier(&pageset_notifier);
+}
+
+#endif
+
 /*
  * Set up the zone data structures:
  *   - mark all pages reserved
@@ -1687,43 +1843,28 @@ static void __init free_area_init_core(struct pglist_data *pgdat,
 
 		zone->temp_priority = zone->prev_priority = DEF_PRIORITY;
 
-		/*
-		 * The per-cpu-pages pools are set to around 1000th of the
-		 * size of the zone.  But no more than 1/4 of a meg - there's
-		 * no point in going beyond the size of L2 cache.
-		 *
-		 * OK, so we don't know how big the cache is.  So guess.
-		 */
-		batch = zone->present_pages / 1024;
-		if (batch * PAGE_SIZE > 256 * 1024)
-			batch = (256 * 1024) / PAGE_SIZE;
-		batch /= 4;		/* We effectively *= 4 below */
-		if (batch < 1)
-			batch = 1;
-
-		/*
-		 * Clamp the batch to a 2^n - 1 value. Having a power
-		 * of 2 value was found to be more likely to have
-		 * suboptimal cache aliasing properties in some cases.
-		 *
-		 * For example if 2 tasks are alternately allocating
-		 * batches of pages, one task can end up with a lot
-		 * of pages of one half of the possible page colors
-		 * and the other with pages of the other colors.
-		 */
-		batch = (1 << fls(batch + batch/2)) - 1;
+		batch = zone_batchsize(zone);
 
 		for (cpu = 0; cpu < NR_CPUS; cpu++) {
 			struct per_cpu_pages *pcp;
+#ifdef CONFIG_NUMA
+			struct per_cpu_pageset *pgset;
+			pgset = &pageset_table[nid*MAX_NR_ZONES*NR_CPUS +
+					(j * NR_CPUS) + cpu];
+
+			zone->pageset[cpu] = pgset;
+#else
+			struct per_cpu_pageset *pgset = zone_pcp(zone, cpu);
+#endif
 
-			pcp = &zone->pageset[cpu].pcp[0];	/* hot */
+			pcp = &pgset->pcp[0];			/* hot */
 			pcp->count = 0;
 			pcp->low = 2 * batch;
 			pcp->high = 6 * batch;
 			pcp->batch = 1 * batch;
 			INIT_LIST_HEAD(&pcp->list);
 
-			pcp = &zone->pageset[cpu].pcp[1];	/* cold */
+			pcp = &pgset->pcp[1];			/* cold */
 			pcp->count = 0;
 			pcp->low = 0;
 			pcp->high = 2 * batch;
@@ -1929,7 +2070,7 @@ static int zoneinfo_show(struct seq_file *m, void *arg)
 			struct per_cpu_pageset *pageset;
 			int j;
 
-			pageset = &zone->pageset[i];
+			pageset = zone_pcp(zone, i);
 			for (j = 0; j < ARRAY_SIZE(pageset->pcp); j++) {
 				if (pageset->pcp[j].count)
 					break;
-- 
cgit v1.2.3-70-g09d2


From 1363c3cd8603a913a27e2995dccbd70d5312d8e6 Mon Sep 17 00:00:00 2001
From: Wolfgang Wander <wwc@rentec.com>
Date: Tue, 21 Jun 2005 17:14:49 -0700
Subject: [PATCH] Avoiding mmap fragmentation

Ingo recently introduced a great speedup for allocating new mmaps using the
free_area_cache pointer which boosts the specweb SSL benchmark by 4-5% and
causes huge performance increases in thread creation.

The downside of this patch is that it does lead to fragmentation in the
mmap-ed areas (visible via /proc/self/maps), such that some applications
that work fine under 2.4 kernels quickly run out of memory on any 2.6
kernel.

The problem is twofold:

  1) the free_area_cache is used to continue a search for memory where
     the last search ended.  Before the change new areas were always
     searched from the base address on.

     So now new small areas are cluttering holes of all sizes
     throughout the whole mmap-able region whereas before small holes
     tended to close holes near the base leaving holes far from the base
     large and available for larger requests.

  2) the free_area_cache also is set to the location of the last
     munmap-ed area so in scenarios where we allocate e.g.  five regions of
     1K each, then free regions 4 2 3 in this order the next request for 1K
     will be placed in the position of the old region 3, whereas before we
     appended it to the still active region 1, placing it at the location
     of the old region 2.  Before we had 1 free region of 2K, now we only
     get two free regions of 1K -> fragmentation.

The patch addresses thes issues by introducing yet another cache descriptor
cached_hole_size that contains the largest known hole size below the
current free_area_cache.  If a new request comes in the size is compared
against the cached_hole_size and if the request can be filled with a hole
below free_area_cache the search is started from the base instead.

The results look promising: Whereas 2.6.12-rc4 fragments quickly and my
(earlier posted) leakme.c test program terminates after 50000+ iterations
with 96 distinct and fragmented maps in /proc/self/maps it performs nicely
(as expected) with thread creation, Ingo's test_str02 with 20000 threads
requires 0.7s system time.

Taking out Ingo's patch (un-patch available per request) by basically
deleting all mentions of free_area_cache from the kernel and starting the
search for new memory always at the respective bases we observe: leakme
terminates successfully with 11 distinctive hardly fragmented areas in
/proc/self/maps but thread creating is gringdingly slow: 30+s(!) system
time for Ingo's test_str02 with 20000 threads.

Now - drumroll ;-) the appended patch works fine with leakme: it ends with
only 7 distinct areas in /proc/self/maps and also thread creation seems
sufficiently fast with 0.71s for 20000 threads.

Signed-off-by: Wolfgang Wander <wwc@rentec.com>
Credit-to: "Richard Purdie" <rpurdie@rpsys.net>
Signed-off-by: Ken Chen <kenneth.w.chen@intel.com>
Acked-by: Ingo Molnar <mingo@elte.hu> (partly)
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/arm/mm/mmap.c              | 10 +++++++-
 arch/i386/mm/hugetlbpage.c      | 34 ++++++++++++++++++++++----
 arch/ppc64/mm/hugetlbpage.c     | 34 ++++++++++++++++++++++----
 arch/sh/kernel/sys_sh.c         |  8 +++++++
 arch/sparc64/kernel/sys_sparc.c |  8 +++++++
 arch/x86_64/ia32/ia32_aout.c    |  1 +
 arch/x86_64/kernel/sys_x86_64.c |  9 +++++++
 fs/binfmt_aout.c                |  1 +
 fs/binfmt_elf.c                 |  1 +
 fs/hugetlbfs/inode.c            |  3 +++
 include/linux/sched.h           | 11 +++++----
 kernel/fork.c                   |  2 ++
 mm/mmap.c                       | 53 +++++++++++++++++++++++++++++++----------
 mm/nommu.c                      |  2 +-
 14 files changed, 147 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mm/mmap.c b/arch/arm/mm/mmap.c
index 32c4b0e35b3..3de7f84b53c 100644
--- a/arch/arm/mm/mmap.c
+++ b/arch/arm/mm/mmap.c
@@ -73,7 +73,12 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr,
 		    (!vma || addr + len <= vma->vm_start))
 			return addr;
 	}
-	start_addr = addr = mm->free_area_cache;
+	if (len > mm->cached_hole_size) {
+	        start_addr = addr = mm->free_area_cache;
+	} else {
+	        start_addr = addr = TASK_UNMAPPED_BASE;
+	        mm->cached_hole_size = 0;
+	}
 
 full_search:
 	if (do_align)
@@ -90,6 +95,7 @@ full_search:
 			 */
 			if (start_addr != TASK_UNMAPPED_BASE) {
 				start_addr = addr = TASK_UNMAPPED_BASE;
+				mm->cached_hole_size = 0;
 				goto full_search;
 			}
 			return -ENOMEM;
@@ -101,6 +107,8 @@ full_search:
 			mm->free_area_cache = addr + len;
 			return addr;
 		}
+		if (addr + mm->cached_hole_size < vma->vm_start)
+		        mm->cached_hole_size = vma->vm_start - addr;
 		addr = vma->vm_end;
 		if (do_align)
 			addr = COLOUR_ALIGN(addr, pgoff);
diff --git a/arch/i386/mm/hugetlbpage.c b/arch/i386/mm/hugetlbpage.c
index 5aa06001a4b..3b099f32b94 100644
--- a/arch/i386/mm/hugetlbpage.c
+++ b/arch/i386/mm/hugetlbpage.c
@@ -140,7 +140,12 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *file,
 	struct vm_area_struct *vma;
 	unsigned long start_addr;
 
-	start_addr = mm->free_area_cache;
+	if (len > mm->cached_hole_size) {
+	        start_addr = mm->free_area_cache;
+	} else {
+	        start_addr = TASK_UNMAPPED_BASE;
+	        mm->cached_hole_size = 0;
+	}
 
 full_search:
 	addr = ALIGN(start_addr, HPAGE_SIZE);
@@ -154,6 +159,7 @@ full_search:
 			 */
 			if (start_addr != TASK_UNMAPPED_BASE) {
 				start_addr = TASK_UNMAPPED_BASE;
+				mm->cached_hole_size = 0;
 				goto full_search;
 			}
 			return -ENOMEM;
@@ -162,6 +168,8 @@ full_search:
 			mm->free_area_cache = addr + len;
 			return addr;
 		}
+		if (addr + mm->cached_hole_size < vma->vm_start)
+		        mm->cached_hole_size = vma->vm_start - addr;
 		addr = ALIGN(vma->vm_end, HPAGE_SIZE);
 	}
 }
@@ -173,12 +181,17 @@ static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file,
 	struct mm_struct *mm = current->mm;
 	struct vm_area_struct *vma, *prev_vma;
 	unsigned long base = mm->mmap_base, addr = addr0;
+	unsigned long largest_hole = mm->cached_hole_size;
 	int first_time = 1;
 
 	/* don't allow allocations above current base */
 	if (mm->free_area_cache > base)
 		mm->free_area_cache = base;
 
+	if (len <= largest_hole) {
+	        largest_hole = 0;
+		mm->free_area_cache  = base;
+	}
 try_again:
 	/* make sure it can fit in the remaining address space */
 	if (mm->free_area_cache < len)
@@ -199,13 +212,21 @@ try_again:
 		 * vma->vm_start, use it:
 		 */
 		if (addr + len <= vma->vm_start &&
-				(!prev_vma || (addr >= prev_vma->vm_end)))
+		            (!prev_vma || (addr >= prev_vma->vm_end))) {
 			/* remember the address as a hint for next time */
-			return (mm->free_area_cache = addr);
-		else
+		        mm->cached_hole_size = largest_hole;
+		        return (mm->free_area_cache = addr);
+		} else {
 			/* pull free_area_cache down to the first hole */
-			if (mm->free_area_cache == vma->vm_end)
+		        if (mm->free_area_cache == vma->vm_end) {
 				mm->free_area_cache = vma->vm_start;
+				mm->cached_hole_size = largest_hole;
+			}
+		}
+
+		/* remember the largest hole we saw so far */
+		if (addr + largest_hole < vma->vm_start)
+		        largest_hole = vma->vm_start - addr;
 
 		/* try just below the current vma->vm_start */
 		addr = (vma->vm_start - len) & HPAGE_MASK;
@@ -218,6 +239,7 @@ fail:
 	 */
 	if (first_time) {
 		mm->free_area_cache = base;
+		largest_hole = 0;
 		first_time = 0;
 		goto try_again;
 	}
@@ -228,6 +250,7 @@ fail:
 	 * allocations.
 	 */
 	mm->free_area_cache = TASK_UNMAPPED_BASE;
+	mm->cached_hole_size = ~0UL;
 	addr = hugetlb_get_unmapped_area_bottomup(file, addr0,
 			len, pgoff, flags);
 
@@ -235,6 +258,7 @@ fail:
 	 * Restore the topdown base:
 	 */
 	mm->free_area_cache = base;
+	mm->cached_hole_size = ~0UL;
 
 	return addr;
 }
diff --git a/arch/ppc64/mm/hugetlbpage.c b/arch/ppc64/mm/hugetlbpage.c
index b4ab766f598..fdcfe97c75c 100644
--- a/arch/ppc64/mm/hugetlbpage.c
+++ b/arch/ppc64/mm/hugetlbpage.c
@@ -292,7 +292,12 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr,
 		    && !is_hugepage_only_range(mm, addr,len))
 			return addr;
 	}
-	start_addr = addr = mm->free_area_cache;
+	if (len > mm->cached_hole_size) {
+	        start_addr = addr = mm->free_area_cache;
+	} else {
+	        start_addr = addr = TASK_UNMAPPED_BASE;
+	        mm->cached_hole_size = 0;
+	}
 
 full_search:
 	vma = find_vma(mm, addr);
@@ -316,6 +321,8 @@ full_search:
 			mm->free_area_cache = addr + len;
 			return addr;
 		}
+		if (addr + mm->cached_hole_size < vma->vm_start)
+		        mm->cached_hole_size = vma->vm_start - addr;
 		addr = vma->vm_end;
 		vma = vma->vm_next;
 	}
@@ -323,6 +330,7 @@ full_search:
 	/* Make sure we didn't miss any holes */
 	if (start_addr != TASK_UNMAPPED_BASE) {
 		start_addr = addr = TASK_UNMAPPED_BASE;
+		mm->cached_hole_size = 0;
 		goto full_search;
 	}
 	return -ENOMEM;
@@ -344,6 +352,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
 	struct vm_area_struct *vma, *prev_vma;
 	struct mm_struct *mm = current->mm;
 	unsigned long base = mm->mmap_base, addr = addr0;
+	unsigned long largest_hole = mm->cached_hole_size;
 	int first_time = 1;
 
 	/* requested length too big for entire address space */
@@ -364,6 +373,10 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
 			return addr;
 	}
 
+	if (len <= largest_hole) {
+	        largest_hole = 0;
+		mm->free_area_cache = base;
+	}
 try_again:
 	/* make sure it can fit in the remaining address space */
 	if (mm->free_area_cache < len)
@@ -392,13 +405,21 @@ hugepage_recheck:
 		 * vma->vm_start, use it:
 		 */
 		if (addr+len <= vma->vm_start &&
-				(!prev_vma || (addr >= prev_vma->vm_end)))
+		          (!prev_vma || (addr >= prev_vma->vm_end))) {
 			/* remember the address as a hint for next time */
-			return (mm->free_area_cache = addr);
-		else
+		        mm->cached_hole_size = largest_hole;
+		        return (mm->free_area_cache = addr);
+		} else {
 			/* pull free_area_cache down to the first hole */
-			if (mm->free_area_cache == vma->vm_end)
+		        if (mm->free_area_cache == vma->vm_end) {
 				mm->free_area_cache = vma->vm_start;
+				mm->cached_hole_size = largest_hole;
+			}
+		}
+
+		/* remember the largest hole we saw so far */
+		if (addr + largest_hole < vma->vm_start)
+		        largest_hole = vma->vm_start - addr;
 
 		/* try just below the current vma->vm_start */
 		addr = vma->vm_start-len;
@@ -411,6 +432,7 @@ fail:
 	 */
 	if (first_time) {
 		mm->free_area_cache = base;
+		largest_hole = 0;
 		first_time = 0;
 		goto try_again;
 	}
@@ -421,11 +443,13 @@ fail:
 	 * allocations.
 	 */
 	mm->free_area_cache = TASK_UNMAPPED_BASE;
+	mm->cached_hole_size = ~0UL;
 	addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags);
 	/*
 	 * Restore the topdown base:
 	 */
 	mm->free_area_cache = base;
+	mm->cached_hole_size = ~0UL;
 
 	return addr;
 }
diff --git a/arch/sh/kernel/sys_sh.c b/arch/sh/kernel/sys_sh.c
index df5ac294c37..917b2f32f26 100644
--- a/arch/sh/kernel/sys_sh.c
+++ b/arch/sh/kernel/sys_sh.c
@@ -79,6 +79,10 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr,
 		    (!vma || addr + len <= vma->vm_start))
 			return addr;
 	}
+	if (len <= mm->cached_hole_size) {
+	        mm->cached_hole_size = 0;
+		mm->free_area_cache = TASK_UNMAPPED_BASE;
+	}
 	if (flags & MAP_PRIVATE)
 		addr = PAGE_ALIGN(mm->free_area_cache);
 	else
@@ -95,6 +99,7 @@ full_search:
 			 */
 			if (start_addr != TASK_UNMAPPED_BASE) {
 				start_addr = addr = TASK_UNMAPPED_BASE;
+				mm->cached_hole_size = 0;
 				goto full_search;
 			}
 			return -ENOMEM;
@@ -106,6 +111,9 @@ full_search:
 			mm->free_area_cache = addr + len;
 			return addr;
 		}
+		if (addr + mm->cached_hole_size < vma->vm_start)
+		        mm->cached_hole_size = vma->vm_start - addr;
+
 		addr = vma->vm_end;
 		if (!(flags & MAP_PRIVATE))
 			addr = COLOUR_ALIGN(addr);
diff --git a/arch/sparc64/kernel/sys_sparc.c b/arch/sparc64/kernel/sys_sparc.c
index 0077f02f4b3..5f8c822a2b4 100644
--- a/arch/sparc64/kernel/sys_sparc.c
+++ b/arch/sparc64/kernel/sys_sparc.c
@@ -84,6 +84,10 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi
 			return addr;
 	}
 
+	if (len <= mm->cached_hole_size) {
+	        mm->cached_hole_size = 0;
+		mm->free_area_cache = TASK_UNMAPPED_BASE;
+	}
 	start_addr = addr = mm->free_area_cache;
 
 	task_size -= len;
@@ -103,6 +107,7 @@ full_search:
 		if (task_size < addr) {
 			if (start_addr != TASK_UNMAPPED_BASE) {
 				start_addr = addr = TASK_UNMAPPED_BASE;
+				mm->cached_hole_size = 0;
 				goto full_search;
 			}
 			return -ENOMEM;
@@ -114,6 +119,9 @@ full_search:
 			mm->free_area_cache = addr + len;
 			return addr;
 		}
+		if (addr + mm->cached_hole_size < vma->vm_start)
+		        mm->cached_hole_size = vma->vm_start - addr;
+
 		addr = vma->vm_end;
 		if (do_color_align)
 			addr = COLOUR_ALIGN(addr, pgoff);
diff --git a/arch/x86_64/ia32/ia32_aout.c b/arch/x86_64/ia32/ia32_aout.c
index 1965efc974d..c12edf5d97f 100644
--- a/arch/x86_64/ia32/ia32_aout.c
+++ b/arch/x86_64/ia32/ia32_aout.c
@@ -312,6 +312,7 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs)
 	current->mm->brk = ex.a_bss +
 		(current->mm->start_brk = N_BSSADDR(ex));
 	current->mm->free_area_cache = TASK_UNMAPPED_BASE;
+	current->mm->cached_hole_size = 0;
 
 	set_mm_counter(current->mm, rss, 0);
 	current->mm->mmap = NULL;
diff --git a/arch/x86_64/kernel/sys_x86_64.c b/arch/x86_64/kernel/sys_x86_64.c
index d9798dd433f..cc7821c6885 100644
--- a/arch/x86_64/kernel/sys_x86_64.c
+++ b/arch/x86_64/kernel/sys_x86_64.c
@@ -105,6 +105,11 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr,
 		    (!vma || addr + len <= vma->vm_start))
 			return addr;
 	}
+	if (((flags & MAP_32BIT) || test_thread_flag(TIF_IA32))
+	    && len <= mm->cached_hole_size) {
+	        mm->cached_hole_size = 0;
+		mm->free_area_cache = begin;
+	}
 	addr = mm->free_area_cache;
 	if (addr < begin) 
 		addr = begin; 
@@ -120,6 +125,7 @@ full_search:
 			 */
 			if (start_addr != begin) {
 				start_addr = addr = begin;
+				mm->cached_hole_size = 0;
 				goto full_search;
 			}
 			return -ENOMEM;
@@ -131,6 +137,9 @@ full_search:
 			mm->free_area_cache = addr + len;
 			return addr;
 		}
+		if (addr + mm->cached_hole_size < vma->vm_start)
+		        mm->cached_hole_size = vma->vm_start - addr;
+
 		addr = vma->vm_end;
 	}
 }
diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c
index 009b8920c1f..dd9baabaf01 100644
--- a/fs/binfmt_aout.c
+++ b/fs/binfmt_aout.c
@@ -316,6 +316,7 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs)
 	current->mm->brk = ex.a_bss +
 		(current->mm->start_brk = N_BSSADDR(ex));
 	current->mm->free_area_cache = current->mm->mmap_base;
+	current->mm->cached_hole_size = 0;
 
 	set_mm_counter(current->mm, rss, 0);
 	current->mm->mmap = NULL;
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index f8f6b6b7617..7976a238f0a 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -775,6 +775,7 @@ static int load_elf_binary(struct linux_binprm * bprm, struct pt_regs * regs)
 	   change some of these later */
 	set_mm_counter(current->mm, rss, 0);
 	current->mm->free_area_cache = current->mm->mmap_base;
+	current->mm->cached_hole_size = 0;
 	retval = setup_arg_pages(bprm, randomize_stack_top(STACK_TOP),
 				 executable_stack);
 	if (retval < 0) {
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index 2af3338f891..3a9b6d179cb 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -122,6 +122,9 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 
 	start_addr = mm->free_area_cache;
 
+	if (len <= mm->cached_hole_size)
+		start_addr = TASK_UNMAPPED_BASE;
+
 full_search:
 	addr = ALIGN(start_addr, HPAGE_SIZE);
 
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 4dbb109022f..b58afd97a18 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -201,8 +201,8 @@ extern unsigned long
 arch_get_unmapped_area_topdown(struct file *filp, unsigned long addr,
 			  unsigned long len, unsigned long pgoff,
 			  unsigned long flags);
-extern void arch_unmap_area(struct vm_area_struct *area);
-extern void arch_unmap_area_topdown(struct vm_area_struct *area);
+extern void arch_unmap_area(struct mm_struct *, unsigned long);
+extern void arch_unmap_area_topdown(struct mm_struct *, unsigned long);
 
 #define set_mm_counter(mm, member, value) (mm)->_##member = (value)
 #define get_mm_counter(mm, member) ((mm)->_##member)
@@ -218,9 +218,10 @@ struct mm_struct {
 	unsigned long (*get_unmapped_area) (struct file *filp,
 				unsigned long addr, unsigned long len,
 				unsigned long pgoff, unsigned long flags);
-	void (*unmap_area) (struct vm_area_struct *area);
-	unsigned long mmap_base;		/* base of mmap area */
-	unsigned long free_area_cache;		/* first hole */
+	void (*unmap_area) (struct mm_struct *mm, unsigned long addr);
+        unsigned long mmap_base;		/* base of mmap area */
+        unsigned long cached_hole_size;         /* if non-zero, the largest hole below free_area_cache */
+	unsigned long free_area_cache;		/* first hole of size cached_hole_size or larger */
 	pgd_t * pgd;
 	atomic_t mm_users;			/* How many users with user space? */
 	atomic_t mm_count;			/* How many references to "struct mm_struct" (users count as 1) */
diff --git a/kernel/fork.c b/kernel/fork.c
index f42a17f8869..876b31cd822 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -194,6 +194,7 @@ static inline int dup_mmap(struct mm_struct * mm, struct mm_struct * oldmm)
 	mm->mmap = NULL;
 	mm->mmap_cache = NULL;
 	mm->free_area_cache = oldmm->mmap_base;
+	mm->cached_hole_size = ~0UL;
 	mm->map_count = 0;
 	set_mm_counter(mm, rss, 0);
 	set_mm_counter(mm, anon_rss, 0);
@@ -322,6 +323,7 @@ static struct mm_struct * mm_init(struct mm_struct * mm)
 	mm->ioctx_list = NULL;
 	mm->default_kioctx = (struct kioctx)INIT_KIOCTX(mm->default_kioctx, *mm);
 	mm->free_area_cache = TASK_UNMAPPED_BASE;
+	mm->cached_hole_size = ~0UL;
 
 	if (likely(!mm_alloc_pgd(mm))) {
 		mm->def_flags = 0;
diff --git a/mm/mmap.c b/mm/mmap.c
index de54acd9942..9da23c1ef9d 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -1175,7 +1175,12 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr,
 		    (!vma || addr + len <= vma->vm_start))
 			return addr;
 	}
-	start_addr = addr = mm->free_area_cache;
+	if (len > mm->cached_hole_size) {
+	        start_addr = addr = mm->free_area_cache;
+	} else {
+	        start_addr = addr = TASK_UNMAPPED_BASE;
+	        mm->cached_hole_size = 0;
+	}
 
 full_search:
 	for (vma = find_vma(mm, addr); ; vma = vma->vm_next) {
@@ -1186,7 +1191,9 @@ full_search:
 			 * some holes.
 			 */
 			if (start_addr != TASK_UNMAPPED_BASE) {
-				start_addr = addr = TASK_UNMAPPED_BASE;
+				addr = TASK_UNMAPPED_BASE;
+			        start_addr = addr;
+				mm->cached_hole_size = 0;
 				goto full_search;
 			}
 			return -ENOMEM;
@@ -1198,19 +1205,22 @@ full_search:
 			mm->free_area_cache = addr + len;
 			return addr;
 		}
+		if (addr + mm->cached_hole_size < vma->vm_start)
+		        mm->cached_hole_size = vma->vm_start - addr;
 		addr = vma->vm_end;
 	}
 }
 #endif	
 
-void arch_unmap_area(struct vm_area_struct *area)
+void arch_unmap_area(struct mm_struct *mm, unsigned long addr)
 {
 	/*
 	 * Is this a new hole at the lowest possible address?
 	 */
-	if (area->vm_start >= TASK_UNMAPPED_BASE &&
-			area->vm_start < area->vm_mm->free_area_cache)
-		area->vm_mm->free_area_cache = area->vm_start;
+	if (addr >= TASK_UNMAPPED_BASE && addr < mm->free_area_cache) {
+		mm->free_area_cache = addr;
+		mm->cached_hole_size = ~0UL;
+	}
 }
 
 /*
@@ -1240,6 +1250,12 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
 			return addr;
 	}
 
+	/* check if free_area_cache is useful for us */
+	if (len <= mm->cached_hole_size) {
+ 	        mm->cached_hole_size = 0;
+ 		mm->free_area_cache = mm->mmap_base;
+ 	}
+
 	/* either no address requested or can't fit in requested address hole */
 	addr = mm->free_area_cache;
 
@@ -1264,6 +1280,10 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
 			/* remember the address as a hint for next time */
 			return (mm->free_area_cache = addr);
 
+ 		/* remember the largest hole we saw so far */
+ 		if (addr + mm->cached_hole_size < vma->vm_start)
+ 		        mm->cached_hole_size = vma->vm_start - addr;
+
 		/* try just below the current vma->vm_start */
 		addr = vma->vm_start-len;
 	} while (len < vma->vm_start);
@@ -1274,28 +1294,30 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
 	 * can happen with large stack limits and large mmap()
 	 * allocations.
 	 */
-	mm->free_area_cache = TASK_UNMAPPED_BASE;
+	mm->cached_hole_size = ~0UL;
+  	mm->free_area_cache = TASK_UNMAPPED_BASE;
 	addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags);
 	/*
 	 * Restore the topdown base:
 	 */
 	mm->free_area_cache = mm->mmap_base;
+	mm->cached_hole_size = ~0UL;
 
 	return addr;
 }
 #endif
 
-void arch_unmap_area_topdown(struct vm_area_struct *area)
+void arch_unmap_area_topdown(struct mm_struct *mm, unsigned long addr)
 {
 	/*
 	 * Is this a new hole at the highest possible address?
 	 */
-	if (area->vm_end > area->vm_mm->free_area_cache)
-		area->vm_mm->free_area_cache = area->vm_end;
+	if (addr > mm->free_area_cache)
+		mm->free_area_cache = addr;
 
 	/* dont allow allocations above current base */
-	if (area->vm_mm->free_area_cache > area->vm_mm->mmap_base)
-		area->vm_mm->free_area_cache = area->vm_mm->mmap_base;
+	if (mm->free_area_cache > mm->mmap_base)
+		mm->free_area_cache = mm->mmap_base;
 }
 
 unsigned long
@@ -1595,7 +1617,6 @@ static void unmap_vma(struct mm_struct *mm, struct vm_area_struct *area)
 	if (area->vm_flags & VM_LOCKED)
 		area->vm_mm->locked_vm -= len >> PAGE_SHIFT;
 	vm_stat_unaccount(area);
-	area->vm_mm->unmap_area(area);
 	remove_vm_struct(area);
 }
 
@@ -1649,6 +1670,7 @@ detach_vmas_to_be_unmapped(struct mm_struct *mm, struct vm_area_struct *vma,
 {
 	struct vm_area_struct **insertion_point;
 	struct vm_area_struct *tail_vma = NULL;
+	unsigned long addr;
 
 	insertion_point = (prev ? &prev->vm_next : &mm->mmap);
 	do {
@@ -1659,6 +1681,11 @@ detach_vmas_to_be_unmapped(struct mm_struct *mm, struct vm_area_struct *vma,
 	} while (vma && vma->vm_start < end);
 	*insertion_point = vma;
 	tail_vma->vm_next = NULL;
+	if (mm->unmap_area == arch_unmap_area)
+		addr = prev ? prev->vm_end : mm->mmap_base;
+	else
+		addr = vma ?  vma->vm_start : mm->mmap_base;
+	mm->unmap_area(mm, addr);
 	mm->mmap_cache = NULL;		/* Kill the cache. */
 }
 
diff --git a/mm/nommu.c b/mm/nommu.c
index c53e9c8f6b4..ce74452c02d 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -1067,7 +1067,7 @@ unsigned long arch_get_unmapped_area(struct file *file, unsigned long addr,
 	return -ENOMEM;
 }
 
-void arch_unmap_area(struct vm_area_struct *area)
+void arch_unmap_area(struct mm_struct *mm, unsigned long addr)
 {
 }
 
-- 
cgit v1.2.3-70-g09d2


From cbe37d093707762fc0abb280781e6a82a9d8d568 Mon Sep 17 00:00:00 2001
From: Badari Pulavarty <pbadari@us.ibm.com>
Date: Tue, 21 Jun 2005 17:14:52 -0700
Subject: [PATCH] mm: remove PG_highmem

Remove PG_highmem, to save a page flag.  Use is_highmem() instead.  It'll
generate a little more code, but we don't use PageHigheMem() in many places.

Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/frv/mm/init.c         |  1 -
 arch/i386/mm/init.c        |  1 -
 arch/mips/mm/init.c        |  1 -
 arch/ppc/mm/init.c         |  1 -
 arch/sparc/mm/init.c       |  1 -
 arch/um/kernel/mem.c       |  1 -
 include/linux/page-flags.h | 31 +++++++++++++++----------------
 7 files changed, 15 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/arch/frv/mm/init.c b/arch/frv/mm/init.c
index 41958f57c83..79433159b5f 100644
--- a/arch/frv/mm/init.c
+++ b/arch/frv/mm/init.c
@@ -169,7 +169,6 @@ void __init mem_init(void)
 		struct page *page = &mem_map[pfn];
 
 		ClearPageReserved(page);
-		set_bit(PG_highmem, &page->flags);
 		set_page_count(page, 1);
 		__free_page(page);
 		totalram_pages++;
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index 7a7ea373726..8766c771bb4 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -269,7 +269,6 @@ void __init one_highpage_init(struct page *page, int pfn, int bad_ppro)
 {
 	if (page_is_ram(pfn) && !(bad_ppro && page_kills_ppro(pfn))) {
 		ClearPageReserved(page);
-		set_bit(PG_highmem, &page->flags);
 		set_page_count(page, 1);
 		__free_page(page);
 		totalhigh_pages++;
diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c
index b027ce7efbc..73843c52877 100644
--- a/arch/mips/mm/init.c
+++ b/arch/mips/mm/init.c
@@ -232,7 +232,6 @@ void __init mem_init(void)
 #ifdef CONFIG_LIMITED_DMA
 		set_page_address(page, lowmem_page_address(page));
 #endif
-		set_bit(PG_highmem, &page->flags);
 		set_page_count(page, 1);
 		__free_page(page);
 		totalhigh_pages++;
diff --git a/arch/ppc/mm/init.c b/arch/ppc/mm/init.c
index 363c157e361..c8529d004cf 100644
--- a/arch/ppc/mm/init.c
+++ b/arch/ppc/mm/init.c
@@ -469,7 +469,6 @@ void __init mem_init(void)
 			struct page *page = mem_map + pfn;
 
 			ClearPageReserved(page);
-			set_bit(PG_highmem, &page->flags);
 			set_page_count(page, 1);
 			__free_page(page);
 			totalhigh_pages++;
diff --git a/arch/sparc/mm/init.c b/arch/sparc/mm/init.c
index a2dea69b2f0..ec2e05028a1 100644
--- a/arch/sparc/mm/init.c
+++ b/arch/sparc/mm/init.c
@@ -384,7 +384,6 @@ void map_high_region(unsigned long start_pfn, unsigned long end_pfn)
 		struct page *page = pfn_to_page(tmp);
 
 		ClearPageReserved(page);
-		set_bit(PG_highmem, &page->flags);
 		set_page_count(page, 1);
 		__free_page(page);
 		totalhigh_pages++;
diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c
index c22825f13e4..5597bd39e6b 100644
--- a/arch/um/kernel/mem.c
+++ b/arch/um/kernel/mem.c
@@ -53,7 +53,6 @@ static void setup_highmem(unsigned long highmem_start,
 	for(i = 0; i < highmem_len >> PAGE_SHIFT; i++){
 		page = &mem_map[highmem_pfn + i];
 		ClearPageReserved(page);
-		set_bit(PG_highmem, &page->flags);
 		set_page_count(page, 1);
 		__free_page(page);
 	}
diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index 39ab8c6b565..df313891db1 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -61,21 +61,20 @@
 #define PG_active		 6
 #define PG_slab			 7	/* slab debug (Suparna wants this) */
 
-#define PG_highmem		 8
-#define PG_checked		 9	/* kill me in 2.5.<early>. */
-#define PG_arch_1		10
-#define PG_reserved		11
-
-#define PG_private		12	/* Has something at ->private */
-#define PG_writeback		13	/* Page is under writeback */
-#define PG_nosave		14	/* Used for system suspend/resume */
-#define PG_compound		15	/* Part of a compound page */
-
-#define PG_swapcache		16	/* Swap page: swp_entry_t in private */
-#define PG_mappedtodisk		17	/* Has blocks allocated on-disk */
-#define PG_reclaim		18	/* To be reclaimed asap */
-#define PG_nosave_free		19	/* Free, should not be written */
-#define PG_uncached		20	/* Page has been mapped as uncached */
+#define PG_checked		 8	/* kill me in 2.5.<early>. */
+#define PG_arch_1		 9
+#define PG_reserved		10
+#define PG_private		11	/* Has something at ->private */
+
+#define PG_writeback		12	/* Page is under writeback */
+#define PG_nosave		13	/* Used for system suspend/resume */
+#define PG_compound		14	/* Part of a compound page */
+#define PG_swapcache		15	/* Swap page: swp_entry_t in private */
+
+#define PG_mappedtodisk		16	/* Has blocks allocated on-disk */
+#define PG_reclaim		17	/* To be reclaimed asap */
+#define PG_nosave_free		18	/* Free, should not be written */
+#define PG_uncached		19	/* Page has been mapped as uncached */
 
 /*
  * Global page accounting.  One instance per CPU.  Only unsigned longs are
@@ -215,7 +214,7 @@ extern void __mod_page_state(unsigned offset, unsigned long delta);
 #define TestSetPageSlab(page)	test_and_set_bit(PG_slab, &(page)->flags)
 
 #ifdef CONFIG_HIGHMEM
-#define PageHighMem(page)	test_bit(PG_highmem, &(page)->flags)
+#define PageHighMem(page)	is_highmem(page_zone(page))
 #else
 #define PageHighMem(page)	0 /* needed to optimize away at compile time */
 #endif
-- 
cgit v1.2.3-70-g09d2


From 1ad539b2bd89bf2e129123eb24d5bcc4484a35de Mon Sep 17 00:00:00 2001
From: Darren Hart <dvhltc@us.ibm.com>
Date: Tue, 21 Jun 2005 17:14:53 -0700
Subject: [PATCH] vm: try_to_free_pages unused argument

try_to_free_pages accepts a third argument, order, but hasn't used it since
before 2.6.0.  The following patch removes the argument and updates all the
calls to try_to_free_pages.

Signed-off-by: Darren Hart <dvhltc@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/buffer.c          | 2 +-
 include/linux/swap.h | 2 +-
 mm/page_alloc.c      | 2 +-
 mm/vmscan.c          | 3 +--
 4 files changed, 4 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/fs/buffer.c b/fs/buffer.c
index 7e9e409feaa..0befa724ab9 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -528,7 +528,7 @@ static void free_more_memory(void)
 	for_each_pgdat(pgdat) {
 		zones = pgdat->node_zonelists[GFP_NOFS&GFP_ZONEMASK].zones;
 		if (*zones)
-			try_to_free_pages(zones, GFP_NOFS, 0);
+			try_to_free_pages(zones, GFP_NOFS);
 	}
 }
 
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 0d21e682d99..2343f999e6e 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -172,7 +172,7 @@ extern int rotate_reclaimable_page(struct page *page);
 extern void swap_setup(void);
 
 /* linux/mm/vmscan.c */
-extern int try_to_free_pages(struct zone **, unsigned int, unsigned int);
+extern int try_to_free_pages(struct zone **, unsigned int);
 extern int zone_reclaim(struct zone *, unsigned int, unsigned int);
 extern int shrink_all_memory(int);
 extern int vm_swappiness;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 95cbd30a67b..70f710286ce 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -859,7 +859,7 @@ rebalance:
 	reclaim_state.reclaimed_slab = 0;
 	p->reclaim_state = &reclaim_state;
 
-	did_some_progress = try_to_free_pages(zones, gfp_mask, order);
+	did_some_progress = try_to_free_pages(zones, gfp_mask);
 
 	p->reclaim_state = NULL;
 	p->flags &= ~PF_MEMALLOC;
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 24da725a30f..4b8e62a1937 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -919,8 +919,7 @@ shrink_caches(struct zone **zones, struct scan_control *sc)
  * holds filesystem locks which prevent writeout this might not work, and the
  * allocation attempt will fail.
  */
-int try_to_free_pages(struct zone **zones,
-		unsigned int gfp_mask, unsigned int order)
+int try_to_free_pages(struct zone **zones, unsigned int gfp_mask)
 {
 	int priority;
 	int ret = 0;
-- 
cgit v1.2.3-70-g09d2


From 83e5d8f7253cb7b14472385a6d57df1e9f848e8e Mon Sep 17 00:00:00 2001
From: Benjamin LaHaise <bcrl@kvack.org>
Date: Tue, 21 Jun 2005 17:14:54 -0700
Subject: [PATCH] __mod_page_state(): pass unsigned long instead of unsigned

By making the offset argument of __mod_page_state an unsigned long instead
of unsigned, we can avoid forcing the compiler to sign extend a usually
constant argument.  This saves 1 instruction on x86-64.

Signed-off-by: Benjamin LaHaise <benjamin.c.lahaise@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/page-flags.h | 2 +-
 mm/page_alloc.c            | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index df313891db1..f2ee9b2332e 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -136,7 +136,7 @@ struct page_state {
 extern void get_page_state(struct page_state *ret);
 extern void get_full_page_state(struct page_state *ret);
 extern unsigned long __read_page_state(unsigned offset);
-extern void __mod_page_state(unsigned offset, unsigned long delta);
+extern void __mod_page_state(unsigned long offset, unsigned long delta);
 
 #define read_page_state(member) \
 	__read_page_state(offsetof(struct page_state, member))
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 70f710286ce..c976628f9d9 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1158,7 +1158,7 @@ unsigned long __read_page_state(unsigned offset)
 	return ret;
 }
 
-void __mod_page_state(unsigned offset, unsigned long delta)
+void __mod_page_state(unsigned long offset, unsigned long delta)
 {
 	unsigned long flags;
 	void* ptr;
-- 
cgit v1.2.3-70-g09d2


From c2f29ea111e3344ed48257c2a142c3db514e1529 Mon Sep 17 00:00:00 2001
From: Benjamin LaHaise <bcrl@kvack.org>
Date: Tue, 21 Jun 2005 17:14:55 -0700
Subject: [PATCH] __read_page_state(): pass unsigned long instead of unsigned

By making the offset argument of __read_page_state an unsigned long instead of
unsigned, we can avoid forcing the compiler to sign extend a usually constant
argument.  This saves 1 instruction on x86-64.

Signed-off-by: Benjamin LaHaise <benjamin.c.lahaise@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/page-flags.h | 2 +-
 mm/page_alloc.c            | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index f2ee9b2332e..f5a6695d4d2 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -135,7 +135,7 @@ struct page_state {
 
 extern void get_page_state(struct page_state *ret);
 extern void get_full_page_state(struct page_state *ret);
-extern unsigned long __read_page_state(unsigned offset);
+extern unsigned long __read_page_state(unsigned long offset);
 extern void __mod_page_state(unsigned long offset, unsigned long delta);
 
 #define read_page_state(member) \
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index c976628f9d9..10446a64637 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1144,7 +1144,7 @@ void get_full_page_state(struct page_state *ret)
 	__get_page_state(ret, sizeof(*ret) / sizeof(unsigned long));
 }
 
-unsigned long __read_page_state(unsigned offset)
+unsigned long __read_page_state(unsigned long offset)
 {
 	unsigned long ret = 0;
 	int cpu;
-- 
cgit v1.2.3-70-g09d2


From 4ae7c03943fca73f23bc0cdb938070f41b98101f Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@engr.sgi.com>
Date: Tue, 21 Jun 2005 17:14:57 -0700
Subject: [PATCH] Periodically drain non local pagesets

The pageset array can potentially acquire a huge amount of memory on large
NUMA systems.  F.e.  on a system with 512 processors and 256 nodes there
will be 256*512 pagesets.  If each pageset only holds 5 pages then we are
talking about 655360 pages.With a 16K page size on IA64 this results in
potentially 10 Gigabytes of memory being trapped in pagesets.  The typical
cases are much less for smaller systems but there is still the potential of
memory being trapped in off node pagesets.  Off node memory may be rarely
used if local memory is available and so we may potentially have memory in
seldom used pagesets without this patch.

The slab allocator flushes its per cpu caches every 2 seconds.  The
following patch flushes the off node pageset caches in the same way by
tying into the slab flush.

The patch also changes /proc/zoneinfo to include the number of pages
currently in each pageset.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/gfp.h |  5 +++++
 mm/page_alloc.c     | 35 +++++++++++++++++++++++++++++++++--
 mm/slab.c           |  1 +
 3 files changed, 39 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 208535fd483..8d6bf608b19 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -133,5 +133,10 @@ extern void FASTCALL(free_cold_page(struct page *page));
 #define free_page(addr) free_pages((addr),0)
 
 void page_alloc_init(void);
+#ifdef CONFIG_NUMA
+void drain_remote_pages(void);
+#else
+static inline void drain_remote_pages(void) { };
+#endif
 
 #endif /* __LINUX_GFP_H */
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index be05d17bd7d..a95e72d7f94 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -516,6 +516,36 @@ static int rmqueue_bulk(struct zone *zone, unsigned int order,
 	return allocated;
 }
 
+#ifdef CONFIG_NUMA
+/* Called from the slab reaper to drain remote pagesets */
+void drain_remote_pages(void)
+{
+	struct zone *zone;
+	int i;
+	unsigned long flags;
+
+	local_irq_save(flags);
+	for_each_zone(zone) {
+		struct per_cpu_pageset *pset;
+
+		/* Do not drain local pagesets */
+		if (zone->zone_pgdat->node_id == numa_node_id())
+			continue;
+
+		pset = zone->pageset[smp_processor_id()];
+		for (i = 0; i < ARRAY_SIZE(pset->pcp); i++) {
+			struct per_cpu_pages *pcp;
+
+			pcp = &pset->pcp[i];
+			if (pcp->count)
+				pcp->count -= free_pages_bulk(zone, pcp->count,
+						&pcp->list, 0);
+		}
+	}
+	local_irq_restore(flags);
+}
+#endif
+
 #if defined(CONFIG_PM) || defined(CONFIG_HOTPLUG_CPU)
 static void __drain_pages(unsigned int cpu)
 {
@@ -1271,12 +1301,13 @@ void show_free_areas(void)
 			pageset = zone_pcp(zone, cpu);
 
 			for (temperature = 0; temperature < 2; temperature++)
-				printk("cpu %d %s: low %d, high %d, batch %d\n",
+				printk("cpu %d %s: low %d, high %d, batch %d used:%d\n",
 					cpu,
 					temperature ? "cold" : "hot",
 					pageset->pcp[temperature].low,
 					pageset->pcp[temperature].high,
-					pageset->pcp[temperature].batch);
+					pageset->pcp[temperature].batch,
+					pageset->pcp[temperature].count);
 		}
 	}
 
diff --git a/mm/slab.c b/mm/slab.c
index c78d343b3c5..93cbbbb39f4 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -2851,6 +2851,7 @@ next:
 	}
 	check_irq_on();
 	up(&cache_chain_sem);
+	drain_remote_pages();
 	/* Setup the next iteration */
 	schedule_delayed_work(&__get_cpu_var(reap_work), REAPTIMEOUT_CPUC + smp_processor_id());
 }
-- 
cgit v1.2.3-70-g09d2


From f14f75b81187cdbe10cc53a521bf9fdf97b59f8c Mon Sep 17 00:00:00 2001
From: Jes Sorensen <jes@wildopensource.com>
Date: Tue, 21 Jun 2005 17:15:02 -0700
Subject: [PATCH] ia64 uncached alloc

This patch contains the ia64 uncached page allocator and the generic
allocator (genalloc).  The uncached allocator was formerly part of the SN2
mspec driver but there are several other users of it so it has been split
off from the driver.

The generic allocator can be used by device driver to manage special memory
etc.  The generic allocator is based on the allocator from the sym53c8xx_2
driver.

Various users on ia64 needs uncached memory.  The SGI SN architecture requires
it for inter-partition communication between partitions within a large NUMA
cluster.  The specific user for this is the XPC code.  Another application is
large MPI style applications which use it for synchronization, on SN this can
be done using special 'fetchop' operations but it also benefits non SN
hardware which may use regular uncached memory for this purpose.  Performance
of doing this through uncached vs cached memory is pretty substantial.  This
is handled by the mspec driver which I will push out in a seperate patch.

Rather than creating a specific allocator for just uncached memory I came up
with genalloc which is a generic purpose allocator that can be used by device
drivers and other subsystems as they please.  For instance to handle onboard
device memory.  It was derived from the sym53c7xx_2 driver's allocator which
is also an example of a potential user (I am refraining from modifying sym2
right now as it seems to have been under fairly heavy development recently).

On ia64 memory has various properties within a granule, ie.  it isn't safe to
access memory as uncached within the same granule as currently has memory
accessed in cached mode.  The regular system therefore doesn't utilize memory
in the lower granules which is mixed in with device PAL code etc.  The
uncached driver walks the EFI memmap and pulls out the spill uncached pages
and sticks them into the uncached pool.  Only after these chunks have been
utilized, will it start converting regular cached memory into uncached memory.
Hence the reason for the EFI related code additions.

Signed-off-by: Jes Sorensen <jes@wildopensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/Kconfig           |   4 +
 arch/ia64/kernel/Makefile   |   1 +
 arch/ia64/kernel/efi.c      |  32 ++++++
 arch/ia64/kernel/uncached.c | 246 ++++++++++++++++++++++++++++++++++++++++++++
 include/asm-ia64/uncached.h |  12 +++
 include/linux/genalloc.h    |  40 +++++++
 lib/Kconfig                 |   6 ++
 lib/Makefile                |   1 +
 lib/genalloc.c              | 188 +++++++++++++++++++++++++++++++++
 9 files changed, 530 insertions(+)
 create mode 100644 arch/ia64/kernel/uncached.c
 create mode 100644 include/asm-ia64/uncached.h
 create mode 100644 include/linux/genalloc.h
 create mode 100644 lib/genalloc.c

(limited to 'include')

diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 3ad2c4af099..295b5abee72 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -50,6 +50,10 @@ config SCHED_NO_NO_OMIT_FRAME_POINTER
 	bool
 	default y
 
+config IA64_UNCACHED_ALLOCATOR
+	bool
+	select GENERIC_ALLOCATOR
+
 choice
 	prompt "System type"
 	default IA64_GENERIC
diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile
index c1a02bbc252..4c73d8ba2e3 100644
--- a/arch/ia64/kernel/Makefile
+++ b/arch/ia64/kernel/Makefile
@@ -20,6 +20,7 @@ obj-$(CONFIG_SMP)		+= smp.o smpboot.o domain.o
 obj-$(CONFIG_PERFMON)		+= perfmon_default_smpl.o
 obj-$(CONFIG_IA64_CYCLONE)	+= cyclone.o
 obj-$(CONFIG_IA64_MCA_RECOVERY)	+= mca_recovery.o
+obj-$(CONFIG_IA64_UNCACHED_ALLOCATOR)	+= uncached.o
 mca_recovery-y			+= mca_drv.o mca_drv_asm.o
 
 # The gate DSO image is built using a special linker script.
diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c
index 4a3b1aac43e..179f230816e 100644
--- a/arch/ia64/kernel/efi.c
+++ b/arch/ia64/kernel/efi.c
@@ -409,6 +409,38 @@ efi_memmap_walk (efi_freemem_callback_t callback, void *arg)
 	}
 }
 
+/*
+ * Walk the EFI memory map to pull out leftover pages in the lower
+ * memory regions which do not end up in the regular memory map and
+ * stick them into the uncached allocator
+ *
+ * The regular walk function is significantly more complex than the
+ * uncached walk which means it really doesn't make sense to try and
+ * marge the two.
+ */
+void __init
+efi_memmap_walk_uc (efi_freemem_callback_t callback)
+{
+	void *efi_map_start, *efi_map_end, *p;
+	efi_memory_desc_t *md;
+	u64 efi_desc_size, start, end;
+
+	efi_map_start = __va(ia64_boot_param->efi_memmap);
+	efi_map_end = efi_map_start + ia64_boot_param->efi_memmap_size;
+	efi_desc_size = ia64_boot_param->efi_memdesc_size;
+
+	for (p = efi_map_start; p < efi_map_end; p += efi_desc_size) {
+		md = p;
+		if (md->attribute == EFI_MEMORY_UC) {
+			start = PAGE_ALIGN(md->phys_addr);
+			end = PAGE_ALIGN((md->phys_addr+(md->num_pages << EFI_PAGE_SHIFT)) & PAGE_MASK);
+			if ((*callback)(start, end, NULL) < 0)
+				return;
+		}
+	}
+}
+
+
 /*
  * Look for the PAL_CODE region reported by EFI and maps it using an
  * ITR to enable safe PAL calls in virtual mode.  See IA-64 Processor
diff --git a/arch/ia64/kernel/uncached.c b/arch/ia64/kernel/uncached.c
new file mode 100644
index 00000000000..490dfc9ab47
--- /dev/null
+++ b/arch/ia64/kernel/uncached.c
@@ -0,0 +1,246 @@
+/*
+ * Copyright (C) 2001-2005 Silicon Graphics, Inc.  All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of version 2 of the GNU General Public License
+ * as published by the Free Software Foundation.
+ *
+ * A simple uncached page allocator using the generic allocator. This
+ * allocator first utilizes the spare (spill) pages found in the EFI
+ * memmap and will then start converting cached pages to uncached ones
+ * at a granule at a time. Node awareness is implemented by having a
+ * pool of pages per node.
+ */
+
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/efi.h>
+#include <linux/genalloc.h>
+#include <asm/page.h>
+#include <asm/pal.h>
+#include <asm/system.h>
+#include <asm/pgtable.h>
+#include <asm/atomic.h>
+#include <asm/tlbflush.h>
+#include <asm/sn/arch.h>
+
+#define DEBUG	0
+
+#if DEBUG
+#define dprintk			printk
+#else
+#define dprintk(x...)		do { } while (0)
+#endif
+
+void __init efi_memmap_walk_uc (efi_freemem_callback_t callback);
+
+#define MAX_UNCACHED_GRANULES	5
+static int allocated_granules;
+
+struct gen_pool *uncached_pool[MAX_NUMNODES];
+
+
+static void uncached_ipi_visibility(void *data)
+{
+	int status;
+
+	status = ia64_pal_prefetch_visibility(PAL_VISIBILITY_PHYSICAL);
+	if ((status != PAL_VISIBILITY_OK) &&
+	    (status != PAL_VISIBILITY_OK_REMOTE_NEEDED))
+		printk(KERN_DEBUG "pal_prefetch_visibility() returns %i on "
+		       "CPU %i\n", status, get_cpu());
+}
+
+
+static void uncached_ipi_mc_drain(void *data)
+{
+	int status;
+	status = ia64_pal_mc_drain();
+	if (status)
+		printk(KERN_WARNING "ia64_pal_mc_drain() failed with %i on "
+		       "CPU %i\n", status, get_cpu());
+}
+
+
+static unsigned long
+uncached_get_new_chunk(struct gen_pool *poolp)
+{
+	struct page *page;
+	void *tmp;
+	int status, i;
+	unsigned long addr, node;
+
+	if (allocated_granules >= MAX_UNCACHED_GRANULES)
+		return 0;
+
+	node = poolp->private;
+	page = alloc_pages_node(node, GFP_KERNEL | __GFP_ZERO,
+				IA64_GRANULE_SHIFT-PAGE_SHIFT);
+
+	dprintk(KERN_INFO "get_new_chunk page %p, addr %lx\n",
+		page, (unsigned long)(page-vmem_map) << PAGE_SHIFT);
+
+	/*
+	 * Do magic if no mem on local node! XXX
+	 */
+	if (!page)
+		return 0;
+	tmp = page_address(page);
+
+	/*
+	 * There's a small race here where it's possible for someone to
+	 * access the page through /dev/mem halfway through the conversion
+	 * to uncached - not sure it's really worth bothering about
+	 */
+	for (i = 0; i < (IA64_GRANULE_SIZE / PAGE_SIZE); i++)
+		SetPageUncached(&page[i]);
+
+	flush_tlb_kernel_range(tmp, tmp + IA64_GRANULE_SIZE);
+
+	status = ia64_pal_prefetch_visibility(PAL_VISIBILITY_PHYSICAL);
+
+	dprintk(KERN_INFO "pal_prefetch_visibility() returns %i on cpu %i\n",
+		status, get_cpu());
+
+	if (!status) {
+		status = smp_call_function(uncached_ipi_visibility, NULL, 0, 1);
+		if (status)
+			printk(KERN_WARNING "smp_call_function failed for "
+			       "uncached_ipi_visibility! (%i)\n", status);
+	}
+
+	if (ia64_platform_is("sn2"))
+		sn_flush_all_caches((unsigned long)tmp, IA64_GRANULE_SIZE);
+	else
+		flush_icache_range((unsigned long)tmp,
+				   (unsigned long)tmp+IA64_GRANULE_SIZE);
+
+	ia64_pal_mc_drain();
+	status = smp_call_function(uncached_ipi_mc_drain, NULL, 0, 1);
+	if (status)
+		printk(KERN_WARNING "smp_call_function failed for "
+		       "uncached_ipi_mc_drain! (%i)\n", status);
+
+	addr = (unsigned long)tmp - PAGE_OFFSET + __IA64_UNCACHED_OFFSET;
+
+	allocated_granules++;
+	return addr;
+}
+
+
+/*
+ * uncached_alloc_page
+ *
+ * Allocate 1 uncached page. Allocates on the requested node. If no
+ * uncached pages are available on the requested node, roundrobin starting
+ * with higher nodes.
+ */
+unsigned long
+uncached_alloc_page(int nid)
+{
+	unsigned long maddr;
+
+	maddr = gen_pool_alloc(uncached_pool[nid], PAGE_SIZE);
+
+	dprintk(KERN_DEBUG "uncached_alloc_page returns %lx on node %i\n",
+		maddr, nid);
+
+	/*
+	 * If no memory is availble on our local node, try the
+	 * remaining nodes in the system.
+	 */
+	if (!maddr) {
+		int i;
+
+		for (i = MAX_NUMNODES - 1; i >= 0; i--) {
+			if (i == nid || !node_online(i))
+				continue;
+			maddr = gen_pool_alloc(uncached_pool[i], PAGE_SIZE);
+			dprintk(KERN_DEBUG "uncached_alloc_page alternate search "
+				"returns %lx on node %i\n", maddr, i);
+			if (maddr) {
+				break;
+			}
+		}
+	}
+
+	return maddr;
+}
+EXPORT_SYMBOL(uncached_alloc_page);
+
+
+/*
+ * uncached_free_page
+ *
+ * Free a single uncached page.
+ */
+void
+uncached_free_page(unsigned long maddr)
+{
+	int node;
+
+	node = nasid_to_cnodeid(NASID_GET(maddr));
+
+	dprintk(KERN_DEBUG "uncached_free_page(%lx) on node %i\n", maddr, node);
+
+	if ((maddr & (0XFUL << 60)) != __IA64_UNCACHED_OFFSET)
+		panic("uncached_free_page invalid address %lx\n", maddr);
+
+	gen_pool_free(uncached_pool[node], maddr, PAGE_SIZE);
+}
+EXPORT_SYMBOL(uncached_free_page);
+
+
+/*
+ * uncached_build_memmap,
+ *
+ * Called at boot time to build a map of pages that can be used for
+ * memory special operations.
+ */
+static int __init
+uncached_build_memmap(unsigned long start, unsigned long end, void *arg)
+{
+	long length;
+	unsigned long vstart, vend;
+	int node;
+
+	length = end - start;
+	vstart = start + __IA64_UNCACHED_OFFSET;
+	vend = end + __IA64_UNCACHED_OFFSET;
+
+	dprintk(KERN_ERR "uncached_build_memmap(%lx %lx)\n", start, end);
+
+	memset((char *)vstart, 0, length);
+
+	node = nasid_to_cnodeid(NASID_GET(start));
+
+	for (; vstart < vend ; vstart += PAGE_SIZE) {
+		dprintk(KERN_INFO "sticking %lx into the pool!\n", vstart);
+		gen_pool_free(uncached_pool[node], vstart, PAGE_SIZE);
+	}
+
+	return 0;
+}
+
+
+static int __init uncached_init(void) {
+	int i;
+
+	for (i = 0; i < MAX_NUMNODES; i++) {
+		if (!node_online(i))
+			continue;
+		uncached_pool[i] = gen_pool_create(0, IA64_GRANULE_SHIFT,
+						   &uncached_get_new_chunk, i);
+	}
+
+	efi_memmap_walk_uc(uncached_build_memmap);
+
+	return 0;
+}
+
+__initcall(uncached_init);
diff --git a/include/asm-ia64/uncached.h b/include/asm-ia64/uncached.h
new file mode 100644
index 00000000000..b82d923b73c
--- /dev/null
+++ b/include/asm-ia64/uncached.h
@@ -0,0 +1,12 @@
+/*
+ * Copyright (C) 2001-2005 Silicon Graphics, Inc.  All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of version 2 of the GNU General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Prototypes for the uncached page allocator
+ */
+
+extern unsigned long uncached_alloc_page(int nid);
+extern void uncached_free_page(unsigned long);
diff --git a/include/linux/genalloc.h b/include/linux/genalloc.h
new file mode 100644
index 00000000000..7fd0576a445
--- /dev/null
+++ b/include/linux/genalloc.h
@@ -0,0 +1,40 @@
+/*
+ * Basic general purpose allocator for managing special purpose memory
+ * not managed by the regular kmalloc/kfree interface.
+ * Uses for this includes on-device special memory, uncached memory
+ * etc.
+ *
+ * This code is based on the buddy allocator found in the sym53c8xx_2
+ * driver, adapted for general purpose use.
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2.  See the file COPYING for more details.
+ */
+
+#include <linux/spinlock.h>
+
+#define ALLOC_MIN_SHIFT		5 /* 32 bytes minimum */
+/*
+ *  Link between free memory chunks of a given size.
+ */
+struct gen_pool_link {
+	struct gen_pool_link *next;
+};
+
+/*
+ *  Memory pool descriptor.
+ */
+struct gen_pool {
+	spinlock_t lock;
+	unsigned long (*get_new_chunk)(struct gen_pool *);
+	struct gen_pool *next;
+	struct gen_pool_link *h;
+	unsigned long private;
+	int max_chunk_shift;
+};
+
+unsigned long gen_pool_alloc(struct gen_pool *poolp, int size);
+void gen_pool_free(struct gen_pool *mp, unsigned long ptr, int size);
+struct gen_pool *gen_pool_create(int nr_chunks, int max_chunk_shift,
+				 unsigned long (*fp)(struct gen_pool *),
+				 unsigned long data);
diff --git a/lib/Kconfig b/lib/Kconfig
index eeb45225248..2d4d4e3bc4a 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -39,6 +39,12 @@ config ZLIB_INFLATE
 config ZLIB_DEFLATE
 	tristate
 
+#
+# Generic allocator support is selected if needed
+#
+config GENERIC_ALLOCATOR
+	boolean
+
 #
 # reed solomon support is select'ed if needed
 #
diff --git a/lib/Makefile b/lib/Makefile
index 5f10cb89840..dcb4231916e 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -30,6 +30,7 @@ obj-$(CONFIG_CRC_CCITT)	+= crc-ccitt.o
 obj-$(CONFIG_CRC32)	+= crc32.o
 obj-$(CONFIG_LIBCRC32C)	+= libcrc32c.o
 obj-$(CONFIG_GENERIC_IOMAP) += iomap.o
+obj-$(CONFIG_GENERIC_ALLOCATOR) += genalloc.o
 
 obj-$(CONFIG_ZLIB_INFLATE) += zlib_inflate/
 obj-$(CONFIG_ZLIB_DEFLATE) += zlib_deflate/
diff --git a/lib/genalloc.c b/lib/genalloc.c
new file mode 100644
index 00000000000..d6d30d2e716
--- /dev/null
+++ b/lib/genalloc.c
@@ -0,0 +1,188 @@
+/*
+ * Basic general purpose allocator for managing special purpose memory
+ * not managed by the regular kmalloc/kfree interface.
+ * Uses for this includes on-device special memory, uncached memory
+ * etc.
+ *
+ * This code is based on the buddy allocator found in the sym53c8xx_2
+ * driver Copyright (C) 1999-2001  Gerard Roudier <groudier@free.fr>,
+ * and adapted for general purpose use.
+ *
+ * Copyright 2005 (C) Jes Sorensen <jes@trained-monkey.org>
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2.  See the file COPYING for more details.
+ */
+
+#include <linux/module.h>
+#include <linux/stddef.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/init.h>
+#include <linux/mm.h>
+#include <linux/spinlock.h>
+#include <linux/genalloc.h>
+
+#include <asm/page.h>
+
+
+struct gen_pool *gen_pool_create(int nr_chunks, int max_chunk_shift,
+				 unsigned long (*fp)(struct gen_pool *),
+				 unsigned long data)
+{
+	struct gen_pool *poolp;
+	unsigned long tmp;
+	int i;
+
+	/*
+	 * This is really an arbitrary limit, +10 is enough for
+	 * IA64_GRANULE_SHIFT, aka 16MB. If anyone needs a large limit
+	 * this can be increased without problems.
+	 */
+	if ((max_chunk_shift > (PAGE_SHIFT + 10)) ||
+	    ((max_chunk_shift < ALLOC_MIN_SHIFT) && max_chunk_shift))
+		return NULL;
+
+	if (!max_chunk_shift)
+		max_chunk_shift = PAGE_SHIFT;
+
+	poolp = kmalloc(sizeof(struct gen_pool), GFP_KERNEL);
+	if (!poolp)
+		return NULL;
+	memset(poolp, 0, sizeof(struct gen_pool));
+	poolp->h = kmalloc(sizeof(struct gen_pool_link) *
+			   (max_chunk_shift - ALLOC_MIN_SHIFT + 1),
+			   GFP_KERNEL);
+	if (!poolp->h) {
+		printk(KERN_WARNING "gen_pool_alloc() failed to allocate\n");
+		kfree(poolp);
+		return NULL;
+	}
+	memset(poolp->h, 0, sizeof(struct gen_pool_link) *
+	       (max_chunk_shift - ALLOC_MIN_SHIFT + 1));
+
+	spin_lock_init(&poolp->lock);
+	poolp->get_new_chunk = fp;
+	poolp->max_chunk_shift = max_chunk_shift;
+	poolp->private = data;
+
+	for (i = 0; i < nr_chunks; i++) {
+		tmp = poolp->get_new_chunk(poolp);
+		printk(KERN_INFO "allocated %lx\n", tmp);
+		if (!tmp)
+			break;
+		gen_pool_free(poolp, tmp, (1 << poolp->max_chunk_shift));
+	}
+
+	return poolp;
+}
+EXPORT_SYMBOL(gen_pool_create);
+
+
+/*
+ *  Simple power of two buddy-like generic allocator.
+ *  Provides naturally aligned memory chunks.
+ */
+unsigned long gen_pool_alloc(struct gen_pool *poolp, int size)
+{
+	int j, i, s, max_chunk_size;
+	unsigned long a, flags;
+	struct gen_pool_link *h = poolp->h;
+
+	max_chunk_size = 1 << poolp->max_chunk_shift;
+
+	if (size > max_chunk_size)
+		return 0;
+
+	i = 0;
+
+	size = max(size, 1 << ALLOC_MIN_SHIFT);
+	s = roundup_pow_of_two(size);
+
+	j = i;
+
+	spin_lock_irqsave(&poolp->lock, flags);
+	while (!h[j].next) {
+		if (s == max_chunk_size) {
+			struct gen_pool_link *ptr;
+			spin_unlock_irqrestore(&poolp->lock, flags);
+			ptr = (struct gen_pool_link *)poolp->get_new_chunk(poolp);
+			spin_lock_irqsave(&poolp->lock, flags);
+			h[j].next = ptr;
+			if (h[j].next)
+				h[j].next->next = NULL;
+			break;
+		}
+		j++;
+		s <<= 1;
+	}
+	a = (unsigned long) h[j].next;
+	if (a) {
+		h[j].next = h[j].next->next;
+		/*
+		 * This should be split into a seperate function doing
+		 * the chunk split in order to support custom
+		 * handling memory not physically accessible by host
+		 */
+		while (j > i) {
+			j -= 1;
+			s >>= 1;
+			h[j].next = (struct gen_pool_link *) (a + s);
+			h[j].next->next = NULL;
+		}
+	}
+	spin_unlock_irqrestore(&poolp->lock, flags);
+	return a;
+}
+EXPORT_SYMBOL(gen_pool_alloc);
+
+
+/*
+ *  Counter-part of the generic allocator.
+ */
+void gen_pool_free(struct gen_pool *poolp, unsigned long ptr, int size)
+{
+	struct gen_pool_link *q;
+	struct gen_pool_link *h = poolp->h;
+	unsigned long a, b, flags;
+	int i, s, max_chunk_size;
+
+	max_chunk_size = 1 << poolp->max_chunk_shift;
+
+	if (size > max_chunk_size)
+		return;
+
+	i = 0;
+
+	size = max(size, 1 << ALLOC_MIN_SHIFT);
+	s = roundup_pow_of_two(size);
+
+	a = ptr;
+
+	spin_lock_irqsave(&poolp->lock, flags);
+	while (1) {
+		if (s == max_chunk_size) {
+			((struct gen_pool_link *)a)->next = h[i].next;
+			h[i].next = (struct gen_pool_link *)a;
+			break;
+		}
+		b = a ^ s;
+		q = &h[i];
+
+		while (q->next && q->next != (struct gen_pool_link *)b)
+			q = q->next;
+
+		if (!q->next) {
+			((struct gen_pool_link *)a)->next = h[i].next;
+			h[i].next = (struct gen_pool_link *)a;
+			break;
+		}
+		q->next = q->next->next;
+		a = a & b;
+		s <<= 1;
+		i++;
+	}
+	spin_unlock_irqrestore(&poolp->lock, flags);
+}
+EXPORT_SYMBOL(gen_pool_free);
-- 
cgit v1.2.3-70-g09d2


From 65ed0b337bd2b47097cf6c772f024772513b7b0d Mon Sep 17 00:00:00 2001
From: Jes Sorensen <jes@trained-monkey.org>
Date: Tue, 21 Jun 2005 17:15:03 -0700
Subject: [PATCH] SN2 XPC build patches

This patch contains the bits to make the XPC code use the uncached
allocator rather than calling into the mspec driver.  It also includes the
mspec.h header which is required to build the XPC modules.

Signed-off-by: Jes Sorensen <jes@wildopensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/Kconfig                   |  2 +-
 arch/ia64/sn/kernel/xpc_partition.c |  6 ++--
 include/asm-ia64/sn/mspec.h         | 59 +++++++++++++++++++++++++++++++++++++
 3 files changed, 64 insertions(+), 3 deletions(-)
 create mode 100644 include/asm-ia64/sn/mspec.h

(limited to 'include')

diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 295b5abee72..34e603cc171 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -227,7 +227,7 @@ config IA64_SGI_SN_SIM
 
 config IA64_SGI_SN_XP
 	tristate "Support communication between SGI SSIs"
-	depends on MSPEC
+	select IA64_UNCACHED_ALLOCATOR
 	help
 	  An SGI machine can be divided into multiple Single System
 	  Images which act independently of each other and have
diff --git a/arch/ia64/sn/kernel/xpc_partition.c b/arch/ia64/sn/kernel/xpc_partition.c
index 2c3c4a8af55..cd7ed73f0e7 100644
--- a/arch/ia64/sn/kernel/xpc_partition.c
+++ b/arch/ia64/sn/kernel/xpc_partition.c
@@ -22,6 +22,7 @@
 #include <linux/cache.h>
 #include <linux/mmzone.h>
 #include <linux/nodemask.h>
+#include <asm/uncached.h>
 #include <asm/sn/bte.h>
 #include <asm/sn/intr.h>
 #include <asm/sn/sn_sal.h>
@@ -183,7 +184,7 @@ xpc_rsvd_page_init(void)
 	 * memory protections are never restricted.
 	 */
 	if ((amos_page = xpc_vars->amos_page) == NULL) {
-		amos_page = (AMO_t *) mspec_kalloc_page(0);
+		amos_page = (AMO_t *) TO_AMO(uncached_alloc_page(0));
 		if (amos_page == NULL) {
 			dev_err(xpc_part, "can't allocate page of AMOs\n");
 			return NULL;
@@ -200,7 +201,8 @@ xpc_rsvd_page_init(void)
 			if (ret != 0) {
 				dev_err(xpc_part, "can't change memory "
 					"protections\n");
-				mspec_kfree_page((unsigned long) amos_page);
+				uncached_free_page(__IA64_UNCACHED_OFFSET |
+						   TO_PHYS((u64) amos_page));
 				return NULL;
 			}
 		}
diff --git a/include/asm-ia64/sn/mspec.h b/include/asm-ia64/sn/mspec.h
new file mode 100644
index 00000000000..dbe13c6121a
--- /dev/null
+++ b/include/asm-ia64/sn/mspec.h
@@ -0,0 +1,59 @@
+/*
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (c) 2001-2004 Silicon Graphics, Inc.  All rights reserved.
+ */
+
+#ifndef _ASM_IA64_SN_MSPEC_H
+#define _ASM_IA64_SN_MSPEC_H
+
+#define FETCHOP_VAR_SIZE 64 /* 64 byte per fetchop variable */
+
+#define FETCHOP_LOAD		0
+#define FETCHOP_INCREMENT	8
+#define FETCHOP_DECREMENT	16
+#define FETCHOP_CLEAR		24
+
+#define FETCHOP_STORE		0
+#define FETCHOP_AND		24
+#define FETCHOP_OR		32
+
+#define FETCHOP_CLEAR_CACHE	56
+
+#define FETCHOP_LOAD_OP(addr, op) ( \
+         *(volatile long *)((char*) (addr) + (op)))
+
+#define FETCHOP_STORE_OP(addr, op, x) ( \
+         *(volatile long *)((char*) (addr) + (op)) = (long) (x))
+
+#ifdef __KERNEL__
+
+/*
+ * Each Atomic Memory Operation (AMO formerly known as fetchop)
+ * variable is 64 bytes long.  The first 8 bytes are used.  The
+ * remaining 56 bytes are unaddressable due to the operation taking
+ * that portion of the address.
+ *
+ * NOTE: The AMO_t _MUST_ be placed in either the first or second half
+ * of the cache line.  The cache line _MUST NOT_ be used for anything
+ * other than additional AMO_t entries.  This is because there are two
+ * addresses which reference the same physical cache line.  One will
+ * be a cached entry with the memory type bits all set.  This address
+ * may be loaded into processor cache.  The AMO_t will be referenced
+ * uncached via the memory special memory type.  If any portion of the
+ * cached cache-line is modified, when that line is flushed, it will
+ * overwrite the uncached value in physical memory and lead to
+ * inconsistency.
+ */
+typedef struct {
+        u64 variable;
+        u64 unused[7];
+} AMO_t;
+
+
+#endif /* __KERNEL__ */
+
+#endif /* _ASM_IA64_SN_MSPEC_H */
-- 
cgit v1.2.3-70-g09d2


From 400e65146c428d2ef677a927786fda2cec545a76 Mon Sep 17 00:00:00 2001
From: Bob Picco <bob.picco@hp.com>
Date: Tue, 21 Jun 2005 17:15:05 -0700
Subject: [PATCH] ia64: pfn_to_nid() implementation

pfn_to_nid is undefined.  We haven't had this interface on ia64.  The
sys_mbind patches need it.

Oh, the paddr_to_nid call could fail when DISCONTIG+NUMA is configured
because there isn't any ACPI SRAT NUMA information.

Signed-off-by: Bob Picco <bob.picco@hp.com>
Acked-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-ia64/mmzone.h | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

(limited to 'include')

diff --git a/include/asm-ia64/mmzone.h b/include/asm-ia64/mmzone.h
index 9491dacc89c..83ca4043fc1 100644
--- a/include/asm-ia64/mmzone.h
+++ b/include/asm-ia64/mmzone.h
@@ -15,6 +15,20 @@
 #include <asm/page.h>
 #include <asm/meminit.h>
 
+static inline int pfn_to_nid(unsigned long pfn)
+{
+#ifdef CONFIG_NUMA
+	extern int paddr_to_nid(unsigned long);
+	int nid = paddr_to_nid(pfn << PAGE_SHIFT);
+	if (nid < 0)
+		return 0;
+	else
+		return nid;
+#else
+	return 0;
+#endif
+}
+
 #ifdef CONFIG_DISCONTIGMEM
 
 #ifdef CONFIG_IA64_DIG /* DIG systems are small */
-- 
cgit v1.2.3-70-g09d2


From b4955ce3dd0818b56da532a16c9a4a3804a558ee Mon Sep 17 00:00:00 2001
From: Abhijit Karmarkar <abhijitk@veritas.com>
Date: Tue, 21 Jun 2005 17:15:13 -0700
Subject: [PATCH] msync: check pte dirty earlier

It's common practice to msync a large address range regularly, in which
often only a few ptes have actually been dirtied since the previous pass.

sync_pte_range then goes much faster if it tests whether pte is dirty
before locating and accessing each struct page cacheline; and it is hardly
slowed by ptep_clear_flush_dirty repeating that test in the opposite case,
when every pte actually is dirty.

But beware, s390's pte_dirty always says false, since its dirty bit is kept
in the storage key, located via the struct page address.  So skip this
optimization in its case: use a pte_maybe_dirty macro which just says true
if page_test_and_clear_dirty is implemented.

Signed-off-by: Abhijit Karmarkar <abhijitk@veritas.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-generic/pgtable.h | 3 +++
 mm/msync.c                    | 2 ++
 2 files changed, 5 insertions(+)

(limited to 'include')

diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index 1f4ec7b7027..f4059356517 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -125,6 +125,9 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addres
 
 #ifndef __HAVE_ARCH_PAGE_TEST_AND_CLEAR_DIRTY
 #define page_test_and_clear_dirty(page) (0)
+#define pte_maybe_dirty(pte)		pte_dirty(pte)
+#else
+#define pte_maybe_dirty(pte)		(1)
 #endif
 
 #ifndef __HAVE_ARCH_PAGE_TEST_AND_CLEAR_YOUNG
diff --git a/mm/msync.c b/mm/msync.c
index 090f426bca7..d0f5a1bce7c 100644
--- a/mm/msync.c
+++ b/mm/msync.c
@@ -34,6 +34,8 @@ static void sync_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
 
 		if (!pte_present(*pte))
 			continue;
+		if (!pte_maybe_dirty(*pte))
+			continue;
 		pfn = pte_pfn(*pte);
 		if (!pfn_valid(pfn))
 			continue;
-- 
cgit v1.2.3-70-g09d2


From 5b37b700f7c491a9320f4e29472bbaf23dded8fd Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Tue, 21 Jun 2005 17:15:18 -0700
Subject: [PATCH] ppc32: Added support for new MPC8548 family of PowerQUICC III
 processors

Added descriptions of the new MPC8548 family processors, e500 core and
peripherals.

Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/kernel/cputable.c        |  14 +++
 arch/ppc/syslib/mpc85xx_devices.c | 185 ++++++++++++++++++++++++++++++++++++++
 arch/ppc/syslib/mpc85xx_sys.c     | 105 ++++++++++++++++++++++
 include/asm-ppc/irq.h             |   6 ++
 include/asm-ppc/mpc85xx.h         |   7 +-
 include/linux/fsl_devices.h       |   8 +-
 6 files changed, 323 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/ppc/kernel/cputable.c b/arch/ppc/kernel/cputable.c
index d44b7dc5390..01c226008db 100644
--- a/arch/ppc/kernel/cputable.c
+++ b/arch/ppc/kernel/cputable.c
@@ -918,6 +918,20 @@ struct cpu_spec	cpu_specs[] = {
 		.dcache_bsize		= 32,
 		.num_pmcs		= 4,
 	},
+	{ 	/* e500v2 */
+		.pvr_mask		= 0xffff0000,
+		.pvr_value		= 0x80210000,
+		.cpu_name		= "e500v2",
+		/* xxx - galak: add CPU_FTR_MAYBE_CAN_DOZE */
+		.cpu_features		= CPU_FTR_SPLIT_ID_CACHE |
+			CPU_FTR_USE_TB | CPU_FTR_BIG_PHYS,
+		.cpu_user_features	= PPC_FEATURE_32 |
+			PPC_FEATURE_HAS_MMU | PPC_FEATURE_SPE_COMP |
+			PPC_FEATURE_HAS_EFP_SINGLE | PPC_FEATURE_HAS_EFP_DOUBLE,
+		.icache_bsize		= 32,
+		.dcache_bsize		= 32,
+		.num_pmcs		= 4,
+	},
 #endif
 #if !CLASSIC_PPC
 	{	/* default match */
diff --git a/arch/ppc/syslib/mpc85xx_devices.c b/arch/ppc/syslib/mpc85xx_devices.c
index 1e658ef57e7..b30c57d4518 100644
--- a/arch/ppc/syslib/mpc85xx_devices.c
+++ b/arch/ppc/syslib/mpc85xx_devices.c
@@ -40,6 +40,42 @@ static struct gianfar_platform_data mpc85xx_tsec2_pdata = {
 	.phy_reg_addr = MPC85xx_ENET1_OFFSET,
 };
 
+static struct gianfar_platform_data mpc85xx_etsec1_pdata = {
+	.device_flags = FSL_GIANFAR_DEV_HAS_GIGABIT |
+	    FSL_GIANFAR_DEV_HAS_COALESCE | FSL_GIANFAR_DEV_HAS_RMON |
+	    FSL_GIANFAR_DEV_HAS_MULTI_INTR |
+	    FSL_GIANFAR_DEV_HAS_CSUM | FSL_GIANFAR_DEV_HAS_VLAN |
+	    FSL_GIANFAR_DEV_HAS_EXTENDED_HASH,
+	.phy_reg_addr = MPC85xx_ENET1_OFFSET,
+};
+
+static struct gianfar_platform_data mpc85xx_etsec2_pdata = {
+	.device_flags = FSL_GIANFAR_DEV_HAS_GIGABIT |
+	    FSL_GIANFAR_DEV_HAS_COALESCE | FSL_GIANFAR_DEV_HAS_RMON |
+	    FSL_GIANFAR_DEV_HAS_MULTI_INTR |
+	    FSL_GIANFAR_DEV_HAS_CSUM | FSL_GIANFAR_DEV_HAS_VLAN |
+	    FSL_GIANFAR_DEV_HAS_EXTENDED_HASH,
+	.phy_reg_addr = MPC85xx_ENET1_OFFSET,
+};
+
+static struct gianfar_platform_data mpc85xx_etsec3_pdata = {
+	.device_flags = FSL_GIANFAR_DEV_HAS_GIGABIT |
+	    FSL_GIANFAR_DEV_HAS_COALESCE | FSL_GIANFAR_DEV_HAS_RMON |
+	    FSL_GIANFAR_DEV_HAS_MULTI_INTR |
+	    FSL_GIANFAR_DEV_HAS_CSUM | FSL_GIANFAR_DEV_HAS_VLAN |
+	    FSL_GIANFAR_DEV_HAS_EXTENDED_HASH,
+	.phy_reg_addr = MPC85xx_ENET1_OFFSET,
+};
+
+static struct gianfar_platform_data mpc85xx_etsec4_pdata = {
+	.device_flags = FSL_GIANFAR_DEV_HAS_GIGABIT |
+	    FSL_GIANFAR_DEV_HAS_COALESCE | FSL_GIANFAR_DEV_HAS_RMON |
+	    FSL_GIANFAR_DEV_HAS_MULTI_INTR |
+	    FSL_GIANFAR_DEV_HAS_CSUM | FSL_GIANFAR_DEV_HAS_VLAN |
+	    FSL_GIANFAR_DEV_HAS_EXTENDED_HASH,
+	.phy_reg_addr = MPC85xx_ENET1_OFFSET,
+};
+
 static struct gianfar_platform_data mpc85xx_fec_pdata = {
 	.phy_reg_addr = MPC85xx_ENET1_OFFSET,
 };
@@ -48,6 +84,10 @@ static struct fsl_i2c_platform_data mpc85xx_fsl_i2c_pdata = {
 	.device_flags = FSL_I2C_DEV_SEPARATE_DFSRR,
 };
 
+static struct fsl_i2c_platform_data mpc85xx_fsl_i2c2_pdata = {
+	.device_flags = FSL_I2C_DEV_SEPARATE_DFSRR,
+};
+
 static struct plat_serial8250_port serial_platform_data[] = {
 	[0] = {
 		.mapbase	= 0x4500,
@@ -536,6 +576,151 @@ struct platform_device ppc_sys_platform_devices[] = {
 		},
 	},
 #endif /* CONFIG_CPM2 */
+	[MPC85xx_eTSEC1] = {
+		.name = "fsl-gianfar",
+		.id	= 1,
+		.dev.platform_data = &mpc85xx_etsec1_pdata,
+		.num_resources	 = 4,
+		.resource = (struct resource[]) {
+			{
+				.start	= MPC85xx_ENET1_OFFSET,
+				.end	= MPC85xx_ENET1_OFFSET +
+						MPC85xx_ENET1_SIZE - 1,
+				.flags	= IORESOURCE_MEM,
+			},
+			{
+				.name	= "tx",
+				.start	= MPC85xx_IRQ_TSEC1_TX,
+				.end	= MPC85xx_IRQ_TSEC1_TX,
+				.flags	= IORESOURCE_IRQ,
+			},
+			{
+				.name	= "rx",
+				.start	= MPC85xx_IRQ_TSEC1_RX,
+				.end	= MPC85xx_IRQ_TSEC1_RX,
+				.flags	= IORESOURCE_IRQ,
+			},
+			{
+				.name	= "error",
+				.start	= MPC85xx_IRQ_TSEC1_ERROR,
+				.end	= MPC85xx_IRQ_TSEC1_ERROR,
+				.flags	= IORESOURCE_IRQ,
+			},
+		},
+	},
+	[MPC85xx_eTSEC2] = {
+		.name = "fsl-gianfar",
+		.id	= 2,
+		.dev.platform_data = &mpc85xx_etsec2_pdata,
+		.num_resources	 = 4,
+		.resource = (struct resource[]) {
+			{
+				.start	= MPC85xx_ENET2_OFFSET,
+				.end	= MPC85xx_ENET2_OFFSET +
+						MPC85xx_ENET2_SIZE - 1,
+				.flags	= IORESOURCE_MEM,
+			},
+			{
+				.name	= "tx",
+				.start	= MPC85xx_IRQ_TSEC2_TX,
+				.end	= MPC85xx_IRQ_TSEC2_TX,
+				.flags	= IORESOURCE_IRQ,
+			},
+			{
+				.name	= "rx",
+				.start	= MPC85xx_IRQ_TSEC2_RX,
+				.end	= MPC85xx_IRQ_TSEC2_RX,
+				.flags	= IORESOURCE_IRQ,
+			},
+			{
+				.name	= "error",
+				.start	= MPC85xx_IRQ_TSEC2_ERROR,
+				.end	= MPC85xx_IRQ_TSEC2_ERROR,
+				.flags	= IORESOURCE_IRQ,
+			},
+		},
+	},
+	[MPC85xx_eTSEC3] = {
+		.name = "fsl-gianfar",
+		.id	= 3,
+		.dev.platform_data = &mpc85xx_etsec3_pdata,
+		.num_resources	 = 4,
+		.resource = (struct resource[]) {
+			{
+				.start	= MPC85xx_ENET3_OFFSET,
+				.end	= MPC85xx_ENET3_OFFSET +
+						MPC85xx_ENET3_SIZE - 1,
+				.flags	= IORESOURCE_MEM,
+			},
+			{
+				.name	= "tx",
+				.start	= MPC85xx_IRQ_TSEC3_TX,
+				.end	= MPC85xx_IRQ_TSEC3_TX,
+				.flags	= IORESOURCE_IRQ,
+			},
+			{
+				.name	= "rx",
+				.start	= MPC85xx_IRQ_TSEC3_RX,
+				.end	= MPC85xx_IRQ_TSEC3_RX,
+				.flags	= IORESOURCE_IRQ,
+			},
+			{
+				.name	= "error",
+				.start	= MPC85xx_IRQ_TSEC3_ERROR,
+				.end	= MPC85xx_IRQ_TSEC3_ERROR,
+				.flags	= IORESOURCE_IRQ,
+			},
+		},
+	},
+	[MPC85xx_eTSEC4] = {
+		.name = "fsl-gianfar",
+		.id	= 4,
+		.dev.platform_data = &mpc85xx_etsec4_pdata,
+		.num_resources	 = 4,
+		.resource = (struct resource[]) {
+			{
+				.start	= 0x27000,
+				.end	= 0x27fff,
+				.flags	= IORESOURCE_MEM,
+			},
+			{
+				.name	= "tx",
+				.start	= MPC85xx_IRQ_TSEC4_TX,
+				.end	= MPC85xx_IRQ_TSEC4_TX,
+				.flags	= IORESOURCE_IRQ,
+			},
+			{
+				.name	= "rx",
+				.start	= MPC85xx_IRQ_TSEC4_RX,
+				.end	= MPC85xx_IRQ_TSEC4_RX,
+				.flags	= IORESOURCE_IRQ,
+			},
+			{
+				.name	= "error",
+				.start	= MPC85xx_IRQ_TSEC4_ERROR,
+				.end	= MPC85xx_IRQ_TSEC4_ERROR,
+				.flags	= IORESOURCE_IRQ,
+			},
+		},
+	},
+	[MPC85xx_IIC2] = {
+		.name = "fsl-i2c",
+		.id	= 2,
+		.dev.platform_data = &mpc85xx_fsl_i2c2_pdata,
+		.num_resources	 = 2,
+		.resource = (struct resource[]) {
+			{
+				.start	= 0x03100,
+				.end	= 0x031ff,
+				.flags	= IORESOURCE_MEM,
+			},
+			{
+				.start	= MPC85xx_IRQ_IIC1,
+				.end	= MPC85xx_IRQ_IIC1,
+				.flags	= IORESOURCE_IRQ,
+			},
+		},
+	},
 };
 
 static int __init mach_mpc85xx_fixup(struct platform_device *pdev)
diff --git a/arch/ppc/syslib/mpc85xx_sys.c b/arch/ppc/syslib/mpc85xx_sys.c
index d806a92a940..6e3184ab354 100644
--- a/arch/ppc/syslib/mpc85xx_sys.c
+++ b/arch/ppc/syslib/mpc85xx_sys.c
@@ -110,6 +110,111 @@ struct ppc_sys_spec ppc_sys_specs[] = {
 			MPC85xx_CPM_USB,
 		},
 	},
+	/* SVRs on 8548 rev1.0 matches for 8548/8547/8545 */
+	{
+		.ppc_sys_name	= "8548E",
+		.mask 		= 0xFFFF00F0,
+		.value 		= 0x80390010,
+		.num_devices	= 13,
+		.device_list	= (enum ppc_sys_devices[])
+		{
+			MPC85xx_eTSEC1, MPC85xx_eTSEC2, MPC85xx_eTSEC3,
+			MPC85xx_eTSEC4, MPC85xx_IIC1, MPC85xx_IIC2,
+			MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3,
+			MPC85xx_PERFMON, MPC85xx_DUART, MPC85xx_SEC2,
+		},
+	},
+	{
+		.ppc_sys_name	= "8548",
+		.mask 		= 0xFFFF00F0,
+		.value 		= 0x80310010,
+		.num_devices	= 12,
+		.device_list	= (enum ppc_sys_devices[])
+		{
+			MPC85xx_eTSEC1, MPC85xx_eTSEC2, MPC85xx_eTSEC3,
+			MPC85xx_eTSEC4, MPC85xx_IIC1, MPC85xx_IIC2,
+			MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3,
+			MPC85xx_PERFMON, MPC85xx_DUART,
+		},
+	},
+	{
+		.ppc_sys_name	= "8547E",
+		.mask 		= 0xFFFF00F0,
+		.value 		= 0x80390010,
+		.num_devices	= 13,
+		.device_list	= (enum ppc_sys_devices[])
+		{
+			MPC85xx_eTSEC1, MPC85xx_eTSEC2, MPC85xx_eTSEC3,
+			MPC85xx_eTSEC4, MPC85xx_IIC1, MPC85xx_IIC2,
+			MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3,
+			MPC85xx_PERFMON, MPC85xx_DUART, MPC85xx_SEC2,
+		},
+	},
+	{
+		.ppc_sys_name	= "8547",
+		.mask 		= 0xFFFF00F0,
+		.value 		= 0x80310010,
+		.num_devices	= 12,
+		.device_list	= (enum ppc_sys_devices[])
+		{
+			MPC85xx_eTSEC1, MPC85xx_eTSEC2, MPC85xx_eTSEC3,
+			MPC85xx_eTSEC4, MPC85xx_IIC1, MPC85xx_IIC2,
+			MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3,
+			MPC85xx_PERFMON, MPC85xx_DUART,
+		},
+	},
+	{
+		.ppc_sys_name	= "8545E",
+		.mask 		= 0xFFFF00F0,
+		.value 		= 0x80390010,
+		.num_devices	= 11,
+		.device_list	= (enum ppc_sys_devices[])
+		{
+			MPC85xx_eTSEC1, MPC85xx_eTSEC2,
+			MPC85xx_IIC1, MPC85xx_IIC2,
+			MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3,
+			MPC85xx_PERFMON, MPC85xx_DUART, MPC85xx_SEC2,
+		},
+	},
+	{
+		.ppc_sys_name	= "8545",
+		.mask 		= 0xFFFF00F0,
+		.value 		= 0x80310010,
+		.num_devices	= 10,
+		.device_list	= (enum ppc_sys_devices[])
+		{
+			MPC85xx_eTSEC1, MPC85xx_eTSEC2,
+			MPC85xx_IIC1, MPC85xx_IIC2,
+			MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3,
+			MPC85xx_PERFMON, MPC85xx_DUART,
+		},
+	},
+	{
+		.ppc_sys_name	= "8543E",
+		.mask 		= 0xFFFF00F0,
+		.value 		= 0x803A0010,
+		.num_devices	= 11,
+		.device_list	= (enum ppc_sys_devices[])
+		{
+			MPC85xx_eTSEC1, MPC85xx_eTSEC2,
+			MPC85xx_IIC1, MPC85xx_IIC2,
+			MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3,
+			MPC85xx_PERFMON, MPC85xx_DUART, MPC85xx_SEC2,
+		},
+	},
+	{
+		.ppc_sys_name	= "8543",
+		.mask 		= 0xFFFF00F0,
+		.value 		= 0x80320010,
+		.num_devices	= 10,
+		.device_list	= (enum ppc_sys_devices[])
+		{
+			MPC85xx_eTSEC1, MPC85xx_eTSEC2,
+			MPC85xx_IIC1, MPC85xx_IIC2,
+			MPC85xx_DMA0, MPC85xx_DMA1, MPC85xx_DMA2, MPC85xx_DMA3,
+			MPC85xx_PERFMON, MPC85xx_DUART,
+		},
+	},
 	{	/* default match */
 		.ppc_sys_name	= "",
 		.mask 		= 0x00000000,
diff --git a/include/asm-ppc/irq.h b/include/asm-ppc/irq.h
index 06b86be61ed..434fc515ba6 100644
--- a/include/asm-ppc/irq.h
+++ b/include/asm-ppc/irq.h
@@ -223,9 +223,15 @@ static __inline__ int irq_canonicalize(int irq)
 #define MPC85xx_IRQ_RIO_RX	(12 + MPC85xx_OPENPIC_IRQ_OFFSET)
 #define MPC85xx_IRQ_TSEC1_TX	(13 + MPC85xx_OPENPIC_IRQ_OFFSET)
 #define MPC85xx_IRQ_TSEC1_RX	(14 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_TSEC3_TX	(15 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_TSEC3_RX	(16 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_TSEC3_ERROR	(17 + MPC85xx_OPENPIC_IRQ_OFFSET)
 #define MPC85xx_IRQ_TSEC1_ERROR	(18 + MPC85xx_OPENPIC_IRQ_OFFSET)
 #define MPC85xx_IRQ_TSEC2_TX	(19 + MPC85xx_OPENPIC_IRQ_OFFSET)
 #define MPC85xx_IRQ_TSEC2_RX	(20 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_TSEC4_TX	(21 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_TSEC4_RX	(22 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_TSEC4_ERROR	(23 + MPC85xx_OPENPIC_IRQ_OFFSET)
 #define MPC85xx_IRQ_TSEC2_ERROR	(24 + MPC85xx_OPENPIC_IRQ_OFFSET)
 #define MPC85xx_IRQ_FEC		(25 + MPC85xx_OPENPIC_IRQ_OFFSET)
 #define MPC85xx_IRQ_DUART	(26 + MPC85xx_OPENPIC_IRQ_OFFSET)
diff --git a/include/asm-ppc/mpc85xx.h b/include/asm-ppc/mpc85xx.h
index 22713e33158..51b61d3305f 100644
--- a/include/asm-ppc/mpc85xx.h
+++ b/include/asm-ppc/mpc85xx.h
@@ -74,7 +74,7 @@ extern unsigned char __res[];
 #define MPC85xx_GUTS_OFFSET	(0xe0000)
 #define MPC85xx_GUTS_SIZE	(0x01000)
 #define MPC85xx_IIC1_OFFSET	(0x03000)
-#define MPC85xx_IIC1_SIZE	(0x01000)
+#define MPC85xx_IIC1_SIZE	(0x00100)
 #define MPC85xx_OPENPIC_OFFSET	(0x40000)
 #define MPC85xx_OPENPIC_SIZE	(0x40000)
 #define MPC85xx_PCI1_OFFSET	(0x08000)
@@ -127,6 +127,11 @@ enum ppc_sys_devices {
 	MPC85xx_CPM_MCC2,
 	MPC85xx_CPM_SMC1,
 	MPC85xx_CPM_SMC2,
+	MPC85xx_eTSEC1,
+	MPC85xx_eTSEC2,
+	MPC85xx_eTSEC3,
+	MPC85xx_eTSEC4,
+	MPC85xx_IIC2,
 };
 
 #endif /* CONFIG_85xx */
diff --git a/include/linux/fsl_devices.h b/include/linux/fsl_devices.h
index faaff4c6455..70f54af87b9 100644
--- a/include/linux/fsl_devices.h
+++ b/include/linux/fsl_devices.h
@@ -51,6 +51,7 @@ struct gianfar_platform_data {
 
 	/* board specific information */
 	u32 board_flags;
+	u32 phy_flags;
 	u32 phyid;
 	u32 interruptPHY;
 	u8 mac_addr[6];
@@ -61,9 +62,14 @@ struct gianfar_platform_data {
 #define FSL_GIANFAR_DEV_HAS_COALESCE		0x00000002
 #define FSL_GIANFAR_DEV_HAS_RMON		0x00000004
 #define FSL_GIANFAR_DEV_HAS_MULTI_INTR		0x00000008
+#define FSL_GIANFAR_DEV_HAS_CSUM		0x00000010
+#define FSL_GIANFAR_DEV_HAS_VLAN		0x00000020
+#define FSL_GIANFAR_DEV_HAS_EXTENDED_HASH	0x00000040
+#define FSL_GIANFAR_DEV_HAS_PADDING		0x00000080
 
 /* Flags in gianfar_platform_data */
-#define FSL_GIANFAR_BRD_HAS_PHY_INTR	0x00000001	/* if not set use a timer */
+#define FSL_GIANFAR_BRD_HAS_PHY_INTR	0x00000001 /* set or use a timer */
+#define FSL_GIANFAR_BRD_IS_REDUCED	0x00000002 /* Set if RGMII, RMII */
 
 struct fsl_i2c_platform_data {
 	/* device specific information */
-- 
cgit v1.2.3-70-g09d2


From c91999bba3ae8b21d4a965f19bf127bac3e5f7f2 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Tue, 21 Jun 2005 17:15:19 -0700
Subject: [PATCH] ppc32: Added preliminary support for the MPC8548 CDS board

Adds support for using the MPC8548 processor on the CDS reference board.
Currently all the major busses (PCI, PCI-X, PCI-Express, sRIO) and eTSEC3
and eTSEC4 are not supported.

Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/configs/mpc8548_cds_defconfig       | 659 +++++++++++++++++++++++++++
 arch/ppc/platforms/85xx/Kconfig              |  10 +
 arch/ppc/platforms/85xx/Makefile             |   1 +
 arch/ppc/platforms/85xx/mpc85xx_cds_common.c |  50 +-
 arch/ppc/syslib/Makefile                     |   1 +
 arch/ppc/syslib/ppc85xx_setup.c              |   6 +-
 include/asm-ppc/mpc85xx.h                    |   2 +-
 7 files changed, 713 insertions(+), 16 deletions(-)
 create mode 100644 arch/ppc/configs/mpc8548_cds_defconfig

(limited to 'include')

diff --git a/arch/ppc/configs/mpc8548_cds_defconfig b/arch/ppc/configs/mpc8548_cds_defconfig
new file mode 100644
index 00000000000..abe034f24b8
--- /dev/null
+++ b/arch/ppc/configs/mpc8548_cds_defconfig
@@ -0,0 +1,659 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.12-rc4
+# Tue May 24 22:36:27 2005
+#
+CONFIG_MMU=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_RWSEM_XCHGADD_ALGORITHM=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_HAVE_DEC_LOCK=y
+CONFIG_PPC=y
+CONFIG_PPC32=y
+CONFIG_GENERIC_NVRAM=y
+CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_CLEAN_COMPILE=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+CONFIG_SWAP=y
+CONFIG_SYSVIPC=y
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
+# CONFIG_HOTPLUG is not set
+CONFIG_KOBJECT_UEVENT=y
+# CONFIG_IKCONFIG is not set
+CONFIG_EMBEDDED=y
+# CONFIG_KALLSYMS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_BASE_FULL=y
+CONFIG_FUTEX=y
+# CONFIG_EPOLL is not set
+# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
+CONFIG_SHMEM=y
+CONFIG_CC_ALIGN_FUNCTIONS=0
+CONFIG_CC_ALIGN_LABELS=0
+CONFIG_CC_ALIGN_LOOPS=0
+CONFIG_CC_ALIGN_JUMPS=0
+# CONFIG_TINY_SHMEM is not set
+CONFIG_BASE_SMALL=0
+
+#
+# Loadable module support
+#
+# CONFIG_MODULES is not set
+
+#
+# Processor
+#
+# CONFIG_6xx is not set
+# CONFIG_40x is not set
+# CONFIG_44x is not set
+# CONFIG_POWER3 is not set
+# CONFIG_POWER4 is not set
+# CONFIG_8xx is not set
+CONFIG_E500=y
+CONFIG_BOOKE=y
+CONFIG_FSL_BOOKE=y
+# CONFIG_PHYS_64BIT is not set
+CONFIG_SPE=y
+CONFIG_MATH_EMULATION=y
+# CONFIG_CPU_FREQ is not set
+# CONFIG_PM is not set
+CONFIG_85xx=y
+CONFIG_PPC_INDIRECT_PCI_BE=y
+
+#
+# Freescale 85xx options
+#
+# CONFIG_MPC8540_ADS is not set
+CONFIG_MPC8548_CDS=y
+# CONFIG_MPC8555_CDS is not set
+# CONFIG_MPC8560_ADS is not set
+# CONFIG_SBC8560 is not set
+# CONFIG_STX_GP3 is not set
+CONFIG_MPC8548=y
+
+#
+# Platform options
+#
+# CONFIG_SMP is not set
+# CONFIG_PREEMPT is not set
+# CONFIG_HIGHMEM is not set
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_MISC is not set
+# CONFIG_CMDLINE_BOOL is not set
+CONFIG_ISA_DMA_API=y
+
+#
+# Bus options
+#
+# CONFIG_PCI is not set
+# CONFIG_PCI_DOMAINS is not set
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+# CONFIG_PCCARD is not set
+
+#
+# Advanced setup
+#
+# CONFIG_ADVANCED_OPTIONS is not set
+
+#
+# Default settings for advanced configuration options are used
+#
+CONFIG_HIGHMEM_START=0xfe000000
+CONFIG_LOWMEM_SIZE=0x30000000
+CONFIG_KERNEL_START=0xc0000000
+CONFIG_TASK_SIZE=0x80000000
+CONFIG_BOOT_LOAD=0x00800000
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+# CONFIG_FW_LOADER is not set
+
+#
+# Memory Technology Devices (MTD)
+#
+# CONFIG_MTD is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+
+#
+# Block devices
+#
+# CONFIG_BLK_DEV_FD is not set
+# CONFIG_BLK_DEV_COW_COMMON is not set
+CONFIG_BLK_DEV_LOOP=y
+# CONFIG_BLK_DEV_CRYPTOLOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=32768
+CONFIG_BLK_DEV_INITRD=y
+CONFIG_INITRAMFS_SOURCE=""
+# CONFIG_LBD is not set
+# CONFIG_CDROM_PKTCDVD is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+CONFIG_IOSCHED_AS=y
+CONFIG_IOSCHED_DEADLINE=y
+CONFIG_IOSCHED_CFQ=y
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_SCSI is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+# CONFIG_MD is not set
+
+#
+# Fusion MPT device support
+#
+
+#
+# IEEE 1394 (FireWire) support
+#
+
+#
+# I2O device support
+#
+
+#
+# Macintosh device drivers
+#
+
+#
+# Networking support
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+# CONFIG_PACKET_MMAP is not set
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+CONFIG_IP_MULTICAST=y
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+CONFIG_IP_PNP_BOOTP=y
+# CONFIG_IP_PNP_RARP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_IP_MROUTE is not set
+# CONFIG_ARPD is not set
+CONFIG_SYN_COOKIES=y
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_TUNNEL is not set
+CONFIG_IP_TCPDIAG=y
+# CONFIG_IP_TCPDIAG_IPV6 is not set
+# CONFIG_IPV6 is not set
+# CONFIG_NETFILTER is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+
+#
+# QoS and/or fair queueing
+#
+# CONFIG_NET_SCHED is not set
+# CONFIG_NET_CLS_ROUTE is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+CONFIG_NETDEVICES=y
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+
+#
+# Ethernet (1000 Mbit)
+#
+CONFIG_GIANFAR=y
+CONFIG_GFAR_NAPI=y
+
+#
+# Ethernet (10000 Mbit)
+#
+
+#
+# Token Ring devices
+#
+
+#
+# Wireless LAN (non-hamradio)
+#
+# CONFIG_NET_RADIO is not set
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+
+#
+# Userland interfaces
+#
+# CONFIG_INPUT_MOUSEDEV is not set
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_TSDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+# CONFIG_INPUT_KEYBOARD is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+CONFIG_SOUND_GAMEPORT=y
+
+#
+# Character devices
+#
+# CONFIG_VT is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+CONFIG_SERIAL_8250=y
+CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_SERIAL_8250_NR_UARTS=4
+# CONFIG_SERIAL_8250_EXTENDED is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+CONFIG_UNIX98_PTYS=y
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+# CONFIG_WATCHDOG is not set
+# CONFIG_NVRAM is not set
+CONFIG_GEN_RTC=y
+# CONFIG_GEN_RTC_X is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_AGP is not set
+# CONFIG_DRM is not set
+# CONFIG_RAW_DRIVER is not set
+
+#
+# TPM devices
+#
+
+#
+# I2C support
+#
+CONFIG_I2C=y
+CONFIG_I2C_CHARDEV=y
+
+#
+# I2C Algorithms
+#
+# CONFIG_I2C_ALGOBIT is not set
+# CONFIG_I2C_ALGOPCF is not set
+# CONFIG_I2C_ALGOPCA is not set
+
+#
+# I2C Hardware Bus support
+#
+# CONFIG_I2C_ISA is not set
+CONFIG_I2C_MPC=y
+# CONFIG_I2C_PARPORT_LIGHT is not set
+# CONFIG_I2C_PCA_ISA is not set
+
+#
+# Hardware Sensors Chip support
+#
+# CONFIG_I2C_SENSOR is not set
+# CONFIG_SENSORS_ADM1021 is not set
+# CONFIG_SENSORS_ADM1025 is not set
+# CONFIG_SENSORS_ADM1026 is not set
+# CONFIG_SENSORS_ADM1031 is not set
+# CONFIG_SENSORS_ASB100 is not set
+# CONFIG_SENSORS_DS1621 is not set
+# CONFIG_SENSORS_FSCHER is not set
+# CONFIG_SENSORS_FSCPOS is not set
+# CONFIG_SENSORS_GL518SM is not set
+# CONFIG_SENSORS_GL520SM is not set
+# CONFIG_SENSORS_IT87 is not set
+# CONFIG_SENSORS_LM63 is not set
+# CONFIG_SENSORS_LM75 is not set
+# CONFIG_SENSORS_LM77 is not set
+# CONFIG_SENSORS_LM78 is not set
+# CONFIG_SENSORS_LM80 is not set
+# CONFIG_SENSORS_LM83 is not set
+# CONFIG_SENSORS_LM85 is not set
+# CONFIG_SENSORS_LM87 is not set
+# CONFIG_SENSORS_LM90 is not set
+# CONFIG_SENSORS_LM92 is not set
+# CONFIG_SENSORS_MAX1619 is not set
+# CONFIG_SENSORS_PC87360 is not set
+# CONFIG_SENSORS_SMSC47B397 is not set
+# CONFIG_SENSORS_SMSC47M1 is not set
+# CONFIG_SENSORS_W83781D is not set
+# CONFIG_SENSORS_W83L785TS is not set
+# CONFIG_SENSORS_W83627HF is not set
+
+#
+# Other I2C Chip support
+#
+# CONFIG_SENSORS_DS1337 is not set
+# CONFIG_SENSORS_EEPROM is not set
+# CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_SENSORS_PCF8591 is not set
+# CONFIG_SENSORS_RTC8564 is not set
+# CONFIG_SENSORS_M41T00 is not set
+# CONFIG_I2C_DEBUG_CORE is not set
+# CONFIG_I2C_DEBUG_ALGO is not set
+# CONFIG_I2C_DEBUG_BUS is not set
+# CONFIG_I2C_DEBUG_CHIP is not set
+
+#
+# Dallas's 1-wire bus
+#
+# CONFIG_W1 is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+# CONFIG_FB is not set
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+
+#
+# USB support
+#
+# CONFIG_USB_ARCH_HAS_HCD is not set
+# CONFIG_USB_ARCH_HAS_OHCI is not set
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# InfiniBand support
+#
+# CONFIG_INFINIBAND is not set
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+# CONFIG_EXT2_FS_XATTR is not set
+CONFIG_EXT3_FS=y
+CONFIG_EXT3_FS_XATTR=y
+# CONFIG_EXT3_FS_POSIX_ACL is not set
+# CONFIG_EXT3_FS_SECURITY is not set
+CONFIG_JBD=y
+# CONFIG_JBD_DEBUG is not set
+CONFIG_FS_MBCACHE=y
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+
+#
+# XFS support
+#
+# CONFIG_XFS_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_ROMFS_FS is not set
+# CONFIG_QUOTA is not set
+CONFIG_DNOTIFY=y
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+# CONFIG_MSDOS_FS is not set
+# CONFIG_VFAT_FS is not set
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_KCORE=y
+CONFIG_SYSFS=y
+# CONFIG_DEVFS_FS is not set
+# CONFIG_DEVPTS_FS_XATTR is not set
+CONFIG_TMPFS=y
+# CONFIG_TMPFS_XATTR is not set
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_RAMFS=y
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+
+#
+# Network File Systems
+#
+CONFIG_NFS_FS=y
+# CONFIG_NFS_V3 is not set
+# CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
+# CONFIG_NFSD is not set
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_SUNRPC=y
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+# CONFIG_SMB_FS is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+
+#
+# Partition Types
+#
+CONFIG_PARTITION_ADVANCED=y
+# CONFIG_ACORN_PARTITION is not set
+# CONFIG_OSF_PARTITION is not set
+# CONFIG_AMIGA_PARTITION is not set
+# CONFIG_ATARI_PARTITION is not set
+# CONFIG_MAC_PARTITION is not set
+# CONFIG_MSDOS_PARTITION is not set
+# CONFIG_LDM_PARTITION is not set
+# CONFIG_SGI_PARTITION is not set
+# CONFIG_ULTRIX_PARTITION is not set
+# CONFIG_SUN_PARTITION is not set
+# CONFIG_EFI_PARTITION is not set
+
+#
+# Native Language Support
+#
+# CONFIG_NLS is not set
+
+#
+# Library routines
+#
+# CONFIG_CRC_CCITT is not set
+CONFIG_CRC32=y
+# CONFIG_LIBCRC32C is not set
+
+#
+# Profiling support
+#
+# CONFIG_PROFILING is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+# CONFIG_DEBUG_KERNEL is not set
+CONFIG_LOG_BUF_SHIFT=14
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+
+#
+# Cryptographic options
+#
+# CONFIG_CRYPTO is not set
+
+#
+# Hardware crypto devices
+#
diff --git a/arch/ppc/platforms/85xx/Kconfig b/arch/ppc/platforms/85xx/Kconfig
index ff92e38e7da..c5bc2821d99 100644
--- a/arch/ppc/platforms/85xx/Kconfig
+++ b/arch/ppc/platforms/85xx/Kconfig
@@ -21,6 +21,11 @@ config MPC8540_ADS
 	help
 	  This option enables support for the MPC 8540 ADS evaluation board.
 
+config MPC8548_CDS
+	bool "Freescale MPC8548 CDS"
+	help
+	  This option enablese support for the MPC8548 CDS evaluation board.
+
 config MPC8555_CDS
 	bool "Freescale MPC8555 CDS"
 	help
@@ -53,6 +58,11 @@ config MPC8540
 	depends on MPC8540_ADS
 	default y
 
+config MPC8548
+	bool
+	depends on MPC8548_CDS
+	default y
+
 config MPC8555
 	bool
 	depends on MPC8555_CDS
diff --git a/arch/ppc/platforms/85xx/Makefile b/arch/ppc/platforms/85xx/Makefile
index 854fbd298ba..efdf813108f 100644
--- a/arch/ppc/platforms/85xx/Makefile
+++ b/arch/ppc/platforms/85xx/Makefile
@@ -2,6 +2,7 @@
 # Makefile for the PowerPC 85xx linux kernel.
 #
 obj-$(CONFIG_MPC8540_ADS)	+= mpc85xx_ads_common.o mpc8540_ads.o
+obj-$(CONFIG_MPC8548_CDS)	+= mpc85xx_cds_common.o
 obj-$(CONFIG_MPC8555_CDS)	+= mpc85xx_cds_common.o
 obj-$(CONFIG_MPC8560_ADS)	+= mpc85xx_ads_common.o mpc8560_ads.o
 obj-$(CONFIG_SBC8560)		+= sbc85xx.o sbc8560.o
diff --git a/arch/ppc/platforms/85xx/mpc85xx_cds_common.c b/arch/ppc/platforms/85xx/mpc85xx_cds_common.c
index e7cfa498568..2d04bf599d2 100644
--- a/arch/ppc/platforms/85xx/mpc85xx_cds_common.c
+++ b/arch/ppc/platforms/85xx/mpc85xx_cds_common.c
@@ -480,21 +480,47 @@ mpc85xx_cds_setup_arch(void)
 
 	/* setup the board related information for the enet controllers */
 	pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC85xx_TSEC1);
-	pdata->board_flags = FSL_GIANFAR_BRD_HAS_PHY_INTR;
-	pdata->interruptPHY = MPC85xx_IRQ_EXT5;
-	pdata->phyid = 0;
-	/* fixup phy address */
-	pdata->phy_reg_addr += binfo->bi_immr_base;
-	memcpy(pdata->mac_addr, binfo->bi_enetaddr, 6);
+	if (pdata) {
+		pdata->board_flags = FSL_GIANFAR_BRD_HAS_PHY_INTR;
+		pdata->interruptPHY = MPC85xx_IRQ_EXT5;
+		pdata->phyid = 0;
+		/* fixup phy address */
+		pdata->phy_reg_addr += binfo->bi_immr_base;
+		memcpy(pdata->mac_addr, binfo->bi_enetaddr, 6);
+	}
 
 	pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC85xx_TSEC2);
-	pdata->board_flags = FSL_GIANFAR_BRD_HAS_PHY_INTR;
-	pdata->interruptPHY = MPC85xx_IRQ_EXT5;
-	pdata->phyid = 1;
-	/* fixup phy address */
-	pdata->phy_reg_addr += binfo->bi_immr_base;
-	memcpy(pdata->mac_addr, binfo->bi_enet1addr, 6);
+	if (pdata) {
+		pdata->board_flags = FSL_GIANFAR_BRD_HAS_PHY_INTR;
+		pdata->interruptPHY = MPC85xx_IRQ_EXT5;
+		pdata->phyid = 1;
+		/* fixup phy address */
+		pdata->phy_reg_addr += binfo->bi_immr_base;
+		memcpy(pdata->mac_addr, binfo->bi_enet1addr, 6);
+	}
+
+	pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC85xx_eTSEC1);
+	if (pdata) {
+		pdata->board_flags = FSL_GIANFAR_BRD_HAS_PHY_INTR;
+		pdata->interruptPHY = MPC85xx_IRQ_EXT5;
+		pdata->phyid = 0;
+		/* fixup phy address */
+		pdata->phy_reg_addr += binfo->bi_immr_base;
+		memcpy(pdata->mac_addr, binfo->bi_enetaddr, 6);
+	}
+
+	pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC85xx_eTSEC2);
+	if (pdata) {
+		pdata->board_flags = FSL_GIANFAR_BRD_HAS_PHY_INTR;
+		pdata->interruptPHY = MPC85xx_IRQ_EXT5;
+		pdata->phyid = 1;
+		/* fixup phy address */
+		pdata->phy_reg_addr += binfo->bi_immr_base;
+		memcpy(pdata->mac_addr, binfo->bi_enet1addr, 6);
+	}
 
+	ppc_sys_device_remove(MPC85xx_eTSEC3);
+	ppc_sys_device_remove(MPC85xx_eTSEC4);
 
 #ifdef CONFIG_BLK_DEV_INITRD
 	if (initrd_start)
diff --git a/arch/ppc/syslib/Makefile b/arch/ppc/syslib/Makefile
index 96acf85800d..6e7ecbc59aa 100644
--- a/arch/ppc/syslib/Makefile
+++ b/arch/ppc/syslib/Makefile
@@ -107,6 +107,7 @@ obj-$(CONFIG_83xx)		+= ipic.o ppc83xx_setup.o ppc_sys.o \
 ifeq ($(CONFIG_83xx),y)
 obj-$(CONFIG_PCI)		+= indirect_pci.o pci_auto.o
 endif
+obj-$(CONFIG_MPC8548_CDS)	+= todc_time.o
 obj-$(CONFIG_MPC8555_CDS)	+= todc_time.o
 obj-$(CONFIG_PPC_MPC52xx)	+= mpc52xx_setup.o mpc52xx_pic.o \
 					mpc52xx_sys.o mpc52xx_devices.o ppc_sys.o
diff --git a/arch/ppc/syslib/ppc85xx_setup.c b/arch/ppc/syslib/ppc85xx_setup.c
index f3277f469e7..455473f589d 100644
--- a/arch/ppc/syslib/ppc85xx_setup.c
+++ b/arch/ppc/syslib/ppc85xx_setup.c
@@ -133,7 +133,7 @@ mpc85xx_halt(void)
 
 #ifdef CONFIG_PCI
 
-#if defined(CONFIG_MPC8555_CDS)
+#if defined(CONFIG_MPC8555_CDS) || defined(CONFIG_MPC8548_CDS)
 extern void mpc85xx_cds_enable_via(struct pci_controller *hose);
 extern void mpc85xx_cds_fixup_via(struct pci_controller *hose);
 #endif
@@ -308,14 +308,14 @@ mpc85xx_setup_hose(void)
 
 	ppc_md.pci_exclude_device = mpc85xx_exclude_device;
 
-#if defined(CONFIG_MPC8555_CDS)
+#if defined(CONFIG_MPC8555_CDS) || defined(CONFIG_MPC8548_CDS)
 	/* Pre pciauto_bus_scan VIA init */
 	mpc85xx_cds_enable_via(hose_a);
 #endif
 
 	hose_a->last_busno = pciauto_bus_scan(hose_a, hose_a->first_busno);
 
-#if defined(CONFIG_MPC8555_CDS)
+#if defined(CONFIG_MPC8555_CDS) || defined(CONFIG_MPC8548_CDS)
 	/* Post pciauto_bus_scan VIA fixup */
 	mpc85xx_cds_fixup_via(hose_a);
 #endif
diff --git a/include/asm-ppc/mpc85xx.h b/include/asm-ppc/mpc85xx.h
index 51b61d3305f..95cc22da807 100644
--- a/include/asm-ppc/mpc85xx.h
+++ b/include/asm-ppc/mpc85xx.h
@@ -25,7 +25,7 @@
 #ifdef CONFIG_MPC8540_ADS
 #include <platforms/85xx/mpc8540_ads.h>
 #endif
-#ifdef CONFIG_MPC8555_CDS
+#if defined(CONFIG_MPC8555_CDS) || defined(CONFIG_MPC8548_CDS)
 #include <platforms/85xx/mpc8555_cds.h>
 #endif
 #ifdef CONFIG_MPC8560_ADS
-- 
cgit v1.2.3-70-g09d2


From b264c3527930ca4812fefc505fde9ad99d8a0117 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Tue, 21 Jun 2005 17:15:21 -0700
Subject: [PATCH] ppc32: Converted MPC10X bridge to use platform devices
 instead of OCP

Converted the MPC10x bridge support (used by MPC10x and 8240/1/5) to used
the standard platform device model.

Signed-off-by: Matt McClintock <msm@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/Kconfig                |   5 --
 arch/ppc/kernel/setup.c         |   4 +-
 arch/ppc/syslib/Makefile        |   2 +-
 arch/ppc/syslib/mpc10x_common.c | 183 ++++++++++++++++++++++++++++++----------
 include/asm-ppc/mpc10x.h        |   6 ++
 include/asm-ppc/ppc_sys.h       |   2 +
 6 files changed, 149 insertions(+), 53 deletions(-)

(limited to 'include')

diff --git a/arch/ppc/Kconfig b/arch/ppc/Kconfig
index 54ce6da2264..10162b187bc 100644
--- a/arch/ppc/Kconfig
+++ b/arch/ppc/Kconfig
@@ -826,11 +826,6 @@ config MPC10X_BRIDGE
 	depends on PCORE || POWERPMC250 || LOPEC || SANDPOINT
 	default y
 
-config FSL_OCP
-	bool
-	depends on MPC10X_BRIDGE
-	default y
-
 config MPC10X_OPENPIC
 	bool
 	depends on POWERPMC250 || LOPEC || SANDPOINT
diff --git a/arch/ppc/kernel/setup.c b/arch/ppc/kernel/setup.c
index 5c20266e3b1..c344c66e68f 100644
--- a/arch/ppc/kernel/setup.c
+++ b/arch/ppc/kernel/setup.c
@@ -41,7 +41,7 @@
 #include <asm/xmon.h>
 #include <asm/ocp.h>
 
-#if defined(CONFIG_85xx) || defined(CONFIG_83xx)
+#if defined(CONFIG_85xx) || defined(CONFIG_83xx) || defined(CONFIG_MPC10X_BRIDGE)
 #include <asm/ppc_sys.h>
 #endif
 
@@ -249,7 +249,7 @@ int show_cpuinfo(struct seq_file *m, void *v)
 	seq_printf(m, "bogomips\t: %lu.%02lu\n",
 		   lpj / (500000/HZ), (lpj / (5000/HZ)) % 100);
 
-#if defined(CONFIG_85xx) || defined(CONFIG_83xx)
+#if defined(CONFIG_85xx) || defined(CONFIG_83xx) || defined(CONFIG_MPC10X_BRIDGE)
 	if (cur_ppc_sys_spec->ppc_sys_name)
 		seq_printf(m, "chipset\t\t: %s\n",
 			cur_ppc_sys_spec->ppc_sys_name);
diff --git a/arch/ppc/syslib/Makefile b/arch/ppc/syslib/Makefile
index 6e7ecbc59aa..dec5bf4f687 100644
--- a/arch/ppc/syslib/Makefile
+++ b/arch/ppc/syslib/Makefile
@@ -92,7 +92,7 @@ ifeq ($(CONFIG_SERIAL_MPSC_CONSOLE),y)
 obj-$(CONFIG_SERIAL_TEXT_DEBUG)	+= mv64x60_dbg.o
 endif
 obj-$(CONFIG_BOOTX_TEXT)	+= btext.o
-obj-$(CONFIG_MPC10X_BRIDGE)     += mpc10x_common.o indirect_pci.o
+obj-$(CONFIG_MPC10X_BRIDGE)	+= mpc10x_common.o indirect_pci.o ppc_sys.o
 obj-$(CONFIG_MPC10X_OPENPIC)	+= open_pic.o
 obj-$(CONFIG_40x)		+= dcr.o
 obj-$(CONFIG_BOOKE)		+= dcr.o
diff --git a/arch/ppc/syslib/mpc10x_common.c b/arch/ppc/syslib/mpc10x_common.c
index fd93adfd464..0130aeb4a2d 100644
--- a/arch/ppc/syslib/mpc10x_common.c
+++ b/arch/ppc/syslib/mpc10x_common.c
@@ -21,6 +21,9 @@
 #include <linux/init.h>
 #include <linux/pci.h>
 #include <linux/slab.h>
+#include <linux/serial_8250.h>
+#include <linux/fsl_devices.h>
+#include <linux/device.h>
 
 #include <asm/byteorder.h>
 #include <asm/io.h>
@@ -30,16 +33,7 @@
 #include <asm/pci-bridge.h>
 #include <asm/open_pic.h>
 #include <asm/mpc10x.h>
-#include <asm/ocp.h>
-
-/* The OCP structure is fixed by code below, before OCP initialises.
-   paddr depends on where the board places the EUMB.
-    - fixed in mpc10x_bridge_init().
-   irq depends on two things:
-    > does the board use the EPIC at all? (PCORE does not).
-    > is the EPIC in serial or parallel mode?
-    - fixed in mpc10x_set_openpic().
-*/
+#include <asm/ppc_sys.h>
 
 #ifdef CONFIG_MPC10X_OPENPIC
 #ifdef CONFIG_EPIC_SERIAL_MODE
@@ -51,34 +45,127 @@
 #define MPC10X_DMA0_IRQ (EPIC_IRQ_BASE + 1 + NUM_8259_INTERRUPTS)
 #define MPC10X_DMA1_IRQ (EPIC_IRQ_BASE + 2 + NUM_8259_INTERRUPTS)
 #else
-#define MPC10X_I2C_IRQ OCP_IRQ_NA
-#define MPC10X_DMA0_IRQ OCP_IRQ_NA
-#define MPC10X_DMA1_IRQ OCP_IRQ_NA
+#define MPC10X_I2C_IRQ -1
+#define MPC10X_DMA0_IRQ -1
+#define MPC10X_DMA1_IRQ -1
 #endif
 
-
-struct ocp_def core_ocp[] = {
-	{ .vendor	= OCP_VENDOR_INVALID
-	}
+static struct fsl_i2c_platform_data mpc10x_i2c_pdata = {
+	.device_flags		= 0,
 };
 
-static struct ocp_fs_i2c_data mpc10x_i2c_data = {
-	.flags		= 0
+static struct plat_serial8250_port serial_platform_data[] = {
+	{ },
 };
-static struct ocp_def mpc10x_i2c_ocp = {
-	.vendor		= OCP_VENDOR_MOTOROLA,
-	.function	= OCP_FUNC_IIC,
-	.index		= 0,
-	.additions	= &mpc10x_i2c_data
+
+struct platform_device ppc_sys_platform_devices[] = {
+	[MPC10X_IIC1] = {
+		.name 	= "fsl-i2c",
+		.id	= 1,
+		.dev.platform_data = &mpc10x_i2c_pdata,
+		.num_resources = 2,
+		.resource = (struct resource[]) {
+			{
+				.start 	= MPC10X_EUMB_I2C_OFFSET,
+				.end	= MPC10X_EUMB_I2C_OFFSET +
+		                            MPC10X_EUMB_I2C_SIZE - 1,
+				.flags	= IORESOURCE_MEM,
+			},
+			{
+				.flags 	= IORESOURCE_IRQ
+			},
+		},
+	},
+	[MPC10X_DMA0] = {
+		.name	= "fsl-dma",
+		.id	= 0,
+		.num_resources = 2,
+		.resource = (struct resource[]) {
+			{
+				.start 	= MPC10X_EUMB_DMA_OFFSET + 0x10,
+				.end	= MPC10X_EUMB_DMA_OFFSET + 0x1f,
+				.flags	= IORESOURCE_MEM,
+			},
+			{
+				.flags	= IORESOURCE_IRQ,
+			},
+		},
+	},
+	[MPC10X_DMA1] = {
+		.name	= "fsl-dma",
+		.id	= 1,
+		.num_resources = 2,
+		.resource = (struct resource[]) {
+			{
+				.start	= MPC10X_EUMB_DMA_OFFSET + 0x20,
+				.end	= MPC10X_EUMB_DMA_OFFSET + 0x2f,
+				.flags	= IORESOURCE_MEM,
+			},
+			{
+				.flags	= IORESOURCE_IRQ,
+			},
+		},
+	},
+	[MPC10X_DMA1] = {
+		.name	= "fsl-dma",
+		.id	= 1,
+		.num_resources = 2,
+		.resource = (struct resource[]) {
+			{
+				.start	= MPC10X_EUMB_DMA_OFFSET + 0x20,
+				.end	= MPC10X_EUMB_DMA_OFFSET + 0x2f,
+				.flags	= IORESOURCE_MEM,
+			},
+			{
+				.flags	= IORESOURCE_IRQ,
+			},
+		},
+	},
+	[MPC10X_DUART] = {
+		.name = "serial8250",
+		.id	= 0,
+		.dev.platform_data = serial_platform_data,
+	},
 };
 
-static struct ocp_def mpc10x_dma_ocp[2] = {
-{	.vendor		= OCP_VENDOR_MOTOROLA,
-	.function	= OCP_FUNC_DMA,
-	.index		= 0 },
-{	.vendor		= OCP_VENDOR_MOTOROLA,
-	.function	= OCP_FUNC_DMA,
-	.index		= 1 }
+/* We use the PCI ID to match on */
+struct ppc_sys_spec *cur_ppc_sys_spec;
+struct ppc_sys_spec ppc_sys_specs[] = {
+	{
+		.ppc_sys_name 	= "8245",
+		.mask		= 0xFFFFFFFF,
+		.value		= MPC10X_BRIDGE_8245,
+		.num_devices	= 4,
+		.device_list	= (enum ppc_sys_devices[])
+		{
+			MPC10X_IIC1, MPC10X_DMA0, MPC10X_DMA1, MPC10X_DUART,
+		},
+	},
+	{
+		.ppc_sys_name 	= "8240",
+		.mask		= 0xFFFFFFFF,
+		.value		= MPC10X_BRIDGE_8240,
+		.num_devices	= 3,
+		.device_list	= (enum ppc_sys_devices[])
+		{
+			MPC10X_IIC1, MPC10X_DMA0, MPC10X_DMA1,
+		},
+	},
+	{
+		.ppc_sys_name	= "107",
+		.mask		= 0xFFFFFFFF,
+		.value		= MPC10X_BRIDGE_107,
+		.num_devices	= 3,
+		.device_list	= (enum ppc_sys_devices[])
+		{
+			MPC10X_IIC1, MPC10X_DMA0, MPC10X_DMA1,
+		},
+	},
+	{       /* default match */
+		.ppc_sys_name   = "",
+		.mask           = 0x00000000,
+		.value          = 0x00000000,
+	},
 };
 
 /* Set resources to match bridge memory map */
@@ -132,7 +219,7 @@ mpc10x_bridge_init(struct pci_controller *hose,
 		   uint new_map,
 		   uint phys_eumb_base)
 {
-	int	host_bridge, picr1, picr1_bit;
+	int	host_bridge, picr1, picr1_bit, i;
 	ulong	pci_config_addr, pci_config_data;
 	u_char	pir, byte;
 
@@ -273,7 +360,7 @@ mpc10x_bridge_init(struct pci_controller *hose,
 			printk("Host bridge in Agent mode\n");
 			/* Read or Set LMBAR & PCSRBAR? */
 		}
-		
+
 		/* Set base addr of the 8240/107 EUMB.  */
 		early_write_config_dword(hose,
 					 0,
@@ -287,17 +374,6 @@ mpc10x_bridge_init(struct pci_controller *hose,
 			ioremap(phys_eumb_base + MPC10X_EUMB_EPIC_OFFSET,
 				MPC10X_EUMB_EPIC_SIZE);
 #endif
-		mpc10x_i2c_ocp.paddr = phys_eumb_base + MPC10X_EUMB_I2C_OFFSET;
-		mpc10x_i2c_ocp.irq = MPC10X_I2C_IRQ;
-		ocp_add_one_device(&mpc10x_i2c_ocp);
-		mpc10x_dma_ocp[0].paddr = phys_eumb_base +
-					MPC10X_EUMB_DMA_OFFSET + 0x100;
-		mpc10x_dma_ocp[0].irq = MPC10X_DMA0_IRQ;
-		ocp_add_one_device(&mpc10x_dma_ocp[0]);
-		mpc10x_dma_ocp[1].paddr = phys_eumb_base +
-					MPC10X_EUMB_DMA_OFFSET + 0x200;
-		mpc10x_dma_ocp[1].irq = MPC10X_DMA1_IRQ;
-		ocp_add_one_device(&mpc10x_dma_ocp[1]);
 	}
 
 #ifdef CONFIG_MPC10X_STORE_GATHERING
@@ -306,6 +382,23 @@ mpc10x_bridge_init(struct pci_controller *hose,
 	mpc10x_disable_store_gathering(hose);
 #endif
 
+	/* setup platform devices for MPC10x bridges */
+	identify_ppc_sys_by_id (host_bridge);
+
+	for (i = 0; i < cur_ppc_sys_spec->num_devices; i++) {
+		unsigned int dev_id = cur_ppc_sys_spec->device_list[i];
+		ppc_sys_fixup_mem_resource(&ppc_sys_platform_devices[dev_id],
+			phys_eumb_base);
+	}
+
+	/* IRQ's are determined at runtime */
+	ppc_sys_platform_devices[MPC10X_IIC1].resource[1].start = MPC10X_I2C_IRQ;
+	ppc_sys_platform_devices[MPC10X_IIC1].resource[1].end = MPC10X_I2C_IRQ;
+	ppc_sys_platform_devices[MPC10X_DMA0].resource[1].start = MPC10X_DMA0_IRQ;
+	ppc_sys_platform_devices[MPC10X_DMA0].resource[1].end = MPC10X_DMA0_IRQ;
+	ppc_sys_platform_devices[MPC10X_DMA1].resource[1].start = MPC10X_DMA1_IRQ;
+	ppc_sys_platform_devices[MPC10X_DMA1].resource[1].end = MPC10X_DMA1_IRQ;
+
 	/*
 	 * 8240 erratum 26, 8241/8245 erratum 29, 107 erratum 23: speculative
 	 * PCI reads may return stale data so turn off.
@@ -330,7 +423,7 @@ mpc10x_bridge_init(struct pci_controller *hose,
 	 * 8245 (Rev 2., dated 10/2003) says PICR2[0] is reserverd.
 	 */
 	if (host_bridge == MPC10X_BRIDGE_8245) {
-		ulong	picr2;
+		u32	picr2;
 
 		early_read_config_dword(hose, 0, PCI_DEVFN(0,0),
 			MPC10X_CFG_PICR2_REG, &picr2);
diff --git a/include/asm-ppc/mpc10x.h b/include/asm-ppc/mpc10x.h
index d8e7e2d6128..f5196a4efbe 100644
--- a/include/asm-ppc/mpc10x.h
+++ b/include/asm-ppc/mpc10x.h
@@ -159,6 +159,12 @@ extern unsigned long			ioremap_base;
 #define	MPC10X_MAPA_EUMB_BASE		(ioremap_base - MPC10X_EUMB_SIZE)
 #define	MPC10X_MAPB_EUMB_BASE		MPC10X_MAPA_EUMB_BASE
 
+enum ppc_sys_devices {
+	MPC10X_IIC1,
+	MPC10X_DMA0,
+	MPC10X_DMA1,
+	MPC10X_DUART,
+};
 
 int mpc10x_bridge_init(struct pci_controller *hose,
 		       uint current_map,
diff --git a/include/asm-ppc/ppc_sys.h b/include/asm-ppc/ppc_sys.h
index 24b991c4276..8ea62456623 100644
--- a/include/asm-ppc/ppc_sys.h
+++ b/include/asm-ppc/ppc_sys.h
@@ -27,6 +27,8 @@
 #include <asm/mpc85xx.h>
 #elif defined(CONFIG_PPC_MPC52xx)
 #include <asm/mpc52xx.h>
+#elif defined(CONFIG_MPC10X_BRIDGE)
+#include <asm/mpc10x.h>
 #else
 #error "need definition of ppc_sys_devices"
 #endif
-- 
cgit v1.2.3-70-g09d2


From 65145e060b1c933ebc4215c3b493f586e08a1d5c Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Tue, 21 Jun 2005 17:15:25 -0700
Subject: [PATCH] ppc32: Added support for all MPC8548 internal interrupts

The MPC8548 has 48 internal interrupts and 12 external interrupts.  The
previous generation PowerQUICC III devices only had 32 internal and 12
external interrupts on the primary interrupt controller.

Expanded the number of internal interrupts to 48 for all PowerQUICC III
processors and moved the interrupt numbers for the external after the 48
internal interrupt lines, rather than putting the 12 new internal
interrupts at the end and ifdef'ng the whole mess.  As parted of this
created a macro which represents the internal interrupt senses since they
are the same on all PQ3 processors.

Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/platforms/85xx/mpc85xx_ads_common.c | 36 ++------------------
 arch/ppc/platforms/85xx/mpc85xx_cds_common.c | 41 ++++------------------
 arch/ppc/platforms/85xx/sbc85xx.c            | 36 ++------------------
 arch/ppc/platforms/85xx/stx_gp3.c            | 36 ++------------------
 include/asm-ppc/irq.h                        | 28 +++++++--------
 include/asm-ppc/mpc85xx.h                    | 51 ++++++++++++++++++++++++++++
 6 files changed, 77 insertions(+), 151 deletions(-)

(limited to 'include')

diff --git a/arch/ppc/platforms/85xx/mpc85xx_ads_common.c b/arch/ppc/platforms/85xx/mpc85xx_ads_common.c
index ba9f9f562c4..30397623fa6 100644
--- a/arch/ppc/platforms/85xx/mpc85xx_ads_common.c
+++ b/arch/ppc/platforms/85xx/mpc85xx_ads_common.c
@@ -59,40 +59,8 @@ extern unsigned long total_memory;	/* in mm/init */
 unsigned char __res[sizeof (bd_t)];
 
 /* Internal interrupts are all Level Sensitive, and Positive Polarity */
-
 static u_char mpc85xx_ads_openpic_initsenses[] __initdata = {
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  0: L2 Cache */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  1: ECM */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  2: DDR DRAM */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  3: LBIU */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  4: DMA 0 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  5: DMA 1 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  6: DMA 2 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  7: DMA 3 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  8: PCI/PCI-X */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  9: RIO Inbound Port Write Error */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 10: RIO Doorbell Inbound */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 11: RIO Outbound Message */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 12: RIO Inbound Message */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 13: TSEC 0 Transmit */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 14: TSEC 0 Receive */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 15: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 16: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 17: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 18: TSEC 0 Receive/Transmit Error */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 19: TSEC 1 Transmit */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 20: TSEC 1 Receive */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 21: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 22: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 23: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 24: TSEC 1 Receive/Transmit Error */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 25: Fast Ethernet */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 26: DUART */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 27: I2C */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 28: Performance Monitor */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 29: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 30: CPM */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 31: Unused */
+	MPC85XX_INTERNAL_IRQ_SENSES,
 	0x0,						/* External  0: */
 #if defined(CONFIG_PCI)
 	(IRQ_SENSE_LEVEL | IRQ_POLARITY_NEGATIVE),	/* External 1: PCI slot 0 */
@@ -159,7 +127,7 @@ mpc85xx_ads_init_IRQ(void)
 	/* Skip reserved space and internal sources */
 	openpic_set_sources(0, 32, OpenPIC_Addr + 0x10200);
 	/* Map PIC IRQs 0-11 */
-	openpic_set_sources(32, 12, OpenPIC_Addr + 0x10000);
+	openpic_set_sources(48, 12, OpenPIC_Addr + 0x10000);
 
 	/* we let openpic interrupts starting from an offset, to
 	 * leave space for cascading interrupts underneath.
diff --git a/arch/ppc/platforms/85xx/mpc85xx_cds_common.c b/arch/ppc/platforms/85xx/mpc85xx_cds_common.c
index 2d04bf599d2..bee6633f460 100644
--- a/arch/ppc/platforms/85xx/mpc85xx_cds_common.c
+++ b/arch/ppc/platforms/85xx/mpc85xx_cds_common.c
@@ -73,40 +73,8 @@ static int cds_pci_slot = 2;
 static volatile u8 * cadmus;
 
 /* Internal interrupts are all Level Sensitive, and Positive Polarity */
-
 static u_char mpc85xx_cds_openpic_initsenses[] __initdata = {
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  0: L2 Cache */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  1: ECM */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  2: DDR DRAM */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  3: LBIU */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  4: DMA 0 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  5: DMA 1 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  6: DMA 2 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  7: DMA 3 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  8: PCI/PCI-X */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  9: RIO Inbound Port Write Error */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 10: RIO Doorbell Inbound */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 11: RIO Outbound Message */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 12: RIO Inbound Message */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 13: TSEC 0 Transmit */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 14: TSEC 0 Receive */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 15: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 16: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 17: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 18: TSEC 0 Receive/Transmit Error */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 19: TSEC 1 Transmit */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 20: TSEC 1 Receive */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 21: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 22: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 23: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 24: TSEC 1 Receive/Transmit Error */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 25: Fast Ethernet */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 26: DUART */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 27: I2C */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 28: Performance Monitor */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 29: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 30: CPM */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 31: Unused */
+	MPC85XX_INTERNAL_IRQ_SENSES,
 #if defined(CONFIG_PCI)
 	(IRQ_SENSE_LEVEL | IRQ_POLARITY_NEGATIVE),	/* External 0: PCI1 slot */
 	(IRQ_SENSE_LEVEL | IRQ_POLARITY_NEGATIVE),	/* External 1: PCI1 slot */
@@ -182,7 +150,6 @@ void __init
 mpc85xx_cds_init_IRQ(void)
 {
 	bd_t *binfo = (bd_t *) __res;
-	int i;
 
 	/* Determine the Physical Address of the OpenPIC regs */
 	phys_addr_t OpenPIC_PAddr = binfo->bi_immr_base + MPC85xx_OPENPIC_OFFSET;
@@ -191,9 +158,13 @@ mpc85xx_cds_init_IRQ(void)
 	OpenPIC_NumInitSenses = sizeof (mpc85xx_cds_openpic_initsenses);
 
 	/* Skip reserved space and internal sources */
+#ifdef CONFIG_MPC8548
+	openpic_set_sources(0, 48, OpenPIC_Addr + 0x10200);
+#else
 	openpic_set_sources(0, 32, OpenPIC_Addr + 0x10200);
+#endif
 	/* Map PIC IRQs 0-11 */
-	openpic_set_sources(32, 12, OpenPIC_Addr + 0x10000);
+	openpic_set_sources(48, 12, OpenPIC_Addr + 0x10000);
 
 	/* we let openpic interrupts starting from an offset, to
 	 * leave space for cascading interrupts underneath.
diff --git a/arch/ppc/platforms/85xx/sbc85xx.c b/arch/ppc/platforms/85xx/sbc85xx.c
index 2d638c1c1bd..a7cb625b2d0 100644
--- a/arch/ppc/platforms/85xx/sbc85xx.c
+++ b/arch/ppc/platforms/85xx/sbc85xx.c
@@ -59,40 +59,8 @@ unsigned long pci_dram_offset = 0;
 extern unsigned long total_memory;	/* in mm/init */
 
 /* Internal interrupts are all Level Sensitive, and Positive Polarity */
-
 static u_char sbc8560_openpic_initsenses[] __initdata = {
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  0: L2 Cache */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  1: ECM */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  2: DDR DRAM */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  3: LBIU */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  4: DMA 0 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  5: DMA 1 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  6: DMA 2 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  7: DMA 3 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  8: PCI/PCI-X */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  9: RIO Inbound Port Write Error */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 10: RIO Doorbell Inbound */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 11: RIO Outbound Message */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 12: RIO Inbound Message */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 13: TSEC 0 Transmit */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 14: TSEC 0 Receive */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 15: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 16: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 17: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 18: TSEC 0 Receive/Transmit Error */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 19: TSEC 1 Transmit */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 20: TSEC 1 Receive */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 21: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 22: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 23: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 24: TSEC 1 Receive/Transmit Error */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 25: Fast Ethernet */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 26: DUART */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 27: I2C */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 28: Performance Monitor */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 29: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 30: CPM */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 31: Unused */
+	MPC85XX_INTERNAL_IRQ_SENSES,
 	0x0,				/* External  0: */
 	0x0,				/* External  1: */
 #if defined(CONFIG_PCI)
@@ -159,7 +127,7 @@ sbc8560_init_IRQ(void)
 	/* Skip reserved space and internal sources */
 	openpic_set_sources(0, 32, OpenPIC_Addr + 0x10200);
 	/* Map PIC IRQs 0-11 */
-	openpic_set_sources(32, 12, OpenPIC_Addr + 0x10000);
+	openpic_set_sources(48, 12, OpenPIC_Addr + 0x10000);
 
 	/* we let openpic interrupts starting from an offset, to 
 	 * leave space for cascading interrupts underneath.
diff --git a/arch/ppc/platforms/85xx/stx_gp3.c b/arch/ppc/platforms/85xx/stx_gp3.c
index bc95836e417..bc0a99638ff 100644
--- a/arch/ppc/platforms/85xx/stx_gp3.c
+++ b/arch/ppc/platforms/85xx/stx_gp3.c
@@ -72,38 +72,7 @@ unsigned long pci_dram_offset = 0;
 
 /* Internal interrupts are all Level Sensitive, and Positive Polarity */
 static u8 gp3_openpic_initsenses[] __initdata = {
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  0: L2 Cache */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  1: ECM */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  2: DDR DRAM */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  3: LBIU */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  4: DMA 0 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  5: DMA 1 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  6: DMA 2 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  7: DMA 3 */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  8: PCI/PCI-X */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  9: RIO Inbound Port Write Error */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 10: RIO Doorbell Inbound */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 11: RIO Outbound Message */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 12: RIO Inbound Message */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 13: TSEC 0 Transmit */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 14: TSEC 0 Receive */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 15: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 16: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 17: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 18: TSEC 0 Receive/Transmit Error */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 19: TSEC 1 Transmit */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 20: TSEC 1 Receive */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 21: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 22: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 23: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 24: TSEC 1 Receive/Transmit Error */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 25: Fast Ethernet */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 26: DUART */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 27: I2C */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 28: Performance Monitor */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 29: Unused */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 30: CPM */
-	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 31: Unused */
+	MPC85XX_INTERNAL_IRQ_SENSES,
 	0x0,						/* External  0: */
 #if defined(CONFIG_PCI)
 	(IRQ_SENSE_LEVEL | IRQ_POLARITY_NEGATIVE),	/* External 1: PCI slot 0 */
@@ -200,7 +169,6 @@ static struct irqaction cpm2_irqaction = {
 static void __init
 gp3_init_IRQ(void)
 {
-	int i;
 	bd_t *binfo = (bd_t *) __res;
 
 	/*
@@ -218,7 +186,7 @@ gp3_init_IRQ(void)
 	openpic_set_sources(0, 32, OpenPIC_Addr + 0x10200);
 
 	/* Map PIC IRQs 0-11 */
-	openpic_set_sources(32, 12, OpenPIC_Addr + 0x10000);
+	openpic_set_sources(48, 12, OpenPIC_Addr + 0x10000);
 
 	/*
 	 * Let openpic interrupts starting from an offset, to
diff --git a/include/asm-ppc/irq.h b/include/asm-ppc/irq.h
index 434fc515ba6..a9b33324f56 100644
--- a/include/asm-ppc/irq.h
+++ b/include/asm-ppc/irq.h
@@ -176,7 +176,7 @@ static __inline__ int irq_canonicalize(int irq)
 */
 #include <asm/mpc85xx.h>
 
-/* The MPC8560 openpic has  32 internal interrupts and 12 external
+/* The MPC8548 openpic has 48 internal interrupts and 12 external
  * interrupts.
  *
  * We are "flattening" the interrupt vectors of the cascaded CPM
@@ -184,7 +184,7 @@ static __inline__ int irq_canonicalize(int irq)
  * single integer.
  */
 #define NR_CPM_INTS	64
-#define NR_EPIC_INTS	44
+#define NR_EPIC_INTS	60
 #ifndef NR_8259_INTS
 #define NR_8259_INTS	0
 #endif
@@ -241,18 +241,18 @@ static __inline__ int irq_canonicalize(int irq)
 #define MPC85xx_IRQ_CPM		(30 + MPC85xx_OPENPIC_IRQ_OFFSET)
 
 /* The 12 external interrupt lines */
-#define MPC85xx_IRQ_EXT0        (32 + MPC85xx_OPENPIC_IRQ_OFFSET)
-#define MPC85xx_IRQ_EXT1        (33 + MPC85xx_OPENPIC_IRQ_OFFSET)
-#define MPC85xx_IRQ_EXT2        (34 + MPC85xx_OPENPIC_IRQ_OFFSET)
-#define MPC85xx_IRQ_EXT3        (35 + MPC85xx_OPENPIC_IRQ_OFFSET)
-#define MPC85xx_IRQ_EXT4        (36 + MPC85xx_OPENPIC_IRQ_OFFSET)
-#define MPC85xx_IRQ_EXT5        (37 + MPC85xx_OPENPIC_IRQ_OFFSET)
-#define MPC85xx_IRQ_EXT6        (38 + MPC85xx_OPENPIC_IRQ_OFFSET)
-#define MPC85xx_IRQ_EXT7        (39 + MPC85xx_OPENPIC_IRQ_OFFSET)
-#define MPC85xx_IRQ_EXT8        (40 + MPC85xx_OPENPIC_IRQ_OFFSET)
-#define MPC85xx_IRQ_EXT9        (41 + MPC85xx_OPENPIC_IRQ_OFFSET)
-#define MPC85xx_IRQ_EXT10       (42 + MPC85xx_OPENPIC_IRQ_OFFSET)
-#define MPC85xx_IRQ_EXT11       (43 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_EXT0        (48 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_EXT1        (49 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_EXT2        (50 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_EXT3        (51 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_EXT4        (52 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_EXT5        (53 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_EXT6        (54 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_EXT7        (55 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_EXT8        (56 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_EXT9        (57 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_EXT10       (58 + MPC85xx_OPENPIC_IRQ_OFFSET)
+#define MPC85xx_IRQ_EXT11       (59 + MPC85xx_OPENPIC_IRQ_OFFSET)
 
 /* CPM related interrupts */
 #define	SIU_INT_ERROR		((uint)0x00+CPM_IRQ_OFFSET)
diff --git a/include/asm-ppc/mpc85xx.h b/include/asm-ppc/mpc85xx.h
index 95cc22da807..516984ee14b 100644
--- a/include/asm-ppc/mpc85xx.h
+++ b/include/asm-ppc/mpc85xx.h
@@ -134,6 +134,57 @@ enum ppc_sys_devices {
 	MPC85xx_IIC2,
 };
 
+/* Internal interrupts are all Level Sensitive, and Positive Polarity */
+#define MPC85XX_INTERNAL_IRQ_SENSES \
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  0 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  1 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  2 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  3 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  4 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  5 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  6 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  7 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  8 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal  9 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 10 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 11 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 12 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 13 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 14 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 15 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 16 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 17 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 18 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 19 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 20 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 21 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 22 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 23 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 24 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 25 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 26 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 27 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 28 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 29 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 30 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 31 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 32 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 33 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 34 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 35 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 36 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 37 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 38 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 39 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 40 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 41 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 42 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 43 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 44 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 45 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE),	/* Internal 46 */	\
+	(IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE)	/* Internal 47 */
+
 #endif /* CONFIG_85xx */
 #endif /* __ASM_MPC85xx_H__ */
 #endif /* __KERNEL__ */
-- 
cgit v1.2.3-70-g09d2


From 5be061eee931db2718feecaf10df17610386202b Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Tue, 21 Jun 2005 17:15:26 -0700
Subject: [PATCH] ppc32: Clean up NUM_TLBCAMS usage for Freescale Book-E PPC's

Made the number of TLB CAM entries private and converted the board
consumers to use num_tlbcam_entries which is setup at boot time from
configuration registers.  This way the only consumers of the #define
NUM_TLBCAMS are the arrays used to manage the TLB.

Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/mm/fsl_booke_mmu.c                  | 2 ++
 arch/ppc/mm/mmu_decl.h                       | 2 ++
 arch/ppc/mm/pgtable.c                        | 1 -
 arch/ppc/platforms/85xx/mpc8540_ads.c        | 4 ++--
 arch/ppc/platforms/85xx/mpc85xx_cds_common.c | 4 ++--
 arch/ppc/platforms/85xx/sbc8560.c            | 4 ++--
 include/asm-ppc/pgtable.h                    | 2 --
 7 files changed, 10 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/arch/ppc/mm/fsl_booke_mmu.c b/arch/ppc/mm/fsl_booke_mmu.c
index 36233bdcdf8..e07990efa04 100644
--- a/arch/ppc/mm/fsl_booke_mmu.c
+++ b/arch/ppc/mm/fsl_booke_mmu.c
@@ -64,6 +64,8 @@ extern unsigned long total_lowmem;
 extern unsigned long __max_low_memory;
 #define MAX_LOW_MEM	CONFIG_LOWMEM_SIZE
 
+#define NUM_TLBCAMS	(16)
+
 struct tlbcam {
    	u32	MAS0;
 	u32	MAS1;
diff --git a/arch/ppc/mm/mmu_decl.h b/arch/ppc/mm/mmu_decl.h
index ffcdb46997d..540f3292b22 100644
--- a/arch/ppc/mm/mmu_decl.h
+++ b/arch/ppc/mm/mmu_decl.h
@@ -43,6 +43,8 @@ extern int mem_init_done;
 extern PTE *Hash, *Hash_end;
 extern unsigned long Hash_size, Hash_mask;
 
+extern unsigned int num_tlbcam_entries;
+
 /* ...and now those things that may be slightly different between processor
  * architectures.  -- Dan
  */
diff --git a/arch/ppc/mm/pgtable.c b/arch/ppc/mm/pgtable.c
index 5d2f3f66aef..81a3d7446d3 100644
--- a/arch/ppc/mm/pgtable.c
+++ b/arch/ppc/mm/pgtable.c
@@ -66,7 +66,6 @@ void setbat(int index, unsigned long virt, unsigned long phys,
 
 #ifdef HAVE_TLBCAM
 extern unsigned int tlbcam_index;
-extern unsigned int num_tlbcam_entries;
 extern unsigned long v_mapped_by_tlbcam(unsigned long va);
 extern unsigned long p_mapped_by_tlbcam(unsigned long pa);
 #else /* !HAVE_TLBCAM */
diff --git a/arch/ppc/platforms/85xx/mpc8540_ads.c b/arch/ppc/platforms/85xx/mpc8540_ads.c
index 583838ab02d..d1c2c6cb300 100644
--- a/arch/ppc/platforms/85xx/mpc8540_ads.c
+++ b/arch/ppc/platforms/85xx/mpc8540_ads.c
@@ -88,7 +88,7 @@ mpc8540ads_setup_arch(void)
 #ifdef CONFIG_SERIAL_TEXT_DEBUG
 	/* Invalidate the entry we stole earlier the serial ports
 	 * should be properly mapped */
-	invalidate_tlbcam_entry(NUM_TLBCAMS - 1);
+	invalidate_tlbcam_entry(num_tlbcam_entries - 1);
 #endif
 
 	/* setup the board related information for the enet controllers */
@@ -150,7 +150,7 @@ platform_init(unsigned long r3, unsigned long r4, unsigned long r5,
 		struct uart_port p;
 
 		/* Use the last TLB entry to map CCSRBAR to allow access to DUART regs */
-		settlbcam(NUM_TLBCAMS - 1, binfo->bi_immr_base,
+		settlbcam(num_tlbcam_entries - 1, binfo->bi_immr_base,
 			  binfo->bi_immr_base, MPC85xx_CCSRBAR_SIZE, _PAGE_IO, 0);
 
 		memset(&p, 0, sizeof (p));
diff --git a/arch/ppc/platforms/85xx/mpc85xx_cds_common.c b/arch/ppc/platforms/85xx/mpc85xx_cds_common.c
index bee6633f460..51964b3081e 100644
--- a/arch/ppc/platforms/85xx/mpc85xx_cds_common.c
+++ b/arch/ppc/platforms/85xx/mpc85xx_cds_common.c
@@ -446,7 +446,7 @@ mpc85xx_cds_setup_arch(void)
 #ifdef CONFIG_SERIAL_TEXT_DEBUG
 	/* Invalidate the entry we stole earlier the serial ports
 	 * should be properly mapped */
-	invalidate_tlbcam_entry(NUM_TLBCAMS - 1);
+	invalidate_tlbcam_entry(num_tlbcam_entries - 1);
 #endif
 
 	/* setup the board related information for the enet controllers */
@@ -528,7 +528,7 @@ platform_init(unsigned long r3, unsigned long r4, unsigned long r5,
 		struct uart_port p;
 
 		/* Use the last TLB entry to map CCSRBAR to allow access to DUART regs */
-		settlbcam(NUM_TLBCAMS - 1, binfo->bi_immr_base,
+		settlbcam(num_tlbcam_entries - 1, binfo->bi_immr_base,
 				binfo->bi_immr_base, MPC85xx_CCSRBAR_SIZE, _PAGE_IO, 0);
 
 		memset(&p, 0, sizeof (p));
diff --git a/arch/ppc/platforms/85xx/sbc8560.c b/arch/ppc/platforms/85xx/sbc8560.c
index 7b9e1543e17..42e960a0bf3 100644
--- a/arch/ppc/platforms/85xx/sbc8560.c
+++ b/arch/ppc/platforms/85xx/sbc8560.c
@@ -125,7 +125,7 @@ sbc8560_setup_arch(void)
 #ifdef CONFIG_SERIAL_TEXT_DEBUG
 	/* Invalidate the entry we stole earlier the serial ports
 	 * should be properly mapped */ 
-	invalidate_tlbcam_entry(NUM_TLBCAMS - 1);
+	invalidate_tlbcam_entry(num_tlbcam_entries - 1);
 #endif
 
 	/* setup the board related information for the enet controllers */
@@ -176,7 +176,7 @@ platform_init(unsigned long r3, unsigned long r4, unsigned long r5,
 
 #ifdef CONFIG_SERIAL_TEXT_DEBUG
 	/* Use the last TLB entry to map CCSRBAR to allow access to DUART regs */
-	settlbcam(NUM_TLBCAMS - 1, UARTA_ADDR,
+	settlbcam(num_tlbcam_entries - 1, UARTA_ADDR,
 		  UARTA_ADDR, 0x1000, _PAGE_IO, 0);
 #endif
 
diff --git a/include/asm-ppc/pgtable.h b/include/asm-ppc/pgtable.h
index a38606d1554..4d4b20c9de7 100644
--- a/include/asm-ppc/pgtable.h
+++ b/include/asm-ppc/pgtable.h
@@ -267,8 +267,6 @@ extern unsigned long ioremap_bot, ioremap_base;
 #define _PMD_PRESENT_MASK (PAGE_MASK)
 #define _PMD_BAD	(~PAGE_MASK)
 
-#define NUM_TLBCAMS	(16)
-
 #elif defined(CONFIG_8xx)
 /* Definitions for 8xx embedded chips. */
 #define _PAGE_PRESENT	0x0001	/* Page is valid */
-- 
cgit v1.2.3-70-g09d2


From 20cee16ced631f70a62c97bdebae08a1c9470448 Mon Sep 17 00:00:00 2001
From: David Gibson <david@gibson.dropbear.id.au>
Date: Tue, 21 Jun 2005 17:15:31 -0700
Subject: [PATCH] ppc64: Abolish ioremap_mm

Currently ppc64 has two mm_structs for the kernel, init_mm and also
ioremap_mm.  The latter really isn't necessary: this patch abolishes it,
instead restricting vmallocs to the lower 1TB of the init_mm's range and
placing io mappings in the upper 1TB.  This simplifies the code in a number
of places and eliminates an unecessary set of pagetables.  It also tweaks
the unmap/free path a little, allowing us to remove the unmap_im_area() set
of page table walkers, replacing them with unmap_vm_area().

Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/eeh.c       |  2 +-
 arch/ppc64/kernel/head.S      |  4 --
 arch/ppc64/kernel/process.c   |  8 ----
 arch/ppc64/mm/hash_utils.c    |  4 --
 arch/ppc64/mm/imalloc.c       | 20 ++++++----
 arch/ppc64/mm/init.c          | 93 +++++--------------------------------------
 include/asm-ppc64/imalloc.h   | 12 +++---
 include/asm-ppc64/page.h      |  2 -
 include/asm-ppc64/pgtable.h   |  9 +----
 include/asm-ppc64/processor.h | 10 -----
 10 files changed, 31 insertions(+), 133 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/eeh.c b/arch/ppc64/kernel/eeh.c
index d63d41f3eec..af5272fedad 100644
--- a/arch/ppc64/kernel/eeh.c
+++ b/arch/ppc64/kernel/eeh.c
@@ -505,7 +505,7 @@ static inline unsigned long eeh_token_to_phys(unsigned long token)
 	pte_t *ptep;
 	unsigned long pa;
 
-	ptep = find_linux_pte(ioremap_mm.pgd, token);
+	ptep = find_linux_pte(init_mm.pgd, token);
 	if (!ptep)
 		return token;
 	pa = pte_pfn(*ptep) << PAGE_SHIFT;
diff --git a/arch/ppc64/kernel/head.S b/arch/ppc64/kernel/head.S
index 346dbf606b5..02c8f4e3e4b 100644
--- a/arch/ppc64/kernel/head.S
+++ b/arch/ppc64/kernel/head.S
@@ -2121,10 +2121,6 @@ empty_zero_page:
 swapper_pg_dir:
 	.space	4096
 
-	.globl	ioremap_dir
-ioremap_dir:
-	.space	4096
-
 #ifdef CONFIG_SMP
 /* 1 page segment table per cpu (max 48, cpu0 allocated at STAB0_PHYS_ADDR) */
 	.globl	stab_array
diff --git a/arch/ppc64/kernel/process.c b/arch/ppc64/kernel/process.c
index cdfecbeb331..aba89554d89 100644
--- a/arch/ppc64/kernel/process.c
+++ b/arch/ppc64/kernel/process.c
@@ -58,14 +58,6 @@ struct task_struct *last_task_used_math = NULL;
 struct task_struct *last_task_used_altivec = NULL;
 #endif
 
-struct mm_struct ioremap_mm = {
-	.pgd		= ioremap_dir,
-	.mm_users	= ATOMIC_INIT(2),
-	.mm_count	= ATOMIC_INIT(1),
-	.cpu_vm_mask	= CPU_MASK_ALL,
-	.page_table_lock = SPIN_LOCK_UNLOCKED,
-};
-
 /*
  * Make sure the floating-point register state in the
  * the thread_struct is up to date for task tsk.
diff --git a/arch/ppc64/mm/hash_utils.c b/arch/ppc64/mm/hash_utils.c
index 0a0f97008d0..87d0525f56f 100644
--- a/arch/ppc64/mm/hash_utils.c
+++ b/arch/ppc64/mm/hash_utils.c
@@ -310,10 +310,6 @@ int hash_page(unsigned long ea, unsigned long access, unsigned long trap)
 
 		vsid = get_vsid(mm->context.id, ea);
 		break;
-	case IO_REGION_ID:
-		mm = &ioremap_mm;
-		vsid = get_kernel_vsid(ea);
-		break;
 	case VMALLOC_REGION_ID:
 		mm = &init_mm;
 		vsid = get_kernel_vsid(ea);
diff --git a/arch/ppc64/mm/imalloc.c b/arch/ppc64/mm/imalloc.c
index cb8727f3267..b6e75b891ac 100644
--- a/arch/ppc64/mm/imalloc.c
+++ b/arch/ppc64/mm/imalloc.c
@@ -15,6 +15,7 @@
 #include <asm/pgtable.h>
 #include <asm/semaphore.h>
 #include <asm/imalloc.h>
+#include <asm/cacheflush.h>
 
 static DECLARE_MUTEX(imlist_sem);
 struct vm_struct * imlist = NULL;
@@ -285,29 +286,32 @@ struct vm_struct * im_get_area(unsigned long v_addr, unsigned long size,
 	return area;
 }
 
-unsigned long im_free(void * addr)
+void im_free(void * addr)
 {
 	struct vm_struct **p, *tmp;
-	unsigned long ret_size = 0;
   
 	if (!addr)
-		return ret_size;
-	if ((PAGE_SIZE-1) & (unsigned long) addr) {
+		return;
+	if ((unsigned long) addr & ~PAGE_MASK) {
 		printk(KERN_ERR "Trying to %s bad address (%p)\n", __FUNCTION__,			addr);
-		return ret_size;
+		return;
 	}
 	down(&imlist_sem);
 	for (p = &imlist ; (tmp = *p) ; p = &tmp->next) {
 		if (tmp->addr == addr) {
-			ret_size = tmp->size;
 			*p = tmp->next;
+
+			/* XXX: do we need the lock? */
+			spin_lock(&init_mm.page_table_lock);
+			unmap_vm_area(tmp);
+			spin_unlock(&init_mm.page_table_lock);
+
 			kfree(tmp);
 			up(&imlist_sem);
-			return ret_size;
+			return;
 		}
 	}
 	up(&imlist_sem);
 	printk(KERN_ERR "Trying to %s nonexistent area (%p)\n", __FUNCTION__,
 			addr);
-	return ret_size;
 }
diff --git a/arch/ppc64/mm/init.c b/arch/ppc64/mm/init.c
index 4b42aff74d7..6fa1e6490b5 100644
--- a/arch/ppc64/mm/init.c
+++ b/arch/ppc64/mm/init.c
@@ -73,9 +73,6 @@ static unsigned long phbs_io_bot = PHBS_IO_BASE;
 extern pgd_t swapper_pg_dir[];
 extern struct task_struct *current_set[NR_CPUS];
 
-extern pgd_t ioremap_dir[];
-pgd_t * ioremap_pgd = (pgd_t *)&ioremap_dir;
-
 unsigned long klimit = (unsigned long)_end;
 
 unsigned long _SDR1=0;
@@ -137,69 +134,6 @@ void iounmap(volatile void __iomem *addr)
 
 #else
 
-static void unmap_im_area_pte(pmd_t *pmd, unsigned long addr,
-				  unsigned long end)
-{
-	pte_t *pte;
-
-	pte = pte_offset_kernel(pmd, addr);
-	do {
-		pte_t ptent = ptep_get_and_clear(&ioremap_mm, addr, pte);
-		WARN_ON(!pte_none(ptent) && !pte_present(ptent));
-	} while (pte++, addr += PAGE_SIZE, addr != end);
-}
-
-static inline void unmap_im_area_pmd(pud_t *pud, unsigned long addr,
-				     unsigned long end)
-{
-	pmd_t *pmd;
-	unsigned long next;
-
-	pmd = pmd_offset(pud, addr);
-	do {
-		next = pmd_addr_end(addr, end);
-		if (pmd_none_or_clear_bad(pmd))
-			continue;
-		unmap_im_area_pte(pmd, addr, next);
-	} while (pmd++, addr = next, addr != end);
-}
-
-static inline void unmap_im_area_pud(pgd_t *pgd, unsigned long addr,
-				     unsigned long end)
-{
-	pud_t *pud;
-	unsigned long next;
-
-	pud = pud_offset(pgd, addr);
-	do {
-		next = pud_addr_end(addr, end);
-		if (pud_none_or_clear_bad(pud))
-			continue;
-		unmap_im_area_pmd(pud, addr, next);
-	} while (pud++, addr = next, addr != end);
-}
-
-static void unmap_im_area(unsigned long addr, unsigned long end)
-{
-	struct mm_struct *mm = &ioremap_mm;
-	unsigned long next;
-	pgd_t *pgd;
-
-	spin_lock(&mm->page_table_lock);
-
-	pgd = pgd_offset_i(addr);
-	flush_cache_vunmap(addr, end);
-	do {
-		next = pgd_addr_end(addr, end);
-		if (pgd_none_or_clear_bad(pgd))
-			continue;
-		unmap_im_area_pud(pgd, addr, next);
-	} while (pgd++, addr = next, addr != end);
-	flush_tlb_kernel_range(start, end);
-
-	spin_unlock(&mm->page_table_lock);
-}
-
 /*
  * map_io_page currently only called by __ioremap
  * map_io_page adds an entry to the ioremap page table
@@ -214,21 +148,21 @@ static int map_io_page(unsigned long ea, unsigned long pa, int flags)
 	unsigned long vsid;
 
 	if (mem_init_done) {
-		spin_lock(&ioremap_mm.page_table_lock);
-		pgdp = pgd_offset_i(ea);
-		pudp = pud_alloc(&ioremap_mm, pgdp, ea);
+		spin_lock(&init_mm.page_table_lock);
+		pgdp = pgd_offset_k(ea);
+		pudp = pud_alloc(&init_mm, pgdp, ea);
 		if (!pudp)
 			return -ENOMEM;
-		pmdp = pmd_alloc(&ioremap_mm, pudp, ea);
+		pmdp = pmd_alloc(&init_mm, pudp, ea);
 		if (!pmdp)
 			return -ENOMEM;
-		ptep = pte_alloc_kernel(&ioremap_mm, pmdp, ea);
+		ptep = pte_alloc_kernel(&init_mm, pmdp, ea);
 		if (!ptep)
 			return -ENOMEM;
 		pa = abs_to_phys(pa);
-		set_pte_at(&ioremap_mm, ea, ptep, pfn_pte(pa >> PAGE_SHIFT,
+		set_pte_at(&init_mm, ea, ptep, pfn_pte(pa >> PAGE_SHIFT,
 							  __pgprot(flags)));
-		spin_unlock(&ioremap_mm.page_table_lock);
+		spin_unlock(&init_mm.page_table_lock);
 	} else {
 		unsigned long va, vpn, hash, hpteg;
 
@@ -267,13 +201,9 @@ static void __iomem * __ioremap_com(unsigned long addr, unsigned long pa,
 
 	for (i = 0; i < size; i += PAGE_SIZE)
 		if (map_io_page(ea+i, pa+i, flags))
-			goto failure;
+			return NULL;
 
 	return (void __iomem *) (ea + (addr & ~PAGE_MASK));
- failure:
-	if (mem_init_done)
-		unmap_im_area(ea, ea + size);
-	return NULL;
 }
 
 
@@ -381,19 +311,14 @@ int __ioremap_explicit(unsigned long pa, unsigned long ea,
  */
 void iounmap(volatile void __iomem *token)
 {
-	unsigned long address, size;
 	void *addr;
 
 	if (!mem_init_done)
 		return;
 	
 	addr = (void *) ((unsigned long __force) token & PAGE_MASK);
-	
-	if ((size = im_free(addr)) == 0)
-		return;
 
-	address = (unsigned long)addr; 
-	unmap_im_area(address, address + size);
+	im_free(addr);
 }
 
 static int iounmap_subset_regions(unsigned long addr, unsigned long size)
diff --git a/include/asm-ppc64/imalloc.h b/include/asm-ppc64/imalloc.h
index 3a45e918bf1..e46ff68a6e4 100644
--- a/include/asm-ppc64/imalloc.h
+++ b/include/asm-ppc64/imalloc.h
@@ -4,9 +4,9 @@
 /*
  * Define the address range of the imalloc VM area.
  */
-#define PHBS_IO_BASE  	  IOREGIONBASE
-#define IMALLOC_BASE      (IOREGIONBASE + 0x80000000ul)	/* Reserve 2 gigs for PHBs */
-#define IMALLOC_END       (IOREGIONBASE + EADDR_MASK)
+#define PHBS_IO_BASE  	  VMALLOC_END
+#define IMALLOC_BASE      (PHBS_IO_BASE + 0x80000000ul)	/* Reserve 2 gigs for PHBs */
+#define IMALLOC_END       (VMALLOC_START + EADDR_MASK)
 
 
 /* imalloc region types */
@@ -18,7 +18,9 @@
 
 extern struct vm_struct * im_get_free_area(unsigned long size);
 extern struct vm_struct * im_get_area(unsigned long v_addr, unsigned long size,
-			int region_type);
-unsigned long im_free(void *addr);
+				      int region_type);
+extern void im_free(void *addr);
+
+extern unsigned long ioremap_bot;
 
 #endif /* _PPC64_IMALLOC_H */
diff --git a/include/asm-ppc64/page.h b/include/asm-ppc64/page.h
index bcd21789d3b..257d87eb7c3 100644
--- a/include/asm-ppc64/page.h
+++ b/include/asm-ppc64/page.h
@@ -202,9 +202,7 @@ extern u64 ppc64_pft_size;		/* Log 2 of page table size */
 #define PAGE_OFFSET     ASM_CONST(0xC000000000000000)
 #define KERNELBASE      PAGE_OFFSET
 #define VMALLOCBASE     ASM_CONST(0xD000000000000000)
-#define IOREGIONBASE    ASM_CONST(0xE000000000000000)
 
-#define IO_REGION_ID       (IOREGIONBASE >> REGION_SHIFT)
 #define VMALLOC_REGION_ID  (VMALLOCBASE >> REGION_SHIFT)
 #define KERNEL_REGION_ID   (KERNELBASE >> REGION_SHIFT)
 #define USER_REGION_ID     (0UL)
diff --git a/include/asm-ppc64/pgtable.h b/include/asm-ppc64/pgtable.h
index 264c4f7993b..46cf61c2ff6 100644
--- a/include/asm-ppc64/pgtable.h
+++ b/include/asm-ppc64/pgtable.h
@@ -53,7 +53,8 @@
  * Define the address range of the vmalloc VM area.
  */
 #define VMALLOC_START (0xD000000000000000ul)
-#define VMALLOC_END   (VMALLOC_START + EADDR_MASK)
+#define VMALLOC_SIZE  (0x10000000000UL)
+#define VMALLOC_END   (VMALLOC_START + VMALLOC_SIZE)
 
 /*
  * Bits in a linux-style PTE.  These match the bits in the
@@ -239,9 +240,6 @@ static inline pte_t pfn_pte(unsigned long pfn, pgprot_t pgprot)
 /* This now only contains the vmalloc pages */
 #define pgd_offset_k(address) pgd_offset(&init_mm, address)
 
-/* to find an entry in the ioremap page-table-directory */
-#define pgd_offset_i(address) (ioremap_pgd + pgd_index(address))
-
 /*
  * The following only work if pte_present() is true.
  * Undefined behaviour if not..
@@ -459,15 +457,12 @@ extern pgprot_t phys_mem_access_prot(struct file *file, unsigned long addr,
 #define __HAVE_ARCH_PTE_SAME
 #define pte_same(A,B)	(((pte_val(A) ^ pte_val(B)) & ~_PAGE_HPTEFLAGS) == 0)
 
-extern unsigned long ioremap_bot, ioremap_base;
-
 #define pmd_ERROR(e) \
 	printk("%s:%d: bad pmd %08x.\n", __FILE__, __LINE__, pmd_val(e))
 #define pgd_ERROR(e) \
 	printk("%s:%d: bad pgd %08x.\n", __FILE__, __LINE__, pgd_val(e))
 
 extern pgd_t swapper_pg_dir[];
-extern pgd_t ioremap_dir[];
 
 extern void paging_init(void);
 
diff --git a/include/asm-ppc64/processor.h b/include/asm-ppc64/processor.h
index 809c634ba1d..3084099086a 100644
--- a/include/asm-ppc64/processor.h
+++ b/include/asm-ppc64/processor.h
@@ -429,16 +429,6 @@ struct thread_struct {
 	.fpexc_mode = MSR_FE0|MSR_FE1, \
 }
 
-/*
- * Note: the vm_start and vm_end fields here should *not*
- * be in kernel space.  (Could vm_end == vm_start perhaps?)
- */
-#define IOREMAP_MMAP { &ioremap_mm, 0, 0x1000, NULL, \
-		    PAGE_SHARED, VM_READ | VM_WRITE | VM_EXEC, \
-		    1, NULL, NULL }
-
-extern struct mm_struct ioremap_mm;
-
 /*
  * Return saved PC of a blocked thread. For now, this is the "user" PC
  */
-- 
cgit v1.2.3-70-g09d2


From 0e3e4a1c4dade7872fdb04a5e735ae26d8e2fbfb Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:33 -0700
Subject: [PATCH] ppc64 iSeries: remove iSeries_proc.h

include/asm-ppc64/iSeries/iSeries_proc.h just contains a declaration of a
function that no longer exists.  Remove it.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/iSeries_proc.c         |  1 -
 arch/ppc64/kernel/iSeries_setup.c        |  1 -
 arch/ppc64/kernel/viopath.c              |  1 -
 include/asm-ppc64/iSeries/iSeries_proc.h | 24 ------------------------
 4 files changed, 27 deletions(-)
 delete mode 100644 include/asm-ppc64/iSeries/iSeries_proc.h

(limited to 'include')

diff --git a/arch/ppc64/kernel/iSeries_proc.c b/arch/ppc64/kernel/iSeries_proc.c
index 0cc58ddf48d..bf85a1f7e87 100644
--- a/arch/ppc64/kernel/iSeries_proc.c
+++ b/arch/ppc64/kernel/iSeries_proc.c
@@ -29,7 +29,6 @@
 #include <asm/iSeries/HvCallXm.h>
 #include <asm/iSeries/IoHriMainStore.h>
 #include <asm/iSeries/LparData.h>
-#include <asm/iSeries/iSeries_proc.h>
 
 static int __init iseries_proc_create(void)
 {
diff --git a/arch/ppc64/kernel/iSeries_setup.c b/arch/ppc64/kernel/iSeries_setup.c
index 6d06eb550a3..e80cfc57092 100644
--- a/arch/ppc64/kernel/iSeries_setup.c
+++ b/arch/ppc64/kernel/iSeries_setup.c
@@ -55,7 +55,6 @@
 #include <asm/iSeries/HvCallXm.h>
 #include <asm/iSeries/ItLpQueue.h>
 #include <asm/iSeries/IoHriMainStore.h>
-#include <asm/iSeries/iSeries_proc.h>
 #include <asm/iSeries/mf.h>
 #include <asm/iSeries/HvLpEvent.h>
 #include <asm/iSeries/iSeries_irq.h>
diff --git a/arch/ppc64/kernel/viopath.c b/arch/ppc64/kernel/viopath.c
index 2ed8ee07568..8caa078a658 100644
--- a/arch/ppc64/kernel/viopath.c
+++ b/arch/ppc64/kernel/viopath.c
@@ -48,7 +48,6 @@
 #include <asm/iSeries/HvLpConfig.h>
 #include <asm/iSeries/HvCallCfg.h>
 #include <asm/iSeries/mf.h>
-#include <asm/iSeries/iSeries_proc.h>
 #include <asm/iSeries/vio.h>
 
 /* Status of the path to each other partition in the system.
diff --git a/include/asm-ppc64/iSeries/iSeries_proc.h b/include/asm-ppc64/iSeries/iSeries_proc.h
deleted file mode 100644
index adb6dc14dd4..00000000000
--- a/include/asm-ppc64/iSeries/iSeries_proc.h
+++ /dev/null
@@ -1,24 +0,0 @@
-/*
- * iSeries_proc.h
- * Copyright (C) 2001  Kyle A. Lucke IBM Corporation
- * 
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- * 
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- * 
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
- */
-#ifndef _ISERIES_PROC_H
-#define _ISERIES_PROC_H
-
-extern void iSeries_proc_early_init(void);
-
-#endif /* _iSeries_PROC_H */
-- 
cgit v1.2.3-70-g09d2


From 45dc76aaf63fc336527da80337d3e3684f50f387 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:33 -0700
Subject: [PATCH] ppc64 iSeries: header file white space cleanups

This patch just contains white space and comment cleanups in the iSeries
headers files.  There are no semantic changes.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-ppc64/iSeries/HvCall.h        | 100 ++++-----
 include/asm-ppc64/iSeries/HvCallCfg.h     | 116 +++++-----
 include/asm-ppc64/iSeries/HvCallEvent.h   |  29 ++-
 include/asm-ppc64/iSeries/HvCallHpt.h     |  89 ++++----
 include/asm-ppc64/iSeries/HvCallPci.h     | 340 +++++++++++++-----------------
 include/asm-ppc64/iSeries/HvCallSc.h      |  38 ++--
 include/asm-ppc64/iSeries/HvCallSm.h      |  31 +--
 include/asm-ppc64/iSeries/HvCallXm.h      |  63 ++----
 include/asm-ppc64/iSeries/HvLpConfig.h    | 291 +++++++++++++------------
 include/asm-ppc64/iSeries/HvLpEvent.h     | 116 +++++-----
 include/asm-ppc64/iSeries/HvReleaseData.h |  76 ++++---
 include/asm-ppc64/iSeries/HvTypes.h       | 108 +++++-----
 12 files changed, 644 insertions(+), 753 deletions(-)

(limited to 'include')

diff --git a/include/asm-ppc64/iSeries/HvCall.h b/include/asm-ppc64/iSeries/HvCall.h
index d9a2e74e239..c903387e912 100644
--- a/include/asm-ppc64/iSeries/HvCall.h
+++ b/include/asm-ppc64/iSeries/HvCall.h
@@ -1,34 +1,28 @@
 /*
  * HvCall.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
  */
-
-//===========================================================================
-//
-//	This file contains the "hypervisor call" interface which is used to
-//	drive the hypervisor from the OS.
-//
-//===========================================================================
+/*
+ * This file contains the "hypervisor call" interface which is used to
+ * drive the hypervisor from the OS.
+ */
 #ifndef _HVCALL_H
 #define _HVCALL_H
 
-//-------------------------------------------------------------------
-// Standard Includes
-//-------------------------------------------------------------------
 #include <asm/iSeries/HvCallSc.h>
 #include <asm/iSeries/HvTypes.h>
 #include <asm/paca.h>
@@ -76,9 +70,9 @@ enum HvCall_VaryOffChunkRc
 */
 
 /* Type of yield for HvCallBaseYieldProcessor */
-#define HvCall_YieldTimed 	0	// Yield until specified time (tb)
-#define HvCall_YieldToActive	1	// Yield until all active procs have run
-#define HvCall_YieldToProc	2	// Yield until the specified processor has run
+#define HvCall_YieldTimed	0	/* Yield until specified time (tb) */
+#define HvCall_YieldToActive	1	/* Yield until all active procs have run */
+#define HvCall_YieldToProc	2	/* Yield until the specified processor has run */
 
 /* interrupt masks for setEnabledInterrupts */
 #define HvCall_MaskIPI		0x00000001
@@ -86,7 +80,7 @@ enum HvCall_VaryOffChunkRc
 #define HvCall_MaskLpProd	0x00000004
 #define HvCall_MaskTimeout	0x00000008
 
-/* Log buffer formats                       */
+/* Log buffer formats */
 #define HvCall_LogBuffer_ASCII          0
 #define HvCall_LogBuffer_EBCDIC         1
 
@@ -95,7 +89,7 @@ enum HvCall_VaryOffChunkRc
 #define HvCallBaseGetHwPatch				HvCallBase +  2
 #define HvCallBaseReIplSpAttn				HvCallBase +  3
 #define HvCallBaseSetASR				HvCallBase +  4
-#define HvCallBaseSetASRAndRfi				HvCallBase +  5 
+#define HvCallBaseSetASRAndRfi				HvCallBase +  5
 #define HvCallBaseSetIMR				HvCallBase +  6
 #define HvCallBaseSendIPI				HvCallBase +  7
 #define HvCallBaseTerminateMachine			HvCallBase +  8
@@ -115,81 +109,75 @@ enum HvCall_VaryOffChunkRc
 #define HvCallBaseGetLogBufferCodePage			HvCallBase + 22
 #define HvCallBaseGetLogBufferFormat			HvCallBase + 23
 #define HvCallBaseGetLogBufferLength			HvCallBase + 24
-#define HvCallBaseReadLogBuffer 			HvCallBase + 25
+#define HvCallBaseReadLogBuffer				HvCallBase + 25
 #define HvCallBaseSetLogBufferFormatAndCodePage		HvCallBase + 26
-#define HvCallBaseWriteLogBuffer               		HvCallBase + 27
+#define HvCallBaseWriteLogBuffer			HvCallBase + 27
 #define HvCallBaseRouter28				HvCallBase + 28
 #define HvCallBaseRouter29				HvCallBase + 29
 #define HvCallBaseRouter30				HvCallBase + 30
-#define HvCallBaseSetDebugBus 				HvCallBase + 31
+#define HvCallBaseSetDebugBus				HvCallBase + 31
 
-#define HvCallCcSetDABR  				HvCallCc + 7
+#define HvCallCcSetDABR					HvCallCc + 7
 
-//=====================================================================================
-static inline void		HvCall_setVirtualDecr(void)
+static inline void HvCall_setVirtualDecr(void)
 {
-	/* Ignore any error return codes - most likely means that the target value for the
-	 * LP has been increased and this vary off would bring us below the new target. */
+	/*
+	 * Ignore any error return codes - most likely means that the
+	 * target value for the LP has been increased and this vary off
+	 * would bring us below the new target.
+	 */
 	HvCall0(HvCallBaseSetVirtualDecr);
 }
-//=====================================================================
-static inline void		HvCall_yieldProcessor(unsigned typeOfYield, u64 yieldParm)
+
+static inline void HvCall_yieldProcessor(unsigned typeOfYield, u64 yieldParm)
 {
-	HvCall2( HvCallBaseYieldProcessor, typeOfYield, yieldParm );
+	HvCall2(HvCallBaseYieldProcessor, typeOfYield, yieldParm);
 }
-//=====================================================================
-static inline void		HvCall_setEnabledInterrupts(u64 enabledInterrupts)
+
+static inline void HvCall_setEnabledInterrupts(u64 enabledInterrupts)
 {
-	HvCall1(HvCallBaseSetEnabledInterrupts,enabledInterrupts);
+	HvCall1(HvCallBaseSetEnabledInterrupts, enabledInterrupts);
 }
 
-//=====================================================================
-static inline void		HvCall_clearLogBuffer(HvLpIndex lpindex)
+static inline void HvCall_clearLogBuffer(HvLpIndex lpindex)
 {
-	HvCall1(HvCallBaseClearLogBuffer,lpindex);
+	HvCall1(HvCallBaseClearLogBuffer, lpindex);
 }
 
-//=====================================================================
-static inline u32  		HvCall_getLogBufferCodePage(HvLpIndex lpindex)
+static inline u32 HvCall_getLogBufferCodePage(HvLpIndex lpindex)
 {
-	u32 retVal = HvCall1(HvCallBaseGetLogBufferCodePage,lpindex);
+	u32 retVal = HvCall1(HvCallBaseGetLogBufferCodePage, lpindex);
 	return retVal;
 }
 
-//=====================================================================
-static inline int  		HvCall_getLogBufferFormat(HvLpIndex lpindex)
+static inline int HvCall_getLogBufferFormat(HvLpIndex lpindex)
 {
-	int retVal = HvCall1(HvCallBaseGetLogBufferFormat,lpindex);
+	int retVal = HvCall1(HvCallBaseGetLogBufferFormat, lpindex);
 	return retVal;
 }
 
-//=====================================================================
-static inline u32  		HvCall_getLogBufferLength(HvLpIndex lpindex)
+static inline u32 HvCall_getLogBufferLength(HvLpIndex lpindex)
 {
-	u32 retVal = HvCall1(HvCallBaseGetLogBufferLength,lpindex);
+	u32 retVal = HvCall1(HvCallBaseGetLogBufferLength, lpindex);
 	return retVal;
 }
 
-//=====================================================================
-static inline void  		HvCall_setLogBufferFormatAndCodepage(int format, u32 codePage)
+static inline void HvCall_setLogBufferFormatAndCodepage(int format, u32 codePage)
 {
-	HvCall2(HvCallBaseSetLogBufferFormatAndCodePage,format, codePage);
+	HvCall2(HvCallBaseSetLogBufferFormatAndCodePage, format, codePage);
 }
 
-//=====================================================================
-int HvCall_readLogBuffer(HvLpIndex lpindex, void *buffer, u64 bufLen);
-void HvCall_writeLogBuffer(const void *buffer, u64 bufLen);
+extern int HvCall_readLogBuffer(HvLpIndex lpindex, void *buffer, u64 bufLen);
+extern void HvCall_writeLogBuffer(const void *buffer, u64 bufLen);
 
-//=====================================================================
-static inline void		HvCall_sendIPI(struct paca_struct * targetPaca)
+static inline void HvCall_sendIPI(struct paca_struct *targetPaca)
 {
-	HvCall1( HvCallBaseSendIPI, targetPaca->paca_index );
+	HvCall1(HvCallBaseSendIPI, targetPaca->paca_index);
 }
 
-//=====================================================================
-static inline void		HvCall_terminateMachineSrc(void)
+static inline void HvCall_terminateMachineSrc(void)
 {
-	HvCall0( HvCallBaseTerminateMachineSrc );
+	HvCall0(HvCallBaseTerminateMachineSrc);
 }
 
 static inline void HvCall_setDABR(unsigned long val)
diff --git a/include/asm-ppc64/iSeries/HvCallCfg.h b/include/asm-ppc64/iSeries/HvCallCfg.h
index 9f40f16de53..2223a9647d0 100644
--- a/include/asm-ppc64/iSeries/HvCallCfg.h
+++ b/include/asm-ppc64/iSeries/HvCallCfg.h
@@ -1,43 +1,32 @@
 /*
  * HvCallCfg.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
  */
-
-//=====================================================================================
-//
-//	This file contains the "hypervisor call" interface which is used to
-//	drive the hypervisor from the OS.
-//
-//=====================================================================================
+/*
+ * This file contains the "hypervisor call" interface which is used to
+ * drive the hypervisor from the OS.
+ */
 #ifndef _HVCALLCFG_H
 #define _HVCALLCFG_H
 
-//-------------------------------------------------------------------
-// Standard Includes
-//-------------------------------------------------------------------
 #include <asm/iSeries/HvCallSc.h>
 #include <asm/iSeries/HvTypes.h>
 
-//-------------------------------------------------------------------------------------
-// Constants
-//-------------------------------------------------------------------------------------
-
-enum HvCallCfg_ReqQual
-{
+enum HvCallCfg_ReqQual {
 	HvCallCfg_Cur	= 0,
 	HvCallCfg_Init	= 1,
 	HvCallCfg_Max	= 2,
@@ -49,7 +38,7 @@ enum HvCallCfg_ReqQual
 #define HvCallCfgGetLpVrmIndex				HvCallCfg +  2
 #define HvCallCfgGetLpMinSupportedPlicVrmIndex		HvCallCfg +  3
 #define HvCallCfgGetLpMinCompatablePlicVrmIndex		HvCallCfg +  4
-#define HvCallCfgGetLpVrmName				HvCallCfg +  5 
+#define HvCallCfgGetLpVrmName				HvCallCfg +  5
 #define HvCallCfgGetSystemPhysicalProcessors		HvCallCfg +  6
 #define HvCallCfgGetPhysicalProcessors			HvCallCfg +  7
 #define HvCallCfgGetSystemMsChunks			HvCallCfg +  8
@@ -76,108 +65,113 @@ enum HvCallCfg_ReqQual
 #define HvCallCfgSetMinRuntimeMsChunks			HvCallCfg + 29
 #define HvCallCfgGetVirtualLanIndexMap			HvCallCfg + 30
 #define HvCallCfgGetLpExecutionMode			HvCallCfg + 31
-#define HvCallCfgGetHostingLpIndex 			HvCallCfg + 32
+#define HvCallCfgGetHostingLpIndex			HvCallCfg + 32
 
-//====================================================================
 static inline HvLpIndex	HvCallCfg_getLps(void)
 {
 	HvLpIndex retVal = HvCall0(HvCallCfgGetLps);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
-//====================================================================
-static inline int		HvCallCfg_isBusDedicated(u64 busIndex)
+
+static inline int HvCallCfg_isBusDedicated(u64 busIndex)
 {
 	int retVal = HvCall1(HvCallCfgIsBusDedicated,busIndex);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
-//====================================================================
+
 static inline HvLpIndex	HvCallCfg_getBusOwner(u64 busIndex)
 {
 	HvLpIndex retVal = HvCall1(HvCallCfgGetBusOwner,busIndex);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
-//====================================================================
-static inline HvLpIndexMap	HvCallCfg_getBusAllocation(u64 busIndex)
+
+static inline HvLpIndexMap HvCallCfg_getBusAllocation(u64 busIndex)
 {
 	HvLpIndexMap retVal = HvCall1(HvCallCfgGetBusAllocation,busIndex);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
-//====================================================================
-static inline HvLpIndexMap	HvCallCfg_getActiveLpMap(void)
+
+static inline HvLpIndexMap HvCallCfg_getActiveLpMap(void)
 {
 	HvLpIndexMap retVal = HvCall0(HvCallCfgGetActiveLpMap);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
-//====================================================================
-static inline HvLpVirtualLanIndexMap	HvCallCfg_getVirtualLanIndexMap(HvLpIndex lp)
+
+static inline HvLpVirtualLanIndexMap HvCallCfg_getVirtualLanIndexMap(
+		HvLpIndex lp)
 {
-	// This is a new function in V5R1 so calls to this on older 
-	// hypervisors will return -1
+	/*
+	 * This is a new function in V5R1 so calls to this on older
+	 * hypervisors will return -1
+	 */
 	u64 retVal = HvCall1(HvCallCfgGetVirtualLanIndexMap, lp);
-	if(retVal == -1)
+	if (retVal == -1)
 		retVal = 0;
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
-//===================================================================
-static inline u64		HvCallCfg_getSystemMsChunks(void)
+
+static inline u64 HvCallCfg_getSystemMsChunks(void)
 {
 	u64 retVal = HvCall0(HvCallCfgGetSystemMsChunks);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
-//===================================================================
-static inline u64		HvCallCfg_getMsChunks(HvLpIndex lp,enum HvCallCfg_ReqQual qual)
+
+static inline u64 HvCallCfg_getMsChunks(HvLpIndex lp,
+		enum HvCallCfg_ReqQual qual)
 {
 	u64 retVal = HvCall2(HvCallCfgGetMsChunks,lp,qual);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
-//===================================================================
-static inline u64		HvCallCfg_getMinRuntimeMsChunks(HvLpIndex lp)
+
+static inline u64 HvCallCfg_getMinRuntimeMsChunks(HvLpIndex lp)
 {
-	// NOTE: This function was added in v5r1 so older hypervisors will return a -1 value
-	u64 retVal = HvCall1(HvCallCfgGetMinRuntimeMsChunks,lp);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	/*
+	 * NOTE: This function was added in v5r1 so older hypervisors
+	 * will return a -1 value
+	 */
+	return HvCall1(HvCallCfgGetMinRuntimeMsChunks, lp);
 }
-//===================================================================
-static inline u64		HvCallCfg_setMinRuntimeMsChunks(u64 chunks)
+
+static inline u64 HvCallCfg_setMinRuntimeMsChunks(u64 chunks)
 {
 	u64 retVal = HvCall1(HvCallCfgSetMinRuntimeMsChunks,chunks);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
-//===================================================================
-static inline u64		HvCallCfg_getSystemPhysicalProcessors(void)
+
+static inline u64 HvCallCfg_getSystemPhysicalProcessors(void)
 {
 	u64 retVal = HvCall0(HvCallCfgGetSystemPhysicalProcessors);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
-//===================================================================
-static inline u64		HvCallCfg_getPhysicalProcessors(HvLpIndex lp,enum HvCallCfg_ReqQual qual)
+
+static inline u64 HvCallCfg_getPhysicalProcessors(HvLpIndex lp,
+		enum HvCallCfg_ReqQual qual)
 {
 	u64 retVal = HvCall2(HvCallCfgGetPhysicalProcessors,lp,qual);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
-//===================================================================
-static inline u64		HvCallCfg_getConfiguredBusUnitsForInterruptProc(HvLpIndex lp,
-										u16 hvLogicalProcIndex)
+
+static inline u64 HvCallCfg_getConfiguredBusUnitsForInterruptProc(HvLpIndex lp,
+		u16 hvLogicalProcIndex)
 {
 	u64 retVal = HvCall2(HvCallCfgGetConfiguredBusUnitsForIntProc,lp,hvLogicalProcIndex);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 
 }
-//==================================================================
-static inline HvLpSharedPoolIndex	HvCallCfg_getSharedPoolIndex(HvLpIndex lp)
+
+static inline HvLpSharedPoolIndex HvCallCfg_getSharedPoolIndex(HvLpIndex lp)
 {
 	HvLpSharedPoolIndex retVal =
 		HvCall1(HvCallCfgGetSharedPoolIndex,lp);
@@ -185,29 +179,29 @@ static inline HvLpSharedPoolIndex	HvCallCfg_getSharedPoolIndex(HvLpIndex lp)
 	return retVal;
 
 }
-//==================================================================
-static inline u64	HvCallCfg_getSharedProcUnits(HvLpIndex lp,enum HvCallCfg_ReqQual qual)
+
+static inline u64 HvCallCfg_getSharedProcUnits(HvLpIndex lp,
+		enum HvCallCfg_ReqQual qual)
 {
 	u64 retVal = HvCall2(HvCallCfgGetSharedProcUnits,lp,qual);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 
 }
-//==================================================================
-static inline u64	HvCallCfg_getNumProcsInSharedPool(HvLpSharedPoolIndex sPI)
+
+static inline u64 HvCallCfg_getNumProcsInSharedPool(HvLpSharedPoolIndex sPI)
 {
 	u16 retVal = HvCall1(HvCallCfgGetNumProcsInSharedPool,sPI);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 
 }
-//==================================================================
+
 static inline HvLpIndex	HvCallCfg_getHostingLpIndex(HvLpIndex lp)
 {
 	u64 retVal = HvCall1(HvCallCfgGetHostingLpIndex,lp);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
-
 }
 
 #endif /* _HVCALLCFG_H */
diff --git a/include/asm-ppc64/iSeries/HvCallEvent.h b/include/asm-ppc64/iSeries/HvCallEvent.h
index 191ddce0c2c..a5954785512 100644
--- a/include/asm-ppc64/iSeries/HvCallEvent.h
+++ b/include/asm-ppc64/iSeries/HvCallEvent.h
@@ -1,32 +1,28 @@
 /*
  * HvCallEvent.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
  */
-
 /*
- *	This file contains the "hypervisor call" interface which is used to
- *	drive the hypervisor from the OS.
+ * This file contains the "hypervisor call" interface which is used to
+ * drive the hypervisor from the OS.
  */
 #ifndef _HVCALLEVENT_H
 #define _HVCALLEVENT_H
 
-/*
- * Standard Includes
- */
 #include <asm/iSeries/HvCallSc.h>
 #include <asm/iSeries/HvTypes.h>
 #include <asm/abs_addr.h>
@@ -71,7 +67,7 @@ typedef u64 HvLpDma_Rc;
 #define HvCallEventCloseLpEventPath			HvCallEvent +  2
 #define HvCallEventDmaBufList				HvCallEvent +  3
 #define HvCallEventDmaSingle				HvCallEvent +  4
-#define HvCallEventDmaToSp				HvCallEvent +  5 
+#define HvCallEventDmaToSp				HvCallEvent +  5
 #define HvCallEventGetOverflowLpEvents			HvCallEvent +  6
 #define HvCallEventGetSourceLpInstanceId		HvCallEvent +  7
 #define HvCallEventGetTargetLpInstanceId		HvCallEvent +  8
@@ -85,13 +81,13 @@ typedef u64 HvLpDma_Rc;
 
 static inline void HvCallEvent_getOverflowLpEvents(u8 queueIndex)
 {
-	HvCall1(HvCallEventGetOverflowLpEvents,queueIndex);
+	HvCall1(HvCallEventGetOverflowLpEvents, queueIndex);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
 
 static inline void HvCallEvent_setInterLpQueueIndex(u8 queueIndex)
 {
-	HvCall1(HvCallEventSetInterLpQueueIndex,queueIndex);
+	HvCall1(HvCallEventSetInterLpQueueIndex, queueIndex);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
 
@@ -138,7 +134,7 @@ static inline HvLpEvent_Rc HvCallEvent_signalLpEventFast(HvLpIndex targetLp,
 {
 	HvLpEvent_Rc retVal;
 
-	// Pack the misc bits into a single Dword to pass to PLIC
+	/* Pack the misc bits into a single Dword to pass to PLIC */
 	union {
 		struct HvCallEvent_PackedParms	parms;
 		u64		dword;
@@ -225,7 +221,7 @@ static inline HvLpDma_Rc HvCallEvent_dmaBufList(HvLpEvent_Type type,
 		u64 localBufList, u64 remoteBufList, u32 transferLength)
 {
 	HvLpDma_Rc retVal;
-	// Pack the misc bits into a single Dword to pass to PLIC
+	/* Pack the misc bits into a single Dword to pass to PLIC */
 	union {
 		struct HvCallEvent_PackedDmaParms	parms;
 		u64		dword;
@@ -257,7 +253,7 @@ static inline HvLpDma_Rc HvCallEvent_dmaSingle(HvLpEvent_Type type,
 		u64 localAddrOrTce, u64 remoteAddrOrTce, u32 transferLength)
 {
 	HvLpDma_Rc retVal;
-	// Pack the misc bits into a single Dword to pass to PLIC
+	/* Pack the misc bits into a single Dword to pass to PLIC */
 	union {
 		struct HvCallEvent_PackedDmaParms	parms;
 		u64		dword;
@@ -280,7 +276,7 @@ static inline HvLpDma_Rc HvCallEvent_dmaSingle(HvLpEvent_Type type,
 	return retVal;
 }
 
-static inline HvLpDma_Rc HvCallEvent_dmaToSp(void* local, u32 remote,
+static inline HvLpDma_Rc HvCallEvent_dmaToSp(void *local, u32 remote,
 		u32 length, HvLpDma_Direction dir)
 {
 	u64 abs_addr;
@@ -293,5 +289,4 @@ static inline HvLpDma_Rc HvCallEvent_dmaToSp(void* local, u32 remote,
 	return retVal;
 }
 
-
 #endif /* _HVCALLEVENT_H */
diff --git a/include/asm-ppc64/iSeries/HvCallHpt.h b/include/asm-ppc64/iSeries/HvCallHpt.h
index da769873d18..827d26c60fc 100644
--- a/include/asm-ppc64/iSeries/HvCallHpt.h
+++ b/include/asm-ppc64/iSeries/HvCallHpt.h
@@ -1,17 +1,17 @@
 /*
  * HvCallHpt.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -19,21 +19,15 @@
 #ifndef _HVCALLHPT_H
 #define _HVCALLHPT_H
 
-//============================================================================
-//
-//	This file contains the "hypervisor call" interface which is used to
-//	drive the hypervisor from the OS.
-//
-//============================================================================
+/*
+ * This file contains the "hypervisor call" interface which is used to
+ * drive the hypervisor from the OS.
+ */
 
 #include <asm/iSeries/HvCallSc.h>
 #include <asm/iSeries/HvTypes.h>
 #include <asm/mmu.h>
 
-//-----------------------------------------------------------------------------
-// Constants
-//-----------------------------------------------------------------------------
-
 #define HvCallHptGetHptAddress		HvCallHpt +  0
 #define HvCallHptGetHptPages		HvCallHpt +  1
 #define HvCallHptSetPp			HvCallHpt +  5
@@ -47,81 +41,76 @@
 #define HvCallHptInvalidateSetSwBitsGet HvCallHpt + 18
 
 
-//============================================================================
-static inline u64		HvCallHpt_getHptAddress(void)
+static inline u64 HvCallHpt_getHptAddress(void)
 {
 	u64 retval = HvCall0(HvCallHptGetHptAddress);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retval;
 }
-//============================================================================
-static inline u64		HvCallHpt_getHptPages(void)
-{	
+
+static inline u64 HvCallHpt_getHptPages(void)
+{
 	u64 retval = HvCall0(HvCallHptGetHptPages);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retval;
 }
-//=============================================================================
-static inline void		HvCallHpt_setPp(u32 hpteIndex, u8 value)
+
+static inline void HvCallHpt_setPp(u32 hpteIndex, u8 value)
 {
-	HvCall2( HvCallHptSetPp, hpteIndex, value );
+	HvCall2(HvCallHptSetPp, hpteIndex, value);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
-//=============================================================================
-static inline void		HvCallHpt_setSwBits(u32 hpteIndex, u8 bitson, u8 bitsoff )
+
+static inline void HvCallHpt_setSwBits(u32 hpteIndex, u8 bitson, u8 bitsoff)
 {
-	HvCall3( HvCallHptSetSwBits, hpteIndex, bitson, bitsoff );
+	HvCall3(HvCallHptSetSwBits, hpteIndex, bitson, bitsoff);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
-//=============================================================================
-static inline void		HvCallHpt_invalidateNoSyncICache(u32 hpteIndex)
-						
+
+static inline void HvCallHpt_invalidateNoSyncICache(u32 hpteIndex)
 {
-	HvCall1( HvCallHptInvalidateNoSyncICache, hpteIndex );
+	HvCall1(HvCallHptInvalidateNoSyncICache, hpteIndex);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
-//=============================================================================
-static inline u64		HvCallHpt_invalidateSetSwBitsGet(u32 hpteIndex, u8 bitson, u8 bitsoff )
-						
+
+static inline u64 HvCallHpt_invalidateSetSwBitsGet(u32 hpteIndex, u8 bitson,
+		u8 bitsoff)
 {
 	u64 compressedStatus;
-	compressedStatus = HvCall4( HvCallHptInvalidateSetSwBitsGet, hpteIndex, bitson, bitsoff, 1 );
-	HvCall1( HvCallHptInvalidateNoSyncICache, hpteIndex );
+
+	compressedStatus = HvCall4(HvCallHptInvalidateSetSwBitsGet,
+			hpteIndex, bitson, bitsoff, 1);
+	HvCall1(HvCallHptInvalidateNoSyncICache, hpteIndex);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return compressedStatus;
 }
-//=============================================================================
-static inline u64		HvCallHpt_findValid( HPTE *hpte, u64 vpn )
+
+static inline u64 HvCallHpt_findValid(HPTE *hpte, u64 vpn)
 {
 	u64 retIndex = HvCall3Ret16( HvCallHptFindValid, hpte, vpn, 0, 0 );
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retIndex;
 }
-//=============================================================================
-static inline u64		HvCallHpt_findNextValid( HPTE *hpte, u32 hpteIndex, u8 bitson, u8 bitsoff )
+
+static inline u64 HvCallHpt_findNextValid(HPTE *hpte, u32 hpteIndex,
+		u8 bitson, u8 bitsoff)
 {
 	u64 retIndex = HvCall3Ret16( HvCallHptFindNextValid, hpte, hpteIndex, bitson, bitsoff );
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retIndex;
 }
-//=============================================================================
-static inline void		HvCallHpt_get( HPTE *hpte, u32 hpteIndex )
+
+static inline void HvCallHpt_get(HPTE *hpte, u32 hpteIndex)
 {
-	HvCall2Ret16( HvCallHptGet, hpte, hpteIndex, 0 );
+	HvCall2Ret16(HvCallHptGet, hpte, hpteIndex, 0);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
-//============================================================================
-static inline void		HvCallHpt_addValidate( u32 hpteIndex,
-						       u32 hBit,
-						       HPTE *hpte )
-						
+
+static inline void HvCallHpt_addValidate(u32 hpteIndex, u32 hBit, HPTE *hpte)
 {
-	HvCall4( HvCallHptAddValidate, hpteIndex,
-		 hBit, (*((u64 *)hpte)), (*(((u64 *)hpte)+1)) );
+	HvCall4(HvCallHptAddValidate, hpteIndex, hBit, (*((u64 *)hpte)),
+			(*(((u64 *)hpte)+1)));
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
 
-
-//=============================================================================
-
 #endif /* _HVCALLHPT_H */
diff --git a/include/asm-ppc64/iSeries/HvCallPci.h b/include/asm-ppc64/iSeries/HvCallPci.h
index 6887b619288..346ab5d39dc 100644
--- a/include/asm-ppc64/iSeries/HvCallPci.h
+++ b/include/asm-ppc64/iSeries/HvCallPci.h
@@ -1,26 +1,26 @@
-/************************************************************************/
-/* Provides the Hypervisor PCI calls for iSeries Linux Parition.        */
-/* Copyright (C) 2001  <Wayne G Holm> <IBM Corporation>                 */
-/*                                                                      */
-/* This program is free software; you can redistribute it and/or modify */
-/* it under the terms of the GNU General Public License as published by */
-/* the Free Software Foundation; either version 2 of the License, or    */
-/* (at your option) any later version.                                  */
-/*                                                                      */
-/* This program is distributed in the hope that it will be useful,      */ 
-/* but WITHOUT ANY WARRANTY; without even the implied warranty of       */
-/* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the        */
-/* GNU General Public License for more details.                         */
-/*                                                                      */
-/* You should have received a copy of the GNU General Public License    */ 
-/* along with this program; if not, write to the:                       */
-/* Free Software Foundation, Inc.,                                      */ 
-/* 59 Temple Place, Suite 330,                                          */ 
-/* Boston, MA  02111-1307  USA                                          */
-/************************************************************************/
-/* Change Activity:                                                     */
-/*   Created, Jan 9, 2001                                               */
-/************************************************************************/
+/*
+ * Provides the Hypervisor PCI calls for iSeries Linux Parition.
+ * Copyright (C) 2001  <Wayne G Holm> <IBM Corporation>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the:
+ * Free Software Foundation, Inc.,
+ * 59 Temple Place, Suite 330,
+ * Boston, MA  02111-1307  USA
+ *
+ * Change Activity:
+ *   Created, Jan 9, 2001
+ */
 
 #ifndef _HVCALLPCI_H
 #define _HVCALLPCI_H
@@ -34,8 +34,8 @@
  */
 struct HvCallPci_DsaAddr {
 	u16		busNumber;		/* PHB index? */
-	u8		subBusNumber; 		/* PCI bus number? */
-	u8		deviceId;     		/* device and function? */
+	u8		subBusNumber;		/* PCI bus number? */
+	u8		deviceId;		/* device and function? */
 	u8		barNumber;
 	u8		reserved[3];
 };
@@ -52,34 +52,37 @@ struct HvCallPci_LoadReturn {
 
 enum HvCallPci_DeviceType {
 	HvCallPci_NodeDevice	= 1,
-	HvCallPci_SpDevice	= 2,	
-	HvCallPci_IopDevice     = 3,	
-	HvCallPci_BridgeDevice	= 4,	
-	HvCallPci_MultiFunctionDevice = 5,	
-	HvCallPci_IoaDevice	= 6	
+	HvCallPci_SpDevice	= 2,
+	HvCallPci_IopDevice     = 3,
+	HvCallPci_BridgeDevice	= 4,
+	HvCallPci_MultiFunctionDevice = 5,
+	HvCallPci_IoaDevice	= 6
 };
 
 
 struct HvCallPci_DeviceInfo {
-	u32	deviceType;		// See DeviceType enum for values
+	u32	deviceType;		/* See DeviceType enum for values */
 };
-    
+
 struct HvCallPci_BusUnitInfo {
-	u32	sizeReturned;		// length of data returned
-	u32	deviceType;		// see DeviceType enum for values
+	u32	sizeReturned;		/* length of data returned */
+	u32	deviceType;		/* see DeviceType enum for values */
 };
 
 struct HvCallPci_BridgeInfo {
-	struct HvCallPci_BusUnitInfo busUnitInfo;  // Generic bus unit info
-	u8		subBusNumber;		// Bus number of secondary bus
-	u8		maxAgents;		// Max idsels on secondary bus
-        u8              maxSubBusNumber;        // Max Sub Bus
-	u8		logicalSlotNumber;	// Logical Slot Number for IOA 
+	struct HvCallPci_BusUnitInfo busUnitInfo;  /* Generic bus unit info */
+	u8		subBusNumber;	/* Bus number of secondary bus */
+	u8		maxAgents;	/* Max idsels on secondary bus */
+        u8              maxSubBusNumber; /* Max Sub Bus */
+	u8		logicalSlotNumber; /* Logical Slot Number for IOA */
 };
-    
 
-//  Maximum BusUnitInfo buffer size.  Provided for clients so they can allocate
-//  a buffer big enough for any type of bus unit.  Increase as needed.
+
+/*
+ * Maximum BusUnitInfo buffer size.  Provided for clients so
+ * they can allocate a buffer big enough for any type of bus
+ * unit.  Increase as needed.
+ */
 enum {HvCallPci_MaxBusUnitInfoSize = 128};
 
 struct HvCallPci_BarParms {
@@ -89,12 +92,12 @@ struct HvCallPci_BarParms {
 	u64		protectStart;
 	u64		protectEnd;
 	u64		relocationOffset;
-	u64		pciAddress;		
+	u64		pciAddress;
 	u64		reserved[3];
-};					
+};
 
 enum HvCallPci_VpdType {
-	HvCallPci_BusVpd		= 1,
+	HvCallPci_BusVpd	= 1,
 	HvCallPci_BusAdapterVpd	= 2
 };
 
@@ -123,15 +126,13 @@ enum HvCallPci_VpdType {
 #define HvCallPciUnmaskInterrupts	HvCallPci + 49
 #define HvCallPciGetBusUnitInfo		HvCallPci + 50
 
-//============================================================================
 static inline u64 HvCallPci_configLoad8(u16 busNumber, u8 subBusNumber,
-					u8 deviceId, u32 offset,
-					u8 *value)
+		u8 deviceId, u32 offset, u8 *value)
 {
 	struct HvCallPci_DsaAddr dsa;
 	struct HvCallPci_LoadReturn retVal;
 
-	*((u64*)&dsa) = 0;				
+	*((u64*)&dsa) = 0;
 
 	dsa.busNumber = busNumber;
 	dsa.subBusNumber = subBusNumber;
@@ -145,15 +146,14 @@ static inline u64 HvCallPci_configLoad8(u16 busNumber, u8 subBusNumber,
 
 	return retVal.rc;
 }
-//============================================================================
+
 static inline u64 HvCallPci_configLoad16(u16 busNumber, u8 subBusNumber,
-					 u8 deviceId, u32 offset,
-					 u16 *value)
+		u8 deviceId, u32 offset, u16 *value)
 {
 	struct HvCallPci_DsaAddr dsa;
 	struct HvCallPci_LoadReturn retVal;
 
-	*((u64*)&dsa) = 0;				
+	*((u64*)&dsa) = 0;
 
 	dsa.busNumber = busNumber;
 	dsa.subBusNumber = subBusNumber;
@@ -167,15 +167,14 @@ static inline u64 HvCallPci_configLoad16(u16 busNumber, u8 subBusNumber,
 
 	return retVal.rc;
 }
-//============================================================================
-static inline u64	HvCallPci_configLoad32(u16 busNumber, u8 subBusNumber,
-					      u8 deviceId, u32 offset,
-					      u32 *value)
+
+static inline u64 HvCallPci_configLoad32(u16 busNumber, u8 subBusNumber,
+		u8 deviceId, u32 offset, u32 *value)
 {
 	struct HvCallPci_DsaAddr dsa;
 	struct HvCallPci_LoadReturn retVal;
 
-	*((u64*)&dsa) = 0;				
+	*((u64*)&dsa) = 0;
 
 	dsa.busNumber = busNumber;
 	dsa.subBusNumber = subBusNumber;
@@ -189,15 +188,14 @@ static inline u64	HvCallPci_configLoad32(u16 busNumber, u8 subBusNumber,
 
 	return retVal.rc;
 }
-//============================================================================
-static inline u64	HvCallPci_configStore8(u16 busNumber, u8 subBusNumber,
-					      u8 deviceId, u32 offset,
-					      u8  value)
+
+static inline u64 HvCallPci_configStore8(u16 busNumber, u8 subBusNumber,
+		u8 deviceId, u32 offset, u8 value)
 {
 	struct HvCallPci_DsaAddr dsa;
 	u64 retVal;
 
-	*((u64*)&dsa) = 0;				
+	*((u64*)&dsa) = 0;
 
 	dsa.busNumber = busNumber;
 	dsa.subBusNumber = subBusNumber;
@@ -209,15 +207,14 @@ static inline u64	HvCallPci_configStore8(u16 busNumber, u8 subBusNumber,
 
 	return retVal;
 }
-//============================================================================
-static inline u64	HvCallPci_configStore16(u16 busNumber, u8 subBusNumber,
-					      u8 deviceId, u32 offset,
-					      u16  value)
+
+static inline u64 HvCallPci_configStore16(u16 busNumber, u8 subBusNumber,
+		u8 deviceId, u32 offset, u16 value)
 {
 	struct HvCallPci_DsaAddr dsa;
 	u64 retVal;
 
-	*((u64*)&dsa) = 0;				
+	*((u64*)&dsa) = 0;
 
 	dsa.busNumber = busNumber;
 	dsa.subBusNumber = subBusNumber;
@@ -229,15 +226,14 @@ static inline u64	HvCallPci_configStore16(u16 busNumber, u8 subBusNumber,
 
 	return retVal;
 }
-//============================================================================
-static inline u64	HvCallPci_configStore32(u16 busNumber, u8 subBusNumber,
-					      u8 deviceId, u32 offset,
-					      u32  value)
+
+static inline u64 HvCallPci_configStore32(u16 busNumber, u8 subBusNumber,
+		u8 deviceId, u32 offset, u32 value)
 {
 	struct HvCallPci_DsaAddr dsa;
 	u64 retVal;
 
-	*((u64*)&dsa) = 0;				
+	*((u64*)&dsa) = 0;
 
 	dsa.busNumber = busNumber;
 	dsa.subBusNumber = subBusNumber;
@@ -249,18 +245,15 @@ static inline u64	HvCallPci_configStore32(u16 busNumber, u8 subBusNumber,
 
 	return retVal;
 }
-//============================================================================
-static inline u64	HvCallPci_barLoad8(u16	busNumberParm,
-					   u8		subBusParm,
-					   u8		deviceIdParm,
-					   u8		barNumberParm,
-					   u64		offsetParm,
-					   u8*		valueParm)
+
+static inline u64 HvCallPci_barLoad8(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm, u8 barNumberParm, u64 offsetParm,
+		u8 *valueParm)
 {
 	struct HvCallPci_DsaAddr dsa;
 	struct HvCallPci_LoadReturn retVal;
 
-	*((u64*)&dsa) = 0;				
+	*((u64*)&dsa) = 0;
 
 	dsa.busNumber = busNumberParm;
 	dsa.subBusNumber = subBusParm;
@@ -275,18 +268,15 @@ static inline u64	HvCallPci_barLoad8(u16	busNumberParm,
 
 	return retVal.rc;
 }
-//============================================================================
-static inline u64	HvCallPci_barLoad16(u16	busNumberParm,
-					   u8		subBusParm,
-					   u8		deviceIdParm,
-					   u8		barNumberParm,
-					   u64		offsetParm,
-					   u16*		valueParm)
+
+static inline u64 HvCallPci_barLoad16(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm, u8 barNumberParm, u64 offsetParm,
+		u16 *valueParm)
 {
 	struct HvCallPci_DsaAddr dsa;
 	struct HvCallPci_LoadReturn retVal;
 
-	*((u64*)&dsa) = 0;				
+	*((u64*)&dsa) = 0;
 
 	dsa.busNumber = busNumberParm;
 	dsa.subBusNumber = subBusParm;
@@ -301,18 +291,15 @@ static inline u64	HvCallPci_barLoad16(u16	busNumberParm,
 
 	return retVal.rc;
 }
-//============================================================================
-static inline u64	HvCallPci_barLoad32(u16	busNumberParm,
-					   u8		subBusParm,
-					   u8		deviceIdParm,
-					   u8		barNumberParm,
-					   u64		offsetParm,
-					   u32*		valueParm)
+
+static inline u64 HvCallPci_barLoad32(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm, u8 barNumberParm, u64 offsetParm,
+		u32 *valueParm)
 {
 	struct HvCallPci_DsaAddr dsa;
 	struct HvCallPci_LoadReturn retVal;
 
-	*((u64*)&dsa) = 0;				
+	*((u64*)&dsa) = 0;
 
 	dsa.busNumber = busNumberParm;
 	dsa.subBusNumber = subBusParm;
@@ -327,18 +314,15 @@ static inline u64	HvCallPci_barLoad32(u16	busNumberParm,
 
 	return retVal.rc;
 }
-//============================================================================
-static inline u64	HvCallPci_barLoad64(u16	busNumberParm,
-					   u8		subBusParm,
-					   u8		deviceIdParm,
-					   u8		barNumberParm,
-					   u64		offsetParm,
-					   u64*		valueParm)
+
+static inline u64 HvCallPci_barLoad64(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm, u8 barNumberParm, u64 offsetParm,
+		u64 *valueParm)
 {
 	struct HvCallPci_DsaAddr dsa;
 	struct HvCallPci_LoadReturn retVal;
 
-	*((u64*)&dsa) = 0;				
+	*((u64*)&dsa) = 0;
 
 	dsa.busNumber = busNumberParm;
 	dsa.subBusNumber = subBusParm;
@@ -353,19 +337,16 @@ static inline u64	HvCallPci_barLoad64(u16	busNumberParm,
 
 	return retVal.rc;
 }
-//============================================================================
-static inline u64	HvCallPci_barStore8(u16	busNumberParm,
-					    u8		subBusParm,
-					    u8		deviceIdParm,
-					    u8		barNumberParm,
-					    u64		offsetParm,
-					    u8		valueParm)
+
+static inline u64 HvCallPci_barStore8(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm, u8 barNumberParm, u64 offsetParm,
+		u8 valueParm)
 {
 	struct HvCallPci_DsaAddr dsa;
 	u64 retVal;
 
 	*((u64*)&dsa) = 0;
-				
+
 	dsa.busNumber = busNumberParm;
 	dsa.subBusNumber = subBusParm;
 	dsa.deviceId = deviceIdParm;
@@ -377,19 +358,16 @@ static inline u64	HvCallPci_barStore8(u16	busNumberParm,
 
 	return retVal;
 }
-//============================================================================
-static inline u64	HvCallPci_barStore16(u16	busNumberParm,
-					     u8		subBusParm,
-					     u8		deviceIdParm,
-					     u8		barNumberParm,
-					     u64	offsetParm,
-					     u16	valueParm)
+
+static inline u64 HvCallPci_barStore16(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm, u8 barNumberParm, u64 offsetParm,
+		u16 valueParm)
 {
 	struct HvCallPci_DsaAddr dsa;
 	u64 retVal;
 
 	*((u64*)&dsa) = 0;
-				
+
 	dsa.busNumber = busNumberParm;
 	dsa.subBusNumber = subBusParm;
 	dsa.deviceId = deviceIdParm;
@@ -401,19 +379,16 @@ static inline u64	HvCallPci_barStore16(u16	busNumberParm,
 
 	return retVal;
 }
-//============================================================================
-static inline u64	HvCallPci_barStore32(u16	busNumberParm,
-					     u8		subBusParm,
-					     u8		deviceIdParm,
-					     u8		barNumberParm,
-					     u64	offsetParm,
-					     u32	valueParm)
+
+static inline u64 HvCallPci_barStore32(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm, u8 barNumberParm, u64 offsetParm,
+		u32 valueParm)
 {
 	struct HvCallPci_DsaAddr dsa;
 	u64 retVal;
 
 	*((u64*)&dsa) = 0;
-				
+
 	dsa.busNumber = busNumberParm;
 	dsa.subBusNumber = subBusParm;
 	dsa.deviceId = deviceIdParm;
@@ -425,19 +400,16 @@ static inline u64	HvCallPci_barStore32(u16	busNumberParm,
 
 	return retVal;
 }
-//============================================================================
-static inline u64	HvCallPci_barStore64(u16	busNumberParm,
-					     u8		subBusParm,
-					     u8		deviceIdParm,
-					     u8		barNumberParm,
-					     u64	offsetParm,
-					     u64	valueParm)
+
+static inline u64 HvCallPci_barStore64(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm, u8 barNumberParm, u64 offsetParm,
+		u64 valueParm)
 {
 	struct HvCallPci_DsaAddr dsa;
 	u64 retVal;
 
 	*((u64*)&dsa) = 0;
-				
+
 	dsa.busNumber = busNumberParm;
 	dsa.subBusNumber = subBusParm;
 	dsa.deviceId = deviceIdParm;
@@ -449,10 +421,9 @@ static inline u64	HvCallPci_barStore64(u16	busNumberParm,
 
 	return retVal;
 }
-//============================================================================
-static inline u64	HvCallPci_eoi(u16	busNumberParm,
-				      u8	subBusParm,  
-				      u8	deviceIdParm)
+
+static inline u64 HvCallPci_eoi(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm)
 {
 	struct HvCallPci_DsaAddr dsa;
 	struct HvCallPci_LoadReturn retVal;
@@ -469,13 +440,9 @@ static inline u64	HvCallPci_eoi(u16	busNumberParm,
 
 	return retVal.rc;
 }
-//============================================================================
-static inline u64	HvCallPci_getBarParms(u16	busNumberParm,
-					      u8	subBusParm,  
-					      u8	deviceIdParm,
-					      u8	barNumberParm,
-					      u64	parms,
-					      u32	sizeofParms)
+
+static inline u64 HvCallPci_getBarParms(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm, u8 barNumberParm, u64 parms, u32 sizeofParms)
 {
 	struct HvCallPci_DsaAddr dsa;
 	u64 retVal;
@@ -493,16 +460,14 @@ static inline u64	HvCallPci_getBarParms(u16	busNumberParm,
 
 	return retVal;
 }
-//============================================================================
-static inline u64	HvCallPci_maskFisr(u16	busNumberParm,
-					   u8	subBusParm,  
-					   u8	deviceIdParm,
-					   u64	fisrMask)
+
+static inline u64 HvCallPci_maskFisr(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm, u64 fisrMask)
 {
 	struct HvCallPci_DsaAddr dsa;
 	u64 retVal;
 
-	*((u64*)&dsa) = 0;		
+	*((u64*)&dsa) = 0;
 
 	dsa.busNumber = busNumberParm;
 	dsa.subBusNumber = subBusParm;
@@ -514,16 +479,14 @@ static inline u64	HvCallPci_maskFisr(u16	busNumberParm,
 
 	return retVal;
 }
-//============================================================================
-static inline u64	HvCallPci_unmaskFisr(u16	busNumberParm,
-					     u8		subBusParm,  
-					     u8		deviceIdParm,
-					     u64	fisrMask)
+
+static inline u64 HvCallPci_unmaskFisr(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm, u64 fisrMask)
 {
 	struct HvCallPci_DsaAddr dsa;
 	u64 retVal;
 
-	*((u64*)&dsa) = 0;		
+	*((u64*)&dsa) = 0;
 
 	dsa.busNumber = busNumberParm;
 	dsa.subBusNumber = subBusParm;
@@ -535,11 +498,9 @@ static inline u64	HvCallPci_unmaskFisr(u16	busNumberParm,
 
 	return retVal;
 }
-//============================================================================
-static inline u64	HvCallPci_setSlotReset(u16		busNumberParm,
-					       u8		subBusParm,
-					       u8		deviceIdParm,
-					       u64		onNotOff)
+
+static inline u64 HvCallPci_setSlotReset(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm, u64 onNotOff)
 {
 	struct HvCallPci_DsaAddr dsa;
 	u64 retVal;
@@ -556,12 +517,9 @@ static inline u64	HvCallPci_setSlotReset(u16		busNumberParm,
 
 	return retVal;
 }
-//============================================================================
-static inline u64	HvCallPci_getDeviceInfo(u16	busNumberParm,
-						u8	subBusParm,  
-						u8	deviceNumberParm,
-						u64     parms,
-						u32	sizeofParms)
+
+static inline u64 HvCallPci_getDeviceInfo(u16 busNumberParm, u8 subBusParm,
+		u8 deviceNumberParm, u64 parms, u32 sizeofParms)
 {
 	struct HvCallPci_DsaAddr dsa;
 	u64 retVal;
@@ -578,16 +536,14 @@ static inline u64	HvCallPci_getDeviceInfo(u16	busNumberParm,
 
 	return retVal;
 }
-//============================================================================
-static inline u64	HvCallPci_maskInterrupts(u16	busNumberParm,
-						 u8	subBusParm,  
-						 u8	deviceIdParm,
-						 u64	interruptMask)
+
+static inline u64 HvCallPci_maskInterrupts(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm, u64 interruptMask)
 {
 	struct HvCallPci_DsaAddr dsa;
 	u64 retVal;
 
-	*((u64*)&dsa) = 0;		
+	*((u64*)&dsa) = 0;
 
 	dsa.busNumber = busNumberParm;
 	dsa.subBusNumber = subBusParm;
@@ -599,16 +555,14 @@ static inline u64	HvCallPci_maskInterrupts(u16	busNumberParm,
 
 	return retVal;
 }
-//============================================================================
-static inline u64	HvCallPci_unmaskInterrupts(u16	busNumberParm,
-						 u8		subBusParm,  
-						 u8		deviceIdParm,
-						 u64		interruptMask)
+
+static inline u64 HvCallPci_unmaskInterrupts(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm, u64 interruptMask)
 {
 	struct HvCallPci_DsaAddr dsa;
 	u64 retVal;
 
-	*((u64*)&dsa) = 0;		
+	*((u64*)&dsa) = 0;
 
 	dsa.busNumber = busNumberParm;
 	dsa.subBusNumber = subBusParm;
@@ -620,18 +574,14 @@ static inline u64	HvCallPci_unmaskInterrupts(u16	busNumberParm,
 
 	return retVal;
 }
-//============================================================================
 
-static inline u64	HvCallPci_getBusUnitInfo(u16		busNumberParm,
-						 u8		subBusParm,  
-						 u8		deviceIdParm,
-						 u64            parms,
-						 u32		sizeofParms)
+static inline u64 HvCallPci_getBusUnitInfo(u16 busNumberParm, u8 subBusParm,
+		u8 deviceIdParm, u64 parms, u32 sizeofParms)
 {
 	struct HvCallPci_DsaAddr dsa;
 	u64 retVal;
 
-	*((u64*)&dsa) = 0;		
+	*((u64*)&dsa) = 0;
 
 	dsa.busNumber = busNumberParm;
 	dsa.subBusNumber = subBusParm;
@@ -643,9 +593,9 @@ static inline u64	HvCallPci_getBusUnitInfo(u16		busNumberParm,
 
 	return retVal;
 }
-//============================================================================
 
-static inline int HvCallPci_getBusVpd(u16 busNumParm, u64 destParm, u16 sizeParm)
+static inline int HvCallPci_getBusVpd(u16 busNumParm, u64 destParm,
+		u16 sizeParm)
 {
 	int xRetSize;
 	u64 xRc = HvCall4(HvCallPciGetCardVpd, busNumParm, destParm, sizeParm, HvCallPci_BusVpd);
@@ -656,9 +606,9 @@ static inline int HvCallPci_getBusVpd(u16 busNumParm, u64 destParm, u16 sizeParm
 		xRetSize = xRc & 0xFFFF;
 	return xRetSize;
 }
-//============================================================================
 
-static inline int HvCallPci_getBusAdapterVpd(u16 busNumParm, u64 destParm, u16 sizeParm)
+static inline int HvCallPci_getBusAdapterVpd(u16 busNumParm, u64 destParm,
+		u16 sizeParm)
 {
 	int xRetSize;
 	u64 xRc = HvCall4(HvCallPciGetCardVpd, busNumParm, destParm, sizeParm, HvCallPci_BusAdapterVpd);
@@ -669,5 +619,5 @@ static inline int HvCallPci_getBusAdapterVpd(u16 busNumParm, u64 destParm, u16 s
 		xRetSize = xRc & 0xFFFF;
 	return xRetSize;
 }
-//============================================================================
+
 #endif /* _HVCALLPCI_H */
diff --git a/include/asm-ppc64/iSeries/HvCallSc.h b/include/asm-ppc64/iSeries/HvCallSc.h
index eea25844764..c32145f9ec7 100644
--- a/include/asm-ppc64/iSeries/HvCallSc.h
+++ b/include/asm-ppc64/iSeries/HvCallSc.h
@@ -1,17 +1,17 @@
 /*
  * HvCallSc.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -30,22 +30,22 @@
 #define HvCallSm		0x8007000000000000ul
 #define HvCallXm		0x8009000000000000ul
 
-u64 HvCall0( u64 );
-u64 HvCall1( u64, u64 );
-u64 HvCall2( u64, u64, u64 );
-u64 HvCall3( u64, u64, u64, u64 );
-u64 HvCall4( u64, u64, u64, u64, u64 );
-u64 HvCall5( u64, u64, u64, u64, u64, u64 );
-u64 HvCall6( u64, u64, u64, u64, u64, u64, u64 );
-u64 HvCall7( u64, u64, u64, u64, u64, u64, u64, u64 );
+u64 HvCall0(u64);
+u64 HvCall1(u64, u64);
+u64 HvCall2(u64, u64, u64);
+u64 HvCall3(u64, u64, u64, u64);
+u64 HvCall4(u64, u64, u64, u64, u64);
+u64 HvCall5(u64, u64, u64, u64, u64, u64);
+u64 HvCall6(u64, u64, u64, u64, u64, u64, u64);
+u64 HvCall7(u64, u64, u64, u64, u64, u64, u64, u64);
 
-u64 HvCall0Ret16( u64, void * );
-u64 HvCall1Ret16( u64, void *, u64 );
-u64 HvCall2Ret16( u64, void *, u64, u64 );
-u64 HvCall3Ret16( u64, void *, u64, u64, u64 );
-u64 HvCall4Ret16( u64, void *, u64, u64, u64, u64 );
-u64 HvCall5Ret16( u64, void *, u64, u64, u64, u64, u64 );
-u64 HvCall6Ret16( u64, void *, u64, u64, u64, u64, u64, u64 );
-u64 HvCall7Ret16( u64, void *, u64, u64 ,u64 ,u64 ,u64 ,u64 ,u64 );
+u64 HvCall0Ret16(u64, void *);
+u64 HvCall1Ret16(u64, void *, u64);
+u64 HvCall2Ret16(u64, void *, u64, u64);
+u64 HvCall3Ret16(u64, void *, u64, u64, u64);
+u64 HvCall4Ret16(u64, void *, u64, u64, u64, u64);
+u64 HvCall5Ret16(u64, void *, u64, u64, u64, u64, u64);
+u64 HvCall6Ret16(u64, void *, u64, u64, u64, u64, u64, u64);
+u64 HvCall7Ret16(u64, void *, u64, u64 ,u64 ,u64 ,u64 ,u64 ,u64);
 
 #endif /* _HVCALLSC_H */
diff --git a/include/asm-ppc64/iSeries/HvCallSm.h b/include/asm-ppc64/iSeries/HvCallSm.h
index 9050c94a529..ef6b303e4d2 100644
--- a/include/asm-ppc64/iSeries/HvCallSm.h
+++ b/include/asm-ppc64/iSeries/HvCallSm.h
@@ -1,17 +1,17 @@
 /*
  * HvCallSm.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -19,34 +19,23 @@
 #ifndef _HVCALLSM_H
 #define _HVCALLSM_H
 
-//============================================================================
-//
-//	This file contains the "hypervisor call" interface which is used to
-//	drive the hypervisor from the OS.
-//
-//============================================================================
+/*
+ * This file contains the "hypervisor call" interface which is used to
+ * drive the hypervisor from the OS.
+ */
 
-//-------------------------------------------------------------------
-// Standard Includes
-//-------------------------------------------------------------------
 #include <asm/iSeries/HvCallSc.h>
 #include <asm/iSeries/HvTypes.h>
 
-//-----------------------------------------------------------------------------
-// Constants
-//-----------------------------------------------------------------------------
-
 #define HvCallSmGet64BitsOfAccessMap	HvCallSm  + 11
 
-
-//============================================================================
-static inline u64		HvCallSm_get64BitsOfAccessMap(
-					HvLpIndex lpIndex, u64 indexIntoBitMap )
+static inline u64 HvCallSm_get64BitsOfAccessMap(HvLpIndex lpIndex,
+		u64 indexIntoBitMap)
 {
 	u64 retval = HvCall2(HvCallSmGet64BitsOfAccessMap, lpIndex,
 			     indexIntoBitMap );
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retval;
 }
-//============================================================================
+
 #endif /* _HVCALLSM_H */
diff --git a/include/asm-ppc64/iSeries/HvCallXm.h b/include/asm-ppc64/iSeries/HvCallXm.h
index bfb898f4db8..a99f29b9361 100644
--- a/include/asm-ppc64/iSeries/HvCallXm.h
+++ b/include/asm-ppc64/iSeries/HvCallXm.h
@@ -1,30 +1,13 @@
-//============================================================================
-//							 Header File Id
-// Name______________:	HvCallXm.H
-//
-// Description_______:
-//
-//	This file contains the "hypervisor call" interface which is used to
-//	drive the hypervisor from SLIC.
-//
-//============================================================================
+/*
+ * This file contains the "hypervisor call" interface which is used to
+ * drive the hypervisor from SLIC.
+ */
 #ifndef _HVCALLXM_H
 #define _HVCALLXM_H
 
-//-------------------------------------------------------------------
-// Forward declarations 
-//-------------------------------------------------------------------
-
-//-------------------------------------------------------------------
-// Standard Includes
-//-------------------------------------------------------------------
 #include <asm/iSeries/HvCallSc.h>
 #include <asm/iSeries/HvTypes.h>
 
-//-----------------------------------------------------------------------------
-// Constants
-//-----------------------------------------------------------------------------
-
 #define HvCallXmGetTceTableParms	HvCallXm +  0
 #define HvCallXmTestBus			HvCallXm +  1
 #define HvCallXmConnectBusUnit		HvCallXm +  2
@@ -33,47 +16,46 @@
 #define HvCallXmSetTce			HvCallXm + 11
 #define HvCallXmSetTces			HvCallXm + 13
 
-
-
-//============================================================================
-static inline void		HvCallXm_getTceTableParms(u64 cb)
+static inline void HvCallXm_getTceTableParms(u64 cb)
 {
 	HvCall1(HvCallXmGetTceTableParms, cb);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
-//============================================================================
-static inline u64		HvCallXm_setTce(u64 tceTableToken, u64 tceOffset, u64 tce)
-{	
+
+static inline u64 HvCallXm_setTce(u64 tceTableToken, u64 tceOffset, u64 tce)
+{
 	u64 retval = HvCall3(HvCallXmSetTce, tceTableToken, tceOffset, tce );
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retval;
 }
-//============================================================================
-static inline u64		HvCallXm_setTces(u64 tceTableToken, u64 tceOffset, u64 numTces, u64 tce1, u64 tce2, u64 tce3, u64 tce4)
-{	
+
+static inline u64 HvCallXm_setTces(u64 tceTableToken, u64 tceOffset,
+		u64 numTces, u64 tce1, u64 tce2, u64 tce3, u64 tce4)
+{
 	u64 retval = HvCall7(HvCallXmSetTces, tceTableToken, tceOffset, numTces,
 			     tce1, tce2, tce3, tce4 );
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retval;
 }
-//=============================================================================
-static inline u64	HvCallXm_testBus(u16 busNumber)
+
+static inline u64 HvCallXm_testBus(u16 busNumber)
 {
 	u64 retVal = HvCall1(HvCallXmTestBus, busNumber);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
-//=====================================================================================
-static inline u64	HvCallXm_testBusUnit(u16 busNumber, u8 subBusNumber, u8 deviceId)
+
+static inline u64 HvCallXm_testBusUnit(u16 busNumber, u8 subBusNumber,
+		u8 deviceId)
 {
 	u64 busUnitNumber = (subBusNumber << 8) | deviceId;
 	u64 retVal = HvCall2(HvCallXmTestBusUnit, busNumber, busUnitNumber);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
-//=====================================================================================
-static inline u64	HvCallXm_connectBusUnit(u16 busNumber, u8 subBusNumber, u8 deviceId,
-						u64 interruptToken)
+
+static inline u64 HvCallXm_connectBusUnit(u16 busNumber, u8 subBusNumber,
+		u8 deviceId, u64 interruptToken)
 {
 	u64 busUnitNumber = (subBusNumber << 8) | deviceId;
 	u64 queueIndex = 0; // HvLpConfig::mapDsaToQueueIndex(HvLpDSA(busNumber, xBoard, xCard));  
@@ -83,13 +65,12 @@ static inline u64	HvCallXm_connectBusUnit(u16 busNumber, u8 subBusNumber, u8 dev
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
-//=====================================================================================
-static inline u64	HvCallXm_loadTod(void)
+
+static inline u64 HvCallXm_loadTod(void)
 {
 	u64 retVal = HvCall0(HvCallXmLoadTod);
 	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
-//=====================================================================================
 
 #endif /* _HVCALLXM_H */
diff --git a/include/asm-ppc64/iSeries/HvLpConfig.h b/include/asm-ppc64/iSeries/HvLpConfig.h
index bdbd70f42c9..47311cc952d 100644
--- a/include/asm-ppc64/iSeries/HvLpConfig.h
+++ b/include/asm-ppc64/iSeries/HvLpConfig.h
@@ -1,17 +1,17 @@
 /*
  * HvLpConfig.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -19,262 +19,285 @@
 #ifndef _HVLPCONFIG_H
 #define _HVLPCONFIG_H
 
-//===========================================================================
-//
-//      This file contains the interface to the LPAR configuration data
-//  to determine which resources should be allocated to each partition.
-//
-//===========================================================================
+/*
+ * This file contains the interface to the LPAR configuration data
+ * to determine which resources should be allocated to each partition.
+ */
 
 #include <asm/iSeries/HvCallCfg.h>
 #include <asm/iSeries/HvTypes.h>
 #include <asm/iSeries/ItLpNaca.h>
 #include <asm/iSeries/LparData.h>
 
-//-------------------------------------------------------------------
-// Constants
-//-------------------------------------------------------------------
-
 extern HvLpIndex HvLpConfig_getLpIndex_outline(void);
 
-//===================================================================
 static inline HvLpIndex	HvLpConfig_getLpIndex(void)
 {
 	return itLpNaca.xLpIndex;
 }
-//===================================================================
+
 static inline HvLpIndex	HvLpConfig_getPrimaryLpIndex(void)
 {
 	return itLpNaca.xPrimaryLpIndex;
 }
-//=================================================================
+
 static inline HvLpIndex	HvLpConfig_getLps(void)
 {
 	return HvCallCfg_getLps();
 }
-//=================================================================
-static inline HvLpIndexMap	HvLpConfig_getActiveLpMap(void)
+
+static inline HvLpIndexMap HvLpConfig_getActiveLpMap(void)
 {
 	return HvCallCfg_getActiveLpMap();
 }
-//=================================================================
-static inline u64		HvLpConfig_getSystemMsMegs(void)
+
+static inline u64 HvLpConfig_getSystemMsMegs(void)
 {
 	return HvCallCfg_getSystemMsChunks() / HVCHUNKSPERMEG;
 }
-//=================================================================
-static inline u64		HvLpConfig_getSystemMsChunks(void)
+
+static inline u64 HvLpConfig_getSystemMsChunks(void)
 {
 	return HvCallCfg_getSystemMsChunks();
 }
-//=================================================================
-static inline u64		HvLpConfig_getSystemMsPages(void)
+
+static inline u64 HvLpConfig_getSystemMsPages(void)
 {
 	return HvCallCfg_getSystemMsChunks() * HVPAGESPERCHUNK;
 }
-//================================================================
-static inline u64		HvLpConfig_getMsMegs(void)
+
+static inline u64 HvLpConfig_getMsMegs(void)
 {
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(),HvCallCfg_Cur) / HVCHUNKSPERMEG;
+	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Cur)
+		/ HVCHUNKSPERMEG;
 }
-//================================================================
-static inline u64		HvLpConfig_getMsChunks(void)
+
+static inline u64 HvLpConfig_getMsChunks(void)
 {
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(),HvCallCfg_Cur);
+	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Cur);
 }
-//================================================================
-static inline u64		HvLpConfig_getMsPages(void)
+
+static inline u64 HvLpConfig_getMsPages(void)
 {
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(),HvCallCfg_Cur) * HVPAGESPERCHUNK;
+	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Cur)
+		* HVPAGESPERCHUNK;
 }
-//================================================================
-static inline u64		HvLpConfig_getMinMsMegs(void)
+
+static inline u64 HvLpConfig_getMinMsMegs(void)
 {
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(),HvCallCfg_Min) / HVCHUNKSPERMEG;
+	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Min)
+		/ HVCHUNKSPERMEG;
 }
-//================================================================
-static inline u64		HvLpConfig_getMinMsChunks(void)
+
+static inline u64 HvLpConfig_getMinMsChunks(void)
 {
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(),HvCallCfg_Min);
+	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Min);
 }
-//================================================================
-static inline u64		HvLpConfig_getMinMsPages(void)
+
+static inline u64 HvLpConfig_getMinMsPages(void)
 {
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(),HvCallCfg_Min) * HVPAGESPERCHUNK;
+	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Min)
+		* HVPAGESPERCHUNK;
 }
-//================================================================
-static inline u64		HvLpConfig_getMinRuntimeMsMegs(void)
+
+static inline u64 HvLpConfig_getMinRuntimeMsMegs(void)
 {
-	return HvCallCfg_getMinRuntimeMsChunks(HvLpConfig_getLpIndex()) / HVCHUNKSPERMEG;
+	return HvCallCfg_getMinRuntimeMsChunks(HvLpConfig_getLpIndex())
+		/ HVCHUNKSPERMEG;
 }
-//===============================================================
-static inline u64		HvLpConfig_getMinRuntimeMsChunks(void)
+
+static inline u64 HvLpConfig_getMinRuntimeMsChunks(void)
 {
 	return HvCallCfg_getMinRuntimeMsChunks(HvLpConfig_getLpIndex());
 }
-//===============================================================
-static inline u64		HvLpConfig_getMinRuntimeMsPages(void)
+
+static inline u64 HvLpConfig_getMinRuntimeMsPages(void)
 {
-	return HvCallCfg_getMinRuntimeMsChunks(HvLpConfig_getLpIndex()) * HVPAGESPERCHUNK;
+	return HvCallCfg_getMinRuntimeMsChunks(HvLpConfig_getLpIndex())
+		* HVPAGESPERCHUNK;
 }
-//===============================================================
-static inline u64		HvLpConfig_getMaxMsMegs(void)
+
+static inline u64 HvLpConfig_getMaxMsMegs(void)
 {
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(),HvCallCfg_Max) / HVCHUNKSPERMEG;
+	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Max)
+		/ HVCHUNKSPERMEG;
 }
-//===============================================================
-static inline u64		HvLpConfig_getMaxMsChunks(void)
+
+static inline u64 HvLpConfig_getMaxMsChunks(void)
 {
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(),HvCallCfg_Max);
+	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Max);
 }
-//===============================================================
-static inline u64		HvLpConfig_getMaxMsPages(void)
+
+static inline u64 HvLpConfig_getMaxMsPages(void)
 {
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(),HvCallCfg_Max) * HVPAGESPERCHUNK;
+	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Max)
+		* HVPAGESPERCHUNK;
 }
-//===============================================================
-static inline u64		HvLpConfig_getInitMsMegs(void)
+
+static inline u64 HvLpConfig_getInitMsMegs(void)
 {
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(),HvCallCfg_Init) / HVCHUNKSPERMEG;
+	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Init)
+		/ HVCHUNKSPERMEG;
 }
-//===============================================================
-static inline u64		HvLpConfig_getInitMsChunks(void)
+
+static inline u64 HvLpConfig_getInitMsChunks(void)
 {
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(),HvCallCfg_Init);
+	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Init);
 }
-//===============================================================
-static inline u64		HvLpConfig_getInitMsPages(void)
-{    return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(),HvCallCfg_Init) * HVPAGESPERCHUNK;
+
+static inline u64 HvLpConfig_getInitMsPages(void)
+{
+	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Init)
+		* HVPAGESPERCHUNK;
 }
-//===============================================================
-static inline u64		HvLpConfig_getSystemPhysicalProcessors(void)
+
+static inline u64 HvLpConfig_getSystemPhysicalProcessors(void)
 {
 	return HvCallCfg_getSystemPhysicalProcessors();
 }
-//===============================================================
-static inline u64		HvLpConfig_getSystemLogicalProcessors(void)
+
+static inline u64 HvLpConfig_getSystemLogicalProcessors(void)
 {
-	return HvCallCfg_getSystemPhysicalProcessors() * (/*getPaca()->getSecondaryThreadCount() +*/ 1);
+	return HvCallCfg_getSystemPhysicalProcessors()
+		* (/*getPaca()->getSecondaryThreadCount() +*/ 1);
 }
-//===============================================================
-static inline u64		HvLpConfig_getNumProcsInSharedPool(HvLpSharedPoolIndex sPI)
+
+static inline u64 HvLpConfig_getNumProcsInSharedPool(HvLpSharedPoolIndex sPI)
 {
 	return HvCallCfg_getNumProcsInSharedPool(sPI);
 }
-//===============================================================
-static inline u64		HvLpConfig_getPhysicalProcessors(void)
+
+static inline u64 HvLpConfig_getPhysicalProcessors(void)
 {
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),HvCallCfg_Cur);
+	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
+			HvCallCfg_Cur);
 }
-//===============================================================
-static inline u64		HvLpConfig_getLogicalProcessors(void)
+
+static inline u64 HvLpConfig_getLogicalProcessors(void)
 {
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),HvCallCfg_Cur) * (/*getPaca()->getSecondaryThreadCount() +*/ 1);
+	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
+			HvCallCfg_Cur)
+		* (/*getPaca()->getSecondaryThreadCount() +*/ 1);
 }
-//===============================================================
-static inline HvLpSharedPoolIndex	HvLpConfig_getSharedPoolIndex(void)
+
+static inline HvLpSharedPoolIndex HvLpConfig_getSharedPoolIndex(void)
 {
 	return HvCallCfg_getSharedPoolIndex(HvLpConfig_getLpIndex());
 }
-//===============================================================
-static inline u64		HvLpConfig_getSharedProcUnits(void)
+
+static inline u64 HvLpConfig_getSharedProcUnits(void)
 {
-	return HvCallCfg_getSharedProcUnits(HvLpConfig_getLpIndex(),HvCallCfg_Cur);
+	return HvCallCfg_getSharedProcUnits(HvLpConfig_getLpIndex(),
+			HvCallCfg_Cur);
 }
-//===============================================================
-static inline u64		HvLpConfig_getMinSharedProcUnits(void)
+
+static inline u64 HvLpConfig_getMinSharedProcUnits(void)
 {
-	return HvCallCfg_getSharedProcUnits(HvLpConfig_getLpIndex(),HvCallCfg_Min);
+	return HvCallCfg_getSharedProcUnits(HvLpConfig_getLpIndex(),
+			HvCallCfg_Min);
 }
-//===============================================================
-static inline u64		HvLpConfig_getMaxSharedProcUnits(void)
+
+static inline u64 HvLpConfig_getMaxSharedProcUnits(void)
 {
-	return HvCallCfg_getSharedProcUnits(HvLpConfig_getLpIndex(),HvCallCfg_Max);
+	return HvCallCfg_getSharedProcUnits(HvLpConfig_getLpIndex(),
+			HvCallCfg_Max);
 }
-//===============================================================
-static inline u64		HvLpConfig_getMinPhysicalProcessors(void)
+
+static inline u64 HvLpConfig_getMinPhysicalProcessors(void)
 {
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),HvCallCfg_Min);
+	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
+			HvCallCfg_Min);
 }
-//===============================================================
-static inline u64		HvLpConfig_getMinLogicalProcessors(void)
+
+static inline u64 HvLpConfig_getMinLogicalProcessors(void)
 {
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),HvCallCfg_Min) * (/*getPaca()->getSecondaryThreadCount() +*/ 1);
+	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
+			HvCallCfg_Min)
+		* (/*getPaca()->getSecondaryThreadCount() +*/ 1);
 }
-//===============================================================
-static inline u64		HvLpConfig_getMaxPhysicalProcessors(void)
+
+static inline u64 HvLpConfig_getMaxPhysicalProcessors(void)
 {
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),HvCallCfg_Max);
+	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
+			HvCallCfg_Max);
 }
-//===============================================================
-static inline u64		HvLpConfig_getMaxLogicalProcessors(void)
+
+static inline u64 HvLpConfig_getMaxLogicalProcessors(void)
 {
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),HvCallCfg_Max) * (/*getPaca()->getSecondaryThreadCount() +*/ 1);
+	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
+			HvCallCfg_Max)
+		* (/*getPaca()->getSecondaryThreadCount() +*/ 1);
 }
-//===============================================================
-static inline u64		HvLpConfig_getInitPhysicalProcessors(void)
+
+static inline u64 HvLpConfig_getInitPhysicalProcessors(void)
 {
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),HvCallCfg_Init);
+	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
+			HvCallCfg_Init);
 }
-//===============================================================
-static inline u64		HvLpConfig_getInitLogicalProcessors(void)
+
+static inline u64 HvLpConfig_getInitLogicalProcessors(void)
 {
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),HvCallCfg_Init) * (/*getPaca()->getSecondaryThreadCount() +*/ 1);
+	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
+			HvCallCfg_Init)
+		* (/*getPaca()->getSecondaryThreadCount() +*/ 1);
 }
-//================================================================
-static inline HvLpVirtualLanIndexMap	HvLpConfig_getVirtualLanIndexMap(void)
+
+static inline HvLpVirtualLanIndexMap HvLpConfig_getVirtualLanIndexMap(void)
 {
 	return HvCallCfg_getVirtualLanIndexMap(HvLpConfig_getLpIndex_outline());
 }
-//===============================================================
-static inline HvLpVirtualLanIndexMap	HvLpConfig_getVirtualLanIndexMapForLp(HvLpIndex lp)
+
+static inline HvLpVirtualLanIndexMap HvLpConfig_getVirtualLanIndexMapForLp(
+		HvLpIndex lp)
 {
 	return HvCallCfg_getVirtualLanIndexMap(lp);
 }
-//================================================================
-static inline HvLpIndex	HvLpConfig_getBusOwner(HvBusNumber busNumber)
+
+static inline HvLpIndex HvLpConfig_getBusOwner(HvBusNumber busNumber)
 {
 	return HvCallCfg_getBusOwner(busNumber);
 }
-//===============================================================
-static inline int		HvLpConfig_isBusDedicated(HvBusNumber busNumber)
+
+static inline int HvLpConfig_isBusDedicated(HvBusNumber busNumber)
 {
 	return HvCallCfg_isBusDedicated(busNumber);
 }
-//================================================================
-static inline HvLpIndexMap	HvLpConfig_getBusAllocation(HvBusNumber busNumber)
+
+static inline HvLpIndexMap HvLpConfig_getBusAllocation(HvBusNumber busNumber)
 {
 	return HvCallCfg_getBusAllocation(busNumber);
 }
-//================================================================
-// returns the absolute real address of the load area
-static inline u64		HvLpConfig_getLoadAddress(void)
+
+/* returns the absolute real address of the load area */
+static inline u64 HvLpConfig_getLoadAddress(void)
 {
 	return itLpNaca.xLoadAreaAddr & 0x7fffffffffffffff;
 }
-//================================================================
-static inline u64		HvLpConfig_getLoadPages(void)
+
+static inline u64 HvLpConfig_getLoadPages(void)
 {
 	return itLpNaca.xLoadAreaChunks * HVPAGESPERCHUNK;
 }
-//================================================================
-static inline int		HvLpConfig_isBusOwnedByThisLp(HvBusNumber busNumber)
+
+static inline int HvLpConfig_isBusOwnedByThisLp(HvBusNumber busNumber)
 {
 	HvLpIndex busOwner = HvLpConfig_getBusOwner(busNumber);
 	return (busOwner == HvLpConfig_getLpIndex());
 }
-//================================================================
-static inline int         HvLpConfig_doLpsCommunicateOnVirtualLan(HvLpIndex lp1, HvLpIndex lp2)
+
+static inline int HvLpConfig_doLpsCommunicateOnVirtualLan(HvLpIndex lp1,
+		HvLpIndex lp2)
 {
-	HvLpVirtualLanIndexMap virtualLanIndexMap1 = HvCallCfg_getVirtualLanIndexMap( lp1 );
-	HvLpVirtualLanIndexMap virtualLanIndexMap2 = HvCallCfg_getVirtualLanIndexMap( lp2 );
+	HvLpVirtualLanIndexMap virtualLanIndexMap1 =
+		HvCallCfg_getVirtualLanIndexMap(lp1);
+	HvLpVirtualLanIndexMap virtualLanIndexMap2 =
+		HvCallCfg_getVirtualLanIndexMap(lp2);
 	return ((virtualLanIndexMap1 & virtualLanIndexMap2) != 0);
 }
-//================================================================
-static inline HvLpIndex		HvLpConfig_getHostingLpIndex(HvLpIndex lp)
+
+static inline HvLpIndex HvLpConfig_getHostingLpIndex(HvLpIndex lp)
 {
 	return HvCallCfg_getHostingLpIndex(lp);
 }
-//================================================================
 
 #endif /* _HVLPCONFIG_H */
diff --git a/include/asm-ppc64/iSeries/HvLpEvent.h b/include/asm-ppc64/iSeries/HvLpEvent.h
index 30936e43306..865000de79b 100644
--- a/include/asm-ppc64/iSeries/HvLpEvent.h
+++ b/include/asm-ppc64/iSeries/HvLpEvent.h
@@ -1,27 +1,24 @@
 /*
  * HvLpEvent.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
  */
 
-//======================================================================
-//
-//	This file contains the class for HV events in the system.
-//
-//=====================================================================
+/* This file contains the class for HV events in the system. */
+
 #ifndef _HVLPEVENT_H
 #define _HVLPEVENT_H
 
@@ -30,69 +27,70 @@
 #include <asm/iSeries/HvTypes.h>
 #include <asm/iSeries/HvCallEvent.h>
 
-//=====================================================================
-//
-// HvLpEvent is the structure for Lp Event messages passed between
-// partitions through PLIC. 
-//
-//=====================================================================
-
-struct HvEventFlags
-{
-	u8	xValid:1;		// Indicates a valid request	x00-x00
-	u8	xRsvd1:4;		// Reserved			...
-	u8	xAckType:1;		// Immediate or deferred	...
-	u8	xAckInd:1;		// Indicates if ACK required	...
-	u8	xFunction:1;		// Interrupt or Acknowledge	...
+/*
+ * HvLpEvent is the structure for Lp Event messages passed between
+ * partitions through PLIC.
+ */
+
+struct HvEventFlags {
+	u8	xValid:1;	/* Indicates a valid request	x00-x00 */
+	u8	xRsvd1:4;	/* Reserved			... */
+	u8	xAckType:1;	/* Immediate or deferred	... */
+	u8	xAckInd:1;	/* Indicates if ACK required	... */
+	u8	xFunction:1;	/* Interrupt or Acknowledge	... */
 };
 
 
-struct HvLpEvent
-{
-	struct HvEventFlags xFlags;	// Event flags			x00-x00
-	u8	xType;			// Type of message		x01-x01
-	u16	xSubtype;		// Subtype for event		x02-x03
-	u8	xSourceLp;		// Source LP			x04-x04
-	u8	xTargetLp;		// Target LP			x05-x05
-	u8	xSizeMinus1;		// Size of Derived class - 1	x06-x06
-	u8	xRc;			// RC for Ack flows		x07-x07
-	u16	xSourceInstanceId;	// Source sides instance id	x08-x09
-	u16	xTargetInstanceId;	// Target sides instance id	x0A-x0B
+struct HvLpEvent {
+	struct HvEventFlags xFlags;	/* Event flags		      x00-x00 */
+	u8	xType;			/* Type of message	      x01-x01 */
+	u16	xSubtype;		/* Subtype for event	      x02-x03 */
+	u8	xSourceLp;		/* Source LP		      x04-x04 */
+	u8	xTargetLp;		/* Target LP		      x05-x05 */
+	u8	xSizeMinus1;		/* Size of Derived class - 1  x06-x06 */
+	u8	xRc;			/* RC for Ack flows	      x07-x07 */
+	u16	xSourceInstanceId;	/* Source sides instance id   x08-x09 */
+	u16	xTargetInstanceId;	/* Target sides instance id   x0A-x0B */
 	union {
-		u32	xSubtypeData;	// Data usable by the subtype	x0C-x0F
-		u16	xSubtypeDataShort[2];	// Data as 2 shorts
-		u8	xSubtypeDataChar[4];	// Data as 4 chars
+		u32	xSubtypeData;	/* Data usable by the subtype x0C-x0F */
+		u16	xSubtypeDataShort[2];	/* Data as 2 shorts */
+		u8	xSubtypeDataChar[4];	/* Data as 4 chars */
 	} x;
 
-	u64	xCorrelationToken;	// Unique value for source/type x10-x17
+	u64	xCorrelationToken;	/* Unique value for source/type x10-x17 */
 };
 
-// Lp Event handler function
 typedef void (*LpEventHandler)(struct HvLpEvent *, struct pt_regs *);
 
-// Register a handler for an event type
-//  returns 0 on success
-extern int HvLpEvent_registerHandler( HvLpEvent_Type eventType, LpEventHandler hdlr);
-
-// Unregister a handler for an event type
-//  This call will sleep until the handler being removed is guaranteed to
-//  be no longer executing on any CPU. Do not call with locks held.
-//
-//  returns 0 on success
-//  Unregister will fail if there are any paths open for the type
-extern int HvLpEvent_unregisterHandler( HvLpEvent_Type eventType );
+/* Register a handler for an event type - returns 0 on success */
+extern int HvLpEvent_registerHandler(HvLpEvent_Type eventType,
+		LpEventHandler hdlr);
 
-// Open an Lp Event Path for an event type
-//  returns 0 on success
-//  openPath will fail if there is no handler registered for the event type.
-//  The lpIndex specified is the partition index for the target partition
-//  (for VirtualIo, VirtualLan and SessionMgr) other types specify zero) 
-extern int HvLpEvent_openPath( HvLpEvent_Type eventType, HvLpIndex lpIndex );
+/*
+ * Unregister a handler for an event type
+ *
+ * This call will sleep until the handler being removed is guaranteed to
+ * be no longer executing on any CPU. Do not call with locks held.
+ *
+ *  returns 0 on success
+ *  Unregister will fail if there are any paths open for the type
+ */
+extern int HvLpEvent_unregisterHandler(HvLpEvent_Type eventType);
 
+/*
+ * Open an Lp Event Path for an event type
+ * returns 0 on success
+ * openPath will fail if there is no handler registered for the event type.
+ * The lpIndex specified is the partition index for the target partition
+ * (for VirtualIo, VirtualLan and SessionMgr) other types specify zero)
+ */
+extern int HvLpEvent_openPath(HvLpEvent_Type eventType, HvLpIndex lpIndex);
 
-// Close an Lp Event Path for a type and partition
-//  returns 0 on sucess
-extern int HvLpEvent_closePath( HvLpEvent_Type eventType, HvLpIndex lpIndex );
+/*
+ * Close an Lp Event Path for a type and partition
+ * returns 0 on sucess
+ */
+extern int HvLpEvent_closePath(HvLpEvent_Type eventType, HvLpIndex lpIndex);
 
 #define HvLpEvent_Type_Hypervisor 0
 #define HvLpEvent_Type_MachineFac 1
@@ -141,4 +139,4 @@ extern int HvLpEvent_closePath( HvLpEvent_Type eventType, HvLpIndex lpIndex );
 #define HvLpDma_Rc_InvalidAddress 4
 #define HvLpDma_Rc_InvalidLength 5
 
-#endif // _HVLPEVENT_H
+#endif /* _HVLPEVENT_H */
diff --git a/include/asm-ppc64/iSeries/HvReleaseData.h b/include/asm-ppc64/iSeries/HvReleaseData.h
index 183e5e738c2..c0cd21caf8f 100644
--- a/include/asm-ppc64/iSeries/HvReleaseData.h
+++ b/include/asm-ppc64/iSeries/HvReleaseData.h
@@ -1,17 +1,17 @@
 /*
  * HvReleaseData.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -19,47 +19,43 @@
 #ifndef _HVRELEASEDATA_H
 #define _HVRELEASEDATA_H
 
-//=============================================================================
-//
-//   This control block contains the critical information about the 
-//   release so that it can be changed in the future (ie, the virtual 
-//   address of the OS's NACA).
-//
+/*
+ * This control block contains the critical information about the
+ * release so that it can be changed in the future (ie, the virtual
+ * address of the OS's NACA).
+ */
 #include <asm/types.h>
 #include <asm/naca.h>
 
-//=============================================================================
-//
-//	When we IPL a secondary partition, we will check if if the 
-//	secondary xMinPlicVrmIndex > the primary xVrmIndex.  
-//	If it is then this tells PLIC that this secondary is not 
-//	supported running on this "old" of a level of PLIC.
-//
-//	Likewise, we will compare the primary xMinSlicVrmIndex to 
-//	the secondary xVrmIndex. 
-//	If the primary xMinSlicVrmDelta > secondary xVrmDelta then we 
-//	know that this PLIC does not support running an OS "that old".
-//
-//=============================================================================
+/*
+ * When we IPL a secondary partition, we will check if if the
+ * secondary xMinPlicVrmIndex > the primary xVrmIndex.
+ * If it is then this tells PLIC that this secondary is not
+ * supported running on this "old" of a level of PLIC.
+ *
+ * Likewise, we will compare the primary xMinSlicVrmIndex to
+ * the secondary xVrmIndex.
+ * If the primary xMinSlicVrmDelta > secondary xVrmDelta then we
+ * know that this PLIC does not support running an OS "that old".
+ */
 
-struct	HvReleaseData
-{
-	u32	xDesc;			// Descriptor	"HvRD" ebcdic	x00-x03
-	u16	xSize;			// Size of this control block	x04-x05
-	u16	xVpdAreasPtrOffset;	// Offset in NACA of ItVpdAreas	x06-x07
-	struct  naca_struct * xSlicNacaAddr; // Virt addr of SLIC NACA  x08-x0F
-	u32	xMsNucDataOffset;	// Offset of Linux Mapping Data x10-x13
-	u32	xRsvd1;			// Reserved			x14-x17
-	u16	xTagsMode:1;		// 0 == tags active, 1 == tags inactive
-	u16	xAddressSize:1;		// 0 == 64-bit, 1 == 32-bit
-	u16	xNoSharedProcs:1;	// 0 == shared procs, 1 == no shared
-	u16	xNoHMT:1;		// 0 == allow HMT, 1 == no HMT
-	u16	xRsvd2:12;		// Reserved			x18-x19
-	u16	xVrmIndex;		// VRM Index of OS image	x1A-x1B
-	u16	xMinSupportedPlicVrmIndex;// Min PLIC level  (soft)	x1C-x1D
-	u16	xMinCompatablePlicVrmIndex;// Min PLIC levelP (hard)	x1E-x1F
-	char	xVrmName[12];		// Displayable name  		x20-x2B
-	char	xRsvd3[20];		// Reserved			x2C-x3F
+struct HvReleaseData {
+	u32	xDesc;		/* Descriptor "HvRD" ebcdic	x00-x03 */
+	u16	xSize;		/* Size of this control block	x04-x05 */
+	u16	xVpdAreasPtrOffset; /* Offset in NACA of ItVpdAreas x06-x07 */
+	struct  naca_struct	*xSlicNacaAddr; /* Virt addr of SLIC NACA x08-x0F */
+	u32	xMsNucDataOffset; /* Offset of Linux Mapping Data x10-x13 */
+	u32	xRsvd1;		/* Reserved			x14-x17 */
+	u16	xTagsMode:1;	/* 0 == tags active, 1 == tags inactive */
+	u16	xAddressSize:1;	/* 0 == 64-bit, 1 == 32-bit */
+	u16	xNoSharedProcs:1; /* 0 == shared procs, 1 == no shared */
+	u16	xNoHMT:1;	/* 0 == allow HMT, 1 == no HMT */
+	u16	xRsvd2:12;	/* Reserved			x18-x19 */
+	u16	xVrmIndex;	/* VRM Index of OS image	x1A-x1B */
+	u16	xMinSupportedPlicVrmIndex; /* Min PLIC level  (soft) x1C-x1D */
+	u16	xMinCompatablePlicVrmIndex; /* Min PLIC levelP (hard) x1E-x1F */
+	char	xVrmName[12];	/* Displayable name		x20-x2B */
+	char	xRsvd3[20];	/* Reserved			x2C-x3F */
 };
 
 #endif /* _HVRELEASEDATA_H */
diff --git a/include/asm-ppc64/iSeries/HvTypes.h b/include/asm-ppc64/iSeries/HvTypes.h
index 3ec49c1aec3..111191ddb18 100644
--- a/include/asm-ppc64/iSeries/HvTypes.h
+++ b/include/asm-ppc64/iSeries/HvTypes.h
@@ -1,17 +1,17 @@
 /*
  * HvTypes.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -19,74 +19,62 @@
 #ifndef _HVTYPES_H
 #define _HVTYPES_H
 
-//===========================================================================
-//                                                             Header File Id
-// Name______________:  HvTypes.H
-//
-// Description_______:
-//
-//	General typedefs for the hypervisor.
-//
-// Declared Class(es):
-//
-//===========================================================================
+/*
+ * General typedefs for the hypervisor.
+ */
 
 #include <asm/types.h>
 
-//-------------------------------------------------------------------
-// Typedefs
-//-------------------------------------------------------------------
 typedef u8	HvLpIndex;
 typedef u16	HvLpInstanceId;
-typedef u64     HvLpTOD;
-typedef u64     HvLpSystemSerialNum;
-typedef u8      HvLpDeviceSerialNum[12];
-typedef u16     HvLpSanHwSet;
-typedef u16     HvLpBus;
-typedef u16     HvLpBoard;
-typedef u16     HvLpCard;
-typedef u8      HvLpDeviceType[4];
-typedef u8      HvLpDeviceModel[3];
-typedef u64     HvIoToken;
-typedef u8      HvLpName[8];
+typedef u64	HvLpTOD;
+typedef u64	HvLpSystemSerialNum;
+typedef u8	HvLpDeviceSerialNum[12];
+typedef u16	HvLpSanHwSet;
+typedef u16	HvLpBus;
+typedef u16	HvLpBoard;
+typedef u16	HvLpCard;
+typedef u8	HvLpDeviceType[4];
+typedef u8	HvLpDeviceModel[3];
+typedef u64	HvIoToken;
+typedef u8	HvLpName[8];
 typedef u32	HvIoId;
 typedef u64	HvRealMemoryIndex;
-typedef u32     HvLpIndexMap;		// Must hold HvMaxArchitectedLps bits!!!
+typedef u32	HvLpIndexMap;	/* Must hold HvMaxArchitectedLps bits!!! */
 typedef u16	HvLpVrmIndex;
 typedef u32	HvXmGenerationId;
-typedef u8	HvLpBusPool;			
-typedef u8	HvLpSharedPoolIndex;		
+typedef u8	HvLpBusPool;
+typedef u8	HvLpSharedPoolIndex;
 typedef u16	HvLpSharedProcUnitsX100;
 typedef u8	HvLpVirtualLanIndex;
-typedef u16	HvLpVirtualLanIndexMap;	// Must hold HvMaxArchitectedVirtualLans bits!!!
-typedef u16	HvBusNumber;		// Hypervisor Bus Number
-typedef u8	HvSubBusNumber;		// Hypervisor SubBus Number
-typedef u8	HvAgentId;		// Hypervisor DevFn
+typedef u16	HvLpVirtualLanIndexMap;	/* Must hold HvMaxArchitectedVirtualLans bits!!! */
+typedef u16	HvBusNumber;	/* Hypervisor Bus Number */
+typedef u8	HvSubBusNumber;	/* Hypervisor SubBus Number */
+typedef u8	HvAgentId;	/* Hypervisor DevFn */
+
 
+#define HVMAXARCHITECTEDLPS		32
+#define HVMAXARCHITECTEDVIRTUALLANS	16
+#define HVMAXARCHITECTEDVIRTUALDISKS	32
+#define HVMAXARCHITECTEDVIRTUALCDROMS	8
+#define HVMAXARCHITECTEDVIRTUALTAPES	8
+#define HVCHUNKSIZE			(256 * 1024)
+#define HVPAGESIZE			(4 * 1024)
+#define HVLPMINMEGSPRIMARY		256
+#define HVLPMINMEGSSECONDARY		64
+#define HVCHUNKSPERMEG			4
+#define HVPAGESPERMEG			256
+#define HVPAGESPERCHUNK			64
 
-#define HVMAXARCHITECTEDLPS 32
-#define HVMAXARCHITECTEDVIRTUALLANS 16
-#define HVMAXARCHITECTEDVIRTUALDISKS 32
-#define HVMAXARCHITECTEDVIRTUALCDROMS 8
-#define HVMAXARCHITECTEDVIRTUALTAPES 8
-#define HVCHUNKSIZE 256 * 1024
-#define HVPAGESIZE 4 * 1024
-#define HVLPMINMEGSPRIMARY 256
-#define HVLPMINMEGSSECONDARY 64
-#define HVCHUNKSPERMEG 4
-#define HVPAGESPERMEG 256
-#define HVPAGESPERCHUNK 64
- 
-#define HvMaxArchitectedLps 		((HvLpIndex)HVMAXARCHITECTEDLPS)
+#define HvMaxArchitectedLps		((HvLpIndex)HVMAXARCHITECTEDLPS)
 #define HvMaxArchitectedVirtualLans	((HvLpVirtualLanIndex)16)
 #define HvLpIndexInvalid		((HvLpIndex)0xff)
 
-//--------------------------------------------------------------------
-// Enums for the sub-components under PLIC
-// Used in HvCall  and HvPrimaryCall
-//--------------------------------------------------------------------
-enum   HvCallCompIds
-{
+/*
+ * Enums for the sub-components under PLIC
+ * Used in HvCall  and HvPrimaryCall
+ */
+enum   HvCallCompIds {
 	HvCallCompId = 0,
 	HvCallCpuCtlsCompId = 1,
 	HvCallCfgCompId = 2,
@@ -97,18 +85,18 @@ enum   HvCallCompIds
 	HvCallSmCompId = 7,
 	HvCallSpdCompId = 8,
 	HvCallXmCompId = 9,
-	HvCallRioCompId = 10, 
+	HvCallRioCompId = 10,
 	HvCallRsvd3CompId = 11,
 	HvCallRsvd2CompId = 12,
 	HvCallRsvd1CompId = 13,
 	HvCallMaxCompId = 14,
-	HvPrimaryCallCompId = 0,    
+	HvPrimaryCallCompId = 0,
 	HvPrimaryCallCfgCompId = 1,
-	HvPrimaryCallPciCompId = 2,    
+	HvPrimaryCallPciCompId = 2,
 	HvPrimaryCallSmCompId = 3,
 	HvPrimaryCallSpdCompId = 4,
 	HvPrimaryCallXmCompId = 5,
-	HvPrimaryCallRioCompId = 6, 
+	HvPrimaryCallRioCompId = 6,
 	HvPrimaryCallRsvd7CompId = 7,
 	HvPrimaryCallRsvd6CompId = 8,
 	HvPrimaryCallRsvd5CompId = 9,
@@ -116,7 +104,7 @@ enum   HvCallCompIds
 	HvPrimaryCallRsvd3CompId = 11,
 	HvPrimaryCallRsvd2CompId = 12,
 	HvPrimaryCallRsvd1CompId = 13,
-	HvPrimaryCallMaxCompId = HvCallMaxCompId     
+	HvPrimaryCallMaxCompId = HvCallMaxCompId
 };
 
 struct HvLpBufferList {
-- 
cgit v1.2.3-70-g09d2


From fcee38952609fccb2bdfe166b3b96bd75a292aa6 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:34 -0700
Subject: [PATCH] ppc64 iSeries: more header file white space cleanups

This patch just contains white space and comment cleanups in the iSeries
headers files.  There are no semantic changes.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-ppc64/iSeries/IoHriMainStore.h    |  33 +++---
 include/asm-ppc64/iSeries/IoHriProcessorVpd.h |  30 +++--
 include/asm-ppc64/iSeries/ItExtVpdPanel.h     |  52 ++++-----
 include/asm-ppc64/iSeries/ItIplParmsReal.h    |  97 ++++++++--------
 include/asm-ppc64/iSeries/ItLpNaca.h          |  42 +++----
 include/asm-ppc64/iSeries/ItLpQueue.h         |  74 ++++++------
 include/asm-ppc64/iSeries/ItLpRegSave.h       |  41 +++----
 include/asm-ppc64/iSeries/ItSpCommArea.h      |  10 +-
 include/asm-ppc64/iSeries/ItVpdAreas.h        | 121 ++++++++++----------
 include/asm-ppc64/iSeries/LparData.h          |  27 +++--
 include/asm-ppc64/iSeries/LparMap.h           |  42 +++----
 include/asm-ppc64/iSeries/XmPciLpEvent.h      |   3 -
 include/asm-ppc64/iSeries/iSeries_io.h        |  59 +++++-----
 include/asm-ppc64/iSeries/iSeries_pci.h       | 159 +++++++++++++-------------
 include/asm-ppc64/iSeries/mf.h                |   5 +-
 include/asm-ppc64/iSeries/vio.h               |  37 +++---
 16 files changed, 395 insertions(+), 437 deletions(-)

(limited to 'include')

diff --git a/include/asm-ppc64/iSeries/IoHriMainStore.h b/include/asm-ppc64/iSeries/IoHriMainStore.h
index ff00e865f62..fbb3ee4d174 100644
--- a/include/asm-ppc64/iSeries/IoHriMainStore.h
+++ b/include/asm-ppc64/iSeries/IoHriMainStore.h
@@ -1,17 +1,17 @@
 /*
  * IoHriMainStore.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -21,7 +21,7 @@
 #define _IOHRIMAINSTORE_H
 
 /* Main Store Vpd for Condor,iStar,sStar */
-struct IoHriMainStoreSegment4 {    
+struct IoHriMainStoreSegment4 {
 	u8	msArea0Exists:1;
 	u8	msArea1Exists:1;
 	u8	msArea2Exists:1;
@@ -51,7 +51,7 @@ struct IoHriMainStoreSegment4 {
 	u8	msArea1HasRiserVpd:1;
 	u8	msArea2HasRiserVpd:1;
 	u8	msArea3HasRiserVpd:1;
-	u8	reserved5:4;	
+	u8	reserved5:4;
 	u8	reserved6;
 	u16	reserved7;
 
@@ -82,8 +82,8 @@ struct IoHriMainStoreVpdFruData {
 };
 
 struct IoHriMainStoreAdrRangeBlock {
-	void *	blockStart      __attribute((packed));
-	void *	blockEnd        __attribute((packed));
+	void	*blockStart      __attribute((packed));
+	void	*blockEnd        __attribute((packed));
 	u32	blockProcChipId __attribute((packed));
 };
 
@@ -102,7 +102,7 @@ struct IoHriMainStoreArea4 {
 	u32	procNodeId			__attribute((packed));
 
 	u32	numAdrRangeBlocks		__attribute((packed));
-	struct IoHriMainStoreAdrRangeBlock xAdrRangeBlock[MaxAreaAdrRangeBlocks] __attribute((packed));
+	struct IoHriMainStoreAdrRangeBlock xAdrRangeBlock[MaxAreaAdrRangeBlocks]	__attribute((packed));
 
 	struct IoHriMainStoreChipInfo1	chipInfo0	__attribute((packed));
 	struct IoHriMainStoreChipInfo1	chipInfo1	__attribute((packed));
@@ -113,17 +113,17 @@ struct IoHriMainStoreArea4 {
 	struct IoHriMainStoreChipInfo1	chipInfo6	__attribute((packed));
 	struct IoHriMainStoreChipInfo1	chipInfo7	__attribute((packed));
 
-	void *   msRamAreaArray			__attribute((packed));
+	void	*msRamAreaArray			__attribute((packed));
 	u32	msRamAreaArrayNumEntries	__attribute((packed));
 	u32	msRamAreaArrayEntrySize		__attribute((packed));
 
 	u32	numaDimmExists			__attribute((packed));
 	u32	numaDimmFunctional		__attribute((packed));
-	void *	numaDimmArray			__attribute((packed));
+	void	*numaDimmArray			__attribute((packed));
 	u32	numaDimmArrayNumEntries		__attribute((packed));
 	u32	numaDimmArrayEntrySize		__attribute((packed));
 
-	struct IoHriMainStoreVpdIdData  idData	__attribute((packed));
+	struct IoHriMainStoreVpdIdData idData	__attribute((packed));
 
 	u64	powerData			__attribute((packed));
 	u64	cardAssemblyPartNum		__attribute((packed));
@@ -143,7 +143,7 @@ struct IoHriMainStoreArea4 {
 };
 
 
-struct IoHriMainStoreSegment5 {    
+struct IoHriMainStoreSegment5 {
 	u16	reserved1;
 	u8	reserved2;
 	u8	msVpdFormat;
@@ -151,17 +151,14 @@ struct IoHriMainStoreSegment5 {
 	u32	totalMainStore;
 	u64	maxConfiguredMsAdr;
 
-	struct IoHriMainStoreArea4*	msAreaArray;
+	struct IoHriMainStoreArea4	*msAreaArray;
 	u32	msAreaArrayNumEntries;
 	u32	msAreaArrayEntrySize;
 
-	u32	msAreaExists;    
+	u32	msAreaExists;
 	u32	msAreaFunctional;
 
 	u64	reserved3;
 };
 
-
-
-#endif // _IOHRIMAINSTORE_H
-
+#endif	/* _IOHRIMAINSTORE_H */
diff --git a/include/asm-ppc64/iSeries/IoHriProcessorVpd.h b/include/asm-ppc64/iSeries/IoHriProcessorVpd.h
index 96543388865..ccb7b7c55a4 100644
--- a/include/asm-ppc64/iSeries/IoHriProcessorVpd.h
+++ b/include/asm-ppc64/iSeries/IoHriProcessorVpd.h
@@ -1,17 +1,17 @@
 /*
  * IoHriProcessorVpd.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -19,16 +19,12 @@
 #ifndef _IOHRIPROCESSORVPD_H
 #define _IOHRIPROCESSORVPD_H
 
-//===================================================================
-//
-//	This struct maps Processor Vpd that is DMAd to SLIC by CSP 
-//
-
 #include <asm/types.h>
 
-struct IoHriProcessorVpd
-{
-
+/*
+ * This struct maps Processor Vpd that is DMAd to SLIC by CSP
+ */
+struct IoHriProcessorVpd {
 	u8	xFormat;		// VPD format indicator		x00-x00
 	u8	xProcStatus:8;		// Processor State		x01-x01
 	u8	xSecondaryThreadCount;	// Secondary thread cnt		x02-x02
@@ -40,12 +36,12 @@ struct IoHriProcessorVpd
 	u16	xRsvd2;			// Reserved			x06-x07
 	u32	xHwNodeId;		// Hardware node id		x08-x0B
 	u32	xHwProcId;		// Hardware processor id	x0C-x0F
-	
+
 	u32	xTypeNum;		// Card Type/CCIN number	x10-x13
 	u32	xModelNum;		// Model/Feature number		x14-x17
 	u64	xSerialNum;		// Serial number		x18-x1F
-	char xPartNum[12];		// Book Part or FPU number	x20-x2B
-	char xMfgID[4];			// Manufacturing ID		x2C-x2F
+	char	xPartNum[12];		// Book Part or FPU number	x20-x2B
+	char	xMfgID[4];		// Manufacturing ID		x2C-x2F
 
 	u32	xProcFreq;		// Processor Frequency		x30-x33
 	u32	xTimeBaseFreq;		// Time Base Frequency		x34-x37
@@ -71,7 +67,7 @@ struct IoHriProcessorVpd
 	u32	xDataL3CacheSizeKB;	// L3 data cache size in KB	x80-x83
 	u32	xDataL3CacheLineSize;	// L3 data cache block size	x84-x87
 	u64	xRsvd6;			// Reserved			x88-x8F
-   
+
 	u64	xFruLabel;		// Card Location Label		x90-x97
 	u8	xSlotsOnCard;		// Slots on card (0=no slots)	x98-x98
 	u8	xPartLocFlag;		// Location flag (0-pluggable 1-imbedded) x99-x99
@@ -79,10 +75,10 @@ struct IoHriProcessorVpd
 	u8	xSmartCardPortNo;	// Smart card port number	x9C-x9C
 	u8	xRsvd7;			// Reserved			x9D-x9D
 	u16	xFrameIdAndRackUnit;	// Frame ID and rack unit adr	x9E-x9F
-    
+
 	u8	xRsvd8[24];		// Reserved			xA0-xB7
 
-	char xProcSrc[72];		// CSP format SRC		xB8-xFF
+	char	xProcSrc[72];		// CSP format SRC		xB8-xFF
 };
 
 #endif /* _IOHRIPROCESSORVPD_H */
diff --git a/include/asm-ppc64/iSeries/ItExtVpdPanel.h b/include/asm-ppc64/iSeries/ItExtVpdPanel.h
index dee6b127303..b7149079d3c 100644
--- a/include/asm-ppc64/iSeries/ItExtVpdPanel.h
+++ b/include/asm-ppc64/iSeries/ItExtVpdPanel.h
@@ -1,17 +1,17 @@
 /*
  * ItExtVpdPanel.h
  * Copyright (C) 2002  Dave Boutcher IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -20,39 +20,31 @@
 #define _ITEXTVPDPANEL_H
 
 /*
- *
- *	This struct maps the panel information 
+ *	This struct maps the panel information
  *
  * Warning:
  *	This data must match the architecture for the panel information
- *
  */
 
-
-/*-------------------------------------------------------------------
- * Standard Includes
- *------------------------------------------------------------------- 
-*/
 #include <asm/types.h>
 
-struct ItExtVpdPanel
-{
-  // Definition of the Extended Vpd On Panel Data Area
-  char                      systemSerial[8];
-  char                      mfgID[4];
-  char                      reserved1[24];
-  char                      machineType[4];
-  char                      systemID[6];
-  char                      somUniqueCnt[4];
-  char                      serialNumberCount;
-  char                      reserved2[7];
-  u16                       bbu3;
-  u16                       bbu2;
-  u16                       bbu1;
-  char                      xLocationLabel[8];
-  u8                        xRsvd1[6];
-  u16                       xFrameId;
-  u8                        xRsvd2[48];
+struct ItExtVpdPanel {
+	/* Definition of the Extended Vpd On Panel Data Area */
+	char	systemSerial[8];
+	char	mfgID[4];
+	char	reserved1[24];
+	char	machineType[4];
+	char	systemID[6];
+	char	somUniqueCnt[4];
+	char	serialNumberCount;
+	char	reserved2[7];
+	u16	bbu3;
+	u16	bbu2;
+	u16	bbu1;
+	char	xLocationLabel[8];
+	u8	xRsvd1[6];
+	u16	xFrameId;
+	u8	xRsvd2[48];
 };
 
-#endif /* _ITEXTVPDPANEL_H  */
+#endif /* _ITEXTVPDPANEL_H */
diff --git a/include/asm-ppc64/iSeries/ItIplParmsReal.h b/include/asm-ppc64/iSeries/ItIplParmsReal.h
index 4d8b430ab12..14f0b0a2a82 100644
--- a/include/asm-ppc64/iSeries/ItIplParmsReal.h
+++ b/include/asm-ppc64/iSeries/ItIplParmsReal.h
@@ -1,17 +1,17 @@
 /*
  * ItIplParmsReal.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -19,58 +19,51 @@
 #ifndef _ITIPLPARMSREAL_H
 #define _ITIPLPARMSREAL_H
 
-//==============================================================================
-//
-//	This struct maps the IPL Parameters DMA'd from the SP.                  
-//
-// Warning:
-//	This data must map in exactly 64 bytes and match the architecture for
-//	the IPL parms
-//
-//=============================================================================
-
+/*
+ *	This struct maps the IPL Parameters DMA'd from the SP.
+ *
+ * Warning:
+ *	This data must map in exactly 64 bytes and match the architecture for
+ *	the IPL parms
+ */
 
-//-------------------------------------------------------------------
-// Standard Includes
-//-------------------------------------------------------------------
 #include <asm/types.h>
 
-struct ItIplParmsReal
-{
-	u8	xFormat;		// Defines format of IplParms		x00-x00
-	u8	xRsvd01:6;		// Reserved				x01-x01
-	u8	xAlternateSearch:1;	// Alternate search indicator		...
-	u8	xUaSupplied:1;		// UA Supplied on programmed IPL	...
-	u8	xLsUaFormat;		// Format byte for UA			x02-x02
-	u8	xRsvd02;		// Reserved				x03-x03
-	u32	xLsUa;			// LS UA				x04-x07
-	u32	xUnusedLsLid;		// First OS LID to load			x08-x0B
-	u16	xLsBusNumber;		// LS Bus Number			x0C-x0D
-	u8	xLsCardAdr;		// LS Card Address			x0E-x0E
-	u8	xLsBoardAdr;		// LS Board Address			x0F-x0F
-	u32	xRsvd03;		// Reserved				x10-x13
-	u8	xSpcnPresent:1;		// SPCN present				x14-x14
-	u8	xCpmPresent:1;		// CPM present				...
-	u8	xRsvd04:6;		// Reserved				...
-	u8	xRsvd05:4;		// Reserved				x15-x15
-	u8	xKeyLock:4;		// Keylock setting			...
-	u8	xRsvd06:6;		// Reserved				x16-x16
-	u8	xIplMode:2;		// Ipl mode (A|B|C|D)			...
-	u8	xHwIplType;		// Fast v slow v slow EC HW IPL		x17-x17
-	u16	xCpmEnabledIpl:1;	// CPM in effect when IPL initiated	x18-x19
-	u16	xPowerOnResetIpl:1;	// Indicate POR condition		...
-	u16	xMainStorePreserved:1;	// Main Storage is preserved		...
-	u16	xRsvd07:13;		// Reserved				...
-	u16	xIplSource:16;		// Ipl source				x1A-x1B
-	u8	xIplReason:8;		// Reason for this IPL			x1C-x1C
-	u8	xRsvd08;		// Reserved				x1D-x1D
-	u16	xRsvd09;		// Reserved				x1E-x1F
-	u16	xSysBoxType;		// System Box Type			x20-x21
-	u16	xSysProcType;		// System Processor Type		x22-x23
-	u32	xRsvd10;		// Reserved				x24-x27
-	u64	xRsvd11;		// Reserved				x28-x2F
-	u64	xRsvd12;		// Reserved				x30-x37
-	u64	xRsvd13;		// Reserved				x38-x3F
+struct ItIplParmsReal {
+	u8	xFormat;		// Defines format of IplParms	x00-x00
+	u8	xRsvd01:6;		// Reserved			x01-x01
+	u8	xAlternateSearch:1;	// Alternate search indicator	...
+	u8	xUaSupplied:1;		// UA Supplied on programmed IPL...
+	u8	xLsUaFormat;		// Format byte for UA		x02-x02
+	u8	xRsvd02;		// Reserved			x03-x03
+	u32	xLsUa;			// LS UA			x04-x07
+	u32	xUnusedLsLid;		// First OS LID to load		x08-x0B
+	u16	xLsBusNumber;		// LS Bus Number		x0C-x0D
+	u8	xLsCardAdr;		// LS Card Address		x0E-x0E
+	u8	xLsBoardAdr;		// LS Board Address		x0F-x0F
+	u32	xRsvd03;		// Reserved			x10-x13
+	u8	xSpcnPresent:1;		// SPCN present			x14-x14
+	u8	xCpmPresent:1;		// CPM present			...
+	u8	xRsvd04:6;		// Reserved			...
+	u8	xRsvd05:4;		// Reserved			x15-x15
+	u8	xKeyLock:4;		// Keylock setting		...
+	u8	xRsvd06:6;		// Reserved			x16-x16
+	u8	xIplMode:2;		// Ipl mode (A|B|C|D)		...
+	u8	xHwIplType;		// Fast v slow v slow EC HW IPL	x17-x17
+	u16	xCpmEnabledIpl:1;	// CPM in effect when IPL initiatedx18-x19
+	u16	xPowerOnResetIpl:1;	// Indicate POR condition	...
+	u16	xMainStorePreserved:1;	// Main Storage is preserved	...
+	u16	xRsvd07:13;		// Reserved			...
+	u16	xIplSource:16;		// Ipl source			x1A-x1B
+	u8	xIplReason:8;		// Reason for this IPL		x1C-x1C
+	u8	xRsvd08;		// Reserved			x1D-x1D
+	u16	xRsvd09;		// Reserved			x1E-x1F
+	u16	xSysBoxType;		// System Box Type		x20-x21
+	u16	xSysProcType;		// System Processor Type	x22-x23
+	u32	xRsvd10;		// Reserved			x24-x27
+	u64	xRsvd11;		// Reserved			x28-x2F
+	u64	xRsvd12;		// Reserved			x30-x37
+	u64	xRsvd13;		// Reserved			x38-x3F
 };
 
 #endif /* _ITIPLPARMSREAL_H */
diff --git a/include/asm-ppc64/iSeries/ItLpNaca.h b/include/asm-ppc64/iSeries/ItLpNaca.h
index 5baffddfae1..942b07d9140 100644
--- a/include/asm-ppc64/iSeries/ItLpNaca.h
+++ b/include/asm-ppc64/iSeries/ItLpNaca.h
@@ -1,17 +1,17 @@
 /*
  * ItLpNaca.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -19,18 +19,13 @@
 #ifndef _ITLPNACA_H
 #define _ITLPNACA_H
 
-//=============================================================================
-//
-//	This control block contains the data that is shared between the
-//	hypervisor (PLIC) and the OS.
-//
-//=============================================================================
+/*
+ *	This control block contains the data that is shared between the
+ *	hypervisor (PLIC) and the OS.
+ */
 
-struct ItLpNaca
-{
-//=============================================================================
+struct ItLpNaca {
 // CACHE_LINE_1 0x0000 - 0x007F Contains read-only data
-//=============================================================================
 	u32	xDesc;			// Eye catcher			x00-x03
 	u16	xSize;			// Size of this class		x04-x05
 	u16	xIntHdlrOffset;		// Offset to IntHdlr array	x06-x07
@@ -59,30 +54,23 @@ struct ItLpNaca
 	u64	xLoadAreaAddr;		// ER address of load area	x28-x2F
 	u32	xLoadAreaChunks;	// Chunks for the load area	x30-x33
 	u32	xPaseSysCallCRMask;	// Mask used to test CR before  x34-x37
-	// doing an ASR switch on PASE
-	// system call.
-	u64	xSlicSegmentTablePtr;   // Pointer to Slic seg table.   x38-x3f
-	u8	xRsvd1_4[64];		//         			x40-x7F 
-   
-//=============================================================================
+					// doing an ASR switch on PASE
+					// system call.
+	u64	xSlicSegmentTablePtr;	// Pointer to Slic seg table.   x38-x3f
+	u8	xRsvd1_4[64];		//				x40-x7F
+
 // CACHE_LINE_2 0x0080 - 0x00FF Contains local read-write data
-//=============================================================================
 	u8	xRsvd2_0[128];		// Reserved			x00-x7F
 
-//=============================================================================
 // CACHE_LINE_3-6 0x0100 - 0x02FF Contains LP Queue indicators
-// NB: Padding required to keep xInterrruptHdlr at x300 which is required 
+// NB: Padding required to keep xInterrruptHdlr at x300 which is required
 // for v4r4 PLIC.
-//=============================================================================
 	u8	xOldLpQueue[128];	// LP Queue needed for v4r4	100-17F
 	u8	xRsvd3_0[384];		// Reserved			180-2FF
-//=============================================================================
+
 // CACHE_LINE_7-8 0x0300 - 0x03FF Contains the address of the OS interrupt
 //  handlers
-//=============================================================================
 	u64	xInterruptHdlr[32];	// Interrupt handlers		300-x3FF
 };
 
-//=============================================================================
-
 #endif /* _ITLPNACA_H */
diff --git a/include/asm-ppc64/iSeries/ItLpQueue.h b/include/asm-ppc64/iSeries/ItLpQueue.h
index 4f4dde2a638..b2bdef42959 100644
--- a/include/asm-ppc64/iSeries/ItLpQueue.h
+++ b/include/asm-ppc64/iSeries/ItLpQueue.h
@@ -1,17 +1,17 @@
 /*
  * ItLpQueue.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -19,47 +19,47 @@
 #ifndef _ITLPQUEUE_H
 #define _ITLPQUEUE_H
 
-//=============================================================================
-//
-//	This control block defines the simple LP queue structure that is 
-//	shared between the hypervisor (PLIC) and the OS in order to send 
-//	events to an LP.  
-//    
+/*
+ *	This control block defines the simple LP queue structure that is
+ *	shared between the hypervisor (PLIC) and the OS in order to send
+ *	events to an LP.
+ */
 
 #include <asm/types.h>
 #include <asm/ptrace.h>
 
 struct HvLpEvent;
 
-#define ITMaxLpQueues 8
+#define ITMaxLpQueues	8
 
 #define NotUsed		0	// Queue will not be used by PLIC
 #define DedicatedIo	1	// Queue dedicated to IO processor specified
 #define DedicatedLp	2	// Queue dedicated to LP specified
 #define Shared		3	// Queue shared for both IO and LP
 
-#define LpEventStackSize 4096
-#define LpEventMaxSize   256
-#define LpEventAlign	 64
+#define LpEventStackSize	4096
+#define LpEventMaxSize		256
+#define LpEventAlign		64
 
-struct ItLpQueue
-{
-//
-//  The xSlicCurEventPtr is the pointer to the next event stack entry that will
-//  become valid.  The OS must peek at this entry to determine if it is valid.
-//  PLIC will set the valid indicator as the very last store into that entry.
-//
-//  When the OS has completed processing of the event then it will mark the event
-//  as invalid so that PLIC knows it can store into that event location again.
-//
-// If the event stack fills and there are overflow events, then PLIC will set 
-// the xPlicOverflowIntPending flag in which case the OS will have to fetch the 
-// additional LP events once they have drained the event stack.
-//
-// The first 16-bytes are known by both the OS and PLIC.  The remainder of the
-// cache line is for use by the OS.
-//
-//=============================================================================
+struct ItLpQueue {
+/*
+ * The xSlicCurEventPtr is the pointer to the next event stack entry
+ * that will become valid.  The OS must peek at this entry to determine
+ * if it is valid.  PLIC will set the valid indicator as the very last
+ * store into that entry.
+ *
+ * When the OS has completed processing of the event then it will mark
+ * the event as invalid so that PLIC knows it can store into that event
+ * location again.
+ *
+ * If the event stack fills and there are overflow events, then PLIC
+ * will set the xPlicOverflowIntPending flag in which case the OS will
+ * have to fetch the additional LP events once they have drained the
+ * event stack.
+ *
+ * The first 16-bytes are known by both the OS and PLIC.  The remainder
+ * of the cache line is for use by the OS.
+ */
 	u8	xPlicOverflowIntPending;// 0x00 Overflow events are pending
 	u8	xPlicStatus;		// 0x01 DedicatedIo or DedicatedLp or NotUsed
 	u16	xSlicLogicalProcIndex;	// 0x02 Logical Proc Index for correlation
@@ -76,17 +76,17 @@ struct ItLpQueue
 
 extern struct ItLpQueue xItLpQueue;
 
-extern struct HvLpEvent * ItLpQueue_getNextLpEvent( struct ItLpQueue * );
-extern int ItLpQueue_isLpIntPending( struct ItLpQueue * ); 
-extern unsigned ItLpQueue_process( struct ItLpQueue *, struct pt_regs * );
-extern void ItLpQueue_clearValid( struct HvLpEvent * );
+extern struct HvLpEvent * ItLpQueue_getNextLpEvent(struct ItLpQueue *);
+extern int ItLpQueue_isLpIntPending(struct ItLpQueue *);
+extern unsigned ItLpQueue_process(struct ItLpQueue *, struct pt_regs *);
+extern void ItLpQueue_clearValid(struct HvLpEvent *);
 
-static __inline__ void process_iSeries_events( void )
+static __inline__ void process_iSeries_events(void)
 {
 	__asm__ __volatile__ (
 	"	li	0,0x5555	\n\
 		sc"
-	: : : "r0", "r3" );	
+	: : : "r0", "r3");
 }
 
 #endif /* _ITLPQUEUE_H */
diff --git a/include/asm-ppc64/iSeries/ItLpRegSave.h b/include/asm-ppc64/iSeries/ItLpRegSave.h
index dafc4c81378..1b3087e7620 100644
--- a/include/asm-ppc64/iSeries/ItLpRegSave.h
+++ b/include/asm-ppc64/iSeries/ItLpRegSave.h
@@ -1,17 +1,17 @@
 /*
  * ItLpRegSave.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -19,33 +19,30 @@
 #ifndef _ITLPREGSAVE_H
 #define _ITLPREGSAVE_H
 
-//=====================================================================================
-//
-//	This control block contains the data that is shared between PLIC
-//	and the OS
-//    
-//
+/*
+ * This control block contains the data that is shared between PLIC
+ * and the OS
+ */
 
-struct ItLpRegSave
-{
+struct ItLpRegSave {
 	u32	xDesc;		// Eye catcher  "LpRS" ebcdic	000-003
 	u16	xSize;		// Size of this class		004-005
 	u8	xInUse;         // Area is live                 006-007
-	u8	xRsvd1[9]; 	// Reserved			007-00F
+	u8	xRsvd1[9];	// Reserved			007-00F
 
-	u8      xFixedRegSave[352]; // Fixed Register Save Area 010-16F 
+	u8      xFixedRegSave[352]; // Fixed Register Save Area 010-16F
 	u32	xCTRL;		// Control Register		170-173
-	u32	xDEC;		// Decrementer			174-177    
+	u32	xDEC;		// Decrementer			174-177
 	u32	xFPSCR;		// FP Status and Control Reg	178-17B
 	u32	xPVR;		// Processor Version Number	17C-17F
-    
+
 	u64	xMMCR0;		// Monitor Mode Control Reg 0	180-187
 	u32	xPMC1;		// Perf Monitor Counter 1	188-18B
 	u32	xPMC2;		// Perf Monitor Counter 2	18C-18F
 	u32	xPMC3;		// Perf Monitor Counter 3	190-193
 	u32	xPMC4;		// Perf Monitor Counter 4	194-197
 	u32	xPIR;		// Processor ID Reg		198-19B
-    
+
 	u32	xMMCR1;		// Monitor Mode Control Reg 1	19C-19F
 	u32	xMMCRA;		// Monitor Mode Control Reg A	1A0-1A3
 	u32	xPMC5;		// Perf Monitor Counter 5	1A4-1A7
@@ -57,17 +54,17 @@ struct ItLpRegSave
 	u32	xRsvd;          // Reserved                     1BC-1BF
 
 	u64	xACCR;		// Address Compare Control Reg	1C0-1C7
-	u64	xIMR;		// Instruction Match Register	1C8-1CF    
-	u64	xSDR1;		// Storage Description Reg 1	1D0-1D7    
+	u64	xIMR;		// Instruction Match Register	1C8-1CF
+	u64	xSDR1;		// Storage Description Reg 1	1D0-1D7
 	u64	xSPRG0;		// Special Purpose Reg General0	1D8-1DF
 	u64	xSPRG1;		// Special Purpose Reg General1	1E0-1E7
 	u64	xSPRG2;		// Special Purpose Reg General2	1E8-1EF
 	u64	xSPRG3;		// Special Purpose Reg General3	1F0-1F7
 	u64	xTB;		// Time Base Register		1F8-1FF
-   
+
 	u64	xFPR[32];	// Floating Point Registers	200-2FF
 
-	u64	xMSR;		// Machine State Register  	300-307
+	u64	xMSR;		// Machine State Register	300-307
 	u64	xNIA;		// Next Instruction Address	308-30F
 
 	u64	xDABR;		// Data Address Breakpoint Reg	310-317
@@ -76,8 +73,8 @@ struct ItLpRegSave
 	u64	xHID0;		// HW Implementation Dependent0	320-327
 
 	u64	xHID4;		// HW Implementation Dependent4	328-32F
-	u64	xSCOMd;		// SCON Data Reg (SPRG4)       	330-337
-	u64	xSCOMc;		// SCON Command Reg (SPRG5)    	338-33F
+	u64	xSCOMd;		// SCON Data Reg (SPRG4)	330-337
+	u64	xSCOMc;		// SCON Command Reg (SPRG5)	338-33F
 	u64	xSDAR;		// Sample Data Address Register	340-347
 	u64	xSIAR;		// Sample Inst Address Register	348-34F
 
diff --git a/include/asm-ppc64/iSeries/ItSpCommArea.h b/include/asm-ppc64/iSeries/ItSpCommArea.h
index f1b56f9e11e..5535f8271c9 100644
--- a/include/asm-ppc64/iSeries/ItSpCommArea.h
+++ b/include/asm-ppc64/iSeries/ItSpCommArea.h
@@ -1,29 +1,27 @@
 /*
  * ItSpCommArea.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
  */
 
-
 #ifndef _ITSPCOMMAREA_H
 #define _ITSPCOMMAREA_H
 
 
-struct SpCommArea
-{
+struct SpCommArea {
 	u32	xDesc;			// Descriptor (only in new formats)	000-003
 	u8	xFormat;		// Format (only in new formats)		004-004
 	u8	xRsvd1[11];		// Reserved				005-00F
diff --git a/include/asm-ppc64/iSeries/ItVpdAreas.h b/include/asm-ppc64/iSeries/ItVpdAreas.h
index d120439f30a..cfb9925e40a 100644
--- a/include/asm-ppc64/iSeries/ItVpdAreas.h
+++ b/include/asm-ppc64/iSeries/ItVpdAreas.h
@@ -1,17 +1,17 @@
 /*
  * ItVpdAreas.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -19,78 +19,75 @@
 #ifndef _ITVPDAREAS_H
 #define _ITVPDAREAS_H
 
-//=====================================================================================
-//
-//      This file defines the address and length of all of the VPD area passed to
-//	the OS from PLIC (most of which start from the SP).
-//
+/*
+ * This file defines the address and length of all of the VPD area passed to
+ * the OS from PLIC (most of which start from the SP).
+ */
 
 #include <asm/types.h>
 
-// VPD Entry index is carved in stone - cannot be changed (easily).
-#define ItVpdCecVpd				   0
-#define ItVpdDynamicSpace			   1
-#define ItVpdExtVpd				   2
-#define ItVpdExtVpdOnPanel			   3
-#define ItVpdFirstPaca				   4
-#define ItVpdIoVpd				   5
-#define ItVpdIplParms				   6
-#define ItVpdMsVpd				   7
-#define ItVpdPanelVpd				   8
-#define ItVpdLpNaca				   9
-#define ItVpdBackplaneAndMaybeClockCardVpd        10 
-#define ItVpdRecoveryLogBuffer		          11
-#define ItVpdSpCommArea				  12
-#define ItVpdSpLogBuffer			  13
-#define ItVpdSpLogBufferSave			  14
-#define ItVpdSpCardVpd				  15
-#define ItVpdFirstProcVpd			  16	
-#define ItVpdApModelVpd				  17
-#define ItVpdClockCardVpd			  18
-#define ItVpdBusExtCardVpd			  19
-#define ItVpdProcCapacityVpd			  20
-#define ItVpdInteractiveCapacityVpd		  21
-#define ItVpdFirstSlotLabel			  22
-#define ItVpdFirstLpQueue			  23
-#define ItVpdFirstL3CacheVpd			  24	
-#define ItVpdFirstProcFruVpd			  25
-
-#define ItVpdMaxEntries				  26
+/* VPD Entry index is carved in stone - cannot be changed (easily). */
+#define ItVpdCecVpd				0
+#define ItVpdDynamicSpace			1
+#define ItVpdExtVpd				2
+#define ItVpdExtVpdOnPanel			3
+#define ItVpdFirstPaca				4
+#define ItVpdIoVpd				5
+#define ItVpdIplParms				6
+#define ItVpdMsVpd				7
+#define ItVpdPanelVpd				8
+#define ItVpdLpNaca				9
+#define ItVpdBackplaneAndMaybeClockCardVpd	10
+#define ItVpdRecoveryLogBuffer			11
+#define ItVpdSpCommArea				12
+#define ItVpdSpLogBuffer			13
+#define ItVpdSpLogBufferSave			14
+#define ItVpdSpCardVpd				15
+#define ItVpdFirstProcVpd			16
+#define ItVpdApModelVpd				17
+#define ItVpdClockCardVpd			18
+#define ItVpdBusExtCardVpd			19
+#define ItVpdProcCapacityVpd			20
+#define ItVpdInteractiveCapacityVpd		21
+#define ItVpdFirstSlotLabel			22
+#define ItVpdFirstLpQueue			23
+#define ItVpdFirstL3CacheVpd			24
+#define ItVpdFirstProcFruVpd			25
 
+#define ItVpdMaxEntries				26
 
-#define ItDmaMaxEntries				  10
+#define ItDmaMaxEntries				10
 
-#define ItVpdAreasMaxSlotLabels		 192 
+#define ItVpdAreasMaxSlotLabels			192
 
 
 struct SlicVpdAdrs {
 	u32	pad1;
-	void *  vpdAddr;
+	void	*vpdAddr;
 };
 
 
-struct	ItVpdAreas
-{
-	u32	xSlicDesc;		// Descriptor				000-003
-	u16	xSlicSize;		// Size of this control block		004-005
-	u16	xPlicAdjustVpdLens:1;	// Flag to indicate new interface 	006-007
-	u16	xRsvd1:15;		// Reserved bits			...
-	u16	xSlicVpdEntries;	// Number of VPD entries		008-009
-	u16	xSlicDmaEntries;	// Number of DMA entries		00A-00B
-	u16	xSlicMaxLogicalProcs;	// Maximum logical processors		00C-00D
-	u16	xSlicMaxPhysicalProcs;	// Maximum physical processors		00E-00F
-	u16	xSlicDmaToksOffset;	// Offset into this of array		010-011
-	u16	xSlicVpdAdrsOffset;	// Offset into this of array		012-013
-	u16	xSlicDmaLensOffset;	// Offset into this of array		014-015
-	u16	xSlicVpdLensOffset;	// Offset into this of array		016-017
-	u16	xSlicMaxSlotLabels;	// Maximum number of slot labels	018-019
-	u16	xSlicMaxLpQueues;	// Maximum number of LP Queues		01A-01B
-	u8	xRsvd2[4];		// Reserved				01C-01F
-	u64	xRsvd3[12];		// Reserved				020-07F
-	u32	xPlicDmaLens[ItDmaMaxEntries];// Array of DMA lengths		080-0A7
-	u32	xPlicDmaToks[ItDmaMaxEntries];// Array of DMA tokens		0A8-0CF
-	u32	xSlicVpdLens[ItVpdMaxEntries];// Array of VPD lengths		0D0-12F
-	void * xSlicVpdAdrs[ItVpdMaxEntries];// Array of VPD buffers	130-1EF
+struct ItVpdAreas {
+	u32	xSlicDesc;		// Descriptor			000-003
+	u16	xSlicSize;		// Size of this control block	004-005
+	u16	xPlicAdjustVpdLens:1;	// Flag to indicate new interface006-007
+	u16	xRsvd1:15;		// Reserved bits		...
+	u16	xSlicVpdEntries;	// Number of VPD entries	008-009
+	u16	xSlicDmaEntries;	// Number of DMA entries	00A-00B
+	u16	xSlicMaxLogicalProcs;	// Maximum logical processors	00C-00D
+	u16	xSlicMaxPhysicalProcs;	// Maximum physical processors	00E-00F
+	u16	xSlicDmaToksOffset;	// Offset into this of array	010-011
+	u16	xSlicVpdAdrsOffset;	// Offset into this of array	012-013
+	u16	xSlicDmaLensOffset;	// Offset into this of array	014-015
+	u16	xSlicVpdLensOffset;	// Offset into this of array	016-017
+	u16	xSlicMaxSlotLabels;	// Maximum number of slot labels018-019
+	u16	xSlicMaxLpQueues;	// Maximum number of LP Queues	01A-01B
+	u8	xRsvd2[4];		// Reserved			01C-01F
+	u64	xRsvd3[12];		// Reserved			020-07F
+	u32	xPlicDmaLens[ItDmaMaxEntries];// Array of DMA lengths	080-0A7
+	u32	xPlicDmaToks[ItDmaMaxEntries];// Array of DMA tokens	0A8-0CF
+	u32	xSlicVpdLens[ItVpdMaxEntries];// Array of VPD lengths	0D0-12F
+	void	*xSlicVpdAdrs[ItVpdMaxEntries];// Array of VPD buffers	130-1EF
 };
 
 #endif /* _ITVPDAREAS_H */
diff --git a/include/asm-ppc64/iSeries/LparData.h b/include/asm-ppc64/iSeries/LparData.h
index e54f3b686e5..a9fd6e287e6 100644
--- a/include/asm-ppc64/iSeries/LparData.h
+++ b/include/asm-ppc64/iSeries/LparData.h
@@ -1,17 +1,17 @@
 /*
  * LparData.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -34,16 +34,15 @@
 #include <asm/iSeries/ItLpQueue.h>
 #include <asm/iSeries/IoHriProcessorVpd.h>
 
-extern struct LparMap	xLparMap;
-extern struct HvReleaseData hvReleaseData;
-extern struct ItLpNaca	itLpNaca;
-extern struct ItIplParmsReal xItIplParmsReal;
-extern struct ItExtVpdPanel xItExtVpdPanel;
-extern struct IoHriProcessorVpd xIoHriProcessorVpd[];
-extern struct ItLpQueue xItLpQueue;
-extern struct ItVpdAreas itVpdAreas;
-extern u64    xMsVpd[];
-extern struct msChunks msChunks;
-
+extern struct LparMap		xLparMap;
+extern struct HvReleaseData	hvReleaseData;
+extern struct ItLpNaca		itLpNaca;
+extern struct ItIplParmsReal	xItIplParmsReal;
+extern struct ItExtVpdPanel	xItExtVpdPanel;
+extern struct IoHriProcessorVpd	xIoHriProcessorVpd[];
+extern struct ItLpQueue		xItLpQueue;
+extern struct ItVpdAreas	itVpdAreas;
+extern u64			xMsVpd[];
+extern struct msChunks		msChunks;
 
 #endif /* _LPARDATA_H */
diff --git a/include/asm-ppc64/iSeries/LparMap.h b/include/asm-ppc64/iSeries/LparMap.h
index 075205bb0f4..1505c4f0f1b 100644
--- a/include/asm-ppc64/iSeries/LparMap.h
+++ b/include/asm-ppc64/iSeries/LparMap.h
@@ -1,17 +1,17 @@
 /*
  * LparMap.h
  * Copyright (C) 2001  Mike Corrigan IBM Corporation
- * 
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
@@ -21,13 +21,14 @@
 
 #include <asm/types.h>
 
-/* The iSeries hypervisor will set up mapping for one or more 
+/*
+ * The iSeries hypervisor will set up mapping for one or more
  * ESID/VSID pairs (in SLB/segment registers) and will set up
  * mappings of one or more ranges of pages to VAs.
  * We will have the hypervisor set up the ESID->VSID mapping
  * for the four kernel segments (C-F).  With shared processors,
  * the hypervisor will clear all segment registers and reload
- * these four whenever the processor is switched from one 
+ * these four whenever the processor is switched from one
  * partition to another.
  */
 
@@ -38,30 +39,29 @@
  * need to be located within the load area (if the total partition size
  * is 64 MB), but cannot be mapped.  Typically, this should specify
  * to map half (32 MB) of the load area.
- * 
- * The hypervisor will set up page table entries for the number of 
+ *
+ * The hypervisor will set up page table entries for the number of
  * pages specified.
  *
  * In 32-bit mode, the hypervisor will load all four of the
- * segment registers (identified by the low-order four bits of the 
+ * segment registers (identified by the low-order four bits of the
  * Esid field.  In 64-bit mode, the hypervisor will load one SLB
  * entry to map the Esid to the Vsid.
 */
 
-// Hypervisor initially maps 32MB of the load area 
-#define HvPagesToMap 8192
+/* Hypervisor initially maps 32MB of the load area */
+#define HvPagesToMap	8192
 
-struct LparMap
-{
-	u64	  xNumberEsids;		// Number of ESID/VSID pairs (1)
-	u64	  xNumberRanges;	// Number of VA ranges to map (1)
-	u64	  xSegmentTableOffs;	// Page number within load area of seg table (0)
-	u64	  xRsvd[5];		// Reserved (0)
-	u64	  xKernelEsid;  	// Esid used to map kernel load (0x0C00000000)	
-	u64	  xKernelVsid;		// Vsid used to map kernel load (0x0C00000000)
-	u64	  xPages;		// Number of pages to be mapped	(8192)
-	u64	  xOffset;		// Offset from start of load area (0)
-	u64	  xVPN;			// Virtual Page Number (0x000C000000000000)
+struct LparMap {
+	u64	xNumberEsids;	// Number of ESID/VSID pairs (1)
+	u64	xNumberRanges;	// Number of VA ranges to map (1)
+	u64	xSegmentTableOffs; // Page number within load area of seg table (0)
+	u64	xRsvd[5];
+	u64	xKernelEsid;	// Esid used to map kernel load (0x0C00000000)
+	u64	xKernelVsid;	// Vsid used to map kernel load (0x0C00000000)
+	u64	xPages;		// Number of pages to be mapped	(8192)
+	u64	xOffset;	// Offset from start of load area (0)
+	u64	xVPN;		// Virtual Page Number (0x000C000000000000)
 };
 
 #endif /* _LPARMAP_H */
diff --git a/include/asm-ppc64/iSeries/XmPciLpEvent.h b/include/asm-ppc64/iSeries/XmPciLpEvent.h
index a3d27f116e4..9c800792bc5 100644
--- a/include/asm-ppc64/iSeries/XmPciLpEvent.h
+++ b/include/asm-ppc64/iSeries/XmPciLpEvent.h
@@ -1,8 +1,6 @@
-
 #ifndef	__XMPCILPEVENT_H__
 #define	__XMPCILPEVENT_H__
 
-
 #ifdef __cplusplus
 extern "C" {
 #endif
@@ -10,7 +8,6 @@ extern "C" {
 int XmPciLpEvent_init(void);
 void ppc_irq_dispatch_handler(struct pt_regs *regs, int irq);
 
-
 #ifdef __cplusplus
 }
 #endif
diff --git a/include/asm-ppc64/iSeries/iSeries_io.h b/include/asm-ppc64/iSeries/iSeries_io.h
index f52b7599b37..9f79413342b 100644
--- a/include/asm-ppc64/iSeries/iSeries_io.h
+++ b/include/asm-ppc64/iSeries/iSeries_io.h
@@ -5,32 +5,33 @@
 
 #ifdef CONFIG_PPC_ISERIES
 #include <linux/types.h>
-/************************************************************************/
-/* File iSeries_io.h created by Allan Trautman on Thu Dec 28 2000.      */
-/************************************************************************/
-/* Remaps the io.h for the iSeries Io                                   */
-/* Copyright (C) 20yy  Allan H Trautman, IBM Corporation                */
-/*                                                                      */
-/* This program is free software; you can redistribute it and/or modify */
-/* it under the terms of the GNU General Public License as published by */
-/* the Free Software Foundation; either version 2 of the License, or    */
-/* (at your option) any later version.                                  */
-/*                                                                      */
-/* This program is distributed in the hope that it will be useful,      */ 
-/* but WITHOUT ANY WARRANTY; without even the implied warranty of       */
-/* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the        */
-/* GNU General Public License for more details.                         */
-/*                                                                      */
-/* You should have received a copy of the GNU General Public License    */ 
-/* along with this program; if not, write to the:                       */
-/* Free Software Foundation, Inc.,                                      */ 
-/* 59 Temple Place, Suite 330,                                          */ 
-/* Boston, MA  02111-1307  USA                                          */
-/************************************************************************/
-/* Change Activity:                                                     */
-/*   Created December 28, 2000                                          */
-/* End Change Activity                                                  */
-/************************************************************************/
+/*
+ * File iSeries_io.h created by Allan Trautman on Thu Dec 28 2000.
+ *
+ * Remaps the io.h for the iSeries Io
+ * Copyright (C) 2000  Allan H Trautman, IBM Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the:
+ * Free Software Foundation, Inc.,
+ * 59 Temple Place, Suite 330,
+ * Boston, MA  02111-1307  USA
+ *
+ * Change Activity:
+ *   Created December 28, 2000
+ * End Change Activity
+ */
+
 extern u8   iSeries_Read_Byte(const volatile void __iomem * IoAddress);
 extern u16  iSeries_Read_Word(const volatile void __iomem * IoAddress);
 extern u32  iSeries_Read_Long(const volatile void __iomem * IoAddress);
@@ -39,8 +40,10 @@ extern void iSeries_Write_Word(u16 IoData, volatile void __iomem * IoAddress);
 extern void iSeries_Write_Long(u32 IoData, volatile void __iomem * IoAddress);
 
 extern void iSeries_memset_io(volatile void __iomem *dest, char x, size_t n);
-extern void iSeries_memcpy_toio(volatile void __iomem *dest, void *source, size_t n);
-extern void iSeries_memcpy_fromio(void *dest, const volatile void __iomem *source, size_t n);
+extern void iSeries_memcpy_toio(volatile void __iomem *dest, void *source,
+		size_t n);
+extern void iSeries_memcpy_fromio(void *dest,
+		const volatile void __iomem *source, size_t n);
 
 #endif /* CONFIG_PPC_ISERIES */
 #endif /* _ISERIES_IO_H */
diff --git a/include/asm-ppc64/iSeries/iSeries_pci.h b/include/asm-ppc64/iSeries/iSeries_pci.h
index 5769cff4c00..9c6d64526d5 100644
--- a/include/asm-ppc64/iSeries/iSeries_pci.h
+++ b/include/asm-ppc64/iSeries/iSeries_pci.h
@@ -1,112 +1,113 @@
 #ifndef _ISERIES_64_PCI_H
 #define _ISERIES_64_PCI_H
 
-/************************************************************************/
-/* File iSeries_pci.h created by Allan Trautman on Tue Feb 20, 2001.    */
-/************************************************************************/
-/* Define some useful macros for the iSeries pci routines.              */
-/* Copyright (C) 2001  Allan H Trautman, IBM Corporation                */
-/*                                                                      */
-/* This program is free software; you can redistribute it and/or modify */
-/* it under the terms of the GNU General Public License as published by */
-/* the Free Software Foundation; either version 2 of the License, or    */
-/* (at your option) any later version.                                  */
-/*                                                                      */
-/* This program is distributed in the hope that it will be useful,      */ 
-/* but WITHOUT ANY WARRANTY; without even the implied warranty of       */
-/* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the        */
-/* GNU General Public License for more details.                         */
-/*                                                                      */
-/* You should have received a copy of the GNU General Public License    */ 
-/* along with this program; if not, write to the:                       */
-/* Free Software Foundation, Inc.,                                      */ 
-/* 59 Temple Place, Suite 330,                                          */ 
-/* Boston, MA  02111-1307  USA                                          */
-/************************************************************************/
-/* Change Activity:                                                     */
-/*   Created Feb 20, 2001                                               */
-/*   Added device reset, March 22, 2001                                 */
-/*   Ported to ppc64, May 25, 2001                                      */
-/* End Change Activity                                                  */
-/************************************************************************/
+/*
+ * File iSeries_pci.h created by Allan Trautman on Tue Feb 20, 2001.
+ *
+ * Define some useful macros for the iSeries pci routines.
+ * Copyright (C) 2001  Allan H Trautman, IBM Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the:
+ * Free Software Foundation, Inc.,
+ * 59 Temple Place, Suite 330,
+ * Boston, MA  02111-1307  USA
+ *
+ * Change Activity:
+ *   Created Feb 20, 2001
+ *   Added device reset, March 22, 2001
+ *   Ported to ppc64, May 25, 2001
+ * End Change Activity
+ */
 
 #include <asm/iSeries/HvCallPci.h>
 #include <asm/abs_addr.h>
 
-struct pci_dev;				/* For Forward Reference        */
+struct pci_dev;				/* For Forward Reference */
 struct iSeries_Device_Node;
 
-/************************************************************************/
-/* Gets iSeries Bus, SubBus, DevFn using iSeries_Device_Node structure */
-/************************************************************************/
+/*
+ * Gets iSeries Bus, SubBus, DevFn using iSeries_Device_Node structure
+ */
 
 #define ISERIES_BUS(DevPtr)	DevPtr->DsaAddr.Dsa.busNumber
 #define ISERIES_SUBBUS(DevPtr)	DevPtr->DsaAddr.Dsa.subBusNumber
 #define ISERIES_DEVICE(DevPtr)	DevPtr->DsaAddr.Dsa.deviceId
 #define ISERIES_DSA(DevPtr)	DevPtr->DsaAddr.DsaAddr
 #define ISERIES_DEVFUN(DevPtr)	DevPtr->DevFn
-#define ISERIES_DEVNODE(PciDev) ((struct iSeries_Device_Node*)PciDev->sysdata)
+#define ISERIES_DEVNODE(PciDev)	((struct iSeries_Device_Node*)PciDev->sysdata)
 
 #define EADsMaxAgents 7
 
-/************************************************************************/
-/* Decodes Linux DevFn to iSeries DevFn, bridge device, or function.    */
-/* For Linux, see PCI_SLOT and PCI_FUNC in include/linux/pci.h          */
-/************************************************************************/
+/*
+ * Decodes Linux DevFn to iSeries DevFn, bridge device, or function.
+ * For Linux, see PCI_SLOT and PCI_FUNC in include/linux/pci.h
+ */
 
-#define ISERIES_PCI_AGENTID(idsel,func)	((idsel & 0x0F) << 4) | (func  & 0x07)
-#define ISERIES_ENCODE_DEVICE(agentid)	((0x10) | ((agentid&0x20)>>2) | (agentid&07))
+#define ISERIES_PCI_AGENTID(idsel, func)	\
+	((idsel & 0x0F) << 4) | (func & 0x07)
+#define ISERIES_ENCODE_DEVICE(agentid)		\
+	((0x10) | ((agentid & 0x20) >> 2) | (agentid & 0x07))
 
-#define ISERIES_GET_DEVICE_FROM_SUBBUS(subbus)   ((subbus >> 5) & 0x7)
-#define ISERIES_GET_FUNCTION_FROM_SUBBUS(subbus) ((subbus >> 2) & 0x7)
+#define ISERIES_GET_DEVICE_FROM_SUBBUS(subbus)		((subbus >> 5) & 0x7)
+#define ISERIES_GET_FUNCTION_FROM_SUBBUS(subbus)	((subbus >> 2) & 0x7)
 
 /*
  * N.B. the ISERIES_DECODE_* macros are not used anywhere, and I think
  * the 0x71 (at least) must be wrong - 0x78 maybe?  -- paulus.
  */
-#define ISERIES_DECODE_DEVFN(linuxdevfn)  (((linuxdevfn & 0x71) << 1) | (linuxdevfn & 0x07))
-#define ISERIES_DECODE_DEVICE(linuxdevfn) (((linuxdevfn & 0x38) >> 3) |(((linuxdevfn & 0x40) >> 2) + 0x10))
-#define ISERIES_DECODE_FUNCTION(linuxdevfn) (linuxdevfn & 0x07)
-
-/************************************************************************/
-/* Converts Virtual Address to Real Address for Hypervisor calls        */
-/************************************************************************/
+#define ISERIES_DECODE_DEVFN(linuxdevfn)	\
+	(((linuxdevfn & 0x71) << 1) | (linuxdevfn & 0x07))
+#define ISERIES_DECODE_DEVICE(linuxdevfn)	\
+	(((linuxdevfn & 0x38) >> 3) | (((linuxdevfn & 0x40) >> 2) + 0x10))
+#define ISERIES_DECODE_FUNCTION(linuxdevfn)	\
+	(linuxdevfn & 0x07)
 
-#define ISERIES_HV_ADDR(virtaddr)  (0x8000000000000000 | virt_to_abs(virtaddr))
-
-/************************************************************************/
-/* iSeries Device Information                                           */
-/************************************************************************/
+/*
+ * Converts Virtual Address to Real Address for Hypervisor calls
+ */
+#define ISERIES_HV_ADDR(virtaddr)	\
+	(0x8000000000000000 | virt_to_abs(virtaddr))
 
+/*
+ * iSeries Device Information
+ */
 struct iSeries_Device_Node {
 	struct list_head Device_List;
-	struct pci_dev* PciDev;         /* Pointer to pci_dev structure*/
-        union HvDsaMap	DsaAddr;	/* Direct Select Address       */
-                                        /* busNumber,subBusNumber,     */ 
-	                                /* deviceId, barNumber         */
-	HvAgentId       AgentId;	/* Hypervisor DevFn            */
-	int             DevFn;          /* Linux devfn                 */
-	int             BarOffset;
-	int             Irq;            /* Assigned IRQ                */
-	int             ReturnCode;	/* Return Code Holder          */
-	int             IoRetry;        /* Current Retry Count         */
-	int             Flags;          /* Possible flags(disable/bist)*/
-	u16             Vendor;         /* Vendor ID                   */
-	u8              LogicalSlot;    /* Hv Slot Index for Tces      */
-	struct iommu_table* iommu_table;/* Device TCE Table            */ 
-	u8              PhbId;          /* Phb Card is on.             */
-	u16             Board;          /* Board Number                */
-	u8              FrameId;	/* iSeries spcn Frame Id       */
-	char            CardLocation[4];/* Char format of planar vpd   */
-	char            Location[20];   /* Frame  1, Card C10          */
+	struct pci_dev	*PciDev;
+	union HvDsaMap	DsaAddr;	/* Direct Select Address */
+					/* busNumber, subBusNumber, */
+					/* deviceId, barNumber */
+	HvAgentId	AgentId;	/* Hypervisor DevFn */
+	int		DevFn;		/* Linux devfn */
+	int		BarOffset;
+	int		Irq;		/* Assigned IRQ */
+	int		ReturnCode;	/* Return Code Holder */
+	int		IoRetry;	/* Current Retry Count */
+	int		Flags;		/* Possible flags(disable/bist)*/
+	u16		Vendor;		/* Vendor ID */
+	u8		LogicalSlot;	/* Hv Slot Index for Tces */
+	struct iommu_table *iommu_table;/* Device TCE Table */
+	u8		PhbId;		/* Phb Card is on. */
+	u16		Board;		/* Board Number */
+	u8		FrameId;	/* iSeries spcn Frame Id */
+	char		CardLocation[4];/* Char format of planar vpd */
+	char		Location[20];	/* Frame	1, Card C10 */
 };
 
-/************************************************************************/
-/* Functions                                                            */
-/************************************************************************/
-
-extern int           iSeries_Device_Information(struct pci_dev*,char*, int);
-extern void          iSeries_Get_Location_Code(struct iSeries_Device_Node*);
-extern int           iSeries_Device_ToggleReset(struct pci_dev* PciDev, int AssertTime, int DelayTime);
+extern int	iSeries_Device_Information(struct pci_dev*, char*, int);
+extern void	iSeries_Get_Location_Code(struct iSeries_Device_Node*);
+extern int	iSeries_Device_ToggleReset(struct pci_dev* PciDev,
+		int AssertTime, int DelayTime);
 
 #endif /* _ISERIES_64_PCI_H */
diff --git a/include/asm-ppc64/iSeries/mf.h b/include/asm-ppc64/iSeries/mf.h
index db333e1ee21..7e6a0d93699 100644
--- a/include/asm-ppc64/iSeries/mf.h
+++ b/include/asm-ppc64/iSeries/mf.h
@@ -9,17 +9,16 @@
  * all partitions in the iSeries.  It also provides miscellaneous low-level
  * machine facility type operations.
  *
- * 
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
- * 
+ *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- * 
+ *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
diff --git a/include/asm-ppc64/iSeries/vio.h b/include/asm-ppc64/iSeries/vio.h
index 3e5766a849d..614088f61b8 100644
--- a/include/asm-ppc64/iSeries/vio.h
+++ b/include/asm-ppc64/iSeries/vio.h
@@ -8,32 +8,32 @@
  *           Colin Devilbiss <devilbis@us.ibm.com>
  *
  * (C) Copyright 2000 IBM Corporation
- * 
+ *
  * This header file is used by the iSeries virtual I/O device
  * drivers.  It defines the interfaces to the common functions
  * (implemented in drivers/char/viopath.h) as well as defining
- * common functions and structures.  Currently (at the time I 
+ * common functions and structures.  Currently (at the time I
  * wrote this comment) the iSeries virtual I/O device drivers
- * that use this are 
- *   drivers/block/viodasd.c 
+ * that use this are
+ *   drivers/block/viodasd.c
  *   drivers/char/viocons.c
  *   drivers/char/viotape.c
  *   drivers/cdrom/viocd.c
  *
  * The iSeries virtual ethernet support (veth.c) uses a whole
  * different set of functions.
- * 
+ *
  * This program is free software;  you can redistribute it and/or
  * modify it under the terms of the GNU General Public License as
  * published by the Free Software Foundation; either version 2 of the
  * License, or (at your option) anyu later version.
  *
  * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of 
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.  
+ * General Public License for more details.
  *
- * You should have received a copy of the GNU General Public License 
+ * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software Foundation,
  * Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
  *
@@ -44,14 +44,16 @@
 #include <asm/iSeries/HvTypes.h>
 #include <asm/iSeries/HvLpEvent.h>
 
-/* iSeries virtual I/O events use the subtype field in
+/*
+ * iSeries virtual I/O events use the subtype field in
  * HvLpEvent to figure out what kind of vio event is coming
  * in.  We use a table to route these, and this defines
  * the maximum number of distinct subtypes
  */
 #define VIO_MAX_SUBTYPES 8
 
-/* Each subtype can register a handler to process their events.
+/*
+ * Each subtype can register a handler to process their events.
  * The handler must have this interface.
  */
 typedef void (vio_event_handler_t) (struct HvLpEvent * event);
@@ -70,13 +72,13 @@ void vio_free_event_buffer(int subtype, void *buffer);
 extern HvLpIndex viopath_hostLp;
 extern HvLpIndex viopath_ourLp;
 
-#define VIOCHAR_MAX_DATA 200
+#define VIOCHAR_MAX_DATA	200
 
-#define VIOMAJOR_SUBTYPE_MASK 0xff00
-#define VIOMINOR_SUBTYPE_MASK 0x00ff
-#define VIOMAJOR_SUBTYPE_SHIFT 8
+#define VIOMAJOR_SUBTYPE_MASK	0xff00
+#define VIOMINOR_SUBTYPE_MASK	0x00ff
+#define VIOMAJOR_SUBTYPE_SHIFT	8
 
-#define VIOVERSION            0x0101
+#define VIOVERSION		0x0101
 
 /*
  * This is the general structure for VIO errors; each module should have
@@ -89,8 +91,8 @@ struct vio_error_entry {
 	int errno;
 	const char *msg;
 };
-const struct vio_error_entry *vio_lookup_rc(const struct vio_error_entry
-					    *local_table, u16 rc);
+extern const struct vio_error_entry *vio_lookup_rc(
+		const struct vio_error_entry *local_table, u16 rc);
 
 enum viosubtypes {
 	viomajorsubtype_monitor = 0x0100,
@@ -102,7 +104,6 @@ enum viosubtypes {
 	viomajorsubtype_scsi = 0x0700
 };
 
-
 enum vioconfigsubtype {
 	vioconfigget = 0x0001,
 };
-- 
cgit v1.2.3-70-g09d2


From 6b7feecb2f8fcab184a38916d10349bd6648e0bc Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:35 -0700
Subject: [PATCH] ppc64 iSeries: obvious code simplifications

This patch does some obvious code cleanups in the iSeries headers files.
	- simplifies the bodies of lots of inline functions
	- parenthesises a macros result
	- removes C++ wrapping
	- adds "extern" to some function declarations
There are no semantic changes.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-ppc64/iSeries/HvCallCfg.h    |  61 ++++---------
 include/asm-ppc64/iSeries/HvCallEvent.h  |  65 +++-----------
 include/asm-ppc64/iSeries/HvCallHpt.h    |  23 ++---
 include/asm-ppc64/iSeries/HvCallPci.h    | 146 ++++++-------------------------
 include/asm-ppc64/iSeries/HvCallSm.h     |   5 +-
 include/asm-ppc64/iSeries/HvCallXm.h     |  35 +++-----
 include/asm-ppc64/iSeries/HvLpConfig.h   |   3 +-
 include/asm-ppc64/iSeries/XmPciLpEvent.h |  12 +--
 include/asm-ppc64/iSeries/iSeries_irq.h  |  18 ++--
 include/asm-ppc64/iSeries/iSeries_pci.h  |   2 +-
 include/asm-ppc64/iSeries/vio.h          |  20 ++---
 11 files changed, 92 insertions(+), 298 deletions(-)

(limited to 'include')

diff --git a/include/asm-ppc64/iSeries/HvCallCfg.h b/include/asm-ppc64/iSeries/HvCallCfg.h
index 2223a9647d0..aa491c3ad0a 100644
--- a/include/asm-ppc64/iSeries/HvCallCfg.h
+++ b/include/asm-ppc64/iSeries/HvCallCfg.h
@@ -69,37 +69,27 @@ enum HvCallCfg_ReqQual {
 
 static inline HvLpIndex	HvCallCfg_getLps(void)
 {
-	HvLpIndex retVal = HvCall0(HvCallCfgGetLps);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall0(HvCallCfgGetLps);
 }
 
 static inline int HvCallCfg_isBusDedicated(u64 busIndex)
 {
-	int retVal = HvCall1(HvCallCfgIsBusDedicated,busIndex);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall1(HvCallCfgIsBusDedicated, busIndex);
 }
 
 static inline HvLpIndex	HvCallCfg_getBusOwner(u64 busIndex)
 {
-	HvLpIndex retVal = HvCall1(HvCallCfgGetBusOwner,busIndex);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall1(HvCallCfgGetBusOwner, busIndex);
 }
 
 static inline HvLpIndexMap HvCallCfg_getBusAllocation(u64 busIndex)
 {
-	HvLpIndexMap retVal = HvCall1(HvCallCfgGetBusAllocation,busIndex);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall1(HvCallCfgGetBusAllocation, busIndex);
 }
 
 static inline HvLpIndexMap HvCallCfg_getActiveLpMap(void)
 {
-	HvLpIndexMap retVal = HvCall0(HvCallCfgGetActiveLpMap);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall0(HvCallCfgGetActiveLpMap);
 }
 
 static inline HvLpVirtualLanIndexMap HvCallCfg_getVirtualLanIndexMap(
@@ -112,23 +102,18 @@ static inline HvLpVirtualLanIndexMap HvCallCfg_getVirtualLanIndexMap(
 	u64 retVal = HvCall1(HvCallCfgGetVirtualLanIndexMap, lp);
 	if (retVal == -1)
 		retVal = 0;
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return retVal;
 }
 
 static inline u64 HvCallCfg_getSystemMsChunks(void)
 {
-	u64 retVal = HvCall0(HvCallCfgGetSystemMsChunks);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall0(HvCallCfgGetSystemMsChunks);
 }
 
 static inline u64 HvCallCfg_getMsChunks(HvLpIndex lp,
 		enum HvCallCfg_ReqQual qual)
 {
-	u64 retVal = HvCall2(HvCallCfgGetMsChunks,lp,qual);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall2(HvCallCfgGetMsChunks, lp, qual);
 }
 
 static inline u64 HvCallCfg_getMinRuntimeMsChunks(HvLpIndex lp)
@@ -142,65 +127,51 @@ static inline u64 HvCallCfg_getMinRuntimeMsChunks(HvLpIndex lp)
 
 static inline u64 HvCallCfg_setMinRuntimeMsChunks(u64 chunks)
 {
-	u64 retVal = HvCall1(HvCallCfgSetMinRuntimeMsChunks,chunks);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall1(HvCallCfgSetMinRuntimeMsChunks, chunks);
 }
 
 static inline u64 HvCallCfg_getSystemPhysicalProcessors(void)
 {
-	u64 retVal = HvCall0(HvCallCfgGetSystemPhysicalProcessors);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall0(HvCallCfgGetSystemPhysicalProcessors);
 }
 
 static inline u64 HvCallCfg_getPhysicalProcessors(HvLpIndex lp,
 		enum HvCallCfg_ReqQual qual)
 {
-	u64 retVal = HvCall2(HvCallCfgGetPhysicalProcessors,lp,qual);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall2(HvCallCfgGetPhysicalProcessors, lp, qual);
 }
 
 static inline u64 HvCallCfg_getConfiguredBusUnitsForInterruptProc(HvLpIndex lp,
 		u16 hvLogicalProcIndex)
 {
-	u64 retVal = HvCall2(HvCallCfgGetConfiguredBusUnitsForIntProc,lp,hvLogicalProcIndex);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall2(HvCallCfgGetConfiguredBusUnitsForIntProc, lp,
+			hvLogicalProcIndex);
 
 }
 
 static inline HvLpSharedPoolIndex HvCallCfg_getSharedPoolIndex(HvLpIndex lp)
 {
-	HvLpSharedPoolIndex retVal =
-		HvCall1(HvCallCfgGetSharedPoolIndex,lp);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall1(HvCallCfgGetSharedPoolIndex, lp);
 
 }
 
 static inline u64 HvCallCfg_getSharedProcUnits(HvLpIndex lp,
 		enum HvCallCfg_ReqQual qual)
 {
-	u64 retVal = HvCall2(HvCallCfgGetSharedProcUnits,lp,qual);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall2(HvCallCfgGetSharedProcUnits, lp, qual);
 
 }
 
 static inline u64 HvCallCfg_getNumProcsInSharedPool(HvLpSharedPoolIndex sPI)
 {
-	u16 retVal = HvCall1(HvCallCfgGetNumProcsInSharedPool,sPI);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
+	u16 retVal = HvCall1(HvCallCfgGetNumProcsInSharedPool, sPI);
 	return retVal;
 
 }
 
 static inline HvLpIndex	HvCallCfg_getHostingLpIndex(HvLpIndex lp)
 {
-	u64 retVal = HvCall1(HvCallCfgGetHostingLpIndex,lp);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
+	u64 retVal = HvCall1(HvCallCfgGetHostingLpIndex, lp);
 	return retVal;
 }
 
diff --git a/include/asm-ppc64/iSeries/HvCallEvent.h b/include/asm-ppc64/iSeries/HvCallEvent.h
index a5954785512..5d9a327d012 100644
--- a/include/asm-ppc64/iSeries/HvCallEvent.h
+++ b/include/asm-ppc64/iSeries/HvCallEvent.h
@@ -82,13 +82,11 @@ typedef u64 HvLpDma_Rc;
 static inline void HvCallEvent_getOverflowLpEvents(u8 queueIndex)
 {
 	HvCall1(HvCallEventGetOverflowLpEvents, queueIndex);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
 
 static inline void HvCallEvent_setInterLpQueueIndex(u8 queueIndex)
 {
 	HvCall1(HvCallEventSetInterLpQueueIndex, queueIndex);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
 
 static inline void HvCallEvent_setLpEventStack(u8 queueIndex,
@@ -99,7 +97,6 @@ static inline void HvCallEvent_setLpEventStack(u8 queueIndex,
 	abs_addr = virt_to_abs(eventStackAddr);
 	HvCall3(HvCallEventSetLpEventStack, queueIndex, abs_addr,
 			eventStackSize);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
 
 static inline void HvCallEvent_setLpEventQueueInterruptProc(u8 queueIndex,
@@ -107,22 +104,18 @@ static inline void HvCallEvent_setLpEventQueueInterruptProc(u8 queueIndex,
 {
 	HvCall2(HvCallEventSetLpEventQueueInterruptProc, queueIndex,
 			lpLogicalProcIndex);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
 
 static inline HvLpEvent_Rc HvCallEvent_signalLpEvent(struct HvLpEvent *event)
 {
 	u64 abs_addr;
-	HvLpEvent_Rc retVal;
 
 #ifdef DEBUG_SENDEVENT
 	printk("HvCallEvent_signalLpEvent: *event = %016lx\n ",
 			(unsigned long)event);
 #endif
 	abs_addr = virt_to_abs(event);
-	retVal = (HvLpEvent_Rc)HvCall1(HvCallEventSignalLpEvent, abs_addr);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall1(HvCallEventSignalLpEvent, abs_addr);
 }
 
 static inline HvLpEvent_Rc HvCallEvent_signalLpEventFast(HvLpIndex targetLp,
@@ -132,8 +125,6 @@ static inline HvLpEvent_Rc HvCallEvent_signalLpEventFast(HvLpIndex targetLp,
 		u64 eventData1, u64 eventData2, u64 eventData3,
 		u64 eventData4, u64 eventData5)
 {
-	HvLpEvent_Rc retVal;
-
 	/* Pack the misc bits into a single Dword to pass to PLIC */
 	union {
 		struct HvCallEvent_PackedParms	parms;
@@ -148,67 +139,49 @@ static inline HvLpEvent_Rc HvCallEvent_signalLpEventFast(HvLpIndex targetLp,
 	packed.parms.xSourceInstId	= sourceInstanceId;
 	packed.parms.xTargetInstId	= targetInstanceId;
 
-	retVal = (HvLpEvent_Rc)HvCall7(HvCallEventSignalLpEventParms,
-			packed.dword, correlationToken, eventData1,eventData2,
-			eventData3,eventData4, eventData5);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall7(HvCallEventSignalLpEventParms, packed.dword,
+			correlationToken, eventData1, eventData2,
+			eventData3, eventData4, eventData5);
 }
 
 static inline HvLpEvent_Rc HvCallEvent_ackLpEvent(struct HvLpEvent *event)
 {
 	u64 abs_addr;
-	HvLpEvent_Rc retVal;
 
 	abs_addr = virt_to_abs(event);
-	retVal = (HvLpEvent_Rc)HvCall1(HvCallEventAckLpEvent, abs_addr);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall1(HvCallEventAckLpEvent, abs_addr);
 }
 
 static inline HvLpEvent_Rc HvCallEvent_cancelLpEvent(struct HvLpEvent *event)
 {
 	u64 abs_addr;
-	HvLpEvent_Rc retVal;
 
 	abs_addr = virt_to_abs(event);
-	retVal = (HvLpEvent_Rc)HvCall1(HvCallEventCancelLpEvent, abs_addr);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall1(HvCallEventCancelLpEvent, abs_addr);
 }
 
 static inline HvLpInstanceId HvCallEvent_getSourceLpInstanceId(
 		HvLpIndex targetLp, HvLpEvent_Type type)
 {
-	HvLpInstanceId retVal;	
-
-	retVal = HvCall2(HvCallEventGetSourceLpInstanceId, targetLp, type);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall2(HvCallEventGetSourceLpInstanceId, targetLp, type);
 }
 
 static inline HvLpInstanceId HvCallEvent_getTargetLpInstanceId(
 		HvLpIndex targetLp, HvLpEvent_Type type)
 {
-	HvLpInstanceId retVal;	
-
-	retVal = HvCall2(HvCallEventGetTargetLpInstanceId, targetLp, type);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall2(HvCallEventGetTargetLpInstanceId, targetLp, type);
 }
 
 static inline void HvCallEvent_openLpEventPath(HvLpIndex targetLp,
 		HvLpEvent_Type type)
 {
 	HvCall2(HvCallEventOpenLpEventPath, targetLp, type);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
 
 static inline void HvCallEvent_closeLpEventPath(HvLpIndex targetLp,
 		HvLpEvent_Type type)
 {
 	HvCall2(HvCallEventCloseLpEventPath, targetLp, type);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
 
 static inline HvLpDma_Rc HvCallEvent_dmaBufList(HvLpEvent_Type type,
@@ -220,7 +193,6 @@ static inline HvLpDma_Rc HvCallEvent_dmaBufList(HvLpEvent_Type type,
 		/* Do these need to be converted to absolute addresses? */
 		u64 localBufList, u64 remoteBufList, u32 transferLength)
 {
-	HvLpDma_Rc retVal;
 	/* Pack the misc bits into a single Dword to pass to PLIC */
 	union {
 		struct HvCallEvent_PackedDmaParms	parms;
@@ -237,11 +209,8 @@ static inline HvLpDma_Rc HvCallEvent_dmaBufList(HvLpEvent_Type type,
 	packed.parms.xLocalInstId	= localInstanceId;
 	packed.parms.xRemoteInstId	= remoteInstanceId;
 
-	retVal = (HvLpDma_Rc)HvCall4(HvCallEventDmaBufList,
-			packed.dword, localBufList, remoteBufList,
-			transferLength);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall4(HvCallEventDmaBufList, packed.dword, localBufList,
+			remoteBufList, transferLength);
 }
 
 static inline HvLpDma_Rc HvCallEvent_dmaSingle(HvLpEvent_Type type,
@@ -252,7 +221,6 @@ static inline HvLpDma_Rc HvCallEvent_dmaSingle(HvLpEvent_Type type,
 		HvLpDma_AddressType remoteAddressType,
 		u64 localAddrOrTce, u64 remoteAddrOrTce, u32 transferLength)
 {
-	HvLpDma_Rc retVal;
 	/* Pack the misc bits into a single Dword to pass to PLIC */
 	union {
 		struct HvCallEvent_PackedDmaParms	parms;
@@ -269,24 +237,17 @@ static inline HvLpDma_Rc HvCallEvent_dmaSingle(HvLpEvent_Type type,
 	packed.parms.xLocalInstId	= localInstanceId;
 	packed.parms.xRemoteInstId	= remoteInstanceId;
 
-	retVal = (HvLpDma_Rc)HvCall4(HvCallEventDmaSingle,
-			packed.dword, localAddrOrTce, remoteAddrOrTce,
-			transferLength);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return (HvLpDma_Rc)HvCall4(HvCallEventDmaSingle, packed.dword,
+			localAddrOrTce, remoteAddrOrTce, transferLength);
 }
 
 static inline HvLpDma_Rc HvCallEvent_dmaToSp(void *local, u32 remote,
 		u32 length, HvLpDma_Direction dir)
 {
 	u64 abs_addr;
-	HvLpDma_Rc retVal;
 
 	abs_addr = virt_to_abs(local);
-	retVal = (HvLpDma_Rc)HvCall4(HvCallEventDmaToSp, abs_addr, remote,
-			length, dir);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall4(HvCallEventDmaToSp, abs_addr, remote, length, dir);
 }
 
 #endif /* _HVCALLEVENT_H */
diff --git a/include/asm-ppc64/iSeries/HvCallHpt.h b/include/asm-ppc64/iSeries/HvCallHpt.h
index 827d26c60fc..66f38222ff7 100644
--- a/include/asm-ppc64/iSeries/HvCallHpt.h
+++ b/include/asm-ppc64/iSeries/HvCallHpt.h
@@ -43,34 +43,27 @@
 
 static inline u64 HvCallHpt_getHptAddress(void)
 {
-	u64 retval = HvCall0(HvCallHptGetHptAddress);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retval;
+	return HvCall0(HvCallHptGetHptAddress);
 }
 
 static inline u64 HvCallHpt_getHptPages(void)
 {
-	u64 retval = HvCall0(HvCallHptGetHptPages);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retval;
+	return HvCall0(HvCallHptGetHptPages);
 }
 
 static inline void HvCallHpt_setPp(u32 hpteIndex, u8 value)
 {
 	HvCall2(HvCallHptSetPp, hpteIndex, value);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
 
 static inline void HvCallHpt_setSwBits(u32 hpteIndex, u8 bitson, u8 bitsoff)
 {
 	HvCall3(HvCallHptSetSwBits, hpteIndex, bitson, bitsoff);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
 
 static inline void HvCallHpt_invalidateNoSyncICache(u32 hpteIndex)
 {
 	HvCall1(HvCallHptInvalidateNoSyncICache, hpteIndex);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
 
 static inline u64 HvCallHpt_invalidateSetSwBitsGet(u32 hpteIndex, u8 bitson,
@@ -81,36 +74,30 @@ static inline u64 HvCallHpt_invalidateSetSwBitsGet(u32 hpteIndex, u8 bitson,
 	compressedStatus = HvCall4(HvCallHptInvalidateSetSwBitsGet,
 			hpteIndex, bitson, bitsoff, 1);
 	HvCall1(HvCallHptInvalidateNoSyncICache, hpteIndex);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 	return compressedStatus;
 }
 
 static inline u64 HvCallHpt_findValid(HPTE *hpte, u64 vpn)
 {
-	u64 retIndex = HvCall3Ret16( HvCallHptFindValid, hpte, vpn, 0, 0 );
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retIndex;
+	return HvCall3Ret16(HvCallHptFindValid, hpte, vpn, 0, 0);
 }
 
 static inline u64 HvCallHpt_findNextValid(HPTE *hpte, u32 hpteIndex,
 		u8 bitson, u8 bitsoff)
 {
-	u64 retIndex = HvCall3Ret16( HvCallHptFindNextValid, hpte, hpteIndex, bitson, bitsoff );
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retIndex;
+	return HvCall3Ret16(HvCallHptFindNextValid, hpte, hpteIndex,
+			bitson, bitsoff);
 }
 
 static inline void HvCallHpt_get(HPTE *hpte, u32 hpteIndex)
 {
 	HvCall2Ret16(HvCallHptGet, hpte, hpteIndex, 0);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
 
 static inline void HvCallHpt_addValidate(u32 hpteIndex, u32 hBit, HPTE *hpte)
 {
 	HvCall4(HvCallHptAddValidate, hpteIndex, hBit, (*((u64 *)hpte)),
 			(*(((u64 *)hpte)+1)));
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
 
 #endif /* _HVCALLHPT_H */
diff --git a/include/asm-ppc64/iSeries/HvCallPci.h b/include/asm-ppc64/iSeries/HvCallPci.h
index 346ab5d39dc..c8d675c40f5 100644
--- a/include/asm-ppc64/iSeries/HvCallPci.h
+++ b/include/asm-ppc64/iSeries/HvCallPci.h
@@ -140,8 +140,6 @@ static inline u64 HvCallPci_configLoad8(u16 busNumber, u8 subBusNumber,
 
 	HvCall3Ret16(HvCallPciConfigLoad8, &retVal, *(u64 *)&dsa, offset, 0);
 
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
 	*value = retVal.value;
 
 	return retVal.rc;
@@ -161,8 +159,6 @@ static inline u64 HvCallPci_configLoad16(u16 busNumber, u8 subBusNumber,
 
 	HvCall3Ret16(HvCallPciConfigLoad16, &retVal, *(u64 *)&dsa, offset, 0);
 
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
 	*value = retVal.value;
 
 	return retVal.rc;
@@ -182,8 +178,6 @@ static inline u64 HvCallPci_configLoad32(u16 busNumber, u8 subBusNumber,
 
 	HvCall3Ret16(HvCallPciConfigLoad32, &retVal, *(u64 *)&dsa, offset, 0);
 
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
 	*value = retVal.value;
 
 	return retVal.rc;
@@ -193,7 +187,6 @@ static inline u64 HvCallPci_configStore8(u16 busNumber, u8 subBusNumber,
 		u8 deviceId, u32 offset, u8 value)
 {
 	struct HvCallPci_DsaAddr dsa;
-	u64 retVal;
 
 	*((u64*)&dsa) = 0;
 
@@ -201,18 +194,13 @@ static inline u64 HvCallPci_configStore8(u16 busNumber, u8 subBusNumber,
 	dsa.subBusNumber = subBusNumber;
 	dsa.deviceId = deviceId;
 
-	retVal = HvCall4(HvCallPciConfigStore8, *(u64 *)&dsa, offset, value, 0);
-
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
-	return retVal;
+	return HvCall4(HvCallPciConfigStore8, *(u64 *)&dsa, offset, value, 0);
 }
 
 static inline u64 HvCallPci_configStore16(u16 busNumber, u8 subBusNumber,
 		u8 deviceId, u32 offset, u16 value)
 {
 	struct HvCallPci_DsaAddr dsa;
-	u64 retVal;
 
 	*((u64*)&dsa) = 0;
 
@@ -220,18 +208,13 @@ static inline u64 HvCallPci_configStore16(u16 busNumber, u8 subBusNumber,
 	dsa.subBusNumber = subBusNumber;
 	dsa.deviceId = deviceId;
 
-	retVal = HvCall4(HvCallPciConfigStore16, *(u64 *)&dsa, offset, value, 0);
-
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
-	return retVal;
+	return HvCall4(HvCallPciConfigStore16, *(u64 *)&dsa, offset, value, 0);
 }
 
 static inline u64 HvCallPci_configStore32(u16 busNumber, u8 subBusNumber,
 		u8 deviceId, u32 offset, u32 value)
 {
 	struct HvCallPci_DsaAddr dsa;
-	u64 retVal;
 
 	*((u64*)&dsa) = 0;
 
@@ -239,11 +222,7 @@ static inline u64 HvCallPci_configStore32(u16 busNumber, u8 subBusNumber,
 	dsa.subBusNumber = subBusNumber;
 	dsa.deviceId = deviceId;
 
-	retVal = HvCall4(HvCallPciConfigStore32, *(u64 *)&dsa, offset, value, 0);
-
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
-	return retVal;
+	return HvCall4(HvCallPciConfigStore32, *(u64 *)&dsa, offset, value, 0);
 }
 
 static inline u64 HvCallPci_barLoad8(u16 busNumberParm, u8 subBusParm,
@@ -262,8 +241,6 @@ static inline u64 HvCallPci_barLoad8(u16 busNumberParm, u8 subBusParm,
 
 	HvCall3Ret16(HvCallPciBarLoad8, &retVal, *(u64 *)&dsa, offsetParm, 0);
 
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
 	*valueParm = retVal.value;
 
 	return retVal.rc;
@@ -285,8 +262,6 @@ static inline u64 HvCallPci_barLoad16(u16 busNumberParm, u8 subBusParm,
 
 	HvCall3Ret16(HvCallPciBarLoad16, &retVal, *(u64 *)&dsa, offsetParm, 0);
 
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
 	*valueParm = retVal.value;
 
 	return retVal.rc;
@@ -308,8 +283,6 @@ static inline u64 HvCallPci_barLoad32(u16 busNumberParm, u8 subBusParm,
 
 	HvCall3Ret16(HvCallPciBarLoad32, &retVal, *(u64 *)&dsa, offsetParm, 0);
 
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
 	*valueParm = retVal.value;
 
 	return retVal.rc;
@@ -331,8 +304,6 @@ static inline u64 HvCallPci_barLoad64(u16 busNumberParm, u8 subBusParm,
 
 	HvCall3Ret16(HvCallPciBarLoad64, &retVal, *(u64 *)&dsa, offsetParm, 0);
 
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
 	*valueParm = retVal.value;
 
 	return retVal.rc;
@@ -343,7 +314,6 @@ static inline u64 HvCallPci_barStore8(u16 busNumberParm, u8 subBusParm,
 		u8 valueParm)
 {
 	struct HvCallPci_DsaAddr dsa;
-	u64 retVal;
 
 	*((u64*)&dsa) = 0;
 
@@ -352,11 +322,8 @@ static inline u64 HvCallPci_barStore8(u16 busNumberParm, u8 subBusParm,
 	dsa.deviceId = deviceIdParm;
 	dsa.barNumber = barNumberParm;
 
-	retVal = HvCall4(HvCallPciBarStore8, *(u64 *)&dsa, offsetParm, valueParm, 0);
-
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
-	return retVal;
+	return HvCall4(HvCallPciBarStore8, *(u64 *)&dsa, offsetParm,
+			valueParm, 0);
 }
 
 static inline u64 HvCallPci_barStore16(u16 busNumberParm, u8 subBusParm,
@@ -364,7 +331,6 @@ static inline u64 HvCallPci_barStore16(u16 busNumberParm, u8 subBusParm,
 		u16 valueParm)
 {
 	struct HvCallPci_DsaAddr dsa;
-	u64 retVal;
 
 	*((u64*)&dsa) = 0;
 
@@ -373,11 +339,8 @@ static inline u64 HvCallPci_barStore16(u16 busNumberParm, u8 subBusParm,
 	dsa.deviceId = deviceIdParm;
 	dsa.barNumber = barNumberParm;
 
-	retVal = HvCall4(HvCallPciBarStore16, *(u64 *)&dsa, offsetParm, valueParm, 0);
-
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
-	return retVal;
+	return HvCall4(HvCallPciBarStore16, *(u64 *)&dsa, offsetParm,
+			valueParm, 0);
 }
 
 static inline u64 HvCallPci_barStore32(u16 busNumberParm, u8 subBusParm,
@@ -385,7 +348,6 @@ static inline u64 HvCallPci_barStore32(u16 busNumberParm, u8 subBusParm,
 		u32 valueParm)
 {
 	struct HvCallPci_DsaAddr dsa;
-	u64 retVal;
 
 	*((u64*)&dsa) = 0;
 
@@ -394,11 +356,8 @@ static inline u64 HvCallPci_barStore32(u16 busNumberParm, u8 subBusParm,
 	dsa.deviceId = deviceIdParm;
 	dsa.barNumber = barNumberParm;
 
-	retVal = HvCall4(HvCallPciBarStore32, *(u64 *)&dsa, offsetParm, valueParm, 0);
-
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
-	return retVal;
+	return HvCall4(HvCallPciBarStore32, *(u64 *)&dsa, offsetParm,
+			valueParm, 0);
 }
 
 static inline u64 HvCallPci_barStore64(u16 busNumberParm, u8 subBusParm,
@@ -406,7 +365,6 @@ static inline u64 HvCallPci_barStore64(u16 busNumberParm, u8 subBusParm,
 		u64 valueParm)
 {
 	struct HvCallPci_DsaAddr dsa;
-	u64 retVal;
 
 	*((u64*)&dsa) = 0;
 
@@ -415,11 +373,8 @@ static inline u64 HvCallPci_barStore64(u16 busNumberParm, u8 subBusParm,
 	dsa.deviceId = deviceIdParm;
 	dsa.barNumber = barNumberParm;
 
-	retVal = HvCall4(HvCallPciBarStore64, *(u64 *)&dsa, offsetParm, valueParm, 0);
-
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
-	return retVal;
+	return HvCall4(HvCallPciBarStore64, *(u64 *)&dsa, offsetParm,
+			valueParm, 0);
 }
 
 static inline u64 HvCallPci_eoi(u16 busNumberParm, u8 subBusParm,
@@ -436,8 +391,6 @@ static inline u64 HvCallPci_eoi(u16 busNumberParm, u8 subBusParm,
 
 	HvCall1Ret16(HvCallPciEoi, &retVal, *(u64*)&dsa);
 
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
 	return retVal.rc;
 }
 
@@ -445,7 +398,6 @@ static inline u64 HvCallPci_getBarParms(u16 busNumberParm, u8 subBusParm,
 		u8 deviceIdParm, u8 barNumberParm, u64 parms, u32 sizeofParms)
 {
 	struct HvCallPci_DsaAddr dsa;
-	u64 retVal;
 
 	*((u64*)&dsa) = 0;
 
@@ -454,18 +406,13 @@ static inline u64 HvCallPci_getBarParms(u16 busNumberParm, u8 subBusParm,
 	dsa.deviceId = deviceIdParm;
 	dsa.barNumber = barNumberParm;
 
-	retVal = HvCall3(HvCallPciGetBarParms, *(u64*)&dsa, parms, sizeofParms);
-
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
-	return retVal;
+	return HvCall3(HvCallPciGetBarParms, *(u64*)&dsa, parms, sizeofParms);
 }
 
 static inline u64 HvCallPci_maskFisr(u16 busNumberParm, u8 subBusParm,
 		u8 deviceIdParm, u64 fisrMask)
 {
 	struct HvCallPci_DsaAddr dsa;
-	u64 retVal;
 
 	*((u64*)&dsa) = 0;
 
@@ -473,18 +420,13 @@ static inline u64 HvCallPci_maskFisr(u16 busNumberParm, u8 subBusParm,
 	dsa.subBusNumber = subBusParm;
 	dsa.deviceId = deviceIdParm;
 
-	retVal = HvCall2(HvCallPciMaskFisr, *(u64*)&dsa, fisrMask);
-
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
-	return retVal;
+	return HvCall2(HvCallPciMaskFisr, *(u64*)&dsa, fisrMask);
 }
 
 static inline u64 HvCallPci_unmaskFisr(u16 busNumberParm, u8 subBusParm,
 		u8 deviceIdParm, u64 fisrMask)
 {
 	struct HvCallPci_DsaAddr dsa;
-	u64 retVal;
 
 	*((u64*)&dsa) = 0;
 
@@ -492,18 +434,13 @@ static inline u64 HvCallPci_unmaskFisr(u16 busNumberParm, u8 subBusParm,
 	dsa.subBusNumber = subBusParm;
 	dsa.deviceId = deviceIdParm;
 
-	retVal = HvCall2(HvCallPciUnmaskFisr, *(u64*)&dsa, fisrMask);
-
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
-	return retVal;
+	return HvCall2(HvCallPciUnmaskFisr, *(u64*)&dsa, fisrMask);
 }
 
 static inline u64 HvCallPci_setSlotReset(u16 busNumberParm, u8 subBusParm,
 		u8 deviceIdParm, u64 onNotOff)
 {
 	struct HvCallPci_DsaAddr dsa;
-	u64 retVal;
 
 	*((u64*)&dsa) = 0;
 
@@ -511,18 +448,13 @@ static inline u64 HvCallPci_setSlotReset(u16 busNumberParm, u8 subBusParm,
 	dsa.subBusNumber = subBusParm;
 	dsa.deviceId = deviceIdParm;
 
-	retVal = HvCall2(HvCallPciSetSlotReset, *(u64*)&dsa, onNotOff);
-
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
-	return retVal;
+	return HvCall2(HvCallPciSetSlotReset, *(u64*)&dsa, onNotOff);
 }
 
 static inline u64 HvCallPci_getDeviceInfo(u16 busNumberParm, u8 subBusParm,
 		u8 deviceNumberParm, u64 parms, u32 sizeofParms)
 {
 	struct HvCallPci_DsaAddr dsa;
-	u64 retVal;
 
 	*((u64*)&dsa) = 0;
 
@@ -530,18 +462,13 @@ static inline u64 HvCallPci_getDeviceInfo(u16 busNumberParm, u8 subBusParm,
 	dsa.subBusNumber = subBusParm;
 	dsa.deviceId = deviceNumberParm << 4;
 
-	retVal = HvCall3(HvCallPciGetDeviceInfo, *(u64*)&dsa, parms, sizeofParms);
-
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
-	return retVal;
+	return HvCall3(HvCallPciGetDeviceInfo, *(u64*)&dsa, parms, sizeofParms);
 }
 
 static inline u64 HvCallPci_maskInterrupts(u16 busNumberParm, u8 subBusParm,
 		u8 deviceIdParm, u64 interruptMask)
 {
 	struct HvCallPci_DsaAddr dsa;
-	u64 retVal;
 
 	*((u64*)&dsa) = 0;
 
@@ -549,18 +476,13 @@ static inline u64 HvCallPci_maskInterrupts(u16 busNumberParm, u8 subBusParm,
 	dsa.subBusNumber = subBusParm;
 	dsa.deviceId = deviceIdParm;
 
-	retVal = HvCall2(HvCallPciMaskInterrupts, *(u64*)&dsa, interruptMask);
-
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
-	return retVal;
+	return HvCall2(HvCallPciMaskInterrupts, *(u64*)&dsa, interruptMask);
 }
 
 static inline u64 HvCallPci_unmaskInterrupts(u16 busNumberParm, u8 subBusParm,
 		u8 deviceIdParm, u64 interruptMask)
 {
 	struct HvCallPci_DsaAddr dsa;
-	u64 retVal;
 
 	*((u64*)&dsa) = 0;
 
@@ -568,18 +490,13 @@ static inline u64 HvCallPci_unmaskInterrupts(u16 busNumberParm, u8 subBusParm,
 	dsa.subBusNumber = subBusParm;
 	dsa.deviceId = deviceIdParm;
 
-	retVal = HvCall2(HvCallPciUnmaskInterrupts, *(u64*)&dsa, interruptMask);
-
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
-	return retVal;
+	return HvCall2(HvCallPciUnmaskInterrupts, *(u64*)&dsa, interruptMask);
 }
 
 static inline u64 HvCallPci_getBusUnitInfo(u16 busNumberParm, u8 subBusParm,
 		u8 deviceIdParm, u64 parms, u32 sizeofParms)
 {
 	struct HvCallPci_DsaAddr dsa;
-	u64 retVal;
 
 	*((u64*)&dsa) = 0;
 
@@ -587,37 +504,30 @@ static inline u64 HvCallPci_getBusUnitInfo(u16 busNumberParm, u8 subBusParm,
 	dsa.subBusNumber = subBusParm;
 	dsa.deviceId = deviceIdParm;
 
-	retVal = HvCall3(HvCallPciGetBusUnitInfo, *(u64*)&dsa, parms, sizeofParms);
-
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-
-	return retVal;
+	return HvCall3(HvCallPciGetBusUnitInfo, *(u64*)&dsa, parms,
+			sizeofParms);
 }
 
 static inline int HvCallPci_getBusVpd(u16 busNumParm, u64 destParm,
 		u16 sizeParm)
 {
-	int xRetSize;
-	u64 xRc = HvCall4(HvCallPciGetCardVpd, busNumParm, destParm, sizeParm, HvCallPci_BusVpd);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
+	u64 xRc = HvCall4(HvCallPciGetCardVpd, busNumParm, destParm,
+			sizeParm, HvCallPci_BusVpd);
 	if (xRc == -1)
-		xRetSize = -1;
+		return -1;
 	else
-		xRetSize = xRc & 0xFFFF;
-	return xRetSize;
+		return xRc & 0xFFFF;
 }
 
 static inline int HvCallPci_getBusAdapterVpd(u16 busNumParm, u64 destParm,
 		u16 sizeParm)
 {
-	int xRetSize;
-	u64 xRc = HvCall4(HvCallPciGetCardVpd, busNumParm, destParm, sizeParm, HvCallPci_BusAdapterVpd);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
+	u64 xRc = HvCall4(HvCallPciGetCardVpd, busNumParm, destParm,
+			sizeParm, HvCallPci_BusAdapterVpd);
 	if (xRc == -1)
-		xRetSize = -1;
+		return -1;
 	else
-		xRetSize = xRc & 0xFFFF;
-	return xRetSize;
+		return xRc & 0xFFFF;
 }
 
 #endif /* _HVCALLPCI_H */
diff --git a/include/asm-ppc64/iSeries/HvCallSm.h b/include/asm-ppc64/iSeries/HvCallSm.h
index ef6b303e4d2..8a3dbb071a4 100644
--- a/include/asm-ppc64/iSeries/HvCallSm.h
+++ b/include/asm-ppc64/iSeries/HvCallSm.h
@@ -32,10 +32,7 @@
 static inline u64 HvCallSm_get64BitsOfAccessMap(HvLpIndex lpIndex,
 		u64 indexIntoBitMap)
 {
-	u64 retval = HvCall2(HvCallSmGet64BitsOfAccessMap, lpIndex,
-			     indexIntoBitMap );
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retval;
+	return HvCall2(HvCallSmGet64BitsOfAccessMap, lpIndex, indexIntoBitMap);
 }
 
 #endif /* _HVCALLSM_H */
diff --git a/include/asm-ppc64/iSeries/HvCallXm.h b/include/asm-ppc64/iSeries/HvCallXm.h
index a99f29b9361..1362a425913 100644
--- a/include/asm-ppc64/iSeries/HvCallXm.h
+++ b/include/asm-ppc64/iSeries/HvCallXm.h
@@ -19,58 +19,43 @@
 static inline void HvCallXm_getTceTableParms(u64 cb)
 {
 	HvCall1(HvCallXmGetTceTableParms, cb);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
 }
 
 static inline u64 HvCallXm_setTce(u64 tceTableToken, u64 tceOffset, u64 tce)
 {
-	u64 retval = HvCall3(HvCallXmSetTce, tceTableToken, tceOffset, tce );
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retval;
+	return HvCall3(HvCallXmSetTce, tceTableToken, tceOffset, tce);
 }
 
 static inline u64 HvCallXm_setTces(u64 tceTableToken, u64 tceOffset,
 		u64 numTces, u64 tce1, u64 tce2, u64 tce3, u64 tce4)
 {
-	u64 retval = HvCall7(HvCallXmSetTces, tceTableToken, tceOffset, numTces,
-			     tce1, tce2, tce3, tce4 );
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retval;
+	return HvCall7(HvCallXmSetTces, tceTableToken, tceOffset, numTces,
+			     tce1, tce2, tce3, tce4);
 }
 
 static inline u64 HvCallXm_testBus(u16 busNumber)
 {
-	u64 retVal = HvCall1(HvCallXmTestBus, busNumber);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall1(HvCallXmTestBus, busNumber);
 }
 
 static inline u64 HvCallXm_testBusUnit(u16 busNumber, u8 subBusNumber,
 		u8 deviceId)
 {
-	u64 busUnitNumber = (subBusNumber << 8) | deviceId;
-	u64 retVal = HvCall2(HvCallXmTestBusUnit, busNumber, busUnitNumber);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall2(HvCallXmTestBusUnit, busNumber,
+			(subBusNumber << 8) | deviceId);
 }
 
 static inline u64 HvCallXm_connectBusUnit(u16 busNumber, u8 subBusNumber,
 		u8 deviceId, u64 interruptToken)
 {
-	u64 busUnitNumber = (subBusNumber << 8) | deviceId;
-	u64 queueIndex = 0; // HvLpConfig::mapDsaToQueueIndex(HvLpDSA(busNumber, xBoard, xCard));  
-
-	u64 retVal = HvCall5(HvCallXmConnectBusUnit, busNumber, busUnitNumber,
-			     interruptToken, 0, queueIndex);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall5(HvCallXmConnectBusUnit, busNumber,
+			(subBusNumber << 8) | deviceId, interruptToken, 0,
+			0 /* HvLpConfig::mapDsaToQueueIndex(HvLpDSA(busNumber, xBoard, xCard)) */);
 }
 
 static inline u64 HvCallXm_loadTod(void)
 {
-	u64 retVal = HvCall0(HvCallXmLoadTod);
-	// getPaca()->adjustHmtForNoOfSpinLocksHeld();
-	return retVal;
+	return HvCall0(HvCallXmLoadTod);
 }
 
 #endif /* _HVCALLXM_H */
diff --git a/include/asm-ppc64/iSeries/HvLpConfig.h b/include/asm-ppc64/iSeries/HvLpConfig.h
index 47311cc952d..817698052ea 100644
--- a/include/asm-ppc64/iSeries/HvLpConfig.h
+++ b/include/asm-ppc64/iSeries/HvLpConfig.h
@@ -281,8 +281,7 @@ static inline u64 HvLpConfig_getLoadPages(void)
 
 static inline int HvLpConfig_isBusOwnedByThisLp(HvBusNumber busNumber)
 {
-	HvLpIndex busOwner = HvLpConfig_getBusOwner(busNumber);
-	return (busOwner == HvLpConfig_getLpIndex());
+	return (HvLpConfig_getBusOwner(busNumber) == HvLpConfig_getLpIndex());
 }
 
 static inline int HvLpConfig_doLpsCommunicateOnVirtualLan(HvLpIndex lp1,
diff --git a/include/asm-ppc64/iSeries/XmPciLpEvent.h b/include/asm-ppc64/iSeries/XmPciLpEvent.h
index 9c800792bc5..4936627b26d 100644
--- a/include/asm-ppc64/iSeries/XmPciLpEvent.h
+++ b/include/asm-ppc64/iSeries/XmPciLpEvent.h
@@ -1,15 +1,7 @@
 #ifndef	__XMPCILPEVENT_H__
 #define	__XMPCILPEVENT_H__
 
-#ifdef __cplusplus
-extern "C" {
-#endif
-
-int XmPciLpEvent_init(void);
-void ppc_irq_dispatch_handler(struct pt_regs *regs, int irq);
-
-#ifdef __cplusplus
-}
-#endif
+extern int XmPciLpEvent_init(void);
+extern void ppc_irq_dispatch_handler(struct pt_regs *regs, int irq);
 
 #endif /* __XMPCILPEVENT_H__ */
diff --git a/include/asm-ppc64/iSeries/iSeries_irq.h b/include/asm-ppc64/iSeries/iSeries_irq.h
index ff8ddede038..fbd5f523ba1 100644
--- a/include/asm-ppc64/iSeries/iSeries_irq.h
+++ b/include/asm-ppc64/iSeries/iSeries_irq.h
@@ -1,19 +1,11 @@
 #ifndef	__ISERIES_IRQ_H__
 #define	__ISERIES_IRQ_H__
 
-#ifdef __cplusplus
-extern "C" {
-#endif
+extern void iSeries_init_IRQ(void);
+extern int  iSeries_allocate_IRQ(HvBusNumber, HvSubBusNumber, HvAgentId);
+extern int  iSeries_assign_IRQ(int, HvBusNumber, HvSubBusNumber, HvAgentId);
+extern void iSeries_activate_IRQs(void);
 
-void iSeries_init_IRQ(void);
-int  iSeries_allocate_IRQ(HvBusNumber, HvSubBusNumber, HvAgentId);
-int  iSeries_assign_IRQ(int, HvBusNumber, HvSubBusNumber, HvAgentId);
-void iSeries_activate_IRQs(void);
-
-int XmPciLpEvent_init(void);
-
-#ifdef __cplusplus
-}
-#endif
+extern int XmPciLpEvent_init(void);
 
 #endif /* __ISERIES_IRQ_H__ */
diff --git a/include/asm-ppc64/iSeries/iSeries_pci.h b/include/asm-ppc64/iSeries/iSeries_pci.h
index 9c6d64526d5..f45cd8a2e7b 100644
--- a/include/asm-ppc64/iSeries/iSeries_pci.h
+++ b/include/asm-ppc64/iSeries/iSeries_pci.h
@@ -55,7 +55,7 @@ struct iSeries_Device_Node;
  */
 
 #define ISERIES_PCI_AGENTID(idsel, func)	\
-	((idsel & 0x0F) << 4) | (func & 0x07)
+	(((idsel & 0x0F) << 4) | (func & 0x07))
 #define ISERIES_ENCODE_DEVICE(agentid)		\
 	((0x10) | ((agentid & 0x20) >> 2) | (agentid & 0x07))
 
diff --git a/include/asm-ppc64/iSeries/vio.h b/include/asm-ppc64/iSeries/vio.h
index 614088f61b8..6c05e6257f5 100644
--- a/include/asm-ppc64/iSeries/vio.h
+++ b/include/asm-ppc64/iSeries/vio.h
@@ -58,16 +58,16 @@
  */
 typedef void (vio_event_handler_t) (struct HvLpEvent * event);
 
-int viopath_open(HvLpIndex remoteLp, int subtype, int numReq);
-int viopath_close(HvLpIndex remoteLp, int subtype, int numReq);
-int vio_setHandler(int subtype, vio_event_handler_t * beh);
-int vio_clearHandler(int subtype);
-int viopath_isactive(HvLpIndex lp);
-HvLpInstanceId viopath_sourceinst(HvLpIndex lp);
-HvLpInstanceId viopath_targetinst(HvLpIndex lp);
-void vio_set_hostlp(void);
-void *vio_get_event_buffer(int subtype);
-void vio_free_event_buffer(int subtype, void *buffer);
+extern int viopath_open(HvLpIndex remoteLp, int subtype, int numReq);
+extern int viopath_close(HvLpIndex remoteLp, int subtype, int numReq);
+extern int vio_setHandler(int subtype, vio_event_handler_t * beh);
+extern int vio_clearHandler(int subtype);
+extern int viopath_isactive(HvLpIndex lp);
+extern HvLpInstanceId viopath_sourceinst(HvLpIndex lp);
+extern HvLpInstanceId viopath_targetinst(HvLpIndex lp);
+extern void vio_set_hostlp(void);
+extern void *vio_get_event_buffer(int subtype);
+extern void vio_free_event_buffer(int subtype, void *buffer);
 
 extern HvLpIndex viopath_hostLp;
 extern HvLpIndex viopath_ourLp;
-- 
cgit v1.2.3-70-g09d2


From 0bc0ffd5f0854b9143606684fb925f4290ae13e7 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:36 -0700
Subject: [PATCH] ppc64 iSeries: remove LparData.h

include/asm-ppc64/iSeries/LparData.h just included a whole lot of other files
to declare variables that would be better declared in those other files.  So,
remove it.  This will reduce that number of things needed to be included in
most cases to access the relevant variables.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/HvLpEvent.c                 |  2 +-
 arch/ppc64/kernel/ItLpQueue.c                 |  1 -
 arch/ppc64/kernel/iSeries_VpdInfo.c           |  1 -
 arch/ppc64/kernel/iSeries_pci.c               |  1 -
 arch/ppc64/kernel/iSeries_proc.c              |  2 +-
 arch/ppc64/kernel/iSeries_setup.c             |  5 ++-
 arch/ppc64/kernel/iSeries_smp.c               |  1 -
 arch/ppc64/kernel/irq.c                       |  2 +-
 arch/ppc64/kernel/lparcfg.c                   |  2 +-
 arch/ppc64/kernel/ras.c                       |  1 -
 arch/ppc64/kernel/rtc.c                       |  1 -
 arch/ppc64/kernel/setup.c                     |  3 +-
 arch/ppc64/kernel/viopath.c                   |  2 +-
 include/asm-ppc64/iSeries/HvLpConfig.h        |  1 -
 include/asm-ppc64/iSeries/HvReleaseData.h     |  2 ++
 include/asm-ppc64/iSeries/IoHriMainStore.h    |  2 ++
 include/asm-ppc64/iSeries/IoHriProcessorVpd.h |  2 ++
 include/asm-ppc64/iSeries/ItExtVpdPanel.h     |  2 ++
 include/asm-ppc64/iSeries/ItIplParmsReal.h    |  2 ++
 include/asm-ppc64/iSeries/ItLpNaca.h          |  4 +++
 include/asm-ppc64/iSeries/ItVpdAreas.h        |  2 ++
 include/asm-ppc64/iSeries/LparData.h          | 48 ---------------------------
 include/asm-ppc64/iSeries/LparMap.h           |  2 ++
 23 files changed, 29 insertions(+), 62 deletions(-)
 delete mode 100644 include/asm-ppc64/iSeries/LparData.h

(limited to 'include')

diff --git a/arch/ppc64/kernel/HvLpEvent.c b/arch/ppc64/kernel/HvLpEvent.c
index f8f19637f73..90032b13890 100644
--- a/arch/ppc64/kernel/HvLpEvent.c
+++ b/arch/ppc64/kernel/HvLpEvent.c
@@ -12,7 +12,7 @@
 #include <asm/system.h>
 #include <asm/iSeries/HvLpEvent.h>
 #include <asm/iSeries/HvCallEvent.h>
-#include <asm/iSeries/LparData.h>
+#include <asm/iSeries/ItLpNaca.h>
 
 /* Array of LpEvent handler functions */
 LpEventHandler lpEventHandler[HvLpEvent_Type_NumTypes];
diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index c923a815760..cdea00d7707 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -16,7 +16,6 @@
 #include <asm/iSeries/ItLpQueue.h>
 #include <asm/iSeries/HvLpEvent.h>
 #include <asm/iSeries/HvCallEvent.h>
-#include <asm/iSeries/LparData.h>
 
 static __inline__ int set_inUse( struct ItLpQueue * lpQueue )
 {
diff --git a/arch/ppc64/kernel/iSeries_VpdInfo.c b/arch/ppc64/kernel/iSeries_VpdInfo.c
index a6f0ff2d023..a82caf98d2e 100644
--- a/arch/ppc64/kernel/iSeries_VpdInfo.c
+++ b/arch/ppc64/kernel/iSeries_VpdInfo.c
@@ -35,7 +35,6 @@
 #include <asm/iSeries/HvCallPci.h>
 #include <asm/iSeries/HvTypes.h>
 #include <asm/iSeries/mf.h>
-#include <asm/iSeries/LparData.h>
 #include <asm/iSeries/iSeries_pci.h>
 #include "pci.h"
 
diff --git a/arch/ppc64/kernel/iSeries_pci.c b/arch/ppc64/kernel/iSeries_pci.c
index bd4c2554f1a..9829cd4666d 100644
--- a/arch/ppc64/kernel/iSeries_pci.c
+++ b/arch/ppc64/kernel/iSeries_pci.c
@@ -40,7 +40,6 @@
 #include <asm/iSeries/HvCallPci.h>
 #include <asm/iSeries/HvCallSm.h>
 #include <asm/iSeries/HvCallXm.h>
-#include <asm/iSeries/LparData.h>
 #include <asm/iSeries/iSeries_irq.h>
 #include <asm/iSeries/iSeries_pci.h>
 #include <asm/iSeries/mf.h>
diff --git a/arch/ppc64/kernel/iSeries_proc.c b/arch/ppc64/kernel/iSeries_proc.c
index bf85a1f7e87..356bd9931fc 100644
--- a/arch/ppc64/kernel/iSeries_proc.c
+++ b/arch/ppc64/kernel/iSeries_proc.c
@@ -28,7 +28,7 @@
 #include <asm/iSeries/ItLpQueue.h>
 #include <asm/iSeries/HvCallXm.h>
 #include <asm/iSeries/IoHriMainStore.h>
-#include <asm/iSeries/LparData.h>
+#include <asm/iSeries/IoHriProcessorVpd.h>
 
 static int __init iseries_proc_create(void)
 {
diff --git a/arch/ppc64/kernel/iSeries_setup.c b/arch/ppc64/kernel/iSeries_setup.c
index e80cfc57092..3d3ed631499 100644
--- a/arch/ppc64/kernel/iSeries_setup.c
+++ b/arch/ppc64/kernel/iSeries_setup.c
@@ -47,7 +47,7 @@
 #include <asm/paca.h>
 #include <asm/cache.h>
 #include <asm/sections.h>
-#include <asm/iSeries/LparData.h>
+#include <asm/abs_addr.h>
 #include <asm/iSeries/HvCallHpt.h>
 #include <asm/iSeries/HvLpConfig.h>
 #include <asm/iSeries/HvCallEvent.h>
@@ -58,6 +58,9 @@
 #include <asm/iSeries/mf.h>
 #include <asm/iSeries/HvLpEvent.h>
 #include <asm/iSeries/iSeries_irq.h>
+#include <asm/iSeries/IoHriProcessorVpd.h>
+#include <asm/iSeries/ItVpdAreas.h>
+#include <asm/iSeries/LparMap.h>
 
 extern void hvlog(char *fmt, ...);
 
diff --git a/arch/ppc64/kernel/iSeries_smp.c b/arch/ppc64/kernel/iSeries_smp.c
index ba1f084d546..ff2281b457e 100644
--- a/arch/ppc64/kernel/iSeries_smp.c
+++ b/arch/ppc64/kernel/iSeries_smp.c
@@ -38,7 +38,6 @@
 #include <asm/io.h>
 #include <asm/smp.h>
 #include <asm/paca.h>
-#include <asm/iSeries/LparData.h>
 #include <asm/iSeries/HvCall.h>
 #include <asm/iSeries/HvCallCfg.h>
 #include <asm/time.h>
diff --git a/arch/ppc64/kernel/irq.c b/arch/ppc64/kernel/irq.c
index 4fd7f203c1e..d860467b8f0 100644
--- a/arch/ppc64/kernel/irq.c
+++ b/arch/ppc64/kernel/irq.c
@@ -52,7 +52,7 @@
 #include <asm/cache.h>
 #include <asm/prom.h>
 #include <asm/ptrace.h>
-#include <asm/iSeries/LparData.h>
+#include <asm/iSeries/ItLpQueue.h>
 #include <asm/machdep.h>
 #include <asm/paca.h>
 
diff --git a/arch/ppc64/kernel/lparcfg.c b/arch/ppc64/kernel/lparcfg.c
index a8fd32df848..387923fcf9b 100644
--- a/arch/ppc64/kernel/lparcfg.c
+++ b/arch/ppc64/kernel/lparcfg.c
@@ -28,12 +28,12 @@
 #include <asm/uaccess.h>
 #include <asm/iSeries/HvLpConfig.h>
 #include <asm/lppaca.h>
-#include <asm/iSeries/LparData.h>
 #include <asm/hvcall.h>
 #include <asm/cputable.h>
 #include <asm/rtas.h>
 #include <asm/system.h>
 #include <asm/time.h>
+#include <asm/iSeries/ItExtVpdPanel.h>
 
 #define MODULE_VERS "1.6"
 #define MODULE_NAME "lparcfg"
diff --git a/arch/ppc64/kernel/ras.c b/arch/ppc64/kernel/ras.c
index 1c4c796b212..3c00f7bfc1b 100644
--- a/arch/ppc64/kernel/ras.c
+++ b/arch/ppc64/kernel/ras.c
@@ -47,7 +47,6 @@
 #include <asm/cache.h>
 #include <asm/prom.h>
 #include <asm/ptrace.h>
-#include <asm/iSeries/LparData.h>
 #include <asm/machdep.h>
 #include <asm/rtas.h>
 #include <asm/ppcdebug.h>
diff --git a/arch/ppc64/kernel/rtc.c b/arch/ppc64/kernel/rtc.c
index 67989055a9f..4cdafb4007b 100644
--- a/arch/ppc64/kernel/rtc.c
+++ b/arch/ppc64/kernel/rtc.c
@@ -42,7 +42,6 @@
 #include <asm/time.h>
 #include <asm/rtas.h>
 
-#include <asm/iSeries/LparData.h>
 #include <asm/iSeries/mf.h>
 #include <asm/machdep.h>
 #include <asm/iSeries/ItSpCommArea.h>
diff --git a/arch/ppc64/kernel/setup.c b/arch/ppc64/kernel/setup.c
index dce198d3932..8e439a81764 100644
--- a/arch/ppc64/kernel/setup.c
+++ b/arch/ppc64/kernel/setup.c
@@ -41,7 +41,6 @@
 #include <asm/smp.h>
 #include <asm/elf.h>
 #include <asm/machdep.h>
-#include <asm/iSeries/LparData.h>
 #include <asm/paca.h>
 #include <asm/ppcdebug.h>
 #include <asm/time.h>
@@ -57,6 +56,8 @@
 #include <asm/cache.h>
 #include <asm/page.h>
 #include <asm/mmu.h>
+#include <asm/lmb.h>
+#include <asm/iSeries/ItLpNaca.h>
 
 #ifdef DEBUG
 #define DBG(fmt...) udbg_printf(fmt)
diff --git a/arch/ppc64/kernel/viopath.c b/arch/ppc64/kernel/viopath.c
index 8caa078a658..4e5c36e81f4 100644
--- a/arch/ppc64/kernel/viopath.c
+++ b/arch/ppc64/kernel/viopath.c
@@ -43,7 +43,7 @@
 #include <asm/system.h>
 #include <asm/uaccess.h>
 #include <asm/iSeries/HvTypes.h>
-#include <asm/iSeries/LparData.h>
+#include <asm/iSeries/ItExtVpdPanel.h>
 #include <asm/iSeries/HvLpEvent.h>
 #include <asm/iSeries/HvLpConfig.h>
 #include <asm/iSeries/HvCallCfg.h>
diff --git a/include/asm-ppc64/iSeries/HvLpConfig.h b/include/asm-ppc64/iSeries/HvLpConfig.h
index 817698052ea..9411b5a04d0 100644
--- a/include/asm-ppc64/iSeries/HvLpConfig.h
+++ b/include/asm-ppc64/iSeries/HvLpConfig.h
@@ -27,7 +27,6 @@
 #include <asm/iSeries/HvCallCfg.h>
 #include <asm/iSeries/HvTypes.h>
 #include <asm/iSeries/ItLpNaca.h>
-#include <asm/iSeries/LparData.h>
 
 extern HvLpIndex HvLpConfig_getLpIndex_outline(void);
 
diff --git a/include/asm-ppc64/iSeries/HvReleaseData.h b/include/asm-ppc64/iSeries/HvReleaseData.h
index c0cd21caf8f..01a1f13ea4a 100644
--- a/include/asm-ppc64/iSeries/HvReleaseData.h
+++ b/include/asm-ppc64/iSeries/HvReleaseData.h
@@ -58,4 +58,6 @@ struct HvReleaseData {
 	char	xRsvd3[20];	/* Reserved			x2C-x3F */
 };
 
+extern struct HvReleaseData	hvReleaseData;
+
 #endif /* _HVRELEASEDATA_H */
diff --git a/include/asm-ppc64/iSeries/IoHriMainStore.h b/include/asm-ppc64/iSeries/IoHriMainStore.h
index fbb3ee4d174..45ed3ea67d0 100644
--- a/include/asm-ppc64/iSeries/IoHriMainStore.h
+++ b/include/asm-ppc64/iSeries/IoHriMainStore.h
@@ -161,4 +161,6 @@ struct IoHriMainStoreSegment5 {
 	u64	reserved3;
 };
 
+extern u64	xMsVpd[];
+
 #endif	/* _IOHRIMAINSTORE_H */
diff --git a/include/asm-ppc64/iSeries/IoHriProcessorVpd.h b/include/asm-ppc64/iSeries/IoHriProcessorVpd.h
index ccb7b7c55a4..73b73d80b8b 100644
--- a/include/asm-ppc64/iSeries/IoHriProcessorVpd.h
+++ b/include/asm-ppc64/iSeries/IoHriProcessorVpd.h
@@ -81,4 +81,6 @@ struct IoHriProcessorVpd {
 	char	xProcSrc[72];		// CSP format SRC		xB8-xFF
 };
 
+extern struct IoHriProcessorVpd	xIoHriProcessorVpd[];
+
 #endif /* _IOHRIPROCESSORVPD_H */
diff --git a/include/asm-ppc64/iSeries/ItExtVpdPanel.h b/include/asm-ppc64/iSeries/ItExtVpdPanel.h
index b7149079d3c..4c546a8802b 100644
--- a/include/asm-ppc64/iSeries/ItExtVpdPanel.h
+++ b/include/asm-ppc64/iSeries/ItExtVpdPanel.h
@@ -47,4 +47,6 @@ struct ItExtVpdPanel {
 	u8	xRsvd2[48];
 };
 
+extern struct ItExtVpdPanel	xItExtVpdPanel;
+
 #endif /* _ITEXTVPDPANEL_H */
diff --git a/include/asm-ppc64/iSeries/ItIplParmsReal.h b/include/asm-ppc64/iSeries/ItIplParmsReal.h
index 14f0b0a2a82..ae3417dc599 100644
--- a/include/asm-ppc64/iSeries/ItIplParmsReal.h
+++ b/include/asm-ppc64/iSeries/ItIplParmsReal.h
@@ -66,4 +66,6 @@ struct ItIplParmsReal {
 	u64	xRsvd13;		// Reserved			x38-x3F
 };
 
+extern struct ItIplParmsReal	xItIplParmsReal;
+
 #endif /* _ITIPLPARMSREAL_H */
diff --git a/include/asm-ppc64/iSeries/ItLpNaca.h b/include/asm-ppc64/iSeries/ItLpNaca.h
index 942b07d9140..225d0176779 100644
--- a/include/asm-ppc64/iSeries/ItLpNaca.h
+++ b/include/asm-ppc64/iSeries/ItLpNaca.h
@@ -19,6 +19,8 @@
 #ifndef _ITLPNACA_H
 #define _ITLPNACA_H
 
+#include <linux/types.h>
+
 /*
  *	This control block contains the data that is shared between the
  *	hypervisor (PLIC) and the OS.
@@ -73,4 +75,6 @@ struct ItLpNaca {
 	u64	xInterruptHdlr[32];	// Interrupt handlers		300-x3FF
 };
 
+extern struct ItLpNaca		itLpNaca;
+
 #endif /* _ITLPNACA_H */
diff --git a/include/asm-ppc64/iSeries/ItVpdAreas.h b/include/asm-ppc64/iSeries/ItVpdAreas.h
index cfb9925e40a..97cfa6c427e 100644
--- a/include/asm-ppc64/iSeries/ItVpdAreas.h
+++ b/include/asm-ppc64/iSeries/ItVpdAreas.h
@@ -90,4 +90,6 @@ struct ItVpdAreas {
 	void	*xSlicVpdAdrs[ItVpdMaxEntries];// Array of VPD buffers	130-1EF
 };
 
+extern struct ItVpdAreas	itVpdAreas;
+
 #endif /* _ITVPDAREAS_H */
diff --git a/include/asm-ppc64/iSeries/LparData.h b/include/asm-ppc64/iSeries/LparData.h
deleted file mode 100644
index a9fd6e287e6..00000000000
--- a/include/asm-ppc64/iSeries/LparData.h
+++ /dev/null
@@ -1,48 +0,0 @@
-/*
- * LparData.h
- * Copyright (C) 2001  Mike Corrigan IBM Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
- */
-
-#ifndef _LPARDATA_H
-#define _LPARDATA_H
-
-#include <asm/types.h>
-#include <asm/page.h>
-#include <asm/abs_addr.h>
-
-#include <asm/iSeries/ItLpNaca.h>
-#include <asm/iSeries/ItLpRegSave.h>
-#include <asm/iSeries/HvReleaseData.h>
-#include <asm/iSeries/LparMap.h>
-#include <asm/iSeries/ItVpdAreas.h>
-#include <asm/iSeries/ItIplParmsReal.h>
-#include <asm/iSeries/ItExtVpdPanel.h>
-#include <asm/iSeries/ItLpQueue.h>
-#include <asm/iSeries/IoHriProcessorVpd.h>
-
-extern struct LparMap		xLparMap;
-extern struct HvReleaseData	hvReleaseData;
-extern struct ItLpNaca		itLpNaca;
-extern struct ItIplParmsReal	xItIplParmsReal;
-extern struct ItExtVpdPanel	xItExtVpdPanel;
-extern struct IoHriProcessorVpd	xIoHriProcessorVpd[];
-extern struct ItLpQueue		xItLpQueue;
-extern struct ItVpdAreas	itVpdAreas;
-extern u64			xMsVpd[];
-extern struct msChunks		msChunks;
-
-#endif /* _LPARDATA_H */
diff --git a/include/asm-ppc64/iSeries/LparMap.h b/include/asm-ppc64/iSeries/LparMap.h
index 1505c4f0f1b..038e5df7e9f 100644
--- a/include/asm-ppc64/iSeries/LparMap.h
+++ b/include/asm-ppc64/iSeries/LparMap.h
@@ -64,4 +64,6 @@ struct LparMap {
 	u64	xVPN;		// Virtual Page Number (0x000C000000000000)
 };
 
+extern struct LparMap		xLparMap;
+
 #endif /* _LPARMAP_H */
-- 
cgit v1.2.3-70-g09d2


From dd61ce922770b299081c3e729ea65758ed676034 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:37 -0700
Subject: [PATCH] ppc64 iSeries: eliminate some unused inline functions

This patch removes from the iSeries header files a large number of inline
functions that are not used.  It also changes the only caller of a HvCallCfg
function that is outside HvLpConfig.h to its equivalent HvLpConfig function
and no longer includes HvCallCfg.h where it is not needed.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/iSeries_smp.c        |   1 -
 arch/ppc64/kernel/viopath.c            |   3 +-
 include/asm-ppc64/iSeries/HvCallCfg.h  |  53 +---------
 include/asm-ppc64/iSeries/HvLpConfig.h | 188 ---------------------------------
 4 files changed, 3 insertions(+), 242 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/iSeries_smp.c b/arch/ppc64/kernel/iSeries_smp.c
index ff2281b457e..f74386e3163 100644
--- a/arch/ppc64/kernel/iSeries_smp.c
+++ b/arch/ppc64/kernel/iSeries_smp.c
@@ -39,7 +39,6 @@
 #include <asm/smp.h>
 #include <asm/paca.h>
 #include <asm/iSeries/HvCall.h>
-#include <asm/iSeries/HvCallCfg.h>
 #include <asm/time.h>
 #include <asm/ppcdebug.h>
 #include <asm/machdep.h>
diff --git a/arch/ppc64/kernel/viopath.c b/arch/ppc64/kernel/viopath.c
index 4e5c36e81f4..ef524042e49 100644
--- a/arch/ppc64/kernel/viopath.c
+++ b/arch/ppc64/kernel/viopath.c
@@ -46,7 +46,6 @@
 #include <asm/iSeries/ItExtVpdPanel.h>
 #include <asm/iSeries/HvLpEvent.h>
 #include <asm/iSeries/HvLpConfig.h>
-#include <asm/iSeries/HvCallCfg.h>
 #include <asm/iSeries/mf.h>
 #include <asm/iSeries/vio.h>
 
@@ -364,7 +363,7 @@ void vio_set_hostlp(void)
 	 * while we're active
 	 */
 	viopath_ourLp = HvLpConfig_getLpIndex();
-	viopath_hostLp = HvCallCfg_getHostingLpIndex(viopath_ourLp);
+	viopath_hostLp = HvLpConfig_getHostingLpIndex(viopath_ourLp);
 
 	if (viopath_hostLp != HvLpIndexInvalid)
 		vio_setHandler(viomajorsubtype_config, handleConfig);
diff --git a/include/asm-ppc64/iSeries/HvCallCfg.h b/include/asm-ppc64/iSeries/HvCallCfg.h
index aa491c3ad0a..ddf2bb59d73 100644
--- a/include/asm-ppc64/iSeries/HvCallCfg.h
+++ b/include/asm-ppc64/iSeries/HvCallCfg.h
@@ -67,31 +67,11 @@ enum HvCallCfg_ReqQual {
 #define HvCallCfgGetLpExecutionMode			HvCallCfg + 31
 #define HvCallCfgGetHostingLpIndex			HvCallCfg + 32
 
-static inline HvLpIndex	HvCallCfg_getLps(void)
-{
-	return HvCall0(HvCallCfgGetLps);
-}
-
-static inline int HvCallCfg_isBusDedicated(u64 busIndex)
-{
-	return HvCall1(HvCallCfgIsBusDedicated, busIndex);
-}
-
 static inline HvLpIndex	HvCallCfg_getBusOwner(u64 busIndex)
 {
 	return HvCall1(HvCallCfgGetBusOwner, busIndex);
 }
 
-static inline HvLpIndexMap HvCallCfg_getBusAllocation(u64 busIndex)
-{
-	return HvCall1(HvCallCfgGetBusAllocation, busIndex);
-}
-
-static inline HvLpIndexMap HvCallCfg_getActiveLpMap(void)
-{
-	return HvCall0(HvCallCfgGetActiveLpMap);
-}
-
 static inline HvLpVirtualLanIndexMap HvCallCfg_getVirtualLanIndexMap(
 		HvLpIndex lp)
 {
@@ -105,31 +85,12 @@ static inline HvLpVirtualLanIndexMap HvCallCfg_getVirtualLanIndexMap(
 	return retVal;
 }
 
-static inline u64 HvCallCfg_getSystemMsChunks(void)
-{
-	return HvCall0(HvCallCfgGetSystemMsChunks);
-}
-
 static inline u64 HvCallCfg_getMsChunks(HvLpIndex lp,
 		enum HvCallCfg_ReqQual qual)
 {
 	return HvCall2(HvCallCfgGetMsChunks, lp, qual);
 }
 
-static inline u64 HvCallCfg_getMinRuntimeMsChunks(HvLpIndex lp)
-{
-	/*
-	 * NOTE: This function was added in v5r1 so older hypervisors
-	 * will return a -1 value
-	 */
-	return HvCall1(HvCallCfgGetMinRuntimeMsChunks, lp);
-}
-
-static inline u64 HvCallCfg_setMinRuntimeMsChunks(u64 chunks)
-{
-	return HvCall1(HvCallCfgSetMinRuntimeMsChunks, chunks);
-}
-
 static inline u64 HvCallCfg_getSystemPhysicalProcessors(void)
 {
 	return HvCall0(HvCallCfgGetSystemPhysicalProcessors);
@@ -141,14 +102,6 @@ static inline u64 HvCallCfg_getPhysicalProcessors(HvLpIndex lp,
 	return HvCall2(HvCallCfgGetPhysicalProcessors, lp, qual);
 }
 
-static inline u64 HvCallCfg_getConfiguredBusUnitsForInterruptProc(HvLpIndex lp,
-		u16 hvLogicalProcIndex)
-{
-	return HvCall2(HvCallCfgGetConfiguredBusUnitsForIntProc, lp,
-			hvLogicalProcIndex);
-
-}
-
 static inline HvLpSharedPoolIndex HvCallCfg_getSharedPoolIndex(HvLpIndex lp)
 {
 	return HvCall1(HvCallCfgGetSharedPoolIndex, lp);
@@ -164,15 +117,13 @@ static inline u64 HvCallCfg_getSharedProcUnits(HvLpIndex lp,
 
 static inline u64 HvCallCfg_getNumProcsInSharedPool(HvLpSharedPoolIndex sPI)
 {
-	u16 retVal = HvCall1(HvCallCfgGetNumProcsInSharedPool, sPI);
-	return retVal;
+	return (u16)HvCall1(HvCallCfgGetNumProcsInSharedPool, sPI);
 
 }
 
 static inline HvLpIndex	HvCallCfg_getHostingLpIndex(HvLpIndex lp)
 {
-	u64 retVal = HvCall1(HvCallCfgGetHostingLpIndex, lp);
-	return retVal;
+	return HvCall1(HvCallCfgGetHostingLpIndex, lp);
 }
 
 #endif /* _HVCALLCFG_H */
diff --git a/include/asm-ppc64/iSeries/HvLpConfig.h b/include/asm-ppc64/iSeries/HvLpConfig.h
index 9411b5a04d0..78d68281cc8 100644
--- a/include/asm-ppc64/iSeries/HvLpConfig.h
+++ b/include/asm-ppc64/iSeries/HvLpConfig.h
@@ -40,127 +40,16 @@ static inline HvLpIndex	HvLpConfig_getPrimaryLpIndex(void)
 	return itLpNaca.xPrimaryLpIndex;
 }
 
-static inline HvLpIndex	HvLpConfig_getLps(void)
-{
-	return HvCallCfg_getLps();
-}
-
-static inline HvLpIndexMap HvLpConfig_getActiveLpMap(void)
-{
-	return HvCallCfg_getActiveLpMap();
-}
-
-static inline u64 HvLpConfig_getSystemMsMegs(void)
-{
-	return HvCallCfg_getSystemMsChunks() / HVCHUNKSPERMEG;
-}
-
-static inline u64 HvLpConfig_getSystemMsChunks(void)
-{
-	return HvCallCfg_getSystemMsChunks();
-}
-
-static inline u64 HvLpConfig_getSystemMsPages(void)
-{
-	return HvCallCfg_getSystemMsChunks() * HVPAGESPERCHUNK;
-}
-
-static inline u64 HvLpConfig_getMsMegs(void)
-{
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Cur)
-		/ HVCHUNKSPERMEG;
-}
-
 static inline u64 HvLpConfig_getMsChunks(void)
 {
 	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Cur);
 }
 
-static inline u64 HvLpConfig_getMsPages(void)
-{
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Cur)
-		* HVPAGESPERCHUNK;
-}
-
-static inline u64 HvLpConfig_getMinMsMegs(void)
-{
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Min)
-		/ HVCHUNKSPERMEG;
-}
-
-static inline u64 HvLpConfig_getMinMsChunks(void)
-{
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Min);
-}
-
-static inline u64 HvLpConfig_getMinMsPages(void)
-{
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Min)
-		* HVPAGESPERCHUNK;
-}
-
-static inline u64 HvLpConfig_getMinRuntimeMsMegs(void)
-{
-	return HvCallCfg_getMinRuntimeMsChunks(HvLpConfig_getLpIndex())
-		/ HVCHUNKSPERMEG;
-}
-
-static inline u64 HvLpConfig_getMinRuntimeMsChunks(void)
-{
-	return HvCallCfg_getMinRuntimeMsChunks(HvLpConfig_getLpIndex());
-}
-
-static inline u64 HvLpConfig_getMinRuntimeMsPages(void)
-{
-	return HvCallCfg_getMinRuntimeMsChunks(HvLpConfig_getLpIndex())
-		* HVPAGESPERCHUNK;
-}
-
-static inline u64 HvLpConfig_getMaxMsMegs(void)
-{
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Max)
-		/ HVCHUNKSPERMEG;
-}
-
-static inline u64 HvLpConfig_getMaxMsChunks(void)
-{
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Max);
-}
-
-static inline u64 HvLpConfig_getMaxMsPages(void)
-{
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Max)
-		* HVPAGESPERCHUNK;
-}
-
-static inline u64 HvLpConfig_getInitMsMegs(void)
-{
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Init)
-		/ HVCHUNKSPERMEG;
-}
-
-static inline u64 HvLpConfig_getInitMsChunks(void)
-{
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Init);
-}
-
-static inline u64 HvLpConfig_getInitMsPages(void)
-{
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Init)
-		* HVPAGESPERCHUNK;
-}
-
 static inline u64 HvLpConfig_getSystemPhysicalProcessors(void)
 {
 	return HvCallCfg_getSystemPhysicalProcessors();
 }
 
-static inline u64 HvLpConfig_getSystemLogicalProcessors(void)
-{
-	return HvCallCfg_getSystemPhysicalProcessors()
-		* (/*getPaca()->getSecondaryThreadCount() +*/ 1);
-}
-
 static inline u64 HvLpConfig_getNumProcsInSharedPool(HvLpSharedPoolIndex sPI)
 {
 	return HvCallCfg_getNumProcsInSharedPool(sPI);
@@ -172,13 +61,6 @@ static inline u64 HvLpConfig_getPhysicalProcessors(void)
 			HvCallCfg_Cur);
 }
 
-static inline u64 HvLpConfig_getLogicalProcessors(void)
-{
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
-			HvCallCfg_Cur)
-		* (/*getPaca()->getSecondaryThreadCount() +*/ 1);
-}
-
 static inline HvLpSharedPoolIndex HvLpConfig_getSharedPoolIndex(void)
 {
 	return HvCallCfg_getSharedPoolIndex(HvLpConfig_getLpIndex());
@@ -190,57 +72,18 @@ static inline u64 HvLpConfig_getSharedProcUnits(void)
 			HvCallCfg_Cur);
 }
 
-static inline u64 HvLpConfig_getMinSharedProcUnits(void)
-{
-	return HvCallCfg_getSharedProcUnits(HvLpConfig_getLpIndex(),
-			HvCallCfg_Min);
-}
-
 static inline u64 HvLpConfig_getMaxSharedProcUnits(void)
 {
 	return HvCallCfg_getSharedProcUnits(HvLpConfig_getLpIndex(),
 			HvCallCfg_Max);
 }
 
-static inline u64 HvLpConfig_getMinPhysicalProcessors(void)
-{
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
-			HvCallCfg_Min);
-}
-
-static inline u64 HvLpConfig_getMinLogicalProcessors(void)
-{
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
-			HvCallCfg_Min)
-		* (/*getPaca()->getSecondaryThreadCount() +*/ 1);
-}
-
 static inline u64 HvLpConfig_getMaxPhysicalProcessors(void)
 {
 	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
 			HvCallCfg_Max);
 }
 
-static inline u64 HvLpConfig_getMaxLogicalProcessors(void)
-{
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
-			HvCallCfg_Max)
-		* (/*getPaca()->getSecondaryThreadCount() +*/ 1);
-}
-
-static inline u64 HvLpConfig_getInitPhysicalProcessors(void)
-{
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
-			HvCallCfg_Init);
-}
-
-static inline u64 HvLpConfig_getInitLogicalProcessors(void)
-{
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
-			HvCallCfg_Init)
-		* (/*getPaca()->getSecondaryThreadCount() +*/ 1);
-}
-
 static inline HvLpVirtualLanIndexMap HvLpConfig_getVirtualLanIndexMap(void)
 {
 	return HvCallCfg_getVirtualLanIndexMap(HvLpConfig_getLpIndex_outline());
@@ -252,37 +95,6 @@ static inline HvLpVirtualLanIndexMap HvLpConfig_getVirtualLanIndexMapForLp(
 	return HvCallCfg_getVirtualLanIndexMap(lp);
 }
 
-static inline HvLpIndex HvLpConfig_getBusOwner(HvBusNumber busNumber)
-{
-	return HvCallCfg_getBusOwner(busNumber);
-}
-
-static inline int HvLpConfig_isBusDedicated(HvBusNumber busNumber)
-{
-	return HvCallCfg_isBusDedicated(busNumber);
-}
-
-static inline HvLpIndexMap HvLpConfig_getBusAllocation(HvBusNumber busNumber)
-{
-	return HvCallCfg_getBusAllocation(busNumber);
-}
-
-/* returns the absolute real address of the load area */
-static inline u64 HvLpConfig_getLoadAddress(void)
-{
-	return itLpNaca.xLoadAreaAddr & 0x7fffffffffffffff;
-}
-
-static inline u64 HvLpConfig_getLoadPages(void)
-{
-	return itLpNaca.xLoadAreaChunks * HVPAGESPERCHUNK;
-}
-
-static inline int HvLpConfig_isBusOwnedByThisLp(HvBusNumber busNumber)
-{
-	return (HvLpConfig_getBusOwner(busNumber) == HvLpConfig_getLpIndex());
-}
-
 static inline int HvLpConfig_doLpsCommunicateOnVirtualLan(HvLpIndex lp1,
 		HvLpIndex lp2)
 {
-- 
cgit v1.2.3-70-g09d2


From 2310c977a9b7a4d08fbaa74e694b49b38d49f5ae Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:38 -0700
Subject: [PATCH] ppc64 iSeries: remove HvCallCfg.h

Now that the only users of things in HvCallCfg.h are in HvLpConfig.h, merge in
the bit we need and remove HvCallCfg.h.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-ppc64/iSeries/HvCallCfg.h  | 129 ---------------------------------
 include/asm-ppc64/iSeries/HvLpConfig.h |  59 ++++++++++-----
 2 files changed, 42 insertions(+), 146 deletions(-)
 delete mode 100644 include/asm-ppc64/iSeries/HvCallCfg.h

(limited to 'include')

diff --git a/include/asm-ppc64/iSeries/HvCallCfg.h b/include/asm-ppc64/iSeries/HvCallCfg.h
deleted file mode 100644
index ddf2bb59d73..00000000000
--- a/include/asm-ppc64/iSeries/HvCallCfg.h
+++ /dev/null
@@ -1,129 +0,0 @@
-/*
- * HvCallCfg.h
- * Copyright (C) 2001  Mike Corrigan IBM Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
- */
-/*
- * This file contains the "hypervisor call" interface which is used to
- * drive the hypervisor from the OS.
- */
-#ifndef _HVCALLCFG_H
-#define _HVCALLCFG_H
-
-#include <asm/iSeries/HvCallSc.h>
-#include <asm/iSeries/HvTypes.h>
-
-enum HvCallCfg_ReqQual {
-	HvCallCfg_Cur	= 0,
-	HvCallCfg_Init	= 1,
-	HvCallCfg_Max	= 2,
-	HvCallCfg_Min	= 3
-};
-
-#define HvCallCfgGetLps					HvCallCfg +  0
-#define HvCallCfgGetActiveLpMap				HvCallCfg +  1
-#define HvCallCfgGetLpVrmIndex				HvCallCfg +  2
-#define HvCallCfgGetLpMinSupportedPlicVrmIndex		HvCallCfg +  3
-#define HvCallCfgGetLpMinCompatablePlicVrmIndex		HvCallCfg +  4
-#define HvCallCfgGetLpVrmName				HvCallCfg +  5
-#define HvCallCfgGetSystemPhysicalProcessors		HvCallCfg +  6
-#define HvCallCfgGetPhysicalProcessors			HvCallCfg +  7
-#define HvCallCfgGetSystemMsChunks			HvCallCfg +  8
-#define HvCallCfgGetMsChunks				HvCallCfg +  9
-#define HvCallCfgGetInteractivePercentage		HvCallCfg + 10
-#define HvCallCfgIsBusDedicated				HvCallCfg + 11
-#define HvCallCfgGetBusOwner				HvCallCfg + 12
-#define HvCallCfgGetBusAllocation			HvCallCfg + 13
-#define HvCallCfgGetBusUnitOwner			HvCallCfg + 14
-#define HvCallCfgGetBusUnitAllocation			HvCallCfg + 15
-#define HvCallCfgGetVirtualBusPool			HvCallCfg + 16
-#define HvCallCfgGetBusUnitInterruptProc		HvCallCfg + 17
-#define HvCallCfgGetConfiguredBusUnitsForIntProc	HvCallCfg + 18
-#define HvCallCfgGetRioSanBusPool			HvCallCfg + 19
-#define HvCallCfgGetSharedPoolIndex			HvCallCfg + 20
-#define HvCallCfgGetSharedProcUnits			HvCallCfg + 21
-#define HvCallCfgGetNumProcsInSharedPool		HvCallCfg + 22
-#define HvCallCfgRouter23				HvCallCfg + 23
-#define HvCallCfgRouter24				HvCallCfg + 24
-#define HvCallCfgRouter25				HvCallCfg + 25
-#define HvCallCfgRouter26				HvCallCfg + 26
-#define HvCallCfgRouter27				HvCallCfg + 27
-#define HvCallCfgGetMinRuntimeMsChunks			HvCallCfg + 28
-#define HvCallCfgSetMinRuntimeMsChunks			HvCallCfg + 29
-#define HvCallCfgGetVirtualLanIndexMap			HvCallCfg + 30
-#define HvCallCfgGetLpExecutionMode			HvCallCfg + 31
-#define HvCallCfgGetHostingLpIndex			HvCallCfg + 32
-
-static inline HvLpIndex	HvCallCfg_getBusOwner(u64 busIndex)
-{
-	return HvCall1(HvCallCfgGetBusOwner, busIndex);
-}
-
-static inline HvLpVirtualLanIndexMap HvCallCfg_getVirtualLanIndexMap(
-		HvLpIndex lp)
-{
-	/*
-	 * This is a new function in V5R1 so calls to this on older
-	 * hypervisors will return -1
-	 */
-	u64 retVal = HvCall1(HvCallCfgGetVirtualLanIndexMap, lp);
-	if (retVal == -1)
-		retVal = 0;
-	return retVal;
-}
-
-static inline u64 HvCallCfg_getMsChunks(HvLpIndex lp,
-		enum HvCallCfg_ReqQual qual)
-{
-	return HvCall2(HvCallCfgGetMsChunks, lp, qual);
-}
-
-static inline u64 HvCallCfg_getSystemPhysicalProcessors(void)
-{
-	return HvCall0(HvCallCfgGetSystemPhysicalProcessors);
-}
-
-static inline u64 HvCallCfg_getPhysicalProcessors(HvLpIndex lp,
-		enum HvCallCfg_ReqQual qual)
-{
-	return HvCall2(HvCallCfgGetPhysicalProcessors, lp, qual);
-}
-
-static inline HvLpSharedPoolIndex HvCallCfg_getSharedPoolIndex(HvLpIndex lp)
-{
-	return HvCall1(HvCallCfgGetSharedPoolIndex, lp);
-
-}
-
-static inline u64 HvCallCfg_getSharedProcUnits(HvLpIndex lp,
-		enum HvCallCfg_ReqQual qual)
-{
-	return HvCall2(HvCallCfgGetSharedProcUnits, lp, qual);
-
-}
-
-static inline u64 HvCallCfg_getNumProcsInSharedPool(HvLpSharedPoolIndex sPI)
-{
-	return (u16)HvCall1(HvCallCfgGetNumProcsInSharedPool, sPI);
-
-}
-
-static inline HvLpIndex	HvCallCfg_getHostingLpIndex(HvLpIndex lp)
-{
-	return HvCall1(HvCallCfgGetHostingLpIndex, lp);
-}
-
-#endif /* _HVCALLCFG_H */
diff --git a/include/asm-ppc64/iSeries/HvLpConfig.h b/include/asm-ppc64/iSeries/HvLpConfig.h
index 78d68281cc8..f1cf1e70ca3 100644
--- a/include/asm-ppc64/iSeries/HvLpConfig.h
+++ b/include/asm-ppc64/iSeries/HvLpConfig.h
@@ -24,10 +24,26 @@
  * to determine which resources should be allocated to each partition.
  */
 
-#include <asm/iSeries/HvCallCfg.h>
+#include <asm/iSeries/HvCallSc.h>
 #include <asm/iSeries/HvTypes.h>
 #include <asm/iSeries/ItLpNaca.h>
 
+enum {
+	HvCallCfg_Cur	= 0,
+	HvCallCfg_Init	= 1,
+	HvCallCfg_Max	= 2,
+	HvCallCfg_Min	= 3
+};
+
+#define HvCallCfgGetSystemPhysicalProcessors		HvCallCfg +  6
+#define HvCallCfgGetPhysicalProcessors			HvCallCfg +  7
+#define HvCallCfgGetMsChunks				HvCallCfg +  9
+#define HvCallCfgGetSharedPoolIndex			HvCallCfg + 20
+#define HvCallCfgGetSharedProcUnits			HvCallCfg + 21
+#define HvCallCfgGetNumProcsInSharedPool		HvCallCfg + 22
+#define HvCallCfgGetVirtualLanIndexMap			HvCallCfg + 30
+#define HvCallCfgGetHostingLpIndex			HvCallCfg + 32
+
 extern HvLpIndex HvLpConfig_getLpIndex_outline(void);
 
 static inline HvLpIndex	HvLpConfig_getLpIndex(void)
@@ -42,72 +58,81 @@ static inline HvLpIndex	HvLpConfig_getPrimaryLpIndex(void)
 
 static inline u64 HvLpConfig_getMsChunks(void)
 {
-	return HvCallCfg_getMsChunks(HvLpConfig_getLpIndex(), HvCallCfg_Cur);
+	return HvCall2(HvCallCfgGetMsChunks, HvLpConfig_getLpIndex(),
+			HvCallCfg_Cur);
 }
 
 static inline u64 HvLpConfig_getSystemPhysicalProcessors(void)
 {
-	return HvCallCfg_getSystemPhysicalProcessors();
+	return HvCall0(HvCallCfgGetSystemPhysicalProcessors);
 }
 
 static inline u64 HvLpConfig_getNumProcsInSharedPool(HvLpSharedPoolIndex sPI)
 {
-	return HvCallCfg_getNumProcsInSharedPool(sPI);
+	return (u16)HvCall1(HvCallCfgGetNumProcsInSharedPool, sPI);
 }
 
 static inline u64 HvLpConfig_getPhysicalProcessors(void)
 {
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
+	return HvCall2(HvCallCfgGetPhysicalProcessors, HvLpConfig_getLpIndex(),
 			HvCallCfg_Cur);
 }
 
 static inline HvLpSharedPoolIndex HvLpConfig_getSharedPoolIndex(void)
 {
-	return HvCallCfg_getSharedPoolIndex(HvLpConfig_getLpIndex());
+	return HvCall1(HvCallCfgGetSharedPoolIndex, HvLpConfig_getLpIndex());
 }
 
 static inline u64 HvLpConfig_getSharedProcUnits(void)
 {
-	return HvCallCfg_getSharedProcUnits(HvLpConfig_getLpIndex(),
+	return HvCall2(HvCallCfgGetSharedProcUnits, HvLpConfig_getLpIndex(),
 			HvCallCfg_Cur);
 }
 
 static inline u64 HvLpConfig_getMaxSharedProcUnits(void)
 {
-	return HvCallCfg_getSharedProcUnits(HvLpConfig_getLpIndex(),
+	return HvCall2(HvCallCfgGetSharedProcUnits, HvLpConfig_getLpIndex(),
 			HvCallCfg_Max);
 }
 
 static inline u64 HvLpConfig_getMaxPhysicalProcessors(void)
 {
-	return HvCallCfg_getPhysicalProcessors(HvLpConfig_getLpIndex(),
+	return HvCall2(HvCallCfgGetPhysicalProcessors, HvLpConfig_getLpIndex(),
 			HvCallCfg_Max);
 }
 
-static inline HvLpVirtualLanIndexMap HvLpConfig_getVirtualLanIndexMap(void)
+static inline HvLpVirtualLanIndexMap HvLpConfig_getVirtualLanIndexMapForLp(
+		HvLpIndex lp)
 {
-	return HvCallCfg_getVirtualLanIndexMap(HvLpConfig_getLpIndex_outline());
+	/*
+	 * This is a new function in V5R1 so calls to this on older
+	 * hypervisors will return -1
+	 */
+	u64 retVal = HvCall1(HvCallCfgGetVirtualLanIndexMap, lp);
+	if (retVal == -1)
+		retVal = 0;
+	return retVal;
 }
 
-static inline HvLpVirtualLanIndexMap HvLpConfig_getVirtualLanIndexMapForLp(
-		HvLpIndex lp)
+static inline HvLpVirtualLanIndexMap HvLpConfig_getVirtualLanIndexMap(void)
 {
-	return HvCallCfg_getVirtualLanIndexMap(lp);
+	return HvLpConfig_getVirtualLanIndexMapForLp(
+			HvLpConfig_getLpIndex_outline());
 }
 
 static inline int HvLpConfig_doLpsCommunicateOnVirtualLan(HvLpIndex lp1,
 		HvLpIndex lp2)
 {
 	HvLpVirtualLanIndexMap virtualLanIndexMap1 =
-		HvCallCfg_getVirtualLanIndexMap(lp1);
+		HvLpConfig_getVirtualLanIndexMapForLp(lp1);
 	HvLpVirtualLanIndexMap virtualLanIndexMap2 =
-		HvCallCfg_getVirtualLanIndexMap(lp2);
+		HvLpConfig_getVirtualLanIndexMapForLp(lp2);
 	return ((virtualLanIndexMap1 & virtualLanIndexMap2) != 0);
 }
 
 static inline HvLpIndex HvLpConfig_getHostingLpIndex(HvLpIndex lp)
 {
-	return HvCallCfg_getHostingLpIndex(lp);
+	return HvCall1(HvCallCfgGetHostingLpIndex, lp);
 }
 
 #endif /* _HVLPCONFIG_H */
-- 
cgit v1.2.3-70-g09d2


From c92877e0a079fe5a488cb244c7afb05137702b49 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:39 -0700
Subject: [PATCH] ppc64 iSeries: cleanup ItLpQueue.h

Just white space cleaups and move process_iSeries_events into its only caller.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/idle.c              |  5 +++++
 include/asm-ppc64/iSeries/ItLpQueue.h | 16 ++++------------
 2 files changed, 9 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/idle.c b/arch/ppc64/kernel/idle.c
index ff8a7db142d..bdf13b4dc1c 100644
--- a/arch/ppc64/kernel/idle.c
+++ b/arch/ppc64/kernel/idle.c
@@ -42,6 +42,11 @@ static int (*idle_loop)(void);
 static unsigned long maxYieldTime = 0;
 static unsigned long minYieldTime = 0xffffffffffffffffUL;
 
+static inline void process_iSeries_events(void)
+{
+	asm volatile ("li 0,0x5555; sc" : : : "r0", "r3");
+}
+
 static void yield_shared_processor(void)
 {
 	unsigned long tb;
diff --git a/include/asm-ppc64/iSeries/ItLpQueue.h b/include/asm-ppc64/iSeries/ItLpQueue.h
index b2bdef42959..393299e04d7 100644
--- a/include/asm-ppc64/iSeries/ItLpQueue.h
+++ b/include/asm-ppc64/iSeries/ItLpQueue.h
@@ -64,9 +64,9 @@ struct ItLpQueue {
 	u8	xPlicStatus;		// 0x01 DedicatedIo or DedicatedLp or NotUsed
 	u16	xSlicLogicalProcIndex;	// 0x02 Logical Proc Index for correlation
 	u8	xPlicRsvd[12];		// 0x04
-	char*	xSlicCurEventPtr;	// 0x10
-	char*	xSlicLastValidEventPtr;	// 0x18
-	char*	xSlicEventStackPtr;	// 0x20
+	char	*xSlicCurEventPtr;	// 0x10
+	char	*xSlicLastValidEventPtr; // 0x18
+	char	*xSlicEventStackPtr;	// 0x20
 	u8	xIndex;			// 0x28 unique sequential index.
 	u8	xSlicRsvd[3];		// 0x29-2b
 	u32	xInUseWord;		// 0x2C
@@ -76,17 +76,9 @@ struct ItLpQueue {
 
 extern struct ItLpQueue xItLpQueue;
 
-extern struct HvLpEvent * ItLpQueue_getNextLpEvent(struct ItLpQueue *);
+extern struct HvLpEvent *ItLpQueue_getNextLpEvent(struct ItLpQueue *);
 extern int ItLpQueue_isLpIntPending(struct ItLpQueue *);
 extern unsigned ItLpQueue_process(struct ItLpQueue *, struct pt_regs *);
 extern void ItLpQueue_clearValid(struct HvLpEvent *);
 
-static __inline__ void process_iSeries_events(void)
-{
-	__asm__ __volatile__ (
-	"	li	0,0x5555	\n\
-		sc"
-	: : : "r0", "r3");
-}
-
 #endif /* _ITLPQUEUE_H */
-- 
cgit v1.2.3-70-g09d2


From 4a5304f5ba04eff979cd108309eae7a00f3fff77 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:40 -0700
Subject: [PATCH] ppc64 iSeries: tidy up some includes and HvCall.h

This patch removes some unused bits from HvCall.h and some unneeded #includes
from other files.  Also includes ItLpQueue.h in paca.h in preference to a stub
declaration of struct ItLpQueue.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/asm-offsets.c    |  1 -
 arch/ppc64/kernel/iSeries_pci.c    |  1 -
 arch/ppc64/kernel/mf.c             |  1 -
 arch/ppc64/kernel/rtc.c            |  1 -
 include/asm-ppc64/iSeries/HvCall.h | 84 +-------------------------------------
 include/asm-ppc64/paca.h           |  2 +-
 6 files changed, 3 insertions(+), 87 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/asm-offsets.c b/arch/ppc64/kernel/asm-offsets.c
index 0094ac79a18..abb9e5b5da0 100644
--- a/arch/ppc64/kernel/asm-offsets.c
+++ b/arch/ppc64/kernel/asm-offsets.c
@@ -31,7 +31,6 @@
 
 #include <asm/paca.h>
 #include <asm/lppaca.h>
-#include <asm/iSeries/ItLpQueue.h>
 #include <asm/iSeries/HvLpEvent.h>
 #include <asm/rtas.h>
 #include <asm/cputable.h>
diff --git a/arch/ppc64/kernel/iSeries_pci.c b/arch/ppc64/kernel/iSeries_pci.c
index 9829cd4666d..1bff515bd29 100644
--- a/arch/ppc64/kernel/iSeries_pci.c
+++ b/arch/ppc64/kernel/iSeries_pci.c
@@ -38,7 +38,6 @@
 #include <asm/iommu.h>
 
 #include <asm/iSeries/HvCallPci.h>
-#include <asm/iSeries/HvCallSm.h>
 #include <asm/iSeries/HvCallXm.h>
 #include <asm/iSeries/iSeries_irq.h>
 #include <asm/iSeries/iSeries_pci.h>
diff --git a/arch/ppc64/kernel/mf.c b/arch/ppc64/kernel/mf.c
index 5aca7e8005a..d98bebf7042 100644
--- a/arch/ppc64/kernel/mf.c
+++ b/arch/ppc64/kernel/mf.c
@@ -40,7 +40,6 @@
 #include <asm/iSeries/vio.h>
 #include <asm/iSeries/mf.h>
 #include <asm/iSeries/HvLpConfig.h>
-#include <asm/iSeries/ItSpCommArea.h>
 #include <asm/iSeries/ItLpQueue.h>
 
 /*
diff --git a/arch/ppc64/kernel/rtc.c b/arch/ppc64/kernel/rtc.c
index 4cdafb4007b..de02aedbe08 100644
--- a/arch/ppc64/kernel/rtc.c
+++ b/arch/ppc64/kernel/rtc.c
@@ -44,7 +44,6 @@
 
 #include <asm/iSeries/mf.h>
 #include <asm/machdep.h>
-#include <asm/iSeries/ItSpCommArea.h>
 
 extern int piranha_simulator;
 
diff --git a/include/asm-ppc64/iSeries/HvCall.h b/include/asm-ppc64/iSeries/HvCall.h
index c903387e912..c3f19475c0d 100644
--- a/include/asm-ppc64/iSeries/HvCall.h
+++ b/include/asm-ppc64/iSeries/HvCall.h
@@ -27,48 +27,6 @@
 #include <asm/iSeries/HvTypes.h>
 #include <asm/paca.h>
 
-/*
-enum HvCall_ReturnCode
-{
-	HvCall_Good		= 0,
-	HvCall_Partial		= 1,
-	HvCall_NotOwned		= 2,
-	HvCall_NotFreed		= 3,
-	HvCall_UnspecifiedError	= 4
-};
-
-enum HvCall_TypeOfSIT
-{
-	HvCall_ReduceOnly	= 0,
-	HvCall_Unconditional	= 1
-};
-
-enum HvCall_TypeOfYield
-{
-	HvCall_YieldTimed	= 0,	// Yield until specified time
-	HvCall_YieldToActive	= 1,	// Yield until all active procs have run
-	HvCall_YieldToProc	= 2	// Yield until the specified processor has run
-};
-
-enum HvCall_InterruptMasks
-{
-	HvCall_MaskIPI		= 0x00000001,
-	HvCall_MaskLpEvent	= 0x00000002,
-	HvCall_MaskLpProd	= 0x00000004,
-	HvCall_MaskTimeout	= 0x00000008
-};
-
-enum HvCall_VaryOffChunkRc
-{
-	HvCall_VaryOffSucceeded		= 0,
-	HvCall_VaryOffWithdrawn		= 1,
-	HvCall_ChunkInLoadArea		= 2,
-	HvCall_ChunkInHPT		= 3,
-	HvCall_ChunkNotAccessible	= 4,
-	HvCall_ChunkInUse		= 5
-};
-*/
-
 /* Type of yield for HvCallBaseYieldProcessor */
 #define HvCall_YieldTimed	0	/* Yield until specified time (tb) */
 #define HvCall_YieldToActive	1	/* Yield until all active procs have run */
@@ -139,35 +97,12 @@ static inline void HvCall_setEnabledInterrupts(u64 enabledInterrupts)
 	HvCall1(HvCallBaseSetEnabledInterrupts, enabledInterrupts);
 }
 
-static inline void HvCall_clearLogBuffer(HvLpIndex lpindex)
-{
-	HvCall1(HvCallBaseClearLogBuffer, lpindex);
-}
-
-static inline u32 HvCall_getLogBufferCodePage(HvLpIndex lpindex)
-{
-	u32 retVal = HvCall1(HvCallBaseGetLogBufferCodePage, lpindex);
-	return retVal;
-}
-
-static inline int HvCall_getLogBufferFormat(HvLpIndex lpindex)
-{
-	int retVal = HvCall1(HvCallBaseGetLogBufferFormat, lpindex);
-	return retVal;
-}
-
-static inline u32 HvCall_getLogBufferLength(HvLpIndex lpindex)
-{
-	u32 retVal = HvCall1(HvCallBaseGetLogBufferLength, lpindex);
-	return retVal;
-}
-
-static inline void HvCall_setLogBufferFormatAndCodepage(int format, u32 codePage)
+static inline void HvCall_setLogBufferFormatAndCodepage(int format,
+		u32 codePage)
 {
 	HvCall2(HvCallBaseSetLogBufferFormatAndCodePage, format, codePage);
 }
 
-extern int HvCall_readLogBuffer(HvLpIndex lpindex, void *buffer, u64 bufLen);
 extern void HvCall_writeLogBuffer(const void *buffer, u64 bufLen);
 
 static inline void HvCall_sendIPI(struct paca_struct *targetPaca)
@@ -175,19 +110,4 @@ static inline void HvCall_sendIPI(struct paca_struct *targetPaca)
 	HvCall1(HvCallBaseSendIPI, targetPaca->paca_index);
 }
 
-static inline void HvCall_terminateMachineSrc(void)
-{
-	HvCall0(HvCallBaseTerminateMachineSrc);
-}
-
-static inline void HvCall_setDABR(unsigned long val)
-{
-	HvCall1(HvCallCcSetDABR, val);
-}
-
-static inline void HvCall_setDebugBus(unsigned long val)
-{
-	HvCall1(HvCallBaseSetDebugBus, val);
-}
-
 #endif /* _HVCALL_H */
diff --git a/include/asm-ppc64/paca.h b/include/asm-ppc64/paca.h
index 1a0223b5f3b..ae76cae1483 100644
--- a/include/asm-ppc64/paca.h
+++ b/include/asm-ppc64/paca.h
@@ -20,13 +20,13 @@
 #include	<asm/types.h>
 #include	<asm/lppaca.h>
 #include	<asm/iSeries/ItLpRegSave.h>
+#include	<asm/iSeries/ItLpQueue.h>
 #include	<asm/mmu.h>
 
 register struct paca_struct *local_paca asm("r13");
 #define get_paca()	local_paca
 
 struct task_struct;
-struct ItLpQueue;
 
 /*
  * Defines the layout of the paca.
-- 
cgit v1.2.3-70-g09d2


From c670b1acd0ed0d9f7a27154759a9825cb5012ae4 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:41 -0700
Subject: [PATCH] ppc64 iSeries: misc header cleanups

Last of this round of the iSeries header cleanups
	- don't have two defines for the same thing (HvMaxArchitectedLps
	  and HvMaxArchitectedVirtualLans)
	- HvCallSc.h only needs linux/types.h
	- remove unused struct definition
	- add "extern" to some more function declarations

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/viopath.c            |  4 ++--
 include/asm-ppc64/iSeries/HvCallSc.h   | 34 +++++++++++++++++-----------------
 include/asm-ppc64/iSeries/HvTypes.h    |  8 +++-----
 include/asm-ppc64/iSeries/ItVpdAreas.h |  6 ------
 4 files changed, 22 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/viopath.c b/arch/ppc64/kernel/viopath.c
index ef524042e49..2a6c4f01c45 100644
--- a/arch/ppc64/kernel/viopath.c
+++ b/arch/ppc64/kernel/viopath.c
@@ -485,7 +485,7 @@ int viopath_open(HvLpIndex remoteLp, int subtype, int numReq)
 	unsigned long flags;
 	int tempNumAllocated;
 
-	if ((remoteLp >= HvMaxArchitectedLps) || (remoteLp == HvLpIndexInvalid))
+	if ((remoteLp >= HVMAXARCHITECTEDLPS) || (remoteLp == HvLpIndexInvalid))
 		return -EINVAL;
 
 	subtype = subtype >> VIOMAJOR_SUBTYPE_SHIFT;
@@ -556,7 +556,7 @@ int viopath_close(HvLpIndex remoteLp, int subtype, int numReq)
 	int numOpen;
 	struct alloc_parms parms;
 
-	if ((remoteLp >= HvMaxArchitectedLps) || (remoteLp == HvLpIndexInvalid))
+	if ((remoteLp >= HVMAXARCHITECTEDLPS) || (remoteLp == HvLpIndexInvalid))
 		return -EINVAL;
 
 	subtype = subtype >> VIOMAJOR_SUBTYPE_SHIFT;
diff --git a/include/asm-ppc64/iSeries/HvCallSc.h b/include/asm-ppc64/iSeries/HvCallSc.h
index c32145f9ec7..a62cef3822f 100644
--- a/include/asm-ppc64/iSeries/HvCallSc.h
+++ b/include/asm-ppc64/iSeries/HvCallSc.h
@@ -19,7 +19,7 @@
 #ifndef _HVCALLSC_H
 #define _HVCALLSC_H
 
-#include <asm/iSeries/HvTypes.h>
+#include <linux/types.h>
 
 #define HvCallBase		0x8000000000000000ul
 #define HvCallCc		0x8001000000000000ul
@@ -30,22 +30,22 @@
 #define HvCallSm		0x8007000000000000ul
 #define HvCallXm		0x8009000000000000ul
 
-u64 HvCall0(u64);
-u64 HvCall1(u64, u64);
-u64 HvCall2(u64, u64, u64);
-u64 HvCall3(u64, u64, u64, u64);
-u64 HvCall4(u64, u64, u64, u64, u64);
-u64 HvCall5(u64, u64, u64, u64, u64, u64);
-u64 HvCall6(u64, u64, u64, u64, u64, u64, u64);
-u64 HvCall7(u64, u64, u64, u64, u64, u64, u64, u64);
+extern u64 HvCall0(u64);
+extern u64 HvCall1(u64, u64);
+extern u64 HvCall2(u64, u64, u64);
+extern u64 HvCall3(u64, u64, u64, u64);
+extern u64 HvCall4(u64, u64, u64, u64, u64);
+extern u64 HvCall5(u64, u64, u64, u64, u64, u64);
+extern u64 HvCall6(u64, u64, u64, u64, u64, u64, u64);
+extern u64 HvCall7(u64, u64, u64, u64, u64, u64, u64, u64);
 
-u64 HvCall0Ret16(u64, void *);
-u64 HvCall1Ret16(u64, void *, u64);
-u64 HvCall2Ret16(u64, void *, u64, u64);
-u64 HvCall3Ret16(u64, void *, u64, u64, u64);
-u64 HvCall4Ret16(u64, void *, u64, u64, u64, u64);
-u64 HvCall5Ret16(u64, void *, u64, u64, u64, u64, u64);
-u64 HvCall6Ret16(u64, void *, u64, u64, u64, u64, u64, u64);
-u64 HvCall7Ret16(u64, void *, u64, u64 ,u64 ,u64 ,u64 ,u64 ,u64);
+extern u64 HvCall0Ret16(u64, void *);
+extern u64 HvCall1Ret16(u64, void *, u64);
+extern u64 HvCall2Ret16(u64, void *, u64, u64);
+extern u64 HvCall3Ret16(u64, void *, u64, u64, u64);
+extern u64 HvCall4Ret16(u64, void *, u64, u64, u64, u64);
+extern u64 HvCall5Ret16(u64, void *, u64, u64, u64, u64, u64);
+extern u64 HvCall6Ret16(u64, void *, u64, u64, u64, u64, u64, u64);
+extern u64 HvCall7Ret16(u64, void *, u64, u64 ,u64 ,u64 ,u64 ,u64 ,u64);
 
 #endif /* _HVCALLSC_H */
diff --git a/include/asm-ppc64/iSeries/HvTypes.h b/include/asm-ppc64/iSeries/HvTypes.h
index 111191ddb18..b1ef2b4cb3e 100644
--- a/include/asm-ppc64/iSeries/HvTypes.h
+++ b/include/asm-ppc64/iSeries/HvTypes.h
@@ -40,14 +40,14 @@ typedef u64	HvIoToken;
 typedef u8	HvLpName[8];
 typedef u32	HvIoId;
 typedef u64	HvRealMemoryIndex;
-typedef u32	HvLpIndexMap;	/* Must hold HvMaxArchitectedLps bits!!! */
+typedef u32	HvLpIndexMap;	/* Must hold HVMAXARCHITECTEDLPS bits!!! */
 typedef u16	HvLpVrmIndex;
 typedef u32	HvXmGenerationId;
 typedef u8	HvLpBusPool;
 typedef u8	HvLpSharedPoolIndex;
 typedef u16	HvLpSharedProcUnitsX100;
 typedef u8	HvLpVirtualLanIndex;
-typedef u16	HvLpVirtualLanIndexMap;	/* Must hold HvMaxArchitectedVirtualLans bits!!! */
+typedef u16	HvLpVirtualLanIndexMap;	/* Must hold HVMAXARCHITECTEDVIRTUALLANS bits!!! */
 typedef u16	HvBusNumber;	/* Hypervisor Bus Number */
 typedef u8	HvSubBusNumber;	/* Hypervisor SubBus Number */
 typedef u8	HvAgentId;	/* Hypervisor DevFn */
@@ -66,15 +66,13 @@ typedef u8	HvAgentId;	/* Hypervisor DevFn */
 #define HVPAGESPERMEG			256
 #define HVPAGESPERCHUNK			64
 
-#define HvMaxArchitectedLps		((HvLpIndex)HVMAXARCHITECTEDLPS)
-#define HvMaxArchitectedVirtualLans	((HvLpVirtualLanIndex)16)
 #define HvLpIndexInvalid		((HvLpIndex)0xff)
 
 /*
  * Enums for the sub-components under PLIC
  * Used in HvCall  and HvPrimaryCall
  */
-enum   HvCallCompIds {
+enum {
 	HvCallCompId = 0,
 	HvCallCpuCtlsCompId = 1,
 	HvCallCfgCompId = 2,
diff --git a/include/asm-ppc64/iSeries/ItVpdAreas.h b/include/asm-ppc64/iSeries/ItVpdAreas.h
index 97cfa6c427e..71b3ad24f95 100644
--- a/include/asm-ppc64/iSeries/ItVpdAreas.h
+++ b/include/asm-ppc64/iSeries/ItVpdAreas.h
@@ -61,12 +61,6 @@
 #define ItVpdAreasMaxSlotLabels			192
 
 
-struct SlicVpdAdrs {
-	u32	pad1;
-	void	*vpdAddr;
-};
-
-
 struct ItVpdAreas {
 	u32	xSlicDesc;		// Descriptor			000-003
 	u16	xSlicSize;		// Size of this control block	004-005
-- 
cgit v1.2.3-70-g09d2


From ea7190d0afebbd922eeb13ee6a8148b17964b1b2 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:42 -0700
Subject: [PATCH] ppc64 iSeries: remove iSeries_pci_reset.c

The file arch/ppc64/kernel/iSeries_pci_reset contains only one function that
is not use anywhere (any more).  Remove it.  This function is the only user of
the ReturnCode member of iSeries_Device_Node, so remove that as well.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/Makefile              |   2 +-
 arch/ppc64/kernel/iSeries_pci_reset.c   | 104 --------------------------------
 include/asm-ppc64/iSeries/iSeries_pci.h |   3 -
 3 files changed, 1 insertion(+), 108 deletions(-)
 delete mode 100644 arch/ppc64/kernel/iSeries_pci_reset.c

(limited to 'include')

diff --git a/arch/ppc64/kernel/Makefile b/arch/ppc64/kernel/Makefile
index 96d90b0c511..9132d927171 100644
--- a/arch/ppc64/kernel/Makefile
+++ b/arch/ppc64/kernel/Makefile
@@ -16,7 +16,7 @@ obj-y += vdso32/ vdso64/
 
 obj-$(CONFIG_PPC_OF) +=	of_device.o
 
-pci-obj-$(CONFIG_PPC_ISERIES)	+= iSeries_pci.o iSeries_pci_reset.o
+pci-obj-$(CONFIG_PPC_ISERIES)	+= iSeries_pci.o
 pci-obj-$(CONFIG_PPC_MULTIPLATFORM)	+= pci_dn.o pci_direct_iommu.o
 
 obj-$(CONFIG_PCI)	+= pci.o pci_iommu.o iomap.o $(pci-obj-y)
diff --git a/arch/ppc64/kernel/iSeries_pci_reset.c b/arch/ppc64/kernel/iSeries_pci_reset.c
deleted file mode 100644
index 0f785e4584f..00000000000
--- a/arch/ppc64/kernel/iSeries_pci_reset.c
+++ /dev/null
@@ -1,104 +0,0 @@
-#define PCIFR(...)
-/************************************************************************/
-/* File iSeries_pci_reset.c created by Allan Trautman on Mar 21 2001.   */
-/************************************************************************/
-/* This code supports the pci interface on the IBM iSeries systems.     */
-/* Copyright (C) 20yy  <Allan H Trautman> <IBM Corp>                    */
-/*                                                                      */
-/* This program is free software; you can redistribute it and/or modify */
-/* it under the terms of the GNU General Public License as published by */
-/* the Free Software Foundation; either version 2 of the License, or    */
-/* (at your option) any later version.                                  */
-/*                                                                      */
-/* This program is distributed in the hope that it will be useful,      */ 
-/* but WITHOUT ANY WARRANTY; without even the implied warranty of       */
-/* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the        */
-/* GNU General Public License for more details.                         */
-/*                                                                      */
-/* You should have received a copy of the GNU General Public License    */ 
-/* along with this program; if not, write to the:                       */
-/* Free Software Foundation, Inc.,                                      */ 
-/* 59 Temple Place, Suite 330,                                          */ 
-/* Boston, MA  02111-1307  USA                                          */
-/************************************************************************/
-/* Change Activity:                                                     */
-/*   Created, March 20, 2001                                            */
-/*   April 30, 2001, Added return codes on functions.                   */
-/*   September 10, 2001, Ported to ppc64.                               */
-/* End Change Activity                                                  */
-/************************************************************************/
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/module.h>
-#include <linux/pci.h>
-#include <linux/irq.h>
-#include <linux/delay.h>
-
-#include <asm/io.h>
-#include <asm/iSeries/HvCallPci.h>
-#include <asm/iSeries/HvTypes.h>
-#include <asm/iSeries/mf.h>
-#include <asm/pci.h>
-
-#include <asm/iSeries/iSeries_pci.h>
-#include "pci.h"
-
-/*
- * Interface to toggle the reset line
- * Time is in .1 seconds, need for seconds.
- */
-int iSeries_Device_ToggleReset(struct pci_dev *PciDev, int AssertTime,
-		int DelayTime)
-{
-	unsigned int AssertDelay, WaitDelay;
-	struct iSeries_Device_Node *DeviceNode =
-		(struct iSeries_Device_Node *)PciDev->sysdata;
-
- 	if (DeviceNode == NULL) { 
-		printk("PCI: Pci Reset Failed, Device Node not found for pci_dev %p\n",
-				PciDev);
-		return -1;
-	}
-	/*
-	 * Set defaults, Assert is .5 second, Wait is 3 seconds.
-	 */
-	if (AssertTime == 0)
-		AssertDelay = 500;
-	else
-		AssertDelay = AssertTime * 100;
-
-	if (DelayTime == 0)
-		WaitDelay = 3000;
-	else
-		WaitDelay = DelayTime * 100;
-
-	/*
-	 * Assert reset
-	 */
-	DeviceNode->ReturnCode = HvCallPci_setSlotReset(ISERIES_BUS(DeviceNode),
-			0x00, DeviceNode->AgentId, 1);
-	if (DeviceNode->ReturnCode == 0) {
-		msleep(AssertDelay);			/* Sleep for the time */
-		DeviceNode->ReturnCode =
-			HvCallPci_setSlotReset(ISERIES_BUS(DeviceNode),
-					0x00, DeviceNode->AgentId, 0);
-
-		/*
-   		 * Wait for device to reset
-		 */
-		msleep(WaitDelay);
-	}
-	if (DeviceNode->ReturnCode == 0)
-		PCIFR("Slot 0x%04X.%02 Reset\n", ISERIES_BUS(DeviceNode),
-				DeviceNode->AgentId);
-	else {
-		printk("PCI: Slot 0x%04X.%02X Reset Failed, RCode: %04X\n",
-				ISERIES_BUS(DeviceNode), DeviceNode->AgentId,
-				DeviceNode->ReturnCode);
-		PCIFR("Slot 0x%04X.%02X Reset Failed, RCode: %04X\n",
-				ISERIES_BUS(DeviceNode), DeviceNode->AgentId,
-				DeviceNode->ReturnCode);
-	}
-	return DeviceNode->ReturnCode;
-}
-EXPORT_SYMBOL(iSeries_Device_ToggleReset);
diff --git a/include/asm-ppc64/iSeries/iSeries_pci.h b/include/asm-ppc64/iSeries/iSeries_pci.h
index f45cd8a2e7b..4313a668402 100644
--- a/include/asm-ppc64/iSeries/iSeries_pci.h
+++ b/include/asm-ppc64/iSeries/iSeries_pci.h
@@ -92,7 +92,6 @@ struct iSeries_Device_Node {
 	int		DevFn;		/* Linux devfn */
 	int		BarOffset;
 	int		Irq;		/* Assigned IRQ */
-	int		ReturnCode;	/* Return Code Holder */
 	int		IoRetry;	/* Current Retry Count */
 	int		Flags;		/* Possible flags(disable/bist)*/
 	u16		Vendor;		/* Vendor ID */
@@ -107,7 +106,5 @@ struct iSeries_Device_Node {
 
 extern int	iSeries_Device_Information(struct pci_dev*, char*, int);
 extern void	iSeries_Get_Location_Code(struct iSeries_Device_Node*);
-extern int	iSeries_Device_ToggleReset(struct pci_dev* PciDev,
-		int AssertTime, int DelayTime);
 
 #endif /* _ISERIES_64_PCI_H */
-- 
cgit v1.2.3-70-g09d2


From e7eb22d201936fb3450fa5d1d30cbf2de2fb918b Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:44 -0700
Subject: [PATCH] ppc64 iSeries: iommu.h cleanups

The iommu_table_cb structure is iSeries specific, so move it to the header
file that declares the function we pass it to.  vio_tce_table and
iommu_setup_iSeries no longer exist, so remove their declarations.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-ppc64/iSeries/HvCallXm.h | 17 +++++++++++++++++
 include/asm-ppc64/iommu.h            | 21 ---------------------
 2 files changed, 17 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/asm-ppc64/iSeries/HvCallXm.h b/include/asm-ppc64/iSeries/HvCallXm.h
index 1362a425913..8b9ba608daa 100644
--- a/include/asm-ppc64/iSeries/HvCallXm.h
+++ b/include/asm-ppc64/iSeries/HvCallXm.h
@@ -16,6 +16,23 @@
 #define HvCallXmSetTce			HvCallXm + 11
 #define HvCallXmSetTces			HvCallXm + 13
 
+/*
+ * Structure passed to HvCallXm_getTceTableParms
+ */
+struct iommu_table_cb {
+	unsigned long	itc_busno;	/* Bus number for this tce table */
+	unsigned long	itc_start;	/* Will be NULL for secondary */
+	unsigned long	itc_totalsize;	/* Size (in pages) of whole table */
+	unsigned long	itc_offset;	/* Index into real tce table of the
+					   start of our section */
+	unsigned long	itc_size;	/* Size (in pages) of our section */
+	unsigned long	itc_index;	/* Index of this tce table */
+	unsigned short	itc_maxtables;	/* Max num of tables for partition */
+	unsigned char	itc_virtbus;	/* Flag to indicate virtual bus */
+	unsigned char	itc_slotno;	/* IOA Tce Slot Index */
+	unsigned char	itc_rsvd[4];
+};
+
 static inline void HvCallXm_getTceTableParms(u64 cb)
 {
 	HvCall1(HvCallXmGetTceTableParms, cb);
diff --git a/include/asm-ppc64/iommu.h b/include/asm-ppc64/iommu.h
index bd53ca4dcfa..d9b53374d23 100644
--- a/include/asm-ppc64/iommu.h
+++ b/include/asm-ppc64/iommu.h
@@ -82,24 +82,6 @@ struct iommu_table {
 	unsigned long *it_map;       /* A simple allocation bitmap for now */
 };
 
-#ifdef CONFIG_PPC_ISERIES
-struct iommu_table_cb {
-	unsigned long	itc_busno;	/* Bus number for this tce table */
-	unsigned long	itc_start;	/* Will be NULL for secondary */
-	unsigned long	itc_totalsize;	/* Size (in pages) of whole table */
-	unsigned long	itc_offset;	/* Index into real tce table of the
-					   start of our section */
-	unsigned long	itc_size;	/* Size (in pages) of our section */
-	unsigned long	itc_index;	/* Index of this tce table */
-	unsigned short	itc_maxtables;	/* Max num of tables for partition */
-	unsigned char	itc_virtbus;	/* Flag to indicate virtual bus */
- 	unsigned char	itc_slotno;	/* IOA Tce Slot Index */
- 	unsigned char	itc_rsvd[4];
-};
-
-extern struct iommu_table vio_tce_table;      /* Tce table for virtual bus */
-#endif /* CONFIG_PPC_ISERIES */
-
 struct scatterlist;
 
 #ifdef CONFIG_PPC_MULTIPLATFORM
@@ -122,9 +104,6 @@ extern void iommu_devnode_init_pSeries(struct device_node *dn);
 
 #ifdef CONFIG_PPC_ISERIES
 
-/* Walks all buses and creates iommu tables */
-extern void iommu_setup_iSeries(void);
-
 /* Initializes tables for bio buses */
 extern void __init iommu_vio_init(void);
 
-- 
cgit v1.2.3-70-g09d2


From aab41dea80b1b00e66533af096f797320be05f13 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:46 -0700
Subject: [PATCH] ppc64 iSeries: iSeries_pci.h cleanups

Remove no longer used things from iSeries_pci.h.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/iSeries_pci.c         |  1 -
 include/asm-ppc64/iSeries/iSeries_pci.h | 17 +----------------
 2 files changed, 1 insertion(+), 17 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/iSeries_pci.c b/arch/ppc64/kernel/iSeries_pci.c
index 1bff515bd29..1313a7a2f00 100644
--- a/arch/ppc64/kernel/iSeries_pci.c
+++ b/arch/ppc64/kernel/iSeries_pci.c
@@ -497,7 +497,6 @@ static int scan_bridge_slot(HvBusNumber Bus,
 
 			++DeviceCount;
 			node = build_device_node(Bus, SubBus, EADsIdSel, Function);
-			node->Vendor = VendorId;
 			node->Irq = Irq;
 			node->LogicalSlot = BridgeInfo->logicalSlotNumber;
 
diff --git a/include/asm-ppc64/iSeries/iSeries_pci.h b/include/asm-ppc64/iSeries/iSeries_pci.h
index 4313a668402..5d127cf29bf 100644
--- a/include/asm-ppc64/iSeries/iSeries_pci.h
+++ b/include/asm-ppc64/iSeries/iSeries_pci.h
@@ -44,8 +44,7 @@ struct iSeries_Device_Node;
 #define ISERIES_SUBBUS(DevPtr)	DevPtr->DsaAddr.Dsa.subBusNumber
 #define ISERIES_DEVICE(DevPtr)	DevPtr->DsaAddr.Dsa.deviceId
 #define ISERIES_DSA(DevPtr)	DevPtr->DsaAddr.DsaAddr
-#define ISERIES_DEVFUN(DevPtr)	DevPtr->DevFn
-#define ISERIES_DEVNODE(PciDev)	((struct iSeries_Device_Node*)PciDev->sysdata)
+#define ISERIES_DEVNODE(PciDev)	((struct iSeries_Device_Node *)PciDev->sysdata)
 
 #define EADsMaxAgents 7
 
@@ -62,17 +61,6 @@ struct iSeries_Device_Node;
 #define ISERIES_GET_DEVICE_FROM_SUBBUS(subbus)		((subbus >> 5) & 0x7)
 #define ISERIES_GET_FUNCTION_FROM_SUBBUS(subbus)	((subbus >> 2) & 0x7)
 
-/*
- * N.B. the ISERIES_DECODE_* macros are not used anywhere, and I think
- * the 0x71 (at least) must be wrong - 0x78 maybe?  -- paulus.
- */
-#define ISERIES_DECODE_DEVFN(linuxdevfn)	\
-	(((linuxdevfn & 0x71) << 1) | (linuxdevfn & 0x07))
-#define ISERIES_DECODE_DEVICE(linuxdevfn)	\
-	(((linuxdevfn & 0x38) >> 3) | (((linuxdevfn & 0x40) >> 2) + 0x10))
-#define ISERIES_DECODE_FUNCTION(linuxdevfn)	\
-	(linuxdevfn & 0x07)
-
 /*
  * Converts Virtual Address to Real Address for Hypervisor calls
  */
@@ -90,15 +78,12 @@ struct iSeries_Device_Node {
 					/* deviceId, barNumber */
 	HvAgentId	AgentId;	/* Hypervisor DevFn */
 	int		DevFn;		/* Linux devfn */
-	int		BarOffset;
 	int		Irq;		/* Assigned IRQ */
 	int		IoRetry;	/* Current Retry Count */
 	int		Flags;		/* Possible flags(disable/bist)*/
-	u16		Vendor;		/* Vendor ID */
 	u8		LogicalSlot;	/* Hv Slot Index for Tces */
 	struct iommu_table *iommu_table;/* Device TCE Table */
 	u8		PhbId;		/* Phb Card is on. */
-	u16		Board;		/* Board Number */
 	u8		FrameId;	/* iSeries spcn Frame Id */
 	char		CardLocation[4];/* Char format of planar vpd */
 	char		Location[20];	/* Frame	1, Card C10 */
-- 
cgit v1.2.3-70-g09d2


From a2ebaf250fabc5c5644b707dbee44c9e0ec442e9 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:47 -0700
Subject: [PATCH] ppc64 iSeries: remove IoRetry from iSeries_Device_Node

The IoRetry member of iSeries_Devide_Node is really only used locally, so
remove it and replace it with a local variable.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/iSeries_pci.c         | 35 +++++++++++++++++----------------
 include/asm-ppc64/iSeries/iSeries_pci.h |  1 -
 2 files changed, 18 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/iSeries_pci.c b/arch/ppc64/kernel/iSeries_pci.c
index 1313a7a2f00..0a28b6ca65d 100644
--- a/arch/ppc64/kernel/iSeries_pci.c
+++ b/arch/ppc64/kernel/iSeries_pci.c
@@ -225,7 +225,6 @@ static struct iSeries_Device_Node *build_device_node(HvBusNumber Bus,
 	node->DsaAddr.Dsa.deviceId = 0x10;
 	node->AgentId = AgentId;
 	node->DevFn = PCI_DEVFN(ISERIES_ENCODE_DEVICE(AgentId), Function);
-	node->IoRetry = 0;
 	iSeries_Get_Location_Code(node);
 	return node;
 }
@@ -658,38 +657,34 @@ static struct pci_ops iSeries_pci_ops = {
  * Check Return Code
  * -> On Failure, print and log information.
  *    Increment Retry Count, if exceeds max, panic partition.
- * -> If in retry, print and log success 
  *
  * PCI: Device 23.90 ReadL I/O Error( 0): 0x1234
  * PCI: Device 23.90 ReadL Retry( 1)
  * PCI: Device 23.90 ReadL Retry Successful(1)
  */
 static int CheckReturnCode(char *TextHdr, struct iSeries_Device_Node *DevNode,
-		u64 ret)
+		int *retry, u64 ret)
 {
 	if (ret != 0)  {
 		++Pci_Error_Count;
-		++DevNode->IoRetry;
+		(*retry)++;
 		printk("PCI: %s: Device 0x%04X:%02X  I/O Error(%2d): 0x%04X\n",
 				TextHdr, DevNode->DsaAddr.Dsa.busNumber, DevNode->DevFn,
-				DevNode->IoRetry, (int)ret);
+				*retry, (int)ret);
 		/*
 		 * Bump the retry and check for retry count exceeded.
 		 * If, Exceeded, panic the system.
 		 */
-		if ((DevNode->IoRetry > Pci_Retry_Max) &&
+		if (((*retry) > Pci_Retry_Max) &&
 				(Pci_Error_Flag > 0)) {
 			mf_display_src(0xB6000103);
-			panic_timeout = 0; 
+			panic_timeout = 0;
 			panic("PCI: Hardware I/O Error, SRC B6000103, "
 					"Automatic Reboot Disabled.\n");
 		}
 		return -1;	/* Retry Try */
 	}
-	/* If retry was in progress, log success and rest retry count */
-	if (DevNode->IoRetry > 0)
-		DevNode->IoRetry = 0;
-	return 0; 
+	return 0;
 }
 
 /*
@@ -735,6 +730,7 @@ u8 iSeries_Read_Byte(const volatile void __iomem *IoAddress)
 {
 	u64 BarOffset;
 	u64 dsa;
+	int retry = 0;
 	struct HvCallPci_LoadReturn ret;
 	struct iSeries_Device_Node *DevNode =
 		xlate_iomm_address(IoAddress, &dsa, &BarOffset);
@@ -754,7 +750,7 @@ u8 iSeries_Read_Byte(const volatile void __iomem *IoAddress)
 	do {
 		++Pci_Io_Read_Count;
 		HvCall3Ret16(HvCallPciBarLoad8, &ret, dsa, BarOffset, 0);
-	} while (CheckReturnCode("RDB", DevNode, ret.rc) != 0);
+	} while (CheckReturnCode("RDB", DevNode, &retry, ret.rc) != 0);
 
 	return (u8)ret.value;
 }
@@ -764,6 +760,7 @@ u16 iSeries_Read_Word(const volatile void __iomem *IoAddress)
 {
 	u64 BarOffset;
 	u64 dsa;
+	int retry = 0;
 	struct HvCallPci_LoadReturn ret;
 	struct iSeries_Device_Node *DevNode =
 		xlate_iomm_address(IoAddress, &dsa, &BarOffset);
@@ -784,7 +781,7 @@ u16 iSeries_Read_Word(const volatile void __iomem *IoAddress)
 		++Pci_Io_Read_Count;
 		HvCall3Ret16(HvCallPciBarLoad16, &ret, dsa,
 				BarOffset, 0);
-	} while (CheckReturnCode("RDW", DevNode, ret.rc) != 0);
+	} while (CheckReturnCode("RDW", DevNode, &retry, ret.rc) != 0);
 
 	return swab16((u16)ret.value);
 }
@@ -794,6 +791,7 @@ u32 iSeries_Read_Long(const volatile void __iomem *IoAddress)
 {
 	u64 BarOffset;
 	u64 dsa;
+	int retry = 0;
 	struct HvCallPci_LoadReturn ret;
 	struct iSeries_Device_Node *DevNode =
 		xlate_iomm_address(IoAddress, &dsa, &BarOffset);
@@ -814,7 +812,7 @@ u32 iSeries_Read_Long(const volatile void __iomem *IoAddress)
 		++Pci_Io_Read_Count;
 		HvCall3Ret16(HvCallPciBarLoad32, &ret, dsa,
 				BarOffset, 0);
-	} while (CheckReturnCode("RDL", DevNode, ret.rc) != 0);
+	} while (CheckReturnCode("RDL", DevNode, &retry, ret.rc) != 0);
 
 	return swab32((u32)ret.value);
 }
@@ -831,6 +829,7 @@ void iSeries_Write_Byte(u8 data, volatile void __iomem *IoAddress)
 {
 	u64 BarOffset;
 	u64 dsa;
+	int retry = 0;
 	u64 rc;
 	struct iSeries_Device_Node *DevNode =
 		xlate_iomm_address(IoAddress, &dsa, &BarOffset);
@@ -850,7 +849,7 @@ void iSeries_Write_Byte(u8 data, volatile void __iomem *IoAddress)
 	do {
 		++Pci_Io_Write_Count;
 		rc = HvCall4(HvCallPciBarStore8, dsa, BarOffset, data, 0);
-	} while (CheckReturnCode("WWB", DevNode, rc) != 0);
+	} while (CheckReturnCode("WWB", DevNode, &retry, rc) != 0);
 }
 EXPORT_SYMBOL(iSeries_Write_Byte);
 
@@ -858,6 +857,7 @@ void iSeries_Write_Word(u16 data, volatile void __iomem *IoAddress)
 {
 	u64 BarOffset;
 	u64 dsa;
+	int retry = 0;
 	u64 rc;
 	struct iSeries_Device_Node *DevNode =
 		xlate_iomm_address(IoAddress, &dsa, &BarOffset);
@@ -877,7 +877,7 @@ void iSeries_Write_Word(u16 data, volatile void __iomem *IoAddress)
 	do {
 		++Pci_Io_Write_Count;
 		rc = HvCall4(HvCallPciBarStore16, dsa, BarOffset, swab16(data), 0);
-	} while (CheckReturnCode("WWW", DevNode, rc) != 0);
+	} while (CheckReturnCode("WWW", DevNode, &retry, rc) != 0);
 }
 EXPORT_SYMBOL(iSeries_Write_Word);
 
@@ -885,6 +885,7 @@ void iSeries_Write_Long(u32 data, volatile void __iomem *IoAddress)
 {
 	u64 BarOffset;
 	u64 dsa;
+	int retry = 0;
 	u64 rc;
 	struct iSeries_Device_Node *DevNode =
 		xlate_iomm_address(IoAddress, &dsa, &BarOffset);
@@ -904,6 +905,6 @@ void iSeries_Write_Long(u32 data, volatile void __iomem *IoAddress)
 	do {
 		++Pci_Io_Write_Count;
 		rc = HvCall4(HvCallPciBarStore32, dsa, BarOffset, swab32(data), 0);
-	} while (CheckReturnCode("WWL", DevNode, rc) != 0);
+	} while (CheckReturnCode("WWL", DevNode, &retry, rc) != 0);
 }
 EXPORT_SYMBOL(iSeries_Write_Long);
diff --git a/include/asm-ppc64/iSeries/iSeries_pci.h b/include/asm-ppc64/iSeries/iSeries_pci.h
index 5d127cf29bf..ee0212f18b9 100644
--- a/include/asm-ppc64/iSeries/iSeries_pci.h
+++ b/include/asm-ppc64/iSeries/iSeries_pci.h
@@ -79,7 +79,6 @@ struct iSeries_Device_Node {
 	HvAgentId	AgentId;	/* Hypervisor DevFn */
 	int		DevFn;		/* Linux devfn */
 	int		Irq;		/* Assigned IRQ */
-	int		IoRetry;	/* Current Retry Count */
 	int		Flags;		/* Possible flags(disable/bist)*/
 	u8		LogicalSlot;	/* Hv Slot Index for Tces */
 	struct iommu_table *iommu_table;/* Device TCE Table */
-- 
cgit v1.2.3-70-g09d2


From 061c063efce96b33f9e26d5f83a8eb0643fa493c Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:48 -0700
Subject: [PATCH] ppc64 iSeries: remove some more members of
 iSeries_Device_Node

The AgentId, PhbId, FrameId, CardLocation and Location members of
iSeries_Device_Node are stored early in the boot process just so that a
message about the device can be printed later in the boot process.  Remove
them and construct the message by doing the VPD parsing at the time the
message is printed.

Also remove a few unused defines in iSeries_VpdInfo.c.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/iSeries_VpdInfo.c     | 137 ++++++++++++++++----------------
 arch/ppc64/kernel/iSeries_pci.c         |   7 +-
 include/asm-ppc64/iSeries/iSeries_pci.h |   8 +-
 3 files changed, 69 insertions(+), 83 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/iSeries_VpdInfo.c b/arch/ppc64/kernel/iSeries_VpdInfo.c
index 9c89ce9a3d0..d11c732daf8 100644
--- a/arch/ppc64/kernel/iSeries_VpdInfo.c
+++ b/arch/ppc64/kernel/iSeries_VpdInfo.c
@@ -44,7 +44,6 @@
 /*
  * Bus Vpd Tags
  */
-#define  VpdEndOfDataTag   0x78
 #define  VpdEndOfAreaTag   0x79
 #define  VpdIdStringTag    0x82
 #define  VpdVendorAreaTag  0x84
@@ -52,11 +51,8 @@
 /*
  * Mfg Area Tags
  */
-#define  VpdFruFlag       0x4647     // "FG"
 #define  VpdFruFrameId    0x4649     // "FI"
 #define  VpdSlotMapFormat 0x4D46     // "MF"
-#define  VpdAsmPartNumber 0x504E     // "PN"
-#define  VpdFruSerial     0x534E     // "SN"
 #define  VpdSlotMap       0x534D     // "SM"
 
 /*
@@ -82,49 +78,11 @@ struct SlotMapStruct {
 typedef struct SlotMapStruct SlotMap;
 #define SLOT_ENTRY_SIZE   16
 
-/*
- * Formats the device information.
- * - Pass in pci_dev* pointer to the device.
- * - Pass in buffer to place the data.  Danger here is the buffer must
- *   be as big as the client says it is.   Should be at least 128 bytes.
- * Return will the length of the string data put in the buffer.
- * Format:
- * PCI: Bus  0, Device 26, Vendor 0x12AE  Frame  1, Card  C10  Ethernet
- * controller
- */
-int iSeries_Device_Information(struct pci_dev *PciDev, char *buffer,
-		int BufferSize)
-{
-	struct iSeries_Device_Node *DevNode =
-		(struct iSeries_Device_Node *)PciDev->sysdata;
-	int len;
-
-	if (DevNode == NULL)
-		return sprintf(buffer,
-				"PCI: iSeries_Device_Information DevNode is NULL");
-
-	if (BufferSize < 128)
-		return 0;
-
-	len = sprintf(buffer, "PCI: Bus%3d, Device%3d, Vendor %04X ",
-			ISERIES_BUS(DevNode), PCI_SLOT(PciDev->devfn),
-			PciDev->vendor);
-	len += sprintf(buffer + len, "Frame%3d, Card %4s  ",
-			DevNode->FrameId, DevNode->CardLocation);
-	if (pci_class_name(PciDev->class >> 8) == 0)
-		len += sprintf(buffer + len, "0x%04X  ",
-				(int)(PciDev->class >> 8));
-	else
-		len += sprintf(buffer + len, "%s",
-				pci_class_name(PciDev->class >> 8));
-	return len;
-}
-
 /*
  * Parse the Slot Area
  */
-static void iSeries_Parse_SlotArea(SlotMap *MapPtr, int MapLen,
-		struct iSeries_Device_Node *DevNode)
+static void __init iSeries_Parse_SlotArea(SlotMap *MapPtr, int MapLen,
+		HvAgentId agent, u8 *PhbId, char card[4])
 {
 	int SlotMapLen = MapLen;
 	SlotMap *SlotMapPtr = MapPtr;
@@ -133,17 +91,16 @@ static void iSeries_Parse_SlotArea(SlotMap *MapPtr, int MapLen,
 	 * Parse Slot label until we find the one requested
 	 */
 	while (SlotMapLen > 0) {
-		if (SlotMapPtr->AgentId == DevNode->AgentId ) {
+		if (SlotMapPtr->AgentId == agent) {
 			/*
 			 * If Phb wasn't found, grab the entry first one found.
 			 */
-			if (DevNode->PhbId == 0xff)
-				DevNode->PhbId = SlotMapPtr->PhbId; 
+			if (*PhbId == 0xff)
+				*PhbId = SlotMapPtr->PhbId;
 			/* Found it, extract the data. */
-			if (SlotMapPtr->PhbId == DevNode->PhbId ) {
-	        		memcpy(&DevNode->CardLocation,
-						&SlotMapPtr->CardLocation, 3);
-				DevNode->CardLocation[3]  = 0;
+			if (SlotMapPtr->PhbId == *PhbId) {
+				memcpy(card, &SlotMapPtr->CardLocation, 3);
+				card[3]  = 0;
 				break;
 			}
 		}
@@ -156,8 +113,9 @@ static void iSeries_Parse_SlotArea(SlotMap *MapPtr, int MapLen,
 /*
  * Parse the Mfg Area
  */
-static void iSeries_Parse_MfgArea(u8 *AreaData, int AreaLen,
-		struct iSeries_Device_Node *DevNode)
+static void __init iSeries_Parse_MfgArea(u8 *AreaData, int AreaLen,
+		HvAgentId agent, u8 *PhbId,
+		u8 *frame, char card[4])
 {
 	MfgArea *MfgAreaPtr = (MfgArea *)AreaData;
 	int MfgAreaLen = AreaLen;
@@ -168,7 +126,7 @@ static void iSeries_Parse_MfgArea(u8 *AreaData, int AreaLen,
 		int MfgTagLen = MfgAreaPtr->TagLength;
 		/* Frame ID         (FI 4649020310 ) */
 		if (MfgAreaPtr->Tag == VpdFruFrameId)		/* FI  */
-			DevNode->FrameId = MfgAreaPtr->AreaData1;
+			*frame = MfgAreaPtr->AreaData1;
 		/* Slot Map Format  (MF 4D46020004 ) */
 		else if (MfgAreaPtr->Tag == VpdSlotMapFormat)	/* MF  */
 			SlotMapFmt = (MfgAreaPtr->AreaData1 * 256)
@@ -183,7 +141,8 @@ static void iSeries_Parse_MfgArea(u8 *AreaData, int AreaLen,
 			else
 				SlotMapPtr = (SlotMap *)((char *)MfgAreaPtr
 						+ MFG_ENTRY_SIZE);
-	    		iSeries_Parse_SlotArea(SlotMapPtr, MfgTagLen, DevNode);
+			iSeries_Parse_SlotArea(SlotMapPtr, MfgTagLen,
+					agent, PhbId, card);
 		}
 		/*
 		 * Point to the next Mfg Area
@@ -199,7 +158,7 @@ static void iSeries_Parse_MfgArea(u8 *AreaData, int AreaLen,
  * Look for "BUS".. Data is not Null terminated.
  * PHBID of 0xFF indicates PHB was not found in VPD Data.
  */
-static int iSeries_Parse_PhbId(u8 *AreaPtr, int AreaLength)
+static int __init iSeries_Parse_PhbId(u8 *AreaPtr, int AreaLength)
 {
 	u8 *PhbPtr = AreaPtr;
 	int DataLen = AreaLength;
@@ -223,27 +182,30 @@ static int iSeries_Parse_PhbId(u8 *AreaPtr, int AreaLength)
 /*
  * Parse out the VPD Areas
  */
-static void iSeries_Parse_Vpd(u8 *VpdData, int VpdDataLen,
-		struct iSeries_Device_Node *DevNode)
+static void __init iSeries_Parse_Vpd(u8 *VpdData, int VpdDataLen,
+		HvAgentId agent, u8 *frame, char card[4])
 {
 	u8 *TagPtr = VpdData;
 	int DataLen = VpdDataLen - 3;
+	u8 PhbId;
 
 	while ((*TagPtr != VpdEndOfAreaTag) && (DataLen > 0)) {
 		int AreaLen = *(TagPtr + 1) + (*(TagPtr + 2) * 256);
 		u8 *AreaData  = TagPtr + 3;
 
 		if (*TagPtr == VpdIdStringTag)
-			DevNode->PhbId = iSeries_Parse_PhbId(AreaData, AreaLen);
+			PhbId = iSeries_Parse_PhbId(AreaData, AreaLen);
 		else if (*TagPtr == VpdVendorAreaTag)
-	    		iSeries_Parse_MfgArea(AreaData, AreaLen, DevNode);
+			iSeries_Parse_MfgArea(AreaData, AreaLen,
+					agent, &PhbId, frame, card);
 		/* Point to next Area. */
 		TagPtr  = AreaData + AreaLen;
 		DataLen -= AreaLen;
 	}
 }
 
-void iSeries_Get_Location_Code(struct iSeries_Device_Node *DevNode)
+static void __init iSeries_Get_Location_Code(u16 bus, HvAgentId agent,
+		u8 *frame, char card[4])
 {
 	int BusVpdLen = 0;
 	u8 *BusVpdPtr = kmalloc(BUS_VPDSIZE, GFP_KERNEL);
@@ -252,23 +214,58 @@ void iSeries_Get_Location_Code(struct iSeries_Device_Node *DevNode)
 		printk("PCI: Bus VPD Buffer allocation failure.\n");
 		return;
 	}
-	BusVpdLen = HvCallPci_getBusVpd(ISERIES_BUS(DevNode),
-					ISERIES_HV_ADDR(BusVpdPtr),
+	BusVpdLen = HvCallPci_getBusVpd(bus, ISERIES_HV_ADDR(BusVpdPtr),
 					BUS_VPDSIZE);
 	if (BusVpdLen == 0) {
-		kfree(BusVpdPtr);
 		printk("PCI: Bus VPD Buffer zero length.\n");
-		return;
+		goto out_free;
 	}
 	/* printk("PCI: BusVpdPtr: %p, %d\n",BusVpdPtr, BusVpdLen); */
 	/* Make sure this is what I think it is */
 	if (*BusVpdPtr != VpdIdStringTag) {	/* 0x82 */
 		printk("PCI: Bus VPD Buffer missing starting tag.\n");
-		kfree(BusVpdPtr);
-		return;
+		goto out_free;
 	}
-	iSeries_Parse_Vpd(BusVpdPtr,BusVpdLen, DevNode);
-	sprintf(DevNode->Location, "Frame%3d, Card %-4s", DevNode->FrameId,
-			DevNode->CardLocation);
+	iSeries_Parse_Vpd(BusVpdPtr, BusVpdLen, agent, frame, card);
+out_free:
 	kfree(BusVpdPtr);
 }
+
+/*
+ * Prints the device information.
+ * - Pass in pci_dev* pointer to the device.
+ * - Pass in the device count
+ *
+ * Format:
+ * PCI: Bus  0, Device 26, Vendor 0x12AE  Frame  1, Card  C10  Ethernet
+ * controller
+ */
+void __init iSeries_Device_Information(struct pci_dev *PciDev, int count)
+{
+	struct iSeries_Device_Node *DevNode = PciDev->sysdata;
+	u16 bus;
+	u8 frame;
+	char card[4];
+	HvSubBusNumber subbus;
+	HvAgentId agent;
+
+	if (DevNode == NULL) {
+		printk("%d. PCI: iSeries_Device_Information DevNode is NULL\n",
+				count);
+		return;
+	}
+
+	bus = ISERIES_BUS(DevNode);
+	subbus = ISERIES_SUBBUS(DevNode);
+	agent = ISERIES_PCI_AGENTID(ISERIES_GET_DEVICE_FROM_SUBBUS(subbus),
+			ISERIES_GET_FUNCTION_FROM_SUBBUS(subbus));
+	iSeries_Get_Location_Code(bus, agent, &frame, card);
+
+	printk("%d. PCI: Bus%3d, Device%3d, Vendor %04X Frame%3d, Card %4s  ",
+			count, bus, PCI_SLOT(PciDev->devfn), PciDev->vendor,
+			frame, card);
+	if (pci_class_name(PciDev->class >> 8) == 0)
+		printk("0x%04X\n", (int)(PciDev->class >> 8));
+	else
+		printk("%s\n", pci_class_name(PciDev->class >> 8));
+}
diff --git a/arch/ppc64/kernel/iSeries_pci.c b/arch/ppc64/kernel/iSeries_pci.c
index 0a28b6ca65d..356e4fd9a94 100644
--- a/arch/ppc64/kernel/iSeries_pci.c
+++ b/arch/ppc64/kernel/iSeries_pci.c
@@ -223,9 +223,7 @@ static struct iSeries_Device_Node *build_device_node(HvBusNumber Bus,
 	node->DsaAddr.Dsa.busNumber = Bus;
 	node->DsaAddr.Dsa.subBusNumber = SubBus;
 	node->DsaAddr.Dsa.deviceId = 0x10;
-	node->AgentId = AgentId;
 	node->DevFn = PCI_DEVFN(ISERIES_ENCODE_DEVICE(AgentId), Function);
-	iSeries_Get_Location_Code(node);
 	return node;
 }
 
@@ -299,7 +297,6 @@ void __init iSeries_pci_final_fixup(void)
 {
 	struct pci_dev *pdev = NULL;
 	struct iSeries_Device_Node *node;
-	char Buffer[256];
     	int DeviceCount = 0;
 
 	PPCDBG(PPCDBG_BUSWALK, "iSeries_pcibios_fixup Entry.\n"); 
@@ -321,9 +318,7 @@ void __init iSeries_pci_final_fixup(void)
 					"pdev 0x%p <==> DevNode 0x%p\n",
 					pdev, node);
 			allocate_device_bars(pdev);
-			iSeries_Device_Information(pdev, Buffer,
-					sizeof(Buffer));
-			printk("%d. %s\n", DeviceCount, Buffer);
+			iSeries_Device_Information(pdev, DeviceCount);
 			iommu_devnode_init_iSeries(node);
 		} else
 			printk("PCI: Device Tree not found for 0x%016lX\n",
diff --git a/include/asm-ppc64/iSeries/iSeries_pci.h b/include/asm-ppc64/iSeries/iSeries_pci.h
index ee0212f18b9..575f611f8b3 100644
--- a/include/asm-ppc64/iSeries/iSeries_pci.h
+++ b/include/asm-ppc64/iSeries/iSeries_pci.h
@@ -76,19 +76,13 @@ struct iSeries_Device_Node {
 	union HvDsaMap	DsaAddr;	/* Direct Select Address */
 					/* busNumber, subBusNumber, */
 					/* deviceId, barNumber */
-	HvAgentId	AgentId;	/* Hypervisor DevFn */
 	int		DevFn;		/* Linux devfn */
 	int		Irq;		/* Assigned IRQ */
 	int		Flags;		/* Possible flags(disable/bist)*/
 	u8		LogicalSlot;	/* Hv Slot Index for Tces */
 	struct iommu_table *iommu_table;/* Device TCE Table */
-	u8		PhbId;		/* Phb Card is on. */
-	u8		FrameId;	/* iSeries spcn Frame Id */
-	char		CardLocation[4];/* Char format of planar vpd */
-	char		Location[20];	/* Frame	1, Card C10 */
 };
 
-extern int	iSeries_Device_Information(struct pci_dev*, char*, int);
-extern void	iSeries_Get_Location_Code(struct iSeries_Device_Node*);
+extern void	iSeries_Device_Information(struct pci_dev*, int);
 
 #endif /* _ISERIES_64_PCI_H */
-- 
cgit v1.2.3-70-g09d2


From 0c3b4f1a8e06584d8a1051a74ed79cf8b41f703e Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:49 -0700
Subject: [PATCH] ppc64 iSeries: irq simple cleanups

This patch is just simple cleanups to the iSeries irq code.
	- whitespace and comments
	- rearrange some functions to avoid forward declarations
	- remove XmPciLpEvent.h as its functions were declared elsewhere
	- remove decaration of function that no longer exists
No semantic changes.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/XmPciLpEvent.c         | 105 ++++++++++-----------
 arch/ppc64/kernel/iSeries_irq.c          | 155 +++++++++++++++----------------
 include/asm-ppc64/iSeries/XmPciLpEvent.h |   7 --
 include/asm-ppc64/iSeries/iSeries_irq.h  |   1 -
 4 files changed, 123 insertions(+), 145 deletions(-)
 delete mode 100644 include/asm-ppc64/iSeries/XmPciLpEvent.h

(limited to 'include')

diff --git a/arch/ppc64/kernel/XmPciLpEvent.c b/arch/ppc64/kernel/XmPciLpEvent.c
index 809c9bc6678..3db53074483 100644
--- a/arch/ppc64/kernel/XmPciLpEvent.c
+++ b/arch/ppc64/kernel/XmPciLpEvent.c
@@ -1,9 +1,8 @@
 /*
- * File XmPciLpEvent.h created by Wayne Holm on Mon Jan 15 2001.
+ * File XmPciLpEvent.c created by Wayne Holm on Mon Jan 15 2001.
  *
  * This module handles PCI interrupt events sent by the iSeries Hypervisor.
-*/
-
+ */
 #include <linux/config.h>
 #include <linux/pci.h>
 #include <linux/init.h>
@@ -17,22 +16,22 @@
 #include <asm/iSeries/HvTypes.h>
 #include <asm/iSeries/HvLpEvent.h>
 #include <asm/iSeries/HvCallPci.h>
-#include <asm/iSeries/XmPciLpEvent.h>
+#include <asm/iSeries/iSeries_irq.h>
 #include <asm/ppcdebug.h>
 
 static long Pci_Interrupt_Count;
 static long Pci_Event_Count;
 
 enum XmPciLpEvent_Subtype {
-	XmPciLpEvent_BusCreated	   = 0,		// PHB has been created
-	XmPciLpEvent_BusError	   = 1,		// PHB has failed
-	XmPciLpEvent_BusFailed	   = 2,		// Msg to Secondary, Primary failed bus
-	XmPciLpEvent_NodeFailed	   = 4,		// Multi-adapter bridge has failed
-	XmPciLpEvent_NodeRecovered = 5,		// Multi-adapter bridge has recovered
-	XmPciLpEvent_BusRecovered  = 12,	// PHB has been recovered
-	XmPciLpEvent_UnQuiesceBus  = 18,	// Secondary bus unqiescing
-	XmPciLpEvent_BridgeError   = 21,	// Bridge Error
-	XmPciLpEvent_SlotInterrupt = 22		// Slot interrupt
+	XmPciLpEvent_BusCreated		= 0,	// PHB has been created
+	XmPciLpEvent_BusError		= 1,	// PHB has failed
+	XmPciLpEvent_BusFailed		= 2,	// Msg to Secondary, Primary failed bus
+	XmPciLpEvent_NodeFailed		= 4,	// Multi-adapter bridge has failed
+	XmPciLpEvent_NodeRecovered	= 5,	// Multi-adapter bridge has recovered
+	XmPciLpEvent_BusRecovered	= 12,	// PHB has been recovered
+	XmPciLpEvent_UnQuiesceBus	= 18,	// Secondary bus unqiescing
+	XmPciLpEvent_BridgeError	= 21,	// Bridge Error
+	XmPciLpEvent_SlotInterrupt	= 22	// Slot interrupt
 };
 
 struct XmPciLpEvent_BusInterrupt {
@@ -70,43 +69,6 @@ struct XmPciLpEvent {
 
 };
 
-static void intReceived(struct XmPciLpEvent *eventParm,
-		struct pt_regs *regsParm);
-
-static void XmPciLpEvent_handler(struct HvLpEvent *eventParm,
-		struct pt_regs *regsParm)
-{
-#ifdef CONFIG_PCI
-#if 0
-	PPCDBG(PPCDBG_BUSWALK, "XmPciLpEvent_handler, type 0x%x\n",
-			eventParm->xType);
-#endif
-	++Pci_Event_Count;
-
-	if (eventParm && (eventParm->xType == HvLpEvent_Type_PciIo)) {
-		switch (eventParm->xFlags.xFunction) {
-		case HvLpEvent_Function_Int:
-			intReceived((struct XmPciLpEvent *)eventParm, regsParm);
-			break;
-		case HvLpEvent_Function_Ack:
-			printk(KERN_ERR
-				"XmPciLpEvent.c: unexpected ack received\n");
-			break;
-		default:
-			printk(KERN_ERR
-				"XmPciLpEvent.c: unexpected event function %d\n",
-				(int)eventParm->xFlags.xFunction);
-			break;
-		}
-	} else if (eventParm)
-		printk(KERN_ERR
-			"XmPciLpEvent.c: Unrecognized PCI event type 0x%x\n",
-			(int)eventParm->xType);
-	else
-		printk(KERN_ERR "XmPciLpEvent.c: NULL event received\n");
-#endif
-}
-
 static void intReceived(struct XmPciLpEvent *eventParm,
 		struct pt_regs *regsParm)
 {
@@ -164,6 +126,39 @@ static void intReceived(struct XmPciLpEvent *eventParm,
 	}
 }
 
+static void XmPciLpEvent_handler(struct HvLpEvent *eventParm,
+		struct pt_regs *regsParm)
+{
+#ifdef CONFIG_PCI
+#if 0
+	PPCDBG(PPCDBG_BUSWALK, "XmPciLpEvent_handler, type 0x%x\n",
+			eventParm->xType);
+#endif
+	++Pci_Event_Count;
+
+	if (eventParm && (eventParm->xType == HvLpEvent_Type_PciIo)) {
+		switch (eventParm->xFlags.xFunction) {
+		case HvLpEvent_Function_Int:
+			intReceived((struct XmPciLpEvent *)eventParm, regsParm);
+			break;
+		case HvLpEvent_Function_Ack:
+			printk(KERN_ERR
+				"XmPciLpEvent.c: unexpected ack received\n");
+			break;
+		default:
+			printk(KERN_ERR
+				"XmPciLpEvent.c: unexpected event function %d\n",
+				(int)eventParm->xFlags.xFunction);
+			break;
+		}
+	} else if (eventParm)
+		printk(KERN_ERR
+			"XmPciLpEvent.c: Unrecognized PCI event type 0x%x\n",
+			(int)eventParm->xType);
+	else
+		printk(KERN_ERR "XmPciLpEvent.c: NULL event received\n");
+#endif
+}
 
 /* This should be called sometime prior to buswalk (init_IRQ would be good) */
 int XmPciLpEvent_init()
@@ -179,12 +174,10 @@ int XmPciLpEvent_init()
 	if (xRc == 0) {
 		xRc = HvLpEvent_openPath(HvLpEvent_Type_PciIo, 0);
 		if (xRc != 0)
-			printk(KERN_ERR
-				"XmPciLpEvent.c: open event path failed with rc 0x%x\n",
-				xRc);
+			printk(KERN_ERR "XmPciLpEvent.c: open event path "
+					"failed with rc 0x%x\n", xRc);
 	} else
-		printk(KERN_ERR
-			"XmPciLpEvent.c: register handler failed with rc 0x%x\n",
-			xRc);
+		printk(KERN_ERR "XmPciLpEvent.c: register handler "
+				"failed with rc 0x%x\n", xRc);
 	return xRc;
 }
diff --git a/arch/ppc64/kernel/iSeries_irq.c b/arch/ppc64/kernel/iSeries_irq.c
index f831d259dbb..41902e358e7 100644
--- a/arch/ppc64/kernel/iSeries_irq.c
+++ b/arch/ppc64/kernel/iSeries_irq.c
@@ -1,27 +1,27 @@
-/************************************************************************/
-/* This module supports the iSeries PCI bus interrupt handling          */
-/* Copyright (C) 20yy  <Robert L Holtorf> <IBM Corp>                    */
-/*                                                                      */
-/* This program is free software; you can redistribute it and/or modify */
-/* it under the terms of the GNU General Public License as published by */
-/* the Free Software Foundation; either version 2 of the License, or    */
-/* (at your option) any later version.                                  */
-/*                                                                      */
-/* This program is distributed in the hope that it will be useful,      */ 
-/* but WITHOUT ANY WARRANTY; without even the implied warranty of       */
-/* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the        */
-/* GNU General Public License for more details.                         */
-/*                                                                      */
-/* You should have received a copy of the GNU General Public License    */ 
-/* along with this program; if not, write to the:                       */
-/* Free Software Foundation, Inc.,                                      */ 
-/* 59 Temple Place, Suite 330,                                          */ 
-/* Boston, MA  02111-1307  USA                                          */
-/************************************************************************/
-/* Change Activity:                                                     */
-/*   Created, December 13, 2000 by Wayne Holm                           */ 
-/* End Change Activity                                                  */
-/************************************************************************/
+/*
+ * This module supports the iSeries PCI bus interrupt handling
+ * Copyright (C) 20yy  <Robert L Holtorf> <IBM Corp>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the:
+ * Free Software Foundation, Inc.,
+ * 59 Temple Place, Suite 330,
+ * Boston, MA  02111-1307  USA
+ *
+ * Change Activity:
+ *   Created, December 13, 2000 by Wayne Holm
+ * End Change Activity
+ */
 #include <linux/pci.h>
 #include <linux/init.h>
 #include <linux/threads.h>
@@ -38,22 +38,6 @@
 #include <asm/iSeries/HvCallPci.h>
 #include <asm/iSeries/HvCallXm.h>
 #include <asm/iSeries/iSeries_irq.h>
-#include <asm/iSeries/XmPciLpEvent.h>
-
-static unsigned int iSeries_startup_IRQ(unsigned int irq);
-static void iSeries_shutdown_IRQ(unsigned int irq);
-static void iSeries_enable_IRQ(unsigned int irq);
-static void iSeries_disable_IRQ(unsigned int irq);
-static void iSeries_end_IRQ(unsigned int irq);
-
-static hw_irq_controller iSeries_IRQ_handler = {
-	.typename = "iSeries irq controller",
-	.startup = iSeries_startup_IRQ,
-	.shutdown = iSeries_shutdown_IRQ,
-	.enable = iSeries_enable_IRQ,
-	.disable = iSeries_disable_IRQ,
-	.end = iSeries_end_IRQ
-};
 
 /* This maps virtual irq numbers to real irqs */
 unsigned int virt_irq_to_real_map[NR_IRQS];
@@ -69,30 +53,32 @@ void __init iSeries_init_IRQ(void)
 	XmPciLpEvent_init();
 }
 
+#define REAL_IRQ_TO_BUS(irq)	((((irq) >> 6) & 0xff) + 1)
+#define REAL_IRQ_TO_IDSEL(irq)	((((irq) >> 3) & 7) + 1)
+#define REAL_IRQ_TO_FUNC(irq)	((irq) & 7)
+
 /*
- * This is called out of iSeries_scan_slot to allocate an IRQ for an EADS slot
- * It calculates the irq value for the slot.
- * Note that subBusNumber is always 0 (at the moment at least).
+ * This will be called by device drivers (via enable_IRQ)
+ * to enable INTA in the bridge interrupt status register.
  */
-int __init iSeries_allocate_IRQ(HvBusNumber busNumber,
-		HvSubBusNumber subBusNumber, HvAgentId deviceId)
+static void iSeries_enable_IRQ(unsigned int irq)
 {
-	unsigned int realirq, virtirq;
-	u8 idsel = (deviceId >> 4);
-	u8 function = deviceId & 7;
+	u32 bus, deviceId, function, mask;
+	const u32 subBus = 0;
+	unsigned int rirq = virt_irq_to_real_map[irq];
 
-	virtirq = next_virtual_irq++;
-	realirq = ((busNumber - 1) << 6) + ((idsel - 1) << 3) + function;
-	virt_irq_to_real_map[virtirq] = realirq;
+	/* The IRQ has already been locked by the caller */
+	bus = REAL_IRQ_TO_BUS(rirq);
+	function = REAL_IRQ_TO_FUNC(rirq);
+	deviceId = (REAL_IRQ_TO_IDSEL(rirq) << 4) + function;
 
-	irq_desc[virtirq].handler = &iSeries_IRQ_handler;
-	return virtirq;
+	/* Unmask secondary INTA */
+	mask = 0x80000000;
+	HvCallPci_unmaskInterrupts(bus, subBus, deviceId, mask);
+	PPCDBG(PPCDBG_BUSWALK, "iSeries_enable_IRQ 0x%02X.%02X.%02X 0x%04X\n",
+			bus, subBus, deviceId, irq);
 }
 
-#define REAL_IRQ_TO_BUS(irq)	((((irq) >> 6) & 0xff) + 1)
-#define REAL_IRQ_TO_IDSEL(irq)	((((irq) >> 3) & 7) + 1)
-#define REAL_IRQ_TO_FUNC(irq)	((irq) & 7)
-
 /* This is called by iSeries_activate_IRQs */
 static unsigned int iSeries_startup_IRQ(unsigned int irq)
 {
@@ -131,7 +117,7 @@ void __init iSeries_activate_IRQs()
 			desc->handler->startup(irq);
 			spin_unlock_irqrestore(&desc->lock, flags);
 		}
-    	}
+	}
 }
 
 /*  this is not called anywhere currently */
@@ -173,29 +159,7 @@ static void iSeries_disable_IRQ(unsigned int irq)
 	mask = 0x80000000;
 	HvCallPci_maskInterrupts(bus, subBus, deviceId, mask);
 	PPCDBG(PPCDBG_BUSWALK, "iSeries_disable_IRQ 0x%02X.%02X.%02X 0x%04X\n",
-	       bus, subBus, deviceId, irq);
-}
-
-/*
- * This will be called by device drivers (via enable_IRQ)
- * to enable INTA in the bridge interrupt status register.
- */
-static void iSeries_enable_IRQ(unsigned int irq)
-{
-	u32 bus, deviceId, function, mask;
-	const u32 subBus = 0;
-	unsigned int rirq = virt_irq_to_real_map[irq];
-
-	/* The IRQ has already been locked by the caller */
-	bus = REAL_IRQ_TO_BUS(rirq);
-	function = REAL_IRQ_TO_FUNC(rirq);
-	deviceId = (REAL_IRQ_TO_IDSEL(rirq) << 4) + function;
-
-	/* Unmask secondary INTA */
-	mask = 0x80000000;
-	HvCallPci_unmaskInterrupts(bus, subBus, deviceId, mask);
-	PPCDBG(PPCDBG_BUSWALK, "iSeries_enable_IRQ 0x%02X.%02X.%02X 0x%04X\n",
-	       bus, subBus, deviceId, irq);
+			bus, subBus, deviceId, irq);
 }
 
 /*
@@ -207,3 +171,32 @@ static void iSeries_enable_IRQ(unsigned int irq)
 static void iSeries_end_IRQ(unsigned int irq)
 {
 }
+
+static hw_irq_controller iSeries_IRQ_handler = {
+	.typename = "iSeries irq controller",
+	.startup = iSeries_startup_IRQ,
+	.shutdown = iSeries_shutdown_IRQ,
+	.enable = iSeries_enable_IRQ,
+	.disable = iSeries_disable_IRQ,
+	.end = iSeries_end_IRQ
+};
+
+/*
+ * This is called out of iSeries_scan_slot to allocate an IRQ for an EADS slot
+ * It calculates the irq value for the slot.
+ * Note that subBusNumber is always 0 (at the moment at least).
+ */
+int __init iSeries_allocate_IRQ(HvBusNumber busNumber,
+		HvSubBusNumber subBusNumber, HvAgentId deviceId)
+{
+	unsigned int realirq, virtirq;
+	u8 idsel = (deviceId >> 4);
+	u8 function = deviceId & 7;
+
+	virtirq = next_virtual_irq++;
+	realirq = ((busNumber - 1) << 6) + ((idsel - 1) << 3) + function;
+	virt_irq_to_real_map[virtirq] = realirq;
+
+	irq_desc[virtirq].handler = &iSeries_IRQ_handler;
+	return virtirq;
+}
diff --git a/include/asm-ppc64/iSeries/XmPciLpEvent.h b/include/asm-ppc64/iSeries/XmPciLpEvent.h
deleted file mode 100644
index 4936627b26d..00000000000
--- a/include/asm-ppc64/iSeries/XmPciLpEvent.h
+++ /dev/null
@@ -1,7 +0,0 @@
-#ifndef	__XMPCILPEVENT_H__
-#define	__XMPCILPEVENT_H__
-
-extern int XmPciLpEvent_init(void);
-extern void ppc_irq_dispatch_handler(struct pt_regs *regs, int irq);
-
-#endif /* __XMPCILPEVENT_H__ */
diff --git a/include/asm-ppc64/iSeries/iSeries_irq.h b/include/asm-ppc64/iSeries/iSeries_irq.h
index fbd5f523ba1..a7c5bceae8f 100644
--- a/include/asm-ppc64/iSeries/iSeries_irq.h
+++ b/include/asm-ppc64/iSeries/iSeries_irq.h
@@ -3,7 +3,6 @@
 
 extern void iSeries_init_IRQ(void);
 extern int  iSeries_allocate_IRQ(HvBusNumber, HvSubBusNumber, HvAgentId);
-extern int  iSeries_assign_IRQ(int, HvBusNumber, HvSubBusNumber, HvAgentId);
 extern void iSeries_activate_IRQs(void);
 
 extern int XmPciLpEvent_init(void);
-- 
cgit v1.2.3-70-g09d2


From 7f74e79fe749da035cc150446f02aec29938a5c8 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:51 -0700
Subject: [PATCH] ppc64 iSeries: tidy up irq code after merge

This patch just removes some dead code, fixes messages that referred to the
file this code used to be in and inserts XmPciLpEvent_init into its caller.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/iSeries_irq.c         | 51 ++++++++++++---------------------
 include/asm-ppc64/iSeries/iSeries_irq.h |  2 --
 2 files changed, 18 insertions(+), 35 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/iSeries_irq.c b/arch/ppc64/kernel/iSeries_irq.c
index 96a1f8cd67e..77376c1bd61 100644
--- a/arch/ppc64/kernel/iSeries_irq.c
+++ b/arch/ppc64/kernel/iSeries_irq.c
@@ -105,9 +105,6 @@ static void intReceived(struct XmPciLpEvent *eventParm,
 	int irq;
 
 	++Pci_Interrupt_Count;
-#if 0
-	PPCDBG(PPCDBG_BUSWALK, "PCI: XmPciLpEvent.c: intReceived\n");
-#endif
 
 	switch (eventParm->hvLpEvent.xSubtype) {
 	case XmPciLpEvent_SlotInterrupt:
@@ -120,37 +117,37 @@ static void intReceived(struct XmPciLpEvent *eventParm,
 		break;
 		/* Ignore error recovery events for now */
 	case XmPciLpEvent_BusCreated:
-		printk(KERN_INFO "XmPciLpEvent.c: system bus %d created\n",
+		printk(KERN_INFO "intReceived: system bus %d created\n",
 			eventParm->eventData.busCreated.busNumber);
 		break;
 	case XmPciLpEvent_BusError:
 	case XmPciLpEvent_BusFailed:
-		printk(KERN_INFO "XmPciLpEvent.c: system bus %d failed\n",
+		printk(KERN_INFO "intReceived: system bus %d failed\n",
 			eventParm->eventData.busFailed.busNumber);
 		break;
 	case XmPciLpEvent_BusRecovered:
 	case XmPciLpEvent_UnQuiesceBus:
-		printk(KERN_INFO "XmPciLpEvent.c: system bus %d recovered\n",
+		printk(KERN_INFO "intReceived: system bus %d recovered\n",
 			eventParm->eventData.busRecovered.busNumber);
 		break;
 	case XmPciLpEvent_NodeFailed:
 	case XmPciLpEvent_BridgeError:
 		printk(KERN_INFO
-			"XmPciLpEvent.c: multi-adapter bridge %d/%d/%d failed\n",
+			"intReceived: multi-adapter bridge %d/%d/%d failed\n",
 			eventParm->eventData.nodeFailed.busNumber,
 			eventParm->eventData.nodeFailed.subBusNumber,
 			eventParm->eventData.nodeFailed.deviceId);
 		break;
 	case XmPciLpEvent_NodeRecovered:
 		printk(KERN_INFO
-			"XmPciLpEvent.c: multi-adapter bridge %d/%d/%d recovered\n",
+			"intReceived: multi-adapter bridge %d/%d/%d recovered\n",
 			eventParm->eventData.nodeRecovered.busNumber,
 			eventParm->eventData.nodeRecovered.subBusNumber,
 			eventParm->eventData.nodeRecovered.deviceId);
 		break;
 	default:
 		printk(KERN_ERR
-			"XmPciLpEvent.c: unrecognized event subtype 0x%x\n",
+			"intReceived: unrecognized event subtype 0x%x\n",
 			eventParm->hvLpEvent.xSubtype);
 		break;
 	}
@@ -160,10 +157,6 @@ static void XmPciLpEvent_handler(struct HvLpEvent *eventParm,
 		struct pt_regs *regsParm)
 {
 #ifdef CONFIG_PCI
-#if 0
-	PPCDBG(PPCDBG_BUSWALK, "XmPciLpEvent_handler, type 0x%x\n",
-			eventParm->xType);
-#endif
 	++Pci_Event_Count;
 
 	if (eventParm && (eventParm->xType == HvLpEvent_Type_PciIo)) {
@@ -173,50 +166,42 @@ static void XmPciLpEvent_handler(struct HvLpEvent *eventParm,
 			break;
 		case HvLpEvent_Function_Ack:
 			printk(KERN_ERR
-				"XmPciLpEvent.c: unexpected ack received\n");
+				"XmPciLpEvent_handler: unexpected ack received\n");
 			break;
 		default:
 			printk(KERN_ERR
-				"XmPciLpEvent.c: unexpected event function %d\n",
+				"XmPciLpEvent_handler: unexpected event function %d\n",
 				(int)eventParm->xFlags.xFunction);
 			break;
 		}
 	} else if (eventParm)
 		printk(KERN_ERR
-			"XmPciLpEvent.c: Unrecognized PCI event type 0x%x\n",
+			"XmPciLpEvent_handler: Unrecognized PCI event type 0x%x\n",
 			(int)eventParm->xType);
 	else
-		printk(KERN_ERR "XmPciLpEvent.c: NULL event received\n");
+		printk(KERN_ERR "XmPciLpEvent_handler: NULL event received\n");
 #endif
 }
 
-/* This should be called sometime prior to buswalk (init_IRQ would be good) */
-int XmPciLpEvent_init()
+/*
+ * This is called by init_IRQ.  set in ppc_md.init_IRQ by iSeries_setup.c
+ * It must be called before the bus walk.
+ */
+void __init iSeries_init_IRQ(void)
 {
+	/* Register PCI event handler and open an event path */
 	int xRc;
 
-	PPCDBG(PPCDBG_BUSWALK,
-			"XmPciLpEvent_init, Register Event type 0x%04X\n",
-			HvLpEvent_Type_PciIo);
-
 	xRc = HvLpEvent_registerHandler(HvLpEvent_Type_PciIo,
 			&XmPciLpEvent_handler);
 	if (xRc == 0) {
 		xRc = HvLpEvent_openPath(HvLpEvent_Type_PciIo, 0);
 		if (xRc != 0)
-			printk(KERN_ERR "XmPciLpEvent.c: open event path "
+			printk(KERN_ERR "iSeries_init_IRQ: open event path "
 					"failed with rc 0x%x\n", xRc);
 	} else
-		printk(KERN_ERR "XmPciLpEvent.c: register handler "
+		printk(KERN_ERR "iSeries_init_IRQ: register handler "
 				"failed with rc 0x%x\n", xRc);
-	return xRc;
-}
-
-/* This is called by init_IRQ.  set in ppc_md.init_IRQ by iSeries_setup.c */
-void __init iSeries_init_IRQ(void)
-{
-	/* Register PCI event handler and open an event path */
-	XmPciLpEvent_init();
 }
 
 #define REAL_IRQ_TO_BUS(irq)	((((irq) >> 6) & 0xff) + 1)
diff --git a/include/asm-ppc64/iSeries/iSeries_irq.h b/include/asm-ppc64/iSeries/iSeries_irq.h
index a7c5bceae8f..6c9767ac130 100644
--- a/include/asm-ppc64/iSeries/iSeries_irq.h
+++ b/include/asm-ppc64/iSeries/iSeries_irq.h
@@ -5,6 +5,4 @@ extern void iSeries_init_IRQ(void);
 extern int  iSeries_allocate_IRQ(HvBusNumber, HvSubBusNumber, HvAgentId);
 extern void iSeries_activate_IRQs(void);
 
-extern int XmPciLpEvent_init(void);
-
 #endif /* __ISERIES_IRQ_H__ */
-- 
cgit v1.2.3-70-g09d2


From 145d01e4287b8cbf50f87c3283e33bf5c84e8468 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:52 -0700
Subject: [PATCH] ppc64 iSeries: allow build with no PCI

This patch allows iSeries to build with CONFIG_PCI=n.  This is useful for
partitions that have only virtual I/O.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/Kconfig                | 2 +-
 arch/ppc64/kernel/Makefile        | 6 +++---
 arch/ppc64/kernel/dma.c           | 4 ++++
 arch/ppc64/kernel/iSeries_iommu.c | 3 ++-
 arch/ppc64/kernel/iSeries_setup.c | 8 ++++++++
 arch/ppc64/kernel/sys_ppc32.c     | 3 ++-
 arch/ppc64/lib/Makefile           | 2 +-
 drivers/char/mem.c                | 8 ++++----
 drivers/serial/Kconfig            | 2 +-
 include/asm-ppc64/dma.h           | 3 +++
 include/asm-ppc64/iommu.h         | 4 ++++
 11 files changed, 33 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig
index 5cb343883e4..0f1fa289744 100644
--- a/arch/ppc64/Kconfig
+++ b/arch/ppc64/Kconfig
@@ -323,7 +323,7 @@ config EISA
 	bool
 
 config PCI
-	bool
+	bool "support for PCI devices" if (EMBEDDED && PPC_ISERIES)
 	default y
 	help
 	  Find out whether your system includes a PCI bus. PCI is the name of
diff --git a/arch/ppc64/kernel/Makefile b/arch/ppc64/kernel/Makefile
index ebf9b3cbf33..b5e167cf1a0 100644
--- a/arch/ppc64/kernel/Makefile
+++ b/arch/ppc64/kernel/Makefile
@@ -16,13 +16,13 @@ obj-y += vdso32/ vdso64/
 
 obj-$(CONFIG_PPC_OF) +=	of_device.o
 
-pci-obj-$(CONFIG_PPC_ISERIES)	+= iSeries_pci.o iSeries_VpdInfo.o
+pci-obj-$(CONFIG_PPC_ISERIES)	+= iSeries_pci.o iSeries_irq.o \
+				iSeries_VpdInfo.o
 pci-obj-$(CONFIG_PPC_MULTIPLATFORM)	+= pci_dn.o pci_direct_iommu.o
 
 obj-$(CONFIG_PCI)	+= pci.o pci_iommu.o iomap.o $(pci-obj-y)
 
-obj-$(CONFIG_PPC_ISERIES) += iSeries_irq.o \
-			     HvCall.o HvLpConfig.o LparData.o \
+obj-$(CONFIG_PPC_ISERIES) += HvCall.o HvLpConfig.o LparData.o \
 			     iSeries_setup.o ItLpQueue.o hvCall.o \
 			     mf.o HvLpEvent.o iSeries_proc.o iSeries_htab.o \
 			     iSeries_iommu.o
diff --git a/arch/ppc64/kernel/dma.c b/arch/ppc64/kernel/dma.c
index ce714c92713..4da8e31b2b6 100644
--- a/arch/ppc64/kernel/dma.c
+++ b/arch/ppc64/kernel/dma.c
@@ -15,8 +15,10 @@
 
 static struct dma_mapping_ops *get_dma_ops(struct device *dev)
 {
+#ifdef CONFIG_PCI
 	if (dev->bus == &pci_bus_type)
 		return &pci_dma_ops;
+#endif
 #ifdef CONFIG_IBMVIO
 	if (dev->bus == &vio_bus_type)
 		return &vio_dma_ops;
@@ -37,8 +39,10 @@ EXPORT_SYMBOL(dma_supported);
 
 int dma_set_mask(struct device *dev, u64 dma_mask)
 {
+#ifdef CONFIG_PCI
 	if (dev->bus == &pci_bus_type)
 		return pci_set_dma_mask(to_pci_dev(dev), dma_mask);
+#endif
 #ifdef CONFIG_IBMVIO
 	if (dev->bus == &vio_bus_type)
 		return -EIO;
diff --git a/arch/ppc64/kernel/iSeries_iommu.c b/arch/ppc64/kernel/iSeries_iommu.c
index 4e1a47c8a80..f8ff1bb054d 100644
--- a/arch/ppc64/kernel/iSeries_iommu.c
+++ b/arch/ppc64/kernel/iSeries_iommu.c
@@ -83,7 +83,7 @@ static void tce_free_iSeries(struct iommu_table *tbl, long index, long npages)
 	}
 }
 
-
+#ifdef CONFIG_PCI
 /*
  * This function compares the known tables to find an iommu_table
  * that has already been built for hardware TCEs.
@@ -159,6 +159,7 @@ void iommu_devnode_init_iSeries(struct iSeries_Device_Node *dn)
 	else
 		kfree(tbl);
 }
+#endif
 
 static void iommu_dev_setup_iSeries(struct pci_dev *dev) { }
 static void iommu_bus_setup_iSeries(struct pci_bus *bus) { }
diff --git a/arch/ppc64/kernel/iSeries_setup.c b/arch/ppc64/kernel/iSeries_setup.c
index 3d3ed631499..b31962436fe 100644
--- a/arch/ppc64/kernel/iSeries_setup.c
+++ b/arch/ppc64/kernel/iSeries_setup.c
@@ -76,7 +76,11 @@ extern void ppcdbg_initialize(void);
 static void build_iSeries_Memory_Map(void);
 static void setup_iSeries_cache_sizes(void);
 static void iSeries_bolt_kernel(unsigned long saddr, unsigned long eaddr);
+#ifdef CONFIG_PCI
 extern void iSeries_pci_final_fixup(void);
+#else
+static void iSeries_pci_final_fixup(void) { }
+#endif
 
 /* Global Variables */
 static unsigned long procFreqHz;
@@ -876,6 +880,10 @@ static int set_spread_lpevents(char *str)
 }
 __setup("spread_lpevents=", set_spread_lpevents);
 
+#ifndef CONFIG_PCI
+void __init iSeries_init_IRQ(void) { }
+#endif
+
 void __init iSeries_early_setup(void)
 {
 	iSeries_fixup_klimit();
diff --git a/arch/ppc64/kernel/sys_ppc32.c b/arch/ppc64/kernel/sys_ppc32.c
index 9c8e317c598..118436e8085 100644
--- a/arch/ppc64/kernel/sys_ppc32.c
+++ b/arch/ppc64/kernel/sys_ppc32.c
@@ -741,6 +741,7 @@ asmlinkage int sys32_pciconfig_write(u32 bus, u32 dfn, u32 off, u32 len, u32 ubu
 
 asmlinkage int sys32_pciconfig_iobase(u32 which, u32 in_bus, u32 in_devfn)
 {
+#ifdef CONFIG_PCI
 	struct pci_controller* hose;
 	struct list_head *ln;
 	struct pci_bus *bus = NULL;
@@ -786,7 +787,7 @@ asmlinkage int sys32_pciconfig_iobase(u32 which, u32 in_bus, u32 in_devfn)
 	case IOBASE_ISA_MEM:
 		return -EINVAL;
 	}
-
+#endif	/* CONFIG_PCI */
 	return -EOPNOTSUPP;
 }
 
diff --git a/arch/ppc64/lib/Makefile b/arch/ppc64/lib/Makefile
index bf7b5bbfc04..76fbfa9f706 100644
--- a/arch/ppc64/lib/Makefile
+++ b/arch/ppc64/lib/Makefile
@@ -12,7 +12,7 @@ lib-$(CONFIG_SMP) += locks.o
 
 # e2a provides EBCDIC to ASCII conversions.
 ifdef CONFIG_PPC_ISERIES
-obj-$(CONFIG_PCI)	+= e2a.o
+obj-y += e2a.o
 endif
 
 lib-$(CONFIG_DEBUG_KERNEL) += sstep.o
diff --git a/drivers/char/mem.c b/drivers/char/mem.c
index 257b8ee605e..e3085b22a36 100644
--- a/drivers/char/mem.c
+++ b/drivers/char/mem.c
@@ -484,7 +484,7 @@ static ssize_t write_kmem(struct file * file, const char __user * buf,
  	return virtr + wrote;
 }
 
-#if defined(CONFIG_ISA) || !defined(__mc68000__)
+#if (defined(CONFIG_ISA) || !defined(__mc68000__)) && (!defined(CONFIG_PPC_ISERIES) || defined(CONFIG_PCI))
 static ssize_t read_port(struct file * file, char __user * buf,
 			 size_t count, loff_t *ppos)
 {
@@ -744,7 +744,7 @@ static struct file_operations null_fops = {
 	.write		= write_null,
 };
 
-#if defined(CONFIG_ISA) || !defined(__mc68000__)
+#if (defined(CONFIG_ISA) || !defined(__mc68000__)) && (!defined(CONFIG_PPC_ISERIES) || defined(CONFIG_PCI))
 static struct file_operations port_fops = {
 	.llseek		= memory_lseek,
 	.read		= read_port,
@@ -804,7 +804,7 @@ static int memory_open(struct inode * inode, struct file * filp)
 		case 3:
 			filp->f_op = &null_fops;
 			break;
-#if defined(CONFIG_ISA) || !defined(__mc68000__)
+#if (defined(CONFIG_ISA) || !defined(__mc68000__)) && (!defined(CONFIG_PPC_ISERIES) || defined(CONFIG_PCI))
 		case 4:
 			filp->f_op = &port_fops;
 			break;
@@ -846,7 +846,7 @@ static const struct {
 	{1, "mem",     S_IRUSR | S_IWUSR | S_IRGRP, &mem_fops},
 	{2, "kmem",    S_IRUSR | S_IWUSR | S_IRGRP, &kmem_fops},
 	{3, "null",    S_IRUGO | S_IWUGO,           &null_fops},
-#if defined(CONFIG_ISA) || !defined(__mc68000__)
+#if (defined(CONFIG_ISA) || !defined(__mc68000__)) && (!defined(CONFIG_PPC_ISERIES) || defined(CONFIG_PCI))
 	{4, "port",    S_IRUSR | S_IWUSR | S_IRGRP, &port_fops},
 #endif
 	{5, "zero",    S_IRUGO | S_IWUGO,           &zero_fops},
diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig
index 6e44b46c9e9..6a15703f1cb 100644
--- a/drivers/serial/Kconfig
+++ b/drivers/serial/Kconfig
@@ -753,7 +753,7 @@ config SERIAL_MPC52xx_CONSOLE_BAUD
 
 config SERIAL_ICOM
 	tristate "IBM Multiport Serial Adapter"
-	depends on PPC_ISERIES || PPC_PSERIES
+	depends on PCI && (PPC_ISERIES || PPC_PSERIES)
 	select SERIAL_CORE
 	help
 	  This driver is for a family of multiport serial adapters
diff --git a/include/asm-ppc64/dma.h b/include/asm-ppc64/dma.h
index d693b8026bc..dfd1f69059b 100644
--- a/include/asm-ppc64/dma.h
+++ b/include/asm-ppc64/dma.h
@@ -27,6 +27,8 @@
 /* Doesn't really apply... */
 #define MAX_DMA_ADDRESS  (~0UL)
 
+#if !defined(CONFIG_PPC_ISERIES) || defined(CONFIG_PCI)
+
 #define dma_outb	outb
 #define dma_inb		inb
 
@@ -323,4 +325,5 @@ extern int isa_dma_bridge_buggy;
 #else                                                         
 #define isa_dma_bridge_buggy   (0)
 #endif
+#endif	/* !defined(CONFIG_PPC_ISERIES) || defined(CONFIG_PCI) */
 #endif /* _ASM_DMA_H */
diff --git a/include/asm-ppc64/iommu.h b/include/asm-ppc64/iommu.h
index d9b53374d23..729de5cc21d 100644
--- a/include/asm-ppc64/iommu.h
+++ b/include/asm-ppc64/iommu.h
@@ -137,8 +137,12 @@ extern void iommu_init_early_pSeries(void);
 extern void iommu_init_early_iSeries(void);
 extern void iommu_init_early_u3(void);
 
+#ifdef CONFIG_PCI
 extern void pci_iommu_init(void);
 extern void pci_direct_iommu_init(void);
+#else
+static inline void pci_iommu_init(void) { }
+#endif
 
 extern void alloc_u3_dart_table(void);
 
-- 
cgit v1.2.3-70-g09d2


From e400bae98499583767da58fb0a1b9ad3e24fcb86 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
Date: Tue, 21 Jun 2005 17:15:56 -0700
Subject: [PATCH] mips: add vr41xx gpio support

Add vr41xx gpio support.

Signed-off-by: Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/Kconfig             |   4 +
 drivers/char/Makefile            |   1 +
 drivers/char/vr41xx_giu.c        | 743 +++++++++++++++++++++++++++++++++++++++
 include/asm-mips/vr41xx/giu.h    |  69 ++++
 include/asm-mips/vr41xx/vr41xx.h |  35 --
 5 files changed, 817 insertions(+), 35 deletions(-)
 create mode 100644 drivers/char/vr41xx_giu.c
 create mode 100644 include/asm-mips/vr41xx/giu.h

(limited to 'include')

diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig
index 5ed6515ae01..7ccf871d3c9 100644
--- a/drivers/char/Kconfig
+++ b/drivers/char/Kconfig
@@ -929,6 +929,10 @@ config SCx200_GPIO
 
 	  If compiled as a module, it will be called scx200_gpio.
 
+config GPIO_VR41XX
+	tristate "NEC VR4100 series General-purpose I/O Unit support"
+	depends on CPU_VR41XX
+
 config RAW_DRIVER
 	tristate "RAW driver (/dev/raw/rawN) (OBSOLETE)"
 	help
diff --git a/drivers/char/Makefile b/drivers/char/Makefile
index e3f5c32aac5..1aff819f383 100644
--- a/drivers/char/Makefile
+++ b/drivers/char/Makefile
@@ -80,6 +80,7 @@ obj-$(CONFIG_PPDEV) += ppdev.o
 obj-$(CONFIG_NWBUTTON) += nwbutton.o
 obj-$(CONFIG_NWFLASH) += nwflash.o
 obj-$(CONFIG_SCx200_GPIO) += scx200_gpio.o
+obj-$(CONFIG_GPIO_VR41XX) += vr41xx_giu.o
 obj-$(CONFIG_TANBAC_TB0219) += tb0219.o
 
 obj-$(CONFIG_WATCHDOG)	+= watchdog/
diff --git a/drivers/char/vr41xx_giu.c b/drivers/char/vr41xx_giu.c
new file mode 100644
index 00000000000..683278bc524
--- /dev/null
+++ b/drivers/char/vr41xx_giu.c
@@ -0,0 +1,743 @@
+/*
+ *  Driver for NEC VR4100 series General-purpose I/O Unit.
+ *
+ *  Copyright (C) 2002 MontaVista Software Inc.
+ *	Author: Yoichi Yuasa <yyuasa@mvista.com or source@mvista.com>
+ *  Copyright (C) 2003-2005  Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+#include <linux/device.h>
+#include <linux/errno.h>
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/spinlock.h>
+#include <linux/types.h>
+
+#include <asm/cpu.h>
+#include <asm/io.h>
+#include <asm/vr41xx/giu.h>
+#include <asm/vr41xx/vr41xx.h>
+
+MODULE_AUTHOR("Yoichi Yuasa <yuasa@hh.iij4u.or.jp>");
+MODULE_DESCRIPTION("NEC VR4100 series General-purpose I/O Unit driver");
+MODULE_LICENSE("GPL");
+
+static int major;	/* default is dynamic major device number */
+module_param(major, int, 0);
+MODULE_PARM_DESC(major, "Major device number");
+
+#define GIU_TYPE1_START		0x0b000100UL
+#define GIU_TYPE1_SIZE		0x20UL
+
+#define GIU_TYPE2_START		0x0f000140UL
+#define GIU_TYPE2_SIZE		0x20UL
+
+#define GIU_TYPE3_START		0x0f000140UL
+#define GIU_TYPE3_SIZE		0x28UL
+
+#define GIU_PULLUPDOWN_START	0x0b0002e0UL
+#define GIU_PULLUPDOWN_SIZE	0x04UL
+
+#define GIUIOSELL	0x00
+#define GIUIOSELH	0x02
+#define GIUPIODL	0x04
+#define GIUPIODH	0x06
+#define GIUINTSTATL	0x08
+#define GIUINTSTATH	0x0a
+#define GIUINTENL	0x0c
+#define GIUINTENH	0x0e
+#define GIUINTTYPL	0x10
+#define GIUINTTYPH	0x12
+#define GIUINTALSELL	0x14
+#define GIUINTALSELH	0x16
+#define GIUINTHTSELL	0x18
+#define GIUINTHTSELH	0x1a
+#define GIUPODATL	0x1c
+#define GIUPODATEN	0x1c
+#define GIUPODATH	0x1e
+ #define PIOEN0		0x0100
+ #define PIOEN1		0x0200
+#define GIUPODAT	0x1e
+#define GIUFEDGEINHL	0x20
+#define GIUFEDGEINHH	0x22
+#define GIUREDGEINHL	0x24
+#define GIUREDGEINHH	0x26
+
+#define GIUUSEUPDN	0x1e0
+#define GIUTERMUPDN	0x1e2
+
+#define GPIO_HAS_PULLUPDOWN_IO		0x0001
+#define GPIO_HAS_OUTPUT_ENABLE		0x0002
+#define GPIO_HAS_INTERRUPT_EDGE_SELECT	0x0100
+
+static spinlock_t giu_lock;
+static struct resource *giu_resource1;
+static struct resource *giu_resource2;
+static unsigned long giu_flags;
+static unsigned int giu_nr_pins;
+
+static void __iomem *giu_base;
+
+#define giu_read(offset)		readw(giu_base + (offset))
+#define giu_write(offset, value)	writew((value), giu_base + (offset))
+
+#define GPIO_PIN_OF_IRQ(irq)	((irq) - GIU_IRQ_BASE)
+#define GIUINT_HIGH_OFFSET	16
+#define GIUINT_HIGH_MAX		32
+
+static inline uint16_t giu_set(uint16_t offset, uint16_t set)
+{
+	uint16_t data;
+
+	data = giu_read(offset);
+	data |= set;
+	giu_write(offset, data);
+
+	return data;
+}
+
+static inline uint16_t giu_clear(uint16_t offset, uint16_t clear)
+{
+	uint16_t data;
+
+	data = giu_read(offset);
+	data &= ~clear;
+	giu_write(offset, data);
+
+	return data;
+}
+
+static unsigned int startup_giuint_low_irq(unsigned int irq)
+{
+	unsigned int pin;
+
+	pin = GPIO_PIN_OF_IRQ(irq);
+	giu_write(GIUINTSTATL, 1 << pin);
+	giu_set(GIUINTENL, 1 << pin);
+
+	return 0;
+}
+
+static void shutdown_giuint_low_irq(unsigned int irq)
+{
+	giu_clear(GIUINTENL, 1 << GPIO_PIN_OF_IRQ(irq));
+}
+
+static void enable_giuint_low_irq(unsigned int irq)
+{
+	giu_set(GIUINTENL, 1 << GPIO_PIN_OF_IRQ(irq));
+}
+
+#define disable_giuint_low_irq	shutdown_giuint_low_irq
+
+static void ack_giuint_low_irq(unsigned int irq)
+{
+	unsigned int pin;
+
+	pin = GPIO_PIN_OF_IRQ(irq);
+	giu_clear(GIUINTENL, 1 << pin);
+	giu_write(GIUINTSTATL, 1 << pin);
+}
+
+static void end_giuint_low_irq(unsigned int irq)
+{
+	if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS)))
+		giu_set(GIUINTENL, 1 << GPIO_PIN_OF_IRQ(irq));
+}
+
+static struct hw_interrupt_type giuint_low_irq_type = {
+	.typename	= "GIUINTL",
+	.startup	= startup_giuint_low_irq,
+	.shutdown	= shutdown_giuint_low_irq,
+	.enable		= enable_giuint_low_irq,
+	.disable	= disable_giuint_low_irq,
+	.ack		= ack_giuint_low_irq,
+	.end		= end_giuint_low_irq,
+};
+
+static unsigned int startup_giuint_high_irq(unsigned int irq)
+{
+	unsigned int pin;
+
+	pin = GPIO_PIN_OF_IRQ(irq) - GIUINT_HIGH_OFFSET;
+	giu_write(GIUINTSTATH, 1 << pin);
+	giu_set(GIUINTENH, 1 << pin);
+
+	return 0;
+}
+
+static void shutdown_giuint_high_irq(unsigned int irq)
+{
+	giu_clear(GIUINTENH, 1 << (GPIO_PIN_OF_IRQ(irq) - GIUINT_HIGH_OFFSET));
+}
+
+static void enable_giuint_high_irq(unsigned int irq)
+{
+	giu_set(GIUINTENH, 1 << (GPIO_PIN_OF_IRQ(irq) - GIUINT_HIGH_OFFSET));
+}
+
+#define disable_giuint_high_irq	shutdown_giuint_high_irq
+
+static void ack_giuint_high_irq(unsigned int irq)
+{
+	unsigned int pin;
+
+	pin = GPIO_PIN_OF_IRQ(irq) - GIUINT_HIGH_OFFSET;
+	giu_clear(GIUINTENH, 1 << pin);
+	giu_write(GIUINTSTATH, 1 << pin);
+}
+
+static void end_giuint_high_irq(unsigned int irq)
+{
+	if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS)))
+		giu_set(GIUINTENH, 1 << (GPIO_PIN_OF_IRQ(irq) - GIUINT_HIGH_OFFSET));
+}
+
+static struct hw_interrupt_type giuint_high_irq_type = {
+	.typename	= "GIUINTH",
+	.startup	= startup_giuint_high_irq,
+	.shutdown	= shutdown_giuint_high_irq,
+	.enable		= enable_giuint_high_irq,
+	.disable	= disable_giuint_high_irq,
+	.ack		= ack_giuint_high_irq,
+	.end		= end_giuint_high_irq,
+};
+
+static int giu_get_irq(unsigned int irq, struct pt_regs *regs)
+{
+	uint16_t pendl, pendh, maskl, maskh;
+	int i;
+
+	pendl = giu_read(GIUINTSTATL);
+	pendh = giu_read(GIUINTSTATH);
+	maskl = giu_read(GIUINTENL);
+	maskh = giu_read(GIUINTENH);
+
+	maskl &= pendl;
+	maskh &= pendh;
+
+	if (maskl) {
+		for (i = 0; i < 16; i++) {
+			if (maskl & (1 << i))
+				return GIU_IRQ(i);
+		}
+	} else if (maskh) {
+		for (i = 0; i < 16; i++) {
+			if (maskh & (1 << i))
+				return GIU_IRQ(i + GIUINT_HIGH_OFFSET);
+		}
+	}
+
+	printk(KERN_ERR "spurious GIU interrupt: %04x(%04x),%04x(%04x)\n",
+	       maskl, pendl, maskh, pendh);
+
+	atomic_inc(&irq_err_count);
+
+	return -EINVAL;
+}
+
+void vr41xx_set_irq_trigger(unsigned int pin, irq_trigger_t trigger, irq_signal_t signal)
+{
+	uint16_t mask;
+
+	if (pin < GIUINT_HIGH_OFFSET) {
+		mask = 1 << pin;
+		if (trigger != IRQ_TRIGGER_LEVEL) {
+        		giu_set(GIUINTTYPL, mask);
+			if (signal == IRQ_SIGNAL_HOLD)
+				giu_set(GIUINTHTSELL, mask);
+			else
+				giu_clear(GIUINTHTSELL, mask);
+			if (current_cpu_data.cputype == CPU_VR4133) {
+				switch (trigger) {
+				case IRQ_TRIGGER_EDGE_FALLING:
+					giu_set(GIUFEDGEINHL, mask);
+					giu_clear(GIUREDGEINHL, mask);
+					break;
+				case IRQ_TRIGGER_EDGE_RISING:
+					giu_clear(GIUFEDGEINHL, mask);
+					giu_set(GIUREDGEINHL, mask);
+					break;
+				default:
+					giu_set(GIUFEDGEINHL, mask);
+					giu_set(GIUREDGEINHL, mask);
+					break;
+				}
+			}
+		} else {
+			giu_clear(GIUINTTYPL, mask);
+			giu_clear(GIUINTHTSELL, mask);
+		}
+		giu_write(GIUINTSTATL, mask);
+	} else if (pin < GIUINT_HIGH_MAX) {
+		mask = 1 << (pin - GIUINT_HIGH_OFFSET);
+		if (trigger != IRQ_TRIGGER_LEVEL) {
+			giu_set(GIUINTTYPH, mask);
+			if (signal == IRQ_SIGNAL_HOLD)
+				giu_set(GIUINTHTSELH, mask);
+			else
+				giu_clear(GIUINTHTSELH, mask);
+			if (current_cpu_data.cputype == CPU_VR4133) {
+				switch (trigger) {
+				case IRQ_TRIGGER_EDGE_FALLING:
+					giu_set(GIUFEDGEINHH, mask);
+					giu_clear(GIUREDGEINHH, mask);
+					break;
+				case IRQ_TRIGGER_EDGE_RISING:
+					giu_clear(GIUFEDGEINHH, mask);
+					giu_set(GIUREDGEINHH, mask);
+					break;
+				default:
+					giu_set(GIUFEDGEINHH, mask);
+					giu_set(GIUREDGEINHH, mask);
+					break;
+				}
+			}
+		} else {
+			giu_clear(GIUINTTYPH, mask);
+			giu_clear(GIUINTHTSELH, mask);
+		}
+		giu_write(GIUINTSTATH, mask);
+	}
+}
+
+EXPORT_SYMBOL_GPL(vr41xx_set_irq_trigger);
+
+void vr41xx_set_irq_level(unsigned int pin, irq_level_t level)
+{
+	uint16_t mask;
+
+	if (pin < GIUINT_HIGH_OFFSET) {
+		mask = 1 << pin;
+		if (level == IRQ_LEVEL_HIGH)
+			giu_set(GIUINTALSELL, mask);
+		else
+			giu_clear(GIUINTALSELL, mask);
+		giu_write(GIUINTSTATL, mask);
+	} else if (pin < GIUINT_HIGH_MAX) {
+		mask = 1 << (pin - GIUINT_HIGH_OFFSET);
+		if (level == IRQ_LEVEL_HIGH)
+			giu_set(GIUINTALSELH, mask);
+		else
+			giu_clear(GIUINTALSELH, mask);
+		giu_write(GIUINTSTATH, mask);
+	}
+}
+
+EXPORT_SYMBOL_GPL(vr41xx_set_irq_level);
+
+gpio_data_t vr41xx_gpio_get_pin(unsigned int pin)
+{
+	uint16_t reg, mask;
+
+	if (pin >= giu_nr_pins)
+		return GPIO_DATA_INVAL;
+
+	if (pin < 16) {
+		reg = giu_read(GIUPIODL);
+		mask = (uint16_t)1 << pin;
+	} else if (pin < 32) {
+		reg = giu_read(GIUPIODH);
+		mask = (uint16_t)1 << (pin - 16);
+	} else if (pin < 48) {
+		reg = giu_read(GIUPODATL);
+		mask = (uint16_t)1 << (pin - 32);
+	} else {
+		reg = giu_read(GIUPODATH);
+		mask = (uint16_t)1 << (pin - 48);
+	}
+
+	if (reg & mask)
+		return GPIO_DATA_HIGH;
+
+	return GPIO_DATA_LOW;
+}
+
+EXPORT_SYMBOL_GPL(vr41xx_gpio_get_pin);
+
+int vr41xx_gpio_set_pin(unsigned int pin, gpio_data_t data)
+{
+	uint16_t offset, mask, reg;
+	unsigned long flags;
+
+	if (pin >= giu_nr_pins)
+		return -EINVAL;
+
+	if (pin < 16) {
+		offset = GIUPIODL;
+		mask = (uint16_t)1 << pin;
+	} else if (pin < 32) {
+		offset = GIUPIODH;
+		mask = (uint16_t)1 << (pin - 16);
+	} else if (pin < 48) {
+		offset = GIUPODATL;
+		mask = (uint16_t)1 << (pin - 32);
+	} else {
+		offset = GIUPODATH;
+		mask = (uint16_t)1 << (pin - 48);
+	}
+
+	spin_lock_irqsave(&giu_lock, flags);
+
+	reg = giu_read(offset);
+	if (data == GPIO_DATA_HIGH)
+		reg |= mask;
+	else
+		reg &= ~mask;
+	giu_write(offset, reg);
+
+	spin_unlock_irqrestore(&giu_lock, flags);
+
+	return 0;
+}
+
+EXPORT_SYMBOL_GPL(vr41xx_gpio_set_pin);
+
+int vr41xx_gpio_set_direction(unsigned int pin, gpio_direction_t dir)
+{
+	uint16_t offset, mask, reg;
+	unsigned long flags;
+
+	if (pin >= giu_nr_pins)
+		return -EINVAL;
+
+	if (pin < 16) {
+		offset = GIUIOSELL;
+		mask = (uint16_t)1 << pin;
+	} else if (pin < 32) {
+		offset = GIUIOSELH;
+		mask = (uint16_t)1 << (pin - 16);
+	} else {
+		if (giu_flags & GPIO_HAS_OUTPUT_ENABLE) {
+			offset = GIUPODATEN;
+			mask = (uint16_t)1 << (pin - 32);
+		} else {
+			switch (pin) {
+			case 48:
+				offset = GIUPODATH;
+				mask = PIOEN0;
+				break;
+			case 49:
+				offset = GIUPODATH;
+				mask = PIOEN1;
+				break;
+			default:
+				return -EINVAL;
+			}
+		}
+	}
+
+	spin_lock_irqsave(&giu_lock, flags);
+
+	reg = giu_read(offset);
+	if (dir == GPIO_OUTPUT)
+		reg |= mask;
+	else
+		reg &= ~mask;
+	giu_write(offset, reg);
+
+	spin_unlock_irqrestore(&giu_lock, flags);
+
+	return 0;
+}
+
+EXPORT_SYMBOL_GPL(vr41xx_gpio_set_direction);
+
+int vr41xx_gpio_pullupdown(unsigned int pin, gpio_pull_t pull)
+{
+	uint16_t reg, mask;
+	unsigned long flags;
+
+	if ((giu_flags & GPIO_HAS_PULLUPDOWN_IO) != GPIO_HAS_PULLUPDOWN_IO)
+		return -EPERM;
+
+	if (pin >= 15)
+		return -EINVAL;
+
+	mask = (uint16_t)1 << pin;
+
+	spin_lock_irqsave(&giu_lock, flags);
+
+	if (pull == GPIO_PULL_UP || pull == GPIO_PULL_DOWN) {
+		reg = giu_read(GIUTERMUPDN);
+		if (pull == GPIO_PULL_UP)
+			reg |= mask;
+		else
+			reg &= ~mask;
+		giu_write(GIUTERMUPDN, reg);
+
+		reg = giu_read(GIUUSEUPDN);
+		reg |= mask;
+		giu_write(GIUUSEUPDN, reg);
+	} else {
+		reg = giu_read(GIUUSEUPDN);
+		reg &= ~mask;
+		giu_write(GIUUSEUPDN, reg);
+	}
+
+	spin_unlock_irqrestore(&giu_lock, flags);
+
+	return 0;
+}
+
+EXPORT_SYMBOL_GPL(vr41xx_gpio_pullupdown);
+
+static ssize_t gpio_read(struct file *file, char __user *buf, size_t len,
+                         loff_t *ppos)
+{
+	unsigned int pin;
+	char value = '0';
+
+	pin = iminor(file->f_dentry->d_inode);
+	if (pin >= giu_nr_pins)
+		return -EBADF;
+
+	if (vr41xx_gpio_get_pin(pin) == GPIO_DATA_HIGH)
+		value = '1';
+
+	if (len <= 0)
+		return -EFAULT;
+
+	if (put_user(value, buf))
+		return -EFAULT;
+
+	return 1;
+}
+
+static ssize_t gpio_write(struct file *file, const char __user *data,
+                          size_t len, loff_t *ppos)
+{
+	unsigned int pin;
+	size_t i;
+	char c;
+	int retval = 0;
+
+	pin = iminor(file->f_dentry->d_inode);
+	if (pin >= giu_nr_pins)
+		return -EBADF;
+
+	for (i = 0; i < len; i++) {
+		if (get_user(c, data + i))
+			return -EFAULT;
+
+		switch (c) {
+		case '0':
+			retval = vr41xx_gpio_set_pin(pin, GPIO_DATA_LOW);
+			break;
+		case '1':
+			retval = vr41xx_gpio_set_pin(pin, GPIO_DATA_HIGH);
+			break;
+		case 'D':
+			printk(KERN_INFO "GPIO%d: pull down\n", pin);
+			retval = vr41xx_gpio_pullupdown(pin, GPIO_PULL_DOWN);
+			break;
+		case 'd':
+			printk(KERN_INFO "GPIO%d: pull up/down disable\n", pin);
+			retval = vr41xx_gpio_pullupdown(pin, GPIO_PULL_DISABLE);
+			break;
+		case 'I':
+			printk(KERN_INFO "GPIO%d: input\n", pin);
+			retval = vr41xx_gpio_set_direction(pin, GPIO_INPUT);
+			break;
+		case 'O':
+			printk(KERN_INFO "GPIO%d: output\n", pin);
+			retval = vr41xx_gpio_set_direction(pin, GPIO_OUTPUT);
+			break;
+		case 'o':
+			printk(KERN_INFO "GPIO%d: output disable\n", pin);
+			retval = vr41xx_gpio_set_direction(pin, GPIO_OUTPUT_DISABLE);
+			break;
+		case 'P':
+			printk(KERN_INFO "GPIO%d: pull up\n", pin);
+			retval = vr41xx_gpio_pullupdown(pin, GPIO_PULL_UP);
+			break;
+		case 'p':
+			printk(KERN_INFO "GPIO%d: pull up/down disable\n", pin);
+			retval = vr41xx_gpio_pullupdown(pin, GPIO_PULL_DISABLE);
+			break;
+		default:
+			break;
+		}
+
+		if (retval < 0)
+			break;
+	}
+
+	return i;
+}
+
+static int gpio_open(struct inode *inode, struct file *file)
+{
+	unsigned int pin;
+
+	pin = iminor(inode);
+	if (pin >= giu_nr_pins)
+		return -EBADF;
+
+	return nonseekable_open(inode, file);
+}
+
+static int gpio_release(struct inode *inode, struct file *file)
+{
+	unsigned int pin;
+
+	pin = iminor(inode);
+	if (pin >= giu_nr_pins)
+		return -EBADF;
+
+	return 0;
+}
+
+static struct file_operations gpio_fops = {
+	.owner		= THIS_MODULE,
+	.read		= gpio_read,
+	.write		= gpio_write,
+	.open		= gpio_open,
+	.release	= gpio_release,
+};
+
+static int giu_probe(struct device *dev)
+{
+	unsigned long start, size, flags = 0;
+	unsigned int nr_pins = 0;
+	struct resource *res1, *res2 = NULL;
+	void *base;
+	int retval, i;
+
+	switch (current_cpu_data.cputype) {
+	case CPU_VR4111:
+	case CPU_VR4121:
+		start = GIU_TYPE1_START;
+		size = GIU_TYPE1_SIZE;
+		flags = GPIO_HAS_PULLUPDOWN_IO;
+		nr_pins = 50;
+		break;
+	case CPU_VR4122:
+	case CPU_VR4131:
+		start = GIU_TYPE2_START;
+		size = GIU_TYPE2_SIZE;
+		nr_pins = 36;
+		break;
+	case CPU_VR4133:
+		start = GIU_TYPE3_START;
+		size = GIU_TYPE3_SIZE;
+		flags = GPIO_HAS_INTERRUPT_EDGE_SELECT;
+		nr_pins = 48;
+		break;
+	default:
+		return -ENODEV;
+	}
+
+	res1 = request_mem_region(start, size, "GIU");
+	if (res1 == NULL)
+		return -EBUSY;
+
+	base = ioremap(start, size);
+	if (base == NULL) {
+		release_resource(res1);
+		return -ENOMEM;
+	}
+
+	if (flags & GPIO_HAS_PULLUPDOWN_IO) {
+		res2 = request_mem_region(GIU_PULLUPDOWN_START, GIU_PULLUPDOWN_SIZE, "GIU");
+		if (res2 == NULL) {
+			iounmap(base);
+			release_resource(res1);
+			return -EBUSY;
+		}
+	}
+
+	retval = register_chrdev(major, "GIU", &gpio_fops);
+	if (retval < 0) {
+		iounmap(base);
+		release_resource(res1);
+		release_resource(res2);
+		return retval;
+	}
+
+	if (major == 0) {
+		major = retval;
+		printk(KERN_INFO "GIU: major number %d\n", major);
+	}
+
+	spin_lock_init(&giu_lock);
+	giu_base = base;
+	giu_resource1 = res1;
+	giu_resource2 = res2;
+	giu_flags = flags;
+	giu_nr_pins = nr_pins;
+
+	giu_write(GIUINTENL, 0);
+	giu_write(GIUINTENH, 0);
+
+	for (i = GIU_IRQ_BASE; i <= GIU_IRQ_LAST; i++) {
+		if (i < GIU_IRQ(GIUINT_HIGH_OFFSET))
+			irq_desc[i].handler = &giuint_low_irq_type;
+		else
+			irq_desc[i].handler = &giuint_high_irq_type;
+	}
+
+	return cascade_irq(GIUINT_IRQ, giu_get_irq);
+}
+
+static int giu_remove(struct device *dev)
+{
+	iounmap(giu_base);
+
+	release_resource(giu_resource1);
+	if (giu_flags & GPIO_HAS_PULLUPDOWN_IO)
+		release_resource(giu_resource2);
+
+	return 0;
+}
+
+static struct platform_device *giu_platform_device;
+
+static struct device_driver giu_device_driver = {
+	.name		= "GIU",
+	.bus		= &platform_bus_type,
+	.probe		= giu_probe,
+	.remove		= giu_remove,
+};
+
+static int __devinit vr41xx_giu_init(void)
+{
+	int retval;
+
+	giu_platform_device = platform_device_register_simple("GIU", -1, NULL, 0);
+	if (IS_ERR(giu_platform_device))
+		return PTR_ERR(giu_platform_device);
+
+	retval = driver_register(&giu_device_driver);
+	if (retval < 0)
+		platform_device_unregister(giu_platform_device);
+
+	return retval;
+}
+
+static void __devexit vr41xx_giu_exit(void)
+{
+	driver_unregister(&giu_device_driver);
+
+	platform_device_unregister(giu_platform_device);
+}
+
+module_init(vr41xx_giu_init);
+module_exit(vr41xx_giu_exit);
diff --git a/include/asm-mips/vr41xx/giu.h b/include/asm-mips/vr41xx/giu.h
new file mode 100644
index 00000000000..8590885a763
--- /dev/null
+++ b/include/asm-mips/vr41xx/giu.h
@@ -0,0 +1,69 @@
+/*
+ *  Include file for NEC VR4100 series General-purpose I/O Unit.
+ *
+ *  Copyright (C) 2005  Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+#ifndef __NEC_VR41XX_GIU_H
+#define __NEC_VR41XX_GIU_H
+
+typedef enum {
+	IRQ_TRIGGER_LEVEL,
+	IRQ_TRIGGER_EDGE,
+	IRQ_TRIGGER_EDGE_FALLING,
+	IRQ_TRIGGER_EDGE_RISING,
+} irq_trigger_t;
+
+typedef enum {
+	IRQ_SIGNAL_THROUGH,
+	IRQ_SIGNAL_HOLD,
+} irq_signal_t;
+
+extern void vr41xx_set_irq_trigger(unsigned int pin, irq_trigger_t trigger, irq_signal_t signal);
+
+typedef enum {
+	IRQ_LEVEL_LOW,
+	IRQ_LEVEL_HIGH,
+} irq_level_t;
+
+extern void vr41xx_set_irq_level(unsigned int pin, irq_level_t level);
+
+typedef enum {
+	GPIO_DATA_LOW,
+	GPIO_DATA_HIGH,
+	GPIO_DATA_INVAL,
+} gpio_data_t;
+
+extern gpio_data_t vr41xx_gpio_get_pin(unsigned int pin);
+extern int vr41xx_gpio_set_pin(unsigned int pin, gpio_data_t data);
+
+typedef enum {
+	GPIO_INPUT,
+	GPIO_OUTPUT,
+	GPIO_OUTPUT_DISABLE,
+} gpio_direction_t;
+
+extern int vr41xx_gpio_set_direction(unsigned int pin, gpio_direction_t dir);
+
+typedef enum {
+	GPIO_PULL_DOWN,
+	GPIO_PULL_UP,
+	GPIO_PULL_DISABLE,
+} gpio_pull_t;
+
+extern int vr41xx_gpio_pullupdown(unsigned int pin, gpio_pull_t pull);
+
+#endif /* __NEC_VR41XX_GIU_H */
diff --git a/include/asm-mips/vr41xx/vr41xx.h b/include/asm-mips/vr41xx/vr41xx.h
index ad0d1ea144f..7d41e44463f 100644
--- a/include/asm-mips/vr41xx/vr41xx.h
+++ b/include/asm-mips/vr41xx/vr41xx.h
@@ -126,7 +126,6 @@ extern void vr41xx_mask_clock(vr41xx_clock_t clock);
 #define GIU_IRQ_BASE		40
 #define GIU_IRQ(x)		(GIU_IRQ_BASE + (x))	/* IRQ 40-71 */
 #define GIU_IRQ_LAST		GIU_IRQ(31)
-#define GIU_IRQ_TO_PIN(x)	((x) - GIU_IRQ_BASE)	/* Pin 0-31 */
 
 extern int vr41xx_set_intassign(unsigned int irq, unsigned char intassign);
 extern int vr41xx_cascade_irq(unsigned int irq, int (*get_irq_number)(int irq));
@@ -197,38 +196,4 @@ extern void vr41xx_disable_csiint(uint16_t mask);
 extern void vr41xx_enable_bcuint(void);
 extern void vr41xx_disable_bcuint(void);
 
-/*
- * General-Purpose I/O Unit
- */
-enum {
-	TRIGGER_LEVEL,
-	TRIGGER_EDGE,
-	TRIGGER_EDGE_FALLING,
-	TRIGGER_EDGE_RISING
-};
-
-enum {
-	SIGNAL_THROUGH,
-	SIGNAL_HOLD
-};
-
-extern void vr41xx_set_irq_trigger(int pin, int trigger, int hold);
-
-enum {
-	LEVEL_LOW,
-	LEVEL_HIGH
-};
-
-extern void vr41xx_set_irq_level(int pin, int level);
-
-enum {
-	PIO_INPUT,
-	PIO_OUTPUT
-};
-
-enum {
-	DATA_LOW,
-	DATA_HIGH
-};
-
 #endif /* __NEC_VR41XX_H */
-- 
cgit v1.2.3-70-g09d2


From 22329b511a97557b293583194037d1f4c71e1504 Mon Sep 17 00:00:00 2001
From: Brent Casavant <bcasavan@sgi.com>
Date: Tue, 21 Jun 2005 17:15:59 -0700
Subject: [PATCH] ioc4: Core driver rewrite

This series of patches reworks the configuration and internal structure
of the SGI IOC4 I/O controller device drivers.

These changes are motivated by several factors:

- The IOC4 chip PCI resources are of mixed use between functions (i.e.
  multiple functions are handled in the same address range, sometimes
  within the same register), muddling resource ownership and initialization
  issues.  Centralizing this ownership in a core driver is desirable.

- The IOC4 chip implements multiple functions (serial, IDE, others not
  yet implemented in the mainline kernel) but is not a multifunction
  PCI device.  In order to properly handle device addition and removal
  as well as module insertion and deletion, an intermediary IOC4-specific
  driver layer is needed to handle these operations cleanly.

- All IOC4 drivers are currently enabled by a single CONFIG value.  As
  not all systems need all IOC4 functions, it is desireable to enable
  these drivers independently.

- The current IOC4 core driver will trigger loading of all function-level
  drivers, as it makes direct calls to them.  This situation should be
  reversed (i.e. function-level drivers cause loading of core driver)
  in order to maintain a clear and least-surprise driver loading model.

- IOC4 hardware design necessitates some driver-level dependency on
  the PCI bus clock speed.  Current code assumes a 66MHz bus, but the
  speed should be autodetected and appropriate compensation taken.

This patch series effects the above changes by a newly and better designed
IOC4 core driver with which the function-level drivers can register and
deregister themselves upon module insertion/removal.  By tracking these
modules, device addition/removal is also handled properly.  PCI resource
management and ownership issues are centralized in this core driver, and
IOC4-wide configuration actions such as bus speed detection are also
handled in this core driver.

This patch:

The SGI IOC4 I/O controller chip implements multiple functions, though it is
not a multi-function PCI device.  Additionally, various PCI resources of the
IOC4 are shared by multiple hardware functions, and thus resource ownership by
driver is not clearly delineated.  Due to the current driver design, all core
and subordinate drivers must be loaded, or none, which is undesirable if not
all IOC4 hardware features are being used.

This patch reorganizes the IOC4 drivers so that the core driver provides a
subdriver registration service.  Through appropriate callbacks the subdrivers
can now handle device addition and removal, as well as module insertion and
deletion (though the IOC4 IDE driver requires further work before module
deletion will work).  The core driver now takes care of allocating PCI
resources and data which must be shared between subdrivers, to clearly
delineate module ownership of these items.

Signed-off-by: Brent Casavant <bcasavan@sgi.com>
Acked-by: Pat Gefre <pfg@sgi.com
Acked-by: Jeremy Higdon <jeremy@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/sgi-ioc4.txt   |  45 +++++++
 drivers/ide/pci/sgiioc4.c    |  30 ++++-
 drivers/serial/ioc4_serial.c | 311 ++++++++++++++++++-------------------------
 drivers/sn/ioc4.c            | 290 ++++++++++++++++++++++++++++++++++++----
 include/linux/ioc4.h         | 156 ++++++++++++++++++++++
 include/linux/ioc4_common.h  |  21 ---
 6 files changed, 617 insertions(+), 236 deletions(-)
 create mode 100644 Documentation/sgi-ioc4.txt
 create mode 100644 include/linux/ioc4.h
 delete mode 100644 include/linux/ioc4_common.h

(limited to 'include')

diff --git a/Documentation/sgi-ioc4.txt b/Documentation/sgi-ioc4.txt
new file mode 100644
index 00000000000..876c96ae38d
--- /dev/null
+++ b/Documentation/sgi-ioc4.txt
@@ -0,0 +1,45 @@
+The SGI IOC4 PCI device is a bit of a strange beast, so some notes on
+it are in order.
+
+First, even though the IOC4 performs multiple functions, such as an
+IDE controller, a serial controller, a PS/2 keyboard/mouse controller,
+and an external interrupt mechanism, it's not implemented as a
+multifunction device.  The consequence of this from a software
+standpoint is that all these functions share a single IRQ, and
+they can't all register to own the same PCI device ID.  To make
+matters a bit worse, some of the register blocks (and even registers
+themselves) present in IOC4 are mixed-purpose between these several
+functions, meaning that there's no clear "owning" device driver.
+
+The solution is to organize the IOC4 driver into several independent
+drivers, "ioc4", "sgiioc4", and "ioc4_serial".  Note that there is no
+PS/2 controller driver as this functionality has never been wired up
+on a shipping IO card.
+
+ioc4
+====
+This is the core (or shim) driver for IOC4.  It is responsible for
+initializing the basic functionality of the chip, and allocating
+the PCI resources that are shared between the IOC4 functions.
+
+This driver also provides registration functions that the other
+IOC4 drivers can call to make their presence known.  Each driver
+needs to provide a probe and remove function, which are invoked
+by the core driver at appropriate times.  The interface of these
+IOC4 function probe and remove operations isn't precisely the same
+as PCI device probe and remove operations, but is logically the
+same operation.
+
+sgiioc4
+=======
+This is the IDE driver for IOC4.  Its name isn't very descriptive
+simply for historical reasons (it used to be the only IOC4 driver
+component).  There's not much to say about it other than it hooks
+up to the ioc4 driver via the appropriate registration, probe, and
+remove functions.
+
+ioc4_serial
+===========
+This is the serial driver for IOC4.  There's not much to say about it
+other than it hooks up to the ioc4 driver via the appropriate registration,
+probe, and remove functions.
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index 4651a22bf12..af526b671c4 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -34,7 +34,7 @@
 #include <linux/mm.h>
 #include <linux/ioport.h>
 #include <linux/blkdev.h>
-#include <linux/ioc4_common.h>
+#include <linux/ioc4.h>
 #include <asm/io.h>
 
 #include <linux/ide.h>
@@ -715,14 +715,34 @@ static ide_pci_device_t sgiioc4_chipsets[] __devinitdata = {
 };
 
 int
-ioc4_ide_attach_one(struct pci_dev *dev, const struct pci_device_id *id)
+ioc4_ide_attach_one(struct ioc4_driver_data *idd)
 {
-	return pci_init_sgiioc4(dev, &sgiioc4_chipsets[id->driver_data]);
+	return pci_init_sgiioc4(idd->idd_pdev,
+				&sgiioc4_chipsets[idd->idd_pci_id->driver_data]);
 }
 
+static struct ioc4_submodule ioc4_ide_submodule = {
+	.is_name = "IOC4_ide",
+	.is_owner = THIS_MODULE,
+	.is_probe = ioc4_ide_attach_one,
+/*	.is_remove = ioc4_ide_remove_one,	*/
+};
+
+static int __devinit
+ioc4_ide_init(void)
+{
+	return ioc4_register_submodule(&ioc4_ide_submodule);
+}
+
+static void __devexit
+ioc4_ide_exit(void)
+{
+	ioc4_unregister_submodule(&ioc4_ide_submodule);
+}
+
+module_init(ioc4_ide_init);
+module_exit(ioc4_ide_exit);
 
 MODULE_AUTHOR("Aniket Malatpure - Silicon Graphics Inc. (SGI)");
 MODULE_DESCRIPTION("IDE PCI driver module for SGI IOC4 Base-IO Card");
 MODULE_LICENSE("GPL");
-
-EXPORT_SYMBOL(ioc4_ide_attach_one);
diff --git a/drivers/serial/ioc4_serial.c b/drivers/serial/ioc4_serial.c
index ba4e13a22a5..da5f10eb484 100644
--- a/drivers/serial/ioc4_serial.c
+++ b/drivers/serial/ioc4_serial.c
@@ -20,7 +20,7 @@
 #include <linux/serial_reg.h>
 #include <linux/module.h>
 #include <linux/pci.h>
-#include <linux/ioc4_common.h>
+#include <linux/ioc4.h>
 #include <linux/serial_core.h>
 
 /*
@@ -130,12 +130,19 @@
 				 IOC4_SIO_IR_S3_TX_EXPLICIT)
 
 /* Bitmasks for IOC4_OTHER_IR, IOC4_OTHER_IEC, and IOC4_OTHER_IES  */
-#define IOC4_OTHER_IR_ATA_INT           0x00000001  /* ATAPI intr pass-thru */
-#define IOC4_OTHER_IR_ATA_MEMERR        0x00000002  /* ATAPI DMA PCI error */
-#define IOC4_OTHER_IR_S0_MEMERR         0x00000004  /* Port 0 PCI error */
-#define IOC4_OTHER_IR_S1_MEMERR         0x00000008  /* Port 1 PCI error */
-#define IOC4_OTHER_IR_S2_MEMERR         0x00000010  /* Port 2 PCI error */
-#define IOC4_OTHER_IR_S3_MEMERR         0x00000020  /* Port 3 PCI error */
+#define IOC4_OTHER_IR_ATA_INT		0x00000001  /* ATAPI intr pass-thru */
+#define IOC4_OTHER_IR_ATA_MEMERR	0x00000002  /* ATAPI DMA PCI error */
+#define IOC4_OTHER_IR_S0_MEMERR		0x00000004  /* Port 0 PCI error */
+#define IOC4_OTHER_IR_S1_MEMERR		0x00000008  /* Port 1 PCI error */
+#define IOC4_OTHER_IR_S2_MEMERR		0x00000010  /* Port 2 PCI error */
+#define IOC4_OTHER_IR_S3_MEMERR		0x00000020  /* Port 3 PCI error */
+#define IOC4_OTHER_IR_KBD_INT		0x00000040  /* Keyboard/mouse */
+#define IOC4_OTHER_IR_RESERVED		0x007fff80  /* Reserved */
+#define IOC4_OTHER_IR_RT_INT		0x00800000  /* INT_OUT section output */
+#define IOC4_OTHER_IR_GEN_INT		0xff000000  /* Generic pins */
+
+#define IOC4_OTHER_IR_SER_MEMERR (IOC4_OTHER_IR_S0_MEMERR | IOC4_OTHER_IR_S1_MEMERR | \
+				  IOC4_OTHER_IR_S2_MEMERR | IOC4_OTHER_IR_S3_MEMERR)
 
 /* Bitmasks for IOC4_SIO_CR */
 #define IOC4_SIO_CR_CMD_PULSE_SHIFT              0  /* byte bus strobe shift */
@@ -274,67 +281,22 @@ struct ioc4_uartregs {
 #define i4u_dlm u2.dlm
 #define i4u_fcr u3.fcr
 
-/* PCI memory space register map addressed using pci_bar0 */
-struct ioc4_memregs {
-	struct ioc4_mem {
-		/* Miscellaneous IOC4  registers */
-		uint32_t pci_err_addr_l;
-		uint32_t pci_err_addr_h;
-		uint32_t sio_ir;
-		uint32_t other_ir;
-
-		/* These registers are read-only for general kernel code.  */
-		uint32_t sio_ies_ro;
-		uint32_t other_ies_ro;
-		uint32_t sio_iec_ro;
-		uint32_t other_iec_ro;
-		uint32_t sio_cr;
-		uint32_t misc_fill1;
-		uint32_t int_out;
-		uint32_t misc_fill2;
-		uint32_t gpcr_s;
-		uint32_t gpcr_c;
-		uint32_t gpdr;
-		uint32_t misc_fill3;
-		uint32_t gppr_0;
-		uint32_t gppr_1;
-		uint32_t gppr_2;
-		uint32_t gppr_3;
-		uint32_t gppr_4;
-		uint32_t gppr_5;
-		uint32_t gppr_6;
-		uint32_t gppr_7;
-	} ioc4_mem;
-
-	char misc_fill4[0x100 - 0x5C - 4];
-
-	/* ATA/ATAP registers */
-	uint32_t ata_notused[9];
-	char ata_fill1[0x140 - 0x120 - 4];
-	uint32_t ata_notused1[8];
-	char ata_fill2[0x200 - 0x15C - 4];
-
-	/* Keyboard and mouse registers */
-	uint32_t km_notused[5];;
-	char km_fill1[0x300 - 0x210 - 4];
-
-	/* Serial port registers used for DMA serial I/O */
-	struct ioc4_serial {
-		uint32_t sbbr01_l;
-		uint32_t sbbr01_h;
-		uint32_t sbbr23_l;
-		uint32_t sbbr23_h;
-
-		struct ioc4_serialregs port_0;
-		struct ioc4_serialregs port_1;
-		struct ioc4_serialregs port_2;
-		struct ioc4_serialregs port_3;
-		struct ioc4_uartregs uart_0;
-		struct ioc4_uartregs uart_1;
-		struct ioc4_uartregs uart_2;
-		struct ioc4_uartregs uart_3;
-	} ioc4_serial;
-};
+/* Serial port registers used for DMA serial I/O */
+struct ioc4_serial {
+	uint32_t sbbr01_l;
+	uint32_t sbbr01_h;
+	uint32_t sbbr23_l;
+	uint32_t sbbr23_h;
+
+	struct ioc4_serialregs port_0;
+	struct ioc4_serialregs port_1;
+	struct ioc4_serialregs port_2;
+	struct ioc4_serialregs port_3;
+	struct ioc4_uartregs uart_0;
+	struct ioc4_uartregs uart_1;
+	struct ioc4_uartregs uart_2;
+	struct ioc4_uartregs uart_3;
+} ioc4_serial;
 
 /* UART clock speed */
 #define IOC4_SER_XIN_CLK        IOC4_SER_XIN_CLK_66
@@ -412,8 +374,8 @@ enum sio_proto {
 					| UART_LCR_WLEN7 | UART_LCR_WLEN8)
 #define LCR_MASK_STOP_BITS	(UART_LCR_STOP)
 
-#define PENDING(_p)	(readl(&(_p)->ip_mem->sio_ir) & _p->ip_ienb)
-#define READ_SIO_IR(_p) readl(&(_p)->ip_mem->sio_ir)
+#define PENDING(_p)	(readl(&(_p)->ip_mem->sio_ir.raw) & _p->ip_ienb)
+#define READ_SIO_IR(_p) readl(&(_p)->ip_mem->sio_ir.raw)
 
 /* Default to 4k buffers */
 #ifdef IOC4_1K_BUFFERS
@@ -447,7 +409,7 @@ struct ioc4_control {
  */
 #define MAX_IOC4_INTR_ENTS	(8 * sizeof(uint32_t))
 struct ioc4_soft {
-	struct ioc4_mem __iomem *is_ioc4_mem_addr;
+	struct ioc4_misc_regs __iomem *is_ioc4_misc_addr;
 	struct ioc4_serial __iomem *is_ioc4_serial_addr;
 
 	/* Each interrupt type has an entry in the array */
@@ -486,7 +448,7 @@ struct ioc4_port {
 	struct ioc4_soft *ip_ioc4_soft;
 
 	/* pci mem addresses */
-	struct ioc4_mem __iomem *ip_mem;
+	struct ioc4_misc_regs __iomem *ip_mem;
 	struct ioc4_serial __iomem *ip_serial;
 	struct ioc4_serialregs __iomem *ip_serial_regs;
 	struct ioc4_uartregs __iomem *ip_uart_regs;
@@ -553,7 +515,7 @@ struct hooks {
 	uint32_t intr_dma_error;
 	uint32_t intr_clear;
 	uint32_t intr_all;
-	char rs422_select_pin;
+	int rs422_select_pin;
 };
 
 static struct hooks hooks_array[IOC4_NUM_SERIAL_PORTS] = {
@@ -669,7 +631,7 @@ static void handle_intr(void *arg, uint32_t sio_ir);
 static inline void
 write_ireg(struct ioc4_soft *ioc4_soft, uint32_t val, int which, int type)
 {
-	struct ioc4_mem __iomem *mem = ioc4_soft->is_ioc4_mem_addr;
+	struct ioc4_misc_regs __iomem *mem = ioc4_soft->is_ioc4_misc_addr;
 	unsigned long flags;
 
 	spin_lock_irqsave(&ioc4_soft->is_ir_lock, flags);
@@ -678,11 +640,11 @@ write_ireg(struct ioc4_soft *ioc4_soft, uint32_t val, int which, int type)
 	case IOC4_SIO_INTR_TYPE:
 		switch (which) {
 		case IOC4_W_IES:
-			writel(val, &mem->sio_ies_ro);
+			writel(val, &mem->sio_ies.raw);
 			break;
 
 		case IOC4_W_IEC:
-			writel(val, &mem->sio_iec_ro);
+			writel(val, &mem->sio_iec.raw);
 			break;
 		}
 		break;
@@ -690,11 +652,11 @@ write_ireg(struct ioc4_soft *ioc4_soft, uint32_t val, int which, int type)
 	case IOC4_OTHER_INTR_TYPE:
 		switch (which) {
 		case IOC4_W_IES:
-			writel(val, &mem->other_ies_ro);
+			writel(val, &mem->other_ies.raw);
 			break;
 
 		case IOC4_W_IEC:
-			writel(val, &mem->other_iec_ro);
+			writel(val, &mem->other_iec.raw);
 			break;
 		}
 		break;
@@ -747,7 +709,8 @@ static int set_baud(struct ioc4_port *port, int baud)
  */
 static struct ioc4_port *get_ioc4_port(struct uart_port *the_port)
 {
-	struct ioc4_control *control = dev_get_drvdata(the_port->dev);
+	struct ioc4_driver_data *idd = dev_get_drvdata(the_port->dev);
+	struct ioc4_control *control = idd->idd_serial_data;
 	int ii;
 
 	if (control) {
@@ -782,7 +745,7 @@ static struct ioc4_port *get_ioc4_port(struct uart_port *the_port)
 static inline uint32_t
 pending_intrs(struct ioc4_soft *soft, int type)
 {
-	struct ioc4_mem __iomem *mem = soft->is_ioc4_mem_addr;
+	struct ioc4_misc_regs __iomem *mem = soft->is_ioc4_misc_addr;
 	unsigned long flag;
 	uint32_t intrs = 0;
 
@@ -793,11 +756,11 @@ pending_intrs(struct ioc4_soft *soft, int type)
 
 	switch (type) {
 	case IOC4_SIO_INTR_TYPE:
-		intrs = readl(&mem->sio_ir) & readl(&mem->sio_ies_ro);
+		intrs = readl(&mem->sio_ir.raw) & readl(&mem->sio_ies.raw);
 		break;
 
 	case IOC4_OTHER_INTR_TYPE:
-		intrs = readl(&mem->other_ir) & readl(&mem->other_ies_ro);
+		intrs = readl(&mem->other_ir.raw) & readl(&mem->other_ies.raw);
 
 		/* Don't process any ATA interrupte */
 		intrs &= ~(IOC4_OTHER_IR_ATA_INT | IOC4_OTHER_IR_ATA_MEMERR);
@@ -826,7 +789,7 @@ static int inline port_init(struct ioc4_port *port)
 
 	/* Wait until any pending bus activity for this port has ceased */
 	do
-		sio_cr = readl(&port->ip_mem->sio_cr);
+		sio_cr = readl(&port->ip_mem->sio_cr.raw);
 	while (!(sio_cr & IOC4_SIO_CR_SIO_DIAG_IDLE));
 
 	/* Finish reset sequence */
@@ -899,7 +862,7 @@ static int inline port_init(struct ioc4_port *port)
 	write_ireg(port->ip_ioc4_soft, hooks->intr_clear,
 		       IOC4_W_IEC, IOC4_SIO_INTR_TYPE);
 	port->ip_ienb &= ~hooks->intr_clear;
-	writel(hooks->intr_clear, &port->ip_mem->sio_ir);
+	writel(hooks->intr_clear, &port->ip_mem->sio_ir.raw);
 	return 0;
 }
 
@@ -918,23 +881,23 @@ static void handle_dma_error_intr(void *arg, uint32_t other_ir)
 	spin_lock_irqsave(&port->ip_lock, flags);
 
 	/* ACK the interrupt */
-	writel(hooks->intr_dma_error, &port->ip_mem->other_ir);
+	writel(hooks->intr_dma_error, &port->ip_mem->other_ir.raw);
 
-	if (readl(&port->ip_mem->pci_err_addr_l) & IOC4_PCI_ERR_ADDR_VLD) {
+	if (readl(&port->ip_mem->pci_err_addr_l.raw) & IOC4_PCI_ERR_ADDR_VLD) {
 		printk(KERN_ERR
 			"PCI error address is 0x%lx, "
 				"master is serial port %c %s\n",
 		     (((uint64_t)readl(&port->ip_mem->pci_err_addr_h)
 							 << 32)
-				| readl(&port->ip_mem->pci_err_addr_l))
+				| readl(&port->ip_mem->pci_err_addr_l.raw))
 					& IOC4_PCI_ERR_ADDR_ADDR_MSK, '1' +
-		     ((char)(readl(&port->ip_mem-> pci_err_addr_l) &
+		     ((char)(readl(&port->ip_mem->pci_err_addr_l.raw) &
 			     IOC4_PCI_ERR_ADDR_MST_NUM_MSK) >> 1),
-		     (readl(&port->ip_mem->pci_err_addr_l)
+		     (readl(&port->ip_mem->pci_err_addr_l.raw)
 				& IOC4_PCI_ERR_ADDR_MST_TYP_MSK)
 				? "RX" : "TX");
 
-		if (readl(&port->ip_mem->pci_err_addr_l)
+		if (readl(&port->ip_mem->pci_err_addr_l.raw)
 						& IOC4_PCI_ERR_ADDR_MUL_ERR) {
 			printk(KERN_ERR
 				"Multiple errors occurred\n");
@@ -1018,26 +981,26 @@ static irqreturn_t ioc4_intr(int irq, void *arg, struct pt_regs *regs)
 				"other_ies = 0x%x\n",
 			       (intr_type == IOC4_SIO_INTR_TYPE) ? "sio" :
 			       "other", this_ir,
-			       readl(&soft->is_ioc4_mem_addr->sio_ir),
-			       readl(&soft->is_ioc4_mem_addr->sio_ies_ro),
-			       readl(&soft->is_ioc4_mem_addr->other_ir),
-			       readl(&soft->is_ioc4_mem_addr->other_ies_ro));
+			       readl(&soft->is_ioc4_misc_addr->sio_ir.raw),
+			       readl(&soft->is_ioc4_misc_addr->sio_ies.raw),
+			       readl(&soft->is_ioc4_misc_addr->other_ir.raw),
+			       readl(&soft->is_ioc4_misc_addr->other_ies.raw));
 		}
 	}
 #ifdef DEBUG_INTERRUPTS
 	{
-		struct ioc4_mem __iomem *mem = soft->is_ioc4_mem_addr;
+		struct ioc4_misc_regs __iomem *mem = soft->is_ioc4_misc_addr;
 		spinlock_t *lp = &soft->is_ir_lock;
 		unsigned long flag;
 
 		spin_lock_irqsave(&soft->is_ir_lock, flag);
-		printk ("%s : %d : mem 0x%p sio_ir 0x%x sio_ies_ro 0x%x "
-				"other_ir 0x%x other_ies_ro 0x%x mask 0x%x\n",
+		printk ("%s : %d : mem 0x%p sio_ir 0x%x sio_ies 0x%x "
+				"other_ir 0x%x other_ies 0x%x mask 0x%x\n",
 		     __FUNCTION__, __LINE__,
-		     (void *)mem, readl(&mem->sio_ir),
-		     readl(&mem->sio_ies_ro),
-		     readl(&mem->other_ir),
-		     readl(&mem->other_ies_ro),
+		     (void *)mem, readl(&mem->sio_ir.raw),
+		     readl(&mem->sio_ies.raw),
+		     readl(&mem->other_ir.raw),
+		     readl(&mem->other_ies.raw),
 		     IOC4_OTHER_IR_ATA_INT | IOC4_OTHER_IR_ATA_MEMERR);
 		spin_unlock_irqrestore(&soft->is_ir_lock, flag);
 	}
@@ -1056,7 +1019,7 @@ static irqreturn_t ioc4_intr(int irq, void *arg, struct pt_regs *regs)
  */
 static int inline ioc4_attach_local(struct pci_dev *pdev,
 			struct ioc4_control *control,
-			struct ioc4_soft *soft, void __iomem *ioc4_mem,
+			struct ioc4_soft *soft, void __iomem *ioc4_misc,
 			void __iomem *ioc4_serial)
 {
 	struct ioc4_port *port;
@@ -1076,7 +1039,7 @@ static int inline ioc4_attach_local(struct pci_dev *pdev,
 				ioc4_revid, ioc4_revid_min);
 		return -EPERM;
 	}
-	BUG_ON(ioc4_mem == NULL);
+	BUG_ON(ioc4_misc == NULL);
 	BUG_ON(ioc4_serial == NULL);
 
 	/* Create port structures for each port */
@@ -1103,7 +1066,7 @@ static int inline ioc4_attach_local(struct pci_dev *pdev,
 		port->ip_pci_bus_speed = IOC4_SER_XIN_CLK;
 		port->ip_baud = 9600;
 		port->ip_control = control;
-		port->ip_mem = ioc4_mem;
+		port->ip_mem = ioc4_misc;
 		port->ip_serial = ioc4_serial;
 
 		/* point to the right hook */
@@ -1604,14 +1567,12 @@ static int ioc4_set_proto(struct ioc4_port *port, enum sio_proto proto)
 	switch (proto) {
 	case PROTO_RS232:
 		/* Clear the appropriate GIO pin */
-		writel(0, (&port->ip_mem->gppr_0 +
-				  hooks->rs422_select_pin));
+		writel(0, (&port->ip_mem->gppr[hooks->rs422_select_pin].raw));
 		break;
 
 	case PROTO_RS422:
 		/* Set the appropriate GIO pin */
-		writel(1, (&port->ip_mem->gppr_0 +
-				  hooks->rs422_select_pin));
+		writel(1, (&port->ip_mem->gppr[hooks->rs422_select_pin].raw));
 		break;
 
 	default:
@@ -1885,7 +1846,7 @@ static void handle_intr(void *arg, uint32_t sio_ir)
 		if (sio_ir & hooks->intr_delta_dcd) {
 			/* ACK the interrupt */
 			writel(hooks->intr_delta_dcd,
-				&port->ip_mem->sio_ir);
+				&port->ip_mem->sio_ir.raw);
 
 			shadow = readl(&port->ip_serial_regs->shadow);
 
@@ -1907,7 +1868,7 @@ static void handle_intr(void *arg, uint32_t sio_ir)
 		if (sio_ir & hooks->intr_delta_cts) {
 			/* ACK the interrupt */
 			writel(hooks->intr_delta_cts,
-					&port->ip_mem->sio_ir);
+					&port->ip_mem->sio_ir.raw);
 
 			shadow = readl(&port->ip_serial_regs->shadow);
 
@@ -1928,7 +1889,7 @@ static void handle_intr(void *arg, uint32_t sio_ir)
 		if (sio_ir & hooks->intr_rx_timer) {
 			/* ACK the interrupt */
 			writel(hooks->intr_rx_timer,
-				&port->ip_mem->sio_ir);
+				&port->ip_mem->sio_ir.raw);
 
 			if ((port->ip_notify & N_DATA_READY)
 					&& (port->ip_port)) {
@@ -1974,7 +1935,7 @@ static void handle_intr(void *arg, uint32_t sio_ir)
 
 			/* ACK the interrupt */
 			writel(hooks->intr_tx_explicit,
-					&port->ip_mem->sio_ir);
+					&port->ip_mem->sio_ir.raw);
 
 			if (port->ip_notify & N_OUTPUT_LOWAT)
 				ioc4_cb_output_lowat(port);
@@ -2634,7 +2595,8 @@ ioc4_serial_core_attach(struct pci_dev *pdev)
 {
 	struct ioc4_port *port;
 	struct uart_port *the_port;
-	struct ioc4_control *control = pci_get_drvdata(pdev);
+	struct ioc4_driver_data *idd = pci_get_drvdata(pdev);
+	struct ioc4_control *control = idd->idd_serial_data;
 	int ii;
 
 	DPRINT_CONFIG(("%s: attach pdev 0x%p - control 0x%p\n",
@@ -2680,55 +2642,29 @@ ioc4_serial_core_attach(struct pci_dev *pdev)
 
 /**
  * ioc4_serial_attach_one - register attach function
- *		called per card found from ioc4_serial_detect as part
- *		of module_init().
- * @pdev: handle for this card
- * @pci_id: pci id for this card
+ *		called per card found from IOC4 master module.
+ * @idd: Master module data for this IOC4
  */
 int
-ioc4_serial_attach_one(struct pci_dev *pdev, const struct pci_device_id *pci_id)
+ioc4_serial_attach_one(struct ioc4_driver_data *idd)
 {
-	struct ioc4_mem __iomem *mem;
-	unsigned long tmp_addr, tmp_addr1;
+	unsigned long tmp_addr1;
 	struct ioc4_serial __iomem *serial;
 	struct ioc4_soft *soft;
 	struct ioc4_control *control;
-	int tmp, ret = 0;
+	int ret = 0;
 
 
-	DPRINT_CONFIG(("%s (0x%p, 0x%p)\n", __FUNCTION__, pdev, pci_id));
-
-	/* Map in the ioc4 memory */
-	tmp_addr = pci_resource_start(pdev, 0);
-	if (!tmp_addr) {
-		printk(KERN_WARNING
-			 "ioc4 (%p) : unable to get PIO mapping for "
-				"MEM space\n", (void *)pdev);
-		return -ENODEV;
-	}
-	if (!request_region(tmp_addr, sizeof(struct ioc4_mem), "sioc4_mem")) {
-		printk(KERN_ALERT
-			"ioc4 (%p): unable to get request region for "
-			"MEM space\n", (void *)pdev);
-		return -ENODEV;
-	}
-	mem = ioremap(tmp_addr, sizeof(struct ioc4_mem));
-	if (!mem) {
-		printk(KERN_WARNING
-			 "ioc4 (%p) : unable to remap ioc4 memory\n",
-				(void *)pdev);
-		ret = -ENODEV;
-		goto out1;
-	}
+	DPRINT_CONFIG(("%s (0x%p, 0x%p)\n", __FUNCTION__, idd->idd_pdev, idd->idd_pci_id));
 
 	/* request serial registers */
-	tmp_addr1 = pci_resource_start(pdev, 0) + IOC4_SERIAL_OFFSET;
+	tmp_addr1 = idd->idd_bar0 + IOC4_SERIAL_OFFSET;
 
 	if (!request_region(tmp_addr1, sizeof(struct ioc4_serial),
 					"sioc4_uart")) {
 		printk(KERN_WARNING
 			"ioc4 (%p): unable to get request region for "
-				"uart space\n", (void *)pdev);
+				"uart space\n", (void *)idd->idd_pdev);
 		ret = -ENODEV;
 		goto out1;
 	}
@@ -2736,12 +2672,12 @@ ioc4_serial_attach_one(struct pci_dev *pdev, const struct pci_device_id *pci_id)
 	if (!serial) {
 		printk(KERN_WARNING
 			 "ioc4 (%p) : unable to remap ioc4 serial register\n",
-				(void *)pdev);
+				(void *)idd->idd_pdev);
 		ret = -ENODEV;
 		goto out2;
 	}
 	DPRINT_CONFIG(("%s : mem 0x%p, serial 0x%p\n",
-				__FUNCTION__, (void *)mem, (void *)serial));
+				__FUNCTION__, (void *)idd->idd_misc_regs, (void *)serial));
 
 	/* Get memory for the new card */
 	control = kmalloc(sizeof(struct ioc4_control) * IOC4_NUM_SERIAL_PORTS,
@@ -2754,59 +2690,57 @@ ioc4_serial_attach_one(struct pci_dev *pdev, const struct pci_device_id *pci_id)
 		goto out2;
 	}
 	memset(control, 0, sizeof(struct ioc4_control));
-	pci_set_drvdata(pdev, control);
+	idd->idd_serial_data = control;
 
 	/* Allocate the soft structure */
 	soft = kmalloc(sizeof(struct ioc4_soft), GFP_KERNEL);
 	if (!soft) {
 		printk(KERN_WARNING
 		       "ioc4 (%p): unable to get memory for the soft struct\n",
-		       (void *)pdev);
+		       (void *)idd->idd_pdev);
 		ret = -ENOMEM;
 		goto out3;
 	}
 	memset(soft, 0, sizeof(struct ioc4_soft));
 
 	spin_lock_init(&soft->is_ir_lock);
-	soft->is_ioc4_mem_addr = mem;
+	soft->is_ioc4_misc_addr = idd->idd_misc_regs;
 	soft->is_ioc4_serial_addr = serial;
 
 	/* Init the IOC4 */
-	pci_read_config_dword(pdev, PCI_COMMAND, &tmp);
-	pci_write_config_dword(pdev, PCI_COMMAND,
-			       tmp | PCI_COMMAND_PARITY | PCI_COMMAND_SERR);
-
-	writel(0xf << IOC4_SIO_CR_CMD_PULSE_SHIFT, &mem->sio_cr);
+	writel(0xf << IOC4_SIO_CR_CMD_PULSE_SHIFT,
+	       &idd->idd_misc_regs->sio_cr.raw);
 
 	/* Enable serial port mode select generic PIO pins as outputs */
 	writel(IOC4_GPCR_UART0_MODESEL | IOC4_GPCR_UART1_MODESEL
 		| IOC4_GPCR_UART2_MODESEL | IOC4_GPCR_UART3_MODESEL,
-		&mem->gpcr_s);
+		&idd->idd_misc_regs->gpcr_s.raw);
 
-	/* Clear and disable all interrupts */
+	/* Clear and disable all serial interrupts */
 	write_ireg(soft, ~0, IOC4_W_IEC, IOC4_SIO_INTR_TYPE);
-	writel(~0, &mem->sio_ir);
-	write_ireg(soft, ~(IOC4_OTHER_IR_ATA_INT | IOC4_OTHER_IR_ATA_MEMERR),
-			IOC4_W_IEC, IOC4_OTHER_INTR_TYPE);
-	writel(~(IOC4_OTHER_IR_ATA_MEMERR | IOC4_OTHER_IR_ATA_MEMERR),
-					&mem->other_ir);
+	writel(~0, &idd->idd_misc_regs->sio_ir.raw);
+	write_ireg(soft, IOC4_OTHER_IR_SER_MEMERR, IOC4_W_IEC,
+		   IOC4_OTHER_INTR_TYPE);
+	writel(IOC4_OTHER_IR_SER_MEMERR, &idd->idd_misc_regs->other_ir.raw);
 	control->ic_soft = soft;
-	if (!request_irq(pdev->irq, ioc4_intr, SA_SHIRQ,
+
+	/* Hook up interrupt handler */
+	if (!request_irq(idd->idd_pdev->irq, ioc4_intr, SA_SHIRQ,
 				"sgi-ioc4serial", (void *)soft)) {
-		control->ic_irq = pdev->irq;
+		control->ic_irq = idd->idd_pdev->irq;
 	} else {
 		printk(KERN_WARNING
 		    "%s : request_irq fails for IRQ 0x%x\n ",
-			__FUNCTION__, pdev->irq);
+			__FUNCTION__, idd->idd_pdev->irq);
 	}
-	if ((ret = ioc4_attach_local(pdev, control, soft,
-				soft->is_ioc4_mem_addr,
+	if ((ret = ioc4_attach_local(idd->idd_pdev, control, soft,
+				soft->is_ioc4_misc_addr,
 				soft->is_ioc4_serial_addr)))
 		goto out4;
 
 	/* register port with the serial core */
 
-	if ((ret = ioc4_serial_core_attach(pdev)))
+	if ((ret = ioc4_serial_core_attach(idd->idd_pdev)))
 		goto out4;
 
 	return ret;
@@ -2819,7 +2753,6 @@ out3:
 out2:
 	release_region(tmp_addr1, sizeof(struct ioc4_serial));
 out1:
-	release_region(tmp_addr, sizeof(struct ioc4_mem));
 
 	return ret;
 }
@@ -2828,11 +2761,10 @@ out1:
 /**
  * ioc4_serial_remove_one - detach function
  *
- * @pdev: handle for this card
+ * @idd: IOC4 master module data for this IOC4
  */
 
-#if 0
-void ioc4_serial_remove_one(struct pci_dev *pdev)
+int ioc4_serial_remove_one(struct ioc4_driver_data *idd)
 {
 	int ii;
 	struct ioc4_control *control;
@@ -2840,7 +2772,7 @@ void ioc4_serial_remove_one(struct pci_dev *pdev)
 	struct ioc4_port *port;
 	struct ioc4_soft *soft;
 
-	control = pci_get_drvdata(pdev);
+	control = idd->idd_serial_data;
 
 	for (ii = 0; ii < IOC4_NUM_SERIAL_PORTS; ii++) {
 		the_port = &control->ic_port[ii].icp_uart_port;
@@ -2867,10 +2799,17 @@ void ioc4_serial_remove_one(struct pci_dev *pdev)
 		kfree(soft);
 	}
 	kfree(control);
-	pci_set_drvdata(pdev, NULL);
-	uart_unregister_driver(&ioc4_uart);
+	idd->idd_serial_data = NULL;
+
+	return 0;
 }
-#endif
+
+static struct ioc4_submodule ioc4_serial_submodule = {
+	.is_name = "IOC4_serial",
+	.is_owner = THIS_MODULE,
+	.is_probe = ioc4_serial_attach_one,
+	.is_remove = ioc4_serial_remove_one,
+};
 
 /**
  * ioc4_serial_init - module init
@@ -2886,12 +2825,20 @@ int ioc4_serial_init(void)
 			__FUNCTION__);
 		return ret;
 	}
-	return 0;
+
+	/* register with IOC4 main module */
+	return ioc4_register_submodule(&ioc4_serial_submodule);
 }
 
+static void __devexit ioc4_serial_exit(void)
+{
+	ioc4_unregister_submodule(&ioc4_serial_submodule);
+	uart_unregister_driver(&ioc4_uart);
+}
+
+module_init(ioc4_serial_init);
+module_exit(ioc4_serial_exit);
+
 MODULE_AUTHOR("Pat Gefre - Silicon Graphics Inc. (SGI) <pfg@sgi.com>");
 MODULE_DESCRIPTION("Serial PCI driver module for SGI IOC4 Base-IO Card");
 MODULE_LICENSE("GPL");
-
-EXPORT_SYMBOL(ioc4_serial_init);
-EXPORT_SYMBOL(ioc4_serial_attach_one);
diff --git a/drivers/sn/ioc4.c b/drivers/sn/ioc4.c
index d9e4ee280e5..70862d72ea9 100644
--- a/drivers/sn/ioc4.c
+++ b/drivers/sn/ioc4.c
@@ -6,60 +6,294 @@
  * Copyright (C) 2005 Silicon Graphics, Inc.  All Rights Reserved.
  */
 
-/*
- * This file contains a shim driver for the IOC4 IDE and serial drivers.
+/* This file contains the master driver module for use by SGI IOC4 subdrivers.
+ *
+ * It allocates any resources shared between multiple subdevices, and
+ * provides accessor functions (where needed) and the like for those
+ * resources.  It also provides a mechanism for the subdevice modules
+ * to support loading and unloading.
+ *
+ * Non-shared resources (e.g. external interrupt A_INT_OUT register page
+ * alias, serial port and UART registers) are handled by the subdevice
+ * modules themselves.
+ *
+ * This is all necessary because IOC4 is not implemented as a multi-function
+ * PCI device, but an amalgamation of disparate registers for several
+ * types of device (ATA, serial, external interrupts).  The normal
+ * resource management in the kernel doesn't have quite the right interfaces
+ * to handle this situation (e.g. multiple modules can't claim the same
+ * PCI ID), thus this IOC4 master module.
  */
 
 #include <linux/errno.h>
 #include <linux/module.h>
 #include <linux/pci.h>
-#include <linux/ioc4_common.h>
-#include <linux/ide.h>
+#include <linux/ioc4.h>
+#include <linux/rwsem.h>
 
+/************************
+ * Submodule management *
+ ************************/
 
-static int __devinit
-ioc4_probe_one(struct pci_dev *pdev, const struct pci_device_id *pci_id)
+static LIST_HEAD(ioc4_devices);
+static DECLARE_RWSEM(ioc4_devices_rwsem);
+
+static LIST_HEAD(ioc4_submodules);
+static DECLARE_RWSEM(ioc4_submodules_rwsem);
+
+/* Register an IOC4 submodule */
+int
+ioc4_register_submodule(struct ioc4_submodule *is)
+{
+	struct ioc4_driver_data *idd;
+
+	down_write(&ioc4_submodules_rwsem);
+	list_add(&is->is_list, &ioc4_submodules);
+	up_write(&ioc4_submodules_rwsem);
+
+	/* Initialize submodule for each IOC4 */
+	if (!is->is_probe)
+		return 0;
+
+	down_read(&ioc4_devices_rwsem);
+	list_for_each_entry(idd, &ioc4_devices, idd_list) {
+		if (is->is_probe(idd)) {
+			printk(KERN_WARNING
+			       "%s: IOC4 submodule %s probe failed "
+			       "for pci_dev %s",
+			       __FUNCTION__, module_name(is->is_owner),
+			       pci_name(idd->idd_pdev));
+		}
+	}
+	up_read(&ioc4_devices_rwsem);
+
+	return 0;
+}
+
+/* Unregister an IOC4 submodule */
+void
+ioc4_unregister_submodule(struct ioc4_submodule *is)
 {
+	struct ioc4_driver_data *idd;
+
+	down_write(&ioc4_submodules_rwsem);
+	list_del(&is->is_list);
+	up_write(&ioc4_submodules_rwsem);
+
+	/* Remove submodule for each IOC4 */
+	if (!is->is_remove)
+		return;
+
+	down_read(&ioc4_devices_rwsem);
+	list_for_each_entry(idd, &ioc4_devices, idd_list) {
+		if (is->is_remove(idd)) {
+			printk(KERN_WARNING
+			       "%s: IOC4 submodule %s remove failed "
+			       "for pci_dev %s.\n",
+			       __FUNCTION__, module_name(is->is_owner),
+			       pci_name(idd->idd_pdev));
+		}
+	}
+	up_read(&ioc4_devices_rwsem);
+}
+
+/*********************
+ * Device management *
+ *********************/
+
+/* Adds a new instance of an IOC4 card */
+static int
+ioc4_probe(struct pci_dev *pdev, const struct pci_device_id *pci_id)
+{
+	struct ioc4_driver_data *idd;
+	struct ioc4_submodule *is;
+	uint32_t pcmd;
 	int ret;
 
+	/* Enable IOC4 and take ownership of it */
 	if ((ret = pci_enable_device(pdev))) {
 		printk(KERN_WARNING
-			 "%s: Failed to enable device with "
-				"pci_dev 0x%p... returning\n",
-				__FUNCTION__, (void *)pdev);
-		return ret;
+		       "%s: Failed to enable IOC4 device for pci_dev %s.\n",
+		       __FUNCTION__, pci_name(pdev));
+		goto out;
 	}
 	pci_set_master(pdev);
 
-	/* attach each sub-device */
-	ret = ioc4_ide_attach_one(pdev, pci_id);
-	if (ret)
-		return ret;
-	return ioc4_serial_attach_one(pdev, pci_id);
+	/* Set up per-IOC4 data */
+	idd = kmalloc(sizeof(struct ioc4_driver_data), GFP_KERNEL);
+	if (!idd) {
+		printk(KERN_WARNING
+		       "%s: Failed to allocate IOC4 data for pci_dev %s.\n",
+		       __FUNCTION__, pci_name(pdev));
+		ret = -ENODEV;
+		goto out_idd;
+	}
+	idd->idd_pdev = pdev;
+	idd->idd_pci_id = pci_id;
+
+	/* Map IOC4 misc registers.  These are shared between subdevices
+	 * so the main IOC4 module manages them.
+	 */
+	idd->idd_bar0 = pci_resource_start(idd->idd_pdev, 0);
+	if (!idd->idd_bar0) {
+		printk(KERN_WARNING
+		       "%s: Unable to find IOC4 misc resource "
+		       "for pci_dev %s.\n",
+		       __FUNCTION__, pci_name(idd->idd_pdev));
+		ret = -ENODEV;
+		goto out_pci;
+	}
+	if (!request_region(idd->idd_bar0, sizeof(struct ioc4_misc_regs),
+			    "ioc4_misc")) {
+		printk(KERN_WARNING
+		       "%s: Unable to request IOC4 misc region "
+		       "for pci_dev %s.\n",
+		       __FUNCTION__, pci_name(idd->idd_pdev));
+		ret = -ENODEV;
+		goto out_pci;
+	}
+	idd->idd_misc_regs = ioremap(idd->idd_bar0,
+				     sizeof(struct ioc4_misc_regs));
+	if (!idd->idd_misc_regs) {
+		printk(KERN_WARNING
+		       "%s: Unable to remap IOC4 misc region "
+		       "for pci_dev %s.\n",
+		       __FUNCTION__, pci_name(idd->idd_pdev));
+		ret = -ENODEV;
+		goto out_misc_region;
+	}
+
+	/* Failsafe portion of per-IOC4 initialization */
+
+	/* Initialize IOC4 */
+	pci_read_config_dword(idd->idd_pdev, PCI_COMMAND, &pcmd);
+	pci_write_config_dword(idd->idd_pdev, PCI_COMMAND,
+			       pcmd | PCI_COMMAND_PARITY | PCI_COMMAND_SERR);
+
+	/* Disable/clear all interrupts.  Need to do this here lest
+	 * one submodule request the shared IOC4 IRQ, but interrupt
+	 * is generated by a different subdevice.
+	 */
+	/* Disable */
+	writel(~0, &idd->idd_misc_regs->other_iec.raw);
+	writel(~0, &idd->idd_misc_regs->sio_iec);
+	/* Clear (i.e. acknowledge) */
+	writel(~0, &idd->idd_misc_regs->other_ir.raw);
+	writel(~0, &idd->idd_misc_regs->sio_ir);
+
+	/* Track PCI-device specific data */
+	idd->idd_serial_data = NULL;
+	pci_set_drvdata(idd->idd_pdev, idd);
+	down_write(&ioc4_devices_rwsem);
+	list_add(&idd->idd_list, &ioc4_devices);
+	up_write(&ioc4_devices_rwsem);
+
+	/* Add this IOC4 to all submodules */
+	down_read(&ioc4_submodules_rwsem);
+	list_for_each_entry(is, &ioc4_submodules, is_list) {
+		if (is->is_probe && is->is_probe(idd)) {
+			printk(KERN_WARNING
+			       "%s: IOC4 submodule 0x%s probe failed "
+			       "for pci_dev %s.\n",
+			       __FUNCTION__, module_name(is->is_owner),
+			       pci_name(idd->idd_pdev));
+		}
+	}
+	up_read(&ioc4_submodules_rwsem);
+
+	return 0;
+
+out_misc_region:
+	release_region(idd->idd_bar0, sizeof(struct ioc4_misc_regs));
+out_pci:
+	kfree(idd);
+out_idd:
+	pci_disable_device(pdev);
+out:
+	return ret;
 }
 
-/* pci device struct */
-static struct pci_device_id ioc4_s_id_table[] = {
+/* Removes a particular instance of an IOC4 card. */
+static void
+ioc4_remove(struct pci_dev *pdev)
+{
+	struct ioc4_submodule *is;
+	struct ioc4_driver_data *idd;
+
+	idd = pci_get_drvdata(pdev);
+
+	/* Remove this IOC4 from all submodules */
+	down_read(&ioc4_submodules_rwsem);
+	list_for_each_entry(is, &ioc4_submodules, is_list) {
+		if (is->is_remove && is->is_remove(idd)) {
+			printk(KERN_WARNING
+			       "%s: IOC4 submodule 0x%s remove failed "
+			       "for pci_dev %s.\n",
+			       __FUNCTION__, module_name(is->is_owner),
+			       pci_name(idd->idd_pdev));
+		}
+	}
+	up_read(&ioc4_submodules_rwsem);
+
+	/* Release resources */
+	iounmap(idd->idd_misc_regs);
+	if (!idd->idd_bar0) {
+		printk(KERN_WARNING
+		       "%s: Unable to get IOC4 misc mapping for pci_dev %s. "
+		       "Device removal may be incomplete.\n",
+		       __FUNCTION__, pci_name(idd->idd_pdev));
+	}
+	release_region(idd->idd_bar0, sizeof(struct ioc4_misc_regs));
+
+	/* Disable IOC4 and relinquish */
+	pci_disable_device(pdev);
+
+	/* Remove and free driver data */
+	down_write(&ioc4_devices_rwsem);
+	list_del(&idd->idd_list);
+	up_write(&ioc4_devices_rwsem);
+	kfree(idd);
+}
+
+static struct pci_device_id ioc4_id_table[] = {
 	{PCI_VENDOR_ID_SGI, PCI_DEVICE_ID_SGI_IOC4, PCI_ANY_ID,
 	 PCI_ANY_ID, 0x0b4000, 0xFFFFFF},
 	{0}
 };
-MODULE_DEVICE_TABLE(pci, ioc4_s_id_table);
 
-static struct pci_driver __devinitdata ioc4_s_driver = {
-	.name	= "IOC4",
-	.id_table = ioc4_s_id_table,
-	.probe	= ioc4_probe_one,
+static struct pci_driver __devinitdata ioc4_driver = {
+	.name = "IOC4",
+	.id_table = ioc4_id_table,
+	.probe = ioc4_probe,
+	.remove = ioc4_remove,
 };
 
-static int __devinit ioc4_detect(void)
+MODULE_DEVICE_TABLE(pci, ioc4_id_table);
+
+/*********************
+ * Module management *
+ *********************/
+
+/* Module load */
+static int __devinit
+ioc4_init(void)
 {
-	ioc4_serial_init();
+	return pci_register_driver(&ioc4_driver);
+}
 
-	return pci_register_driver(&ioc4_s_driver);
+/* Module unload */
+static void __devexit
+ioc4_exit(void)
+{
+	pci_unregister_driver(&ioc4_driver);
 }
-module_init(ioc4_detect);
 
-MODULE_AUTHOR("Pat Gefre - Silicon Graphics Inc. (SGI) <pfg@sgi.com>");
-MODULE_DESCRIPTION("PCI driver module for SGI IOC4 Base-IO Card");
+module_init(ioc4_init);
+module_exit(ioc4_exit);
+
+MODULE_AUTHOR("Brent Casavant - Silicon Graphics, Inc. <bcasavan@sgi.com>");
+MODULE_DESCRIPTION("PCI driver master module for SGI IOC4 Base-IO Card");
 MODULE_LICENSE("GPL");
+
+EXPORT_SYMBOL(ioc4_register_submodule);
+EXPORT_SYMBOL(ioc4_unregister_submodule);
diff --git a/include/linux/ioc4.h b/include/linux/ioc4.h
new file mode 100644
index 00000000000..729bfa4c4ac
--- /dev/null
+++ b/include/linux/ioc4.h
@@ -0,0 +1,156 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (c) 2005 Silicon Graphics, Inc.  All Rights Reserved.
+ */
+
+#ifndef _LINUX_IOC4_H
+#define _LINUX_IOC4_H
+
+#include <linux/interrupt.h>
+
+/***********************************
+ * Structures needed by subdrivers *
+ ***********************************/
+
+/* This structure fully describes the IOC4 miscellaneous registers which
+ * appear at bar[0]+0x00000 through bar[0]+0x0005c.  The corresponding
+ * PCI resource is managed by the main IOC4 driver because it contains
+ * registers of interest to many different IOC4 subdrivers.
+ */
+struct ioc4_misc_regs {
+	/* Miscellaneous IOC4 registers */
+	union ioc4_pci_err_addr_l {
+		uint32_t raw;
+		struct {
+			uint32_t valid:1;	/* Address captured */
+			uint32_t master_id:4;	/* Unit causing error
+						 * 0/1: Serial port 0 TX/RX
+						 * 2/3: Serial port 1 TX/RX
+						 * 4/5: Serial port 2 TX/RX
+						 * 6/7: Serial port 3 TX/RX
+						 * 8: ATA/ATAPI
+						 * 9-15: Undefined
+						 */
+			uint32_t mul_err:1;	/* Multiple errors occurred */
+			uint32_t addr:26;	/* Bits 31-6 of error addr */
+		} fields;
+	} pci_err_addr_l;
+	uint32_t pci_err_addr_h;	/* Bits 63-32 of error addr */
+	union ioc4_sio_int {
+		uint32_t raw;
+		struct {
+			uint8_t tx_mt:1;	/* TX ring buffer empty */
+			uint8_t rx_full:1;	/* RX ring buffer full */
+			uint8_t rx_high:1;	/* RX high-water exceeded */
+			uint8_t rx_timer:1;	/* RX timer has triggered */
+			uint8_t delta_dcd:1;	/* DELTA_DCD seen */
+			uint8_t delta_cts:1;	/* DELTA_CTS seen */
+			uint8_t intr_pass:1;	/* Interrupt pass-through */
+			uint8_t tx_explicit:1;	/* TX, MCW, or delay complete */
+		} fields[4];
+	} sio_ir;		/* Serial interrupt state */
+	union ioc4_other_int {
+		uint32_t raw;
+		struct {
+			uint32_t ata_int:1;	/* ATA port passthru */
+			uint32_t ata_memerr:1;	/* ATA halted by mem error */
+			uint32_t memerr:4;	/* Serial halted by mem err */
+			uint32_t kbd_int:1;	/* kbd/mouse intr asserted */
+			uint32_t reserved:16;	/* zero */
+			uint32_t rt_int:1;	/* INT_OUT section latch */
+			uint32_t gen_int:8;	/* Intr. from generic pins */
+		} fields;
+	} other_ir;		/* Other interrupt state */
+	union ioc4_sio_int sio_ies;	/* Serial interrupt enable set */
+	union ioc4_other_int other_ies;	/* Other interrupt enable set */
+	union ioc4_sio_int sio_iec;	/* Serial interrupt enable clear */
+	union ioc4_other_int other_iec;	/* Other interrupt enable clear */
+	union ioc4_sio_cr {
+		uint32_t raw;
+		struct {
+			uint32_t cmd_pulse:4;	/* Bytebus strobe width */
+			uint32_t arb_diag:3;	/* PCI bus requester */
+			uint32_t sio_diag_idle:1;	/* Active ser req? */
+			uint32_t ata_diag_idle:1;	/* Active ATA req? */
+			uint32_t ata_diag_active:1;	/* ATA req is winner */
+			uint32_t reserved:22;	/* zero */
+		} fields;
+	} sio_cr;
+	uint32_t unused1;
+	union ioc4_int_out {
+		uint32_t raw;
+		struct {
+			uint32_t count:16;	/* Period control */
+			uint32_t mode:3;	/* Output signal shape */
+			uint32_t reserved:11;	/* zero */
+			uint32_t diag:1;	/* Timebase control */
+			uint32_t int_out:1;	/* Current value */
+		} fields;
+	} int_out;		/* External interrupt output control */
+	uint32_t unused2;
+	union ioc4_gpcr {
+		uint32_t raw;
+		struct {
+			uint32_t dir:8;	/* Pin direction */
+			uint32_t edge:8;	/* Edge/level mode */
+			uint32_t reserved1:4;	/* zero */
+			uint32_t int_out_en:1;	/* INT_OUT enable */
+			uint32_t reserved2:11;	/* zero */
+		} fields;
+	} gpcr_s;		/* Generic PIO control set */
+	union ioc4_gpcr gpcr_c;	/* Generic PIO control clear */
+	union ioc4_gpdr {
+		uint32_t raw;
+		struct {
+			uint32_t gen_pin:8;	/* State of pins */
+			uint32_t reserved:24;
+		} fields;
+	} gpdr;			/* Generic PIO data */
+	uint32_t unused3;
+	union ioc4_gppr {
+		uint32_t raw;
+		struct {
+			uint32_t gen_pin:1;	/* Single pin state */
+			uint32_t reserved:31;
+		} fields;
+	} gppr[8];		/* Generic PIO pins */
+};
+
+/* One of these per IOC4
+ *
+ * The idd_serial_data field is present here, even though it's used
+ * solely by the serial subdriver, because the main IOC4 module
+ * properly owns pci_{get,set}_drvdata functionality.  This field
+ * allows that subdriver to stash its own drvdata somewhere.
+ */
+struct ioc4_driver_data {
+	struct list_head idd_list;
+	unsigned long idd_bar0;
+	struct pci_dev *idd_pdev;
+	const struct pci_device_id *idd_pci_id;
+	struct __iomem ioc4_misc_regs *idd_misc_regs;
+	void *idd_serial_data;
+};
+
+/* One per submodule */
+struct ioc4_submodule {
+	struct list_head is_list;
+	char *is_name;
+	struct module *is_owner;
+	int (*is_probe) (struct ioc4_driver_data *);
+	int (*is_remove) (struct ioc4_driver_data *);
+};
+
+#define IOC4_NUM_CARDS		8	/* max cards per partition */
+
+/**********************************
+ * Functions needed by submodules *
+ **********************************/
+
+extern int ioc4_register_submodule(struct ioc4_submodule *);
+extern void ioc4_unregister_submodule(struct ioc4_submodule *);
+
+#endif				/* _LINUX_IOC4_H */
diff --git a/include/linux/ioc4_common.h b/include/linux/ioc4_common.h
deleted file mode 100644
index b03bcc46df5..00000000000
--- a/include/linux/ioc4_common.h
+++ /dev/null
@@ -1,21 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (c) 2005 Silicon Graphics, Inc.  All Rights Reserved.
- */
-
-#ifndef _LINUX_IOC4_COMMON_H
-#define _LINUX_IOC4_COMMON_H
-
-/* prototypes */
-
-int ioc4_serial_init(void);
-
-int ioc4_serial_attach_one(struct pci_dev *pdev, const struct
-				pci_device_id *pci_id);
-int ioc4_ide_attach_one(struct pci_dev *pdev, const struct
-				pci_device_id *pci_id);
-
-#endif	/* _LINUX_IOC4_COMMON_H */
-- 
cgit v1.2.3-70-g09d2


From d4c477ca5448f19afaaf6c0cfd655009ea9e614d Mon Sep 17 00:00:00 2001
From: Brent Casavant <bcasavan@sgi.com>
Date: Tue, 21 Jun 2005 17:16:01 -0700
Subject: [PATCH] ioc4: PCI bus speed detection

Several hardware features of SGI's IOC4 I/O controller chip require
timing-related driver calculations dependent upon the PCI bus speed.  This
patch enables the core IOC4 driver code to detect the actual bus speed and
store a value that can later be used by the IOC4 subdrivers as needed.

Signed-off-by: Brent Casavant <bcasavan@sgi.com>
Acked-by: Pat Gefre <pfg@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/serial/ioc4_serial.c |  31 ++++++-----
 drivers/sn/Kconfig           |   2 +-
 drivers/sn/ioc4.c            | 128 +++++++++++++++++++++++++++++++++++++++++++
 include/linux/ioc4.h         |  37 ++++++++++---
 4 files changed, 177 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/drivers/serial/ioc4_serial.c b/drivers/serial/ioc4_serial.c
index da5f10eb484..793c3a7cbe4 100644
--- a/drivers/serial/ioc4_serial.c
+++ b/drivers/serial/ioc4_serial.c
@@ -299,7 +299,6 @@ struct ioc4_serial {
 } ioc4_serial;
 
 /* UART clock speed */
-#define IOC4_SER_XIN_CLK        IOC4_SER_XIN_CLK_66
 #define IOC4_SER_XIN_CLK_66     66666667
 #define IOC4_SER_XIN_CLK_33     33333333
 
@@ -1012,21 +1011,20 @@ static irqreturn_t ioc4_intr(int irq, void *arg, struct pt_regs *regs)
  * ioc4_attach_local - Device initialization.
  *			Called at *_attach() time for each
  *			IOC4 with serial ports in the system.
- * @control: ioc4_control ptr
- * @pdev: PCI handle for this device
- * @soft: soft struct for this device
- * @ioc4: ioc4 mem space
+ * @idd: Master module data for this IOC4
  */
-static int inline ioc4_attach_local(struct pci_dev *pdev,
-			struct ioc4_control *control,
-			struct ioc4_soft *soft, void __iomem *ioc4_misc,
-			void __iomem *ioc4_serial)
+static int inline ioc4_attach_local(struct ioc4_driver_data *idd)
 {
 	struct ioc4_port *port;
 	struct ioc4_port *ports[IOC4_NUM_SERIAL_PORTS];
 	int port_number;
 	uint16_t ioc4_revid_min = 62;
 	uint16_t ioc4_revid;
+	struct pci_dev *pdev = idd->idd_pdev;
+	struct ioc4_control* control = idd->idd_serial_data;
+	struct ioc4_soft *soft = control->ic_soft;
+	void __iomem *ioc4_misc = idd->idd_misc_regs;
+	void __iomem *ioc4_serial = soft->is_ioc4_serial_addr;
 
 	/* IOC4 firmware must be at least rev 62 */
 	pci_read_config_word(pdev, PCI_COMMAND_SPECIAL, &ioc4_revid);
@@ -1063,7 +1061,15 @@ static int inline ioc4_attach_local(struct pci_dev *pdev,
 		port->ip_ioc4_soft = soft;
 		port->ip_pdev = pdev;
 		port->ip_ienb = 0;
-		port->ip_pci_bus_speed = IOC4_SER_XIN_CLK;
+		/* Use baud rate calculations based on detected PCI
+		 * bus speed.  Simply test whether the PCI clock is
+		 * running closer to 66MHz or 33MHz.
+		 */
+		if (idd->count_period/IOC4_EXTINT_COUNT_DIVISOR < 20) {
+			port->ip_pci_bus_speed = IOC4_SER_XIN_CLK_66;
+		} else {
+			port->ip_pci_bus_speed = IOC4_SER_XIN_CLK_33;
+		}
 		port->ip_baud = 9600;
 		port->ip_control = control;
 		port->ip_mem = ioc4_misc;
@@ -2733,9 +2739,8 @@ ioc4_serial_attach_one(struct ioc4_driver_data *idd)
 		    "%s : request_irq fails for IRQ 0x%x\n ",
 			__FUNCTION__, idd->idd_pdev->irq);
 	}
-	if ((ret = ioc4_attach_local(idd->idd_pdev, control, soft,
-				soft->is_ioc4_misc_addr,
-				soft->is_ioc4_serial_addr)))
+	ret = ioc4_attach_local(idd);
+	if (ret)
 		goto out4;
 
 	/* register port with the serial core */
diff --git a/drivers/sn/Kconfig b/drivers/sn/Kconfig
index 20f7515ab83..13b8d249da5 100644
--- a/drivers/sn/Kconfig
+++ b/drivers/sn/Kconfig
@@ -6,7 +6,7 @@ menu "SN Devices"
 
 config SGI_IOC4
 	tristate "SGI IOC4 Base IO support"
-	depends on IA64_GENERIC || IA64_SGI_SN2
+	depends on (IA64_GENERIC || IA64_SGI_SN2) && MMTIMER
 	default m
 	---help---
 	This option enables basic support for the SGI IOC4-based Base IO
diff --git a/drivers/sn/ioc4.c b/drivers/sn/ioc4.c
index 70862d72ea9..ea75b3d0612 100644
--- a/drivers/sn/ioc4.c
+++ b/drivers/sn/ioc4.c
@@ -29,7 +29,26 @@
 #include <linux/module.h>
 #include <linux/pci.h>
 #include <linux/ioc4.h>
+#include <linux/mmtimer.h>
+#include <linux/rtc.h>
 #include <linux/rwsem.h>
+#include <asm/sn/addrs.h>
+#include <asm/sn/clksupport.h>
+#include <asm/sn/shub_mmr.h>
+
+/***************
+ * Definitions *
+ ***************/
+
+/* Tweakable values */
+
+/* PCI bus speed detection/calibration */
+#define IOC4_CALIBRATE_COUNT 63	/* Calibration cycle period */
+#define IOC4_CALIBRATE_CYCLES 256	/* Average over this many cycles */
+#define IOC4_CALIBRATE_DISCARD 2	/* Discard first few cycles */
+#define IOC4_CALIBRATE_LOW_MHZ 25	/* Lower bound on bus speed sanity */
+#define IOC4_CALIBRATE_HIGH_MHZ 75	/* Upper bound on bus speed sanity */
+#define IOC4_CALIBRATE_DEFAULT_MHZ 66	/* Assumed if sanity check fails */
 
 /************************
  * Submodule management *
@@ -101,6 +120,112 @@ ioc4_unregister_submodule(struct ioc4_submodule *is)
  * Device management *
  *********************/
 
+#define IOC4_CALIBRATE_LOW_LIMIT \
+	(1000*IOC4_EXTINT_COUNT_DIVISOR/IOC4_CALIBRATE_LOW_MHZ)
+#define IOC4_CALIBRATE_HIGH_LIMIT \
+	(1000*IOC4_EXTINT_COUNT_DIVISOR/IOC4_CALIBRATE_HIGH_MHZ)
+#define IOC4_CALIBRATE_DEFAULT \
+	(1000*IOC4_EXTINT_COUNT_DIVISOR/IOC4_CALIBRATE_DEFAULT_MHZ)
+
+#define IOC4_CALIBRATE_END \
+	(IOC4_CALIBRATE_CYCLES + IOC4_CALIBRATE_DISCARD)
+
+#define IOC4_INT_OUT_MODE_TOGGLE 0x7	/* Toggle INT_OUT every COUNT+1 ticks */
+
+/* Determines external interrupt output clock period of the PCI bus an
+ * IOC4 is attached to.  This value can be used to determine the PCI
+ * bus speed.
+ *
+ * IOC4 has a design feature that various internal timers are derived from
+ * the PCI bus clock.  This causes IOC4 device drivers to need to take the
+ * bus speed into account when setting various register values (e.g. INT_OUT
+ * register COUNT field, UART divisors, etc).  Since this information is
+ * needed by several subdrivers, it is determined by the main IOC4 driver,
+ * even though the following code utilizes external interrupt registers
+ * to perform the speed calculation.
+ */
+static void
+ioc4_clock_calibrate(struct ioc4_driver_data *idd)
+{
+	extern unsigned long sn_rtc_cycles_per_second;
+	union ioc4_int_out int_out;
+	union ioc4_gpcr gpcr;
+	unsigned int state, last_state = 1;
+	uint64_t start = 0, end, period;
+	unsigned int count = 0;
+
+	/* Enable output */
+	gpcr.raw = 0;
+	gpcr.fields.dir = IOC4_GPCR_DIR_0;
+	gpcr.fields.int_out_en = 1;
+	writel(gpcr.raw, &idd->idd_misc_regs->gpcr_s.raw);
+
+	/* Reset to power-on state */
+	writel(0, &idd->idd_misc_regs->int_out.raw);
+	mmiowb();
+
+	printk(KERN_INFO
+	       "%s: Calibrating PCI bus speed "
+	       "for pci_dev %s ... ", __FUNCTION__, pci_name(idd->idd_pdev));
+	/* Set up square wave */
+	int_out.raw = 0;
+	int_out.fields.count = IOC4_CALIBRATE_COUNT;
+	int_out.fields.mode = IOC4_INT_OUT_MODE_TOGGLE;
+	int_out.fields.diag = 0;
+	writel(int_out.raw, &idd->idd_misc_regs->int_out.raw);
+	mmiowb();
+
+	/* Check square wave period averaged over some number of cycles */
+	do {
+		int_out.raw = readl(&idd->idd_misc_regs->int_out.raw);
+		state = int_out.fields.int_out;
+		if (!last_state && state) {
+			count++;
+			if (count == IOC4_CALIBRATE_END) {
+				end = rtc_time();
+				break;
+			} else if (count == IOC4_CALIBRATE_DISCARD)
+				start = rtc_time();
+		}
+		last_state = state;
+	} while (1);
+
+	/* Calculation rearranged to preserve intermediate precision.
+	 * Logically:
+	 * 1. "end - start" gives us number of RTC cycles over all the
+	 *    square wave cycles measured.
+	 * 2. Divide by number of square wave cycles to get number of
+	 *    RTC cycles per square wave cycle.
+	 * 3. Divide by 2*(int_out.fields.count+1), which is the formula
+	 *    by which the IOC4 generates the square wave, to get the
+	 *    number of RTC cycles per IOC4 INT_OUT count.
+	 * 4. Divide by sn_rtc_cycles_per_second to get seconds per
+	 *    count.
+	 * 5. Multiply by 1E9 to get nanoseconds per count.
+	 */
+	period = ((end - start) * 1000000000) /
+	    (IOC4_CALIBRATE_CYCLES * 2 * (IOC4_CALIBRATE_COUNT + 1)
+	     * sn_rtc_cycles_per_second);
+
+	/* Bounds check the result. */
+	if (period > IOC4_CALIBRATE_LOW_LIMIT ||
+	    period < IOC4_CALIBRATE_HIGH_LIMIT) {
+		printk("failed. Assuming PCI clock ticks are %d ns.\n",
+		       IOC4_CALIBRATE_DEFAULT / IOC4_EXTINT_COUNT_DIVISOR);
+		period = IOC4_CALIBRATE_DEFAULT;
+	} else {
+		printk("succeeded. PCI clock ticks are %ld ns.\n",
+		       period / IOC4_EXTINT_COUNT_DIVISOR);
+	}
+
+	/* Remember results.  We store the extint clock period rather
+	 * than the PCI clock period so that greater precision is
+	 * retained.  Divide by IOC4_EXTINT_COUNT_DIVISOR to get
+	 * PCI clock period.
+	 */
+	idd->count_period = period;
+}
+
 /* Adds a new instance of an IOC4 card */
 static int
 ioc4_probe(struct pci_dev *pdev, const struct pci_device_id *pci_id)
@@ -170,6 +295,9 @@ ioc4_probe(struct pci_dev *pdev, const struct pci_device_id *pci_id)
 	pci_write_config_dword(idd->idd_pdev, PCI_COMMAND,
 			       pcmd | PCI_COMMAND_PARITY | PCI_COMMAND_SERR);
 
+	/* Determine PCI clock */
+	ioc4_clock_calibrate(idd);
+
 	/* Disable/clear all interrupts.  Need to do this here lest
 	 * one submodule request the shared IOC4 IRQ, but interrupt
 	 * is generated by a different subdevice.
diff --git a/include/linux/ioc4.h b/include/linux/ioc4.h
index 729bfa4c4ac..3dd18b785eb 100644
--- a/include/linux/ioc4.h
+++ b/include/linux/ioc4.h
@@ -11,6 +11,14 @@
 
 #include <linux/interrupt.h>
 
+/***************
+ * Definitions *
+ ***************/
+
+/* Miscellaneous values inherent to hardware */
+
+#define IOC4_EXTINT_COUNT_DIVISOR 520	/* PCI clocks per COUNT tick */
+
 /***********************************
  * Structures needed by subdrivers *
  ***********************************/
@@ -119,19 +127,34 @@ struct ioc4_misc_regs {
 	} gppr[8];		/* Generic PIO pins */
 };
 
-/* One of these per IOC4
- *
- * The idd_serial_data field is present here, even though it's used
- * solely by the serial subdriver, because the main IOC4 module
- * properly owns pci_{get,set}_drvdata functionality.  This field
- * allows that subdriver to stash its own drvdata somewhere.
- */
+/* Masks for GPCR DIR pins */
+#define IOC4_GPCR_DIR_0 0x01	/* External interrupt output */
+#define IOC4_GPCR_DIR_1 0x02	/* External interrupt input */
+#define IOC4_GPCR_DIR_2 0x04
+#define IOC4_GPCR_DIR_3 0x08	/* Keyboard/mouse presence */
+#define IOC4_GPCR_DIR_4 0x10	/* Ser. port 0 xcvr select (0=232, 1=422) */
+#define IOC4_GPCR_DIR_5 0x20	/* Ser. port 1 xcvr select (0=232, 1=422) */
+#define IOC4_GPCR_DIR_6 0x40	/* Ser. port 2 xcvr select (0=232, 1=422) */
+#define IOC4_GPCR_DIR_7 0x80	/* Ser. port 3 xcvr select (0=232, 1=422) */
+
+/* Masks for GPCR EDGE pins */
+#define IOC4_GPCR_EDGE_0 0x01
+#define IOC4_GPCR_EDGE_1 0x02	/* External interrupt input */
+#define IOC4_GPCR_EDGE_2 0x04
+#define IOC4_GPCR_EDGE_3 0x08
+#define IOC4_GPCR_EDGE_4 0x10
+#define IOC4_GPCR_EDGE_5 0x20
+#define IOC4_GPCR_EDGE_6 0x40
+#define IOC4_GPCR_EDGE_7 0x80
+
+/* One of these per IOC4 */
 struct ioc4_driver_data {
 	struct list_head idd_list;
 	unsigned long idd_bar0;
 	struct pci_dev *idd_pdev;
 	const struct pci_device_id *idd_pci_id;
 	struct __iomem ioc4_misc_regs *idd_misc_regs;
+	unsigned long count_period;
 	void *idd_serial_data;
 };
 
-- 
cgit v1.2.3-70-g09d2


From 2368086344c3d67b0f4aecac39d620fb9b8795c3 Mon Sep 17 00:00:00 2001
From: Hirokazu Takata <takata@linux-m32r.org>
Date: Tue, 21 Jun 2005 17:16:10 -0700
Subject: [PATCH] m32r: Support M3A-2170(Mappi-III) platform

This patchset is for supporting a new m32r platform, M3A-2170(Mappi-III)
evaluation board.  An M32R chip multiprocessor is equipped on the board.
http://http://www.linux-m32r.org/eng/platform/platform.html

	* arch/m32r/Kconfig: Support Mappi-III platform.
	* arch/m32r/kernel/Makefile: ditto.
	* arch/m32r/kernel/io_mappi3.c: ditto.
	* arch/m32r/kernel/setup.c: ditto.
	* arch/m32r/kernel/setup_mappi3.c: ditto.
	* include/asm-m32r/m32102.h: ditto.
	* include/asm-m32r/m32r.h: ditto.
	* include/asm-m32r/mappi3/mappi3_pld.h: ditto.

	* include/asm-m32r/ide.h: CF support for Mappi-III.
	* arch/m32r/kernel/setup_mappi3.c: ditto.

	* arch/m32r/mappi3/defconfig.smp: A default config file for Mappi-III.
	* arch/m32r/mappi3/dot.gdbinit: A default .gdbinit file for Mappi-III.

	* arch/m32r/boot/compressed/m32r_sio.c: Modified for Mappi-III
	  - At boot time, m32r-g00ff bootloader makes MMU off for Mappi-III,
	    on the contrary it makes MMU on for Mappi-II.

	* arch/m32r/kernel/io_mappi2.c: Update comments.
	* arch/m32r/kernel/setup_mappi2.c: ditto.

Signed-off-by: Mamoru Sakugawa <sakugawa@linux-m32r.org>
Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/m32r/Kconfig                    |   7 +-
 arch/m32r/boot/compressed/m32r_sio.c |  19 +-
 arch/m32r/kernel/Makefile            |   2 +-
 arch/m32r/kernel/io_m32700ut.c       |  92 +----
 arch/m32r/kernel/io_mappi.c          |  74 +---
 arch/m32r/kernel/io_mappi2.c         |  92 +----
 arch/m32r/kernel/io_mappi3.c         | 378 ++++++++++++++++++
 arch/m32r/kernel/io_oaks32r.c        |  36 +-
 arch/m32r/kernel/io_opsput.c         |  79 +---
 arch/m32r/kernel/io_usrv.c           |  51 +--
 arch/m32r/kernel/setup.c             |   2 +
 arch/m32r/kernel/setup_mappi2.c      |   6 +-
 arch/m32r/kernel/setup_mappi3.c      | 208 ++++++++++
 arch/m32r/mappi3/defconfig.smp       | 751 +++++++++++++++++++++++++++++++++++
 arch/m32r/mappi3/dot.gdbinit         | 224 +++++++++++
 include/asm-m32r/ide.h               |   2 +-
 include/asm-m32r/m32102.h            |   1 +
 include/asm-m32r/m32r.h              |   4 +
 include/asm-m32r/mappi3/mappi3_pld.h | 143 +++++++
 19 files changed, 1785 insertions(+), 386 deletions(-)
 create mode 100644 arch/m32r/kernel/io_mappi3.c
 create mode 100644 arch/m32r/kernel/setup_mappi3.c
 create mode 100644 arch/m32r/mappi3/defconfig.smp
 create mode 100644 arch/m32r/mappi3/dot.gdbinit
 create mode 100644 include/asm-m32r/mappi3/mappi3_pld.h

(limited to 'include')

diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig
index 0d90ea58a0c..64c133344af 100644
--- a/arch/m32r/Kconfig
+++ b/arch/m32r/Kconfig
@@ -78,6 +78,9 @@ config PLAT_OAKS32R
 config PLAT_MAPPI2
        bool "Mappi-II(M3A-ZA36/M3A-ZA52)"
 
+config PLAT_MAPPI3
+       bool "Mappi-III(M3A-2170)"
+
 endchoice
 
 choice
@@ -134,6 +137,7 @@ config BUS_CLOCK
 	int "Bus Clock [Hz] (integer)"
 	default "70000000" if PLAT_MAPPI
 	default "25000000" if PLAT_USRV
+	default "50000000" if PLAT_MAPPI3
 	default "50000000" if PLAT_M32700UT
 	default "50000000" if PLAT_OPSPUT
 	default "33333333" if PLAT_OAKS32R
@@ -149,7 +153,7 @@ config CPU_LITTLE_ENDIAN
 
 config MEMORY_START
 	hex "Physical memory start address (hex)"
-	default "08000000" if PLAT_MAPPI || PLAT_MAPPI2
+	default "08000000" if PLAT_MAPPI || PLAT_MAPPI2 || PLAT_MAPPI3
 	default "08000000" if PLAT_USRV
 	default "08000000" if PLAT_M32700UT
 	default "08000000" if PLAT_OPSPUT
@@ -157,6 +161,7 @@ config MEMORY_START
 
 config MEMORY_SIZE
 	hex "Physical memory size (hex)"
+	default "08000000" if PLAT_MAPPI3
 	default "04000000" if PLAT_MAPPI || PLAT_MAPPI2
 	default "02000000" if PLAT_USRV
 	default "01000000" if PLAT_M32700UT
diff --git a/arch/m32r/boot/compressed/m32r_sio.c b/arch/m32r/boot/compressed/m32r_sio.c
index bad5475eff9..8f9a57271f8 100644
--- a/arch/m32r/boot/compressed/m32r_sio.c
+++ b/arch/m32r/boot/compressed/m32r_sio.c
@@ -6,6 +6,7 @@
  */
 
 #include <linux/config.h>
+#include <asm/processor.h>
 
 static void putc(char c);
 
@@ -38,16 +39,17 @@ static int puts(const char *s)
 
 static void putc(char c)
 {
-
-	while ((*BOOT_SIO0STS & 0x3) != 0x3) ;
+	while ((*BOOT_SIO0STS & 0x3) != 0x3)
+		cpu_relax();
 	if (c == '\n') {
 		*BOOT_SIO0TXB = '\r';
-		while ((*BOOT_SIO0STS & 0x3) != 0x3) ;
+		while ((*BOOT_SIO0STS & 0x3) != 0x3)
+			cpu_relax();
 	}
 	*BOOT_SIO0TXB = c;
 }
-#else /* defined(CONFIG_PLAT_M32700UT_Alpha) || defined(CONFIG_PLAT_M32700UT) */
-#ifdef CONFIG_MMU
+#else /* !(CONFIG_PLAT_M32700UT_Alpha) && !(CONFIG_PLAT_M32700UT) */
+#if defined(CONFIG_PLAT_MAPPI2)
 #define SIO0STS	(volatile unsigned short *)(0xa0efd000 + 14)
 #define SIO0TXB	(volatile unsigned short *)(0xa0efd000 + 30)
 #else
@@ -57,11 +59,12 @@ static void putc(char c)
 
 static void putc(char c)
 {
-
-	while ((*SIO0STS & 0x1) == 0) ;
+	while ((*SIO0STS & 0x1) == 0)
+		cpu_relax();
 	if (c == '\n') {
 		*SIO0TXB = '\r';
-		while ((*SIO0STS & 0x1) == 0) ;
+		while ((*SIO0STS & 0x1) == 0)
+			cpu_relax();
 	}
 	*SIO0TXB = c;
 }
diff --git a/arch/m32r/kernel/Makefile b/arch/m32r/kernel/Makefile
index cfd690bf6d8..6c6b6c37663 100644
--- a/arch/m32r/kernel/Makefile
+++ b/arch/m32r/kernel/Makefile
@@ -10,6 +10,7 @@ obj-y	:= process.o entry.o traps.o align.o irq.o setup.o time.o \
 obj-$(CONFIG_SMP)		+= smp.o smpboot.o
 obj-$(CONFIG_PLAT_MAPPI)	+= setup_mappi.o io_mappi.o
 obj-$(CONFIG_PLAT_MAPPI2)	+= setup_mappi2.o io_mappi2.o
+obj-$(CONFIG_PLAT_MAPPI3)	+= setup_mappi3.o io_mappi3.o
 obj-$(CONFIG_PLAT_USRV)		+= setup_usrv.o io_usrv.o
 obj-$(CONFIG_PLAT_M32700UT)	+= setup_m32700ut.o io_m32700ut.o
 obj-$(CONFIG_PLAT_OPSPUT)	+= setup_opsput.o io_opsput.o
@@ -17,4 +18,3 @@ obj-$(CONFIG_MODULES)		+= module.o
 obj-$(CONFIG_PLAT_OAKS32R)	+= setup_oaks32r.o io_oaks32r.o
 
 EXTRA_AFLAGS	:= -traditional
-
diff --git a/arch/m32r/kernel/io_m32700ut.c b/arch/m32r/kernel/io_m32700ut.c
index 371ba904e96..e545b065f7e 100644
--- a/arch/m32r/kernel/io_m32700ut.c
+++ b/arch/m32r/kernel/io_m32700ut.c
@@ -3,8 +3,8 @@
  *
  *  Typical I/O routines for M32700UT board.
  *
- *  Copyright (c) 2001, 2002  Hiroyuki Kondo, Hirokazu Takata,
- *                            Hitoshi Yamamoto, Takeo Takahashi
+ *  Copyright (c) 2001-2005  Hiroyuki Kondo, Hirokazu Takata,
+ *                           Hitoshi Yamamoto, Takeo Takahashi
  *
  *  This file is subject to the terms and conditions of the GNU General
  *  Public License.  See the file "COPYING" in the main directory of this
@@ -172,64 +172,21 @@ unsigned long _inl(unsigned long port)
 
 unsigned char _inb_p(unsigned long port)
 {
-	unsigned char  v;
-
-	if (port >= LAN_IOSTART && port < LAN_IOEND)
-		v = _ne_inb(PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
-	if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
-		return *(volatile unsigned char *)__port2addr_ata(port);
-	} else
-#endif
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		unsigned char b;
-		pcc_ioread_byte(0, port, &b, sizeof(b), 1, 0);
-		return b;
-	} else
-#endif
-		v = *(volatile unsigned char *)PORT2ADDR(port);
-
+	unsigned char v = _inb(port);
 	delay();
 	return (v);
 }
 
 unsigned short _inw_p(unsigned long port)
 {
-	unsigned short  v;
-
-	if (port >= LAN_IOSTART && port < LAN_IOEND)
-		v = _ne_inw(PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
-	if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
-		return *(volatile unsigned short *)__port2addr_ata(port);
-	} else
-#endif
-#if defined(CONFIG_USB)
-	if(port >= 0x340 && port < 0x3a0)
-		return *(volatile unsigned short *)PORT2ADDR_USB(port);
-	else
-#endif
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		unsigned short w;
-		pcc_ioread_word(0, port, &w, sizeof(w), 1, 0);
-		return w;
-	} else
-#endif
-		v = *(volatile unsigned short *)PORT2ADDR(port);
-
+	unsigned short v = _inw(port);
 	delay();
 	return (v);
 }
 
 unsigned long _inl_p(unsigned long port)
 {
-	unsigned long  v;
-
-	v = *(volatile unsigned long *)PORT2ADDR(port);
+	unsigned long v = _inl(port);
 	delay();
 	return (v);
 }
@@ -287,52 +244,19 @@ void _outl(unsigned long l, unsigned long port)
 
 void _outb_p(unsigned char b, unsigned long port)
 {
-	if (port >= LAN_IOSTART && port < LAN_IOEND)
-		_ne_outb(b, PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
-	if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
-		*(volatile unsigned char *)__port2addr_ata(port) = b;
-	} else
-#endif
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		pcc_iowrite_byte(0, port, &b, sizeof(b), 1, 0);
-	} else
-#endif
-		*(volatile unsigned char *)PORT2ADDR(port) = b;
-
+	_outb(b, port);
 	delay();
 }
 
 void _outw_p(unsigned short w, unsigned long port)
 {
-	if (port >= LAN_IOSTART && port < LAN_IOEND)
-		_ne_outw(w, PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
-	if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
-		*(volatile unsigned short *)__port2addr_ata(port) = w;
-	} else
-#endif
-#if defined(CONFIG_USB)
-	if(port >= 0x340 && port < 0x3a0)
-		*(volatile unsigned short *)PORT2ADDR_USB(port) = w;
-	else
-#endif
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		pcc_iowrite_word(0, port, &w, sizeof(w), 1, 0);
-	} else
-#endif
-		*(volatile unsigned short *)PORT2ADDR(port) = w;
-
+	_outw(w, port);
 	delay();
 }
 
 void _outl_p(unsigned long l, unsigned long port)
 {
-	*(volatile unsigned long *)PORT2ADDR(port) = l;
+	_outl(l, port);
 	delay();
 }
 
diff --git a/arch/m32r/kernel/io_mappi.c b/arch/m32r/kernel/io_mappi.c
index 85688ffb52f..78033165fb5 100644
--- a/arch/m32r/kernel/io_mappi.c
+++ b/arch/m32r/kernel/io_mappi.c
@@ -3,8 +3,8 @@
  *
  *  Typical I/O routines for Mappi board.
  *
- *  Copyright (c) 2001, 2002  Hiroyuki Kondo, Hirokazu Takata,
- *                            Hitoshi Yamamoto
+ *  Copyright (c) 2001-2005  Hiroyuki Kondo, Hirokazu Takata,
+ *                           Hitoshi Yamamoto
  */
 
 #include <linux/config.h>
@@ -130,57 +130,21 @@ unsigned long _inl(unsigned long port)
 
 unsigned char _inb_p(unsigned long port)
 {
-	unsigned char  v;
-
-	if (port >= 0x300 && port < 0x320)
-		v = _ne_inb(PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_PCC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		unsigned char b;
-		pcc_ioread(0, port, &b, sizeof(b), 1, 0);
-		return b;
-	} else 	if (port >= M32R_PCC_IOSTART1 && port <= M32R_PCC_IOEND1) {
-		unsigned char b;
-		pcc_ioread(1, port, &b, sizeof(b), 1, 0);
-		return b;
-	} else
-#endif
-		v = *(volatile unsigned char *)PORT2ADDR(port);
-
+	unsigned char v = _inb(port);
 	delay();
 	return (v);
 }
 
 unsigned short _inw_p(unsigned long port)
 {
-	unsigned short  v;
-
-	if (port >= 0x300 && port < 0x320)
-		v = _ne_inw(PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_PCC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		unsigned short w;
-		pcc_ioread(0, port, &w, sizeof(w), 1, 0);
-		return w;
-	} else 	if (port >= M32R_PCC_IOSTART1 && port <= M32R_PCC_IOEND1) {
-		unsigned short w;
-		pcc_ioread(1, port, &w, sizeof(w), 1, 0);
-		return w;
-	} else
-#endif
-		v = *(volatile unsigned short *)PORT2ADDR(port);
-
+	unsigned short v = _inw(port);
 	delay();
 	return (v);
 }
 
 unsigned long _inl_p(unsigned long port)
 {
-	unsigned long  v;
-
-	v = *(volatile unsigned long *)PORT2ADDR(port);
+	unsigned long v = _inl(port);
 	delay();
 	return (v);
 }
@@ -229,41 +193,19 @@ void _outl(unsigned long l, unsigned long port)
 
 void _outb_p(unsigned char b, unsigned long port)
 {
-	if (port >= 0x300 && port < 0x320)
-		_ne_outb(b, PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_PCC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		pcc_iowrite(0, port, &b, sizeof(b), 1, 0);
-	} else 	if (port >= M32R_PCC_IOSTART1 && port <= M32R_PCC_IOEND1) {
-		pcc_iowrite(1, port, &b, sizeof(b), 1, 0);
-	} else
-#endif
-		*(volatile unsigned char *)PORT2ADDR(port) = b;
-
+	_outb(b, port);
 	delay();
 }
 
 void _outw_p(unsigned short w, unsigned long port)
 {
-	if (port >= 0x300 && port < 0x320)
-		_ne_outw(w, PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_PCC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		pcc_iowrite(0, port, &w, sizeof(w), 1, 0);
-	} else 	if (port >= M32R_PCC_IOSTART1 && port <= M32R_PCC_IOEND1) {
-		pcc_iowrite(1, port, &w, sizeof(w), 1, 0);
-	} else
-#endif
-		*(volatile unsigned short *)PORT2ADDR(port) = w;
-
+	_outw(w, port);
 	delay();
 }
 
 void _outl_p(unsigned long l, unsigned long port)
 {
-	*(volatile unsigned long *)PORT2ADDR(port) = l;
+	_outl(l, port);
 	delay();
 }
 
diff --git a/arch/m32r/kernel/io_mappi2.c b/arch/m32r/kernel/io_mappi2.c
index 4182cd4f97c..5c03504bf65 100644
--- a/arch/m32r/kernel/io_mappi2.c
+++ b/arch/m32r/kernel/io_mappi2.c
@@ -3,7 +3,7 @@
  *
  *  Typical I/O routines for Mappi2 board.
  *
- *  Copyright (c) 2001-2003  Hiroyuki Kondo, Hirokazu Takata,
+ *  Copyright (c) 2001-2005  Hiroyuki Kondo, Hirokazu Takata,
  *                           Hitoshi Yamamoto, Mamoru Sakugawa
  */
 
@@ -25,7 +25,7 @@ extern void pcc_ioread_byte(int, unsigned long, void *, size_t, size_t, int);
 extern void pcc_ioread_word(int, unsigned long, void *, size_t, size_t, int);
 extern void pcc_iowrite_byte(int, unsigned long, void *, size_t, size_t, int);
 extern void pcc_iowrite_word(int, unsigned long, void *, size_t, size_t, int);
-#endif /* CONFIG_PCMCIA && CONFIG_MAPPI2_CFC */
+#endif /* CONFIG_PCMCIA && CONFIG_M32R_CFC */
 
 #define PORT2ADDR(port)      _port2addr(port)
 #define PORT2ADDR_NE(port)   _port2addr_ne(port)
@@ -169,64 +169,21 @@ unsigned long _inl(unsigned long port)
 
 unsigned char _inb_p(unsigned long port)
 {
-	unsigned char  v;
-
-	if (port >= LAN_IOSTART && port < LAN_IOEND)
-		v = _ne_inb(PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
-	if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
-		return *(volatile unsigned char *)__port2addr_ata(port);
-	} else
-#endif
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		unsigned char b;
-		pcc_ioread_byte(0, port, &b, sizeof(b), 1, 0);
-		return b;
-	} else
-#endif
-		v = *(volatile unsigned char *)PORT2ADDR(port);
-
+	unsigned char v = _inb(port);
 	delay();
 	return (v);
 }
 
 unsigned short _inw_p(unsigned long port)
 {
-	unsigned short  v;
-
-	if (port >= LAN_IOSTART && port < LAN_IOEND)
-		v = _ne_inw(PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
-	if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
-		return *(volatile unsigned short *)__port2addr_ata(port);
-	} else
-#endif
-#if defined(CONFIG_USB)
-	if (port >= 0x340 && port < 0x3a0)
-		v = *(volatile unsigned short *)PORT2ADDR_USB(port);
-	else
-#endif
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		unsigned short w;
-		pcc_ioread_word(0, port, &w, sizeof(w), 1, 0);
-		return w;
-	} else
-#endif
-		v = *(volatile unsigned short *)PORT2ADDR(port);
-
+	unsigned short v = _inw(port);
 	delay();
 	return (v);
 }
 
 unsigned long _inl_p(unsigned long port)
 {
-	unsigned long  v;
-
-	v = *(volatile unsigned long *)PORT2ADDR(port);
+	unsigned long v = _inl(port);
 	delay();
 	return (v);
 }
@@ -284,52 +241,19 @@ void _outl(unsigned long l, unsigned long port)
 
 void _outb_p(unsigned char b, unsigned long port)
 {
-	if (port >= LAN_IOSTART && port < LAN_IOEND)
-		_ne_outb(b, PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
-	if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
-		*(volatile unsigned char *)__port2addr_ata(port) = b;
-	} else
-#endif
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		pcc_iowrite_byte(0, port, &b, sizeof(b), 1, 0);
-	} else
-#endif
-		*(volatile unsigned char *)PORT2ADDR(port) = b;
-
+	_outb(b, port);
 	delay();
 }
 
 void _outw_p(unsigned short w, unsigned long port)
 {
-	if (port >= LAN_IOSTART && port < LAN_IOEND)
-		_ne_outw(w, PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
-	if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
-		*(volatile unsigned short *)__port2addr_ata(port) = w;
-	} else
-#endif
-#if defined(CONFIG_USB)
-	  if (port >= 0x340 && port < 0x3a0)
-		*(volatile unsigned short *)PORT2ADDR_USB(port) = w;
-	else
-#endif
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		pcc_iowrite_word(0, port, &w, sizeof(w), 1, 0);
-	} else
-#endif
-		*(volatile unsigned short *)PORT2ADDR(port) = w;
-
+	_outw(w, port);
 	delay();
 }
 
 void _outl_p(unsigned long l, unsigned long port)
 {
-	*(volatile unsigned long *)PORT2ADDR(port) = l;
+	_outl(l, port);
 	delay();
 }
 
diff --git a/arch/m32r/kernel/io_mappi3.c b/arch/m32r/kernel/io_mappi3.c
new file mode 100644
index 00000000000..c80bde65785
--- /dev/null
+++ b/arch/m32r/kernel/io_mappi3.c
@@ -0,0 +1,378 @@
+/*
+ *  linux/arch/m32r/kernel/io_mappi3.c
+ *
+ *  Typical I/O routines for Mappi3 board.
+ *
+ *  Copyright (c) 2001-2005  Hiroyuki Kondo, Hirokazu Takata,
+ *                           Hitoshi Yamamoto, Mamoru Sakugawa
+ */
+
+#include <linux/config.h>
+#include <asm/m32r.h>
+#include <asm/page.h>
+#include <asm/io.h>
+#include <asm/byteorder.h>
+
+#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
+#include <linux/types.h>
+
+#define M32R_PCC_IOMAP_SIZE 0x1000
+
+#define M32R_PCC_IOSTART0 0x1000
+#define M32R_PCC_IOEND0   (M32R_PCC_IOSTART0 + M32R_PCC_IOMAP_SIZE - 1)
+
+extern void pcc_ioread_byte(int, unsigned long, void *, size_t, size_t, int);
+extern void pcc_ioread_word(int, unsigned long, void *, size_t, size_t, int);
+extern void pcc_iowrite_byte(int, unsigned long, void *, size_t, size_t, int);
+extern void pcc_iowrite_word(int, unsigned long, void *, size_t, size_t, int);
+#endif /* CONFIG_PCMCIA && CONFIG_M32R_CFC */
+
+#define PORT2ADDR(port)      _port2addr(port)
+#define PORT2ADDR_NE(port)   _port2addr_ne(port)
+#define PORT2ADDR_USB(port)  _port2addr_usb(port)
+
+static inline void *_port2addr(unsigned long port)
+{
+	return (void *)(port + NONCACHE_OFFSET);
+}
+
+#define LAN_IOSTART	0x300
+#define LAN_IOEND	0x320
+
+#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
+static inline void *__port2addr_ata(unsigned long port)
+{
+	static int	dummy_reg;
+
+	switch (port) {
+	case 0x1f0:	return (void *)0xb4002000;
+	case 0x1f1:	return (void *)0xb4012800;
+	case 0x1f2:	return (void *)0xb4012002;
+	case 0x1f3:	return (void *)0xb4012802;
+	case 0x1f4:	return (void *)0xb4012004;
+	case 0x1f5:	return (void *)0xb4012804;
+	case 0x1f6:	return (void *)0xb4012006;
+	case 0x1f7:	return (void *)0xb4012806;
+	case 0x3f6:	return (void *)0xb401200e;
+	default: 	return (void *)&dummy_reg;
+	}
+}
+#endif
+
+static inline void *_port2addr_ne(unsigned long port)
+{
+	return (void *)(port + NONCACHE_OFFSET + 0x10000000);
+}
+
+static inline void *_port2addr_usb(unsigned long port)
+{
+	return (void *)(port + NONCACHE_OFFSET + 0x12000000);
+}
+static inline void delay(void)
+{
+	__asm__ __volatile__ ("push r0; \n\t pop r0;" : : :"memory");
+}
+
+/*
+ * NIC I/O function
+ */
+
+static inline unsigned char _ne_inb(void *portp)
+{
+	return (unsigned char) *(volatile unsigned char *)portp;
+}
+
+static inline unsigned short _ne_inw(void *portp)
+{
+	return (unsigned short)le16_to_cpu(*(volatile unsigned short *)portp);
+}
+
+static inline void _ne_insb(void *portp, void * addr, unsigned long count)
+{
+	unsigned char *buf = addr;
+
+	while (count--)
+		*buf++ = *(volatile unsigned char *)portp;
+}
+
+static inline void _ne_outb(unsigned char b, void *portp)
+{
+	*(volatile unsigned char *)portp = (unsigned char)b;
+}
+
+static inline void _ne_outw(unsigned short w, void *portp)
+{
+	*(volatile unsigned short *)portp = cpu_to_le16(w);
+}
+
+unsigned char _inb(unsigned long port)
+{
+	if (port >= LAN_IOSTART && port < LAN_IOEND)
+		return _ne_inb(PORT2ADDR_NE(port));
+#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
+	else if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
+		return *(volatile unsigned char *)__port2addr_ata(port);
+	}
+#endif
+#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
+	else if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
+		unsigned char b;
+		pcc_ioread_byte(0, port, &b, sizeof(b), 1, 0);
+		return b;
+	} else
+#endif
+	return *(volatile unsigned char *)PORT2ADDR(port);
+}
+
+unsigned short _inw(unsigned long port)
+{
+	if (port >= LAN_IOSTART && port < LAN_IOEND)
+		return _ne_inw(PORT2ADDR_NE(port));
+#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
+	else if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
+		return *(volatile unsigned short *)__port2addr_ata(port);
+	}
+#endif
+#if defined(CONFIG_USB)
+	else if (port >= 0x340 && port < 0x3a0)
+		return *(volatile unsigned short *)PORT2ADDR_USB(port);
+#endif
+
+#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
+	else if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
+		unsigned short w;
+		pcc_ioread_word(0, port, &w, sizeof(w), 1, 0);
+		return w;
+	} else
+#endif
+	return *(volatile unsigned short *)PORT2ADDR(port);
+}
+
+unsigned long _inl(unsigned long port)
+{
+#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
+	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
+		unsigned long l;
+		pcc_ioread_word(0, port, &l, sizeof(l), 1, 0);
+		return l;
+	} else
+#endif
+	return *(volatile unsigned long *)PORT2ADDR(port);
+}
+
+unsigned char _inb_p(unsigned long port)
+{
+	unsigned char v = _inb(port);
+	delay();
+	return (v);
+}
+
+unsigned short _inw_p(unsigned long port)
+{
+	unsigned short v = _inw(port);
+	delay();
+	return (v);
+}
+
+unsigned long _inl_p(unsigned long port)
+{
+	unsigned long v = _inl(port);
+	delay();
+	return (v);
+}
+
+void _outb(unsigned char b, unsigned long port)
+{
+	if (port >= LAN_IOSTART && port < LAN_IOEND)
+		_ne_outb(b, PORT2ADDR_NE(port));
+	else
+#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
+	if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
+		*(volatile unsigned char *)__port2addr_ata(port) = b;
+	} else
+#endif
+#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
+	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
+		pcc_iowrite_byte(0, port, &b, sizeof(b), 1, 0);
+	} else
+#endif
+		*(volatile unsigned char *)PORT2ADDR(port) = b;
+}
+
+void _outw(unsigned short w, unsigned long port)
+{
+	if (port >= LAN_IOSTART && port < LAN_IOEND)
+		_ne_outw(w, PORT2ADDR_NE(port));
+	else
+#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
+	if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
+		*(volatile unsigned short *)__port2addr_ata(port) = w;
+	} else
+#endif
+#if defined(CONFIG_USB)
+	if (port >= 0x340 && port < 0x3a0)
+		*(volatile unsigned short *)PORT2ADDR_USB(port) = w;
+	else
+#endif
+#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
+	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
+		pcc_iowrite_word(0, port, &w, sizeof(w), 1, 0);
+	} else
+#endif
+		*(volatile unsigned short *)PORT2ADDR(port) = w;
+}
+
+void _outl(unsigned long l, unsigned long port)
+{
+#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
+	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
+		pcc_iowrite_word(0, port, &l, sizeof(l), 1, 0);
+	} else
+#endif
+	*(volatile unsigned long *)PORT2ADDR(port) = l;
+}
+
+void _outb_p(unsigned char b, unsigned long port)
+{
+	_outb(b, port);
+	delay();
+}
+
+void _outw_p(unsigned short w, unsigned long port)
+{
+	_outw(w, port);
+	delay();
+}
+
+void _outl_p(unsigned long l, unsigned long port)
+{
+	_outl(l, port);
+	delay();
+}
+
+void _insb(unsigned int port, void * addr, unsigned long count)
+{
+	if (port >= LAN_IOSTART && port < LAN_IOEND)
+		_ne_insb(PORT2ADDR_NE(port), addr, count);
+#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
+	else if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
+		unsigned char *buf = addr;
+		unsigned char *portp = __port2addr_ata(port);
+		while (count--)
+			*buf++ = *(volatile unsigned char *)portp;
+	}
+#endif
+#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
+	else if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
+		pcc_ioread_byte(0, port, (void *)addr, sizeof(unsigned char),
+				count, 1);
+	}
+#endif
+	else {
+		unsigned char *buf = addr;
+		unsigned char *portp = PORT2ADDR(port);
+		while (count--)
+			*buf++ = *(volatile unsigned char *)portp;
+	}
+}
+
+void _insw(unsigned int port, void * addr, unsigned long count)
+{
+	unsigned short *buf = addr;
+	unsigned short *portp;
+
+	if (port >= LAN_IOSTART && port < LAN_IOEND) {
+		portp = PORT2ADDR_NE(port);
+		while (count--)
+			*buf++ = *(volatile unsigned short *)portp;
+#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
+	} else if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
+		pcc_ioread_word(9, port, (void *)addr, sizeof(unsigned short),
+				count, 1);
+#endif
+#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
+	} else if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
+		portp = __port2addr_ata(port);
+		while (count--)
+			*buf++ = *(volatile unsigned short *)portp;
+#endif
+	} else {
+		portp = PORT2ADDR(port);
+		while (count--)
+			*buf++ = *(volatile unsigned short *)portp;
+	}
+}
+
+void _insl(unsigned int port, void * addr, unsigned long count)
+{
+	unsigned long *buf = addr;
+	unsigned long *portp;
+
+	portp = PORT2ADDR(port);
+	while (count--)
+		*buf++ = *(volatile unsigned long *)portp;
+}
+
+void _outsb(unsigned int port, const void * addr, unsigned long count)
+{
+	const unsigned char *buf = addr;
+	unsigned char *portp;
+
+	if (port >= LAN_IOSTART && port < LAN_IOEND) {
+		portp = PORT2ADDR_NE(port);
+		while (count--)
+			_ne_outb(*buf++, portp);
+#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
+	} else if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
+		portp = __port2addr_ata(port);
+		while (count--)
+			*(volatile unsigned char *)portp = *buf++;
+#endif
+#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
+	} else if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
+		pcc_iowrite_byte(0, port, (void *)addr, sizeof(unsigned char),
+				 count, 1);
+#endif
+	} else {
+		portp = PORT2ADDR(port);
+		while (count--)
+			*(volatile unsigned char *)portp = *buf++;
+	}
+}
+
+void _outsw(unsigned int port, const void * addr, unsigned long count)
+{
+	const unsigned short *buf = addr;
+	unsigned short *portp;
+
+	if (port >= LAN_IOSTART && port < LAN_IOEND) {
+		portp = PORT2ADDR_NE(port);
+		while (count--)
+			*(volatile unsigned short *)portp = *buf++;
+#if defined(CONFIG_IDE) && !defined(CONFIG_M32R_CFC)
+	} else if ((port >= 0x1f0 && port <=0x1f7) || port == 0x3f6) {
+		portp = __port2addr_ata(port);
+		while (count--)
+			*(volatile unsigned short *)portp = *buf++;
+#endif
+#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
+	} else if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
+		pcc_iowrite_word(9, port, (void *)addr, sizeof(unsigned short),
+				 count, 1);
+#endif
+	} else {
+		portp = PORT2ADDR(port);
+		while (count--)
+			*(volatile unsigned short *)portp = *buf++;
+	}
+}
+
+void _outsl(unsigned int port, const void * addr, unsigned long count)
+{
+	const unsigned long *buf = addr;
+	unsigned char *portp;
+
+	portp = PORT2ADDR(port);
+	while (count--)
+		*(volatile unsigned long *)portp = *buf++;
+}
diff --git a/arch/m32r/kernel/io_oaks32r.c b/arch/m32r/kernel/io_oaks32r.c
index 286964794d5..9997dddd24d 100644
--- a/arch/m32r/kernel/io_oaks32r.c
+++ b/arch/m32r/kernel/io_oaks32r.c
@@ -3,7 +3,7 @@
  *
  *  Typical I/O routines for OAKS32R board.
  *
- *  Copyright (c) 2001-2004  Hiroyuki Kondo, Hirokazu Takata,
+ *  Copyright (c) 2001-2005  Hiroyuki Kondo, Hirokazu Takata,
  *                           Hitoshi Yamamoto, Mamoru Sakugawa
  */
 
@@ -90,35 +90,21 @@ unsigned long _inl(unsigned long port)
 
 unsigned char _inb_p(unsigned long port)
 {
-	unsigned char  v;
-
-	if (port >= 0x300 && port < 0x320)
-		v = _ne_inb(PORT2ADDR_NE(port));
-	else
-		v = *(volatile unsigned char *)PORT2ADDR(port);
-
+	unsigned char v = _inb(port);
 	delay();
 	return (v);
 }
 
 unsigned short _inw_p(unsigned long port)
 {
-	unsigned short  v;
-
-	if (port >= 0x300 && port < 0x320)
-		v = _ne_inw(PORT2ADDR_NE(port));
-	else
-		v = *(volatile unsigned short *)PORT2ADDR(port);
-
+	unsigned short v = _inw(port);
 	delay();
 	return (v);
 }
 
 unsigned long _inl_p(unsigned long port)
 {
-	unsigned long  v;
-
-	v = *(volatile unsigned long *)PORT2ADDR(port);
+	unsigned long v = _inl(port);
 	delay();
 	return (v);
 }
@@ -146,27 +132,19 @@ void _outl(unsigned long l, unsigned long port)
 
 void _outb_p(unsigned char b, unsigned long port)
 {
-	if (port >= 0x300 && port < 0x320)
-		_ne_outb(b, PORT2ADDR_NE(port));
-	else
-		*(volatile unsigned char *)PORT2ADDR(port) = b;
-
+	_outb(b, port);
 	delay();
 }
 
 void _outw_p(unsigned short w, unsigned long port)
 {
-	if (port >= 0x300 && port < 0x320)
-		_ne_outw(w, PORT2ADDR_NE(port));
-	else
-		*(volatile unsigned short *)PORT2ADDR(port) = w;
-
+	_outw(w, port);
 	delay();
 }
 
 void _outl_p(unsigned long l, unsigned long port)
 {
-	*(volatile unsigned long *)PORT2ADDR(port) = l;
+	_outl(l, port);
 	delay();
 }
 
diff --git a/arch/m32r/kernel/io_opsput.c b/arch/m32r/kernel/io_opsput.c
index aaf42f9f76d..e34951e8156 100644
--- a/arch/m32r/kernel/io_opsput.c
+++ b/arch/m32r/kernel/io_opsput.c
@@ -1,10 +1,10 @@
 /*
- *  linux/arch/m32r/kernel/io_mappi.c
+ *  linux/arch/m32r/kernel/io_opsput.c
  *
  *  Typical I/O routines for OPSPUT board.
  *
- *  Copyright (c) 2001, 2002  Hiroyuki Kondo, Hirokazu Takata,
- *                            Hitoshi Yamamoto, Takeo Takahashi
+ *  Copyright (c) 2001-2005  Hiroyuki Kondo, Hirokazu Takata,
+ *                           Hitoshi Yamamoto, Takeo Takahashi
  *
  *  This file is subject to the terms and conditions of the GNU General
  *  Public License.  See the file "COPYING" in the main directory of this
@@ -98,7 +98,6 @@ unsigned char _inb(unsigned long port)
 {
 	if (port >= LAN_IOSTART && port < LAN_IOEND)
 		return _ne_inb(PORT2ADDR_NE(port));
-
 #if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
 	else if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
 		unsigned char b;
@@ -118,7 +117,6 @@ unsigned short _inw(unsigned long port)
 	else if(port >= 0x340 && port < 0x3a0)
 		return *(volatile unsigned short *)PORT2ADDR_USB(port);
 #endif
-
 #if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
 	else if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
 		unsigned short w;
@@ -143,55 +141,21 @@ unsigned long _inl(unsigned long port)
 
 unsigned char _inb_p(unsigned long port)
 {
-	unsigned char  v;
-
-	if (port >= LAN_IOSTART && port < LAN_IOEND)
-		v = _ne_inb(PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		unsigned char b;
-		pcc_ioread_byte(0, port, &b, sizeof(b), 1, 0);
-		return b;
-	} else
-#endif
-		v = *(volatile unsigned char *)PORT2ADDR(port);
-
+	unsigned char v = _inb(port);
 	delay();
 	return (v);
 }
 
 unsigned short _inw_p(unsigned long port)
 {
-	unsigned short  v;
-
-	if (port >= LAN_IOSTART && port < LAN_IOEND)
-		v = _ne_inw(PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_USB)
-	if(port >= 0x340 && port < 0x3a0)
-		return *(volatile unsigned short *)PORT2ADDR_USB(port);
-	else
-#endif
-
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		unsigned short w;
-		pcc_ioread_word(0, port, &w, sizeof(w), 1, 0);
-		return w;
-	} else
-#endif
-		v = *(volatile unsigned short *)PORT2ADDR(port);
-
+	unsigned short v = _inw(port);
 	delay();
 	return (v);
 }
 
 unsigned long _inl_p(unsigned long port)
 {
-	unsigned long  v;
-
-	v = *(volatile unsigned long *)PORT2ADDR(port);
+	unsigned long v = _inl(port);
 	delay();
 	return (v);
 }
@@ -219,7 +183,6 @@ void _outw(unsigned short w, unsigned long port)
 		*(volatile unsigned short *)PORT2ADDR_USB(port) = w;
 	else
 #endif
-
 #if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
 	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
 		pcc_iowrite_word(0, port, &w, sizeof(w), 1, 0);
@@ -240,43 +203,19 @@ void _outl(unsigned long l, unsigned long port)
 
 void _outb_p(unsigned char b, unsigned long port)
 {
-	if (port >= LAN_IOSTART && port < LAN_IOEND)
-		_ne_outb(b, PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		pcc_iowrite_byte(0, port, &b, sizeof(b), 1, 0);
-	} else
-#endif
-		*(volatile unsigned char *)PORT2ADDR(port) = b;
-
+	_outb(b, port);
 	delay();
 }
 
 void _outw_p(unsigned short w, unsigned long port)
 {
-	if (port >= LAN_IOSTART && port < LAN_IOEND)
-		_ne_outw(w, PORT2ADDR_NE(port));
-	else
-#if defined(CONFIG_USB)
-	if(port >= 0x340 && port < 0x3a0)
-		*(volatile unsigned short *)PORT2ADDR_USB(port) = w;
-	else
-#endif
-
-#if defined(CONFIG_PCMCIA) && defined(CONFIG_M32R_CFC)
-	if (port >= M32R_PCC_IOSTART0 && port <= M32R_PCC_IOEND0) {
-		pcc_iowrite_word(0, port, &w, sizeof(w), 1, 0);
-	} else
-#endif
-		*(volatile unsigned short *)PORT2ADDR(port) = w;
-
+	_outw(w, port);
 	delay();
 }
 
 void _outl_p(unsigned long l, unsigned long port)
 {
-	*(volatile unsigned long *)PORT2ADDR(port) = l;
+	_outl(l, port);
 	delay();
 }
 
diff --git a/arch/m32r/kernel/io_usrv.c b/arch/m32r/kernel/io_usrv.c
index 27928a0b99e..9eb161dcc10 100644
--- a/arch/m32r/kernel/io_usrv.c
+++ b/arch/m32r/kernel/io_usrv.c
@@ -3,8 +3,8 @@
  *
  *  Typical I/O routines for uServer board.
  *
- *  Copyright (c) 2001 - 2003  Hiroyuki Kondo, Hirokazu Takata,
- *                             Hitoshi Yamamoto, Takeo Takahashi
+ *  Copyright (c) 2001-2005  Hiroyuki Kondo, Hirokazu Takata,
+ *                           Hitoshi Yamamoto, Takeo Takahashi
  *
  *  This file is subject to the terms and conditions of the GNU General
  *  Public License.  See the file "COPYING" in the main directory of this
@@ -39,7 +39,7 @@ extern void pcc_iowrite_word(int, unsigned long, void *, size_t, size_t, int);
 
 #define PORT2ADDR(port)	_port2addr(port)
 
-static __inline__ void *_port2addr(unsigned long port)
+static inline void *_port2addr(unsigned long port)
 {
 #if defined(CONFIG_SERIAL_8250) || defined(CONFIG_SERIAL_8250_MODULE)
 	if (port >= UART0_IOSTART && port <= UART0_IOEND)
@@ -50,7 +50,7 @@ static __inline__ void *_port2addr(unsigned long port)
 	return (void *)(port + NONCACHE_OFFSET);
 }
 
-static __inline__ void delay(void)
+static inline void delay(void)
 {
 	__asm__ __volatile__ ("push r0; \n\t pop r0;" : : :"memory");
 }
@@ -87,39 +87,22 @@ unsigned long _inl(unsigned long port)
 
 unsigned char _inb_p(unsigned long port)
 {
-	unsigned char b;
-
-	if (port >= CFC_IOSTART && port <= CFC_IOEND) {
-		pcc_ioread_byte(0, port, &b, sizeof(b), 1, 0);
-		return b;
-	} else {
-		b = *(volatile unsigned char *)PORT2ADDR(port);
-		delay();
-		return b;
-	}
+	unsigned char v = _inb(port);
+	delay();
+	return v;
 }
 
 unsigned short _inw_p(unsigned long port)
 {
-	unsigned short w;
-
-	if (port >= CFC_IOSTART && port <= CFC_IOEND) {
-		pcc_ioread_word(0, port, &w, sizeof(w), 1, 0);
-		return w;
-	} else {
-		w = *(volatile unsigned short *)PORT2ADDR(port);
-		delay();
-		return w;
-	}
+	unsigned short v = _inw(port);
+	delay();
+	return v;
 }
 
 unsigned long _inl_p(unsigned long port)
 {
-	unsigned long v;
-
-	v = *(volatile unsigned long *)PORT2ADDR(port);
+	unsigned long v = _inl(port);
 	delay();
-
 	return v;
 }
 
@@ -149,25 +132,19 @@ void _outl(unsigned long l, unsigned long port)
 
 void _outb_p(unsigned char b, unsigned long port)
 {
-	if (port >= CFC_IOSTART && port <= CFC_IOEND)
-		pcc_iowrite_byte(0, port, &b, sizeof(b), 1, 0);
-	else
-		*(volatile unsigned char *)PORT2ADDR(port) = b;
+	_outb(b, port);
 	delay();
 }
 
 void _outw_p(unsigned short w, unsigned long port)
 {
-	if (port >= CFC_IOSTART && port <= CFC_IOEND)
-		pcc_iowrite_word(0, port, &w, sizeof(w), 1, 0);
-	else
-		*(volatile unsigned short *)PORT2ADDR(port) = w;
+	_outw(w, port);
 	delay();
 }
 
 void _outl_p(unsigned long l, unsigned long port)
 {
-	*(volatile unsigned long *)PORT2ADDR(port) = l;
+	_outl(l, port);
 	delay();
 }
 
diff --git a/arch/m32r/kernel/setup.c b/arch/m32r/kernel/setup.c
index 4826cd6e40e..ec5674727e7 100644
--- a/arch/m32r/kernel/setup.c
+++ b/arch/m32r/kernel/setup.c
@@ -330,6 +330,8 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 	seq_printf(m, "Machine\t\t: Mappi Evaluation board\n");
 #elif CONFIG_PLAT_MAPPI2
 	seq_printf(m, "Machine\t\t: Mappi-II Evaluation board\n");
+#elif CONFIG_PLAT_MAPPI3
+	seq_printf(m, "Machine\t\t: Mappi-III Evaluation board\n");
 #elif  CONFIG_PLAT_M32700UT
 	seq_printf(m, "Machine\t\t: M32700UT Evaluation board\n");
 #elif  CONFIG_PLAT_OPSPUT
diff --git a/arch/m32r/kernel/setup_mappi2.c b/arch/m32r/kernel/setup_mappi2.c
index 1904d465a50..248fc658dbd 100644
--- a/arch/m32r/kernel/setup_mappi2.c
+++ b/arch/m32r/kernel/setup_mappi2.c
@@ -1,5 +1,5 @@
 /*
- *  linux/arch/m32r/kernel/setup_mappi.c
+ *  linux/arch/m32r/kernel/setup_mappi2.c
  *
  *  Setup routines for Renesas MAPPI-II(M3A-ZA36) Board
  *
@@ -156,7 +156,6 @@ void __init init_IRQ(void)
 	irq_desc[PLD_IRQ_CFIREQ].handler = &mappi2_irq_type;
 	irq_desc[PLD_IRQ_CFIREQ].action = 0;
 	irq_desc[PLD_IRQ_CFIREQ].depth = 1;	/* disable nested irq */
-//	icu_data[PLD_IRQ_CFIREQ].icucr = M32R_ICUCR_IEN|M32R_ICUCR_ISMOD00;
 	icu_data[PLD_IRQ_CFIREQ].icucr = M32R_ICUCR_IEN|M32R_ICUCR_ISMOD01;
 	disable_mappi2_irq(PLD_IRQ_CFIREQ);
 
@@ -167,7 +166,6 @@ void __init init_IRQ(void)
 	irq_desc[PLD_IRQ_CFC_INSERT].action = 0;
 	irq_desc[PLD_IRQ_CFC_INSERT].depth = 1;	/* disable nested irq */
 	icu_data[PLD_IRQ_CFC_INSERT].icucr = M32R_ICUCR_IEN|M32R_ICUCR_ISMOD00;
-//	icu_data[PLD_IRQ_CFC_INSERT].icucr = 0;
 	disable_mappi2_irq(PLD_IRQ_CFC_INSERT);
 
 	/* ICUCR42: CFC Eject */
@@ -176,9 +174,7 @@ void __init init_IRQ(void)
 	irq_desc[PLD_IRQ_CFC_EJECT].action = 0;
 	irq_desc[PLD_IRQ_CFC_EJECT].depth = 1;	/* disable nested irq */
 	icu_data[PLD_IRQ_CFC_EJECT].icucr = M32R_ICUCR_IEN|M32R_ICUCR_ISMOD10;
-//	icu_data[PLD_IRQ_CFC_EJECT].icucr = 0;
 	disable_mappi2_irq(PLD_IRQ_CFC_EJECT);
-
 #endif /* CONFIG_MAPPI2_CFC */
 }
 
diff --git a/arch/m32r/kernel/setup_mappi3.c b/arch/m32r/kernel/setup_mappi3.c
new file mode 100644
index 00000000000..72b5b7609ce
--- /dev/null
+++ b/arch/m32r/kernel/setup_mappi3.c
@@ -0,0 +1,208 @@
+/*
+ *  linux/arch/m32r/kernel/setup_mappi3.c
+ *
+ *  Setup routines for Renesas MAPPI-III(M3A-2170) Board
+ *
+ *  Copyright (c) 2001-2005   Hiroyuki Kondo, Hirokazu Takata,
+ *                            Hitoshi Yamamoto, Mamoru Sakugawa
+ */
+
+#include <linux/config.h>
+#include <linux/irq.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/device.h>
+
+#include <asm/system.h>
+#include <asm/m32r.h>
+#include <asm/io.h>
+
+#define irq2port(x) (M32R_ICU_CR1_PORTL + ((x - 1) * sizeof(unsigned long)))
+
+#ifndef CONFIG_SMP
+typedef struct {
+	unsigned long icucr;  /* ICU Control Register */
+} icu_data_t;
+#endif /* CONFIG_SMP */
+
+icu_data_t icu_data[NR_IRQS];
+
+static void disable_mappi3_irq(unsigned int irq)
+{
+	unsigned long port, data;
+
+	if ((irq == 0) ||(irq >= NR_IRQS))  {
+		printk("bad irq 0x%08x\n", irq);
+		return;
+	}
+	port = irq2port(irq);
+	data = icu_data[irq].icucr|M32R_ICUCR_ILEVEL7;
+	outl(data, port);
+}
+
+static void enable_mappi3_irq(unsigned int irq)
+{
+	unsigned long port, data;
+
+	if ((irq == 0) ||(irq >= NR_IRQS))  {
+		printk("bad irq 0x%08x\n", irq);
+		return;
+	}
+	port = irq2port(irq);
+	data = icu_data[irq].icucr|M32R_ICUCR_IEN|M32R_ICUCR_ILEVEL6;
+	outl(data, port);
+}
+
+static void mask_and_ack_mappi3(unsigned int irq)
+{
+	disable_mappi3_irq(irq);
+}
+
+static void end_mappi3_irq(unsigned int irq)
+{
+	enable_mappi3_irq(irq);
+}
+
+static unsigned int startup_mappi3_irq(unsigned int irq)
+{
+	enable_mappi3_irq(irq);
+	return (0);
+}
+
+static void shutdown_mappi3_irq(unsigned int irq)
+{
+	unsigned long port;
+
+	port = irq2port(irq);
+	outl(M32R_ICUCR_ILEVEL7, port);
+}
+
+static struct hw_interrupt_type mappi3_irq_type =
+{
+	"MAPPI3-IRQ",
+	startup_mappi3_irq,
+	shutdown_mappi3_irq,
+	enable_mappi3_irq,
+	disable_mappi3_irq,
+	mask_and_ack_mappi3,
+	end_mappi3_irq
+};
+
+void __init init_IRQ(void)
+{
+#if defined(CONFIG_SMC91X)
+	/* INT0 : LAN controller (SMC91111) */
+	irq_desc[M32R_IRQ_INT0].status = IRQ_DISABLED;
+	irq_desc[M32R_IRQ_INT0].handler = &mappi3_irq_type;
+	irq_desc[M32R_IRQ_INT0].action = 0;
+	irq_desc[M32R_IRQ_INT0].depth = 1;
+	icu_data[M32R_IRQ_INT0].icucr = M32R_ICUCR_IEN|M32R_ICUCR_ISMOD10;
+	disable_mappi3_irq(M32R_IRQ_INT0);
+#endif  /* CONFIG_SMC91X */
+
+	/* MFT2 : system timer */
+	irq_desc[M32R_IRQ_MFT2].status = IRQ_DISABLED;
+	irq_desc[M32R_IRQ_MFT2].handler = &mappi3_irq_type;
+	irq_desc[M32R_IRQ_MFT2].action = 0;
+	irq_desc[M32R_IRQ_MFT2].depth = 1;
+	icu_data[M32R_IRQ_MFT2].icucr = M32R_ICUCR_IEN;
+	disable_mappi3_irq(M32R_IRQ_MFT2);
+
+#ifdef CONFIG_SERIAL_M32R_SIO
+	/* SIO0_R : uart receive data */
+	irq_desc[M32R_IRQ_SIO0_R].status = IRQ_DISABLED;
+	irq_desc[M32R_IRQ_SIO0_R].handler = &mappi3_irq_type;
+	irq_desc[M32R_IRQ_SIO0_R].action = 0;
+	irq_desc[M32R_IRQ_SIO0_R].depth = 1;
+	icu_data[M32R_IRQ_SIO0_R].icucr = 0;
+	disable_mappi3_irq(M32R_IRQ_SIO0_R);
+
+	/* SIO0_S : uart send data */
+	irq_desc[M32R_IRQ_SIO0_S].status = IRQ_DISABLED;
+	irq_desc[M32R_IRQ_SIO0_S].handler = &mappi3_irq_type;
+	irq_desc[M32R_IRQ_SIO0_S].action = 0;
+	irq_desc[M32R_IRQ_SIO0_S].depth = 1;
+	icu_data[M32R_IRQ_SIO0_S].icucr = 0;
+	disable_mappi3_irq(M32R_IRQ_SIO0_S);
+	/* SIO1_R : uart receive data */
+	irq_desc[M32R_IRQ_SIO1_R].status = IRQ_DISABLED;
+	irq_desc[M32R_IRQ_SIO1_R].handler = &mappi3_irq_type;
+	irq_desc[M32R_IRQ_SIO1_R].action = 0;
+	irq_desc[M32R_IRQ_SIO1_R].depth = 1;
+	icu_data[M32R_IRQ_SIO1_R].icucr = 0;
+	disable_mappi3_irq(M32R_IRQ_SIO1_R);
+
+	/* SIO1_S : uart send data */
+	irq_desc[M32R_IRQ_SIO1_S].status = IRQ_DISABLED;
+	irq_desc[M32R_IRQ_SIO1_S].handler = &mappi3_irq_type;
+	irq_desc[M32R_IRQ_SIO1_S].action = 0;
+	irq_desc[M32R_IRQ_SIO1_S].depth = 1;
+	icu_data[M32R_IRQ_SIO1_S].icucr = 0;
+	disable_mappi3_irq(M32R_IRQ_SIO1_S);
+#endif  /* CONFIG_M32R_USE_DBG_CONSOLE */
+
+#if defined(CONFIG_USB)
+	/* INT1 : USB Host controller interrupt */
+	irq_desc[M32R_IRQ_INT1].status = IRQ_DISABLED;
+	irq_desc[M32R_IRQ_INT1].handler = &mappi3_irq_type;
+	irq_desc[M32R_IRQ_INT1].action = 0;
+	irq_desc[M32R_IRQ_INT1].depth = 1;
+	icu_data[M32R_IRQ_INT1].icucr = M32R_ICUCR_ISMOD01;
+	disable_mappi3_irq(M32R_IRQ_INT1);
+#endif /* CONFIG_USB */
+
+	/* ICUCR40: CFC IREQ */
+	irq_desc[PLD_IRQ_CFIREQ].status = IRQ_DISABLED;
+	irq_desc[PLD_IRQ_CFIREQ].handler = &mappi3_irq_type;
+	irq_desc[PLD_IRQ_CFIREQ].action = 0;
+	irq_desc[PLD_IRQ_CFIREQ].depth = 1;	/* disable nested irq */
+	icu_data[PLD_IRQ_CFIREQ].icucr = M32R_ICUCR_IEN|M32R_ICUCR_ISMOD01;
+	disable_mappi3_irq(PLD_IRQ_CFIREQ);
+
+#if defined(CONFIG_M32R_CFC)
+	/* ICUCR41: CFC Insert */
+	irq_desc[PLD_IRQ_CFC_INSERT].status = IRQ_DISABLED;
+	irq_desc[PLD_IRQ_CFC_INSERT].handler = &mappi3_irq_type;
+	irq_desc[PLD_IRQ_CFC_INSERT].action = 0;
+	irq_desc[PLD_IRQ_CFC_INSERT].depth = 1;	/* disable nested irq */
+	icu_data[PLD_IRQ_CFC_INSERT].icucr = M32R_ICUCR_IEN|M32R_ICUCR_ISMOD00;
+	disable_mappi3_irq(PLD_IRQ_CFC_INSERT);
+
+	/* ICUCR42: CFC Eject */
+	irq_desc[PLD_IRQ_CFC_EJECT].status = IRQ_DISABLED;
+	irq_desc[PLD_IRQ_CFC_EJECT].handler = &mappi3_irq_type;
+	irq_desc[PLD_IRQ_CFC_EJECT].action = 0;
+	irq_desc[PLD_IRQ_CFC_EJECT].depth = 1;	/* disable nested irq */
+	icu_data[PLD_IRQ_CFC_EJECT].icucr = M32R_ICUCR_IEN|M32R_ICUCR_ISMOD10;
+	disable_mappi3_irq(PLD_IRQ_CFC_EJECT);
+#endif /* CONFIG_M32R_CFC */
+}
+
+#define LAN_IOSTART     0x300
+#define LAN_IOEND       0x320
+static struct resource smc91x_resources[] = {
+	[0] = {
+		.start  = (LAN_IOSTART),
+		.end    = (LAN_IOEND),
+		.flags  = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start  = M32R_IRQ_INT0,
+		.end    = M32R_IRQ_INT0,
+		.flags  = IORESOURCE_IRQ,
+	}
+};
+
+static struct platform_device smc91x_device = {
+	.name		= "smc91x",
+	.id		= 0,
+	.num_resources  = ARRAY_SIZE(smc91x_resources),
+	.resource       = smc91x_resources,
+};
+
+static int __init platform_init(void)
+{
+	platform_device_register(&smc91x_device);
+	return 0;
+}
+arch_initcall(platform_init);
diff --git a/arch/m32r/mappi3/defconfig.smp b/arch/m32r/mappi3/defconfig.smp
new file mode 100644
index 00000000000..2eebe75fc07
--- /dev/null
+++ b/arch/m32r/mappi3/defconfig.smp
@@ -0,0 +1,751 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.12-rc5
+# Tue May 31 17:55:34 2005
+#
+CONFIG_M32R=y
+# CONFIG_UID16 is not set
+CONFIG_GENERIC_ISA_DMA=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+# CONFIG_CLEAN_COMPILE is not set
+CONFIG_BROKEN=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+CONFIG_SWAP=y
+CONFIG_SYSVIPC=y
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
+CONFIG_HOTPLUG=y
+CONFIG_KOBJECT_UEVENT=y
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+# CONFIG_CPUSETS is not set
+CONFIG_EMBEDDED=y
+# CONFIG_KALLSYMS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_BASE_FULL=y
+# CONFIG_FUTEX is not set
+# CONFIG_EPOLL is not set
+# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
+CONFIG_SHMEM=y
+CONFIG_CC_ALIGN_FUNCTIONS=0
+CONFIG_CC_ALIGN_LABELS=0
+CONFIG_CC_ALIGN_LOOPS=0
+CONFIG_CC_ALIGN_JUMPS=0
+# CONFIG_TINY_SHMEM is not set
+CONFIG_BASE_SMALL=0
+
+#
+# Loadable module support
+#
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_MODULE_FORCE_UNLOAD is not set
+CONFIG_OBSOLETE_MODPARM=y
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+CONFIG_KMOD=y
+CONFIG_STOP_MACHINE=y
+
+#
+# Processor type and features
+#
+# CONFIG_PLAT_MAPPI is not set
+# CONFIG_PLAT_USRV is not set
+# CONFIG_PLAT_M32700UT is not set
+# CONFIG_PLAT_OPSPUT is not set
+# CONFIG_PLAT_OAKS32R is not set
+# CONFIG_PLAT_MAPPI2 is not set
+CONFIG_PLAT_MAPPI3=y
+CONFIG_CHIP_M32700=y
+# CONFIG_CHIP_M32102 is not set
+# CONFIG_CHIP_VDEC2 is not set
+# CONFIG_CHIP_OPSP is not set
+CONFIG_MMU=y
+CONFIG_TLB_ENTRIES=32
+CONFIG_ISA_M32R2=y
+CONFIG_ISA_DSP_LEVEL2=y
+CONFIG_ISA_DUAL_ISSUE=y
+CONFIG_BUS_CLOCK=10000000
+CONFIG_TIMER_DIVIDE=128
+# CONFIG_CPU_LITTLE_ENDIAN is not set
+CONFIG_MEMORY_START=0x08000000
+CONFIG_MEMORY_SIZE=0x08000000
+CONFIG_NOHIGHMEM=y
+CONFIG_DISCONTIGMEM=y
+CONFIG_IRAM_START=0x00f00000
+CONFIG_IRAM_SIZE=0x00080000
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_PREEMPT=y
+# CONFIG_HAVE_DEC_LOCK is not set
+CONFIG_SMP=y
+# CONFIG_CHIP_M32700_TS1 is not set
+CONFIG_NR_CPUS=2
+# CONFIG_NUMA is not set
+
+#
+# Bus options (PCI, PCMCIA, EISA, MCA, ISA)
+#
+# CONFIG_PCI is not set
+# CONFIG_ISA is not set
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+CONFIG_PCCARD=y
+# CONFIG_PCMCIA_DEBUG is not set
+CONFIG_PCMCIA=y
+
+#
+# PC-card bridges
+#
+# CONFIG_TCIC is not set
+# CONFIG_M32R_PCC is not set
+# CONFIG_M32R_CFC is not set
+
+#
+# PCI Hotplug Support
+#
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+CONFIG_FW_LOADER=y
+
+#
+# Memory Technology Devices (MTD)
+#
+CONFIG_MTD=y
+# CONFIG_MTD_DEBUG is not set
+# CONFIG_MTD_CONCAT is not set
+CONFIG_MTD_PARTITIONS=y
+CONFIG_MTD_REDBOOT_PARTS=y
+CONFIG_MTD_REDBOOT_DIRECTORY_BLOCK=-1
+# CONFIG_MTD_REDBOOT_PARTS_UNALLOCATED is not set
+# CONFIG_MTD_REDBOOT_PARTS_READONLY is not set
+# CONFIG_MTD_CMDLINE_PARTS is not set
+
+#
+# User Modules And Translation Layers
+#
+CONFIG_MTD_CHAR=y
+CONFIG_MTD_BLOCK=y
+# CONFIG_FTL is not set
+# CONFIG_NFTL is not set
+# CONFIG_INFTL is not set
+
+#
+# RAM/ROM/Flash chip drivers
+#
+# CONFIG_MTD_CFI is not set
+# CONFIG_MTD_JEDECPROBE is not set
+CONFIG_MTD_MAP_BANK_WIDTH_1=y
+CONFIG_MTD_MAP_BANK_WIDTH_2=y
+CONFIG_MTD_MAP_BANK_WIDTH_4=y
+# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
+CONFIG_MTD_CFI_I1=y
+CONFIG_MTD_CFI_I2=y
+# CONFIG_MTD_CFI_I4 is not set
+# CONFIG_MTD_CFI_I8 is not set
+# CONFIG_MTD_RAM is not set
+# CONFIG_MTD_ROM is not set
+# CONFIG_MTD_ABSENT is not set
+# CONFIG_MTD_OBSOLETE_CHIPS is not set
+
+#
+# Mapping drivers for chip access
+#
+# CONFIG_MTD_COMPLEX_MAPPINGS is not set
+
+#
+# Self-contained MTD device drivers
+#
+# CONFIG_MTD_SLRAM is not set
+# CONFIG_MTD_PHRAM is not set
+# CONFIG_MTD_MTDRAM is not set
+# CONFIG_MTD_BLKMTD is not set
+# CONFIG_MTD_BLOCK2MTD is not set
+
+#
+# Disk-On-Chip Device Drivers
+#
+# CONFIG_MTD_DOC2000 is not set
+# CONFIG_MTD_DOC2001 is not set
+# CONFIG_MTD_DOC2001PLUS is not set
+
+#
+# NAND Flash Device Drivers
+#
+# CONFIG_MTD_NAND is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+
+#
+# Block devices
+#
+# CONFIG_BLK_DEV_FD is not set
+# CONFIG_BLK_DEV_COW_COMMON is not set
+CONFIG_BLK_DEV_LOOP=y
+# CONFIG_BLK_DEV_CRYPTOLOOP is not set
+CONFIG_BLK_DEV_NBD=m
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+CONFIG_BLK_DEV_INITRD=y
+CONFIG_INITRAMFS_SOURCE=""
+# CONFIG_CDROM_PKTCDVD is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+CONFIG_IOSCHED_DEADLINE=y
+CONFIG_IOSCHED_CFQ=y
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+CONFIG_IDE=y
+CONFIG_BLK_DEV_IDE=y
+
+#
+# Please see Documentation/ide.txt for help/info on IDE drives
+#
+# CONFIG_BLK_DEV_IDE_SATA is not set
+CONFIG_BLK_DEV_IDEDISK=y
+# CONFIG_IDEDISK_MULTI_MODE is not set
+CONFIG_BLK_DEV_IDECS=y
+CONFIG_BLK_DEV_IDECD=m
+# CONFIG_BLK_DEV_IDETAPE is not set
+# CONFIG_BLK_DEV_IDEFLOPPY is not set
+# CONFIG_IDE_TASK_IOCTL is not set
+
+#
+# IDE chipset support/bugfixes
+#
+CONFIG_IDE_GENERIC=y
+# CONFIG_IDE_ARM is not set
+# CONFIG_BLK_DEV_IDEDMA is not set
+# CONFIG_IDEDMA_AUTO is not set
+# CONFIG_BLK_DEV_HD is not set
+
+#
+# SCSI device support
+#
+# CONFIG_SCSI is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+# CONFIG_MD is not set
+
+#
+# Fusion MPT device support
+#
+
+#
+# IEEE 1394 (FireWire) support
+#
+# CONFIG_IEEE1394 is not set
+
+#
+# I2O device support
+#
+
+#
+# Networking support
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+# CONFIG_PACKET is not set
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+# CONFIG_IP_PNP_BOOTP is not set
+# CONFIG_IP_PNP_RARP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_TUNNEL is not set
+CONFIG_IP_TCPDIAG=y
+# CONFIG_IP_TCPDIAG_IPV6 is not set
+# CONFIG_IPV6 is not set
+# CONFIG_NETFILTER is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+
+#
+# QoS and/or fair queueing
+#
+# CONFIG_NET_SCHED is not set
+# CONFIG_NET_CLS_ROUTE is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+CONFIG_NETDEVICES=y
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+CONFIG_SMC91X=y
+# CONFIG_NE2000 is not set
+
+#
+# Ethernet (1000 Mbit)
+#
+
+#
+# Ethernet (10000 Mbit)
+#
+
+#
+# Token Ring devices
+#
+
+#
+# Wireless LAN (non-hamradio)
+#
+# CONFIG_NET_RADIO is not set
+
+#
+# PCMCIA network device support
+#
+# CONFIG_NET_PCMCIA is not set
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+
+#
+# Userland interfaces
+#
+CONFIG_INPUT_MOUSEDEV=y
+CONFIG_INPUT_MOUSEDEV_PSAUX=y
+CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024
+CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_TSDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+# CONFIG_INPUT_KEYBOARD is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+CONFIG_SERIO=y
+# CONFIG_SERIO_I8042 is not set
+# CONFIG_SERIO_SERPORT is not set
+# CONFIG_SERIO_LIBPS2 is not set
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+CONFIG_SOUND_GAMEPORT=y
+
+#
+# Character devices
+#
+# CONFIG_VT is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+CONFIG_SERIAL_M32R_SIO=y
+CONFIG_SERIAL_M32R_SIO_CONSOLE=y
+# CONFIG_SERIAL_M32R_PLDSIO is not set
+CONFIG_UNIX98_PTYS=y
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+# CONFIG_WATCHDOG is not set
+# CONFIG_RTC is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_DRM is not set
+
+#
+# PCMCIA character devices
+#
+# CONFIG_SYNCLINK_CS is not set
+# CONFIG_RAW_DRIVER is not set
+
+#
+# TPM devices
+#
+
+#
+# I2C support
+#
+# CONFIG_I2C is not set
+
+#
+# Dallas's 1-wire bus
+#
+# CONFIG_W1 is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+# CONFIG_FB is not set
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+
+#
+# USB support
+#
+# CONFIG_USB_ARCH_HAS_HCD is not set
+# CONFIG_USB_ARCH_HAS_OHCI is not set
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# InfiniBand support
+#
+# CONFIG_INFINIBAND is not set
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+# CONFIG_EXT2_FS_XATTR is not set
+CONFIG_EXT3_FS=y
+CONFIG_EXT3_FS_XATTR=y
+# CONFIG_EXT3_FS_POSIX_ACL is not set
+# CONFIG_EXT3_FS_SECURITY is not set
+CONFIG_JBD=y
+# CONFIG_JBD_DEBUG is not set
+CONFIG_FS_MBCACHE=y
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+
+#
+# XFS support
+#
+# CONFIG_XFS_FS is not set
+# CONFIG_MINIX_FS is not set
+CONFIG_ROMFS_FS=y
+# CONFIG_QUOTA is not set
+CONFIG_DNOTIFY=y
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+CONFIG_ISO9660_FS=y
+# CONFIG_JOLIET is not set
+# CONFIG_ZISOFS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+CONFIG_FAT_FS=m
+CONFIG_MSDOS_FS=m
+CONFIG_VFAT_FS=m
+CONFIG_FAT_DEFAULT_CODEPAGE=437
+CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_KCORE=y
+CONFIG_SYSFS=y
+CONFIG_DEVFS_FS=y
+CONFIG_DEVFS_MOUNT=y
+# CONFIG_DEVFS_DEBUG is not set
+# CONFIG_DEVPTS_FS_XATTR is not set
+CONFIG_TMPFS=y
+# CONFIG_TMPFS_XATTR is not set
+# CONFIG_HUGETLBFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_RAMFS=y
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+CONFIG_JFFS_FS=y
+CONFIG_JFFS_FS_VERBOSE=0
+CONFIG_JFFS_PROC_FS=y
+CONFIG_JFFS2_FS=y
+CONFIG_JFFS2_FS_DEBUG=0
+# CONFIG_JFFS2_FS_NAND is not set
+# CONFIG_JFFS2_FS_NOR_ECC is not set
+# CONFIG_JFFS2_COMPRESSION_OPTIONS is not set
+CONFIG_JFFS2_ZLIB=y
+CONFIG_JFFS2_RTIME=y
+# CONFIG_JFFS2_RUBIN is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+
+#
+# Network File Systems
+#
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
+# CONFIG_NFSD is not set
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_SUNRPC=y
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+# CONFIG_SMB_FS is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+
+#
+# Native Language Support
+#
+CONFIG_NLS=y
+CONFIG_NLS_DEFAULT="iso8859-1"
+# CONFIG_NLS_CODEPAGE_437 is not set
+# CONFIG_NLS_CODEPAGE_737 is not set
+# CONFIG_NLS_CODEPAGE_775 is not set
+# CONFIG_NLS_CODEPAGE_850 is not set
+# CONFIG_NLS_CODEPAGE_852 is not set
+# CONFIG_NLS_CODEPAGE_855 is not set
+# CONFIG_NLS_CODEPAGE_857 is not set
+# CONFIG_NLS_CODEPAGE_860 is not set
+# CONFIG_NLS_CODEPAGE_861 is not set
+# CONFIG_NLS_CODEPAGE_862 is not set
+# CONFIG_NLS_CODEPAGE_863 is not set
+# CONFIG_NLS_CODEPAGE_864 is not set
+# CONFIG_NLS_CODEPAGE_865 is not set
+# CONFIG_NLS_CODEPAGE_866 is not set
+# CONFIG_NLS_CODEPAGE_869 is not set
+# CONFIG_NLS_CODEPAGE_936 is not set
+# CONFIG_NLS_CODEPAGE_950 is not set
+# CONFIG_NLS_CODEPAGE_932 is not set
+# CONFIG_NLS_CODEPAGE_949 is not set
+# CONFIG_NLS_CODEPAGE_874 is not set
+# CONFIG_NLS_ISO8859_8 is not set
+# CONFIG_NLS_CODEPAGE_1250 is not set
+# CONFIG_NLS_CODEPAGE_1251 is not set
+# CONFIG_NLS_ASCII is not set
+# CONFIG_NLS_ISO8859_1 is not set
+# CONFIG_NLS_ISO8859_2 is not set
+# CONFIG_NLS_ISO8859_3 is not set
+# CONFIG_NLS_ISO8859_4 is not set
+# CONFIG_NLS_ISO8859_5 is not set
+# CONFIG_NLS_ISO8859_6 is not set
+# CONFIG_NLS_ISO8859_7 is not set
+# CONFIG_NLS_ISO8859_9 is not set
+# CONFIG_NLS_ISO8859_13 is not set
+# CONFIG_NLS_ISO8859_14 is not set
+# CONFIG_NLS_ISO8859_15 is not set
+# CONFIG_NLS_KOI8_R is not set
+# CONFIG_NLS_KOI8_U is not set
+# CONFIG_NLS_UTF8 is not set
+
+#
+# Profiling support
+#
+# CONFIG_PROFILING is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+# CONFIG_DEBUG_KERNEL is not set
+CONFIG_LOG_BUF_SHIFT=15
+# CONFIG_DEBUG_BUGVERBOSE is not set
+# CONFIG_FRAME_POINTER is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+
+#
+# Cryptographic options
+#
+# CONFIG_CRYPTO is not set
+
+#
+# Hardware crypto devices
+#
+
+#
+# Library routines
+#
+# CONFIG_CRC_CCITT is not set
+CONFIG_CRC32=y
+# CONFIG_LIBCRC32C is not set
+CONFIG_ZLIB_INFLATE=y
+CONFIG_ZLIB_DEFLATE=y
diff --git a/arch/m32r/mappi3/dot.gdbinit b/arch/m32r/mappi3/dot.gdbinit
new file mode 100644
index 00000000000..89c22184e13
--- /dev/null
+++ b/arch/m32r/mappi3/dot.gdbinit
@@ -0,0 +1,224 @@
+# .gdbinit file
+# $Id: dot.gdbinit,v 1.1 2005/04/11 02:21:08 sakugawa Exp $
+
+# setting
+set width 0d70
+set radix 0d16
+use_debug_dma
+
+# Initialize SDRAM controller for Mappi
+define sdram_init
+  # SDIR0
+  set *(unsigned long *)0x00ef6008 = 0x00000182
+  # SDIR1
+  set *(unsigned long *)0x00ef600c = 0x00000001
+  # Initialize wait
+  shell sleep 0.1
+  # MOD
+  set *(unsigned long *)0x00ef602c = 0x00000020
+  set *(unsigned long *)0x00ef604c = 0x00000020
+  # TR
+  set *(unsigned long *)0x00ef6028 = 0x00051502
+  set *(unsigned long *)0x00ef6048 = 0x00051502
+  # ADR
+  set *(unsigned long *)0x00ef6020 = 0x08000004
+  set *(unsigned long *)0x00ef6040 = 0x0c000004
+  # AutoRef On
+  set *(unsigned long *)0x00ef6004 = 0x00010517
+  # Access enable
+  set *(unsigned long *)0x00ef6024 = 0x00000001
+  set *(unsigned long *)0x00ef6044 = 0x00000001
+end
+
+# Initialize LAN controller for Mappi
+define lanc_init
+  # Set BSEL4
+  #set *(unsigned long *)0x00ef5004 = 0x0fff330f
+  #set *(unsigned long *)0x00ef5004 = 0x01113301
+
+#  set *(unsigned long *)0x00ef5004 = 0x02011101
+#  set *(unsigned long *)0x00ef5004 = 0x04441104
+end
+
+define clock_init
+  set *(unsigned long *)0x00ef4010 = 2
+  set *(unsigned long *)0x00ef4014 = 2
+  set *(unsigned long *)0x00ef4020 = 3
+  set *(unsigned long *)0x00ef4024 = 3
+  set *(unsigned long *)0x00ef4004 = 0x7
+#  shell sleep 0.1
+#  set *(unsigned long *)0x00ef4004 = 0x5
+  shell sleep 0.1
+  set *(unsigned long *)0x00ef4008 = 0x0200
+end
+
+define port_init
+  set $sfrbase = 0x00ef0000
+  set *(unsigned short *)0x00ef1060 = 0x5555
+  set *(unsigned short *)0x00ef1062 = 0x5555
+  set *(unsigned short *)0x00ef1064 = 0x5555
+  set *(unsigned short *)0x00ef1066 = 0x5555
+  set *(unsigned short *)0x00ef1068 = 0x5555
+  set *(unsigned short *)0x00ef106a = 0x0000
+  set *(unsigned short *)0x00ef106e = 0x5555
+  set *(unsigned short *)0x00ef1070 = 0x5555
+end
+
+# MMU enable
+define mmu_enable
+  set $evb=0x88000000
+  set *(unsigned long *)0xffff0024=1
+end
+
+# MMU disable
+define mmu_disable
+  set $evb=0
+  set *(unsigned long *)0xffff0024=0
+end
+
+# Show TLB entries
+define show_tlb_entries
+  set $i = 0
+  set $addr = $arg0
+  while ($i < 0d16 )
+    set $tlb_tag = *(unsigned long*)$addr
+    set $tlb_data = *(unsigned long*)($addr + 4)
+    printf " [%2d] 0x%08lx : 0x%08lx - 0x%08lx\n", $i, $addr, $tlb_tag, $tlb_data
+    set $i = $i + 1
+    set $addr = $addr + 8
+  end
+end
+define itlb
+  set $itlb=0xfe000000
+  show_tlb_entries $itlb
+end
+define dtlb
+  set $dtlb=0xfe000800
+  show_tlb_entries $dtlb
+end
+
+# Cache ON
+define set_cache_type
+  set $mctype = (void*)0xfffffff8
+# chaos
+# set *(unsigned long *)($mctype) = 0x0000c000
+# m32102 i-cache only
+  set *(unsigned long *)($mctype) = 0x00008000
+# m32102 d-cache only
+#  set *(unsigned long *)($mctype) = 0x00004000
+end
+define cache_on
+  set $param = (void*)0x08001000
+  set *(unsigned long *)($param) = 0x60ff6102
+end
+
+
+# Show current task structure
+define show_current
+  set $current = $spi & 0xffffe000
+  printf "$current=0x%08lX\n",$current
+  print *(struct task_struct *)$current
+end
+
+# Show user assigned task structure
+define show_task
+  set $task = $arg0 & 0xffffe000
+  printf "$task=0x%08lX\n",$task
+  print *(struct task_struct *)$task
+end
+document show_task
+  Show user assigned task structure
+  arg0 : task structure address
+end
+
+# Show M32R registers
+define show_regs
+  printf " R0[0x%08lX]   R1[0x%08lX]   R2[0x%08lX]   R3[0x%08lX]\n",$r0,$r1,$r2,$r3
+  printf " R4[0x%08lX]   R5[0x%08lX]   R6[0x%08lX]   R7[0x%08lX]\n",$r4,$r5,$r6,$r7
+  printf " R8[0x%08lX]   R9[0x%08lX]  R10[0x%08lX]  R11[0x%08lX]\n",$r8,$r9,$r10,$r11
+  printf "R12[0x%08lX]   FP[0x%08lX]   LR[0x%08lX]   SP[0x%08lX]\n",$r12,$fp,$lr,$sp
+  printf "PSW[0x%08lX]  CBR[0x%08lX]  SPI[0x%08lX]  SPU[0x%08lX]\n",$psw,$cbr,$spi,$spu
+  printf "BPC[0x%08lX]   PC[0x%08lX] ACCL[0x%08lX] ACCH[0x%08lX]\n",$bpc,$pc,$accl,$acch
+  printf "EVB[0x%08lX]\n",$evb
+
+  set $mests = *(unsigned long *)0xffff000c
+  set $mdeva = *(unsigned long *)0xffff0010
+  printf "MESTS[0x%08lX] MDEVA[0x%08lX]\n",$mests,$mdeva
+end
+
+
+# Setup all
+define setup
+  clock_init
+  shell sleep 0.1
+  port_init
+  sdram_init
+#  lanc_init
+#  dispc_init
+#  set $evb=0x08000000
+end
+
+# Load modules
+define load_modules
+  use_debug_dma
+  load
+#  load busybox.mot
+end
+
+# Set kernel parameters
+define set_kernel_parameters
+  set $param = (void*)0x08001000
+
+  ## MOUNT_ROOT_RDONLY
+  set {long}($param+0x00)=0
+  ## RAMDISK_FLAGS
+  #set {long}($param+0x04)=0
+  ## ORIG_ROOT_DEV
+  #set {long}($param+0x08)=0x00000100
+  ## LOADER_TYPE
+  #set {long}($param+0x0C)=0
+  ## INITRD_START
+  set {long}($param+0x10)=0x082a0000
+  ## INITRD_SIZE
+  set {long}($param+0x14)=0d6200000
+
+  # M32R_CPUCLK
+  set *(unsigned long *)($param + 0x0018) = 0d100000000
+  # M32R_BUSCLK
+  set *(unsigned long *)($param + 0x001c) = 0d50000000
+  # M32R_TIMER_DIVIDE
+  set *(unsigned long *)($param + 0x0020) = 0d128
+
+
+ set {char[0x200]}($param + 0x100) = "console=ttyS0,115200n8x root=/dev/nfsroot nfsroot=192.168.0.1:/project/m32r-linux/export/root.2.6_04 nfsaddrs=192.168.0.102:192.168.0.1:192.168.0.1:255.255.255.0:mappi: \0"
+
+
+end
+
+# Boot
+define boot
+  set_kernel_parameters
+  debug_chaos
+  set *(unsigned long *)0x00f00000=0x08002000
+  set $pc=0x08002000
+  set $fp=0
+  del b
+  si
+end
+
+# Restart
+define restart
+  sdireset
+  sdireset
+  setup
+  load_modules
+  boot
+end
+
+sdireset
+sdireset
+file vmlinux
+target m32rsdi
+
+restart
+boot
diff --git a/include/asm-m32r/ide.h b/include/asm-m32r/ide.h
index be64f24e37e..194393bd8be 100644
--- a/include/asm-m32r/ide.h
+++ b/include/asm-m32r/ide.h
@@ -35,7 +35,7 @@
 static __inline__ int ide_default_irq(unsigned long base)
 {
 	switch (base) {
-#if defined(CONFIG_PLAT_M32700UT) || defined(CONFIG_PLAT_MAPPI2)
+#if defined(CONFIG_PLAT_M32700UT) || defined(CONFIG_PLAT_MAPPI2) || defined(CONFIG_PLAT_MAPPI3)
 		case 0x1f0: return PLD_IRQ_CFIREQ;
 		default:
 			return 0;
diff --git a/include/asm-m32r/m32102.h b/include/asm-m32r/m32102.h
index b56034026bf..cb98101f4f6 100644
--- a/include/asm-m32r/m32102.h
+++ b/include/asm-m32r/m32102.h
@@ -175,6 +175,7 @@
 #define M32R_ICU_CR5_PORTL    (0x210+M32R_ICU_OFFSET)  /* INT4 */
 #define M32R_ICU_CR6_PORTL    (0x214+M32R_ICU_OFFSET)  /* INT5 */
 #define M32R_ICU_CR7_PORTL    (0x218+M32R_ICU_OFFSET)  /* INT6 */
+#define M32R_ICU_CR8_PORTL    (0x219+M32R_ICU_OFFSET)  /* INT7 */
 #define M32R_ICU_CR16_PORTL   (0x23C+M32R_ICU_OFFSET)  /* MFT0 */
 #define M32R_ICU_CR17_PORTL   (0x240+M32R_ICU_OFFSET)  /* MFT1 */
 #define M32R_ICU_CR18_PORTL   (0x244+M32R_ICU_OFFSET)  /* MFT2 */
diff --git a/include/asm-m32r/m32r.h b/include/asm-m32r/m32r.h
index f116649bbef..fd2b2823762 100644
--- a/include/asm-m32r/m32r.h
+++ b/include/asm-m32r/m32r.h
@@ -36,6 +36,10 @@
 #include <asm/mappi2/mappi2_pld.h>
 #endif	/* CONFIG_PLAT_MAPPI2 */
 
+#if defined(CONFIG_PLAT_MAPPI3)
+#include <asm/mappi3/mappi3_pld.h>
+#endif	/* CONFIG_PLAT_MAPPI3 */
+
 #if defined(CONFIG_PLAT_USRV)
 #include <asm/m32700ut/m32700ut_pld.h>
 #endif
diff --git a/include/asm-m32r/mappi3/mappi3_pld.h b/include/asm-m32r/mappi3/mappi3_pld.h
new file mode 100644
index 00000000000..3f1551f7f01
--- /dev/null
+++ b/include/asm-m32r/mappi3/mappi3_pld.h
@@ -0,0 +1,143 @@
+/*
+ * include/asm/mappi3/mappi3_pld.h
+ *
+ * Definitions for Extended IO Logic on MAPPI3 board.
+ *  based on m32700ut_pld.h
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ *
+ */
+
+#ifndef _MAPPI3_PLD_H
+#define _MAPPI3_PLD_H
+
+#ifndef __ASSEMBLY__
+/* FIXME:
+ * Some C functions use non-cache address, so can't define non-cache address.
+ */
+#define PLD_BASE		(0x1c000000 /* + NONCACHE_OFFSET */)
+#define __reg8			(volatile unsigned char *)
+#define __reg16			(volatile unsigned short *)
+#define __reg32			(volatile unsigned int *)
+#else
+#define PLD_BASE		(0x1c000000 + NONCACHE_OFFSET)
+#define __reg8
+#define __reg16
+#define __reg32
+#endif	/* __ASSEMBLY__ */
+
+/* CFC */
+#define	PLD_CFRSTCR		__reg16(PLD_BASE + 0x0000)
+#define PLD_CFSTS		__reg16(PLD_BASE + 0x0002)
+#define PLD_CFIMASK		__reg16(PLD_BASE + 0x0004)
+#define PLD_CFBUFCR		__reg16(PLD_BASE + 0x0006)
+#define PLD_CFCR0		__reg16(PLD_BASE + 0x000a)
+#define PLD_CFCR1		__reg16(PLD_BASE + 0x000c)
+
+/* MMC */
+#define PLD_MMCCR		__reg16(PLD_BASE + 0x4000)
+#define PLD_MMCMOD		__reg16(PLD_BASE + 0x4002)
+#define PLD_MMCSTS		__reg16(PLD_BASE + 0x4006)
+#define PLD_MMCBAUR		__reg16(PLD_BASE + 0x400a)
+#define PLD_MMCCMDBCUT		__reg16(PLD_BASE + 0x400c)
+#define PLD_MMCCDTBCUT		__reg16(PLD_BASE + 0x400e)
+#define PLD_MMCDET		__reg16(PLD_BASE + 0x4010)
+#define PLD_MMCWP		__reg16(PLD_BASE + 0x4012)
+#define PLD_MMCWDATA		__reg16(PLD_BASE + 0x5000)
+#define PLD_MMCRDATA		__reg16(PLD_BASE + 0x6000)
+#define PLD_MMCCMDDATA		__reg16(PLD_BASE + 0x7000)
+#define PLD_MMCRSPDATA		__reg16(PLD_BASE + 0x7006)
+
+/* Power Control of MMC and CF */
+#define PLD_CPCR		__reg16(PLD_BASE + 0x14000)
+
+
+/*==== ICU ====*/
+#define  M32R_IRQ_PC104        (5)   /* INT4(PC/104) */
+#define  M32R_IRQ_I2C          (28)  /* I2C-BUS     */
+#define  PLD_IRQ_CFIREQ       (6)  /* INT5 CFC Card Interrupt */
+#define  PLD_IRQ_CFC_INSERT   (7)  /* INT6 CFC Card Insert */
+#define  PLD_IRQ_CFC_EJECT    (8)  /* INT7 CFC Card Eject */
+#define  PLD_IRQ_MMCCARD      (43)  /* MMC Card Insert */
+#define  PLD_IRQ_MMCIRQ       (44)  /* MMC Transfer Done */
+
+
+#if 0
+/* LED Control
+ *
+ * 1: DIP swich side
+ * 2: Reset switch side
+ */
+#define PLD_IOLEDCR		__reg16(PLD_BASE + 0x14002)
+#define PLD_IOLED_1_ON		0x001
+#define PLD_IOLED_1_OFF		0x000
+#define PLD_IOLED_2_ON		0x002
+#define PLD_IOLED_2_OFF		0x000
+
+/* DIP Switch
+ *  0: Write-protect of Flash Memory (0:protected, 1:non-protected)
+ *  1: -
+ *  2: -
+ *  3: -
+ */
+#define PLD_IOSWSTS		__reg16(PLD_BASE + 0x14004)
+#define	PLD_IOSWSTS_IOSW2	0x0200
+#define	PLD_IOSWSTS_IOSW1	0x0100
+#define	PLD_IOSWSTS_IOWP0	0x0001
+
+#endif
+
+/* CRC */
+#define PLD_CRC7DATA		__reg16(PLD_BASE + 0x18000)
+#define PLD_CRC7INDATA		__reg16(PLD_BASE + 0x18002)
+#define PLD_CRC16DATA		__reg16(PLD_BASE + 0x18004)
+#define PLD_CRC16INDATA		__reg16(PLD_BASE + 0x18006)
+#define PLD_CRC16ADATA		__reg16(PLD_BASE + 0x18008)
+#define PLD_CRC16AINDATA	__reg16(PLD_BASE + 0x1800a)
+
+
+#if 0
+/* RTC */
+#define PLD_RTCCR		__reg16(PLD_BASE + 0x1c000)
+#define PLD_RTCBAUR		__reg16(PLD_BASE + 0x1c002)
+#define PLD_RTCWRDATA		__reg16(PLD_BASE + 0x1c004)
+#define PLD_RTCRDDATA		__reg16(PLD_BASE + 0x1c006)
+#define PLD_RTCRSTODT		__reg16(PLD_BASE + 0x1c008)
+
+/* SIO0 */
+#define PLD_ESIO0CR		__reg16(PLD_BASE + 0x20000)
+#define	PLD_ESIO0CR_TXEN	0x0001
+#define	PLD_ESIO0CR_RXEN	0x0002
+#define PLD_ESIO0MOD0		__reg16(PLD_BASE + 0x20002)
+#define	PLD_ESIO0MOD0_CTSS	0x0040
+#define	PLD_ESIO0MOD0_RTSS	0x0080
+#define PLD_ESIO0MOD1		__reg16(PLD_BASE + 0x20004)
+#define	PLD_ESIO0MOD1_LMFS	0x0010
+#define PLD_ESIO0STS		__reg16(PLD_BASE + 0x20006)
+#define	PLD_ESIO0STS_TEMP	0x0001
+#define	PLD_ESIO0STS_TXCP	0x0002
+#define	PLD_ESIO0STS_RXCP	0x0004
+#define	PLD_ESIO0STS_TXSC	0x0100
+#define	PLD_ESIO0STS_RXSC	0x0200
+#define PLD_ESIO0STS_TXREADY	(PLD_ESIO0STS_TXCP | PLD_ESIO0STS_TEMP)
+#define PLD_ESIO0INTCR		__reg16(PLD_BASE + 0x20008)
+#define	PLD_ESIO0INTCR_TXIEN	0x0002
+#define	PLD_ESIO0INTCR_RXCEN	0x0004
+#define PLD_ESIO0BAUR		__reg16(PLD_BASE + 0x2000a)
+#define PLD_ESIO0TXB		__reg16(PLD_BASE + 0x2000c)
+#define PLD_ESIO0RXB		__reg16(PLD_BASE + 0x2000e)
+
+/* SIM Card */
+#define PLD_SCCR		__reg16(PLD_BASE + 0x38000)
+#define PLD_SCMOD		__reg16(PLD_BASE + 0x38004)
+#define PLD_SCSTS		__reg16(PLD_BASE + 0x38006)
+#define PLD_SCINTCR		__reg16(PLD_BASE + 0x38008)
+#define PLD_SCBAUR		__reg16(PLD_BASE + 0x3800a)
+#define PLD_SCTXB		__reg16(PLD_BASE + 0x3800c)
+#define PLD_SCRXB		__reg16(PLD_BASE + 0x3800e)
+
+#endif
+
+#endif	/* _MAPPI3_PLD.H */
-- 
cgit v1.2.3-70-g09d2


From 0adbb44a146d6dff3f74ea7a9d3826f8942ed709 Mon Sep 17 00:00:00 2001
From: Hirokazu Takata <takata@linux-m32r.org>
Date: Tue, 21 Jun 2005 17:16:16 -0700
Subject: [PATCH] m32r: Remove include/asm-m32r/m32102peri.h

This patch removes an obsolete header file include/asm-m32r/m32102peri.h.
In this header, there are some undesirable single character types, like V.
And the header is almost no longer used.

Signed-off-by: Hayato Fujiwara <fujiwara@linux-m32r.org>
Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/net/8390.c            |   4 +-
 include/asm-m32r/m32102peri.h | 468 ------------------------------------------
 include/asm-m32r/m32r.h       |   1 -
 3 files changed, 2 insertions(+), 471 deletions(-)
 delete mode 100644 include/asm-m32r/m32102peri.h

(limited to 'include')

diff --git a/drivers/net/8390.c b/drivers/net/8390.c
index bab16bcc9ae..6d76f3a99b1 100644
--- a/drivers/net/8390.c
+++ b/drivers/net/8390.c
@@ -225,9 +225,9 @@ void ei_tx_timeout(struct net_device *dev)
 	unsigned long icucr;
 
 	local_irq_save(flags);
-	icucr = inl(ICUCR1);
+	icucr = inl(M32R_ICU_CR1_PORTL);
 	icucr |= M32R_ICUCR_ISMOD11;
-	outl(icucr, ICUCR1);
+	outl(icucr, M32R_ICU_CR1_PORTL);
 	local_irq_restore(flags);
 #endif
 	ei_local->stat.tx_errors++;
diff --git a/include/asm-m32r/m32102peri.h b/include/asm-m32r/m32102peri.h
deleted file mode 100644
index 3c12955ad0f..00000000000
--- a/include/asm-m32r/m32102peri.h
+++ /dev/null
@@ -1,468 +0,0 @@
-/* $Id$
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 2000,2001 by Hiroyuki Kondo
- */
-
-#ifndef __ASSEMBLY__
-
-typedef	void	V;
-typedef	char	B;
-typedef	short	S;
-typedef	int		W;
-typedef	long	L;
-typedef	float	F;
-typedef	double	D;
-typedef	unsigned char	UB;
-typedef	unsigned short	US;
-typedef	unsigned int	UW;
-typedef	unsigned long	UL;
-typedef	const unsigned int	CUW;
-
-/*********************************
-
-M32102 ICU
-
-*********************************/
-#define		ICUISTS		(UW *)0xa0EFF004
-#define		ICUIREQ0	(UW *)0xa0EFF008
-#define		ICUIREQ1	(UW *)0xa0EFF00C
-
-#define		ICUSBICR	(UW *)0xa0EFF018
-#define		ICUIMASK	(UW *)0xa0EFF01C
-
-#define		ICUCR1		(UW *)0xa0EFF200	/* INT0 */
-#define		ICUCR2		(UW *)0xa0EFF204	/* INT1 */
-#define		ICUCR3		(UW *)0xa0EFF208	/* INT2 */
-#define		ICUCR4		(UW *)0xa0EFF20C	/* INT3 */
-#define		ICUCR5		(UW *)0xa0EFF210	/* INT4 */
-#define		ICUCR6		(UW *)0xa0EFF214	/* INT5 */
-#define		ICUCR7		(UW *)0xa0EFF218	/* INT6 */
-
-#define		ICUCR16		(UW *)0xa0EFF23C	/* MFT0 */
-#define		ICUCR17		(UW *)0xa0EFF240	/* MFT1 */
-#define		ICUCR18		(UW *)0xa0EFF244	/* MFT2 */
-#define		ICUCR19		(UW *)0xa0EFF248	/* MFT3 */
-#define		ICUCR20		(UW *)0xa0EFF24C	/* MFT4 */
-#define		ICUCR21		(UW *)0xa0EFF250	/* MFT5 */
-
-#define		ICUCR32		(UW *)0xa0EFF27C	/* DMA0 */
-#define		ICUCR33		(UW *)0xa0EFF280	/* DMA1 */
-
-#define		ICUCR48		(UW *)0xa0EFF2BC	/* SIO0R */
-#define		ICUCR49		(UW *)0xa0EFF2C0	/* SIO0S */
-#define		ICUCR50		(UW *)0xa0EFF2C4	/* SIO1R */
-#define		ICUCR51		(UW *)0xa0EFF2C8	/* SIO1S */
-#define		ICUCR52		(UW *)0xa0EFF2CC	/* SIO2R */
-#define		ICUCR53		(UW *)0xa0EFF2D0	/* SIO2S */
-#define		ICUCR54		(UW *)0xa0EFF2D4	/* SIO3R */
-#define		ICUCR55		(UW *)0xa0EFF2D8	/* SIO3S */
-#define		ICUCR56		(UW *)0xa0EFF2DC	/* SIO4R */
-#define		ICUCR57		(UW *)0xa0EFF2E0	/* SIO4S */
-
-/*********************************
-
-M32102 MFT
-
-*********************************/
-#define		MFTCR		(US *)0xa0EFC002
-#define		MFTRPR		(UB *)0xa0EFC006
-
-#define		MFT0MOD		(US *)0xa0EFC102
-#define		MFT0BOS		(US *)0xa0EFC106
-#define		MFT0CUT		(US *)0xa0EFC10A
-#define		MFT0RLD		(US *)0xa0EFC10E
-#define		MFT0CRLD	(US *)0xa0EFC112
-
-#define		MFT1MOD		(US *)0xa0EFC202
-#define		MFT1BOS		(US *)0xa0EFC206
-#define		MFT1CUT		(US *)0xa0EFC20A
-#define		MFT1RLD		(US *)0xa0EFC20E
-#define		MFT1CRLD	(US *)0xa0EFC212
-
-#define		MFT2MOD		(US *)0xa0EFC302
-#define		MFT2BOS		(US *)0xa0EFC306
-#define		MFT2CUT		(US *)0xa0EFC30A
-#define		MFT2RLD		(US *)0xa0EFC30E
-#define		MFT2CRLD	(US *)0xa0EFC312
-
-#define		MFT3MOD		(US *)0xa0EFC402
-#define		MFT3CUT		(US *)0xa0EFC40A
-#define		MFT3RLD		(US *)0xa0EFC40E
-#define		MFT3CRLD	(US *)0xa0EFC412
-
-#define		MFT4MOD		(US *)0xa0EFC502
-#define		MFT4CUT		(US *)0xa0EFC50A
-#define		MFT4RLD		(US *)0xa0EFC50E
-#define		MFT4CRLD	(US *)0xa0EFC512
-
-#define		MFT5MOD		(US *)0xa0EFC602
-#define		MFT5CUT		(US *)0xa0EFC60A
-#define		MFT5RLD		(US *)0xa0EFC60E
-#define		MFT5CRLD	(US *)0xa0EFC612
-
-/*********************************
-
-M32102 SIO
-
-*********************************/
-
-#define SIO0CR     (volatile int *)0xa0efd000
-#define SIO0MOD0   (volatile int *)0xa0efd004
-#define SIO0MOD1   (volatile int *)0xa0efd008
-#define SIO0STS    (volatile int *)0xa0efd00c
-#define SIO0IMASK  (volatile int *)0xa0efd010
-#define SIO0BAUR   (volatile int *)0xa0efd014
-#define SIO0RBAUR  (volatile int *)0xa0efd018
-#define SIO0TXB    (volatile int *)0xa0efd01c
-#define SIO0RXB    (volatile int *)0xa0efd020
-
-#define SIO1CR     (volatile int *)0xa0efd100
-#define SIO1MOD0   (volatile int *)0xa0efd104
-#define SIO1MOD1   (volatile int *)0xa0efd108
-#define SIO1STS    (volatile int *)0xa0efd10c
-#define SIO1IMASK  (volatile int *)0xa0efd110
-#define SIO1BAUR   (volatile int *)0xa0efd114
-#define SIO1RBAUR  (volatile int *)0xa0efd118
-#define SIO1TXB    (volatile int *)0xa0efd11c
-#define SIO1RXB    (volatile int *)0xa0efd120
-/*********************************
-
-M32102 PORT
-
-*********************************/
-#define		PIEN		(UB *)0xa0EF1003	/* input enable */
-
-#define		P0DATA		(UB *)0xa0EF1020	/* data */
-#define		P1DATA		(UB *)0xa0EF1021
-#define		P2DATA		(UB *)0xa0EF1022
-#define		P3DATA		(UB *)0xa0EF1023
-#define		P4DATA		(UB *)0xa0EF1024
-#define		P5DATA		(UB *)0xa0EF1025
-#define		P6DATA		(UB *)0xa0EF1026
-#define		P7DATA		(UB *)0xa0EF1027
-
-#define		P0DIR		(UB *)0xa0EF1040	/* direction */
-#define		P1DIR		(UB *)0xa0EF1041
-#define		P2DIR		(UB *)0xa0EF1042
-#define		P3DIR		(UB *)0xa0EF1043
-#define		P4DIR		(UB *)0xa0EF1044
-#define		P5DIR		(UB *)0xa0EF1045
-#define		P6DIR		(UB *)0xa0EF1046
-#define		P7DIR		(UB *)0xa0EF1047
-
-#define		P0MOD		(US *)0xa0EF1060	/* mode control */
-#define		P1MOD		(US *)0xa0EF1062
-#define		P2MOD		(US *)0xa0EF1064
-#define		P3MOD		(US *)0xa0EF1066
-#define		P4MOD		(US *)0xa0EF1068
-#define		P5MOD		(US *)0xa0EF106A
-#define		P6MOD		(US *)0xa0EF106C
-#define		P7MOD		(US *)0xa0EF106E
-
-#define		P0ODCR		(UB *)0xa0EF1080	/* open-drain control */
-#define		P1ODCR		(UB *)0xa0EF1081
-#define		P2ODCR		(UB *)0xa0EF1082
-#define		P3ODCR		(UB *)0xa0EF1083
-#define		P4ODCR		(UB *)0xa0EF1084
-#define		P5ODCR		(UB *)0xa0EF1085
-#define		P6ODCR		(UB *)0xa0EF1086
-#define		P7ODCR		(UB *)0xa0EF1087
-
-/*********************************
-
-M32102 Cache
-
-********************************/
-
-#define		MCCR	(US *)0xFFFFFFFE
-
-
-#else  /* __ASSEMBLY__ */
-
-;;
-;; PIO     0x80ef1000
-;;
-
-#define PIEN          0xa0ef1000
-
-#define P0DATA        0xa0ef1020
-#define P1DATA        0xa0ef1021
-#define P2DATA        0xa0ef1022
-#define P3DATA        0xa0ef1023
-#define P4DATA        0xa0ef1024
-#define P5DATA        0xa0ef1025
-#define P6DATA        0xa0ef1026
-#define P7DATA        0xa0ef1027
-
-#define P0DIR         0xa0ef1040
-#define P1DIR         0xa0ef1041
-#define P2DIR         0xa0ef1042
-#define P3DIR         0xa0ef1043
-#define P4DIR         0xa0ef1044
-#define P5DIR         0xa0ef1045
-#define P6DIR         0xa0ef1046
-#define P7DIR         0xa0ef1047
-
-#define P0MOD         0xa0ef1060
-#define P1MOD         0xa0ef1062
-#define P2MOD         0xa0ef1064
-#define P3MOD         0xa0ef1066
-#define P4MOD         0xa0ef1068
-#define P5MOD         0xa0ef106a
-#define P6MOD         0xa0ef106c
-#define P7MOD         0xa0ef106e
-;
-#define P0ODCR        0xa0ef1080
-#define P1ODCR        0xa0ef1081
-#define P2ODCR        0xa0ef1082
-#define P3ODCR        0xa0ef1083
-#define P4ODCR        0xa0ef1084
-#define P5ODCR        0xa0ef1085
-#define P6ODCR        0xa0ef1086
-#define P7ODCR        0xa0ef1087
-
-;;
-;; WDT     0xa0ef2000
-;;
-
-#define WDTCR         0xa0ef2000
-
-
-;;
-;; CLK     0xa0ef4000
-;;
-
-#define CPUCLKCR      0xa0ef4000
-#define CLKMOD        0xa0ef4004
-#define PLLCR         0xa0ef4008
-
-
-;;
-;; BSEL    0xa0ef5000
-;;
-
-#define BSEL0CR       0xa0ef5000
-#define BSEL1CR       0xa0ef5004
-#define BSEL2CR       0xa0ef5008
-#define BSEL3CR       0xa0ef500c
-#define BSEL4CR       0xa0ef5010
-#define BSEL5CR       0xa0ef5014
-
-
-;;
-;; SDRAMC  0xa0ef6000
-;;
-
-#define SDRF0         0xa0ef6000
-#define SDRF1         0xa0ef6004
-#define SDIR0         0xa0ef6008
-#define SDIR1         0xa0ef600c
-#define SDBR          0xa0ef6010
-
-;; CH0
-#define SD0ADR        0xa0ef6020
-#define SD0SZ         0xa0ef6022
-#define SD0ER         0xa0ef6024
-#define SD0TR         0xa0ef6028
-#define SD0MOD        0xa0ef602c
-
-;; CH1
-#define SD1ADR        0xa0ef6040
-#define SD1SZ         0xa0ef6042
-#define SD1ER         0xa0ef6044
-#define SD1TR         0xa0ef6048
-#define SD1MOD        0xa0ef604c
-
-
-;;
-;; DMAC    0xa0ef8000
-;;
-
-#define DMAEN         0xa0ef8000
-#define DMAISTS       0xa0ef8004
-#define DMAEDET       0xa0ef8008
-#define DMAASTS       0xa0ef800c
-
-;; CH0
-#define DMA0CR0       0xa0ef8100
-#define DMA0CR1       0xa0ef8104
-#define DMA0CSA       0xa0ef8108
-#define DMA0RSA       0xa0ef810c
-#define DMA0CDA       0xa0ef8110
-#define DMA0RDA       0xa0ef8114
-#define DMA0CBCUT     0xa0ef8118
-#define DMA0RBCUT     0xa0ef811c
-
-;; CH1
-#define DMA1CR0       0xa0ef8200
-#define DMA1CR1       0xa0ef8204
-#define DMA1CSA       0xa0ef8208
-#define DMA1RSA       0xa0ef820c
-#define DMA1CDA       0xa0ef8210
-#define DMA1RDA       0xa0ef8214
-#define DMA1CBCUT     0xa0ef8218
-#define DMA1RBCUT     0xa0ef821c
-
-
-;;
-;; MFT     0xa0efc000
-;;
-
-#define MFTCR        0xa0efc000
-#define MFTRPR       0xa0efc004
-
-;; CH0
-#define MFT0MOD      0xa0efc100
-#define MFT0BOS      0xa0efc104
-#define MFT0CUT      0xa0efc108
-#define MFT0RLD      0xa0efc10c
-#define MFT0CMPRLD   0xa0efc110
-
-;; CH1
-#define MFT1MOD      0xa0efc200
-#define MFT1BOS      0xa0efc204
-#define MFT1CUT      0xa0efc208
-#define MFT1RLD      0xa0efc20c
-#define MFT1CMPRLD   0xa0efc210
-
-;; CH2
-#define MFT2MOD      0xa0efc300
-#define MFT2BOS      0xa0efc304
-#define MFT2CUT      0xa0efc308
-#define MFT2RLD      0xa0efc30c
-#define MFT2CMPRLD   0xa0efc310
-
-;; CH3
-#define MFT3MOD      0xa0efc400
-#define MFT3BOS      0xa0efc404
-#define MFT3CUT      0xa0efc408
-#define MFT3RLD      0xa0efc40c
-#define MFT3CMPRLD   0xa0efc410
-
-;; CH4
-#define MFT4MOD      0xa0efc500
-#define MFT4BOS      0xa0efc504
-#define MFT4CUT      0xa0efc508
-#define MFT4RLD      0xa0efc50c
-#define MFT4CMPRLD   0xa0efc510
-
-;; CH5
-#define MFT5MOD      0xa0efc600
-#define MFT5BOS      0xa0efc604
-#define MFT5CUT      0xa0efc608
-#define MFT5RLD      0xa0efc60c
-#define MFT5CMPRLD   0xa0efc610
-
-
-;;
-;; SIO     0xa0efd000
-;;
-
-;; CH0
-#define SIO0CR        0xa0efd000
-#define SIO0MOD0      0xa0efd004
-#define SIO0MOD1      0xa0efd008
-#define SIO0STS       0xa0efd00c
-#define SIO0IMASK     0xa0efd010
-#define SIO0BAUR      0xa0efd014
-#define SIO0RBAUR     0xa0efd018
-#define SIO0TXB       0xa0efd01c
-#define SIO0RXB       0xa0efd020
-
-;; CH1
-#define SIO1CR        0xa0efd100
-#define SIO1MOD0      0xa0efd104
-#define SIO1MOD1      0xa0efd108
-#define SIO1STS       0xa0efd10c
-#define SIO1IMASK     0xa0efd110
-#define SIO1BAUR      0xa0efd114
-#define SIO1RBAUR     0xa0efd118
-#define SIO1TXB       0xa0efd11c
-#define SIO1RXB       0xa0efd120
-
-;; CH2
-#define SIO2CR        0xa0efd200
-#define SIO2MOD0      0xa0efd204
-#define SIO2MOD1      0xa0efd208
-#define SIO2STS       0xa0efd20c
-#define SIO2IMASK     0xa0efd210
-#define SIO2BAUR      0xa0efd214
-#define SIO2RBAUR     0xa0efd218
-#define SIO2TXB       0xa0efd21c
-#define SIO2RXB       0xa0efd220
-
-;; CH3
-#define SIO3CR        0xa0efd300
-#define SIO3MOD0      0xa0efd304
-#define SIO3MOD1      0xa0efd308
-#define SIO3STS       0xa0efd30c
-#define SIO3IMASK     0xa0efd310
-#define SIO3BAUR      0xa0efd314
-#define SIO3RBAUR     0xa0efd318
-#define SIO3TXB       0xa0efd31c
-#define SIO3RXB       0xa0efd320
-
-;; CH4
-#define SIO4CR        0xa0efd400
-#define SIO4MOD0      0xa0efd404
-#define SIO4MOD1      0xa0efd408
-#define SIO4STS       0xa0efd40c
-#define SIO4IMASK     0xa0efd410
-#define SIO4BAUR      0xa0efd414
-#define SIO4RBAUR     0xa0efd418
-#define SIO4TXB       0xa0efd41c
-#define SIO4RXB       0xa0efd420
-
-
-;;
-;; ICU     0xa0eff000
-;;
-
-#define ICUISTS       0xa0eff004
-#define ICUIREQ0      0xa0eff008
-#define ICUIREQ1      0xa0eff00c
-
-#define ICUSBICR      0xa0eff018
-#define ICUIMASK      0xa0eff01c
-
-#define ICUCR1        0xa0eff200
-#define ICUCR2        0xa0eff204
-#define ICUCR3        0xa0eff208
-#define ICUCR4        0xa0eff20c
-#define ICUCR5        0xa0eff210
-#define ICUCR6        0xa0eff214
-#define ICUCR7        0xa0eff218
-
-#define ICUCR16       0xa0eff23c
-#define ICUCR17       0xa0eff240
-#define ICUCR18       0xa0eff244
-#define ICUCR19       0xa0eff248
-#define ICUCR20       0xa0eff24c
-#define ICUCR21       0xa0eff250
-
-#define ICUCR32       0xa0eff27c
-#define ICUCR33       0xa0eff280
-
-#define ICUCR48       0xa0eff2bc
-#define ICUCR49       0xa0eff2c0
-#define ICUCR50       0xa0eff2c4
-#define ICUCR51       0xa0eff2c8
-#define ICUCR52       0xa0eff2cc
-#define ICUCR53       0xa0eff2d0
-#define ICUCR54       0xa0eff2d4
-#define ICUCR55       0xa0eff2d8
-#define ICUCR56       0xa0eff2dc
-#define ICUCR57       0xa0eff2e0
-
-;;
-;; CACHE
-;;
-
-#define MCCR		  0xfffffffc
-
-
-#endif  /* __ASSEMBLY__ */
diff --git a/include/asm-m32r/m32r.h b/include/asm-m32r/m32r.h
index fd2b2823762..ec142be0086 100644
--- a/include/asm-m32r/m32r.h
+++ b/include/asm-m32r/m32r.h
@@ -16,7 +16,6 @@
 	|| defined(CONFIG_CHIP_M32700) || defined(CONFIG_CHIP_M32102) \
         || defined(CONFIG_CHIP_OPSP)
 #include <asm/m32102.h>
-#include <asm/m32102peri.h>
 #endif
 
 /* Platform type */
-- 
cgit v1.2.3-70-g09d2


From 5757b284a300e0e5d2173750906625b6470bd9f0 Mon Sep 17 00:00:00 2001
From: Hirokazu Takata <takata@linux-m32r.org>
Date: Tue, 21 Jun 2005 17:16:17 -0700
Subject: [PATCH] m32r: Use asm-generic/div64.h

The current include/asm-m32r/div64.h of 2.6.12-rc5 looks buggy.  Here is a
patch for updating it to use asm-generic/div64.h for m32r like other
architectures.

Signed-off-by: Hitoshi Yamamoto <hitoshiy@isl.melco.co.jp>
Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-m32r/div64.h | 39 +--------------------------------------
 1 file changed, 1 insertion(+), 38 deletions(-)

(limited to 'include')

diff --git a/include/asm-m32r/div64.h b/include/asm-m32r/div64.h
index 417a51bd552..6cd978cefb2 100644
--- a/include/asm-m32r/div64.h
+++ b/include/asm-m32r/div64.h
@@ -1,38 +1 @@
-#ifndef _ASM_M32R_DIV64
-#define _ASM_M32R_DIV64
-
-/* $Id$ */
-
-/* unsigned long long division.
- * Input:
- *  unsigned long long  n
- *  unsigned long  base
- * Output:
- *  n = n / base;
- *  return value = n % base;
- */
-#define do_div(n, base)						\
-({								\
-	unsigned long _res, _high, _mid, _low;			\
-								\
-	_low = (n) & 0xffffffffUL;				\
-	_high = (n) >> 32;					\
-	if (_high) {						\
-		_mid = (_high % (unsigned long)(base)) << 16;	\
-		_high = _high / (unsigned long)(base);		\
-		_mid += _low >> 16;				\
-		_low &= 0x0000ffffUL;				\
-		_low += (_mid % (unsigned long)(base)) << 16;	\
-		_mid = _mid / (unsigned long)(base);		\
-		_res = _low % (unsigned long)(base);		\
-		_low = _low / (unsigned long)(base);		\
-		n = _low + ((long long)_mid << 16) +		\
-			((long long)_high << 32);		\
-	} else {						\
-		_res = _low % (unsigned long)(base);		\
-		n = (_low / (unsigned long)(base));		\
-	}							\
-	_res;							\
-})
-
-#endif  /* _ASM_M32R_DIV64 */
+#include <asm-generic/div64.h>
-- 
cgit v1.2.3-70-g09d2


From dbce706e2550253c5ab6043f4f5dfde0cd02470f Mon Sep 17 00:00:00 2001
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Date: Tue, 21 Jun 2005 17:16:19 -0700
Subject: [PATCH] uml: add and use generic hw_controller_type->release

With Chris Wedgwood <cw@f00f.org>

Currently UML must explicitly call the UML-specific
free_irq_by_irq_and_dev() for each free_irq call it's done.

This is needed because ->shutdown and/or ->disable are only called when the
last "action" for that irq is removed.

Instead, for UML shared IRQs (UML IRQs are very often, if not always,
shared), for each dev_id some setup is done, which must be cleared on the
release of that fd.  For instance, for each open console a new instance
(i.e.  new dev_id) of the same IRQ is requested().

Exactly, a fd is stored in an array (pollfds), which is after read by a
host thread and passed to poll().  Each event registered by poll() triggers
an interrupt.  So, for each free_irq() we must remove the corresponding
host fd from the table, which we do via this -release() method.

In this patch we add an appropriate hook for this, and remove all uses of
it by pointing the hook to the said procedure; this is safe to do since the
said procedure.

Also some cosmetic improvements are included.

This is heavily based on some work by Chris Wedgwood, which however didn't
get the patch merged for something I'd call a "misunderstanding" (the need
for this patch wasn't cleanly explained, thus adding the generic hook was
felt as undesirable).

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
CC: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/drivers/line.c       |  2 --
 arch/um/drivers/net_kern.c   |  1 -
 arch/um/drivers/port_kern.c  |  1 -
 arch/um/drivers/xterm_kern.c |  1 -
 arch/um/kernel/irq.c         | 11 +++++++----
 arch/um/kernel/irq_user.c    |  2 --
 include/linux/irq.h          |  1 +
 kernel/irq/manage.c          |  4 ++++
 8 files changed, 12 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c
index 025d3be8aca..562f864254b 100644
--- a/arch/um/drivers/line.c
+++ b/arch/um/drivers/line.c
@@ -406,14 +406,12 @@ void line_disable(struct tty_struct *tty, int current_irq)
 	if(line->driver->read_irq == current_irq)
 		free_irq_later(line->driver->read_irq, tty);
 	else {
-		free_irq_by_irq_and_dev(line->driver->read_irq, tty);
 		free_irq(line->driver->read_irq, tty);
 	}
 
 	if(line->driver->write_irq == current_irq)
 		free_irq_later(line->driver->write_irq, tty);
 	else {
-		free_irq_by_irq_and_dev(line->driver->write_irq, tty);
 		free_irq(line->driver->write_irq, tty);
 	}
 
diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c
index 4eeaf88c1e9..5388a742869 100644
--- a/arch/um/drivers/net_kern.c
+++ b/arch/um/drivers/net_kern.c
@@ -146,7 +146,6 @@ static int uml_net_close(struct net_device *dev)
 	netif_stop_queue(dev);
 	spin_lock(&lp->lock);
 
-	free_irq_by_irq_and_dev(dev->irq, dev);
 	free_irq(dev->irq, dev);
 	if(lp->close != NULL)
 		(*lp->close)(lp->fd, &lp->user);
diff --git a/arch/um/drivers/port_kern.c b/arch/um/drivers/port_kern.c
index b5ee07472f7..c41efd207fc 100644
--- a/arch/um/drivers/port_kern.c
+++ b/arch/um/drivers/port_kern.c
@@ -257,7 +257,6 @@ int port_wait(void *data)
 		 * connection.  Then we loop here throwing out failed 
 		 * connections until a good one is found.
 		 */
-		free_irq_by_irq_and_dev(TELNETD_IRQ, conn);
 		free_irq(TELNETD_IRQ, conn);
 
 		if(conn->fd >= 0) break;
diff --git a/arch/um/drivers/xterm_kern.c b/arch/um/drivers/xterm_kern.c
index a4fdf3584ad..d269a80f4b0 100644
--- a/arch/um/drivers/xterm_kern.c
+++ b/arch/um/drivers/xterm_kern.c
@@ -69,7 +69,6 @@ int xterm_fd(int socket, int *pid_out)
 	 * isn't set) this will hang... */
 	wait_for_completion(&data->ready);
 
-	free_irq_by_irq_and_dev(XTERM_IRQ, data);
 	free_irq(XTERM_IRQ, data);
 
 	ret = data->new_fd;
diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c
index d44fb528254..9f18061ef4c 100644
--- a/arch/um/kernel/irq.c
+++ b/arch/um/kernel/irq.c
@@ -124,14 +124,16 @@ void irq_unlock(unsigned long flags)
 	spin_unlock_irqrestore(&irq_spinlock, flags);
 }
 
-/*  presently hw_interrupt_type must define (startup || enable) &&
- *  disable && end */
+/* hw_interrupt_type must define (startup || enable) &&
+ * (shutdown || disable) && end */
 static void dummy(unsigned int irq)
 {
 }
 
-static struct hw_interrupt_type SIGIO_irq_type = {
+/* This is used for everything else than the timer. */
+static struct hw_interrupt_type normal_irq_type = {
 	.typename = "SIGIO",
+	.release = free_irq_by_irq_and_dev,
 	.disable = dummy,
 	.enable = dummy,
 	.ack = dummy,
@@ -140,6 +142,7 @@ static struct hw_interrupt_type SIGIO_irq_type = {
 
 static struct hw_interrupt_type SIGVTALRM_irq_type = {
 	.typename = "SIGVTALRM",
+	.release = free_irq_by_irq_and_dev,
 	.shutdown = dummy, /* never called */
 	.disable = dummy,
 	.enable = dummy,
@@ -160,7 +163,7 @@ void __init init_IRQ(void)
 		irq_desc[i].status = IRQ_DISABLED;
 		irq_desc[i].action = NULL;
 		irq_desc[i].depth = 1;
-		irq_desc[i].handler = &SIGIO_irq_type;
+		irq_desc[i].handler = &normal_irq_type;
 		enable_irq(i);
 	}
 }
diff --git a/arch/um/kernel/irq_user.c b/arch/um/kernel/irq_user.c
index b3074cbaa47..c3ccaf24f3e 100644
--- a/arch/um/kernel/irq_user.c
+++ b/arch/um/kernel/irq_user.c
@@ -85,8 +85,6 @@ void sigio_handler(int sig, union uml_pt_regs *regs)
 				next = irq_fd->next;
 				if(irq_fd->freed){
 					free_irq(irq_fd->irq, irq_fd->id);
-					free_irq_by_irq_and_dev(irq_fd->irq,
-								irq_fd->id);
 				}
 			}
 		}
diff --git a/include/linux/irq.h b/include/linux/irq.h
index c3ff4d10166..b68ad80e246 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -47,6 +47,7 @@ struct hw_interrupt_type {
 	void (*ack)(unsigned int irq);
 	void (*end)(unsigned int irq);
 	void (*set_affinity)(unsigned int irq, cpumask_t dest);
+	void (*release)(unsigned int irq, void *dev_id);
 };
 
 typedef struct hw_interrupt_type  hw_irq_controller;
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 5202e4c4a5b..5fde8177eed 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -255,6 +255,10 @@ void free_irq(unsigned int irq, void *dev_id)
 
 			/* Found it - now remove it from the list of entries */
 			*pp = action->next;
+
+			if (desc->handler->release)
+				desc->handler->release(irq, dev_id);
+
 			if (!desc->action) {
 				desc->status |= IRQ_DISABLED;
 				if (desc->handler->shutdown)
-- 
cgit v1.2.3-70-g09d2


From b77d6adc922b8bbf8b16b67f567958c42962cf88 Mon Sep 17 00:00:00 2001
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Date: Tue, 21 Jun 2005 17:16:24 -0700
Subject: [PATCH] uml: make hw_controller_type->release exist only for archs
 needing it

With Chris Wedgwood <cw@f00f.org>

As suggested by Chris, we can make the "just added" method ->release
conditional to UML only (better: to archs requesting it, i.e.  only UML
currently), so that other archs don't get this unneeded crud, and if UML
won't need it any more we can kill this.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
CC: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/Kconfig     | 5 +++++
 include/linux/irq.h | 3 +++
 kernel/irq/manage.c | 4 ++++
 3 files changed, 12 insertions(+)

(limited to 'include')

diff --git a/arch/um/Kconfig b/arch/um/Kconfig
index c5292181a66..b8e952c88fd 100644
--- a/arch/um/Kconfig
+++ b/arch/um/Kconfig
@@ -35,6 +35,11 @@ config GENERIC_CALIBRATE_DELAY
 	bool
 	default y
 
+# Used in kernel/irq/manage.c and include/linux/irq.h
+config IRQ_RELEASE_METHOD
+	bool
+	default y
+
 menu "UML-specific options"
 
 config MODE_TT
diff --git a/include/linux/irq.h b/include/linux/irq.h
index b68ad80e246..7fc1022be9e 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -47,7 +47,10 @@ struct hw_interrupt_type {
 	void (*ack)(unsigned int irq);
 	void (*end)(unsigned int irq);
 	void (*set_affinity)(unsigned int irq, cpumask_t dest);
+	/* Currently used only by UML, might disappear one day.*/
+#ifdef CONFIG_IRQ_RELEASE_METHOD
 	void (*release)(unsigned int irq, void *dev_id);
+#endif
 };
 
 typedef struct hw_interrupt_type  hw_irq_controller;
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 5fde8177eed..ac670098570 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -6,6 +6,7 @@
  * This file contains driver APIs to the irq subsystem.
  */
 
+#include <linux/config.h>
 #include <linux/irq.h>
 #include <linux/module.h>
 #include <linux/random.h>
@@ -256,8 +257,11 @@ void free_irq(unsigned int irq, void *dev_id)
 			/* Found it - now remove it from the list of entries */
 			*pp = action->next;
 
+			/* Currently used only by UML, might disappear one day.*/
+#ifdef CONFIG_IRQ_RELEASE_METHOD
 			if (desc->handler->release)
 				desc->handler->release(irq, dev_id);
+#endif
 
 			if (!desc->action) {
 				desc->status |= IRQ_DISABLED;
-- 
cgit v1.2.3-70-g09d2


From 8a96619145840c6eb50d85759f72d9337db411f7 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Tue, 21 Jun 2005 17:16:41 -0700
Subject: [PATCH] autofs4: subversion bump to identify these changes

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/auto_fs4.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/auto_fs4.h b/include/linux/auto_fs4.h
index a1657fb9951..9343c89d843 100644
--- a/include/linux/auto_fs4.h
+++ b/include/linux/auto_fs4.h
@@ -23,7 +23,7 @@
 #define AUTOFS_MIN_PROTO_VERSION	3
 #define AUTOFS_MAX_PROTO_VERSION	4
 
-#define AUTOFS_PROTO_SUBVERSION		6
+#define AUTOFS_PROTO_SUBVERSION		7
 
 /* Mask for expire behaviour */
 #define AUTOFS_EXP_IMMEDIATE		1
-- 
cgit v1.2.3-70-g09d2


From f5a9951c94e7a285a3d00648e3d790a7f016bd11 Mon Sep 17 00:00:00 2001
From: James Simmons <jsimmons@pentafluge.infradead.org>
Date: Tue, 21 Jun 2005 17:16:58 -0700
Subject: [PATCH] fbdev: iomove removal

Since no one is using the inbuf, outbuf of struct fb_pixmap I removed their
use in the framebuffer console.  The idea is instead move the pixmap
functionality below the accelerated functions intead of on top as the way
it is now.  If there is no objection please apply.  This is against Linus
latestr GIT tree.  Thank you.

Signed-off-by: James Simmons <jsimmons@www.infradead.org>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/console/bitblit.c | 18 ++-----------
 drivers/video/fbmem.c           | 57 ++++-------------------------------------
 drivers/video/softcursor.c      |  8 ++----
 include/linux/fb.h              | 12 +++------
 4 files changed, 12 insertions(+), 83 deletions(-)

(limited to 'include')

diff --git a/drivers/video/console/bitblit.c b/drivers/video/console/bitblit.c
index b28a4b0e395..4eef20790c3 100644
--- a/drivers/video/console/bitblit.c
+++ b/drivers/video/console/bitblit.c
@@ -107,13 +107,6 @@ static void bit_putcs(struct vc_data *vc, struct fb_info *info,
 		      const unsigned short *s, int count, int yy, int xx,
 		      int fg, int bg)
 {
-	void (*move_unaligned)(struct fb_info *info, struct fb_pixmap *buf,
-			       u8 *dst, u32 d_pitch, u8 *src, u32 idx,
-			       u32 height, u32 shift_high, u32 shift_low,
-			       u32 mod);
-	void (*move_aligned)(struct fb_info *info, struct fb_pixmap *buf,
-			     u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch,
-			     u32 height);
 	unsigned short charmask = vc->vc_hi_font_mask ? 0x1ff : 0xff;
 	unsigned int width = (vc->vc_font.width + 7) >> 3;
 	unsigned int cellsize = vc->vc_font.height * width;
@@ -141,13 +134,6 @@ static void bit_putcs(struct vc_data *vc, struct fb_info *info,
 	image.height = vc->vc_font.height;
 	image.depth = 1;
 
-	if (info->pixmap.outbuf && info->pixmap.inbuf) {
-		move_aligned = fb_iomove_buf_aligned;
-		move_unaligned = fb_iomove_buf_unaligned;
-	} else {
-		move_aligned = fb_sysmove_buf_aligned;
-		move_unaligned = fb_sysmove_buf_unaligned;
-	}
 	while (count) {
 		if (count > maxcnt)
 			cnt = k = maxcnt;
@@ -171,7 +157,7 @@ static void bit_putcs(struct vc_data *vc, struct fb_info *info,
 					src = buf;
 				}
 
-				move_unaligned(info, &info->pixmap, dst, pitch,
+				fb_sysmove_buf_unaligned(info, &info->pixmap, dst, pitch,
 					       src, idx, image.height,
 					       shift_high, shift_low, mod);
 				shift_low += mod;
@@ -189,7 +175,7 @@ static void bit_putcs(struct vc_data *vc, struct fb_info *info,
 					src = buf;
 				}
 
-				move_aligned(info, &info->pixmap, dst, pitch,
+				fb_sysmove_buf_aligned(info, &info->pixmap, dst, pitch,
 					     src, idx, image.height);
 				dst += width;
 			}
diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c
index 8cef020d180..78907a87349 100644
--- a/drivers/video/fbmem.c
+++ b/drivers/video/fbmem.c
@@ -76,65 +76,21 @@ int fb_get_color_depth(struct fb_var_screeninfo *var)
 EXPORT_SYMBOL(fb_get_color_depth);
 
 /*
- * Drawing helpers.
+ * Data padding functions.
  */
-void fb_iomove_buf_aligned(struct fb_info *info, struct fb_pixmap *buf,
-			   u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch,
-			   u32 height)
-{
-	int i;
-
-	for (i = height; i--; ) {
-		buf->outbuf(info, dst, src, s_pitch);
-		src += s_pitch;
-		dst += d_pitch;
-	}
-}
-
 void fb_sysmove_buf_aligned(struct fb_info *info, struct fb_pixmap *buf,
 			    u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch,
 			    u32 height)
 {
-	int i, j;
+	int i;
 
 	for (i = height; i--; ) {
-		for (j = 0; j < s_pitch; j++)
-			dst[j] = src[j];
+		memcpy(dst, src, s_pitch);
 		src += s_pitch;
 		dst += d_pitch;
 	}
 }
-
-void fb_iomove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf,
-			     u8 *dst, u32 d_pitch, u8 *src, u32 idx,
-			     u32 height, u32 shift_high, u32 shift_low,
-			     u32 mod)
-{
-	u8 mask = (u8) (0xfff << shift_high), tmp;
-	int i, j;
-
-	for (i = height; i--; ) {
-		for (j = 0; j < idx; j++) {
-			tmp = buf->inbuf(info, dst+j);
-			tmp &= mask;
-			tmp |= *src >> shift_low;
-			buf->outbuf(info, dst+j, &tmp, 1);
-			tmp = *src << shift_high;
-			buf->outbuf(info, dst+j+1, &tmp, 1);
-			src++;
-		}
-		tmp = buf->inbuf(info, dst+idx);
-		tmp &= mask;
-		tmp |= *src >> shift_low;
-		buf->outbuf(info, dst+idx, &tmp, 1);
-		if (shift_high < mod) {
-			tmp = *src << shift_high;
-			buf->outbuf(info, dst+idx+1, &tmp, 1);
-		}	
-		src++;
-		dst += d_pitch;
-	}
-}
+EXPORT_SYMBOL(fb_sysmove_buf_aligned);
 
 void fb_sysmove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf,
 			      u8 *dst, u32 d_pitch, u8 *src, u32 idx,
@@ -166,6 +122,7 @@ void fb_sysmove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf,
 		dst += d_pitch;
 	}
 }
+EXPORT_SYMBOL(fb_sysmove_buf_unaligned);
 
 /*
  * we need to lock this section since fb_cursor
@@ -1357,10 +1314,6 @@ EXPORT_SYMBOL(fb_set_var);
 EXPORT_SYMBOL(fb_blank);
 EXPORT_SYMBOL(fb_pan_display);
 EXPORT_SYMBOL(fb_get_buffer_offset);
-EXPORT_SYMBOL(fb_iomove_buf_unaligned);
-EXPORT_SYMBOL(fb_iomove_buf_aligned);
-EXPORT_SYMBOL(fb_sysmove_buf_unaligned);
-EXPORT_SYMBOL(fb_sysmove_buf_aligned);
 EXPORT_SYMBOL(fb_set_suspend);
 EXPORT_SYMBOL(fb_register_client);
 EXPORT_SYMBOL(fb_unregister_client);
diff --git a/drivers/video/softcursor.c b/drivers/video/softcursor.c
index 13a4511539a..a6c5ca88d6b 100644
--- a/drivers/video/softcursor.c
+++ b/drivers/video/softcursor.c
@@ -58,12 +58,8 @@ int soft_cursor(struct fb_info *info, struct fb_cursor *cursor)
 	} else 
 		memcpy(src, image->data, dsize);
 	
-	if (info->pixmap.outbuf)
-		fb_iomove_buf_aligned(info, &info->pixmap, dst, d_pitch, src,
-				  s_pitch, image->height);
-	else
-		fb_sysmove_buf_aligned(info, &info->pixmap, dst, d_pitch, src,
-				   s_pitch, image->height);
+	fb_sysmove_buf_aligned(info, &info->pixmap, dst, d_pitch, src,
+				s_pitch, image->height);
 
 	image->data = dst;
 	info->fbops->fb_imageblit(info, image);
diff --git a/include/linux/fb.h b/include/linux/fb.h
index b468bf49654..e14942805db 100644
--- a/include/linux/fb.h
+++ b/include/linux/fb.h
@@ -524,11 +524,11 @@ struct fb_pixmap {
 	u32 offset;		/* current offset to buffer		*/
 	u32 buf_align;		/* byte alignment of each bitmap	*/
 	u32 scan_align;		/* alignment per scanline		*/
-	u32 access_align;	/* alignment per read/write		*/
+	u32 access_align;	/* alignment per read/write (bits)	*/
 	u32 flags;		/* see FB_PIXMAP_*			*/
 	/* access methods */
-	void (*outbuf)(struct fb_info *info, u8 *addr, u8 *src, unsigned int size);
-	u8   (*inbuf) (struct fb_info *info, u8 *addr);
+	void (*writeio)(struct fb_info *info, void __iomem *dst, void *src, unsigned int size);
+	void (*readio) (struct fb_info *info, void *dst, void __iomem *src, unsigned int size);
 };
 
 
@@ -816,12 +816,6 @@ extern int unregister_framebuffer(struct fb_info *fb_info);
 extern int fb_prepare_logo(struct fb_info *fb_info);
 extern int fb_show_logo(struct fb_info *fb_info);
 extern char* fb_get_buffer_offset(struct fb_info *info, struct fb_pixmap *buf, u32 size);
-extern void fb_iomove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf,
-				u8 *dst, u32 d_pitch, u8 *src, u32 idx,
-				u32 height, u32 shift_high, u32 shift_low, u32 mod);
-extern void fb_iomove_buf_aligned(struct fb_info *info, struct fb_pixmap *buf,
-				u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch,
-				u32 height);
 extern void fb_sysmove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf,
 				u8 *dst, u32 d_pitch, u8 *src, u32 idx,
 				u32 height, u32 shift_high, u32 shift_low, u32 mod);
-- 
cgit v1.2.3-70-g09d2


From 1154ea7dcd8eed758fb5ec47393a79d5a1f0bc43 Mon Sep 17 00:00:00 2001
From: Jaya Kumar <jayalk@intworks.biz>
Date: Tue, 21 Jun 2005 17:17:04 -0700
Subject: [PATCH] Framebuffer driver for Arc LCD board

Add support for the Arc monochrome LCD board.

The board uses KS108 controllers to drive individual 64x64 LCD matrices.
The board can be paneled in a variety of setups such as 2x1=128x64,
4x4=256x256 and so on.  The board/host interface is through GPIO.

Signed-off-by: Jaya Kumar <jayalk@intworks.biz>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: <linux-fbdev-devel@lists.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 CREDITS                |   7 +
 MAINTAINERS            |   5 +
 drivers/video/Kconfig  |  16 ++
 drivers/video/Makefile |   1 +
 drivers/video/arcfb.c  | 684 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/arcfb.h  |   8 +
 6 files changed, 721 insertions(+)
 create mode 100644 drivers/video/arcfb.c
 create mode 100644 include/linux/arcfb.h

(limited to 'include')

diff --git a/CREDITS b/CREDITS
index d65ffe5a4d0..2993348ce6c 100644
--- a/CREDITS
+++ b/CREDITS
@@ -1880,6 +1880,13 @@ S: Schlehenweg 9
 S: D-91080 Uttenreuth
 S: Germany
 
+N: Jaya Kumar
+E: jayalk@intworks.biz
+W: http://www.intworks.biz
+D: Arc monochrome LCD framebuffer driver, x86 reboot fixups
+S: Gurgaon, India
+S: Kuala Lumpur, Malaysia
+
 N: Gabor Kuti
 M: seasons@falcon.sch.bme.hu
 M: seasons@makosteszta.sote.hu
diff --git a/MAINTAINERS b/MAINTAINERS
index e3f0f3f157c..0f88a70a15d 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -265,6 +265,11 @@ P:	Arnaldo Carvalho de Melo
 M:	acme@conectiva.com.br
 S:	Maintained
 
+ARC FRAMEBUFFER DRIVER
+P:	Jaya Kumar
+M:	jayalk@intworks.biz
+S:	Maintained
+
 ARM26 ARCHITECTURE
 P:	Ian Molton
 M:	spyro@f2s.com
diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index 6be8fbec0a0..04d3120f723 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -322,6 +322,22 @@ config FB_FM2
 	  This is the frame buffer device driver for the Amiga FrameMaster
 	  card from BSC (exhibited 1992 but not shipped as a CBM product).
 
+config FB_ARC
+	tristate "Arc Monochrome LCD board support"
+	depends on FB && X86
+	select FB_CFB_FILLRECT
+	select FB_CFB_COPYAREA
+	select FB_CFB_IMAGEBLIT
+	select FB_SOFT_CURSOR
+	help
+	  This enables support for the Arc Monochrome LCD board. The board
+	  is based on the KS-108 lcd controller and is typically a matrix
+	  of 2*n chips. This driver was tested with a 128x64 panel. This
+	  driver supports it for use with x86 SBCs through a 16 bit GPIO
+	  interface (8 bit data, 8 bit control). If you anticpate using
+	  this driver, say Y or M; otherwise say N. You must specify the
+	  GPIO IO address to be used for setting control and data.
+
 config FB_ATARI
 	bool "Atari native chipset support"
 	depends on (FB = y) && ATARI && BROKEN
diff --git a/drivers/video/Makefile b/drivers/video/Makefile
index bd8dc0ffe72..b018df4e95c 100644
--- a/drivers/video/Makefile
+++ b/drivers/video/Makefile
@@ -21,6 +21,7 @@ obj-$(CONFIG_FB_MACMODES)      += macmodes.o
 # Hardware specific drivers go first
 obj-$(CONFIG_FB_RETINAZ3)         += retz3fb.o
 obj-$(CONFIG_FB_AMIGA)            += amifb.o c2p.o
+obj-$(CONFIG_FB_ARC)              += arcfb.o
 obj-$(CONFIG_FB_CLPS711X)         += clps711xfb.o
 obj-$(CONFIG_FB_CYBER)            += cyberfb.o
 obj-$(CONFIG_FB_CYBER2000)        += cyber2000fb.o
diff --git a/drivers/video/arcfb.c b/drivers/video/arcfb.c
new file mode 100644
index 00000000000..d28457e0c06
--- /dev/null
+++ b/drivers/video/arcfb.c
@@ -0,0 +1,684 @@
+/*
+ * linux/drivers/video/arcfb.c -- FB driver for Arc monochrome LCD board
+ *
+ * Copyright (C) 2005, Jaya Kumar <jayalk@intworks.biz>
+ * http://www.intworks.biz/arclcd
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file COPYING in the main directory of this archive for
+ * more details.
+ *
+ * Layout is based on skeletonfb.c by James Simmons and Geert Uytterhoeven.
+ *
+ * This driver was written to be used with the Arc LCD board. Arc uses a
+ * set of KS108 chips that control individual 64x64 LCD matrices. The board
+ * can be paneled in a variety of setups such as 2x1=128x64, 4x4=256x256 and
+ * so on. The interface between the board and the host is TTL based GPIO. The
+ * GPIO requirements are 8 writable data lines and 4+n lines for control. On a
+ * GPIO-less system, the board can be tested by connecting the respective sigs
+ * up to a parallel port connector. The driver requires the IO addresses for
+ * data and control GPIO at load time. It is unable to probe for the
+ * existence of the LCD so it must be told at load time whether it should
+ * be enabled or not.
+ *
+ * Todo:
+ * - testing with 4x4
+ * - testing with interrupt hw
+ *
+ * General notes:
+ * - User must set tuhold. It's in microseconds. According to the 108 spec,
+ *   the hold time is supposed to be at least 1 microsecond.
+ * - User must set num_cols=x num_rows=y, eg: x=2 means 128
+ * - User must set arcfb_enable=1 to enable it
+ * - User must set dio_addr=0xIOADDR cio_addr=0xIOADDR
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/mm.h>
+#include <linux/tty.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/fb.h>
+#include <linux/init.h>
+#include <linux/arcfb.h>
+
+#include <asm/uaccess.h>
+
+#define floor8(a) (a&(~0x07))
+#define floorXres(a,xres) (a&(~(xres - 1)))
+#define iceil8(a) (((int)((a+7)/8))*8)
+#define ceil64(a) (a|0x3F)
+#define ceilXres(a,xres) (a|(xres - 1))
+
+/* ks108 chipset specific defines and code */
+
+#define KS_SET_DPY_START_LINE 	0xC0
+#define KS_SET_PAGE_NUM 	0xB8
+#define KS_SET_X 		0x40
+#define KS_CEHI 		0x01
+#define KS_CELO 		0x00
+#define KS_SEL_CMD 		0x08
+#define KS_SEL_DATA 		0x00
+#define KS_DPY_ON 		0x3F
+#define KS_DPY_OFF 		0x3E
+#define KS_INTACK 		0x40
+#define KS_CLRINT		0x02
+
+struct arcfb_par {
+	unsigned long dio_addr;
+	unsigned long cio_addr;
+	unsigned long c2io_addr;
+	atomic_t ref_count;
+	unsigned char cslut[9];
+	struct fb_info *info;
+	unsigned int irq;
+	spinlock_t lock;
+};
+
+static struct fb_fix_screeninfo arcfb_fix __initdata = {
+	.id =		"arcfb",
+	.type =		FB_TYPE_PACKED_PIXELS,
+	.visual =	FB_VISUAL_MONO01,
+	.xpanstep =	0,
+	.ypanstep =	1,
+	.ywrapstep =	0,
+	.accel =	FB_ACCEL_NONE,
+};
+
+static struct fb_var_screeninfo arcfb_var __initdata = {
+	.xres		= 128,
+	.yres		= 64,
+	.xres_virtual	= 128,
+	.yres_virtual	= 64,
+	.bits_per_pixel	= 1,
+	.nonstd		= 1,
+};
+
+static unsigned long num_cols;
+static unsigned long num_rows;
+static unsigned long dio_addr;
+static unsigned long cio_addr;
+static unsigned long c2io_addr;
+static unsigned long splashval;
+static unsigned long tuhold;
+static unsigned int nosplash;
+static unsigned int arcfb_enable;
+static unsigned int irq;
+
+static DECLARE_WAIT_QUEUE_HEAD(arcfb_waitq);
+
+static void ks108_writeb_ctl(struct arcfb_par *par,
+				unsigned int chipindex, unsigned char value)
+{
+	unsigned char chipselval = par->cslut[chipindex];
+
+	outb(chipselval|KS_CEHI|KS_SEL_CMD, par->cio_addr);
+	outb(value, par->dio_addr);
+	udelay(tuhold);
+	outb(chipselval|KS_CELO|KS_SEL_CMD, par->cio_addr);
+}
+
+static void ks108_writeb_mainctl(struct arcfb_par *par, unsigned char value)
+{
+
+	outb(value, par->cio_addr);
+	udelay(tuhold);
+}
+
+static unsigned char ks108_readb_ctl2(struct arcfb_par *par)
+{
+	return inb(par->c2io_addr);
+}
+
+static void ks108_writeb_data(struct arcfb_par *par,
+				unsigned int chipindex, unsigned char value)
+{
+	unsigned char chipselval = par->cslut[chipindex];
+
+	outb(chipselval|KS_CEHI|KS_SEL_DATA, par->cio_addr);
+	outb(value, par->dio_addr);
+	udelay(tuhold);
+	outb(chipselval|KS_CELO|KS_SEL_DATA, par->cio_addr);
+}
+
+static void ks108_set_start_line(struct arcfb_par *par,
+				unsigned int chipindex, unsigned char y)
+{
+	ks108_writeb_ctl(par, chipindex, KS_SET_DPY_START_LINE|y);
+}
+
+static void ks108_set_yaddr(struct arcfb_par *par,
+				unsigned int chipindex, unsigned char y)
+{
+	ks108_writeb_ctl(par, chipindex, KS_SET_PAGE_NUM|y);
+}
+
+static void ks108_set_xaddr(struct arcfb_par *par,
+				unsigned int chipindex, unsigned char x)
+{
+	ks108_writeb_ctl(par, chipindex, KS_SET_X|x);
+}
+
+static void ks108_clear_lcd(struct arcfb_par *par, unsigned int chipindex)
+{
+	int i,j;
+
+	for (i = 0; i <= 8; i++) {
+		ks108_set_yaddr(par, chipindex, i);
+		ks108_set_xaddr(par, chipindex, 0);
+		for (j = 0; j < 64; j++) {
+			ks108_writeb_data(par, chipindex,
+				(unsigned char) splashval);
+		}
+	}
+}
+
+/* main arcfb functions */
+
+static int arcfb_open(struct fb_info *info, int user)
+{
+	struct arcfb_par *par = info->par;
+
+	atomic_inc(&par->ref_count);
+	return 0;
+}
+
+static int arcfb_release(struct fb_info *info, int user)
+{
+	struct arcfb_par *par = info->par;
+	int count = atomic_read(&par->ref_count);
+
+	if (!count)
+		return -EINVAL;
+	atomic_dec(&par->ref_count);
+	return 0;
+}
+
+static int arcfb_pan_display(struct fb_var_screeninfo *var,
+				struct fb_info *info)
+{
+	int i;
+	struct arcfb_par *par = info->par;
+
+	if ((var->vmode & FB_VMODE_YWRAP) && (var->yoffset < 64)
+		&& (info->var.yres <= 64)) {
+		for (i = 0; i < num_cols; i++) {
+			ks108_set_start_line(par, i, var->yoffset);
+		}
+		info->var.yoffset = var->yoffset;
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
+static irqreturn_t arcfb_interrupt(int vec, void *dev_instance,
+		struct pt_regs *regs)
+{
+	struct fb_info *info = dev_instance;
+	unsigned char ctl2status;
+	struct arcfb_par *par = info->par;
+
+	ctl2status = ks108_readb_ctl2(par);
+
+	if (!(ctl2status & KS_INTACK)) /* not arc generated interrupt */
+		return IRQ_NONE;
+
+	ks108_writeb_mainctl(par, KS_CLRINT);
+
+	spin_lock(&par->lock);
+        if (waitqueue_active(&arcfb_waitq)) {
+                wake_up(&arcfb_waitq);
+        }
+	spin_unlock(&par->lock);
+
+	return IRQ_HANDLED;
+}
+
+/*
+ * here we handle a specific page on the lcd. the complexity comes from
+ * the fact that the fb is laidout in 8xX vertical columns. we extract
+ * each write of 8 vertical pixels. then we shift out as we move along
+ * X. That's what rightshift does. bitmask selects the desired input bit.
+ */
+static void arcfb_lcd_update_page(struct arcfb_par *par, unsigned int upper,
+		unsigned int left, unsigned int right, unsigned int distance)
+{
+	unsigned char *src;
+	unsigned int xindex, yindex, chipindex, linesize;
+	int i, count;
+	unsigned char val;
+	unsigned char bitmask, rightshift;
+
+	xindex = left >> 6;
+	yindex = upper >> 6;
+	chipindex = (xindex + (yindex*num_cols));
+
+	ks108_set_yaddr(par, chipindex, upper/8);
+
+	linesize = par->info->var.xres/8;
+	src = par->info->screen_base + (left/8) + (upper * linesize);
+	ks108_set_xaddr(par, chipindex, left);
+
+	bitmask=1;
+	rightshift=0;
+	while (left <= right) {
+		val = 0;
+		for (i = 0; i < 8; i++) {
+			if ( i > rightshift) {
+				val |= (*(src + (i*linesize)) & bitmask)
+						<< (i - rightshift);
+			} else {
+				val |= (*(src + (i*linesize)) & bitmask)
+						 >> (rightshift - i);
+			}
+		}
+		ks108_writeb_data(par, chipindex, val);
+		left++;
+		count++;
+		if (bitmask == 0x80) {
+			bitmask = 1;
+			src++;
+			rightshift=0;
+		} else {
+			bitmask <<= 1;
+			rightshift++;
+		}
+	}
+}
+
+/*
+ * here we handle the entire vertical page of the update. we write across
+ * lcd chips. update_page uses the upper/left values to decide which
+ * chip to select for the right. upper is needed for setting the page
+ * desired for the write.
+ */
+static void arcfb_lcd_update_vert(struct arcfb_par *par, unsigned int top,
+		unsigned int bottom, unsigned int left, unsigned int right)
+{
+	unsigned int distance, upper, lower;
+
+	distance = (bottom - top) + 1;
+	upper = top;
+	lower = top + 7;
+
+	while (distance > 0) {
+		distance -= 8;
+		arcfb_lcd_update_page(par, upper, left, right, 8);
+		upper = lower + 1;
+		lower = upper + 7;
+	}
+}
+
+/*
+ * here we handle horizontal blocks for the update. update_vert will
+ * handle spaning multiple pages. we break out each horizontal
+ * block in to individual blocks no taller than 64 pixels.
+ */
+static void arcfb_lcd_update_horiz(struct arcfb_par *par, unsigned int left,
+			unsigned int right, unsigned int top, unsigned int h)
+{
+	unsigned int distance, upper, lower;
+
+	distance = h;
+	upper = floor8(top);
+	lower = min(upper + distance - 1, ceil64(upper));
+
+	while (distance > 0) {
+		distance -= ((lower - upper) + 1 );
+		arcfb_lcd_update_vert(par, upper, lower, left, right);
+		upper = lower + 1;
+		lower = min(upper + distance - 1, ceil64(upper));
+	}
+}
+
+/*
+ * here we start the process of spliting out the fb update into
+ * individual blocks of pixels. we end up spliting into 64x64 blocks
+ * and finally down to 64x8 pages.
+ */
+static void arcfb_lcd_update(struct arcfb_par *par, unsigned int dx,
+			unsigned int dy, unsigned int w, unsigned int h)
+{
+	unsigned int left, right, distance, y;
+
+	/* align the request first */
+	y = floor8(dy);
+	h += dy - y;
+	h = iceil8(h);
+
+	distance = w;
+	left = dx;
+	right = min(left + w - 1, ceil64(left));
+
+	while (distance > 0) {
+		arcfb_lcd_update_horiz(par, left, right, y, h);
+		distance -= ((right - left) + 1);
+		left = right + 1;
+		right = min(left + distance - 1, ceil64(left));
+	}
+}
+
+void arcfb_fillrect(struct fb_info *info, const struct fb_fillrect *rect)
+{
+	struct arcfb_par *par = info->par;
+
+	cfb_fillrect(info, rect);
+
+	/* update the physical lcd */
+	arcfb_lcd_update(par, rect->dx, rect->dy, rect->width, rect->height);
+}
+
+void arcfb_copyarea(struct fb_info *info, const struct fb_copyarea *area)
+{
+	struct arcfb_par *par = info->par;
+
+	cfb_copyarea(info, area);
+
+	/* update the physical lcd */
+	arcfb_lcd_update(par, area->dx, area->dy, area->width, area->height);
+}
+
+void arcfb_imageblit(struct fb_info *info, const struct fb_image *image)
+{
+	struct arcfb_par *par = info->par;
+
+	cfb_imageblit(info, image);
+
+	/* update the physical lcd */
+	arcfb_lcd_update(par, image->dx, image->dy, image->width,
+				image->height);
+}
+
+static int arcfb_ioctl(struct inode *inode, struct file *file,
+			  unsigned int cmd, unsigned long arg,
+			  struct fb_info *info)
+{
+	void __user *argp = (void __user *)arg;
+	struct arcfb_par *par = info->par;
+	unsigned long flags;
+
+	switch (cmd) {
+		case FBIO_WAITEVENT:
+		{
+			DEFINE_WAIT(wait);
+			/* illegal to wait on arc if no irq will occur */
+			if (!par->irq)
+				return -EINVAL;
+
+			/* wait until the Arc has generated an interrupt
+			 * which will wake us up */
+			spin_lock_irqsave(&par->lock, flags);
+			prepare_to_wait(&arcfb_waitq, &wait,
+					TASK_INTERRUPTIBLE);
+			spin_unlock_irqrestore(&par->lock, flags);
+			schedule();
+			finish_wait(&arcfb_waitq, &wait);
+		}
+		case FBIO_GETCONTROL2:
+		{
+			unsigned char ctl2;
+
+			ctl2 = ks108_readb_ctl2(info->par);
+			if (copy_to_user(argp, &ctl2, sizeof(ctl2)))
+				return -EFAULT;
+			return 0;
+		}
+		default:
+			return -EINVAL;
+	}
+}
+
+/*
+ * this is the access path from userspace. they can seek and write to
+ * the fb. it's inefficient for them to do anything less than 64*8
+ * writes since we update the lcd in each write() anyway.
+ */
+static ssize_t arcfb_write(struct file *file, const char *buf, size_t count,
+				loff_t *ppos)
+{
+	/* modded from epson 1355 */
+
+	struct inode *inode;
+	int fbidx;
+	struct fb_info *info;
+	unsigned long p;
+	int err=-EINVAL;
+	unsigned int fbmemlength,x,y,w,h, bitppos, startpos, endpos, bitcount;
+	struct arcfb_par *par;
+	unsigned int xres;
+
+	p = *ppos;
+	inode = file->f_dentry->d_inode;
+	fbidx = iminor(inode);
+	info = registered_fb[fbidx];
+	par = info->par;
+
+	if (!info || !info->screen_base)
+		return -ENODEV;
+
+	xres = info->var.xres;
+	fbmemlength = (xres * info->var.yres)/8;
+
+	if (p > fbmemlength)
+		return -ENOSPC;
+
+	err = 0;
+	if ((count + p) > fbmemlength) {
+		count = fbmemlength - p;
+		err = -ENOSPC;
+	}
+
+	if (count) {
+		char *base_addr;
+
+		base_addr = info->screen_base;
+		count -= copy_from_user(base_addr + p, buf, count);
+		*ppos += count;
+		err = -EFAULT;
+	}
+
+
+	bitppos = p*8;
+	startpos = floorXres(bitppos, xres);
+	endpos = ceilXres((bitppos + (count*8)), xres);
+	bitcount = endpos - startpos;
+
+	x = startpos % xres;
+	y = startpos / xres;
+	w = xres;
+	h = bitcount / xres;
+	arcfb_lcd_update(par, x, y, w, h);
+
+	if (count)
+		return count;
+	return err;
+}
+
+static void arcfb_platform_release(struct device *device)
+{
+}
+
+static struct fb_ops arcfb_ops = {
+	.owner		= THIS_MODULE,
+	.fb_open	= arcfb_open,
+	.fb_write	= arcfb_write,
+	.fb_release	= arcfb_release,
+	.fb_pan_display	= arcfb_pan_display,
+	.fb_fillrect	= arcfb_fillrect,
+	.fb_copyarea	= arcfb_copyarea,
+	.fb_imageblit	= arcfb_imageblit,
+	.fb_cursor	= soft_cursor,
+	.fb_ioctl 	= arcfb_ioctl,
+};
+
+static int __init arcfb_probe(struct device *device)
+{
+	struct platform_device *dev = to_platform_device(device);
+	struct fb_info *info;
+	int retval = -ENOMEM;
+	int videomemorysize;
+	unsigned char *videomemory;
+	struct arcfb_par *par;
+	int i;
+
+	videomemorysize = (((64*64)*num_cols)*num_rows)/8;
+
+	/* We need a flat backing store for the Arc's
+	   less-flat actual paged framebuffer */
+	if (!(videomemory = vmalloc(videomemorysize)))
+		return retval;
+
+	memset(videomemory, 0, videomemorysize);
+
+	info = framebuffer_alloc(sizeof(struct arcfb_par), &dev->dev);
+	if (!info)
+		goto err;
+
+	info->screen_base = (char __iomem *)videomemory;
+	info->fbops = &arcfb_ops;
+
+	info->var = arcfb_var;
+	info->fix = arcfb_fix;
+	par = info->par;
+	par->info = info;
+
+	if (!dio_addr || !cio_addr || !c2io_addr) {
+		printk(KERN_WARNING "no IO addresses supplied\n");
+		goto err1;
+	}
+	par->dio_addr = dio_addr;
+	par->cio_addr = cio_addr;
+	par->c2io_addr = c2io_addr;
+	par->cslut[0] = 0x00;
+	par->cslut[1] = 0x06;
+	info->flags = FBINFO_FLAG_DEFAULT;
+	spin_lock_init(&par->lock);
+	retval = register_framebuffer(info);
+	if (retval < 0)
+		goto err1;
+	dev_set_drvdata(&dev->dev, info);
+	if (irq) {
+		par->irq = irq;
+		if (request_irq(par->irq, &arcfb_interrupt, SA_SHIRQ,
+				"arcfb", info)) {
+			printk(KERN_INFO
+				"arcfb: Failed req IRQ %d\n", par->irq);
+			goto err1;
+		}
+	}
+	printk(KERN_INFO
+	       "fb%d: Arc frame buffer device, using %dK of video memory\n",
+	       info->node, videomemorysize >> 10);
+
+	/* this inits the lcd but doesn't clear dirty pixels */
+	for (i = 0; i < num_cols * num_rows; i++) {
+		ks108_writeb_ctl(par, i, KS_DPY_OFF);
+		ks108_set_start_line(par, i, 0);
+		ks108_set_yaddr(par, i, 0);
+		ks108_set_xaddr(par, i, 0);
+		ks108_writeb_ctl(par, i, KS_DPY_ON);
+	}
+
+	/* if we were told to splash the screen, we just clear it */
+	if (!nosplash) {
+		for (i = 0; i < num_cols * num_rows; i++) {
+			printk(KERN_INFO "fb%d: splashing lcd %d\n",
+				info->node, i);
+			ks108_set_start_line(par, i, 0);
+			ks108_clear_lcd(par, i);
+		}
+	}
+
+	return 0;
+err1:
+	framebuffer_release(info);
+err:
+	vfree(videomemory);
+	return retval;
+}
+
+static int arcfb_remove(struct device *device)
+{
+	struct fb_info *info = dev_get_drvdata(device);
+
+	if (info) {
+		unregister_framebuffer(info);
+		vfree(info->screen_base);
+		framebuffer_release(info);
+	}
+	return 0;
+}
+
+static struct device_driver arcfb_driver = {
+	.name	= "arcfb",
+	.bus	= &platform_bus_type,
+	.probe	= arcfb_probe,
+	.remove = arcfb_remove,
+};
+
+static struct platform_device arcfb_device = {
+	.name	= "arcfb",
+	.id	= 0,
+	.dev	= {
+		.release = arcfb_platform_release,
+	}
+};
+
+static int __init arcfb_init(void)
+{
+	int ret;
+
+	if (!arcfb_enable)
+		return -ENXIO;
+
+	ret = driver_register(&arcfb_driver);
+	if (!ret) {
+		ret = platform_device_register(&arcfb_device);
+		if (ret)
+			driver_unregister(&arcfb_driver);
+	}
+	return ret;
+
+}
+
+static void __exit arcfb_exit(void)
+{
+	platform_device_unregister(&arcfb_device);
+	driver_unregister(&arcfb_driver);
+}
+
+module_param(num_cols, ulong, 0);
+MODULE_PARM_DESC(num_cols, "Num horiz panels, eg: 2 = 128 bit wide");
+module_param(num_rows, ulong, 0);
+MODULE_PARM_DESC(num_rows, "Num vert panels, eg: 1 = 64 bit high");
+module_param(nosplash, uint, 0);
+MODULE_PARM_DESC(nosplash, "Disable doing the splash screen");
+module_param(arcfb_enable, uint, 0);
+MODULE_PARM_DESC(arcfb_enable, "Enable communication with Arc board");
+module_param(dio_addr, ulong, 0);
+MODULE_PARM_DESC(dio_addr, "IO address for data, eg: 0x480");
+module_param(cio_addr, ulong, 0);
+MODULE_PARM_DESC(cio_addr, "IO address for control, eg: 0x400");
+module_param(c2io_addr, ulong, 0);
+MODULE_PARM_DESC(c2io_addr, "IO address for secondary control, eg: 0x408");
+module_param(splashval, ulong, 0);
+MODULE_PARM_DESC(splashval, "Splash pattern: 0xFF is black, 0x00 is green");
+module_param(tuhold, ulong, 0);
+MODULE_PARM_DESC(tuhold, "Time to hold between strobing data to Arc board");
+module_param(irq, uint, 0);
+MODULE_PARM_DESC(irq, "IRQ for the Arc board");
+
+module_init(arcfb_init);
+module_exit(arcfb_exit);
+
+MODULE_DESCRIPTION("fbdev driver for Arc monochrome LCD board");
+MODULE_AUTHOR("Jaya Kumar");
+MODULE_LICENSE("GPL");
+
diff --git a/include/linux/arcfb.h b/include/linux/arcfb.h
new file mode 100644
index 00000000000..721e7654dae
--- /dev/null
+++ b/include/linux/arcfb.h
@@ -0,0 +1,8 @@
+#ifndef __LINUX_ARCFB_H__
+#define __LINUX_ARCFB_H__
+
+#define FBIO_WAITEVENT		_IO('F', 0x88)
+#define FBIO_GETCONTROL2	_IOR('F', 0x89, size_t)
+
+#endif
+
-- 
cgit v1.2.3-70-g09d2


From d5881eb4883ef7dd28a4dcea237714817c4b2f8e Mon Sep 17 00:00:00 2001
From: James Simmons <jsimmons@www.infradead.org>
Date: Tue, 21 Jun 2005 17:17:05 -0700
Subject: [PATCH] fbdev: new pci id for chipsfb

Patch adds pci ID for CT 69000 chipset.

Signed-off-by: James Simmons <jsimmons@www.infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/pci_ids.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index b8b4ebf9abf..63e89e47b8e 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -575,6 +575,7 @@
 #define PCI_DEVICE_ID_CT_65550		0x00e0
 #define PCI_DEVICE_ID_CT_65554		0x00e4
 #define PCI_DEVICE_ID_CT_65555		0x00e5
+#define PCI_DEVICE_ID_CT_69000		0x00c0
 
 #define PCI_VENDOR_ID_MIRO		0x1031
 #define PCI_DEVICE_ID_MIRO_36050	0x5601
-- 
cgit v1.2.3-70-g09d2


From 303b86d9913eca0cbfc3c5cb41e7006f6e13b755 Mon Sep 17 00:00:00 2001
From: Jurriaan <thunder7@xs4all.nl>
Date: Tue, 21 Jun 2005 17:17:06 -0700
Subject: [PATCH] New framebuffer fonts + updated 12x22 font available

Improve the fonts for use with the framebuffer.

I've added all the characters marked 'FIXME' in the sun12x22 font and
created a 10x18 font (based on the sun12x22 font) and a 7x14 font (based
on the vga8x16 font).

This patch is non-intrusive, no options are enabled by default so most
users won't notice a thing.

I am placing my changes under the GPL, however, I've not seen any copyright
notices on the sun12x22 font and the vga8x16 font which I derived my new
fonts from so I don't know what the copyright status is.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/console/Kconfig         |   17 +
 drivers/video/console/Makefile        |    2 +
 drivers/video/console/font_10x18.c    | 5146 +++++++++++++++++++++++++++++++++
 drivers/video/console/font_7x14.c     | 4118 ++++++++++++++++++++++++++
 drivers/video/console/font_sun12x22.c | 1579 +++++-----
 drivers/video/console/fonts.c         |    8 +
 include/linux/font.h                  |   26 +-
 7 files changed, 10068 insertions(+), 828 deletions(-)
 create mode 100644 drivers/video/console/font_10x18.c
 create mode 100644 drivers/video/console/font_7x14.c

(limited to 'include')

diff --git a/drivers/video/console/Kconfig b/drivers/video/console/Kconfig
index ccf55811d24..e4b91a4b936 100644
--- a/drivers/video/console/Kconfig
+++ b/drivers/video/console/Kconfig
@@ -153,6 +153,15 @@ config FONT_6x11
 	  Small console font with Macintosh-style high-half glyphs.  Some Mac
 	  framebuffer drivers don't support this one at all.
 
+config FONT_7x14
+	bool "console 7x14 font (not supported by all drivers)" if FONTS
+	depends on FRAMEBUFFER_CONSOLE
+	default y if !SPARC32 && !SPARC64 && !FONTS
+	help
+	  Console font with characters just a bit smaller than the default.
+	  If the standard 8x16 font is a little too big for you, say Y.
+	  Otherwise, say N.
+
 config FONT_PEARL_8x8
 	bool "Pearl (old m68k) console 8x8 font" if FONTS
 	depends on FRAMEBUFFER_CONSOLE
@@ -187,5 +196,13 @@ config FONT_SUN12x22
 	  big letters (like the letters used in the SPARC PROM). If the
 	  standard font is unreadable for you, say Y, otherwise say N.
 
+config FONT_10x18
+	bool "console 10x18 font (not supported by all drivers)"
+	depends on FONTS
+	help
+	  This is a high resolution console font for machines with very
+	  big letters. It fits between the sun 12x22 and the normal 8x16 font.
+	  If other fonts are too big or too small for you, say Y, otherwise say N.
+
 endmenu
 
diff --git a/drivers/video/console/Makefile b/drivers/video/console/Makefile
index 33516447f9f..b562f6bb9d3 100644
--- a/drivers/video/console/Makefile
+++ b/drivers/video/console/Makefile
@@ -10,6 +10,8 @@ font-objs-$(CONFIG_FONT_SUN12x22)  += font_sun12x22.o
 font-objs-$(CONFIG_FONT_8x8)       += font_8x8.o
 font-objs-$(CONFIG_FONT_8x16)      += font_8x16.o
 font-objs-$(CONFIG_FONT_6x11)      += font_6x11.o
+font-objs-$(CONFIG_FONT_7x14)      += font_7x14.o
+font-objs-$(CONFIG_FONT_10x18)     += font_10x18.o
 font-objs-$(CONFIG_FONT_PEARL_8x8) += font_pearl_8x8.o
 font-objs-$(CONFIG_FONT_ACORN_8x8) += font_acorn_8x8.o
 font-objs-$(CONFIG_FONT_MINI_4x6)  += font_mini_4x6.o
diff --git a/drivers/video/console/font_10x18.c b/drivers/video/console/font_10x18.c
new file mode 100644
index 00000000000..ff0af96e4df
--- /dev/null
+++ b/drivers/video/console/font_10x18.c
@@ -0,0 +1,5146 @@
+/********************************
+ * adapted from font_sun12x22.c *
+ * by Jurriaan Kalkman 06-2005  *
+ ********************************/
+
+#include <linux/font.h>
+
+#define FONTDATAMAX 9216
+
+static unsigned char fontdata_10x18[FONTDATAMAX] = {
+
+	/* 0 0x00 '^@' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 1 0x01 '^A' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x3f, 0x80, /* 0011111110 */
+	0x40, 0x40, /* 0100000001 */
+	0x5b, 0x40, /* 0101101101 */
+	0x40, 0x40, /* 0100000001 */
+	0x44, 0x40, /* 0100010001 */
+	0x44, 0x40, /* 0100010001 */
+	0x51, 0x40, /* 0101000101 */
+	0x4e, 0x40, /* 0100111001 */
+	0x40, 0x40, /* 0100000001 */
+	0x3f, 0x80, /* 0011111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 2 0x02 '^B' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x3f, 0x80, /* 0011111110 */
+	0x7f, 0xc0, /* 0111111111 */
+	0x64, 0xc0, /* 0110010011 */
+	0x7f, 0xc0, /* 0111111111 */
+	0x7b, 0xc0, /* 0111101111 */
+	0x7b, 0xc0, /* 0111101111 */
+	0x6e, 0xc0, /* 0110111011 */
+	0x71, 0xc0, /* 0111000111 */
+	0x7f, 0xc0, /* 0111111111 */
+	0x3f, 0x80, /* 0011111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 3 0x03 '^C' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x11, 0x00, /* 0001000100 */
+	0x3b, 0x80, /* 0011101110 */
+	0x7f, 0xc0, /* 0111111111 */
+	0x3f, 0x80, /* 0011111110 */
+	0x3f, 0x80, /* 0011111110 */
+	0x1f, 0x00, /* 0001111100 */
+	0x1f, 0x00, /* 0001111100 */
+	0x0e, 0x00, /* 0000111000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x04, 0x00, /* 0000010000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 4 0x04 '^D' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x04, 0x00, /* 0000010000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x1f, 0x00, /* 0001111100 */
+	0x3f, 0x80, /* 0011111110 */
+	0x1f, 0x00, /* 0001111100 */
+	0x1f, 0x00, /* 0001111100 */
+	0x0e, 0x00, /* 0000111000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x04, 0x00, /* 0000010000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 5 0x05 '^E' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x31, 0x80, /* 0011000110 */
+	0x7b, 0xc0, /* 0111101111 */
+	0x35, 0x80, /* 0011010110 */
+	0x04, 0x00, /* 0000010000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 6 0x06 '^F' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x04, 0x00, /* 0000010000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x1f, 0x00, /* 0001111100 */
+	0x3f, 0x80, /* 0011111110 */
+	0x7f, 0xc0, /* 0111111111 */
+	0x7f, 0xc0, /* 0111111111 */
+	0x35, 0x80, /* 0011010110 */
+	0x04, 0x00, /* 0000010000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 7 0x07 '^G' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 8 0x08 '^H' */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xf3, 0xc0, /* 1111001111 */
+	0xe1, 0xc0, /* 1110000111 */
+	0xe1, 0xc0, /* 1110000111 */
+	0xc0, 0xc0, /* 1100000011 */
+	0xc0, 0xc0, /* 1100000011 */
+	0xe1, 0xc0, /* 1110000111 */
+	0xe1, 0xc0, /* 1110000111 */
+	0xf3, 0xc0, /* 1111001111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+
+	/* 9 0x09 '^I' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x12, 0x00, /* 0001001000 */
+	0x12, 0x00, /* 0001001000 */
+	0x21, 0x00, /* 0010000100 */
+	0x21, 0x00, /* 0010000100 */
+	0x12, 0x00, /* 0001001000 */
+	0x12, 0x00, /* 0001001000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 10 0x0a '^J' */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xf3, 0xc0, /* 1111001111 */
+	0xed, 0xc0, /* 1110110111 */
+	0xed, 0xc0, /* 1110110111 */
+	0xde, 0xc0, /* 1101111011 */
+	0xde, 0xc0, /* 1101111011 */
+	0xed, 0xc0, /* 1110110111 */
+	0xed, 0xc0, /* 1110110111 */
+	0xf3, 0xc0, /* 1111001111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+
+	/* 11 0x0b '^K' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x03, 0xc0, /* 0000001111 */
+	0x06, 0xc0, /* 0000011011 */
+	0x0c, 0xc0, /* 0000110011 */
+	0x3c, 0x00, /* 0011110000 */
+	0x66, 0x00, /* 0110011000 */
+	0xc3, 0x00, /* 1100001100 */
+	0xc3, 0x00, /* 1100001100 */
+	0xc3, 0x00, /* 1100001100 */
+	0x66, 0x00, /* 0110011000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 12 0x0c '^L' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x33, 0x00, /* 0011001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x33, 0x00, /* 0011001100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 13 0x0d '^M' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0f, 0x80, /* 0000111110 */
+	0x08, 0x80, /* 0000100010 */
+	0x0f, 0x80, /* 0000111110 */
+	0x08, 0x00, /* 0000100000 */
+	0x08, 0x00, /* 0000100000 */
+	0x08, 0x00, /* 0000100000 */
+	0x08, 0x00, /* 0000100000 */
+	0x08, 0x00, /* 0000100000 */
+	0x08, 0x00, /* 0000100000 */
+	0x38, 0x00, /* 0011100000 */
+	0x78, 0x00, /* 0111100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 14 0x0e '^N' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0x80, /* 0001111110 */
+	0x10, 0x80, /* 0001000010 */
+	0x1f, 0x80, /* 0001111110 */
+	0x10, 0x80, /* 0001000010 */
+	0x10, 0x80, /* 0001000010 */
+	0x10, 0x80, /* 0001000010 */
+	0x10, 0x80, /* 0001000010 */
+	0x13, 0x80, /* 0001001110 */
+	0x17, 0x80, /* 0001011110 */
+	0x73, 0x00, /* 0111001100 */
+	0xf0, 0x00, /* 1111000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 15 0x0f '^O' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x04, 0x00, /* 0000010000 */
+	0x24, 0x80, /* 0010010010 */
+	0x15, 0x00, /* 0001010100 */
+	0x55, 0x40, /* 0101010101 */
+	0x3f, 0x80, /* 0011111110 */
+	0x0e, 0x00, /* 0000111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x3f, 0x80, /* 0011111110 */
+	0x55, 0x40, /* 0101010101 */
+	0x15, 0x00, /* 0001010100 */
+	0x24, 0x80, /* 0010010010 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 16 0x10 '^P' */
+	0x00, 0x80, /* 0000000010 */
+	0x01, 0x80, /* 0000000110 */
+	0x03, 0x80, /* 0000001110 */
+	0x07, 0x80, /* 0000011110 */
+	0x0f, 0x80, /* 0000111110 */
+	0x1f, 0x80, /* 0001111110 */
+	0x3f, 0x80, /* 0011111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0xff, 0x80, /* 1111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x3f, 0x80, /* 0011111110 */
+	0x1f, 0x80, /* 0001111110 */
+	0x0f, 0x80, /* 0000111110 */
+	0x07, 0x80, /* 0000011110 */
+	0x03, 0x80, /* 0000001110 */
+	0x01, 0x80, /* 0000000110 */
+	0x00, 0x80, /* 0000000010 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 17 0x11 '^Q' */
+	0x40, 0x00, /* 0100000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x70, 0x00, /* 0111000000 */
+	0x78, 0x00, /* 0111100000 */
+	0x7c, 0x00, /* 0111110000 */
+	0x7e, 0x00, /* 0111111000 */
+	0x7f, 0x00, /* 0111111100 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0xc0, /* 0111111111 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x00, /* 0111111100 */
+	0x7e, 0x00, /* 0111111000 */
+	0x7c, 0x00, /* 0111110000 */
+	0x78, 0x00, /* 0111100000 */
+	0x70, 0x00, /* 0111000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x40, 0x00, /* 0100000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 18 0x12 '^R' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x7f, 0x80, /* 0111111110 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x3f, 0x00, /* 0011111100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 19 0x13 '^S' */
+	0x00, 0x00, /* 0000000000 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 20 0x14 '^T' */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0xc0, /* 0001111111 */
+	0x39, 0x80, /* 0011100110 */
+	0x79, 0x80, /* 0111100110 */
+	0x79, 0x80, /* 0111100110 */
+	0x79, 0x80, /* 0111100110 */
+	0x39, 0x80, /* 0011100110 */
+	0x19, 0x80, /* 0001100110 */
+	0x19, 0x80, /* 0001100110 */
+	0x19, 0x80, /* 0001100110 */
+	0x19, 0x80, /* 0001100110 */
+	0x19, 0x80, /* 0001100110 */
+	0x19, 0x80, /* 0001100110 */
+	0x39, 0xc0, /* 0011100111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 21 0x15 '^U' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x3e, 0x00, /* 0011111000 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x3e, 0x00, /* 0011111000 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x3e, 0x00, /* 0011111000 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x3e, 0x00, /* 0011111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 22 0x16 '^V' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 23 0x17 '^W' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x7f, 0x80, /* 0111111110 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x3f, 0x00, /* 0011111100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 24 0x18 '^X' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x7f, 0x80, /* 0111111110 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 25 0x19 '^Y' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x3f, 0x00, /* 0011111100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 26 0x1a '^Z' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x04, 0x00, /* 0000010000 */
+	0x06, 0x00, /* 0000011000 */
+	0x07, 0x00, /* 0000011100 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x07, 0x00, /* 0000011100 */
+	0x06, 0x00, /* 0000011000 */
+	0x04, 0x00, /* 0000010000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 27 0x1b '^[' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x08, 0x00, /* 0000100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x38, 0x00, /* 0011100000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x38, 0x00, /* 0011100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x08, 0x00, /* 0000100000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 28 0x1c '^\' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 29 0x1d '^]' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x12, 0x00, /* 0001001000 */
+	0x33, 0x00, /* 0011001100 */
+	0x73, 0x80, /* 0111001110 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x73, 0x80, /* 0111001110 */
+	0x33, 0x00, /* 0011001100 */
+	0x12, 0x00, /* 0001001000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 30 0x1e '^^' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x04, 0x00, /* 0000010000 */
+	0x04, 0x00, /* 0000010000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x1f, 0x00, /* 0001111100 */
+	0x3f, 0x80, /* 0011111110 */
+	0x3f, 0x80, /* 0011111110 */
+	0x7f, 0xc0, /* 0111111111 */
+	0x7f, 0xc0, /* 0111111111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 31 0x1f '^_' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0xc0, /* 0111111111 */
+	0x7f, 0xc0, /* 0111111111 */
+	0x3f, 0x80, /* 0011111110 */
+	0x3f, 0x80, /* 0011111110 */
+	0x1f, 0x00, /* 0001111100 */
+	0x1f, 0x00, /* 0001111100 */
+	0x0e, 0x00, /* 0000111000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x04, 0x00, /* 0000010000 */
+	0x04, 0x00, /* 0000010000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 32 0x20 ' ' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 33 0x21 '!' */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 34 0x22 '"' */
+	0x00, 0x00, /* 0000000000 */
+	0x63, 0x00, /* 0110001100 */
+	0xf7, 0x80, /* 1111011110 */
+	0xf7, 0x80, /* 1111011110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x63, 0x00, /* 0110001100 */
+	0x42, 0x00, /* 0100001000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 35 0x23 '#' */
+	0x00, 0x00, /* 0000000000 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 36 0x24 '$' */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x3e, 0x00, /* 0011111000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x6f, 0x80, /* 0110111110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6c, 0x80, /* 0110110010 */
+	0x3c, 0x00, /* 0011110000 */
+	0x0f, 0x00, /* 0000111100 */
+	0x0d, 0x80, /* 0000110110 */
+	0x4d, 0x80, /* 0100110110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x7f, 0x00, /* 0111111100 */
+	0x3e, 0x00, /* 0011111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 37 0x25 '%' */
+	0x00, 0x00, /* 0000000000 */
+	0x31, 0x80, /* 0011000110 */
+	0x7b, 0x00, /* 0111101100 */
+	0x7b, 0x00, /* 0111101100 */
+	0x36, 0x00, /* 0011011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x1b, 0x00, /* 0001101100 */
+	0x37, 0x80, /* 0011011110 */
+	0x37, 0x80, /* 0011011110 */
+	0x63, 0x00, /* 0110001100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 38 0x26 '&' */
+	0x00, 0x00, /* 0000000000 */
+	0x07, 0x00, /* 0000011100 */
+	0x0f, 0x80, /* 0000111110 */
+	0x19, 0x80, /* 0001100110 */
+	0x19, 0x80, /* 0001100110 */
+	0x0f, 0x80, /* 0000111110 */
+	0x1e, 0x00, /* 0001111000 */
+	0x3e, 0x00, /* 0011111000 */
+	0x76, 0x00, /* 0111011000 */
+	0x66, 0x40, /* 0110011001 */
+	0x63, 0xc0, /* 0110001111 */
+	0x63, 0x80, /* 0110001110 */
+	0x63, 0x00, /* 0110001100 */
+	0x3f, 0x80, /* 0011111110 */
+	0x1c, 0xc0, /* 0001110011 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 39 0x27 ''' */
+	0x00, 0x00, /* 0000000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x78, 0x00, /* 0111100000 */
+	0x78, 0x00, /* 0111100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x20, 0x00, /* 0010000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 40 0x28 '(' */
+	0x00, 0x00, /* 0000000000 */
+	0x03, 0x00, /* 0000001100 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x06, 0x00, /* 0000011000 */
+	0x03, 0x00, /* 0000001100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 41 0x29 ')' */
+	0x00, 0x00, /* 0000000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 42 0x2a '*' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x4c, 0x80, /* 0100110010 */
+	0x6d, 0x80, /* 0110110110 */
+	0x3f, 0x00, /* 0011111100 */
+	0x7f, 0x80, /* 0111111110 */
+	0x3f, 0x00, /* 0011111100 */
+	0x6d, 0x80, /* 0110110110 */
+	0x4c, 0x80, /* 0100110010 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 43 0x2b '+' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 44 0x2c ',' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x78, 0x00, /* 0111100000 */
+	0x78, 0x00, /* 0111100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x40, 0x00, /* 0100000000 */
+
+	/* 45 0x2d '-' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 46 0x2e '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 47 0x2f '/' */
+	0x00, 0x00, /* 0000000000 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 48 0x30 '0' */
+	0x00, 0x00, /* 0000000000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x23, 0x00, /* 0010001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x63, 0x80, /* 0110001110 */
+	0x65, 0x80, /* 0110010110 */
+	0x65, 0x80, /* 0110010110 */
+	0x69, 0x80, /* 0110100110 */
+	0x69, 0x80, /* 0110100110 */
+	0x71, 0x80, /* 0111000110 */
+	0x61, 0x00, /* 0110000100 */
+	0x31, 0x00, /* 0011000100 */
+	0x3e, 0x00, /* 0011111000 */
+	0x1c, 0x00, /* 0001110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 49 0x31 '1' */
+	0x00, 0x00, /* 0000000000 */
+	0x04, 0x00, /* 0000010000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x1c, 0x00, /* 0001110000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x6c, 0x00, /* 0110110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 50 0x32 '2' */
+	0x00, 0x00, /* 0000000000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x63, 0x80, /* 0110001110 */
+	0x41, 0x80, /* 0100000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x03, 0x00, /* 0000001100 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x80, /* 0011000010 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 51 0x33 '3' */
+	0x00, 0x00, /* 0000000000 */
+	0x1c, 0x00, /* 0001110000 */
+	0x3e, 0x00, /* 0011111000 */
+	0x47, 0x00, /* 0100011100 */
+	0x03, 0x00, /* 0000001100 */
+	0x07, 0x00, /* 0000011100 */
+	0x06, 0x00, /* 0000011000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x07, 0x00, /* 0000011100 */
+	0x03, 0x00, /* 0000001100 */
+	0x01, 0x80, /* 0000000110 */
+	0x41, 0x80, /* 0100000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x3f, 0x00, /* 0011111100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 52 0x34 '4' */
+	0x00, 0x00, /* 0000000000 */
+	0x06, 0x00, /* 0000011000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x36, 0x00, /* 0011011000 */
+	0x36, 0x00, /* 0011011000 */
+	0x66, 0x00, /* 0110011000 */
+	0x66, 0x00, /* 0110011000 */
+	0xc6, 0x00, /* 1100011000 */
+	0xc6, 0x00, /* 1100011000 */
+	0xff, 0x80, /* 1111111110 */
+	0xff, 0x80, /* 1111111110 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 53 0x35 '5' */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x1f, 0x00, /* 0001111100 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x7e, 0x00, /* 0111111000 */
+	0x67, 0x00, /* 0110011100 */
+	0x03, 0x80, /* 0000001110 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x41, 0x80, /* 0100000110 */
+	0x63, 0x00, /* 0110001100 */
+	0x3e, 0x00, /* 0011111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 54 0x36 '6' */
+	0x00, 0x00, /* 0000000000 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x6e, 0x00, /* 0110111000 */
+	0x7f, 0x00, /* 0111111100 */
+	0x73, 0x80, /* 0111001110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x71, 0x00, /* 0111000100 */
+	0x3e, 0x00, /* 0011111000 */
+	0x1c, 0x00, /* 0001110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 55 0x37 '7' */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0x80, /* 0001111110 */
+	0x3f, 0x80, /* 0011111110 */
+	0x61, 0x80, /* 0110000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 56 0x38 '8' */
+	0x00, 0x00, /* 0000000000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x23, 0x00, /* 0010001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x31, 0x00, /* 0011000100 */
+	0x1a, 0x00, /* 0001101000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x16, 0x00, /* 0001011000 */
+	0x23, 0x00, /* 0010001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x31, 0x00, /* 0011000100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 57 0x39 '9' */
+	0x00, 0x00, /* 0000000000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x17, 0x00, /* 0001011100 */
+	0x23, 0x80, /* 0010001110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x73, 0x80, /* 0111001110 */
+	0x3d, 0x80, /* 0011110110 */
+	0x19, 0x80, /* 0001100110 */
+	0x01, 0x80, /* 0000000110 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 58 0x3a ':' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 59 0x3b ';' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x00, 0x00, /* 0000000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x20, 0x00, /* 0010000000 */
+
+	/* 60 0x3c '<' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x03, 0x00, /* 0000001100 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x06, 0x00, /* 0000011000 */
+	0x03, 0x00, /* 0000001100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 61 0x3d '=' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 62 0x3e '>' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x06, 0x00, /* 0000011000 */
+	0x03, 0x00, /* 0000001100 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x03, 0x00, /* 0000001100 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 63 0x3f '?' */
+	0x00, 0x00, /* 0000000000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x3b, 0x80, /* 0011101110 */
+	0x21, 0x80, /* 0010000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x03, 0x00, /* 0000001100 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 64 0x40 '@' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x31, 0x80, /* 0011000110 */
+	0x65, 0x80, /* 0110010110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6f, 0x80, /* 0110111110 */
+	0x60, 0x00, /* 0110000000 */
+	0x31, 0x80, /* 0011000110 */
+	0x3f, 0x80, /* 0011111110 */
+	0x0f, 0x00, /* 0000111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 65 0x41 'A' */
+	0x00, 0x00, /* 0000000000 */
+	0x04, 0x00, /* 0000010000 */
+	0x04, 0x00, /* 0000010000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x19, 0x80, /* 0001100110 */
+	0x31, 0x80, /* 0011000110 */
+	0x3f, 0x80, /* 0011111110 */
+	0x31, 0x80, /* 0011000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x60, 0xc0, /* 0110000011 */
+	0x60, 0xc0, /* 0110000011 */
+	0xf1, 0xc0, /* 1111000111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 66 0x42 'B' */
+	0x00, 0x00, /* 0000000000 */
+	0xfc, 0x00, /* 1111110000 */
+	0x62, 0x00, /* 0110001000 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x66, 0x00, /* 0110011000 */
+	0x7e, 0x00, /* 0111111000 */
+	0x63, 0x00, /* 0110001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x63, 0x00, /* 0110001100 */
+	0xfe, 0x00, /* 1111111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 67 0x43 'C' */
+	0x00, 0x00, /* 0000000000 */
+	0x0f, 0x00, /* 0000111100 */
+	0x11, 0x80, /* 0001000110 */
+	0x20, 0x80, /* 0010000010 */
+	0x20, 0x00, /* 0010000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x20, 0x00, /* 0010000000 */
+	0x30, 0x80, /* 0011000010 */
+	0x19, 0x00, /* 0001100100 */
+	0x0e, 0x00, /* 0000111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 68 0x44 'D' */
+	0x00, 0x00, /* 0000000000 */
+	0xfc, 0x00, /* 1111110000 */
+	0x67, 0x00, /* 0110011100 */
+	0x63, 0x00, /* 0110001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x00, /* 0110000100 */
+	0x66, 0x00, /* 0110011000 */
+	0xf8, 0x00, /* 1111100000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 69 0x45 'E' */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x30, 0x80, /* 0011000010 */
+	0x30, 0x80, /* 0011000010 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x31, 0x00, /* 0011000100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x31, 0x00, /* 0011000100 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x80, /* 0011000010 */
+	0x30, 0x80, /* 0011000010 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 70 0x46 'F' */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x30, 0x80, /* 0011000010 */
+	0x30, 0x80, /* 0011000010 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x31, 0x00, /* 0011000100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x31, 0x00, /* 0011000100 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x78, 0x00, /* 0111100000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 71 0x47 'G' */
+	0x00, 0x00, /* 0000000000 */
+	0x0f, 0x00, /* 0000111100 */
+	0x11, 0x80, /* 0001000110 */
+	0x20, 0x80, /* 0010000010 */
+	0x20, 0x00, /* 0010000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x67, 0xc0, /* 0110011111 */
+	0x61, 0x80, /* 0110000110 */
+	0x21, 0x80, /* 0010000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x19, 0x80, /* 0001100110 */
+	0x0e, 0x00, /* 0000111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 72 0x48 'H' */
+	0x00, 0x00, /* 0000000000 */
+	0xf3, 0xc0, /* 1111001111 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0xf3, 0xc0, /* 1111001111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 73 0x49 'I' */
+	0x00, 0x00, /* 0000000000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 74 0x4a 'J' */
+	0x00, 0x00, /* 0000000000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x08, 0x00, /* 0000100000 */
+	0x70, 0x00, /* 0111000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 75 0x4b 'K' */
+	0x00, 0x00, /* 0000000000 */
+	0xf1, 0x80, /* 1111000110 */
+	0x63, 0x00, /* 0110001100 */
+	0x66, 0x00, /* 0110011000 */
+	0x6c, 0x00, /* 0110110000 */
+	0x78, 0x00, /* 0111100000 */
+	0x70, 0x00, /* 0111000000 */
+	0x70, 0x00, /* 0111000000 */
+	0x78, 0x00, /* 0111100000 */
+	0x78, 0x00, /* 0111100000 */
+	0x6c, 0x00, /* 0110110000 */
+	0x66, 0x00, /* 0110011000 */
+	0x63, 0x00, /* 0110001100 */
+	0x61, 0x80, /* 0110000110 */
+	0xf0, 0xc0, /* 1111000011 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 76 0x4c 'L' */
+	0x00, 0x00, /* 0000000000 */
+	0x78, 0x00, /* 0111100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x80, /* 0011000010 */
+	0x30, 0x80, /* 0011000010 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 77 0x4d 'M' */
+	0x00, 0x00, /* 0000000000 */
+	0xe0, 0xc0, /* 1110000011 */
+	0x61, 0x80, /* 0110000110 */
+	0x73, 0x80, /* 0111001110 */
+	0x73, 0x80, /* 0111001110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0xf3, 0xc0, /* 1111001111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 78 0x4e 'N' */
+	0x00, 0x00, /* 0000000000 */
+	0xf3, 0xc0, /* 1111001111 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x71, 0x80, /* 0111000110 */
+	0x79, 0x80, /* 0111100110 */
+	0x79, 0x80, /* 0111100110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x67, 0x80, /* 0110011110 */
+	0x67, 0x80, /* 0110011110 */
+	0x63, 0x80, /* 0110001110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0xf3, 0xc0, /* 1111001111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 79 0x4f 'O' */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x17, 0x00, /* 0001011100 */
+	0x23, 0x00, /* 0010001100 */
+	0x21, 0x80, /* 0010000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x21, 0x00, /* 0010000100 */
+	0x31, 0x00, /* 0011000100 */
+	0x1a, 0x00, /* 0001101000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 80 0x50 'P' */
+	0x00, 0x00, /* 0000000000 */
+	0xfe, 0x00, /* 1111111000 */
+	0x63, 0x00, /* 0110001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x63, 0x00, /* 0110001100 */
+	0x7e, 0x00, /* 0111111000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0xf0, 0x00, /* 1111000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 81 0x51 'Q' */
+	0x00, 0x00, /* 0000000000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x13, 0x00, /* 0001001100 */
+	0x23, 0x00, /* 0010001100 */
+	0x21, 0x80, /* 0010000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x3b, 0x00, /* 0011101100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x26, 0x00, /* 0010011000 */
+	0x03, 0x80, /* 0000001110 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 82 0x52 'R' */
+	0x00, 0x00, /* 0000000000 */
+	0xfe, 0x00, /* 1111111000 */
+	0x63, 0x00, /* 0110001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x00, /* 0110000100 */
+	0x7e, 0x00, /* 0111111000 */
+	0x78, 0x00, /* 0111100000 */
+	0x6c, 0x00, /* 0110110000 */
+	0x6e, 0x00, /* 0110111000 */
+	0x67, 0x00, /* 0110011100 */
+	0x63, 0x80, /* 0110001110 */
+	0x61, 0xc0, /* 0110000111 */
+	0xf0, 0xc0, /* 1111000011 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 83 0x53 'S' */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x31, 0x80, /* 0011000110 */
+	0x60, 0x80, /* 0110000010 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x06, 0x00, /* 0000011000 */
+	0x03, 0x00, /* 0000001100 */
+	0x01, 0x80, /* 0000000110 */
+	0x41, 0x80, /* 0100000110 */
+	0x63, 0x00, /* 0110001100 */
+	0x3e, 0x00, /* 0011111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 84 0x54 'T' */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x4c, 0x80, /* 0100110010 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 85 0x55 'U' */
+	0x00, 0x00, /* 0000000000 */
+	0xf3, 0xc0, /* 1111001111 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x73, 0x00, /* 0111001100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 86 0x56 'V' */
+	0x00, 0x00, /* 0000000000 */
+	0xe1, 0xc0, /* 1110000111 */
+	0xc0, 0xc0, /* 1100000011 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x12, 0x00, /* 0001001000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 87 0x57 'W' */
+	0x00, 0x00, /* 0000000000 */
+	0xe1, 0xc0, /* 1110000111 */
+	0xc0, 0xc0, /* 1100000011 */
+	0xc0, 0xc0, /* 1100000011 */
+	0xc0, 0xc0, /* 1100000011 */
+	0xe0, 0xc0, /* 1110000011 */
+	0x61, 0x80, /* 0110000110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x77, 0x00, /* 0111011100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 88 0x58 'X' */
+	0x00, 0x00, /* 0000000000 */
+	0xf7, 0x80, /* 1111011110 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x36, 0x00, /* 0011011000 */
+	0x36, 0x00, /* 0011011000 */
+	0x36, 0x00, /* 0011011000 */
+	0x1c, 0x00, /* 0001110000 */
+	0x1c, 0x00, /* 0001110000 */
+	0x36, 0x00, /* 0011011000 */
+	0x36, 0x00, /* 0011011000 */
+	0x36, 0x00, /* 0011011000 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0xf7, 0x80, /* 1111011110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 89 0x59 'Y' */
+	0x00, 0x00, /* 0000000000 */
+	0xf3, 0xc0, /* 1111001111 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 90 0x5a 'Z' */
+	0x00, 0x00, /* 0000000000 */
+	0x3f, 0x80, /* 0011111110 */
+	0x21, 0x80, /* 0010000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x80, /* 0011000010 */
+	0x3f, 0x80, /* 0011111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 91 0x5b '[' */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x1f, 0x00, /* 0001111100 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x1f, 0x00, /* 0001111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 92 0x5c '\' */
+	0x00, 0x00, /* 0000000000 */
+	0xc0, 0x00, /* 1100000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x00, 0xc0, /* 0000000011 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 93 0x5d ']' */
+	0x00, 0x00, /* 0000000000 */
+	0x3e, 0x00, /* 0011111000 */
+	0x3e, 0x00, /* 0011111000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x3e, 0x00, /* 0011111000 */
+	0x3e, 0x00, /* 0011111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 94 0x5e '^' */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x33, 0x00, /* 0011001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 95 0x5f '_' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 96 0x60 '`' */
+	0x04, 0x00, /* 0000010000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 97 0x61 'a' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x31, 0x80, /* 0011000110 */
+	0x21, 0x80, /* 0010000110 */
+	0x07, 0x80, /* 0000011110 */
+	0x39, 0x80, /* 0011100110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x73, 0x80, /* 0111001110 */
+	0x3d, 0xc0, /* 0011110111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 98 0x62 'b' */
+	0x20, 0x00, /* 0010000000 */
+	0x60, 0x00, /* 0110000000 */
+	0xe0, 0x00, /* 1110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x66, 0x00, /* 0110011000 */
+	0x6f, 0x00, /* 0110111100 */
+	0x73, 0x80, /* 0111001110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x71, 0x80, /* 0111000110 */
+	0x7b, 0x00, /* 0111101100 */
+	0x4e, 0x00, /* 0100111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 99 0x63 'c' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x37, 0x00, /* 0011011100 */
+	0x23, 0x00, /* 0010001100 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x71, 0x00, /* 0111000100 */
+	0x33, 0x00, /* 0011001100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 100 0x64 'd' */
+	0x01, 0x80, /* 0000000110 */
+	0x03, 0x80, /* 0000001110 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x0d, 0x80, /* 0000110110 */
+	0x37, 0x80, /* 0011011110 */
+	0x23, 0x80, /* 0010001110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x73, 0x80, /* 0111001110 */
+	0x35, 0x80, /* 0011010110 */
+	0x19, 0xc0, /* 0001100111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 101 0x65 'e' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x33, 0x00, /* 0011001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x19, 0x80, /* 0001100110 */
+	0x0e, 0x00, /* 0000111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 102 0x66 'f' */
+	0x07, 0x00, /* 0000011100 */
+	0x09, 0x80, /* 0000100110 */
+	0x09, 0x80, /* 0000100110 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x7f, 0x00, /* 0111111100 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 103 0x67 'g' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1c, 0x80, /* 0001110010 */
+	0x37, 0x80, /* 0011011110 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x36, 0x00, /* 0011011000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x60, 0x00, /* 0110000000 */
+	0x7f, 0x00, /* 0111111100 */
+	0x3f, 0x80, /* 0011111110 */
+	0x21, 0x80, /* 0010000110 */
+	0x40, 0x80, /* 0100000010 */
+	0x7f, 0x00, /* 0111111100 */
+	0x3e, 0x00, /* 0011111000 */
+
+	/* 104 0x68 'h' */
+	0x10, 0x00, /* 0001000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x70, 0x00, /* 0111000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x37, 0x00, /* 0011011100 */
+	0x3b, 0x80, /* 0011101110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x7b, 0xc0, /* 0111101111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 105 0x69 'i' */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 106 0x6a 'j' */
+	0x00, 0x00, /* 0000000000 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x07, 0x80, /* 0000011110 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x41, 0x80, /* 0100000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x71, 0x80, /* 0111000110 */
+	0x3f, 0x00, /* 0011111100 */
+	0x1c, 0x00, /* 0001110000 */
+
+	/* 107 0x6b 'k' */
+	0x60, 0x00, /* 0110000000 */
+	0xe0, 0x00, /* 1110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x63, 0x80, /* 0110001110 */
+	0x66, 0x00, /* 0110011000 */
+	0x6c, 0x00, /* 0110110000 */
+	0x78, 0x00, /* 0111100000 */
+	0x70, 0x00, /* 0111000000 */
+	0x78, 0x00, /* 0111100000 */
+	0x6c, 0x00, /* 0110110000 */
+	0x6e, 0x00, /* 0110111000 */
+	0x67, 0x00, /* 0110011100 */
+	0xf3, 0x80, /* 1111001110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 108 0x6c 'l' */
+	0x3c, 0x00, /* 0011110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 109 0x6d 'm' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xdb, 0x80, /* 1101101110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6d, 0x80, /* 0110110110 */
+	0xed, 0xc0, /* 1110110111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 110 0x6e 'n' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x6f, 0x00, /* 0110111100 */
+	0x7b, 0x80, /* 0111101110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x7b, 0xc0, /* 0111101111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 111 0x6f 'o' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x66, 0x00, /* 0110011000 */
+	0xc3, 0x00, /* 1100001100 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xe1, 0x80, /* 1110000110 */
+	0x73, 0x00, /* 0111001100 */
+	0x3c, 0x00, /* 0011110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 112 0x70 'p' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xde, 0x00, /* 1101111000 */
+	0x76, 0x00, /* 0111011000 */
+	0x63, 0x00, /* 0110001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x71, 0x80, /* 0111000110 */
+	0x7b, 0x00, /* 0111101100 */
+	0x7e, 0x00, /* 0111111000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0xf0, 0x00, /* 1111000000 */
+
+	/* 113 0x71 'q' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0e, 0xc0, /* 0000111011 */
+	0x1b, 0x80, /* 0001101110 */
+	0x33, 0x80, /* 0011001110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x71, 0x80, /* 0111000110 */
+	0x3b, 0x80, /* 0011101110 */
+	0x1f, 0x80, /* 0001111110 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x03, 0xc0, /* 0000001111 */
+
+	/* 114 0x72 'r' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x73, 0x00, /* 0111001100 */
+	0x35, 0x80, /* 0011010110 */
+	0x39, 0x80, /* 0011100110 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x78, 0x00, /* 0111100000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 115 0x73 's' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x63, 0x00, /* 0110001100 */
+	0x61, 0x00, /* 0110000100 */
+	0x70, 0x00, /* 0111000000 */
+	0x38, 0x00, /* 0011100000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x07, 0x00, /* 0000011100 */
+	0x43, 0x00, /* 0100001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x7e, 0x00, /* 0111111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 116 0x74 't' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x08, 0x00, /* 0000100000 */
+	0x08, 0x00, /* 0000100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x1c, 0x80, /* 0001110010 */
+	0x0f, 0x00, /* 0000111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 117 0x75 'u' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xf7, 0x80, /* 1111011110 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x77, 0x00, /* 0111011100 */
+	0x3d, 0x80, /* 0011110110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 118 0x76 'v' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xf1, 0xc0, /* 1111000111 */
+	0x60, 0xc0, /* 0110000011 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x19, 0x80, /* 0001100110 */
+	0x1b, 0x00, /* 0001101100 */
+	0x0f, 0x00, /* 0000111100 */
+	0x0f, 0x00, /* 0000111100 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 119 0x77 'w' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xe3, 0xc0, /* 1110001111 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0x6b, 0x00, /* 0110101100 */
+	0x6b, 0x00, /* 0110101100 */
+	0x7e, 0x00, /* 0111111000 */
+	0x36, 0x00, /* 0011011000 */
+	0x36, 0x00, /* 0011011000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 120 0x78 'x' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xf7, 0x80, /* 1111011110 */
+	0x63, 0x00, /* 0110001100 */
+	0x36, 0x00, /* 0011011000 */
+	0x36, 0x00, /* 0011011000 */
+	0x1c, 0x00, /* 0001110000 */
+	0x1c, 0x00, /* 0001110000 */
+	0x36, 0x00, /* 0011011000 */
+	0x66, 0x00, /* 0110011000 */
+	0x63, 0x00, /* 0110001100 */
+	0xf7, 0x80, /* 1111011110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 121 0x79 'y' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xf3, 0xc0, /* 1111001111 */
+	0x61, 0x80, /* 0110000110 */
+	0x33, 0x00, /* 0011001100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x78, 0x00, /* 0111100000 */
+	0x70, 0x00, /* 0111000000 */
+
+	/* 122 0x7a 'z' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x61, 0x80, /* 0110000110 */
+	0x43, 0x00, /* 0100001100 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x60, 0x80, /* 0110000010 */
+	0x61, 0x80, /* 0110000110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 123 0x7b '{' */
+	0x07, 0x00, /* 0000011100 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x70, 0x00, /* 0111000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x07, 0x00, /* 0000011100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 124 0x7c '|' */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 125 0x7d '}' */
+	0x38, 0x00, /* 0011100000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x06, 0x00, /* 0000011000 */
+	0x03, 0x80, /* 0000001110 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x38, 0x00, /* 0011100000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 126 0x7e '~' */
+	0x00, 0x00, /* 0000000000 */
+	0x18, 0x80, /* 0001100010 */
+	0x3d, 0x80, /* 0011110110 */
+	0x6f, 0x00, /* 0110111100 */
+	0x46, 0x00, /* 0100011000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 127 0x7f '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x12, 0x00, /* 0001001000 */
+	0x21, 0x00, /* 0010000100 */
+	0x40, 0x80, /* 0100000010 */
+	0x40, 0x80, /* 0100000010 */
+	0x40, 0x80, /* 0100000010 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 128 0x80 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x21, 0x80, /* 0010000110 */
+	0x40, 0x80, /* 0100000010 */
+	0x40, 0x00, /* 0100000000 */
+	0x40, 0x00, /* 0100000000 */
+	0x40, 0x00, /* 0100000000 */
+	0x40, 0x00, /* 0100000000 */
+	0x40, 0x00, /* 0100000000 */
+	0x40, 0x00, /* 0100000000 */
+	0x60, 0x80, /* 0110000010 */
+	0x31, 0x00, /* 0011000100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x08, 0x00, /* 0000100000 */
+	0x04, 0x00, /* 0000010000 */
+	0x02, 0x00, /* 0000001000 */
+	0x02, 0x00, /* 0000001000 */
+	0x1c, 0x00, /* 0001110000 */
+
+	/* 129 0x81 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7b, 0x80, /* 0111101110 */
+	0x31, 0x00, /* 0011000100 */
+	0x31, 0x00, /* 0011000100 */
+	0x31, 0x00, /* 0011000100 */
+	0x31, 0x00, /* 0011000100 */
+	0x31, 0x00, /* 0011000100 */
+	0x31, 0x00, /* 0011000100 */
+	0x31, 0x00, /* 0011000100 */
+	0x3b, 0x00, /* 0011101100 */
+	0x1c, 0x80, /* 0001110010 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 130 0x82 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x01, 0x00, /* 0000000100 */
+	0x02, 0x00, /* 0000001000 */
+	0x04, 0x00, /* 0000010000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x33, 0x00, /* 0011001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x19, 0x80, /* 0001100110 */
+	0x0e, 0x00, /* 0000111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 131 0x83 '.' */
+	0x04, 0x00, /* 0000010000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x1b, 0x00, /* 0001101100 */
+	0x31, 0x80, /* 0011000110 */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x31, 0x80, /* 0011000110 */
+	0x21, 0x80, /* 0010000110 */
+	0x07, 0x80, /* 0000011110 */
+	0x39, 0x80, /* 0011100110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x73, 0x80, /* 0111001110 */
+	0x3d, 0xc0, /* 0011110111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 132 0x84 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x31, 0x80, /* 0011000110 */
+	0x21, 0x80, /* 0010000110 */
+	0x07, 0x80, /* 0000011110 */
+	0x39, 0x80, /* 0011100110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x73, 0x80, /* 0111001110 */
+	0x3d, 0xc0, /* 0011110111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 133 0x85 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x06, 0x00, /* 0000011000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x31, 0x80, /* 0011000110 */
+	0x21, 0x80, /* 0010000110 */
+	0x07, 0x80, /* 0000011110 */
+	0x39, 0x80, /* 0011100110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x73, 0x80, /* 0111001110 */
+	0x3d, 0xc0, /* 0011110111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 134 0x86 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x1b, 0x00, /* 0001101100 */
+	0x0e, 0x00, /* 0000111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x31, 0x80, /* 0011000110 */
+	0x21, 0x80, /* 0010000110 */
+	0x07, 0x80, /* 0000011110 */
+	0x39, 0x80, /* 0011100110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x73, 0x80, /* 0111001110 */
+	0x3d, 0xc0, /* 0011110111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 135 0x87 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x31, 0x80, /* 0011000110 */
+	0x20, 0x80, /* 0010000010 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x70, 0x80, /* 0111000010 */
+	0x30, 0x80, /* 0011000010 */
+	0x1f, 0x00, /* 0001111100 */
+	0x04, 0x00, /* 0000010000 */
+	0x02, 0x00, /* 0000001000 */
+	0x01, 0x00, /* 0000000100 */
+	0x0e, 0x00, /* 0000111000 */
+
+	/* 136 0x88 '.' */
+	0x04, 0x00, /* 0000010000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x1b, 0x00, /* 0001101100 */
+	0x31, 0x80, /* 0011000110 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x33, 0x00, /* 0011001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x19, 0x80, /* 0001100110 */
+	0x0e, 0x00, /* 0000111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 137 0x89 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x36, 0x00, /* 0011011000 */
+	0x36, 0x00, /* 0011011000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x33, 0x00, /* 0011001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x19, 0x80, /* 0001100110 */
+	0x0e, 0x00, /* 0000111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 138 0x8a '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x33, 0x00, /* 0011001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x19, 0x80, /* 0001100110 */
+	0x0e, 0x00, /* 0000111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 139 0x8b '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x36, 0x00, /* 0011011000 */
+	0x36, 0x00, /* 0011011000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 140 0x8c '.' */
+	0x08, 0x00, /* 0000100000 */
+	0x1c, 0x00, /* 0001110000 */
+	0x36, 0x00, /* 0011011000 */
+	0x63, 0x00, /* 0110001100 */
+	0x00, 0x00, /* 0000000000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 141 0x8d '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 142 0x8e '.' */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x04, 0x00, /* 0000010000 */
+	0x04, 0x00, /* 0000010000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x19, 0x00, /* 0001100100 */
+	0x19, 0x00, /* 0001100100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x31, 0x00, /* 0011000100 */
+	0x31, 0x00, /* 0011000100 */
+	0x60, 0x80, /* 0110000010 */
+	0x60, 0x80, /* 0110000010 */
+	0xf3, 0xc0, /* 1111001111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 143 0x8f '.' */
+	0x04, 0x00, /* 0000010000 */
+	0x0a, 0x00, /* 0000101000 */
+	0x0a, 0x00, /* 0000101000 */
+	0x04, 0x00, /* 0000010000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x19, 0x00, /* 0001100100 */
+	0x19, 0x00, /* 0001100100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x31, 0x00, /* 0011000100 */
+	0x31, 0x00, /* 0011000100 */
+	0x60, 0x80, /* 0110000010 */
+	0x60, 0x80, /* 0110000010 */
+	0xf3, 0xc0, /* 1111001111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 144 0x90 '.' */
+	0x03, 0x00, /* 0000001100 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x30, 0x80, /* 0011000010 */
+	0x30, 0x80, /* 0011000010 */
+	0x30, 0x00, /* 0011000000 */
+	0x31, 0x00, /* 0011000100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x31, 0x00, /* 0011000100 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x80, /* 0011000010 */
+	0x30, 0x80, /* 0011000010 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 145 0x91 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x3b, 0x80, /* 0011101110 */
+	0x6c, 0xc0, /* 0110110011 */
+	0x4c, 0xc0, /* 0100110011 */
+	0x0c, 0xc0, /* 0000110011 */
+	0x3f, 0xc0, /* 0011111111 */
+	0x6c, 0x00, /* 0110110000 */
+	0xcc, 0x00, /* 1100110000 */
+	0xcc, 0x00, /* 1100110000 */
+	0xee, 0xc0, /* 1110111011 */
+	0x7b, 0x80, /* 0111101110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 146 0x92 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x07, 0xc0, /* 0000011111 */
+	0x0e, 0x40, /* 0000111001 */
+	0x0e, 0x40, /* 0000111001 */
+	0x0e, 0x00, /* 0000111000 */
+	0x16, 0x00, /* 0001011000 */
+	0x16, 0x80, /* 0001011010 */
+	0x17, 0x80, /* 0001011110 */
+	0x16, 0x80, /* 0001011010 */
+	0x3e, 0x00, /* 0011111000 */
+	0x26, 0x00, /* 0010011000 */
+	0x26, 0x00, /* 0010011000 */
+	0x46, 0x40, /* 0100011001 */
+	0x46, 0x40, /* 0100011001 */
+	0xef, 0xc0, /* 1110111111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 147 0x93 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x08, 0x00, /* 0000100000 */
+	0x1c, 0x00, /* 0001110000 */
+	0x36, 0x00, /* 0011011000 */
+	0x00, 0x00, /* 0000000000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x66, 0x00, /* 0110011000 */
+	0xc3, 0x00, /* 1100001100 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xe1, 0x80, /* 1110000110 */
+	0x73, 0x00, /* 0111001100 */
+	0x3c, 0x00, /* 0011110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 148 0x94 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x00, 0x00, /* 0000000000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x66, 0x00, /* 0110011000 */
+	0xc3, 0x00, /* 1100001100 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xe1, 0x80, /* 1110000110 */
+	0x73, 0x00, /* 0111001100 */
+	0x3c, 0x00, /* 0011110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 149 0x95 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x06, 0x00, /* 0000011000 */
+	0x00, 0x00, /* 0000000000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x66, 0x00, /* 0110011000 */
+	0xc3, 0x00, /* 1100001100 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xe1, 0x80, /* 1110000110 */
+	0x73, 0x00, /* 0111001100 */
+	0x3c, 0x00, /* 0011110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 150 0x96 '.' */
+	0x08, 0x00, /* 0000100000 */
+	0x1c, 0x00, /* 0001110000 */
+	0x36, 0x00, /* 0011011000 */
+	0x63, 0x00, /* 0110001100 */
+	0x00, 0x00, /* 0000000000 */
+	0xf7, 0x80, /* 1111011110 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x77, 0x00, /* 0111011100 */
+	0x3d, 0x80, /* 0011110110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 151 0x97 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x00, 0x00, /* 0000000000 */
+	0xf7, 0x80, /* 1111011110 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x77, 0x00, /* 0111011100 */
+	0x3d, 0x80, /* 0011110110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 152 0x98 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xf3, 0xc0, /* 1111001111 */
+	0x61, 0x80, /* 0110000110 */
+	0x33, 0x00, /* 0011001100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x78, 0x00, /* 0111100000 */
+	0x70, 0x00, /* 0111000000 */
+
+	/* 153 0x99 '.' */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x0c, 0x00, /* 0000110000 */
+	0x17, 0x00, /* 0001011100 */
+	0x23, 0x00, /* 0010001100 */
+	0x21, 0x80, /* 0010000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x21, 0x00, /* 0010000100 */
+	0x31, 0x00, /* 0011000100 */
+	0x1a, 0x00, /* 0001101000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 154 0x9a '.' */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x00, 0x00, /* 0000000000 */
+	0xf1, 0xc0, /* 1111000111 */
+	0x60, 0x80, /* 0110000010 */
+	0x60, 0x80, /* 0110000010 */
+	0x60, 0x80, /* 0110000010 */
+	0x60, 0x80, /* 0110000010 */
+	0x60, 0x80, /* 0110000010 */
+	0x60, 0x80, /* 0110000010 */
+	0x60, 0x80, /* 0110000010 */
+	0x60, 0x80, /* 0110000010 */
+	0x71, 0x00, /* 0111000100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 155 0x9b '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x1f, 0x80, /* 0001111110 */
+	0x36, 0x80, /* 0011011010 */
+	0x26, 0x00, /* 0010011000 */
+	0x66, 0x00, /* 0110011000 */
+	0x66, 0x00, /* 0110011000 */
+	0x66, 0x00, /* 0110011000 */
+	0x66, 0x00, /* 0110011000 */
+	0x76, 0x00, /* 0111011000 */
+	0x36, 0x80, /* 0011011010 */
+	0x1f, 0x80, /* 0001111110 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 156 0x9c '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x3b, 0x00, /* 0011101100 */
+	0x33, 0x00, /* 0011001100 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x7e, 0x00, /* 0111111000 */
+	0x7e, 0x00, /* 0111111000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x7c, 0x80, /* 0111110010 */
+	0x7f, 0x80, /* 0111111110 */
+	0x43, 0x00, /* 0100001100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 157 0x9d '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x40, 0x80, /* 0100000010 */
+	0x40, 0x80, /* 0100000010 */
+	0x21, 0x00, /* 0010000100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x0c, 0x00, /* 0000110000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 158 0x9e '.' */
+	0x00, 0x00, /* 0000000000 */
+	0xbf, 0x00, /* 1011111100 */
+	0x40, 0x80, /* 0100000010 */
+	0x40, 0x80, /* 0100000010 */
+	0x7f, 0x00, /* 0111111100 */
+	0x40, 0x00, /* 0100000000 */
+	0x48, 0x00, /* 0100100000 */
+	0x48, 0x00, /* 0100100000 */
+	0x5e, 0x00, /* 0101111000 */
+	0x48, 0x00, /* 0100100000 */
+	0x48, 0x00, /* 0100100000 */
+	0x48, 0x00, /* 0100100000 */
+	0x48, 0x80, /* 0100100010 */
+	0x47, 0x00, /* 0100011100 */
+	0xe0, 0x00, /* 1110000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 159 0x9f '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x03, 0x00, /* 0000001100 */
+	0x04, 0x80, /* 0000010010 */
+	0x08, 0x00, /* 0000100000 */
+	0x08, 0x00, /* 0000100000 */
+	0x08, 0x00, /* 0000100000 */
+	0x08, 0x00, /* 0000100000 */
+	0x09, 0x00, /* 0000100100 */
+	0x3e, 0x00, /* 0011111000 */
+	0x48, 0x00, /* 0100100000 */
+	0x08, 0x00, /* 0000100000 */
+	0x08, 0x00, /* 0000100000 */
+	0x08, 0x00, /* 0000100000 */
+	0x08, 0x00, /* 0000100000 */
+	0x08, 0x00, /* 0000100000 */
+	0x90, 0x00, /* 1001000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 160 0xa0 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x03, 0x00, /* 0000001100 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x31, 0x80, /* 0011000110 */
+	0x21, 0x80, /* 0010000110 */
+	0x07, 0x80, /* 0000011110 */
+	0x39, 0x80, /* 0011100110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x73, 0x80, /* 0111001110 */
+	0x3d, 0xc0, /* 0011110111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 161 0xa1 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x03, 0x00, /* 0000001100 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 162 0xa2 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x00, 0x00, /* 0000000000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x66, 0x00, /* 0110011000 */
+	0xc3, 0x00, /* 1100001100 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xc1, 0x80, /* 1100000110 */
+	0xe1, 0x80, /* 1110000110 */
+	0x73, 0x00, /* 0111001100 */
+	0x3c, 0x00, /* 0011110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 163 0xa3 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x00, 0x00, /* 0000000000 */
+	0xf7, 0x80, /* 1111011110 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x77, 0x00, /* 0111011100 */
+	0x3d, 0x80, /* 0011110110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 164 0xa4 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x38, 0x80, /* 0011100010 */
+	0x7f, 0x80, /* 0111111110 */
+	0x47, 0x00, /* 0100011100 */
+	0x00, 0x00, /* 0000000000 */
+	0x6f, 0x00, /* 0110111100 */
+	0x7b, 0x80, /* 0111101110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x7b, 0xc0, /* 0111101111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 165 0xa5 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x38, 0x80, /* 0011100010 */
+	0x7f, 0x80, /* 0111111110 */
+	0x47, 0x00, /* 0100011100 */
+	0x00, 0x00, /* 0000000000 */
+	0xe3, 0xc0, /* 1110001111 */
+	0x71, 0x80, /* 0111000110 */
+	0x79, 0x80, /* 0111100110 */
+	0x79, 0x80, /* 0111100110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x67, 0x80, /* 0110011110 */
+	0x63, 0x80, /* 0110001110 */
+	0x61, 0x80, /* 0110000110 */
+	0xf0, 0xc0, /* 1111000011 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 166 0xa6 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x3e, 0x00, /* 0011111000 */
+	0x63, 0x00, /* 0110001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x0f, 0x00, /* 0000111100 */
+	0x33, 0x00, /* 0011001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x67, 0x00, /* 0110011100 */
+	0x3b, 0x80, /* 0011101110 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 167 0xa7 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x33, 0x00, /* 0011001100 */
+	0x21, 0x80, /* 0010000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x00, /* 0110000100 */
+	0x33, 0x00, /* 0011001100 */
+	0x1c, 0x00, /* 0001110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 168 0xa8 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x00, 0x00, /* 0000000000 */
+	0x06, 0x00, /* 0000011000 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x80, /* 0110000010 */
+	0x73, 0x80, /* 0111001110 */
+	0x3f, 0x00, /* 0011111100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 169 0xa9 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 170 0xaa '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 171 0xab '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x20, 0x00, /* 0010000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x20, 0x00, /* 0010000000 */
+	0x20, 0x80, /* 0010000010 */
+	0x21, 0x00, /* 0010000100 */
+	0x22, 0x00, /* 0010001000 */
+	0x74, 0x00, /* 0111010000 */
+	0x08, 0x00, /* 0000100000 */
+	0x17, 0x00, /* 0001011100 */
+	0x28, 0x80, /* 0010100010 */
+	0x43, 0x00, /* 0100001100 */
+	0x04, 0x00, /* 0000010000 */
+	0x08, 0x00, /* 0000100000 */
+	0x0f, 0x80, /* 0000111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 172 0xac '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x20, 0x00, /* 0010000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x20, 0x00, /* 0010000000 */
+	0x20, 0x80, /* 0010000010 */
+	0x21, 0x00, /* 0010000100 */
+	0x22, 0x00, /* 0010001000 */
+	0x74, 0x00, /* 0111010000 */
+	0x09, 0x00, /* 0000100100 */
+	0x13, 0x00, /* 0001001100 */
+	0x25, 0x00, /* 0010010100 */
+	0x49, 0x00, /* 0100100100 */
+	0x1f, 0x80, /* 0001111110 */
+	0x01, 0x00, /* 0000000100 */
+	0x01, 0x00, /* 0000000100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 173 0xad '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 174 0xae '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0d, 0x80, /* 0000110110 */
+	0x1b, 0x00, /* 0001101100 */
+	0x36, 0x00, /* 0011011000 */
+	0x6c, 0x00, /* 0110110000 */
+	0xd8, 0x00, /* 1101100000 */
+	0x6c, 0x00, /* 0110110000 */
+	0x36, 0x00, /* 0011011000 */
+	0x1b, 0x00, /* 0001101100 */
+	0x0d, 0x80, /* 0000110110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 175 0xaf '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x6c, 0x00, /* 0110110000 */
+	0x36, 0x00, /* 0011011000 */
+	0x1b, 0x00, /* 0001101100 */
+	0x0d, 0x80, /* 0000110110 */
+	0x06, 0xc0, /* 0000011011 */
+	0x0d, 0x80, /* 0000110110 */
+	0x1b, 0x00, /* 0001101100 */
+	0x36, 0x00, /* 0011011000 */
+	0x6c, 0x00, /* 0110110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 176 0xb0 '.' */
+	0xc3, 0x00, /* 1100001100 */
+	0x41, 0x00, /* 0100000100 */
+	0x18, 0x40, /* 0001100001 */
+	0x10, 0x40, /* 0001000001 */
+	0xc3, 0x00, /* 1100001100 */
+	0x41, 0x00, /* 0100000100 */
+	0x18, 0x40, /* 0001100001 */
+	0x10, 0x40, /* 0001000001 */
+	0xc3, 0x00, /* 1100001100 */
+	0x41, 0x00, /* 0100000100 */
+	0x18, 0x40, /* 0001100001 */
+	0x10, 0x40, /* 0001000001 */
+	0xc3, 0x00, /* 1100001100 */
+	0x41, 0x00, /* 0100000100 */
+	0x18, 0x40, /* 0001100001 */
+	0x10, 0x40, /* 0001000001 */
+	0xc3, 0x00, /* 1100001100 */
+	0x41, 0x00, /* 0100000100 */
+
+	/* 177 0xb1 '.' */
+	0x11, 0x00, /* 0001000100 */
+	0xbb, 0x80, /* 1011101110 */
+	0x11, 0x00, /* 0001000100 */
+	0x44, 0x40, /* 0100010001 */
+	0xee, 0xc0, /* 1110111011 */
+	0x44, 0x40, /* 0100010001 */
+	0x11, 0x00, /* 0001000100 */
+	0xbb, 0x80, /* 1011101110 */
+	0x11, 0x00, /* 0001000100 */
+	0x44, 0x40, /* 0100010001 */
+	0xee, 0xc0, /* 1110111011 */
+	0x44, 0x40, /* 0100010001 */
+	0x11, 0x00, /* 0001000100 */
+	0xbb, 0x80, /* 1011101110 */
+	0x11, 0x00, /* 0001000100 */
+	0x44, 0x40, /* 0100010001 */
+	0xee, 0xc0, /* 1110111011 */
+	0x44, 0x40, /* 0100010001 */
+
+	/* 178 0xb2 '.' */
+	0x3c, 0xc0, /* 0011110011 */
+	0xbe, 0xc0, /* 1011111011 */
+	0xe7, 0x80, /* 1110011110 */
+	0xef, 0x80, /* 1110111110 */
+	0x3c, 0xc0, /* 0011110011 */
+	0xbe, 0xc0, /* 1011111011 */
+	0xe7, 0x80, /* 1110011110 */
+	0xef, 0x80, /* 1110111110 */
+	0x3c, 0xc0, /* 0011110011 */
+	0xbe, 0xc0, /* 1011111011 */
+	0xe7, 0x80, /* 1110011110 */
+	0xef, 0x80, /* 1110111110 */
+	0x3c, 0xc0, /* 0011110011 */
+	0xbe, 0xc0, /* 1011111011 */
+	0xe7, 0x80, /* 1110011110 */
+	0xef, 0x80, /* 1110111110 */
+	0x3c, 0xc0, /* 0011110011 */
+	0xbe, 0xc0, /* 1011111011 */
+
+	/* 179 0xb3 '.' */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+
+	/* 180 0xb4 '.' */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+
+	/* 181 0xb5 '.' */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+
+	/* 182 0xb6 '.' */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0xfb, 0x00, /* 1111101100 */
+	0xfb, 0x00, /* 1111101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+
+	/* 183 0xb7 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0x00, /* 1111111100 */
+	0xff, 0x00, /* 1111111100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+
+	/* 184 0xb8 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xfc, 0x00, /* 1111110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+
+	/* 185 0xb9 '.' */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0xfb, 0x00, /* 1111101100 */
+	0xfb, 0x00, /* 1111101100 */
+	0x03, 0x00, /* 0000001100 */
+	0xfb, 0x00, /* 1111101100 */
+	0xfb, 0x00, /* 1111101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+
+	/* 186 0xba '.' */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+
+	/* 187 0xbb '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0x00, /* 1111111100 */
+	0xff, 0x00, /* 1111111100 */
+	0x03, 0x00, /* 0000001100 */
+	0xfb, 0x00, /* 1111101100 */
+	0xfb, 0x00, /* 1111101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+
+	/* 188 0xbc '.' */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0xfb, 0x00, /* 1111101100 */
+	0xfb, 0x00, /* 1111101100 */
+	0x03, 0x00, /* 0000001100 */
+	0xff, 0x00, /* 1111111100 */
+	0xff, 0x00, /* 1111111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 189 0xbd '.' */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0xff, 0x00, /* 1111111100 */
+	0xff, 0x00, /* 1111111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 190 0xbe '.' */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 191 0xbf '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xfc, 0x00, /* 1111110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+
+	/* 192 0xc0 '.' */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 193 0xc1 '.' */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 194 0xc2 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+
+	/* 195 0xc3 '.' */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+
+	/* 196 0xc4 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 197 0xc5 '.' */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+
+	/* 198 0xc6 '.' */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+
+	/* 199 0xc7 '.' */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0xc0, /* 0001101111 */
+	0x1b, 0xc0, /* 0001101111 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+
+	/* 200 0xc8 '.' */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0xc0, /* 0001101111 */
+	0x1b, 0xc0, /* 0001101111 */
+	0x18, 0x00, /* 0001100000 */
+	0x1f, 0xc0, /* 0001111111 */
+	0x1f, 0xc0, /* 0001111111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 201 0xc9 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0xc0, /* 0001111111 */
+	0x1f, 0xc0, /* 0001111111 */
+	0x18, 0x00, /* 0001100000 */
+	0x1b, 0xc0, /* 0001101111 */
+	0x1b, 0xc0, /* 0001101111 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+
+	/* 202 0xca '.' */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0xfb, 0xc0, /* 1111101111 */
+	0xfb, 0xc0, /* 1111101111 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 203 0xcb '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x00, 0x00, /* 0000000000 */
+	0xfb, 0xc0, /* 1111101111 */
+	0xfb, 0xc0, /* 1111101111 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+
+	/* 204 0xcc '.' */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0xc0, /* 0001101111 */
+	0x1b, 0xc0, /* 0001101111 */
+	0x18, 0x00, /* 0001100000 */
+	0x1b, 0xc0, /* 0001101111 */
+	0x1b, 0xc0, /* 0001101111 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+
+	/* 205 0xcd '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 206 0xce '.' */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0xfb, 0xc0, /* 1111101111 */
+	0xfb, 0xc0, /* 1111101111 */
+	0x00, 0x00, /* 0000000000 */
+	0xfb, 0xc0, /* 1111101111 */
+	0xfb, 0xc0, /* 1111101111 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+
+	/* 207 0xcf '.' */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 208 0xd0 '.' */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 209 0xd1 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+
+	/* 210 0xd2 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+
+	/* 211 0xd3 '.' */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1f, 0xc0, /* 0001111111 */
+	0x1f, 0xc0, /* 0001111111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 212 0xd4 '.' */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 213 0xd5 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+
+	/* 214 0xd6 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0xc0, /* 0001111111 */
+	0x1f, 0xc0, /* 0001111111 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+
+	/* 215 0xd7 '.' */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+	0x1b, 0x00, /* 0001101100 */
+
+	/* 216 0xd8 '.' */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x0c, 0x00, /* 0000110000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+
+	/* 217 0xd9 '.' */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0xfc, 0x00, /* 1111110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 218 0xda '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+
+	/* 219 0xdb '.' */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+
+	/* 220 0xdc '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+
+	/* 221 0xdd '.' */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+	0xf8, 0x00, /* 1111100000 */
+
+	/* 222 0xde '.' */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+	0x07, 0xc0, /* 0000011111 */
+
+	/* 223 0xdf '.' */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0xff, 0xc0, /* 1111111111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 224 0xe0 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1c, 0x80, /* 0001110010 */
+	0x35, 0x80, /* 0011010110 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x63, 0x00, /* 0110001100 */
+	0x37, 0x80, /* 0011011110 */
+	0x1c, 0x80, /* 0001110010 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 225 0xe1 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x33, 0x00, /* 0011001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x63, 0x00, /* 0110001100 */
+	0x6f, 0x00, /* 0110111100 */
+	0x63, 0x00, /* 0110001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x63, 0x00, /* 0110001100 */
+	0x6e, 0x00, /* 0110111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 226 0xe2 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 227 0xe3 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 228 0xe4 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0x80, /* 1111111110 */
+	0x60, 0x00, /* 0110000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x80, /* 0011000010 */
+	0x61, 0x80, /* 0110000110 */
+	0xff, 0x80, /* 1111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 229 0xe5 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1f, 0xc0, /* 0001111111 */
+	0x36, 0x00, /* 0011011000 */
+	0x63, 0x00, /* 0110001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x33, 0x00, /* 0011001100 */
+	0x3e, 0x00, /* 0011111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 230 0xe6 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x73, 0x80, /* 0111001110 */
+	0x6d, 0x80, /* 0110110110 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0xc0, 0x00, /* 1100000000 */
+
+	/* 231 0xe7 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x01, 0x80, /* 0000000110 */
+	0x36, 0x40, /* 0011011001 */
+	0x5e, 0x00, /* 0101111000 */
+	0x8c, 0x00, /* 1000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 232 0xe8 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x33, 0x00, /* 0011001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x33, 0x00, /* 0011001100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 233 0xe9 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x1f, 0x00, /* 0001111100 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x60, 0xc0, /* 0110000011 */
+	0x7f, 0xc0, /* 0111111111 */
+	0x7f, 0xc0, /* 0111111111 */
+	0x60, 0xc0, /* 0110000011 */
+	0x31, 0x80, /* 0011000110 */
+	0x31, 0x80, /* 0011000110 */
+	0x1f, 0x00, /* 0001111100 */
+	0x0e, 0x00, /* 0000111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 234 0xea '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x33, 0x00, /* 0011001100 */
+	0x61, 0x80, /* 0110000110 */
+	0xc0, 0xc0, /* 1100000011 */
+	0xc0, 0xc0, /* 1100000011 */
+	0xc0, 0xc0, /* 1100000011 */
+	0x61, 0x80, /* 0110000110 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0xf3, 0xc0, /* 1111001111 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 235 0xeb '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x07, 0x00, /* 0000011100 */
+	0x1f, 0x80, /* 0001111110 */
+	0x30, 0xc0, /* 0011000011 */
+	0x30, 0x00, /* 0011000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x3e, 0x00, /* 0011111000 */
+	0x66, 0x00, /* 0110011000 */
+	0xc3, 0x00, /* 1100001100 */
+	0xc3, 0x00, /* 1100001100 */
+	0xc3, 0x00, /* 1100001100 */
+	0x66, 0x00, /* 0110011000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 236 0xec '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x33, 0x00, /* 0011001100 */
+	0x6d, 0x80, /* 0110110110 */
+	0xcc, 0xc0, /* 1100110011 */
+	0xcc, 0xc0, /* 1100110011 */
+	0xcc, 0xc0, /* 1100110011 */
+	0xcc, 0xc0, /* 1100110011 */
+	0x6d, 0x80, /* 0110110110 */
+	0x33, 0x00, /* 0011001100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 237 0xed '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x01, 0x80, /* 0000000110 */
+	0x01, 0x80, /* 0000000110 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x37, 0x00, /* 0011011100 */
+	0x6d, 0x80, /* 0110110110 */
+	0xcc, 0xc0, /* 1100110011 */
+	0xcc, 0xc0, /* 1100110011 */
+	0xcc, 0xc0, /* 1100110011 */
+	0xcc, 0xc0, /* 1100110011 */
+	0x6d, 0x80, /* 0110110110 */
+	0x3b, 0x00, /* 0011101100 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x60, 0x00, /* 0110000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 238 0xee '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x03, 0x80, /* 0000001110 */
+	0x0e, 0x00, /* 0000111000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x3f, 0x80, /* 0011111110 */
+	0x3f, 0x80, /* 0011111110 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x18, 0x00, /* 0001100000 */
+	0x18, 0x00, /* 0001100000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x03, 0x80, /* 0000001110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 239 0xef '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x33, 0x00, /* 0011001100 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x61, 0x80, /* 0110000110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 240 0xf0 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 241 0xf1 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 242 0xf2 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xe0, 0x00, /* 1110000000 */
+	0x38, 0x00, /* 0011100000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x03, 0x80, /* 0000001110 */
+	0x0e, 0x00, /* 0000111000 */
+	0x38, 0x00, /* 0011100000 */
+	0xe0, 0x00, /* 1110000000 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0x00, /* 1111111100 */
+	0xff, 0x00, /* 1111111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 243 0xf3 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x03, 0x80, /* 0000001110 */
+	0x0e, 0x00, /* 0000111000 */
+	0x38, 0x00, /* 0011100000 */
+	0xe0, 0x00, /* 1110000000 */
+	0x38, 0x00, /* 0011100000 */
+	0x0e, 0x00, /* 0000111000 */
+	0x03, 0x80, /* 0000001110 */
+	0x00, 0x00, /* 0000000000 */
+	0xff, 0x80, /* 1111111110 */
+	0xff, 0x80, /* 1111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 244 0xf4 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x30, 0x00, /* 0011000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 245 0xf5 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x03, 0x00, /* 0000001100 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 246 0xf6 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 247 0xf7 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x38, 0x00, /* 0011100000 */
+	0x6c, 0x00, /* 0110110000 */
+	0x06, 0xc0, /* 0000011011 */
+	0x03, 0x80, /* 0000001110 */
+	0x38, 0x00, /* 0011100000 */
+	0x6c, 0x00, /* 0110110000 */
+	0x06, 0xc0, /* 0000011011 */
+	0x03, 0x80, /* 0000001110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 248 0xf8 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x33, 0x00, /* 0011001100 */
+	0x33, 0x00, /* 0011001100 */
+	0x1e, 0x00, /* 0001111000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 249 0xf9 '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 250 0xfa '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 251 0xfb '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0f, 0xc0, /* 0000111111 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0xcc, 0x00, /* 1100110000 */
+	0x6c, 0x00, /* 0110110000 */
+	0x3c, 0x00, /* 0011110000 */
+	0x1c, 0x00, /* 0001110000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 252 0xfc '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x27, 0x00, /* 0010011100 */
+	0x7b, 0x00, /* 0111101100 */
+	0x31, 0x00, /* 0011000100 */
+	0x31, 0x00, /* 0011000100 */
+	0x31, 0x00, /* 0011000100 */
+	0x7b, 0x80, /* 0111101110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 253 0xfd '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x1e, 0x00, /* 0001111000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x63, 0x00, /* 0110001100 */
+	0x43, 0x00, /* 0100001100 */
+	0x06, 0x00, /* 0000011000 */
+	0x0c, 0x00, /* 0000110000 */
+	0x18, 0x00, /* 0001100000 */
+	0x30, 0x80, /* 0011000010 */
+	0x7f, 0x80, /* 0111111110 */
+	0x7f, 0x80, /* 0111111110 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 254 0xfe '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x3f, 0x00, /* 0011111100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x3f, 0x00, /* 0011111100 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+	/* 255 0xff '.' */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+	0x00, 0x00, /* 0000000000 */
+
+};
+
+
+struct font_desc font_10x18 = {
+	FONT10x18_IDX,
+	"10x18",
+	10,
+	18,
+	fontdata_10x18,
+#ifdef __sparc__
+	5
+#else
+	-1
+#endif
+};
diff --git a/drivers/video/console/font_7x14.c b/drivers/video/console/font_7x14.c
new file mode 100644
index 00000000000..1fa7fcf2ff7
--- /dev/null
+++ b/drivers/video/console/font_7x14.c
@@ -0,0 +1,4118 @@
+/**************************************/
+/* this file adapted from font_8x16.c */
+/* by Jurriaan Kalkman 05-2005        */
+/**************************************/
+
+#include <linux/font.h>
+
+#define FONTDATAMAX 3584
+
+static unsigned char fontdata_7x14[FONTDATAMAX] = {
+
+	/* 0 0x00 '^@' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 1 0x01 '^A' */
+	0x00, /* 0000000 */
+	0x7c, /* 0111110 */
+	0x82, /* 1000001 */
+	0xaa, /* 1010101 */
+	0x82, /* 1000001 */
+	0x82, /* 1000001 */
+	0xba, /* 1011101 */
+	0x92, /* 1001001 */
+	0x82, /* 1000001 */
+	0x7c, /* 0111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 2 0x02 '^B' */
+	0x00, /* 0000000 */
+	0x7c, /* 0111110 */
+	0xfe, /* 1111111 */
+	0xd6, /* 1101011 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xc6, /* 1100011 */
+	0xee, /* 1110111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0x7c, /* 0111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 3 0x03 '^C' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x6c, /* 0110110 */
+	0x7c, /* 0111110 */
+	0xfe, /* 1111111 */
+	0x7c, /* 0111110 */
+	0x38, /* 0011100 */
+	0x18, /* 0001100 */
+	0x10, /* 0001000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 4 0x04 '^D' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x10, /* 0001000 */
+	0x38, /* 0011100 */
+	0x7c, /* 0111110 */
+	0xfe, /* 1111111 */
+	0x7c, /* 0111110 */
+	0x38, /* 0011100 */
+	0x10, /* 0001000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 5 0x05 '^E' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x38, /* 0011100 */
+	0x38, /* 0011100 */
+	0x38, /* 0011100 */
+	0xee, /* 1110111 */
+	0xee, /* 1110111 */
+	0xee, /* 1110111 */
+	0x10, /* 0001000 */
+	0x10, /* 0001000 */
+	0x38, /* 0011100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 6 0x06 '^F' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x10, /* 0001000 */
+	0x38, /* 0011100 */
+	0x7c, /* 0111110 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0x7c, /* 0111110 */
+	0x10, /* 0001000 */
+	0x10, /* 0001000 */
+	0x38, /* 0011100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 7 0x07 '^G' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x18, /* 0001100 */
+	0x3c, /* 0011110 */
+	0x3c, /* 0011110 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 8 0x08 '^H' */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xe6, /* 1110011 */
+	0xc2, /* 1100001 */
+	0xc2, /* 1100001 */
+	0xe6, /* 1110011 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+
+	/* 9 0x09 '^I' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0x44, /* 0100010 */
+	0x6c, /* 0110110 */
+	0x38, /* 0011100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 10 0x0a '^J' */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xc6, /* 1100011 */
+	0x92, /* 1001001 */
+	0xba, /* 1011101 */
+	0x92, /* 1001001 */
+	0xc6, /* 1100011 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+
+	/* 11 0x0b '^K' */
+	0x00, /* 0000000 */
+	0x1e, /* 0001111 */
+	0x0e, /* 0000111 */
+	0x1a, /* 0001101 */
+	0x1a, /* 0001101 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 12 0x0c '^L' */
+	0x00, /* 0000000 */
+	0x3c, /* 0011110 */
+	0x66, /* 0110011 */
+	0x66, /* 0110011 */
+	0x66, /* 0110011 */
+	0x66, /* 0110011 */
+	0x3c, /* 0011110 */
+	0x18, /* 0001100 */
+	0x7e, /* 0111111 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 13 0x0d '^M' */
+	0x00, /* 0000000 */
+	0x3e, /* 0011111 */
+	0x36, /* 0011011 */
+	0x3e, /* 0011111 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x70, /* 0111000 */
+	0xf0, /* 1111000 */
+	0xe0, /* 1110000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 14 0x0e '^N' */
+	0x00, /* 0000000 */
+	0x7e, /* 0111111 */
+	0x66, /* 0110011 */
+	0x7e, /* 0111111 */
+	0x66, /* 0110011 */
+	0x66, /* 0110011 */
+	0x66, /* 0110011 */
+	0x66, /* 0110011 */
+	0x6e, /* 0110111 */
+	0xee, /* 1110111 */
+	0xec, /* 1110110 */
+	0xc0, /* 1100000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 15 0x0f '^O' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x10, /* 0001000 */
+	0x10, /* 0001000 */
+	0xd6, /* 1101011 */
+	0x38, /* 0011100 */
+	0xee, /* 1110111 */
+	0x38, /* 0011100 */
+	0xd6, /* 1101011 */
+	0x10, /* 0001000 */
+	0x10, /* 0001000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 16 0x10 '^P' */
+	0x00, /* 0000000 */
+	0x80, /* 1000000 */
+	0xc0, /* 1100000 */
+	0xe0, /* 1110000 */
+	0xf0, /* 1111000 */
+	0xfc, /* 1111110 */
+	0xf0, /* 1111000 */
+	0xe0, /* 1110000 */
+	0xc0, /* 1100000 */
+	0x80, /* 1000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 17 0x11 '^Q' */
+	0x00, /* 0000000 */
+	0x04, /* 0000010 */
+	0x0c, /* 0000110 */
+	0x1c, /* 0001110 */
+	0x3c, /* 0011110 */
+	0xfc, /* 1111110 */
+	0x3c, /* 0011110 */
+	0x1c, /* 0001110 */
+	0x0c, /* 0000110 */
+	0x04, /* 0000010 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 18 0x12 '^R' */
+	0x00, /* 0000000 */
+	0x18, /* 0001100 */
+	0x3c, /* 0011110 */
+	0x7e, /* 0111111 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x7e, /* 0111111 */
+	0x3c, /* 0011110 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 19 0x13 '^S' */
+	0x00, /* 0000000 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x00, /* 0000000 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 20 0x14 '^T' */
+	0x00, /* 0000000 */
+	0x7e, /* 0111111 */
+	0xd4, /* 1101010 */
+	0xd4, /* 1101010 */
+	0xd4, /* 1101010 */
+	0x74, /* 0111010 */
+	0x14, /* 0001010 */
+	0x14, /* 0001010 */
+	0x14, /* 0001010 */
+	0x14, /* 0001010 */
+	0x16, /* 0001011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 21 0x15 '^U' */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0x60, /* 0110000 */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0xc6, /* 1100011 */
+	0xc6, /* 1100011 */
+	0x6c, /* 0110110 */
+	0x38, /* 0011100 */
+	0x18, /* 0001100 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 22 0x16 '^V' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0xfc, /* 1111110 */
+	0xfc, /* 1111110 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 23 0x17 '^W' */
+	0x00, /* 0000000 */
+	0x18, /* 0001100 */
+	0x3c, /* 0011110 */
+	0x7e, /* 0111111 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x7e, /* 0111111 */
+	0x3c, /* 0011110 */
+	0x18, /* 0001100 */
+	0x7e, /* 0111111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 24 0x18 '^X' */
+	0x00, /* 0000000 */
+	0x18, /* 0001100 */
+	0x3c, /* 0011110 */
+	0x7e, /* 0111111 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 25 0x19 '^Y' */
+	0x00, /* 0000000 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x7e, /* 0111111 */
+	0x3c, /* 0011110 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 26 0x1a '^Z' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x18, /* 0001100 */
+	0xfc, /* 1111110 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 27 0x1b '^[' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0xfc, /* 1111110 */
+	0x60, /* 0110000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 28 0x1c '^\' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 29 0x1d '^]' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x28, /* 0010100 */
+	0x6c, /* 0110110 */
+	0xfe, /* 1111111 */
+	0x6c, /* 0110110 */
+	0x28, /* 0010100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 30 0x1e '^^' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0x78, /* 0111100 */
+	0xfc, /* 1111110 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 31 0x1f '^_' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0xfc, /* 1111110 */
+	0x78, /* 0111100 */
+	0x78, /* 0111100 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 32 0x20 ' ' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 33 0x21 '!' */
+	0x00, /* 0000000 */
+	0x18, /* 0001100 */
+	0x3c, /* 0011110 */
+	0x3c, /* 0011110 */
+	0x3c, /* 0011110 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 34 0x22 '"' */
+	0x00, /* 0000000 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x28, /* 0010100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 35 0x23 '#' */
+	0x00, /* 0000000 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 36 0x24 '$' */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xc4, /* 1100010 */
+	0xc0, /* 1100000 */
+	0x78, /* 0111100 */
+	0x0c, /* 0000110 */
+	0x8c, /* 1000110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+
+	/* 37 0x25 '%' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xc0, /* 1100000 */
+	0xc4, /* 1100010 */
+	0x0c, /* 0000110 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0xcc, /* 1100110 */
+	0x8c, /* 1000110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 38 0x26 '&' */
+	0x00, /* 0000000 */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x38, /* 0011100 */
+	0x78, /* 0111100 */
+	0xde, /* 1101111 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xdc, /* 1101110 */
+	0x76, /* 0111011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 39 0x27 ''' */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 40 0x28 '(' */
+	0x00, /* 0000000 */
+	0x0c, /* 0000110 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x18, /* 0001100 */
+	0x0c, /* 0000110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 41 0x29 ')' */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x18, /* 0001100 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 42 0x2a '*' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x6c, /* 0110110 */
+	0x38, /* 0011100 */
+	0xfe, /* 1111111 */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 43 0x2b '+' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x10, /* 0001000 */
+	0x10, /* 0001000 */
+	0x7c, /* 0111110 */
+	0x10, /* 0001000 */
+	0x10, /* 0001000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 44 0x2c ',' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 45 0x2d '-' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 46 0x2e '.' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 47 0x2f '/' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x04, /* 0000010 */
+	0x0c, /* 0000110 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0xc0, /* 1100000 */
+	0x80, /* 1000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 48 0x30 '0' */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xdc, /* 1101110 */
+	0xec, /* 1110110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 49 0x31 '1' */
+	0x00, /* 0000000 */
+	0x18, /* 0001100 */
+	0x38, /* 0011100 */
+	0x78, /* 0111100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x7c, /* 0111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 50 0x32 '2' */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0x0c, /* 0000110 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0xc0, /* 1100000 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 51 0x33 '3' */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x38, /* 0011100 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 52 0x34 '4' */
+	0x00, /* 0000000 */
+	0x0c, /* 0000110 */
+	0x1c, /* 0001110 */
+	0x3c, /* 0011110 */
+	0x6c, /* 0110110 */
+	0xcc, /* 1100110 */
+	0xfe, /* 1111111 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 53 0x35 '5' */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xf8, /* 1111100 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 54 0x36 '6' */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xf8, /* 1111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 55 0x37 '7' */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0xcc, /* 1100110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 56 0x38 '8' */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 57 0x39 '9' */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x7c, /* 0111110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x18, /* 0001100 */
+	0x70, /* 0111000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 58 0x3a ':' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 59 0x3b ';' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 60 0x3c '<' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x04, /* 0000010 */
+	0x0c, /* 0000110 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0x30, /* 0011000 */
+	0x18, /* 0001100 */
+	0x0c, /* 0000110 */
+	0x04, /* 0000010 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 61 0x3d '=' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x7c, /* 0111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x7c, /* 0111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 62 0x3e '>' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x40, /* 0100000 */
+	0x60, /* 0110000 */
+	0x30, /* 0011000 */
+	0x18, /* 0001100 */
+	0x0c, /* 0000110 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0x40, /* 0100000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 63 0x3f '?' */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 64 0x40 '@' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xdc, /* 1101110 */
+	0xdc, /* 1101110 */
+	0xd8, /* 1101100 */
+	0xc0, /* 1100000 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 65 0x41 'A' */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 66 0x42 'B' */
+	0x00, /* 0000000 */
+	0xf8, /* 1111100 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x78, /* 0111100 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0xf8, /* 1111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 67 0x43 'C' */
+	0x00, /* 0000000 */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0xc4, /* 1100010 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc4, /* 1100010 */
+	0x6c, /* 0110110 */
+	0x38, /* 0011100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 68 0x44 'D' */
+	0x00, /* 0000000 */
+	0xf0, /* 1111000 */
+	0xd8, /* 1101100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xd8, /* 1101100 */
+	0xf0, /* 1111000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 69 0x45 'E' */
+	0x00, /* 0000000 */
+	0x7c, /* 0111110 */
+	0x6c, /* 0110110 */
+	0x64, /* 0110010 */
+	0x68, /* 0110100 */
+	0x78, /* 0111100 */
+	0x68, /* 0110100 */
+	0x60, /* 0110000 */
+	0x64, /* 0110010 */
+	0x6c, /* 0110110 */
+	0x7c, /* 0111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 70 0x46 'F' */
+	0x00, /* 0000000 */
+	0x7c, /* 0111110 */
+	0x64, /* 0110010 */
+	0x60, /* 0110000 */
+	0x68, /* 0110100 */
+	0x78, /* 0111100 */
+	0x68, /* 0110100 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x70, /* 0111000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 71 0x47 'G' */
+	0x00, /* 0000000 */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0xc4, /* 1100010 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xdc, /* 1101110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x6c, /* 0110110 */
+	0x34, /* 0011010 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 72 0x48 'H' */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 73 0x49 'I' */
+	0x00, /* 0000000 */
+	0x3c, /* 0011110 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x3c, /* 0011110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 74 0x4a 'J' */
+	0x00, /* 0000000 */
+	0x1c, /* 0001110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 75 0x4b 'K' */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xd8, /* 1101100 */
+	0xf0, /* 1111000 */
+	0xf0, /* 1111000 */
+	0xd8, /* 1101100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 76 0x4c 'L' */
+	0x00, /* 0000000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc4, /* 1100010 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 77 0x4d 'M' */
+	0x00, /* 0000000 */
+	0xc6, /* 1100011 */
+	0xee, /* 1110111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xd6, /* 1101011 */
+	0xc6, /* 1100011 */
+	0xc6, /* 1100011 */
+	0xc6, /* 1100011 */
+	0xc6, /* 1100011 */
+	0xc6, /* 1100011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 78 0x4e 'N' */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xec, /* 1110110 */
+	0xec, /* 1110110 */
+	0xfc, /* 1111110 */
+	0xdc, /* 1101110 */
+	0xdc, /* 1101110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 79 0x4f 'O' */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 80 0x50 'P' */
+	0x00, /* 0000000 */
+	0xf8, /* 1111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xf8, /* 1111100 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 81 0x51 'Q' */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xdc, /* 1101110 */
+	0x78, /* 0111100 */
+	0x18, /* 0001100 */
+	0x1c, /* 0001110 */
+	0x00, /* 0000000 */
+
+	/* 82 0x52 'R' */
+	0x00, /* 0000000 */
+	0xf8, /* 1111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xf8, /* 1111100 */
+	0xd8, /* 1101100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 83 0x53 'S' */
+	0x00, /* 0000000 */
+	0x7c, /* 0111110 */
+	0xc4, /* 1100010 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0x60, /* 0110000 */
+	0x38, /* 0011100 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x8c, /* 1000110 */
+	0xf8, /* 1111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 84 0x54 'T' */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0xfc, /* 1111110 */
+	0xb4, /* 1011010 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 85 0x55 'U' */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 86 0x56 'V' */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x78, /* 0111100 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 87 0x57 'W' */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0xfc, /* 1111110 */
+	0x48, /* 0100100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 88 0x58 'X' */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x78, /* 0111100 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 89 0x59 'Y' */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 90 0x5a 'Z' */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0xcc, /* 1100110 */
+	0x8c, /* 1000110 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0xc4, /* 1100010 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 91 0x5b '[' */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 92 0x5c '\' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x80, /* 1000000 */
+	0xc0, /* 1100000 */
+	0xe0, /* 1110000 */
+	0x70, /* 0111000 */
+	0x38, /* 0011100 */
+	0x1c, /* 0001110 */
+	0x0c, /* 0000110 */
+	0x04, /* 0000010 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 93 0x5d ']' */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 94 0x5e '^' */
+	0x10, /* 0001000 */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0xc6, /* 1100011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 95 0x5f '_' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfe, /* 1111111 */
+	0x00, /* 0000000 */
+
+	/* 96 0x60 '`' */
+	0x00, /* 0000000 */
+	0x60, /* 0110000 */
+	0x30, /* 0011000 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 97 0x61 'a' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0x0c, /* 0000110 */
+	0x7c, /* 0111110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x76, /* 0111011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 98 0x62 'b' */
+	0x00, /* 0000000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xf0, /* 1111000 */
+	0xd8, /* 1101100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xf8, /* 1111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 99 0x63 'c' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 100 0x64 'd' */
+	0x00, /* 0000000 */
+	0x1c, /* 0001110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x3c, /* 0011110 */
+	0x6c, /* 0110110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x76, /* 0111011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 101 0x65 'e' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 102 0x66 'f' */
+	0x00, /* 0000000 */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0x64, /* 0110010 */
+	0x60, /* 0110000 */
+	0xf0, /* 1111000 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0xf0, /* 1111000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 103 0x67 'g' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x76, /* 0111011 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x7c, /* 0111110 */
+	0x0c, /* 0000110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+
+	/* 104 0x68 'h' */
+	0x00, /* 0000000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xd8, /* 1101100 */
+	0xec, /* 1110110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 105 0x69 'i' */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x70, /* 0111000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 106 0x6a 'j' */
+	0x00, /* 0000000 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x00, /* 0000000 */
+	0x1c, /* 0001110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+
+	/* 107 0x6b 'k' */
+	0x00, /* 0000000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xcc, /* 1100110 */
+	0xd8, /* 1101100 */
+	0xf0, /* 1111000 */
+	0xf0, /* 1111000 */
+	0xd8, /* 1101100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 108 0x6c 'l' */
+	0x00, /* 0000000 */
+	0x70, /* 0111000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 109 0x6d 'm' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xec, /* 1110110 */
+	0xfe, /* 1111111 */
+	0xd6, /* 1101011 */
+	0xd6, /* 1101011 */
+	0xd6, /* 1101011 */
+	0xd6, /* 1101011 */
+	0xd6, /* 1101011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 110 0x6e 'n' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xb8, /* 1011100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 111 0x6f 'o' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 112 0x70 'p' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xb8, /* 1011100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xf8, /* 1111100 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+
+	/* 113 0x71 'q' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x74, /* 0111010 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x7c, /* 0111110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+
+	/* 114 0x72 'r' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xb8, /* 1011100 */
+	0xec, /* 1110110 */
+	0xcc, /* 1100110 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 115 0x73 's' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0x60, /* 0110000 */
+	0x30, /* 0011000 */
+	0x18, /* 0001100 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 116 0x74 't' */
+	0x00, /* 0000000 */
+	0x10, /* 0001000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0xfc, /* 1111110 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x36, /* 0011011 */
+	0x1c, /* 0001110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 117 0x75 'u' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x76, /* 0111011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 118 0x76 'v' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 119 0x77 'w' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xc6, /* 1100011 */
+	0xc6, /* 1100011 */
+	0xd6, /* 1101011 */
+	0xd6, /* 1101011 */
+	0xd6, /* 1101011 */
+	0xfe, /* 1111111 */
+	0x6c, /* 0110110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 120 0x78 'x' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 121 0x79 'y' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x7c, /* 0111110 */
+	0x0c, /* 0000110 */
+	0x18, /* 0001100 */
+	0xf0, /* 1111000 */
+
+	/* 122 0x7a 'z' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0xcc, /* 1100110 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 123 0x7b '{' */
+	0x00, /* 0000000 */
+	0x1c, /* 0001110 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0xe0, /* 1110000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x1c, /* 0001110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 124 0x7c '|' */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 125 0x7d '}' */
+	0x00, /* 0000000 */
+	0x70, /* 0111000 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x0e, /* 0000111 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x70, /* 0111000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 126 0x7e '~' */
+	0x00, /* 0000000 */
+	0xec, /* 1110110 */
+	0xb8, /* 1011100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 127 0x7f '' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x10, /* 0001000 */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0xc6, /* 1100011 */
+	0xc6, /* 1100011 */
+	0xc6, /* 1100011 */
+	0xfe, /* 1111111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 128 0x80 '€' */
+	0x00, /* 0000000 */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0xc4, /* 1100010 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc4, /* 1100010 */
+	0x6c, /* 0110110 */
+	0x38, /* 0011100 */
+	0x18, /* 0001100 */
+	0x70, /* 0111000 */
+	0x00, /* 0000000 */
+
+	/* 129 0x81 '�' */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x76, /* 0111011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 130 0x82 '‚' */
+	0x0c, /* 0000110 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 131 0x83 'ƒ' */
+	0x10, /* 0001000 */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0x0c, /* 0000110 */
+	0x7c, /* 0111110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x76, /* 0111011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 132 0x84 '„' */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0x0c, /* 0000110 */
+	0x7c, /* 0111110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x76, /* 0111011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 133 0x85 '…' */
+	0x60, /* 0110000 */
+	0x30, /* 0011000 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0x0c, /* 0000110 */
+	0x7c, /* 0111110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x76, /* 0111011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 134 0x86 '†' */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0x38, /* 0011100 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0x0c, /* 0000110 */
+	0x7c, /* 0111110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x76, /* 0111011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 135 0x87 '‡' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0xe0, /* 1110000 */
+
+	/* 136 0x88 'ˆ' */
+	0x10, /* 0001000 */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 137 0x89 '‰' */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 138 0x8a 'Š' */
+	0xc0, /* 1100000 */
+	0x60, /* 0110000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 139 0x8b '‹' */
+	0x00, /* 0000000 */
+	0x6c, /* 0110110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x38, /* 0011100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x3c, /* 0011110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 140 0x8c 'Œ' */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x70, /* 0111000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 141 0x8d '�' */
+	0xc0, /* 1100000 */
+	0x60, /* 0110000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x70, /* 0111000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 142 0x8e 'Ž' */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 143 0x8f '�' */
+	0x30, /* 0011000 */
+	0x48, /* 0100100 */
+	0x48, /* 0100100 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 144 0x90 '�' */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0xfc, /* 1111110 */
+	0xcc, /* 1100110 */
+	0xc4, /* 1100010 */
+	0xd0, /* 1101000 */
+	0xf0, /* 1111000 */
+	0xd0, /* 1101000 */
+	0xc4, /* 1100010 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 145 0x91 '‘' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xec, /* 1110110 */
+	0x36, /* 0011011 */
+	0x36, /* 0011011 */
+	0x7e, /* 0111111 */
+	0xd8, /* 1101100 */
+	0xd8, /* 1101100 */
+	0x6e, /* 0110111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 146 0x92 '’' */
+	0x00, /* 0000000 */
+	0x3e, /* 0011111 */
+	0x6c, /* 0110110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xfe, /* 1111111 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xce, /* 1100111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 147 0x93 '“' */
+	0x10, /* 0001000 */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 148 0x94 '”' */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 149 0x95 '•' */
+	0xc0, /* 1100000 */
+	0x60, /* 0110000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 150 0x96 '–' */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x76, /* 0111011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 151 0x97 '—' */
+	0x60, /* 0110000 */
+	0x30, /* 0011000 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x76, /* 0111011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 152 0x98 '˜' */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x7c, /* 0111110 */
+	0x0c, /* 0000110 */
+	0x18, /* 0001100 */
+	0x70, /* 0111000 */
+
+	/* 153 0x99 '™' */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 154 0x9a 'š' */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 155 0x9b '›' */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x7c, /* 0111110 */
+	0xcc, /* 1100110 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xcc, /* 1100110 */
+	0x7c, /* 0111110 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 156 0x9c 'œ' */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0x64, /* 0110010 */
+	0x60, /* 0110000 */
+	0xf0, /* 1111000 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0xe6, /* 1110011 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 157 0x9d '�' */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x30, /* 0011000 */
+	0xfc, /* 1111110 */
+	0x30, /* 0011000 */
+	0xfc, /* 1111110 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 158 0x9e 'ž' */
+	0xf8, /* 1111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xf8, /* 1111100 */
+	0xc4, /* 1100010 */
+	0xcc, /* 1100110 */
+	0xde, /* 1101111 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xc6, /* 1100011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 159 0x9f 'Ÿ' */
+	0x1c, /* 0001110 */
+	0x36, /* 0011011 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0xfc, /* 1111110 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0xb0, /* 1011000 */
+	0xe0, /* 1110000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 160 0xa0 ' ' */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0x0c, /* 0000110 */
+	0x7c, /* 0111110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x76, /* 0111011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 161 0xa1 '¡' */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0x00, /* 0000000 */
+	0x70, /* 0111000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 162 0xa2 '¢' */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 163 0xa3 '£' */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x76, /* 0111011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 164 0xa4 '¤' */
+	0x00, /* 0000000 */
+	0x76, /* 0111011 */
+	0xdc, /* 1101110 */
+	0x00, /* 0000000 */
+	0xb8, /* 1011100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 165 0xa5 '¥' */
+	0x76, /* 0111011 */
+	0xdc, /* 1101110 */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xec, /* 1110110 */
+	0xec, /* 1110110 */
+	0xfc, /* 1111110 */
+	0xdc, /* 1101110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 166 0xa6 '¦' */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xd8, /* 1101100 */
+	0xd8, /* 1101100 */
+	0x7c, /* 0111110 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 167 0xa7 '§' */
+	0x00, /* 0000000 */
+	0x70, /* 0111000 */
+	0xd8, /* 1101100 */
+	0xd8, /* 1101100 */
+	0x70, /* 0111000 */
+	0x00, /* 0000000 */
+	0xf8, /* 1111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 168 0xa8 '¨' */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0xc0, /* 1100000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 169 0xa9 '©' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 170 0xaa 'ª' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 171 0xab '«' */
+	0x60, /* 0110000 */
+	0xe0, /* 1110000 */
+	0x62, /* 0110001 */
+	0x66, /* 0110011 */
+	0x6c, /* 0110110 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0xc0, /* 1100000 */
+	0xb8, /* 1011100 */
+	0x4c, /* 0100110 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x7c, /* 0111110 */
+
+	/* 172 0xac '¬' */
+	0x60, /* 0110000 */
+	0xe0, /* 1110000 */
+	0x62, /* 0110001 */
+	0x66, /* 0110011 */
+	0x6c, /* 0110110 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x6c, /* 0110110 */
+	0xdc, /* 1101110 */
+	0xb4, /* 1011010 */
+	0x7e, /* 0111111 */
+	0x0c, /* 0000110 */
+	0x0c, /* 0000110 */
+	0x00, /* 0000000 */
+
+	/* 173 0xad '­' */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0x78, /* 0111100 */
+	0x78, /* 0111100 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 174 0xae '®' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x36, /* 0011011 */
+	0x6c, /* 0110110 */
+	0xd8, /* 1101100 */
+	0x6c, /* 0110110 */
+	0x36, /* 0011011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 175 0xaf '¯' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xd8, /* 1101100 */
+	0x6c, /* 0110110 */
+	0x36, /* 0011011 */
+	0x6c, /* 0110110 */
+	0xd8, /* 1101100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 176 0xb0 '°' */
+	0x88, /* 1000100 */
+	0x22, /* 0010001 */
+	0x88, /* 1000100 */
+	0x22, /* 0010001 */
+	0x88, /* 1000100 */
+	0x22, /* 0010001 */
+	0x88, /* 1000100 */
+	0x22, /* 0010001 */
+	0x88, /* 1000100 */
+	0x22, /* 0010001 */
+	0x88, /* 1000100 */
+	0x22, /* 0010001 */
+	0x88, /* 1000100 */
+	0x22, /* 0010001 */
+
+	/* 177 0xb1 '±' */
+	0x54, /* 0101010 */
+	0xaa, /* 1010101 */
+	0x54, /* 0101010 */
+	0xaa, /* 1010101 */
+	0x54, /* 0101010 */
+	0xaa, /* 1010101 */
+	0x54, /* 0101010 */
+	0xaa, /* 1010101 */
+	0x54, /* 0101010 */
+	0xaa, /* 1010101 */
+	0x54, /* 0101010 */
+	0xaa, /* 1010101 */
+	0x54, /* 0101010 */
+	0xaa, /* 1010101 */
+
+	/* 178 0xb2 '²' */
+	0xee, /* 1110111 */
+	0xba, /* 1011101 */
+	0xee, /* 1110111 */
+	0xba, /* 1011101 */
+	0xee, /* 1110111 */
+	0xba, /* 1011101 */
+	0xee, /* 1110111 */
+	0xba, /* 1011101 */
+	0xee, /* 1110111 */
+	0xba, /* 1011101 */
+	0xee, /* 1110111 */
+	0xba, /* 1011101 */
+	0xee, /* 1110111 */
+	0xba, /* 1011101 */
+
+	/* 179 0xb3 '³' */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+
+	/* 180 0xb4 '´' */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0xf0, /* 1111000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+
+	/* 181 0xb5 'µ' */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0xf0, /* 1111000 */
+	0x30, /* 0011000 */
+	0xf0, /* 1111000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+
+	/* 182 0xb6 '¶' */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0xec, /* 1110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+
+	/* 183 0xb7 '·' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+
+	/* 184 0xb8 '¸' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xf0, /* 1111000 */
+	0x30, /* 0011000 */
+	0xf0, /* 1111000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+
+	/* 185 0xb9 '¹' */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0xec, /* 1110110 */
+	0x0c, /* 0000110 */
+	0xec, /* 1110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+
+	/* 186 0xba 'º' */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+
+	/* 187 0xbb '»' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0x0c, /* 0000110 */
+	0xec, /* 1110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+
+	/* 188 0xbc '¼' */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0xec, /* 1110110 */
+	0x0c, /* 0000110 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 189 0xbd '½' */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 190 0xbe '¾' */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0xf0, /* 1111000 */
+	0x30, /* 0011000 */
+	0xf0, /* 1111000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 191 0xbf '¿' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xf0, /* 1111000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+
+	/* 192 0xc0 'À' */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x3e, /* 0011111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 193 0xc1 'Á' */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0xfe, /* 1111111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 194 0xc2 'Â' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfe, /* 1111111 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+
+	/* 195 0xc3 'Ã' */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x3e, /* 0011111 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+
+	/* 196 0xc4 'Ä' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfe, /* 1111111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 197 0xc5 'Å' */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0xfe, /* 1111111 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+
+	/* 198 0xc6 'Æ' */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x3e, /* 0011111 */
+	0x30, /* 0011000 */
+	0x3e, /* 0011111 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+
+	/* 199 0xc7 'Ç' */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6e, /* 0110111 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+
+	/* 200 0xc8 'È' */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6e, /* 0110111 */
+	0x60, /* 0110000 */
+	0x7e, /* 0111111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 201 0xc9 'É' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x7e, /* 0111111 */
+	0x60, /* 0110000 */
+	0x6e, /* 0110111 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+
+	/* 202 0xca 'Ê' */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0xee, /* 1110111 */
+	0x00, /* 0000000 */
+	0xfe, /* 1111111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 203 0xcb 'Ë' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfe, /* 1111111 */
+	0x00, /* 0000000 */
+	0xee, /* 1110111 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+
+	/* 204 0xcc 'Ì' */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6e, /* 0110111 */
+	0x60, /* 0110000 */
+	0x6e, /* 0110111 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+
+	/* 205 0xcd 'Í' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfe, /* 1111111 */
+	0x00, /* 0000000 */
+	0xfe, /* 1111111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 206 0xce 'Î' */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0xee, /* 1110111 */
+	0x00, /* 0000000 */
+	0xee, /* 1110111 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+
+	/* 207 0xcf 'Ï' */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0xfe, /* 1111111 */
+	0x00, /* 0000000 */
+	0xfe, /* 1111111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 208 0xd0 'Ð' */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0xfe, /* 1111111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 209 0xd1 'Ñ' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfe, /* 1111111 */
+	0x00, /* 0000000 */
+	0xfe, /* 1111111 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+
+	/* 210 0xd2 'Ò' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfe, /* 1111111 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+
+	/* 211 0xd3 'Ó' */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x7e, /* 0111111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 212 0xd4 'Ô' */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x3e, /* 0011111 */
+	0x30, /* 0011000 */
+	0x3e, /* 0011111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 213 0xd5 'Õ' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x3e, /* 0011111 */
+	0x30, /* 0011000 */
+	0x3e, /* 0011111 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+
+	/* 214 0xd6 'Ö' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x7e, /* 0111111 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+
+	/* 215 0xd7 '×' */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0xfe, /* 1111111 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+
+	/* 216 0xd8 'Ø' */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0xfe, /* 1111111 */
+	0x30, /* 0011000 */
+	0xfe, /* 1111111 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+
+	/* 217 0xd9 'Ù' */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0xf0, /* 1111000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 218 0xda 'Ú' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x3e, /* 0011111 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+
+	/* 219 0xdb 'Û' */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+
+	/* 220 0xdc 'Ü' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+
+	/* 221 0xdd 'Ý' */
+	0xe0, /* 1110000 */
+	0xe0, /* 1110000 */
+	0xe0, /* 1110000 */
+	0xe0, /* 1110000 */
+	0xe0, /* 1110000 */
+	0xe0, /* 1110000 */
+	0xe0, /* 1110000 */
+	0xe0, /* 1110000 */
+	0xe0, /* 1110000 */
+	0xe0, /* 1110000 */
+	0xe0, /* 1110000 */
+	0xe0, /* 1110000 */
+	0xe0, /* 1110000 */
+	0xe0, /* 1110000 */
+
+	/* 222 0xde 'Þ' */
+	0x1e, /* 0001111 */
+	0x1e, /* 0001111 */
+	0x1e, /* 0001111 */
+	0x1e, /* 0001111 */
+	0x1e, /* 0001111 */
+	0x1e, /* 0001111 */
+	0x1e, /* 0001111 */
+	0x1e, /* 0001111 */
+	0x1e, /* 0001111 */
+	0x1e, /* 0001111 */
+	0x1e, /* 0001111 */
+	0x1e, /* 0001111 */
+	0x1e, /* 0001111 */
+	0x1e, /* 0001111 */
+
+	/* 223 0xdf 'ß' */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 224 0xe0 'à' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x76, /* 0111011 */
+	0xdc, /* 1101110 */
+	0xd8, /* 1101100 */
+	0xd8, /* 1101100 */
+	0xd8, /* 1101100 */
+	0xdc, /* 1101110 */
+	0x76, /* 0111011 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 225 0xe1 'á' */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xd8, /* 1101100 */
+	0xcc, /* 1100110 */
+	0xc6, /* 1100011 */
+	0xc6, /* 1100011 */
+	0xc6, /* 1100011 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 226 0xe2 'â' */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 227 0xe3 'ã' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfe, /* 1111111 */
+	0xfe, /* 1111111 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 228 0xe4 'ä' */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0xcc, /* 1100110 */
+	0x60, /* 0110000 */
+	0x30, /* 0011000 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 229 0xe5 'å' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x7e, /* 0111111 */
+	0xd8, /* 1101100 */
+	0xd8, /* 1101100 */
+	0xd8, /* 1101100 */
+	0xd8, /* 1101100 */
+	0xd8, /* 1101100 */
+	0x70, /* 0111000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 230 0xe6 'æ' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xf8, /* 1111100 */
+	0xc0, /* 1100000 */
+	0xc0, /* 1100000 */
+	0x80, /* 1000000 */
+
+	/* 231 0xe7 'ç' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x76, /* 0111011 */
+	0xdc, /* 1101110 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 232 0xe8 'è' */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0x30, /* 0011000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x30, /* 0011000 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 233 0xe9 'é' */
+	0x00, /* 0000000 */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xfc, /* 1111110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x6c, /* 0110110 */
+	0x38, /* 0011100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 234 0xea 'ê' */
+	0x00, /* 0000000 */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0xc6, /* 1100011 */
+	0xc6, /* 1100011 */
+	0xc6, /* 1100011 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0xee, /* 1110111 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 235 0xeb 'ë' */
+	0x00, /* 0000000 */
+	0x3c, /* 0011110 */
+	0x60, /* 0110000 */
+	0x30, /* 0011000 */
+	0x18, /* 0001100 */
+	0x7c, /* 0111110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x78, /* 0111100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 236 0xec 'ì' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x7c, /* 0111110 */
+	0xd6, /* 1101011 */
+	0xd6, /* 1101011 */
+	0xd6, /* 1101011 */
+	0x7c, /* 0111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 237 0xed 'í' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x06, /* 0000011 */
+	0x0c, /* 0000110 */
+	0x7c, /* 0111110 */
+	0xd6, /* 1101011 */
+	0xd6, /* 1101011 */
+	0xe6, /* 1110011 */
+	0x7c, /* 0111110 */
+	0x60, /* 0110000 */
+	0xc0, /* 1100000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 238 0xee 'î' */
+	0x00, /* 0000000 */
+	0x1c, /* 0001110 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x7c, /* 0111110 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x60, /* 0110000 */
+	0x30, /* 0011000 */
+	0x1c, /* 0001110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 239 0xef 'ï' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0xcc, /* 1100110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 240 0xf0 'ð' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 241 0xf1 'ñ' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0xfc, /* 1111110 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 242 0xf2 'ò' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x60, /* 0110000 */
+	0x30, /* 0011000 */
+	0x18, /* 0001100 */
+	0x0c, /* 0000110 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 243 0xf3 'ó' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x60, /* 0110000 */
+	0xc0, /* 1100000 */
+	0x60, /* 0110000 */
+	0x30, /* 0011000 */
+	0x18, /* 0001100 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 244 0xf4 'ô' */
+	0x00, /* 0000000 */
+	0x1c, /* 0001110 */
+	0x36, /* 0011011 */
+	0x36, /* 0011011 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+
+	/* 245 0xf5 'õ' */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0xd8, /* 1101100 */
+	0xd8, /* 1101100 */
+	0xd8, /* 1101100 */
+	0x70, /* 0111000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 246 0xf6 'ö' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 247 0xf7 '÷' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x76, /* 0111011 */
+	0xdc, /* 1101110 */
+	0x00, /* 0000000 */
+	0x76, /* 0111011 */
+	0xdc, /* 1101110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 248 0xf8 'ø' */
+	0x38, /* 0011100 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x38, /* 0011100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 249 0xf9 'ù' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 250 0xfa 'ú' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x30, /* 0011000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 251 0xfb 'û' */
+	0x1e, /* 0001111 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0x18, /* 0001100 */
+	0xd8, /* 1101100 */
+	0xd8, /* 1101100 */
+	0xd8, /* 1101100 */
+	0x78, /* 0111100 */
+	0x38, /* 0011100 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 252 0xfc 'ü' */
+	0xd8, /* 1101100 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x6c, /* 0110110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 253 0xfd 'ý' */
+	0x78, /* 0111100 */
+	0xcc, /* 1100110 */
+	0x18, /* 0001100 */
+	0x30, /* 0011000 */
+	0x64, /* 0110010 */
+	0xfc, /* 1111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 254 0xfe 'þ' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x7c, /* 0111110 */
+	0x7c, /* 0111110 */
+	0x7c, /* 0111110 */
+	0x7c, /* 0111110 */
+	0x7c, /* 0111110 */
+	0x7c, /* 0111110 */
+	0x7c, /* 0111110 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+	/* 255 0xff 'ÿ' */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+	0x00, /* 0000000 */
+
+};
+
+
+struct font_desc font_7x14 = {
+	FONT7x14_IDX,
+	"7x14",
+	7,
+	14,
+	fontdata_7x14,
+	0
+};
diff --git a/drivers/video/console/font_sun12x22.c b/drivers/video/console/font_sun12x22.c
index 05215d0c3e0..c7bd967ea10 100644
--- a/drivers/video/console/font_sun12x22.c
+++ b/drivers/video/console/font_sun12x22.c
@@ -29,24 +29,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 1 0x01 '^A' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x1f, 0xc0, /* 000111111100 */
+	0x30, 0x60, /* 001100000110 */
+	0x65, 0x30, /* 011001010011 */
+	0x6d, 0xb0, /* 011011011011 */
+	0x60, 0x30, /* 011000000011 */
+	0x62, 0x30, /* 011000100011 */
+	0x62, 0x30, /* 011000100011 */
+	0x60, 0x30, /* 011000000011 */
+	0x6f, 0xb0, /* 011011111011 */
+	0x67, 0x30, /* 011001110011 */
+	0x30, 0x60, /* 001100000110 */
+	0x1f, 0xc0, /* 000111111100 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -54,24 +53,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 2 0x02 '^B' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x1f, 0xc0, /* 000111111100 */
+	0x3f, 0xe0, /* 001111111110 */
+	0x7a, 0xf0, /* 011110101111 */
+	0x72, 0x70, /* 011100100111 */
+	0x7f, 0xf0, /* 011111111111 */
+	0x7d, 0xf0, /* 011111011111 */
+	0x7d, 0xf0, /* 011111011111 */
+	0x7f, 0xf0, /* 011111111111 */
+	0x70, 0x70, /* 011100000111 */
+	0x78, 0xf0, /* 011110001111 */
+	0x3f, 0xe0, /* 001111111110 */
+	0x1f, 0xc0, /* 000111111100 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -79,24 +77,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 3 0x03 '^C' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x19, 0x80, /* 000110011000 */
 	0x3f, 0xc0, /* 001111111100 */
+	0x7f, 0xe0, /* 011111111110 */
 	0x3f, 0xc0, /* 001111111100 */
 	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x1f, 0x80, /* 000111111000 */
+	0x1f, 0x80, /* 000111111000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x06, 0x00, /* 000001100000 */
+	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -104,24 +101,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 4 0x04 '^D' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x02, 0x00, /* 000000100000 */
+	0x07, 0x00, /* 000001110000 */
+	0x0f, 0x80, /* 000011111000 */
+	0x0f, 0x80, /* 000011111000 */
+	0x1f, 0xc0, /* 000111111100 */
+	0x1f, 0xc0, /* 000111111100 */
+	0x3f, 0xe0, /* 001111111110 */
+	0x1f, 0xc0, /* 000111111100 */
+	0x1f, 0xc0, /* 000111111100 */
+	0x0f, 0x80, /* 000011111000 */
+	0x0f, 0x80, /* 000011111000 */
+	0x07, 0x00, /* 000001110000 */
+	0x02, 0x00, /* 000000100000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -129,24 +125,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 5 0x05 '^E' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x02, 0x00, /* 000000100000 */
+	0x07, 0x00, /* 000001110000 */
+	0x07, 0x00, /* 000001110000 */
+	0x02, 0x00, /* 000000100000 */
+	0x18, 0xc0, /* 000110001100 */
+	0x3d, 0xe0, /* 001111011110 */
+	0x3d, 0xe0, /* 001111011110 */
+	0x1a, 0xc0, /* 000110101100 */
+	0x02, 0x00, /* 000000100000 */
+	0x07, 0x00, /* 000001110000 */
+	0x0f, 0x80, /* 000011111000 */
+	0x1f, 0xc0, /* 000111111100 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -154,23 +149,22 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 6 0x06 '^F' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x06, 0x00, /* 000001100000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x1f, 0x80, /* 000111111000 */
+	0x1f, 0x80, /* 000111111000 */
 	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x36, 0xc0, /* 001101101100 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x1f, 0x80, /* 000111111000 */
 	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -179,24 +173,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 7 0x07 '^G' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x06, 0x00, /* 000001100000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x1f, 0x80, /* 000111111000 */
+	0x1f, 0x80, /* 000111111000 */
 	0x3f, 0xc0, /* 001111111100 */
 	0x3f, 0xc0, /* 001111111100 */
+	0x1f, 0x80, /* 000111111000 */
+	0x1f, 0x80, /* 000111111000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x06, 0x00, /* 000001100000 */
+	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -204,49 +197,47 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 8 0x08 '^H' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xf9, 0xf0, /* 111110011111 */
+	0xf0, 0xf0, /* 111100001111 */
+	0xf0, 0xf0, /* 111100001111 */
+	0xe0, 0x70, /* 111000000111 */
+	0xe0, 0x70, /* 111000000111 */
+	0xc0, 0x30, /* 110000000011 */
+	0xc0, 0x30, /* 110000000011 */
+	0xe0, 0x70, /* 111000000111 */
+	0xe0, 0x70, /* 111000000111 */
+	0xf0, 0xf0, /* 111100001111 */
+	0xf0, 0xf0, /* 111100001111 */
+	0xf9, 0xf0, /* 111110011111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
 
 	/* 9 0x09 '^I' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x06, 0x00, /* 000001100000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x19, 0x80, /* 000110011000 */
+	0x19, 0x80, /* 000110011000 */
+	0x30, 0xc0, /* 001100001100 */
+	0x30, 0xc0, /* 001100001100 */
+	0x19, 0x80, /* 000110011000 */
+	0x19, 0x80, /* 000110011000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x06, 0x00, /* 000001100000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -254,74 +245,71 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 10 0x0a '^J' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xf9, 0xf0, /* 111110011111 */
+	0xf0, 0xf0, /* 111100001111 */
+	0xf0, 0xf0, /* 111100001111 */
+	0xe6, 0x70, /* 111001100111 */
+	0xe6, 0x70, /* 111001100111 */
+	0xcf, 0x30, /* 110011110011 */
+	0xcf, 0x30, /* 110011110011 */
+	0xe6, 0x70, /* 111001100111 */
+	0xe6, 0x70, /* 111001100111 */
+	0xf0, 0xf0, /* 111100001111 */
+	0xf0, 0xf0, /* 111100001111 */
+	0xf9, 0xf0, /* 111110011111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
 
 	/* 11 0x0b '^K' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x0f, 0xe0, /* 000011111110 */
+	0x0f, 0xe0, /* 000011111110 */
+	0x01, 0xe0, /* 000000011110 */
+	0x03, 0x60, /* 000000110110 */
+	0x06, 0x60, /* 000001100110 */
+	0x1e, 0x00, /* 000111100000 */
+	0x33, 0x00, /* 001100110000 */
+	0x33, 0x00, /* 001100110000 */
+	0x61, 0x80, /* 011000011000 */
+	0x61, 0x80, /* 011000011000 */
+	0x33, 0x00, /* 001100110000 */
+	0x33, 0x00, /* 001100110000 */
+	0x1e, 0x00, /* 000111100000 */
+	0x0c, 0x00, /* 000011000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 
 	/* 12 0x0c '^L' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x06, 0x00, /* 000001100000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x19, 0x80, /* 000110011000 */
+	0x30, 0xc0, /* 001100001100 */
+	0x30, 0xc0, /* 001100001100 */
+	0x19, 0x80, /* 000110011000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
 	0x3f, 0xc0, /* 001111111100 */
 	0x3f, 0xc0, /* 001111111100 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -329,149 +317,143 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 13 0x0d '^M' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x0f, 0xe0, /* 000011111110 */
+	0x0c, 0x60, /* 000011000110 */
+	0x0c, 0x60, /* 000011000110 */
+	0x0f, 0xe0, /* 000011111110 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x3c, 0x00, /* 001111000000 */
+	0x7c, 0x00, /* 011111000000 */
+	0x78, 0x00, /* 011110000000 */
+	0x30, 0x00, /* 001100000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 
 	/* 14 0x0e '^N' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x1f, 0xe0, /* 000111111110 */
+	0x18, 0x60, /* 000110000110 */
+	0x18, 0x60, /* 000110000110 */
+	0x1f, 0xe0, /* 000111111110 */
+	0x18, 0x60, /* 000110000110 */
+	0x18, 0x60, /* 000110000110 */
+	0x18, 0x60, /* 000110000110 */
+	0x18, 0x60, /* 000110000110 */
+	0x18, 0x60, /* 000110000110 */
+	0x18, 0x60, /* 000110000110 */
+	0x19, 0xe0, /* 000110011110 */
+	0x1b, 0xe0, /* 000110111110 */
+	0x1b, 0xc0, /* 000110111100 */
+	0x79, 0x80, /* 011110011000 */
+	0xf8, 0x00, /* 111110000000 */
+	0xf0, 0x00, /* 111100000000 */
+	0x60, 0x00, /* 011000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 
 	/* 15 0x0f '^O' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x18, 0xc0, /* 000110001100 */
+	0x0d, 0x80, /* 000011011000 */
+	0x6d, 0xb0, /* 011011011011 */
+	0x3d, 0xe0, /* 001111011110 */
 	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-
-	/* 16 0x10 '^P' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
+	0x3d, 0xe0, /* 001111011110 */
+	0x6d, 0xb0, /* 011011011011 */
+	0x0d, 0x80, /* 000011011000 */
+	0x18, 0xc0, /* 000110001100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+
+	/* 16 0x10 '^P' */
 	0x00, 0x00, /* 000000000000 */
+	0x00, 0x20, /* 000000000010 */
+	0x00, 0x60, /* 000000000110 */
+	0x00, 0xe0, /* 000000001110 */
+	0x01, 0xe0, /* 000000011110 */
+	0x03, 0xe0, /* 000000111110 */
+	0x07, 0xe0, /* 000001111110 */
+	0x0f, 0xe0, /* 000011111110 */
+	0x1f, 0xe0, /* 000111111110 */
+	0x3f, 0xe0, /* 001111111110 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x3f, 0xe0, /* 001111111110 */
+	0x1f, 0xe0, /* 000111111110 */
+	0x0f, 0xe0, /* 000011111110 */
+	0x07, 0xe0, /* 000001111110 */
+	0x03, 0xe0, /* 000000111110 */
+	0x01, 0xe0, /* 000000011110 */
+	0x00, 0xe0, /* 000000001110 */
+	0x00, 0x60, /* 000000000110 */
+	0x00, 0x20, /* 000000000010 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 
 	/* 17 0x11 '^Q' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x40, 0x00, /* 010000000000 */
+	0x60, 0x00, /* 011000000000 */
+	0x70, 0x00, /* 011100000000 */
+	0x78, 0x00, /* 011110000000 */
+	0x7c, 0x00, /* 011111000000 */
+	0x7e, 0x00, /* 011111100000 */
+	0x7f, 0x00, /* 011111110000 */
+	0x7f, 0x80, /* 011111111000 */
+	0x7f, 0xc0, /* 011111111100 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xc0, /* 011111111100 */
+	0x7f, 0x80, /* 011111111000 */
+	0x7f, 0x00, /* 011111110000 */
+	0x7e, 0x00, /* 011111100000 */
+	0x7c, 0x00, /* 011111000000 */
+	0x78, 0x00, /* 011110000000 */
+	0x70, 0x00, /* 011100000000 */
+	0x60, 0x00, /* 011000000000 */
+	0x40, 0x00, /* 010000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 
 	/* 18 0x12 '^R' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x04, 0x00, /* 000001000000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x1f, 0x00, /* 000111110000 */
+	0x3f, 0x80, /* 001111111000 */
+	0x7f, 0xc0, /* 011111111100 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x7f, 0xc0, /* 011111111100 */
+	0x3f, 0x80, /* 001111111000 */
+	0x1f, 0x00, /* 000111110000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x04, 0x00, /* 000001000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -551,99 +533,95 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 22 0x16 '^V' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-
-	/* 23 0x17 '^W' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xe0, /* 011111111110 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 
-	/* 24 0x18 '^X' */
-	/* FIXME */
+	/* 23 0x17 '^W' */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x04, 0x00, /* 000001000000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x1f, 0x00, /* 000111110000 */
+	0x3f, 0x80, /* 001111111000 */
+	0x7f, 0xc0, /* 011111111100 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x7f, 0xc0, /* 011111111100 */
+	0x3f, 0x80, /* 001111111000 */
+	0x1f, 0x00, /* 000111110000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x04, 0x00, /* 000001000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xe0, /* 011111111110 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+
+	/* 24 0x18 '^X' */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-
-	/* 25 0x19 '^Y' */
-	/* FIXME */
+	0x04, 0x00, /* 000001000000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x1f, 0x00, /* 000111110000 */
+	0x3f, 0x80, /* 001111111000 */
+	0x7f, 0xc0, /* 011111111100 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x00, 0x00, /* 000000000000 */
+
+	/* 25 0x19 '^Y' */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x7f, 0xc0, /* 011111111100 */
+	0x3f, 0x80, /* 001111111000 */
+	0x1f, 0x00, /* 000111110000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x04, 0x00, /* 000001000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -651,24 +629,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 26 0x1a '^Z' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x00, 0x00, /* 000000000000 */
+	0x08, 0x00, /* 000010000000 */
+	0x18, 0x00, /* 000110000000 */
+	0x38, 0x00, /* 001110000000 */
+	0x7f, 0xe0, /* 011111111110 */
+	0xff, 0xe0, /* 111111111110 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x38, 0x00, /* 001110000000 */
+	0x18, 0x00, /* 000110000000 */
+	0x08, 0x00, /* 000010000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -676,24 +653,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 27 0x1b '^[' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x00, 0x00, /* 000000000000 */
+	0x01, 0x00, /* 000000010000 */
+	0x01, 0x80, /* 000000011000 */
+	0x01, 0xc0, /* 000000011100 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xf0, /* 011111111111 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x01, 0xc0, /* 000000011100 */
+	0x01, 0x80, /* 000000011000 */
+	0x01, 0x00, /* 000000010000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -701,24 +677,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 28 0x1c '^\' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x30, 0x00, /* 001100000000 */
+	0x30, 0x00, /* 001100000000 */
+	0x3f, 0xe0, /* 001111111110 */
+	0x3f, 0xe0, /* 001111111110 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -726,24 +701,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 29 0x1d '^]' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x00, 0x00, /* 000000000000 */
+	0x09, 0x00, /* 000010010000 */
+	0x19, 0x80, /* 000110011000 */
+	0x39, 0xc0, /* 001110011100 */
+	0x7f, 0xe0, /* 011111111110 */
+	0xff, 0xf0, /* 111111111111 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x39, 0xc0, /* 001110011100 */
+	0x19, 0x80, /* 000110011000 */
+	0x09, 0x00, /* 000010010000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -751,24 +725,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 30 0x1e '^^' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x00, 0x00, /* 000000000000 */
+	0x04, 0x00, /* 000001000000 */
+	0x04, 0x00, /* 000001000000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x1f, 0x00, /* 000111110000 */
+	0x1f, 0x00, /* 000111110000 */
+	0x3f, 0x80, /* 001111111000 */
+	0x3f, 0x80, /* 001111111000 */
+	0x7f, 0xc0, /* 011111111100 */
+	0x7f, 0xc0, /* 011111111100 */
+	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -776,24 +749,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 31 0x1f '^_' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x00, 0x00, /* 000000000000 */
+	0x7f, 0xc0, /* 011111111100 */
+	0x7f, 0xc0, /* 011111111100 */
+	0x3f, 0x80, /* 001111111000 */
+	0x3f, 0x80, /* 001111111000 */
+	0x1f, 0x00, /* 000111110000 */
+	0x1f, 0x00, /* 000111110000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x04, 0x00, /* 000001000000 */
+	0x04, 0x00, /* 000001000000 */
+	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -3081,29 +3053,28 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 127 0x7f '.' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0xff, 0xf0, /* 111111111111 */
+	0xff, 0xf0, /* 111111111111 */
+	0x00, 0x00, /* 000000000000 */
 
 	/* 128 0x80 '.' */
 	0x00, 0x00, /* 000000000000 */
@@ -3826,24 +3797,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 158 0x9e '.' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x7f, 0x80, /* 011111111000 */
+	0x30, 0xc0, /* 001100001100 */
+	0x30, 0x60, /* 001100000110 */
+	0x30, 0x60, /* 001100000110 */
+	0x30, 0x60, /* 001100000110 */
+	0x30, 0xc0, /* 001100001100 */
+	0x37, 0x80, /* 001101111000 */
+	0x30, 0x00, /* 001100000000 */
+	0x33, 0x00, /* 001100110000 */
+	0x37, 0x80, /* 001101111000 */
+	0x33, 0x00, /* 001100110000 */
+	0x33, 0x00, /* 001100110000 */
+	0x33, 0x30, /* 001100110011 */
+	0x31, 0xe0, /* 001100011110 */
+	0x78, 0xc0, /* 011110001100 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -3851,28 +3821,27 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 159 0x9f '.' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x00, 0xc0, /* 000000001100 */
+	0x01, 0xe0, /* 000000011110 */
+	0x03, 0x30, /* 000000110011 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x3f, 0xe0, /* 001111111110 */
+	0x7f, 0xc0, /* 011111111100 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0xcc, 0x00, /* 110011000000 */
+	0x78, 0x00, /* 011110000000 */
+	0x30, 0x00, /* 001100000000 */
 	0x00, 0x00, /* 000000000000 */
 
 	/* 160 0xa0 '.' */
@@ -4092,24 +4061,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 169 0xa9 '.' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x3f, 0xc0, /* 001111111100 */
 	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x30, 0x00, /* 001100000000 */
+	0x30, 0x00, /* 001100000000 */
+	0x30, 0x00, /* 001100000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -5413,24 +5381,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 224 0xe0 '.' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x00, 0x00, /* 000000000000 */
+	0x0f, 0x60, /* 000011110110 */
+	0x13, 0xe0, /* 000100111110 */
+	0x21, 0xc0, /* 001000011100 */
+	0x60, 0xc0, /* 011000001100 */
+	0x60, 0xc0, /* 011000001100 */
+	0x60, 0xc0, /* 011000001100 */
+	0x60, 0xc0, /* 011000001100 */
+	0x70, 0x80, /* 011100001000 */
+	0x39, 0xc0, /* 001110011100 */
+	0x1f, 0x60, /* 000111110110 */
+	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -5462,24 +5429,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 226 0xe2 '.' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x3f, 0xe0, /* 001111111110 */
+	0x3f, 0xe0, /* 001111111110 */
+	0x30, 0x60, /* 001100000110 */
+	0x30, 0x60, /* 001100000110 */
+	0x30, 0x00, /* 001100000000 */
+	0x30, 0x00, /* 001100000000 */
+	0x30, 0x00, /* 001100000000 */
+	0x30, 0x00, /* 001100000000 */
+	0x30, 0x00, /* 001100000000 */
+	0x30, 0x00, /* 001100000000 */
+	0x30, 0x00, /* 001100000000 */
+	0x30, 0x00, /* 001100000000 */
+	0x30, 0x00, /* 001100000000 */
+	0x30, 0x00, /* 001100000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -5487,49 +5453,47 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 227 0xe3 '.' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x19, 0x80, /* 000110011000 */
+	0x19, 0x80, /* 000110011000 */
+	0x19, 0x80, /* 000110011000 */
+	0x19, 0x80, /* 000110011000 */
+	0x19, 0x80, /* 000110011000 */
+	0x19, 0x80, /* 000110011000 */
+	0x19, 0x80, /* 000110011000 */
+	0x19, 0x80, /* 000110011000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-
-	/* 228 0xe4 '.' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+
+	/* 228 0xe4 '.' */
 	0x00, 0x00, /* 000000000000 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x60, 0x60, /* 011000000110 */
+	0x30, 0x60, /* 001100000110 */
+	0x30, 0x00, /* 001100000000 */
+	0x18, 0x00, /* 000110000000 */
+	0x18, 0x00, /* 000110000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x18, 0x00, /* 000110000000 */
+	0x18, 0x00, /* 000110000000 */
+	0x30, 0x00, /* 001100000000 */
+	0x30, 0x60, /* 001100000110 */
+	0x60, 0x60, /* 011000000110 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xe0, /* 011111111110 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -5537,24 +5501,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 229 0xe5 '.' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x07, 0xe0, /* 000001111110 */
+	0x0f, 0xe0, /* 000011111110 */
+	0x13, 0x80, /* 000100111000 */
+	0x21, 0xc0, /* 001000011100 */
+	0x60, 0xc0, /* 011000001100 */
+	0x60, 0xc0, /* 011000001100 */
+	0x60, 0xc0, /* 011000001100 */
+	0x60, 0xc0, /* 011000001100 */
+	0x70, 0x80, /* 011100001000 */
+	0x39, 0x00, /* 001110010000 */
+	0x1e, 0x00, /* 000111100000 */
+	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -5586,24 +5549,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 231 0xe7 '.' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x19, 0x80, /* 000110011000 */
 	0x3f, 0xc0, /* 001111111100 */
+	0x66, 0x60, /* 011001100110 */
+	0x66, 0x60, /* 011001100110 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -5611,24 +5573,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 232 0xe8 '.' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x19, 0x80, /* 000110011000 */
+	0x30, 0xc0, /* 001100001100 */
+	0x30, 0xc0, /* 001100001100 */
+	0x19, 0x80, /* 000110011000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xe0, /* 011111111110 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -5636,24 +5597,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 233 0xe9 '.' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x0f, 0x00, /* 000011110000 */
+	0x1f, 0x80, /* 000111111000 */
+	0x30, 0xc0, /* 001100001100 */
+	0x60, 0x60, /* 011000000110 */
+	0x60, 0x60, /* 011000000110 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x60, 0x60, /* 011000000110 */
+	0x60, 0x60, /* 011000000110 */
+	0x30, 0xc0, /* 001100001100 */
+	0x1f, 0x80, /* 000111111000 */
+	0x0f, 0x00, /* 000011110000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -5661,24 +5621,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 234 0xea '.' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x1f, 0x00, /* 000111110000 */
+	0x31, 0x80, /* 001100011000 */
+	0x30, 0xc0, /* 001100001100 */
+	0x30, 0xc0, /* 001100001100 */
+	0x60, 0x60, /* 011000000110 */
+	0x60, 0x60, /* 011000000110 */
+	0x30, 0xc0, /* 001100001100 */
+	0x30, 0xc0, /* 001100001100 */
+	0x19, 0x80, /* 000110011000 */
+	0x19, 0x80, /* 000110011000 */
+	0xd9, 0xb0, /* 110110011011 */
+	0x79, 0xe0, /* 011110011110 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -5686,24 +5645,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 235 0xeb '.' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x07, 0x80, /* 000001111000 */
+	0x0c, 0xc0, /* 000011001100 */
+	0x18, 0x60, /* 000110000110 */
+	0x18, 0x00, /* 000110000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x06, 0x00, /* 000001100000 */
+	0x03, 0x00, /* 000000110000 */
+	0x0f, 0x80, /* 000011111000 */
+	0x11, 0xc0, /* 000100011100 */
+	0x20, 0xe0, /* 001000001110 */
+	0x60, 0x60, /* 011000000110 */
+	0x60, 0x60, /* 011000000110 */
+	0x70, 0x40, /* 011100000100 */
+	0x38, 0x80, /* 001110001000 */
+	0x1f, 0x00, /* 000111110000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -5711,74 +5669,71 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 236 0xec '.' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x39, 0xc0, /* 001110011100 */
+	0x6f, 0x60, /* 011011110110 */
+	0x66, 0x60, /* 011001100110 */
+	0xc6, 0x30, /* 110001100011 */
+	0xc6, 0x30, /* 110001100011 */
+	0x66, 0x60, /* 011001100110 */
+	0x6f, 0x60, /* 011011110110 */
+	0x39, 0xc0, /* 001110011100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-
-	/* 237 0xed '.' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+
+	/* 237 0xed '.' */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x00, 0xc0, /* 000000001100 */
+	0x00, 0xc0, /* 000000001100 */
+	0x01, 0x80, /* 000000011000 */
+	0x01, 0x80, /* 000000011000 */
+	0x3b, 0xc0, /* 001110111100 */
+	0x6f, 0x60, /* 011011110110 */
+	0x66, 0x60, /* 011001100110 */
+	0xc6, 0x30, /* 110001100011 */
+	0xc6, 0x30, /* 110001100011 */
+	0x66, 0x60, /* 011001100110 */
+	0x6f, 0x60, /* 011011110110 */
+	0x3d, 0xc0, /* 001111011100 */
+	0x18, 0x00, /* 000110000000 */
+	0x18, 0x00, /* 000110000000 */
+	0x30, 0x00, /* 001100000000 */
+	0x30, 0x00, /* 001100000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 
 	/* 238 0xee '.' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x01, 0xc0, /* 000000011100 */
+	0x03, 0x00, /* 000000110000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x18, 0x00, /* 000110000000 */
+	0x1f, 0xc0, /* 000111111100 */
+	0x18, 0x00, /* 000110000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x03, 0x00, /* 000000110000 */
+	0x01, 0xc0, /* 000000011100 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -5786,24 +5741,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 239 0xef '.' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x0f, 0x00, /* 000011110000 */
+	0x1f, 0x80, /* 000111111000 */
+	0x39, 0xc0, /* 001110011100 */
+	0x30, 0xc0, /* 001100001100 */
+	0x30, 0xc0, /* 001100001100 */
+	0x30, 0xc0, /* 001100001100 */
+	0x30, 0xc0, /* 001100001100 */
+	0x30, 0xc0, /* 001100001100 */
+	0x30, 0xc0, /* 001100001100 */
+	0x30, 0xc0, /* 001100001100 */
+	0x30, 0xc0, /* 001100001100 */
+	0x30, 0xc0, /* 001100001100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -5811,24 +5765,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 240 0xf0 '.' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xe0, /* 011111111110 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -5860,24 +5813,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 242 0xf2 '.' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x60, 0x00, /* 011000000000 */
+	0x38, 0x00, /* 001110000000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x03, 0x80, /* 000000111000 */
+	0x00, 0xe0, /* 000000001110 */
+	0x00, 0xe0, /* 000000001110 */
+	0x03, 0x80, /* 000000111000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x38, 0x00, /* 001110000000 */
+	0x60, 0x00, /* 011000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xe0, /* 011111111110 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -5885,24 +5837,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 243 0xf3 '.' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x00, 0x60, /* 000000000110 */
+	0x01, 0xc0, /* 000000011100 */
+	0x07, 0x00, /* 000001110000 */
+	0x1c, 0x00, /* 000111000000 */
+	0x70, 0x00, /* 011100000000 */
+	0x70, 0x00, /* 011100000000 */
+	0x1c, 0x00, /* 000111000000 */
+	0x07, 0x00, /* 000001110000 */
+	0x01, 0xc0, /* 000000011100 */
+	0x00, 0x60, /* 000000000110 */
 	0x00, 0x00, /* 000000000000 */
+	0x7f, 0xe0, /* 011111111110 */
+	0x7f, 0xe0, /* 011111111110 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -5910,54 +5861,52 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 244 0xf4 '.' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x03, 0x80, /* 000000111000 */
+	0x07, 0xc0, /* 000001111100 */
+	0x0c, 0x60, /* 000011000110 */
+	0x0c, 0x60, /* 000011000110 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
+	0x0c, 0x00, /* 000011000000 */
 
 	/* 245 0xf5 '.' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x1c, 0x00, /* 000111000000 */
+	0x3e, 0x00, /* 001111100000 */
+	0x63, 0x00, /* 011000110000 */
+	0x63, 0x00, /* 011000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
+	0x03, 0x00, /* 000000110000 */
 
 	/* 246 0xf6 '.' */
 	0x00, 0x00, /* 000000000000 */
@@ -5984,24 +5933,23 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 247 0xf7 '.' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x38, 0x00, /* 001110000000 */
+	0x6c, 0x00, /* 011011000000 */
+	0x06, 0x30, /* 000001100011 */
+	0x03, 0x60, /* 000000110110 */
+	0x39, 0xc0, /* 001110011100 */
+	0x6c, 0x00, /* 011011000000 */
+	0x06, 0x30, /* 000001100011 */
+	0x03, 0x60, /* 000000110110 */
+	0x01, 0xc0, /* 000000011100 */
+	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -6033,44 +5981,31 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 
 	/* 249 0xf9 '.' */
-	/* FIXME */
-	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-
-	/* 250 0xfa '.' */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x1c, 0x00, /* 000111000000 */
+	0x3e, 0x00, /* 001111100000 */
+	0x3e, 0x00, /* 001111100000 */
+	0x3e, 0x00, /* 001111100000 */
+	0x1c, 0x00, /* 000111000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+
+	/* 250 0xfa '.' */
 	0x00, 0x00, /* 000000000000 */
-	0x06, 0x00, /* 000001100000 */
-	0x0f, 0x00, /* 000011110000 */
-	0x0f, 0x00, /* 000011110000 */
-	0x06, 0x00, /* 000001100000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
@@ -6080,51 +6015,61 @@ static unsigned char fontdata_sun12x22[FONTDATAMAX] = {
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-
-	/* 251 0xfb '.' */
-	/* FIXME */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+	0x18, 0x00, /* 000110000000 */
+	0x3c, 0x00, /* 001111000000 */
+	0x3c, 0x00, /* 001111000000 */
+	0x18, 0x00, /* 000110000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
+
+	/* 251 0xfb '.' */
 	0x00, 0x00, /* 000000000000 */
+	0x07, 0xe0, /* 000001111110 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x06, 0x00, /* 000001100000 */
+	0xc6, 0x00, /* 110001100000 */
+	0x66, 0x00, /* 011001100000 */
+	0x36, 0x00, /* 001101100000 */
+	0x1e, 0x00, /* 000111100000 */
+	0x0e, 0x00, /* 000011100000 */
+	0x06, 0x00, /* 000001100000 */
+	0x02, 0x00, /* 000000100000 */
 	0x00, 0x00, /* 000000000000 */
 
 	/* 252 0xfc '.' */
-	/* FIXME */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x13, 0x80, /* 000100111000 */
+	0x3d, 0xc0, /* 001111011100 */
+	0x18, 0xc0, /* 000110001100 */
+	0x18, 0xc0, /* 000110001100 */
+	0x18, 0xc0, /* 000110001100 */
+	0x18, 0xc0, /* 000110001100 */
+	0x3d, 0xe0, /* 001111011110 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
+	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
-	0x3f, 0xc0, /* 001111111100 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
 	0x00, 0x00, /* 000000000000 */
diff --git a/drivers/video/console/fonts.c b/drivers/video/console/fonts.c
index 465d678230a..e79b2970264 100644
--- a/drivers/video/console/fonts.c
+++ b/drivers/video/console/fonts.c
@@ -36,6 +36,10 @@ static struct font_desc *fonts[] = {
 #undef NO_FONTS
     &font_vga_6x11,
 #endif
+#ifdef CONFIG_FONT_7x14
+#undef NO_FONTS
+    &font_7x14,
+#endif
 #ifdef CONFIG_FONT_SUN8x16
 #undef NO_FONTS
     &font_sun_8x16,
@@ -44,6 +48,10 @@ static struct font_desc *fonts[] = {
 #undef NO_FONTS
     &font_sun_12x22,
 #endif
+#ifdef CONFIG_FONT_10x18
+#undef NO_FONTS
+    &font_10x18,
+#endif
 #ifdef CONFIG_FONT_ACORN_8x8
 #undef NO_FONTS
     &font_acorn_8x8,
diff --git a/include/linux/font.h b/include/linux/font.h
index fc2d690c9d5..8fc80a7d78a 100644
--- a/include/linux/font.h
+++ b/include/linux/font.h
@@ -25,19 +25,23 @@ struct font_desc {
 #define VGA8x16_IDX	1
 #define PEARL8x8_IDX	2
 #define VGA6x11_IDX	3
-#define SUN8x16_IDX	4
-#define SUN12x22_IDX	5
-#define ACORN8x8_IDX	6
-#define	MINI4x6_IDX	7
+#define FONT7x14_IDX	4
+#define	FONT10x18_IDX	5
+#define SUN8x16_IDX	6
+#define SUN12x22_IDX	7
+#define ACORN8x8_IDX	8
+#define	MINI4x6_IDX	9
 
 extern struct font_desc	font_vga_8x8,
-				font_vga_8x16,
-				font_pearl_8x8,
-				font_vga_6x11,
-				font_sun_8x16,
-				font_sun_12x22,
-				font_acorn_8x8,
-				font_mini_4x6;
+			font_vga_8x16,
+			font_pearl_8x8,
+			font_vga_6x11,
+			font_7x14,
+			font_10x18,
+			font_sun_8x16,
+			font_sun_12x22,
+			font_acorn_8x8,
+			font_mini_4x6;
 
 /* Find a font with a specific name */
 
-- 
cgit v1.2.3-70-g09d2


From f1ab5dac251bb4514607918b0019a3b3f5f5fb48 Mon Sep 17 00:00:00 2001
From: James Simmons <jsimmons@pentafluge.infradead.org>
Date: Tue, 21 Jun 2005 17:17:07 -0700
Subject: [PATCH] fbdev: stack reduction

Shrink the stack when calling the drawing alignment functions.

Signed-off-by: James Simmons <jsimmons@www.infradead.org>
Cc: "Antonino A. Daplas" <adaplas@hotpop.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/console/bitblit.c |  9 ++++-----
 drivers/video/fbmem.c           | 14 +++++---------
 drivers/video/nvidia/nvidia.c   | 17 +++++++----------
 drivers/video/riva/fbdev.c      | 18 +++++++-----------
 drivers/video/softcursor.c      |  5 +----
 include/linux/fb.h              |  7 ++-----
 6 files changed, 26 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/drivers/video/console/bitblit.c b/drivers/video/console/bitblit.c
index 4eef20790c3..3c731577fed 100644
--- a/drivers/video/console/bitblit.c
+++ b/drivers/video/console/bitblit.c
@@ -157,9 +157,9 @@ static void bit_putcs(struct vc_data *vc, struct fb_info *info,
 					src = buf;
 				}
 
-				fb_sysmove_buf_unaligned(info, &info->pixmap, dst, pitch,
-					       src, idx, image.height,
-					       shift_high, shift_low, mod);
+				fb_pad_unaligned_buffer(dst, pitch, src, idx,
+						image.height, shift_high,
+						shift_low, mod);
 				shift_low += mod;
 				dst += (shift_low >= 8) ? width : width - 1;
 				shift_low &= 7;
@@ -175,8 +175,7 @@ static void bit_putcs(struct vc_data *vc, struct fb_info *info,
 					src = buf;
 				}
 
-				fb_sysmove_buf_aligned(info, &info->pixmap, dst, pitch,
-					     src, idx, image.height);
+				fb_pad_aligned_buffer(dst, pitch, src, idx, image.height);
 				dst += width;
 			}
 		}
diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c
index 78907a87349..6f871a8b905 100644
--- a/drivers/video/fbmem.c
+++ b/drivers/video/fbmem.c
@@ -78,9 +78,7 @@ EXPORT_SYMBOL(fb_get_color_depth);
 /*
  * Data padding functions.
  */
-void fb_sysmove_buf_aligned(struct fb_info *info, struct fb_pixmap *buf,
-			    u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch,
-			    u32 height)
+void fb_pad_aligned_buffer(u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch, u32 height)
 {
 	int i;
 
@@ -90,12 +88,10 @@ void fb_sysmove_buf_aligned(struct fb_info *info, struct fb_pixmap *buf,
 		dst += d_pitch;
 	}
 }
-EXPORT_SYMBOL(fb_sysmove_buf_aligned);
+EXPORT_SYMBOL(fb_pad_aligned_buffer);
 
-void fb_sysmove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf,
-			      u8 *dst, u32 d_pitch, u8 *src, u32 idx,
-			      u32 height, u32 shift_high, u32 shift_low,
-			      u32 mod)
+void fb_pad_unaligned_buffer(u8 *dst, u32 d_pitch, u8 *src, u32 idx, u32 height,
+				u32 shift_high, u32 shift_low, u32 mod)
 {
 	u8 mask = (u8) (0xfff << shift_high), tmp;
 	int i, j;
@@ -122,7 +118,7 @@ void fb_sysmove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf,
 		dst += d_pitch;
 	}
 }
-EXPORT_SYMBOL(fb_sysmove_buf_unaligned);
+EXPORT_SYMBOL(fb_pad_unaligned_buffer);
 
 /*
  * we need to lock this section since fb_cursor
diff --git a/drivers/video/nvidia/nvidia.c b/drivers/video/nvidia/nvidia.c
index 47733f58153..266ff5d7ac3 100644
--- a/drivers/video/nvidia/nvidia.c
+++ b/drivers/video/nvidia/nvidia.c
@@ -516,9 +516,9 @@ static struct backlight_controller nvidia_backlight_controller = {
 static void nvidiafb_load_cursor_image(struct nvidia_par *par, u8 * data8,
 				       u16 bg, u16 fg, u32 w, u32 h)
 {
+	u32 *data = (u32 *) data8;
 	int i, j, k = 0;
 	u32 b, tmp;
-	u32 *data = (u32 *) data8;
 
 	w = (w + 1) & ~1;
 
@@ -890,11 +890,11 @@ static int nvidiafb_cursor(struct fb_info *info, struct fb_cursor *cursor)
 {
 	struct nvidia_par *par = info->par;
 	u8 data[MAX_CURS * MAX_CURS / 8];
-	u16 fg, bg;
 	int i, set = cursor->set;
+	u16 fg, bg;
 
 	if (cursor->image.width > MAX_CURS || cursor->image.height > MAX_CURS)
-		return soft_cursor(info, cursor);
+		return -ENXIO;
 
 	NVShowHideCursor(par, 0);
 
@@ -931,21 +931,18 @@ static int nvidiafb_cursor(struct fb_info *info, struct fb_cursor *cursor)
 		if (src) {
 			switch (cursor->rop) {
 			case ROP_XOR:
-				for (i = 0; i < s_pitch * cursor->image.height;
-				     i++)
+				for (i = 0; i < s_pitch * cursor->image.height; i++)
 					src[i] = dat[i] ^ msk[i];
 				break;
 			case ROP_COPY:
 			default:
-				for (i = 0; i < s_pitch * cursor->image.height;
-				     i++)
+				for (i = 0; i < s_pitch * cursor->image.height; i++)
 					src[i] = dat[i] & msk[i];
 				break;
 			}
 
-			fb_sysmove_buf_aligned(info, &info->pixmap, data,
-					       d_pitch, src, s_pitch,
-					       cursor->image.height);
+			fb_pad_aligned_buffer(data, d_pitch, src, s_pitch,
+						cursor->image.height);
 
 			bg = ((info->cmap.red[bg_idx] & 0xf8) << 7) |
 			    ((info->cmap.green[bg_idx] & 0xf8) << 2) |
diff --git a/drivers/video/riva/fbdev.c b/drivers/video/riva/fbdev.c
index b0c886de040..7540f60af94 100644
--- a/drivers/video/riva/fbdev.c
+++ b/drivers/video/riva/fbdev.c
@@ -1582,12 +1582,11 @@ static int rivafb_cursor(struct fb_info *info, struct fb_cursor *cursor)
 {
 	struct riva_par *par = (struct riva_par *) info->par;
 	u8 data[MAX_CURS * MAX_CURS/8];
-	u16 fg, bg;
 	int i, set = cursor->set;
+	u16 fg, bg;
 
-	if (cursor->image.width > MAX_CURS ||
-	    cursor->image.height > MAX_CURS)
-		return soft_cursor(info, cursor);
+	if (cursor->image.width > MAX_CURS || cursor->image.height > MAX_CURS)
+		return -ENXIO;
 
 	par->riva.ShowHideCursor(&par->riva, 0);
 
@@ -1625,21 +1624,18 @@ static int rivafb_cursor(struct fb_info *info, struct fb_cursor *cursor)
 		if (src) {
 			switch (cursor->rop) {
 			case ROP_XOR:
-				for (i = 0; i < s_pitch * cursor->image.height;
-				     i++)
+				for (i = 0; i < s_pitch * cursor->image.height; i++)
 					src[i] = dat[i] ^ msk[i];
 				break;
 			case ROP_COPY:
 			default:
-				for (i = 0; i < s_pitch * cursor->image.height;
-				     i++)
+				for (i = 0; i < s_pitch * cursor->image.height; i++)
 					src[i] = dat[i] & msk[i];
 				break;
 			}
 
-			fb_sysmove_buf_aligned(info, &info->pixmap, data,
-					       d_pitch, src, s_pitch,
-					       cursor->image.height);
+			fb_pad_aligned_buffer(data, d_pitch, src, s_pitch,
+						cursor->image.height);
 
 			bg = ((info->cmap.red[bg_idx] & 0xf8) << 7) |
 				((info->cmap.green[bg_idx] & 0xf8) << 2) |
diff --git a/drivers/video/softcursor.c b/drivers/video/softcursor.c
index a6c5ca88d6b..229c4bc3507 100644
--- a/drivers/video/softcursor.c
+++ b/drivers/video/softcursor.c
@@ -58,13 +58,10 @@ int soft_cursor(struct fb_info *info, struct fb_cursor *cursor)
 	} else 
 		memcpy(src, image->data, dsize);
 	
-	fb_sysmove_buf_aligned(info, &info->pixmap, dst, d_pitch, src,
-				s_pitch, image->height);
-
+	fb_pad_aligned_buffer(dst, d_pitch, src, s_pitch, image->height);
 	image->data = dst;
 	info->fbops->fb_imageblit(info, image);
 	kfree(src);
-
 	return 0;
 }
 
diff --git a/include/linux/fb.h b/include/linux/fb.h
index e14942805db..bc24beeed97 100644
--- a/include/linux/fb.h
+++ b/include/linux/fb.h
@@ -816,12 +816,9 @@ extern int unregister_framebuffer(struct fb_info *fb_info);
 extern int fb_prepare_logo(struct fb_info *fb_info);
 extern int fb_show_logo(struct fb_info *fb_info);
 extern char* fb_get_buffer_offset(struct fb_info *info, struct fb_pixmap *buf, u32 size);
-extern void fb_sysmove_buf_unaligned(struct fb_info *info, struct fb_pixmap *buf,
-				u8 *dst, u32 d_pitch, u8 *src, u32 idx,
+extern void fb_pad_unaligned_buffer(u8 *dst, u32 d_pitch, u8 *src, u32 idx,
 				u32 height, u32 shift_high, u32 shift_low, u32 mod);
-extern void fb_sysmove_buf_aligned(struct fb_info *info, struct fb_pixmap *buf,
-				u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch,
-				u32 height);
+extern void fb_pad_aligned_buffer(u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch, u32 height);
 extern void fb_set_suspend(struct fb_info *info, int state);
 extern int fb_get_color_depth(struct fb_var_screeninfo *var);
 extern int fb_get_options(char *name, char **option);
-- 
cgit v1.2.3-70-g09d2


From 06d91a5fe0b50c9060e70bdf7786f8a3c66249db Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Tue, 21 Jun 2005 17:17:12 -0700
Subject: [PATCH] md: improve locking on 'safemode' and move superblock writes

When md marks the superblock dirty before a write, it calls
generic_make_request (to write the superblock) from within
generic_make_request (to write the first dirty block), which could cause
problems later.

With this patch, the superblock write is always done by the helper thread, and
write request are delayed until that write completes.

Also, the locking around marking the array dirty and writing the superblock is
improved to avoid possible races.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/md/md.c           | 73 ++++++++++++++++++++++++++++++++++++++---------
 drivers/md/raid1.c        |  4 ++-
 drivers/md/raid10.c       |  5 +++-
 drivers/md/raid5.c        |  6 ++--
 drivers/md/raid6main.c    |  6 ++--
 include/linux/raid/md.h   |  2 +-
 include/linux/raid/md_k.h |  7 +++++
 7 files changed, 82 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index c842e34d850..177d2a7d7ce 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -218,6 +218,8 @@ static mddev_t * mddev_find(dev_t unit)
 	INIT_LIST_HEAD(&new->all_mddevs);
 	init_timer(&new->safemode_timer);
 	atomic_set(&new->active, 1);
+	bio_list_init(&new->write_list);
+	spin_lock_init(&new->write_lock);
 
 	new->queue = blk_alloc_queue(GFP_KERNEL);
 	if (!new->queue) {
@@ -1251,9 +1253,11 @@ static void md_update_sb(mddev_t * mddev)
 	int err, count = 100;
 	struct list_head *tmp;
 	mdk_rdev_t *rdev;
+	int sync_req;
 
-	mddev->sb_dirty = 0;
 repeat:
+	spin_lock(&mddev->write_lock);
+	sync_req = mddev->in_sync;
 	mddev->utime = get_seconds();
 	mddev->events ++;
 
@@ -1272,8 +1276,12 @@ repeat:
 	 * do not write anything to disk if using
 	 * nonpersistent superblocks
 	 */
-	if (!mddev->persistent)
+	if (!mddev->persistent) {
+		mddev->sb_dirty = 0;
+		spin_unlock(&mddev->write_lock);
 		return;
+	}
+	spin_unlock(&mddev->write_lock);
 
 	dprintk(KERN_INFO 
 		"md: updating %s RAID superblock on device (in sync %d)\n",
@@ -1304,6 +1312,15 @@ repeat:
 		printk(KERN_ERR \
 			"md: excessive errors occurred during superblock update, exiting\n");
 	}
+	spin_lock(&mddev->write_lock);
+	if (mddev->in_sync != sync_req) {
+		/* have to write it out again */
+		spin_unlock(&mddev->write_lock);
+		goto repeat;
+	}
+	mddev->sb_dirty = 0;
+	spin_unlock(&mddev->write_lock);
+
 }
 
 /*
@@ -3178,19 +3195,31 @@ void md_done_sync(mddev_t *mddev, int blocks, int ok)
 }
 
 
-void md_write_start(mddev_t *mddev)
+/* md_write_start(mddev, bi)
+ * If we need to update some array metadata (e.g. 'active' flag
+ * in superblock) before writing, queue bi for later writing
+ * and return 0, else return 1 and it will be written now
+ */
+int md_write_start(mddev_t *mddev, struct bio *bi)
 {
-	if (!atomic_read(&mddev->writes_pending)) {
-		mddev_lock_uninterruptible(mddev);
-		if (mddev->in_sync) {
-			mddev->in_sync = 0;
- 			del_timer(&mddev->safemode_timer);
-			md_update_sb(mddev);
-		}
-		atomic_inc(&mddev->writes_pending);
-		mddev_unlock(mddev);
-	} else
-		atomic_inc(&mddev->writes_pending);
+	if (bio_data_dir(bi) != WRITE)
+		return 1;
+
+	atomic_inc(&mddev->writes_pending);
+	spin_lock(&mddev->write_lock);
+	if (mddev->in_sync == 0 && mddev->sb_dirty == 0) {
+		spin_unlock(&mddev->write_lock);
+		return 1;
+	}
+	bio_list_add(&mddev->write_list, bi);
+
+	if (mddev->in_sync) {
+		mddev->in_sync = 0;
+		mddev->sb_dirty = 1;
+	}
+	spin_unlock(&mddev->write_lock);
+	md_wakeup_thread(mddev->thread);
+	return 0;
 }
 
 void md_write_end(mddev_t *mddev)
@@ -3472,6 +3501,7 @@ void md_check_recovery(mddev_t *mddev)
 		mddev->sb_dirty ||
 		test_bit(MD_RECOVERY_NEEDED, &mddev->recovery) ||
 		test_bit(MD_RECOVERY_DONE, &mddev->recovery) ||
+		mddev->write_list.head ||
 		(mddev->safemode == 1) ||
 		(mddev->safemode == 2 && ! atomic_read(&mddev->writes_pending)
 		 && !mddev->in_sync && mddev->recovery_cp == MaxSector)
@@ -3480,7 +3510,9 @@ void md_check_recovery(mddev_t *mddev)
 
 	if (mddev_trylock(mddev)==0) {
 		int spares =0;
+		struct bio *blist;
 
+		spin_lock(&mddev->write_lock);
 		if (mddev->safemode && !atomic_read(&mddev->writes_pending) &&
 		    !mddev->in_sync && mddev->recovery_cp == MaxSector) {
 			mddev->in_sync = 1;
@@ -3488,9 +3520,22 @@ void md_check_recovery(mddev_t *mddev)
 		}
 		if (mddev->safemode == 1)
 			mddev->safemode = 0;
+		blist = bio_list_get(&mddev->write_list);
+		spin_unlock(&mddev->write_lock);
 
 		if (mddev->sb_dirty)
 			md_update_sb(mddev);
+
+		while (blist) {
+			struct bio *b = blist;
+			blist = blist->bi_next;
+			b->bi_next = NULL;
+			generic_make_request(b);
+			/* we already counted this, so need to un-count */
+			md_write_end(mddev);
+		}
+
+
 		if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) &&
 		    !test_bit(MD_RECOVERY_DONE, &mddev->recovery)) {
 			/* resync/recovery still happening */
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index b34ad56362d..3f1280bbaf3 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -530,6 +530,8 @@ static int make_request(request_queue_t *q, struct bio * bio)
 	 * thread has put up a bar for new requests.
 	 * Continue immediately if no resync is active currently.
 	 */
+	if (md_write_start(mddev, bio)==0)
+		return 0;
 	spin_lock_irq(&conf->resync_lock);
 	wait_event_lock_irq(conf->wait_resume, !conf->barrier, conf->resync_lock, );
 	conf->nr_pending++;
@@ -611,7 +613,7 @@ static int make_request(request_queue_t *q, struct bio * bio)
 	rcu_read_unlock();
 
 	atomic_set(&r1_bio->remaining, 1);
-	md_write_start(mddev);
+
 	for (i = 0; i < disks; i++) {
 		struct bio *mbio;
 		if (!r1_bio->bios[i])
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 9ae21504db8..bfc9f52f0ec 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -700,6 +700,9 @@ static int make_request(request_queue_t *q, struct bio * bio)
 		return 0;
 	}
 
+	if (md_write_start(mddev, bio) == 0)
+		return 0;
+
 	/*
 	 * Register the new request and wait if the reconstruction
 	 * thread has put up a bar for new requests.
@@ -774,7 +777,7 @@ static int make_request(request_queue_t *q, struct bio * bio)
 	rcu_read_unlock();
 
 	atomic_set(&r10_bio->remaining, 1);
-	md_write_start(mddev);
+
 	for (i = 0; i < conf->copies; i++) {
 		struct bio *mbio;
 		int d = r10_bio->devs[i].devnum;
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 63b1c59d36f..677ce49078d 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -1411,6 +1411,9 @@ static int make_request (request_queue_t *q, struct bio * bi)
 	sector_t logical_sector, last_sector;
 	struct stripe_head *sh;
 
+	if (md_write_start(mddev, bi)==0)
+		return 0;
+
 	if (bio_data_dir(bi)==WRITE) {
 		disk_stat_inc(mddev->gendisk, writes);
 		disk_stat_add(mddev->gendisk, write_sectors, bio_sectors(bi));
@@ -1423,8 +1426,7 @@ static int make_request (request_queue_t *q, struct bio * bi)
 	last_sector = bi->bi_sector + (bi->bi_size>>9);
 	bi->bi_next = NULL;
 	bi->bi_phys_segments = 1;	/* over-loaded to count active stripes */
-	if ( bio_data_dir(bi) == WRITE )
-		md_write_start(mddev);
+
 	for (;logical_sector < last_sector; logical_sector += STRIPE_SECTORS) {
 		DEFINE_WAIT(w);
 		
diff --git a/drivers/md/raid6main.c b/drivers/md/raid6main.c
index 9d0e0e42a3b..fede16c4e8f 100644
--- a/drivers/md/raid6main.c
+++ b/drivers/md/raid6main.c
@@ -1570,6 +1570,9 @@ static int make_request (request_queue_t *q, struct bio * bi)
 	sector_t logical_sector, last_sector;
 	struct stripe_head *sh;
 
+	if (md_write_start(mddev, bi)==0)
+		return 0;
+
 	if (bio_data_dir(bi)==WRITE) {
 		disk_stat_inc(mddev->gendisk, writes);
 		disk_stat_add(mddev->gendisk, write_sectors, bio_sectors(bi));
@@ -1583,8 +1586,7 @@ static int make_request (request_queue_t *q, struct bio * bi)
 
 	bi->bi_next = NULL;
 	bi->bi_phys_segments = 1;	/* over-loaded to count active stripes */
-	if ( bio_data_dir(bi) == WRITE )
-		md_write_start(mddev);
+
 	for (;logical_sector < last_sector; logical_sector += STRIPE_SECTORS) {
 		DEFINE_WAIT(w);
 
diff --git a/include/linux/raid/md.h b/include/linux/raid/md.h
index a6a67d102bf..cfde8f497d6 100644
--- a/include/linux/raid/md.h
+++ b/include/linux/raid/md.h
@@ -69,7 +69,7 @@ extern mdk_thread_t * md_register_thread (void (*run) (mddev_t *mddev),
 extern void md_unregister_thread (mdk_thread_t *thread);
 extern void md_wakeup_thread(mdk_thread_t *thread);
 extern void md_check_recovery(mddev_t *mddev);
-extern void md_write_start(mddev_t *mddev);
+extern int md_write_start(mddev_t *mddev, struct bio *bi);
 extern void md_write_end(mddev_t *mddev);
 extern void md_handle_safemode(mddev_t *mddev);
 extern void md_done_sync(mddev_t *mddev, int blocks, int ok);
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index c9a0d4013be..d92db54255a 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -15,6 +15,9 @@
 #ifndef _MD_K_H
 #define _MD_K_H
 
+/* and dm-bio-list.h is not under include/linux because.... ??? */
+#include "../../../drivers/md/dm-bio-list.h"
+
 #define MD_RESERVED       0UL
 #define LINEAR            1UL
 #define RAID0             2UL
@@ -252,6 +255,10 @@ struct mddev_s
 	atomic_t			recovery_active; /* blocks scheduled, but not written */
 	wait_queue_head_t		recovery_wait;
 	sector_t			recovery_cp;
+
+	spinlock_t			write_lock;
+	struct bio_list			write_list;
+
 	unsigned int			safemode;	/* if set, update "clean" superblock
 							 * when no writes pending.
 							 */ 
-- 
cgit v1.2.3-70-g09d2


From 57afd89f98a990747445f01c458ecae64263b2f8 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Tue, 21 Jun 2005 17:17:13 -0700
Subject: [PATCH] md: improve the interface to sync_request

1/ change the return value (which is number-of-sectors synced)
 from 'int' to 'sector_t'.
 The number of sectors is usually easily small enough to fit
 in an int, but if resync needs to abort, it may want to return
 the total number of remaining sectors, which could be large.
 Also errors cannot be returned as negative numbers now, so use
 0 instead
2/ Add a 'skipped' return parameter to allow the array to report
 that it skipped the sectors.  This allows md to take this into account
 in the speed calculations.
 Currently there is no important skipping, but the bitmap-based-resync
 that is coming will use this.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/md/md.c           | 35 +++++++++++++++++++++++------------
 drivers/md/raid1.c        |  8 ++++----
 drivers/md/raid10.c       | 19 ++++++++++++-------
 drivers/md/raid5.c        |  6 +++---
 drivers/md/raid6main.c    |  6 +++---
 include/linux/raid/md_k.h |  2 +-
 6 files changed, 46 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 177d2a7d7ce..fa608a1a5c2 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -3241,12 +3241,13 @@ static void md_do_sync(mddev_t *mddev)
 	mddev_t *mddev2;
 	unsigned int currspeed = 0,
 		 window;
-	sector_t max_sectors,j;
+	sector_t max_sectors,j, io_sectors;
 	unsigned long mark[SYNC_MARKS];
 	sector_t mark_cnt[SYNC_MARKS];
 	int last_mark,m;
 	struct list_head *tmp;
 	sector_t last_check;
+	int skipped = 0;
 
 	/* just incase thread restarts... */
 	if (test_bit(MD_RECOVERY_DONE, &mddev->recovery))
@@ -3312,7 +3313,7 @@ static void md_do_sync(mddev_t *mddev)
 
 	if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery))
 		/* resync follows the size requested by the personality,
-		 * which default to physical size, but can be virtual size
+		 * which defaults to physical size, but can be virtual size
 		 */
 		max_sectors = mddev->resync_max_sectors;
 	else
@@ -3331,9 +3332,10 @@ static void md_do_sync(mddev_t *mddev)
 		j = mddev->recovery_cp;
 	else
 		j = 0;
+	io_sectors = 0;
 	for (m = 0; m < SYNC_MARKS; m++) {
 		mark[m] = jiffies;
-		mark_cnt[m] = j;
+		mark_cnt[m] = io_sectors;
 	}
 	last_mark = 0;
 	mddev->resync_mark = mark[last_mark];
@@ -3358,21 +3360,29 @@ static void md_do_sync(mddev_t *mddev)
 	}
 
 	while (j < max_sectors) {
-		int sectors;
+		sector_t sectors;
 
-		sectors = mddev->pers->sync_request(mddev, j, currspeed < sysctl_speed_limit_min);
-		if (sectors < 0) {
+		skipped = 0;
+		sectors = mddev->pers->sync_request(mddev, j, &skipped,
+					    currspeed < sysctl_speed_limit_min);
+		if (sectors == 0) {
 			set_bit(MD_RECOVERY_ERR, &mddev->recovery);
 			goto out;
 		}
-		atomic_add(sectors, &mddev->recovery_active);
+
+		if (!skipped) { /* actual IO requested */
+			io_sectors += sectors;
+			atomic_add(sectors, &mddev->recovery_active);
+		}
+
 		j += sectors;
 		if (j>1) mddev->curr_resync = j;
 
-		if (last_check + window > j || j == max_sectors)
+
+		if (last_check + window > io_sectors || j == max_sectors)
 			continue;
 
-		last_check = j;
+		last_check = io_sectors;
 
 		if (test_bit(MD_RECOVERY_INTR, &mddev->recovery) ||
 		    test_bit(MD_RECOVERY_ERR, &mddev->recovery))
@@ -3386,7 +3396,7 @@ static void md_do_sync(mddev_t *mddev)
 			mddev->resync_mark = mark[next];
 			mddev->resync_mark_cnt = mark_cnt[next];
 			mark[next] = jiffies;
-			mark_cnt[next] = j - atomic_read(&mddev->recovery_active);
+			mark_cnt[next] = io_sectors - atomic_read(&mddev->recovery_active);
 			last_mark = next;
 		}
 
@@ -3413,7 +3423,8 @@ static void md_do_sync(mddev_t *mddev)
 		mddev->queue->unplug_fn(mddev->queue);
 		cond_resched();
 
-		currspeed = ((unsigned long)(j-mddev->resync_mark_cnt))/2/((jiffies-mddev->resync_mark)/HZ +1) +1;
+		currspeed = ((unsigned long)(io_sectors-mddev->resync_mark_cnt))/2
+			/((jiffies-mddev->resync_mark)/HZ +1) +1;
 
 		if (currspeed > sysctl_speed_limit_min) {
 			if ((currspeed > sysctl_speed_limit_max) ||
@@ -3433,7 +3444,7 @@ static void md_do_sync(mddev_t *mddev)
 	wait_event(mddev->recovery_wait, !atomic_read(&mddev->recovery_active));
 
 	/* tell personality that we are finished */
-	mddev->pers->sync_request(mddev, max_sectors, 1);
+	mddev->pers->sync_request(mddev, max_sectors, &skipped, 1);
 
 	if (!test_bit(MD_RECOVERY_ERR, &mddev->recovery) &&
 	    mddev->curr_resync > 2 &&
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 3f1280bbaf3..3c5c916cb09 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1010,7 +1010,7 @@ static int init_resync(conf_t *conf)
  * that can be installed to exclude normal IO requests.
  */
 
-static int sync_request(mddev_t *mddev, sector_t sector_nr, int go_faster)
+static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster)
 {
 	conf_t *conf = mddev_to_conf(mddev);
 	mirror_info_t *mirror;
@@ -1023,7 +1023,7 @@ static int sync_request(mddev_t *mddev, sector_t sector_nr, int go_faster)
 
 	if (!conf->r1buf_pool)
 		if (init_resync(conf))
-			return -ENOMEM;
+			return 0;
 
 	max_sector = mddev->size << 1;
 	if (sector_nr >= max_sector) {
@@ -1107,8 +1107,8 @@ static int sync_request(mddev_t *mddev, sector_t sector_nr, int go_faster)
 		/* There is nowhere to write, so all non-sync
 		 * drives must be failed - so we are finished
 		 */
-		int rv = max_sector - sector_nr;
-		md_done_sync(mddev, rv, 1);
+		sector_t rv = max_sector - sector_nr;
+		*skipped = 1;
 		put_buf(r1_bio);
 		rdev_dec_pending(conf->mirrors[disk].rdev, mddev);
 		return rv;
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index bfc9f52f0ec..8476515bfdc 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -1321,7 +1321,7 @@ static int init_resync(conf_t *conf)
  *
  */
 
-static int sync_request(mddev_t *mddev, sector_t sector_nr, int go_faster)
+static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster)
 {
 	conf_t *conf = mddev_to_conf(mddev);
 	r10bio_t *r10_bio;
@@ -1335,7 +1335,7 @@ static int sync_request(mddev_t *mddev, sector_t sector_nr, int go_faster)
 
 	if (!conf->r10buf_pool)
 		if (init_resync(conf))
-			return -ENOMEM;
+			return 0;
 
  skipped:
 	max_sector = mddev->size << 1;
@@ -1343,15 +1343,15 @@ static int sync_request(mddev_t *mddev, sector_t sector_nr, int go_faster)
 		max_sector = mddev->resync_max_sectors;
 	if (sector_nr >= max_sector) {
 		close_sync(conf);
+		*skipped = 1;
 		return sectors_skipped;
 	}
 	if (chunks_skipped >= conf->raid_disks) {
 		/* if there has been nothing to do on any drive,
 		 * then there is nothing to do at all..
 		 */
-		sector_t sec = max_sector - sector_nr;
-		md_done_sync(mddev, sec, 1);
-		return sec + sectors_skipped;
+		*skipped = 1;
+		return (max_sector - sector_nr) + sectors_skipped;
 	}
 
 	/* make sure whole request will fit in a chunk - if chunks
@@ -1565,17 +1565,22 @@ static int sync_request(mddev_t *mddev, sector_t sector_nr, int go_faster)
 		}
 	}
 
+	if (sectors_skipped)
+		/* pretend they weren't skipped, it makes
+		 * no important difference in this case
+		 */
+		md_done_sync(mddev, sectors_skipped, 1);
+
 	return sectors_skipped + nr_sectors;
  giveup:
 	/* There is nowhere to write, so all non-sync
 	 * drives must be failed, so try the next chunk...
 	 */
 	{
-	int sec = max_sector - sector_nr;
+	sector_t sec = max_sector - sector_nr;
 	sectors_skipped += sec;
 	chunks_skipped ++;
 	sector_nr = max_sector;
-	md_done_sync(mddev, sec, 1);
 	goto skipped;
 	}
 }
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 677ce49078d..1ce3f5aaa98 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -1477,7 +1477,7 @@ static int make_request (request_queue_t *q, struct bio * bi)
 }
 
 /* FIXME go_faster isn't used */
-static int sync_request (mddev_t *mddev, sector_t sector_nr, int go_faster)
+static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster)
 {
 	raid5_conf_t *conf = (raid5_conf_t *) mddev->private;
 	struct stripe_head *sh;
@@ -1500,8 +1500,8 @@ static int sync_request (mddev_t *mddev, sector_t sector_nr, int go_faster)
 	 * nothing we can do.
 	 */
 	if (mddev->degraded >= 1 && test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) {
-		int rv = (mddev->size << 1) - sector_nr;
-		md_done_sync(mddev, rv, 1);
+		sector_t rv = (mddev->size << 1) - sector_nr;
+		*skipped = 1;
 		return rv;
 	}
 
diff --git a/drivers/md/raid6main.c b/drivers/md/raid6main.c
index fede16c4e8f..d9c385496dc 100644
--- a/drivers/md/raid6main.c
+++ b/drivers/md/raid6main.c
@@ -1636,7 +1636,7 @@ static int make_request (request_queue_t *q, struct bio * bi)
 }
 
 /* FIXME go_faster isn't used */
-static int sync_request (mddev_t *mddev, sector_t sector_nr, int go_faster)
+static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster)
 {
 	raid6_conf_t *conf = (raid6_conf_t *) mddev->private;
 	struct stripe_head *sh;
@@ -1659,8 +1659,8 @@ static int sync_request (mddev_t *mddev, sector_t sector_nr, int go_faster)
 	 * nothing we can do.
 	 */
 	if (mddev->degraded >= 2 && test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) {
-		int rv = (mddev->size << 1) - sector_nr;
-		md_done_sync(mddev, rv, 1);
+		sector_t rv = (mddev->size << 1) - sector_nr;
+		*skipped = 1;
 		return rv;
 	}
 
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index d92db54255a..bce0032decf 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -298,7 +298,7 @@ struct mdk_personality_s
 	int (*hot_add_disk) (mddev_t *mddev, mdk_rdev_t *rdev);
 	int (*hot_remove_disk) (mddev_t *mddev, int number);
 	int (*spare_active) (mddev_t *mddev);
-	int (*sync_request)(mddev_t *mddev, sector_t sector_nr, int go_faster);
+	sector_t (*sync_request)(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster);
 	int (*resize) (mddev_t *mddev, sector_t sectors);
 	int (*reshape) (mddev_t *mddev, int raid_disks);
 	int (*reconfig) (mddev_t *mddev, int layout, int chunk_size);
-- 
cgit v1.2.3-70-g09d2


From 32a7627cf3a35396a8e834faf34e38ae9f3b1309 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Tue, 21 Jun 2005 17:17:14 -0700
Subject: [PATCH] md: optimised resync using Bitmap based intent logging

With this patch, the intent to write to some block in the array can be logged
to a bitmap file.  Each bit represents some number of sectors and is set
before any update happens, and only cleared when all writes relating to all
sectors are complete.

After an unclean shutdown, information in this bitmap can be used to optimise
resync - only sectors which could be out-of-sync need to be updated.

Also if a drive is removed and then added back into an array, the recovery can
make use of the bitmap to optimise reconstruction.  This is not implemented in
this patch.

Currently the bitmap is stored in a file which must (obviously) be stored on a
separate device.

The patch only provided infrastructure.  It does not update any personalities
to bitmap intent logging.

Md arrays can still be used with no bitmap file.  This patch has minimal
impact on such arrays.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/md/Makefile         |    3 +-
 drivers/md/bitmap.c         | 1519 +++++++++++++++++++++++++++++++++++++++++++
 drivers/md/md.c             |  172 ++++-
 include/linux/raid/bitmap.h |  280 ++++++++
 include/linux/raid/md_k.h   |    4 +
 include/linux/raid/md_u.h   |    7 +
 6 files changed, 1970 insertions(+), 15 deletions(-)
 create mode 100644 drivers/md/bitmap.c
 create mode 100644 include/linux/raid/bitmap.h

(limited to 'include')

diff --git a/drivers/md/Makefile b/drivers/md/Makefile
index 90de9c146a5..d3efedf6a6a 100644
--- a/drivers/md/Makefile
+++ b/drivers/md/Makefile
@@ -7,6 +7,7 @@ dm-mod-objs	:= dm.o dm-table.o dm-target.o dm-linear.o dm-stripe.o \
 dm-multipath-objs := dm-hw-handler.o dm-path-selector.o dm-mpath.o
 dm-snapshot-objs := dm-snap.o dm-exception-store.o
 dm-mirror-objs	:= dm-log.o dm-raid1.o
+md-mod-objs     := md.o bitmap.o
 raid6-objs	:= raid6main.o raid6algos.o raid6recov.o raid6tables.o \
 		   raid6int1.o raid6int2.o raid6int4.o \
 		   raid6int8.o raid6int16.o raid6int32.o \
@@ -28,7 +29,7 @@ obj-$(CONFIG_MD_RAID5)		+= raid5.o xor.o
 obj-$(CONFIG_MD_RAID6)		+= raid6.o xor.o
 obj-$(CONFIG_MD_MULTIPATH)	+= multipath.o
 obj-$(CONFIG_MD_FAULTY)		+= faulty.o
-obj-$(CONFIG_BLK_DEV_MD)	+= md.o
+obj-$(CONFIG_BLK_DEV_MD)	+= md-mod.o
 obj-$(CONFIG_BLK_DEV_DM)	+= dm-mod.o
 obj-$(CONFIG_DM_CRYPT)		+= dm-crypt.o
 obj-$(CONFIG_DM_MULTIPATH)	+= dm-multipath.o dm-round-robin.o
diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
new file mode 100644
index 00000000000..34ffc133db0
--- /dev/null
+++ b/drivers/md/bitmap.c
@@ -0,0 +1,1519 @@
+/*
+ * bitmap.c two-level bitmap (C) Peter T. Breuer (ptb@ot.uc3m.es) 2003
+ *
+ * bitmap_create  - sets up the bitmap structure
+ * bitmap_destroy - destroys the bitmap structure
+ *
+ * additions, Copyright (C) 2003-2004, Paul Clements, SteelEye Technology, Inc.:
+ * - added disk storage for bitmap
+ * - changes to allow various bitmap chunk sizes
+ * - added bitmap daemon (to asynchronously clear bitmap bits from disk)
+ */
+
+/*
+ * Still to do:
+ *
+ * flush after percent set rather than just time based. (maybe both).
+ * wait if count gets too high, wake when it drops to half.
+ * allow bitmap to be mirrored with superblock (before or after...)
+ * allow hot-add to re-instate a current device.
+ * allow hot-add of bitmap after quiessing device
+ */
+
+#include <linux/module.h>
+#include <linux/version.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/init.h>
+#include <linux/config.h>
+#include <linux/timer.h>
+#include <linux/sched.h>
+#include <linux/list.h>
+#include <linux/file.h>
+#include <linux/mount.h>
+#include <linux/buffer_head.h>
+#include <linux/raid/md.h>
+#include <linux/raid/bitmap.h>
+
+/* debug macros */
+
+#define DEBUG 0
+
+#if DEBUG
+/* these are for debugging purposes only! */
+
+/* define one and only one of these */
+#define INJECT_FAULTS_1 0 /* cause bitmap_alloc_page to fail always */
+#define INJECT_FAULTS_2 0 /* cause bitmap file to be kicked when first bit set*/
+#define INJECT_FAULTS_3 0 /* treat bitmap file as kicked at init time */
+#define INJECT_FAULTS_4 0 /* undef */
+#define INJECT_FAULTS_5 0 /* undef */
+#define INJECT_FAULTS_6 0
+
+/* if these are defined, the driver will fail! debug only */
+#define INJECT_FATAL_FAULT_1 0 /* fail kmalloc, causing bitmap_create to fail */
+#define INJECT_FATAL_FAULT_2 0 /* undef */
+#define INJECT_FATAL_FAULT_3 0 /* undef */
+#endif
+
+//#define DPRINTK PRINTK /* set this NULL to avoid verbose debug output */
+#define DPRINTK(x...) do { } while(0)
+
+#ifndef PRINTK
+#  if DEBUG > 0
+#    define PRINTK(x...) printk(KERN_DEBUG x)
+#  else
+#    define PRINTK(x...)
+#  endif
+#endif
+
+static inline char * bmname(struct bitmap *bitmap)
+{
+	return bitmap->mddev ? mdname(bitmap->mddev) : "mdX";
+}
+
+
+/*
+ * test if the bitmap is active
+ */
+int bitmap_active(struct bitmap *bitmap)
+{
+	unsigned long flags;
+	int res = 0;
+
+	if (!bitmap)
+		return res;
+	spin_lock_irqsave(&bitmap->lock, flags);
+	res = bitmap->flags & BITMAP_ACTIVE;
+	spin_unlock_irqrestore(&bitmap->lock, flags);
+	return res;
+}
+
+#define WRITE_POOL_SIZE 256
+/* mempool for queueing pending writes on the bitmap file */
+static void *write_pool_alloc(unsigned int gfp_flags, void *data)
+{
+	return kmalloc(sizeof(struct page_list), gfp_flags);
+}
+
+static void write_pool_free(void *ptr, void *data)
+{
+	kfree(ptr);
+}
+
+/*
+ * just a placeholder - calls kmalloc for bitmap pages
+ */
+static unsigned char *bitmap_alloc_page(struct bitmap *bitmap)
+{
+	unsigned char *page;
+
+#if INJECT_FAULTS_1
+	page = NULL;
+#else
+	page = kmalloc(PAGE_SIZE, GFP_NOIO);
+#endif
+	if (!page)
+		printk("%s: bitmap_alloc_page FAILED\n", bmname(bitmap));
+	else
+		printk("%s: bitmap_alloc_page: allocated page at %p\n",
+			bmname(bitmap), page);
+	return page;
+}
+
+/*
+ * for now just a placeholder -- just calls kfree for bitmap pages
+ */
+static void bitmap_free_page(struct bitmap *bitmap, unsigned char *page)
+{
+	PRINTK("%s: bitmap_free_page: free page %p\n", bmname(bitmap), page);
+	kfree(page);
+}
+
+/*
+ * check a page and, if necessary, allocate it (or hijack it if the alloc fails)
+ *
+ * 1) check to see if this page is allocated, if it's not then try to alloc
+ * 2) if the alloc fails, set the page's hijacked flag so we'll use the
+ *    page pointer directly as a counter
+ *
+ * if we find our page, we increment the page's refcount so that it stays
+ * allocated while we're using it
+ */
+static int bitmap_checkpage(struct bitmap *bitmap, unsigned long page, int create)
+{
+	unsigned char *mappage;
+
+	if (page >= bitmap->pages) {
+		printk(KERN_ALERT
+			"%s: invalid bitmap page request: %lu (> %lu)\n",
+			bmname(bitmap), page, bitmap->pages-1);
+		return -EINVAL;
+	}
+
+
+	if (bitmap->bp[page].hijacked) /* it's hijacked, don't try to alloc */
+		return 0;
+
+	if (bitmap->bp[page].map) /* page is already allocated, just return */
+		return 0;
+
+	if (!create)
+		return -ENOENT;
+
+	spin_unlock_irq(&bitmap->lock);
+
+	/* this page has not been allocated yet */
+
+	if ((mappage = bitmap_alloc_page(bitmap)) == NULL) {
+		PRINTK("%s: bitmap map page allocation failed, hijacking\n",
+			bmname(bitmap));
+		/* failed - set the hijacked flag so that we can use the
+		 * pointer as a counter */
+		spin_lock_irq(&bitmap->lock);
+		if (!bitmap->bp[page].map)
+			bitmap->bp[page].hijacked = 1;
+		goto out;
+	}
+
+	/* got a page */
+
+	spin_lock_irq(&bitmap->lock);
+
+	/* recheck the page */
+
+	if (bitmap->bp[page].map || bitmap->bp[page].hijacked) {
+		/* somebody beat us to getting the page */
+		bitmap_free_page(bitmap, mappage);
+		return 0;
+	}
+
+	/* no page was in place and we have one, so install it */
+
+	memset(mappage, 0, PAGE_SIZE);
+	bitmap->bp[page].map = mappage;
+	bitmap->missing_pages--;
+out:
+	return 0;
+}
+
+
+/* if page is completely empty, put it back on the free list, or dealloc it */
+/* if page was hijacked, unmark the flag so it might get alloced next time */
+/* Note: lock should be held when calling this */
+static inline void bitmap_checkfree(struct bitmap *bitmap, unsigned long page)
+{
+	char *ptr;
+
+	if (bitmap->bp[page].count) /* page is still busy */
+		return;
+
+	/* page is no longer in use, it can be released */
+
+	if (bitmap->bp[page].hijacked) { /* page was hijacked, undo this now */
+		bitmap->bp[page].hijacked = 0;
+		bitmap->bp[page].map = NULL;
+		return;
+	}
+
+	/* normal case, free the page */
+
+#if 0
+/* actually ... let's not.  We will probably need the page again exactly when
+ * memory is tight and we are flusing to disk
+ */
+	return;
+#else
+	ptr = bitmap->bp[page].map;
+	bitmap->bp[page].map = NULL;
+	bitmap->missing_pages++;
+	bitmap_free_page(bitmap, ptr);
+	return;
+#endif
+}
+
+
+/*
+ * bitmap file handling - read and write the bitmap file and its superblock
+ */
+
+/* copy the pathname of a file to a buffer */
+char *file_path(struct file *file, char *buf, int count)
+{
+	struct dentry *d;
+	struct vfsmount *v;
+
+	if (!buf)
+		return NULL;
+
+	d = file->f_dentry;
+	v = file->f_vfsmnt;
+
+	buf = d_path(d, v, buf, count);
+
+	return IS_ERR(buf) ? NULL : buf;
+}
+
+/*
+ * basic page I/O operations
+ */
+
+/*
+ * write out a page
+ */
+static int write_page(struct page *page, int wait)
+{
+	int ret = -ENOMEM;
+
+	lock_page(page);
+
+	if (page->mapping == NULL)
+		goto unlock_out;
+	else if (i_size_read(page->mapping->host) < page->index << PAGE_SHIFT) {
+		ret = -ENOENT;
+		goto unlock_out;
+	}
+
+	ret = page->mapping->a_ops->prepare_write(NULL, page, 0, PAGE_SIZE);
+	if (!ret)
+		ret = page->mapping->a_ops->commit_write(NULL, page, 0,
+			PAGE_SIZE);
+	if (ret) {
+unlock_out:
+		unlock_page(page);
+		return ret;
+	}
+
+	set_page_dirty(page); /* force it to be written out */
+	return write_one_page(page, wait);
+}
+
+/* read a page from a file, pinning it into cache, and return bytes_read */
+static struct page *read_page(struct file *file, unsigned long index,
+					unsigned long *bytes_read)
+{
+	struct inode *inode = file->f_mapping->host;
+	struct page *page = NULL;
+	loff_t isize = i_size_read(inode);
+	unsigned long end_index = isize >> PAGE_CACHE_SHIFT;
+
+	PRINTK("read bitmap file (%dB @ %Lu)\n", (int)PAGE_CACHE_SIZE,
+			(unsigned long long)index << PAGE_CACHE_SHIFT);
+
+	page = read_cache_page(inode->i_mapping, index,
+			(filler_t *)inode->i_mapping->a_ops->readpage, file);
+	if (IS_ERR(page))
+		goto out;
+	wait_on_page_locked(page);
+	if (!PageUptodate(page) || PageError(page)) {
+		page_cache_release(page);
+		page = ERR_PTR(-EIO);
+		goto out;
+	}
+
+	if (index > end_index) /* we have read beyond EOF */
+		*bytes_read = 0;
+	else if (index == end_index) /* possible short read */
+		*bytes_read = isize & ~PAGE_CACHE_MASK;
+	else
+		*bytes_read = PAGE_CACHE_SIZE; /* got a full page */
+out:
+	if (IS_ERR(page))
+		printk(KERN_ALERT "md: bitmap read error: (%dB @ %Lu): %ld\n",
+			(int)PAGE_CACHE_SIZE,
+			(unsigned long long)index << PAGE_CACHE_SHIFT,
+			PTR_ERR(page));
+	return page;
+}
+
+/*
+ * bitmap file superblock operations
+ */
+
+/* update the event counter and sync the superblock to disk */
+int bitmap_update_sb(struct bitmap *bitmap)
+{
+	bitmap_super_t *sb;
+	unsigned long flags;
+
+	if (!bitmap || !bitmap->mddev) /* no bitmap for this array */
+		return 0;
+	spin_lock_irqsave(&bitmap->lock, flags);
+	if (!bitmap->sb_page) { /* no superblock */
+		spin_unlock_irqrestore(&bitmap->lock, flags);
+		return 0;
+	}
+	page_cache_get(bitmap->sb_page);
+	spin_unlock_irqrestore(&bitmap->lock, flags);
+	sb = (bitmap_super_t *)kmap(bitmap->sb_page);
+	sb->events = cpu_to_le64(bitmap->mddev->events);
+	if (!bitmap->mddev->degraded)
+		sb->events_cleared = cpu_to_le64(bitmap->mddev->events);
+	kunmap(bitmap->sb_page);
+	write_page(bitmap->sb_page, 0);
+	return 0;
+}
+
+/* print out the bitmap file superblock */
+void bitmap_print_sb(struct bitmap *bitmap)
+{
+	bitmap_super_t *sb;
+
+	if (!bitmap || !bitmap->sb_page)
+		return;
+	sb = (bitmap_super_t *)kmap(bitmap->sb_page);
+	printk(KERN_DEBUG "%s: bitmap file superblock:\n", bmname(bitmap));
+	printk(KERN_DEBUG "       magic: %08x\n", le32_to_cpu(sb->magic));
+	printk(KERN_DEBUG "     version: %d\n", le32_to_cpu(sb->version));
+	printk(KERN_DEBUG "        uuid: %08x.%08x.%08x.%08x\n",
+					*(__u32 *)(sb->uuid+0),
+					*(__u32 *)(sb->uuid+4),
+					*(__u32 *)(sb->uuid+8),
+					*(__u32 *)(sb->uuid+12));
+	printk(KERN_DEBUG "      events: %llu\n",
+			(unsigned long long) le64_to_cpu(sb->events));
+	printk(KERN_DEBUG "events_clred: %llu\n",
+			(unsigned long long) le64_to_cpu(sb->events_cleared));
+	printk(KERN_DEBUG "       state: %08x\n", le32_to_cpu(sb->state));
+	printk(KERN_DEBUG "   chunksize: %d B\n", le32_to_cpu(sb->chunksize));
+	printk(KERN_DEBUG "daemon sleep: %ds\n", le32_to_cpu(sb->daemon_sleep));
+	printk(KERN_DEBUG "   sync size: %llu KB\n", le64_to_cpu(sb->sync_size));
+	kunmap(bitmap->sb_page);
+}
+
+/* read the superblock from the bitmap file and initialize some bitmap fields */
+static int bitmap_read_sb(struct bitmap *bitmap)
+{
+	char *reason = NULL;
+	bitmap_super_t *sb;
+	unsigned long chunksize, daemon_sleep;
+	unsigned long bytes_read;
+	unsigned long long events;
+	int err = -EINVAL;
+
+	/* page 0 is the superblock, read it... */
+	bitmap->sb_page = read_page(bitmap->file, 0, &bytes_read);
+	if (IS_ERR(bitmap->sb_page)) {
+		err = PTR_ERR(bitmap->sb_page);
+		bitmap->sb_page = NULL;
+		return err;
+	}
+
+	sb = (bitmap_super_t *)kmap(bitmap->sb_page);
+
+	if (bytes_read < sizeof(*sb)) { /* short read */
+		printk(KERN_INFO "%s: bitmap file superblock truncated\n",
+			bmname(bitmap));
+		err = -ENOSPC;
+		goto out;
+	}
+
+	chunksize = le32_to_cpu(sb->chunksize);
+	daemon_sleep = le32_to_cpu(sb->daemon_sleep);
+
+	/* verify that the bitmap-specific fields are valid */
+	if (sb->magic != cpu_to_le32(BITMAP_MAGIC))
+		reason = "bad magic";
+	else if (sb->version != cpu_to_le32(BITMAP_MAJOR))
+		reason = "unrecognized superblock version";
+	else if (chunksize < 512 || chunksize > (1024 * 1024 * 4))
+		reason = "bitmap chunksize out of range (512B - 4MB)";
+	else if ((1 << ffz(~chunksize)) != chunksize)
+		reason = "bitmap chunksize not a power of 2";
+	else if (daemon_sleep < 1 || daemon_sleep > 15)
+		reason = "daemon sleep period out of range";
+	if (reason) {
+		printk(KERN_INFO "%s: invalid bitmap file superblock: %s\n",
+			bmname(bitmap), reason);
+		goto out;
+	}
+
+	/* keep the array size field of the bitmap superblock up to date */
+	sb->sync_size = cpu_to_le64(bitmap->mddev->resync_max_sectors);
+
+	if (!bitmap->mddev->persistent)
+		goto success;
+
+	/*
+	 * if we have a persistent array superblock, compare the
+	 * bitmap's UUID and event counter to the mddev's
+	 */
+	if (memcmp(sb->uuid, bitmap->mddev->uuid, 16)) {
+		printk(KERN_INFO "%s: bitmap superblock UUID mismatch\n",
+			bmname(bitmap));
+		goto out;
+	}
+	events = le64_to_cpu(sb->events);
+	if (events < bitmap->mddev->events) {
+		printk(KERN_INFO "%s: bitmap file is out of date (%llu < %llu) "
+			"-- forcing full recovery\n", bmname(bitmap), events,
+			(unsigned long long) bitmap->mddev->events);
+		sb->state |= BITMAP_STALE;
+	}
+success:
+	/* assign fields using values from superblock */
+	bitmap->chunksize = chunksize;
+	bitmap->daemon_sleep = daemon_sleep;
+	bitmap->flags |= sb->state;
+	bitmap->events_cleared = le64_to_cpu(sb->events_cleared);
+	err = 0;
+out:
+	kunmap(bitmap->sb_page);
+	if (err)
+		bitmap_print_sb(bitmap);
+	return err;
+}
+
+enum bitmap_mask_op {
+	MASK_SET,
+	MASK_UNSET
+};
+
+/* record the state of the bitmap in the superblock */
+static void bitmap_mask_state(struct bitmap *bitmap, enum bitmap_state bits,
+				enum bitmap_mask_op op)
+{
+	bitmap_super_t *sb;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bitmap->lock, flags);
+	if (!bitmap || !bitmap->sb_page) { /* can't set the state */
+		spin_unlock_irqrestore(&bitmap->lock, flags);
+		return;
+	}
+	page_cache_get(bitmap->sb_page);
+	spin_unlock_irqrestore(&bitmap->lock, flags);
+	sb = (bitmap_super_t *)kmap(bitmap->sb_page);
+	switch (op) {
+		case MASK_SET: sb->state |= bits;
+				break;
+		case MASK_UNSET: sb->state &= ~bits;
+				break;
+		default: BUG();
+	}
+	kunmap(bitmap->sb_page);
+	page_cache_release(bitmap->sb_page);
+}
+
+/*
+ * general bitmap file operations
+ */
+
+/* calculate the index of the page that contains this bit */
+static inline unsigned long file_page_index(unsigned long chunk)
+{
+	return CHUNK_BIT_OFFSET(chunk) >> PAGE_BIT_SHIFT;
+}
+
+/* calculate the (bit) offset of this bit within a page */
+static inline unsigned long file_page_offset(unsigned long chunk)
+{
+	return CHUNK_BIT_OFFSET(chunk) & (PAGE_BITS - 1);
+}
+
+/*
+ * return a pointer to the page in the filemap that contains the given bit
+ *
+ * this lookup is complicated by the fact that the bitmap sb might be exactly
+ * 1 page (e.g., x86) or less than 1 page -- so the bitmap might start on page
+ * 0 or page 1
+ */
+static inline struct page *filemap_get_page(struct bitmap *bitmap,
+					unsigned long chunk)
+{
+	return bitmap->filemap[file_page_index(chunk) - file_page_index(0)];
+}
+
+
+static void bitmap_file_unmap(struct bitmap *bitmap)
+{
+	struct page **map, *sb_page;
+	unsigned long *attr;
+	int pages;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bitmap->lock, flags);
+	map = bitmap->filemap;
+	bitmap->filemap = NULL;
+	attr = bitmap->filemap_attr;
+	bitmap->filemap_attr = NULL;
+	pages = bitmap->file_pages;
+	bitmap->file_pages = 0;
+	sb_page = bitmap->sb_page;
+	bitmap->sb_page = NULL;
+	spin_unlock_irqrestore(&bitmap->lock, flags);
+
+	while (pages--)
+		if (map[pages]->index != 0) /* 0 is sb_page, release it below */
+			page_cache_release(map[pages]);
+	kfree(map);
+	kfree(attr);
+
+	if (sb_page)
+		page_cache_release(sb_page);
+}
+
+static void bitmap_stop_daemons(struct bitmap *bitmap);
+
+/* dequeue the next item in a page list -- don't call from irq context */
+static struct page_list *dequeue_page(struct bitmap *bitmap,
+					struct list_head *head)
+{
+	struct page_list *item = NULL;
+
+	spin_lock(&bitmap->write_lock);
+	if (list_empty(head))
+		goto out;
+	item = list_entry(head->prev, struct page_list, list);
+	list_del(head->prev);
+out:
+	spin_unlock(&bitmap->write_lock);
+	return item;
+}
+
+static void drain_write_queues(struct bitmap *bitmap)
+{
+	struct list_head *queues[] = { 	&bitmap->complete_pages, NULL };
+	struct list_head *head;
+	struct page_list *item;
+	int i;
+
+	for (i = 0; queues[i]; i++) {
+		head = queues[i];
+		while ((item = dequeue_page(bitmap, head))) {
+			page_cache_release(item->page);
+			mempool_free(item, bitmap->write_pool);
+		}
+	}
+
+	spin_lock(&bitmap->write_lock);
+	bitmap->writes_pending = 0; /* make sure waiters continue */
+	wake_up(&bitmap->write_wait);
+	spin_unlock(&bitmap->write_lock);
+}
+
+static void bitmap_file_put(struct bitmap *bitmap)
+{
+	struct file *file;
+	struct inode *inode;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bitmap->lock, flags);
+	file = bitmap->file;
+	bitmap->file = NULL;
+	spin_unlock_irqrestore(&bitmap->lock, flags);
+
+	bitmap_stop_daemons(bitmap);
+
+	drain_write_queues(bitmap);
+
+	bitmap_file_unmap(bitmap);
+
+	if (file) {
+		inode = file->f_mapping->host;
+		spin_lock(&inode->i_lock);
+		atomic_set(&inode->i_writecount, 1); /* allow writes again */
+		spin_unlock(&inode->i_lock);
+		fput(file);
+	}
+}
+
+
+/*
+ * bitmap_file_kick - if an error occurs while manipulating the bitmap file
+ * then it is no longer reliable, so we stop using it and we mark the file
+ * as failed in the superblock
+ */
+static void bitmap_file_kick(struct bitmap *bitmap)
+{
+	char *path, *ptr = NULL;
+
+	bitmap_mask_state(bitmap, BITMAP_STALE, MASK_SET);
+	bitmap_update_sb(bitmap);
+
+	path = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	if (path)
+		ptr = file_path(bitmap->file, path, PAGE_SIZE);
+
+	printk(KERN_ALERT "%s: kicking failed bitmap file %s from array!\n",
+		bmname(bitmap), ptr ? ptr : "");
+
+	kfree(path);
+
+	bitmap_file_put(bitmap);
+
+	return;
+}
+
+enum bitmap_page_attr {
+	BITMAP_PAGE_DIRTY = 1, // there are set bits that need to be synced
+	BITMAP_PAGE_CLEAN = 2, // there are bits that might need to be cleared
+	BITMAP_PAGE_NEEDWRITE=4, // there are cleared bits that need to be synced
+};
+
+static inline void set_page_attr(struct bitmap *bitmap, struct page *page,
+				enum bitmap_page_attr attr)
+{
+	bitmap->filemap_attr[page->index] |= attr;
+}
+
+static inline void clear_page_attr(struct bitmap *bitmap, struct page *page,
+				enum bitmap_page_attr attr)
+{
+	bitmap->filemap_attr[page->index] &= ~attr;
+}
+
+static inline unsigned long get_page_attr(struct bitmap *bitmap, struct page *page)
+{
+	return bitmap->filemap_attr[page->index];
+}
+
+/*
+ * bitmap_file_set_bit -- called before performing a write to the md device
+ * to set (and eventually sync) a particular bit in the bitmap file
+ *
+ * we set the bit immediately, then we record the page number so that
+ * when an unplug occurs, we can flush the dirty pages out to disk
+ */
+static void bitmap_file_set_bit(struct bitmap *bitmap, sector_t block)
+{
+	unsigned long bit;
+	struct page *page;
+	void *kaddr;
+	unsigned long chunk = block >> CHUNK_BLOCK_SHIFT(bitmap);
+
+	if (!bitmap->file || !bitmap->filemap) {
+		return;
+	}
+
+	page = filemap_get_page(bitmap, chunk);
+	bit = file_page_offset(chunk);
+
+
+	/* make sure the page stays cached until it gets written out */
+	if (! (get_page_attr(bitmap, page) & BITMAP_PAGE_DIRTY))
+		page_cache_get(page);
+
+ 	/* set the bit */
+	kaddr = kmap_atomic(page, KM_USER0);
+	set_bit(bit, kaddr);
+	kunmap_atomic(kaddr, KM_USER0);
+	PRINTK("set file bit %lu page %lu\n", bit, page->index);
+
+	/* record page number so it gets flushed to disk when unplug occurs */
+	set_page_attr(bitmap, page, BITMAP_PAGE_DIRTY);
+
+}
+
+/* this gets called when the md device is ready to unplug its underlying
+ * (slave) device queues -- before we let any writes go down, we need to
+ * sync the dirty pages of the bitmap file to disk */
+int bitmap_unplug(struct bitmap *bitmap)
+{
+	unsigned long i, attr, flags;
+	struct page *page;
+	int wait = 0;
+
+	if (!bitmap)
+		return 0;
+
+	/* look at each page to see if there are any set bits that need to be
+	 * flushed out to disk */
+	for (i = 0; i < bitmap->file_pages; i++) {
+		spin_lock_irqsave(&bitmap->lock, flags);
+		if (!bitmap->file || !bitmap->filemap) {
+			spin_unlock_irqrestore(&bitmap->lock, flags);
+			return 0;
+		}
+		page = bitmap->filemap[i];
+		attr = get_page_attr(bitmap, page);
+		clear_page_attr(bitmap, page, BITMAP_PAGE_DIRTY);
+		clear_page_attr(bitmap, page, BITMAP_PAGE_NEEDWRITE);
+		if ((attr & BITMAP_PAGE_DIRTY))
+			wait = 1;
+		spin_unlock_irqrestore(&bitmap->lock, flags);
+
+		if (attr & (BITMAP_PAGE_DIRTY | BITMAP_PAGE_NEEDWRITE))
+			write_page(page, 0);
+	}
+	if (wait) { /* if any writes were performed, we need to wait on them */
+		spin_lock_irq(&bitmap->write_lock);
+		wait_event_lock_irq(bitmap->write_wait,
+			bitmap->writes_pending == 0, bitmap->write_lock,
+			wake_up_process(bitmap->writeback_daemon->tsk));
+		spin_unlock_irq(&bitmap->write_lock);
+	}
+	return 0;
+}
+
+static void bitmap_set_memory_bits(struct bitmap *bitmap, sector_t offset,
+	unsigned long sectors, int set);
+/* * bitmap_init_from_disk -- called at bitmap_create time to initialize
+ * the in-memory bitmap from the on-disk bitmap -- also, sets up the
+ * memory mapping of the bitmap file
+ * Special cases:
+ *   if there's no bitmap file, or if the bitmap file had been
+ *   previously kicked from the array, we mark all the bits as
+ *   1's in order to cause a full resync.
+ */
+static int bitmap_init_from_disk(struct bitmap *bitmap)
+{
+	unsigned long i, chunks, index, oldindex, bit;
+	struct page *page = NULL, *oldpage = NULL;
+	unsigned long num_pages, bit_cnt = 0;
+	struct file *file;
+	unsigned long bytes, offset, dummy;
+	int outofdate;
+	int ret = -ENOSPC;
+
+	chunks = bitmap->chunks;
+	file = bitmap->file;
+
+	if (!file) { /* no file, dirty all the in-memory bits */
+		printk(KERN_INFO "%s: no bitmap file, doing full recovery\n",
+			bmname(bitmap));
+		bitmap_set_memory_bits(bitmap, 0,
+				       chunks << CHUNK_BLOCK_SHIFT(bitmap), 1);
+		return 0;
+	}
+
+#if INJECT_FAULTS_3
+	outofdate = 1;
+#else
+	outofdate = bitmap->flags & BITMAP_STALE;
+#endif
+	if (outofdate)
+		printk(KERN_INFO "%s: bitmap file is out of date, doing full "
+			"recovery\n", bmname(bitmap));
+
+	bytes = (chunks + 7) / 8;
+	num_pages = (bytes + PAGE_SIZE - 1) / PAGE_SIZE;
+	if (i_size_read(file->f_mapping->host) < bytes + sizeof(bitmap_super_t)) {
+		printk(KERN_INFO "%s: bitmap file too short %lu < %lu\n",
+			bmname(bitmap),
+			(unsigned long) i_size_read(file->f_mapping->host),
+			bytes + sizeof(bitmap_super_t));
+		goto out;
+	}
+	num_pages++;
+	bitmap->filemap = kmalloc(sizeof(struct page *) * num_pages, GFP_KERNEL);
+	if (!bitmap->filemap) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	bitmap->filemap_attr = kmalloc(sizeof(long) * num_pages, GFP_KERNEL);
+	if (!bitmap->filemap_attr) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	memset(bitmap->filemap_attr, 0, sizeof(long) * num_pages);
+
+	oldindex = ~0L;
+
+	for (i = 0; i < chunks; i++) {
+		index = file_page_index(i);
+		bit = file_page_offset(i);
+		if (index != oldindex) { /* this is a new page, read it in */
+			/* unmap the old page, we're done with it */
+			if (oldpage != NULL)
+				kunmap(oldpage);
+			if (index == 0) {
+				/*
+				 * if we're here then the superblock page
+				 * contains some bits (PAGE_SIZE != sizeof sb)
+				 * we've already read it in, so just use it
+				 */
+				page = bitmap->sb_page;
+				offset = sizeof(bitmap_super_t);
+			} else {
+				page = read_page(file, index, &dummy);
+				if (IS_ERR(page)) { /* read error */
+					ret = PTR_ERR(page);
+					goto out;
+				}
+				offset = 0;
+			}
+			oldindex = index;
+			oldpage = page;
+			kmap(page);
+
+			if (outofdate) {
+				/*
+				 * if bitmap is out of date, dirty the
+			 	 * whole page and write it out
+				 */
+				memset(page_address(page) + offset, 0xff,
+					PAGE_SIZE - offset);
+				ret = write_page(page, 1);
+				if (ret) {
+					kunmap(page);
+					/* release, page not in filemap yet */
+					page_cache_release(page);
+					goto out;
+				}
+			}
+
+			bitmap->filemap[bitmap->file_pages++] = page;
+		}
+		if (test_bit(bit, page_address(page))) {
+			/* if the disk bit is set, set the memory bit */
+			bitmap_set_memory_bits(bitmap,
+					i << CHUNK_BLOCK_SHIFT(bitmap), 1, 1);
+			bit_cnt++;
+		}
+#if 0
+		else
+			bitmap_set_memory_bits(bitmap,
+				       i << CHUNK_BLOCK_SHIFT(bitmap), 1, 0);
+#endif
+	}
+
+ 	/* everything went OK */
+	ret = 0;
+	bitmap_mask_state(bitmap, BITMAP_STALE, MASK_UNSET);
+
+	if (page) /* unmap the last page */
+		kunmap(page);
+
+	if (bit_cnt) { /* Kick recovery if any bits were set */
+		set_bit(MD_RECOVERY_NEEDED, &bitmap->mddev->recovery);
+		md_wakeup_thread(bitmap->mddev->thread);
+	}
+
+out:
+	printk(KERN_INFO "%s: bitmap initialized from disk: "
+		"read %lu/%lu pages, set %lu bits, status: %d\n",
+		bmname(bitmap), bitmap->file_pages, num_pages, bit_cnt, ret);
+
+	return ret;
+}
+
+
+static void bitmap_count_page(struct bitmap *bitmap, sector_t offset, int inc)
+{
+	sector_t chunk = offset >> CHUNK_BLOCK_SHIFT(bitmap);
+	unsigned long page = chunk >> PAGE_COUNTER_SHIFT;
+	bitmap->bp[page].count += inc;
+/*
+	if (page == 0) printk("count page 0, offset %llu: %d gives %d\n",
+			      (unsigned long long)offset, inc, bitmap->bp[page].count);
+*/
+	bitmap_checkfree(bitmap, page);
+}
+static bitmap_counter_t *bitmap_get_counter(struct bitmap *bitmap,
+					    sector_t offset, int *blocks,
+					    int create);
+
+/*
+ * bitmap daemon -- periodically wakes up to clean bits and flush pages
+ *			out to disk
+ */
+
+int bitmap_daemon_work(struct bitmap *bitmap)
+{
+	unsigned long bit, j;
+	unsigned long flags;
+	struct page *page = NULL, *lastpage = NULL;
+	int err = 0;
+	int blocks;
+	int attr;
+
+	if (bitmap == NULL)
+		return 0;
+	if (time_before(jiffies, bitmap->daemon_lastrun + bitmap->daemon_sleep*HZ))
+		return 0;
+	bitmap->daemon_lastrun = jiffies;
+
+	for (j = 0; j < bitmap->chunks; j++) {
+		bitmap_counter_t *bmc;
+		spin_lock_irqsave(&bitmap->lock, flags);
+		if (!bitmap->file || !bitmap->filemap) {
+			/* error or shutdown */
+			spin_unlock_irqrestore(&bitmap->lock, flags);
+			break;
+		}
+
+		page = filemap_get_page(bitmap, j);
+		/* skip this page unless it's marked as needing cleaning */
+		if (!((attr=get_page_attr(bitmap, page)) & BITMAP_PAGE_CLEAN)) {
+			if (attr & BITMAP_PAGE_NEEDWRITE) {
+				page_cache_get(page);
+				clear_page_attr(bitmap, page, BITMAP_PAGE_NEEDWRITE);
+			}
+			spin_unlock_irqrestore(&bitmap->lock, flags);
+			if (attr & BITMAP_PAGE_NEEDWRITE) {
+				if (write_page(page, 0))
+					bitmap_file_kick(bitmap);
+				page_cache_release(page);
+			}
+			continue;
+		}
+
+		bit = file_page_offset(j);
+
+		if (page != lastpage) {
+			/* grab the new page, sync and release the old */
+			page_cache_get(page);
+			if (lastpage != NULL) {
+				if (get_page_attr(bitmap, lastpage) & BITMAP_PAGE_NEEDWRITE) {
+					clear_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
+					spin_unlock_irqrestore(&bitmap->lock, flags);
+					write_page(lastpage, 0);
+				} else {
+					set_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
+					spin_unlock_irqrestore(&bitmap->lock, flags);
+				}
+				kunmap(lastpage);
+				page_cache_release(lastpage);
+				if (err)
+					bitmap_file_kick(bitmap);
+			} else
+				spin_unlock_irqrestore(&bitmap->lock, flags);
+			lastpage = page;
+			kmap(page);
+/*
+			printk("bitmap clean at page %lu\n", j);
+*/
+			spin_lock_irqsave(&bitmap->lock, flags);
+			clear_page_attr(bitmap, page, BITMAP_PAGE_CLEAN);
+		}
+		bmc = bitmap_get_counter(bitmap, j << CHUNK_BLOCK_SHIFT(bitmap),
+					&blocks, 0);
+		if (bmc) {
+/*
+  if (j < 100) printk("bitmap: j=%lu, *bmc = 0x%x\n", j, *bmc);
+*/
+			if (*bmc == 2) {
+				*bmc=1; /* maybe clear the bit next time */
+				set_page_attr(bitmap, page, BITMAP_PAGE_CLEAN);
+			} else if (*bmc == 1) {
+				/* we can clear the bit */
+				*bmc = 0;
+				bitmap_count_page(bitmap, j << CHUNK_BLOCK_SHIFT(bitmap),
+						  -1);
+
+				/* clear the bit */
+				clear_bit(bit, page_address(page));
+			}
+		}
+		spin_unlock_irqrestore(&bitmap->lock, flags);
+	}
+
+	/* now sync the final page */
+	if (lastpage != NULL) {
+		kunmap(lastpage);
+		spin_lock_irqsave(&bitmap->lock, flags);
+		if (get_page_attr(bitmap, lastpage) &BITMAP_PAGE_NEEDWRITE) {
+			clear_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
+			spin_unlock_irqrestore(&bitmap->lock, flags);
+			write_page(lastpage, 0);
+		} else {
+			set_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
+			spin_unlock_irqrestore(&bitmap->lock, flags);
+		}
+
+		page_cache_release(lastpage);
+	}
+
+	return err;
+}
+
+static void daemon_exit(struct bitmap *bitmap, mdk_thread_t **daemon)
+{
+	mdk_thread_t *dmn;
+	unsigned long flags;
+
+	/* if no one is waiting on us, we'll free the md thread struct
+	 * and exit, otherwise we let the waiter clean things up */
+	spin_lock_irqsave(&bitmap->lock, flags);
+	if ((dmn = *daemon)) { /* no one is waiting, cleanup and exit */
+		*daemon = NULL;
+		spin_unlock_irqrestore(&bitmap->lock, flags);
+		kfree(dmn);
+		complete_and_exit(NULL, 0); /* do_exit not exported */
+	}
+	spin_unlock_irqrestore(&bitmap->lock, flags);
+}
+
+static void bitmap_writeback_daemon(mddev_t *mddev)
+{
+	struct bitmap *bitmap = mddev->bitmap;
+	struct page *page;
+	struct page_list *item;
+	int err = 0;
+
+	while (1) {
+		PRINTK("%s: bitmap writeback daemon waiting...\n", bmname(bitmap));
+		down_interruptible(&bitmap->write_done);
+		if (signal_pending(current)) {
+			printk(KERN_INFO
+			    "%s: bitmap writeback daemon got signal, exiting...\n",
+			    bmname(bitmap));
+			break;
+		}
+
+		PRINTK("%s: bitmap writeback daemon woke up...\n", bmname(bitmap));
+		/* wait on bitmap page writebacks */
+		while ((item = dequeue_page(bitmap, &bitmap->complete_pages))) {
+			page = item->page;
+			mempool_free(item, bitmap->write_pool);
+			PRINTK("wait on page writeback: %p %lu\n", page, bitmap->writes_pending);
+			wait_on_page_writeback(page);
+			PRINTK("finished page writeback: %p %lu\n", page, bitmap->writes_pending);
+			spin_lock(&bitmap->write_lock);
+			if (!--bitmap->writes_pending)
+				wake_up(&bitmap->write_wait);
+			spin_unlock(&bitmap->write_lock);
+			err = PageError(page);
+			page_cache_release(page);
+			if (err) {
+				printk(KERN_WARNING "%s: bitmap file writeback "
+					"failed (page %lu): %d\n",
+					bmname(bitmap), page->index, err);
+				bitmap_file_kick(bitmap);
+				goto out;
+			}
+		}
+	}
+out:
+	if (err) {
+		printk(KERN_INFO "%s: bitmap writeback daemon exiting (%d)\n",
+			bmname(bitmap), err);
+		daemon_exit(bitmap, &bitmap->writeback_daemon);
+	}
+	return;
+}
+
+static int bitmap_start_daemon(struct bitmap *bitmap, mdk_thread_t **ptr,
+				void (*func)(mddev_t *), char *name)
+{
+	mdk_thread_t *daemon;
+	unsigned long flags;
+	char namebuf[32];
+
+	spin_lock_irqsave(&bitmap->lock, flags);
+	*ptr = NULL;
+	if (!bitmap->file) /* no need for daemon if there's no backing file */
+		goto out_unlock;
+
+	spin_unlock_irqrestore(&bitmap->lock, flags);
+
+#if INJECT_FATAL_FAULT_2
+	daemon = NULL;
+#else
+	sprintf(namebuf, "%%s_%s", name);
+	daemon = md_register_thread(func, bitmap->mddev, namebuf);
+#endif
+	if (!daemon) {
+		printk(KERN_ERR "%s: failed to start bitmap daemon\n",
+			bmname(bitmap));
+		return -ECHILD;
+	}
+
+	spin_lock_irqsave(&bitmap->lock, flags);
+	*ptr = daemon;
+
+	md_wakeup_thread(daemon); /* start it running */
+
+	PRINTK("%s: %s daemon (pid %d) started...\n",
+		bmname(bitmap), name, bitmap->daemon->tsk->pid);
+out_unlock:
+	spin_unlock_irqrestore(&bitmap->lock, flags);
+	return 0;
+}
+
+static int bitmap_start_daemons(struct bitmap *bitmap)
+{
+	int err = bitmap_start_daemon(bitmap, &bitmap->writeback_daemon,
+					bitmap_writeback_daemon, "bitmap_wb");
+	return err;
+}
+
+static void bitmap_stop_daemon(struct bitmap *bitmap, mdk_thread_t **ptr)
+{
+	mdk_thread_t *daemon;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bitmap->lock, flags);
+	daemon = *ptr;
+	*ptr = NULL;
+	spin_unlock_irqrestore(&bitmap->lock, flags);
+	if (daemon)
+		md_unregister_thread(daemon); /* destroy the thread */
+}
+
+static void bitmap_stop_daemons(struct bitmap *bitmap)
+{
+	/* the daemons can't stop themselves... they'll just exit instead... */
+	if (bitmap->writeback_daemon &&
+	    current->pid != bitmap->writeback_daemon->tsk->pid)
+		bitmap_stop_daemon(bitmap, &bitmap->writeback_daemon);
+}
+
+static bitmap_counter_t *bitmap_get_counter(struct bitmap *bitmap,
+					    sector_t offset, int *blocks,
+					    int create)
+{
+	/* If 'create', we might release the lock and reclaim it.
+	 * The lock must have been taken with interrupts enabled.
+	 * If !create, we don't release the lock.
+	 */
+	sector_t chunk = offset >> CHUNK_BLOCK_SHIFT(bitmap);
+	unsigned long page = chunk >> PAGE_COUNTER_SHIFT;
+	unsigned long pageoff = (chunk & PAGE_COUNTER_MASK) << COUNTER_BYTE_SHIFT;
+	sector_t csize;
+
+	if (bitmap_checkpage(bitmap, page, create) < 0) {
+		csize = ((sector_t)1) << (CHUNK_BLOCK_SHIFT(bitmap));
+		*blocks = csize - (offset & (csize- 1));
+		return NULL;
+	}
+	/* now locked ... */
+
+	if (bitmap->bp[page].hijacked) { /* hijacked pointer */
+		/* should we use the first or second counter field
+		 * of the hijacked pointer? */
+		int hi = (pageoff > PAGE_COUNTER_MASK);
+		csize = ((sector_t)1) << (CHUNK_BLOCK_SHIFT(bitmap) +
+					  PAGE_COUNTER_SHIFT - 1);
+		*blocks = csize - (offset & (csize- 1));
+		return  &((bitmap_counter_t *)
+			  &bitmap->bp[page].map)[hi];
+	} else { /* page is allocated */
+		csize = ((sector_t)1) << (CHUNK_BLOCK_SHIFT(bitmap));
+		*blocks = csize - (offset & (csize- 1));
+		return (bitmap_counter_t *)
+			&(bitmap->bp[page].map[pageoff]);
+	}
+}
+
+int bitmap_startwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors)
+{
+	if (!bitmap) return 0;
+	while (sectors) {
+		int blocks;
+		bitmap_counter_t *bmc;
+
+		spin_lock_irq(&bitmap->lock);
+		bmc = bitmap_get_counter(bitmap, offset, &blocks, 1);
+		if (!bmc) {
+			spin_unlock_irq(&bitmap->lock);
+			return 0;
+		}
+
+		switch(*bmc) {
+		case 0:
+			bitmap_file_set_bit(bitmap, offset);
+			bitmap_count_page(bitmap,offset, 1);
+			blk_plug_device(bitmap->mddev->queue);
+			/* fall through */
+		case 1:
+			*bmc = 2;
+		}
+		if ((*bmc & COUNTER_MAX) == COUNTER_MAX) BUG();
+		(*bmc)++;
+
+		spin_unlock_irq(&bitmap->lock);
+
+		offset += blocks;
+		if (sectors > blocks)
+			sectors -= blocks;
+		else sectors = 0;
+	}
+	return 0;
+}
+
+void bitmap_endwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors,
+		     int success)
+{
+	if (!bitmap) return;
+	while (sectors) {
+		int blocks;
+		unsigned long flags;
+		bitmap_counter_t *bmc;
+
+		spin_lock_irqsave(&bitmap->lock, flags);
+		bmc = bitmap_get_counter(bitmap, offset, &blocks, 0);
+		if (!bmc) {
+			spin_unlock_irqrestore(&bitmap->lock, flags);
+			return;
+		}
+
+		if (!success && ! (*bmc & NEEDED_MASK))
+			*bmc |= NEEDED_MASK;
+
+		(*bmc)--;
+		if (*bmc <= 2) {
+			set_page_attr(bitmap,
+				      filemap_get_page(bitmap, offset >> CHUNK_BLOCK_SHIFT(bitmap)),
+				      BITMAP_PAGE_CLEAN);
+		}
+		spin_unlock_irqrestore(&bitmap->lock, flags);
+		offset += blocks;
+		if (sectors > blocks)
+			sectors -= blocks;
+		else sectors = 0;
+	}
+}
+
+int bitmap_start_sync(struct bitmap *bitmap, sector_t offset, int *blocks)
+{
+	bitmap_counter_t *bmc;
+	int rv;
+	if (bitmap == NULL) {/* FIXME or bitmap set as 'failed' */
+		*blocks = 1024;
+		return 1; /* always resync if no bitmap */
+	}
+	spin_lock_irq(&bitmap->lock);
+	bmc = bitmap_get_counter(bitmap, offset, blocks, 0);
+	rv = 0;
+	if (bmc) {
+		/* locked */
+		if (RESYNC(*bmc))
+			rv = 1;
+		else if (NEEDED(*bmc)) {
+			rv = 1;
+			*bmc |= RESYNC_MASK;
+			*bmc &= ~NEEDED_MASK;
+		}
+	}
+	spin_unlock_irq(&bitmap->lock);
+	return rv;
+}
+
+void bitmap_end_sync(struct bitmap *bitmap, sector_t offset, int *blocks, int aborted)
+{
+	bitmap_counter_t *bmc;
+	unsigned long flags;
+/*
+	if (offset == 0) printk("bitmap_end_sync 0 (%d)\n", aborted);
+*/	if (bitmap == NULL) {
+		*blocks = 1024;
+		return;
+	}
+	spin_lock_irqsave(&bitmap->lock, flags);
+	bmc = bitmap_get_counter(bitmap, offset, blocks, 0);
+	if (bmc == NULL)
+		goto unlock;
+	/* locked */
+/*
+	if (offset == 0) printk("bitmap_end sync found 0x%x, blocks %d\n", *bmc, *blocks);
+*/
+	if (RESYNC(*bmc)) {
+		*bmc &= ~RESYNC_MASK;
+
+		if (!NEEDED(*bmc) && aborted)
+			*bmc |= NEEDED_MASK;
+		else {
+			if (*bmc <= 2) {
+				set_page_attr(bitmap,
+					      filemap_get_page(bitmap, offset >> CHUNK_BLOCK_SHIFT(bitmap)),
+					      BITMAP_PAGE_CLEAN);
+			}
+		}
+	}
+ unlock:
+	spin_unlock_irqrestore(&bitmap->lock, flags);
+}
+
+void bitmap_close_sync(struct bitmap *bitmap)
+{
+	/* Sync has finished, and any bitmap chunks that weren't synced
+	 * properly have been aborted.  It remains to us to clear the
+	 * RESYNC bit wherever it is still on
+	 */
+	sector_t sector = 0;
+	int blocks;
+	if (!bitmap) return;
+	while (sector < bitmap->mddev->resync_max_sectors) {
+		bitmap_end_sync(bitmap, sector, &blocks, 0);
+/*
+		if (sector < 500) printk("bitmap_close_sync: sec %llu blks %d\n",
+					 (unsigned long long)sector, blocks);
+*/		sector += blocks;
+	}
+}
+
+static void bitmap_set_memory_bits(struct bitmap *bitmap, sector_t offset,
+				   unsigned long sectors, int set)
+{
+	/* For each chunk covered by any of these sectors, set the
+	 * resync needed bit, and the counter to 1.  They should all
+	 * be 0 at this point
+	 */
+	while (sectors) {
+		int secs;
+		bitmap_counter_t *bmc;
+		spin_lock_irq(&bitmap->lock);
+		bmc = bitmap_get_counter(bitmap, offset, &secs, 1);
+		if (!bmc) {
+			spin_unlock_irq(&bitmap->lock);
+			return;
+		}
+		if (set && !NEEDED(*bmc)) {
+			BUG_ON(*bmc);
+			*bmc = NEEDED_MASK | 1;
+			bitmap_count_page(bitmap, offset, 1);
+		}
+		spin_unlock_irq(&bitmap->lock);
+		if (sectors > secs)
+			sectors -= secs;
+		else
+			sectors = 0;
+	}
+}
+
+/* dirty the entire bitmap */
+int bitmap_setallbits(struct bitmap *bitmap)
+{
+	unsigned long flags;
+	unsigned long j;
+
+	/* dirty the in-memory bitmap */
+	bitmap_set_memory_bits(bitmap, 0, bitmap->chunks << CHUNK_BLOCK_SHIFT(bitmap), 1);
+
+	/* dirty the bitmap file */
+	for (j = 0; j < bitmap->file_pages; j++) {
+		struct page *page = bitmap->filemap[j];
+
+		spin_lock_irqsave(&bitmap->lock, flags);
+		page_cache_get(page);
+		spin_unlock_irqrestore(&bitmap->lock, flags);
+		memset(kmap(page), 0xff, PAGE_SIZE);
+		kunmap(page);
+		write_page(page, 0);
+	}
+
+	return 0;
+}
+
+/*
+ * free memory that was allocated
+ */
+void bitmap_destroy(mddev_t *mddev)
+{
+	unsigned long k, pages;
+	struct bitmap_page *bp;
+	struct bitmap *bitmap = mddev->bitmap;
+
+	if (!bitmap) /* there was no bitmap */
+		return;
+
+	mddev->bitmap = NULL; /* disconnect from the md device */
+
+	/* release the bitmap file and kill the daemon */
+	bitmap_file_put(bitmap);
+
+	bp = bitmap->bp;
+	pages = bitmap->pages;
+
+	/* free all allocated memory */
+
+	mempool_destroy(bitmap->write_pool);
+
+	if (bp) /* deallocate the page memory */
+		for (k = 0; k < pages; k++)
+			if (bp[k].map && !bp[k].hijacked)
+				kfree(bp[k].map);
+	kfree(bp);
+	kfree(bitmap);
+}
+
+/*
+ * initialize the bitmap structure
+ * if this returns an error, bitmap_destroy must be called to do clean up
+ */
+int bitmap_create(mddev_t *mddev)
+{
+	struct bitmap *bitmap;
+	unsigned long blocks = mddev->resync_max_sectors;
+	unsigned long chunks;
+	unsigned long pages;
+	struct file *file = mddev->bitmap_file;
+	int err;
+
+	BUG_ON(sizeof(bitmap_super_t) != 256);
+
+	if (!file) /* bitmap disabled, nothing to do */
+		return 0;
+
+	bitmap = kmalloc(sizeof(*bitmap), GFP_KERNEL);
+	if (!bitmap)
+		return -ENOMEM;
+
+	memset(bitmap, 0, sizeof(*bitmap));
+
+	spin_lock_init(&bitmap->lock);
+	bitmap->mddev = mddev;
+	mddev->bitmap = bitmap;
+
+	spin_lock_init(&bitmap->write_lock);
+	init_MUTEX_LOCKED(&bitmap->write_done);
+	INIT_LIST_HEAD(&bitmap->complete_pages);
+	init_waitqueue_head(&bitmap->write_wait);
+	bitmap->write_pool = mempool_create(WRITE_POOL_SIZE, write_pool_alloc,
+				write_pool_free, NULL);
+	if (!bitmap->write_pool)
+		return -ENOMEM;
+
+	bitmap->file = file;
+	get_file(file);
+	/* read superblock from bitmap file (this sets bitmap->chunksize) */
+	err = bitmap_read_sb(bitmap);
+	if (err)
+		return err;
+
+	bitmap->chunkshift = find_first_bit(&bitmap->chunksize,
+					sizeof(bitmap->chunksize));
+
+	/* now that chunksize and chunkshift are set, we can use these macros */
+ 	chunks = (blocks + CHUNK_BLOCK_RATIO(bitmap) - 1) /
+			CHUNK_BLOCK_RATIO(bitmap);
+ 	pages = (chunks + PAGE_COUNTER_RATIO - 1) / PAGE_COUNTER_RATIO;
+
+	BUG_ON(!pages);
+
+	bitmap->chunks = chunks;
+	bitmap->pages = pages;
+	bitmap->missing_pages = pages;
+	bitmap->counter_bits = COUNTER_BITS;
+
+	bitmap->syncchunk = ~0UL;
+
+#if INJECT_FATAL_FAULT_1
+	bitmap->bp = NULL;
+#else
+	bitmap->bp = kmalloc(pages * sizeof(*bitmap->bp), GFP_KERNEL);
+#endif
+	if (!bitmap->bp)
+		return -ENOMEM;
+	memset(bitmap->bp, 0, pages * sizeof(*bitmap->bp));
+
+	bitmap->flags |= BITMAP_ACTIVE;
+
+	/* now that we have some pages available, initialize the in-memory
+	 * bitmap from the on-disk bitmap */
+	err = bitmap_init_from_disk(bitmap);
+	if (err)
+		return err;
+
+	printk(KERN_INFO "created bitmap (%lu pages) for device %s\n",
+		pages, bmname(bitmap));
+
+	/* kick off the bitmap daemons */
+	err = bitmap_start_daemons(bitmap);
+	if (err)
+		return err;
+	return bitmap_update_sb(bitmap);
+}
+
+/* the bitmap API -- for raid personalities */
+EXPORT_SYMBOL(bitmap_startwrite);
+EXPORT_SYMBOL(bitmap_endwrite);
+EXPORT_SYMBOL(bitmap_start_sync);
+EXPORT_SYMBOL(bitmap_end_sync);
+EXPORT_SYMBOL(bitmap_unplug);
+EXPORT_SYMBOL(bitmap_close_sync);
+EXPORT_SYMBOL(bitmap_daemon_work);
diff --git a/drivers/md/md.c b/drivers/md/md.c
index fa608a1a5c2..c402f6cc704 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -19,6 +19,9 @@
 
      Neil Brown <neilb@cse.unsw.edu.au>.
 
+   - persistent bitmap code
+     Copyright (C) 2003-2004, Paul Clements, SteelEye Technology, Inc.
+
    This program is free software; you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation; either version 2, or (at your option)
@@ -33,6 +36,7 @@
 #include <linux/config.h>
 #include <linux/linkage.h>
 #include <linux/raid/md.h>
+#include <linux/raid/bitmap.h>
 #include <linux/sysctl.h>
 #include <linux/devfs_fs_kernel.h>
 #include <linux/buffer_head.h> /* for invalidate_bdev */
@@ -40,6 +44,8 @@
 
 #include <linux/init.h>
 
+#include <linux/file.h>
+
 #ifdef CONFIG_KMOD
 #include <linux/kmod.h>
 #endif
@@ -1198,8 +1204,11 @@ void md_print_devices(void)
 	printk("md:	* <COMPLETE RAID STATE PRINTOUT> *\n");
 	printk("md:	**********************************\n");
 	ITERATE_MDDEV(mddev,tmp) {
-		printk("%s: ", mdname(mddev));
 
+		if (mddev->bitmap)
+			bitmap_print_sb(mddev->bitmap);
+		else
+			printk("%s: ", mdname(mddev));
 		ITERATE_RDEV(mddev,rdev,tmp2)
 			printk("<%s>", bdevname(rdev->bdev,b));
 		printk("\n");
@@ -1287,7 +1296,7 @@ repeat:
 		"md: updating %s RAID superblock on device (in sync %d)\n",
 		mdname(mddev),mddev->in_sync);
 
-	err = 0;
+	err = bitmap_update_sb(mddev->bitmap);
 	ITERATE_RDEV(mddev,rdev,tmp) {
 		char b[BDEVNAME_SIZE];
 		dprintk(KERN_INFO "md: ");
@@ -1624,12 +1633,19 @@ static int do_md_run(mddev_t * mddev)
 
 	mddev->resync_max_sectors = mddev->size << 1; /* may be over-ridden by personality */
 
-	err = mddev->pers->run(mddev);
+	/* before we start the array running, initialise the bitmap */
+	err = bitmap_create(mddev);
+	if (err)
+		printk(KERN_ERR "%s: failed to create bitmap (%d)\n",
+			mdname(mddev), err);
+	else
+		err = mddev->pers->run(mddev);
 	if (err) {
 		printk(KERN_ERR "md: pers->run() failed ...\n");
 		module_put(mddev->pers->owner);
 		mddev->pers = NULL;
-		return -EINVAL;
+		bitmap_destroy(mddev);
+		return err;
 	}
  	atomic_set(&mddev->writes_pending,0);
 	mddev->safemode = 0;
@@ -1742,6 +1758,14 @@ static int do_md_stop(mddev_t * mddev, int ro)
 		if (ro)
 			set_disk_ro(disk, 1);
 	}
+
+	bitmap_destroy(mddev);
+	if (mddev->bitmap_file) {
+		atomic_set(&mddev->bitmap_file->f_dentry->d_inode->i_writecount, 1);
+		fput(mddev->bitmap_file);
+		mddev->bitmap_file = NULL;
+	}
+
 	/*
 	 * Free resources if final stop
 	 */
@@ -2000,6 +2024,42 @@ static int get_array_info(mddev_t * mddev, void __user * arg)
 	return 0;
 }
 
+static int get_bitmap_file(mddev_t * mddev, void * arg)
+{
+	mdu_bitmap_file_t *file = NULL; /* too big for stack allocation */
+	char *ptr, *buf = NULL;
+	int err = -ENOMEM;
+
+	file = kmalloc(sizeof(*file), GFP_KERNEL);
+	if (!file)
+		goto out;
+
+	/* bitmap disabled, zero the first byte and copy out */
+	if (!mddev->bitmap || !mddev->bitmap->file) {
+		file->pathname[0] = '\0';
+		goto copy_out;
+	}
+
+	buf = kmalloc(sizeof(file->pathname), GFP_KERNEL);
+	if (!buf)
+		goto out;
+
+	ptr = file_path(mddev->bitmap->file, buf, sizeof(file->pathname));
+	if (!ptr)
+		goto out;
+
+	strcpy(file->pathname, ptr);
+
+copy_out:
+	err = 0;
+	if (copy_to_user(arg, file, sizeof(*file)))
+		err = -EFAULT;
+out:
+	kfree(buf);
+	kfree(file);
+	return err;
+}
+
 static int get_disk_info(mddev_t * mddev, void __user * arg)
 {
 	mdu_disk_info_t info;
@@ -2275,6 +2335,48 @@ abort_export:
 	return err;
 }
 
+/* similar to deny_write_access, but accounts for our holding a reference
+ * to the file ourselves */
+static int deny_bitmap_write_access(struct file * file)
+{
+	struct inode *inode = file->f_mapping->host;
+
+	spin_lock(&inode->i_lock);
+	if (atomic_read(&inode->i_writecount) > 1) {
+		spin_unlock(&inode->i_lock);
+		return -ETXTBSY;
+	}
+	atomic_set(&inode->i_writecount, -1);
+	spin_unlock(&inode->i_lock);
+
+	return 0;
+}
+
+static int set_bitmap_file(mddev_t *mddev, int fd)
+{
+	int err;
+
+	if (mddev->pers)
+		return -EBUSY;
+
+	mddev->bitmap_file = fget(fd);
+
+	if (mddev->bitmap_file == NULL) {
+		printk(KERN_ERR "%s: error: failed to get bitmap file\n",
+			mdname(mddev));
+		return -EBADF;
+	}
+
+	err = deny_bitmap_write_access(mddev->bitmap_file);
+	if (err) {
+		printk(KERN_ERR "%s: error: bitmap file is already in use\n",
+			mdname(mddev));
+		fput(mddev->bitmap_file);
+		mddev->bitmap_file = NULL;
+	}
+	return err;
+}
+
 /*
  * set_array_info is used two different ways
  * The original usage is when creating a new array.
@@ -2586,8 +2688,10 @@ static int md_ioctl(struct inode *inode, struct file *file,
 	/*
 	 * Commands querying/configuring an existing array:
 	 */
-	/* if we are initialised yet, only ADD_NEW_DISK or STOP_ARRAY is allowed */
-	if (!mddev->raid_disks && cmd != ADD_NEW_DISK && cmd != STOP_ARRAY && cmd != RUN_ARRAY) {
+	/* if we are not initialised yet, only ADD_NEW_DISK, STOP_ARRAY,
+	 * RUN_ARRAY, and SET_BITMAP_FILE are allowed */
+	if (!mddev->raid_disks && cmd != ADD_NEW_DISK && cmd != STOP_ARRAY
+			&& cmd != RUN_ARRAY && cmd != SET_BITMAP_FILE) {
 		err = -ENODEV;
 		goto abort_unlock;
 	}
@@ -2601,6 +2705,10 @@ static int md_ioctl(struct inode *inode, struct file *file,
 			err = get_array_info(mddev, argp);
 			goto done_unlock;
 
+		case GET_BITMAP_FILE:
+			err = get_bitmap_file(mddev, (void *)arg);
+			goto done_unlock;
+
 		case GET_DISK_INFO:
 			err = get_disk_info(mddev, argp);
 			goto done_unlock;
@@ -2681,6 +2789,10 @@ static int md_ioctl(struct inode *inode, struct file *file,
 			err = do_md_run (mddev);
 			goto done_unlock;
 
+		case SET_BITMAP_FILE:
+			err = set_bitmap_file(mddev, (int)arg);
+			goto done_unlock;
+
 		default:
 			if (_IOC_TYPE(cmd) == MD_MAJOR)
 				printk(KERN_WARNING "md: %s(pid %d) used"
@@ -2792,8 +2904,9 @@ static int md_thread(void * arg)
 	while (thread->run) {
 		void (*run)(mddev_t *);
 
-		wait_event_interruptible(thread->wqueue,
-					 test_bit(THREAD_WAKEUP, &thread->flags));
+		wait_event_interruptible_timeout(thread->wqueue,
+						 test_bit(THREAD_WAKEUP, &thread->flags),
+						 thread->timeout);
 		if (current->flags & PF_FREEZE)
 			refrigerator(PF_FREEZE);
 
@@ -2839,6 +2952,7 @@ mdk_thread_t *md_register_thread(void (*run) (mddev_t *), mddev_t *mddev,
 	thread->run = run;
 	thread->mddev = mddev;
 	thread->name = name;
+	thread->timeout = MAX_SCHEDULE_TIMEOUT;
 	ret = kernel_thread(md_thread, thread, 0);
 	if (ret < 0) {
 		kfree(thread);
@@ -2877,13 +2991,13 @@ void md_error(mddev_t *mddev, mdk_rdev_t *rdev)
 
 	if (!rdev || rdev->faulty)
 		return;
-
+/*
 	dprintk("md_error dev:%s, rdev:(%d:%d), (caller: %p,%p,%p,%p).\n",
 		mdname(mddev),
 		MAJOR(rdev->bdev->bd_dev), MINOR(rdev->bdev->bd_dev),
 		__builtin_return_address(0),__builtin_return_address(1),
 		__builtin_return_address(2),__builtin_return_address(3));
-
+*/
 	if (!mddev->pers->error_handler)
 		return;
 	mddev->pers->error_handler(mddev,rdev);
@@ -3037,6 +3151,7 @@ static int md_seq_show(struct seq_file *seq, void *v)
 	struct list_head *tmp2;
 	mdk_rdev_t *rdev;
 	int i;
+	struct bitmap *bitmap;
 
 	if (v == (void*)1) {
 		seq_printf(seq, "Personalities : ");
@@ -3089,10 +3204,36 @@ static int md_seq_show(struct seq_file *seq, void *v)
 		if (mddev->pers) {
 			mddev->pers->status (seq, mddev);
 	 		seq_printf(seq, "\n      ");
-			if (mddev->curr_resync > 2)
+			if (mddev->curr_resync > 2) {
 				status_resync (seq, mddev);
-			else if (mddev->curr_resync == 1 || mddev->curr_resync == 2)
-				seq_printf(seq, "	resync=DELAYED");
+				seq_printf(seq, "\n      ");
+			} else if (mddev->curr_resync == 1 || mddev->curr_resync == 2)
+				seq_printf(seq, "	resync=DELAYED\n      ");
+		} else
+			seq_printf(seq, "\n       ");
+
+		if ((bitmap = mddev->bitmap)) {
+			char *buf, *path;
+			unsigned long chunk_kb;
+			unsigned long flags;
+			buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
+			spin_lock_irqsave(&bitmap->lock, flags);
+			chunk_kb = bitmap->chunksize >> 10;
+			seq_printf(seq, "bitmap: %lu/%lu pages [%luKB], "
+				"%lu%s chunk",
+				bitmap->pages - bitmap->missing_pages,
+				bitmap->pages,
+				(bitmap->pages - bitmap->missing_pages)
+					<< (PAGE_SHIFT - 10),
+				chunk_kb ? chunk_kb : bitmap->chunksize,
+				chunk_kb ? "KB" : "B");
+			if (bitmap->file && buf) {
+				path = file_path(bitmap->file, buf, PAGE_SIZE);
+				seq_printf(seq, ", file: %s", path ? path : "");
+			}
+			seq_printf(seq, "\n");
+			spin_unlock_irqrestore(&bitmap->lock, flags);
+			kfree(buf);
 		}
 
 		seq_printf(seq, "\n");
@@ -3328,7 +3469,8 @@ static void md_do_sync(mddev_t *mddev)
 	       sysctl_speed_limit_max);
 
 	is_mddev_idle(mddev); /* this also initializes IO event counters */
-	if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery))
+	/* we don't use the checkpoint if there's a bitmap */
+	if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery) && !mddev->bitmap)
 		j = mddev->recovery_cp;
 	else
 		j = 0;
@@ -3673,6 +3815,8 @@ static int __init md_init(void)
 			" MD_SB_DISKS=%d\n",
 			MD_MAJOR_VERSION, MD_MINOR_VERSION,
 			MD_PATCHLEVEL_VERSION, MAX_MD_DEVS, MD_SB_DISKS);
+	printk(KERN_INFO "md: bitmap version %d.%d\n", BITMAP_MAJOR,
+			BITMAP_MINOR);
 
 	if (register_blkdev(MAJOR_NR, "md"))
 		return -1;
diff --git a/include/linux/raid/bitmap.h b/include/linux/raid/bitmap.h
new file mode 100644
index 00000000000..f785cf26cba
--- /dev/null
+++ b/include/linux/raid/bitmap.h
@@ -0,0 +1,280 @@
+/*
+ * bitmap.h: Copyright (C) Peter T. Breuer (ptb@ot.uc3m.es) 2003
+ *
+ * additions: Copyright (C) 2003-2004, Paul Clements, SteelEye Technology, Inc.
+ */
+#ifndef BITMAP_H
+#define BITMAP_H 1
+
+#define BITMAP_MAJOR 3
+#define BITMAP_MINOR 38
+
+/*
+ * in-memory bitmap:
+ *
+ * Use 16 bit block counters to track pending writes to each "chunk".
+ * The 2 high order bits are special-purpose, the first is a flag indicating
+ * whether a resync is needed.  The second is a flag indicating whether a
+ * resync is active.
+ * This means that the counter is actually 14 bits:
+ *
+ * +--------+--------+------------------------------------------------+
+ * | resync | resync |               counter                          |
+ * | needed | active |                                                |
+ * |  (0-1) |  (0-1) |              (0-16383)                         |
+ * +--------+--------+------------------------------------------------+
+ *
+ * The "resync needed" bit is set when:
+ *    a '1' bit is read from storage at startup.
+ *    a write request fails on some drives
+ *    a resync is aborted on a chunk with 'resync active' set
+ * It is cleared (and resync-active set) when a resync starts across all drives
+ * of the chunk.
+ *
+ *
+ * The "resync active" bit is set when:
+ *    a resync is started on all drives, and resync_needed is set.
+ *       resync_needed will be cleared (as long as resync_active wasn't already set).
+ * It is cleared when a resync completes.
+ *
+ * The counter counts pending write requests, plus the on-disk bit.
+ * When the counter is '1' and the resync bits are clear, the on-disk
+ * bit can be cleared aswell, thus setting the counter to 0.
+ * When we set a bit, or in the counter (to start a write), if the fields is
+ * 0, we first set the disk bit and set the counter to 1.
+ *
+ * If the counter is 0, the on-disk bit is clear and the stipe is clean
+ * Anything that dirties the stipe pushes the counter to 2 (at least)
+ * and sets the on-disk bit (lazily).
+ * If a periodic sweep find the counter at 2, it is decremented to 1.
+ * If the sweep find the counter at 1, the on-disk bit is cleared and the
+ * counter goes to zero.
+ *
+ * Also, we'll hijack the "map" pointer itself and use it as two 16 bit block
+ * counters as a fallback when "page" memory cannot be allocated:
+ *
+ * Normal case (page memory allocated):
+ *
+ *     page pointer (32-bit)
+ *
+ *     [ ] ------+
+ *               |
+ *               +-------> [   ][   ]..[   ] (4096 byte page == 2048 counters)
+ *                          c1   c2    c2048
+ *
+ * Hijacked case (page memory allocation failed):
+ *
+ *     hijacked page pointer (32-bit)
+ *
+ *     [		  ][		  ] (no page memory allocated)
+ *      counter #1 (16-bit) counter #2 (16-bit)
+ *
+ */
+
+#ifdef __KERNEL__
+
+#define PAGE_BITS (PAGE_SIZE << 3)
+#define PAGE_BIT_SHIFT (PAGE_SHIFT + 3)
+
+typedef __u16 bitmap_counter_t;
+#define COUNTER_BITS 16
+#define COUNTER_BIT_SHIFT 4
+#define COUNTER_BYTE_RATIO (COUNTER_BITS / 8)
+#define COUNTER_BYTE_SHIFT (COUNTER_BIT_SHIFT - 3)
+
+#define NEEDED_MASK ((bitmap_counter_t) (1 << (COUNTER_BITS - 1)))
+#define RESYNC_MASK ((bitmap_counter_t) (1 << (COUNTER_BITS - 2)))
+#define COUNTER_MAX ((bitmap_counter_t) RESYNC_MASK - 1)
+#define NEEDED(x) (((bitmap_counter_t) x) & NEEDED_MASK)
+#define RESYNC(x) (((bitmap_counter_t) x) & RESYNC_MASK)
+#define COUNTER(x) (((bitmap_counter_t) x) & COUNTER_MAX)
+
+/* how many counters per page? */
+#define PAGE_COUNTER_RATIO (PAGE_BITS / COUNTER_BITS)
+/* same, except a shift value for more efficient bitops */
+#define PAGE_COUNTER_SHIFT (PAGE_BIT_SHIFT - COUNTER_BIT_SHIFT)
+/* same, except a mask value for more efficient bitops */
+#define PAGE_COUNTER_MASK  (PAGE_COUNTER_RATIO - 1)
+
+#define BITMAP_BLOCK_SIZE 512
+#define BITMAP_BLOCK_SHIFT 9
+
+/* how many blocks per chunk? (this is variable) */
+#define CHUNK_BLOCK_RATIO(bitmap) ((bitmap)->chunksize >> BITMAP_BLOCK_SHIFT)
+#define CHUNK_BLOCK_SHIFT(bitmap) ((bitmap)->chunkshift - BITMAP_BLOCK_SHIFT)
+#define CHUNK_BLOCK_MASK(bitmap) (CHUNK_BLOCK_RATIO(bitmap) - 1)
+
+/* when hijacked, the counters and bits represent even larger "chunks" */
+/* there will be 1024 chunks represented by each counter in the page pointers */
+#define PAGEPTR_BLOCK_RATIO(bitmap) \
+			(CHUNK_BLOCK_RATIO(bitmap) << PAGE_COUNTER_SHIFT >> 1)
+#define PAGEPTR_BLOCK_SHIFT(bitmap) \
+			(CHUNK_BLOCK_SHIFT(bitmap) + PAGE_COUNTER_SHIFT - 1)
+#define PAGEPTR_BLOCK_MASK(bitmap) (PAGEPTR_BLOCK_RATIO(bitmap) - 1)
+
+/*
+ * on-disk bitmap:
+ *
+ * Use one bit per "chunk" (block set). We do the disk I/O on the bitmap
+ * file a page at a time. There's a superblock at the start of the file.
+ */
+
+/* map chunks (bits) to file pages - offset by the size of the superblock */
+#define CHUNK_BIT_OFFSET(chunk) ((chunk) + (sizeof(bitmap_super_t) << 3))
+
+#endif
+
+/*
+ * bitmap structures:
+ */
+
+#define BITMAP_MAGIC 0x6d746962
+
+/* use these for bitmap->flags and bitmap->sb->state bit-fields */
+enum bitmap_state {
+	BITMAP_ACTIVE = 0x001, /* the bitmap is in use */
+	BITMAP_STALE  = 0x002  /* the bitmap file is out of date or had -EIO */
+};
+
+/* the superblock at the front of the bitmap file -- little endian */
+typedef struct bitmap_super_s {
+	__u32 magic;        /*  0  BITMAP_MAGIC */
+	__u32 version;      /*  4  the bitmap major for now, could change... */
+	__u8  uuid[16];     /*  8  128 bit uuid - must match md device uuid */
+	__u64 events;       /* 24  event counter for the bitmap (1)*/
+	__u64 events_cleared;/*32  event counter when last bit cleared (2) */
+	__u64 sync_size;    /* 40  the size of the md device's sync range(3) */
+	__u32 state;        /* 48  bitmap state information */
+	__u32 chunksize;    /* 52  the bitmap chunk size in bytes */
+	__u32 daemon_sleep; /* 56  seconds between disk flushes */
+
+	__u8  pad[256 - 60]; /* set to zero */
+} bitmap_super_t;
+
+/* notes:
+ * (1) This event counter is updated before the eventcounter in the md superblock
+ *    When a bitmap is loaded, it is only accepted if this event counter is equal
+ *    to, or one greater than, the event counter in the superblock.
+ * (2) This event counter is updated when the other one is *if*and*only*if* the
+ *    array is not degraded.  As bits are not cleared when the array is degraded,
+ *    this represents the last time that any bits were cleared.
+ *    If a device is being added that has an event count with this value or
+ *    higher, it is accepted as conforming to the bitmap.
+ * (3)This is the number of sectors represented by the bitmap, and is the range that
+ *    resync happens across.  For raid1 and raid5/6 it is the size of individual
+ *    devices.  For raid10 it is the size of the array.
+ */
+
+#ifdef __KERNEL__
+
+/* the in-memory bitmap is represented by bitmap_pages */
+struct bitmap_page {
+	/*
+	 * map points to the actual memory page
+	 */
+	char *map;
+	/*
+	 * in emergencies (when map cannot be alloced), hijack the map
+	 * pointer and use it as two counters itself
+	 */
+	unsigned int hijacked:1;
+	/*
+	 * count of dirty bits on the page
+	 */
+	unsigned int  count:31;
+};
+
+/* keep track of bitmap file pages that have pending writes on them */
+struct page_list {
+	struct list_head list;
+	struct page *page;
+};
+
+/* the main bitmap structure - one per mddev */
+struct bitmap {
+	struct bitmap_page *bp;
+	unsigned long pages; /* total number of pages in the bitmap */
+	unsigned long missing_pages; /* number of pages not yet allocated */
+
+	mddev_t *mddev; /* the md device that the bitmap is for */
+
+	int counter_bits; /* how many bits per block counter */
+
+	/* bitmap chunksize -- how much data does each bit represent? */
+	unsigned long chunksize;
+	unsigned long chunkshift; /* chunksize = 2^chunkshift (for bitops) */
+	unsigned long chunks; /* total number of data chunks for the array */
+
+	/* We hold a count on the chunk currently being synced, and drop
+	 * it when the last block is started.  If the resync is aborted
+	 * midway, we need to be able to drop that count, so we remember
+	 * the counted chunk..
+	 */
+	unsigned long syncchunk;
+
+	__u64	events_cleared;
+
+	/* bitmap spinlock */
+	spinlock_t lock;
+
+	struct file *file; /* backing disk file */
+	struct page *sb_page; /* cached copy of the bitmap file superblock */
+	struct page **filemap; /* list of cache pages for the file */
+	unsigned long *filemap_attr; /* attributes associated w/ filemap pages */
+	unsigned long file_pages; /* number of pages in the file */
+
+	unsigned long flags;
+
+	/*
+	 * the bitmap daemon - periodically wakes up and sweeps the bitmap
+	 * file, cleaning up bits and flushing out pages to disk as necessary
+	 */
+	unsigned long daemon_lastrun; /* jiffies of last run */
+	unsigned long daemon_sleep; /* how many seconds between updates? */
+
+	/*
+	 * bitmap write daemon - this daemon performs writes to the bitmap file
+	 * this thread is only needed because of a limitation in ext3 (jbd)
+	 * that does not allow a task to have two journal transactions ongoing
+	 * simultaneously (even if the transactions are for two different
+	 * filesystems) -- in the case of bitmap, that would be the filesystem
+	 * that the bitmap file resides on and the filesystem that is mounted
+	 * on the md device -- see current->journal_info in jbd/transaction.c
+	 */
+	mdk_thread_t *writeback_daemon;
+	spinlock_t write_lock;
+	struct semaphore write_ready;
+	struct semaphore write_done;
+	unsigned long writes_pending;
+	wait_queue_head_t write_wait;
+	struct list_head write_pages;
+	struct list_head complete_pages;
+	mempool_t *write_pool;
+};
+
+/* the bitmap API */
+
+/* these are used only by md/bitmap */
+int  bitmap_create(mddev_t *mddev);
+void bitmap_destroy(mddev_t *mddev);
+int  bitmap_active(struct bitmap *bitmap);
+
+char *file_path(struct file *file, char *buf, int count);
+void bitmap_print_sb(struct bitmap *bitmap);
+int bitmap_update_sb(struct bitmap *bitmap);
+
+int  bitmap_setallbits(struct bitmap *bitmap);
+
+/* these are exported */
+int bitmap_startwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors);
+void bitmap_endwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors,
+		     int success);
+int bitmap_start_sync(struct bitmap *bitmap, sector_t offset, int *blocks);
+void bitmap_end_sync(struct bitmap *bitmap, sector_t offset, int *blocks, int aborted);
+void bitmap_close_sync(struct bitmap *bitmap);
+
+int bitmap_unplug(struct bitmap *bitmap);
+int bitmap_daemon_work(struct bitmap *bitmap);
+#endif
+
+#endif
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index bce0032decf..16e94a9f0f8 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -267,6 +267,9 @@ struct mddev_s
 	atomic_t			writes_pending; 
 	request_queue_t			*queue;	/* for plugging ... */
 
+	struct bitmap                   *bitmap; /* the bitmap for the device */
+	struct file			*bitmap_file; /* the bitmap file */
+
 	struct list_head		all_mddevs;
 };
 
@@ -341,6 +344,7 @@ typedef struct mdk_thread_s {
 	unsigned long           flags;
 	struct completion	*event;
 	struct task_struct	*tsk;
+	unsigned long		timeout;
 	const char		*name;
 } mdk_thread_t;
 
diff --git a/include/linux/raid/md_u.h b/include/linux/raid/md_u.h
index a2df5c2a42a..81da20ccec4 100644
--- a/include/linux/raid/md_u.h
+++ b/include/linux/raid/md_u.h
@@ -23,6 +23,7 @@
 #define GET_DISK_INFO		_IOR (MD_MAJOR, 0x12, mdu_disk_info_t)
 #define PRINT_RAID_DEBUG	_IO (MD_MAJOR, 0x13)
 #define RAID_AUTORUN		_IO (MD_MAJOR, 0x14)
+#define GET_BITMAP_FILE		_IOR (MD_MAJOR, 0x15, mdu_bitmap_file_t)
 
 /* configuration */
 #define CLEAR_ARRAY		_IO (MD_MAJOR, 0x20)
@@ -36,6 +37,7 @@
 #define HOT_ADD_DISK		_IO (MD_MAJOR, 0x28)
 #define SET_DISK_FAULTY		_IO (MD_MAJOR, 0x29)
 #define HOT_GENERATE_ERROR	_IO (MD_MAJOR, 0x2a)
+#define SET_BITMAP_FILE		_IOW (MD_MAJOR, 0x2b, int)
 
 /* usage */
 #define RUN_ARRAY		_IOW (MD_MAJOR, 0x30, mdu_param_t)
@@ -106,6 +108,11 @@ typedef struct mdu_start_info_s {
 
 } mdu_start_info_t;
 
+typedef struct mdu_bitmap_file_s
+{
+	char pathname[4096];
+} mdu_bitmap_file_t;
+
 typedef struct mdu_param_s
 {
 	int			personality;	/* 1,2,3,4 */
-- 
cgit v1.2.3-70-g09d2


From 77ad4bc706fe6c52ab953f31c287a6af712d080c Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Tue, 21 Jun 2005 17:17:21 -0700
Subject: [PATCH] md: enable the bitmap write-back daemon and wait for it.

Currently we don't wait for updates to the bitmap to be flushed to disk
properly.  The infrastructure all there, but it isn't being used....

A separate kernel thread (bitmap_writeback_daemon) is needed to wait for each
page as we cannot get callbacks when a page write completes.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/md/bitmap.c         | 119 ++++++++++++++++++++------------------------
 include/linux/raid/bitmap.h |  13 +----
 2 files changed, 55 insertions(+), 77 deletions(-)

(limited to 'include')

diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index 9462fdd517c..86b6b037fa4 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -261,30 +261,33 @@ char *file_path(struct file *file, char *buf, int count)
 /*
  * write out a page
  */
-static int write_page(struct page *page, int wait)
+static int write_page(struct bitmap *bitmap, struct page *page, int wait)
 {
 	int ret = -ENOMEM;
 
 	lock_page(page);
 
-	if (page->mapping == NULL)
-		goto unlock_out;
-	else if (i_size_read(page->mapping->host) < page->index << PAGE_SHIFT) {
-		ret = -ENOENT;
-		goto unlock_out;
-	}
-
 	ret = page->mapping->a_ops->prepare_write(NULL, page, 0, PAGE_SIZE);
 	if (!ret)
 		ret = page->mapping->a_ops->commit_write(NULL, page, 0,
 			PAGE_SIZE);
 	if (ret) {
-unlock_out:
 		unlock_page(page);
 		return ret;
 	}
 
 	set_page_dirty(page); /* force it to be written out */
+
+	if (!wait) {
+		/* add to list to be waited for by daemon */
+		struct page_list *item = mempool_alloc(bitmap->write_pool, GFP_NOIO);
+		item->page = page;
+		page_cache_get(page);
+		spin_lock(&bitmap->write_lock);
+		list_add(&item->list, &bitmap->complete_pages);
+		spin_unlock(&bitmap->write_lock);
+		md_wakeup_thread(bitmap->writeback_daemon);
+	}
 	return write_one_page(page, wait);
 }
 
@@ -343,14 +346,13 @@ int bitmap_update_sb(struct bitmap *bitmap)
 		spin_unlock_irqrestore(&bitmap->lock, flags);
 		return 0;
 	}
-	page_cache_get(bitmap->sb_page);
 	spin_unlock_irqrestore(&bitmap->lock, flags);
 	sb = (bitmap_super_t *)kmap(bitmap->sb_page);
 	sb->events = cpu_to_le64(bitmap->mddev->events);
 	if (!bitmap->mddev->degraded)
 		sb->events_cleared = cpu_to_le64(bitmap->mddev->events);
 	kunmap(bitmap->sb_page);
-	return write_page(bitmap->sb_page, 0);
+	return write_page(bitmap, bitmap->sb_page, 0);
 }
 
 /* print out the bitmap file superblock */
@@ -556,10 +558,10 @@ static void bitmap_file_unmap(struct bitmap *bitmap)
 static void bitmap_stop_daemons(struct bitmap *bitmap);
 
 /* dequeue the next item in a page list -- don't call from irq context */
-static struct page_list *dequeue_page(struct bitmap *bitmap,
-					struct list_head *head)
+static struct page_list *dequeue_page(struct bitmap *bitmap)
 {
 	struct page_list *item = NULL;
+	struct list_head *head = &bitmap->complete_pages;
 
 	spin_lock(&bitmap->write_lock);
 	if (list_empty(head))
@@ -573,23 +575,15 @@ out:
 
 static void drain_write_queues(struct bitmap *bitmap)
 {
-	struct list_head *queues[] = { 	&bitmap->complete_pages, NULL };
-	struct list_head *head;
 	struct page_list *item;
-	int i;
 
-	for (i = 0; queues[i]; i++) {
-		head = queues[i];
-		while ((item = dequeue_page(bitmap, head))) {
-			page_cache_release(item->page);
-			mempool_free(item, bitmap->write_pool);
-		}
+	while ((item = dequeue_page(bitmap))) {
+		/* don't bother to wait */
+		page_cache_release(item->page);
+		mempool_free(item, bitmap->write_pool);
 	}
 
-	spin_lock(&bitmap->write_lock);
-	bitmap->writes_pending = 0; /* make sure waiters continue */
 	wake_up(&bitmap->write_wait);
-	spin_unlock(&bitmap->write_lock);
 }
 
 static void bitmap_file_put(struct bitmap *bitmap)
@@ -734,13 +728,13 @@ int bitmap_unplug(struct bitmap *bitmap)
 		spin_unlock_irqrestore(&bitmap->lock, flags);
 
 		if (attr & (BITMAP_PAGE_DIRTY | BITMAP_PAGE_NEEDWRITE))
-			if (write_page(page, 0))
+			if (write_page(bitmap, page, 0))
 				return 1;
 	}
 	if (wait) { /* if any writes were performed, we need to wait on them */
 		spin_lock_irq(&bitmap->write_lock);
 		wait_event_lock_irq(bitmap->write_wait,
-			bitmap->writes_pending == 0, bitmap->write_lock,
+			list_empty(&bitmap->complete_pages), bitmap->write_lock,
 			wake_up_process(bitmap->writeback_daemon->tsk));
 		spin_unlock_irq(&bitmap->write_lock);
 	}
@@ -841,7 +835,7 @@ static int bitmap_init_from_disk(struct bitmap *bitmap, int in_sync)
 				 */
 				memset(page_address(page) + offset, 0xff,
 					PAGE_SIZE - offset);
-				ret = write_page(page, 1);
+				ret = write_page(bitmap, page, 1);
 				if (ret) {
 					kunmap(page);
 					/* release, page not in filemap yet */
@@ -934,7 +928,7 @@ int bitmap_daemon_work(struct bitmap *bitmap)
 			}
 			spin_unlock_irqrestore(&bitmap->lock, flags);
 			if (attr & BITMAP_PAGE_NEEDWRITE) {
-				if (write_page(page, 0))
+				if (write_page(bitmap, page, 0))
 					bitmap_file_kick(bitmap);
 				page_cache_release(page);
 			}
@@ -950,7 +944,7 @@ int bitmap_daemon_work(struct bitmap *bitmap)
 				if (get_page_attr(bitmap, lastpage) & BITMAP_PAGE_NEEDWRITE) {
 					clear_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
 					spin_unlock_irqrestore(&bitmap->lock, flags);
-					err = write_page(lastpage, 0);
+					err = write_page(bitmap, lastpage, 0);
 				} else {
 					set_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
 					spin_unlock_irqrestore(&bitmap->lock, flags);
@@ -998,7 +992,7 @@ int bitmap_daemon_work(struct bitmap *bitmap)
 		if (get_page_attr(bitmap, lastpage) &BITMAP_PAGE_NEEDWRITE) {
 			clear_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
 			spin_unlock_irqrestore(&bitmap->lock, flags);
-			err = write_page(lastpage, 0);
+			err = write_page(bitmap, lastpage, 0);
 		} else {
 			set_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
 			spin_unlock_irqrestore(&bitmap->lock, flags);
@@ -1034,46 +1028,40 @@ static void bitmap_writeback_daemon(mddev_t *mddev)
 	struct page_list *item;
 	int err = 0;
 
-	while (1) {
-		PRINTK("%s: bitmap writeback daemon waiting...\n", bmname(bitmap));
-		down_interruptible(&bitmap->write_done);
-		if (signal_pending(current)) {
-			printk(KERN_INFO
-			    "%s: bitmap writeback daemon got signal, exiting...\n",
-			    bmname(bitmap));
-			break;
-		}
+	if (signal_pending(current)) {
+		printk(KERN_INFO
+		       "%s: bitmap writeback daemon got signal, exiting...\n",
+		       bmname(bitmap));
+		err = -EINTR;
+		goto out;
+	}
 
-		PRINTK("%s: bitmap writeback daemon woke up...\n", bmname(bitmap));
-		/* wait on bitmap page writebacks */
-		while ((item = dequeue_page(bitmap, &bitmap->complete_pages))) {
-			page = item->page;
-			mempool_free(item, bitmap->write_pool);
-			PRINTK("wait on page writeback: %p %lu\n", page, bitmap->writes_pending);
-			wait_on_page_writeback(page);
-			PRINTK("finished page writeback: %p %lu\n", page, bitmap->writes_pending);
-			spin_lock(&bitmap->write_lock);
-			if (!--bitmap->writes_pending)
-				wake_up(&bitmap->write_wait);
-			spin_unlock(&bitmap->write_lock);
-			err = PageError(page);
-			page_cache_release(page);
-			if (err) {
-				printk(KERN_WARNING "%s: bitmap file writeback "
-					"failed (page %lu): %d\n",
-					bmname(bitmap), page->index, err);
-				bitmap_file_kick(bitmap);
-				goto out;
-			}
+	PRINTK("%s: bitmap writeback daemon woke up...\n", bmname(bitmap));
+	/* wait on bitmap page writebacks */
+	while ((item = dequeue_page(bitmap))) {
+		page = item->page;
+		mempool_free(item, bitmap->write_pool);
+		PRINTK("wait on page writeback: %p\n", page);
+		wait_on_page_writeback(page);
+		PRINTK("finished page writeback: %p\n", page);
+
+		err = PageError(page);
+		page_cache_release(page);
+		if (err) {
+			printk(KERN_WARNING "%s: bitmap file writeback "
+			       "failed (page %lu): %d\n",
+			       bmname(bitmap), page->index, err);
+			bitmap_file_kick(bitmap);
+			goto out;
 		}
 	}
-out:
+ out:
+	wake_up(&bitmap->write_wait);
 	if (err) {
 		printk(KERN_INFO "%s: bitmap writeback daemon exiting (%d)\n",
-			bmname(bitmap), err);
+		       bmname(bitmap), err);
 		daemon_exit(bitmap, &bitmap->writeback_daemon);
 	}
-	return;
 }
 
 static int bitmap_start_daemon(struct bitmap *bitmap, mdk_thread_t **ptr,
@@ -1375,7 +1363,7 @@ int bitmap_setallbits(struct bitmap *bitmap)
 		spin_unlock_irqrestore(&bitmap->lock, flags);
 		memset(kmap(page), 0xff, PAGE_SIZE);
 		kunmap(page);
-		if (write_page(page, 0))
+		if (write_page(bitmap, page, 0))
 			return 1;
 	}
 
@@ -1443,7 +1431,6 @@ int bitmap_create(mddev_t *mddev)
 	mddev->bitmap = bitmap;
 
 	spin_lock_init(&bitmap->write_lock);
-	init_MUTEX_LOCKED(&bitmap->write_done);
 	INIT_LIST_HEAD(&bitmap->complete_pages);
 	init_waitqueue_head(&bitmap->write_wait);
 	bitmap->write_pool = mempool_create(WRITE_POOL_SIZE, write_pool_alloc,
diff --git a/include/linux/raid/bitmap.h b/include/linux/raid/bitmap.h
index f785cf26cba..cfe60cfc8f3 100644
--- a/include/linux/raid/bitmap.h
+++ b/include/linux/raid/bitmap.h
@@ -233,21 +233,12 @@ struct bitmap {
 	unsigned long daemon_sleep; /* how many seconds between updates? */
 
 	/*
-	 * bitmap write daemon - this daemon performs writes to the bitmap file
-	 * this thread is only needed because of a limitation in ext3 (jbd)
-	 * that does not allow a task to have two journal transactions ongoing
-	 * simultaneously (even if the transactions are for two different
-	 * filesystems) -- in the case of bitmap, that would be the filesystem
-	 * that the bitmap file resides on and the filesystem that is mounted
-	 * on the md device -- see current->journal_info in jbd/transaction.c
+	 * bitmap_writeback_daemon waits for file-pages that have been written,
+	 * as there is no way to get a call-back when a page write completes.
 	 */
 	mdk_thread_t *writeback_daemon;
 	spinlock_t write_lock;
-	struct semaphore write_ready;
-	struct semaphore write_done;
-	unsigned long writes_pending;
 	wait_queue_head_t write_wait;
-	struct list_head write_pages;
 	struct list_head complete_pages;
 	mempool_t *write_pool;
 };
-- 
cgit v1.2.3-70-g09d2


From 191ea9b2c7cc3ebbe0678834ab710d7d95ad3f9a Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Tue, 21 Jun 2005 17:17:23 -0700
Subject: [PATCH] md: raid1 support for bitmap intent logging

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/md/raid1.c         | 182 ++++++++++++++++++++++++++++++++++++++-------
 include/linux/raid/raid1.h |  16 +++-
 2 files changed, 168 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 3c5c916cb09..0fd4c3bfc85 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -12,6 +12,15 @@
  * Fixes to reconstruction by Jakob Østergaard" <jakob@ostenfeld.dk>
  * Various fixes by Neil Brown <neilb@cse.unsw.edu.au>
  *
+ * Changes by Peter T. Breuer <ptb@it.uc3m.es> 31/1/2003 to support
+ * bitmapped intelligence in resync:
+ *
+ *      - bitmap marked during normal i/o
+ *      - bitmap used to skip nondirty blocks during sync
+ *
+ * Additions to bitmap code, (C) 2003-2004 Paul Clements, SteelEye Technology:
+ * - persistent bitmap code
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2, or (at your option)
@@ -22,7 +31,16 @@
  * Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
+#include "dm-bio-list.h"
 #include <linux/raid/raid1.h>
+#include <linux/raid/bitmap.h>
+
+#define DEBUG 0
+#if DEBUG
+#define PRINTK(x...) printk(x)
+#else
+#define PRINTK(x...)
+#endif
 
 /*
  * Number of guaranteed r1bios in case of extreme VM load:
@@ -287,9 +305,11 @@ static int raid1_end_write_request(struct bio *bio, unsigned int bytes_done, int
 	/*
 	 * this branch is our 'one mirror IO has finished' event handler:
 	 */
-	if (!uptodate)
+	if (!uptodate) {
 		md_error(r1_bio->mddev, conf->mirrors[mirror].rdev);
-	else
+		/* an I/O failed, we can't clear the bitmap */
+		set_bit(R1BIO_Degraded, &r1_bio->state);
+	} else
 		/*
 		 * Set R1BIO_Uptodate in our master bio, so that
 		 * we will return a good error code for to the higher
@@ -309,6 +329,10 @@ static int raid1_end_write_request(struct bio *bio, unsigned int bytes_done, int
 	 * already.
 	 */
 	if (atomic_dec_and_test(&r1_bio->remaining)) {
+		/* clear the bitmap if all writes complete successfully */
+		bitmap_endwrite(r1_bio->mddev->bitmap, r1_bio->sector,
+				r1_bio->sectors,
+				!test_bit(R1BIO_Degraded, &r1_bio->state));
 		md_write_end(r1_bio->mddev);
 		raid_end_bio_io(r1_bio);
 	}
@@ -458,7 +482,10 @@ static void unplug_slaves(mddev_t *mddev)
 
 static void raid1_unplug(request_queue_t *q)
 {
-	unplug_slaves(q->queuedata);
+	mddev_t *mddev = q->queuedata;
+
+	unplug_slaves(mddev);
+	md_wakeup_thread(mddev->thread);
 }
 
 static int raid1_issue_flush(request_queue_t *q, struct gendisk *disk,
@@ -501,16 +528,16 @@ static void device_barrier(conf_t *conf, sector_t sect)
 {
 	spin_lock_irq(&conf->resync_lock);
 	wait_event_lock_irq(conf->wait_idle, !waitqueue_active(&conf->wait_resume),
-			    conf->resync_lock, unplug_slaves(conf->mddev));
+			    conf->resync_lock, raid1_unplug(conf->mddev->queue));
 	
 	if (!conf->barrier++) {
 		wait_event_lock_irq(conf->wait_idle, !conf->nr_pending,
-				    conf->resync_lock, unplug_slaves(conf->mddev));
+				    conf->resync_lock, raid1_unplug(conf->mddev->queue));
 		if (conf->nr_pending)
 			BUG();
 	}
 	wait_event_lock_irq(conf->wait_resume, conf->barrier < RESYNC_DEPTH,
-			    conf->resync_lock, unplug_slaves(conf->mddev));
+			    conf->resync_lock, raid1_unplug(conf->mddev->queue));
 	conf->next_resync = sect;
 	spin_unlock_irq(&conf->resync_lock);
 }
@@ -522,8 +549,12 @@ static int make_request(request_queue_t *q, struct bio * bio)
 	mirror_info_t *mirror;
 	r1bio_t *r1_bio;
 	struct bio *read_bio;
-	int i, disks;
+	int i, targets = 0, disks;
 	mdk_rdev_t *rdev;
+	struct bitmap *bitmap = mddev->bitmap;
+	unsigned long flags;
+	struct bio_list bl;
+
 
 	/*
 	 * Register the new request and wait if the reconstruction
@@ -554,7 +585,7 @@ static int make_request(request_queue_t *q, struct bio * bio)
 
 	r1_bio->master_bio = bio;
 	r1_bio->sectors = bio->bi_size >> 9;
-
+	r1_bio->state = 0;
 	r1_bio->mddev = mddev;
 	r1_bio->sector = bio->bi_sector;
 
@@ -597,6 +628,13 @@ static int make_request(request_queue_t *q, struct bio * bio)
 	 * bios[x] to bio
 	 */
 	disks = conf->raid_disks;
+#if 0
+	{ static int first=1;
+	if (first) printk("First Write sector %llu disks %d\n",
+			  (unsigned long long)r1_bio->sector, disks);
+	first = 0;
+	}
+#endif
 	rcu_read_lock();
 	for (i = 0;  i < disks; i++) {
 		if ((rdev=conf->mirrors[i].rdev) != NULL &&
@@ -607,13 +645,21 @@ static int make_request(request_queue_t *q, struct bio * bio)
 				r1_bio->bios[i] = NULL;
 			} else
 				r1_bio->bios[i] = bio;
+			targets++;
 		} else
 			r1_bio->bios[i] = NULL;
 	}
 	rcu_read_unlock();
 
-	atomic_set(&r1_bio->remaining, 1);
+	if (targets < conf->raid_disks) {
+		/* array is degraded, we will not clear the bitmap
+		 * on I/O completion (see raid1_end_write_request) */
+		set_bit(R1BIO_Degraded, &r1_bio->state);
+	}
+
+	atomic_set(&r1_bio->remaining, 0);
 
+	bio_list_init(&bl);
 	for (i = 0; i < disks; i++) {
 		struct bio *mbio;
 		if (!r1_bio->bios[i])
@@ -629,14 +675,23 @@ static int make_request(request_queue_t *q, struct bio * bio)
 		mbio->bi_private = r1_bio;
 
 		atomic_inc(&r1_bio->remaining);
-		generic_make_request(mbio);
-	}
 
-	if (atomic_dec_and_test(&r1_bio->remaining)) {
-		md_write_end(mddev);
-		raid_end_bio_io(r1_bio);
+		bio_list_add(&bl, mbio);
 	}
 
+	bitmap_startwrite(bitmap, bio->bi_sector, r1_bio->sectors);
+	spin_lock_irqsave(&conf->device_lock, flags);
+	bio_list_merge(&conf->pending_bio_list, &bl);
+	bio_list_init(&bl);
+
+	blk_plug_device(mddev->queue);
+	spin_unlock_irqrestore(&conf->device_lock, flags);
+
+#if 0
+	while ((bio = bio_list_pop(&bl)) != NULL)
+		generic_make_request(bio);
+#endif
+
 	return 0;
 }
 
@@ -716,7 +771,7 @@ static void close_sync(conf_t *conf)
 {
 	spin_lock_irq(&conf->resync_lock);
 	wait_event_lock_irq(conf->wait_resume, !conf->barrier,
-			    conf->resync_lock, 	unplug_slaves(conf->mddev));
+			    conf->resync_lock, 	raid1_unplug(conf->mddev->queue));
 	spin_unlock_irq(&conf->resync_lock);
 
 	if (conf->barrier) BUG();
@@ -830,10 +885,11 @@ static int end_sync_read(struct bio *bio, unsigned int bytes_done, int error)
 	 * or re-read if the read failed.
 	 * We don't do much here, just schedule handling by raid1d
 	 */
-	if (!uptodate)
+	if (!uptodate) {
 		md_error(r1_bio->mddev,
 			 conf->mirrors[r1_bio->read_disk].rdev);
-	else
+		set_bit(R1BIO_Degraded, &r1_bio->state);
+	} else
 		set_bit(R1BIO_Uptodate, &r1_bio->state);
 	rdev_dec_pending(conf->mirrors[r1_bio->read_disk].rdev, conf->mddev);
 	reschedule_retry(r1_bio);
@@ -857,8 +913,10 @@ static int end_sync_write(struct bio *bio, unsigned int bytes_done, int error)
 			mirror = i;
 			break;
 		}
-	if (!uptodate)
+	if (!uptodate) {
 		md_error(mddev, conf->mirrors[mirror].rdev);
+		set_bit(R1BIO_Degraded, &r1_bio->state);
+	}
 	update_head_pos(mirror, r1_bio);
 
 	if (atomic_dec_and_test(&r1_bio->remaining)) {
@@ -878,6 +936,9 @@ static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio)
 
 	bio = r1_bio->bios[r1_bio->read_disk];
 
+/*
+	if (r1_bio->sector == 0) printk("First sync write startss\n");
+*/
 	/*
 	 * schedule writes
 	 */
@@ -905,10 +966,12 @@ static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio)
 		atomic_inc(&conf->mirrors[i].rdev->nr_pending);
 		atomic_inc(&r1_bio->remaining);
 		md_sync_acct(conf->mirrors[i].rdev->bdev, wbio->bi_size >> 9);
+
 		generic_make_request(wbio);
 	}
 
 	if (atomic_dec_and_test(&r1_bio->remaining)) {
+		/* if we're here, all write(s) have completed, so clean up */
 		md_done_sync(mddev, r1_bio->sectors, 1);
 		put_buf(r1_bio);
 	}
@@ -937,6 +1000,26 @@ static void raid1d(mddev_t *mddev)
 	for (;;) {
 		char b[BDEVNAME_SIZE];
 		spin_lock_irqsave(&conf->device_lock, flags);
+
+		if (conf->pending_bio_list.head) {
+			bio = bio_list_get(&conf->pending_bio_list);
+			blk_remove_plug(mddev->queue);
+			spin_unlock_irqrestore(&conf->device_lock, flags);
+			/* flush any pending bitmap writes to disk before proceeding w/ I/O */
+			if (bitmap_unplug(mddev->bitmap) != 0)
+				printk("%s: bitmap file write failed!\n", mdname(mddev));
+
+			while (bio) { /* submit pending writes */
+				struct bio *next = bio->bi_next;
+				bio->bi_next = NULL;
+				generic_make_request(bio);
+				bio = next;
+			}
+			unplug = 1;
+
+			continue;
+		}
+
 		if (list_empty(head))
 			break;
 		r1_bio = list_entry(head->prev, r1bio_t, retry_list);
@@ -1020,17 +1103,43 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i
 	int disk;
 	int i;
 	int write_targets = 0;
+	int sync_blocks;
 
 	if (!conf->r1buf_pool)
+	{
+/*
+		printk("sync start - bitmap %p\n", mddev->bitmap);
+*/
 		if (init_resync(conf))
 			return 0;
+	}
 
 	max_sector = mddev->size << 1;
 	if (sector_nr >= max_sector) {
+		/* If we aborted, we need to abort the
+		 * sync on the 'current' bitmap chunk (there will
+		 * only be one in raid1 resync.
+		 * We can find the current addess in mddev->curr_resync
+		 */
+		if (!conf->fullsync) {
+			if (mddev->curr_resync < max_sector)
+				bitmap_end_sync(mddev->bitmap,
+						mddev->curr_resync,
+						&sync_blocks, 1);
+			bitmap_close_sync(mddev->bitmap);
+		}
+		if (mddev->curr_resync >= max_sector)
+			conf->fullsync = 0;
 		close_sync(conf);
 		return 0;
 	}
 
+	if (!conf->fullsync &&
+	    !bitmap_start_sync(mddev->bitmap, sector_nr, &sync_blocks)) {
+		/* We can skip this block, and probably several more */
+		*skipped = 1;
+		return sync_blocks;
+	}
 	/*
 	 * If there is non-resync activity waiting for us then
 	 * put in a delay to throttle resync.
@@ -1069,6 +1178,7 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i
 
 	r1_bio->mddev = mddev;
 	r1_bio->sector = sector_nr;
+	r1_bio->state = 0;
 	set_bit(R1BIO_IsSync, &r1_bio->state);
 	r1_bio->read_disk = disk;
 
@@ -1103,6 +1213,11 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i
 		bio->bi_bdev = conf->mirrors[i].rdev->bdev;
 		bio->bi_private = r1_bio;
 	}
+
+	if (write_targets + 1 < conf->raid_disks)
+		/* array degraded, can't clear bitmap */
+		set_bit(R1BIO_Degraded, &r1_bio->state);
+
 	if (write_targets == 0) {
 		/* There is nowhere to write, so all non-sync
 		 * drives must be failed - so we are finished
@@ -1122,6 +1237,14 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i
 			len = (max_sector - sector_nr) << 9;
 		if (len == 0)
 			break;
+		if (!conf->fullsync && sync_blocks == 0)
+			if (!bitmap_start_sync(mddev->bitmap,
+					       sector_nr, &sync_blocks))
+				break;
+		if (sync_blocks < (PAGE_SIZE>>9))
+			BUG();
+		if (len > (sync_blocks<<9)) len = sync_blocks<<9;
+
 		for (i=0 ; i < conf->raid_disks; i++) {
 			bio = r1_bio->bios[i];
 			if (bio->bi_end_io) {
@@ -1144,6 +1267,7 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i
 		}
 		nr_sectors += len>>9;
 		sector_nr += len>>9;
+		sync_blocks -= (len>>9);
 	} while (r1_bio->bios[disk]->bi_vcnt < RESYNC_PAGES);
  bio_full:
 	bio = r1_bio->bios[disk];
@@ -1232,6 +1356,9 @@ static int run(mddev_t *mddev)
 	init_waitqueue_head(&conf->wait_idle);
 	init_waitqueue_head(&conf->wait_resume);
 
+	bio_list_init(&conf->pending_bio_list);
+	bio_list_init(&conf->flushing_bio_list);
+
 	if (!conf->working_disks) {
 		printk(KERN_ERR "raid1: no operational mirrors for %s\n",
 			mdname(mddev));
@@ -1260,16 +1387,15 @@ static int run(mddev_t *mddev)
 	conf->last_used = j;
 
 
-
-	{
-		mddev->thread = md_register_thread(raid1d, mddev, "%s_raid1");
-		if (!mddev->thread) {
-			printk(KERN_ERR 
-				"raid1: couldn't allocate thread for %s\n", 
-				mdname(mddev));
-			goto out_free_conf;
-		}
+	mddev->thread = md_register_thread(raid1d, mddev, "%s_raid1");
+	if (!mddev->thread) {
+		printk(KERN_ERR
+		       "raid1: couldn't allocate thread for %s\n",
+		       mdname(mddev));
+		goto out_free_conf;
 	}
+	if (mddev->bitmap) mddev->thread->timeout = mddev->bitmap->daemon_sleep * HZ;
+
 	printk(KERN_INFO 
 		"raid1: raid set %s active with %d out of %d mirrors\n",
 		mdname(mddev), mddev->raid_disks - mddev->degraded, 
@@ -1394,7 +1520,7 @@ static int raid1_reshape(mddev_t *mddev, int raid_disks)
 	spin_lock_irq(&conf->resync_lock);
 	conf->barrier++;
 	wait_event_lock_irq(conf->wait_idle, !conf->nr_pending,
-			    conf->resync_lock, unplug_slaves(mddev));
+			    conf->resync_lock, raid1_unplug(mddev->queue));
 	spin_unlock_irq(&conf->resync_lock);
 
 	/* ok, everything is stopped */
diff --git a/include/linux/raid/raid1.h b/include/linux/raid/raid1.h
index abbfdd9afe1..9d93cf12e89 100644
--- a/include/linux/raid/raid1.h
+++ b/include/linux/raid/raid1.h
@@ -36,12 +36,21 @@ struct r1_private_data_s {
 	spinlock_t		device_lock;
 
 	struct list_head	retry_list;
+	/* queue pending writes and submit them on unplug */
+	struct bio_list		pending_bio_list;
+	/* queue of writes that have been unplugged */
+	struct bio_list		flushing_bio_list;
+
 	/* for use when syncing mirrors: */
 
 	spinlock_t		resync_lock;
-	int nr_pending;
-	int barrier;
+	int			nr_pending;
+	int			barrier;
 	sector_t		next_resync;
+	int			fullsync;  /* set to 1 if a full sync is needed,
+					    * (fresh device added).
+					    * Cleared when a sync completes.
+					    */
 
 	wait_queue_head_t	wait_idle;
 	wait_queue_head_t	wait_resume;
@@ -85,14 +94,17 @@ struct r1bio_s {
 	int			read_disk;
 
 	struct list_head	retry_list;
+	struct bitmap_update	*bitmap_update;
 	/*
 	 * if the IO is in WRITE direction, then multiple bios are used.
 	 * We choose the number when they are allocated.
 	 */
 	struct bio		*bios[0];
+	/* DO NOT PUT ANY NEW FIELDS HERE - bios array is contiguously alloced*/
 };
 
 /* bits for r1bio.state */
 #define	R1BIO_Uptodate	0
 #define	R1BIO_IsSync	1
+#define	R1BIO_Degraded	2
 #endif
-- 
cgit v1.2.3-70-g09d2


From 41158c7eb22312cfaa256744e1553bb4042ff085 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Tue, 21 Jun 2005 17:17:25 -0700
Subject: [PATCH] md: optimise reconstruction when re-adding a recently failed
 drive.

When an array is degraded, bit in the intent-bitmap are never cleared.  So if
a recently failed drive is re-added, we only need to reconstruct the block
that are still reflected in the bitmap.

This patch adds support for this re-adding.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/md/md.c           | 71 ++++++++++++++++++++++++++++++++++++-----------
 drivers/md/raid1.c        |  7 ++++-
 include/linux/raid/md_k.h |  4 +++
 3 files changed, 65 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index b02f8d1d77e..789b114f860 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -577,6 +577,8 @@ static int super_90_validate(mddev_t *mddev, mdk_rdev_t *rdev)
 	mdp_disk_t *desc;
 	mdp_super_t *sb = (mdp_super_t *)page_address(rdev->sb_page);
 
+	rdev->raid_disk = -1;
+	rdev->in_sync = 0;
 	if (mddev->raid_disks == 0) {
 		mddev->major_version = 0;
 		mddev->minor_version = sb->minor_version;
@@ -607,16 +609,24 @@ static int super_90_validate(mddev_t *mddev, mdk_rdev_t *rdev)
 		memcpy(mddev->uuid+12,&sb->set_uuid3, 4);
 
 		mddev->max_disks = MD_SB_DISKS;
-	} else {
-		__u64 ev1;
-		ev1 = md_event(sb);
+	} else if (mddev->pers == NULL) {
+		/* Insist on good event counter while assembling */
+		__u64 ev1 = md_event(sb);
 		++ev1;
 		if (ev1 < mddev->events) 
 			return -EINVAL;
-	}
+	} else if (mddev->bitmap) {
+		/* if adding to array with a bitmap, then we can accept an
+		 * older device ... but not too old.
+		 */
+		__u64 ev1 = md_event(sb);
+		if (ev1 < mddev->bitmap->events_cleared)
+			return 0;
+	} else /* just a hot-add of a new device, leave raid_disk at -1 */
+		return 0;
+
 	if (mddev->level != LEVEL_MULTIPATH) {
-		rdev->raid_disk = -1;
-		rdev->in_sync = rdev->faulty = 0;
+		rdev->faulty = 0;
 		desc = sb->disks + rdev->desc_nr;
 
 		if (desc->state & (1<<MD_DISK_FAULTY))
@@ -626,7 +636,8 @@ static int super_90_validate(mddev_t *mddev, mdk_rdev_t *rdev)
 			rdev->in_sync = 1;
 			rdev->raid_disk = desc->raid_disk;
 		}
-	}
+	} else /* MULTIPATH are always insync */
+		rdev->in_sync = 1;
 	return 0;
 }
 
@@ -868,6 +879,8 @@ static int super_1_validate(mddev_t *mddev, mdk_rdev_t *rdev)
 {
 	struct mdp_superblock_1 *sb = (struct mdp_superblock_1*)page_address(rdev->sb_page);
 
+	rdev->raid_disk = -1;
+	rdev->in_sync = 0;
 	if (mddev->raid_disks == 0) {
 		mddev->major_version = 1;
 		mddev->patch_version = 0;
@@ -885,13 +898,21 @@ static int super_1_validate(mddev_t *mddev, mdk_rdev_t *rdev)
 		memcpy(mddev->uuid, sb->set_uuid, 16);
 
 		mddev->max_disks =  (4096-256)/2;
-	} else {
-		__u64 ev1;
-		ev1 = le64_to_cpu(sb->events);
+	} else if (mddev->pers == NULL) {
+		/* Insist of good event counter while assembling */
+		__u64 ev1 = le64_to_cpu(sb->events);
 		++ev1;
 		if (ev1 < mddev->events)
 			return -EINVAL;
-	}
+	} else if (mddev->bitmap) {
+		/* If adding to array with a bitmap, then we can accept an
+		 * older device, but not too old.
+		 */
+		__u64 ev1 = le64_to_cpu(sb->events);
+		if (ev1 < mddev->bitmap->events_cleared)
+			return 0;
+	} else /* just a hot-add of a new device, leave raid_disk at -1 */
+		return 0;
 
 	if (mddev->level != LEVEL_MULTIPATH) {
 		int role;
@@ -899,14 +920,10 @@ static int super_1_validate(mddev_t *mddev, mdk_rdev_t *rdev)
 		role = le16_to_cpu(sb->dev_roles[rdev->desc_nr]);
 		switch(role) {
 		case 0xffff: /* spare */
-			rdev->in_sync = 0;
 			rdev->faulty = 0;
-			rdev->raid_disk = -1;
 			break;
 		case 0xfffe: /* faulty */
-			rdev->in_sync = 0;
 			rdev->faulty = 1;
-			rdev->raid_disk = -1;
 			break;
 		default:
 			rdev->in_sync = 1;
@@ -914,7 +931,9 @@ static int super_1_validate(mddev_t *mddev, mdk_rdev_t *rdev)
 			rdev->raid_disk = role;
 			break;
 		}
-	}
+	} else /* MULTIPATH are always insync */
+		rdev->in_sync = 1;
+
 	return 0;
 }
 
@@ -2155,6 +2174,18 @@ static int add_new_disk(mddev_t * mddev, mdu_disk_info_t *info)
 				PTR_ERR(rdev));
 			return PTR_ERR(rdev);
 		}
+		/* set save_raid_disk if appropriate */
+		if (!mddev->persistent) {
+			if (info->state & (1<<MD_DISK_SYNC)  &&
+			    info->raid_disk < mddev->raid_disks)
+				rdev->raid_disk = info->raid_disk;
+			else
+				rdev->raid_disk = -1;
+		} else
+			super_types[mddev->major_version].
+				validate_super(mddev, rdev);
+		rdev->saved_raid_disk = rdev->raid_disk;
+
 		rdev->in_sync = 0; /* just to be sure */
 		rdev->raid_disk = -1;
 		err = bind_rdev_to_array(rdev, mddev);
@@ -3706,6 +3737,14 @@ void md_check_recovery(mddev_t *mddev)
 				mddev->pers->spare_active(mddev);
 			}
 			md_update_sb(mddev);
+
+			/* if array is no-longer degraded, then any saved_raid_disk
+			 * information must be scrapped
+			 */
+			if (!mddev->degraded)
+				ITERATE_RDEV(mddev,rdev,rtmp)
+					rdev->saved_raid_disk = -1;
+
 			mddev->recovery = 0;
 			/* flag recovery needed just to double check */
 			set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index c3b4772cfae..3f5234fe359 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -811,9 +811,12 @@ static int raid1_add_disk(mddev_t *mddev, mdk_rdev_t *rdev)
 {
 	conf_t *conf = mddev->private;
 	int found = 0;
-	int mirror;
+	int mirror = 0;
 	mirror_info_t *p;
 
+	if (rdev->saved_raid_disk >= 0 &&
+	    conf->mirrors[rdev->saved_raid_disk].rdev == NULL)
+		mirror = rdev->saved_raid_disk;
 	for (mirror=0; mirror < mddev->raid_disks; mirror++)
 		if ( !(p=conf->mirrors+mirror)->rdev) {
 
@@ -830,6 +833,8 @@ static int raid1_add_disk(mddev_t *mddev, mdk_rdev_t *rdev)
 			p->head_position = 0;
 			rdev->raid_disk = mirror;
 			found = 1;
+			if (rdev->saved_raid_disk != mirror)
+				conf->fullsync = 1;
 			p->rdev = rdev;
 			break;
 		}
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index 16e94a9f0f8..6cdcb4434c6 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -183,6 +183,10 @@ struct mdk_rdev_s
 
 	int desc_nr;			/* descriptor index in the superblock */
 	int raid_disk;			/* role of device in array */
+	int saved_raid_disk;		/* role that device used to have in the
+					 * array and could again if we did a partial
+					 * resync from the bitmap
+					 */
 
 	atomic_t	nr_pending;	/* number of pending requests.
 					 * only maintained for arrays that
-- 
cgit v1.2.3-70-g09d2


From 3d310eb7b3df1252e8595d059d982b0a9825a137 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Tue, 21 Jun 2005 17:17:26 -0700
Subject: [PATCH] md: fix deadlock due to md thread processing delayed
 requests.

Before completing a 'write' the md superblock might need to be updated.
This is best done by the md_thread.

The current code schedules this up and queues the write request for later
handling by the md_thread.

However some personalities (Raid5/raid6) will deadlock if the md_thread
tries to submit requests to its own array.

So this patch changes things so the processes submitting the request waits
for the superblock to be written and then submits the request itself.

This fixes a recently-created deadlock in raid5/raid6

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/md/md.c           | 45 ++++++++++++++++-----------------------------
 drivers/md/raid1.c        |  4 ++--
 drivers/md/raid10.c       |  3 +--
 drivers/md/raid5.c        |  3 +--
 drivers/md/raid6main.c    |  3 +--
 include/linux/raid/md.h   |  2 +-
 include/linux/raid/md_k.h |  2 +-
 7 files changed, 23 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 789b114f860..7075bebb7f3 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -224,8 +224,8 @@ static mddev_t * mddev_find(dev_t unit)
 	INIT_LIST_HEAD(&new->all_mddevs);
 	init_timer(&new->safemode_timer);
 	atomic_set(&new->active, 1);
-	bio_list_init(&new->write_list);
 	spin_lock_init(&new->write_lock);
+	init_waitqueue_head(&new->sb_wait);
 
 	new->queue = blk_alloc_queue(GFP_KERNEL);
 	if (!new->queue) {
@@ -1307,6 +1307,7 @@ repeat:
 	if (!mddev->persistent) {
 		mddev->sb_dirty = 0;
 		spin_unlock(&mddev->write_lock);
+		wake_up(&mddev->sb_wait);
 		return;
 	}
 	spin_unlock(&mddev->write_lock);
@@ -1348,6 +1349,7 @@ repeat:
 	}
 	mddev->sb_dirty = 0;
 	spin_unlock(&mddev->write_lock);
+	wake_up(&mddev->sb_wait);
 
 }
 
@@ -3368,29 +3370,26 @@ void md_done_sync(mddev_t *mddev, int blocks, int ok)
 
 /* md_write_start(mddev, bi)
  * If we need to update some array metadata (e.g. 'active' flag
- * in superblock) before writing, queue bi for later writing
- * and return 0, else return 1 and it will be written now
+ * in superblock) before writing, schedule a superblock update
+ * and wait for it to complete.
  */
-int md_write_start(mddev_t *mddev, struct bio *bi)
+void md_write_start(mddev_t *mddev, struct bio *bi)
 {
+	DEFINE_WAIT(w);
 	if (bio_data_dir(bi) != WRITE)
-		return 1;
+		return;
 
 	atomic_inc(&mddev->writes_pending);
-	spin_lock(&mddev->write_lock);
-	if (mddev->in_sync == 0 && mddev->sb_dirty == 0) {
-		spin_unlock(&mddev->write_lock);
-		return 1;
-	}
-	bio_list_add(&mddev->write_list, bi);
-
 	if (mddev->in_sync) {
-		mddev->in_sync = 0;
-		mddev->sb_dirty = 1;
+		spin_lock(&mddev->write_lock);
+		if (mddev->in_sync) {
+			mddev->in_sync = 0;
+			mddev->sb_dirty = 1;
+			md_wakeup_thread(mddev->thread);
+		}
+		spin_unlock(&mddev->write_lock);
 	}
-	spin_unlock(&mddev->write_lock);
-	md_wakeup_thread(mddev->thread);
-	return 0;
+	wait_event(mddev->sb_wait, mddev->sb_dirty==0);
 }
 
 void md_write_end(mddev_t *mddev)
@@ -3685,7 +3684,6 @@ void md_check_recovery(mddev_t *mddev)
 		mddev->sb_dirty ||
 		test_bit(MD_RECOVERY_NEEDED, &mddev->recovery) ||
 		test_bit(MD_RECOVERY_DONE, &mddev->recovery) ||
-		mddev->write_list.head ||
 		(mddev->safemode == 1) ||
 		(mddev->safemode == 2 && ! atomic_read(&mddev->writes_pending)
 		 && !mddev->in_sync && mddev->recovery_cp == MaxSector)
@@ -3694,7 +3692,6 @@ void md_check_recovery(mddev_t *mddev)
 
 	if (mddev_trylock(mddev)==0) {
 		int spares =0;
-		struct bio *blist;
 
 		spin_lock(&mddev->write_lock);
 		if (mddev->safemode && !atomic_read(&mddev->writes_pending) &&
@@ -3704,21 +3701,11 @@ void md_check_recovery(mddev_t *mddev)
 		}
 		if (mddev->safemode == 1)
 			mddev->safemode = 0;
-		blist = bio_list_get(&mddev->write_list);
 		spin_unlock(&mddev->write_lock);
 
 		if (mddev->sb_dirty)
 			md_update_sb(mddev);
 
-		while (blist) {
-			struct bio *b = blist;
-			blist = blist->bi_next;
-			b->bi_next = NULL;
-			generic_make_request(b);
-			/* we already counted this, so need to un-count */
-			md_write_end(mddev);
-		}
-
 
 		if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) &&
 		    !test_bit(MD_RECOVERY_DONE, &mddev->recovery)) {
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 3f5234fe359..98b09773e79 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -561,8 +561,8 @@ static int make_request(request_queue_t *q, struct bio * bio)
 	 * thread has put up a bar for new requests.
 	 * Continue immediately if no resync is active currently.
 	 */
-	if (md_write_start(mddev, bio)==0)
-		return 0;
+	md_write_start(mddev, bio); /* wait on superblock update early */
+
 	spin_lock_irq(&conf->resync_lock);
 	wait_event_lock_irq(conf->wait_resume, !conf->barrier, conf->resync_lock, );
 	conf->nr_pending++;
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 8476515bfdc..fd7324a86d1 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -700,8 +700,7 @@ static int make_request(request_queue_t *q, struct bio * bio)
 		return 0;
 	}
 
-	if (md_write_start(mddev, bio) == 0)
-		return 0;
+	md_write_start(mddev, bio);
 
 	/*
 	 * Register the new request and wait if the reconstruction
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 1ce3f5aaa98..93a9726cc2d 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -1411,8 +1411,7 @@ static int make_request (request_queue_t *q, struct bio * bi)
 	sector_t logical_sector, last_sector;
 	struct stripe_head *sh;
 
-	if (md_write_start(mddev, bi)==0)
-		return 0;
+	md_write_start(mddev, bi);
 
 	if (bio_data_dir(bi)==WRITE) {
 		disk_stat_inc(mddev->gendisk, writes);
diff --git a/drivers/md/raid6main.c b/drivers/md/raid6main.c
index d9c385496dc..f62ea1a73d0 100644
--- a/drivers/md/raid6main.c
+++ b/drivers/md/raid6main.c
@@ -1570,8 +1570,7 @@ static int make_request (request_queue_t *q, struct bio * bi)
 	sector_t logical_sector, last_sector;
 	struct stripe_head *sh;
 
-	if (md_write_start(mddev, bi)==0)
-		return 0;
+	md_write_start(mddev, bi);
 
 	if (bio_data_dir(bi)==WRITE) {
 		disk_stat_inc(mddev->gendisk, writes);
diff --git a/include/linux/raid/md.h b/include/linux/raid/md.h
index cfde8f497d6..75f41d8faed 100644
--- a/include/linux/raid/md.h
+++ b/include/linux/raid/md.h
@@ -69,7 +69,7 @@ extern mdk_thread_t * md_register_thread (void (*run) (mddev_t *mddev),
 extern void md_unregister_thread (mdk_thread_t *thread);
 extern void md_wakeup_thread(mdk_thread_t *thread);
 extern void md_check_recovery(mddev_t *mddev);
-extern int md_write_start(mddev_t *mddev, struct bio *bi);
+extern void md_write_start(mddev_t *mddev, struct bio *bi);
 extern void md_write_end(mddev_t *mddev);
 extern void md_handle_safemode(mddev_t *mddev);
 extern void md_done_sync(mddev_t *mddev, int blocks, int ok);
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index 6cdcb4434c6..3e977025cf4 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -261,7 +261,7 @@ struct mddev_s
 	sector_t			recovery_cp;
 
 	spinlock_t			write_lock;
-	struct bio_list			write_list;
+	wait_queue_head_t		sb_wait;	/* for waiting on superblock updates */
 
 	unsigned int			safemode;	/* if set, update "clean" superblock
 							 * when no writes pending.
-- 
cgit v1.2.3-70-g09d2


From a654b9d8f851f4ca02649d5825cbe6c608adb10c Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Tue, 21 Jun 2005 17:17:27 -0700
Subject: [PATCH] md: allow md intent bitmap to be stored near the superblock.

This provides an alternate to storing the bitmap in a separate file.  The
bitmap can be stored at a given offset from the superblock.  Obviously the
creator of the array must make sure this doesn't intersect with data....
After is good for version-0.90 superblocks.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/md/bitmap.c         | 132 +++++++++++++++++++++++++++++++++++---------
 drivers/md/md.c             |  40 +++++++++++++-
 include/linux/raid/bitmap.h |   2 +
 include/linux/raid/md.h     |  15 ++++-
 include/linux/raid/md_k.h   |   4 ++
 include/linux/raid/md_p.h   |   7 ++-
 6 files changed, 170 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index 204564dc6a0..030d6861051 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -116,7 +116,7 @@ static unsigned char *bitmap_alloc_page(struct bitmap *bitmap)
 	if (!page)
 		printk("%s: bitmap_alloc_page FAILED\n", bmname(bitmap));
 	else
-		printk("%s: bitmap_alloc_page: allocated page at %p\n",
+		PRINTK("%s: bitmap_alloc_page: allocated page at %p\n",
 			bmname(bitmap), page);
 	return page;
 }
@@ -258,13 +258,61 @@ char *file_path(struct file *file, char *buf, int count)
  * basic page I/O operations
  */
 
+/* IO operations when bitmap is stored near all superblocks */
+static struct page *read_sb_page(mddev_t *mddev, long offset, unsigned long index)
+{
+	/* choose a good rdev and read the page from there */
+
+	mdk_rdev_t *rdev;
+	struct list_head *tmp;
+	struct page *page = alloc_page(GFP_KERNEL);
+	sector_t target;
+
+	if (!page)
+		return ERR_PTR(-ENOMEM);
+	do {
+		ITERATE_RDEV(mddev, rdev, tmp)
+			if (rdev->in_sync && !rdev->faulty)
+				goto found;
+		return ERR_PTR(-EIO);
+
+	found:
+		target = (rdev->sb_offset << 1) + offset + index * (PAGE_SIZE/512);
+
+	} while (!sync_page_io(rdev->bdev, target, PAGE_SIZE, page, READ));
+
+	page->index = index;
+	return page;
+}
+
+static int write_sb_page(mddev_t *mddev, long offset, struct page *page, int wait)
+{
+	mdk_rdev_t *rdev;
+	struct list_head *tmp;
+
+	ITERATE_RDEV(mddev, rdev, tmp)
+		if (rdev->in_sync && !rdev->faulty)
+			md_super_write(mddev, rdev,
+				       (rdev->sb_offset<<1) + offset
+				       + page->index * (PAGE_SIZE/512),
+				       PAGE_SIZE,
+				       page);
+
+	if (wait)
+		wait_event(mddev->sb_wait, atomic_read(&mddev->pending_writes)==0);
+	return 0;
+}
+
 /*
- * write out a page
+ * write out a page to a file
  */
 static int write_page(struct bitmap *bitmap, struct page *page, int wait)
 {
 	int ret = -ENOMEM;
 
+	if (bitmap->file == NULL)
+		return write_sb_page(bitmap->mddev, bitmap->offset, page, wait);
+
 	lock_page(page);
 
 	ret = page->mapping->a_ops->prepare_write(NULL, page, 0, PAGE_SIZE);
@@ -394,7 +442,12 @@ static int bitmap_read_sb(struct bitmap *bitmap)
 	int err = -EINVAL;
 
 	/* page 0 is the superblock, read it... */
-	bitmap->sb_page = read_page(bitmap->file, 0, &bytes_read);
+	if (bitmap->file)
+		bitmap->sb_page = read_page(bitmap->file, 0, &bytes_read);
+	else {
+		bitmap->sb_page = read_sb_page(bitmap->mddev, bitmap->offset, 0);
+		bytes_read = PAGE_SIZE;
+	}
 	if (IS_ERR(bitmap->sb_page)) {
 		err = PTR_ERR(bitmap->sb_page);
 		bitmap->sb_page = NULL;
@@ -625,14 +678,16 @@ static void bitmap_file_kick(struct bitmap *bitmap)
 	bitmap_mask_state(bitmap, BITMAP_STALE, MASK_SET);
 	bitmap_update_sb(bitmap);
 
-	path = kmalloc(PAGE_SIZE, GFP_KERNEL);
-	if (path)
-		ptr = file_path(bitmap->file, path, PAGE_SIZE);
+	if (bitmap->file) {
+		path = kmalloc(PAGE_SIZE, GFP_KERNEL);
+		if (path)
+			ptr = file_path(bitmap->file, path, PAGE_SIZE);
 
-	printk(KERN_ALERT "%s: kicking failed bitmap file %s from array!\n",
-		bmname(bitmap), ptr ? ptr : "");
+		printk(KERN_ALERT "%s: kicking failed bitmap file %s from array!\n",
+		       bmname(bitmap), ptr ? ptr : "");
 
-	kfree(path);
+		kfree(path);
+	}
 
 	bitmap_file_put(bitmap);
 
@@ -676,7 +731,7 @@ static void bitmap_file_set_bit(struct bitmap *bitmap, sector_t block)
 	void *kaddr;
 	unsigned long chunk = block >> CHUNK_BLOCK_SHIFT(bitmap);
 
-	if (!bitmap->file || !bitmap->filemap) {
+	if (!bitmap->filemap) {
 		return;
 	}
 
@@ -715,7 +770,7 @@ int bitmap_unplug(struct bitmap *bitmap)
 	 * flushed out to disk */
 	for (i = 0; i < bitmap->file_pages; i++) {
 		spin_lock_irqsave(&bitmap->lock, flags);
-		if (!bitmap->file || !bitmap->filemap) {
+		if (!bitmap->filemap) {
 			spin_unlock_irqrestore(&bitmap->lock, flags);
 			return 0;
 		}
@@ -732,11 +787,15 @@ int bitmap_unplug(struct bitmap *bitmap)
 				return 1;
 	}
 	if (wait) { /* if any writes were performed, we need to wait on them */
-		spin_lock_irq(&bitmap->write_lock);
-		wait_event_lock_irq(bitmap->write_wait,
-			list_empty(&bitmap->complete_pages), bitmap->write_lock,
-			wake_up_process(bitmap->writeback_daemon->tsk));
-		spin_unlock_irq(&bitmap->write_lock);
+		if (bitmap->file) {
+			spin_lock_irq(&bitmap->write_lock);
+			wait_event_lock_irq(bitmap->write_wait,
+					    list_empty(&bitmap->complete_pages), bitmap->write_lock,
+					    wake_up_process(bitmap->writeback_daemon->tsk));
+			spin_unlock_irq(&bitmap->write_lock);
+		} else
+			wait_event(bitmap->mddev->sb_wait,
+				   atomic_read(&bitmap->mddev->pending_writes)==0);
 	}
 	return 0;
 }
@@ -764,7 +823,7 @@ static int bitmap_init_from_disk(struct bitmap *bitmap, int in_sync)
 	chunks = bitmap->chunks;
 	file = bitmap->file;
 
-	BUG_ON(!file);
+	BUG_ON(!file && !bitmap->offset);
 
 #if INJECT_FAULTS_3
 	outofdate = 1;
@@ -779,7 +838,7 @@ static int bitmap_init_from_disk(struct bitmap *bitmap, int in_sync)
 
 	num_pages = (bytes + sizeof(bitmap_super_t) + PAGE_SIZE - 1) / PAGE_SIZE;
 
-	if (i_size_read(file->f_mapping->host) < bytes + sizeof(bitmap_super_t)) {
+	if (file && i_size_read(file->f_mapping->host) < bytes + sizeof(bitmap_super_t)) {
 		printk(KERN_INFO "%s: bitmap file too short %lu < %lu\n",
 			bmname(bitmap),
 			(unsigned long) i_size_read(file->f_mapping->host),
@@ -816,14 +875,18 @@ static int bitmap_init_from_disk(struct bitmap *bitmap, int in_sync)
 				 */
 				page = bitmap->sb_page;
 				offset = sizeof(bitmap_super_t);
-			} else {
+			} else if (file) {
 				page = read_page(file, index, &dummy);
-				if (IS_ERR(page)) { /* read error */
-					ret = PTR_ERR(page);
-					goto out;
-				}
+				offset = 0;
+			} else {
+				page = read_sb_page(bitmap->mddev, bitmap->offset, index);
 				offset = 0;
 			}
+			if (IS_ERR(page)) { /* read error */
+				ret = PTR_ERR(page);
+				goto out;
+			}
+
 			oldindex = index;
 			oldpage = page;
 			kmap(page);
@@ -874,6 +937,19 @@ out:
 	return ret;
 }
 
+void bitmap_write_all(struct bitmap *bitmap)
+{
+	/* We don't actually write all bitmap blocks here,
+	 * just flag them as needing to be written
+	 */
+
+	unsigned long chunks = bitmap->chunks;
+	unsigned long bytes = (chunks+7)/8 + sizeof(bitmap_super_t);
+	unsigned long num_pages = (bytes + PAGE_SIZE-1) / PAGE_SIZE;
+	while (num_pages--)
+		bitmap->filemap_attr[num_pages] |= BITMAP_PAGE_NEEDWRITE;
+}
+
 
 static void bitmap_count_page(struct bitmap *bitmap, sector_t offset, int inc)
 {
@@ -913,7 +989,7 @@ int bitmap_daemon_work(struct bitmap *bitmap)
 	for (j = 0; j < bitmap->chunks; j++) {
 		bitmap_counter_t *bmc;
 		spin_lock_irqsave(&bitmap->lock, flags);
-		if (!bitmap->file || !bitmap->filemap) {
+		if (!bitmap->filemap) {
 			/* error or shutdown */
 			spin_unlock_irqrestore(&bitmap->lock, flags);
 			break;
@@ -1072,6 +1148,7 @@ static int bitmap_start_daemon(struct bitmap *bitmap, mdk_thread_t **ptr,
 
 	spin_lock_irqsave(&bitmap->lock, flags);
 	*ptr = NULL;
+
 	if (!bitmap->file) /* no need for daemon if there's no backing file */
 		goto out_unlock;
 
@@ -1416,9 +1493,11 @@ int bitmap_create(mddev_t *mddev)
 
 	BUG_ON(sizeof(bitmap_super_t) != 256);
 
-	if (!file) /* bitmap disabled, nothing to do */
+	if (!file && !mddev->bitmap_offset) /* bitmap disabled, nothing to do */
 		return 0;
 
+	BUG_ON(file && mddev->bitmap_offset);
+
 	bitmap = kmalloc(sizeof(*bitmap), GFP_KERNEL);
 	if (!bitmap)
 		return -ENOMEM;
@@ -1438,7 +1517,8 @@ int bitmap_create(mddev_t *mddev)
 		return -ENOMEM;
 
 	bitmap->file = file;
-	get_file(file);
+	bitmap->offset = mddev->bitmap_offset;
+	if (file) get_file(file);
 	/* read superblock from bitmap file (this sets bitmap->chunksize) */
 	err = bitmap_read_sb(bitmap);
 	if (err)
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 7075bebb7f3..fde8acfac32 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -337,7 +337,7 @@ static int bi_complete(struct bio *bio, unsigned int bytes_done, int error)
 	return 0;
 }
 
-static int sync_page_io(struct block_device *bdev, sector_t sector, int size,
+int sync_page_io(struct block_device *bdev, sector_t sector, int size,
 		   struct page *page, int rw)
 {
 	struct bio *bio = bio_alloc(GFP_NOIO, 1);
@@ -609,6 +609,17 @@ static int super_90_validate(mddev_t *mddev, mdk_rdev_t *rdev)
 		memcpy(mddev->uuid+12,&sb->set_uuid3, 4);
 
 		mddev->max_disks = MD_SB_DISKS;
+
+		if (sb->state & (1<<MD_SB_BITMAP_PRESENT) &&
+		    mddev->bitmap_file == NULL) {
+			if (mddev->level != 1) {
+				/* FIXME use a better test */
+				printk(KERN_WARNING "md: bitmaps only support for raid1\n");
+				return -EINVAL;
+			}
+			mddev->bitmap_offset = (MD_SB_BYTES >> 9);
+		}
+
 	} else if (mddev->pers == NULL) {
 		/* Insist on good event counter while assembling */
 		__u64 ev1 = md_event(sb);
@@ -702,6 +713,9 @@ static void super_90_sync(mddev_t *mddev, mdk_rdev_t *rdev)
 	sb->layout = mddev->layout;
 	sb->chunk_size = mddev->chunk_size;
 
+	if (mddev->bitmap && mddev->bitmap_file == NULL)
+		sb->state |= (1<<MD_SB_BITMAP_PRESENT);
+
 	sb->disks[0].state = (1<<MD_DISK_REMOVED);
 	ITERATE_RDEV(mddev,rdev2,tmp) {
 		mdp_disk_t *d;
@@ -898,6 +912,15 @@ static int super_1_validate(mddev_t *mddev, mdk_rdev_t *rdev)
 		memcpy(mddev->uuid, sb->set_uuid, 16);
 
 		mddev->max_disks =  (4096-256)/2;
+
+		if ((le32_to_cpu(sb->feature_map) & 1) &&
+		    mddev->bitmap_file == NULL ) {
+			if (mddev->level != 1) {
+				printk(KERN_WARNING "md: bitmaps only supported for raid1\n");
+				return -EINVAL;
+			}
+			mddev->bitmap_offset = (__s32)le32_to_cpu(sb->bitmap_offset);
+		}
 	} else if (mddev->pers == NULL) {
 		/* Insist of good event counter while assembling */
 		__u64 ev1 = le64_to_cpu(sb->events);
@@ -960,6 +983,11 @@ static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev)
 	else
 		sb->resync_offset = cpu_to_le64(0);
 
+	if (mddev->bitmap && mddev->bitmap_file == NULL) {
+		sb->bitmap_offset = cpu_to_le32((__u32)mddev->bitmap_offset);
+		sb->feature_map = cpu_to_le32(1);
+	}
+
 	max_dev = 0;
 	ITERATE_RDEV(mddev,rdev2,tmp)
 		if (rdev2->desc_nr+1 > max_dev)
@@ -2406,7 +2434,8 @@ static int set_bitmap_file(mddev_t *mddev, int fd)
 			mdname(mddev));
 		fput(mddev->bitmap_file);
 		mddev->bitmap_file = NULL;
-	}
+	} else
+		mddev->bitmap_offset = 0; /* file overrides offset */
 	return err;
 }
 
@@ -3774,6 +3803,13 @@ void md_check_recovery(mddev_t *mddev)
 			set_bit(MD_RECOVERY_RUNNING, &mddev->recovery);
 			if (!spares)
 				set_bit(MD_RECOVERY_SYNC, &mddev->recovery);
+			if (spares && mddev->bitmap && ! mddev->bitmap->file) {
+				/* We are adding a device or devices to an array
+				 * which has the bitmap stored on all devices.
+				 * So make sure all bitmap pages get written
+				 */
+				bitmap_write_all(mddev->bitmap);
+			}
 			mddev->sync_thread = md_register_thread(md_do_sync,
 								mddev,
 								"%s_resync");
diff --git a/include/linux/raid/bitmap.h b/include/linux/raid/bitmap.h
index cfe60cfc8f3..e24b74b1115 100644
--- a/include/linux/raid/bitmap.h
+++ b/include/linux/raid/bitmap.h
@@ -217,6 +217,7 @@ struct bitmap {
 	/* bitmap spinlock */
 	spinlock_t lock;
 
+	long offset; /* offset from superblock if file is NULL */
 	struct file *file; /* backing disk file */
 	struct page *sb_page; /* cached copy of the bitmap file superblock */
 	struct page **filemap; /* list of cache pages for the file */
@@ -255,6 +256,7 @@ void bitmap_print_sb(struct bitmap *bitmap);
 int bitmap_update_sb(struct bitmap *bitmap);
 
 int  bitmap_setallbits(struct bitmap *bitmap);
+void bitmap_write_all(struct bitmap *bitmap);
 
 /* these are exported */
 int bitmap_startwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors);
diff --git a/include/linux/raid/md.h b/include/linux/raid/md.h
index 75f41d8faed..ffa316ce4dc 100644
--- a/include/linux/raid/md.h
+++ b/include/linux/raid/md.h
@@ -60,7 +60,14 @@
  */
 #define MD_MAJOR_VERSION                0
 #define MD_MINOR_VERSION                90
-#define MD_PATCHLEVEL_VERSION           1
+/*
+ * MD_PATCHLEVEL_VERSION indicates kernel functionality.
+ * >=1 means different superblock formats are selectable using SET_ARRAY_INFO
+ *     and major_version/minor_version accordingly
+ * >=2 means that Internal bitmaps are supported by setting MD_SB_BITMAP_PRESENT
+ *     in the super status byte
+ */
+#define MD_PATCHLEVEL_VERSION           2
 
 extern int register_md_personality (int p_num, mdk_personality_t *p);
 extern int unregister_md_personality (int p_num);
@@ -78,6 +85,12 @@ extern void md_unplug_mddev(mddev_t *mddev);
 
 extern void md_print_devices (void);
 
+extern void md_super_write(mddev_t *mddev, mdk_rdev_t *rdev,
+			   sector_t sector, int size, struct page *page);
+extern int sync_page_io(struct block_device *bdev, sector_t sector, int size,
+			struct page *page, int rw);
+
+
 #define MD_BUG(x...) { printk("md: bug in file %s, line %d\n", __FILE__, __LINE__); md_print_devices(); }
 
 #endif 
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index 3e977025cf4..a3725b57fb7 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -273,6 +273,10 @@ struct mddev_s
 
 	struct bitmap                   *bitmap; /* the bitmap for the device */
 	struct file			*bitmap_file; /* the bitmap file */
+	long				bitmap_offset; /* offset from superblock of
+							* start of bitmap. May be
+							* negative, but not '0'
+							*/
 
 	struct list_head		all_mddevs;
 };
diff --git a/include/linux/raid/md_p.h b/include/linux/raid/md_p.h
index 8ba95d67329..8e592a25a8b 100644
--- a/include/linux/raid/md_p.h
+++ b/include/linux/raid/md_p.h
@@ -96,6 +96,7 @@ typedef struct mdp_device_descriptor_s {
 #define MD_SB_CLEAN		0
 #define MD_SB_ERRORS		1
 
+#define	MD_SB_BITMAP_PRESENT	8 /* bitmap may be present nearby */
 typedef struct mdp_superblock_s {
 	/*
 	 * Constant generic information
@@ -184,7 +185,7 @@ struct mdp_superblock_1 {
 	/* constant array information - 128 bytes */
 	__u32	magic;		/* MD_SB_MAGIC: 0xa92b4efc - little endian */
 	__u32	major_version;	/* 1 */
-	__u32	feature_map;	/* 0 for now */
+	__u32	feature_map;	/* bit 0 set if 'bitmap_offset' is meaningful */
 	__u32	pad0;		/* always set to 0 when writing */
 
 	__u8	set_uuid[16];	/* user-space generated. */
@@ -197,6 +198,10 @@ struct mdp_superblock_1 {
 
 	__u32	chunksize;	/* in 512byte sectors */
 	__u32	raid_disks;
+	__u32	bitmap_offset;	/* sectors after start of superblock that bitmap starts
+				 * NOTE: signed, so bitmap can be before superblock
+				 * only meaningful of feature_map[0] is set.
+				 */
 	__u8	pad1[128-96];	/* set to 0 when written */
 
 	/* constant this-device information - 64 bytes */
-- 
cgit v1.2.3-70-g09d2


From 7bfa19f2748000d646dbdf8f48258cfe1d257b52 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Tue, 21 Jun 2005 17:17:28 -0700
Subject: [PATCH] md: allow md to update multiple superblocks in parallel.

currently, md updates all superblocks (one on each device) in series.  It
waits for one write to complete before starting the next.  This isn't a big
problem as superblock updates don't happen that often.

However it is neater to do it in parallel, and if the drives in the array have
gone to "sleep" after a period of idleness, then waking them is parallel is
faster (and someone else should be worrying about power drain).

Futher, we will need parallel superblock updates for a future patch which
keeps the intent-logging bitmap near the superblock.

Also remove the silly code that retired superblock updates 100 times.  This
simply never made sense.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/md/md.c           | 85 ++++++++++++++++++++++++++---------------------
 include/linux/raid/md_k.h |  1 +
 2 files changed, 49 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index fde8acfac32..ef3ad99562c 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -328,6 +328,40 @@ static void free_disk_sb(mdk_rdev_t * rdev)
 }
 
 
+static int super_written(struct bio *bio, unsigned int bytes_done, int error)
+{
+	mdk_rdev_t *rdev = bio->bi_private;
+	if (bio->bi_size)
+		return 1;
+
+	if (error || !test_bit(BIO_UPTODATE, &bio->bi_flags))
+		md_error(rdev->mddev, rdev);
+
+	if (atomic_dec_and_test(&rdev->mddev->pending_writes))
+		wake_up(&rdev->mddev->sb_wait);
+	return 0;
+}
+
+void md_super_write(mddev_t *mddev, mdk_rdev_t *rdev,
+		   sector_t sector, int size, struct page *page)
+{
+	/* write first size bytes of page to sector of rdev
+	 * Increment mddev->pending_writes before returning
+	 * and decrement it on completion, waking up sb_wait
+	 * if zero is reached.
+	 * If an error occurred, call md_error
+	 */
+	struct bio *bio = bio_alloc(GFP_NOIO, 1);
+
+	bio->bi_bdev = rdev->bdev;
+	bio->bi_sector = sector;
+	bio_add_page(bio, page, size, 0);
+	bio->bi_private = rdev;
+	bio->bi_end_io = super_written;
+	atomic_inc(&mddev->pending_writes);
+	submit_bio((1<<BIO_RW)|(1<<BIO_RW_SYNC), bio);
+}
+
 static int bi_complete(struct bio *bio, unsigned int bytes_done, int error)
 {
 	if (bio->bi_size)
@@ -1268,30 +1302,6 @@ void md_print_devices(void)
 }
 
 
-static int write_disk_sb(mdk_rdev_t * rdev)
-{
-	char b[BDEVNAME_SIZE];
-	if (!rdev->sb_loaded) {
-		MD_BUG();
-		return 1;
-	}
-	if (rdev->faulty) {
-		MD_BUG();
-		return 1;
-	}
-
-	dprintk(KERN_INFO "(write) %s's sb offset: %llu\n",
-		bdevname(rdev->bdev,b),
-	       (unsigned long long)rdev->sb_offset);
-  
-	if (sync_page_io(rdev->bdev, rdev->sb_offset<<1, MD_SB_BYTES, rdev->sb_page, WRITE))
-		return 0;
-
-	printk("md: write_disk_sb failed for device %s\n", 
-		bdevname(rdev->bdev,b));
-	return 1;
-}
-
 static void sync_sbs(mddev_t * mddev)
 {
 	mdk_rdev_t *rdev;
@@ -1306,7 +1316,7 @@ static void sync_sbs(mddev_t * mddev)
 
 static void md_update_sb(mddev_t * mddev)
 {
-	int err, count = 100;
+	int err;
 	struct list_head *tmp;
 	mdk_rdev_t *rdev;
 	int sync_req;
@@ -1326,6 +1336,7 @@ repeat:
 		MD_BUG();
 		mddev->events --;
 	}
+	mddev->sb_dirty = 2;
 	sync_sbs(mddev);
 
 	/*
@@ -1353,24 +1364,24 @@ repeat:
 
 		dprintk("%s ", bdevname(rdev->bdev,b));
 		if (!rdev->faulty) {
-			err += write_disk_sb(rdev);
+			md_super_write(mddev,rdev,
+				       rdev->sb_offset<<1, MD_SB_BYTES,
+				       rdev->sb_page);
+			dprintk(KERN_INFO "(write) %s's sb offset: %llu\n",
+				bdevname(rdev->bdev,b),
+				(unsigned long long)rdev->sb_offset);
+
 		} else
 			dprintk(")\n");
-		if (!err && mddev->level == LEVEL_MULTIPATH)
+		if (mddev->level == LEVEL_MULTIPATH)
 			/* only need to write one superblock... */
 			break;
 	}
-	if (err) {
-		if (--count) {
-			printk(KERN_ERR "md: errors occurred during superblock"
-				" update, repeating\n");
-			goto repeat;
-		}
-		printk(KERN_ERR \
-			"md: excessive errors occurred during superblock update, exiting\n");
-	}
+	wait_event(mddev->sb_wait, atomic_read(&mddev->pending_writes)==0);
+	/* if there was a failure, sb_dirty was set to 1, and we re-write super */
+
 	spin_lock(&mddev->write_lock);
-	if (mddev->in_sync != sync_req) {
+	if (mddev->in_sync != sync_req|| mddev->sb_dirty == 1) {
 		/* have to write it out again */
 		spin_unlock(&mddev->write_lock);
 		goto repeat;
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index a3725b57fb7..8c14ba565a4 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -262,6 +262,7 @@ struct mddev_s
 
 	spinlock_t			write_lock;
 	wait_queue_head_t		sb_wait;	/* for waiting on superblock updates */
+	atomic_t			pending_writes;	/* number of active superblock writes */
 
 	unsigned int			safemode;	/* if set, update "clean" superblock
 							 * when no writes pending.
-- 
cgit v1.2.3-70-g09d2


From 39730960d94306d7be414e8d54f4e5c071af1278 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Tue, 21 Jun 2005 17:17:28 -0700
Subject: [PATCH] Two small fixes for md verion-1 superblocks.

1/ Must typecast int to (sector_t) before inverting or we
 might not invert enough bits.

2/ When "bitmap_offset" was added to mdp_superblock_1, we didn't increase
   the count of words-used (96 to 100).

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/md/md.c           | 2 +-
 include/linux/raid/md_p.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index ef3ad99562c..65fab74ad73 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -847,7 +847,7 @@ static int super_1_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version)
 	case 0:
 		sb_offset = rdev->bdev->bd_inode->i_size >> 9;
 		sb_offset -= 8*2;
-		sb_offset &= ~(4*2-1);
+		sb_offset &= ~(sector_t)(4*2-1);
 		/* convert from sectors to K */
 		sb_offset /= 2;
 		break;
diff --git a/include/linux/raid/md_p.h b/include/linux/raid/md_p.h
index 8e592a25a8b..dc65cd43549 100644
--- a/include/linux/raid/md_p.h
+++ b/include/linux/raid/md_p.h
@@ -202,7 +202,7 @@ struct mdp_superblock_1 {
 				 * NOTE: signed, so bitmap can be before superblock
 				 * only meaningful of feature_map[0] is set.
 				 */
-	__u8	pad1[128-96];	/* set to 0 when written */
+	__u8	pad1[128-100];	/* set to 0 when written */
 
 	/* constant this-device information - 64 bytes */
 	__u64	data_offset;	/* sector start of data, often 0 */
-- 
cgit v1.2.3-70-g09d2


From b3d5496ea5915fa4848fe307af9f7097f312e932 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sat, 2 Apr 2005 20:31:02 +0200
Subject: [PATCH] I2C: Kill address ranges in non-sensors i2c chip drivers

Some months ago, you killed the address ranges mechanism from all
sensors i2c chip drivers (both the module parameters and the in-code
address lists). I think it was a very good move, as the ranges can
easily be replaced by individual addresses, and this allowed for
significant cleanups in the i2c core (let alone the impressive size
shrink for all these drivers).

Unfortunately you did not do the same for non-sensors i2c chip drivers.
These need the address ranges even less, so we could get rid of the
ranges here as well for another significant i2c core cleanup. Here comes
a patch which does just that. Since the process is exactly the same as
what you did for the other drivers set already, I did not split this one
in parts.

A documentation update is included.

The change saves 308 bytes in the i2c core, and an average 1382 bytes
for chip drivers which use I2C_CLIENT_INSMOD, 126 bytes for those which
do not.

This change is required if we want to merge the sensors and non-sensors
i2c code (and we want to do this).

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Index: gregkh-2.6/Documentation/i2c/writing-clients
===================================================================
---
 Documentation/i2c/writing-clients       | 62 +++++++--------------------------
 drivers/acorn/char/pcf8583.c            |  3 --
 drivers/i2c/chips/isp1301_omap.c        |  1 -
 drivers/i2c/chips/m41t00.c              |  3 --
 drivers/i2c/chips/rtc8564.c             |  3 --
 drivers/i2c/i2c-core.c                  | 35 -------------------
 drivers/macintosh/therm_windtunnel.c    |  6 ++--
 drivers/media/video/adv7170.c           |  6 ----
 drivers/media/video/adv7175.c           |  6 ----
 drivers/media/video/bt819.c             |  6 ----
 drivers/media/video/bt832.c             |  4 +--
 drivers/media/video/bt856.c             |  6 ----
 drivers/media/video/msp3400.c           |  1 -
 drivers/media/video/saa5246a.c          |  1 -
 drivers/media/video/saa5249.c           |  1 -
 drivers/media/video/saa7110.c           |  6 ----
 drivers/media/video/saa7111.c           |  6 ----
 drivers/media/video/saa7114.c           |  6 ----
 drivers/media/video/saa7134/saa6752hs.c |  1 -
 drivers/media/video/saa7185.c           |  6 ----
 drivers/media/video/tda7432.c           |  1 -
 drivers/media/video/tda9840.c           |  1 -
 drivers/media/video/tda9875.c           |  1 -
 drivers/media/video/tda9887.c           |  1 -
 drivers/media/video/tea6415c.c          |  1 -
 drivers/media/video/tea6420.c           |  1 -
 drivers/media/video/tuner-3036.c        | 13 +++----
 drivers/media/video/tuner-core.c        | 11 +++---
 drivers/media/video/tvaudio.c           |  1 -
 drivers/media/video/tveeprom.c          |  1 -
 drivers/media/video/vpx3220.c           |  6 ----
 drivers/video/matrox/matroxfb_maven.c   |  1 -
 include/linux/i2c.h                     | 12 -------
 33 files changed, 27 insertions(+), 193 deletions(-)

(limited to 'include')

diff --git a/Documentation/i2c/writing-clients b/Documentation/i2c/writing-clients
index ad27511e3c7..f482dae81de 100644
--- a/Documentation/i2c/writing-clients
+++ b/Documentation/i2c/writing-clients
@@ -171,45 +171,31 @@ The following lists are used internally:
 
   normal_i2c: filled in by the module writer. 
      A list of I2C addresses which should normally be examined.
-   normal_i2c_range: filled in by the module writer.
-     A list of pairs of I2C addresses, each pair being an inclusive range of
-     addresses which should normally be examined.
    probe: insmod parameter. 
      A list of pairs. The first value is a bus number (-1 for any I2C bus), 
      the second is the address. These addresses are also probed, as if they 
      were in the 'normal' list.
-   probe_range: insmod parameter. 
-     A list of triples. The first value is a bus number (-1 for any I2C bus), 
-     the second and third are addresses.  These form an inclusive range of 
-     addresses that are also probed, as if they were in the 'normal' list.
    ignore: insmod parameter.
      A list of pairs. The first value is a bus number (-1 for any I2C bus), 
      the second is the I2C address. These addresses are never probed. 
      This parameter overrules 'normal' and 'probe', but not the 'force' lists.
-   ignore_range: insmod parameter. 
-     A list of triples. The first value is a bus number (-1 for any I2C bus), 
-     the second and third are addresses. These form an inclusive range of 
-     I2C addresses that are never probed.
-     This parameter overrules 'normal' and 'probe', but not the 'force' lists.
    force: insmod parameter. 
      A list of pairs. The first value is a bus number (-1 for any I2C bus),
      the second is the I2C address. A device is blindly assumed to be on
      the given address, no probing is done. 
 
-Fortunately, as a module writer, you just have to define the `normal' 
-and/or `normal_range' parameters. The complete declaration could look
-like this:
+Fortunately, as a module writer, you just have to define the `normal_i2c' 
+parameter. The complete declaration could look like this:
 
-  /* Scan 0x20 to 0x2f, 0x37, and 0x40 to 0x4f */
-  static unsigned short normal_i2c[] = { 0x37,I2C_CLIENT_END }; 
-  static unsigned short normal_i2c_range[] = { 0x20, 0x2f, 0x40, 0x4f, 
-                                               I2C_CLIENT_END };
+  /* Scan 0x37, and 0x48 to 0x4f */
+  static unsigned short normal_i2c[] = { 0x37, 0x48, 0x49, 0x4a, 0x4b, 0x4c,
+                                         0x4d, 0x4e, 0x4f, I2C_CLIENT_END };
 
   /* Magic definition of all other variables and things */
   I2C_CLIENT_INSMOD;
 
-Note that you *have* to call the two defined variables `normal_i2c' and
-`normal_i2c_range', without any prefix!
+Note that you *have* to call the defined variable `normal_i2c',
+without any prefix!
 
 
 Probing classes (sensors)
@@ -223,39 +209,17 @@ The following lists are used internally. They are all lists of integers.
 
    normal_i2c: filled in by the module writer. Terminated by SENSORS_I2C_END.
      A list of I2C addresses which should normally be examined.
-   normal_i2c_range: filled in by the module writer. Terminated by 
-     SENSORS_I2C_END
-     A list of pairs of I2C addresses, each pair being an inclusive range of
-     addresses which should normally be examined.
    normal_isa: filled in by the module writer. Terminated by SENSORS_ISA_END.
      A list of ISA addresses which should normally be examined.
-   normal_isa_range: filled in by the module writer. Terminated by 
-     SENSORS_ISA_END
-     A list of triples. The first two elements are ISA addresses, being an
-     range of addresses which should normally be examined. The third is the
-     modulo parameter: only addresses which are 0 module this value relative
-     to the first address of the range are actually considered.
    probe: insmod parameter. Initialize this list with SENSORS_I2C_END values.
      A list of pairs. The first value is a bus number (SENSORS_ISA_BUS for
      the ISA bus, -1 for any I2C bus), the second is the address. These
      addresses are also probed, as if they were in the 'normal' list.
-   probe_range: insmod parameter. Initialize this list with SENSORS_I2C_END 
-     values.
-     A list of triples. The first value is a bus number (SENSORS_ISA_BUS for
-     the ISA bus, -1 for any I2C bus), the second and third are addresses. 
-     These form an inclusive range of addresses that are also probed, as
-     if they were in the 'normal' list.
    ignore: insmod parameter. Initialize this list with SENSORS_I2C_END values.
      A list of pairs. The first value is a bus number (SENSORS_ISA_BUS for
      the ISA bus, -1 for any I2C bus), the second is the I2C address. These
      addresses are never probed. This parameter overrules 'normal' and 
      'probe', but not the 'force' lists.
-   ignore_range: insmod parameter. Initialize this list with SENSORS_I2C_END 
-      values.
-     A list of triples. The first value is a bus number (SENSORS_ISA_BUS for
-     the ISA bus, -1 for any I2C bus), the second and third are addresses. 
-     These form an inclusive range of I2C addresses that are never probed.
-     This parameter overrules 'normal' and 'probe', but not the 'force' lists.
 
 Also used is a list of pointers to sensors_force_data structures:
    force_data: insmod parameters. A list, ending with an element of which
@@ -269,16 +233,14 @@ Also used is a list of pointers to sensors_force_data structures:
 So we have a generic insmod variabled `force', and chip-specific variables
 `force_CHIPNAME'.
 
-Fortunately, as a module writer, you just have to define the `normal' 
-and/or `normal_range' parameters, and define what chip names are used. 
+Fortunately, as a module writer, you just have to define the `normal_i2c' 
+and `normal_isa' parameters, and define what chip names are used. 
 The complete declaration could look like this:
-  /* Scan i2c addresses 0x20 to 0x2f, 0x37, and 0x40 to 0x4f
-  static unsigned short normal_i2c[] = {0x37,SENSORS_I2C_END};
-  static unsigned short normal_i2c_range[] = {0x20,0x2f,0x40,0x4f,
-                                              SENSORS_I2C_END};
+  /* Scan i2c addresses 0x37, and 0x48 to 0x4f */
+  static unsigned short normal_i2c[] = { 0x37, 0x48, 0x49, 0x4a, 0x4b, 0x4c,
+                                         0x4d, 0x4e, 0x4f, I2C_CLIENT_END };
   /* Scan ISA address 0x290 */
   static unsigned int normal_isa[] = {0x0290,SENSORS_ISA_END};
-  static unsigned int normal_isa_range[] = {SENSORS_ISA_END};
 
   /* Define chips foo and bar, as well as all module parameters and things */
   SENSORS_INSMOD_2(foo,bar);
diff --git a/drivers/acorn/char/pcf8583.c b/drivers/acorn/char/pcf8583.c
index ad7ae7ab892..141b4c237a5 100644
--- a/drivers/acorn/char/pcf8583.c
+++ b/drivers/acorn/char/pcf8583.c
@@ -26,11 +26,8 @@ static unsigned short normal_addr[] = { 0x50, I2C_CLIENT_END };
 
 static struct i2c_client_address_data addr_data = {
 	.normal_i2c		= normal_addr,
-	.normal_i2c_range	= ignore,
 	.probe			= ignore,
-	.probe_range		= ignore,
 	.ignore			= ignore,
-	.ignore_range		= ignore,
 	.force			= ignore,
 };
 
diff --git a/drivers/i2c/chips/isp1301_omap.c b/drivers/i2c/chips/isp1301_omap.c
index 7f29a8aff16..354a2629567 100644
--- a/drivers/i2c/chips/isp1301_omap.c
+++ b/drivers/i2c/chips/isp1301_omap.c
@@ -145,7 +145,6 @@ static inline void notresponding(struct isp1301 *isp)
 static unsigned short normal_i2c[] = {
 	ISP_BASE, ISP_BASE + 1,
 	I2C_CLIENT_END };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 
 I2C_CLIENT_INSMOD;
 
diff --git a/drivers/i2c/chips/m41t00.c b/drivers/i2c/chips/m41t00.c
index e771566dffa..5e463c47bfb 100644
--- a/drivers/i2c/chips/m41t00.c
+++ b/drivers/i2c/chips/m41t00.c
@@ -40,11 +40,8 @@ static unsigned short normal_addr[] = { 0x68, I2C_CLIENT_END };
 
 static struct i2c_client_address_data addr_data = {
 	.normal_i2c		= normal_addr,
-	.normal_i2c_range	= ignore,
 	.probe			= ignore,
-	.probe_range		= ignore,
 	.ignore			= ignore,
-	.ignore_range		= ignore,
 	.force			= ignore,
 };
 
diff --git a/drivers/i2c/chips/rtc8564.c b/drivers/i2c/chips/rtc8564.c
index 5a9deddb626..30f553e7370 100644
--- a/drivers/i2c/chips/rtc8564.c
+++ b/drivers/i2c/chips/rtc8564.c
@@ -66,11 +66,8 @@ static unsigned short normal_addr[] = { 0x51, I2C_CLIENT_END };
 
 static struct i2c_client_address_data addr_data = {
 	.normal_i2c		= normal_addr,
-	.normal_i2c_range	= ignore,
 	.probe			= ignore,
-	.probe_range		= ignore,
 	.ignore			= ignore,
-	.ignore_range		= ignore,
 	.force			= ignore,
 };
 
diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index a22e53badac..4cc8c9f7211 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -742,18 +742,6 @@ int i2c_probe(struct i2c_adapter *adapter,
 				found = 1;
 			}
 		}
-		for (i = 0;
-		     !found && (address_data->ignore_range[i] != I2C_CLIENT_END);
-		     i += 3) {
-			if (((adap_id == address_data->ignore_range[i]) ||
-			    ((address_data->ignore_range[i]==ANY_I2C_BUS))) &&
-			    (addr >= address_data->ignore_range[i+1]) &&
-			    (addr <= address_data->ignore_range[i+2])) {
-				dev_dbg(&adapter->dev, "found ignore_range parameter for adapter %d, "
-					"addr %04x\n", adap_id,addr);
-				found = 1;
-			}
-		}
 		if (found) 
 			continue;
 
@@ -769,17 +757,6 @@ int i2c_probe(struct i2c_adapter *adapter,
 			}
 		}
 
-		for (i = 0;
-		     !found && (address_data->normal_i2c_range[i] != I2C_CLIENT_END);
-		     i += 2) {
-			if ((addr >= address_data->normal_i2c_range[i]) &&
-			    (addr <= address_data->normal_i2c_range[i+1])) {
-				found = 1;
-				dev_dbg(&adapter->dev, "found normal i2c_range entry for adapter %d, "
-					"addr %04x\n", adap_id,addr);
-			}
-		}
-
 		for (i = 0;
 		     !found && (address_data->probe[i] != I2C_CLIENT_END);
 		     i += 2) {
@@ -791,18 +768,6 @@ int i2c_probe(struct i2c_adapter *adapter,
 					"addr %04x\n", adap_id,addr);
 			}
 		}
-		for (i = 0;
-		     !found && (address_data->probe_range[i] != I2C_CLIENT_END);
-		     i += 3) {
-			if (((adap_id == address_data->probe_range[i]) ||
-			   (address_data->probe_range[i] == ANY_I2C_BUS)) &&
-			   (addr >= address_data->probe_range[i+1]) &&
-			   (addr <= address_data->probe_range[i+2])) {
-				found = 1;
-				dev_dbg(&adapter->dev, "found probe_range parameter for adapter %d, "
-					"addr %04x\n", adap_id,addr);
-			}
-		}
 		if (!found) 
 			continue;
 
diff --git a/drivers/macintosh/therm_windtunnel.c b/drivers/macintosh/therm_windtunnel.c
index 0bdb47f08c2..61400f04015 100644
--- a/drivers/macintosh/therm_windtunnel.c
+++ b/drivers/macintosh/therm_windtunnel.c
@@ -51,8 +51,10 @@
 static int 			do_probe( struct i2c_adapter *adapter, int addr, int kind);
 
 /* scan 0x48-0x4f (DS1775) and 0x2c-2x2f (ADM1030) */
-static unsigned short		normal_i2c[] = { 0x49, 0x2c, I2C_CLIENT_END };
-static unsigned short		normal_i2c_range[] = { 0x48, 0x4f, 0x2c, 0x2f, I2C_CLIENT_END };
+static unsigned short		normal_i2c[] = { 0x48, 0x49, 0x4a, 0x4b,
+						 0x4c, 0x4d, 0x4e, 0x4f,
+						 0x2c, 0x2d, 0x2e, 0x2f,
+						 I2C_CLIENT_END };
 
 I2C_CLIENT_INSMOD;
 
diff --git a/drivers/media/video/adv7170.c b/drivers/media/video/adv7170.c
index 80254caa444..e9bf3394296 100644
--- a/drivers/media/video/adv7170.c
+++ b/drivers/media/video/adv7170.c
@@ -384,21 +384,15 @@ static unsigned short normal_i2c[] =
 	I2C_ADV7171 >> 1, (I2C_ADV7171 >> 1) + 1,
 	I2C_CLIENT_END
 };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 
 static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END };
                                                                                 
 static struct i2c_client_address_data addr_data = {
 	.normal_i2c		= normal_i2c,
-	.normal_i2c_range	= normal_i2c_range,
 	.probe			= probe,
-	.probe_range		= probe_range,
 	.ignore			= ignore,
-	.ignore_range		= ignore_range,
 	.force			= force
 };
 
diff --git a/drivers/media/video/adv7175.c b/drivers/media/video/adv7175.c
index 95d0974b0ab..2d5fa44fcd4 100644
--- a/drivers/media/video/adv7175.c
+++ b/drivers/media/video/adv7175.c
@@ -434,21 +434,15 @@ static unsigned short normal_i2c[] =
 	I2C_ADV7176 >> 1, (I2C_ADV7176 >> 1) + 1,
 	I2C_CLIENT_END
 };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 
 static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END };
                                                                                 
 static struct i2c_client_address_data addr_data = {
 	.normal_i2c		= normal_i2c,
-	.normal_i2c_range	= normal_i2c_range,
 	.probe			= probe,
-	.probe_range		= probe_range,
 	.ignore			= ignore,
-	.ignore_range		= ignore_range,
 	.force			= force
 };
 
diff --git a/drivers/media/video/bt819.c b/drivers/media/video/bt819.c
index cf0db2554a8..31d51851bb4 100644
--- a/drivers/media/video/bt819.c
+++ b/drivers/media/video/bt819.c
@@ -500,21 +500,15 @@ static unsigned short normal_i2c[] = {
 	I2C_BT819 >> 1,
 	I2C_CLIENT_END,
 };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 
 static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END };
                                                                                 
 static struct i2c_client_address_data addr_data = {
 	.normal_i2c		= normal_i2c,
-	.normal_i2c_range	= normal_i2c_range,
 	.probe			= probe,
-	.probe_range		= probe_range,
 	.ignore			= ignore,
-	.ignore_range		= ignore_range,
 	.force			= force
 };
 
diff --git a/drivers/media/video/bt832.c b/drivers/media/video/bt832.c
index efe605a113a..07f72f64c5f 100644
--- a/drivers/media/video/bt832.c
+++ b/drivers/media/video/bt832.c
@@ -39,8 +39,8 @@
 MODULE_LICENSE("GPL");
 
 /* Addresses to scan */
-static unsigned short normal_i2c[] = {I2C_CLIENT_END};
-static unsigned short normal_i2c_range[] = {I2C_BT832_ALT1>>1,I2C_BT832_ALT2>>1,I2C_CLIENT_END};
+static unsigned short normal_i2c[] = { I2C_BT832_ALT1>>1, I2C_BT832_ALT2>>1,
+				       I2C_CLIENT_END };
 I2C_CLIENT_INSMOD;
 
 /* ---------------------------------------------------------------------- */
diff --git a/drivers/media/video/bt856.c b/drivers/media/video/bt856.c
index 72c7eb0f8c2..59121a0ec81 100644
--- a/drivers/media/video/bt856.c
+++ b/drivers/media/video/bt856.c
@@ -288,21 +288,15 @@ bt856_command (struct i2c_client *client,
  * concerning the addresses: i2c wants 7 bit (without the r/w bit), so '>>1'
  */
 static unsigned short normal_i2c[] = { I2C_BT856 >> 1, I2C_CLIENT_END };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 
 static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END };
                                                                                 
 static struct i2c_client_address_data addr_data = {
 	.normal_i2c		= normal_i2c,
-	.normal_i2c_range	= normal_i2c_range,
 	.probe			= probe,
-	.probe_range		= probe_range,
 	.ignore			= ignore,
-	.ignore_range		= ignore_range,
 	.force			= force
 };
 
diff --git a/drivers/media/video/msp3400.c b/drivers/media/video/msp3400.c
index 7fbb8581a87..09464d624a6 100644
--- a/drivers/media/video/msp3400.c
+++ b/drivers/media/video/msp3400.c
@@ -147,7 +147,6 @@ static unsigned short normal_i2c[] = {
 	I2C_MSP3400C_ALT  >> 1,
 	I2C_CLIENT_END
 };
-static unsigned short normal_i2c_range[] = {I2C_CLIENT_END,I2C_CLIENT_END};
 I2C_CLIENT_INSMOD;
 
 /* ----------------------------------------------------------------------- */
diff --git a/drivers/media/video/saa5246a.c b/drivers/media/video/saa5246a.c
index ba69f09cbdd..b8054da31ff 100644
--- a/drivers/media/video/saa5246a.c
+++ b/drivers/media/video/saa5246a.c
@@ -64,7 +64,6 @@ static struct video_device saa_template;	/* Declared near bottom */
 
 /* Addresses to scan */
 static unsigned short normal_i2c[]	 = { I2C_ADDRESS, I2C_CLIENT_END };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 I2C_CLIENT_INSMOD;
 
 static struct i2c_client client_template;
diff --git a/drivers/media/video/saa5249.c b/drivers/media/video/saa5249.c
index d74caa139f0..7ffa2e9a9bf 100644
--- a/drivers/media/video/saa5249.c
+++ b/drivers/media/video/saa5249.c
@@ -132,7 +132,6 @@ static struct video_device saa_template;	/* Declared near bottom */
 
 /* Addresses to scan */
 static unsigned short normal_i2c[] = {34>>1,I2C_CLIENT_END};
-static unsigned short normal_i2c_range[] = {I2C_CLIENT_END};
 I2C_CLIENT_INSMOD;
 
 static struct i2c_client client_template;
diff --git a/drivers/media/video/saa7110.c b/drivers/media/video/saa7110.c
index 64273b43853..90b0a0b34f3 100644
--- a/drivers/media/video/saa7110.c
+++ b/drivers/media/video/saa7110.c
@@ -463,21 +463,15 @@ static unsigned short normal_i2c[] = {
 	(I2C_SAA7110 >> 1) + 1,
 	I2C_CLIENT_END
 };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 
 static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END };
                                                                                 
 static struct i2c_client_address_data addr_data = {
 	.normal_i2c		= normal_i2c,
-	.normal_i2c_range	= normal_i2c_range,
 	.probe			= probe,
-	.probe_range		= probe_range,
 	.ignore			= ignore,
-	.ignore_range		= ignore_range,
 	.force			= force
 };
 
diff --git a/drivers/media/video/saa7111.c b/drivers/media/video/saa7111.c
index 0a873112ae2..e305a89f7cd 100644
--- a/drivers/media/video/saa7111.c
+++ b/drivers/media/video/saa7111.c
@@ -482,21 +482,15 @@ saa7111_command (struct i2c_client *client,
  * concerning the addresses: i2c wants 7 bit (without the r/w bit), so '>>1'
  */
 static unsigned short normal_i2c[] = { I2C_SAA7111 >> 1, I2C_CLIENT_END };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 
 static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END };
                                                                                 
 static struct i2c_client_address_data addr_data = {
 	.normal_i2c		= normal_i2c,
-	.normal_i2c_range	= normal_i2c_range,
 	.probe			= probe,
-	.probe_range		= probe_range,
 	.ignore			= ignore,
-	.ignore_range		= ignore_range,
 	.force			= force
 };
 
diff --git a/drivers/media/video/saa7114.c b/drivers/media/video/saa7114.c
index e73023695e5..1ca4e70fed7 100644
--- a/drivers/media/video/saa7114.c
+++ b/drivers/media/video/saa7114.c
@@ -820,21 +820,15 @@ saa7114_command (struct i2c_client *client,
  */
 static unsigned short normal_i2c[] =
     { I2C_SAA7114 >> 1, I2C_SAA7114A >> 1, I2C_CLIENT_END };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 
 static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END };
                                                                                 
 static struct i2c_client_address_data addr_data = {
 	.normal_i2c		= normal_i2c,
-	.normal_i2c_range	= normal_i2c_range,
 	.probe			= probe,
-	.probe_range		= probe_range,
 	.ignore			= ignore,
-	.ignore_range		= ignore_range,
 	.force			= force
 };
 
diff --git a/drivers/media/video/saa7134/saa6752hs.c b/drivers/media/video/saa7134/saa6752hs.c
index 1db02268298..42c2b565c9f 100644
--- a/drivers/media/video/saa7134/saa6752hs.c
+++ b/drivers/media/video/saa7134/saa6752hs.c
@@ -22,7 +22,6 @@
 
 /* Addresses to scan */
 static unsigned short normal_i2c[] = {0x20, I2C_CLIENT_END};
-static unsigned short normal_i2c_range[] = {I2C_CLIENT_END};
 I2C_CLIENT_INSMOD;
 
 MODULE_DESCRIPTION("device driver for saa6752hs MPEG2 encoder");
diff --git a/drivers/media/video/saa7185.c b/drivers/media/video/saa7185.c
index 5f0b224c3cb..5c623fadc8f 100644
--- a/drivers/media/video/saa7185.c
+++ b/drivers/media/video/saa7185.c
@@ -380,21 +380,15 @@ saa7185_command (struct i2c_client *client,
  * concerning the addresses: i2c wants 7 bit (without the r/w bit), so '>>1'
  */
 static unsigned short normal_i2c[] = { I2C_SAA7185 >> 1, I2C_CLIENT_END };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 
 static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END };
                                                                                 
 static struct i2c_client_address_data addr_data = {
 	.normal_i2c		= normal_i2c,
-	.normal_i2c_range	= normal_i2c_range,
 	.probe			= probe,
-	.probe_range		= probe_range,
 	.ignore			= ignore,
-	.ignore_range		= ignore_range,
 	.force			= force
 };
 
diff --git a/drivers/media/video/tda7432.c b/drivers/media/video/tda7432.c
index 376a4a439e9..07ba6d3ed08 100644
--- a/drivers/media/video/tda7432.c
+++ b/drivers/media/video/tda7432.c
@@ -74,7 +74,6 @@ static unsigned short normal_i2c[] = {
 	I2C_TDA7432 >> 1,
 	I2C_CLIENT_END,
 };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END, I2C_CLIENT_END };
 I2C_CLIENT_INSMOD;
 
 /* Structure of address and subaddresses for the tda7432 */
diff --git a/drivers/media/video/tda9840.c b/drivers/media/video/tda9840.c
index b5177c6f54f..c29bdfc3244 100644
--- a/drivers/media/video/tda9840.c
+++ b/drivers/media/video/tda9840.c
@@ -43,7 +43,6 @@ MODULE_PARM_DESC(debug, "Turn on/off device debugging (default:off).");
 
 /* addresses to scan, found only at 0x42 (7-Bit) */
 static unsigned short normal_i2c[] = { I2C_TDA9840, I2C_CLIENT_END };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 
 /* magic definition of all other variables and things */
 I2C_CLIENT_INSMOD;
diff --git a/drivers/media/video/tda9875.c b/drivers/media/video/tda9875.c
index 4f1114c033a..97b113e070f 100644
--- a/drivers/media/video/tda9875.c
+++ b/drivers/media/video/tda9875.c
@@ -44,7 +44,6 @@ static unsigned short normal_i2c[] =  {
     I2C_TDA9875 >> 1,
     I2C_CLIENT_END
 };
-static unsigned short normal_i2c_range[] = {I2C_CLIENT_END};
 I2C_CLIENT_INSMOD;
 
 /* This is a superset of the TDA9875 */
diff --git a/drivers/media/video/tda9887.c b/drivers/media/video/tda9887.c
index debef1910c3..7e6e6dd966a 100644
--- a/drivers/media/video/tda9887.c
+++ b/drivers/media/video/tda9887.c
@@ -33,7 +33,6 @@ static unsigned short normal_i2c[] = {
 	0x96 >>1,
 	I2C_CLIENT_END,
 };
-static unsigned short normal_i2c_range[] = {I2C_CLIENT_END,I2C_CLIENT_END};
 I2C_CLIENT_INSMOD;
 
 /* insmod options */
diff --git a/drivers/media/video/tea6415c.c b/drivers/media/video/tea6415c.c
index 3ec39550bf4..b44db8a7b94 100644
--- a/drivers/media/video/tea6415c.c
+++ b/drivers/media/video/tea6415c.c
@@ -43,7 +43,6 @@ MODULE_PARM_DESC(debug, "Turn on/off device debugging (default:off).");
 
 /* addresses to scan, found only at 0x03 and/or 0x43 (7-bit) */
 static unsigned short normal_i2c[] = { I2C_TEA6415C_1, I2C_TEA6415C_2, I2C_CLIENT_END };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 
 /* magic definition of all other variables and things */
 I2C_CLIENT_INSMOD;
diff --git a/drivers/media/video/tea6420.c b/drivers/media/video/tea6420.c
index bd10710fd90..48d4db7d507 100644
--- a/drivers/media/video/tea6420.c
+++ b/drivers/media/video/tea6420.c
@@ -40,7 +40,6 @@ MODULE_PARM_DESC(debug, "Turn on/off device debugging (default:off).");
 
 /* addresses to scan, found only at 0x4c and/or 0x4d (7-Bit) */
 static unsigned short normal_i2c[] = { I2C_TEA6420_1, I2C_TEA6420_2, I2C_CLIENT_END };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 
 /* magic definition of all other variables and things */
 I2C_CLIENT_INSMOD;
diff --git a/drivers/media/video/tuner-3036.c b/drivers/media/video/tuner-3036.c
index 6b20aa902a8..bedb15e2f23 100644
--- a/drivers/media/video/tuner-3036.c
+++ b/drivers/media/video/tuner-3036.c
@@ -34,19 +34,16 @@ static int this_adap;
 static struct i2c_client client_template;
 
 /* Addresses to scan */
-static unsigned short normal_i2c[] = {I2C_CLIENT_END};
-static unsigned short normal_i2c_range[] = {0x60, 0x61, I2C_CLIENT_END};
+static unsigned short normal_i2c[] = { 0x60, 0x61, I2C_CLIENT_END };
 static unsigned short probe[2]        = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short probe_range[2]  = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short ignore[2]       = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short force[2]        = { I2C_CLIENT_END, I2C_CLIENT_END };
 
 static struct i2c_client_address_data addr_data = {
-	normal_i2c, normal_i2c_range, 
-	probe, probe_range, 
-	ignore, ignore_range, 
-	force
+	.normal_i2c	= normal_i2c,
+	.probe		= probe,
+	.ignore		= ignore,
+	.force		= force,
 };
 
 /* ---------------------------------------------------------------------- */
diff --git a/drivers/media/video/tuner-core.c b/drivers/media/video/tuner-core.c
index 6212388edb7..81882ddab85 100644
--- a/drivers/media/video/tuner-core.c
+++ b/drivers/media/video/tuner-core.c
@@ -28,10 +28,8 @@
 /* standard i2c insmod options */
 static unsigned short normal_i2c[] = {
 	0x4b, /* tda8290 */
-	I2C_CLIENT_END
-};
-static unsigned short normal_i2c_range[] = {
-	0x60, 0x6f,
+	0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67,
+	0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f,
 	I2C_CLIENT_END
 };
 I2C_CLIENT_INSMOD;
@@ -225,9 +223,8 @@ static int tuner_attach(struct i2c_adapter *adap, int addr, int kind)
 static int tuner_probe(struct i2c_adapter *adap)
 {
 	if (0 != addr) {
-		normal_i2c[0]       = addr;
-		normal_i2c_range[0] = addr;
-		normal_i2c_range[1] = addr;
+		normal_i2c[0] = addr;
+		normal_i2c[1] = I2C_CLIENT_END;
 	}
 	this_adap = 0;
 
diff --git a/drivers/media/video/tvaudio.c b/drivers/media/video/tvaudio.c
index 80dc34f18c2..41b635e0d3c 100644
--- a/drivers/media/video/tvaudio.c
+++ b/drivers/media/video/tvaudio.c
@@ -148,7 +148,6 @@ static unsigned short normal_i2c[] = {
 	I2C_TDA9874   >> 1,
 	I2C_PIC16C54  >> 1,
 	I2C_CLIENT_END };
-static unsigned short normal_i2c_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 I2C_CLIENT_INSMOD;
 
 static struct i2c_driver driver;
diff --git a/drivers/media/video/tveeprom.c b/drivers/media/video/tveeprom.c
index e1443a0937e..3d216973798 100644
--- a/drivers/media/video/tveeprom.c
+++ b/drivers/media/video/tveeprom.c
@@ -482,7 +482,6 @@ static unsigned short normal_i2c[] = {
 	0xa0 >> 1,
 	I2C_CLIENT_END,
 };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 I2C_CLIENT_INSMOD;
 
 struct i2c_driver i2c_driver_tveeprom;
diff --git a/drivers/media/video/vpx3220.c b/drivers/media/video/vpx3220.c
index 0fd6c9a7091..b97036910fa 100644
--- a/drivers/media/video/vpx3220.c
+++ b/drivers/media/video/vpx3220.c
@@ -569,21 +569,15 @@ static unsigned short normal_i2c[] =
     { I2C_VPX3220 >> 1, (I2C_VPX3220 >> 1) + 4,
 	I2C_CLIENT_END
 };
-static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 
 static unsigned short probe[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short probe_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short ignore[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
-static unsigned short ignore_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 static unsigned short force[2] = { I2C_CLIENT_END , I2C_CLIENT_END };
                                                                                 
 static struct i2c_client_address_data addr_data = {
 	.normal_i2c		= normal_i2c,
-	.normal_i2c_range	= normal_i2c_range,
 	.probe			= probe,
-	.probe_range		= probe_range,
 	.ignore			= ignore,
-	.ignore_range		= ignore_range,
 	.force			= force
 };
 
diff --git a/drivers/video/matrox/matroxfb_maven.c b/drivers/video/matrox/matroxfb_maven.c
index e529841cd83..67f85344f0c 100644
--- a/drivers/video/matrox/matroxfb_maven.c
+++ b/drivers/video/matrox/matroxfb_maven.c
@@ -1230,7 +1230,6 @@ static int maven_shutdown_client(struct i2c_client* clnt) {
 }
 
 static unsigned short normal_i2c[] = { MAVEN_I2CID, I2C_CLIENT_END };
-static unsigned short normal_i2c_range[] = { MAVEN_I2CID, MAVEN_I2CID, I2C_CLIENT_END };
 I2C_CLIENT_INSMOD;
 
 static struct i2c_driver maven_driver;
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index ebcd745f4cd..be837b13f29 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -290,11 +290,8 @@ static inline void i2c_set_adapdata (struct i2c_adapter *dev, void *data)
  */
 struct i2c_client_address_data {
 	unsigned short *normal_i2c;
-	unsigned short *normal_i2c_range;
 	unsigned short *probe;
-	unsigned short *probe_range;
 	unsigned short *ignore;
-	unsigned short *ignore_range;
 	unsigned short *force;
 };
 
@@ -563,24 +560,15 @@ union i2c_smbus_data {
 #define I2C_CLIENT_INSMOD \
   I2C_CLIENT_MODULE_PARM(probe, \
                       "List of adapter,address pairs to scan additionally"); \
-  I2C_CLIENT_MODULE_PARM(probe_range, \
-                      "List of adapter,start-addr,end-addr triples to scan " \
-                      "additionally"); \
   I2C_CLIENT_MODULE_PARM(ignore, \
                       "List of adapter,address pairs not to scan"); \
-  I2C_CLIENT_MODULE_PARM(ignore_range, \
-                      "List of adapter,start-addr,end-addr triples not to " \
-                      "scan"); \
   I2C_CLIENT_MODULE_PARM(force, \
                       "List of adapter,address pairs to boldly assume " \
                       "to be present"); \
 	static struct i2c_client_address_data addr_data = {		\
 			.normal_i2c = 		normal_i2c,		\
-			.normal_i2c_range =	normal_i2c_range,	\
 			.probe =		probe,			\
-			.probe_range =		probe_range,		\
 			.ignore =		ignore,			\
-			.ignore_range =		ignore_range,		\
 			.force =		force,			\
 		}
 
-- 
cgit v1.2.3-70-g09d2


From 3886246a257e828248ce1e72ced00408a3557f0d Mon Sep 17 00:00:00 2001
From: Sebastian Witt <se.witt@gmx.net>
Date: Wed, 13 Apr 2005 22:25:39 +0200
Subject: [PATCH] I2C: i2c-vid.h: Support for VID to reg conversion

Adds conversion from VID (mV) to register value. Used by the atxp1 I2C module.
Removed uneeded switch case.

Signed-off-by: Sebastian Witt <se.witt@gmx.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/i2c-vid.h | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'include')

diff --git a/include/linux/i2c-vid.h b/include/linux/i2c-vid.h
index 974835e3530..41d0635e0ba 100644
--- a/include/linux/i2c-vid.h
+++ b/include/linux/i2c-vid.h
@@ -97,3 +97,15 @@ static inline int vid_from_reg(int val, int vrm)
 		                     2050 - (val) * 50);
 	}
 }
+
+static inline int vid_to_reg(int val, int vrm)
+{
+	switch (vrm) {
+	case 91:		/* VRM 9.1 */
+	case 90:		/* VRM 9.0 */
+		return ((val >= 1100) && (val <= 1850) ?
+			((18499 - val * 10) / 25 + 5) / 10 : -1);
+	default:
+		return -1;
+	}
+}
-- 
cgit v1.2.3-70-g09d2


From 72cd799544f2b36c2f07ceaeed6d984cb130d4f3 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Tue, 24 May 2005 17:34:51 -0700
Subject: [PATCH] I2C: add i2c driver for TPS6501x

This adds an I2C driver for the TPS6501x series of power management chips.
It's used on many OMAP based boards, and this driver has been widely used
in the Linux-OMAP trees over the last year or so.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/chips/Kconfig            |   19 +-
 drivers/i2c/chips/Makefile           |    4 +-
 drivers/i2c/chips/tps65010.c         | 1072 ++++++++++++++++++++++++++++++++++
 include/asm-arm/arch-omap/tps65010.h |   76 +++
 4 files changed, 1169 insertions(+), 2 deletions(-)
 create mode 100644 drivers/i2c/chips/tps65010.c

(limited to 'include')

diff --git a/drivers/i2c/chips/Kconfig b/drivers/i2c/chips/Kconfig
index 4da7641975f..028a6feebea 100644
--- a/drivers/i2c/chips/Kconfig
+++ b/drivers/i2c/chips/Kconfig
@@ -1,5 +1,5 @@
 #
-# I2C Sensor device configuration
+# I2C Sensor and "other" chip configuration
 #
 
 menu "Hardware Sensors Chip support"
@@ -472,6 +472,23 @@ config ISP1301_OMAP
 	  This driver can also be built as a module.  If so, the module
 	  will be called isp1301_omap.
 
+# NOTE:  This isn't really OMAP-specific, except for the current
+# interface location in  <include/asm-arm/arch-omap/tps65010.h>
+# and having mostly OMAP-specific board support
+config TPS65010
+	tristate "TPS6501x Power Management chips"
+	depends on I2C && ARCH_OMAP
+	default y if MACH_OMAP_H2 || MACH_OMAP_H3 || MACH_OMAP_OSK
+	help
+	  If you say yes here you get support for the TPS6501x series of
+	  Power Management chips.  These include voltage regulators,
+	  lithium ion/polymer battery charging, and other features that
+	  are often used in portable devices like cell phones and cameras.
+
+	  This driver can also be built as a module.  If so, the module
+	  will be called tps65010.
+
+
 config SENSORS_M41T00
 	tristate "ST M41T00 RTC chip"
 	depends on I2C && PPC32
diff --git a/drivers/i2c/chips/Makefile b/drivers/i2c/chips/Makefile
index 2281435fc4c..5054ba5e470 100644
--- a/drivers/i2c/chips/Makefile
+++ b/drivers/i2c/chips/Makefile
@@ -1,5 +1,5 @@
 #
-# Makefile for the kernel hardware sensors chip drivers.
+# Makefile for sensor and "other" I2C chip drivers.
 #
 
 # asb100, then w83781d go first, as they can override other drivers' addresses.
@@ -43,7 +43,9 @@ obj-$(CONFIG_SENSORS_SMSC47M1)	+= smsc47m1.o
 obj-$(CONFIG_SENSORS_VIA686A)	+= via686a.o
 obj-$(CONFIG_SENSORS_W83627EHF)	+= w83627ehf.o
 obj-$(CONFIG_SENSORS_W83L785TS)	+= w83l785ts.o
+
 obj-$(CONFIG_ISP1301_OMAP)	+= isp1301_omap.o
+obj-$(CONFIG_TPS65010)		+= tps65010.o
 
 ifeq ($(CONFIG_I2C_DEBUG_CHIP),y)
 EXTRA_CFLAGS += -DDEBUG
diff --git a/drivers/i2c/chips/tps65010.c b/drivers/i2c/chips/tps65010.c
new file mode 100644
index 00000000000..c0ac01b6003
--- /dev/null
+++ b/drivers/i2c/chips/tps65010.c
@@ -0,0 +1,1072 @@
+/*
+ * tps65010 - driver for tps6501x power management chips
+ *
+ * Copyright (C) 2004 Texas Instruments
+ * Copyright (C) 2004-2005 David Brownell
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+#undef	DEBUG
+
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/interrupt.h>
+#include <linux/device.h>
+#include <linux/i2c.h>
+#include <linux/delay.h>
+#include <linux/workqueue.h>
+#include <linux/suspend.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+
+#include <asm/irq.h>
+#include <asm/mach-types.h>
+
+#include <asm/arch/gpio.h>
+#include <asm/arch/mux.h>
+#include <asm/arch/tps65010.h>
+
+/*-------------------------------------------------------------------------*/
+
+#define	DRIVER_VERSION	"2 May 2005"
+#define	DRIVER_NAME	(tps65010_driver.name)
+
+MODULE_DESCRIPTION("TPS6501x Power Management Driver");
+MODULE_LICENSE("GPL");
+
+/* only two addresses possible */
+#define	TPS_BASE	0x48
+static unsigned short normal_i2c[] = {
+	TPS_BASE,
+	I2C_CLIENT_END };
+static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
+
+I2C_CLIENT_INSMOD;
+
+static struct i2c_driver tps65010_driver;
+
+/*-------------------------------------------------------------------------*/
+
+/* This driver handles a family of multipurpose chips, which incorporate
+ * voltage regulators, lithium ion/polymer battery charging, GPIOs, LEDs,
+ * and other features often needed in portable devices like cell phones
+ * or digital cameras.
+ *
+ * The tps65011 and tps65013 have different voltage settings compared
+ * to tps65010 and tps65012.  The tps65013 has a NO_CHG status/irq.
+ * All except tps65010 have "wait" mode, possibly defaulted so that
+ * battery-insert != device-on.
+ *
+ * We could distinguish between some models by checking VDCDC1.UVLO or
+ * other registers, unless they've been changed already after powerup
+ * as part of board setup by a bootloader.
+ */
+enum tps_model {
+	TPS_UNKNOWN = 0,
+	TPS65010,
+	TPS65011,
+	TPS65012,
+	TPS65013,
+};
+
+struct tps65010 {
+	struct i2c_client	client;
+	struct semaphore	lock;
+	int			irq;
+	struct work_struct	work;
+	struct dentry		*file;
+	unsigned		charging:1;
+	unsigned		por:1;
+	unsigned		model:8;
+	u16			vbus;
+	unsigned long		flags;
+#define	FLAG_VBUS_CHANGED	0
+#define	FLAG_IRQ_ENABLE		1
+
+	/* copies of last register state */
+	u8			chgstatus, regstatus, chgconf;
+	u8			nmask1, nmask2;
+
+	/* plus four GPIOs, probably used to switch power */
+};
+
+#define	POWER_POLL_DELAY	msecs_to_jiffies(800)
+
+/*-------------------------------------------------------------------------*/
+
+#if	defined(DEBUG) || defined(CONFIG_DEBUG_FS)
+
+static void dbg_chgstat(char *buf, size_t len, u8 chgstatus)
+{
+	snprintf(buf, len, "%02x%s%s%s%s%s%s%s%s\n",
+		chgstatus,
+		(chgstatus & TPS_CHG_USB) ? " USB" : "",
+		(chgstatus & TPS_CHG_AC) ? " AC" : "",
+		(chgstatus & TPS_CHG_THERM) ? " therm" : "",
+		(chgstatus & TPS_CHG_TERM) ? " done" :
+			((chgstatus & (TPS_CHG_USB|TPS_CHG_AC))
+				? " (charging)" : ""),
+		(chgstatus & TPS_CHG_TAPER_TMO) ? " taper_tmo" : "",
+		(chgstatus & TPS_CHG_CHG_TMO) ? " charge_tmo" : "",
+		(chgstatus & TPS_CHG_PRECHG_TMO) ? " prechg_tmo" : "",
+		(chgstatus & TPS_CHG_TEMP_ERR) ? " temp_err" : "");
+}
+
+static void dbg_regstat(char *buf, size_t len, u8 regstatus)
+{
+	snprintf(buf, len, "%02x %s%s%s%s%s%s%s%s\n",
+		regstatus,
+		(regstatus & TPS_REG_ONOFF) ? "off" : "(on)",
+		(regstatus & TPS_REG_COVER) ? " uncover" : "",
+		(regstatus & TPS_REG_UVLO) ? " UVLO" : "",
+		(regstatus & TPS_REG_NO_CHG) ? " NO_CHG" : "",
+		(regstatus & TPS_REG_PG_LD02) ? " ld01_bad" : "",
+		(regstatus & TPS_REG_PG_LD01) ? " ld01_bad" : "",
+		(regstatus & TPS_REG_PG_MAIN) ? " main_bad" : "",
+		(regstatus & TPS_REG_PG_CORE) ? " core_bad" : "");
+}
+
+static void dbg_chgconf(int por, char *buf, size_t len, u8 chgconfig)
+{
+	char *hibit;
+
+	if (por)
+		hibit = (chgconfig & TPS_CHARGE_POR)
+				? "POR=69ms" : "POR=1sec";
+	else
+		hibit = (chgconfig & TPS65013_AUA) ? "AUA" : "";
+
+	snprintf(buf, len, "%02x %s%s%s AC=%d%% USB=%dmA %sCharge\n",
+		chgconfig, hibit,
+		(chgconfig & TPS_CHARGE_RESET) ? " reset" : "",
+		(chgconfig & TPS_CHARGE_FAST) ? " fast" : "",
+		({int p; switch ((chgconfig >> 3) & 3) {
+		case 3:		p = 100; break;
+		case 2:		p = 75; break;
+		case 1:		p = 50; break;
+		default:	p = 25; break;
+		}; p; }),
+		(chgconfig & TPS_VBUS_CHARGING)
+			? ((chgconfig & TPS_VBUS_500MA) ? 500 : 100)
+			: 0,
+		(chgconfig & TPS_CHARGE_ENABLE) ? "" : "No");
+}
+
+#endif
+
+#ifdef	DEBUG
+
+static void show_chgstatus(const char *label, u8 chgstatus)
+{
+	char buf [100];
+
+	dbg_chgstat(buf, sizeof buf, chgstatus);
+	pr_debug("%s: %s %s", DRIVER_NAME, label, buf);
+}
+
+static void show_regstatus(const char *label, u8 regstatus)
+{
+	char buf [100];
+
+	dbg_regstat(buf, sizeof buf, regstatus);
+	pr_debug("%s: %s %s", DRIVER_NAME, label, buf);
+}
+
+static void show_chgconfig(int por, const char *label, u8 chgconfig)
+{
+	char buf [100];
+
+	dbg_chgconf(por, buf, sizeof buf, chgconfig);
+	pr_debug("%s: %s %s", DRIVER_NAME, label, buf);
+}
+
+#else
+
+static inline void show_chgstatus(const char *label, u8 chgstatus) { }
+static inline void show_regstatus(const char *label, u8 chgstatus) { }
+static inline void show_chgconfig(int por, const char *label, u8 chgconfig) { }
+
+#endif
+
+#ifdef	CONFIG_DEBUG_FS
+
+static int dbg_show(struct seq_file *s, void *_)
+{
+	struct tps65010	*tps = s->private;
+	u8		value, v2;
+	unsigned	i;
+	char		buf[100];
+	const char	*chip;
+
+	switch (tps->model) {
+	case TPS65010:	chip = "tps65010"; break;
+	case TPS65011:	chip = "tps65011"; break;
+	case TPS65012:	chip = "tps65012"; break;
+	case TPS65013:	chip = "tps65013"; break;
+	default:	chip = NULL; break;
+	}
+	seq_printf(s, "driver  %s\nversion %s\nchip    %s\n\n",
+			DRIVER_NAME, DRIVER_VERSION, chip);
+
+	down(&tps->lock);
+
+	/* FIXME how can we tell whether a battery is present?
+	 * likely involves a charge gauging chip (like BQ26501).
+	 */
+
+	seq_printf(s, "%scharging\n\n", tps->charging ? "" : "(not) ");
+
+
+	/* registers for monitoring battery charging and status; note
+	 * that reading chgstat and regstat may ack IRQs...
+	 */
+	value = i2c_smbus_read_byte_data(&tps->client, TPS_CHGCONFIG);
+	dbg_chgconf(tps->por, buf, sizeof buf, value);
+	seq_printf(s, "chgconfig %s", buf);
+
+	value = i2c_smbus_read_byte_data(&tps->client, TPS_CHGSTATUS);
+	dbg_chgstat(buf, sizeof buf, value);
+	seq_printf(s, "chgstat   %s", buf);
+	value = i2c_smbus_read_byte_data(&tps->client, TPS_MASK1);
+	dbg_chgstat(buf, sizeof buf, value);
+	seq_printf(s, "mask1     %s", buf);
+	/* ignore ackint1 */
+
+	value = i2c_smbus_read_byte_data(&tps->client, TPS_REGSTATUS);
+	dbg_regstat(buf, sizeof buf, value);
+	seq_printf(s, "regstat   %s", buf);
+	value = i2c_smbus_read_byte_data(&tps->client, TPS_MASK2);
+	dbg_regstat(buf, sizeof buf, value);
+	seq_printf(s, "mask2     %s\n", buf);
+	/* ignore ackint2 */
+
+	(void) schedule_delayed_work(&tps->work, POWER_POLL_DELAY);
+
+
+	/* VMAIN voltage, enable lowpower, etc */
+	value = i2c_smbus_read_byte_data(&tps->client, TPS_VDCDC1);
+	seq_printf(s, "vdcdc1    %02x\n", value);
+
+	/* VCORE voltage, vibrator on/off */
+	value = i2c_smbus_read_byte_data(&tps->client, TPS_VDCDC2);
+	seq_printf(s, "vdcdc2    %02x\n", value);
+
+	/* both LD0s, and their lowpower behavior */
+	value = i2c_smbus_read_byte_data(&tps->client, TPS_VREGS1);
+	seq_printf(s, "vregs1    %02x\n\n", value);
+
+
+	/* LEDs and GPIOs */
+	value = i2c_smbus_read_byte_data(&tps->client, TPS_LED1_ON);
+	v2 = i2c_smbus_read_byte_data(&tps->client, TPS_LED1_PER);
+	seq_printf(s, "led1 %s, on=%02x, per=%02x, %d/%d msec\n",
+		(value & 0x80)
+			? ((v2 & 0x80) ? "on" : "off")
+			: ((v2 & 0x80) ? "blink" : "(nPG)"),
+		value, v2,
+		(value & 0x7f) * 10, (v2 & 0x7f) * 100);
+
+	value = i2c_smbus_read_byte_data(&tps->client, TPS_LED2_ON);
+	v2 = i2c_smbus_read_byte_data(&tps->client, TPS_LED2_PER);
+	seq_printf(s, "led2 %s, on=%02x, per=%02x, %d/%d msec\n",
+		(value & 0x80)
+			? ((v2 & 0x80) ? "on" : "off")
+			: ((v2 & 0x80) ? "blink" : "off"),
+		value, v2,
+		(value & 0x7f) * 10, (v2 & 0x7f) * 100);
+
+	value = i2c_smbus_read_byte_data(&tps->client, TPS_DEFGPIO);
+	v2 = i2c_smbus_read_byte_data(&tps->client, TPS_MASK3);
+	seq_printf(s, "defgpio %02x mask3 %02x\n", value, v2);
+
+	for (i = 0; i < 4; i++) {
+		if (value & (1 << (4 +i)))
+			seq_printf(s, "  gpio%d-out %s\n", i + 1,
+				(value & (1 << i)) ? "low" : "hi ");
+		else
+			seq_printf(s, "  gpio%d-in  %s %s %s\n", i + 1,
+				(value & (1 << i)) ? "hi " : "low",
+				(v2 & (1 << i)) ? "no-irq" : "irq",
+				(v2 & (1 << (4 + i))) ? "rising" : "falling");
+	}
+
+	up(&tps->lock);
+	return 0;
+}
+
+static int dbg_tps_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, dbg_show, inode->u.generic_ip);
+}
+
+static struct file_operations debug_fops = {
+	.open		= dbg_tps_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+#define	DEBUG_FOPS	&debug_fops
+
+#else
+#define	DEBUG_FOPS	NULL
+#endif
+
+/*-------------------------------------------------------------------------*/
+
+/* handle IRQS in a task context, so we can use I2C calls */
+static void tps65010_interrupt(struct tps65010 *tps)
+{
+	u8 tmp = 0, mask, poll;
+
+	/* IRQs won't trigger irqs for certain events, but we can get
+	 * others by polling (normally, with external power applied).
+	 */
+	poll = 0;
+
+	/* regstatus irqs */
+	if (tps->nmask2) {
+		tmp = i2c_smbus_read_byte_data(&tps->client, TPS_REGSTATUS);
+		mask = tmp ^ tps->regstatus;
+		tps->regstatus = tmp;
+		mask &= tps->nmask2;
+	} else
+		mask = 0;
+	if (mask) {
+		tps->regstatus =  tmp;
+		/* may need to shut something down ... */
+
+		/* "off" usually means deep sleep */
+		if (tmp & TPS_REG_ONOFF) {
+			pr_info("%s: power off button\n", DRIVER_NAME);
+#if 0
+			/* REVISIT:  this might need its own workqueue
+			 * plus tweaks including deadlock avoidance ...
+			 */
+			software_suspend();
+#endif
+			poll = 1;
+		}
+	}
+
+	/* chgstatus irqs */
+	if (tps->nmask1) {
+		tmp = i2c_smbus_read_byte_data(&tps->client, TPS_CHGSTATUS);
+		mask = tmp ^ tps->chgstatus;
+		tps->chgstatus = tmp;
+		mask &= tps->nmask1;
+	} else
+		mask = 0;
+	if (mask) {
+		unsigned	charging = 0;
+
+		show_chgstatus("chg/irq", tmp);
+		if (tmp & (TPS_CHG_USB|TPS_CHG_AC))
+			show_chgconfig(tps->por, "conf", tps->chgconf);
+
+		/* Unless it was turned off or disabled, we charge any
+		 * battery whenever there's power available for it
+		 * and the charger hasn't been disabled.
+		 */
+		if (!(tps->chgstatus & ~(TPS_CHG_USB|TPS_CHG_AC))
+				&& (tps->chgstatus & (TPS_CHG_USB|TPS_CHG_AC))
+				&& (tps->chgconf & TPS_CHARGE_ENABLE)
+				) {
+			if (tps->chgstatus & TPS_CHG_USB) {
+				/* VBUS options are readonly until reconnect */
+				if (mask & TPS_CHG_USB)
+					set_bit(FLAG_VBUS_CHANGED, &tps->flags);
+				charging = 1;
+			} else if (tps->chgstatus & TPS_CHG_AC)
+				charging = 1;
+		}
+		if (charging != tps->charging) {
+			tps->charging = charging;
+			pr_info("%s: battery %scharging\n",
+				DRIVER_NAME, charging ? "" :
+				((tps->chgstatus & (TPS_CHG_USB|TPS_CHG_AC))
+					? "NOT " : "dis"));
+		}
+	}
+
+	/* always poll to detect (a) power removal, without tps65013
+	 * NO_CHG IRQ; or (b) restart of charging after stop.
+	 */
+	if ((tps->model != TPS65013 || !tps->charging)
+			&& (tps->chgstatus & (TPS_CHG_USB|TPS_CHG_AC)))
+		poll = 1;
+	if (poll)
+		(void) schedule_delayed_work(&tps->work, POWER_POLL_DELAY);
+
+	/* also potentially gpio-in rise or fall */
+}
+
+/* handle IRQs and polling using keventd for now */
+static void tps65010_work(void *_tps)
+{
+	struct tps65010		*tps = _tps;
+
+	down(&tps->lock);
+
+	tps65010_interrupt(tps);
+
+	if (test_and_clear_bit(FLAG_VBUS_CHANGED, &tps->flags)) {
+		int	status;
+		u8	chgconfig, tmp;
+
+		chgconfig = i2c_smbus_read_byte_data(&tps->client,
+					TPS_CHGCONFIG);
+		chgconfig &= ~(TPS_VBUS_500MA | TPS_VBUS_CHARGING);
+		if (tps->vbus == 500)
+			chgconfig |= TPS_VBUS_500MA | TPS_VBUS_CHARGING;
+		else if (tps->vbus >= 100)
+			chgconfig |= TPS_VBUS_CHARGING;
+
+		status = i2c_smbus_write_byte_data(&tps->client,
+				TPS_CHGCONFIG, chgconfig);
+
+		/* vbus update fails unless VBUS is connected! */
+		tmp = i2c_smbus_read_byte_data(&tps->client, TPS_CHGCONFIG);
+		tps->chgconf = tmp;
+		show_chgconfig(tps->por, "update vbus", tmp);
+	}
+
+	if (test_and_clear_bit(FLAG_IRQ_ENABLE, &tps->flags))
+		enable_irq(tps->irq);
+
+	up(&tps->lock);
+}
+
+static irqreturn_t tps65010_irq(int irq, void *_tps, struct pt_regs *regs)
+{
+	struct tps65010		*tps = _tps;
+
+	disable_irq_nosync(irq);
+	set_bit(FLAG_IRQ_ENABLE, &tps->flags);
+	(void) schedule_work(&tps->work);
+	return IRQ_HANDLED;
+}
+
+/*-------------------------------------------------------------------------*/
+
+static struct tps65010 *the_tps;
+
+static int __exit tps65010_detach_client(struct i2c_client *client)
+{
+	struct tps65010		*tps;
+
+	tps = container_of(client, struct tps65010, client);
+#ifdef	CONFIG_ARM
+	if (machine_is_omap_h2())
+		omap_free_gpio(58);
+	if (machine_is_omap_osk())
+		omap_free_gpio(OMAP_MPUIO(1));
+#endif
+	free_irq(tps->irq, tps);
+	debugfs_remove(tps->file);
+	if (i2c_detach_client(client) == 0)
+		kfree(tps);
+	the_tps = 0;
+	return 0;
+}
+
+static int tps65010_noscan(struct i2c_adapter *bus)
+{
+	/* pure paranoia, in case someone adds another i2c bus
+	 * after our init section's gone...
+	 */
+	return -ENODEV;
+}
+
+/* no error returns, they'd just make bus scanning stop */
+static int __init
+tps65010_probe(struct i2c_adapter *bus, int address, int kind)
+{
+	struct tps65010		*tps;
+	int			status;
+
+	if (the_tps) {
+		dev_dbg(&bus->dev, "only one %s for now\n", DRIVER_NAME);
+		return 0;
+	}
+
+	tps = kmalloc(sizeof *tps, GFP_KERNEL);
+	if (!tps)
+		return 0;
+
+	memset(tps, 0, sizeof *tps);
+	init_MUTEX(&tps->lock);
+	INIT_WORK(&tps->work, tps65010_work, tps);
+	tps->irq = -1;
+	tps->client.addr = address;
+	i2c_set_clientdata(&tps->client, tps);
+	tps->client.adapter = bus;
+	tps->client.driver = &tps65010_driver;
+	strlcpy(tps->client.name, DRIVER_NAME, I2C_NAME_SIZE);
+
+	status = i2c_attach_client(&tps->client);
+	if (status < 0) {
+		dev_dbg(&bus->dev, "can't attach %s to device %d, err %d\n",
+				DRIVER_NAME, address, status);
+fail1:
+		kfree(tps);
+		return 0;
+	}
+
+#ifdef	CONFIG_ARM
+	if (machine_is_omap_h2()) {
+		tps->model = TPS65010;
+		omap_cfg_reg(W4_GPIO58);
+		tps->irq = OMAP_GPIO_IRQ(58);
+		omap_request_gpio(58);
+		omap_set_gpio_direction(58, 1);
+		omap_set_gpio_edge_ctrl(58, OMAP_GPIO_FALLING_EDGE);
+	}
+	if (machine_is_omap_osk()) {
+		tps->model = TPS65010;
+		// omap_cfg_reg(U19_1610_MPUIO1);
+		tps->irq = OMAP_GPIO_IRQ(OMAP_MPUIO(1));
+		omap_request_gpio(OMAP_MPUIO(1));
+		omap_set_gpio_direction(OMAP_MPUIO(1), 1);
+		omap_set_gpio_edge_ctrl(OMAP_MPUIO(1), OMAP_GPIO_FALLING_EDGE);
+	}
+	if (machine_is_omap_h3()) {
+		tps->model = TPS65013;
+
+		// FIXME set up this board's IRQ ...
+	}
+#else
+#define set_irq_type(num,trigger)	do{}while(0)
+#endif
+
+	if (tps->irq > 0) {
+		set_irq_type(tps->irq, IRQT_LOW);
+		status = request_irq(tps->irq, tps65010_irq,
+			SA_SAMPLE_RANDOM, DRIVER_NAME, tps);
+		if (status < 0) {
+			dev_dbg(&tps->client.dev, "can't get IRQ %d, err %d\n",
+					tps->irq, status);
+			i2c_detach_client(&tps->client);
+			goto fail1;
+		}
+#ifdef	CONFIG_ARM
+		/* annoying race here, ideally we'd have an option
+		 * to claim the irq now and enable it later.
+		 */
+		disable_irq(tps->irq);
+		set_bit(FLAG_IRQ_ENABLE, &tps->flags);
+#endif
+	} else
+		printk(KERN_WARNING "%s: IRQ not configured!\n",
+				DRIVER_NAME);
+
+
+	switch (tps->model) {
+	case TPS65010:
+	case TPS65012:
+		tps->por = 1;
+		break;
+	case TPS_UNKNOWN:
+		printk(KERN_WARNING "%s: unknown TPS chip\n", DRIVER_NAME);
+		break;
+	/* else CHGCONFIG.POR is replaced by AUA, enabling a WAIT mode */
+	}
+	tps->chgconf = i2c_smbus_read_byte_data(&tps->client, TPS_CHGCONFIG);
+	show_chgconfig(tps->por, "conf/init", tps->chgconf);
+
+	show_chgstatus("chg/init",
+		i2c_smbus_read_byte_data(&tps->client, TPS_CHGSTATUS));
+	show_regstatus("reg/init",
+		i2c_smbus_read_byte_data(&tps->client, TPS_REGSTATUS));
+
+	pr_debug("%s: vdcdc1 0x%02x, vdcdc2 %02x, vregs1 %02x\n", DRIVER_NAME,
+		i2c_smbus_read_byte_data(&tps->client, TPS_VDCDC1),
+		i2c_smbus_read_byte_data(&tps->client, TPS_VDCDC2),
+		i2c_smbus_read_byte_data(&tps->client, TPS_VREGS1));
+	pr_debug("%s: defgpio 0x%02x, mask3 0x%02x\n", DRIVER_NAME,
+		i2c_smbus_read_byte_data(&tps->client, TPS_DEFGPIO),
+		i2c_smbus_read_byte_data(&tps->client, TPS_MASK3));
+
+	tps65010_driver.attach_adapter = tps65010_noscan;
+	the_tps = tps;
+
+#if	defined(CONFIG_USB_GADGET) && !defined(CONFIG_USB_OTG)
+	/* USB hosts can't draw VBUS.  OTG devices could, later
+	 * when OTG infrastructure enables it.  USB peripherals
+	 * could be relying on VBUS while booting, though.
+	 */
+	tps->vbus = 100;
+#endif
+
+	/* unmask the "interesting" irqs, then poll once to
+	 * kickstart monitoring, initialize shadowed status
+	 * registers, and maybe disable VBUS draw.
+	 */
+	tps->nmask1 = ~0;
+	(void) i2c_smbus_write_byte_data(&tps->client, TPS_MASK1, ~tps->nmask1);
+
+	tps->nmask2 = TPS_REG_ONOFF;
+	if (tps->model == TPS65013)
+		tps->nmask2 |= TPS_REG_NO_CHG;
+	(void) i2c_smbus_write_byte_data(&tps->client, TPS_MASK2, ~tps->nmask2);
+
+	(void) i2c_smbus_write_byte_data(&tps->client, TPS_MASK3, 0x0f
+		| i2c_smbus_read_byte_data(&tps->client, TPS_MASK3));
+
+	tps65010_work(tps);
+
+	tps->file = debugfs_create_file(DRIVER_NAME, S_IRUGO, NULL,
+				tps, DEBUG_FOPS);
+	return 0;
+}
+
+static int __init tps65010_scan_bus(struct i2c_adapter *bus)
+{
+	if (!i2c_check_functionality(bus, I2C_FUNC_SMBUS_BYTE_DATA))
+		return -EINVAL;
+	return i2c_probe(bus, &addr_data, tps65010_probe);
+}
+
+static struct i2c_driver tps65010_driver = {
+	.owner		= THIS_MODULE,
+	.name		= "tps65010",
+	.id		= 888,		/* FIXME assign "official" value */
+	.flags		= I2C_DF_NOTIFY,
+	.attach_adapter	= tps65010_scan_bus,
+	.detach_client	= __exit_p(tps65010_detach_client),
+};
+
+/*-------------------------------------------------------------------------*/
+
+/* Draw from VBUS:
+ *   0 mA -- DON'T DRAW (might supply power instead)
+ * 100 mA -- usb unit load (slowest charge rate)
+ * 500 mA -- usb high power (fast battery charge)
+ */
+int tps65010_set_vbus_draw(unsigned mA)
+{
+	unsigned long	flags;
+
+	if (!the_tps)
+		return -ENODEV;
+
+	/* assumes non-SMP */
+	local_irq_save(flags);
+	if (mA >= 500)
+		mA = 500;
+	else if (mA >= 100)
+		mA = 100;
+	else
+		mA = 0;
+	the_tps->vbus = mA;
+	if ((the_tps->chgstatus & TPS_CHG_USB)
+			&& test_and_set_bit(
+				FLAG_VBUS_CHANGED, &the_tps->flags)) {
+		/* gadget drivers call this in_irq() */
+		(void) schedule_work(&the_tps->work);
+	}
+	local_irq_restore(flags);
+
+	return 0;
+}
+EXPORT_SYMBOL(tps65010_set_vbus_draw);
+
+/*-------------------------------------------------------------------------*/
+/* tps65010_set_gpio_out_value parameter:
+ * gpio:  GPIO1, GPIO2, GPIO3 or GPIO4
+ * value: LOW or HIGH
+ */
+int tps65010_set_gpio_out_value(unsigned gpio, unsigned value)
+{
+	int	 status;
+	unsigned defgpio;
+
+	if (!the_tps)
+		return -ENODEV;
+	if ((gpio < GPIO1) || (gpio > GPIO4))
+		return -EINVAL;
+
+	down(&the_tps->lock);
+
+	defgpio = i2c_smbus_read_byte_data(&the_tps->client, TPS_DEFGPIO);
+
+	/* Configure GPIO for output */
+	defgpio |= 1 << (gpio + 3);
+
+	/* Writing 1 forces a logic 0 on that GPIO and vice versa */
+	switch (value) {
+	case LOW:
+		defgpio |= 1 << (gpio - 1);    /* set GPIO low by writing 1 */
+		break;
+	/* case HIGH: */
+	default:
+		defgpio &= ~(1 << (gpio - 1)); /* set GPIO high by writing 0 */
+		break;
+	}
+
+	status = i2c_smbus_write_byte_data(&the_tps->client,
+		TPS_DEFGPIO, defgpio);
+
+	pr_debug("%s: gpio%dout = %s, defgpio 0x%02x\n", DRIVER_NAME,
+		gpio, value ? "high" : "low",
+		i2c_smbus_read_byte_data(&the_tps->client, TPS_DEFGPIO));
+
+	up(&the_tps->lock);
+	return status;
+}
+EXPORT_SYMBOL(tps65010_set_gpio_out_value);
+
+/*-------------------------------------------------------------------------*/
+/* tps65010_set_led parameter:
+ * led:  LED1 or LED2
+ * mode: ON, OFF or BLINK
+ */
+int tps65010_set_led(unsigned led, unsigned mode)
+{
+	int	 status;
+	unsigned led_on, led_per, offs;
+
+	if (!the_tps)
+		return -ENODEV;
+
+	if(led == LED1)
+		offs = 0;
+	else {
+		offs = 2;
+		led = LED2;
+	}
+
+	down(&the_tps->lock);
+
+	dev_dbg (&the_tps->client.dev, "led%i_on   0x%02x\n", led,
+		i2c_smbus_read_byte_data(&the_tps->client, TPS_LED1_ON + offs));
+
+	dev_dbg (&the_tps->client.dev, "led%i_per  0x%02x\n", led,
+		i2c_smbus_read_byte_data(&the_tps->client, TPS_LED1_PER + offs));
+
+	switch (mode) {
+	case OFF:
+		led_on  = 1 << 7;
+		led_per = 0 << 7;
+		break;
+	case ON:
+		led_on  = 1 << 7;
+		led_per = 1 << 7;
+		break;
+	case BLINK:
+		led_on  = 0x30 | (0 << 7);
+		led_per = 0x08 | (1 << 7);
+		break;
+	default:
+		printk(KERN_ERR "%s: Wrong mode parameter for tps65010_set_led()\n",
+		       DRIVER_NAME);
+		up(&the_tps->lock);
+		return -EINVAL;
+	}
+
+	status = i2c_smbus_write_byte_data(&the_tps->client,
+			TPS_LED1_ON + offs, led_on);
+
+	if (status != 0) {
+		printk(KERN_ERR "%s: Failed to write led%i_on register\n",
+		       DRIVER_NAME, led);
+		up(&the_tps->lock);
+		return status;
+	}
+
+	dev_dbg (&the_tps->client.dev, "led%i_on   0x%02x\n", led,
+		i2c_smbus_read_byte_data(&the_tps->client, TPS_LED1_ON + offs));
+
+	status = i2c_smbus_write_byte_data(&the_tps->client,
+			TPS_LED1_PER + offs, led_per);
+
+	if (status != 0) {
+		printk(KERN_ERR "%s: Failed to write led%i_per register\n",
+		       DRIVER_NAME, led);
+		up(&the_tps->lock);
+		return status;
+	}
+
+	dev_dbg (&the_tps->client.dev, "led%i_per  0x%02x\n", led,
+		i2c_smbus_read_byte_data(&the_tps->client, TPS_LED1_PER + offs));
+
+	up(&the_tps->lock);
+
+	return status;
+}
+EXPORT_SYMBOL(tps65010_set_led);
+
+/*-------------------------------------------------------------------------*/
+/* tps65010_set_vib parameter:
+ * value: ON or OFF
+ */
+int tps65010_set_vib(unsigned value)
+{
+	int	 status;
+	unsigned vdcdc2;
+
+	if (!the_tps)
+		return -ENODEV;
+
+	down(&the_tps->lock);
+
+	vdcdc2 = i2c_smbus_read_byte_data(&the_tps->client, TPS_VDCDC2);
+	vdcdc2 &= ~(1 << 1);
+	if (value)
+		vdcdc2 |= (1 << 1);
+	status = i2c_smbus_write_byte_data(&the_tps->client,
+		TPS_VDCDC2, vdcdc2);
+
+	pr_debug("%s: vibrator %s\n", DRIVER_NAME, value ? "on" : "off");
+
+	up(&the_tps->lock);
+	return status;
+}
+EXPORT_SYMBOL(tps65010_set_vib);
+
+/*-------------------------------------------------------------------------*/
+/* tps65010_set_low_pwr parameter:
+ * mode: ON or OFF
+ */
+int tps65010_set_low_pwr(unsigned mode)
+{
+	int	 status;
+	unsigned vdcdc1;
+
+	if (!the_tps)
+		return -ENODEV;
+
+	down(&the_tps->lock);
+
+	pr_debug("%s: %s low_pwr, vdcdc1 0x%02x\n", DRIVER_NAME,
+		mode ? "enable" : "disable",
+		i2c_smbus_read_byte_data(&the_tps->client, TPS_VDCDC1));
+
+	vdcdc1 = i2c_smbus_read_byte_data(&the_tps->client, TPS_VDCDC1);
+
+	switch (mode) {
+	case OFF:
+		vdcdc1 &= ~TPS_ENABLE_LP; /* disable ENABLE_LP bit */
+		break;
+	/* case ON: */
+	default:
+		vdcdc1 |= TPS_ENABLE_LP;  /* enable ENABLE_LP bit */
+		break;
+	}
+
+	status = i2c_smbus_write_byte_data(&the_tps->client,
+			TPS_VDCDC1, vdcdc1);
+
+	if (status != 0)
+		printk(KERN_ERR "%s: Failed to write vdcdc1 register\n",
+		       DRIVER_NAME);
+	else
+		pr_debug("%s: vdcdc1 0x%02x\n", DRIVER_NAME,
+			i2c_smbus_read_byte_data(&the_tps->client, TPS_VDCDC1));
+
+	up(&the_tps->lock);
+
+	return status;
+}
+EXPORT_SYMBOL(tps65010_set_low_pwr);
+
+/*-------------------------------------------------------------------------*/
+/* tps65010_config_vregs1 parameter:
+ * value to be written to VREGS1 register
+ * Note: The complete register is written, set all bits you need
+ */
+int tps65010_config_vregs1(unsigned value)
+{
+	int	 status;
+
+	if (!the_tps)
+		return -ENODEV;
+
+	down(&the_tps->lock);
+
+	pr_debug("%s: vregs1 0x%02x\n", DRIVER_NAME,
+		        i2c_smbus_read_byte_data(&the_tps->client, TPS_VREGS1));
+
+	status = i2c_smbus_write_byte_data(&the_tps->client,
+			TPS_VREGS1, value);
+
+	if (status != 0)
+		printk(KERN_ERR "%s: Failed to write vregs1 register\n",
+		        DRIVER_NAME);
+	else
+		pr_debug("%s: vregs1 0x%02x\n", DRIVER_NAME,
+			i2c_smbus_read_byte_data(&the_tps->client, TPS_VREGS1));
+
+	up(&the_tps->lock);
+
+	return status;
+}
+EXPORT_SYMBOL(tps65010_config_vregs1);
+
+/*-------------------------------------------------------------------------*/
+/* tps65013_set_low_pwr parameter:
+ * mode: ON or OFF
+ */
+
+/* FIXME: Assumes AC or USB power is present. Setting AUA bit is not
+	required if power supply is through a battery */
+
+int tps65013_set_low_pwr(unsigned mode)
+{
+	int	 status;
+	unsigned vdcdc1, chgconfig;
+
+	if (!the_tps || the_tps->por)
+		return -ENODEV;
+
+	down(&the_tps->lock);
+
+	pr_debug("%s: %s low_pwr, chgconfig 0x%02x vdcdc1 0x%02x\n",
+		DRIVER_NAME,
+		mode ? "enable" : "disable",
+		i2c_smbus_read_byte_data(&the_tps->client, TPS_CHGCONFIG),
+		i2c_smbus_read_byte_data(&the_tps->client, TPS_VDCDC1));
+
+	chgconfig = i2c_smbus_read_byte_data(&the_tps->client, TPS_CHGCONFIG);
+	vdcdc1 = i2c_smbus_read_byte_data(&the_tps->client, TPS_VDCDC1);
+
+	switch (mode) {
+	case OFF:
+		chgconfig &= ~TPS65013_AUA; /* disable AUA bit */
+		vdcdc1 &= ~TPS_ENABLE_LP; /* disable ENABLE_LP bit */
+		break;
+	/* case ON: */
+	default:
+		chgconfig |= TPS65013_AUA;  /* enable AUA bit */
+		vdcdc1 |= TPS_ENABLE_LP;  /* enable ENABLE_LP bit */
+		break;
+	}
+
+	status = i2c_smbus_write_byte_data(&the_tps->client,
+			TPS_CHGCONFIG, chgconfig);
+	if (status != 0) {
+		printk(KERN_ERR "%s: Failed to write chconfig register\n",
+	 DRIVER_NAME);
+		up(&the_tps->lock);
+		return status;
+	}
+
+	chgconfig = i2c_smbus_read_byte_data(&the_tps->client, TPS_CHGCONFIG);
+	the_tps->chgconf = chgconfig;
+	show_chgconfig(0, "chgconf", chgconfig);
+
+	status = i2c_smbus_write_byte_data(&the_tps->client,
+			TPS_VDCDC1, vdcdc1);
+
+	if (status != 0)
+		printk(KERN_ERR "%s: Failed to write vdcdc1 register\n",
+	 DRIVER_NAME);
+	else
+		pr_debug("%s: vdcdc1 0x%02x\n", DRIVER_NAME,
+			i2c_smbus_read_byte_data(&the_tps->client, TPS_VDCDC1));
+
+	up(&the_tps->lock);
+
+	return status;
+}
+EXPORT_SYMBOL(tps65013_set_low_pwr);
+
+/*-------------------------------------------------------------------------*/
+
+static int __init tps_init(void)
+{
+	u32	tries = 3;
+	int	status = -ENODEV;
+
+	printk(KERN_INFO "%s: version %s\n", DRIVER_NAME, DRIVER_VERSION);
+
+	/* some boards have startup glitches */
+	while (tries--) {
+		status = i2c_add_driver(&tps65010_driver);
+		if (the_tps)
+			break;
+		i2c_del_driver(&tps65010_driver);
+		if (!tries) {
+			printk(KERN_ERR "%s: no chip?\n", DRIVER_NAME);
+			return -ENODEV;
+		}
+		pr_debug("%s: re-probe ...\n", DRIVER_NAME);
+		msleep(10);
+	}
+
+#if defined(CONFIG_ARM)
+	if (machine_is_omap_osk()) {
+
+		// FIXME: More should be placed in the initialization code
+		//	  of the submodules (DSP, ethernet, power management,
+		//	  board-osk.c). Careful: I2C is initialized "late".
+
+		/* Let LED1 (D9) blink */
+		tps65010_set_led(LED1, BLINK);
+
+		/* Disable LED 2 (D2) */
+		tps65010_set_led(LED2, OFF);
+
+		/* Set GPIO 1 HIGH to disable VBUS power supply;
+		 * OHCI driver powers it up/down as needed.
+		 */
+		tps65010_set_gpio_out_value(GPIO1, HIGH);
+
+		/* Set GPIO 2 low to turn on LED D3 */
+		tps65010_set_gpio_out_value(GPIO2, HIGH);
+
+		/* Set GPIO 3 low to take ethernet out of reset */
+		tps65010_set_gpio_out_value(GPIO3, LOW);
+
+		/* gpio4 for VDD_DSP */
+
+		/* Enable LOW_PWR */
+		tps65010_set_low_pwr(ON);
+
+		/* Switch VLDO2 to 3.0V for AIC23 */
+		tps65010_config_vregs1(TPS_LDO2_ENABLE | TPS_VLDO2_3_0V | TPS_LDO1_ENABLE);
+
+	} else if (machine_is_omap_h2()) {
+		/* gpio3 for SD, gpio4 for VDD_DSP */
+
+		/* Enable LOW_PWR */
+		tps65010_set_low_pwr(ON);
+	} else if (machine_is_omap_h3()) {
+		/* gpio4 for SD, gpio3 for VDD_DSP */
+#ifdef CONFIG_PM
+		/* Enable LOW_PWR */
+		tps65013_set_low_pwr(ON);
+#endif
+	}
+#endif
+
+	return status;
+}
+/* NOTE:  this MUST be initialized before the other parts of the system
+ * that rely on it ... but after the i2c bus on which this relies.
+ * That is, much earlier than on PC-type systems, which don't often use
+ * I2C as a core system bus.
+ */
+subsys_initcall(tps_init);
+
+static void __exit tps_exit(void)
+{
+	i2c_del_driver(&tps65010_driver);
+}
+module_exit(tps_exit);
+
diff --git a/include/asm-arm/arch-omap/tps65010.h b/include/asm-arm/arch-omap/tps65010.h
index 0f97bb2e8fc..b9aa2b3a390 100644
--- a/include/asm-arm/arch-omap/tps65010.h
+++ b/include/asm-arm/arch-omap/tps65010.h
@@ -28,6 +28,66 @@
 #ifndef __ASM_ARCH_TPS65010_H
 #define __ASM_ARCH_TPS65010_H
 
+/*
+ * ----------------------------------------------------------------------------
+ * Registers, all 8 bits
+ * ----------------------------------------------------------------------------
+ */
+
+#define	TPS_CHGSTATUS		0x01
+#	define	TPS_CHG_USB		(1 << 7)
+#	define	TPS_CHG_AC		(1 << 6)
+#	define	TPS_CHG_THERM		(1 << 5)
+#	define	TPS_CHG_TERM		(1 << 4)
+#	define	TPS_CHG_TAPER_TMO	(1 << 3)
+#	define	TPS_CHG_CHG_TMO		(1 << 2)
+#	define	TPS_CHG_PRECHG_TMO	(1 << 1)
+#	define	TPS_CHG_TEMP_ERR	(1 << 0)
+#define	TPS_REGSTATUS		0x02
+#	define	TPS_REG_ONOFF		(1 << 7)
+#	define	TPS_REG_COVER		(1 << 6)
+#	define	TPS_REG_UVLO		(1 << 5)
+#	define	TPS_REG_NO_CHG		(1 << 4)	/* tps65013 */
+#	define	TPS_REG_PG_LD02		(1 << 3)
+#	define	TPS_REG_PG_LD01		(1 << 2)
+#	define	TPS_REG_PG_MAIN		(1 << 1)
+#	define	TPS_REG_PG_CORE		(1 << 0)
+#define	TPS_MASK1		0x03
+#define	TPS_MASK2		0x04
+#define	TPS_ACKINT1		0x05
+#define	TPS_ACKINT2		0x06
+#define	TPS_CHGCONFIG		0x07
+#	define	TPS_CHARGE_POR		(1 << 7)	/* 65010/65012 */
+#	define	TPS65013_AUA		(1 << 7)	/* 65011/65013 */
+#	define	TPS_CHARGE_RESET	(1 << 6)
+#	define	TPS_CHARGE_FAST		(1 << 5)
+#	define	TPS_CHARGE_CURRENT	(3 << 3)
+#	define	TPS_VBUS_500MA		(1 << 2)
+#	define	TPS_VBUS_CHARGING	(1 << 1)
+#	define	TPS_CHARGE_ENABLE	(1 << 0)
+#define	TPS_LED1_ON		0x08
+#define	TPS_LED1_PER		0x09
+#define	TPS_LED2_ON		0x0a
+#define	TPS_LED2_PER		0x0b
+#define	TPS_VDCDC1		0x0c
+#	define	TPS_ENABLE_LP		(1 << 3)
+#define	TPS_VDCDC2		0x0d
+#define	TPS_VREGS1		0x0e
+#	define	TPS_LDO2_ENABLE	(1 << 7)
+#	define	TPS_LDO2_OFF	(1 << 6)
+#	define	TPS_VLDO2_3_0V	(3 << 4)
+#	define	TPS_VLDO2_2_75V	(2 << 4)
+#	define	TPS_VLDO2_2_5V	(1 << 4)
+#	define	TPS_VLDO2_1_8V	(0 << 4)
+#	define	TPS_LDO1_ENABLE	(1 << 3)
+#	define	TPS_LDO1_OFF	(1 << 2)
+#	define	TPS_VLDO1_3_0V	(3 << 0)
+#	define	TPS_VLDO1_2_75V	(2 << 0)
+#	define	TPS_VLDO1_2_5V	(1 << 0)
+#	define	TPS_VLDO1_ADJ	(0 << 0)
+#define	TPS_MASK3		0x0f
+#define	TPS_DEFGPIO		0x10
+
 /*
  * ----------------------------------------------------------------------------
  * Macros used by exported functions
@@ -71,10 +131,26 @@ extern int tps65010_set_gpio_out_value(unsigned gpio, unsigned value);
  */
 extern int tps65010_set_led(unsigned led, unsigned mode);
 
+/* tps65010_set_vib parameter:
+ * value: ON or OFF
+ */
+extern int tps65010_set_vib(unsigned value);
+
 /* tps65010_set_low_pwr parameter:
  * mode: ON or OFF
  */
 extern int tps65010_set_low_pwr(unsigned mode);
 
+/* tps65010_config_vregs1 parameter:
+ * value to be written to VREGS1 register
+ * Note: The complete register is written, set all bits you need
+ */
+extern int tps65010_config_vregs1(unsigned value);
+
+/* tps65013_set_low_pwr parameter:
+ * mode: ON or OFF
+ */
+extern int tps65013_set_low_pwr(unsigned mode);
+
 #endif /*  __ASM_ARCH_TPS65010_H */
 
-- 
cgit v1.2.3-70-g09d2


From 10c08f8100ee2c4d27b862635574cdf4ef439e67 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Mon, 6 Jun 2005 19:34:45 +0200
Subject: [PATCH] I2C: rename i2c-sysfs.h to hwmon-sysfs.h

This patch renames the new linux/i2c-sysfs.h header file to
linux/hwmon-sysfs.h. This names seems to be more appropriate since this
file defines macros and structures not related to i2c but to hardware
monitoring drivers. The patch also updates the five hardware monitoring
driver which include that header file already.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/chips/adm1026.c |  2 +-
 drivers/i2c/chips/it87.c    |  2 +-
 drivers/i2c/chips/lm63.c    |  2 +-
 drivers/i2c/chips/lm83.c    |  2 +-
 drivers/i2c/chips/lm90.c    |  2 +-
 include/linux/hwmon-sysfs.h | 36 ++++++++++++++++++++++++++++++++++++
 include/linux/i2c-sysfs.h   | 36 ------------------------------------
 7 files changed, 41 insertions(+), 41 deletions(-)
 create mode 100644 include/linux/hwmon-sysfs.h
 delete mode 100644 include/linux/i2c-sysfs.h

(limited to 'include')

diff --git a/drivers/i2c/chips/adm1026.c b/drivers/i2c/chips/adm1026.c
index ddbc01505ed..3c85fe150cd 100644
--- a/drivers/i2c/chips/adm1026.c
+++ b/drivers/i2c/chips/adm1026.c
@@ -29,8 +29,8 @@
 #include <linux/jiffies.h>
 #include <linux/i2c.h>
 #include <linux/i2c-sensor.h>
-#include <linux/i2c-sysfs.h>
 #include <linux/i2c-vid.h>
+#include <linux/hwmon-sysfs.h>
 
 /* Addresses to scan */
 static unsigned short normal_i2c[] = { 0x2c, 0x2d, 0x2e, I2C_CLIENT_END };
diff --git a/drivers/i2c/chips/it87.c b/drivers/i2c/chips/it87.c
index 6a9b65a10bb..db20c9e4739 100644
--- a/drivers/i2c/chips/it87.c
+++ b/drivers/i2c/chips/it87.c
@@ -37,8 +37,8 @@
 #include <linux/jiffies.h>
 #include <linux/i2c.h>
 #include <linux/i2c-sensor.h>
-#include <linux/i2c-sysfs.h>
 #include <linux/i2c-vid.h>
+#include <linux/hwmon-sysfs.h>
 #include <asm/io.h>
 
 
diff --git a/drivers/i2c/chips/lm63.c b/drivers/i2c/chips/lm63.c
index a1fd12bd615..7c6f9ea5a25 100644
--- a/drivers/i2c/chips/lm63.c
+++ b/drivers/i2c/chips/lm63.c
@@ -43,7 +43,7 @@
 #include <linux/jiffies.h>
 #include <linux/i2c.h>
 #include <linux/i2c-sensor.h>
-#include <linux/i2c-sysfs.h>
+#include <linux/hwmon-sysfs.h>
 
 /*
  * Addresses to scan
diff --git a/drivers/i2c/chips/lm83.c b/drivers/i2c/chips/lm83.c
index 0e0eae4dcea..a49008b444c 100644
--- a/drivers/i2c/chips/lm83.c
+++ b/drivers/i2c/chips/lm83.c
@@ -33,7 +33,7 @@
 #include <linux/jiffies.h>
 #include <linux/i2c.h>
 #include <linux/i2c-sensor.h>
-#include <linux/i2c-sysfs.h>
+#include <linux/hwmon-sysfs.h>
 
 /*
  * Addresses to scan
diff --git a/drivers/i2c/chips/lm90.c b/drivers/i2c/chips/lm90.c
index ebd99dfbf9c..a67dcadf7cb 100644
--- a/drivers/i2c/chips/lm90.c
+++ b/drivers/i2c/chips/lm90.c
@@ -76,7 +76,7 @@
 #include <linux/jiffies.h>
 #include <linux/i2c.h>
 #include <linux/i2c-sensor.h>
-#include <linux/i2c-sysfs.h>
+#include <linux/hwmon-sysfs.h>
 
 /*
  * Addresses to scan
diff --git a/include/linux/hwmon-sysfs.h b/include/linux/hwmon-sysfs.h
new file mode 100644
index 00000000000..1b5018a965f
--- /dev/null
+++ b/include/linux/hwmon-sysfs.h
@@ -0,0 +1,36 @@
+/*
+ *  hwmon-sysfs.h - hardware monitoring chip driver sysfs defines
+ *
+ *  Copyright (C) 2005 Yani Ioannou <yani.ioannou@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+#ifndef _LINUX_HWMON_SYSFS_H
+#define _LINUX_HWMON_SYSFS_H
+
+struct sensor_device_attribute{
+	struct device_attribute dev_attr;
+	int index;
+};
+#define to_sensor_dev_attr(_dev_attr) \
+	container_of(_dev_attr, struct sensor_device_attribute, dev_attr)
+
+#define SENSOR_DEVICE_ATTR(_name,_mode,_show,_store,_index)	\
+struct sensor_device_attribute sensor_dev_attr_##_name = {	\
+	.dev_attr =	__ATTR(_name,_mode,_show,_store),	\
+	.index =	_index,					\
+}
+
+#endif /* _LINUX_HWMON_SYSFS_H */
diff --git a/include/linux/i2c-sysfs.h b/include/linux/i2c-sysfs.h
deleted file mode 100644
index d7bf6ce1167..00000000000
--- a/include/linux/i2c-sysfs.h
+++ /dev/null
@@ -1,36 +0,0 @@
-/*
- *  i2c-sysfs.h - i2c chip driver sysfs defines
- *
- *  Copyright (C) 2005 Yani Ioannou <yani.ioannou@gmail.com>
- *
- *  This program is free software; you can redistribute it and/or modify
- *  it under the terms of the GNU General Public License as published by
- *  the Free Software Foundation; either version 2 of the License, or
- *  (at your option) any later version.
- *
- *  This program is distributed in the hope that it will be useful,
- *  but WITHOUT ANY WARRANTY; without even the implied warranty of
- *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- *  GNU General Public License for more details.
- *
- *  You should have received a copy of the GNU General Public License
- *  along with this program; if not, write to the Free Software
- *  Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#ifndef _LINUX_I2C_SYSFS_H
-#define _LINUX_I2C_SYSFS_H
-
-struct sensor_device_attribute{
-	struct device_attribute dev_attr;
-	int index;
-};
-#define to_sensor_dev_attr(_dev_attr) \
-	container_of(_dev_attr, struct sensor_device_attribute, dev_attr)
-
-#define SENSOR_DEVICE_ATTR(_name,_mode,_show,_store,_index)	\
-struct sensor_device_attribute sensor_dev_attr_##_name = {	\
-	.dev_attr =	__ATTR(_name,_mode,_show,_store),	\
-	.index =	_index,					\
-}
-
-#endif /* _LINUX_I2C_SYSFS_H */
-- 
cgit v1.2.3-70-g09d2


From c124a78d8c7475ecc43f385f34112b638c4228d9 Mon Sep 17 00:00:00 2001
From: Randy Vinson <rvinson@mvista.com>
Date: Fri, 3 Jun 2005 14:36:06 -0700
Subject: [PATCH] I2C: Add support for Maxim/Dallas DS1374 Real-Time Clock Chip
 (1/2)

Add support for Maxim/Dallas DS1374 Real-Time Clock Chip

This change adds support for the Maxim/Dallas DS1374 RTC chip. This chip
is an I2C-based RTC that maintains a simple 32-bit binary seconds count
with battery backup support.

Signed-off-by: Randy Vinson <rvinson@mvista.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/chips/Kconfig  |  11 ++
 drivers/i2c/chips/Makefile |   1 +
 drivers/i2c/chips/ds1374.c | 266 +++++++++++++++++++++++++++++++++++++++++++++
 include/linux/i2c-id.h     |   1 +
 4 files changed, 279 insertions(+)
 create mode 100644 drivers/i2c/chips/ds1374.c

(limited to 'include')

diff --git a/drivers/i2c/chips/Kconfig b/drivers/i2c/chips/Kconfig
index 33de80afd6c..a0982da0980 100644
--- a/drivers/i2c/chips/Kconfig
+++ b/drivers/i2c/chips/Kconfig
@@ -417,6 +417,17 @@ config SENSORS_DS1337
 	  This driver can also be built as a module.  If so, the module
 	  will be called ds1337.
 
+config SENSORS_DS1374
+	tristate "Maxim/Dallas Semiconductor DS1374 Real Time Clock"
+	depends on I2C && EXPERIMENTAL
+	select I2C_SENSOR
+	help
+	  If you say yes here you get support for Dallas Semiconductor
+	  DS1374 real-time clock chips.
+
+	  This driver can also be built as a module.  If so, the module
+	  will be called ds1374.
+
 config SENSORS_EEPROM
 	tristate "EEPROM reader"
 	depends on I2C && EXPERIMENTAL
diff --git a/drivers/i2c/chips/Makefile b/drivers/i2c/chips/Makefile
index 6bebdc10416..b5e6d2f84f9 100644
--- a/drivers/i2c/chips/Makefile
+++ b/drivers/i2c/chips/Makefile
@@ -14,6 +14,7 @@ obj-$(CONFIG_SENSORS_ADM1031)	+= adm1031.o
 obj-$(CONFIG_SENSORS_ADM9240)	+= adm9240.o
 obj-$(CONFIG_SENSORS_ATXP1)	+= atxp1.o
 obj-$(CONFIG_SENSORS_DS1337)	+= ds1337.o
+obj-$(CONFIG_SENSORS_DS1374)	+= ds1374.o
 obj-$(CONFIG_SENSORS_DS1621)	+= ds1621.o
 obj-$(CONFIG_SENSORS_EEPROM)	+= eeprom.o
 obj-$(CONFIG_SENSORS_FSCHER)	+= fscher.o
diff --git a/drivers/i2c/chips/ds1374.c b/drivers/i2c/chips/ds1374.c
new file mode 100644
index 00000000000..1278d979db2
--- /dev/null
+++ b/drivers/i2c/chips/ds1374.c
@@ -0,0 +1,266 @@
+/*
+ * drivers/i2c/chips/ds1374.c
+ *
+ * I2C client/driver for the Maxim/Dallas DS1374 Real-Time Clock
+ *
+ * Author: Randy Vinson <rvinson@mvista.com>
+ *
+ * Based on the m41t00.c by Mark Greer <mgreer@mvista.com>
+ *
+ * 2005 (c) MontaVista Software, Inc. This file is licensed under
+ * the terms of the GNU General Public License version 2. This program
+ * is licensed "as is" without any warranty of any kind, whether express
+ * or implied.
+ */
+/*
+ * This i2c client/driver wedges between the drivers/char/genrtc.c RTC
+ * interface and the SMBus interface of the i2c subsystem.
+ * It would be more efficient to use i2c msgs/i2c_transfer directly but, as
+ * recommened in .../Documentation/i2c/writing-clients section
+ * "Sending and receiving", using SMBus level communication is preferred.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/interrupt.h>
+#include <linux/i2c.h>
+#include <linux/rtc.h>
+#include <linux/bcd.h>
+
+#include <asm/time.h>
+#include <asm/rtc.h>
+
+#define DS1374_REG_TOD0		0x00
+#define DS1374_REG_TOD1		0x01
+#define DS1374_REG_TOD2		0x02
+#define DS1374_REG_TOD3		0x03
+#define DS1374_REG_WDALM0	0x04
+#define DS1374_REG_WDALM1	0x05
+#define DS1374_REG_WDALM2	0x06
+#define DS1374_REG_CR		0x07
+#define DS1374_REG_SR		0x08
+#define DS1374_REG_SR_OSF	0x80
+#define DS1374_REG_TCR		0x09
+
+#define	DS1374_DRV_NAME		"ds1374"
+
+static DECLARE_MUTEX(ds1374_mutex);
+
+static struct i2c_driver ds1374_driver;
+static struct i2c_client *save_client;
+
+static unsigned short ignore[] = { I2C_CLIENT_END };
+static unsigned short normal_addr[] = { 0x68, I2C_CLIENT_END };
+
+static struct i2c_client_address_data addr_data = {
+	.normal_i2c = normal_addr,
+	.normal_i2c_range = ignore,
+	.probe = ignore,
+	.probe_range = ignore,
+	.ignore = ignore,
+	.ignore_range = ignore,
+	.force = ignore,
+};
+
+static ulong ds1374_read_rtc(void)
+{
+	ulong time = 0;
+	int reg = DS1374_REG_WDALM0;
+
+	while (reg--) {
+		s32 tmp;
+		if ((tmp = i2c_smbus_read_byte_data(save_client, reg)) < 0) {
+			dev_warn(&save_client->dev,
+				 "can't read from rtc chip\n");
+			return 0;
+		}
+		time = (time << 8) | (tmp & 0xff);
+	}
+	return time;
+}
+
+static void ds1374_write_rtc(ulong time)
+{
+	int reg;
+
+	for (reg = DS1374_REG_TOD0; reg < DS1374_REG_WDALM0; reg++) {
+		if (i2c_smbus_write_byte_data(save_client, reg, time & 0xff)
+		    < 0) {
+			dev_warn(&save_client->dev,
+				 "can't write to rtc chip\n");
+			break;
+		}
+		time = time >> 8;
+	}
+}
+
+static void ds1374_check_rtc_status(void)
+{
+	s32 tmp;
+
+	tmp = i2c_smbus_read_byte_data(save_client, DS1374_REG_SR);
+	if (tmp < 0) {
+		dev_warn(&save_client->dev,
+			 "can't read status from rtc chip\n");
+		return;
+	}
+	if (tmp & DS1374_REG_SR_OSF) {
+		dev_warn(&save_client->dev,
+			 "oscillator discontinuity flagged, time unreliable\n");
+		tmp &= ~DS1374_REG_SR_OSF;
+		tmp = i2c_smbus_write_byte_data(save_client, DS1374_REG_SR,
+						tmp & 0xff);
+		if (tmp < 0)
+			dev_warn(&save_client->dev,
+				 "can't clear discontinuity notification\n");
+	}
+}
+
+ulong ds1374_get_rtc_time(void)
+{
+	ulong t1, t2;
+	int limit = 10;		/* arbitrary retry limit */
+
+	down(&ds1374_mutex);
+
+	/*
+	 * Since the reads are being performed one byte at a time using
+	 * the SMBus vs a 4-byte i2c transfer, there is a chance that a
+	 * carry will occur during the read. To detect this, 2 reads are
+	 * performed and compared.
+	 */
+	do {
+		t1 = ds1374_read_rtc();
+		t2 = ds1374_read_rtc();
+	} while (t1 != t2 && limit--);
+
+	up(&ds1374_mutex);
+
+	if (t1 != t2) {
+		dev_warn(&save_client->dev,
+			 "can't get consistent time from rtc chip\n");
+		t1 = 0;
+	}
+
+	return t1;
+}
+
+static void ds1374_set_tlet(ulong arg)
+{
+	ulong t1, t2;
+	int limit = 10;		/* arbitrary retry limit */
+
+	t1 = *(ulong *) arg;
+
+	down(&ds1374_mutex);
+
+	/*
+	 * Since the writes are being performed one byte at a time using
+	 * the SMBus vs a 4-byte i2c transfer, there is a chance that a
+	 * carry will occur during the write. To detect this, the write
+	 * value is read back and compared.
+	 */
+	do {
+		ds1374_write_rtc(t1);
+		t2 = ds1374_read_rtc();
+	} while (t1 != t2 && limit--);
+
+	up(&ds1374_mutex);
+
+	if (t1 != t2)
+		dev_warn(&save_client->dev,
+			 "can't confirm time set from rtc chip\n");
+}
+
+ulong new_time;
+
+DECLARE_TASKLET_DISABLED(ds1374_tasklet, ds1374_set_tlet, (ulong) & new_time);
+
+int ds1374_set_rtc_time(ulong nowtime)
+{
+	new_time = nowtime;
+
+	if (in_interrupt())
+		tasklet_schedule(&ds1374_tasklet);
+	else
+		ds1374_set_tlet((ulong) & new_time);
+
+	return 0;
+}
+
+/*
+ *****************************************************************************
+ *
+ *	Driver Interface
+ *
+ *****************************************************************************
+ */
+static int ds1374_probe(struct i2c_adapter *adap, int addr, int kind)
+{
+	struct i2c_client *client;
+	int rc;
+
+	client = kmalloc(sizeof(struct i2c_client), GFP_KERNEL);
+	if (!client)
+		return -ENOMEM;
+
+	memset(client, 0, sizeof(struct i2c_client));
+	strncpy(client->name, DS1374_DRV_NAME, I2C_NAME_SIZE);
+	client->flags = I2C_DF_NOTIFY;
+	client->addr = addr;
+	client->adapter = adap;
+	client->driver = &ds1374_driver;
+
+	if ((rc = i2c_attach_client(client)) != 0) {
+		kfree(client);
+		return rc;
+	}
+
+	save_client = client;
+
+	ds1374_check_rtc_status();
+
+	return 0;
+}
+
+static int ds1374_attach(struct i2c_adapter *adap)
+{
+	return i2c_probe(adap, &addr_data, ds1374_probe);
+}
+
+static int ds1374_detach(struct i2c_client *client)
+{
+	int rc;
+
+	if ((rc = i2c_detach_client(client)) == 0) {
+		kfree(i2c_get_clientdata(client));
+		tasklet_kill(&ds1374_tasklet);
+	}
+	return rc;
+}
+
+static struct i2c_driver ds1374_driver = {
+	.owner = THIS_MODULE,
+	.name = DS1374_DRV_NAME,
+	.id = I2C_DRIVERID_DS1374,
+	.flags = I2C_DF_NOTIFY,
+	.attach_adapter = ds1374_attach,
+	.detach_client = ds1374_detach,
+};
+
+static int __init ds1374_init(void)
+{
+	return i2c_add_driver(&ds1374_driver);
+}
+
+static void __exit ds1374_exit(void)
+{
+	i2c_del_driver(&ds1374_driver);
+}
+
+module_init(ds1374_init);
+module_exit(ds1374_exit);
+
+MODULE_AUTHOR("Randy Vinson <rvinson@mvista.com>");
+MODULE_DESCRIPTION("Maxim/Dallas DS1374 RTC I2C Client Driver");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h
index 89270ce5147..33f08258f22 100644
--- a/include/linux/i2c-id.h
+++ b/include/linux/i2c-id.h
@@ -108,6 +108,7 @@
 #define I2C_DRIVERID_TDA7313	62	/* TDA7313 audio processor	*/
 #define I2C_DRIVERID_MAX6900	63	/* MAX6900 real-time clock	*/
 #define I2C_DRIVERID_SAA7114H	64	/* video decoder		*/
+#define I2C_DRIVERID_DS1374	65	/* DS1374 real time clock	*/
 
 
 #define I2C_DRIVERID_EXP0	0xF0	/* experimental use id's	*/
-- 
cgit v1.2.3-70-g09d2


From 69ad07cf98d0ef65cac67bac2ea4381bb499bea8 Mon Sep 17 00:00:00 2001
From: Jaroslav Kysela <perex@suse.cz>
Date: Mon, 30 May 2005 14:48:16 +0200
Subject: [ALSA] AC97 - renamed vendor/device to subvendor/subdevice where
 appropriate

AC97 Codec,ATIIXP driver,VIA82xx driver
To avoid confusion, the structure members vendor/device were renamed
to subvendor/subdevice, because we compare them with PCI subsystem vendor
and subsystem device.

Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/ac97_codec.h  |   4 +-
 sound/pci/ac97/ac97_codec.c |   8 +--
 sound/pci/atiixp.c          |   4 +-
 sound/pci/via82xx.c         | 118 ++++++++++++++++++++++----------------------
 4 files changed, 67 insertions(+), 67 deletions(-)

(limited to 'include')

diff --git a/include/sound/ac97_codec.h b/include/sound/ac97_codec.h
index 996eeab683b..1309c12b8f7 100644
--- a/include/sound/ac97_codec.h
+++ b/include/sound/ac97_codec.h
@@ -573,8 +573,8 @@ enum {
 };
 
 struct ac97_quirk {
-	unsigned short vendor;	/* PCI vendor id */
-	unsigned short device;	/* PCI device id */
+	unsigned short subvendor; /* PCI subsystem vendor id */
+	unsigned short subdevice; /* PCI sybsystem device id */
 	unsigned short mask;	/* device id bit mask, 0 = accept all */
 	unsigned int codec_id;	/* codec id (if any), 0 = accept all */
 	const char *name;	/* name shown as info */
diff --git a/sound/pci/ac97/ac97_codec.c b/sound/pci/ac97/ac97_codec.c
index 996fcfb0953..a4b72cd2eea 100644
--- a/sound/pci/ac97/ac97_codec.c
+++ b/sound/pci/ac97/ac97_codec.c
@@ -2542,11 +2542,11 @@ int snd_ac97_tune_hardware(ac97_t *ac97, struct ac97_quirk *quirk, const char *o
 		return result;
 	}
 
-	for (; quirk->vendor; quirk++) {
-		if (quirk->vendor != ac97->subsystem_vendor)
+	for (; quirk->subvendor; quirk++) {
+		if (quirk->subvendor != ac97->subsystem_vendor)
 			continue;
-		if ((! quirk->mask && quirk->device == ac97->subsystem_device) ||
-		    quirk->device == (quirk->mask & ac97->subsystem_device)) {
+		if ((! quirk->mask && quirk->subdevice == ac97->subsystem_device) ||
+		    quirk->subdevice == (quirk->mask & ac97->subsystem_device)) {
 			if (quirk->codec_id && quirk->codec_id != ac97->id)
 				continue;
 			snd_printdd("ac97 quirk for %s (%04x:%04x)\n", quirk->name, ac97->subsystem_vendor, ac97->subsystem_device);
diff --git a/sound/pci/atiixp.c b/sound/pci/atiixp.c
index 06551a69fb4..cafab4af5c5 100644
--- a/sound/pci/atiixp.c
+++ b/sound/pci/atiixp.c
@@ -1334,8 +1334,8 @@ static irqreturn_t snd_atiixp_interrupt(int irq, void *dev_id, struct pt_regs *r
 
 static struct ac97_quirk ac97_quirks[] __devinitdata = {
 	{
-		.vendor = 0x103c,
-		.device = 0x006b,
+		.subvendor = 0x103c,
+		.subdevice = 0x006b,
 		.name = "HP Pavilion ZV5030US",
 		.type = AC97_TUNE_MUTE_LED
 	},
diff --git a/sound/pci/via82xx.c b/sound/pci/via82xx.c
index 4c47fc83797..52d1074f869 100644
--- a/sound/pci/via82xx.c
+++ b/sound/pci/via82xx.c
@@ -1560,51 +1560,51 @@ static void snd_via82xx_mixer_free_ac97(ac97_t *ac97)
 
 static struct ac97_quirk ac97_quirks[] = {
 	{
-		.vendor = 0x1106,
-		.device = 0x4161,
+		.subvendor = 0x1106,
+		.subdevice = 0x4161,
 		.codec_id = 0x56494161, /* VT1612A */
 		.name = "Soltek SL-75DRV5",
 		.type = AC97_TUNE_NONE
 	},
 	{	/* FIXME: which codec? */
-		.vendor = 0x1106,
-		.device = 0x4161,
+		.subvendor = 0x1106,
+		.subdevice = 0x4161,
 		.name = "ASRock K7VT2",
 		.type = AC97_TUNE_HP_ONLY
 	},
 	{
-		.vendor = 0x1019,
-		.device = 0x0a81,
+		.subvendor = 0x1019,
+		.subdevice = 0x0a81,
 		.name = "ECS K7VTA3",
 		.type = AC97_TUNE_HP_ONLY
 	},
 	{
-		.vendor = 0x1019,
-		.device = 0x0a85,
+		.subvendor = 0x1019,
+		.subdevice = 0x0a85,
 		.name = "ECS L7VMM2",
 		.type = AC97_TUNE_HP_ONLY
 	},
 	{
-		.vendor = 0x1849,
-		.device = 0x3059,
+		.subvendor = 0x1849,
+		.subdevice = 0x3059,
 		.name = "ASRock K7VM2",
 		.type = AC97_TUNE_HP_ONLY	/* VT1616 */
 	},
 	{
-		.vendor = 0x14cd,
-		.device = 0x7002,
+		.subvendor = 0x14cd,
+		.subdevice = 0x7002,
 		.name = "Unknown",
 		.type = AC97_TUNE_ALC_JACK
 	},
 	{
-		.vendor = 0x1071,
-		.device = 0x8590,
+		.subvendor = 0x1071,
+		.subdevice = 0x8590,
 		.name = "Mitac Mobo",
 		.type = AC97_TUNE_ALC_JACK
 	},
 	{
-		.vendor = 0x161f,
-		.device = 0x202b,
+		.subvendor = 0x161f,
+		.subdevice = 0x202b,
 		.name = "Arima Notebook",
 		.type = AC97_TUNE_HP_ONLY,
 	},
@@ -2142,8 +2142,8 @@ static struct via823x_info via823x_cards[] __devinitdata = {
  * auto detection of DXS channel supports.
  */
 struct dxs_whitelist {
-	unsigned short vendor;
-	unsigned short device; 
+	unsigned short subvendor;
+	unsigned short subdevice; 
 	unsigned short mask; 
 	short action;	/* new dxs_support value */
 };
@@ -2151,43 +2151,43 @@ struct dxs_whitelist {
 static int __devinit check_dxs_list(struct pci_dev *pci)
 {
 	static struct dxs_whitelist whitelist[] = {
-		{ .vendor = 0x1005, .device = 0x4710, .action = VIA_DXS_ENABLE }, /* Avance Logic Mobo */
-		{ .vendor = 0x1019, .device = 0x0996, .action = VIA_DXS_48K },
-		{ .vendor = 0x1019, .device = 0x0a81, .action = VIA_DXS_NO_VRA }, /* ECS K7VTA3 v8.0 */
-		{ .vendor = 0x1019, .device = 0x0a85, .action = VIA_DXS_NO_VRA }, /* ECS L7VMM2 */
-		{ .vendor = 0x1025, .device = 0x0033, .action = VIA_DXS_NO_VRA }, /* Acer Inspire 1353LM */
-		{ .vendor = 0x1043, .device = 0x8095, .action = VIA_DXS_NO_VRA }, /* ASUS A7V8X (FIXME: possibly VIA_DXS_ENABLE?)*/
-		{ .vendor = 0x1043, .device = 0x80a1, .action = VIA_DXS_NO_VRA }, /* ASUS A7V8-X */
-		{ .vendor = 0x1043, .device = 0x80b0, .action = VIA_DXS_NO_VRA }, /* ASUS A7V600 & K8V*/ 
-		{ .vendor = 0x1043, .device = 0x812a, .action = VIA_DXS_SRC    }, /* ASUS A8V Deluxe */ 
-		{ .vendor = 0x1071, .device = 0x8375, .action = VIA_DXS_NO_VRA }, /* Vobis/Yakumo/Mitac notebook */
-		{ .vendor = 0x1071, .device = 0x8399, .action = VIA_DXS_ENABLE }, /* Umax AB 595T (VIA K8N800A - VT8237) */
-		{ .vendor = 0x10cf, .device = 0x118e, .action = VIA_DXS_ENABLE }, /* FSC laptop */
-		{ .vendor = 0x1106, .device = 0x4161, .action = VIA_DXS_NO_VRA }, /* ASRock K7VT2 */
-		{ .vendor = 0x1106, .device = 0x4552, .action = VIA_DXS_NO_VRA }, /* QDI Kudoz 7X/600-6AL */
-		{ .vendor = 0x1106, .device = 0xaa01, .action = VIA_DXS_NO_VRA }, /* EPIA MII */
-		{ .vendor = 0x1106, .device = 0xc001, .action = VIA_DXS_SRC }, /* Insight P4-ITX */
-		{ .vendor = 0x1297, .device = 0xa232, .action = VIA_DXS_ENABLE }, /* Shuttle ?? */
-		{ .vendor = 0x1297, .device = 0xc160, .action = VIA_DXS_ENABLE }, /* Shuttle SK41G */
-		{ .vendor = 0x1458, .device = 0xa002, .action = VIA_DXS_ENABLE }, /* Gigabyte GA-7VAXP */
-		{ .vendor = 0x1462, .device = 0x0080, .action = VIA_DXS_SRC }, /* MSI K8T Neo-FIS2R */
-		{ .vendor = 0x1462, .device = 0x3800, .action = VIA_DXS_ENABLE }, /* MSI KT266 */
-		{ .vendor = 0x1462, .device = 0x5901, .action = VIA_DXS_NO_VRA }, /* MSI KT6 Delta-SR */
-		{ .vendor = 0x1462, .device = 0x7023, .action = VIA_DXS_NO_VRA }, /* MSI K8T Neo2-FI */
-		{ .vendor = 0x1462, .device = 0x7120, .action = VIA_DXS_ENABLE }, /* MSI KT4V */
-		{ .vendor = 0x147b, .device = 0x1401, .action = VIA_DXS_ENABLE }, /* ABIT KD7(-RAID) */
-		{ .vendor = 0x147b, .device = 0x1411, .action = VIA_DXS_ENABLE }, /* ABIT VA-20 */
-		{ .vendor = 0x147b, .device = 0x1413, .action = VIA_DXS_ENABLE }, /* ABIT KV8 Pro */
-		{ .vendor = 0x147b, .device = 0x1415, .action = VIA_DXS_NO_VRA }, /* Abit AV8 */
-		{ .vendor = 0x14ff, .device = 0x0403, .action = VIA_DXS_ENABLE }, /* Twinhead mobo */
-		{ .vendor = 0x14ff, .device = 0x0408, .action = VIA_DXS_NO_VRA }, /* Twinhead mobo */
-		{ .vendor = 0x1584, .device = 0x8120, .action = VIA_DXS_ENABLE }, /* Gericom/Targa/Vobis/Uniwill laptop */
-		{ .vendor = 0x1584, .device = 0x8123, .action = VIA_DXS_NO_VRA }, /* Uniwill (Targa Visionary XP-210) */
-		{ .vendor = 0x161f, .device = 0x202b, .action = VIA_DXS_NO_VRA }, /* Amira Note book */
-		{ .vendor = 0x161f, .device = 0x2032, .action = VIA_DXS_48K }, /* m680x machines */
-		{ .vendor = 0x1631, .device = 0xe004, .action = VIA_DXS_ENABLE }, /* Easy Note 3174, Packard Bell */
-		{ .vendor = 0x1695, .device = 0x3005, .action = VIA_DXS_ENABLE }, /* EPoX EP-8K9A */
-		{ .vendor = 0x1849, .device = 0x3059, .action = VIA_DXS_NO_VRA }, /* ASRock K7VM2 */
+		{ .subvendor = 0x1005, .subdevice = 0x4710, .action = VIA_DXS_ENABLE }, /* Avance Logic Mobo */
+		{ .subvendor = 0x1019, .subdevice = 0x0996, .action = VIA_DXS_48K },
+		{ .subvendor = 0x1019, .subdevice = 0x0a81, .action = VIA_DXS_NO_VRA }, /* ECS K7VTA3 v8.0 */
+		{ .subvendor = 0x1019, .subdevice = 0x0a85, .action = VIA_DXS_NO_VRA }, /* ECS L7VMM2 */
+		{ .subvendor = 0x1025, .subdevice = 0x0033, .action = VIA_DXS_NO_VRA }, /* Acer Inspire 1353LM */
+		{ .subvendor = 0x1043, .subdevice = 0x8095, .action = VIA_DXS_NO_VRA }, /* ASUS A7V8X (FIXME: possibly VIA_DXS_ENABLE?)*/
+		{ .subvendor = 0x1043, .subdevice = 0x80a1, .action = VIA_DXS_NO_VRA }, /* ASUS A7V8-X */
+		{ .subvendor = 0x1043, .subdevice = 0x80b0, .action = VIA_DXS_NO_VRA }, /* ASUS A7V600 & K8V*/ 
+		{ .subvendor = 0x1043, .subdevice = 0x812a, .action = VIA_DXS_SRC    }, /* ASUS A8V Deluxe */ 
+		{ .subvendor = 0x1071, .subdevice = 0x8375, .action = VIA_DXS_NO_VRA }, /* Vobis/Yakumo/Mitac notebook */
+		{ .subvendor = 0x1071, .subdevice = 0x8399, .action = VIA_DXS_ENABLE }, /* Umax AB 595T (VIA K8N800A - VT8237) */
+		{ .subvendor = 0x10cf, .subdevice = 0x118e, .action = VIA_DXS_ENABLE }, /* FSC laptop */
+		{ .subvendor = 0x1106, .subdevice = 0x4161, .action = VIA_DXS_NO_VRA }, /* ASRock K7VT2 */
+		{ .subvendor = 0x1106, .subdevice = 0x4552, .action = VIA_DXS_NO_VRA }, /* QDI Kudoz 7X/600-6AL */
+		{ .subvendor = 0x1106, .subdevice = 0xaa01, .action = VIA_DXS_NO_VRA }, /* EPIA MII */
+		{ .subvendor = 0x1106, .subdevice = 0xc001, .action = VIA_DXS_SRC }, /* Insight P4-ITX */
+		{ .subvendor = 0x1297, .subdevice = 0xa232, .action = VIA_DXS_ENABLE }, /* Shuttle ?? */
+		{ .subvendor = 0x1297, .subdevice = 0xc160, .action = VIA_DXS_ENABLE }, /* Shuttle SK41G */
+		{ .subvendor = 0x1458, .subdevice = 0xa002, .action = VIA_DXS_ENABLE }, /* Gigabyte GA-7VAXP */
+		{ .subvendor = 0x1462, .subdevice = 0x0080, .action = VIA_DXS_SRC }, /* MSI K8T Neo-FIS2R */
+		{ .subvendor = 0x1462, .subdevice = 0x3800, .action = VIA_DXS_ENABLE }, /* MSI KT266 */
+		{ .subvendor = 0x1462, .subdevice = 0x5901, .action = VIA_DXS_NO_VRA }, /* MSI KT6 Delta-SR */
+		{ .subvendor = 0x1462, .subdevice = 0x7023, .action = VIA_DXS_NO_VRA }, /* MSI K8T Neo2-FI */
+		{ .subvendor = 0x1462, .subdevice = 0x7120, .action = VIA_DXS_ENABLE }, /* MSI KT4V */
+		{ .subvendor = 0x147b, .subdevice = 0x1401, .action = VIA_DXS_ENABLE }, /* ABIT KD7(-RAID) */
+		{ .subvendor = 0x147b, .subdevice = 0x1411, .action = VIA_DXS_ENABLE }, /* ABIT VA-20 */
+		{ .subvendor = 0x147b, .subdevice = 0x1413, .action = VIA_DXS_ENABLE }, /* ABIT KV8 Pro */
+		{ .subvendor = 0x147b, .subdevice = 0x1415, .action = VIA_DXS_NO_VRA }, /* Abit AV8 */
+		{ .subvendor = 0x14ff, .subdevice = 0x0403, .action = VIA_DXS_ENABLE }, /* Twinhead mobo */
+		{ .subvendor = 0x14ff, .subdevice = 0x0408, .action = VIA_DXS_NO_VRA }, /* Twinhead mobo */
+		{ .subvendor = 0x1584, .subdevice = 0x8120, .action = VIA_DXS_ENABLE }, /* Gericom/Targa/Vobis/Uniwill laptop */
+		{ .subvendor = 0x1584, .subdevice = 0x8123, .action = VIA_DXS_NO_VRA }, /* Uniwill (Targa Visionary XP-210) */
+		{ .subvendor = 0x161f, .subdevice = 0x202b, .action = VIA_DXS_NO_VRA }, /* Amira Note book */
+		{ .subvendor = 0x161f, .subdevice = 0x2032, .action = VIA_DXS_48K }, /* m680x machines */
+		{ .subvendor = 0x1631, .subdevice = 0xe004, .action = VIA_DXS_ENABLE }, /* Easy Note 3174, Packard Bell */
+		{ .subvendor = 0x1695, .subdevice = 0x3005, .action = VIA_DXS_ENABLE }, /* EPoX EP-8K9A */
+		{ .subvendor = 0x1849, .subdevice = 0x3059, .action = VIA_DXS_NO_VRA }, /* ASRock K7VM2 */
 		{ } /* terminator */
 	};
 	struct dxs_whitelist *w;
@@ -2197,14 +2197,14 @@ static int __devinit check_dxs_list(struct pci_dev *pci)
 	pci_read_config_word(pci, PCI_SUBSYSTEM_VENDOR_ID, &subsystem_vendor);
 	pci_read_config_word(pci, PCI_SUBSYSTEM_ID, &subsystem_device);
 
-	for (w = whitelist; w->vendor; w++) {
-		if (w->vendor != subsystem_vendor)
+	for (w = whitelist; w->subvendor; w++) {
+		if (w->subvendor != subsystem_vendor)
 			continue;
 		if (w->mask) {
-			if ((w->mask & subsystem_device) == w->device)
+			if ((w->mask & subsystem_device) == w->subdevice)
 				return w->action;
 		} else {
-			if (subsystem_device == w->device)
+			if (subsystem_device == w->subdevice)
 				return w->action;
 		}
 	}
-- 
cgit v1.2.3-70-g09d2


From 763f356cd8de9e158836d236b3fd9dd149d696f9 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Fri, 3 Jun 2005 11:25:34 +0200
Subject: [ALSA] Add HDSP MADI driver

HDSPM driver,PCI drivers,RME9652 driver
Added RME Hammerfall DSP MADI driver by Winfried Ritsch.
(Moved from alsa-driver tree to mainline.)

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 include/sound/hdspm.h      |  131 ++
 sound/pci/Kconfig          |   13 +
 sound/pci/rme9652/Makefile |    2 +
 sound/pci/rme9652/hdspm.c  | 3671 ++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 3817 insertions(+)
 create mode 100644 include/sound/hdspm.h
 create mode 100644 sound/pci/rme9652/hdspm.c

(limited to 'include')

diff --git a/include/sound/hdspm.h b/include/sound/hdspm.h
new file mode 100644
index 00000000000..c34427ccd0b
--- /dev/null
+++ b/include/sound/hdspm.h
@@ -0,0 +1,131 @@
+#ifndef __SOUND_HDSPM_H		/* -*- linux-c -*- */
+#define __SOUND_HDSPM_H
+/*
+ *   Copyright (C) 2003 Winfried Ritsch (IEM)
+ *   based on hdsp.h from Thomas Charbonnel (thomas@undata.org)
+ *                      
+ *    
+ *   This program is free software; you can redistribute it and/or modify
+ *   it under the terms of the GNU General Public License as published by
+ *   the Free Software Foundation; either version 2 of the License, or
+ *   (at your option) any later version.
+ *
+ *   This program is distributed in the hope that it will be useful,
+ *   but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *   GNU General Public License for more details.
+ *
+ *   You should have received a copy of the GNU General Public License
+ *   along with this program; if not, write to the Free Software
+ *   Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+/* Maximum channels is 64 even on 56Mode you have 64playbacks to matrix */
+#define HDSPM_MAX_CHANNELS      64
+
+/* -------------------- IOCTL Peak/RMS Meters -------------------- */
+
+typedef struct _snd_hdspm_peak_rms hdspm_peak_rms_t;
+
+/* peam rms level structure like we get from hardware 
+  
+   maybe in future we can memory map it so I just copy it
+   to user on ioctl call now an dont change anything
+   rms are made out of low and high values
+   where (long) ????_rms = (????_rms_l >> 8) + ((????_rms_h & 0xFFFFFF00)<<24)
+   (i asume so from the code)
+*/
+
+struct _snd_hdspm_peak_rms {
+
+	unsigned int level_offset[1024];
+
+	unsigned int input_peak[64];
+	unsigned int playback_peak[64];
+	unsigned int output_peak[64];
+	unsigned int xxx_peak[64];	/* not used */
+
+	unsigned int reserved[256];	/* not used */
+
+	unsigned int input_rms_l[64];
+	unsigned int playback_rms_l[64];
+	unsigned int output_rms_l[64];
+	unsigned int xxx_rms_l[64];	/* not used */
+
+	unsigned int input_rms_h[64];
+	unsigned int playback_rms_h[64];
+	unsigned int output_rms_h[64];
+	unsigned int xxx_rms_h[64];	/* not used */
+};
+
+struct sndrv_hdspm_peak_rms_ioctl {
+	hdspm_peak_rms_t *peak;
+};
+
+/* use indirect access due to the limit of ioctl bit size */
+#define SNDRV_HDSPM_IOCTL_GET_PEAK_RMS _IOR('H', 0x40, struct sndrv_hdspm_peak_rms_ioctl)
+
+/* ------------ CONFIG block IOCTL ---------------------- */
+
+typedef struct _snd_hdspm_config_info hdspm_config_info_t;
+
+struct _snd_hdspm_config_info {
+	unsigned char pref_sync_ref;
+	unsigned char wordclock_sync_check;
+	unsigned char madi_sync_check;
+	unsigned int system_sample_rate;
+	unsigned int autosync_sample_rate;
+	unsigned char system_clock_mode;
+	unsigned char clock_source;
+	unsigned char autosync_ref;
+	unsigned char line_out;
+	unsigned int passthru;
+	unsigned int analog_out;
+};
+
+#define SNDRV_HDSPM_IOCTL_GET_CONFIG_INFO _IOR('H', 0x41, hdspm_config_info_t)
+
+
+/* get Soundcard Version */
+
+typedef struct _snd_hdspm_version hdspm_version_t;
+
+struct _snd_hdspm_version {
+	unsigned short firmware_rev;
+};
+
+#define SNDRV_HDSPM_IOCTL_GET_VERSION _IOR('H', 0x43, hdspm_version_t)
+
+
+/* ------------- get Matrix Mixer IOCTL --------------- */
+
+/* MADI mixer: 64inputs+64playback in 64outputs = 8192 => *4Byte = 32768 Bytes */
+
+/* organisation is 64 channelfader in a continous memory block */
+/* equivalent to hardware definition, maybe for future feature of mmap of them */
+/* each of 64 outputs has 64 infader and 64 outfader: 
+   Ins to Outs mixer[out].in[in], Outstreams to Outs mixer[out].pb[pb] */
+
+#define HDSPM_MIXER_CHANNELS HDSPM_MAX_CHANNELS
+
+typedef struct _snd_hdspm_channelfader snd_hdspm_channelfader_t;
+
+struct _snd_hdspm_channelfader {
+	unsigned int in[HDSPM_MIXER_CHANNELS];
+	unsigned int pb[HDSPM_MIXER_CHANNELS];
+};
+
+typedef struct _snd_hdspm_mixer hdspm_mixer_t;
+
+struct _snd_hdspm_mixer {
+	snd_hdspm_channelfader_t ch[HDSPM_MIXER_CHANNELS];
+};
+
+struct sndrv_hdspm_mixer_ioctl {
+	hdspm_mixer_t *mixer;
+};
+
+/* use indirect access due to the limit of ioctl bit size */
+#define SNDRV_HDSPM_IOCTL_GET_MIXER _IOR('H', 0x44, struct sndrv_hdspm_mixer_ioctl)
+
+#endif				/* __SOUND_HDSPM_H */
diff --git a/sound/pci/Kconfig b/sound/pci/Kconfig
index 428efdbd70a..6d7a00f34d8 100644
--- a/sound/pci/Kconfig
+++ b/sound/pci/Kconfig
@@ -274,6 +274,19 @@ config SND_HDSP
 	  To compile this driver as a module, choose M here: the module
 	  will be called snd-hdsp.
 
+config SND_HDSPM
+	tristate "RME Hammerfall DSP MADI"
+	depends on SND
+	select SND_HWDEP
+	select SND_RAWMIDI
+	select SND_PCM
+	help
+	  Say Y here to include support for RME Hammerfall DSP MADI
+	  soundcards.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called snd-hdspm.
+
 config SND_TRIDENT
 	tristate "Trident 4D-Wave DX/NX; SiS 7018"
 	depends on SND
diff --git a/sound/pci/rme9652/Makefile b/sound/pci/rme9652/Makefile
index 917374c9cd4..d2c294e136f 100644
--- a/sound/pci/rme9652/Makefile
+++ b/sound/pci/rme9652/Makefile
@@ -5,7 +5,9 @@
 
 snd-rme9652-objs := rme9652.o
 snd-hdsp-objs := hdsp.o
+snd-hdspm-objs := hdspm.o
 
 # Toplevel Module Dependency
 obj-$(CONFIG_SND_RME9652) += snd-rme9652.o
 obj-$(CONFIG_SND_HDSP) += snd-hdsp.o
+obj-$(CONFIG_SND_HDSPM) +=snd-hdspm.o
diff --git a/sound/pci/rme9652/hdspm.c b/sound/pci/rme9652/hdspm.c
new file mode 100644
index 00000000000..9e86d0eb41c
--- /dev/null
+++ b/sound/pci/rme9652/hdspm.c
@@ -0,0 +1,3671 @@
+/*   -*- linux-c -*-
+ *
+ *   ALSA driver for RME Hammerfall DSP MADI audio interface(s)
+ *
+ *      Copyright (c) 2003 Winfried Ritsch (IEM)
+ *      code based on hdsp.c   Paul Davis
+ *                             Marcus Andersson
+ *                             Thomas Charbonnel
+ *
+ *   This program is free software; you can redistribute it and/or modify
+ *   it under the terms of the GNU General Public License as published by
+ *   the Free Software Foundation; either version 2 of the License, or
+ *   (at your option) any later version.
+ *
+ *   This program is distributed in the hope that it will be useful,
+ *   but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *   GNU General Public License for more details.
+ *
+ *   You should have received a copy of the GNU General Public License
+ *   along with this program; if not, write to the Free Software
+ *   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ *
+ */
+#include <sound/driver.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/moduleparam.h>
+#include <linux/slab.h>
+#include <linux/pci.h>
+#include <asm/io.h>
+
+#include <sound/core.h>
+#include <sound/control.h>
+#include <sound/pcm.h>
+#include <sound/info.h>
+#include <sound/asoundef.h>
+#include <sound/rawmidi.h>
+#include <sound/hwdep.h>
+#include <sound/initval.h>
+
+#include <sound/hdspm.h>
+
+static int index[SNDRV_CARDS] = SNDRV_DEFAULT_IDX;	  /* Index 0-MAX */
+static char *id[SNDRV_CARDS] = SNDRV_DEFAULT_STR;	  /* ID for this card */
+static int enable[SNDRV_CARDS] = SNDRV_DEFAULT_ENABLE_PNP;/* Enable this card */
+
+/* Disable precise pointer at start */
+static int precise_ptr[SNDRV_CARDS];
+
+/* Send all playback to line outs */
+static int line_outs_monitor[SNDRV_CARDS];
+
+/* Enable Analog Outs on Channel 63/64 by default */
+static int enable_monitor[SNDRV_CARDS];
+
+module_param_array(index, int, NULL, 0444);
+MODULE_PARM_DESC(index, "Index value for RME HDSPM interface.");
+
+module_param_array(id, charp, NULL, 0444);
+MODULE_PARM_DESC(id, "ID string for RME HDSPM interface.");
+
+module_param_array(enable, bool, NULL, 0444);
+MODULE_PARM_DESC(enable, "Enable/disable specific HDSPM soundcards.");
+
+module_param_array(precise_ptr, bool, NULL, 0444);
+MODULE_PARM_DESC(precise_ptr, "Enable precise pointer, or disable.");
+
+module_param_array(line_outs_monitor, bool, NULL, 0444);
+MODULE_PARM_DESC(line_outs_monitor,
+		 "Send playback streams to analog outs by default.");
+
+module_param_array(enable_monitor, bool, NULL, 0444);
+MODULE_PARM_DESC(enable_monitor,
+		 "Enable Analog Out on Channel 63/64 by default.");
+
+MODULE_AUTHOR
+      ("Winfried Ritsch <ritsch_AT_iem.at>, Paul Davis <paul@linuxaudiosystems.com>, "
+       "Marcus Andersson, Thomas Charbonnel <thomas@undata.org>");
+MODULE_DESCRIPTION("RME HDSPM");
+MODULE_LICENSE("GPL");
+MODULE_SUPPORTED_DEVICE("{{RME HDSPM-MADI}}");
+
+/* --- Write registers. --- 
+  These are defined as byte-offsets from the iobase value.  */
+
+#define HDSPM_controlRegister	     64
+#define HDSPM_interruptConfirmation  96
+#define HDSPM_control2Reg	     256  /* not in specs ???????? */
+#define HDSPM_midiDataOut0  	     352  /* just believe in old code */
+#define HDSPM_midiDataOut1  	     356
+
+/* DMA enable for 64 channels, only Bit 0 is relevant */
+#define HDSPM_outputEnableBase       512  /* 512-767  input  DMA */ 
+#define HDSPM_inputEnableBase        768  /* 768-1023 output DMA */
+
+/* 16 page addresses for each of the 64 channels DMA buffer in and out 
+   (each 64k=16*4k) Buffer must be 4k aligned (which is default i386 ????) */
+#define HDSPM_pageAddressBufferOut       8192
+#define HDSPM_pageAddressBufferIn        (HDSPM_pageAddressBufferOut+64*16*4)
+
+#define HDSPM_MADI_mixerBase    32768	/* 32768-65535 for 2x64x64 Fader */
+
+#define HDSPM_MATRIX_MIXER_SIZE  8192	/* = 2*64*64 * 4 Byte => 32kB */
+
+/* --- Read registers. ---
+   These are defined as byte-offsets from the iobase value */
+#define HDSPM_statusRegister    0
+#define HDSPM_statusRegister2  96
+
+#define HDSPM_midiDataIn0     360
+#define HDSPM_midiDataIn1     364
+
+/* status is data bytes in MIDI-FIFO (0-128) */
+#define HDSPM_midiStatusOut0  384	
+#define HDSPM_midiStatusOut1  388	
+#define HDSPM_midiStatusIn0   392	
+#define HDSPM_midiStatusIn1   396	
+
+
+/* the meters are regular i/o-mapped registers, but offset
+   considerably from the rest. the peak registers are reset
+   when read; the least-significant 4 bits are full-scale counters; 
+   the actual peak value is in the most-significant 24 bits.
+*/
+#define HDSPM_MADI_peakrmsbase 	4096	/* 4096-8191 2x64x32Bit Meters */
+
+/* --- Control Register bits --------- */
+#define HDSPM_Start                (1<<0) /* start engine */
+
+#define HDSPM_Latency0             (1<<1) /* buffer size = 2^n */
+#define HDSPM_Latency1             (1<<2) /* where n is defined */
+#define HDSPM_Latency2             (1<<3) /* by Latency{2,1,0} */
+
+#define HDSPM_ClockModeMaster      (1<<4) /* 1=Master, 0=Slave/Autosync */
+
+#define HDSPM_AudioInterruptEnable (1<<5) /* what do you think ? */
+
+#define HDSPM_Frequency0  (1<<6)  /* 0=44.1kHz/88.2kHz 1=48kHz/96kHz */
+#define HDSPM_Frequency1  (1<<7)  /* 0=32kHz/64kHz */
+#define HDSPM_DoubleSpeed (1<<8)  /* 0=normal speed, 1=double speed */
+#define HDSPM_QuadSpeed   (1<<31) /* quad speed bit, not implemented now */
+
+#define HDSPM_TX_64ch     (1<<10) /* Output 64channel MODE=1,
+				     56channelMODE=0 */
+
+#define HDSPM_AutoInp     (1<<11) /* Auto Input (takeover) == Safe Mode, 
+                                     0=off, 1=on  */
+
+#define HDSPM_InputSelect0 (1<<14) /* Input select 0= optical, 1=coax */
+#define HDSPM_InputSelect1 (1<<15) /* should be 0 */
+
+#define HDSPM_SyncRef0     (1<<16) /* 0=WOrd, 1=MADI */
+#define HDSPM_SyncRef1     (1<<17) /* should be 0 */
+
+#define HDSPM_clr_tms      (1<<19) /* clear track marker, do not use 
+                                      AES additional bits in
+				      lower 5 Audiodatabits ??? */
+
+#define HDSPM_Midi0InterruptEnable (1<<22)
+#define HDSPM_Midi1InterruptEnable (1<<23)
+
+#define HDSPM_LineOut (1<<24) /* Analog Out on channel 63/64 on=1, mute=0 */
+
+
+/* --- bit helper defines */
+#define HDSPM_LatencyMask    (HDSPM_Latency0|HDSPM_Latency1|HDSPM_Latency2)
+#define HDSPM_FrequencyMask  (HDSPM_Frequency0|HDSPM_Frequency1)
+#define HDSPM_InputMask      (HDSPM_InputSelect0|HDSPM_InputSelect1)
+#define HDSPM_InputOptical   0
+#define HDSPM_InputCoaxial   (HDSPM_InputSelect0)
+#define HDSPM_SyncRefMask    (HDSPM_SyncRef0|HDSPM_SyncRef1)
+#define HDSPM_SyncRef_Word   0
+#define HDSPM_SyncRef_MADI   (HDSPM_SyncRef0)
+
+#define HDSPM_SYNC_FROM_WORD 0	/* Preferred sync reference */
+#define HDSPM_SYNC_FROM_MADI 1	/* choices - used by "pref_sync_ref" */
+
+#define HDSPM_Frequency32KHz    HDSPM_Frequency0
+#define HDSPM_Frequency44_1KHz  HDSPM_Frequency1
+#define HDSPM_Frequency48KHz   (HDSPM_Frequency1|HDSPM_Frequency0)
+#define HDSPM_Frequency64KHz   (HDSPM_DoubleSpeed|HDSPM_Frequency0)
+#define HDSPM_Frequency88_2KHz (HDSPM_DoubleSpeed|HDSPM_Frequency1)
+#define HDSPM_Frequency96KHz   (HDSPM_DoubleSpeed|HDSPM_Frequency1|HDSPM_Frequency0)
+
+/* --- for internal discrimination */
+#define HDSPM_CLOCK_SOURCE_AUTOSYNC          0	/* Sample Clock Sources */
+#define HDSPM_CLOCK_SOURCE_INTERNAL_32KHZ    1
+#define HDSPM_CLOCK_SOURCE_INTERNAL_44_1KHZ  2
+#define HDSPM_CLOCK_SOURCE_INTERNAL_48KHZ    3
+#define HDSPM_CLOCK_SOURCE_INTERNAL_64KHZ    4
+#define HDSPM_CLOCK_SOURCE_INTERNAL_88_2KHZ  5
+#define HDSPM_CLOCK_SOURCE_INTERNAL_96KHZ    6
+#define HDSPM_CLOCK_SOURCE_INTERNAL_128KHZ   7
+#define HDSPM_CLOCK_SOURCE_INTERNAL_176_4KHZ 8
+#define HDSPM_CLOCK_SOURCE_INTERNAL_192KHZ   9
+
+/* Synccheck Status */
+#define HDSPM_SYNC_CHECK_NO_LOCK 0
+#define HDSPM_SYNC_CHECK_LOCK    1
+#define HDSPM_SYNC_CHECK_SYNC	 2
+
+/* AutoSync References - used by "autosync_ref" control switch */
+#define HDSPM_AUTOSYNC_FROM_WORD      0
+#define HDSPM_AUTOSYNC_FROM_MADI      1
+#define HDSPM_AUTOSYNC_FROM_NONE      2
+
+/* Possible sources of MADI input */
+#define HDSPM_OPTICAL 0		/* optical   */
+#define HDSPM_COAXIAL 1		/* BNC */
+
+#define hdspm_encode_latency(x)       (((x)<<1) & HDSPM_LatencyMask)
+#define hdspm_decode_latency(x)       (((x) & HDSPM_LatencyMask)>>1)
+
+#define hdspm_encode_in(x) (((x)&0x3)<<14)
+#define hdspm_decode_in(x) (((x)>>14)&0x3)
+
+/* --- control2 register bits --- */
+#define HDSPM_TMS             (1<<0)
+#define HDSPM_TCK             (1<<1)
+#define HDSPM_TDI             (1<<2)
+#define HDSPM_JTAG            (1<<3)
+#define HDSPM_PWDN            (1<<4)
+#define HDSPM_PROGRAM	      (1<<5)
+#define HDSPM_CONFIG_MODE_0   (1<<6)
+#define HDSPM_CONFIG_MODE_1   (1<<7)
+/*#define HDSPM_VERSION_BIT     (1<<8) not defined any more*/
+#define HDSPM_BIGENDIAN_MODE  (1<<9)
+#define HDSPM_RD_MULTIPLE     (1<<10)
+
+/* --- Status Register bits --- */
+#define HDSPM_audioIRQPending    (1<<0)	/* IRQ is high and pending */
+#define HDSPM_RX_64ch            (1<<1)	/* Input 64chan. MODE=1, 56chn. MODE=0 */
+#define HDSPM_AB_int             (1<<2)	/* InputChannel Opt=0, Coax=1 (like inp0) */
+#define HDSPM_madiLock           (1<<3)	/* MADI Locked =1, no=0 */
+
+#define HDSPM_BufferPositionMask 0x000FFC0 /* Bit 6..15 : h/w buffer pointer */
+                                           /* since 64byte accurate last 6 bits 
+                                              are not used */
+
+#define HDSPM_madiSync          (1<<18) /* MADI is in sync */
+#define HDSPM_DoubleSpeedStatus (1<<19) /* (input) card in double speed */
+
+#define HDSPM_madiFreq0         (1<<22)	/* system freq 0=error */
+#define HDSPM_madiFreq1         (1<<23)	/* 1=32, 2=44.1 3=48 */
+#define HDSPM_madiFreq2         (1<<24)	/* 4=64, 5=88.2 6=96 */
+#define HDSPM_madiFreq3         (1<<25)	/* 7=128, 8=176.4 9=192 */
+
+#define HDSPM_BufferID          (1<<26)	/* (Double)Buffer ID toggles with Interrupt */
+#define HDSPM_midi0IRQPending   (1<<30)	/* MIDI IRQ is pending  */
+#define HDSPM_midi1IRQPending   (1<<31)	/* and aktiv */
+
+/* --- status bit helpers */
+#define HDSPM_madiFreqMask  (HDSPM_madiFreq0|HDSPM_madiFreq1|HDSPM_madiFreq2|HDSPM_madiFreq3)
+#define HDSPM_madiFreq32    (HDSPM_madiFreq0)
+#define HDSPM_madiFreq44_1  (HDSPM_madiFreq1)
+#define HDSPM_madiFreq48    (HDSPM_madiFreq0|HDSPM_madiFreq1)
+#define HDSPM_madiFreq64    (HDSPM_madiFreq2)
+#define HDSPM_madiFreq88_2  (HDSPM_madiFreq0|HDSPM_madiFreq2)
+#define HDSPM_madiFreq96    (HDSPM_madiFreq1|HDSPM_madiFreq2)
+#define HDSPM_madiFreq128   (HDSPM_madiFreq0|HDSPM_madiFreq1|HDSPM_madiFreq2)
+#define HDSPM_madiFreq176_4 (HDSPM_madiFreq3)
+#define HDSPM_madiFreq192   (HDSPM_madiFreq3|HDSPM_madiFreq0)
+
+/* Status2 Register bits */
+
+#define HDSPM_version0 (1<<0)	/* not realy defined but I guess */
+#define HDSPM_version1 (1<<1)	/* in former cards it was ??? */
+#define HDSPM_version2 (1<<2)
+
+#define HDSPM_wcLock (1<<3)	/* Wordclock is detected and locked */
+#define HDSPM_wcSync (1<<4)	/* Wordclock is in sync with systemclock */
+
+#define HDSPM_wc_freq0 (1<<5)	/* input freq detected via autosync  */
+#define HDSPM_wc_freq1 (1<<6)	/* 001=32, 010==44.1, 011=48, */
+#define HDSPM_wc_freq2 (1<<7)	/* 100=64, 101=88.2, 110=96, */
+/* missing Bit   for               111=128, 1000=176.4, 1001=192 */
+
+#define HDSPM_SelSyncRef0 (1<<8)	/* Sync Source in slave mode */
+#define HDSPM_SelSyncRef1 (1<<9)	/* 000=word, 001=MADI, */
+#define HDSPM_SelSyncRef2 (1<<10)	/* 111=no valid signal */
+
+#define HDSPM_wc_valid (HDSPM_wcLock|HDSPM_wcSync)
+
+#define HDSPM_wcFreqMask  (HDSPM_wc_freq0|HDSPM_wc_freq1|HDSPM_wc_freq2)
+#define HDSPM_wcFreq32    (HDSPM_wc_freq0)
+#define HDSPM_wcFreq44_1  (HDSPM_wc_freq1)
+#define HDSPM_wcFreq48    (HDSPM_wc_freq0|HDSPM_wc_freq1)
+#define HDSPM_wcFreq64    (HDSPM_wc_freq2)
+#define HDSPM_wcFreq88_2  (HDSPM_wc_freq0|HDSPM_wc_freq2)
+#define HDSPM_wcFreq96    (HDSPM_wc_freq1|HDSPM_wc_freq2)
+
+
+#define HDSPM_SelSyncRefMask       (HDSPM_SelSyncRef0|HDSPM_SelSyncRef1|HDSPM_SelSyncRef2)
+#define HDSPM_SelSyncRef_WORD      0
+#define HDSPM_SelSyncRef_MADI      (HDSPM_SelSyncRef0)
+#define HDSPM_SelSyncRef_NVALID    (HDSPM_SelSyncRef0|HDSPM_SelSyncRef1|HDSPM_SelSyncRef2)
+
+/* Mixer Values */
+#define UNITY_GAIN          32768	/* = 65536/2 */
+#define MINUS_INFINITY_GAIN 0
+
+/* PCI info */
+#ifndef PCI_VENDOR_ID_XILINX
+#define PCI_VENDOR_ID_XILINX		0x10ee
+#endif
+#ifndef PCI_DEVICE_ID_XILINX_HAMMERFALL_DSP
+#define PCI_DEVICE_ID_XILINX_HAMMERFALL_DSP 0x3fc5
+#endif
+#ifndef PCI_DEVICE_ID_XILINX_HAMMERFALL_DSP_MADI
+#define PCI_DEVICE_ID_XILINX_HAMMERFALL_DSP_MADI 0x3fc6
+#endif
+
+
+/* Number of channels for different Speed Modes */
+#define MADI_SS_CHANNELS       64
+#define MADI_DS_CHANNELS       32
+#define MADI_QS_CHANNELS       16
+
+/* the size of a substream (1 mono data stream) */
+#define HDSPM_CHANNEL_BUFFER_SAMPLES  (16*1024)
+#define HDSPM_CHANNEL_BUFFER_BYTES    (4*HDSPM_CHANNEL_BUFFER_SAMPLES)
+
+/* the size of the area we need to allocate for DMA transfers. the
+   size is the same regardless of the number of channels, and
+   also the latency to use. 
+   for one direction !!!
+*/
+#define HDSPM_DMA_AREA_BYTES (HDSPM_MAX_CHANNELS * HDSPM_CHANNEL_BUFFER_BYTES)
+#define HDSPM_DMA_AREA_KILOBYTES (HDSPM_DMA_AREA_BYTES/1024)
+
+typedef struct _hdspm hdspm_t;
+typedef struct _hdspm_midi hdspm_midi_t;
+
+struct _hdspm_midi {
+	hdspm_t *hdspm;
+	int id;
+	snd_rawmidi_t *rmidi;
+	snd_rawmidi_substream_t *input;
+	snd_rawmidi_substream_t *output;
+	char istimer;		/* timer in use */
+	struct timer_list timer;
+	spinlock_t lock;
+	int pending;
+};
+
+struct _hdspm {
+        spinlock_t lock;
+        snd_pcm_substream_t *capture_substream;	 /* only one playback */
+        snd_pcm_substream_t *playback_substream; /* and/or capture stream */
+
+	char *card_name;	     /* for procinfo */
+	unsigned short firmware_rev; /* dont know if relevant */
+
+	int precise_ptr;	/* use precise pointers, to be tested */
+	int monitor_outs;	/* set up monitoring outs init flag */
+
+	u32 control_register;	/* cached value */
+	u32 control2_register;	/* cached value */
+
+	hdspm_midi_t midi[2];
+	struct tasklet_struct midi_tasklet;
+
+	size_t period_bytes;
+	unsigned char ss_channels;	/* channels of card in single speed */
+	unsigned char ds_channels;	/* Double Speed */
+	unsigned char qs_channels;	/* Quad Speed */
+
+	unsigned char *playback_buffer;	/* suitably aligned address */
+	unsigned char *capture_buffer;	/* suitably aligned address */
+
+	pid_t capture_pid;	/* process id which uses capture */
+	pid_t playback_pid;	/* process id which uses capture */
+	int running;		/* running status */
+
+	int last_external_sample_rate;	/* samplerate mystic ... */
+	int last_internal_sample_rate;
+	int system_sample_rate;
+
+	char *channel_map;	/* channel map for DS and Quadspeed */
+
+	int dev;		/* Hardware vars... */
+	int irq;
+	unsigned long port;
+	void __iomem *iobase;
+
+	int irq_count;		/* for debug */
+
+	snd_card_t *card;	/* one card */
+	snd_pcm_t *pcm;		/* has one pcm */
+	snd_hwdep_t *hwdep;	/* and a hwdep for additional ioctl */
+	struct pci_dev *pci;	/* and an pci info */
+
+	/* Mixer vars */
+	snd_kcontrol_t *playback_mixer_ctls[HDSPM_MAX_CHANNELS];	/* fast alsa mixer */
+	snd_kcontrol_t *input_mixer_ctls[HDSPM_MAX_CHANNELS];	/* but input to much, so not used */
+	hdspm_mixer_t *mixer;	/* full mixer accessable over mixer ioctl or hwdep-device */
+
+};
+
+/* These tables map the ALSA channels 1..N to the channels that we
+   need to use in order to find the relevant channel buffer. RME
+   refer to this kind of mapping as between "the ADAT channel and
+   the DMA channel." We index it using the logical audio channel,
+   and the value is the DMA channel (i.e. channel buffer number)
+   where the data for that channel can be read/written from/to.
+*/
+
+static char channel_map_madi_ss[HDSPM_MAX_CHANNELS] = {
+   0, 1, 2, 3, 4, 5, 6, 7,
+   8, 9, 10, 11, 12, 13, 14, 15,
+   16, 17, 18, 19, 20, 21, 22, 23,
+   24, 25, 26, 27, 28, 29, 30, 31,
+   32, 33, 34, 35, 36, 37, 38, 39,
+   40, 41, 42, 43, 44, 45, 46, 47,
+   48, 49, 50, 51, 52, 53, 54, 55,
+   56, 57, 58, 59, 60, 61, 62, 63
+};
+
+static char channel_map_madi_ds[HDSPM_MAX_CHANNELS] = {
+  0, 2, 4, 6, 8, 10, 12, 14,
+  16, 18, 20, 22, 24, 26, 28, 30,
+  32, 34, 36, 38, 40, 42, 44, 46,
+  48, 50, 52, 54, 56, 58, 60, 62,
+  -1, -1, -1, -1, -1, -1, -1, -1,
+  -1, -1, -1, -1, -1, -1, -1, -1,
+  -1, -1, -1, -1, -1, -1, -1, -1,
+  -1, -1, -1, -1, -1, -1, -1, -1
+};
+
+static char channel_map_madi_qs[HDSPM_MAX_CHANNELS] = {
+  0,   4,  8, 12, 16, 20, 24,  28,  
+  32, 36, 40, 44, 48, 52, 56,  60
+  -1, -1, -1, -1, -1, -1, -1, -1,  
+  -1, -1, -1, -1, -1, -1, -1, -1,  
+  -1, -1, -1, -1, -1, -1, -1, -1, 
+  -1, -1, -1, -1, -1, -1, -1, -1, 
+  -1, -1, -1, -1, -1, -1, -1, -1, 
+  -1, -1, -1, -1, -1, -1, -1, -1
+};
+
+
+static struct pci_device_id snd_hdspm_ids[] = {
+	{
+	 .vendor = PCI_VENDOR_ID_XILINX,
+	 .device = PCI_DEVICE_ID_XILINX_HAMMERFALL_DSP_MADI,
+	 .subvendor = PCI_ANY_ID,
+	 .subdevice = PCI_ANY_ID,
+	 .class = 0,
+	 .class_mask = 0,
+	 .driver_data = 0},
+	{0,}
+};
+
+MODULE_DEVICE_TABLE(pci, snd_hdspm_ids);
+
+/* prototypes */
+static int __devinit snd_hdspm_create_alsa_devices(snd_card_t * card,
+						   hdspm_t * hdspm);
+static int __devinit snd_hdspm_create_pcm(snd_card_t * card,
+					  hdspm_t * hdspm);
+
+static inline void snd_hdspm_initialize_midi_flush(hdspm_t * hdspm);
+static int hdspm_update_simple_mixer_controls(hdspm_t * hdspm);
+static int hdspm_autosync_ref(hdspm_t * hdspm);
+static int snd_hdspm_set_defaults(hdspm_t * hdspm);
+static void hdspm_set_sgbuf(hdspm_t * hdspm, struct snd_sg_buf *sgbuf,
+			     unsigned int reg, int channels);
+
+/* Write/read to/from HDSPM with Adresses in Bytes
+   not words but only 32Bit writes are allowed */
+
+static inline void hdspm_write(hdspm_t * hdspm, unsigned int reg,
+			       unsigned int val)
+{
+	writel(val, hdspm->iobase + reg);
+}
+
+static inline unsigned int hdspm_read(hdspm_t * hdspm, unsigned int reg)
+{
+	return readl(hdspm->iobase + reg);
+}
+
+/* for each output channel (chan) I have an Input (in) and Playback (pb) Fader 
+   mixer is write only on hardware so we have to cache him for read 
+   each fader is a u32, but uses only the first 16 bit */
+
+static inline int hdspm_read_in_gain(hdspm_t * hdspm, unsigned int chan,
+				     unsigned int in)
+{
+	if (chan > HDSPM_MIXER_CHANNELS || in > HDSPM_MIXER_CHANNELS)
+		return 0;
+
+	return hdspm->mixer->ch[chan].in[in];
+}
+
+static inline int hdspm_read_pb_gain(hdspm_t * hdspm, unsigned int chan,
+				     unsigned int pb)
+{
+	if (chan > HDSPM_MIXER_CHANNELS || pb > HDSPM_MIXER_CHANNELS)
+		return 0;
+	return hdspm->mixer->ch[chan].pb[pb];
+}
+
+static inline int hdspm_write_in_gain(hdspm_t * hdspm, unsigned int chan,
+				      unsigned int in, unsigned short data)
+{
+	if (chan >= HDSPM_MIXER_CHANNELS || in >= HDSPM_MIXER_CHANNELS)
+		return -1;
+
+	hdspm_write(hdspm,
+		    HDSPM_MADI_mixerBase +
+		    ((in + 128 * chan) * sizeof(u32)),
+		    (hdspm->mixer->ch[chan].in[in] = data & 0xFFFF));
+	return 0;
+}
+
+static inline int hdspm_write_pb_gain(hdspm_t * hdspm, unsigned int chan,
+				      unsigned int pb, unsigned short data)
+{
+	if (chan >= HDSPM_MIXER_CHANNELS || pb >= HDSPM_MIXER_CHANNELS)
+		return -1;
+
+	hdspm_write(hdspm,
+		    HDSPM_MADI_mixerBase +
+		    ((64 + pb + 128 * chan) * sizeof(u32)),
+		    (hdspm->mixer->ch[chan].pb[pb] = data & 0xFFFF));
+	return 0;
+}
+
+
+/* enable DMA for specific channels, now available for DSP-MADI */
+static inline void snd_hdspm_enable_in(hdspm_t * hdspm, int i, int v)
+{
+	hdspm_write(hdspm, HDSPM_inputEnableBase + (4 * i), v);
+}
+
+static inline void snd_hdspm_enable_out(hdspm_t * hdspm, int i, int v)
+{
+	hdspm_write(hdspm, HDSPM_outputEnableBase + (4 * i), v);
+}
+
+/* check if same process is writing and reading */
+static inline int snd_hdspm_use_is_exclusive(hdspm_t * hdspm)
+{
+	unsigned long flags;
+	int ret = 1;
+
+	spin_lock_irqsave(&hdspm->lock, flags);
+	if ((hdspm->playback_pid != hdspm->capture_pid) &&
+	    (hdspm->playback_pid >= 0) && (hdspm->capture_pid >= 0)) {
+		ret = 0;
+	}
+	spin_unlock_irqrestore(&hdspm->lock, flags);
+	return ret;
+}
+
+/* check for external sample rate */
+static inline int hdspm_external_sample_rate(hdspm_t * hdspm)
+{
+	unsigned int status2 = hdspm_read(hdspm, HDSPM_statusRegister2);
+	unsigned int status = hdspm_read(hdspm, HDSPM_statusRegister);
+	unsigned int rate_bits;
+	int rate = 0;
+
+	/* if wordclock has synced freq and wordclock is valid */
+	if ((status2 & HDSPM_wcLock) != 0 &&
+	    (status & HDSPM_SelSyncRef0) == 0) {
+
+		rate_bits = status2 & HDSPM_wcFreqMask;
+
+		switch (rate_bits) {
+		case HDSPM_wcFreq32:
+			rate = 32000;
+			break;
+		case HDSPM_wcFreq44_1:
+			rate = 44100;
+			break;
+		case HDSPM_wcFreq48:
+			rate = 48000;
+			break;
+		case HDSPM_wcFreq64:
+			rate = 64000;
+			break;
+		case HDSPM_wcFreq88_2:
+			rate = 88200;
+			break;
+		case HDSPM_wcFreq96:
+			rate = 96000;
+			break;
+			/* Quadspeed Bit missing ???? */
+		default:
+			rate = 0;
+			break;
+		}
+	}
+
+	/* if rate detected and Syncref is Word than have it, word has priority to MADI */
+	if (rate != 0
+	    && (status2 & HDSPM_SelSyncRefMask) == HDSPM_SelSyncRef_WORD)
+		return rate;
+
+	/* maby a madi input (which is taken if sel sync is madi) */
+	if (status & HDSPM_madiLock) {
+		rate_bits = status & HDSPM_madiFreqMask;
+
+		switch (rate_bits) {
+		case HDSPM_madiFreq32:
+			rate = 32000;
+			break;
+		case HDSPM_madiFreq44_1:
+			rate = 44100;
+			break;
+		case HDSPM_madiFreq48:
+			rate = 48000;
+			break;
+		case HDSPM_madiFreq64:
+			rate = 64000;
+			break;
+		case HDSPM_madiFreq88_2:
+			rate = 88200;
+			break;
+		case HDSPM_madiFreq96:
+			rate = 96000;
+			break;
+		case HDSPM_madiFreq128:
+			rate = 128000;
+			break;
+		case HDSPM_madiFreq176_4:
+			rate = 176400;
+			break;
+		case HDSPM_madiFreq192:
+			rate = 192000;
+			break;
+		default:
+			rate = 0;
+			break;
+		}
+	}
+	return rate;
+}
+
+/* Latency function */
+static inline void hdspm_compute_period_size(hdspm_t * hdspm)
+{
+	hdspm->period_bytes =
+	    1 << ((hdspm_decode_latency(hdspm->control_register) + 8));
+}
+
+static snd_pcm_uframes_t hdspm_hw_pointer(hdspm_t * hdspm)
+{
+	int position;
+
+	position = hdspm_read(hdspm, HDSPM_statusRegister);
+
+	if (!hdspm->precise_ptr) {
+		return (position & HDSPM_BufferID) ? (hdspm->period_bytes /
+						      4) : 0;
+	}
+
+	/* hwpointer comes in bytes and is 64Bytes accurate (by docu since PCI Burst)
+	   i have experimented that it is at most 64 Byte to much for playing 
+	   so substraction of 64 byte should be ok for ALSA, but use it only
+	   for application where you know what you do since if you come to
+	   near with record pointer it can be a disaster */
+
+	position &= HDSPM_BufferPositionMask;
+	position = ((position - 64) % (2 * hdspm->period_bytes)) / 4;
+
+	return position;
+}
+
+
+static inline void hdspm_start_audio(hdspm_t * s)
+{
+	s->control_register |= (HDSPM_AudioInterruptEnable | HDSPM_Start);
+	hdspm_write(s, HDSPM_controlRegister, s->control_register);
+}
+
+static inline void hdspm_stop_audio(hdspm_t * s)
+{
+	s->control_register &= ~(HDSPM_Start | HDSPM_AudioInterruptEnable);
+	hdspm_write(s, HDSPM_controlRegister, s->control_register);
+}
+
+/* should I silence all or only opened ones ? doit all for first even is 4MB*/
+static inline void hdspm_silence_playback(hdspm_t * hdspm)
+{
+	int i;
+	int n = hdspm->period_bytes;
+	void *buf = hdspm->playback_buffer;
+
+	snd_assert(buf != NULL, return);
+
+	for (i = 0; i < HDSPM_MAX_CHANNELS; i++) {
+		memset(buf, 0, n);
+		buf += HDSPM_CHANNEL_BUFFER_BYTES;
+	}
+}
+
+static int hdspm_set_interrupt_interval(hdspm_t * s, unsigned int frames)
+{
+	int n;
+
+	spin_lock_irq(&s->lock);
+
+	frames >>= 7;
+	n = 0;
+	while (frames) {
+		n++;
+		frames >>= 1;
+	}
+	s->control_register &= ~HDSPM_LatencyMask;
+	s->control_register |= hdspm_encode_latency(n);
+
+	hdspm_write(s, HDSPM_controlRegister, s->control_register);
+
+	hdspm_compute_period_size(s);
+
+	spin_unlock_irq(&s->lock);
+
+	return 0;
+}
+
+
+/* dummy set rate lets see what happens */
+static int hdspm_set_rate(hdspm_t * hdspm, int rate, int called_internally)
+{
+	int reject_if_open = 0;
+	int current_rate;
+	int rate_bits;
+	int not_set = 0;
+
+	/* ASSUMPTION: hdspm->lock is either set, or there is no need for
+	   it (e.g. during module initialization).
+	 */
+
+	if (!(hdspm->control_register & HDSPM_ClockModeMaster)) {
+
+	        /* SLAVE --- */ 
+		if (called_internally) {
+
+        	  /* request from ctl or card initialization 
+	             just make a warning an remember setting 
+		     for future master mode switching */
+    
+			snd_printk
+			    (KERN_WARNING "HDSPM: Warning: device is not running as a clock master.\n");
+			not_set = 1;
+		} else {
+
+			/* hw_param request while in AutoSync mode */
+			int external_freq =
+			    hdspm_external_sample_rate(hdspm);
+
+			if ((hdspm_autosync_ref(hdspm) ==
+			     HDSPM_AUTOSYNC_FROM_NONE)) {
+
+				snd_printk(KERN_WARNING "HDSPM: Detected no Externel Sync \n");
+				not_set = 1;
+
+			} else if (rate != external_freq) {
+
+				snd_printk
+				    (KERN_WARNING "HDSPM: Warning: No AutoSync source for requested rate\n");
+				not_set = 1;
+			}
+		}
+	}
+
+	current_rate = hdspm->system_sample_rate;
+
+	/* Changing between Singe, Double and Quad speed is not
+	   allowed if any substreams are open. This is because such a change
+	   causes a shift in the location of the DMA buffers and a reduction
+	   in the number of available buffers.
+
+	   Note that a similar but essentially insoluble problem exists for
+	   externally-driven rate changes. All we can do is to flag rate
+	   changes in the read/write routines.  
+	 */
+
+	switch (rate) {
+	case 32000:
+		if (current_rate > 48000) {
+			reject_if_open = 1;
+		}
+		rate_bits = HDSPM_Frequency32KHz;
+		break;
+	case 44100:
+		if (current_rate > 48000) {
+			reject_if_open = 1;
+		}
+		rate_bits = HDSPM_Frequency44_1KHz;
+		break;
+	case 48000:
+		if (current_rate > 48000) {
+			reject_if_open = 1;
+		}
+		rate_bits = HDSPM_Frequency48KHz;
+		break;
+	case 64000:
+		if (current_rate <= 48000) {
+			reject_if_open = 1;
+		}
+		rate_bits = HDSPM_Frequency64KHz;
+		break;
+	case 88200:
+		if (current_rate <= 48000) {
+			reject_if_open = 1;
+		}
+		rate_bits = HDSPM_Frequency88_2KHz;
+		break;
+	case 96000:
+		if (current_rate <= 48000) {
+			reject_if_open = 1;
+		}
+		rate_bits = HDSPM_Frequency96KHz;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (reject_if_open
+	    && (hdspm->capture_pid >= 0 || hdspm->playback_pid >= 0)) {
+		snd_printk
+		    (KERN_ERR "HDSPM: cannot change between single- and double-speed mode (capture PID = %d, playback PID = %d)\n",
+		     hdspm->capture_pid, hdspm->playback_pid);
+		return -EBUSY;
+	}
+
+	hdspm->control_register &= ~HDSPM_FrequencyMask;
+	hdspm->control_register |= rate_bits;
+	hdspm_write(hdspm, HDSPM_controlRegister, hdspm->control_register);
+
+	if (rate > 64000)
+		hdspm->channel_map = channel_map_madi_qs;
+	else if (rate > 48000)
+		hdspm->channel_map = channel_map_madi_ds;
+	else 
+		hdspm->channel_map = channel_map_madi_ss;
+
+	hdspm->system_sample_rate = rate;
+
+	if (not_set != 0)
+		return -1;
+
+	return 0;
+}
+
+/* mainly for init to 0 on load */
+static void all_in_all_mixer(hdspm_t * hdspm, int sgain)
+{
+	int i, j;
+	unsigned int gain =
+	    (sgain > UNITY_GAIN) ? UNITY_GAIN : (sgain < 0) ? 0 : sgain;
+
+	for (i = 0; i < HDSPM_MIXER_CHANNELS; i++)
+		for (j = 0; j < HDSPM_MIXER_CHANNELS; j++) {
+			hdspm_write_in_gain(hdspm, i, j, gain);
+			hdspm_write_pb_gain(hdspm, i, j, gain);
+		}
+}
+
+/*----------------------------------------------------------------------------
+   MIDI
+  ----------------------------------------------------------------------------*/
+
+static inline unsigned char snd_hdspm_midi_read_byte (hdspm_t *hdspm, int id)
+{
+	/* the hardware already does the relevant bit-mask with 0xff */
+	if (id)
+		return hdspm_read(hdspm, HDSPM_midiDataIn1);
+	else
+		return hdspm_read(hdspm, HDSPM_midiDataIn0);
+}
+
+static inline void snd_hdspm_midi_write_byte (hdspm_t *hdspm, int id, int val)
+{
+	/* the hardware already does the relevant bit-mask with 0xff */
+	if (id)
+		return hdspm_write(hdspm, HDSPM_midiDataOut1, val);
+	else
+		return hdspm_write(hdspm, HDSPM_midiDataOut0, val);
+}
+
+static inline int snd_hdspm_midi_input_available (hdspm_t *hdspm, int id)
+{
+	if (id)
+		return (hdspm_read(hdspm, HDSPM_midiStatusIn1) & 0xff);
+	else
+		return (hdspm_read(hdspm, HDSPM_midiStatusIn0) & 0xff);
+}
+
+static inline int snd_hdspm_midi_output_possible (hdspm_t *hdspm, int id)
+{
+	int fifo_bytes_used;
+
+	if (id)
+		fifo_bytes_used = hdspm_read(hdspm, HDSPM_midiStatusOut1) & 0xff;
+	else
+		fifo_bytes_used = hdspm_read(hdspm, HDSPM_midiStatusOut0) & 0xff;
+
+	if (fifo_bytes_used < 128)
+		return  128 - fifo_bytes_used;
+	else
+		return 0;
+}
+
+static inline void snd_hdspm_flush_midi_input (hdspm_t *hdspm, int id)
+{
+	while (snd_hdspm_midi_input_available (hdspm, id))
+		snd_hdspm_midi_read_byte (hdspm, id);
+}
+
+static int snd_hdspm_midi_output_write (hdspm_midi_t *hmidi)
+{
+	unsigned long flags;
+	int n_pending;
+	int to_write;
+	int i;
+	unsigned char buf[128];
+
+	/* Output is not interrupt driven */
+		
+	spin_lock_irqsave (&hmidi->lock, flags);
+	if (hmidi->output) {
+		if (!snd_rawmidi_transmit_empty (hmidi->output)) {
+			if ((n_pending = snd_hdspm_midi_output_possible (hmidi->hdspm, hmidi->id)) > 0) {
+				if (n_pending > (int)sizeof (buf))
+					n_pending = sizeof (buf);
+				
+				if ((to_write = snd_rawmidi_transmit (hmidi->output, buf, n_pending)) > 0) {
+					for (i = 0; i < to_write; ++i) 
+						snd_hdspm_midi_write_byte (hmidi->hdspm, hmidi->id, buf[i]);
+				}
+			}
+		}
+	}
+	spin_unlock_irqrestore (&hmidi->lock, flags);
+	return 0;
+}
+
+static int snd_hdspm_midi_input_read (hdspm_midi_t *hmidi)
+{
+	unsigned char buf[128]; /* this buffer is designed to match the MIDI input FIFO size */
+	unsigned long flags;
+	int n_pending;
+	int i;
+
+	spin_lock_irqsave (&hmidi->lock, flags);
+	if ((n_pending = snd_hdspm_midi_input_available (hmidi->hdspm, hmidi->id)) > 0) {
+		if (hmidi->input) {
+			if (n_pending > (int)sizeof (buf)) {
+				n_pending = sizeof (buf);
+			}
+			for (i = 0; i < n_pending; ++i) {
+				buf[i] = snd_hdspm_midi_read_byte (hmidi->hdspm, hmidi->id);
+			}
+			if (n_pending) {
+				snd_rawmidi_receive (hmidi->input, buf, n_pending);
+			}
+		} else {
+			/* flush the MIDI input FIFO */
+			while (n_pending--) {
+				snd_hdspm_midi_read_byte (hmidi->hdspm, hmidi->id);
+			}
+		}
+	}
+	hmidi->pending = 0;
+	if (hmidi->id) {
+		hmidi->hdspm->control_register |= HDSPM_Midi1InterruptEnable;
+	} else {
+		hmidi->hdspm->control_register |= HDSPM_Midi0InterruptEnable;
+	}
+	hdspm_write(hmidi->hdspm, HDSPM_controlRegister, hmidi->hdspm->control_register);
+	spin_unlock_irqrestore (&hmidi->lock, flags);
+	return snd_hdspm_midi_output_write (hmidi);
+}
+
+static void snd_hdspm_midi_input_trigger(snd_rawmidi_substream_t * substream, int up)
+{
+	hdspm_t *hdspm;
+	hdspm_midi_t *hmidi;
+	unsigned long flags;
+	u32 ie;
+
+	hmidi = (hdspm_midi_t *) substream->rmidi->private_data;
+	hdspm = hmidi->hdspm;
+	ie = hmidi->id ? HDSPM_Midi1InterruptEnable : HDSPM_Midi0InterruptEnable;
+	spin_lock_irqsave (&hdspm->lock, flags);
+	if (up) {
+		if (!(hdspm->control_register & ie)) {
+			snd_hdspm_flush_midi_input (hdspm, hmidi->id);
+			hdspm->control_register |= ie;
+		}
+	} else {
+		hdspm->control_register &= ~ie;
+	}
+
+	hdspm_write(hdspm, HDSPM_controlRegister, hdspm->control_register);
+	spin_unlock_irqrestore (&hdspm->lock, flags);
+}
+
+static void snd_hdspm_midi_output_timer(unsigned long data)
+{
+	hdspm_midi_t *hmidi = (hdspm_midi_t *) data;
+	unsigned long flags;
+	
+	snd_hdspm_midi_output_write(hmidi);
+	spin_lock_irqsave (&hmidi->lock, flags);
+
+	/* this does not bump hmidi->istimer, because the
+	   kernel automatically removed the timer when it
+	   expired, and we are now adding it back, thus
+	   leaving istimer wherever it was set before.  
+	*/
+
+	if (hmidi->istimer) {
+		hmidi->timer.expires = 1 + jiffies;
+		add_timer(&hmidi->timer);
+	}
+
+	spin_unlock_irqrestore (&hmidi->lock, flags);
+}
+
+static void snd_hdspm_midi_output_trigger(snd_rawmidi_substream_t * substream, int up)
+{
+	hdspm_midi_t *hmidi;
+	unsigned long flags;
+
+	hmidi = (hdspm_midi_t *) substream->rmidi->private_data;
+	spin_lock_irqsave (&hmidi->lock, flags);
+	if (up) {
+		if (!hmidi->istimer) {
+			init_timer(&hmidi->timer);
+			hmidi->timer.function = snd_hdspm_midi_output_timer;
+			hmidi->timer.data = (unsigned long) hmidi;
+			hmidi->timer.expires = 1 + jiffies;
+			add_timer(&hmidi->timer);
+			hmidi->istimer++;
+		}
+	} else {
+		if (hmidi->istimer && --hmidi->istimer <= 0) {
+			del_timer (&hmidi->timer);
+		}
+	}
+	spin_unlock_irqrestore (&hmidi->lock, flags);
+	if (up)
+		snd_hdspm_midi_output_write(hmidi);
+}
+
+static int snd_hdspm_midi_input_open(snd_rawmidi_substream_t * substream)
+{
+	hdspm_midi_t *hmidi;
+
+	hmidi = (hdspm_midi_t *) substream->rmidi->private_data;
+	spin_lock_irq (&hmidi->lock);
+	snd_hdspm_flush_midi_input (hmidi->hdspm, hmidi->id);
+	hmidi->input = substream;
+	spin_unlock_irq (&hmidi->lock);
+
+	return 0;
+}
+
+static int snd_hdspm_midi_output_open(snd_rawmidi_substream_t * substream)
+{
+	hdspm_midi_t *hmidi;
+
+	hmidi = (hdspm_midi_t *) substream->rmidi->private_data;
+	spin_lock_irq (&hmidi->lock);
+	hmidi->output = substream;
+	spin_unlock_irq (&hmidi->lock);
+
+	return 0;
+}
+
+static int snd_hdspm_midi_input_close(snd_rawmidi_substream_t * substream)
+{
+	hdspm_midi_t *hmidi;
+
+	snd_hdspm_midi_input_trigger (substream, 0);
+
+	hmidi = (hdspm_midi_t *) substream->rmidi->private_data;
+	spin_lock_irq (&hmidi->lock);
+	hmidi->input = NULL;
+	spin_unlock_irq (&hmidi->lock);
+
+	return 0;
+}
+
+static int snd_hdspm_midi_output_close(snd_rawmidi_substream_t * substream)
+{
+	hdspm_midi_t *hmidi;
+
+	snd_hdspm_midi_output_trigger (substream, 0);
+
+	hmidi = (hdspm_midi_t *) substream->rmidi->private_data;
+	spin_lock_irq (&hmidi->lock);
+	hmidi->output = NULL;
+	spin_unlock_irq (&hmidi->lock);
+
+	return 0;
+}
+
+snd_rawmidi_ops_t snd_hdspm_midi_output =
+{
+	.open =		snd_hdspm_midi_output_open,
+	.close =	snd_hdspm_midi_output_close,
+	.trigger =	snd_hdspm_midi_output_trigger,
+};
+
+snd_rawmidi_ops_t snd_hdspm_midi_input =
+{
+	.open =		snd_hdspm_midi_input_open,
+	.close =	snd_hdspm_midi_input_close,
+	.trigger =	snd_hdspm_midi_input_trigger,
+};
+
+static int __devinit snd_hdspm_create_midi (snd_card_t *card, hdspm_t *hdspm, int id)
+{
+	int err;
+	char buf[32];
+
+	hdspm->midi[id].id = id;
+	hdspm->midi[id].rmidi = NULL;
+	hdspm->midi[id].input = NULL;
+	hdspm->midi[id].output = NULL;
+	hdspm->midi[id].hdspm = hdspm;
+	hdspm->midi[id].istimer = 0;
+	hdspm->midi[id].pending = 0;
+	spin_lock_init (&hdspm->midi[id].lock);
+
+	sprintf (buf, "%s MIDI %d", card->shortname, id+1);
+	if ((err = snd_rawmidi_new (card, buf, id, 1, 1, &hdspm->midi[id].rmidi)) < 0)
+		return err;
+
+	sprintf (hdspm->midi[id].rmidi->name, "%s MIDI %d", card->id, id+1);
+	hdspm->midi[id].rmidi->private_data = &hdspm->midi[id];
+
+	snd_rawmidi_set_ops (hdspm->midi[id].rmidi, SNDRV_RAWMIDI_STREAM_OUTPUT, &snd_hdspm_midi_output);
+	snd_rawmidi_set_ops (hdspm->midi[id].rmidi, SNDRV_RAWMIDI_STREAM_INPUT, &snd_hdspm_midi_input);
+
+	hdspm->midi[id].rmidi->info_flags |= SNDRV_RAWMIDI_INFO_OUTPUT |
+		SNDRV_RAWMIDI_INFO_INPUT |
+		SNDRV_RAWMIDI_INFO_DUPLEX;
+
+	return 0;
+}
+
+
+static void hdspm_midi_tasklet(unsigned long arg)
+{
+	hdspm_t *hdspm = (hdspm_t *)arg;
+	
+	if (hdspm->midi[0].pending)
+		snd_hdspm_midi_input_read (&hdspm->midi[0]);
+	if (hdspm->midi[1].pending)
+		snd_hdspm_midi_input_read (&hdspm->midi[1]);
+} 
+
+
+/*-----------------------------------------------------------------------------
+  Status Interface
+  ----------------------------------------------------------------------------*/
+
+/* get the system sample rate which is set */
+
+#define HDSPM_SYSTEM_SAMPLE_RATE(xname, xindex) \
+{ .iface = SNDRV_CTL_ELEM_IFACE_HWDEP, \
+  .name = xname, \
+  .index = xindex, \
+  .access = SNDRV_CTL_ELEM_ACCESS_READ, \
+  .info = snd_hdspm_info_system_sample_rate, \
+  .get = snd_hdspm_get_system_sample_rate \
+}
+
+static int snd_hdspm_info_system_sample_rate(snd_kcontrol_t * kcontrol,
+					     snd_ctl_elem_info_t * uinfo)
+{
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_INTEGER;
+	uinfo->count = 1;
+	return 0;
+}
+
+static int snd_hdspm_get_system_sample_rate(snd_kcontrol_t * kcontrol,
+					    snd_ctl_elem_value_t *
+					    ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+
+	ucontrol->value.enumerated.item[0] = hdspm->system_sample_rate;
+	return 0;
+}
+
+#define HDSPM_AUTOSYNC_SAMPLE_RATE(xname, xindex) \
+{ .iface = SNDRV_CTL_ELEM_IFACE_PCM, \
+  .name = xname, \
+  .index = xindex, \
+  .access = SNDRV_CTL_ELEM_ACCESS_READ, \
+  .info = snd_hdspm_info_autosync_sample_rate, \
+  .get = snd_hdspm_get_autosync_sample_rate \
+}
+
+static int snd_hdspm_info_autosync_sample_rate(snd_kcontrol_t * kcontrol,
+					       snd_ctl_elem_info_t * uinfo)
+{
+	static char *texts[] = { "32000", "44100", "48000",
+		"64000", "88200", "96000",
+		"128000", "176400", "192000",
+		"None"
+	};
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED;
+	uinfo->count = 1;
+	uinfo->value.enumerated.items = 10;
+	if (uinfo->value.enumerated.item >= uinfo->value.enumerated.items)
+		uinfo->value.enumerated.item =
+		    uinfo->value.enumerated.items - 1;
+	strcpy(uinfo->value.enumerated.name,
+	       texts[uinfo->value.enumerated.item]);
+	return 0;
+}
+
+static int snd_hdspm_get_autosync_sample_rate(snd_kcontrol_t * kcontrol,
+					      snd_ctl_elem_value_t *
+					      ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+
+	switch (hdspm_external_sample_rate(hdspm)) {
+	case 32000:
+		ucontrol->value.enumerated.item[0] = 0;
+		break;
+	case 44100:
+		ucontrol->value.enumerated.item[0] = 1;
+		break;
+	case 48000:
+		ucontrol->value.enumerated.item[0] = 2;
+		break;
+	case 64000:
+		ucontrol->value.enumerated.item[0] = 3;
+		break;
+	case 88200:
+		ucontrol->value.enumerated.item[0] = 4;
+		break;
+	case 96000:
+		ucontrol->value.enumerated.item[0] = 5;
+		break;
+	case 128000:
+		ucontrol->value.enumerated.item[0] = 6;
+		break;
+	case 176400:
+		ucontrol->value.enumerated.item[0] = 7;
+		break;
+	case 192000:
+		ucontrol->value.enumerated.item[0] = 8;
+		break;
+
+	default:
+		ucontrol->value.enumerated.item[0] = 9;
+	}
+	return 0;
+}
+
+#define HDSPM_SYSTEM_CLOCK_MODE(xname, xindex) \
+{ .iface = SNDRV_CTL_ELEM_IFACE_HWDEP, \
+  .name = xname, \
+  .index = xindex, \
+  .access = SNDRV_CTL_ELEM_ACCESS_READ, \
+  .info = snd_hdspm_info_system_clock_mode, \
+  .get = snd_hdspm_get_system_clock_mode, \
+}
+
+
+
+static int hdspm_system_clock_mode(hdspm_t * hdspm)
+{
+        /* Always reflect the hardware info, rme is never wrong !!!! */
+
+	if (hdspm->control_register & HDSPM_ClockModeMaster)
+		return 0;
+	return 1;
+}
+
+static int snd_hdspm_info_system_clock_mode(snd_kcontrol_t * kcontrol,
+					    snd_ctl_elem_info_t * uinfo)
+{
+	static char *texts[] = { "Master", "Slave" };
+
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED;
+	uinfo->count = 1;
+	uinfo->value.enumerated.items = 2;
+	if (uinfo->value.enumerated.item >= uinfo->value.enumerated.items)
+		uinfo->value.enumerated.item =
+		    uinfo->value.enumerated.items - 1;
+	strcpy(uinfo->value.enumerated.name,
+	       texts[uinfo->value.enumerated.item]);
+	return 0;
+}
+
+static int snd_hdspm_get_system_clock_mode(snd_kcontrol_t * kcontrol,
+					   snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+
+	ucontrol->value.enumerated.item[0] =
+	    hdspm_system_clock_mode(hdspm);
+	return 0;
+}
+
+#define HDSPM_CLOCK_SOURCE(xname, xindex) \
+{ .iface = SNDRV_CTL_ELEM_IFACE_PCM, \
+  .name = xname, \
+  .index = xindex, \
+  .info = snd_hdspm_info_clock_source, \
+  .get = snd_hdspm_get_clock_source, \
+  .put = snd_hdspm_put_clock_source \
+}
+
+static int hdspm_clock_source(hdspm_t * hdspm)
+{
+	if (hdspm->control_register & HDSPM_ClockModeMaster) {
+		switch (hdspm->system_sample_rate) {
+		case 32000:
+			return 1;
+		case 44100:
+			return 2;
+		case 48000:
+			return 3;
+		case 64000:
+			return 4;
+		case 88200:
+			return 5;
+		case 96000:
+			return 6;
+		case 128000:
+			return 7;
+		case 176400:
+			return 8;
+		case 192000:
+			return 9;
+		default:
+			return 3;
+		}
+	} else {
+		return 0;
+	}
+}
+
+static int hdspm_set_clock_source(hdspm_t * hdspm, int mode)
+{
+	int rate;
+	switch (mode) {
+
+	case HDSPM_CLOCK_SOURCE_AUTOSYNC:
+		if (hdspm_external_sample_rate(hdspm) != 0) {
+			hdspm->control_register &= ~HDSPM_ClockModeMaster;
+			hdspm_write(hdspm, HDSPM_controlRegister,
+				    hdspm->control_register);
+			return 0;
+		}
+		return -1;
+	case HDSPM_CLOCK_SOURCE_INTERNAL_32KHZ:
+		rate = 32000;
+		break;
+	case HDSPM_CLOCK_SOURCE_INTERNAL_44_1KHZ:
+		rate = 44100;
+		break;
+	case HDSPM_CLOCK_SOURCE_INTERNAL_48KHZ:
+		rate = 48000;
+		break;
+	case HDSPM_CLOCK_SOURCE_INTERNAL_64KHZ:
+		rate = 64000;
+		break;
+	case HDSPM_CLOCK_SOURCE_INTERNAL_88_2KHZ:
+		rate = 88200;
+		break;
+	case HDSPM_CLOCK_SOURCE_INTERNAL_96KHZ:
+		rate = 96000;
+		break;
+	case HDSPM_CLOCK_SOURCE_INTERNAL_128KHZ:
+		rate = 128000;
+		break;
+	case HDSPM_CLOCK_SOURCE_INTERNAL_176_4KHZ:
+		rate = 176400;
+		break;
+	case HDSPM_CLOCK_SOURCE_INTERNAL_192KHZ:
+		rate = 192000;
+		break;
+
+	default:
+		rate = 44100;
+	}
+	hdspm->control_register |= HDSPM_ClockModeMaster;
+	hdspm_write(hdspm, HDSPM_controlRegister, hdspm->control_register);
+	hdspm_set_rate(hdspm, rate, 1);
+	return 0;
+}
+
+static int snd_hdspm_info_clock_source(snd_kcontrol_t * kcontrol,
+				       snd_ctl_elem_info_t * uinfo)
+{
+	static char *texts[] = { "AutoSync",
+		"Internal 32.0 kHz", "Internal 44.1 kHz",
+		    "Internal 48.0 kHz",
+		"Internal 64.0 kHz", "Internal 88.2 kHz",
+		    "Internal 96.0 kHz",
+		"Internal 128.0 kHz", "Internal 176.4 kHz",
+		    "Internal 192.0 kHz"
+	};
+
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED;
+	uinfo->count = 1;
+	uinfo->value.enumerated.items = 10;
+
+	if (uinfo->value.enumerated.item >= uinfo->value.enumerated.items)
+		uinfo->value.enumerated.item =
+		    uinfo->value.enumerated.items - 1;
+
+	strcpy(uinfo->value.enumerated.name,
+	       texts[uinfo->value.enumerated.item]);
+
+	return 0;
+}
+
+static int snd_hdspm_get_clock_source(snd_kcontrol_t * kcontrol,
+				      snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+
+	ucontrol->value.enumerated.item[0] = hdspm_clock_source(hdspm);
+	return 0;
+}
+
+static int snd_hdspm_put_clock_source(snd_kcontrol_t * kcontrol,
+				      snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+	int change;
+	int val;
+
+	if (!snd_hdspm_use_is_exclusive(hdspm))
+		return -EBUSY;
+	val = ucontrol->value.enumerated.item[0];
+	if (val < 0)
+		val = 0;
+	if (val > 6)
+		val = 6;
+	spin_lock_irq(&hdspm->lock);
+	if (val != hdspm_clock_source(hdspm))
+		change = (hdspm_set_clock_source(hdspm, val) == 0) ? 1 : 0;
+	else
+		change = 0;
+	spin_unlock_irq(&hdspm->lock);
+	return change;
+}
+
+#define HDSPM_PREF_SYNC_REF(xname, xindex) \
+{ .iface = SNDRV_CTL_ELEM_IFACE_HWDEP, \
+  .name = xname, \
+  .index = xindex, \
+  .info = snd_hdspm_info_pref_sync_ref, \
+  .get = snd_hdspm_get_pref_sync_ref, \
+  .put = snd_hdspm_put_pref_sync_ref \
+}
+
+static int hdspm_pref_sync_ref(hdspm_t * hdspm)
+{
+	/* Notice that this looks at the requested sync source,
+	   not the one actually in use.
+	 */
+	switch (hdspm->control_register & HDSPM_SyncRefMask) {
+	case HDSPM_SyncRef_Word:
+		return HDSPM_SYNC_FROM_WORD;
+	case HDSPM_SyncRef_MADI:
+		return HDSPM_SYNC_FROM_MADI;
+	}
+
+	return HDSPM_SYNC_FROM_WORD;
+}
+
+static int hdspm_set_pref_sync_ref(hdspm_t * hdspm, int pref)
+{
+	hdspm->control_register &= ~HDSPM_SyncRefMask;
+
+	switch (pref) {
+	case HDSPM_SYNC_FROM_MADI:
+		hdspm->control_register |= HDSPM_SyncRef_MADI;
+		break;
+	case HDSPM_SYNC_FROM_WORD:
+		hdspm->control_register |= HDSPM_SyncRef_Word;
+		break;
+	default:
+		return -1;
+	}
+	hdspm_write(hdspm, HDSPM_controlRegister, hdspm->control_register);
+	return 0;
+}
+
+static int snd_hdspm_info_pref_sync_ref(snd_kcontrol_t * kcontrol,
+					snd_ctl_elem_info_t * uinfo)
+{
+	static char *texts[] = { "Word", "MADI" };
+
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED;
+	uinfo->count = 1;
+
+	uinfo->value.enumerated.items = 2;
+
+	if (uinfo->value.enumerated.item >= uinfo->value.enumerated.items)
+		uinfo->value.enumerated.item =
+		    uinfo->value.enumerated.items - 1;
+	strcpy(uinfo->value.enumerated.name,
+	       texts[uinfo->value.enumerated.item]);
+	return 0;
+}
+
+static int snd_hdspm_get_pref_sync_ref(snd_kcontrol_t * kcontrol,
+				       snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+
+	ucontrol->value.enumerated.item[0] = hdspm_pref_sync_ref(hdspm);
+	return 0;
+}
+
+static int snd_hdspm_put_pref_sync_ref(snd_kcontrol_t * kcontrol,
+				       snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+	int change, max;
+	unsigned int val;
+
+	max = 2;
+
+	if (!snd_hdspm_use_is_exclusive(hdspm))
+		return -EBUSY;
+
+	val = ucontrol->value.enumerated.item[0] % max;
+
+	spin_lock_irq(&hdspm->lock);
+	change = (int) val != hdspm_pref_sync_ref(hdspm);
+	hdspm_set_pref_sync_ref(hdspm, val);
+	spin_unlock_irq(&hdspm->lock);
+	return change;
+}
+
+#define HDSPM_AUTOSYNC_REF(xname, xindex) \
+{ .iface = SNDRV_CTL_ELEM_IFACE_HWDEP, \
+  .name = xname, \
+  .index = xindex, \
+  .access = SNDRV_CTL_ELEM_ACCESS_READ, \
+  .info = snd_hdspm_info_autosync_ref, \
+  .get = snd_hdspm_get_autosync_ref, \
+}
+
+static int hdspm_autosync_ref(hdspm_t * hdspm)
+{
+	/* This looks at the autosync selected sync reference */
+	unsigned int status2 = hdspm_read(hdspm, HDSPM_statusRegister2);
+
+	switch (status2 & HDSPM_SelSyncRefMask) {
+
+	case HDSPM_SelSyncRef_WORD:
+		return HDSPM_AUTOSYNC_FROM_WORD;
+
+	case HDSPM_SelSyncRef_MADI:
+		return HDSPM_AUTOSYNC_FROM_MADI;
+
+	case HDSPM_SelSyncRef_NVALID:
+		return HDSPM_AUTOSYNC_FROM_NONE;
+
+	default:
+		return 0;
+	}
+
+	return 0;
+}
+
+static int snd_hdspm_info_autosync_ref(snd_kcontrol_t * kcontrol,
+				       snd_ctl_elem_info_t * uinfo)
+{
+	static char *texts[] = { "WordClock", "MADI", "None" };
+
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED;
+	uinfo->count = 1;
+	uinfo->value.enumerated.items = 3;
+	if (uinfo->value.enumerated.item >= uinfo->value.enumerated.items)
+		uinfo->value.enumerated.item =
+		    uinfo->value.enumerated.items - 1;
+	strcpy(uinfo->value.enumerated.name,
+	       texts[uinfo->value.enumerated.item]);
+	return 0;
+}
+
+static int snd_hdspm_get_autosync_ref(snd_kcontrol_t * kcontrol,
+				      snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+
+	ucontrol->value.enumerated.item[0] = hdspm_pref_sync_ref(hdspm);
+	return 0;
+}
+
+#define HDSPM_LINE_OUT(xname, xindex) \
+{ .iface = SNDRV_CTL_ELEM_IFACE_HWDEP, \
+  .name = xname, \
+  .index = xindex, \
+  .info = snd_hdspm_info_line_out, \
+  .get = snd_hdspm_get_line_out, \
+  .put = snd_hdspm_put_line_out \
+}
+
+static int hdspm_line_out(hdspm_t * hdspm)
+{
+	return (hdspm->control_register & HDSPM_LineOut) ? 1 : 0;
+}
+
+
+static int hdspm_set_line_output(hdspm_t * hdspm, int out)
+{
+	if (out)
+		hdspm->control_register |= HDSPM_LineOut;
+	else
+		hdspm->control_register &= ~HDSPM_LineOut;
+	hdspm_write(hdspm, HDSPM_controlRegister, hdspm->control_register);
+
+	return 0;
+}
+
+static int snd_hdspm_info_line_out(snd_kcontrol_t * kcontrol,
+				   snd_ctl_elem_info_t * uinfo)
+{
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_BOOLEAN;
+	uinfo->count = 1;
+	uinfo->value.integer.min = 0;
+	uinfo->value.integer.max = 1;
+	return 0;
+}
+
+static int snd_hdspm_get_line_out(snd_kcontrol_t * kcontrol,
+				  snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+
+	spin_lock_irq(&hdspm->lock);
+	ucontrol->value.integer.value[0] = hdspm_line_out(hdspm);
+	spin_unlock_irq(&hdspm->lock);
+	return 0;
+}
+
+static int snd_hdspm_put_line_out(snd_kcontrol_t * kcontrol,
+				  snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+	int change;
+	unsigned int val;
+
+	if (!snd_hdspm_use_is_exclusive(hdspm))
+		return -EBUSY;
+	val = ucontrol->value.integer.value[0] & 1;
+	spin_lock_irq(&hdspm->lock);
+	change = (int) val != hdspm_line_out(hdspm);
+	hdspm_set_line_output(hdspm, val);
+	spin_unlock_irq(&hdspm->lock);
+	return change;
+}
+
+#define HDSPM_TX_64(xname, xindex) \
+{ .iface = SNDRV_CTL_ELEM_IFACE_HWDEP, \
+  .name = xname, \
+  .index = xindex, \
+  .info = snd_hdspm_info_tx_64, \
+  .get = snd_hdspm_get_tx_64, \
+  .put = snd_hdspm_put_tx_64 \
+}
+
+static int hdspm_tx_64(hdspm_t * hdspm)
+{
+	return (hdspm->control_register & HDSPM_TX_64ch) ? 1 : 0;
+}
+
+static int hdspm_set_tx_64(hdspm_t * hdspm, int out)
+{
+	if (out)
+		hdspm->control_register |= HDSPM_TX_64ch;
+	else
+		hdspm->control_register &= ~HDSPM_TX_64ch;
+	hdspm_write(hdspm, HDSPM_controlRegister, hdspm->control_register);
+
+	return 0;
+}
+
+static int snd_hdspm_info_tx_64(snd_kcontrol_t * kcontrol,
+				snd_ctl_elem_info_t * uinfo)
+{
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_BOOLEAN;
+	uinfo->count = 1;
+	uinfo->value.integer.min = 0;
+	uinfo->value.integer.max = 1;
+	return 0;
+}
+
+static int snd_hdspm_get_tx_64(snd_kcontrol_t * kcontrol,
+			       snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+
+	spin_lock_irq(&hdspm->lock);
+	ucontrol->value.integer.value[0] = hdspm_tx_64(hdspm);
+	spin_unlock_irq(&hdspm->lock);
+	return 0;
+}
+
+static int snd_hdspm_put_tx_64(snd_kcontrol_t * kcontrol,
+			       snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+	int change;
+	unsigned int val;
+
+	if (!snd_hdspm_use_is_exclusive(hdspm))
+		return -EBUSY;
+	val = ucontrol->value.integer.value[0] & 1;
+	spin_lock_irq(&hdspm->lock);
+	change = (int) val != hdspm_tx_64(hdspm);
+	hdspm_set_tx_64(hdspm, val);
+	spin_unlock_irq(&hdspm->lock);
+	return change;
+}
+
+#define HDSPM_C_TMS(xname, xindex) \
+{ .iface = SNDRV_CTL_ELEM_IFACE_HWDEP, \
+  .name = xname, \
+  .index = xindex, \
+  .info = snd_hdspm_info_c_tms, \
+  .get = snd_hdspm_get_c_tms, \
+  .put = snd_hdspm_put_c_tms \
+}
+
+static int hdspm_c_tms(hdspm_t * hdspm)
+{
+	return (hdspm->control_register & HDSPM_clr_tms) ? 1 : 0;
+}
+
+static int hdspm_set_c_tms(hdspm_t * hdspm, int out)
+{
+	if (out)
+		hdspm->control_register |= HDSPM_clr_tms;
+	else
+		hdspm->control_register &= ~HDSPM_clr_tms;
+	hdspm_write(hdspm, HDSPM_controlRegister, hdspm->control_register);
+
+	return 0;
+}
+
+static int snd_hdspm_info_c_tms(snd_kcontrol_t * kcontrol,
+				snd_ctl_elem_info_t * uinfo)
+{
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_BOOLEAN;
+	uinfo->count = 1;
+	uinfo->value.integer.min = 0;
+	uinfo->value.integer.max = 1;
+	return 0;
+}
+
+static int snd_hdspm_get_c_tms(snd_kcontrol_t * kcontrol,
+			       snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+
+	spin_lock_irq(&hdspm->lock);
+	ucontrol->value.integer.value[0] = hdspm_c_tms(hdspm);
+	spin_unlock_irq(&hdspm->lock);
+	return 0;
+}
+
+static int snd_hdspm_put_c_tms(snd_kcontrol_t * kcontrol,
+			       snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+	int change;
+	unsigned int val;
+
+	if (!snd_hdspm_use_is_exclusive(hdspm))
+		return -EBUSY;
+	val = ucontrol->value.integer.value[0] & 1;
+	spin_lock_irq(&hdspm->lock);
+	change = (int) val != hdspm_c_tms(hdspm);
+	hdspm_set_c_tms(hdspm, val);
+	spin_unlock_irq(&hdspm->lock);
+	return change;
+}
+
+#define HDSPM_SAFE_MODE(xname, xindex) \
+{ .iface = SNDRV_CTL_ELEM_IFACE_HWDEP, \
+  .name = xname, \
+  .index = xindex, \
+  .info = snd_hdspm_info_safe_mode, \
+  .get = snd_hdspm_get_safe_mode, \
+  .put = snd_hdspm_put_safe_mode \
+}
+
+static int hdspm_safe_mode(hdspm_t * hdspm)
+{
+	return (hdspm->control_register & HDSPM_AutoInp) ? 1 : 0;
+}
+
+static int hdspm_set_safe_mode(hdspm_t * hdspm, int out)
+{
+	if (out)
+		hdspm->control_register |= HDSPM_AutoInp;
+	else
+		hdspm->control_register &= ~HDSPM_AutoInp;
+	hdspm_write(hdspm, HDSPM_controlRegister, hdspm->control_register);
+
+	return 0;
+}
+
+static int snd_hdspm_info_safe_mode(snd_kcontrol_t * kcontrol,
+				    snd_ctl_elem_info_t * uinfo)
+{
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_BOOLEAN;
+	uinfo->count = 1;
+	uinfo->value.integer.min = 0;
+	uinfo->value.integer.max = 1;
+	return 0;
+}
+
+static int snd_hdspm_get_safe_mode(snd_kcontrol_t * kcontrol,
+				   snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+
+	spin_lock_irq(&hdspm->lock);
+	ucontrol->value.integer.value[0] = hdspm_safe_mode(hdspm);
+	spin_unlock_irq(&hdspm->lock);
+	return 0;
+}
+
+static int snd_hdspm_put_safe_mode(snd_kcontrol_t * kcontrol,
+				   snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+	int change;
+	unsigned int val;
+
+	if (!snd_hdspm_use_is_exclusive(hdspm))
+		return -EBUSY;
+	val = ucontrol->value.integer.value[0] & 1;
+	spin_lock_irq(&hdspm->lock);
+	change = (int) val != hdspm_safe_mode(hdspm);
+	hdspm_set_safe_mode(hdspm, val);
+	spin_unlock_irq(&hdspm->lock);
+	return change;
+}
+
+#define HDSPM_INPUT_SELECT(xname, xindex) \
+{ .iface = SNDRV_CTL_ELEM_IFACE_HWDEP, \
+  .name = xname, \
+  .index = xindex, \
+  .info = snd_hdspm_info_input_select, \
+  .get = snd_hdspm_get_input_select, \
+  .put = snd_hdspm_put_input_select \
+}
+
+static int hdspm_input_select(hdspm_t * hdspm)
+{
+	return (hdspm->control_register & HDSPM_InputSelect0) ? 1 : 0;
+}
+
+static int hdspm_set_input_select(hdspm_t * hdspm, int out)
+{
+	if (out)
+		hdspm->control_register |= HDSPM_InputSelect0;
+	else
+		hdspm->control_register &= ~HDSPM_InputSelect0;
+	hdspm_write(hdspm, HDSPM_controlRegister, hdspm->control_register);
+
+	return 0;
+}
+
+static int snd_hdspm_info_input_select(snd_kcontrol_t * kcontrol,
+				       snd_ctl_elem_info_t * uinfo)
+{
+	static char *texts[] = { "optical", "coaxial" };
+
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED;
+	uinfo->count = 1;
+	uinfo->value.enumerated.items = 2;
+
+	if (uinfo->value.enumerated.item >= uinfo->value.enumerated.items)
+		uinfo->value.enumerated.item =
+		    uinfo->value.enumerated.items - 1;
+	strcpy(uinfo->value.enumerated.name,
+	       texts[uinfo->value.enumerated.item]);
+
+	return 0;
+}
+
+static int snd_hdspm_get_input_select(snd_kcontrol_t * kcontrol,
+				      snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+
+	spin_lock_irq(&hdspm->lock);
+	ucontrol->value.enumerated.item[0] = hdspm_input_select(hdspm);
+	spin_unlock_irq(&hdspm->lock);
+	return 0;
+}
+
+static int snd_hdspm_put_input_select(snd_kcontrol_t * kcontrol,
+				      snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+	int change;
+	unsigned int val;
+
+	if (!snd_hdspm_use_is_exclusive(hdspm))
+		return -EBUSY;
+	val = ucontrol->value.integer.value[0] & 1;
+	spin_lock_irq(&hdspm->lock);
+	change = (int) val != hdspm_input_select(hdspm);
+	hdspm_set_input_select(hdspm, val);
+	spin_unlock_irq(&hdspm->lock);
+	return change;
+}
+
+/*           Simple Mixer
+  deprecated since to much faders ???
+  MIXER interface says output (source, destination, value)
+   where source > MAX_channels are playback channels 
+   on MADICARD 
+  - playback mixer matrix: [channelout+64] [output] [value]
+  - input(thru) mixer matrix: [channelin] [output] [value]
+  (better do 2 kontrols for seperation ?)
+*/
+
+#define HDSPM_MIXER(xname, xindex) \
+{ .iface = SNDRV_CTL_ELEM_IFACE_HWDEP, \
+  .name = xname, \
+  .index = xindex, \
+  .access = SNDRV_CTL_ELEM_ACCESS_READWRITE | \
+		 SNDRV_CTL_ELEM_ACCESS_VOLATILE, \
+  .info = snd_hdspm_info_mixer, \
+  .get = snd_hdspm_get_mixer, \
+  .put = snd_hdspm_put_mixer \
+}
+
+static int snd_hdspm_info_mixer(snd_kcontrol_t * kcontrol,
+				snd_ctl_elem_info_t * uinfo)
+{
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_INTEGER;
+	uinfo->count = 3;
+	uinfo->value.integer.min = 0;
+	uinfo->value.integer.max = 65535;
+	uinfo->value.integer.step = 1;
+	return 0;
+}
+
+static int snd_hdspm_get_mixer(snd_kcontrol_t * kcontrol,
+			       snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+	int source;
+	int destination;
+
+	source = ucontrol->value.integer.value[0];
+	if (source < 0)
+		source = 0;
+	else if (source >= 2 * HDSPM_MAX_CHANNELS)
+		source = 2 * HDSPM_MAX_CHANNELS - 1;
+
+	destination = ucontrol->value.integer.value[1];
+	if (destination < 0)
+		destination = 0;
+	else if (destination >= HDSPM_MAX_CHANNELS)
+		destination = HDSPM_MAX_CHANNELS - 1;
+
+	spin_lock_irq(&hdspm->lock);
+	if (source >= HDSPM_MAX_CHANNELS)
+		ucontrol->value.integer.value[2] =
+		    hdspm_read_pb_gain(hdspm, destination,
+				       source - HDSPM_MAX_CHANNELS);
+	else
+		ucontrol->value.integer.value[2] =
+		    hdspm_read_in_gain(hdspm, destination, source);
+
+	spin_unlock_irq(&hdspm->lock);
+
+	return 0;
+}
+
+static int snd_hdspm_put_mixer(snd_kcontrol_t * kcontrol,
+			       snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+	int change;
+	int source;
+	int destination;
+	int gain;
+
+	if (!snd_hdspm_use_is_exclusive(hdspm))
+		return -EBUSY;
+
+	source = ucontrol->value.integer.value[0];
+	destination = ucontrol->value.integer.value[1];
+
+	if (source < 0 || source >= 2 * HDSPM_MAX_CHANNELS)
+		return -1;
+	if (destination < 0 || destination >= HDSPM_MAX_CHANNELS)
+		return -1;
+
+	gain = ucontrol->value.integer.value[2];
+
+	spin_lock_irq(&hdspm->lock);
+
+	if (source >= HDSPM_MAX_CHANNELS)
+		change = gain != hdspm_read_pb_gain(hdspm, destination,
+						    source -
+						    HDSPM_MAX_CHANNELS);
+	else
+		change =
+		    gain != hdspm_read_in_gain(hdspm, destination, source);
+
+	if (change) {
+		if (source >= HDSPM_MAX_CHANNELS)
+			hdspm_write_pb_gain(hdspm, destination,
+					    source - HDSPM_MAX_CHANNELS,
+					    gain);
+		else
+			hdspm_write_in_gain(hdspm, destination, source,
+					    gain);
+	}
+	spin_unlock_irq(&hdspm->lock);
+
+	return change;
+}
+
+/* The simple mixer control(s) provide gain control for the
+   basic 1:1 mappings of playback streams to output
+   streams. 
+*/
+
+#define HDSPM_PLAYBACK_MIXER \
+{ .iface = SNDRV_CTL_ELEM_IFACE_MIXER, \
+  .access = SNDRV_CTL_ELEM_ACCESS_READ | SNDRV_CTL_ELEM_ACCESS_WRITE | \
+		 SNDRV_CTL_ELEM_ACCESS_VOLATILE, \
+  .info = snd_hdspm_info_playback_mixer, \
+  .get = snd_hdspm_get_playback_mixer, \
+  .put = snd_hdspm_put_playback_mixer \
+}
+
+static int snd_hdspm_info_playback_mixer(snd_kcontrol_t * kcontrol,
+					 snd_ctl_elem_info_t * uinfo)
+{
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_INTEGER;
+	uinfo->count = 1;
+	uinfo->value.integer.min = 0;
+	uinfo->value.integer.max = 65536;
+	uinfo->value.integer.step = 1;
+	return 0;
+}
+
+static int snd_hdspm_get_playback_mixer(snd_kcontrol_t * kcontrol,
+					snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+	int channel;
+	int mapped_channel;
+
+	channel = ucontrol->id.index - 1;
+
+	snd_assert(channel >= 0
+		   || channel < HDSPM_MAX_CHANNELS, return -EINVAL);
+
+	if ((mapped_channel = hdspm->channel_map[channel]) < 0)
+		return -EINVAL;
+
+	spin_lock_irq(&hdspm->lock);
+	ucontrol->value.integer.value[0] =
+	    hdspm_read_pb_gain(hdspm, mapped_channel, mapped_channel);
+	spin_unlock_irq(&hdspm->lock);
+
+	/*    snd_printdd("get pb mixer index %d, channel %d, mapped_channel %d, value %d\n",
+	   ucontrol->id.index,        channel, mapped_channel,  ucontrol->value.integer.value[0]); 
+	 */
+
+	return 0;
+}
+
+static int snd_hdspm_put_playback_mixer(snd_kcontrol_t * kcontrol,
+					snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+	int change;
+	int channel;
+	int mapped_channel;
+	int gain;
+
+	if (!snd_hdspm_use_is_exclusive(hdspm))
+		return -EBUSY;
+
+	channel = ucontrol->id.index - 1;
+
+	snd_assert(channel >= 0
+		   || channel < HDSPM_MAX_CHANNELS, return -EINVAL);
+
+	if ((mapped_channel = hdspm->channel_map[channel]) < 0)
+		return -EINVAL;
+
+	gain = ucontrol->value.integer.value[0];
+
+	spin_lock_irq(&hdspm->lock);
+	change =
+	    gain != hdspm_read_pb_gain(hdspm, mapped_channel,
+				       mapped_channel);
+	if (change)
+		hdspm_write_pb_gain(hdspm, mapped_channel, mapped_channel,
+				    gain);
+	spin_unlock_irq(&hdspm->lock);
+	return change;
+}
+
+#define HDSPM_WC_SYNC_CHECK(xname, xindex) \
+{ .iface = SNDRV_CTL_ELEM_IFACE_HWDEP, \
+  .name = xname, \
+  .index = xindex, \
+  .access = SNDRV_CTL_ELEM_ACCESS_READ | SNDRV_CTL_ELEM_ACCESS_VOLATILE, \
+  .info = snd_hdspm_info_sync_check, \
+  .get = snd_hdspm_get_wc_sync_check \
+}
+
+static int snd_hdspm_info_sync_check(snd_kcontrol_t * kcontrol,
+				     snd_ctl_elem_info_t * uinfo)
+{
+	static char *texts[] = { "No Lock", "Lock", "Sync" };
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED;
+	uinfo->count = 1;
+	uinfo->value.enumerated.items = 3;
+	if (uinfo->value.enumerated.item >= uinfo->value.enumerated.items)
+		uinfo->value.enumerated.item =
+		    uinfo->value.enumerated.items - 1;
+	strcpy(uinfo->value.enumerated.name,
+	       texts[uinfo->value.enumerated.item]);
+	return 0;
+}
+
+static int hdspm_wc_sync_check(hdspm_t * hdspm)
+{
+	int status2 = hdspm_read(hdspm, HDSPM_statusRegister2);
+	if (status2 & HDSPM_wcLock) {
+		if (status2 & HDSPM_wcSync)
+			return 2;
+		else
+			return 1;
+	}
+	return 0;
+}
+
+static int snd_hdspm_get_wc_sync_check(snd_kcontrol_t * kcontrol,
+				       snd_ctl_elem_value_t * ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+
+	ucontrol->value.enumerated.item[0] = hdspm_wc_sync_check(hdspm);
+	return 0;
+}
+
+
+#define HDSPM_MADI_SYNC_CHECK(xname, xindex) \
+{ .iface = SNDRV_CTL_ELEM_IFACE_HWDEP, \
+  .name = xname, \
+  .index = xindex, \
+  .access = SNDRV_CTL_ELEM_ACCESS_READ | SNDRV_CTL_ELEM_ACCESS_VOLATILE, \
+  .info = snd_hdspm_info_sync_check, \
+  .get = snd_hdspm_get_madisync_sync_check \
+}
+
+static int hdspm_madisync_sync_check(hdspm_t * hdspm)
+{
+	int status = hdspm_read(hdspm, HDSPM_statusRegister);
+	if (status & HDSPM_madiLock) {
+		if (status & HDSPM_madiSync)
+			return 2;
+		else
+			return 1;
+	}
+	return 0;
+}
+
+static int snd_hdspm_get_madisync_sync_check(snd_kcontrol_t * kcontrol,
+					     snd_ctl_elem_value_t *
+					     ucontrol)
+{
+	hdspm_t *hdspm = snd_kcontrol_chip(kcontrol);
+
+	ucontrol->value.enumerated.item[0] =
+	    hdspm_madisync_sync_check(hdspm);
+	return 0;
+}
+
+
+
+
+static snd_kcontrol_new_t snd_hdspm_controls[] = {
+
+	HDSPM_MIXER("Mixer", 0),
+/* 'Sample Clock Source' complies with the alsa control naming scheme */
+	HDSPM_CLOCK_SOURCE("Sample Clock Source", 0),
+
+	HDSPM_SYSTEM_CLOCK_MODE("System Clock Mode", 0),
+	HDSPM_PREF_SYNC_REF("Preferred Sync Reference", 0),
+	HDSPM_AUTOSYNC_REF("AutoSync Reference", 0),
+	HDSPM_SYSTEM_SAMPLE_RATE("System Sample Rate", 0),
+/* 'External Rate' complies with the alsa control naming scheme */
+	HDSPM_AUTOSYNC_SAMPLE_RATE("External Rate", 0),
+	HDSPM_WC_SYNC_CHECK("Word Clock Lock Status", 0),
+	HDSPM_MADI_SYNC_CHECK("MADI Sync Lock Status", 0),
+	HDSPM_LINE_OUT("Line Out", 0),
+	HDSPM_TX_64("TX 64 channels mode", 0),
+	HDSPM_C_TMS("Clear Track Marker", 0),
+	HDSPM_SAFE_MODE("Safe Mode", 0),
+	HDSPM_INPUT_SELECT("Input Select", 0),
+};
+
+static snd_kcontrol_new_t snd_hdspm_playback_mixer = HDSPM_PLAYBACK_MIXER;
+
+
+static int hdspm_update_simple_mixer_controls(hdspm_t * hdspm)
+{
+	int i;
+
+	for (i = hdspm->ds_channels; i < hdspm->ss_channels; ++i) {
+		if (hdspm->system_sample_rate > 48000) {
+			hdspm->playback_mixer_ctls[i]->vd[0].access =
+			    SNDRV_CTL_ELEM_ACCESS_INACTIVE |
+			    SNDRV_CTL_ELEM_ACCESS_READ |
+			    SNDRV_CTL_ELEM_ACCESS_VOLATILE;
+		} else {
+			hdspm->playback_mixer_ctls[i]->vd[0].access =
+			    SNDRV_CTL_ELEM_ACCESS_READWRITE |
+			    SNDRV_CTL_ELEM_ACCESS_VOLATILE;
+		}
+		snd_ctl_notify(hdspm->card, SNDRV_CTL_EVENT_MASK_VALUE |
+			       SNDRV_CTL_EVENT_MASK_INFO,
+			       &hdspm->playback_mixer_ctls[i]->id);
+	}
+
+	return 0;
+}
+
+
+static int snd_hdspm_create_controls(snd_card_t * card, hdspm_t * hdspm)
+{
+	unsigned int idx, limit;
+	int err;
+	snd_kcontrol_t *kctl;
+
+	/* add control list first */
+
+	for (idx = 0; idx < ARRAY_SIZE(snd_hdspm_controls); idx++) {
+		if ((err =
+		     snd_ctl_add(card, kctl =
+				 snd_ctl_new1(&snd_hdspm_controls[idx],
+					      hdspm))) < 0) {
+			return err;
+		}
+	}
+
+	/* Channel playback mixer as default control 
+	   Note: the whole matrix would be 128*HDSPM_MIXER_CHANNELS Faders, thats to big for any alsamixer 
+	   they are accesible via special IOCTL on hwdep
+	   and the mixer 2dimensional mixer control */
+
+	snd_hdspm_playback_mixer.name = "Chn";
+	limit = HDSPM_MAX_CHANNELS;
+
+	/* The index values are one greater than the channel ID so that alsamixer
+	   will display them correctly. We want to use the index for fast lookup
+	   of the relevant channel, but if we use it at all, most ALSA software
+	   does the wrong thing with it ...
+	 */
+
+	for (idx = 0; idx < limit; ++idx) {
+		snd_hdspm_playback_mixer.index = idx + 1;
+		if ((err = snd_ctl_add(card,
+				       kctl =
+				       snd_ctl_new1
+				       (&snd_hdspm_playback_mixer,
+					hdspm)))) {
+			return err;
+		}
+		hdspm->playback_mixer_ctls[idx] = kctl;
+	}
+
+	return 0;
+}
+
+/*------------------------------------------------------------
+   /proc interface 
+ ------------------------------------------------------------*/
+
+static void
+snd_hdspm_proc_read(snd_info_entry_t * entry, snd_info_buffer_t * buffer)
+{
+	hdspm_t *hdspm = (hdspm_t *) entry->private_data;
+	unsigned int status;
+	unsigned int status2;
+	char *pref_sync_ref;
+	char *autosync_ref;
+	char *system_clock_mode;
+	char *clock_source;
+	char *insel;
+	char *syncref;
+	int x, x2;
+
+	status = hdspm_read(hdspm, HDSPM_statusRegister);
+	status2 = hdspm_read(hdspm, HDSPM_statusRegister2);
+
+	snd_iprintf(buffer, "%s (Card #%d) Rev.%x Status2first3bits: %x\n",
+		    hdspm->card_name, hdspm->card->number + 1,
+		    hdspm->firmware_rev,
+		    (status2 & HDSPM_version0) |
+		    (status2 & HDSPM_version1) | (status2 &
+						  HDSPM_version2));
+
+	snd_iprintf(buffer, "IRQ: %d Registers bus: 0x%lx VM: 0x%lx\n",
+		    hdspm->irq, hdspm->port, (unsigned long)hdspm->iobase);
+
+	snd_iprintf(buffer, "--- System ---\n");
+
+	snd_iprintf(buffer,
+		    "IRQ Pending: Audio=%d, MIDI0=%d, MIDI1=%d, IRQcount=%d\n",
+		    status & HDSPM_audioIRQPending,
+		    (status & HDSPM_midi0IRQPending) ? 1 : 0,
+		    (status & HDSPM_midi1IRQPending) ? 1 : 0,
+		    hdspm->irq_count);
+	snd_iprintf(buffer,
+		    "HW pointer: id = %d, rawptr = %d (%d->%d) estimated= %ld (bytes)\n",
+		    ((status & HDSPM_BufferID) ? 1 : 0),
+		    (status & HDSPM_BufferPositionMask),
+		    (status & HDSPM_BufferPositionMask) % (2 *
+							   (int)hdspm->
+							   period_bytes),
+		    ((status & HDSPM_BufferPositionMask) -
+		     64) % (2 * (int)hdspm->period_bytes),
+		    (long) hdspm_hw_pointer(hdspm) * 4);
+
+	snd_iprintf(buffer,
+		    "MIDI FIFO: Out1=0x%x, Out2=0x%x, In1=0x%x, In2=0x%x \n",
+		    hdspm_read(hdspm, HDSPM_midiStatusOut0) & 0xFF,
+		    hdspm_read(hdspm, HDSPM_midiStatusOut1) & 0xFF,
+		    hdspm_read(hdspm, HDSPM_midiStatusIn0) & 0xFF,
+		    hdspm_read(hdspm, HDSPM_midiStatusIn1) & 0xFF);
+	snd_iprintf(buffer,
+		    "Register: ctrl1=0x%x, ctrl2=0x%x, status1=0x%x, status2=0x%x\n",
+		    hdspm->control_register, hdspm->control2_register,
+		    status, status2);
+
+	snd_iprintf(buffer, "--- Settings ---\n");
+
+	x = 1 << (6 +
+		  hdspm_decode_latency(hdspm->
+				       control_register &
+				       HDSPM_LatencyMask));
+
+	snd_iprintf(buffer,
+		    "Size (Latency): %d samples (2 periods of %lu bytes)\n",
+		    x, (unsigned long) hdspm->period_bytes);
+
+	snd_iprintf(buffer, "Line out: %s,   Precise Pointer: %s\n",
+		    (hdspm->
+		     control_register & HDSPM_LineOut) ? "on " : "off",
+		    (hdspm->precise_ptr) ? "on" : "off");
+
+	switch (hdspm->control_register & HDSPM_InputMask) {
+	case HDSPM_InputOptical:
+		insel = "Optical";
+		break;
+	case HDSPM_InputCoaxial:
+		insel = "Coaxial";
+		break;
+	default:
+		insel = "Unkown";
+	}
+
+	switch (hdspm->control_register & HDSPM_SyncRefMask) {
+	case HDSPM_SyncRef_Word:
+		syncref = "WordClock";
+		break;
+	case HDSPM_SyncRef_MADI:
+		syncref = "MADI";
+		break;
+	default:
+		syncref = "Unkown";
+	}
+	snd_iprintf(buffer, "Inputsel = %s, SyncRef = %s\n", insel,
+		    syncref);
+
+	snd_iprintf(buffer,
+		    "ClearTrackMarker = %s, Transmit in %s Channel Mode, Auto Input %s\n",
+		    (hdspm->
+		     control_register & HDSPM_clr_tms) ? "on" : "off",
+		    (hdspm->
+		     control_register & HDSPM_TX_64ch) ? "64" : "56",
+		    (hdspm->
+		     control_register & HDSPM_AutoInp) ? "on" : "off");
+
+	switch (hdspm_clock_source(hdspm)) {
+	case HDSPM_CLOCK_SOURCE_AUTOSYNC:
+		clock_source = "AutoSync";
+		break;
+	case HDSPM_CLOCK_SOURCE_INTERNAL_32KHZ:
+		clock_source = "Internal 32 kHz";
+		break;
+	case HDSPM_CLOCK_SOURCE_INTERNAL_44_1KHZ:
+		clock_source = "Internal 44.1 kHz";
+		break;
+	case HDSPM_CLOCK_SOURCE_INTERNAL_48KHZ:
+		clock_source = "Internal 48 kHz";
+		break;
+	case HDSPM_CLOCK_SOURCE_INTERNAL_64KHZ:
+		clock_source = "Internal 64 kHz";
+		break;
+	case HDSPM_CLOCK_SOURCE_INTERNAL_88_2KHZ:
+		clock_source = "Internal 88.2 kHz";
+		break;
+	case HDSPM_CLOCK_SOURCE_INTERNAL_96KHZ:
+		clock_source = "Internal 96 kHz";
+		break;
+	default:
+		clock_source = "Error";
+	}
+	snd_iprintf(buffer, "Sample Clock Source: %s\n", clock_source);
+	if (!(hdspm->control_register & HDSPM_ClockModeMaster)) {
+		system_clock_mode = "Slave";
+	} else {
+		system_clock_mode = "Master";
+	}
+	snd_iprintf(buffer, "System Clock Mode: %s\n", system_clock_mode);
+
+	switch (hdspm_pref_sync_ref(hdspm)) {
+	case HDSPM_SYNC_FROM_WORD:
+		pref_sync_ref = "Word Clock";
+		break;
+	case HDSPM_SYNC_FROM_MADI:
+		pref_sync_ref = "MADI Sync";
+		break;
+	default:
+		pref_sync_ref = "XXXX Clock";
+		break;
+	}
+	snd_iprintf(buffer, "Preferred Sync Reference: %s\n",
+		    pref_sync_ref);
+
+	snd_iprintf(buffer, "System Clock Frequency: %d\n",
+		    hdspm->system_sample_rate);
+
+
+	snd_iprintf(buffer, "--- Status:\n");
+
+	x = status & HDSPM_madiSync;
+	x2 = status2 & HDSPM_wcSync;
+
+	snd_iprintf(buffer, "Inputs MADI=%s, WordClock=%s\n",
+		    (status & HDSPM_madiLock) ? (x ? "Sync" : "Lock") :
+		    "NoLock",
+		    (status2 & HDSPM_wcLock) ? (x2 ? "Sync" : "Lock") :
+		    "NoLock");
+
+	switch (hdspm_autosync_ref(hdspm)) {
+	case HDSPM_AUTOSYNC_FROM_WORD:
+		autosync_ref = "Word Clock";
+		break;
+	case HDSPM_AUTOSYNC_FROM_MADI:
+		autosync_ref = "MADI Sync";
+		break;
+	case HDSPM_AUTOSYNC_FROM_NONE:
+		autosync_ref = "Input not valid";
+		break;
+	default:
+		autosync_ref = "---";
+		break;
+	}
+	snd_iprintf(buffer,
+		    "AutoSync: Reference= %s, Freq=%d (MADI = %d, Word = %d)\n",
+		    autosync_ref, hdspm_external_sample_rate(hdspm),
+		    (status & HDSPM_madiFreqMask) >> 22,
+		    (status2 & HDSPM_wcFreqMask) >> 5);
+
+	snd_iprintf(buffer, "Input: %s, Mode=%s\n",
+		    (status & HDSPM_AB_int) ? "Coax" : "Optical",
+		    (status & HDSPM_RX_64ch) ? "64 channels" :
+		    "56 channels");
+
+	snd_iprintf(buffer, "\n");
+}
+
+static void __devinit snd_hdspm_proc_init(hdspm_t * hdspm)
+{
+	snd_info_entry_t *entry;
+
+	if (!snd_card_proc_new(hdspm->card, "hdspm", &entry))
+		snd_info_set_text_ops(entry, hdspm, 1024,
+				      snd_hdspm_proc_read);
+}
+
+/*------------------------------------------------------------
+   hdspm intitialize 
+ ------------------------------------------------------------*/
+
+static int snd_hdspm_set_defaults(hdspm_t * hdspm)
+{
+	unsigned int i;
+
+	/* ASSUMPTION: hdspm->lock is either held, or there is no need to
+	   hold it (e.g. during module initalization).
+	 */
+
+	/* set defaults:       */
+
+	hdspm->control_register = HDSPM_ClockModeMaster |	/* Master Cloack Mode on */
+	    hdspm_encode_latency(7) |	/* latency maximum = 8192 samples */
+	    HDSPM_InputCoaxial |	/* Input Coax not Optical */
+	    HDSPM_SyncRef_MADI |	/* Madi is syncclock */
+	    HDSPM_LineOut |	/* Analog output in */
+	    HDSPM_TX_64ch |	/* transmit in 64ch mode */
+	    HDSPM_AutoInp;	/* AutoInput chossing (takeover) */
+
+	/* ! HDSPM_Frequency0|HDSPM_Frequency1 = 44.1khz */
+	/* !  HDSPM_DoubleSpeed HDSPM_QuadSpeed = normal speed */
+	/* ! HDSPM_clr_tms = do not clear bits in track marks */
+
+	hdspm_write(hdspm, HDSPM_controlRegister, hdspm->control_register);
+
+#ifdef SNDRV_BIG_ENDIAN
+	hdspm->control2_register = HDSPM_BIGENDIAN_MODE;
+#else
+	hdspm->control2_register = 0;
+#endif
+
+	hdspm_write(hdspm, HDSPM_control2Reg, hdspm->control2_register);
+	hdspm_compute_period_size(hdspm);
+
+	/* silence everything */
+
+	all_in_all_mixer(hdspm, 0 * UNITY_GAIN);
+
+	if (line_outs_monitor[hdspm->dev]) {
+
+		snd_printk(KERN_INFO "HDSPM: sending all playback streams to line outs.\n");
+
+		for (i = 0; i < HDSPM_MIXER_CHANNELS; i++) {
+			if (hdspm_write_pb_gain(hdspm, i, i, UNITY_GAIN))
+				return -EIO;
+		}
+	}
+
+	/* set a default rate so that the channel map is set up. */
+	hdspm->channel_map = channel_map_madi_ss;
+	hdspm_set_rate(hdspm, 44100, 1);
+
+	return 0;
+}
+
+
+/*------------------------------------------------------------
+   interupt 
+ ------------------------------------------------------------*/
+
+static irqreturn_t snd_hdspm_interrupt(int irq, void *dev_id,
+				       struct pt_regs *regs)
+{
+	hdspm_t *hdspm = (hdspm_t *) dev_id;
+	unsigned int status;
+	int audio;
+	int midi0;
+	int midi1;
+	unsigned int midi0status;
+	unsigned int midi1status;
+	int schedule = 0;
+
+	status = hdspm_read(hdspm, HDSPM_statusRegister);
+
+	audio = status & HDSPM_audioIRQPending;
+	midi0 = status & HDSPM_midi0IRQPending;
+	midi1 = status & HDSPM_midi1IRQPending;
+
+	if (!audio && !midi0 && !midi1)
+		return IRQ_NONE;
+
+	hdspm_write(hdspm, HDSPM_interruptConfirmation, 0);
+	hdspm->irq_count++;
+
+	midi0status = hdspm_read(hdspm, HDSPM_midiStatusIn0) & 0xff;
+	midi1status = hdspm_read(hdspm, HDSPM_midiStatusIn1) & 0xff;
+
+	if (audio) {
+
+		if (hdspm->capture_substream)
+			snd_pcm_period_elapsed(hdspm->pcm->
+					       streams
+					       [SNDRV_PCM_STREAM_CAPTURE].
+					       substream);
+
+		if (hdspm->playback_substream)
+			snd_pcm_period_elapsed(hdspm->pcm->
+					       streams
+					       [SNDRV_PCM_STREAM_PLAYBACK].
+					       substream);
+	}
+
+	if (midi0 && midi0status) {
+		/* we disable interrupts for this input until processing is done */
+		hdspm->control_register &= ~HDSPM_Midi0InterruptEnable;
+		hdspm_write(hdspm, HDSPM_controlRegister,
+			    hdspm->control_register);
+		hdspm->midi[0].pending = 1;
+		schedule = 1;
+	}
+	if (midi1 && midi1status) {
+		/* we disable interrupts for this input until processing is done */
+		hdspm->control_register &= ~HDSPM_Midi1InterruptEnable;
+		hdspm_write(hdspm, HDSPM_controlRegister,
+			    hdspm->control_register);
+		hdspm->midi[1].pending = 1;
+		schedule = 1;
+	}
+	if (schedule)
+		tasklet_hi_schedule(&hdspm->midi_tasklet);
+	return IRQ_HANDLED;
+}
+
+/*------------------------------------------------------------
+   pcm interface 
+  ------------------------------------------------------------*/
+
+
+static snd_pcm_uframes_t snd_hdspm_hw_pointer(snd_pcm_substream_t *
+					      substream)
+{
+	hdspm_t *hdspm = snd_pcm_substream_chip(substream);
+	return hdspm_hw_pointer(hdspm);
+}
+
+static char *hdspm_channel_buffer_location(hdspm_t * hdspm,
+					   int stream, int channel)
+{
+	int mapped_channel;
+
+	snd_assert(channel >= 0
+		   || channel < HDSPM_MAX_CHANNELS, return NULL);
+
+	if ((mapped_channel = hdspm->channel_map[channel]) < 0)
+		return NULL;
+
+	if (stream == SNDRV_PCM_STREAM_CAPTURE) {
+		return hdspm->capture_buffer +
+		    mapped_channel * HDSPM_CHANNEL_BUFFER_BYTES;
+	} else {
+		return hdspm->playback_buffer +
+		    mapped_channel * HDSPM_CHANNEL_BUFFER_BYTES;
+	}
+}
+
+
+/* dont know why need it ??? */
+static int snd_hdspm_playback_copy(snd_pcm_substream_t * substream,
+				   int channel, snd_pcm_uframes_t pos,
+				   void __user *src, snd_pcm_uframes_t count)
+{
+	hdspm_t *hdspm = snd_pcm_substream_chip(substream);
+	char *channel_buf;
+
+	snd_assert(pos + count <= HDSPM_CHANNEL_BUFFER_BYTES / 4,
+		   return -EINVAL);
+
+	channel_buf = hdspm_channel_buffer_location(hdspm,
+						    substream->pstr->
+						    stream, channel);
+
+	snd_assert(channel_buf != NULL, return -EIO);
+
+	return copy_from_user(channel_buf + pos * 4, src, count * 4);
+}
+
+static int snd_hdspm_capture_copy(snd_pcm_substream_t * substream,
+				  int channel, snd_pcm_uframes_t pos,
+				  void __user *dst, snd_pcm_uframes_t count)
+{
+	hdspm_t *hdspm = snd_pcm_substream_chip(substream);
+	char *channel_buf;
+
+	snd_assert(pos + count <= HDSPM_CHANNEL_BUFFER_BYTES / 4,
+		   return -EINVAL);
+
+	channel_buf = hdspm_channel_buffer_location(hdspm,
+						    substream->pstr->
+						    stream, channel);
+	snd_assert(channel_buf != NULL, return -EIO);
+	return copy_to_user(dst, channel_buf + pos * 4, count * 4);
+}
+
+static int snd_hdspm_hw_silence(snd_pcm_substream_t * substream,
+				int channel, snd_pcm_uframes_t pos,
+				snd_pcm_uframes_t count)
+{
+	hdspm_t *hdspm = snd_pcm_substream_chip(substream);
+	char *channel_buf;
+
+	channel_buf =
+	    hdspm_channel_buffer_location(hdspm, substream->pstr->stream,
+					  channel);
+	snd_assert(channel_buf != NULL, return -EIO);
+	memset(channel_buf + pos * 4, 0, count * 4);
+	return 0;
+}
+
+static int snd_hdspm_reset(snd_pcm_substream_t * substream)
+{
+	snd_pcm_runtime_t *runtime = substream->runtime;
+	hdspm_t *hdspm = snd_pcm_substream_chip(substream);
+	snd_pcm_substream_t *other;
+
+	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
+		other = hdspm->capture_substream;
+	else
+		other = hdspm->playback_substream;
+
+	if (hdspm->running)
+		runtime->status->hw_ptr = hdspm_hw_pointer(hdspm);
+	else
+		runtime->status->hw_ptr = 0;
+	if (other) {
+		struct list_head *pos;
+		snd_pcm_substream_t *s;
+		snd_pcm_runtime_t *oruntime = other->runtime;
+		snd_pcm_group_for_each(pos, substream) {
+			s = snd_pcm_group_substream_entry(pos);
+			if (s == other) {
+				oruntime->status->hw_ptr =
+				    runtime->status->hw_ptr;
+				break;
+			}
+		}
+	}
+	return 0;
+}
+
+static int snd_hdspm_hw_params(snd_pcm_substream_t * substream,
+			       snd_pcm_hw_params_t * params)
+{
+	hdspm_t *hdspm = snd_pcm_substream_chip(substream);
+	int err;
+	int i;
+	pid_t this_pid;
+	pid_t other_pid;
+	struct snd_sg_buf *sgbuf;
+
+
+	spin_lock_irq(&hdspm->lock);
+
+	if (substream->pstr->stream == SNDRV_PCM_STREAM_PLAYBACK) {
+		this_pid = hdspm->playback_pid;
+		other_pid = hdspm->capture_pid;
+	} else {
+		this_pid = hdspm->capture_pid;
+		other_pid = hdspm->playback_pid;
+	}
+
+	if ((other_pid > 0) && (this_pid != other_pid)) {
+
+		/* The other stream is open, and not by the same
+		   task as this one. Make sure that the parameters
+		   that matter are the same.
+		 */
+
+		if (params_rate(params) != hdspm->system_sample_rate) {
+			spin_unlock_irq(&hdspm->lock);
+			_snd_pcm_hw_param_setempty(params,
+						   SNDRV_PCM_HW_PARAM_RATE);
+			return -EBUSY;
+		}
+
+		if (params_period_size(params) != hdspm->period_bytes / 4) {
+			spin_unlock_irq(&hdspm->lock);
+			_snd_pcm_hw_param_setempty(params,
+						   SNDRV_PCM_HW_PARAM_PERIOD_SIZE);
+			return -EBUSY;
+		}
+
+	}
+	/* We're fine. */
+	spin_unlock_irq(&hdspm->lock);
+
+	/* how to make sure that the rate matches an externally-set one ?   */
+
+	spin_lock_irq(&hdspm->lock);
+	if ((err = hdspm_set_rate(hdspm, params_rate(params), 0)) < 0) {
+		spin_unlock_irq(&hdspm->lock);
+		_snd_pcm_hw_param_setempty(params,
+					   SNDRV_PCM_HW_PARAM_RATE);
+		return err;
+	}
+	spin_unlock_irq(&hdspm->lock);
+
+	if ((err =
+	     hdspm_set_interrupt_interval(hdspm,
+					  params_period_size(params))) <
+	    0) {
+		_snd_pcm_hw_param_setempty(params,
+					   SNDRV_PCM_HW_PARAM_PERIOD_SIZE);
+		return err;
+	}
+
+	/* Memory allocation, takashi's method, dont know if we should spinlock  */
+	/* malloc all buffer even if not enabled to get sure */
+	/* malloc only needed bytes */
+	err =
+	    snd_pcm_lib_malloc_pages(substream,
+				     HDSPM_CHANNEL_BUFFER_BYTES *
+				     params_channels(params));
+	if (err < 0)
+		return err;
+
+	sgbuf = snd_pcm_substream_sgbuf(substream);
+
+	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) {
+
+		hdspm_set_sgbuf(hdspm, sgbuf, HDSPM_pageAddressBufferOut,
+				params_channels(params));
+
+		for (i = 0; i < params_channels(params); ++i)
+			snd_hdspm_enable_out(hdspm, i, 1);
+
+		hdspm->playback_buffer =
+		    (unsigned char *) substream->runtime->dma_area;
+	} else {
+		hdspm_set_sgbuf(hdspm, sgbuf, HDSPM_pageAddressBufferIn,
+				params_channels(params));
+
+		for (i = 0; i < params_channels(params); ++i)
+			snd_hdspm_enable_in(hdspm, i, 1);
+
+		hdspm->capture_buffer =
+		    (unsigned char *) substream->runtime->dma_area;
+	}
+	return 0;
+}
+
+static int snd_hdspm_hw_free(snd_pcm_substream_t * substream)
+{
+	int i;
+	hdspm_t *hdspm = snd_pcm_substream_chip(substream);
+
+	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) {
+
+		/* params_channels(params) should be enough, 
+		   but to get sure in case of error */
+		for (i = 0; i < HDSPM_MAX_CHANNELS; ++i)
+			snd_hdspm_enable_out(hdspm, i, 0);
+
+		hdspm->playback_buffer = NULL;
+	} else {
+		for (i = 0; i < HDSPM_MAX_CHANNELS; ++i)
+			snd_hdspm_enable_in(hdspm, i, 0);
+
+		hdspm->capture_buffer = NULL;
+
+	}
+
+	snd_pcm_lib_free_pages(substream);
+
+	return 0;
+}
+
+static int snd_hdspm_channel_info(snd_pcm_substream_t * substream,
+				  snd_pcm_channel_info_t * info)
+{
+	hdspm_t *hdspm = snd_pcm_substream_chip(substream);
+	int mapped_channel;
+
+	snd_assert(info->channel < HDSPM_MAX_CHANNELS, return -EINVAL);
+
+	if ((mapped_channel = hdspm->channel_map[info->channel]) < 0)
+		return -EINVAL;
+
+	info->offset = mapped_channel * HDSPM_CHANNEL_BUFFER_BYTES;
+	info->first = 0;
+	info->step = 32;
+	return 0;
+}
+
+static int snd_hdspm_ioctl(snd_pcm_substream_t * substream,
+			   unsigned int cmd, void *arg)
+{
+	switch (cmd) {
+	case SNDRV_PCM_IOCTL1_RESET:
+		{
+			return snd_hdspm_reset(substream);
+		}
+
+	case SNDRV_PCM_IOCTL1_CHANNEL_INFO:
+		{
+			snd_pcm_channel_info_t *info = arg;
+			return snd_hdspm_channel_info(substream, info);
+		}
+	default:
+		break;
+	}
+
+	return snd_pcm_lib_ioctl(substream, cmd, arg);
+}
+
+static int snd_hdspm_trigger(snd_pcm_substream_t * substream, int cmd)
+{
+	hdspm_t *hdspm = snd_pcm_substream_chip(substream);
+	snd_pcm_substream_t *other;
+	int running;
+
+	spin_lock(&hdspm->lock);
+	running = hdspm->running;
+	switch (cmd) {
+	case SNDRV_PCM_TRIGGER_START:
+		running |= 1 << substream->stream;
+		break;
+	case SNDRV_PCM_TRIGGER_STOP:
+		running &= ~(1 << substream->stream);
+		break;
+	default:
+		snd_BUG();
+		spin_unlock(&hdspm->lock);
+		return -EINVAL;
+	}
+	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
+		other = hdspm->capture_substream;
+	else
+		other = hdspm->playback_substream;
+
+	if (other) {
+		struct list_head *pos;
+		snd_pcm_substream_t *s;
+		snd_pcm_group_for_each(pos, substream) {
+			s = snd_pcm_group_substream_entry(pos);
+			if (s == other) {
+				snd_pcm_trigger_done(s, substream);
+				if (cmd == SNDRV_PCM_TRIGGER_START)
+					running |= 1 << s->stream;
+				else
+					running &= ~(1 << s->stream);
+				goto _ok;
+			}
+		}
+		if (cmd == SNDRV_PCM_TRIGGER_START) {
+			if (!(running & (1 << SNDRV_PCM_STREAM_PLAYBACK))
+			    && substream->stream ==
+			    SNDRV_PCM_STREAM_CAPTURE)
+				hdspm_silence_playback(hdspm);
+		} else {
+			if (running &&
+			    substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
+				hdspm_silence_playback(hdspm);
+		}
+	} else {
+		if (substream->stream == SNDRV_PCM_STREAM_CAPTURE)
+			hdspm_silence_playback(hdspm);
+	}
+      _ok:
+	snd_pcm_trigger_done(substream, substream);
+	if (!hdspm->running && running)
+		hdspm_start_audio(hdspm);
+	else if (hdspm->running && !running)
+		hdspm_stop_audio(hdspm);
+	hdspm->running = running;
+	spin_unlock(&hdspm->lock);
+
+	return 0;
+}
+
+static int snd_hdspm_prepare(snd_pcm_substream_t * substream)
+{
+	return 0;
+}
+
+static unsigned int period_sizes[] =
+    { 64, 128, 256, 512, 1024, 2048, 4096, 8192 };
+
+static snd_pcm_hardware_t snd_hdspm_playback_subinfo = {
+	.info = (SNDRV_PCM_INFO_MMAP |
+		 SNDRV_PCM_INFO_MMAP_VALID |
+		 SNDRV_PCM_INFO_NONINTERLEAVED |
+		 SNDRV_PCM_INFO_SYNC_START | SNDRV_PCM_INFO_DOUBLE),
+	.formats = SNDRV_PCM_FMTBIT_S32_LE,
+	.rates = (SNDRV_PCM_RATE_32000 |
+		  SNDRV_PCM_RATE_44100 |
+		  SNDRV_PCM_RATE_48000 |
+		  SNDRV_PCM_RATE_64000 |
+		  SNDRV_PCM_RATE_88200 | SNDRV_PCM_RATE_96000),
+	.rate_min = 32000,
+	.rate_max = 96000,
+	.channels_min = 1,
+	.channels_max = HDSPM_MAX_CHANNELS,
+	.buffer_bytes_max =
+	    HDSPM_CHANNEL_BUFFER_BYTES * HDSPM_MAX_CHANNELS,
+	.period_bytes_min = (64 * 4),
+	.period_bytes_max = (8192 * 4) * HDSPM_MAX_CHANNELS,
+	.periods_min = 2,
+	.periods_max = 2,
+	.fifo_size = 0
+};
+
+static snd_pcm_hardware_t snd_hdspm_capture_subinfo = {
+	.info = (SNDRV_PCM_INFO_MMAP |
+		 SNDRV_PCM_INFO_MMAP_VALID |
+		 SNDRV_PCM_INFO_NONINTERLEAVED |
+		 SNDRV_PCM_INFO_SYNC_START),
+	.formats = SNDRV_PCM_FMTBIT_S32_LE,
+	.rates = (SNDRV_PCM_RATE_32000 |
+		  SNDRV_PCM_RATE_44100 |
+		  SNDRV_PCM_RATE_48000 |
+		  SNDRV_PCM_RATE_64000 |
+		  SNDRV_PCM_RATE_88200 | SNDRV_PCM_RATE_96000),
+	.rate_min = 32000,
+	.rate_max = 96000,
+	.channels_min = 1,
+	.channels_max = HDSPM_MAX_CHANNELS,
+	.buffer_bytes_max =
+	    HDSPM_CHANNEL_BUFFER_BYTES * HDSPM_MAX_CHANNELS,
+	.period_bytes_min = (64 * 4),
+	.period_bytes_max = (8192 * 4) * HDSPM_MAX_CHANNELS,
+	.periods_min = 2,
+	.periods_max = 2,
+	.fifo_size = 0
+};
+
+static snd_pcm_hw_constraint_list_t hw_constraints_period_sizes = {
+	.count = ARRAY_SIZE(period_sizes),
+	.list = period_sizes,
+	.mask = 0
+};
+
+
+static int snd_hdspm_hw_rule_channels_rate(snd_pcm_hw_params_t * params,
+					   snd_pcm_hw_rule_t * rule)
+{
+	hdspm_t *hdspm = rule->private;
+	snd_interval_t *c =
+	    hw_param_interval(params, SNDRV_PCM_HW_PARAM_CHANNELS);
+	snd_interval_t *r =
+	    hw_param_interval(params, SNDRV_PCM_HW_PARAM_RATE);
+
+	if (r->min > 48000) {
+		snd_interval_t t = {
+			.min = 1,
+			.max = hdspm->ds_channels,
+			.integer = 1,
+		};
+		return snd_interval_refine(c, &t);
+	} else if (r->max < 64000) {
+		snd_interval_t t = {
+			.min = 1,
+			.max = hdspm->ss_channels,
+			.integer = 1,
+		};
+		return snd_interval_refine(c, &t);
+	}
+	return 0;
+}
+
+static int snd_hdspm_hw_rule_rate_channels(snd_pcm_hw_params_t * params,
+					   snd_pcm_hw_rule_t * rule)
+{
+	hdspm_t *hdspm = rule->private;
+	snd_interval_t *c =
+	    hw_param_interval(params, SNDRV_PCM_HW_PARAM_CHANNELS);
+	snd_interval_t *r =
+	    hw_param_interval(params, SNDRV_PCM_HW_PARAM_RATE);
+
+	if (c->min <= hdspm->ss_channels) {
+		snd_interval_t t = {
+			.min = 32000,
+			.max = 48000,
+			.integer = 1,
+		};
+		return snd_interval_refine(r, &t);
+	} else if (c->max > hdspm->ss_channels) {
+		snd_interval_t t = {
+			.min = 64000,
+			.max = 96000,
+			.integer = 1,
+		};
+
+		return snd_interval_refine(r, &t);
+	}
+	return 0;
+}
+
+static int snd_hdspm_playback_open(snd_pcm_substream_t * substream)
+{
+	hdspm_t *hdspm = snd_pcm_substream_chip(substream);
+	snd_pcm_runtime_t *runtime = substream->runtime;
+
+	snd_printdd("Open device substream %d\n", substream->stream);
+
+	spin_lock_irq(&hdspm->lock);
+
+	snd_pcm_set_sync(substream);
+
+	runtime->hw = snd_hdspm_playback_subinfo;
+
+	if (hdspm->capture_substream == NULL)
+		hdspm_stop_audio(hdspm);
+
+	hdspm->playback_pid = current->pid;
+	hdspm->playback_substream = substream;
+
+	spin_unlock_irq(&hdspm->lock);
+
+	snd_pcm_hw_constraint_msbits(runtime, 0, 32, 24);
+
+	snd_pcm_hw_constraint_list(runtime, 0,
+				   SNDRV_PCM_HW_PARAM_PERIOD_SIZE,
+				   &hw_constraints_period_sizes);
+
+	snd_pcm_hw_rule_add(runtime, 0, SNDRV_PCM_HW_PARAM_CHANNELS,
+			    snd_hdspm_hw_rule_channels_rate, hdspm,
+			    SNDRV_PCM_HW_PARAM_RATE, -1);
+
+	snd_pcm_hw_rule_add(runtime, 0, SNDRV_PCM_HW_PARAM_RATE,
+			    snd_hdspm_hw_rule_rate_channels, hdspm,
+			    SNDRV_PCM_HW_PARAM_CHANNELS, -1);
+
+	return 0;
+}
+
+static int snd_hdspm_playback_release(snd_pcm_substream_t * substream)
+{
+	hdspm_t *hdspm = snd_pcm_substream_chip(substream);
+
+	spin_lock_irq(&hdspm->lock);
+
+	hdspm->playback_pid = -1;
+	hdspm->playback_substream = NULL;
+
+	spin_unlock_irq(&hdspm->lock);
+
+	return 0;
+}
+
+
+static int snd_hdspm_capture_open(snd_pcm_substream_t * substream)
+{
+	hdspm_t *hdspm = snd_pcm_substream_chip(substream);
+	snd_pcm_runtime_t *runtime = substream->runtime;
+
+	spin_lock_irq(&hdspm->lock);
+	snd_pcm_set_sync(substream);
+	runtime->hw = snd_hdspm_capture_subinfo;
+
+	if (hdspm->playback_substream == NULL)
+		hdspm_stop_audio(hdspm);
+
+	hdspm->capture_pid = current->pid;
+	hdspm->capture_substream = substream;
+
+	spin_unlock_irq(&hdspm->lock);
+
+	snd_pcm_hw_constraint_msbits(runtime, 0, 32, 24);
+	snd_pcm_hw_constraint_list(runtime, 0,
+				   SNDRV_PCM_HW_PARAM_PERIOD_SIZE,
+				   &hw_constraints_period_sizes);
+
+	snd_pcm_hw_rule_add(runtime, 0, SNDRV_PCM_HW_PARAM_CHANNELS,
+			    snd_hdspm_hw_rule_channels_rate, hdspm,
+			    SNDRV_PCM_HW_PARAM_RATE, -1);
+
+	snd_pcm_hw_rule_add(runtime, 0, SNDRV_PCM_HW_PARAM_RATE,
+			    snd_hdspm_hw_rule_rate_channels, hdspm,
+			    SNDRV_PCM_HW_PARAM_CHANNELS, -1);
+	return 0;
+}
+
+static int snd_hdspm_capture_release(snd_pcm_substream_t * substream)
+{
+	hdspm_t *hdspm = snd_pcm_substream_chip(substream);
+
+	spin_lock_irq(&hdspm->lock);
+
+	hdspm->capture_pid = -1;
+	hdspm->capture_substream = NULL;
+
+	spin_unlock_irq(&hdspm->lock);
+	return 0;
+}
+
+static int snd_hdspm_hwdep_dummy_op(snd_hwdep_t * hw, struct file *file)
+{
+	/* we have nothing to initialize but the call is required */
+	return 0;
+}
+
+
+static int snd_hdspm_hwdep_ioctl(snd_hwdep_t * hw, struct file *file,
+				 unsigned int cmd, unsigned long arg)
+{
+	hdspm_t *hdspm = (hdspm_t *) hw->private_data;
+	struct sndrv_hdspm_mixer_ioctl mixer;
+	hdspm_config_info_t info;
+	hdspm_version_t hdspm_version;
+	struct sndrv_hdspm_peak_rms_ioctl rms;
+
+	switch (cmd) {
+
+		
+	case SNDRV_HDSPM_IOCTL_GET_PEAK_RMS:
+		if (copy_from_user(&rms, (void __user *)arg, sizeof(rms)))
+			return -EFAULT;
+		/* maybe there is a chance to memorymap in future so dont touch just copy */
+		if(copy_to_user_fromio((void __user *)rms.peak,
+				       hdspm->iobase+HDSPM_MADI_peakrmsbase,
+				       sizeof(hdspm_peak_rms_t)) != 0 )
+			return -EFAULT;
+
+		break;
+		
+
+	case SNDRV_HDSPM_IOCTL_GET_CONFIG_INFO:
+
+		spin_lock_irq(&hdspm->lock);
+		info.pref_sync_ref =
+		    (unsigned char) hdspm_pref_sync_ref(hdspm);
+		info.wordclock_sync_check =
+		    (unsigned char) hdspm_wc_sync_check(hdspm);
+
+		info.system_sample_rate = hdspm->system_sample_rate;
+		info.autosync_sample_rate =
+		    hdspm_external_sample_rate(hdspm);
+		info.system_clock_mode =
+		    (unsigned char) hdspm_system_clock_mode(hdspm);
+		info.clock_source =
+		    (unsigned char) hdspm_clock_source(hdspm);
+		info.autosync_ref =
+		    (unsigned char) hdspm_autosync_ref(hdspm);
+		info.line_out = (unsigned char) hdspm_line_out(hdspm);
+		info.passthru = 0;
+		spin_unlock_irq(&hdspm->lock);
+		if (copy_to_user((void __user *) arg, &info, sizeof(info)))
+			return -EFAULT;
+		break;
+
+	case SNDRV_HDSPM_IOCTL_GET_VERSION:
+		hdspm_version.firmware_rev = hdspm->firmware_rev;
+		if (copy_to_user((void __user *) arg, &hdspm_version,
+				 sizeof(hdspm_version)))
+			return -EFAULT;
+		break;
+
+	case SNDRV_HDSPM_IOCTL_GET_MIXER:
+		if (copy_from_user(&mixer, (void __user *)arg, sizeof(mixer)))
+			return -EFAULT;
+		if (copy_to_user
+		    ((void __user *)mixer.mixer, hdspm->mixer, sizeof(hdspm_mixer_t)))
+			return -EFAULT;
+		break;
+
+	default:
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static snd_pcm_ops_t snd_hdspm_playback_ops = {
+	.open = snd_hdspm_playback_open,
+	.close = snd_hdspm_playback_release,
+	.ioctl = snd_hdspm_ioctl,
+	.hw_params = snd_hdspm_hw_params,
+	.hw_free = snd_hdspm_hw_free,
+	.prepare = snd_hdspm_prepare,
+	.trigger = snd_hdspm_trigger,
+	.pointer = snd_hdspm_hw_pointer,
+	.copy = snd_hdspm_playback_copy,
+	.silence = snd_hdspm_hw_silence,
+	.page = snd_pcm_sgbuf_ops_page,
+};
+
+static snd_pcm_ops_t snd_hdspm_capture_ops = {
+	.open = snd_hdspm_capture_open,
+	.close = snd_hdspm_capture_release,
+	.ioctl = snd_hdspm_ioctl,
+	.hw_params = snd_hdspm_hw_params,
+	.hw_free = snd_hdspm_hw_free,
+	.prepare = snd_hdspm_prepare,
+	.trigger = snd_hdspm_trigger,
+	.pointer = snd_hdspm_hw_pointer,
+	.copy = snd_hdspm_capture_copy,
+	.page = snd_pcm_sgbuf_ops_page,
+};
+
+static int __devinit snd_hdspm_create_hwdep(snd_card_t * card,
+					    hdspm_t * hdspm)
+{
+	snd_hwdep_t *hw;
+	int err;
+
+	if ((err = snd_hwdep_new(card, "HDSPM hwdep", 0, &hw)) < 0)
+		return err;
+
+	hdspm->hwdep = hw;
+	hw->private_data = hdspm;
+	strcpy(hw->name, "HDSPM hwdep interface");
+
+	hw->ops.open = snd_hdspm_hwdep_dummy_op;
+	hw->ops.ioctl = snd_hdspm_hwdep_ioctl;
+	hw->ops.release = snd_hdspm_hwdep_dummy_op;
+
+	return 0;
+}
+
+
+/*------------------------------------------------------------
+   memory interface 
+ ------------------------------------------------------------*/
+static int __devinit snd_hdspm_preallocate_memory(hdspm_t * hdspm)
+{
+	int err;
+	snd_pcm_t *pcm;
+	size_t wanted;
+
+	pcm = hdspm->pcm;
+
+	wanted = HDSPM_DMA_AREA_BYTES + 4096;	/* dont know why, but it works */
+
+	if ((err =
+	     snd_pcm_lib_preallocate_pages_for_all(pcm,
+	     					   SNDRV_DMA_TYPE_DEV_SG,
+						   snd_dma_pci_data(hdspm->pci),
+						   wanted,
+						   wanted)) < 0) {
+		snd_printdd("Could not preallocate %d  Bytes\n", wanted);
+
+		return err;
+	} else
+		snd_printdd(" Preallocated %d  Bytes\n", wanted);
+
+	return 0;
+}
+
+static int snd_hdspm_memory_free(hdspm_t * hdspm)
+{
+	snd_printdd("memory_free_for_all %p\n", hdspm->pcm);
+
+	snd_pcm_lib_preallocate_free_for_all(hdspm->pcm);
+	return 0;
+}
+
+
+static void hdspm_set_sgbuf(hdspm_t * hdspm, struct snd_sg_buf *sgbuf,
+			     unsigned int reg, int channels)
+{
+	int i;
+	for (i = 0; i < (channels * 16); i++)
+		hdspm_write(hdspm, reg + 4 * i,
+			    snd_pcm_sgbuf_get_addr(sgbuf,
+						   (size_t) 4096 * i));
+}
+
+/* ------------- ALSA Devices ---------------------------- */
+static int __devinit snd_hdspm_create_pcm(snd_card_t * card,
+					  hdspm_t * hdspm)
+{
+	snd_pcm_t *pcm;
+	int err;
+
+	if ((err = snd_pcm_new(card, hdspm->card_name, 0, 1, 1, &pcm)) < 0)
+		return err;
+
+	hdspm->pcm = pcm;
+	pcm->private_data = hdspm;
+	strcpy(pcm->name, hdspm->card_name);
+
+	snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_PLAYBACK,
+			&snd_hdspm_playback_ops);
+	snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_CAPTURE,
+			&snd_hdspm_capture_ops);
+
+	pcm->info_flags = SNDRV_PCM_INFO_JOINT_DUPLEX;
+
+	if ((err = snd_hdspm_preallocate_memory(hdspm)) < 0)
+		return err;
+
+	return 0;
+}
+
+static inline void snd_hdspm_initialize_midi_flush(hdspm_t * hdspm)
+{
+	snd_hdspm_flush_midi_input(hdspm, 0);
+	snd_hdspm_flush_midi_input(hdspm, 1);
+}
+
+static int __devinit snd_hdspm_create_alsa_devices(snd_card_t * card,
+						   hdspm_t * hdspm)
+{
+	int err;
+
+	snd_printdd("Create card...\n");
+	if ((err = snd_hdspm_create_pcm(card, hdspm)) < 0)
+		return err;
+
+	if ((err = snd_hdspm_create_midi(card, hdspm, 0)) < 0)
+		return err;
+
+	if ((err = snd_hdspm_create_midi(card, hdspm, 1)) < 0)
+		return err;
+
+	if ((err = snd_hdspm_create_controls(card, hdspm)) < 0)
+		return err;
+
+	if ((err = snd_hdspm_create_hwdep(card, hdspm)) < 0)
+		return err;
+
+	snd_printdd("proc init...\n");
+	snd_hdspm_proc_init(hdspm);
+
+	hdspm->system_sample_rate = -1;
+	hdspm->last_external_sample_rate = -1;
+	hdspm->last_internal_sample_rate = -1;
+	hdspm->playback_pid = -1;
+	hdspm->capture_pid = -1;
+	hdspm->capture_substream = NULL;
+	hdspm->playback_substream = NULL;
+
+	snd_printdd("Set defaults...\n");
+	if ((err = snd_hdspm_set_defaults(hdspm)) < 0)
+		return err;
+
+	snd_printdd("Update mixer controls...\n");
+	hdspm_update_simple_mixer_controls(hdspm);
+
+	snd_printdd("Initializeing complete ???\n");
+
+	if ((err = snd_card_register(card)) < 0) {
+		snd_printk(KERN_ERR "HDSPM: error registering card\n");
+		return err;
+	}
+
+	snd_printdd("... yes now\n");
+
+	return 0;
+}
+
+static int __devinit snd_hdspm_create(snd_card_t * card, hdspm_t * hdspm,
+				      int precise_ptr, int enable_monitor)
+{
+	struct pci_dev *pci = hdspm->pci;
+	int err;
+	int i;
+
+	unsigned long io_extent;
+
+	hdspm->irq = -1;
+	hdspm->irq_count = 0;
+
+	hdspm->midi[0].rmidi = NULL;
+	hdspm->midi[1].rmidi = NULL;
+	hdspm->midi[0].input = NULL;
+	hdspm->midi[1].input = NULL;
+	hdspm->midi[0].output = NULL;
+	hdspm->midi[1].output = NULL;
+	spin_lock_init(&hdspm->midi[0].lock);
+	spin_lock_init(&hdspm->midi[1].lock);
+	hdspm->iobase = NULL;
+	hdspm->control_register = 0;
+	hdspm->control2_register = 0;
+
+	hdspm->playback_buffer = NULL;
+	hdspm->capture_buffer = NULL;
+
+	for (i = 0; i < HDSPM_MAX_CHANNELS; ++i)
+		hdspm->playback_mixer_ctls[i] = NULL;
+	hdspm->mixer = NULL;
+
+	hdspm->card = card;
+
+	spin_lock_init(&hdspm->lock);
+
+	tasklet_init(&hdspm->midi_tasklet,
+		     hdspm_midi_tasklet, (unsigned long) hdspm);
+
+	pci_read_config_word(hdspm->pci,
+			     PCI_CLASS_REVISION, &hdspm->firmware_rev);
+
+	strcpy(card->driver, "HDSPM");
+	strcpy(card->mixername, "Xilinx FPGA");
+	hdspm->card_name = "RME HDSPM MADI";
+
+	if ((err = pci_enable_device(pci)) < 0)
+		return err;
+
+	pci_set_master(hdspm->pci);
+
+	if ((err = pci_request_regions(pci, "hdspm")) < 0)
+		return err;
+
+	hdspm->port = pci_resource_start(pci, 0);
+	io_extent = pci_resource_len(pci, 0);
+
+	snd_printdd("grabbed memory region 0x%lx-0x%lx\n",
+		   hdspm->port, hdspm->port + io_extent - 1);
+
+
+	if ((hdspm->iobase = ioremap_nocache(hdspm->port, io_extent)) == NULL) {
+		snd_printk(KERN_ERR "HDSPM: unable to remap region 0x%lx-0x%lx\n",
+			   hdspm->port, hdspm->port + io_extent - 1);
+		return -EBUSY;
+	}
+	snd_printdd("remapped region (0x%lx) 0x%lx-0x%lx\n",
+		   (unsigned long)hdspm->iobase, hdspm->port,
+		   hdspm->port + io_extent - 1);
+
+	if (request_irq(pci->irq, snd_hdspm_interrupt,
+			SA_INTERRUPT | SA_SHIRQ, "hdspm",
+			(void *) hdspm)) {
+		snd_printk(KERN_ERR "HDSPM: unable to use IRQ %d\n", pci->irq);
+		return -EBUSY;
+	}
+
+	snd_printdd("use IRQ %d\n", pci->irq);
+
+	hdspm->irq = pci->irq;
+	hdspm->precise_ptr = precise_ptr;
+
+	hdspm->monitor_outs = enable_monitor;
+
+	snd_printdd("kmalloc Mixer memory of %d Bytes\n",
+		   sizeof(hdspm_mixer_t));
+	if ((hdspm->mixer =
+	     (hdspm_mixer_t *) kmalloc(sizeof(hdspm_mixer_t), GFP_KERNEL))
+	    == NULL) {
+		snd_printk(KERN_ERR "HDSPM: unable to kmalloc Mixer memory of %d Bytes\n",
+			   (int)sizeof(hdspm_mixer_t));
+		return err;
+	}
+
+	hdspm->ss_channels = MADI_SS_CHANNELS;
+	hdspm->ds_channels = MADI_DS_CHANNELS;
+	hdspm->qs_channels = MADI_QS_CHANNELS;
+
+	snd_printdd("create alsa devices.\n");
+	if ((err = snd_hdspm_create_alsa_devices(card, hdspm)) < 0)
+		return err;
+
+	snd_hdspm_initialize_midi_flush(hdspm);
+
+	return 0;
+}
+
+static int snd_hdspm_free(hdspm_t * hdspm)
+{
+
+	if (hdspm->port) {
+
+		/* stop th audio, and cancel all interrupts */
+		hdspm->control_register &=
+		    ~(HDSPM_Start | HDSPM_AudioInterruptEnable
+		      | HDSPM_Midi0InterruptEnable |
+		      HDSPM_Midi1InterruptEnable);
+		hdspm_write(hdspm, HDSPM_controlRegister,
+			    hdspm->control_register);
+	}
+
+	if (hdspm->irq >= 0)
+		free_irq(hdspm->irq, (void *) hdspm);
+
+
+	if (hdspm->mixer)
+		kfree(hdspm->mixer);
+
+	if (hdspm->iobase)
+		iounmap(hdspm->iobase);
+
+	snd_hdspm_memory_free(hdspm);
+
+	if (hdspm->port)
+		pci_release_regions(hdspm->pci);
+
+	pci_disable_device(hdspm->pci);
+	return 0;
+}
+
+static void snd_hdspm_card_free(snd_card_t * card)
+{
+	hdspm_t *hdspm = (hdspm_t *) card->private_data;
+
+	if (hdspm)
+		snd_hdspm_free(hdspm);
+}
+
+static int __devinit snd_hdspm_probe(struct pci_dev *pci,
+				     const struct pci_device_id *pci_id)
+{
+	static int dev;
+	hdspm_t *hdspm;
+	snd_card_t *card;
+	int err;
+
+	if (dev >= SNDRV_CARDS)
+		return -ENODEV;
+	if (!enable[dev]) {
+		dev++;
+		return -ENOENT;
+	}
+
+	if (!(card = snd_card_new(index[dev], id[dev],
+				  THIS_MODULE, sizeof(hdspm_t))))
+		return -ENOMEM;
+
+	hdspm = (hdspm_t *) card->private_data;
+	card->private_free = snd_hdspm_card_free;
+	hdspm->dev = dev;
+	hdspm->pci = pci;
+
+	if ((err =
+	     snd_hdspm_create(card, hdspm, precise_ptr[dev],
+			      enable_monitor[dev])) < 0) {
+		snd_card_free(card);
+		return err;
+	}
+
+	strcpy(card->shortname, "HDSPM MADI");
+	sprintf(card->longname, "%s at 0x%lx, irq %d", hdspm->card_name,
+		hdspm->port, hdspm->irq);
+
+	if ((err = snd_card_register(card)) < 0) {
+		snd_card_free(card);
+		return err;
+	}
+
+	pci_set_drvdata(pci, card);
+
+	dev++;
+	return 0;
+}
+
+static void __devexit snd_hdspm_remove(struct pci_dev *pci)
+{
+	snd_card_free(pci_get_drvdata(pci));
+	pci_set_drvdata(pci, NULL);
+}
+
+static struct pci_driver driver = {
+	.name = "RME Hammerfall DSP MADI",
+	.id_table = snd_hdspm_ids,
+	.probe = snd_hdspm_probe,
+	.remove = __devexit_p(snd_hdspm_remove),
+};
+
+
+static int __init alsa_card_hdspm_init(void)
+{
+	return pci_register_driver(&driver);
+}
+
+static void __exit alsa_card_hdspm_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
+module_init(alsa_card_hdspm_init)
+module_exit(alsa_card_hdspm_exit)
-- 
cgit v1.2.3-70-g09d2


From b636a71d9b9525ee51ca872d461817a5bd5c39fd Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Mon, 13 Jun 2005 14:13:09 +0200
Subject: [ALSA] Add const prefix

Control Midlevel
Add const prefix to snd_kcontrol_new_t pointer for better protection.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 include/sound/control.h | 2 +-
 sound/core/control.c    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/sound/control.h b/include/sound/control.h
index 7b9444cd02f..ef7903c7a32 100644
--- a/include/sound/control.h
+++ b/include/sound/control.h
@@ -106,7 +106,7 @@ typedef int (*snd_kctl_ioctl_func_t) (snd_card_t * card,
 void snd_ctl_notify(snd_card_t * card, unsigned int mask, snd_ctl_elem_id_t * id);
 
 snd_kcontrol_t *snd_ctl_new(snd_kcontrol_t * kcontrol, unsigned int access);
-snd_kcontrol_t *snd_ctl_new1(snd_kcontrol_new_t * kcontrolnew, void * private_data);
+snd_kcontrol_t *snd_ctl_new1(const snd_kcontrol_new_t * kcontrolnew, void * private_data);
 void snd_ctl_free_one(snd_kcontrol_t * kcontrol);
 int snd_ctl_add(snd_card_t * card, snd_kcontrol_t * kcontrol);
 int snd_ctl_remove(snd_card_t * card, snd_kcontrol_t * kcontrol);
diff --git a/sound/core/control.c b/sound/core/control.c
index 4e39a2103d0..227f3cf0277 100644
--- a/sound/core/control.c
+++ b/sound/core/control.c
@@ -215,7 +215,7 @@ snd_kcontrol_t *snd_ctl_new(snd_kcontrol_t * control, unsigned int access)
  *
  * Returns the pointer of the newly generated instance, or NULL on failure.
  */
-snd_kcontrol_t *snd_ctl_new1(snd_kcontrol_new_t * ncontrol, void *private_data)
+snd_kcontrol_t *snd_ctl_new1(const snd_kcontrol_new_t * ncontrol, void *private_data)
 {
 	snd_kcontrol_t kctl;
 	unsigned int access;
-- 
cgit v1.2.3-70-g09d2


From dd7f0b80926befc8c70a873b5b0c0c7b5fd1e7b9 Mon Sep 17 00:00:00 2001
From: Harald Welte <laforge@netfilter.org>
Date: Wed, 22 Jun 2005 12:38:33 -0700
Subject: [NETFILTER]: Fix "iptables -D" rule deletion with ipt_CLUSTERIP
 target.

The patch just changes the order of structure members.

Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4/ipt_CLUSTERIP.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h b/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h
index baa83e75715..d9bceedfb3d 100644
--- a/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h
+++ b/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h
@@ -18,7 +18,6 @@ struct clusterip_config;
 struct ipt_clusterip_tgt_info {
 
 	u_int32_t flags;
-	struct clusterip_config *config;
 	
 	/* only relevant for new ones */
 	u_int8_t clustermac[6];
@@ -27,6 +26,8 @@ struct ipt_clusterip_tgt_info {
 	u_int16_t local_nodes[CLUSTERIP_MAX_NODES];
 	enum clusterip_hashmode hash_mode;
 	u_int32_t hash_initval;
+
+	struct clusterip_config *config;
 };
 
 #endif /*_IPT_CLUSTERIP_H_target*/
-- 
cgit v1.2.3-70-g09d2


From bdf042486a01aefaf29d74be1b4526daa70a5f0f Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 22 Jun 2005 20:58:29 +0100
Subject: [PATCH] ARM: Factor out common pmd_populate functionality

Both pmd_populate variants set two pmd entries before
ensuring that they are flushed from the cache.  Separate
this functionality into __pmd_populate().

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/pgalloc.h | 24 +++++++++---------------
 1 file changed, 9 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/pgalloc.h b/include/asm-arm/pgalloc.h
index e814f8144f8..bc18ff40518 100644
--- a/include/asm-arm/pgalloc.h
+++ b/include/asm-arm/pgalloc.h
@@ -89,6 +89,13 @@ static inline void pte_free(struct page *pte)
 	__free_page(pte);
 }
 
+static inline void __pmd_populate(pmd_t *pmdp, unsigned long pmdval)
+{
+	pmdp[0] = __pmd(pmdval);
+	pmdp[1] = __pmd(pmdval + 256 * sizeof(pte_t));
+	flush_pmd_entry(pmdp);
+}
+
 /*
  * Populate the pmdp entry with a pointer to the pte.  This pmd is part
  * of the mm address space.
@@ -99,32 +106,19 @@ static inline void
 pmd_populate_kernel(struct mm_struct *mm, pmd_t *pmdp, pte_t *ptep)
 {
 	unsigned long pte_ptr = (unsigned long)ptep;
-	unsigned long pmdval;
-
-	BUG_ON(mm != &init_mm);
 
 	/*
 	 * The pmd must be loaded with the physical
 	 * address of the PTE table
 	 */
 	pte_ptr -= PTRS_PER_PTE * sizeof(void *);
-	pmdval = __pa(pte_ptr) | _PAGE_KERNEL_TABLE;
-	pmdp[0] = __pmd(pmdval);
-	pmdp[1] = __pmd(pmdval + 256 * sizeof(pte_t));
-	flush_pmd_entry(pmdp);
+	__pmd_populate(pmdp, __pa(pte_ptr) | _PAGE_KERNEL_TABLE);
 }
 
 static inline void
 pmd_populate(struct mm_struct *mm, pmd_t *pmdp, struct page *ptep)
 {
-	unsigned long pmdval;
-
-	BUG_ON(mm == &init_mm);
-
-	pmdval = page_to_pfn(ptep) << PAGE_SHIFT | _PAGE_USER_TABLE;
-	pmdp[0] = __pmd(pmdval);
-	pmdp[1] = __pmd(pmdval + 256 * sizeof(pte_t));
-	flush_pmd_entry(pmdp);
+	__pmd_populate(pmdp, page_to_pfn(ptep) << PAGE_SHIFT | _PAGE_USER_TABLE);
 }
 
 #endif
-- 
cgit v1.2.3-70-g09d2


From 5ee0ed7d3ab620a764740fb018f469d45f561931 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 22 Jun 2005 17:16:20 +0000
Subject: [PATCH] RPC: Make rpc_create_client() probe server for RPC
 program+version support

 Ensure that we don't create an RPC client without checking that the server
 does indeed support the RPC program + version that we are trying to set up.

 This enables us to immediately return an error to "mount" if it turns out
 that the server is only supporting NFSv2, when we requested NFSv3 or NFSv4.

 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/lockd/host.c             |  4 +--
 fs/lockd/mon.c              |  2 +-
 include/linux/sunrpc/clnt.h |  4 +++
 net/sunrpc/clnt.c           | 59 ++++++++++++++++++++++++++++++++++++++++++++-
 net/sunrpc/pmap_clnt.c      |  2 +-
 5 files changed, 66 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index 90a62f27914..82c77df81c5 100644
--- a/fs/lockd/host.c
+++ b/fs/lockd/host.c
@@ -189,6 +189,8 @@ nlm_bind_host(struct nlm_host *host)
 			goto forgetit;
 
 		xprt_set_timeout(&xprt->timeout, 5, nlmsvc_timeout);
+		xprt->nocong = 1;	/* No congestion control for NLM */
+		xprt->resvport = 1;	/* NLM requires a reserved port */
 
 		/* Existing NLM servers accept AUTH_UNIX only */
 		clnt = rpc_create_client(xprt, host->h_name, &nlm_program,
@@ -196,8 +198,6 @@ nlm_bind_host(struct nlm_host *host)
 		if (IS_ERR(clnt))
 			goto forgetit;
 		clnt->cl_autobind = 1;	/* turn on pmap queries */
-		xprt->nocong = 1;	/* No congestion control for NLM */
-		xprt->resvport = 1;	/* NLM requires a reserved port */
 
 		host->h_rpcclnt = clnt;
 	}
diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index 81b5e7778d7..2d144abe84a 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -115,6 +115,7 @@ nsm_create(void)
 	xprt = xprt_create_proto(IPPROTO_UDP, &sin, NULL);
 	if (IS_ERR(xprt))
 		return (struct rpc_clnt *)xprt;
+	xprt->resvport = 1;	/* NSM requires a reserved port */
 
 	clnt = rpc_create_client(xprt, "localhost",
 				&nsm_program, SM_VERSION,
@@ -124,7 +125,6 @@ nsm_create(void)
 	clnt->cl_softrtry = 1;
 	clnt->cl_chatty   = 1;
 	clnt->cl_oneshot  = 1;
-	xprt->resvport = 1;	/* NSM requires a reserved port */
 	return clnt;
 
 out_err:
diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index 2709caf4d12..d25e80f77ff 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -111,6 +111,9 @@ struct rpc_procinfo {
 struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname,
 				struct rpc_program *info,
 				u32 version, rpc_authflavor_t authflavor);
+struct rpc_clnt *rpc_new_client(struct rpc_xprt *xprt, char *servname,
+				struct rpc_program *info,
+				u32 version, rpc_authflavor_t authflavor);
 struct rpc_clnt *rpc_clone_client(struct rpc_clnt *);
 int		rpc_shutdown_client(struct rpc_clnt *);
 int		rpc_destroy_client(struct rpc_clnt *);
@@ -129,6 +132,7 @@ void		rpc_clnt_sigmask(struct rpc_clnt *clnt, sigset_t *oldset);
 void		rpc_clnt_sigunmask(struct rpc_clnt *clnt, sigset_t *oldset);
 void		rpc_setbufsize(struct rpc_clnt *, unsigned int, unsigned int);
 size_t		rpc_max_payload(struct rpc_clnt *);
+int		rpc_ping(struct rpc_clnt *clnt, int flags);
 
 static __inline__
 int rpc_call(struct rpc_clnt *clnt, u32 proc, void *argp, void *resp, int flags)
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 99515d7727a..b36797ad808 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -97,7 +97,7 @@ rpc_setup_pipedir(struct rpc_clnt *clnt, char *dir_name)
  * made to sleep too long.
  */
 struct rpc_clnt *
-rpc_create_client(struct rpc_xprt *xprt, char *servname,
+rpc_new_client(struct rpc_xprt *xprt, char *servname,
 		  struct rpc_program *program, u32 vers,
 		  rpc_authflavor_t flavor)
 {
@@ -182,6 +182,36 @@ out_err:
 	return ERR_PTR(err);
 }
 
+/**
+ * Create an RPC client
+ * @xprt - pointer to xprt struct
+ * @servname - name of server
+ * @info - rpc_program
+ * @version - rpc_program version
+ * @authflavor - rpc_auth flavour to use
+ *
+ * Creates an RPC client structure, then pings the server in order to
+ * determine if it is up, and if it supports this program and version.
+ *
+ * This function should never be called by asynchronous tasks such as
+ * the portmapper.
+ */
+struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname,
+		struct rpc_program *info, u32 version, rpc_authflavor_t authflavor)
+{
+	struct rpc_clnt *clnt;
+	int err;
+	
+	clnt = rpc_new_client(xprt, servname, info, version, authflavor);
+	if (IS_ERR(clnt))
+		return clnt;
+	err = rpc_ping(clnt, RPC_TASK_SOFT|RPC_TASK_NOINTR);
+	if (err == 0)
+		return clnt;
+	rpc_shutdown_client(clnt);
+	return ERR_PTR(err);
+}
+
 /*
  * This function clones the RPC client structure. It allows us to share the
  * same transport while varying parameters such as the authentication
@@ -1086,3 +1116,30 @@ out_overflow:
 	printk(KERN_WARNING "RPC %s: server reply was truncated.\n", __FUNCTION__);
 	goto out_retry;
 }
+
+static int rpcproc_encode_null(void *rqstp, u32 *data, void *obj)
+{
+	return 0;
+}
+
+static int rpcproc_decode_null(void *rqstp, u32 *data, void *obj)
+{
+	return 0;
+}
+
+static struct rpc_procinfo rpcproc_null = {
+	.p_encode = rpcproc_encode_null,
+	.p_decode = rpcproc_decode_null,
+};
+
+int rpc_ping(struct rpc_clnt *clnt, int flags)
+{
+	struct rpc_message msg = {
+		.rpc_proc = &rpcproc_null,
+	};
+	int err;
+	msg.rpc_cred = authnull_ops.lookup_cred(NULL, NULL, 0);
+	err = rpc_call_sync(clnt, &msg, flags);
+	put_rpccred(msg.rpc_cred);
+	return err;
+}
diff --git a/net/sunrpc/pmap_clnt.c b/net/sunrpc/pmap_clnt.c
index 97c420ff1ee..df4d84c9020 100644
--- a/net/sunrpc/pmap_clnt.c
+++ b/net/sunrpc/pmap_clnt.c
@@ -207,7 +207,7 @@ pmap_create(char *hostname, struct sockaddr_in *srvaddr, int proto)
 	xprt->addr.sin_port = htons(RPC_PMAP_PORT);
 
 	/* printk("pmap: create clnt\n"); */
-	clnt = rpc_create_client(xprt, hostname,
+	clnt = rpc_new_client(xprt, hostname,
 				&pmap_program, RPC_PMAP_VERSION,
 				RPC_AUTH_UNIX);
 	if (!IS_ERR(clnt)) {
-- 
cgit v1.2.3-70-g09d2


From 4ce79717ce32a9f88c1ddce4b9658556cb59d37a Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 22 Jun 2005 17:16:21 +0000
Subject: [PATCH] NFS: Header file cleanup...

 - Move NFSv4 state definitions into a private header file.
 - Clean up gunk in nfs_fs.h

 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/callback.c      |   1 +
 fs/nfs/callback_proc.c |   1 +
 fs/nfs/callback_xdr.c  |   1 +
 fs/nfs/delegation.c    |   1 +
 fs/nfs/dir.c           |   1 +
 fs/nfs/idmap.c         |   1 +
 fs/nfs/inode.c         |   1 +
 fs/nfs/nfs4_fs.h       | 250 +++++++++++++++++++++++++++++++++++++++++++++++++
 fs/nfs/nfs4proc.c      |   5 +-
 fs/nfs/nfs4renewd.c    |   1 +
 fs/nfs/nfs4state.c     |  12 +--
 fs/nfs/nfs4xdr.c       |   8 +-
 include/linux/nfs_fs.h | 241 -----------------------------------------------
 13 files changed, 264 insertions(+), 260 deletions(-)
 create mode 100644 fs/nfs/nfs4_fs.h

(limited to 'include')

diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c
index 560d6175dd5..f2ca782aba3 100644
--- a/fs/nfs/callback.c
+++ b/fs/nfs/callback.c
@@ -14,6 +14,7 @@
 #include <linux/sunrpc/svc.h>
 #include <linux/sunrpc/svcsock.h>
 #include <linux/nfs_fs.h>
+#include "nfs4_fs.h"
 #include "callback.h"
 
 #define NFSDBG_FACILITY NFSDBG_CALLBACK
diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index ece27e42b93..65f1e19e4d1 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -8,6 +8,7 @@
 #include <linux/config.h>
 #include <linux/nfs4.h>
 #include <linux/nfs_fs.h>
+#include "nfs4_fs.h"
 #include "callback.h"
 #include "delegation.h"
 
diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c
index d271df9df2b..c99677ec58f 100644
--- a/fs/nfs/callback_xdr.c
+++ b/fs/nfs/callback_xdr.c
@@ -10,6 +10,7 @@
 #include <linux/sunrpc/svc.h>
 #include <linux/nfs4.h>
 #include <linux/nfs_fs.h>
+#include "nfs4_fs.h"
 #include "callback.h"
 
 #define CB_OP_TAGLEN_MAXSZ	(512)
diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index 5b9c60f9779..d7f7eb669d0 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -16,6 +16,7 @@
 #include <linux/nfs_fs.h>
 #include <linux/nfs_xdr.h>
 
+#include "nfs4_fs.h"
 #include "delegation.h"
 
 static struct nfs_delegation *nfs_alloc_delegation(void)
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index ff6155f5e8d..9ccb15e8696 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -32,6 +32,7 @@
 #include <linux/smp_lock.h>
 #include <linux/namei.h>
 
+#include "nfs4_fs.h"
 #include "delegation.h"
 
 #define NFS_PARANOIA 1
diff --git a/fs/nfs/idmap.c b/fs/nfs/idmap.c
index 87f4f9aeac8..ffb8df91dc3 100644
--- a/fs/nfs/idmap.c
+++ b/fs/nfs/idmap.c
@@ -50,6 +50,7 @@
 #include <linux/nfs_fs.h>
 
 #include <linux/nfs_idmap.h>
+#include "nfs4_fs.h"
 
 #define IDMAP_HASH_SZ          128
 
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 32ddcf69e9a..c80a81ff59c 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -39,6 +39,7 @@
 #include <asm/system.h>
 #include <asm/uaccess.h>
 
+#include "nfs4_fs.h"
 #include "delegation.h"
 
 #define NFSDBG_FACILITY		NFSDBG_VFS
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
new file mode 100644
index 00000000000..85cf3bd3692
--- /dev/null
+++ b/fs/nfs/nfs4_fs.h
@@ -0,0 +1,250 @@
+/*
+ * linux/fs/nfs/nfs4_fs.h
+ *
+ * Copyright (C) 2005 Trond Myklebust
+ *
+ * NFSv4-specific filesystem definitions and declarations
+ */
+
+#ifndef __LINUX_FS_NFS_NFS4_FS_H
+#define __LINUX_FS_NFS_NFS4_FS_H
+
+#ifdef CONFIG_NFS_V4
+
+struct idmap;
+
+/*
+ * In a seqid-mutating op, this macro controls which error return
+ * values trigger incrementation of the seqid.
+ *
+ * from rfc 3010:
+ * The client MUST monotonically increment the sequence number for the
+ * CLOSE, LOCK, LOCKU, OPEN, OPEN_CONFIRM, and OPEN_DOWNGRADE
+ * operations.  This is true even in the event that the previous
+ * operation that used the sequence number received an error.  The only
+ * exception to this rule is if the previous operation received one of
+ * the following errors: NFSERR_STALE_CLIENTID, NFSERR_STALE_STATEID,
+ * NFSERR_BAD_STATEID, NFSERR_BAD_SEQID, NFSERR_BADXDR,
+ * NFSERR_RESOURCE, NFSERR_NOFILEHANDLE.
+ *
+ */
+#define seqid_mutating_err(err)       \
+(((err) != NFSERR_STALE_CLIENTID) &&  \
+ ((err) != NFSERR_STALE_STATEID)  &&  \
+ ((err) != NFSERR_BAD_STATEID)    &&  \
+ ((err) != NFSERR_BAD_SEQID)      &&  \
+ ((err) != NFSERR_BAD_XDR)        &&  \
+ ((err) != NFSERR_RESOURCE)       &&  \
+ ((err) != NFSERR_NOFILEHANDLE))
+
+enum nfs4_client_state {
+	NFS4CLNT_OK  = 0,
+};
+
+/*
+ * The nfs4_client identifies our client state to the server.
+ */
+struct nfs4_client {
+	struct list_head	cl_servers;	/* Global list of servers */
+	struct in_addr		cl_addr;	/* Server identifier */
+	u64			cl_clientid;	/* constant */
+	nfs4_verifier		cl_confirm;
+	unsigned long		cl_state;
+
+	u32			cl_lockowner_id;
+
+	/*
+	 * The following rwsem ensures exclusive access to the server
+	 * while we recover the state following a lease expiration.
+	 */
+	struct rw_semaphore	cl_sem;
+
+	struct list_head	cl_delegations;
+	struct list_head	cl_state_owners;
+	struct list_head	cl_unused;
+	int			cl_nunused;
+	spinlock_t		cl_lock;
+	atomic_t		cl_count;
+
+	struct rpc_clnt *	cl_rpcclient;
+	struct rpc_cred *	cl_cred;
+
+	struct list_head	cl_superblocks;	/* List of nfs_server structs */
+
+	unsigned long		cl_lease_time;
+	unsigned long		cl_last_renewal;
+	struct work_struct	cl_renewd;
+	struct work_struct	cl_recoverd;
+
+	wait_queue_head_t	cl_waitq;
+	struct rpc_wait_queue	cl_rpcwaitq;
+
+	/* used for the setclientid verifier */
+	struct timespec		cl_boot_time;
+
+	/* idmapper */
+	struct idmap *		cl_idmap;
+
+	/* Our own IP address, as a null-terminated string.
+	 * This is used to generate the clientid, and the callback address.
+	 */
+	char			cl_ipaddr[16];
+	unsigned char		cl_id_uniquifier;
+};
+
+/*
+ * NFS4 state_owners and lock_owners are simply labels for ordered
+ * sequences of RPC calls. Their sole purpose is to provide once-only
+ * semantics by allowing the server to identify replayed requests.
+ *
+ * The ->so_sema is held during all state_owner seqid-mutating operations:
+ * OPEN, OPEN_DOWNGRADE, and CLOSE. Its purpose is to properly serialize
+ * so_seqid.
+ */
+struct nfs4_state_owner {
+	struct list_head     so_list;	 /* per-clientid list of state_owners */
+	struct nfs4_client   *so_client;
+	u32                  so_id;      /* 32-bit identifier, unique */
+	struct semaphore     so_sema;
+	u32                  so_seqid;   /* protected by so_sema */
+	atomic_t	     so_count;
+
+	struct rpc_cred	     *so_cred;	 /* Associated cred */
+	struct list_head     so_states;
+	struct list_head     so_delegations;
+};
+
+/*
+ * struct nfs4_state maintains the client-side state for a given
+ * (state_owner,inode) tuple (OPEN) or state_owner (LOCK).
+ *
+ * OPEN:
+ * In order to know when to OPEN_DOWNGRADE or CLOSE the state on the server,
+ * we need to know how many files are open for reading or writing on a
+ * given inode. This information too is stored here.
+ *
+ * LOCK: one nfs4_state (LOCK) to hold the lock stateid nfs4_state(OPEN)
+ */
+
+struct nfs4_lock_state {
+	struct list_head	ls_locks;	/* Other lock stateids */
+	fl_owner_t		ls_owner;	/* POSIX lock owner */
+#define NFS_LOCK_INITIALIZED 1
+	int			ls_flags;
+	u32			ls_seqid;
+	u32			ls_id;
+	nfs4_stateid		ls_stateid;
+	atomic_t		ls_count;
+};
+
+/* bits for nfs4_state->flags */
+enum {
+	LK_STATE_IN_USE,
+	NFS_DELEGATED_STATE,
+};
+
+struct nfs4_state {
+	struct list_head open_states;	/* List of states for the same state_owner */
+	struct list_head inode_states;	/* List of states for the same inode */
+	struct list_head lock_states;	/* List of subservient lock stateids */
+
+	struct nfs4_state_owner *owner;	/* Pointer to the open owner */
+	struct inode *inode;		/* Pointer to the inode */
+
+	unsigned long flags;		/* Do we hold any locks? */
+	struct semaphore lock_sema;	/* Serializes file locking operations */
+	rwlock_t state_lock;		/* Protects the lock_states list */
+
+	nfs4_stateid stateid;
+
+	unsigned int nreaders;
+	unsigned int nwriters;
+	int state;			/* State on the server (R,W, or RW) */
+	atomic_t count;
+};
+
+
+struct nfs4_exception {
+	long timeout;
+	int retry;
+};
+
+struct nfs4_state_recovery_ops {
+	int (*recover_open)(struct nfs4_state_owner *, struct nfs4_state *);
+	int (*recover_lock)(struct nfs4_state *, struct file_lock *);
+};
+
+extern struct dentry_operations nfs4_dentry_operations;
+extern struct inode_operations nfs4_dir_inode_operations;
+
+/* nfs4proc.c */
+extern int nfs4_map_errors(int err);
+extern int nfs4_proc_setclientid(struct nfs4_client *, u32, unsigned short);
+extern int nfs4_proc_setclientid_confirm(struct nfs4_client *);
+extern int nfs4_proc_async_renew(struct nfs4_client *);
+extern int nfs4_proc_renew(struct nfs4_client *);
+extern int nfs4_do_close(struct inode *inode, struct nfs4_state *state, mode_t mode);
+extern struct inode *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *);
+extern int nfs4_open_revalidate(struct inode *, struct dentry *, int);
+
+extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops;
+extern struct nfs4_state_recovery_ops nfs4_network_partition_recovery_ops;
+
+extern const u32 nfs4_fattr_bitmap[2];
+extern const u32 nfs4_statfs_bitmap[2];
+extern const u32 nfs4_pathconf_bitmap[2];
+extern const u32 nfs4_fsinfo_bitmap[2];
+
+/* nfs4renewd.c */
+extern void nfs4_schedule_state_renewal(struct nfs4_client *);
+extern void nfs4_renewd_prepare_shutdown(struct nfs_server *);
+extern void nfs4_kill_renewd(struct nfs4_client *);
+extern void nfs4_renew_state(void *);
+
+/* nfs4state.c */
+extern void init_nfsv4_state(struct nfs_server *);
+extern void destroy_nfsv4_state(struct nfs_server *);
+extern struct nfs4_client *nfs4_get_client(struct in_addr *);
+extern void nfs4_put_client(struct nfs4_client *clp);
+extern int nfs4_init_client(struct nfs4_client *clp);
+extern struct nfs4_client *nfs4_find_client(struct in_addr *);
+extern u32 nfs4_alloc_lockowner_id(struct nfs4_client *);
+
+extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *);
+extern void nfs4_put_state_owner(struct nfs4_state_owner *);
+extern void nfs4_drop_state_owner(struct nfs4_state_owner *);
+extern struct nfs4_state * nfs4_get_open_state(struct inode *, struct nfs4_state_owner *);
+extern void nfs4_put_open_state(struct nfs4_state *);
+extern void nfs4_close_state(struct nfs4_state *, mode_t);
+extern struct nfs4_state *nfs4_find_state(struct inode *, struct rpc_cred *, mode_t mode);
+extern void nfs4_increment_seqid(int status, struct nfs4_state_owner *sp);
+extern void nfs4_schedule_state_recovery(struct nfs4_client *);
+extern struct nfs4_lock_state *nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t);
+extern struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t);
+extern void nfs4_put_lock_state(struct nfs4_lock_state *state);
+extern void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *ls);
+extern void nfs4_notify_setlk(struct nfs4_state *, struct file_lock *, struct nfs4_lock_state *);
+extern void nfs4_notify_unlck(struct nfs4_state *, struct file_lock *, struct nfs4_lock_state *);
+extern void nfs4_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t);
+
+extern const nfs4_stateid zero_stateid;
+
+/* nfs4xdr.c */
+extern uint32_t *nfs4_decode_dirent(uint32_t *p, struct nfs_entry *entry, int plus);
+extern struct rpc_procinfo nfs4_procedures[];
+
+struct nfs4_mount_data;
+
+/* callback_xdr.c */
+extern struct svc_version nfs4_callback_version1;
+
+#else
+
+#define init_nfsv4_state(server)  do { } while (0)
+#define destroy_nfsv4_state(server)       do { } while (0)
+#define nfs4_put_state_owner(inode, owner) do { } while (0)
+#define nfs4_put_open_state(state) do { } while (0)
+#define nfs4_close_state(a, b) do { } while (0)
+
+#endif /* CONFIG_NFS_V4 */
+#endif /* __LINUX_FS_NFS_NFS4_FS.H */
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 1d5cb3e80c3..a69c02b206c 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -48,6 +48,7 @@
 #include <linux/smp_lock.h>
 #include <linux/namei.h>
 
+#include "nfs4_fs.h"
 #include "delegation.h"
 
 #define NFSDBG_FACILITY		NFSDBG_PROC
@@ -62,8 +63,6 @@ static int nfs4_handle_exception(struct nfs_server *server, int errorcode, struc
 extern u32 *nfs4_decode_dirent(u32 *p, struct nfs_entry *entry, int plus);
 extern struct rpc_procinfo nfs4_procedures[];
 
-extern nfs4_stateid zero_stateid;
-
 /* Prevent leaks of NFSv4 errors into userland */
 int nfs4_map_errors(int err)
 {
@@ -104,7 +103,7 @@ const u32 nfs4_statfs_bitmap[2] = {
 	| FATTR4_WORD1_SPACE_TOTAL
 };
 
-u32 nfs4_pathconf_bitmap[2] = {
+const u32 nfs4_pathconf_bitmap[2] = {
 	FATTR4_WORD0_MAXLINK
 	| FATTR4_WORD0_MAXNAME,
 	0
diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c
index 667e06f1c64..a3001628ad3 100644
--- a/fs/nfs/nfs4renewd.c
+++ b/fs/nfs/nfs4renewd.c
@@ -53,6 +53,7 @@
 #include <linux/nfs.h>
 #include <linux/nfs4.h>
 #include <linux/nfs_fs.h>
+#include "nfs4_fs.h"
 
 #define NFSDBG_FACILITY	NFSDBG_PROC
 
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 231cebce3c8..17b187f2d77 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -46,24 +46,18 @@
 #include <linux/workqueue.h>
 #include <linux/bitops.h>
 
+#include "nfs4_fs.h"
 #include "callback.h"
 #include "delegation.h"
 
 #define OPENOWNER_POOL_SIZE	8
 
-static DEFINE_SPINLOCK(state_spinlock);
-
-nfs4_stateid zero_stateid;
-
-#if 0
-nfs4_stateid one_stateid =
-	{ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
-#endif
+const nfs4_stateid zero_stateid;
 
+static DEFINE_SPINLOCK(state_spinlock);
 static LIST_HEAD(nfs4_clientid_list);
 
 static void nfs4_recover_state(void *);
-extern void nfs4_renew_state(void *);
 
 void
 init_nfsv4_state(struct nfs_server *server)
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 5f4de05763c..e86406eff0e 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -51,6 +51,7 @@
 #include <linux/nfs4.h>
 #include <linux/nfs_fs.h>
 #include <linux/nfs_idmap.h>
+#include "nfs4_fs.h"
 
 #define NFSDBG_FACILITY		NFSDBG_XDR
 
@@ -660,8 +661,6 @@ static int encode_getattr_two(struct xdr_stream *xdr, uint32_t bm0, uint32_t bm1
 
 static int encode_getfattr(struct xdr_stream *xdr, const u32* bitmask)
 {
-	extern u32 nfs4_fattr_bitmap[];
-
 	return encode_getattr_two(xdr,
 			bitmask[0] & nfs4_fattr_bitmap[0],
 			bitmask[1] & nfs4_fattr_bitmap[1]);
@@ -669,8 +668,6 @@ static int encode_getfattr(struct xdr_stream *xdr, const u32* bitmask)
 
 static int encode_fsinfo(struct xdr_stream *xdr, const u32* bitmask)
 {
-	extern u32 nfs4_fsinfo_bitmap[];
-
 	return encode_getattr_two(xdr, bitmask[0] & nfs4_fsinfo_bitmap[0],
 			bitmask[1] & nfs4_fsinfo_bitmap[1]);
 }
@@ -969,7 +966,6 @@ static int encode_putrootfh(struct xdr_stream *xdr)
 
 static void encode_stateid(struct xdr_stream *xdr, const struct nfs_open_context *ctx)
 {
-	extern nfs4_stateid zero_stateid;
 	nfs4_stateid stateid;
 	uint32_t *p;
 
@@ -1697,7 +1693,6 @@ static int nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, uint32_t *p, struct nfs4_fs
  */
 static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, uint32_t *p, const struct nfs4_pathconf_arg *args)
 {
-	extern u32 nfs4_pathconf_bitmap[2];
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
 		.nops = 2,
@@ -1718,7 +1713,6 @@ static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, uint32_t *p, const struct
  */
 static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, uint32_t *p, const struct nfs4_statfs_arg *args)
 {
-	extern u32 nfs4_statfs_bitmap[];
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
 		.nops = 2,
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index dbac7f363e5..fb33e7655cf 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -15,7 +15,6 @@
 #include <linux/pagemap.h>
 #include <linux/rwsem.h>
 #include <linux/wait.h>
-#include <linux/uio.h>
 
 #include <linux/nfs_fs_sb.h>
 
@@ -29,7 +28,6 @@
 #include <linux/nfs4.h>
 #include <linux/nfs_xdr.h>
 #include <linux/rwsem.h>
-#include <linux/workqueue.h>
 #include <linux/mempool.h>
 
 /*
@@ -43,13 +41,6 @@
 #define NFS_MAX_FILE_IO_BUFFER_SIZE	32768
 #define NFS_DEF_FILE_IO_BUFFER_SIZE	4096
 
-/*
- * The upper limit on timeouts for the exponential backoff algorithm.
- */
-#define NFS_WRITEBACK_DELAY		(5*HZ)
-#define NFS_WRITEBACK_LOCKDELAY		(60*HZ)
-#define NFS_COMMIT_DELAY		(5*HZ)
-
 /*
  * superblock magic number for NFS
  */
@@ -60,9 +51,6 @@
  */
 #define NFS_RPC_SWAPFLAGS		(RPC_TASK_SWAPPER|RPC_TASK_ROOTCREDS)
 
-#define NFS_RW_SYNC		0x0001	/* O_SYNC handling */
-#define NFS_RW_SWAP		0x0002	/* This is a swap request */
-
 /*
  * When flushing a cluster of dirty pages, there can be different
  * strategies:
@@ -434,11 +422,6 @@ static inline void nfs_writedata_free(struct nfs_write_data *p)
 	mempool_free(p, nfs_wdata_mempool);
 }
 
-/* Hack for future NFS swap support */
-#ifndef IS_SWAPFILE
-# define IS_SWAPFILE(inode)	(0)
-#endif
-
 /*
  * linux/fs/nfs/read.c
  */
@@ -515,230 +498,6 @@ extern void * nfs_root_data(void);
 
 #define NFS_JUKEBOX_RETRY_TIME (5 * HZ)
 
-#ifdef CONFIG_NFS_V4
-
-struct idmap;
-
-/*
- * In a seqid-mutating op, this macro controls which error return
- * values trigger incrementation of the seqid.
- *
- * from rfc 3010:
- * The client MUST monotonically increment the sequence number for the
- * CLOSE, LOCK, LOCKU, OPEN, OPEN_CONFIRM, and OPEN_DOWNGRADE
- * operations.  This is true even in the event that the previous
- * operation that used the sequence number received an error.  The only
- * exception to this rule is if the previous operation received one of
- * the following errors: NFSERR_STALE_CLIENTID, NFSERR_STALE_STATEID,
- * NFSERR_BAD_STATEID, NFSERR_BAD_SEQID, NFSERR_BADXDR,
- * NFSERR_RESOURCE, NFSERR_NOFILEHANDLE.
- *
- */
-#define seqid_mutating_err(err)       \
-(((err) != NFSERR_STALE_CLIENTID) &&  \
- ((err) != NFSERR_STALE_STATEID)  &&  \
- ((err) != NFSERR_BAD_STATEID)    &&  \
- ((err) != NFSERR_BAD_SEQID)      &&  \
- ((err) != NFSERR_BAD_XDR)        &&  \
- ((err) != NFSERR_RESOURCE)       &&  \
- ((err) != NFSERR_NOFILEHANDLE))
-
-enum nfs4_client_state {
-	NFS4CLNT_OK  = 0,
-};
-
-/*
- * The nfs4_client identifies our client state to the server.
- */
-struct nfs4_client {
-	struct list_head	cl_servers;	/* Global list of servers */
-	struct in_addr		cl_addr;	/* Server identifier */
-	u64			cl_clientid;	/* constant */
-	nfs4_verifier		cl_confirm;
-	unsigned long		cl_state;
-
-	u32			cl_lockowner_id;
-
-	/*
-	 * The following rwsem ensures exclusive access to the server
-	 * while we recover the state following a lease expiration.
-	 */
-	struct rw_semaphore	cl_sem;
-
-	struct list_head	cl_delegations;
-	struct list_head	cl_state_owners;
-	struct list_head	cl_unused;
-	int			cl_nunused;
-	spinlock_t		cl_lock;
-	atomic_t		cl_count;
-
-	struct rpc_clnt *	cl_rpcclient;
-	struct rpc_cred *	cl_cred;
-
-	struct list_head	cl_superblocks;	/* List of nfs_server structs */
-
-	unsigned long		cl_lease_time;
-	unsigned long		cl_last_renewal;
-	struct work_struct	cl_renewd;
-	struct work_struct	cl_recoverd;
-
-	wait_queue_head_t	cl_waitq;
-	struct rpc_wait_queue	cl_rpcwaitq;
-
-	/* used for the setclientid verifier */
-	struct timespec		cl_boot_time;
-
-	/* idmapper */
-	struct idmap *		cl_idmap;
-
-	/* Our own IP address, as a null-terminated string.
-	 * This is used to generate the clientid, and the callback address.
-	 */
-	char			cl_ipaddr[16];
-	unsigned char		cl_id_uniquifier;
-};
-
-/*
- * NFS4 state_owners and lock_owners are simply labels for ordered
- * sequences of RPC calls. Their sole purpose is to provide once-only
- * semantics by allowing the server to identify replayed requests.
- *
- * The ->so_sema is held during all state_owner seqid-mutating operations:
- * OPEN, OPEN_DOWNGRADE, and CLOSE. Its purpose is to properly serialize
- * so_seqid.
- */
-struct nfs4_state_owner {
-	struct list_head     so_list;	 /* per-clientid list of state_owners */
-	struct nfs4_client   *so_client;
-	u32                  so_id;      /* 32-bit identifier, unique */
-	struct semaphore     so_sema;
-	u32                  so_seqid;   /* protected by so_sema */
-	atomic_t	     so_count;
-
-	struct rpc_cred	     *so_cred;	 /* Associated cred */
-	struct list_head     so_states;
-	struct list_head     so_delegations;
-};
-
-/*
- * struct nfs4_state maintains the client-side state for a given
- * (state_owner,inode) tuple (OPEN) or state_owner (LOCK).
- *
- * OPEN:
- * In order to know when to OPEN_DOWNGRADE or CLOSE the state on the server,
- * we need to know how many files are open for reading or writing on a
- * given inode. This information too is stored here.
- *
- * LOCK: one nfs4_state (LOCK) to hold the lock stateid nfs4_state(OPEN)
- */
-
-struct nfs4_lock_state {
-	struct list_head	ls_locks;	/* Other lock stateids */
-	fl_owner_t		ls_owner;	/* POSIX lock owner */
-#define NFS_LOCK_INITIALIZED 1
-	int			ls_flags;
-	u32			ls_seqid;
-	u32			ls_id;
-	nfs4_stateid		ls_stateid;
-	atomic_t		ls_count;
-};
-
-/* bits for nfs4_state->flags */
-enum {
-	LK_STATE_IN_USE,
-	NFS_DELEGATED_STATE,
-};
-
-struct nfs4_state {
-	struct list_head open_states;	/* List of states for the same state_owner */
-	struct list_head inode_states;	/* List of states for the same inode */
-	struct list_head lock_states;	/* List of subservient lock stateids */
-
-	struct nfs4_state_owner *owner;	/* Pointer to the open owner */
-	struct inode *inode;		/* Pointer to the inode */
-
-	unsigned long flags;		/* Do we hold any locks? */
-	struct semaphore lock_sema;	/* Serializes file locking operations */
-	rwlock_t state_lock;		/* Protects the lock_states list */
-
-	nfs4_stateid stateid;
-
-	unsigned int nreaders;
-	unsigned int nwriters;
-	int state;			/* State on the server (R,W, or RW) */
-	atomic_t count;
-};
-
-
-struct nfs4_exception {
-	long timeout;
-	int retry;
-};
-
-struct nfs4_state_recovery_ops {
-	int (*recover_open)(struct nfs4_state_owner *, struct nfs4_state *);
-	int (*recover_lock)(struct nfs4_state *, struct file_lock *);
-};
-
-extern struct dentry_operations nfs4_dentry_operations;
-extern struct inode_operations nfs4_dir_inode_operations;
-
-/* nfs4proc.c */
-extern int nfs4_map_errors(int err);
-extern int nfs4_proc_setclientid(struct nfs4_client *, u32, unsigned short);
-extern int nfs4_proc_setclientid_confirm(struct nfs4_client *);
-extern int nfs4_proc_async_renew(struct nfs4_client *);
-extern int nfs4_proc_renew(struct nfs4_client *);
-extern int nfs4_do_close(struct inode *inode, struct nfs4_state *state, mode_t mode);
-extern struct inode *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *);
-extern int nfs4_open_revalidate(struct inode *, struct dentry *, int);
-
-extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops;
-extern struct nfs4_state_recovery_ops nfs4_network_partition_recovery_ops;
-
-/* nfs4renewd.c */
-extern void nfs4_schedule_state_renewal(struct nfs4_client *);
-extern void nfs4_renewd_prepare_shutdown(struct nfs_server *);
-extern void nfs4_kill_renewd(struct nfs4_client *);
-
-/* nfs4state.c */
-extern void init_nfsv4_state(struct nfs_server *);
-extern void destroy_nfsv4_state(struct nfs_server *);
-extern struct nfs4_client *nfs4_get_client(struct in_addr *);
-extern void nfs4_put_client(struct nfs4_client *clp);
-extern int nfs4_init_client(struct nfs4_client *clp);
-extern struct nfs4_client *nfs4_find_client(struct in_addr *);
-extern u32 nfs4_alloc_lockowner_id(struct nfs4_client *);
-
-extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *);
-extern void nfs4_put_state_owner(struct nfs4_state_owner *);
-extern void nfs4_drop_state_owner(struct nfs4_state_owner *);
-extern struct nfs4_state * nfs4_get_open_state(struct inode *, struct nfs4_state_owner *);
-extern void nfs4_put_open_state(struct nfs4_state *);
-extern void nfs4_close_state(struct nfs4_state *, mode_t);
-extern struct nfs4_state *nfs4_find_state(struct inode *, struct rpc_cred *, mode_t mode);
-extern void nfs4_increment_seqid(int status, struct nfs4_state_owner *sp);
-extern void nfs4_schedule_state_recovery(struct nfs4_client *);
-extern struct nfs4_lock_state *nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t);
-extern struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t);
-extern void nfs4_put_lock_state(struct nfs4_lock_state *state);
-extern void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *ls);
-extern void nfs4_notify_setlk(struct nfs4_state *, struct file_lock *, struct nfs4_lock_state *);
-extern void nfs4_notify_unlck(struct nfs4_state *, struct file_lock *, struct nfs4_lock_state *);
-extern void nfs4_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t);
-
-
-
-struct nfs4_mount_data;
-#else
-#define init_nfsv4_state(server)  do { } while (0)
-#define destroy_nfsv4_state(server)       do { } while (0)
-#define nfs4_put_state_owner(inode, owner) do { } while (0)
-#define nfs4_put_open_state(state) do { } while (0)
-#define nfs4_close_state(a, b) do { } while (0)
-#define nfs4_renewd_prepare_shutdown(server) do { } while (0)
-#endif
-
 #endif /* __KERNEL__ */
 
 /*
-- 
cgit v1.2.3-70-g09d2


From a656db998785324a818005bcf71bae6dcbbb3cf5 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 22 Jun 2005 17:16:21 +0000
Subject: [PATCH] NFS: Remove unused NFS inode field readdir_timestamp.

 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/dir.c           | 8 +++-----
 include/linux/nfs_fs.h | 1 -
 2 files changed, 3 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 9ccb15e8696..dffa21abd3e 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -165,12 +165,10 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page *page)
 	NFS_FLAGS(inode) |= NFS_INO_INVALID_ATIME;
 	/* Ensure consistent page alignment of the data.
 	 * Note: assumes we have exclusive access to this mapping either
-	 *	 throught inode->i_sem or some other mechanism.
+	 *	 through inode->i_sem or some other mechanism.
 	 */
-	if (page->index == 0) {
-		invalidate_inode_pages(inode->i_mapping);
-		NFS_I(inode)->readdir_timestamp = timestamp;
-	}
+	if (page->index == 0)
+		invalidate_inode_pages2_range(inode->i_mapping, PAGE_CACHE_SIZE, -1);
 	unlock_page(page);
 	return 0;
  error:
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index fb33e7655cf..68d5aae8997 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -128,7 +128,6 @@ struct nfs_inode {
 	 *
 	 *	mtime != read_cache_mtime
 	 */
-	unsigned long		readdir_timestamp;
 	unsigned long		read_cache_jiffies;
 	unsigned long		attrtimeo;
 	unsigned long		attrtimeo_timestamp;
-- 
cgit v1.2.3-70-g09d2


From 96651ab341cde0fee940ec837f323d711cbfa7d5 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 22 Jun 2005 17:16:21 +0000
Subject: [PATCH] RPC: Shrink struct rpc_task by switching to wait_on_bit()

 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/sched.h |  1 -
 net/sunrpc/sched.c           | 31 ++++++++++++++++++-------------
 2 files changed, 18 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h
index 99d17ed7ceb..4d77e90d0b3 100644
--- a/include/linux/sunrpc/sched.h
+++ b/include/linux/sunrpc/sched.h
@@ -31,7 +31,6 @@ struct rpc_wait_queue;
 struct rpc_wait {
 	struct list_head	list;		/* wait queue links */
 	struct list_head	links;		/* Links to related tasks */
-	wait_queue_head_t	waitq;		/* sync: sleep on this q */
 	struct rpc_wait_queue *	rpc_waitq;	/* RPC wait queue we're on */
 };
 
diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index cc298fa4b81..2d9eb7fbd52 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -290,7 +290,7 @@ static void rpc_make_runnable(struct rpc_task *task)
 			return;
 		}
 	} else
-		wake_up(&task->u.tk_wait.waitq);
+		wake_up_bit(&task->tk_runstate, RPC_TASK_QUEUED);
 }
 
 /*
@@ -578,6 +578,14 @@ static inline int __rpc_do_exit(struct rpc_task *task)
 	return 1;
 }
 
+static int rpc_wait_bit_interruptible(void *word)
+{
+	if (signal_pending(current))
+		return -ERESTARTSYS;
+	schedule();
+	return 0;
+}
+
 /*
  * This is the RPC `scheduler' (or rather, the finite state machine).
  */
@@ -648,22 +656,21 @@ static int __rpc_execute(struct rpc_task *task)
 
 		/* sync task: sleep here */
 		dprintk("RPC: %4d sync task going to sleep\n", task->tk_pid);
-		if (RPC_TASK_UNINTERRUPTIBLE(task)) {
-			__wait_event(task->u.tk_wait.waitq, !RPC_IS_QUEUED(task));
-		} else {
-			__wait_event_interruptible(task->u.tk_wait.waitq, !RPC_IS_QUEUED(task), status);
+		/* Note: Caller should be using rpc_clnt_sigmask() */
+		status = out_of_line_wait_on_bit(&task->tk_runstate,
+				RPC_TASK_QUEUED, rpc_wait_bit_interruptible,
+				TASK_INTERRUPTIBLE);
+		if (status == -ERESTARTSYS) {
 			/*
 			 * When a sync task receives a signal, it exits with
 			 * -ERESTARTSYS. In order to catch any callbacks that
 			 * clean up after sleeping on some queue, we don't
 			 * break the loop here, but go around once more.
 			 */
-			if (status == -ERESTARTSYS) {
-				dprintk("RPC: %4d got signal\n", task->tk_pid);
-				task->tk_flags |= RPC_TASK_KILLED;
-				rpc_exit(task, -ERESTARTSYS);
-				rpc_wake_up_task(task);
-			}
+			dprintk("RPC: %4d got signal\n", task->tk_pid);
+			task->tk_flags |= RPC_TASK_KILLED;
+			rpc_exit(task, -ERESTARTSYS);
+			rpc_wake_up_task(task);
 		}
 		rpc_set_running(task);
 		dprintk("RPC: %4d sync task resuming\n", task->tk_pid);
@@ -766,8 +773,6 @@ void rpc_init_task(struct rpc_task *task, struct rpc_clnt *clnt, rpc_action call
 
 	/* Initialize workqueue for async tasks */
 	task->tk_workqueue = rpciod_workqueue;
-	if (!RPC_IS_ASYNC(task))
-		init_waitqueue_head(&task->u.tk_wait.waitq);
 
 	if (clnt) {
 		atomic_inc(&clnt->cl_users);
-- 
cgit v1.2.3-70-g09d2


From 464a98bd70bae8c559cfc82af799faf44824ce64 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 22 Jun 2005 17:16:21 +0000
Subject: [PATCH] NFS: cleanup: shrink struct nfs_open_context

 Remove the wait queue, and replace the functions that depended on it
 with wait_on_bit().

 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/inode.c         |  1 -
 fs/nfs/pagelist.c      | 35 ++++++++++++++++++++++++++++-------
 include/linux/nfs_fs.h |  1 -
 3 files changed, 28 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index c80a81ff59c..a38d4b22d1f 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -848,7 +848,6 @@ struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, struct rp
 		ctx->state = NULL;
 		ctx->lockowner = current->files;
 		ctx->error = 0;
-		init_waitqueue_head(&ctx->waitq);
 	}
 	return ctx;
 }
diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c
index 4f1ba723848..80777f99a58 100644
--- a/fs/nfs/pagelist.c
+++ b/fs/nfs/pagelist.c
@@ -107,7 +107,7 @@ void nfs_unlock_request(struct nfs_page *req)
 	smp_mb__before_clear_bit();
 	clear_bit(PG_BUSY, &req->wb_flags);
 	smp_mb__after_clear_bit();
-	wake_up_all(&req->wb_context->waitq);
+	wake_up_bit(&req->wb_flags, PG_BUSY);
 	nfs_release_request(req);
 }
 
@@ -180,6 +180,17 @@ nfs_list_add_request(struct nfs_page *req, struct list_head *head)
 	req->wb_list_head = head;
 }
 
+static int nfs_wait_bit_interruptible(void *word)
+{
+	int ret = 0;
+
+	if (signal_pending(current))
+		ret = -ERESTARTSYS;
+	else
+		schedule();
+	return ret;
+}
+
 /**
  * nfs_wait_on_request - Wait for a request to complete.
  * @req: request to wait upon.
@@ -190,12 +201,22 @@ nfs_list_add_request(struct nfs_page *req, struct list_head *head)
 int
 nfs_wait_on_request(struct nfs_page *req)
 {
-	struct inode	*inode = req->wb_context->dentry->d_inode;
-        struct rpc_clnt	*clnt = NFS_CLIENT(inode);
-
-	if (!NFS_WBACK_BUSY(req))
-		return 0;
-	return nfs_wait_event(clnt, req->wb_context->waitq, !NFS_WBACK_BUSY(req));
+        struct rpc_clnt	*clnt = NFS_CLIENT(req->wb_context->dentry->d_inode);
+	sigset_t oldmask;
+	int ret = 0;
+
+	if (!test_bit(PG_BUSY, &req->wb_flags))
+		goto out;
+	/*
+	 * Note: the call to rpc_clnt_sigmask() suffices to ensure that we
+	 *	 are not interrupted if intr flag is not set
+	 */
+	rpc_clnt_sigmask(clnt, &oldmask);
+	ret = out_of_line_wait_on_bit(&req->wb_flags, PG_BUSY,
+			nfs_wait_bit_interruptible, TASK_INTERRUPTIBLE);
+	rpc_clnt_sigunmask(clnt, &oldmask);
+out:
+	return ret;
 }
 
 /**
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 68d5aae8997..0b01b96337f 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -84,7 +84,6 @@ struct nfs_open_context {
 	int error;
 
 	struct list_head list;
-	wait_queue_head_t waitq;
 };
 
 /*
-- 
cgit v1.2.3-70-g09d2


From 92cfc62cb8412c9563860b1bf70cd4701f03092e Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Wed, 22 Jun 2005 17:16:22 +0000
Subject: [PATCH] NFS: Allow NFS versions to support different sets of inode
 operations.

 ACL support will require supporting additional inode operations in v4
 (getxattr, setxattr, listxattr).  This patch allows different protocol versions
 to support different inode operations by adding a file_inode_ops to the
 nfs_rpc_ops (to match the existing dir_inode_ops).

 Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/inode.c          | 2 +-
 fs/nfs/nfs3proc.c       | 1 +
 fs/nfs/nfs4proc.c       | 1 +
 fs/nfs/proc.c           | 1 +
 include/linux/nfs_xdr.h | 1 +
 5 files changed, 5 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index a38d4b22d1f..a82f0340744 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -686,7 +686,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr)
 		/* Why so? Because we want revalidate for devices/FIFOs, and
 		 * that's precisely what we have in nfs_file_inode_operations.
 		 */
-		inode->i_op = &nfs_file_inode_operations;
+		inode->i_op = NFS_SB(sb)->rpc_ops->file_inode_ops;
 		if (S_ISREG(inode->i_mode)) {
 			inode->i_fop = &nfs_file_operations;
 			inode->i_data.a_ops = &nfs_file_aops;
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index 3878494dfc2..53953a77571 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -826,6 +826,7 @@ struct nfs_rpc_ops	nfs_v3_clientops = {
 	.version	= 3,			/* protocol version */
 	.dentry_ops	= &nfs_dentry_operations,
 	.dir_inode_ops	= &nfs_dir_inode_operations,
+	.file_inode_ops	= &nfs_file_inode_operations,
 	.getroot	= nfs3_proc_get_root,
 	.getattr	= nfs3_proc_getattr,
 	.setattr	= nfs3_proc_setattr,
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index a69c02b206c..a5a8cb3159a 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2746,6 +2746,7 @@ struct nfs_rpc_ops	nfs_v4_clientops = {
 	.version	= 4,			/* protocol version */
 	.dentry_ops	= &nfs4_dentry_operations,
 	.dir_inode_ops	= &nfs4_dir_inode_operations,
+	.file_inode_ops	= &nfs_file_inode_operations,
 	.getroot	= nfs4_proc_get_root,
 	.getattr	= nfs4_proc_getattr,
 	.setattr	= nfs4_proc_setattr,
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index d31b4d6e5a5..cedf636bcf3 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -622,6 +622,7 @@ struct nfs_rpc_ops	nfs_v2_clientops = {
 	.version	= 2,		       /* protocol version */
 	.dentry_ops	= &nfs_dentry_operations,
 	.dir_inode_ops	= &nfs_dir_inode_operations,
+	.file_inode_ops	= &nfs_file_inode_operations,
 	.getroot	= nfs_proc_get_root,
 	.getattr	= nfs_proc_getattr,
 	.setattr	= nfs_proc_setattr,
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 47037d9521c..5b45bafd9db 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -667,6 +667,7 @@ struct nfs_rpc_ops {
 	int	version;		/* Protocol version */
 	struct dentry_operations *dentry_ops;
 	struct inode_operations *dir_inode_ops;
+	struct inode_operations *file_inode_ops;
 
 	int	(*getroot) (struct nfs_server *, struct nfs_fh *,
 			    struct nfs_fsinfo *);
-- 
cgit v1.2.3-70-g09d2


From ada70d9425bcc5e376fef8591e4e76e204c0834c Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 22 Jun 2005 17:16:22 +0000
Subject: [PATCH] NFS: Add hooks to allow common NFS attribute code to clear
 cached acls

 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/inode.c          | 33 ++++++++++++++++++++++++++-------
 include/linux/nfs_fs.h  |  1 +
 include/linux/nfs_xdr.h |  1 +
 3 files changed, 28 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index a82f0340744..c45bd52cc1d 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -64,6 +64,7 @@ static void nfs_clear_inode(struct inode *);
 static void nfs_umount_begin(struct super_block *);
 static int  nfs_statfs(struct super_block *, struct kstatfs *);
 static int  nfs_show_options(struct seq_file *, struct vfsmount *);
+static void nfs_zap_acl_cache(struct inode *);
 
 static struct rpc_program	nfs_program;
 
@@ -153,6 +154,7 @@ nfs_clear_inode(struct inode *inode)
 
 	nfs_wb_all(inode);
 	BUG_ON (!list_empty(&nfsi->open_files));
+	nfs_zap_acl_cache(inode);
 	cred = nfsi->cache_access.cred;
 	if (cred)
 		put_rpccred(cred);
@@ -587,9 +589,19 @@ nfs_zap_caches(struct inode *inode)
 
 	memset(NFS_COOKIEVERF(inode), 0, sizeof(NFS_COOKIEVERF(inode)));
 	if (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode))
-		nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS;
+		nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
 	else
-		nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS;
+		nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
+}
+
+static void nfs_zap_acl_cache(struct inode *inode)
+{
+	void (*clear_acl_cache)(struct inode *);
+
+	clear_acl_cache = NFS_PROTO(inode)->clear_acl_cache;
+	if (clear_acl_cache != NULL)
+		clear_acl_cache(inode);
+	NFS_I(inode)->flags &= ~NFS_INO_INVALID_ACL;
 }
 
 /*
@@ -789,7 +801,7 @@ nfs_setattr(struct dentry *dentry, struct iattr *attr)
 		}
 	}
 	if ((attr->ia_valid & (ATTR_MODE|ATTR_UID|ATTR_GID)) != 0)
-		NFS_FLAGS(inode) |= NFS_INO_INVALID_ACCESS;
+		NFS_FLAGS(inode) |= NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
 	nfs_end_data_update(inode);
 	unlock_kernel();
 	return error;
@@ -1033,6 +1045,8 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 		/* This ensures we revalidate dentries */
 		nfsi->cache_change_attribute++;
 	}
+	if (flags & NFS_INO_INVALID_ACL)
+		nfs_zap_acl_cache(inode);
 	dfprintk(PAGECACHE, "NFS: (%s/%Ld) revalidation complete\n",
 		inode->i_sb->s_id,
 		(long long)NFS_FILEID(inode));
@@ -1183,7 +1197,7 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr)
 	if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO)
 			|| inode->i_uid != fattr->uid
 			|| inode->i_gid != fattr->gid)
-		nfsi->flags |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS;
+		nfsi->flags |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL;
 
 	/* Has the link count changed? */
 	if (inode->i_nlink != fattr->nlink)
@@ -1292,16 +1306,21 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign
 #endif
 		nfsi->change_attr = fattr->change_attr;
 		if (!data_unstable)
-			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS;
+			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
 	}
 
-	memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime));
+	/* If ctime has changed we should definitely clear access+acl caches */
+	if (!timespec_equal(&inode->i_ctime, &fattr->ctime)) {
+		if (!data_unstable)
+			invalid |= NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
+		memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime));
+	}
 	memcpy(&inode->i_atime, &fattr->atime, sizeof(inode->i_atime));
 
 	if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO) ||
 	    inode->i_uid != fattr->uid ||
 	    inode->i_gid != fattr->gid)
-		invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS;
+		invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
 
 	inode->i_mode = fattr->mode;
 	inode->i_nlink = fattr->nlink;
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 0b01b96337f..140bdf489f7 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -189,6 +189,7 @@ struct nfs_inode {
 #define NFS_INO_INVALID_DATA	0x0010		/* cached data is invalid */
 #define NFS_INO_INVALID_ATIME	0x0020		/* cached atime is invalid */
 #define NFS_INO_INVALID_ACCESS	0x0040		/* cached access cred invalid */
+#define NFS_INO_INVALID_ACL	0x0080		/* cached acls are invalid */
 
 static inline struct nfs_inode *NFS_I(struct inode *inode)
 {
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 5b45bafd9db..cf38db59f34 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -714,6 +714,7 @@ struct nfs_rpc_ops {
 	int	(*file_open)   (struct inode *, struct file *);
 	int	(*file_release) (struct inode *, struct file *);
 	int	(*lock)(struct file *, int, struct file_lock *);
+	void	(*clear_acl_cache)(struct inode *);
 };
 
 /*
-- 
cgit v1.2.3-70-g09d2


From 029d105e66e5a90850d5a09dad76815d0bcfcaa3 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Wed, 22 Jun 2005 17:16:22 +0000
Subject: [PATCH] NFSv4: Client-side xdr for reading NFSv4 acls

 Client-side support for NFSv4 acls: xdr encoding and decoding routines for
 reading acls

 Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4xdr.c        | 100 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/nfs4.h    |   1 +
 include/linux/nfs_xdr.h |   7 ++++
 3 files changed, 108 insertions(+)

(limited to 'include')

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 8204926bb46..6f1c003ee33 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -365,6 +365,13 @@ static int nfs_stat_to_errno(int);
 				encode_delegreturn_maxsz)
 #define NFS4_dec_delegreturn_sz (compound_decode_hdr_maxsz + \
 				decode_delegreturn_maxsz)
+#define NFS4_enc_getacl_sz	(compound_encode_hdr_maxsz + \
+				encode_putfh_maxsz + \
+				encode_getattr_maxsz)
+#define NFS4_dec_getacl_sz	(compound_decode_hdr_maxsz + \
+				decode_putfh_maxsz + \
+				op_decode_hdr_maxsz + \
+				nfs4_fattr_bitmap_maxsz + 1)
 
 static struct {
 	unsigned int	mode;
@@ -1631,6 +1638,34 @@ out:
         return status;
 }
 
+/*
+ * Encode a GETACL request
+ */
+static int
+nfs4_xdr_enc_getacl(struct rpc_rqst *req, uint32_t *p,
+		struct nfs_getaclargs *args)
+{
+	struct xdr_stream xdr;
+	struct rpc_auth *auth = req->rq_task->tk_auth;
+	struct compound_hdr hdr = {
+		.nops   = 2,
+	};
+	int replen, status;
+
+	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
+	encode_compound_hdr(&xdr, &hdr);
+	status = encode_putfh(&xdr, args->fh);
+	if (status)
+		goto out;
+	status = encode_getattr_two(&xdr, FATTR4_WORD0_ACL, 0);
+	/* set up reply buffer: */
+	replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_getacl_sz) << 2;
+	xdr_inline_pages(&req->rq_rcv_buf, replen,
+		args->acl_pages, args->acl_pgbase, args->acl_len);
+out:
+	return status;
+}
+
 /*
  * Encode a WRITE request
  */
@@ -3125,6 +3160,47 @@ static int decode_renew(struct xdr_stream *xdr)
 	return decode_op_hdr(xdr, OP_RENEW);
 }
 
+static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req,
+		size_t *acl_len)
+{
+	uint32_t *savep;
+	uint32_t attrlen,
+		 bitmap[2] = {0};
+	struct kvec *iov = req->rq_rcv_buf.head;
+	int status;
+
+	*acl_len = 0;
+	if ((status = decode_op_hdr(xdr, OP_GETATTR)) != 0)
+		goto out;
+	if ((status = decode_attr_bitmap(xdr, bitmap)) != 0)
+		goto out;
+	if ((status = decode_attr_length(xdr, &attrlen, &savep)) != 0)
+		goto out;
+
+	if (unlikely(bitmap[0] & (FATTR4_WORD0_ACL - 1U)))
+		return -EIO;
+	if (likely(bitmap[0] & FATTR4_WORD0_ACL)) {
+		int hdrlen, recvd;
+
+		/* We ignore &savep and don't do consistency checks on
+		 * the attr length.  Let userspace figure it out.... */
+		hdrlen = (u8 *)xdr->p - (u8 *)iov->iov_base;
+		recvd = req->rq_rcv_buf.len - hdrlen;
+		if (attrlen > recvd) {
+			printk(KERN_WARNING "NFS: server cheating in getattr"
+					" acl reply: attrlen %u > recvd %u\n",
+					attrlen, recvd);
+			return -EINVAL;
+		}
+		if (attrlen <= *acl_len)
+			xdr_read_pages(xdr, attrlen);
+		*acl_len = attrlen;
+	}
+
+out:
+	return status;
+}
+
 static int
 decode_savefh(struct xdr_stream *xdr)
 {
@@ -3417,6 +3493,29 @@ out:
 }
 
 
+/*
+ * Decode GETACL response
+ */
+static int
+nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, uint32_t *p, size_t *acl_len)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr;
+	int status;
+
+	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_putfh(&xdr);
+	if (status)
+		goto out;
+	status = decode_getacl(&xdr, rqstp, acl_len);
+
+out:
+	return status;
+}
+
 /*
  * Decode CLOSE response
  */
@@ -4017,6 +4116,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
   PROC(READDIR,		enc_readdir,	dec_readdir),
   PROC(SERVER_CAPS,	enc_server_caps, dec_server_caps),
   PROC(DELEGRETURN,	enc_delegreturn, dec_delegreturn),
+  PROC(GETACL,		enc_getacl,	dec_getacl),
 };
 
 struct rpc_version		nfs_version4 = {
diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index 5ca8a8d8ccd..6ee7e2585af 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -382,6 +382,7 @@ enum {
 	NFSPROC4_CLNT_READDIR,
 	NFSPROC4_CLNT_SERVER_CAPS,
 	NFSPROC4_CLNT_DELEGRETURN,
+	NFSPROC4_CLNT_GETACL,
 };
 
 #endif
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index cf38db59f34..9f5e1d407c7 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -326,6 +326,13 @@ struct nfs_setattrargs {
 	const u32 *			bitmask;
 };
 
+struct nfs_getaclargs {
+	struct nfs_fh *			fh;
+	size_t				acl_len;
+	unsigned int			acl_pgbase;
+	struct page **			acl_pages;
+};
+
 struct nfs_setattrres {
 	struct nfs_fattr *              fattr;
 	const struct nfs_server *	server;
-- 
cgit v1.2.3-70-g09d2


From 23ec6965c20db96bc8ea7af0ec178f074dd31c40 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Wed, 22 Jun 2005 17:16:22 +0000
Subject: [PATCH] NFSv4: Client-side xdr for writing NFSv4 acls

 Client-side support for NFSv4 acls: xdr encoding and decoding routines for
 writing acls

 Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4xdr.c        | 71 ++++++++++++++++++++++++++++++++++++++++++++++++-
 include/linux/nfs4.h    |  1 +
 include/linux/nfs_xdr.h |  7 +++++
 3 files changed, 78 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 6f1c003ee33..325cd6d4f23 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -372,6 +372,13 @@ static int nfs_stat_to_errno(int);
 				decode_putfh_maxsz + \
 				op_decode_hdr_maxsz + \
 				nfs4_fattr_bitmap_maxsz + 1)
+#define NFS4_enc_setacl_sz	(compound_encode_hdr_maxsz + \
+				encode_putfh_maxsz + \
+				op_encode_hdr_maxsz + 4 + \
+				nfs4_fattr_bitmap_maxsz + 1)
+#define NFS4_dec_setacl_sz	(compound_decode_hdr_maxsz + \
+				decode_putfh_maxsz + \
+				op_decode_hdr_maxsz + nfs4_fattr_bitmap_maxsz)
 
 static struct {
 	unsigned int	mode;
@@ -471,7 +478,7 @@ static int encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const s
 	 * In the worst-case, this would be
 	 *   12(bitmap) + 4(attrlen) + 8(size) + 4(mode) + 4(atime) + 4(mtime)
 	 *          = 36 bytes, plus any contribution from variable-length fields
-	 *            such as owner/group/acl's.
+	 *            such as owner/group.
 	 */
 	len = 16;
 
@@ -1095,6 +1102,25 @@ static int encode_renew(struct xdr_stream *xdr, const struct nfs4_client *client
 	return 0;
 }
 
+static int
+encode_setacl(struct xdr_stream *xdr, struct nfs_setaclargs *arg)
+{
+	uint32_t *p;
+
+	RESERVE_SPACE(4+sizeof(zero_stateid.data));
+	WRITE32(OP_SETATTR);
+	WRITEMEM(zero_stateid.data, sizeof(zero_stateid.data));
+	RESERVE_SPACE(2*4);
+	WRITE32(1);
+	WRITE32(FATTR4_WORD0_ACL);
+	if (arg->acl_len % 4)
+		return -EINVAL;
+	RESERVE_SPACE(4);
+	WRITE32(arg->acl_len);
+	xdr_write_pages(xdr, arg->acl_pages, arg->acl_pgbase, arg->acl_len);
+	return 0;
+}
+
 static int
 encode_savefh(struct xdr_stream *xdr)
 {
@@ -3492,6 +3518,48 @@ out:
 
 }
 
+/*
+ * Encode an SETACL request
+ */
+static int
+nfs4_xdr_enc_setacl(struct rpc_rqst *req, uint32_t *p, struct nfs_setaclargs *args)
+{
+        struct xdr_stream xdr;
+        struct compound_hdr hdr = {
+                .nops   = 2,
+        };
+        int status;
+
+        xdr_init_encode(&xdr, &req->rq_snd_buf, p);
+        encode_compound_hdr(&xdr, &hdr);
+        status = encode_putfh(&xdr, args->fh);
+        if (status)
+                goto out;
+        status = encode_setacl(&xdr, args);
+out:
+        return status;
+}
+/*
+ * Decode SETACL response
+ */
+static int
+nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, uint32_t *p, void *res)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr;
+	int status;
+
+	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_putfh(&xdr);
+	if (status)
+		goto out;
+	status = decode_setattr(&xdr, res);
+out:
+	return status;
+}
 
 /*
  * Decode GETACL response
@@ -4117,6 +4185,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
   PROC(SERVER_CAPS,	enc_server_caps, dec_server_caps),
   PROC(DELEGRETURN,	enc_delegreturn, dec_delegreturn),
   PROC(GETACL,		enc_getacl,	dec_getacl),
+  PROC(SETACL,		enc_setacl,	dec_setacl),
 };
 
 struct rpc_version		nfs_version4 = {
diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index 6ee7e2585af..5bb5b2fd7ba 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -383,6 +383,7 @@ enum {
 	NFSPROC4_CLNT_SERVER_CAPS,
 	NFSPROC4_CLNT_DELEGRETURN,
 	NFSPROC4_CLNT_GETACL,
+	NFSPROC4_CLNT_SETACL,
 };
 
 #endif
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 9f5e1d407c7..46b206b460c 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -326,6 +326,13 @@ struct nfs_setattrargs {
 	const u32 *			bitmask;
 };
 
+struct nfs_setaclargs {
+	struct nfs_fh *			fh;
+	size_t				acl_len;
+	unsigned int			acl_pgbase;
+	struct page **			acl_pages;
+};
+
 struct nfs_getaclargs {
 	struct nfs_fh *			fh;
 	size_t				acl_len;
-- 
cgit v1.2.3-70-g09d2


From e50a1c2e1f816c81eed6a589019052cb44189267 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Wed, 22 Jun 2005 17:16:23 +0000
Subject: [PATCH] NFSv4: client-side caching NFSv4 ACLs

 Add nfs4_acl field to the nfs_inode, and use it to cache acls.  Only cache
 acls of size up to a page.  Also prepare for up to a page of acl data even
 when the user doesn't pass in a buffer, as when they want to get the acl
 length to decide what size buffer to allocate.

 Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/inode.c         |   7 ++-
 fs/nfs/nfs4proc.c      | 129 +++++++++++++++++++++++++++++++++++++++++++++----
 include/linux/nfs_fs.h |   2 +-
 3 files changed, 124 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index c45bd52cc1d..350c48c1263 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -142,10 +142,6 @@ nfs_delete_inode(struct inode * inode)
 	clear_inode(inode);
 }
 
-/*
- * For the moment, the only task for the NFS clear_inode method is to
- * release the mmap credential
- */
 static void
 nfs_clear_inode(struct inode *inode)
 {
@@ -1923,6 +1919,9 @@ static struct inode *nfs_alloc_inode(struct super_block *sb)
 	if (!nfsi)
 		return NULL;
 	nfsi->flags = 0;
+#ifdef CONFIG_NFS_V4
+	nfsi->nfs4_acl = NULL;
+#endif /* CONFIG_NFS_V4 */
 	return &nfsi->vfs_inode;
 }
 
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index d969dd13e7d..128d01cfea1 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2188,9 +2188,75 @@ static void buf_to_pages(const void *buf, size_t buflen,
 	}
 }
 
-static ssize_t nfs4_proc_get_acl(struct inode *inode, void *buf, size_t buflen)
+struct nfs4_cached_acl {
+	int cached;
+	size_t len;
+	char data[];
+};
+
+static void nfs4_set_cached_acl(struct inode *inode, struct nfs4_cached_acl *acl)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+
+	spin_lock(&inode->i_lock);
+	kfree(nfsi->nfs4_acl);
+	nfsi->nfs4_acl = acl;
+	spin_unlock(&inode->i_lock);
+}
+
+static void nfs4_zap_acl_attr(struct inode *inode)
+{
+	nfs4_set_cached_acl(inode, NULL);
+}
+
+static inline ssize_t nfs4_read_cached_acl(struct inode *inode, char *buf, size_t buflen)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+	struct nfs4_cached_acl *acl;
+	int ret = -ENOENT;
+
+	spin_lock(&inode->i_lock);
+	acl = nfsi->nfs4_acl;
+	if (acl == NULL)
+		goto out;
+	if (buf == NULL) /* user is just asking for length */
+		goto out_len;
+	if (acl->cached == 0)
+		goto out;
+	ret = -ERANGE; /* see getxattr(2) man page */
+	if (acl->len > buflen)
+		goto out;
+	memcpy(buf, acl->data, acl->len);
+out_len:
+	ret = acl->len;
+out:
+	spin_unlock(&inode->i_lock);
+	return ret;
+}
+
+static void nfs4_write_cached_acl(struct inode *inode, const char *buf, size_t acl_len)
+{
+	struct nfs4_cached_acl *acl;
+
+	if (buf && acl_len <= PAGE_SIZE) {
+		acl = kmalloc(sizeof(*acl) + acl_len, GFP_KERNEL);
+		if (acl == NULL)
+			goto out;
+		acl->cached = 1;
+		memcpy(acl->data, buf, acl_len);
+	} else {
+		acl = kmalloc(sizeof(*acl), GFP_KERNEL);
+		if (acl == NULL)
+			goto out;
+		acl->cached = 0;
+	}
+	acl->len = acl_len;
+out:
+	nfs4_set_cached_acl(inode, acl);
+}
+
+static inline ssize_t nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t buflen)
 {
-	struct nfs_server *server = NFS_SERVER(inode);
 	struct page *pages[NFS4ACL_MAXPAGES];
 	struct nfs_getaclargs args = {
 		.fh = NFS_FH(inode),
@@ -2198,24 +2264,66 @@ static ssize_t nfs4_proc_get_acl(struct inode *inode, void *buf, size_t buflen)
 		.acl_len = buflen,
 	};
 	size_t resp_len = buflen;
+	void *resp_buf;
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GETACL],
 		.rpc_argp = &args,
 		.rpc_resp = &resp_len,
 	};
+	struct page *localpage = NULL;
 	int ret;
 
-	if (!nfs4_server_supports_acls(server))
-		return -EOPNOTSUPP;
-	buf_to_pages(buf, buflen, args.acl_pages, &args.acl_pgbase);
+	if (buflen < PAGE_SIZE) {
+		/* As long as we're doing a round trip to the server anyway,
+		 * let's be prepared for a page of acl data. */
+		localpage = alloc_page(GFP_KERNEL);
+		resp_buf = page_address(localpage);
+		if (localpage == NULL)
+			return -ENOMEM;
+		args.acl_pages[0] = localpage;
+		args.acl_pgbase = 0;
+		args.acl_len = PAGE_SIZE;
+	} else {
+		resp_buf = buf;
+		buf_to_pages(buf, buflen, args.acl_pages, &args.acl_pgbase);
+	}
 	ret = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
-	if (buflen && resp_len > buflen)
-		return -ERANGE;
-	if (ret == 0)
-		ret = resp_len;
+	if (ret)
+		goto out_free;
+	if (resp_len > args.acl_len)
+		nfs4_write_cached_acl(inode, NULL, resp_len);
+	else
+		nfs4_write_cached_acl(inode, resp_buf, resp_len);
+	if (buf) {
+		ret = -ERANGE;
+		if (resp_len > buflen)
+			goto out_free;
+		if (localpage)
+			memcpy(buf, resp_buf, resp_len);
+	}
+	ret = resp_len;
+out_free:
+	if (localpage)
+		__free_page(localpage);
 	return ret;
 }
 
+static ssize_t nfs4_proc_get_acl(struct inode *inode, void *buf, size_t buflen)
+{
+	struct nfs_server *server = NFS_SERVER(inode);
+	int ret;
+
+	if (!nfs4_server_supports_acls(server))
+		return -EOPNOTSUPP;
+	ret = nfs_revalidate_inode(server, inode);
+	if (ret < 0)
+		return ret;
+	ret = nfs4_read_cached_acl(inode, buf, buflen);
+	if (ret != -ENOENT)
+		return ret;
+	return nfs4_get_acl_uncached(inode, buf, buflen);
+}
+
 static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen)
 {
 	struct nfs_server *server = NFS_SERVER(inode);
@@ -2236,6 +2344,8 @@ static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen
 		return -EOPNOTSUPP;
 	buf_to_pages(buf, buflen, arg.acl_pages, &arg.acl_pgbase);
 	ret = rpc_call_sync(NFS_SERVER(inode)->client, &msg, 0);
+	if (ret == 0)
+		nfs4_write_cached_acl(inode, buf, buflen);
 	return ret;
 }
 
@@ -2907,6 +3017,7 @@ struct nfs_rpc_ops	nfs_v4_clientops = {
 	.file_open      = nfs4_proc_file_open,
 	.file_release   = nfs4_proc_file_release,
 	.lock		= nfs4_proc_lock,
+	.clear_acl_cache = nfs4_zap_acl_attr,
 };
 
 /*
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 140bdf489f7..d2b5d7e0e85 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -169,13 +169,13 @@ struct nfs_inode {
 	wait_queue_head_t	nfs_i_wait;
 
 #ifdef CONFIG_NFS_V4
+	struct nfs4_cached_acl	*nfs4_acl;
         /* NFSv4 state */
 	struct list_head	open_states;
 	struct nfs_delegation	*delegation;
 	int			 delegation_state;
 	struct rw_semaphore	rwsem;
 #endif /* CONFIG_NFS_V4*/
-
 	struct inode		vfs_inode;
 };
 
-- 
cgit v1.2.3-70-g09d2


From 007e251f2b2760f738c92adc8c80cbae0bed3ce5 Mon Sep 17 00:00:00 2001
From: Andreas Gruenbacher <agruen@suse.de>
Date: Wed, 22 Jun 2005 17:16:23 +0000
Subject: [PATCH] RPC: Allow multiple RPC client programs to share the same
 transport

 Signed-off-by: Andreas Gruenbacher <agruen@suse.de>
 Acked-by: Olaf Kirch <okir@suse.de>
 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/clnt.h |  2 ++
 net/sunrpc/clnt.c           | 40 ++++++++++++++++++++++++++++++++++++++++
 net/sunrpc/pmap_clnt.c      |  3 +++
 net/sunrpc/sunrpc_syms.c    |  1 +
 4 files changed, 46 insertions(+)

(limited to 'include')

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index d25e80f77ff..ab151bbb66d 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -114,6 +114,8 @@ struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname,
 struct rpc_clnt *rpc_new_client(struct rpc_xprt *xprt, char *servname,
 				struct rpc_program *info,
 				u32 version, rpc_authflavor_t authflavor);
+struct rpc_clnt	*rpc_bind_new_program(struct rpc_clnt *,
+				struct rpc_program *, int);
 struct rpc_clnt *rpc_clone_client(struct rpc_clnt *);
 int		rpc_shutdown_client(struct rpc_clnt *);
 int		rpc_destroy_client(struct rpc_clnt *);
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 33f12b84e26..c979fcf8879 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -241,6 +241,8 @@ rpc_clone_client(struct rpc_clnt *clnt)
 	rpc_init_rtt(&new->cl_rtt_default, clnt->cl_xprt->timeout.to_initval);
 	if (new->cl_auth)
 		atomic_inc(&new->cl_auth->au_count);
+	new->cl_pmap		= &new->cl_pmap_default;
+	rpc_init_wait_queue(&new->cl_pmap_default.pm_bindwait, "bindwait");
 	return new;
 out_no_clnt:
 	printk(KERN_INFO "RPC: out of memory in %s\n", __FUNCTION__);
@@ -329,6 +331,44 @@ rpc_release_client(struct rpc_clnt *clnt)
 		rpc_destroy_client(clnt);
 }
 
+/**
+ * rpc_bind_new_program - bind a new RPC program to an existing client
+ * @old - old rpc_client
+ * @program - rpc program to set
+ * @vers - rpc program version
+ *
+ * Clones the rpc client and sets up a new RPC program. This is mainly
+ * of use for enabling different RPC programs to share the same transport.
+ * The Sun NFSv2/v3 ACL protocol can do this.
+ */
+struct rpc_clnt *rpc_bind_new_program(struct rpc_clnt *old,
+				      struct rpc_program *program,
+				      int vers)
+{
+	struct rpc_clnt *clnt;
+	struct rpc_version *version;
+	int err;
+
+	BUG_ON(vers >= program->nrvers || !program->version[vers]);
+	version = program->version[vers];
+	clnt = rpc_clone_client(old);
+	if (IS_ERR(clnt))
+		goto out;
+	clnt->cl_procinfo = version->procs;
+	clnt->cl_maxproc  = version->nrprocs;
+	clnt->cl_protname = program->name;
+	clnt->cl_prog     = program->number;
+	clnt->cl_vers     = version->number;
+	clnt->cl_stats    = program->stats;
+	err = rpc_ping(clnt, RPC_TASK_SOFT|RPC_TASK_NOINTR);
+	if (err != 0) {
+		rpc_shutdown_client(clnt);
+		clnt = ERR_PTR(err);
+	}
+out:	
+	return clnt;
+}
+
 /*
  * Default callback for async RPC calls
  */
diff --git a/net/sunrpc/pmap_clnt.c b/net/sunrpc/pmap_clnt.c
index df4d84c9020..4e81f276692 100644
--- a/net/sunrpc/pmap_clnt.c
+++ b/net/sunrpc/pmap_clnt.c
@@ -53,6 +53,9 @@ rpc_getport(struct rpc_task *task, struct rpc_clnt *clnt)
 			task->tk_pid, clnt->cl_server,
 			map->pm_prog, map->pm_vers, map->pm_prot);
 
+	/* Autobind on cloned rpc clients is discouraged */
+	BUG_ON(clnt->cl_parent != clnt);
+
 	spin_lock(&pmap_lock);
 	if (map->pm_binding) {
 		rpc_sleep_on(&map->pm_bindwait, task, NULL, NULL);
diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c
index 1b0ff7e0e86..d8673f66acc 100644
--- a/net/sunrpc/sunrpc_syms.c
+++ b/net/sunrpc/sunrpc_syms.c
@@ -42,6 +42,7 @@ EXPORT_SYMBOL(rpc_release_task);
 /* RPC client functions */
 EXPORT_SYMBOL(rpc_create_client);
 EXPORT_SYMBOL(rpc_clone_client);
+EXPORT_SYMBOL(rpc_bind_new_program);
 EXPORT_SYMBOL(rpc_destroy_client);
 EXPORT_SYMBOL(rpc_shutdown_client);
 EXPORT_SYMBOL(rpc_release_client);
-- 
cgit v1.2.3-70-g09d2


From e053d1ab62c8ef0eff3dd4c95448cad3c6d2fbf4 Mon Sep 17 00:00:00 2001
From: Olaf Kirch <okir@suse.de>
Date: Wed, 22 Jun 2005 17:16:24 +0000
Subject: [PATCH] RPC: Lazy RPC receive buffer allocation

 Signed-off-by: Olaf Kirch <okir@suse.de>
 Signed-off-by: Andreas Gruenbacher <agruen@suse.de>
 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/xdr.h |  2 +-
 net/sunrpc/xdr.c           | 16 +++++++++++++---
 net/sunrpc/xprt.c          | 26 ++++++++++++++++++++++----
 3 files changed, 36 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h
index 541dcf838ab..0f5b7a5a743 100644
--- a/include/linux/sunrpc/xdr.h
+++ b/include/linux/sunrpc/xdr.h
@@ -160,7 +160,7 @@ typedef struct {
 
 typedef size_t (*skb_read_actor_t)(skb_reader_t *desc, void *to, size_t len);
 
-extern void xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int,
+extern int xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int,
 		skb_reader_t *, skb_read_actor_t);
 
 struct socket;
diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c
index f86d1baa630..65b268d3978 100644
--- a/net/sunrpc/xdr.c
+++ b/net/sunrpc/xdr.c
@@ -176,7 +176,7 @@ xdr_inline_pages(struct xdr_buf *xdr, unsigned int offset,
 	xdr->buflen += len;
 }
 
-void
+int
 xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base,
 			  skb_reader_t *desc,
 			  skb_read_actor_t copy_actor)
@@ -190,7 +190,7 @@ xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base,
 		len -= base;
 		ret = copy_actor(desc, (char *)xdr->head[0].iov_base + base, len);
 		if (ret != len || !desc->count)
-			return;
+			return 0;
 		base = 0;
 	} else
 		base -= len;
@@ -210,6 +210,14 @@ xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base,
 	do {
 		char *kaddr;
 
+		/* ACL likes to be lazy in allocating pages - ACLs
+		 * are small by default but can get huge. */
+		if (unlikely(*ppage == NULL)) {
+			*ppage = alloc_page(GFP_ATOMIC);
+			if (unlikely(*ppage == NULL))
+				return -ENOMEM;
+		}
+
 		len = PAGE_CACHE_SIZE;
 		kaddr = kmap_atomic(*ppage, KM_SKB_SUNRPC_DATA);
 		if (base) {
@@ -226,13 +234,15 @@ xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base,
 		flush_dcache_page(*ppage);
 		kunmap_atomic(kaddr, KM_SKB_SUNRPC_DATA);
 		if (ret != len || !desc->count)
-			return;
+			return 0;
 		ppage++;
 	} while ((pglen -= len) != 0);
 copy_tail:
 	len = xdr->tail[0].iov_len;
 	if (base < len)
 		copy_actor(desc, (char *)xdr->tail[0].iov_base + base, len - base);
+
+	return 0;
 }
 
 
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index c74a6bb9407..a180ed4952d 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -725,7 +725,8 @@ csum_partial_copy_to_xdr(struct xdr_buf *xdr, struct sk_buff *skb)
 		goto no_checksum;
 
 	desc.csum = csum_partial(skb->data, desc.offset, skb->csum);
-	xdr_partial_copy_from_skb(xdr, 0, &desc, skb_read_and_csum_bits);
+	if (xdr_partial_copy_from_skb(xdr, 0, &desc, skb_read_and_csum_bits) < 0)
+		return -1;
 	if (desc.offset != skb->len) {
 		unsigned int csum2;
 		csum2 = skb_checksum(skb, desc.offset, skb->len - desc.offset, 0);
@@ -737,7 +738,8 @@ csum_partial_copy_to_xdr(struct xdr_buf *xdr, struct sk_buff *skb)
 		return -1;
 	return 0;
 no_checksum:
-	xdr_partial_copy_from_skb(xdr, 0, &desc, skb_read_bits);
+	if (xdr_partial_copy_from_skb(xdr, 0, &desc, skb_read_bits) < 0)
+		return -1;
 	if (desc.count)
 		return -1;
 	return 0;
@@ -907,6 +909,7 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc)
 	struct rpc_rqst *req;
 	struct xdr_buf *rcvbuf;
 	size_t len;
+	int r;
 
 	/* Find and lock the request corresponding to this xid */
 	spin_lock(&xprt->sock_lock);
@@ -927,16 +930,30 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc)
 		len = xprt->tcp_reclen - xprt->tcp_offset;
 		memcpy(&my_desc, desc, sizeof(my_desc));
 		my_desc.count = len;
-		xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied,
+		r = xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied,
 					  &my_desc, tcp_copy_data);
 		desc->count -= len;
 		desc->offset += len;
 	} else
-		xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied,
+		r = xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied,
 					  desc, tcp_copy_data);
 	xprt->tcp_copied += len;
 	xprt->tcp_offset += len;
 
+	if (r < 0) {
+		/* Error when copying to the receive buffer,
+		 * usually because we weren't able to allocate
+		 * additional buffer pages. All we can do now
+		 * is turn off XPRT_COPY_DATA, so the request
+		 * will not receive any additional updates,
+		 * and time out.
+		 * Any remaining data from this record will
+		 * be discarded.
+		 */
+		xprt->tcp_flags &= ~XPRT_COPY_DATA;
+		goto out;
+	}
+
 	if (xprt->tcp_copied == req->rq_private_buf.buflen)
 		xprt->tcp_flags &= ~XPRT_COPY_DATA;
 	else if (xprt->tcp_offset == xprt->tcp_reclen) {
@@ -949,6 +966,7 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc)
 				req->rq_task->tk_pid);
 		xprt_complete_rqst(xprt, req, xprt->tcp_copied);
 	}
+out:
 	spin_unlock(&xprt->sock_lock);
 	tcp_check_recm(xprt);
 }
-- 
cgit v1.2.3-70-g09d2


From 7e06b53d796a3740307b54aa2799077f8a0c84e7 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 22 Jun 2005 17:16:24 +0000
Subject: [PATCH] RPC: fix accounting bug in the case of a truncated RPC
 message

 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/xdr.h |  2 +-
 net/sunrpc/xdr.c           | 22 ++++++++++++++--------
 net/sunrpc/xprt.c          | 35 +++++++++++++++++++++++++++--------
 3 files changed, 42 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h
index 0f5b7a5a743..5d1eed2b58a 100644
--- a/include/linux/sunrpc/xdr.h
+++ b/include/linux/sunrpc/xdr.h
@@ -160,7 +160,7 @@ typedef struct {
 
 typedef size_t (*skb_read_actor_t)(skb_reader_t *desc, void *to, size_t len);
 
-extern int xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int,
+extern ssize_t xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int,
 		skb_reader_t *, skb_read_actor_t);
 
 struct socket;
diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c
index 65b268d3978..b3ac3f72bf9 100644
--- a/net/sunrpc/xdr.c
+++ b/net/sunrpc/xdr.c
@@ -176,21 +176,23 @@ xdr_inline_pages(struct xdr_buf *xdr, unsigned int offset,
 	xdr->buflen += len;
 }
 
-int
+ssize_t
 xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base,
 			  skb_reader_t *desc,
 			  skb_read_actor_t copy_actor)
 {
 	struct page	**ppage = xdr->pages;
 	unsigned int	len, pglen = xdr->page_len;
+	ssize_t		copied = 0;
 	int		ret;
 
 	len = xdr->head[0].iov_len;
 	if (base < len) {
 		len -= base;
 		ret = copy_actor(desc, (char *)xdr->head[0].iov_base + base, len);
+		copied += ret;
 		if (ret != len || !desc->count)
-			return 0;
+			goto out;
 		base = 0;
 	} else
 		base -= len;
@@ -214,8 +216,11 @@ xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base,
 		 * are small by default but can get huge. */
 		if (unlikely(*ppage == NULL)) {
 			*ppage = alloc_page(GFP_ATOMIC);
-			if (unlikely(*ppage == NULL))
-				return -ENOMEM;
+			if (unlikely(*ppage == NULL)) {
+				if (copied == 0)
+					copied = -ENOMEM;
+				goto out;
+			}
 		}
 
 		len = PAGE_CACHE_SIZE;
@@ -233,16 +238,17 @@ xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base,
 		}
 		flush_dcache_page(*ppage);
 		kunmap_atomic(kaddr, KM_SKB_SUNRPC_DATA);
+		copied += ret;
 		if (ret != len || !desc->count)
-			return 0;
+			goto out;
 		ppage++;
 	} while ((pglen -= len) != 0);
 copy_tail:
 	len = xdr->tail[0].iov_len;
 	if (base < len)
-		copy_actor(desc, (char *)xdr->tail[0].iov_base + base, len - base);
-
-	return 0;
+		copied += copy_actor(desc, (char *)xdr->tail[0].iov_base + base, len - base);
+out:
+	return copied;
 }
 
 
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index a180ed4952d..ef941e7de8b 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -823,10 +823,15 @@ tcp_copy_data(skb_reader_t *desc, void *p, size_t len)
 {
 	if (len > desc->count)
 		len = desc->count;
-	if (skb_copy_bits(desc->skb, desc->offset, p, len))
+	if (skb_copy_bits(desc->skb, desc->offset, p, len)) {
+		dprintk("RPC:      failed to copy %zu bytes from skb. %zu bytes remain\n",
+				len, desc->count);
 		return 0;
+	}
 	desc->offset += len;
 	desc->count -= len;
+	dprintk("RPC:      copied %zu bytes from skb. %zu bytes remain\n",
+			len, desc->count);
 	return len;
 }
 
@@ -865,6 +870,8 @@ tcp_read_fraghdr(struct rpc_xprt *xprt, skb_reader_t *desc)
 static void
 tcp_check_recm(struct rpc_xprt *xprt)
 {
+	dprintk("RPC:      xprt = %p, tcp_copied = %lu, tcp_offset = %u, tcp_reclen = %u, tcp_flags = %lx\n",
+			xprt, xprt->tcp_copied, xprt->tcp_offset, xprt->tcp_reclen, xprt->tcp_flags);
 	if (xprt->tcp_offset == xprt->tcp_reclen) {
 		xprt->tcp_flags |= XPRT_COPY_RECM;
 		xprt->tcp_offset = 0;
@@ -909,7 +916,7 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc)
 	struct rpc_rqst *req;
 	struct xdr_buf *rcvbuf;
 	size_t len;
-	int r;
+	ssize_t r;
 
 	/* Find and lock the request corresponding to this xid */
 	spin_lock(&xprt->sock_lock);
@@ -932,15 +939,17 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc)
 		my_desc.count = len;
 		r = xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied,
 					  &my_desc, tcp_copy_data);
-		desc->count -= len;
-		desc->offset += len;
+		desc->count -= r;
+		desc->offset += r;
 	} else
 		r = xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied,
 					  desc, tcp_copy_data);
-	xprt->tcp_copied += len;
-	xprt->tcp_offset += len;
 
-	if (r < 0) {
+	if (r > 0) {
+		xprt->tcp_copied += r;
+		xprt->tcp_offset += r;
+	}
+	if (r != len) {
 		/* Error when copying to the receive buffer,
 		 * usually because we weren't able to allocate
 		 * additional buffer pages. All we can do now
@@ -951,9 +960,18 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc)
 		 * be discarded.
 		 */
 		xprt->tcp_flags &= ~XPRT_COPY_DATA;
+		dprintk("RPC:      XID %08x truncated request\n",
+				ntohl(xprt->tcp_xid));
+		dprintk("RPC:      xprt = %p, tcp_copied = %lu, tcp_offset = %u, tcp_reclen = %u\n",
+				xprt, xprt->tcp_copied, xprt->tcp_offset, xprt->tcp_reclen);
 		goto out;
 	}
 
+	dprintk("RPC:      XID %08x read %u bytes\n",
+			ntohl(xprt->tcp_xid), r);
+	dprintk("RPC:      xprt = %p, tcp_copied = %lu, tcp_offset = %u, tcp_reclen = %u\n",
+			xprt, xprt->tcp_copied, xprt->tcp_offset, xprt->tcp_reclen);
+
 	if (xprt->tcp_copied == req->rq_private_buf.buflen)
 		xprt->tcp_flags &= ~XPRT_COPY_DATA;
 	else if (xprt->tcp_offset == xprt->tcp_reclen) {
@@ -961,12 +979,12 @@ tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc)
 			xprt->tcp_flags &= ~XPRT_COPY_DATA;
 	}
 
+out:
 	if (!(xprt->tcp_flags & XPRT_COPY_DATA)) {
 		dprintk("RPC: %4d received reply complete\n",
 				req->rq_task->tk_pid);
 		xprt_complete_rqst(xprt, req, xprt->tcp_copied);
 	}
-out:
 	spin_unlock(&xprt->sock_lock);
 	tcp_check_recm(xprt);
 }
@@ -985,6 +1003,7 @@ tcp_read_discard(struct rpc_xprt *xprt, skb_reader_t *desc)
 	desc->count -= len;
 	desc->offset += len;
 	xprt->tcp_offset += len;
+	dprintk("RPC:      discarded %u bytes\n", len);
 	tcp_check_recm(xprt);
 }
 
-- 
cgit v1.2.3-70-g09d2


From bd8100e7eda87507649c6ba4cb32173b34e49986 Mon Sep 17 00:00:00 2001
From: Andreas Gruenbacher <agruen@suse.de>
Date: Wed, 22 Jun 2005 17:16:24 +0000
Subject: [PATCH] RPC: Encode and decode arbitrary XDR arrays

 Signed-off-by: Andreas Gruenbacher <agruen@suse.de>
 Acked-by: Olaf Kirch <okir@suse.de>
 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/xdr.h |  19 +++-
 net/sunrpc/sunrpc_syms.c   |   4 +
 net/sunrpc/xdr.c           | 256 ++++++++++++++++++++++++++++++++++++++++++++-
 3 files changed, 275 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h
index 5d1eed2b58a..34ec3e8d99b 100644
--- a/include/linux/sunrpc/xdr.h
+++ b/include/linux/sunrpc/xdr.h
@@ -146,7 +146,8 @@ extern void xdr_shift_buf(struct xdr_buf *, size_t);
 extern void xdr_buf_from_iov(struct kvec *, struct xdr_buf *);
 extern int xdr_buf_subsegment(struct xdr_buf *, struct xdr_buf *, int, int);
 extern int xdr_buf_read_netobj(struct xdr_buf *, struct xdr_netobj *, int);
-extern int read_bytes_from_xdr_buf(struct xdr_buf *buf, int base, void *obj, int len);
+extern int read_bytes_from_xdr_buf(struct xdr_buf *, int, void *, int);
+extern int write_bytes_to_xdr_buf(struct xdr_buf *, int, void *, int);
 
 /*
  * Helper structure for copying from an sk_buff.
@@ -168,6 +169,22 @@ struct sockaddr;
 extern int xdr_sendpages(struct socket *, struct sockaddr *, int,
 		struct xdr_buf *, unsigned int, int);
 
+extern int xdr_encode_word(struct xdr_buf *, int, u32);
+extern int xdr_decode_word(struct xdr_buf *, int, u32 *);
+
+struct xdr_array2_desc;
+typedef int (*xdr_xcode_elem_t)(struct xdr_array2_desc *desc, void *elem);
+struct xdr_array2_desc {
+	unsigned int elem_size;
+	unsigned int array_len;
+	xdr_xcode_elem_t xcode;
+};
+
+extern int xdr_decode_array2(struct xdr_buf *buf, unsigned int base,
+                             struct xdr_array2_desc *desc);
+extern int xdr_encode_array2(struct xdr_buf *buf, unsigned int base,
+			     struct xdr_array2_desc *desc);
+
 /*
  * Provide some simple tools for XDR buffer overflow-checking etc.
  */
diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c
index d8673f66acc..32e8acbc60f 100644
--- a/net/sunrpc/sunrpc_syms.c
+++ b/net/sunrpc/sunrpc_syms.c
@@ -129,6 +129,10 @@ EXPORT_SYMBOL(xdr_encode_netobj);
 EXPORT_SYMBOL(xdr_encode_pages);
 EXPORT_SYMBOL(xdr_inline_pages);
 EXPORT_SYMBOL(xdr_shift_buf);
+EXPORT_SYMBOL(xdr_encode_word);
+EXPORT_SYMBOL(xdr_decode_word);
+EXPORT_SYMBOL(xdr_encode_array2);
+EXPORT_SYMBOL(xdr_decode_array2);
 EXPORT_SYMBOL(xdr_buf_from_iov);
 EXPORT_SYMBOL(xdr_buf_subsegment);
 EXPORT_SYMBOL(xdr_buf_read_netobj);
diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c
index b3ac3f72bf9..8a4d9c106af 100644
--- a/net/sunrpc/xdr.c
+++ b/net/sunrpc/xdr.c
@@ -887,8 +887,34 @@ out:
 	return status;
 }
 
-static int
-read_u32_from_xdr_buf(struct xdr_buf *buf, int base, u32 *obj)
+/* obj is assumed to point to allocated memory of size at least len: */
+int
+write_bytes_to_xdr_buf(struct xdr_buf *buf, int base, void *obj, int len)
+{
+	struct xdr_buf subbuf;
+	int this_len;
+	int status;
+
+	status = xdr_buf_subsegment(buf, &subbuf, base, len);
+	if (status)
+		goto out;
+	this_len = min(len, (int)subbuf.head[0].iov_len);
+	memcpy(subbuf.head[0].iov_base, obj, this_len);
+	len -= this_len;
+	obj += this_len;
+	this_len = min(len, (int)subbuf.page_len);
+	if (this_len)
+		_copy_to_pages(subbuf.pages, subbuf.page_base, obj, this_len);
+	len -= this_len;
+	obj += this_len;
+	this_len = min(len, (int)subbuf.tail[0].iov_len);
+	memcpy(subbuf.tail[0].iov_base, obj, this_len);
+out:
+	return status;
+}
+
+int
+xdr_decode_word(struct xdr_buf *buf, int base, u32 *obj)
 {
 	u32	raw;
 	int	status;
@@ -900,6 +926,14 @@ read_u32_from_xdr_buf(struct xdr_buf *buf, int base, u32 *obj)
 	return 0;
 }
 
+int
+xdr_encode_word(struct xdr_buf *buf, int base, u32 obj)
+{
+	u32	raw = htonl(obj);
+
+	return write_bytes_to_xdr_buf(buf, base, &raw, sizeof(obj));
+}
+
 /* If the netobj starting offset bytes from the start of xdr_buf is contained
  * entirely in the head or the tail, set object to point to it; otherwise
  * try to find space for it at the end of the tail, copy it there, and
@@ -910,7 +944,7 @@ xdr_buf_read_netobj(struct xdr_buf *buf, struct xdr_netobj *obj, int offset)
 	u32	tail_offset = buf->head[0].iov_len + buf->page_len;
 	u32	obj_end_offset;
 
-	if (read_u32_from_xdr_buf(buf, offset, &obj->len))
+	if (xdr_decode_word(buf, offset, &obj->len))
 		goto out;
 	obj_end_offset = offset + 4 + obj->len;
 
@@ -943,3 +977,219 @@ xdr_buf_read_netobj(struct xdr_buf *buf, struct xdr_netobj *obj, int offset)
 out:
 	return -1;
 }
+
+/* Returns 0 on success, or else a negative error code. */
+static int
+xdr_xcode_array2(struct xdr_buf *buf, unsigned int base,
+		 struct xdr_array2_desc *desc, int encode)
+{
+	char *elem = NULL, *c;
+	unsigned int copied = 0, todo, avail_here;
+	struct page **ppages = NULL;
+	int err;
+
+	if (encode) {
+		if (xdr_encode_word(buf, base, desc->array_len) != 0)
+			return -EINVAL;
+	} else {
+		if (xdr_decode_word(buf, base, &desc->array_len) != 0 ||
+		    (unsigned long) base + 4 + desc->array_len *
+				    desc->elem_size > buf->len)
+			return -EINVAL;
+	}
+	base += 4;
+
+	if (!desc->xcode)
+		return 0;
+
+	todo = desc->array_len * desc->elem_size;
+
+	/* process head */
+	if (todo && base < buf->head->iov_len) {
+		c = buf->head->iov_base + base;
+		avail_here = min_t(unsigned int, todo,
+				   buf->head->iov_len - base);
+		todo -= avail_here;
+
+		while (avail_here >= desc->elem_size) {
+			err = desc->xcode(desc, c);
+			if (err)
+				goto out;
+			c += desc->elem_size;
+			avail_here -= desc->elem_size;
+		}
+		if (avail_here) {
+			if (!elem) {
+				elem = kmalloc(desc->elem_size, GFP_KERNEL);
+				err = -ENOMEM;
+				if (!elem)
+					goto out;
+			}
+			if (encode) {
+				err = desc->xcode(desc, elem);
+				if (err)
+					goto out;
+				memcpy(c, elem, avail_here);
+			} else
+				memcpy(elem, c, avail_here);
+			copied = avail_here;
+		}
+		base = buf->head->iov_len;  /* align to start of pages */
+	}
+
+	/* process pages array */
+	base -= buf->head->iov_len;
+	if (todo && base < buf->page_len) {
+		unsigned int avail_page;
+
+		avail_here = min(todo, buf->page_len - base);
+		todo -= avail_here;
+
+		base += buf->page_base;
+		ppages = buf->pages + (base >> PAGE_CACHE_SHIFT);
+		base &= ~PAGE_CACHE_MASK;
+		avail_page = min_t(unsigned int, PAGE_CACHE_SIZE - base,
+					avail_here);
+		c = kmap(*ppages) + base;
+
+		while (avail_here) {
+			avail_here -= avail_page;
+			if (copied || avail_page < desc->elem_size) {
+				unsigned int l = min(avail_page,
+					desc->elem_size - copied);
+				if (!elem) {
+					elem = kmalloc(desc->elem_size,
+						       GFP_KERNEL);
+					err = -ENOMEM;
+					if (!elem)
+						goto out;
+				}
+				if (encode) {
+					if (!copied) {
+						err = desc->xcode(desc, elem);
+						if (err)
+							goto out;
+					}
+					memcpy(c, elem + copied, l);
+					copied += l;
+					if (copied == desc->elem_size)
+						copied = 0;
+				} else {
+					memcpy(elem + copied, c, l);
+					copied += l;
+					if (copied == desc->elem_size) {
+						err = desc->xcode(desc, elem);
+						if (err)
+							goto out;
+						copied = 0;
+					}
+				}
+				avail_page -= l;
+				c += l;
+			}
+			while (avail_page >= desc->elem_size) {
+				err = desc->xcode(desc, c);
+				if (err)
+					goto out;
+				c += desc->elem_size;
+				avail_page -= desc->elem_size;
+			}
+			if (avail_page) {
+				unsigned int l = min(avail_page,
+					    desc->elem_size - copied);
+				if (!elem) {
+					elem = kmalloc(desc->elem_size,
+						       GFP_KERNEL);
+					err = -ENOMEM;
+					if (!elem)
+						goto out;
+				}
+				if (encode) {
+					if (!copied) {
+						err = desc->xcode(desc, elem);
+						if (err)
+							goto out;
+					}
+					memcpy(c, elem + copied, l);
+					copied += l;
+					if (copied == desc->elem_size)
+						copied = 0;
+				} else {
+					memcpy(elem + copied, c, l);
+					copied += l;
+					if (copied == desc->elem_size) {
+						err = desc->xcode(desc, elem);
+						if (err)
+							goto out;
+						copied = 0;
+					}
+				}
+			}
+			if (avail_here) {
+				kunmap(*ppages);
+				ppages++;
+				c = kmap(*ppages);
+			}
+
+			avail_page = min(avail_here,
+				 (unsigned int) PAGE_CACHE_SIZE);
+		}
+		base = buf->page_len;  /* align to start of tail */
+	}
+
+	/* process tail */
+	base -= buf->page_len;
+	if (todo) {
+		c = buf->tail->iov_base + base;
+		if (copied) {
+			unsigned int l = desc->elem_size - copied;
+
+			if (encode)
+				memcpy(c, elem + copied, l);
+			else {
+				memcpy(elem + copied, c, l);
+				err = desc->xcode(desc, elem);
+				if (err)
+					goto out;
+			}
+			todo -= l;
+			c += l;
+		}
+		while (todo) {
+			err = desc->xcode(desc, c);
+			if (err)
+				goto out;
+			c += desc->elem_size;
+			todo -= desc->elem_size;
+		}
+	}
+	err = 0;
+
+out:
+	if (elem)
+		kfree(elem);
+	if (ppages)
+		kunmap(*ppages);
+	return err;
+}
+
+int
+xdr_decode_array2(struct xdr_buf *buf, unsigned int base,
+		  struct xdr_array2_desc *desc)
+{
+	if (base >= buf->len)
+		return -EINVAL;
+
+	return xdr_xcode_array2(buf, base, desc, 0);
+}
+
+int
+xdr_encode_array2(struct xdr_buf *buf, unsigned int base,
+		  struct xdr_array2_desc *desc)
+{
+	if ((unsigned long) base + 4 + desc->array_len * desc->elem_size >
+	    buf->head->iov_len + buf->page_len + buf->tail->iov_len)
+		return -EINVAL;
+
+	return xdr_xcode_array2(buf, base, desc, 1);
+}
-- 
cgit v1.2.3-70-g09d2


From 9ba02638e4be28dd4ff724202a640264427c62d1 Mon Sep 17 00:00:00 2001
From: Andreas Gruenbacher <agruen@suse.de>
Date: Wed, 22 Jun 2005 17:16:24 +0000
Subject: [PATCH] RPC: Allow the sunrpc server to multiplex serveral programs
 on a single port

 The NFS and NFSACL programs run on the same RPC transport.  This patch adds
 support for this by converting svc_program into a chained list of programs
 (server-side).

 Signed-off-by: Andreas Gruenbacher <agruen@suse.de>
 Signed-off-by: Olaf Kirch <okir@suse.de>
 Signed-off-by: Andrew Morton <akpm@osdl.org>
 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/svc.h |  3 ++-
 net/sunrpc/svc.c           | 35 ++++++++++++++++++-----------------
 2 files changed, 20 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h
index 37003970cf2..facb94488bb 100644
--- a/include/linux/sunrpc/svc.h
+++ b/include/linux/sunrpc/svc.h
@@ -240,9 +240,10 @@ struct svc_deferred_req {
 };
 
 /*
- * RPC program
+ * List of RPC programs on the same transport endpoint
  */
 struct svc_program {
+	struct svc_program *	pg_next;	/* other programs (same xprt) */
 	u32			pg_prog;	/* program number */
 	unsigned int		pg_lovers;	/* lowest version */
 	unsigned int		pg_hivers;	/* lowest version */
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index a02d424a740..e9bd91265f7 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -35,20 +35,24 @@ svc_create(struct svc_program *prog, unsigned int bufsize)
 	if (!(serv = (struct svc_serv *) kmalloc(sizeof(*serv), GFP_KERNEL)))
 		return NULL;
 	memset(serv, 0, sizeof(*serv));
+	serv->sv_name      = prog->pg_name;
 	serv->sv_program   = prog;
 	serv->sv_nrthreads = 1;
 	serv->sv_stats     = prog->pg_stats;
 	serv->sv_bufsz	   = bufsize? bufsize : 4096;
-	prog->pg_lovers = prog->pg_nvers-1;
 	xdrsize = 0;
-	for (vers=0; vers<prog->pg_nvers ; vers++)
-		if (prog->pg_vers[vers]) {
-			prog->pg_hivers = vers;
-			if (prog->pg_lovers > vers)
-				prog->pg_lovers = vers;
-			if (prog->pg_vers[vers]->vs_xdrsize > xdrsize)
-				xdrsize = prog->pg_vers[vers]->vs_xdrsize;
-		}
+	while (prog) {
+		prog->pg_lovers = prog->pg_nvers-1;
+		for (vers=0; vers<prog->pg_nvers ; vers++)
+			if (prog->pg_vers[vers]) {
+				prog->pg_hivers = vers;
+				if (prog->pg_lovers > vers)
+					prog->pg_lovers = vers;
+				if (prog->pg_vers[vers]->vs_xdrsize > xdrsize)
+					xdrsize = prog->pg_vers[vers]->vs_xdrsize;
+			}
+		prog = prog->pg_next;
+	}
 	serv->sv_xdrsize   = xdrsize;
 	INIT_LIST_HEAD(&serv->sv_threads);
 	INIT_LIST_HEAD(&serv->sv_sockets);
@@ -56,8 +60,6 @@ svc_create(struct svc_program *prog, unsigned int bufsize)
 	INIT_LIST_HEAD(&serv->sv_permsocks);
 	spin_lock_init(&serv->sv_lock);
 
-	serv->sv_name      = prog->pg_name;
-
 	/* Remove any stale portmap registrations */
 	svc_register(serv, 0, 0);
 
@@ -339,7 +341,10 @@ svc_process(struct svc_serv *serv, struct svc_rqst *rqstp)
 		goto sendit;
 	}
 		
-	if (prog != progp->pg_prog)
+	for (progp = serv->sv_program; progp; progp = progp->pg_next)
+		if (prog == progp->pg_prog)
+			break;
+	if (progp == NULL)
 		goto err_bad_prog;
 
 	if (vers >= progp->pg_nvers ||
@@ -452,11 +457,7 @@ err_bad_auth:
 	goto sendit;
 
 err_bad_prog:
-#ifdef RPC_PARANOIA
-	if (prog != 100227 || progp->pg_prog != 100003)
-		printk("svc: unknown program %d (me %d)\n", prog, progp->pg_prog);
-	/* else it is just a Solaris client seeing if ACLs are supported */
-#endif
+	dprintk("svc: unknown program %d\n", prog);
 	serv->sv_stats->rpcbadfmt++;
 	svc_putu32(resv, rpc_prog_unavail);
 	goto sendit;
-- 
cgit v1.2.3-70-g09d2


From a257cdd0e2179630d3201c32ba14d7fcb3c3a055 Mon Sep 17 00:00:00 2001
From: Andreas Gruenbacher <agruen@suse.de>
Date: Wed, 22 Jun 2005 17:16:26 +0000
Subject: [PATCH] NFSD: Add server support for NFSv3 ACLs.

 This adds functions for encoding and decoding POSIX ACLs for the NFSACL
 protocol extension, and the GETACL and SETACL RPCs.  The implementation is
 compatible with NFSACL in Solaris.

 Signed-off-by: Andreas Gruenbacher <agruen@suse.de>
 Acked-by: Olaf Kirch <okir@suse.de>
 Signed-off-by: Andrew Morton <akpm@osdl.org>
 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/Kconfig                 |  24 ++++
 fs/Makefile                |   1 +
 fs/nfs_common/Makefile     |   7 +
 fs/nfs_common/nfsacl.c     | 257 ++++++++++++++++++++++++++++++++++
 fs/nfsd/Makefile           |   2 +
 fs/nfsd/nfs2acl.c          | 336 +++++++++++++++++++++++++++++++++++++++++++++
 fs/nfsd/nfs3acl.c          | 267 +++++++++++++++++++++++++++++++++++
 fs/nfsd/nfs3xdr.c          |  13 ++
 fs/nfsd/nfssvc.c           |  27 ++++
 fs/nfsd/nfsxdr.c           |  11 ++
 fs/nfsd/vfs.c              | 107 ++++++++++++++-
 include/linux/nfsacl.h     |  58 ++++++++
 include/linux/nfsd/nfsd.h  |  16 +++
 include/linux/nfsd/xdr.h   |   4 +
 include/linux/nfsd/xdr3.h  |  26 ++++
 include/linux/sunrpc/svc.h |  11 ++
 16 files changed, 1166 insertions(+), 1 deletion(-)
 create mode 100644 fs/nfs_common/Makefile
 create mode 100644 fs/nfs_common/nfsacl.c
 create mode 100644 fs/nfsd/nfs2acl.c
 create mode 100644 fs/nfsd/nfs3acl.c
 create mode 100644 include/linux/nfsacl.h

(limited to 'include')

diff --git a/fs/Kconfig b/fs/Kconfig
index 178e27494b7..d44b04d9b0a 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -1353,6 +1353,7 @@ config NFSD
 	select LOCKD
 	select SUNRPC
 	select EXPORTFS
+	select NFS_ACL_SUPPORT if NFSD_V3_ACL || NFSD_V2_ACL
 	help
 	  If you want your Linux box to act as an NFS *server*, so that other
 	  computers on your local network which support NFS can access certain
@@ -1376,6 +1377,10 @@ config NFSD
 	  To compile the NFS server support as a module, choose M here: the
 	  module will be called nfsd.  If unsure, say N.
 
+config NFSD_V2_ACL
+	bool
+	depends on NFSD
+
 config NFSD_V3
 	bool "Provide NFSv3 server support"
 	depends on NFSD
@@ -1383,6 +1388,16 @@ config NFSD_V3
 	  If you would like to include the NFSv3 server as well as the NFSv2
 	  server, say Y here.  If unsure, say Y.
 
+config NFSD_V3_ACL
+	bool "Provide server support for the NFSv3 ACL protocol extension"
+	depends on NFSD_V3
+	select NFSD_V2_ACL
+	help
+	  Implement the NFSv3 ACL protocol extension for manipulating POSIX
+	  Access Control Lists on exported file systems. NFS clients should
+	  be compiled with the NFSv3 ACL protocol extension; see the
+	  CONFIG_NFS_V3_ACL option.  If unsure, say N.
+
 config NFSD_V4
 	bool "Provide NFSv4 server support (EXPERIMENTAL)"
 	depends on NFSD_V3 && EXPERIMENTAL
@@ -1427,6 +1442,15 @@ config LOCKD_V4
 config EXPORTFS
 	tristate
 
+config NFS_ACL_SUPPORT
+	tristate
+	select FS_POSIX_ACL
+
+config NFS_COMMON
+	bool
+	depends on NFSD || NFS_FS
+	default y
+
 config SUNRPC
 	tristate
 
diff --git a/fs/Makefile b/fs/Makefile
index 443f2bc56cc..fc92e59e9fa 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -31,6 +31,7 @@ obj-$(CONFIG_BINFMT_FLAT)	+= binfmt_flat.o
 
 obj-$(CONFIG_FS_MBCACHE)	+= mbcache.o
 obj-$(CONFIG_FS_POSIX_ACL)	+= posix_acl.o xattr_acl.o
+obj-$(CONFIG_NFS_COMMON)	+= nfs_common/
 
 obj-$(CONFIG_QUOTA)		+= dquot.o
 obj-$(CONFIG_QFMT_V1)		+= quota_v1.o
diff --git a/fs/nfs_common/Makefile b/fs/nfs_common/Makefile
new file mode 100644
index 00000000000..f689ed82af3
--- /dev/null
+++ b/fs/nfs_common/Makefile
@@ -0,0 +1,7 @@
+#
+# Makefile for Linux filesystem routines that are shared by client and server.
+#
+
+obj-$(CONFIG_NFS_ACL_SUPPORT) += nfs_acl.o
+
+nfs_acl-objs := nfsacl.o
diff --git a/fs/nfs_common/nfsacl.c b/fs/nfs_common/nfsacl.c
new file mode 100644
index 00000000000..18c58c32e32
--- /dev/null
+++ b/fs/nfs_common/nfsacl.c
@@ -0,0 +1,257 @@
+/*
+ * fs/nfs_common/nfsacl.c
+ *
+ *  Copyright (C) 2002-2003 Andreas Gruenbacher <agruen@suse.de>
+ */
+
+/*
+ * The Solaris nfsacl protocol represents some ACLs slightly differently
+ * than POSIX 1003.1e draft 17 does (and we do):
+ *
+ *  - Minimal ACLs always have an ACL_MASK entry, so they have
+ *    four instead of three entries.
+ *  - The ACL_MASK entry in such minimal ACLs always has the same
+ *    permissions as the ACL_GROUP_OBJ entry. (In extended ACLs
+ *    the ACL_MASK and ACL_GROUP_OBJ entries may differ.)
+ *  - The identifier fields of the ACL_USER_OBJ and ACL_GROUP_OBJ
+ *    entries contain the identifiers of the owner and owning group.
+ *    (In POSIX ACLs we always set them to ACL_UNDEFINED_ID).
+ *  - ACL entries in the kernel are kept sorted in ascending order
+ *    of (e_tag, e_id). Solaris ACLs are unsorted.
+ */
+
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/sunrpc/xdr.h>
+#include <linux/nfsacl.h>
+#include <linux/nfs3.h>
+#include <linux/sort.h>
+
+MODULE_LICENSE("GPL");
+
+EXPORT_SYMBOL(nfsacl_encode);
+EXPORT_SYMBOL(nfsacl_decode);
+
+struct nfsacl_encode_desc {
+	struct xdr_array2_desc desc;
+	unsigned int count;
+	struct posix_acl *acl;
+	int typeflag;
+	uid_t uid;
+	gid_t gid;
+};
+
+static int
+xdr_nfsace_encode(struct xdr_array2_desc *desc, void *elem)
+{
+	struct nfsacl_encode_desc *nfsacl_desc =
+		(struct nfsacl_encode_desc *) desc;
+	u32 *p = (u32 *) elem;
+
+	if (nfsacl_desc->count < nfsacl_desc->acl->a_count) {
+		struct posix_acl_entry *entry =
+			&nfsacl_desc->acl->a_entries[nfsacl_desc->count++];
+
+		*p++ = htonl(entry->e_tag | nfsacl_desc->typeflag);
+		switch(entry->e_tag) {
+			case ACL_USER_OBJ:
+				*p++ = htonl(nfsacl_desc->uid);
+				break;
+			case ACL_GROUP_OBJ:
+				*p++ = htonl(nfsacl_desc->gid);
+				break;
+			case ACL_USER:
+			case ACL_GROUP:
+				*p++ = htonl(entry->e_id);
+				break;
+			default:  /* Solaris depends on that! */
+				*p++ = 0;
+				break;
+		}
+		*p++ = htonl(entry->e_perm & S_IRWXO);
+	} else {
+		const struct posix_acl_entry *pa, *pe;
+		int group_obj_perm = ACL_READ|ACL_WRITE|ACL_EXECUTE;
+
+		FOREACH_ACL_ENTRY(pa, nfsacl_desc->acl, pe) {
+			if (pa->e_tag == ACL_GROUP_OBJ) {
+				group_obj_perm = pa->e_perm & S_IRWXO;
+				break;
+			}
+		}
+		/* fake up ACL_MASK entry */
+		*p++ = htonl(ACL_MASK | nfsacl_desc->typeflag);
+		*p++ = htonl(0);
+		*p++ = htonl(group_obj_perm);
+	}
+
+	return 0;
+}
+
+unsigned int
+nfsacl_encode(struct xdr_buf *buf, unsigned int base, struct inode *inode,
+	      struct posix_acl *acl, int encode_entries, int typeflag)
+{
+	int entries = (acl && acl->a_count) ? max_t(int, acl->a_count, 4) : 0;
+	struct nfsacl_encode_desc nfsacl_desc = {
+		.desc = {
+			.elem_size = 12,
+			.array_len = encode_entries ? entries : 0,
+			.xcode = xdr_nfsace_encode,
+		},
+		.acl = acl,
+		.typeflag = typeflag,
+		.uid = inode->i_uid,
+		.gid = inode->i_gid,
+	};
+	int err;
+
+	if (entries > NFS_ACL_MAX_ENTRIES ||
+	    xdr_encode_word(buf, base, entries))
+		return -EINVAL;
+	err = xdr_encode_array2(buf, base + 4, &nfsacl_desc.desc);
+	if (!err)
+		err = 8 + nfsacl_desc.desc.elem_size *
+			  nfsacl_desc.desc.array_len;
+	return err;
+}
+
+struct nfsacl_decode_desc {
+	struct xdr_array2_desc desc;
+	unsigned int count;
+	struct posix_acl *acl;
+};
+
+static int
+xdr_nfsace_decode(struct xdr_array2_desc *desc, void *elem)
+{
+	struct nfsacl_decode_desc *nfsacl_desc =
+		(struct nfsacl_decode_desc *) desc;
+	u32 *p = (u32 *) elem;
+	struct posix_acl_entry *entry;
+
+	if (!nfsacl_desc->acl) {
+		if (desc->array_len > NFS_ACL_MAX_ENTRIES)
+			return -EINVAL;
+		nfsacl_desc->acl = posix_acl_alloc(desc->array_len, GFP_KERNEL);
+		if (!nfsacl_desc->acl)
+			return -ENOMEM;
+		nfsacl_desc->count = 0;
+	}
+
+	entry = &nfsacl_desc->acl->a_entries[nfsacl_desc->count++];
+	entry->e_tag = ntohl(*p++) & ~NFS_ACL_DEFAULT;
+	entry->e_id = ntohl(*p++);
+	entry->e_perm = ntohl(*p++);
+
+	switch(entry->e_tag) {
+		case ACL_USER_OBJ:
+		case ACL_USER:
+		case ACL_GROUP_OBJ:
+		case ACL_GROUP:
+		case ACL_OTHER:
+			if (entry->e_perm & ~S_IRWXO)
+				return -EINVAL;
+			break;
+		case ACL_MASK:
+			/* Solaris sometimes sets additonal bits in the mask */
+			entry->e_perm &= S_IRWXO;
+			break;
+		default:
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int
+cmp_acl_entry(const void *x, const void *y)
+{
+	const struct posix_acl_entry *a = x, *b = y;
+
+	if (a->e_tag != b->e_tag)
+		return a->e_tag - b->e_tag;
+	else if (a->e_id > b->e_id)
+		return 1;
+	else if (a->e_id < b->e_id)
+		return -1;
+	else
+		return 0;
+}
+
+/*
+ * Convert from a Solaris ACL to a POSIX 1003.1e draft 17 ACL.
+ */
+static int
+posix_acl_from_nfsacl(struct posix_acl *acl)
+{
+	struct posix_acl_entry *pa, *pe,
+	       *group_obj = NULL, *mask = NULL;
+
+	if (!acl)
+		return 0;
+
+	sort(acl->a_entries, acl->a_count, sizeof(struct posix_acl_entry),
+	     cmp_acl_entry, NULL);
+
+	/* Clear undefined identifier fields and find the ACL_GROUP_OBJ
+	   and ACL_MASK entries. */
+	FOREACH_ACL_ENTRY(pa, acl, pe) {
+		switch(pa->e_tag) {
+			case ACL_USER_OBJ:
+				pa->e_id = ACL_UNDEFINED_ID;
+				break;
+			case ACL_GROUP_OBJ:
+				pa->e_id = ACL_UNDEFINED_ID;
+				group_obj = pa;
+				break;
+			case ACL_MASK:
+				mask = pa;
+				/* fall through */
+			case ACL_OTHER:
+				pa->e_id = ACL_UNDEFINED_ID;
+				break;
+		}
+	}
+	if (acl->a_count == 4 && group_obj && mask &&
+	    mask->e_perm == group_obj->e_perm) {
+		/* remove bogus ACL_MASK entry */
+		memmove(mask, mask+1, (3 - (mask - acl->a_entries)) *
+				      sizeof(struct posix_acl_entry));
+		acl->a_count = 3;
+	}
+	return 0;
+}
+
+unsigned int
+nfsacl_decode(struct xdr_buf *buf, unsigned int base, unsigned int *aclcnt,
+	      struct posix_acl **pacl)
+{
+	struct nfsacl_decode_desc nfsacl_desc = {
+		.desc = {
+			.elem_size = 12,
+			.xcode = pacl ? xdr_nfsace_decode : NULL,
+		},
+	};
+	u32 entries;
+	int err;
+
+	if (xdr_decode_word(buf, base, &entries) ||
+	    entries > NFS_ACL_MAX_ENTRIES)
+		return -EINVAL;
+	err = xdr_decode_array2(buf, base + 4, &nfsacl_desc.desc);
+	if (err)
+		return err;
+	if (pacl) {
+		if (entries != nfsacl_desc.desc.array_len ||
+		    posix_acl_from_nfsacl(nfsacl_desc.acl) != 0) {
+			posix_acl_release(nfsacl_desc.acl);
+			return -EINVAL;
+		}
+		*pacl = nfsacl_desc.acl;
+	}
+	if (aclcnt)
+		*aclcnt = entries;
+	return 8 + nfsacl_desc.desc.elem_size *
+		   nfsacl_desc.desc.array_len;
+}
diff --git a/fs/nfsd/Makefile b/fs/nfsd/Makefile
index b8680a247f8..9f043f44c92 100644
--- a/fs/nfsd/Makefile
+++ b/fs/nfsd/Makefile
@@ -6,7 +6,9 @@ obj-$(CONFIG_NFSD)	+= nfsd.o
 
 nfsd-y 			:= nfssvc.o nfsctl.o nfsproc.o nfsfh.o vfs.o \
 			   export.o auth.o lockd.o nfscache.o nfsxdr.o stats.o
+nfsd-$(CONFIG_NFSD_V2_ACL) += nfs2acl.o
 nfsd-$(CONFIG_NFSD_V3)	+= nfs3proc.o nfs3xdr.o
+nfsd-$(CONFIG_NFSD_V3_ACL) += nfs3acl.o
 nfsd-$(CONFIG_NFSD_V4)	+= nfs4proc.o nfs4xdr.o nfs4state.o nfs4idmap.o \
 			   nfs4acl.o nfs4callback.o
 nfsd-objs		:= $(nfsd-y)
diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c
new file mode 100644
index 00000000000..7cbf0682b2f
--- /dev/null
+++ b/fs/nfsd/nfs2acl.c
@@ -0,0 +1,336 @@
+/*
+ * linux/fs/nfsd/nfsacl.c
+ *
+ * Process version 2 NFSACL requests.
+ *
+ * Copyright (C) 2002-2003 Andreas Gruenbacher <agruen@suse.de>
+ */
+
+#include <linux/sunrpc/svc.h>
+#include <linux/nfs.h>
+#include <linux/nfsd/nfsd.h>
+#include <linux/nfsd/cache.h>
+#include <linux/nfsd/xdr.h>
+#include <linux/nfsd/xdr3.h>
+#include <linux/posix_acl.h>
+#include <linux/nfsacl.h>
+
+#define NFSDDBG_FACILITY		NFSDDBG_PROC
+#define RETURN_STATUS(st)	{ resp->status = (st); return (st); }
+
+/*
+ * NULL call.
+ */
+static int
+nfsacld_proc_null(struct svc_rqst *rqstp, void *argp, void *resp)
+{
+	return nfs_ok;
+}
+
+/*
+ * Get the Access and/or Default ACL of a file.
+ */
+static int nfsacld_proc_getacl(struct svc_rqst * rqstp,
+		struct nfsd3_getaclargs *argp, struct nfsd3_getaclres *resp)
+{
+	svc_fh *fh;
+	struct posix_acl *acl;
+	int nfserr = 0;
+
+	dprintk("nfsd: GETACL(2acl)   %s\n", SVCFH_fmt(&argp->fh));
+
+	fh = fh_copy(&resp->fh, &argp->fh);
+	if ((nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP)))
+		RETURN_STATUS(nfserr_inval);
+
+	if (argp->mask & ~(NFS_ACL|NFS_ACLCNT|NFS_DFACL|NFS_DFACLCNT))
+		RETURN_STATUS(nfserr_inval);
+	resp->mask = argp->mask;
+
+	if (resp->mask & (NFS_ACL|NFS_ACLCNT)) {
+		acl = nfsd_get_posix_acl(fh, ACL_TYPE_ACCESS);
+		if (IS_ERR(acl)) {
+			int err = PTR_ERR(acl);
+
+			if (err == -ENODATA || err == -EOPNOTSUPP)
+				acl = NULL;
+			else {
+				nfserr = nfserrno(err);
+				goto fail;
+			}
+		}
+		if (acl == NULL) {
+			/* Solaris returns the inode's minimum ACL. */
+
+			struct inode *inode = fh->fh_dentry->d_inode;
+			acl = posix_acl_from_mode(inode->i_mode, GFP_KERNEL);
+		}
+		resp->acl_access = acl;
+	}
+	if (resp->mask & (NFS_DFACL|NFS_DFACLCNT)) {
+		/* Check how Solaris handles requests for the Default ACL
+		   of a non-directory! */
+
+		acl = nfsd_get_posix_acl(fh, ACL_TYPE_DEFAULT);
+		if (IS_ERR(acl)) {
+			int err = PTR_ERR(acl);
+
+			if (err == -ENODATA || err == -EOPNOTSUPP)
+				acl = NULL;
+			else {
+				nfserr = nfserrno(err);
+				goto fail;
+			}
+		}
+		resp->acl_default = acl;
+	}
+
+	/* resp->acl_{access,default} are released in nfssvc_release_getacl. */
+	RETURN_STATUS(0);
+
+fail:
+	posix_acl_release(resp->acl_access);
+	posix_acl_release(resp->acl_default);
+	RETURN_STATUS(nfserr);
+}
+
+/*
+ * Set the Access and/or Default ACL of a file.
+ */
+static int nfsacld_proc_setacl(struct svc_rqst * rqstp,
+		struct nfsd3_setaclargs *argp,
+		struct nfsd_attrstat *resp)
+{
+	svc_fh *fh;
+	int nfserr = 0;
+
+	dprintk("nfsd: SETACL(2acl)   %s\n", SVCFH_fmt(&argp->fh));
+
+	fh = fh_copy(&resp->fh, &argp->fh);
+	nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP);
+
+	if (!nfserr) {
+		nfserr = nfserrno( nfsd_set_posix_acl(
+			fh, ACL_TYPE_ACCESS, argp->acl_access) );
+	}
+	if (!nfserr) {
+		nfserr = nfserrno( nfsd_set_posix_acl(
+			fh, ACL_TYPE_DEFAULT, argp->acl_default) );
+	}
+
+	/* argp->acl_{access,default} may have been allocated in
+	   nfssvc_decode_setaclargs. */
+	posix_acl_release(argp->acl_access);
+	posix_acl_release(argp->acl_default);
+	return nfserr;
+}
+
+/*
+ * Check file attributes
+ */
+static int nfsacld_proc_getattr(struct svc_rqst * rqstp,
+		struct nfsd_fhandle *argp, struct nfsd_attrstat *resp)
+{
+	dprintk("nfsd: GETATTR  %s\n", SVCFH_fmt(&argp->fh));
+
+	fh_copy(&resp->fh, &argp->fh);
+	return fh_verify(rqstp, &resp->fh, 0, MAY_NOP);
+}
+
+/*
+ * Check file access
+ */
+static int nfsacld_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp,
+		struct nfsd3_accessres *resp)
+{
+	int nfserr;
+
+	dprintk("nfsd: ACCESS(2acl)   %s 0x%x\n",
+			SVCFH_fmt(&argp->fh),
+			argp->access);
+
+	fh_copy(&resp->fh, &argp->fh);
+	resp->access = argp->access;
+	nfserr = nfsd_access(rqstp, &resp->fh, &resp->access, NULL);
+	return nfserr;
+}
+
+/*
+ * XDR decode functions
+ */
+static int nfsaclsvc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p,
+		struct nfsd3_getaclargs *argp)
+{
+	if (!(p = nfs2svc_decode_fh(p, &argp->fh)))
+		return 0;
+	argp->mask = ntohl(*p); p++;
+
+	return xdr_argsize_check(rqstp, p);
+}
+
+
+static int nfsaclsvc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p,
+		struct nfsd3_setaclargs *argp)
+{
+	struct kvec *head = rqstp->rq_arg.head;
+	unsigned int base;
+	int n;
+
+	if (!(p = nfs2svc_decode_fh(p, &argp->fh)))
+		return 0;
+	argp->mask = ntohl(*p++);
+	if (argp->mask & ~(NFS_ACL|NFS_ACLCNT|NFS_DFACL|NFS_DFACLCNT) ||
+	    !xdr_argsize_check(rqstp, p))
+		return 0;
+
+	base = (char *)p - (char *)head->iov_base;
+	n = nfsacl_decode(&rqstp->rq_arg, base, NULL,
+			  (argp->mask & NFS_ACL) ?
+			  &argp->acl_access : NULL);
+	if (n > 0)
+		n = nfsacl_decode(&rqstp->rq_arg, base + n, NULL,
+				  (argp->mask & NFS_DFACL) ?
+				  &argp->acl_default : NULL);
+	return (n > 0);
+}
+
+static int nfsaclsvc_decode_fhandleargs(struct svc_rqst *rqstp, u32 *p,
+		struct nfsd_fhandle *argp)
+{
+	if (!(p = nfs2svc_decode_fh(p, &argp->fh)))
+		return 0;
+	return xdr_argsize_check(rqstp, p);
+}
+
+static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, u32 *p,
+		struct nfsd3_accessargs *argp)
+{
+	if (!(p = nfs2svc_decode_fh(p, &argp->fh)))
+		return 0;
+	argp->access = ntohl(*p++);
+
+	return xdr_argsize_check(rqstp, p);
+}
+
+/*
+ * XDR encode functions
+ */
+
+/* GETACL */
+static int nfsaclsvc_encode_getaclres(struct svc_rqst *rqstp, u32 *p,
+		struct nfsd3_getaclres *resp)
+{
+	struct dentry *dentry = resp->fh.fh_dentry;
+	struct inode *inode = dentry->d_inode;
+	int w = nfsacl_size(
+		(resp->mask & NFS_ACL)   ? resp->acl_access  : NULL,
+		(resp->mask & NFS_DFACL) ? resp->acl_default : NULL);
+	struct kvec *head = rqstp->rq_res.head;
+	unsigned int base;
+	int n;
+
+	if (dentry == NULL || dentry->d_inode == NULL)
+		return 0;
+	inode = dentry->d_inode;
+
+	p = nfs2svc_encode_fattr(rqstp, p, &resp->fh);
+	*p++ = htonl(resp->mask);
+	if (!xdr_ressize_check(rqstp, p))
+		return 0;
+	base = (char *)p - (char *)head->iov_base;
+
+	rqstp->rq_res.page_len = w;
+	while (w > 0) {
+		if (!svc_take_res_page(rqstp))
+			return 0;
+		w -= PAGE_SIZE;
+	}
+
+	n = nfsacl_encode(&rqstp->rq_res, base, inode,
+			  resp->acl_access,
+			  resp->mask & NFS_ACL, 0);
+	if (n > 0)
+		n = nfsacl_encode(&rqstp->rq_res, base + n, inode,
+				  resp->acl_default,
+				  resp->mask & NFS_DFACL,
+				  NFS_ACL_DEFAULT);
+	if (n <= 0)
+		return 0;
+	return 1;
+}
+
+static int nfsaclsvc_encode_attrstatres(struct svc_rqst *rqstp, u32 *p,
+		struct nfsd_attrstat *resp)
+{
+	p = nfs2svc_encode_fattr(rqstp, p, &resp->fh);
+	return xdr_ressize_check(rqstp, p);
+}
+
+/* ACCESS */
+static int nfsaclsvc_encode_accessres(struct svc_rqst *rqstp, u32 *p,
+		struct nfsd3_accessres *resp)
+{
+	p = nfs2svc_encode_fattr(rqstp, p, &resp->fh);
+	*p++ = htonl(resp->access);
+	return xdr_ressize_check(rqstp, p);
+}
+
+/*
+ * XDR release functions
+ */
+static int nfsaclsvc_release_getacl(struct svc_rqst *rqstp, u32 *p,
+		struct nfsd3_getaclres *resp)
+{
+	fh_put(&resp->fh);
+	posix_acl_release(resp->acl_access);
+	posix_acl_release(resp->acl_default);
+	return 1;
+}
+
+static int nfsaclsvc_release_fhandle(struct svc_rqst *rqstp, u32 *p,
+		struct nfsd_fhandle *resp)
+{
+	fh_put(&resp->fh);
+	return 1;
+}
+
+#define nfsaclsvc_decode_voidargs	NULL
+#define nfsaclsvc_encode_voidres	NULL
+#define nfsaclsvc_release_void		NULL
+#define nfsd3_fhandleargs	nfsd_fhandle
+#define nfsd3_attrstatres	nfsd_attrstat
+#define nfsd3_voidres		nfsd3_voidargs
+struct nfsd3_voidargs { int dummy; };
+
+#define PROC(name, argt, rest, relt, cache, respsize)	\
+ { (svc_procfunc) nfsacld_proc_##name,		\
+   (kxdrproc_t) nfsaclsvc_decode_##argt##args,	\
+   (kxdrproc_t) nfsaclsvc_encode_##rest##res,	\
+   (kxdrproc_t) nfsaclsvc_release_##relt,		\
+   sizeof(struct nfsd3_##argt##args),		\
+   sizeof(struct nfsd3_##rest##res),		\
+   0,						\
+   cache,					\
+   respsize,					\
+ }
+
+#define ST 1		/* status*/
+#define AT 21		/* attributes */
+#define pAT (1+AT)	/* post attributes - conditional */
+#define ACL (1+NFS_ACL_MAX_ENTRIES*3)  /* Access Control List */
+
+static struct svc_procedure		nfsd_acl_procedures2[] = {
+  PROC(null,	void,		void,		void,	  RC_NOCACHE, ST),
+  PROC(getacl,	getacl,		getacl,		getacl,	  RC_NOCACHE, ST+1+2*(1+ACL)),
+  PROC(setacl,	setacl,		attrstat,	fhandle,  RC_NOCACHE, ST+AT),
+  PROC(getattr, fhandle,	attrstat,	fhandle,  RC_NOCACHE, ST+AT),
+  PROC(access,	access,		access,		fhandle,  RC_NOCACHE, ST+AT+1),
+};
+
+struct svc_version	nfsd_acl_version2 = {
+		.vs_vers	= 2,
+		.vs_nproc	= 5,
+		.vs_proc	= nfsd_acl_procedures2,
+		.vs_dispatch	= nfsd_dispatch,
+		.vs_xdrsize	= NFS3_SVC_XDRSIZE,
+};
diff --git a/fs/nfsd/nfs3acl.c b/fs/nfsd/nfs3acl.c
new file mode 100644
index 00000000000..64ba40572fe
--- /dev/null
+++ b/fs/nfsd/nfs3acl.c
@@ -0,0 +1,267 @@
+/*
+ * linux/fs/nfsd/nfs3acl.c
+ *
+ * Process version 3 NFSACL requests.
+ *
+ * Copyright (C) 2002-2003 Andreas Gruenbacher <agruen@suse.de>
+ */
+
+#include <linux/sunrpc/svc.h>
+#include <linux/nfs3.h>
+#include <linux/nfsd/nfsd.h>
+#include <linux/nfsd/cache.h>
+#include <linux/nfsd/xdr3.h>
+#include <linux/posix_acl.h>
+#include <linux/nfsacl.h>
+
+#define RETURN_STATUS(st)	{ resp->status = (st); return (st); }
+
+/*
+ * NULL call.
+ */
+static int
+nfsd3_proc_null(struct svc_rqst *rqstp, void *argp, void *resp)
+{
+	return nfs_ok;
+}
+
+/*
+ * Get the Access and/or Default ACL of a file.
+ */
+static int nfsd3_proc_getacl(struct svc_rqst * rqstp,
+		struct nfsd3_getaclargs *argp, struct nfsd3_getaclres *resp)
+{
+	svc_fh *fh;
+	struct posix_acl *acl;
+	int nfserr = 0;
+
+	fh = fh_copy(&resp->fh, &argp->fh);
+	if ((nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP)))
+		RETURN_STATUS(nfserr_inval);
+
+	if (argp->mask & ~(NFS_ACL|NFS_ACLCNT|NFS_DFACL|NFS_DFACLCNT))
+		RETURN_STATUS(nfserr_inval);
+	resp->mask = argp->mask;
+
+	if (resp->mask & (NFS_ACL|NFS_ACLCNT)) {
+		acl = nfsd_get_posix_acl(fh, ACL_TYPE_ACCESS);
+		if (IS_ERR(acl)) {
+			int err = PTR_ERR(acl);
+
+			if (err == -ENODATA || err == -EOPNOTSUPP)
+				acl = NULL;
+			else {
+				nfserr = nfserrno(err);
+				goto fail;
+			}
+		}
+		if (acl == NULL) {
+			/* Solaris returns the inode's minimum ACL. */
+
+			struct inode *inode = fh->fh_dentry->d_inode;
+			acl = posix_acl_from_mode(inode->i_mode, GFP_KERNEL);
+		}
+		resp->acl_access = acl;
+	}
+	if (resp->mask & (NFS_DFACL|NFS_DFACLCNT)) {
+		/* Check how Solaris handles requests for the Default ACL
+		   of a non-directory! */
+
+		acl = nfsd_get_posix_acl(fh, ACL_TYPE_DEFAULT);
+		if (IS_ERR(acl)) {
+			int err = PTR_ERR(acl);
+
+			if (err == -ENODATA || err == -EOPNOTSUPP)
+				acl = NULL;
+			else {
+				nfserr = nfserrno(err);
+				goto fail;
+			}
+		}
+		resp->acl_default = acl;
+	}
+
+	/* resp->acl_{access,default} are released in nfs3svc_release_getacl. */
+	RETURN_STATUS(0);
+
+fail:
+	posix_acl_release(resp->acl_access);
+	posix_acl_release(resp->acl_default);
+	RETURN_STATUS(nfserr);
+}
+
+/*
+ * Set the Access and/or Default ACL of a file.
+ */
+static int nfsd3_proc_setacl(struct svc_rqst * rqstp,
+		struct nfsd3_setaclargs *argp,
+		struct nfsd3_attrstat *resp)
+{
+	svc_fh *fh;
+	int nfserr = 0;
+
+	fh = fh_copy(&resp->fh, &argp->fh);
+	nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP);
+
+	if (!nfserr) {
+		nfserr = nfserrno( nfsd_set_posix_acl(
+			fh, ACL_TYPE_ACCESS, argp->acl_access) );
+	}
+	if (!nfserr) {
+		nfserr = nfserrno( nfsd_set_posix_acl(
+			fh, ACL_TYPE_DEFAULT, argp->acl_default) );
+	}
+
+	/* argp->acl_{access,default} may have been allocated in
+	   nfs3svc_decode_setaclargs. */
+	posix_acl_release(argp->acl_access);
+	posix_acl_release(argp->acl_default);
+	RETURN_STATUS(nfserr);
+}
+
+/*
+ * XDR decode functions
+ */
+static int nfs3svc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p,
+		struct nfsd3_getaclargs *args)
+{
+	if (!(p = nfs3svc_decode_fh(p, &args->fh)))
+		return 0;
+	args->mask = ntohl(*p); p++;
+
+	return xdr_argsize_check(rqstp, p);
+}
+
+
+static int nfs3svc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p,
+		struct nfsd3_setaclargs *args)
+{
+	struct kvec *head = rqstp->rq_arg.head;
+	unsigned int base;
+	int n;
+
+	if (!(p = nfs3svc_decode_fh(p, &args->fh)))
+		return 0;
+	args->mask = ntohl(*p++);
+	if (args->mask & ~(NFS_ACL|NFS_ACLCNT|NFS_DFACL|NFS_DFACLCNT) ||
+	    !xdr_argsize_check(rqstp, p))
+		return 0;
+
+	base = (char *)p - (char *)head->iov_base;
+	n = nfsacl_decode(&rqstp->rq_arg, base, NULL,
+			  (args->mask & NFS_ACL) ?
+			  &args->acl_access : NULL);
+	if (n > 0)
+		n = nfsacl_decode(&rqstp->rq_arg, base + n, NULL,
+				  (args->mask & NFS_DFACL) ?
+				  &args->acl_default : NULL);
+	return (n > 0);
+}
+
+/*
+ * XDR encode functions
+ */
+
+/* GETACL */
+static int nfs3svc_encode_getaclres(struct svc_rqst *rqstp, u32 *p,
+		struct nfsd3_getaclres *resp)
+{
+	struct dentry *dentry = resp->fh.fh_dentry;
+
+	p = nfs3svc_encode_post_op_attr(rqstp, p, &resp->fh);
+	if (resp->status == 0 && dentry && dentry->d_inode) {
+		struct inode *inode = dentry->d_inode;
+		int w = nfsacl_size(
+			(resp->mask & NFS_ACL)   ? resp->acl_access  : NULL,
+			(resp->mask & NFS_DFACL) ? resp->acl_default : NULL);
+		struct kvec *head = rqstp->rq_res.head;
+		unsigned int base;
+		int n;
+
+		*p++ = htonl(resp->mask);
+		if (!xdr_ressize_check(rqstp, p))
+			return 0;
+		base = (char *)p - (char *)head->iov_base;
+
+		rqstp->rq_res.page_len = w;
+		while (w > 0) {
+			if (!svc_take_res_page(rqstp))
+				return 0;
+			w -= PAGE_SIZE;
+		}
+
+		n = nfsacl_encode(&rqstp->rq_res, base, inode,
+				  resp->acl_access,
+				  resp->mask & NFS_ACL, 0);
+		if (n > 0)
+			n = nfsacl_encode(&rqstp->rq_res, base + n, inode,
+					  resp->acl_default,
+					  resp->mask & NFS_DFACL,
+					  NFS_ACL_DEFAULT);
+		if (n <= 0)
+			return 0;
+	} else
+		if (!xdr_ressize_check(rqstp, p))
+			return 0;
+
+	return 1;
+}
+
+/* SETACL */
+static int nfs3svc_encode_setaclres(struct svc_rqst *rqstp, u32 *p,
+		struct nfsd3_attrstat *resp)
+{
+	p = nfs3svc_encode_post_op_attr(rqstp, p, &resp->fh);
+
+	return xdr_ressize_check(rqstp, p);
+}
+
+/*
+ * XDR release functions
+ */
+static int nfs3svc_release_getacl(struct svc_rqst *rqstp, u32 *p,
+		struct nfsd3_getaclres *resp)
+{
+	fh_put(&resp->fh);
+	posix_acl_release(resp->acl_access);
+	posix_acl_release(resp->acl_default);
+	return 1;
+}
+
+#define nfs3svc_decode_voidargs		NULL
+#define nfs3svc_release_void		NULL
+#define nfsd3_setaclres			nfsd3_attrstat
+#define nfsd3_voidres			nfsd3_voidargs
+struct nfsd3_voidargs { int dummy; };
+
+#define PROC(name, argt, rest, relt, cache, respsize)	\
+ { (svc_procfunc) nfsd3_proc_##name,		\
+   (kxdrproc_t) nfs3svc_decode_##argt##args,	\
+   (kxdrproc_t) nfs3svc_encode_##rest##res,	\
+   (kxdrproc_t) nfs3svc_release_##relt,		\
+   sizeof(struct nfsd3_##argt##args),		\
+   sizeof(struct nfsd3_##rest##res),		\
+   0,						\
+   cache,					\
+   respsize,					\
+ }
+
+#define ST 1		/* status*/
+#define AT 21		/* attributes */
+#define pAT (1+AT)	/* post attributes - conditional */
+#define ACL (1+NFS_ACL_MAX_ENTRIES*3)  /* Access Control List */
+
+static struct svc_procedure		nfsd_acl_procedures3[] = {
+  PROC(null,	void,		void,		void,	  RC_NOCACHE, ST),
+  PROC(getacl,	getacl,		getacl,		getacl,	  RC_NOCACHE, ST+1+2*(1+ACL)),
+  PROC(setacl,	setacl,		setacl,		fhandle,  RC_NOCACHE, ST+pAT),
+};
+
+struct svc_version	nfsd_acl_version3 = {
+		.vs_vers	= 3,
+		.vs_nproc	= 3,
+		.vs_proc	= nfsd_acl_procedures3,
+		.vs_dispatch	= nfsd_dispatch,
+		.vs_xdrsize	= NFS3_SVC_XDRSIZE,
+};
+
diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c
index 11f806835c5..e0e134d6bab 100644
--- a/fs/nfsd/nfs3xdr.c
+++ b/fs/nfsd/nfs3xdr.c
@@ -71,6 +71,12 @@ decode_fh(u32 *p, struct svc_fh *fhp)
 	return p + XDR_QUADLEN(size);
 }
 
+/* Helper function for NFSv3 ACL code */
+u32 *nfs3svc_decode_fh(u32 *p, struct svc_fh *fhp)
+{
+	return decode_fh(p, fhp);
+}
+
 static inline u32 *
 encode_fh(u32 *p, struct svc_fh *fhp)
 {
@@ -233,6 +239,13 @@ encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp)
 	return p;
 }
 
+/* Helper for NFSv3 ACLs */
+u32 *
+nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp)
+{
+	return encode_post_op_attr(rqstp, p, fhp);
+}
+
 /*
  * Enocde weak cache consistency data
  */
diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index 02ded7cfbdc..79b25b19fec 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -31,6 +31,7 @@
 #include <linux/nfsd/stats.h>
 #include <linux/nfsd/cache.h>
 #include <linux/lockd/bind.h>
+#include <linux/nfsacl.h>
 
 #define NFSDDBG_FACILITY	NFSDDBG_SVC
 
@@ -362,6 +363,31 @@ nfsd_dispatch(struct svc_rqst *rqstp, u32 *statp)
 	return 1;
 }
 
+#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL)
+static struct svc_stat	nfsd_acl_svcstats;
+static struct svc_version *	nfsd_acl_version[] = {
+	[2] = &nfsd_acl_version2,
+	[3] = &nfsd_acl_version3,
+};
+
+#define NFSD_ACL_NRVERS		(sizeof(nfsd_acl_version)/sizeof(nfsd_acl_version[0]))
+static struct svc_program	nfsd_acl_program = {
+	.pg_prog		= NFS_ACL_PROGRAM,
+	.pg_nvers		= NFSD_ACL_NRVERS,
+	.pg_vers		= nfsd_acl_version,
+	.pg_name		= "nfsd",
+	.pg_stats		= &nfsd_acl_svcstats,
+};
+
+static struct svc_stat	nfsd_acl_svcstats = {
+	.program	= &nfsd_acl_program,
+};
+
+#define nfsd_acl_program_p	&nfsd_acl_program
+#else
+#define nfsd_acl_program_p	NULL
+#endif /* defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) */
+
 extern struct svc_version nfsd_version2, nfsd_version3, nfsd_version4;
 
 static struct svc_version *	nfsd_version[] = {
@@ -376,6 +402,7 @@ static struct svc_version *	nfsd_version[] = {
 
 #define NFSD_NRVERS		(sizeof(nfsd_version)/sizeof(nfsd_version[0]))
 struct svc_program		nfsd_program = {
+	.pg_next		= nfsd_acl_program_p,
 	.pg_prog		= NFS_PROGRAM,		/* program number */
 	.pg_nvers		= NFSD_NRVERS,		/* nr of entries in nfsd_version */
 	.pg_vers		= nfsd_version,		/* version table */
diff --git a/fs/nfsd/nfsxdr.c b/fs/nfsd/nfsxdr.c
index 948b08287c9..b45999ff33e 100644
--- a/fs/nfsd/nfsxdr.c
+++ b/fs/nfsd/nfsxdr.c
@@ -49,6 +49,12 @@ decode_fh(u32 *p, struct svc_fh *fhp)
 	return p + (NFS_FHSIZE >> 2);
 }
 
+/* Helper function for NFSv2 ACL code */
+u32 *nfs2svc_decode_fh(u32 *p, struct svc_fh *fhp)
+{
+	return decode_fh(p, fhp);
+}
+
 static inline u32 *
 encode_fh(u32 *p, struct svc_fh *fhp)
 {
@@ -190,6 +196,11 @@ encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp)
 	return p;
 }
 
+/* Helper function for NFSv2 ACL code */
+u32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp)
+{
+	return encode_fattr(rqstp, p, fhp);
+}
 
 /*
  * XDR decode functions
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index e3e9d217236..ae3940dc85c 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -46,8 +46,9 @@
 #include <linux/nfsd/nfsfh.h>
 #include <linux/quotaops.h>
 #include <linux/dnotify.h>
-#ifdef CONFIG_NFSD_V4
+#include <linux/xattr_acl.h>
 #include <linux/posix_acl.h>
+#ifdef CONFIG_NFSD_V4
 #include <linux/posix_acl_xattr.h>
 #include <linux/xattr_acl.h>
 #include <linux/xattr.h>
@@ -1857,3 +1858,107 @@ nfsd_racache_init(int cache_size)
 	nfsdstats.ra_size = cache_size;
 	return 0;
 }
+
+#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL)
+struct posix_acl *
+nfsd_get_posix_acl(struct svc_fh *fhp, int type)
+{
+	struct inode *inode = fhp->fh_dentry->d_inode;
+	char *name;
+	void *value = NULL;
+	ssize_t size;
+	struct posix_acl *acl;
+
+	if (!IS_POSIXACL(inode) || !inode->i_op || !inode->i_op->getxattr)
+		return ERR_PTR(-EOPNOTSUPP);
+	switch(type) {
+		case ACL_TYPE_ACCESS:
+			name = XATTR_NAME_ACL_ACCESS;
+			break;
+		case ACL_TYPE_DEFAULT:
+			name = XATTR_NAME_ACL_DEFAULT;
+			break;
+		default:
+			return ERR_PTR(-EOPNOTSUPP);
+	}
+
+	size = inode->i_op->getxattr(fhp->fh_dentry, name, NULL, 0);
+
+	if (size < 0) {
+		acl = ERR_PTR(size);
+		goto getout;
+	} else if (size > 0) {
+		value = kmalloc(size, GFP_KERNEL);
+		if (!value) {
+			acl = ERR_PTR(-ENOMEM);
+			goto getout;
+		}
+		size = inode->i_op->getxattr(fhp->fh_dentry, name, value, size);
+		if (size < 0) {
+			acl = ERR_PTR(size);
+			goto getout;
+		}
+	}
+	acl = posix_acl_from_xattr(value, size);
+
+getout:
+	kfree(value);
+	return acl;
+}
+
+int
+nfsd_set_posix_acl(struct svc_fh *fhp, int type, struct posix_acl *acl)
+{
+	struct inode *inode = fhp->fh_dentry->d_inode;
+	char *name;
+	void *value = NULL;
+	size_t size;
+	int error;
+
+	if (!IS_POSIXACL(inode) || !inode->i_op ||
+	    !inode->i_op->setxattr || !inode->i_op->removexattr)
+		return -EOPNOTSUPP;
+	switch(type) {
+		case ACL_TYPE_ACCESS:
+			name = XATTR_NAME_ACL_ACCESS;
+			break;
+		case ACL_TYPE_DEFAULT:
+			name = XATTR_NAME_ACL_DEFAULT;
+			break;
+		default:
+			return -EOPNOTSUPP;
+	}
+
+	if (acl && acl->a_count) {
+		size = xattr_acl_size(acl->a_count);
+		value = kmalloc(size, GFP_KERNEL);
+		if (!value)
+			return -ENOMEM;
+		size = posix_acl_to_xattr(acl, value, size);
+		if (size < 0) {
+			error = size;
+			goto getout;
+		}
+	} else
+		size = 0;
+
+	if (!fhp->fh_locked)
+		fh_lock(fhp);  /* unlocking is done automatically */
+	if (size)
+		error = inode->i_op->setxattr(fhp->fh_dentry, name,
+					      value, size, 0);
+	else {
+		if (!S_ISDIR(inode->i_mode) && type == ACL_TYPE_DEFAULT)
+			error = 0;
+		else {
+			error = inode->i_op->removexattr(fhp->fh_dentry, name);
+			if (error == -ENODATA)
+				error = 0;
+		}
+	}
+
+getout:
+	kfree(value);
+	return error;
+}
+#endif  /* defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) */
diff --git a/include/linux/nfsacl.h b/include/linux/nfsacl.h
new file mode 100644
index 00000000000..54487a99beb
--- /dev/null
+++ b/include/linux/nfsacl.h
@@ -0,0 +1,58 @@
+/*
+ * File: linux/nfsacl.h
+ *
+ * (C) 2003 Andreas Gruenbacher <agruen@suse.de>
+ */
+#ifndef __LINUX_NFSACL_H
+#define __LINUX_NFSACL_H
+
+#define NFS_ACL_PROGRAM	100227
+
+#define ACLPROC2_GETACL		1
+#define ACLPROC2_SETACL		2
+#define ACLPROC2_GETATTR	3
+#define ACLPROC2_ACCESS		4
+
+#define ACLPROC3_GETACL		1
+#define ACLPROC3_SETACL		2
+
+
+/* Flags for the getacl/setacl mode */
+#define NFS_ACL			0x0001
+#define NFS_ACLCNT		0x0002
+#define NFS_DFACL		0x0004
+#define NFS_DFACLCNT		0x0008
+
+/* Flag for Default ACL entries */
+#define NFS_ACL_DEFAULT		0x1000
+
+#ifdef __KERNEL__
+
+#include <linux/posix_acl.h>
+
+/* Maximum number of ACL entries over NFS */
+#define NFS_ACL_MAX_ENTRIES	1024
+
+#define NFSACL_MAXWORDS		(2*(2+3*NFS_ACL_MAX_ENTRIES))
+#define NFSACL_MAXPAGES		((2*(8+12*NFS_ACL_MAX_ENTRIES) + PAGE_SIZE-1) \
+				 >> PAGE_SHIFT)
+
+static inline unsigned int
+nfsacl_size(struct posix_acl *acl_access, struct posix_acl *acl_default)
+{
+	unsigned int w = 16;
+	w += max(acl_access ? (int)acl_access->a_count : 3, 4) * 12;
+	if (acl_default)
+		w += max((int)acl_default->a_count, 4) * 12;
+	return w;
+}
+
+extern unsigned int
+nfsacl_encode(struct xdr_buf *buf, unsigned int base, struct inode *inode,
+	      struct posix_acl *acl, int encode_entries, int typeflag);
+extern unsigned int
+nfsacl_decode(struct xdr_buf *buf, unsigned int base, unsigned int *aclcnt,
+	      struct posix_acl **pacl);
+
+#endif /* __KERNEL__ */
+#endif  /* __LINUX_NFSACL_H */
diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h
index 8f85d9a5960..4bf931d5ff5 100644
--- a/include/linux/nfsd/nfsd.h
+++ b/include/linux/nfsd/nfsd.h
@@ -15,6 +15,7 @@
 #include <linux/unistd.h>
 #include <linux/dirent.h>
 #include <linux/fs.h>
+#include <linux/posix_acl.h>
 #include <linux/mount.h>
 
 #include <linux/nfsd/debug.h>
@@ -124,6 +125,21 @@ int		nfsd_statfs(struct svc_rqst *, struct svc_fh *,
 int		nfsd_notify_change(struct inode *, struct iattr *);
 int		nfsd_permission(struct svc_export *, struct dentry *, int);
 
+#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL)
+#ifdef CONFIG_NFSD_V2_ACL
+extern struct svc_version nfsd_acl_version2;
+#else
+#define nfsd_acl_version2 NULL
+#endif
+#ifdef CONFIG_NFSD_V3_ACL
+extern struct svc_version nfsd_acl_version3;
+#else
+#define nfsd_acl_version3 NULL
+#endif
+struct posix_acl *nfsd_get_posix_acl(struct svc_fh *, int);
+int nfsd_set_posix_acl(struct svc_fh *, int, struct posix_acl *);
+#endif
+
 
 /* 
  * NFSv4 State
diff --git a/include/linux/nfsd/xdr.h b/include/linux/nfsd/xdr.h
index ecccef777da..130d4f588a3 100644
--- a/include/linux/nfsd/xdr.h
+++ b/include/linux/nfsd/xdr.h
@@ -169,4 +169,8 @@ int nfssvc_encode_entry(struct readdir_cd *, const char *name,
 
 int nfssvc_release_fhandle(struct svc_rqst *, u32 *, struct nfsd_fhandle *);
 
+/* Helper functions for NFSv2 ACL code */
+u32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp);
+u32 *nfs2svc_decode_fh(u32 *p, struct svc_fh *fhp);
+
 #endif /* LINUX_NFSD_H */
diff --git a/include/linux/nfsd/xdr3.h b/include/linux/nfsd/xdr3.h
index 0ae9e0ef5f6..21e18ce7ca6 100644
--- a/include/linux/nfsd/xdr3.h
+++ b/include/linux/nfsd/xdr3.h
@@ -110,6 +110,19 @@ struct nfsd3_commitargs {
 	__u32			count;
 };
 
+struct nfsd3_getaclargs {
+	struct svc_fh		fh;
+	int			mask;
+};
+
+struct posix_acl;
+struct nfsd3_setaclargs {
+	struct svc_fh		fh;
+	int			mask;
+	struct posix_acl	*acl_access;
+	struct posix_acl	*acl_default;
+};
+
 struct nfsd3_attrstat {
 	__u32			status;
 	struct svc_fh		fh;
@@ -209,6 +222,14 @@ struct nfsd3_commitres {
 	struct svc_fh		fh;
 };
 
+struct nfsd3_getaclres {
+	__u32			status;
+	struct svc_fh		fh;
+	int			mask;
+	struct posix_acl	*acl_access;
+	struct posix_acl	*acl_default;
+};
+
 /* dummy type for release */
 struct nfsd3_fhandle_pair {
 	__u32			dummy;
@@ -241,6 +262,7 @@ union nfsd3_xdrstore {
 	struct nfsd3_fsinfores		fsinfores;
 	struct nfsd3_pathconfres	pathconfres;
 	struct nfsd3_commitres		commitres;
+	struct nfsd3_getaclres		getaclres;
 };
 
 #define NFS3_SVC_XDRSIZE		sizeof(union nfsd3_xdrstore)
@@ -316,6 +338,10 @@ int nfs3svc_encode_entry(struct readdir_cd *, const char *name,
 int nfs3svc_encode_entry_plus(struct readdir_cd *, const char *name,
 				int namlen, loff_t offset, ino_t ino,
 				unsigned int);
+/* Helper functions for NFSv3 ACL code */
+u32 *nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, u32 *p,
+				struct svc_fh *fhp);
+u32 *nfs3svc_decode_fh(u32 *p, struct svc_fh *fhp);
 
 
 #endif /* _LINUX_NFSD_XDR3_H */
diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h
index facb94488bb..5af8800e0ce 100644
--- a/include/linux/sunrpc/svc.h
+++ b/include/linux/sunrpc/svc.h
@@ -185,6 +185,17 @@ xdr_ressize_check(struct svc_rqst *rqstp, u32 *p)
 	return vec->iov_len <= PAGE_SIZE;
 }
 
+static inline struct page *
+svc_take_res_page(struct svc_rqst *rqstp)
+{
+	if (rqstp->rq_arghi <= rqstp->rq_argused)
+		return NULL;
+	rqstp->rq_arghi--;
+	rqstp->rq_respages[rqstp->rq_resused] =
+		rqstp->rq_argpages[rqstp->rq_arghi];
+	return rqstp->rq_respages[rqstp->rq_resused++];
+}
+
 static inline int svc_take_page(struct svc_rqst *rqstp)
 {
 	if (rqstp->rq_arghi <= rqstp->rq_argused)
-- 
cgit v1.2.3-70-g09d2


From b7fa0554cf1ba6d6895cd0a5b02989a26e0bc704 Mon Sep 17 00:00:00 2001
From: Andreas Gruenbacher <agruen@suse.de>
Date: Wed, 22 Jun 2005 17:16:27 +0000
Subject: [PATCH] NFS: Add support for NFSv3 ACLs

 This adds acl support fo nfs clients via the NFSACL protocol extension, by
 implementing the getxattr, listxattr, setxattr, and removexattr iops for the
 system.posix_acl_access and system.posix_acl_default attributes.  This patch
 implements a dumb version that uses no caching (and thus adds some overhead).
 (Another patch in this patchset adds caching as well.)

 Signed-off-by: Andreas Gruenbacher <agruen@suse.de>
 Acked-by: Olaf Kirch <okir@suse.de>
 Signed-off-by: Andrew Morton <akpm@osdl.org>
 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/Kconfig                |  11 ++
 fs/nfs/Makefile           |   1 +
 fs/nfs/dir.c              |  21 ++++
 fs/nfs/file.c             |  12 ++
 fs/nfs/inode.c            |  36 +++++-
 fs/nfs/nfs3acl.c          | 303 ++++++++++++++++++++++++++++++++++++++++++++++
 fs/nfs/nfs3proc.c         |   7 +-
 fs/nfs/nfs3xdr.c          | 147 ++++++++++++++++++++++
 fs/nfs/nfsroot.c          |   9 ++
 include/linux/nfs_fs.h    |  31 +++++
 include/linux/nfs_fs_sb.h |   1 +
 include/linux/nfs_mount.h |   1 +
 include/linux/nfs_xdr.h   |  27 +++++
 13 files changed, 601 insertions(+), 6 deletions(-)
 create mode 100644 fs/nfs/nfs3acl.c

(limited to 'include')

diff --git a/fs/Kconfig b/fs/Kconfig
index d44b04d9b0a..a7c0cc3203c 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -1268,6 +1268,7 @@ config NFS_FS
 	depends on INET
 	select LOCKD
 	select SUNRPC
+	select NFS_ACL_SUPPORT if NFS_V3_ACL
 	help
 	  If you are connected to some other (usually local) Unix computer
 	  (using SLIP, PLIP, PPP or Ethernet) and want to mount files residing
@@ -1310,6 +1311,16 @@ config NFS_V3
 
 	  If unsure, say Y.
 
+config NFS_V3_ACL
+	bool "Provide client support for the NFSv3 ACL protocol extension"
+	depends on NFS_V3
+	help
+	  Implement the NFSv3 ACL protocol extension for manipulating POSIX
+	  Access Control Lists.  The server should also be compiled with
+	  the NFSv3 ACL protocol extension; see the CONFIG_NFSD_V3_ACL option.
+
+	  If unsure, say N.
+
 config NFS_V4
 	bool "Provide NFSv4 client support (EXPERIMENTAL)"
 	depends on NFS_FS && EXPERIMENTAL
diff --git a/fs/nfs/Makefile b/fs/nfs/Makefile
index b4baa031edf..8b3bb715d17 100644
--- a/fs/nfs/Makefile
+++ b/fs/nfs/Makefile
@@ -8,6 +8,7 @@ nfs-y 			:= dir.o file.o inode.o nfs2xdr.o pagelist.o \
 			   proc.o read.o symlink.o unlink.o write.o
 nfs-$(CONFIG_ROOT_NFS)	+= nfsroot.o mount_clnt.o      
 nfs-$(CONFIG_NFS_V3)	+= nfs3proc.o nfs3xdr.o
+nfs-$(CONFIG_NFS_V3_ACL)	+= nfs3acl.o
 nfs-$(CONFIG_NFS_V4)	+= nfs4proc.o nfs4xdr.o nfs4state.o nfs4renewd.o \
 			   delegation.o idmap.o \
 			   callback.o callback_xdr.o callback_proc.o
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 5720537bffd..2c6a9594568 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -75,6 +75,27 @@ struct inode_operations nfs_dir_inode_operations = {
 	.setattr	= nfs_setattr,
 };
 
+#ifdef CONFIG_NFS_V3
+struct inode_operations nfs3_dir_inode_operations = {
+	.create		= nfs_create,
+	.lookup		= nfs_lookup,
+	.link		= nfs_link,
+	.unlink		= nfs_unlink,
+	.symlink	= nfs_symlink,
+	.mkdir		= nfs_mkdir,
+	.rmdir		= nfs_rmdir,
+	.mknod		= nfs_mknod,
+	.rename		= nfs_rename,
+	.permission	= nfs_permission,
+	.getattr	= nfs_getattr,
+	.setattr	= nfs_setattr,
+	.listxattr	= nfs3_listxattr,
+	.getxattr	= nfs3_getxattr,
+	.setxattr	= nfs3_setxattr,
+	.removexattr	= nfs3_removexattr,
+};
+#endif  /* CONFIG_NFS_V3 */
+
 #ifdef CONFIG_NFS_V4
 
 static struct dentry *nfs_atomic_lookup(struct inode *, struct dentry *, struct nameidata *);
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 55c90759249..a606708264e 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -71,6 +71,18 @@ struct inode_operations nfs_file_inode_operations = {
 	.setattr	= nfs_setattr,
 };
 
+#ifdef CONFIG_NFS_V3
+struct inode_operations nfs3_file_inode_operations = {
+	.permission	= nfs_permission,
+	.getattr	= nfs_getattr,
+	.setattr	= nfs_setattr,
+	.listxattr	= nfs3_listxattr,
+	.getxattr	= nfs3_getxattr,
+	.setxattr	= nfs3_setxattr,
+	.removexattr	= nfs3_removexattr,
+};
+#endif  /* CONFIG_NFS_v3 */
+
 /* Hack for future NFS swap support */
 #ifndef IS_SWAPFILE
 # define IS_SWAPFILE(inode)	(0)
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 97b3fe7ece6..440b9cbb6f8 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -108,6 +108,21 @@ static struct rpc_program	nfs_program = {
 	.pipe_dir_name		= "/nfs",
 };
 
+#ifdef CONFIG_NFS_V3_ACL
+static struct rpc_stat		nfsacl_rpcstat = { &nfsacl_program };
+static struct rpc_version *	nfsacl_version[] = {
+	[3]			= &nfsacl_version3,
+};
+
+struct rpc_program		nfsacl_program = {
+	.name =			"nfsacl",
+	.number =		NFS_ACL_PROGRAM,
+	.nrvers =		sizeof(nfsacl_version) / sizeof(nfsacl_version[0]),
+	.version =		nfsacl_version,
+	.stats =		&nfsacl_rpcstat,
+};
+#endif  /* CONFIG_NFS_V3_ACL */
+
 static inline unsigned long
 nfs_fattr_to_ino_t(struct nfs_fattr *fattr)
 {
@@ -165,6 +180,9 @@ nfs_umount_begin(struct super_block *sb)
 	/* -EIO all pending I/O */
 	if (!IS_ERR(rpc))
 		rpc_killall_tasks(rpc);
+	rpc = NFS_SB(sb)->client_acl;
+	if (!IS_ERR(rpc))
+		rpc_killall_tasks(rpc);
 }
 
 
@@ -461,8 +479,17 @@ nfs_fill_super(struct super_block *sb, struct nfs_mount_data *data, int silent)
 		atomic_inc(&server->client->cl_count);
 		server->client_sys = server->client;
 	}
-
 	if (server->flags & NFS_MOUNT_VER3) {
+#ifdef CONFIG_NFS_V3_ACL
+		if (!(server->flags & NFS_MOUNT_NOACL)) {
+			server->client_acl = rpc_bind_new_program(server->client, &nfsacl_program, 3);
+			/* No errors! Assume that Sun nfsacls are supported */
+			if (!IS_ERR(server->client_acl))
+				server->caps |= NFS_CAP_ACLS;
+		}
+#else
+		server->flags &= ~NFS_MOUNT_NOACL;
+#endif /* CONFIG_NFS_V3_ACL */
 		if (server->namelen == 0 || server->namelen > NFS3_MAXNAMLEN)
 			server->namelen = NFS3_MAXNAMLEN;
 		sb->s_time_gran = 1;
@@ -546,6 +573,7 @@ static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt)
 		{ NFS_MOUNT_NOCTO, ",nocto", "" },
 		{ NFS_MOUNT_NOAC, ",noac", "" },
 		{ NFS_MOUNT_NONLM, ",nolock", ",lock" },
+		{ NFS_MOUNT_NOACL, ",noacl", "" },
 		{ 0, NULL, NULL }
 	};
 	struct proc_nfs_info *nfs_infop;
@@ -1452,7 +1480,7 @@ static struct super_block *nfs_get_sb(struct file_system_type *fs_type,
 	memset(server, 0, sizeof(struct nfs_server));
 	/* Zero out the NFS state stuff */
 	init_nfsv4_state(server);
-	server->client = server->client_sys = ERR_PTR(-EINVAL);
+	server->client = server->client_sys = server->client_acl = ERR_PTR(-EINVAL);
 
 	root = &server->fh;
 	if (data->flags & NFS_MOUNT_VER3)
@@ -1513,6 +1541,8 @@ static void nfs_kill_super(struct super_block *s)
 		rpc_shutdown_client(server->client);
 	if (!IS_ERR(server->client_sys))
 		rpc_shutdown_client(server->client_sys);
+	if (!IS_ERR(server->client_acl))
+		rpc_shutdown_client(server->client_acl);
 
 	if (!(server->flags & NFS_MOUNT_NONLM))
 		lockd_down();	/* release rpc.lockd */
@@ -1794,7 +1824,7 @@ static struct super_block *nfs4_get_sb(struct file_system_type *fs_type,
 	memset(server, 0, sizeof(struct nfs_server));
 	/* Zero out the NFS state stuff */
 	init_nfsv4_state(server);
-	server->client = server->client_sys = ERR_PTR(-EINVAL);
+	server->client = server->client_sys = server->client_acl = ERR_PTR(-EINVAL);
 
 	p = nfs_copy_user_string(NULL, &data->hostname, 256);
 	if (IS_ERR(p))
diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c
new file mode 100644
index 00000000000..393ba79fc14
--- /dev/null
+++ b/fs/nfs/nfs3acl.c
@@ -0,0 +1,303 @@
+#include <linux/fs.h>
+#include <linux/nfs.h>
+#include <linux/nfs3.h>
+#include <linux/nfs_fs.h>
+#include <linux/xattr_acl.h>
+#include <linux/nfsacl.h>
+
+#define NFSDBG_FACILITY	NFSDBG_PROC
+
+ssize_t nfs3_listxattr(struct dentry *dentry, char *buffer, size_t size)
+{
+	struct inode *inode = dentry->d_inode;
+	struct posix_acl *acl;
+	int pos=0, len=0;
+
+#	define output(s) do {						\
+			if (pos + sizeof(s) <= size) {			\
+				memcpy(buffer + pos, s, sizeof(s));	\
+				pos += sizeof(s);			\
+			}						\
+			len += sizeof(s);				\
+		} while(0)
+
+	acl = nfs3_proc_getacl(inode, ACL_TYPE_ACCESS);
+	if (IS_ERR(acl))
+		return PTR_ERR(acl);
+	if (acl) {
+		output("system.posix_acl_access");
+		posix_acl_release(acl);
+	}
+
+	if (S_ISDIR(inode->i_mode)) {
+		acl = nfs3_proc_getacl(inode, ACL_TYPE_DEFAULT);
+		if (IS_ERR(acl))
+			return PTR_ERR(acl);
+		if (acl) {
+			output("system.posix_acl_default");
+			posix_acl_release(acl);
+		}
+	}
+
+#	undef output
+
+	if (!buffer || len <= size)
+		return len;
+	return -ERANGE;
+}
+
+ssize_t nfs3_getxattr(struct dentry *dentry, const char *name,
+		void *buffer, size_t size)
+{
+	struct inode *inode = dentry->d_inode;
+	struct posix_acl *acl;
+	int type, error = 0;
+
+	if (strcmp(name, XATTR_NAME_ACL_ACCESS) == 0)
+		type = ACL_TYPE_ACCESS;
+	else if (strcmp(name, XATTR_NAME_ACL_DEFAULT) == 0)
+		type = ACL_TYPE_DEFAULT;
+	else
+		return -EOPNOTSUPP;
+
+	acl = nfs3_proc_getacl(inode, type);
+	if (IS_ERR(acl))
+		return PTR_ERR(acl);
+	else if (acl) {
+		if (type == ACL_TYPE_ACCESS && acl->a_count == 0)
+			error = -ENODATA;
+		else
+			error = posix_acl_to_xattr(acl, buffer, size);
+		posix_acl_release(acl);
+	} else
+		error = -ENODATA;
+
+	return error;
+}
+
+int nfs3_setxattr(struct dentry *dentry, const char *name,
+	     const void *value, size_t size, int flags)
+{
+	struct inode *inode = dentry->d_inode;
+	struct posix_acl *acl;
+	int type, error;
+
+	if (strcmp(name, XATTR_NAME_ACL_ACCESS) == 0)
+		type = ACL_TYPE_ACCESS;
+	else if (strcmp(name, XATTR_NAME_ACL_DEFAULT) == 0)
+		type = ACL_TYPE_DEFAULT;
+	else
+		return -EOPNOTSUPP;
+
+	acl = posix_acl_from_xattr(value, size);
+	if (IS_ERR(acl))
+		return PTR_ERR(acl);
+	error = nfs3_proc_setacl(inode, type, acl);
+	posix_acl_release(acl);
+
+	return error;
+}
+
+int nfs3_removexattr(struct dentry *dentry, const char *name)
+{
+	struct inode *inode = dentry->d_inode;
+	int type;
+
+	if (strcmp(name, XATTR_NAME_ACL_ACCESS) == 0)
+		type = ACL_TYPE_ACCESS;
+	else if (strcmp(name, XATTR_NAME_ACL_DEFAULT) == 0)
+		type = ACL_TYPE_DEFAULT;
+	else
+		return -EOPNOTSUPP;
+
+	return nfs3_proc_setacl(inode, type, NULL);
+}
+
+struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type)
+{
+	struct nfs_server *server = NFS_SERVER(inode);
+	struct nfs_fattr fattr;
+	struct page *pages[NFSACL_MAXPAGES] = { };
+	struct nfs3_getaclargs args = {
+		.fh = NFS_FH(inode),
+		/* The xdr layer may allocate pages here. */
+		.pages = pages,
+	};
+	struct nfs3_getaclres res = {
+		.fattr =	&fattr,
+	};
+	struct posix_acl *acl = NULL;
+	int status, count;
+
+	if (!nfs_server_capable(inode, NFS_CAP_ACLS))
+		return ERR_PTR(-EOPNOTSUPP);
+
+	switch (type) {
+		case ACL_TYPE_ACCESS:
+			args.mask = NFS_ACLCNT|NFS_ACL;
+			break;
+
+		case ACL_TYPE_DEFAULT:
+			if (!S_ISDIR(inode->i_mode))
+				return NULL;
+			args.mask = NFS_DFACLCNT|NFS_DFACL;
+			break;
+
+		default:
+			return ERR_PTR(-EINVAL);
+	}
+
+	dprintk("NFS call getacl\n");
+	status = rpc_call(server->client_acl, ACLPROC3_GETACL,
+			  &args, &res, 0);
+	dprintk("NFS reply getacl: %d\n", status);
+
+	/* pages may have been allocated at the xdr layer. */
+	for (count = 0; count < NFSACL_MAXPAGES && args.pages[count]; count++)
+		__free_page(args.pages[count]);
+
+	switch (status) {
+		case 0:
+			status = nfs_refresh_inode(inode, &fattr);
+			break;
+		case -EPFNOSUPPORT:
+		case -EPROTONOSUPPORT:
+			dprintk("NFS_V3_ACL extension not supported; disabling\n");
+			server->caps &= ~NFS_CAP_ACLS;
+		case -ENOTSUPP:
+			status = -EOPNOTSUPP;
+		default:
+			goto getout;
+	}
+	if ((args.mask & res.mask) != args.mask) {
+		status = -EIO;
+		goto getout;
+	}
+
+	if (res.acl_access != NULL) {
+		if (posix_acl_equiv_mode(res.acl_access, NULL) == 0) {
+			posix_acl_release(res.acl_access);
+			res.acl_access = NULL;
+		}
+	}
+
+	switch(type) {
+		case ACL_TYPE_ACCESS:
+			acl = res.acl_access;
+			res.acl_access = NULL;
+			break;
+
+		case ACL_TYPE_DEFAULT:
+			acl = res.acl_default;
+			res.acl_default = NULL;
+	}
+
+getout:
+	posix_acl_release(res.acl_access);
+	posix_acl_release(res.acl_default);
+
+	if (status != 0) {
+		posix_acl_release(acl);
+		acl = ERR_PTR(status);
+	}
+	return acl;
+}
+
+static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl,
+		  struct posix_acl *dfacl)
+{
+	struct nfs_server *server = NFS_SERVER(inode);
+	struct nfs_fattr fattr;
+	struct page *pages[NFSACL_MAXPAGES] = { };
+	struct nfs3_setaclargs args = {
+		.inode = inode,
+		.mask = NFS_ACL,
+		.acl_access = acl,
+		.pages = pages,
+	};
+	int status, count;
+
+	status = -EOPNOTSUPP;
+	if (!nfs_server_capable(inode, NFS_CAP_ACLS))
+		goto out;
+
+	/* We are doing this here, because XDR marshalling can only
+	   return -ENOMEM. */
+	status = -ENOSPC;
+	if (acl != NULL && acl->a_count > NFS_ACL_MAX_ENTRIES)
+		goto out;
+	if (dfacl != NULL && dfacl->a_count > NFS_ACL_MAX_ENTRIES)
+		goto out;
+	if (S_ISDIR(inode->i_mode)) {
+		args.mask |= NFS_DFACL;
+		args.acl_default = dfacl;
+	}
+
+	dprintk("NFS call setacl\n");
+	nfs_begin_data_update(inode);
+	status = rpc_call(server->client_acl, ACLPROC3_SETACL,
+			  &args, &fattr, 0);
+	NFS_FLAGS(inode) |= NFS_INO_INVALID_ACCESS;
+	nfs_end_data_update(inode);
+	dprintk("NFS reply setacl: %d\n", status);
+
+	/* pages may have been allocated at the xdr layer. */
+	for (count = 0; count < NFSACL_MAXPAGES && args.pages[count]; count++)
+		__free_page(args.pages[count]);
+
+	switch (status) {
+		case 0:
+			status = nfs_refresh_inode(inode, &fattr);
+			break;
+		case -EPFNOSUPPORT:
+		case -EPROTONOSUPPORT:
+			dprintk("NFS_V3_ACL SETACL RPC not supported"
+					"(will not retry)\n");
+			server->caps &= ~NFS_CAP_ACLS;
+		case -ENOTSUPP:
+			status = -EOPNOTSUPP;
+	}
+out:
+	return status;
+}
+
+int nfs3_proc_setacl(struct inode *inode, int type, struct posix_acl *acl)
+{
+	struct posix_acl *alloc = NULL, *dfacl = NULL;
+	int status;
+
+	if (S_ISDIR(inode->i_mode)) {
+		switch(type) {
+			case ACL_TYPE_ACCESS:
+				alloc = dfacl = nfs3_proc_getacl(inode,
+						ACL_TYPE_DEFAULT);
+				if (IS_ERR(alloc))
+					goto fail;
+				break;
+
+			case ACL_TYPE_DEFAULT:
+				dfacl = acl;
+				alloc = acl = nfs3_proc_getacl(inode,
+						ACL_TYPE_ACCESS);
+				if (IS_ERR(alloc))
+					goto fail;
+				break;
+
+			default:
+				return -EINVAL;
+		}
+	} else if (type != ACL_TYPE_ACCESS)
+			return -EINVAL;
+
+	if (acl == NULL) {
+		alloc = acl = posix_acl_from_mode(inode->i_mode, GFP_KERNEL);
+		if (IS_ERR(alloc))
+			goto fail;
+	}
+	status = nfs3_proc_setacls(inode, acl, dfacl);
+	posix_acl_release(alloc);
+	return status;
+
+fail:
+	return PTR_ERR(alloc);
+}
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index 53953a77571..d03bac0cc42 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -17,6 +17,7 @@
 #include <linux/nfs_page.h>
 #include <linux/lockd/bind.h>
 #include <linux/smp_lock.h>
+#include <linux/nfs_mount.h>
 
 #define NFSDBG_FACILITY		NFSDBG_PROC
 
@@ -45,7 +46,7 @@ static inline int
 nfs3_rpc_call_wrapper(struct rpc_clnt *clnt, u32 proc, void *argp, void *resp, int flags)
 {
 	struct rpc_message msg = {
-		.rpc_proc	= &nfs3_procedures[proc],
+		.rpc_proc	= &clnt->cl_procinfo[proc],
 		.rpc_argp	= argp,
 		.rpc_resp	= resp,
 	};
@@ -825,8 +826,8 @@ nfs3_proc_lock(struct file *filp, int cmd, struct file_lock *fl)
 struct nfs_rpc_ops	nfs_v3_clientops = {
 	.version	= 3,			/* protocol version */
 	.dentry_ops	= &nfs_dentry_operations,
-	.dir_inode_ops	= &nfs_dir_inode_operations,
-	.file_inode_ops	= &nfs_file_inode_operations,
+	.dir_inode_ops	= &nfs3_dir_inode_operations,
+	.file_inode_ops	= &nfs3_file_inode_operations,
 	.getroot	= nfs3_proc_get_root,
 	.getattr	= nfs3_proc_getattr,
 	.setattr	= nfs3_proc_setattr,
diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c
index a3593d47e5a..a4437fb177f 100644
--- a/fs/nfs/nfs3xdr.c
+++ b/fs/nfs/nfs3xdr.c
@@ -21,6 +21,7 @@
 #include <linux/nfs.h>
 #include <linux/nfs3.h>
 #include <linux/nfs_fs.h>
+#include <linux/nfsacl.h>
 
 #define NFSDBG_FACILITY		NFSDBG_XDR
 
@@ -79,6 +80,11 @@ extern int			nfs_stat_to_errno(int);
 #define NFS3_pathconfres_sz	(1+NFS3_post_op_attr_sz+6)
 #define NFS3_commitres_sz	(1+NFS3_wcc_data_sz+2)
 
+#define ACL3_getaclargs_sz	(NFS3_fh_sz+1)
+#define ACL3_setaclargs_sz	(NFS3_fh_sz+1+2*(2+5*3))
+#define ACL3_getaclres_sz	(1+NFS3_post_op_attr_sz+1+2*(2+5*3))
+#define ACL3_setaclres_sz	(1+NFS3_post_op_attr_sz)
+
 /*
  * Map file type to S_IFMT bits
  */
@@ -627,6 +633,74 @@ nfs3_xdr_commitargs(struct rpc_rqst *req, u32 *p, struct nfs_writeargs *args)
 	return 0;
 }
 
+#ifdef CONFIG_NFS_V3_ACL
+/*
+ * Encode GETACL arguments
+ */
+static int
+nfs3_xdr_getaclargs(struct rpc_rqst *req, u32 *p,
+		    struct nfs3_getaclargs *args)
+{
+	struct rpc_auth *auth = req->rq_task->tk_auth;
+	unsigned int replen;
+
+	p = xdr_encode_fhandle(p, args->fh);
+	*p++ = htonl(args->mask);
+	req->rq_slen = xdr_adjust_iovec(req->rq_svec, p);
+
+	if (args->mask & (NFS_ACL | NFS_DFACL)) {
+		/* Inline the page array */
+		replen = (RPC_REPHDRSIZE + auth->au_rslack +
+			  ACL3_getaclres_sz) << 2;
+		xdr_inline_pages(&req->rq_rcv_buf, replen, args->pages, 0,
+				 NFSACL_MAXPAGES << PAGE_SHIFT);
+	}
+	return 0;
+}
+
+/*
+ * Encode SETACL arguments
+ */
+static int
+nfs3_xdr_setaclargs(struct rpc_rqst *req, u32 *p,
+                   struct nfs3_setaclargs *args)
+{
+	struct xdr_buf *buf = &req->rq_snd_buf;
+	unsigned int base, len_in_head, len = nfsacl_size(
+		(args->mask & NFS_ACL)   ? args->acl_access  : NULL,
+		(args->mask & NFS_DFACL) ? args->acl_default : NULL);
+	int count, err;
+
+	p = xdr_encode_fhandle(p, NFS_FH(args->inode));
+	*p++ = htonl(args->mask);
+	base = (char *)p - (char *)buf->head->iov_base;
+	/* put as much of the acls into head as possible. */
+	len_in_head = min_t(unsigned int, buf->head->iov_len - base, len);
+	len -= len_in_head;
+	req->rq_slen = xdr_adjust_iovec(req->rq_svec, p + len_in_head);
+
+	for (count = 0; (count << PAGE_SHIFT) < len; count++) {
+		args->pages[count] = alloc_page(GFP_KERNEL);
+		if (!args->pages[count]) {
+			while (count)
+				__free_page(args->pages[--count]);
+			return -ENOMEM;
+		}
+	}
+	xdr_encode_pages(buf, args->pages, 0, len);
+
+	err = nfsacl_encode(buf, base, args->inode,
+			    (args->mask & NFS_ACL) ?
+			    args->acl_access : NULL, 1, 0);
+	if (err > 0)
+		err = nfsacl_encode(buf, base + err, args->inode,
+				    (args->mask & NFS_DFACL) ?
+				    args->acl_default : NULL, 1,
+				    NFS_ACL_DEFAULT);
+	return (err > 0) ? 0 : err;
+}
+#endif  /* CONFIG_NFS_V3_ACL */
+
 /*
  * NFS XDR decode functions
  */
@@ -978,6 +1052,54 @@ nfs3_xdr_commitres(struct rpc_rqst *req, u32 *p, struct nfs_writeres *res)
 	return 0;
 }
 
+#ifdef CONFIG_NFS_V3_ACL
+/*
+ * Decode GETACL reply
+ */
+static int
+nfs3_xdr_getaclres(struct rpc_rqst *req, u32 *p,
+		   struct nfs3_getaclres *res)
+{
+	struct xdr_buf *buf = &req->rq_rcv_buf;
+	int status = ntohl(*p++);
+	struct posix_acl **acl;
+	unsigned int *aclcnt;
+	int err, base;
+
+	if (status != 0)
+		return -nfs_stat_to_errno(status);
+	p = xdr_decode_post_op_attr(p, res->fattr);
+	res->mask = ntohl(*p++);
+	if (res->mask & ~(NFS_ACL|NFS_ACLCNT|NFS_DFACL|NFS_DFACLCNT))
+		return -EINVAL;
+	base = (char *)p - (char *)req->rq_rcv_buf.head->iov_base;
+
+	acl = (res->mask & NFS_ACL) ? &res->acl_access : NULL;
+	aclcnt = (res->mask & NFS_ACLCNT) ? &res->acl_access_count : NULL;
+	err = nfsacl_decode(buf, base, aclcnt, acl);
+
+	acl = (res->mask & NFS_DFACL) ? &res->acl_default : NULL;
+	aclcnt = (res->mask & NFS_DFACLCNT) ? &res->acl_default_count : NULL;
+	if (err > 0)
+		err = nfsacl_decode(buf, base + err, aclcnt, acl);
+	return (err > 0) ? 0 : err;
+}
+
+/*
+ * Decode setacl reply.
+ */
+static int
+nfs3_xdr_setaclres(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr)
+{
+	int status = ntohl(*p++);
+
+	if (status)
+		return -nfs_stat_to_errno(status);
+	xdr_decode_post_op_attr(p, fattr);
+	return 0;
+}
+#endif  /* CONFIG_NFS_V3_ACL */
+
 #ifndef MAX
 # define MAX(a, b)	(((a) > (b))? (a) : (b))
 #endif
@@ -1021,3 +1143,28 @@ struct rpc_version		nfs_version3 = {
 	.procs			= nfs3_procedures
 };
 
+#ifdef CONFIG_NFS_V3_ACL
+static struct rpc_procinfo	nfs3_acl_procedures[] = {
+	[ACLPROC3_GETACL] = {
+		.p_proc = ACLPROC3_GETACL,
+		.p_encode = (kxdrproc_t) nfs3_xdr_getaclargs,
+		.p_decode = (kxdrproc_t) nfs3_xdr_getaclres,
+		.p_bufsiz = MAX(ACL3_getaclargs_sz, ACL3_getaclres_sz) << 2,
+		.p_timer = 1,
+	},
+	[ACLPROC3_SETACL] = {
+		.p_proc = ACLPROC3_SETACL,
+		.p_encode = (kxdrproc_t) nfs3_xdr_setaclargs,
+		.p_decode = (kxdrproc_t) nfs3_xdr_setaclres,
+		.p_bufsiz = MAX(ACL3_setaclargs_sz, ACL3_setaclres_sz) << 2,
+		.p_timer = 0,
+	},
+};
+
+struct rpc_version		nfsacl_version3 = {
+	.number			= 3,
+	.nrprocs		= sizeof(nfs3_acl_procedures)/
+				  sizeof(nfs3_acl_procedures[0]),
+	.procs			= nfs3_acl_procedures,
+};
+#endif  /* CONFIG_NFS_V3_ACL */
diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c
index fd5bc596fe8..1b272a135a3 100644
--- a/fs/nfs/nfsroot.c
+++ b/fs/nfs/nfsroot.c
@@ -124,6 +124,7 @@ enum {
 	Opt_soft, Opt_hard, Opt_intr,
 	Opt_nointr, Opt_posix, Opt_noposix, Opt_cto, Opt_nocto, Opt_ac, 
 	Opt_noac, Opt_lock, Opt_nolock, Opt_v2, Opt_v3, Opt_udp, Opt_tcp,
+	Opt_acl, Opt_noacl,
 	/* Error token */
 	Opt_err
 };
@@ -158,6 +159,8 @@ static match_table_t __initdata tokens = {
 	{Opt_udp, "udp"},
 	{Opt_tcp, "proto=tcp"},
 	{Opt_tcp, "tcp"},
+	{Opt_acl, "acl"},
+	{Opt_noacl, "noacl"},
 	{Opt_err, NULL}
 	
 };
@@ -266,6 +269,12 @@ static int __init root_nfs_parse(char *name, char *buf)
 			case Opt_tcp:
 				nfs_data.flags |= NFS_MOUNT_TCP;
 				break;
+			case Opt_acl:
+				nfs_data.flags &= ~NFS_MOUNT_NOACL;
+				break;
+			case Opt_noacl:
+				nfs_data.flags |= NFS_MOUNT_NOACL;
+				break;
 			default : 
 				return 0;
 		}
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index d2b5d7e0e85..3a5e442ac77 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -301,6 +301,9 @@ extern u32 root_nfs_parse_addr(char *name); /*__init*/
  * linux/fs/nfs/file.c
  */
 extern struct inode_operations nfs_file_inode_operations;
+#ifdef CONFIG_NFS_V3
+extern struct inode_operations nfs3_file_inode_operations;
+#endif /* CONFIG_NFS_V3 */
 extern struct file_operations nfs_file_operations;
 extern struct address_space_operations nfs_file_aops;
 
@@ -315,6 +318,22 @@ static inline struct rpc_cred *nfs_file_cred(struct file *file)
 	return NULL;
 }
 
+/*
+ * linux/fs/nfs/xattr.c
+ */
+#ifdef CONFIG_NFS_V3_ACL
+extern ssize_t nfs3_listxattr(struct dentry *, char *, size_t);
+extern ssize_t nfs3_getxattr(struct dentry *, const char *, void *, size_t);
+extern int nfs3_setxattr(struct dentry *, const char *,
+			const void *, size_t, int);
+extern int nfs3_removexattr (struct dentry *, const char *name);
+#else
+# define nfs3_listxattr NULL
+# define nfs3_getxattr NULL
+# define nfs3_setxattr NULL
+# define nfs3_removexattr NULL
+#endif
+
 /*
  * linux/fs/nfs/direct.c
  */
@@ -329,6 +348,9 @@ extern ssize_t nfs_file_direct_write(struct kiocb *iocb, const char __user *buf,
  * linux/fs/nfs/dir.c
  */
 extern struct inode_operations nfs_dir_inode_operations;
+#ifdef CONFIG_NFS_V3
+extern struct inode_operations nfs3_dir_inode_operations;
+#endif /* CONFIG_NFS_V3 */
 extern struct file_operations nfs_dir_operations;
 extern struct dentry_operations nfs_dentry_operations;
 
@@ -449,6 +471,15 @@ static inline void nfs_readdata_free(struct nfs_read_data *p)
 
 extern void  nfs_readdata_release(struct rpc_task *task);
 
+/*
+ * linux/fs/nfs3proc.c
+ */
+#ifdef CONFIG_NFS_V3_ACL
+extern struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type);
+extern int nfs3_proc_setacl(struct inode *inode, int type,
+			    struct posix_acl *acl);
+#endif /* CONFIG_NFS_V3_ACL */
+
 /*
  * linux/fs/mount_clnt.c
  * (Used only by nfsroot module)
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index fc51645d61e..3d3a305488c 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -10,6 +10,7 @@
 struct nfs_server {
 	struct rpc_clnt *	client;		/* RPC client handle */
 	struct rpc_clnt *	client_sys;	/* 2nd handle for FSINFO */
+	struct rpc_clnt *	client_acl;	/* ACL RPC client handle */
 	struct nfs_rpc_ops *	rpc_ops;	/* NFS protocol vector */
 	struct backing_dev_info	backing_dev_info;
 	int			flags;		/* various flags */
diff --git a/include/linux/nfs_mount.h b/include/linux/nfs_mount.h
index 0071428231f..659c7543845 100644
--- a/include/linux/nfs_mount.h
+++ b/include/linux/nfs_mount.h
@@ -58,6 +58,7 @@ struct nfs_mount_data {
 #define NFS_MOUNT_KERBEROS	0x0100	/* 3 */
 #define NFS_MOUNT_NONLM		0x0200	/* 3 */
 #define NFS_MOUNT_BROKEN_SUID	0x0400	/* 4 */
+#define NFS_MOUNT_NOACL		0x0800	/* 4 */
 #define NFS_MOUNT_STRICTLOCK	0x1000	/* reserved for NFSv4 */
 #define NFS_MOUNT_SECFLAVOUR	0x2000	/* 5 */
 #define NFS_MOUNT_FLAGMASK	0xFFFF
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 46b206b460c..a2bf6914ff1 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -2,6 +2,7 @@
 #define _LINUX_NFS_XDR_H
 
 #include <linux/sunrpc/xprt.h>
+#include <linux/nfsacl.h>
 
 struct nfs4_fsid {
 	__u64 major;
@@ -368,6 +369,20 @@ struct nfs_readdirargs {
 	struct page **		pages;
 };
 
+struct nfs3_getaclargs {
+	struct nfs_fh *		fh;
+	int			mask;
+	struct page **		pages;
+};
+
+struct nfs3_setaclargs {
+	struct inode *		inode;
+	int			mask;
+	struct posix_acl *	acl_access;
+	struct posix_acl *	acl_default;
+	struct page **		pages;
+};
+
 struct nfs_diropok {
 	struct nfs_fh *		fh;
 	struct nfs_fattr *	fattr;
@@ -491,6 +506,15 @@ struct nfs3_readdirres {
 	int			plus;
 };
 
+struct nfs3_getaclres {
+	struct nfs_fattr *	fattr;
+	int			mask;
+	unsigned int		acl_access_count;
+	unsigned int		acl_default_count;
+	struct posix_acl *	acl_access;
+	struct posix_acl *	acl_default;
+};
+
 #ifdef CONFIG_NFS_V4
 
 typedef u64 clientid4;
@@ -748,4 +772,7 @@ extern struct rpc_version	nfs_version2;
 extern struct rpc_version	nfs_version3;
 extern struct rpc_version	nfs_version4;
 
+extern struct rpc_version	nfsacl_version3;
+extern struct rpc_program	nfsacl_program;
+
 #endif
-- 
cgit v1.2.3-70-g09d2


From 055ffbea0596942579b0dae71d5dab78de8135f6 Mon Sep 17 00:00:00 2001
From: Andreas Gruenbacher <agruen@suse.de>
Date: Wed, 22 Jun 2005 17:16:27 +0000
Subject: [PATCH] NFS: Fix handling of the umask when an NFSv3 default acl is
 present.

 NFSv3 has no concept of a umask on the server side: The client applies
 the umask locally, and sends the effective permissions to the server.
 This behavior is wrong when files are created in a directory that has a
 default ACL.  In this case, the umask is supposed to be ignored, and
 only the default ACL determines the file's effective permissions.

 Usually its the server's task to conditionally apply the umask.  But
 since the server knows nothing about the umask, we have to do it on the
 client side.  This patch tries to fetch the parent directory's default
 ACL before creating a new file, computes the appropriate create mode to
 send to the server, and finally sets the new file's access and default
 acl appropriately.

 Many thanks to Buck Huppmann <buchk@pobox.com> for sending the initial
 version of this patch, as well as for arguing why we need this change.

 Signed-off-by: Andreas Gruenbacher <agruen@suse.de>
 Acked-by: Olaf Kirch <okir@suse.de>
 Signed-off-by: Andrew Morton <akpm@osdl.org>
 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/inode.c         |  5 +++++
 fs/nfs/nfs3acl.c       | 29 +++++++++++++++++++++++++++++
 fs/nfs/nfs3proc.c      | 36 ++++++++++++++++++++++++++++++------
 include/linux/nfs_fs.h |  9 +++++++++
 4 files changed, 73 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 440b9cbb6f8..50a03f1504a 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -490,6 +490,11 @@ nfs_fill_super(struct super_block *sb, struct nfs_mount_data *data, int silent)
 #else
 		server->flags &= ~NFS_MOUNT_NOACL;
 #endif /* CONFIG_NFS_V3_ACL */
+		/*
+		 * The VFS shouldn't apply the umask to mode bits. We will
+		 * do so ourselves when necessary.
+		 */
+		sb->s_flags |= MS_POSIXACL;
 		if (server->namelen == 0 || server->namelen > NFS3_MAXNAMLEN)
 			server->namelen = NFS3_MAXNAMLEN;
 		sb->s_time_gran = 1;
diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c
index 393ba79fc14..89b6468700e 100644
--- a/fs/nfs/nfs3acl.c
+++ b/fs/nfs/nfs3acl.c
@@ -301,3 +301,32 @@ int nfs3_proc_setacl(struct inode *inode, int type, struct posix_acl *acl)
 fail:
 	return PTR_ERR(alloc);
 }
+
+int nfs3_proc_set_default_acl(struct inode *dir, struct inode *inode,
+		mode_t mode)
+{
+	struct posix_acl *dfacl, *acl;
+	int error = 0;
+
+	dfacl = nfs3_proc_getacl(dir, ACL_TYPE_DEFAULT);
+	if (IS_ERR(dfacl)) {
+		error = PTR_ERR(dfacl);
+		return (error == -EOPNOTSUPP) ? 0 : error;
+	}
+	if (!dfacl)
+		return 0;
+	acl = posix_acl_clone(dfacl, GFP_KERNEL);
+	error = -ENOMEM;
+	if (!acl)
+		goto out_release_dfacl;
+	error = posix_acl_create_masq(acl, &mode);
+	if (error < 0)
+		goto out_release_acl;
+	error = nfs3_proc_setacls(inode, acl, S_ISDIR(inode->i_mode) ?
+						      dfacl : NULL);
+out_release_acl:
+	posix_acl_release(acl);
+out_release_dfacl:
+	posix_acl_release(dfacl);
+	return error;
+}
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index d03bac0cc42..a9ddc196224 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -314,7 +314,8 @@ nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
 		.fh		= &fhandle,
 		.fattr		= &fattr
 	};
-	int			status;
+	mode_t mode = sattr->ia_mode;
+	int status;
 
 	dprintk("NFS call  create %s\n", dentry->d_name.name);
 	arg.createmode = NFS3_CREATE_UNCHECKED;
@@ -324,6 +325,8 @@ nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
 		arg.verifier[1] = current->pid;
 	}
 
+	sattr->ia_mode &= ~current->fs->umask;
+
 again:
 	dir_attr.valid = 0;
 	fattr.valid = 0;
@@ -370,6 +373,9 @@ again:
 		nfs_refresh_inode(dentry->d_inode, &fattr);
 		dprintk("NFS reply setattr (post-create): %d\n", status);
 	}
+	if (status != 0)
+		goto out;
+	status = nfs3_proc_set_default_acl(dir, dentry->d_inode, mode);
 out:
 	dprintk("NFS reply create: %d\n", status);
 	return status;
@@ -539,15 +545,24 @@ nfs3_proc_mkdir(struct inode *dir, struct dentry *dentry, struct iattr *sattr)
 		.fh		= &fhandle,
 		.fattr		= &fattr
 	};
-	int			status;
+	int mode = sattr->ia_mode;
+	int status;
 
 	dprintk("NFS call  mkdir %s\n", dentry->d_name.name);
 	dir_attr.valid = 0;
 	fattr.valid = 0;
+
+	sattr->ia_mode &= ~current->fs->umask;
+
 	status = rpc_call(NFS_CLIENT(dir), NFS3PROC_MKDIR, &arg, &res, 0);
 	nfs_refresh_inode(dir, &dir_attr);
-	if (status == 0)
-		status = nfs_instantiate(dentry, &fhandle, &fattr);
+	if (status != 0)
+		goto out;
+	status = nfs_instantiate(dentry, &fhandle, &fattr);
+	if (status != 0)
+		goto out;
+	status = nfs3_proc_set_default_acl(dir, dentry->d_inode, mode);
+out:
 	dprintk("NFS reply mkdir: %d\n", status);
 	return status;
 }
@@ -642,6 +657,7 @@ nfs3_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
 		.fh		= &fh,
 		.fattr		= &fattr
 	};
+	mode_t mode = sattr->ia_mode;
 	int status;
 
 	switch (sattr->ia_mode & S_IFMT) {
@@ -654,12 +670,20 @@ nfs3_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
 
 	dprintk("NFS call  mknod %s %u:%u\n", dentry->d_name.name,
 			MAJOR(rdev), MINOR(rdev));
+
+	sattr->ia_mode &= ~current->fs->umask;
+
 	dir_attr.valid = 0;
 	fattr.valid = 0;
 	status = rpc_call(NFS_CLIENT(dir), NFS3PROC_MKNOD, &arg, &res, 0);
 	nfs_refresh_inode(dir, &dir_attr);
-	if (status == 0)
-		status = nfs_instantiate(dentry, &fh, &fattr);
+	if (status != 0)
+		goto out;
+	status = nfs_instantiate(dentry, &fh, &fattr);
+	if (status != 0)
+		goto out;
+	status = nfs3_proc_set_default_acl(dir, dentry->d_inode, mode);
+out:
 	dprintk("NFS reply mknod: %d\n", status);
 	return status;
 }
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 3a5e442ac77..7662c5131b4 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -478,6 +478,15 @@ extern void  nfs_readdata_release(struct rpc_task *task);
 extern struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type);
 extern int nfs3_proc_setacl(struct inode *inode, int type,
 			    struct posix_acl *acl);
+extern int nfs3_proc_set_default_acl(struct inode *dir, struct inode *inode,
+		mode_t mode);
+#else
+static inline int nfs3_proc_set_default_acl(struct inode *dir,
+					    struct inode *inode,
+					    mode_t mode)
+{
+	return 0;
+}
 #endif /* CONFIG_NFS_V3_ACL */
 
 /*
-- 
cgit v1.2.3-70-g09d2


From 5c6a9f7d92291c832d47e792ed1fafa44acb066e Mon Sep 17 00:00:00 2001
From: Andreas Gruenbacher <agruen@suse.de>
Date: Wed, 22 Jun 2005 17:16:27 +0000
Subject: [PATCH] NFS: Cache the NFSv3 acls.

 Attach acls to inodes in the icache to avoid unnecessary GETACL RPC
 round-trips.  As long as the client doesn't retrieve any acls itself, only the
 default acls of exiting directories and the default and access acls of new
 directories will end up in the cache, which preserves some memory compared to
 always caching the access and default acl of all files.

 Signed-off-by: Andreas Gruenbacher <agruen@suse.de>
 Acked-by: Olaf Kirch <okir@suse.de>
 Signed-off-by: Andrew Morton <akpm@osdl.org>
 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs3acl.c       | 100 +++++++++++++++++++++++++++++++++++++++++--------
 fs/nfs/nfs3proc.c      |   1 +
 include/linux/nfs_fs.h |  11 ++++++
 3 files changed, 97 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c
index 89b6468700e..451112ff9aa 100644
--- a/fs/nfs/nfs3acl.c
+++ b/fs/nfs/nfs3acl.c
@@ -113,6 +113,69 @@ int nfs3_removexattr(struct dentry *dentry, const char *name)
 	return nfs3_proc_setacl(inode, type, NULL);
 }
 
+static void __nfs3_forget_cached_acls(struct nfs_inode *nfsi)
+{
+	if (nfsi->acl_access != ERR_PTR(-EAGAIN)) {
+		posix_acl_release(nfsi->acl_access);
+		nfsi->acl_access = ERR_PTR(-EAGAIN);
+	}
+	if (nfsi->acl_default != ERR_PTR(-EAGAIN)) {
+		posix_acl_release(nfsi->acl_default);
+		nfsi->acl_default = ERR_PTR(-EAGAIN);
+	}
+}
+
+void nfs3_forget_cached_acls(struct inode *inode)
+{
+	dprintk("NFS: nfs3_forget_cached_acls(%s/%ld)\n", inode->i_sb->s_id,
+		inode->i_ino);
+	spin_lock(&inode->i_lock);
+	__nfs3_forget_cached_acls(NFS_I(inode));
+	spin_unlock(&inode->i_lock);
+}
+
+static struct posix_acl *nfs3_get_cached_acl(struct inode *inode, int type)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+	struct posix_acl *acl = ERR_PTR(-EAGAIN);
+
+	spin_lock(&inode->i_lock);
+	switch(type) {
+		case ACL_TYPE_ACCESS:
+			acl = nfsi->acl_access;
+			break;
+
+		case ACL_TYPE_DEFAULT:
+			acl = nfsi->acl_default;
+			break;
+
+		default:
+			return ERR_PTR(-EINVAL);
+	}
+	if (acl == ERR_PTR(-EAGAIN))
+		acl = ERR_PTR(-EAGAIN);
+	else
+		acl = posix_acl_dup(acl);
+	spin_unlock(&inode->i_lock);
+	dprintk("NFS: nfs3_get_cached_acl(%s/%ld, %d) = %p\n", inode->i_sb->s_id,
+		inode->i_ino, type, acl);
+	return acl;
+}
+
+static void nfs3_cache_acls(struct inode *inode, struct posix_acl *acl,
+		    struct posix_acl *dfacl)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+
+	dprintk("nfs3_cache_acls(%s/%ld, %p, %p)\n", inode->i_sb->s_id,
+		inode->i_ino, acl, dfacl);
+	spin_lock(&inode->i_lock);
+	__nfs3_forget_cached_acls(NFS_I(inode));
+	nfsi->acl_access = posix_acl_dup(acl);
+	nfsi->acl_default = posix_acl_dup(dfacl);
+	spin_unlock(&inode->i_lock);
+}
+
 struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type)
 {
 	struct nfs_server *server = NFS_SERVER(inode);
@@ -126,26 +189,32 @@ struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type)
 	struct nfs3_getaclres res = {
 		.fattr =	&fattr,
 	};
-	struct posix_acl *acl = NULL;
+	struct posix_acl *acl;
 	int status, count;
 
 	if (!nfs_server_capable(inode, NFS_CAP_ACLS))
 		return ERR_PTR(-EOPNOTSUPP);
 
-	switch (type) {
-		case ACL_TYPE_ACCESS:
-			args.mask = NFS_ACLCNT|NFS_ACL;
-			break;
-
-		case ACL_TYPE_DEFAULT:
-			if (!S_ISDIR(inode->i_mode))
-				return NULL;
-			args.mask = NFS_DFACLCNT|NFS_DFACL;
-			break;
-
-		default:
-			return ERR_PTR(-EINVAL);
-	}
+	status = nfs_revalidate_inode(server, inode);
+	if (status < 0)
+		return ERR_PTR(status);
+	acl = nfs3_get_cached_acl(inode, type);
+	if (acl != ERR_PTR(-EAGAIN))
+		return acl;
+	acl = NULL;
+
+	/*
+	 * Only get the access acl when explicitly requested: We don't
+	 * need it for access decisions, and only some applications use
+	 * it. Applications which request the access acl first are not
+	 * penalized from this optimization.
+	 */
+	if (type == ACL_TYPE_ACCESS)
+		args.mask |= NFS_ACLCNT|NFS_ACL;
+	if (S_ISDIR(inode->i_mode))
+		args.mask |= NFS_DFACLCNT|NFS_DFACL;
+	if (args.mask == 0)
+		return NULL;
 
 	dprintk("NFS call getacl\n");
 	status = rpc_call(server->client_acl, ACLPROC3_GETACL,
@@ -180,6 +249,7 @@ struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type)
 			res.acl_access = NULL;
 		}
 	}
+	nfs3_cache_acls(inode, res.acl_access, res.acl_default);
 
 	switch(type) {
 		case ACL_TYPE_ACCESS:
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index a9ddc196224..7851569b31c 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -882,4 +882,5 @@ struct nfs_rpc_ops	nfs_v3_clientops = {
 	.file_open	= nfs_open,
 	.file_release	= nfs_release,
 	.lock		= nfs3_proc_lock,
+	.clear_acl_cache = nfs3_forget_cached_acls,
 };
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 7662c5131b4..4ceac9ddac9 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -91,6 +91,8 @@ struct nfs_open_context {
  */
 struct nfs_delegation;
 
+struct posix_acl;
+
 /*
  * nfs fs inode data in memory
  */
@@ -144,6 +146,10 @@ struct nfs_inode {
 	atomic_t		data_updates;
 
 	struct nfs_access_entry	cache_access;
+#ifdef CONFIG_NFS_V3_ACL
+	struct posix_acl	*acl_access;
+	struct posix_acl	*acl_default;
+#endif
 
 	/*
 	 * This is the cookie verifier used for NFSv3 readdir
@@ -480,6 +486,7 @@ extern int nfs3_proc_setacl(struct inode *inode, int type,
 			    struct posix_acl *acl);
 extern int nfs3_proc_set_default_acl(struct inode *dir, struct inode *inode,
 		mode_t mode);
+extern void nfs3_forget_cached_acls(struct inode *inode);
 #else
 static inline int nfs3_proc_set_default_acl(struct inode *dir,
 					    struct inode *inode,
@@ -487,6 +494,10 @@ static inline int nfs3_proc_set_default_acl(struct inode *dir,
 {
 	return 0;
 }
+
+static inline void nfs3_forget_cached_acls(struct inode *inode)
+{
+}
 #endif /* CONFIG_NFS_V3_ACL */
 
 /*
-- 
cgit v1.2.3-70-g09d2


From 00a926422765064cb28e218d4837411c88bf6a3e Mon Sep 17 00:00:00 2001
From: Olivier Galibert <galibert@pobox.com>
Date: Wed, 22 Jun 2005 17:16:29 +0000
Subject: [PATCH] NFS: Hide NFS server-generated readdir cookies from userland

 NFSv3 currently returns the unsigned 64-bit cookie directly to
 userspace. The following patch causes the kernel to generate
 loff_t offsets for the benefit of userland.
 The current server-generated READDIR cookie is cached in the
 nfs_open_context instead of in filp->f_pos, so we still end up work
 correctly under directory insertions/deletion.

 Signed-off-by: Olivier Galibert <galibert@pobox.com>
 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/dir.c           | 114 ++++++++++++++++++++++++++++++++++++++-----------
 fs/nfs/inode.c         |   2 +
 include/linux/nfs_fs.h |   3 ++
 3 files changed, 95 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 2c6a9594568..fceef29c65a 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -141,7 +141,9 @@ typedef struct {
 	struct page	*page;
 	unsigned long	page_index;
 	u32		*ptr;
-	u64		target;
+	u64		target_cookie;
+	int		target_index;
+	int		current_index;
 	struct nfs_entry *entry;
 	decode_dirent_t	decode;
 	int		plus;
@@ -225,14 +227,14 @@ void dir_page_release(nfs_readdir_descriptor_t *desc)
 
 /*
  * Given a pointer to a buffer that has already been filled by a call
- * to readdir, find the next entry.
+ * to readdir, find the next entry with cookie 'desc->target_cookie'.
  *
  * If the end of the buffer has been reached, return -EAGAIN, if not,
  * return the offset within the buffer of the next entry to be
  * read.
  */
 static inline
-int find_dirent(nfs_readdir_descriptor_t *desc, struct page *page)
+int find_dirent(nfs_readdir_descriptor_t *desc)
 {
 	struct nfs_entry *entry = desc->entry;
 	int		loop_count = 0,
@@ -240,7 +242,7 @@ int find_dirent(nfs_readdir_descriptor_t *desc, struct page *page)
 
 	while((status = dir_decode(desc)) == 0) {
 		dfprintk(VFS, "NFS: found cookie %Lu\n", (long long)entry->cookie);
-		if (entry->prev_cookie == desc->target)
+		if (entry->prev_cookie == desc->target_cookie)
 			break;
 		if (loop_count++ > 200) {
 			loop_count = 0;
@@ -252,8 +254,44 @@ int find_dirent(nfs_readdir_descriptor_t *desc, struct page *page)
 }
 
 /*
- * Find the given page, and call find_dirent() in order to try to
- * return the next entry.
+ * Given a pointer to a buffer that has already been filled by a call
+ * to readdir, find the entry at offset 'desc->target_index'.
+ *
+ * If the end of the buffer has been reached, return -EAGAIN, if not,
+ * return the offset within the buffer of the next entry to be
+ * read.
+ */
+static inline
+int find_dirent_index(nfs_readdir_descriptor_t *desc)
+{
+	struct nfs_entry *entry = desc->entry;
+	int		loop_count = 0,
+			status;
+
+	for(;;) {
+		status = dir_decode(desc);
+		if (status)
+			break;
+
+		dfprintk(VFS, "NFS: found cookie %Lu at index %d\n", (long long)entry->cookie, desc->current_index);
+
+		if (desc->target_index == desc->current_index) {
+			desc->target_cookie = entry->cookie;
+			break;
+		}
+		desc->current_index++;
+		if (loop_count++ > 200) {
+			loop_count = 0;
+			schedule();
+		}
+	}
+	dfprintk(VFS, "NFS: find_dirent_index() returns %d\n", status);
+	return status;
+}
+
+/*
+ * Find the given page, and call find_dirent() or find_dirent_index in
+ * order to try to return the next entry.
  */
 static inline
 int find_dirent_page(nfs_readdir_descriptor_t *desc)
@@ -276,7 +314,10 @@ int find_dirent_page(nfs_readdir_descriptor_t *desc)
 	/* NOTE: Someone else may have changed the READDIRPLUS flag */
 	desc->page = page;
 	desc->ptr = kmap(page);		/* matching kunmap in nfs_do_filldir */
-	status = find_dirent(desc, page);
+	if (desc->target_cookie)
+		status = find_dirent(desc);
+	else
+		status = find_dirent_index(desc);
 	if (status < 0)
 		dir_page_release(desc);
  out:
@@ -291,7 +332,8 @@ int find_dirent_page(nfs_readdir_descriptor_t *desc)
  * Recurse through the page cache pages, and return a
  * filled nfs_entry structure of the next directory entry if possible.
  *
- * The target for the search is 'desc->target'.
+ * The target for the search is 'desc->target_cookie' if non-0,
+ * 'desc->target_index' otherwise
  */
 static inline
 int readdir_search_pagecache(nfs_readdir_descriptor_t *desc)
@@ -299,7 +341,19 @@ int readdir_search_pagecache(nfs_readdir_descriptor_t *desc)
 	int		loop_count = 0;
 	int		res;
 
-	dfprintk(VFS, "NFS: readdir_search_pagecache() searching for cookie %Lu\n", (long long)desc->target);
+	if (desc->target_cookie)
+		dfprintk(VFS, "NFS: readdir_search_pagecache() searching for cookie %Lu\n", (long long)desc->target_cookie);
+	else
+		dfprintk(VFS, "NFS: readdir_search_pagecache() searching for cookie number %d\n", desc->target_index);
+
+	/* Always search-by-index from the beginning of the cache */
+	if (!(desc->target_cookie)) {
+		desc->page_index = 0;
+		desc->entry->cookie = desc->entry->prev_cookie = 0;
+		desc->entry->eof = 0;
+		desc->current_index = 0;
+	}
+
 	for (;;) {
 		res = find_dirent_page(desc);
 		if (res != -EAGAIN)
@@ -332,11 +386,12 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent,
 	struct file	*file = desc->file;
 	struct nfs_entry *entry = desc->entry;
 	struct dentry	*dentry = NULL;
+	struct nfs_open_context *ctx = file->private_data;
 	unsigned long	fileid;
 	int		loop_count = 0,
 			res;
 
-	dfprintk(VFS, "NFS: nfs_do_filldir() filling starting @ cookie %Lu\n", (long long)desc->target);
+	dfprintk(VFS, "NFS: nfs_do_filldir() filling starting @ cookie %Lu\n", (long long)entry->cookie);
 
 	for(;;) {
 		unsigned d_type = DT_UNKNOWN;
@@ -356,10 +411,11 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent,
 		}
 
 		res = filldir(dirent, entry->name, entry->len, 
-			      entry->prev_cookie, fileid, d_type);
+			      file->f_pos, fileid, d_type);
 		if (res < 0)
 			break;
-		file->f_pos = desc->target = entry->cookie;
+		file->f_pos++;
+		desc->target_cookie = entry->cookie;
 		if (dir_decode(desc) != 0) {
 			desc->page_index ++;
 			break;
@@ -369,10 +425,12 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent,
 			schedule();
 		}
 	}
+	ctx->dir_pos        = file->f_pos;
+	ctx->dir_cookie     = desc->target_cookie;
 	dir_page_release(desc);
 	if (dentry != NULL)
 		dput(dentry);
-	dfprintk(VFS, "NFS: nfs_do_filldir() filling ended @ cookie %Lu; returning = %d\n", (long long)desc->target, res);
+	dfprintk(VFS, "NFS: nfs_do_filldir() filling ended @ cookie %Lu; returning = %d\n", (long long)desc->target_cookie, res);
 	return res;
 }
 
@@ -398,14 +456,14 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent,
 	struct page	*page = NULL;
 	int		status;
 
-	dfprintk(VFS, "NFS: uncached_readdir() searching for cookie %Lu\n", (long long)desc->target);
+	dfprintk(VFS, "NFS: uncached_readdir() searching for cookie %Lu\n", (long long)desc->target_cookie);
 
 	page = alloc_page(GFP_HIGHUSER);
 	if (!page) {
 		status = -ENOMEM;
 		goto out;
 	}
-	desc->error = NFS_PROTO(inode)->readdir(file->f_dentry, cred, desc->target,
+	desc->error = NFS_PROTO(inode)->readdir(file->f_dentry, cred, desc->target_cookie,
 						page,
 						NFS_SERVER(inode)->dtsize,
 						desc->plus);
@@ -414,7 +472,7 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent,
 	desc->ptr = kmap(page);		/* matching kunmap in nfs_do_filldir */
 	if (desc->error >= 0) {
 		if ((status = dir_decode(desc)) == 0)
-			desc->entry->prev_cookie = desc->target;
+			desc->entry->prev_cookie = desc->target_cookie;
 	} else
 		status = -EIO;
 	if (status < 0)
@@ -435,13 +493,15 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent,
 	goto out;
 }
 
-/* The file offset position is now represented as a true offset into the
- * page cache as is the case in most of the other filesystems.
+/* The file offset position represents the dirent entry number.  A
+   last cookie cache takes care of the common case of reading the
+   whole directory.
  */
 static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 {
 	struct dentry	*dentry = filp->f_dentry;
 	struct inode	*inode = dentry->d_inode;
+	struct nfs_open_context *ctx = filp->private_data;
 	nfs_readdir_descriptor_t my_desc,
 			*desc = &my_desc;
 	struct nfs_entry my_entry;
@@ -458,17 +518,22 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	}
 
 	/*
-	 * filp->f_pos points to the file offset in the page cache.
-	 * but if the cache has meanwhile been zapped, we need to
-	 * read from the last dirent to revalidate f_pos
-	 * itself.
+	 * filp->f_pos points to the dirent entry number.
+	 * ctx->dir_pos has the number of the cached cookie.  We have
+	 * to either find the entry with the appropriate number or
+	 * revalidate the cookie.
 	 */
 	memset(desc, 0, sizeof(*desc));
 
 	desc->file = filp;
-	desc->target = filp->f_pos;
 	desc->decode = NFS_PROTO(inode)->decode_dirent;
 	desc->plus = NFS_USE_READDIRPLUS(inode);
+	desc->target_index = filp->f_pos;
+
+	if (filp->f_pos == ctx->dir_pos)
+		desc->target_cookie = ctx->dir_cookie;
+	else
+		desc->target_cookie = 0;
 
 	my_entry.cookie = my_entry.prev_cookie = 0;
 	my_entry.eof = 0;
@@ -478,9 +543,10 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 
 	while(!desc->entry->eof) {
 		res = readdir_search_pagecache(desc);
+
 		if (res == -EBADCOOKIE) {
 			/* This means either end of directory */
-			if (desc->entry->cookie != desc->target) {
+			if (desc->target_cookie && desc->entry->cookie != desc->target_cookie) {
 				/* Or that the server has 'lost' a cookie */
 				res = uncached_readdir(desc, dirent, filldir);
 				if (res >= 0)
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 8a8d57d9d66..9fa02e7984a 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -891,6 +891,8 @@ struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, struct rp
 		ctx->state = NULL;
 		ctx->lockowner = current->files;
 		ctx->error = 0;
+		ctx->dir_pos = 0;
+		ctx->dir_cookie = 0;
 	}
 	return ctx;
 }
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 4ceac9ddac9..f810195ef7a 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -84,6 +84,9 @@ struct nfs_open_context {
 	int error;
 
 	struct list_head list;
+
+	int dir_pos;		/* Directory cookie cache */
+	__u64 dir_cookie;
 };
 
 /*
-- 
cgit v1.2.3-70-g09d2


From f0dd2136da6d2070e12bfa6d199b136318e666c7 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 22 Jun 2005 17:16:29 +0000
Subject: [PATCH] NFS: Clean up readdir changes.

 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/dir.c           | 85 ++++++++++++++++++++++++++++----------------------
 fs/nfs/inode.c         |  1 -
 include/linux/nfs_fs.h |  1 -
 3 files changed, 48 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index fceef29c65a..b38a57e78a6 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -51,8 +51,10 @@ static int nfs_mknod(struct inode *, struct dentry *, int, dev_t);
 static int nfs_rename(struct inode *, struct dentry *,
 		      struct inode *, struct dentry *);
 static int nfs_fsync_dir(struct file *, struct dentry *, int);
+static loff_t nfs_llseek_dir(struct file *, loff_t, int);
 
 struct file_operations nfs_dir_operations = {
+	.llseek		= nfs_llseek_dir,
 	.read		= generic_read_dir,
 	.readdir	= nfs_readdir,
 	.open		= nfs_opendir,
@@ -141,9 +143,8 @@ typedef struct {
 	struct page	*page;
 	unsigned long	page_index;
 	u32		*ptr;
-	u64		target_cookie;
-	int		target_index;
-	int		current_index;
+	u64		*dir_cookie;
+	loff_t		current_index;
 	struct nfs_entry *entry;
 	decode_dirent_t	decode;
 	int		plus;
@@ -227,7 +228,7 @@ void dir_page_release(nfs_readdir_descriptor_t *desc)
 
 /*
  * Given a pointer to a buffer that has already been filled by a call
- * to readdir, find the next entry with cookie 'desc->target_cookie'.
+ * to readdir, find the next entry with cookie '*desc->dir_cookie'.
  *
  * If the end of the buffer has been reached, return -EAGAIN, if not,
  * return the offset within the buffer of the next entry to be
@@ -241,8 +242,8 @@ int find_dirent(nfs_readdir_descriptor_t *desc)
 			status;
 
 	while((status = dir_decode(desc)) == 0) {
-		dfprintk(VFS, "NFS: found cookie %Lu\n", (long long)entry->cookie);
-		if (entry->prev_cookie == desc->target_cookie)
+		dfprintk(VFS, "NFS: found cookie %Lu\n", (unsigned long long)entry->cookie);
+		if (entry->prev_cookie == *desc->dir_cookie)
 			break;
 		if (loop_count++ > 200) {
 			loop_count = 0;
@@ -255,7 +256,7 @@ int find_dirent(nfs_readdir_descriptor_t *desc)
 
 /*
  * Given a pointer to a buffer that has already been filled by a call
- * to readdir, find the entry at offset 'desc->target_index'.
+ * to readdir, find the entry at offset 'desc->file->f_pos'.
  *
  * If the end of the buffer has been reached, return -EAGAIN, if not,
  * return the offset within the buffer of the next entry to be
@@ -273,10 +274,10 @@ int find_dirent_index(nfs_readdir_descriptor_t *desc)
 		if (status)
 			break;
 
-		dfprintk(VFS, "NFS: found cookie %Lu at index %d\n", (long long)entry->cookie, desc->current_index);
+		dfprintk(VFS, "NFS: found cookie %Lu at index %Ld\n", (unsigned long long)entry->cookie, desc->current_index);
 
-		if (desc->target_index == desc->current_index) {
-			desc->target_cookie = entry->cookie;
+		if (desc->file->f_pos == desc->current_index) {
+			*desc->dir_cookie = entry->cookie;
 			break;
 		}
 		desc->current_index++;
@@ -314,7 +315,7 @@ int find_dirent_page(nfs_readdir_descriptor_t *desc)
 	/* NOTE: Someone else may have changed the READDIRPLUS flag */
 	desc->page = page;
 	desc->ptr = kmap(page);		/* matching kunmap in nfs_do_filldir */
-	if (desc->target_cookie)
+	if (*desc->dir_cookie != 0)
 		status = find_dirent(desc);
 	else
 		status = find_dirent_index(desc);
@@ -332,8 +333,8 @@ int find_dirent_page(nfs_readdir_descriptor_t *desc)
  * Recurse through the page cache pages, and return a
  * filled nfs_entry structure of the next directory entry if possible.
  *
- * The target for the search is 'desc->target_cookie' if non-0,
- * 'desc->target_index' otherwise
+ * The target for the search is '*desc->dir_cookie' if non-0,
+ * 'desc->file->f_pos' otherwise
  */
 static inline
 int readdir_search_pagecache(nfs_readdir_descriptor_t *desc)
@@ -341,18 +342,15 @@ int readdir_search_pagecache(nfs_readdir_descriptor_t *desc)
 	int		loop_count = 0;
 	int		res;
 
-	if (desc->target_cookie)
-		dfprintk(VFS, "NFS: readdir_search_pagecache() searching for cookie %Lu\n", (long long)desc->target_cookie);
-	else
-		dfprintk(VFS, "NFS: readdir_search_pagecache() searching for cookie number %d\n", desc->target_index);
-
 	/* Always search-by-index from the beginning of the cache */
-	if (!(desc->target_cookie)) {
+	if (*desc->dir_cookie == 0) {
+		dfprintk(VFS, "NFS: readdir_search_pagecache() searching for offset %Ld\n", (long long)desc->file->f_pos);
 		desc->page_index = 0;
 		desc->entry->cookie = desc->entry->prev_cookie = 0;
 		desc->entry->eof = 0;
 		desc->current_index = 0;
-	}
+	} else
+		dfprintk(VFS, "NFS: readdir_search_pagecache() searching for cookie %Lu\n", (unsigned long long)*desc->dir_cookie);
 
 	for (;;) {
 		res = find_dirent_page(desc);
@@ -386,7 +384,6 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent,
 	struct file	*file = desc->file;
 	struct nfs_entry *entry = desc->entry;
 	struct dentry	*dentry = NULL;
-	struct nfs_open_context *ctx = file->private_data;
 	unsigned long	fileid;
 	int		loop_count = 0,
 			res;
@@ -415,7 +412,7 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent,
 		if (res < 0)
 			break;
 		file->f_pos++;
-		desc->target_cookie = entry->cookie;
+		*desc->dir_cookie = entry->cookie;
 		if (dir_decode(desc) != 0) {
 			desc->page_index ++;
 			break;
@@ -425,12 +422,10 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent,
 			schedule();
 		}
 	}
-	ctx->dir_pos        = file->f_pos;
-	ctx->dir_cookie     = desc->target_cookie;
 	dir_page_release(desc);
 	if (dentry != NULL)
 		dput(dentry);
-	dfprintk(VFS, "NFS: nfs_do_filldir() filling ended @ cookie %Lu; returning = %d\n", (long long)desc->target_cookie, res);
+	dfprintk(VFS, "NFS: nfs_do_filldir() filling ended @ cookie %Lu; returning = %d\n", (unsigned long long)*desc->dir_cookie, res);
 	return res;
 }
 
@@ -456,14 +451,14 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent,
 	struct page	*page = NULL;
 	int		status;
 
-	dfprintk(VFS, "NFS: uncached_readdir() searching for cookie %Lu\n", (long long)desc->target_cookie);
+	dfprintk(VFS, "NFS: uncached_readdir() searching for cookie %Lu\n", (unsigned long long)*desc->dir_cookie);
 
 	page = alloc_page(GFP_HIGHUSER);
 	if (!page) {
 		status = -ENOMEM;
 		goto out;
 	}
-	desc->error = NFS_PROTO(inode)->readdir(file->f_dentry, cred, desc->target_cookie,
+	desc->error = NFS_PROTO(inode)->readdir(file->f_dentry, cred, *desc->dir_cookie,
 						page,
 						NFS_SERVER(inode)->dtsize,
 						desc->plus);
@@ -472,7 +467,7 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent,
 	desc->ptr = kmap(page);		/* matching kunmap in nfs_do_filldir */
 	if (desc->error >= 0) {
 		if ((status = dir_decode(desc)) == 0)
-			desc->entry->prev_cookie = desc->target_cookie;
+			desc->entry->prev_cookie = *desc->dir_cookie;
 	} else
 		status = -EIO;
 	if (status < 0)
@@ -501,7 +496,6 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 {
 	struct dentry	*dentry = filp->f_dentry;
 	struct inode	*inode = dentry->d_inode;
-	struct nfs_open_context *ctx = filp->private_data;
 	nfs_readdir_descriptor_t my_desc,
 			*desc = &my_desc;
 	struct nfs_entry my_entry;
@@ -519,21 +513,16 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 
 	/*
 	 * filp->f_pos points to the dirent entry number.
-	 * ctx->dir_pos has the number of the cached cookie.  We have
+	 * *desc->dir_cookie has the cookie for the next entry. We have
 	 * to either find the entry with the appropriate number or
 	 * revalidate the cookie.
 	 */
 	memset(desc, 0, sizeof(*desc));
 
 	desc->file = filp;
+	desc->dir_cookie = &((struct nfs_open_context *)filp->private_data)->dir_cookie;
 	desc->decode = NFS_PROTO(inode)->decode_dirent;
 	desc->plus = NFS_USE_READDIRPLUS(inode);
-	desc->target_index = filp->f_pos;
-
-	if (filp->f_pos == ctx->dir_pos)
-		desc->target_cookie = ctx->dir_cookie;
-	else
-		desc->target_cookie = 0;
 
 	my_entry.cookie = my_entry.prev_cookie = 0;
 	my_entry.eof = 0;
@@ -546,7 +535,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 
 		if (res == -EBADCOOKIE) {
 			/* This means either end of directory */
-			if (desc->target_cookie && desc->entry->cookie != desc->target_cookie) {
+			if (*desc->dir_cookie && desc->entry->cookie != *desc->dir_cookie) {
 				/* Or that the server has 'lost' a cookie */
 				res = uncached_readdir(desc, dirent, filldir);
 				if (res >= 0)
@@ -579,6 +568,28 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	return 0;
 }
 
+loff_t nfs_llseek_dir(struct file *filp, loff_t offset, int origin)
+{
+	down(&filp->f_dentry->d_inode->i_sem);
+	switch (origin) {
+		case 1:
+			offset += filp->f_pos;
+		case 0:
+			if (offset >= 0)
+				break;
+		default:
+			offset = -EINVAL;
+			goto out;
+	}
+	if (offset != filp->f_pos) {
+		filp->f_pos = offset;
+		((struct nfs_open_context *)filp->private_data)->dir_cookie = 0;
+	}
+out:
+	up(&filp->f_dentry->d_inode->i_sem);
+	return offset;
+}
+
 /*
  * All directory operations under NFS are synchronous, so fsync()
  * is a dummy operation.
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 9fa02e7984a..6300e05e946 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -891,7 +891,6 @@ struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, struct rp
 		ctx->state = NULL;
 		ctx->lockowner = current->files;
 		ctx->error = 0;
-		ctx->dir_pos = 0;
 		ctx->dir_cookie = 0;
 	}
 	return ctx;
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index f810195ef7a..c90313bfa43 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -85,7 +85,6 @@ struct nfs_open_context {
 
 	struct list_head list;
 
-	int dir_pos;		/* Directory cookie cache */
 	__u64 dir_cookie;
 };
 
-- 
cgit v1.2.3-70-g09d2


From 951a143b3fcf15cfa9d38250b7462f821db241db Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 22 Jun 2005 17:16:30 +0000
Subject: [PATCH] NFS: Fix the file size revalidation

 Instead of looking at whether or not the file is open for writes before
 we accept to update the length using the server value, we should rather
 be looking at whether or not we are currently caching any writes.

 Failure to do so means in particular that we're not updating the file
 length correctly after obtaining a POSIX or BSD lock.

 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/direct.c        |  2 +-
 fs/nfs/inode.c         | 69 +++++++++++++-------------------------------------
 fs/nfs/write.c         |  4 +--
 include/linux/nfs_fs.h |  1 -
 4 files changed, 21 insertions(+), 55 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 68df803f27c..d6a30c844de 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -517,7 +517,7 @@ retry:
 	result = tot_bytes;
 
 out:
-	nfs_end_data_update_defer(inode);
+	nfs_end_data_update(inode);
 	nfs_writedata_free(wdata);
 	return result;
 
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 6300e05e946..b2d16758ced 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -1147,27 +1147,6 @@ void nfs_end_data_update(struct inode *inode)
 	atomic_dec(&nfsi->data_updates);
 }
 
-/**
- * nfs_end_data_update_defer
- * @inode - pointer to inode
- * Declare end of the operations that will update file data
- * This will defer marking the inode as needing revalidation
- * unless there are no other pending updates.
- */
-void nfs_end_data_update_defer(struct inode *inode)
-{
-	struct nfs_inode *nfsi = NFS_I(inode);
-
-	if (atomic_dec_and_test(&nfsi->data_updates)) {
-		/* Mark the attribute cache for revalidation */
-		nfsi->flags |= NFS_INO_INVALID_ATTR;
-		/* Directories and symlinks: invalidate page cache too */
-		if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))
-			nfsi->flags |= NFS_INO_INVALID_DATA;
-		nfsi->cache_change_attribute ++;
-	}
-}
-
 /**
  * nfs_refresh_inode - verify consistency of the inode attribute cache
  * @inode - pointer to inode
@@ -1222,8 +1201,8 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr)
 		if (!timespec_equal(&inode->i_mtime, &fattr->mtime)
 				|| cur_size != new_isize)
 			nfsi->flags |= NFS_INO_INVALID_ATTR;
-	} else if (S_ISREG(inode->i_mode) && new_isize > cur_size)
-			nfsi->flags |= NFS_INO_INVALID_ATTR;
+	} else if (new_isize != cur_size && nfsi->npages == 0)
+		nfsi->flags |= NFS_INO_INVALID_ATTR;
 
 	/* Have any file permissions changed? */
 	if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO)
@@ -1257,10 +1236,8 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr)
 static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsigned long verifier)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
-	__u64		new_size;
-	loff_t		new_isize;
+	loff_t cur_isize, new_isize;
 	unsigned int	invalid = 0;
-	loff_t		cur_isize;
 	int data_unstable;
 
 	dfprintk(VFS, "NFS: %s(%s/%ld ct=%d info=0x%x)\n",
@@ -1293,49 +1270,39 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign
 	/* Are we racing with known updates of the metadata on the server? */
 	data_unstable = ! nfs_verify_change_attribute(inode, verifier);
 
-	/* Check if the file size agrees */
-	new_size = fattr->size;
+	/* Check if our cached file size is stale */
  	new_isize = nfs_size_to_loff_t(fattr->size);
 	cur_isize = i_size_read(inode);
-	if (cur_isize != new_size) {
-#ifdef NFS_DEBUG_VERBOSE
-		printk(KERN_DEBUG "NFS: isize change on %s/%ld\n", inode->i_sb->s_id, inode->i_ino);
-#endif
-		/*
-		 * If we have pending writebacks, things can get
-		 * messy.
-		 */
-		if (S_ISREG(inode->i_mode) && data_unstable) {
-			if (new_isize > cur_isize) {
+	if (new_isize != cur_isize) {
+		/* Do we perhaps have any outstanding writes? */
+		if (nfsi->npages == 0) {
+			/* No, but did we race with nfs_end_data_update()? */
+			if (verifier  ==  nfsi->cache_change_attribute) {
 				inode->i_size = new_isize;
-				invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA;
+				invalid |= NFS_INO_INVALID_DATA;
 			}
-		} else {
+			invalid |= NFS_INO_INVALID_ATTR;
+		} else if (new_isize > cur_isize) {
 			inode->i_size = new_isize;
 			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA;
 		}
+		dprintk("NFS: isize change on server for file %s/%ld\n",
+				inode->i_sb->s_id, inode->i_ino);
 	}
 
-	/*
-	 * Note: we don't check inode->i_mtime since pipes etc.
-	 *       can change this value in VFS without requiring a
-	 *	 cache revalidation.
-	 */
+	/* Check if the mtime agrees */
 	if (!timespec_equal(&inode->i_mtime, &fattr->mtime)) {
 		memcpy(&inode->i_mtime, &fattr->mtime, sizeof(inode->i_mtime));
-#ifdef NFS_DEBUG_VERBOSE
-		printk(KERN_DEBUG "NFS: mtime change on %s/%ld\n", inode->i_sb->s_id, inode->i_ino);
-#endif
+		dprintk("NFS: mtime change on server for file %s/%ld\n",
+				inode->i_sb->s_id, inode->i_ino);
 		if (!data_unstable)
 			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA;
 	}
 
 	if ((fattr->valid & NFS_ATTR_FATTR_V4)
 	    && nfsi->change_attr != fattr->change_attr) {
-#ifdef NFS_DEBUG_VERBOSE
-		printk(KERN_DEBUG "NFS: change_attr change on %s/%ld\n",
+		dprintk("NFS: change_attr change on server for file %s/%ld\n",
 		       inode->i_sb->s_id, inode->i_ino);
-#endif
 		nfsi->change_attr = fattr->change_attr;
 		if (!data_unstable)
 			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 6f7a4af3bc4..c574d551f02 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -220,7 +220,7 @@ static int nfs_writepage_sync(struct nfs_open_context *ctx, struct inode *inode,
 		ClearPageError(page);
 
 io_error:
-	nfs_end_data_update_defer(inode);
+	nfs_end_data_update(inode);
 	nfs_writedata_free(wdata);
 	return written ? written : result;
 }
@@ -401,7 +401,7 @@ static void nfs_inode_remove_request(struct nfs_page *req)
 	nfsi->npages--;
 	if (!nfsi->npages) {
 		spin_unlock(&nfsi->req_lock);
-		nfs_end_data_update_defer(inode);
+		nfs_end_data_update(inode);
 		iput(inode);
 	} else
 		spin_unlock(&nfsi->req_lock);
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index c90313bfa43..211266c56ce 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -294,7 +294,6 @@ extern void nfs_begin_attr_update(struct inode *);
 extern void nfs_end_attr_update(struct inode *);
 extern void nfs_begin_data_update(struct inode *);
 extern void nfs_end_data_update(struct inode *);
-extern void nfs_end_data_update_defer(struct inode *);
 extern struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, struct rpc_cred *cred);
 extern struct nfs_open_context *get_nfs_open_context(struct nfs_open_context *ctx);
 extern void put_nfs_open_context(struct nfs_open_context *ctx);
-- 
cgit v1.2.3-70-g09d2


From 7d52e86274e09fce8ac8f963e3605a84d0a305a7 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 22 Jun 2005 17:16:30 +0000
Subject: [PATCH] NFS: Cleanup of caching code, and slight optimization of
 writes.

 Unless we're doing O_APPEND writes, we really don't care about revalidating
 the file length. Just make sure that we catch any page cache invalidations.

 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/file.c          | 12 +++++++++---
 fs/nfs/inode.c         | 44 +++++++++++++++++++++++++++++---------------
 include/linux/nfs_fs.h |  1 +
 3 files changed, 39 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index a606708264e..40436857ed4 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -333,9 +333,15 @@ nfs_file_write(struct kiocb *iocb, const char __user *buf, size_t count, loff_t
 	result = -EBUSY;
 	if (IS_SWAPFILE(inode))
 		goto out_swapfile;
-	result = nfs_revalidate_inode(NFS_SERVER(inode), inode);
-	if (result)
-		goto out;
+	/*
+	 * O_APPEND implies that we must revalidate the file length.
+	 */
+	if (iocb->ki_filp->f_flags & O_APPEND) {
+		result = nfs_revalidate_file_size(inode, iocb->ki_filp);
+		if (result)
+			goto out;
+	} else
+		nfs_revalidate_mapping(inode, iocb->ki_filp->f_mapping);
 
 	result = count;
 	if (!count)
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index b2d16758ced..a3922f4cc0a 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -1062,21 +1062,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 	if (verifier == nfsi->cache_change_attribute)
 		nfsi->flags &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME);
 	/* Do the page cache invalidation */
-	if (flags & NFS_INO_INVALID_DATA) {
-		if (S_ISREG(inode->i_mode)) {
-			if (filemap_fdatawrite(inode->i_mapping) == 0)
-				filemap_fdatawait(inode->i_mapping);
-			nfs_wb_all(inode);
-		}
-		nfsi->flags &= ~NFS_INO_INVALID_DATA;
-		invalidate_inode_pages2(inode->i_mapping);
-		memset(NFS_COOKIEVERF(inode), 0, sizeof(NFS_COOKIEVERF(inode)));
-		dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n",
-				inode->i_sb->s_id,
-				(long long)NFS_FILEID(inode));
-		/* This ensures we revalidate dentries */
-		nfsi->cache_change_attribute++;
-	}
+	nfs_revalidate_mapping(inode, inode->i_mapping);
 	if (flags & NFS_INO_INVALID_ACL)
 		nfs_zap_acl_cache(inode);
 	dfprintk(PAGECACHE, "NFS: (%s/%Ld) revalidation complete\n",
@@ -1115,6 +1101,34 @@ int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 	return __nfs_revalidate_inode(server, inode);
 }
 
+/**
+ * nfs_revalidate_mapping - Revalidate the pagecache
+ * @inode - pointer to host inode
+ * @mapping - pointer to mapping
+ */
+void nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+
+	if (nfsi->flags & NFS_INO_INVALID_DATA) {
+		if (S_ISREG(inode->i_mode)) {
+			if (filemap_fdatawrite(mapping) == 0)
+				filemap_fdatawait(mapping);
+			nfs_wb_all(inode);
+		}
+		invalidate_inode_pages2(mapping);
+		nfsi->flags &= ~NFS_INO_INVALID_DATA;
+		if (S_ISDIR(inode->i_mode)) {
+			memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf));
+			/* This ensures we revalidate child dentries */
+			nfsi->cache_change_attribute++;
+		}
+		dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n",
+				inode->i_sb->s_id,
+				(long long)NFS_FILEID(inode));
+	}
+}
+
 /**
  * nfs_begin_data_update
  * @inode - pointer to inode
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 211266c56ce..443103c13e5 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -289,6 +289,7 @@ extern int nfs_release(struct inode *, struct file *);
 extern int nfs_attribute_timeout(struct inode *inode);
 extern int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode);
 extern int __nfs_revalidate_inode(struct nfs_server *, struct inode *);
+extern void nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping);
 extern int nfs_setattr(struct dentry *, struct iattr *);
 extern void nfs_begin_attr_update(struct inode *);
 extern void nfs_end_attr_update(struct inode *);
-- 
cgit v1.2.3-70-g09d2


From fe51beecc55d0b0dce289e4758e7c529a642f63e Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 22 Jun 2005 17:16:30 +0000
Subject: [PATCH] NFS: Ensure that fstat() always returns the correct mtime

 Even if the file is open for writes.

 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/file.c          | 28 ++++++++++++++++++++++------
 fs/nfs/inode.c         | 24 ++++++++++++++++--------
 include/linux/nfs_fs.h |  1 +
 3 files changed, 39 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 40436857ed4..5621ba9885f 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -127,6 +127,21 @@ nfs_file_release(struct inode *inode, struct file *filp)
 	return NFS_PROTO(inode)->file_release(inode, filp);
 }
 
+/**
+ * nfs_revalidate_file - Revalidate the page cache & related metadata
+ * @inode - pointer to inode struct
+ * @file - pointer to file
+ */
+static int nfs_revalidate_file(struct inode *inode, struct file *filp)
+{
+	int retval = 0;
+
+	if ((NFS_FLAGS(inode) & NFS_INO_REVAL_PAGECACHE) || nfs_attribute_timeout(inode))
+		retval = __nfs_revalidate_inode(NFS_SERVER(inode), inode);
+	nfs_revalidate_mapping(inode, filp->f_mapping);
+	return 0;
+}
+
 /**
  * nfs_revalidate_size - Revalidate the file size
  * @inode - pointer to inode struct
@@ -149,7 +164,8 @@ static int nfs_revalidate_file_size(struct inode *inode, struct file *filp)
 		goto force_reval;
 	if (nfsi->npages != 0)
 		return 0;
-	return nfs_revalidate_inode(server, inode);
+	if (!(NFS_FLAGS(inode) & NFS_INO_REVAL_PAGECACHE) && !nfs_attribute_timeout(inode))
+		return 0;
 force_reval:
 	return __nfs_revalidate_inode(server, inode);
 }
@@ -210,7 +226,7 @@ nfs_file_read(struct kiocb *iocb, char __user * buf, size_t count, loff_t pos)
 		dentry->d_parent->d_name.name, dentry->d_name.name,
 		(unsigned long) count, (unsigned long) pos);
 
-	result = nfs_revalidate_inode(NFS_SERVER(inode), inode);
+	result = nfs_revalidate_file(inode, iocb->ki_filp);
 	if (!result)
 		result = generic_file_aio_read(iocb, buf, count, pos);
 	return result;
@@ -228,7 +244,7 @@ nfs_file_sendfile(struct file *filp, loff_t *ppos, size_t count,
 		dentry->d_parent->d_name.name, dentry->d_name.name,
 		(unsigned long) count, (unsigned long long) *ppos);
 
-	res = nfs_revalidate_inode(NFS_SERVER(inode), inode);
+	res = nfs_revalidate_file(inode, filp);
 	if (!res)
 		res = generic_file_sendfile(filp, ppos, count, actor, target);
 	return res;
@@ -244,7 +260,7 @@ nfs_file_mmap(struct file * file, struct vm_area_struct * vma)
 	dfprintk(VFS, "nfs: mmap(%s/%s)\n",
 		dentry->d_parent->d_name.name, dentry->d_name.name);
 
-	status = nfs_revalidate_inode(NFS_SERVER(inode), inode);
+	status = nfs_revalidate_file(inode, file);
 	if (!status)
 		status = generic_file_mmap(file, vma);
 	return status;
@@ -340,8 +356,8 @@ nfs_file_write(struct kiocb *iocb, const char __user *buf, size_t count, loff_t
 		result = nfs_revalidate_file_size(inode, iocb->ki_filp);
 		if (result)
 			goto out;
-	} else
-		nfs_revalidate_mapping(inode, iocb->ki_filp->f_mapping);
+	}
+	nfs_revalidate_mapping(inode, iocb->ki_filp->f_mapping);
 
 	result = count;
 	if (!count)
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index a3922f4cc0a..4f545f382ba 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -620,9 +620,9 @@ nfs_zap_caches(struct inode *inode)
 
 	memset(NFS_COOKIEVERF(inode), 0, sizeof(NFS_COOKIEVERF(inode)));
 	if (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode))
-		nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
+		nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL|NFS_INO_REVAL_PAGECACHE;
 	else
-		nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
+		nfsi->flags |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL|NFS_INO_REVAL_PAGECACHE;
 }
 
 static void nfs_zap_acl_cache(struct inode *inode)
@@ -1055,6 +1055,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 		goto out;
 	}
 	flags = nfsi->flags;
+	nfsi->flags &= ~NFS_INO_REVAL_PAGECACHE;
 	/*
 	 * We may need to keep the attributes marked as invalid if
 	 * we raced with nfs_end_attr_update().
@@ -1187,8 +1188,11 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr)
 		if ((fattr->valid & NFS_ATTR_PRE_CHANGE) != 0
 				&& nfsi->change_attr == fattr->pre_change_attr)
 			nfsi->change_attr = fattr->change_attr;
-		if (!data_unstable && nfsi->change_attr != fattr->change_attr)
+		if (nfsi->change_attr != fattr->change_attr) {
 			nfsi->flags |= NFS_INO_INVALID_ATTR;
+			if (!data_unstable)
+				nfsi->flags |= NFS_INO_REVAL_PAGECACHE;
+		}
 	}
 
 	if ((fattr->valid & NFS_ATTR_FATTR) == 0)
@@ -1211,12 +1215,16 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr)
 	}
 
 	/* Verify a few of the more important attributes */
-	if (!data_unstable) {
-		if (!timespec_equal(&inode->i_mtime, &fattr->mtime)
-				|| cur_size != new_isize)
-			nfsi->flags |= NFS_INO_INVALID_ATTR;
-	} else if (new_isize != cur_size && nfsi->npages == 0)
+	if (!timespec_equal(&inode->i_mtime, &fattr->mtime)) {
 		nfsi->flags |= NFS_INO_INVALID_ATTR;
+		if (!data_unstable)
+			nfsi->flags |= NFS_INO_REVAL_PAGECACHE;
+	}
+	if (cur_size != new_isize) {
+		nfsi->flags |= NFS_INO_INVALID_ATTR;
+		if (nfsi->npages == 0)
+			nfsi->flags |= NFS_INO_REVAL_PAGECACHE;
+	}
 
 	/* Have any file permissions changed? */
 	if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO)
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 443103c13e5..2954e44ed49 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -198,6 +198,7 @@ struct nfs_inode {
 #define NFS_INO_INVALID_ATIME	0x0020		/* cached atime is invalid */
 #define NFS_INO_INVALID_ACCESS	0x0040		/* cached access cred invalid */
 #define NFS_INO_INVALID_ACL	0x0080		/* cached acls are invalid */
+#define NFS_INO_REVAL_PAGECACHE	0x1000		/* must revalidate pagecache */
 
 static inline struct nfs_inode *NFS_I(struct inode *inode)
 {
-- 
cgit v1.2.3-70-g09d2


From c6a556b88adfacd2af90be84357c8165d716c27d Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 22 Jun 2005 17:16:30 +0000
Subject: [PATCH] NFS: Make searching and waiting on busy writeback requests
 more efficient.

 Basically copies the VFS's method for tracking writebacks and applies
 it to the struct nfs_page.

 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/pagelist.c        | 29 ++++++++++++++++++++++++++++-
 fs/nfs/read.c            |  3 ---
 fs/nfs/write.c           | 19 +++++++++----------
 include/linux/nfs_page.h | 12 ++++++++----
 4 files changed, 45 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c
index 80777f99a58..356a33bb38a 100644
--- a/fs/nfs/pagelist.c
+++ b/fs/nfs/pagelist.c
@@ -111,6 +111,33 @@ void nfs_unlock_request(struct nfs_page *req)
 	nfs_release_request(req);
 }
 
+/**
+ * nfs_set_page_writeback_locked - Lock a request for writeback
+ * @req:
+ */
+int nfs_set_page_writeback_locked(struct nfs_page *req)
+{
+	struct nfs_inode *nfsi = NFS_I(req->wb_context->dentry->d_inode);
+
+	if (!nfs_lock_request(req))
+		return 0;
+	radix_tree_tag_set(&nfsi->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_WRITEBACK);
+	return 1;
+}
+
+/**
+ * nfs_clear_page_writeback - Unlock request and wake up sleepers
+ */
+void nfs_clear_page_writeback(struct nfs_page *req)
+{
+	struct nfs_inode *nfsi = NFS_I(req->wb_context->dentry->d_inode);
+
+	spin_lock(&nfsi->req_lock);
+	radix_tree_tag_clear(&nfsi->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_WRITEBACK);
+	spin_unlock(&nfsi->req_lock);
+	nfs_unlock_request(req);
+}
+
 /**
  * nfs_clear_request - Free up all resources allocated to the request
  * @req:
@@ -301,7 +328,7 @@ nfs_scan_list(struct list_head *head, struct list_head *dst,
 		if (req->wb_index > idx_end)
 			break;
 
-		if (!nfs_lock_request(req))
+		if (!nfs_set_page_writeback_locked(req))
 			continue;
 		nfs_list_remove_request(req);
 		nfs_list_add_request(req, dst);
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index a0042fb5863..6f866b8aa2d 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -173,7 +173,6 @@ static int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode,
 	if (len < PAGE_CACHE_SIZE)
 		memclear_highpage_flush(page, len, PAGE_CACHE_SIZE - len);
 
-	nfs_lock_request(new);
 	nfs_list_add_request(new, &one_request);
 	nfs_pagein_one(&one_request, inode);
 	return 0;
@@ -185,7 +184,6 @@ static void nfs_readpage_release(struct nfs_page *req)
 
 	nfs_clear_request(req);
 	nfs_release_request(req);
-	nfs_unlock_request(req);
 
 	dprintk("NFS: read done (%s/%Ld %d@%Ld)\n",
 			req->wb_context->dentry->d_inode->i_sb->s_id,
@@ -553,7 +551,6 @@ readpage_async_filler(void *data, struct page *page)
 	}
 	if (len < PAGE_CACHE_SIZE)
 		memclear_highpage_flush(page, len, PAGE_CACHE_SIZE - len);
-	nfs_lock_request(new);
 	nfs_list_add_request(new, desc->head);
 	return 0;
 }
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 79b621a545b..58a39b0486a 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -503,13 +503,12 @@ nfs_wait_on_requests(struct inode *inode, unsigned long idx_start, unsigned int
 
 	spin_lock(&nfsi->req_lock);
 	next = idx_start;
-	while (radix_tree_gang_lookup(&nfsi->nfs_page_tree, (void **)&req, next, 1)) {
+	while (radix_tree_gang_lookup_tag(&nfsi->nfs_page_tree, (void **)&req, next, 1, NFS_PAGE_TAG_WRITEBACK)) {
 		if (req->wb_index > idx_end)
 			break;
 
 		next = req->wb_index + 1;
-		if (!NFS_WBACK_BUSY(req))
-			continue;
+		BUG_ON(!NFS_WBACK_BUSY(req));
 
 		atomic_inc(&req->wb_count);
 		spin_unlock(&nfsi->req_lock);
@@ -821,7 +820,7 @@ out:
 #else
 	nfs_inode_remove_request(req);
 #endif
-	nfs_unlock_request(req);
+	nfs_clear_page_writeback(req);
 }
 
 static inline int flush_task_priority(int how)
@@ -952,7 +951,7 @@ out_bad:
 		nfs_writedata_free(data);
 	}
 	nfs_mark_request_dirty(req);
-	nfs_unlock_request(req);
+	nfs_clear_page_writeback(req);
 	return -ENOMEM;
 }
 
@@ -1002,7 +1001,7 @@ static int nfs_flush_one(struct list_head *head, struct inode *inode, int how)
 		struct nfs_page *req = nfs_list_entry(head->next);
 		nfs_list_remove_request(req);
 		nfs_mark_request_dirty(req);
-		nfs_unlock_request(req);
+		nfs_clear_page_writeback(req);
 	}
 	return -ENOMEM;
 }
@@ -1029,7 +1028,7 @@ nfs_flush_list(struct list_head *head, int wpages, int how)
 		req = nfs_list_entry(head->next);
 		nfs_list_remove_request(req);
 		nfs_mark_request_dirty(req);
-		nfs_unlock_request(req);
+		nfs_clear_page_writeback(req);
 	}
 	return error;
 }
@@ -1121,7 +1120,7 @@ static void nfs_writeback_done_full(struct nfs_write_data *data, int status)
 		nfs_inode_remove_request(req);
 #endif
 	next:
-		nfs_unlock_request(req);
+		nfs_clear_page_writeback(req);
 	}
 }
 
@@ -1278,7 +1277,7 @@ nfs_commit_list(struct list_head *head, int how)
 		req = nfs_list_entry(head->next);
 		nfs_list_remove_request(req);
 		nfs_mark_request_commit(req);
-		nfs_unlock_request(req);
+		nfs_clear_page_writeback(req);
 	}
 	return -ENOMEM;
 }
@@ -1324,7 +1323,7 @@ nfs_commit_done(struct rpc_task *task)
 		dprintk(" mismatch\n");
 		nfs_mark_request_dirty(req);
 	next:
-		nfs_unlock_request(req);
+		nfs_clear_page_writeback(req);
 		res++;
 	}
 	sub_page_state(nr_unstable,res);
diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h
index 39e4895bcdb..db40e4590ba 100644
--- a/include/linux/nfs_page.h
+++ b/include/linux/nfs_page.h
@@ -19,6 +19,11 @@
 
 #include <asm/atomic.h>
 
+/*
+ * Valid flags for the radix tree
+ */
+#define NFS_PAGE_TAG_WRITEBACK	1
+
 /*
  * Valid flags for a dirty buffer
  */
@@ -62,6 +67,9 @@ extern	int nfs_coalesce_requests(struct list_head *, struct list_head *,
 				  unsigned int);
 extern  int nfs_wait_on_request(struct nfs_page *);
 extern	void nfs_unlock_request(struct nfs_page *req);
+extern  int nfs_set_page_writeback_locked(struct nfs_page *req);
+extern  void nfs_clear_page_writeback(struct nfs_page *req);
+
 
 /*
  * Lock the page of an asynchronous request without incrementing the wb_count
@@ -96,10 +104,6 @@ nfs_list_remove_request(struct nfs_page *req)
 {
 	if (list_empty(&req->wb_list))
 		return;
-	if (!NFS_WBACK_BUSY(req)) {
-		printk(KERN_ERR "NFS: unlocked request attempted removed from list!\n");
-		BUG();
-	}
 	list_del_init(&req->wb_list);
 	req->wb_list_head = NULL;
 }
-- 
cgit v1.2.3-70-g09d2


From 3da28eb1c6545fe73263a24eba0996217490e1eb Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 22 Jun 2005 17:16:31 +0000
Subject: [PATCH] NFS: Replace nfs_page insertion sort with a radix sort

 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/inode.c           |  2 +-
 fs/nfs/pagelist.c        | 86 +++++++++++++++++++++++++++++++-----------------
 fs/nfs/write.c           | 71 ++++++++++++++++++---------------------
 include/linux/nfs_fs.h   |  4 +--
 include/linux/nfs_page.h | 18 ++++++++--
 5 files changed, 107 insertions(+), 74 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 4f545f382ba..4845911f1c6 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -135,7 +135,7 @@ nfs_write_inode(struct inode *inode, int sync)
 	int flags = sync ? FLUSH_WAIT : 0;
 	int ret;
 
-	ret = nfs_commit_inode(inode, 0, 0, flags);
+	ret = nfs_commit_inode(inode, flags);
 	if (ret < 0)
 		return ret;
 	return 0;
diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c
index 356a33bb38a..d53857b148e 100644
--- a/fs/nfs/pagelist.c
+++ b/fs/nfs/pagelist.c
@@ -177,36 +177,6 @@ nfs_release_request(struct nfs_page *req)
 	nfs_page_free(req);
 }
 
-/**
- * nfs_list_add_request - Insert a request into a sorted list
- * @req: request
- * @head: head of list into which to insert the request.
- *
- * Note that the wb_list is sorted by page index in order to facilitate
- * coalescing of requests.
- * We use an insertion sort that is optimized for the case of appended
- * writes.
- */
-void
-nfs_list_add_request(struct nfs_page *req, struct list_head *head)
-{
-	struct list_head *pos;
-
-#ifdef NFS_PARANOIA
-	if (!list_empty(&req->wb_list)) {
-		printk(KERN_ERR "NFS: Add to list failed!\n");
-		BUG();
-	}
-#endif
-	list_for_each_prev(pos, head) {
-		struct nfs_page	*p = nfs_list_entry(pos);
-		if (p->wb_index < req->wb_index)
-			break;
-	}
-	list_add(&req->wb_list, pos);
-	req->wb_list_head = head;
-}
-
 static int nfs_wait_bit_interruptible(void *word)
 {
 	int ret = 0;
@@ -291,6 +261,62 @@ nfs_coalesce_requests(struct list_head *head, struct list_head *dst,
 	return npages;
 }
 
+#define NFS_SCAN_MAXENTRIES 16
+/**
+ * nfs_scan_lock_dirty - Scan the radix tree for dirty requests
+ * @nfsi: NFS inode
+ * @dst: Destination list
+ * @idx_start: lower bound of page->index to scan
+ * @npages: idx_start + npages sets the upper bound to scan.
+ *
+ * Moves elements from one of the inode request lists.
+ * If the number of requests is set to 0, the entire address_space
+ * starting at index idx_start, is scanned.
+ * The requests are *not* checked to ensure that they form a contiguous set.
+ * You must be holding the inode's req_lock when calling this function
+ */
+int
+nfs_scan_lock_dirty(struct nfs_inode *nfsi, struct list_head *dst,
+	      unsigned long idx_start, unsigned int npages)
+{
+	struct nfs_page *pgvec[NFS_SCAN_MAXENTRIES];
+	struct nfs_page *req;
+	unsigned long idx_end;
+	int found, i;
+	int res;
+
+	res = 0;
+	if (npages == 0)
+		idx_end = ~0;
+	else
+		idx_end = idx_start + npages - 1;
+
+	for (;;) {
+		found = radix_tree_gang_lookup_tag(&nfsi->nfs_page_tree,
+				(void **)&pgvec[0], idx_start, NFS_SCAN_MAXENTRIES,
+				NFS_PAGE_TAG_DIRTY);
+		if (found <= 0)
+			break;
+		for (i = 0; i < found; i++) {
+			req = pgvec[i];
+			if (req->wb_index > idx_end)
+				goto out;
+
+			idx_start = req->wb_index + 1;
+
+			if (nfs_set_page_writeback_locked(req)) {
+				radix_tree_tag_clear(&nfsi->nfs_page_tree,
+						req->wb_index, NFS_PAGE_TAG_DIRTY);
+				nfs_list_remove_request(req);
+				nfs_list_add_request(req, dst);
+				res++;
+			}
+		}
+	}
+out:
+	return res;
+}
+
 /**
  * nfs_scan_list - Scan a list for matching requests
  * @head: One of the NFS inode request lists
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 58a39b0486a..5130eda231d 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -352,7 +352,7 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc)
 		if (err < 0)
 			goto out;
 	}
-	err = nfs_commit_inode(inode, 0, 0, wb_priority(wbc));
+	err = nfs_commit_inode(inode, wb_priority(wbc));
 	if (err > 0) {
 		wbc->nr_to_write -= err;
 		err = 0;
@@ -446,6 +446,8 @@ nfs_mark_request_dirty(struct nfs_page *req)
 	struct nfs_inode *nfsi = NFS_I(inode);
 
 	spin_lock(&nfsi->req_lock);
+	radix_tree_tag_set(&nfsi->nfs_page_tree,
+			req->wb_index, NFS_PAGE_TAG_DIRTY);
 	nfs_list_add_request(req, &nfsi->dirty);
 	nfsi->ndirty++;
 	spin_unlock(&nfsi->req_lock);
@@ -537,12 +539,15 @@ static int
 nfs_scan_dirty(struct inode *inode, struct list_head *dst, unsigned long idx_start, unsigned int npages)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
-	int	res;
-	res = nfs_scan_list(&nfsi->dirty, dst, idx_start, npages);
-	nfsi->ndirty -= res;
-	sub_page_state(nr_dirty,res);
-	if ((nfsi->ndirty == 0) != list_empty(&nfsi->dirty))
-		printk(KERN_ERR "NFS: desynchronized value of nfs_i.ndirty.\n");
+	int res = 0;
+
+	if (nfsi->ndirty != 0) {
+		res = nfs_scan_lock_dirty(nfsi, dst, idx_start, npages);
+		nfsi->ndirty -= res;
+		sub_page_state(nr_dirty,res);
+		if ((nfsi->ndirty == 0) != list_empty(&nfsi->dirty))
+			printk(KERN_ERR "NFS: desynchronized value of nfs_i.ndirty.\n");
+	}
 	return res;
 }
 
@@ -561,11 +566,14 @@ static int
 nfs_scan_commit(struct inode *inode, struct list_head *dst, unsigned long idx_start, unsigned int npages)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
-	int	res;
-	res = nfs_scan_list(&nfsi->commit, dst, idx_start, npages);
-	nfsi->ncommit -= res;
-	if ((nfsi->ncommit == 0) != list_empty(&nfsi->commit))
-		printk(KERN_ERR "NFS: desynchronized value of nfs_i.ncommit.\n");
+	int res = 0;
+
+	if (nfsi->ncommit != 0) {
+		res = nfs_scan_list(&nfsi->commit, dst, idx_start, npages);
+		nfsi->ncommit -= res;
+		if ((nfsi->ncommit == 0) != list_empty(&nfsi->commit))
+			printk(KERN_ERR "NFS: desynchronized value of nfs_i.ncommit.\n");
+	}
 	return res;
 }
 #endif
@@ -1209,36 +1217,24 @@ static void nfs_commit_rpcsetup(struct list_head *head,
 		struct nfs_write_data *data, int how)
 {
 	struct rpc_task		*task = &data->task;
-	struct nfs_page		*first, *last;
+	struct nfs_page		*first;
 	struct inode		*inode;
-	loff_t			start, end, len;
 
 	/* Set up the RPC argument and reply structs
 	 * NB: take care not to mess about with data->commit et al. */
 
 	list_splice_init(head, &data->pages);
 	first = nfs_list_entry(data->pages.next);
-	last = nfs_list_entry(data->pages.prev);
 	inode = first->wb_context->dentry->d_inode;
 
-	/*
-	 * Determine the offset range of requests in the COMMIT call.
-	 * We rely on the fact that data->pages is an ordered list...
-	 */
-	start = req_offset(first);
-	end = req_offset(last) + last->wb_bytes;
-	len = end - start;
-	/* If 'len' is not a 32-bit quantity, pass '0' in the COMMIT call */
-	if (end >= i_size_read(inode) || len < 0 || len > (~((u32)0) >> 1))
-		len = 0;
-
 	data->inode	  = inode;
 	data->cred	  = first->wb_context->cred;
 
 	data->args.fh     = NFS_FH(data->inode);
-	data->args.offset = start;
-	data->args.count  = len;
-	data->res.count   = len;
+	/* Note: we always request a commit of the entire inode */
+	data->args.offset = 0;
+	data->args.count  = 0;
+	data->res.count   = 0;
 	data->res.fattr   = &data->fattr;
 	data->res.verf    = &data->verf;
 	
@@ -1357,8 +1353,7 @@ static int nfs_flush_inode(struct inode *inode, unsigned long idx_start,
 }
 
 #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
-int nfs_commit_inode(struct inode *inode, unsigned long idx_start,
-		    unsigned int npages, int how)
+int nfs_commit_inode(struct inode *inode, int how)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
 	LIST_HEAD(head);
@@ -1366,15 +1361,13 @@ int nfs_commit_inode(struct inode *inode, unsigned long idx_start,
 				error = 0;
 
 	spin_lock(&nfsi->req_lock);
-	res = nfs_scan_commit(inode, &head, idx_start, npages);
+	res = nfs_scan_commit(inode, &head, 0, 0);
+	spin_unlock(&nfsi->req_lock);
 	if (res) {
-		res += nfs_scan_commit(inode, &head, 0, 0);
-		spin_unlock(&nfsi->req_lock);
 		error = nfs_commit_list(&head, how);
-	} else
-		spin_unlock(&nfsi->req_lock);
-	if (error < 0)
-		return error;
+		if (error < 0)
+			return error;
+	}
 	return res;
 }
 #endif
@@ -1396,7 +1389,7 @@ int nfs_sync_inode(struct inode *inode, unsigned long idx_start,
 			error = nfs_flush_inode(inode, idx_start, npages, how);
 #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
 		if (error == 0)
-			error = nfs_commit_inode(inode, idx_start, npages, how);
+			error = nfs_commit_inode(inode, how);
 #endif
 	} while (error > 0);
 	return error;
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 2954e44ed49..8ea249110fb 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -395,10 +395,10 @@ extern void nfs_commit_done(struct rpc_task *);
  */
 extern int  nfs_sync_inode(struct inode *, unsigned long, unsigned int, int);
 #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
-extern int  nfs_commit_inode(struct inode *, unsigned long, unsigned int, int);
+extern int  nfs_commit_inode(struct inode *, int);
 #else
 static inline int
-nfs_commit_inode(struct inode *inode, unsigned long idx_start, unsigned int npages, int how)
+nfs_commit_inode(struct inode *inode, int how)
 {
 	return 0;
 }
diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h
index db40e4590ba..da2e077b65e 100644
--- a/include/linux/nfs_page.h
+++ b/include/linux/nfs_page.h
@@ -22,6 +22,7 @@
 /*
  * Valid flags for the radix tree
  */
+#define NFS_PAGE_TAG_DIRTY	0
 #define NFS_PAGE_TAG_WRITEBACK	1
 
 /*
@@ -31,6 +32,7 @@
 #define PG_NEED_COMMIT		1
 #define PG_NEED_RESCHED		2
 
+struct nfs_inode;
 struct nfs_page {
 	struct list_head	wb_list,	/* Defines state of page: */
 				*wb_list_head;	/*      read/write/commit */
@@ -59,8 +61,8 @@ extern	void nfs_clear_request(struct nfs_page *req);
 extern	void nfs_release_request(struct nfs_page *req);
 
 
-extern	void nfs_list_add_request(struct nfs_page *, struct list_head *);
-
+extern  int nfs_scan_lock_dirty(struct nfs_inode *nfsi, struct list_head *dst,
+				unsigned long idx_start, unsigned int npages);
 extern	int nfs_scan_list(struct list_head *, struct list_head *,
 			  unsigned long, unsigned int);
 extern	int nfs_coalesce_requests(struct list_head *, struct list_head *,
@@ -94,6 +96,18 @@ nfs_lock_request(struct nfs_page *req)
 	return 1;
 }
 
+/**
+ * nfs_list_add_request - Insert a request into a list
+ * @req: request
+ * @head: head of list into which to insert the request.
+ */
+static inline void
+nfs_list_add_request(struct nfs_page *req, struct list_head *head)
+{
+	list_add_tail(&req->wb_list, head);
+	req->wb_list_head = head;
+}
+
 
 /**
  * nfs_list_remove_request - Remove a request from its wb_list
-- 
cgit v1.2.3-70-g09d2


From ecdbf769b2cb8903e07cd482334c714d89fd1146 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 22 Jun 2005 17:16:31 +0000
Subject: [PATCH] NLM: fix a client-side race on blocking locks.

 If the lock blocks, the server may send us a GRANTED message that
 races with the reply to our LOCK request. Make sure that we catch
 the GRANTED by queueing up our request on the nlm_blocked list
 before we send off the first LOCK rpc call.

 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/lockd/clntlock.c         | 99 ++++++++++++++++++++++++++-------------------
 fs/lockd/clntproc.c         | 40 ++++++++++++++----
 include/linux/lockd/lockd.h |  7 +++-
 3 files changed, 96 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c
index 44adb84183b..006bb9e1457 100644
--- a/fs/lockd/clntlock.c
+++ b/fs/lockd/clntlock.c
@@ -42,23 +42,51 @@ struct nlm_wait {
 static LIST_HEAD(nlm_blocked);
 
 /*
- * Block on a lock
+ * Queue up a lock for blocking so that the GRANTED request can see it
  */
-int
-nlmclnt_block(struct nlm_host *host, struct file_lock *fl, u32 *statp)
+int nlmclnt_prepare_block(struct nlm_rqst *req, struct nlm_host *host, struct file_lock *fl)
+{
+	struct nlm_wait *block;
+
+	BUG_ON(req->a_block != NULL);
+	block = kmalloc(sizeof(*block), GFP_KERNEL);
+	if (block == NULL)
+		return -ENOMEM;
+	block->b_host = host;
+	block->b_lock = fl;
+	init_waitqueue_head(&block->b_wait);
+	block->b_status = NLM_LCK_BLOCKED;
+
+	list_add(&block->b_list, &nlm_blocked);
+	req->a_block = block;
+
+	return 0;
+}
+
+void nlmclnt_finish_block(struct nlm_rqst *req)
 {
-	struct nlm_wait	block, **head;
-	int		err;
-	u32		pstate;
+	struct nlm_wait *block = req->a_block;
 
-	block.b_host   = host;
-	block.b_lock   = fl;
-	init_waitqueue_head(&block.b_wait);
-	block.b_status = NLM_LCK_BLOCKED;
-	list_add(&block.b_list, &nlm_blocked);
+	if (block == NULL)
+		return;
+	req->a_block = NULL;
+	list_del(&block->b_list);
+	kfree(block);
+}
 
-	/* Remember pseudo nsm state */
-	pstate = host->h_state;
+/*
+ * Block on a lock
+ */
+long nlmclnt_block(struct nlm_rqst *req, long timeout)
+{
+	struct nlm_wait	*block = req->a_block;
+	long ret;
+
+	/* A borken server might ask us to block even if we didn't
+	 * request it. Just say no!
+	 */
+	if (!req->a_args.block)
+		return -EAGAIN;
 
 	/* Go to sleep waiting for GRANT callback. Some servers seem
 	 * to lose callbacks, however, so we're going to poll from
@@ -68,23 +96,16 @@ nlmclnt_block(struct nlm_host *host, struct file_lock *fl, u32 *statp)
 	 * a 1 minute timeout would do. See the comment before
 	 * nlmclnt_lock for an explanation.
 	 */
-	sleep_on_timeout(&block.b_wait, 30*HZ);
+	ret = wait_event_interruptible_timeout(block->b_wait,
+			block->b_status != NLM_LCK_BLOCKED,
+			timeout);
 
-	list_del(&block.b_list);
-
-	if (!signalled()) {
-		*statp = block.b_status;
-		return 0;
+	if (block->b_status != NLM_LCK_BLOCKED) {
+		req->a_res.status = block->b_status;
+		block->b_status = NLM_LCK_BLOCKED;
 	}
 
-	/* Okay, we were interrupted. Cancel the pending request
-	 * unless the server has rebooted.
-	 */
-	if (pstate == host->h_state && (err = nlmclnt_cancel(host, fl)) < 0)
-		printk(KERN_NOTICE
-			"lockd: CANCEL call failed (errno %d)\n", -err);
-
-	return -ERESTARTSYS;
+	return ret;
 }
 
 /*
@@ -94,27 +115,23 @@ u32
 nlmclnt_grant(struct nlm_lock *lock)
 {
 	struct nlm_wait	*block;
+	u32 res = nlm_lck_denied;
 
 	/*
 	 * Look up blocked request based on arguments. 
 	 * Warning: must not use cookie to match it!
 	 */
 	list_for_each_entry(block, &nlm_blocked, b_list) {
-		if (nlm_compare_locks(block->b_lock, &lock->fl))
-			break;
+		if (nlm_compare_locks(block->b_lock, &lock->fl)) {
+			/* Alright, we found a lock. Set the return status
+			 * and wake up the caller
+			 */
+			block->b_status = NLM_LCK_GRANTED;
+			wake_up(&block->b_wait);
+			res = nlm_granted;
+		}
 	}
-
-	/* Ooops, no blocked request found. */
-	if (block == NULL)
-		return nlm_lck_denied;
-
-	/* Alright, we found the lock. Set the return status and
-	 * wake up the caller.
-	 */
-	block->b_status = NLM_LCK_GRANTED;
-	wake_up(&block->b_wait);
-
-	return nlm_granted;
+	return res;
 }
 
 /*
diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c
index a4407619b1f..fd77ed1d710 100644
--- a/fs/lockd/clntproc.c
+++ b/fs/lockd/clntproc.c
@@ -21,6 +21,7 @@
 
 #define NLMDBG_FACILITY		NLMDBG_CLIENT
 #define NLMCLNT_GRACE_WAIT	(5*HZ)
+#define NLMCLNT_POLL_TIMEOUT	(30*HZ)
 
 static int	nlmclnt_test(struct nlm_rqst *, struct file_lock *);
 static int	nlmclnt_lock(struct nlm_rqst *, struct file_lock *);
@@ -553,7 +554,8 @@ nlmclnt_lock(struct nlm_rqst *req, struct file_lock *fl)
 {
 	struct nlm_host	*host = req->a_host;
 	struct nlm_res	*resp = &req->a_res;
-	int		status;
+	long timeout;
+	int status;
 
 	if (!host->h_monitored && nsm_monitor(host) < 0) {
 		printk(KERN_NOTICE "lockd: failed to monitor %s\n",
@@ -562,15 +564,32 @@ nlmclnt_lock(struct nlm_rqst *req, struct file_lock *fl)
 		goto out;
 	}
 
-	do {
-		if ((status = nlmclnt_call(req, NLMPROC_LOCK)) >= 0) {
-			if (resp->status != NLM_LCK_BLOCKED)
-				break;
-			status = nlmclnt_block(host, fl, &resp->status);
-		}
+	if (req->a_args.block) {
+		status = nlmclnt_prepare_block(req, host, fl);
 		if (status < 0)
 			goto out;
-	} while (resp->status == NLM_LCK_BLOCKED && req->a_args.block);
+	}
+	for(;;) {
+		status = nlmclnt_call(req, NLMPROC_LOCK);
+		if (status < 0)
+			goto out_unblock;
+		if (resp->status != NLM_LCK_BLOCKED)
+			break;
+		/* Wait on an NLM blocking lock */
+		timeout = nlmclnt_block(req, NLMCLNT_POLL_TIMEOUT);
+		/* Did a reclaimer thread notify us of a server reboot? */
+		if (resp->status ==  NLM_LCK_DENIED_GRACE_PERIOD)
+			continue;
+		if (resp->status != NLM_LCK_BLOCKED)
+			break;
+		if (timeout >= 0)
+			continue;
+		/* We were interrupted. Send a CANCEL request to the server
+		 * and exit
+		 */
+		status = (int)timeout;
+		goto out_unblock;
+	}
 
 	if (resp->status == NLM_LCK_GRANTED) {
 		fl->fl_u.nfs_fl.state = host->h_state;
@@ -579,6 +598,11 @@ nlmclnt_lock(struct nlm_rqst *req, struct file_lock *fl)
 		do_vfs_lock(fl);
 	}
 	status = nlm_stat_to_errno(resp->status);
+out_unblock:
+	nlmclnt_finish_block(req);
+	/* Cancel the blocked request if it is still pending */
+	if (resp->status == NLM_LCK_BLOCKED)
+		nlmclnt_cancel(host, fl);
 out:
 	nlmclnt_release_lockargs(req);
 	return status;
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 0d9d2257821..16d4e5a08e1 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -72,6 +72,8 @@ struct nlm_lockowner {
 	uint32_t pid;
 };
 
+struct nlm_wait;
+
 /*
  * Memory chunk for NLM client RPC request.
  */
@@ -81,6 +83,7 @@ struct nlm_rqst {
 	struct nlm_host *	a_host;		/* host handle */
 	struct nlm_args		a_args;		/* arguments */
 	struct nlm_res		a_res;		/* result */
+	struct nlm_wait *	a_block;
 	char			a_owner[NLMCLNT_OHSIZE];
 };
 
@@ -142,7 +145,9 @@ extern unsigned long		nlmsvc_timeout;
  * Lockd client functions
  */
 struct nlm_rqst * nlmclnt_alloc_call(void);
-int		  nlmclnt_block(struct nlm_host *, struct file_lock *, u32 *);
+int		  nlmclnt_prepare_block(struct nlm_rqst *req, struct nlm_host *host, struct file_lock *fl);
+void		  nlmclnt_finish_block(struct nlm_rqst *req);
+long		  nlmclnt_block(struct nlm_rqst *req, long timeout);
 int		  nlmclnt_cancel(struct nlm_host *, struct file_lock *);
 u32		  nlmclnt_grant(struct nlm_lock *);
 void		  nlmclnt_recovery(struct nlm_host *, u32);
-- 
cgit v1.2.3-70-g09d2


From 8d0a8a9d0ec790086c64d210af413ac351d89e35 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 22 Jun 2005 17:16:32 +0000
Subject: [PATCH] NFSv4: Clean up nfs4 lock state accounting

 Ensure that lock owner structures are not released prematurely.

 Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4_fs.h         |   9 +--
 fs/nfs/nfs4proc.c        |  69 ++++++++----------
 fs/nfs/nfs4state.c       | 178 +++++++++++++++++++++--------------------------
 include/linux/fs.h       |   1 +
 include/linux/nfs_fs_i.h |   5 ++
 5 files changed, 118 insertions(+), 144 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 7c6f1d668fb..ec1a22d7b87 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -128,6 +128,7 @@ struct nfs4_state_owner {
 
 struct nfs4_lock_state {
 	struct list_head	ls_locks;	/* Other lock stateids */
+	struct nfs4_state *	ls_state;	/* Pointer to open state */
 	fl_owner_t		ls_owner;	/* POSIX lock owner */
 #define NFS_LOCK_INITIALIZED 1
 	int			ls_flags;
@@ -153,7 +154,7 @@ struct nfs4_state {
 
 	unsigned long flags;		/* Do we hold any locks? */
 	struct semaphore lock_sema;	/* Serializes file locking operations */
-	rwlock_t state_lock;		/* Protects the lock_states list */
+	spinlock_t state_lock;		/* Protects the lock_states list */
 
 	nfs4_stateid stateid;
 
@@ -225,12 +226,8 @@ extern void nfs4_close_state(struct nfs4_state *, mode_t);
 extern struct nfs4_state *nfs4_find_state(struct inode *, struct rpc_cred *, mode_t mode);
 extern void nfs4_increment_seqid(int status, struct nfs4_state_owner *sp);
 extern void nfs4_schedule_state_recovery(struct nfs4_client *);
-extern struct nfs4_lock_state *nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t);
-extern struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t);
-extern void nfs4_put_lock_state(struct nfs4_lock_state *state);
+extern int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl);
 extern void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *ls);
-extern void nfs4_notify_setlk(struct nfs4_state *, struct file_lock *, struct nfs4_lock_state *);
-extern void nfs4_notify_unlck(struct nfs4_state *, struct file_lock *, struct nfs4_lock_state *);
 extern void nfs4_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t);
 
 extern const nfs4_stateid zero_stateid;
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index af80b598148..0ddc20102d4 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2626,14 +2626,11 @@ static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock
 	down_read(&clp->cl_sem);
 	nlo.clientid = clp->cl_clientid;
 	down(&state->lock_sema);
-	lsp = nfs4_find_lock_state(state, request->fl_owner);
-	if (lsp)
-		nlo.id = lsp->ls_id; 
-	else {
-		spin_lock(&clp->cl_lock);
-		nlo.id = nfs4_alloc_lockowner_id(clp);
-		spin_unlock(&clp->cl_lock);
-	}
+	status = nfs4_set_lock_state(state, request);
+	if (status != 0)
+		goto out;
+	lsp = request->fl_u.nfs4_fl.owner;
+	nlo.id = lsp->ls_id; 
 	arg.u.lockt = &nlo;
 	status = rpc_call_sync(server->client, &msg, 0);
 	if (!status) {
@@ -2654,8 +2651,7 @@ static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock
 		request->fl_pid = 0;
 		status = 0;
 	}
-	if (lsp)
-		nfs4_put_lock_state(lsp);
+out:
 	up(&state->lock_sema);
 	up_read(&clp->cl_sem);
 	return status;
@@ -2715,28 +2711,26 @@ static int _nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock
 	};
 	struct nfs4_lock_state *lsp;
 	struct nfs_locku_opargs luargs;
-	int status = 0;
+	int status;
 			
 	down_read(&clp->cl_sem);
 	down(&state->lock_sema);
-	lsp = nfs4_find_lock_state(state, request->fl_owner);
-	if (!lsp)
+	status = nfs4_set_lock_state(state, request);
+	if (status != 0)
 		goto out;
+	lsp = request->fl_u.nfs4_fl.owner;
 	/* We might have lost the locks! */
-	if ((lsp->ls_flags & NFS_LOCK_INITIALIZED) != 0) {
-		luargs.seqid = lsp->ls_seqid;
-		memcpy(&luargs.stateid, &lsp->ls_stateid, sizeof(luargs.stateid));
-		arg.u.locku = &luargs;
-		status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR);
-		nfs4_increment_lock_seqid(status, lsp);
-	}
+	if ((lsp->ls_flags & NFS_LOCK_INITIALIZED) == 0)
+		goto out;
+	luargs.seqid = lsp->ls_seqid;
+	memcpy(&luargs.stateid, &lsp->ls_stateid, sizeof(luargs.stateid));
+	arg.u.locku = &luargs;
+	status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR);
+	nfs4_increment_lock_seqid(status, lsp);
 
-	if (status == 0) {
+	if (status == 0)
 		memcpy(&lsp->ls_stateid,  &res.u.stateid, 
 				sizeof(lsp->ls_stateid));
-		nfs4_notify_unlck(state, request, lsp);
-	}
-	nfs4_put_lock_state(lsp);
 out:
 	up(&state->lock_sema);
 	if (status == 0)
@@ -2762,7 +2756,7 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *r
 {
 	struct inode *inode = state->inode;
 	struct nfs_server *server = NFS_SERVER(inode);
-	struct nfs4_lock_state *lsp;
+	struct nfs4_lock_state *lsp = request->fl_u.nfs4_fl.owner;
 	struct nfs_lockargs arg = {
 		.fh = NFS_FH(inode),
 		.type = nfs4_lck_type(cmd, request),
@@ -2784,9 +2778,6 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *r
 	};
 	int status;
 
-	lsp = nfs4_get_lock_state(state, request->fl_owner);
-	if (lsp == NULL)
-		return -ENOMEM;
 	if (!(lsp->ls_flags & NFS_LOCK_INITIALIZED)) {
 		struct nfs4_state_owner *owner = state->owner;
 		struct nfs_open_to_lock otl = {
@@ -2808,27 +2799,26 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *r
 		* seqid mutating errors */
 		nfs4_increment_seqid(status, owner);
 		up(&owner->so_sema);
+		if (status == 0) {
+			lsp->ls_flags |= NFS_LOCK_INITIALIZED;
+			lsp->ls_seqid++;
+		}
 	} else {
 		struct nfs_exist_lock el = {
 			.seqid = lsp->ls_seqid,
 		};
 		memcpy(&el.stateid, &lsp->ls_stateid, sizeof(el.stateid));
 		largs.u.exist_lock = &el;
-		largs.new_lock_owner = 0;
 		arg.u.lock = &largs;
 		status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR);
+		/* increment seqid on success, and * seqid mutating errors*/
+		nfs4_increment_lock_seqid(status, lsp);
 	}
-	/* increment seqid on success, and * seqid mutating errors*/
-	nfs4_increment_lock_seqid(status, lsp);
 	/* save the returned stateid. */
-	if (status == 0) {
+	if (status == 0)
 		memcpy(&lsp->ls_stateid, &res.u.stateid, sizeof(nfs4_stateid));
-		lsp->ls_flags |= NFS_LOCK_INITIALIZED;
-		if (!reclaim)
-			nfs4_notify_setlk(state, request, lsp);
-	} else if (status == -NFS4ERR_DENIED)
+	else if (status == -NFS4ERR_DENIED)
 		status = -EAGAIN;
-	nfs4_put_lock_state(lsp);
 	return status;
 }
 
@@ -2869,7 +2859,9 @@ static int _nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock
 
 	down_read(&clp->cl_sem);
 	down(&state->lock_sema);
-	status = _nfs4_do_setlk(state, cmd, request, 0);
+	status = nfs4_set_lock_state(state, request);
+	if (status == 0)
+		status = _nfs4_do_setlk(state, cmd, request, 0);
 	up(&state->lock_sema);
 	if (status == 0) {
 		/* Note: we always want to sleep here! */
@@ -2927,7 +2919,6 @@ nfs4_proc_lock(struct file *filp, int cmd, struct file_lock *request)
 		if (signalled())
 			break;
 	} while(status < 0);
-
 	return status;
 }
 
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 591ad1d5188..afe587d82f1 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -360,7 +360,7 @@ nfs4_alloc_open_state(void)
 	atomic_set(&state->count, 1);
 	INIT_LIST_HEAD(&state->lock_states);
 	init_MUTEX(&state->lock_sema);
-	rwlock_init(&state->state_lock);
+	spin_lock_init(&state->state_lock);
 	return state;
 }
 
@@ -542,16 +542,6 @@ __nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t fl_owner)
 	return NULL;
 }
 
-struct nfs4_lock_state *
-nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t fl_owner)
-{
-	struct nfs4_lock_state *lsp;
-	read_lock(&state->state_lock);
-	lsp = __nfs4_find_lock_state(state, fl_owner);
-	read_unlock(&state->state_lock);
-	return lsp;
-}
-
 /*
  * Return a compatible lock_state. If no initialized lock_state structure
  * exists, return an uninitialized one.
@@ -568,14 +558,13 @@ static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, f
 		return NULL;
 	lsp->ls_flags = 0;
 	lsp->ls_seqid = 0;	/* arbitrary */
-	lsp->ls_id = -1; 
 	memset(lsp->ls_stateid.data, 0, sizeof(lsp->ls_stateid.data));
 	atomic_set(&lsp->ls_count, 1);
 	lsp->ls_owner = fl_owner;
-	INIT_LIST_HEAD(&lsp->ls_locks);
 	spin_lock(&clp->cl_lock);
 	lsp->ls_id = nfs4_alloc_lockowner_id(clp);
 	spin_unlock(&clp->cl_lock);
+	INIT_LIST_HEAD(&lsp->ls_locks);
 	return lsp;
 }
 
@@ -585,121 +574,112 @@ static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, f
  *
  * The caller must be holding state->lock_sema and clp->cl_sem
  */
-struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t owner)
+static struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t owner)
 {
-	struct nfs4_lock_state * lsp;
+	struct nfs4_lock_state *lsp, *new = NULL;
 	
-	lsp = nfs4_find_lock_state(state, owner);
-	if (lsp == NULL)
-		lsp = nfs4_alloc_lock_state(state, owner);
+	for(;;) {
+		spin_lock(&state->state_lock);
+		lsp = __nfs4_find_lock_state(state, owner);
+		if (lsp != NULL)
+			break;
+		if (new != NULL) {
+			new->ls_state = state;
+			list_add(&new->ls_locks, &state->lock_states);
+			set_bit(LK_STATE_IN_USE, &state->flags);
+			lsp = new;
+			new = NULL;
+			break;
+		}
+		spin_unlock(&state->state_lock);
+		new = nfs4_alloc_lock_state(state, owner);
+		if (new == NULL)
+			return NULL;
+	}
+	spin_unlock(&state->state_lock);
+	kfree(new);
 	return lsp;
 }
 
 /*
- * Byte-range lock aware utility to initialize the stateid of read/write
- * requests.
+ * Release reference to lock_state, and free it if we see that
+ * it is no longer in use
  */
-void
-nfs4_copy_stateid(nfs4_stateid *dst, struct nfs4_state *state, fl_owner_t fl_owner)
+static void nfs4_put_lock_state(struct nfs4_lock_state *lsp)
 {
-	if (test_bit(LK_STATE_IN_USE, &state->flags)) {
-		struct nfs4_lock_state *lsp;
+	struct nfs4_state *state;
 
-		lsp = nfs4_find_lock_state(state, fl_owner);
-		if (lsp) {
-			memcpy(dst, &lsp->ls_stateid, sizeof(*dst));
-			nfs4_put_lock_state(lsp);
-			return;
-		}
-	}
-	memcpy(dst, &state->stateid, sizeof(*dst));
+	if (lsp == NULL)
+		return;
+	state = lsp->ls_state;
+	if (!atomic_dec_and_lock(&lsp->ls_count, &state->state_lock))
+		return;
+	list_del(&lsp->ls_locks);
+	if (list_empty(&state->lock_states))
+		clear_bit(LK_STATE_IN_USE, &state->flags);
+	spin_unlock(&state->state_lock);
+	kfree(lsp);
 }
 
-/*
-* Called with state->lock_sema and clp->cl_sem held.
-*/
-void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *lsp)
+static void nfs4_fl_copy_lock(struct file_lock *dst, struct file_lock *src)
 {
-	if (status == NFS_OK || seqid_mutating_err(-status))
-		lsp->ls_seqid++;
-}
+	struct nfs4_lock_state *lsp = src->fl_u.nfs4_fl.owner;
 
-/* 
-* Check to see if the request lock (type FL_UNLK) effects the fl lock.
-*
-* fl and request must have the same posix owner
-*
-* return: 
-* 0 -> fl not effected by request
-* 1 -> fl consumed by request
-*/
+	dst->fl_u.nfs4_fl.owner = lsp;
+	atomic_inc(&lsp->ls_count);
+}
 
-static int
-nfs4_check_unlock(struct file_lock *fl, struct file_lock *request)
+static void nfs4_fl_release_lock(struct file_lock *fl)
 {
-	if (fl->fl_start >= request->fl_start && fl->fl_end <= request->fl_end)
-		return 1;
-	return 0;
+	nfs4_put_lock_state(fl->fl_u.nfs4_fl.owner);
 }
 
-/*
- * Post an initialized lock_state on the state->lock_states list.
- */
-void nfs4_notify_setlk(struct nfs4_state *state, struct file_lock *request, struct nfs4_lock_state *lsp)
+static struct file_lock_operations nfs4_fl_lock_ops = {
+	.fl_copy_lock = nfs4_fl_copy_lock,
+	.fl_release_private = nfs4_fl_release_lock,
+};
+
+int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl)
 {
-	if (!list_empty(&lsp->ls_locks))
-		return;
-	atomic_inc(&lsp->ls_count);
-	write_lock(&state->state_lock);
-	list_add(&lsp->ls_locks, &state->lock_states);
-	set_bit(LK_STATE_IN_USE, &state->flags);
-	write_unlock(&state->state_lock);
+	struct nfs4_lock_state *lsp;
+
+	if (fl->fl_ops != NULL)
+		return 0;
+	lsp = nfs4_get_lock_state(state, fl->fl_owner);
+	if (lsp == NULL)
+		return -ENOMEM;
+	fl->fl_u.nfs4_fl.owner = lsp;
+	fl->fl_ops = &nfs4_fl_lock_ops;
+	return 0;
 }
 
-/* 
- * to decide to 'reap' lock state:
- * 1) search i_flock for file_locks with fl.lock_state = to ls.
- * 2) determine if unlock will consume found lock. 
- * 	if so, reap
- *
- * 	else, don't reap.
- *
+/*
+ * Byte-range lock aware utility to initialize the stateid of read/write
+ * requests.
  */
-void
-nfs4_notify_unlck(struct nfs4_state *state, struct file_lock *request, struct nfs4_lock_state *lsp)
+void nfs4_copy_stateid(nfs4_stateid *dst, struct nfs4_state *state, fl_owner_t fl_owner)
 {
-	struct inode *inode = state->inode;
-	struct file_lock *fl;
+	struct nfs4_lock_state *lsp;
 
-	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
-		if (!(fl->fl_flags & FL_POSIX))
-			continue;
-		if (fl->fl_owner != lsp->ls_owner)
-			continue;
-		/* Exit if we find at least one lock which is not consumed */
-		if (nfs4_check_unlock(fl,request) == 0)
-			return;
-	}
+	memcpy(dst, &state->stateid, sizeof(*dst));
+	if (test_bit(LK_STATE_IN_USE, &state->flags) == 0)
+		return;
 
-	write_lock(&state->state_lock);
-	list_del_init(&lsp->ls_locks);
-	if (list_empty(&state->lock_states))
-		clear_bit(LK_STATE_IN_USE, &state->flags);
-	write_unlock(&state->state_lock);
+	spin_lock(&state->state_lock);
+	lsp = __nfs4_find_lock_state(state, fl_owner);
+	if (lsp != NULL && (lsp->ls_flags & NFS_LOCK_INITIALIZED) != 0)
+		memcpy(dst, &lsp->ls_stateid, sizeof(*dst));
+	spin_unlock(&state->state_lock);
 	nfs4_put_lock_state(lsp);
 }
 
 /*
- * Release reference to lock_state, and free it if we see that
- * it is no longer in use
- */
-void
-nfs4_put_lock_state(struct nfs4_lock_state *lsp)
+* Called with state->lock_sema and clp->cl_sem held.
+*/
+void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *lsp)
 {
-	if (!atomic_dec_and_test(&lsp->ls_count))
-		return;
-	BUG_ON (!list_empty(&lsp->ls_locks));
-	kfree(lsp);
+	if (status == NFS_OK || seqid_mutating_err(-status))
+		lsp->ls_seqid++;
 }
 
 /*
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9b8b696d4f1..e5a8db00df2 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -674,6 +674,7 @@ struct file_lock {
 	struct lock_manager_operations *fl_lmops;	/* Callbacks for lockmanagers */
 	union {
 		struct nfs_lock_info	nfs_fl;
+		struct nfs4_lock_info	nfs4_fl;
 	} fl_u;
 };
 
diff --git a/include/linux/nfs_fs_i.h b/include/linux/nfs_fs_i.h
index e9a749588a7..e2c18dabff8 100644
--- a/include/linux/nfs_fs_i.h
+++ b/include/linux/nfs_fs_i.h
@@ -16,6 +16,11 @@ struct nfs_lock_info {
 	struct nlm_lockowner *owner;
 };
 
+struct nfs4_lock_state;
+struct nfs4_lock_info {
+	struct nfs4_lock_state *owner;
+};
+
 /*
  * Lock flag values
  */
-- 
cgit v1.2.3-70-g09d2


From 10f7e7c15e6ce41799c5dba6925ae4bf8048c870 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 23 Jun 2005 09:43:07 +1000
Subject: [PATCH] ppc64: consolidate calibrate_decr implementations

pSeries and maple have almost the same code for calibrate_decr,
and BPA would need yet another copy. Instead, I'm moving the
code to arch/ppc64/kernel/time.c.

Some of the related declarations were missing from header
files, so I'm moving those as well.

It makes sense to merge this with the pmac function of the
same name, so we end up having just one implemetation for
iSeries and one for Open Firmware based machines.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/iSeries_setup.c |  5 ---
 arch/ppc64/kernel/maple_setup.c   |  2 +-
 arch/ppc64/kernel/maple_time.c    | 51 -----------------------------
 arch/ppc64/kernel/pSeries_setup.c | 69 +--------------------------------------
 arch/ppc64/kernel/pmac_time.c     |  8 +----
 arch/ppc64/kernel/setup.c         |  5 +--
 arch/ppc64/kernel/time.c          | 63 +++++++++++++++++++++++++++++++++++
 include/asm-ppc64/time.h          |  9 +++++
 8 files changed, 76 insertions(+), 136 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/iSeries_setup.c b/arch/ppc64/kernel/iSeries_setup.c
index b31962436fe..86966ce76b5 100644
--- a/arch/ppc64/kernel/iSeries_setup.c
+++ b/arch/ppc64/kernel/iSeries_setup.c
@@ -671,9 +671,6 @@ static void __init iSeries_bolt_kernel(unsigned long saddr, unsigned long eaddr)
 	}
 }
 
-extern unsigned long ppc_proc_freq;
-extern unsigned long ppc_tb_freq;
-
 /*
  * Document me.
  */
@@ -772,8 +769,6 @@ static void iSeries_halt(void)
 	mf_power_off();
 }
 
-extern void setup_default_decr(void);
-
 /*
  * void __init iSeries_calibrate_decr()
  *
diff --git a/arch/ppc64/kernel/maple_setup.c b/arch/ppc64/kernel/maple_setup.c
index 8cf95a27178..3dabedb8553 100644
--- a/arch/ppc64/kernel/maple_setup.c
+++ b/arch/ppc64/kernel/maple_setup.c
@@ -235,6 +235,6 @@ struct machdep_calls __initdata maple_md = {
        	.get_boot_time		= maple_get_boot_time,
        	.set_rtc_time		= maple_set_rtc_time,
        	.get_rtc_time		= maple_get_rtc_time,
-      	.calibrate_decr		= maple_calibrate_decr,
+      	.calibrate_decr		= generic_calibrate_decr,
 	.progress		= maple_progress,
 };
diff --git a/arch/ppc64/kernel/maple_time.c b/arch/ppc64/kernel/maple_time.c
index 07ce7895b43..d65210abcd0 100644
--- a/arch/ppc64/kernel/maple_time.c
+++ b/arch/ppc64/kernel/maple_time.c
@@ -42,11 +42,8 @@
 #define DBG(x...)
 #endif
 
-extern void setup_default_decr(void);
 extern void GregorianDay(struct rtc_time * tm);
 
-extern unsigned long ppc_tb_freq;
-extern unsigned long ppc_proc_freq;
 static int maple_rtc_addr;
 
 static int maple_clock_read(int addr)
@@ -176,51 +173,3 @@ void __init maple_get_boot_time(struct rtc_time *tm)
 	maple_get_rtc_time(tm);
 }
 
-/* XXX FIXME: Some sane defaults: 125 MHz timebase, 1GHz processor */
-#define DEFAULT_TB_FREQ		125000000UL
-#define DEFAULT_PROC_FREQ	(DEFAULT_TB_FREQ * 8)
-
-void __init maple_calibrate_decr(void)
-{
-	struct device_node *cpu;
-	struct div_result divres;
-	unsigned int *fp = NULL;
-
-	/*
-	 * The cpu node should have a timebase-frequency property
-	 * to tell us the rate at which the decrementer counts.
-	 */
-	cpu = of_find_node_by_type(NULL, "cpu");
-
-	ppc_tb_freq = DEFAULT_TB_FREQ;
-	if (cpu != 0)
-		fp = (unsigned int *)get_property(cpu, "timebase-frequency", NULL);
-	if (fp != NULL)
-		ppc_tb_freq = *fp;
-	else
-		printk(KERN_ERR "WARNING: Estimating decrementer frequency (not found)\n");
-	fp = NULL;
-	ppc_proc_freq = DEFAULT_PROC_FREQ;
-	if (cpu != 0)
-		fp = (unsigned int *)get_property(cpu, "clock-frequency", NULL);
-	if (fp != NULL)
-		ppc_proc_freq = *fp;
-	else
-		printk(KERN_ERR "WARNING: Estimating processor frequency (not found)\n");
-
-	of_node_put(cpu);
-
-	printk(KERN_INFO "time_init: decrementer frequency = %lu.%.6lu MHz\n",
-	       ppc_tb_freq/1000000, ppc_tb_freq%1000000);
-	printk(KERN_INFO "time_init: processor frequency   = %lu.%.6lu MHz\n",
-	       ppc_proc_freq/1000000, ppc_proc_freq%1000000);
-
-	tb_ticks_per_jiffy = ppc_tb_freq / HZ;
-	tb_ticks_per_sec = tb_ticks_per_jiffy * HZ;
-	tb_ticks_per_usec = ppc_tb_freq / 1000000;
-	tb_to_us = mulhwu_scale_factor(ppc_tb_freq, 1000000);
-	div128_by_32(1024*1024, 0, tb_ticks_per_sec, &divres);
-	tb_to_xs = divres.result_low;
-
-	setup_default_decr();
-}
diff --git a/arch/ppc64/kernel/pSeries_setup.c b/arch/ppc64/kernel/pSeries_setup.c
index 6c0d1d58a55..f9a310c0c8d 100644
--- a/arch/ppc64/kernel/pSeries_setup.c
+++ b/arch/ppc64/kernel/pSeries_setup.c
@@ -84,9 +84,6 @@ extern void generic_find_legacy_serial_ports(u64 *physport,
 
 int fwnmi_active;  /* TRUE if an FWNMI handler is present */
 
-extern unsigned long ppc_proc_freq;
-extern unsigned long ppc_tb_freq;
-
 extern void pSeries_system_reset_exception(struct pt_regs *regs);
 extern int pSeries_machine_check_exception(struct pt_regs *regs);
 
@@ -482,70 +479,6 @@ static void pSeries_progress(char *s, unsigned short hex)
 	spin_unlock(&progress_lock);
 }
 
-extern void setup_default_decr(void);
-
-/* Some sane defaults: 125 MHz timebase, 1GHz processor */
-#define DEFAULT_TB_FREQ		125000000UL
-#define DEFAULT_PROC_FREQ	(DEFAULT_TB_FREQ * 8)
-
-static void __init pSeries_calibrate_decr(void)
-{
-	struct device_node *cpu;
-	struct div_result divres;
-	unsigned int *fp;
-	int node_found;
-
-	/*
-	 * The cpu node should have a timebase-frequency property
-	 * to tell us the rate at which the decrementer counts.
-	 */
-	cpu = of_find_node_by_type(NULL, "cpu");
-
-	ppc_tb_freq = DEFAULT_TB_FREQ;		/* hardcoded default */
-	node_found = 0;
-	if (cpu != 0) {
-		fp = (unsigned int *)get_property(cpu, "timebase-frequency",
-						  NULL);
-		if (fp != 0) {
-			node_found = 1;
-			ppc_tb_freq = *fp;
-		}
-	}
-	if (!node_found)
-		printk(KERN_ERR "WARNING: Estimating decrementer frequency "
-				"(not found)\n");
-
-	ppc_proc_freq = DEFAULT_PROC_FREQ;
-	node_found = 0;
-	if (cpu != 0) {
-		fp = (unsigned int *)get_property(cpu, "clock-frequency",
-						  NULL);
-		if (fp != 0) {
-			node_found = 1;
-			ppc_proc_freq = *fp;
-		}
-	}
-	if (!node_found)
-		printk(KERN_ERR "WARNING: Estimating processor frequency "
-				"(not found)\n");
-
-	of_node_put(cpu);
-
-	printk(KERN_INFO "time_init: decrementer frequency = %lu.%.6lu MHz\n",
-	       ppc_tb_freq/1000000, ppc_tb_freq%1000000);
-	printk(KERN_INFO "time_init: processor frequency   = %lu.%.6lu MHz\n",
-	       ppc_proc_freq/1000000, ppc_proc_freq%1000000);
-
-	tb_ticks_per_jiffy = ppc_tb_freq / HZ;
-	tb_ticks_per_sec = tb_ticks_per_jiffy * HZ;
-	tb_ticks_per_usec = ppc_tb_freq / 1000000;
-	tb_to_us = mulhwu_scale_factor(ppc_tb_freq, 1000000);
-	div128_by_32(1024*1024, 0, tb_ticks_per_sec, &divres);
-	tb_to_xs = divres.result_low;
-
-	setup_default_decr();
-}
-
 static int pSeries_check_legacy_ioport(unsigned int baseport)
 {
 	struct device_node *np;
@@ -604,7 +537,7 @@ struct machdep_calls __initdata pSeries_md = {
 	.get_boot_time		= pSeries_get_boot_time,
 	.get_rtc_time		= pSeries_get_rtc_time,
 	.set_rtc_time		= pSeries_set_rtc_time,
-	.calibrate_decr		= pSeries_calibrate_decr,
+	.calibrate_decr		= generic_calibrate_decr,
 	.progress		= pSeries_progress,
 	.check_legacy_ioport	= pSeries_check_legacy_ioport,
 	.system_reset_exception = pSeries_system_reset_exception,
diff --git a/arch/ppc64/kernel/pmac_time.c b/arch/ppc64/kernel/pmac_time.c
index f24827581dd..3059edb09cc 100644
--- a/arch/ppc64/kernel/pmac_time.c
+++ b/arch/ppc64/kernel/pmac_time.c
@@ -40,11 +40,6 @@
 #define DBG(x...)
 #endif
 
-extern void setup_default_decr(void);
-
-extern unsigned long ppc_tb_freq;
-extern unsigned long ppc_proc_freq;
-
 /* Apparently the RTC stores seconds since 1 Jan 1904 */
 #define RTC_OFFSET	2082844800
 
@@ -161,8 +156,7 @@ void __init pmac_get_boot_time(struct rtc_time *tm)
 
 /*
  * Query the OF and get the decr frequency.
- * This was taken from the pmac time_init() when merging the prep/pmac
- * time functions.
+ * FIXME: merge this with generic_calibrate_decr
  */
 void __init pmac_calibrate_decr(void)
 {
diff --git a/arch/ppc64/kernel/setup.c b/arch/ppc64/kernel/setup.c
index 8e439a81764..93b0ee88cda 100644
--- a/arch/ppc64/kernel/setup.c
+++ b/arch/ppc64/kernel/setup.c
@@ -701,9 +701,6 @@ void machine_halt(void)
 
 EXPORT_SYMBOL(machine_halt);
 
-unsigned long ppc_proc_freq;
-unsigned long ppc_tb_freq;
-
 static int ppc64_panic_event(struct notifier_block *this,
                              unsigned long event, void *ptr)
 {
@@ -1117,7 +1114,7 @@ void ppc64_dump_msg(unsigned int src, const char *msg)
 }
 
 /* This should only be called on processor 0 during calibrate decr */
-void setup_default_decr(void)
+void __init setup_default_decr(void)
 {
 	struct paca_struct *lpaca = get_paca();
 
diff --git a/arch/ppc64/kernel/time.c b/arch/ppc64/kernel/time.c
index 33364a7d2cd..2348a75e050 100644
--- a/arch/ppc64/kernel/time.c
+++ b/arch/ppc64/kernel/time.c
@@ -107,6 +107,9 @@ void ppc_adjtimex(void);
 
 static unsigned adjusting_time = 0;
 
+unsigned long ppc_proc_freq;
+unsigned long ppc_tb_freq;
+
 static __inline__ void timer_check_rtc(void)
 {
         /*
@@ -472,6 +475,66 @@ int do_settimeofday(struct timespec *tv)
 
 EXPORT_SYMBOL(do_settimeofday);
 
+#if defined(CONFIG_PPC_PSERIES) || defined(CONFIG_PPC_MAPLE) || defined(CONFIG_PPC_BPA)
+void __init generic_calibrate_decr(void)
+{
+	struct device_node *cpu;
+	struct div_result divres;
+	unsigned int *fp;
+	int node_found;
+
+	/*
+	 * The cpu node should have a timebase-frequency property
+	 * to tell us the rate at which the decrementer counts.
+	 */
+	cpu = of_find_node_by_type(NULL, "cpu");
+
+	ppc_tb_freq = DEFAULT_TB_FREQ;		/* hardcoded default */
+	node_found = 0;
+	if (cpu != 0) {
+		fp = (unsigned int *)get_property(cpu, "timebase-frequency",
+						  NULL);
+		if (fp != 0) {
+			node_found = 1;
+			ppc_tb_freq = *fp;
+		}
+	}
+	if (!node_found)
+		printk(KERN_ERR "WARNING: Estimating decrementer frequency "
+				"(not found)\n");
+
+	ppc_proc_freq = DEFAULT_PROC_FREQ;
+	node_found = 0;
+	if (cpu != 0) {
+		fp = (unsigned int *)get_property(cpu, "clock-frequency",
+						  NULL);
+		if (fp != 0) {
+			node_found = 1;
+			ppc_proc_freq = *fp;
+		}
+	}
+	if (!node_found)
+		printk(KERN_ERR "WARNING: Estimating processor frequency "
+				"(not found)\n");
+
+	of_node_put(cpu);
+
+	printk(KERN_INFO "time_init: decrementer frequency = %lu.%.6lu MHz\n",
+	       ppc_tb_freq/1000000, ppc_tb_freq%1000000);
+	printk(KERN_INFO "time_init: processor frequency   = %lu.%.6lu MHz\n",
+	       ppc_proc_freq/1000000, ppc_proc_freq%1000000);
+
+	tb_ticks_per_jiffy = ppc_tb_freq / HZ;
+	tb_ticks_per_sec = tb_ticks_per_jiffy * HZ;
+	tb_ticks_per_usec = ppc_tb_freq / 1000000;
+	tb_to_us = mulhwu_scale_factor(ppc_tb_freq, 1000000);
+	div128_by_32(1024*1024, 0, tb_ticks_per_sec, &divres);
+	tb_to_xs = divres.result_low;
+
+	setup_default_decr();
+}
+#endif
+
 void __init time_init(void)
 {
 	/* This function is only called on the boot processor */
diff --git a/include/asm-ppc64/time.h b/include/asm-ppc64/time.h
index 8d6e3760ee1..c6c762cad8b 100644
--- a/include/asm-ppc64/time.h
+++ b/include/asm-ppc64/time.h
@@ -34,6 +34,15 @@ struct rtc_time;
 extern void to_tm(int tim, struct rtc_time * tm);
 extern time_t last_rtc_update;
 
+void generic_calibrate_decr(void);
+void setup_default_decr(void);
+
+/* Some sane defaults: 125 MHz timebase, 1GHz processor */
+extern unsigned long ppc_proc_freq;
+#define DEFAULT_PROC_FREQ	(DEFAULT_TB_FREQ * 8)
+extern unsigned long ppc_tb_freq;
+#define DEFAULT_TB_FREQ		125000000UL
+
 /*
  * By putting all of this stuff into a single struct we 
  * reduce the number of cache lines touched by do_gettimeofday.
-- 
cgit v1.2.3-70-g09d2


From 773bf9c469c01f01280c9bd45ec2462dd94d08a0 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 23 Jun 2005 09:43:18 +1000
Subject: [PATCH] ppc64: rename pSeries rtc functions into rtas_*

The rtc rtas functions are not pSeries specific but can
also be used by BPA and other SLOF based platforms

Signed-off-by: Arnd Bergmann <arndb@de.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/pSeries_setup.c | 9 +++------
 arch/ppc64/kernel/rtc.c           | 6 +++---
 include/asm-ppc64/rtas.h          | 5 +++++
 3 files changed, 11 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/pSeries_setup.c b/arch/ppc64/kernel/pSeries_setup.c
index f9a310c0c8d..c5f3ccac7cd 100644
--- a/arch/ppc64/kernel/pSeries_setup.c
+++ b/arch/ppc64/kernel/pSeries_setup.c
@@ -73,9 +73,6 @@
 
 extern void pSeries_final_fixup(void);
 
-extern void pSeries_get_boot_time(struct rtc_time *rtc_time);
-extern void pSeries_get_rtc_time(struct rtc_time *rtc_time);
-extern int  pSeries_set_rtc_time(struct rtc_time *rtc_time);
 extern void find_udbg_vterm(void);
 extern void system_reset_fwnmi(void);	/* from head.S */
 extern void machine_check_fwnmi(void);	/* from head.S */
@@ -534,9 +531,9 @@ struct machdep_calls __initdata pSeries_md = {
 	.halt			= rtas_halt,
 	.panic			= rtas_os_term,
 	.cpu_die		= pSeries_mach_cpu_die,
-	.get_boot_time		= pSeries_get_boot_time,
-	.get_rtc_time		= pSeries_get_rtc_time,
-	.set_rtc_time		= pSeries_set_rtc_time,
+	.get_boot_time		= rtas_get_boot_time,
+	.get_rtc_time		= rtas_get_rtc_time,
+	.set_rtc_time		= rtas_set_rtc_time,
 	.calibrate_decr		= generic_calibrate_decr,
 	.progress		= pSeries_progress,
 	.check_legacy_ioport	= pSeries_check_legacy_ioport,
diff --git a/arch/ppc64/kernel/rtc.c b/arch/ppc64/kernel/rtc.c
index de02aedbe08..d729fefa0df 100644
--- a/arch/ppc64/kernel/rtc.c
+++ b/arch/ppc64/kernel/rtc.c
@@ -301,7 +301,7 @@ void iSeries_get_boot_time(struct rtc_time *tm)
 #ifdef CONFIG_PPC_RTAS
 #define MAX_RTC_WAIT 5000	/* 5 sec */
 #define RTAS_CLOCK_BUSY (-2)
-void pSeries_get_boot_time(struct rtc_time *rtc_tm)
+void rtas_get_boot_time(struct rtc_time *rtc_tm)
 {
 	int ret[8];
 	int error, wait_time;
@@ -336,7 +336,7 @@ void pSeries_get_boot_time(struct rtc_time *rtc_tm)
  * and if a delay is needed to read the clock.  In this case we just
  * silently return without updating rtc_tm.
  */
-void pSeries_get_rtc_time(struct rtc_time *rtc_tm)
+void rtas_get_rtc_time(struct rtc_time *rtc_tm)
 {
         int ret[8];
 	int error, wait_time;
@@ -371,7 +371,7 @@ void pSeries_get_rtc_time(struct rtc_time *rtc_tm)
 	rtc_tm->tm_year = ret[0] - 1900;
 }
 
-int pSeries_set_rtc_time(struct rtc_time *tm)
+int rtas_set_rtc_time(struct rtc_time *tm)
 {
 	int error, wait_time;
 	unsigned long max_wait_tb;
diff --git a/include/asm-ppc64/rtas.h b/include/asm-ppc64/rtas.h
index a8ab0e9db84..c46ceb2ffc1 100644
--- a/include/asm-ppc64/rtas.h
+++ b/include/asm-ppc64/rtas.h
@@ -188,6 +188,11 @@ extern int rtas_set_power_level(int powerdomain, int level, int *setlevel);
 extern int rtas_set_indicator(int indicator, int index, int new_value);
 extern void rtas_initialize(void);
 
+struct rtc_time;
+extern void rtas_get_boot_time(struct rtc_time *rtc_time);
+extern void rtas_get_rtc_time(struct rtc_time *rtc_time);
+extern int rtas_set_rtc_time(struct rtc_time *rtc_time);
+
 /* Given an RTAS status code of 9900..9905 compute the hinted delay */
 unsigned int rtas_extended_busy_delay_time(int status);
 static inline int rtas_is_extended_busy(int status)
-- 
cgit v1.2.3-70-g09d2


From 6566c6f1f18d42affe73ccdd403e290b64d10473 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 23 Jun 2005 09:43:28 +1000
Subject: [PATCH] ppc64: pSeries_progress -> rtas_progress

The pSeries_progress function is called from some places in the rtas code,
which may also be used by non-pSeries platforms.
Though pSeries is currently the only platform type that implements
display-character, the code is actually generic enough to be part of
the rtas subsystem.

I hit a bug here because the generic rtas code tried calling ppc_md.progress,
which points to an __init function on most platforms.

We could also clear the ppc_md.progress pointer when freeing the init memory
to make it more explicit that ppc_md.progress must not be called after
bootup.

Signed-off-by: Arnd Bergmann <arndb@de.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/pSeries_setup.c | 103 +-----------------------------------
 arch/ppc64/kernel/rtas-proc.c     |   4 +-
 arch/ppc64/kernel/rtas.c          | 106 +++++++++++++++++++++++++++++++++++++-
 include/asm-ppc64/rtas.h          |   1 +
 4 files changed, 108 insertions(+), 106 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/pSeries_setup.c b/arch/ppc64/kernel/pSeries_setup.c
index c5f3ccac7cd..41e6de2c915 100644
--- a/arch/ppc64/kernel/pSeries_setup.c
+++ b/arch/ppc64/kernel/pSeries_setup.c
@@ -375,107 +375,6 @@ static void __init pSeries_init_early(void)
 }
 
 
-static void pSeries_progress(char *s, unsigned short hex)
-{
-	struct device_node *root;
-	int width, *p;
-	char *os;
-	static int display_character, set_indicator;
-	static int max_width;
-	static DEFINE_SPINLOCK(progress_lock);
-	static int pending_newline = 0;  /* did last write end with unprinted newline? */
-
-	if (!rtas.base)
-		return;
-
-	if (max_width == 0) {
-		if ((root = find_path_device("/rtas")) &&
-		     (p = (unsigned int *)get_property(root,
-						       "ibm,display-line-length",
-						       NULL)))
-			max_width = *p;
-		else
-			max_width = 0x10;
-		display_character = rtas_token("display-character");
-		set_indicator = rtas_token("set-indicator");
-	}
-
-	if (display_character == RTAS_UNKNOWN_SERVICE) {
-		/* use hex display if available */
-		if (set_indicator != RTAS_UNKNOWN_SERVICE)
-			rtas_call(set_indicator, 3, 1, NULL, 6, 0, hex);
-		return;
-	}
-
-	spin_lock(&progress_lock);
-
-	/*
-	 * Last write ended with newline, but we didn't print it since
-	 * it would just clear the bottom line of output. Print it now
-	 * instead.
-	 *
-	 * If no newline is pending, print a CR to start output at the
-	 * beginning of the line.
-	 */
-	if (pending_newline) {
-		rtas_call(display_character, 1, 1, NULL, '\r');
-		rtas_call(display_character, 1, 1, NULL, '\n');
-		pending_newline = 0;
-	} else {
-		rtas_call(display_character, 1, 1, NULL, '\r');
-	}
- 
-	width = max_width;
-	os = s;
-	while (*os) {
-		if (*os == '\n' || *os == '\r') {
-			/* Blank to end of line. */
-			while (width-- > 0)
-				rtas_call(display_character, 1, 1, NULL, ' ');
- 
-			/* If newline is the last character, save it
-			 * until next call to avoid bumping up the
-			 * display output.
-			 */
-			if (*os == '\n' && !os[1]) {
-				pending_newline = 1;
-				spin_unlock(&progress_lock);
-				return;
-			}
- 
-			/* RTAS wants CR-LF, not just LF */
- 
-			if (*os == '\n') {
-				rtas_call(display_character, 1, 1, NULL, '\r');
-				rtas_call(display_character, 1, 1, NULL, '\n');
-			} else {
-				/* CR might be used to re-draw a line, so we'll
-				 * leave it alone and not add LF.
-				 */
-				rtas_call(display_character, 1, 1, NULL, *os);
-			}
- 
-			width = max_width;
-		} else {
-			width--;
-			rtas_call(display_character, 1, 1, NULL, *os);
-		}
- 
-		os++;
- 
-		/* if we overwrite the screen length */
-		if (width <= 0)
-			while ((*os != 0) && (*os != '\n') && (*os != '\r'))
-				os++;
-	}
- 
-	/* Blank to end of line. */
-	while (width-- > 0)
-		rtas_call(display_character, 1, 1, NULL, ' ');
-
-	spin_unlock(&progress_lock);
-}
-
 static int pSeries_check_legacy_ioport(unsigned int baseport)
 {
 	struct device_node *np;
@@ -535,7 +434,7 @@ struct machdep_calls __initdata pSeries_md = {
 	.get_rtc_time		= rtas_get_rtc_time,
 	.set_rtc_time		= rtas_set_rtc_time,
 	.calibrate_decr		= generic_calibrate_decr,
-	.progress		= pSeries_progress,
+	.progress		= rtas_progress,
 	.check_legacy_ioport	= pSeries_check_legacy_ioport,
 	.system_reset_exception = pSeries_system_reset_exception,
 	.machine_check_exception = pSeries_machine_check_exception,
diff --git a/arch/ppc64/kernel/rtas-proc.c b/arch/ppc64/kernel/rtas-proc.c
index 28b1f1521f2..1f3ff860fdf 100644
--- a/arch/ppc64/kernel/rtas-proc.c
+++ b/arch/ppc64/kernel/rtas-proc.c
@@ -371,11 +371,11 @@ static ssize_t ppc_rtas_progress_write(struct file *file,
 	/* Lets see if the user passed hexdigits */
 	hex = simple_strtoul(progress_led, NULL, 10);
 
-	ppc_md.progress ((char *)progress_led, hex);
+	rtas_progress ((char *)progress_led, hex);
 	return count;
 
 	/* clear the line */
-	/* ppc_md.progress("                   ", 0xffff);*/
+	/* rtas_progress("                   ", 0xffff);*/
 }
 /* ****************************************************************** */
 static int ppc_rtas_progress_show(struct seq_file *m, void *v)
diff --git a/arch/ppc64/kernel/rtas.c b/arch/ppc64/kernel/rtas.c
index 5575603def2..43e1518653d 100644
--- a/arch/ppc64/kernel/rtas.c
+++ b/arch/ppc64/kernel/rtas.c
@@ -91,6 +91,108 @@ call_rtas_display_status_delay(unsigned char c)
 	}
 }
 
+void
+rtas_progress(char *s, unsigned short hex)
+{
+	struct device_node *root;
+	int width, *p;
+	char *os;
+	static int display_character, set_indicator;
+	static int max_width;
+	static DEFINE_SPINLOCK(progress_lock);
+	static int pending_newline = 0;  /* did last write end with unprinted newline? */
+
+	if (!rtas.base)
+		return;
+
+	if (max_width == 0) {
+		if ((root = find_path_device("/rtas")) &&
+		     (p = (unsigned int *)get_property(root,
+						       "ibm,display-line-length",
+						       NULL)))
+			max_width = *p;
+		else
+			max_width = 0x10;
+		display_character = rtas_token("display-character");
+		set_indicator = rtas_token("set-indicator");
+	}
+
+	if (display_character == RTAS_UNKNOWN_SERVICE) {
+		/* use hex display if available */
+		if (set_indicator != RTAS_UNKNOWN_SERVICE)
+			rtas_call(set_indicator, 3, 1, NULL, 6, 0, hex);
+		return;
+	}
+
+	spin_lock(&progress_lock);
+
+	/*
+	 * Last write ended with newline, but we didn't print it since
+	 * it would just clear the bottom line of output. Print it now
+	 * instead.
+	 *
+	 * If no newline is pending, print a CR to start output at the
+	 * beginning of the line.
+	 */
+	if (pending_newline) {
+		rtas_call(display_character, 1, 1, NULL, '\r');
+		rtas_call(display_character, 1, 1, NULL, '\n');
+		pending_newline = 0;
+	} else {
+		rtas_call(display_character, 1, 1, NULL, '\r');
+	}
+ 
+	width = max_width;
+	os = s;
+	while (*os) {
+		if (*os == '\n' || *os == '\r') {
+			/* Blank to end of line. */
+			while (width-- > 0)
+				rtas_call(display_character, 1, 1, NULL, ' ');
+ 
+			/* If newline is the last character, save it
+			 * until next call to avoid bumping up the
+			 * display output.
+			 */
+			if (*os == '\n' && !os[1]) {
+				pending_newline = 1;
+				spin_unlock(&progress_lock);
+				return;
+			}
+ 
+			/* RTAS wants CR-LF, not just LF */
+ 
+			if (*os == '\n') {
+				rtas_call(display_character, 1, 1, NULL, '\r');
+				rtas_call(display_character, 1, 1, NULL, '\n');
+			} else {
+				/* CR might be used to re-draw a line, so we'll
+				 * leave it alone and not add LF.
+				 */
+				rtas_call(display_character, 1, 1, NULL, *os);
+			}
+ 
+			width = max_width;
+		} else {
+			width--;
+			rtas_call(display_character, 1, 1, NULL, *os);
+		}
+ 
+		os++;
+ 
+		/* if we overwrite the screen length */
+		if (width <= 0)
+			while ((*os != 0) && (*os != '\n') && (*os != '\r'))
+				os++;
+	}
+ 
+	/* Blank to end of line. */
+	while (width-- > 0)
+		rtas_call(display_character, 1, 1, NULL, ' ');
+
+	spin_unlock(&progress_lock);
+}
+
 int
 rtas_token(const char *service)
 {
@@ -425,8 +527,8 @@ rtas_flash_firmware(void)
 
 	printk(KERN_ALERT "FLASH: flash image is %ld bytes\n", image_size);
 	printk(KERN_ALERT "FLASH: performing flash and reboot\n");
-	ppc_md.progress("Flashing        \n", 0x0);
-	ppc_md.progress("Please Wait...  ", 0x0);
+	rtas_progress("Flashing        \n", 0x0);
+	rtas_progress("Please Wait...  ", 0x0);
 	printk(KERN_ALERT "FLASH: this will take several minutes.  Do not power off!\n");
 	status = rtas_call(update_token, 1, 1, NULL, rtas_block_list);
 	switch (status) {	/* should only get "bad" status */
diff --git a/include/asm-ppc64/rtas.h b/include/asm-ppc64/rtas.h
index c46ceb2ffc1..e7d1b522280 100644
--- a/include/asm-ppc64/rtas.h
+++ b/include/asm-ppc64/rtas.h
@@ -186,6 +186,7 @@ extern int rtas_get_sensor(int sensor, int index, int *state);
 extern int rtas_get_power_level(int powerdomain, int *level);
 extern int rtas_set_power_level(int powerdomain, int level, int *setlevel);
 extern int rtas_set_indicator(int indicator, int index, int new_value);
+extern void rtas_progress(char *s, unsigned short hex);
 extern void rtas_initialize(void);
 
 struct rtc_time;
-- 
cgit v1.2.3-70-g09d2


From 5f5b4e669a59be1cf8fc9d6d04ff1ccad8ab6de0 Mon Sep 17 00:00:00 2001
From: Utz Bacher <utz.bacher@de.ibm.com>
Date: Thu, 23 Jun 2005 09:43:31 +1000
Subject: [PATCH] ppc64: add a minimal nvram driver

The firmware provides the location and size of the nvram
in the device tree, so it does not really contain any
hardware specific bits and could be used on other
machines as well.

Signed-off-by: Arnd Bergmann <arndb@de.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/bpa_nvram.c | 118 ++++++++++++++++++++++++++++++++++++++++++
 include/asm-ppc64/nvram.h     |   1 +
 2 files changed, 119 insertions(+)
 create mode 100644 arch/ppc64/kernel/bpa_nvram.c

(limited to 'include')

diff --git a/arch/ppc64/kernel/bpa_nvram.c b/arch/ppc64/kernel/bpa_nvram.c
new file mode 100644
index 00000000000..06a119cfceb
--- /dev/null
+++ b/arch/ppc64/kernel/bpa_nvram.c
@@ -0,0 +1,118 @@
+/*
+ * NVRAM for CPBW
+ *
+ * (C) Copyright IBM Corp. 2005
+ *
+ * Authors : Utz Bacher <utz.bacher@de.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2, or (at your option)
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/spinlock.h>
+#include <linux/types.h>
+
+#include <asm/machdep.h>
+#include <asm/nvram.h>
+#include <asm/prom.h>
+
+static void __iomem *bpa_nvram_start;
+static long bpa_nvram_len;
+static spinlock_t bpa_nvram_lock = SPIN_LOCK_UNLOCKED;
+
+static ssize_t bpa_nvram_read(char *buf, size_t count, loff_t *index)
+{
+	unsigned long flags;
+
+	if (*index >= bpa_nvram_len)
+		return 0;
+	if (*index + count > bpa_nvram_len)
+		count = bpa_nvram_len - *index;
+
+	spin_lock_irqsave(&bpa_nvram_lock, flags);
+
+	memcpy_fromio(buf, bpa_nvram_start + *index, count);
+
+	spin_unlock_irqrestore(&bpa_nvram_lock, flags);
+	
+	*index += count;
+	return count;
+}
+
+static ssize_t bpa_nvram_write(char *buf, size_t count, loff_t *index)
+{
+	unsigned long flags;
+
+	if (*index >= bpa_nvram_len)
+		return 0;
+	if (*index + count > bpa_nvram_len)
+		count = bpa_nvram_len - *index;
+
+	spin_lock_irqsave(&bpa_nvram_lock, flags);
+
+	memcpy_toio(bpa_nvram_start + *index, buf, count);
+
+	spin_unlock_irqrestore(&bpa_nvram_lock, flags);
+	
+	*index += count;
+	return count;
+}
+
+static ssize_t bpa_nvram_get_size(void)
+{
+	return bpa_nvram_len;
+}
+
+int __init bpa_nvram_init(void)
+{
+	struct device_node *nvram_node;
+	unsigned long *buffer;
+	int proplen;
+	unsigned long nvram_addr;
+	int ret;
+
+	ret = -ENODEV;
+	nvram_node = of_find_node_by_type(NULL, "nvram");
+	if (!nvram_node)
+		goto out;
+
+	ret = -EIO;
+	buffer = (unsigned long *)get_property(nvram_node, "reg", &proplen);
+	if (proplen != 2*sizeof(unsigned long))
+		goto out;
+
+	ret = -ENODEV;
+	nvram_addr = buffer[0];
+	bpa_nvram_len = buffer[1];
+	if ( (!bpa_nvram_len) || (!nvram_addr) )
+		goto out;
+
+	bpa_nvram_start = ioremap(nvram_addr, bpa_nvram_len);
+	if (!bpa_nvram_start)
+		goto out;
+
+	printk(KERN_INFO "BPA NVRAM, %luk mapped to %p\n",
+	       bpa_nvram_len >> 10, bpa_nvram_start);
+
+	ppc_md.nvram_read	= bpa_nvram_read;
+	ppc_md.nvram_write	= bpa_nvram_write;
+	ppc_md.nvram_size	= bpa_nvram_get_size;
+
+out:
+	of_node_put(nvram_node);
+	return ret;
+}
diff --git a/include/asm-ppc64/nvram.h b/include/asm-ppc64/nvram.h
index 4e6dd370d93..dfaa21566c9 100644
--- a/include/asm-ppc64/nvram.h
+++ b/include/asm-ppc64/nvram.h
@@ -70,6 +70,7 @@ extern struct nvram_partition *nvram_find_partition(int sig, const char *name);
 
 extern int pSeries_nvram_init(void);
 extern int pmac_nvram_init(void);
+extern int bpa_nvram_init(void);
 
 /* PowerMac specific nvram stuffs */
 
-- 
cgit v1.2.3-70-g09d2


From fef1c772fa154c16e0a54577e9ecb5480f7b937e Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 23 Jun 2005 09:43:37 +1000
Subject: [PATCH] ppc64: add BPA platform type

This adds the basic support for running on BPA machines.
So far, this is only the IBM workstation, and it will
not run on others without a little more generalization.

It should be possible to configure a kernel for any
combination of CONFIG_PPC_BPA with any of the other
multiplatform targets.

Signed-off-by: Arnd Bergmann <arndb@de.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 MAINTAINERS                          |   7 ++
 arch/ppc64/Kconfig                   |   6 +-
 arch/ppc64/Makefile                  |   2 +
 arch/ppc64/kernel/Makefile           |   3 +
 arch/ppc64/kernel/bpa_setup.c        | 135 +++++++++++++++++++++++++++++++++++
 arch/ppc64/kernel/cpu_setup_power4.S |  16 ++++-
 arch/ppc64/kernel/cputable.c         |  11 +++
 arch/ppc64/kernel/irq.c              |   3 +
 arch/ppc64/kernel/proc_ppc64.c       |   2 +-
 arch/ppc64/kernel/prom_init.c        |   4 +-
 arch/ppc64/kernel/setup.c            |   4 ++
 arch/ppc64/kernel/traps.c            |   4 ++
 include/asm-ppc64/mmu.h              |   5 +-
 include/asm-ppc64/processor.h        |  15 +++-
 include/asm-ppc64/smp.h              |   8 +++
 15 files changed, 216 insertions(+), 9 deletions(-)
 create mode 100644 arch/ppc64/kernel/bpa_setup.c

(limited to 'include')

diff --git a/MAINTAINERS b/MAINTAINERS
index 4d44824884e..5eaa6807cdb 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -504,6 +504,13 @@ L:   bonding-devel@lists.sourceforge.net
 W:   http://sourceforge.net/projects/bonding/
 S:   Supported
 
+BROADBAND PROCESSOR ARCHITECTURE
+P:	Arnd Bergmann
+M:	arnd@arndb.de
+L:	linuxppc64-dev@ozlabs.org
+W:	http://linuxppc64.org
+S:	Supported
+
 BTTV VIDEO4LINUX DRIVER
 P:	Gerd Knorr
 M:	kraxel@bytesex.org
diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig
index 0f1fa289744..c7f2f0a4d85 100644
--- a/arch/ppc64/Kconfig
+++ b/arch/ppc64/Kconfig
@@ -77,6 +77,10 @@ config PPC_PSERIES
 	bool "  IBM pSeries & new iSeries"
 	default y
 
+config PPC_BPA
+	bool "  Broadband Processor Architecture"
+	depends on PPC_MULTIPLATFORM
+
 config PPC_PMAC
 	depends on PPC_MULTIPLATFORM
 	bool "  Apple G5 based machines"
@@ -256,7 +260,7 @@ config MSCHUNKS
 
 config PPC_RTAS
 	bool
-	depends on PPC_PSERIES
+	depends on PPC_PSERIES || PPC_BPA
 	default y
 
 config RTAS_PROC
diff --git a/arch/ppc64/Makefile b/arch/ppc64/Makefile
index 33c752ceca4..731b8475833 100644
--- a/arch/ppc64/Makefile
+++ b/arch/ppc64/Makefile
@@ -90,12 +90,14 @@ boot := arch/ppc64/boot
 boottarget-$(CONFIG_PPC_PSERIES) := zImage zImage.initrd
 boottarget-$(CONFIG_PPC_MAPLE) := zImage zImage.initrd
 boottarget-$(CONFIG_PPC_ISERIES) := vmlinux.sminitrd vmlinux.initrd vmlinux.sm
+boottarget-$(CONFIG_PPC_BPA) := zImage zImage.initrd
 $(boottarget-y): vmlinux
 	$(Q)$(MAKE) $(build)=$(boot) $(boot)/$@
 
 bootimage-$(CONFIG_PPC_PSERIES) := $(boot)/zImage
 bootimage-$(CONFIG_PPC_PMAC) := vmlinux
 bootimage-$(CONFIG_PPC_MAPLE) := $(boot)/zImage
+bootimage-$(CONFIG_PPC_BPA) := zImage
 bootimage-$(CONFIG_PPC_ISERIES) := vmlinux
 BOOTIMAGE := $(bootimage-y)
 install: vmlinux
diff --git a/arch/ppc64/kernel/Makefile b/arch/ppc64/kernel/Makefile
index f389f2453da..c89983ab909 100644
--- a/arch/ppc64/kernel/Makefile
+++ b/arch/ppc64/kernel/Makefile
@@ -33,6 +33,8 @@ obj-$(CONFIG_PPC_PSERIES) += pSeries_pci.o pSeries_lpar.o pSeries_hvCall.o \
 			     pSeries_nvram.o rtasd.o ras.o pSeries_reconfig.o \
 			     xics.o pSeries_setup.o pSeries_iommu.o
 
+obj-$(CONFIG_PPC_BPA) += bpa_setup.o bpa_nvram.o
+
 obj-$(CONFIG_EEH)		+= eeh.o
 obj-$(CONFIG_PROC_FS)		+= proc_ppc64.o
 obj-$(CONFIG_RTAS_FLASH)	+= rtas_flash.o
@@ -59,6 +61,7 @@ ifdef CONFIG_SMP
 obj-$(CONFIG_PPC_PMAC)		+= pmac_smp.o smp-tbsync.o
 obj-$(CONFIG_PPC_ISERIES)	+= iSeries_smp.o
 obj-$(CONFIG_PPC_PSERIES)	+= pSeries_smp.o
+obj-$(CONFIG_PPC_BPA)		+= pSeries_smp.o
 obj-$(CONFIG_PPC_MAPLE)		+= smp-tbsync.o
 endif
 
diff --git a/arch/ppc64/kernel/bpa_setup.c b/arch/ppc64/kernel/bpa_setup.c
new file mode 100644
index 00000000000..d1992dd2d61
--- /dev/null
+++ b/arch/ppc64/kernel/bpa_setup.c
@@ -0,0 +1,135 @@
+/*
+ *  linux/arch/ppc/kernel/bpa_setup.c
+ *
+ *  Copyright (C) 1995  Linus Torvalds
+ *  Adapted from 'alpha' version by Gary Thomas
+ *  Modified by Cort Dougan (cort@cs.nmt.edu)
+ *  Modified by PPC64 Team, IBM Corp
+ *  Modified by BPA Team, IBM Deutschland Entwicklung GmbH
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+#undef DEBUG
+
+#include <linux/config.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/stddef.h>
+#include <linux/unistd.h>
+#include <linux/slab.h>
+#include <linux/user.h>
+#include <linux/reboot.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/irq.h>
+#include <linux/seq_file.h>
+#include <linux/root_dev.h>
+#include <linux/console.h>
+
+#include <asm/mmu.h>
+#include <asm/processor.h>
+#include <asm/io.h>
+#include <asm/pgtable.h>
+#include <asm/prom.h>
+#include <asm/rtas.h>
+#include <asm/pci-bridge.h>
+#include <asm/iommu.h>
+#include <asm/dma.h>
+#include <asm/machdep.h>
+#include <asm/time.h>
+#include <asm/nvram.h>
+#include <asm/cputable.h>
+
+#include "pci.h"
+
+#ifdef DEBUG
+#define DBG(fmt...) udbg_printf(fmt)
+#else
+#define DBG(fmt...)
+#endif
+
+void bpa_get_cpuinfo(struct seq_file *m)
+{
+	struct device_node *root;
+	const char *model = "";
+
+	root = of_find_node_by_path("/");
+	if (root)
+		model = get_property(root, "model", NULL);
+	seq_printf(m, "machine\t\t: BPA %s\n", model);
+	of_node_put(root);
+}
+
+static void bpa_progress(char *s, unsigned short hex)
+{
+	printk("*** %04x : %s\n", hex, s ? s : "");
+}
+
+static void __init bpa_setup_arch(void)
+{
+#ifdef CONFIG_SMP
+	smp_init_pSeries();
+#endif
+
+	/* init to some ~sane value until calibrate_delay() runs */
+	loops_per_jiffy = 50000000;
+
+	if (ROOT_DEV == 0) {
+		printk("No ramdisk, default root is /dev/hda2\n");
+		ROOT_DEV = Root_HDA2;
+	}
+
+	/* Find and initialize PCI host bridges */
+	init_pci_config_tokens();
+	find_and_init_phbs();
+
+#ifdef CONFIG_DUMMY_CONSOLE
+	conswitchp = &dummy_con;
+#endif
+
+	bpa_nvram_init();
+}
+
+/*
+ * Early initialization.  Relocation is on but do not reference unbolted pages
+ */
+static void __init bpa_init_early(void)
+{
+	DBG(" -> bpa_init_early()\n");
+
+	hpte_init_native();
+
+	pci_direct_iommu_init();
+
+	ppc64_interrupt_controller = IC_BPA_IIC;
+
+	DBG(" <- bpa_init_early()\n");
+}
+
+
+static int __init bpa_probe(int platform)
+{
+	if (platform != PLATFORM_BPA)
+		return 0;
+
+	return 1;
+}
+
+struct machdep_calls __initdata bpa_md = {
+	.probe			= bpa_probe,
+	.setup_arch		= bpa_setup_arch,
+	.init_early		= bpa_init_early,
+	.get_cpuinfo		= bpa_get_cpuinfo,
+	.restart		= rtas_restart,
+	.power_off		= rtas_power_off,
+	.halt			= rtas_halt,
+	.get_boot_time		= rtas_get_boot_time,
+	.get_rtc_time		= rtas_get_rtc_time,
+	.set_rtc_time		= rtas_set_rtc_time,
+	.calibrate_decr		= generic_calibrate_decr,
+	.progress		= bpa_progress,
+};
diff --git a/arch/ppc64/kernel/cpu_setup_power4.S b/arch/ppc64/kernel/cpu_setup_power4.S
index 3bd95182085..42fc08cf87a 100644
--- a/arch/ppc64/kernel/cpu_setup_power4.S
+++ b/arch/ppc64/kernel/cpu_setup_power4.S
@@ -73,7 +73,21 @@ _GLOBAL(__970_cpu_preinit)
 
 _GLOBAL(__setup_cpu_power4)
 	blr
-	
+
+_GLOBAL(__setup_cpu_be)
+        /* Set large page sizes LP=0: 16MB, LP=1: 64KB */
+        addi    r3, 0,  0
+        ori     r3, r3, HID6_LB
+        sldi    r3, r3, 32
+        nor     r3, r3, r3
+        mfspr   r4, SPRN_HID6
+        and     r4, r4, r3
+        addi    r3, 0, 0x02000
+        sldi    r3, r3, 32
+        or      r4, r4, r3
+        mtspr   SPRN_HID6, r4
+	blr
+
 _GLOBAL(__setup_cpu_ppc970)
 	mfspr	r0,SPRN_HID0
 	li	r11,5			/* clear DOZE and SLEEP */
diff --git a/arch/ppc64/kernel/cputable.c b/arch/ppc64/kernel/cputable.c
index 8644a864805..1d162c7c59d 100644
--- a/arch/ppc64/kernel/cputable.c
+++ b/arch/ppc64/kernel/cputable.c
@@ -34,6 +34,7 @@ EXPORT_SYMBOL(cur_cpu_spec);
 extern void __setup_cpu_power3(unsigned long offset, struct cpu_spec* spec);
 extern void __setup_cpu_power4(unsigned long offset, struct cpu_spec* spec);
 extern void __setup_cpu_ppc970(unsigned long offset, struct cpu_spec* spec);
+extern void __setup_cpu_be(unsigned long offset, struct cpu_spec* spec);
 
 
 /* We only set the altivec features if the kernel was compiled with altivec
@@ -162,6 +163,16 @@ struct cpu_spec	cpu_specs[] = {
 	    __setup_cpu_power4,
 	    COMMON_PPC64_FW
     },
+    {	/* BE DD1.x  */
+	    0xffff0000, 0x00700000, "Broadband Engine",
+	    CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE |
+		    CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_ALTIVEC_COMP |
+		    CPU_FTR_SMT,
+	    COMMON_USER_PPC64 | PPC_FEATURE_HAS_ALTIVEC_COMP,
+	    128, 128,
+	    __setup_cpu_be,
+	    COMMON_PPC64_FW
+    },
     {	/* default match */
 	    0x00000000, 0x00000000, "POWER4 (compatible)",
   	    CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE |
diff --git a/arch/ppc64/kernel/irq.c b/arch/ppc64/kernel/irq.c
index d860467b8f0..3defc8c33ad 100644
--- a/arch/ppc64/kernel/irq.c
+++ b/arch/ppc64/kernel/irq.c
@@ -395,6 +395,9 @@ int virt_irq_create_mapping(unsigned int real_irq)
 	if (ppc64_interrupt_controller == IC_OPEN_PIC)
 		return real_irq;	/* no mapping for openpic (for now) */
 
+	if (ppc64_interrupt_controller == IC_BPA_IIC)
+		return real_irq;	/* no mapping for iic either */
+
 	/* don't map interrupts < MIN_VIRT_IRQ */
 	if (real_irq < MIN_VIRT_IRQ) {
 		virt_irq_to_real_map[real_irq] = real_irq;
diff --git a/arch/ppc64/kernel/proc_ppc64.c b/arch/ppc64/kernel/proc_ppc64.c
index 0914b0669b0..a87c66a9652 100644
--- a/arch/ppc64/kernel/proc_ppc64.c
+++ b/arch/ppc64/kernel/proc_ppc64.c
@@ -53,7 +53,7 @@ static int __init proc_ppc64_create(void)
 	if (!root)
 		return 1;
 
-	if (!(systemcfg->platform & PLATFORM_PSERIES))
+	if (!(systemcfg->platform & (PLATFORM_PSERIES | PLATFORM_BPA)))
 		return 0;
 
 	if (!proc_mkdir("rtas", root))
diff --git a/arch/ppc64/kernel/prom_init.c b/arch/ppc64/kernel/prom_init.c
index b7683abfbe6..e248a7950ae 100644
--- a/arch/ppc64/kernel/prom_init.c
+++ b/arch/ppc64/kernel/prom_init.c
@@ -1915,9 +1915,9 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4, unsigned long
 		prom_send_capabilities();
 
 	/*
-	 * On pSeries, copy the CPU hold code
+	 * On pSeries and BPA, copy the CPU hold code
 	 */
-       	if (RELOC(of_platform) & PLATFORM_PSERIES)
+       	if (RELOC(of_platform) & (PLATFORM_PSERIES | PLATFORM_BPA))
        		copy_and_flush(0, KERNELBASE - offset, 0x100, 0);
 
 	/*
diff --git a/arch/ppc64/kernel/setup.c b/arch/ppc64/kernel/setup.c
index 93b0ee88cda..10222008fe2 100644
--- a/arch/ppc64/kernel/setup.c
+++ b/arch/ppc64/kernel/setup.c
@@ -344,6 +344,7 @@ static void __init setup_cpu_maps(void)
 extern struct machdep_calls pSeries_md;
 extern struct machdep_calls pmac_md;
 extern struct machdep_calls maple_md;
+extern struct machdep_calls bpa_md;
 
 /* Ultimately, stuff them in an elf section like initcalls... */
 static struct machdep_calls __initdata *machines[] = {
@@ -356,6 +357,9 @@ static struct machdep_calls __initdata *machines[] = {
 #ifdef CONFIG_PPC_MAPLE
 	&maple_md,
 #endif /* CONFIG_PPC_MAPLE */
+#ifdef CONFIG_PPC_BPA
+	&bpa_md,
+#endif
 	NULL
 };
 
diff --git a/arch/ppc64/kernel/traps.c b/arch/ppc64/kernel/traps.c
index 7e52cb2605e..a8d5e83ee89 100644
--- a/arch/ppc64/kernel/traps.c
+++ b/arch/ppc64/kernel/traps.c
@@ -126,6 +126,10 @@ int die(const char *str, struct pt_regs *regs, long err)
 			printk("POWERMAC ");
 			nl = 1;
 			break;
+		case PLATFORM_BPA:
+			printk("BPA ");
+			nl = 1;
+			break;
 	}
 	if (nl)
 		printk("\n");
diff --git a/include/asm-ppc64/mmu.h b/include/asm-ppc64/mmu.h
index c78282a67d8..9d03a98a4fa 100644
--- a/include/asm-ppc64/mmu.h
+++ b/include/asm-ppc64/mmu.h
@@ -47,9 +47,10 @@
 #define SLB_VSID_KS		ASM_CONST(0x0000000000000800)
 #define SLB_VSID_KP		ASM_CONST(0x0000000000000400)
 #define SLB_VSID_N		ASM_CONST(0x0000000000000200) /* no-execute */
-#define SLB_VSID_L		ASM_CONST(0x0000000000000100) /* largepage 16M */
+#define SLB_VSID_L		ASM_CONST(0x0000000000000100) /* largepage */
 #define SLB_VSID_C		ASM_CONST(0x0000000000000080) /* class */
-
+#define SLB_VSID_LS		ASM_CONST(0x0000000000000070) /* size of largepage */
+ 
 #define SLB_VSID_KERNEL		(SLB_VSID_KP|SLB_VSID_C)
 #define SLB_VSID_USER		(SLB_VSID_KP|SLB_VSID_KS)
 
diff --git a/include/asm-ppc64/processor.h b/include/asm-ppc64/processor.h
index 3084099086a..af28aa55d8c 100644
--- a/include/asm-ppc64/processor.h
+++ b/include/asm-ppc64/processor.h
@@ -138,8 +138,16 @@
 #define	SPRN_NIADORM	0x3F3	/* Hardware Implementation Register 2 */
 #define SPRN_HID4	0x3F4	/* 970 HID4 */
 #define SPRN_HID5	0x3F6	/* 970 HID5 */
-#define	SPRN_TSC 	0x3FD	/* Thread switch control */
-#define	SPRN_TST 	0x3FC	/* Thread switch timeout */
+#define	SPRN_HID6	0x3F9	/* BE HID 6 */
+#define	  HID6_LB	(0x0F<<12) /* Concurrent Large Page Modes */
+#define	  HID6_DLP	(1<<20)	/* Disable all large page modes (4K only) */
+#define	SPRN_TSCR	0x399   /* Thread switch control on BE */
+#define	SPRN_TTR	0x39A   /* Thread switch timeout on BE */
+#define	  TSCR_DEC_ENABLE	0x200000 /* Decrementer Interrupt */
+#define	  TSCR_EE_ENABLE	0x100000 /* External Interrupt */
+#define	  TSCR_EE_BOOST		0x080000 /* External Interrupt Boost */
+#define	SPRN_TSC 	0x3FD	/* Thread switch control on others */
+#define	SPRN_TST 	0x3FC	/* Thread switch timeout on others */
 #define	SPRN_L2CR	0x3F9	/* Level 2 Cache Control Regsiter */
 #define	SPRN_LR		0x008	/* Link Register */
 #define	SPRN_PIR	0x3FF	/* Processor Identification Register */
@@ -259,6 +267,7 @@
 #define PV_970FX	0x003C
 #define	PV_630        	0x0040
 #define	PV_630p	        0x0041
+#define	PV_BE		0x0070
 
 /* Platforms supported by PPC64 */
 #define PLATFORM_PSERIES      0x0100
@@ -267,6 +276,7 @@
 #define PLATFORM_LPAR         0x0001
 #define PLATFORM_POWERMAC     0x0400
 #define PLATFORM_MAPLE        0x0500
+#define PLATFORM_BPA          0x1000
 
 /* Compatibility with drivers coming from PPC32 world */
 #define _machine	(systemcfg->platform)
@@ -278,6 +288,7 @@
 #define IC_INVALID    0
 #define IC_OPEN_PIC   1
 #define IC_PPC_XIC    2
+#define IC_BPA_IIC    3
 
 #define XGLUE(a,b) a##b
 #define GLUE(a,b) XGLUE(a,b)
diff --git a/include/asm-ppc64/smp.h b/include/asm-ppc64/smp.h
index 8115ecb8fee..d86f742e9a2 100644
--- a/include/asm-ppc64/smp.h
+++ b/include/asm-ppc64/smp.h
@@ -85,6 +85,14 @@ extern void smp_generic_take_timebase(void);
 
 extern struct smp_ops_t *smp_ops;
 
+#ifdef CONFIG_PPC_PSERIES
+void vpa_init(int cpu);
+#else
+static inline void vpa_init(int cpu)
+{
+}
+#endif /* CONFIG_PPC_PSERIES */
+
 #endif /* __ASSEMBLY__ */
 
 #endif /* !(_PPC64_SMP_H) */
-- 
cgit v1.2.3-70-g09d2


From 6ca4f65e6b390d09e1de7280cf9fd4f5d8e4b48b Mon Sep 17 00:00:00 2001
From: Jeff Moyer <jmoyer@redhat.com>
Date: Wed, 22 Jun 2005 22:04:55 -0700
Subject: [NETPOLL]: Set poll_owner to -1 before unlocking in
 netpoll_poll_unlock()

This trivial patch moves the assignment of poll_owner to -1 inside of
the lock.  This fixes a potential SMP race in the code.

Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netpoll.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h
index c0d8b90c520..449a4fde658 100644
--- a/include/linux/netpoll.h
+++ b/include/linux/netpoll.h
@@ -53,8 +53,8 @@ static inline void netpoll_poll_lock(struct net_device *dev)
 static inline void netpoll_poll_unlock(struct net_device *dev)
 {
 	if (dev->np) {
-		spin_unlock(&dev->np->poll_lock);
 		dev->np->poll_owner = -1;
+		spin_unlock(&dev->np->poll_lock);
 	}
 }
 
-- 
cgit v1.2.3-70-g09d2


From 115c1d6e61b70851d9a363328c3b8d4c2559a1d3 Mon Sep 17 00:00:00 2001
From: Jeff Moyer <jmoyer@redhat.com>
Date: Wed, 22 Jun 2005 22:05:31 -0700
Subject: [NETPOLL]: Introduce a netpoll_info struct

This patch introduces a netpoll_info structure, which the struct net_device
will now point to instead of pointing to a struct netpoll.  The reason for
this is two-fold: 1) fields such as the rx_flags, poll_owner, and poll_lock
should be maintained per net_device, not per netpoll;  and 2) this is a first
step in providing support for multiple netpoll clients to register against the
same net_device.

The struct netpoll is now pointed to by the netpoll_info structure.  As
such, the previous behaviour of the code is preserved.

Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  4 ++--
 include/linux/netpoll.h   | 25 ++++++++++++++-------
 net/core/netpoll.c        | 57 +++++++++++++++++++++++++++++++----------------
 3 files changed, 57 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index ba5d1236aa1..d6afd440cf7 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -41,7 +41,7 @@
 struct divert_blk;
 struct vlan_group;
 struct ethtool_ops;
-struct netpoll;
+struct netpoll_info;
 					/* source back-compat hooks */
 #define SET_ETHTOOL_OPS(netdev,ops) \
 	( (netdev)->ethtool_ops = (ops) )
@@ -468,7 +468,7 @@ struct net_device
 						     unsigned char *haddr);
 	int			(*neigh_setup)(struct net_device *dev, struct neigh_parms *);
 #ifdef CONFIG_NETPOLL
-	struct netpoll		*np;
+	struct netpoll_info	*npinfo;
 #endif
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	void                    (*poll_controller)(struct net_device *dev);
diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h
index 449a4fde658..388cd91bc7a 100644
--- a/include/linux/netpoll.h
+++ b/include/linux/netpoll.h
@@ -16,14 +16,18 @@ struct netpoll;
 struct netpoll {
 	struct net_device *dev;
 	char dev_name[16], *name;
-	int rx_flags;
 	void (*rx_hook)(struct netpoll *, int, char *, int);
 	void (*drop)(struct sk_buff *skb);
 	u32 local_ip, remote_ip;
 	u16 local_port, remote_port;
 	unsigned char local_mac[6], remote_mac[6];
+};
+
+struct netpoll_info {
 	spinlock_t poll_lock;
 	int poll_owner;
+	int rx_flags;
+	struct netpoll *np;
 };
 
 void netpoll_poll(struct netpoll *np);
@@ -39,22 +43,27 @@ void netpoll_queue(struct sk_buff *skb);
 #ifdef CONFIG_NETPOLL
 static inline int netpoll_rx(struct sk_buff *skb)
 {
-	return skb->dev->np && skb->dev->np->rx_flags && __netpoll_rx(skb);
+	struct netpoll_info *npinfo = skb->dev->npinfo;
+
+	if (!npinfo || !npinfo->rx_flags)
+		return 0;
+
+	return npinfo->np && __netpoll_rx(skb);
 }
 
 static inline void netpoll_poll_lock(struct net_device *dev)
 {
-	if (dev->np) {
-		spin_lock(&dev->np->poll_lock);
-		dev->np->poll_owner = smp_processor_id();
+	if (dev->npinfo) {
+		spin_lock(&dev->npinfo->poll_lock);
+		dev->npinfo->poll_owner = smp_processor_id();
 	}
 }
 
 static inline void netpoll_poll_unlock(struct net_device *dev)
 {
-	if (dev->np) {
-		dev->np->poll_owner = -1;
-		spin_unlock(&dev->np->poll_lock);
+	if (dev->npinfo) {
+		dev->npinfo->poll_owner = -1;
+		spin_unlock(&dev->npinfo->poll_lock);
 	}
 }
 
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index a119696d552..ab3c0c9713b 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -130,19 +130,20 @@ static int checksum_udp(struct sk_buff *skb, struct udphdr *uh,
  */
 static void poll_napi(struct netpoll *np)
 {
+	struct netpoll_info *npinfo = np->dev->npinfo;
 	int budget = 16;
 
 	if (test_bit(__LINK_STATE_RX_SCHED, &np->dev->state) &&
-	    np->poll_owner != smp_processor_id() &&
-	    spin_trylock(&np->poll_lock)) {
-		np->rx_flags |= NETPOLL_RX_DROP;
+	    npinfo->poll_owner != smp_processor_id() &&
+	    spin_trylock(&npinfo->poll_lock)) {
+		npinfo->rx_flags |= NETPOLL_RX_DROP;
 		atomic_inc(&trapped);
 
 		np->dev->poll(np->dev, &budget);
 
 		atomic_dec(&trapped);
-		np->rx_flags &= ~NETPOLL_RX_DROP;
-		spin_unlock(&np->poll_lock);
+		npinfo->rx_flags &= ~NETPOLL_RX_DROP;
+		spin_unlock(&npinfo->poll_lock);
 	}
 }
 
@@ -245,6 +246,7 @@ repeat:
 static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb)
 {
 	int status;
+	struct netpoll_info *npinfo;
 
 repeat:
 	if(!np || !np->dev || !netif_running(np->dev)) {
@@ -253,8 +255,9 @@ repeat:
 	}
 
 	/* avoid recursion */
-	if(np->poll_owner == smp_processor_id() ||
-	   np->dev->xmit_lock_owner == smp_processor_id()) {
+	npinfo = np->dev->npinfo;
+	if (npinfo->poll_owner == smp_processor_id() ||
+	    np->dev->xmit_lock_owner == smp_processor_id()) {
 		if (np->drop)
 			np->drop(skb);
 		else
@@ -341,14 +344,18 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len)
 
 static void arp_reply(struct sk_buff *skb)
 {
+	struct netpoll_info *npinfo = skb->dev->npinfo;
 	struct arphdr *arp;
 	unsigned char *arp_ptr;
 	int size, type = ARPOP_REPLY, ptype = ETH_P_ARP;
 	u32 sip, tip;
 	struct sk_buff *send_skb;
-	struct netpoll *np = skb->dev->np;
+	struct netpoll *np = NULL;
 
-	if (!np) return;
+	if (npinfo)
+		np = npinfo->np;
+	if (!np)
+		return;
 
 	/* No arp on this interface */
 	if (skb->dev->flags & IFF_NOARP)
@@ -429,7 +436,7 @@ int __netpoll_rx(struct sk_buff *skb)
 	int proto, len, ulen;
 	struct iphdr *iph;
 	struct udphdr *uh;
-	struct netpoll *np = skb->dev->np;
+	struct netpoll *np = skb->dev->npinfo->np;
 
 	if (!np->rx_hook)
 		goto out;
@@ -611,9 +618,7 @@ int netpoll_setup(struct netpoll *np)
 {
 	struct net_device *ndev = NULL;
 	struct in_device *in_dev;
-
-	np->poll_lock = SPIN_LOCK_UNLOCKED;
-	np->poll_owner = -1;
+	struct netpoll_info *npinfo;
 
 	if (np->dev_name)
 		ndev = dev_get_by_name(np->dev_name);
@@ -624,7 +629,16 @@ int netpoll_setup(struct netpoll *np)
 	}
 
 	np->dev = ndev;
-	ndev->np = np;
+	if (!ndev->npinfo) {
+		npinfo = kmalloc(sizeof(*npinfo), GFP_KERNEL);
+		if (!npinfo)
+			goto release;
+
+		npinfo->np = NULL;
+		npinfo->poll_lock = SPIN_LOCK_UNLOCKED;
+		npinfo->poll_owner = -1;
+	} else
+		npinfo = ndev->npinfo;
 
 	if (!ndev->poll_controller) {
 		printk(KERN_ERR "%s: %s doesn't support polling, aborting.\n",
@@ -693,12 +707,15 @@ int netpoll_setup(struct netpoll *np)
 	}
 
 	if(np->rx_hook)
-		np->rx_flags = NETPOLL_RX_ENABLED;
+		npinfo->rx_flags = NETPOLL_RX_ENABLED;
+	npinfo->np = np;
+	ndev->npinfo = npinfo;
 
 	return 0;
 
  release:
-	ndev->np = NULL;
+	if (!ndev->npinfo)
+		kfree(npinfo);
 	np->dev = NULL;
 	dev_put(ndev);
 	return -1;
@@ -706,9 +723,11 @@ int netpoll_setup(struct netpoll *np)
 
 void netpoll_cleanup(struct netpoll *np)
 {
-	if (np->dev)
-		np->dev->np = NULL;
-	dev_put(np->dev);
+	if (np->dev) {
+		if (np->dev->npinfo)
+			np->dev->npinfo->np = NULL;
+		dev_put(np->dev);
+	}
 	np->dev = NULL;
 }
 
-- 
cgit v1.2.3-70-g09d2


From fbeec2e1552949002065435c9829dc244ad85407 Mon Sep 17 00:00:00 2001
From: Jeff Moyer <jmoyer@redhat.com>
Date: Wed, 22 Jun 2005 22:05:59 -0700
Subject: [NETPOLL]: allow multiple netpoll_clients to register against one
 interface

This patch provides support for registering multiple netpoll clients to the
same network device.  Only one of these clients may register an rx_hook,
however.  In practice, this restriction has not been problematic.  It is
worth mentioning, though, that the current design can be easily extended to
allow for the registration of multiple rx_hooks.

The basic idea of the patch is that the rx_np pointer in the netpoll_info
structure points to the struct netpoll that has rx_hook filled in.  Aside
from this one case, there is no need for a pointer from the struct
net_device to an individual struct netpoll.

A lock is introduced to protect the setting and clearing of the np_rx
pointer.  The pointer will only be cleared upon netpoll client module
removal, and the lock should be uncontested.

Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netpoll.h | 15 ++++++++++++---
 net/core/netpoll.c      | 39 +++++++++++++++++++++++++++++----------
 2 files changed, 41 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h
index 388cd91bc7a..bcd0ac33f59 100644
--- a/include/linux/netpoll.h
+++ b/include/linux/netpoll.h
@@ -27,7 +27,8 @@ struct netpoll_info {
 	spinlock_t poll_lock;
 	int poll_owner;
 	int rx_flags;
-	struct netpoll *np;
+	spinlock_t rx_lock;
+	struct netpoll *rx_np; /* netpoll that registered an rx_hook */
 };
 
 void netpoll_poll(struct netpoll *np);
@@ -44,11 +45,19 @@ void netpoll_queue(struct sk_buff *skb);
 static inline int netpoll_rx(struct sk_buff *skb)
 {
 	struct netpoll_info *npinfo = skb->dev->npinfo;
+	unsigned long flags;
+	int ret = 0;
 
-	if (!npinfo || !npinfo->rx_flags)
+	if (!npinfo || (!npinfo->rx_np && !npinfo->rx_flags))
 		return 0;
 
-	return npinfo->np && __netpoll_rx(skb);
+	spin_lock_irqsave(&npinfo->rx_lock, flags);
+	/* check rx_flags again with the lock held */
+	if (npinfo->rx_flags && __netpoll_rx(skb))
+		ret = 1;
+	spin_unlock_irqrestore(&npinfo->rx_lock, flags);
+
+	return ret;
 }
 
 static inline void netpoll_poll_lock(struct net_device *dev)
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index ab3c0c9713b..c327c9edadc 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -349,11 +349,15 @@ static void arp_reply(struct sk_buff *skb)
 	unsigned char *arp_ptr;
 	int size, type = ARPOP_REPLY, ptype = ETH_P_ARP;
 	u32 sip, tip;
+	unsigned long flags;
 	struct sk_buff *send_skb;
 	struct netpoll *np = NULL;
 
-	if (npinfo)
-		np = npinfo->np;
+	spin_lock_irqsave(&npinfo->rx_lock, flags);
+	if (npinfo->rx_np && npinfo->rx_np->dev == skb->dev)
+		np = npinfo->rx_np;
+	spin_unlock_irqrestore(&npinfo->rx_lock, flags);
+
 	if (!np)
 		return;
 
@@ -436,9 +440,9 @@ int __netpoll_rx(struct sk_buff *skb)
 	int proto, len, ulen;
 	struct iphdr *iph;
 	struct udphdr *uh;
-	struct netpoll *np = skb->dev->npinfo->np;
+	struct netpoll *np = skb->dev->npinfo->rx_np;
 
-	if (!np->rx_hook)
+	if (!np)
 		goto out;
 	if (skb->dev->type != ARPHRD_ETHER)
 		goto out;
@@ -619,6 +623,7 @@ int netpoll_setup(struct netpoll *np)
 	struct net_device *ndev = NULL;
 	struct in_device *in_dev;
 	struct netpoll_info *npinfo;
+	unsigned long flags;
 
 	if (np->dev_name)
 		ndev = dev_get_by_name(np->dev_name);
@@ -634,9 +639,10 @@ int netpoll_setup(struct netpoll *np)
 		if (!npinfo)
 			goto release;
 
-		npinfo->np = NULL;
+		npinfo->rx_np = NULL;
 		npinfo->poll_lock = SPIN_LOCK_UNLOCKED;
 		npinfo->poll_owner = -1;
+		npinfo->rx_lock = SPIN_LOCK_UNLOCKED;
 	} else
 		npinfo = ndev->npinfo;
 
@@ -706,9 +712,13 @@ int netpoll_setup(struct netpoll *np)
 		       np->name, HIPQUAD(np->local_ip));
 	}
 
-	if(np->rx_hook)
-		npinfo->rx_flags = NETPOLL_RX_ENABLED;
-	npinfo->np = np;
+	if (np->rx_hook) {
+		spin_lock_irqsave(&npinfo->rx_lock, flags);
+		npinfo->rx_flags |= NETPOLL_RX_ENABLED;
+		npinfo->rx_np = np;
+		spin_unlock_irqrestore(&npinfo->rx_lock, flags);
+	}
+	/* last thing to do is link it to the net device structure */
 	ndev->npinfo = npinfo;
 
 	return 0;
@@ -723,11 +733,20 @@ int netpoll_setup(struct netpoll *np)
 
 void netpoll_cleanup(struct netpoll *np)
 {
+	struct netpoll_info *npinfo;
+	unsigned long flags;
+
 	if (np->dev) {
-		if (np->dev->npinfo)
-			np->dev->npinfo->np = NULL;
+		npinfo = np->dev->npinfo;
+		if (npinfo && npinfo->rx_np == np) {
+			spin_lock_irqsave(&npinfo->rx_lock, flags);
+			npinfo->rx_np = NULL;
+			npinfo->rx_flags &= ~NETPOLL_RX_ENABLED;
+			spin_unlock_irqrestore(&npinfo->rx_lock, flags);
+		}
 		dev_put(np->dev);
 	}
+
 	np->dev = NULL;
 }
 
-- 
cgit v1.2.3-70-g09d2


From cb65d506c34c86df5bcef939ce5a8666a451bd8b Mon Sep 17 00:00:00 2001
From: Shaun Pereira <spereira@tusc.com.au>
Date: Wed, 22 Jun 2005 22:15:01 -0700
Subject: [X25]: Selective sub-address matching with call user data.

From: Shaun Pereira <spereira@tusc.com.au>

This is the first (independent of the second) patch of two that I am
working on with x25 on linux (tested with xot on a cisco router).  Details
are as follows.

Current state of module:

A server using the current implementation (2.6.11.7) of the x25 module will
accept a call request/ incoming call packet at the listening x.25 address,
from all callers to that address, as long as NO call user data is present
in the packet header.

If the server needs to choose to accept a particular call request/ incoming
call packet arriving at its listening x25 address, then the kernel has to
allow a match of call user data present in the call request packet with its
own.  This is required when multiple servers listen at the same x25 address
and device interface.  The kernel currently matches ALL call user data, if
present.

Current Changes:

This patch is a follow up to the patch submitted previously by Andrew
Hendry, and allows the user to selectively control the number of octets of
call user data in the call request packet, that the kernel will match.  By
default no call user data is matched, even if call user data is present.
To allow call user data matching, a cudmatchlength > 0 has to be passed
into the kernel after which the passed number of octets will be matched.
Otherwise the kernel behavior is exactly as the original implementation.

This patch also ensures that as is normally the case, no call user data
will be present in the Call accepted / call connected packet sent back to
the caller

Future Changes on next patch:

There are cases however when call user data may be present in the call
accepted packet.  According to the X.25 recommendation (ITU-T 10/96)
section 5.2.3.2 call user data may be present in the call accepted packet
provided the fast select facility is used.  My next patch will include this
fast select utility and the ability to send up to 128 octets call user data
in the call accepted packet provided the fast select facility is used.  I
am currently testing this, again with xot on linux and cisco.

Signed-off-by: Shaun Pereira <spereira@tusc.com.au>

(With a fix from Alexey Dobriyan <adobriyan@gmail.com>)
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/x25.h | 10 ++++++++
 include/net/x25.h   |  3 +--
 net/x25/af_x25.c    | 73 +++++++++++++++++++++++++++++++++++------------------
 net/x25/x25_subr.c  | 18 -------------
 4 files changed, 59 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/include/linux/x25.h b/include/linux/x25.h
index 7531cfed588..6f43b3d2024 100644
--- a/include/linux/x25.h
+++ b/include/linux/x25.h
@@ -4,6 +4,8 @@
  * 	History
  *	mar/20/00	Daniela Squassoni Disabling/enabling of facilities 
  *					  negotiation.
+ *	apr/02/05	Shaun Pereira Selective sub address matching with
+ *					call user data
  */
 
 #ifndef	X25_KERNEL_H
@@ -16,6 +18,7 @@
 #define	SIOCX25GCALLUSERDATA	(SIOCPROTOPRIVATE + 4)
 #define	SIOCX25SCALLUSERDATA	(SIOCPROTOPRIVATE + 5)
 #define	SIOCX25GCAUSEDIAG	(SIOCPROTOPRIVATE + 6)
+#define SIOCX25SCUDMATCHLEN	(SIOCPROTOPRIVATE + 7)
 
 /*
  *	Values for {get,set}sockopt.
@@ -109,4 +112,11 @@ struct x25_causediag {
 	unsigned char	diagnostic;
 };
 
+/*
+ *	Further optional call user data match length selection
+ */
+struct x25_subaddr {
+	unsigned int cudmatchlength;
+};
+
 #endif
diff --git a/include/net/x25.h b/include/net/x25.h
index 7a1ba5bbb86..9dd70dd4a9b 100644
--- a/include/net/x25.h
+++ b/include/net/x25.h
@@ -134,7 +134,7 @@ struct x25_sock {
 	struct sock		sk;
 	struct x25_address	source_addr, dest_addr;
 	struct x25_neigh	*neighbour;
-	unsigned int		lci;
+	unsigned int		lci, cudmatchlength;
 	unsigned char		state, condition, qbitincl, intflag;
 	unsigned short		vs, vr, va, vl;
 	unsigned long		t2, t21, t22, t23;
@@ -242,7 +242,6 @@ extern int  x25_validate_nr(struct sock *, unsigned short);
 extern void x25_write_internal(struct sock *, int);
 extern int  x25_decode(struct sock *, struct sk_buff *, int *, int *, int *, int *, int *);
 extern void x25_disconnect(struct sock *, int, unsigned char, unsigned char);
-extern int x25_check_calluserdata(struct x25_calluserdata *,struct x25_calluserdata *);
 
 /* x25_timer.c */
 extern void x25_start_heartbeat(struct sock *);
diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c
index 2a24b243b84..e17d84a55d5 100644
--- a/net/x25/af_x25.c
+++ b/net/x25/af_x25.c
@@ -29,6 +29,8 @@
  *	2000-11-14	Henner Eisen    Closing datalink from NETDEV_GOING_DOWN
  *	2002-10-06	Arnaldo C. Melo Get rid of cli/sti, move proc stuff to
  *					x25_proc.c, using seq_file
+ *	2005-04-02	Shaun Pereira	Selective sub address matching
+ *					with call user data
  */
 
 #include <linux/config.h>
@@ -219,7 +221,8 @@ static void x25_insert_socket(struct sock *sk)
  *	Note: if a listening socket has cud set it must only get calls
  *	with matching cud.
  */
-static struct sock *x25_find_listener(struct x25_address *addr, struct x25_calluserdata *calluserdata)
+static struct sock *x25_find_listener(struct x25_address *addr,
+					struct sk_buff *skb)
 {
 	struct sock *s;
 	struct sock *next_best;
@@ -230,22 +233,23 @@ static struct sock *x25_find_listener(struct x25_address *addr, struct x25_callu
 
 	sk_for_each(s, node, &x25_list)
 		if ((!strcmp(addr->x25_addr,
-			     x25_sk(s)->source_addr.x25_addr) ||
-		     !strcmp(addr->x25_addr,
-			     null_x25_address.x25_addr)) &&
-		     s->sk_state == TCP_LISTEN) {
-
+			x25_sk(s)->source_addr.x25_addr) ||
+				!strcmp(addr->x25_addr,
+					null_x25_address.x25_addr)) &&
+					s->sk_state == TCP_LISTEN) {
 			/*
 			 * Found a listening socket, now check the incoming
 			 * call user data vs this sockets call user data
 			 */
-			if (x25_check_calluserdata(&x25_sk(s)->calluserdata, calluserdata)) {
-				sock_hold(s);
-				goto found;
-			}
-			if (x25_sk(s)->calluserdata.cudlength == 0) {
+			if(skb->len > 0 && x25_sk(s)->cudmatchlength > 0) {
+			 	if((memcmp(x25_sk(s)->calluserdata.cuddata,
+			 		skb->data,
+					x25_sk(s)->cudmatchlength)) == 0) {
+					sock_hold(s);
+					goto found;
+				 }
+			} else
 				next_best = s;
-			}
 		}
 	if (next_best) {
 		s = next_best;
@@ -497,6 +501,7 @@ static int x25_create(struct socket *sock, int protocol)
 	x25->t23   = sysctl_x25_clear_request_timeout;
 	x25->t2    = sysctl_x25_ack_holdback_timeout;
 	x25->state = X25_STATE_0;
+	x25->cudmatchlength = 0;
 
 	x25->facilities.winsize_in  = X25_DEFAULT_WINDOW_SIZE;
 	x25->facilities.winsize_out = X25_DEFAULT_WINDOW_SIZE;
@@ -545,6 +550,7 @@ static struct sock *x25_make_new(struct sock *osk)
 	x25->t2         = ox25->t2;
 	x25->facilities = ox25->facilities;
 	x25->qbitincl   = ox25->qbitincl;
+	x25->cudmatchlength = ox25->cudmatchlength;
 
 	x25_init_timers(sk);
 out:
@@ -822,7 +828,6 @@ int x25_rx_call_request(struct sk_buff *skb, struct x25_neigh *nb,
 	struct x25_sock *makex25;
 	struct x25_address source_addr, dest_addr;
 	struct x25_facilities facilities;
-	struct x25_calluserdata calluserdata;
 	int len, rc;
 
 	/*
@@ -844,20 +849,11 @@ int x25_rx_call_request(struct sk_buff *skb, struct x25_neigh *nb,
 	len = skb->data[0] + 1;
 	skb_pull(skb,len);
 
-	/*
-	 *	Incoming Call User Data.
-	 */
-	if (skb->len >= 0) {
-		memcpy(calluserdata.cuddata, skb->data, skb->len);
-		calluserdata.cudlength = skb->len;
-	}
-
-	skb_push(skb,len);
-
 	/*
 	 *	Find a listener for the particular address/cud pair.
 	 */
-	sk = x25_find_listener(&source_addr,&calluserdata);
+	sk = x25_find_listener(&source_addr,skb);
+	skb_push(skb,len);
 
 	/*
 	 *	We can't accept the Call Request.
@@ -900,12 +896,22 @@ int x25_rx_call_request(struct sk_buff *skb, struct x25_neigh *nb,
 	makex25->neighbour     = nb;
 	makex25->facilities    = facilities;
 	makex25->vc_facil_mask = x25_sk(sk)->vc_facil_mask;
-	makex25->calluserdata  = calluserdata;
+	/* ensure no reverse facil on accept */
+	makex25->vc_facil_mask &= ~X25_MASK_REVERSE;
+	makex25->cudmatchlength = x25_sk(sk)->cudmatchlength;
 
 	x25_write_internal(make, X25_CALL_ACCEPTED);
 
 	makex25->state = X25_STATE_3;
 
+	/*
+	 *	Incoming Call User Data.
+	 */
+	if (skb->len >= 0) {
+		memcpy(makex25->calluserdata.cuddata, skb->data, skb->len);
+		makex25->calluserdata.cudlength = skb->len;
+	}
+
 	sk->sk_ack_backlog++;
 
 	x25_insert_socket(make);
@@ -1325,6 +1331,23 @@ static int x25_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 			break;
 		}
 
+		case SIOCX25SCUDMATCHLEN: {
+			struct x25_subaddr sub_addr;
+			rc = -EINVAL;
+			if(sk->sk_state != TCP_CLOSE)
+				break;
+			rc = -EFAULT;
+			if (copy_from_user(&sub_addr, argp,
+					sizeof(sub_addr)))
+				break;
+		 	rc = -EINVAL;
+			if(sub_addr.cudmatchlength > X25_MAX_CUD_LEN)
+				break;
+			x25->cudmatchlength = sub_addr.cudmatchlength;
+			rc = 0;
+			break;
+		}
+
  		default:
 			rc = dev_ioctl(cmd, argp);
 			break;
diff --git a/net/x25/x25_subr.c b/net/x25/x25_subr.c
index 183fea3bba6..c349bbd6168 100644
--- a/net/x25/x25_subr.c
+++ b/net/x25/x25_subr.c
@@ -354,21 +354,3 @@ void x25_check_rbuf(struct sock *sk)
 	}
 }
 
-/*
- * Compare 2 calluserdata structures, used to find correct listening sockets
- * when call user data is used.
- */
-int x25_check_calluserdata(struct x25_calluserdata *ours, struct x25_calluserdata *theirs)
-{
-	int i;
-	if (ours->cudlength != theirs->cudlength)
-		return 0;
-
-	for (i=0;i<ours->cudlength;i++) {
-		if (ours->cuddata[i] != theirs->cuddata[i]) {
-			return 0;
-		}
-	}
-	return 1;
-}
-
-- 
cgit v1.2.3-70-g09d2


From ebc3f64b864fc16a594c2e63bf55a55c7d42084b Mon Sep 17 00:00:00 2001
From: Shaun Pereira <spereira@tusc.com.au>
Date: Wed, 22 Jun 2005 22:16:17 -0700
Subject: [X25]: Fast select with no restriction on response

This patch is a follow up to patch 1 regarding "Selective Sub Address
matching with call user data".  It allows use of the Fast-Select-Acceptance
optional user facility for X.25.

This patch just implements fast select with no restriction on response
(NRR).  What this means (according to ITU-T Recomendation 10/96 section
6.16) is that if in an incoming call packet, the relevant facility bits are
set for fast-select-NRR, then the called DTE can issue a direct response to
the incoming packet using a call-accepted packet that contains
call-user-data.  This patch allows such a response.

The called DTE can also respond with a clear-request packet that contains
call-user-data.  However, this feature is currently not implemented by the
patch.

How is Fast Select Acceptance used?
By default, the system does not allow fast select acceptance (as before).
To enable a response to fast select acceptance,
After a listen socket in created and bound as follows
	socket(AF_X25, SOCK_SEQPACKET, 0);
	bind(call_soc, (struct sockaddr *)&locl_addr, sizeof(locl_addr));
but before a listen system call is made, the following ioctl should be used.
	ioctl(call_soc,SIOCX25CALLACCPTAPPRV);
Now the listen system call can be made
	listen(call_soc, 4);
After this, an incoming-call packet will be accepted, but no call-accepted
packet will be sent back until the following system call is made on the socket
that accepts the call
	ioctl(vc_soc,SIOCX25SENDCALLACCPT);
The network (or cisco xot router used for testing here) will allow the
application server's call-user-data in the call-accepted packet,
provided the call-request was made with Fast-select NRR.

Signed-off-by: Shaun Pereira <spereira@tusc.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/x25.h      |  2 ++
 include/net/x25.h        |  6 ++++--
 net/x25/af_x25.c         | 37 +++++++++++++++++++++++++++++++++----
 net/x25/x25_facilities.c | 34 +++++++++++++++++++++++++++++-----
 net/x25/x25_subr.c       | 23 ++++++++++++++++++-----
 5 files changed, 86 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/linux/x25.h b/include/linux/x25.h
index 6f43b3d2024..16d44931afa 100644
--- a/include/linux/x25.h
+++ b/include/linux/x25.h
@@ -19,6 +19,8 @@
 #define	SIOCX25SCALLUSERDATA	(SIOCPROTOPRIVATE + 5)
 #define	SIOCX25GCAUSEDIAG	(SIOCPROTOPRIVATE + 6)
 #define SIOCX25SCUDMATCHLEN	(SIOCPROTOPRIVATE + 7)
+#define SIOCX25CALLACCPTAPPRV   (SIOCPROTOPRIVATE + 8)
+#define SIOCX25SENDCALLACCPT    (SIOCPROTOPRIVATE + 9)
 
 /*
  *	Values for {get,set}sockopt.
diff --git a/include/net/x25.h b/include/net/x25.h
index 9dd70dd4a9b..8b39b98876e 100644
--- a/include/net/x25.h
+++ b/include/net/x25.h
@@ -79,6 +79,8 @@ enum {
 #define	X25_DEFAULT_PACKET_SIZE	X25_PS128		/* Default Packet Size */
 #define	X25_DEFAULT_THROUGHPUT	0x0A			/* Deafult Throughput */
 #define	X25_DEFAULT_REVERSE	0x00			/* Default Reverse Charging */
+#define X25_DENY_ACCPT_APPRV   0x01			/* Default value */
+#define X25_ALLOW_ACCPT_APPRV  0x00			/* Control enabled */
 
 #define X25_SMODULUS 		8
 #define	X25_EMODULUS		128
@@ -94,7 +96,7 @@ enum {
 #define	X25_FAC_CLASS_C		0x80
 #define	X25_FAC_CLASS_D		0xC0
 
-#define	X25_FAC_REVERSE		0x01
+#define	X25_FAC_REVERSE		0x01			/* also fast select */
 #define	X25_FAC_THROUGHPUT	0x02
 #define	X25_FAC_PACKET_SIZE	0x42
 #define	X25_FAC_WINDOW_SIZE	0x43
@@ -135,7 +137,7 @@ struct x25_sock {
 	struct x25_address	source_addr, dest_addr;
 	struct x25_neigh	*neighbour;
 	unsigned int		lci, cudmatchlength;
-	unsigned char		state, condition, qbitincl, intflag;
+	unsigned char		state, condition, qbitincl, intflag, accptapprv;
 	unsigned short		vs, vr, va, vl;
 	unsigned long		t2, t21, t22, t23;
 	unsigned short		fraglen;
diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c
index e17d84a55d5..04bec047fa9 100644
--- a/net/x25/af_x25.c
+++ b/net/x25/af_x25.c
@@ -31,6 +31,8 @@
  *					x25_proc.c, using seq_file
  *	2005-04-02	Shaun Pereira	Selective sub address matching
  *					with call user data
+ *	2005-04-15	Shaun Pereira	Fast select with no restriction on
+ *					response
  */
 
 #include <linux/config.h>
@@ -502,6 +504,8 @@ static int x25_create(struct socket *sock, int protocol)
 	x25->t2    = sysctl_x25_ack_holdback_timeout;
 	x25->state = X25_STATE_0;
 	x25->cudmatchlength = 0;
+	x25->accptapprv = X25_DENY_ACCPT_APPRV;		/* normally no cud  */
+							/* on call accept   */
 
 	x25->facilities.winsize_in  = X25_DEFAULT_WINDOW_SIZE;
 	x25->facilities.winsize_out = X25_DEFAULT_WINDOW_SIZE;
@@ -551,6 +555,7 @@ static struct sock *x25_make_new(struct sock *osk)
 	x25->facilities = ox25->facilities;
 	x25->qbitincl   = ox25->qbitincl;
 	x25->cudmatchlength = ox25->cudmatchlength;
+	x25->accptapprv = ox25->accptapprv;
 
 	x25_init_timers(sk);
 out:
@@ -900,9 +905,11 @@ int x25_rx_call_request(struct sk_buff *skb, struct x25_neigh *nb,
 	makex25->vc_facil_mask &= ~X25_MASK_REVERSE;
 	makex25->cudmatchlength = x25_sk(sk)->cudmatchlength;
 
-	x25_write_internal(make, X25_CALL_ACCEPTED);
-
-	makex25->state = X25_STATE_3;
+	/* Normally all calls are accepted immediatly */
+	if(makex25->accptapprv & X25_DENY_ACCPT_APPRV) {
+		x25_write_internal(make, X25_CALL_ACCEPTED);
+		makex25->state = X25_STATE_3;
+	}
 
 	/*
 	 *	Incoming Call User Data.
@@ -1294,7 +1301,8 @@ static int x25_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 			if (facilities.throughput < 0x03 ||
 			    facilities.throughput > 0xDD)
 				break;
-			if (facilities.reverse && facilities.reverse != 1)
+			if (facilities.reverse &&
+				(facilities.reverse | 0x81)!= 0x81)
 				break;
 			x25->facilities = facilities;
 			rc = 0;
@@ -1348,6 +1356,27 @@ static int x25_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 			break;
 		}
 
+		case SIOCX25CALLACCPTAPPRV: {
+			rc = -EINVAL;
+			if (sk->sk_state != TCP_CLOSE)
+				break;
+			x25->accptapprv = X25_ALLOW_ACCPT_APPRV;
+			rc = 0;
+			break;
+		}
+
+		case SIOCX25SENDCALLACCPT:  {
+			rc = -EINVAL;
+			if (sk->sk_state != TCP_ESTABLISHED)
+				break;
+			if (x25->accptapprv)	/* must call accptapprv above */
+				break;
+			x25_write_internal(sk, X25_CALL_ACCEPTED);
+			x25->state = X25_STATE_3;
+			rc = 0;
+			break;
+		}
+
  		default:
 			rc = dev_ioctl(cmd, argp);
 			break;
diff --git a/net/x25/x25_facilities.c b/net/x25/x25_facilities.c
index a21bdb95f9a..54278b962f4 100644
--- a/net/x25/x25_facilities.c
+++ b/net/x25/x25_facilities.c
@@ -17,6 +17,8 @@
  *	X.25 001	Split from x25_subr.c
  *	mar/20/00	Daniela Squassoni Disabling/enabling of facilities 
  *					  negotiation.
+ *	apr/14/05	Shaun Pereira - Allow fast select with no restriction
+ *					on response.
  */
 
 #include <linux/kernel.h>
@@ -43,9 +45,31 @@ int x25_parse_facilities(struct sk_buff *skb,
 		case X25_FAC_CLASS_A:
 			switch (*p) {
 			case X25_FAC_REVERSE:
-				facilities->reverse = p[1] & 0x01;
-				*vc_fac_mask |= X25_MASK_REVERSE;
-				break;
+				if((p[1] & 0x81) == 0x81) {
+					facilities->reverse = p[1] & 0x81;
+					*vc_fac_mask |= X25_MASK_REVERSE;
+					break;
+				}
+
+				if((p[1] & 0x01) == 0x01) {
+					facilities->reverse = p[1] & 0x01;
+					*vc_fac_mask |= X25_MASK_REVERSE;
+					break;
+				}
+
+				if((p[1] & 0x80) == 0x80) {
+					facilities->reverse = p[1] & 0x80;
+					*vc_fac_mask |= X25_MASK_REVERSE;
+					break;
+				}
+
+				if(p[1] == 0x00) {
+					facilities->reverse
+						= X25_DEFAULT_REVERSE;
+					*vc_fac_mask |= X25_MASK_REVERSE;
+					break;
+				}
+
 			case X25_FAC_THROUGHPUT:
 				facilities->throughput = p[1];
 				*vc_fac_mask |= X25_MASK_THROUGHPUT;
@@ -122,7 +146,7 @@ int x25_create_facilities(unsigned char *buffer,
 
 	if (facilities->reverse && (facil_mask & X25_MASK_REVERSE)) {
 		*p++ = X25_FAC_REVERSE;
-		*p++ = !!facilities->reverse;
+		*p++ = facilities->reverse;
 	}
 
 	if (facilities->throughput && (facil_mask & X25_MASK_THROUGHPUT)) {
@@ -171,7 +195,7 @@ int x25_negotiate_facilities(struct sk_buff *skb, struct sock *sk,
 	/*
 	 *	They want reverse charging, we won't accept it.
 	 */
-	if (theirs.reverse && ours->reverse) {
+	if ((theirs.reverse & 0x01 ) && (ours->reverse & 0x01)) {
 		SOCK_DEBUG(sk, "X.25: rejecting reverse charging request");
 		return -1;
 	}
diff --git a/net/x25/x25_subr.c b/net/x25/x25_subr.c
index c349bbd6168..7fd872ad0c2 100644
--- a/net/x25/x25_subr.c
+++ b/net/x25/x25_subr.c
@@ -19,6 +19,8 @@
  *	mar/20/00	Daniela Squassoni Disabling/enabling of facilities
  *					  negotiation.
  *	jun/24/01	Arnaldo C. Melo	  use skb_queue_purge, cleanups
+ *	apr/04/15	Shaun Pereira		Fast select with no
+ *						restriction on response.
  */
 
 #include <linux/kernel.h>
@@ -127,8 +129,12 @@ void x25_write_internal(struct sock *sk, int frametype)
 			len += 1 + X25_ADDR_LEN + X25_MAX_FAC_LEN +
 			       X25_MAX_CUD_LEN;
 			break;
-		case X25_CALL_ACCEPTED:
-			len += 1 + X25_MAX_FAC_LEN + X25_MAX_CUD_LEN;
+		case X25_CALL_ACCEPTED: /* fast sel with no restr on resp */
+			if(x25->facilities.reverse & 0x80) {
+				len += 1 + X25_MAX_FAC_LEN + X25_MAX_CUD_LEN;
+			} else {
+				len += 1 + X25_MAX_FAC_LEN;
+			}
 			break;
 		case X25_CLEAR_REQUEST:
 		case X25_RESET_REQUEST:
@@ -203,9 +209,16 @@ void x25_write_internal(struct sock *sk, int frametype)
 							x25->vc_facil_mask);
 			dptr    = skb_put(skb, len);
 			memcpy(dptr, facilities, len);
-			dptr = skb_put(skb, x25->calluserdata.cudlength);
-			memcpy(dptr, x25->calluserdata.cuddata,
-			       x25->calluserdata.cudlength);
+
+			/* fast select with no restriction on response
+				allows call user data. Userland must
+				ensure it is ours and not theirs */
+			if(x25->facilities.reverse & 0x80) {
+				dptr = skb_put(skb,
+					x25->calluserdata.cudlength);
+				memcpy(dptr, x25->calluserdata.cuddata,
+				       x25->calluserdata.cudlength);
+			}
 			x25->calluserdata.cudlength = 0;
 			break;
 
-- 
cgit v1.2.3-70-g09d2


From dad32bbf43b496bcd32a83f73a1e7fd0a02cfd3e Mon Sep 17 00:00:00 2001
From: John Rose <johnrose@austin.ibm.com>
Date: Thu, 23 Jun 2005 17:09:54 +1000
Subject: [PATCH] pSeries - read irqs dynamically

For I/O DLPAR to work properly, the kernel needs to allow for dynamic
assignment of the irq field of the pci_dev structure upon dynamic bus
addition.  This patch moves the assignment of that field from
pSeries_final_fixup() to pcibios_fixup_bus(), which enables dynamic
assignment for the children of a newly added bus.

Currently, pci_devs receive their irq numbers in one of two ways.  The
irq line is either read at boot for all pci_devs, or read by the rpaphp
module at slot enable time.  The latter is no longer sufficient for
DLPAR addition of slots that don't qualify as PCI-hotplug capable.
This solution handles the cases of boot and dynamic add.

Signed-off-by: John Rose <johnrose@austin.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/pSeries_pci.c   | 35 ++++++++++++++++++++---------------
 arch/ppc64/kernel/pSeries_setup.c |  3 +--
 arch/ppc64/kernel/pci.c           |  3 +++
 arch/ppc64/kernel/pci.h           |  6 +++++-
 include/asm-ppc64/machdep.h       |  1 +
 5 files changed, 30 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/pSeries_pci.c b/arch/ppc64/kernel/pSeries_pci.c
index dfa6d3d3e9f..1f5f141fb7a 100644
--- a/arch/ppc64/kernel/pSeries_pci.c
+++ b/arch/ppc64/kernel/pSeries_pci.c
@@ -32,7 +32,7 @@
 
 #include "pci.h"
 
-static int __initdata s7a_workaround;
+static int __initdata s7a_workaround = -1;
 
 #if 0
 void pcibios_name_device(struct pci_dev *dev)
@@ -65,6 +65,7 @@ static void __init check_s7a(void)
 	struct device_node *root;
 	char *model;
 
+	s7a_workaround = 0;
 	root = of_find_node_by_path("/");
 	if (root) {
 		model = get_property(root, "model", NULL);
@@ -74,6 +75,24 @@ static void __init check_s7a(void)
 	}
 }
 
+void __devinit pSeries_irq_bus_setup(struct pci_bus *bus)
+{
+	struct pci_dev *dev;
+
+	if (s7a_workaround < 0)
+		check_s7a();
+	list_for_each_entry(dev, &bus->devices, bus_list) {
+		pci_read_irq_line(dev);
+		if (s7a_workaround) {
+			if (dev->irq > 16) {
+				dev->irq -= 3;
+				pci_write_config_byte(dev, PCI_INTERRUPT_LINE,
+					dev->irq);
+			}
+		}
+	}
+}
+
 static void __init pSeries_request_regions(void)
 {
 	if (!isa_io_base)
@@ -89,20 +108,6 @@ static void __init pSeries_request_regions(void)
 
 void __init pSeries_final_fixup(void)
 {
-	struct pci_dev *dev = NULL;
-
-	check_s7a();
-
-	for_each_pci_dev(dev) {
-		pci_read_irq_line(dev);
-		if (s7a_workaround) {
-			if (dev->irq > 16) {
-				dev->irq -= 3;
-				pci_write_config_byte(dev, PCI_INTERRUPT_LINE, dev->irq);
-			}
-		}
-	}
-
 	phbs_remap_io();
 	pSeries_request_regions();
 
diff --git a/arch/ppc64/kernel/pSeries_setup.c b/arch/ppc64/kernel/pSeries_setup.c
index 41e6de2c915..f2b41243342 100644
--- a/arch/ppc64/kernel/pSeries_setup.c
+++ b/arch/ppc64/kernel/pSeries_setup.c
@@ -71,8 +71,6 @@
 #define DBG(fmt...)
 #endif
 
-extern void pSeries_final_fixup(void);
-
 extern void find_udbg_vterm(void);
 extern void system_reset_fwnmi(void);	/* from head.S */
 extern void machine_check_fwnmi(void);	/* from head.S */
@@ -425,6 +423,7 @@ struct machdep_calls __initdata pSeries_md = {
 	.get_cpuinfo		= pSeries_get_cpuinfo,
 	.log_error		= pSeries_log_error,
 	.pcibios_fixup		= pSeries_final_fixup,
+	.irq_bus_setup		= pSeries_irq_bus_setup,
 	.restart		= rtas_restart,
 	.power_off		= rtas_power_off,
 	.halt			= rtas_halt,
diff --git a/arch/ppc64/kernel/pci.c b/arch/ppc64/kernel/pci.c
index 2bf0513f3ec..580676f87d2 100644
--- a/arch/ppc64/kernel/pci.c
+++ b/arch/ppc64/kernel/pci.c
@@ -902,6 +902,9 @@ void __devinit pcibios_fixup_bus(struct pci_bus *bus)
 	list_for_each_entry(dev, &bus->devices, bus_list)
 		ppc_md.iommu_dev_setup(dev);
 
+	if (ppc_md.irq_bus_setup)
+		ppc_md.irq_bus_setup(bus);
+
 	if (!pci_probe_only)
 		return;
 
diff --git a/arch/ppc64/kernel/pci.h b/arch/ppc64/kernel/pci.h
index 0fd7d849aa7..26be78b13af 100644
--- a/arch/ppc64/kernel/pci.h
+++ b/arch/ppc64/kernel/pci.h
@@ -40,10 +40,14 @@ struct device_node *fetch_dev_dn(struct pci_dev *dev);
 void pci_addr_cache_insert_device(struct pci_dev *dev);
 void pci_addr_cache_remove_device(struct pci_dev *dev);
 
-/* From pSeries_pci.h */
+/* From rtas_pci.h */
 void init_pci_config_tokens (void);
 unsigned long get_phb_buid (struct device_node *);
 
+/* From pSeries_pci.h */
+extern void pSeries_final_fixup(void);
+extern void pSeries_irq_bus_setup(struct pci_bus *bus);
+
 extern unsigned long pci_probe_only;
 extern unsigned long pci_assign_all_buses;
 extern int pci_read_irq_line(struct pci_dev *pci_dev);
diff --git a/include/asm-ppc64/machdep.h b/include/asm-ppc64/machdep.h
index 5d3cd9d042e..553b2ea23be 100644
--- a/include/asm-ppc64/machdep.h
+++ b/include/asm-ppc64/machdep.h
@@ -76,6 +76,7 @@ struct machdep_calls {
 	void		(*tce_flush)(struct iommu_table *tbl);
 	void		(*iommu_dev_setup)(struct pci_dev *dev);
 	void		(*iommu_bus_setup)(struct pci_bus *bus);
+	void		(*irq_bus_setup)(struct pci_bus *bus);
 
 	int		(*probe)(int platform);
 	void		(*setup_arch)(void);
-- 
cgit v1.2.3-70-g09d2


From 408fde81c1bff15c875a3618481e93a01dcc79ea Mon Sep 17 00:00:00 2001
From: Dave Hansen <haveblue@us.ibm.com>
Date: Thu, 23 Jun 2005 00:07:37 -0700
Subject: [PATCH] remove non-DISCONTIG use of pgdat->node_mem_map

This patch effectively eliminates direct use of pgdat->node_mem_map outside
of the DISCONTIG code.  On a flat memory system, these fields aren't
currently used, neither are they on a sparsemem system.

There was also a node_mem_map(nid) macro on many architectures.  Its use
along with the use of ->node_mem_map itself was not consistent.  It has
been removed in favor of two new, more explicit, arch-independent macros:

	pgdat_page_nr(pgdat, pagenr)
	nid_page_nr(nid, pagenr)

I called them "pgdat" and "nid" because we overload the term "node" to mean
"NUMA node", "DISCONTIG node" or "pg_data_t" in very confusing ways.  I
believe the newer names are much clearer.

These macros can be overridden in the sparsemem case with a theoretically
slower operation using node_start_pfn and pfn_to_page(), instead.  We could
make this the only behavior if people want, but I don't want to change too
much at once.  One thing at a time.

This patch removes more code than it adds.

Compile tested on alpha, alpha discontig, arm, arm-discontig, i386, i386
generic, NUMAQ, Summit, ppc64, ppc64 discontig, and x86_64.  Full list
here: http://sr71.net/patches/2.6.12/2.6.12-rc1-mhp2/configs/

Boot tested on NUMAQ, x86 SMP and ppc64 power4/5 LPARs.

Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Martin J. Bligh <mbligh@aracnet.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/alpha/mm/numa.c             | 16 +++++++---------
 arch/i386/mm/pgtable.c           |  2 +-
 arch/ia64/mm/discontig.c         |  9 +++++----
 arch/m32r/mm/init.c              |  4 ++--
 arch/mips/sgi-ip27/ip27-memory.c |  5 ++---
 arch/parisc/mm/init.c            |  2 +-
 arch/ppc64/mm/init.c             |  4 ++--
 include/asm-alpha/mmzone.h       |  3 +--
 include/asm-i386/mmzone.h        |  3 +--
 include/asm-m32r/mmzone.h        |  3 +--
 include/asm-parisc/mmzone.h      |  3 +--
 include/asm-ppc64/mmzone.h       |  3 +--
 include/asm-x86_64/mmzone.h      |  5 +----
 include/linux/mmzone.h           |  2 ++
 14 files changed, 28 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/arch/alpha/mm/numa.c b/arch/alpha/mm/numa.c
index ba81c4422aa..c7481d59b6d 100644
--- a/arch/alpha/mm/numa.c
+++ b/arch/alpha/mm/numa.c
@@ -327,8 +327,6 @@ void __init mem_init(void)
 	extern char _text, _etext, _data, _edata;
 	extern char __init_begin, __init_end;
 	unsigned long nid, i;
-	struct page * lmem_map;
-
 	high_memory = (void *) __va(max_low_pfn << PAGE_SHIFT);
 
 	reservedpages = 0;
@@ -338,10 +336,10 @@ void __init mem_init(void)
 		 */
 		totalram_pages += free_all_bootmem_node(NODE_DATA(nid));
 
-		lmem_map = node_mem_map(nid);
 		pfn = NODE_DATA(nid)->node_start_pfn;
 		for (i = 0; i < node_spanned_pages(nid); i++, pfn++)
-			if (page_is_ram(pfn) && PageReserved(lmem_map+i))
+			if (page_is_ram(pfn) &&
+			    PageReserved(nid_page_nr(nid, i)))
 				reservedpages++;
 	}
 
@@ -373,18 +371,18 @@ show_mem(void)
 	show_free_areas();
 	printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
 	for_each_online_node(nid) {
-		struct page * lmem_map = node_mem_map(nid);
 		i = node_spanned_pages(nid);
 		while (i-- > 0) {
+			struct page *page = nid_page_nr(nid, i);
 			total++;
-			if (PageReserved(lmem_map+i))
+			if (PageReserved(page))
 				reserved++;
-			else if (PageSwapCache(lmem_map+i))
+			else if (PageSwapCache(page))
 				cached++;
-			else if (!page_count(lmem_map+i))
+			else if (!page_count(page))
 				free++;
 			else
-				shared += page_count(lmem_map + i) - 1;
+				shared += page_count(page) - 1;
 		}
 	}
 	printk("%ld pages of RAM\n",total);
diff --git a/arch/i386/mm/pgtable.c b/arch/i386/mm/pgtable.c
index dd81479ff88..80c84cdf22e 100644
--- a/arch/i386/mm/pgtable.c
+++ b/arch/i386/mm/pgtable.c
@@ -36,7 +36,7 @@ void show_mem(void)
 	printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
 	for_each_pgdat(pgdat) {
 		for (i = 0; i < pgdat->node_spanned_pages; ++i) {
-			page = pgdat->node_mem_map + i;
+			page = pgdat_page_nr(pgdat, i);
 			total++;
 			if (PageHighMem(page))
 				highmem++;
diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
index c0071092939..f3fd528ead3 100644
--- a/arch/ia64/mm/discontig.c
+++ b/arch/ia64/mm/discontig.c
@@ -560,14 +560,15 @@ void show_mem(void)
 		int shared = 0, cached = 0, reserved = 0;
 		printk("Node ID: %d\n", pgdat->node_id);
 		for(i = 0; i < pgdat->node_spanned_pages; i++) {
+			struct page *page = pgdat_page_nr(pgdat, i);
 			if (!ia64_pfn_valid(pgdat->node_start_pfn+i))
 				continue;
-			if (PageReserved(pgdat->node_mem_map+i))
+			if (PageReserved(page))
 				reserved++;
-			else if (PageSwapCache(pgdat->node_mem_map+i))
+			else if (PageSwapCache(page))
 				cached++;
-			else if (page_count(pgdat->node_mem_map+i))
-				shared += page_count(pgdat->node_mem_map+i)-1;
+			else if (page_count(page))
+				shared += page_count(page)-1;
 		}
 		total_present += present;
 		total_reserved += reserved;
diff --git a/arch/m32r/mm/init.c b/arch/m32r/mm/init.c
index bc423d838fb..d9a40b1fe8b 100644
--- a/arch/m32r/mm/init.c
+++ b/arch/m32r/mm/init.c
@@ -49,7 +49,7 @@ void show_mem(void)
 	printk("Free swap:       %6ldkB\n",nr_swap_pages<<(PAGE_SHIFT-10));
 	for_each_pgdat(pgdat) {
 		for (i = 0; i < pgdat->node_spanned_pages; ++i) {
-			page = pgdat->node_mem_map + i;
+			page = pgdat_page_nr(pgdat, i);
 			total++;
 			if (PageHighMem(page))
 				highmem++;
@@ -152,7 +152,7 @@ int __init reservedpages_count(void)
 	reservedpages = 0;
 	for_each_online_node(nid)
 		for (i = 0 ; i < MAX_LOW_PFN(nid) - START_PFN(nid) ; i++)
-			if (PageReserved(NODE_DATA(nid)->node_mem_map + i))
+			if (PageReserved(nid_page_nr(nid, i)))
 				reservedpages++;
 
 	return reservedpages;
diff --git a/arch/mips/sgi-ip27/ip27-memory.c b/arch/mips/sgi-ip27/ip27-memory.c
index 0a44a98d7ad..a160d04f7db 100644
--- a/arch/mips/sgi-ip27/ip27-memory.c
+++ b/arch/mips/sgi-ip27/ip27-memory.c
@@ -549,9 +549,8 @@ void __init mem_init(void)
 		 */
 		numslots = node_getlastslot(node);
 		for (slot = 1; slot <= numslots; slot++) {
-			p = NODE_DATA(node)->node_mem_map +
-				(slot_getbasepfn(node, slot) -
-				 slot_getbasepfn(node, 0));
+			p = nid_page_nr(node, slot_getbasepfn(node, slot) -
+					      slot_getbasepfn(node, 0));
 
 			/*
 			 * Free valid memory in current slot.
diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c
index cac37589e35..2886ad70db4 100644
--- a/arch/parisc/mm/init.c
+++ b/arch/parisc/mm/init.c
@@ -506,7 +506,7 @@ void show_mem(void)
 		for (j = node_start_pfn(i); j < node_end_pfn(i); j++) {
 			struct page *p;
 
-			p = node_mem_map(i) + j - node_start_pfn(i);
+			p = nid_page_nr(i, j) - node_start_pfn(i);
 
 			total++;
 			if (PageReserved(p))
diff --git a/arch/ppc64/mm/init.c b/arch/ppc64/mm/init.c
index 6fa1e6490b5..29dbe084c21 100644
--- a/arch/ppc64/mm/init.c
+++ b/arch/ppc64/mm/init.c
@@ -98,7 +98,7 @@ void show_mem(void)
 	printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
 	for_each_pgdat(pgdat) {
 		for (i = 0; i < pgdat->node_spanned_pages; i++) {
-			page = pgdat->node_mem_map + i;
+			page = pgdat_page_nr(pgdat, i);
 			total++;
 			if (PageReserved(page))
 				reserved++;
@@ -654,7 +654,7 @@ void __init mem_init(void)
 
 	for_each_pgdat(pgdat) {
 		for (i = 0; i < pgdat->node_spanned_pages; i++) {
-			page = pgdat->node_mem_map + i;
+			page = pgdat_page_nr(pgdat, i);
 			if (PageReserved(page))
 				reservedpages++;
 		}
diff --git a/include/asm-alpha/mmzone.h b/include/asm-alpha/mmzone.h
index 726c150dcbe..a011ef4cf3d 100644
--- a/include/asm-alpha/mmzone.h
+++ b/include/asm-alpha/mmzone.h
@@ -57,7 +57,6 @@ PLAT_NODE_DATA_LOCALNR(unsigned long p, int n)
  * Given a kernel address, find the home node of the underlying memory.
  */
 #define kvaddr_to_nid(kaddr)	pa_to_nid(__pa(kaddr))
-#define node_mem_map(nid)	(NODE_DATA(nid)->node_mem_map)
 #define node_start_pfn(nid)	(NODE_DATA(nid)->node_start_pfn)
 
 #define local_mapnr(kvaddr) \
@@ -108,7 +107,7 @@ PLAT_NODE_DATA_LOCALNR(unsigned long p, int n)
 #define pfn_to_page(pfn)						\
 ({									\
  	unsigned long kaddr = (unsigned long)__va((pfn) << PAGE_SHIFT);	\
-	(node_mem_map(kvaddr_to_nid(kaddr)) + local_mapnr(kaddr));	\
+	(NODE_DATA(kvaddr_to_nid(kaddr))->node_mem_map + local_mapnr(kaddr));	\
 })
 
 #define page_to_pfn(page)						\
diff --git a/include/asm-i386/mmzone.h b/include/asm-i386/mmzone.h
index 13830ae67ca..9cec191f462 100644
--- a/include/asm-i386/mmzone.h
+++ b/include/asm-i386/mmzone.h
@@ -79,7 +79,6 @@ static inline int pfn_to_nid(unsigned long pfn)
  */
 #define kvaddr_to_nid(kaddr)	pfn_to_nid(__pa(kaddr) >> PAGE_SHIFT)
 
-#define node_mem_map(nid)	(NODE_DATA(nid)->node_mem_map)
 #define node_start_pfn(nid)	(NODE_DATA(nid)->node_start_pfn)
 #define node_end_pfn(nid)						\
 ({									\
@@ -100,7 +99,7 @@ static inline int pfn_to_nid(unsigned long pfn)
 ({									\
 	unsigned long __pfn = pfn;					\
 	int __node  = pfn_to_nid(__pfn);				\
-	&node_mem_map(__node)[node_localnr(__pfn,__node)];		\
+	&NODE_DATA(__node)->node_mem_map[node_localnr(__pfn,__node)];	\
 })
 
 #define page_to_pfn(pg)							\
diff --git a/include/asm-m32r/mmzone.h b/include/asm-m32r/mmzone.h
index ebf0228fec4..d58878ec899 100644
--- a/include/asm-m32r/mmzone.h
+++ b/include/asm-m32r/mmzone.h
@@ -14,7 +14,6 @@ extern struct pglist_data *node_data[];
 #define NODE_DATA(nid)		(node_data[nid])
 
 #define node_localnr(pfn, nid)	((pfn) - NODE_DATA(nid)->node_start_pfn)
-#define node_mem_map(nid)	(NODE_DATA(nid)->node_mem_map)
 #define node_start_pfn(nid)	(NODE_DATA(nid)->node_start_pfn)
 #define node_end_pfn(nid)						\
 ({									\
@@ -32,7 +31,7 @@ extern struct pglist_data *node_data[];
 ({									\
 	unsigned long __pfn = pfn;					\
 	int __node  = pfn_to_nid(__pfn);				\
-	&node_mem_map(__node)[node_localnr(__pfn,__node)];		\
+	&NODE_DATA(__node)->node_mem_map[node_localnr(__pfn,__node)];	\
 })
 
 #define page_to_pfn(pg)							\
diff --git a/include/asm-parisc/mmzone.h b/include/asm-parisc/mmzone.h
index 928bf50c469..595d3dce120 100644
--- a/include/asm-parisc/mmzone.h
+++ b/include/asm-parisc/mmzone.h
@@ -19,7 +19,6 @@ extern struct node_map_data node_data[];
  */
 #define kvaddr_to_nid(kaddr)	pfn_to_nid(__pa(kaddr) >> PAGE_SHIFT)
 
-#define node_mem_map(nid)	(NODE_DATA(nid)->node_mem_map)
 #define node_start_pfn(nid)	(NODE_DATA(nid)->node_start_pfn)
 #define node_end_pfn(nid)						\
 ({									\
@@ -38,7 +37,7 @@ extern struct node_map_data node_data[];
 ({									\
 	unsigned long __pfn = (pfn);					\
 	int __node  = pfn_to_nid(__pfn);				\
-	&node_mem_map(__node)[node_localnr(__pfn,__node)];		\
+	&NODE_DATA(__node)->node_mem_map[node_localnr(__pfn,__node)];	\
 })
 
 #define page_to_pfn(pg)							\
diff --git a/include/asm-ppc64/mmzone.h b/include/asm-ppc64/mmzone.h
index 0619a41a3c9..cbfc5ecfe87 100644
--- a/include/asm-ppc64/mmzone.h
+++ b/include/asm-ppc64/mmzone.h
@@ -65,7 +65,6 @@ static inline int pa_to_nid(unsigned long pa)
  */
 #define kvaddr_to_nid(kaddr)	pa_to_nid(__pa(kaddr))
 
-#define node_mem_map(nid)	(NODE_DATA(nid)->node_mem_map)
 #define node_start_pfn(nid)	(NODE_DATA(nid)->node_start_pfn)
 #define node_end_pfn(nid)	(NODE_DATA(nid)->node_end_pfn)
 
@@ -76,7 +75,7 @@ static inline int pa_to_nid(unsigned long pa)
 #define discontigmem_pfn_to_page(pfn) \
 ({ \
 	unsigned long __tmp = pfn; \
-	(node_mem_map(pfn_to_nid(__tmp)) + \
+	(NODE_DATA(pfn_to_nid(__tmp))->node_mem_map + \
 	 node_localnr(__tmp, pfn_to_nid(__tmp))); \
 })
 
diff --git a/include/asm-x86_64/mmzone.h b/include/asm-x86_64/mmzone.h
index d95b7c24083..ca4fc3fe0de 100644
--- a/include/asm-x86_64/mmzone.h
+++ b/include/asm-x86_64/mmzone.h
@@ -35,9 +35,6 @@ static inline __attribute__((pure)) int phys_to_nid(unsigned long addr)
 #define kvaddr_to_nid(kaddr)	phys_to_nid(__pa(kaddr))
 #define NODE_DATA(nid)		(node_data[nid])
 
-#define node_mem_map(nid)	(NODE_DATA(nid)->node_mem_map)
-
-#define node_mem_map(nid)	(NODE_DATA(nid)->node_mem_map)
 #define node_start_pfn(nid)	(NODE_DATA(nid)->node_start_pfn)
 #define node_end_pfn(nid)       (NODE_DATA(nid)->node_start_pfn + \
 				 NODE_DATA(nid)->node_spanned_pages)
@@ -50,7 +47,7 @@ static inline __attribute__((pure)) int phys_to_nid(unsigned long addr)
    (2.4 used to). */
 #define pfn_to_page(pfn) ({ \
 	int nid = phys_to_nid(((unsigned long)(pfn)) << PAGE_SHIFT); 	\
-	((pfn) - node_start_pfn(nid)) + node_mem_map(nid);		\
+	((pfn) - node_start_pfn(nid)) + NODE_DATA(nid)->node_mem_map;	\
 })
 
 #define page_to_pfn(page) \
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 4733d35d822..b79633d3a97 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -284,6 +284,8 @@ typedef struct pglist_data {
 
 #define node_present_pages(nid)	(NODE_DATA(nid)->node_present_pages)
 #define node_spanned_pages(nid)	(NODE_DATA(nid)->node_spanned_pages)
+#define pgdat_page_nr(pgdat, pagenr)	((pgdat)->node_mem_map + (pagenr))
+#define nid_page_nr(nid, pagenr) 	pgdat_page_nr(NODE_DATA(nid),(pagenr))
 
 extern struct pglist_data *pgdat_list;
 
-- 
cgit v1.2.3-70-g09d2


From 6f167ec721108c9282d54424516a12c805e3c306 Mon Sep 17 00:00:00 2001
From: Dave Hansen <haveblue@us.ibm.com>
Date: Thu, 23 Jun 2005 00:07:39 -0700
Subject: [PATCH] sparsemem base: simple NUMA remap space allocator

Introduce a simple allocator for the NUMA remap space.  This space is very
scarce, used for structures which are best allocated node local.

This mechanism is also used on non-NUMA ia64 systems with a vmem_map to keep
the pgdat->node_mem_map initialized in a consistent place for all
architectures.

Issues:
o alloc_remap takes a node_id where we might expect a pgdat which was intended
  to allow us to allocate the pgdat's using this mechanism; which we do not yet
  do.  Could have alloc_remap_node() and alloc_remap_nid() for this purpose.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig        |  5 ++++
 arch/i386/mm/discontig.c | 59 +++++++++++++++++++++++++-----------------------
 include/linux/bootmem.h  |  9 ++++++++
 mm/page_alloc.c          |  6 ++++-
 4 files changed, 50 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index dfd904f6883..35ca3a17ed2 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -803,6 +803,11 @@ config NEED_NODE_MEMMAP_SIZE
 	depends on DISCONTIGMEM
 	default y
 
+config HAVE_ARCH_ALLOC_REMAP
+	bool
+	depends on NUMA
+	default y
+
 config HIGHPTE
 	bool "Allocate 3rd-level pagetables from highmem"
 	depends on HIGHMEM4G || HIGHMEM64G
diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c
index 85d2fcbe107..dcc71f969b0 100644
--- a/arch/i386/mm/discontig.c
+++ b/arch/i386/mm/discontig.c
@@ -108,6 +108,9 @@ unsigned long node_remap_offset[MAX_NUMNODES];
 void *node_remap_start_vaddr[MAX_NUMNODES];
 void set_pmd_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags);
 
+void *node_remap_end_vaddr[MAX_NUMNODES];
+void *node_remap_alloc_vaddr[MAX_NUMNODES];
+
 /*
  * FLAT - support for basic PC memory model with discontig enabled, essentially
  *        a single node with all available processors in it with a flat
@@ -178,6 +181,21 @@ static void __init allocate_pgdat(int nid)
 	}
 }
 
+void *alloc_remap(int nid, unsigned long size)
+{
+	void *allocation = node_remap_alloc_vaddr[nid];
+
+	size = ALIGN(size, L1_CACHE_BYTES);
+
+	if (!allocation || (allocation + size) >= node_remap_end_vaddr[nid])
+		return 0;
+
+	node_remap_alloc_vaddr[nid] += size;
+	memset(allocation, 0, size);
+
+	return allocation;
+}
+
 void __init remap_numa_kva(void)
 {
 	void *vaddr;
@@ -185,8 +203,6 @@ void __init remap_numa_kva(void)
 	int node;
 
 	for_each_online_node(node) {
-		if (node == 0)
-			continue;
 		for (pfn=0; pfn < node_remap_size[node]; pfn += PTRS_PER_PTE) {
 			vaddr = node_remap_start_vaddr[node]+(pfn<<PAGE_SHIFT);
 			set_pmd_pfn((ulong) vaddr, 
@@ -202,11 +218,6 @@ static unsigned long calculate_numa_remap_pages(void)
 	unsigned long size, reserve_pages = 0;
 
 	for_each_online_node(nid) {
-		if (nid == 0)
-			continue;
-		if (!node_remap_size[nid])
-			continue;
-
 		/*
 		 * The acpi/srat node info can show hot-add memroy zones
 		 * where memory could be added but not currently present.
@@ -226,8 +237,8 @@ static unsigned long calculate_numa_remap_pages(void)
 		printk("Reserving %ld pages of KVA for lmem_map of node %d\n",
 				size, nid);
 		node_remap_size[nid] = size;
-		reserve_pages += size;
 		node_remap_offset[nid] = reserve_pages;
+		reserve_pages += size;
 		printk("Shrinking node %d from %ld pages to %ld pages\n",
 			nid, node_end_pfn[nid], node_end_pfn[nid] - size);
 		node_end_pfn[nid] -= size;
@@ -280,12 +291,18 @@ unsigned long __init setup_memory(void)
 			(ulong) pfn_to_kaddr(max_low_pfn));
 	for_each_online_node(nid) {
 		node_remap_start_vaddr[nid] = pfn_to_kaddr(
-			(highstart_pfn + reserve_pages) - node_remap_offset[nid]);
+				highstart_pfn + node_remap_offset[nid]);
+		/* Init the node remap allocator */
+		node_remap_end_vaddr[nid] = node_remap_start_vaddr[nid] +
+			(node_remap_size[nid] * PAGE_SIZE);
+		node_remap_alloc_vaddr[nid] = node_remap_start_vaddr[nid] +
+			ALIGN(sizeof(pg_data_t), PAGE_SIZE);
+
 		allocate_pgdat(nid);
 		printk ("node %d will remap to vaddr %08lx - %08lx\n", nid,
 			(ulong) node_remap_start_vaddr[nid],
-			(ulong) pfn_to_kaddr(highstart_pfn + reserve_pages
-			    - node_remap_offset[nid] + node_remap_size[nid]));
+			(ulong) pfn_to_kaddr(highstart_pfn
+			   + node_remap_offset[nid] + node_remap_size[nid]));
 	}
 	printk("High memory starts at vaddr %08lx\n",
 			(ulong) pfn_to_kaddr(highstart_pfn));
@@ -348,23 +365,9 @@ void __init zone_sizes_init(void)
 		}
 
 		zholes_size = get_zholes_size(nid);
-		/*
-		 * We let the lmem_map for node 0 be allocated from the
-		 * normal bootmem allocator, but other nodes come from the
-		 * remapped KVA area - mbligh
-		 */
-		if (!nid)
-			free_area_init_node(nid, NODE_DATA(nid),
-					zones_size, start, zholes_size);
-		else {
-			unsigned long lmem_map;
-			lmem_map = (unsigned long)node_remap_start_vaddr[nid];
-			lmem_map += sizeof(pg_data_t) + PAGE_SIZE - 1;
-			lmem_map &= PAGE_MASK;
-			NODE_DATA(nid)->node_mem_map = (struct page *)lmem_map;
-			free_area_init_node(nid, NODE_DATA(nid), zones_size,
-				start, zholes_size);
-		}
+
+		free_area_init_node(nid, NODE_DATA(nid), zones_size, start,
+				zholes_size);
 	}
 	return;
 }
diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index 0dd8ca1a3d5..500f451ce0c 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -67,6 +67,15 @@ extern void * __init __alloc_bootmem_node (pg_data_t *pgdat, unsigned long size,
 	__alloc_bootmem_node((pgdat), (x), PAGE_SIZE, 0)
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
 
+#ifdef CONFIG_HAVE_ARCH_ALLOC_REMAP
+extern void *alloc_remap(int nid, unsigned long size);
+#else
+static inline void *alloc_remap(int nid, unsigned long size)
+{
+	return NULL;
+}
+#endif
+
 extern unsigned long __initdata nr_kernel_pages;
 extern unsigned long __initdata nr_all_pages;
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 559336de968..bf1dd881909 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1936,6 +1936,7 @@ static void __init free_area_init_core(struct pglist_data *pgdat,
 static void __init alloc_node_mem_map(struct pglist_data *pgdat)
 {
 	unsigned long size;
+	struct page *map;
 
 	/* Skip empty nodes */
 	if (!pgdat->node_spanned_pages)
@@ -1944,7 +1945,10 @@ static void __init alloc_node_mem_map(struct pglist_data *pgdat)
 	/* ia64 gets its own node_mem_map, before this, without bootmem */
 	if (!pgdat->node_mem_map) {
 		size = (pgdat->node_spanned_pages + 1) * sizeof(struct page);
-		pgdat->node_mem_map = alloc_bootmem_node(pgdat, size);
+		map = alloc_remap(pgdat->node_id, size);
+		if (!map)
+			map = alloc_bootmem_node(pgdat, size);
+		pgdat->node_mem_map = map;
 	}
 #ifndef CONFIG_DISCONTIGMEM
 	/*
-- 
cgit v1.2.3-70-g09d2


From 348f8b6c4837a07304d2f72b11ce8d96588065e0 Mon Sep 17 00:00:00 2001
From: Dave Hansen <haveblue@us.ibm.com>
Date: Thu, 23 Jun 2005 00:07:40 -0700
Subject: [PATCH] sparsemem base: reorganize page->flags bit operations

Generify the value fields in the page_flags.  The aim is to allow the location
and size of these fields to be varied.  Additionally we want to move away from
fixed allocations per field whilst still enforcing the overall bit utilisation
limits.  We rely on the compiler to spot and optimise the accessor functions.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h     | 53 +++++++++++++++++++++++++++++++++++++++++---------
 include/linux/mmzone.h | 19 +++++++-----------
 mm/page_alloc.c        |  2 +-
 3 files changed, 52 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 1813b162b0a..57b2ead51db 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -395,19 +395,41 @@ static inline void put_page(struct page *page)
 /*
  * The zone field is never updated after free_area_init_core()
  * sets it, so none of the operations on it need to be atomic.
- * We'll have up to (MAX_NUMNODES * MAX_NR_ZONES) zones total,
- * so we use (MAX_NODES_SHIFT + MAX_ZONES_SHIFT) here to get enough bits.
  */
-#define NODEZONE_SHIFT (sizeof(page_flags_t)*8 - MAX_NODES_SHIFT - MAX_ZONES_SHIFT)
+
+/* Page flags: | NODE | ZONE | ... | FLAGS | */
+#define NODES_PGOFF		((sizeof(page_flags_t)*8) - NODES_SHIFT)
+#define ZONES_PGOFF		(NODES_PGOFF - ZONES_SHIFT)
+
+/*
+ * Define the bit shifts to access each section.  For non-existant
+ * sections we define the shift as 0; that plus a 0 mask ensures
+ * the compiler will optimise away reference to them.
+ */
+#define NODES_PGSHIFT		(NODES_PGOFF * (NODES_SHIFT != 0))
+#define ZONES_PGSHIFT		(ZONES_PGOFF * (ZONES_SHIFT != 0))
+
+/* NODE:ZONE is used to lookup the zone from a page. */
+#define ZONETABLE_SHIFT		(NODES_SHIFT + ZONES_SHIFT)
+#define ZONETABLE_PGSHIFT	ZONES_PGSHIFT
+
+#if NODES_SHIFT+ZONES_SHIFT > FLAGS_RESERVED
+#error NODES_SHIFT+ZONES_SHIFT > FLAGS_RESERVED
+#endif
+
 #define NODEZONE(node, zone)	((node << ZONES_SHIFT) | zone)
 
+#define ZONES_MASK		((1UL << ZONES_SHIFT) - 1)
+#define NODES_MASK		((1UL << NODES_SHIFT) - 1)
+#define ZONETABLE_MASK		((1UL << ZONETABLE_SHIFT) - 1)
+
 static inline unsigned long page_zonenum(struct page *page)
 {
-	return (page->flags >> NODEZONE_SHIFT) & (~(~0UL << ZONES_SHIFT));
+	return (page->flags >> ZONES_PGSHIFT) & ZONES_MASK;
 }
 static inline unsigned long page_to_nid(struct page *page)
 {
-	return (page->flags >> (NODEZONE_SHIFT + ZONES_SHIFT));
+	return (page->flags >> NODES_PGSHIFT) & NODES_MASK;
 }
 
 struct zone;
@@ -415,13 +437,26 @@ extern struct zone *zone_table[];
 
 static inline struct zone *page_zone(struct page *page)
 {
-	return zone_table[page->flags >> NODEZONE_SHIFT];
+	return zone_table[(page->flags >> ZONETABLE_PGSHIFT) &
+			ZONETABLE_MASK];
+}
+
+static inline void set_page_zone(struct page *page, unsigned long zone)
+{
+	page->flags &= ~(ZONES_MASK << ZONES_PGSHIFT);
+	page->flags |= (zone & ZONES_MASK) << ZONES_PGSHIFT;
+}
+static inline void set_page_node(struct page *page, unsigned long node)
+{
+	page->flags &= ~(NODES_MASK << NODES_PGSHIFT);
+	page->flags |= (node & NODES_MASK) << NODES_PGSHIFT;
 }
 
-static inline void set_page_zone(struct page *page, unsigned long nodezone_num)
+static inline void set_page_links(struct page *page, unsigned long zone,
+	unsigned long node)
 {
-	page->flags &= ~(~0UL << NODEZONE_SHIFT);
-	page->flags |= nodezone_num << NODEZONE_SHIFT;
+	set_page_zone(page, zone);
+	set_page_node(page, node);
 }
 
 #ifndef CONFIG_DISCONTIGMEM
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index b79633d3a97..39e912708e2 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -414,30 +414,25 @@ extern struct pglist_data contig_page_data;
 
 #include <asm/mmzone.h>
 
+#endif /* !CONFIG_DISCONTIGMEM */
+
 #if BITS_PER_LONG == 32 || defined(ARCH_HAS_ATOMIC_UNSIGNED)
 /*
  * with 32 bit page->flags field, we reserve 8 bits for node/zone info.
  * there are 3 zones (2 bits) and this leaves 8-2=6 bits for nodes.
  */
-#define MAX_NODES_SHIFT		6
+#define FLAGS_RESERVED		8
+
 #elif BITS_PER_LONG == 64
 /*
  * with 64 bit flags field, there's plenty of room.
  */
-#define MAX_NODES_SHIFT		10
-#endif
+#define FLAGS_RESERVED		32
 
-#endif /* !CONFIG_DISCONTIGMEM */
-
-#if NODES_SHIFT > MAX_NODES_SHIFT
-#error NODES_SHIFT > MAX_NODES_SHIFT
-#endif
+#else
 
-/* There are currently 3 zones: DMA, Normal & Highmem, thus we need 2 bits */
-#define MAX_ZONES_SHIFT		2
+#error BITS_PER_LONG not defined
 
-#if ZONES_SHIFT > MAX_ZONES_SHIFT
-#error ZONES_SHIFT > MAX_ZONES_SHIFT
 #endif
 
 #endif /* !__ASSEMBLY__ */
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index bf1dd881909..1958358e29b 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1653,7 +1653,7 @@ void __init memmap_init_zone(unsigned long size, int nid, unsigned long zone,
 	struct page *page;
 
 	for (page = start; page < (start + size); page++) {
-		set_page_zone(page, NODEZONE(nid, zone));
+		set_page_links(page, zone, nid);
 		set_page_count(page, 0);
 		reset_page_mapcount(page);
 		SetPageReserved(page);
-- 
cgit v1.2.3-70-g09d2


From 5b505b90b2d54e526cc8d123bdef3b98c9f0bbc6 Mon Sep 17 00:00:00 2001
From: Dave Hansen <haveblue@us.ibm.com>
Date: Thu, 23 Jun 2005 00:07:41 -0700
Subject: [PATCH] sparsemem base: teach discontig about sparse ranges

discontig.c has some assumptions that mem_map[]s inside of a node are
contiguous.  Teach it to make sure that each region that it's bringing online
is actually made up of valid ranges of ram.

Written-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mm/discontig.c | 14 ++++++++++++++
 arch/i386/mm/init.c      |  2 +-
 include/asm-i386/page.h  |  2 ++
 3 files changed, 17 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c
index dcc71f969b0..088ca472218 100644
--- a/arch/i386/mm/discontig.c
+++ b/arch/i386/mm/discontig.c
@@ -216,6 +216,7 @@ static unsigned long calculate_numa_remap_pages(void)
 {
 	int nid;
 	unsigned long size, reserve_pages = 0;
+	unsigned long pfn;
 
 	for_each_online_node(nid) {
 		/*
@@ -234,6 +235,19 @@ static unsigned long calculate_numa_remap_pages(void)
 		size = (size + LARGE_PAGE_BYTES - 1) / LARGE_PAGE_BYTES;
 		/* now the roundup is correct, convert to PAGE_SIZE pages */
 		size = size * PTRS_PER_PTE;
+
+		/*
+		 * Validate the region we are allocating only contains valid
+		 * pages.
+		 */
+		for (pfn = node_end_pfn[nid] - size;
+		     pfn < node_end_pfn[nid]; pfn++)
+			if (!page_is_ram(pfn))
+				break;
+
+		if (pfn != node_end_pfn[nid])
+			size = 0;
+
 		printk("Reserving %ld pages of KVA for lmem_map of node %d\n",
 				size, nid);
 		node_remap_size[nid] = size;
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index 8766c771bb4..666ca79fb50 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -191,7 +191,7 @@ static inline int page_kills_ppro(unsigned long pagenr)
 
 extern int is_available_memory(efi_memory_desc_t *);
 
-static inline int page_is_ram(unsigned long pagenr)
+int page_is_ram(unsigned long pagenr)
 {
 	int i;
 	unsigned long addr, end;
diff --git a/include/asm-i386/page.h b/include/asm-i386/page.h
index 41400d342d4..8f3dded01bf 100644
--- a/include/asm-i386/page.h
+++ b/include/asm-i386/page.h
@@ -120,6 +120,8 @@ static __inline__ int get_order(unsigned long size)
 
 extern int sysctl_legacy_va_layout;
 
+extern int page_is_ram(unsigned long pagenr);
+
 #endif /* __ASSEMBLY__ */
 
 #ifdef __ASSEMBLY__
-- 
cgit v1.2.3-70-g09d2


From 93b7504e3e6c1d98586854806e51bea329ea3aa9 Mon Sep 17 00:00:00 2001
From: Dave Hansen <haveblue@us.ibm.com>
Date: Thu, 23 Jun 2005 00:07:47 -0700
Subject: [PATCH] Introduce new Kconfig option for NUMA or DISCONTIG

There is some confusion that arose when working on SPARSEMEM patch between
what is needed for DISCONTIG vs. NUMA.

Multiple pg_data_t's are needed for DISCONTIGMEM or NUMA, independently.
All of the current NUMA implementations require an implementation of
DISCONTIG.  Because of this, quite a lot of code which is really needed for
NUMA is actually under DISCONTIG #ifdefs.  For SPARSEMEM, we changed some
of these #ifdefs to CONFIG_NUMA, but that broke the DISCONTIG=y and NUMA=n
case.

Introducing this new NEED_MULTIPLE_NODES config option allows code that is
needed for both NUMA or DISCONTIG to be separated out from code that is
specific to DISCONTIG.

One great advantage of this approach is that it doesn't require every
architecture to be converted over.  All of the current implementations
should "just work", only the ones implementing SPARSEMEM will have to be
fixed up.

The change to free_area_init() makes it work inside, or out of the new
config option.

Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mmzone.h | 6 +++---
 mm/Kconfig             | 8 ++++++++
 mm/page_alloc.c        | 6 +++---
 3 files changed, 14 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 39e912708e2..95f4a780ea6 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -402,7 +402,7 @@ int lowmem_reserve_ratio_sysctl_handler(struct ctl_table *, int, struct file *,
 /* Returns the number of the current Node. */
 #define numa_node_id()		(cpu_to_node(raw_smp_processor_id()))
 
-#ifndef CONFIG_DISCONTIGMEM
+#ifndef CONFIG_NEED_MULTIPLE_NODES
 
 extern struct pglist_data contig_page_data;
 #define NODE_DATA(nid)		(&contig_page_data)
@@ -410,11 +410,11 @@ extern struct pglist_data contig_page_data;
 #define MAX_NODES_SHIFT		1
 #define pfn_to_nid(pfn)		(0)
 
-#else /* CONFIG_DISCONTIGMEM */
+#else /* CONFIG_NEED_MULTIPLE_NODES */
 
 #include <asm/mmzone.h>
 
-#endif /* !CONFIG_DISCONTIGMEM */
+#endif /* !CONFIG_NEED_MULTIPLE_NODES */
 
 #if BITS_PER_LONG == 32 || defined(ARCH_HAS_ATOMIC_UNSIGNED)
 /*
diff --git a/mm/Kconfig b/mm/Kconfig
index 69caa9d8674..15c13139363 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -23,3 +23,11 @@ config DISCONTIGMEM
 
 endchoice
 
+#
+# Both the NUMA code and DISCONTIGMEM use arrays of pg_data_t's
+# to represent different areas of memory.  This variable allows
+# those dependencies to exist individually.
+#
+config NEED_MULTIPLE_NODES
+	def_bool y
+	depends on DISCONTIGMEM || NUMA
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 1958358e29b..20e239599db 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1972,18 +1972,18 @@ void __init free_area_init_node(int nid, struct pglist_data *pgdat,
 	free_area_init_core(pgdat, zones_size, zholes_size);
 }
 
-#ifndef CONFIG_DISCONTIGMEM
+#ifndef CONFIG_NEED_MULTIPLE_NODES
 static bootmem_data_t contig_bootmem_data;
 struct pglist_data contig_page_data = { .bdata = &contig_bootmem_data };
 
 EXPORT_SYMBOL(contig_page_data);
+#endif
 
 void __init free_area_init(unsigned long *zones_size)
 {
-	free_area_init_node(0, &contig_page_data, zones_size,
+	free_area_init_node(0, NODE_DATA(0), zones_size,
 			__pa(PAGE_OFFSET) >> PAGE_SHIFT, NULL);
 }
-#endif
 
 #ifdef CONFIG_PROC_FS
 
-- 
cgit v1.2.3-70-g09d2


From b159d43fbf7eaaac6ecc647f51cf4257332db47b Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Thu, 23 Jun 2005 00:07:52 -0700
Subject: [PATCH] generify early_pfn_to_nid

Provide a default implementation for early_pfn_to_nid returning node 0.  Allow
architectures to override this with their own implementation out of
asm/mmzone.h.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Martin Bligh <mbligh@aracnet.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig         | 4 ++++
 include/asm-i386/mmzone.h | 3 +++
 include/linux/mmzone.h    | 4 ++++
 3 files changed, 11 insertions(+)

(limited to 'include')

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index 8e5242c8e09..a8128f99733 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -810,6 +810,10 @@ config HAVE_ARCH_ALLOC_REMAP
 
 source "mm/Kconfig"
 
+config HAVE_ARCH_EARLY_PFN_TO_NID
+	bool
+	default y
+
 config HIGHPTE
 	bool "Allocate 3rd-level pagetables from highmem"
 	depends on HIGHMEM4G || HIGHMEM64G
diff --git a/include/asm-i386/mmzone.h b/include/asm-i386/mmzone.h
index 9cec191f462..48e46d403aa 100644
--- a/include/asm-i386/mmzone.h
+++ b/include/asm-i386/mmzone.h
@@ -143,4 +143,7 @@ static inline void get_memcfg_numa(void)
 }
 
 #endif /* CONFIG_DISCONTIGMEM */
+
+extern int early_pfn_to_nid(unsigned long pfn);
+
 #endif /* _ASM_MMZONE_H_ */
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 95f4a780ea6..6ef07de98d6 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -435,6 +435,10 @@ extern struct pglist_data contig_page_data;
 
 #endif
 
+#ifndef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID
+#define early_pfn_to_nid(nid)  (0UL)
+#endif
+
 #endif /* !__ASSEMBLY__ */
 #endif /* __KERNEL__ */
 #endif /* _LINUX_MMZONE_H */
-- 
cgit v1.2.3-70-g09d2


From d41dee369bff3b9dcb6328d4d822926c28cc2594 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Thu, 23 Jun 2005 00:07:54 -0700
Subject: [PATCH] sparsemem memory model

Sparsemem abstracts the use of discontiguous mem_maps[].  This kind of
mem_map[] is needed by discontiguous memory machines (like in the old
CONFIG_DISCONTIGMEM case) as well as memory hotplug systems.  Sparsemem
replaces DISCONTIGMEM when enabled, and it is hoped that it can eventually
become a complete replacement.

A significant advantage over DISCONTIGMEM is that it's completely separated
from CONFIG_NUMA.  When producing this patch, it became apparent in that NUMA
and DISCONTIG are often confused.

Another advantage is that sparse doesn't require each NUMA node's ranges to be
contiguous.  It can handle overlapping ranges between nodes with no problems,
where DISCONTIGMEM currently throws away that memory.

Sparsemem uses an array to provide different pfn_to_page() translations for
each SECTION_SIZE area of physical memory.  This is what allows the mem_map[]
to be chopped up.

In order to do quick pfn_to_page() operations, the section number of the page
is encoded in page->flags.  Part of the sparsemem infrastructure enables
sharing of these bits more dynamically (at compile-time) between the
page_zone() and sparsemem operations.  However, on 32-bit architectures, the
number of bits is quite limited, and may require growing the size of the
page->flags type in certain conditions.  Several things might force this to
occur: a decrease in the SECTION_SIZE (if you want to hotplug smaller areas of
memory), an increase in the physical address space, or an increase in the
number of used page->flags.

One thing to note is that, once sparsemem is present, the NUMA node
information no longer needs to be stored in the page->flags.  It might provide
speed increases on certain platforms and will be stored there if there is
room.  But, if out of room, an alternate (theoretically slower) mechanism is
used.

This patch introduces CONFIG_FLATMEM.  It is used in almost all cases where
there used to be an #ifndef DISCONTIG, because SPARSEMEM and DISCONTIGMEM
often have to compile out the same areas of code.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Martin Bligh <mbligh@aracnet.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Bob Picco <bob.picco@hp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig      |  1 +
 include/linux/mm.h     | 92 ++++++++++++++++++++++++++++++++++++++---------
 include/linux/mmzone.h | 96 ++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/numa.h   |  2 +-
 mm/Kconfig             | 38 ++++++++++++++++++--
 mm/Makefile            |  1 +
 mm/bootmem.c           |  9 +++--
 mm/memory.c            |  2 +-
 mm/page_alloc.c        | 39 +++++++++++++++-----
 mm/sparse.c            | 85 ++++++++++++++++++++++++++++++++++++++++++++
 10 files changed, 332 insertions(+), 33 deletions(-)
 create mode 100644 mm/sparse.c

(limited to 'include')

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index 3b7248126d2..f0064b5e370 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -813,6 +813,7 @@ source "mm/Kconfig"
 config HAVE_ARCH_EARLY_PFN_TO_NID
 	bool
 	default y
+	depends on NUMA
 
 config HIGHPTE
 	bool "Allocate 3rd-level pagetables from highmem"
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 57b2ead51db..6eb7f48317f 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -397,40 +397,80 @@ static inline void put_page(struct page *page)
  * sets it, so none of the operations on it need to be atomic.
  */
 
-/* Page flags: | NODE | ZONE | ... | FLAGS | */
-#define NODES_PGOFF		((sizeof(page_flags_t)*8) - NODES_SHIFT)
-#define ZONES_PGOFF		(NODES_PGOFF - ZONES_SHIFT)
+
+/*
+ * page->flags layout:
+ *
+ * There are three possibilities for how page->flags get
+ * laid out.  The first is for the normal case, without
+ * sparsemem.  The second is for sparsemem when there is
+ * plenty of space for node and section.  The last is when
+ * we have run out of space and have to fall back to an
+ * alternate (slower) way of determining the node.
+ *
+ *        No sparsemem: |       NODE     | ZONE | ... | FLAGS |
+ * with space for node: | SECTION | NODE | ZONE | ... | FLAGS |
+ *   no space for node: | SECTION |     ZONE    | ... | FLAGS |
+ */
+#ifdef CONFIG_SPARSEMEM
+#define SECTIONS_WIDTH		SECTIONS_SHIFT
+#else
+#define SECTIONS_WIDTH		0
+#endif
+
+#define ZONES_WIDTH		ZONES_SHIFT
+
+#if SECTIONS_WIDTH+ZONES_WIDTH+NODES_SHIFT <= FLAGS_RESERVED
+#define NODES_WIDTH		NODES_SHIFT
+#else
+#define NODES_WIDTH		0
+#endif
+
+/* Page flags: | [SECTION] | [NODE] | ZONE | ... | FLAGS | */
+#define SECTIONS_PGOFF		((sizeof(page_flags_t)*8) - SECTIONS_WIDTH)
+#define NODES_PGOFF		(SECTIONS_PGOFF - NODES_WIDTH)
+#define ZONES_PGOFF		(NODES_PGOFF - ZONES_WIDTH)
+
+/*
+ * We are going to use the flags for the page to node mapping if its in
+ * there.  This includes the case where there is no node, so it is implicit.
+ */
+#define FLAGS_HAS_NODE		(NODES_WIDTH > 0 || NODES_SHIFT == 0)
+
+#ifndef PFN_SECTION_SHIFT
+#define PFN_SECTION_SHIFT 0
+#endif
 
 /*
  * Define the bit shifts to access each section.  For non-existant
  * sections we define the shift as 0; that plus a 0 mask ensures
  * the compiler will optimise away reference to them.
  */
-#define NODES_PGSHIFT		(NODES_PGOFF * (NODES_SHIFT != 0))
-#define ZONES_PGSHIFT		(ZONES_PGOFF * (ZONES_SHIFT != 0))
+#define SECTIONS_PGSHIFT	(SECTIONS_PGOFF * (SECTIONS_WIDTH != 0))
+#define NODES_PGSHIFT		(NODES_PGOFF * (NODES_WIDTH != 0))
+#define ZONES_PGSHIFT		(ZONES_PGOFF * (ZONES_WIDTH != 0))
 
-/* NODE:ZONE is used to lookup the zone from a page. */
+/* NODE:ZONE or SECTION:ZONE is used to lookup the zone from a page. */
+#if FLAGS_HAS_NODE
 #define ZONETABLE_SHIFT		(NODES_SHIFT + ZONES_SHIFT)
+#else
+#define ZONETABLE_SHIFT		(SECTIONS_SHIFT + ZONES_SHIFT)
+#endif
 #define ZONETABLE_PGSHIFT	ZONES_PGSHIFT
 
-#if NODES_SHIFT+ZONES_SHIFT > FLAGS_RESERVED
-#error NODES_SHIFT+ZONES_SHIFT > FLAGS_RESERVED
+#if SECTIONS_WIDTH+NODES_WIDTH+ZONES_WIDTH > FLAGS_RESERVED
+#error SECTIONS_WIDTH+NODES_WIDTH+ZONES_WIDTH > FLAGS_RESERVED
 #endif
 
-#define NODEZONE(node, zone)	((node << ZONES_SHIFT) | zone)
-
-#define ZONES_MASK		((1UL << ZONES_SHIFT) - 1)
-#define NODES_MASK		((1UL << NODES_SHIFT) - 1)
+#define ZONES_MASK		((1UL << ZONES_WIDTH) - 1)
+#define NODES_MASK		((1UL << NODES_WIDTH) - 1)
+#define SECTIONS_MASK		((1UL << SECTIONS_WIDTH) - 1)
 #define ZONETABLE_MASK		((1UL << ZONETABLE_SHIFT) - 1)
 
 static inline unsigned long page_zonenum(struct page *page)
 {
 	return (page->flags >> ZONES_PGSHIFT) & ZONES_MASK;
 }
-static inline unsigned long page_to_nid(struct page *page)
-{
-	return (page->flags >> NODES_PGSHIFT) & NODES_MASK;
-}
 
 struct zone;
 extern struct zone *zone_table[];
@@ -441,6 +481,18 @@ static inline struct zone *page_zone(struct page *page)
 			ZONETABLE_MASK];
 }
 
+static inline unsigned long page_to_nid(struct page *page)
+{
+	if (FLAGS_HAS_NODE)
+		return (page->flags >> NODES_PGSHIFT) & NODES_MASK;
+	else
+		return page_zone(page)->zone_pgdat->node_id;
+}
+static inline unsigned long page_to_section(struct page *page)
+{
+	return (page->flags >> SECTIONS_PGSHIFT) & SECTIONS_MASK;
+}
+
 static inline void set_page_zone(struct page *page, unsigned long zone)
 {
 	page->flags &= ~(ZONES_MASK << ZONES_PGSHIFT);
@@ -451,12 +503,18 @@ static inline void set_page_node(struct page *page, unsigned long node)
 	page->flags &= ~(NODES_MASK << NODES_PGSHIFT);
 	page->flags |= (node & NODES_MASK) << NODES_PGSHIFT;
 }
+static inline void set_page_section(struct page *page, unsigned long section)
+{
+	page->flags &= ~(SECTIONS_MASK << SECTIONS_PGSHIFT);
+	page->flags |= (section & SECTIONS_MASK) << SECTIONS_PGSHIFT;
+}
 
 static inline void set_page_links(struct page *page, unsigned long zone,
-	unsigned long node)
+	unsigned long node, unsigned long pfn)
 {
 	set_page_zone(page, zone);
 	set_page_node(page, node);
+	set_page_section(page, pfn_to_section_nr(pfn));
 }
 
 #ifndef CONFIG_DISCONTIGMEM
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 6ef07de98d6..19860d317ec 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -269,7 +269,9 @@ typedef struct pglist_data {
 	struct zone node_zones[MAX_NR_ZONES];
 	struct zonelist node_zonelists[GFP_ZONETYPES];
 	int nr_zones;
+#ifdef CONFIG_FLAT_NODE_MEM_MAP
 	struct page *node_mem_map;
+#endif
 	struct bootmem_data *bdata;
 	unsigned long node_start_pfn;
 	unsigned long node_present_pages; /* total number of physical pages */
@@ -284,7 +286,11 @@ typedef struct pglist_data {
 
 #define node_present_pages(nid)	(NODE_DATA(nid)->node_present_pages)
 #define node_spanned_pages(nid)	(NODE_DATA(nid)->node_spanned_pages)
+#ifdef CONFIG_FLAT_NODE_MEM_MAP
 #define pgdat_page_nr(pgdat, pagenr)	((pgdat)->node_mem_map + (pagenr))
+#else
+#define pgdat_page_nr(pgdat, pagenr)	pfn_to_page((pgdat)->node_start_pfn + (pagenr))
+#endif
 #define nid_page_nr(nid, pagenr) 	pgdat_page_nr(NODE_DATA(nid),(pagenr))
 
 extern struct pglist_data *pgdat_list;
@@ -416,6 +422,10 @@ extern struct pglist_data contig_page_data;
 
 #endif /* !CONFIG_NEED_MULTIPLE_NODES */
 
+#ifdef CONFIG_SPARSEMEM
+#include <asm/sparsemem.h>
+#endif
+
 #if BITS_PER_LONG == 32 || defined(ARCH_HAS_ATOMIC_UNSIGNED)
 /*
  * with 32 bit page->flags field, we reserve 8 bits for node/zone info.
@@ -439,6 +449,92 @@ extern struct pglist_data contig_page_data;
 #define early_pfn_to_nid(nid)  (0UL)
 #endif
 
+#define pfn_to_section_nr(pfn) ((pfn) >> PFN_SECTION_SHIFT)
+#define section_nr_to_pfn(sec) ((sec) << PFN_SECTION_SHIFT)
+
+#ifdef CONFIG_SPARSEMEM
+
+/*
+ * SECTION_SHIFT    		#bits space required to store a section #
+ *
+ * PA_SECTION_SHIFT		physical address to/from section number
+ * PFN_SECTION_SHIFT		pfn to/from section number
+ */
+#define SECTIONS_SHIFT		(MAX_PHYSMEM_BITS - SECTION_SIZE_BITS)
+
+#define PA_SECTION_SHIFT	(SECTION_SIZE_BITS)
+#define PFN_SECTION_SHIFT	(SECTION_SIZE_BITS - PAGE_SHIFT)
+
+#define NR_MEM_SECTIONS		(1UL << SECTIONS_SHIFT)
+
+#define PAGES_PER_SECTION       (1UL << PFN_SECTION_SHIFT)
+#define PAGE_SECTION_MASK	(~(PAGES_PER_SECTION-1))
+
+#if (MAX_ORDER - 1 + PAGE_SHIFT) > SECTION_SIZE_BITS
+#error Allocator MAX_ORDER exceeds SECTION_SIZE
+#endif
+
+struct page;
+struct mem_section {
+	struct page *section_mem_map;
+};
+
+extern struct mem_section mem_section[NR_MEM_SECTIONS];
+
+/*
+ * Given a kernel address, find the home node of the underlying memory.
+ */
+#define kvaddr_to_nid(kaddr)	pfn_to_nid(__pa(kaddr) >> PAGE_SHIFT)
+
+static inline struct mem_section *__pfn_to_section(unsigned long pfn)
+{
+	return &mem_section[pfn_to_section_nr(pfn)];
+}
+
+#define pfn_to_page(pfn) 						\
+({ 									\
+	unsigned long __pfn = (pfn);					\
+	__pfn_to_section(__pfn)->section_mem_map + __pfn;		\
+})
+#define page_to_pfn(page)						\
+({									\
+	page - mem_section[page_to_section(page)].section_mem_map;	\
+})
+
+static inline int pfn_valid(unsigned long pfn)
+{
+	if (pfn_to_section_nr(pfn) >= NR_MEM_SECTIONS)
+		return 0;
+	return mem_section[pfn_to_section_nr(pfn)].section_mem_map != 0;
+}
+
+/*
+ * These are _only_ used during initialisation, therefore they
+ * can use __initdata ...  They could have names to indicate
+ * this restriction.
+ */
+#ifdef CONFIG_NUMA
+#define pfn_to_nid		early_pfn_to_nid
+#endif
+
+#define pfn_to_pgdat(pfn)						\
+({									\
+	NODE_DATA(pfn_to_nid(pfn));					\
+})
+
+#define early_pfn_valid(pfn)	pfn_valid(pfn)
+void sparse_init(void);
+#else
+#define sparse_init()	do {} while (0)
+#endif /* CONFIG_SPARSEMEM */
+
+#ifndef early_pfn_valid
+#define early_pfn_valid(pfn)	(1)
+#endif
+
+void memory_present(int nid, unsigned long start, unsigned long end);
+unsigned long __init node_memmap_size_bytes(int, unsigned long, unsigned long);
+
 #endif /* !__ASSEMBLY__ */
 #endif /* __KERNEL__ */
 #endif /* _LINUX_MMZONE_H */
diff --git a/include/linux/numa.h b/include/linux/numa.h
index bd0c8c4e9a9..f0c539bd3cf 100644
--- a/include/linux/numa.h
+++ b/include/linux/numa.h
@@ -3,7 +3,7 @@
 
 #include <linux/config.h>
 
-#ifdef CONFIG_DISCONTIGMEM
+#ifndef CONFIG_FLATMEM
 #include <asm/numnodes.h>
 #endif
 
diff --git a/mm/Kconfig b/mm/Kconfig
index 5127441561b..cd379936cac 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -6,6 +6,7 @@ choice
 	prompt "Memory model"
 	depends on SELECT_MEMORY_MODEL
 	default DISCONTIGMEM_MANUAL if ARCH_DISCONTIGMEM_DEFAULT
+	default SPARSEMEM_MANUAL if ARCH_SPARSEMEM_DEFAULT
 	default FLATMEM_MANUAL
 
 config FLATMEM_MANUAL
@@ -17,7 +18,15 @@ config FLATMEM_MANUAL
 	  only have one option here: FLATMEM.  This is normal
 	  and a correct option.
 
-	  If unsure, choose this option over any other.
+	  Some users of more advanced features like NUMA and
+	  memory hotplug may have different options here.
+	  DISCONTIGMEM is an more mature, better tested system,
+	  but is incompatible with memory hotplug and may suffer
+	  decreased performance over SPARSEMEM.  If unsure between
+	  "Sparse Memory" and "Discontiguous Memory", choose
+	  "Discontiguous Memory".
+
+	  If unsure, choose this option (Flat Memory) over any other.
 
 config DISCONTIGMEM_MANUAL
 	bool "Discontigious Memory"
@@ -35,15 +44,38 @@ config DISCONTIGMEM_MANUAL
 
 	  If unsure, choose "Flat Memory" over this option.
 
+config SPARSEMEM_MANUAL
+	bool "Sparse Memory"
+	depends on ARCH_SPARSEMEM_ENABLE
+	help
+	  This will be the only option for some systems, including
+	  memory hotplug systems.  This is normal.
+
+	  For many other systems, this will be an alternative to
+	  "Discontigious Memory".  This option provides some potential
+	  performance benefits, along with decreased code complexity,
+	  but it is newer, and more experimental.
+
+	  If unsure, choose "Discontiguous Memory" or "Flat Memory"
+	  over this option.
+
 endchoice
 
 config DISCONTIGMEM
 	def_bool y
 	depends on (!SELECT_MEMORY_MODEL && ARCH_DISCONTIGMEM_ENABLE) || DISCONTIGMEM_MANUAL
 
+config SPARSEMEM
+	def_bool y
+	depends on SPARSEMEM_MANUAL
+
 config FLATMEM
 	def_bool y
-	depends on !DISCONTIGMEM || FLATMEM_MANUAL
+	depends on (!DISCONTIGMEM && !SPARSEMEM) || FLATMEM_MANUAL
+
+config FLAT_NODE_MEM_MAP
+	def_bool y
+	depends on !SPARSEMEM
 
 #
 # Both the NUMA code and DISCONTIGMEM use arrays of pg_data_t's
@@ -56,4 +88,4 @@ config NEED_MULTIPLE_NODES
 
 config HAVE_MEMORY_PRESENT
 	def_bool y
-	depends on ARCH_HAVE_MEMORY_PRESENT
+	depends on ARCH_HAVE_MEMORY_PRESENT || SPARSEMEM
diff --git a/mm/Makefile b/mm/Makefile
index 097408064f6..8f70ffd763c 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -15,6 +15,7 @@ obj-y			:= bootmem.o filemap.o mempool.o oom_kill.o fadvise.o \
 obj-$(CONFIG_SWAP)	+= page_io.o swap_state.o swapfile.o thrash.o
 obj-$(CONFIG_HUGETLBFS)	+= hugetlb.o
 obj-$(CONFIG_NUMA) 	+= mempolicy.o
+obj-$(CONFIG_SPARSEMEM)	+= sparse.o
 obj-$(CONFIG_SHMEM) += shmem.o
 obj-$(CONFIG_TINY_SHMEM) += tiny-shmem.o
 
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 260e703850d..f82f7aebbee 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -256,6 +256,7 @@ found:
 static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 {
 	struct page *page;
+	unsigned long pfn;
 	bootmem_data_t *bdata = pgdat->bdata;
 	unsigned long i, count, total = 0;
 	unsigned long idx;
@@ -266,7 +267,7 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 
 	count = 0;
 	/* first extant page of the node */
-	page = virt_to_page(phys_to_virt(bdata->node_boot_start));
+	pfn = bdata->node_boot_start >> PAGE_SHIFT;
 	idx = bdata->node_low_pfn - (bdata->node_boot_start >> PAGE_SHIFT);
 	map = bdata->node_bootmem_map;
 	/* Check physaddr is O(LOG2(BITS_PER_LONG)) page aligned */
@@ -275,9 +276,11 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 		gofast = 1;
 	for (i = 0; i < idx; ) {
 		unsigned long v = ~map[i / BITS_PER_LONG];
+
 		if (gofast && v == ~0UL) {
 			int j, order;
 
+			page = pfn_to_page(pfn);
 			count += BITS_PER_LONG;
 			__ClearPageReserved(page);
 			order = ffs(BITS_PER_LONG) - 1;
@@ -292,6 +295,8 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 			page += BITS_PER_LONG;
 		} else if (v) {
 			unsigned long m;
+
+			page = pfn_to_page(pfn);
 			for (m = 1; m && i < idx; m<<=1, page++, i++) {
 				if (v & m) {
 					count++;
@@ -302,8 +307,8 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 			}
 		} else {
 			i+=BITS_PER_LONG;
-			page += BITS_PER_LONG;
 		}
+		pfn += BITS_PER_LONG;
 	}
 	total += count;
 
diff --git a/mm/memory.c b/mm/memory.c
index da91b7bf998..30975ef4872 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -58,7 +58,7 @@
 #include <linux/swapops.h>
 #include <linux/elf.h>
 
-#ifndef CONFIG_DISCONTIGMEM
+#ifndef CONFIG_NEED_MULTIPLE_NODES
 /* use the per-pgdat data instead for discontigmem - mbligh */
 unsigned long max_mapnr;
 struct page *mem_map;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 20e239599db..5c1b8982a6d 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -68,7 +68,7 @@ EXPORT_SYMBOL(nr_swap_pages);
  * Used by page_zone() to look up the address of the struct zone whose
  * id is encoded in the upper bits of page->flags
  */
-struct zone *zone_table[1 << (ZONES_SHIFT + NODES_SHIFT)];
+struct zone *zone_table[1 << ZONETABLE_SHIFT];
 EXPORT_SYMBOL(zone_table);
 
 static char *zone_names[MAX_NR_ZONES] = { "DMA", "Normal", "HighMem" };
@@ -1649,11 +1649,15 @@ static void __init calculate_zone_totalpages(struct pglist_data *pgdat,
 void __init memmap_init_zone(unsigned long size, int nid, unsigned long zone,
 		unsigned long start_pfn)
 {
-	struct page *start = pfn_to_page(start_pfn);
 	struct page *page;
+	int end_pfn = start_pfn + size;
+	int pfn;
 
-	for (page = start; page < (start + size); page++) {
-		set_page_links(page, zone, nid);
+	for (pfn = start_pfn; pfn < end_pfn; pfn++, page++) {
+		if (!early_pfn_valid(pfn))
+			continue;
+		page = pfn_to_page(pfn);
+		set_page_links(page, zone, nid, pfn);
 		set_page_count(page, 0);
 		reset_page_mapcount(page);
 		SetPageReserved(page);
@@ -1677,6 +1681,20 @@ void zone_init_free_lists(struct pglist_data *pgdat, struct zone *zone,
 	}
 }
 
+#define ZONETABLE_INDEX(x, zone_nr)	((x << ZONES_SHIFT) | zone_nr)
+void zonetable_add(struct zone *zone, int nid, int zid, unsigned long pfn,
+		unsigned long size)
+{
+	unsigned long snum = pfn_to_section_nr(pfn);
+	unsigned long end = pfn_to_section_nr(pfn + size);
+
+	if (FLAGS_HAS_NODE)
+		zone_table[ZONETABLE_INDEX(nid, zid)] = zone;
+	else
+		for (; snum <= end; snum++)
+			zone_table[ZONETABLE_INDEX(snum, zid)] = zone;
+}
+
 #ifndef __HAVE_ARCH_MEMMAP_INIT
 #define memmap_init(size, nid, zone, start_pfn) \
 	memmap_init_zone((size), (nid), (zone), (start_pfn))
@@ -1861,7 +1879,6 @@ static void __init free_area_init_core(struct pglist_data *pgdat,
 		unsigned long size, realsize;
 		unsigned long batch;
 
-		zone_table[NODEZONE(nid, j)] = zone;
 		realsize = size = zones_size[j];
 		if (zholes_size)
 			realsize -= zholes_size[j];
@@ -1927,6 +1944,8 @@ static void __init free_area_init_core(struct pglist_data *pgdat,
 
 		memmap_init(size, nid, j, zone_start_pfn);
 
+		zonetable_add(zone, nid, j, zone_start_pfn, size);
+
 		zone_start_pfn += size;
 
 		zone_init_free_lists(pgdat, zone, zone->spanned_pages);
@@ -1935,28 +1954,30 @@ static void __init free_area_init_core(struct pglist_data *pgdat,
 
 static void __init alloc_node_mem_map(struct pglist_data *pgdat)
 {
-	unsigned long size;
-	struct page *map;
-
 	/* Skip empty nodes */
 	if (!pgdat->node_spanned_pages)
 		return;
 
+#ifdef CONFIG_FLAT_NODE_MEM_MAP
 	/* ia64 gets its own node_mem_map, before this, without bootmem */
 	if (!pgdat->node_mem_map) {
+		unsigned long size;
+		struct page *map;
+
 		size = (pgdat->node_spanned_pages + 1) * sizeof(struct page);
 		map = alloc_remap(pgdat->node_id, size);
 		if (!map)
 			map = alloc_bootmem_node(pgdat, size);
 		pgdat->node_mem_map = map;
 	}
-#ifndef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_FLATMEM
 	/*
 	 * With no DISCONTIG, the global mem_map is just set as node 0's
 	 */
 	if (pgdat == NODE_DATA(0))
 		mem_map = NODE_DATA(0)->node_mem_map;
 #endif
+#endif /* CONFIG_FLAT_NODE_MEM_MAP */
 }
 
 void __init free_area_init_node(int nid, struct pglist_data *pgdat,
diff --git a/mm/sparse.c b/mm/sparse.c
new file mode 100644
index 00000000000..f888385b9e1
--- /dev/null
+++ b/mm/sparse.c
@@ -0,0 +1,85 @@
+/*
+ * sparse memory mappings.
+ */
+#include <linux/config.h>
+#include <linux/mm.h>
+#include <linux/mmzone.h>
+#include <linux/bootmem.h>
+#include <linux/module.h>
+#include <asm/dma.h>
+
+/*
+ * Permanent SPARSEMEM data:
+ *
+ * 1) mem_section	- memory sections, mem_map's for valid memory
+ */
+struct mem_section mem_section[NR_MEM_SECTIONS];
+EXPORT_SYMBOL(mem_section);
+
+/* Record a memory area against a node. */
+void memory_present(int nid, unsigned long start, unsigned long end)
+{
+	unsigned long pfn;
+
+	start &= PAGE_SECTION_MASK;
+	for (pfn = start; pfn < end; pfn += PAGES_PER_SECTION) {
+		unsigned long section = pfn_to_section_nr(pfn);
+		if (!mem_section[section].section_mem_map)
+			mem_section[section].section_mem_map = (void *) -1;
+	}
+}
+
+/*
+ * Only used by the i386 NUMA architecures, but relatively
+ * generic code.
+ */
+unsigned long __init node_memmap_size_bytes(int nid, unsigned long start_pfn,
+						     unsigned long end_pfn)
+{
+	unsigned long pfn;
+	unsigned long nr_pages = 0;
+
+	for (pfn = start_pfn; pfn < end_pfn; pfn += PAGES_PER_SECTION) {
+		if (nid != early_pfn_to_nid(pfn))
+			continue;
+
+		if (pfn_valid(pfn))
+			nr_pages += PAGES_PER_SECTION;
+	}
+
+	return nr_pages * sizeof(struct page);
+}
+
+/*
+ * Allocate the accumulated non-linear sections, allocate a mem_map
+ * for each and record the physical to section mapping.
+ */
+void sparse_init(void)
+{
+	unsigned long pnum;
+	struct page *map;
+	int nid;
+
+	for (pnum = 0; pnum < NR_MEM_SECTIONS; pnum++) {
+		if (!mem_section[pnum].section_mem_map)
+			continue;
+
+		nid = early_pfn_to_nid(section_nr_to_pfn(pnum));
+		map = alloc_remap(nid, sizeof(struct page) * PAGES_PER_SECTION);
+		if (!map)
+			map = alloc_bootmem_node(NODE_DATA(nid),
+				sizeof(struct page) * PAGES_PER_SECTION);
+		if (!map) {
+			mem_section[pnum].section_mem_map = 0;
+			continue;
+		}
+
+		/*
+		 * Subtle, we encode the real pfn into the mem_map such that
+		 * the identity pfn - section_mem_map will return the actual
+		 * physical page frame number.
+		 */
+		mem_section[pnum].section_mem_map = map -
+						section_nr_to_pfn(pnum);
+	}
+}
-- 
cgit v1.2.3-70-g09d2


From 05b79bdcb48c18cd9b580c39e3efb9a1ab078151 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Thu, 23 Jun 2005 00:07:57 -0700
Subject: [PATCH] sparsemem memory model for i386

Provide the architecture specific implementation for SPARSEMEM for i386 SMP
and NUMA systems.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Martin Bligh <mbligh@aracnet.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig            | 24 ++++++++----
 arch/i386/kernel/setup.c     |  8 ++--
 arch/i386/mm/Makefile        |  2 +-
 arch/i386/mm/discontig.c     | 36 ++++++++++--------
 arch/i386/mm/init.c          | 18 ++++-----
 include/asm-i386/mmzone.h    | 89 ++++++++++++++++++++++++--------------------
 include/asm-i386/page.h      |  4 +-
 include/asm-i386/pgtable.h   |  4 +-
 include/asm-i386/sparsemem.h | 31 +++++++++++++++
 9 files changed, 135 insertions(+), 81 deletions(-)
 create mode 100644 include/asm-i386/sparsemem.h

(limited to 'include')

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index f0064b5e370..bfdcedef06e 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -68,7 +68,6 @@ config X86_VOYAGER
 
 config X86_NUMAQ
 	bool "NUMAQ (IBM/Sequent)"
-	select DISCONTIGMEM
 	select NUMA
 	help
 	  This option is used for getting Linux to run on a (IBM/Sequent) NUMA
@@ -783,11 +782,6 @@ comment "NUMA (NUMA-Q) requires SMP, 64GB highmem support"
 comment "NUMA (Summit) requires SMP, 64GB highmem support, ACPI"
 	depends on X86_SUMMIT && (!HIGHMEM64G || !ACPI)
 
-config ARCH_DISCONTIGMEM_ENABLE
-	bool
-	depends on NUMA
-	default y
-
 config HAVE_ARCH_BOOTMEM_NODE
 	bool
 	depends on NUMA
@@ -800,7 +794,7 @@ config ARCH_HAVE_MEMORY_PRESENT
 
 config NEED_NODE_MEMMAP_SIZE
 	bool
-	depends on DISCONTIGMEM
+	depends on DISCONTIGMEM || SPARSEMEM
 	default y
 
 config HAVE_ARCH_ALLOC_REMAP
@@ -808,6 +802,22 @@ config HAVE_ARCH_ALLOC_REMAP
 	depends on NUMA
 	default y
 
+config ARCH_DISCONTIGMEM_ENABLE
+	def_bool y
+	depends on NUMA
+
+config ARCH_DISCONTIGMEM_DEFAULT
+	def_bool y
+	depends on NUMA
+
+config ARCH_SPARSEMEM_ENABLE
+	def_bool y
+	depends on NUMA
+
+config ARCH_SELECT_MEMORY_MODEL
+	def_bool y
+	depends on ARCH_SPARSEMEM_ENABLE
+
 source "mm/Kconfig"
 
 config HAVE_ARCH_EARLY_PFN_TO_NID
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index 2bfbddebdbf..0d689335c8d 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -25,6 +25,7 @@
 
 #include <linux/sched.h>
 #include <linux/mm.h>
+#include <linux/mmzone.h>
 #include <linux/tty.h>
 #include <linux/ioport.h>
 #include <linux/acpi.h>
@@ -1022,7 +1023,7 @@ static void __init reserve_ebda_region(void)
 		reserve_bootmem(addr, PAGE_SIZE);	
 }
 
-#ifndef CONFIG_DISCONTIGMEM
+#ifndef CONFIG_NEED_MULTIPLE_NODES
 void __init setup_bootmem_allocator(void);
 static unsigned long __init setup_memory(void)
 {
@@ -1072,9 +1073,9 @@ void __init zone_sizes_init(void)
 	free_area_init(zones_size);
 }
 #else
-extern unsigned long setup_memory(void);
+extern unsigned long __init setup_memory(void);
 extern void zone_sizes_init(void);
-#endif /* !CONFIG_DISCONTIGMEM */
+#endif /* !CONFIG_NEED_MULTIPLE_NODES */
 
 void __init setup_bootmem_allocator(void)
 {
@@ -1475,6 +1476,7 @@ void __init setup_arch(char **cmdline_p)
 #endif
 	paging_init();
 	remapped_pgdat_init();
+	sparse_init();
 	zone_sizes_init();
 
 	/*
diff --git a/arch/i386/mm/Makefile b/arch/i386/mm/Makefile
index fc327250684..80908b5aa60 100644
--- a/arch/i386/mm/Makefile
+++ b/arch/i386/mm/Makefile
@@ -4,7 +4,7 @@
 
 obj-y	:= init.o pgtable.o fault.o ioremap.o extable.o pageattr.o mmap.o
 
-obj-$(CONFIG_DISCONTIGMEM)	+= discontig.o
+obj-$(CONFIG_NUMA) += discontig.o
 obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o
 obj-$(CONFIG_HIGHMEM) += highmem.o
 obj-$(CONFIG_BOOT_IOREMAP) += boot_ioremap.o
diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c
index 088ca472218..0efeb96ba5d 100644
--- a/arch/i386/mm/discontig.c
+++ b/arch/i386/mm/discontig.c
@@ -42,12 +42,16 @@ bootmem_data_t node0_bdata;
  *                  populated the following initialisation.
  *
  * 1) node_online_map  - the map of all nodes configured (online) in the system
- * 2) physnode_map     - the mapping between a pfn and owning node
- * 3) node_start_pfn   - the starting page frame number for a node
+ * 2) node_start_pfn   - the starting page frame number for a node
  * 3) node_end_pfn     - the ending page fram number for a node
  */
+unsigned long node_start_pfn[MAX_NUMNODES];
+unsigned long node_end_pfn[MAX_NUMNODES];
+
 
+#ifdef CONFIG_DISCONTIGMEM
 /*
+ * 4) physnode_map     - the mapping between a pfn and owning node
  * physnode_map keeps track of the physical memory layout of a generic
  * numa node on a 256Mb break (each element of the array will
  * represent 256Mb of memory and will be marked by the node id.  so,
@@ -85,9 +89,7 @@ unsigned long node_memmap_size_bytes(int nid, unsigned long start_pfn,
 
 	return (nr_pages + 1) * sizeof(struct page);
 }
-
-unsigned long node_start_pfn[MAX_NUMNODES];
-unsigned long node_end_pfn[MAX_NUMNODES];
+#endif
 
 extern unsigned long find_max_low_pfn(void);
 extern void find_max_pfn(void);
@@ -390,24 +392,26 @@ void __init set_highmem_pages_init(int bad_ppro)
 {
 #ifdef CONFIG_HIGHMEM
 	struct zone *zone;
+	struct page *page;
 
 	for_each_zone(zone) {
-		unsigned long node_pfn, node_high_size, zone_start_pfn;
-		struct page * zone_mem_map;
-		
+		unsigned long node_pfn, zone_start_pfn, zone_end_pfn;
+
 		if (!is_highmem(zone))
 			continue;
 
-		printk("Initializing %s for node %d\n", zone->name,
-			zone->zone_pgdat->node_id);
-
-		node_high_size = zone->spanned_pages;
-		zone_mem_map = zone->zone_mem_map;
 		zone_start_pfn = zone->zone_start_pfn;
+		zone_end_pfn = zone_start_pfn + zone->spanned_pages;
+
+		printk("Initializing %s for node %d (%08lx:%08lx)\n",
+				zone->name, zone->zone_pgdat->node_id,
+				zone_start_pfn, zone_end_pfn);
 
-		for (node_pfn = 0; node_pfn < node_high_size; node_pfn++) {
-			one_highpage_init((struct page *)(zone_mem_map + node_pfn),
-					  zone_start_pfn + node_pfn, bad_ppro);
+		for (node_pfn = zone_start_pfn; node_pfn < zone_end_pfn; node_pfn++) {
+			if (!pfn_valid(node_pfn))
+				continue;
+			page = pfn_to_page(node_pfn);
+			one_highpage_init(page, node_pfn, bad_ppro);
 		}
 	}
 	totalram_pages += totalhigh_pages;
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index 666ca79fb50..48ebfab77a3 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -276,7 +276,9 @@ void __init one_highpage_init(struct page *page, int pfn, int bad_ppro)
 		SetPageReserved(page);
 }
 
-#ifndef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_NUMA
+extern void set_highmem_pages_init(int);
+#else
 static void __init set_highmem_pages_init(int bad_ppro)
 {
 	int pfn;
@@ -284,9 +286,7 @@ static void __init set_highmem_pages_init(int bad_ppro)
 		one_highpage_init(pfn_to_page(pfn), pfn, bad_ppro);
 	totalram_pages += totalhigh_pages;
 }
-#else
-extern void set_highmem_pages_init(int);
-#endif /* !CONFIG_DISCONTIGMEM */
+#endif /* CONFIG_FLATMEM */
 
 #else
 #define kmap_init() do { } while (0)
@@ -297,10 +297,10 @@ extern void set_highmem_pages_init(int);
 unsigned long long __PAGE_KERNEL = _PAGE_KERNEL;
 unsigned long long __PAGE_KERNEL_EXEC = _PAGE_KERNEL_EXEC;
 
-#ifndef CONFIG_DISCONTIGMEM
-#define remap_numa_kva() do {} while (0)
-#else
+#ifdef CONFIG_NUMA
 extern void __init remap_numa_kva(void);
+#else
+#define remap_numa_kva() do {} while (0)
 #endif
 
 static void __init pagetable_init (void)
@@ -525,7 +525,7 @@ static void __init set_max_mapnr_init(void)
 #else
 	num_physpages = max_low_pfn;
 #endif
-#ifndef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_FLATMEM
 	max_mapnr = num_physpages;
 #endif
 }
@@ -539,7 +539,7 @@ void __init mem_init(void)
 	int tmp;
 	int bad_ppro;
 
-#ifndef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_FLATMEM
 	if (!mem_map)
 		BUG();
 #endif
diff --git a/include/asm-i386/mmzone.h b/include/asm-i386/mmzone.h
index 48e46d403aa..33ce5d37e89 100644
--- a/include/asm-i386/mmzone.h
+++ b/include/asm-i386/mmzone.h
@@ -8,7 +8,9 @@
 
 #include <asm/smp.h>
 
-#ifdef CONFIG_DISCONTIGMEM
+#if CONFIG_NUMA
+extern struct pglist_data *node_data[];
+#define NODE_DATA(nid)	(node_data[nid])
 
 #ifdef CONFIG_NUMA
 	#ifdef CONFIG_X86_NUMAQ
@@ -21,8 +23,28 @@
 	#define get_zholes_size(n) (0)
 #endif /* CONFIG_NUMA */
 
-extern struct pglist_data *node_data[];
-#define NODE_DATA(nid)		(node_data[nid])
+extern int get_memcfg_numa_flat(void );
+/*
+ * This allows any one NUMA architecture to be compiled
+ * for, and still fall back to the flat function if it
+ * fails.
+ */
+static inline void get_memcfg_numa(void)
+{
+#ifdef CONFIG_X86_NUMAQ
+	if (get_memcfg_numaq())
+		return;
+#elif CONFIG_ACPI_SRAT
+	if (get_memcfg_from_srat())
+		return;
+#endif
+
+	get_memcfg_numa_flat();
+}
+
+#endif /* CONFIG_NUMA */
+
+#ifdef CONFIG_DISCONTIGMEM
 
 /*
  * generic node memory support, the following assumptions apply:
@@ -48,26 +70,6 @@ static inline int pfn_to_nid(unsigned long pfn)
 #endif
 }
 
-/*
- * Following are macros that are specific to this numa platform.
- */
-#define reserve_bootmem(addr, size) \
-	reserve_bootmem_node(NODE_DATA(0), (addr), (size))
-#define alloc_bootmem(x) \
-	__alloc_bootmem_node(NODE_DATA(0), (x), SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
-#define alloc_bootmem_low(x) \
-	__alloc_bootmem_node(NODE_DATA(0), (x), SMP_CACHE_BYTES, 0)
-#define alloc_bootmem_pages(x) \
-	__alloc_bootmem_node(NODE_DATA(0), (x), PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
-#define alloc_bootmem_low_pages(x) \
-	__alloc_bootmem_node(NODE_DATA(0), (x), PAGE_SIZE, 0)
-#define alloc_bootmem_node(ignore, x) \
-	__alloc_bootmem_node(NODE_DATA(0), (x), SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
-#define alloc_bootmem_pages_node(ignore, x) \
-	__alloc_bootmem_node(NODE_DATA(0), (x), PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
-#define alloc_bootmem_low_pages_node(ignore, x) \
-	__alloc_bootmem_node(NODE_DATA(0), (x), PAGE_SIZE, 0)
-
 #define node_localnr(pfn, nid)		((pfn) - node_data[nid]->node_start_pfn)
 
 /*
@@ -121,28 +123,33 @@ static inline int pfn_valid(int pfn)
 		return (pfn < node_end_pfn(nid));
 	return 0;
 }
-#endif
+#endif /* CONFIG_X86_NUMAQ */
+
+#endif /* CONFIG_DISCONTIGMEM */
+
+#ifdef CONFIG_NEED_MULTIPLE_NODES
 
-extern int get_memcfg_numa_flat(void );
 /*
- * This allows any one NUMA architecture to be compiled
- * for, and still fall back to the flat function if it
- * fails.
+ * Following are macros that are specific to this numa platform.
  */
-static inline void get_memcfg_numa(void)
-{
-#ifdef CONFIG_X86_NUMAQ
-	if (get_memcfg_numaq())
-		return;
-#elif CONFIG_ACPI_SRAT
-	if (get_memcfg_from_srat())
-		return;
-#endif
-
-	get_memcfg_numa_flat();
-}
+#define reserve_bootmem(addr, size) \
+	reserve_bootmem_node(NODE_DATA(0), (addr), (size))
+#define alloc_bootmem(x) \
+	__alloc_bootmem_node(NODE_DATA(0), (x), SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
+#define alloc_bootmem_low(x) \
+	__alloc_bootmem_node(NODE_DATA(0), (x), SMP_CACHE_BYTES, 0)
+#define alloc_bootmem_pages(x) \
+	__alloc_bootmem_node(NODE_DATA(0), (x), PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
+#define alloc_bootmem_low_pages(x) \
+	__alloc_bootmem_node(NODE_DATA(0), (x), PAGE_SIZE, 0)
+#define alloc_bootmem_node(ignore, x) \
+	__alloc_bootmem_node(NODE_DATA(0), (x), SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
+#define alloc_bootmem_pages_node(ignore, x) \
+	__alloc_bootmem_node(NODE_DATA(0), (x), PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
+#define alloc_bootmem_low_pages_node(ignore, x) \
+	__alloc_bootmem_node(NODE_DATA(0), (x), PAGE_SIZE, 0)
 
-#endif /* CONFIG_DISCONTIGMEM */
+#endif /* CONFIG_NEED_MULTIPLE_NODES */
 
 extern int early_pfn_to_nid(unsigned long pfn);
 
diff --git a/include/asm-i386/page.h b/include/asm-i386/page.h
index 8f3dded01bf..dea8f8e6d86 100644
--- a/include/asm-i386/page.h
+++ b/include/asm-i386/page.h
@@ -137,11 +137,11 @@ extern int page_is_ram(unsigned long pagenr);
 #define __pa(x)			((unsigned long)(x)-PAGE_OFFSET)
 #define __va(x)			((void *)((unsigned long)(x)+PAGE_OFFSET))
 #define pfn_to_kaddr(pfn)      __va((pfn) << PAGE_SHIFT)
-#ifndef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_FLATMEM
 #define pfn_to_page(pfn)	(mem_map + (pfn))
 #define page_to_pfn(page)	((unsigned long)((page) - mem_map))
 #define pfn_valid(pfn)		((pfn) < max_mapnr)
-#endif /* !CONFIG_DISCONTIGMEM */
+#endif /* CONFIG_FLATMEM */
 #define virt_to_page(kaddr)	pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
 
 #define virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h
index e9efe148fdf..77c6497f416 100644
--- a/include/asm-i386/pgtable.h
+++ b/include/asm-i386/pgtable.h
@@ -398,9 +398,9 @@ extern void noexec_setup(const char *str);
 
 #endif /* !__ASSEMBLY__ */
 
-#ifndef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_FLATMEM
 #define kern_addr_valid(addr)	(1)
-#endif /* !CONFIG_DISCONTIGMEM */
+#endif /* CONFIG_FLATMEM */
 
 #define io_remap_page_range(vma, vaddr, paddr, size, prot)		\
 		remap_pfn_range(vma, vaddr, (paddr) >> PAGE_SHIFT, size, prot)
diff --git a/include/asm-i386/sparsemem.h b/include/asm-i386/sparsemem.h
new file mode 100644
index 00000000000..cfeed990585
--- /dev/null
+++ b/include/asm-i386/sparsemem.h
@@ -0,0 +1,31 @@
+#ifndef _I386_SPARSEMEM_H
+#define _I386_SPARSEMEM_H
+#ifdef CONFIG_SPARSEMEM
+
+/*
+ * generic non-linear memory support:
+ *
+ * 1) we will not split memory into more chunks than will fit into the
+ *    flags field of the struct page
+ */
+
+/*
+ * SECTION_SIZE_BITS		2^N: how big each section will be
+ * MAX_PHYSADDR_BITS		2^N: how much physical address space we have
+ * MAX_PHYSMEM_BITS		2^N: how much memory we can have in that space
+ */
+#ifdef CONFIG_X86_PAE
+#define SECTION_SIZE_BITS       30
+#define MAX_PHYSADDR_BITS       36
+#define MAX_PHYSMEM_BITS	36
+#else
+#define SECTION_SIZE_BITS       26
+#define MAX_PHYSADDR_BITS       32
+#define MAX_PHYSMEM_BITS	32
+#endif
+
+/* XXX: FIXME -- wli */
+#define kern_addr_valid(kaddr)  (0)
+
+#endif /* CONFIG_SPARSEMEM */
+#endif /* _I386_SPARSEMEM_H */
-- 
cgit v1.2.3-70-g09d2


From 641c767389b19859a45e6de46d8e18cd935bdb60 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Thu, 23 Jun 2005 00:07:59 -0700
Subject: [PATCH] sparsemem swiss cheese numa layouts

The part of the sparsemem patch which modifies memmap_init_zone() has recently
become a problem.  It changes behavior so that there is a call to
pfn_to_page() for each individual page inside of a node's range:
node_start_pfn through node_end_pfn.  It used to simply do this once, at the
beginning of the node, but having sparsemem's non-contiguous mem_map[]s inside
of a node made it necessary to change.

Mike Kravetz recently wrote a patch which made the NUMA code accept some new
kinds of layouts.  The system's memory was laid out like this, with node 0's
memory in two pieces: one before and one after node 1's memory:

	Node 0: +++++     +++++
	Node 1:      +++++

Previous behavior before Mike's patch was to assign nodes like this:

	Node 0: 00000     XXXXX
	Node 1:      11111

Where the 'X' areas were simply thrown away.  The new behavior was to make the
pg_data_t span node 0 across all of its areas, including areas that are really
node 1's: Node 0: 000000000000000 Node 1: 11111

This wastes a little bit of mem_map space, but ends up being OK, and more
fully utilizes the system's memory.  memmap_init_zone() initializes all of the
"struct page"s for node 0, even for the "hole", but those never get used,
because there is no pfn_to_page() that resolves to those pages.  However, only
calling pfn_to_page() once, memmap_init_zone() always uses the pages that were
allocated for node0->node_mem_map because:

	struct page *start = pfn_to_page(start_pfn);
	// effectively start = &node->node_mem_map[0]
	for (page = start; page < (start + size); page++) {
		init_page_here();...
		page++;
	}

Slow, and wasteful, but generally harmless.

But, modify that to call pfn_to_page() for each loop iteration (like sparsemem
does):

	for (pfn = start_pfn; pfn < < (start_pfn + size); pfn++++) {
		page = pfn_to_page(pfn);
	}

And you end up trying to initialize node 1's pages too early, along with bogus
data from node 0.  This patch checks for those weird layouts and declines to
touch the pages, making the more frequent pfn_to_page() calls OK to do.

Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/Kconfig     | 12 ++++++++++++
 include/linux/mmzone.h |  6 ++++++
 mm/page_alloc.c        |  2 ++
 3 files changed, 20 insertions(+)

(limited to 'include')

diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig
index 011b5c0bf1d..85f8fcf44b6 100644
--- a/arch/ppc64/Kconfig
+++ b/arch/ppc64/Kconfig
@@ -211,6 +211,18 @@ config ARCH_FLATMEM_ENABLE
 
 source "mm/Kconfig"
 
+# Some NUMA nodes have memory ranges that span
+# other nodes.  Even though a pfn is valid and
+# between a node's start and end pfns, it may not
+# reside on that node.
+#
+# This is a relatively temporary hack that should
+# be able to go away when sparsemem is fully in
+# place
+config NODES_SPAN_OTHER_NODES
+	def_bool y
+	depends on NEED_MULTIPLE_NODES
+
 config NUMA
 	bool "NUMA support"
 	depends on DISCONTIGMEM
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 19860d317ec..746b57e3d37 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -528,6 +528,12 @@ void sparse_init(void);
 #define sparse_init()	do {} while (0)
 #endif /* CONFIG_SPARSEMEM */
 
+#ifdef CONFIG_NODES_SPAN_OTHER_NODES
+#define early_pfn_in_nid(pfn, nid)	(early_pfn_to_nid(pfn) == (nid))
+#else
+#define early_pfn_in_nid(pfn, nid)	(1)
+#endif
+
 #ifndef early_pfn_valid
 #define early_pfn_valid(pfn)	(1)
 #endif
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 5c1b8982a6d..1eb683f9b3a 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1656,6 +1656,8 @@ void __init memmap_init_zone(unsigned long size, int nid, unsigned long zone,
 	for (pfn = start_pfn; pfn < end_pfn; pfn++, page++) {
 		if (!early_pfn_valid(pfn))
 			continue;
+		if (!early_pfn_in_nid(pfn, nid))
+			continue;
 		page = pfn_to_page(pfn);
 		set_page_links(page, zone, nid, pfn);
 		set_page_count(page, 0);
-- 
cgit v1.2.3-70-g09d2


From 29751f6991e845f7d002a6ae520bf996b38c8dcd Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Thu, 23 Jun 2005 00:08:00 -0700
Subject: [PATCH] sparsemem hotplug base

Make sparse's initalization be accessible at runtime.  This allows sparse
mappings to be created after boot in a hotplug situation.

This patch is separated from the previous one just to give an indication how
much of the sparse infrastructure is *just* for hotplug memory.

The section_mem_map doesn't really store a pointer.  It stores something that
is convenient to do some math against to get a pointer.  It isn't valid to
just do *section_mem_map, so I don't think it should be stored as a pointer.

There are a couple of things I'd like to store about a section.  First of all,
the fact that it is !NULL does not mean that it is present.  There could be
such a combination where section_mem_map *is* NULL, but the math gets you
properly to a real mem_map.  So, I don't think that check is safe.

Since we're storing 32-bit-aligned structures, we have a few bits in the
bottom of the pointer to play with.  Use one bit to encode whether there's
really a mem_map there, and the other one to tell whether there's a valid
section there.  We need to distinguish between the two because sometimes
there's a gap between when a section is discovered to be present and when we
can get the mem_map for it.

Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Bob Picco <bob.picco@hp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mmzone.h | 56 +++++++++++++++++++++++++++---
 mm/page_alloc.c        |  4 +--
 mm/sparse.c            | 92 +++++++++++++++++++++++++++++++++++++++-----------
 3 files changed, 125 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 746b57e3d37..6c90461ed99 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -476,11 +476,56 @@ extern struct pglist_data contig_page_data;
 
 struct page;
 struct mem_section {
-	struct page *section_mem_map;
+	/*
+	 * This is, logically, a pointer to an array of struct
+	 * pages.  However, it is stored with some other magic.
+	 * (see sparse.c::sparse_init_one_section())
+	 *
+	 * Making it a UL at least makes someone do a cast
+	 * before using it wrong.
+	 */
+	unsigned long section_mem_map;
 };
 
 extern struct mem_section mem_section[NR_MEM_SECTIONS];
 
+static inline struct mem_section *__nr_to_section(unsigned long nr)
+{
+	return &mem_section[nr];
+}
+
+/*
+ * We use the lower bits of the mem_map pointer to store
+ * a little bit of information.  There should be at least
+ * 3 bits here due to 32-bit alignment.
+ */
+#define	SECTION_MARKED_PRESENT	(1UL<<0)
+#define SECTION_HAS_MEM_MAP	(1UL<<1)
+#define SECTION_MAP_LAST_BIT	(1UL<<2)
+#define SECTION_MAP_MASK	(~(SECTION_MAP_LAST_BIT-1))
+
+static inline struct page *__section_mem_map_addr(struct mem_section *section)
+{
+	unsigned long map = section->section_mem_map;
+	map &= SECTION_MAP_MASK;
+	return (struct page *)map;
+}
+
+static inline int valid_section(struct mem_section *section)
+{
+	return (section->section_mem_map & SECTION_MARKED_PRESENT);
+}
+
+static inline int section_has_mem_map(struct mem_section *section)
+{
+	return (section->section_mem_map & SECTION_HAS_MEM_MAP);
+}
+
+static inline int valid_section_nr(unsigned long nr)
+{
+	return valid_section(__nr_to_section(nr));
+}
+
 /*
  * Given a kernel address, find the home node of the underlying memory.
  */
@@ -488,24 +533,25 @@ extern struct mem_section mem_section[NR_MEM_SECTIONS];
 
 static inline struct mem_section *__pfn_to_section(unsigned long pfn)
 {
-	return &mem_section[pfn_to_section_nr(pfn)];
+	return __nr_to_section(pfn_to_section_nr(pfn));
 }
 
 #define pfn_to_page(pfn) 						\
 ({ 									\
 	unsigned long __pfn = (pfn);					\
-	__pfn_to_section(__pfn)->section_mem_map + __pfn;		\
+	__section_mem_map_addr(__pfn_to_section(__pfn)) + __pfn;	\
 })
 #define page_to_pfn(page)						\
 ({									\
-	page - mem_section[page_to_section(page)].section_mem_map;	\
+	page - __section_mem_map_addr(__nr_to_section(			\
+		page_to_section(page)));				\
 })
 
 static inline int pfn_valid(unsigned long pfn)
 {
 	if (pfn_to_section_nr(pfn) >= NR_MEM_SECTIONS)
 		return 0;
-	return mem_section[pfn_to_section_nr(pfn)].section_mem_map != 0;
+	return valid_section(__nr_to_section(pfn_to_section_nr(pfn)));
 }
 
 /*
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 1eb683f9b3a..7ee675ad101 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1650,8 +1650,8 @@ void __init memmap_init_zone(unsigned long size, int nid, unsigned long zone,
 		unsigned long start_pfn)
 {
 	struct page *page;
-	int end_pfn = start_pfn + size;
-	int pfn;
+	unsigned long end_pfn = start_pfn + size;
+	unsigned long pfn;
 
 	for (pfn = start_pfn; pfn < end_pfn; pfn++, page++) {
 		if (!early_pfn_valid(pfn))
diff --git a/mm/sparse.c b/mm/sparse.c
index f888385b9e1..b54e304df4a 100644
--- a/mm/sparse.c
+++ b/mm/sparse.c
@@ -25,7 +25,7 @@ void memory_present(int nid, unsigned long start, unsigned long end)
 	for (pfn = start; pfn < end; pfn += PAGES_PER_SECTION) {
 		unsigned long section = pfn_to_section_nr(pfn);
 		if (!mem_section[section].section_mem_map)
-			mem_section[section].section_mem_map = (void *) -1;
+			mem_section[section].section_mem_map = SECTION_MARKED_PRESENT;
 	}
 }
 
@@ -50,6 +50,56 @@ unsigned long __init node_memmap_size_bytes(int nid, unsigned long start_pfn,
 	return nr_pages * sizeof(struct page);
 }
 
+/*
+ * Subtle, we encode the real pfn into the mem_map such that
+ * the identity pfn - section_mem_map will return the actual
+ * physical page frame number.
+ */
+static unsigned long sparse_encode_mem_map(struct page *mem_map, unsigned long pnum)
+{
+	return (unsigned long)(mem_map - (section_nr_to_pfn(pnum)));
+}
+
+/*
+ * We need this if we ever free the mem_maps.  While not implemented yet,
+ * this function is included for parity with its sibling.
+ */
+static __attribute((unused))
+struct page *sparse_decode_mem_map(unsigned long coded_mem_map, unsigned long pnum)
+{
+	return ((struct page *)coded_mem_map) + section_nr_to_pfn(pnum);
+}
+
+static int sparse_init_one_section(struct mem_section *ms,
+		unsigned long pnum, struct page *mem_map)
+{
+	if (!valid_section(ms))
+		return -EINVAL;
+
+	ms->section_mem_map |= sparse_encode_mem_map(mem_map, pnum);
+
+	return 1;
+}
+
+static struct page *sparse_early_mem_map_alloc(unsigned long pnum)
+{
+	struct page *map;
+	int nid = early_pfn_to_nid(section_nr_to_pfn(pnum));
+
+	map = alloc_remap(nid, sizeof(struct page) * PAGES_PER_SECTION);
+	if (map)
+		return map;
+
+	map = alloc_bootmem_node(NODE_DATA(nid),
+			sizeof(struct page) * PAGES_PER_SECTION);
+	if (map)
+		return map;
+
+	printk(KERN_WARNING "%s: allocation failed\n", __FUNCTION__);
+	mem_section[pnum].section_mem_map = 0;
+	return NULL;
+}
+
 /*
  * Allocate the accumulated non-linear sections, allocate a mem_map
  * for each and record the physical to section mapping.
@@ -58,28 +108,30 @@ void sparse_init(void)
 {
 	unsigned long pnum;
 	struct page *map;
-	int nid;
 
 	for (pnum = 0; pnum < NR_MEM_SECTIONS; pnum++) {
-		if (!mem_section[pnum].section_mem_map)
+		if (!valid_section_nr(pnum))
 			continue;
 
-		nid = early_pfn_to_nid(section_nr_to_pfn(pnum));
-		map = alloc_remap(nid, sizeof(struct page) * PAGES_PER_SECTION);
-		if (!map)
-			map = alloc_bootmem_node(NODE_DATA(nid),
-				sizeof(struct page) * PAGES_PER_SECTION);
-		if (!map) {
-			mem_section[pnum].section_mem_map = 0;
-			continue;
-		}
-
-		/*
-		 * Subtle, we encode the real pfn into the mem_map such that
-		 * the identity pfn - section_mem_map will return the actual
-		 * physical page frame number.
-		 */
-		mem_section[pnum].section_mem_map = map -
-						section_nr_to_pfn(pnum);
+		map = sparse_early_mem_map_alloc(pnum);
+		if (map)
+			sparse_init_one_section(&mem_section[pnum], pnum, map);
 	}
 }
+
+/*
+ * returns the number of sections whose mem_maps were properly
+ * set.  If this is <=0, then that means that the passed-in
+ * map was not consumed and must be freed.
+ */
+int sparse_add_one_section(unsigned long start_pfn, int nr_pages, struct page *map)
+{
+	struct mem_section *ms = __pfn_to_section(start_pfn);
+
+	if (ms->section_mem_map & SECTION_MARKED_PRESENT)
+		return -EEXIST;
+
+	ms->section_mem_map |= SECTION_MARKED_PRESENT;
+
+	return sparse_init_one_section(ms, pfn_to_section_nr(start_pfn), map);
+}
-- 
cgit v1.2.3-70-g09d2


From 510f8fa7ba18320d408dd3093663e58f5664f2f0 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Thu, 23 Jun 2005 00:08:01 -0700
Subject: [PATCH] ppc64: add early_pfn_to_nid

Provide an implementation of early_pfn_to_nid for PPC64.  This is used by
memory models to determine the node from which to take allocations before the
memory allocators are fully initialised.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Martin Bligh <mbligh@aracnet.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/Kconfig         | 4 ++++
 include/asm-ppc64/mmzone.h | 5 +++++
 2 files changed, 9 insertions(+)

(limited to 'include')

diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig
index 85f8fcf44b6..3c8bfd0345c 100644
--- a/arch/ppc64/Kconfig
+++ b/arch/ppc64/Kconfig
@@ -211,6 +211,10 @@ config ARCH_FLATMEM_ENABLE
 
 source "mm/Kconfig"
 
+config HAVE_ARCH_EARLY_PFN_TO_NID
+	bool
+	default y
+
 # Some NUMA nodes have memory ranges that span
 # other nodes.  Even though a pfn is valid and
 # between a node's start and end pfns, it may not
diff --git a/include/asm-ppc64/mmzone.h b/include/asm-ppc64/mmzone.h
index cbfc5ecfe87..ecc4b07507d 100644
--- a/include/asm-ppc64/mmzone.h
+++ b/include/asm-ppc64/mmzone.h
@@ -90,4 +90,9 @@ static inline int pa_to_nid(unsigned long pa)
 #define discontigmem_pfn_valid(pfn)		((pfn) < num_physpages)
 
 #endif /* CONFIG_DISCONTIGMEM */
+
+#ifdef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID
+#define early_pfn_to_nid(pfn)  pa_to_nid(((unsigned long)pfn) << PAGE_SHIFT)
+#endif
+
 #endif /* _ASM_MMZONE_H_ */
-- 
cgit v1.2.3-70-g09d2


From 145e664231648121026d470094c200851a446a73 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Thu, 23 Jun 2005 00:08:03 -0700
Subject: [PATCH] ppc64: sparsemem memory model

Provide the architecture specific implementation for SPARSEMEM for PPC64
systems.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Mike Kravetz <kravetz@us.ibm.com> (in part)
Signed-off-by: Martin Bligh <mbligh@aracnet.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/Kconfig            | 13 ++++++++++++-
 arch/ppc64/kernel/setup.c     |  1 +
 arch/ppc64/mm/Makefile        |  2 +-
 arch/ppc64/mm/init.c          | 18 +++++++++++++-----
 include/asm-ppc64/mmzone.h    | 36 +++++++++++++++++++++++-------------
 include/asm-ppc64/page.h      |  3 ++-
 include/asm-ppc64/sparsemem.h | 16 ++++++++++++++++
 7 files changed, 68 insertions(+), 21 deletions(-)
 create mode 100644 include/asm-ppc64/sparsemem.h

(limited to 'include')

diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig
index c92d48fe06e..6448231cb10 100644
--- a/arch/ppc64/Kconfig
+++ b/arch/ppc64/Kconfig
@@ -198,6 +198,13 @@ config HMT
 	  This option enables hardware multithreading on RS64 cpus.
 	  pSeries systems p620 and p660 have such a cpu type.
 
+config ARCH_SELECT_MEMORY_MODEL
+	def_bool y
+
+config ARCH_FLATMEM_ENABLE
+       def_bool y
+       depends on !NUMA
+
 config ARCH_DISCONTIGMEM_ENABLE
 	def_bool y
 	depends on SMP && PPC_PSERIES
@@ -209,6 +216,10 @@ config ARCH_DISCONTIGMEM_DEFAULT
 config ARCH_FLATMEM_ENABLE
 	def_bool y
 
+config ARCH_SPARSEMEM_ENABLE
+	def_bool y
+	depends on ARCH_DISCONTIGMEM_ENABLE
+
 source "mm/Kconfig"
 
 config HAVE_ARCH_EARLY_PFN_TO_NID
@@ -229,7 +240,7 @@ config NODES_SPAN_OTHER_NODES
 
 config NUMA
 	bool "NUMA support"
-	depends on DISCONTIGMEM
+	default y if DISCONTIGMEM || SPARSEMEM
 
 config SCHED_SMT
 	bool "SMT (Hyperthreading) scheduler support"
diff --git a/arch/ppc64/kernel/setup.c b/arch/ppc64/kernel/setup.c
index 8e439a81764..9e70ac90dec 100644
--- a/arch/ppc64/kernel/setup.c
+++ b/arch/ppc64/kernel/setup.c
@@ -1055,6 +1055,7 @@ void __init setup_arch(char **cmdline_p)
 
 	/* set up the bootmem stuff with available memory */
 	do_init_bootmem();
+	sparse_init();
 
 	/* initialize the syscall map in systemcfg */
 	setup_syscall_map();
diff --git a/arch/ppc64/mm/Makefile b/arch/ppc64/mm/Makefile
index ac522d57b2a..3695d00d347 100644
--- a/arch/ppc64/mm/Makefile
+++ b/arch/ppc64/mm/Makefile
@@ -6,6 +6,6 @@ EXTRA_CFLAGS += -mno-minimal-toc
 
 obj-y := fault.o init.o imalloc.o hash_utils.o hash_low.o tlb.o \
 	slb_low.o slb.o stab.o mmap.o
-obj-$(CONFIG_DISCONTIGMEM) += numa.o
+obj-$(CONFIG_NEED_MULTIPLE_NODES) += numa.o
 obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o
 obj-$(CONFIG_PPC_MULTIPLATFORM) += hash_native.o
diff --git a/arch/ppc64/mm/init.c b/arch/ppc64/mm/init.c
index 29dbe084c21..b50b3a446db 100644
--- a/arch/ppc64/mm/init.c
+++ b/arch/ppc64/mm/init.c
@@ -531,7 +531,7 @@ EXPORT_SYMBOL(page_is_ram);
  * Initialize the bootmem system and give it all the memory we
  * have available.
  */
-#ifndef CONFIG_DISCONTIGMEM
+#ifndef CONFIG_NEED_MULTIPLE_NODES
 void __init do_init_bootmem(void)
 {
 	unsigned long i;
@@ -553,12 +553,20 @@ void __init do_init_bootmem(void)
 
 	max_pfn = max_low_pfn;
 
-	/* add all physical memory to the bootmem map. Also find the first */
+	/* Add all physical memory to the bootmem map, mark each area
+	 * present.
+	 */
 	for (i=0; i < lmb.memory.cnt; i++) {
 		unsigned long physbase, size;
+		unsigned long start_pfn, end_pfn;
 
 		physbase = lmb.memory.region[i].physbase;
 		size = lmb.memory.region[i].size;
+
+		start_pfn = physbase >> PAGE_SHIFT;
+		end_pfn = start_pfn + (size >> PAGE_SHIFT);
+		memory_present(0, start_pfn, end_pfn);
+
 		free_bootmem(physbase, size);
 	}
 
@@ -597,7 +605,7 @@ void __init paging_init(void)
 	free_area_init_node(0, NODE_DATA(0), zones_size,
 			    __pa(PAGE_OFFSET) >> PAGE_SHIFT, zholes_size);
 }
-#endif /* CONFIG_DISCONTIGMEM */
+#endif /* ! CONFIG_NEED_MULTIPLE_NODES */
 
 static struct kcore_list kcore_vmem;
 
@@ -628,7 +636,7 @@ module_init(setup_kcore);
 
 void __init mem_init(void)
 {
-#ifdef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_NEED_MULTIPLE_NODES
 	int nid;
 #endif
 	pg_data_t *pgdat;
@@ -639,7 +647,7 @@ void __init mem_init(void)
 	num_physpages = max_low_pfn;	/* RAM is assumed contiguous */
 	high_memory = (void *) __va(max_low_pfn * PAGE_SIZE);
 
-#ifdef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_NEED_MULTIPLE_NODES
         for_each_online_node(nid) {
 		if (NODE_DATA(nid)->node_spanned_pages != 0) {
 			printk("freeing bootmem node %x\n", nid);
diff --git a/include/asm-ppc64/mmzone.h b/include/asm-ppc64/mmzone.h
index ecc4b07507d..ed473f4b015 100644
--- a/include/asm-ppc64/mmzone.h
+++ b/include/asm-ppc64/mmzone.h
@@ -10,9 +10,20 @@
 #include <linux/config.h>
 #include <asm/smp.h>
 
-#ifdef CONFIG_DISCONTIGMEM
+/* generic non-linear memory support:
+ *
+ * 1) we will not split memory into more chunks than will fit into the
+ *    flags field of the struct page
+ */
+
+
+#ifdef CONFIG_NEED_MULTIPLE_NODES
 
 extern struct pglist_data *node_data[];
+/*
+ * Return a pointer to the node data for node n.
+ */
+#define NODE_DATA(nid)		(node_data[nid])
 
 /*
  * Following are specific to this numa platform.
@@ -47,30 +58,27 @@ static inline int pa_to_nid(unsigned long pa)
 	return nid;
 }
 
-#define pfn_to_nid(pfn)		pa_to_nid((pfn) << PAGE_SHIFT)
-
-/*
- * Return a pointer to the node data for node n.
- */
-#define NODE_DATA(nid)		(node_data[nid])
-
 #define node_localnr(pfn, nid)	((pfn) - NODE_DATA(nid)->node_start_pfn)
 
 /*
  * Following are macros that each numa implmentation must define.
  */
 
-/*
- * Given a kernel address, find the home node of the underlying memory.
- */
-#define kvaddr_to_nid(kaddr)	pa_to_nid(__pa(kaddr))
-
 #define node_start_pfn(nid)	(NODE_DATA(nid)->node_start_pfn)
 #define node_end_pfn(nid)	(NODE_DATA(nid)->node_end_pfn)
 
 #define local_mapnr(kvaddr) \
 	( (__pa(kvaddr) >> PAGE_SHIFT) - node_start_pfn(kvaddr_to_nid(kvaddr)) 
 
+#ifdef CONFIG_DISCONTIGMEM
+
+/*
+ * Given a kernel address, find the home node of the underlying memory.
+ */
+#define kvaddr_to_nid(kaddr)	pa_to_nid(__pa(kaddr))
+
+#define pfn_to_nid(pfn)		pa_to_nid((unsigned long)(pfn) << PAGE_SHIFT)
+
 /* Written this way to avoid evaluating arguments twice */
 #define discontigmem_pfn_to_page(pfn) \
 ({ \
@@ -91,6 +99,8 @@ static inline int pa_to_nid(unsigned long pa)
 
 #endif /* CONFIG_DISCONTIGMEM */
 
+#endif /* CONFIG_NEED_MULTIPLE_NODES */
+
 #ifdef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID
 #define early_pfn_to_nid(pfn)  pa_to_nid(((unsigned long)pfn) << PAGE_SHIFT)
 #endif
diff --git a/include/asm-ppc64/page.h b/include/asm-ppc64/page.h
index 257d87eb7c3..a5893a305a0 100644
--- a/include/asm-ppc64/page.h
+++ b/include/asm-ppc64/page.h
@@ -217,7 +217,8 @@ extern u64 ppc64_pft_size;		/* Log 2 of page table size */
 #define page_to_pfn(page)	discontigmem_page_to_pfn(page)
 #define pfn_to_page(pfn)	discontigmem_pfn_to_page(pfn)
 #define pfn_valid(pfn)		discontigmem_pfn_valid(pfn)
-#else
+#endif
+#ifdef CONFIG_FLATMEM
 #define pfn_to_page(pfn)	(mem_map + (pfn))
 #define page_to_pfn(page)	((unsigned long)((page) - mem_map))
 #define pfn_valid(pfn)		((pfn) < max_mapnr)
diff --git a/include/asm-ppc64/sparsemem.h b/include/asm-ppc64/sparsemem.h
new file mode 100644
index 00000000000..c5bd47e57f1
--- /dev/null
+++ b/include/asm-ppc64/sparsemem.h
@@ -0,0 +1,16 @@
+#ifndef _ASM_PPC64_SPARSEMEM_H
+#define _ASM_PPC64_SPARSEMEM_H 1
+
+#ifdef CONFIG_SPARSEMEM
+/*
+ * SECTION_SIZE_BITS		2^N: how big each section will be
+ * MAX_PHYSADDR_BITS		2^N: how much physical address space we have
+ * MAX_PHYSMEM_BITS		2^N: how much memory we can have in that space
+ */
+#define SECTION_SIZE_BITS       24
+#define MAX_PHYSADDR_BITS       38
+#define MAX_PHYSMEM_BITS        36
+
+#endif /* CONFIG_SPARSEMEM */
+
+#endif /* _ASM_PPC64_SPARSEMEM_H */
-- 
cgit v1.2.3-70-g09d2


From 2b97690f4cd960779fb351b7cd9974390afabb36 Mon Sep 17 00:00:00 2001
From: Matt Tolentino <metolent@snoqualmie.dp.intel.com>
Date: Thu, 23 Jun 2005 00:08:06 -0700
Subject: [PATCH] reorganize x86-64 NUMA and DISCONTIGMEM config options

In order to use the alternative sparsemem implmentation for NUMA kernels,
we need to reorganize the config options.  This patch effectively abstracts
out the CONFIG_DISCONTIGMEM options to CONFIG_NUMA in most cases.  Thus,
the discontigmem implementation may be employed as always, but the
sparsemem implementation may be used alternatively.

Signed-off-by: Matt Tolentino <matthew.e.tolentino@intel.com>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/head64.c   |  2 +-
 arch/x86_64/kernel/setup.c    |  6 +++---
 arch/x86_64/mm/Makefile       |  2 +-
 arch/x86_64/mm/init.c         |  9 ++++++---
 arch/x86_64/mm/ioremap.c      |  2 +-
 include/asm-x86_64/io.h       |  5 -----
 include/asm-x86_64/mmzone.h   | 15 +++++++++------
 include/asm-x86_64/page.h     |  4 +++-
 include/asm-x86_64/topology.h |  4 +---
 9 files changed, 25 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/head64.c b/arch/x86_64/kernel/head64.c
index 0f8c78dcd38..cf6ab147a2a 100644
--- a/arch/x86_64/kernel/head64.c
+++ b/arch/x86_64/kernel/head64.c
@@ -94,7 +94,7 @@ void __init x86_64_start_kernel(char * real_mode_data)
 	s = strstr(saved_command_line, "earlyprintk=");
 	if (s != NULL)
 		setup_early_printk(s);
-#ifdef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_NUMA
 	s = strstr(saved_command_line, "numa=");
 	if (s != NULL)
 		numa_setup(s+5);
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 125a0bb1f5d..dd8419b37a7 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -348,7 +348,7 @@ static __init void parse_cmdline_early (char ** cmdline_p)
 		if (!memcmp(from, "mem=", 4))
 			parse_memopt(from+4, &from); 
 
-#ifdef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_NUMA
 		if (!memcmp(from, "numa=", 5))
 			numa_setup(from+5); 
 #endif
@@ -377,7 +377,7 @@ static __init void parse_cmdline_early (char ** cmdline_p)
 	*cmdline_p = command_line;
 }
 
-#ifndef CONFIG_DISCONTIGMEM
+#ifndef CONFIG_NUMA
 static void __init contig_initmem_init(void)
 {
         unsigned long bootmap_size, bootmap; 
@@ -554,7 +554,7 @@ void __init setup_arch(char **cmdline_p)
 	acpi_numa_init();
 #endif
 
-#ifdef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_NUMA
 	numa_initmem_init(0, end_pfn); 
 #else
 	contig_initmem_init(); 
diff --git a/arch/x86_64/mm/Makefile b/arch/x86_64/mm/Makefile
index 66c354ad80c..1d232a87f11 100644
--- a/arch/x86_64/mm/Makefile
+++ b/arch/x86_64/mm/Makefile
@@ -4,7 +4,7 @@
 
 obj-y	 := init.o fault.o ioremap.o extable.o pageattr.o
 obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o
-obj-$(CONFIG_DISCONTIGMEM) += numa.o
+obj-$(CONFIG_NUMA) += numa.o
 obj-$(CONFIG_K8_NUMA) += k8topology.o
 obj-$(CONFIG_ACPI_NUMA) += srat.o
 
diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index dbe53b4c7e6..72e4b364ed7 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -318,7 +318,7 @@ void zap_low_mappings(void)
 	flush_tlb_all();
 }
 
-#ifndef CONFIG_DISCONTIGMEM
+#ifndef CONFIG_NUMA
 void __init paging_init(void)
 {
 	{
@@ -427,13 +427,16 @@ void __init mem_init(void)
 	reservedpages = 0;
 
 	/* this will put all low memory onto the freelists */
-#ifdef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_NUMA
 	totalram_pages += numa_free_all_bootmem();
 	tmp = 0;
 	/* should count reserved pages here for all nodes */ 
 #else
+
+#ifdef CONFIG_FLATMEM
 	max_mapnr = end_pfn;
 	if (!mem_map) BUG();
+#endif
 
 	totalram_pages += free_all_bootmem();
 
@@ -515,7 +518,7 @@ void free_initrd_mem(unsigned long start, unsigned long end)
 void __init reserve_bootmem_generic(unsigned long phys, unsigned len) 
 { 
 	/* Should check here against the e820 map to avoid double free */ 
-#ifdef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_NUMA
 	int nid = phys_to_nid(phys);
   	reserve_bootmem_node(NODE_DATA(nid), phys, len);
 #else       		
diff --git a/arch/x86_64/mm/ioremap.c b/arch/x86_64/mm/ioremap.c
index 58aac23760e..6972df480d2 100644
--- a/arch/x86_64/mm/ioremap.c
+++ b/arch/x86_64/mm/ioremap.c
@@ -178,7 +178,7 @@ void __iomem * __ioremap(unsigned long phys_addr, unsigned long size, unsigned l
 	if (phys_addr >= ISA_START_ADDRESS && last_addr < ISA_END_ADDRESS)
 		return (__force void __iomem *)phys_to_virt(phys_addr);
 
-#ifndef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_FLATMEM
 	/*
 	 * Don't allow anybody to remap normal RAM that we're using..
 	 */
diff --git a/include/asm-x86_64/io.h b/include/asm-x86_64/io.h
index 94202703fae..37fc3f149a5 100644
--- a/include/asm-x86_64/io.h
+++ b/include/asm-x86_64/io.h
@@ -124,12 +124,7 @@ extern inline void * phys_to_virt(unsigned long address)
 /*
  * Change "struct page" to physical address.
  */
-#ifdef CONFIG_DISCONTIGMEM
-#include <asm/mmzone.h>
 #define page_to_phys(page)    ((dma_addr_t)page_to_pfn(page) << PAGE_SHIFT)
-#else
-#define page_to_phys(page)	((page - mem_map) << PAGE_SHIFT)
-#endif
 
 #include <asm-generic/iomap.h>
 
diff --git a/include/asm-x86_64/mmzone.h b/include/asm-x86_64/mmzone.h
index ca4fc3fe0de..768413751b3 100644
--- a/include/asm-x86_64/mmzone.h
+++ b/include/asm-x86_64/mmzone.h
@@ -6,7 +6,7 @@
 
 #include <linux/config.h>
 
-#ifdef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_NUMA
 
 #define VIRTUAL_BUG_ON(x) 
 
@@ -30,17 +30,16 @@ static inline __attribute__((pure)) int phys_to_nid(unsigned long addr)
 	return nid; 
 } 
 
-#define pfn_to_nid(pfn) phys_to_nid((unsigned long)(pfn) << PAGE_SHIFT)
-
-#define kvaddr_to_nid(kaddr)	phys_to_nid(__pa(kaddr))
 #define NODE_DATA(nid)		(node_data[nid])
 
 #define node_start_pfn(nid)	(NODE_DATA(nid)->node_start_pfn)
 #define node_end_pfn(nid)       (NODE_DATA(nid)->node_start_pfn + \
 				 NODE_DATA(nid)->node_spanned_pages)
 
-#define local_mapnr(kvaddr) \
-	( (__pa(kvaddr) >> PAGE_SHIFT) - node_start_pfn(kvaddr_to_nid(kvaddr)) )
+#ifdef CONFIG_DISCONTIGMEM
+
+#define pfn_to_nid(pfn) phys_to_nid((unsigned long)(pfn) << PAGE_SHIFT)
+#define kvaddr_to_nid(kaddr)	phys_to_nid(__pa(kaddr))
 
 /* AK: this currently doesn't deal with invalid addresses. We'll see 
    if the 2.5 kernel doesn't pass them
@@ -57,4 +56,8 @@ static inline __attribute__((pure)) int phys_to_nid(unsigned long addr)
 			({ u8 nid__ = pfn_to_nid(pfn); \
 			   nid__ != 0xff && (pfn) >= node_start_pfn(nid__) && (pfn) <= node_end_pfn(nid__); }))
 #endif
+
+#define local_mapnr(kvaddr) \
+	( (__pa(kvaddr) >> PAGE_SHIFT) - node_start_pfn(kvaddr_to_nid(kvaddr)) )
+#endif
 #endif
diff --git a/include/asm-x86_64/page.h b/include/asm-x86_64/page.h
index 9ce338c3a71..60130f4ca98 100644
--- a/include/asm-x86_64/page.h
+++ b/include/asm-x86_64/page.h
@@ -119,7 +119,9 @@ extern __inline__ int get_order(unsigned long size)
 	  __pa(v); })
 
 #define __va(x)			((void *)((unsigned long)(x)+PAGE_OFFSET))
-#ifndef CONFIG_DISCONTIGMEM
+#define __boot_va(x)		__va(x)
+#define __boot_pa(x)		__pa(x)
+#ifdef CONFIG_FLATMEM
 #define pfn_to_page(pfn)	(mem_map + (pfn))
 #define page_to_pfn(page)	((unsigned long)((page) - mem_map))
 #define pfn_valid(pfn)		((pfn) < max_mapnr)
diff --git a/include/asm-x86_64/topology.h b/include/asm-x86_64/topology.h
index 67f24e0ea81..da21573ec73 100644
--- a/include/asm-x86_64/topology.h
+++ b/include/asm-x86_64/topology.h
@@ -3,7 +3,7 @@
 
 #include <linux/config.h>
 
-#ifdef CONFIG_DISCONTIGMEM
+#ifdef CONFIG_NUMA
 
 #include <asm/mpspec.h>
 #include <asm/bitops.h>
@@ -37,7 +37,6 @@ static inline cpumask_t __pcibus_to_cpumask(int bus)
 }
 #define pcibus_to_cpumask(bus) __pcibus_to_cpumask(bus->number)
 
-#ifdef CONFIG_NUMA
 /* sched_domains SD_NODE_INIT for x86_64 machines */
 #define SD_NODE_INIT (struct sched_domain) {		\
 	.span			= CPU_MASK_NONE,	\
@@ -59,7 +58,6 @@ static inline cpumask_t __pcibus_to_cpumask(int bus)
 	.balance_interval	= 1,			\
 	.nr_balance_failed	= 0,			\
 }
-#endif
 
 #endif
 
-- 
cgit v1.2.3-70-g09d2


From bbfceef47fb9467424113a004070bf37a806a97c Mon Sep 17 00:00:00 2001
From: Matt Tolentino <metolent@snoqualmie.dp.intel.com>
Date: Thu, 23 Jun 2005 00:08:07 -0700
Subject: [PATCH] add x86-64 specific support for sparsemem

This patch adds in the necessary support for sparsemem such that x86-64
kernels may use sparsemem as an alternative to discontigmem for NUMA
kernels.  Note that this does no preclude one from continuing to build NUMA
kernels using discontigmem, but merely allows the option to build NUMA
kernels with sparsemem.

Interestingly, the use of sparsemem in lieu of discontigmem in NUMA kernels
results in reduced text size for otherwise equivalent kernels as shown in
the example builds below:

   text	   data	    bss	    dec	    hex	filename
2371036	 765884	1237108	4374028	 42be0c	vmlinux.discontig
2366549	 776484	1302772	4445805	 43d66d	vmlinux.sparse

Signed-off-by: Matt Tolentino <matthew.e.tolentino@intel.com>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/setup.c     | 27 +++++++++++++++++----------
 arch/x86_64/mm/numa.c          |  8 ++++++++
 include/asm-x86_64/bitops.h    |  2 --
 include/asm-x86_64/sparsemem.h | 26 ++++++++++++++++++++++++++
 4 files changed, 51 insertions(+), 12 deletions(-)
 create mode 100644 include/asm-x86_64/sparsemem.h

(limited to 'include')

diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index dd8419b37a7..000015dd5a8 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -40,6 +40,8 @@
 #include <linux/acpi.h>
 #include <linux/kallsyms.h>
 #include <linux/edd.h>
+#include <linux/mmzone.h>
+
 #include <asm/mtrr.h>
 #include <asm/uaccess.h>
 #include <asm/system.h>
@@ -378,16 +380,19 @@ static __init void parse_cmdline_early (char ** cmdline_p)
 }
 
 #ifndef CONFIG_NUMA
-static void __init contig_initmem_init(void)
+static void __init
+contig_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
 {
-        unsigned long bootmap_size, bootmap; 
-        bootmap_size = bootmem_bootmap_pages(end_pfn)<<PAGE_SHIFT;
-        bootmap = find_e820_area(0, end_pfn<<PAGE_SHIFT, bootmap_size);
-        if (bootmap == -1L) 
-                panic("Cannot find bootmem map of size %ld\n",bootmap_size);
-        bootmap_size = init_bootmem(bootmap >> PAGE_SHIFT, end_pfn);
-        e820_bootmem_free(NODE_DATA(0), 0, end_pfn << PAGE_SHIFT);
-        reserve_bootmem(bootmap, bootmap_size);
+	unsigned long bootmap_size, bootmap;
+
+	memory_present(0, start_pfn, end_pfn);
+	bootmap_size = bootmem_bootmap_pages(end_pfn)<<PAGE_SHIFT;
+	bootmap = find_e820_area(0, end_pfn<<PAGE_SHIFT, bootmap_size);
+	if (bootmap == -1L)
+		panic("Cannot find bootmem map of size %ld\n",bootmap_size);
+	bootmap_size = init_bootmem(bootmap >> PAGE_SHIFT, end_pfn);
+	e820_bootmem_free(NODE_DATA(0), 0, end_pfn << PAGE_SHIFT);
+	reserve_bootmem(bootmap, bootmap_size);
 } 
 #endif
 
@@ -557,7 +562,7 @@ void __init setup_arch(char **cmdline_p)
 #ifdef CONFIG_NUMA
 	numa_initmem_init(0, end_pfn); 
 #else
-	contig_initmem_init(); 
+	contig_initmem_init(0, end_pfn);
 #endif
 
 	/* Reserve direct mapping */
@@ -618,6 +623,8 @@ void __init setup_arch(char **cmdline_p)
 		}
 	}
 #endif
+
+	sparse_init();
 	paging_init();
 
 	check_ioapic();
diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c
index fd9f25d7a6c..84cde796ecb 100644
--- a/arch/x86_64/mm/numa.c
+++ b/arch/x86_64/mm/numa.c
@@ -66,6 +66,13 @@ int __init compute_hash_shift(struct node *nodes, int numnodes)
 	return -1; 
 }
 
+#ifdef CONFIG_SPARSEMEM
+int early_pfn_to_nid(unsigned long pfn)
+{
+	return phys_to_nid(pfn << PAGE_SHIFT);
+}
+#endif
+
 /* Initialize bootmem allocator for a node */
 void __init setup_node_bootmem(int nodeid, unsigned long start, unsigned long end)
 { 
@@ -80,6 +87,7 @@ void __init setup_node_bootmem(int nodeid, unsigned long start, unsigned long en
 	start_pfn = start >> PAGE_SHIFT;
 	end_pfn = end >> PAGE_SHIFT;
 
+	memory_present(nodeid, start_pfn, end_pfn);
 	nodedata_phys = find_e820_area(start, end, pgdat_size); 
 	if (nodedata_phys == -1L) 
 		panic("Cannot find memory pgdat in node %d\n", nodeid);
diff --git a/include/asm-x86_64/bitops.h b/include/asm-x86_64/bitops.h
index 5dd7727c756..a31bb99be53 100644
--- a/include/asm-x86_64/bitops.h
+++ b/include/asm-x86_64/bitops.h
@@ -411,8 +411,6 @@ static __inline__ int ffs(int x)
 /* find last set bit */
 #define fls(x) generic_fls(x)
 
-#define ARCH_HAS_ATOMIC_UNSIGNED 1
-
 #endif /* __KERNEL__ */
 
 #endif /* _X86_64_BITOPS_H */
diff --git a/include/asm-x86_64/sparsemem.h b/include/asm-x86_64/sparsemem.h
new file mode 100644
index 00000000000..dabb16714a7
--- /dev/null
+++ b/include/asm-x86_64/sparsemem.h
@@ -0,0 +1,26 @@
+#ifndef _ASM_X86_64_SPARSEMEM_H
+#define _ASM_X86_64_SPARSEMEM_H 1
+
+#ifdef CONFIG_SPARSEMEM
+
+/*
+ * generic non-linear memory support:
+ *
+ * 1) we will not split memory into more chunks than will fit into the flags
+ *    field of the struct page
+ *
+ * SECTION_SIZE_BITS		2^n: size of each section
+ * MAX_PHYSADDR_BITS		2^n: max size of physical address space
+ * MAX_PHYSMEM_BITS		2^n: how much memory we can have in that space
+ *
+ */
+
+#define SECTION_SIZE_BITS	27 /* matt - 128 is convenient right now */
+#define MAX_PHYSADDR_BITS	40
+#define MAX_PHYSMEM_BITS	40
+
+extern int early_pfn_to_nid(unsigned long pfn);
+
+#endif /* CONFIG_SPARSEMEM */
+
+#endif /* _ASM_X86_64_SPARSEMEM_H */
-- 
cgit v1.2.3-70-g09d2


From 8a9e1b0f564615bd92ba50162623e25c2904e564 Mon Sep 17 00:00:00 2001
From: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Date: Thu, 23 Jun 2005 00:08:13 -0700
Subject: [PATCH] Platform SMIs and their interferance with tsc based delay
 calibration

Issue:
Current tsc based delay_calibration can result in significant errors in
loops_per_jiffy count when the platform events like SMIs
(System Management Interrupts that are non-maskable) are present. This could
lead to potential kernel panic(). This issue is becoming more visible with 2.6
kernel (as default HZ is 1000) and on platforms with higher SMI handling
latencies. During the boot time, SMIs are mostly used by BIOS (for things
like legacy keyboard emulation).

Description:
The psuedocode for current delay calibration with tsc based delay looks like
(0) Estimate a value for loops_per_jiffy
(1) While (loops_per_jiffy estimate is accurate enough)
(2)   wait for jiffy transition (jiffy1)
(3)   Note down current tsc (tsc1)
(4)   loop until tsc becomes tsc1 + loops_per_jiffy
(5)   check whether jiffy changed since jiffy1 or not and refine
loops_per_jiffy estimate

Consider the following cases
Case 1:
If SMIs happen between (2) and (3) above, we can end up with a
loops_per_jiffy value that is too low. This results in shorted delays and
kernel can panic () during boot (Mostly at IOAPIC timer initialization
timer_irq_works() as we don't have enough timer interrupts in a specified
interval).

Case 2:
If SMIs happen between (3) and (4) above, then we can end up with a
loops_per_jiffy value that is too high. And with current i386 code, too
high lpj value (greater than 17M) can result in a overflow in
delay.c:__const_udelay() again resulting in shorter delay and panic().

Solution:
The patch below makes the calibration routine aware of asynchronous events
like SMIs. We increase the delay calibration time and also identify any
significant errors (greater than 12.5%) in the calibration and notify it to
user.

Patch below changes both i386 and x86-64 architectures to use this
new and improved calibrate_delay_direct() routine.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/timers/common.c     |  9 ++++
 arch/i386/kernel/timers/timer.c      |  9 ++++
 arch/i386/kernel/timers/timer_hpet.c |  1 +
 arch/i386/kernel/timers/timer_pm.c   |  1 +
 arch/i386/kernel/timers/timer_tsc.c  |  1 +
 arch/x86_64/lib/delay.c              |  7 +++
 include/asm-i386/timer.h             |  2 +
 include/asm-i386/timex.h             |  3 ++
 include/asm-x86_64/timex.h           |  3 ++
 init/calibrate.c                     | 94 ++++++++++++++++++++++++++++++++++++
 10 files changed, 130 insertions(+)

(limited to 'include')

diff --git a/arch/i386/kernel/timers/common.c b/arch/i386/kernel/timers/common.c
index 8e201219f52..b38cc0d0c71 100644
--- a/arch/i386/kernel/timers/common.c
+++ b/arch/i386/kernel/timers/common.c
@@ -139,6 +139,15 @@ bad_calibration:
 }
 #endif
 
+
+unsigned long read_timer_tsc(void)
+{
+	unsigned long retval;
+	rdtscl(retval);
+	return retval;
+}
+
+
 /* calculate cpu_khz */
 void init_cpu_khz(void)
 {
diff --git a/arch/i386/kernel/timers/timer.c b/arch/i386/kernel/timers/timer.c
index a3d6a288088..7e39ed8e33f 100644
--- a/arch/i386/kernel/timers/timer.c
+++ b/arch/i386/kernel/timers/timer.c
@@ -64,3 +64,12 @@ struct timer_opts* __init select_timer(void)
 	panic("select_timer: Cannot find a suitable timer\n");
 	return NULL;
 }
+
+int read_current_timer(unsigned long *timer_val)
+{
+	if (cur_timer->read_timer) {
+		*timer_val = cur_timer->read_timer();
+		return 0;
+	}
+	return -1;
+}
diff --git a/arch/i386/kernel/timers/timer_hpet.c b/arch/i386/kernel/timers/timer_hpet.c
index f778f471a09..15a7d727bd6 100644
--- a/arch/i386/kernel/timers/timer_hpet.c
+++ b/arch/i386/kernel/timers/timer_hpet.c
@@ -186,6 +186,7 @@ static struct timer_opts timer_hpet = {
 	.get_offset =		get_offset_hpet,
 	.monotonic_clock =	monotonic_clock_hpet,
 	.delay = 		delay_hpet,
+	.read_timer = 		read_timer_tsc,
 };
 
 struct init_timer_opts __initdata timer_hpet_init = {
diff --git a/arch/i386/kernel/timers/timer_pm.c b/arch/i386/kernel/timers/timer_pm.c
index d77f22030fe..4ef20e66349 100644
--- a/arch/i386/kernel/timers/timer_pm.c
+++ b/arch/i386/kernel/timers/timer_pm.c
@@ -246,6 +246,7 @@ static struct timer_opts timer_pmtmr = {
 	.get_offset		= get_offset_pmtmr,
 	.monotonic_clock 	= monotonic_clock_pmtmr,
 	.delay 			= delay_pmtmr,
+	.read_timer 		= read_timer_tsc,
 };
 
 struct init_timer_opts __initdata timer_pmtmr_init = {
diff --git a/arch/i386/kernel/timers/timer_tsc.c b/arch/i386/kernel/timers/timer_tsc.c
index 180444d8782..27f08ae9120 100644
--- a/arch/i386/kernel/timers/timer_tsc.c
+++ b/arch/i386/kernel/timers/timer_tsc.c
@@ -572,6 +572,7 @@ static struct timer_opts timer_tsc = {
 	.get_offset = get_offset_tsc,
 	.monotonic_clock = monotonic_clock_tsc,
 	.delay = delay_tsc,
+	.read_timer = read_timer_tsc,
 };
 
 struct init_timer_opts __initdata timer_tsc_init = {
diff --git a/arch/x86_64/lib/delay.c b/arch/x86_64/lib/delay.c
index aed61a668a1..33a873a3c22 100644
--- a/arch/x86_64/lib/delay.c
+++ b/arch/x86_64/lib/delay.c
@@ -12,6 +12,7 @@
 #include <linux/sched.h>
 #include <linux/delay.h>
 #include <asm/delay.h>
+#include <asm/msr.h>
 
 #ifdef CONFIG_SMP
 #include <asm/smp.h>
@@ -19,6 +20,12 @@
 
 int x86_udelay_tsc = 0;		/* Delay via TSC */
 
+int read_current_timer(unsigned long *timer_value)
+{
+	rdtscll(*timer_value);
+	return 0;
+}
+
 void __delay(unsigned long loops)
 {
 	unsigned bclock, now;
diff --git a/include/asm-i386/timer.h b/include/asm-i386/timer.h
index c3470984983..dcf1e07db08 100644
--- a/include/asm-i386/timer.h
+++ b/include/asm-i386/timer.h
@@ -22,6 +22,7 @@ struct timer_opts {
 	unsigned long (*get_offset)(void);
 	unsigned long long (*monotonic_clock)(void);
 	void (*delay)(unsigned long);
+	unsigned long (*read_timer)(void);
 };
 
 struct init_timer_opts {
@@ -52,6 +53,7 @@ extern struct init_timer_opts timer_cyclone_init;
 #endif
 
 extern unsigned long calibrate_tsc(void);
+extern unsigned long read_timer_tsc(void);
 extern void init_cpu_khz(void);
 extern int recalibrate_cpu_khz(void);
 #ifdef CONFIG_HPET_TIMER
diff --git a/include/asm-i386/timex.h b/include/asm-i386/timex.h
index b41e484c344..e370f907bd3 100644
--- a/include/asm-i386/timex.h
+++ b/include/asm-i386/timex.h
@@ -49,4 +49,7 @@ static inline cycles_t get_cycles (void)
 
 extern unsigned long cpu_khz;
 
+extern int read_current_timer(unsigned long *timer_value);
+#define ARCH_HAS_READ_CURRENT_TIMER	1
+
 #endif
diff --git a/include/asm-x86_64/timex.h b/include/asm-x86_64/timex.h
index 34f31a18f90..24ecf6a637c 100644
--- a/include/asm-x86_64/timex.h
+++ b/include/asm-x86_64/timex.h
@@ -26,6 +26,9 @@ static inline cycles_t get_cycles (void)
 
 extern unsigned int cpu_khz;
 
+extern int read_current_timer(unsigned long *timer_value);
+#define ARCH_HAS_READ_CURRENT_TIMER	1
+
 extern struct vxtime_data vxtime;
 
 #endif
diff --git a/init/calibrate.c b/init/calibrate.c
index c698e04a3db..d206c7548fe 100644
--- a/init/calibrate.c
+++ b/init/calibrate.c
@@ -8,6 +8,8 @@
 #include <linux/delay.h>
 #include <linux/init.h>
 
+#include <asm/timex.h>
+
 static unsigned long preset_lpj;
 static int __init lpj_setup(char *str)
 {
@@ -17,6 +19,92 @@ static int __init lpj_setup(char *str)
 
 __setup("lpj=", lpj_setup);
 
+#ifdef ARCH_HAS_READ_CURRENT_TIMER
+
+/* This routine uses the read_current_timer() routine and gets the
+ * loops per jiffy directly, instead of guessing it using delay().
+ * Also, this code tries to handle non-maskable asynchronous events
+ * (like SMIs)
+ */
+#define DELAY_CALIBRATION_TICKS			((HZ < 100) ? 1 : (HZ/100))
+#define MAX_DIRECT_CALIBRATION_RETRIES		5
+
+static unsigned long __devinit calibrate_delay_direct(void)
+{
+	unsigned long pre_start, start, post_start;
+	unsigned long pre_end, end, post_end;
+	unsigned long start_jiffies;
+	unsigned long tsc_rate_min, tsc_rate_max;
+	unsigned long good_tsc_sum = 0;
+	unsigned long good_tsc_count = 0;
+	int i;
+
+	if (read_current_timer(&pre_start) < 0 )
+		return 0;
+
+	/*
+	 * A simple loop like
+	 *	while ( jiffies < start_jiffies+1)
+	 *		start = read_current_timer();
+	 * will not do. As we don't really know whether jiffy switch
+	 * happened first or timer_value was read first. And some asynchronous
+	 * event can happen between these two events introducing errors in lpj.
+	 *
+	 * So, we do
+	 * 1. pre_start <- When we are sure that jiffy switch hasn't happened
+	 * 2. check jiffy switch
+	 * 3. start <- timer value before or after jiffy switch
+	 * 4. post_start <- When we are sure that jiffy switch has happened
+	 *
+	 * Note, we don't know anything about order of 2 and 3.
+	 * Now, by looking at post_start and pre_start difference, we can
+	 * check whether any asynchronous event happened or not
+	 */
+
+	for (i = 0; i < MAX_DIRECT_CALIBRATION_RETRIES; i++) {
+		pre_start = 0;
+		read_current_timer(&start);
+		start_jiffies = jiffies;
+		while (jiffies <= (start_jiffies + 1)) {
+			pre_start = start;
+			read_current_timer(&start);
+		}
+		read_current_timer(&post_start);
+
+		pre_end = 0;
+		end = post_start;
+		while (jiffies <=
+		       (start_jiffies + 1 + DELAY_CALIBRATION_TICKS)) {
+			pre_end = end;
+			read_current_timer(&end);
+		}
+		read_current_timer(&post_end);
+
+		tsc_rate_max = (post_end - pre_start) / DELAY_CALIBRATION_TICKS;
+		tsc_rate_min = (pre_end - post_start) / DELAY_CALIBRATION_TICKS;
+
+		/*
+	 	 * If the upper limit and lower limit of the tsc_rate is
+		 * >= 12.5% apart, redo calibration.
+		 */
+		if (pre_start != 0 && pre_end != 0 &&
+		    (tsc_rate_max - tsc_rate_min) < (tsc_rate_max >> 3)) {
+			good_tsc_count++;
+			good_tsc_sum += tsc_rate_max;
+		}
+	}
+
+	if (good_tsc_count)
+		return (good_tsc_sum/good_tsc_count);
+
+	printk(KERN_WARNING "calibrate_delay_direct() failed to get a good "
+	       "estimate for loops_per_jiffy.\nProbably due to long platform interrupts. Consider using \"lpj=\" boot option.\n");
+	return 0;
+}
+#else
+static unsigned long __devinit calibrate_delay_direct(void) {return 0;}
+#endif
+
 /*
  * This is the number of bits of precision for the loops_per_jiffy.  Each
  * bit takes on average 1.5/HZ seconds.  This (like the original) is a little
@@ -35,6 +123,12 @@ void __devinit calibrate_delay(void)
 			"%lu.%02lu BogoMIPS preset\n",
 			loops_per_jiffy/(500000/HZ),
 			(loops_per_jiffy/(5000/HZ)) % 100);
+	} else if ((loops_per_jiffy = calibrate_delay_direct()) != 0) {
+		printk("Calibrating delay using timer specific routine.. ");
+		printk("%lu.%02lu BogoMIPS (lpj=%lu)\n",
+			loops_per_jiffy/(500000/HZ),
+			(loops_per_jiffy/(5000/HZ)) % 100,
+			loops_per_jiffy);
 	} else {
 		loops_per_jiffy = (1<<12);
 
-- 
cgit v1.2.3-70-g09d2


From 4a35293667f8691b45fdb72770f087c086fb9702 Mon Sep 17 00:00:00 2001
From: Hirokazu Takata <takata@linux-m32r.org>
Date: Thu, 23 Jun 2005 00:08:14 -0700
Subject: [PATCH] m32r: build fix for asm-m32r/topology.h

Use asm-generic/topology.h to fix yet another pcibus_to_node() build error.

Cc: Christoph Lameter <clameter@engr.sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-m32r/topology.h | 44 +-------------------------------------------
 1 file changed, 1 insertion(+), 43 deletions(-)

(limited to 'include')

diff --git a/include/asm-m32r/topology.h b/include/asm-m32r/topology.h
index 299a89d91bd..d607eb32bd7 100644
--- a/include/asm-m32r/topology.h
+++ b/include/asm-m32r/topology.h
@@ -1,48 +1,6 @@
-/*
- * linux/include/asm-generic/topology.h
- *
- * Written by: Matthew Dobson, IBM Corporation
- *
- * Copyright (C) 2002, IBM Corp.
- *
- * All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
- * NON INFRINGEMENT.  See the GNU General Public License for more
- * details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- *
- * Send feedback to <colpatch@us.ibm.com>
- */
 #ifndef _ASM_M32R_TOPOLOGY_H
 #define _ASM_M32R_TOPOLOGY_H
 
-/* Other architectures wishing to use this simple topology API should fill
-   in the below functions as appropriate in their own <asm/topology.h> file. */
-
-#define cpu_to_node(cpu)	(0)
-
-#ifndef parent_node
-#define parent_node(node)	(0)
-#endif
-#ifndef node_to_cpumask
-#define node_to_cpumask(node)	(cpu_online_map)
-#endif
-#ifndef node_to_first_cpu
-#define node_to_first_cpu(node)	(0)
-#endif
-#ifndef pcibus_to_cpumask
-#define pcibus_to_cpumask(bus)	(cpu_online_map)
-#endif
+#include <asm-generic/topology.h>
 
 #endif /* _ASM_M32R_TOPOLOGY_H */
-- 
cgit v1.2.3-70-g09d2


From e164f5573bef0e6caf53519719cf0228c9c15ce3 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <christoph@lameter.com>
Date: Thu, 23 Jun 2005 00:08:17 -0700
Subject: [PATCH] ppc64: pcibus_to_node fix

asm-generic/topology.h must also be included if CONFIG_NUMA is set in order to
provide the fall back pcibus_to_node function.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-ppc64/topology.h | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-ppc64/topology.h b/include/asm-ppc64/topology.h
index d58d9dd7999..fcdcfd26a26 100644
--- a/include/asm-ppc64/topology.h
+++ b/include/asm-ppc64/topology.h
@@ -59,10 +59,8 @@ static inline int node_to_first_cpu(int node)
 	.nr_balance_failed	= 0,			\
 }
 
-#else /* !CONFIG_NUMA */
+#endif /* CONFIG_NUMA */
 
 #include <asm-generic/topology.h>
 
-#endif /* CONFIG_NUMA */
-
 #endif /* _ASM_PPC64_TOPOLOGY_H */
-- 
cgit v1.2.3-70-g09d2


From 8c5a09082f4e61a176382e96a831a0636b918602 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <christoph@lameter.com>
Date: Thu, 23 Jun 2005 00:08:18 -0700
Subject: [PATCH] x86/x86_64: pcibus_to_node

Define pcibus_to_node to be able to figure out which NUMA node contains a
given PCI device.  This defines pcibus_to_node(bus) in
include/linux/topology.h and adjusts the macros for i386 and x86_64 that
already provided a way to determine the cpumask of a pci device.

x86_64 was changed to not build an array of cpumasks anymore.  Instead an
array of nodes is build which can be used to generate the cpumask via
node_to_cpumask.

Signed-off-by: Christoph Lameter <christoph@lameter.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/mpparse.c   |  4 +++-
 arch/x86_64/pci/k8-bus.c       | 16 +---------------
 include/asm-generic/topology.h |  9 ++++++++-
 include/asm-i386/topology.h    |  8 ++------
 include/asm-x86_64/topology.h  | 14 +++-----------
 5 files changed, 17 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index 61a63be6b29..ed6a5588146 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -23,6 +23,7 @@
 #include <linux/kernel_stat.h>
 #include <linux/mc146818rtc.h>
 #include <linux/acpi.h>
+#include <linux/module.h>
 
 #include <asm/smp.h>
 #include <asm/mtrr.h>
@@ -45,7 +46,8 @@ int acpi_found_madt;
 int apic_version [MAX_APICS];
 unsigned char mp_bus_id_to_type [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
 int mp_bus_id_to_pci_bus [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
-cpumask_t pci_bus_to_cpumask [256] = { [0 ... 255] = CPU_MASK_ALL };
+unsigned char pci_bus_to_node [256];
+EXPORT_SYMBOL(pci_bus_to_node);
 
 static int mp_current_pci_id = 0;
 /* I/O APIC entries */
diff --git a/arch/x86_64/pci/k8-bus.c b/arch/x86_64/pci/k8-bus.c
index 62349c78db5..7e7d0c2a002 100644
--- a/arch/x86_64/pci/k8-bus.c
+++ b/arch/x86_64/pci/k8-bus.c
@@ -53,25 +53,11 @@ fill_mp_bus_to_cpumask(void)
 				for (j = SECONDARY_LDT_BUS_NUMBER(ldtbus);
 				     j <= SUBORDINATE_LDT_BUS_NUMBER(ldtbus);
 				     j++)
-					pci_bus_to_cpumask[j] =
-						node_to_cpumask(NODE_ID(nid));
+					pci_bus_to_node[j] = NODE_ID(nid);
 			}
 		}
 	}
 
-	/* quick sanity check */
-	printed = 0;
-	for (i = 0; i < 256; i++) {
-		if (cpus_empty(pci_bus_to_cpumask[i])) {
-			pci_bus_to_cpumask[i] = CPU_MASK_ALL;
-			if (printed)
-				continue;
-			printk(KERN_ERR
-			       "k8-bus.c: some busses have empty cpu mask\n");
-			printed = 1;
-		}
-	}
-
 	return 0;
 }
 
diff --git a/include/asm-generic/topology.h b/include/asm-generic/topology.h
index ec96e8b0f19..5d9d70cd17f 100644
--- a/include/asm-generic/topology.h
+++ b/include/asm-generic/topology.h
@@ -41,8 +41,15 @@
 #ifndef node_to_first_cpu
 #define node_to_first_cpu(node)	(0)
 #endif
+#ifndef pcibus_to_node
+#define pcibus_to_node(node)	(-1)
+#endif
+
 #ifndef pcibus_to_cpumask
-#define pcibus_to_cpumask(bus)	(cpu_online_map)
+#define pcibus_to_cpumask(bus)	(pcibus_to_node(bus) == -1 ? \
+					CPU_MASK_ALL : \
+					node_to_cpumask(pcibus_to_node(bus)) \
+				)
 #endif
 
 #endif /* _ASM_GENERIC_TOPOLOGY_H */
diff --git a/include/asm-i386/topology.h b/include/asm-i386/topology.h
index 98f9e6850cb..6d0f67507b2 100644
--- a/include/asm-i386/topology.h
+++ b/include/asm-i386/topology.h
@@ -60,12 +60,8 @@ static inline int node_to_first_cpu(int node)
 	return first_cpu(mask);
 }
 
-/* Returns the number of the node containing PCI bus number 'busnr' */
-static inline cpumask_t __pcibus_to_cpumask(int busnr)
-{
-	return node_to_cpumask(mp_bus_id_to_node[busnr]);
-}
-#define pcibus_to_cpumask(bus)	__pcibus_to_cpumask(bus->number)
+#define pcibus_to_node(bus) mp_bus_id_to_node[(bus)->number]
+#define pcibus_to_cpumask(bus) node_to_cpumask(pcibus_to_node(bus))
 
 /* sched_domains SD_NODE_INIT for NUMAQ machines */
 #define SD_NODE_INIT (struct sched_domain) {		\
diff --git a/include/asm-x86_64/topology.h b/include/asm-x86_64/topology.h
index da21573ec73..8f77e9f6bc2 100644
--- a/include/asm-x86_64/topology.h
+++ b/include/asm-x86_64/topology.h
@@ -13,8 +13,8 @@
 extern cpumask_t cpu_online_map;
 
 extern unsigned char cpu_to_node[];
+extern unsigned char pci_bus_to_node[];
 extern cpumask_t     node_to_cpumask[];
-extern cpumask_t pci_bus_to_cpumask[];
 
 #ifdef CONFIG_ACPI_NUMA
 extern int __node_distance(int, int);
@@ -26,16 +26,8 @@ extern int __node_distance(int, int);
 #define parent_node(node)		(node)
 #define node_to_first_cpu(node) 	(__ffs(node_to_cpumask[node]))
 #define node_to_cpumask(node)		(node_to_cpumask[node])
-
-static inline cpumask_t __pcibus_to_cpumask(int bus)
-{
-	cpumask_t busmask = pci_bus_to_cpumask[bus];
-	cpumask_t online = cpu_online_map;
-	cpumask_t res;
-	cpus_and(res, busmask, online);
-	return res;
-}
-#define pcibus_to_cpumask(bus) __pcibus_to_cpumask(bus->number)
+#define pcibus_to_node(bus)		pci_bus_to_node[(bus)->number]
+#define pcibus_to_cpumask(bus)		node_to_cpumask(pcibus_to_node(bus));
 
 /* sched_domains SD_NODE_INIT for x86_64 machines */
 #define SD_NODE_INIT (struct sched_domain) {		\
-- 
cgit v1.2.3-70-g09d2


From 1946089a109251655c5438d92c539bd2930e71ea Mon Sep 17 00:00:00 2001
From: Christoph Lameter <christoph@lameter.com>
Date: Thu, 23 Jun 2005 00:08:19 -0700
Subject: [PATCH] NUMA aware block device control structure allocation

Patch to allocate the control structures for for ide devices on the node of
the device itself (for NUMA systems).  The patch depends on the Slab API
change patch by Manfred and me (in mm) and the pcidev_to_node patch that I
posted today.

Does some realignment too.

Signed-off-by: Justin M. Forbes <jmforbes@linuxtx.org>
Signed-off-by: Christoph Lameter <christoph@lameter.com>
Signed-off-by: Pravin Shelar <pravin@calsoftinc.com>
Signed-off-by: Shobhit Dayal <shobhit@calsoftinc.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/as-iosched.c       |  8 +++++---
 drivers/block/deadline-iosched.c |  8 +++++---
 drivers/block/genhd.c            | 13 ++++++++++---
 drivers/block/ll_rw_blk.c        | 30 +++++++++++++++++++++++-------
 drivers/ide/ide-disk.c           |  3 ++-
 drivers/ide/ide-probe.c          |  8 +++++---
 include/linux/blkdev.h           |  6 +++++-
 include/linux/genhd.h            |  1 +
 include/linux/ide.h              |  2 +-
 include/linux/mempool.h          | 11 ++++++++---
 mm/mempool.c                     | 17 ++++++++++++-----
 11 files changed, 77 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/drivers/block/as-iosched.c b/drivers/block/as-iosched.c
index 638db06de2b..3410b4d294b 100644
--- a/drivers/block/as-iosched.c
+++ b/drivers/block/as-iosched.c
@@ -1871,20 +1871,22 @@ static int as_init_queue(request_queue_t *q, elevator_t *e)
 	if (!arq_pool)
 		return -ENOMEM;
 
-	ad = kmalloc(sizeof(*ad), GFP_KERNEL);
+	ad = kmalloc_node(sizeof(*ad), GFP_KERNEL, q->node);
 	if (!ad)
 		return -ENOMEM;
 	memset(ad, 0, sizeof(*ad));
 
 	ad->q = q; /* Identify what queue the data belongs to */
 
-	ad->hash = kmalloc(sizeof(struct list_head)*AS_HASH_ENTRIES,GFP_KERNEL);
+	ad->hash = kmalloc_node(sizeof(struct list_head)*AS_HASH_ENTRIES,
+				GFP_KERNEL, q->node);
 	if (!ad->hash) {
 		kfree(ad);
 		return -ENOMEM;
 	}
 
-	ad->arq_pool = mempool_create(BLKDEV_MIN_RQ, mempool_alloc_slab, mempool_free_slab, arq_pool);
+	ad->arq_pool = mempool_create_node(BLKDEV_MIN_RQ, mempool_alloc_slab,
+				mempool_free_slab, arq_pool, q->node);
 	if (!ad->arq_pool) {
 		kfree(ad->hash);
 		kfree(ad);
diff --git a/drivers/block/deadline-iosched.c b/drivers/block/deadline-iosched.c
index 7f79f3dd016..4bc2fea7327 100644
--- a/drivers/block/deadline-iosched.c
+++ b/drivers/block/deadline-iosched.c
@@ -711,18 +711,20 @@ static int deadline_init_queue(request_queue_t *q, elevator_t *e)
 	if (!drq_pool)
 		return -ENOMEM;
 
-	dd = kmalloc(sizeof(*dd), GFP_KERNEL);
+	dd = kmalloc_node(sizeof(*dd), GFP_KERNEL, q->node);
 	if (!dd)
 		return -ENOMEM;
 	memset(dd, 0, sizeof(*dd));
 
-	dd->hash = kmalloc(sizeof(struct list_head)*DL_HASH_ENTRIES,GFP_KERNEL);
+	dd->hash = kmalloc_node(sizeof(struct list_head)*DL_HASH_ENTRIES,
+				GFP_KERNEL, q->node);
 	if (!dd->hash) {
 		kfree(dd);
 		return -ENOMEM;
 	}
 
-	dd->drq_pool = mempool_create(BLKDEV_MIN_RQ, mempool_alloc_slab, mempool_free_slab, drq_pool);
+	dd->drq_pool = mempool_create_node(BLKDEV_MIN_RQ, mempool_alloc_slab,
+					mempool_free_slab, drq_pool, q->node);
 	if (!dd->drq_pool) {
 		kfree(dd->hash);
 		kfree(dd);
diff --git a/drivers/block/genhd.c b/drivers/block/genhd.c
index 53f7d846b74..43805e4d31e 100644
--- a/drivers/block/genhd.c
+++ b/drivers/block/genhd.c
@@ -582,10 +582,16 @@ struct seq_operations diskstats_op = {
 	.show	= diskstats_show
 };
 
-
 struct gendisk *alloc_disk(int minors)
 {
-	struct gendisk *disk = kmalloc(sizeof(struct gendisk), GFP_KERNEL);
+	return alloc_disk_node(minors, -1);
+}
+
+struct gendisk *alloc_disk_node(int minors, int node_id)
+{
+	struct gendisk *disk;
+
+	disk = kmalloc_node(sizeof(struct gendisk), GFP_KERNEL, node_id);
 	if (disk) {
 		memset(disk, 0, sizeof(struct gendisk));
 		if (!init_disk_stats(disk)) {
@@ -594,7 +600,7 @@ struct gendisk *alloc_disk(int minors)
 		}
 		if (minors > 1) {
 			int size = (minors - 1) * sizeof(struct hd_struct *);
-			disk->part = kmalloc(size, GFP_KERNEL);
+			disk->part = kmalloc_node(size, GFP_KERNEL, node_id);
 			if (!disk->part) {
 				kfree(disk);
 				return NULL;
@@ -610,6 +616,7 @@ struct gendisk *alloc_disk(int minors)
 }
 
 EXPORT_SYMBOL(alloc_disk);
+EXPORT_SYMBOL(alloc_disk_node);
 
 struct kobject *get_disk(struct gendisk *disk)
 {
diff --git a/drivers/block/ll_rw_blk.c b/drivers/block/ll_rw_blk.c
index 81fe3a0c1fe..cd8cf302068 100644
--- a/drivers/block/ll_rw_blk.c
+++ b/drivers/block/ll_rw_blk.c
@@ -28,6 +28,7 @@
 #include <linux/slab.h>
 #include <linux/swap.h>
 #include <linux/writeback.h>
+#include <linux/blkdev.h>
 
 /*
  * for max sense size
@@ -1645,7 +1646,8 @@ static int blk_init_free_list(request_queue_t *q)
 	init_waitqueue_head(&rl->wait[WRITE]);
 	init_waitqueue_head(&rl->drain);
 
-	rl->rq_pool = mempool_create(BLKDEV_MIN_RQ, mempool_alloc_slab, mempool_free_slab, request_cachep);
+	rl->rq_pool = mempool_create_node(BLKDEV_MIN_RQ, mempool_alloc_slab,
+				mempool_free_slab, request_cachep, q->node);
 
 	if (!rl->rq_pool)
 		return -ENOMEM;
@@ -1657,8 +1659,15 @@ static int __make_request(request_queue_t *, struct bio *);
 
 request_queue_t *blk_alloc_queue(int gfp_mask)
 {
-	request_queue_t *q = kmem_cache_alloc(requestq_cachep, gfp_mask);
+	return blk_alloc_queue_node(gfp_mask, -1);
+}
+EXPORT_SYMBOL(blk_alloc_queue);
+
+request_queue_t *blk_alloc_queue_node(int gfp_mask, int node_id)
+{
+	request_queue_t *q;
 
+	q = kmem_cache_alloc_node(requestq_cachep, gfp_mask, node_id);
 	if (!q)
 		return NULL;
 
@@ -1671,8 +1680,7 @@ request_queue_t *blk_alloc_queue(int gfp_mask)
 
 	return q;
 }
-
-EXPORT_SYMBOL(blk_alloc_queue);
+EXPORT_SYMBOL(blk_alloc_queue_node);
 
 /**
  * blk_init_queue  - prepare a request queue for use with a block device
@@ -1705,13 +1713,22 @@ EXPORT_SYMBOL(blk_alloc_queue);
  *    blk_init_queue() must be paired with a blk_cleanup_queue() call
  *    when the block device is deactivated (such as at module unload).
  **/
+
 request_queue_t *blk_init_queue(request_fn_proc *rfn, spinlock_t *lock)
 {
-	request_queue_t *q = blk_alloc_queue(GFP_KERNEL);
+	return blk_init_queue_node(rfn, lock, -1);
+}
+EXPORT_SYMBOL(blk_init_queue);
+
+request_queue_t *
+blk_init_queue_node(request_fn_proc *rfn, spinlock_t *lock, int node_id)
+{
+	request_queue_t *q = blk_alloc_queue_node(GFP_KERNEL, node_id);
 
 	if (!q)
 		return NULL;
 
+	q->node = node_id;
 	if (blk_init_free_list(q))
 		goto out_init;
 
@@ -1754,8 +1771,7 @@ out_init:
 	kmem_cache_free(requestq_cachep, q);
 	return NULL;
 }
-
-EXPORT_SYMBOL(blk_init_queue);
+EXPORT_SYMBOL(blk_init_queue_node);
 
 int blk_get_queue(request_queue_t *q)
 {
diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 3302cd8eab4..d6f934886b0 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -1215,7 +1215,8 @@ static int ide_disk_probe(struct device *dev)
 	if (!idkp)
 		goto failed;
 
-	g = alloc_disk(1 << PARTN_BITS);
+	g = alloc_disk_node(1 << PARTN_BITS,
+			pcibus_to_node(drive->hwif->pci_dev->bus));
 	if (!g)
 		goto out_free_idkp;
 
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 5d876f53c69..7df85af7537 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -977,8 +977,9 @@ static int ide_init_queue(ide_drive_t *drive)
 	 *	limits and LBA48 we could raise it but as yet
 	 *	do not.
 	 */
-	 
-	q = blk_init_queue(do_ide_request, &ide_lock);
+
+	q = blk_init_queue_node(do_ide_request, &ide_lock,
+				pcibus_to_node(drive->hwif->pci_dev->bus));
 	if (!q)
 		return 1;
 
@@ -1095,7 +1096,8 @@ static int init_irq (ide_hwif_t *hwif)
 		hwgroup->hwif->next = hwif;
 		spin_unlock_irq(&ide_lock);
 	} else {
-		hwgroup = kmalloc(sizeof(ide_hwgroup_t),GFP_KERNEL);
+		hwgroup = kmalloc_node(sizeof(ide_hwgroup_t), GFP_KERNEL,
+			pcibus_to_node(hwif->drives[0].hwif->pci_dev->bus));
 		if (!hwgroup)
 	       		goto out_up;
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 4a99b76c5a3..235c3414d26 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -396,6 +396,7 @@ struct request_queue
 	 */
 	unsigned int		sg_timeout;
 	unsigned int		sg_reserved_size;
+	int			node;
 
 	struct list_head	drain_list;
 
@@ -615,6 +616,8 @@ static inline void blkdev_dequeue_request(struct request *req)
 /*
  * Access functions for manipulating queue properties
  */
+extern request_queue_t *blk_init_queue_node(request_fn_proc *rfn,
+					spinlock_t *lock, int node_id);
 extern request_queue_t *blk_init_queue(request_fn_proc *, spinlock_t *);
 extern void blk_cleanup_queue(request_queue_t *);
 extern void blk_queue_make_request(request_queue_t *, make_request_fn *);
@@ -646,7 +649,8 @@ extern void blk_wait_queue_drained(request_queue_t *, int);
 extern void blk_finish_queue_drain(request_queue_t *);
 
 int blk_get_queue(request_queue_t *);
-request_queue_t *blk_alloc_queue(int);
+request_queue_t *blk_alloc_queue(int gfp_mask);
+request_queue_t *blk_alloc_queue_node(int,int);
 #define blk_put_queue(q) blk_cleanup_queue((q))
 
 /*
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 47dedaf971d..af26dc718ef 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -403,6 +403,7 @@ extern int rescan_partitions(struct gendisk *disk, struct block_device *bdev);
 extern void add_partition(struct gendisk *, int, sector_t, sector_t);
 extern void delete_partition(struct gendisk *, int);
 
+extern struct gendisk *alloc_disk_node(int minors, int node_id);
 extern struct gendisk *alloc_disk(int minors);
 extern struct kobject *get_disk(struct gendisk *disk);
 extern void put_disk(struct gendisk *disk);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 336d6e509f5..92129078d4f 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -917,7 +917,7 @@ typedef struct hwif_s {
 	unsigned dma;
 
 	void (*led_act)(void *data, int rw);
-} ide_hwif_t;
+} ____cacheline_maxaligned_in_smp ide_hwif_t;
 
 /*
  *  internal ide interrupt handler type
diff --git a/include/linux/mempool.h b/include/linux/mempool.h
index 4a36edf1c97..796220ce47c 100644
--- a/include/linux/mempool.h
+++ b/include/linux/mempool.h
@@ -20,9 +20,14 @@ typedef struct mempool_s {
 	mempool_free_t *free;
 	wait_queue_head_t wait;
 } mempool_t;
-extern mempool_t * mempool_create(int min_nr, mempool_alloc_t *alloc_fn,
-				 mempool_free_t *free_fn, void *pool_data);
-extern int mempool_resize(mempool_t *pool, int new_min_nr, unsigned int __nocast gfp_mask);
+
+extern mempool_t *mempool_create(int min_nr, mempool_alloc_t *alloc_fn,
+			mempool_free_t *free_fn, void *pool_data);
+extern mempool_t *mempool_create_node(int min_nr, mempool_alloc_t *alloc_fn,
+			mempool_free_t *free_fn, void *pool_data, int nid);
+
+extern int mempool_resize(mempool_t *pool, int new_min_nr,
+			unsigned int __nocast gfp_mask);
 extern void mempool_destroy(mempool_t *pool);
 extern void * mempool_alloc(mempool_t *pool, unsigned int __nocast gfp_mask);
 extern void mempool_free(void *element, mempool_t *pool);
diff --git a/mm/mempool.c b/mm/mempool.c
index c9f3d462042..920c8c3ab1b 100644
--- a/mm/mempool.c
+++ b/mm/mempool.c
@@ -51,16 +51,23 @@ static void free_pool(mempool_t *pool)
  * functions might sleep - as long as the mempool_alloc function is not called
  * from IRQ contexts.
  */
-mempool_t * mempool_create(int min_nr, mempool_alloc_t *alloc_fn,
+mempool_t *mempool_create(int min_nr, mempool_alloc_t *alloc_fn,
 				mempool_free_t *free_fn, void *pool_data)
 {
-	mempool_t *pool;
+	return  mempool_create_node(min_nr,alloc_fn,free_fn, pool_data,-1);
+}
+EXPORT_SYMBOL(mempool_create);
 
-	pool = kmalloc(sizeof(*pool), GFP_KERNEL);
+mempool_t *mempool_create_node(int min_nr, mempool_alloc_t *alloc_fn,
+			mempool_free_t *free_fn, void *pool_data, int node_id)
+{
+	mempool_t *pool;
+	pool = kmalloc_node(sizeof(*pool), GFP_KERNEL, node_id);
 	if (!pool)
 		return NULL;
 	memset(pool, 0, sizeof(*pool));
-	pool->elements = kmalloc(min_nr * sizeof(void *), GFP_KERNEL);
+	pool->elements = kmalloc_node(min_nr * sizeof(void *),
+					GFP_KERNEL, node_id);
 	if (!pool->elements) {
 		kfree(pool);
 		return NULL;
@@ -87,7 +94,7 @@ mempool_t * mempool_create(int min_nr, mempool_alloc_t *alloc_fn,
 	}
 	return pool;
 }
-EXPORT_SYMBOL(mempool_create);
+EXPORT_SYMBOL(mempool_create_node);
 
 /**
  * mempool_resize - resize an existing memory pool
-- 
cgit v1.2.3-70-g09d2


From ca05fea6db5259c6d62e517c41d448a4249175f4 Mon Sep 17 00:00:00 2001
From: Natalie Protasevich <Natalie.Protasevich@unisys.com>
Date: Thu, 23 Jun 2005 00:08:22 -0700
Subject: [PATCH] Do not enforce unique IO_APIC_ID check for xAPIC systems
 (i386)

This patch is per Andi's request to remove NO_IOAPIC_CHECK from genapic and
use heuristics to prevent unique I/O APIC ID check for systems that don't
need it.  The patch disables unique I/O APIC ID check for Xeon-based and
other platforms that don't use serial APIC bus for interrupt delivery.
Andi stated that AMD systems don't need unique IO_APIC_IDs either.

Signed-off-by: Natalie Protasevich <Natalie.Protasevich@unisys.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/io_apic.c                | 10 ++++++----
 arch/i386/kernel/mpparse.c                |  5 ++++-
 include/asm-i386/genapic.h                |  1 -
 include/asm-i386/mach-bigsmp/mach_apic.h  |  2 --
 include/asm-i386/mach-default/mach_apic.h |  2 --
 include/asm-i386/mach-es7000/mach_apic.h  |  2 --
 include/asm-i386/mach-generic/mach_apic.h |  1 -
 include/asm-i386/mach-numaq/mach_apic.h   |  2 --
 include/asm-i386/mach-summit/mach_apic.h  |  2 --
 include/asm-i386/mach-visws/mach_apic.h   |  2 --
 10 files changed, 10 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index 7a324e8b86f..51f7a5d8721 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -1658,6 +1658,12 @@ static void __init setup_ioapic_ids_from_mpc(void)
 	unsigned char old_id;
 	unsigned long flags;
 
+	/*
+	 * Don't check I/O APIC IDs for xAPIC systems.  They have
+	 * no meaning without the serial APIC bus.
+	 */
+	if (!(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL && boot_cpu_data.x86 < 15))
+		return;
 	/*
 	 * This is broken; anything with a real cpu count has to
 	 * circumvent this idiocy regardless.
@@ -1684,10 +1690,6 @@ static void __init setup_ioapic_ids_from_mpc(void)
 			mp_ioapics[apic].mpc_apicid = reg_00.bits.ID;
 		}
 
-		/* Don't check I/O APIC IDs for some xAPIC systems.  They have
-		 * no meaning without the serial APIC bus. */
-		if (NO_IOAPIC_CHECK)
-			continue;
 		/*
 		 * Sanity check, is the ID really free? Every APIC in a
 		 * system must have a unique ID or we get lots of nice
diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index 1347ab4939e..0a061056b82 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -914,7 +914,10 @@ void __init mp_register_ioapic (
 	mp_ioapics[idx].mpc_apicaddr = address;
 
 	set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
-	mp_ioapics[idx].mpc_apicid = io_apic_get_unique_id(idx, id);
+	if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) && (boot_cpu_data.x86 < 15))
+		mp_ioapics[idx].mpc_apicid = io_apic_get_unique_id(idx, id);
+	else
+		mp_ioapics[idx].mpc_apicid = id;
 	mp_ioapics[idx].mpc_apicver = io_apic_get_version(idx);
 	
 	/* 
diff --git a/include/asm-i386/genapic.h b/include/asm-i386/genapic.h
index fc813b2e827..b3783a32abe 100644
--- a/include/asm-i386/genapic.h
+++ b/include/asm-i386/genapic.h
@@ -78,7 +78,6 @@ struct genapic {
 	.int_delivery_mode = INT_DELIVERY_MODE, \
 	.int_dest_mode = INT_DEST_MODE, \
 	.no_balance_irq = NO_BALANCE_IRQ, \
-	.no_ioapic_check = NO_IOAPIC_CHECK, \
 	.ESR_DISABLE = esr_disable, \
 	.apic_destination_logical = APIC_DEST_LOGICAL, \
 	APICFUNC(apic_id_registered), \
diff --git a/include/asm-i386/mach-bigsmp/mach_apic.h b/include/asm-i386/mach-bigsmp/mach_apic.h
index 2339868270e..ba936d4daed 100644
--- a/include/asm-i386/mach-bigsmp/mach_apic.h
+++ b/include/asm-i386/mach-bigsmp/mach_apic.h
@@ -14,8 +14,6 @@
 #define NO_BALANCE_IRQ (1)
 #define esr_disable (1)
 
-#define NO_IOAPIC_CHECK (0)
-
 static inline int apic_id_registered(void)
 {
 	return (1);
diff --git a/include/asm-i386/mach-default/mach_apic.h b/include/asm-i386/mach-default/mach_apic.h
index 627f1cd084b..3ef6292db78 100644
--- a/include/asm-i386/mach-default/mach_apic.h
+++ b/include/asm-i386/mach-default/mach_apic.h
@@ -19,8 +19,6 @@ static inline cpumask_t target_cpus(void)
 #define NO_BALANCE_IRQ (0)
 #define esr_disable (0)
 
-#define NO_IOAPIC_CHECK (0)
-
 #define INT_DELIVERY_MODE dest_LowestPrio
 #define INT_DEST_MODE 1     /* logical delivery broadcast to all procs */
 
diff --git a/include/asm-i386/mach-es7000/mach_apic.h b/include/asm-i386/mach-es7000/mach_apic.h
index ceab2c464b1..b5f3f0d0b2b 100644
--- a/include/asm-i386/mach-es7000/mach_apic.h
+++ b/include/asm-i386/mach-es7000/mach_apic.h
@@ -38,8 +38,6 @@ static inline cpumask_t target_cpus(void)
 #define WAKE_SECONDARY_VIA_INIT
 #endif
 
-#define NO_IOAPIC_CHECK (1)
-
 static inline unsigned long check_apicid_used(physid_mask_t bitmap, int apicid)
 { 
 	return 0;
diff --git a/include/asm-i386/mach-generic/mach_apic.h b/include/asm-i386/mach-generic/mach_apic.h
index ab36d02ebed..b13767a4e93 100644
--- a/include/asm-i386/mach-generic/mach_apic.h
+++ b/include/asm-i386/mach-generic/mach_apic.h
@@ -5,7 +5,6 @@
 
 #define esr_disable (genapic->ESR_DISABLE)
 #define NO_BALANCE_IRQ (genapic->no_balance_irq)
-#define NO_IOAPIC_CHECK	(genapic->no_ioapic_check)
 #define INT_DELIVERY_MODE (genapic->int_delivery_mode)
 #define INT_DEST_MODE (genapic->int_dest_mode)
 #undef APIC_DEST_LOGICAL
diff --git a/include/asm-i386/mach-numaq/mach_apic.h b/include/asm-i386/mach-numaq/mach_apic.h
index e1a04494764..9d158095da8 100644
--- a/include/asm-i386/mach-numaq/mach_apic.h
+++ b/include/asm-i386/mach-numaq/mach_apic.h
@@ -17,8 +17,6 @@ static inline cpumask_t target_cpus(void)
 #define NO_BALANCE_IRQ (1)
 #define esr_disable (1)
 
-#define NO_IOAPIC_CHECK (0)
-
 #define INT_DELIVERY_MODE dest_LowestPrio
 #define INT_DEST_MODE 0     /* physical delivery on LOCAL quad */
  
diff --git a/include/asm-i386/mach-summit/mach_apic.h b/include/asm-i386/mach-summit/mach_apic.h
index 74e9cbc8c01..3d6d12937e1 100644
--- a/include/asm-i386/mach-summit/mach_apic.h
+++ b/include/asm-i386/mach-summit/mach_apic.h
@@ -7,8 +7,6 @@
 #define esr_disable (1)
 #define NO_BALANCE_IRQ (0)
 
-#define NO_IOAPIC_CHECK (1)	/* Don't check I/O APIC ID for xAPIC */
-
 /* In clustered mode, the high nibble of APIC ID is a cluster number.
  * The low nibble is a 4-bit bitmap. */
 #define XAPIC_DEST_CPUS_SHIFT	4
diff --git a/include/asm-i386/mach-visws/mach_apic.h b/include/asm-i386/mach-visws/mach_apic.h
index 4e6cdfb8b09..de438c7147a 100644
--- a/include/asm-i386/mach-visws/mach_apic.h
+++ b/include/asm-i386/mach-visws/mach_apic.h
@@ -9,8 +9,6 @@
 #define no_balance_irq (0)
 #define esr_disable (0)
 
-#define NO_IOAPIC_CHECK (0)
-
 #define INT_DELIVERY_MODE dest_LowestPrio
 #define INT_DEST_MODE 1     /* logical delivery broadcast to all procs */
 
-- 
cgit v1.2.3-70-g09d2


From 59121003721a8fad11ee72e646fd9d3076b5679c Mon Sep 17 00:00:00 2001
From: Christoph Lameter <christoph@lameter.com>
Date: Thu, 23 Jun 2005 00:08:25 -0700
Subject: [PATCH] i386: Selectable Frequency of the Timer Interrupt

Make the timer frequency selectable. The timer interrupt may cause bus
and memory contention in large NUMA systems since the interrupt occurs
on each processor HZ times per second.

Signed-off-by: Christoph Lameter <christoph@lameter.com>
Signed-off-by: Shai Fultheim <shai@scalex86.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig          |  2 ++
 arch/x86_64/Kconfig        |  2 ++
 include/asm-i386/param.h   |  4 +++-
 include/asm-x86_64/param.h |  6 ++++--
 kernel/Kconfig.hz          | 46 ++++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 57 insertions(+), 3 deletions(-)
 create mode 100644 kernel/Kconfig.hz

(limited to 'include')

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index bfdcedef06e..d4ae5f9ceae 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -961,6 +961,8 @@ config SECCOMP
 
 	  If unsure, say Y. Only embedded should say N here.
 
+source kernel/Kconfig.hz
+
 endmenu
 
 
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 61ed1665234..db259757dc8 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -402,6 +402,8 @@ config SECCOMP
 
 	  If unsure, say Y. Only embedded should say N here.
 
+source kernel/Kconfig.hz
+
 endmenu
 
 #
diff --git a/include/asm-i386/param.h b/include/asm-i386/param.h
index b6440526e42..fa02e67ea86 100644
--- a/include/asm-i386/param.h
+++ b/include/asm-i386/param.h
@@ -1,8 +1,10 @@
+#include <linux/config.h>
+
 #ifndef _ASMi386_PARAM_H
 #define _ASMi386_PARAM_H
 
 #ifdef __KERNEL__
-# define HZ		1000		/* Internal kernel timer frequency */
+# define HZ		CONFIG_HZ	/* Internal kernel timer frequency */
 # define USER_HZ	100		/* .. some user interfaces are in "ticks" */
 # define CLOCKS_PER_SEC		(USER_HZ)	/* like times() */
 #endif
diff --git a/include/asm-x86_64/param.h b/include/asm-x86_64/param.h
index b707f0568c9..40b11937180 100644
--- a/include/asm-x86_64/param.h
+++ b/include/asm-x86_64/param.h
@@ -1,9 +1,11 @@
+#include <linux/config.h>
+
 #ifndef _ASMx86_64_PARAM_H
 #define _ASMx86_64_PARAM_H
 
 #ifdef __KERNEL__
-# define HZ            1000            /* Internal kernel timer frequency */
-# define USER_HZ       100          /* .. some user interfaces are in "ticks */
+# define HZ            CONFIG_HZ	/* Internal kernel timer frequency */
+# define USER_HZ       100		/* .. some user interfaces are in "ticks */
 #define CLOCKS_PER_SEC        (USER_HZ)       /* like times() */
 #endif
 
diff --git a/kernel/Kconfig.hz b/kernel/Kconfig.hz
new file mode 100644
index 00000000000..248e1c396f8
--- /dev/null
+++ b/kernel/Kconfig.hz
@@ -0,0 +1,46 @@
+#
+# Timer Interrupt Frequency Configuration
+#
+
+choice
+	prompt "Timer frequency"
+	default HZ_250
+	help
+	 Allows the configuration of the timer frequency. It is customary
+	 to have the timer interrupt run at 1000 HZ but 100 HZ may be more
+	 beneficial for servers and NUMA systems that do not need to have
+	 a fast response for user interaction and that may experience bus
+	 contention and cacheline bounces as a result of timer interrupts.
+	 Note that the timer interrupt occurs on each processor in an SMP
+	 environment leading to NR_CPUS * HZ number of timer interrupts
+	 per second.
+
+
+	config HZ_100
+		bool "100 HZ"
+	help
+	  100 HZ is a typical choice for servers, SMP and NUMA systems
+	  with lots of processors that may show reduced performance if
+	  too many timer interrupts are occurring.
+
+	config HZ_250
+		bool "250 HZ"
+	help
+	 250 HZ is a good compromise choice allowing server performance
+	 while also showing good interactive responsiveness even
+	 on SMP and NUMA systems.
+
+	config HZ_1000
+		bool "1000 HZ"
+	help
+	 1000 HZ is the preferred choice for desktop systems and other
+	 systems requiring fast interactive responses to events.
+
+endchoice
+
+config HZ
+	int
+	default 100 if HZ_100
+	default 250 if HZ_250
+	default 1000 if HZ_1000
+
-- 
cgit v1.2.3-70-g09d2


From b5d23e5b8c7ecd97d32f6ad7680d9909977580a7 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@engr.sgi.com>
Date: Thu, 23 Jun 2005 00:08:27 -0700
Subject: [PATCH] ia64: Selectable Timer Interrupt Frequency

It allows a selectable timer interrupt frequency of 100, 250 and 1000 HZ.
Reducing the timer frequency may have important performance benefits on
large systems.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/Kconfig        | 2 ++
 include/asm-ia64/param.h | 2 +-
 2 files changed, 3 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 9f6a46caba9..01b78e7f992 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -161,6 +161,8 @@ config IA64_PAGE_SIZE_64KB
 
 endchoice
 
+source kernel/Kconfig.hz
+
 config IA64_BRL_EMU
 	bool
 	depends on ITANIUM
diff --git a/include/asm-ia64/param.h b/include/asm-ia64/param.h
index 6c6b679b7a9..5e1e0d2d7ba 100644
--- a/include/asm-ia64/param.h
+++ b/include/asm-ia64/param.h
@@ -27,7 +27,7 @@
    */
 #  define HZ	  32
 # else
-#  define HZ	1024
+#  define HZ	CONFIG_HZ
 # endif
 # define USER_HZ	HZ
 # define CLOCKS_PER_SEC	HZ	/* frequency at which times() counts */
-- 
cgit v1.2.3-70-g09d2


From a9ed8817966dd723754a990f1003264a21b5747e Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Thu, 23 Jun 2005 00:08:32 -0700
Subject: [PATCH] x86: #include asm/uaccess.h in asm/checksum.h

csum_and_copy_to_user is static inline and uses VERIFY_WRITE.  Patch allows
to remove asm/uaccess.h from i386_ksyms.c without dependency surprises.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/checksum.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/asm-i386/checksum.h b/include/asm-i386/checksum.h
index 641342002bc..f949e44c2a3 100644
--- a/include/asm-i386/checksum.h
+++ b/include/asm-i386/checksum.h
@@ -3,6 +3,8 @@
 
 #include <linux/in6.h>
 
+#include <asm/uaccess.h>
+
 /*
  * computes the checksum of a memory block at buff, length len,
  * and adds in "sum" (32-bit)
-- 
cgit v1.2.3-70-g09d2


From a3a255e744dfa672e741dc24306491139d0de2d8 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Thu, 23 Jun 2005 00:08:34 -0700
Subject: [PATCH] x86: cpu_khz type fix

x86_64's cpu_khz is unsigned int and there is no reason why x86 needs to use
unsigned long.

So make cpu_khz unsigned int on x86 as well.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/cpu/proc.c          | 2 +-
 arch/i386/kernel/smpboot.c           | 2 +-
 arch/i386/kernel/time.c              | 2 +-
 arch/i386/kernel/timers/common.c     | 3 ++-
 arch/i386/kernel/timers/timer_hpet.c | 2 +-
 arch/i386/kernel/timers/timer_tsc.c  | 7 ++++---
 include/asm-i386/timex.h             | 2 +-
 7 files changed, 11 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/cpu/proc.c b/arch/i386/kernel/cpu/proc.c
index 7323c19f354..8bd77d948a8 100644
--- a/arch/i386/kernel/cpu/proc.c
+++ b/arch/i386/kernel/cpu/proc.c
@@ -86,7 +86,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 		seq_printf(m, "stepping\t: unknown\n");
 
 	if ( cpu_has(c, X86_FEATURE_TSC) ) {
-		seq_printf(m, "cpu MHz\t\t: %lu.%03lu\n",
+		seq_printf(m, "cpu MHz\t\t: %u.%03u\n",
 			cpu_khz / 1000, (cpu_khz % 1000));
 	}
 
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index aaad95e7f87..c20d96d5c15 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -205,7 +205,7 @@ static void __init synchronize_tsc_bp (void)
 	unsigned long long t0;
 	unsigned long long sum, avg;
 	long long delta;
-	unsigned long one_usec;
+	unsigned int one_usec;
 	int buggy = 0;
 
 	printk(KERN_INFO "checking TSC synchronization across %u CPUs: ", num_booting_cpus());
diff --git a/arch/i386/kernel/time.c b/arch/i386/kernel/time.c
index 8bc8363fbb4..e68d9fdb075 100644
--- a/arch/i386/kernel/time.c
+++ b/arch/i386/kernel/time.c
@@ -77,7 +77,7 @@ u64 jiffies_64 = INITIAL_JIFFIES;
 
 EXPORT_SYMBOL(jiffies_64);
 
-unsigned long cpu_khz;	/* Detected as we calibrate the TSC */
+unsigned int cpu_khz;	/* Detected as we calibrate the TSC */
 EXPORT_SYMBOL(cpu_khz);
 
 extern unsigned long wall_jiffies;
diff --git a/arch/i386/kernel/timers/common.c b/arch/i386/kernel/timers/common.c
index b38cc0d0c71..37353bd3180 100644
--- a/arch/i386/kernel/timers/common.c
+++ b/arch/i386/kernel/timers/common.c
@@ -163,7 +163,8 @@ void init_cpu_khz(void)
 		       		:"=a" (cpu_khz), "=d" (edx)
         	       		:"r" (tsc_quotient),
 	                	"0" (eax), "1" (edx));
-				printk("Detected %lu.%03lu MHz processor.\n", cpu_khz / 1000, cpu_khz % 1000);
+				printk("Detected %u.%03u MHz processor.\n",
+					cpu_khz / 1000, cpu_khz % 1000);
 			}
 		}
 	}
diff --git a/arch/i386/kernel/timers/timer_hpet.c b/arch/i386/kernel/timers/timer_hpet.c
index 15a7d727bd6..d766e0963ac 100644
--- a/arch/i386/kernel/timers/timer_hpet.c
+++ b/arch/i386/kernel/timers/timer_hpet.c
@@ -158,7 +158,7 @@ static int __init init_hpet(char* override)
 			{	unsigned long eax=0, edx=1000;
 				ASM_DIV64_REG(cpu_khz, edx, tsc_quotient,
 						eax, edx);
-				printk("Detected %lu.%03lu MHz processor.\n",
+				printk("Detected %u.%03u MHz processor.\n",
 					cpu_khz / 1000, cpu_khz % 1000);
 			}
 			set_cyc2ns_scale(cpu_khz/1000);
diff --git a/arch/i386/kernel/timers/timer_tsc.c b/arch/i386/kernel/timers/timer_tsc.c
index 27f08ae9120..54c36b18202 100644
--- a/arch/i386/kernel/timers/timer_tsc.c
+++ b/arch/i386/kernel/timers/timer_tsc.c
@@ -256,7 +256,7 @@ static unsigned long loops_per_jiffy_ref = 0;
 
 #ifndef CONFIG_SMP
 static unsigned long fast_gettimeoffset_ref = 0;
-static unsigned long cpu_khz_ref = 0;
+static unsigned int cpu_khz_ref = 0;
 #endif
 
 static int
@@ -323,7 +323,7 @@ static inline void cpufreq_delayed_get(void) { return; }
 int recalibrate_cpu_khz(void)
 {
 #ifndef CONFIG_SMP
-	unsigned long cpu_khz_old = cpu_khz;
+	unsigned int cpu_khz_old = cpu_khz;
 
 	if (cpu_has_tsc) {
 		init_cpu_khz();
@@ -534,7 +534,8 @@ static int __init init_tsc(char* override)
 		       		:"=a" (cpu_khz), "=d" (edx)
         	       		:"r" (tsc_quotient),
 	                	"0" (eax), "1" (edx));
-				printk("Detected %lu.%03lu MHz processor.\n", cpu_khz / 1000, cpu_khz % 1000);
+				printk("Detected %u.%03u MHz processor.\n",
+					cpu_khz / 1000, cpu_khz % 1000);
 			}
 			set_cyc2ns_scale(cpu_khz/1000);
 			return 0;
diff --git a/include/asm-i386/timex.h b/include/asm-i386/timex.h
index e370f907bd3..292b5a68f62 100644
--- a/include/asm-i386/timex.h
+++ b/include/asm-i386/timex.h
@@ -47,7 +47,7 @@ static inline cycles_t get_cycles (void)
 	return ret;
 }
 
-extern unsigned long cpu_khz;
+extern unsigned int cpu_khz;
 
 extern int read_current_timer(unsigned long *timer_value);
 #define ARCH_HAS_READ_CURRENT_TIMER	1
-- 
cgit v1.2.3-70-g09d2


From 32ecd42b6f94d3ee320a22827b46bd19ccf924e5 Mon Sep 17 00:00:00 2001
From: Jan Beulich <JBeulich@novell.com>
Date: Thu, 23 Jun 2005 00:08:38 -0700
Subject: [PATCH] eliminate duplicate rdpmc definition

Eliminate duplicate definition of rdpmc in x86-64's mtrr.h.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-x86_64/msr.h | 5 -----
 1 file changed, 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86_64/msr.h b/include/asm-x86_64/msr.h
index 513e52c7182..bc700232728 100644
--- a/include/asm-x86_64/msr.h
+++ b/include/asm-x86_64/msr.h
@@ -57,11 +57,6 @@
      (val) = ((unsigned long)__a) | (((unsigned long)__d)<<32); \
 } while(0)
 
-#define rdpmc(counter,low,high) \
-     __asm__ __volatile__("rdpmc" \
-			  : "=a" (low), "=d" (high) \
-			  : "c" (counter))
-
 #define write_tsc(val1,val2) wrmsr(0x10, val1, val2)
 
 #define rdpmc(counter,low,high) \
-- 
cgit v1.2.3-70-g09d2


From f5012310e35bd62fd39fce338ee44422c975ff3c Mon Sep 17 00:00:00 2001
From: Vincent Hanquez <vincent.hanquez@cl.cam.ac.uk>
Date: Thu, 23 Jun 2005 00:08:42 -0700
Subject: [PATCH] xen: x86: add macro for debugreg

Add 2 macros to set and get debugreg on x86.  This is useful for Xen because
it will need only to redefine each macro to a hypervisor call.

Signed-off-by: Vincent Hanquez <vincent.hanquez@cl.cam.ac.uk>
Cc: Ian Pratt <m+Ian.Pratt@cl.cam.ac.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/processor.h | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h
index 359bb015174..c76c50e9622 100644
--- a/include/asm-i386/processor.h
+++ b/include/asm-i386/processor.h
@@ -501,12 +501,16 @@ static inline void load_esp0(struct tss_struct *tss, struct thread_struct *threa
 } while (0)
 
 /*
- * This special macro can be used to load a debugging register
+ * These special macros can be used to get or set a debugging register
  */
-#define loaddebug(thread,register) \
-               __asm__("movl %0,%%db" #register  \
-                       : /* no output */ \
-                       :"r" ((thread)->debugreg[register]))
+#define get_debugreg(var, register)				\
+		__asm__("movl %%db" #register ", %0"		\
+			:"=r" (var))
+#define set_debugreg(value, register)			\
+		__asm__("movl %0,%%db" #register		\
+			: /* no output */			\
+			:"r" (value))
+
 
 /* Forward declaration, a strange C thing */
 struct task_struct;
-- 
cgit v1.2.3-70-g09d2


From fa1e1bdf78d405f9905b8290ee9211e7a7bbc99b Mon Sep 17 00:00:00 2001
From: Vincent Hanquez <vincent.hanquez@cl.cam.ac.uk>
Date: Thu, 23 Jun 2005 00:08:44 -0700
Subject: [PATCH] xen: x86: Rename usermode macro

Rename user_mode to user_mode_vm and add a user_mode macro similar to the
x86-64 one.

This is useful for Xen because the linux xen kernel does not runs on the same
priviledge that a vanilla linux kernel, and with this we just need to redefine
user_mode().

Signed-off-by: Vincent Hanquez <vincent.hanquez@cl.cam.ac.uk>
Cc: Ian Pratt <m+Ian.Pratt@cl.cam.ac.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/apic.c              | 2 +-
 arch/i386/kernel/ptrace.c            | 2 +-
 arch/i386/mach-voyager/voyager_smp.c | 2 +-
 arch/i386/oprofile/backtrace.c       | 2 +-
 include/asm-i386/ptrace.h            | 3 ++-
 include/asm-x86_64/ptrace.h          | 1 +
 6 files changed, 7 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index d509836b70c..8d993fa7175 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -1133,7 +1133,7 @@ inline void smp_local_timer_interrupt(struct pt_regs * regs)
 		}
 
 #ifdef CONFIG_SMP
-		update_process_times(user_mode(regs));
+		update_process_times(user_mode_vm(regs));
 #endif
 	}
 
diff --git a/arch/i386/kernel/ptrace.c b/arch/i386/kernel/ptrace.c
index e34f651fa13..0da59b42843 100644
--- a/arch/i386/kernel/ptrace.c
+++ b/arch/i386/kernel/ptrace.c
@@ -668,7 +668,7 @@ void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code)
 	info.si_code = TRAP_BRKPT;
 
 	/* User-mode eip? */
-	info.si_addr = user_mode(regs) ? (void __user *) regs->eip : NULL;
+	info.si_addr = user_mode_vm(regs) ? (void __user *) regs->eip : NULL;
 
 	/* Send us the fakey SIGTRAP */
 	force_sig_info(SIGTRAP, &info, tsk);
diff --git a/arch/i386/mach-voyager/voyager_smp.c b/arch/i386/mach-voyager/voyager_smp.c
index a6e0ddd65bd..8c8527593da 100644
--- a/arch/i386/mach-voyager/voyager_smp.c
+++ b/arch/i386/mach-voyager/voyager_smp.c
@@ -1288,7 +1288,7 @@ smp_local_timer_interrupt(struct pt_regs * regs)
 						per_cpu(prof_counter, cpu);
 		}
 
-		update_process_times(user_mode(regs));
+		update_process_times(user_mode_vm(regs));
 	}
 
 	if( ((1<<cpu) & voyager_extended_vic_processors) == 0)
diff --git a/arch/i386/oprofile/backtrace.c b/arch/i386/oprofile/backtrace.c
index 52d72e074f7..65dfd2edb67 100644
--- a/arch/i386/oprofile/backtrace.c
+++ b/arch/i386/oprofile/backtrace.c
@@ -91,7 +91,7 @@ x86_backtrace(struct pt_regs * const regs, unsigned int depth)
 	head = (struct frame_head *)regs->ebp;
 #endif
 
-	if (!user_mode(regs)) {
+	if (!user_mode_vm(regs)) {
 		while (depth-- && valid_kernel_stack(head, regs))
 			head = dump_backtrace(head);
 		return;
diff --git a/include/asm-i386/ptrace.h b/include/asm-i386/ptrace.h
index 8618914b352..eef9f93870d 100644
--- a/include/asm-i386/ptrace.h
+++ b/include/asm-i386/ptrace.h
@@ -57,7 +57,8 @@ struct pt_regs {
 #ifdef __KERNEL__
 struct task_struct;
 extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code);
-#define user_mode(regs) ((VM_MASK & (regs)->eflags) || (3 & (regs)->xcs))
+#define user_mode(regs)		(3 & (regs)->xcs)
+#define user_mode_vm(regs)	((VM_MASK & (regs)->eflags) || user_mode(regs))
 #define instruction_pointer(regs) ((regs)->eip)
 #if defined(CONFIG_SMP) && defined(CONFIG_FRAME_POINTER)
 extern unsigned long profile_pc(struct pt_regs *regs);
diff --git a/include/asm-x86_64/ptrace.h b/include/asm-x86_64/ptrace.h
index 5bbc8d3141c..ca6f15ff61d 100644
--- a/include/asm-x86_64/ptrace.h
+++ b/include/asm-x86_64/ptrace.h
@@ -82,6 +82,7 @@ struct pt_regs {
 
 #if defined(__KERNEL__) && !defined(__ASSEMBLY__) 
 #define user_mode(regs) (!!((regs)->cs & 3))
+#define user_mode_vm(regs) user_mode(regs)
 #define instruction_pointer(regs) ((regs)->rip)
 extern unsigned long profile_pc(struct pt_regs *regs);
 void signal_fault(struct pt_regs *regs, void __user *frame, char *where);
-- 
cgit v1.2.3-70-g09d2


From e9129e56e9ec50c0689eb4cf7a3ca132f1e776db Mon Sep 17 00:00:00 2001
From: Vincent Hanquez <vincent.hanquez@cl.cam.ac.uk>
Date: Thu, 23 Jun 2005 00:08:46 -0700
Subject: [PATCH] xen: x86_64: Add macro for debugreg

Add 2 macros to set and get debugreg on x86_64.  This is useful for Xen
because it will need only to redefine each macro to a hypervisor call.

Signed-off-by: Vincent Hanquez <vincent.hanquez@cl.cam.ac.uk>
Cc: Ian Pratt <m+Ian.Pratt@cl.cam.ac.uk>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/signal.c    | 2 +-
 arch/x86_64/kernel/traps.c     | 4 ++--
 include/asm-x86_64/processor.h | 8 ++++++++
 3 files changed, 11 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/signal.c b/arch/x86_64/kernel/signal.c
index 3c5f3089371..0446e8b96a2 100644
--- a/arch/x86_64/kernel/signal.c
+++ b/arch/x86_64/kernel/signal.c
@@ -439,7 +439,7 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
 		 * inside the kernel.
 		 */
 		if (current->thread.debugreg7)
-			asm volatile("movq %0,%%db7"	: : "r" (current->thread.debugreg7));
+			set_debugreg(current->thread.debugreg7, 7);
 
 		/* Whee!  Actually deliver the signal.  */
 		return handle_signal(signr, &info, &ka, oldset, regs);
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 3dfec8fdabc..21706c1a0f9 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -669,7 +669,7 @@ asmlinkage void do_debug(struct pt_regs * regs, unsigned long error_code)
        }
 #endif
 
-	asm("movq %%db6,%0" : "=r" (condition));
+	get_debugreg(condition, 6);
 
 	if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code,
 						SIGTRAP) == NOTIFY_STOP)
@@ -721,7 +721,7 @@ asmlinkage void do_debug(struct pt_regs * regs, unsigned long error_code)
 	info.si_addr = (void __user *)regs->rip;
 	force_sig_info(SIGTRAP, &info, tsk);	
 clear_dr7:
-	asm volatile("movq %0,%%db7"::"r"(0UL));
+	set_debugreg(0UL, 7);
 	return;
 
 clear_TF_reenable:
diff --git a/include/asm-x86_64/processor.h b/include/asm-x86_64/processor.h
index 8b55f139968..106f666517b 100644
--- a/include/asm-x86_64/processor.h
+++ b/include/asm-x86_64/processor.h
@@ -280,6 +280,14 @@ struct thread_struct {
 	set_fs(USER_DS);							 \
 } while(0) 
 
+#define get_debugreg(var, register)				\
+		__asm__("movq %%db" #register ", %0"		\
+			:"=r" (var))
+#define set_debugreg(value, register)			\
+		__asm__("movq %0,%%db" #register		\
+			: /* no output */			\
+			:"r" (value))
+
 struct task_struct;
 struct mm_struct;
 
-- 
cgit v1.2.3-70-g09d2


From fa72b903f75e4f0f0b2c2feed093005167da4023 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Thu, 23 Jun 2005 00:08:49 -0700
Subject: [PATCH] blk: remove blk_queue_tag->real_max_depth optimization

blk_queue_tag->real_max_depth was used to optimize out unnecessary
allocations/frees on tag resize.  However, the whole thing was very broken -
tag_map was never allocated to real_max_depth resulting in access beyond the
end of the map, bits in [max_depth..real_max_depth] were set when initializing
a map and copied when resizing resulting in pre-occupied tags.

As the gain of the optimization is very small, well, almost nill, remove the
whole thing.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Acked-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/ll_rw_blk.c | 35 ++++++++++-------------------------
 include/linux/blkdev.h    |  1 -
 2 files changed, 10 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/drivers/block/ll_rw_blk.c b/drivers/block/ll_rw_blk.c
index 808390c7420..896d17c28f4 100644
--- a/drivers/block/ll_rw_blk.c
+++ b/drivers/block/ll_rw_blk.c
@@ -717,7 +717,7 @@ struct request *blk_queue_find_tag(request_queue_t *q, int tag)
 {
 	struct blk_queue_tag *bqt = q->queue_tags;
 
-	if (unlikely(bqt == NULL || tag >= bqt->real_max_depth))
+	if (unlikely(bqt == NULL || tag >= bqt->max_depth))
 		return NULL;
 
 	return bqt->tag_index[tag];
@@ -775,9 +775,9 @@ EXPORT_SYMBOL(blk_queue_free_tags);
 static int
 init_tag_map(request_queue_t *q, struct blk_queue_tag *tags, int depth)
 {
-	int bits, i;
 	struct request **tag_index;
 	unsigned long *tag_map;
+	int nr_ulongs;
 
 	if (depth > q->nr_requests * 2) {
 		depth = q->nr_requests * 2;
@@ -789,24 +789,17 @@ init_tag_map(request_queue_t *q, struct blk_queue_tag *tags, int depth)
 	if (!tag_index)
 		goto fail;
 
-	bits = (depth / BLK_TAGS_PER_LONG) + 1;
-	tag_map = kmalloc(bits * sizeof(unsigned long), GFP_ATOMIC);
+	nr_ulongs = ALIGN(depth, BLK_TAGS_PER_LONG) / BLK_TAGS_PER_LONG;
+	tag_map = kmalloc(nr_ulongs * sizeof(unsigned long), GFP_ATOMIC);
 	if (!tag_map)
 		goto fail;
 
 	memset(tag_index, 0, depth * sizeof(struct request *));
-	memset(tag_map, 0, bits * sizeof(unsigned long));
+	memset(tag_map, 0, nr_ulongs * sizeof(unsigned long));
 	tags->max_depth = depth;
-	tags->real_max_depth = bits * BITS_PER_LONG;
 	tags->tag_index = tag_index;
 	tags->tag_map = tag_map;
 
-	/*
-	 * set the upper bits if the depth isn't a multiple of the word size
-	 */
-	for (i = depth; i < bits * BLK_TAGS_PER_LONG; i++)
-		__set_bit(i, tag_map);
-
 	return 0;
 fail:
 	kfree(tag_index);
@@ -871,32 +864,24 @@ int blk_queue_resize_tags(request_queue_t *q, int new_depth)
 	struct blk_queue_tag *bqt = q->queue_tags;
 	struct request **tag_index;
 	unsigned long *tag_map;
-	int bits, max_depth;
+	int max_depth, nr_ulongs;
 
 	if (!bqt)
 		return -ENXIO;
 
-	/*
-	 * don't bother sizing down
-	 */
-	if (new_depth <= bqt->real_max_depth) {
-		bqt->max_depth = new_depth;
-		return 0;
-	}
-
 	/*
 	 * save the old state info, so we can copy it back
 	 */
 	tag_index = bqt->tag_index;
 	tag_map = bqt->tag_map;
-	max_depth = bqt->real_max_depth;
+	max_depth = bqt->max_depth;
 
 	if (init_tag_map(q, bqt, new_depth))
 		return -ENOMEM;
 
 	memcpy(bqt->tag_index, tag_index, max_depth * sizeof(struct request *));
-	bits = max_depth / BLK_TAGS_PER_LONG;
-	memcpy(bqt->tag_map, tag_map, bits * sizeof(unsigned long));
+	nr_ulongs = ALIGN(max_depth, BLK_TAGS_PER_LONG) / BLK_TAGS_PER_LONG;
+	memcpy(bqt->tag_map, tag_map, nr_ulongs * sizeof(unsigned long));
 
 	kfree(tag_index);
 	kfree(tag_map);
@@ -926,7 +911,7 @@ void blk_queue_end_tag(request_queue_t *q, struct request *rq)
 
 	BUG_ON(tag == -1);
 
-	if (unlikely(tag >= bqt->real_max_depth))
+	if (unlikely(tag >= bqt->max_depth))
 		return;
 
 	if (unlikely(!__test_and_clear_bit(tag, bqt->tag_map))) {
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 235c3414d26..8d7e2f4151d 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -294,7 +294,6 @@ struct blk_queue_tag {
 	struct list_head busy_list;	/* fifo list of busy tags */
 	int busy;			/* current depth */
 	int max_depth;			/* what we will send to device */
-	int real_max_depth;		/* what the array can hold */
 	atomic_t refcnt;		/* map can be shared */
 };
 
-- 
cgit v1.2.3-70-g09d2


From f7d37d028dfba90b1b747f8ac685bf0959aeda8b Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Thu, 23 Jun 2005 00:08:50 -0700
Subject: [PATCH] blk: remove BLK_TAGS_{PER_LONG|MASK}

Replace BLK_TAGS_PER_LONG with BITS_PER_LONG and remove unused BLK_TAGS_MASK.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Acked-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/ll_rw_blk.c | 4 ++--
 include/linux/blkdev.h    | 3 ---
 2 files changed, 2 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/drivers/block/ll_rw_blk.c b/drivers/block/ll_rw_blk.c
index 896d17c28f4..99afeec1031 100644
--- a/drivers/block/ll_rw_blk.c
+++ b/drivers/block/ll_rw_blk.c
@@ -789,7 +789,7 @@ init_tag_map(request_queue_t *q, struct blk_queue_tag *tags, int depth)
 	if (!tag_index)
 		goto fail;
 
-	nr_ulongs = ALIGN(depth, BLK_TAGS_PER_LONG) / BLK_TAGS_PER_LONG;
+	nr_ulongs = ALIGN(depth, BITS_PER_LONG) / BITS_PER_LONG;
 	tag_map = kmalloc(nr_ulongs * sizeof(unsigned long), GFP_ATOMIC);
 	if (!tag_map)
 		goto fail;
@@ -880,7 +880,7 @@ int blk_queue_resize_tags(request_queue_t *q, int new_depth)
 		return -ENOMEM;
 
 	memcpy(bqt->tag_index, tag_index, max_depth * sizeof(struct request *));
-	nr_ulongs = ALIGN(max_depth, BLK_TAGS_PER_LONG) / BLK_TAGS_PER_LONG;
+	nr_ulongs = ALIGN(max_depth, BITS_PER_LONG) / BITS_PER_LONG;
 	memcpy(bqt->tag_map, tag_map, nr_ulongs * sizeof(unsigned long));
 
 	kfree(tag_index);
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 8d7e2f4151d..60272141ff1 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -285,9 +285,6 @@ enum blk_queue_state {
 	Queue_up,
 };
 
-#define BLK_TAGS_PER_LONG	(sizeof(unsigned long) * 8)
-#define BLK_TAGS_MASK		(BLK_TAGS_PER_LONG - 1)
-
 struct blk_queue_tag {
 	struct request **tag_index;	/* map of busy tags */
 	unsigned long *tag_map;		/* bit map of free/busy tags */
-- 
cgit v1.2.3-70-g09d2


From 55c888d6d09a0df236adfaf8ccf06ff5d0646775 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Thu, 23 Jun 2005 00:08:56 -0700
Subject: [PATCH] timers fixes/improvements

This patch tries to solve following problems:

1. del_timer_sync() is racy. The timer can be fired again after
   del_timer_sync have checked all cpus and before it will recheck
   timer_pending().

2. It has scalability problems. All cpus are scanned to determine
   if the timer is running on that cpu.

   With this patch del_timer_sync is O(1) and no slower than plain
   del_timer(pending_timer), unless it has to actually wait for
   completion of the currently running timer.

   The only restriction is that the recurring timer should not use
   add_timer_on().

3. The timers are not serialized wrt to itself.

   If CPU_0 does mod_timer(jiffies+1) while the timer is currently
   running on CPU 1, it is quite possible that local interrupt on
   CPU_0 will start that timer before it finished on CPU_1.

4. The timers locking is suboptimal. __mod_timer() takes 3 locks
   at once and still requires wmb() in del_timer/run_timers.

   The new implementation takes 2 locks sequentially and does not
   need memory barriers.

Currently ->base != NULL means that the timer is pending. In that case
->base.lock is used to lock the timer. __mod_timer also takes timer->lock
because ->base can be == NULL.

This patch uses timer->entry.next != NULL as indication that the timer is
pending. So it does __list_del(), entry->next = NULL instead of list_del()
when the timer is deleted.

The ->base field is used for hashed locking only, it is initialized
in init_timer() which sets ->base = per_cpu(tvec_bases). When the
tvec_bases.lock is locked, it means that all timers which are tied
to this base via timer->base are locked, and the base itself is locked
too.

So __run_timers/migrate_timers can safely modify all timers which could
be found on ->tvX lists (pending timers).

When the timer's base is locked, and the timer removed from ->entry list
(which means that _run_timers/migrate_timers can't see this timer), it is
possible to set timer->base = NULL and drop the lock: the timer remains
locked.

This patch adds lock_timer_base() helper, which waits for ->base != NULL,
locks the ->base, and checks it is still the same.

__mod_timer() schedules the timer on the local CPU and changes it's base.
However, it does not lock both old and new bases at once. It locks the
timer via lock_timer_base(), deletes the timer, sets ->base = NULL, and
unlocks old base. Then __mod_timer() locks new_base, sets ->base = new_base,
and adds this timer. This simplifies the code, because AB-BA deadlock is not
possible. __mod_timer() also ensures that the timer's base is not changed
while the timer's handler is running on the old base.

__run_timers(), del_timer() do not change ->base anymore, they only clear
pending flag.

So del_timer_sync() can test timer->base->running_timer == timer to detect
whether it is running or not.

We don't need timer_list->lock anymore, this patch kills it.

We also don't need barriers. del_timer() and __run_timers() used smp_wmb()
before clearing timer's pending flag. It was needed because __mod_timer()
did not lock old_base if the timer is not pending, so __mod_timer()->list_add()
could race with del_timer()->list_del(). With this patch these functions are
serialized through base->lock.

One problem. TIMER_INITIALIZER can't use per_cpu(tvec_bases). So this patch
adds global

        struct timer_base_s {
                spinlock_t lock;
                struct timer_list *running_timer;
        } __init_timer_base;

which is used by TIMER_INITIALIZER. The corresponding fields in tvec_t_base_s
struct are replaced by struct timer_base_s t_base.

It is indeed ugly. But this can't have scalability problems. The global
__init_timer_base.lock is used only when __mod_timer() is called for the first
time AND the timer was compile time initialized. After that the timer migrates
to the local CPU.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Renaud Lienhart <renaud.lienhart@free.fr>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/timer.h |  30 ++---
 kernel/timer.c        | 328 ++++++++++++++++++++++++--------------------------
 2 files changed, 166 insertions(+), 192 deletions(-)

(limited to 'include')

diff --git a/include/linux/timer.h b/include/linux/timer.h
index 90db1cc62dd..2e78fedfc06 100644
--- a/include/linux/timer.h
+++ b/include/linux/timer.h
@@ -6,45 +6,33 @@
 #include <linux/spinlock.h>
 #include <linux/stddef.h>
 
-struct tvec_t_base_s;
+struct timer_base_s;
 
 struct timer_list {
 	struct list_head entry;
 	unsigned long expires;
 
-	spinlock_t lock;
 	unsigned long magic;
 
 	void (*function)(unsigned long);
 	unsigned long data;
 
-	struct tvec_t_base_s *base;
+	struct timer_base_s *base;
 };
 
 #define TIMER_MAGIC	0x4b87ad6e
 
+extern struct timer_base_s __init_timer_base;
+
 #define TIMER_INITIALIZER(_function, _expires, _data) {		\
 		.function = (_function),			\
 		.expires = (_expires),				\
 		.data = (_data),				\
-		.base = NULL,					\
+		.base = &__init_timer_base,			\
 		.magic = TIMER_MAGIC,				\
-		.lock = SPIN_LOCK_UNLOCKED,			\
 	}
 
-/***
- * init_timer - initialize a timer.
- * @timer: the timer to be initialized
- *
- * init_timer() must be done to a timer prior calling *any* of the
- * other timer functions.
- */
-static inline void init_timer(struct timer_list * timer)
-{
-	timer->base = NULL;
-	timer->magic = TIMER_MAGIC;
-	spin_lock_init(&timer->lock);
-}
+void fastcall init_timer(struct timer_list * timer);
 
 /***
  * timer_pending - is a timer pending?
@@ -58,7 +46,7 @@ static inline void init_timer(struct timer_list * timer)
  */
 static inline int timer_pending(const struct timer_list * timer)
 {
-	return timer->base != NULL;
+	return timer->entry.next != NULL;
 }
 
 extern void add_timer_on(struct timer_list *timer, int cpu);
@@ -89,12 +77,12 @@ static inline void add_timer(struct timer_list * timer)
 
 #ifdef CONFIG_SMP
   extern int del_timer_sync(struct timer_list *timer);
-  extern int del_singleshot_timer_sync(struct timer_list *timer);
 #else
 # define del_timer_sync(t) del_timer(t)
-# define del_singleshot_timer_sync(t) del_timer(t)
 #endif
 
+#define del_singleshot_timer_sync(t) del_timer_sync(t)
+
 extern void init_timers(void);
 extern void run_local_timers(void);
 extern void it_real_fn(unsigned long);
diff --git a/kernel/timer.c b/kernel/timer.c
index 207aa4f0aa1..8aadc62efd6 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -57,6 +57,11 @@ static void time_interpolator_update(long delta_nsec);
 #define TVN_MASK (TVN_SIZE - 1)
 #define TVR_MASK (TVR_SIZE - 1)
 
+struct timer_base_s {
+	spinlock_t lock;
+	struct timer_list *running_timer;
+};
+
 typedef struct tvec_s {
 	struct list_head vec[TVN_SIZE];
 } tvec_t;
@@ -66,9 +71,8 @@ typedef struct tvec_root_s {
 } tvec_root_t;
 
 struct tvec_t_base_s {
-	spinlock_t lock;
+	struct timer_base_s t_base;
 	unsigned long timer_jiffies;
-	struct timer_list *running_timer;
 	tvec_root_t tv1;
 	tvec_t tv2;
 	tvec_t tv3;
@@ -77,18 +81,16 @@ struct tvec_t_base_s {
 } ____cacheline_aligned_in_smp;
 
 typedef struct tvec_t_base_s tvec_base_t;
+static DEFINE_PER_CPU(tvec_base_t, tvec_bases);
 
 static inline void set_running_timer(tvec_base_t *base,
 					struct timer_list *timer)
 {
 #ifdef CONFIG_SMP
-	base->running_timer = timer;
+	base->t_base.running_timer = timer;
 #endif
 }
 
-/* Fake initialization */
-static DEFINE_PER_CPU(tvec_base_t, tvec_bases) = { SPIN_LOCK_UNLOCKED };
-
 static void check_timer_failed(struct timer_list *timer)
 {
 	static int whine_count;
@@ -103,7 +105,6 @@ static void check_timer_failed(struct timer_list *timer)
 	/*
 	 * Now fix it up
 	 */
-	spin_lock_init(&timer->lock);
 	timer->magic = TIMER_MAGIC;
 }
 
@@ -156,65 +157,113 @@ static void internal_add_timer(tvec_base_t *base, struct timer_list *timer)
 	list_add_tail(&timer->entry, vec);
 }
 
+typedef struct timer_base_s timer_base_t;
+/*
+ * Used by TIMER_INITIALIZER, we can't use per_cpu(tvec_bases)
+ * at compile time, and we need timer->base to lock the timer.
+ */
+timer_base_t __init_timer_base
+	____cacheline_aligned_in_smp = { .lock = SPIN_LOCK_UNLOCKED };
+EXPORT_SYMBOL(__init_timer_base);
+
+/***
+ * init_timer - initialize a timer.
+ * @timer: the timer to be initialized
+ *
+ * init_timer() must be done to a timer prior calling *any* of the
+ * other timer functions.
+ */
+void fastcall init_timer(struct timer_list *timer)
+{
+	timer->entry.next = NULL;
+	timer->base = &per_cpu(tvec_bases, raw_smp_processor_id()).t_base;
+	timer->magic = TIMER_MAGIC;
+}
+EXPORT_SYMBOL(init_timer);
+
+static inline void detach_timer(struct timer_list *timer,
+					int clear_pending)
+{
+	struct list_head *entry = &timer->entry;
+
+	__list_del(entry->prev, entry->next);
+	if (clear_pending)
+		entry->next = NULL;
+	entry->prev = LIST_POISON2;
+}
+
+/*
+ * We are using hashed locking: holding per_cpu(tvec_bases).t_base.lock
+ * means that all timers which are tied to this base via timer->base are
+ * locked, and the base itself is locked too.
+ *
+ * So __run_timers/migrate_timers can safely modify all timers which could
+ * be found on ->tvX lists.
+ *
+ * When the timer's base is locked, and the timer removed from list, it is
+ * possible to set timer->base = NULL and drop the lock: the timer remains
+ * locked.
+ */
+static timer_base_t *lock_timer_base(struct timer_list *timer,
+					unsigned long *flags)
+{
+	timer_base_t *base;
+
+	for (;;) {
+		base = timer->base;
+		if (likely(base != NULL)) {
+			spin_lock_irqsave(&base->lock, *flags);
+			if (likely(base == timer->base))
+				return base;
+			/* The timer has migrated to another CPU */
+			spin_unlock_irqrestore(&base->lock, *flags);
+		}
+		cpu_relax();
+	}
+}
+
 int __mod_timer(struct timer_list *timer, unsigned long expires)
 {
-	tvec_base_t *old_base, *new_base;
+	timer_base_t *base;
+	tvec_base_t *new_base;
 	unsigned long flags;
 	int ret = 0;
 
 	BUG_ON(!timer->function);
-
 	check_timer(timer);
 
-	spin_lock_irqsave(&timer->lock, flags);
+	base = lock_timer_base(timer, &flags);
+
+	if (timer_pending(timer)) {
+		detach_timer(timer, 0);
+		ret = 1;
+	}
+
 	new_base = &__get_cpu_var(tvec_bases);
-repeat:
-	old_base = timer->base;
 
-	/*
-	 * Prevent deadlocks via ordering by old_base < new_base.
-	 */
-	if (old_base && (new_base != old_base)) {
-		if (old_base < new_base) {
-			spin_lock(&new_base->lock);
-			spin_lock(&old_base->lock);
-		} else {
-			spin_lock(&old_base->lock);
-			spin_lock(&new_base->lock);
-		}
+	if (base != &new_base->t_base) {
 		/*
-		 * The timer base might have been cancelled while we were
-		 * trying to take the lock(s):
+		 * We are trying to schedule the timer on the local CPU.
+		 * However we can't change timer's base while it is running,
+		 * otherwise del_timer_sync() can't detect that the timer's
+		 * handler yet has not finished. This also guarantees that
+		 * the timer is serialized wrt itself.
 		 */
-		if (timer->base != old_base) {
-			spin_unlock(&new_base->lock);
-			spin_unlock(&old_base->lock);
-			goto repeat;
-		}
-	} else {
-		spin_lock(&new_base->lock);
-		if (timer->base != old_base) {
-			spin_unlock(&new_base->lock);
-			goto repeat;
+		if (unlikely(base->running_timer == timer)) {
+			/* The timer remains on a former base */
+			new_base = container_of(base, tvec_base_t, t_base);
+		} else {
+			/* See the comment in lock_timer_base() */
+			timer->base = NULL;
+			spin_unlock(&base->lock);
+			spin_lock(&new_base->t_base.lock);
+			timer->base = &new_base->t_base;
 		}
 	}
 
-	/*
-	 * Delete the previous timeout (if there was any), and install
-	 * the new one:
-	 */
-	if (old_base) {
-		list_del(&timer->entry);
-		ret = 1;
-	}
 	timer->expires = expires;
 	internal_add_timer(new_base, timer);
-	timer->base = new_base;
-
-	if (old_base && (new_base != old_base))
-		spin_unlock(&old_base->lock);
-	spin_unlock(&new_base->lock);
-	spin_unlock_irqrestore(&timer->lock, flags);
+	spin_unlock_irqrestore(&new_base->t_base.lock, flags);
 
 	return ret;
 }
@@ -232,15 +281,15 @@ void add_timer_on(struct timer_list *timer, int cpu)
 {
 	tvec_base_t *base = &per_cpu(tvec_bases, cpu);
   	unsigned long flags;
-  
+
   	BUG_ON(timer_pending(timer) || !timer->function);
 
 	check_timer(timer);
 
-	spin_lock_irqsave(&base->lock, flags);
+	spin_lock_irqsave(&base->t_base.lock, flags);
+	timer->base = &base->t_base;
 	internal_add_timer(base, timer);
-	timer->base = base;
-	spin_unlock_irqrestore(&base->lock, flags);
+	spin_unlock_irqrestore(&base->t_base.lock, flags);
 }
 
 
@@ -295,27 +344,22 @@ EXPORT_SYMBOL(mod_timer);
  */
 int del_timer(struct timer_list *timer)
 {
+	timer_base_t *base;
 	unsigned long flags;
-	tvec_base_t *base;
+	int ret = 0;
 
 	check_timer(timer);
 
-repeat:
- 	base = timer->base;
-	if (!base)
-		return 0;
-	spin_lock_irqsave(&base->lock, flags);
-	if (base != timer->base) {
+	if (timer_pending(timer)) {
+		base = lock_timer_base(timer, &flags);
+		if (timer_pending(timer)) {
+			detach_timer(timer, 1);
+			ret = 1;
+		}
 		spin_unlock_irqrestore(&base->lock, flags);
-		goto repeat;
 	}
-	list_del(&timer->entry);
-	/* Need to make sure that anybody who sees a NULL base also sees the list ops */
-	smp_wmb();
-	timer->base = NULL;
-	spin_unlock_irqrestore(&base->lock, flags);
 
-	return 1;
+	return ret;
 }
 
 EXPORT_SYMBOL(del_timer);
@@ -332,72 +376,39 @@ EXPORT_SYMBOL(del_timer);
  * Synchronization rules: callers must prevent restarting of the timer,
  * otherwise this function is meaningless. It must not be called from
  * interrupt contexts. The caller must not hold locks which would prevent
- * completion of the timer's handler.  Upon exit the timer is not queued and
- * the handler is not running on any CPU.
+ * completion of the timer's handler. The timer's handler must not call
+ * add_timer_on(). Upon exit the timer is not queued and the handler is
+ * not running on any CPU.
  *
  * The function returns whether it has deactivated a pending timer or not.
- *
- * del_timer_sync() is slow and complicated because it copes with timer
- * handlers which re-arm the timer (periodic timers).  If the timer handler
- * is known to not do this (a single shot timer) then use
- * del_singleshot_timer_sync() instead.
  */
 int del_timer_sync(struct timer_list *timer)
 {
-	tvec_base_t *base;
-	int i, ret = 0;
+	timer_base_t *base;
+	unsigned long flags;
+	int ret = -1;
 
 	check_timer(timer);
 
-del_again:
-	ret += del_timer(timer);
+	do {
+		base = lock_timer_base(timer, &flags);
 
-	for_each_online_cpu(i) {
-		base = &per_cpu(tvec_bases, i);
-		if (base->running_timer == timer) {
-			while (base->running_timer == timer) {
-				cpu_relax();
-				preempt_check_resched();
-			}
-			break;
+		if (base->running_timer == timer)
+			goto unlock;
+
+		ret = 0;
+		if (timer_pending(timer)) {
+			detach_timer(timer, 1);
+			ret = 1;
 		}
-	}
-	smp_rmb();
-	if (timer_pending(timer))
-		goto del_again;
+unlock:
+		spin_unlock_irqrestore(&base->lock, flags);
+	} while (ret < 0);
 
 	return ret;
 }
-EXPORT_SYMBOL(del_timer_sync);
 
-/***
- * del_singleshot_timer_sync - deactivate a non-recursive timer
- * @timer: the timer to be deactivated
- *
- * This function is an optimization of del_timer_sync for the case where the
- * caller can guarantee the timer does not reschedule itself in its timer
- * function.
- *
- * Synchronization rules: callers must prevent restarting of the timer,
- * otherwise this function is meaningless. It must not be called from
- * interrupt contexts. The caller must not hold locks which wold prevent
- * completion of the timer's handler.  Upon exit the timer is not queued and
- * the handler is not running on any CPU.
- *
- * The function returns whether it has deactivated a pending timer or not.
- */
-int del_singleshot_timer_sync(struct timer_list *timer)
-{
-	int ret = del_timer(timer);
-
-	if (!ret) {
-		ret = del_timer_sync(timer);
-		BUG_ON(ret);
-	}
-
-	return ret;
-}
-EXPORT_SYMBOL(del_singleshot_timer_sync);
+EXPORT_SYMBOL(del_timer_sync);
 #endif
 
 static int cascade(tvec_base_t *base, tvec_t *tv, int index)
@@ -415,7 +426,7 @@ static int cascade(tvec_base_t *base, tvec_t *tv, int index)
 		struct timer_list *tmp;
 
 		tmp = list_entry(curr, struct timer_list, entry);
-		BUG_ON(tmp->base != base);
+		BUG_ON(tmp->base != &base->t_base);
 		curr = curr->next;
 		internal_add_timer(base, tmp);
 	}
@@ -437,7 +448,7 @@ static inline void __run_timers(tvec_base_t *base)
 {
 	struct timer_list *timer;
 
-	spin_lock_irq(&base->lock);
+	spin_lock_irq(&base->t_base.lock);
 	while (time_after_eq(jiffies, base->timer_jiffies)) {
 		struct list_head work_list = LIST_HEAD_INIT(work_list);
 		struct list_head *head = &work_list;
@@ -453,8 +464,7 @@ static inline void __run_timers(tvec_base_t *base)
 			cascade(base, &base->tv5, INDEX(3));
 		++base->timer_jiffies; 
 		list_splice_init(base->tv1.vec + index, &work_list);
-repeat:
-		if (!list_empty(head)) {
+		while (!list_empty(head)) {
 			void (*fn)(unsigned long);
 			unsigned long data;
 
@@ -462,11 +472,9 @@ repeat:
  			fn = timer->function;
  			data = timer->data;
 
-			list_del(&timer->entry);
 			set_running_timer(base, timer);
-			smp_wmb();
-			timer->base = NULL;
-			spin_unlock_irq(&base->lock);
+			detach_timer(timer, 1);
+			spin_unlock_irq(&base->t_base.lock);
 			{
 				u32 preempt_count = preempt_count();
 				fn(data);
@@ -475,12 +483,11 @@ repeat:
 					BUG();
 				}
 			}
-			spin_lock_irq(&base->lock);
-			goto repeat;
+			spin_lock_irq(&base->t_base.lock);
 		}
 	}
 	set_running_timer(base, NULL);
-	spin_unlock_irq(&base->lock);
+	spin_unlock_irq(&base->t_base.lock);
 }
 
 #ifdef CONFIG_NO_IDLE_HZ
@@ -499,7 +506,7 @@ unsigned long next_timer_interrupt(void)
 	int i, j;
 
 	base = &__get_cpu_var(tvec_bases);
-	spin_lock(&base->lock);
+	spin_lock(&base->t_base.lock);
 	expires = base->timer_jiffies + (LONG_MAX >> 1);
 	list = 0;
 
@@ -547,7 +554,7 @@ found:
 				expires = nte->expires;
 		}
 	}
-	spin_unlock(&base->lock);
+	spin_unlock(&base->t_base.lock);
 	return expires;
 }
 #endif
@@ -1286,9 +1293,9 @@ static void __devinit init_timers_cpu(int cpu)
 {
 	int j;
 	tvec_base_t *base;
-       
+
 	base = &per_cpu(tvec_bases, cpu);
-	spin_lock_init(&base->lock);
+	spin_lock_init(&base->t_base.lock);
 	for (j = 0; j < TVN_SIZE; j++) {
 		INIT_LIST_HEAD(base->tv5.vec + j);
 		INIT_LIST_HEAD(base->tv4.vec + j);
@@ -1302,22 +1309,16 @@ static void __devinit init_timers_cpu(int cpu)
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
-static int migrate_timer_list(tvec_base_t *new_base, struct list_head *head)
+static void migrate_timer_list(tvec_base_t *new_base, struct list_head *head)
 {
 	struct timer_list *timer;
 
 	while (!list_empty(head)) {
 		timer = list_entry(head->next, struct timer_list, entry);
-		/* We're locking backwards from __mod_timer order here,
-		   beware deadlock. */
-		if (!spin_trylock(&timer->lock))
-			return 0;
-		list_del(&timer->entry);
+		detach_timer(timer, 0);
+		timer->base = &new_base->t_base;
 		internal_add_timer(new_base, timer);
-		timer->base = new_base;
-		spin_unlock(&timer->lock);
 	}
-	return 1;
 }
 
 static void __devinit migrate_timers(int cpu)
@@ -1331,39 +1332,24 @@ static void __devinit migrate_timers(int cpu)
 	new_base = &get_cpu_var(tvec_bases);
 
 	local_irq_disable();
-again:
-	/* Prevent deadlocks via ordering by old_base < new_base. */
-	if (old_base < new_base) {
-		spin_lock(&new_base->lock);
-		spin_lock(&old_base->lock);
-	} else {
-		spin_lock(&old_base->lock);
-		spin_lock(&new_base->lock);
-	}
+	spin_lock(&new_base->t_base.lock);
+	spin_lock(&old_base->t_base.lock);
 
-	if (old_base->running_timer)
+	if (old_base->t_base.running_timer)
 		BUG();
 	for (i = 0; i < TVR_SIZE; i++)
-		if (!migrate_timer_list(new_base, old_base->tv1.vec + i))
-			goto unlock_again;
-	for (i = 0; i < TVN_SIZE; i++)
-		if (!migrate_timer_list(new_base, old_base->tv2.vec + i)
-		    || !migrate_timer_list(new_base, old_base->tv3.vec + i)
-		    || !migrate_timer_list(new_base, old_base->tv4.vec + i)
-		    || !migrate_timer_list(new_base, old_base->tv5.vec + i))
-			goto unlock_again;
-	spin_unlock(&old_base->lock);
-	spin_unlock(&new_base->lock);
+		migrate_timer_list(new_base, old_base->tv1.vec + i);
+	for (i = 0; i < TVN_SIZE; i++) {
+		migrate_timer_list(new_base, old_base->tv2.vec + i);
+		migrate_timer_list(new_base, old_base->tv3.vec + i);
+		migrate_timer_list(new_base, old_base->tv4.vec + i);
+		migrate_timer_list(new_base, old_base->tv5.vec + i);
+	}
+
+	spin_unlock(&old_base->t_base.lock);
+	spin_unlock(&new_base->t_base.lock);
 	local_irq_enable();
 	put_cpu_var(tvec_bases);
-	return;
-
-unlock_again:
-	/* Avoid deadlock with __mod_timer, by backing off. */
-	spin_unlock(&old_base->lock);
-	spin_unlock(&new_base->lock);
-	cpu_relax();
-	goto again;
 }
 #endif /* CONFIG_HOTPLUG_CPU */
 
-- 
cgit v1.2.3-70-g09d2


From fd450b7318b75343fd76b3d95416853e34e72c95 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Thu, 23 Jun 2005 00:08:59 -0700
Subject: [PATCH] timers: introduce try_to_del_timer_sync()

This patch splits del_timer_sync() into 2 functions.  The new one,
try_to_del_timer_sync(), returns -1 when it hits executing timer.

It can be used in interrupt context, or when the caller hold locks which
can prevent completion of the timer's handler.

NOTE.  Currently it can't be used in interrupt context in UP case, because
->running_timer is used only with CONFIG_SMP.

Should the need arise, it is possible to kill #ifdef CONFIG_SMP in
set_running_timer(), it is cheap.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/timer.h |  4 +++-
 kernel/timer.c        | 53 ++++++++++++++++++++++++++++++++-------------------
 2 files changed, 36 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/linux/timer.h b/include/linux/timer.h
index 2e78fedfc06..221f81ac200 100644
--- a/include/linux/timer.h
+++ b/include/linux/timer.h
@@ -76,9 +76,11 @@ static inline void add_timer(struct timer_list * timer)
 }
 
 #ifdef CONFIG_SMP
+  extern int try_to_del_timer_sync(struct timer_list *timer);
   extern int del_timer_sync(struct timer_list *timer);
 #else
-# define del_timer_sync(t) del_timer(t)
+# define try_to_del_timer_sync(t)	del_timer(t)
+# define del_timer_sync(t)		del_timer(t)
 #endif
 
 #define del_singleshot_timer_sync(t) del_timer_sync(t)
diff --git a/kernel/timer.c b/kernel/timer.c
index 8aadc62efd6..1f986c16d89 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -365,6 +365,34 @@ int del_timer(struct timer_list *timer)
 EXPORT_SYMBOL(del_timer);
 
 #ifdef CONFIG_SMP
+/*
+ * This function tries to deactivate a timer. Upon successful (ret >= 0)
+ * exit the timer is not queued and the handler is not running on any CPU.
+ *
+ * It must not be called from interrupt contexts.
+ */
+int try_to_del_timer_sync(struct timer_list *timer)
+{
+	timer_base_t *base;
+	unsigned long flags;
+	int ret = -1;
+
+	base = lock_timer_base(timer, &flags);
+
+	if (base->running_timer == timer)
+		goto out;
+
+	ret = 0;
+	if (timer_pending(timer)) {
+		detach_timer(timer, 1);
+		ret = 1;
+	}
+out:
+	spin_unlock_irqrestore(&base->lock, flags);
+
+	return ret;
+}
+
 /***
  * del_timer_sync - deactivate a timer and wait for the handler to finish.
  * @timer: the timer to be deactivated
@@ -384,28 +412,13 @@ EXPORT_SYMBOL(del_timer);
  */
 int del_timer_sync(struct timer_list *timer)
 {
-	timer_base_t *base;
-	unsigned long flags;
-	int ret = -1;
-
 	check_timer(timer);
 
-	do {
-		base = lock_timer_base(timer, &flags);
-
-		if (base->running_timer == timer)
-			goto unlock;
-
-		ret = 0;
-		if (timer_pending(timer)) {
-			detach_timer(timer, 1);
-			ret = 1;
-		}
-unlock:
-		spin_unlock_irqrestore(&base->lock, flags);
-	} while (ret < 0);
-
-	return ret;
+	for (;;) {
+		int ret = try_to_del_timer_sync(timer);
+		if (ret >= 0)
+			return ret;
+	}
 }
 
 EXPORT_SYMBOL(del_timer_sync);
-- 
cgit v1.2.3-70-g09d2


From 991114c6fa6a21d1fa4d544abe78592352860c82 Mon Sep 17 00:00:00 2001
From: Alexander Viro <aviro@redhat.com>
Date: Thu, 23 Jun 2005 00:09:01 -0700
Subject: [PATCH] fix for prune_icache()/forced final iput() races

Based on analysis and a patch from Russ Weight <rweight@us.ibm.com>

There is a race condition that can occur if an inode is allocated and then
released (using iput) during the ->fill_super functions.  The race
condition is between kswapd and mount.

For most filesystems this can only happen in an error path when kswapd is
running concurrently.  For isofs, however, the error can occur in a more
common code path (which is how the bug was found).

The logic here is "we want final iput() to free inode *now* instead of
letting it sit in cache if fs is going down or had not quite come up".  The
problem is with kswapd seeing such inodes in the middle of being killed and
happily taking over.

The clean solution would be to tell kswapd to leave those inodes alone and
let our final iput deal with them.  I.e.  add a new flag
(I_FORCED_FREEING), set it before write_inode_now() there and make
prune_icache() leave those alone.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/inode.c         | 16 ++++++++++------
 include/linux/fs.h |  1 +
 2 files changed, 11 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/fs/inode.c b/fs/inode.c
index 801fe7f3628..1f9a3a2b89b 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -500,7 +500,7 @@ repeat:
 			continue;
 		if (!test(inode, data))
 			continue;
-		if (inode->i_state & (I_FREEING|I_CLEAR)) {
+		if (inode->i_state & (I_FREEING|I_CLEAR|I_WILL_FREE)) {
 			__wait_on_freeing_inode(inode);
 			goto repeat;
 		}
@@ -525,7 +525,7 @@ repeat:
 			continue;
 		if (inode->i_sb != sb)
 			continue;
-		if (inode->i_state & (I_FREEING|I_CLEAR)) {
+		if (inode->i_state & (I_FREEING|I_CLEAR|I_WILL_FREE)) {
 			__wait_on_freeing_inode(inode);
 			goto repeat;
 		}
@@ -727,7 +727,7 @@ EXPORT_SYMBOL(iunique);
 struct inode *igrab(struct inode *inode)
 {
 	spin_lock(&inode_lock);
-	if (!(inode->i_state & I_FREEING))
+	if (!(inode->i_state & (I_FREEING|I_WILL_FREE)))
 		__iget(inode);
 	else
 		/*
@@ -1024,17 +1024,21 @@ static void generic_forget_inode(struct inode *inode)
 		if (!(inode->i_state & (I_DIRTY|I_LOCK)))
 			list_move(&inode->i_list, &inode_unused);
 		inodes_stat.nr_unused++;
-		spin_unlock(&inode_lock);
-		if (!sb || (sb->s_flags & MS_ACTIVE))
+		if (!sb || (sb->s_flags & MS_ACTIVE)) {
+			spin_unlock(&inode_lock);
 			return;
+		}
+		inode->i_state |= I_WILL_FREE;
+		spin_unlock(&inode_lock);
 		write_inode_now(inode, 1);
 		spin_lock(&inode_lock);
+		inode->i_state &= ~I_WILL_FREE;
 		inodes_stat.nr_unused--;
 		hlist_del_init(&inode->i_hash);
 	}
 	list_del_init(&inode->i_list);
 	list_del_init(&inode->i_sb_list);
-	inode->i_state|=I_FREEING;
+	inode->i_state |= I_FREEING;
 	inodes_stat.nr_inodes--;
 	spin_unlock(&inode_lock);
 	if (inode->i_data.nrpages)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index e5a8db00df2..3622e952e98 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1025,6 +1025,7 @@ struct super_operations {
 #define I_FREEING		16
 #define I_CLEAR			32
 #define I_NEW			64
+#define I_WILL_FREE		128
 
 #define I_DIRTY (I_DIRTY_SYNC | I_DIRTY_DATASYNC | I_DIRTY_PAGES)
 
-- 
cgit v1.2.3-70-g09d2


From 543537bd922692bc978e2e356fcd8bfc9c2ee7d5 Mon Sep 17 00:00:00 2001
From: Paulo Marques <pmarques@grupopie.com>
Date: Thu, 23 Jun 2005 00:09:02 -0700
Subject: [PATCH] create a kstrdup library function

This patch creates a new kstrdup library function and changes the "local"
implementations in several places to use this function.

Most of the changes come from the sound and net subsystems.  The sound part
had already been acknowledged by Takashi Iwai and the net part by David S.
Miller.

I left UML alone for now because I would need more time to read the code
carefully before making changes there.

Signed-off-by: Paulo Marques <pmarques@grupopie.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/md/dm-ioctl.c         | 14 +++-----------
 drivers/parport/probe.c       | 18 +++++-------------
 include/linux/netdevice.h     |  4 ----
 include/linux/string.h        |  2 ++
 include/sound/core.h          |  3 ++-
 mm/slab.c                     | 24 ++++++++++++++++++++++++
 net/core/neighbour.c          |  3 ++-
 net/core/sysctl_net_core.c    | 15 ---------------
 net/ipv4/devinet.c            |  2 +-
 net/ipv6/addrconf.c           |  3 ++-
 net/sunrpc/svcauth_unix.c     | 11 ++---------
 sound/core/info.c             |  3 ++-
 sound/core/info_oss.c         |  3 ++-
 sound/core/memory.c           | 41 ++++++++++++++---------------------------
 sound/core/oss/mixer_oss.c    |  3 ++-
 sound/core/oss/pcm_oss.c      |  3 ++-
 sound/core/sound.c            |  2 +-
 sound/core/timer.c            |  3 ++-
 sound/isa/gus/gus_mem.c       |  7 ++++---
 sound/pci/hda/patch_realtek.c |  2 +-
 sound/synth/emux/emux.c       |  3 ++-
 21 files changed, 75 insertions(+), 94 deletions(-)

(limited to 'include')

diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c
index ee3c869d970..200a0688f71 100644
--- a/drivers/md/dm-ioctl.c
+++ b/drivers/md/dm-ioctl.c
@@ -122,14 +122,6 @@ static struct hash_cell *__get_uuid_cell(const char *str)
 /*-----------------------------------------------------------------
  * Inserting, removing and renaming a device.
  *---------------------------------------------------------------*/
-static inline char *kstrdup(const char *str)
-{
-	char *r = kmalloc(strlen(str) + 1, GFP_KERNEL);
-	if (r)
-		strcpy(r, str);
-	return r;
-}
-
 static struct hash_cell *alloc_cell(const char *name, const char *uuid,
 				    struct mapped_device *md)
 {
@@ -139,7 +131,7 @@ static struct hash_cell *alloc_cell(const char *name, const char *uuid,
 	if (!hc)
 		return NULL;
 
-	hc->name = kstrdup(name);
+	hc->name = kstrdup(name, GFP_KERNEL);
 	if (!hc->name) {
 		kfree(hc);
 		return NULL;
@@ -149,7 +141,7 @@ static struct hash_cell *alloc_cell(const char *name, const char *uuid,
 		hc->uuid = NULL;
 
 	else {
-		hc->uuid = kstrdup(uuid);
+		hc->uuid = kstrdup(uuid, GFP_KERNEL);
 		if (!hc->uuid) {
 			kfree(hc->name);
 			kfree(hc);
@@ -273,7 +265,7 @@ static int dm_hash_rename(const char *old, const char *new)
 	/*
 	 * duplicate new.
 	 */
-	new_name = kstrdup(new);
+	new_name = kstrdup(new, GFP_KERNEL);
 	if (!new_name)
 		return -ENOMEM;
 
diff --git a/drivers/parport/probe.c b/drivers/parport/probe.c
index c94963145e1..6e6f42d01e6 100644
--- a/drivers/parport/probe.c
+++ b/drivers/parport/probe.c
@@ -48,14 +48,6 @@ static void pretty_print(struct parport *port, int device)
 	printk("\n");
 }
 
-static char *strdup(char *str)
-{
-	int n = strlen(str)+1;
-	char *s = kmalloc(n, GFP_KERNEL);
-	if (!s) return NULL;
-	return strcpy(s, str);
-}
-
 static void parse_data(struct parport *port, int device, char *str)
 {
 	char *txt = kmalloc(strlen(str)+1, GFP_KERNEL);
@@ -88,16 +80,16 @@ static void parse_data(struct parport *port, int device, char *str)
 			if (!strcmp(p, "MFG") || !strcmp(p, "MANUFACTURER")) {
 				if (info->mfr)
 					kfree (info->mfr);
-				info->mfr = strdup(sep);
+				info->mfr = kstrdup(sep, GFP_KERNEL);
 			} else if (!strcmp(p, "MDL") || !strcmp(p, "MODEL")) {
 				if (info->model)
 					kfree (info->model);
-				info->model = strdup(sep);
+				info->model = kstrdup(sep, GFP_KERNEL);
 			} else if (!strcmp(p, "CLS") || !strcmp(p, "CLASS")) {
 				int i;
 				if (info->class_name)
 					kfree (info->class_name);
-				info->class_name = strdup(sep);
+				info->class_name = kstrdup(sep, GFP_KERNEL);
 				for (u = sep; *u; u++)
 					*u = toupper(*u);
 				for (i = 0; classes[i].token; i++) {
@@ -112,7 +104,7 @@ static void parse_data(struct parport *port, int device, char *str)
 				   !strcmp(p, "COMMAND SET")) {
 				if (info->cmdset)
 					kfree (info->cmdset);
-				info->cmdset = strdup(sep);
+				info->cmdset = kstrdup(sep, GFP_KERNEL);
 				/* if it speaks printer language, it's
 				   probably a printer */
 				if (strstr(sep, "PJL") || strstr(sep, "PCL"))
@@ -120,7 +112,7 @@ static void parse_data(struct parport *port, int device, char *str)
 			} else if (!strcmp(p, "DES") || !strcmp(p, "DESCRIPTION")) {
 				if (info->description)
 					kfree (info->description);
-				info->description = strdup(sep);
+				info->description = kstrdup(sep, GFP_KERNEL);
 			}
 		}
 	rock_on:
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index d6afd440cf7..d89816ad642 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -925,10 +925,6 @@ extern int skb_checksum_help(struct sk_buff *skb, int inward);
 extern void		net_enable_timestamp(void);
 extern void		net_disable_timestamp(void);
 
-#ifdef CONFIG_SYSCTL
-extern char *net_sysctl_strdup(const char *s);
-#endif
-
 #endif /* __KERNEL__ */
 
 #endif	/* _LINUX_DEV_H */
diff --git a/include/linux/string.h b/include/linux/string.h
index b9fc5946995..93994c61309 100644
--- a/include/linux/string.h
+++ b/include/linux/string.h
@@ -88,6 +88,8 @@ extern int memcmp(const void *,const void *,__kernel_size_t);
 extern void * memchr(const void *,int,__kernel_size_t);
 #endif
 
+extern char *kstrdup(const char *s, int gfp);
+
 #ifdef __cplusplus
 }
 #endif
diff --git a/include/sound/core.h b/include/sound/core.h
index 9117c23e3a0..f8c4ef0aa35 100644
--- a/include/sound/core.h
+++ b/include/sound/core.h
@@ -292,6 +292,7 @@ void *snd_hidden_kcalloc(size_t n, size_t size, int flags);
 void snd_hidden_kfree(const void *obj);
 void *snd_hidden_vmalloc(unsigned long size);
 void snd_hidden_vfree(void *obj);
+char *snd_hidden_kstrdup(const char *s, int flags);
 #define kmalloc(size, flags) snd_hidden_kmalloc(size, flags)
 #define kcalloc(n, size, flags) snd_hidden_kcalloc(n, size, flags)
 #define kfree(obj) snd_hidden_kfree(obj)
@@ -301,6 +302,7 @@ void snd_hidden_vfree(void *obj);
 #define vmalloc_nocheck(size) snd_wrapper_vmalloc(size)
 #define kfree_nocheck(obj) snd_wrapper_kfree(obj)
 #define vfree_nocheck(obj) snd_wrapper_vfree(obj)
+#define kstrdup(s, flags)  snd_hidden_kstrdup(s, flags)
 #else
 #define snd_memory_init() /*NOP*/
 #define snd_memory_done() /*NOP*/
@@ -311,7 +313,6 @@ void snd_hidden_vfree(void *obj);
 #define kfree_nocheck(obj) kfree(obj)
 #define vfree_nocheck(obj) vfree(obj)
 #endif
-char *snd_kmalloc_strdup(const char *string, int flags);
 int copy_to_user_fromio(void __user *dst, const volatile void __iomem *src, size_t count);
 int copy_from_user_toio(volatile void __iomem *dst, const void __user *src, size_t count);
 
diff --git a/mm/slab.c b/mm/slab.c
index 93cbbbb39f4..122d031baab 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -92,6 +92,7 @@
 #include	<linux/sysctl.h>
 #include	<linux/module.h>
 #include	<linux/rcupdate.h>
+#include	<linux/string.h>
 
 #include	<asm/uaccess.h>
 #include	<asm/cacheflush.h>
@@ -3082,3 +3083,26 @@ unsigned int ksize(const void *objp)
 
 	return size;
 }
+
+
+/*
+ * kstrdup - allocate space for and copy an existing string
+ *
+ * @s: the string to duplicate
+ * @gfp: the GFP mask used in the kmalloc() call when allocating memory
+ */
+char *kstrdup(const char *s, int gfp)
+{
+	size_t len;
+	char *buf;
+
+	if (!s)
+		return NULL;
+
+	len = strlen(s) + 1;
+	buf = kmalloc(len, gfp);
+	if (buf)
+		memcpy(buf, s, len);
+	return buf;
+}
+EXPORT_SYMBOL(kstrdup);
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index f6bdcad47da..851eb927ed9 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -32,6 +32,7 @@
 #include <net/sock.h>
 #include <linux/rtnetlink.h>
 #include <linux/random.h>
+#include <linux/string.h>
 
 #define NEIGH_DEBUG 1
 
@@ -2592,7 +2593,7 @@ int neigh_sysctl_register(struct net_device *dev, struct neigh_parms *p,
 		t->neigh_vars[17].extra1 = dev;
 	}
 
-	dev_name = net_sysctl_strdup(dev_name_source);
+	dev_name = kstrdup(dev_name_source, GFP_KERNEL);
 	if (!dev_name) {
 		err = -ENOBUFS;
 		goto free;
diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c
index c8be646cb19..880a8881521 100644
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -35,19 +35,6 @@ extern int sysctl_somaxconn;
 extern char sysctl_divert_version[];
 #endif /* CONFIG_NET_DIVERT */
 
-/*
- * This strdup() is used for creating copies of network 
- * device names to be handed over to sysctl.
- */
- 
-char *net_sysctl_strdup(const char *s)
-{
-	char *rv = kmalloc(strlen(s)+1, GFP_KERNEL);
-	if (rv)
-		strcpy(rv, s);
-	return rv;
-}
-
 ctl_table core_table[] = {
 #ifdef CONFIG_NET
 	{
@@ -177,6 +164,4 @@ ctl_table core_table[] = {
 	{ .ctl_name = 0 }
 };
 
-EXPORT_SYMBOL(net_sysctl_strdup);
-
 #endif
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 650dcb12d9a..d8a10e3dd77 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1471,7 +1471,7 @@ static void devinet_sysctl_register(struct in_device *in_dev,
 	 * by sysctl and we wouldn't want anyone to change it under our feet
 	 * (see SIOCSIFNAME).
 	 */	
-	dev_name = net_sysctl_strdup(dev_name);
+	dev_name = kstrdup(dev_name, GFP_KERNEL);
 	if (!dev_name)
 	    goto free;
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 14f5c53235f..a54d4ef3fd3 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -57,6 +57,7 @@
 #endif
 #include <linux/delay.h>
 #include <linux/notifier.h>
+#include <linux/string.h>
 
 #include <net/sock.h>
 #include <net/snmp.h>
@@ -3437,7 +3438,7 @@ static void addrconf_sysctl_register(struct inet6_dev *idev, struct ipv6_devconf
 	 * by sysctl and we wouldn't want anyone to change it under our feet
 	 * (see SIOCSIFNAME).
 	 */	
-	dev_name = net_sysctl_strdup(dev_name);
+	dev_name = kstrdup(dev_name, GFP_KERNEL);
 	if (!dev_name)
 	    goto free;
 
diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c
index 2b99b4028d3..d6baf6fdf8a 100644
--- a/net/sunrpc/svcauth_unix.c
+++ b/net/sunrpc/svcauth_unix.c
@@ -8,6 +8,7 @@
 #include <linux/err.h>
 #include <linux/seq_file.h>
 #include <linux/hash.h>
+#include <linux/string.h>
 
 #define RPCDBG_FACILITY	RPCDBG_AUTH
 
@@ -20,14 +21,6 @@
  */
 
 
-static char *strdup(char *s)
-{
-	char *rv = kmalloc(strlen(s)+1, GFP_KERNEL);
-	if (rv)
-		strcpy(rv, s);
-	return rv;
-}
-
 struct unix_domain {
 	struct auth_domain	h;
 	int	addr_changes;
@@ -55,7 +48,7 @@ struct auth_domain *unix_domain_find(char *name)
 	if (new == NULL)
 		return NULL;
 	cache_init(&new->h.h);
-	new->h.name = strdup(name);
+	new->h.name = kstrdup(name, GFP_KERNEL);
 	new->h.flavour = RPC_AUTH_UNIX;
 	new->addr_changes = 0;
 	new->h.h.expiry_time = NEVER;
diff --git a/sound/core/info.c b/sound/core/info.c
index 31faffe01cb..5e122bbe7c9 100644
--- a/sound/core/info.c
+++ b/sound/core/info.c
@@ -24,6 +24,7 @@
 #include <linux/vmalloc.h>
 #include <linux/time.h>
 #include <linux/smp_lock.h>
+#include <linux/string.h>
 #include <sound/core.h>
 #include <sound/minors.h>
 #include <sound/info.h>
@@ -754,7 +755,7 @@ static snd_info_entry_t *snd_info_create_entry(const char *name)
 	entry = kcalloc(1, sizeof(*entry), GFP_KERNEL);
 	if (entry == NULL)
 		return NULL;
-	entry->name = snd_kmalloc_strdup(name, GFP_KERNEL);
+	entry->name = kstrdup(name, GFP_KERNEL);
 	if (entry->name == NULL) {
 		kfree(entry);
 		return NULL;
diff --git a/sound/core/info_oss.c b/sound/core/info_oss.c
index f9e4ce44345..12107968d40 100644
--- a/sound/core/info_oss.c
+++ b/sound/core/info_oss.c
@@ -22,6 +22,7 @@
 #include <sound/driver.h>
 #include <linux/slab.h>
 #include <linux/time.h>
+#include <linux/string.h>
 #include <sound/core.h>
 #include <sound/minors.h>
 #include <sound/info.h>
@@ -51,7 +52,7 @@ int snd_oss_info_register(int dev, int num, char *string)
 			x = NULL;
 		}
 	} else {
-		x = snd_kmalloc_strdup(string, GFP_KERNEL);
+		x = kstrdup(string, GFP_KERNEL);
 		if (x == NULL) {
 			up(&strings);
 			return -ENOMEM;
diff --git a/sound/core/memory.c b/sound/core/memory.c
index 20860fec936..c1fb28e8433 100644
--- a/sound/core/memory.c
+++ b/sound/core/memory.c
@@ -184,6 +184,20 @@ void snd_hidden_vfree(void *obj)
 	snd_wrapper_vfree(obj);
 }
 
+char *snd_hidden_kstrdup(const char *s, int flags)
+{
+	int len;
+	char *buf;
+
+	if (!s) return NULL;
+
+	len = strlen(s) + 1;
+	buf = _snd_kmalloc(len, flags);
+	if (buf)
+		memcpy(buf, s, len);
+	return buf;
+}
+
 static void snd_memory_info_read(snd_info_entry_t *entry, snd_info_buffer_t * buffer)
 {
 	snd_iprintf(buffer, "kmalloc: %li bytes\n", snd_alloc_kmalloc);
@@ -214,35 +228,8 @@ int __exit snd_memory_info_done(void)
 	return 0;
 }
 
-#else
-
-#define _snd_kmalloc kmalloc
-
 #endif /* CONFIG_SND_DEBUG_MEMORY */
 
-/**
- * snd_kmalloc_strdup - copy the string
- * @string: the original string
- * @flags: allocation conditions, GFP_XXX
- *
- * Allocates a memory chunk via kmalloc() and copies the string to it.
- *
- * Returns the pointer, or NULL if no enoguh memory.
- */
-char *snd_kmalloc_strdup(const char *string, int flags)
-{
-	size_t len;
-	char *ptr;
-
-	if (!string)
-		return NULL;
-	len = strlen(string) + 1;
-	ptr = _snd_kmalloc(len, flags);
-	if (ptr)
-		memcpy(ptr, string, len);
-	return ptr;
-}
-
 /**
  * copy_to_user_fromio - copy data from mmio-space to user-space
  * @dst: the destination pointer on user-space
diff --git a/sound/core/oss/mixer_oss.c b/sound/core/oss/mixer_oss.c
index 98ed9a9f0da..98fc0766f88 100644
--- a/sound/core/oss/mixer_oss.c
+++ b/sound/core/oss/mixer_oss.c
@@ -24,6 +24,7 @@
 #include <linux/smp_lock.h>
 #include <linux/slab.h>
 #include <linux/time.h>
+#include <linux/string.h>
 #include <sound/core.h>
 #include <sound/minors.h>
 #include <sound/control.h>
@@ -1137,7 +1138,7 @@ static void snd_mixer_oss_proc_write(snd_info_entry_t *entry,
 			goto __unlock;
 		}
 		tbl->oss_id = ch;
-		tbl->name = snd_kmalloc_strdup(str, GFP_KERNEL);
+		tbl->name = kstrdup(str, GFP_KERNEL);
 		if (! tbl->name) {
 			kfree(tbl);
 			goto __unlock;
diff --git a/sound/core/oss/pcm_oss.c b/sound/core/oss/pcm_oss.c
index cab30977e7c..de7444c586f 100644
--- a/sound/core/oss/pcm_oss.c
+++ b/sound/core/oss/pcm_oss.c
@@ -33,6 +33,7 @@
 #include <linux/time.h>
 #include <linux/vmalloc.h>
 #include <linux/moduleparam.h>
+#include <linux/string.h>
 #include <sound/core.h>
 #include <sound/minors.h>
 #include <sound/pcm.h>
@@ -2360,7 +2361,7 @@ static void snd_pcm_oss_proc_write(snd_info_entry_t *entry,
 					for (setup1 = pstr->oss.setup_list; setup1->next; setup1 = setup1->next);
 					setup1->next = setup;
 				}
-				template.task_name = snd_kmalloc_strdup(task_name, GFP_KERNEL);
+				template.task_name = kstrdup(task_name, GFP_KERNEL);
 			} else {
 				buffer->error = -ENOMEM;
 			}
diff --git a/sound/core/sound.c b/sound/core/sound.c
index 0815fadeb3e..7612884f530 100644
--- a/sound/core/sound.c
+++ b/sound/core/sound.c
@@ -399,8 +399,8 @@ EXPORT_SYMBOL(snd_hidden_kcalloc);
 EXPORT_SYMBOL(snd_hidden_kfree);
 EXPORT_SYMBOL(snd_hidden_vmalloc);
 EXPORT_SYMBOL(snd_hidden_vfree);
+EXPORT_SYMBOL(snd_hidden_kstrdup);
 #endif
-EXPORT_SYMBOL(snd_kmalloc_strdup);
 EXPORT_SYMBOL(copy_to_user_fromio);
 EXPORT_SYMBOL(copy_from_user_toio);
   /* init.c */
diff --git a/sound/core/timer.c b/sound/core/timer.c
index b498e5482d7..cfaccd415b3 100644
--- a/sound/core/timer.c
+++ b/sound/core/timer.c
@@ -26,6 +26,7 @@
 #include <linux/slab.h>
 #include <linux/time.h>
 #include <linux/moduleparam.h>
+#include <linux/string.h>
 #include <sound/core.h>
 #include <sound/timer.h>
 #include <sound/control.h>
@@ -100,7 +101,7 @@ static snd_timer_instance_t *snd_timer_instance_new(char *owner, snd_timer_t *ti
 	timeri = kcalloc(1, sizeof(*timeri), GFP_KERNEL);
 	if (timeri == NULL)
 		return NULL;
-	timeri->owner = snd_kmalloc_strdup(owner, GFP_KERNEL);
+	timeri->owner = kstrdup(owner, GFP_KERNEL);
 	if (! timeri->owner) {
 		kfree(timeri);
 		return NULL;
diff --git a/sound/isa/gus/gus_mem.c b/sound/isa/gus/gus_mem.c
index 609838e8ef6..5eb766dd564 100644
--- a/sound/isa/gus/gus_mem.c
+++ b/sound/isa/gus/gus_mem.c
@@ -21,6 +21,7 @@
 
 #include <sound/driver.h>
 #include <linux/slab.h>
+#include <linux/string.h>
 #include <sound/core.h>
 #include <sound/gus.h>
 #include <sound/info.h>
@@ -213,7 +214,7 @@ snd_gf1_mem_block_t *snd_gf1_mem_alloc(snd_gf1_mem_t * alloc, int owner,
 	if (share_id != NULL)
 		memcpy(&block.share_id, share_id, sizeof(block.share_id));
 	block.owner = owner;
-	block.name = snd_kmalloc_strdup(name, GFP_KERNEL);
+	block.name = kstrdup(name, GFP_KERNEL);
 	nblock = snd_gf1_mem_xalloc(alloc, &block);
 	snd_gf1_mem_lock(alloc, 1);
 	return nblock;
@@ -253,13 +254,13 @@ int snd_gf1_mem_init(snd_gus_card_t * gus)
 	if (gus->gf1.enh_mode) {
 		block.ptr = 0;
 		block.size = 1024;
-		block.name = snd_kmalloc_strdup("InterWave LFOs", GFP_KERNEL);
+		block.name = kstrdup("InterWave LFOs", GFP_KERNEL);
 		if (snd_gf1_mem_xalloc(alloc, &block) == NULL)
 			return -ENOMEM;
 	}
 	block.ptr = gus->gf1.default_voice_address;
 	block.size = 4;
-	block.name = snd_kmalloc_strdup("Voice default (NULL's)", GFP_KERNEL);
+	block.name = kstrdup("Voice default (NULL's)", GFP_KERNEL);
 	if (snd_gf1_mem_xalloc(alloc, &block) == NULL)
 		return -ENOMEM;
 #ifdef CONFIG_SND_DEBUG
diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index 9edd558d6bd..bab89843d85 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -1781,7 +1781,7 @@ static int add_control(struct alc_spec *spec, int type, const char *name, unsign
 
 	knew = &spec->kctl_alloc[spec->num_kctl_used];
 	*knew = alc880_control_templates[type];
-	knew->name = snd_kmalloc_strdup(name, GFP_KERNEL);
+	knew->name = kstrdup(name, GFP_KERNEL);
 	if (! knew->name)
 		return -ENOMEM;
 	knew->private_value = val;
diff --git a/sound/synth/emux/emux.c b/sound/synth/emux/emux.c
index 16f3b461627..60d0b2c6669 100644
--- a/sound/synth/emux/emux.c
+++ b/sound/synth/emux/emux.c
@@ -22,6 +22,7 @@
 #include <linux/wait.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
+#include <linux/string.h>
 #include <sound/core.h>
 #include <sound/emux_synth.h>
 #include <linux/init.h>
@@ -76,7 +77,7 @@ int snd_emux_register(snd_emux_t *emu, snd_card_t *card, int index, char *name)
 	snd_assert(name != NULL, return -EINVAL);
 
 	emu->card = card;
-	emu->name = snd_kmalloc_strdup(name, GFP_KERNEL);
+	emu->name = kstrdup(name, GFP_KERNEL);
 	emu->voices = kcalloc(emu->max_voices, sizeof(snd_emux_voice_t), GFP_KERNEL);
 	if (emu->voices == NULL)
 		return -ENOMEM;
-- 
cgit v1.2.3-70-g09d2


From 35a82d1a53e1a9ad54efafcc940f9335beaed5c3 Mon Sep 17 00:00:00 2001
From: Nick Piggin <nickpiggin@yahoo.com.au>
Date: Thu, 23 Jun 2005 00:09:06 -0700
Subject: [PATCH] optimise loop driver a bit

Looks like locking can be optimised quite a lot.  Increase lock widths
slightly so lo_lock is taken fewer times per request.  Also it was quite
trivial to cover lo_pending with that lock, and remove the atomic
requirement.  This also makes memory ordering explicitly correct, which is
nice (not that I particularly saw any mem ordering bugs).

Test was reading 4 250MB files in parallel on ext2-on-tmpfs filesystem (1K
block size, 4K page size).  System is 2 socket Xeon with HT (4 thread).

intel:/home/npiggin# umount /dev/loop0 ; mount /dev/loop0 /mnt/loop ; /usr/bin/time ./mtloop.sh

Before:
0.24user 5.51system 0:02.84elapsed 202%CPU (0avgtext+0avgdata 0maxresident)k
0.19user 5.52system 0:02.88elapsed 198%CPU (0avgtext+0avgdata 0maxresident)k
0.19user 5.57system 0:02.89elapsed 198%CPU (0avgtext+0avgdata 0maxresident)k
0.22user 5.51system 0:02.90elapsed 197%CPU (0avgtext+0avgdata 0maxresident)k
0.19user 5.44system 0:02.91elapsed 193%CPU (0avgtext+0avgdata 0maxresident)k

After:
0.07user 2.34system 0:01.68elapsed 143%CPU (0avgtext+0avgdata 0maxresident)k
0.06user 2.37system 0:01.68elapsed 144%CPU (0avgtext+0avgdata 0maxresident)k
0.06user 2.39system 0:01.68elapsed 145%CPU (0avgtext+0avgdata 0maxresident)k
0.06user 2.36system 0:01.68elapsed 144%CPU (0avgtext+0avgdata 0maxresident)k
0.06user 2.42system 0:01.68elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k

Signed-off-by: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/loop.c | 81 ++++++++++++++++++++++++----------------------------
 include/linux/loop.h |  2 +-
 2 files changed, 39 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 6f011d0d8e9..b35e08876dd 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -472,17 +472,11 @@ static int do_bio_filebacked(struct loop_device *lo, struct bio *bio)
  */
 static void loop_add_bio(struct loop_device *lo, struct bio *bio)
 {
-	unsigned long flags;
-
-	spin_lock_irqsave(&lo->lo_lock, flags);
 	if (lo->lo_biotail) {
 		lo->lo_biotail->bi_next = bio;
 		lo->lo_biotail = bio;
 	} else
 		lo->lo_bio = lo->lo_biotail = bio;
-	spin_unlock_irqrestore(&lo->lo_lock, flags);
-
-	up(&lo->lo_bh_mutex);
 }
 
 /*
@@ -492,14 +486,12 @@ static struct bio *loop_get_bio(struct loop_device *lo)
 {
 	struct bio *bio;
 
-	spin_lock_irq(&lo->lo_lock);
 	if ((bio = lo->lo_bio)) {
 		if (bio == lo->lo_biotail)
 			lo->lo_biotail = NULL;
 		lo->lo_bio = bio->bi_next;
 		bio->bi_next = NULL;
 	}
-	spin_unlock_irq(&lo->lo_lock);
 
 	return bio;
 }
@@ -509,35 +501,28 @@ static int loop_make_request(request_queue_t *q, struct bio *old_bio)
 	struct loop_device *lo = q->queuedata;
 	int rw = bio_rw(old_bio);
 
-	if (!lo)
-		goto out;
+	if (rw == READA)
+		rw = READ;
+
+	BUG_ON(!lo || (rw != READ && rw != WRITE));
 
 	spin_lock_irq(&lo->lo_lock);
 	if (lo->lo_state != Lo_bound)
-		goto inactive;
-	atomic_inc(&lo->lo_pending);
-	spin_unlock_irq(&lo->lo_lock);
-
-	if (rw == WRITE) {
-		if (lo->lo_flags & LO_FLAGS_READ_ONLY)
-			goto err;
-	} else if (rw == READA) {
-		rw = READ;
-	} else if (rw != READ) {
-		printk(KERN_ERR "loop: unknown command (%x)\n", rw);
-		goto err;
-	}
+		goto out;
+	if (unlikely(rw == WRITE && (lo->lo_flags & LO_FLAGS_READ_ONLY)))
+		goto out;
+	lo->lo_pending++;
 	loop_add_bio(lo, old_bio);
+	spin_unlock_irq(&lo->lo_lock);
+	up(&lo->lo_bh_mutex);
 	return 0;
-err:
-	if (atomic_dec_and_test(&lo->lo_pending))
-		up(&lo->lo_bh_mutex);
+
 out:
+	if (lo->lo_pending == 0)
+		up(&lo->lo_bh_mutex);
+	spin_unlock_irq(&lo->lo_lock);
 	bio_io_error(old_bio, old_bio->bi_size);
 	return 0;
-inactive:
-	spin_unlock_irq(&lo->lo_lock);
-	goto out;
 }
 
 /*
@@ -560,13 +545,11 @@ static void do_loop_switch(struct loop_device *, struct switch_request *);
 
 static inline void loop_handle_bio(struct loop_device *lo, struct bio *bio)
 {
-	int ret;
-
 	if (unlikely(!bio->bi_bdev)) {
 		do_loop_switch(lo, bio->bi_private);
 		bio_put(bio);
 	} else {
-		ret = do_bio_filebacked(lo, bio);
+		int ret = do_bio_filebacked(lo, bio);
 		bio_endio(bio, bio->bi_size, ret);
 	}
 }
@@ -594,7 +577,7 @@ static int loop_thread(void *data)
 	set_user_nice(current, -20);
 
 	lo->lo_state = Lo_bound;
-	atomic_inc(&lo->lo_pending);
+	lo->lo_pending = 1;
 
 	/*
 	 * up sem, we are running
@@ -602,26 +585,37 @@ static int loop_thread(void *data)
 	up(&lo->lo_sem);
 
 	for (;;) {
-		down_interruptible(&lo->lo_bh_mutex);
+		int pending;
+
 		/*
-		 * could be upped because of tear-down, not because of
-		 * pending work
+		 * interruptible just to not contribute to load avg
 		 */
-		if (!atomic_read(&lo->lo_pending))
+		if (down_interruptible(&lo->lo_bh_mutex))
+			continue;
+
+		spin_lock_irq(&lo->lo_lock);
+
+		/*
+		 * could be upped because of tear-down, not pending work
+		 */
+		if (unlikely(!lo->lo_pending)) {
+			spin_unlock_irq(&lo->lo_lock);
 			break;
+		}
 
 		bio = loop_get_bio(lo);
-		if (!bio) {
-			printk("loop: missing bio\n");
-			continue;
-		}
+		lo->lo_pending--;
+		pending = lo->lo_pending;
+		spin_unlock_irq(&lo->lo_lock);
+
+		BUG_ON(!bio);
 		loop_handle_bio(lo, bio);
 
 		/*
 		 * upped both for pending work and tear-down, lo_pending
 		 * will hit zero then
 		 */
-		if (atomic_dec_and_test(&lo->lo_pending))
+		if (unlikely(!pending))
 			break;
 	}
 
@@ -900,7 +894,8 @@ static int loop_clr_fd(struct loop_device *lo, struct block_device *bdev)
 
 	spin_lock_irq(&lo->lo_lock);
 	lo->lo_state = Lo_rundown;
-	if (atomic_dec_and_test(&lo->lo_pending))
+	lo->lo_pending--;
+	if (!lo->lo_pending)
 		up(&lo->lo_bh_mutex);
 	spin_unlock_irq(&lo->lo_lock);
 
diff --git a/include/linux/loop.h b/include/linux/loop.h
index 8220d9c9da0..53fa5159544 100644
--- a/include/linux/loop.h
+++ b/include/linux/loop.h
@@ -61,7 +61,7 @@ struct loop_device {
 	struct semaphore	lo_sem;
 	struct semaphore	lo_ctl_mutex;
 	struct semaphore	lo_bh_mutex;
-	atomic_t		lo_pending;
+	int			lo_pending;
 
 	request_queue_t		*lo_queue;
 };
-- 
cgit v1.2.3-70-g09d2


From dcd497f99a1ef29a7c5e76142965be77e9dacabd Mon Sep 17 00:00:00 2001
From: Jesper Juhl <juhl-lkml@dif.dk>
Date: Thu, 23 Jun 2005 00:09:07 -0700
Subject: [PATCH] streamline preempt_count type across archs

The preempt_count member of struct thread_info is currently either defined
as int, unsigned int or __s32 depending on arch.  This patch makes the type
of preempt_count an int on all archs.

Having preempt_count be an unsigned type prevents the catching of
preempt_count < 0 bugs, and using int on some archs and __s32 on others is
not exactely "neat" - much nicer when it's just int all over.

A previous version of this patch was already ACK'ed by Robert Love, and the
only change in this version of the patch compared to the one he ACK'ed is
that this one also makes sure the preempt_count member is consistently
commented.

Signed-off-by: Jesper Juhl <juhl-lkml@dif.dk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-arm/thread_info.h       | 2 +-
 include/asm-arm26/thread_info.h     | 2 +-
 include/asm-cris/thread_info.h      | 2 +-
 include/asm-frv/thread_info.h       | 2 +-
 include/asm-h8300/thread_info.h     | 2 +-
 include/asm-i386/thread_info.h      | 2 +-
 include/asm-ia64/thread_info.h      | 2 +-
 include/asm-m32r/thread_info.h      | 2 +-
 include/asm-m68k/thread_info.h      | 2 +-
 include/asm-m68knommu/thread_info.h | 2 +-
 include/asm-mips/thread_info.h      | 2 +-
 include/asm-parisc/thread_info.h    | 2 +-
 include/asm-ppc/thread_info.h       | 3 ++-
 include/asm-ppc64/thread_info.h     | 2 +-
 include/asm-s390/thread_info.h      | 2 +-
 include/asm-sh/thread_info.h        | 2 +-
 include/asm-sh64/thread_info.h      | 2 +-
 include/asm-sparc/thread_info.h     | 4 ++--
 include/asm-sparc64/thread_info.h   | 2 +-
 include/asm-um/thread_info.h        | 2 +-
 include/asm-v850/thread_info.h      | 3 ++-
 include/asm-x86_64/thread_info.h    | 2 +-
 22 files changed, 25 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/thread_info.h b/include/asm-arm/thread_info.h
index 66c585c50cf..8252a4cd860 100644
--- a/include/asm-arm/thread_info.h
+++ b/include/asm-arm/thread_info.h
@@ -49,7 +49,7 @@ struct cpu_context_save {
  */
 struct thread_info {
 	unsigned long		flags;		/* low level flags */
-	__s32			preempt_count;	/* 0 => preemptable, <0 => bug */
+	int			preempt_count;	/* 0 => preemptable, <0 => bug */
 	mm_segment_t		addr_limit;	/* address limit */
 	struct task_struct	*task;		/* main task structure */
 	struct exec_domain	*exec_domain;	/* execution domain */
diff --git a/include/asm-arm26/thread_info.h b/include/asm-arm26/thread_info.h
index 50f41b50268..aff3e5699c6 100644
--- a/include/asm-arm26/thread_info.h
+++ b/include/asm-arm26/thread_info.h
@@ -44,7 +44,7 @@ struct cpu_context_save {
  */
 struct thread_info {
 	unsigned long		flags;		/* low level flags */
-	__s32			preempt_count;	/* 0 => preemptable, <0 => bug */
+	int			preempt_count;	/* 0 => preemptable, <0 => bug */
 	mm_segment_t		addr_limit;	/* address limit */
 	struct task_struct	*task;		/* main task structure */
 	struct exec_domain      *exec_domain;   /* execution domain */
diff --git a/include/asm-cris/thread_info.h b/include/asm-cris/thread_info.h
index 53193feb082..5ba4b7865cc 100644
--- a/include/asm-cris/thread_info.h
+++ b/include/asm-cris/thread_info.h
@@ -31,7 +31,7 @@ struct thread_info {
 	struct exec_domain	*exec_domain;	/* execution domain */
 	unsigned long		flags;		/* low level flags */
 	__u32			cpu;		/* current CPU */
-	__s32			preempt_count; /* 0 => preemptable, <0 => BUG */
+	int			preempt_count;	/* 0 => preemptable, <0 => BUG */
 
 	mm_segment_t		addr_limit;	/* thread address space:
 					 	   0-0xBFFFFFFF for user-thead
diff --git a/include/asm-frv/thread_info.h b/include/asm-frv/thread_info.h
index b80a97f50af..c8cba7836f0 100644
--- a/include/asm-frv/thread_info.h
+++ b/include/asm-frv/thread_info.h
@@ -33,7 +33,7 @@ struct thread_info {
 	unsigned long		flags;		/* low level flags */
 	unsigned long		status;		/* thread-synchronous flags */
 	__u32			cpu;		/* current CPU */
-	__s32			preempt_count;	/* 0 => preemptable, <0 => BUG */
+	int			preempt_count;	/* 0 => preemptable, <0 => BUG */
 
 	mm_segment_t		addr_limit;	/* thread address space:
 					 	   0-0xBFFFFFFF for user-thead
diff --git a/include/asm-h8300/thread_info.h b/include/asm-h8300/thread_info.h
index b07c9344776..bfcc755c3bb 100644
--- a/include/asm-h8300/thread_info.h
+++ b/include/asm-h8300/thread_info.h
@@ -23,7 +23,7 @@ struct thread_info {
 	struct exec_domain *exec_domain;	/* execution domain */
 	unsigned long	   flags;		/* low level flags */
 	int		   cpu;			/* cpu we're on */
-	int		   preempt_count;	/* 0 => preemptable, <0 => BUG*/
+	int		   preempt_count;	/* 0 => preemptable, <0 => BUG */
 	struct restart_block restart_block;
 };
 
diff --git a/include/asm-i386/thread_info.h b/include/asm-i386/thread_info.h
index 2cd57271801..95add81237e 100644
--- a/include/asm-i386/thread_info.h
+++ b/include/asm-i386/thread_info.h
@@ -31,7 +31,7 @@ struct thread_info {
 	unsigned long		flags;		/* low level flags */
 	unsigned long		status;		/* thread-synchronous flags */
 	__u32			cpu;		/* current CPU */
-	__s32			preempt_count; /* 0 => preemptable, <0 => BUG */
+	int			preempt_count;	/* 0 => preemptable, <0 => BUG */
 
 
 	mm_segment_t		addr_limit;	/* thread address space:
diff --git a/include/asm-ia64/thread_info.h b/include/asm-ia64/thread_info.h
index 8d5b7e77028..7dc8951708a 100644
--- a/include/asm-ia64/thread_info.h
+++ b/include/asm-ia64/thread_info.h
@@ -25,7 +25,7 @@ struct thread_info {
 	__u32 flags;			/* thread_info flags (see TIF_*) */
 	__u32 cpu;			/* current CPU */
 	mm_segment_t addr_limit;	/* user-level address space limit */
-	__s32 preempt_count;		/* 0=premptable, <0=BUG; will also serve as bh-counter */
+	int preempt_count;		/* 0=premptable, <0=BUG; will also serve as bh-counter */
 	struct restart_block restart_block;
 	struct {
 		int signo;
diff --git a/include/asm-m32r/thread_info.h b/include/asm-m32r/thread_info.h
index 9f3a0fcf6e2..7a6be7727a9 100644
--- a/include/asm-m32r/thread_info.h
+++ b/include/asm-m32r/thread_info.h
@@ -28,7 +28,7 @@ struct thread_info {
 	unsigned long		flags;		/* low level flags */
 	unsigned long		status;		/* thread-synchronous flags */
 	__u32			cpu;		/* current CPU */
-	__s32			preempt_count; /* 0 => preemptable, <0 => BUG */
+	int			preempt_count;	/* 0 => preemptable, <0 => BUG */
 
 	mm_segment_t		addr_limit;	/* thread address space:
 					 	   0-0xBFFFFFFF for user-thread
diff --git a/include/asm-m68k/thread_info.h b/include/asm-m68k/thread_info.h
index 5f58939c59d..2aed24f6fd2 100644
--- a/include/asm-m68k/thread_info.h
+++ b/include/asm-m68k/thread_info.h
@@ -8,7 +8,7 @@
 struct thread_info {
 	struct task_struct	*task;		/* main task structure */
 	struct exec_domain	*exec_domain;	/* execution domain */
-	__s32			preempt_count; /* 0 => preemptable, <0 => BUG */
+	int			preempt_count;	/* 0 => preemptable, <0 => BUG */
 	__u32 cpu; /* should always be 0 on m68k */
 	struct restart_block    restart_block;
 
diff --git a/include/asm-m68knommu/thread_info.h b/include/asm-m68knommu/thread_info.h
index c8153b7c1f5..7b9a3fa3af5 100644
--- a/include/asm-m68knommu/thread_info.h
+++ b/include/asm-m68knommu/thread_info.h
@@ -36,7 +36,7 @@ struct thread_info {
 	struct exec_domain *exec_domain;	/* execution domain */
 	unsigned long	   flags;		/* low level flags */
 	int		   cpu;			/* cpu we're on */
-	int		   preempt_count;	/* 0 => preemptable, <0 => BUG*/
+	int		   preempt_count;	/* 0 => preemptable, <0 => BUG */
 	struct restart_block restart_block;
 };
 
diff --git a/include/asm-mips/thread_info.h b/include/asm-mips/thread_info.h
index 768900305e2..42fcd6f2c20 100644
--- a/include/asm-mips/thread_info.h
+++ b/include/asm-mips/thread_info.h
@@ -27,7 +27,7 @@ struct thread_info {
 	struct exec_domain	*exec_domain;	/* execution domain */
 	unsigned long		flags;		/* low level flags */
 	__u32			cpu;		/* current CPU */
-	__s32			preempt_count; /* 0 => preemptable, <0 => BUG */
+	int			preempt_count;	/* 0 => preemptable, <0 => BUG */
 
 	mm_segment_t		addr_limit;	/* thread address space:
 					 	   0-0xBFFFFFFF for user-thead
diff --git a/include/asm-parisc/thread_info.h b/include/asm-parisc/thread_info.h
index fe9b7f8ae4c..57bbb76cb6c 100644
--- a/include/asm-parisc/thread_info.h
+++ b/include/asm-parisc/thread_info.h
@@ -12,7 +12,7 @@ struct thread_info {
 	unsigned long flags;		/* thread_info flags (see TIF_*) */
 	mm_segment_t addr_limit;	/* user-level address space limit */
 	__u32 cpu;			/* current CPU */
-	__s32 preempt_count;		/* 0=premptable, <0=BUG; will also serve as bh-counter */
+	int preempt_count;		/* 0=premptable, <0=BUG; will also serve as bh-counter */
 	struct restart_block restart_block;
 };
 
diff --git a/include/asm-ppc/thread_info.h b/include/asm-ppc/thread_info.h
index e3b5284a6f9..27903db42ef 100644
--- a/include/asm-ppc/thread_info.h
+++ b/include/asm-ppc/thread_info.h
@@ -20,7 +20,8 @@ struct thread_info {
 	unsigned long		flags;		/* low level flags */
 	unsigned long		local_flags;	/* non-racy flags */
 	int			cpu;		/* cpu we're on */
-	int			preempt_count;
+	int			preempt_count;	/* 0 => preemptable,
+						   <0 => BUG */
 	struct restart_block	restart_block;
 };
 
diff --git a/include/asm-ppc64/thread_info.h b/include/asm-ppc64/thread_info.h
index 48b7900e90e..0494df6fca7 100644
--- a/include/asm-ppc64/thread_info.h
+++ b/include/asm-ppc64/thread_info.h
@@ -24,7 +24,7 @@ struct thread_info {
 	struct task_struct *task;		/* main task structure */
 	struct exec_domain *exec_domain;	/* execution domain */
 	int		cpu;			/* cpu we're on */
-	int		preempt_count;
+	int		preempt_count;		/* 0 => preemptable, <0 => BUG */
 	struct restart_block restart_block;
 	/* set by force_successful_syscall_return */
 	unsigned char	syscall_noerror;
diff --git a/include/asm-s390/thread_info.h b/include/asm-s390/thread_info.h
index aade85c53a6..fe101d41e84 100644
--- a/include/asm-s390/thread_info.h
+++ b/include/asm-s390/thread_info.h
@@ -50,7 +50,7 @@ struct thread_info {
 	struct exec_domain	*exec_domain;	/* execution domain */
 	unsigned long		flags;		/* low level flags */
 	unsigned int		cpu;		/* current CPU */
-	unsigned int		preempt_count; /* 0 => preemptable */
+	int			preempt_count;	/* 0 => preemptable, <0 => BUG */
 	struct restart_block	restart_block;
 };
 
diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h
index 4bbbd9f3c37..46080cefaff 100644
--- a/include/asm-sh/thread_info.h
+++ b/include/asm-sh/thread_info.h
@@ -20,7 +20,7 @@ struct thread_info {
 	struct exec_domain	*exec_domain;	/* execution domain */
 	__u32			flags;		/* low level flags */
 	__u32			cpu;
-	__s32			preempt_count; /* 0 => preemptable, <0 => BUG */
+	int			preempt_count; /* 0 => preemptable, <0 => BUG */
 	struct restart_block	restart_block;
 	__u8			supervisor_stack[0];
 };
diff --git a/include/asm-sh64/thread_info.h b/include/asm-sh64/thread_info.h
index 8a32d6bd0b7..10f024c6a2e 100644
--- a/include/asm-sh64/thread_info.h
+++ b/include/asm-sh64/thread_info.h
@@ -22,7 +22,7 @@ struct thread_info {
 	struct exec_domain	*exec_domain;	/* execution domain */
 	unsigned long		flags;		/* low level flags */
 	/* Put the 4 32-bit fields together to make asm offsetting easier. */
-	__s32			preempt_count; /* 0 => preemptable, <0 => BUG */
+	int			preempt_count;	/* 0 => preemptable, <0 => BUG */
 	__u16			cpu;
 
 	mm_segment_t		addr_limit;
diff --git a/include/asm-sparc/thread_info.h b/include/asm-sparc/thread_info.h
index 104f03c5541..ff6ccb3d24c 100644
--- a/include/asm-sparc/thread_info.h
+++ b/include/asm-sparc/thread_info.h
@@ -30,9 +30,9 @@ struct thread_info {
 	struct task_struct	*task;		/* main task structure */
 	struct exec_domain	*exec_domain;	/* execution domain */
 	unsigned long		flags;		/* low level flags */
-
 	int			cpu;		/* cpu we're on */
-	int			preempt_count;
+	int			preempt_count;	/* 0 => preemptable,
+						   <0 => BUG */
 	int			softirq_count;
 	int			hardirq_count;
 
diff --git a/include/asm-sparc64/thread_info.h b/include/asm-sparc64/thread_info.h
index 517caaba1c8..0cd65295692 100644
--- a/include/asm-sparc64/thread_info.h
+++ b/include/asm-sparc64/thread_info.h
@@ -46,7 +46,7 @@ struct thread_info {
 	unsigned long		fault_address;
 	struct pt_regs		*kregs;
 	struct exec_domain	*exec_domain;
-	int			preempt_count;
+	int			preempt_count;	/* 0 => preemptable, <0 => BUG */
 	int			__pad;
 
 	unsigned long		*utraps;
diff --git a/include/asm-um/thread_info.h b/include/asm-um/thread_info.h
index 1feaaf148ef..97267f059ef 100644
--- a/include/asm-um/thread_info.h
+++ b/include/asm-um/thread_info.h
@@ -17,7 +17,7 @@ struct thread_info {
 	struct exec_domain	*exec_domain;	/* execution domain */
 	unsigned long		flags;		/* low level flags */
 	__u32			cpu;		/* current CPU */
-	__s32			preempt_count;  /* 0 => preemptable, 
+	int			preempt_count;  /* 0 => preemptable,
 						   <0 => BUG */
 	mm_segment_t		addr_limit;	/* thread address space:
 					 	   0-0xBFFFFFFF for user
diff --git a/include/asm-v850/thread_info.h b/include/asm-v850/thread_info.h
index e2ef4459375..e4cfad94a55 100644
--- a/include/asm-v850/thread_info.h
+++ b/include/asm-v850/thread_info.h
@@ -30,7 +30,8 @@ struct thread_info {
 	struct exec_domain	*exec_domain;	/* execution domain */
 	unsigned long		flags;		/* low level flags */
 	int			cpu;		/* cpu we're on */
-	int			preempt_count;
+	int			preempt_count;	/* 0 => preemptable,
+						   <0 => BUG */
 	struct restart_block	restart_block;
 };
 
diff --git a/include/asm-x86_64/thread_info.h b/include/asm-x86_64/thread_info.h
index f4b3b249639..08eb6e4f373 100644
--- a/include/asm-x86_64/thread_info.h
+++ b/include/asm-x86_64/thread_info.h
@@ -29,7 +29,7 @@ struct thread_info {
 	__u32			flags;		/* low level flags */
 	__u32			status;		/* thread synchronous flags */
 	__u32			cpu;		/* current CPU */
-	int 			preempt_count;
+	int 			preempt_count;	/* 0 => preemptable, <0 => BUG */
 
 	mm_segment_t		addr_limit;	
 	struct restart_block    restart_block;
-- 
cgit v1.2.3-70-g09d2


From ac20427ef6aa63da663bdc88b71d16f7394f5e23 Mon Sep 17 00:00:00 2001
From: Neil Horman <nhorman@redhat.com>
Date: Thu, 23 Jun 2005 00:09:11 -0700
Subject: [PATCH] add check to /proc/devices read routines

Patch to add check to get_chrdev_list and get_blkdev_list to prevent reads
of /proc/devices from spilling over the provided page if more than 4096
bytes of string data are generated from all the registered character and
block devices in a system

Signed-off-by: Neil Horman <nhorman@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: <viro@parcelfarce.linux.theplanet.co.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/genhd.c | 12 ++++++++++--
 fs/char_dev.c         | 13 ++++++++++++-
 fs/proc/proc_misc.c   |  2 +-
 include/linux/genhd.h |  2 +-
 4 files changed, 24 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/drivers/block/genhd.c b/drivers/block/genhd.c
index 43805e4d31e..47fd3659a06 100644
--- a/drivers/block/genhd.c
+++ b/drivers/block/genhd.c
@@ -40,7 +40,7 @@ static inline int major_to_index(int major)
 
 #ifdef CONFIG_PROC_FS
 /* get block device names in somewhat random order */
-int get_blkdev_list(char *p)
+int get_blkdev_list(char *p, int used)
 {
 	struct blk_major_name *n;
 	int i, len;
@@ -49,10 +49,18 @@ int get_blkdev_list(char *p)
 
 	down(&block_subsys_sem);
 	for (i = 0; i < ARRAY_SIZE(major_names); i++) {
-		for (n = major_names[i]; n; n = n->next)
+		for (n = major_names[i]; n; n = n->next) {
+			/*
+			 * If the curent string plus the 5 extra characters
+			 * in the line would run us off the page, then we're done
+			 */
+			if ((len + used + strlen(n->name) + 5) >= PAGE_SIZE)
+				goto page_full;
 			len += sprintf(p+len, "%3d %s\n",
 				       n->major, n->name);
+		}
 	}
+page_full:
 	up(&block_subsys_sem);
 
 	return len;
diff --git a/fs/char_dev.c b/fs/char_dev.c
index c1e3537909f..e82aac9cc2f 100644
--- a/fs/char_dev.c
+++ b/fs/char_dev.c
@@ -56,10 +56,21 @@ int get_chrdev_list(char *page)
 
 	down(&chrdevs_lock);
 	for (i = 0; i < ARRAY_SIZE(chrdevs) ; i++) {
-		for (cd = chrdevs[i]; cd; cd = cd->next)
+		for (cd = chrdevs[i]; cd; cd = cd->next) {
+			/*
+			 * if the current name, plus the 5 extra characters
+			 * in the device line for this entry
+			 * would run us off the page, we're done
+			 */
+			if ((len+strlen(cd->name) + 5) >= PAGE_SIZE)
+				goto page_full;
+
+
 			len += sprintf(page+len, "%3d %s\n",
 				       cd->major, cd->name);
+		}
 	}
+page_full:
 	up(&chrdevs_lock);
 
 	return len;
diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c
index 63a9fbf1ac5..94b570ad037 100644
--- a/fs/proc/proc_misc.c
+++ b/fs/proc/proc_misc.c
@@ -451,7 +451,7 @@ static int devices_read_proc(char *page, char **start, off_t off,
 				 int count, int *eof, void *data)
 {
 	int len = get_chrdev_list(page);
-	len += get_blkdev_list(page+len);
+	len += get_blkdev_list(page+len, len);
 	return proc_calc_metrics(page, start, off, count, eof, len);
 }
 
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index af26dc718ef..01796c41c95 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -224,7 +224,7 @@ static inline void free_disk_stats(struct gendisk *disk)
 extern void disk_round_stats(struct gendisk *disk);
 
 /* drivers/block/genhd.c */
-extern int get_blkdev_list(char *);
+extern int get_blkdev_list(char *, int);
 extern void add_disk(struct gendisk *disk);
 extern void del_gendisk(struct gendisk *gp);
 extern void unlink_gendisk(struct gendisk *gp);
-- 
cgit v1.2.3-70-g09d2


From 84de856ed30c568c2bb7b9ac0679772bd2737d9b Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Thu, 23 Jun 2005 00:09:16 -0700
Subject: [PATCH] quota: consolidate code surrounding vfs_quota_on_mount

Move some code duplicated in both callers into vfs_quota_on_mount

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jan Kara <jack@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/dquot.c               | 23 +++++++++++++++++++----
 fs/ext3/super.c          | 18 ++----------------
 fs/reiserfs/super.c      | 21 +++------------------
 include/linux/quotaops.h |  3 ++-
 4 files changed, 26 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/fs/dquot.c b/fs/dquot.c
index 3995ce7907c..343c0365561 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -1519,14 +1519,29 @@ out_path:
  * This function is used when filesystem needs to initialize quotas
  * during mount time.
  */
-int vfs_quota_on_mount(int type, int format_id, struct dentry *dentry)
+int vfs_quota_on_mount(struct super_block *sb, char *qf_name,
+		int format_id, int type)
 {
+	struct qstr name = {.name = qf_name, .len = 0, .len = strlen(qf_name)};
+	struct dentry *dentry;
 	int error;
 
+	dentry = lookup_hash(&name, sb->s_root);
+	if (IS_ERR(dentry))
+		return PTR_ERR(dentry);
+
 	error = security_quota_on(dentry);
-	if (error)
-		return error;
-	return vfs_quota_on_inode(dentry->d_inode, type, format_id);
+	if (!error)
+		error = vfs_quota_on_inode(dentry->d_inode, type, format_id);
+
+	/*
+	 * Now invalidate and put the dentry - quota got its own reference
+	 * to inode and dentry has at least wrong hash so we had better
+	 * throw it away.
+	 */
+	d_invalidate(dentry);
+	dput(dentry);
+	return error;
 }
 
 /* Generic routine for getting common part of quota structure */
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 981ccb233ef..9630fbfdc24 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -2348,22 +2348,8 @@ static int ext3_write_info(struct super_block *sb, int type)
  */
 static int ext3_quota_on_mount(struct super_block *sb, int type)
 {
-	int err;
-	struct dentry *dentry;
-	struct qstr name = { .name = EXT3_SB(sb)->s_qf_names[type],
-			     .hash = 0,
-			     .len = strlen(EXT3_SB(sb)->s_qf_names[type])};
-
-	dentry = lookup_hash(&name, sb->s_root);
-	if (IS_ERR(dentry))
-		return PTR_ERR(dentry);
-	err = vfs_quota_on_mount(type, EXT3_SB(sb)->s_jquota_fmt, dentry);
-	/* Now invalidate and put the dentry - quota got its own reference
-	 * to inode and dentry has at least wrong hash so we had better
-	 * throw it away */
-	d_invalidate(dentry);
-	dput(dentry);
-	return err;
+	return vfs_quota_on_mount(sb, EXT3_SB(sb)->s_qf_names[type],
+			EXT3_SB(sb)->s_jquota_fmt, type);
 }
 
 /*
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index b35b8774498..aae0779ed5b 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -1932,27 +1932,12 @@ static int reiserfs_write_info(struct super_block *sb, int type)
 }
 
 /*
- * Turn on quotas during mount time - we need to find
- * the quota file and such...
+ * Turn on quotas during mount time - we need to find the quota file and such...
  */
 static int reiserfs_quota_on_mount(struct super_block *sb, int type)
 {
-    int err;
-    struct dentry *dentry;
-    struct qstr name = { .name = REISERFS_SB(sb)->s_qf_names[type],
-                         .hash = 0,
-                         .len = strlen(REISERFS_SB(sb)->s_qf_names[type])};
-
-    dentry = lookup_hash(&name, sb->s_root);
-    if (IS_ERR(dentry))
-            return PTR_ERR(dentry);
-    err = vfs_quota_on_mount(type, REISERFS_SB(sb)->s_jquota_fmt, dentry);
-    /* Now invalidate and put the dentry - quota got its own reference
-     * to inode and dentry has at least wrong hash so we had better
-     * throw it away */
-    d_invalidate(dentry);
-    dput(dentry);
-    return err;
+	return vfs_quota_on_mount(sb, REISERFS_SB(sb)->s_qf_names[type],
+			REISERFS_SB(sb)->s_jquota_fmt, type);
 }
 
 /*
diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h
index e57baa85e74..d211507ab24 100644
--- a/include/linux/quotaops.h
+++ b/include/linux/quotaops.h
@@ -39,7 +39,8 @@ extern int dquot_commit_info(struct super_block *sb, int type);
 extern int dquot_mark_dquot_dirty(struct dquot *dquot);
 
 extern int vfs_quota_on(struct super_block *sb, int type, int format_id, char *path);
-extern int vfs_quota_on_mount(int type, int format_id, struct dentry *dentry);
+extern int vfs_quota_on_mount(struct super_block *sb, char *qf_name,
+		int format_id, int type);
 extern int vfs_quota_off(struct super_block *sb, int type);
 #define vfs_quota_off_mount(sb, type) vfs_quota_off(sb, type)
 extern int vfs_quota_sync(struct super_block *sb, int type);
-- 
cgit v1.2.3-70-g09d2


From b94cce926b2b902b79380ccba370d6f9f2980de0 Mon Sep 17 00:00:00 2001
From: Hien Nguyen <hien@us.ibm.com>
Date: Thu, 23 Jun 2005 00:09:19 -0700
Subject: [PATCH] kprobes: function-return probes

This patch adds function-return probes to kprobes for the i386
architecture.  This enables you to establish a handler to be run when a
function returns.

1. API

Two new functions are added to kprobes:

	int register_kretprobe(struct kretprobe *rp);
	void unregister_kretprobe(struct kretprobe *rp);

2. Registration and unregistration

2.1 Register

  To register a function-return probe, the user populates the following
  fields in a kretprobe object and calls register_kretprobe() with the
  kretprobe address as an argument:

  kp.addr - the function's address

  handler - this function is run after the ret instruction executes, but
  before control returns to the return address in the caller.

  maxactive - The maximum number of instances of the probed function that
  can be active concurrently.  For example, if the function is non-
  recursive and is called with a spinlock or mutex held, maxactive = 1
  should be enough.  If the function is non-recursive and can never
  relinquish the CPU (e.g., via a semaphore or preemption), NR_CPUS should
  be enough.  maxactive is used to determine how many kretprobe_instance
  objects to allocate for this particular probed function.  If maxactive <=
  0, it is set to a default value (if CONFIG_PREEMPT maxactive=max(10, 2 *
  NR_CPUS) else maxactive=NR_CPUS)

  For example:

    struct kretprobe rp;
    rp.kp.addr = /* entrypoint address */
    rp.handler = /*return probe handler */
    rp.maxactive = /* e.g., 1 or NR_CPUS or 0, see the above explanation */
    register_kretprobe(&rp);

  The following field may also be of interest:

  nmissed - Initialized to zero when the function-return probe is
  registered, and incremented every time the probed function is entered but
  there is no kretprobe_instance object available for establishing the
  function-return probe (i.e., because maxactive was set too low).

2.2 Unregister

  To unregiter a function-return probe, the user calls
  unregister_kretprobe() with the same kretprobe object as registered
  previously.  If a probed function is running when the return probe is
  unregistered, the function will return as expected, but the handler won't
  be run.

3. Limitations

3.1 This patch supports only the i386 architecture, but patches for
    x86_64 and ppc64 are anticipated soon.

3.2 Return probes operates by replacing the return address in the stack
    (or in a known register, such as the lr register for ppc).  This may
    cause __builtin_return_address(0), when invoked from the return-probed
    function, to return the address of the return-probes trampoline.

3.3 This implementation uses the "Multiprobes at an address" feature in
    2.6.12-rc3-mm3.

3.4 Due to a limitation in multi-probes, you cannot currently establish
    a return probe and a jprobe on the same function.  A patch to remove
    this limitation is being tested.

This feature is required by SystemTap (http://sourceware.org/systemtap),
and reflects ideas contributed by several SystemTap developers, including
Will Cohen and Ananth Mavinakayanahalli.

Signed-off-by: Hien Nguyen <hien@us.ibm.com>
Signed-off-by: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Signed-off-by: Frederik Deweerdt <frederik.deweerdt@laposte.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/kprobes.c | 102 +++++++++++++++++++++-
 arch/i386/kernel/process.c |  15 ++++
 include/asm-i386/kprobes.h |   3 +
 include/linux/kprobes.h    |  90 ++++++++++++++++++-
 kernel/kprobes.c           | 213 +++++++++++++++++++++++++++++++++++++++++++--
 5 files changed, 415 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c
index 59ff9b45506..048f754bbe2 100644
--- a/arch/i386/kernel/kprobes.c
+++ b/arch/i386/kernel/kprobes.c
@@ -23,6 +23,9 @@
  *		Rusty Russell).
  * 2004-July	Suparna Bhattacharya <suparna@in.ibm.com> added jumper probes
  *		interface to access function arguments.
+ * 2005-May	Hien Nguyen <hien@us.ibm.com>, Jim Keniston
+ *		<jkenisto@us.ibm.com> and Prasanna S Panchamukhi
+ *		<prasanna@in.ibm.com> added function-return probes.
  */
 
 #include <linux/config.h>
@@ -91,6 +94,53 @@ static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 		regs->eip = (unsigned long)&p->ainsn.insn;
 }
 
+struct task_struct  *arch_get_kprobe_task(void *ptr)
+{
+	return ((struct thread_info *) (((unsigned long) ptr) &
+					(~(THREAD_SIZE -1))))->task;
+}
+
+void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs)
+{
+	unsigned long *sara = (unsigned long *)&regs->esp;
+	struct kretprobe_instance *ri;
+	static void *orig_ret_addr;
+
+	/*
+	 * Save the return address when the return probe hits
+	 * the first time, and use it to populate the (krprobe
+	 * instance)->ret_addr for subsequent return probes at
+	 * the same addrress since stack address would have
+	 * the kretprobe_trampoline by then.
+	 */
+	if (((void*) *sara) != kretprobe_trampoline)
+		orig_ret_addr = (void*) *sara;
+
+	if ((ri = get_free_rp_inst(rp)) != NULL) {
+		ri->rp = rp;
+		ri->stack_addr = sara;
+		ri->ret_addr = orig_ret_addr;
+		add_rp_inst(ri);
+		/* Replace the return addr with trampoline addr */
+		*sara = (unsigned long) &kretprobe_trampoline;
+	} else {
+		rp->nmissed++;
+	}
+}
+
+void arch_kprobe_flush_task(struct task_struct *tk, spinlock_t *kp_lock)
+{
+	unsigned long flags = 0;
+	struct kretprobe_instance *ri;
+	spin_lock_irqsave(kp_lock, flags);
+	while ((ri = get_rp_inst_tsk(tk)) != NULL) {
+		*((unsigned long *)(ri->stack_addr)) =
+					(unsigned long) ri->ret_addr;
+		recycle_rp_inst(ri);
+	}
+	spin_unlock_irqrestore(kp_lock, flags);
+}
+
 /*
  * Interrupts are disabled on entry as trap3 is an interrupt gate and they
  * remain disabled thorough out this function.
@@ -183,6 +233,55 @@ no_kprobe:
 	return ret;
 }
 
+/*
+ * For function-return probes, init_kprobes() establishes a probepoint
+ * here. When a retprobed function returns, this probe is hit and
+ * trampoline_probe_handler() runs, calling the kretprobe's handler.
+ */
+ void kretprobe_trampoline_holder(void)
+ {
+ 	asm volatile (  ".global kretprobe_trampoline\n"
+ 			"kretprobe_trampoline: \n"
+ 			"nop\n");
+ }
+
+/*
+ * Called when we hit the probe point at kretprobe_trampoline
+ */
+int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct task_struct *tsk;
+	struct kretprobe_instance *ri;
+	struct hlist_head *head;
+	struct hlist_node *node;
+	unsigned long *sara = ((unsigned long *) &regs->esp) - 1;
+
+	tsk = arch_get_kprobe_task(sara);
+	head = kretprobe_inst_table_head(tsk);
+
+	hlist_for_each_entry(ri, node, head, hlist) {
+		if (ri->stack_addr == sara && ri->rp) {
+			if (ri->rp->handler)
+				ri->rp->handler(ri, regs);
+		}
+	}
+	return 0;
+}
+
+void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs,
+						unsigned long flags)
+{
+	struct kretprobe_instance *ri;
+	/* RA already popped */
+	unsigned long *sara = ((unsigned long *)&regs->esp) - 1;
+
+	while ((ri = get_rp_inst(sara))) {
+		regs->eip = (unsigned long)ri->ret_addr;
+		recycle_rp_inst(ri);
+	}
+	regs->eflags &= ~TF_MASK;
+}
+
 /*
  * Called after single-stepping.  p->addr is the address of the
  * instruction whose first byte has been replaced by the "int 3"
@@ -266,7 +365,8 @@ static inline int post_kprobe_handler(struct pt_regs *regs)
 	if (current_kprobe->post_handler)
 		current_kprobe->post_handler(current_kprobe, regs, 0);
 
-	resume_execution(current_kprobe, regs);
+	if (current_kprobe->post_handler != trampoline_post_handler)
+		resume_execution(current_kprobe, regs);
 	regs->eflags |= kprobe_saved_eflags;
 
 	unlock_kprobes();
diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index be3efba7caf..aea2ce1145d 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -37,6 +37,7 @@
 #include <linux/kallsyms.h>
 #include <linux/ptrace.h>
 #include <linux/random.h>
+#include <linux/kprobes.h>
 
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
@@ -339,6 +340,13 @@ void exit_thread(void)
 	struct task_struct *tsk = current;
 	struct thread_struct *t = &tsk->thread;
 
+	/*
+	 * Remove function-return probe instances associated with this task
+	 * and put them back on the free list. Do not insert an exit probe for
+	 * this function, it will be disabled by kprobe_flush_task if you do.
+	 */
+	kprobe_flush_task(tsk);
+
 	/* The process may have allocated an io port bitmap... nuke it. */
 	if (unlikely(NULL != t->io_bitmap_ptr)) {
 		int cpu = get_cpu();
@@ -362,6 +370,13 @@ void flush_thread(void)
 {
 	struct task_struct *tsk = current;
 
+	/*
+	 * Remove function-return probe instances associated with this task
+	 * and put them back on the free list. Do not insert an exit probe for
+	 * this function, it will be disabled by kprobe_flush_task if you do.
+	 */
+	kprobe_flush_task(tsk);
+
 	memset(tsk->thread.debugreg, 0, sizeof(unsigned long)*8);
 	memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array));	
 	/*
diff --git a/include/asm-i386/kprobes.h b/include/asm-i386/kprobes.h
index 4092f68d123..8b6d3a90cd7 100644
--- a/include/asm-i386/kprobes.h
+++ b/include/asm-i386/kprobes.h
@@ -39,6 +39,9 @@ typedef u8 kprobe_opcode_t;
 	: (((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR)))
 
 #define JPROBE_ENTRY(pentry)	(kprobe_opcode_t *)pentry
+#define ARCH_SUPPORTS_KRETPROBES
+
+void kretprobe_trampoline(void);
 
 /* Architecture specific copy of original instruction*/
 struct arch_specific_insn {
diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h
index 99ddba5a4e0..fba39f87efe 100644
--- a/include/linux/kprobes.h
+++ b/include/linux/kprobes.h
@@ -25,21 +25,31 @@
  *		Rusty Russell).
  * 2004-July	Suparna Bhattacharya <suparna@in.ibm.com> added jumper probes
  *		interface to access function arguments.
+ * 2005-May	Hien Nguyen <hien@us.ibm.com> and Jim Keniston
+ *		<jkenisto@us.ibm.com>  and Prasanna S Panchamukhi
+ *		<prasanna@in.ibm.com> added function-return probes.
  */
 #include <linux/config.h>
 #include <linux/list.h>
 #include <linux/notifier.h>
 #include <linux/smp.h>
+#include <linux/spinlock.h>
+
 #include <asm/kprobes.h>
 
 struct kprobe;
 struct pt_regs;
+struct kretprobe;
+struct kretprobe_instance;
 typedef int (*kprobe_pre_handler_t) (struct kprobe *, struct pt_regs *);
 typedef int (*kprobe_break_handler_t) (struct kprobe *, struct pt_regs *);
 typedef void (*kprobe_post_handler_t) (struct kprobe *, struct pt_regs *,
 				       unsigned long flags);
 typedef int (*kprobe_fault_handler_t) (struct kprobe *, struct pt_regs *,
 				       int trapnr);
+typedef int (*kretprobe_handler_t) (struct kretprobe_instance *,
+				    struct pt_regs *);
+
 struct kprobe {
 	struct hlist_node hlist;
 
@@ -85,6 +95,62 @@ struct jprobe {
 	kprobe_opcode_t *entry;	/* probe handling code to jump to */
 };
 
+#ifdef ARCH_SUPPORTS_KRETPROBES
+extern int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs);
+extern void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs,
+							unsigned long flags);
+extern struct task_struct *arch_get_kprobe_task(void *ptr);
+extern void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs);
+extern void arch_kprobe_flush_task(struct task_struct *tk, spinlock_t *kp_lock);
+#else /* ARCH_SUPPORTS_KRETPROBES */
+static inline void kretprobe_trampoline(void)
+{
+}
+static inline int trampoline_probe_handler(struct kprobe *p,
+						struct pt_regs *regs)
+{
+	return 0;
+}
+static inline void trampoline_post_handler(struct kprobe *p,
+				struct pt_regs *regs, unsigned long flags)
+{
+}
+static inline void arch_prepare_kretprobe(struct kretprobe *rp,
+					struct pt_regs *regs)
+{
+}
+static inline void arch_kprobe_flush_task(struct task_struct *tk)
+{
+}
+#define arch_get_kprobe_task(ptr) ((struct task_struct *)NULL)
+#endif /* ARCH_SUPPORTS_KRETPROBES */
+/*
+ * Function-return probe -
+ * Note:
+ * User needs to provide a handler function, and initialize maxactive.
+ * maxactive - The maximum number of instances of the probed function that
+ * can be active concurrently.
+ * nmissed - tracks the number of times the probed function's return was
+ * ignored, due to maxactive being too low.
+ *
+ */
+struct kretprobe {
+	struct kprobe kp;
+	kretprobe_handler_t handler;
+	int maxactive;
+	int nmissed;
+	struct hlist_head free_instances;
+	struct hlist_head used_instances;
+};
+
+struct kretprobe_instance {
+	struct hlist_node uflist; /* either on free list or used list */
+	struct hlist_node hlist;
+	struct kretprobe *rp;
+	void *ret_addr;
+	void *stack_addr;
+};
+
 #ifdef CONFIG_KPROBES
 /* Locks kprobe: irq must be disabled */
 void lock_kprobes(void);
@@ -104,6 +170,7 @@ extern void show_registers(struct pt_regs *regs);
 
 /* Get the kprobe at this addr (if any).  Must have called lock_kprobes */
 struct kprobe *get_kprobe(void *addr);
+struct hlist_head * kretprobe_inst_table_head(struct task_struct *tsk);
 
 int register_kprobe(struct kprobe *p);
 void unregister_kprobe(struct kprobe *p);
@@ -113,7 +180,16 @@ int register_jprobe(struct jprobe *p);
 void unregister_jprobe(struct jprobe *p);
 void jprobe_return(void);
 
-#else
+int register_kretprobe(struct kretprobe *rp);
+void unregister_kretprobe(struct kretprobe *rp);
+
+struct kretprobe_instance *get_free_rp_inst(struct kretprobe *rp);
+struct kretprobe_instance *get_rp_inst(void *sara);
+struct kretprobe_instance *get_rp_inst_tsk(struct task_struct *tk);
+void add_rp_inst(struct kretprobe_instance *ri);
+void kprobe_flush_task(struct task_struct *tk);
+void recycle_rp_inst(struct kretprobe_instance *ri);
+#else /* CONFIG_KPROBES */
 static inline int kprobe_running(void)
 {
 	return 0;
@@ -135,5 +211,15 @@ static inline void unregister_jprobe(struct jprobe *p)
 static inline void jprobe_return(void)
 {
 }
-#endif
+static inline int register_kretprobe(struct kretprobe *rp)
+{
+	return -ENOSYS;
+}
+static inline void unregister_kretprobe(struct kretprobe *rp)
+{
+}
+static inline void kprobe_flush_task(struct task_struct *tk)
+{
+}
+#endif				/* CONFIG_KPROBES */
 #endif				/* _LINUX_KPROBES_H */
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 037142b72a4..692fbf75ab4 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -27,6 +27,9 @@
  *		interface to access function arguments.
  * 2004-Sep	Prasanna S Panchamukhi <prasanna@in.ibm.com> Changed Kprobes
  *		exceptions notifier to be first on the priority list.
+ * 2005-May	Hien Nguyen <hien@us.ibm.com>, Jim Keniston
+ *		<jkenisto@us.ibm.com> and Prasanna S Panchamukhi
+ *		<prasanna@in.ibm.com> added function-return probes.
  */
 #include <linux/kprobes.h>
 #include <linux/spinlock.h>
@@ -41,6 +44,7 @@
 #define KPROBE_TABLE_SIZE (1 << KPROBE_HASH_BITS)
 
 static struct hlist_head kprobe_table[KPROBE_TABLE_SIZE];
+static struct hlist_head kretprobe_inst_table[KPROBE_TABLE_SIZE];
 
 unsigned int kprobe_cpu = NR_CPUS;
 static DEFINE_SPINLOCK(kprobe_lock);
@@ -78,7 +82,7 @@ struct kprobe *get_kprobe(void *addr)
  * Aggregate handlers for multiple kprobes support - these handlers
  * take care of invoking the individual kprobe handlers on p->list
  */
-int aggr_pre_handler(struct kprobe *p, struct pt_regs *regs)
+static int aggr_pre_handler(struct kprobe *p, struct pt_regs *regs)
 {
 	struct kprobe *kp;
 
@@ -92,8 +96,8 @@ int aggr_pre_handler(struct kprobe *p, struct pt_regs *regs)
 	return 0;
 }
 
-void aggr_post_handler(struct kprobe *p, struct pt_regs *regs,
-		unsigned long flags)
+static void aggr_post_handler(struct kprobe *p, struct pt_regs *regs,
+			      unsigned long flags)
 {
 	struct kprobe *kp;
 
@@ -107,7 +111,8 @@ void aggr_post_handler(struct kprobe *p, struct pt_regs *regs,
 	return;
 }
 
-int aggr_fault_handler(struct kprobe *p, struct pt_regs *regs, int trapnr)
+static int aggr_fault_handler(struct kprobe *p, struct pt_regs *regs,
+			      int trapnr)
 {
 	/*
 	 * if we faulted "during" the execution of a user specified
@@ -120,6 +125,135 @@ int aggr_fault_handler(struct kprobe *p, struct pt_regs *regs, int trapnr)
 	return 0;
 }
 
+struct kprobe trampoline_p = {
+		.addr = (kprobe_opcode_t *) &kretprobe_trampoline,
+		.pre_handler = trampoline_probe_handler,
+		.post_handler = trampoline_post_handler
+};
+
+struct kretprobe_instance *get_free_rp_inst(struct kretprobe *rp)
+{
+	struct hlist_node *node;
+	struct kretprobe_instance *ri;
+	hlist_for_each_entry(ri, node, &rp->free_instances, uflist)
+		return ri;
+	return NULL;
+}
+
+static struct kretprobe_instance *get_used_rp_inst(struct kretprobe *rp)
+{
+	struct hlist_node *node;
+	struct kretprobe_instance *ri;
+	hlist_for_each_entry(ri, node, &rp->used_instances, uflist)
+		return ri;
+	return NULL;
+}
+
+struct kretprobe_instance *get_rp_inst(void *sara)
+{
+	struct hlist_head *head;
+	struct hlist_node *node;
+	struct task_struct *tsk;
+	struct kretprobe_instance *ri;
+
+	tsk = arch_get_kprobe_task(sara);
+	head = &kretprobe_inst_table[hash_ptr(tsk, KPROBE_HASH_BITS)];
+	hlist_for_each_entry(ri, node, head, hlist) {
+		if (ri->stack_addr == sara)
+			return ri;
+	}
+	return NULL;
+}
+
+void add_rp_inst(struct kretprobe_instance *ri)
+{
+	struct task_struct *tsk;
+	/*
+	 * Remove rp inst off the free list -
+	 * Add it back when probed function returns
+	 */
+	hlist_del(&ri->uflist);
+	tsk = arch_get_kprobe_task(ri->stack_addr);
+	/* Add rp inst onto table */
+	INIT_HLIST_NODE(&ri->hlist);
+	hlist_add_head(&ri->hlist,
+			&kretprobe_inst_table[hash_ptr(tsk, KPROBE_HASH_BITS)]);
+
+	/* Also add this rp inst to the used list. */
+	INIT_HLIST_NODE(&ri->uflist);
+	hlist_add_head(&ri->uflist, &ri->rp->used_instances);
+}
+
+void recycle_rp_inst(struct kretprobe_instance *ri)
+{
+	/* remove rp inst off the rprobe_inst_table */
+	hlist_del(&ri->hlist);
+	if (ri->rp) {
+		/* remove rp inst off the used list */
+		hlist_del(&ri->uflist);
+		/* put rp inst back onto the free list */
+		INIT_HLIST_NODE(&ri->uflist);
+		hlist_add_head(&ri->uflist, &ri->rp->free_instances);
+	} else
+		/* Unregistering */
+		kfree(ri);
+}
+
+struct hlist_head * kretprobe_inst_table_head(struct task_struct *tsk)
+{
+	return &kretprobe_inst_table[hash_ptr(tsk, KPROBE_HASH_BITS)];
+}
+
+struct kretprobe_instance *get_rp_inst_tsk(struct task_struct *tk)
+{
+	struct task_struct *tsk;
+	struct hlist_head *head;
+	struct hlist_node *node;
+	struct kretprobe_instance *ri;
+
+	head = &kretprobe_inst_table[hash_ptr(tk, KPROBE_HASH_BITS)];
+
+	hlist_for_each_entry(ri, node, head, hlist) {
+		tsk = arch_get_kprobe_task(ri->stack_addr);
+		if (tsk == tk)
+			return ri;
+	}
+	return NULL;
+}
+
+/*
+ * This function is called from do_exit or do_execv when task tk's stack is
+ * about to be recycled. Recycle any function-return probe instances
+ * associated with this task. These represent probed functions that have
+ * been called but may never return.
+ */
+void kprobe_flush_task(struct task_struct *tk)
+{
+	arch_kprobe_flush_task(tk, &kprobe_lock);
+}
+
+/*
+ * This kprobe pre_handler is registered with every kretprobe. When probe
+ * hits it will set up the return probe.
+ */
+static int pre_handler_kretprobe(struct kprobe *p, struct pt_regs *regs)
+{
+	struct kretprobe *rp = container_of(p, struct kretprobe, kp);
+
+	/*TODO: consider to only swap the RA after the last pre_handler fired */
+	arch_prepare_kretprobe(rp, regs);
+	return 0;
+}
+
+static inline void free_rp_inst(struct kretprobe *rp)
+{
+	struct kretprobe_instance *ri;
+	while ((ri = get_free_rp_inst(rp)) != NULL) {
+		hlist_del(&ri->uflist);
+		kfree(ri);
+	}
+}
+
 /*
  * Fill in the required fields of the "manager kprobe". Replace the
  * earlier kprobe in the hlist with the manager kprobe
@@ -257,16 +391,82 @@ void unregister_jprobe(struct jprobe *jp)
 	unregister_kprobe(&jp->kp);
 }
 
+#ifdef ARCH_SUPPORTS_KRETPROBES
+
+int register_kretprobe(struct kretprobe *rp)
+{
+	int ret = 0;
+	struct kretprobe_instance *inst;
+	int i;
+
+	rp->kp.pre_handler = pre_handler_kretprobe;
+
+	/* Pre-allocate memory for max kretprobe instances */
+	if (rp->maxactive <= 0) {
+#ifdef CONFIG_PREEMPT
+		rp->maxactive = max(10, 2 * NR_CPUS);
+#else
+		rp->maxactive = NR_CPUS;
+#endif
+	}
+	INIT_HLIST_HEAD(&rp->used_instances);
+	INIT_HLIST_HEAD(&rp->free_instances);
+	for (i = 0; i < rp->maxactive; i++) {
+		inst = kmalloc(sizeof(struct kretprobe_instance), GFP_KERNEL);
+		if (inst == NULL) {
+			free_rp_inst(rp);
+			return -ENOMEM;
+		}
+		INIT_HLIST_NODE(&inst->uflist);
+		hlist_add_head(&inst->uflist, &rp->free_instances);
+	}
+
+	rp->nmissed = 0;
+	/* Establish function entry probe point */
+	if ((ret = register_kprobe(&rp->kp)) != 0)
+		free_rp_inst(rp);
+	return ret;
+}
+
+#else /* ARCH_SUPPORTS_KRETPROBES */
+
+int register_kretprobe(struct kretprobe *rp)
+{
+	return -ENOSYS;
+}
+
+#endif /* ARCH_SUPPORTS_KRETPROBES */
+
+void unregister_kretprobe(struct kretprobe *rp)
+{
+	unsigned long flags;
+	struct kretprobe_instance *ri;
+
+	unregister_kprobe(&rp->kp);
+	/* No race here */
+	spin_lock_irqsave(&kprobe_lock, flags);
+	free_rp_inst(rp);
+	while ((ri = get_used_rp_inst(rp)) != NULL) {
+		ri->rp = NULL;
+		hlist_del(&ri->uflist);
+	}
+	spin_unlock_irqrestore(&kprobe_lock, flags);
+}
+
 static int __init init_kprobes(void)
 {
 	int i, err = 0;
 
 	/* FIXME allocate the probe table, currently defined statically */
 	/* initialize all list heads */
-	for (i = 0; i < KPROBE_TABLE_SIZE; i++)
+	for (i = 0; i < KPROBE_TABLE_SIZE; i++) {
 		INIT_HLIST_HEAD(&kprobe_table[i]);
+		INIT_HLIST_HEAD(&kretprobe_inst_table[i]);
+	}
 
 	err = register_die_notifier(&kprobe_exceptions_nb);
+	/* Register the trampoline probe for return probe */
+	register_kprobe(&trampoline_p);
 	return err;
 }
 
@@ -277,3 +477,6 @@ EXPORT_SYMBOL_GPL(unregister_kprobe);
 EXPORT_SYMBOL_GPL(register_jprobe);
 EXPORT_SYMBOL_GPL(unregister_jprobe);
 EXPORT_SYMBOL_GPL(jprobe_return);
+EXPORT_SYMBOL_GPL(register_kretprobe);
+EXPORT_SYMBOL_GPL(unregister_kretprobe);
+
-- 
cgit v1.2.3-70-g09d2


From 73649dab0fd524cb8545a8cb83c6eaf77b107105 Mon Sep 17 00:00:00 2001
From: Rusty Lynch <rusty.lynch@intel.com>
Date: Thu, 23 Jun 2005 00:09:23 -0700
Subject: [PATCH] x86_64 specific function return probes

The following patch adds the x86_64 architecture specific implementation
for function return probes.

Function return probes is a mechanism built on top of kprobes that allows
a caller to register a handler to be called when a given function exits.
For example, to instrument the return path of sys_mkdir:

static int sys_mkdir_exit(struct kretprobe_instance *i, struct pt_regs *regs)
{
	printk("sys_mkdir exited\n");
	return 0;
}
static struct kretprobe return_probe = {
	.handler = sys_mkdir_exit,
};

<inside setup function>

return_probe.kp.addr = (kprobe_opcode_t *) kallsyms_lookup_name("sys_mkdir");
if (register_kretprobe(&return_probe)) {
	printk(KERN_DEBUG "Unable to register return probe!\n");
	/* do error path */
}

<inside cleanup function>
unregister_kretprobe(&return_probe);

The way this works is that:

* At system initialization time, kernel/kprobes.c installs a kprobe
  on a function called kretprobe_trampoline() that is implemented in
  the arch/x86_64/kernel/kprobes.c  (More on this later)

* When a return probe is registered using register_kretprobe(),
  kernel/kprobes.c will install a kprobe on the first instruction of the
  targeted function with the pre handler set to arch_prepare_kretprobe()
  which is implemented in arch/x86_64/kernel/kprobes.c.

* arch_prepare_kretprobe() will prepare a kretprobe instance that stores:
  - nodes for hanging this instance in an empty or free list
  - a pointer to the return probe
  - the original return address
  - a pointer to the stack address

  With all this stowed away, arch_prepare_kretprobe() then sets the return
  address for the targeted function to a special trampoline function called
  kretprobe_trampoline() implemented in arch/x86_64/kernel/kprobes.c

* The kprobe completes as normal, with control passing back to the target
  function that executes as normal, and eventually returns to our trampoline
  function.

* Since a kprobe was installed on kretprobe_trampoline() during system
  initialization, control passes back to kprobes via the architecture
  specific function trampoline_probe_handler() which will lookup the
  instance in an hlist maintained by kernel/kprobes.c, and then call
  the handler function.

* When trampoline_probe_handler() is done, the kprobes infrastructure
  single steps the original instruction (in this case just a top), and
  then calls trampoline_post_handler().  trampoline_post_handler() then
  looks up the instance again, puts the instance back on the free list,
  and then makes a long jump back to the original return instruction.

So to recap, to instrument the exit path of a function this implementation
will cause four interruptions:

  - A breakpoint at the very beginning of the function allowing us to
    switch out the return address
  - A single step interruption to execute the original instruction that
    we replaced with the break instruction (normal kprobe flow)
  - A breakpoint in the trampoline function where our instrumented function
    returned to
  - A single step interruption to execute the original instruction that
    we replaced with the break instruction (normal kprobe flow)

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/kprobes.c | 98 +++++++++++++++++++++++++++++++++++++++++++-
 arch/x86_64/kernel/process.c | 16 ++++++++
 include/asm-x86_64/kprobes.h |  3 ++
 3 files changed, 116 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c
index f77f8a0ff18..203672ca740 100644
--- a/arch/x86_64/kernel/kprobes.c
+++ b/arch/x86_64/kernel/kprobes.c
@@ -27,6 +27,8 @@
  *		<prasanna@in.ibm.com> adapted for x86_64
  * 2005-Mar	Roland McGrath <roland@redhat.com>
  *		Fixed to handle %rip-relative addressing mode correctly.
+ * 2005-May     Rusty Lynch <rusty.lynch@intel.com>
+ *              Added function return probes functionality
  */
 
 #include <linux/config.h>
@@ -240,6 +242,50 @@ static void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 		regs->rip = (unsigned long)p->ainsn.insn;
 }
 
+struct task_struct  *arch_get_kprobe_task(void *ptr)
+{
+	return ((struct thread_info *) (((unsigned long) ptr) &
+					(~(THREAD_SIZE -1))))->task;
+}
+
+void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs)
+{
+	unsigned long *sara = (unsigned long *)regs->rsp;
+	struct kretprobe_instance *ri;
+	static void *orig_ret_addr;
+
+	/*
+	 * Save the return address when the return probe hits
+	 * the first time, and use it to populate the (krprobe
+	 * instance)->ret_addr for subsequent return probes at
+	 * the same addrress since stack address would have
+	 * the kretprobe_trampoline by then.
+	 */
+	if (((void*) *sara) != kretprobe_trampoline)
+		orig_ret_addr = (void*) *sara;
+
+	if ((ri = get_free_rp_inst(rp)) != NULL) {
+		ri->rp = rp;
+		ri->stack_addr = sara;
+		ri->ret_addr = orig_ret_addr;
+		add_rp_inst(ri);
+		/* Replace the return addr with trampoline addr */
+		*sara = (unsigned long) &kretprobe_trampoline;
+	} else {
+		rp->nmissed++;
+	}
+}
+
+void arch_kprobe_flush_task(struct task_struct *tk)
+{
+	struct kretprobe_instance *ri;
+	while ((ri = get_rp_inst_tsk(tk)) != NULL) {
+		*((unsigned long *)(ri->stack_addr)) =
+					(unsigned long) ri->ret_addr;
+		recycle_rp_inst(ri);
+	}
+}
+
 /*
  * Interrupts are disabled on entry as trap3 is an interrupt gate and they
  * remain disabled thorough out this function.
@@ -316,6 +362,55 @@ no_kprobe:
 	return ret;
 }
 
+/*
+ * For function-return probes, init_kprobes() establishes a probepoint
+ * here. When a retprobed function returns, this probe is hit and
+ * trampoline_probe_handler() runs, calling the kretprobe's handler.
+ */
+ void kretprobe_trampoline_holder(void)
+ {
+ 	asm volatile (  ".global kretprobe_trampoline\n"
+ 			"kretprobe_trampoline: \n"
+ 			"nop\n");
+ }
+
+/*
+ * Called when we hit the probe point at kretprobe_trampoline
+ */
+int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct task_struct *tsk;
+	struct kretprobe_instance *ri;
+	struct hlist_head *head;
+	struct hlist_node *node;
+	unsigned long *sara = (unsigned long *)regs->rsp - 1;
+
+	tsk = arch_get_kprobe_task(sara);
+	head = kretprobe_inst_table_head(tsk);
+
+	hlist_for_each_entry(ri, node, head, hlist) {
+		if (ri->stack_addr == sara && ri->rp) {
+			if (ri->rp->handler)
+				ri->rp->handler(ri, regs);
+		}
+	}
+	return 0;
+}
+
+void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs,
+						unsigned long flags)
+{
+	struct kretprobe_instance *ri;
+	/* RA already popped */
+	unsigned long *sara = ((unsigned long *)regs->rsp) - 1;
+
+	while ((ri = get_rp_inst(sara))) {
+		regs->rip = (unsigned long)ri->ret_addr;
+		recycle_rp_inst(ri);
+	}
+	regs->eflags &= ~TF_MASK;
+}
+
 /*
  * Called after single-stepping.  p->addr is the address of the
  * instruction whose first byte has been replaced by the "int 3"
@@ -404,7 +499,8 @@ int post_kprobe_handler(struct pt_regs *regs)
 	if (current_kprobe->post_handler)
 		current_kprobe->post_handler(current_kprobe, regs, 0);
 
-	resume_execution(current_kprobe, regs);
+	if (current_kprobe->post_handler != trampoline_post_handler)
+		resume_execution(current_kprobe, regs);
 	regs->eflags |= kprobe_saved_rflags;
 
 	unlock_kprobes();
diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index dce8bab4306..e59d1f9d616 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -34,6 +34,7 @@
 #include <linux/ptrace.h>
 #include <linux/utsname.h>
 #include <linux/random.h>
+#include <linux/kprobes.h>
 
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
@@ -293,6 +294,14 @@ void exit_thread(void)
 {
 	struct task_struct *me = current;
 	struct thread_struct *t = &me->thread;
+
+	/*
+	 * Remove function-return probe instances associated with this task
+	 * and put them back on the free list. Do not insert an exit probe for
+	 * this function, it will be disabled by kprobe_flush_task if you do.
+	 */
+	kprobe_flush_task(me);
+
 	if (me->thread.io_bitmap_ptr) { 
 		struct tss_struct *tss = &per_cpu(init_tss, get_cpu());
 
@@ -312,6 +321,13 @@ void flush_thread(void)
 	struct task_struct *tsk = current;
 	struct thread_info *t = current_thread_info();
 
+	/*
+	 * Remove function-return probe instances associated with this task
+	 * and put them back on the free list. Do not insert an exit probe for
+	 * this function, it will be disabled by kprobe_flush_task if you do.
+	 */
+	kprobe_flush_task(tsk);
+
 	if (t->flags & _TIF_ABI_PENDING)
 		t->flags ^= (_TIF_ABI_PENDING | _TIF_IA32);
 
diff --git a/include/asm-x86_64/kprobes.h b/include/asm-x86_64/kprobes.h
index bfea52d516f..6d6d883fdf6 100644
--- a/include/asm-x86_64/kprobes.h
+++ b/include/asm-x86_64/kprobes.h
@@ -38,6 +38,9 @@ typedef u8 kprobe_opcode_t;
 	: (((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR)))
 
 #define JPROBE_ENTRY(pentry)	(kprobe_opcode_t *)pentry
+#define ARCH_SUPPORTS_KRETPROBES
+
+void kretprobe_trampoline(void);
 
 /* Architecture specific copy of original instruction*/
 struct arch_specific_insn {
-- 
cgit v1.2.3-70-g09d2


From 7e1048b11c5afe79aac46a42e3ccec86b8365c6d Mon Sep 17 00:00:00 2001
From: Rusty Lynch <rusty.lynch@intel.com>
Date: Thu, 23 Jun 2005 00:09:25 -0700
Subject: [PATCH] Move kprobe [dis]arming into arch specific code

The architecture independent code of the current kprobes implementation is
arming and disarming kprobes at registration time.  The problem is that the
code is assuming that arming and disarming is a just done by a simple write
of some magic value to an address.  This is problematic for ia64 where our
instructions look more like structures, and we can not insert break points
by just doing something like:

*p->addr = BREAKPOINT_INSTRUCTION;

The following patch to 2.6.12-rc4-mm2 adds two new architecture dependent
functions:

     * void arch_arm_kprobe(struct kprobe *p)
     * void arch_disarm_kprobe(struct kprobe *p)

and then adds the new functions for each of the architectures that already
implement kprobes (spar64/ppc64/i386/x86_64).

I thought arch_[dis]arm_kprobe was the most descriptive of what was really
happening, but each of the architectures already had a disarm_kprobe()
function that was really a "disarm and do some other clean-up items as
needed when you stumble across a recursive kprobe." So...  I took the
liberty of changing the code that was calling disarm_kprobe() to call
arch_disarm_kprobe(), and then do the cleanup in the block of code dealing
with the recursive kprobe case.

So far this patch as been tested on i386, x86_64, and ppc64, but still
needs to be tested in sparc64.

Signed-off-by: Rusty Lynch <rusty.lynch@intel.com>
Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/kprobes.c    | 19 +++++++++++++++----
 arch/ppc64/kernel/kprobes.c   | 19 +++++++++++++++----
 arch/sparc64/kernel/kprobes.c | 31 ++++++++++++++++++-------------
 arch/x86_64/kernel/kprobes.c  | 26 ++++++++++++++++++--------
 include/linux/kprobes.h       |  2 ++
 kernel/kprobes.c              | 12 ++++--------
 6 files changed, 72 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c
index 048f754bbe2..2314d8d306f 100644
--- a/arch/i386/kernel/kprobes.c
+++ b/arch/i386/kernel/kprobes.c
@@ -33,6 +33,7 @@
 #include <linux/ptrace.h>
 #include <linux/spinlock.h>
 #include <linux/preempt.h>
+#include <asm/cacheflush.h>
 #include <asm/kdebug.h>
 #include <asm/desc.h>
 
@@ -71,16 +72,25 @@ int arch_prepare_kprobe(struct kprobe *p)
 void arch_copy_kprobe(struct kprobe *p)
 {
 	memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
+	p->opcode = *p->addr;
 }
 
-void arch_remove_kprobe(struct kprobe *p)
+void arch_arm_kprobe(struct kprobe *p)
 {
+	*p->addr = BREAKPOINT_INSTRUCTION;
+	flush_icache_range((unsigned long) p->addr,
+			   (unsigned long) p->addr + sizeof(kprobe_opcode_t));
 }
 
-static inline void disarm_kprobe(struct kprobe *p, struct pt_regs *regs)
+void arch_disarm_kprobe(struct kprobe *p)
 {
 	*p->addr = p->opcode;
-	regs->eip = (unsigned long)p->addr;
+	flush_icache_range((unsigned long) p->addr,
+			   (unsigned long) p->addr + sizeof(kprobe_opcode_t));
+}
+
+void arch_remove_kprobe(struct kprobe *p)
+{
 }
 
 static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
@@ -177,7 +187,8 @@ static int kprobe_handler(struct pt_regs *regs)
 				unlock_kprobes();
 				goto no_kprobe;
 			}
-			disarm_kprobe(p, regs);
+			arch_disarm_kprobe(p);
+			regs->eip = (unsigned long)p->addr;
 			ret = 1;
 		} else {
 			p = current_kprobe;
diff --git a/arch/ppc64/kernel/kprobes.c b/arch/ppc64/kernel/kprobes.c
index e950a2058a1..8c0920a6d03 100644
--- a/arch/ppc64/kernel/kprobes.c
+++ b/arch/ppc64/kernel/kprobes.c
@@ -32,6 +32,7 @@
 #include <linux/ptrace.h>
 #include <linux/spinlock.h>
 #include <linux/preempt.h>
+#include <asm/cacheflush.h>
 #include <asm/kdebug.h>
 #include <asm/sstep.h>
 
@@ -61,16 +62,25 @@ int arch_prepare_kprobe(struct kprobe *p)
 void arch_copy_kprobe(struct kprobe *p)
 {
 	memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
+	p->opcode = *p->addr;
 }
 
-void arch_remove_kprobe(struct kprobe *p)
+void arch_arm_kprobe(struct kprobe *p)
 {
+	*p->addr = BREAKPOINT_INSTRUCTION;
+	flush_icache_range((unsigned long) p->addr,
+			   (unsigned long) p->addr + sizeof(kprobe_opcode_t));
 }
 
-static inline void disarm_kprobe(struct kprobe *p, struct pt_regs *regs)
+void arch_disarm_kprobe(struct kprobe *p)
 {
 	*p->addr = p->opcode;
-	regs->nip = (unsigned long)p->addr;
+	flush_icache_range((unsigned long) p->addr,
+			   (unsigned long) p->addr + sizeof(kprobe_opcode_t));
+}
+
+void arch_remove_kprobe(struct kprobe *p)
+{
 }
 
 static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
@@ -101,7 +111,8 @@ static inline int kprobe_handler(struct pt_regs *regs)
 				unlock_kprobes();
 				goto no_kprobe;
 			}
-			disarm_kprobe(p, regs);
+			arch_disarm_kprobe(p);
+			regs->nip = (unsigned long)p->addr;
 			ret = 1;
 		} else {
 			p = current_kprobe;
diff --git a/arch/sparc64/kernel/kprobes.c b/arch/sparc64/kernel/kprobes.c
index 7066d7ba667..d67195ba3fa 100644
--- a/arch/sparc64/kernel/kprobes.c
+++ b/arch/sparc64/kernel/kprobes.c
@@ -6,7 +6,6 @@
 #include <linux/config.h>
 #include <linux/kernel.h>
 #include <linux/kprobes.h>
-
 #include <asm/kdebug.h>
 #include <asm/signal.h>
 
@@ -47,6 +46,19 @@ void arch_copy_kprobe(struct kprobe *p)
 {
 	p->ainsn.insn[0] = *p->addr;
 	p->ainsn.insn[1] = BREAKPOINT_INSTRUCTION_2;
+	p->opcode = *p->addr;
+}
+
+void arch_arm_kprobe(struct kprobe *p)
+{
+	*p->addr = BREAKPOINT_INSTRUCTION;
+	flushi(p->addr);
+}
+
+void arch_disarm_kprobe(struct kprobe *p)
+{
+	*p->addr = p->opcode;
+	flushi(p->addr);
 }
 
 void arch_remove_kprobe(struct kprobe *p)
@@ -78,17 +90,6 @@ static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 	}
 }
 
-static inline void disarm_kprobe(struct kprobe *p, struct pt_regs *regs)
-{
-	*p->addr = p->opcode;
-	flushi(p->addr);
-
-	regs->tpc = (unsigned long) p->addr;
-	regs->tnpc = current_kprobe_orig_tnpc;
-	regs->tstate = ((regs->tstate & ~TSTATE_PIL) |
-			current_kprobe_orig_tstate_pil);
-}
-
 static int kprobe_handler(struct pt_regs *regs)
 {
 	struct kprobe *p;
@@ -109,7 +110,11 @@ static int kprobe_handler(struct pt_regs *regs)
 				unlock_kprobes();
 				goto no_kprobe;
 			}
-			disarm_kprobe(p, regs);
+			arch_disarm_kprobe(p);
+			regs->tpc = (unsigned long) p->addr;
+			regs->tnpc = current_kprobe_orig_tnpc;
+			regs->tstate = ((regs->tstate & ~TSTATE_PIL) |
+					current_kprobe_orig_tstate_pil);
 			ret = 1;
 		} else {
 			p = current_kprobe;
diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c
index 203672ca740..324bf57925a 100644
--- a/arch/x86_64/kernel/kprobes.c
+++ b/arch/x86_64/kernel/kprobes.c
@@ -39,7 +39,7 @@
 #include <linux/slab.h>
 #include <linux/preempt.h>
 #include <linux/moduleloader.h>
-
+#include <asm/cacheflush.h>
 #include <asm/pgtable.h>
 #include <asm/kdebug.h>
 
@@ -216,19 +216,28 @@ void arch_copy_kprobe(struct kprobe *p)
 		BUG_ON((s64) (s32) disp != disp); /* Sanity check.  */
 		*ripdisp = disp;
 	}
+	p->opcode = *p->addr;
 }
 
-void arch_remove_kprobe(struct kprobe *p)
+void arch_arm_kprobe(struct kprobe *p)
 {
-	up(&kprobe_mutex);
-	free_insn_slot(p->ainsn.insn);
-	down(&kprobe_mutex);
+	*p->addr = BREAKPOINT_INSTRUCTION;
+	flush_icache_range((unsigned long) p->addr,
+			   (unsigned long) p->addr + sizeof(kprobe_opcode_t));
 }
 
-static inline void disarm_kprobe(struct kprobe *p, struct pt_regs *regs)
+void arch_disarm_kprobe(struct kprobe *p)
 {
 	*p->addr = p->opcode;
-	regs->rip = (unsigned long)p->addr;
+	flush_icache_range((unsigned long) p->addr,
+			   (unsigned long) p->addr + sizeof(kprobe_opcode_t));
+}
+
+void arch_remove_kprobe(struct kprobe *p)
+{
+	up(&kprobe_mutex);
+	free_insn_slot(p->ainsn.insn);
+	down(&kprobe_mutex);
 }
 
 static void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
@@ -311,7 +320,8 @@ int kprobe_handler(struct pt_regs *regs)
 				unlock_kprobes();
 				goto no_kprobe;
 			}
-			disarm_kprobe(p, regs);
+			arch_disarm_kprobe(p);
+			regs->rip = (unsigned long)p->addr;
 			ret = 1;
 		} else {
 			p = current_kprobe;
diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h
index fba39f87efe..0f90466fb8b 100644
--- a/include/linux/kprobes.h
+++ b/include/linux/kprobes.h
@@ -165,6 +165,8 @@ static inline int kprobe_running(void)
 
 extern int arch_prepare_kprobe(struct kprobe *p);
 extern void arch_copy_kprobe(struct kprobe *p);
+extern void arch_arm_kprobe(struct kprobe *p);
+extern void arch_disarm_kprobe(struct kprobe *p);
 extern void arch_remove_kprobe(struct kprobe *p);
 extern void show_registers(struct pt_regs *regs);
 
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 692fbf75ab4..e8e0ae8a6e1 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -261,7 +261,7 @@ static inline void free_rp_inst(struct kretprobe *rp)
 static inline void add_aggr_kprobe(struct kprobe *ap, struct kprobe *p)
 {
 	ap->addr = p->addr;
-	ap->opcode = p->opcode;
+	memcpy(&ap->opcode, &p->opcode, sizeof(kprobe_opcode_t));
 	memcpy(&ap->ainsn, &p->ainsn, sizeof(struct arch_specific_insn));
 
 	ap->pre_handler = aggr_pre_handler;
@@ -304,10 +304,8 @@ static int register_aggr_kprobe(struct kprobe *old_p, struct kprobe *p)
 /* kprobe removal house-keeping routines */
 static inline void cleanup_kprobe(struct kprobe *p, unsigned long flags)
 {
-	*p->addr = p->opcode;
+	arch_disarm_kprobe(p);
 	hlist_del(&p->hlist);
-	flush_icache_range((unsigned long) p->addr,
-		   (unsigned long) p->addr + sizeof(kprobe_opcode_t));
 	spin_unlock_irqrestore(&kprobe_lock, flags);
 	arch_remove_kprobe(p);
 }
@@ -344,10 +342,8 @@ int register_kprobe(struct kprobe *p)
 	hlist_add_head(&p->hlist,
 		       &kprobe_table[hash_ptr(p->addr, KPROBE_HASH_BITS)]);
 
-	p->opcode = *p->addr;
-	*p->addr = BREAKPOINT_INSTRUCTION;
-	flush_icache_range((unsigned long) p->addr,
-			   (unsigned long) p->addr + sizeof(kprobe_opcode_t));
+  	arch_arm_kprobe(p);
+
 out:
 	spin_unlock_irqrestore(&kprobe_lock, flags);
 rm_kprobe:
-- 
cgit v1.2.3-70-g09d2


From 0aa55e4d7db822059fe8132fe9f2b7773c48216c Mon Sep 17 00:00:00 2001
From: Hien Nguyen <hien@us.ibm.com>
Date: Thu, 23 Jun 2005 00:09:26 -0700
Subject: [PATCH] kprobes: moves lock-unlock to non-arch kprobe_flush_task

This patch moves the lock/unlock of the arch specific kprobe_flush_task()
to the non-arch specific kprobe_flusk_task().

Signed-off-by: Hien Nguyen <hien@us.ibm.com>
Acked-by: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/kprobes.c | 5 +----
 include/linux/kprobes.h    | 3 +--
 kernel/kprobes.c           | 5 ++++-
 3 files changed, 6 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c
index 2314d8d306f..b8e2bae0ab4 100644
--- a/arch/i386/kernel/kprobes.c
+++ b/arch/i386/kernel/kprobes.c
@@ -138,17 +138,14 @@ void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs)
 	}
 }
 
-void arch_kprobe_flush_task(struct task_struct *tk, spinlock_t *kp_lock)
+void arch_kprobe_flush_task(struct task_struct *tk)
 {
-	unsigned long flags = 0;
 	struct kretprobe_instance *ri;
-	spin_lock_irqsave(kp_lock, flags);
 	while ((ri = get_rp_inst_tsk(tk)) != NULL) {
 		*((unsigned long *)(ri->stack_addr)) =
 					(unsigned long) ri->ret_addr;
 		recycle_rp_inst(ri);
 	}
-	spin_unlock_irqrestore(kp_lock, flags);
 }
 
 /*
diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h
index 0f90466fb8b..461391decc4 100644
--- a/include/linux/kprobes.h
+++ b/include/linux/kprobes.h
@@ -33,7 +33,6 @@
 #include <linux/list.h>
 #include <linux/notifier.h>
 #include <linux/smp.h>
-#include <linux/spinlock.h>
 
 #include <asm/kprobes.h>
 
@@ -101,7 +100,7 @@ extern void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs,
 							unsigned long flags);
 extern struct task_struct *arch_get_kprobe_task(void *ptr);
 extern void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs);
-extern void arch_kprobe_flush_task(struct task_struct *tk, spinlock_t *kp_lock);
+extern void arch_kprobe_flush_task(struct task_struct *tk);
 #else /* ARCH_SUPPORTS_KRETPROBES */
 static inline void kretprobe_trampoline(void)
 {
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index e8e0ae8a6e1..dd42e717dd3 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -229,7 +229,10 @@ struct kretprobe_instance *get_rp_inst_tsk(struct task_struct *tk)
  */
 void kprobe_flush_task(struct task_struct *tk)
 {
-	arch_kprobe_flush_task(tk, &kprobe_lock);
+	unsigned long flags = 0;
+	spin_lock_irqsave(&kprobe_lock, flags);
+	arch_kprobe_flush_task(tk);
+	spin_unlock_irqrestore(&kprobe_lock, flags);
 }
 
 /*
-- 
cgit v1.2.3-70-g09d2


From 7213b2521889eb087eed8abaa48d1a692575da3e Mon Sep 17 00:00:00 2001
From: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Date: Thu, 23 Jun 2005 00:09:27 -0700
Subject: [PATCH] Kprobes/IA64: kdebug die notification mechanism

As many of you know that kprobes exist in the main line kernel for various
architecture including i386, x86_64, ppc64 and sparc64.  Attached patches
following this mail are a port of Kprobes and Jprobes for IA64.

I have tesed this patches for kprobes and Jprobes and this seems to work fine.
 I have tested this patch by inserting kprobes on various slots and various
templates including various types of branch instructions.

I have also tested this patch using the tool
http://marc.theaimsgroup.com/?l=linux-kernel&m=111657358022586&w=2 and the
kprobes for IA64 works great.

Here is list of TODO things and pathes for the same will appear soon.

1) Support kprobes on "mov r1=ip" type of instruction
2) Support Kprobes and Jprobes to exist on the same address
3) Support Return probes
3) Architecture independent cleanup of kprobes

This patch adds the kdebug die notification mechanism needed by Kprobes.

For break instruction on Branch type slot, imm21 is ignored and value
zero is placed in IIM register, hence we need to handle kprobes
for switch case zero.

Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Rusty Lynch <Rusty.lynch@intel.com>

From: Rusty Lynch <rusty.lynch@intel.com>

At the point in traps.c where we recieve a break with a zero value, we can
not say if the break was a result of a kprobe or some other debug facility.

This simple patch changes the informational string to a more correct "break
0" value, and applies to the 2.6.12-rc2-mm2 tree with all the kprobes
patches that were just recently included for the next mm cut.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/traps.c  | 33 ++++++++++++++++++++++++-
 arch/ia64/mm/fault.c      |  8 +++++++
 include/asm-ia64/break.h  |  2 ++
 include/asm-ia64/kdebug.h | 61 +++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 103 insertions(+), 1 deletion(-)
 create mode 100644 include/asm-ia64/kdebug.h

(limited to 'include')

diff --git a/arch/ia64/kernel/traps.c b/arch/ia64/kernel/traps.c
index 1861173bd4f..e7e520d90f0 100644
--- a/arch/ia64/kernel/traps.c
+++ b/arch/ia64/kernel/traps.c
@@ -21,12 +21,26 @@
 #include <asm/intrinsics.h>
 #include <asm/processor.h>
 #include <asm/uaccess.h>
+#include <asm/kdebug.h>
 
 extern spinlock_t timerlist_lock;
 
 fpswa_interface_t *fpswa_interface;
 EXPORT_SYMBOL(fpswa_interface);
 
+struct notifier_block *ia64die_chain;
+static DEFINE_SPINLOCK(die_notifier_lock);
+
+int register_die_notifier(struct notifier_block *nb)
+{
+	int err = 0;
+	unsigned long flags;
+	spin_lock_irqsave(&die_notifier_lock, flags);
+	err = notifier_chain_register(&ia64die_chain, nb);
+	spin_unlock_irqrestore(&die_notifier_lock, flags);
+	return err;
+}
+
 void __init
 trap_init (void)
 {
@@ -137,6 +151,10 @@ ia64_bad_break (unsigned long break_num, struct pt_regs *regs)
 
 	switch (break_num) {
 	      case 0: /* unknown error (used by GCC for __builtin_abort()) */
+		if (notify_die(DIE_BREAK, "break 0", regs, break_num, TRAP_BRKPT, SIGTRAP)
+			       	== NOTIFY_STOP) {
+			return;
+		}
 		die_if_kernel("bugcheck!", regs, break_num);
 		sig = SIGILL; code = ILL_ILLOPC;
 		break;
@@ -189,6 +207,15 @@ ia64_bad_break (unsigned long break_num, struct pt_regs *regs)
 		sig = SIGILL; code = __ILL_BNDMOD;
 		break;
 
+	      case 0x80200:
+	      case 0x80300:
+		if (notify_die(DIE_BREAK, "kprobe", regs, break_num, TRAP_BRKPT, SIGTRAP)
+			       	== NOTIFY_STOP) {
+			return;
+		}
+		sig = SIGTRAP; code = TRAP_BRKPT;
+		break;
+
 	      default:
 		if (break_num < 0x40000 || break_num > 0x100000)
 			die_if_kernel("Bad break", regs, break_num);
@@ -548,7 +575,11 @@ ia64_fault (unsigned long vector, unsigned long isr, unsigned long ifa,
 #endif
 			break;
 		      case 35: siginfo.si_code = TRAP_BRANCH; ifa = 0; break;
-		      case 36: siginfo.si_code = TRAP_TRACE; ifa = 0; break;
+		      case 36:
+			      if (notify_die(DIE_SS, "ss", &regs, vector,
+					     vector, SIGTRAP) == NOTIFY_STOP)
+				      return;
+			      siginfo.si_code = TRAP_TRACE; ifa = 0; break;
 		}
 		siginfo.si_signo = SIGTRAP;
 		siginfo.si_errno = 0;
diff --git a/arch/ia64/mm/fault.c b/arch/ia64/mm/fault.c
index 4174ec999dd..ff62551eb3a 100644
--- a/arch/ia64/mm/fault.c
+++ b/arch/ia64/mm/fault.c
@@ -14,6 +14,7 @@
 #include <asm/processor.h>
 #include <asm/system.h>
 #include <asm/uaccess.h>
+#include <asm/kdebug.h>
 
 extern void die (char *, struct pt_regs *, long);
 
@@ -102,6 +103,13 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re
 		goto bad_area_no_up;
 #endif
 
+	/*
+	 * This is to handle the kprobes on user space access instructions
+	 */
+	if (notify_die(DIE_PAGE_FAULT, "page fault", regs, code, TRAP_BRKPT,
+					SIGSEGV) == NOTIFY_STOP)
+		return;
+
 	down_read(&mm->mmap_sem);
 
 	vma = find_vma_prev(mm, address, &prev_vma);
diff --git a/include/asm-ia64/break.h b/include/asm-ia64/break.h
index 97c7b2d7960..8167828edc4 100644
--- a/include/asm-ia64/break.h
+++ b/include/asm-ia64/break.h
@@ -12,6 +12,8 @@
  * OS-specific debug break numbers:
  */
 #define __IA64_BREAK_KDB		0x80100
+#define __IA64_BREAK_KPROBE		0x80200
+#define __IA64_BREAK_JPROBE		0x80300
 
 /*
  * OS-specific break numbers:
diff --git a/include/asm-ia64/kdebug.h b/include/asm-ia64/kdebug.h
new file mode 100644
index 00000000000..4d376e1663f
--- /dev/null
+++ b/include/asm-ia64/kdebug.h
@@ -0,0 +1,61 @@
+#ifndef _IA64_KDEBUG_H
+#define _IA64_KDEBUG_H 1
+/*
+ * include/asm-ia64/kdebug.h
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) Intel Corporation, 2005
+ *
+ * 2005-Apr     Rusty Lynch <rusty.lynch@intel.com> and Anil S Keshavamurthy
+ *              <anil.s.keshavamurthy@intel.com> adopted from
+ *              include/asm-x86_64/kdebug.h
+ */
+#include <linux/notifier.h>
+
+struct pt_regs;
+
+struct die_args {
+	struct pt_regs *regs;
+	const char *str;
+	long err;
+	int trapnr;
+	int signr;
+};
+
+int register_die_notifier(struct notifier_block *nb);
+extern struct notifier_block *ia64die_chain;
+
+enum die_val {
+	DIE_BREAK = 1,
+	DIE_SS,
+	DIE_PAGE_FAULT,
+};
+
+static inline int notify_die(enum die_val val, char *str, struct pt_regs *regs,
+			     long err, int trap, int sig)
+{
+	struct die_args args = {
+		.regs   = regs,
+		.str    = str,
+		.err    = err,
+		.trapnr = trap,
+		.signr  = sig
+	};
+
+	return notifier_call_chain(&ia64die_chain, val, &args);
+}
+
+#endif
-- 
cgit v1.2.3-70-g09d2


From fd7b231ff98578308d8f5fb76a25a369ce1074ae Mon Sep 17 00:00:00 2001
From: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Date: Thu, 23 Jun 2005 00:09:28 -0700
Subject: [PATCH] Kprobes/IA64: arch specific handling

This is an IA64 arch specific handling of Kprobes

Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Rusty Lynch <Rusty.lynch@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/Kconfig.debug    |  11 ++
 arch/ia64/kernel/Makefile  |   1 +
 arch/ia64/kernel/kprobes.c | 332 +++++++++++++++++++++++++++++++++++++++++++++
 include/asm-ia64/kprobes.h |  72 ++++++++++
 4 files changed, 416 insertions(+)
 create mode 100644 arch/ia64/kernel/kprobes.c
 create mode 100644 include/asm-ia64/kprobes.h

(limited to 'include')

diff --git a/arch/ia64/Kconfig.debug b/arch/ia64/Kconfig.debug
index de9d507ba0f..fda67ac993d 100644
--- a/arch/ia64/Kconfig.debug
+++ b/arch/ia64/Kconfig.debug
@@ -2,6 +2,17 @@ menu "Kernel hacking"
 
 source "lib/Kconfig.debug"
 
+config KPROBES
+        bool "Kprobes"
+        depends on DEBUG_KERNEL
+        help
+          Kprobes allows you to trap at almost any kernel address and
+          execute a callback function.  register_kprobe() establishes
+          a probepoint and specifies the callback.  Kprobes is useful
+          for kernel debugging, non-intrusive instrumentation and testing.
+          If in doubt, say "N".
+
+
 choice
 	prompt "Physical memory granularity"
 	default IA64_GRANULE_64MB
diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile
index 4c73d8ba2e3..5290fa4c337 100644
--- a/arch/ia64/kernel/Makefile
+++ b/arch/ia64/kernel/Makefile
@@ -20,6 +20,7 @@ obj-$(CONFIG_SMP)		+= smp.o smpboot.o domain.o
 obj-$(CONFIG_PERFMON)		+= perfmon_default_smpl.o
 obj-$(CONFIG_IA64_CYCLONE)	+= cyclone.o
 obj-$(CONFIG_IA64_MCA_RECOVERY)	+= mca_recovery.o
+obj-$(CONFIG_KPROBES)		+= kprobes.o
 obj-$(CONFIG_IA64_UNCACHED_ALLOCATOR)	+= uncached.o
 mca_recovery-y			+= mca_drv.o mca_drv_asm.o
 
diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
new file mode 100644
index 00000000000..81a53f99a57
--- /dev/null
+++ b/arch/ia64/kernel/kprobes.c
@@ -0,0 +1,332 @@
+/*
+ *  Kernel Probes (KProbes)
+ *  arch/ia64/kernel/kprobes.c
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2002, 2004
+ * Copyright (C) Intel Corporation, 2005
+ *
+ * 2005-Apr     Rusty Lynch <rusty.lynch@intel.com> and Anil S Keshavamurthy
+ *              <anil.s.keshavamurthy@intel.com> adapted from i386
+ */
+
+#include <linux/config.h>
+#include <linux/kprobes.h>
+#include <linux/ptrace.h>
+#include <linux/spinlock.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/preempt.h>
+#include <linux/moduleloader.h>
+
+#include <asm/pgtable.h>
+#include <asm/kdebug.h>
+
+/* kprobe_status settings */
+#define KPROBE_HIT_ACTIVE	0x00000001
+#define KPROBE_HIT_SS		0x00000002
+
+static struct kprobe *current_kprobe;
+static unsigned long kprobe_status;
+
+enum instruction_type {A, I, M, F, B, L, X, u};
+static enum instruction_type bundle_encoding[32][3] = {
+  { M, I, I },				/* 00 */
+  { M, I, I },				/* 01 */
+  { M, I, I },				/* 02 */
+  { M, I, I },				/* 03 */
+  { M, L, X },				/* 04 */
+  { M, L, X },				/* 05 */
+  { u, u, u },  			/* 06 */
+  { u, u, u },  			/* 07 */
+  { M, M, I },				/* 08 */
+  { M, M, I },				/* 09 */
+  { M, M, I },				/* 0A */
+  { M, M, I },				/* 0B */
+  { M, F, I },				/* 0C */
+  { M, F, I },				/* 0D */
+  { M, M, F },				/* 0E */
+  { M, M, F },				/* 0F */
+  { M, I, B },				/* 10 */
+  { M, I, B },				/* 11 */
+  { M, B, B },				/* 12 */
+  { M, B, B },				/* 13 */
+  { u, u, u },  			/* 14 */
+  { u, u, u },  			/* 15 */
+  { B, B, B },				/* 16 */
+  { B, B, B },				/* 17 */
+  { M, M, B },				/* 18 */
+  { M, M, B },				/* 19 */
+  { u, u, u },  			/* 1A */
+  { u, u, u },  			/* 1B */
+  { M, F, B },				/* 1C */
+  { M, F, B },				/* 1D */
+  { u, u, u },  			/* 1E */
+  { u, u, u },  			/* 1F */
+};
+
+int arch_prepare_kprobe(struct kprobe *p)
+{
+	unsigned long addr = (unsigned long) p->addr;
+	unsigned long bundle_addr = addr & ~0xFULL;
+	unsigned long slot = addr & 0xf;
+	bundle_t bundle;
+	unsigned long template;
+
+	/*
+	 * TODO: Verify that a probe is not being inserted
+	 *       in sensitive regions of code
+	 * TODO: Verify that the memory holding the probe is rwx
+	 * TODO: verify this is a kernel address
+	 */
+	memcpy(&bundle, (unsigned long *)bundle_addr, sizeof(bundle_t));
+	template = bundle.quad0.template;
+	if (((bundle_encoding[template][1] == L) && slot > 1) || (slot > 2)) {
+		printk(KERN_WARNING "Attempting to insert unaligned kprobe at 0x%lx\n", addr);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+void arch_copy_kprobe(struct kprobe *p)
+{
+	unsigned long addr = (unsigned long)p->addr;
+	unsigned long bundle_addr = addr & ~0xFULL;
+
+	memcpy(&p->ainsn.insn.bundle, (unsigned long *)bundle_addr,
+				sizeof(bundle_t));
+	memcpy(&p->opcode.bundle, &p->ainsn.insn.bundle, sizeof(bundle_t));
+}
+
+void arch_arm_kprobe(struct kprobe *p)
+{
+	unsigned long addr = (unsigned long)p->addr;
+	unsigned long arm_addr = addr & ~0xFULL;
+	unsigned long slot = addr & 0xf;
+	unsigned long template;
+	bundle_t bundle;
+
+	memcpy(&bundle, &p->ainsn.insn.bundle, sizeof(bundle_t));
+
+	template = bundle.quad0.template;
+	if (slot == 1 && bundle_encoding[template][1] == L)
+		slot = 2;
+	switch (slot) {
+	case 0:
+		bundle.quad0.slot0 = BREAK_INST;
+		break;
+	case 1:
+		bundle.quad0.slot1_p0 = BREAK_INST;
+		bundle.quad1.slot1_p1 = (BREAK_INST >> (64-46));
+		break;
+	case 2:
+		bundle.quad1.slot2 = BREAK_INST;
+		break;
+	}
+
+ 	/* Flush icache for the instruction at the emulated address */
+	flush_icache_range((unsigned long)&p->ainsn.insn.bundle,
+			(unsigned long)&p->ainsn.insn.bundle +
+			sizeof(bundle_t));
+	/*
+	 * Patch the original instruction with the probe instruction
+	 * and flush the instruction cache
+	 */
+	memcpy((char *) arm_addr, (char *) &bundle, sizeof(bundle_t));
+	flush_icache_range(arm_addr, arm_addr + sizeof(bundle_t));
+}
+
+void arch_disarm_kprobe(struct kprobe *p)
+{
+	unsigned long addr = (unsigned long)p->addr;
+	unsigned long arm_addr = addr & ~0xFULL;
+
+	/* p->opcode contains the original unaltered bundle */
+	memcpy((char *) arm_addr, (char *) &p->opcode.bundle, sizeof(bundle_t));
+	flush_icache_range(arm_addr, arm_addr + sizeof(bundle_t));
+}
+
+void arch_remove_kprobe(struct kprobe *p)
+{
+}
+
+/*
+ * We are resuming execution after a single step fault, so the pt_regs
+ * structure reflects the register state after we executed the instruction
+ * located in the kprobe (p->ainsn.insn.bundle).  We still need to adjust
+ * the ip to point back to the original stack address, and if we see that
+ * the slot has incremented back to zero, then we need to point to the next
+ * slot location.
+ */
+static void resume_execution(struct kprobe *p, struct pt_regs *regs)
+{
+	unsigned long bundle = (unsigned long)p->addr & ~0xFULL;
+
+	/*
+	 * TODO: Handle cases where kprobe was inserted on a branch instruction
+	 */
+
+	if (!ia64_psr(regs)->ri)
+		regs->cr_iip = bundle + 0x10;
+	else
+		regs->cr_iip = bundle;
+
+	ia64_psr(regs)->ss = 0;
+}
+
+static void prepare_ss(struct kprobe *p, struct pt_regs *regs)
+{
+	unsigned long bundle_addr = (unsigned long) &p->ainsn.insn.bundle;
+	unsigned long slot = (unsigned long)p->addr & 0xf;
+
+	/* Update instruction pointer (IIP) and slot number (IPSR.ri) */
+	regs->cr_iip = bundle_addr & ~0xFULL;
+
+	if (slot > 2)
+		slot = 0;
+
+	ia64_psr(regs)->ri = slot;
+
+	/* turn on single stepping */
+	ia64_psr(regs)->ss = 1;
+}
+
+static int pre_kprobes_handler(struct pt_regs *regs)
+{
+	struct kprobe *p;
+	int ret = 0;
+	kprobe_opcode_t *addr = (kprobe_opcode_t *)instruction_pointer(regs);
+
+	preempt_disable();
+
+	/* Handle recursion cases */
+	if (kprobe_running()) {
+		p = get_kprobe(addr);
+		if (p) {
+			if (kprobe_status == KPROBE_HIT_SS) {
+				unlock_kprobes();
+				goto no_kprobe;
+			}
+			arch_disarm_kprobe(p);
+			ret = 1;
+		} else {
+			/*
+			 * jprobe instrumented function just completed
+			 */
+			p = current_kprobe;
+			if (p->break_handler && p->break_handler(p, regs)) {
+				goto ss_probe;
+			}
+		}
+	}
+
+	lock_kprobes();
+	p = get_kprobe(addr);
+	if (!p) {
+		unlock_kprobes();
+		goto no_kprobe;
+	}
+
+	kprobe_status = KPROBE_HIT_ACTIVE;
+	current_kprobe = p;
+
+	if (p->pre_handler && p->pre_handler(p, regs))
+		/*
+		 * Our pre-handler is specifically requesting that we just
+		 * do a return.  This is handling the case where the
+		 * pre-handler is really our special jprobe pre-handler.
+		 */
+		return 1;
+
+ss_probe:
+	prepare_ss(p, regs);
+	kprobe_status = KPROBE_HIT_SS;
+	return 1;
+
+no_kprobe:
+	preempt_enable_no_resched();
+	return ret;
+}
+
+static int post_kprobes_handler(struct pt_regs *regs)
+{
+	if (!kprobe_running())
+		return 0;
+
+	if (current_kprobe->post_handler)
+		current_kprobe->post_handler(current_kprobe, regs, 0);
+
+	resume_execution(current_kprobe, regs);
+
+	unlock_kprobes();
+	preempt_enable_no_resched();
+	return 1;
+}
+
+static int kprobes_fault_handler(struct pt_regs *regs, int trapnr)
+{
+	if (!kprobe_running())
+		return 0;
+
+	if (current_kprobe->fault_handler &&
+	    current_kprobe->fault_handler(current_kprobe, regs, trapnr))
+		return 1;
+
+	if (kprobe_status & KPROBE_HIT_SS) {
+		resume_execution(current_kprobe, regs);
+		unlock_kprobes();
+		preempt_enable_no_resched();
+	}
+
+	return 0;
+}
+
+int kprobe_exceptions_notify(struct notifier_block *self, unsigned long val,
+			     void *data)
+{
+	struct die_args *args = (struct die_args *)data;
+	switch(val) {
+	case DIE_BREAK:
+		if (pre_kprobes_handler(args->regs))
+			return NOTIFY_STOP;
+		break;
+	case DIE_SS:
+		if (post_kprobes_handler(args->regs))
+			return NOTIFY_STOP;
+		break;
+	case DIE_PAGE_FAULT:
+		if (kprobes_fault_handler(args->regs, args->trapnr))
+			return NOTIFY_STOP;
+	default:
+		break;
+	}
+	return NOTIFY_DONE;
+}
+
+int setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	printk(KERN_WARNING "Jprobes is not supported\n");
+	return 0;
+}
+
+void jprobe_return(void)
+{
+}
+
+int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	return 0;
+}
diff --git a/include/asm-ia64/kprobes.h b/include/asm-ia64/kprobes.h
new file mode 100644
index 00000000000..2a6f2a14889
--- /dev/null
+++ b/include/asm-ia64/kprobes.h
@@ -0,0 +1,72 @@
+#ifndef _ASM_KPROBES_H
+#define _ASM_KPROBES_H
+/*
+ *  Kernel Probes (KProbes)
+ *  include/asm-ia64/kprobes.h
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2002, 2004
+ * Copyright (C) Intel Corporation, 2005
+ *
+ * 2005-Apr     Rusty Lynch <rusty.lynch@intel.com> and Anil S Keshavamurthy
+ *              <anil.s.keshavamurthy@intel.com> adapted from i386
+ */
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <asm/break.h>
+
+#define BREAK_INST	(long)(__IA64_BREAK_KPROBE << 6)
+
+typedef struct _bundle {
+	struct {
+		unsigned long long template : 5;
+		unsigned long long slot0 : 41;
+		unsigned long long slot1_p0 : 64-46;
+	} quad0;
+	struct {
+		unsigned long long slot1_p1 : 41 - (64-46);
+		unsigned long long slot2 : 41;
+	} quad1;
+} __attribute__((__aligned__(16)))  bundle_t;
+
+#define JPROBE_ENTRY(pentry)	(kprobe_opcode_t *)pentry
+
+typedef struct kprobe_opcode {
+	bundle_t bundle;
+} kprobe_opcode_t;
+
+struct fnptr {
+	unsigned long ip;
+	unsigned long gp;
+};
+
+/* Architecture specific copy of original instruction*/
+struct arch_specific_insn {
+	/* copy of the original instruction */
+	kprobe_opcode_t insn;
+};
+
+#ifdef CONFIG_KPROBES
+extern int kprobe_exceptions_notify(struct notifier_block *self,
+				    unsigned long val, void *data);
+#else				/* !CONFIG_KPROBES */
+static inline int kprobe_exceptions_notify(struct notifier_block *self,
+					   unsigned long val, void *data)
+{
+	return 0;
+}
+#endif
+#endif				/* _ASM_KPROBES_H */
-- 
cgit v1.2.3-70-g09d2


From b2761dc262b428475890fffd979687051beb12ba Mon Sep 17 00:00:00 2001
From: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Date: Thu, 23 Jun 2005 00:09:28 -0700
Subject: [PATCH] Kprobes/IA64: architecture specific JProbes support

This patch adds IA64 architecture specific JProbes support on top of Kprobes

Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Rusty Lynch <Rusty.lynch@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/Makefile  |  2 +-
 arch/ia64/kernel/jprobes.S | 61 ++++++++++++++++++++++++++++++++++++++++++++++
 arch/ia64/kernel/kprobes.c | 28 ++++++++++++++++-----
 include/asm-ia64/kprobes.h |  5 ++++
 4 files changed, 89 insertions(+), 7 deletions(-)
 create mode 100644 arch/ia64/kernel/jprobes.S

(limited to 'include')

diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile
index 5290fa4c337..b2e2f6509eb 100644
--- a/arch/ia64/kernel/Makefile
+++ b/arch/ia64/kernel/Makefile
@@ -20,7 +20,7 @@ obj-$(CONFIG_SMP)		+= smp.o smpboot.o domain.o
 obj-$(CONFIG_PERFMON)		+= perfmon_default_smpl.o
 obj-$(CONFIG_IA64_CYCLONE)	+= cyclone.o
 obj-$(CONFIG_IA64_MCA_RECOVERY)	+= mca_recovery.o
-obj-$(CONFIG_KPROBES)		+= kprobes.o
+obj-$(CONFIG_KPROBES)		+= kprobes.o jprobes.o
 obj-$(CONFIG_IA64_UNCACHED_ALLOCATOR)	+= uncached.o
 mca_recovery-y			+= mca_drv.o mca_drv_asm.o
 
diff --git a/arch/ia64/kernel/jprobes.S b/arch/ia64/kernel/jprobes.S
new file mode 100644
index 00000000000..b7fa3ccd2b0
--- /dev/null
+++ b/arch/ia64/kernel/jprobes.S
@@ -0,0 +1,61 @@
+/*
+ * Jprobe specific operations
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) Intel Corporation, 2005
+ *
+ * 2005-May     Rusty Lynch <rusty.lynch@intel.com> and Anil S Keshavamurthy
+ *              <anil.s.keshavamurthy@intel.com> initial implementation
+ *
+ * Jprobes (a.k.a. "jump probes" which is built on-top of kprobes) allow a
+ * probe to be inserted into the beginning of a function call.  The fundamental
+ * difference between a jprobe and a kprobe is the jprobe handler is executed
+ * in the same context as the target function, while the kprobe handlers
+ * are executed in interrupt context.
+ *
+ * For jprobes we initially gain control by placing a break point in the
+ * first instruction of the targeted function.  When we catch that specific
+ * break, we:
+ *        * set the return address to our jprobe_inst_return() function
+ *        * jump to the jprobe handler function
+ *
+ * Since we fixed up the return address, the jprobe handler will return to our
+ * jprobe_inst_return() function, giving us control again.  At this point we
+ * are back in the parents frame marker, so we do yet another call to our
+ * jprobe_break() function to fix up the frame marker as it would normally
+ * exist in the target function.
+ *
+ * Our jprobe_return function then transfers control back to kprobes.c by
+ * executing a break instruction using one of our reserved numbers.  When we
+ * catch that break in kprobes.c, we continue like we do for a normal kprobe
+ * by single stepping the emulated instruction, and then returning execution
+ * to the correct location.
+ */
+#include <asm/asmmacro.h>
+
+	/*
+	 * void jprobe_break(void)
+	 */
+ENTRY(jprobe_break)
+	break.m 0x80300
+END(jprobe_break)
+
+	/*
+	 * void jprobe_inst_return(void)
+	 */
+GLOBAL_ENTRY(jprobe_inst_return)
+	br.call.sptk.many b0=jprobe_break
+END(jprobe_inst_return)
diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index 81a53f99a57..6683a44f419 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -35,12 +35,15 @@
 #include <asm/pgtable.h>
 #include <asm/kdebug.h>
 
+extern void jprobe_inst_return(void);
+
 /* kprobe_status settings */
 #define KPROBE_HIT_ACTIVE	0x00000001
 #define KPROBE_HIT_SS		0x00000002
 
 static struct kprobe *current_kprobe;
 static unsigned long kprobe_status;
+static struct pt_regs jprobe_saved_regs;
 
 enum instruction_type {A, I, M, F, B, L, X, u};
 static enum instruction_type bundle_encoding[32][3] = {
@@ -318,15 +321,28 @@ int kprobe_exceptions_notify(struct notifier_block *self, unsigned long val,
 
 int setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
 {
-	printk(KERN_WARNING "Jprobes is not supported\n");
-	return 0;
-}
+	struct jprobe *jp = container_of(p, struct jprobe, kp);
+	unsigned long addr = ((struct fnptr *)(jp->entry))->ip;
 
-void jprobe_return(void)
-{
+	/* save architectural state */
+	jprobe_saved_regs = *regs;
+
+	/* after rfi, execute the jprobe instrumented function */
+	regs->cr_iip = addr & ~0xFULL;
+	ia64_psr(regs)->ri = addr & 0xf;
+	regs->r1 = ((struct fnptr *)(jp->entry))->gp;
+
+	/*
+	 * fix the return address to our jprobe_inst_return() function
+	 * in the jprobes.S file
+	 */
+ 	regs->b0 = ((struct fnptr *)(jprobe_inst_return))->ip;
+
+	return 1;
 }
 
 int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 {
-	return 0;
+	*regs = jprobe_saved_regs;
+	return 1;
 }
diff --git a/include/asm-ia64/kprobes.h b/include/asm-ia64/kprobes.h
index 2a6f2a14889..fec3506e53f 100644
--- a/include/asm-ia64/kprobes.h
+++ b/include/asm-ia64/kprobes.h
@@ -59,6 +59,11 @@ struct arch_specific_insn {
 	kprobe_opcode_t insn;
 };
 
+/* ia64 does not need this */
+static inline void jprobe_return(void)
+{
+}
+
 #ifdef CONFIG_KPROBES
 extern int kprobe_exceptions_notify(struct notifier_block *self,
 				    unsigned long val, void *data);
-- 
cgit v1.2.3-70-g09d2


From cd2675bf65455a45b54228b7acc0c6a26a164cb6 Mon Sep 17 00:00:00 2001
From: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Date: Thu, 23 Jun 2005 00:09:29 -0700
Subject: [PATCH] Kprobes/IA64: support kprobe on branch/call instructions

This patch is required to support kprobe on branch/call instructions.

Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/kprobes.c | 131 +++++++++++++++++++++++++++++++++++++++------
 include/asm-ia64/kprobes.h |  17 +++++-
 2 files changed, 132 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index 6683a44f419..20a250e2e9b 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -120,25 +120,75 @@ void arch_arm_kprobe(struct kprobe *p)
 	unsigned long arm_addr = addr & ~0xFULL;
 	unsigned long slot = addr & 0xf;
 	unsigned long template;
+    	unsigned long major_opcode = 0;
+    	unsigned long lx_type_inst = 0;
+    	unsigned long kprobe_inst = 0;
 	bundle_t bundle;
 
+   	p->ainsn.inst_flag = 0;
+   	p->ainsn.target_br_reg = 0;
+
 	memcpy(&bundle, &p->ainsn.insn.bundle, sizeof(bundle_t));
+    	template = bundle.quad0.template;
+    	if (slot == 1 && bundle_encoding[template][1] == L) {
+    		lx_type_inst = 1;
+     		slot = 2;
+    	}
+
 
-	template = bundle.quad0.template;
-	if (slot == 1 && bundle_encoding[template][1] == L)
-		slot = 2;
 	switch (slot) {
 	case 0:
+   		major_opcode = (bundle.quad0.slot0 >> SLOT0_OPCODE_SHIFT);
+    		kprobe_inst = bundle.quad0.slot0;
 		bundle.quad0.slot0 = BREAK_INST;
 		break;
 	case 1:
+    		major_opcode = (bundle.quad1.slot1_p1 >> SLOT1_p1_OPCODE_SHIFT);
+    		kprobe_inst = (bundle.quad0.slot1_p0 |
+    				(bundle.quad1.slot1_p1 << (64-46)));
 		bundle.quad0.slot1_p0 = BREAK_INST;
 		bundle.quad1.slot1_p1 = (BREAK_INST >> (64-46));
 		break;
 	case 2:
+    		major_opcode = (bundle.quad1.slot2 >> SLOT2_OPCODE_SHIFT);
+    		kprobe_inst = bundle.quad1.slot2;
 		bundle.quad1.slot2 = BREAK_INST;
 		break;
 	}
+    	/*
+    	 * Look for IP relative Branches, IP relative call or
+    	 * IP relative predicate instructions
+    	 */
+    	if (bundle_encoding[template][slot] == B) {
+    		switch (major_opcode) {
+    			case INDIRECT_CALL_OPCODE:
+    				p->ainsn.inst_flag |= INST_FLAG_FIX_BRANCH_REG;
+    				p->ainsn.target_br_reg = ((kprobe_inst >> 6) & 0x7);
+    				break;
+    			case IP_RELATIVE_PREDICT_OPCODE:
+    			case IP_RELATIVE_BRANCH_OPCODE:
+    				p->ainsn.inst_flag |= INST_FLAG_FIX_RELATIVE_IP_ADDR;
+    				break;
+    			case IP_RELATIVE_CALL_OPCODE:
+    				p->ainsn.inst_flag |= INST_FLAG_FIX_RELATIVE_IP_ADDR;
+    				p->ainsn.inst_flag |= INST_FLAG_FIX_BRANCH_REG;
+    				p->ainsn.target_br_reg = ((kprobe_inst >> 6) & 0x7);
+    				break;
+    			default:
+    				/* Do nothing */
+    				break;
+    		}
+    	} else if (lx_type_inst) {
+    		switch (major_opcode) {
+    			case LONG_CALL_OPCODE:
+    				p->ainsn.inst_flag |= INST_FLAG_FIX_BRANCH_REG;
+    				p->ainsn.target_br_reg = ((kprobe_inst >> 6) & 0x7);
+   				break;
+    			default:
+    				/* Do nothing */
+    				break;
+    		}
+    	}
 
  	/* Flush icache for the instruction at the emulated address */
 	flush_icache_range((unsigned long)&p->ainsn.insn.bundle,
@@ -170,24 +220,75 @@ void arch_remove_kprobe(struct kprobe *p)
  * We are resuming execution after a single step fault, so the pt_regs
  * structure reflects the register state after we executed the instruction
  * located in the kprobe (p->ainsn.insn.bundle).  We still need to adjust
- * the ip to point back to the original stack address, and if we see that
- * the slot has incremented back to zero, then we need to point to the next
- * slot location.
+ * the ip to point back to the original stack address. To set the IP address
+ * to original stack address, handle the case where we need to fixup the
+ * relative IP address and/or fixup branch register.
  */
 static void resume_execution(struct kprobe *p, struct pt_regs *regs)
 {
-	unsigned long bundle = (unsigned long)p->addr & ~0xFULL;
+  	unsigned long bundle_addr = ((unsigned long) (&p->ainsn.insn.bundle)) & ~0xFULL;
+  	unsigned long resume_addr = (unsigned long)p->addr & ~0xFULL;
+ 	unsigned long template;
+ 	int slot = ((unsigned long)p->addr & 0xf);
 
-	/*
-	 * TODO: Handle cases where kprobe was inserted on a branch instruction
-	 */
+	template = p->opcode.bundle.quad0.template;
+
+ 	if (slot == 1 && bundle_encoding[template][1] == L)
+ 		slot = 2;
+
+	if (p->ainsn.inst_flag) {
 
-	if (!ia64_psr(regs)->ri)
-		regs->cr_iip = bundle + 0x10;
-	else
-		regs->cr_iip = bundle;
+		if (p->ainsn.inst_flag & INST_FLAG_FIX_RELATIVE_IP_ADDR) {
+			/* Fix relative IP address */
+ 			regs->cr_iip = (regs->cr_iip - bundle_addr) + resume_addr;
+		}
+
+		if (p->ainsn.inst_flag & INST_FLAG_FIX_BRANCH_REG) {
+		/*
+		 * Fix target branch register, software convention is
+		 * to use either b0 or b6 or b7, so just checking
+		 * only those registers
+		 */
+			switch (p->ainsn.target_br_reg) {
+			case 0:
+				if ((regs->b0 == bundle_addr) ||
+					(regs->b0 == bundle_addr + 0x10)) {
+					regs->b0 = (regs->b0 - bundle_addr) +
+						resume_addr;
+				}
+				break;
+			case 6:
+				if ((regs->b6 == bundle_addr) ||
+					(regs->b6 == bundle_addr + 0x10)) {
+					regs->b6 = (regs->b6 - bundle_addr) +
+						resume_addr;
+				}
+				break;
+			case 7:
+				if ((regs->b7 == bundle_addr) ||
+					(regs->b7 == bundle_addr + 0x10)) {
+					regs->b7 = (regs->b7 - bundle_addr) +
+						resume_addr;
+				}
+				break;
+			} /* end switch */
+		}
+		goto turn_ss_off;
+	}
 
-	ia64_psr(regs)->ss = 0;
+	if (slot == 2) {
+ 		if (regs->cr_iip == bundle_addr + 0x10) {
+ 			regs->cr_iip = resume_addr + 0x10;
+ 		}
+ 	} else {
+ 		if (regs->cr_iip == bundle_addr) {
+ 			regs->cr_iip = resume_addr;
+ 		}
+ 	}
+
+turn_ss_off:
+  	/* Turn off Single Step bit */
+  	ia64_psr(regs)->ss = 0;
 }
 
 static void prepare_ss(struct kprobe *p, struct pt_regs *regs)
diff --git a/include/asm-ia64/kprobes.h b/include/asm-ia64/kprobes.h
index fec3506e53f..d30af77a0b1 100644
--- a/include/asm-ia64/kprobes.h
+++ b/include/asm-ia64/kprobes.h
@@ -44,6 +44,17 @@ typedef struct _bundle {
 
 #define JPROBE_ENTRY(pentry)	(kprobe_opcode_t *)pentry
 
+#define SLOT0_OPCODE_SHIFT	(37)
+#define SLOT1_p1_OPCODE_SHIFT	(37 - (64-46))
+#define SLOT2_OPCODE_SHIFT 	(37)
+
+#define INDIRECT_CALL_OPCODE		(1)
+#define IP_RELATIVE_CALL_OPCODE		(5)
+#define IP_RELATIVE_BRANCH_OPCODE	(4)
+#define IP_RELATIVE_PREDICT_OPCODE	(7)
+#define LONG_BRANCH_OPCODE		(0xC)
+#define LONG_CALL_OPCODE		(0xD)
+
 typedef struct kprobe_opcode {
 	bundle_t bundle;
 } kprobe_opcode_t;
@@ -55,8 +66,12 @@ struct fnptr {
 
 /* Architecture specific copy of original instruction*/
 struct arch_specific_insn {
-	/* copy of the original instruction */
+	/* copy of the instruction to be emulated */
 	kprobe_opcode_t insn;
+ #define INST_FLAG_FIX_RELATIVE_IP_ADDR		1
+ #define INST_FLAG_FIX_BRANCH_REG		2
+ 	unsigned long inst_flag;
+ 	unsigned short target_br_reg;
 };
 
 /* ia64 does not need this */
-- 
cgit v1.2.3-70-g09d2


From 8bc76772ad653bcaad1b0af72aafb6072ef0fa87 Mon Sep 17 00:00:00 2001
From: Rusty Lynch <rusty.lynch@intel.com>
Date: Thu, 23 Jun 2005 00:09:30 -0700
Subject: [PATCH] Kprobes ia64 cleanup

A cleanup of the ia64 kprobes implementation such that all of the bundle
manipulation logic is concentrated in arch_prepare_kprobe().

With the current design for kprobes, the arch specific code only has a
chance to return failure inside the arch_prepare_kprobe() function.

This patch moves all of the work that was happening in arch_copy_kprobe()
and most of the work that was happening in arch_arm_kprobe() into
arch_prepare_kprobe().  By doing this we can add further robustness checks
in arch_arm_kprobe() and refuse to insert kprobes that will cause problems.

Signed-off-by: Rusty Lynch <Rusty.lynch@intel.com>
Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/kprobes.c | 178 ++++++++++++++++++++-------------------------
 include/asm-ia64/kprobes.h |   7 ++
 2 files changed, 84 insertions(+), 101 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index 20a250e2e9b..b7a204137fb 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -84,121 +84,97 @@ static enum instruction_type bundle_encoding[32][3] = {
 int arch_prepare_kprobe(struct kprobe *p)
 {
 	unsigned long addr = (unsigned long) p->addr;
-	unsigned long bundle_addr = addr & ~0xFULL;
+	unsigned long *bundle_addr = (unsigned long *)(addr & ~0xFULL);
 	unsigned long slot = addr & 0xf;
-	bundle_t bundle;
 	unsigned long template;
+ 	unsigned long major_opcode = 0;
+ 	unsigned long lx_type_inst = 0;
+ 	unsigned long kprobe_inst = 0;
+	bundle_t *bundle = &p->ainsn.insn.bundle;
 
-	/*
-	 * TODO: Verify that a probe is not being inserted
-	 *       in sensitive regions of code
-	 * TODO: Verify that the memory holding the probe is rwx
-	 * TODO: verify this is a kernel address
-	 */
-	memcpy(&bundle, (unsigned long *)bundle_addr, sizeof(bundle_t));
-	template = bundle.quad0.template;
-	if (((bundle_encoding[template][1] == L) && slot > 1) || (slot > 2)) {
-		printk(KERN_WARNING "Attempting to insert unaligned kprobe at 0x%lx\n", addr);
-		return -EINVAL;
-	}
-	return 0;
-}
+	memcpy(&p->opcode.bundle, bundle_addr, sizeof(bundle_t));
+	memcpy(&p->ainsn.insn.bundle, bundle_addr, sizeof(bundle_t));
 
-void arch_copy_kprobe(struct kprobe *p)
-{
-	unsigned long addr = (unsigned long)p->addr;
-	unsigned long bundle_addr = addr & ~0xFULL;
+	p->ainsn.inst_flag = 0;
+	p->ainsn.target_br_reg = 0;
 
-	memcpy(&p->ainsn.insn.bundle, (unsigned long *)bundle_addr,
-				sizeof(bundle_t));
-	memcpy(&p->opcode.bundle, &p->ainsn.insn.bundle, sizeof(bundle_t));
-}
+ 	template = bundle->quad0.template;
 
-void arch_arm_kprobe(struct kprobe *p)
-{
-	unsigned long addr = (unsigned long)p->addr;
-	unsigned long arm_addr = addr & ~0xFULL;
-	unsigned long slot = addr & 0xf;
-	unsigned long template;
-    	unsigned long major_opcode = 0;
-    	unsigned long lx_type_inst = 0;
-    	unsigned long kprobe_inst = 0;
-	bundle_t bundle;
-
-   	p->ainsn.inst_flag = 0;
-   	p->ainsn.target_br_reg = 0;
-
-	memcpy(&bundle, &p->ainsn.insn.bundle, sizeof(bundle_t));
-    	template = bundle.quad0.template;
-    	if (slot == 1 && bundle_encoding[template][1] == L) {
-    		lx_type_inst = 1;
-     		slot = 2;
-    	}
+	if (((bundle_encoding[template][1] == L) && slot > 1) || (slot > 2)) {
+		printk(KERN_WARNING "Attempting to insert unaligned kprobe at 0x%lx\n",
+				addr);
+		return -EINVAL;
+	}
 
+ 	if (slot == 1 && bundle_encoding[template][1] == L) {
+ 		lx_type_inst = 1;
+  		slot = 2;
+ 	}
 
 	switch (slot) {
 	case 0:
-   		major_opcode = (bundle.quad0.slot0 >> SLOT0_OPCODE_SHIFT);
-    		kprobe_inst = bundle.quad0.slot0;
-		bundle.quad0.slot0 = BREAK_INST;
+ 		major_opcode = (bundle->quad0.slot0 >> SLOT0_OPCODE_SHIFT);
+ 		kprobe_inst = bundle->quad0.slot0;
+		bundle->quad0.slot0 = BREAK_INST;
 		break;
 	case 1:
-    		major_opcode = (bundle.quad1.slot1_p1 >> SLOT1_p1_OPCODE_SHIFT);
-    		kprobe_inst = (bundle.quad0.slot1_p0 |
-    				(bundle.quad1.slot1_p1 << (64-46)));
-		bundle.quad0.slot1_p0 = BREAK_INST;
-		bundle.quad1.slot1_p1 = (BREAK_INST >> (64-46));
+ 		major_opcode = (bundle->quad1.slot1_p1 >> SLOT1_p1_OPCODE_SHIFT);
+ 		kprobe_inst = (bundle->quad0.slot1_p0 |
+ 				(bundle->quad1.slot1_p1 << (64-46)));
+		bundle->quad0.slot1_p0 = BREAK_INST;
+		bundle->quad1.slot1_p1 = (BREAK_INST >> (64-46));
 		break;
 	case 2:
-    		major_opcode = (bundle.quad1.slot2 >> SLOT2_OPCODE_SHIFT);
-    		kprobe_inst = bundle.quad1.slot2;
-		bundle.quad1.slot2 = BREAK_INST;
+ 		major_opcode = (bundle->quad1.slot2 >> SLOT2_OPCODE_SHIFT);
+ 		kprobe_inst = bundle->quad1.slot2;
+		bundle->quad1.slot2 = BREAK_INST;
 		break;
 	}
-    	/*
-    	 * Look for IP relative Branches, IP relative call or
-    	 * IP relative predicate instructions
-    	 */
-    	if (bundle_encoding[template][slot] == B) {
-    		switch (major_opcode) {
-    			case INDIRECT_CALL_OPCODE:
-    				p->ainsn.inst_flag |= INST_FLAG_FIX_BRANCH_REG;
-    				p->ainsn.target_br_reg = ((kprobe_inst >> 6) & 0x7);
-    				break;
-    			case IP_RELATIVE_PREDICT_OPCODE:
-    			case IP_RELATIVE_BRANCH_OPCODE:
-    				p->ainsn.inst_flag |= INST_FLAG_FIX_RELATIVE_IP_ADDR;
-    				break;
-    			case IP_RELATIVE_CALL_OPCODE:
-    				p->ainsn.inst_flag |= INST_FLAG_FIX_RELATIVE_IP_ADDR;
-    				p->ainsn.inst_flag |= INST_FLAG_FIX_BRANCH_REG;
-    				p->ainsn.target_br_reg = ((kprobe_inst >> 6) & 0x7);
-    				break;
-    			default:
-    				/* Do nothing */
-    				break;
-    		}
-    	} else if (lx_type_inst) {
-    		switch (major_opcode) {
-    			case LONG_CALL_OPCODE:
-    				p->ainsn.inst_flag |= INST_FLAG_FIX_BRANCH_REG;
-    				p->ainsn.target_br_reg = ((kprobe_inst >> 6) & 0x7);
-   				break;
-    			default:
-    				/* Do nothing */
-    				break;
-    		}
-    	}
-
- 	/* Flush icache for the instruction at the emulated address */
-	flush_icache_range((unsigned long)&p->ainsn.insn.bundle,
-			(unsigned long)&p->ainsn.insn.bundle +
-			sizeof(bundle_t));
-	/*
-	 * Patch the original instruction with the probe instruction
-	 * and flush the instruction cache
-	 */
-	memcpy((char *) arm_addr, (char *) &bundle, sizeof(bundle_t));
+
+ 	/*
+ 	 * Look for IP relative Branches, IP relative call or
+ 	 * IP relative predicate instructions
+ 	 */
+ 	if (bundle_encoding[template][slot] == B) {
+ 		switch (major_opcode) {
+ 			case INDIRECT_CALL_OPCODE:
+ 				p->ainsn.inst_flag |= INST_FLAG_FIX_BRANCH_REG;
+ 				p->ainsn.target_br_reg = ((kprobe_inst >> 6) & 0x7);
+ 				break;
+ 			case IP_RELATIVE_PREDICT_OPCODE:
+ 			case IP_RELATIVE_BRANCH_OPCODE:
+ 				p->ainsn.inst_flag |= INST_FLAG_FIX_RELATIVE_IP_ADDR;
+ 				break;
+ 			case IP_RELATIVE_CALL_OPCODE:
+ 				p->ainsn.inst_flag |= INST_FLAG_FIX_RELATIVE_IP_ADDR;
+ 				p->ainsn.inst_flag |= INST_FLAG_FIX_BRANCH_REG;
+ 				p->ainsn.target_br_reg = ((kprobe_inst >> 6) & 0x7);
+ 				break;
+ 			default:
+ 				/* Do nothing */
+ 				break;
+ 		}
+ 	} else if (lx_type_inst) {
+ 		switch (major_opcode) {
+ 			case LONG_CALL_OPCODE:
+ 				p->ainsn.inst_flag |= INST_FLAG_FIX_BRANCH_REG;
+ 				p->ainsn.target_br_reg = ((kprobe_inst >> 6) & 0x7);
+				break;
+ 			default:
+ 				/* Do nothing */
+ 				break;
+ 		}
+	}
+
+	return 0;
+}
+
+void arch_arm_kprobe(struct kprobe *p)
+{
+	unsigned long addr = (unsigned long)p->addr;
+	unsigned long arm_addr = addr & ~0xFULL;
+
+	memcpy((char *)arm_addr, &p->ainsn.insn.bundle, sizeof(bundle_t));
 	flush_icache_range(arm_addr, arm_addr + sizeof(bundle_t));
 }
 
@@ -226,7 +202,7 @@ void arch_remove_kprobe(struct kprobe *p)
  */
 static void resume_execution(struct kprobe *p, struct pt_regs *regs)
 {
-  	unsigned long bundle_addr = ((unsigned long) (&p->ainsn.insn.bundle)) & ~0xFULL;
+  	unsigned long bundle_addr = ((unsigned long) (&p->opcode.bundle)) & ~0xFULL;
   	unsigned long resume_addr = (unsigned long)p->addr & ~0xFULL;
  	unsigned long template;
  	int slot = ((unsigned long)p->addr & 0xf);
@@ -293,7 +269,7 @@ turn_ss_off:
 
 static void prepare_ss(struct kprobe *p, struct pt_regs *regs)
 {
-	unsigned long bundle_addr = (unsigned long) &p->ainsn.insn.bundle;
+	unsigned long bundle_addr = (unsigned long) &p->opcode.bundle;
 	unsigned long slot = (unsigned long)p->addr & 0xf;
 
 	/* Update instruction pointer (IIP) and slot number (IPSR.ri) */
diff --git a/include/asm-ia64/kprobes.h b/include/asm-ia64/kprobes.h
index d30af77a0b1..cec4d995830 100644
--- a/include/asm-ia64/kprobes.h
+++ b/include/asm-ia64/kprobes.h
@@ -30,6 +30,8 @@
 
 #define BREAK_INST	(long)(__IA64_BREAK_KPROBE << 6)
 
+struct kprobe;
+
 typedef struct _bundle {
 	struct {
 		unsigned long long template : 5;
@@ -79,6 +81,11 @@ static inline void jprobe_return(void)
 {
 }
 
+/* ia64 does not need this */
+static inline void arch_copy_kprobe(struct kprobe *p)
+{
+}
+
 #ifdef CONFIG_KPROBES
 extern int kprobe_exceptions_notify(struct notifier_block *self,
 				    unsigned long val, void *data);
-- 
cgit v1.2.3-70-g09d2


From 1674eafcbd3e3c68556cf19fbf4a2c30f7add729 Mon Sep 17 00:00:00 2001
From: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Date: Thu, 23 Jun 2005 00:09:33 -0700
Subject: [PATCH] Kprobes IA64: cmp ctype unc support

The current Kprobes when patching the original instruction with the break
instruction tries to retain the original qualifying predicate(qp), however
for cmp.crel.ctype where ctype == unc, which is a special instruction
always needs to be executed irrespective of qp.  Hence, if the instruction
we are patching is of this type, then we should not copy the original qp to
the break instruction, this is because we always want the break fault to
happen so that we can emulate the instruction.

This patch is based on the feedback given by David Mosberger

Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/kprobes.c | 43 +++++++++++++++++++++++++++++++++++++++++--
 include/asm-ia64/kprobes.h | 17 +++++++++++++++++
 2 files changed, 58 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index 5d534468155..51bc08bd046 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -119,6 +119,41 @@ static void update_kprobe_inst_flag(uint template, uint  slot, uint major_opcode
 	return;
 }
 
+/*
+ * In this function we check to see if the instruction
+ * (qp) cmpx.crel.ctype p1,p2=r2,r3
+ * on which we are inserting kprobe is cmp instruction
+ * with ctype as unc.
+ */
+static uint is_cmp_ctype_unc_inst(uint template, uint slot, uint major_opcode,
+unsigned long kprobe_inst)
+{
+	cmp_inst_t cmp_inst;
+	uint ctype_unc = 0;
+
+	if (!((bundle_encoding[template][slot] == I) ||
+		(bundle_encoding[template][slot] == M)))
+		goto out;
+
+	if (!((major_opcode == 0xC) || (major_opcode == 0xD) ||
+		(major_opcode == 0xE)))
+		goto out;
+
+	cmp_inst.l = kprobe_inst;
+	if ((cmp_inst.f.x2 == 0) || (cmp_inst.f.x2 == 1)) {
+		/* Integere compare - Register Register (A6 type)*/
+		if ((cmp_inst.f.tb == 0) && (cmp_inst.f.ta == 0)
+				&&(cmp_inst.f.c == 1))
+			ctype_unc = 1;
+	} else if ((cmp_inst.f.x2 == 2)||(cmp_inst.f.x2 == 3)) {
+		/* Integere compare - Immediate Register (A8 type)*/
+		if ((cmp_inst.f.ta == 0) &&(cmp_inst.f.c == 1))
+			ctype_unc = 1;
+	}
+out:
+	return ctype_unc;
+}
+
 /*
  * In this function we override the bundle with
  * the break instruction at the given slot.
@@ -131,9 +166,13 @@ static void prepare_break_inst(uint template, uint  slot, uint major_opcode,
 
 	/*
 	 * Copy the original kprobe_inst qualifying predicate(qp)
-	 * to the break instruction
+	 * to the break instruction iff !is_cmp_ctype_unc_inst
+	 * because for cmp instruction with ctype equal to unc,
+	 * which is a special instruction always needs to be
+	 * executed regradless of qp
 	 */
-	break_inst |= (0x3f & kprobe_inst);
+	if (!is_cmp_ctype_unc_inst(template, slot, major_opcode, kprobe_inst))
+		break_inst |= (0x3f & kprobe_inst);
 
 	switch (slot) {
 	  case 0:
diff --git a/include/asm-ia64/kprobes.h b/include/asm-ia64/kprobes.h
index cec4d995830..7b700035e36 100644
--- a/include/asm-ia64/kprobes.h
+++ b/include/asm-ia64/kprobes.h
@@ -30,6 +30,23 @@
 
 #define BREAK_INST	(long)(__IA64_BREAK_KPROBE << 6)
 
+typedef union cmp_inst {
+	struct {
+	unsigned long long qp : 6;
+	unsigned long long p1 : 6;
+	unsigned long long c  : 1;
+	unsigned long long r2 : 7;
+	unsigned long long r3 : 7;
+	unsigned long long p2 : 6;
+	unsigned long long ta : 1;
+	unsigned long long x2 : 2;
+	unsigned long long tb : 1;
+	unsigned long long opcode : 4;
+	unsigned long long reserved : 23;
+	}f;
+	unsigned long long l;
+} cmp_inst_t;
+
 struct kprobe;
 
 typedef struct _bundle {
-- 
cgit v1.2.3-70-g09d2


From ea32c65cc2d2294c04e9f81d0578a6f51febfdbf Mon Sep 17 00:00:00 2001
From: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Date: Thu, 23 Jun 2005 00:09:36 -0700
Subject: [PATCH] kprobes: Temporary disarming of reentrant probe

In situations where a kprobes handler calls a routine which has a probe on it,
then kprobes_handler() disarms the new probe forever.  This patch removes the
above limitation by temporarily disarming the new probe.  When the another
probe hits while handling the old probe, the kprobes_handler() saves previous
kprobes state and handles the new probe without calling the new kprobes
registered handlers.  kprobe_post_handler() restores back the previous kprobes
state and the normal execution continues.

However on x86_64 architecture, re-rentrancy is provided only through
pre_handler().  If a routine having probe is referenced through
post_handler(), then the probes on that routine are disarmed forever, since
the exception stack is gets changed after the processor single steps the
instruction of the new probe.

This patch includes generic changes to support temporary disarming on
reentrancy of probes.

Signed-of-by: Prasanna S Panchamukhi <prasanna@in.ibm.com>

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/kprobes.h | 9 +++++++++
 kernel/kprobes.c        | 1 +
 2 files changed, 10 insertions(+)

(limited to 'include')

diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h
index 461391decc4..5e1a7b0d7b3 100644
--- a/include/linux/kprobes.h
+++ b/include/linux/kprobes.h
@@ -36,6 +36,12 @@
 
 #include <asm/kprobes.h>
 
+/* kprobe_status settings */
+#define KPROBE_HIT_ACTIVE	0x00000001
+#define KPROBE_HIT_SS		0x00000002
+#define KPROBE_REENTER		0x00000004
+#define KPROBE_HIT_SSDONE	0x00000008
+
 struct kprobe;
 struct pt_regs;
 struct kretprobe;
@@ -55,6 +61,9 @@ struct kprobe {
 	/* list of kprobes for multi-handler support */
 	struct list_head list;
 
+	/*count the number of times this probe was temporarily disarmed */
+	unsigned long nmissed;
+
 	/* location of the probe point */
 	kprobe_opcode_t *addr;
 
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index dd42e717dd3..456ecedff2d 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -335,6 +335,7 @@ int register_kprobe(struct kprobe *p)
 	}
 	spin_lock_irqsave(&kprobe_lock, flags);
 	old_p = get_kprobe(p->addr);
+	p->nmissed = 0;
 	if (old_p) {
 		ret = register_aggr_kprobe(old_p, p);
 		goto out;
-- 
cgit v1.2.3-70-g09d2


From d6e711448137ca3301512cec41a2c2ce852b3d0a Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Thu, 23 Jun 2005 00:09:43 -0700
Subject: [PATCH] setuid core dump

Add a new `suid_dumpable' sysctl:

This value can be used to query and set the core dump mode for setuid
or otherwise protected/tainted binaries. The modes are

0 - (default) - traditional behaviour.  Any process which has changed
    privilege levels or is execute only will not be dumped

1 - (debug) - all processes dump core when possible.  The core dump is
    owned by the current user and no security is applied.  This is intended
    for system debugging situations only.  Ptrace is unchecked.

2 - (suidsafe) - any binary which normally would not be dumped is dumped
    readable by root only.  This allows the end user to remove such a dump but
    not access it directly.  For security reasons core dumps in this mode will
    not overwrite one another or other files.  This mode is appropriate when
    adminstrators are attempting to debug problems in a normal environment.

(akpm:

> > +EXPORT_SYMBOL(suid_dumpable);
>
> EXPORT_SYMBOL_GPL?

No problem to me.

> >  	if (current->euid == current->uid && current->egid == current->gid)
> >  		current->mm->dumpable = 1;
>
> Should this be SUID_DUMP_USER?

Actually the feedback I had from last time was that the SUID_ defines
should go because its clearer to follow the numbers. They can go
everywhere (and there are lots of places where dumpable is tested/used
as a bool in untouched code)

> Maybe this should be renamed to `dump_policy' or something.  Doing that
> would help us catch any code which isn't using the #defines, too.

Fair comment. The patch was designed to be easy to maintain for Red Hat
rather than for merging. Changing that field would create a gigantic
diff because it is used all over the place.

)

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/sysctl/kernel.txt | 20 ++++++++++++++++++++
 fs/exec.c                       | 23 +++++++++++++++++++++--
 fs/proc/base.c                  |  6 ++++--
 include/linux/binfmts.h         |  5 +++++
 include/linux/sched.h           |  2 +-
 include/linux/sysctl.h          |  1 +
 kernel/sys.c                    | 22 +++++++++++-----------
 kernel/sysctl.c                 |  9 +++++++++
 security/commoncap.c            |  2 +-
 security/dummy.c                |  2 +-
 10 files changed, 74 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/Documentation/sysctl/kernel.txt b/Documentation/sysctl/kernel.txt
index 35159176997..9f11d36a8c1 100644
--- a/Documentation/sysctl/kernel.txt
+++ b/Documentation/sysctl/kernel.txt
@@ -49,6 +49,7 @@ show up in /proc/sys/kernel:
 - shmmax                      [ sysv ipc ]
 - shmmni
 - stop-a                      [ SPARC only ]
+- suid_dumpable
 - sysrq                       ==> Documentation/sysrq.txt
 - tainted
 - threads-max
@@ -300,6 +301,25 @@ kernel.  This value defaults to SHMMAX.
 
 ==============================================================
 
+suid_dumpable:
+
+This value can be used to query and set the core dump mode for setuid
+or otherwise protected/tainted binaries. The modes are
+
+0 - (default) - traditional behaviour. Any process which has changed
+	privilege levels or is execute only will not be dumped
+1 - (debug) - all processes dump core when possible. The core dump is
+	owned by the current user and no security is applied. This is
+	intended for system debugging situations only. Ptrace is unchecked.
+2 - (suidsafe) - any binary which normally would not be dumped is dumped
+	readable by root only. This allows the end user to remove
+	such a dump but not access it directly. For security reasons
+	core dumps in this mode will not overwrite one another or
+	other files. This mode is appropriate when adminstrators are
+	attempting to debug problems in a normal environment.
+
+==============================================================
+
 tainted: 
 
 Non-zero if the kernel has been tainted.  Numeric values, which
diff --git a/fs/exec.c b/fs/exec.c
index 3a4b35a14c0..48871917d36 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -58,6 +58,9 @@
 
 int core_uses_pid;
 char core_pattern[65] = "core";
+int suid_dumpable = 0;
+
+EXPORT_SYMBOL(suid_dumpable);
 /* The maximal length of core_pattern is also specified in sysctl.c */
 
 static struct linux_binfmt *formats;
@@ -864,6 +867,9 @@ int flush_old_exec(struct linux_binprm * bprm)
 
 	if (current->euid == current->uid && current->egid == current->gid)
 		current->mm->dumpable = 1;
+	else
+		current->mm->dumpable = suid_dumpable;
+
 	name = bprm->filename;
 
 	/* Copies the binary name from after last slash */
@@ -884,7 +890,7 @@ int flush_old_exec(struct linux_binprm * bprm)
 	    permission(bprm->file->f_dentry->d_inode,MAY_READ, NULL) ||
 	    (bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP)) {
 		suid_keys(current);
-		current->mm->dumpable = 0;
+		current->mm->dumpable = suid_dumpable;
 	}
 
 	/* An exec changes our domain. We are no longer part of the thread
@@ -1432,6 +1438,8 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs)
 	struct inode * inode;
 	struct file * file;
 	int retval = 0;
+	int fsuid = current->fsuid;
+	int flag = 0;
 
 	binfmt = current->binfmt;
 	if (!binfmt || !binfmt->core_dump)
@@ -1441,6 +1449,16 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs)
 		up_write(&mm->mmap_sem);
 		goto fail;
 	}
+
+	/*
+	 *	We cannot trust fsuid as being the "true" uid of the
+	 *	process nor do we know its entire history. We only know it
+	 *	was tainted so we dump it as root in mode 2.
+	 */
+	if (mm->dumpable == 2) {	/* Setuid core dump mode */
+		flag = O_EXCL;		/* Stop rewrite attacks */
+		current->fsuid = 0;	/* Dump root private */
+	}
 	mm->dumpable = 0;
 	init_completion(&mm->core_done);
 	spin_lock_irq(&current->sighand->siglock);
@@ -1466,7 +1484,7 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs)
  	lock_kernel();
 	format_corename(corename, core_pattern, signr);
 	unlock_kernel();
-	file = filp_open(corename, O_CREAT | 2 | O_NOFOLLOW | O_LARGEFILE, 0600);
+	file = filp_open(corename, O_CREAT | 2 | O_NOFOLLOW | O_LARGEFILE | flag, 0600);
 	if (IS_ERR(file))
 		goto fail_unlock;
 	inode = file->f_dentry->d_inode;
@@ -1491,6 +1509,7 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs)
 close_fail:
 	filp_close(file, NULL);
 fail_unlock:
+	current->fsuid = fsuid;
 	complete_all(&mm->core_done);
 fail:
 	return retval;
diff --git a/fs/proc/base.c b/fs/proc/base.c
index e31903aadd9..ace151fa487 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -314,7 +314,7 @@ static int may_ptrace_attach(struct task_struct *task)
 	     (current->gid != task->gid)) && !capable(CAP_SYS_PTRACE))
 		goto out;
 	rmb();
-	if (!task->mm->dumpable && !capable(CAP_SYS_PTRACE))
+	if (task->mm->dumpable != 1 && !capable(CAP_SYS_PTRACE))
 		goto out;
 	if (security_ptrace(current, task))
 		goto out;
@@ -1113,7 +1113,9 @@ static int task_dumpable(struct task_struct *task)
 	if (mm)
 		dumpable = mm->dumpable;
 	task_unlock(task);
-	return dumpable;
+	if(dumpable == 1)
+		return 1;
+	return 0;
 }
 
 
diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h
index 7e736e201c4..c1e82c51444 100644
--- a/include/linux/binfmts.h
+++ b/include/linux/binfmts.h
@@ -69,6 +69,11 @@ extern void remove_arg_zero(struct linux_binprm *);
 extern int search_binary_handler(struct linux_binprm *,struct pt_regs *);
 extern int flush_old_exec(struct linux_binprm * bprm);
 
+extern int suid_dumpable;
+#define SUID_DUMP_DISABLE	0	/* No setuid dumping */
+#define SUID_DUMP_USER		1	/* Dump as user of process */
+#define SUID_DUMP_ROOT		2	/* Dump as root */
+
 /* Stack area protections */
 #define EXSTACK_DEFAULT   0	/* Whatever the arch defaults to */
 #define EXSTACK_DISABLE_X 1	/* Disable executable stacks */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index b58afd97a18..901742f9238 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -246,7 +246,7 @@ struct mm_struct {
 
 	unsigned long saved_auxv[42]; /* for /proc/PID/auxv */
 
-	unsigned dumpable:1;
+	unsigned dumpable:2;
 	cpumask_t cpu_vm_mask;
 
 	/* Architecture-specific MM context */
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index a17745c80a9..614e939c78a 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -136,6 +136,7 @@ enum
 	KERN_UNKNOWN_NMI_PANIC=66, /* int: unknown nmi panic flag */
 	KERN_BOOTLOADER_TYPE=67, /* int: boot loader type */
 	KERN_RANDOMIZE=68, /* int: randomize virtual address space */
+	KERN_SETUID_DUMPABLE=69, /* int: behaviour of dumps for setuid core */
 };
 
 
diff --git a/kernel/sys.c b/kernel/sys.c
index f006632c2ba..0a2c8cda963 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -525,7 +525,7 @@ asmlinkage long sys_setregid(gid_t rgid, gid_t egid)
 	}
 	if (new_egid != old_egid)
 	{
-		current->mm->dumpable = 0;
+		current->mm->dumpable = suid_dumpable;
 		smp_wmb();
 	}
 	if (rgid != (gid_t) -1 ||
@@ -556,7 +556,7 @@ asmlinkage long sys_setgid(gid_t gid)
 	{
 		if(old_egid != gid)
 		{
-			current->mm->dumpable=0;
+			current->mm->dumpable = suid_dumpable;
 			smp_wmb();
 		}
 		current->gid = current->egid = current->sgid = current->fsgid = gid;
@@ -565,7 +565,7 @@ asmlinkage long sys_setgid(gid_t gid)
 	{
 		if(old_egid != gid)
 		{
-			current->mm->dumpable=0;
+			current->mm->dumpable = suid_dumpable;
 			smp_wmb();
 		}
 		current->egid = current->fsgid = gid;
@@ -596,7 +596,7 @@ static int set_user(uid_t new_ruid, int dumpclear)
 
 	if(dumpclear)
 	{
-		current->mm->dumpable = 0;
+		current->mm->dumpable = suid_dumpable;
 		smp_wmb();
 	}
 	current->uid = new_ruid;
@@ -653,7 +653,7 @@ asmlinkage long sys_setreuid(uid_t ruid, uid_t euid)
 
 	if (new_euid != old_euid)
 	{
-		current->mm->dumpable=0;
+		current->mm->dumpable = suid_dumpable;
 		smp_wmb();
 	}
 	current->fsuid = current->euid = new_euid;
@@ -703,7 +703,7 @@ asmlinkage long sys_setuid(uid_t uid)
 
 	if (old_euid != uid)
 	{
-		current->mm->dumpable = 0;
+		current->mm->dumpable = suid_dumpable;
 		smp_wmb();
 	}
 	current->fsuid = current->euid = uid;
@@ -748,7 +748,7 @@ asmlinkage long sys_setresuid(uid_t ruid, uid_t euid, uid_t suid)
 	if (euid != (uid_t) -1) {
 		if (euid != current->euid)
 		{
-			current->mm->dumpable = 0;
+			current->mm->dumpable = suid_dumpable;
 			smp_wmb();
 		}
 		current->euid = euid;
@@ -798,7 +798,7 @@ asmlinkage long sys_setresgid(gid_t rgid, gid_t egid, gid_t sgid)
 	if (egid != (gid_t) -1) {
 		if (egid != current->egid)
 		{
-			current->mm->dumpable = 0;
+			current->mm->dumpable = suid_dumpable;
 			smp_wmb();
 		}
 		current->egid = egid;
@@ -845,7 +845,7 @@ asmlinkage long sys_setfsuid(uid_t uid)
 	{
 		if (uid != old_fsuid)
 		{
-			current->mm->dumpable = 0;
+			current->mm->dumpable = suid_dumpable;
 			smp_wmb();
 		}
 		current->fsuid = uid;
@@ -875,7 +875,7 @@ asmlinkage long sys_setfsgid(gid_t gid)
 	{
 		if (gid != old_fsgid)
 		{
-			current->mm->dumpable = 0;
+			current->mm->dumpable = suid_dumpable;
 			smp_wmb();
 		}
 		current->fsgid = gid;
@@ -1652,7 +1652,7 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3,
 				error = 1;
 			break;
 		case PR_SET_DUMPABLE:
-			if (arg2 != 0 && arg2 != 1) {
+			if (arg2 < 0 || arg2 > 2) {
 				error = -EINVAL;
 				break;
 			}
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 701d12c6306..24a4d12d5aa 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -58,6 +58,7 @@ extern int sysctl_overcommit_ratio;
 extern int max_threads;
 extern int sysrq_enabled;
 extern int core_uses_pid;
+extern int suid_dumpable;
 extern char core_pattern[];
 extern int cad_pid;
 extern int pid_max;
@@ -950,6 +951,14 @@ static ctl_table fs_table[] = {
 		.proc_handler	= &proc_dointvec,
 	},
 #endif
+	{
+		.ctl_name	= KERN_SETUID_DUMPABLE,
+		.procname	= "suid_dumpable",
+		.data		= &suid_dumpable,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
 	{ .ctl_name = 0 }
 };
 
diff --git a/security/commoncap.c b/security/commoncap.c
index 849b8c338ee..04c12f58d65 100644
--- a/security/commoncap.c
+++ b/security/commoncap.c
@@ -149,7 +149,7 @@ void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe)
 
 	if (bprm->e_uid != current->uid || bprm->e_gid != current->gid ||
 	    !cap_issubset (new_permitted, current->cap_permitted)) {
-		current->mm->dumpable = 0;
+		current->mm->dumpable = suid_dumpable;
 
 		if (unsafe & ~LSM_UNSAFE_PTRACE_CAP) {
 			if (!capable(CAP_SETUID)) {
diff --git a/security/dummy.c b/security/dummy.c
index b32eff14654..6ff88758647 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -130,7 +130,7 @@ static void dummy_bprm_free_security (struct linux_binprm *bprm)
 static void dummy_bprm_apply_creds (struct linux_binprm *bprm, int unsafe)
 {
 	if (bprm->e_uid != current->uid || bprm->e_gid != current->gid) {
-		current->mm->dumpable = 0;
+		current->mm->dumpable = suid_dumpable;
 
 		if ((unsafe & ~LSM_UNSAFE_PTRACE_CAP) && !capable(CAP_SETUID)) {
 			bprm->e_uid = current->uid;
-- 
cgit v1.2.3-70-g09d2


From ef3daeda7b58f046f94b26637d500354038d39f4 Mon Sep 17 00:00:00 2001
From: Yoav Zach <yoav_zach@yahoo.com>
Date: Thu, 23 Jun 2005 00:09:58 -0700
Subject: [PATCH] Don't force O_LARGEFILE for 32 bit processes on ia64

In ia64 kernel, the O_LARGEFILE flag is forced when opening a file.  This
is problematic for execution of 32 bit processes, which are not largefile
aware, either by SW emulation or by HW execution.

For such processes, the problem is two-fold:

1) When trying to open a file that is larger than 4G
   the operation should fail, but it's not
2) Writing to offset larger than 4G should fail, but
   it's not

The proposed patch takes advantage of the way 32 bit processes are
identified in ia64 systems.  Such processes have PER_LINUX32 for their
personality.  With the patch, the ia64 kernel will not enforce the
O_LARGEFILE flag if the current process has PER_LINUX32 set.  The behavior
for all other architectures remains unchanged.

Signed-off-by: Yoav Zach <yoav.zach@intel.com>
Acked-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/open.c                | 7 ++++---
 include/asm-ia64/fcntl.h | 2 ++
 include/linux/fcntl.h    | 4 ++++
 3 files changed, 10 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/fs/open.c b/fs/open.c
index 963bd81a44c..2ebb72c1a87 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -21,6 +21,7 @@
 #include <linux/vfs.h>
 #include <asm/uaccess.h>
 #include <linux/fs.h>
+#include <linux/personality.h>
 #include <linux/pagemap.h>
 #include <linux/syscalls.h>
 
@@ -935,9 +936,9 @@ asmlinkage long sys_open(const char __user * filename, int flags, int mode)
 	char * tmp;
 	int fd, error;
 
-#if BITS_PER_LONG != 32
-	flags |= O_LARGEFILE;
-#endif
+	if (force_o_largefile())
+		flags |= O_LARGEFILE;
+
 	tmp = getname(filename);
 	fd = PTR_ERR(tmp);
 	if (!IS_ERR(tmp)) {
diff --git a/include/asm-ia64/fcntl.h b/include/asm-ia64/fcntl.h
index d193981bb1d..c9f8d835d0c 100644
--- a/include/asm-ia64/fcntl.h
+++ b/include/asm-ia64/fcntl.h
@@ -81,4 +81,6 @@ struct flock {
 
 #define F_LINUX_SPECIFIC_BASE	1024
 
+#define force_o_largefile() ( ! (current->personality & PER_LINUX32) )
+
 #endif /* _ASM_IA64_FCNTL_H */
diff --git a/include/linux/fcntl.h b/include/linux/fcntl.h
index 704fb76b633..8a7c82151de 100644
--- a/include/linux/fcntl.h
+++ b/include/linux/fcntl.h
@@ -25,6 +25,10 @@
 
 #ifdef __KERNEL__
 
+#ifndef force_o_largefile
+#define force_o_largefile() (BITS_PER_LONG != 32)
+#endif
+
 #if BITS_PER_LONG == 32
 #define IS_GETLK32(cmd)		((cmd) == F_GETLK)
 #define IS_SETLK32(cmd)		((cmd) == F_SETLK)
-- 
cgit v1.2.3-70-g09d2


From 11c80c8367db0a9d342529ed74464670cd86a1f6 Mon Sep 17 00:00:00 2001
From: Jan Beulich <JBeulich@novell.com>
Date: Thu, 23 Jun 2005 00:09:59 -0700
Subject: [PATCH] adjust per_cpu definition in non-SMP case

Fix (in the architectures I'm actually building for) the UP definition of
per_cpu so that the cpu specified may be any expression, not just an
identifier or a suffix expression.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-generic/percpu.h | 2 +-
 include/asm-ia64/percpu.h    | 2 +-
 include/asm-x86_64/percpu.h  | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h
index 3b709b84934..9044aeb3782 100644
--- a/include/asm-generic/percpu.h
+++ b/include/asm-generic/percpu.h
@@ -29,7 +29,7 @@ do {								\
 #define DEFINE_PER_CPU(type, name) \
     __typeof__(type) per_cpu__##name
 
-#define per_cpu(var, cpu)			(*((void)cpu, &per_cpu__##var))
+#define per_cpu(var, cpu)			(*((void)(cpu), &per_cpu__##var))
 #define __get_cpu_var(var)			per_cpu__##var
 
 #endif	/* SMP */
diff --git a/include/asm-ia64/percpu.h b/include/asm-ia64/percpu.h
index 1e87f19dad5..2b14dee29ce 100644
--- a/include/asm-ia64/percpu.h
+++ b/include/asm-ia64/percpu.h
@@ -50,7 +50,7 @@ extern void *per_cpu_init(void);
 
 #else /* ! SMP */
 
-#define per_cpu(var, cpu)			(*((void)cpu, &per_cpu__##var))
+#define per_cpu(var, cpu)			(*((void)(cpu), &per_cpu__##var))
 #define __get_cpu_var(var)			per_cpu__##var
 #define per_cpu_init()				(__phys_per_cpu_start)
 
diff --git a/include/asm-x86_64/percpu.h b/include/asm-x86_64/percpu.h
index 415d73f3c8e..9c71855736f 100644
--- a/include/asm-x86_64/percpu.h
+++ b/include/asm-x86_64/percpu.h
@@ -39,7 +39,7 @@ extern void setup_per_cpu_areas(void);
 #define DEFINE_PER_CPU(type, name) \
     __typeof__(type) per_cpu__##name
 
-#define per_cpu(var, cpu)			(*((void)cpu, &per_cpu__##var))
+#define per_cpu(var, cpu)			(*((void)(cpu), &per_cpu__##var))
 #define __get_cpu_var(var)			per_cpu__##var
 
 #endif	/* SMP */
-- 
cgit v1.2.3-70-g09d2


From 46c271bedd2c8444b1d05bc44928beec0c07debc Mon Sep 17 00:00:00 2001
From: Peter Osterlund <petero2@telia.com>
Date: Thu, 23 Jun 2005 00:10:02 -0700
Subject: [PATCH] Improve CD/DVD packet driver write performance

This patch improves write performance for the CD/DVD packet writing driver.
 The logic for switching between reading and writing has been changed so
that streaming writes are no longer interrupted by read requests.

Signed-off-by: Peter Osterlund <petero2@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/pktcdvd.c | 36 ++++++++++++++++++++----------------
 include/linux/pktcdvd.h |  2 +-
 2 files changed, 21 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index bc56770bcc9..7f3d78de265 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -467,14 +467,12 @@ static int pkt_set_speed(struct pktcdvd_device *pd, unsigned write_speed, unsign
  * Queue a bio for processing by the low-level CD device. Must be called
  * from process context.
  */
-static void pkt_queue_bio(struct pktcdvd_device *pd, struct bio *bio, int high_prio_read)
+static void pkt_queue_bio(struct pktcdvd_device *pd, struct bio *bio)
 {
 	spin_lock(&pd->iosched.lock);
 	if (bio_data_dir(bio) == READ) {
 		pkt_add_list_last(bio, &pd->iosched.read_queue,
 				  &pd->iosched.read_queue_tail);
-		if (high_prio_read)
-			pd->iosched.high_prio_read = 1;
 	} else {
 		pkt_add_list_last(bio, &pd->iosched.write_queue,
 				  &pd->iosched.write_queue_tail);
@@ -490,15 +488,16 @@ static void pkt_queue_bio(struct pktcdvd_device *pd, struct bio *bio, int high_p
  * requirements for CDRW drives:
  * - A cache flush command must be inserted before a read request if the
  *   previous request was a write.
- * - Switching between reading and writing is slow, so don't it more often
+ * - Switching between reading and writing is slow, so don't do it more often
  *   than necessary.
+ * - Optimize for throughput at the expense of latency. This means that streaming
+ *   writes will never be interrupted by a read, but if the drive has to seek
+ *   before the next write, switch to reading instead if there are any pending
+ *   read requests.
  * - Set the read speed according to current usage pattern. When only reading
  *   from the device, it's best to use the highest possible read speed, but
  *   when switching often between reading and writing, it's better to have the
  *   same read and write speeds.
- * - Reads originating from user space should have higher priority than reads
- *   originating from pkt_gather_data, because some process is usually waiting
- *   on reads of the first kind.
  */
 static void pkt_iosched_process_queue(struct pktcdvd_device *pd)
 {
@@ -512,21 +511,24 @@ static void pkt_iosched_process_queue(struct pktcdvd_device *pd)
 
 	for (;;) {
 		struct bio *bio;
-		int reads_queued, writes_queued, high_prio_read;
+		int reads_queued, writes_queued;
 
 		spin_lock(&pd->iosched.lock);
 		reads_queued = (pd->iosched.read_queue != NULL);
 		writes_queued = (pd->iosched.write_queue != NULL);
-		if (!reads_queued)
-			pd->iosched.high_prio_read = 0;
-		high_prio_read = pd->iosched.high_prio_read;
 		spin_unlock(&pd->iosched.lock);
 
 		if (!reads_queued && !writes_queued)
 			break;
 
 		if (pd->iosched.writing) {
-			if (high_prio_read || (!writes_queued && reads_queued)) {
+			int need_write_seek = 1;
+			spin_lock(&pd->iosched.lock);
+			bio = pd->iosched.write_queue;
+			spin_unlock(&pd->iosched.lock);
+			if (bio && (bio->bi_sector == pd->iosched.last_write))
+				need_write_seek = 0;
+			if (need_write_seek && reads_queued) {
 				if (atomic_read(&pd->cdrw.pending_bios) > 0) {
 					VPRINTK("pktcdvd: write, waiting\n");
 					break;
@@ -559,8 +561,10 @@ static void pkt_iosched_process_queue(struct pktcdvd_device *pd)
 
 		if (bio_data_dir(bio) == READ)
 			pd->iosched.successive_reads += bio->bi_size >> 10;
-		else
+		else {
 			pd->iosched.successive_reads = 0;
+			pd->iosched.last_write = bio->bi_sector + bio_sectors(bio);
+		}
 		if (pd->iosched.successive_reads >= HI_SPEED_SWITCH) {
 			if (pd->read_speed == pd->write_speed) {
 				pd->read_speed = MAX_SPEED;
@@ -765,7 +769,7 @@ static void pkt_gather_data(struct pktcdvd_device *pd, struct packet_data *pkt)
 
 		atomic_inc(&pkt->io_wait);
 		bio->bi_rw = READ;
-		pkt_queue_bio(pd, bio, 0);
+		pkt_queue_bio(pd, bio);
 		frames_read++;
 	}
 
@@ -1062,7 +1066,7 @@ static void pkt_start_write(struct pktcdvd_device *pd, struct packet_data *pkt)
 
 	atomic_set(&pkt->io_wait, 1);
 	pkt->w_bio->bi_rw = WRITE;
-	pkt_queue_bio(pd, pkt->w_bio, 0);
+	pkt_queue_bio(pd, pkt->w_bio);
 }
 
 static void pkt_finish_packet(struct packet_data *pkt, int uptodate)
@@ -2120,7 +2124,7 @@ static int pkt_make_request(request_queue_t *q, struct bio *bio)
 		cloned_bio->bi_private = psd;
 		cloned_bio->bi_end_io = pkt_end_io_read_cloned;
 		pd->stats.secs_r += bio->bi_size >> 9;
-		pkt_queue_bio(pd, cloned_bio, 1);
+		pkt_queue_bio(pd, cloned_bio);
 		return 0;
 	}
 
diff --git a/include/linux/pktcdvd.h b/include/linux/pktcdvd.h
index 4e2d2a942ec..4b32bce9a28 100644
--- a/include/linux/pktcdvd.h
+++ b/include/linux/pktcdvd.h
@@ -159,7 +159,7 @@ struct packet_iosched
 	struct bio		*read_queue_tail;
 	struct bio		*write_queue;
 	struct bio		*write_queue_tail;
-	int			high_prio_read;	/* An important read request has been queued */
+	sector_t		last_write;	/* The sector where the last write ended */
 	int			successive_reads;
 };
 
-- 
cgit v1.2.3-70-g09d2


From fa912bcb06d5dc9525d8912a145db2bf4b7668c5 Mon Sep 17 00:00:00 2001
From: Daniel Ritz <daniel.ritz@gmx.ch>
Date: Thu, 23 Jun 2005 00:10:12 -0700
Subject: [PATCH] yenta TI: turn off interrupts during card power-on #2

- make boot-up card recognition more reliable (ie.  redo interrogation
  always if there is no valid 'card inserted' state) (and yes, i saw it
  happening on an o2micro controller that both CB_CBARD and CB_16BITCARD
  bits were set at the same time)

- also redo interrogation before probing the ISA interrupts.  it's safer
  to do the probing with the socket in a clean state.

- make card insert detect more reliable.  yenta_get_status() now returns
  SS_PENDING as long as the card is not completley inserted and one of the
  voltage bits is set.  also !CB_CBARD doesn't mean CB_16BITCARD.  there is
  CB_NOTACARD as well, so make an explicit check for CB_16BITCARD.

- for TI bridges: disable IRQs during power-on.  in all-serial and tied
  interrupt mode the interrupts are always disabled for single-slot
  controllers.  for two-slot contollers the disabling is only done when the
  other slot is empty.  to force disabling there is a new module parameter
  now: pwr_irqs_off=Y (which is a regression for working setups.  that's
  why it's an option, only use when required)

- modparm to disable ISA interrupt probing (isa_probe, defaults to on)

- remove unneeded code/cleanups (ie.  merge yenta_events() into
  yenta_interrupts())

Signed-off-by: Daniel Ritz <daniel.ritz@gmx.ch>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/pcmcia/cs.c           |  11 ++-
 drivers/pcmcia/ti113x.h       | 167 ++++++++++++++++++++++++++++++++++++++++++
 drivers/pcmcia/yenta_socket.c |  60 +++++++++------
 include/pcmcia/ss.h           |   8 ++
 4 files changed, 223 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/drivers/pcmcia/cs.c b/drivers/pcmcia/cs.c
index 03fc885db1c..d136b3c8fac 100644
--- a/drivers/pcmcia/cs.c
+++ b/drivers/pcmcia/cs.c
@@ -508,6 +508,10 @@ static int socket_setup(struct pcmcia_socket *skt, int initial_delay)
 		cs_err(skt, "unsupported voltage key.\n");
 		return CS_BAD_TYPE;
 	}
+
+	if (skt->power_hook)
+		skt->power_hook(skt, HOOK_POWER_PRE);
+
 	skt->socket.flags = 0;
 	skt->ops->set_socket(skt, &skt->socket);
 
@@ -522,7 +526,12 @@ static int socket_setup(struct pcmcia_socket *skt, int initial_delay)
 		return CS_BAD_TYPE;
 	}
 
-	return socket_reset(skt);
+	status = socket_reset(skt);
+
+	if (skt->power_hook)
+		skt->power_hook(skt, HOOK_POWER_POST);
+
+	return status;
 }
 
 /*
diff --git a/drivers/pcmcia/ti113x.h b/drivers/pcmcia/ti113x.h
index a8a1d104524..c7ba99871ac 100644
--- a/drivers/pcmcia/ti113x.h
+++ b/drivers/pcmcia/ti113x.h
@@ -611,6 +611,170 @@ out:
 	}
 }
 
+
+/* Returns true value if the second slot of a two-slot controller is empty */
+static int ti12xx_2nd_slot_empty(struct yenta_socket *socket)
+{
+	struct pci_dev *func;
+	struct yenta_socket *slot2;
+	int devfn;
+	unsigned int state;
+	int ret = 1;
+
+	/* catch the two-slot controllers */
+	switch (socket->dev->device) {
+	case PCI_DEVICE_ID_TI_1220:
+	case PCI_DEVICE_ID_TI_1221:
+	case PCI_DEVICE_ID_TI_1225:
+	case PCI_DEVICE_ID_TI_1251A:
+	case PCI_DEVICE_ID_TI_1251B:
+	case PCI_DEVICE_ID_TI_1420:
+	case PCI_DEVICE_ID_TI_1450:
+	case PCI_DEVICE_ID_TI_1451A:
+	case PCI_DEVICE_ID_TI_1520:
+	case PCI_DEVICE_ID_TI_1620:
+	case PCI_DEVICE_ID_TI_4520:
+	case PCI_DEVICE_ID_TI_4450:
+	case PCI_DEVICE_ID_TI_4451:
+		/*
+		 * there are way more, but they need to be added in yenta_socket.c
+		 * and pci_ids.h first anyway.
+		 */
+		break;
+
+	/* single-slot controllers have the 2nd slot empty always :) */
+	default:
+		return 1;
+	}
+
+	/* get other slot */
+	devfn = socket->dev->devfn & ~0x07;
+	func = pci_get_slot(socket->dev->bus,
+	                    (socket->dev->devfn & 0x07) ? devfn : devfn | 0x01);
+	if (!func)
+		return 1;
+
+	slot2 = pci_get_drvdata(func);
+	if (!slot2)
+		goto out;
+
+	/* check state */
+	yenta_get_status(&socket->socket, &state);
+	if (state & SS_DETECT) {
+		ret = 0;
+		goto out;
+	}
+
+out:
+	pci_dev_put(func);
+	return ret;
+}
+
+/*
+ * TI specifiy parts for the power hook.
+ *
+ * some TI's with some CB's produces interrupt storm on power on. it has been
+ * seen with atheros wlan cards on TI1225 and TI1410. solution is simply to
+ * disable any CB interrupts during this time.
+ */
+static int ti12xx_power_hook(struct pcmcia_socket *sock, int operation)
+{
+	struct yenta_socket *socket = container_of(sock, struct yenta_socket, socket);
+	u32 mfunc, devctl, sysctl;
+	u8 gpio3;
+
+	/* only POWER_PRE and POWER_POST are interesting */
+	if ((operation != HOOK_POWER_PRE) && (operation != HOOK_POWER_POST))
+		return 0;
+
+	devctl = config_readb(socket, TI113X_DEVICE_CONTROL);
+	sysctl = config_readl(socket, TI113X_SYSTEM_CONTROL);
+	mfunc = config_readl(socket, TI122X_MFUNC);
+
+	/*
+	 * all serial/tied: only disable when modparm set. always doing it
+	 * would mean a regression for working setups 'cos it disables the
+	 * interrupts for both both slots on 2-slot controllers
+	 * (and users of single slot controllers where it's save have to
+	 * live with setting the modparm, most don't have to anyway)
+	 */
+	if (((devctl & TI113X_DCR_IMODE_MASK) == TI12XX_DCR_IMODE_ALL_SERIAL) &&
+	    (pwr_irqs_off || ti12xx_2nd_slot_empty(socket))) {
+		switch (socket->dev->device) {
+		case PCI_DEVICE_ID_TI_1250:
+		case PCI_DEVICE_ID_TI_1251A:
+		case PCI_DEVICE_ID_TI_1251B:
+		case PCI_DEVICE_ID_TI_1450:
+		case PCI_DEVICE_ID_TI_1451A:
+		case PCI_DEVICE_ID_TI_4450:
+		case PCI_DEVICE_ID_TI_4451:
+			/* these chips have no IRQSER setting in MFUNC3  */
+			break;
+
+		default:
+			if (operation == HOOK_POWER_PRE)
+				mfunc = (mfunc & ~TI122X_MFUNC3_MASK);
+			else
+				mfunc = (mfunc & ~TI122X_MFUNC3_MASK) | TI122X_MFUNC3_IRQSER;
+		}
+
+		return 0;
+	}
+
+	/* do the job differently for func0/1 */
+	if ((PCI_FUNC(socket->dev->devfn) == 0) ||
+	    ((sysctl & TI122X_SCR_INTRTIE) &&
+	     (pwr_irqs_off || ti12xx_2nd_slot_empty(socket)))) {
+		/* some bridges are different */
+		switch (socket->dev->device) {
+		case PCI_DEVICE_ID_TI_1250:
+		case PCI_DEVICE_ID_TI_1251A:
+		case PCI_DEVICE_ID_TI_1251B:
+		case PCI_DEVICE_ID_TI_1450:
+			/* those oldies use gpio3 for INTA */
+			gpio3 = config_readb(socket, TI1250_GPIO3_CONTROL);
+			if (operation == HOOK_POWER_PRE)
+				gpio3 = (gpio3 & ~TI1250_GPIO_MODE_MASK) | 0x40;
+			else
+				gpio3 &= ~TI1250_GPIO_MODE_MASK;
+			config_writeb(socket, TI1250_GPIO3_CONTROL, gpio3);
+			break;
+
+		default:
+			/* all new bridges are the same */
+			if (operation == HOOK_POWER_PRE)
+				mfunc &= ~TI122X_MFUNC0_MASK;
+			else
+				mfunc |= TI122X_MFUNC0_INTA;
+			config_writel(socket, TI122X_MFUNC, mfunc);
+		}
+	} else {
+		switch (socket->dev->device) {
+		case PCI_DEVICE_ID_TI_1251A:
+		case PCI_DEVICE_ID_TI_1251B:
+		case PCI_DEVICE_ID_TI_1450:
+			/* those have INTA elsewhere and INTB in MFUNC0 */
+			if (operation == HOOK_POWER_PRE)
+				mfunc &= ~TI122X_MFUNC0_MASK;
+			else
+				mfunc |= TI125X_MFUNC0_INTB;
+			config_writel(socket, TI122X_MFUNC, mfunc);
+
+			break;
+
+		default:
+			/* all new bridges are the same */
+			if (operation == HOOK_POWER_PRE)
+				mfunc &= ~TI122X_MFUNC1_MASK;
+			else
+				mfunc |= TI122X_MFUNC1_INTB;
+			config_writel(socket, TI122X_MFUNC, mfunc);
+		}
+	}
+
+	return 0;
+}
+
 static int ti12xx_override(struct yenta_socket *socket)
 {
 	u32 val, val_orig;
@@ -654,6 +818,9 @@ static int ti12xx_override(struct yenta_socket *socket)
 	else
 		ti12xx_irqroute_func1(socket);
 
+	/* install power hook */
+	socket->socket.power_hook = ti12xx_power_hook;
+
 	return ti_override(socket);
 }
 
diff --git a/drivers/pcmcia/yenta_socket.c b/drivers/pcmcia/yenta_socket.c
index 6404d97a12e..bee05362fd2 100644
--- a/drivers/pcmcia/yenta_socket.c
+++ b/drivers/pcmcia/yenta_socket.c
@@ -32,6 +32,14 @@ static int disable_clkrun;
 module_param(disable_clkrun, bool, 0444);
 MODULE_PARM_DESC(disable_clkrun, "If PC card doesn't function properly, please try this option");
 
+static int isa_probe = 1;
+module_param(isa_probe, bool, 0444);
+MODULE_PARM_DESC(isa_probe, "If set ISA interrupts are probed (default). Set to N to disable probing");
+
+static int pwr_irqs_off;
+module_param(pwr_irqs_off, bool, 0644);
+MODULE_PARM_DESC(pwr_irqs_off, "Force IRQs off during power-on of slot. Use only when seeing IRQ storms!");
+
 #if 0
 #define debug(x,args...) printk(KERN_DEBUG "%s: " x, __func__ , ##args)
 #else
@@ -150,15 +158,16 @@ static int yenta_get_status(struct pcmcia_socket *sock, unsigned int *value)
 
 	val  = (state & CB_3VCARD) ? SS_3VCARD : 0;
 	val |= (state & CB_XVCARD) ? SS_XVCARD : 0;
-	val |= (state & (CB_CDETECT1 | CB_CDETECT2 | CB_5VCARD | CB_3VCARD
-			 | CB_XVCARD | CB_YVCARD)) ? 0 : SS_PENDING;
+	val |= (state & (CB_5VCARD | CB_3VCARD | CB_XVCARD | CB_YVCARD)) ? 0 : SS_PENDING;
+	val |= (state & (CB_CDETECT1 | CB_CDETECT2)) ? SS_PENDING : 0;
+
 
 	if (state & CB_CBCARD) {
 		val |= SS_CARDBUS;	
 		val |= (state & CB_CARDSTS) ? SS_STSCHG : 0;
 		val |= (state & (CB_CDETECT1 | CB_CDETECT2)) ? 0 : SS_DETECT;
 		val |= (state & CB_PWRCYCLE) ? SS_POWERON | SS_READY : 0;
-	} else {
+	} else if (state & CB_16BITCARD) {
 		u8 status = exca_readb(socket, I365_STATUS);
 		val |= ((status & I365_CS_DETECT) == I365_CS_DETECT) ? SS_DETECT : 0;
 		if (exca_readb(socket, I365_INTCTL) & I365_PC_IOCARD) {
@@ -405,11 +414,13 @@ static int yenta_set_mem_map(struct pcmcia_socket *sock, struct pccard_mem_map *
 }
 
 
-static unsigned int yenta_events(struct yenta_socket *socket)
+
+static irqreturn_t yenta_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
+	unsigned int events;
+	struct yenta_socket *socket = (struct yenta_socket *) dev_id;
 	u8 csc;
 	u32 cb_event;
-	unsigned int events;
 
 	/* Clear interrupt status for the event */
 	cb_event = cb_readl(socket, CB_SOCKET_EVENT);
@@ -426,20 +437,13 @@ static unsigned int yenta_events(struct yenta_socket *socket)
 		events |= (csc & I365_CSC_BVD2) ? SS_BATWARN : 0;
 		events |= (csc & I365_CSC_READY) ? SS_READY : 0;
 	}
-	return events;
-}
-
-
-static irqreturn_t yenta_interrupt(int irq, void *dev_id, struct pt_regs *regs)
-{
-	unsigned int events;
-	struct yenta_socket *socket = (struct yenta_socket *) dev_id;
 
-	events = yenta_events(socket);
-	if (events) {
+	if (events)
 		pcmcia_parse_events(&socket->socket, events);
+
+	if (cb_event || csc)
 		return IRQ_HANDLED;
-	}
+
 	return IRQ_NONE;
 }
 
@@ -470,11 +474,22 @@ static void yenta_clear_maps(struct yenta_socket *socket)
 	}
 }
 
+/* redoes voltage interrogation if required */
+static void yenta_interrogate(struct yenta_socket *socket)
+{
+	u32 state;
+
+	state = cb_readl(socket, CB_SOCKET_STATE);
+	if (!(state & (CB_5VCARD | CB_3VCARD | CB_XVCARD | CB_YVCARD)) ||
+	    (state & (CB_CDETECT1 | CB_CDETECT2 | CB_NOTACARD | CB_BADVCCREQ)) ||
+	    ((state & (CB_16BITCARD | CB_CBCARD)) == (CB_16BITCARD | CB_CBCARD)))
+		cb_writel(socket, CB_SOCKET_FORCE, CB_CVSTEST);
+}
+
 /* Called at resume and initialization events */
 static int yenta_sock_init(struct pcmcia_socket *sock)
 {
 	struct yenta_socket *socket = container_of(sock, struct yenta_socket, socket);
-	u32 state;
 	u16 bridge;
 
 	bridge = config_readw(socket, CB_BRIDGE_CONTROL) & ~CB_BRIDGE_INTR;
@@ -486,10 +501,7 @@ static int yenta_sock_init(struct pcmcia_socket *sock)
 	exca_writeb(socket, I365_GENCTL, 0x00);
 
 	/* Redo card voltage interrogation */
-	state = cb_readl(socket, CB_SOCKET_STATE);
-	if (!(state & (CB_CDETECT1 | CB_CDETECT2 | CB_5VCARD |
-	               CB_3VCARD | CB_XVCARD | CB_YVCARD)))
-		cb_writel(socket, CB_SOCKET_FORCE, CB_CVSTEST);
+	yenta_interrogate(socket);
 
 	yenta_clear_maps(socket);
 
@@ -856,7 +868,10 @@ static void yenta_get_socket_capabilities(struct yenta_socket *socket, u32 isa_i
 	socket->socket.features |= SS_CAP_PAGE_REGS | SS_CAP_PCCARD | SS_CAP_CARDBUS;
 	socket->socket.map_size = 0x1000;
 	socket->socket.pci_irq = socket->cb_irq;
-	socket->socket.irq_mask = yenta_probe_irq(socket, isa_irq_mask);
+	if (isa_probe)
+		socket->socket.irq_mask = yenta_probe_irq(socket, isa_irq_mask);
+	else
+		socket->socket.irq_mask = 0;
 	socket->socket.cb_dev = socket->dev;
 
 	printk(KERN_INFO "Yenta: ISA IRQ mask 0x%04x, PCI irq %d\n",
@@ -996,6 +1011,7 @@ static int __devinit yenta_probe (struct pci_dev *dev, const struct pci_device_i
 	}
 
 	/* Figure out what the dang thing can do for the PCMCIA layer... */
+	yenta_interrogate(socket);
 	yenta_get_socket_capabilities(socket, isa_interrupts);
 	printk(KERN_INFO "Socket status: %08x\n", cb_readl(socket, CB_SOCKET_STATE));
 
diff --git a/include/pcmcia/ss.h b/include/pcmcia/ss.h
index 6d3413a5670..67b867f31fe 100644
--- a/include/pcmcia/ss.h
+++ b/include/pcmcia/ss.h
@@ -77,6 +77,11 @@ extern socket_state_t dead_socket;
 /* Use this just for bridge windows */
 #define MAP_IOSPACE	0x20
 
+/* power hook operations */
+#define HOOK_POWER_PRE	0x01
+#define HOOK_POWER_POST	0x02
+
+
 typedef struct pccard_io_map {
     u_char	map;
     u_char	flags;
@@ -222,6 +227,9 @@ struct pcmcia_socket {
 	/* Zoom video behaviour is so chip specific its not worth adding
 	   this to _ops */
 	void 				(*zoom_video)(struct pcmcia_socket *, int);
+
+	/* so is power hook */
+	int (*power_hook)(struct pcmcia_socket *sock, int operation);
                            
 	/* state thread */
 	struct semaphore		skt_sem;	/* protects socket h/w state */
-- 
cgit v1.2.3-70-g09d2


From 0d77e5a2c23da734f5a7925f64afa1c2ed92e0f9 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Thu, 23 Jun 2005 00:10:14 -0700
Subject: [PATCH] compat: introduce compat_time_t

This patch is based on work by Carlos O'Donell and Matthew Wilcox.  It
introduces/updates the compat_time_t type and uses it for compat siginfo
structures.  I have built this on ppc64 and x86_64.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/ia32/ia32priv.h       | 2 +-
 arch/s390/kernel/compat_linux.h | 2 +-
 arch/sparc64/kernel/signal32.c  | 2 +-
 include/asm-ia64/compat.h       | 1 +
 include/asm-mips/compat.h       | 1 +
 include/asm-parisc/compat.h     | 2 +-
 include/asm-ppc64/compat.h      | 1 +
 include/asm-ppc64/ppc32.h       | 2 +-
 include/asm-sparc64/compat.h    | 1 +
 include/asm-x86_64/ia32.h       | 2 +-
 10 files changed, 10 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/ia32/ia32priv.h b/arch/ia64/ia32/ia32priv.h
index b2de948bdae..e3e9290e3ff 100644
--- a/arch/ia64/ia32/ia32priv.h
+++ b/arch/ia64/ia32/ia32priv.h
@@ -241,7 +241,7 @@ typedef struct compat_siginfo {
 
 		/* POSIX.1b timers */
 		struct {
-			timer_t _tid;		/* timer id */
+			compat_timer_t _tid;		/* timer id */
 			int _overrun;		/* overrun count */
 			char _pad[sizeof(unsigned int) - sizeof(int)];
 			compat_sigval_t _sigval;	/* same as below */
diff --git a/arch/s390/kernel/compat_linux.h b/arch/s390/kernel/compat_linux.h
index bf33dcfec7d..3898f66d0b2 100644
--- a/arch/s390/kernel/compat_linux.h
+++ b/arch/s390/kernel/compat_linux.h
@@ -45,7 +45,7 @@ typedef struct compat_siginfo {
 
 		/* POSIX.1b timers */
 		struct {
-			timer_t _tid;		/* timer id */
+			compat_timer_t _tid;		/* timer id */
 			int _overrun;		/* overrun count */
 			compat_sigval_t _sigval;	/* same as below */
 			int _sys_private;       /* not to be passed to user */
diff --git a/arch/sparc64/kernel/signal32.c b/arch/sparc64/kernel/signal32.c
index 9a375e975cf..f28428f4170 100644
--- a/arch/sparc64/kernel/signal32.c
+++ b/arch/sparc64/kernel/signal32.c
@@ -102,7 +102,7 @@ typedef struct compat_siginfo{
 
 		/* POSIX.1b timers */
 		struct {
-			timer_t _tid;			/* timer id */
+			compat_timer_t _tid;			/* timer id */
 			int _overrun;			/* overrun count */
 			compat_sigval_t _sigval;		/* same as below */
 			int _sys_private;		/* not to be passed to user */
diff --git a/include/asm-ia64/compat.h b/include/asm-ia64/compat.h
index cc0ff0a4bdd..0c05e5bad8a 100644
--- a/include/asm-ia64/compat.h
+++ b/include/asm-ia64/compat.h
@@ -27,6 +27,7 @@ typedef u16		compat_ipc_pid_t;
 typedef s32		compat_daddr_t;
 typedef u32		compat_caddr_t;
 typedef __kernel_fsid_t	compat_fsid_t;
+typedef s32		compat_timer_t;
 
 typedef s32		compat_int_t;
 typedef s32		compat_long_t;
diff --git a/include/asm-mips/compat.h b/include/asm-mips/compat.h
index dce92079e7f..d78002afb1e 100644
--- a/include/asm-mips/compat.h
+++ b/include/asm-mips/compat.h
@@ -29,6 +29,7 @@ typedef s32		compat_caddr_t;
 typedef struct {
 	s32	val[2];
 } compat_fsid_t;
+typedef s32		compat_timer_t;
 
 typedef s32		compat_int_t;
 typedef s32		compat_long_t;
diff --git a/include/asm-parisc/compat.h b/include/asm-parisc/compat.h
index ca0eac647a0..7630d1ad239 100644
--- a/include/asm-parisc/compat.h
+++ b/include/asm-parisc/compat.h
@@ -24,7 +24,7 @@ typedef u16	compat_nlink_t;
 typedef u16	compat_ipc_pid_t;
 typedef s32	compat_daddr_t;
 typedef u32	compat_caddr_t;
-typedef u32	compat_timer_t;
+typedef s32	compat_timer_t;
 
 typedef s32	compat_int_t;
 typedef s32	compat_long_t;
diff --git a/include/asm-ppc64/compat.h b/include/asm-ppc64/compat.h
index 09c28d28ce6..12414f5fc66 100644
--- a/include/asm-ppc64/compat.h
+++ b/include/asm-ppc64/compat.h
@@ -26,6 +26,7 @@ typedef s32		compat_daddr_t;
 typedef u32		compat_caddr_t;
 typedef __kernel_fsid_t	compat_fsid_t;
 typedef s32		compat_key_t;
+typedef s32		compat_timer_t;
 
 typedef s32		compat_int_t;
 typedef s32		compat_long_t;
diff --git a/include/asm-ppc64/ppc32.h b/include/asm-ppc64/ppc32.h
index 1d040489755..6b44a8caf39 100644
--- a/include/asm-ppc64/ppc32.h
+++ b/include/asm-ppc64/ppc32.h
@@ -32,7 +32,7 @@ typedef struct compat_siginfo {
 
 		/* POSIX.1b timers */
 		struct {
-			timer_t _tid;			/* timer id */
+			compat_timer_t _tid;			/* timer id */
 			int _overrun;			/* overrun count */
 			compat_sigval_t _sigval;		/* same as below */
 			int _sys_private;		/* not to be passed to user */
diff --git a/include/asm-sparc64/compat.h b/include/asm-sparc64/compat.h
index 22f58055b8a..b59122dd176 100644
--- a/include/asm-sparc64/compat.h
+++ b/include/asm-sparc64/compat.h
@@ -25,6 +25,7 @@ typedef s32		compat_daddr_t;
 typedef u32		compat_caddr_t;
 typedef __kernel_fsid_t	compat_fsid_t;
 typedef s32		compat_key_t;
+typedef s32		compat_timer_t;
 
 typedef s32		compat_int_t;
 typedef s32		compat_long_t;
diff --git a/include/asm-x86_64/ia32.h b/include/asm-x86_64/ia32.h
index c0a7717923e..6efa00fe4e7 100644
--- a/include/asm-x86_64/ia32.h
+++ b/include/asm-x86_64/ia32.h
@@ -94,7 +94,7 @@ typedef struct compat_siginfo{
 
 		/* POSIX.1b timers */
 		struct {
-			int _tid;		/* timer id */
+			compat_timer_t _tid;	/* timer id */
 			int _overrun;		/* overrun count */
 			compat_sigval_t _sigval;	/* same as below */
 			int _sys_private;	/* not to be passed to user */
-- 
cgit v1.2.3-70-g09d2


From bb93e3a52f8db7210258a1a2134cced0b78a46e1 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 23 Jun 2005 00:10:15 -0700
Subject: [PATCH] block: add unlocked_ioctl support for block devices

This patch allows block device drivers to convert their ioctl functions to
unlocked_ioctl() like character devices and other subsystems.  All
functions that were called with the BKL held before are still used that
way, but I would not be surprised if it could be removed from the ioctl
functions in drivers/block/ioctl.c themselves.

As a side note, I found that compat_blkdev_ioctl() acquires the BKL as
well, which looks like a bug.  I have checked that every user of
disk->fops->compat_ioctl() in the current git tree gets the BKL itself, so
it could easily be removed from compat_blkdev_ioctl().

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/ioctl.c | 74 +++++++++++++++++++++++++++++++++++++--------------
 fs/block_dev.c        |  5 ++--
 include/linux/fs.h    |  1 +
 3 files changed, 57 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/drivers/block/ioctl.c b/drivers/block/ioctl.c
index 6d7bcc9da9e..6e278474f9a 100644
--- a/drivers/block/ioctl.c
+++ b/drivers/block/ioctl.c
@@ -133,11 +133,9 @@ static int put_u64(unsigned long arg, u64 val)
 	return put_user(val, (u64 __user *)arg);
 }
 
-int blkdev_ioctl(struct inode *inode, struct file *file, unsigned cmd,
-			unsigned long arg)
+static int blkdev_locked_ioctl(struct file *file, struct block_device *bdev,
+				unsigned cmd, unsigned long arg)
 {
-	struct block_device *bdev = inode->i_bdev;
-	struct gendisk *disk = bdev->bd_disk;
 	struct backing_dev_info *bdi;
 	int ret, n;
 
@@ -190,36 +188,72 @@ int blkdev_ioctl(struct inode *inode, struct file *file, unsigned cmd,
 		return put_ulong(arg, bdev->bd_inode->i_size >> 9);
 	case BLKGETSIZE64:
 		return put_u64(arg, bdev->bd_inode->i_size);
+	}
+	return -ENOIOCTLCMD;
+}
+
+static int blkdev_driver_ioctl(struct inode *inode, struct file *file,
+		struct gendisk *disk, unsigned cmd, unsigned long arg)
+{
+	int ret;
+	if (disk->fops->unlocked_ioctl)
+		return disk->fops->unlocked_ioctl(file, cmd, arg);
+
+	if (disk->fops->ioctl) {
+		lock_kernel();
+		ret = disk->fops->ioctl(inode, file, cmd, arg);
+		unlock_kernel();
+		return ret;
+	}
+
+	return -ENOTTY;
+}
+
+int blkdev_ioctl(struct inode *inode, struct file *file, unsigned cmd,
+			unsigned long arg)
+{
+	struct block_device *bdev = inode->i_bdev;
+	struct gendisk *disk = bdev->bd_disk;
+	int ret, n;
+
+	switch(cmd) {
 	case BLKFLSBUF:
 		if (!capable(CAP_SYS_ADMIN))
 			return -EACCES;
-		if (disk->fops->ioctl) {
-			ret = disk->fops->ioctl(inode, file, cmd, arg);
-			/* -EINVAL to handle old uncorrected drivers */
-			if (ret != -EINVAL && ret != -ENOTTY)
-				return ret;
-		}
+
+		ret = blkdev_driver_ioctl(inode, file, disk, cmd, arg);
+		/* -EINVAL to handle old uncorrected drivers */
+		if (ret != -EINVAL && ret != -ENOTTY)
+			return ret;
+
+		lock_kernel();
 		fsync_bdev(bdev);
 		invalidate_bdev(bdev, 0);
+		unlock_kernel();
 		return 0;
+
 	case BLKROSET:
-		if (disk->fops->ioctl) {
-			ret = disk->fops->ioctl(inode, file, cmd, arg);
-			/* -EINVAL to handle old uncorrected drivers */
-			if (ret != -EINVAL && ret != -ENOTTY)
-				return ret;
-		}
+		ret = blkdev_driver_ioctl(inode, file, disk, cmd, arg);
+		/* -EINVAL to handle old uncorrected drivers */
+		if (ret != -EINVAL && ret != -ENOTTY)
+			return ret;
 		if (!capable(CAP_SYS_ADMIN))
 			return -EACCES;
 		if (get_user(n, (int __user *)(arg)))
 			return -EFAULT;
+		lock_kernel();
 		set_device_ro(bdev, n);
+		unlock_kernel();
 		return 0;
-	default:
-		if (disk->fops->ioctl)
-			return disk->fops->ioctl(inode, file, cmd, arg);
 	}
-	return -ENOTTY;
+
+	lock_kernel();
+	ret = blkdev_locked_ioctl(file, bdev, cmd, arg);
+	unlock_kernel();
+	if (ret != -ENOIOCTLCMD)
+		return ret;
+
+	return blkdev_driver_ioctl(inode, file, disk, cmd, arg);
 }
 
 /* Most of the generic ioctls are handled in the normal fallback path.
diff --git a/fs/block_dev.c b/fs/block_dev.c
index c0cbd1bc1a0..e0df94c37b7 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -777,8 +777,7 @@ static ssize_t blkdev_file_aio_write(struct kiocb *iocb, const char __user *buf,
 	return generic_file_aio_write_nolock(iocb, &local_iov, 1, &iocb->ki_pos);
 }
 
-static int block_ioctl(struct inode *inode, struct file *file, unsigned cmd,
-			unsigned long arg)
+static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 {
 	return blkdev_ioctl(file->f_mapping->host, file, cmd, arg);
 }
@@ -803,7 +802,7 @@ struct file_operations def_blk_fops = {
   	.aio_write	= blkdev_file_aio_write, 
 	.mmap		= generic_file_mmap,
 	.fsync		= block_fsync,
-	.ioctl		= block_ioctl,
+	.unlocked_ioctl	= block_ioctl,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl	= compat_blkdev_ioctl,
 #endif
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 3622e952e98..9b1278e2127 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -884,6 +884,7 @@ struct block_device_operations {
 	int (*open) (struct inode *, struct file *);
 	int (*release) (struct inode *, struct file *);
 	int (*ioctl) (struct inode *, struct file *, unsigned, unsigned long);
+	long (*unlocked_ioctl) (struct file *, unsigned, unsigned long);
 	long (*compat_ioctl) (struct file *, unsigned, unsigned long);
 	int (*media_changed) (struct gendisk *);
 	int (*revalidate_disk) (struct gendisk *);
-- 
cgit v1.2.3-70-g09d2


From 45778ca819accab1a4a3378b3566cab0f189164f Mon Sep 17 00:00:00 2001
From: Christoph Lameter <christoph@graphe.net>
Date: Thu, 23 Jun 2005 00:10:17 -0700
Subject: [PATCH] Remove f_error field from struct file

The following patch removes the f_error field and all checks of f_error.

Trond said:

  f_error was introduced for NFS, and made sense when we were guaranteed
  always to have a file pointer around when write errors occurred.  Since
  then, we have (for various reasons) had to introduce the nfs_open_context in
  order to track the file read/write state, and it made sense to move our
  f_error tracking there too.

Signed-off-by: Christoph Lameter <christoph@lameter.com>
Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfs/direct.c    |  5 -----
 fs/open.c          | 16 ++++------------
 include/linux/fs.h |  1 -
 mm/filemap.c       |  6 ------
 4 files changed, 4 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index d6a30c844de..6537f2c4ae4 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -751,11 +751,6 @@ nfs_file_direct_write(struct kiocb *iocb, const char __user *buf, size_t count,
 	retval = -EFAULT;
 	if (!access_ok(VERIFY_READ, iov.iov_base, iov.iov_len))
 		goto out;
-        if (file->f_error) {
-                retval = file->f_error;
-                file->f_error = 0;
-                goto out;
-        }
 	retval = -EFBIG;
 	if (limit != RLIM_INFINITY) {
 		if (pos >= limit) {
diff --git a/fs/open.c b/fs/open.c
index 2ebb72c1a87..5dd411b084b 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -981,23 +981,15 @@ asmlinkage long sys_creat(const char __user * pathname, int mode)
  */
 int filp_close(struct file *filp, fl_owner_t id)
 {
-	int retval;
-
-	/* Report and clear outstanding errors */
-	retval = filp->f_error;
-	if (retval)
-		filp->f_error = 0;
+	int retval = 0;
 
 	if (!file_count(filp)) {
 		printk(KERN_ERR "VFS: Close: file count is 0\n");
-		return retval;
+		return 0;
 	}
 
-	if (filp->f_op && filp->f_op->flush) {
-		int err = filp->f_op->flush(filp);
-		if (!retval)
-			retval = err;
-	}
+	if (filp->f_op && filp->f_op->flush)
+		retval = filp->f_op->flush(filp);
 
 	dnotify_flush(filp, id);
 	locks_remove_posix(filp, id);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9b1278e2127..517bf4966bf 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -581,7 +581,6 @@ struct file {
 	atomic_t		f_count;
 	unsigned int 		f_flags;
 	mode_t			f_mode;
-	int			f_error;
 	loff_t			f_pos;
 	struct fown_struct	f_owner;
 	unsigned int		f_uid, f_gid;
diff --git a/mm/filemap.c b/mm/filemap.c
index 4a2fee2cb62..a3598b542a3 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1827,12 +1827,6 @@ inline int generic_write_checks(struct file *file, loff_t *pos, size_t *count, i
         if (unlikely(*pos < 0))
                 return -EINVAL;
 
-        if (unlikely(file->f_error)) {
-                int err = file->f_error;
-                file->f_error = 0;
-                return err;
-        }
-
 	if (!isblk) {
 		/* FIXME: this is for backwards compatibility with 2.4 */
 		if (file->f_flags & O_APPEND)
-- 
cgit v1.2.3-70-g09d2


From f9fd27a253d5e0b23531d12ce7ad15b6535d4486 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Thu, 23 Jun 2005 00:10:19 -0700
Subject: [PATCH] acl endianess annotations

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/posix_acl_xattr.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/posix_acl_xattr.h b/include/linux/posix_acl_xattr.h
index 5efd0a6dad9..fe271c1947b 100644
--- a/include/linux/posix_acl_xattr.h
+++ b/include/linux/posix_acl_xattr.h
@@ -23,13 +23,13 @@
 #define ACL_UNDEFINED_ID	(-1)
 
 typedef struct {
-	__u16			e_tag;
-	__u16			e_perm;
-	__u32			e_id;
+	__le16			e_tag;
+	__le16			e_perm;
+	__le32			e_id;
 } posix_acl_xattr_entry;
 
 typedef struct {
-	__u32			a_version;
+	__le32			a_version;
 	posix_acl_xattr_entry	a_entries[0];
 } posix_acl_xattr_header;
 
-- 
cgit v1.2.3-70-g09d2


From 9a59f452abe11f569e13ec16c51e6d61c54b9838 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Thu, 23 Jun 2005 00:10:19 -0700
Subject: [PATCH] remove <linux/xattr_acl.h>

This file duplicates <linux/posix_acl_xattr.h>, using slightly different
names.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/boot/simple/misc.c          |  2 +-
 arch/ppc/boot/simple/mpc10x_memory.c |  2 +-
 fs/ext2/acl.c                        | 12 ++++++------
 fs/ext2/acl.h                        |  2 +-
 fs/ext3/acl.c                        | 12 ++++++------
 fs/ext3/acl.h                        |  2 +-
 fs/jfs/acl.c                         | 11 ++++++-----
 fs/jfs/jfs_acl.h                     |  2 --
 fs/jfs/super.c                       |  1 +
 fs/jfs/xattr.c                       |  7 ++++---
 fs/nfsd/vfs.c                        |  9 ++++-----
 fs/reiserfs/xattr_acl.c              | 26 +++++++++++++-------------
 include/linux/posix_acl_xattr.h      |  3 +++
 include/linux/reiserfs_acl.h         |  1 -
 14 files changed, 47 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/arch/ppc/boot/simple/misc.c b/arch/ppc/boot/simple/misc.c
index ab0f9902cb6..e02de5b467a 100644
--- a/arch/ppc/boot/simple/misc.c
+++ b/arch/ppc/boot/simple/misc.c
@@ -222,7 +222,7 @@ decompress_kernel(unsigned long load_addr, int num_words, unsigned long cksum)
 	puts("\n");
 
 	puts("Uncompressing Linux...");
-	gunzip(0x0, 0x400000, zimage_start, &zimage_size);
+	gunzip(NULL, 0x400000, zimage_start, &zimage_size);
 	puts("done.\n");
 
 	/* get the bi_rec address */
diff --git a/arch/ppc/boot/simple/mpc10x_memory.c b/arch/ppc/boot/simple/mpc10x_memory.c
index 977daedc14c..20d92a34ceb 100644
--- a/arch/ppc/boot/simple/mpc10x_memory.c
+++ b/arch/ppc/boot/simple/mpc10x_memory.c
@@ -33,7 +33,7 @@
 
 #define MPC10X_PCI_OP(rw, size, type, op, mask)			 	\
 static void								\
-mpc10x_##rw##_config_##size(unsigned int *cfg_addr, 			\
+mpc10x_##rw##_config_##size(unsigned int __iomem *cfg_addr, 			\
 		unsigned int *cfg_data, int devfn, int offset,		\
 		type val)						\
 {									\
diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c
index 25f4a64fd6b..213148c36eb 100644
--- a/fs/ext2/acl.c
+++ b/fs/ext2/acl.c
@@ -396,12 +396,12 @@ static size_t
 ext2_xattr_list_acl_access(struct inode *inode, char *list, size_t list_size,
 			   const char *name, size_t name_len)
 {
-	const size_t size = sizeof(XATTR_NAME_ACL_ACCESS);
+	const size_t size = sizeof(POSIX_ACL_XATTR_ACCESS);
 
 	if (!test_opt(inode->i_sb, POSIX_ACL))
 		return 0;
 	if (list && size <= list_size)
-		memcpy(list, XATTR_NAME_ACL_ACCESS, size);
+		memcpy(list, POSIX_ACL_XATTR_ACCESS, size);
 	return size;
 }
 
@@ -409,12 +409,12 @@ static size_t
 ext2_xattr_list_acl_default(struct inode *inode, char *list, size_t list_size,
 			    const char *name, size_t name_len)
 {
-	const size_t size = sizeof(XATTR_NAME_ACL_DEFAULT);
+	const size_t size = sizeof(POSIX_ACL_XATTR_DEFAULT);
 
 	if (!test_opt(inode->i_sb, POSIX_ACL))
 		return 0;
 	if (list && size <= list_size)
-		memcpy(list, XATTR_NAME_ACL_DEFAULT, size);
+		memcpy(list, POSIX_ACL_XATTR_DEFAULT, size);
 	return size;
 }
 
@@ -506,14 +506,14 @@ ext2_xattr_set_acl_default(struct inode *inode, const char *name,
 }
 
 struct xattr_handler ext2_xattr_acl_access_handler = {
-	.prefix	= XATTR_NAME_ACL_ACCESS,
+	.prefix	= POSIX_ACL_XATTR_ACCESS,
 	.list	= ext2_xattr_list_acl_access,
 	.get	= ext2_xattr_get_acl_access,
 	.set	= ext2_xattr_set_acl_access,
 };
 
 struct xattr_handler ext2_xattr_acl_default_handler = {
-	.prefix	= XATTR_NAME_ACL_DEFAULT,
+	.prefix	= POSIX_ACL_XATTR_DEFAULT,
 	.list	= ext2_xattr_list_acl_default,
 	.get	= ext2_xattr_get_acl_default,
 	.set	= ext2_xattr_set_acl_default,
diff --git a/fs/ext2/acl.h b/fs/ext2/acl.h
index fed96ae81a7..0bde85bafe3 100644
--- a/fs/ext2/acl.h
+++ b/fs/ext2/acl.h
@@ -4,7 +4,7 @@
   (C) 2001 Andreas Gruenbacher, <a.gruenbacher@computer.org>
 */
 
-#include <linux/xattr_acl.h>
+#include <linux/posix_acl_xattr.h>
 
 #define EXT2_ACL_VERSION	0x0001
 
diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c
index 638c13a26c0..133f5aa581b 100644
--- a/fs/ext3/acl.c
+++ b/fs/ext3/acl.c
@@ -417,12 +417,12 @@ static size_t
 ext3_xattr_list_acl_access(struct inode *inode, char *list, size_t list_len,
 			   const char *name, size_t name_len)
 {
-	const size_t size = sizeof(XATTR_NAME_ACL_ACCESS);
+	const size_t size = sizeof(POSIX_ACL_XATTR_ACCESS);
 
 	if (!test_opt(inode->i_sb, POSIX_ACL))
 		return 0;
 	if (list && size <= list_len)
-		memcpy(list, XATTR_NAME_ACL_ACCESS, size);
+		memcpy(list, POSIX_ACL_XATTR_ACCESS, size);
 	return size;
 }
 
@@ -430,12 +430,12 @@ static size_t
 ext3_xattr_list_acl_default(struct inode *inode, char *list, size_t list_len,
 			    const char *name, size_t name_len)
 {
-	const size_t size = sizeof(XATTR_NAME_ACL_DEFAULT);
+	const size_t size = sizeof(POSIX_ACL_XATTR_DEFAULT);
 
 	if (!test_opt(inode->i_sb, POSIX_ACL))
 		return 0;
 	if (list && size <= list_len)
-		memcpy(list, XATTR_NAME_ACL_DEFAULT, size);
+		memcpy(list, POSIX_ACL_XATTR_DEFAULT, size);
 	return size;
 }
 
@@ -535,14 +535,14 @@ ext3_xattr_set_acl_default(struct inode *inode, const char *name,
 }
 
 struct xattr_handler ext3_xattr_acl_access_handler = {
-	.prefix	= XATTR_NAME_ACL_ACCESS,
+	.prefix	= POSIX_ACL_XATTR_ACCESS,
 	.list	= ext3_xattr_list_acl_access,
 	.get	= ext3_xattr_get_acl_access,
 	.set	= ext3_xattr_set_acl_access,
 };
 
 struct xattr_handler ext3_xattr_acl_default_handler = {
-	.prefix	= XATTR_NAME_ACL_DEFAULT,
+	.prefix	= POSIX_ACL_XATTR_DEFAULT,
 	.list	= ext3_xattr_list_acl_default,
 	.get	= ext3_xattr_get_acl_default,
 	.set	= ext3_xattr_set_acl_default,
diff --git a/fs/ext3/acl.h b/fs/ext3/acl.h
index 98af0c0d0ba..92d50b53a93 100644
--- a/fs/ext3/acl.h
+++ b/fs/ext3/acl.h
@@ -4,7 +4,7 @@
   (C) 2001 Andreas Gruenbacher, <a.gruenbacher@computer.org>
 */
 
-#include <linux/xattr_acl.h>
+#include <linux/posix_acl_xattr.h>
 
 #define EXT3_ACL_VERSION	0x0001
 
diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c
index 30a2bf9eeda..e892dab40c2 100644
--- a/fs/jfs/acl.c
+++ b/fs/jfs/acl.c
@@ -21,6 +21,7 @@
 #include <linux/sched.h>
 #include <linux/fs.h>
 #include <linux/quotaops.h>
+#include <linux/posix_acl_xattr.h>
 #include "jfs_incore.h"
 #include "jfs_xattr.h"
 #include "jfs_acl.h"
@@ -36,11 +37,11 @@ static struct posix_acl *jfs_get_acl(struct inode *inode, int type)
 
 	switch(type) {
 		case ACL_TYPE_ACCESS:
-			ea_name = XATTR_NAME_ACL_ACCESS;
+			ea_name = POSIX_ACL_XATTR_ACCESS;
 			p_acl = &ji->i_acl;
 			break;
 		case ACL_TYPE_DEFAULT:
-			ea_name = XATTR_NAME_ACL_DEFAULT;
+			ea_name = POSIX_ACL_XATTR_DEFAULT;
 			p_acl = &ji->i_default_acl;
 			break;
 		default:
@@ -88,11 +89,11 @@ static int jfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 
 	switch(type) {
 		case ACL_TYPE_ACCESS:
-			ea_name = XATTR_NAME_ACL_ACCESS;
+			ea_name = POSIX_ACL_XATTR_ACCESS;
 			p_acl = &ji->i_acl;
 			break;
 		case ACL_TYPE_DEFAULT:
-			ea_name = XATTR_NAME_ACL_DEFAULT;
+			ea_name = POSIX_ACL_XATTR_DEFAULT;
 			p_acl = &ji->i_default_acl;
 			if (!S_ISDIR(inode->i_mode))
 				return acl ? -EACCES : 0;
@@ -101,7 +102,7 @@ static int jfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 			return -EINVAL;
 	}
 	if (acl) {
-		size = xattr_acl_size(acl->a_count);
+		size = posix_acl_xattr_size(acl->a_count);
 		value = kmalloc(size, GFP_KERNEL);
 		if (!value)
 			return -ENOMEM;
diff --git a/fs/jfs/jfs_acl.h b/fs/jfs/jfs_acl.h
index d2ae430adec..a3acd3eec05 100644
--- a/fs/jfs/jfs_acl.h
+++ b/fs/jfs/jfs_acl.h
@@ -20,8 +20,6 @@
 
 #ifdef CONFIG_JFS_POSIX_ACL
 
-#include <linux/xattr_acl.h>
-
 int jfs_permission(struct inode *, int, struct nameidata *);
 int jfs_init_acl(struct inode *, struct inode *);
 int jfs_setattr(struct dentry *, struct iattr *);
diff --git a/fs/jfs/super.c b/fs/jfs/super.c
index 810a3653d8b..ee32211288c 100644
--- a/fs/jfs/super.c
+++ b/fs/jfs/super.c
@@ -24,6 +24,7 @@
 #include <linux/completion.h>
 #include <linux/vfs.h>
 #include <linux/moduleparam.h>
+#include <linux/posix_acl.h>
 #include <asm/uaccess.h>
 
 #include "jfs_incore.h"
diff --git a/fs/jfs/xattr.c b/fs/jfs/xattr.c
index 6016373701a..ee438d429d4 100644
--- a/fs/jfs/xattr.c
+++ b/fs/jfs/xattr.c
@@ -19,6 +19,7 @@
 
 #include <linux/fs.h>
 #include <linux/xattr.h>
+#include <linux/posix_acl_xattr.h>
 #include <linux/quotaops.h>
 #include "jfs_incore.h"
 #include "jfs_superblock.h"
@@ -718,9 +719,9 @@ static int can_set_system_xattr(struct inode *inode, const char *name,
 		return -EPERM;
 
 	/*
-	 * XATTR_NAME_ACL_ACCESS is tied to i_mode
+	 * POSIX_ACL_XATTR_ACCESS is tied to i_mode
 	 */
-	if (strcmp(name, XATTR_NAME_ACL_ACCESS) == 0) {
+	if (strcmp(name, POSIX_ACL_XATTR_ACCESS) == 0) {
 		acl = posix_acl_from_xattr(value, value_len);
 		if (IS_ERR(acl)) {
 			rc = PTR_ERR(acl);
@@ -750,7 +751,7 @@ static int can_set_system_xattr(struct inode *inode, const char *name,
 		JFS_IP(inode)->i_acl = JFS_ACL_NOT_CACHED;
 
 		return 0;
-	} else if (strcmp(name, XATTR_NAME_ACL_DEFAULT) == 0) {
+	} else if (strcmp(name, POSIX_ACL_XATTR_DEFAULT) == 0) {
 		acl = posix_acl_from_xattr(value, value_len);
 		if (IS_ERR(acl)) {
 			rc = PTR_ERR(acl);
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index ae3940dc85c..de340ffd33c 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -50,7 +50,6 @@
 #include <linux/posix_acl.h>
 #ifdef CONFIG_NFSD_V4
 #include <linux/posix_acl_xattr.h>
-#include <linux/xattr_acl.h>
 #include <linux/xattr.h>
 #include <linux/nfs4.h>
 #include <linux/nfs4_acl.h>
@@ -425,13 +424,13 @@ nfsd4_set_nfs4_acl(struct svc_rqst *rqstp, struct svc_fh *fhp,
 		goto out_nfserr;
 
 	if (pacl) {
-		error = set_nfsv4_acl_one(dentry, pacl, XATTR_NAME_ACL_ACCESS);
+		error = set_nfsv4_acl_one(dentry, pacl, POSIX_ACL_XATTR_ACCESS);
 		if (error < 0)
 			goto out_nfserr;
 	}
 
 	if (dpacl) {
-		error = set_nfsv4_acl_one(dentry, dpacl, XATTR_NAME_ACL_DEFAULT);
+		error = set_nfsv4_acl_one(dentry, dpacl, POSIX_ACL_XATTR_DEFAULT);
 		if (error < 0)
 			goto out_nfserr;
 	}
@@ -498,7 +497,7 @@ nfsd4_get_nfs4_acl(struct svc_rqst *rqstp, struct dentry *dentry, struct nfs4_ac
 	struct posix_acl *pacl = NULL, *dpacl = NULL;
 	unsigned int flags = 0;
 
-	pacl = _get_posix_acl(dentry, XATTR_NAME_ACL_ACCESS);
+	pacl = _get_posix_acl(dentry, POSIX_ACL_XATTR_ACCESS);
 	if (IS_ERR(pacl) && PTR_ERR(pacl) == -ENODATA)
 		pacl = posix_acl_from_mode(inode->i_mode, GFP_KERNEL);
 	if (IS_ERR(pacl)) {
@@ -508,7 +507,7 @@ nfsd4_get_nfs4_acl(struct svc_rqst *rqstp, struct dentry *dentry, struct nfs4_ac
 	}
 
 	if (S_ISDIR(inode->i_mode)) {
-		dpacl = _get_posix_acl(dentry, XATTR_NAME_ACL_DEFAULT);
+		dpacl = _get_posix_acl(dentry, POSIX_ACL_XATTR_DEFAULT);
 		if (IS_ERR(dpacl) && PTR_ERR(dpacl) == -ENODATA)
 			dpacl = NULL;
 		else if (IS_ERR(dpacl)) {
diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c
index e302071903a..c312881c5f5 100644
--- a/fs/reiserfs/xattr_acl.c
+++ b/fs/reiserfs/xattr_acl.c
@@ -4,7 +4,7 @@
 #include <linux/errno.h>
 #include <linux/pagemap.h>
 #include <linux/xattr.h>
-#include <linux/xattr_acl.h>
+#include <linux/posix_acl_xattr.h>
 #include <linux/reiserfs_xattr.h>
 #include <linux/reiserfs_acl.h>
 #include <asm/uaccess.h>
@@ -192,11 +192,11 @@ reiserfs_get_acl(struct inode *inode, int type)
 
         switch (type) {
             case ACL_TYPE_ACCESS:
-                name = XATTR_NAME_ACL_ACCESS;
+                name = POSIX_ACL_XATTR_ACCESS;
                 p_acl = &reiserfs_i->i_acl_access;
                 break;
             case ACL_TYPE_DEFAULT:
-                name = XATTR_NAME_ACL_DEFAULT;
+                name = POSIX_ACL_XATTR_DEFAULT;
                 p_acl = &reiserfs_i->i_acl_default;
                 break;
             default:
@@ -260,7 +260,7 @@ reiserfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 
         switch (type) {
             case ACL_TYPE_ACCESS:
-                name = XATTR_NAME_ACL_ACCESS;
+                name = POSIX_ACL_XATTR_ACCESS;
                 p_acl = &reiserfs_i->i_acl_access;
                 if (acl) {
                     mode_t mode = inode->i_mode;
@@ -275,7 +275,7 @@ reiserfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
                 }
                 break;
             case ACL_TYPE_DEFAULT:
-                name = XATTR_NAME_ACL_DEFAULT;
+                name = POSIX_ACL_XATTR_DEFAULT;
                 p_acl = &reiserfs_i->i_acl_default;
                 if (!S_ISDIR (inode->i_mode))
                     return acl ? -EACCES : 0;
@@ -468,7 +468,7 @@ static int
 posix_acl_access_get(struct inode *inode, const char *name,
 			  void *buffer, size_t size)
 {
-	if (strlen(name) != sizeof(XATTR_NAME_ACL_ACCESS)-1)
+	if (strlen(name) != sizeof(POSIX_ACL_XATTR_ACCESS)-1)
 		return -EINVAL;
 	return xattr_get_acl(inode, ACL_TYPE_ACCESS, buffer, size);
 }
@@ -477,7 +477,7 @@ static int
 posix_acl_access_set(struct inode *inode, const char *name,
 			  const void *value, size_t size, int flags)
 {
-	if (strlen(name) != sizeof(XATTR_NAME_ACL_ACCESS)-1)
+	if (strlen(name) != sizeof(POSIX_ACL_XATTR_ACCESS)-1)
 		return -EINVAL;
 	return xattr_set_acl(inode, ACL_TYPE_ACCESS, value, size);
 }
@@ -487,7 +487,7 @@ posix_acl_access_del (struct inode *inode, const char *name)
 {
     struct reiserfs_inode_info *reiserfs_i = REISERFS_I(inode);
     struct posix_acl **acl = &reiserfs_i->i_acl_access;
-    if (strlen(name) != sizeof(XATTR_NAME_ACL_ACCESS)-1)
+    if (strlen(name) != sizeof(POSIX_ACL_XATTR_ACCESS)-1)
 	return -EINVAL;
     if (!IS_ERR (*acl) && *acl) {
         posix_acl_release (*acl);
@@ -510,7 +510,7 @@ posix_acl_access_list (struct inode *inode, const char *name, int namelen, char
 }
 
 struct reiserfs_xattr_handler posix_acl_access_handler = {
-	.prefix = XATTR_NAME_ACL_ACCESS,
+	.prefix = POSIX_ACL_XATTR_ACCESS,
 	.get = posix_acl_access_get,
 	.set = posix_acl_access_set,
 	.del = posix_acl_access_del,
@@ -521,7 +521,7 @@ static int
 posix_acl_default_get (struct inode *inode, const char *name,
 			   void *buffer, size_t size)
 {
-	if (strlen(name) != sizeof(XATTR_NAME_ACL_DEFAULT)-1)
+	if (strlen(name) != sizeof(POSIX_ACL_XATTR_DEFAULT)-1)
 		return -EINVAL;
 	return xattr_get_acl(inode, ACL_TYPE_DEFAULT, buffer, size);
 }
@@ -530,7 +530,7 @@ static int
 posix_acl_default_set(struct inode *inode, const char *name,
 			   const void *value, size_t size, int flags)
 {
-	if (strlen(name) != sizeof(XATTR_NAME_ACL_DEFAULT)-1)
+	if (strlen(name) != sizeof(POSIX_ACL_XATTR_DEFAULT)-1)
 		return -EINVAL;
 	return xattr_set_acl(inode, ACL_TYPE_DEFAULT, value, size);
 }
@@ -540,7 +540,7 @@ posix_acl_default_del (struct inode *inode, const char *name)
 {
     struct reiserfs_inode_info *reiserfs_i = REISERFS_I(inode);
     struct posix_acl **acl = &reiserfs_i->i_acl_default;
-    if (strlen(name) != sizeof(XATTR_NAME_ACL_DEFAULT)-1)
+    if (strlen(name) != sizeof(POSIX_ACL_XATTR_DEFAULT)-1)
 	return -EINVAL;
     if (!IS_ERR (*acl) && *acl) {
         posix_acl_release (*acl);
@@ -563,7 +563,7 @@ posix_acl_default_list (struct inode *inode, const char *name, int namelen, char
 }
 
 struct reiserfs_xattr_handler posix_acl_default_handler = {
-	.prefix = XATTR_NAME_ACL_DEFAULT,
+	.prefix = POSIX_ACL_XATTR_DEFAULT,
 	.get = posix_acl_default_get,
 	.set = posix_acl_default_set,
 	.del = posix_acl_default_del,
diff --git a/include/linux/posix_acl_xattr.h b/include/linux/posix_acl_xattr.h
index fe271c1947b..6e53c34035c 100644
--- a/include/linux/posix_acl_xattr.h
+++ b/include/linux/posix_acl_xattr.h
@@ -52,4 +52,7 @@ posix_acl_xattr_count(size_t size)
 	return size / sizeof(posix_acl_xattr_entry);
 }
 
+struct posix_acl *posix_acl_from_xattr(const void *value, size_t size);
+int posix_acl_to_xattr(const struct posix_acl *acl, void *buffer, size_t size);
+
 #endif	/* _POSIX_ACL_XATTR_H */
diff --git a/include/linux/reiserfs_acl.h b/include/linux/reiserfs_acl.h
index 2aef9c3f5ce..0760507a545 100644
--- a/include/linux/reiserfs_acl.h
+++ b/include/linux/reiserfs_acl.h
@@ -1,6 +1,5 @@
 #include <linux/init.h>
 #include <linux/posix_acl.h>
-#include <linux/xattr_acl.h>
 
 #define REISERFS_ACL_VERSION	0x0001
 
-- 
cgit v1.2.3-70-g09d2


From c43dc2fd885b5658cfd7cedb7bcca20910c517a4 Mon Sep 17 00:00:00 2001
From: Benjamin LaHaise <bcrl@kvack.org>
Date: Thu, 23 Jun 2005 00:10:27 -0700
Subject: [PATCH] aio: make wait_queue ->task ->private

In the upcoming aio_down patch, it is useful to store a private data
pointer in the kiocb's wait_queue.  Since we provide our own wake up
function and do not require the task_struct pointer, it makes sense to
convert the task pointer into a generic private pointer.

Signed-off-by: Benjamin LaHaise <benjamin.c.lahaise@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/wait.h | 16 ++++++++--------
 kernel/sched.c       |  2 +-
 2 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/linux/wait.h b/include/linux/wait.h
index c9486c3efb4..d38c9fecdc3 100644
--- a/include/linux/wait.h
+++ b/include/linux/wait.h
@@ -33,7 +33,7 @@ int default_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key
 struct __wait_queue {
 	unsigned int flags;
 #define WQ_FLAG_EXCLUSIVE	0x01
-	struct task_struct * task;
+	void *private;
 	wait_queue_func_t func;
 	struct list_head task_list;
 };
@@ -60,7 +60,7 @@ typedef struct __wait_queue_head wait_queue_head_t;
  */
 
 #define __WAITQUEUE_INITIALIZER(name, tsk) {				\
-	.task		= tsk,						\
+	.private	= tsk,						\
 	.func		= default_wake_function,			\
 	.task_list	= { NULL, NULL } }
 
@@ -86,7 +86,7 @@ static inline void init_waitqueue_head(wait_queue_head_t *q)
 static inline void init_waitqueue_entry(wait_queue_t *q, struct task_struct *p)
 {
 	q->flags = 0;
-	q->task = p;
+	q->private = p;
 	q->func = default_wake_function;
 }
 
@@ -94,7 +94,7 @@ static inline void init_waitqueue_func_entry(wait_queue_t *q,
 					wait_queue_func_t func)
 {
 	q->flags = 0;
-	q->task = NULL;
+	q->private = NULL;
 	q->func = func;
 }
 
@@ -110,7 +110,7 @@ static inline int waitqueue_active(wait_queue_head_t *q)
  * aio specifies a wait queue entry with an async notification
  * callback routine, not associated with any task.
  */
-#define is_sync_wait(wait)	(!(wait) || ((wait)->task))
+#define is_sync_wait(wait)	(!(wait) || ((wait)->private))
 
 extern void FASTCALL(add_wait_queue(wait_queue_head_t *q, wait_queue_t * wait));
 extern void FASTCALL(add_wait_queue_exclusive(wait_queue_head_t *q, wait_queue_t * wait));
@@ -384,7 +384,7 @@ int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *key);
 
 #define DEFINE_WAIT(name)						\
 	wait_queue_t name = {						\
-		.task		= current,				\
+		.private	= current,				\
 		.func		= autoremove_wake_function,		\
 		.task_list	= LIST_HEAD_INIT((name).task_list),	\
 	}
@@ -393,7 +393,7 @@ int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *key);
 	struct wait_bit_queue name = {					\
 		.key = __WAIT_BIT_KEY_INITIALIZER(word, bit),		\
 		.wait	= {						\
-			.task		= current,			\
+			.private	= current,			\
 			.func		= wake_bit_function,		\
 			.task_list	=				\
 				LIST_HEAD_INIT((name).wait.task_list),	\
@@ -402,7 +402,7 @@ int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *key);
 
 #define init_wait(wait)							\
 	do {								\
-		(wait)->task = current;					\
+		(wait)->private = current;				\
 		(wait)->func = autoremove_wake_function;		\
 		INIT_LIST_HEAD(&(wait)->task_list);			\
 	} while (0)
diff --git a/kernel/sched.c b/kernel/sched.c
index 6ee4515d5a2..76080d142e3 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -2869,7 +2869,7 @@ need_resched:
 
 int default_wake_function(wait_queue_t *curr, unsigned mode, int sync, void *key)
 {
-	task_t *p = curr->task;
+	task_t *p = curr->private;
 	return try_to_wake_up(p, mode, sync);
 }
 
-- 
cgit v1.2.3-70-g09d2


From bfb07599da289881d3bcbb601a110e997fc7444b Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Thu, 23 Jun 2005 00:10:32 -0700
Subject: [PATCH] Introduce tty_unregister_ldisc()

It's a bit strange to see tty_register_ldisc call in modules' exit
functions.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/tty.txt |  2 +-
 drivers/char/tty_io.c | 37 ++++++++++++++++++++++++-------------
 include/linux/tty.h   |  1 +
 3 files changed, 26 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/Documentation/tty.txt b/Documentation/tty.txt
index 3958cf746dd..8ff7bc2a081 100644
--- a/Documentation/tty.txt
+++ b/Documentation/tty.txt
@@ -22,7 +22,7 @@ copy of the structure. You must not re-register over the top of the line
 discipline even with the same data or your computer again will be eaten by
 demons.
 
-In order to remove a line discipline call tty_register_ldisc passing NULL.
+In order to remove a line discipline call tty_unregister_ldisc().
 In ancient times this always worked. In modern times the function will
 return -EBUSY if the ldisc is currently in use. Since the ldisc referencing
 code manages the module counts this should not usually be a concern.
diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index 31831030f73..cc4b43bad70 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -251,7 +251,7 @@ static void tty_set_termios_ldisc(struct tty_struct *tty, int num)
  
 static DEFINE_SPINLOCK(tty_ldisc_lock);
 static DECLARE_WAIT_QUEUE_HEAD(tty_ldisc_wait);
-static struct tty_ldisc tty_ldiscs[NR_LDISCS];	/* line disc dispatch table	*/
+static struct tty_ldisc tty_ldiscs[NR_LDISCS];	/* line disc dispatch table */
 
 int tty_register_ldisc(int disc, struct tty_ldisc *new_ldisc)
 {
@@ -262,24 +262,35 @@ int tty_register_ldisc(int disc, struct tty_ldisc *new_ldisc)
 		return -EINVAL;
 	
 	spin_lock_irqsave(&tty_ldisc_lock, flags);
-	if (new_ldisc) {
-		tty_ldiscs[disc] = *new_ldisc;
-		tty_ldiscs[disc].num = disc;
-		tty_ldiscs[disc].flags |= LDISC_FLAG_DEFINED;
-		tty_ldiscs[disc].refcount = 0;
-	} else {
-		if(tty_ldiscs[disc].refcount)
-			ret = -EBUSY;
-		else
-			tty_ldiscs[disc].flags &= ~LDISC_FLAG_DEFINED;
-	}
+	tty_ldiscs[disc] = *new_ldisc;
+	tty_ldiscs[disc].num = disc;
+	tty_ldiscs[disc].flags |= LDISC_FLAG_DEFINED;
+	tty_ldiscs[disc].refcount = 0;
 	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
 	
 	return ret;
 }
-
 EXPORT_SYMBOL(tty_register_ldisc);
 
+int tty_unregister_ldisc(int disc)
+{
+	unsigned long flags;
+	int ret = 0;
+
+	if (disc < N_TTY || disc >= NR_LDISCS)
+		return -EINVAL;
+
+	spin_lock_irqsave(&tty_ldisc_lock, flags);
+	if (tty_ldiscs[disc].refcount)
+		ret = -EBUSY;
+	else
+		tty_ldiscs[disc].flags &= ~LDISC_FLAG_DEFINED;
+	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
+
+	return ret;
+}
+EXPORT_SYMBOL(tty_unregister_ldisc);
+
 struct tty_ldisc *tty_ldisc_get(int disc)
 {
 	unsigned long flags;
diff --git a/include/linux/tty.h b/include/linux/tty.h
index 1b76106272d..59ff42c629e 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -345,6 +345,7 @@ extern int tty_check_change(struct tty_struct * tty);
 extern void stop_tty(struct tty_struct * tty);
 extern void start_tty(struct tty_struct * tty);
 extern int tty_register_ldisc(int disc, struct tty_ldisc *new_ldisc);
+extern int tty_unregister_ldisc(int disc);
 extern int tty_register_driver(struct tty_driver *driver);
 extern int tty_unregister_driver(struct tty_driver *driver);
 extern void tty_register_device(struct tty_driver *driver, unsigned index, struct device *dev);
-- 
cgit v1.2.3-70-g09d2


From 4749f32da939d4e4160541b2cadc22492bb507ec Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Thu, 23 Jun 2005 11:36:56 +0200
Subject: [PATCH] better USB_MON dependencies

This makes the USB_MON less confusing.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/usb/core/hcd.c   |  2 +-
 drivers/usb/core/hcd.h   |  2 +-
 drivers/usb/mon/Kconfig  | 13 ++++---------
 drivers/usb/mon/Makefile |  2 +-
 include/linux/usb.h      |  2 +-
 5 files changed, 8 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index d041782e0c8..0da23732e80 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -1794,7 +1794,7 @@ EXPORT_SYMBOL (usb_remove_hcd);
 
 /*-------------------------------------------------------------------------*/
 
-#if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE)
+#if defined(CONFIG_USB_MON)
 
 struct usb_mon_operations *mon_ops;
 
diff --git a/drivers/usb/core/hcd.h b/drivers/usb/core/hcd.h
index f67cf1e634f..325a51656c3 100644
--- a/drivers/usb/core/hcd.h
+++ b/drivers/usb/core/hcd.h
@@ -399,7 +399,7 @@ static inline void usbfs_cleanup(void) { }
 
 /*-------------------------------------------------------------------------*/
 
-#if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE)
+#if defined(CONFIG_USB_MON)
 
 struct usb_mon_operations {
 	void (*urb_submit)(struct usb_bus *bus, struct urb *urb);
diff --git a/drivers/usb/mon/Kconfig b/drivers/usb/mon/Kconfig
index 4e6152aa5f1..777642e26b9 100644
--- a/drivers/usb/mon/Kconfig
+++ b/drivers/usb/mon/Kconfig
@@ -2,13 +2,9 @@
 # USB Monitor configuration
 #
 
-# In normal life, it makes little sense to have usbmon as a module, and in fact
-# it is harmful, because there is no way to autoload the module.
-# The 'm' option is allowed for hackers who debug the usbmon itself,
-# and for those who have usbcore as a module.
 config USB_MON
-	tristate "USB Monitor"
-	depends on USB
+	bool "USB Monitor"
+	depends on USB!=n
 	default y
 	help
 	  If you say Y here, a component which captures the USB traffic
@@ -17,6 +13,5 @@ config USB_MON
 	  Harding's USBMon.
 
 	  This is somewhat experimental at this time, but it should be safe,
-	  as long as you aren't building this as a module and then removing it.
-
-	  If unsure, say Y. Do not say M.
+	  as long as you aren't using modular USB and try to remove this
+	  module.
diff --git a/drivers/usb/mon/Makefile b/drivers/usb/mon/Makefile
index 3cff8d444bb..f18d10ce91f 100644
--- a/drivers/usb/mon/Makefile
+++ b/drivers/usb/mon/Makefile
@@ -4,4 +4,4 @@
 
 usbmon-objs	:= mon_main.o mon_stat.o mon_text.o
 
-obj-$(CONFIG_USB_MON)	+= usbmon.o
+obj-$(CONFIG_USB)	+= usbmon.o
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 3d508bf0840..eb282b58154 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -290,7 +290,7 @@ struct usb_bus {
 	struct class_device *class_dev;	/* class device for this bus */
 	struct kref kref;		/* handles reference counting this bus */
 	void (*release)(struct usb_bus *bus);	/* function to destroy this bus's memory */
-#if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE)
+#if defined(CONFIG_USB_MON)
 	struct mon_bus *mon_bus;	/* non-null when associated */
 	int monitored;			/* non-zero when monitored */
 #endif
-- 
cgit v1.2.3-70-g09d2


From 317a76f9a44b437d6301718f4e5d08bd93f98da7 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Thu, 23 Jun 2005 12:19:55 -0700
Subject: [TCP]: Add pluggable congestion control algorithm infrastructure.

Allow TCP to have multiple pluggable congestion control algorithms.
Algorithms are defined by a set of operations and can be built in
or modules.  The legacy "new RENO" algorithm is used as a starting
point and fallback.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/sysctl.h     |   9 +-
 include/linux/tcp.h        |  49 +--
 include/net/tcp.h          | 237 +++++----------
 net/ipv4/Makefile          |   3 +-
 net/ipv4/sysctl_net_ipv4.c | 114 +++----
 net/ipv4/tcp.c             |   2 +
 net/ipv4/tcp_cong.c        | 195 ++++++++++++
 net/ipv4/tcp_diag.c        |  20 +-
 net/ipv4/tcp_input.c       | 737 ++++-----------------------------------------
 net/ipv4/tcp_ipv4.c        |   3 +
 net/ipv4/tcp_minisocks.c   |   4 +-
 net/ipv4/tcp_output.c      |  23 +-
 net/ipv6/tcp_ipv6.c        |   2 +-
 13 files changed, 399 insertions(+), 999 deletions(-)
 create mode 100644 net/ipv4/tcp_cong.c

(limited to 'include')

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 614e939c78a..72965bfe6cf 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -333,21 +333,14 @@ enum
 	NET_TCP_FRTO=92,
 	NET_TCP_LOW_LATENCY=93,
 	NET_IPV4_IPFRAG_SECRET_INTERVAL=94,
-	NET_TCP_WESTWOOD=95,
 	NET_IPV4_IGMP_MAX_MSF=96,
 	NET_TCP_NO_METRICS_SAVE=97,
-	NET_TCP_VEGAS=98,
-	NET_TCP_VEGAS_ALPHA=99,
-	NET_TCP_VEGAS_BETA=100,
-	NET_TCP_VEGAS_GAMMA=101,
- 	NET_TCP_BIC=102,
- 	NET_TCP_BIC_FAST_CONVERGENCE=103,
-	NET_TCP_BIC_LOW_WINDOW=104,
 	NET_TCP_DEFAULT_WIN_SCALE=105,
 	NET_TCP_MODERATE_RCVBUF=106,
 	NET_TCP_TSO_WIN_DIVISOR=107,
 	NET_TCP_BIC_BETA=108,
 	NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR=109,
+	NET_TCP_CONG_CONTROL=110,
 };
 
 enum {
diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 97a7c9e03df..3ea75dd6640 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -203,13 +203,6 @@ struct tcp_sack_block {
 	__u32	end_seq;
 };
 
-enum tcp_congestion_algo {
-	TCP_RENO=0,
-	TCP_VEGAS,
-	TCP_WESTWOOD,
-	TCP_BIC,
-};
-
 struct tcp_options_received {
 /*	PAWS/RTTM data	*/
 	long	ts_recent_stamp;/* Time we stored ts_recent (for aging) */
@@ -305,7 +298,7 @@ struct tcp_sock {
 	__u8	reordering;	/* Packet reordering metric.		*/
 	__u8	frto_counter;	/* Number of new acks after RTO */
 
-	__u8	adv_cong;	/* Using Vegas, Westwood, or BIC */
+	__u8	unused;
 	__u8	defer_accept;	/* User waits for some data after accept() */
 
 /* RTT measurement */
@@ -401,37 +394,10 @@ struct tcp_sock {
 		__u32	time;
 	} rcvq_space;
 
-/* TCP Westwood structure */
-        struct {
-                __u32    bw_ns_est;        /* first bandwidth estimation..not too smoothed 8) */
-                __u32    bw_est;           /* bandwidth estimate */
-                __u32    rtt_win_sx;       /* here starts a new evaluation... */
-                __u32    bk;
-                __u32    snd_una;          /* used for evaluating the number of acked bytes */
-                __u32    cumul_ack;
-                __u32    accounted;
-                __u32    rtt;
-                __u32    rtt_min;          /* minimum observed RTT */
-        } westwood;
-
-/* Vegas variables */
-	struct {
-		__u32	beg_snd_nxt;	/* right edge during last RTT */
-		__u32	beg_snd_una;	/* left edge  during last RTT */
-		__u32	beg_snd_cwnd;	/* saves the size of the cwnd */
-		__u8	doing_vegas_now;/* if true, do vegas for this RTT */
-		__u16	cntRTT;		/* # of RTTs measured within last RTT */
-		__u32	minRTT;		/* min of RTTs measured within last RTT (in usec) */
-		__u32	baseRTT;	/* the min of all Vegas RTT measurements seen (in usec) */
-	} vegas;
-
-	/* BI TCP Parameters */
-	struct {
-		__u32	cnt;		/* increase cwnd by 1 after this number of ACKs */
-		__u32 	last_max_cwnd;	/* last maximium snd_cwnd */
-		__u32	last_cwnd;	/* the last snd_cwnd */
-		__u32   last_stamp;     /* time when updated last_cwnd */
-	} bictcp;
+	/* Pluggable TCP congestion control hook */
+	struct tcp_congestion_ops *ca_ops;
+	u32	ca_priv[16];
+#define TCP_CA_PRIV_SIZE	(16*sizeof(u32))
 };
 
 static inline struct tcp_sock *tcp_sk(const struct sock *sk)
@@ -439,6 +405,11 @@ static inline struct tcp_sock *tcp_sk(const struct sock *sk)
 	return (struct tcp_sock *)sk;
 }
 
+static inline void *tcp_ca(const struct tcp_sock *tp)
+{
+	return (void *) tp->ca_priv;
+}
+
 #endif
 
 #endif	/* _LINUX_TCP_H */
diff --git a/include/net/tcp.h b/include/net/tcp.h
index f730935b824..e427cf35915 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -505,25 +505,6 @@ static __inline__ int tcp_sk_listen_hashfn(struct sock *sk)
 #else
 # define TCP_TW_RECYCLE_TICK (12+2-TCP_TW_RECYCLE_SLOTS_LOG)
 #endif
-
-#define BICTCP_BETA_SCALE    1024	/* Scale factor beta calculation
-					 * max_cwnd = snd_cwnd * beta
-					 */
-#define BICTCP_MAX_INCREMENT 32		/*
-					 * Limit on the amount of
-					 * increment allowed during
-					 * binary search.
-					 */
-#define BICTCP_FUNC_OF_MIN_INCR 11	/*
-					 * log(B/Smin)/log(B/(B-1))+1,
-					 * Smin:min increment
-					 * B:log factor
-					 */
-#define BICTCP_B		4	 /*
-					  * In binary search,
-					  * go to point (max+min)/N
-					  */
-
 /*
  *	TCP option
  */
@@ -596,16 +577,7 @@ extern int sysctl_tcp_adv_win_scale;
 extern int sysctl_tcp_tw_reuse;
 extern int sysctl_tcp_frto;
 extern int sysctl_tcp_low_latency;
-extern int sysctl_tcp_westwood;
-extern int sysctl_tcp_vegas_cong_avoid;
-extern int sysctl_tcp_vegas_alpha;
-extern int sysctl_tcp_vegas_beta;
-extern int sysctl_tcp_vegas_gamma;
 extern int sysctl_tcp_nometrics_save;
-extern int sysctl_tcp_bic;
-extern int sysctl_tcp_bic_fast_convergence;
-extern int sysctl_tcp_bic_low_window;
-extern int sysctl_tcp_bic_beta;
 extern int sysctl_tcp_moderate_rcvbuf;
 extern int sysctl_tcp_tso_win_divisor;
 
@@ -1136,6 +1108,80 @@ static inline void tcp_packets_out_dec(struct tcp_sock *tp,
 	tp->packets_out -= tcp_skb_pcount(skb);
 }
 
+/* Events passed to congestion control interface */
+enum tcp_ca_event {
+	CA_EVENT_TX_START,	/* first transmit when no packets in flight */
+	CA_EVENT_CWND_RESTART,	/* congestion window restart */
+	CA_EVENT_COMPLETE_CWR,	/* end of congestion recovery */
+	CA_EVENT_FRTO,		/* fast recovery timeout */
+	CA_EVENT_LOSS,		/* loss timeout */
+	CA_EVENT_FAST_ACK,	/* in sequence ack */
+	CA_EVENT_SLOW_ACK,	/* other ack */
+};
+
+/*
+ * Interface for adding new TCP congestion control handlers
+ */
+#define TCP_CA_NAME_MAX	16
+struct tcp_congestion_ops {
+	struct list_head	list;
+
+	/* initialize private data (optional) */
+	void (*init)(struct tcp_sock *tp);
+	/* cleanup private data  (optional) */
+	void (*release)(struct tcp_sock *tp);
+
+	/* return slow start threshold (required) */
+	u32 (*ssthresh)(struct tcp_sock *tp);
+	/* lower bound for congestion window (optional) */
+	u32 (*min_cwnd)(struct tcp_sock *tp);
+	/* do new cwnd calculation (required) */
+	void (*cong_avoid)(struct tcp_sock *tp, u32 ack,
+			   u32 rtt, u32 in_flight, int good_ack);
+	/* round trip time sample per acked packet (optional) */
+	void (*rtt_sample)(struct tcp_sock *tp, u32 usrtt);
+	/* call before changing ca_state (optional) */
+	void (*set_state)(struct tcp_sock *tp, u8 new_state);
+	/* call when cwnd event occurs (optional) */
+	void (*cwnd_event)(struct tcp_sock *tp, enum tcp_ca_event ev);
+	/* new value of cwnd after loss (optional) */
+	u32  (*undo_cwnd)(struct tcp_sock *tp);
+	/* hook for packet ack accounting (optional) */
+	void (*pkts_acked)(struct tcp_sock *tp, u32 num_acked);
+	/* get info for tcp_diag (optional) */
+	void (*get_info)(struct tcp_sock *tp, u32 ext, struct sk_buff *skb);
+
+	char 		name[TCP_CA_NAME_MAX];
+	struct module 	*owner;
+};
+
+extern int tcp_register_congestion_control(struct tcp_congestion_ops *type);
+extern void tcp_unregister_congestion_control(struct tcp_congestion_ops *type);
+
+extern void tcp_init_congestion_control(struct tcp_sock *tp);
+extern void tcp_cleanup_congestion_control(struct tcp_sock *tp);
+extern int tcp_set_default_congestion_control(const char *name);
+extern void tcp_get_default_congestion_control(char *name);
+
+extern struct tcp_congestion_ops tcp_reno;
+extern u32 tcp_reno_ssthresh(struct tcp_sock *tp);
+extern void tcp_reno_cong_avoid(struct tcp_sock *tp, u32 ack,
+				u32 rtt, u32 in_flight, int flag);
+extern u32 tcp_reno_min_cwnd(struct tcp_sock *tp);
+
+static inline void tcp_set_ca_state(struct tcp_sock *tp, u8 ca_state)
+{
+	if (tp->ca_ops->set_state)
+		tp->ca_ops->set_state(tp, ca_state);
+	tp->ca_state = ca_state;
+}
+
+static inline void tcp_ca_event(struct tcp_sock *tp, enum tcp_ca_event event)
+{
+	if (tp->ca_ops->cwnd_event)
+		tp->ca_ops->cwnd_event(tp, event);
+}
+
 /* This determines how many packets are "in the network" to the best
  * of our knowledge.  In many cases it is conservative, but where
  * detailed information is available from the receiver (via SACK
@@ -1155,91 +1201,6 @@ static __inline__ unsigned int tcp_packets_in_flight(const struct tcp_sock *tp)
 	return (tp->packets_out - tp->left_out + tp->retrans_out);
 }
 
-/*
- * Which congestion algorithim is in use on the connection.
- */
-#define tcp_is_vegas(__tp)	((__tp)->adv_cong == TCP_VEGAS)
-#define tcp_is_westwood(__tp)	((__tp)->adv_cong == TCP_WESTWOOD)
-#define tcp_is_bic(__tp)	((__tp)->adv_cong == TCP_BIC)
-
-/* Recalculate snd_ssthresh, we want to set it to:
- *
- * Reno:
- * 	one half the current congestion window, but no
- *	less than two segments
- *
- * BIC:
- *	behave like Reno until low_window is reached,
- *	then increase congestion window slowly
- */
-static inline __u32 tcp_recalc_ssthresh(struct tcp_sock *tp)
-{
-	if (tcp_is_bic(tp)) {
-		if (sysctl_tcp_bic_fast_convergence &&
-		    tp->snd_cwnd < tp->bictcp.last_max_cwnd)
-			tp->bictcp.last_max_cwnd = (tp->snd_cwnd * 
-						    (BICTCP_BETA_SCALE
-						     + sysctl_tcp_bic_beta))
-				/ (2 * BICTCP_BETA_SCALE);
-		else
-			tp->bictcp.last_max_cwnd = tp->snd_cwnd;
-
-		if (tp->snd_cwnd > sysctl_tcp_bic_low_window)
-			return max((tp->snd_cwnd * sysctl_tcp_bic_beta)
-				   / BICTCP_BETA_SCALE, 2U);
-	}
-
-	return max(tp->snd_cwnd >> 1U, 2U);
-}
-
-/* Stop taking Vegas samples for now. */
-#define tcp_vegas_disable(__tp)	((__tp)->vegas.doing_vegas_now = 0)
-    
-static inline void tcp_vegas_enable(struct tcp_sock *tp)
-{
-	/* There are several situations when we must "re-start" Vegas:
-	 *
-	 *  o when a connection is established
-	 *  o after an RTO
-	 *  o after fast recovery
-	 *  o when we send a packet and there is no outstanding
-	 *    unacknowledged data (restarting an idle connection)
-	 *
-	 * In these circumstances we cannot do a Vegas calculation at the
-	 * end of the first RTT, because any calculation we do is using
-	 * stale info -- both the saved cwnd and congestion feedback are
-	 * stale.
-	 *
-	 * Instead we must wait until the completion of an RTT during
-	 * which we actually receive ACKs.
-	 */
-    
-	/* Begin taking Vegas samples next time we send something. */
-	tp->vegas.doing_vegas_now = 1;
-     
-	/* Set the beginning of the next send window. */
-	tp->vegas.beg_snd_nxt = tp->snd_nxt;
-
-	tp->vegas.cntRTT = 0;
-	tp->vegas.minRTT = 0x7fffffff;
-}
-
-/* Should we be taking Vegas samples right now? */
-#define tcp_vegas_enabled(__tp)	((__tp)->vegas.doing_vegas_now)
-
-extern void tcp_ca_init(struct tcp_sock *tp);
-
-static inline void tcp_set_ca_state(struct tcp_sock *tp, u8 ca_state)
-{
-	if (tcp_is_vegas(tp)) {
-		if (ca_state == TCP_CA_Open) 
-			tcp_vegas_enable(tp);
-		else
-			tcp_vegas_disable(tp);
-	}
-	tp->ca_state = ca_state;
-}
-
 /* If cwnd > ssthresh, we may raise ssthresh to be half-way to cwnd.
  * The exception is rate halving phase, when cwnd is decreasing towards
  * ssthresh.
@@ -1288,7 +1249,7 @@ static inline void tcp_cwnd_validate(struct sock *sk, struct tcp_sock *tp)
 static inline void __tcp_enter_cwr(struct tcp_sock *tp)
 {
 	tp->undo_marker = 0;
-	tp->snd_ssthresh = tcp_recalc_ssthresh(tp);
+	tp->snd_ssthresh = tp->ca_ops->ssthresh(tp);
 	tp->snd_cwnd = min(tp->snd_cwnd,
 			   tcp_packets_in_flight(tp) + 1U);
 	tp->snd_cwnd_cnt = 0;
@@ -1876,52 +1837,4 @@ struct tcp_iter_state {
 extern int tcp_proc_register(struct tcp_seq_afinfo *afinfo);
 extern void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo);
 
-/* TCP Westwood functions and constants */
-
-#define TCP_WESTWOOD_INIT_RTT  (20*HZ)           /* maybe too conservative?! */
-#define TCP_WESTWOOD_RTT_MIN   (HZ/20)           /* 50ms */
-
-static inline void tcp_westwood_update_rtt(struct tcp_sock *tp, __u32 rtt_seq)
-{
-        if (tcp_is_westwood(tp))
-                tp->westwood.rtt = rtt_seq;
-}
-
-static inline __u32 __tcp_westwood_bw_rttmin(const struct tcp_sock *tp)
-{
-        return max((tp->westwood.bw_est) * (tp->westwood.rtt_min) /
-		   (__u32) (tp->mss_cache_std),
-		   2U);
-}
-
-static inline __u32 tcp_westwood_bw_rttmin(const struct tcp_sock *tp)
-{
-	return tcp_is_westwood(tp) ? __tcp_westwood_bw_rttmin(tp) : 0;
-}
-
-static inline int tcp_westwood_ssthresh(struct tcp_sock *tp)
-{
-	__u32 ssthresh = 0;
-
-	if (tcp_is_westwood(tp)) {
-		ssthresh = __tcp_westwood_bw_rttmin(tp);
-		if (ssthresh)
-			tp->snd_ssthresh = ssthresh;  
-	}
-
-	return (ssthresh != 0);
-}
-
-static inline int tcp_westwood_cwnd(struct tcp_sock *tp)
-{
-	__u32 cwnd = 0;
-
-	if (tcp_is_westwood(tp)) {
-		cwnd = __tcp_westwood_bw_rttmin(tp);
-		if (cwnd)
-			tp->snd_cwnd = cwnd;
-	}
-
-	return (cwnd != 0);
-}
 #endif	/* _TCP_H */
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index 65d57d8e1ad..89c0b4cb470 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -5,7 +5,8 @@
 obj-y     := utils.o route.o inetpeer.o protocol.o \
 	     ip_input.o ip_fragment.o ip_forward.o ip_options.o \
 	     ip_output.o ip_sockglue.o \
-	     tcp.o tcp_input.o tcp_output.o tcp_timer.o tcp_ipv4.o tcp_minisocks.o \
+	     tcp.o tcp_input.o tcp_output.o tcp_timer.o tcp_ipv4.o \
+	     tcp_minisocks.o tcp_cong.o \
 	     datagram.o raw.o udp.o arp.o icmp.o devinet.o af_inet.o igmp.o \
 	     sysctl_net_ipv4.o fib_frontend.o fib_semantics.o
 
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 23068bddbf0..e3289453241 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -118,6 +118,45 @@ static int ipv4_sysctl_forward_strategy(ctl_table *table,
 	return 1;
 }
 
+static int proc_tcp_congestion_control(ctl_table *ctl, int write, struct file * filp,
+				       void __user *buffer, size_t *lenp, loff_t *ppos)
+{
+	char val[TCP_CA_NAME_MAX];
+	ctl_table tbl = {
+		.data = val,
+		.maxlen = TCP_CA_NAME_MAX,
+	};
+	int ret;
+
+	tcp_get_default_congestion_control(val);
+
+	ret = proc_dostring(&tbl, write, filp, buffer, lenp, ppos);
+	if (write && ret == 0)
+		ret = tcp_set_default_congestion_control(val);
+	return ret;
+}
+
+int sysctl_tcp_congestion_control(ctl_table *table, int __user *name, int nlen,
+				  void __user *oldval, size_t __user *oldlenp,
+				  void __user *newval, size_t newlen,
+				  void **context)
+{
+	char val[TCP_CA_NAME_MAX];
+	ctl_table tbl = {
+		.data = val,
+		.maxlen = TCP_CA_NAME_MAX,
+	};
+	int ret;
+
+	tcp_get_default_congestion_control(val);
+	ret = sysctl_string(&tbl, name, nlen, oldval, oldlenp, newval, newlen,
+			    context);
+	if (ret == 0 && newval && newlen)
+		ret = tcp_set_default_congestion_control(val);
+	return ret;
+}
+
+
 ctl_table ipv4_table[] = {
         {
 		.ctl_name	= NET_IPV4_TCP_TIMESTAMPS,
@@ -611,70 +650,6 @@ ctl_table ipv4_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
 	},
-	{
-		.ctl_name	= NET_TCP_WESTWOOD, 
-		.procname	= "tcp_westwood",
-		.data		= &sysctl_tcp_westwood,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec,
-	},
-	{
-		.ctl_name	= NET_TCP_VEGAS,
-		.procname	= "tcp_vegas_cong_avoid",
-		.data		= &sysctl_tcp_vegas_cong_avoid,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec,
-	},
-	{
-		.ctl_name	= NET_TCP_VEGAS_ALPHA,
-		.procname	= "tcp_vegas_alpha",
-		.data		= &sysctl_tcp_vegas_alpha,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec,
-	},
-	{
-		.ctl_name	= NET_TCP_VEGAS_BETA,
-		.procname	= "tcp_vegas_beta",
-		.data		= &sysctl_tcp_vegas_beta,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec,
-	},
-	{
-		.ctl_name	= NET_TCP_VEGAS_GAMMA,
-		.procname	= "tcp_vegas_gamma",
-		.data		= &sysctl_tcp_vegas_gamma,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec,
-	},
-	{
-		.ctl_name	= NET_TCP_BIC,
-		.procname	= "tcp_bic",
-		.data		= &sysctl_tcp_bic,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec,
-	},
-	{
-		.ctl_name	= NET_TCP_BIC_FAST_CONVERGENCE,
-		.procname	= "tcp_bic_fast_convergence",
-		.data		= &sysctl_tcp_bic_fast_convergence,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec,
-	},
-	{
-		.ctl_name	= NET_TCP_BIC_LOW_WINDOW,
-		.procname	= "tcp_bic_low_window",
-		.data		= &sysctl_tcp_bic_low_window,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec,
-	},
 	{
 		.ctl_name	= NET_TCP_MODERATE_RCVBUF,
 		.procname	= "tcp_moderate_rcvbuf",
@@ -692,13 +667,14 @@ ctl_table ipv4_table[] = {
 		.proc_handler	= &proc_dointvec,
 	},
 	{
-		.ctl_name	= NET_TCP_BIC_BETA,
-		.procname	= "tcp_bic_beta",
-		.data		= &sysctl_tcp_bic_beta,
-		.maxlen		= sizeof(int),
+		.ctl_name	= NET_TCP_CONG_CONTROL,
+		.procname	= "tcp_congestion_control",
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec,
+		.maxlen		= TCP_CA_NAME_MAX,
+		.proc_handler	= &proc_tcp_congestion_control,
+		.strategy	= &sysctl_tcp_congestion_control,
 	},
+
 	{ .ctl_name = 0 }
 };
 
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 674bbd8cfd3..f3dbc8dc126 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2333,6 +2333,8 @@ void __init tcp_init(void)
 	printk(KERN_INFO "TCP: Hash tables configured "
 	       "(established %d bind %d)\n",
 	       tcp_ehash_size << 1, tcp_bhash_size);
+
+	tcp_register_congestion_control(&tcp_reno);
 }
 
 EXPORT_SYMBOL(tcp_accept);
diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c
new file mode 100644
index 00000000000..665394a63ae
--- /dev/null
+++ b/net/ipv4/tcp_cong.c
@@ -0,0 +1,195 @@
+/*
+ * Plugable TCP congestion control support and newReno
+ * congestion control.
+ * Based on ideas from I/O scheduler suport and Web100.
+ *
+ * Copyright (C) 2005 Stephen Hemminger <shemminger@osdl.org>
+ */
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/mm.h>
+#include <linux/types.h>
+#include <linux/list.h>
+#include <net/tcp.h>
+
+static DEFINE_SPINLOCK(tcp_cong_list_lock);
+static LIST_HEAD(tcp_cong_list);
+
+/* Simple linear search, don't expect many entries! */
+static struct tcp_congestion_ops *tcp_ca_find(const char *name)
+{
+	struct tcp_congestion_ops *e;
+
+	list_for_each_entry(e, &tcp_cong_list, list) {
+		if (strcmp(e->name, name) == 0)
+			return e;
+	}
+
+	return NULL;
+}
+
+/*
+ * Attach new congestion control algorthim to the list
+ * of available options.
+ */
+int tcp_register_congestion_control(struct tcp_congestion_ops *ca)
+{
+	int ret = 0;
+
+	/* all algorithms must implement ssthresh and cong_avoid ops */
+	if (!ca->ssthresh || !ca->cong_avoid || !ca->min_cwnd) {
+		printk(KERN_ERR "TCP %s does not implement required ops\n",
+		       ca->name);
+		return -EINVAL;
+	}
+
+	spin_lock(&tcp_cong_list_lock);
+	if (tcp_ca_find(ca->name)) {
+		printk(KERN_NOTICE "TCP %s already registered\n", ca->name);
+		ret = -EEXIST;
+	} else {
+		list_add_rcu(&ca->list, &tcp_cong_list);
+		printk(KERN_INFO "TCP %s registered\n", ca->name);
+	}
+	spin_unlock(&tcp_cong_list_lock);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(tcp_register_congestion_control);
+
+/*
+ * Remove congestion control algorithm, called from
+ * the module's remove function.  Module ref counts are used
+ * to ensure that this can't be done till all sockets using
+ * that method are closed.
+ */
+void tcp_unregister_congestion_control(struct tcp_congestion_ops *ca)
+{
+	spin_lock(&tcp_cong_list_lock);
+	list_del_rcu(&ca->list);
+	spin_unlock(&tcp_cong_list_lock);
+}
+EXPORT_SYMBOL_GPL(tcp_unregister_congestion_control);
+
+/* Assign choice of congestion control. */
+void tcp_init_congestion_control(struct tcp_sock *tp)
+{
+	struct tcp_congestion_ops *ca;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(ca, &tcp_cong_list, list) {
+		if (try_module_get(ca->owner)) {
+			tp->ca_ops = ca;
+			break;
+		}
+
+	}
+	rcu_read_unlock();
+
+	if (tp->ca_ops->init)
+		tp->ca_ops->init(tp);
+}
+
+/* Manage refcounts on socket close. */
+void tcp_cleanup_congestion_control(struct tcp_sock *tp)
+{
+	if (tp->ca_ops->release)
+		tp->ca_ops->release(tp);
+	module_put(tp->ca_ops->owner);
+}
+
+/* Used by sysctl to change default congestion control */
+int tcp_set_default_congestion_control(const char *name)
+{
+	struct tcp_congestion_ops *ca;
+	int ret = -ENOENT;
+
+	spin_lock(&tcp_cong_list_lock);
+	ca = tcp_ca_find(name);
+#ifdef CONFIG_KMOD
+	if (!ca) {
+		spin_unlock(&tcp_cong_list_lock);
+
+		request_module("tcp_%s", name);
+		spin_lock(&tcp_cong_list_lock);
+		ca = tcp_ca_find(name);
+	}
+#endif
+
+	if (ca) {
+		list_move(&ca->list, &tcp_cong_list);
+		ret = 0;
+	}
+	spin_unlock(&tcp_cong_list_lock);
+
+	return ret;
+}
+
+/* Get current default congestion control */
+void tcp_get_default_congestion_control(char *name)
+{
+	struct tcp_congestion_ops *ca;
+	/* We will always have reno... */
+	BUG_ON(list_empty(&tcp_cong_list));
+
+	rcu_read_lock();
+	ca = list_entry(tcp_cong_list.next, struct tcp_congestion_ops, list);
+	strncpy(name, ca->name, TCP_CA_NAME_MAX);
+	rcu_read_unlock();
+}
+
+/*
+ * TCP Reno congestion control
+ * This is special case used for fallback as well.
+ */
+/* This is Jacobson's slow start and congestion avoidance.
+ * SIGCOMM '88, p. 328.
+ */
+void tcp_reno_cong_avoid(struct tcp_sock *tp, u32 ack, u32 rtt, u32 in_flight,
+			 int flag)
+{
+	if (in_flight < tp->snd_cwnd)
+		return;
+
+        if (tp->snd_cwnd <= tp->snd_ssthresh) {
+                /* In "safe" area, increase. */
+		if (tp->snd_cwnd < tp->snd_cwnd_clamp)
+			tp->snd_cwnd++;
+	} else {
+                /* In dangerous area, increase slowly.
+		 * In theory this is tp->snd_cwnd += 1 / tp->snd_cwnd
+		 */
+		if (tp->snd_cwnd_cnt >= tp->snd_cwnd) {
+			if (tp->snd_cwnd < tp->snd_cwnd_clamp)
+				tp->snd_cwnd++;
+			tp->snd_cwnd_cnt = 0;
+		} else
+			tp->snd_cwnd_cnt++;
+	}
+}
+EXPORT_SYMBOL_GPL(tcp_reno_cong_avoid);
+
+/* Slow start threshold is half the congestion window (min 2) */
+u32 tcp_reno_ssthresh(struct tcp_sock *tp)
+{
+	return max(tp->snd_cwnd >> 1U, 2U);
+}
+EXPORT_SYMBOL_GPL(tcp_reno_ssthresh);
+
+/* Lower bound on congestion window. */
+u32 tcp_reno_min_cwnd(struct tcp_sock *tp)
+{
+	return tp->snd_ssthresh/2;
+}
+EXPORT_SYMBOL_GPL(tcp_reno_min_cwnd);
+
+struct tcp_congestion_ops tcp_reno = {
+	.name		= "reno",
+	.owner		= THIS_MODULE,
+	.ssthresh	= tcp_reno_ssthresh,
+	.cong_avoid	= tcp_reno_cong_avoid,
+	.min_cwnd	= tcp_reno_min_cwnd,
+};
+
+EXPORT_SYMBOL_GPL(tcp_reno);
diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c
index 634befc0792..867acc0f79d 100644
--- a/net/ipv4/tcp_diag.c
+++ b/net/ipv4/tcp_diag.c
@@ -42,7 +42,6 @@ struct tcpdiag_entry
 
 static struct sock *tcpnl;
 
-
 #define TCPDIAG_PUT(skb, attrtype, attrlen) \
 ({ int rtalen = RTA_LENGTH(attrlen);        \
    struct rtattr *rta;                      \
@@ -61,7 +60,6 @@ static int tcpdiag_fill(struct sk_buff *skb, struct sock *sk,
 	struct nlmsghdr  *nlh;
 	struct tcp_info  *info = NULL;
 	struct tcpdiag_meminfo  *minfo = NULL;
-	struct tcpvegas_info *vinfo = NULL;
 	unsigned char	 *b = skb->tail;
 
 	nlh = NLMSG_PUT(skb, pid, seq, TCPDIAG_GETSOCK, sizeof(*r));
@@ -73,9 +71,6 @@ static int tcpdiag_fill(struct sk_buff *skb, struct sock *sk,
 		if (ext & (1<<(TCPDIAG_INFO-1)))
 			info = TCPDIAG_PUT(skb, TCPDIAG_INFO, sizeof(*info));
 		
-		if ((tcp_is_westwood(tp) || tcp_is_vegas(tp))
-		    && (ext & (1<<(TCPDIAG_VEGASINFO-1))))
-			vinfo = TCPDIAG_PUT(skb, TCPDIAG_VEGASINFO, sizeof(*vinfo));
 	}
 	r->tcpdiag_family = sk->sk_family;
 	r->tcpdiag_state = sk->sk_state;
@@ -166,19 +161,8 @@ static int tcpdiag_fill(struct sk_buff *skb, struct sock *sk,
 	if (info) 
 		tcp_get_info(sk, info);
 
-	if (vinfo) {
-		if (tcp_is_vegas(tp)) {
-			vinfo->tcpv_enabled = tp->vegas.doing_vegas_now;
-			vinfo->tcpv_rttcnt = tp->vegas.cntRTT;
-			vinfo->tcpv_rtt = jiffies_to_usecs(tp->vegas.baseRTT);
-			vinfo->tcpv_minrtt = jiffies_to_usecs(tp->vegas.minRTT);
-		} else {
-			vinfo->tcpv_enabled = 0;
-			vinfo->tcpv_rttcnt = 0;
-			vinfo->tcpv_rtt = jiffies_to_usecs(tp->westwood.rtt);
-			vinfo->tcpv_minrtt = jiffies_to_usecs(tp->westwood.rtt_min);
-		}
-	}
+	if (sk->sk_state < TCP_TIME_WAIT && tp->ca_ops->get_info)
+		tp->ca_ops->get_info(tp, ext, skb);
 
 	nlh->nlmsg_len = skb->tail - b;
 	return skb->len;
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 5bad504630a..7bbbbc33eb4 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -61,7 +61,6 @@
  *		Panu Kuhlberg:		Experimental audit of TCP (re)transmission
  *					engine. Lots of bugs are found.
  *		Pasi Sarolahti:		F-RTO for dealing with spurious RTOs
- *		Angelo Dell'Aera:	TCP Westwood+ support
  */
 
 #include <linux/config.h>
@@ -88,23 +87,9 @@ int sysctl_tcp_rfc1337;
 int sysctl_tcp_max_orphans = NR_FILE;
 int sysctl_tcp_frto;
 int sysctl_tcp_nometrics_save;
-int sysctl_tcp_westwood;
-int sysctl_tcp_vegas_cong_avoid;
 
 int sysctl_tcp_moderate_rcvbuf = 1;
 
-/* Default values of the Vegas variables, in fixed-point representation
- * with V_PARAM_SHIFT bits to the right of the binary point.
- */
-#define V_PARAM_SHIFT 1
-int sysctl_tcp_vegas_alpha = 1<<V_PARAM_SHIFT;
-int sysctl_tcp_vegas_beta  = 3<<V_PARAM_SHIFT;
-int sysctl_tcp_vegas_gamma = 1<<V_PARAM_SHIFT;
-int sysctl_tcp_bic = 1;
-int sysctl_tcp_bic_fast_convergence = 1;
-int sysctl_tcp_bic_low_window = 14;
-int sysctl_tcp_bic_beta = 819;		/* = 819/1024 (BICTCP_BETA_SCALE) */
-
 #define FLAG_DATA		0x01 /* Incoming frame contained data.		*/
 #define FLAG_WIN_UPDATE		0x02 /* Incoming ACK was a window update.	*/
 #define FLAG_DATA_ACKED		0x04 /* This ACK acknowledged new data.		*/
@@ -333,15 +318,6 @@ static void tcp_init_buffer_space(struct sock *sk)
 	tp->snd_cwnd_stamp = tcp_time_stamp;
 }
 
-static void init_bictcp(struct tcp_sock *tp)
-{
-	tp->bictcp.cnt = 0;
-
-	tp->bictcp.last_max_cwnd = 0;
-	tp->bictcp.last_cwnd = 0;
-	tp->bictcp.last_stamp = 0;
-}
-
 /* 5. Recalculate window clamp after socket hit its memory bounds. */
 static void tcp_clamp_window(struct sock *sk, struct tcp_sock *tp)
 {
@@ -558,45 +534,6 @@ static void tcp_event_data_recv(struct sock *sk, struct tcp_sock *tp, struct sk_
 		tcp_grow_window(sk, tp, skb);
 }
 
-/* When starting a new connection, pin down the current choice of 
- * congestion algorithm.
- */
-void tcp_ca_init(struct tcp_sock *tp)
-{
-	if (sysctl_tcp_westwood) 
-		tp->adv_cong = TCP_WESTWOOD;
-	else if (sysctl_tcp_bic)
-		tp->adv_cong = TCP_BIC;
-	else if (sysctl_tcp_vegas_cong_avoid) {
-		tp->adv_cong = TCP_VEGAS;
-		tp->vegas.baseRTT = 0x7fffffff;
-		tcp_vegas_enable(tp);
-	} 
-}
-
-/* Do RTT sampling needed for Vegas.
- * Basically we:
- *   o min-filter RTT samples from within an RTT to get the current
- *     propagation delay + queuing delay (we are min-filtering to try to
- *     avoid the effects of delayed ACKs)
- *   o min-filter RTT samples from a much longer window (forever for now)
- *     to find the propagation delay (baseRTT)
- */
-static inline void vegas_rtt_calc(struct tcp_sock *tp, __u32 rtt)
-{
-	__u32 vrtt = rtt + 1; /* Never allow zero rtt or baseRTT */
-
-	/* Filter to find propagation delay: */
-	if (vrtt < tp->vegas.baseRTT) 
-		tp->vegas.baseRTT = vrtt;
-
-	/* Find the min RTT during the last RTT to find
-	 * the current prop. delay + queuing delay:
-	 */
-	tp->vegas.minRTT = min(tp->vegas.minRTT, vrtt);
-	tp->vegas.cntRTT++;
-}
-
 /* Called to compute a smoothed rtt estimate. The data fed to this
  * routine either comes from timestamps, or from segments that were
  * known _not_ to have been retransmitted [see Karn/Partridge
@@ -606,13 +543,10 @@ static inline void vegas_rtt_calc(struct tcp_sock *tp, __u32 rtt)
  * To save cycles in the RFC 1323 implementation it was better to break
  * it up into three procedures. -- erics
  */
-static void tcp_rtt_estimator(struct tcp_sock *tp, __u32 mrtt)
+static void tcp_rtt_estimator(struct tcp_sock *tp, __u32 mrtt, u32 *usrtt)
 {
 	long m = mrtt; /* RTT */
 
-	if (tcp_vegas_enabled(tp))
-		vegas_rtt_calc(tp, mrtt);
-
 	/*	The following amusing code comes from Jacobson's
 	 *	article in SIGCOMM '88.  Note that rtt and mdev
 	 *	are scaled versions of rtt and mean deviation.
@@ -670,7 +604,8 @@ static void tcp_rtt_estimator(struct tcp_sock *tp, __u32 mrtt)
 		tp->rtt_seq = tp->snd_nxt;
 	}
 
-	tcp_westwood_update_rtt(tp, tp->srtt >> 3);
+	if (tp->ca_ops->rtt_sample)
+		tp->ca_ops->rtt_sample(tp, *usrtt);
 }
 
 /* Calculate rto without backoff.  This is the second half of Van Jacobson's
@@ -1185,8 +1120,8 @@ void tcp_enter_frto(struct sock *sk)
             tp->snd_una == tp->high_seq ||
             (tp->ca_state == TCP_CA_Loss && !tp->retransmits)) {
 		tp->prior_ssthresh = tcp_current_ssthresh(tp);
-		if (!tcp_westwood_ssthresh(tp))
-			tp->snd_ssthresh = tcp_recalc_ssthresh(tp);
+		tp->snd_ssthresh = tp->ca_ops->ssthresh(tp);
+		tcp_ca_event(tp, CA_EVENT_FRTO);
 	}
 
 	/* Have to clear retransmission markers here to keep the bookkeeping
@@ -1252,8 +1187,6 @@ static void tcp_enter_frto_loss(struct sock *sk)
 	tcp_set_ca_state(tp, TCP_CA_Loss);
 	tp->high_seq = tp->frto_highmark;
 	TCP_ECN_queue_cwr(tp);
-
-	init_bictcp(tp);
 }
 
 void tcp_clear_retrans(struct tcp_sock *tp)
@@ -1283,7 +1216,8 @@ void tcp_enter_loss(struct sock *sk, int how)
 	if (tp->ca_state <= TCP_CA_Disorder || tp->snd_una == tp->high_seq ||
 	    (tp->ca_state == TCP_CA_Loss && !tp->retransmits)) {
 		tp->prior_ssthresh = tcp_current_ssthresh(tp);
-		tp->snd_ssthresh = tcp_recalc_ssthresh(tp);
+		tp->snd_ssthresh = tp->ca_ops->ssthresh(tp);
+		tcp_ca_event(tp, CA_EVENT_LOSS);
 	}
 	tp->snd_cwnd	   = 1;
 	tp->snd_cwnd_cnt   = 0;
@@ -1596,28 +1530,14 @@ static inline void tcp_moderate_cwnd(struct tcp_sock *tp)
 }
 
 /* Decrease cwnd each second ack. */
-
 static void tcp_cwnd_down(struct tcp_sock *tp)
 {
 	int decr = tp->snd_cwnd_cnt + 1;
-	__u32 limit;
-
-	/*
-	 * TCP Westwood
-	 * Here limit is evaluated as BWestimation*RTTmin (for obtaining it
-	 * in packets we use mss_cache). If sysctl_tcp_westwood is off
-	 * tcp_westwood_bw_rttmin() returns 0. In such case snd_ssthresh is
-	 * still used as usual. It prevents other strange cases in which
-	 * BWE*RTTmin could assume value 0. It should not happen but...
-	 */
-
-	if (!(limit = tcp_westwood_bw_rttmin(tp)))
-		limit = tp->snd_ssthresh/2;
 
 	tp->snd_cwnd_cnt = decr&1;
 	decr >>= 1;
 
-	if (decr && tp->snd_cwnd > limit)
+	if (decr && tp->snd_cwnd > tp->ca_ops->min_cwnd(tp))
 		tp->snd_cwnd -= decr;
 
 	tp->snd_cwnd = min(tp->snd_cwnd, tcp_packets_in_flight(tp)+1);
@@ -1654,8 +1574,8 @@ static void DBGUNDO(struct sock *sk, struct tcp_sock *tp, const char *msg)
 static void tcp_undo_cwr(struct tcp_sock *tp, int undo)
 {
 	if (tp->prior_ssthresh) {
-		if (tcp_is_bic(tp))
-			tp->snd_cwnd = max(tp->snd_cwnd, tp->bictcp.last_max_cwnd);
+		if (tp->ca_ops->undo_cwnd)
+			tp->snd_cwnd = tp->ca_ops->undo_cwnd(tp);
 		else
 			tp->snd_cwnd = max(tp->snd_cwnd, tp->snd_ssthresh<<1);
 
@@ -1767,11 +1687,9 @@ static int tcp_try_undo_loss(struct sock *sk, struct tcp_sock *tp)
 
 static inline void tcp_complete_cwr(struct tcp_sock *tp)
 {
-	if (tcp_westwood_cwnd(tp)) 
-		tp->snd_ssthresh = tp->snd_cwnd;
-	else
-		tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_ssthresh);
+	tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_ssthresh);
 	tp->snd_cwnd_stamp = tcp_time_stamp;
+	tcp_ca_event(tp, CA_EVENT_COMPLETE_CWR);
 }
 
 static void tcp_try_to_open(struct sock *sk, struct tcp_sock *tp, int flag)
@@ -1946,7 +1864,7 @@ tcp_fastretrans_alert(struct sock *sk, u32 prior_snd_una,
 		if (tp->ca_state < TCP_CA_CWR) {
 			if (!(flag&FLAG_ECE))
 				tp->prior_ssthresh = tcp_current_ssthresh(tp);
-			tp->snd_ssthresh = tcp_recalc_ssthresh(tp);
+			tp->snd_ssthresh = tp->ca_ops->ssthresh(tp);
 			TCP_ECN_queue_cwr(tp);
 		}
 
@@ -1963,7 +1881,7 @@ tcp_fastretrans_alert(struct sock *sk, u32 prior_snd_una,
 /* Read draft-ietf-tcplw-high-performance before mucking
  * with this code. (Superceeds RFC1323)
  */
-static void tcp_ack_saw_tstamp(struct tcp_sock *tp, int flag)
+static void tcp_ack_saw_tstamp(struct tcp_sock *tp, u32 *usrtt, int flag)
 {
 	__u32 seq_rtt;
 
@@ -1983,13 +1901,13 @@ static void tcp_ack_saw_tstamp(struct tcp_sock *tp, int flag)
 	 * in window is lost... Voila.	 			--ANK (010210)
 	 */
 	seq_rtt = tcp_time_stamp - tp->rx_opt.rcv_tsecr;
-	tcp_rtt_estimator(tp, seq_rtt);
+	tcp_rtt_estimator(tp, seq_rtt, usrtt);
 	tcp_set_rto(tp);
 	tp->backoff = 0;
 	tcp_bound_rto(tp);
 }
 
-static void tcp_ack_no_tstamp(struct tcp_sock *tp, u32 seq_rtt, int flag)
+static void tcp_ack_no_tstamp(struct tcp_sock *tp, u32 seq_rtt, u32 *usrtt, int flag)
 {
 	/* We don't have a timestamp. Can only use
 	 * packets that are not retransmitted to determine
@@ -2003,338 +1921,29 @@ static void tcp_ack_no_tstamp(struct tcp_sock *tp, u32 seq_rtt, int flag)
 	if (flag & FLAG_RETRANS_DATA_ACKED)
 		return;
 
-	tcp_rtt_estimator(tp, seq_rtt);
+	tcp_rtt_estimator(tp, seq_rtt, usrtt);
 	tcp_set_rto(tp);
 	tp->backoff = 0;
 	tcp_bound_rto(tp);
 }
 
 static inline void tcp_ack_update_rtt(struct tcp_sock *tp,
-				      int flag, s32 seq_rtt)
+				      int flag, s32 seq_rtt, u32 *usrtt)
 {
 	/* Note that peer MAY send zero echo. In this case it is ignored. (rfc1323) */
 	if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr)
-		tcp_ack_saw_tstamp(tp, flag);
+		tcp_ack_saw_tstamp(tp, usrtt, flag);
 	else if (seq_rtt >= 0)
-		tcp_ack_no_tstamp(tp, seq_rtt, flag);
+		tcp_ack_no_tstamp(tp, seq_rtt, usrtt, flag);
 }
 
-/*
- * Compute congestion window to use.
- *
- * This is from the implementation of BICTCP in
- * Lison-Xu, Kahaled Harfoush, and Injog Rhee.
- *  "Binary Increase Congestion Control for Fast, Long Distance
- *  Networks" in InfoComm 2004
- * Available from:
- *  http://www.csc.ncsu.edu/faculty/rhee/export/bitcp.pdf
- *
- * Unless BIC is enabled and congestion window is large
- * this behaves the same as the original Reno.
- */
-static inline __u32 bictcp_cwnd(struct tcp_sock *tp)
-{
-	/* orignal Reno behaviour */
-	if (!tcp_is_bic(tp))
-		return tp->snd_cwnd;
-
-	if (tp->bictcp.last_cwnd == tp->snd_cwnd &&
-	   (s32)(tcp_time_stamp - tp->bictcp.last_stamp) <= (HZ>>5))
-		return tp->bictcp.cnt;
-
-	tp->bictcp.last_cwnd = tp->snd_cwnd;
-	tp->bictcp.last_stamp = tcp_time_stamp;
-      
-	/* start off normal */
-	if (tp->snd_cwnd <= sysctl_tcp_bic_low_window)
-		tp->bictcp.cnt = tp->snd_cwnd;
-
-	/* binary increase */
-	else if (tp->snd_cwnd < tp->bictcp.last_max_cwnd) {
-		__u32 	dist = (tp->bictcp.last_max_cwnd - tp->snd_cwnd)
-			/ BICTCP_B;
-
-		if (dist > BICTCP_MAX_INCREMENT)
-			/* linear increase */
-			tp->bictcp.cnt = tp->snd_cwnd / BICTCP_MAX_INCREMENT;
-		else if (dist <= 1U)
-			/* binary search increase */
-			tp->bictcp.cnt = tp->snd_cwnd * BICTCP_FUNC_OF_MIN_INCR
-				/ BICTCP_B;
-		else
-			/* binary search increase */
-			tp->bictcp.cnt = tp->snd_cwnd / dist;
-	} else {
-		/* slow start amd linear increase */
-		if (tp->snd_cwnd < tp->bictcp.last_max_cwnd + BICTCP_B)
-			/* slow start */
-			tp->bictcp.cnt = tp->snd_cwnd * BICTCP_FUNC_OF_MIN_INCR
-				/ BICTCP_B;
-		else if (tp->snd_cwnd < tp->bictcp.last_max_cwnd
-			 		+ BICTCP_MAX_INCREMENT*(BICTCP_B-1))
-			/* slow start */
-			tp->bictcp.cnt = tp->snd_cwnd * (BICTCP_B-1)
-				/ (tp->snd_cwnd-tp->bictcp.last_max_cwnd);
-		else
-			/* linear increase */
-			tp->bictcp.cnt = tp->snd_cwnd / BICTCP_MAX_INCREMENT;
-	}
-	return tp->bictcp.cnt;
-}
-
-/* This is Jacobson's slow start and congestion avoidance. 
- * SIGCOMM '88, p. 328.
- */
-static inline void reno_cong_avoid(struct tcp_sock *tp)
+static inline void tcp_cong_avoid(struct tcp_sock *tp, u32 ack, u32 rtt,
+				  u32 in_flight, int good)
 {
-        if (tp->snd_cwnd <= tp->snd_ssthresh) {
-                /* In "safe" area, increase. */
-		if (tp->snd_cwnd < tp->snd_cwnd_clamp)
-			tp->snd_cwnd++;
-	} else {
-                /* In dangerous area, increase slowly.
-		 * In theory this is tp->snd_cwnd += 1 / tp->snd_cwnd
-		 */
-		if (tp->snd_cwnd_cnt >= bictcp_cwnd(tp)) {
-			if (tp->snd_cwnd < tp->snd_cwnd_clamp)
-				tp->snd_cwnd++;
-			tp->snd_cwnd_cnt=0;
-		} else
-			tp->snd_cwnd_cnt++;
-        }
+	tp->ca_ops->cong_avoid(tp, ack, rtt, in_flight, good);
 	tp->snd_cwnd_stamp = tcp_time_stamp;
 }
 
-/* This is based on the congestion detection/avoidance scheme described in
- *    Lawrence S. Brakmo and Larry L. Peterson.
- *    "TCP Vegas: End to end congestion avoidance on a global internet."
- *    IEEE Journal on Selected Areas in Communication, 13(8):1465--1480,
- *    October 1995. Available from:
- *	ftp://ftp.cs.arizona.edu/xkernel/Papers/jsac.ps
- *
- * See http://www.cs.arizona.edu/xkernel/ for their implementation.
- * The main aspects that distinguish this implementation from the
- * Arizona Vegas implementation are:
- *   o We do not change the loss detection or recovery mechanisms of
- *     Linux in any way. Linux already recovers from losses quite well,
- *     using fine-grained timers, NewReno, and FACK.
- *   o To avoid the performance penalty imposed by increasing cwnd
- *     only every-other RTT during slow start, we increase during
- *     every RTT during slow start, just like Reno.
- *   o Largely to allow continuous cwnd growth during slow start,
- *     we use the rate at which ACKs come back as the "actual"
- *     rate, rather than the rate at which data is sent.
- *   o To speed convergence to the right rate, we set the cwnd
- *     to achieve the right ("actual") rate when we exit slow start.
- *   o To filter out the noise caused by delayed ACKs, we use the
- *     minimum RTT sample observed during the last RTT to calculate
- *     the actual rate.
- *   o When the sender re-starts from idle, it waits until it has
- *     received ACKs for an entire flight of new data before making
- *     a cwnd adjustment decision. The original Vegas implementation
- *     assumed senders never went idle.
- */
-static void vegas_cong_avoid(struct tcp_sock *tp, u32 ack, u32 seq_rtt)
-{
-	/* The key players are v_beg_snd_una and v_beg_snd_nxt.
-	 *
-	 * These are so named because they represent the approximate values
-	 * of snd_una and snd_nxt at the beginning of the current RTT. More
-	 * precisely, they represent the amount of data sent during the RTT.
-	 * At the end of the RTT, when we receive an ACK for v_beg_snd_nxt,
-	 * we will calculate that (v_beg_snd_nxt - v_beg_snd_una) outstanding
-	 * bytes of data have been ACKed during the course of the RTT, giving
-	 * an "actual" rate of:
-	 *
-	 *     (v_beg_snd_nxt - v_beg_snd_una) / (rtt duration)
-	 *
-	 * Unfortunately, v_beg_snd_una is not exactly equal to snd_una,
-	 * because delayed ACKs can cover more than one segment, so they
-	 * don't line up nicely with the boundaries of RTTs.
-	 *
-	 * Another unfortunate fact of life is that delayed ACKs delay the
-	 * advance of the left edge of our send window, so that the number
-	 * of bytes we send in an RTT is often less than our cwnd will allow.
-	 * So we keep track of our cwnd separately, in v_beg_snd_cwnd.
-	 */
-
-	if (after(ack, tp->vegas.beg_snd_nxt)) {
-		/* Do the Vegas once-per-RTT cwnd adjustment. */
-		u32 old_wnd, old_snd_cwnd;
-
-		
-		/* Here old_wnd is essentially the window of data that was
-		 * sent during the previous RTT, and has all
-		 * been acknowledged in the course of the RTT that ended
-		 * with the ACK we just received. Likewise, old_snd_cwnd
-		 * is the cwnd during the previous RTT.
-		 */
-		old_wnd = (tp->vegas.beg_snd_nxt - tp->vegas.beg_snd_una) /
-			tp->mss_cache_std;
-		old_snd_cwnd = tp->vegas.beg_snd_cwnd;
-
-		/* Save the extent of the current window so we can use this
-		 * at the end of the next RTT.
-		 */
-		tp->vegas.beg_snd_una  = tp->vegas.beg_snd_nxt;
-		tp->vegas.beg_snd_nxt  = tp->snd_nxt;
-		tp->vegas.beg_snd_cwnd = tp->snd_cwnd;
-
-		/* Take into account the current RTT sample too, to
-		 * decrease the impact of delayed acks. This double counts
-		 * this sample since we count it for the next window as well,
-		 * but that's not too awful, since we're taking the min,
-		 * rather than averaging.
-		 */
-		vegas_rtt_calc(tp, seq_rtt);
-
-		/* We do the Vegas calculations only if we got enough RTT
-		 * samples that we can be reasonably sure that we got
-		 * at least one RTT sample that wasn't from a delayed ACK.
-		 * If we only had 2 samples total,
-		 * then that means we're getting only 1 ACK per RTT, which
-		 * means they're almost certainly delayed ACKs.
-		 * If  we have 3 samples, we should be OK.
-		 */
-
-		if (tp->vegas.cntRTT <= 2) {
-			/* We don't have enough RTT samples to do the Vegas
-			 * calculation, so we'll behave like Reno.
-			 */
-			if (tp->snd_cwnd > tp->snd_ssthresh)
-				tp->snd_cwnd++;
-		} else {
-			u32 rtt, target_cwnd, diff;
-
-			/* We have enough RTT samples, so, using the Vegas
-			 * algorithm, we determine if we should increase or
-			 * decrease cwnd, and by how much.
-			 */
-
-			/* Pluck out the RTT we are using for the Vegas
-			 * calculations. This is the min RTT seen during the
-			 * last RTT. Taking the min filters out the effects
-			 * of delayed ACKs, at the cost of noticing congestion
-			 * a bit later.
-			 */
-			rtt = tp->vegas.minRTT;
-
-			/* Calculate the cwnd we should have, if we weren't
-			 * going too fast.
-			 *
-			 * This is:
-			 *     (actual rate in segments) * baseRTT
-			 * We keep it as a fixed point number with
-			 * V_PARAM_SHIFT bits to the right of the binary point.
-			 */
-			target_cwnd = ((old_wnd * tp->vegas.baseRTT)
-				       << V_PARAM_SHIFT) / rtt;
-
-			/* Calculate the difference between the window we had,
-			 * and the window we would like to have. This quantity
-			 * is the "Diff" from the Arizona Vegas papers.
-			 *
-			 * Again, this is a fixed point number with
-			 * V_PARAM_SHIFT bits to the right of the binary
-			 * point.
-			 */
-			diff = (old_wnd << V_PARAM_SHIFT) - target_cwnd;
-
-			if (tp->snd_cwnd < tp->snd_ssthresh) {
-				/* Slow start.  */
-				if (diff > sysctl_tcp_vegas_gamma) {
-					/* Going too fast. Time to slow down
-					 * and switch to congestion avoidance.
-					 */
-					tp->snd_ssthresh = 2;
-
-					/* Set cwnd to match the actual rate
-					 * exactly:
-					 *   cwnd = (actual rate) * baseRTT
-					 * Then we add 1 because the integer
-					 * truncation robs us of full link
-					 * utilization.
-					 */
-					tp->snd_cwnd = min(tp->snd_cwnd,
-							   (target_cwnd >>
-							    V_PARAM_SHIFT)+1);
-
-				}
-			} else {
-				/* Congestion avoidance. */
-				u32 next_snd_cwnd;
-
-				/* Figure out where we would like cwnd
-				 * to be.
-				 */
-				if (diff > sysctl_tcp_vegas_beta) {
-					/* The old window was too fast, so
-					 * we slow down.
-					 */
-					next_snd_cwnd = old_snd_cwnd - 1;
-				} else if (diff < sysctl_tcp_vegas_alpha) {
-					/* We don't have enough extra packets
-					 * in the network, so speed up.
-					 */
-					next_snd_cwnd = old_snd_cwnd + 1;
-				} else {
-					/* Sending just as fast as we
-					 * should be.
-					 */
-					next_snd_cwnd = old_snd_cwnd;
-				}
-
-				/* Adjust cwnd upward or downward, toward the
-				 * desired value.
-				 */
-				if (next_snd_cwnd > tp->snd_cwnd)
-					tp->snd_cwnd++;
-				else if (next_snd_cwnd < tp->snd_cwnd)
-					tp->snd_cwnd--;
-			}
-		}
-
-		/* Wipe the slate clean for the next RTT. */
-		tp->vegas.cntRTT = 0;
-		tp->vegas.minRTT = 0x7fffffff;
-	}
-
-	/* The following code is executed for every ack we receive,
-	 * except for conditions checked in should_advance_cwnd()
-	 * before the call to tcp_cong_avoid(). Mainly this means that
-	 * we only execute this code if the ack actually acked some
-	 * data.
-	 */
-
-	/* If we are in slow start, increase our cwnd in response to this ACK.
-	 * (If we are not in slow start then we are in congestion avoidance,
-	 * and adjust our congestion window only once per RTT. See the code
-	 * above.)
-	 */
-	if (tp->snd_cwnd <= tp->snd_ssthresh) 
-		tp->snd_cwnd++;
-
-	/* to keep cwnd from growing without bound */
-	tp->snd_cwnd = min_t(u32, tp->snd_cwnd, tp->snd_cwnd_clamp);
-
-	/* Make sure that we are never so timid as to reduce our cwnd below
-	 * 2 MSS.
-	 *
-	 * Going below 2 MSS would risk huge delayed ACKs from our receiver.
-	 */
-	tp->snd_cwnd = max(tp->snd_cwnd, 2U);
-
-	tp->snd_cwnd_stamp = tcp_time_stamp;
-}
-
-static inline void tcp_cong_avoid(struct tcp_sock *tp, u32 ack, u32 seq_rtt)
-{
-	if (tcp_vegas_enabled(tp))
-		vegas_cong_avoid(tp, ack, seq_rtt);
-	else
-		reno_cong_avoid(tp);
-}
-
 /* Restart timer after forward progress on connection.
  * RFC2988 recommends to restart timer to now+rto.
  */
@@ -2415,13 +2024,18 @@ static int tcp_tso_acked(struct sock *sk, struct sk_buff *skb,
 
 
 /* Remove acknowledged frames from the retransmission queue. */
-static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p)
+static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p, s32 *seq_usrtt)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct sk_buff *skb;
 	__u32 now = tcp_time_stamp;
 	int acked = 0;
 	__s32 seq_rtt = -1;
+	struct timeval usnow;
+	u32 pkts_acked = 0;
+
+	if (seq_usrtt)
+		do_gettimeofday(&usnow);
 
 	while ((skb = skb_peek(&sk->sk_write_queue)) &&
 	       skb != sk->sk_send_head) {
@@ -2448,6 +2062,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p)
 		 */
 		if (!(scb->flags & TCPCB_FLAG_SYN)) {
 			acked |= FLAG_DATA_ACKED;
+			++pkts_acked;
 		} else {
 			acked |= FLAG_SYN_ACKED;
 			tp->retrans_stamp = 0;
@@ -2461,6 +2076,10 @@ static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p)
 				seq_rtt = -1;
 			} else if (seq_rtt < 0)
 				seq_rtt = now - scb->when;
+			if (seq_usrtt)
+				*seq_usrtt = (usnow.tv_sec - skb->stamp.tv_sec) * 1000000
+					+ (usnow.tv_usec - skb->stamp.tv_usec);
+
 			if (sacked & TCPCB_SACKED_ACKED)
 				tp->sacked_out -= tcp_skb_pcount(skb);
 			if (sacked & TCPCB_LOST)
@@ -2479,8 +2098,11 @@ static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p)
 	}
 
 	if (acked&FLAG_ACKED) {
-		tcp_ack_update_rtt(tp, acked, seq_rtt);
+		tcp_ack_update_rtt(tp, acked, seq_rtt, seq_usrtt);
 		tcp_ack_packets_out(sk, tp);
+
+		if (tp->ca_ops->pkts_acked)
+			tp->ca_ops->pkts_acked(tp, pkts_acked);
 	}
 
 #if FASTRETRANS_DEBUG > 0
@@ -2624,257 +2246,6 @@ static void tcp_process_frto(struct sock *sk, u32 prior_snd_una)
 	tp->frto_counter = (tp->frto_counter + 1) % 3;
 }
 
-/*
- * TCP Westwood+
- */
-
-/*
- * @init_westwood
- * This function initializes fields used in TCP Westwood+. We can't
- * get no information about RTTmin at this time so we simply set it to
- * TCP_WESTWOOD_INIT_RTT. This value was chosen to be too conservative
- * since in this way we're sure it will be updated in a consistent
- * way as soon as possible. It will reasonably happen within the first
- * RTT period of the connection lifetime.
- */
-
-static void init_westwood(struct sock *sk)
-{
-        struct tcp_sock *tp = tcp_sk(sk);
-
-        tp->westwood.bw_ns_est = 0;
-        tp->westwood.bw_est = 0;
-        tp->westwood.accounted = 0;
-        tp->westwood.cumul_ack = 0;
-        tp->westwood.rtt_win_sx = tcp_time_stamp;
-        tp->westwood.rtt = TCP_WESTWOOD_INIT_RTT;
-        tp->westwood.rtt_min = TCP_WESTWOOD_INIT_RTT;
-        tp->westwood.snd_una = tp->snd_una;
-}
-
-/*
- * @westwood_do_filter
- * Low-pass filter. Implemented using constant coeffients.
- */
-
-static inline __u32 westwood_do_filter(__u32 a, __u32 b)
-{
-	return (((7 * a) + b) >> 3);
-}
-
-static void westwood_filter(struct sock *sk, __u32 delta)
-{
-	struct tcp_sock *tp = tcp_sk(sk);
-
-	tp->westwood.bw_ns_est =
-		westwood_do_filter(tp->westwood.bw_ns_est, 
-				   tp->westwood.bk / delta);
-	tp->westwood.bw_est =
-		westwood_do_filter(tp->westwood.bw_est,
-				   tp->westwood.bw_ns_est);
-}
-
-/* 
- * @westwood_update_rttmin
- * It is used to update RTTmin. In this case we MUST NOT use
- * WESTWOOD_RTT_MIN minimum bound since we could be on a LAN!
- */
-
-static inline __u32 westwood_update_rttmin(const struct sock *sk)
-{
-	const struct tcp_sock *tp = tcp_sk(sk);
-	__u32 rttmin = tp->westwood.rtt_min;
-
-	if (tp->westwood.rtt != 0 &&
-	    (tp->westwood.rtt < tp->westwood.rtt_min || !rttmin))
-		rttmin = tp->westwood.rtt;
-
-	return rttmin;
-}
-
-/*
- * @westwood_acked
- * Evaluate increases for dk. 
- */
-
-static inline __u32 westwood_acked(const struct sock *sk)
-{
-	const struct tcp_sock *tp = tcp_sk(sk);
-
-	return tp->snd_una - tp->westwood.snd_una;
-}
-
-/*
- * @westwood_new_window
- * It evaluates if we are receiving data inside the same RTT window as
- * when we started.
- * Return value:
- * It returns 0 if we are still evaluating samples in the same RTT
- * window, 1 if the sample has to be considered in the next window.
- */
-
-static int westwood_new_window(const struct sock *sk)
-{
-	const struct tcp_sock *tp = tcp_sk(sk);
-	__u32 left_bound;
-	__u32 rtt;
-	int ret = 0;
-
-	left_bound = tp->westwood.rtt_win_sx;
-	rtt = max(tp->westwood.rtt, (u32) TCP_WESTWOOD_RTT_MIN);
-
-	/*
-	 * A RTT-window has passed. Be careful since if RTT is less than
-	 * 50ms we don't filter but we continue 'building the sample'.
-	 * This minimum limit was choosen since an estimation on small
-	 * time intervals is better to avoid...
-	 * Obvioulsy on a LAN we reasonably will always have
-	 * right_bound = left_bound + WESTWOOD_RTT_MIN
-         */
-
-	if ((left_bound + rtt) < tcp_time_stamp)
-		ret = 1;
-
-	return ret;
-}
-
-/*
- * @westwood_update_window
- * It updates RTT evaluation window if it is the right moment to do
- * it. If so it calls filter for evaluating bandwidth. 
- */
-
-static void __westwood_update_window(struct sock *sk, __u32 now)
-{
-	struct tcp_sock *tp = tcp_sk(sk);
-	__u32 delta = now - tp->westwood.rtt_win_sx;
-
-        if (delta) {
-		if (tp->westwood.rtt)
-			westwood_filter(sk, delta);
-
-		tp->westwood.bk = 0;
-		tp->westwood.rtt_win_sx = tcp_time_stamp;
-	}
-}
-
-
-static void westwood_update_window(struct sock *sk, __u32 now)
-{
-	if (westwood_new_window(sk)) 
-		__westwood_update_window(sk, now);
-}
-
-/*
- * @__tcp_westwood_fast_bw
- * It is called when we are in fast path. In particular it is called when
- * header prediction is successfull. In such case infact update is
- * straight forward and doesn't need any particular care.
- */
-
-static void __tcp_westwood_fast_bw(struct sock *sk, struct sk_buff *skb)
-{
-	struct tcp_sock *tp = tcp_sk(sk);
-
-	westwood_update_window(sk, tcp_time_stamp);
-
-	tp->westwood.bk += westwood_acked(sk);
-	tp->westwood.snd_una = tp->snd_una;
-	tp->westwood.rtt_min = westwood_update_rttmin(sk);
-}
-
-static inline void tcp_westwood_fast_bw(struct sock *sk, struct sk_buff *skb)
-{
-        if (tcp_is_westwood(tcp_sk(sk)))
-                __tcp_westwood_fast_bw(sk, skb);
-}
-
-
-/*
- * @westwood_dupack_update
- * It updates accounted and cumul_ack when receiving a dupack.
- */
-
-static void westwood_dupack_update(struct sock *sk)
-{
-	struct tcp_sock *tp = tcp_sk(sk);
-
-	tp->westwood.accounted += tp->mss_cache_std;
-	tp->westwood.cumul_ack = tp->mss_cache_std;
-}
-
-static inline int westwood_may_change_cumul(struct tcp_sock *tp)
-{
-	return (tp->westwood.cumul_ack > tp->mss_cache_std);
-}
-
-static inline void westwood_partial_update(struct tcp_sock *tp)
-{
-	tp->westwood.accounted -= tp->westwood.cumul_ack;
-	tp->westwood.cumul_ack = tp->mss_cache_std;
-}
-
-static inline void westwood_complete_update(struct tcp_sock *tp)
-{
-	tp->westwood.cumul_ack -= tp->westwood.accounted;
-	tp->westwood.accounted = 0;
-}
-
-/*
- * @westwood_acked_count
- * This function evaluates cumul_ack for evaluating dk in case of
- * delayed or partial acks.
- */
-
-static inline __u32 westwood_acked_count(struct sock *sk)
-{
-	struct tcp_sock *tp = tcp_sk(sk);
-
-	tp->westwood.cumul_ack = westwood_acked(sk);
-
-        /* If cumul_ack is 0 this is a dupack since it's not moving
-         * tp->snd_una.
-         */
-        if (!(tp->westwood.cumul_ack))
-                westwood_dupack_update(sk);
-
-        if (westwood_may_change_cumul(tp)) {
-		/* Partial or delayed ack */
-		if (tp->westwood.accounted >= tp->westwood.cumul_ack)
-			westwood_partial_update(tp);
-		else
-			westwood_complete_update(tp);
-	}
-
-	tp->westwood.snd_una = tp->snd_una;
-
-	return tp->westwood.cumul_ack;
-}
-
-
-/*
- * @__tcp_westwood_slow_bw
- * It is called when something is going wrong..even if there could
- * be no problems! Infact a simple delayed packet may trigger a
- * dupack. But we need to be careful in such case.
- */
-
-static void __tcp_westwood_slow_bw(struct sock *sk, struct sk_buff *skb)
-{
-	struct tcp_sock *tp = tcp_sk(sk);
-
-	westwood_update_window(sk, tcp_time_stamp);
-
-	tp->westwood.bk += westwood_acked_count(sk);
-	tp->westwood.rtt_min = westwood_update_rttmin(sk);
-}
-
-static inline void tcp_westwood_slow_bw(struct sock *sk, struct sk_buff *skb)
-{
-        if (tcp_is_westwood(tcp_sk(sk)))
-                __tcp_westwood_slow_bw(sk, skb);
-}
-
 /* This routine deals with incoming acks, but not outgoing ones. */
 static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag)
 {
@@ -2884,6 +2255,7 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag)
 	u32 ack = TCP_SKB_CB(skb)->ack_seq;
 	u32 prior_in_flight;
 	s32 seq_rtt;
+	s32 seq_usrtt = 0;
 	int prior_packets;
 
 	/* If the ack is newer than sent or older than previous acks
@@ -2902,9 +2274,10 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag)
 		 */
 		tcp_update_wl(tp, ack, ack_seq);
 		tp->snd_una = ack;
-		tcp_westwood_fast_bw(sk, skb);
 		flag |= FLAG_WIN_UPDATE;
 
+		tcp_ca_event(tp, CA_EVENT_FAST_ACK);
+
 		NET_INC_STATS_BH(LINUX_MIB_TCPHPACKS);
 	} else {
 		if (ack_seq != TCP_SKB_CB(skb)->end_seq)
@@ -2920,7 +2293,7 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag)
 		if (TCP_ECN_rcv_ecn_echo(tp, skb->h.th))
 			flag |= FLAG_ECE;
 
-		tcp_westwood_slow_bw(sk,skb);
+		tcp_ca_event(tp, CA_EVENT_SLOW_ACK);
 	}
 
 	/* We passed data and got it acked, remove any soft error
@@ -2935,22 +2308,20 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag)
 	prior_in_flight = tcp_packets_in_flight(tp);
 
 	/* See if we can take anything off of the retransmit queue. */
-	flag |= tcp_clean_rtx_queue(sk, &seq_rtt);
+	flag |= tcp_clean_rtx_queue(sk, &seq_rtt,
+				    tp->ca_ops->rtt_sample ? &seq_usrtt : NULL);
 
 	if (tp->frto_counter)
 		tcp_process_frto(sk, prior_snd_una);
 
 	if (tcp_ack_is_dubious(tp, flag)) {
 		/* Advanve CWND, if state allows this. */
-		if ((flag & FLAG_DATA_ACKED) &&
-		    (tcp_vegas_enabled(tp) || prior_in_flight >= tp->snd_cwnd) &&
-		    tcp_may_raise_cwnd(tp, flag))
-			tcp_cong_avoid(tp, ack, seq_rtt);
+		if ((flag & FLAG_DATA_ACKED) && tcp_may_raise_cwnd(tp, flag))
+			tcp_cong_avoid(tp, ack,  seq_rtt, prior_in_flight, 0);
 		tcp_fastretrans_alert(sk, prior_snd_una, prior_packets, flag);
 	} else {
-		if ((flag & FLAG_DATA_ACKED) && 
-		    (tcp_vegas_enabled(tp) || prior_in_flight >= tp->snd_cwnd))
-			tcp_cong_avoid(tp, ack, seq_rtt);
+		if ((flag & FLAG_DATA_ACKED))
+			tcp_cong_avoid(tp, ack, seq_rtt, prior_in_flight, 1);
 	}
 
 	if ((flag & FLAG_FORWARD_PROGRESS) || !(flag&FLAG_NOT_DUP))
@@ -4552,6 +3923,8 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
 
 		tcp_init_metrics(sk);
 
+		tcp_init_congestion_control(tp);
+
 		/* Prevent spurious tcp_cwnd_restart() on first data
 		 * packet.
 		 */
@@ -4708,9 +4081,6 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
 			if(tp->af_specific->conn_request(sk, skb) < 0)
 				return 1;
 
-			init_westwood(sk);
-			init_bictcp(tp);
-
 			/* Now we have several options: In theory there is 
 			 * nothing else in the frame. KA9Q has an option to 
 			 * send data with the syn, BSD accepts data with the
@@ -4732,9 +4102,6 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
 		goto discard;
 
 	case TCP_SYN_SENT:
-		init_westwood(sk);
-		init_bictcp(tp);
-
 		queued = tcp_rcv_synsent_state_process(sk, skb, th, len);
 		if (queued >= 0)
 			return queued;
@@ -4816,7 +4183,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
 				 */
 				if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr &&
 				    !tp->srtt)
-					tcp_ack_saw_tstamp(tp, 0);
+					tcp_ack_saw_tstamp(tp, 0, 0);
 
 				if (tp->rx_opt.tstamp_ok)
 					tp->advmss -= TCPOLEN_TSTAMP_ALIGNED;
@@ -4828,6 +4195,8 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
 
 				tcp_init_metrics(sk);
 
+				tcp_init_congestion_control(tp);
+
 				/* Prevent spurious tcp_cwnd_restart() on
 				 * first data packet.
 				 */
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 2d41d5d6ad1..9122814c13a 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2048,6 +2048,7 @@ static int tcp_v4_init_sock(struct sock *sk)
 	tp->mss_cache_std = tp->mss_cache = 536;
 
 	tp->reordering = sysctl_tcp_reordering;
+	tp->ca_ops = &tcp_reno;
 
 	sk->sk_state = TCP_CLOSE;
 
@@ -2070,6 +2071,8 @@ int tcp_v4_destroy_sock(struct sock *sk)
 
 	tcp_clear_xmit_timers(sk);
 
+	tcp_cleanup_congestion_control(tp);
+
 	/* Cleanup up the write buffer. */
   	sk_stream_writequeue_purge(sk);
 
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index b3943e7562f..f42a284164b 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -774,6 +774,8 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req,
 		newtp->frto_counter = 0;
 		newtp->frto_highmark = 0;
 
+		newtp->ca_ops = &tcp_reno;
+
 		tcp_set_ca_state(newtp, TCP_CA_Open);
 		tcp_init_xmit_timers(newsk);
 		skb_queue_head_init(&newtp->out_of_order_queue);
@@ -842,8 +844,6 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req,
 		if (newtp->ecn_flags&TCP_ECN_OK)
 			sock_set_flag(newsk, SOCK_NO_LARGESEND);
 
-		tcp_ca_init(newtp);
-
 		TCP_INC_STATS_BH(TCP_MIB_PASSIVEOPENS);
 	}
 	return newsk;
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index f17c6577e33..0e17c244875 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -111,8 +111,7 @@ static void tcp_cwnd_restart(struct tcp_sock *tp, struct dst_entry *dst)
 	u32 restart_cwnd = tcp_init_cwnd(tp, dst);
 	u32 cwnd = tp->snd_cwnd;
 
-	if (tcp_is_vegas(tp)) 
-		tcp_vegas_enable(tp);
+	tcp_ca_event(tp, CA_EVENT_CWND_RESTART);
 
 	tp->snd_ssthresh = tcp_current_ssthresh(tp);
 	restart_cwnd = min(restart_cwnd, cwnd);
@@ -280,6 +279,10 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb)
 #define SYSCTL_FLAG_WSCALE	0x2
 #define SYSCTL_FLAG_SACK	0x4
 
+		/* If congestion control is doing timestamping */
+		if (tp->ca_ops->rtt_sample)
+			do_gettimeofday(&skb->stamp);
+
 		sysctl_flags = 0;
 		if (tcb->flags & TCPCB_FLAG_SYN) {
 			tcp_header_size = sizeof(struct tcphdr) + TCPOLEN_MSS;
@@ -304,17 +307,8 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb)
 					    (tp->rx_opt.eff_sacks * TCPOLEN_SACK_PERBLOCK));
 		}
 		
-		/*
-		 * If the connection is idle and we are restarting,
-		 * then we don't want to do any Vegas calculations
-		 * until we get fresh RTT samples.  So when we
-		 * restart, we reset our Vegas state to a clean
-		 * slate. After we get acks for this flight of
-		 * packets, _then_ we can make Vegas calculations
-		 * again.
-		 */
-		if (tcp_is_vegas(tp) && tcp_packets_in_flight(tp) == 0)
-			tcp_vegas_enable(tp);
+		if (tcp_packets_in_flight(tp) == 0)
+			tcp_ca_event(tp, CA_EVENT_TX_START);
 
 		th = (struct tcphdr *) skb_push(skb, tcp_header_size);
 		skb->h.th = th;
@@ -521,6 +515,7 @@ static int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len)
 	 * skbs, which it never sent before. --ANK
 	 */
 	TCP_SKB_CB(buff)->when = TCP_SKB_CB(skb)->when;
+	buff->stamp = skb->stamp;
 
 	if (TCP_SKB_CB(skb)->sacked & TCPCB_LOST) {
 		tp->lost_out -= tcp_skb_pcount(skb);
@@ -1449,7 +1444,6 @@ static inline void tcp_connect_init(struct sock *sk)
 		tp->window_clamp = dst_metric(dst, RTAX_WINDOW);
 	tp->advmss = dst_metric(dst, RTAX_ADVMSS);
 	tcp_initialize_rcv_mss(sk);
-	tcp_ca_init(tp);
 
 	tcp_select_initial_window(tcp_full_space(sk),
 				  tp->advmss - (tp->rx_opt.ts_recent_stamp ? tp->tcp_header_len - sizeof(struct tcphdr) : 0),
@@ -1503,7 +1497,6 @@ int tcp_connect(struct sock *sk)
 	TCP_SKB_CB(buff)->end_seq = tp->write_seq;
 	tp->snd_nxt = tp->write_seq;
 	tp->pushed_seq = tp->write_seq;
-	tcp_ca_init(tp);
 
 	/* Send it off. */
 	TCP_SKB_CB(buff)->when = tcp_time_stamp;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 2414937f2a8..fce56039b0e 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2025,7 +2025,7 @@ static int tcp_v6_init_sock(struct sock *sk)
 	sk->sk_state = TCP_CLOSE;
 
 	tp->af_specific = &ipv6_specific;
-
+	tp->ca_ops = &tcp_reno;
 	sk->sk_write_space = sk_stream_write_space;
 	sock_set_flag(sk, SOCK_USE_WRITE_QUEUE);
 
-- 
cgit v1.2.3-70-g09d2


From 056ede6cface66b400cd3b8e60ed077cc5b85c18 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Thu, 23 Jun 2005 12:21:28 -0700
Subject: [TCP]: Report congestion control algorithm in tcp_diag.

Enhancement to the tcp_diag interface used by the iproute2 ss command
to report the tcp congestion control being used by a socket.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp_diag.h | 4 ++--
 net/ipv4/tcp_diag.c      | 5 +++++
 2 files changed, 7 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/tcp_diag.h b/include/linux/tcp_diag.h
index ceee962e1d1..7a599674394 100644
--- a/include/linux/tcp_diag.h
+++ b/include/linux/tcp_diag.h
@@ -99,9 +99,10 @@ enum
 	TCPDIAG_MEMINFO,
 	TCPDIAG_INFO,
 	TCPDIAG_VEGASINFO,
+	TCPDIAG_CONG,
 };
 
-#define TCPDIAG_MAX TCPDIAG_VEGASINFO
+#define TCPDIAG_MAX TCPDIAG_CONG
 
 
 /* TCPDIAG_MEM */
@@ -123,5 +124,4 @@ struct tcpvegas_info {
 	__u32	tcpv_minrtt;
 };
 
-
 #endif /* _TCP_DIAG_H_ */
diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c
index a4e512036d8..f66945cb158 100644
--- a/net/ipv4/tcp_diag.c
+++ b/net/ipv4/tcp_diag.c
@@ -65,6 +65,11 @@ static int tcpdiag_fill(struct sk_buff *skb, struct sock *sk,
 		if (ext & (1<<(TCPDIAG_INFO-1)))
 			info = TCPDIAG_PUT(skb, TCPDIAG_INFO, sizeof(*info));
 		
+		if (ext & (1<<(TCPDIAG_CONG-1))) {
+			size_t len = strlen(tp->ca_ops->name);
+			strcpy(TCPDIAG_PUT(skb, TCPDIAG_CONG, len+1),
+			       tp->ca_ops->name);
+		}
 	}
 	r->tcpdiag_family = sk->sk_family;
 	r->tcpdiag_state = sk->sk_state;
-- 
cgit v1.2.3-70-g09d2


From e608a8072b10258aa18c2e33324def225199ba1d Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 22 Jun 2005 22:24:00 -0700
Subject: [IA64] Fix pfn_to_nid() so the kernel compiles again for
 !CONFIG_DISCONTIGMEM.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 include/asm-ia64/mmzone.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-ia64/mmzone.h b/include/asm-ia64/mmzone.h
index 83ca4043fc1..d32f51e3d6c 100644
--- a/include/asm-ia64/mmzone.h
+++ b/include/asm-ia64/mmzone.h
@@ -15,6 +15,8 @@
 #include <asm/page.h>
 #include <asm/meminit.h>
 
+#ifdef CONFIG_DISCONTIGMEM
+
 static inline int pfn_to_nid(unsigned long pfn)
 {
 #ifdef CONFIG_NUMA
@@ -29,8 +31,6 @@ static inline int pfn_to_nid(unsigned long pfn)
 #endif
 }
 
-#ifdef CONFIG_DISCONTIGMEM
-
 #ifdef CONFIG_IA64_DIG /* DIG systems are small */
 # define MAX_PHYSNODE_ID	8
 # define NR_NODE_MEMBLKS	(MAX_NUMNODES * 8)
-- 
cgit v1.2.3-70-g09d2


From c1ebcdb8c422cd73f54bcd2b9953e443a47667e5 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Thu, 23 Jun 2005 20:08:59 -0700
Subject: [NET]: Remove obsolete fastroute stats.

Remove last vestiages of fastroute code that is no longer used.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h  |  5 -----
 net/core/dev.c             | 10 ++--------
 net/core/sysctl_net_core.c |  1 -
 3 files changed, 2 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index d89816ad642..c2e15e381a5 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -165,11 +165,6 @@ struct netif_rx_stats
 	unsigned dropped;
 	unsigned time_squeeze;
 	unsigned throttled;
-	unsigned fastroute_hit;
-	unsigned fastroute_success;
-	unsigned fastroute_defer;
-	unsigned fastroute_deferred_out;
-	unsigned fastroute_latency_reduction;
 	unsigned cpu_collision;
 };
 
diff --git a/net/core/dev.c b/net/core/dev.c
index ab935778ce8..4f1ae2efe87 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2056,14 +2056,8 @@ static int softnet_seq_show(struct seq_file *seq, void *v)
 
 	seq_printf(seq, "%08x %08x %08x %08x %08x %08x %08x %08x %08x\n",
 		   s->total, s->dropped, s->time_squeeze, s->throttled,
-		   s->fastroute_hit, s->fastroute_success, s->fastroute_defer,
-		   s->fastroute_deferred_out,
-#if 0
-		   s->fastroute_latency_reduction
-#else
-		   s->cpu_collision
-#endif
-		  );
+		   0, 0, 0, 0, /* was fastroute */
+		   s->cpu_collision );
 	return 0;
 }
 
diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c
index 880a8881521..76e9987474c 100644
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -18,7 +18,6 @@ extern int no_cong_thresh;
 extern int no_cong;
 extern int lo_cong;
 extern int mod_cong;
-extern int netdev_fastroute;
 extern int net_msg_cost;
 extern int net_msg_burst;
 
-- 
cgit v1.2.3-70-g09d2


From 34008d8c631d067caffa136313260525f3ae48a2 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Thu, 23 Jun 2005 20:10:00 -0700
Subject: [NET]: Remove obsolete netif_rx congestion sensing mechanism.

Remove the congestion sensing mechanism from netif_rx, and always
return either full or empty.  Almost no driver checks the return value
from netif_rx, and those that do only use it for debug messages.

The original design of netif_rx was to do flow control based on the
receive queue, but NAPI has supplanted this and no driver uses the
feedback.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h  |  2 --
 net/core/dev.c             | 88 +---------------------------------------------
 net/core/sysctl_net_core.c | 36 -------------------
 3 files changed, 1 insertion(+), 125 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c2e15e381a5..718ad579c65 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -558,8 +558,6 @@ static inline int unregister_gifconf(unsigned int family)
 struct softnet_data
 {
 	int			throttle;
-	int			cng_level;
-	int			avg_blog;
 	struct sk_buff_head	input_pkt_queue;
 	struct list_head	poll_list;
 	struct net_device	*output_queue;
diff --git a/net/core/dev.c b/net/core/dev.c
index 4f1ae2efe87..3156df699f0 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -115,18 +115,6 @@
 #endif	/* CONFIG_NET_RADIO */
 #include <asm/current.h>
 
-/* This define, if set, will randomly drop a packet when congestion
- * is more than moderate.  It helps fairness in the multi-interface
- * case when one of them is a hog, but it kills performance for the
- * single interface case so it is off now by default.
- */
-#undef RAND_LIE
-
-/* Setting this will sample the queue lengths and thus congestion
- * via a timer instead of as each packet is received.
- */
-#undef OFFLINE_SAMPLE
-
 /*
  *	The list of packet types we will receive (as opposed to discard)
  *	and the routines to invoke.
@@ -159,11 +147,6 @@ static DEFINE_SPINLOCK(ptype_lock);
 static struct list_head ptype_base[16];	/* 16 way hashed list */
 static struct list_head ptype_all;		/* Taps */
 
-#ifdef OFFLINE_SAMPLE
-static void sample_queue(unsigned long dummy);
-static struct timer_list samp_timer = TIMER_INITIALIZER(sample_queue, 0, 0);
-#endif
-
 /*
  * The @dev_base list is protected by @dev_base_lock and the rtln
  * semaphore.
@@ -1365,69 +1348,10 @@ out:
 
 int netdev_max_backlog = 300;
 int weight_p = 64;            /* old backlog weight */
-/* These numbers are selected based on intuition and some
- * experimentatiom, if you have more scientific way of doing this
- * please go ahead and fix things.
- */
-int no_cong_thresh = 10;
-int no_cong = 20;
-int lo_cong = 100;
-int mod_cong = 290;
 
 DEFINE_PER_CPU(struct netif_rx_stats, netdev_rx_stat) = { 0, };
 
 
-static void get_sample_stats(int cpu)
-{
-#ifdef RAND_LIE
-	unsigned long rd;
-	int rq;
-#endif
-	struct softnet_data *sd = &per_cpu(softnet_data, cpu);
-	int blog = sd->input_pkt_queue.qlen;
-	int avg_blog = sd->avg_blog;
-
-	avg_blog = (avg_blog >> 1) + (blog >> 1);
-
-	if (avg_blog > mod_cong) {
-		/* Above moderate congestion levels. */
-		sd->cng_level = NET_RX_CN_HIGH;
-#ifdef RAND_LIE
-		rd = net_random();
-		rq = rd % netdev_max_backlog;
-		if (rq < avg_blog) /* unlucky bastard */
-			sd->cng_level = NET_RX_DROP;
-#endif
-	} else if (avg_blog > lo_cong) {
-		sd->cng_level = NET_RX_CN_MOD;
-#ifdef RAND_LIE
-		rd = net_random();
-		rq = rd % netdev_max_backlog;
-			if (rq < avg_blog) /* unlucky bastard */
-				sd->cng_level = NET_RX_CN_HIGH;
-#endif
-	} else if (avg_blog > no_cong)
-		sd->cng_level = NET_RX_CN_LOW;
-	else  /* no congestion */
-		sd->cng_level = NET_RX_SUCCESS;
-
-	sd->avg_blog = avg_blog;
-}
-
-#ifdef OFFLINE_SAMPLE
-static void sample_queue(unsigned long dummy)
-{
-/* 10 ms 0r 1ms -- i don't care -- JHS */
-	int next_tick = 1;
-	int cpu = smp_processor_id();
-
-	get_sample_stats(cpu);
-	next_tick += jiffies;
-	mod_timer(&samp_timer, next_tick);
-}
-#endif
-
-
 /**
  *	netif_rx	-	post buffer to the network code
  *	@skb: buffer to post
@@ -1476,11 +1400,8 @@ int netif_rx(struct sk_buff *skb)
 enqueue:
 			dev_hold(skb->dev);
 			__skb_queue_tail(&queue->input_pkt_queue, skb);
-#ifndef OFFLINE_SAMPLE
-			get_sample_stats(this_cpu);
-#endif
 			local_irq_restore(flags);
-			return queue->cng_level;
+			return NET_RX_SUCCESS;
 		}
 
 		if (queue->throttle)
@@ -3300,8 +3221,6 @@ static int __init net_dev_init(void)
 		queue = &per_cpu(softnet_data, i);
 		skb_queue_head_init(&queue->input_pkt_queue);
 		queue->throttle = 0;
-		queue->cng_level = 0;
-		queue->avg_blog = 10; /* arbitrary non-zero */
 		queue->completion_queue = NULL;
 		INIT_LIST_HEAD(&queue->poll_list);
 		set_bit(__LINK_STATE_START, &queue->backlog_dev.state);
@@ -3310,11 +3229,6 @@ static int __init net_dev_init(void)
 		atomic_set(&queue->backlog_dev.refcnt, 1);
 	}
 
-#ifdef OFFLINE_SAMPLE
-	samp_timer.expires = jiffies + (10 * HZ);
-	add_timer(&samp_timer);
-#endif
-
 	dev_boot_phase = 0;
 
 	open_softirq(NET_TX_SOFTIRQ, net_tx_action, NULL);
diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c
index 76e9987474c..fff63643a35 100644
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -14,10 +14,6 @@
 
 extern int netdev_max_backlog;
 extern int weight_p;
-extern int no_cong_thresh;
-extern int no_cong;
-extern int lo_cong;
-extern int mod_cong;
 extern int net_msg_cost;
 extern int net_msg_burst;
 
@@ -84,38 +80,6 @@ ctl_table core_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
 	},
-	{
-		.ctl_name	= NET_CORE_NO_CONG_THRESH,
-		.procname	= "no_cong_thresh",
-		.data		= &no_cong_thresh,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
-	},
-	{
-		.ctl_name	= NET_CORE_NO_CONG,
-		.procname	= "no_cong",
-		.data		= &no_cong,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
-	},
-	{
-		.ctl_name	= NET_CORE_LO_CONG,
-		.procname	= "lo_cong",
-		.data		= &lo_cong,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
-	},
-	{
-		.ctl_name	= NET_CORE_MOD_CONG,
-		.procname	= "mod_cong",
-		.data		= &mod_cong,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
-	},
 	{
 		.ctl_name	= NET_CORE_MSG_COST,
 		.procname	= "message_cost",
-- 
cgit v1.2.3-70-g09d2


From 31aa02c53c84658f6694f319f09e232ede27be5a Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Thu, 23 Jun 2005 20:12:48 -0700
Subject: [NET]: Eliminate netif_rx massive packet drops.

Eliminate the throttling behaviour when the netif receive queue fills
because it behaves badly when using high speed networks under load.
The throttling cause multiple packet drops that cause TCP to go into
slow start mode. The same effective patch has been part of BIC TCP and
H-TCP as well as part of Web100.

The existing code drops 100's of packets when the queue fills;
this changes it to individual packet drop-tail.

Signed-off-by: Stephen Hemmminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  4 +---
 net/core/dev.c            | 21 ++-------------------
 2 files changed, 3 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 718ad579c65..3a0ed7f9e80 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -164,7 +164,6 @@ struct netif_rx_stats
 	unsigned total;
 	unsigned dropped;
 	unsigned time_squeeze;
-	unsigned throttled;
 	unsigned cpu_collision;
 };
 
@@ -557,10 +556,9 @@ static inline int unregister_gifconf(unsigned int family)
 
 struct softnet_data
 {
-	int			throttle;
+	struct net_device	*output_queue;
 	struct sk_buff_head	input_pkt_queue;
 	struct list_head	poll_list;
-	struct net_device	*output_queue;
 	struct sk_buff		*completion_queue;
 
 	struct net_device	backlog_dev;	/* Sorry. 8) */
diff --git a/net/core/dev.c b/net/core/dev.c
index 3156df699f0..1a64508e527 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -198,7 +198,7 @@ static struct notifier_block *netdev_chain;
  *	Device drivers call our routines to queue packets here. We empty the
  *	queue in the local softnet handler.
  */
-DEFINE_PER_CPU(struct softnet_data, softnet_data) = { 0, };
+DEFINE_PER_CPU(struct softnet_data, softnet_data) = { NULL };
 
 #ifdef CONFIG_SYSFS
 extern int netdev_sysfs_init(void);
@@ -1372,7 +1372,6 @@ DEFINE_PER_CPU(struct netif_rx_stats, netdev_rx_stat) = { 0, };
 
 int netif_rx(struct sk_buff *skb)
 {
-	int this_cpu;
 	struct softnet_data *queue;
 	unsigned long flags;
 
@@ -1388,15 +1387,11 @@ int netif_rx(struct sk_buff *skb)
 	 * short when CPU is congested, but is still operating.
 	 */
 	local_irq_save(flags);
-	this_cpu = smp_processor_id();
 	queue = &__get_cpu_var(softnet_data);
 
 	__get_cpu_var(netdev_rx_stat).total++;
 	if (queue->input_pkt_queue.qlen <= netdev_max_backlog) {
 		if (queue->input_pkt_queue.qlen) {
-			if (queue->throttle)
-				goto drop;
-
 enqueue:
 			dev_hold(skb->dev);
 			__skb_queue_tail(&queue->input_pkt_queue, skb);
@@ -1404,19 +1399,10 @@ enqueue:
 			return NET_RX_SUCCESS;
 		}
 
-		if (queue->throttle)
-			queue->throttle = 0;
-
 		netif_rx_schedule(&queue->backlog_dev);
 		goto enqueue;
 	}
 
-	if (!queue->throttle) {
-		queue->throttle = 1;
-		__get_cpu_var(netdev_rx_stat).throttled++;
-	}
-
-drop:
 	__get_cpu_var(netdev_rx_stat).dropped++;
 	local_irq_restore(flags);
 
@@ -1701,8 +1687,6 @@ job_done:
 	smp_mb__before_clear_bit();
 	netif_poll_enable(backlog_dev);
 
-	if (queue->throttle)
-		queue->throttle = 0;
 	local_irq_enable();
 	return 0;
 }
@@ -1976,7 +1960,7 @@ static int softnet_seq_show(struct seq_file *seq, void *v)
 	struct netif_rx_stats *s = v;
 
 	seq_printf(seq, "%08x %08x %08x %08x %08x %08x %08x %08x %08x\n",
-		   s->total, s->dropped, s->time_squeeze, s->throttled,
+		   s->total, s->dropped, s->time_squeeze, 0,
 		   0, 0, 0, 0, /* was fastroute */
 		   s->cpu_collision );
 	return 0;
@@ -3220,7 +3204,6 @@ static int __init net_dev_init(void)
 
 		queue = &per_cpu(softnet_data, i);
 		skb_queue_head_init(&queue->input_pkt_queue);
-		queue->throttle = 0;
 		queue->completion_queue = NULL;
 		INIT_LIST_HEAD(&queue->poll_list);
 		set_bit(__LINK_STATE_START, &queue->backlog_dev.state);
-- 
cgit v1.2.3-70-g09d2


From 51b0bdedb8e784d0d969a6b77151911130812400 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Thu, 23 Jun 2005 20:14:40 -0700
Subject: [NET]: Separate two usages of netdev_max_backlog.

Separate out the two uses of netdev_max_backlog. One controls the
upper bound on packets processed per softirq, the new name for this is
netdev_budget; the other controls the limit on packets queued via
netif_rx.

Increase the max_backlog default to account for faster processors.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/sysctl.h     | 1 +
 net/core/dev.c             | 6 +++---
 net/core/sysctl_net_core.c | 9 +++++++++
 3 files changed, 13 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 72965bfe6cf..ebfe1250f0a 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -243,6 +243,7 @@ enum
 	NET_CORE_MOD_CONG=16,
 	NET_CORE_DEV_WEIGHT=17,
 	NET_CORE_SOMAXCONN=18,
+	NET_CORE_BUDGET=19,
 };
 
 /* /proc/sys/net/ethernet */
diff --git a/net/core/dev.c b/net/core/dev.c
index 1a64508e527..7016e0c36b3 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1346,7 +1346,8 @@ out:
 			Receiver routines
   =======================================================================*/
 
-int netdev_max_backlog = 300;
+int netdev_max_backlog = 1000;
+int netdev_budget = 300;
 int weight_p = 64;            /* old backlog weight */
 
 DEFINE_PER_CPU(struct netif_rx_stats, netdev_rx_stat) = { 0, };
@@ -1695,8 +1696,7 @@ static void net_rx_action(struct softirq_action *h)
 {
 	struct softnet_data *queue = &__get_cpu_var(softnet_data);
 	unsigned long start_time = jiffies;
-	int budget = netdev_max_backlog;
-
+	int budget = netdev_budget;
 	
 	local_irq_disable();
 
diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c
index fff63643a35..8f817ad9f54 100644
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -13,6 +13,7 @@
 #ifdef CONFIG_SYSCTL
 
 extern int netdev_max_backlog;
+extern int netdev_budget;
 extern int weight_p;
 extern int net_msg_cost;
 extern int net_msg_burst;
@@ -124,6 +125,14 @@ ctl_table core_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
 	},
+	{
+		.ctl_name	= NET_CORE_BUDGET,
+		.procname	= "netdev_budget",
+		.data		= &netdev_budget,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec
+	},
 	{ .ctl_name = 0 }
 };
 
-- 
cgit v1.2.3-70-g09d2


From 5f8ef48d240963093451bcf83df89f1a1364f51d Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Thu, 23 Jun 2005 20:37:36 -0700
Subject: [TCP]: Allow choosing TCP congestion control via sockopt.

Allow using setsockopt to set TCP congestion control to use on a per
socket basis.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h |  1 +
 include/net/tcp.h   |  3 ++-
 net/ipv4/tcp.c      | 31 ++++++++++++++++++++++++++++++-
 net/ipv4/tcp_cong.c | 46 ++++++++++++++++++++++++++++++++++++++++++++--
 net/ipv4/tcp_ipv4.c |  2 +-
 net/ipv6/tcp_ipv6.c |  2 +-
 6 files changed, 79 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 3ea75dd6640..dfd93d03f5d 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -127,6 +127,7 @@ enum {
 #define TCP_WINDOW_CLAMP	10	/* Bound advertised window */
 #define TCP_INFO		11	/* Information about this connection. */
 #define TCP_QUICKACK		12	/* Block/reenable quick acks */
+#define TCP_CONGESTION		13	/* Congestion control algorithm */
 
 #define TCPI_OPT_TIMESTAMPS	1
 #define TCPI_OPT_SACK		2
diff --git a/include/net/tcp.h b/include/net/tcp.h
index e427cf35915..d04b21188cc 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1162,8 +1162,9 @@ extern void tcp_init_congestion_control(struct tcp_sock *tp);
 extern void tcp_cleanup_congestion_control(struct tcp_sock *tp);
 extern int tcp_set_default_congestion_control(const char *name);
 extern void tcp_get_default_congestion_control(char *name);
+extern int tcp_set_congestion_control(struct tcp_sock *tp, const char *name);
 
-extern struct tcp_congestion_ops tcp_reno;
+extern struct tcp_congestion_ops tcp_init_congestion_ops;
 extern u32 tcp_reno_ssthresh(struct tcp_sock *tp);
 extern void tcp_reno_cong_avoid(struct tcp_sock *tp, u32 ack,
 				u32 rtt, u32 in_flight, int flag);
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index f3dbc8dc126..882436da9a3 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1927,6 +1927,25 @@ int tcp_setsockopt(struct sock *sk, int level, int optname, char __user *optval,
 		return tp->af_specific->setsockopt(sk, level, optname,
 						   optval, optlen);
 
+	/* This is a string value all the others are int's */
+	if (optname == TCP_CONGESTION) {
+		char name[TCP_CA_NAME_MAX];
+
+		if (optlen < 1)
+			return -EINVAL;
+
+		val = strncpy_from_user(name, optval,
+					min(TCP_CA_NAME_MAX-1, optlen));
+		if (val < 0)
+			return -EFAULT;
+		name[val] = 0;
+
+		lock_sock(sk);
+		err = tcp_set_congestion_control(tp, name);
+		release_sock(sk);
+		return err;
+	}
+
 	if (optlen < sizeof(int))
 		return -EINVAL;
 
@@ -2211,6 +2230,16 @@ int tcp_getsockopt(struct sock *sk, int level, int optname, char __user *optval,
 	case TCP_QUICKACK:
 		val = !tp->ack.pingpong;
 		break;
+
+	case TCP_CONGESTION:
+		if (get_user(len, optlen))
+			return -EFAULT;
+		len = min_t(unsigned int, len, TCP_CA_NAME_MAX);
+		if (put_user(len, optlen))
+			return -EFAULT;
+		if (copy_to_user(optval, tp->ca_ops->name, len))
+			return -EFAULT;
+		return 0;
 	default:
 		return -ENOPROTOOPT;
 	};
@@ -2224,7 +2253,7 @@ int tcp_getsockopt(struct sock *sk, int level, int optname, char __user *optval,
 
 
 extern void __skb_cb_too_small_for_tcp(int, int);
-extern void tcpdiag_init(void);
+extern struct tcp_congestion_ops tcp_reno;
 
 static __initdata unsigned long thash_entries;
 static int __init set_thash_entries(char *str)
diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c
index 665394a63ae..4970d10a778 100644
--- a/net/ipv4/tcp_cong.c
+++ b/net/ipv4/tcp_cong.c
@@ -21,7 +21,7 @@ static struct tcp_congestion_ops *tcp_ca_find(const char *name)
 {
 	struct tcp_congestion_ops *e;
 
-	list_for_each_entry(e, &tcp_cong_list, list) {
+	list_for_each_entry_rcu(e, &tcp_cong_list, list) {
 		if (strcmp(e->name, name) == 0)
 			return e;
 	}
@@ -77,6 +77,9 @@ void tcp_init_congestion_control(struct tcp_sock *tp)
 {
 	struct tcp_congestion_ops *ca;
 
+	if (tp->ca_ops != &tcp_init_congestion_ops)
+		return;
+
 	rcu_read_lock();
 	list_for_each_entry_rcu(ca, &tcp_cong_list, list) {
 		if (try_module_get(ca->owner)) {
@@ -139,6 +142,34 @@ void tcp_get_default_congestion_control(char *name)
 	rcu_read_unlock();
 }
 
+/* Change congestion control for socket */
+int tcp_set_congestion_control(struct tcp_sock *tp, const char *name)
+{
+	struct tcp_congestion_ops *ca;
+	int err = 0;
+
+	rcu_read_lock();
+	ca = tcp_ca_find(name);
+	if (ca == tp->ca_ops)
+		goto out;
+
+	if (!ca)
+		err = -ENOENT;
+
+	else if (!try_module_get(ca->owner))
+		err = -EBUSY;
+
+	else {
+		tcp_cleanup_congestion_control(tp);
+		tp->ca_ops = ca;
+		if (tp->ca_ops->init)
+			tp->ca_ops->init(tp);
+	}
+ out:
+	rcu_read_unlock();
+	return err;
+}
+
 /*
  * TCP Reno congestion control
  * This is special case used for fallback as well.
@@ -192,4 +223,15 @@ struct tcp_congestion_ops tcp_reno = {
 	.min_cwnd	= tcp_reno_min_cwnd,
 };
 
-EXPORT_SYMBOL_GPL(tcp_reno);
+/* Initial congestion control used (until SYN)
+ * really reno under another name so we can tell difference
+ * during tcp_set_default_congestion_control
+ */
+struct tcp_congestion_ops tcp_init_congestion_ops  = {
+	.name		= "",
+	.owner		= THIS_MODULE,
+	.ssthresh	= tcp_reno_ssthresh,
+	.cong_avoid	= tcp_reno_cong_avoid,
+	.min_cwnd	= tcp_reno_min_cwnd,
+};
+EXPORT_SYMBOL_GPL(tcp_init_congestion_ops);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 9122814c13a..ebf112347a9 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2048,7 +2048,7 @@ static int tcp_v4_init_sock(struct sock *sk)
 	tp->mss_cache_std = tp->mss_cache = 536;
 
 	tp->reordering = sysctl_tcp_reordering;
-	tp->ca_ops = &tcp_reno;
+	tp->ca_ops = &tcp_init_congestion_ops;
 
 	sk->sk_state = TCP_CLOSE;
 
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index fce56039b0e..9dac7fdf472 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2025,7 +2025,7 @@ static int tcp_v6_init_sock(struct sock *sk)
 	sk->sk_state = TCP_CLOSE;
 
 	tp->af_specific = &ipv6_specific;
-	tp->ca_ops = &tcp_reno;
+	tp->ca_ops = &tcp_init_congestion_ops;
 	sk->sk_write_space = sk_stream_write_space;
 	sock_set_flag(sk, SOCK_USE_WRITE_QUEUE);
 
-- 
cgit v1.2.3-70-g09d2


From 2de4ff7bd658c97fb357efa3095a509674dacb5a Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Thu, 23 Jun 2005 20:49:30 -0700
Subject: [LIB]: Textsearch infrastructure.

The textsearch infrastructure provides text searching
facitilies for both linear and non-linear data.
Individual search algorithms are implemented in modules
and chosen by the user.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/textsearch.h | 180 +++++++++++++++++++++++++
 lib/Kconfig                |   8 +-
 lib/Makefile               |   2 +
 lib/textsearch.c           | 317 +++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 506 insertions(+), 1 deletion(-)
 create mode 100644 include/linux/textsearch.h
 create mode 100644 lib/textsearch.c

(limited to 'include')

diff --git a/include/linux/textsearch.h b/include/linux/textsearch.h
new file mode 100644
index 00000000000..941f45ac117
--- /dev/null
+++ b/include/linux/textsearch.h
@@ -0,0 +1,180 @@
+#ifndef __LINUX_TEXTSEARCH_H
+#define __LINUX_TEXTSEARCH_H
+
+#ifdef __KERNEL__
+
+#include <linux/types.h>
+#include <linux/list.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/err.h>
+
+struct ts_config;
+
+/**
+ * TS_AUTOLOAD - Automatically load textsearch modules when needed
+ */
+#define TS_AUTOLOAD	1
+
+/**
+ * struct ts_state - search state
+ * @offset: offset for next match
+ * @cb: control buffer, for persistant variables of get_next_block()
+ */
+struct ts_state
+{
+	unsigned int		offset;
+	char			cb[40];
+};
+
+/**
+ * struct ts_ops - search module operations
+ * @name: name of search algorithm
+ * @init: initialization function to prepare a search
+ * @find: find the next occurrence of the pattern
+ * @destroy: destroy algorithm specific parts of a search configuration
+ * @get_pattern: return head of pattern
+ * @get_pattern_len: return length of pattern
+ * @owner: module reference to algorithm
+ */
+struct ts_ops
+{
+	const char		*name;
+	struct ts_config *	(*init)(const void *, unsigned int, int);
+	unsigned int		(*find)(struct ts_config *,
+					struct ts_state *);
+	void			(*destroy)(struct ts_config *);
+	void *			(*get_pattern)(struct ts_config *);
+	unsigned int		(*get_pattern_len)(struct ts_config *);
+	struct module		*owner;
+	struct list_head	list;
+};
+
+/**
+ * struct ts_config - search configuration
+ * @ops: operations of chosen algorithm
+ * @get_next_block: callback to fetch the next block to search in
+ * @finish: callback to finalize a search
+ */
+struct ts_config
+{
+	struct ts_ops		*ops;
+
+	/**
+	 * get_next_block - fetch next block of data
+	 * @consumed: number of bytes consumed by the caller
+	 * @dst: destination buffer
+	 * @conf: search configuration
+	 * @state: search state
+	 *
+	 * Called repeatedly until 0 is returned. Must assign the
+	 * head of the next block of data to &*dst and return the length
+	 * of the block or 0 if at the end. consumed == 0 indicates
+	 * a new search. May store/read persistant values in state->cb.
+	 */
+	unsigned int		(*get_next_block)(unsigned int consumed,
+						  const u8 **dst,
+						  struct ts_config *conf,
+						  struct ts_state *state);
+
+	/**
+	 * finish - finalize/clean a series of get_next_block() calls
+	 * @conf: search configuration
+	 * @state: search state
+	 *
+	 * Called after the last use of get_next_block(), may be used
+	 * to cleanup any leftovers.
+	 */
+	void			(*finish)(struct ts_config *conf,
+					  struct ts_state *state);
+};
+
+/**
+ * textsearch_next - continue searching for a pattern
+ * @conf: search configuration
+ * @state: search state
+ *
+ * Continues a search looking for more occurrences of the pattern.
+ * textsearch_find() must be called to find the first occurrence
+ * in order to reset the state.
+ *
+ * Returns the position of the next occurrence of the pattern or
+ * UINT_MAX if not match was found.
+ */ 
+static inline unsigned int textsearch_next(struct ts_config *conf,
+					   struct ts_state *state)
+{
+	unsigned int ret = conf->ops->find(conf, state);
+
+	if (conf->finish)
+		conf->finish(conf, state);
+
+	return ret;
+}
+
+/**
+ * textsearch_find - start searching for a pattern
+ * @conf: search configuration
+ * @state: search state
+ *
+ * Returns the position of first occurrence of the pattern or
+ * UINT_MAX if no match was found.
+ */ 
+static inline unsigned int textsearch_find(struct ts_config *conf,
+					   struct ts_state *state)
+{
+	state->offset = 0;
+	return textsearch_next(conf, state);
+}
+
+/**
+ * textsearch_get_pattern - return head of the pattern
+ * @conf: search configuration
+ */
+static inline void *textsearch_get_pattern(struct ts_config *conf)
+{
+	return conf->ops->get_pattern(conf);
+}
+
+/**
+ * textsearch_get_pattern_len - return length of the pattern
+ * @conf: search configuration
+ */
+static inline unsigned int textsearch_get_pattern_len(struct ts_config *conf)
+{
+	return conf->ops->get_pattern_len(conf);
+}
+
+extern int textsearch_register(struct ts_ops *);
+extern int textsearch_unregister(struct ts_ops *);
+extern struct ts_config *textsearch_prepare(const char *, const void *,
+					    unsigned int, int, int);
+extern void textsearch_destroy(struct ts_config *conf);
+extern unsigned int textsearch_find_continuous(struct ts_config *,
+					       struct ts_state *,
+					       const void *, unsigned int);
+
+
+#define TS_PRIV_ALIGNTO	8
+#define TS_PRIV_ALIGN(len) (((len) + TS_PRIV_ALIGNTO-1) & ~(TS_PRIV_ALIGNTO-1))
+
+static inline struct ts_config *alloc_ts_config(size_t payload, int gfp_mask)
+{
+	struct ts_config *conf;
+
+	conf = kmalloc(TS_PRIV_ALIGN(sizeof(*conf)) + payload, gfp_mask);
+	if (conf == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	memset(conf, 0, TS_PRIV_ALIGN(sizeof(*conf)) + payload);
+	return conf;
+}
+
+static inline void *ts_config_priv(struct ts_config *conf)
+{
+	return ((u8 *) conf + TS_PRIV_ALIGN(sizeof(struct ts_config)));
+}
+
+#endif /* __KERNEL__ */
+
+#endif
diff --git a/lib/Kconfig b/lib/Kconfig
index 2d4d4e3bc4a..5bc2d523e6d 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -63,5 +63,11 @@ config REED_SOLOMON_ENC16
 config REED_SOLOMON_DEC16
 	boolean
 
-endmenu
+config TEXTSEARCH
+	boolean "Textsearch infrastructure"
+	default y
+	help
+	  Say Y here if you want to provide a textsearch infrastructure
+	  to other subsystems.
 
+endmenu
diff --git a/lib/Makefile b/lib/Makefile
index dcb4231916e..3e917436ad6 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -36,6 +36,8 @@ obj-$(CONFIG_ZLIB_INFLATE) += zlib_inflate/
 obj-$(CONFIG_ZLIB_DEFLATE) += zlib_deflate/
 obj-$(CONFIG_REED_SOLOMON) += reed_solomon/
 
+lib-$(CONFIG_TEXTSEARCH) += textsearch.o
+
 hostprogs-y	:= gen_crc32table
 clean-files	:= crc32table.h
 
diff --git a/lib/textsearch.c b/lib/textsearch.c
new file mode 100644
index 00000000000..1e934c196f0
--- /dev/null
+++ b/lib/textsearch.c
@@ -0,0 +1,317 @@
+/*
+ * lib/textsearch.c	Generic text search interface
+ *
+ *		This program is free software; you can redistribute it and/or
+ *		modify it under the terms of the GNU General Public License
+ *		as published by the Free Software Foundation; either version
+ *		2 of the License, or (at your option) any later version.
+ *
+ * Authors:	Thomas Graf <tgraf@suug.ch>
+ * 		Pablo Neira Ayuso <pablo@eurodev.net>
+ *
+ * ==========================================================================
+ *
+ * INTRODUCTION
+ *
+ *   The textsearch infrastructure provides text searching facitilies for
+ *   both linear and non-linear data. Individual search algorithms are
+ *   implemented in modules and chosen by the user.
+ *
+ * ARCHITECTURE
+ *
+ *      User
+ *     +----------------+
+ *     |        finish()|<--------------(6)-----------------+
+ *     |get_next_block()|<--------------(5)---------------+ |
+ *     |                |                     Algorithm   | |
+ *     |                |                    +------------------------------+
+ *     |                |                    |  init()   find()   destroy() |
+ *     |                |                    +------------------------------+
+ *     |                |       Core API           ^       ^          ^
+ *     |                |      +---------------+  (2)     (4)        (8)
+ *     |             (1)|----->| prepare()     |---+       |          |
+ *     |             (3)|----->| find()/next() |-----------+          |
+ *     |             (7)|----->| destroy()     |----------------------+
+ *     +----------------+      +---------------+
+ *  
+ *   (1) User configures a search by calling _prepare() specifying the
+ *       search parameters such as the pattern and algorithm name.
+ *   (2) Core requests the algorithm to allocate and initialize a search
+ *       configuration according to the specified parameters.
+ *   (3) User starts the search(es) by calling _find() or _next() to
+ *       fetch subsequent occurrences. A state variable is provided
+ *       to the algorihtm to store persistant variables.
+ *   (4) Core eventually resets the search offset and forwards the find()
+ *       request to the algorithm.
+ *   (5) Algorithm calls get_next_block() provided by the user continously
+ *       to fetch the data to be searched in block by block.
+ *   (6) Algorithm invokes finish() after the last call to get_next_block
+ *       to clean up any leftovers from get_next_block. (Optional)
+ *   (7) User destroys the configuration by calling _destroy().
+ *   (8) Core notifies the algorithm to destroy algorithm specific
+ *       allocations. (Optional)
+ *
+ * USAGE
+ *
+ *   Before a search can be performed, a configuration must be created
+ *   by calling textsearch_prepare() specyfing the searching algorithm and
+ *   the pattern to look for. The returned configuration may then be used
+ *   for an arbitary amount of times and even in parallel as long as a
+ *   separate struct ts_state variable is provided to every instance.
+ *
+ *   The actual search is performed by either calling textsearch_find_-
+ *   continuous() for linear data or by providing an own get_next_block()
+ *   implementation and calling textsearch_find(). Both functions return
+ *   the position of the first occurrence of the patern or UINT_MAX if
+ *   no match was found. Subsequent occurences can be found by calling
+ *   textsearch_next() regardless of the linearity of the data.
+ *
+ *   Once you're done using a configuration it must be given back via
+ *   textsearch_destroy.
+ *
+ * EXAMPLE
+ *
+ *   int pos;
+ *   struct ts_config *conf;
+ *   struct ts_state state;
+ *   const char *pattern = "chicken";
+ *   const char *example = "We dance the funky chicken";
+ *
+ *   conf = textsearch_prepare("kmp", pattern, strlen(pattern),
+ *                             GFP_KERNEL, TS_AUTOLOAD);
+ *   if (IS_ERR(conf)) {
+ *       err = PTR_ERR(conf);
+ *       goto errout;
+ *   }
+ *
+ *   pos = textsearch_find_continuous(conf, &state, example, strlen(example));
+ *   if (pos != UINT_MAX)
+ *       panic("Oh my god, dancing chickens at %d\n", pos);
+ *
+ *   textsearch_destroy(conf);
+ *
+ * ==========================================================================
+ */
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/string.h>
+#include <linux/init.h>
+#include <linux/rcupdate.h>
+#include <linux/err.h>
+#include <linux/textsearch.h>
+
+static LIST_HEAD(ts_ops);
+static DEFINE_SPINLOCK(ts_mod_lock);
+
+static inline struct ts_ops *lookup_ts_algo(const char *name)
+{
+	struct ts_ops *o;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(o, &ts_ops, list) {
+		if (!strcmp(name, o->name)) {
+			if (!try_module_get(o->owner))
+				o = NULL;
+			rcu_read_unlock();
+			return o;
+		}
+	}
+	rcu_read_unlock();
+
+	return NULL;
+}
+
+/**
+ * textsearch_register - register a textsearch module
+ * @ops: operations lookup table
+ *
+ * This function must be called by textsearch modules to announce
+ * their presence. The specified &@ops must have %name set to a
+ * unique identifier and the callbacks find(), init(), get_pattern(),
+ * and get_pattern_len() must be implemented.
+ *
+ * Returns 0 or -EEXISTS if another module has already registered
+ * with same name.
+ */
+int textsearch_register(struct ts_ops *ops)
+{
+	int err = -EEXIST;
+	struct ts_ops *o;
+
+	if (ops->name == NULL || ops->find == NULL || ops->init == NULL ||
+	    ops->get_pattern == NULL || ops->get_pattern_len == NULL)
+		return -EINVAL;
+
+	spin_lock(&ts_mod_lock);
+	list_for_each_entry(o, &ts_ops, list) {
+		if (!strcmp(ops->name, o->name))
+			goto errout;
+	}
+
+	list_add_tail_rcu(&ops->list, &ts_ops);
+	err = 0;
+errout:
+	spin_unlock(&ts_mod_lock);
+	return err;
+}
+
+/**
+ * textsearch_unregister - unregister a textsearch module
+ * @ops: operations lookup table
+ *
+ * This function must be called by textsearch modules to announce
+ * their disappearance for examples when the module gets unloaded.
+ * The &ops parameter must be the same as the one during the
+ * registration.
+ *
+ * Returns 0 on success or -ENOENT if no matching textsearch
+ * registration was found.
+ */
+int textsearch_unregister(struct ts_ops *ops)
+{
+	int err = 0;
+	struct ts_ops *o;
+
+	spin_lock(&ts_mod_lock);
+	list_for_each_entry(o, &ts_ops, list) {
+		if (o == ops) {
+			list_del_rcu(&o->list);
+			goto out;
+		}
+	}
+
+	err = -ENOENT;
+out:
+	spin_unlock(&ts_mod_lock);
+	return err;
+}
+
+struct ts_linear_state
+{
+	unsigned int	len;
+	const void	*data;
+};
+
+static unsigned int get_linear_data(unsigned int consumed, const u8 **dst,
+				    struct ts_config *conf,
+				    struct ts_state *state)
+{
+	struct ts_linear_state *st = (struct ts_linear_state *) state->cb;
+
+	if (likely(consumed < st->len)) {
+		*dst = st->data + consumed;
+		return st->len - consumed;
+	}
+
+	return 0;
+}
+
+/**
+ * textsearch_find_continuous - search a pattern in continuous/linear data
+ * @conf: search configuration
+ * @state: search state
+ * @data: data to search in
+ * @len: length of data
+ *
+ * A simplified version of textsearch_find() for continuous/linear data.
+ * Call textsearch_next() to retrieve subsequent matches.
+ *
+ * Returns the position of first occurrence of the pattern or
+ * UINT_MAX if no occurrence was found.
+ */ 
+unsigned int textsearch_find_continuous(struct ts_config *conf,
+					struct ts_state *state,
+					const void *data, unsigned int len)
+{
+	struct ts_linear_state *st = (struct ts_linear_state *) state->cb;
+
+	conf->get_next_block = get_linear_data;
+	st->data = data;
+	st->len = len;
+
+	return textsearch_find(conf, state);
+}
+
+/**
+ * textsearch_prepare - Prepare a search
+ * @algo: name of search algorithm
+ * @pattern: pattern data
+ * @len: length of pattern
+ * @gfp_mask: allocation mask
+ * @flags: search flags
+ *
+ * Looks up the search algorithm module and creates a new textsearch
+ * configuration for the specified pattern. Upon completion all
+ * necessary refcnts are held and the configuration must be put back
+ * using textsearch_put() after usage.
+ *
+ * Note: The format of the pattern may not be compatible between
+ *       the various search algorithms.
+ *
+ * Returns a new textsearch configuration according to the specified
+ *         parameters or a ERR_PTR().
+ */
+struct ts_config *textsearch_prepare(const char *algo, const void *pattern,
+				     unsigned int len, int gfp_mask, int flags)
+{
+	int err = -ENOENT;
+	struct ts_config *conf;
+	struct ts_ops *ops;
+	
+	ops = lookup_ts_algo(algo);
+#ifdef CONFIG_KMOD
+	/*
+	 * Why not always autoload you may ask. Some users are
+	 * in a situation where requesting a module may deadlock,
+	 * especially when the module is located on a NFS mount.
+	 */
+	if (ops == NULL && flags & TS_AUTOLOAD) {
+		request_module("ts_%s", algo);
+		ops = lookup_ts_algo(algo);
+	}
+#endif
+
+	if (ops == NULL)
+		goto errout;
+
+	conf = ops->init(pattern, len, gfp_mask);
+	if (IS_ERR(conf)) {
+		err = PTR_ERR(conf);
+		goto errout;
+	}
+
+	conf->ops = ops;
+	return conf;
+
+errout:
+	if (ops)
+		module_put(ops->owner);
+		
+	return ERR_PTR(err);
+}
+
+/**
+ * textsearch_destroy - destroy a search configuration
+ * @conf: search configuration
+ *
+ * Releases all references of the configuration and frees
+ * up the memory.
+ */
+void textsearch_destroy(struct ts_config *conf)
+{
+	if (conf->ops) {
+		if (conf->ops->destroy)
+			conf->ops->destroy(conf);
+		module_put(conf->ops->owner);
+	}
+
+	kfree(conf);
+}
+
+EXPORT_SYMBOL(textsearch_register);
+EXPORT_SYMBOL(textsearch_unregister);
+EXPORT_SYMBOL(textsearch_prepare);
+EXPORT_SYMBOL(textsearch_find_continuous);
+EXPORT_SYMBOL(textsearch_destroy);
-- 
cgit v1.2.3-70-g09d2


From 6408f79cce401e1bfecf923e7156f84f96e021e3 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Thu, 23 Jun 2005 20:59:16 -0700
Subject: [LIB]: Naive finite state machine based textsearch

A finite state machine consists of n states (struct ts_fsm_token)
representing the pattern as a finite automation. The data is read
sequentially on a octet basis. Every state token specifies the number
of recurrences and the type of value accepted which can be either a
specific character or ctype based set of characters. The available
type of recurrences include 1, (0|1), [0 n], and [1 n].

The algorithm differs between strict/non-strict mode specyfing
whether the pattern has to start at the first octect. Strict mode
is enabled by default and can be disabled by inserting
TS_FSM_HEAD_IGNORE as the first token in the chain.

The runtime performance of the algorithm should be around O(n),
however while in strict mode the average runtime can be better.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/textsearch_fsm.h |  48 ++++++
 lib/Kconfig                    |  11 ++
 lib/Makefile                   |   1 +
 lib/ts_fsm.c                   | 338 +++++++++++++++++++++++++++++++++++++++++
 4 files changed, 398 insertions(+)
 create mode 100644 include/linux/textsearch_fsm.h
 create mode 100644 lib/ts_fsm.c

(limited to 'include')

diff --git a/include/linux/textsearch_fsm.h b/include/linux/textsearch_fsm.h
new file mode 100644
index 00000000000..fdfa078c66e
--- /dev/null
+++ b/include/linux/textsearch_fsm.h
@@ -0,0 +1,48 @@
+#ifndef __LINUX_TEXTSEARCH_FSM_H
+#define __LINUX_TEXTSEARCH_FSM_H
+
+#include <linux/types.h>
+
+enum {
+	TS_FSM_SPECIFIC,	/* specific character */
+	TS_FSM_WILDCARD,	/* any character */
+	TS_FSM_DIGIT,		/* isdigit() */
+	TS_FSM_XDIGIT,		/* isxdigit() */
+	TS_FSM_PRINT,		/* isprint() */
+	TS_FSM_ALPHA,		/* isalpha() */
+	TS_FSM_ALNUM,		/* isalnum() */
+	TS_FSM_ASCII,		/* isascii() */
+	TS_FSM_CNTRL,		/* iscntrl() */
+	TS_FSM_GRAPH,		/* isgraph() */
+	TS_FSM_LOWER,		/* islower() */
+	TS_FSM_UPPER,		/* isupper() */
+	TS_FSM_PUNCT,		/* ispunct() */
+	TS_FSM_SPACE,		/* isspace() */
+	__TS_FSM_TYPE_MAX,
+};
+#define TS_FSM_TYPE_MAX (__TS_FSM_TYPE_MAX - 1)
+
+enum {
+	TS_FSM_SINGLE,		/* 1 occurrence */
+	TS_FSM_PERHAPS,		/* 1 or 0 occurrence */
+	TS_FSM_ANY,		/* 0..n occurrences */
+	TS_FSM_MULTI,		/* 1..n occurrences */
+	TS_FSM_HEAD_IGNORE,	/* 0..n ignored occurrences at head */
+	__TS_FSM_RECUR_MAX,
+};
+#define TS_FSM_RECUR_MAX (__TS_FSM_RECUR_MAX - 1)
+
+/**
+ * struct ts_fsm_token - state machine token (state)
+ * @type: type of token
+ * @recur: number of recurrences
+ * @value: character value for TS_FSM_SPECIFIC
+ */
+struct ts_fsm_token
+{
+	__u16		type;
+	__u8		recur;
+	__u8		value;
+};
+
+#endif
diff --git a/lib/Kconfig b/lib/Kconfig
index 16b8fa2175e..455833a9e31 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -80,4 +80,15 @@ config TEXTSEARCH_KMP
 	  To compile this code as a module, choose M here: the
 	  module will be called ts_kmp.
 
+config TEXTSEARCH_FSM
+	depends on TEXTSEARCH
+	tristate "Finite state machine"
+	help
+	  Say Y here if you want to be able to search text using a
+	  naive finite state machine approach implementing a subset
+	  of regular expressions.
+
+	  To compile this code as a module, choose M here: the
+	  module will be called ts_fsm.
+
 endmenu
diff --git a/lib/Makefile b/lib/Makefile
index 6cdb10f312d..7f6eda44910 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -38,6 +38,7 @@ obj-$(CONFIG_REED_SOLOMON) += reed_solomon/
 
 lib-$(CONFIG_TEXTSEARCH) += textsearch.o
 obj-$(CONFIG_TEXTSEARCH_KMP) += ts_kmp.o
+obj-$(CONFIG_TEXTSEARCH_FSM) += ts_fsm.o
 
 hostprogs-y	:= gen_crc32table
 clean-files	:= crc32table.h
diff --git a/lib/ts_fsm.c b/lib/ts_fsm.c
new file mode 100644
index 00000000000..d27c0a07294
--- /dev/null
+++ b/lib/ts_fsm.c
@@ -0,0 +1,338 @@
+/*
+ * lib/ts_fsm.c	   A naive finite state machine text search approach
+ *
+ *		This program is free software; you can redistribute it and/or
+ *		modify it under the terms of the GNU General Public License
+ *		as published by the Free Software Foundation; either version
+ *		2 of the License, or (at your option) any later version.
+ *
+ * Authors:	Thomas Graf <tgraf@suug.ch>
+ *
+ * ==========================================================================
+ *
+ *   A finite state machine consists of n states (struct ts_fsm_token)
+ *   representing the pattern as a finite automation. The data is read
+ *   sequentially on a octet basis. Every state token specifies the number
+ *   of recurrences and the type of value accepted which can be either a
+ *   specific character or ctype based set of characters. The available
+ *   type of recurrences include 1, (0|1), [0 n], and [1 n].
+ *
+ *   The algorithm differs between strict/non-strict mode specyfing
+ *   whether the pattern has to start at the first octect. Strict mode
+ *   is enabled by default and can be disabled by inserting
+ *   TS_FSM_HEAD_IGNORE as the first token in the chain.
+ *
+ *   The runtime performance of the algorithm should be around O(n),
+ *   however while in strict mode the average runtime can be better.
+ */
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/string.h>
+#include <linux/ctype.h>
+#include <linux/textsearch.h>
+#include <linux/textsearch_fsm.h>
+
+struct ts_fsm
+{
+	unsigned int		ntokens;
+	struct ts_fsm_token	tokens[0];
+};
+
+/* other values derived from ctype.h */
+#define _A		0x100 /* ascii */
+#define _W		0x200 /* wildcard */
+
+/* Map to _ctype flags and some magic numbers */
+static u16 token_map[TS_FSM_TYPE_MAX+1] = {
+	[TS_FSM_SPECIFIC] = 0,
+	[TS_FSM_WILDCARD] = _W,
+	[TS_FSM_CNTRL]	  = _C,
+	[TS_FSM_LOWER]	  = _L,
+	[TS_FSM_UPPER]	  = _U,
+	[TS_FSM_PUNCT]	  = _P,
+	[TS_FSM_SPACE]	  = _S,
+	[TS_FSM_DIGIT]	  = _D,
+	[TS_FSM_XDIGIT]	  = _D | _X,
+	[TS_FSM_ALPHA]	  = _U | _L,
+	[TS_FSM_ALNUM]	  = _U | _L | _D,
+	[TS_FSM_PRINT]	  = _P | _U | _L | _D | _SP,
+	[TS_FSM_GRAPH]	  = _P | _U | _L | _D,
+	[TS_FSM_ASCII]	  = _A,
+};
+
+static u16 token_lookup_tbl[256] = {
+_W|_A|_C,      _W|_A|_C,     _W|_A|_C,     _W|_A|_C,		/*   0-  3 */
+_W|_A|_C,      _W|_A|_C,     _W|_A|_C,     _W|_A|_C,		/*   4-  7 */
+_W|_A|_C,      _W|_A|_C|_S,  _W|_A|_C|_S,  _W|_A|_C|_S,		/*   8- 11 */
+_W|_A|_C|_S,   _W|_A|_C|_S,  _W|_A|_C,     _W|_A|_C,		/*  12- 15 */
+_W|_A|_C,      _W|_A|_C,     _W|_A|_C,     _W|_A|_C,		/*  16- 19 */
+_W|_A|_C,      _W|_A|_C,     _W|_A|_C,     _W|_A|_C,		/*  20- 23 */
+_W|_A|_C,      _W|_A|_C,     _W|_A|_C,     _W|_A|_C,		/*  24- 27 */
+_W|_A|_C,      _W|_A|_C,     _W|_A|_C,     _W|_A|_C,		/*  28- 31 */
+_W|_A|_S|_SP,  _W|_A|_P,     _W|_A|_P,     _W|_A|_P,		/*  32- 35 */
+_W|_A|_P,      _W|_A|_P,     _W|_A|_P,     _W|_A|_P,		/*  36- 39 */
+_W|_A|_P,      _W|_A|_P,     _W|_A|_P,     _W|_A|_P,		/*  40- 43 */
+_W|_A|_P,      _W|_A|_P,     _W|_A|_P,     _W|_A|_P,		/*  44- 47 */
+_W|_A|_D,      _W|_A|_D,     _W|_A|_D,     _W|_A|_D,		/*  48- 51 */
+_W|_A|_D,      _W|_A|_D,     _W|_A|_D,     _W|_A|_D,		/*  52- 55 */
+_W|_A|_D,      _W|_A|_D,     _W|_A|_P,     _W|_A|_P,		/*  56- 59 */
+_W|_A|_P,      _W|_A|_P,     _W|_A|_P,     _W|_A|_P,		/*  60- 63 */
+_W|_A|_P,      _W|_A|_U|_X,  _W|_A|_U|_X,  _W|_A|_U|_X,		/*  64- 67 */
+_W|_A|_U|_X,   _W|_A|_U|_X,  _W|_A|_U|_X,  _W|_A|_U,		/*  68- 71 */
+_W|_A|_U,      _W|_A|_U,     _W|_A|_U,     _W|_A|_U,		/*  72- 75 */
+_W|_A|_U,      _W|_A|_U,     _W|_A|_U,     _W|_A|_U,		/*  76- 79 */
+_W|_A|_U,      _W|_A|_U,     _W|_A|_U,     _W|_A|_U,		/*  80- 83 */
+_W|_A|_U,      _W|_A|_U,     _W|_A|_U,     _W|_A|_U,		/*  84- 87 */
+_W|_A|_U,      _W|_A|_U,     _W|_A|_U,     _W|_A|_P,		/*  88- 91 */
+_W|_A|_P,      _W|_A|_P,     _W|_A|_P,     _W|_A|_P,		/*  92- 95 */
+_W|_A|_P,      _W|_A|_L|_X,  _W|_A|_L|_X,  _W|_A|_L|_X,		/*  96- 99 */
+_W|_A|_L|_X,   _W|_A|_L|_X,  _W|_A|_L|_X,  _W|_A|_L,		/* 100-103 */
+_W|_A|_L,      _W|_A|_L,     _W|_A|_L,     _W|_A|_L,		/* 104-107 */
+_W|_A|_L,      _W|_A|_L,     _W|_A|_L,     _W|_A|_L,		/* 108-111 */
+_W|_A|_L,      _W|_A|_L,     _W|_A|_L,     _W|_A|_L,		/* 112-115 */
+_W|_A|_L,      _W|_A|_L,     _W|_A|_L,     _W|_A|_L,		/* 116-119 */
+_W|_A|_L,      _W|_A|_L,     _W|_A|_L,     _W|_A|_P,		/* 120-123 */
+_W|_A|_P,      _W|_A|_P,     _W|_A|_P,     _W|_A|_C,		/* 124-127 */
+_W,            _W,           _W,           _W,			/* 128-131 */
+_W,            _W,           _W,           _W,			/* 132-135 */
+_W,            _W,           _W,           _W,			/* 136-139 */
+_W,            _W,           _W,           _W,			/* 140-143 */
+_W,            _W,           _W,           _W,			/* 144-147 */
+_W,            _W,           _W,           _W,			/* 148-151 */
+_W,            _W,           _W,           _W,			/* 152-155 */
+_W,            _W,           _W,           _W,			/* 156-159 */
+_W|_S|_SP,     _W|_P,        _W|_P,        _W|_P,		/* 160-163 */
+_W|_P,         _W|_P,        _W|_P,        _W|_P,		/* 164-167 */
+_W|_P,         _W|_P,        _W|_P,        _W|_P,		/* 168-171 */
+_W|_P,         _W|_P,        _W|_P,        _W|_P,		/* 172-175 */
+_W|_P,         _W|_P,        _W|_P,        _W|_P,		/* 176-179 */
+_W|_P,         _W|_P,        _W|_P,        _W|_P,		/* 180-183 */
+_W|_P,         _W|_P,        _W|_P,        _W|_P,		/* 184-187 */
+_W|_P,         _W|_P,        _W|_P,        _W|_P,		/* 188-191 */
+_W|_U,         _W|_U,        _W|_U,        _W|_U,		/* 192-195 */
+_W|_U,         _W|_U,        _W|_U,        _W|_U,		/* 196-199 */
+_W|_U,         _W|_U,        _W|_U,        _W|_U,		/* 200-203 */
+_W|_U,         _W|_U,        _W|_U,        _W|_U,		/* 204-207 */
+_W|_U,         _W|_U,        _W|_U,        _W|_U,		/* 208-211 */
+_W|_U,         _W|_U,        _W|_U,        _W|_P,		/* 212-215 */
+_W|_U,         _W|_U,        _W|_U,        _W|_U,		/* 216-219 */
+_W|_U,         _W|_U,        _W|_U,        _W|_L,		/* 220-223 */
+_W|_L,         _W|_L,        _W|_L,        _W|_L,		/* 224-227 */
+_W|_L,         _W|_L,        _W|_L,        _W|_L,		/* 228-231 */
+_W|_L,         _W|_L,        _W|_L,        _W|_L,		/* 232-235 */
+_W|_L,         _W|_L,        _W|_L,        _W|_L,		/* 236-239 */
+_W|_L,         _W|_L,        _W|_L,        _W|_L,		/* 240-243 */
+_W|_L,         _W|_L,        _W|_L,        _W|_P,		/* 244-247 */
+_W|_L,         _W|_L,        _W|_L,        _W|_L,		/* 248-251 */
+_W|_L,         _W|_L,        _W|_L,        _W|_L};		/* 252-255 */
+
+static inline int match_token(struct ts_fsm_token *t, u8 d)
+{
+	if (t->type)
+		return (token_lookup_tbl[d] & t->type) != 0;
+	else
+		return t->value == d;
+}
+
+static unsigned int fsm_find(struct ts_config *conf, struct ts_state *state)
+{
+	struct ts_fsm *fsm = ts_config_priv(conf);
+	struct ts_fsm_token *cur = NULL, *next;
+	unsigned int match_start, block_idx = 0, tok_idx;
+	unsigned block_len = 0, strict, consumed = state->offset;
+	const u8 *data;
+
+#define GET_NEXT_BLOCK()		\
+({	consumed += block_idx;		\
+	block_idx = 0;			\
+	block_len = conf->get_next_block(consumed, &data, conf, state); })
+
+#define TOKEN_MISMATCH()		\
+	do {				\
+		if (strict)		\
+			goto no_match;	\
+		block_idx++;		\
+		goto startover;		\
+	} while(0)
+
+#define end_of_data() unlikely(block_idx >= block_len && !GET_NEXT_BLOCK())
+
+	if (end_of_data())
+		goto no_match;
+
+	strict = fsm->tokens[0].recur != TS_FSM_HEAD_IGNORE;
+
+startover:
+	match_start = consumed + block_idx;
+
+	for (tok_idx = 0; tok_idx < fsm->ntokens; tok_idx++) {
+		cur = &fsm->tokens[tok_idx];
+
+		if (likely(tok_idx < (fsm->ntokens - 1)))
+			next = &fsm->tokens[tok_idx + 1];
+		else
+			next = NULL;
+
+		switch (cur->recur) {
+		case TS_FSM_SINGLE:
+			if (end_of_data())
+				goto no_match;
+
+			if (!match_token(cur, data[block_idx]))
+				TOKEN_MISMATCH();
+			break;
+
+		case TS_FSM_PERHAPS:
+			if (end_of_data() ||
+			    !match_token(cur, data[block_idx]))
+				continue;
+			break;
+
+		case TS_FSM_MULTI:
+			if (end_of_data())
+				goto no_match;
+
+			if (!match_token(cur, data[block_idx]))
+				TOKEN_MISMATCH();
+
+			block_idx++;
+			/* fall through */
+
+		case TS_FSM_ANY:
+			if (next == NULL)
+				goto found_match;
+
+			if (end_of_data())
+				continue;
+
+			while (!match_token(next, data[block_idx])) {
+				if (!match_token(cur, data[block_idx]))
+					TOKEN_MISMATCH();
+				block_idx++;
+				if (end_of_data())
+					goto no_match;
+			}
+			continue;
+
+		/*
+		 * Optimization: Prefer small local loop over jumping
+		 * back and forth until garbage at head is munched.
+		 */
+		case TS_FSM_HEAD_IGNORE:
+			if (end_of_data())
+				continue;
+
+			while (!match_token(next, data[block_idx])) {
+				/*
+				 * Special case, don't start over upon
+				 * a mismatch, give the user the
+				 * chance to specify the type of data
+				 * allowed to be ignored.
+				 */
+				if (!match_token(cur, data[block_idx]))
+					goto no_match;
+
+				block_idx++;
+				if (end_of_data())
+					goto no_match;
+			}
+
+			match_start = consumed + block_idx;
+			continue;
+		}
+
+		block_idx++;
+	}
+
+	if (end_of_data())
+		goto found_match;
+
+no_match:
+	return UINT_MAX;
+
+found_match:
+	state->offset = consumed + block_idx;
+	return match_start;
+}
+
+static struct ts_config *fsm_init(const void *pattern, unsigned int len,
+				     int gfp_mask)
+{
+	int i, err = -EINVAL;
+	struct ts_config *conf;
+	struct ts_fsm *fsm;
+	struct ts_fsm_token *tokens = (struct ts_fsm_token *) pattern;
+	unsigned int ntokens = len / sizeof(*tokens);
+	size_t priv_size = sizeof(*fsm) + len;
+
+	if (len  % sizeof(struct ts_fsm_token) || ntokens < 1)
+		goto errout;
+
+	for (i = 0; i < ntokens; i++) {
+		struct ts_fsm_token *t = &tokens[i];
+
+		if (t->type > TS_FSM_TYPE_MAX || t->recur > TS_FSM_RECUR_MAX)
+			goto errout;
+
+		if (t->recur == TS_FSM_HEAD_IGNORE &&
+		    (i != 0 || i == (ntokens - 1)))
+			goto errout;
+	}
+
+	conf = alloc_ts_config(priv_size, gfp_mask);
+	if (IS_ERR(conf))
+		return conf;
+
+	fsm = ts_config_priv(conf);
+	fsm->ntokens = ntokens;
+	memcpy(fsm->tokens, pattern, len);
+
+	for (i = 0; i < fsm->ntokens; i++) {
+		struct ts_fsm_token *t = &fsm->tokens[i];
+		t->type = token_map[t->type];
+	}
+
+	return conf;
+
+errout:
+	return ERR_PTR(err);
+}
+
+static void *fsm_get_pattern(struct ts_config *conf)
+{
+	struct ts_fsm *fsm = ts_config_priv(conf);
+	return fsm->tokens;
+}
+
+static unsigned int fsm_get_pattern_len(struct ts_config *conf)
+{
+	struct ts_fsm *fsm = ts_config_priv(conf);
+	return fsm->ntokens * sizeof(struct ts_fsm_token);
+}
+
+static struct ts_ops fsm_ops = {
+	.name		  = "fsm",
+	.find		  = fsm_find,
+	.init		  = fsm_init,
+	.get_pattern	  = fsm_get_pattern,
+	.get_pattern_len  = fsm_get_pattern_len,
+	.owner		  = THIS_MODULE,
+	.list		  = LIST_HEAD_INIT(fsm_ops.list)
+};
+
+static int __init init_fsm(void)
+{
+	return textsearch_register(&fsm_ops);
+}
+
+static void __exit exit_fsm(void)
+{
+	textsearch_unregister(&fsm_ops);
+}
+
+MODULE_LICENSE("GPL");
+
+module_init(init_fsm);
+module_exit(exit_fsm);
-- 
cgit v1.2.3-70-g09d2


From 677e90eda3bd8cfde0b748daaa46476162a03950 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Thu, 23 Jun 2005 20:59:51 -0700
Subject: [NET]: Zerocopy sequential reading of skb data

Implements sequential reading for both linear and non-linear
skb data at zerocopy cost. The data is returned in chunks of
arbitary length, therefore random access is not possible.

Usage:
	from	 := 0
	to	 := 128
	state	 := undef
	data	 := undef
	len	 := undef
	consumed := 0

	skb_prepare_seq_read(skb, from, to, &state)
	while (len = skb_seq_read(consumed, &data, &state)) != 0 do
		/* do something with 'data' of length 'len' */
		if abort then
			/* abort read if we don't wait for
			 * skb_seq_read() to return 0 */
			skb_abort_seq_read(&state)
			return
		endif
		/* not necessary to consume all of 'len' */
		consumed += len
	done

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h |  18 ++++++++
 net/core/skbuff.c      | 117 +++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 135 insertions(+)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index d7c839a2184..171a37dff83 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -321,6 +321,24 @@ extern void	      skb_over_panic(struct sk_buff *skb, int len,
 extern void	      skb_under_panic(struct sk_buff *skb, int len,
 				      void *here);
 
+struct skb_seq_state
+{
+	__u32		lower_offset;
+	__u32		upper_offset;
+	__u32		frag_idx;
+	__u32		stepped_offset;
+	struct sk_buff	*root_skb;
+	struct sk_buff	*cur_skb;
+	__u8		*frag_data;
+};
+
+extern void	      skb_prepare_seq_read(struct sk_buff *skb,
+					   unsigned int from, unsigned int to,
+					   struct skb_seq_state *st);
+extern unsigned int   skb_seq_read(unsigned int consumed, const u8 **data,
+				   struct skb_seq_state *st);
+extern void	      skb_abort_seq_read(struct skb_seq_state *st);
+
 /* Internal */
 #define skb_shinfo(SKB)		((struct skb_shared_info *)((SKB)->end))
 
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 6d68c03bc05..d285f2f7e81 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -1500,6 +1500,120 @@ void skb_split(struct sk_buff *skb, struct sk_buff *skb1, const u32 len)
 		skb_split_no_header(skb, skb1, len, pos);
 }
 
+/**
+ * skb_prepare_seq_read - Prepare a sequential read of skb data
+ * @skb: the buffer to read
+ * @from: lower offset of data to be read
+ * @to: upper offset of data to be read
+ * @st: state variable
+ *
+ * Initializes the specified state variable. Must be called before
+ * invoking skb_seq_read() for the first time.
+ */
+void skb_prepare_seq_read(struct sk_buff *skb, unsigned int from,
+			  unsigned int to, struct skb_seq_state *st)
+{
+	st->lower_offset = from;
+	st->upper_offset = to;
+	st->root_skb = st->cur_skb = skb;
+	st->frag_idx = st->stepped_offset = 0;
+	st->frag_data = NULL;
+}
+
+/**
+ * skb_seq_read - Sequentially read skb data
+ * @consumed: number of bytes consumed by the caller so far
+ * @data: destination pointer for data to be returned
+ * @st: state variable
+ *
+ * Reads a block of skb data at &consumed relative to the
+ * lower offset specified to skb_prepare_seq_read(). Assigns
+ * the head of the data block to &data and returns the length
+ * of the block or 0 if the end of the skb data or the upper
+ * offset has been reached.
+ *
+ * The caller is not required to consume all of the data
+ * returned, i.e. &consumed is typically set to the number
+ * of bytes already consumed and the next call to
+ * skb_seq_read() will return the remaining part of the block.
+ *
+ * Note: The size of each block of data returned can be arbitary,
+ *       this limitation is the cost for zerocopy seqeuental
+ *       reads of potentially non linear data.
+ *
+ * Note: Fragment lists within fragments are not implemented
+ *       at the moment, state->root_skb could be replaced with
+ *       a stack for this purpose.
+ */
+unsigned int skb_seq_read(unsigned int consumed, const u8 **data,
+			  struct skb_seq_state *st)
+{
+	unsigned int block_limit, abs_offset = consumed + st->lower_offset;
+	skb_frag_t *frag;
+
+	if (unlikely(abs_offset >= st->upper_offset))
+		return 0;
+
+next_skb:
+	block_limit = skb_headlen(st->cur_skb);
+
+	if (abs_offset < block_limit) {
+		*data = st->cur_skb->data + abs_offset;
+		return block_limit - abs_offset;
+	}
+
+	if (st->frag_idx == 0 && !st->frag_data)
+		st->stepped_offset += skb_headlen(st->cur_skb);
+
+	while (st->frag_idx < skb_shinfo(st->cur_skb)->nr_frags) {
+		frag = &skb_shinfo(st->cur_skb)->frags[st->frag_idx];
+		block_limit = frag->size + st->stepped_offset;
+
+		if (abs_offset < block_limit) {
+			if (!st->frag_data)
+				st->frag_data = kmap_skb_frag(frag);
+
+			*data = (u8 *) st->frag_data + frag->page_offset +
+				(abs_offset - st->stepped_offset);
+
+			return block_limit - abs_offset;
+		}
+
+		if (st->frag_data) {
+			kunmap_skb_frag(st->frag_data);
+			st->frag_data = NULL;
+		}
+
+		st->frag_idx++;
+		st->stepped_offset += frag->size;
+	}
+
+	if (st->cur_skb->next) {
+		st->cur_skb = st->cur_skb->next;
+		st->frag_idx = 0;
+		goto next_skb;
+	} else if (st->root_skb == st->cur_skb &&
+		   skb_shinfo(st->root_skb)->frag_list) {
+		st->cur_skb = skb_shinfo(st->root_skb)->frag_list;
+		goto next_skb;
+	}
+
+	return 0;
+}
+
+/**
+ * skb_abort_seq_read - Abort a sequential read of skb data
+ * @st: state variable
+ *
+ * Must be called if skb_seq_read() was not called until it
+ * returned 0.
+ */
+void skb_abort_seq_read(struct skb_seq_state *st)
+{
+	if (st->frag_data)
+		kunmap_skb_frag(st->frag_data);
+}
+
 void __init skb_init(void)
 {
 	skbuff_head_cache = kmem_cache_create("skbuff_head_cache",
@@ -1538,3 +1652,6 @@ EXPORT_SYMBOL(skb_queue_tail);
 EXPORT_SYMBOL(skb_unlink);
 EXPORT_SYMBOL(skb_append);
 EXPORT_SYMBOL(skb_split);
+EXPORT_SYMBOL(skb_prepare_seq_read);
+EXPORT_SYMBOL(skb_seq_read);
+EXPORT_SYMBOL(skb_abort_seq_read);
-- 
cgit v1.2.3-70-g09d2


From 3fc7e8a6d842f72d16d2623b1022814a635ab961 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Thu, 23 Jun 2005 21:00:17 -0700
Subject: [NET]: skb_find_text() - Find a text pattern in skb data

Finds a pattern in the skb data according to the specified
textsearch configuration. Use textsearch_next() to retrieve
subsequent occurrences of the pattern. Returns the offset
to the first occurrence or UINT_MAX if no match was found.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h |  5 +++++
 net/core/skbuff.c      | 40 ++++++++++++++++++++++++++++++++++++++++
 2 files changed, 45 insertions(+)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 171a37dff83..416a2e4024b 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -27,6 +27,7 @@
 #include <linux/highmem.h>
 #include <linux/poll.h>
 #include <linux/net.h>
+#include <linux/textsearch.h>
 #include <net/checksum.h>
 
 #define HAVE_ALLOC_SKB		/* For the drivers to know */
@@ -339,6 +340,10 @@ extern unsigned int   skb_seq_read(unsigned int consumed, const u8 **data,
 				   struct skb_seq_state *st);
 extern void	      skb_abort_seq_read(struct skb_seq_state *st);
 
+extern unsigned int   skb_find_text(struct sk_buff *skb, unsigned int from,
+				    unsigned int to, struct ts_config *config,
+				    struct ts_state *state);
+
 /* Internal */
 #define skb_shinfo(SKB)		((struct skb_shared_info *)((SKB)->end))
 
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index d285f2f7e81..bb73b2190ec 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -1614,6 +1614,45 @@ void skb_abort_seq_read(struct skb_seq_state *st)
 		kunmap_skb_frag(st->frag_data);
 }
 
+#define TS_SKB_CB(state)	((struct skb_seq_state *) &((state)->cb))
+
+static unsigned int skb_ts_get_next_block(unsigned int offset, const u8 **text,
+					  struct ts_config *conf,
+					  struct ts_state *state)
+{
+	return skb_seq_read(offset, text, TS_SKB_CB(state));
+}
+
+static void skb_ts_finish(struct ts_config *conf, struct ts_state *state)
+{
+	skb_abort_seq_read(TS_SKB_CB(state));
+}
+
+/**
+ * skb_find_text - Find a text pattern in skb data
+ * @skb: the buffer to look in
+ * @from: search offset
+ * @to: search limit
+ * @config: textsearch configuration
+ * @state: uninitialized textsearch state variable
+ *
+ * Finds a pattern in the skb data according to the specified
+ * textsearch configuration. Use textsearch_next() to retrieve
+ * subsequent occurrences of the pattern. Returns the offset
+ * to the first occurrence or UINT_MAX if no match was found.
+ */
+unsigned int skb_find_text(struct sk_buff *skb, unsigned int from,
+			   unsigned int to, struct ts_config *config,
+			   struct ts_state *state)
+{
+	config->get_next_block = skb_ts_get_next_block;
+	config->finish = skb_ts_finish;
+
+	skb_prepare_seq_read(skb, from, to, TS_SKB_CB(state));
+
+	return textsearch_find(config, state);
+}
+
 void __init skb_init(void)
 {
 	skbuff_head_cache = kmem_cache_create("skbuff_head_cache",
@@ -1655,3 +1694,4 @@ EXPORT_SYMBOL(skb_split);
 EXPORT_SYMBOL(skb_prepare_seq_read);
 EXPORT_SYMBOL(skb_seq_read);
 EXPORT_SYMBOL(skb_abort_seq_read);
+EXPORT_SYMBOL(skb_find_text);
-- 
cgit v1.2.3-70-g09d2


From d675c989ed2d4ba23dff615330b04371aea83534 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Thu, 23 Jun 2005 21:00:58 -0700
Subject: [PKT_SCHED]: Packet classification based on textsearch (ematch)

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pkt_cls.h              |   1 +
 include/linux/rtnetlink.h            |   7 +-
 include/linux/tc_ematch/tc_em_text.h |  19 +++++
 net/sched/Kconfig                    |  11 +++
 net/sched/Makefile                   |   1 +
 net/sched/em_text.c                  | 157 +++++++++++++++++++++++++++++++++++
 6 files changed, 194 insertions(+), 2 deletions(-)
 create mode 100644 include/linux/tc_ematch/tc_em_text.h
 create mode 100644 net/sched/em_text.c

(limited to 'include')

diff --git a/include/linux/pkt_cls.h b/include/linux/pkt_cls.h
index d2aa214d680..25d2d67c1fa 100644
--- a/include/linux/pkt_cls.h
+++ b/include/linux/pkt_cls.h
@@ -408,6 +408,7 @@ enum
 	TCF_EM_NBYTE,
 	TCF_EM_U32,
 	TCF_EM_META,
+	TCF_EM_TEXT,
 	__TCF_EM_MAX
 };
 
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index e68dbf0bf57..d021888b58f 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -892,10 +892,13 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi
 		 goto rtattr_failure; \
    	__rta_fill(skb, attrtype, attrlen, data); }) 
 
-#define RTA_PUT_NOHDR(skb, attrlen, data) \
+#define RTA_APPEND(skb, attrlen, data) \
 ({	if (unlikely(skb_tailroom(skb) < (int)(attrlen))) \
 		goto rtattr_failure; \
-	memcpy(skb_put(skb, RTA_ALIGN(attrlen)), data, attrlen); })
+	memcpy(skb_put(skb, attrlen), data, attrlen); })
+
+#define RTA_PUT_NOHDR(skb, attrlen, data) \
+	RTA_APPEND(skb, RTA_ALIGN(attrlen), data)
 
 #define RTA_PUT_U8(skb, attrtype, value) \
 ({	u8 _tmp = (value); \
diff --git a/include/linux/tc_ematch/tc_em_text.h b/include/linux/tc_ematch/tc_em_text.h
new file mode 100644
index 00000000000..7cd43e99c7f
--- /dev/null
+++ b/include/linux/tc_ematch/tc_em_text.h
@@ -0,0 +1,19 @@
+#ifndef __LINUX_TC_EM_TEXT_H
+#define __LINUX_TC_EM_TEXT_H
+
+#include <linux/pkt_cls.h>
+
+#define TC_EM_TEXT_ALGOSIZ	16
+
+struct tcf_em_text
+{
+	char		algo[TC_EM_TEXT_ALGOSIZ];
+	__u16		from_offset;
+	__u16		to_offset;
+	__u16		pattern_len;
+	__u8		from_layer:4;
+	__u8		to_layer:4;
+	__u8		pad;
+};
+
+#endif
diff --git a/net/sched/Kconfig b/net/sched/Kconfig
index b22c9beb604..95d9bc5d862 100644
--- a/net/sched/Kconfig
+++ b/net/sched/Kconfig
@@ -449,6 +449,17 @@ config NET_EMATCH_META
 	  To compile this code as a module, choose M here: the
 	  module will be called em_meta.
 
+config NET_EMATCH_TEXT
+	tristate "Textsearch"
+	depends on NET_EMATCH
+	---help---
+	  Say Y here if you want to be ablt to classify packets based on
+	  textsearch comparisons. Please select the appropriate textsearch
+	  algorithms in the Library section.
+
+	  To compile this code as a module, choose M here: the
+	  module will be called em_text.
+
 config NET_CLS_ACT
 	bool "Packet ACTION"
 	depends on EXPERIMENTAL && NET_CLS && NET_QOS
diff --git a/net/sched/Makefile b/net/sched/Makefile
index eb3fe583eba..8f58cecd626 100644
--- a/net/sched/Makefile
+++ b/net/sched/Makefile
@@ -40,3 +40,4 @@ obj-$(CONFIG_NET_EMATCH_CMP)	+= em_cmp.o
 obj-$(CONFIG_NET_EMATCH_NBYTE)	+= em_nbyte.o
 obj-$(CONFIG_NET_EMATCH_U32)	+= em_u32.o
 obj-$(CONFIG_NET_EMATCH_META)	+= em_meta.o
+obj-$(CONFIG_NET_EMATCH_TEXT)	+= em_text.o
diff --git a/net/sched/em_text.c b/net/sched/em_text.c
new file mode 100644
index 00000000000..873840d8d07
--- /dev/null
+++ b/net/sched/em_text.c
@@ -0,0 +1,157 @@
+/*
+ * net/sched/em_text.c	Textsearch ematch
+ *
+ *		This program is free software; you can redistribute it and/or
+ *		modify it under the terms of the GNU General Public License
+ *		as published by the Free Software Foundation; either version
+ *		2 of the License, or (at your option) any later version.
+ *
+ * Authors:	Thomas Graf <tgraf@suug.ch>
+ */
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/string.h>
+#include <linux/skbuff.h>
+#include <linux/textsearch.h>
+#include <linux/tc_ematch/tc_em_text.h>
+#include <net/pkt_cls.h>
+
+struct text_match
+{
+	u16			from_offset;
+	u16			to_offset;
+	u8			from_layer;
+	u8			to_layer;
+	struct ts_config	*config;
+};
+
+#define EM_TEXT_PRIV(m) ((struct text_match *) (m)->data)
+
+static int em_text_match(struct sk_buff *skb, struct tcf_ematch *m,
+			 struct tcf_pkt_info *info)
+{
+	struct text_match *tm = EM_TEXT_PRIV(m);
+	int from, to;
+	struct ts_state state;
+
+	from = tcf_get_base_ptr(skb, tm->from_layer) - skb->data;
+	from += tm->from_offset;
+
+	to = tcf_get_base_ptr(skb, tm->to_layer) - skb->data;
+	to += tm->to_offset;
+
+	return skb_find_text(skb, from, to, tm->config, &state) != UINT_MAX;
+}
+
+static int em_text_change(struct tcf_proto *tp, void *data, int len,
+			  struct tcf_ematch *m)
+{
+	struct text_match *tm;
+	struct tcf_em_text *conf = data;
+	struct ts_config *ts_conf;
+	int flags = 0;
+
+	printk("Configuring text: %s from %d:%d to %d:%d len %d\n", conf->algo, conf->from_offset,
+	    conf->from_layer, conf->to_offset, conf->to_layer, conf->pattern_len);
+
+	if (len < sizeof(*conf) || len < (sizeof(*conf) + conf->pattern_len))
+		return -EINVAL;
+
+	if (conf->from_layer > conf->to_layer)
+		return -EINVAL;
+
+	if (conf->from_layer == conf->to_layer &&
+	    conf->from_offset > conf->to_offset)
+		return -EINVAL;
+
+retry:
+	ts_conf = textsearch_prepare(conf->algo, (u8 *) conf + sizeof(*conf),
+				     conf->pattern_len, GFP_KERNEL, flags);
+
+	if (flags & TS_AUTOLOAD)
+		rtnl_lock();
+
+	if (IS_ERR(ts_conf)) {
+		if (PTR_ERR(ts_conf) == -ENOENT && !(flags & TS_AUTOLOAD)) {
+			rtnl_unlock();
+			flags |= TS_AUTOLOAD;
+			goto retry;
+		} else
+			return PTR_ERR(ts_conf);
+	} else if (flags & TS_AUTOLOAD) {
+		textsearch_destroy(ts_conf);
+		return -EAGAIN;
+	}
+
+	tm = kmalloc(sizeof(*tm), GFP_KERNEL);
+	if (tm == NULL) {
+		textsearch_destroy(ts_conf);
+		return -ENOBUFS;
+	}
+
+	tm->from_offset = conf->from_offset;
+	tm->to_offset   = conf->to_offset;
+	tm->from_layer  = conf->from_layer;
+	tm->to_layer    = conf->to_layer;
+	tm->config      = ts_conf;
+
+	m->datalen = sizeof(*tm);
+	m->data = (unsigned long) tm;
+
+	return 0;
+}
+
+static void em_text_destroy(struct tcf_proto *tp, struct tcf_ematch *m)
+{
+	textsearch_destroy(EM_TEXT_PRIV(m)->config);
+}
+
+static int em_text_dump(struct sk_buff *skb, struct tcf_ematch *m)
+{
+	struct text_match *tm = EM_TEXT_PRIV(m);
+	struct tcf_em_text conf;
+
+	strncpy(conf.algo, tm->config->ops->name, sizeof(conf.algo) - 1);
+	conf.from_offset = tm->from_offset;
+	conf.to_offset = tm->to_offset;
+	conf.from_layer = tm->from_layer;
+	conf.to_layer = tm->to_layer;
+	conf.pattern_len = textsearch_get_pattern_len(tm->config);
+	conf.pad = 0;
+
+	RTA_PUT_NOHDR(skb, sizeof(conf), &conf);
+	RTA_APPEND(skb, conf.pattern_len, textsearch_get_pattern(tm->config));
+	return 0;
+
+rtattr_failure:
+	return -1;
+}		
+
+static struct tcf_ematch_ops em_text_ops = {
+	.kind	  = TCF_EM_TEXT,
+	.change	  = em_text_change,
+	.match	  = em_text_match,
+	.destroy  = em_text_destroy,
+	.dump	  = em_text_dump,
+	.owner	  = THIS_MODULE,
+	.link	  = LIST_HEAD_INIT(em_text_ops.link)
+};
+
+static int __init init_em_text(void)
+{
+	return tcf_em_register(&em_text_ops);
+}
+
+static void __exit exit_em_text(void) 
+{
+	tcf_em_unregister(&em_text_ops);
+}
+
+MODULE_LICENSE("GPL");
+
+module_init(init_em_text);
+module_exit(exit_em_text);
-- 
cgit v1.2.3-70-g09d2


From a8acfbac75c2ffdd66fb5dfcdb7ab5aaced94fd8 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 23 Jun 2005 23:45:02 -0700
Subject: [TCP]: Need to declare 'tcp_reno' in net/tcp.h

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index d04b21188cc..ec9e20c2717 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1169,6 +1169,7 @@ extern u32 tcp_reno_ssthresh(struct tcp_sock *tp);
 extern void tcp_reno_cong_avoid(struct tcp_sock *tp, u32 ack,
 				u32 rtt, u32 in_flight, int flag);
 extern u32 tcp_reno_min_cwnd(struct tcp_sock *tp);
+extern struct tcp_congestion_ops tcp_reno;
 
 static inline void tcp_set_ca_state(struct tcp_sock *tp, u8 ca_state)
 {
-- 
cgit v1.2.3-70-g09d2


From 76d8aeabfeb1c42641a81c44280177b9a08670d8 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Thu, 23 Jun 2005 22:00:49 -0700
Subject: [PATCH] keys: Discard key spinlock and use RCU for key payload

The attached patch changes the key implementation in a number of ways:

 (1) It removes the spinlock from the key structure.

 (2) The key flags are now accessed using atomic bitops instead of
     write-locking the key spinlock and using C bitwise operators.

     The three instantiation flags are dealt with with the construction
     semaphore held during the request_key/instantiate/negate sequence, thus
     rendering the spinlock superfluous.

     The key flags are also now bit numbers not bit masks.

 (3) The key payload is now accessed using RCU. This permits the recursive
     keyring search algorithm to be simplified greatly since no locks need be
     taken other than the usual RCU preemption disablement. Searching now does
     not require any locks or semaphores to be held; merely that the starting
     keyring be pinned.

 (4) The keyring payload now includes an RCU head so that it can be disposed
     of by call_rcu(). This requires that the payload be copied on unlink to
     prevent introducing races in copy-down vs search-up.

 (5) The user key payload is now a structure with the data following it. It
     includes an RCU head like the keyring payload and for the same reason. It
     also contains a data length because the data length in the key may be
     changed on another CPU whilst an RCU protected read is in progress on the
     payload. This would then see the supposed RCU payload and the on-key data
     length getting out of sync.

     I'm tempted to drop the key's datalen entirely, except that it's used in
     conjunction with quota management and so is a little tricky to get rid
     of.

 (6) Update the keys documentation.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/keys.txt       | 285 ++++++++++++++++++++++++++-----------------
 include/linux/key-ui.h       |   6 +-
 include/linux/key.h          |  25 ++--
 security/keys/key.c          |  94 +++++++-------
 security/keys/keyctl.c       |  23 ++--
 security/keys/keyring.c      | 245 ++++++++++++++++++++++---------------
 security/keys/proc.c         |  21 ++--
 security/keys/process_keys.c |  12 +-
 security/keys/request_key.c  |  32 +++--
 security/keys/user_defined.c |  85 +++++++++----
 10 files changed, 480 insertions(+), 348 deletions(-)

(limited to 'include')

diff --git a/Documentation/keys.txt b/Documentation/keys.txt
index 36d80aeeaf2..3df40c1fe15 100644
--- a/Documentation/keys.txt
+++ b/Documentation/keys.txt
@@ -22,6 +22,7 @@ This document has the following sections:
 	- New procfs files
 	- Userspace system call interface
 	- Kernel services
+	- Notes on accessing payload contents
 	- Defining a key type
 	- Request-key callback service
 	- Key access filesystem
@@ -45,27 +46,26 @@ Each key has a number of attributes:
 	- State.
 
 
- (*) Each key is issued a serial number of type key_serial_t that is unique
-     for the lifetime of that key. All serial numbers are positive non-zero
-     32-bit integers.
+ (*) Each key is issued a serial number of type key_serial_t that is unique for
+     the lifetime of that key. All serial numbers are positive non-zero 32-bit
+     integers.
 
      Userspace programs can use a key's serial numbers as a way to gain access
      to it, subject to permission checking.
 
  (*) Each key is of a defined "type". Types must be registered inside the
-     kernel by a kernel service (such as a filesystem) before keys of that
-     type can be added or used. Userspace programs cannot define new types
-     directly.
+     kernel by a kernel service (such as a filesystem) before keys of that type
+     can be added or used. Userspace programs cannot define new types directly.
 
-     Key types are represented in the kernel by struct key_type. This defines
-     a number of operations that can be performed on a key of that type.
+     Key types are represented in the kernel by struct key_type. This defines a
+     number of operations that can be performed on a key of that type.
 
      Should a type be removed from the system, all the keys of that type will
      be invalidated.
 
  (*) Each key has a description. This should be a printable string. The key
-     type provides an operation to perform a match between the description on
-     a key and a criterion string.
+     type provides an operation to perform a match between the description on a
+     key and a criterion string.
 
  (*) Each key has an owner user ID, a group ID and a permissions mask. These
      are used to control what a process may do to a key from userspace, and
@@ -74,10 +74,10 @@ Each key has a number of attributes:
  (*) Each key can be set to expire at a specific time by the key type's
      instantiation function. Keys can also be immortal.
 
- (*) Each key can have a payload. This is a quantity of data that represent
-     the actual "key". In the case of a keyring, this is a list of keys to
-     which the keyring links; in the case of a user-defined key, it's an
-     arbitrary blob of data.
+ (*) Each key can have a payload. This is a quantity of data that represent the
+     actual "key". In the case of a keyring, this is a list of keys to which
+     the keyring links; in the case of a user-defined key, it's an arbitrary
+     blob of data.
 
      Having a payload is not required; and the payload can, in fact, just be a
      value stored in the struct key itself.
@@ -92,8 +92,8 @@ Each key has a number of attributes:
 
  (*) Each key can be in one of a number of basic states:
 
-     (*) Uninstantiated. The key exists, but does not have any data
-	 attached. Keys being requested from userspace will be in this state.
+     (*) Uninstantiated. The key exists, but does not have any data attached.
+     	 Keys being requested from userspace will be in this state.
 
      (*) Instantiated. This is the normal state. The key is fully formed, and
 	 has data attached.
@@ -140,10 +140,10 @@ The key service provides a number of features besides keys:
      clone, fork, vfork or execve occurs. A new keyring is created only when
      required.
 
-     The process-specific keyring is replaced with an empty one in the child
-     on clone, fork, vfork unless CLONE_THREAD is supplied, in which case it
-     is shared. execve also discards the process's process keyring and creates
-     a new one.
+     The process-specific keyring is replaced with an empty one in the child on
+     clone, fork, vfork unless CLONE_THREAD is supplied, in which case it is
+     shared. execve also discards the process's process keyring and creates a
+     new one.
 
      The session-specific keyring is persistent across clone, fork, vfork and
      execve, even when the latter executes a set-UID or set-GID binary. A
@@ -177,11 +177,11 @@ The key service provides a number of features besides keys:
      If a system call that modifies a key or keyring in some way would put the
      user over quota, the operation is refused and error EDQUOT is returned.
 
- (*) There's a system call interface by which userspace programs can create
-     and manipulate keys and keyrings.
+ (*) There's a system call interface by which userspace programs can create and
+     manipulate keys and keyrings.
 
- (*) There's a kernel interface by which services can register types and
-     search for keys.
+ (*) There's a kernel interface by which services can register types and search
+     for keys.
 
  (*) There's a way for the a search done from the kernel to call back to
      userspace to request a key that can't be found in a process's keyrings.
@@ -194,9 +194,9 @@ The key service provides a number of features besides keys:
 KEY ACCESS PERMISSIONS
 ======================
 
-Keys have an owner user ID, a group access ID, and a permissions mask. The
-mask has up to eight bits each for user, group and other access. Only five of
-each set of eight bits are defined. These permissions granted are:
+Keys have an owner user ID, a group access ID, and a permissions mask. The mask
+has up to eight bits each for user, group and other access. Only five of each
+set of eight bits are defined. These permissions granted are:
 
  (*) View
 
@@ -210,8 +210,8 @@ each set of eight bits are defined. These permissions granted are:
 
  (*) Write
 
-     This permits a key's payload to be instantiated or updated, or it allows
-     a link to be added to or removed from a keyring.
+     This permits a key's payload to be instantiated or updated, or it allows a
+     link to be added to or removed from a keyring.
 
  (*) Search
 
@@ -238,8 +238,8 @@ about the status of the key service:
  (*) /proc/keys
 
      This lists all the keys on the system, giving information about their
-     type, description and permissions. The payload of the key is not
-     available this way:
+     type, description and permissions. The payload of the key is not available
+     this way:
 
 	SERIAL   FLAGS  USAGE EXPY PERM   UID   GID   TYPE      DESCRIPTION: SUMMARY
 	00000001 I-----    39 perm 1f0000     0     0 keyring   _uid_ses.0: 1/4
@@ -318,21 +318,21 @@ The main syscalls are:
      If a key of the same type and description as that proposed already exists
      in the keyring, this will try to update it with the given payload, or it
      will return error EEXIST if that function is not supported by the key
-     type. The process must also have permission to write to the key to be
-     able to update it. The new key will have all user permissions granted and
-     no group or third party permissions.
+     type. The process must also have permission to write to the key to be able
+     to update it. The new key will have all user permissions granted and no
+     group or third party permissions.
 
-     Otherwise, this will attempt to create a new key of the specified type
-     and description, and to instantiate it with the supplied payload and
-     attach it to the keyring. In this case, an error will be generated if the
-     process does not have permission to write to the keyring.
+     Otherwise, this will attempt to create a new key of the specified type and
+     description, and to instantiate it with the supplied payload and attach it
+     to the keyring. In this case, an error will be generated if the process
+     does not have permission to write to the keyring.
 
      The payload is optional, and the pointer can be NULL if not required by
      the type. The payload is plen in size, and plen can be zero for an empty
      payload.
 
-     A new keyring can be generated by setting type "keyring", the keyring
-     name as the description (or NULL) and setting the payload to NULL.
+     A new keyring can be generated by setting type "keyring", the keyring name
+     as the description (or NULL) and setting the payload to NULL.
 
      User defined keys can be created by specifying type "user". It is
      recommended that a user defined key's description by prefixed with a type
@@ -369,9 +369,9 @@ The keyctl syscall functions are:
 	key_serial_t keyctl(KEYCTL_GET_KEYRING_ID, key_serial_t id,
 			    int create);
 
-     The special key specified by "id" is looked up (with the key being
-     created if necessary) and the ID of the key or keyring thus found is
-     returned if it exists.
+     The special key specified by "id" is looked up (with the key being created
+     if necessary) and the ID of the key or keyring thus found is returned if
+     it exists.
 
      If the key does not yet exist, the key will be created if "create" is
      non-zero; and the error ENOKEY will be returned if "create" is zero.
@@ -402,8 +402,8 @@ The keyctl syscall functions are:
 
      This will try to update the specified key with the given payload, or it
      will return error EOPNOTSUPP if that function is not supported by the key
-     type. The process must also have permission to write to the key to be
-     able to update it.
+     type. The process must also have permission to write to the key to be able
+     to update it.
 
      The payload is of length plen, and may be absent or empty as for
      add_key().
@@ -422,8 +422,8 @@ The keyctl syscall functions are:
 
 	long keyctl(KEYCTL_CHOWN, key_serial_t key, uid_t uid, gid_t gid);
 
-     This function permits a key's owner and group ID to be changed. Either
-     one of uid or gid can be set to -1 to suppress that change.
+     This function permits a key's owner and group ID to be changed. Either one
+     of uid or gid can be set to -1 to suppress that change.
 
      Only the superuser can change a key's owner to something other than the
      key's current owner. Similarly, only the superuser can change a key's
@@ -484,12 +484,12 @@ The keyctl syscall functions are:
 
 	long keyctl(KEYCTL_LINK, key_serial_t keyring, key_serial_t key);
 
-     This function creates a link from the keyring to the key. The process
-     must have write permission on the keyring and must have link permission
-     on the key.
+     This function creates a link from the keyring to the key. The process must
+     have write permission on the keyring and must have link permission on the
+     key.
 
-     Should the keyring not be a keyring, error ENOTDIR will result; and if
-     the keyring is full, error ENFILE will result.
+     Should the keyring not be a keyring, error ENOTDIR will result; and if the
+     keyring is full, error ENFILE will result.
 
      The link procedure checks the nesting of the keyrings, returning ELOOP if
      it appears to deep or EDEADLK if the link would introduce a cycle.
@@ -503,8 +503,8 @@ The keyctl syscall functions are:
      specified key, and removes it if found. Subsequent links to that key are
      ignored. The process must have write permission on the keyring.
 
-     If the keyring is not a keyring, error ENOTDIR will result; and if the
-     key is not present, error ENOENT will be the result.
+     If the keyring is not a keyring, error ENOTDIR will result; and if the key
+     is not present, error ENOENT will be the result.
 
 
  (*) Search a keyring tree for a key:
@@ -513,9 +513,9 @@ The keyctl syscall functions are:
 			    const char *type, const char *description,
 			    key_serial_t dest_keyring);
 
-     This searches the keyring tree headed by the specified keyring until a
-     key is found that matches the type and description criteria. Each keyring
-     is checked for keys before recursion into its children occurs.
+     This searches the keyring tree headed by the specified keyring until a key
+     is found that matches the type and description criteria. Each keyring is
+     checked for keys before recursion into its children occurs.
 
      The process must have search permission on the top level keyring, or else
      error EACCES will result. Only keyrings that the process has search
@@ -549,8 +549,8 @@ The keyctl syscall functions are:
      As much of the data as can be fitted into the buffer will be copied to
      userspace if the buffer pointer is not NULL.
 
-     On a successful return, the function will always return the amount of
-     data available rather than the amount copied.
+     On a successful return, the function will always return the amount of data
+     available rather than the amount copied.
 
 
  (*) Instantiate a partially constructed key.
@@ -568,8 +568,8 @@ The keyctl syscall functions are:
      it, and the key must be uninstantiated.
 
      If a keyring is specified (non-zero), the key will also be linked into
-     that keyring, however all the constraints applying in KEYCTL_LINK apply
-     in this case too.
+     that keyring, however all the constraints applying in KEYCTL_LINK apply in
+     this case too.
 
      The payload and plen arguments describe the payload data as for add_key().
 
@@ -587,8 +587,8 @@ The keyctl syscall functions are:
      it, and the key must be uninstantiated.
 
      If a keyring is specified (non-zero), the key will also be linked into
-     that keyring, however all the constraints applying in KEYCTL_LINK apply
-     in this case too.
+     that keyring, however all the constraints applying in KEYCTL_LINK apply in
+     this case too.
 
 
 ===============
@@ -601,17 +601,14 @@ be broken down into two areas: keys and key types.
 Dealing with keys is fairly straightforward. Firstly, the kernel service
 registers its type, then it searches for a key of that type. It should retain
 the key as long as it has need of it, and then it should release it. For a
-filesystem or device file, a search would probably be performed during the
-open call, and the key released upon close. How to deal with conflicting keys
-due to two different users opening the same file is left to the filesystem
-author to solve.
-
-When accessing a key's payload data, key->lock should be at least read locked,
-or else the data may be changed by an update being performed from userspace
-whilst the driver or filesystem is trying to access it. If no update method is
-supplied, then the key's payload may be accessed without holding a lock as
-there is no way to change it, provided it can be guaranteed that the key's
-type definition won't go away.
+filesystem or device file, a search would probably be performed during the open
+call, and the key released upon close. How to deal with conflicting keys due to
+two different users opening the same file is left to the filesystem author to
+solve.
+
+When accessing a key's payload contents, certain precautions must be taken to
+prevent access vs modification races. See the section "Notes on accessing
+payload contents" for more information.
 
 (*) To search for a key, call:
 
@@ -690,6 +687,54 @@ type definition won't go away.
 	void unregister_key_type(struct key_type *type);
 
 
+===================================
+NOTES ON ACCESSING PAYLOAD CONTENTS
+===================================
+
+The simplest payload is just a number in key->payload.value. In this case,
+there's no need to indulge in RCU or locking when accessing the payload.
+
+More complex payload contents must be allocated and a pointer to them set in
+key->payload.data. One of the following ways must be selected to access the
+data:
+
+ (1) Unmodifyable key type.
+
+     If the key type does not have a modify method, then the key's payload can
+     be accessed without any form of locking, provided that it's known to be
+     instantiated (uninstantiated keys cannot be "found").
+
+ (2) The key's semaphore.
+
+     The semaphore could be used to govern access to the payload and to control
+     the payload pointer. It must be write-locked for modifications and would
+     have to be read-locked for general access. The disadvantage of doing this
+     is that the accessor may be required to sleep.
+
+ (3) RCU.
+
+     RCU must be used when the semaphore isn't already held; if the semaphore
+     is held then the contents can't change under you unexpectedly as the
+     semaphore must still be used to serialise modifications to the key. The
+     key management code takes care of this for the key type.
+
+     However, this means using:
+
+	rcu_read_lock() ... rcu_dereference() ... rcu_read_unlock()
+
+     to read the pointer, and:
+
+	rcu_dereference() ... rcu_assign_pointer() ... call_rcu()
+
+     to set the pointer and dispose of the old contents after a grace period.
+     Note that only the key type should ever modify a key's payload.
+
+     Furthermore, an RCU controlled payload must hold a struct rcu_head for the
+     use of call_rcu() and, if the payload is of variable size, the length of
+     the payload. key->datalen cannot be relied upon to be consistent with the
+     payload just dereferenced if the key's semaphore is not held.
+
+
 ===================
 DEFINING A KEY TYPE
 ===================
@@ -717,15 +762,15 @@ The structure has a number of fields, some of which are mandatory:
 
 	int key_payload_reserve(struct key *key, size_t datalen);
 
-     With the revised data length. Error EDQUOT will be returned if this is
-     not viable.
+     With the revised data length. Error EDQUOT will be returned if this is not
+     viable.
 
 
  (*) int (*instantiate)(struct key *key, const void *data, size_t datalen);
 
      This method is called to attach a payload to a key during construction.
-     The payload attached need not bear any relation to the data passed to
-     this function.
+     The payload attached need not bear any relation to the data passed to this
+     function.
 
      If the amount of data attached to the key differs from the size in
      keytype->def_datalen, then key_payload_reserve() should be called.
@@ -734,38 +779,47 @@ The structure has a number of fields, some of which are mandatory:
      The fact that KEY_FLAG_INSTANTIATED is not set in key->flags prevents
      anything else from gaining access to the key.
 
-     This method may sleep if it wishes.
+     It is safe to sleep in this method.
 
 
  (*) int (*duplicate)(struct key *key, const struct key *source);
 
      If this type of key can be duplicated, then this method should be
-     provided. It is called to copy the payload attached to the source into
-     the new key. The data length on the new key will have been updated and
-     the quota adjusted already.
+     provided. It is called to copy the payload attached to the source into the
+     new key. The data length on the new key will have been updated and the
+     quota adjusted already.
 
      This method will be called with the source key's semaphore read-locked to
-     prevent its payload from being changed. It is safe to sleep here.
+     prevent its payload from being changed, thus RCU constraints need not be
+     applied to the source key.
+
+     This method does not have to lock the destination key in order to attach a
+     payload. The fact that KEY_FLAG_INSTANTIATED is not set in key->flags
+     prevents anything else from gaining access to the key.
+
+     It is safe to sleep in this method.
 
 
  (*) int (*update)(struct key *key, const void *data, size_t datalen);
 
-     If this type of key can be updated, then this method should be
-     provided. It is called to update a key's payload from the blob of data
-     provided.
+     If this type of key can be updated, then this method should be provided.
+     It is called to update a key's payload from the blob of data provided.
 
      key_payload_reserve() should be called if the data length might change
-     before any changes are actually made. Note that if this succeeds, the
-     type is committed to changing the key because it's already been altered,
-     so all memory allocation must be done first.
+     before any changes are actually made. Note that if this succeeds, the type
+     is committed to changing the key because it's already been altered, so all
+     memory allocation must be done first.
+
+     The key will have its semaphore write-locked before this method is called,
+     but this only deters other writers; any changes to the key's payload must
+     be made under RCU conditions, and call_rcu() must be used to dispose of
+     the old payload.
 
-     key_payload_reserve() should be called with the key->lock write locked,
-     and the changes to the key's attached payload should be made before the
-     key is locked.
+     key_payload_reserve() should be called before the changes are made, but
+     after all allocations and other potentially failing function calls are
+     made.
 
-     The key will have its semaphore write-locked before this method is
-     called. Any changes to the key should be made with the key's rwlock
-     write-locked also. It is safe to sleep here.
+     It is safe to sleep in this method.
 
 
  (*) int (*match)(const struct key *key, const void *desc);
@@ -782,12 +836,12 @@ The structure has a number of fields, some of which are mandatory:
 
  (*) void (*destroy)(struct key *key);
 
-     This method is optional. It is called to discard the payload data on a
-     key when it is being destroyed.
+     This method is optional. It is called to discard the payload data on a key
+     when it is being destroyed.
 
-     This method does not need to lock the key; it can consider the key as
-     being inaccessible. Note that the key's type may have changed before this
-     function is called.
+     This method does not need to lock the key to access the payload; it can
+     consider the key as being inaccessible at this time. Note that the key's
+     type may have been changed before this function is called.
 
      It is not safe to sleep in this method; the caller may hold spinlocks.
 
@@ -797,26 +851,31 @@ The structure has a number of fields, some of which are mandatory:
      This method is optional. It is called during /proc/keys reading to
      summarise a key's description and payload in text form.
 
-     This method will be called with the key's rwlock read-locked. This will
-     prevent the key's payload and state changing; also the description should
-     not change. This also means it is not safe to sleep in this method.
+     This method will be called with the RCU read lock held. rcu_dereference()
+     should be used to read the payload pointer if the payload is to be
+     accessed. key->datalen cannot be trusted to stay consistent with the
+     contents of the payload.
+
+     The description will not change, though the key's state may.
+
+     It is not safe to sleep in this method; the RCU read lock is held by the
+     caller.
 
 
  (*) long (*read)(const struct key *key, char __user *buffer, size_t buflen);
 
      This method is optional. It is called by KEYCTL_READ to translate the
-     key's payload into something a blob of data for userspace to deal
-     with. Ideally, the blob should be in the same format as that passed in to
-     the instantiate and update methods.
+     key's payload into something a blob of data for userspace to deal with.
+     Ideally, the blob should be in the same format as that passed in to the
+     instantiate and update methods.
 
      If successful, the blob size that could be produced should be returned
      rather than the size copied.
 
-     This method will be called with the key's semaphore read-locked. This
-     will prevent the key's payload changing. It is not necessary to also
-     read-lock key->lock when accessing the key's payload. It is safe to sleep
-     in this method, such as might happen when the userspace buffer is
-     accessed.
+     This method will be called with the key's semaphore read-locked. This will
+     prevent the key's payload changing. It is not necessary to use RCU locking
+     when accessing the key's payload. It is safe to sleep in this method, such
+     as might happen when the userspace buffer is accessed.
 
 
 ============================
@@ -853,8 +912,8 @@ If it returns with the key remaining in the unconstructed state, the key will
 be marked as being negative, it will be added to the session keyring, and an
 error will be returned to the key requestor.
 
-Supplementary information may be provided from whoever or whatever invoked
-this service. This will be passed as the <callout_info> parameter. If no such
+Supplementary information may be provided from whoever or whatever invoked this
+service. This will be passed as the <callout_info> parameter. If no such
 information was made available, then "-" will be passed as this parameter
 instead.
 
diff --git a/include/linux/key-ui.h b/include/linux/key-ui.h
index 60cc7b762e7..159ca8d54e9 100644
--- a/include/linux/key-ui.h
+++ b/include/linux/key-ui.h
@@ -31,8 +31,10 @@ extern spinlock_t key_serial_lock;
  * subscribed
  */
 struct keyring_list {
-	unsigned	maxkeys;	/* max keys this list can hold */
-	unsigned	nkeys;		/* number of keys currently held */
+	struct rcu_head	rcu;		/* RCU deletion hook */
+	unsigned short	maxkeys;	/* max keys this list can hold */
+	unsigned short	nkeys;		/* number of keys currently held */
+	unsigned short	delkey;		/* key to be unlinked by RCU */
 	struct key	*keys[0];
 };
 
diff --git a/include/linux/key.h b/include/linux/key.h
index 6aa46d0e812..2c24ffaca86 100644
--- a/include/linux/key.h
+++ b/include/linux/key.h
@@ -18,7 +18,7 @@
 #include <linux/types.h>
 #include <linux/list.h>
 #include <linux/rbtree.h>
-#include <linux/spinlock.h>
+#include <linux/rcupdate.h>
 #include <asm/atomic.h>
 
 #ifdef __KERNEL__
@@ -78,7 +78,6 @@ struct key {
 	key_serial_t		serial;		/* key serial number */
 	struct rb_node		serial_node;
 	struct key_type		*type;		/* type of key */
-	rwlock_t		lock;		/* examination vs change lock */
 	struct rw_semaphore	sem;		/* change vs change sem */
 	struct key_user		*user;		/* owner of this key */
 	time_t			expiry;		/* time at which key expires (or 0) */
@@ -86,14 +85,10 @@ struct key {
 	gid_t			gid;
 	key_perm_t		perm;		/* access permissions */
 	unsigned short		quotalen;	/* length added to quota */
-	unsigned short		datalen;	/* payload data length */
-	unsigned short		flags;		/* status flags (change with lock writelocked) */
-#define KEY_FLAG_INSTANTIATED	0x00000001	/* set if key has been instantiated */
-#define KEY_FLAG_DEAD		0x00000002	/* set if key type has been deleted */
-#define KEY_FLAG_REVOKED	0x00000004	/* set if key had been revoked */
-#define KEY_FLAG_IN_QUOTA	0x00000008	/* set if key consumes quota */
-#define KEY_FLAG_USER_CONSTRUCT	0x00000010	/* set if key is being constructed in userspace */
-#define KEY_FLAG_NEGATIVE	0x00000020	/* set if key is negative */
+	unsigned short		datalen;	/* payload data length
+						 * - may not match RCU dereferenced payload
+						 * - payload should contain own length
+						 */
 
 #ifdef KEY_DEBUGGING
 	unsigned		magic;
@@ -101,6 +96,14 @@ struct key {
 #define KEY_DEBUG_MAGIC_X	0xf8e9dacbu
 #endif
 
+	unsigned long		flags;		/* status flags (change with bitops) */
+#define KEY_FLAG_INSTANTIATED	0	/* set if key has been instantiated */
+#define KEY_FLAG_DEAD		1	/* set if key type has been deleted */
+#define KEY_FLAG_REVOKED	2	/* set if key had been revoked */
+#define KEY_FLAG_IN_QUOTA	3	/* set if key consumes quota */
+#define KEY_FLAG_USER_CONSTRUCT	4	/* set if key is being constructed in userspace */
+#define KEY_FLAG_NEGATIVE	5	/* set if key is negative */
+
 	/* the description string
 	 * - this is used to match a key against search criteria
 	 * - this should be a printable string
@@ -250,6 +253,8 @@ extern int keyring_add_key(struct key *keyring,
 
 extern struct key *key_lookup(key_serial_t id);
 
+extern void keyring_replace_payload(struct key *key, void *replacement);
+
 #define key_serial(key) ((key) ? (key)->serial : 0)
 
 /*
diff --git a/security/keys/key.c b/security/keys/key.c
index 59402c84320..1fdfccb3fe4 100644
--- a/security/keys/key.c
+++ b/security/keys/key.c
@@ -294,7 +294,6 @@ struct key *key_alloc(struct key_type *type, const char *desc,
 	}
 
 	atomic_set(&key->usage, 1);
-	rwlock_init(&key->lock);
 	init_rwsem(&key->sem);
 	key->type = type;
 	key->user = user;
@@ -308,7 +307,7 @@ struct key *key_alloc(struct key_type *type, const char *desc,
 	key->payload.data = NULL;
 
 	if (!not_in_quota)
-		key->flags |= KEY_FLAG_IN_QUOTA;
+		key->flags |= 1 << KEY_FLAG_IN_QUOTA;
 
 	memset(&key->type_data, 0, sizeof(key->type_data));
 
@@ -359,7 +358,7 @@ int key_payload_reserve(struct key *key, size_t datalen)
 	key_check(key);
 
 	/* contemplate the quota adjustment */
-	if (delta != 0 && key->flags & KEY_FLAG_IN_QUOTA) {
+	if (delta != 0 && test_bit(KEY_FLAG_IN_QUOTA, &key->flags)) {
 		spin_lock(&key->user->lock);
 
 		if (delta > 0 &&
@@ -405,23 +404,17 @@ static int __key_instantiate_and_link(struct key *key,
 	down_write(&key_construction_sem);
 
 	/* can't instantiate twice */
-	if (!(key->flags & KEY_FLAG_INSTANTIATED)) {
+	if (!test_bit(KEY_FLAG_INSTANTIATED, &key->flags)) {
 		/* instantiate the key */
 		ret = key->type->instantiate(key, data, datalen);
 
 		if (ret == 0) {
 			/* mark the key as being instantiated */
-			write_lock(&key->lock);
-
 			atomic_inc(&key->user->nikeys);
-			key->flags |= KEY_FLAG_INSTANTIATED;
+			set_bit(KEY_FLAG_INSTANTIATED, &key->flags);
 
-			if (key->flags & KEY_FLAG_USER_CONSTRUCT) {
-				key->flags &= ~KEY_FLAG_USER_CONSTRUCT;
+			if (test_and_clear_bit(KEY_FLAG_USER_CONSTRUCT, &key->flags))
 				awaken = 1;
-			}
-
-			write_unlock(&key->lock);
 
 			/* and link it into the destination keyring */
 			if (keyring)
@@ -486,21 +479,17 @@ int key_negate_and_link(struct key *key,
 	down_write(&key_construction_sem);
 
 	/* can't instantiate twice */
-	if (!(key->flags & KEY_FLAG_INSTANTIATED)) {
+	if (!test_bit(KEY_FLAG_INSTANTIATED, &key->flags)) {
 		/* mark the key as being negatively instantiated */
-		write_lock(&key->lock);
-
 		atomic_inc(&key->user->nikeys);
-		key->flags |= KEY_FLAG_INSTANTIATED | KEY_FLAG_NEGATIVE;
+		set_bit(KEY_FLAG_NEGATIVE, &key->flags);
+		set_bit(KEY_FLAG_INSTANTIATED, &key->flags);
 		now = current_kernel_time();
 		key->expiry = now.tv_sec + timeout;
 
-		if (key->flags & KEY_FLAG_USER_CONSTRUCT) {
-			key->flags &= ~KEY_FLAG_USER_CONSTRUCT;
+		if (test_and_clear_bit(KEY_FLAG_USER_CONSTRUCT, &key->flags))
 			awaken = 1;
-		}
 
-		write_unlock(&key->lock);
 		ret = 0;
 
 		/* and link it into the destination keyring */
@@ -553,8 +542,10 @@ static void key_cleanup(void *data)
 	rb_erase(&key->serial_node, &key_serial_tree);
 	spin_unlock(&key_serial_lock);
 
+	key_check(key);
+
 	/* deal with the user's key tracking and quota */
-	if (key->flags & KEY_FLAG_IN_QUOTA) {
+	if (test_bit(KEY_FLAG_IN_QUOTA, &key->flags)) {
 		spin_lock(&key->user->lock);
 		key->user->qnkeys--;
 		key->user->qnbytes -= key->quotalen;
@@ -562,7 +553,7 @@ static void key_cleanup(void *data)
 	}
 
 	atomic_dec(&key->user->nkeys);
-	if (key->flags & KEY_FLAG_INSTANTIATED)
+	if (test_bit(KEY_FLAG_INSTANTIATED, &key->flags))
 		atomic_dec(&key->user->nikeys);
 
 	key_user_put(key->user);
@@ -631,9 +622,9 @@ struct key *key_lookup(key_serial_t id)
 	goto error;
 
  found:
-	/* pretent doesn't exist if it's dead */
+	/* pretend it doesn't exist if it's dead */
 	if (atomic_read(&key->usage) == 0 ||
-	    (key->flags & KEY_FLAG_DEAD) ||
+	    test_bit(KEY_FLAG_DEAD, &key->flags) ||
 	    key->type == &key_type_dead)
 		goto not_found;
 
@@ -708,12 +699,9 @@ static inline struct key *__key_update(struct key *key, const void *payload,
 
 	ret = key->type->update(key, payload, plen);
 
-	if (ret == 0) {
+	if (ret == 0)
 		/* updating a negative key instantiates it */
-		write_lock(&key->lock);
-		key->flags &= ~KEY_FLAG_NEGATIVE;
-		write_unlock(&key->lock);
-	}
+		clear_bit(KEY_FLAG_NEGATIVE, &key->flags);
 
 	up_write(&key->sem);
 
@@ -841,12 +829,9 @@ int key_update(struct key *key, const void *payload, size_t plen)
 		down_write(&key->sem);
 		ret = key->type->update(key, payload, plen);
 
-		if (ret == 0) {
+		if (ret == 0)
 			/* updating a negative key instantiates it */
-			write_lock(&key->lock);
-			key->flags &= ~KEY_FLAG_NEGATIVE;
-			write_unlock(&key->lock);
-		}
+			clear_bit(KEY_FLAG_NEGATIVE, &key->flags);
 
 		up_write(&key->sem);
 	}
@@ -892,10 +877,7 @@ struct key *key_duplicate(struct key *source, const char *desc)
 		goto error2;
 
 	atomic_inc(&key->user->nikeys);
-
-	write_lock(&key->lock);
-	key->flags |= KEY_FLAG_INSTANTIATED;
-	write_unlock(&key->lock);
+	set_bit(KEY_FLAG_INSTANTIATED, &key->flags);
 
  error_k:
 	up_read(&key_types_sem);
@@ -922,9 +904,7 @@ void key_revoke(struct key *key)
 	/* make sure no one's trying to change or use the key when we mark
 	 * it */
 	down_write(&key->sem);
-	write_lock(&key->lock);
-	key->flags |= KEY_FLAG_REVOKED;
-	write_unlock(&key->lock);
+	set_bit(KEY_FLAG_REVOKED, &key->flags);
 	up_write(&key->sem);
 
 } /* end key_revoke() */
@@ -975,24 +955,33 @@ void unregister_key_type(struct key_type *ktype)
 	/* withdraw the key type */
 	list_del_init(&ktype->link);
 
-	/* need to withdraw all keys of this type */
+	/* mark all the keys of this type dead */
 	spin_lock(&key_serial_lock);
 
 	for (_n = rb_first(&key_serial_tree); _n; _n = rb_next(_n)) {
 		key = rb_entry(_n, struct key, serial_node);
 
-		if (key->type != ktype)
-			continue;
+		if (key->type == ktype)
+			key->type = &key_type_dead;
+	}
+
+	spin_unlock(&key_serial_lock);
+
+	/* make sure everyone revalidates their keys */
+	synchronize_kernel();
+
+	/* we should now be able to destroy the payloads of all the keys of
+	 * this type with impunity */
+	spin_lock(&key_serial_lock);
 
-		write_lock(&key->lock);
-		key->type = &key_type_dead;
-		write_unlock(&key->lock);
+	for (_n = rb_first(&key_serial_tree); _n; _n = rb_next(_n)) {
+		key = rb_entry(_n, struct key, serial_node);
 
-		/* there shouldn't be anyone looking at the description or
-		 * payload now */
-		if (ktype->destroy)
-			ktype->destroy(key);
-		memset(&key->payload, 0xbd, sizeof(key->payload));
+		if (key->type == ktype) {
+			if (ktype->destroy)
+				ktype->destroy(key);
+			memset(&key->payload, 0xbd, sizeof(key->payload));
+		}
 	}
 
 	spin_unlock(&key_serial_lock);
@@ -1037,4 +1026,5 @@ void __init key_init(void)
 
 	/* link the two root keyrings together */
 	key_link(&root_session_keyring, &root_user_keyring);
+
 } /* end key_init() */
diff --git a/security/keys/keyctl.c b/security/keys/keyctl.c
index dc0011b3fac..cedb7326de2 100644
--- a/security/keys/keyctl.c
+++ b/security/keys/keyctl.c
@@ -728,7 +728,6 @@ long keyctl_chown_key(key_serial_t id, uid_t uid, gid_t gid)
 	/* make the changes with the locks held to prevent chown/chown races */
 	ret = -EACCES;
 	down_write(&key->sem);
-	write_lock(&key->lock);
 
 	if (!capable(CAP_SYS_ADMIN)) {
 		/* only the sysadmin can chown a key to some other UID */
@@ -755,7 +754,6 @@ long keyctl_chown_key(key_serial_t id, uid_t uid, gid_t gid)
 	ret = 0;
 
  no_access:
-	write_unlock(&key->lock);
 	up_write(&key->sem);
 	key_put(key);
  error:
@@ -784,26 +782,19 @@ long keyctl_setperm_key(key_serial_t id, key_perm_t perm)
 		goto error;
 	}
 
-	/* make the changes with the locks held to prevent chown/chmod
-	 * races */
+	/* make the changes with the locks held to prevent chown/chmod races */
 	ret = -EACCES;
 	down_write(&key->sem);
-	write_lock(&key->lock);
 
-	/* if we're not the sysadmin, we can only chmod a key that we
-	 * own */
-	if (!capable(CAP_SYS_ADMIN) && key->uid != current->fsuid)
-		goto no_access;
-
-	/* changing the permissions mask */
-	key->perm = perm;
-	ret = 0;
+	/* if we're not the sysadmin, we can only change a key that we own */
+	if (capable(CAP_SYS_ADMIN) || key->uid == current->fsuid) {
+		key->perm = perm;
+		ret = 0;
+	}
 
- no_access:
-	write_unlock(&key->lock);
 	up_write(&key->sem);
 	key_put(key);
- error:
+error:
 	return ret;
 
 } /* end keyctl_setperm_key() */
diff --git a/security/keys/keyring.c b/security/keys/keyring.c
index e2ab4f8e748..c9a5de19748 100644
--- a/security/keys/keyring.c
+++ b/security/keys/keyring.c
@@ -132,10 +132,17 @@ static int keyring_duplicate(struct key *keyring, const struct key *source)
 		(PAGE_SIZE - sizeof(*klist)) / sizeof(struct key);
 
 	ret = 0;
-	sklist = source->payload.subscriptions;
 
-	if (sklist && sklist->nkeys > 0) {
+	/* find out how many keys are currently linked */
+	rcu_read_lock();
+	sklist = rcu_dereference(source->payload.subscriptions);
+	max = 0;
+	if (sklist)
 		max = sklist->nkeys;
+	rcu_read_unlock();
+
+	/* allocate a new payload and stuff load with key links */
+	if (max > 0) {
 		BUG_ON(max > limit);
 
 		max = (max + 3) & ~3;
@@ -148,6 +155,10 @@ static int keyring_duplicate(struct key *keyring, const struct key *source)
 		if (!klist)
 			goto error;
 
+		/* set links */
+		rcu_read_lock();
+		sklist = rcu_dereference(source->payload.subscriptions);
+
 		klist->maxkeys = max;
 		klist->nkeys = sklist->nkeys;
 		memcpy(klist->keys,
@@ -157,7 +168,9 @@ static int keyring_duplicate(struct key *keyring, const struct key *source)
 		for (loop = klist->nkeys - 1; loop >= 0; loop--)
 			atomic_inc(&klist->keys[loop]->usage);
 
-		keyring->payload.subscriptions = klist;
+		rcu_read_unlock();
+
+		rcu_assign_pointer(keyring->payload.subscriptions, klist);
 		ret = 0;
 	}
 
@@ -192,7 +205,7 @@ static void keyring_destroy(struct key *keyring)
 		write_unlock(&keyring_name_lock);
 	}
 
-	klist = keyring->payload.subscriptions;
+	klist = rcu_dereference(keyring->payload.subscriptions);
 	if (klist) {
 		for (loop = klist->nkeys - 1; loop >= 0; loop--)
 			key_put(klist->keys[loop]);
@@ -216,17 +229,20 @@ static void keyring_describe(const struct key *keyring, struct seq_file *m)
 		seq_puts(m, "[anon]");
 	}
 
-	klist = keyring->payload.subscriptions;
+	rcu_read_lock();
+	klist = rcu_dereference(keyring->payload.subscriptions);
 	if (klist)
 		seq_printf(m, ": %u/%u", klist->nkeys, klist->maxkeys);
 	else
 		seq_puts(m, ": empty");
+	rcu_read_unlock();
 
 } /* end keyring_describe() */
 
 /*****************************************************************************/
 /*
  * read a list of key IDs from the keyring's contents
+ * - the keyring's semaphore is read-locked
  */
 static long keyring_read(const struct key *keyring,
 			 char __user *buffer, size_t buflen)
@@ -237,7 +253,7 @@ static long keyring_read(const struct key *keyring,
 	int loop, ret;
 
 	ret = 0;
-	klist = keyring->payload.subscriptions;
+	klist = rcu_dereference(keyring->payload.subscriptions);
 
 	if (klist) {
 		/* calculate how much data we could return */
@@ -320,7 +336,7 @@ struct key *keyring_search_aux(struct key *keyring,
 			       key_match_func_t match)
 {
 	struct {
-		struct key *keyring;
+		struct keyring_list *keylist;
 		int kix;
 	} stack[KEYRING_SEARCH_MAX_DEPTH];
 
@@ -328,10 +344,12 @@ struct key *keyring_search_aux(struct key *keyring,
 	struct timespec now;
 	struct key *key;
 	long err;
-	int sp, psp, kix;
+	int sp, kix;
 
 	key_check(keyring);
 
+	rcu_read_lock();
+
 	/* top keyring must have search permission to begin the search */
 	key = ERR_PTR(-EACCES);
 	if (!key_permission(keyring, KEY_SEARCH))
@@ -347,11 +365,10 @@ struct key *keyring_search_aux(struct key *keyring,
 
 	/* start processing a new keyring */
  descend:
-	read_lock(&keyring->lock);
-	if (keyring->flags & KEY_FLAG_REVOKED)
+	if (test_bit(KEY_FLAG_REVOKED, &keyring->flags))
 		goto not_this_keyring;
 
-	keylist = keyring->payload.subscriptions;
+	keylist = rcu_dereference(keyring->payload.subscriptions);
 	if (!keylist)
 		goto not_this_keyring;
 
@@ -364,7 +381,7 @@ struct key *keyring_search_aux(struct key *keyring,
 			continue;
 
 		/* skip revoked keys and expired keys */
-		if (key->flags & KEY_FLAG_REVOKED)
+		if (test_bit(KEY_FLAG_REVOKED, &key->flags))
 			continue;
 
 		if (key->expiry && now.tv_sec >= key->expiry)
@@ -379,7 +396,7 @@ struct key *keyring_search_aux(struct key *keyring,
 			continue;
 
 		/* we set a different error code if we find a negative key */
-		if (key->flags & KEY_FLAG_NEGATIVE) {
+		if (test_bit(KEY_FLAG_NEGATIVE, &key->flags)) {
 			err = -ENOKEY;
 			continue;
 		}
@@ -390,48 +407,37 @@ struct key *keyring_search_aux(struct key *keyring,
 	/* search through the keyrings nested in this one */
 	kix = 0;
  ascend:
-	while (kix < keylist->nkeys) {
+	for (; kix < keylist->nkeys; kix++) {
 		key = keylist->keys[kix];
 		if (key->type != &key_type_keyring)
-			goto next;
+			continue;
 
 		/* recursively search nested keyrings
 		 * - only search keyrings for which we have search permission
 		 */
 		if (sp >= KEYRING_SEARCH_MAX_DEPTH)
-			goto next;
+			continue;
 
 		if (!key_permission(key, KEY_SEARCH))
-			goto next;
-
-		/* evade loops in the keyring tree */
-		for (psp = 0; psp < sp; psp++)
-			if (stack[psp].keyring == keyring)
-				goto next;
+			continue;
 
 		/* stack the current position */
-		stack[sp].keyring = keyring;
+		stack[sp].keylist = keylist;
 		stack[sp].kix = kix;
 		sp++;
 
 		/* begin again with the new keyring */
 		keyring = key;
 		goto descend;
-
-	next:
-		kix++;
 	}
 
 	/* the keyring we're looking at was disqualified or didn't contain a
 	 * matching key */
  not_this_keyring:
-	read_unlock(&keyring->lock);
-
 	if (sp > 0) {
 		/* resume the processing of a keyring higher up in the tree */
 		sp--;
-		keyring = stack[sp].keyring;
-		keylist = keyring->payload.subscriptions;
+		keylist = stack[sp].keylist;
 		kix = stack[sp].kix + 1;
 		goto ascend;
 	}
@@ -442,16 +448,9 @@ struct key *keyring_search_aux(struct key *keyring,
 	/* we found a viable match */
  found:
 	atomic_inc(&key->usage);
-	read_unlock(&keyring->lock);
-
-	/* unwind the keyring stack */
-	while (sp > 0) {
-		sp--;
-		read_unlock(&stack[sp].keyring->lock);
-	}
-
 	key_check(key);
  error:
+	rcu_read_unlock();
 	return key;
 
 } /* end keyring_search_aux() */
@@ -489,7 +488,9 @@ struct key *__keyring_search_one(struct key *keyring,
 	struct key *key;
 	int loop;
 
-	klist = keyring->payload.subscriptions;
+	rcu_read_lock();
+
+	klist = rcu_dereference(keyring->payload.subscriptions);
 	if (klist) {
 		for (loop = 0; loop < klist->nkeys; loop++) {
 			key = klist->keys[loop];
@@ -497,7 +498,7 @@ struct key *__keyring_search_one(struct key *keyring,
 			if (key->type == ktype &&
 			    key->type->match(key, description) &&
 			    key_permission(key, perm) &&
-			    !(key->flags & KEY_FLAG_REVOKED)
+			    !test_bit(KEY_FLAG_REVOKED, &key->flags)
 			    )
 				goto found;
 		}
@@ -509,6 +510,7 @@ struct key *__keyring_search_one(struct key *keyring,
  found:
 	atomic_inc(&key->usage);
  error:
+	rcu_read_unlock();
 	return key;
 
 } /* end __keyring_search_one() */
@@ -540,7 +542,7 @@ struct key *find_keyring_by_name(const char *name, key_serial_t bound)
 				    &keyring_name_hash[bucket],
 				    type_data.link
 				    ) {
-			if (keyring->flags & KEY_FLAG_REVOKED)
+			if (test_bit(KEY_FLAG_REVOKED, &keyring->flags))
 				continue;
 
 			if (strcmp(keyring->description, name) != 0)
@@ -579,7 +581,7 @@ struct key *find_keyring_by_name(const char *name, key_serial_t bound)
 static int keyring_detect_cycle(struct key *A, struct key *B)
 {
 	struct {
-		struct key *subtree;
+		struct keyring_list *keylist;
 		int kix;
 	} stack[KEYRING_SEARCH_MAX_DEPTH];
 
@@ -587,20 +589,21 @@ static int keyring_detect_cycle(struct key *A, struct key *B)
 	struct key *subtree, *key;
 	int sp, kix, ret;
 
+	rcu_read_lock();
+
 	ret = -EDEADLK;
 	if (A == B)
-		goto error;
+		goto cycle_detected;
 
 	subtree = B;
 	sp = 0;
 
 	/* start processing a new keyring */
  descend:
-	read_lock(&subtree->lock);
-	if (subtree->flags & KEY_FLAG_REVOKED)
+	if (test_bit(KEY_FLAG_REVOKED, &subtree->flags))
 		goto not_this_keyring;
 
-	keylist = subtree->payload.subscriptions;
+	keylist = rcu_dereference(subtree->payload.subscriptions);
 	if (!keylist)
 		goto not_this_keyring;
 	kix = 0;
@@ -619,7 +622,7 @@ static int keyring_detect_cycle(struct key *A, struct key *B)
 				goto too_deep;
 
 			/* stack the current position */
-			stack[sp].subtree = subtree;
+			stack[sp].keylist = keylist;
 			stack[sp].kix = kix;
 			sp++;
 
@@ -632,13 +635,10 @@ static int keyring_detect_cycle(struct key *A, struct key *B)
 	/* the keyring we're looking at was disqualified or didn't contain a
 	 * matching key */
  not_this_keyring:
-	read_unlock(&subtree->lock);
-
 	if (sp > 0) {
 		/* resume the checking of a keyring higher up in the tree */
 		sp--;
-		subtree = stack[sp].subtree;
-		keylist = subtree->payload.subscriptions;
+		keylist = stack[sp].keylist;
 		kix = stack[sp].kix + 1;
 		goto ascend;
 	}
@@ -646,30 +646,36 @@ static int keyring_detect_cycle(struct key *A, struct key *B)
 	ret = 0; /* no cycles detected */
 
  error:
+	rcu_read_unlock();
 	return ret;
 
  too_deep:
 	ret = -ELOOP;
-	goto error_unwind;
+	goto error;
+
  cycle_detected:
 	ret = -EDEADLK;
- error_unwind:
-	read_unlock(&subtree->lock);
-
-	/* unwind the keyring stack */
-	while (sp > 0) {
-		sp--;
-		read_unlock(&stack[sp].subtree->lock);
-	}
-
 	goto error;
 
 } /* end keyring_detect_cycle() */
 
+/*****************************************************************************/
+/*
+ * dispose of a keyring list after the RCU grace period
+ */
+static void keyring_link_rcu_disposal(struct rcu_head *rcu)
+{
+	struct keyring_list *klist =
+		container_of(rcu, struct keyring_list, rcu);
+
+	kfree(klist);
+
+} /* end keyring_link_rcu_disposal() */
+
 /*****************************************************************************/
 /*
  * link a key into to a keyring
- * - must be called with the keyring's semaphore held
+ * - must be called with the keyring's semaphore write-locked
  */
 int __key_link(struct key *keyring, struct key *key)
 {
@@ -679,7 +685,7 @@ int __key_link(struct key *keyring, struct key *key)
 	int ret;
 
 	ret = -EKEYREVOKED;
-	if (keyring->flags & KEY_FLAG_REVOKED)
+	if (test_bit(KEY_FLAG_REVOKED, &keyring->flags))
 		goto error;
 
 	ret = -ENOTDIR;
@@ -710,9 +716,10 @@ int __key_link(struct key *keyring, struct key *key)
 		/* there's sufficient slack space to add directly */
 		atomic_inc(&key->usage);
 
-		write_lock(&keyring->lock);
-		klist->keys[klist->nkeys++] = key;
-		write_unlock(&keyring->lock);
+		klist->keys[klist->nkeys] = key;
+		smp_wmb();
+		klist->nkeys++;
+		smp_wmb();
 
 		ret = 0;
 	}
@@ -723,6 +730,8 @@ int __key_link(struct key *keyring, struct key *key)
 			max += klist->maxkeys;
 
 		ret = -ENFILE;
+		if (max > 65535)
+			goto error3;
 		size = sizeof(*klist) + sizeof(*key) * max;
 		if (size > PAGE_SIZE)
 			goto error3;
@@ -743,14 +752,13 @@ int __key_link(struct key *keyring, struct key *key)
 
 		/* add the key into the new space */
 		atomic_inc(&key->usage);
-
-		write_lock(&keyring->lock);
-		keyring->payload.subscriptions = nklist;
 		nklist->keys[nklist->nkeys++] = key;
-		write_unlock(&keyring->lock);
+
+		rcu_assign_pointer(keyring->payload.subscriptions, nklist);
 
 		/* dispose of the old keyring list */
-		kfree(klist);
+		if (klist)
+			call_rcu(&klist->rcu, keyring_link_rcu_disposal);
 
 		ret = 0;
 	}
@@ -789,13 +797,28 @@ int key_link(struct key *keyring, struct key *key)
 
 EXPORT_SYMBOL(key_link);
 
+/*****************************************************************************/
+/*
+ * dispose of a keyring list after the RCU grace period, freeing the unlinked
+ * key
+ */
+static void keyring_unlink_rcu_disposal(struct rcu_head *rcu)
+{
+	struct keyring_list *klist =
+		container_of(rcu, struct keyring_list, rcu);
+
+	key_put(klist->keys[klist->delkey]);
+	kfree(klist);
+
+} /* end keyring_unlink_rcu_disposal() */
+
 /*****************************************************************************/
 /*
  * unlink the first link to a key from a keyring
  */
 int key_unlink(struct key *keyring, struct key *key)
 {
-	struct keyring_list *klist;
+	struct keyring_list *klist, *nklist;
 	int loop, ret;
 
 	key_check(keyring);
@@ -819,36 +842,69 @@ int key_unlink(struct key *keyring, struct key *key)
 	ret = -ENOENT;
 	goto error;
 
- key_is_present:
+key_is_present:
+	/* we need to copy the key list for RCU purposes */
+	nklist = kmalloc(sizeof(*klist) + sizeof(*key) * klist->maxkeys,
+			 GFP_KERNEL);
+	if (!nklist)
+		goto nomem;
+	nklist->maxkeys = klist->maxkeys;
+	nklist->nkeys = klist->nkeys - 1;
+
+	if (loop > 0)
+		memcpy(&nklist->keys[0],
+		       &klist->keys[0],
+		       loop * sizeof(klist->keys[0]));
+
+	if (loop < nklist->nkeys)
+		memcpy(&nklist->keys[loop],
+		       &klist->keys[loop + 1],
+		       (nklist->nkeys - loop) * sizeof(klist->keys[0]));
+
 	/* adjust the user's quota */
 	key_payload_reserve(keyring,
 			    keyring->datalen - KEYQUOTA_LINK_BYTES);
 
-	/* shuffle down the key pointers
-	 * - it might be worth shrinking the allocated memory, but that runs
-	 *   the risk of ENOMEM as we would have to copy
-	 */
-	write_lock(&keyring->lock);
+	rcu_assign_pointer(keyring->payload.subscriptions, nklist);
 
-	klist->nkeys--;
-	if (loop < klist->nkeys)
-		memcpy(&klist->keys[loop],
-		       &klist->keys[loop + 1],
-		       (klist->nkeys - loop) * sizeof(struct key *));
+	up_write(&keyring->sem);
 
-	write_unlock(&keyring->lock);
+	/* schedule for later cleanup */
+	klist->delkey = loop;
+	call_rcu(&klist->rcu, keyring_unlink_rcu_disposal);
 
-	up_write(&keyring->sem);
-	key_put(key);
 	ret = 0;
 
- error:
+error:
 	return ret;
+nomem:
+	ret = -ENOMEM;
+	up_write(&keyring->sem);
+	goto error;
 
 } /* end key_unlink() */
 
 EXPORT_SYMBOL(key_unlink);
 
+/*****************************************************************************/
+/*
+ * dispose of a keyring list after the RCU grace period, releasing the keys it
+ * links to
+ */
+static void keyring_clear_rcu_disposal(struct rcu_head *rcu)
+{
+	struct keyring_list *klist;
+	int loop;
+
+	klist = container_of(rcu, struct keyring_list, rcu);
+
+	for (loop = klist->nkeys - 1; loop >= 0; loop--)
+		key_put(klist->keys[loop]);
+
+	kfree(klist);
+
+} /* end keyring_clear_rcu_disposal() */
+
 /*****************************************************************************/
 /*
  * clear the specified process keyring
@@ -857,7 +913,7 @@ EXPORT_SYMBOL(key_unlink);
 int keyring_clear(struct key *keyring)
 {
 	struct keyring_list *klist;
-	int loop, ret;
+	int ret;
 
 	ret = -ENOTDIR;
 	if (keyring->type == &key_type_keyring) {
@@ -870,20 +926,15 @@ int keyring_clear(struct key *keyring)
 			key_payload_reserve(keyring,
 					    sizeof(struct keyring_list));
 
-			write_lock(&keyring->lock);
-			keyring->payload.subscriptions = NULL;
-			write_unlock(&keyring->lock);
+			rcu_assign_pointer(keyring->payload.subscriptions,
+					   NULL);
 		}
 
 		up_write(&keyring->sem);
 
 		/* free the keys after the locks have been dropped */
-		if (klist) {
-			for (loop = klist->nkeys - 1; loop >= 0; loop--)
-				key_put(klist->keys[loop]);
-
-			kfree(klist);
-		}
+		if (klist)
+			call_rcu(&klist->rcu, keyring_clear_rcu_disposal);
 
 		ret = 0;
 	}
diff --git a/security/keys/proc.c b/security/keys/proc.c
index 91343b85c39..c55cf1fd082 100644
--- a/security/keys/proc.c
+++ b/security/keys/proc.c
@@ -140,7 +140,7 @@ static int proc_keys_show(struct seq_file *m, void *v)
 
 	now = current_kernel_time();
 
-	read_lock(&key->lock);
+	rcu_read_lock();
 
 	/* come up with a suitable timeout value */
 	if (key->expiry == 0) {
@@ -164,14 +164,17 @@ static int proc_keys_show(struct seq_file *m, void *v)
 			sprintf(xbuf, "%luw", timo / (60*60*24*7));
 	}
 
+#define showflag(KEY, LETTER, FLAG) \
+	(test_bit(FLAG,	&(KEY)->flags) ? LETTER : '-')
+
 	seq_printf(m, "%08x %c%c%c%c%c%c %5d %4s %06x %5d %5d %-9.9s ",
 		   key->serial,
-		   key->flags & KEY_FLAG_INSTANTIATED	? 'I' : '-',
-		   key->flags & KEY_FLAG_REVOKED	? 'R' : '-',
-		   key->flags & KEY_FLAG_DEAD		? 'D' : '-',
-		   key->flags & KEY_FLAG_IN_QUOTA	? 'Q' : '-',
-		   key->flags & KEY_FLAG_USER_CONSTRUCT	? 'U' : '-',
-		   key->flags & KEY_FLAG_NEGATIVE	? 'N' : '-',
+		   showflag(key, 'I', KEY_FLAG_INSTANTIATED),
+		   showflag(key, 'R', KEY_FLAG_REVOKED),
+		   showflag(key, 'D', KEY_FLAG_DEAD),
+		   showflag(key, 'Q', KEY_FLAG_IN_QUOTA),
+		   showflag(key, 'U', KEY_FLAG_USER_CONSTRUCT),
+		   showflag(key, 'N', KEY_FLAG_NEGATIVE),
 		   atomic_read(&key->usage),
 		   xbuf,
 		   key->perm,
@@ -179,11 +182,13 @@ static int proc_keys_show(struct seq_file *m, void *v)
 		   key->gid,
 		   key->type->name);
 
+#undef showflag
+
 	if (key->type->describe)
 		key->type->describe(key, m);
 	seq_putc(m, '\n');
 
-	read_unlock(&key->lock);
+	rcu_read_unlock();
 
 	return 0;
 
diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c
index 2eb0e471cd4..059c350cac4 100644
--- a/security/keys/process_keys.c
+++ b/security/keys/process_keys.c
@@ -38,10 +38,9 @@ struct key root_user_keyring = {
 	.serial		= 2,
 	.type		= &key_type_keyring,
 	.user		= &root_key_user,
-	.lock		= RW_LOCK_UNLOCKED,
 	.sem		= __RWSEM_INITIALIZER(root_user_keyring.sem),
 	.perm		= KEY_USR_ALL,
-	.flags		= KEY_FLAG_INSTANTIATED,
+	.flags		= 1 << KEY_FLAG_INSTANTIATED,
 	.description	= "_uid.0",
 #ifdef KEY_DEBUGGING
 	.magic		= KEY_DEBUG_MAGIC,
@@ -54,10 +53,9 @@ struct key root_session_keyring = {
 	.serial		= 1,
 	.type		= &key_type_keyring,
 	.user		= &root_key_user,
-	.lock		= RW_LOCK_UNLOCKED,
 	.sem		= __RWSEM_INITIALIZER(root_session_keyring.sem),
 	.perm		= KEY_USR_ALL,
-	.flags		= KEY_FLAG_INSTANTIATED,
+	.flags		= 1 << KEY_FLAG_INSTANTIATED,
 	.description	= "_uid_ses.0",
 #ifdef KEY_DEBUGGING
 	.magic		= KEY_DEBUG_MAGIC,
@@ -349,9 +347,7 @@ void key_fsuid_changed(struct task_struct *tsk)
 	/* update the ownership of the thread keyring */
 	if (tsk->thread_keyring) {
 		down_write(&tsk->thread_keyring->sem);
-		write_lock(&tsk->thread_keyring->lock);
 		tsk->thread_keyring->uid = tsk->fsuid;
-		write_unlock(&tsk->thread_keyring->lock);
 		up_write(&tsk->thread_keyring->sem);
 	}
 
@@ -366,9 +362,7 @@ void key_fsgid_changed(struct task_struct *tsk)
 	/* update the ownership of the thread keyring */
 	if (tsk->thread_keyring) {
 		down_write(&tsk->thread_keyring->sem);
-		write_lock(&tsk->thread_keyring->lock);
 		tsk->thread_keyring->gid = tsk->fsgid;
-		write_unlock(&tsk->thread_keyring->lock);
 		up_write(&tsk->thread_keyring->sem);
 	}
 
@@ -588,7 +582,7 @@ struct key *lookup_user_key(key_serial_t id, int create, int partial,
 	}
 
 	ret = -EIO;
-	if (!partial && !(key->flags & KEY_FLAG_INSTANTIATED))
+	if (!partial && !test_bit(KEY_FLAG_INSTANTIATED, &key->flags))
 		goto invalid_key;
 
 	ret = -EACCES;
diff --git a/security/keys/request_key.c b/security/keys/request_key.c
index 9705b1aeba5..1f6c0940297 100644
--- a/security/keys/request_key.c
+++ b/security/keys/request_key.c
@@ -105,7 +105,7 @@ static struct key *__request_key_construction(struct key_type *type,
 	struct key_construction cons;
 	struct timespec now;
 	struct key *key;
-	int ret, negative;
+	int ret, negated;
 
 	/* create a key and add it to the queue */
 	key = key_alloc(type, description,
@@ -113,9 +113,7 @@ static struct key *__request_key_construction(struct key_type *type,
 	if (IS_ERR(key))
 		goto alloc_failed;
 
-	write_lock(&key->lock);
-	key->flags |= KEY_FLAG_USER_CONSTRUCT;
-	write_unlock(&key->lock);
+	set_bit(KEY_FLAG_USER_CONSTRUCT, &key->flags);
 
 	cons.key = key;
 	list_add_tail(&cons.link, &key->user->consq);
@@ -130,7 +128,7 @@ static struct key *__request_key_construction(struct key_type *type,
 
 	/* if the key wasn't instantiated, then we want to give an error */
 	ret = -ENOKEY;
-	if (!(key->flags & KEY_FLAG_INSTANTIATED))
+	if (!test_bit(KEY_FLAG_INSTANTIATED, &key->flags))
 		goto request_failed;
 
 	down_write(&key_construction_sem);
@@ -139,7 +137,7 @@ static struct key *__request_key_construction(struct key_type *type,
 
 	/* also give an error if the key was negatively instantiated */
  check_not_negative:
-	if (key->flags & KEY_FLAG_NEGATIVE) {
+	if (test_bit(KEY_FLAG_NEGATIVE, &key->flags)) {
 		key_put(key);
 		key = ERR_PTR(-ENOKEY);
 	}
@@ -152,24 +150,23 @@ static struct key *__request_key_construction(struct key_type *type,
 	 * - remove from construction queue
 	 * - mark the key as dead
 	 */
-	negative = 0;
+	negated = 0;
 	down_write(&key_construction_sem);
 
 	list_del(&cons.link);
 
-	write_lock(&key->lock);
-	key->flags &= ~KEY_FLAG_USER_CONSTRUCT;
-
 	/* check it didn't get instantiated between the check and the down */
-	if (!(key->flags & KEY_FLAG_INSTANTIATED)) {
-		key->flags |= KEY_FLAG_INSTANTIATED | KEY_FLAG_NEGATIVE;
-		negative = 1;
+	if (!test_bit(KEY_FLAG_INSTANTIATED, &key->flags)) {
+		set_bit(KEY_FLAG_NEGATIVE, &key->flags);
+		set_bit(KEY_FLAG_INSTANTIATED, &key->flags);
+		negated = 1;
 	}
 
-	write_unlock(&key->lock);
+	clear_bit(KEY_FLAG_USER_CONSTRUCT, &key->flags);
+
 	up_write(&key_construction_sem);
 
-	if (!negative)
+	if (!negated)
 		goto check_not_negative; /* surprisingly, the key got
 					  * instantiated */
 
@@ -250,7 +247,7 @@ static struct key *request_key_construction(struct key_type *type,
 
 	for (;;) {
 		set_current_state(TASK_UNINTERRUPTIBLE);
-		if (!(ckey->flags & KEY_FLAG_USER_CONSTRUCT))
+		if (!test_bit(KEY_FLAG_USER_CONSTRUCT, &ckey->flags))
 			break;
 		schedule();
 	}
@@ -339,7 +336,8 @@ int key_validate(struct key *key)
 	if (key) {
 		/* check it's still accessible */
 		ret = -EKEYREVOKED;
-		if (key->flags & (KEY_FLAG_REVOKED | KEY_FLAG_DEAD))
+		if (test_bit(KEY_FLAG_REVOKED, &key->flags) ||
+		    test_bit(KEY_FLAG_DEAD, &key->flags))
 			goto error;
 
 		/* check it hasn't expired */
diff --git a/security/keys/user_defined.c b/security/keys/user_defined.c
index 8d65b3a2812..c33d3614a0d 100644
--- a/security/keys/user_defined.c
+++ b/security/keys/user_defined.c
@@ -42,12 +42,19 @@ struct key_type key_type_user = {
 	.read		= user_read,
 };
 
+struct user_key_payload {
+	struct rcu_head	rcu;		/* RCU destructor */
+	unsigned short	datalen;	/* length of this data */
+	char		data[0];	/* actual data */
+};
+
 /*****************************************************************************/
 /*
  * instantiate a user defined key
  */
 static int user_instantiate(struct key *key, const void *data, size_t datalen)
 {
+	struct user_key_payload *upayload;
 	int ret;
 
 	ret = -EINVAL;
@@ -58,13 +65,15 @@ static int user_instantiate(struct key *key, const void *data, size_t datalen)
 	if (ret < 0)
 		goto error;
 
-	/* attach the data */
 	ret = -ENOMEM;
-	key->payload.data = kmalloc(datalen, GFP_KERNEL);
-	if (!key->payload.data)
+	upayload = kmalloc(sizeof(*upayload) + datalen, GFP_KERNEL);
+	if (!upayload)
 		goto error;
 
-	memcpy(key->payload.data, data, datalen);
+	/* attach the data */
+	upayload->datalen = datalen;
+	memcpy(upayload->data, data, datalen);
+	rcu_assign_pointer(key->payload.data, upayload);
 	ret = 0;
 
  error:
@@ -75,18 +84,25 @@ static int user_instantiate(struct key *key, const void *data, size_t datalen)
 /*****************************************************************************/
 /*
  * duplicate a user defined key
+ * - both keys' semaphores are locked against further modification
+ * - the new key cannot yet be accessed
  */
 static int user_duplicate(struct key *key, const struct key *source)
 {
+	struct user_key_payload *upayload, *spayload;
 	int ret;
 
 	/* just copy the payload */
 	ret = -ENOMEM;
-	key->payload.data = kmalloc(source->datalen, GFP_KERNEL);
+	upayload = kmalloc(sizeof(*upayload) + source->datalen, GFP_KERNEL);
+	if (upayload) {
+		spayload = rcu_dereference(source->payload.data);
+		BUG_ON(source->datalen != spayload->datalen);
 
-	if (key->payload.data) {
-		key->datalen = source->datalen;
-		memcpy(key->payload.data, source->payload.data, source->datalen);
+		upayload->datalen = key->datalen = spayload->datalen;
+		memcpy(upayload->data, spayload->data, key->datalen);
+
+		key->payload.data = upayload;
 		ret = 0;
 	}
 
@@ -94,42 +110,56 @@ static int user_duplicate(struct key *key, const struct key *source)
 
 } /* end user_duplicate() */
 
+/*****************************************************************************/
+/*
+ * dispose of the old data from an updated user defined key
+ */
+static void user_update_rcu_disposal(struct rcu_head *rcu)
+{
+	struct user_key_payload *upayload;
+
+	upayload = container_of(rcu, struct user_key_payload, rcu);
+
+	kfree(upayload);
+
+} /* end user_update_rcu_disposal() */
+
 /*****************************************************************************/
 /*
  * update a user defined key
+ * - the key's semaphore is write-locked
  */
 static int user_update(struct key *key, const void *data, size_t datalen)
 {
-	void *new, *zap;
+	struct user_key_payload *upayload, *zap;
 	int ret;
 
 	ret = -EINVAL;
 	if (datalen <= 0 || datalen > 32767 || !data)
 		goto error;
 
-	/* copy the data */
+	/* construct a replacement payload */
 	ret = -ENOMEM;
-	new = kmalloc(datalen, GFP_KERNEL);
-	if (!new)
+	upayload = kmalloc(sizeof(*upayload) + datalen, GFP_KERNEL);
+	if (!upayload)
 		goto error;
 
-	memcpy(new, data, datalen);
+	upayload->datalen = datalen;
+	memcpy(upayload->data, data, datalen);
 
 	/* check the quota and attach the new data */
-	zap = new;
-	write_lock(&key->lock);
+	zap = upayload;
 
 	ret = key_payload_reserve(key, datalen);
 
 	if (ret == 0) {
 		/* attach the new data, displacing the old */
 		zap = key->payload.data;
-		key->payload.data = new;
+		rcu_assign_pointer(key->payload.data, upayload);
 		key->expiry = 0;
 	}
 
-	write_unlock(&key->lock);
-	kfree(zap);
+	call_rcu(&zap->rcu, user_update_rcu_disposal);
 
  error:
 	return ret;
@@ -152,13 +182,15 @@ static int user_match(const struct key *key, const void *description)
  */
 static void user_destroy(struct key *key)
 {
-	kfree(key->payload.data);
+	struct user_key_payload *upayload = key->payload.data;
+
+	kfree(upayload);
 
 } /* end user_destroy() */
 
 /*****************************************************************************/
 /*
- * describe the user
+ * describe the user key
  */
 static void user_describe(const struct key *key, struct seq_file *m)
 {
@@ -171,18 +203,23 @@ static void user_describe(const struct key *key, struct seq_file *m)
 /*****************************************************************************/
 /*
  * read the key data
+ * - the key's semaphore is read-locked
  */
 static long user_read(const struct key *key,
 		      char __user *buffer, size_t buflen)
 {
-	long ret = key->datalen;
+	struct user_key_payload *upayload;
+	long ret;
+
+	upayload = rcu_dereference(key->payload.data);
+	ret = upayload->datalen;
 
 	/* we can return the data as is */
 	if (buffer && buflen > 0) {
-		if (buflen > key->datalen)
-			buflen = key->datalen;
+		if (buflen > upayload->datalen)
+			buflen = upayload->datalen;
 
-		if (copy_to_user(buffer, key->payload.data, buflen) != 0)
+		if (copy_to_user(buffer, upayload->data, buflen) != 0)
 			ret = -EFAULT;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 7888e7ff4ee579442128d7d12a9c9dbf2cf7de6a Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Thu, 23 Jun 2005 22:00:51 -0700
Subject: [PATCH] Keys: Pass session keyring to call_usermodehelper()

The attached patch makes it possible to pass a session keyring through to the
process spawned by call_usermodehelper().  This allows patch 3/3 to pass an
authorisation key through to /sbin/request-key, thus permitting better access
controls when doing just-in-time key creation.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/key.h         | 10 +++++++++-
 include/linux/kmod.h        | 13 ++++++++++++-
 kernel/kmod.c               | 17 +++++++++++++----
 security/keys/request_key.c |  2 +-
 4 files changed, 35 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/key.h b/include/linux/key.h
index 2c24ffaca86..2bfbf88d274 100644
--- a/include/linux/key.h
+++ b/include/linux/key.h
@@ -273,14 +273,22 @@ extern void key_fsuid_changed(struct task_struct *tsk);
 extern void key_fsgid_changed(struct task_struct *tsk);
 extern void key_init(void);
 
+#define __install_session_keyring(tsk, keyring)			\
+({								\
+	struct key *old_session = tsk->signal->session_keyring;	\
+	tsk->signal->session_keyring = keyring;			\
+	old_session;						\
+})
+
 #else /* CONFIG_KEYS */
 
 #define key_validate(k)			0
 #define key_serial(k)			0
-#define key_get(k) 			NULL
+#define key_get(k) 			({ NULL; })
 #define key_put(k)			do { } while(0)
 #define alloc_uid_keyring(u)		0
 #define switch_uid_keyring(u)		do { } while(0)
+#define __install_session_keyring(t, k)	({ NULL; })
 #define copy_keys(f,t)			0
 #define copy_thread_group_keys(t)	0
 #define exit_keys(t)			do { } while(0)
diff --git a/include/linux/kmod.h b/include/linux/kmod.h
index 95d0e4b0814..e4a23154940 100644
--- a/include/linux/kmod.h
+++ b/include/linux/kmod.h
@@ -19,6 +19,7 @@
  *      Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
+#include <linux/stddef.h>
 #include <linux/config.h>
 #include <linux/errno.h>
 #include <linux/compiler.h>
@@ -34,7 +35,17 @@ static inline int request_module(const char * name, ...) { return -ENOSYS; }
 #endif
 
 #define try_then_request_module(x, mod...) ((x) ?: (request_module(mod), (x)))
-extern int call_usermodehelper(char *path, char *argv[], char *envp[], int wait);
+
+struct key;
+extern int call_usermodehelper_keys(char *path, char *argv[], char *envp[],
+				    struct key *session_keyring, int wait);
+
+static inline int
+call_usermodehelper(char *path, char **argv, char **envp, int wait)
+{
+	return call_usermodehelper_keys(path, argv, envp, NULL, wait);
+}
+
 extern void usermodehelper_init(void);
 
 #endif /* __LINUX_KMOD_H__ */
diff --git a/kernel/kmod.c b/kernel/kmod.c
index eed53d4f523..44166e3bb8a 100644
--- a/kernel/kmod.c
+++ b/kernel/kmod.c
@@ -120,6 +120,7 @@ struct subprocess_info {
 	char *path;
 	char **argv;
 	char **envp;
+	struct key *ring;
 	int wait;
 	int retval;
 };
@@ -130,16 +131,21 @@ struct subprocess_info {
 static int ____call_usermodehelper(void *data)
 {
 	struct subprocess_info *sub_info = data;
+	struct key *old_session;
 	int retval;
 
-	/* Unblock all signals. */
+	/* Unblock all signals and set the session keyring. */
+	key_get(sub_info->ring);
 	flush_signals(current);
 	spin_lock_irq(&current->sighand->siglock);
+	old_session = __install_session_keyring(current, sub_info->ring);
 	flush_signal_handlers(current, 1);
 	sigemptyset(&current->blocked);
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 
+	key_put(old_session);
+
 	/* We can run anywhere, unlike our parent keventd(). */
 	set_cpus_allowed(current, CPU_MASK_ALL);
 
@@ -211,10 +217,11 @@ static void __call_usermodehelper(void *data)
 }
 
 /**
- * call_usermodehelper - start a usermode application
+ * call_usermodehelper_keys - start a usermode application
  * @path: pathname for the application
  * @argv: null-terminated argument list
  * @envp: null-terminated environment list
+ * @session_keyring: session keyring for process (NULL for an empty keyring)
  * @wait: wait for the application to finish and return status.
  *
  * Runs a user-space application.  The application is started
@@ -224,7 +231,8 @@ static void __call_usermodehelper(void *data)
  * Must be called from process context.  Returns a negative error code
  * if program was not execed successfully, or 0.
  */
-int call_usermodehelper(char *path, char **argv, char **envp, int wait)
+int call_usermodehelper_keys(char *path, char **argv, char **envp,
+			     struct key *session_keyring, int wait)
 {
 	DECLARE_COMPLETION(done);
 	struct subprocess_info sub_info = {
@@ -232,6 +240,7 @@ int call_usermodehelper(char *path, char **argv, char **envp, int wait)
 		.path		= path,
 		.argv		= argv,
 		.envp		= envp,
+		.ring		= session_keyring,
 		.wait		= wait,
 		.retval		= 0,
 	};
@@ -247,7 +256,7 @@ int call_usermodehelper(char *path, char **argv, char **envp, int wait)
 	wait_for_completion(&done);
 	return sub_info.retval;
 }
-EXPORT_SYMBOL(call_usermodehelper);
+EXPORT_SYMBOL(call_usermodehelper_keys);
 
 void __init usermodehelper_init(void)
 {
diff --git a/security/keys/request_key.c b/security/keys/request_key.c
index 1f6c0940297..1919540f047 100644
--- a/security/keys/request_key.c
+++ b/security/keys/request_key.c
@@ -88,7 +88,7 @@ static int call_request_key(struct key *key,
 	argv[i] = NULL;
 
 	/* do it */
-	return call_usermodehelper(argv[0], argv, envp, 1);
+	return call_usermodehelper_keys(argv[0], argv, envp, NULL, 1);
 
 } /* end call_request_key() */
 
-- 
cgit v1.2.3-70-g09d2


From 3e30148c3d524a9c1c63ca28261bc24c457eb07a Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Thu, 23 Jun 2005 22:00:56 -0700
Subject: [PATCH] Keys: Make request-key create an authorisation key

The attached patch makes the following changes:

 (1) There's a new special key type called ".request_key_auth".

     This is an authorisation key for when one process requests a key and
     another process is started to construct it. This type of key cannot be
     created by the user; nor can it be requested by kernel services.

     Authorisation keys hold two references:

     (a) Each refers to a key being constructed. When the key being
     	 constructed is instantiated the authorisation key is revoked,
     	 rendering it of no further use.

     (b) The "authorising process". This is either:

     	 (i) the process that called request_key(), or:

     	 (ii) if the process that called request_key() itself had an
     	      authorisation key in its session keyring, then the authorising
     	      process referred to by that authorisation key will also be
     	      referred to by the new authorisation key.

	 This means that the process that initiated a chain of key requests
	 will authorise the lot of them, and will, by default, wind up with
	 the keys obtained from them in its keyrings.

 (2) request_key() creates an authorisation key which is then passed to
     /sbin/request-key in as part of a new session keyring.

 (3) When request_key() is searching for a key to hand back to the caller, if
     it comes across an authorisation key in the session keyring of the
     calling process, it will also search the keyrings of the process
     specified therein and it will use the specified process's credentials
     (fsuid, fsgid, groups) to do that rather than the calling process's
     credentials.

     This allows a process started by /sbin/request-key to find keys belonging
     to the authorising process.

 (4) A key can be read, even if the process executing KEYCTL_READ doesn't have
     direct read or search permission if that key is contained within the
     keyrings of a process specified by an authorisation key found within the
     calling process's session keyring, and is searchable using the
     credentials of the authorising process.

     This allows a process started by /sbin/request-key to read keys belonging
     to the authorising process.

 (5) The magic KEY_SPEC_*_KEYRING key IDs when passed to KEYCTL_INSTANTIATE or
     KEYCTL_NEGATE will specify a keyring of the authorising process, rather
     than the process doing the instantiation.

 (6) One of the process keyrings can be nominated as the default to which
     request_key() should attach new keys if not otherwise specified. This is
     done with KEYCTL_SET_REQKEY_KEYRING and one of the KEY_REQKEY_DEFL_*
     constants. The current setting can also be read using this call.

 (7) request_key() is partially interruptible. If it is waiting for another
     process to finish constructing a key, it can be interrupted. This permits
     a request-key cycle to be broken without recourse to rebooting.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-Off-By: Benoit Boissinot <benoit.boissinot@ens-lyon.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/keys.txt           |  34 ++++++++
 include/linux/key-ui.h           |  41 ++++++++-
 include/linux/key.h              |   9 +-
 include/linux/keyctl.h           |  11 +++
 include/linux/sched.h            |   8 +-
 kernel/sys.c                     |   2 +-
 security/keys/Makefile           |   5 +-
 security/keys/compat.c           |   7 +-
 security/keys/internal.h         |  45 +++++++++-
 security/keys/key.c              |  24 ++++--
 security/keys/keyctl.c           | 176 ++++++++++++++++++++++++-------------
 security/keys/keyring.c          |  67 ++++++++++++--
 security/keys/process_keys.c     | 179 +++++++++++++++++++++++---------------
 security/keys/request_key.c      | 182 ++++++++++++++++++++++++++++++++-------
 security/keys/request_key_auth.c | 180 ++++++++++++++++++++++++++++++++++++++
 15 files changed, 779 insertions(+), 191 deletions(-)
 create mode 100644 security/keys/request_key_auth.c

(limited to 'include')

diff --git a/Documentation/keys.txt b/Documentation/keys.txt
index 3df40c1fe15..0321ded4b9a 100644
--- a/Documentation/keys.txt
+++ b/Documentation/keys.txt
@@ -591,6 +591,37 @@ The keyctl syscall functions are:
      this case too.
 
 
+ (*) Set the default request-key destination keyring.
+
+	long keyctl(KEYCTL_SET_REQKEY_KEYRING, int reqkey_defl);
+
+     This sets the default keyring to which implicitly requested keys will be
+     attached for this thread. reqkey_defl should be one of these constants:
+
+	CONSTANT				VALUE	NEW DEFAULT KEYRING
+	======================================	======	=======================
+	KEY_REQKEY_DEFL_NO_CHANGE		-1	No change
+	KEY_REQKEY_DEFL_DEFAULT			0	Default[1]
+	KEY_REQKEY_DEFL_THREAD_KEYRING		1	Thread keyring
+	KEY_REQKEY_DEFL_PROCESS_KEYRING		2	Process keyring
+	KEY_REQKEY_DEFL_SESSION_KEYRING		3	Session keyring
+	KEY_REQKEY_DEFL_USER_KEYRING		4	User keyring
+	KEY_REQKEY_DEFL_USER_SESSION_KEYRING	5	User session keyring
+	KEY_REQKEY_DEFL_GROUP_KEYRING		6	Group keyring
+
+     The old default will be returned if successful and error EINVAL will be
+     returned if reqkey_defl is not one of the above values.
+
+     The default keyring can be overridden by the keyring indicated to the
+     request_key() system call.
+
+     Note that this setting is inherited across fork/exec.
+
+     [1] The default default is: the thread keyring if there is one, otherwise
+     the process keyring if there is one, otherwise the session keyring if
+     there is one, otherwise the user default session keyring.
+
+
 ===============
 KERNEL SERVICES
 ===============
@@ -626,6 +657,9 @@ payload contents" for more information.
     Should the function fail error ENOKEY, EKEYEXPIRED or EKEYREVOKED will be
     returned.
 
+    If successful, the key will have been attached to the default keyring for
+    implicitly obtained request-key keys, as set by KEYCTL_SET_REQKEY_KEYRING.
+
 
 (*) When it is no longer required, the key should be released using:
 
diff --git a/include/linux/key-ui.h b/include/linux/key-ui.h
index 159ca8d54e9..cc326174a80 100644
--- a/include/linux/key-ui.h
+++ b/include/linux/key-ui.h
@@ -1,4 +1,4 @@
-/* key-ui.h: key userspace interface stuff for use by keyfs
+/* key-ui.h: key userspace interface stuff
  *
  * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
@@ -84,8 +84,45 @@ static inline int key_any_permission(const struct key *key, key_perm_t perm)
 	return kperm != 0;
 }
 
+static inline int key_task_groups_search(struct task_struct *tsk, gid_t gid)
+{
+	int ret;
+
+	task_lock(tsk);
+	ret = groups_search(tsk->group_info, gid);
+	task_unlock(tsk);
+	return ret;
+}
+
+static inline int key_task_permission(const struct key *key,
+				      struct task_struct *context,
+				      key_perm_t perm)
+{
+	key_perm_t kperm;
+
+	if (key->uid == context->fsuid) {
+		kperm = key->perm >> 16;
+	}
+	else if (key->gid != -1 &&
+		 key->perm & KEY_GRP_ALL && (
+			 key->gid == context->fsgid ||
+			 key_task_groups_search(context, key->gid)
+			 )
+		 ) {
+		kperm = key->perm >> 8;
+	}
+	else {
+		kperm = key->perm;
+	}
+
+	kperm = kperm & perm & KEY_ALL;
+
+	return kperm == perm;
+
+}
 
-extern struct key *lookup_user_key(key_serial_t id, int create, int part,
+extern struct key *lookup_user_key(struct task_struct *context,
+				   key_serial_t id, int create, int partial,
 				   key_perm_t perm);
 
 extern long join_session_keyring(const char *name);
diff --git a/include/linux/key.h b/include/linux/key.h
index 2bfbf88d274..970bbd916cf 100644
--- a/include/linux/key.h
+++ b/include/linux/key.h
@@ -199,10 +199,12 @@ extern int key_payload_reserve(struct key *key, size_t datalen);
 extern int key_instantiate_and_link(struct key *key,
 				    const void *data,
 				    size_t datalen,
-				    struct key *keyring);
+				    struct key *keyring,
+				    struct key *instkey);
 extern int key_negate_and_link(struct key *key,
 			       unsigned timeout,
-			       struct key *keyring);
+			       struct key *keyring,
+			       struct key *instkey);
 extern void key_revoke(struct key *key);
 extern void key_put(struct key *key);
 
@@ -245,9 +247,6 @@ extern struct key *keyring_search(struct key *keyring,
 				  struct key_type *type,
 				  const char *description);
 
-extern struct key *search_process_keyrings(struct key_type *type,
-					   const char *description);
-
 extern int keyring_add_key(struct key *keyring,
 			   struct key *key);
 
diff --git a/include/linux/keyctl.h b/include/linux/keyctl.h
index 381dedc370a..8d7c59a29e0 100644
--- a/include/linux/keyctl.h
+++ b/include/linux/keyctl.h
@@ -20,6 +20,16 @@
 #define KEY_SPEC_USER_SESSION_KEYRING	-5	/* - key ID for UID-session keyring */
 #define KEY_SPEC_GROUP_KEYRING		-6	/* - key ID for GID-specific keyring */
 
+/* request-key default keyrings */
+#define KEY_REQKEY_DEFL_NO_CHANGE		-1
+#define KEY_REQKEY_DEFL_DEFAULT			0
+#define KEY_REQKEY_DEFL_THREAD_KEYRING		1
+#define KEY_REQKEY_DEFL_PROCESS_KEYRING		2
+#define KEY_REQKEY_DEFL_SESSION_KEYRING		3
+#define KEY_REQKEY_DEFL_USER_KEYRING		4
+#define KEY_REQKEY_DEFL_USER_SESSION_KEYRING	5
+#define KEY_REQKEY_DEFL_GROUP_KEYRING		6
+
 /* keyctl commands */
 #define KEYCTL_GET_KEYRING_ID		0	/* ask for a keyring's ID */
 #define KEYCTL_JOIN_SESSION_KEYRING	1	/* join or start named session keyring */
@@ -35,5 +45,6 @@
 #define KEYCTL_READ			11	/* read a key or keyring's contents */
 #define KEYCTL_INSTANTIATE		12	/* instantiate a partially constructed key */
 #define KEYCTL_NEGATE			13	/* negate a partially constructed key */
+#define KEYCTL_SET_REQKEY_KEYRING	14	/* set default request-key keyring */
 
 #endif /*  _LINUX_KEYCTL_H */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 901742f9238..2c69682b044 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -561,9 +561,10 @@ struct group_info {
 		groups_free(group_info); \
 } while (0)
 
-struct group_info *groups_alloc(int gidsetsize);
-void groups_free(struct group_info *group_info);
-int set_current_groups(struct group_info *group_info);
+extern struct group_info *groups_alloc(int gidsetsize);
+extern void groups_free(struct group_info *group_info);
+extern int set_current_groups(struct group_info *group_info);
+extern int groups_search(struct group_info *group_info, gid_t grp);
 /* access the groups "array" with this macro */
 #define GROUP_AT(gi, i) \
     ((gi)->blocks[(i)/NGROUPS_PER_BLOCK][(i)%NGROUPS_PER_BLOCK])
@@ -660,6 +661,7 @@ struct task_struct {
 	struct user_struct *user;
 #ifdef CONFIG_KEYS
 	struct key *thread_keyring;	/* keyring private to this thread */
+	unsigned char jit_keyring;	/* default keyring to attach requested keys to */
 #endif
 	int oomkilladj; /* OOM kill score adjustment (bit shift). */
 	char comm[TASK_COMM_LEN]; /* executable name excluding path
diff --git a/kernel/sys.c b/kernel/sys.c
index 5a9d6b07501..da24bc1292d 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -1259,7 +1259,7 @@ static void groups_sort(struct group_info *group_info)
 }
 
 /* a simple bsearch */
-static int groups_search(struct group_info *group_info, gid_t grp)
+int groups_search(struct group_info *group_info, gid_t grp)
 {
 	int left, right;
 
diff --git a/security/keys/Makefile b/security/keys/Makefile
index ddb495d6506..c392d750b20 100644
--- a/security/keys/Makefile
+++ b/security/keys/Makefile
@@ -7,8 +7,9 @@ obj-y := \
 	keyring.o \
 	keyctl.o \
 	process_keys.o \
-	user_defined.o \
-	request_key.o
+	request_key.o \
+	request_key_auth.o \
+	user_defined.o
 
 obj-$(CONFIG_KEYS_COMPAT) += compat.o
 obj-$(CONFIG_PROC_FS) += proc.o
diff --git a/security/keys/compat.c b/security/keys/compat.c
index aff8b22dcb5..3303673c636 100644
--- a/security/keys/compat.c
+++ b/security/keys/compat.c
@@ -1,6 +1,6 @@
 /* compat.c: 32-bit compatibility syscall for 64-bit systems
  *
- * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
+ * Copyright (C) 2004-5 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
  * This program is free software; you can redistribute it and/or
@@ -24,7 +24,7 @@
  * - if you can, you should call sys_keyctl directly
  */
 asmlinkage long compat_sys_keyctl(u32 option,
-			      u32 arg2, u32 arg3, u32 arg4, u32 arg5)
+				  u32 arg2, u32 arg3, u32 arg4, u32 arg5)
 {
 	switch (option) {
 	case KEYCTL_GET_KEYRING_ID:
@@ -71,6 +71,9 @@ asmlinkage long compat_sys_keyctl(u32 option,
 	case KEYCTL_NEGATE:
 		return keyctl_negate_key(arg2, arg3, arg4);
 
+	case KEYCTL_SET_REQKEY_KEYRING:
+		return keyctl_set_reqkey_keyring(arg2);
+
 	default:
 		return -EOPNOTSUPP;
 	}
diff --git a/security/keys/internal.h b/security/keys/internal.h
index 67b2b93a748..46c8602661c 100644
--- a/security/keys/internal.h
+++ b/security/keys/internal.h
@@ -1,6 +1,6 @@
 /* internal.h: authentication token and access key management internal defs
  *
- * Copyright (C) 2003 Red Hat, Inc. All Rights Reserved.
+ * Copyright (C) 2003-5 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
  * This program is free software; you can redistribute it and/or
@@ -15,6 +15,16 @@
 #include <linux/key.h>
 #include <linux/key-ui.h>
 
+#if 0
+#define kenter(FMT, a...)	printk("==> %s("FMT")\n",__FUNCTION__ , ## a)
+#define kleave(FMT, a...)	printk("<== %s()"FMT"\n",__FUNCTION__ , ## a)
+#define kdebug(FMT, a...)	printk(FMT"\n" , ## a)
+#else
+#define kenter(FMT, a...)	do {} while(0)
+#define kleave(FMT, a...)	do {} while(0)
+#define kdebug(FMT, a...)	do {} while(0)
+#endif
+
 extern struct key_type key_type_dead;
 extern struct key_type key_type_user;
 
@@ -66,20 +76,46 @@ extern struct key *__keyring_search_one(struct key *keyring,
 					const char *description,
 					key_perm_t perm);
 
+extern struct key *keyring_search_instkey(struct key *keyring,
+					  key_serial_t target_id);
+
 typedef int (*key_match_func_t)(const struct key *, const void *);
 
 extern struct key *keyring_search_aux(struct key *keyring,
+				      struct task_struct *tsk,
 				      struct key_type *type,
 				      const void *description,
 				      key_match_func_t match);
 
-extern struct key *search_process_keyrings_aux(struct key_type *type,
-					       const void *description,
-					       key_match_func_t match);
+extern struct key *search_process_keyrings(struct key_type *type,
+					   const void *description,
+					   key_match_func_t match,
+					   struct task_struct *tsk);
 
 extern struct key *find_keyring_by_name(const char *name, key_serial_t bound);
 
 extern int install_thread_keyring(struct task_struct *tsk);
+extern int install_process_keyring(struct task_struct *tsk);
+
+extern struct key *request_key_and_link(struct key_type *type,
+					const char *description,
+					const char *callout_info,
+					struct key *dest_keyring);
+
+/*
+ * request_key authorisation
+ */
+struct request_key_auth {
+	struct key		*target_key;
+	struct task_struct	*context;
+	pid_t			pid;
+};
+
+extern struct key_type key_type_request_key_auth;
+extern struct key *request_key_auth_new(struct key *target,
+					struct key **_rkakey);
+
+extern struct key *key_get_instantiation_authkey(key_serial_t target_id);
 
 /*
  * keyctl functions
@@ -100,6 +136,7 @@ extern long keyctl_setperm_key(key_serial_t, key_perm_t);
 extern long keyctl_instantiate_key(key_serial_t, const void __user *,
 				   size_t, key_serial_t);
 extern long keyctl_negate_key(key_serial_t, unsigned, key_serial_t);
+extern long keyctl_set_reqkey_keyring(int);
 
 
 /*
diff --git a/security/keys/key.c b/security/keys/key.c
index 1fdfccb3fe4..3304d37bb37 100644
--- a/security/keys/key.c
+++ b/security/keys/key.c
@@ -1,6 +1,6 @@
 /* key.c: basic authentication token and access key management
  *
- * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
+ * Copyright (C) 2004-5 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
  * This program is free software; you can redistribute it and/or
@@ -391,7 +391,8 @@ EXPORT_SYMBOL(key_payload_reserve);
 static int __key_instantiate_and_link(struct key *key,
 				      const void *data,
 				      size_t datalen,
-				      struct key *keyring)
+				      struct key *keyring,
+				      struct key *instkey)
 {
 	int ret, awaken;
 
@@ -419,6 +420,10 @@ static int __key_instantiate_and_link(struct key *key,
 			/* and link it into the destination keyring */
 			if (keyring)
 				ret = __key_link(keyring, key);
+
+			/* disable the authorisation key */
+			if (instkey)
+				key_revoke(instkey);
 		}
 	}
 
@@ -439,19 +444,21 @@ static int __key_instantiate_and_link(struct key *key,
 int key_instantiate_and_link(struct key *key,
 			     const void *data,
 			     size_t datalen,
-			     struct key *keyring)
+			     struct key *keyring,
+			     struct key *instkey)
 {
 	int ret;
 
 	if (keyring)
 		down_write(&keyring->sem);
 
-	ret = __key_instantiate_and_link(key, data, datalen, keyring);
+	ret = __key_instantiate_and_link(key, data, datalen, keyring, instkey);
 
 	if (keyring)
 		up_write(&keyring->sem);
 
 	return ret;
+
 } /* end key_instantiate_and_link() */
 
 EXPORT_SYMBOL(key_instantiate_and_link);
@@ -462,7 +469,8 @@ EXPORT_SYMBOL(key_instantiate_and_link);
  */
 int key_negate_and_link(struct key *key,
 			unsigned timeout,
-			struct key *keyring)
+			struct key *keyring,
+			struct key *instkey)
 {
 	struct timespec now;
 	int ret, awaken;
@@ -495,6 +503,10 @@ int key_negate_and_link(struct key *key,
 		/* and link it into the destination keyring */
 		if (keyring)
 			ret = __key_link(keyring, key);
+
+		/* disable the authorisation key */
+		if (instkey)
+			key_revoke(instkey);
 	}
 
 	up_write(&key_construction_sem);
@@ -781,7 +793,7 @@ struct key *key_create_or_update(struct key *keyring,
 	}
 
 	/* instantiate it and link it into the target keyring */
-	ret = __key_instantiate_and_link(key, payload, plen, keyring);
+	ret = __key_instantiate_and_link(key, payload, plen, keyring, NULL);
 	if (ret < 0) {
 		key_put(key);
 		key = ERR_PTR(ret);
diff --git a/security/keys/keyctl.c b/security/keys/keyctl.c
index cedb7326de2..fea262860ea 100644
--- a/security/keys/keyctl.c
+++ b/security/keys/keyctl.c
@@ -1,6 +1,6 @@
 /* keyctl.c: userspace keyctl operations
  *
- * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
+ * Copyright (C) 2004-5 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
  * This program is free software; you can redistribute it and/or
@@ -49,6 +49,13 @@ asmlinkage long sys_add_key(const char __user *_type,
 		goto error;
 	type[31] = '\0';
 
+	if (!type[0])
+		goto error;
+
+	ret = -EPERM;
+	if (type[0] == '.')
+		goto error;
+
 	ret = -EFAULT;
 	dlen = strnlen_user(_description, PAGE_SIZE - 1);
 	if (dlen <= 0)
@@ -82,7 +89,7 @@ asmlinkage long sys_add_key(const char __user *_type,
 	}
 
 	/* find the target keyring (which must be writable) */
-	keyring = lookup_user_key(ringid, 1, 0, KEY_WRITE);
+	keyring = lookup_user_key(NULL, ringid, 1, 0, KEY_WRITE);
 	if (IS_ERR(keyring)) {
 		ret = PTR_ERR(keyring);
 		goto error3;
@@ -181,7 +188,7 @@ asmlinkage long sys_request_key(const char __user *_type,
 	/* get the destination keyring if specified */
 	dest = NULL;
 	if (destringid) {
-		dest = lookup_user_key(destringid, 1, 0, KEY_WRITE);
+		dest = lookup_user_key(NULL, destringid, 1, 0, KEY_WRITE);
 		if (IS_ERR(dest)) {
 			ret = PTR_ERR(dest);
 			goto error3;
@@ -196,23 +203,15 @@ asmlinkage long sys_request_key(const char __user *_type,
 	}
 
 	/* do the search */
-	key = request_key(ktype, description, callout_info);
+	key = request_key_and_link(ktype, description, callout_info, dest);
 	if (IS_ERR(key)) {
 		ret = PTR_ERR(key);
 		goto error5;
 	}
 
-	/* link the resulting key to the destination keyring */
-	if (dest) {
-		ret = key_link(dest, key);
-		if (ret < 0)
-			goto error6;
-	}
-
 	ret = key->serial;
 
- error6:
-	key_put(key);
+ 	key_put(key);
  error5:
 	key_type_put(ktype);
  error4:
@@ -237,7 +236,7 @@ long keyctl_get_keyring_ID(key_serial_t id, int create)
 	struct key *key;
 	long ret;
 
-	key = lookup_user_key(id, create, 0, KEY_SEARCH);
+	key = lookup_user_key(NULL, id, create, 0, KEY_SEARCH);
 	if (IS_ERR(key)) {
 		ret = PTR_ERR(key);
 		goto error;
@@ -324,7 +323,7 @@ long keyctl_update_key(key_serial_t id,
 	}
 
 	/* find the target key (which must be writable) */
-	key = lookup_user_key(id, 0, 0, KEY_WRITE);
+	key = lookup_user_key(NULL, id, 0, 0, KEY_WRITE);
 	if (IS_ERR(key)) {
 		ret = PTR_ERR(key);
 		goto error2;
@@ -352,7 +351,7 @@ long keyctl_revoke_key(key_serial_t id)
 	struct key *key;
 	long ret;
 
-	key = lookup_user_key(id, 0, 0, KEY_WRITE);
+	key = lookup_user_key(NULL, id, 0, 0, KEY_WRITE);
 	if (IS_ERR(key)) {
 		ret = PTR_ERR(key);
 		goto error;
@@ -378,7 +377,7 @@ long keyctl_keyring_clear(key_serial_t ringid)
 	struct key *keyring;
 	long ret;
 
-	keyring = lookup_user_key(ringid, 1, 0, KEY_WRITE);
+	keyring = lookup_user_key(NULL, ringid, 1, 0, KEY_WRITE);
 	if (IS_ERR(keyring)) {
 		ret = PTR_ERR(keyring);
 		goto error;
@@ -404,13 +403,13 @@ long keyctl_keyring_link(key_serial_t id, key_serial_t ringid)
 	struct key *keyring, *key;
 	long ret;
 
-	keyring = lookup_user_key(ringid, 1, 0, KEY_WRITE);
+	keyring = lookup_user_key(NULL, ringid, 1, 0, KEY_WRITE);
 	if (IS_ERR(keyring)) {
 		ret = PTR_ERR(keyring);
 		goto error;
 	}
 
-	key = lookup_user_key(id, 1, 0, KEY_LINK);
+	key = lookup_user_key(NULL, id, 1, 0, KEY_LINK);
 	if (IS_ERR(key)) {
 		ret = PTR_ERR(key);
 		goto error2;
@@ -438,13 +437,13 @@ long keyctl_keyring_unlink(key_serial_t id, key_serial_t ringid)
 	struct key *keyring, *key;
 	long ret;
 
-	keyring = lookup_user_key(ringid, 0, 0, KEY_WRITE);
+	keyring = lookup_user_key(NULL, ringid, 0, 0, KEY_WRITE);
 	if (IS_ERR(keyring)) {
 		ret = PTR_ERR(keyring);
 		goto error;
 	}
 
-	key = lookup_user_key(id, 0, 0, 0);
+	key = lookup_user_key(NULL, id, 0, 0, 0);
 	if (IS_ERR(key)) {
 		ret = PTR_ERR(key);
 		goto error2;
@@ -475,16 +474,29 @@ long keyctl_describe_key(key_serial_t keyid,
 			 char __user *buffer,
 			 size_t buflen)
 {
-	struct key *key;
+	struct key *key, *instkey;
 	char *tmpbuf;
 	long ret;
 
-	key = lookup_user_key(keyid, 0, 1, KEY_VIEW);
+	key = lookup_user_key(NULL, keyid, 0, 1, KEY_VIEW);
 	if (IS_ERR(key)) {
+		/* viewing a key under construction is permitted if we have the
+		 * authorisation token handy */
+		if (PTR_ERR(key) == -EACCES) {
+			instkey = key_get_instantiation_authkey(keyid);
+			if (!IS_ERR(instkey)) {
+				key_put(instkey);
+				key = lookup_user_key(NULL, keyid, 0, 1, 0);
+				if (!IS_ERR(key))
+					goto okay;
+			}
+		}
+
 		ret = PTR_ERR(key);
 		goto error;
 	}
 
+okay:
 	/* calculate how much description we're going to return */
 	ret = -ENOMEM;
 	tmpbuf = kmalloc(PAGE_SIZE, GFP_KERNEL);
@@ -568,7 +580,7 @@ long keyctl_keyring_search(key_serial_t ringid,
 		goto error2;
 
 	/* get the keyring at which to begin the search */
-	keyring = lookup_user_key(ringid, 0, 0, KEY_SEARCH);
+	keyring = lookup_user_key(NULL, ringid, 0, 0, KEY_SEARCH);
 	if (IS_ERR(keyring)) {
 		ret = PTR_ERR(keyring);
 		goto error2;
@@ -577,7 +589,7 @@ long keyctl_keyring_search(key_serial_t ringid,
 	/* get the destination keyring if specified */
 	dest = NULL;
 	if (destringid) {
-		dest = lookup_user_key(destringid, 1, 0, KEY_WRITE);
+		dest = lookup_user_key(NULL, destringid, 1, 0, KEY_WRITE);
 		if (IS_ERR(dest)) {
 			ret = PTR_ERR(dest);
 			goto error3;
@@ -656,24 +668,23 @@ long keyctl_read_key(key_serial_t keyid, char __user *buffer, size_t buflen)
 	long ret;
 
 	/* find the key first */
-	key = lookup_user_key(keyid, 0, 0, 0);
+	key = lookup_user_key(NULL, keyid, 0, 0, 0);
 	if (!IS_ERR(key)) {
 		/* see if we can read it directly */
 		if (key_permission(key, KEY_READ))
 			goto can_read_key;
 
-		/* can't; see if it's searchable from this process's
-		 * keyrings */
-		ret = -ENOKEY;
-		if (key_permission(key, KEY_SEARCH)) {
-			/* okay - we do have search permission on the key
-			 * itself, but do we have the key? */
-			skey = search_process_keyrings_aux(key->type, key,
-							   keyctl_read_key_same);
-			if (!IS_ERR(skey))
-				goto can_read_key2;
-		}
-
+		/* we can't; see if it's searchable from this process's
+		 * keyrings
+		 * - we automatically take account of the fact that it may be
+		 *   dangling off an instantiation key
+		 */
+		skey = search_process_keyrings(key->type, key,
+					       keyctl_read_key_same, current);
+		if (!IS_ERR(skey))
+			goto can_read_key2;
+
+		ret = PTR_ERR(skey);
 		goto error2;
 	}
 
@@ -719,7 +730,7 @@ long keyctl_chown_key(key_serial_t id, uid_t uid, gid_t gid)
 	if (uid == (uid_t) -1 && gid == (gid_t) -1)
 		goto error;
 
-	key = lookup_user_key(id, 1, 1, 0);
+	key = lookup_user_key(NULL, id, 1, 1, 0);
 	if (IS_ERR(key)) {
 		ret = PTR_ERR(key);
 		goto error;
@@ -776,7 +787,7 @@ long keyctl_setperm_key(key_serial_t id, key_perm_t perm)
 	if (perm & ~(KEY_USR_ALL | KEY_GRP_ALL | KEY_OTH_ALL))
 		goto error;
 
-	key = lookup_user_key(id, 1, 1, 0);
+	key = lookup_user_key(NULL, id, 1, 1, 0);
 	if (IS_ERR(key)) {
 		ret = PTR_ERR(key);
 		goto error;
@@ -809,7 +820,8 @@ long keyctl_instantiate_key(key_serial_t id,
 			    size_t plen,
 			    key_serial_t ringid)
 {
-	struct key *key, *keyring;
+	struct request_key_auth *rka;
+	struct key *instkey, *keyring;
 	void *payload;
 	long ret;
 
@@ -831,18 +843,21 @@ long keyctl_instantiate_key(key_serial_t id,
 			goto error2;
 	}
 
-	/* find the target key (which must be writable) */
-	key = lookup_user_key(id, 0, 1, KEY_WRITE);
-	if (IS_ERR(key)) {
-		ret = PTR_ERR(key);
+	/* find the instantiation authorisation key */
+	instkey = key_get_instantiation_authkey(id);
+	if (IS_ERR(instkey)) {
+		ret = PTR_ERR(instkey);
 		goto error2;
 	}
 
-	/* find the destination keyring if present (which must also be
-	 * writable) */
+	rka = instkey->payload.data;
+
+	/* find the destination keyring amongst those belonging to the
+	 * requesting task */
 	keyring = NULL;
 	if (ringid) {
-		keyring = lookup_user_key(ringid, 1, 0, KEY_WRITE);
+		keyring = lookup_user_key(rka->context, ringid, 1, 0,
+					  KEY_WRITE);
 		if (IS_ERR(keyring)) {
 			ret = PTR_ERR(keyring);
 			goto error3;
@@ -850,11 +865,12 @@ long keyctl_instantiate_key(key_serial_t id,
 	}
 
 	/* instantiate the key and link it into a keyring */
-	ret = key_instantiate_and_link(key, payload, plen, keyring);
+	ret = key_instantiate_and_link(rka->target_key, payload, plen,
+				       keyring, instkey);
 
 	key_put(keyring);
  error3:
-	key_put(key);
+	key_put(instkey);
  error2:
 	kfree(payload);
  error:
@@ -869,21 +885,24 @@ long keyctl_instantiate_key(key_serial_t id,
  */
 long keyctl_negate_key(key_serial_t id, unsigned timeout, key_serial_t ringid)
 {
-	struct key *key, *keyring;
+	struct request_key_auth *rka;
+	struct key *instkey, *keyring;
 	long ret;
 
-	/* find the target key (which must be writable) */
-	key = lookup_user_key(id, 0, 1, KEY_WRITE);
-	if (IS_ERR(key)) {
-		ret = PTR_ERR(key);
+	/* find the instantiation authorisation key */
+	instkey = key_get_instantiation_authkey(id);
+	if (IS_ERR(instkey)) {
+		ret = PTR_ERR(instkey);
 		goto error;
 	}
 
+	rka = instkey->payload.data;
+
 	/* find the destination keyring if present (which must also be
 	 * writable) */
 	keyring = NULL;
 	if (ringid) {
-		keyring = lookup_user_key(ringid, 1, 0, KEY_WRITE);
+		keyring = lookup_user_key(NULL, ringid, 1, 0, KEY_WRITE);
 		if (IS_ERR(keyring)) {
 			ret = PTR_ERR(keyring);
 			goto error2;
@@ -891,16 +910,54 @@ long keyctl_negate_key(key_serial_t id, unsigned timeout, key_serial_t ringid)
 	}
 
 	/* instantiate the key and link it into a keyring */
-	ret = key_negate_and_link(key, timeout, keyring);
+	ret = key_negate_and_link(rka->target_key, timeout, keyring, instkey);
 
 	key_put(keyring);
  error2:
-	key_put(key);
+	key_put(instkey);
  error:
 	return ret;
 
 } /* end keyctl_negate_key() */
 
+/*****************************************************************************/
+/*
+ * set the default keyring in which request_key() will cache keys
+ * - return the old setting
+ */
+long keyctl_set_reqkey_keyring(int reqkey_defl)
+{
+	int ret;
+
+	switch (reqkey_defl) {
+	case KEY_REQKEY_DEFL_THREAD_KEYRING:
+		ret = install_thread_keyring(current);
+		if (ret < 0)
+			return ret;
+		goto set;
+
+	case KEY_REQKEY_DEFL_PROCESS_KEYRING:
+		ret = install_process_keyring(current);
+		if (ret < 0)
+			return ret;
+
+	case KEY_REQKEY_DEFL_DEFAULT:
+	case KEY_REQKEY_DEFL_SESSION_KEYRING:
+	case KEY_REQKEY_DEFL_USER_KEYRING:
+	case KEY_REQKEY_DEFL_USER_SESSION_KEYRING:
+	set:
+		current->jit_keyring = reqkey_defl;
+
+	case KEY_REQKEY_DEFL_NO_CHANGE:
+		return current->jit_keyring;
+
+	case KEY_REQKEY_DEFL_GROUP_KEYRING:
+	default:
+		return -EINVAL;
+	}
+
+} /* end keyctl_set_reqkey_keyring() */
+
 /*****************************************************************************/
 /*
  * the key control system call
@@ -971,6 +1028,9 @@ asmlinkage long sys_keyctl(int option, unsigned long arg2, unsigned long arg3,
 					 (unsigned) arg3,
 					 (key_serial_t) arg4);
 
+	case KEYCTL_SET_REQKEY_KEYRING:
+		return keyctl_set_reqkey_keyring(arg2);
+
 	default:
 		return -EOPNOTSUPP;
 	}
diff --git a/security/keys/keyring.c b/security/keys/keyring.c
index c9a5de19748..90a551e4da6 100644
--- a/security/keys/keyring.c
+++ b/security/keys/keyring.c
@@ -1,6 +1,6 @@
 /* keyring.c: keyring handling
  *
- * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
+ * Copyright (C) 2004-5 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
  * This program is free software; you can redistribute it and/or
@@ -308,7 +308,7 @@ struct key *keyring_alloc(const char *description, uid_t uid, gid_t gid,
 			    uid, gid, KEY_USR_ALL, not_in_quota);
 
 	if (!IS_ERR(keyring)) {
-		ret = key_instantiate_and_link(keyring, NULL, 0, dest);
+		ret = key_instantiate_and_link(keyring, NULL, 0, dest, NULL);
 		if (ret < 0) {
 			key_put(keyring);
 			keyring = ERR_PTR(ret);
@@ -326,11 +326,12 @@ struct key *keyring_alloc(const char *description, uid_t uid, gid_t gid,
  * - we only find keys on which we have search permission
  * - we use the supplied match function to see if the description (or other
  *   feature of interest) matches
- * - we readlock the keyrings as we search down the tree
+ * - we rely on RCU to prevent the keyring lists from disappearing on us
  * - we return -EAGAIN if we didn't find any matching key
  * - we return -ENOKEY if we only found negative matching keys
  */
 struct key *keyring_search_aux(struct key *keyring,
+			       struct task_struct *context,
 			       struct key_type *type,
 			       const void *description,
 			       key_match_func_t match)
@@ -352,7 +353,7 @@ struct key *keyring_search_aux(struct key *keyring,
 
 	/* top keyring must have search permission to begin the search */
 	key = ERR_PTR(-EACCES);
-	if (!key_permission(keyring, KEY_SEARCH))
+	if (!key_task_permission(keyring, context, KEY_SEARCH))
 		goto error;
 
 	key = ERR_PTR(-ENOTDIR);
@@ -392,7 +393,7 @@ struct key *keyring_search_aux(struct key *keyring,
 			continue;
 
 		/* key must have search permissions */
-		if (!key_permission(key, KEY_SEARCH))
+		if (!key_task_permission(key, context, KEY_SEARCH))
 			continue;
 
 		/* we set a different error code if we find a negative key */
@@ -418,7 +419,7 @@ struct key *keyring_search_aux(struct key *keyring,
 		if (sp >= KEYRING_SEARCH_MAX_DEPTH)
 			continue;
 
-		if (!key_permission(key, KEY_SEARCH))
+		if (!key_task_permission(key, context, KEY_SEARCH))
 			continue;
 
 		/* stack the current position */
@@ -468,7 +469,11 @@ struct key *keyring_search(struct key *keyring,
 			   struct key_type *type,
 			   const char *description)
 {
-	return keyring_search_aux(keyring, type, description, type->match);
+	if (!type->match)
+		return ERR_PTR(-ENOKEY);
+
+	return keyring_search_aux(keyring, current,
+				  type, description, type->match);
 
 } /* end keyring_search() */
 
@@ -496,7 +501,8 @@ struct key *__keyring_search_one(struct key *keyring,
 			key = klist->keys[loop];
 
 			if (key->type == ktype &&
-			    key->type->match(key, description) &&
+			    (!key->type->match ||
+			     key->type->match(key, description)) &&
 			    key_permission(key, perm) &&
 			    !test_bit(KEY_FLAG_REVOKED, &key->flags)
 			    )
@@ -515,6 +521,51 @@ struct key *__keyring_search_one(struct key *keyring,
 
 } /* end __keyring_search_one() */
 
+/*****************************************************************************/
+/*
+ * search for an instantiation authorisation key matching a target key
+ * - the RCU read lock must be held by the caller
+ * - a target_id of zero specifies any valid token
+ */
+struct key *keyring_search_instkey(struct key *keyring,
+				   key_serial_t target_id)
+{
+	struct request_key_auth *rka;
+	struct keyring_list *klist;
+	struct key *instkey;
+	int loop;
+
+	klist = rcu_dereference(keyring->payload.subscriptions);
+	if (klist) {
+		for (loop = 0; loop < klist->nkeys; loop++) {
+			instkey = klist->keys[loop];
+
+			if (instkey->type != &key_type_request_key_auth)
+				continue;
+
+			rka = instkey->payload.data;
+			if (target_id && rka->target_key->serial != target_id)
+				continue;
+
+			/* the auth key is revoked during instantiation */
+			if (!test_bit(KEY_FLAG_REVOKED, &instkey->flags))
+				goto found;
+
+			instkey = ERR_PTR(-EKEYREVOKED);
+			goto error;
+		}
+	}
+
+	instkey = ERR_PTR(-EACCES);
+	goto error;
+
+found:
+	atomic_inc(&instkey->usage);
+error:
+	return instkey;
+
+} /* end keyring_search_instkey() */
+
 /*****************************************************************************/
 /*
  * find a keyring with the specified name
diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c
index 972e3017268..34db087bbcc 100644
--- a/security/keys/process_keys.c
+++ b/security/keys/process_keys.c
@@ -165,7 +165,7 @@ int install_thread_keyring(struct task_struct *tsk)
 /*
  * make sure a process keyring is installed
  */
-static int install_process_keyring(struct task_struct *tsk)
+int install_process_keyring(struct task_struct *tsk)
 {
 	unsigned long flags;
 	struct key *keyring;
@@ -376,12 +376,13 @@ void key_fsgid_changed(struct task_struct *tsk)
  * - we return -EAGAIN if we didn't find any matching key
  * - we return -ENOKEY if we found only negative matching keys
  */
-struct key *search_process_keyrings_aux(struct key_type *type,
-					const void *description,
-					key_match_func_t match)
+struct key *search_process_keyrings(struct key_type *type,
+				    const void *description,
+				    key_match_func_t match,
+				    struct task_struct *context)
 {
-	struct task_struct *tsk = current;
-	struct key *key, *ret, *err;
+	struct request_key_auth *rka;
+	struct key *key, *ret, *err, *instkey;
 
 	/* we want to return -EAGAIN or -ENOKEY if any of the keyrings were
 	 * searchable, but we failed to find a key or we found a negative key;
@@ -395,9 +396,9 @@ struct key *search_process_keyrings_aux(struct key_type *type,
 	err = ERR_PTR(-EAGAIN);
 
 	/* search the thread keyring first */
-	if (tsk->thread_keyring) {
-		key = keyring_search_aux(tsk->thread_keyring, type,
-					 description, match);
+	if (context->thread_keyring) {
+		key = keyring_search_aux(context->thread_keyring,
+					 context, type, description, match);
 		if (!IS_ERR(key))
 			goto found;
 
@@ -415,9 +416,9 @@ struct key *search_process_keyrings_aux(struct key_type *type,
 	}
 
 	/* search the process keyring second */
-	if (tsk->signal->process_keyring) {
-		key = keyring_search_aux(tsk->signal->process_keyring,
-					 type, description, match);
+	if (context->signal->process_keyring) {
+		key = keyring_search_aux(context->signal->process_keyring,
+					 context, type, description, match);
 		if (!IS_ERR(key))
 			goto found;
 
@@ -434,53 +435,93 @@ struct key *search_process_keyrings_aux(struct key_type *type,
 		}
 	}
 
-	/* search the session keyring last */
-	if (tsk->signal->session_keyring) {
+	/* search the session keyring */
+	if (context->signal->session_keyring) {
 		rcu_read_lock();
 		key = keyring_search_aux(
-			rcu_dereference(tsk->signal->session_keyring),
-			type, description, match);
+			rcu_dereference(context->signal->session_keyring),
+			context, type, description, match);
 		rcu_read_unlock();
+
+		if (!IS_ERR(key))
+			goto found;
+
+		switch (PTR_ERR(key)) {
+		case -EAGAIN: /* no key */
+			if (ret)
+				break;
+		case -ENOKEY: /* negative key */
+			ret = key;
+			break;
+		default:
+			err = key;
+			break;
+		}
+
+		/* if this process has a session keyring and that has an
+		 * instantiation authorisation key in the bottom level, then we
+		 * also search the keyrings of the process mentioned there */
+		if (context != current)
+			goto no_key;
+
+		rcu_read_lock();
+		instkey = __keyring_search_one(
+			rcu_dereference(context->signal->session_keyring),
+			&key_type_request_key_auth, NULL, 0);
+		rcu_read_unlock();
+
+		if (IS_ERR(instkey))
+			goto no_key;
+
+		rka = instkey->payload.data;
+
+		key = search_process_keyrings(type, description, match,
+					      rka->context);
+		key_put(instkey);
+
+		if (!IS_ERR(key))
+			goto found;
+
+		switch (PTR_ERR(key)) {
+		case -EAGAIN: /* no key */
+			if (ret)
+				break;
+		case -ENOKEY: /* negative key */
+			ret = key;
+			break;
+		default:
+			err = key;
+			break;
+		}
 	}
+	/* or search the user-session keyring */
 	else {
-		key = keyring_search_aux(tsk->user->session_keyring,
-					 type, description, match);
-	}
-
-	if (!IS_ERR(key))
-		goto found;
+		key = keyring_search_aux(context->user->session_keyring,
+					 context, type, description, match);
+		if (!IS_ERR(key))
+			goto found;
 
-	switch (PTR_ERR(key)) {
-	case -EAGAIN: /* no key */
-		if (ret)
+		switch (PTR_ERR(key)) {
+		case -EAGAIN: /* no key */
+			if (ret)
+				break;
+		case -ENOKEY: /* negative key */
+			ret = key;
 			break;
-	case -ENOKEY: /* negative key */
-		ret = key;
-		break;
-	default:
-		err = key;
-		break;
+		default:
+			err = key;
+			break;
+		}
 	}
 
+
+no_key:
 	/* no key - decide on the error we're going to go for */
 	key = ret ? ret : err;
 
- found:
+found:
 	return key;
 
-} /* end search_process_keyrings_aux() */
-
-/*****************************************************************************/
-/*
- * search the process keyrings for the first matching key
- * - we return -EAGAIN if we didn't find any matching key
- * - we return -ENOKEY if we found only negative matching keys
- */
-struct key *search_process_keyrings(struct key_type *type,
-				    const char *description)
-{
-	return search_process_keyrings_aux(type, description, type->match);
-
 } /* end search_process_keyrings() */
 
 /*****************************************************************************/
@@ -489,72 +530,73 @@ struct key *search_process_keyrings(struct key_type *type,
  * - don't create special keyrings unless so requested
  * - partially constructed keys aren't found unless requested
  */
-struct key *lookup_user_key(key_serial_t id, int create, int partial,
-			    key_perm_t perm)
+struct key *lookup_user_key(struct task_struct *context, key_serial_t id,
+			    int create, int partial, key_perm_t perm)
 {
-	struct task_struct *tsk = current;
-	unsigned long flags;
 	struct key *key;
 	int ret;
 
+	if (!context)
+		context = current;
+
 	key = ERR_PTR(-ENOKEY);
 
 	switch (id) {
 	case KEY_SPEC_THREAD_KEYRING:
-		if (!tsk->thread_keyring) {
+		if (!context->thread_keyring) {
 			if (!create)
 				goto error;
 
-			ret = install_thread_keyring(tsk);
+			ret = install_thread_keyring(context);
 			if (ret < 0) {
 				key = ERR_PTR(ret);
 				goto error;
 			}
 		}
 
-		key = tsk->thread_keyring;
+		key = context->thread_keyring;
 		atomic_inc(&key->usage);
 		break;
 
 	case KEY_SPEC_PROCESS_KEYRING:
-		if (!tsk->signal->process_keyring) {
+		if (!context->signal->process_keyring) {
 			if (!create)
 				goto error;
 
-			ret = install_process_keyring(tsk);
+			ret = install_process_keyring(context);
 			if (ret < 0) {
 				key = ERR_PTR(ret);
 				goto error;
 			}
 		}
 
-		key = tsk->signal->process_keyring;
+		key = context->signal->process_keyring;
 		atomic_inc(&key->usage);
 		break;
 
 	case KEY_SPEC_SESSION_KEYRING:
-		if (!tsk->signal->session_keyring) {
+		if (!context->signal->session_keyring) {
 			/* always install a session keyring upon access if one
 			 * doesn't exist yet */
 			ret = install_session_keyring(
-			       tsk, tsk->user->session_keyring);
+			       context, context->user->session_keyring);
 			if (ret < 0)
 				goto error;
 		}
 
-		spin_lock_irqsave(&tsk->sighand->siglock, flags);
-		key = tsk->signal->session_keyring;
+		rcu_read_lock();
+		key = rcu_dereference(context->signal->session_keyring);
 		atomic_inc(&key->usage);
-		spin_unlock_irqrestore(&tsk->sighand->siglock, flags);
+		rcu_read_unlock();
 		break;
 
 	case KEY_SPEC_USER_KEYRING:
-		key = tsk->user->uid_keyring;
+		key = context->user->uid_keyring;
 		atomic_inc(&key->usage);
 		break;
 
 	case KEY_SPEC_USER_SESSION_KEYRING:
-		key = tsk->user->session_keyring;
+		key = context->user->session_keyring;
 		atomic_inc(&key->usage);
 		break;
 
@@ -574,7 +616,7 @@ struct key *lookup_user_key(key_serial_t id, int create, int partial,
 		break;
 	}
 
-	/* check the status and permissions */
+	/* check the status */
 	if (perm) {
 		ret = key_validate(key);
 		if (ret < 0)
@@ -585,8 +627,10 @@ struct key *lookup_user_key(key_serial_t id, int create, int partial,
 	if (!partial && !test_bit(KEY_FLAG_INSTANTIATED, &key->flags))
 		goto invalid_key;
 
+	/* check the permissions */
 	ret = -EACCES;
-	if (!key_permission(key, perm))
+
+	if (!key_task_permission(key, context, perm))
 		goto invalid_key;
 
  error:
@@ -609,7 +653,6 @@ struct key *lookup_user_key(key_serial_t id, int create, int partial,
 long join_session_keyring(const char *name)
 {
 	struct task_struct *tsk = current;
-	unsigned long flags;
 	struct key *keyring;
 	long ret;
 
@@ -619,9 +662,9 @@ long join_session_keyring(const char *name)
 		if (ret < 0)
 			goto error;
 
-		spin_lock_irqsave(&tsk->sighand->siglock, flags);
-		ret = tsk->signal->session_keyring->serial;
-		spin_unlock_irqrestore(&tsk->sighand->siglock, flags);
+		rcu_read_lock();
+		ret = rcu_dereference(tsk->signal->session_keyring)->serial;
+		rcu_read_unlock();
 		goto error;
 	}
 
diff --git a/security/keys/request_key.c b/security/keys/request_key.c
index 54aa7b70e63..dfcd983af1f 100644
--- a/security/keys/request_key.c
+++ b/security/keys/request_key.c
@@ -1,6 +1,6 @@
 /* request_key.c: request a key from userspace
  *
- * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
+ * Copyright (C) 2004-5 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
  * This program is free software; you can redistribute it and/or
@@ -13,6 +13,7 @@
 #include <linux/sched.h>
 #include <linux/kmod.h>
 #include <linux/err.h>
+#include <linux/keyctl.h>
 #include "internal.h"
 
 struct key_construction {
@@ -27,18 +28,26 @@ DECLARE_WAIT_QUEUE_HEAD(request_key_conswq);
 /*
  * request userspace finish the construction of a key
  * - execute "/sbin/request-key <op> <key> <uid> <gid> <keyring> <keyring> <keyring> <info>"
- * - if callout_info is an empty string, it'll be rendered as a "-" instead
  */
 static int call_request_key(struct key *key,
 			    const char *op,
 			    const char *callout_info)
 {
 	struct task_struct *tsk = current;
-	unsigned long flags;
 	key_serial_t prkey, sskey;
+	struct key *session_keyring, *rkakey;
 	char *argv[10], *envp[3], uid_str[12], gid_str[12];
 	char key_str[12], keyring_str[3][12];
-	int i;
+	int ret, i;
+
+	kenter("{%d},%s,%s", key->serial, op, callout_info);
+
+	/* generate a new session keyring with an auth key in it */
+	session_keyring = request_key_auth_new(key, &rkakey);
+	if (IS_ERR(session_keyring)) {
+		ret = PTR_ERR(session_keyring);
+		goto error;
+	}
 
 	/* record the UID and GID */
 	sprintf(uid_str, "%d", current->fsuid);
@@ -55,17 +64,17 @@ static int call_request_key(struct key *key,
 	if (tsk->signal->process_keyring)
 		prkey = tsk->signal->process_keyring->serial;
 
-	sskey = 0;
-	spin_lock_irqsave(&tsk->sighand->siglock, flags);
-	if (tsk->signal->session_keyring)
-		sskey = tsk->signal->session_keyring->serial;
-	spin_unlock_irqrestore(&tsk->sighand->siglock, flags);
-
+	sprintf(keyring_str[1], "%d", prkey);
 
-	if (!sskey)
+	if (tsk->signal->session_keyring) {
+		rcu_read_lock();
+		sskey = rcu_dereference(tsk->signal->session_keyring)->serial;
+		rcu_read_unlock();
+	}
+	else {
 		sskey = tsk->user->session_keyring->serial;
+	}
 
-	sprintf(keyring_str[1], "%d", prkey);
 	sprintf(keyring_str[2], "%d", sskey);
 
 	/* set up a minimal environment */
@@ -84,11 +93,20 @@ static int call_request_key(struct key *key,
 	argv[i++] = keyring_str[0];
 	argv[i++] = keyring_str[1];
 	argv[i++] = keyring_str[2];
-	argv[i++] = callout_info[0] ? (char *) callout_info : "-";
+	argv[i++] = (char *) callout_info;
 	argv[i] = NULL;
 
 	/* do it */
-	return call_usermodehelper_keys(argv[0], argv, envp, NULL, 1);
+	ret = call_usermodehelper_keys(argv[0], argv, envp, session_keyring, 1);
+
+	/* dispose of the special keys */
+	key_revoke(rkakey);
+	key_put(rkakey);
+	key_put(session_keyring);
+
+ error:
+	kleave(" = %d", ret);
+	return ret;
 
 } /* end call_request_key() */
 
@@ -107,6 +125,8 @@ static struct key *__request_key_construction(struct key_type *type,
 	struct key *key;
 	int ret, negated;
 
+	kenter("%s,%s,%s", type->name, description, callout_info);
+
 	/* create a key and add it to the queue */
 	key = key_alloc(type, description,
 			current->fsuid, current->fsgid, KEY_USR_ALL, 0);
@@ -143,6 +163,7 @@ static struct key *__request_key_construction(struct key_type *type,
 	}
 
  out:
+	kleave(" = %p", key);
 	return key;
 
  request_failed:
@@ -216,6 +237,9 @@ static struct key *request_key_construction(struct key_type *type,
 
 	DECLARE_WAITQUEUE(myself, current);
 
+	kenter("%s,%s,{%d},%s",
+	       type->name, description, user->uid, callout_info);
+
 	/* see if there's such a key under construction already */
 	down_write(&key_construction_sem);
 
@@ -232,6 +256,7 @@ static struct key *request_key_construction(struct key_type *type,
 	/* see about getting userspace to construct the key */
 	key = __request_key_construction(type, description, callout_info);
  error:
+	kleave(" = %p", key);
 	return key;
 
 	/* someone else has the same key under construction
@@ -245,9 +270,11 @@ static struct key *request_key_construction(struct key_type *type,
 	add_wait_queue(&request_key_conswq, &myself);
 
 	for (;;) {
-		set_current_state(TASK_UNINTERRUPTIBLE);
+		set_current_state(TASK_INTERRUPTIBLE);
 		if (!test_bit(KEY_FLAG_USER_CONSTRUCT, &ckey->flags))
 			break;
+		if (signal_pending(current))
+			break;
 		schedule();
 	}
 
@@ -265,23 +292,85 @@ static struct key *request_key_construction(struct key_type *type,
 
 } /* end request_key_construction() */
 
+/*****************************************************************************/
+/*
+ * link a freshly minted key to an appropriate destination keyring
+ */
+static void request_key_link(struct key *key, struct key *dest_keyring)
+{
+	struct task_struct *tsk = current;
+	struct key *drop = NULL;
+
+	kenter("{%d},%p", key->serial, dest_keyring);
+
+	/* find the appropriate keyring */
+	if (!dest_keyring) {
+		switch (tsk->jit_keyring) {
+		case KEY_REQKEY_DEFL_DEFAULT:
+		case KEY_REQKEY_DEFL_THREAD_KEYRING:
+			dest_keyring = tsk->thread_keyring;
+			if (dest_keyring)
+				break;
+
+		case KEY_REQKEY_DEFL_PROCESS_KEYRING:
+			dest_keyring = tsk->signal->process_keyring;
+			if (dest_keyring)
+				break;
+
+		case KEY_REQKEY_DEFL_SESSION_KEYRING:
+			rcu_read_lock();
+			dest_keyring = key_get(
+				rcu_dereference(tsk->signal->session_keyring));
+			rcu_read_unlock();
+			drop = dest_keyring;
+
+			if (dest_keyring)
+				break;
+
+		case KEY_REQKEY_DEFL_USER_SESSION_KEYRING:
+			dest_keyring = current->user->session_keyring;
+			break;
+
+		case KEY_REQKEY_DEFL_USER_KEYRING:
+			dest_keyring = current->user->uid_keyring;
+			break;
+
+		case KEY_REQKEY_DEFL_GROUP_KEYRING:
+		default:
+			BUG();
+		}
+	}
+
+	/* and attach the key to it */
+	key_link(dest_keyring, key);
+
+	key_put(drop);
+
+	kleave("");
+
+} /* end request_key_link() */
+
 /*****************************************************************************/
 /*
  * request a key
  * - search the process's keyrings
  * - check the list of keys being created or updated
- * - call out to userspace for a key if requested (supplementary info can be
- *   passed)
+ * - call out to userspace for a key if supplementary info was provided
+ * - cache the key in an appropriate keyring
  */
-struct key *request_key(struct key_type *type,
-			const char *description,
-			const char *callout_info)
+struct key *request_key_and_link(struct key_type *type,
+				 const char *description,
+				 const char *callout_info,
+				 struct key *dest_keyring)
 {
 	struct key_user *user;
 	struct key *key;
 
+	kenter("%s,%s,%s,%p",
+	       type->name, description, callout_info, dest_keyring);
+
 	/* search all the process keyrings for a key */
-	key = search_process_keyrings_aux(type, description, type->match);
+	key = search_process_keyrings(type, description, type->match, current);
 
 	if (PTR_ERR(key) == -EAGAIN) {
 		/* the search failed, but the keyrings were searchable, so we
@@ -292,12 +381,13 @@ struct key *request_key(struct key_type *type,
 
 		/* - get hold of the user's construction queue */
 		user = key_user_lookup(current->fsuid);
-		if (!user) {
-			key = ERR_PTR(-ENOMEM);
-			goto error;
-		}
+		if (!user)
+			goto nomem;
+
+		do {
+			if (signal_pending(current))
+				goto interrupted;
 
-		for (;;) {
 			/* ask userspace (returns NULL if it waited on a key
 			 * being constructed) */
 			key = request_key_construction(type, description,
@@ -307,18 +397,46 @@ struct key *request_key(struct key_type *type,
 
 			/* someone else made the key we want, so we need to
 			 * search again as it might now be available to us */
-			key = search_process_keyrings_aux(type, description,
-							  type->match);
-			if (PTR_ERR(key) != -EAGAIN)
-				break;
-		}
+			key = search_process_keyrings(type, description,
+						      type->match, current);
+
+		} while (PTR_ERR(key) == -EAGAIN);
 
 		key_user_put(user);
+
+		/* link the new key into the appropriate keyring */
+		if (!PTR_ERR(key))
+			request_key_link(key, dest_keyring);
 	}
 
- error:
+error:
+	kleave(" = %p", key);
 	return key;
 
+nomem:
+	key = ERR_PTR(-ENOMEM);
+	goto error;
+
+interrupted:
+	key_user_put(user);
+	key = ERR_PTR(-EINTR);
+	goto error;
+
+} /* end request_key_and_link() */
+
+/*****************************************************************************/
+/*
+ * request a key
+ * - search the process's keyrings
+ * - check the list of keys being created or updated
+ * - call out to userspace for a key if supplementary info was provided
+ */
+struct key *request_key(struct key_type *type,
+			const char *description,
+			const char *callout_info)
+{
+	return request_key_and_link(type, description, callout_info, NULL);
+
 } /* end request_key() */
 
 EXPORT_SYMBOL(request_key);
diff --git a/security/keys/request_key_auth.c b/security/keys/request_key_auth.c
new file mode 100644
index 00000000000..f2226463222
--- /dev/null
+++ b/security/keys/request_key_auth.c
@@ -0,0 +1,180 @@
+/* request_key_auth.c: request key authorisation controlling key def
+ *
+ * Copyright (C) 2005 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/err.h>
+#include <linux/seq_file.h>
+#include "internal.h"
+
+static int request_key_auth_instantiate(struct key *, const void *, size_t);
+static void request_key_auth_describe(const struct key *, struct seq_file *);
+static void request_key_auth_destroy(struct key *);
+
+/*
+ * the request-key authorisation key type definition
+ */
+struct key_type key_type_request_key_auth = {
+	.name		= ".request_key_auth",
+	.def_datalen	= sizeof(struct request_key_auth),
+	.instantiate	= request_key_auth_instantiate,
+	.describe	= request_key_auth_describe,
+	.destroy	= request_key_auth_destroy,
+};
+
+/*****************************************************************************/
+/*
+ * instantiate a request-key authorisation record
+ */
+static int request_key_auth_instantiate(struct key *key,
+					const void *data,
+					size_t datalen)
+{
+	struct request_key_auth *rka, *irka;
+	struct key *instkey;
+	int ret;
+
+	ret = -ENOMEM;
+	rka = kmalloc(sizeof(*rka), GFP_KERNEL);
+	if (rka) {
+		/* see if the calling process is already servicing the key
+		 * request of another process */
+		instkey = key_get_instantiation_authkey(0);
+		if (!IS_ERR(instkey)) {
+			/* it is - use that instantiation context here too */
+			irka = instkey->payload.data;
+			rka->context = irka->context;
+			rka->pid = irka->pid;
+			key_put(instkey);
+		}
+		else {
+			/* it isn't - use this process as the context */
+			rka->context = current;
+			rka->pid = current->pid;
+		}
+
+		rka->target_key = key_get((struct key *) data);
+		key->payload.data = rka;
+		ret = 0;
+	}
+
+	return ret;
+
+} /* end request_key_auth_instantiate() */
+
+/*****************************************************************************/
+/*
+ *
+ */
+static void request_key_auth_describe(const struct key *key,
+				      struct seq_file *m)
+{
+	struct request_key_auth *rka = key->payload.data;
+
+	seq_puts(m, "key:");
+	seq_puts(m, key->description);
+	seq_printf(m, " pid:%d", rka->pid);
+
+} /* end request_key_auth_describe() */
+
+/*****************************************************************************/
+/*
+ * destroy an instantiation authorisation token key
+ */
+static void request_key_auth_destroy(struct key *key)
+{
+	struct request_key_auth *rka = key->payload.data;
+
+	kenter("{%d}", key->serial);
+
+	key_put(rka->target_key);
+
+} /* end request_key_auth_destroy() */
+
+/*****************************************************************************/
+/*
+ * create a session keyring to be for the invokation of /sbin/request-key and
+ * stick an authorisation token in it
+ */
+struct key *request_key_auth_new(struct key *target, struct key **_rkakey)
+{
+	struct key *keyring, *rkakey = NULL;
+	char desc[20];
+	int ret;
+
+	kenter("%d,", target->serial);
+
+	/* allocate a new session keyring */
+	sprintf(desc, "_req.%u", target->serial);
+
+	keyring = keyring_alloc(desc, current->fsuid, current->fsgid, 1, NULL);
+	if (IS_ERR(keyring)) {
+		kleave("= %ld", PTR_ERR(keyring));
+		return keyring;
+	}
+
+	/* allocate the auth key */
+	sprintf(desc, "%x", target->serial);
+
+	rkakey = key_alloc(&key_type_request_key_auth, desc,
+			   current->fsuid, current->fsgid,
+			   KEY_USR_VIEW, 1);
+	if (IS_ERR(rkakey)) {
+		key_put(keyring);
+		kleave("= %ld", PTR_ERR(rkakey));
+		return rkakey;
+	}
+
+	/* construct and attach to the keyring */
+	ret = key_instantiate_and_link(rkakey, target, 0, keyring, NULL);
+	if (ret < 0) {
+		key_revoke(rkakey);
+		key_put(rkakey);
+		key_put(keyring);
+		kleave("= %d", ret);
+		return ERR_PTR(ret);
+	}
+
+	*_rkakey = rkakey;
+	kleave(" = {%d} ({%d})", keyring->serial, rkakey->serial);
+	return keyring;
+
+} /* end request_key_auth_new() */
+
+/*****************************************************************************/
+/*
+ * get the authorisation key for instantiation of a specific key if attached to
+ * the current process's keyrings
+ * - this key is inserted into a keyring and that is set as /sbin/request-key's
+ *   session keyring
+ * - a target_id of zero specifies any valid token
+ */
+struct key *key_get_instantiation_authkey(key_serial_t target_id)
+{
+	struct task_struct *tsk = current;
+	struct key *instkey;
+
+	/* we must have our own personal session keyring */
+	if (!tsk->signal->session_keyring)
+		return ERR_PTR(-EACCES);
+
+	/* and it must contain a suitable request authorisation key
+	 * - lock RCU against session keyring changing
+	 */
+	rcu_read_lock();
+
+	instkey = keyring_search_instkey(
+		rcu_dereference(tsk->signal->session_keyring), target_id);
+
+	rcu_read_unlock();
+	return instkey;
+
+} /* end key_get_instantiation_authkey() */
-- 
cgit v1.2.3-70-g09d2


From 92198f7eaa5df3479341dd8fa20c2c81aa3b1e25 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Thu, 23 Jun 2005 22:00:59 -0700
Subject: [PATCH] pass iocb to dio_iodone_t

XFS will have to look at iocb->private to fix aio+dio.  No other filesystem
is using the blockdev_direct_IO* end_io callback.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/direct-io.c              | 2 +-
 fs/xfs/linux-2.6/xfs_aops.c | 3 ++-
 include/linux/fs.h          | 4 ++--
 3 files changed, 5 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/fs/direct-io.c b/fs/direct-io.c
index 1d55e7e6734..0d06097bc99 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -215,7 +215,7 @@ static struct page *dio_get_page(struct dio *dio)
 static void dio_complete(struct dio *dio, loff_t offset, ssize_t bytes)
 {
 	if (dio->end_io && dio->result)
-		dio->end_io(dio->inode, offset, bytes, dio->map_bh.b_private);
+		dio->end_io(dio->iocb, offset, bytes, dio->map_bh.b_private);
 	if (dio->lock_type == DIO_LOCKING)
 		up_read(&dio->inode->i_alloc_sem);
 }
diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c
index 93ce257cd14..a3a4b5aaf5d 100644
--- a/fs/xfs/linux-2.6/xfs_aops.c
+++ b/fs/xfs/linux-2.6/xfs_aops.c
@@ -149,11 +149,12 @@ linvfs_unwritten_convert(
  */
 STATIC void
 linvfs_unwritten_convert_direct(
-	struct inode	*inode,
+	struct kiocb	*iocb,
 	loff_t		offset,
 	ssize_t		size,
 	void		*private)
 {
+	struct inode	*inode = iocb->ki_filp->f_dentry->d_inode;
 	ASSERT(!private || inode == (struct inode *)private);
 
 	/* private indicates an unwritten extent lay beneath this IO */
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 517bf4966bf..83857d8070d 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -220,6 +220,7 @@ extern int dir_notify_enable;
 
 struct iovec;
 struct nameidata;
+struct kiocb;
 struct pipe_inode_info;
 struct poll_table_struct;
 struct kstatfs;
@@ -240,7 +241,7 @@ typedef int (get_block_t)(struct inode *inode, sector_t iblock,
 typedef int (get_blocks_t)(struct inode *inode, sector_t iblock,
 			unsigned long max_blocks,
 			struct buffer_head *bh_result, int create);
-typedef void (dio_iodone_t)(struct inode *inode, loff_t offset,
+typedef void (dio_iodone_t)(struct kiocb *iocb, loff_t offset,
 			ssize_t bytes, void *private);
 
 /*
@@ -302,7 +303,6 @@ struct iattr {
 struct page;
 struct address_space;
 struct writeback_control;
-struct kiocb;
 
 struct address_space_operations {
 	int (*writepage)(struct page *page, struct writeback_control *wbc);
-- 
cgit v1.2.3-70-g09d2


From 4e5117ba0af4582b6ec9164874f719d7f3f1eb2b Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Thu, 23 Jun 2005 22:01:03 -0700
Subject: [PATCH] quota: improve credits estimates

Improve estimates on the number of needed credits for quota transaction.
Now we distinguish blocks that might need to be allocated and blocks that
only need to be rewritten.  Also we distinguish deleting of a quota
structure and creating of a new one.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/dqblk_v1.h | 6 ++++++
 include/linux/dqblk_v2.h | 6 ++++++
 include/linux/quota.h    | 7 +++++--
 3 files changed, 17 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/dqblk_v1.h b/include/linux/dqblk_v1.h
index 42fbf479715..57f1250d5a5 100644
--- a/include/linux/dqblk_v1.h
+++ b/include/linux/dqblk_v1.h
@@ -11,6 +11,12 @@
 /* Root squash turned on */
 #define V1_DQF_RSQUASH 1
 
+/* Numbers of blocks needed for updates */
+#define V1_INIT_ALLOC 1
+#define V1_INIT_REWRITE 1
+#define V1_DEL_ALLOC 0
+#define V1_DEL_REWRITE 2
+
 /* Special information about quotafile */
 struct v1_mem_dqinfo {
 };
diff --git a/include/linux/dqblk_v2.h b/include/linux/dqblk_v2.h
index 4a6c5f6867b..4f853322cb7 100644
--- a/include/linux/dqblk_v2.h
+++ b/include/linux/dqblk_v2.h
@@ -10,6 +10,12 @@
 /* id numbers of quota format */
 #define QFMT_VFS_V0 2
 
+/* Numbers of blocks needed for updates */
+#define V2_INIT_ALLOC 4
+#define V2_INIT_REWRITE 2
+#define V2_DEL_ALLOC 0
+#define V2_DEL_REWRITE 6
+
 /* Inmemory copy of version specific information */
 struct v2_mem_dqinfo {
 	unsigned int dqi_blocks;
diff --git a/include/linux/quota.h b/include/linux/quota.h
index ac5b90f4f25..700ead45084 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -138,8 +138,11 @@ struct if_dqinfo {
 #include <linux/dqblk_v2.h>
 
 /* Maximal numbers of writes for quota operation (insert/delete/update)
- * (over all formats) - info block, 4 pointer blocks, data block */
-#define DQUOT_MAX_WRITES	6
+ * (over VFS all formats) */
+#define DQUOT_INIT_ALLOC max(V1_INIT_ALLOC, V2_INIT_ALLOC)
+#define DQUOT_INIT_REWRITE max(V1_INIT_REWRITE, V2_INIT_REWRITE)
+#define DQUOT_DEL_ALLOC max(V1_DEL_ALLOC, V2_DEL_ALLOC)
+#define DQUOT_DEL_REWRITE max(V1_DEL_REWRITE, V2_DEL_REWRITE)
 
 /*
  * Data for one user/group kept in memory
-- 
cgit v1.2.3-70-g09d2


From 1f54587bea84a35125c95e19b98c2f464c50871b Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Thu, 23 Jun 2005 22:01:04 -0700
Subject: [PATCH] quota: ext3: Improve quota credit estimates

Use improved credits estimates for quota operations.  Also reserve a space
for a quota operation in a transaction only if filesystem was mounted with
some quota options.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext3/acl.c            |  5 +++--
 fs/ext3/inode.c          |  7 ++++---
 fs/ext3/namei.c          | 25 +++++++++++++------------
 fs/ext3/super.c          | 37 +++++++++++++++++++++++++++----------
 fs/ext3/xattr.c          |  2 +-
 include/linux/ext3_fs.h  |  1 +
 include/linux/ext3_jbd.h | 19 +++++++++++--------
 7 files changed, 60 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c
index 133f5aa581b..3ac38266fc9 100644
--- a/fs/ext3/acl.c
+++ b/fs/ext3/acl.c
@@ -393,7 +393,8 @@ ext3_acl_chmod(struct inode *inode)
 		int retries = 0;
 
 	retry:
-		handle = ext3_journal_start(inode, EXT3_DATA_TRANS_BLOCKS);
+		handle = ext3_journal_start(inode,
+				EXT3_DATA_TRANS_BLOCKS(inode->i_sb));
 		if (IS_ERR(handle)) {
 			error = PTR_ERR(handle);
 			ext3_std_error(inode->i_sb, error);
@@ -503,7 +504,7 @@ ext3_xattr_set_acl(struct inode *inode, int type, const void *value,
 		acl = NULL;
 
 retry:
-	handle = ext3_journal_start(inode, EXT3_DATA_TRANS_BLOCKS);
+	handle = ext3_journal_start(inode, EXT3_DATA_TRANS_BLOCKS(inode->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 	error = ext3_set_acl(handle, inode, type, acl);
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 0d5fa73b18d..0b2db4f618c 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -128,7 +128,7 @@ static unsigned long blocks_for_truncate(struct inode *inode)
 	if (needed > EXT3_MAX_TRANS_DATA) 
 		needed = EXT3_MAX_TRANS_DATA;
 
-	return EXT3_DATA_TRANS_BLOCKS + needed;
+	return EXT3_DATA_TRANS_BLOCKS(inode->i_sb) + needed;
 }
 
 /* 
@@ -2763,7 +2763,8 @@ int ext3_setattr(struct dentry *dentry, struct iattr *attr)
 
 		/* (user+group)*(old+new) structure, inode write (sb,
 		 * inode block, ? - but truncate inode update has it) */
-		handle = ext3_journal_start(inode, 4*EXT3_QUOTA_INIT_BLOCKS+3);
+		handle = ext3_journal_start(inode, 2*(EXT3_QUOTA_INIT_BLOCKS(inode->i_sb)+
+					EXT3_QUOTA_DEL_BLOCKS(inode->i_sb))+3);
 		if (IS_ERR(handle)) {
 			error = PTR_ERR(handle);
 			goto err_out;
@@ -2861,7 +2862,7 @@ static int ext3_writepage_trans_blocks(struct inode *inode)
 #ifdef CONFIG_QUOTA
 	/* We know that structure was already allocated during DQUOT_INIT so
 	 * we will be updating only the data blocks + inodes */
-	ret += 2*EXT3_QUOTA_TRANS_BLOCKS;
+	ret += 2*EXT3_QUOTA_TRANS_BLOCKS(inode->i_sb);
 #endif
 
 	return ret;
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index 60e44e6dd7a..50378d8ff84 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -1645,9 +1645,9 @@ static int ext3_create (struct inode * dir, struct dentry * dentry, int mode,
 	int err, retries = 0;
 
 retry:
-	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS +
+	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) +
 					EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 +
-					2*EXT3_QUOTA_INIT_BLOCKS);
+					2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
@@ -1679,9 +1679,9 @@ static int ext3_mknod (struct inode * dir, struct dentry *dentry,
 		return -EINVAL;
 
 retry:
-	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS +
+	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) +
 			 		EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 +
-					2*EXT3_QUOTA_INIT_BLOCKS);
+					2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
@@ -1715,9 +1715,9 @@ static int ext3_mkdir(struct inode * dir, struct dentry * dentry, int mode)
 		return -EMLINK;
 
 retry:
-	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS +
+	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) +
 					EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 +
-					2*EXT3_QUOTA_INIT_BLOCKS);
+					2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
@@ -2006,7 +2006,7 @@ static int ext3_rmdir (struct inode * dir, struct dentry *dentry)
 	/* Initialize quotas before so that eventual writes go in
 	 * separate transaction */
 	DQUOT_INIT(dentry->d_inode);
-	handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS);
+	handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
@@ -2065,7 +2065,7 @@ static int ext3_unlink(struct inode * dir, struct dentry *dentry)
 	/* Initialize quotas before so that eventual writes go
 	 * in separate transaction */
 	DQUOT_INIT(dentry->d_inode);
-	handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS);
+	handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
@@ -2120,9 +2120,9 @@ static int ext3_symlink (struct inode * dir,
 		return -ENAMETOOLONG;
 
 retry:
-	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS +
+	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) +
 			 		EXT3_INDEX_EXTRA_TRANS_BLOCKS + 5 +
-					2*EXT3_QUOTA_INIT_BLOCKS);
+					2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
@@ -2174,7 +2174,7 @@ static int ext3_link (struct dentry * old_dentry,
 		return -EMLINK;
 
 retry:
-	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS +
+	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) +
 					EXT3_INDEX_EXTRA_TRANS_BLOCKS);
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
@@ -2216,7 +2216,8 @@ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry,
 	 * in separate transaction */
 	if (new_dentry->d_inode)
 		DQUOT_INIT(new_dentry->d_inode);
-	handle = ext3_journal_start(old_dir, 2 * EXT3_DATA_TRANS_BLOCKS +
+	handle = ext3_journal_start(old_dir, 2 *
+					EXT3_DATA_TRANS_BLOCKS(old_dir->i_sb) +
 			 		EXT3_INDEX_EXTRA_TRANS_BLOCKS + 2);
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 9630fbfdc24..b4b3e8a3913 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -589,7 +589,7 @@ enum {
 	Opt_commit, Opt_journal_update, Opt_journal_inum,
 	Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback,
 	Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
-	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0,
+	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota,
 	Opt_ignore, Opt_barrier, Opt_err, Opt_resize,
 };
 
@@ -634,10 +634,10 @@ static match_table_t tokens = {
 	{Opt_grpjquota, "grpjquota=%s"},
 	{Opt_jqfmt_vfsold, "jqfmt=vfsold"},
 	{Opt_jqfmt_vfsv0, "jqfmt=vfsv0"},
-	{Opt_ignore, "grpquota"},
-	{Opt_ignore, "noquota"},
-	{Opt_ignore, "quota"},
-	{Opt_ignore, "usrquota"},
+	{Opt_quota, "grpquota"},
+	{Opt_noquota, "noquota"},
+	{Opt_quota, "quota"},
+	{Opt_quota, "usrquota"},
 	{Opt_barrier, "barrier=%u"},
 	{Opt_err, NULL},
 	{Opt_resize, "resize"},
@@ -876,6 +876,7 @@ set_qf_name:
 				sbi->s_qf_names[qtype] = NULL;
 				return 0;
 			}
+			set_opt(sbi->s_mount_opt, QUOTA);
 			break;
 		case Opt_offusrjquota:
 			qtype = USRQUOTA;
@@ -898,6 +899,17 @@ clear_qf_name:
 		case Opt_jqfmt_vfsv0:
 			sbi->s_jquota_fmt = QFMT_VFS_V0;
 			break;
+		case Opt_quota:
+			set_opt(sbi->s_mount_opt, QUOTA);
+			break;
+		case Opt_noquota:
+			if (sb_any_quota_enabled(sb)) {
+				printk(KERN_ERR "EXT3-fs: Cannot change quota "
+					"options when quota turned on.\n");
+				return 0;
+			}
+			clear_opt(sbi->s_mount_opt, QUOTA);
+			break;
 #else
 		case Opt_usrjquota:
 		case Opt_grpjquota:
@@ -909,6 +921,9 @@ clear_qf_name:
 				"EXT3-fs: journalled quota options not "
 				"supported.\n");
 			break;
+		case Opt_quota:
+		case Opt_noquota:
+			break;
 #endif
 		case Opt_abort:
 			set_opt(sbi->s_mount_opt, ABORT);
@@ -2238,7 +2253,7 @@ static int ext3_dquot_initialize(struct inode *inode, int type)
 	int ret, err;
 
 	/* We may create quota structure so we need to reserve enough blocks */
-	handle = ext3_journal_start(inode, 2*EXT3_QUOTA_INIT_BLOCKS);
+	handle = ext3_journal_start(inode, 2*EXT3_QUOTA_INIT_BLOCKS(inode->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 	ret = dquot_initialize(inode, type);
@@ -2254,7 +2269,7 @@ static int ext3_dquot_drop(struct inode *inode)
 	int ret, err;
 
 	/* We may delete quota structure so we need to reserve enough blocks */
-	handle = ext3_journal_start(inode, 2*EXT3_QUOTA_INIT_BLOCKS);
+	handle = ext3_journal_start(inode, 2*EXT3_QUOTA_DEL_BLOCKS(inode->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 	ret = dquot_drop(inode);
@@ -2272,7 +2287,7 @@ static int ext3_write_dquot(struct dquot *dquot)
 
 	inode = dquot_to_inode(dquot);
 	handle = ext3_journal_start(inode,
-					EXT3_QUOTA_TRANS_BLOCKS);
+					EXT3_QUOTA_TRANS_BLOCKS(dquot->dq_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 	ret = dquot_commit(dquot);
@@ -2288,7 +2303,7 @@ static int ext3_acquire_dquot(struct dquot *dquot)
 	handle_t *handle;
 
 	handle = ext3_journal_start(dquot_to_inode(dquot),
-					EXT3_QUOTA_INIT_BLOCKS);
+					EXT3_QUOTA_INIT_BLOCKS(dquot->dq_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 	ret = dquot_acquire(dquot);
@@ -2304,7 +2319,7 @@ static int ext3_release_dquot(struct dquot *dquot)
 	handle_t *handle;
 
 	handle = ext3_journal_start(dquot_to_inode(dquot),
-					EXT3_QUOTA_INIT_BLOCKS);
+					EXT3_QUOTA_DEL_BLOCKS(dquot->dq_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 	ret = dquot_release(dquot);
@@ -2361,6 +2376,8 @@ static int ext3_quota_on(struct super_block *sb, int type, int format_id,
 	int err;
 	struct nameidata nd;
 
+	if (!test_opt(sb, QUOTA))
+		return -EINVAL;
 	/* Not journalling quota? */
 	if (!EXT3_SB(sb)->s_qf_names[USRQUOTA] &&
 	    !EXT3_SB(sb)->s_qf_names[GRPQUOTA])
diff --git a/fs/ext3/xattr.c b/fs/ext3/xattr.c
index 4cbc6d0212d..3f9dfa643b1 100644
--- a/fs/ext3/xattr.c
+++ b/fs/ext3/xattr.c
@@ -1044,7 +1044,7 @@ ext3_xattr_set(struct inode *inode, int name_index, const char *name,
 	int error, retries = 0;
 
 retry:
-	handle = ext3_journal_start(inode, EXT3_DATA_TRANS_BLOCKS);
+	handle = ext3_journal_start(inode, EXT3_DATA_TRANS_BLOCKS(inode->i_sb));
 	if (IS_ERR(handle)) {
 		error = PTR_ERR(handle);
 	} else {
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index 74ad31781e3..4b6e1ab216a 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -358,6 +358,7 @@ struct ext3_inode {
 #define EXT3_MOUNT_RESERVATION		0x10000	/* Preallocation */
 #define EXT3_MOUNT_BARRIER		0x20000 /* Use block barriers */
 #define EXT3_MOUNT_NOBH			0x40000 /* No bufferheads */
+#define EXT3_MOUNT_QUOTA		0x80000 /* Some quota option set */
 
 /* Compatibility, for having both ext2_fs.h and ext3_fs.h included at once */
 #ifndef _LINUX_EXT2_FS_H
diff --git a/include/linux/ext3_jbd.h b/include/linux/ext3_jbd.h
index e8292af9033..c8307c02dd0 100644
--- a/include/linux/ext3_jbd.h
+++ b/include/linux/ext3_jbd.h
@@ -42,15 +42,15 @@
  * superblock only gets updated once, of course, so don't bother
  * counting that again for the quota updates. */
 
-#define EXT3_DATA_TRANS_BLOCKS		(EXT3_SINGLEDATA_TRANS_BLOCKS + \
+#define EXT3_DATA_TRANS_BLOCKS(sb)	(EXT3_SINGLEDATA_TRANS_BLOCKS + \
 					 EXT3_XATTR_TRANS_BLOCKS - 2 + \
-					 2*EXT3_QUOTA_TRANS_BLOCKS)
+					 2*EXT3_QUOTA_TRANS_BLOCKS(sb))
 
 /* Delete operations potentially hit one directory's namespace plus an
  * entire inode, plus arbitrary amounts of bitmap/indirection data.  Be
  * generous.  We can grow the delete transaction later if necessary. */
 
-#define EXT3_DELETE_TRANS_BLOCKS	(2 * EXT3_DATA_TRANS_BLOCKS + 64)
+#define EXT3_DELETE_TRANS_BLOCKS(sb)	(2 * EXT3_DATA_TRANS_BLOCKS(sb) + 64)
 
 /* Define an arbitrary limit for the amount of data we will anticipate
  * writing to any given transaction.  For unbounded transactions such as
@@ -74,14 +74,17 @@
 #ifdef CONFIG_QUOTA
 /* Amount of blocks needed for quota update - we know that the structure was
  * allocated so we need to update only inode+data */
-#define EXT3_QUOTA_TRANS_BLOCKS 2
+#define EXT3_QUOTA_TRANS_BLOCKS(sb) (test_opt(sb, QUOTA) ? 2 : 0)
 /* Amount of blocks needed for quota insert/delete - we do some block writes
  * but inode, sb and group updates are done only once */
-#define EXT3_QUOTA_INIT_BLOCKS (DQUOT_MAX_WRITES*\
-				(EXT3_SINGLEDATA_TRANS_BLOCKS-3)+3)
+#define EXT3_QUOTA_INIT_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_INIT_ALLOC*\
+		(EXT3_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_INIT_REWRITE) : 0)
+#define EXT3_QUOTA_DEL_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_DEL_ALLOC*\
+		(EXT3_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_DEL_REWRITE) : 0)
 #else
-#define EXT3_QUOTA_TRANS_BLOCKS 0
-#define EXT3_QUOTA_INIT_BLOCKS 0
+#define EXT3_QUOTA_TRANS_BLOCKS(sb) 0
+#define EXT3_QUOTA_INIT_BLOCKS(sb) 0
+#define EXT3_QUOTA_DEL_BLOCKS(sb) 0
 #endif
 
 int
-- 
cgit v1.2.3-70-g09d2


From 556a2a45bce1740f035befaa7201e4ad836c7257 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Thu, 23 Jun 2005 22:01:06 -0700
Subject: [PATCH] quota: reiserfs: improve quota credit estimates

Use improved credits estimates for quota operations.  Also reserve space
for a quota operation in a transaction only if filesystem was mounted with
some quota option.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/reiserfs/file.c             |  4 ++--
 fs/reiserfs/inode.c            | 11 ++++++-----
 fs/reiserfs/namei.c            | 25 ++++++++++++++-----------
 fs/reiserfs/super.c            | 35 +++++++++++++++++++++++------------
 include/linux/reiserfs_fs.h    | 15 +++++++++++----
 include/linux/reiserfs_fs_sb.h |  2 ++
 6 files changed, 58 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c
index 2230afff187..12e91209544 100644
--- a/fs/reiserfs/file.c
+++ b/fs/reiserfs/file.c
@@ -201,7 +201,7 @@ static int reiserfs_allocate_blocks_for_region(
     /* If we came here, it means we absolutely need to open a transaction,
        since we need to allocate some blocks */
     reiserfs_write_lock(inode->i_sb); // Journaling stuff and we need that.
-    res = journal_begin(th, inode->i_sb, JOURNAL_PER_BALANCE_CNT * 3 + 1 + 2 * REISERFS_QUOTA_TRANS_BLOCKS); // Wish I know if this number enough
+    res = journal_begin(th, inode->i_sb, JOURNAL_PER_BALANCE_CNT * 3 + 1 + 2 * REISERFS_QUOTA_TRANS_BLOCKS(inode->i_sb)); // Wish I know if this number enough
     if (res)
         goto error_exit;
     reiserfs_update_inode_transaction(inode) ;
@@ -576,7 +576,7 @@ error_exit:
         int err;
         // update any changes we made to blk count
         reiserfs_update_sd(th, inode);
-        err = journal_end(th, inode->i_sb, JOURNAL_PER_BALANCE_CNT * 3 + 1 + 2 * REISERFS_QUOTA_TRANS_BLOCKS);
+        err = journal_end(th, inode->i_sb, JOURNAL_PER_BALANCE_CNT * 3 + 1 + 2 * REISERFS_QUOTA_TRANS_BLOCKS(inode->i_sb));
         if (err)
             res = err;
     }
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index 073425e6e0a..0d5817f8197 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -28,7 +28,7 @@ static int reiserfs_prepare_write(struct file *f, struct page *page,
 void reiserfs_delete_inode (struct inode * inode)
 {
     /* We need blocks for transaction + (user+group) quota update (possibly delete) */
-    int jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 * REISERFS_QUOTA_INIT_BLOCKS;
+    int jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 * REISERFS_QUOTA_INIT_BLOCKS(inode->i_sb);
     struct reiserfs_transaction_handle th ;
   
     reiserfs_write_lock(inode->i_sb);
@@ -591,7 +591,7 @@ int reiserfs_get_block (struct inode * inode, sector_t block,
        XXX in practically impossible worst case direct2indirect()
        can incur (much) more than 3 balancings.
        quota update for user, group */
-    int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 1 + 2 * REISERFS_QUOTA_TRANS_BLOCKS;
+    int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 1 + 2 * REISERFS_QUOTA_TRANS_BLOCKS(inode->i_sb);
     int version;
     int dangle = 1;
     loff_t new_offset = (((loff_t)block) << inode->i_sb->s_blocksize_bits) + 1 ;
@@ -2796,14 +2796,15 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr) {
 
                 if (!error) {
 		    struct reiserfs_transaction_handle th;
+		    int jbegin_count = 2*(REISERFS_QUOTA_INIT_BLOCKS(inode->i_sb)+REISERFS_QUOTA_DEL_BLOCKS(inode->i_sb))+2;
 
 		    /* (user+group)*(old+new) structure - we count quota info and , inode write (sb, inode) */
-		    error = journal_begin(&th, inode->i_sb, 4*REISERFS_QUOTA_INIT_BLOCKS+2);
+		    error = journal_begin(&th, inode->i_sb, jbegin_count);
  		    if (error)
  			goto out;
                     error = DQUOT_TRANSFER(inode, attr) ? -EDQUOT : 0;
 		    if (error) {
-			journal_end(&th, inode->i_sb, 4*REISERFS_QUOTA_INIT_BLOCKS+2);
+			journal_end(&th, inode->i_sb, jbegin_count);
 			goto out;
 		    }
 		    /* Update corresponding info in inode so that everything is in
@@ -2813,7 +2814,7 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr) {
 		    if (attr->ia_valid & ATTR_GID)
 			inode->i_gid = attr->ia_gid;
 		    mark_inode_dirty(inode);
-		    error = journal_end(&th, inode->i_sb, 4*REISERFS_QUOTA_INIT_BLOCKS+2);
+		    error = journal_end(&th, inode->i_sb, jbegin_count);
 		}
         }
         if (!error)
diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c
index 7d4dc5f5aa8..4a333255f27 100644
--- a/fs/reiserfs/namei.c
+++ b/fs/reiserfs/namei.c
@@ -586,7 +586,7 @@ static int reiserfs_create (struct inode * dir, struct dentry *dentry, int mode,
     int retval;
     struct inode * inode;
     /* We need blocks for transaction + (user+group)*(quotas for new inode + update of quota for directory owner) */
-    int jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 * (REISERFS_QUOTA_INIT_BLOCKS+REISERFS_QUOTA_TRANS_BLOCKS);
+    int jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 * (REISERFS_QUOTA_INIT_BLOCKS(dir->i_sb)+REISERFS_QUOTA_TRANS_BLOCKS(dir->i_sb));
     struct reiserfs_transaction_handle th ;
     int locked;
 
@@ -653,7 +653,7 @@ static int reiserfs_mknod (struct inode * dir, struct dentry *dentry, int mode,
     struct inode * inode;
     struct reiserfs_transaction_handle th ;
     /* We need blocks for transaction + (user+group)*(quotas for new inode + update of quota for directory owner) */
-    int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * (REISERFS_QUOTA_INIT_BLOCKS+REISERFS_QUOTA_TRANS_BLOCKS);
+    int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * (REISERFS_QUOTA_INIT_BLOCKS(dir->i_sb)+REISERFS_QUOTA_TRANS_BLOCKS(dir->i_sb));
     int locked;
 
     if (!new_valid_dev(rdev))
@@ -727,7 +727,7 @@ static int reiserfs_mkdir (struct inode * dir, struct dentry *dentry, int mode)
     struct inode * inode;
     struct reiserfs_transaction_handle th ;
     /* We need blocks for transaction + (user+group)*(quotas for new inode + update of quota for directory owner) */
-    int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * (REISERFS_QUOTA_INIT_BLOCKS+REISERFS_QUOTA_TRANS_BLOCKS);
+    int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * (REISERFS_QUOTA_INIT_BLOCKS(dir->i_sb)+REISERFS_QUOTA_TRANS_BLOCKS(dir->i_sb));
     int locked;
 
 #ifdef DISPLACE_NEW_PACKING_LOCALITIES
@@ -829,8 +829,10 @@ static int reiserfs_rmdir (struct inode * dir, struct dentry *dentry)
 
 
     /* we will be doing 2 balancings and update 2 stat data, we change quotas
-     * of the owner of the directory and of the owner of the parent directory */
-    jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 + 2 * (REISERFS_QUOTA_INIT_BLOCKS+REISERFS_QUOTA_TRANS_BLOCKS);
+     * of the owner of the directory and of the owner of the parent directory.
+     * The quota structure is possibly deleted only on last iput => outside
+     * of this transaction */
+    jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 + 4 * REISERFS_QUOTA_TRANS_BLOCKS(dir->i_sb);
 
     reiserfs_write_lock(dir->i_sb);
     retval = journal_begin(&th, dir->i_sb, jbegin_count) ;
@@ -913,9 +915,10 @@ static int reiserfs_unlink (struct inode * dir, struct dentry *dentry)
     inode = dentry->d_inode;
 
     /* in this transaction we can be doing at max two balancings and update
-       two stat datas, we change quotas of the owner of the directory and of
-       the owner of the parent directory */
-    jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 + 2 * (REISERFS_QUOTA_INIT_BLOCKS+REISERFS_QUOTA_TRANS_BLOCKS);
+     * two stat datas, we change quotas of the owner of the directory and of
+     * the owner of the parent directory. The quota structure is possibly
+     * deleted only on iput => outside of this transaction */
+    jbegin_count = JOURNAL_PER_BALANCE_CNT * 2 + 2 + 4 * REISERFS_QUOTA_TRANS_BLOCKS(dir->i_sb);
 
     reiserfs_write_lock(dir->i_sb);
     retval = journal_begin(&th, dir->i_sb, jbegin_count) ;
@@ -1000,7 +1003,7 @@ static int reiserfs_symlink (struct inode * parent_dir,
     struct reiserfs_transaction_handle th ;
     int mode = S_IFLNK | S_IRWXUGO;
     /* We need blocks for transaction + (user+group)*(quotas for new inode + update of quota for directory owner) */
-    int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * (REISERFS_QUOTA_INIT_BLOCKS+REISERFS_QUOTA_TRANS_BLOCKS);
+    int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * (REISERFS_QUOTA_INIT_BLOCKS(parent_dir->i_sb)+REISERFS_QUOTA_TRANS_BLOCKS(parent_dir->i_sb));
 
     if (!(inode = new_inode(parent_dir->i_sb))) {
 	return -ENOMEM ;
@@ -1076,7 +1079,7 @@ static int reiserfs_link (struct dentry * old_dentry, struct inode * dir, struct
     struct inode *inode = old_dentry->d_inode;
     struct reiserfs_transaction_handle th ;
     /* We need blocks for transaction + update of quotas for the owners of the directory */
-    int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * REISERFS_QUOTA_TRANS_BLOCKS;
+    int jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 2 * REISERFS_QUOTA_TRANS_BLOCKS(dir->i_sb);
 
     reiserfs_write_lock(dir->i_sb);
     if (inode->i_nlink >= REISERFS_LINK_MAX) {
@@ -1196,7 +1199,7 @@ static int reiserfs_rename (struct inode * old_dir, struct dentry *old_dentry,
        pointed initially and (5) maybe block containing ".." of
        renamed directory
        quota updates: two parent directories */
-    jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 5 + 4 * REISERFS_QUOTA_TRANS_BLOCKS;
+    jbegin_count = JOURNAL_PER_BALANCE_CNT * 3 + 5 + 4 * REISERFS_QUOTA_TRANS_BLOCKS(old_dir->i_sb);
 
     old_inode = old_dentry->d_inode;
     new_dentry_inode = new_dentry->d_inode;
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index 031577fb41a..660aefca1fd 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -866,8 +866,9 @@ static int reiserfs_parse_options (struct super_block * s, char * options, /* st
 	{"jdev",	.arg_required = 'j', .values = NULL},
 	{"nolargeio",	.arg_required = 'w', .values = NULL},
 	{"commit",	.arg_required = 'c', .values = NULL},
-	{"usrquota",},
-	{"grpquota",},
+	{"usrquota",	.setmask = 1<<REISERFS_QUOTA},
+	{"grpquota",	.setmask = 1<<REISERFS_QUOTA},
+	{"noquota",	.clrmask = 1<<REISERFS_QUOTA},
 	{"errors", 	.arg_required = 'e', .values = error_actions},
 	{"usrjquota",	.arg_required = 'u'|(1<<REISERFS_OPT_ALLOWEMPTY), .values = NULL},
 	{"grpjquota",	.arg_required = 'g'|(1<<REISERFS_OPT_ALLOWEMPTY), .values = NULL},
@@ -964,6 +965,7 @@ static int reiserfs_parse_options (struct super_block * s, char * options, /* st
 		    return 0;
 		}
 		strcpy(REISERFS_SB(s)->s_qf_names[qtype], arg);
+		*mount_options |= 1<<REISERFS_QUOTA;
 	    }
 	    else {
 		if (REISERFS_SB(s)->s_qf_names[qtype]) {
@@ -995,7 +997,13 @@ static int reiserfs_parse_options (struct super_block * s, char * options, /* st
 	reiserfs_warning(s, "reiserfs_parse_options: journalled quota format not specified.");
 	return 0;
     }
+    /* This checking is not precise wrt the quota type but for our purposes it is sufficient */
+    if (!(*mount_options & (1<<REISERFS_QUOTA)) && sb_any_quota_enabled(s)) {
+	reiserfs_warning(s, "reiserfs_parse_options: quota options must be present when quota is turned on.");
+	return 0;
+    }
 #endif
+
     return 1;
 }
 
@@ -1105,6 +1113,7 @@ static int reiserfs_remount (struct super_block * s, int * mount_flags, char * a
   safe_mask |= 1 << REISERFS_ERROR_RO;
   safe_mask |= 1 << REISERFS_ERROR_CONTINUE;
   safe_mask |= 1 << REISERFS_ERROR_PANIC;
+  safe_mask |= 1 << REISERFS_QUOTA;
 
   /* Update the bitmask, taking care to keep
    * the bits we're not allowed to change here */
@@ -1845,11 +1854,11 @@ static int reiserfs_dquot_initialize(struct inode *inode, int type)
 
     /* We may create quota structure so we need to reserve enough blocks */
     reiserfs_write_lock(inode->i_sb);
-    ret = journal_begin(&th, inode->i_sb, 2*REISERFS_QUOTA_INIT_BLOCKS);
+    ret = journal_begin(&th, inode->i_sb, 2*REISERFS_QUOTA_INIT_BLOCKS(inode->i_sb));
     if (ret)
 	goto out;
     ret = dquot_initialize(inode, type);
-    err = journal_end(&th, inode->i_sb, 2*REISERFS_QUOTA_INIT_BLOCKS);
+    err = journal_end(&th, inode->i_sb, 2*REISERFS_QUOTA_INIT_BLOCKS(inode->i_sb));
     if (!ret && err)
 	ret = err;
 out:
@@ -1864,11 +1873,11 @@ static int reiserfs_dquot_drop(struct inode *inode)
 
     /* We may delete quota structure so we need to reserve enough blocks */
     reiserfs_write_lock(inode->i_sb);
-    ret = journal_begin(&th, inode->i_sb, 2*REISERFS_QUOTA_INIT_BLOCKS);
+    ret = journal_begin(&th, inode->i_sb, 2*REISERFS_QUOTA_DEL_BLOCKS(inode->i_sb));
     if (ret)
  	goto out;
     ret = dquot_drop(inode);
-    err = journal_end(&th, inode->i_sb, 2*REISERFS_QUOTA_INIT_BLOCKS);
+    err = journal_end(&th, inode->i_sb, 2*REISERFS_QUOTA_DEL_BLOCKS(inode->i_sb));
     if (!ret && err)
 	ret = err;
 out:
@@ -1882,11 +1891,11 @@ static int reiserfs_write_dquot(struct dquot *dquot)
     int ret, err;
 
     reiserfs_write_lock(dquot->dq_sb);
-    ret = journal_begin(&th, dquot->dq_sb, REISERFS_QUOTA_TRANS_BLOCKS);
+    ret = journal_begin(&th, dquot->dq_sb, REISERFS_QUOTA_TRANS_BLOCKS(dquot->dq_sb));
     if (ret)
 	goto out;
     ret = dquot_commit(dquot);
-    err = journal_end(&th, dquot->dq_sb, REISERFS_QUOTA_TRANS_BLOCKS);
+    err = journal_end(&th, dquot->dq_sb, REISERFS_QUOTA_TRANS_BLOCKS(dquot->dq_sb));
     if (!ret && err)
 	ret = err;
 out:
@@ -1900,11 +1909,11 @@ static int reiserfs_acquire_dquot(struct dquot *dquot)
     int ret, err;
 
     reiserfs_write_lock(dquot->dq_sb);
-    ret = journal_begin(&th, dquot->dq_sb, REISERFS_QUOTA_INIT_BLOCKS);
+    ret = journal_begin(&th, dquot->dq_sb, REISERFS_QUOTA_INIT_BLOCKS(dquot->dq_sb));
     if (ret)
 	goto out;
     ret = dquot_acquire(dquot);
-    err = journal_end(&th, dquot->dq_sb, REISERFS_QUOTA_INIT_BLOCKS);
+    err = journal_end(&th, dquot->dq_sb, REISERFS_QUOTA_INIT_BLOCKS(dquot->dq_sb));
     if (!ret && err)
 	ret = err;
 out:
@@ -1918,11 +1927,11 @@ static int reiserfs_release_dquot(struct dquot *dquot)
     int ret, err;
 
     reiserfs_write_lock(dquot->dq_sb);
-    ret = journal_begin(&th, dquot->dq_sb, REISERFS_QUOTA_INIT_BLOCKS);
+    ret = journal_begin(&th, dquot->dq_sb, REISERFS_QUOTA_DEL_BLOCKS(dquot->dq_sb));
     if (ret)
  	goto out;
     ret = dquot_release(dquot);
-    err = journal_end(&th, dquot->dq_sb, REISERFS_QUOTA_INIT_BLOCKS);
+    err = journal_end(&th, dquot->dq_sb, REISERFS_QUOTA_DEL_BLOCKS(dquot->dq_sb));
     if (!ret && err)
 	ret = err;
 out:
@@ -1978,6 +1987,8 @@ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id, ch
     int err;
     struct nameidata nd;
 
+    if (!(REISERFS_SB(sb)->s_mount_opt & (1<<REISERFS_QUOTA)))
+	return -EINVAL;
     err = path_lookup(path, LOOKUP_FOLLOW, &nd);
     if (err)
         return err;
diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h
index 32148625fc2..4c7c5689ad9 100644
--- a/include/linux/reiserfs_fs.h
+++ b/include/linux/reiserfs_fs.h
@@ -1644,11 +1644,18 @@ struct reiserfs_journal_header {
 #define JOURNAL_MAX_TRANS_AGE 30
 #define JOURNAL_PER_BALANCE_CNT (3 * (MAX_HEIGHT-2) + 9)
 #ifdef CONFIG_QUOTA
-#define REISERFS_QUOTA_TRANS_BLOCKS 2	/* We need to update data and inode (atime) */
-#define REISERFS_QUOTA_INIT_BLOCKS (DQUOT_MAX_WRITES*(JOURNAL_PER_BALANCE_CNT+2)+1)	/* 1 balancing, 1 bitmap, 1 data per write + stat data update */
+/* We need to update data and inode (atime) */
+#define REISERFS_QUOTA_TRANS_BLOCKS(s) (REISERFS_SB(s)->s_mount_opt & (1<<REISERFS_QUOTA) ? 2 : 0)
+/* 1 balancing, 1 bitmap, 1 data per write + stat data update */
+#define REISERFS_QUOTA_INIT_BLOCKS(s) (REISERFS_SB(s)->s_mount_opt & (1<<REISERFS_QUOTA) ? \
+(DQUOT_INIT_ALLOC*(JOURNAL_PER_BALANCE_CNT+2)+DQUOT_INIT_REWRITE+1) : 0)
+/* same as with INIT */
+#define REISERFS_QUOTA_DEL_BLOCKS(s) (REISERFS_SB(s)->s_mount_opt & (1<<REISERFS_QUOTA) ? \
+(DQUOT_DEL_ALLOC*(JOURNAL_PER_BALANCE_CNT+2)+DQUOT_DEL_REWRITE+1) : 0)
 #else
-#define REISERFS_QUOTA_TRANS_BLOCKS 0
-#define REISERFS_QUOTA_INIT_BLOCKS 0
+#define REISERFS_QUOTA_TRANS_BLOCKS(s) 0
+#define REISERFS_QUOTA_INIT_BLOCKS(s) 0
+#define REISERFS_QUOTA_DEL_BLOCKS(s) 0
 #endif
 
 /* both of these can be as low as 1, or as high as you want.  The min is the
diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h
index 37a3a7afbec..31c709d0fe1 100644
--- a/include/linux/reiserfs_fs_sb.h
+++ b/include/linux/reiserfs_fs_sb.h
@@ -467,6 +467,8 @@ enum reiserfs_mount_options {
     REISERFS_ERROR_RO,
     REISERFS_ERROR_CONTINUE,
 
+    REISERFS_QUOTA,		/* Some quota option specified */
+
     REISERFS_TEST1,
     REISERFS_TEST2,
     REISERFS_TEST3,
-- 
cgit v1.2.3-70-g09d2


From 9a8fd5589902153a134111ed7a40f9cca1f83254 Mon Sep 17 00:00:00 2001
From: Chris Zankel <czankel@tensilica.com>
Date: Thu, 23 Jun 2005 22:01:26 -0700
Subject: [PATCH] xtensa: Architecture support for Tensilica Xtensa Part 6

The attached patches provides part 6 of an architecture implementation for the
Tensilica Xtensa CPU series.

Signed-off-by: Chris Zankel <chris@zankel.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-xtensa/a.out.h       |  33 +++
 include/asm-xtensa/atomic.h      | 272 ++++++++++++++++++++
 include/asm-xtensa/bitops.h      | 446 ++++++++++++++++++++++++++++++++
 include/asm-xtensa/bootparam.h   |  61 +++++
 include/asm-xtensa/bug.h         |  41 +++
 include/asm-xtensa/bugs.h        |  22 ++
 include/asm-xtensa/byteorder.h   |  82 ++++++
 include/asm-xtensa/cache.h       |  32 +++
 include/asm-xtensa/cacheflush.h  | 122 +++++++++
 include/asm-xtensa/checksum.h    | 264 +++++++++++++++++++
 include/asm-xtensa/coprocessor.h |  70 +++++
 include/asm-xtensa/cpumask.h     |  16 ++
 include/asm-xtensa/cputime.h     |   6 +
 include/asm-xtensa/current.h     |  38 +++
 include/asm-xtensa/delay.h       |  50 ++++
 include/asm-xtensa/div64.h       |  19 ++
 include/asm-xtensa/dma-mapping.h | 182 +++++++++++++
 include/asm-xtensa/dma.h         |  61 +++++
 include/asm-xtensa/elf.h         | 222 ++++++++++++++++
 include/asm-xtensa/errno.h       | 142 +++++++++++
 include/asm-xtensa/fcntl.h       | 101 ++++++++
 include/asm-xtensa/fixmap.h      | 252 ++++++++++++++++++
 include/asm-xtensa/hardirq.h     |  28 ++
 include/asm-xtensa/hdreg.h       |  17 ++
 include/asm-xtensa/highmem.h     |  17 ++
 include/asm-xtensa/hw_irq.h      |  18 ++
 include/asm-xtensa/ide.h         |  36 +++
 include/asm-xtensa/io.h          | 197 ++++++++++++++
 include/asm-xtensa/ioctl.h       |  83 ++++++
 include/asm-xtensa/ioctls.h      | 112 ++++++++
 include/asm-xtensa/ipc.h         |  34 +++
 include/asm-xtensa/ipcbuf.h      |  37 +++
 include/asm-xtensa/irq.h         |  37 +++
 include/asm-xtensa/kmap_types.h  |  31 +++
 include/asm-xtensa/linkage.h     |  16 ++
 include/asm-xtensa/local.h       |  16 ++
 include/asm-xtensa/mman.h        |  80 ++++++
 include/asm-xtensa/mmu.h         |  17 ++
 include/asm-xtensa/mmu_context.h | 330 ++++++++++++++++++++++++
 include/asm-xtensa/module.h      |  25 ++
 include/asm-xtensa/msgbuf.h      |  48 ++++
 include/asm-xtensa/namei.h       |  26 ++
 include/asm-xtensa/page.h        | 133 ++++++++++
 include/asm-xtensa/page.h.n      | 135 ++++++++++
 include/asm-xtensa/param.h       |  34 +++
 include/asm-xtensa/pci-bridge.h  |  88 +++++++
 include/asm-xtensa/pci.h         |  89 +++++++
 include/asm-xtensa/percpu.h      |  16 ++
 include/asm-xtensa/pgalloc.h     | 116 +++++++++
 include/asm-xtensa/pgtable.h     | 468 ++++++++++++++++++++++++++++++++++
 include/asm-xtensa/poll.h        |  37 +++
 include/asm-xtensa/posix_types.h | 123 +++++++++
 include/asm-xtensa/processor.h   | 205 +++++++++++++++
 include/asm-xtensa/ptrace.h      | 135 ++++++++++
 include/asm-xtensa/resource.h    |  16 ++
 include/asm-xtensa/rmap.h        |  16 ++
 include/asm-xtensa/rwsem.h       | 175 +++++++++++++
 include/asm-xtensa/scatterlist.h |  34 +++
 include/asm-xtensa/sections.h    |  16 ++
 include/asm-xtensa/segment.h     |  16 ++
 include/asm-xtensa/semaphore.h   | 129 ++++++++++
 include/asm-xtensa/sembuf.h      |  44 ++++
 include/asm-xtensa/serial.h      |  18 ++
 include/asm-xtensa/setup.h       |  16 ++
 include/asm-xtensa/shmbuf.h      |  50 ++++
 include/asm-xtensa/shmparam.h    |  23 ++
 include/asm-xtensa/sigcontext.h  |  44 ++++
 include/asm-xtensa/siginfo.h     |  16 ++
 include/asm-xtensa/signal.h      | 187 ++++++++++++++
 include/asm-xtensa/smp.h         |  27 ++
 include/asm-xtensa/socket.h      |  61 +++++
 include/asm-xtensa/sockios.h     |  30 +++
 include/asm-xtensa/spinlock.h    |  16 ++
 include/asm-xtensa/stat.h        | 105 ++++++++
 include/asm-xtensa/statfs.h      |  17 ++
 include/asm-xtensa/string.h      | 124 +++++++++
 include/asm-xtensa/system.h      | 252 ++++++++++++++++++
 include/asm-xtensa/termbits.h    | 194 ++++++++++++++
 include/asm-xtensa/termios.h     | 122 +++++++++
 include/asm-xtensa/thread_info.h | 146 +++++++++++
 include/asm-xtensa/timex.h       |  94 +++++++
 include/asm-xtensa/tlb.h         |  25 ++
 include/asm-xtensa/tlbflush.h    | 200 +++++++++++++++
 include/asm-xtensa/topology.h    |  16 ++
 include/asm-xtensa/types.h       |  66 +++++
 include/asm-xtensa/uaccess.h     | 532 ++++++++++++++++++++++++++++++++++++++
 include/asm-xtensa/ucontext.h    |  22 ++
 include/asm-xtensa/unaligned.h   |  28 ++
 include/asm-xtensa/unistd.h      | 537 +++++++++++++++++++++++++++++++++++++++
 include/asm-xtensa/user.h        |  20 ++
 include/asm-xtensa/vga.h         |  19 ++
 include/asm-xtensa/xor.h         |  16 ++
 92 files changed, 8787 insertions(+)
 create mode 100644 include/asm-xtensa/a.out.h
 create mode 100644 include/asm-xtensa/atomic.h
 create mode 100644 include/asm-xtensa/bitops.h
 create mode 100644 include/asm-xtensa/bootparam.h
 create mode 100644 include/asm-xtensa/bug.h
 create mode 100644 include/asm-xtensa/bugs.h
 create mode 100644 include/asm-xtensa/byteorder.h
 create mode 100644 include/asm-xtensa/cache.h
 create mode 100644 include/asm-xtensa/cacheflush.h
 create mode 100644 include/asm-xtensa/checksum.h
 create mode 100644 include/asm-xtensa/coprocessor.h
 create mode 100644 include/asm-xtensa/cpumask.h
 create mode 100644 include/asm-xtensa/cputime.h
 create mode 100644 include/asm-xtensa/current.h
 create mode 100644 include/asm-xtensa/delay.h
 create mode 100644 include/asm-xtensa/div64.h
 create mode 100644 include/asm-xtensa/dma-mapping.h
 create mode 100644 include/asm-xtensa/dma.h
 create mode 100644 include/asm-xtensa/elf.h
 create mode 100644 include/asm-xtensa/errno.h
 create mode 100644 include/asm-xtensa/fcntl.h
 create mode 100644 include/asm-xtensa/fixmap.h
 create mode 100644 include/asm-xtensa/hardirq.h
 create mode 100644 include/asm-xtensa/hdreg.h
 create mode 100644 include/asm-xtensa/highmem.h
 create mode 100644 include/asm-xtensa/hw_irq.h
 create mode 100644 include/asm-xtensa/ide.h
 create mode 100644 include/asm-xtensa/io.h
 create mode 100644 include/asm-xtensa/ioctl.h
 create mode 100644 include/asm-xtensa/ioctls.h
 create mode 100644 include/asm-xtensa/ipc.h
 create mode 100644 include/asm-xtensa/ipcbuf.h
 create mode 100644 include/asm-xtensa/irq.h
 create mode 100644 include/asm-xtensa/kmap_types.h
 create mode 100644 include/asm-xtensa/linkage.h
 create mode 100644 include/asm-xtensa/local.h
 create mode 100644 include/asm-xtensa/mman.h
 create mode 100644 include/asm-xtensa/mmu.h
 create mode 100644 include/asm-xtensa/mmu_context.h
 create mode 100644 include/asm-xtensa/module.h
 create mode 100644 include/asm-xtensa/msgbuf.h
 create mode 100644 include/asm-xtensa/namei.h
 create mode 100644 include/asm-xtensa/page.h
 create mode 100644 include/asm-xtensa/page.h.n
 create mode 100644 include/asm-xtensa/param.h
 create mode 100644 include/asm-xtensa/pci-bridge.h
 create mode 100644 include/asm-xtensa/pci.h
 create mode 100644 include/asm-xtensa/percpu.h
 create mode 100644 include/asm-xtensa/pgalloc.h
 create mode 100644 include/asm-xtensa/pgtable.h
 create mode 100644 include/asm-xtensa/poll.h
 create mode 100644 include/asm-xtensa/posix_types.h
 create mode 100644 include/asm-xtensa/processor.h
 create mode 100644 include/asm-xtensa/ptrace.h
 create mode 100644 include/asm-xtensa/resource.h
 create mode 100644 include/asm-xtensa/rmap.h
 create mode 100644 include/asm-xtensa/rwsem.h
 create mode 100644 include/asm-xtensa/scatterlist.h
 create mode 100644 include/asm-xtensa/sections.h
 create mode 100644 include/asm-xtensa/segment.h
 create mode 100644 include/asm-xtensa/semaphore.h
 create mode 100644 include/asm-xtensa/sembuf.h
 create mode 100644 include/asm-xtensa/serial.h
 create mode 100644 include/asm-xtensa/setup.h
 create mode 100644 include/asm-xtensa/shmbuf.h
 create mode 100644 include/asm-xtensa/shmparam.h
 create mode 100644 include/asm-xtensa/sigcontext.h
 create mode 100644 include/asm-xtensa/siginfo.h
 create mode 100644 include/asm-xtensa/signal.h
 create mode 100644 include/asm-xtensa/smp.h
 create mode 100644 include/asm-xtensa/socket.h
 create mode 100644 include/asm-xtensa/sockios.h
 create mode 100644 include/asm-xtensa/spinlock.h
 create mode 100644 include/asm-xtensa/stat.h
 create mode 100644 include/asm-xtensa/statfs.h
 create mode 100644 include/asm-xtensa/string.h
 create mode 100644 include/asm-xtensa/system.h
 create mode 100644 include/asm-xtensa/termbits.h
 create mode 100644 include/asm-xtensa/termios.h
 create mode 100644 include/asm-xtensa/thread_info.h
 create mode 100644 include/asm-xtensa/timex.h
 create mode 100644 include/asm-xtensa/tlb.h
 create mode 100644 include/asm-xtensa/tlbflush.h
 create mode 100644 include/asm-xtensa/topology.h
 create mode 100644 include/asm-xtensa/types.h
 create mode 100644 include/asm-xtensa/uaccess.h
 create mode 100644 include/asm-xtensa/ucontext.h
 create mode 100644 include/asm-xtensa/unaligned.h
 create mode 100644 include/asm-xtensa/unistd.h
 create mode 100644 include/asm-xtensa/user.h
 create mode 100644 include/asm-xtensa/vga.h
 create mode 100644 include/asm-xtensa/xor.h

(limited to 'include')

diff --git a/include/asm-xtensa/a.out.h b/include/asm-xtensa/a.out.h
new file mode 100644
index 00000000000..3be701dfe09
--- /dev/null
+++ b/include/asm-xtensa/a.out.h
@@ -0,0 +1,33 @@
+/*
+ * include/asm-xtensa/addrspace.h
+ *
+ * Dummy a.out file. Xtensa does not support the a.out format, but the kernel
+ * seems to depend on it.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_A_OUT_H
+#define _XTENSA_A_OUT_H
+
+/* Note: the kernel needs the a.out definitions, even if only ELF is used. */
+
+#define STACK_TOP	TASK_SIZE
+
+struct exec
+{
+  unsigned long a_info;
+  unsigned a_text;
+  unsigned a_data;
+  unsigned a_bss;
+  unsigned a_syms;
+  unsigned a_entry;
+  unsigned a_trsize;
+  unsigned a_drsize;
+};
+
+#endif /* _XTENSA_A_OUT_H */
diff --git a/include/asm-xtensa/atomic.h b/include/asm-xtensa/atomic.h
new file mode 100644
index 00000000000..d72bcb32ba4
--- /dev/null
+++ b/include/asm-xtensa/atomic.h
@@ -0,0 +1,272 @@
+/*
+ * include/asm-xtensa/atomic.h
+ *
+ * Atomic operations that C can't guarantee us.  Useful for resource counting..
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_ATOMIC_H
+#define _XTENSA_ATOMIC_H
+
+#include <linux/config.h>
+#include <linux/stringify.h>
+
+typedef struct { volatile int counter; } atomic_t;
+
+#ifdef __KERNEL__
+#include <asm/processor.h>
+#include <asm/system.h>
+
+#define ATOMIC_INIT(i)	( (atomic_t) { (i) } )
+
+/*
+ * This Xtensa implementation assumes that the right mechanism
+ * for exclusion is for locking interrupts to level 1.
+ *
+ * Locking interrupts looks like this:
+ *
+ *    rsil a15, 1
+ *    <code>
+ *    wsr  a15, PS
+ *    rsync
+ *
+ * Note that a15 is used here because the register allocation
+ * done by the compiler is not guaranteed and a window overflow
+ * may not occur between the rsil and wsr instructions. By using
+ * a15 in the rsil, the machine is guaranteed to be in a state
+ * where no register reference will cause an overflow.
+ */
+
+/**
+ * atomic_read - read atomic variable
+ * @v: pointer of type atomic_t
+ *
+ * Atomically reads the value of @v.
+ */
+#define atomic_read(v)		((v)->counter)
+
+/**
+ * atomic_set - set atomic variable
+ * @v: pointer of type atomic_t
+ * @i: required value
+ *
+ * Atomically sets the value of @v to @i.
+ */
+#define atomic_set(v,i)		((v)->counter = (i))
+
+/**
+ * atomic_add - add integer to atomic variable
+ * @i: integer value to add
+ * @v: pointer of type atomic_t
+ *
+ * Atomically adds @i to @v.
+ */
+extern __inline__ void atomic_add(int i, atomic_t * v)
+{
+    unsigned int vval;
+
+    __asm__ __volatile__(
+	"rsil    a15, "__stringify(LOCKLEVEL)"\n\t"
+	"l32i    %0, %2, 0              \n\t"
+	"add     %0, %0, %1             \n\t"
+	"s32i    %0, %2, 0              \n\t"
+	"wsr     a15, "__stringify(PS)"       \n\t"
+	"rsync                          \n"
+	: "=&a" (vval)
+	: "a" (i), "a" (v)
+	: "a15", "memory"
+	);
+}
+
+/**
+ * atomic_sub - subtract the atomic variable
+ * @i: integer value to subtract
+ * @v: pointer of type atomic_t
+ *
+ * Atomically subtracts @i from @v.
+ */
+extern __inline__ void atomic_sub(int i, atomic_t *v)
+{
+    unsigned int vval;
+
+    __asm__ __volatile__(
+	"rsil    a15, "__stringify(LOCKLEVEL)"\n\t"
+	"l32i    %0, %2, 0              \n\t"
+	"sub     %0, %0, %1             \n\t"
+	"s32i    %0, %2, 0              \n\t"
+	"wsr     a15, "__stringify(PS)"       \n\t"
+	"rsync                          \n"
+	: "=&a" (vval)
+	: "a" (i), "a" (v)
+	: "a15", "memory"
+	);
+}
+
+/*
+ * We use atomic_{add|sub}_return to define other functions.
+ */
+
+extern __inline__ int atomic_add_return(int i, atomic_t * v)
+{
+     unsigned int vval;
+
+    __asm__ __volatile__(
+	"rsil    a15,"__stringify(LOCKLEVEL)"\n\t"
+	"l32i    %0, %2, 0             \n\t"
+	"add     %0, %0, %1            \n\t"
+	"s32i    %0, %2, 0             \n\t"
+	"wsr     a15, "__stringify(PS)"      \n\t"
+	"rsync                         \n"
+	: "=&a" (vval)
+	: "a" (i), "a" (v)
+	: "a15", "memory"
+	);
+
+    return vval;
+}
+
+extern __inline__ int atomic_sub_return(int i, atomic_t * v)
+{
+    unsigned int vval;
+
+    __asm__ __volatile__(
+	"rsil    a15,"__stringify(LOCKLEVEL)"\n\t"
+	"l32i    %0, %2, 0             \n\t"
+	"sub     %0, %0, %1            \n\t"
+	"s32i    %0, %2, 0             \n\t"
+	"wsr     a15, "__stringify(PS)"       \n\t"
+	"rsync                         \n"
+	: "=&a" (vval)
+	: "a" (i), "a" (v)
+	: "a15", "memory"
+	);
+
+    return vval;
+}
+
+/**
+ * atomic_sub_and_test - subtract value from variable and test result
+ * @i: integer value to subtract
+ * @v: pointer of type atomic_t
+ *
+ * Atomically subtracts @i from @v and returns
+ * true if the result is zero, or false for all
+ * other cases.
+ */
+#define atomic_sub_and_test(i,v) (atomic_sub_return((i),(v)) == 0)
+
+/**
+ * atomic_inc - increment atomic variable
+ * @v: pointer of type atomic_t
+ *
+ * Atomically increments @v by 1.
+ */
+#define atomic_inc(v) atomic_add(1,(v))
+
+/**
+ * atomic_inc - increment atomic variable
+ * @v: pointer of type atomic_t
+ *
+ * Atomically increments @v by 1.
+ */
+#define atomic_inc_return(v) atomic_add_return(1,(v))
+
+/**
+ * atomic_dec - decrement atomic variable
+ * @v: pointer of type atomic_t
+ *
+ * Atomically decrements @v by 1.
+ */
+#define atomic_dec(v) atomic_sub(1,(v))
+
+/**
+ * atomic_dec_return - decrement atomic variable
+ * @v: pointer of type atomic_t
+ *
+ * Atomically decrements @v by 1.
+ */
+#define atomic_dec_return(v) atomic_sub_return(1,(v))
+
+/**
+ * atomic_dec_and_test - decrement and test
+ * @v: pointer of type atomic_t
+ *
+ * Atomically decrements @v by 1 and
+ * returns true if the result is 0, or false for all other
+ * cases.
+ */
+#define atomic_dec_and_test(v) (atomic_sub_return(1,(v)) == 0)
+
+/**
+ * atomic_inc_and_test - increment and test
+ * @v: pointer of type atomic_t
+ *
+ * Atomically increments @v by 1
+ * and returns true if the result is zero, or false for all
+ * other cases.
+ */
+#define atomic_inc_and_test(v) (atomic_add_return(1,(v)) == 0)
+
+/**
+ * atomic_add_negative - add and test if negative
+ * @v: pointer of type atomic_t
+ * @i: integer value to add
+ *
+ * Atomically adds @i to @v and returns true
+ * if the result is negative, or false when
+ * result is greater than or equal to zero.
+ */
+#define atomic_add_negative(i,v) (atomic_add_return((i),(v)) < 0)
+
+
+extern __inline__ void atomic_clear_mask(unsigned int mask, atomic_t *v)
+{
+    unsigned int all_f = -1;
+    unsigned int vval;
+
+    __asm__ __volatile__(
+	"rsil    a15,"__stringify(LOCKLEVEL)"\n\t"
+	"l32i    %0, %2, 0             \n\t"
+	"xor     %1, %4, %3            \n\t"
+	"and     %0, %0, %4            \n\t"
+	"s32i    %0, %2, 0             \n\t"
+	"wsr     a15, "__stringify(PS)"      \n\t"
+	"rsync                         \n"
+	: "=&a" (vval), "=a" (mask)
+	: "a" (v), "a" (all_f), "1" (mask)
+	: "a15", "memory"
+	);
+}
+
+extern __inline__ void atomic_set_mask(unsigned int mask, atomic_t *v)
+{
+    unsigned int vval;
+
+    __asm__ __volatile__(
+	"rsil    a15,"__stringify(LOCKLEVEL)"\n\t"
+	"l32i    %0, %2, 0             \n\t"
+	"or      %0, %0, %1            \n\t"
+	"s32i    %0, %2, 0             \n\t"
+	"wsr     a15, "__stringify(PS)"       \n\t"
+	"rsync                         \n"
+	: "=&a" (vval)
+	: "a" (mask), "a" (v)
+	: "a15", "memory"
+	);
+}
+
+/* Atomic operations are already serializing */
+#define smp_mb__before_atomic_dec()	barrier()
+#define smp_mb__after_atomic_dec()	barrier()
+#define smp_mb__before_atomic_inc()	barrier()
+#define smp_mb__after_atomic_inc()	barrier()
+
+#endif /* __KERNEL__ */
+
+#endif /* _XTENSA_ATOMIC_H */
+
diff --git a/include/asm-xtensa/bitops.h b/include/asm-xtensa/bitops.h
new file mode 100644
index 00000000000..d395ef226c3
--- /dev/null
+++ b/include/asm-xtensa/bitops.h
@@ -0,0 +1,446 @@
+/*
+ * include/asm-xtensa/bitops.h
+ *
+ * Atomic operations that C can't guarantee us.Useful for resource counting etc.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_BITOPS_H
+#define _XTENSA_BITOPS_H
+
+#ifdef __KERNEL__
+
+#include <asm/processor.h>
+#include <asm/byteorder.h>
+#include <asm/system.h>
+
+#ifdef CONFIG_SMP
+# error SMP not supported on this architecture
+#endif
+
+static __inline__ void set_bit(int nr, volatile void * addr)
+{
+	unsigned long mask = 1 << (nr & 0x1f);
+	unsigned long *a = ((unsigned long *)addr) + (nr >> 5);
+	unsigned long flags;
+
+	local_irq_save(flags);
+	*a |= mask;
+	local_irq_restore(flags);
+}
+
+static __inline__ void __set_bit(int nr, volatile unsigned long * addr)
+{
+	unsigned long mask = 1 << (nr & 0x1f);
+	unsigned long *a = ((unsigned long *)addr) + (nr >> 5);
+
+	*a |= mask;
+}
+
+static __inline__ void clear_bit(int nr, volatile void * addr)
+{
+	unsigned long mask = 1 << (nr & 0x1f);
+	unsigned long *a = ((unsigned long *)addr) + (nr >> 5);
+	unsigned long flags;
+
+	local_irq_save(flags);
+	*a &= ~mask;
+	local_irq_restore(flags);
+}
+
+static __inline__ void __clear_bit(int nr, volatile unsigned long *addr)
+{
+	unsigned long mask = 1 << (nr & 0x1f);
+	unsigned long *a = ((unsigned long *)addr) + (nr >> 5);
+
+	*a &= ~mask;
+}
+
+/*
+ * clear_bit() doesn't provide any barrier for the compiler.
+ */
+
+#define smp_mb__before_clear_bit()	barrier()
+#define smp_mb__after_clear_bit()	barrier()
+
+static __inline__ void change_bit(int nr, volatile void * addr)
+{
+	unsigned long mask = 1 << (nr & 0x1f);
+	unsigned long *a = ((unsigned long *)addr) + (nr >> 5);
+	unsigned long flags;
+
+	local_irq_save(flags);
+	*a ^= mask;
+	local_irq_restore(flags);
+}
+
+static __inline__ void __change_bit(int nr, volatile void * addr)
+{
+	unsigned long mask = 1 << (nr & 0x1f);
+	unsigned long *a = ((unsigned long *)addr) + (nr >> 5);
+
+	*a ^= mask;
+}
+
+static __inline__ int test_and_set_bit(int nr, volatile void * addr)
+{
+  	unsigned long retval;
+	unsigned long mask = 1 << (nr & 0x1f);
+	unsigned long *a = ((unsigned long *)addr) + (nr >> 5);
+	unsigned long flags;
+
+	local_irq_save(flags);
+	retval = (mask & *a) != 0;
+	*a |= mask;
+	local_irq_restore(flags);
+
+	return retval;
+}
+
+static __inline__ int __test_and_set_bit(int nr, volatile void * addr)
+{
+  	unsigned long retval;
+	unsigned long mask = 1 << (nr & 0x1f);
+	unsigned long *a = ((unsigned long *)addr) + (nr >> 5);
+
+	retval = (mask & *a) != 0;
+	*a |= mask;
+
+	return retval;
+}
+
+static __inline__ int test_and_clear_bit(int nr, volatile void * addr)
+{
+  	unsigned long retval;
+	unsigned long mask = 1 << (nr & 0x1f);
+	unsigned long *a = ((unsigned long *)addr) + (nr >> 5);
+	unsigned long flags;
+
+	local_irq_save(flags);
+	retval = (mask & *a) != 0;
+	*a &= ~mask;
+	local_irq_restore(flags);
+
+	return retval;
+}
+
+static __inline__ int __test_and_clear_bit(int nr, volatile void * addr)
+{
+	unsigned long mask = 1 << (nr & 0x1f);
+	unsigned long *a = ((unsigned long *)addr) + (nr >> 5);
+  	unsigned long old = *a;
+
+	*a = old & ~mask;
+	return (old & mask) != 0;
+}
+
+static __inline__ int test_and_change_bit(int nr, volatile void * addr)
+{
+  	unsigned long retval;
+	unsigned long mask = 1 << (nr & 0x1f);
+	unsigned long *a = ((unsigned long *)addr) + (nr >> 5);
+	unsigned long flags;
+
+	local_irq_save(flags);
+
+	retval = (mask & *a) != 0;
+	*a ^= mask;
+	local_irq_restore(flags);
+
+	return retval;
+}
+
+/*
+ * non-atomic version; can be reordered
+ */
+
+static __inline__ int __test_and_change_bit(int nr, volatile void *addr)
+{
+	unsigned long mask = 1 << (nr & 0x1f);
+	unsigned long *a = ((unsigned long *)addr) + (nr >> 5);
+	unsigned long old = *a;
+
+	*a = old ^ mask;
+	return (old & mask) != 0;
+}
+
+static __inline__ int test_bit(int nr, const volatile void *addr)
+{
+	return 1UL & (((const volatile unsigned int *)addr)[nr>>5] >> (nr&31));
+}
+
+#if XCHAL_HAVE_NSAU
+
+static __inline__ int __cntlz (unsigned long x)
+{
+	int lz;
+	asm ("nsau %0, %1" : "=r" (lz) : "r" (x));
+	return 31 - lz;
+}
+
+#else
+
+static __inline__ int __cntlz (unsigned long x)
+{
+	unsigned long sum, x1, x2, x4, x8, x16;
+	x1  = x & 0xAAAAAAAA;
+	x2  = x & 0xCCCCCCCC;
+	x4  = x & 0xF0F0F0F0;
+	x8  = x & 0xFF00FF00;
+	x16 = x & 0xFFFF0000;
+	sum = x2 ? 2 : 0;
+	sum += (x16 != 0) * 16;
+	sum += (x8 != 0) * 8;
+	sum += (x4 != 0) * 4;
+	sum += (x1 != 0);
+
+	return sum;
+}
+
+#endif
+
+/*
+ * ffz: Find first zero in word. Undefined if no zero exists.
+ * bit 0 is the LSB of addr; bit 32 is the LSB of (addr+1).
+ */
+
+static __inline__ int ffz(unsigned long x)
+{
+	if ((x = ~x) == 0)
+		return 32;
+	return __cntlz(x & -x);
+}
+
+/*
+ * __ffs: Find first bit set in word. Return 0 for bit 0
+ */
+
+static __inline__ int __ffs(unsigned long x)
+{
+	return __cntlz(x & -x);
+}
+
+/*
+ * ffs: Find first bit set in word. This is defined the same way as
+ * the libc and compiler builtin ffs routines, therefore
+ * differs in spirit from the above ffz (man ffs).
+ */
+
+static __inline__ int ffs(unsigned long x)
+{
+	return __cntlz(x & -x) + 1;
+}
+
+/*
+ * fls: Find last (most-significant) bit set in word.
+ * Note fls(0) = 0, fls(1) = 1, fls(0x80000000) = 32.
+ */
+
+static __inline__ int fls (unsigned int x)
+{
+	return __cntlz(x);
+}
+
+static __inline__ int
+find_next_bit(const unsigned long *addr, int size, int offset)
+{
+	const unsigned long *p = addr + (offset >> 5);
+	unsigned long result = offset & ~31UL;
+	unsigned long tmp;
+
+	if (offset >= size)
+		return size;
+	size -= result;
+	offset &= 31UL;
+	if (offset) {
+		tmp = *p++;
+		tmp &= ~0UL << offset;
+		if (size < 32)
+			goto found_first;
+		if (tmp)
+			goto found_middle;
+		size -= 32;
+		result += 32;
+	}
+	while (size >= 32) {
+		if ((tmp = *p++) != 0)
+			goto found_middle;
+		result += 32;
+		size -= 32;
+	}
+	if (!size)
+		return result;
+	tmp = *p;
+
+found_first:
+	tmp &= ~0UL >> (32 - size);
+	if (tmp == 0UL)	/* Are any bits set? */
+		return result + size;	/* Nope. */
+found_middle:
+	return result + __ffs(tmp);
+}
+
+/**
+ * find_first_bit - find the first set bit in a memory region
+ * @addr: The address to start the search at
+ * @size: The maximum size to search
+ *
+ * Returns the bit-number of the first set bit, not the number of the byte
+ * containing a bit.
+ */
+
+#define find_first_bit(addr, size) \
+        find_next_bit((addr), (size), 0)
+
+static __inline__ int
+find_next_zero_bit(const unsigned long *addr, int size, int offset)
+{
+	const unsigned long *p = addr + (offset >> 5);
+	unsigned long result = offset & ~31UL;
+	unsigned long tmp;
+
+	if (offset >= size)
+		return size;
+	size -= result;
+	offset &= 31UL;
+	if (offset) {
+		tmp = *p++;
+		tmp |= ~0UL >> (32-offset);
+		if (size < 32)
+			goto found_first;
+		if (~tmp)
+			goto found_middle;
+		size -= 32;
+		result += 32;
+	}
+	while (size & ~31UL) {
+		if (~(tmp = *p++))
+			goto found_middle;
+		result += 32;
+		size -= 32;
+	}
+	if (!size)
+		return result;
+	tmp = *p;
+
+found_first:
+	tmp |= ~0UL << size;
+found_middle:
+	return result + ffz(tmp);
+}
+
+#define find_first_zero_bit(addr, size) \
+        find_next_zero_bit((addr), (size), 0)
+
+#ifdef __XTENSA_EL__
+# define ext2_set_bit(nr,addr) __test_and_set_bit((nr), (addr))
+# define ext2_set_bit_atomic(lock,nr,addr) test_and_set_bit((nr),(addr))
+# define ext2_clear_bit(nr,addr) __test_and_clear_bit((nr), (addr))
+# define ext2_clear_bit_atomic(lock,nr,addr) test_and_clear_bit((nr),(addr))
+# define ext2_test_bit(nr,addr) test_bit((nr), (addr))
+# define ext2_find_first_zero_bit(addr, size) find_first_zero_bit((addr),(size))
+# define ext2_find_next_zero_bit(addr, size, offset) \
+                find_next_zero_bit((addr), (size), (offset))
+#elif defined(__XTENSA_EB__)
+# define ext2_set_bit(nr,addr) __test_and_set_bit((nr) ^ 0x18, (addr))
+# define ext2_set_bit_atomic(lock,nr,addr) test_and_set_bit((nr) ^ 0x18, (addr))
+# define ext2_clear_bit(nr,addr) __test_and_clear_bit((nr) ^ 18, (addr))
+# define ext2_clear_bit_atomic(lock,nr,addr) test_and_clear_bit((nr)^0x18,(addr))
+# define ext2_test_bit(nr,addr) test_bit((nr) ^ 0x18, (addr))
+# define ext2_find_first_zero_bit(addr, size) \
+        ext2_find_next_zero_bit((addr), (size), 0)
+
+static __inline__ unsigned long ext2_find_next_zero_bit(void *addr, unsigned long size, unsigned long offset)
+{
+	unsigned long *p = ((unsigned long *) addr) + (offset >> 5);
+	unsigned long result = offset & ~31UL;
+	unsigned long tmp;
+
+	if (offset >= size)
+		return size;
+	size -= result;
+	offset &= 31UL;
+	if(offset) {
+		/* We hold the little endian value in tmp, but then the
+		 * shift is illegal. So we could keep a big endian value
+		 * in tmp, like this:
+		 *
+		 * tmp = __swab32(*(p++));
+		 * tmp |= ~0UL >> (32-offset);
+		 *
+		 * but this would decrease preformance, so we change the
+		 * shift:
+		 */
+		tmp = *(p++);
+		tmp |= __swab32(~0UL >> (32-offset));
+		if(size < 32)
+			goto found_first;
+		if(~tmp)
+			goto found_middle;
+		size -= 32;
+		result += 32;
+	}
+	while(size & ~31UL) {
+		if(~(tmp = *(p++)))
+			goto found_middle;
+		result += 32;
+		size -= 32;
+	}
+	if(!size)
+		return result;
+	tmp = *p;
+
+found_first:
+	/* tmp is little endian, so we would have to swab the shift,
+	 * see above. But then we have to swab tmp below for ffz, so
+	 * we might as well do this here.
+	 */
+	return result + ffz(__swab32(tmp) | (~0UL << size));
+found_middle:
+	return result + ffz(__swab32(tmp));
+}
+
+#else
+# error processor byte order undefined!
+#endif
+
+
+#define hweight32(x)	generic_hweight32(x)
+#define hweight16(x)	generic_hweight16(x)
+#define hweight8(x)	generic_hweight8(x)
+
+/*
+ * Find the first bit set in a 140-bit bitmap.
+ * The first 100 bits are unlikely to be set.
+ */
+
+static inline int sched_find_first_bit(const unsigned long *b)
+{
+	if (unlikely(b[0]))
+		return __ffs(b[0]);
+	if (unlikely(b[1]))
+		return __ffs(b[1]) + 32;
+	if (unlikely(b[2]))
+		return __ffs(b[2]) + 64;
+	if (b[3])
+		return __ffs(b[3]) + 96;
+	return __ffs(b[4]) + 128;
+}
+
+
+/* Bitmap functions for the minix filesystem.  */
+
+#define minix_test_and_set_bit(nr,addr) test_and_set_bit(nr,addr)
+#define minix_set_bit(nr,addr) set_bit(nr,addr)
+#define minix_test_and_clear_bit(nr,addr) test_and_clear_bit(nr,addr)
+#define minix_test_bit(nr,addr) test_bit(nr,addr)
+#define minix_find_first_zero_bit(addr,size) find_first_zero_bit(addr,size)
+
+#endif	/* __KERNEL__ */
+
+#endif	/* _XTENSA_BITOPS_H */
diff --git a/include/asm-xtensa/bootparam.h b/include/asm-xtensa/bootparam.h
new file mode 100644
index 00000000000..9983f2c1b7e
--- /dev/null
+++ b/include/asm-xtensa/bootparam.h
@@ -0,0 +1,61 @@
+/*
+ * include/asm-xtensa/bootparam.h
+ *
+ * Definition of the Linux/Xtensa boot parameter structure
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005  Tensilica Inc.
+ *
+ * (Concept borrowed from the 68K port)
+ */
+
+#ifndef _XTENSA_BOOTPARAM_H
+#define _XTENSA_BOOTPARAM_H
+
+#define BP_VERSION 0x0001
+
+#define BP_TAG_COMMAND_LINE	0x1001	/* command line (0-terminated string)*/
+#define BP_TAG_INITRD		0x1002	/* ramdisk addr and size (bp_meminfo) */
+#define BP_TAG_MEMORY		0x1003	/* memory addr and size (bp_meminfo) */
+#define BP_TAG_SERIAL_BAUSRATE	0x1004	/* baud rate of current console. */
+#define BP_TAG_SERIAL_PORT	0x1005	/* serial device of current console */
+
+#define BP_TAG_FIRST		0x7B0B  /* first tag with a version number */
+#define BP_TAG_LAST 		0x7E0B	/* last tag */
+
+#ifndef __ASSEMBLY__
+
+/* All records are aligned to 4 bytes */
+
+typedef struct bp_tag {
+  unsigned short id;		/* tag id */
+  unsigned short size;		/* size of this record excluding the structure*/
+  unsigned long data[0];	/* data */
+} bp_tag_t;
+
+typedef struct meminfo {
+  unsigned long type;
+  unsigned long start;
+  unsigned long end;
+} meminfo_t;
+
+#define SYSMEM_BANKS_MAX 5
+
+#define MEMORY_TYPE_CONVENTIONAL	0x1000
+#define MEMORY_TYPE_NONE		0x2000
+
+typedef struct sysmem_info {
+  int nr_banks;
+  meminfo_t bank[SYSMEM_BANKS_MAX];
+} sysmem_info_t;
+
+extern sysmem_info_t sysmem;
+
+#endif
+#endif
+
+
+
diff --git a/include/asm-xtensa/bug.h b/include/asm-xtensa/bug.h
new file mode 100644
index 00000000000..56703659b20
--- /dev/null
+++ b/include/asm-xtensa/bug.h
@@ -0,0 +1,41 @@
+/*
+ * include/asm-xtensa/bug.h
+ *
+ * Macros to cause a 'bug' message.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_BUG_H
+#define _XTENSA_BUG_H
+
+#include <linux/stringify.h>
+
+#define ILL	__asm__ __volatile__ (".byte 0,0,0\n")
+
+#ifdef CONFIG_KALLSYMS
+# define BUG() do {							\
+	printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__);		\
+	ILL;								\
+} while (0)
+#else
+# define BUG() do {							\
+	printk("kernel BUG!\n");					\
+      	ILL;								\
+} while (0)
+#endif
+
+#define BUG_ON(condition) do { if (unlikely((condition)!=0)) BUG(); } while(0)
+#define PAGE_BUG(page) do {  BUG(); } while (0)
+#define WARN_ON(condition) do {						   \
+  if (unlikely((condition)!=0)) {					   \
+    printk ("Warning in %s at %s:%d\n", __FUNCTION__, __FILE__, __LINE__); \
+      dump_stack();							   \
+  }									   \
+} while (0)
+
+#endif	/* _XTENSA_BUG_H */
diff --git a/include/asm-xtensa/bugs.h b/include/asm-xtensa/bugs.h
new file mode 100644
index 00000000000..c4228532013
--- /dev/null
+++ b/include/asm-xtensa/bugs.h
@@ -0,0 +1,22 @@
+/*
+ * include/asm-xtensa/bugs.h
+ *
+ * This is included by init/main.c to check for architecture-dependent bugs.
+ *
+ * Xtensa processors don't have any bugs.  :)
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ */
+
+#ifndef _XTENSA_BUGS_H
+#define _XTENSA_BUGS_H
+
+#include <asm/processor.h>
+
+static void __init check_bugs(void)
+{
+}
+
+#endif /* _XTENSA_BUGS_H */
diff --git a/include/asm-xtensa/byteorder.h b/include/asm-xtensa/byteorder.h
new file mode 100644
index 00000000000..0b1552569aa
--- /dev/null
+++ b/include/asm-xtensa/byteorder.h
@@ -0,0 +1,82 @@
+/*
+ * include/asm-xtensa/byteorder.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_BYTEORDER_H
+#define _XTENSA_BYTEORDER_H
+
+#include <asm/processor.h>
+#include <asm/types.h>
+
+static __inline__ __const__ __u32 ___arch__swab32(__u32 x)
+{
+    __u32 res;
+    /* instruction sequence from Xtensa ISA release 2/2000 */
+    __asm__("ssai     8           \n\t"
+	    "srli     %0, %1, 16  \n\t"
+	    "src      %0, %0, %1  \n\t"
+	    "src      %0, %0, %0  \n\t"
+	    "src      %0, %1, %0  \n"
+	    : "=&a" (res)
+	    : "a" (x)
+	    );
+    return res;
+}
+
+static __inline__ __const__ __u16 ___arch__swab16(__u16 x)
+{
+    /* Given that 'short' values are signed (i.e., can be negative),
+     * we cannot assume that the upper 16-bits of the register are
+     * zero.  We are careful to mask values after shifting.
+     */
+
+    /* There exists an anomaly between xt-gcc and xt-xcc.  xt-gcc
+     * inserts an extui instruction after putting this function inline
+     * to ensure that it uses only the least-significant 16 bits of
+     * the result.  xt-xcc doesn't use an extui, but assumes the
+     * __asm__ macro follows convention that the upper 16 bits of an
+     * 'unsigned short' result are still zero.  This macro doesn't
+     * follow convention; indeed, it leaves garbage in the upport 16
+     * bits of the register.
+
+     * Declaring the temporary variables 'res' and 'tmp' to be 32-bit
+     * types while the return type of the function is a 16-bit type
+     * forces both compilers to insert exactly one extui instruction
+     * (or equivalent) to mask off the upper 16 bits. */
+
+    __u32 res;
+    __u32 tmp;
+
+    __asm__("extui    %1, %2, 8, 8\n\t"
+	    "slli     %0, %2, 8   \n\t"
+	    "or       %0, %0, %1  \n"
+	    : "=&a" (res), "=&a" (tmp)
+	    : "a" (x)
+	    );
+
+    return res;
+}
+
+#define __arch__swab32(x) ___arch__swab32(x)
+#define __arch__swab16(x) ___arch__swab16(x)
+
+#if !defined(__STRICT_ANSI__) || defined(__KERNEL__)
+#  define __BYTEORDER_HAS_U64__
+#  define __SWAB_64_THRU_32__
+#endif
+
+#ifdef __XTENSA_EL__
+# include <linux/byteorder/little_endian.h>
+#elif defined(__XTENSA_EB__)
+# include <linux/byteorder/big_endian.h>
+#else
+# error processor byte order undefined!
+#endif
+
+#endif /* __ASM_XTENSA_BYTEORDER_H */
diff --git a/include/asm-xtensa/cache.h b/include/asm-xtensa/cache.h
new file mode 100644
index 00000000000..5aae3f12407
--- /dev/null
+++ b/include/asm-xtensa/cache.h
@@ -0,0 +1,32 @@
+/*
+ * include/asm-xtensa/cacheflush.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ * 2 of the License, or (at your option) any later version.
+ *
+ * (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_CACHE_H
+#define _XTENSA_CACHE_H
+
+#include <xtensa/config/core.h>
+
+#if XCHAL_ICACHE_SIZE > 0
+# if (XCHAL_ICACHE_SIZE % (XCHAL_ICACHE_LINESIZE*XCHAL_ICACHE_WAYS*4)) != 0
+#  error cache configuration outside expected/supported range!
+# endif
+#endif
+
+#if XCHAL_DCACHE_SIZE > 0
+# if (XCHAL_DCACHE_SIZE % (XCHAL_DCACHE_LINESIZE*XCHAL_DCACHE_WAYS*4)) != 0
+#  error cache configuration outside expected/supported range!
+# endif
+#endif
+
+#define L1_CACHE_SHIFT		XCHAL_CACHE_LINEWIDTH_MAX
+#define L1_CACHE_BYTES		XCHAL_CACHE_LINESIZE_MAX
+
+#endif	/* _XTENSA_CACHE_H */
diff --git a/include/asm-xtensa/cacheflush.h b/include/asm-xtensa/cacheflush.h
new file mode 100644
index 00000000000..44a36e08784
--- /dev/null
+++ b/include/asm-xtensa/cacheflush.h
@@ -0,0 +1,122 @@
+/*
+ * include/asm-xtensa/cacheflush.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_CACHEFLUSH_H
+#define _XTENSA_CACHEFLUSH_H
+
+#ifdef __KERNEL__
+
+#include <linux/mm.h>
+#include <asm/processor.h>
+#include <asm/page.h>
+
+/*
+ * flush and invalidate data cache, invalidate instruction cache:
+ *
+ * __flush_invalidate_cache_all()
+ * __flush_invalidate_cache_range(from,sze)
+ *
+ * invalidate data or instruction cache:
+ *
+ * __invalidate_icache_all()
+ * __invalidate_icache_page(adr)
+ * __invalidate_dcache_page(adr)
+ * __invalidate_icache_range(from,size)
+ * __invalidate_dcache_range(from,size)
+ *
+ * flush data cache:
+ *
+ * __flush_dcache_page(adr)
+ *
+ * flush and invalidate data cache:
+ *
+ * __flush_invalidate_dcache_all()
+ * __flush_invalidate_dcache_page(adr)
+ * __flush_invalidate_dcache_range(from,size)
+ */
+
+extern void __flush_invalidate_cache_all(void);
+extern void __flush_invalidate_cache_range(unsigned long, unsigned long);
+extern void __flush_invalidate_dcache_all(void);
+extern void __invalidate_icache_all(void);
+
+extern void __invalidate_dcache_page(unsigned long);
+extern void __invalidate_icache_page(unsigned long);
+extern void __invalidate_icache_range(unsigned long, unsigned long);
+extern void __invalidate_dcache_range(unsigned long, unsigned long);
+
+#if XCHAL_DCACHE_IS_WRITEBACK
+extern void __flush_dcache_page(unsigned long);
+extern void __flush_invalidate_dcache_page(unsigned long);
+extern void __flush_invalidate_dcache_range(unsigned long, unsigned long);
+#else
+# define __flush_dcache_page(p)				do { } while(0)
+# define __flush_invalidate_dcache_page(p) 		do { } while(0)
+# define __flush_invalidate_dcache_range(p,s)		do { } while(0)
+#endif
+
+/*
+ * We have physically tagged caches - nothing to do here -
+ * unless we have cache aliasing.
+ *
+ * Pages can get remapped. Because this might change the 'color' of that page,
+ * we have to flush the cache before the PTE is changed.
+ * (see also Documentation/cachetlb.txt)
+ */
+
+#if (DCACHE_WAY_SIZE > PAGE_SIZE) && XCHAL_DCACHE_IS_WRITEBACK
+
+#define flush_cache_all()		__flush_invalidate_cache_all();
+#define flush_cache_mm(mm)		__flush_invalidate_cache_all();
+
+#define flush_cache_vmap(start,end)	__flush_invalidate_cache_all();
+#define flush_cache_vunmap(start,end)	__flush_invalidate_cache_all();
+
+extern void flush_dcache_page(struct page*);
+
+extern void flush_cache_range(struct vm_area_struct*, ulong, ulong);
+extern void flush_cache_page(struct vm_area_struct*, unsigned long, unsigned long);
+
+#else
+
+#define flush_cache_all()				do { } while (0)
+#define flush_cache_mm(mm)				do { } while (0)
+
+#define flush_cache_vmap(start,end)			do { } while (0)
+#define flush_cache_vunmap(start,end)			do { } while (0)
+
+#define flush_dcache_page(page)				do { } while (0)
+
+#define flush_cache_page(vma,addr,pfn)			do { } while (0)
+#define flush_cache_range(vma,start,end)		do { } while (0)
+
+#endif
+
+#define flush_icache_range(start,end) 					\
+	__invalidate_icache_range(start,(end)-(start))
+
+/* This is not required, see Documentation/cachetlb.txt */
+
+#define	flush_icache_page(vma,page)			do { } while(0)
+
+#define flush_dcache_mmap_lock(mapping)			do { } while (0)
+#define flush_dcache_mmap_unlock(mapping)		do { } while (0)
+
+
+#define copy_to_user_page(vma, page, vaddr, dst, src, len) \
+	memcpy(dst, src, len)
+
+#define copy_from_user_page(vma, page, vaddr, dst, src, len) \
+	memcpy(dst, src, len)
+
+#endif /* __KERNEL__ */
+
+#endif /* _XTENSA_CACHEFLUSH_H */
+
diff --git a/include/asm-xtensa/checksum.h b/include/asm-xtensa/checksum.h
new file mode 100644
index 00000000000..1a00fad1992
--- /dev/null
+++ b/include/asm-xtensa/checksum.h
@@ -0,0 +1,264 @@
+/*
+ * include/asm-xtensa/checksum.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_CHECKSUM_H
+#define _XTENSA_CHECKSUM_H
+
+#include <linux/config.h>
+#include <linux/in6.h>
+#include <xtensa/config/core.h>
+
+/*
+ * computes the checksum of a memory block at buff, length len,
+ * and adds in "sum" (32-bit)
+ *
+ * returns a 32-bit number suitable for feeding into itself
+ * or csum_tcpudp_magic
+ *
+ * this function must be called with even lengths, except
+ * for the last fragment, which may be odd
+ *
+ * it's best to have buff aligned on a 32-bit boundary
+ */
+asmlinkage unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum);
+
+/*
+ * the same as csum_partial, but copies from src while it
+ * checksums, and handles user-space pointer exceptions correctly, when needed.
+ *
+ * here even more important to align src and dst on a 32-bit (or even
+ * better 64-bit) boundary
+ */
+
+asmlinkage unsigned int csum_partial_copy_generic( const char *src, char *dst, int len, int sum,
+						   int *src_err_ptr, int *dst_err_ptr);
+
+/*
+ *	Note: when you get a NULL pointer exception here this means someone
+ *	passed in an incorrect kernel address to one of these functions.
+ *
+ *	If you use these functions directly please don't forget the
+ *	verify_area().
+ */
+extern __inline__
+unsigned int csum_partial_copy_nocheck ( const char *src, char *dst,
+					int len, int sum)
+{
+	return csum_partial_copy_generic ( src, dst, len, sum, NULL, NULL);
+}
+
+extern __inline__
+unsigned int csum_partial_copy_from_user ( const char *src, char *dst,
+						int len, int sum, int *err_ptr)
+{
+	return csum_partial_copy_generic ( src, dst, len, sum, err_ptr, NULL);
+}
+
+/*
+ * These are the old (and unsafe) way of doing checksums, a warning message will be
+ * printed if they are used and an exeption occurs.
+ *
+ * these functions should go away after some time.
+ */
+
+#define csum_partial_copy_fromuser csum_partial_copy
+unsigned int csum_partial_copy( const char *src, char *dst, int len, int sum);
+
+/*
+ *	Fold a partial checksum
+ */
+
+static __inline__ unsigned int csum_fold(unsigned int sum)
+{
+	unsigned int __dummy;
+	__asm__("extui	%1, %0, 16, 16\n\t"
+		"extui	%0 ,%0, 0, 16\n\t"
+		"add	%0, %0, %1\n\t"
+		"slli	%1, %0, 16\n\t"
+		"add	%0, %0, %1\n\t"
+		"extui	%0, %0, 16, 16\n\t"
+		"neg	%0, %0\n\t"
+		"addi	%0, %0, -1\n\t"
+		"extui	%0, %0, 0, 16\n\t"
+		: "=r" (sum), "=&r" (__dummy)
+		: "0" (sum));
+	return sum;
+}
+
+/*
+ *	This is a version of ip_compute_csum() optimized for IP headers,
+ *	which always checksum on 4 octet boundaries.
+ */
+static __inline__ unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl)
+{
+	unsigned int sum, tmp, endaddr;
+
+	__asm__ __volatile__(
+		"sub		%0, %0, %0\n\t"
+#if XCHAL_HAVE_LOOPS
+		"loopgtz	%2, 2f\n\t"
+#else
+		"beqz		%2, 2f\n\t"
+		"slli		%4, %2, 2\n\t"
+		"add		%4, %4, %1\n\t"
+		"0:\t"
+#endif
+		"l32i		%3, %1, 0\n\t"
+		"add		%0, %0, %3\n\t"
+		"bgeu		%0, %3, 1f\n\t"
+		"addi		%0, %0, 1\n\t"
+		"1:\t"
+		"addi		%1, %1, 4\n\t"
+#if !XCHAL_HAVE_LOOPS
+		"blt		%1, %4, 0b\n\t"
+#endif
+		"2:\t"
+	/* Since the input registers which are loaded with iph and ihl
+	   are modified, we must also specify them as outputs, or gcc
+	   will assume they contain their original values. */
+		: "=r" (sum), "=r" (iph), "=r" (ihl), "=&r" (tmp), "=&r" (endaddr)
+		: "1" (iph), "2" (ihl));
+
+	return	csum_fold(sum);
+}
+
+static __inline__ unsigned long csum_tcpudp_nofold(unsigned long saddr,
+						   unsigned long daddr,
+						   unsigned short len,
+						   unsigned short proto,
+						   unsigned int sum)
+{
+
+#ifdef __XTENSA_EL__
+	unsigned long len_proto = (ntohs(len)<<16)+proto*256;
+#elif defined(__XTENSA_EB__)
+	unsigned long len_proto = (proto<<16)+len;
+#else
+# error processor byte order undefined!
+#endif
+	__asm__("add	%0, %0, %1\n\t"
+		"bgeu	%0, %1, 1f\n\t"
+		"addi	%0, %0, 1\n\t"
+		"1:\t"
+		"add	%0, %0, %2\n\t"
+		"bgeu	%0, %2, 1f\n\t"
+		"addi	%0, %0, 1\n\t"
+		"1:\t"
+		"add	%0, %0, %3\n\t"
+		"bgeu	%0, %3, 1f\n\t"
+		"addi	%0, %0, 1\n\t"
+		"1:\t"
+		: "=r" (sum), "=r" (len_proto)
+		: "r" (daddr), "r" (saddr), "1" (len_proto), "0" (sum));
+	return sum;
+}
+
+/*
+ * computes the checksum of the TCP/UDP pseudo-header
+ * returns a 16-bit checksum, already complemented
+ */
+static __inline__ unsigned short int csum_tcpudp_magic(unsigned long saddr,
+						       unsigned long daddr,
+						       unsigned short len,
+						       unsigned short proto,
+						       unsigned int sum)
+{
+	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
+}
+
+/*
+ * this routine is used for miscellaneous IP-like checksums, mainly
+ * in icmp.c
+ */
+
+static __inline__ unsigned short ip_compute_csum(unsigned char * buff, int len)
+{
+    return csum_fold (csum_partial(buff, len, 0));
+}
+
+#define _HAVE_ARCH_IPV6_CSUM
+static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr,
+						     struct in6_addr *daddr,
+						     __u32 len,
+						     unsigned short proto,
+						     unsigned int sum)
+{
+	unsigned int __dummy;
+	__asm__("l32i	%1, %2, 0\n\t"
+		"add	%0, %0, %1\n\t"
+		"bgeu	%0, %1, 1f\n\t"
+		"addi	%0, %0, 1\n\t"
+		"1:\t"
+		"l32i	%1, %2, 4\n\t"
+		"add	%0, %0, %1\n\t"
+		"bgeu	%0, %1, 1f\n\t"
+		"addi	%0, %0, 1\n\t"
+		"1:\t"
+		"l32i	%1, %2, 8\n\t"
+		"add	%0, %0, %1\n\t"
+		"bgeu	%0, %1, 1f\n\t"
+		"addi	%0, %0, 1\n\t"
+		"1:\t"
+		"l32i	%1, %2, 12\n\t"
+		"add	%0, %0, %1\n\t"
+		"bgeu	%0, %1, 1f\n\t"
+		"addi	%0, %0, 1\n\t"
+		"1:\t"
+		"l32i	%1, %3, 0\n\t"
+		"add	%0, %0, %1\n\t"
+		"bgeu	%0, %1, 1f\n\t"
+		"addi	%0, %0, 1\n\t"
+		"1:\t"
+		"l32i	%1, %3, 4\n\t"
+		"add	%0, %0, %1\n\t"
+		"bgeu	%0, %1, 1f\n\t"
+		"addi	%0, %0, 1\n\t"
+		"1:\t"
+		"l32i	%1, %3, 8\n\t"
+		"add	%0, %0, %1\n\t"
+		"bgeu	%0, %1, 1f\n\t"
+		"addi	%0, %0, 1\n\t"
+		"1:\t"
+		"l32i	%1, %3, 12\n\t"
+		"add	%0, %0, %1\n\t"
+		"bgeu	%0, %1, 1f\n\t"
+		"addi	%0, %0, 1\n\t"
+		"1:\t"
+		"add	%0, %0, %4\n\t"
+		"bgeu	%0, %4, 1f\n\t"
+		"addi	%0, %0, 1\n\t"
+		"1:\t"
+		"add	%0, %0, %5\n\t"
+		"bgeu	%0, %5, 1f\n\t"
+		"addi	%0, %0, 1\n\t"
+		"1:\t"
+		: "=r" (sum), "=&r" (__dummy)
+		: "r" (saddr), "r" (daddr),
+		  "r" (htonl(len)), "r" (htonl(proto)), "0" (sum));
+
+	return csum_fold(sum);
+}
+
+/*
+ *	Copy and checksum to user
+ */
+#define HAVE_CSUM_COPY_USER
+static __inline__ unsigned int csum_and_copy_to_user (const char *src, char *dst,
+				    int len, int sum, int *err_ptr)
+{
+	if (access_ok(VERIFY_WRITE, dst, len))
+		return csum_partial_copy_generic(src, dst, len, sum, NULL, err_ptr);
+
+	if (len)
+		*err_ptr = -EFAULT;
+
+	return -1; /* invalid checksum */
+}
+#endif
diff --git a/include/asm-xtensa/coprocessor.h b/include/asm-xtensa/coprocessor.h
new file mode 100644
index 00000000000..a91b96dc0ef
--- /dev/null
+++ b/include/asm-xtensa/coprocessor.h
@@ -0,0 +1,70 @@
+/*
+ * include/asm-xtensa/cpextra.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2003 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_COPROCESSOR_H
+#define _XTENSA_COPROCESSOR_H
+
+#include <xtensa/config/core.h>
+
+#define XTOFS(last_start,last_size,align) \
+	((last_start+last_size+align-1) & -align)
+
+#define XTENSA_CP_EXTRA_OFFSET	0
+#define XTENSA_CP_EXTRA_ALIGN	XCHAL_EXTRA_SA_ALIGN
+
+#define XTENSA_CPE_CP0_OFFSET	\
+	XTOFS(XTENSA_CP_EXTRA_OFFSET, XCHAL_EXTRA_SA_SIZE, XCHAL_CP0_SA_ALIGN)
+#define XTENSA_CPE_CP1_OFFSET	\
+	XTOFS(XTENSA_CPE_CP0_OFFSET, XCHAL_CP0_SA_SIZE, XCHAL_CP1_SA_ALIGN)
+#define XTENSA_CPE_CP2_OFFSET	\
+	XTOFS(XTENSA_CPE_CP1_OFFSET, XCHAL_CP1_SA_SIZE, XCHAL_CP2_SA_ALIGN)
+#define XTENSA_CPE_CP3_OFFSET	\
+	XTOFS(XTENSA_CPE_CP2_OFFSET, XCHAL_CP2_SA_SIZE, XCHAL_CP3_SA_ALIGN)
+#define XTENSA_CPE_CP4_OFFSET	\
+	XTOFS(XTENSA_CPE_CP3_OFFSET, XCHAL_CP3_SA_SIZE, XCHAL_CP4_SA_ALIGN)
+#define XTENSA_CPE_CP5_OFFSET	\
+	XTOFS(XTENSA_CPE_CP4_OFFSET, XCHAL_CP4_SA_SIZE, XCHAL_CP5_SA_ALIGN)
+#define XTENSA_CPE_CP6_OFFSET	\
+	XTOFS(XTENSA_CPE_CP5_OFFSET, XCHAL_CP5_SA_SIZE, XCHAL_CP6_SA_ALIGN)
+#define XTENSA_CPE_CP7_OFFSET	\
+	XTOFS(XTENSA_CPE_CP6_OFFSET, XCHAL_CP6_SA_SIZE, XCHAL_CP7_SA_ALIGN)
+#define XTENSA_CP_EXTRA_SIZE	\
+	XTOFS(XTENSA_CPE_CP7_OFFSET, XCHAL_CP7_SA_SIZE, 16)
+
+#if XCHAL_CP_NUM > 0
+# ifndef __ASSEMBLY__
+/*
+ * Tasks that own contents of (last user) each coprocessor.
+ * Entries are 0 for not-owned or non-existent coprocessors.
+ * Note: The size of this structure is fixed to 8 bytes in entry.S
+ */
+typedef struct {
+	struct task_struct *owner;	/* owner */
+	int offset;			/* offset in cpextra space. */
+} coprocessor_info_t;
+# else
+#  define COPROCESSOR_INFO_OWNER 0
+#  define COPROCESSOR_INFO_OFFSET 4
+#  define COPROCESSOR_INFO_SIZE 8
+# endif
+#endif
+
+
+#ifndef __ASSEMBLY__
+# if XCHAL_CP_NUM > 0
+struct task_struct;
+extern void release_coprocessors (struct task_struct*);
+extern void save_coprocessor_registers(void*, int);
+# else
+#  define release_coprocessors(task)
+# endif
+#endif
+
+#endif	/* _XTENSA_COPROCESSOR_H */
diff --git a/include/asm-xtensa/cpumask.h b/include/asm-xtensa/cpumask.h
new file mode 100644
index 00000000000..ebeede397db
--- /dev/null
+++ b/include/asm-xtensa/cpumask.h
@@ -0,0 +1,16 @@
+/*
+ * include/asm-xtensa/cpumask.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_CPUMASK_H
+#define _XTENSA_CPUMASK_H
+
+#include <asm-generic/cpumask.h>
+
+#endif /* _XTENSA_CPUMASK_H */
diff --git a/include/asm-xtensa/cputime.h b/include/asm-xtensa/cputime.h
new file mode 100644
index 00000000000..a7fb864a50a
--- /dev/null
+++ b/include/asm-xtensa/cputime.h
@@ -0,0 +1,6 @@
+#ifndef _XTENSA_CPUTIME_H
+#define _XTENSA_CPUTIME_H
+
+#include <asm-generic/cputime.h>
+
+#endif /* _XTENSA_CPUTIME_H */
diff --git a/include/asm-xtensa/current.h b/include/asm-xtensa/current.h
new file mode 100644
index 00000000000..8d1eb5d7864
--- /dev/null
+++ b/include/asm-xtensa/current.h
@@ -0,0 +1,38 @@
+/*
+ * include/asm-xtensa/current.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_CURRENT_H
+#define _XTENSA_CURRENT_H
+
+#ifndef __ASSEMBLY__
+
+#include <linux/thread_info.h>
+
+struct task_struct;
+
+static inline struct task_struct *get_current(void)
+{
+	return current_thread_info()->task;
+}
+
+#define current get_current()
+
+#else
+
+#define CURRENT_SHIFT 13
+
+#define GET_CURRENT(reg,sp)		\
+	GET_THREAD_INFO(reg,sp);	\
+  	l32i reg, reg, TI_TASK		\
+
+#endif
+
+
+#endif /* XTENSA_CURRENT_H */
diff --git a/include/asm-xtensa/delay.h b/include/asm-xtensa/delay.h
new file mode 100644
index 00000000000..6359c55e77a
--- /dev/null
+++ b/include/asm-xtensa/delay.h
@@ -0,0 +1,50 @@
+/*
+ * include/asm-xtensa/delay.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ *
+ */
+
+#ifndef _XTENSA_DELAY_H
+#define _XTENSA_DELAY_H
+
+#include <linux/config.h>
+#include <asm/processor.h>
+#include <asm/param.h>
+
+extern unsigned long loops_per_jiffy;
+
+extern __inline__ void __delay(unsigned long loops)
+{
+  /* 2 cycles per loop. */
+  __asm__ __volatile__ ("1: addi %0, %0, -2; bgeui %0, 1, 1b"
+			: "=r" (loops) : "0" (loops));
+}
+
+static __inline__ u32 xtensa_get_ccount(void)
+{
+	u32 ccount;
+	asm volatile ("rsr %0, 234; # CCOUNT\n" : "=r" (ccount));
+	return ccount;
+}
+
+/* For SMP/NUMA systems, change boot_cpu_data to something like
+ * local_cpu_data->... where local_cpu_data points to the current
+ * cpu. */
+
+static __inline__ void udelay (unsigned long usecs)
+{
+	unsigned long start = xtensa_get_ccount();
+	unsigned long cycles = usecs * (loops_per_jiffy / (1000000UL / HZ));
+
+	/* Note: all variables are unsigned (can wrap around)! */
+	while (((unsigned long)xtensa_get_ccount()) - start < cycles)
+		;
+}
+
+#endif
+
diff --git a/include/asm-xtensa/div64.h b/include/asm-xtensa/div64.h
new file mode 100644
index 00000000000..c4a10577638
--- /dev/null
+++ b/include/asm-xtensa/div64.h
@@ -0,0 +1,19 @@
+/*
+ * include/asm-xtensa/div64.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_DIV64_H
+#define _XTENSA_DIV64_H
+
+#define do_div(n,base) ({ \
+	int __res = n % ((unsigned int) base); \
+	n /= (unsigned int) base; \
+	__res; })
+
+#endif
diff --git a/include/asm-xtensa/dma-mapping.h b/include/asm-xtensa/dma-mapping.h
new file mode 100644
index 00000000000..e86a206f120
--- /dev/null
+++ b/include/asm-xtensa/dma-mapping.h
@@ -0,0 +1,182 @@
+/*
+ * include/asm-xtensa/dma_mapping.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2003 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_DMA_MAPPING_H
+#define _XTENSA_DMA_MAPPING_H
+
+#include <asm/scatterlist.h>
+#include <asm/cache.h>
+#include <asm/io.h>
+#include <linux/mm.h>
+
+/*
+ * DMA-consistent mapping functions.
+ */
+
+extern void *consistent_alloc(int, size_t, dma_addr_t, unsigned long);
+extern void consistent_free(void*, size_t, dma_addr_t);
+extern void consistent_sync(void*, size_t, int);
+
+#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
+#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
+
+void *dma_alloc_coherent(struct device *dev, size_t size,
+			   dma_addr_t *dma_handle, int flag);
+
+void dma_free_coherent(struct device *dev, size_t size,
+			 void *vaddr, dma_addr_t dma_handle);
+
+static inline dma_addr_t
+dma_map_single(struct device *dev, void *ptr, size_t size,
+	       enum dma_data_direction direction)
+{
+	BUG_ON(direction == DMA_NONE);
+	consistent_sync(ptr, size, direction);
+	return virt_to_phys(ptr);
+}
+
+static inline void
+dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
+		 enum dma_data_direction direction)
+{
+	BUG_ON(direction == DMA_NONE);
+}
+
+static inline int
+dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
+	   enum dma_data_direction direction)
+{
+	int i;
+
+	BUG_ON(direction == DMA_NONE);
+
+	for (i = 0; i < nents; i++, sg++ ) {
+		BUG_ON(!sg->page);
+
+		sg->dma_address = page_to_phys(sg->page) + sg->offset;
+		consistent_sync(page_address(sg->page) + sg->offset,
+				sg->length, direction);
+	}
+
+	return nents;
+}
+
+static inline dma_addr_t
+dma_map_page(struct device *dev, struct page *page, unsigned long offset,
+	     size_t size, enum dma_data_direction direction)
+{
+	BUG_ON(direction == DMA_NONE);
+	return (dma_addr_t)(page_to_pfn(page)) * PAGE_SIZE + offset;
+}
+
+static inline void
+dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
+	       enum dma_data_direction direction)
+{
+	BUG_ON(direction == DMA_NONE);
+}
+
+
+static inline void
+dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nhwentries,
+	     enum dma_data_direction direction)
+{
+	BUG_ON(direction == DMA_NONE);
+}
+
+static inline void
+dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size,
+		enum dma_data_direction direction)
+{
+	consistent_sync((void *)bus_to_virt(dma_handle), size, direction);
+}
+
+static inline void
+dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size,
+		enum dma_data_direction direction)
+{
+	consistent_sync((void *)bus_to_virt(dma_handle), size, direction);
+}
+
+static inline void
+dma_sync_single_range_for_cpu(struct device *dev, dma_addr_t dma_handle,
+		      unsigned long offset, size_t size,
+		      enum dma_data_direction direction)
+{
+
+	consistent_sync((void *)bus_to_virt(dma_handle)+offset,size,direction);
+}
+
+static inline void
+dma_sync_single_range_for_device(struct device *dev, dma_addr_t dma_handle,
+		      unsigned long offset, size_t size,
+		      enum dma_data_direction direction)
+{
+
+	consistent_sync((void *)bus_to_virt(dma_handle)+offset,size,direction);
+}
+static inline void
+dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems,
+		 enum dma_data_direction dir)
+{
+	int i;
+	for (i = 0; i < nelems; i++, sg++)
+		consistent_sync(page_address(sg->page) + sg->offset,
+				sg->length, dir);
+}
+
+static inline void
+dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems,
+		 enum dma_data_direction dir)
+{
+	int i;
+	for (i = 0; i < nelems; i++, sg++)
+		consistent_sync(page_address(sg->page) + sg->offset,
+				sg->length, dir);
+}
+static inline int
+dma_mapping_error(dma_addr_t dma_addr)
+{
+	return 0;
+}
+
+static inline int
+dma_supported(struct device *dev, u64 mask)
+{
+	return 1;
+}
+
+static inline int
+dma_set_mask(struct device *dev, u64 mask)
+{
+	if(!dev->dma_mask || !dma_supported(dev, mask))
+		return -EIO;
+
+	*dev->dma_mask = mask;
+
+	return 0;
+}
+
+static inline int
+dma_get_cache_alignment(void)
+{
+	return L1_CACHE_BYTES;
+}
+
+#define dma_is_consistent(d)	(1)
+
+static inline void
+dma_cache_sync(void *vaddr, size_t size,
+	       enum dma_data_direction direction)
+{
+	consistent_sync(vaddr, size, direction);
+}
+
+#endif	/* _XTENSA_DMA_MAPPING_H */
diff --git a/include/asm-xtensa/dma.h b/include/asm-xtensa/dma.h
new file mode 100644
index 00000000000..1c22b023458
--- /dev/null
+++ b/include/asm-xtensa/dma.h
@@ -0,0 +1,61 @@
+/*
+ * include/asm-xtensa/dma.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2003 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_DMA_H
+#define _XTENSA_DMA_H
+
+#include <linux/config.h>
+#include <asm/io.h>		/* need byte IO */
+#include <xtensa/config/core.h>
+
+/*
+ * This is only to be defined if we have PC-like DMA.
+ * By default this is not true on an Xtensa processor,
+ * however on boards with a PCI bus, such functionality
+ * might be emulated externally.
+ *
+ * NOTE:  there still exists driver code that assumes
+ * this is defined, eg. drivers/sound/soundcard.c (as of 2.4).
+ */
+#define MAX_DMA_CHANNELS	8
+
+/*
+ * The maximum virtual address to which DMA transfers
+ * can be performed on this platform.
+ *
+ * NOTE: This is board (platform) specific, not processor-specific!
+ *
+ * NOTE: This assumes DMA transfers can only be performed on
+ *	the section of physical memory contiguously mapped in virtual
+ *	space for the kernel.  For the Xtensa architecture, this
+ *	means the maximum possible size of this DMA area is
+ *	the size of the statically mapped kernel segment
+ *	(XCHAL_KSEG_{CACHED,BYPASS}_SIZE), ie. 128 MB.
+ *
+ * NOTE: When the entire KSEG area is DMA capable, we substract
+ *	one from the max address so that the virt_to_phys() macro
+ *	works correctly on the address (otherwise the address
+ *	enters another area, and virt_to_phys() may not return
+ *	the value desired).
+ */
+#define MAX_DMA_ADDRESS		(PAGE_OFFSET + XCHAL_KSEG_CACHED_SIZE - 1)
+
+/* Reserve and release a DMA channel */
+extern int request_dma(unsigned int dmanr, const char * device_id);
+extern void free_dma(unsigned int dmanr);
+
+#ifdef CONFIG_PCI
+extern int isa_dma_bridge_buggy;
+#else
+#define isa_dma_bridge_buggy 	(0)
+#endif
+
+
+#endif
diff --git a/include/asm-xtensa/elf.h b/include/asm-xtensa/elf.h
new file mode 100644
index 00000000000..64f1f53874f
--- /dev/null
+++ b/include/asm-xtensa/elf.h
@@ -0,0 +1,222 @@
+/*
+ * include/asm-xtensa/elf.h
+ *
+ * ELF register definitions
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_ELF_H
+#define _XTENSA_ELF_H
+
+#include <asm/ptrace.h>
+#include <asm/coprocessor.h>
+#include <xtensa/config/core.h>
+
+/* Xtensa processor ELF architecture-magic number */
+
+#define EM_XTENSA	94
+#define EM_XTENSA_OLD	0xABC7
+
+/* ELF register definitions. This is needed for core dump support.  */
+
+/*
+ * elf_gregset_t contains the application-level state in the following order:
+ * Processor info: 	config_version, cpuxy
+ * Processor state:	pc, ps, exccause, excvaddr, wb, ws,
+ *			lbeg, lend, lcount, sar
+ * GP regs:		ar0 - arXX
+ */
+
+typedef unsigned long elf_greg_t;
+
+typedef struct {
+	elf_greg_t xchal_config_id0;
+	elf_greg_t xchal_config_id1;
+	elf_greg_t cpux;
+	elf_greg_t cpuy;
+	elf_greg_t pc;
+	elf_greg_t ps;
+	elf_greg_t exccause;
+	elf_greg_t excvaddr;
+	elf_greg_t windowbase;
+	elf_greg_t windowstart;
+	elf_greg_t lbeg;
+	elf_greg_t lend;
+	elf_greg_t lcount;
+	elf_greg_t sar;
+	elf_greg_t syscall;
+	elf_greg_t ar[XCHAL_NUM_AREGS];
+} xtensa_gregset_t;
+
+#define ELF_NGREG	(sizeof(xtensa_gregset_t) / sizeof(elf_greg_t))
+
+typedef elf_greg_t elf_gregset_t[ELF_NGREG];
+
+/*
+ *  Compute the size of the coprocessor and extra state layout (register info)
+ *  table (in bytes).
+ *  This is actually the maximum size of the table, as opposed to the size,
+ *  which is available from the _xtensa_reginfo_table_size global variable.
+ *
+ *  (See also arch/xtensa/kernel/coprocessor.S)
+ *
+ */
+
+#ifndef XCHAL_EXTRA_SA_CONTENTS_LIBDB_NUM
+# define XTENSA_CPE_LTABLE_SIZE		0
+#else
+# define XTENSA_CPE_SEGMENT(num)	(num ? (1+num) : 0)
+# define XTENSA_CPE_LTABLE_ENTRIES	\
+		( XTENSA_CPE_SEGMENT(XCHAL_EXTRA_SA_CONTENTS_LIBDB_NUM)	\
+		+ XTENSA_CPE_SEGMENT(XCHAL_CP0_SA_CONTENTS_LIBDB_NUM)	\
+		+ XTENSA_CPE_SEGMENT(XCHAL_CP1_SA_CONTENTS_LIBDB_NUM)	\
+		+ XTENSA_CPE_SEGMENT(XCHAL_CP2_SA_CONTENTS_LIBDB_NUM)	\
+		+ XTENSA_CPE_SEGMENT(XCHAL_CP3_SA_CONTENTS_LIBDB_NUM)	\
+		+ XTENSA_CPE_SEGMENT(XCHAL_CP4_SA_CONTENTS_LIBDB_NUM)	\
+		+ XTENSA_CPE_SEGMENT(XCHAL_CP5_SA_CONTENTS_LIBDB_NUM)	\
+		+ XTENSA_CPE_SEGMENT(XCHAL_CP6_SA_CONTENTS_LIBDB_NUM)	\
+		+ XTENSA_CPE_SEGMENT(XCHAL_CP7_SA_CONTENTS_LIBDB_NUM)	\
+		+ 1		/* final entry */			\
+		)
+# define XTENSA_CPE_LTABLE_SIZE		(XTENSA_CPE_LTABLE_ENTRIES * 8)
+#endif
+
+
+/*
+ * Instantiations of the elf_fpregset_t type contain, in most
+ * architectures, the floating point (FPU) register set.
+ * For Xtensa, this type is extended to contain all custom state,
+ * ie. coprocessor and "extra" (non-coprocessor) state (including,
+ * for example, TIE-defined states and register files; as well
+ * as other optional processor state).
+ * This includes FPU state if a floating-point coprocessor happens
+ * to have been configured within the Xtensa processor.
+ *
+ * TOTAL_FPREGS_SIZE is the required size (without rounding)
+ * of elf_fpregset_t.  It provides space for the following:
+ *
+ *  a)	32-bit mask of active coprocessors for this task (similar
+ *	to CPENABLE in single-threaded Xtensa processor systems)
+ *
+ *  b)	table describing the layout of custom states (ie. of
+ *      individual registers, etc) within the save areas
+ *
+ *  c)  save areas for each coprocessor and for non-coprocessor
+ *      ("extra") state
+ *
+ * Note that save areas may require up to 16-byte alignment when
+ * accessed by save/restore sequences.  We do not need to ensure
+ * such alignment in an elf_fpregset_t structure because custom
+ * state is not directly loaded/stored into it; rather, save area
+ * contents are copied to elf_fpregset_t from the active save areas
+ * (see 'struct task_struct' definition in processor.h for that)
+ * using memcpy().  But we do allow space for such alignment,
+ * to allow optimizations of layout and copying.
+ */
+
+#define TOTAL_FPREGS_SIZE						\
+  	(4 + XTENSA_CPE_LTABLE_SIZE + XTENSA_CP_EXTRA_SIZE)
+#define ELF_NFPREG							\
+	((TOTAL_FPREGS_SIZE + sizeof(elf_fpreg_t) - 1) / sizeof(elf_fpreg_t))
+
+typedef unsigned int elf_fpreg_t;
+typedef elf_fpreg_t elf_fpregset_t[ELF_NFPREG];
+
+#define ELF_CORE_COPY_REGS(_eregs, _pregs) 				\
+	xtensa_elf_core_copy_regs (&_eregs, _pregs);
+
+extern void xtensa_elf_core_copy_regs (xtensa_gregset_t *, struct pt_regs *);
+
+/*
+ * This is used to ensure we don't load something for the wrong architecture.
+ */
+
+#define elf_check_arch(x) ( ( (x)->e_machine == EM_XTENSA )  || \
+			    ( (x)->e_machine == EM_XTENSA_OLD ) )
+
+/*
+ * These are used to set parameters in the core dumps.
+ */
+
+#ifdef __XTENSA_EL__
+# define ELF_DATA	ELFDATA2LSB
+#elif defined(__XTENSA_EB__)
+# define ELF_DATA	ELFDATA2MSB
+#else
+# error processor byte order undefined!
+#endif
+
+#define ELF_CLASS	ELFCLASS32
+#define ELF_ARCH	EM_XTENSA
+
+#define USE_ELF_CORE_DUMP
+#define ELF_EXEC_PAGESIZE	PAGE_SIZE
+
+/*
+ * This is the location that an ET_DYN program is loaded if exec'ed.  Typical
+ * use of this is to invoke "./ld.so someprog" to test out a new version of
+ * the loader.  We need to make sure that it is out of the way of the program
+ * that it will "exec", and that there is sufficient room for the brk.
+ */
+
+#define ELF_ET_DYN_BASE         (2 * TASK_SIZE / 3)
+
+/*
+ * This yields a mask that user programs can use to figure out what
+ * instruction set this CPU supports.  This could be done in user space,
+ * but it's not easy, and we've already done it here.
+ */
+
+#define ELF_HWCAP	(0)
+
+/*
+ * This yields a string that ld.so will use to load implementation
+ * specific libraries for optimization.  This is more specific in
+ * intent than poking at uname or /proc/cpuinfo.
+ * For the moment, we have only optimizations for the Intel generations,
+ * but that could change...
+ */
+
+#define ELF_PLATFORM  (NULL)
+
+/*
+ * The Xtensa processor ABI says that when the program starts, a2
+ * contains a pointer to a function which might be registered using
+ * `atexit'.  This provides a mean for the dynamic linker to call
+ * DT_FINI functions for shared libraries that have been loaded before
+ * the code runs.
+ *
+ * A value of 0 tells we have no such handler.
+ *
+ * We might as well make sure everything else is cleared too (except
+ * for the stack pointer in a1), just to make things more
+ * deterministic.  Also, clearing a0 terminates debugger backtraces.
+ */
+
+#define ELF_PLAT_INIT(_r, load_addr) \
+  do { _r->areg[0]=0; /*_r->areg[1]=0;*/ _r->areg[2]=0;  _r->areg[3]=0;  \
+       _r->areg[4]=0;  _r->areg[5]=0;    _r->areg[6]=0;  _r->areg[7]=0;  \
+       _r->areg[8]=0;  _r->areg[9]=0;    _r->areg[10]=0; _r->areg[11]=0; \
+       _r->areg[12]=0; _r->areg[13]=0;   _r->areg[14]=0; _r->areg[15]=0; \
+  } while (0)
+
+#ifdef __KERNEL__
+
+#define SET_PERSONALITY(ex, ibcs2) set_personality(PER_LINUX_32BIT)
+
+extern void do_copy_regs (xtensa_gregset_t*, struct pt_regs*,
+			  struct task_struct*);
+extern void do_restore_regs (xtensa_gregset_t*, struct pt_regs*,
+			     struct task_struct*);
+extern void do_save_fpregs (elf_fpregset_t*, struct pt_regs*,
+			    struct task_struct*);
+extern int do_restore_fpregs (elf_fpregset_t*, struct pt_regs*,
+			      struct task_struct*);
+
+#endif	/* __KERNEL__ */
+#endif	/* _XTENSA_ELF_H */
diff --git a/include/asm-xtensa/errno.h b/include/asm-xtensa/errno.h
new file mode 100644
index 00000000000..ced5194d275
--- /dev/null
+++ b/include/asm-xtensa/errno.h
@@ -0,0 +1,142 @@
+/*
+ * include/asm-xtensa/errno.h
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ *
+ * Copyright (C) 2002 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_ERRNO_H
+#define _XTENSA_ERRNO_H
+
+#define	EPERM		 1	/* Operation not permitted */
+#define	ENOENT		 2	/* No such file or directory */
+#define	ESRCH		 3	/* No such process */
+#define	EINTR		 4	/* Interrupted system call */
+#define	EIO		 5	/* I/O error */
+#define	ENXIO		 6	/* No such device or address */
+#define	E2BIG		 7	/* Arg list too long */
+#define	ENOEXEC		 8	/* Exec format error */
+#define	EBADF		 9	/* Bad file number */
+#define	ECHILD		10	/* No child processes */
+#define	EAGAIN		11	/* Try again */
+#define	ENOMEM		12	/* Out of memory */
+#define	EACCES		13	/* Permission denied */
+#define	EFAULT		14	/* Bad address */
+#define	ENOTBLK		15	/* Block device required */
+#define	EBUSY		16	/* Device or resource busy */
+#define	EEXIST		17	/* File exists */
+#define	EXDEV		18	/* Cross-device link */
+#define	ENODEV		19	/* No such device */
+#define	ENOTDIR		20	/* Not a directory */
+#define	EISDIR		21	/* Is a directory */
+#define	EINVAL		22	/* Invalid argument */
+#define	ENFILE		23	/* File table overflow */
+#define	EMFILE		24	/* Too many open files */
+#define	ENOTTY		25	/* Not a typewriter */
+#define	ETXTBSY		26	/* Text file busy */
+#define	EFBIG		27	/* File too large */
+#define	ENOSPC		28	/* No space left on device */
+#define	ESPIPE		29	/* Illegal seek */
+#define	EROFS		30	/* Read-only file system */
+#define	EMLINK		31	/* Too many links */
+#define	EPIPE		32	/* Broken pipe */
+#define	EDOM		33	/* Math argument out of domain of func */
+#define	ERANGE		34	/* Math result not representable */
+#define	EDEADLK		35	/* Resource deadlock would occur */
+#define	ENAMETOOLONG	36	/* File name too long */
+#define	ENOLCK		37	/* No record locks available */
+#define	ENOSYS		38	/* Function not implemented */
+#define	ENOTEMPTY	39	/* Directory not empty */
+#define	ELOOP		40	/* Too many symbolic links encountered */
+#define	EWOULDBLOCK	EAGAIN	/* Operation would block */
+#define	ENOMSG		42	/* No message of desired type */
+#define	EIDRM		43	/* Identifier removed */
+#define	ECHRNG		44	/* Channel number out of range */
+#define	EL2NSYNC	45	/* Level 2 not synchronized */
+#define	EL3HLT		46	/* Level 3 halted */
+#define	EL3RST		47	/* Level 3 reset */
+#define	ELNRNG		48	/* Link number out of range */
+#define	EUNATCH		49	/* Protocol driver not attached */
+#define	ENOCSI		50	/* No CSI structure available */
+#define	EL2HLT		51	/* Level 2 halted */
+#define	EBADE		52	/* Invalid exchange */
+#define	EBADR		53	/* Invalid request descriptor */
+#define	EXFULL		54	/* Exchange full */
+#define	ENOANO		55	/* No anode */
+#define	EBADRQC		56	/* Invalid request code */
+#define	EBADSLT		57	/* Invalid slot */
+
+#define	EDEADLOCK	EDEADLK
+
+#define	EBFONT		59	/* Bad font file format */
+#define	ENOSTR		60	/* Device not a stream */
+#define	ENODATA		61	/* No data available */
+#define	ETIME		62	/* Timer expired */
+#define	ENOSR		63	/* Out of streams resources */
+#define	ENONET		64	/* Machine is not on the network */
+#define	ENOPKG		65	/* Package not installed */
+#define	EREMOTE		66	/* Object is remote */
+#define	ENOLINK		67	/* Link has been severed */
+#define	EADV		68	/* Advertise error */
+#define	ESRMNT		69	/* Srmount error */
+#define	ECOMM		70	/* Communication error on send */
+#define	EPROTO		71	/* Protocol error */
+#define	EMULTIHOP	72	/* Multihop attempted */
+#define	EDOTDOT		73	/* RFS specific error */
+#define	EBADMSG		74	/* Not a data message */
+#define	EOVERFLOW	75	/* Value too large for defined data type */
+#define	ENOTUNIQ	76	/* Name not unique on network */
+#define	EBADFD		77	/* File descriptor in bad state */
+#define	EREMCHG		78	/* Remote address changed */
+#define	ELIBACC		79	/* Can not access a needed shared library */
+#define	ELIBBAD		80	/* Accessing a corrupted shared library */
+#define	ELIBSCN		81	/* .lib section in a.out corrupted */
+#define	ELIBMAX		82	/* Attempting to link in too many shared libraries */
+#define	ELIBEXEC	83	/* Cannot exec a shared library directly */
+#define	EILSEQ		84	/* Illegal byte sequence */
+#define	ERESTART	85	/* Interrupted system call should be restarted */
+#define	ESTRPIPE	86	/* Streams pipe error */
+#define	EUSERS		87	/* Too many users */
+#define	ENOTSOCK	88	/* Socket operation on non-socket */
+#define	EDESTADDRREQ	89	/* Destination address required */
+#define	EMSGSIZE	90	/* Message too long */
+#define	EPROTOTYPE	91	/* Protocol wrong type for socket */
+#define	ENOPROTOOPT	92	/* Protocol not available */
+#define	EPROTONOSUPPORT	93	/* Protocol not supported */
+#define	ESOCKTNOSUPPORT	94	/* Socket type not supported */
+#define	EOPNOTSUPP	95	/* Operation not supported on transport endpoint */
+#define	EPFNOSUPPORT	96	/* Protocol family not supported */
+#define	EAFNOSUPPORT	97	/* Address family not supported by protocol */
+#define	EADDRINUSE	98	/* Address already in use */
+#define	EADDRNOTAVAIL	99	/* Cannot assign requested address */
+#define	ENETDOWN	100	/* Network is down */
+#define	ENETUNREACH	101	/* Network is unreachable */
+#define	ENETRESET	102	/* Network dropped connection because of reset */
+#define	ECONNABORTED	103	/* Software caused connection abort */
+#define	ECONNRESET	104	/* Connection reset by peer */
+#define	ENOBUFS		105	/* No buffer space available */
+#define	EISCONN		106	/* Transport endpoint is already connected */
+#define	ENOTCONN	107	/* Transport endpoint is not connected */
+#define	ESHUTDOWN	108	/* Cannot send after transport endpoint shutdown */
+#define	ETOOMANYREFS	109	/* Too many references: cannot splice */
+#define	ETIMEDOUT	110	/* Connection timed out */
+#define	ECONNREFUSED	111	/* Connection refused */
+#define	EHOSTDOWN	112	/* Host is down */
+#define	EHOSTUNREACH	113	/* No route to host */
+#define	EALREADY	114	/* Operation already in progress */
+#define	EINPROGRESS	115	/* Operation now in progress */
+#define	ESTALE		116	/* Stale NFS file handle */
+#define	EUCLEAN		117	/* Structure needs cleaning */
+#define	ENOTNAM		118	/* Not a XENIX named type file */
+#define	ENAVAIL		119	/* No XENIX semaphores available */
+#define	EISNAM		120	/* Is a named type file */
+#define	EREMOTEIO	121	/* Remote I/O error */
+#define	EDQUOT		122	/* Quota exceeded */
+
+#define	ENOMEDIUM	123	/* No medium found */
+#define	EMEDIUMTYPE	124	/* Wrong medium type */
+
+#endif	/* _XTENSA_ERRNO_H */
diff --git a/include/asm-xtensa/fcntl.h b/include/asm-xtensa/fcntl.h
new file mode 100644
index 00000000000..48876bb727d
--- /dev/null
+++ b/include/asm-xtensa/fcntl.h
@@ -0,0 +1,101 @@
+/*
+ * include/asm-xtensa/fcntl.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1995, 1996, 1997, 1998 by Ralf Baechle
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_FCNTL_H
+#define _XTENSA_FCNTL_H
+
+/* open/fcntl - O_SYNC is only implemented on blocks devices and on files
+   located on an ext2 file system */
+#define O_ACCMODE	0x0003
+#define O_RDONLY	0x0000
+#define O_WRONLY	0x0001
+#define O_RDWR		0x0002
+#define O_APPEND	0x0008
+#define O_SYNC		0x0010
+#define O_NONBLOCK	0x0080
+#define O_CREAT         0x0100	/* not fcntl */
+#define O_TRUNC		0x0200	/* not fcntl */
+#define O_EXCL		0x0400	/* not fcntl */
+#define O_NOCTTY	0x0800	/* not fcntl */
+#define FASYNC		0x1000	/* fcntl, for BSD compatibility */
+#define O_LARGEFILE	0x2000	/* allow large file opens - currently ignored */
+#define O_DIRECT	0x8000	/* direct disk access hint - currently ignored*/
+#define O_DIRECTORY	0x10000	/* must be a directory */
+#define O_NOFOLLOW	0x20000	/* don't follow links */
+#define O_NOATIME	0x100000
+
+#define O_NDELAY	O_NONBLOCK
+
+#define F_DUPFD		0	/* dup */
+#define F_GETFD		1	/* get close_on_exec */
+#define F_SETFD		2	/* set/clear close_on_exec */
+#define F_GETFL		3	/* get file->f_flags */
+#define F_SETFL		4	/* set file->f_flags */
+#define F_GETLK		14
+#define F_GETLK64       15
+#define F_SETLK		6
+#define F_SETLKW	7
+#define F_SETLK64       16
+#define F_SETLKW64      17
+
+#define F_SETOWN	24	/*  for sockets. */
+#define F_GETOWN	23	/*  for sockets. */
+#define F_SETSIG	10	/*  for sockets. */
+#define F_GETSIG	11	/*  for sockets. */
+
+/* for F_[GET|SET]FL */
+#define FD_CLOEXEC	1	/* actually anything with low bit set goes */
+
+/* for posix fcntl() and lockf() */
+#define F_RDLCK		0
+#define F_WRLCK		1
+#define F_UNLCK		2
+
+/* for old implementation of bsd flock () */
+#define F_EXLCK		4	/* or 3 */
+#define F_SHLCK		8	/* or 4 */
+
+/* for leases */
+#define F_INPROGRESS	16
+
+/* operations for bsd flock(), also used by the kernel implementation */
+#define LOCK_SH		1	/* shared lock */
+#define LOCK_EX		2	/* exclusive lock */
+#define LOCK_NB		4	/* or'd with one of the above to prevent
+				   blocking */
+#define LOCK_UN		8	/* remove lock */
+
+#define LOCK_MAND	32	/* This is a mandatory flock ... */
+#define LOCK_READ	64	/*  which allows concurrent read operations */
+#define LOCK_WRITE	128	/*  which allows concurrent write operations */
+#define LOCK_RW		192	/*  which allows concurrent read & write ops */
+
+typedef struct flock {
+	short l_type;
+	short l_whence;
+	__kernel_off_t l_start;
+	__kernel_off_t l_len;
+	long  l_sysid;
+	__kernel_pid_t l_pid;
+	long  pad[4];
+} flock_t;
+
+struct flock64 {
+	short  l_type;
+	short  l_whence;
+	__kernel_off_t l_start;
+	__kernel_off_t l_len;
+	pid_t  l_pid;
+};
+
+#define F_LINUX_SPECIFIC_BASE	1024
+
+#endif /* _XTENSA_FCNTL_H */
diff --git a/include/asm-xtensa/fixmap.h b/include/asm-xtensa/fixmap.h
new file mode 100644
index 00000000000..4423b8ad495
--- /dev/null
+++ b/include/asm-xtensa/fixmap.h
@@ -0,0 +1,252 @@
+/*
+ * include/asm-xtensa/fixmap.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_FIXMAP_H
+#define _XTENSA_FIXMAP_H
+
+#include <asm/processor.h>
+
+#ifdef CONFIG_MMU
+
+/*
+ * Here we define all the compile-time virtual addresses.
+ */
+
+#if XCHAL_SEG_MAPPABLE_VADDR != 0
+# error "Current port requires virtual user space starting at 0"
+#endif
+#if XCHAL_SEG_MAPPABLE_SIZE < 0x80000000
+# error "Current port requires at least 0x8000000 bytes for user space"
+#endif
+
+/* Verify instruction/data ram/rom and xlmi don't overlay vmalloc space. */
+
+#define __IN_VMALLOC(addr)						\
+	(((addr) >= VMALLOC_START) && ((addr) < VMALLOC_END))
+#define __SPAN_VMALLOC(start,end)					\
+	(((start) < VMALLOC_START) && ((end) >= VMALLOC_END))
+#define INSIDE_VMALLOC(start,end) 					\
+	(__IN_VMALLOC((start)) || __IN_VMALLOC(end) || __SPAN_VMALLOC((start),(end)))
+
+#if XCHAL_NUM_INSTROM
+# if XCHAL_NUM_INSTROM == 1
+#  if INSIDE_VMALLOC(XCHAL_INSTROM0_VADDR,XCHAL_INSTROM0_VADDR+XCHAL_INSTROM0_SIZE)
+#   error vmalloc range conflicts with instrom0
+#  endif
+# endif
+# if XCHAL_NUM_INSTROM == 2
+#  if INSIDE_VMALLOC(XCHAL_INSTROM1_VADDR,XCHAL_INSTROM1_VADDR+XCHAL_INSTROM1_SIZE)
+#   error vmalloc range conflicts with instrom1
+#  endif
+# endif
+#endif
+
+#if XCHAL_NUM_INSTRAM
+# if XCHAL_NUM_INSTRAM == 1
+#  if INSIDE_VMALLOC(XCHAL_INSTRAM0_VADDR,XCHAL_INSTRAM0_VADDR+XCHAL_INSTRAM0_SIZE)
+#   error vmalloc range conflicts with instram0
+#  endif
+# endif
+# if XCHAL_NUM_INSTRAM == 2
+#  if INSIDE_VMALLOC(XCHAL_INSTRAM1_VADDR,XCHAL_INSTRAM1_VADDR+XCHAL_INSTRAM1_SIZE)
+#   error vmalloc range conflicts with instram1
+#  endif
+# endif
+#endif
+
+#if XCHAL_NUM_DATAROM
+# if XCHAL_NUM_DATAROM == 1
+#  if INSIDE_VMALLOC(XCHAL_DATAROM0_VADDR,XCHAL_DATAROM0_VADDR+XCHAL_DATAROM0_SIZE)
+#   error vmalloc range conflicts with datarom0
+#  endif
+# endif
+# if XCHAL_NUM_DATAROM == 2
+#  if INSIDE_VMALLOC(XCHAL_DATAROM1_VADDR,XCHAL_DATAROM1_VADDR+XCHAL_DATAROM1_SIZE)
+#   error vmalloc range conflicts with datarom1
+#  endif
+# endif
+#endif
+
+#if XCHAL_NUM_DATARAM
+# if XCHAL_NUM_DATARAM == 1
+#  if INSIDE_VMALLOC(XCHAL_DATARAM0_VADDR,XCHAL_DATARAM0_VADDR+XCHAL_DATARAM0_SIZE)
+#   error vmalloc range conflicts with dataram0
+#  endif
+# endif
+# if XCHAL_NUM_DATARAM == 2
+#  if INSIDE_VMALLOC(XCHAL_DATARAM1_VADDR,XCHAL_DATARAM1_VADDR+XCHAL_DATARAM1_SIZE)
+#   error vmalloc range conflicts with dataram1
+#  endif
+# endif
+#endif
+
+#if XCHAL_NUM_XLMI
+# if XCHAL_NUM_XLMI == 1
+#  if INSIDE_VMALLOC(XCHAL_XLMI0_VADDR,XCHAL_XLMI0_VADDR+XCHAL_XLMI0_SIZE)
+#   error vmalloc range conflicts with xlmi0
+#  endif
+# endif
+# if XCHAL_NUM_XLMI == 2
+#  if INSIDE_VMALLOC(XCHAL_XLMI1_VADDR,XCHAL_XLMI1_VADDR+XCHAL_XLMI1_SIZE)
+#   error vmalloc range conflicts with xlmi1
+#  endif
+# endif
+#endif
+
+#if (XCHAL_NUM_INSTROM > 2) || \
+    (XCHAL_NUM_INSTRAM > 2) || \
+    (XCHAL_NUM_DATARAM > 2) || \
+    (XCHAL_NUM_DATAROM > 2) || \
+    (XCHAL_NUM_XLMI    > 2)
+# error Insufficient checks on vmalloc above for more than 2 devices
+#endif
+
+/*
+ * USER_VM_SIZE does not necessarily equal TASK_SIZE.  We bumped
+ * TASK_SIZE down to 0x4000000 to simplify the handling of windowed
+ * call instructions (currently limited to a range of 1 GByte).  User
+ * tasks may very well reclaim the VM space from 0x40000000 to
+ * 0x7fffffff in the future, so we do not want the kernel becoming
+ * accustomed to having any of its stuff (e.g., page tables) in this
+ * region.  This VM region is no-man's land for now.
+ */
+
+#define USER_VM_START		XCHAL_SEG_MAPPABLE_VADDR
+#define USER_VM_SIZE		0x80000000
+
+/*  Size of page table:  */
+
+#define PGTABLE_SIZE_BITS	(32 - XCHAL_MMU_MIN_PTE_PAGE_SIZE + 2)
+#define PGTABLE_SIZE		(1L << PGTABLE_SIZE_BITS)
+
+/*  All kernel-mappable space:  */
+
+#define KERNEL_ALLMAP_START	(USER_VM_START + USER_VM_SIZE)
+#define KERNEL_ALLMAP_SIZE	(XCHAL_SEG_MAPPABLE_SIZE - KERNEL_ALLMAP_START)
+
+/*  Carve out page table at start of kernel-mappable area:  */
+
+#if KERNEL_ALLMAP_SIZE < PGTABLE_SIZE
+#error "Gimme some space for page table!"
+#endif
+#define PGTABLE_START		KERNEL_ALLMAP_START
+
+/*  Remaining kernel-mappable space:  */
+
+#define KERNEL_MAPPED_START	(KERNEL_ALLMAP_START + PGTABLE_SIZE)
+#define KERNEL_MAPPED_SIZE	(KERNEL_ALLMAP_SIZE - PGTABLE_SIZE)
+
+#if KERNEL_MAPPED_SIZE < 0x01000000	/* 16 MB is arbitrary for now */
+# error "Shouldn't the kernel have at least *some* mappable space?"
+#endif
+
+#define MAX_LOW_MEMORY		XCHAL_KSEG_CACHED_SIZE
+
+#endif
+
+/*
+ *  Some constants used elsewhere, but perhaps only in Xtensa header
+ *  files, so maybe we can get rid of some and access compile-time HAL
+ *  directly...
+ *
+ *  Note:  We assume that system RAM is located at the very start of the
+ *  	   kernel segments !!
+ */
+#define KERNEL_VM_LOW           XCHAL_KSEG_CACHED_VADDR
+#define KERNEL_VM_HIGH          XCHAL_KSEG_BYPASS_VADDR
+#define KERNEL_SPACE            XCHAL_KSEG_CACHED_VADDR
+
+/*
+ * Returns the physical/virtual addresses of the kernel space
+ * (works with the cached kernel segment only, which is the
+ *  one normally used for kernel operation).
+ */
+
+/*			PHYSICAL	BYPASS		CACHED
+ *
+ *  bypass vaddr	bypass paddr	*		cached vaddr
+ *  cached vaddr	cached paddr	bypass vaddr	*
+ *  bypass paddr	*		bypass vaddr	cached vaddr
+ *  cached paddr	*		bypass vaddr	cached vaddr
+ *  other		*		*		*
+ */
+
+#define PHYSADDR(a)							      \
+(((unsigned)(a) >= XCHAL_KSEG_BYPASS_VADDR				      \
+  && (unsigned)(a) < XCHAL_KSEG_BYPASS_VADDR + XCHAL_KSEG_BYPASS_SIZE) ?      \
+    (unsigned)(a) - XCHAL_KSEG_BYPASS_VADDR + XCHAL_KSEG_BYPASS_PADDR :       \
+    ((unsigned)(a) >= XCHAL_KSEG_CACHED_VADDR				      \
+     && (unsigned)(a) < XCHAL_KSEG_CACHED_VADDR + XCHAL_KSEG_CACHED_SIZE) ?   \
+        (unsigned)(a) - XCHAL_KSEG_CACHED_VADDR + XCHAL_KSEG_CACHED_PADDR :   \
+	(unsigned)(a))
+
+#define BYPASS_ADDR(a)							      \
+(((unsigned)(a) >= XCHAL_KSEG_BYPASS_PADDR				      \
+  && (unsigned)(a) < XCHAL_KSEG_BYPASS_PADDR + XCHAL_KSEG_BYPASS_SIZE) ?      \
+    (unsigned)(a) - XCHAL_KSEG_BYPASS_PADDR + XCHAL_KSEG_BYPASS_VADDR :       \
+    ((unsigned)(a) >= XCHAL_KSEG_CACHED_PADDR				      \
+     && (unsigned)(a) < XCHAL_KSEG_CACHED_PADDR + XCHAL_KSEG_CACHED_SIZE) ?   \
+        (unsigned)(a) - XCHAL_KSEG_CACHED_PADDR + XCHAL_KSEG_BYPASS_VADDR :   \
+        ((unsigned)(a) >= XCHAL_KSEG_CACHED_VADDR			      \
+         && (unsigned)(a) < XCHAL_KSEG_CACHED_VADDR+XCHAL_KSEG_CACHED_SIZE)?  \
+            (unsigned)(a) - XCHAL_KSEG_CACHED_VADDR+XCHAL_KSEG_BYPASS_VADDR:  \
+	    (unsigned)(a))
+
+#define CACHED_ADDR(a)							      \
+(((unsigned)(a) >= XCHAL_KSEG_BYPASS_PADDR				      \
+  && (unsigned)(a) < XCHAL_KSEG_BYPASS_PADDR + XCHAL_KSEG_BYPASS_SIZE) ?      \
+    (unsigned)(a) - XCHAL_KSEG_BYPASS_PADDR + XCHAL_KSEG_CACHED_VADDR :       \
+    ((unsigned)(a) >= XCHAL_KSEG_CACHED_PADDR			              \
+     && (unsigned)(a) < XCHAL_KSEG_CACHED_PADDR + XCHAL_KSEG_CACHED_SIZE) ?   \
+        (unsigned)(a) - XCHAL_KSEG_CACHED_PADDR + XCHAL_KSEG_CACHED_VADDR :   \
+        ((unsigned)(a) >= XCHAL_KSEG_BYPASS_VADDR			      \
+         && (unsigned)(a) < XCHAL_KSEG_BYPASS_VADDR+XCHAL_KSEG_BYPASS_SIZE) ? \
+            (unsigned)(a) - XCHAL_KSEG_BYPASS_VADDR+XCHAL_KSEG_CACHED_VADDR : \
+	    (unsigned)(a))
+
+#define PHYSADDR_IO(a)							      \
+(((unsigned)(a) >= XCHAL_KIO_BYPASS_VADDR				      \
+  && (unsigned)(a) < XCHAL_KIO_BYPASS_VADDR + XCHAL_KIO_BYPASS_SIZE) ?	      \
+    (unsigned)(a) - XCHAL_KIO_BYPASS_VADDR + XCHAL_KIO_BYPASS_PADDR :	      \
+    ((unsigned)(a) >= XCHAL_KIO_CACHED_VADDR				      \
+     && (unsigned)(a) < XCHAL_KIO_CACHED_VADDR + XCHAL_KIO_CACHED_SIZE) ?     \
+        (unsigned)(a) - XCHAL_KIO_CACHED_VADDR + XCHAL_KIO_CACHED_PADDR :     \
+	(unsigned)(a))
+
+#define BYPASS_ADDR_IO(a)						      \
+(((unsigned)(a) >= XCHAL_KIO_BYPASS_PADDR				      \
+  && (unsigned)(a) < XCHAL_KIO_BYPASS_PADDR + XCHAL_KIO_BYPASS_SIZE) ?	      \
+    (unsigned)(a) - XCHAL_KIO_BYPASS_PADDR + XCHAL_KIO_BYPASS_VADDR :	      \
+    ((unsigned)(a) >= XCHAL_KIO_CACHED_PADDR				      \
+     && (unsigned)(a) < XCHAL_KIO_CACHED_PADDR + XCHAL_KIO_CACHED_SIZE) ?     \
+        (unsigned)(a) - XCHAL_KIO_CACHED_PADDR + XCHAL_KIO_BYPASS_VADDR :     \
+        ((unsigned)(a) >= XCHAL_KIO_CACHED_VADDR			      \
+         && (unsigned)(a) < XCHAL_KIO_CACHED_VADDR + XCHAL_KIO_CACHED_SIZE) ? \
+            (unsigned)(a) - XCHAL_KIO_CACHED_VADDR + XCHAL_KIO_BYPASS_VADDR : \
+	    (unsigned)(a))
+
+#define CACHED_ADDR_IO(a)						      \
+(((unsigned)(a) >= XCHAL_KIO_BYPASS_PADDR				      \
+  && (unsigned)(a) < XCHAL_KIO_BYPASS_PADDR + XCHAL_KIO_BYPASS_SIZE) ?	      \
+    (unsigned)(a) - XCHAL_KIO_BYPASS_PADDR + XCHAL_KIO_CACHED_VADDR :	      \
+    ((unsigned)(a) >= XCHAL_KIO_CACHED_PADDR				      \
+     && (unsigned)(a) < XCHAL_KIO_CACHED_PADDR + XCHAL_KIO_CACHED_SIZE) ?     \
+        (unsigned)(a) - XCHAL_KIO_CACHED_PADDR + XCHAL_KIO_CACHED_VADDR :     \
+        ((unsigned)(a) >= XCHAL_KIO_BYPASS_VADDR			      \
+         && (unsigned)(a) < XCHAL_KIO_BYPASS_VADDR + XCHAL_KIO_BYPASS_SIZE) ? \
+            (unsigned)(a) - XCHAL_KIO_BYPASS_VADDR + XCHAL_KIO_CACHED_VADDR : \
+	    (unsigned)(a))
+
+#endif /* _XTENSA_ADDRSPACE_H */
+
+
+
+
+
diff --git a/include/asm-xtensa/hardirq.h b/include/asm-xtensa/hardirq.h
new file mode 100644
index 00000000000..e07c76c36b9
--- /dev/null
+++ b/include/asm-xtensa/hardirq.h
@@ -0,0 +1,28 @@
+/*
+ * include/asm-xtensa/hardirq.h
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ *
+ * Copyright (C) 2002 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_HARDIRQ_H
+#define _XTENSA_HARDIRQ_H
+
+#include <linux/config.h>
+#include <linux/cache.h>
+#include <asm/irq.h>
+
+/* headers.S is sensitive to the offsets of these fields */
+typedef struct {
+	unsigned int __softirq_pending;
+	unsigned int __syscall_count;
+	struct task_struct * __ksoftirqd_task; /* waitqueue is too large */
+	unsigned int __nmi_count;	       /* arch dependent */
+} ____cacheline_aligned irq_cpustat_t;
+
+#include <linux/irq_cpustat.h>	/* Standard mappings for irq_cpustat_t above */
+
+#endif	/* _XTENSA_HARDIRQ_H */
diff --git a/include/asm-xtensa/hdreg.h b/include/asm-xtensa/hdreg.h
new file mode 100644
index 00000000000..64b80607b80
--- /dev/null
+++ b/include/asm-xtensa/hdreg.h
@@ -0,0 +1,17 @@
+/*
+ * include/asm-xtensa/hdreg.h
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ *
+ * Copyright (C) 2002 - 2005 Tensilica Inc.
+ * Copyright (C) 1994-1996  Linus Torvalds & authors
+ */
+
+#ifndef _XTENSA_HDREG_H
+#define _XTENSA_HDREG_H
+
+typedef unsigned int ide_ioreg_t;
+
+#endif
diff --git a/include/asm-xtensa/highmem.h b/include/asm-xtensa/highmem.h
new file mode 100644
index 00000000000..0a046ca5a68
--- /dev/null
+++ b/include/asm-xtensa/highmem.h
@@ -0,0 +1,17 @@
+/*
+ * include/asm-xtensa/highmem.h
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ *
+ * Copyright (C) 2003 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_HIGHMEM_H
+#define _XTENSA_HIGHMEM_H
+
+extern void flush_cache_kmaps(void);
+
+#endif
+
diff --git a/include/asm-xtensa/hw_irq.h b/include/asm-xtensa/hw_irq.h
new file mode 100644
index 00000000000..ccf436249ea
--- /dev/null
+++ b/include/asm-xtensa/hw_irq.h
@@ -0,0 +1,18 @@
+/*
+ * include/asm-xtensa/hw_irq.h
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ *
+ * Copyright (C) 2002 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_HW_IRQ_H
+#define _XTENSA_HW_IRQ_H
+
+static inline void hw_resend_irq(struct hw_interrupt_type *h, unsigned int i)
+{
+}
+
+#endif
diff --git a/include/asm-xtensa/ide.h b/include/asm-xtensa/ide.h
new file mode 100644
index 00000000000..b523cd4a486
--- /dev/null
+++ b/include/asm-xtensa/ide.h
@@ -0,0 +1,36 @@
+/*
+ * include/asm-xtensa/ide.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1994 - 1996  Linus Torvalds & authors
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_IDE_H
+#define _XTENSA_IDE_H
+
+#ifdef __KERNEL__
+
+#include <linux/config.h>
+
+#ifndef MAX_HWIFS
+# define MAX_HWIFS	1
+#endif
+
+static __inline__ int ide_default_irq(unsigned long base)
+{
+	/* Unsupported! */
+  	return 0;
+}
+
+static __inline__ unsigned long ide_default_io_base(int index)
+{
+	/* Unsupported! */
+  	return 0;
+}
+
+#endif	/* __KERNEL__ */
+#endif	/* _XTENSA_IDE_H */
diff --git a/include/asm-xtensa/io.h b/include/asm-xtensa/io.h
new file mode 100644
index 00000000000..2c471c42ecf
--- /dev/null
+++ b/include/asm-xtensa/io.h
@@ -0,0 +1,197 @@
+/*
+ * linux/include/asm-xtensa/io.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_IO_H
+#define _XTENSA_IO_H
+
+#ifdef __KERNEL__
+#include <linux/config.h>
+#include <asm/byteorder.h>
+
+#include <linux/types.h>
+#include <asm/fixmap.h>
+
+#define _IO_BASE 0
+
+
+/*
+ * swap functions to change byte order from little-endian to big-endian and
+ * vice versa.
+ */
+
+static inline unsigned short _swapw (unsigned short v)
+{
+	return (v << 8) | (v >> 8);
+}
+
+static inline unsigned int _swapl (unsigned int v)
+{
+	return (v << 24) | ((v & 0xff00) << 8) | ((v >> 8) & 0xff00) | (v >> 24);
+}
+
+/*
+ * Change virtual addresses to physical addresses and vv.
+ * These are trivial on the 1:1 Linux/Xtensa mapping
+ */
+
+extern inline unsigned long virt_to_phys(volatile void * address)
+{
+	return PHYSADDR((unsigned long)address);
+}
+
+extern inline void * phys_to_virt(unsigned long address)
+{
+	return (void*) CACHED_ADDR(address);
+}
+
+/*
+ * IO bus memory addresses are also 1:1 with the physical address
+ */
+
+extern inline unsigned long virt_to_bus(volatile void * address)
+{
+	return PHYSADDR((unsigned long)address);
+}
+
+extern inline void * bus_to_virt (unsigned long address)
+{
+	return (void *) CACHED_ADDR(address);
+}
+
+/*
+ * Change "struct page" to physical address.
+ */
+
+extern inline void *ioremap(unsigned long offset, unsigned long size)
+{
+        return (void *) CACHED_ADDR_IO(offset);
+}
+
+extern inline void *ioremap_nocache(unsigned long offset, unsigned long size)
+{
+        return (void *) BYPASS_ADDR_IO(offset);
+}
+
+extern inline void iounmap(void *addr)
+{
+}
+
+/*
+ * Generic I/O
+ */
+
+#define readb(addr) \
+	({ unsigned char __v = (*(volatile unsigned char *)(addr)); __v; })
+#define readw(addr) \
+	({ unsigned short __v = (*(volatile unsigned short *)(addr)); __v; })
+#define readl(addr) \
+	({ unsigned int __v = (*(volatile unsigned int *)(addr)); __v; })
+#define writeb(b, addr) (void)((*(volatile unsigned char *)(addr)) = (b))
+#define writew(b, addr) (void)((*(volatile unsigned short *)(addr)) = (b))
+#define writel(b, addr) (void)((*(volatile unsigned int *)(addr)) = (b))
+
+static inline __u8 __raw_readb(const volatile void __iomem *addr)
+{
+          return *(__force volatile __u8 *)(addr);
+}
+static inline __u16 __raw_readw(const volatile void __iomem *addr)
+{
+          return *(__force volatile __u16 *)(addr);
+}
+static inline __u32 __raw_readl(const volatile void __iomem *addr)
+{
+          return *(__force volatile __u32 *)(addr);
+}
+static inline void __raw_writeb(__u8 b, volatile void __iomem *addr)
+{
+          *(__force volatile __u8 *)(addr) = b;
+}
+static inline void __raw_writew(__u16 b, volatile void __iomem *addr)
+{
+          *(__force volatile __u16 *)(addr) = b;
+}
+static inline void __raw_writel(__u32 b, volatile void __iomem *addr)
+{
+          *(__force volatile __u32 *)(addr) = b;
+}
+
+
+
+
+/* These are the definitions for the x86 IO instructions
+ * inb/inw/inl/outb/outw/outl, the "string" versions
+ * insb/insw/insl/outsb/outsw/outsl, and the "pausing" versions
+ * inb_p/inw_p/...
+ * The macros don't do byte-swapping.
+ */
+
+#define inb(port)		readb((u8 *)((port)+_IO_BASE))
+#define outb(val, port)		writeb((val),(u8 *)((unsigned long)(port)+_IO_BASE))
+#define inw(port)		readw((u16 *)((port)+_IO_BASE))
+#define outw(val, port)		writew((val),(u16 *)((unsigned long)(port)+_IO_BASE))
+#define inl(port)		readl((u32 *)((port)+_IO_BASE))
+#define outl(val, port)		writel((val),(u32 *)((unsigned long)(port)))
+
+#define inb_p(port)		inb((port))
+#define outb_p(val, port)	outb((val), (port))
+#define inw_p(port)		inw((port))
+#define outw_p(val, port)	outw((val), (port))
+#define inl_p(port)		inl((port))
+#define outl_p(val, port)	outl((val), (port))
+
+extern void insb (unsigned long port, void *dst, unsigned long count);
+extern void insw (unsigned long port, void *dst, unsigned long count);
+extern void insl (unsigned long port, void *dst, unsigned long count);
+extern void outsb (unsigned long port, const void *src, unsigned long count);
+extern void outsw (unsigned long port, const void *src, unsigned long count);
+extern void outsl (unsigned long port, const void *src, unsigned long count);
+
+#define IO_SPACE_LIMIT ~0
+
+#define memset_io(a,b,c)       memset((void *)(a),(b),(c))
+#define memcpy_fromio(a,b,c)   memcpy((a),(void *)(b),(c))
+#define memcpy_toio(a,b,c)      memcpy((void *)(a),(b),(c))
+
+/* At this point the Xtensa doesn't provide byte swap instructions */
+
+#ifdef __XTENSA_EB__
+# define in_8(addr) (*(u8*)(addr))
+# define in_le16(addr) _swapw(*(u16*)(addr))
+# define in_le32(addr) _swapl(*(u32*)(addr))
+# define out_8(b, addr) *(u8*)(addr) = (b)
+# define out_le16(b, addr) *(u16*)(addr) = _swapw(b)
+# define out_le32(b, addr) *(u32*)(addr) = _swapl(b)
+#elif defined(__XTENSA_EL__)
+# define in_8(addr)  (*(u8*)(addr))
+# define in_le16(addr) (*(u16*)(addr))
+# define in_le32(addr) (*(u32*)(addr))
+# define out_8(b, addr) *(u8*)(addr) = (b)
+# define out_le16(b, addr) *(u16*)(addr) = (b)
+# define out_le32(b, addr) *(u32*)(addr) = (b)
+#else
+# error processor byte order undefined!
+#endif
+
+
+/*
+ *  * Convert a physical pointer to a virtual kernel pointer for /dev/mem
+ *   * access
+ *    */
+#define xlate_dev_mem_ptr(p)    __va(p)
+
+/*
+ *  * Convert a virtual cached pointer to an uncached pointer
+ *   */
+#define xlate_dev_kmem_ptr(p)   p
+
+
+#endif	/* __KERNEL__ */
+
+#endif	/* _XTENSA_IO_H */
diff --git a/include/asm-xtensa/ioctl.h b/include/asm-xtensa/ioctl.h
new file mode 100644
index 00000000000..856c605d62b
--- /dev/null
+++ b/include/asm-xtensa/ioctl.h
@@ -0,0 +1,83 @@
+/*
+ * include/asm-xtensa/ioctl.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2003 - 2005 Tensilica Inc.
+ *
+ * Derived from "include/asm-i386/ioctl.h"
+ */
+
+#ifndef _XTENSA_IOCTL_H
+#define _XTENSA_IOCTL_H
+
+
+/* ioctl command encoding: 32 bits total, command in lower 16 bits,
+ * size of the parameter structure in the lower 14 bits of the
+ * upper 16 bits.
+ * Encoding the size of the parameter structure in the ioctl request
+ * is useful for catching programs compiled with old versions
+ * and to avoid overwriting user space outside the user buffer area.
+ * The highest 2 bits are reserved for indicating the ``access mode''.
+ * NOTE: This limits the max parameter size to 16kB -1 !
+ */
+
+/*
+ * The following is for compatibility across the various Linux
+ * platforms.  The i386 ioctl numbering scheme doesn't really enforce
+ * a type field.  De facto, however, the top 8 bits of the lower 16
+ * bits are indeed used as a type field, so we might just as well make
+ * this explicit here.  Please be sure to use the decoding macros
+ * below from now on.
+ */
+#define _IOC_NRBITS	8
+#define _IOC_TYPEBITS	8
+#define _IOC_SIZEBITS	14
+#define _IOC_DIRBITS	2
+
+#define _IOC_NRMASK	((1 << _IOC_NRBITS)-1)
+#define _IOC_TYPEMASK	((1 << _IOC_TYPEBITS)-1)
+#define _IOC_SIZEMASK	((1 << _IOC_SIZEBITS)-1)
+#define _IOC_DIRMASK	((1 << _IOC_DIRBITS)-1)
+
+#define _IOC_NRSHIFT	0
+#define _IOC_TYPESHIFT	(_IOC_NRSHIFT+_IOC_NRBITS)
+#define _IOC_SIZESHIFT	(_IOC_TYPESHIFT+_IOC_TYPEBITS)
+#define _IOC_DIRSHIFT	(_IOC_SIZESHIFT+_IOC_SIZEBITS)
+
+/*
+ * Direction bits.
+ */
+#define _IOC_NONE	0U
+#define _IOC_WRITE	1U
+#define _IOC_READ	2U
+
+#define _IOC(dir,type,nr,size) \
+	(((dir)  << _IOC_DIRSHIFT) | \
+	 ((type) << _IOC_TYPESHIFT) | \
+	 ((nr)   << _IOC_NRSHIFT) | \
+	 ((size) << _IOC_SIZESHIFT))
+
+/* used to create numbers */
+#define _IO(type,nr)		_IOC(_IOC_NONE,(type),(nr),0)
+#define _IOR(type,nr,size)	_IOC(_IOC_READ,(type),(nr),sizeof(size))
+#define _IOW(type,nr,size)	_IOC(_IOC_WRITE,(type),(nr),sizeof(size))
+#define _IOWR(type,nr,size)	_IOC(_IOC_READ|_IOC_WRITE,(type),(nr),sizeof(size))
+
+/* used to decode ioctl numbers.. */
+#define _IOC_DIR(nr)		(((nr) >> _IOC_DIRSHIFT) & _IOC_DIRMASK)
+#define _IOC_TYPE(nr)		(((nr) >> _IOC_TYPESHIFT) & _IOC_TYPEMASK)
+#define _IOC_NR(nr)		(((nr) >> _IOC_NRSHIFT) & _IOC_NRMASK)
+#define _IOC_SIZE(nr)		(((nr) >> _IOC_SIZESHIFT) & _IOC_SIZEMASK)
+
+/* ...and for the drivers/sound files... */
+
+#define IOC_IN		(_IOC_WRITE << _IOC_DIRSHIFT)
+#define IOC_OUT		(_IOC_READ << _IOC_DIRSHIFT)
+#define IOC_INOUT	((_IOC_WRITE|_IOC_READ) << _IOC_DIRSHIFT)
+#define IOCSIZE_MASK	(_IOC_SIZEMASK << _IOC_SIZESHIFT)
+#define IOCSIZE_SHIFT	(_IOC_SIZESHIFT)
+
+#endif
diff --git a/include/asm-xtensa/ioctls.h b/include/asm-xtensa/ioctls.h
new file mode 100644
index 00000000000..10c443435c1
--- /dev/null
+++ b/include/asm-xtensa/ioctls.h
@@ -0,0 +1,112 @@
+/*
+ * include/asm-xtensa/ioctl.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2003 - 2005 Tensilica Inc.
+ *
+ * Derived from "include/asm-i386/ioctls.h"
+ */
+
+#ifndef _XTENSA_IOCTLS_H
+#define _XTENSA_IOCTLS_H
+
+#include <asm/ioctl.h>
+
+#define FIOCLEX		_IO('f', 1)
+#define FIONCLEX	_IO('f', 2)
+#define FIOASYNC	_IOW('f', 125, int)
+#define FIONBIO		_IOW('f', 126, int)
+#define FIONREAD	_IOR('f', 127, int)
+#define TIOCINQ		FIONREAD
+#define FIOQSIZE	_IOR('f', 128, loff_t)
+
+#define TCGETS		0x5401
+#define TCSETS		0x5402
+#define TCSETSW		0x5403
+#define TCSETSF		0x5404
+
+#define TCGETA		_IOR('t', 23, struct termio)
+#define TCSETA		_IOW('t', 24, struct termio)
+#define TCSETAW		_IOW('t', 25, struct termio)
+#define TCSETAF		_IOW('t', 28, struct termio)
+
+#define TCSBRK		_IO('t', 29)
+#define TCXONC		_IO('t', 30)
+#define TCFLSH		_IO('t', 31)
+
+#define TIOCSWINSZ	_IOW('t', 103, struct winsize)
+#define TIOCGWINSZ	_IOR('t', 104, struct winsize)
+#define	TIOCSTART	_IO('t', 110)		/* start output, like ^Q */
+#define	TIOCSTOP	_IO('t', 111)		/* stop output, like ^S */
+#define TIOCOUTQ        _IOR('t', 115, int)     /* output queue size */
+
+#define TIOCSPGRP	_IOW('t', 118, int)
+#define TIOCGPGRP	_IOR('t', 119, int)
+
+#define TIOCEXCL	_IO('T', 12)
+#define TIOCNXCL	_IO('T', 13)
+#define TIOCSCTTY	_IO('T', 14)
+
+#define TIOCSTI		_IOW('T', 18, char)
+#define TIOCMGET	_IOR('T', 21, unsigned int)
+#define TIOCMBIS	_IOW('T', 22, unsigned int)
+#define TIOCMBIC	_IOW('T', 23, unsigned int)
+#define TIOCMSET	_IOW('T', 24, unsigned int)
+# define TIOCM_LE	0x001
+# define TIOCM_DTR	0x002
+# define TIOCM_RTS	0x004
+# define TIOCM_ST	0x008
+# define TIOCM_SR	0x010
+# define TIOCM_CTS	0x020
+# define TIOCM_CAR	0x040
+# define TIOCM_RNG	0x080
+# define TIOCM_DSR	0x100
+# define TIOCM_CD	TIOCM_CAR
+# define TIOCM_RI	TIOCM_RNG
+
+#define TIOCGSOFTCAR	_IOR('T', 25, unsigned int)
+#define TIOCSSOFTCAR	_IOW('T', 26, unsigned int)
+#define TIOCLINUX	_IOW('T', 28, char)
+#define TIOCCONS	_IO('T', 29)
+#define TIOCGSERIAL	_IOR('T', 30, struct serial_struct)
+#define TIOCSSERIAL	_IOW('T', 31, struct serial_struct)
+#define TIOCPKT		_IOW('T', 32, int)
+# define TIOCPKT_DATA		 0
+# define TIOCPKT_FLUSHREAD	 1
+# define TIOCPKT_FLUSHWRITE	 2
+# define TIOCPKT_STOP		 4
+# define TIOCPKT_START		 8
+# define TIOCPKT_NOSTOP		16
+# define TIOCPKT_DOSTOP		32
+
+
+#define TIOCNOTTY	_IO('T', 34)
+#define TIOCSETD	_IOW('T', 35, int)
+#define TIOCGETD	_IOR('T', 36, int)
+#define TCSBRKP		_IOW('T', 37, int)   /* Needed for POSIX tcsendbreak()*/
+#define TIOCTTYGSTRUCT	_IOR('T', 38, struct tty_struct) /* For debugging only*/
+#define TIOCSBRK	_IO('T', 39) 	     /* BSD compatibility */
+#define TIOCCBRK	_IO('T', 40)	     /* BSD compatibility */
+#define TIOCGSID	_IOR('T', 41, pid_t) /* Return the session ID of FD*/
+#define TIOCGPTN	_IOR('T',0x30, unsigned int) /* Get Pty Number (of pty-mux device) */
+#define TIOCSPTLCK	_IOW('T',0x31, int)  /* Lock/unlock Pty */
+
+#define TIOCSERCONFIG	_IO('T', 83)
+#define TIOCSERGWILD	_IOR('T', 84,  int)
+#define TIOCSERSWILD	_IOW('T', 85,  int)
+#define TIOCGLCKTRMIOS	0x5456
+#define TIOCSLCKTRMIOS	0x5457
+#define TIOCSERGSTRUCT	0x5458		     /* For debugging only */
+#define TIOCSERGETLSR   _IOR('T', 89, unsigned int) /* Get line status reg. */
+  /* ioctl (fd, TIOCSERGETLSR, &result) where result may be as below */
+# define TIOCSER_TEMT    0x01		     /* Transmitter physically empty */
+#define TIOCSERGETMULTI _IOR('T', 90, struct serial_multiport_struct) /* Get multiport config  */
+#define TIOCSERSETMULTI _IOW('T', 91, struct serial_multiport_struct) /* Set multiport config */
+
+#define TIOCMIWAIT	_IO('T', 92) /* wait for a change on serial input line(s) */
+#define TIOCGICOUNT	_IOR('T', 93, struct async_icount) /* read serial port inline interrupt counts */
+
+#endif /* _XTENSA_IOCTLS_H */
diff --git a/include/asm-xtensa/ipc.h b/include/asm-xtensa/ipc.h
new file mode 100644
index 00000000000..d37bdb4d4c9
--- /dev/null
+++ b/include/asm-xtensa/ipc.h
@@ -0,0 +1,34 @@
+/*
+ * include/asm-xtensa/ipc.h
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_IPC_H
+#define _XTENSA_IPC_H
+
+struct ipc_kludge {
+	struct msgbuf __user *msgp;
+	long msgtyp;
+};
+
+#define SEMOP		 1
+#define SEMGET		 2
+#define SEMCTL		 3
+#define SEMTIMEDOP	 4
+#define MSGSND		11
+#define MSGRCV		12
+#define MSGGET		13
+#define MSGCTL		14
+#define SHMAT		21
+#define SHMDT		22
+#define SHMGET		23
+#define SHMCTL		24
+
+#define IPCCALL(version,op)	((version)<<16 | (op))
+
+#endif	/* _XTENSA_IPC_H */
diff --git a/include/asm-xtensa/ipcbuf.h b/include/asm-xtensa/ipcbuf.h
new file mode 100644
index 00000000000..c33aa6a4214
--- /dev/null
+++ b/include/asm-xtensa/ipcbuf.h
@@ -0,0 +1,37 @@
+/*
+ * include/asm-xtensa/ipcbuf.h
+ *
+ * The ipc64_perm structure for the Xtensa architecture.
+ * Note extra padding because this structure is passed back and forth
+ * between kernel and user space.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_IPCBUF_H
+#define _XTENSA_IPCBUF_H
+
+/*
+ * Pad space is left for:
+ * - 32-bit mode_t and seq
+ * - 2 miscellaneous 32-bit values
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ */
+
+struct ipc64_perm
+{
+	__kernel_key_t		key;
+	__kernel_uid32_t	uid;
+	__kernel_gid32_t	gid;
+	__kernel_uid32_t	cuid;
+	__kernel_gid32_t	cgid;
+	__kernel_mode_t		mode;
+	unsigned long		seq;
+	unsigned long		__unused1;
+	unsigned long		__unused2;
+};
+
+#endif /* _XTENSA_IPCBUF_H */
diff --git a/include/asm-xtensa/irq.h b/include/asm-xtensa/irq.h
new file mode 100644
index 00000000000..d984e955938
--- /dev/null
+++ b/include/asm-xtensa/irq.h
@@ -0,0 +1,37 @@
+/*
+ * include/asm-xtensa/irq.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_IRQ_H
+#define _XTENSA_IRQ_H
+
+#include <linux/config.h>
+#include <asm/platform/hardware.h>
+
+#include <xtensa/config/core.h>
+
+#ifndef PLATFORM_NR_IRQS
+# define PLATFORM_NR_IRQS 0
+#endif
+#define XTENSA_NR_IRQS XCHAL_NUM_INTERRUPTS
+#define NR_IRQS (XTENSA_NR_IRQS + PLATFORM_NR_IRQS)
+
+static __inline__ int irq_canonicalize(int irq)
+{
+	return (irq);
+}
+
+struct irqaction;
+#if 0 // FIXME
+extern void disable_irq_nosync(unsigned int);
+extern void disable_irq(unsigned int);
+extern void enable_irq(unsigned int);
+#endif
+
+#endif	/* _XTENSA_IRQ_H */
diff --git a/include/asm-xtensa/kmap_types.h b/include/asm-xtensa/kmap_types.h
new file mode 100644
index 00000000000..9e822d2e3bc
--- /dev/null
+++ b/include/asm-xtensa/kmap_types.h
@@ -0,0 +1,31 @@
+/*
+ * include/asm-xtensa/kmap_types.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_KMAP_TYPES_H
+#define _XTENSA_KMAP_TYPES_H
+
+enum km_type {
+  KM_BOUNCE_READ,
+  KM_SKB_SUNRPC_DATA,
+  KM_SKB_DATA_SOFTIRQ,
+  KM_USER0,
+  KM_USER1,
+  KM_BIO_SRC_IRQ,
+  KM_BIO_DST_IRQ,
+  KM_PTE0,
+  KM_PTE1,
+  KM_IRQ0,
+  KM_IRQ1,
+  KM_SOFTIRQ0,
+  KM_SOFTIRQ1,
+  KM_TYPE_NR
+};
+
+#endif	/* _XTENSA_KMAP_TYPES_H */
diff --git a/include/asm-xtensa/linkage.h b/include/asm-xtensa/linkage.h
new file mode 100644
index 00000000000..bf2128a99d7
--- /dev/null
+++ b/include/asm-xtensa/linkage.h
@@ -0,0 +1,16 @@
+/*
+ * include/asm-xtensa/linkage.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_LINKAGE_H
+#define _XTENSA_LINKAGE_H
+
+/* Nothing to do here ... */
+
+#endif	/* _XTENSA_LINKAGE_H */
diff --git a/include/asm-xtensa/local.h b/include/asm-xtensa/local.h
new file mode 100644
index 00000000000..48723e550d1
--- /dev/null
+++ b/include/asm-xtensa/local.h
@@ -0,0 +1,16 @@
+/*
+ * include/asm-xtensa/local.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_LOCAL_H
+#define _XTENSA_LOCAL_H
+
+#include <asm-generic/local.h>
+
+#endif /* _XTENSA_LOCAL_H */
diff --git a/include/asm-xtensa/mman.h b/include/asm-xtensa/mman.h
new file mode 100644
index 00000000000..9a95a45df99
--- /dev/null
+++ b/include/asm-xtensa/mman.h
@@ -0,0 +1,80 @@
+/*
+ * include/asm-xtensa/mman.h
+ *
+ * Xtensa Processor memory-manager definitions
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1995 by Ralf Baechle
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_MMAN_H
+#define _XTENSA_MMAN_H
+
+/*
+ * Protections are chosen from these bits, OR'd together.  The
+ * implementation does not necessarily support PROT_EXEC or PROT_WRITE
+ * without PROT_READ.  The only guarantees are that no writing will be
+ * allowed without PROT_WRITE and no access will be allowed for PROT_NONE.
+ */
+
+#define PROT_NONE	0x0		/* page can not be accessed */
+#define PROT_READ	0x1		/* page can be read */
+#define PROT_WRITE	0x2		/* page can be written */
+#define PROT_EXEC	0x4		/* page can be executed */
+
+#define PROT_SEM	0x10		/* page may be used for atomic ops */
+#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
+#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end fo growsup vma */
+
+/*
+ * Flags for mmap
+ */
+#define MAP_SHARED	0x001		/* Share changes */
+#define MAP_PRIVATE	0x002		/* Changes are private */
+#define MAP_TYPE	0x00f		/* Mask for type of mapping */
+#define MAP_FIXED	0x010		/* Interpret addr exactly */
+
+/* not used by linux, but here to make sure we don't clash with ABI defines */
+#define MAP_RENAME	0x020		/* Assign page to file */
+#define MAP_AUTOGROW	0x040		/* File may grow by writing */
+#define MAP_LOCAL	0x080		/* Copy on fork/sproc */
+#define MAP_AUTORSRV	0x100		/* Logical swap reserved on demand */
+
+/* These are linux-specific */
+#define MAP_NORESERVE	0x0400		/* don't check for reservations */
+#define MAP_ANONYMOUS	0x0800		/* don't use a file */
+#define MAP_GROWSDOWN	0x1000		/* stack-like segment */
+#define MAP_DENYWRITE	0x2000		/* ETXTBSY */
+#define MAP_EXECUTABLE	0x4000		/* mark it as an executable */
+#define MAP_LOCKED	0x8000		/* pages are locked */
+#define MAP_POPULATE	0x10000		/* populate (prefault) pagetables */
+#define MAP_NONBLOCK	0x20000		/* do not block on IO */
+
+/*
+ * Flags for msync
+ */
+#define MS_ASYNC	0x0001		/* sync memory asynchronously */
+#define MS_INVALIDATE	0x0002		/* invalidate mappings & caches */
+#define MS_SYNC		0x0004		/* synchronous memory sync */
+
+/*
+ * Flags for mlockall
+ */
+#define MCL_CURRENT	1		/* lock all current mappings */
+#define MCL_FUTURE	2		/* lock all future mappings */
+
+#define MADV_NORMAL	0x0		/* default page-in behavior */
+#define MADV_RANDOM	0x1		/* page-in minimum required */
+#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
+#define MADV_WILLNEED	0x3		/* pre-fault pages */
+#define MADV_DONTNEED	0x4		/* discard these pages */
+
+/* compatibility flags */
+#define MAP_ANON       MAP_ANONYMOUS
+#define MAP_FILE       0
+
+#endif /* _XTENSA_MMAN_H */
diff --git a/include/asm-xtensa/mmu.h b/include/asm-xtensa/mmu.h
new file mode 100644
index 00000000000..44c5bb04c55
--- /dev/null
+++ b/include/asm-xtensa/mmu.h
@@ -0,0 +1,17 @@
+/*
+ * include/asm-xtensa/mmu.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_MMU_H
+#define _XTENSA_MMU_H
+
+/* Default "unsigned long" context */
+typedef unsigned long mm_context_t;
+
+#endif	/* _XTENSA_MMU_H */
diff --git a/include/asm-xtensa/mmu_context.h b/include/asm-xtensa/mmu_context.h
new file mode 100644
index 00000000000..1b0801548cd
--- /dev/null
+++ b/include/asm-xtensa/mmu_context.h
@@ -0,0 +1,330 @@
+/*
+ * include/asm-xtensa/mmu_context.h
+ *
+ * Switch an MMU context.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_MMU_CONTEXT_H
+#define _XTENSA_MMU_CONTEXT_H
+
+#include <linux/config.h>
+#include <linux/stringify.h>
+
+#include <asm/pgtable.h>
+#include <asm/mmu_context.h>
+#include <asm/cacheflush.h>
+#include <asm/tlbflush.h>
+
+/*
+ * Linux was ported to Xtensa assuming all auto-refill ways in set 0
+ * had the same properties (a very likely assumption).  Multiple sets
+ * of auto-refill ways will still work properly, but not as optimally
+ * as the Xtensa designer may have assumed.
+ *
+ * We make this case a hard #error, killing the kernel build, to alert
+ * the developer to this condition (which is more likely an error).
+ * You super-duper clever developers can change it to a warning or
+ * remove it altogether if you think you know what you're doing.  :)
+ */
+
+#if (XCHAL_HAVE_TLBS != 1)
+# error "Linux must have an MMU!"
+#endif
+
+#if ((XCHAL_ITLB_ARF_WAYS == 0) || (XCHAL_DTLB_ARF_WAYS == 0))
+# error "MMU must have auto-refill ways"
+#endif
+
+#if ((XCHAL_ITLB_ARF_SETS != 1) || (XCHAL_DTLB_ARF_SETS != 1))
+# error Linux may not use all auto-refill ways as efficiently as you think
+#endif
+
+#if (XCHAL_MMU_MAX_PTE_PAGE_SIZE != XCHAL_MMU_MIN_PTE_PAGE_SIZE)
+# error Only one page size allowed!
+#endif
+
+extern unsigned long asid_cache;
+extern pgd_t *current_pgd;
+
+/*
+ * Define the number of entries per auto-refill way in set 0 of both I and D
+ * TLBs.  We deal only with set 0 here (an assumption further explained in
+ * assertions.h).  Also, define the total number of ARF entries in both TLBs.
+ */
+
+#define ITLB_ENTRIES_PER_ARF_WAY  (XCHAL_ITLB_SET(XCHAL_ITLB_ARF_SET0,ENTRIES))
+#define DTLB_ENTRIES_PER_ARF_WAY  (XCHAL_DTLB_SET(XCHAL_DTLB_ARF_SET0,ENTRIES))
+
+#define ITLB_ENTRIES							\
+ 	(ITLB_ENTRIES_PER_ARF_WAY * (XCHAL_ITLB_SET(XCHAL_ITLB_ARF_SET0,WAYS)))
+#define DTLB_ENTRIES							\
+	(DTLB_ENTRIES_PER_ARF_WAY * (XCHAL_DTLB_SET(XCHAL_DTLB_ARF_SET0,WAYS)))
+
+
+/*
+ * SMALLEST_NTLB_ENTRIES is the smaller of ITLB_ENTRIES and DTLB_ENTRIES.
+ * In practice, they are probably equal.  This macro simplifies function
+ * flush_tlb_range().
+ */
+
+#if (DTLB_ENTRIES < ITLB_ENTRIES)
+# define SMALLEST_NTLB_ENTRIES  DTLB_ENTRIES
+#else
+# define SMALLEST_NTLB_ENTRIES  ITLB_ENTRIES
+#endif
+
+
+/*
+ * asid_cache tracks only the ASID[USER_RING] field of the RASID special
+ * register, which is the current user-task asid allocation value.
+ * mm->context has the same meaning.  When it comes time to write the
+ * asid_cache or mm->context values to the RASID special register, we first
+ * shift the value left by 8, then insert the value.
+ * ASID[0] always contains the kernel's asid value, and we reserve three
+ * other asid values that we never assign to user tasks.
+ */
+
+#define ASID_INC	0x1
+#define ASID_MASK	((1 << XCHAL_MMU_ASID_BITS) - 1)
+
+/*
+ * XCHAL_MMU_ASID_INVALID is a configurable Xtensa processor constant
+ * indicating invalid address space.  XCHAL_MMU_ASID_KERNEL is a configurable
+ * Xtensa processor constant indicating the kernel address space.  They can
+ * be arbitrary values.
+ *
+ * We identify three more unique, reserved ASID values to use in the unused
+ * ring positions.  No other user process will be assigned these reserved
+ * ASID values.
+ *
+ * For example, given that
+ *
+ *	XCHAL_MMU_ASID_INVALID == 0
+ *	XCHAL_MMU_ASID_KERNEL  == 1
+ *
+ * the following maze of #if statements would generate
+ *
+ *	ASID_RESERVED_1        == 2
+ *	ASID_RESERVED_2        == 3
+ *	ASID_RESERVED_3        == 4
+ *	ASID_FIRST_NONRESERVED == 5
+ */
+
+#if (XCHAL_MMU_ASID_INVALID != XCHAL_MMU_ASID_KERNEL + 1)
+# define ASID_RESERVED_1         ((XCHAL_MMU_ASID_KERNEL + 1) & ASID_MASK)
+#else
+# define ASID_RESERVED_1         ((XCHAL_MMU_ASID_KERNEL + 2) & ASID_MASK)
+#endif
+
+#if (XCHAL_MMU_ASID_INVALID != ASID_RESERVED_1 + 1)
+# define ASID_RESERVED_2         ((ASID_RESERVED_1 + 1) & ASID_MASK)
+#else
+# define ASID_RESERVED_2         ((ASID_RESERVED_1 + 2) & ASID_MASK)
+#endif
+
+#if (XCHAL_MMU_ASID_INVALID != ASID_RESERVED_2 + 1)
+# define ASID_RESERVED_3         ((ASID_RESERVED_2 + 1) & ASID_MASK)
+#else
+# define ASID_RESERVED_3         ((ASID_RESERVED_2 + 2) & ASID_MASK)
+#endif
+
+#if (XCHAL_MMU_ASID_INVALID != ASID_RESERVED_3 + 1)
+# define ASID_FIRST_NONRESERVED  ((ASID_RESERVED_3 + 1) & ASID_MASK)
+#else
+# define ASID_FIRST_NONRESERVED  ((ASID_RESERVED_3 + 2) & ASID_MASK)
+#endif
+
+#define ASID_ALL_RESERVED ( ((ASID_RESERVED_1) << 24) + \
+                            ((ASID_RESERVED_2) << 16) + \
+                            ((ASID_RESERVED_3) <<  8) + \
+                            ((XCHAL_MMU_ASID_KERNEL))   )
+
+
+/*
+ * NO_CONTEXT is the invalid ASID value that we don't ever assign to
+ * any user or kernel context.  NO_CONTEXT is a better mnemonic than
+ * XCHAL_MMU_ASID_INVALID, so we use it in code instead.
+ */
+
+#define NO_CONTEXT   XCHAL_MMU_ASID_INVALID
+
+#if (KERNEL_RING != 0)
+# error The KERNEL_RING really should be zero.
+#endif
+
+#if (USER_RING >= XCHAL_MMU_RINGS)
+# error USER_RING cannot be greater than the highest numbered ring.
+#endif
+
+#if (USER_RING == KERNEL_RING)
+# error The user and kernel rings really should not be equal.
+#endif
+
+#if (USER_RING == 1)
+#define ASID_INSERT(x) ( ((ASID_RESERVED_1)   << 24) + \
+                         ((ASID_RESERVED_2)   << 16) + \
+                         (((x) & (ASID_MASK)) <<  8) + \
+                         ((XCHAL_MMU_ASID_KERNEL))   )
+
+#elif (USER_RING == 2)
+#define ASID_INSERT(x) ( ((ASID_RESERVED_1)   << 24) + \
+                         (((x) & (ASID_MASK)) << 16) + \
+                         ((ASID_RESERVED_2)   <<  8) + \
+                         ((XCHAL_MMU_ASID_KERNEL))   )
+
+#elif (USER_RING == 3)
+#define ASID_INSERT(x) ( (((x) & (ASID_MASK)) << 24) + \
+			 ((ASID_RESERVED_1)   << 16) + \
+                         ((ASID_RESERVED_2)   <<  8) + \
+                         ((XCHAL_MMU_ASID_KERNEL))   )
+
+#else
+#error Goofy value for USER_RING
+
+#endif /* USER_RING == 1 */
+
+
+/*
+ *  All unused by hardware upper bits will be considered
+ *  as a software asid extension.
+ */
+
+#define ASID_VERSION_MASK  ((unsigned long)~(ASID_MASK|(ASID_MASK-1)))
+#define ASID_FIRST_VERSION						\
+	((unsigned long)(~ASID_VERSION_MASK) + 1 + ASID_FIRST_NONRESERVED)
+
+extern inline void set_rasid_register (unsigned long val)
+{
+	__asm__ __volatile__ (" wsr %0, "__stringify(RASID)"\n\t"
+			      " isync\n" : : "a" (val));
+}
+
+extern inline unsigned long get_rasid_register (void)
+{
+	unsigned long tmp;
+	__asm__ __volatile__ (" rsr %0, "__stringify(RASID)"\n\t" : "=a" (tmp));
+	return tmp;
+}
+
+
+#if ((XCHAL_MMU_ASID_INVALID == 0) && (XCHAL_MMU_ASID_KERNEL == 1))
+
+extern inline void
+get_new_mmu_context(struct mm_struct *mm, unsigned long asid)
+{
+	extern void flush_tlb_all(void);
+	if (! ((asid += ASID_INC) & ASID_MASK) ) {
+		flush_tlb_all(); /* start new asid cycle */
+		if (!asid)      /* fix version if needed */
+			asid = ASID_FIRST_VERSION - ASID_FIRST_NONRESERVED;
+		asid += ASID_FIRST_NONRESERVED;
+	}
+	mm->context = asid_cache = asid;
+}
+
+#else
+#warning ASID_{INVALID,KERNEL} values impose non-optimal get_new_mmu_context implementation
+
+/* XCHAL_MMU_ASID_INVALID == 0 and XCHAL_MMU_ASID_KERNEL ==1 are
+   really the best, but if you insist... */
+
+extern inline int validate_asid (unsigned long asid)
+{
+	switch (asid) {
+	case XCHAL_MMU_ASID_INVALID:
+	case XCHAL_MMU_ASID_KERNEL:
+	case ASID_RESERVED_1:
+	case ASID_RESERVED_2:
+	case ASID_RESERVED_3:
+		return 0; /* can't use these values as ASIDs */
+	}
+	return 1; /* valid */
+}
+
+extern inline void
+get_new_mmu_context(struct mm_struct *mm, unsigned long asid)
+{
+	extern void flush_tlb_all(void);
+	while (1) {
+		asid += ASID_INC;
+		if ( ! (asid & ASID_MASK) ) {
+			flush_tlb_all(); /* start new asid cycle */
+			if (!asid)      /* fix version if needed */
+				asid = ASID_FIRST_VERSION - ASID_FIRST_NONRESERVED;
+			asid += ASID_FIRST_NONRESERVED;
+			break; /* no need to validate here */
+		}
+		if (validate_asid (asid & ASID_MASK))
+			break;
+	}
+	mm->context = asid_cache = asid;
+}
+
+#endif
+
+
+/*
+ * Initialize the context related info for a new mm_struct
+ * instance.
+ */
+
+extern inline int
+init_new_context(struct task_struct *tsk, struct mm_struct *mm)
+{
+	mm->context = NO_CONTEXT;
+	return 0;
+}
+
+extern inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
+                             struct task_struct *tsk)
+{
+	unsigned long asid = asid_cache;
+
+	/* Check if our ASID is of an older version and thus invalid */
+
+	if ((next->context ^ asid) & ASID_VERSION_MASK)
+		get_new_mmu_context(next, asid);
+
+	set_rasid_register (ASID_INSERT(next->context));
+	invalidate_page_directory();
+}
+
+#define deactivate_mm(tsk, mm)	do { } while(0)
+
+/*
+ * Destroy context related info for an mm_struct that is about
+ * to be put to rest.
+ */
+extern inline void destroy_context(struct mm_struct *mm)
+{
+	/* Nothing to do. */
+}
+
+/*
+ * After we have set current->mm to a new value, this activates
+ * the context for the new mm so we see the new mappings.
+ */
+extern inline void
+activate_mm(struct mm_struct *prev, struct mm_struct *next)
+{
+	/* Unconditionally get a new ASID.  */
+
+	get_new_mmu_context(next, asid_cache);
+	set_rasid_register (ASID_INSERT(next->context));
+	invalidate_page_directory();
+}
+
+
+static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
+{
+	/* Nothing to do. */
+
+}
+
+#endif /* _XTENSA_MMU_CONTEXT_H */
diff --git a/include/asm-xtensa/module.h b/include/asm-xtensa/module.h
new file mode 100644
index 00000000000..ffb25bfdf6a
--- /dev/null
+++ b/include/asm-xtensa/module.h
@@ -0,0 +1,25 @@
+/*
+ * include/asm-xtensa/module.h
+ *
+ * This file contains the module code specific to the Xtensa architecture.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_MODULE_H
+#define _XTENSA_MODULE_H
+
+struct mod_arch_specific
+{
+	/* Module support is not completely implemented. */
+};
+
+#define Elf_Shdr Elf32_Shdr
+#define Elf_Sym Elf32_Sym
+#define Elf_Ehdr Elf32_Ehdr
+
+#endif	/* _XTENSA_MODULE_H */
diff --git a/include/asm-xtensa/msgbuf.h b/include/asm-xtensa/msgbuf.h
new file mode 100644
index 00000000000..693c9675528
--- /dev/null
+++ b/include/asm-xtensa/msgbuf.h
@@ -0,0 +1,48 @@
+/*
+ * include/asm-xtensa/msgbuf.h
+ *
+ * The msqid64_ds structure for the Xtensa architecture.
+ * Note extra padding because this structure is passed back and forth
+ * between kernel and user space.
+ *
+ * Pad space is left for:
+ * - 64-bit time_t to solve y2038 problem
+ * - 2 miscellaneous 32-bit values
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ */
+
+#ifndef _XTENSA_MSGBUF_H
+#define _XTENSA_MSGBUF_H
+
+struct msqid64_ds {
+	struct ipc64_perm msg_perm;
+#ifdef __XTENSA_EB__
+	unsigned int	__unused1;
+	__kernel_time_t msg_stime;	/* last msgsnd time */
+	unsigned int	__unused2;
+	__kernel_time_t msg_rtime;	/* last msgrcv time */
+	unsigned int	__unused3;
+	__kernel_time_t msg_ctime;	/* last change time */
+#elif defined(__XTENSA_EL__)
+	__kernel_time_t msg_stime;	/* last msgsnd time */
+	unsigned int	__unused1;
+	__kernel_time_t msg_rtime;	/* last msgrcv time */
+	unsigned int	__unused2;
+	__kernel_time_t msg_ctime;	/* last change time */
+	unsigned int	__unused3;
+#else
+# error processor byte order undefined!
+#endif
+	unsigned long  msg_cbytes;	/* current number of bytes on queue */
+	unsigned long  msg_qnum;	/* number of messages in queue */
+	unsigned long  msg_qbytes;	/* max number of bytes on queue */
+	__kernel_pid_t msg_lspid;	/* pid of last msgsnd */
+	__kernel_pid_t msg_lrpid;	/* last receive pid */
+	unsigned long  __unused4;
+	unsigned long  __unused5;
+};
+
+#endif	/* _XTENSA_MSGBUF_H */
diff --git a/include/asm-xtensa/namei.h b/include/asm-xtensa/namei.h
new file mode 100644
index 00000000000..3fdff039d27
--- /dev/null
+++ b/include/asm-xtensa/namei.h
@@ -0,0 +1,26 @@
+/*
+ * include/asm-xtensa/namei.h
+ *
+ * Included from linux/fs/namei.c
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_NAMEI_H
+#define _XTENSA_NAMEI_H
+
+#ifdef __KERNEL__
+
+/* This dummy routine maybe changed to something useful
+ * for /usr/gnemul/ emulation stuff.
+ * Look at asm-sparc/namei.h for details.
+ */
+
+#define __emul_prefix() NULL
+
+#endif	/* __KERNEL__ */
+#endif	/* _XTENSA_NAMEI_H */
diff --git a/include/asm-xtensa/page.h b/include/asm-xtensa/page.h
new file mode 100644
index 00000000000..b495e5b5a94
--- /dev/null
+++ b/include/asm-xtensa/page.h
@@ -0,0 +1,133 @@
+/*
+ * linux/include/asm-xtensa/page.h
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version2 as
+ * published by the Free Software Foundation.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_PAGE_H
+#define _XTENSA_PAGE_H
+
+#ifdef __KERNEL__
+
+#include <asm/processor.h>
+#include <linux/config.h>
+
+/*
+ * PAGE_SHIFT determines the page size
+ * PAGE_ALIGN(x) aligns the pointer to the (next) page boundary
+ */
+
+#define PAGE_SHIFT		XCHAL_MMU_MIN_PTE_PAGE_SIZE
+#define PAGE_SIZE		(1 << PAGE_SHIFT)
+#define PAGE_MASK		(~(PAGE_SIZE-1))
+#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE - 1) & PAGE_MASK)
+
+#define DCACHE_WAY_SIZE		(XCHAL_DCACHE_SIZE / XCHAL_DCACHE_WAYS)
+#define PAGE_OFFSET		XCHAL_KSEG_CACHED_VADDR
+
+#ifdef __ASSEMBLY__
+
+#define __pgprot(x)	(x)
+
+#else
+
+/*
+ * These are used to make use of C type-checking..
+ */
+
+typedef struct { unsigned long pte; } pte_t;		/* page table entry */
+typedef struct { unsigned long pgd; } pgd_t;		/* PGD table entry */
+typedef struct { unsigned long pgprot; } pgprot_t;
+
+#define pte_val(x)	((x).pte)
+#define pgd_val(x)	((x).pgd)
+#define pgprot_val(x)	((x).pgprot)
+
+#define __pte(x)	((pte_t) { (x) } )
+#define __pgd(x)	((pgd_t) { (x) } )
+#define __pgprot(x)	((pgprot_t) { (x) } )
+
+/*
+ * Pure 2^n version of get_order
+ */
+
+extern __inline__ int get_order(unsigned long size)
+{
+	int order;
+#ifndef XCHAL_HAVE_NSU
+	unsigned long x1, x2, x4, x8, x16;
+
+	size = (size + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	x1  = size & 0xAAAAAAAA;
+	x2  = size & 0xCCCCCCCC;
+	x4  = size & 0xF0F0F0F0;
+	x8  = size & 0xFF00FF00;
+	x16 = size & 0xFFFF0000;
+	order = x2 ? 2 : 0;
+	order += (x16 != 0) * 16;
+	order += (x8 != 0) * 8;
+	order += (x4 != 0) * 4;
+	order += (x1 != 0);
+
+	return order;
+#else
+	size = (size - 1) >> PAGE_SHIFT;
+	asm ("nsau %0, %1" : "=r" (order) : "r" (size));
+	return 32 - order;
+#endif
+}
+
+
+struct page;
+extern void clear_page(void *page);
+extern void copy_page(void *to, void *from);
+
+/*
+ * If we have cache aliasing and writeback caches, we might have to do
+ * some extra work
+ */
+
+#if (DCACHE_WAY_SIZE > PAGE_SIZE)
+void clear_user_page(void *addr, unsigned long vaddr, struct page* page);
+void copy_user_page(void *to,void* from,unsigned long vaddr,struct page* page);
+#else
+# define clear_user_page(page,vaddr,pg)		clear_page(page)
+# define copy_user_page(to, from, vaddr, pg)	copy_page(to, from)
+#endif
+
+/*
+ * This handles the memory map.  We handle pages at
+ * XCHAL_KSEG_CACHED_VADDR for kernels with 32 bit address space.
+ * These macros are for conversion of kernel address, not user
+ * addresses.
+ */
+
+#define __pa(x)			((unsigned long) (x) - PAGE_OFFSET)
+#define __va(x)			((void *)((unsigned long) (x) + PAGE_OFFSET))
+#define pfn_valid(pfn)		((unsigned long)pfn < max_mapnr)
+#ifndef CONFIG_DISCONTIGMEM
+# define pfn_to_page(pfn)	(mem_map + (pfn))
+# define page_to_pfn(page)	((unsigned long)((page) - mem_map))
+#else
+# error CONFIG_DISCONTIGMEM not supported
+#endif
+
+#define virt_to_page(kaddr)	pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
+#define page_to_virt(page)	__va(page_to_pfn(page) << PAGE_SHIFT)
+#define virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
+#define page_to_phys(page)	(page_to_pfn(page) << PAGE_SHIFT)
+
+#define WANT_PAGE_VIRTUAL
+
+
+#endif /* __ASSEMBLY__ */
+
+#define VM_DATA_DEFAULT_FLAGS	(VM_READ | VM_WRITE | VM_EXEC | \
+				 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+
+#endif /* __KERNEL__ */
+#endif /* _XTENSA_PAGE_H */
diff --git a/include/asm-xtensa/page.h.n b/include/asm-xtensa/page.h.n
new file mode 100644
index 00000000000..546cc6624f2
--- /dev/null
+++ b/include/asm-xtensa/page.h.n
@@ -0,0 +1,135 @@
+/*
+ * linux/include/asm-xtensa/page.h
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version2 as
+ * published by the Free Software Foundation.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_PAGE_H
+#define _XTENSA_PAGE_H
+
+#ifdef __KERNEL__
+
+#include <asm/processor.h>
+#include <linux/config.h>
+
+/*
+ * PAGE_SHIFT determines the page size
+ * PAGE_ALIGN(x) aligns the pointer to the (next) page boundary
+ */
+#define PAGE_SHIFT		XCHAL_MMU_MIN_PTE_PAGE_SIZE
+#define PAGE_SIZE		(1 << PAGE_SHIFT)
+#define PAGE_MASK		(~(PAGE_SIZE-1))
+#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE - 1) & PAGE_MASK)
+
+#define DCACHE_WAY_SIZE		(XCHAL_DCACHE_SIZE / XCHAL_DCACHE_WAYS)
+#define PAGE_OFFSET		XCHAL_KSEG_CACHED_VADDR
+
+#ifdef __ASSEMBLY__
+
+#define __pgprot(x)	(x)
+
+#else
+
+
+/*
+ * These are used to make use of C type-checking..
+ */
+typedef struct { unsigned long pte; } pte_t;		/* page table entry */
+typedef struct { unsigned long pmd; } pmd_t;		/* PMD table entry */
+typedef struct { unsigned long pgd; } pgd_t;		/* PGD table entry */
+typedef struct { unsigned long pgprot; } pgprot_t;
+
+#define pte_val(x)	((x).pte)
+#define pmd_val(x)	((x).pmd)
+#define pgd_val(x)	((x).pgd)
+#define pgprot_val(x)	((x).pgprot)
+
+#define __pte(x)	((pte_t) { (x) } )
+#define __pmd(x)	((pmd_t) { (x) } )
+#define __pgd(x)	((pgd_t) { (x) } )
+#define __pgprot(x)	((pgprot_t) { (x) } )
+
+/*
+ * Pure 2^n version of get_order
+ */
+extern __inline__ int get_order(unsigned long size)
+{
+	int order;
+#ifndef XCHAL_HAVE_NSU
+	unsigned long x1, x2, x4, x8, x16;
+
+	size = (size + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	x1  = size & 0xAAAAAAAA;
+	x2  = size & 0xCCCCCCCC;
+	x4  = size & 0xF0F0F0F0;
+	x8  = size & 0xFF00FF00;
+	x16 = size & 0xFFFF0000;
+	order = x2 ? 2 : 0;
+	order += (x16 != 0) * 16;
+	order += (x8 != 0) * 8;
+	order += (x4 != 0) * 4;
+	order += (x1 != 0);
+
+	return order;
+#else
+	size = (size - 1) >> PAGE_SHIFT;
+	asm ("nsau %0, %1" : "=r" (order) : "r" (size));
+	return 32 - order;
+#endif
+}
+
+
+struct page;
+extern void clear_page(void *page);
+extern void copy_page(void *to, void *from);
+
+/*
+ * If we have cache aliasing and writeback caches, we might have to do
+ * some extra work
+ */
+
+#if (DCACHE_WAY_SIZE > PAGE_SIZE) && XCHAL_DCACHE_IS_WRITEBACK
+void clear_user_page(void *addr, unsigned long vaddr, struct page* page);
+void copy_user_page(void *to, void* from, unsigned long vaddr, struct page* page);
+#else
+# define clear_user_page(page,vaddr,pg)		clear_page(page)
+# define copy_user_page(to, from, vaddr, pg)	copy_page(to, from)
+#endif
+
+
+/*
+ * This handles the memory map.  We handle pages at
+ * XCHAL_KSEG_CACHED_VADDR for kernels with 32 bit address space.
+ * These macros are for conversion of kernel address, not user
+ * addresses.
+ */
+
+#define __pa(x)			((unsigned long) (x) - PAGE_OFFSET)
+#define __va(x)			((void *)((unsigned long) (x) + PAGE_OFFSET))
+#define pfn_valid(pfn)		((unsigned long)pfn < max_mapnr)
+#ifndef CONFIG_DISCONTIGMEM
+# define pfn_to_page(pfn)	(mem_map + (pfn))
+# define page_to_pfn(page)	((unsigned long)((page) - mem_map))
+#else
+# error CONFIG_DISCONTIGMEM not supported
+#endif
+
+#define virt_to_page(kaddr)	pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
+#define page_to_virt(page)	__va(page_to_pfn(page) << PAGE_SHIFT)
+#define virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
+#define page_to_phys(page)	(page_to_pfn(page) << PAGE_SHIFT)
+
+#define WANT_PAGE_VIRTUAL
+
+
+#endif /* __ASSEMBLY__ */
+
+#define VM_DATA_DEFAULT_FLAGS	(VM_READ | VM_WRITE | VM_EXEC | \
+				 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+
+#endif /* __KERNEL__ */
+#endif /* _XTENSA_PAGE_H */
diff --git a/include/asm-xtensa/param.h b/include/asm-xtensa/param.h
new file mode 100644
index 00000000000..c0eec8260b0
--- /dev/null
+++ b/include/asm-xtensa/param.h
@@ -0,0 +1,34 @@
+/*
+ * include/asm-xtensa/param.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_PARAM_H
+#define _XTENSA_PARAM_H
+
+#include <xtensa/config/core.h>
+
+#ifdef __KERNEL__
+# define HZ		100		/* internal timer frequency */
+# define USER_HZ	100		/* for user interfaces in "ticks" */
+# define CLOCKS_PER_SEC (USER_HZ)	/* frequnzy at which times() counts */
+#endif
+
+#define EXEC_PAGESIZE	(1 << XCHAL_MMU_MIN_PTE_PAGE_SIZE)
+
+#ifndef NGROUPS
+#define NGROUPS		32
+#endif
+
+#ifndef NOGROUP
+#define NOGROUP		(-1)
+#endif
+
+#define MAXHOSTNAMELEN	64	/* max length of hostname */
+
+#endif /* _XTENSA_PARAM_H */
diff --git a/include/asm-xtensa/pci-bridge.h b/include/asm-xtensa/pci-bridge.h
new file mode 100644
index 00000000000..00fcbd7c534
--- /dev/null
+++ b/include/asm-xtensa/pci-bridge.h
@@ -0,0 +1,88 @@
+/*
+ * include/asm-xtensa/pci-bridge.h
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ *
+ * Copyright (C) 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_PCI_BRIDGE_H
+#define _XTENSA_PCI_BRIDGE_H
+
+#ifdef __KERNEL__
+
+struct device_node;
+struct pci_controller;
+
+/*
+ * pciauto_bus_scan() enumerates the pci space.
+ */
+
+extern int pciauto_bus_scan(struct pci_controller *, int);
+
+struct pci_space {
+	unsigned long start;
+	unsigned long end;
+	unsigned long base;
+};
+
+/*
+ * Structure of a PCI controller (host bridge)
+ */
+
+struct pci_controller {
+	int index;			/* used for pci_controller_num */
+	struct pci_controller *next;
+        struct pci_bus *bus;
+	void *arch_data;
+
+	int first_busno;
+	int last_busno;
+
+	struct pci_ops *ops;
+	volatile unsigned int *cfg_addr;
+	volatile unsigned char *cfg_data;
+
+	/* Currently, we limit ourselves to 1 IO range and 3 mem
+	 * ranges since the common pci_bus structure can't handle more
+	 */
+	struct resource	io_resource;
+	struct resource mem_resources[3];
+	int mem_resource_count;
+
+	/* Host bridge I/O and Memory space
+	 * Used for BAR placement algorithms
+	 */
+	struct pci_space io_space;
+	struct pci_space mem_space;
+
+	/* Return the interrupt number fo a device. */
+	int (*map_irq)(struct pci_dev*, u8, u8);
+
+};
+
+static inline void pcibios_init_resource(struct resource *res,
+		unsigned long start, unsigned long end, int flags, char *name)
+{
+	res->start = start;
+	res->end = end;
+	res->flags = flags;
+	res->name = name;
+	res->parent = NULL;
+	res->sibling = NULL;
+	res->child = NULL;
+}
+
+
+/* These are used for config access before all the PCI probing has been done. */
+int early_read_config_byte(struct pci_controller*, int, int, int, u8*);
+int early_read_config_word(struct pci_controller*, int, int, int, u16*);
+int early_read_config_dword(struct pci_controller*, int, int, int, u32*);
+int early_write_config_byte(struct pci_controller*, int, int, int, u8);
+int early_write_config_word(struct pci_controller*, int, int, int, u16);
+int early_write_config_dword(struct pci_controller*, int, int, int, u32);
+
+#endif	/* __KERNEL__ */
+#endif	/* _XTENSA_PCI_BRIDGE_H */
diff --git a/include/asm-xtensa/pci.h b/include/asm-xtensa/pci.h
new file mode 100644
index 00000000000..6817742301c
--- /dev/null
+++ b/include/asm-xtensa/pci.h
@@ -0,0 +1,89 @@
+/*
+ * linux/include/asm-xtensa/pci.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_PCI_H
+#define _XTENSA_PCI_H
+
+#ifdef __KERNEL__
+
+/* Can be used to override the logic in pci_scan_bus for skipping
+ * already-configured bus numbers - to be used for buggy BIOSes
+ * or architectures with incomplete PCI setup by the loader
+ */
+
+#define pcibios_assign_all_busses()	0
+
+extern struct pci_controller* pcibios_alloc_controller(void);
+
+extern inline void pcibios_set_master(struct pci_dev *dev)
+{
+	/* No special bus mastering setup handling */
+}
+
+extern inline void pcibios_penalize_isa_irq(int irq)
+{
+	/* We don't do dynamic PCI IRQ allocation */
+}
+
+/* Assume some values. (We should revise them, if necessary) */
+
+#define PCIBIOS_MIN_IO		0x2000
+#define PCIBIOS_MIN_MEM		0x10000000
+
+/* Dynamic DMA mapping stuff.
+ * Xtensa has everything mapped statically like x86.
+ */
+
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <asm/scatterlist.h>
+#include <linux/string.h>
+#include <asm/io.h>
+
+struct pci_dev;
+
+/* The PCI address space does equal the physical memory address space.
+ * The networking and block device layers use this boolean for bounce buffer
+ * decisions.
+ */
+
+#define PCI_DMA_BUS_IS_PHYS	(1)
+
+/* pci_unmap_{page,single} is a no-op, so */
+#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME)
+#define DECLARE_PCI_UNMAP_LEN(LEN_NAME)
+#define pci_unmap_addr(PTR, ADDR_NAME)		(0)
+#define pci_unmap_addr_set(PTR, ADDR_NAME, VAL)	do { } while (0)
+#define pci_ubnmap_len(PTR, LEN_NAME)		(0)
+#define pci_unmap_len_set(PTR, LEN_NAME, VAL)	do { } while (0)
+
+/* We cannot access memory above 4GB */
+#define pci_dac_dma_supported(pci_dev, mask)	(0)
+
+/* Map a range of PCI memory or I/O space for a device into user space */
+int pci_mmap_page_range(struct pci_dev *pdev, struct vm_area_struct *vma,
+                        enum pci_mmap_state mmap_state, int write_combine);
+
+/* Tell drivers/pci/proc.c that we have pci_mmap_page_range() */
+#define HAVE_PCI_MMAP	1
+
+static inline void pcibios_add_platform_entries(struct pci_dev *dev)
+{
+}
+
+#endif /* __KERNEL__ */
+
+/* Implement the pci_ DMA API in terms of the generic device dma_ one */
+#include <asm-generic/pci-dma-compat.h>
+
+/* Generic PCI */
+#include <asm-generic/pci.h>
+
+#endif	/* _XTENSA_PCI_H */
diff --git a/include/asm-xtensa/percpu.h b/include/asm-xtensa/percpu.h
new file mode 100644
index 00000000000..6d2bc2ada9d
--- /dev/null
+++ b/include/asm-xtensa/percpu.h
@@ -0,0 +1,16 @@
+/*
+ * linux/include/asm-xtensa/percpu.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_PERCPU__
+#define _XTENSA_PERCPU__
+
+#include <asm-generic/percpu.h>
+
+#endif /* _XTENSA_PERCPU__ */
diff --git a/include/asm-xtensa/pgalloc.h b/include/asm-xtensa/pgalloc.h
new file mode 100644
index 00000000000..734a8d06039
--- /dev/null
+++ b/include/asm-xtensa/pgalloc.h
@@ -0,0 +1,116 @@
+/*
+ * linux/include/asm-xtensa/pgalloc.h
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Copyright (C) 2001-2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_PGALLOC_H
+#define _XTENSA_PGALLOC_H
+
+#ifdef __KERNEL__
+
+#include <linux/config.h>
+#include <linux/threads.h>
+#include <linux/highmem.h>
+#include <asm/processor.h>
+#include <asm/cacheflush.h>
+
+
+/* Cache aliasing:
+ *
+ * If the cache size for one way is greater than the page size, we have to
+ * deal with cache aliasing. The cache index is wider than the page size:
+ *
+ *      |cache |
+ * |pgnum |page|	virtual address
+ * |xxxxxX|zzzz|
+ * |      |    |
+ *   \  / |    |
+ *  trans.|    |
+ *   /  \ |    |
+ * |yyyyyY|zzzz|	physical address
+ *
+ * When the page number is translated to the physical page address, the lowest
+ * bit(s) (X) that are also part of the cache index are also translated (Y).
+ * If this translation changes this bit (X), the cache index is also afected,
+ * thus resulting in a different cache line than before.
+ * The kernel does not provide a mechanism to ensure that the page color
+ * (represented by this bit) remains the same when allocated or when pages
+ * are remapped. When user pages are mapped into kernel space, the color of
+ * the page might also change.
+ *
+ * We use the address space VMALLOC_END ... VMALLOC_END + DCACHE_WAY_SIZE * 2
+ * to temporarily map a patch so we can match the color.
+ */
+
+#if (DCACHE_WAY_SIZE > PAGE_SIZE)
+# define PAGE_COLOR_MASK	(PAGE_MASK & (DCACHE_WAY_SIZE-1))
+# define PAGE_COLOR(a)		\
+	(((unsigned long)(a)&PAGE_COLOR_MASK) >> PAGE_SHIFT)
+# define PAGE_COLOR_EQ(a,b)	\
+	((((unsigned long)(a) ^ (unsigned long)(b)) & PAGE_COLOR_MASK) == 0)
+# define PAGE_COLOR_MAP0(v)	\
+	(VMALLOC_END + ((unsigned long)(v) & PAGE_COLOR_MASK))
+# define PAGE_COLOR_MAP1(v)	\
+	(VMALLOC_END + ((unsigned long)(v) & PAGE_COLOR_MASK) + DCACHE_WAY_SIZE)
+#endif
+
+/*
+ * Allocating and freeing a pmd is trivial: the 1-entry pmd is
+ * inside the pgd, so has no extra memory associated with it.
+ */
+
+#define pgd_free(pgd)	free_page((unsigned long)(pgd))
+
+#if (DCACHE_WAY_SIZE > PAGE_SIZE) && XCHAL_DCACHE_IS_WRITEBACK
+
+static inline void
+pmd_populate_kernel(struct mm_struct *mm, pmd_t *pmdp, pte_t *pte)
+{
+	pmd_val(*(pmdp)) = (unsigned long)(pte);
+	__asm__ __volatile__ ("memw; dhwb %0, 0; dsync" :: "a" (pmdp));
+}
+
+static inline void
+pmd_populate(struct mm_struct *mm, pmd_t *pmdp, struct page *page)
+{
+	pmd_val(*(pmdp)) = (unsigned long)page_to_virt(page);
+	__asm__ __volatile__ ("memw; dhwb %0, 0; dsync" :: "a" (pmdp));
+}
+
+
+
+#else
+
+# define pmd_populate_kernel(mm, pmdp, pte)				     \
+	(pmd_val(*(pmdp)) = (unsigned long)(pte))
+# define pmd_populate(mm, pmdp, page)					     \
+	(pmd_val(*(pmdp)) = (unsigned long)page_to_virt(page))
+
+#endif
+
+static inline pgd_t*
+pgd_alloc(struct mm_struct *mm)
+{
+	pgd_t *pgd;
+
+	pgd = (pgd_t *)__get_free_pages(GFP_KERNEL|__GFP_ZERO, PGD_ORDER);
+
+	if (likely(pgd != NULL))
+		__flush_dcache_page((unsigned long)pgd);
+
+	return pgd;
+}
+
+extern pte_t* pte_alloc_one_kernel(struct mm_struct* mm, unsigned long addr);
+extern struct page* pte_alloc_one(struct mm_struct* mm, unsigned long addr);
+
+#define pte_free_kernel(pte) free_page((unsigned long)pte)
+#define pte_free(pte) __free_page(pte)
+
+#endif /* __KERNEL__ */
+#endif /* _XTENSA_PGALLOC_H */
diff --git a/include/asm-xtensa/pgtable.h b/include/asm-xtensa/pgtable.h
new file mode 100644
index 00000000000..0bb6416ae26
--- /dev/null
+++ b/include/asm-xtensa/pgtable.h
@@ -0,0 +1,468 @@
+/*
+ * linux/include/asm-xtensa/page.h
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version2 as
+ * published by the Free Software Foundation.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_PGTABLE_H
+#define _XTENSA_PGTABLE_H
+
+#include <asm-generic/pgtable-nopmd.h>
+#include <asm/page.h>
+
+/* Assertions. */
+
+#ifdef CONFIG_MMU
+
+
+#if (XCHAL_MMU_RINGS < 2)
+# error Linux build assumes at least 2 ring levels.
+#endif
+
+#if (XCHAL_MMU_CA_BITS != 4)
+# error We assume exactly four bits for CA.
+#endif
+
+#if (XCHAL_MMU_SR_BITS != 0)
+# error We have no room for SR bits.
+#endif
+
+/*
+ * Use the first min-wired way for mapping page-table pages.
+ * Page coloring requires a second min-wired way.
+ */
+
+#if (XCHAL_DTLB_MINWIRED_SETS == 0)
+# error Need a min-wired way for mapping page-table pages
+#endif
+
+#define DTLB_WAY_PGTABLE XCHAL_DTLB_SET(XCHAL_DTLB_MINWIRED_SET0, WAY)
+
+#if (DCACHE_WAY_SIZE > PAGE_SIZE) && XCHAL_DCACHE_IS_WRITEBACK
+# if XCHAL_DTLB_SET(XCHAL_DTLB_MINWIRED_SET0, WAYS) >= 2
+#  define DTLB_WAY_DCACHE_ALIAS0 (DTLB_WAY_PGTABLE + 1)
+#  define DTLB_WAY_DCACHE_ALIAS1 (DTLB_WAY_PGTABLE + 2)
+# else
+#  error Page coloring requires its own wired dtlb way!
+# endif
+#endif
+
+#endif /* CONFIG_MMU */
+
+/*
+ * We only use two ring levels, user and kernel space.
+ */
+
+#define USER_RING		1	/* user ring level */
+#define KERNEL_RING		0	/* kernel ring level */
+
+/*
+ * The Xtensa architecture port of Linux has a two-level page table system,
+ * i.e. the logical three-level Linux page table layout are folded.
+ * Each task has the following memory page tables:
+ *
+ *   PGD table (page directory), ie. 3rd-level page table:
+ *	One page (4 kB) of 1024 (PTRS_PER_PGD) pointers to PTE tables
+ *	(Architectures that don't have the PMD folded point to the PMD tables)
+ *
+ *	The pointer to the PGD table for a given task can be retrieved from
+ *	the task structure (struct task_struct*) t, e.g. current():
+ *	  (t->mm ? t->mm : t->active_mm)->pgd
+ *
+ *   PMD tables (page middle-directory), ie. 2nd-level page tables:
+ *	Absent for the Xtensa architecture (folded, PTRS_PER_PMD == 1).
+ *
+ *   PTE tables (page table entry), ie. 1st-level page tables:
+ *	One page (4 kB) of 1024 (PTRS_PER_PTE) PTEs with a special PTE
+ *	invalid_pte_table for absent mappings.
+ *
+ * The individual pages are 4 kB big with special pages for the empty_zero_page.
+ */
+#define PGDIR_SHIFT	22
+#define PGDIR_SIZE	(1UL << PGDIR_SHIFT)
+#define PGDIR_MASK	(~(PGDIR_SIZE-1))
+
+/*
+ * Entries per page directory level: we use two-level, so
+ * we don't really have any PMD directory physically.
+ */
+#define PTRS_PER_PTE		1024
+#define PTRS_PER_PTE_SHIFT	10
+#define PTRS_PER_PMD		1
+#define PTRS_PER_PGD		1024
+#define PGD_ORDER		0
+#define PMD_ORDER		0
+#define USER_PTRS_PER_PGD	(TASK_SIZE/PGDIR_SIZE)
+#define FIRST_USER_ADDRESS      XCHAL_SEG_MAPPABLE_VADDR
+#define FIRST_USER_PGD_NR	(FIRST_USER_ADDRESS >> PGDIR_SHIFT)
+
+/* virtual memory area. We keep a distance to other memory regions to be
+ * on the safe side. We also use this area for cache aliasing.
+ */
+
+// FIXME: virtual memory area must be configuration-dependent
+
+#define VMALLOC_START		0xC0000000
+#define VMALLOC_END		0xC7FF0000
+
+/* Xtensa Linux config PTE layout (when present):
+ *	31-12:	PPN
+ *	11-6:	Software
+ *	5-4:	RING
+ *	3-0:	CA
+ *
+ * Similar to the Alpha and MIPS ports, we need to keep track of the ref
+ * and mod bits in software.  We have a software "you can read
+ * from this page" bit, and a hardware one which actually lets the
+ * process read from the page.  On the same token we have a software
+ * writable bit and the real hardware one which actually lets the
+ * process write to the page.
+ *
+ * See further below for PTE layout for swapped-out pages.
+ */
+
+#define _PAGE_VALID		(1<<0)	/* hardware: page is accessible */
+#define _PAGE_WRENABLE		(1<<1)	/* hardware: page is writable */
+
+/* None of these cache modes include MP coherency:  */
+#define _PAGE_NO_CACHE		(0<<2)	/* bypass, non-speculative */
+#if XCHAL_DCACHE_IS_WRITEBACK
+# define _PAGE_WRITEBACK	(1<<2)	/* write back */
+# define _PAGE_WRITETHRU	(2<<2)	/* write through */
+#else
+# define _PAGE_WRITEBACK	(1<<2)	/* assume write through */
+# define _PAGE_WRITETHRU	(1<<2)
+#endif
+#define _PAGE_NOALLOC		(3<<2)	/* don't allocate cache,if not cached */
+#define _CACHE_MASK		(3<<2)
+
+#define _PAGE_USER		(1<<4)	/* user access (ring=1) */
+#define _PAGE_KERNEL		(0<<4)	/* kernel access (ring=0) */
+
+/* Software */
+#define _PAGE_RW		(1<<6)	/* software: page writable */
+#define _PAGE_DIRTY		(1<<7)	/* software: page dirty */
+#define _PAGE_ACCESSED		(1<<8)	/* software: page accessed (read) */
+#define _PAGE_FILE		(1<<9)	/* nonlinear file mapping*/
+
+#define _PAGE_CHG_MASK	(PAGE_MASK | _PAGE_ACCESSED | _CACHE_MASK | _PAGE_DIRTY)
+#define _PAGE_PRESENT	( _PAGE_VALID | _PAGE_WRITEBACK | _PAGE_ACCESSED)
+
+#ifdef CONFIG_MMU
+
+# define PAGE_NONE	__pgprot(_PAGE_PRESENT)
+# define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_RW)
+# define PAGE_COPY	__pgprot(_PAGE_PRESENT | _PAGE_USER)
+# define PAGE_READONLY	__pgprot(_PAGE_PRESENT | _PAGE_USER)
+# define PAGE_KERNEL	__pgprot(_PAGE_PRESENT | _PAGE_KERNEL | _PAGE_WRENABLE)
+# define PAGE_INVALID	__pgprot(_PAGE_USER)
+
+# if (DCACHE_WAY_SIZE > PAGE_SIZE)
+#  define PAGE_DIRECTORY  __pgprot(_PAGE_VALID | _PAGE_ACCESSED | _PAGE_KERNEL)
+# else
+#  define PAGE_DIRECTORY  __pgprot(_PAGE_PRESENT | _PAGE_KERNEL)
+# endif
+
+#else /* no mmu */
+
+# define PAGE_NONE       __pgprot(0)
+# define PAGE_SHARED     __pgprot(0)
+# define PAGE_COPY       __pgprot(0)
+# define PAGE_READONLY   __pgprot(0)
+# define PAGE_KERNEL     __pgprot(0)
+
+#endif
+
+/*
+ * On certain configurations of Xtensa MMUs (eg. the initial Linux config),
+ * the MMU can't do page protection for execute, and considers that the same as
+ * read.  Also, write permissions may imply read permissions.
+ * What follows is the closest we can get by reasonable means..
+ * See linux/mm/mmap.c for protection_map[] array that uses these definitions.
+ */
+#define __P000	PAGE_NONE	/* private --- */
+#define __P001	PAGE_READONLY	/* private --r */
+#define __P010	PAGE_COPY	/* private -w- */
+#define __P011	PAGE_COPY	/* private -wr */
+#define __P100	PAGE_READONLY	/* private x-- */
+#define __P101	PAGE_READONLY	/* private x-r */
+#define __P110	PAGE_COPY	/* private xw- */
+#define __P111	PAGE_COPY	/* private xwr */
+
+#define __S000	PAGE_NONE	/* shared  --- */
+#define __S001	PAGE_READONLY	/* shared  --r */
+#define __S010	PAGE_SHARED	/* shared  -w- */
+#define __S011	PAGE_SHARED	/* shared  -wr */
+#define __S100	PAGE_READONLY	/* shared  x-- */
+#define __S101	PAGE_READONLY	/* shared  x-r */
+#define __S110	PAGE_SHARED	/* shared  xw- */
+#define __S111	PAGE_SHARED	/* shared  xwr */
+
+#ifndef __ASSEMBLY__
+
+#define pte_ERROR(e) \
+	printk("%s:%d: bad pte %08lx.\n", __FILE__, __LINE__, pte_val(e))
+#define pgd_ERROR(e) \
+	printk("%s:%d: bad pgd entry %08lx.\n", __FILE__, __LINE__, pgd_val(e))
+
+extern unsigned long empty_zero_page[1024];
+
+#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
+
+extern pgd_t swapper_pg_dir[PAGE_SIZE/sizeof(pgd_t)];
+
+/*
+ * The pmd contains the kernel virtual address of the pte page.
+ */
+#define pmd_page_kernel(pmd) ((unsigned long)(pmd_val(pmd) & PAGE_MASK))
+#define pmd_page(pmd) virt_to_page(pmd_val(pmd))
+
+/*
+ * The following only work if pte_present() is true.
+ */
+#define pte_none(pte)	 (!(pte_val(pte) ^ _PAGE_USER))
+#define pte_present(pte) (pte_val(pte) & _PAGE_VALID)
+#define pte_clear(mm,addr,ptep)						\
+	do { update_pte(ptep, __pte(_PAGE_USER)); } while(0)
+
+#define pmd_none(pmd)	 (!pmd_val(pmd))
+#define pmd_present(pmd) (pmd_val(pmd) & PAGE_MASK)
+#define pmd_clear(pmdp)	 do { set_pmd(pmdp, __pmd(0)); } while (0)
+#define pmd_bad(pmd)	 (pmd_val(pmd) & ~PAGE_MASK)
+
+/* Note: We use the _PAGE_USER bit to indicate write-protect kernel memory */
+
+static inline int pte_read(pte_t pte)  { return pte_val(pte) & _PAGE_USER; }
+static inline int pte_write(pte_t pte) { return pte_val(pte) & _PAGE_RW; }
+static inline int pte_dirty(pte_t pte) { return pte_val(pte) & _PAGE_DIRTY; }
+static inline int pte_young(pte_t pte) { return pte_val(pte) & _PAGE_ACCESSED; }
+static inline int pte_file(pte_t pte)  { return pte_val(pte) & _PAGE_FILE; }
+static inline pte_t pte_wrprotect(pte_t pte)	{ pte_val(pte) &= ~(_PAGE_RW | _PAGE_WRENABLE); return pte; }
+static inline pte_t pte_rdprotect(pte_t pte)	{ pte_val(pte) &= ~_PAGE_USER; return pte; }
+static inline pte_t pte_mkclean(pte_t pte)	{ pte_val(pte) &= ~_PAGE_DIRTY; return pte; }
+static inline pte_t pte_mkold(pte_t pte)	{ pte_val(pte) &= ~_PAGE_ACCESSED; return pte; }
+static inline pte_t pte_mkread(pte_t pte)	{ pte_val(pte) |= _PAGE_USER; return pte; }
+static inline pte_t pte_mkdirty(pte_t pte)	{ pte_val(pte) |= _PAGE_DIRTY; return pte; }
+static inline pte_t pte_mkyoung(pte_t pte)	{ pte_val(pte) |= _PAGE_ACCESSED; return pte; }
+static inline pte_t pte_mkwrite(pte_t pte)	{ pte_val(pte) |= _PAGE_RW; return pte; }
+
+/*
+ * Conversion functions: convert a page and protection to a page entry,
+ * and a page entry and page directory to the page they refer to.
+ */
+#define pte_pfn(pte)		(pte_val(pte) >> PAGE_SHIFT)
+#define pte_same(a,b)		(pte_val(a) == pte_val(b))
+#define pte_page(x)		pfn_to_page(pte_pfn(x))
+#define pfn_pte(pfn, prot)	__pte(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
+#define mk_pte(page, prot)	pfn_pte(page_to_pfn(page), prot)
+
+extern inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
+{
+	return __pte((pte_val(pte) & _PAGE_CHG_MASK) | pgprot_val(newprot));
+}
+
+/*
+ * Certain architectures need to do special things when pte's
+ * within a page table are directly modified.  Thus, the following
+ * hook is made available.
+ */
+static inline void update_pte(pte_t *ptep, pte_t pteval)
+{
+	*ptep = pteval;
+#if (DCACHE_WAY_SIZE > PAGE_SIZE) && XCHAL_DCACHE_IS_WRITEBACK
+	__asm__ __volatile__ ("memw; dhwb %0, 0; dsync" :: "a" (ptep));
+#endif
+}
+
+extern inline void
+set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pteval)
+{
+	update_pte(ptep, pteval);
+}
+
+
+extern inline void
+set_pmd(pmd_t *pmdp, pmd_t pmdval)
+{
+	*pmdp = pmdval;
+#if (DCACHE_WAY_SIZE > PAGE_SIZE) && XCHAL_DCACHE_IS_WRITEBACK
+	__asm__ __volatile__ ("memw; dhwb %0, 0; dsync" :: "a" (pmdp));
+#endif
+}
+
+
+static inline int
+ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr,
+    			  pte_t *ptep)
+{
+	pte_t pte = *ptep;
+	if (!pte_young(pte))
+		return 0;
+	update_pte(ptep, pte_mkold(pte));
+	return 1;
+}
+
+static inline int
+ptep_test_and_clear_dirty(struct vm_area_struct *vma, unsigned long addr,
+   			  pte_t *ptep)
+{
+	pte_t pte = *ptep;
+	if (!pte_dirty(pte))
+		return 0;
+	update_pte(ptep, pte_mkclean(pte));
+	return 1;
+}
+
+static inline pte_t
+ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+{
+	pte_t pte = *ptep;
+	pte_clear(mm, addr, ptep);
+	return pte;
+}
+
+static inline void
+ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+{
+  	pte_t pte = *ptep;
+  	update_pte(ptep, pte_wrprotect(pte));
+}
+
+/* to find an entry in a kernel page-table-directory */
+#define pgd_offset_k(address)	pgd_offset(&init_mm, address)
+
+/* to find an entry in a page-table-directory */
+#define pgd_offset(mm,address)	((mm)->pgd + pgd_index(address))
+
+#define pgd_index(address)	((address) >> PGDIR_SHIFT)
+
+/* Find an entry in the second-level page table.. */
+#define pmd_offset(dir,address) ((pmd_t*)(dir))
+
+/* Find an entry in the third-level page table.. */
+#define pte_index(address)	(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
+#define pte_offset_kernel(dir,addr) 					\
+	((pte_t*) pmd_page_kernel(*(dir)) + pte_index(addr))
+#define pte_offset_map(dir,addr)	pte_offset_kernel((dir),(addr))
+#define pte_offset_map_nested(dir,addr)	pte_offset_kernel((dir),(addr))
+
+#define pte_unmap(pte)		do { } while (0)
+#define pte_unmap_nested(pte)	do { } while (0)
+
+
+/*
+ * Encode and decode a swap entry.
+ * Each PTE in a process VM's page table is either:
+ *   "present" -- valid and not swapped out, protection bits are meaningful;
+ *   "not present" -- which further subdivides in these two cases:
+ *      "none" -- no mapping at all; identified by pte_none(), set by pte_clear(
+ *      "swapped out" -- the page is swapped out, and the SWP macros below
+ *                      are used to store swap file info in the PTE itself.
+ *
+ * In the Xtensa processor MMU, any PTE entries in user space (or anywhere
+ * in virtual memory that can map differently across address spaces)
+ * must have a correct ring value that represents the RASID field that
+ * is changed when switching address spaces.  Eg. such PTE entries cannot
+ * be set to ring zero, because that can cause a (global) kernel ASID
+ * entry to be created in the TLBs (even with invalid cache attribute),
+ * potentially causing a multihit exception when going back to another
+ * address space that mapped the same virtual address at another ring.
+ *
+ * SO: we avoid using ring bits (_PAGE_RING_MASK) in "not present" PTEs.
+ * We also avoid using the _PAGE_VALID bit which must be zero for non-present
+ * pages.
+ *
+ * We end up with the following available bits:  1..3 and 7..31.
+ * We don't bother with 1..3 for now (we can use them later if needed),
+ * and chose to allocate 6 bits for SWP_TYPE and the remaining 19 bits
+ * for SWP_OFFSET.  At least 5 bits are needed for SWP_TYPE, because it
+ * is currently implemented as an index into swap_info[MAX_SWAPFILES]
+ * and MAX_SWAPFILES is currently defined as 32 in <linux/swap.h>.
+ * However, for some reason all other architectures in the 2.4 kernel
+ * reserve either 6, 7, or 8 bits so I'll not detract from that for now.  :)
+ * SWP_OFFSET is an offset into the swap file in page-size units, so
+ * with 4 kB pages, 19 bits supports a maximum swap file size of 2 GB.
+ *
+ * FIXME:  2 GB isn't very big.  Other bits can be used to allow
+ * larger swap sizes.  In the meantime, it appears relatively easy to get
+ * around the 2 GB limitation by simply using multiple swap files.
+ */
+
+#define __swp_type(entry)	(((entry).val >> 7) & 0x3f)
+#define __swp_offset(entry)	((entry).val >> 13)
+#define __swp_entry(type,offs)	((swp_entry_t) {((type) << 7) | ((offs) << 13)})
+#define __pte_to_swp_entry(pte)	((swp_entry_t) { pte_val(pte) })
+#define __swp_entry_to_pte(x)	((pte_t) { (x).val })
+
+#define PTE_FILE_MAX_BITS	29
+#define pte_to_pgoff(pte)	(pte_val(pte) >> 3)
+#define pgoff_to_pte(off)	((pte_t) { ((off) << 3) | _PAGE_FILE })
+
+
+#endif /*  !defined (__ASSEMBLY__) */
+
+
+#ifdef __ASSEMBLY__
+
+/* Assembly macro _PGD_INDEX is the same as C pgd_index(unsigned long),
+ *                _PGD_OFFSET as C pgd_offset(struct mm_struct*, unsigned long),
+ *                _PMD_OFFSET as C pmd_offset(pgd_t*, unsigned long)
+ *                _PTE_OFFSET as C pte_offset(pmd_t*, unsigned long)
+ *
+ * Note: We require an additional temporary register which can be the same as
+ *       the register that holds the address.
+ *
+ * ((pte_t*) ((unsigned long)(pmd_val(*pmd) & PAGE_MASK)) + pte_index(addr))
+ *
+ */
+#define _PGD_INDEX(rt,rs)	extui	rt, rs, PGDIR_SHIFT, 32-PGDIR_SHIFT
+#define _PTE_INDEX(rt,rs)	extui	rt, rs, PAGE_SHIFT, PTRS_PER_PTE_SHIFT
+
+#define _PGD_OFFSET(mm,adr,tmp)		l32i	mm, mm, MM_PGD;		\
+					_PGD_INDEX(tmp, adr);		\
+					addx4	mm, tmp, mm
+
+#define _PTE_OFFSET(pmd,adr,tmp)	_PTE_INDEX(tmp, adr);		\
+					srli	pmd, pmd, PAGE_SHIFT;	\
+					slli	pmd, pmd, PAGE_SHIFT;	\
+					addx4	pmd, tmp, pmd
+
+#else
+
+extern void paging_init(void);
+
+#define kern_addr_valid(addr)	(1)
+
+extern  void update_mmu_cache(struct vm_area_struct * vma,
+			      unsigned long address, pte_t pte);
+
+/*
+ * remap a physical address `phys' of size `size' with page protection `prot'
+ * into virtual address `from'
+ */
+#define io_remap_page_range(vma,from,phys,size,prot) \
+                remap_pfn_range(vma, from, (phys) >> PAGE_SHIFT, size, prot)
+
+
+/* No page table caches to init */
+
+#define pgtable_cache_init()	do { } while (0)
+
+typedef pte_t *pte_addr_t;
+
+#endif /* !defined (__ASSEMBLY__) */
+
+#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
+#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
+#define __HAVE_ARCH_PTEP_SET_WRPROTECT
+#define __HAVE_ARCH_PTEP_MKDIRTY
+#define __HAVE_ARCH_PTE_SAME
+
+#include <asm-generic/pgtable.h>
+
+#endif /* _XTENSA_PGTABLE_H */
diff --git a/include/asm-xtensa/poll.h b/include/asm-xtensa/poll.h
new file mode 100644
index 00000000000..dffe447534e
--- /dev/null
+++ b/include/asm-xtensa/poll.h
@@ -0,0 +1,37 @@
+/*
+ * include/asm-xtensa/poll.h
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_POLL_H
+#define _XTENSA_POLL_H
+
+
+#define POLLIN		0x0001
+#define POLLPRI		0x0002
+#define POLLOUT		0x0004
+
+#define POLLERR		0x0008
+#define POLLHUP		0x0010
+#define POLLNVAL	0x0020
+
+#define POLLRDNORM	0x0040
+#define POLLRDBAND	0x0080
+#define POLLWRNORM	POLLOUT
+#define POLLWRBAND	0x0100
+
+#define POLLMSG		0x0400
+#define POLLREMOVE	0x0800
+
+struct pollfd {
+	int fd;
+	short events;
+	short revents;
+};
+
+#endif /* _XTENSA_POLL_H */
diff --git a/include/asm-xtensa/posix_types.h b/include/asm-xtensa/posix_types.h
new file mode 100644
index 00000000000..2c816b0e776
--- /dev/null
+++ b/include/asm-xtensa/posix_types.h
@@ -0,0 +1,123 @@
+/*
+ * include/asm-xtensa/posix_types.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Largely copied from include/asm-ppc/posix_types.h
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_POSIX_TYPES_H
+#define _XTENSA_POSIX_TYPES_H
+
+/*
+ * This file is generally used by user-level software, so you need to
+ * be a little careful about namespace pollution etc.  Also, we cannot
+ * assume GCC is being used.
+ */
+
+typedef unsigned long	__kernel_ino_t;
+typedef unsigned int	__kernel_mode_t;
+typedef unsigned short	__kernel_nlink_t;
+typedef long		__kernel_off_t;
+typedef int		__kernel_pid_t;
+typedef unsigned short	__kernel_ipc_pid_t;
+typedef unsigned int	__kernel_uid_t;
+typedef unsigned int	__kernel_gid_t;
+typedef unsigned int	__kernel_size_t;
+typedef int		__kernel_ssize_t;
+typedef long		__kernel_ptrdiff_t;
+typedef long		__kernel_time_t;
+typedef long		__kernel_suseconds_t;
+typedef long		__kernel_clock_t;
+typedef int		__kernel_timer_t;
+typedef int		__kernel_clockid_t;
+typedef int		__kernel_daddr_t;
+typedef char *		__kernel_caddr_t;
+typedef unsigned short	__kernel_uid16_t;
+typedef unsigned short	__kernel_gid16_t;
+typedef unsigned int	__kernel_uid32_t;
+typedef unsigned int	__kernel_gid32_t;
+
+typedef unsigned short	__kernel_old_uid_t;
+typedef unsigned short	__kernel_old_gid_t;
+typedef unsigned short	__kernel_old_dev_t;
+
+#ifdef __GNUC__
+typedef long long	__kernel_loff_t;
+#endif
+
+typedef struct {
+	int	val[2];
+} __kernel_fsid_t;
+
+#ifndef __GNUC__
+
+#define	__FD_SET(d, set)	((set)->fds_bits[__FDELT(d)] |= __FDMASK(d))
+#define	__FD_CLR(d, set)	((set)->fds_bits[__FDELT(d)] &= ~__FDMASK(d))
+#define	__FD_ISSET(d, set)	((set)->fds_bits[__FDELT(d)] & __FDMASK(d))
+#define	__FD_ZERO(set)	\
+  ((void) memset ((__ptr_t) (set), 0, sizeof (__kernel_fd_set)))
+
+#else /* __GNUC__ */
+
+#if defined(__KERNEL__) || !defined(__GLIBC__) || (__GLIBC__ < 2) \
+    || (__GLIBC__ == 2 && __GLIBC_MINOR__ == 0)
+/* With GNU C, use inline functions instead so args are evaluated only once: */
+
+#undef __FD_SET
+static __inline__ void __FD_SET(unsigned long fd, __kernel_fd_set *fdsetp)
+{
+	unsigned long _tmp = fd / __NFDBITS;
+	unsigned long _rem = fd % __NFDBITS;
+	fdsetp->fds_bits[_tmp] |= (1UL<<_rem);
+}
+
+#undef __FD_CLR
+static __inline__ void __FD_CLR(unsigned long fd, __kernel_fd_set *fdsetp)
+{
+	unsigned long _tmp = fd / __NFDBITS;
+	unsigned long _rem = fd % __NFDBITS;
+	fdsetp->fds_bits[_tmp] &= ~(1UL<<_rem);
+}
+
+#undef __FD_ISSET
+static __inline__ int __FD_ISSET(unsigned long fd, __kernel_fd_set *p)
+{
+	unsigned long _tmp = fd / __NFDBITS;
+	unsigned long _rem = fd % __NFDBITS;
+	return (p->fds_bits[_tmp] & (1UL<<_rem)) != 0;
+}
+
+/*
+ * This will unroll the loop for the normal constant case (8 ints,
+ * for a 256-bit fd_set)
+ */
+#undef __FD_ZERO
+static __inline__ void __FD_ZERO(__kernel_fd_set *p)
+{
+	unsigned int *tmp = (unsigned int *)p->fds_bits;
+	int i;
+
+	if (__builtin_constant_p(__FDSET_LONGS)) {
+		switch (__FDSET_LONGS) {
+			case 8:
+				tmp[0] = 0; tmp[1] = 0; tmp[2] = 0; tmp[3] = 0;
+				tmp[4] = 0; tmp[5] = 0; tmp[6] = 0; tmp[7] = 0;
+				return;
+		}
+	}
+	i = __FDSET_LONGS;
+	while (i) {
+		i--;
+		*tmp = 0;
+		tmp++;
+	}
+}
+
+#endif /* defined(__KERNEL__) || !defined(__GLIBC__) || (__GLIBC__ < 2) */
+#endif /* __GNUC__ */
+#endif /* _XTENSA_POSIX_TYPES_H */
diff --git a/include/asm-xtensa/processor.h b/include/asm-xtensa/processor.h
new file mode 100644
index 00000000000..9cab5e4298b
--- /dev/null
+++ b/include/asm-xtensa/processor.h
@@ -0,0 +1,205 @@
+/*
+ * include/asm-xtensa/processor.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_PROCESSOR_H
+#define _XTENSA_PROCESSOR_H
+
+#ifdef __ASSEMBLY__
+#define _ASMLANGUAGE
+#endif
+
+#include <xtensa/config/core.h>
+#include <xtensa/config/specreg.h>
+#include <xtensa/config/tie.h>
+#include <xtensa/config/system.h>
+
+#include <asm/ptrace.h>
+#include <asm/types.h>
+#include <asm/coprocessor.h>
+
+/* Assertions. */
+
+#if (XCHAL_HAVE_WINDOWED != 1)
+#error Linux requires the Xtensa Windowed Registers Option.
+#endif
+
+/*
+ * User space process size: 1 GB.
+ * Windowed call ABI requires caller and callee to be located within the same
+ * 1 GB region. The C compiler places trampoline code on the stack for sources
+ * that take the address of a nested C function (a feature used by glibc), so
+ * the 1 GB requirement applies to the stack as well.
+ */
+
+#define TASK_SIZE	0x40000000
+
+/*
+ * General exception cause assigned to debug exceptions. Debug exceptions go
+ * to their own vector, rather than the general exception vectors (user,
+ * kernel, double); and their specific causes are reported via DEBUGCAUSE
+ * rather than EXCCAUSE.  However it is sometimes convenient to redirect debug
+ * exceptions to the general exception mechanism.  To do this, an otherwise
+ * unused EXCCAUSE value was assigned to debug exceptions for this purpose.
+ */
+
+#define EXCCAUSE_MAPPED_DEBUG	63
+
+/*
+ * We use DEPC also as a flag to distinguish between double and regular
+ * exceptions. For performance reasons, DEPC might contain the value of
+ * EXCCAUSE for regular exceptions, so we use this definition to mark a
+ * valid double exception address.
+ * (Note: We use it in bgeui, so it should be 64, 128, or 256)
+ */
+
+#define VALID_DOUBLE_EXCEPTION_ADDRESS	64
+
+/* LOCKLEVEL defines the interrupt level that masks all
+ * general-purpose interrupts.
+ */
+#define LOCKLEVEL 1
+
+/* WSBITS and WBBITS are the width of the WINDOWSTART and WINDOWBASE
+ * registers
+ */
+#define WSBITS  (XCHAL_NUM_AREGS / 4)      /* width of WINDOWSTART in bits */
+#define WBBITS  (XCHAL_NUM_AREGS_LOG2 - 2) /* width of WINDOWBASE in bits */
+
+#ifndef __ASSEMBLY__
+
+/* Build a valid return address for the specified call winsize.
+ * winsize must be 1 (call4), 2 (call8), or 3 (call12)
+ */
+#define MAKE_RA_FOR_CALL(ra,ws)   (((ra) & 0x3fffffff) | (ws) << 30)
+
+/* Convert return address to a valid pc
+ * Note: We assume that the stack pointer is in the same 1GB ranges as the ra
+ */
+#define MAKE_PC_FROM_RA(ra,sp)    (((ra) & 0x3fffffff) | ((sp) & 0xc0000000))
+
+typedef struct {
+    unsigned long seg;
+} mm_segment_t;
+
+struct thread_struct {
+
+	/* kernel's return address and stack pointer for context switching */
+	unsigned long ra; /* kernel's a0: return address and window call size */
+	unsigned long sp; /* kernel's a1: stack pointer */
+
+	mm_segment_t current_ds;    /* see uaccess.h for example uses */
+
+	/* struct xtensa_cpuinfo info; */
+
+	unsigned long bad_vaddr; /* last user fault */
+	unsigned long bad_uaddr; /* last kernel fault accessing user space */
+	unsigned long error_code;
+
+	unsigned long ibreak[XCHAL_NUM_IBREAK];
+	unsigned long dbreaka[XCHAL_NUM_DBREAK];
+	unsigned long dbreakc[XCHAL_NUM_DBREAK];
+
+	/* Allocate storage for extra state and coprocessor state. */
+	unsigned char cp_save[XTENSA_CP_EXTRA_SIZE]
+		__attribute__ ((aligned(XTENSA_CP_EXTRA_ALIGN)));
+
+	/* Make structure 16 bytes aligned. */
+	int align[0] __attribute__ ((aligned(16)));
+};
+
+
+/*
+ * Default implementation of macro that returns current
+ * instruction pointer ("program counter").
+ */
+#define current_text_addr()  ({ __label__ _l; _l: &&_l;})
+
+
+/* This decides where the kernel will search for a free chunk of vm
+ * space during mmap's.
+ */
+#define TASK_UNMAPPED_BASE	(TASK_SIZE / 2)
+
+#define INIT_THREAD  \
+{									\
+	ra:		0, 						\
+	sp:		sizeof(init_stack) + (long) &init_stack,	\
+	current_ds:	{0},						\
+	/*info:		{0}, */						\
+	bad_vaddr:	0,						\
+	bad_uaddr:	0,						\
+	error_code:	0,						\
+}
+
+
+/*
+ * Do necessary setup to start up a newly executed thread.
+ * Note: We set-up ps as if we did a call4 to the new pc.
+ *       set_thread_state in signal.c depends on it.
+ */
+#define USER_PS_VALUE ( (1 << XCHAL_PS_WOE_SHIFT) + \
+                        (1 << XCHAL_PS_CALLINC_SHIFT) + \
+                        (USER_RING << XCHAL_PS_RING_SHIFT) + \
+                        (1 << XCHAL_PS_PROGSTACK_SHIFT) + \
+                        (1 << XCHAL_PS_EXCM_SHIFT) )
+
+/* Clearing a0 terminates the backtrace. */
+#define start_thread(regs, new_pc, new_sp) \
+	regs->pc = new_pc; \
+	regs->ps = USER_PS_VALUE; \
+	regs->areg[1] = new_sp; \
+	regs->areg[0] = 0; \
+	regs->wmask = 1; \
+	regs->depc = 0; \
+	regs->windowbase = 0; \
+	regs->windowstart = 1;
+
+/* Forward declaration */
+struct task_struct;
+struct mm_struct;
+
+// FIXME: do we need release_thread for CP??
+/* Free all resources held by a thread. */
+#define release_thread(thread) do { } while(0)
+
+// FIXME: do we need prepare_to_copy (lazy status) for CP??
+/* Prepare to copy thread state - unlazy all lazy status */
+#define prepare_to_copy(tsk)	do { } while (0)
+
+/*
+ * create a kernel thread without removing it from tasklists
+ */
+extern int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
+
+/* Copy and release all segment info associated with a VM */
+
+#define copy_segments(p, mm)	do { } while(0)
+#define release_segments(mm)	do { } while(0)
+#define forget_segments()	do { } while (0)
+
+#define thread_saved_pc(tsk)	(xtensa_pt_regs(tsk)->pc)
+
+extern unsigned long get_wchan(struct task_struct *p);
+
+#define KSTK_EIP(tsk)		(xtensa_pt_regs(tsk)->pc)
+#define KSTK_ESP(tsk)		(xtensa_pt_regs(tsk)->areg[1])
+
+#define cpu_relax()  do { } while (0)
+
+/* Special register access. */
+
+#define WSR(v,sr) __asm__ __volatile__ ("wsr %0,"__stringify(sr) :: "a"(v));
+#define RSR(v,sr) __asm__ __volatile__ ("rsr %0,"__stringify(sr) : "=a"(v));
+
+#define set_sr(x,sr) ({unsigned int v=(unsigned int)x; WSR(v,sr);})
+#define get_sr(sr) ({unsigned int v; RSR(v,sr); v; })
+
+#endif	/* __ASSEMBLY__ */
+#endif	/* _XTENSA_PROCESSOR_H */
diff --git a/include/asm-xtensa/ptrace.h b/include/asm-xtensa/ptrace.h
new file mode 100644
index 00000000000..2848a5ff834
--- /dev/null
+++ b/include/asm-xtensa/ptrace.h
@@ -0,0 +1,135 @@
+/*
+ * include/asm-xtensa/ptrace.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_PTRACE_H
+#define _XTENSA_PTRACE_H
+
+#include <xtensa/config/core.h>
+
+/*
+ * Kernel stack
+ *
+ * 		+-----------------------+  -------- STACK_SIZE
+ * 		|     register file     |  |
+ * 		+-----------------------+  |
+ * 		|    struct pt_regs     |  |
+ * 		+-----------------------+  | ------ PT_REGS_OFFSET
+ * double 	:  16 bytes spill area  :  |  ^
+ * excetion 	:- - - - - - - - - - - -:  |  |
+ * frame	:    struct pt_regs     :  |  |
+ * 		:- - - - - - - - - - - -:  |  |
+ * 		|                       |  |  |
+ * 		|     memory stack      |  |  |
+ * 		|                       |  |  |
+ * 		~                       ~  ~  ~
+ * 		~                       ~  ~  ~
+ * 		|                       |  |  |
+ * 		|                       |  |  |
+ * 		+-----------------------+  |  | --- STACK_BIAS
+ * 		|  struct task_struct   |  |  |  ^
+ *  current --> +-----------------------+  |  |  |
+ * 		|  struct thread_info   |  |  |  |
+ *		+-----------------------+ --------
+ */
+
+#define KERNEL_STACK_SIZE (2 * PAGE_SIZE)
+
+/*  Offsets for exception_handlers[] (3 x 64-entries x 4-byte tables). */
+
+#define EXC_TABLE_KSTK		0x004	/* Kernel Stack */
+#define EXC_TABLE_DOUBLE_SAVE	0x008	/* Double exception save area for a0 */
+#define EXC_TABLE_FIXUP		0x00c	/* Fixup handler */
+#define EXC_TABLE_PARAM		0x010	/* For passing a parameter to fixup */
+#define EXC_TABLE_SYSCALL_SAVE	0x014	/* For fast syscall handler */
+#define EXC_TABLE_FAST_USER	0x100	/* Fast user exception handler */
+#define EXC_TABLE_FAST_KERNEL	0x200	/* Fast kernel exception handler */
+#define EXC_TABLE_DEFAULT	0x300	/* Default C-Handler */
+#define EXC_TABLE_SIZE		0x400
+
+/* Registers used by strace */
+
+#define REG_A_BASE	0xfc000000
+#define REG_AR_BASE	0x04000000
+#define REG_PC		0x14000000
+#define REG_PS		0x080000e6
+#define REG_WB		0x08000048
+#define REG_WS		0x08000049
+#define REG_LBEG	0x08000000
+#define REG_LEND	0x08000001
+#define REG_LCOUNT	0x08000002
+#define REG_SAR		0x08000003
+#define REG_DEPC	0x080000c0
+#define	REG_EXCCAUSE	0x080000e8
+#define REG_EXCVADDR	0x080000ee
+#define SYSCALL_NR	0x1
+
+#define AR_REGNO_TO_A_REGNO(ar, wb) (ar - wb*4) & ~(XCHAL_NUM_AREGS - 1)
+
+/* Other PTRACE_ values defined in <linux/ptrace.h> using values 0-9,16,17,24 */
+
+#define PTRACE_GETREGS            12
+#define PTRACE_SETREGS            13
+#define PTRACE_GETFPREGS          14
+#define PTRACE_SETFPREGS          15
+#define PTRACE_GETFPREGSIZE       18
+
+#ifndef __ASSEMBLY__
+
+/*
+ * This struct defines the way the registers are stored on the
+ * kernel stack during a system call or other kernel entry.
+ */
+struct pt_regs {
+	unsigned long pc;		/*   4 */
+	unsigned long ps;		/*   8 */
+	unsigned long depc;		/*  12 */
+	unsigned long exccause;		/*  16 */
+	unsigned long excvaddr;		/*  20 */
+	unsigned long debugcause;	/*  24 */
+	unsigned long wmask;		/*  28 */
+	unsigned long lbeg;		/*  32 */
+	unsigned long lend;		/*  36 */
+	unsigned long lcount;		/*  40 */
+	unsigned long sar;		/*  44 */
+	unsigned long windowbase;	/*  48 */
+	unsigned long windowstart;	/*  52 */
+	unsigned long syscall;		/*  56 */
+	int reserved[2];		/*  64 */
+
+	/* Make sure the areg field is 16 bytes aligned. */
+	int align[0] __attribute__ ((aligned(16)));
+
+	/* current register frame.
+	 * Note: The ESF for kernel exceptions ends after 16 registers!
+	 */
+	unsigned long areg[16];		/* 128 (64) */
+};
+
+#ifdef __KERNEL__
+# define xtensa_pt_regs(tsk) ((struct pt_regs*) \
+  (((long)(tsk)->thread_info + KERNEL_STACK_SIZE - (XCHAL_NUM_AREGS-16)*4)) - 1)
+# define user_mode(regs) (((regs)->ps & 0x00000020)!=0)
+# define instruction_pointer(regs) ((regs)->pc)
+extern void show_regs(struct pt_regs *);
+
+# ifndef CONFIG_SMP
+#  define profile_pc(regs) instruction_pointer(regs)
+# endif
+#endif /* __KERNEL__ */
+
+#else	/* __ASSEMBLY__ */
+
+#ifdef __KERNEL__
+# include <asm/offsets.h>
+#define PT_REGS_OFFSET	  (KERNEL_STACK_SIZE - PT_USER_SIZE)
+#endif
+
+#endif	/* !__ASSEMBLY__ */
+#endif	/* _XTENSA_PTRACE_H */
diff --git a/include/asm-xtensa/resource.h b/include/asm-xtensa/resource.h
new file mode 100644
index 00000000000..17b5ab31177
--- /dev/null
+++ b/include/asm-xtensa/resource.h
@@ -0,0 +1,16 @@
+/*
+ * include/asm-xtensa/resource.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_RESOURCE_H
+#define _XTENSA_RESOURCE_H
+
+#include <asm-generic/resource.h>
+
+#endif	/* _XTENSA_RESOURCE_H */
diff --git a/include/asm-xtensa/rmap.h b/include/asm-xtensa/rmap.h
new file mode 100644
index 00000000000..649588b7e9a
--- /dev/null
+++ b/include/asm-xtensa/rmap.h
@@ -0,0 +1,16 @@
+/*
+ * include/asm-xtensa/rmap.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_RMAP_H
+#define _XTENSA_RMAP_H
+
+#include <asm-generic/rmap.h>
+
+#endif
diff --git a/include/asm-xtensa/rwsem.h b/include/asm-xtensa/rwsem.h
new file mode 100644
index 00000000000..3c02b0e033f
--- /dev/null
+++ b/include/asm-xtensa/rwsem.h
@@ -0,0 +1,175 @@
+/*
+ * include/asm-xtensa/rwsem.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Largely copied from include/asm-ppc/rwsem.h
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_RWSEM_H
+#define _XTENSA_RWSEM_H
+
+#include <linux/list.h>
+#include <linux/spinlock.h>
+#include <asm/atomic.h>
+#include <asm/system.h>
+
+/*
+ * the semaphore definition
+ */
+struct rw_semaphore {
+	signed long		count;
+#define RWSEM_UNLOCKED_VALUE		0x00000000
+#define RWSEM_ACTIVE_BIAS		0x00000001
+#define RWSEM_ACTIVE_MASK		0x0000ffff
+#define RWSEM_WAITING_BIAS		(-0x00010000)
+#define RWSEM_ACTIVE_READ_BIAS		RWSEM_ACTIVE_BIAS
+#define RWSEM_ACTIVE_WRITE_BIAS		(RWSEM_WAITING_BIAS + RWSEM_ACTIVE_BIAS)
+	spinlock_t		wait_lock;
+	struct list_head	wait_list;
+#if RWSEM_DEBUG
+	int			debug;
+#endif
+};
+
+/*
+ * initialisation
+ */
+#if RWSEM_DEBUG
+#define __RWSEM_DEBUG_INIT      , 0
+#else
+#define __RWSEM_DEBUG_INIT	/* */
+#endif
+
+#define __RWSEM_INITIALIZER(name) \
+	{ RWSEM_UNLOCKED_VALUE, SPIN_LOCK_UNLOCKED, \
+	  LIST_HEAD_INIT((name).wait_list) \
+	  __RWSEM_DEBUG_INIT }
+
+#define DECLARE_RWSEM(name)		\
+	struct rw_semaphore name = __RWSEM_INITIALIZER(name)
+
+extern struct rw_semaphore *rwsem_down_read_failed(struct rw_semaphore *sem);
+extern struct rw_semaphore *rwsem_down_write_failed(struct rw_semaphore *sem);
+extern struct rw_semaphore *rwsem_wake(struct rw_semaphore *sem);
+extern struct rw_semaphore *rwsem_downgrade_wake(struct rw_semaphore *sem);
+
+static inline void init_rwsem(struct rw_semaphore *sem)
+{
+	sem->count = RWSEM_UNLOCKED_VALUE;
+	spin_lock_init(&sem->wait_lock);
+	INIT_LIST_HEAD(&sem->wait_list);
+#if RWSEM_DEBUG
+	sem->debug = 0;
+#endif
+}
+
+/*
+ * lock for reading
+ */
+static inline void __down_read(struct rw_semaphore *sem)
+{
+	if (atomic_add_return(1,(atomic_t *)(&sem->count)) > 0)
+		smp_wmb();
+	else
+		rwsem_down_read_failed(sem);
+}
+
+static inline int __down_read_trylock(struct rw_semaphore *sem)
+{
+	int tmp;
+
+	while ((tmp = sem->count) >= 0) {
+		if (tmp == cmpxchg(&sem->count, tmp,
+				   tmp + RWSEM_ACTIVE_READ_BIAS)) {
+			smp_wmb();
+			return 1;
+		}
+	}
+	return 0;
+}
+
+/*
+ * lock for writing
+ */
+static inline void __down_write(struct rw_semaphore *sem)
+{
+	int tmp;
+
+	tmp = atomic_add_return(RWSEM_ACTIVE_WRITE_BIAS,
+				(atomic_t *)(&sem->count));
+	if (tmp == RWSEM_ACTIVE_WRITE_BIAS)
+		smp_wmb();
+	else
+		rwsem_down_write_failed(sem);
+}
+
+static inline int __down_write_trylock(struct rw_semaphore *sem)
+{
+	int tmp;
+
+	tmp = cmpxchg(&sem->count, RWSEM_UNLOCKED_VALUE,
+		      RWSEM_ACTIVE_WRITE_BIAS);
+	smp_wmb();
+	return tmp == RWSEM_UNLOCKED_VALUE;
+}
+
+/*
+ * unlock after reading
+ */
+static inline void __up_read(struct rw_semaphore *sem)
+{
+	int tmp;
+
+	smp_wmb();
+	tmp = atomic_sub_return(1,(atomic_t *)(&sem->count));
+	if (tmp < -1 && (tmp & RWSEM_ACTIVE_MASK) == 0)
+		rwsem_wake(sem);
+}
+
+/*
+ * unlock after writing
+ */
+static inline void __up_write(struct rw_semaphore *sem)
+{
+	smp_wmb();
+	if (atomic_sub_return(RWSEM_ACTIVE_WRITE_BIAS,
+			      (atomic_t *)(&sem->count)) < 0)
+		rwsem_wake(sem);
+}
+
+/*
+ * implement atomic add functionality
+ */
+static inline void rwsem_atomic_add(int delta, struct rw_semaphore *sem)
+{
+	atomic_add(delta, (atomic_t *)(&sem->count));
+}
+
+/*
+ * downgrade write lock to read lock
+ */
+static inline void __downgrade_write(struct rw_semaphore *sem)
+{
+	int tmp;
+
+	smp_wmb();
+	tmp = atomic_add_return(-RWSEM_WAITING_BIAS, (atomic_t *)(&sem->count));
+	if (tmp < 0)
+		rwsem_downgrade_wake(sem);
+}
+
+/*
+ * implement exchange and add functionality
+ */
+static inline int rwsem_atomic_update(int delta, struct rw_semaphore *sem)
+{
+	smp_mb();
+	return atomic_add_return(delta, (atomic_t *)(&sem->count));
+}
+
+#endif	/* _XTENSA_RWSEM_XADD_H */
diff --git a/include/asm-xtensa/scatterlist.h b/include/asm-xtensa/scatterlist.h
new file mode 100644
index 00000000000..38a2b9acd65
--- /dev/null
+++ b/include/asm-xtensa/scatterlist.h
@@ -0,0 +1,34 @@
+/*
+ * include/asm-xtensa/scatterlist.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_SCATTERLIST_H
+#define _XTENSA_SCATTERLIST_H
+
+struct scatterlist {
+	struct page 	*page;
+	unsigned int	offset;
+	dma_addr_t	dma_address;
+	unsigned int	length;
+};
+
+/*
+ * These macros should be used after a pci_map_sg call has been done
+ * to get bus addresses of each of the SG entries and their lengths.
+ * You should only work with the number of sg entries pci_map_sg
+ * returns, or alternatively stop on the first sg_dma_len(sg) which
+ * is 0.
+ */
+#define sg_dma_address(sg)      ((sg)->dma_address)
+#define sg_dma_len(sg)          ((sg)->length)
+
+
+#define ISA_DMA_THRESHOLD (~0UL)
+
+#endif	/* _XTENSA_SCATTERLIST_H */
diff --git a/include/asm-xtensa/sections.h b/include/asm-xtensa/sections.h
new file mode 100644
index 00000000000..40b5191b55a
--- /dev/null
+++ b/include/asm-xtensa/sections.h
@@ -0,0 +1,16 @@
+/*
+ * include/asm-xtensa/sections.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_SECTIONS_H
+#define _XTENSA_SECTIONS_H
+
+#include <asm-generic/sections.h>
+
+#endif	/* _XTENSA_SECTIONS_H */
diff --git a/include/asm-xtensa/segment.h b/include/asm-xtensa/segment.h
new file mode 100644
index 00000000000..a2eb547a1a7
--- /dev/null
+++ b/include/asm-xtensa/segment.h
@@ -0,0 +1,16 @@
+/*
+ * include/asm-xtensa/segment.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_SEGMENT_H
+#define _XTENSA_SEGMENT_H
+
+#include <asm/uaccess.h>
+
+#endif	/* _XTENSA_SEGEMENT_H */
diff --git a/include/asm-xtensa/semaphore.h b/include/asm-xtensa/semaphore.h
new file mode 100644
index 00000000000..c8a7574a9a5
--- /dev/null
+++ b/include/asm-xtensa/semaphore.h
@@ -0,0 +1,129 @@
+/*
+ * linux/include/asm-xtensa/semaphore.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_SEMAPHORE_H
+#define _XTENSA_SEMAPHORE_H
+
+#include <asm/atomic.h>
+#include <asm/system.h>
+#include <linux/wait.h>
+#include <linux/rwsem.h>
+
+struct semaphore {
+	atomic_t count;
+	int sleepers;
+	wait_queue_head_t wait;
+#if WAITQUEUE_DEBUG
+	long __magic;
+#endif
+};
+
+#if WAITQUEUE_DEBUG
+# define __SEM_DEBUG_INIT(name) \
+		, (int)&(name).__magic
+#else
+# define __SEM_DEBUG_INIT(name)
+#endif
+
+#define __SEMAPHORE_INITIALIZER(name,count)			\
+	{ ATOMIC_INIT(count), 					\
+	  0,							\
+	  __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)		\
+	__SEM_DEBUG_INIT(name) }
+
+#define __MUTEX_INITIALIZER(name) \
+	__SEMAPHORE_INITIALIZER(name, 1)
+
+#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
+	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
+
+#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
+#define DECLARE_MUTEX_LOCKED(name) __DECLARE_SEMAPHORE_GENERIC(name,0)
+
+extern inline void sema_init (struct semaphore *sem, int val)
+{
+/*
+ *	*sem = (struct semaphore)__SEMAPHORE_INITIALIZER((*sem),val);
+ *
+ * i'd rather use the more flexible initialization above, but sadly
+ * GCC 2.7.2.3 emits a bogus warning. EGCS doesnt. Oh well.
+ */
+	atomic_set(&sem->count, val);
+	init_waitqueue_head(&sem->wait);
+#if WAITQUEUE_DEBUG
+	sem->__magic = (int)&sem->__magic;
+#endif
+}
+
+static inline void init_MUTEX (struct semaphore *sem)
+{
+	sema_init(sem, 1);
+}
+
+static inline void init_MUTEX_LOCKED (struct semaphore *sem)
+{
+	sema_init(sem, 0);
+}
+
+asmlinkage void __down(struct semaphore * sem);
+asmlinkage int  __down_interruptible(struct semaphore * sem);
+asmlinkage int  __down_trylock(struct semaphore * sem);
+asmlinkage void __up(struct semaphore * sem);
+
+extern spinlock_t semaphore_wake_lock;
+
+extern __inline__ void down(struct semaphore * sem)
+{
+#if WAITQUEUE_DEBUG
+	CHECK_MAGIC(sem->__magic);
+#endif
+
+	if (atomic_sub_return(1, &sem->count) < 0)
+		__down(sem);
+}
+
+extern __inline__ int down_interruptible(struct semaphore * sem)
+{
+	int ret = 0;
+#if WAITQUEUE_DEBUG
+	CHECK_MAGIC(sem->__magic);
+#endif
+
+	if (atomic_sub_return(1, &sem->count) < 0)
+		ret = __down_interruptible(sem);
+	return ret;
+}
+
+extern __inline__ int down_trylock(struct semaphore * sem)
+{
+	int ret = 0;
+#if WAITQUEUE_DEBUG
+	CHECK_MAGIC(sem->__magic);
+#endif
+
+	if (atomic_sub_return(1, &sem->count) < 0)
+		ret = __down_trylock(sem);
+	return ret;
+}
+
+/*
+ * Note! This is subtle. We jump to wake people up only if
+ * the semaphore was negative (== somebody was waiting on it).
+ */
+extern __inline__ void up(struct semaphore * sem)
+{
+#if WAITQUEUE_DEBUG
+	CHECK_MAGIC(sem->__magic);
+#endif
+	if (atomic_add_return(1, &sem->count) <= 0)
+		__up(sem);
+}
+
+#endif /* _XTENSA_SEMAPHORE_H */
diff --git a/include/asm-xtensa/sembuf.h b/include/asm-xtensa/sembuf.h
new file mode 100644
index 00000000000..2d26c47666f
--- /dev/null
+++ b/include/asm-xtensa/sembuf.h
@@ -0,0 +1,44 @@
+/*
+ * include/asm-xtensa/sembuf.h
+ *
+ * The semid64_ds structure for Xtensa architecture.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ *
+ * Note extra padding because this structure is passed back and forth
+ * between kernel and user space.
+ *
+ * Pad space is left for:
+ * - 64-bit time_t to solve y2038 problem
+ * - 2 miscellaneous 32-bit values
+ *
+ */
+
+#ifndef _XTENSA_SEMBUF_H
+#define _XTENSA_SEMBUF_H
+
+#include <asm/byteorder.h>
+
+struct semid64_ds {
+	struct ipc64_perm sem_perm;		/* permissions .. see ipc.h */
+#if XCHAL_HAVE_LE
+	__kernel_time_t	sem_otime;		/* last semop time */
+	unsigned long	__unused1;
+	__kernel_time_t	sem_ctime;		/* last change time */
+	unsigned long	__unused2;
+#else
+	unsigned long	__unused1;
+	__kernel_time_t	sem_otime;		/* last semop time */
+	unsigned long	__unused2;
+	__kernel_time_t	sem_ctime;		/* last change time */
+#endif
+	unsigned long	sem_nsems;		/* no. of semaphores in array */
+	unsigned long	__unused3;
+	unsigned long	__unused4;
+};
+
+#endif /* __ASM_XTENSA_SEMBUF_H */
diff --git a/include/asm-xtensa/serial.h b/include/asm-xtensa/serial.h
new file mode 100644
index 00000000000..ec04114fcf0
--- /dev/null
+++ b/include/asm-xtensa/serial.h
@@ -0,0 +1,18 @@
+/*
+ * include/asm-xtensa/serial.h
+ *
+ * Configuration details for 8250, 16450, 16550, etc. serial ports
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_SERIAL_H
+#define _XTENSA_SERIAL_H
+
+#include <asm/platform/serial.h>
+
+#endif	/* _XTENSA_SERIAL_H */
diff --git a/include/asm-xtensa/setup.h b/include/asm-xtensa/setup.h
new file mode 100644
index 00000000000..e3636520d8c
--- /dev/null
+++ b/include/asm-xtensa/setup.h
@@ -0,0 +1,16 @@
+/*
+ * include/asm-xtensa/setup.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_SETUP_H
+#define _XTENSA_SETUP_H
+
+#define COMMAND_LINE_SIZE	256
+
+#endif
diff --git a/include/asm-xtensa/shmbuf.h b/include/asm-xtensa/shmbuf.h
new file mode 100644
index 00000000000..a30b81a4b93
--- /dev/null
+++ b/include/asm-xtensa/shmbuf.h
@@ -0,0 +1,50 @@
+/*
+ * include/asm-xtensa/shmbuf.h
+ *
+ * The shmid64_ds structure for Xtensa architecture.
+ * Note extra padding because this structure is passed back and forth
+ * between kernel and user space.
+ *
+ * Pad space is left for:
+ * - 64-bit time_t to solve y2038 problem
+ * - 2 miscellaneous 32-bit values
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_SHMBUF_H
+#define _XTENSA_SHMBUF_H
+
+struct shmid64_ds {
+	struct ipc64_perm	shm_perm;	/* operation perms */
+	size_t			shm_segsz;	/* size of segment (bytes) */
+	__kernel_time_t		shm_atime;	/* last attach time */
+	unsigned long		__unused1;
+	__kernel_time_t		shm_dtime;	/* last detach time */
+	unsigned long		__unused2;
+	__kernel_time_t		shm_ctime;	/* last change time */
+	unsigned long		__unused3;
+	__kernel_pid_t		shm_cpid;	/* pid of creator */
+	__kernel_pid_t		shm_lpid;	/* pid of last operator */
+	unsigned long		shm_nattch;	/* no. of current attaches */
+	unsigned long		__unused4;
+	unsigned long		__unused5;
+};
+
+struct shminfo64 {
+	unsigned long	shmmax;
+	unsigned long	shmmin;
+	unsigned long	shmmni;
+	unsigned long	shmseg;
+	unsigned long	shmall;
+	unsigned long	__unused1;
+	unsigned long	__unused2;
+	unsigned long	__unused3;
+	unsigned long	__unused4;
+};
+
+#endif	/* _XTENSA_SHMBUF_H */
diff --git a/include/asm-xtensa/shmparam.h b/include/asm-xtensa/shmparam.h
new file mode 100644
index 00000000000..d3b65bfa71c
--- /dev/null
+++ b/include/asm-xtensa/shmparam.h
@@ -0,0 +1,23 @@
+/*
+ * include/asm-xtensa/shmparam.h
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ */
+
+#ifndef _XTENSA_SHMPARAM_H
+#define _XTENSA_SHMPARAM_H
+
+#include <asm/processor.h>
+
+/*
+ * Xtensa can have variable size caches, and if
+ * the size of single way is larger than the page size,
+ * then we have to start worrying about cache aliasing
+ * problems.
+ */
+
+#define SHMLBA	((PAGE_SIZE > DCACHE_WAY_SIZE)? PAGE_SIZE : DCACHE_WAY_SIZE)
+
+#endif /* _XTENSA_SHMPARAM_H */
diff --git a/include/asm-xtensa/sigcontext.h b/include/asm-xtensa/sigcontext.h
new file mode 100644
index 00000000000..a7517729141
--- /dev/null
+++ b/include/asm-xtensa/sigcontext.h
@@ -0,0 +1,44 @@
+/*
+ * include/asm-xtensa/sigcontext.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2003 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_SIGCONTEXT_H
+#define _XTENSA_SIGCONTEXT_H
+
+#define _ASMLANGUAGE
+#include <asm/processor.h>
+#include <asm/coprocessor.h>
+
+
+struct _cpstate {
+	unsigned char _cpstate[XTENSA_CP_EXTRA_SIZE];
+} __attribute__ ((aligned (XTENSA_CP_EXTRA_ALIGN)));
+
+
+struct sigcontext {
+	unsigned long	oldmask;
+
+	/* CPU registers */
+	unsigned long sc_pc;
+	unsigned long sc_ps;
+	unsigned long sc_wmask;
+	unsigned long sc_windowbase;
+	unsigned long sc_windowstart;
+	unsigned long sc_lbeg;
+	unsigned long sc_lend;
+	unsigned long sc_lcount;
+	unsigned long sc_sar;
+	unsigned long sc_depc;
+	unsigned long sc_dareg0;
+	unsigned long sc_treg[4];
+	unsigned long sc_areg[XCHAL_NUM_AREGS];
+	struct _cpstate *sc_cpstate;
+};
+
+#endif /* __ASM_XTENSA_SIGCONTEXT_H */
diff --git a/include/asm-xtensa/siginfo.h b/include/asm-xtensa/siginfo.h
new file mode 100644
index 00000000000..44f0ae77b53
--- /dev/null
+++ b/include/asm-xtensa/siginfo.h
@@ -0,0 +1,16 @@
+/*
+ * include/asm-xtensa/processor.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_SIGINFO_H
+#define _XTENSA_SIGINFO_H
+
+#include <asm-generic/siginfo.h>
+
+#endif	/* _XTENSA_SIGINFO_H */
diff --git a/include/asm-xtensa/signal.h b/include/asm-xtensa/signal.h
new file mode 100644
index 00000000000..5d6fc9cdf58
--- /dev/null
+++ b/include/asm-xtensa/signal.h
@@ -0,0 +1,187 @@
+/*
+ * include/asm-xtensa/signal.h
+ *
+ * Swiped from SH.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_SIGNAL_H
+#define _XTENSA_SIGNAL_H
+
+
+#define _NSIG		64
+#define _NSIG_BPW	32
+#define _NSIG_WORDS	(_NSIG / _NSIG_BPW)
+
+#ifndef __ASSEMBLY__
+
+#include <linux/types.h>
+
+/* Avoid too many header ordering problems.  */
+struct siginfo;
+typedef unsigned long old_sigset_t;		/* at least 32 bits */
+typedef struct {
+	unsigned long sig[_NSIG_WORDS];
+} sigset_t;
+
+#endif
+
+#define SIGHUP		 1
+#define SIGINT		 2
+#define SIGQUIT		 3
+#define SIGILL		 4
+#define SIGTRAP		 5
+#define SIGABRT		 6
+#define SIGIOT		 6
+#define SIGBUS		 7
+#define SIGFPE		 8
+#define SIGKILL		 9
+#define SIGUSR1		10
+#define SIGSEGV		11
+#define SIGUSR2		12
+#define SIGPIPE		13
+#define SIGALRM		14
+#define SIGTERM		15
+#define SIGSTKFLT	16
+#define SIGCHLD		17
+#define SIGCONT		18
+#define SIGSTOP		19
+#define SIGTSTP		20
+#define SIGTTIN		21
+#define SIGTTOU		22
+#define SIGURG		23
+#define SIGXCPU		24
+#define SIGXFSZ		25
+#define SIGVTALRM	26
+#define SIGPROF		27
+#define SIGWINCH	28
+#define SIGIO		29
+#define SIGPOLL		SIGIO
+/* #define SIGLOST		29 */
+#define SIGPWR		30
+#define SIGSYS		31
+#define	SIGUNUSED	31
+
+/* These should not be considered constants from userland.  */
+#define SIGRTMIN	32
+#define SIGRTMAX	(_NSIG-1)
+
+/*
+ * SA_FLAGS values:
+ *
+ * SA_ONSTACK indicates that a registered stack_t will be used.
+ * SA_INTERRUPT is a no-op, but left due to historical reasons. Use the
+ * SA_RESTART flag to get restarting signals (which were the default long ago)
+ * SA_NOCLDSTOP flag to turn off SIGCHLD when children stop.
+ * SA_RESETHAND clears the handler when the signal is delivered.
+ * SA_NOCLDWAIT flag on SIGCHLD to inhibit zombies.
+ * SA_NODEFER prevents the current signal from being masked in the handler.
+ *
+ * SA_ONESHOT and SA_NOMASK are the historical Linux names for the Single
+ * Unix names RESETHAND and NODEFER respectively.
+ */
+#define SA_NOCLDSTOP	0x00000001
+#define SA_NOCLDWAIT	0x00000002 /* not supported yet */
+#define SA_SIGINFO	0x00000004
+#define SA_ONSTACK	0x08000000
+#define SA_RESTART	0x10000000
+#define SA_NODEFER	0x40000000
+#define SA_RESETHAND	0x80000000
+
+#define SA_NOMASK	SA_NODEFER
+#define SA_ONESHOT	SA_RESETHAND
+#define SA_INTERRUPT	0x20000000 /* dummy -- ignored */
+
+#define SA_RESTORER	0x04000000
+
+/*
+ * sigaltstack controls
+ */
+#define SS_ONSTACK	1
+#define SS_DISABLE	2
+
+#define MINSIGSTKSZ	2048
+#define SIGSTKSZ	8192
+
+#ifndef __ASSEMBLY__
+#ifdef __KERNEL__
+
+/*
+ * These values of sa_flags are used only by the kernel as part of the
+ * irq handling routines.
+ *
+ * SA_INTERRUPT is also used by the irq handling routines.
+ * SA_SHIRQ is for shared interrupt support on PCI and EISA.
+ */
+#define SA_PROBE		SA_ONESHOT
+#define SA_SAMPLE_RANDOM	SA_RESTART
+#define SA_SHIRQ		0x04000000
+#endif
+
+#define SIG_BLOCK          0	/* for blocking signals */
+#define SIG_UNBLOCK        1	/* for unblocking signals */
+#define SIG_SETMASK        2	/* for setting the signal mask */
+
+/* Type of a signal handler.  */
+typedef void (*__sighandler_t)(int);
+
+#define SIG_DFL	((__sighandler_t)0)	/* default signal handling */
+#define SIG_IGN	((__sighandler_t)1)	/* ignore signal */
+#define SIG_ERR	((__sighandler_t)-1)	/* error return from signal */
+
+#ifdef __KERNEL__
+struct old_sigaction {
+	__sighandler_t sa_handler;
+	old_sigset_t sa_mask;
+	unsigned long sa_flags;
+	void (*sa_restorer)(void);
+};
+
+struct sigaction {
+	__sighandler_t sa_handler;
+	unsigned long sa_flags;
+	void (*sa_restorer)(void);
+	sigset_t sa_mask;		/* mask last for extensibility */
+};
+
+struct k_sigaction {
+	struct sigaction sa;
+};
+
+#else
+
+/* Here we must cater to libcs that poke about in kernel headers.  */
+
+struct sigaction {
+	union {
+	  __sighandler_t _sa_handler;
+	  void (*_sa_sigaction)(int, struct siginfo *, void *);
+	} _u;
+	sigset_t sa_mask;
+	unsigned long sa_flags;
+	void (*sa_restorer)(void);
+};
+
+#define sa_handler	_u._sa_handler
+#define sa_sigaction	_u._sa_sigaction
+
+#endif /* __KERNEL__ */
+
+typedef struct sigaltstack {
+	void *ss_sp;
+	int ss_flags;
+	size_t ss_size;
+} stack_t;
+
+#ifdef __KERNEL__
+#include <asm/sigcontext.h>
+#define ptrace_signal_deliver(regs, cookie) do { } while (0)
+
+#endif	/* __KERNEL__ */
+#endif	/* __ASSEMBLY__ */
+#endif	/* _XTENSA_SIGNAL_H */
diff --git a/include/asm-xtensa/smp.h b/include/asm-xtensa/smp.h
new file mode 100644
index 00000000000..83c569e3bdb
--- /dev/null
+++ b/include/asm-xtensa/smp.h
@@ -0,0 +1,27 @@
+/*
+ * include/asm-xtensa/smp.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_SMP_H
+#define _XTENSA_SMP_H
+
+extern struct xtensa_cpuinfo boot_cpu_data;
+
+#define cpu_data (&boot_cpu_data)
+#define current_cpu_data boot_cpu_data
+
+struct xtensa_cpuinfo {
+	unsigned long	*pgd_cache;
+	unsigned long	*pte_cache;
+	unsigned long	pgtable_cache_sz;
+};
+
+#define cpu_logical_map(cpu)	(cpu)
+
+#endif	/* _XTENSA_SMP_H */
diff --git a/include/asm-xtensa/socket.h b/include/asm-xtensa/socket.h
new file mode 100644
index 00000000000..daccd05a14c
--- /dev/null
+++ b/include/asm-xtensa/socket.h
@@ -0,0 +1,61 @@
+/*
+ * include/asm-xtensa/socket.h
+ *
+ * Copied from i386.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#ifndef _XTENSA_SOCKET_H
+#define _XTENSA_SOCKET_H
+
+#include <asm/sockios.h>
+
+/* For setsockoptions(2) */
+#define SOL_SOCKET	1
+
+#define SO_DEBUG	1
+#define SO_REUSEADDR	2
+#define SO_TYPE		3
+#define SO_ERROR	4
+#define SO_DONTROUTE	5
+#define SO_BROADCAST	6
+#define SO_SNDBUF	7
+#define SO_RCVBUF	8
+#define SO_KEEPALIVE	9
+#define SO_OOBINLINE	10
+#define SO_NO_CHECK	11
+#define SO_PRIORITY	12
+#define SO_LINGER	13
+#define SO_BSDCOMPAT	14
+/* To add :#define SO_REUSEPORT 15 */
+#define SO_PASSCRED	16
+#define SO_PEERCRED	17
+#define SO_RCVLOWAT	18
+#define SO_SNDLOWAT	19
+#define SO_RCVTIMEO	20
+#define SO_SNDTIMEO	21
+
+/* Security levels - as per NRL IPv6 - don't actually do anything */
+
+#define SO_SECURITY_AUTHENTICATION		22
+#define SO_SECURITY_ENCRYPTION_TRANSPORT	23
+#define SO_SECURITY_ENCRYPTION_NETWORK		24
+
+#define SO_BINDTODEVICE	25
+
+/* Socket filtering */
+
+#define SO_ATTACH_FILTER        26
+#define SO_DETACH_FILTER        27
+
+#define SO_PEERNAME		28
+#define SO_TIMESTAMP		29
+#define SCM_TIMESTAMP		SO_TIMESTAMP
+
+#define SO_ACCEPTCONN		30
+#define SO_PEERSEC		31
+
+#endif	/* _XTENSA_SOCKET_H */
diff --git a/include/asm-xtensa/sockios.h b/include/asm-xtensa/sockios.h
new file mode 100644
index 00000000000..20d2ba10ecd
--- /dev/null
+++ b/include/asm-xtensa/sockios.h
@@ -0,0 +1,30 @@
+/*
+ * include/asm-xtensa/sockios.h
+ *
+ * Socket-level I/O control calls.  Copied from MIPS.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1995 by Ralf Baechle
+ * Copyright (C) 2001 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_SOCKIOS_H
+#define _XTENSA_SOCKIOS_H
+
+#include <asm/ioctl.h>
+
+/* Socket-level I/O control calls. */
+
+#define FIOGETOWN	_IOR('f', 123, int)
+#define FIOSETOWN 	_IOW('f', 124, int)
+
+#define SIOCATMARK	_IOR('s', 7, int)
+#define SIOCSPGRP	_IOW('s', 8, pid_t)
+#define SIOCGPGRP	_IOR('s', 9, pid_t)
+
+#define SIOCGSTAMP	0x8906		/* Get stamp - linux-specific */
+
+#endif	/* _XTENSA_SOCKIOS_H */
diff --git a/include/asm-xtensa/spinlock.h b/include/asm-xtensa/spinlock.h
new file mode 100644
index 00000000000..8ff23649581
--- /dev/null
+++ b/include/asm-xtensa/spinlock.h
@@ -0,0 +1,16 @@
+/*
+ * include/asm-xtensa/spinlock.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_SPINLOCK_H
+#define _XTENSA_SPINLOCK_H
+
+#include <linux/spinlock.h>
+
+#endif	/* _XTENSA_SPINLOCK_H */
diff --git a/include/asm-xtensa/stat.h b/include/asm-xtensa/stat.h
new file mode 100644
index 00000000000..2f4662ff6c3
--- /dev/null
+++ b/include/asm-xtensa/stat.h
@@ -0,0 +1,105 @@
+/*
+ * include/asm-xtensa/stat.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_STAT_H
+#define _XTENSA_STAT_H
+
+#include <linux/types.h>
+
+struct __old_kernel_stat {
+	unsigned short st_dev;
+	unsigned short st_ino;
+	unsigned short st_mode;
+	unsigned short st_nlink;
+	unsigned short st_uid;
+	unsigned short st_gid;
+	unsigned short st_rdev;
+	unsigned long  st_size;
+	unsigned long  st_atime;
+	unsigned long  st_mtime;
+	unsigned long  st_ctime;
+};
+
+#define STAT_HAVE_NSEC 1
+
+struct stat {
+	unsigned short st_dev;
+	unsigned short __pad1;
+	unsigned long st_ino;
+	unsigned short st_mode;
+	unsigned short st_nlink;
+	unsigned short st_uid;
+	unsigned short st_gid;
+	unsigned short st_rdev;
+	unsigned short __pad2;
+	unsigned long  st_size;
+	unsigned long  st_blksize;
+	unsigned long  st_blocks;
+	unsigned long  st_atime;
+	unsigned long  st_atime_nsec;
+	unsigned long  st_mtime;
+	unsigned long  st_mtime_nsec;
+	unsigned long  st_ctime;
+	unsigned long  st_ctime_nsec;
+	unsigned long  __unused4;
+	unsigned long  __unused5;
+};
+
+/* This matches struct stat64 in glibc-2.2.3. */
+
+struct stat64  {
+#ifdef __XTENSA_EL__
+	unsigned short	st_dev;		/* Device */
+	unsigned char	__pad0[10];
+#else
+	unsigned char	__pad0[6];
+	unsigned short	st_dev;
+	unsigned char	__pad1[2];
+#endif
+
+#define STAT64_HAS_BROKEN_ST_INO	1
+	unsigned long __st_ino;		/* 32bit file serial number. */
+
+	unsigned int  st_mode;		/* File mode. */
+	unsigned int  st_nlink;		/* Link count. */
+	unsigned int  st_uid;		/* User ID of the file's owner. */
+	unsigned int  st_gid;		/* Group ID of the file's group. */
+
+#ifdef __XTENSA_EL__
+	unsigned short	st_rdev;	/* Device number, if device. */
+	unsigned char	__pad3[10];
+#else
+	unsigned char	__pad2[6];
+	unsigned short	st_rdev;
+	unsigned char	__pad3[2];
+#endif
+
+	long long int  st_size;		/* Size of file, in bytes. */
+	long int st_blksize;		/* Optimal block size for I/O. */
+
+#ifdef __XTENSA_EL__
+	unsigned long  st_blocks;	/* Number 512-byte blocks allocated. */
+	unsigned long  __pad4;
+#else
+	unsigned long  __pad4;
+	unsigned long  st_blocks;
+#endif
+
+	unsigned long  __pad5;
+	long int st_atime;		/* Time of last access. */
+	unsigned long  st_atime_nsec;
+	long int st_mtime;		/* Time of last modification. */
+	unsigned long  st_mtime_nsec;
+	long int  st_ctime;		/* Time of last status change. */
+	unsigned long  st_ctime_nsec;
+	unsigned long long int st_ino;	/* File serial number. */
+};
+
+#endif	/* _XTENSA_STAT_H */
diff --git a/include/asm-xtensa/statfs.h b/include/asm-xtensa/statfs.h
new file mode 100644
index 00000000000..9c3d1a21313
--- /dev/null
+++ b/include/asm-xtensa/statfs.h
@@ -0,0 +1,17 @@
+/*
+ * include/asm-xtensa/statfs.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_STATFS_H
+#define _XTENSA_STATFS_H
+
+#include <asm-generic/statfs.h>
+
+#endif	/* _XTENSA_STATFS_H */
+
diff --git a/include/asm-xtensa/string.h b/include/asm-xtensa/string.h
new file mode 100644
index 00000000000..3f81b27d980
--- /dev/null
+++ b/include/asm-xtensa/string.h
@@ -0,0 +1,124 @@
+/*
+ * include/asm-xtensa/string.h
+ *
+ * These trivial string functions are considered part of the public domain.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+/* We should optimize these. See arch/xtensa/lib/strncpy_user.S */
+
+#ifndef _XTENSA_STRING_H
+#define _XTENSA_STRING_H
+
+#define __HAVE_ARCH_STRCPY
+extern __inline__ char *strcpy(char *__dest, const char *__src)
+{
+	register char *__xdest = __dest;
+	unsigned long __dummy;
+
+	__asm__ __volatile__("1:\n\t"
+		"l8ui	%2, %1, 0\n\t"
+		"s8i	%2, %0, 0\n\t"
+		"addi	%1, %1, 1\n\t"
+		"addi	%0, %0, 1\n\t"
+		"bnez	%2, 1b\n\t"
+		: "=r" (__dest), "=r" (__src), "=&r" (__dummy)
+		: "0" (__dest), "1" (__src)
+		: "memory");
+
+	return __xdest;
+}
+
+#define __HAVE_ARCH_STRNCPY
+extern __inline__ char *strncpy(char *__dest, const char *__src, size_t __n)
+{
+	register char *__xdest = __dest;
+	unsigned long __dummy;
+
+	if (__n == 0)
+		return __xdest;
+
+	__asm__ __volatile__(
+		"1:\n\t"
+		"l8ui	%2, %1, 0\n\t"
+		"s8i	%2, %0, 0\n\t"
+		"addi	%1, %1, 1\n\t"
+		"addi	%0, %0, 1\n\t"
+		"beqz	%2, 2f\n\t"
+		"bne	%1, %5, 1b\n"
+		"2:"
+		: "=r" (__dest), "=r" (__src), "=&r" (__dummy)
+		: "0" (__dest), "1" (__src), "r" (__src+__n)
+		: "memory");
+
+	return __xdest;
+}
+
+#define __HAVE_ARCH_STRCMP
+extern __inline__ int strcmp(const char *__cs, const char *__ct)
+{
+	register int __res;
+	unsigned long __dummy;
+
+	__asm__ __volatile__(
+		"1:\n\t"
+		"l8ui	%3, %1, 0\n\t"
+		"addi	%1, %1, 1\n\t"
+		"l8ui	%2, %0, 0\n\t"
+		"addi	%0, %0, 1\n\t"
+		"beqz	%2, 2f\n\t"
+		"beq	%2, %3, 1b\n"
+		"2:\n\t"
+		"sub	%2, %3, %2"
+		: "=r" (__cs), "=r" (__ct), "=&r" (__res), "=&r" (__dummy)
+		: "0" (__cs), "1" (__ct));
+
+	return __res;
+}
+
+#define __HAVE_ARCH_STRNCMP
+extern __inline__ int strncmp(const char *__cs, const char *__ct, size_t __n)
+{
+	register int __res;
+	unsigned long __dummy;
+
+	__asm__ __volatile__(
+		"mov	%2, %3\n"
+		"1:\n\t"
+		"beq	%0, %6, 2f\n\t"
+		"l8ui	%3, %1, 0\n\t"
+		"addi	%1, %1, 1\n\t"
+		"l8ui	%2, %0, 0\n\t"
+		"addi	%0, %0, 1\n\t"
+		"beqz	%2, 2f\n\t"
+		"beqz	%3, 2f\n\t"
+		"beq	%2, %3, 1b\n"
+		"2:\n\t"
+		"sub	%2, %3, %2"
+		: "=r" (__cs), "=r" (__ct), "=&r" (__res), "=&r" (__dummy)
+		: "0" (__cs), "1" (__ct), "r" (__cs+__n));
+
+	return __res;
+}
+
+#define __HAVE_ARCH_MEMSET
+extern void *memset(void *__s, int __c, size_t __count);
+
+#define __HAVE_ARCH_MEMCPY
+extern void *memcpy(void *__to, __const__ void *__from, size_t __n);
+
+#define __HAVE_ARCH_MEMMOVE
+extern void *memmove(void *__dest, __const__ void *__src, size_t __n);
+
+/* Don't build bcopy at all ...  */
+#define __HAVE_ARCH_BCOPY
+
+#define __HAVE_ARCH_MEMSCAN
+#define memscan memchr
+
+#endif	/* _XTENSA_STRING_H */
diff --git a/include/asm-xtensa/system.h b/include/asm-xtensa/system.h
new file mode 100644
index 00000000000..690fe325e67
--- /dev/null
+++ b/include/asm-xtensa/system.h
@@ -0,0 +1,252 @@
+/*
+ * include/asm-xtensa/system.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_SYSTEM_H
+#define _XTENSA_SYSTEM_H
+
+#include <linux/config.h>
+#include <linux/stringify.h>
+
+#include <asm/processor.h>
+
+/* interrupt control */
+
+#define local_save_flags(x)						\
+	__asm__ __volatile__ ("rsr %0,"__stringify(PS) : "=a" (x));
+#define local_irq_restore(x)	do {					\
+	__asm__ __volatile__ ("wsr %0, "__stringify(PS)" ; rsync" 	\
+	    		      :: "a" (x) : "memory"); } while(0);
+#define local_irq_save(x)	do {					\
+	__asm__ __volatile__ ("rsil %0, "__stringify(LOCKLEVEL) 	\
+	    		      : "=a" (x) :: "memory");} while(0);
+
+static inline void local_irq_disable(void)
+{
+	unsigned long flags;
+	__asm__ __volatile__ ("rsil %0, "__stringify(LOCKLEVEL)
+	    		      : "=a" (flags) :: "memory");
+}
+static inline void local_irq_enable(void)
+{
+	unsigned long flags;
+	__asm__ __volatile__ ("rsil %0, 0" : "=a" (flags) :: "memory");
+
+}
+
+static inline int irqs_disabled(void)
+{
+	unsigned long flags;
+	local_save_flags(flags);
+	return flags & 0xf;
+}
+
+#define RSR_CPENABLE(x)	do {						  \
+	__asm__ __volatile__("rsr %0," __stringify(CPENABLE) : "=a" (x)); \
+	} while(0);
+#define WSR_CPENABLE(x)	do {						  \
+  	__asm__ __volatile__("wsr %0," __stringify(CPENABLE)";rsync" 	  \
+	    		     :: "a" (x));} while(0);
+
+#define clear_cpenable() __clear_cpenable()
+
+extern __inline__ void __clear_cpenable(void)
+{
+#if XCHAL_HAVE_CP
+	unsigned long i = 0;
+	WSR_CPENABLE(i);
+#endif
+}
+
+extern __inline__ void enable_coprocessor(int i)
+{
+#if XCHAL_HAVE_CP
+	int cp;
+	RSR_CPENABLE(cp);
+	cp |= 1 << i;
+	WSR_CPENABLE(cp);
+#endif
+}
+
+extern __inline__ void disable_coprocessor(int i)
+{
+#if XCHAL_HAVE_CP
+	int cp;
+	RSR_CPENABLE(cp);
+	cp &= ~(1 << i);
+	WSR_CPENABLE(cp);
+#endif
+}
+
+#define smp_read_barrier_depends() do { } while(0)
+#define read_barrier_depends() do { } while(0)
+
+#define mb()  barrier()
+#define rmb() mb()
+#define wmb() mb()
+
+#ifdef CONFIG_SMP
+#error smp_* not defined
+#else
+#define smp_mb()	barrier()
+#define smp_rmb()	barrier()
+#define smp_wmb()	barrier()
+#endif
+
+#define set_mb(var, value)	do { var = value; mb(); } while (0)
+#define set_wmb(var, value)	do { var = value; wmb(); } while (0)
+
+#if !defined (__ASSEMBLY__)
+
+/* * switch_to(n) should switch tasks to task nr n, first
+ * checking that n isn't the current task, in which case it does nothing.
+ */
+extern void *_switch_to(void *last, void *next);
+
+#endif	/* __ASSEMBLY__ */
+
+#define prepare_to_switch()	do { } while(0)
+
+#define switch_to(prev,next,last)		\
+do {						\
+	clear_cpenable();			\
+	(last) = _switch_to(prev, next);	\
+} while(0)
+
+/*
+ * cmpxchg
+ */
+
+extern __inline__ unsigned long
+__cmpxchg_u32(volatile int *p, int old, int new)
+{
+  __asm__ __volatile__("rsil    a15, "__stringify(LOCKLEVEL)"\n\t"
+		       "l32i    %0, %1, 0              \n\t"
+		       "bne	%0, %2, 1f             \n\t"
+		       "s32i    %3, %1, 0              \n\t"
+		       "1:                             \n\t"
+		       "wsr     a15, "__stringify(PS)" \n\t"
+		       "rsync                          \n\t"
+		       : "=&a" (old)
+		       : "a" (p), "a" (old), "r" (new)
+		       : "a15", "memory");
+  return old;
+}
+/* This function doesn't exist, so you'll get a linker error
+ * if something tries to do an invalid cmpxchg(). */
+
+extern void __cmpxchg_called_with_bad_pointer(void);
+
+static __inline__ unsigned long
+__cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size)
+{
+	switch (size) {
+	case 4:  return __cmpxchg_u32(ptr, old, new);
+	default: __cmpxchg_called_with_bad_pointer();
+		 return old;
+	}
+}
+
+#define cmpxchg(ptr,o,n)						      \
+	({ __typeof__(*(ptr)) _o_ = (o);				      \
+	   __typeof__(*(ptr)) _n_ = (n);				      \
+	   (__typeof__(*(ptr))) __cmpxchg((ptr), (unsigned long)_o_,	      \
+	   			        (unsigned long)_n_, sizeof (*(ptr))); \
+	})
+
+
+
+
+/*
+ * xchg_u32
+ *
+ * Note that a15 is used here because the register allocation
+ * done by the compiler is not guaranteed and a window overflow
+ * may not occur between the rsil and wsr instructions. By using
+ * a15 in the rsil, the machine is guaranteed to be in a state
+ * where no register reference will cause an overflow.
+ */
+
+extern __inline__ unsigned long xchg_u32(volatile int * m, unsigned long val)
+{
+  unsigned long tmp;
+  __asm__ __volatile__("rsil    a15, "__stringify(LOCKLEVEL)"\n\t"
+		       "l32i    %0, %1, 0              \n\t"
+		       "s32i    %2, %1, 0              \n\t"
+		       "wsr     a15, "__stringify(PS)" \n\t"
+		       "rsync                          \n\t"
+		       : "=&a" (tmp)
+		       : "a" (m), "a" (val)
+		       : "a15", "memory");
+  return tmp;
+}
+
+#define tas(ptr) (xchg((ptr),1))
+
+#if ( __XCC__ == 1 )
+
+/* xt-xcc processes __inline__ differently than xt-gcc and decides to
+ * insert an out-of-line copy of function __xchg.  This presents the
+ * unresolved symbol at link time of __xchg_called_with_bad_pointer,
+ * even though such a function would never be called at run-time.
+ * xt-gcc always inlines __xchg, and optimizes away the undefined
+ * bad_pointer function.
+ */
+
+#define xchg(ptr,x) xchg_u32(ptr,x)
+
+#else  /* assume xt-gcc */
+
+#define xchg(ptr,x) ((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr))))
+
+/*
+ * This only works if the compiler isn't horribly bad at optimizing.
+ * gcc-2.5.8 reportedly can't handle this, but I define that one to
+ * be dead anyway.
+ */
+
+extern void __xchg_called_with_bad_pointer(void);
+
+static __inline__ unsigned long
+__xchg(unsigned long x, volatile void * ptr, int size)
+{
+	switch (size) {
+		case 4:
+			return xchg_u32(ptr, x);
+	}
+	__xchg_called_with_bad_pointer();
+	return x;
+}
+
+#endif
+
+extern void set_except_vector(int n, void *addr);
+
+static inline void spill_registers(void)
+{
+	unsigned int a0, ps;
+
+	__asm__ __volatile__ (
+		"movi	a14," __stringify (PS_EXCM_MASK) " | 1\n\t"
+		"mov	a12, a0\n\t"
+		"rsr	a13," __stringify(SAR) "\n\t"
+		"xsr	a14," __stringify(PS) "\n\t"
+		"movi	a0, _spill_registers\n\t"
+		"rsync\n\t"
+		"callx0 a0\n\t"
+		"mov	a0, a12\n\t"
+		"wsr	a13," __stringify(SAR) "\n\t"
+		"wsr	a14," __stringify(PS) "\n\t"
+		:: "a" (&a0), "a" (&ps)
+		: "a2", "a3", "a12", "a13", "a14", "a15", "memory");
+}
+
+#define arch_align_stack(x) (x)
+
+#endif	/* _XTENSA_SYSTEM_H */
diff --git a/include/asm-xtensa/termbits.h b/include/asm-xtensa/termbits.h
new file mode 100644
index 00000000000..c780593ff5f
--- /dev/null
+++ b/include/asm-xtensa/termbits.h
@@ -0,0 +1,194 @@
+/*
+ * include/asm-xtensa/termbits.h
+ *
+ * Copied from SH.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_TERMBITS_H
+#define _XTENSA_TERMBITS_H
+
+
+#include <linux/posix_types.h>
+
+typedef unsigned char	cc_t;
+typedef unsigned int	speed_t;
+typedef unsigned int	tcflag_t;
+
+#define NCCS 19
+struct termios {
+	tcflag_t c_iflag;		/* input mode flags */
+	tcflag_t c_oflag;		/* output mode flags */
+	tcflag_t c_cflag;		/* control mode flags */
+	tcflag_t c_lflag;		/* local mode flags */
+	cc_t c_line;			/* line discipline */
+	cc_t c_cc[NCCS];		/* control characters */
+};
+
+/* c_cc characters */
+
+#define VINTR 0
+#define VQUIT 1
+#define VERASE 2
+#define VKILL 3
+#define VEOF 4
+#define VTIME 5
+#define VMIN 6
+#define VSWTC 7
+#define VSTART 8
+#define VSTOP 9
+#define VSUSP 10
+#define VEOL 11
+#define VREPRINT 12
+#define VDISCARD 13
+#define VWERASE 14
+#define VLNEXT 15
+#define VEOL2 16
+
+/* c_iflag bits */
+
+#define IGNBRK	0000001
+#define BRKINT	0000002
+#define IGNPAR	0000004
+#define PARMRK	0000010
+#define INPCK	0000020
+#define ISTRIP	0000040
+#define INLCR	0000100
+#define IGNCR	0000200
+#define ICRNL	0000400
+#define IUCLC	0001000
+#define IXON	0002000
+#define IXANY	0004000
+#define IXOFF	0010000
+#define IMAXBEL	0020000
+#define IUTF8	0040000
+
+/* c_oflag bits */
+
+#define OPOST	0000001
+#define OLCUC	0000002
+#define ONLCR	0000004
+#define OCRNL	0000010
+#define ONOCR	0000020
+#define ONLRET	0000040
+#define OFILL	0000100
+#define OFDEL	0000200
+#define NLDLY	0000400
+#define   NL0	0000000
+#define   NL1	0000400
+#define CRDLY	0003000
+#define   CR0	0000000
+#define   CR1	0001000
+#define   CR2	0002000
+#define   CR3	0003000
+#define TABDLY	0014000
+#define   TAB0	0000000
+#define   TAB1	0004000
+#define   TAB2	0010000
+#define   TAB3	0014000
+#define   XTABS	0014000
+#define BSDLY	0020000
+#define   BS0	0000000
+#define   BS1	0020000
+#define VTDLY	0040000
+#define   VT0	0000000
+#define   VT1	0040000
+#define FFDLY	0100000
+#define   FF0	0000000
+#define   FF1	0100000
+
+/* c_cflag bit meaning */
+
+#define CBAUD	0010017
+#define  B0	0000000		/* hang up */
+#define  B50	0000001
+#define  B75	0000002
+#define  B110	0000003
+#define  B134	0000004
+#define  B150	0000005
+#define  B200	0000006
+#define  B300	0000007
+#define  B600	0000010
+#define  B1200	0000011
+#define  B1800	0000012
+#define  B2400	0000013
+#define  B4800	0000014
+#define  B9600	0000015
+#define  B19200	0000016
+#define  B38400	0000017
+#define EXTA B19200
+#define EXTB B38400
+#define CSIZE	0000060
+#define   CS5	0000000
+#define   CS6	0000020
+#define   CS7	0000040
+#define   CS8	0000060
+#define CSTOPB	0000100
+#define CREAD	0000200
+#define PARENB	0000400
+#define PARODD	0001000
+#define HUPCL	0002000
+#define CLOCAL	0004000
+#define CBAUDEX 0010000
+#define    B57600 0010001
+#define   B115200 0010002
+#define   B230400 0010003
+#define   B460800 0010004
+#define   B500000 0010005
+#define   B576000 0010006
+#define   B921600 0010007
+#define  B1000000 0010010
+#define  B1152000 0010011
+#define  B1500000 0010012
+#define  B2000000 0010013
+#define  B2500000 0010014
+#define  B3000000 0010015
+#define  B3500000 0010016
+#define  B4000000 0010017
+#define CIBAUD	  002003600000	/* input baud rate (not used) */
+#define CMSPAR	  010000000000		/* mark or space (stick) parity */
+#define CRTSCTS	  020000000000		/* flow control */
+
+/* c_lflag bits */
+
+#define ISIG	0000001
+#define ICANON	0000002
+#define XCASE	0000004
+#define ECHO	0000010
+#define ECHOE	0000020
+#define ECHOK	0000040
+#define ECHONL	0000100
+#define NOFLSH	0000200
+#define TOSTOP	0000400
+#define ECHOCTL	0001000
+#define ECHOPRT	0002000
+#define ECHOKE	0004000
+#define FLUSHO	0010000
+#define PENDIN	0040000
+#define IEXTEN	0100000
+
+/* tcflow() and TCXONC use these */
+
+#define	TCOOFF		0
+#define	TCOON		1
+#define	TCIOFF		2
+#define	TCION		3
+
+/* tcflush() and TCFLSH use these */
+
+#define	TCIFLUSH	0
+#define	TCOFLUSH	1
+#define	TCIOFLUSH	2
+
+/* tcsetattr uses these */
+
+#define	TCSANOW		0
+#define	TCSADRAIN	1
+#define	TCSAFLUSH	2
+
+#endif	/* _XTENSA_TERMBITS_H */
diff --git a/include/asm-xtensa/termios.h b/include/asm-xtensa/termios.h
new file mode 100644
index 00000000000..83c6aed1d11
--- /dev/null
+++ b/include/asm-xtensa/termios.h
@@ -0,0 +1,122 @@
+/*
+ * include/asm-xtensa/termios.h
+ *
+ * Copied from SH.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_TERMIOS_H
+#define _XTENSA_TERMIOS_H
+
+#include <asm/termbits.h>
+#include <asm/ioctls.h>
+
+struct winsize {
+	unsigned short ws_row;
+	unsigned short ws_col;
+	unsigned short ws_xpixel;
+	unsigned short ws_ypixel;
+};
+
+#define NCC 8
+struct termio {
+	unsigned short c_iflag;		/* input mode flags */
+	unsigned short c_oflag;		/* output mode flags */
+	unsigned short c_cflag;		/* control mode flags */
+	unsigned short c_lflag;		/* local mode flags */
+	unsigned char c_line;		/* line discipline */
+	unsigned char c_cc[NCC];	/* control characters */
+};
+
+/* Modem lines */
+
+#define TIOCM_LE	0x001
+#define TIOCM_DTR	0x002
+#define TIOCM_RTS	0x004
+#define TIOCM_ST	0x008
+#define TIOCM_SR	0x010
+#define TIOCM_CTS	0x020
+#define TIOCM_CAR	0x040
+#define TIOCM_RNG	0x080
+#define TIOCM_DSR	0x100
+#define TIOCM_CD	TIOCM_CAR
+#define TIOCM_RI	TIOCM_RNG
+#define TIOCM_OUT1	0x2000
+#define TIOCM_OUT2	0x4000
+#define TIOCM_LOOP	0x8000
+
+/* ioctl (fd, TIOCSERGETLSR, &result) where result may be as below */
+
+/* Line disciplines */
+
+#define N_TTY		0
+#define N_SLIP		1
+#define N_MOUSE		2
+#define N_PPP		3
+#define N_STRIP		4
+#define N_AX25		5
+#define N_X25		6	/* X.25 async */
+#define N_6PACK		7
+#define N_MASC		8	/* Reserved for Mobitex module <kaz@cafe.net> */
+#define N_R3964		9	/* Reserved for Simatic R3964 module */
+#define N_PROFIBUS_FDL	10	/* Reserved for Profibus <Dave@mvhi.com> */
+#define N_IRDA		11	/* Linux IR - http://irda.sourceforge.net/ */
+#define N_SMSBLOCK	12	/* SMS block mode - for talking to GSM data cards about SMS messages */
+#define N_HDLC		13	/* synchronous HDLC */
+#define N_SYNC_PPP	14
+#define N_HCI		15      /* Bluetooth HCI UART */
+
+#ifdef __KERNEL__
+
+/*	intr=^C		quit=^\		erase=del	kill=^U
+	eof=^D		vtime=\0	vmin=\1		sxtc=\0
+	start=^Q	stop=^S		susp=^Z		eol=\0
+	reprint=^R	discard=^U	werase=^W	lnext=^V
+	eol2=\0
+*/
+#define INIT_C_CC "\003\034\177\025\004\0\1\0\021\023\032\0\022\017\027\026\0"
+
+/*
+ * Translate a "termio" structure into a "termios". Ugh.
+ */
+
+#define SET_LOW_TERMIOS_BITS(termios, termio, x) { \
+	unsigned short __tmp; \
+	get_user(__tmp,&(termio)->x); \
+	*(unsigned short *) &(termios)->x = __tmp; \
+}
+
+#define user_termio_to_kernel_termios(termios, termio) \
+({ \
+	SET_LOW_TERMIOS_BITS(termios, termio, c_iflag); \
+	SET_LOW_TERMIOS_BITS(termios, termio, c_oflag); \
+	SET_LOW_TERMIOS_BITS(termios, termio, c_cflag); \
+	SET_LOW_TERMIOS_BITS(termios, termio, c_lflag); \
+	copy_from_user((termios)->c_cc, (termio)->c_cc, NCC); \
+})
+
+/*
+ * Translate a "termios" structure into a "termio". Ugh.
+ */
+
+#define kernel_termios_to_user_termio(termio, termios) \
+({ \
+	put_user((termios)->c_iflag, &(termio)->c_iflag); \
+	put_user((termios)->c_oflag, &(termio)->c_oflag); \
+	put_user((termios)->c_cflag, &(termio)->c_cflag); \
+	put_user((termios)->c_lflag, &(termio)->c_lflag); \
+	put_user((termios)->c_line,  &(termio)->c_line); \
+	copy_to_user((termio)->c_cc, (termios)->c_cc, NCC); \
+})
+
+#define user_termios_to_kernel_termios(k, u) copy_from_user(k, u, sizeof(struct termios))
+#define kernel_termios_to_user_termios(u, k) copy_to_user(u, k, sizeof(struct termios))
+
+#endif	/* __KERNEL__ */
+
+#endif	/* _XTENSA_TERMIOS_H */
diff --git a/include/asm-xtensa/thread_info.h b/include/asm-xtensa/thread_info.h
new file mode 100644
index 00000000000..af208d41fd8
--- /dev/null
+++ b/include/asm-xtensa/thread_info.h
@@ -0,0 +1,146 @@
+/*
+ * include/asm-xtensa/thread_info.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_THREAD_INFO_H
+#define _XTENSA_THREAD_INFO_H
+
+#ifdef __KERNEL__
+
+#ifndef __ASSEMBLY__
+# include <asm/processor.h>
+#endif
+
+/*
+ * low level task data that entry.S needs immediate access to
+ * - this struct should fit entirely inside of one cache line
+ * - this struct shares the supervisor stack pages
+ * - if the contents of this structure are changed, the assembly constants
+ *   must also be changed
+ */
+
+#ifndef __ASSEMBLY__
+
+struct thread_info {
+	struct task_struct	*task;		/* main task structure */
+	struct exec_domain	*exec_domain;	/* execution domain */
+	unsigned long		flags;		/* low level flags */
+	unsigned long		status;		/* thread-synchronous flags */
+	__u32			cpu;		/* current CPU */
+	__s32			preempt_count;	/* 0 => preemptable,< 0 => BUG*/
+
+	mm_segment_t		addr_limit;	/* thread address space */
+	struct restart_block    restart_block;
+
+
+};
+
+#else /* !__ASSEMBLY__ */
+
+/* offsets into the thread_info struct for assembly code access */
+#define TI_TASK		 0x00000000
+#define TI_EXEC_DOMAIN	 0x00000004
+#define TI_FLAGS	 0x00000008
+#define TI_STATUS	 0x0000000C
+#define TI_CPU		 0x00000010
+#define TI_PRE_COUNT	 0x00000014
+#define TI_ADDR_LIMIT	 0x00000018
+#define TI_RESTART_BLOCK 0x000001C
+
+#endif
+
+#define PREEMPT_ACTIVE		0x10000000
+
+/*
+ * macros/functions for gaining access to the thread information structure
+ *
+ * preempt_count needs to be 1 initially, until the scheduler is functional.
+ */
+
+#ifndef __ASSEMBLY__
+
+#define INIT_THREAD_INFO(tsk)			\
+{						\
+	.task		= &tsk,			\
+	.exec_domain	= &default_exec_domain,	\
+	.flags		= 0,			\
+	.cpu		= 0,			\
+	.preempt_count	= 1,			\
+	.addr_limit	= KERNEL_DS,		\
+	.restart_block = {			\
+		.fn = do_no_restart_syscall,	\
+	},					\
+}
+
+#define init_thread_info	(init_thread_union.thread_info)
+#define init_stack		(init_thread_union.stack)
+
+/* how to get the thread information struct from C */
+static inline struct thread_info *current_thread_info(void)
+{
+	struct thread_info *ti;
+	 __asm__("extui %0,a1,0,13\n\t"
+	         "xor %0, a1, %0" : "=&r" (ti) : );
+	return ti;
+}
+
+/* thread information allocation */
+#define alloc_thread_info(tsk) ((struct thread_info *) __get_free_pages(GFP_KERNEL,1))
+#define free_thread_info(ti) free_pages((unsigned long) (ti), 1)
+#define get_thread_info(ti) get_task_struct((ti)->task)
+#define put_thread_info(ti) put_task_struct((ti)->task)
+
+#else /* !__ASSEMBLY__ */
+
+/* how to get the thread information struct from ASM */
+#define GET_THREAD_INFO(reg,sp) \
+	extui reg, sp, 0, 13; \
+	xor   reg, sp, reg
+#endif
+
+
+/*
+ * thread information flags
+ * - these are process state flags that various assembly files may need to access
+ * - pending work-to-be-done flags are in LSW
+ * - other flags in MSW
+ */
+#define TIF_SYSCALL_TRACE	0	/* syscall trace active */
+#define TIF_NOTIFY_RESUME	1	/* resumption notification requested */
+#define TIF_SIGPENDING		2	/* signal pending */
+#define TIF_NEED_RESCHED	3	/* rescheduling necessary */
+#define TIF_SINGLESTEP		4	/* restore singlestep on return to user mode */
+#define TIF_IRET		5	/* return with iret */
+#define TIF_MEMDIE		6
+#define TIF_POLLING_NRFLAG	16	/* true if poll_idle() is polling TIF_NEED_RESCHED */
+
+#define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
+#define _TIF_NOTIFY_RESUME	(1<<TIF_NOTIFY_RESUME)
+#define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
+#define _TIF_NEED_RESCHED	(1<<TIF_NEED_RESCHED)
+#define _TIF_SINGLESTEP		(1<<TIF_SINGLESTEP)
+#define _TIF_IRET		(1<<TIF_IRET)
+#define _TIF_POLLING_NRFLAG	(1<<TIF_POLLING_NRFLAG)
+
+#define _TIF_WORK_MASK		0x0000FFFE	/* work to do on interrupt/exception return */
+#define _TIF_ALLWORK_MASK	0x0000FFFF	/* work to do on any return to u-space */
+
+/*
+ * Thread-synchronous status.
+ *
+ * This is different from the flags in that nobody else
+ * ever touches our thread-synchronous status, so we don't
+ * have to worry about atomic accesses.
+ */
+#define TS_USEDFPU		0x0001	/* FPU was used by this task this quantum (SMP) */
+
+#define THREAD_SIZE 8192	//(2*PAGE_SIZE)
+
+#endif	/* __KERNEL__ */
+#endif	/* _XTENSA_THREAD_INFO */
diff --git a/include/asm-xtensa/timex.h b/include/asm-xtensa/timex.h
new file mode 100644
index 00000000000..d14a3755a12
--- /dev/null
+++ b/include/asm-xtensa/timex.h
@@ -0,0 +1,94 @@
+/*
+ * include/asm-xtensa/timex.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_TIMEX_H
+#define _XTENSA_TIMEX_H
+
+#ifdef __KERNEL__
+
+#include <asm/processor.h>
+#include <linux/stringify.h>
+
+#if XCHAL_INT_LEVEL(XCHAL_TIMER0_INTERRUPT) == 1
+# define LINUX_TIMER     0
+#elif XCHAL_INT_LEVEL(XCHAL_TIMER1_INTERRUPT) == 1
+# define LINUX_TIMER     1
+#elif XCHAL_INT_LEVEL(XCHAL_TIMER2_INTERRUPT) == 1
+# define LINUX_TIMER     2
+#else
+# error "Bad timer number for Linux configurations!"
+#endif
+
+#define LINUX_TIMER_INT         XCHAL_TIMER_INTERRUPT(LINUX_TIMER)
+#define LINUX_TIMER_MASK        (1L << LINUX_TIMER_INT)
+
+#define CLOCK_TICK_RATE 	1193180	/* (everyone is using this value) */
+#define CLOCK_TICK_FACTOR       20 /* Factor of both 10^6 and CLOCK_TICK_RATE */
+#define FINETUNE ((((((long)LATCH * HZ - CLOCK_TICK_RATE) << SHIFT_HZ) * \
+	(1000000/CLOCK_TICK_FACTOR) / (CLOCK_TICK_RATE/CLOCK_TICK_FACTOR)) \
+		<< (SHIFT_SCALE-SHIFT_HZ)) / HZ)
+
+#ifdef CONFIG_XTENSA_CALIBRATE_CCOUNT
+extern unsigned long ccount_per_jiffy;
+extern unsigned long ccount_nsec;
+#define CCOUNT_PER_JIFFY ccount_per_jiffy
+#define CCOUNT_NSEC ccount_nsec
+#else
+#define CCOUNT_PER_JIFFY (CONFIG_XTENSA_CPU_CLOCK*(1000000UL/HZ))
+#define CCOUNT_NSEC (1000000000UL / CONFIG_XTENSA_CPU_CLOCK)
+#endif
+
+
+typedef unsigned long long cycles_t;
+
+/*
+ * Only used for SMP.
+ */
+
+extern cycles_t cacheflush_time;
+
+#define get_cycles()	(0)
+
+
+/*
+ * Register access.
+ */
+
+#define WSR_CCOUNT(r)	  __asm__("wsr %0,"__stringify(CCOUNT) :: "a" (r))
+#define RSR_CCOUNT(r)	  __asm__("rsr %0,"__stringify(CCOUNT) : "=a" (r))
+#define WSR_CCOMPARE(x,r) __asm__("wsr %0,"__stringify(CCOMPARE_0)"+"__stringify(x) :: "a"(r))
+#define RSR_CCOMPARE(x,r) __asm__("rsr %0,"__stringify(CCOMPARE_0)"+"__stringify(x) : "=a"(r))
+
+static inline unsigned long get_ccount (void)
+{
+	unsigned long ccount;
+	RSR_CCOUNT(ccount);
+	return ccount;
+}
+
+static inline void set_ccount (unsigned long ccount)
+{
+	WSR_CCOUNT(ccount);
+}
+
+static inline unsigned long get_linux_timer (void)
+{
+	unsigned ccompare;
+	RSR_CCOMPARE(LINUX_TIMER, ccompare);
+	return ccompare;
+}
+
+static inline void set_linux_timer (unsigned long ccompare)
+{
+	WSR_CCOMPARE(LINUX_TIMER, ccompare);
+}
+
+#endif	/* __KERNEL__ */
+#endif	/* _XTENSA_TIMEX_H */
diff --git a/include/asm-xtensa/tlb.h b/include/asm-xtensa/tlb.h
new file mode 100644
index 00000000000..4562b2dcfbc
--- /dev/null
+++ b/include/asm-xtensa/tlb.h
@@ -0,0 +1,25 @@
+/*
+ * include/asm-xtensa/tlb.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_TLB_H
+#define _XTENSA_TLB_H
+
+#define tlb_start_vma(tlb,vma)			do { } while (0)
+#define tlb_end_vma(tlb,vma)			do { } while (0)
+#define __tlb_remove_tlb_entry(tlb,pte,addr)	do { } while (0)
+
+#define tlb_flush(tlb)				flush_tlb_mm((tlb)->mm)
+
+#include <asm-generic/tlb.h>
+#include <asm/page.h>
+
+#define __pte_free_tlb(tlb,pte)			pte_free(pte)
+
+#endif	/* _XTENSA_TLB_H */
diff --git a/include/asm-xtensa/tlbflush.h b/include/asm-xtensa/tlbflush.h
new file mode 100644
index 00000000000..23bfe9db45f
--- /dev/null
+++ b/include/asm-xtensa/tlbflush.h
@@ -0,0 +1,200 @@
+/*
+ * include/asm-xtensa/tlbflush.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_TLBFLUSH_H
+#define _XTENSA_TLBFLUSH_H
+
+#define DEBUG_TLB
+
+#ifdef __KERNEL__
+
+#include <asm/processor.h>
+#include <linux/stringify.h>
+
+/* TLB flushing:
+ *
+ *  - flush_tlb_all() flushes all processes TLB entries
+ *  - flush_tlb_mm(mm) flushes the specified mm context TLB entries
+ *  - flush_tlb_page(mm, vmaddr) flushes a single page
+ *  - flush_tlb_range(mm, start, end) flushes a range of pages
+ */
+
+extern void flush_tlb_all(void);
+extern void flush_tlb_mm(struct mm_struct*);
+extern void flush_tlb_page(struct vm_area_struct*,unsigned long);
+extern void flush_tlb_range(struct vm_area_struct*,unsigned long,unsigned long);
+
+#define flush_tlb_kernel_range(start,end) flush_tlb_all()
+
+
+/* This is calld in munmap when we have freed up some page-table pages.
+ * We don't need to do anything here, there's nothing special about our
+ * page-table pages.
+ */
+
+extern inline void flush_tlb_pgtables(struct mm_struct *mm,
+                                      unsigned long start, unsigned long end)
+{
+}
+
+/* TLB operations. */
+
+#define ITLB_WAYS_LOG2      XCHAL_ITLB_WAY_BITS
+#define DTLB_WAYS_LOG2      XCHAL_DTLB_WAY_BITS
+#define ITLB_PROBE_SUCCESS  (1 << ITLB_WAYS_LOG2)
+#define DTLB_PROBE_SUCCESS  (1 << DTLB_WAYS_LOG2)
+
+extern inline unsigned long itlb_probe(unsigned long addr)
+{
+	unsigned long tmp;
+	__asm__ __volatile__("pitlb  %0, %1\n\t" : "=a" (tmp) : "a" (addr));
+	return tmp;
+}
+
+extern inline unsigned long dtlb_probe(unsigned long addr)
+{
+	unsigned long tmp;
+	__asm__ __volatile__("pdtlb  %0, %1\n\t" : "=a" (tmp) : "a" (addr));
+	return tmp;
+}
+
+extern inline void invalidate_itlb_entry (unsigned long probe)
+{
+	__asm__ __volatile__("iitlb  %0; isync\n\t" : : "a" (probe));
+}
+
+extern inline void invalidate_dtlb_entry (unsigned long probe)
+{
+	__asm__ __volatile__("idtlb  %0; dsync\n\t" : : "a" (probe));
+}
+
+/* Use the .._no_isync functions with caution.  Generally, these are
+ * handy for bulk invalidates followed by a single 'isync'.  The
+ * caller must follow up with an 'isync', which can be relatively
+ * expensive on some Xtensa implementations.
+ */
+extern inline void invalidate_itlb_entry_no_isync (unsigned entry)
+{
+	/* Caller must follow up with 'isync'. */
+	__asm__ __volatile__ ("iitlb  %0\n" : : "a" (entry) );
+}
+
+extern inline void invalidate_dtlb_entry_no_isync (unsigned entry)
+{
+	/* Caller must follow up with 'isync'. */
+	__asm__ __volatile__ ("idtlb  %0\n" : : "a" (entry) );
+}
+
+extern inline void set_itlbcfg_register (unsigned long val)
+{
+	__asm__ __volatile__("wsr  %0, "__stringify(ITLBCFG)"\n\t" "isync\n\t"
+			     : : "a" (val));
+}
+
+extern inline void set_dtlbcfg_register (unsigned long val)
+{
+	__asm__ __volatile__("wsr  %0, "__stringify(DTLBCFG)"; dsync\n\t"
+	    		     : : "a" (val));
+}
+
+extern inline void set_ptevaddr_register (unsigned long val)
+{
+	__asm__ __volatile__(" wsr  %0, "__stringify(PTEVADDR)"; isync\n"
+			     : : "a" (val));
+}
+
+extern inline unsigned long read_ptevaddr_register (void)
+{
+	unsigned long tmp;
+	__asm__ __volatile__("rsr  %0, "__stringify(PTEVADDR)"\n\t" : "=a" (tmp));
+	return tmp;
+}
+
+extern inline void write_dtlb_entry (pte_t entry, int way)
+{
+	__asm__ __volatile__("wdtlb  %1, %0; dsync\n\t"
+			     : : "r" (way), "r" (entry) );
+}
+
+extern inline void write_itlb_entry (pte_t entry, int way)
+{
+	__asm__ __volatile__("witlb  %1, %0; isync\n\t"
+	                     : : "r" (way), "r" (entry) );
+}
+
+extern inline void invalidate_page_directory (void)
+{
+	invalidate_dtlb_entry (DTLB_WAY_PGTABLE);
+}
+
+extern inline void invalidate_itlb_mapping (unsigned address)
+{
+	unsigned long tlb_entry;
+	while ((tlb_entry = itlb_probe (address)) & ITLB_PROBE_SUCCESS)
+		invalidate_itlb_entry (tlb_entry);
+}
+
+extern inline void invalidate_dtlb_mapping (unsigned address)
+{
+	unsigned long tlb_entry;
+	while ((tlb_entry = dtlb_probe (address)) & DTLB_PROBE_SUCCESS)
+		invalidate_dtlb_entry (tlb_entry);
+}
+
+#define check_pgt_cache()	do { } while (0)
+
+
+#ifdef DEBUG_TLB
+
+/* DO NOT USE THESE FUNCTIONS.  These instructions aren't part of the Xtensa
+ * ISA and exist only for test purposes..
+ * You may find it helpful for MMU debugging, however.
+ *
+ * 'at' is the unmodified input register
+ * 'as' is the output register, as follows (specific to the Linux config):
+ *
+ *      as[31..12] contain the virtual address
+ *      as[11..08] are meaningless
+ *      as[07..00] contain the asid
+ */
+
+extern inline unsigned long read_dtlb_virtual (int way)
+{
+	unsigned long tmp;
+	__asm__ __volatile__("rdtlb0  %0, %1\n\t" : "=a" (tmp), "+a" (way));
+	return tmp;
+}
+
+extern inline unsigned long read_dtlb_translation (int way)
+{
+	unsigned long tmp;
+	__asm__ __volatile__("rdtlb1  %0, %1\n\t" : "=a" (tmp), "+a" (way));
+	return tmp;
+}
+
+extern inline unsigned long read_itlb_virtual (int way)
+{
+	unsigned long tmp;
+	__asm__ __volatile__("ritlb0  %0, %1\n\t" : "=a" (tmp), "+a" (way));
+	return tmp;
+}
+
+extern inline unsigned long read_itlb_translation (int way)
+{
+	unsigned long tmp;
+	__asm__ __volatile__("ritlb1  %0, %1\n\t" : "=a" (tmp), "+a" (way));
+	return tmp;
+}
+
+#endif	/* DEBUG_TLB */
+
+
+#endif	/* __KERNEL__ */
+#endif	/* _XTENSA_PGALLOC_H */
diff --git a/include/asm-xtensa/topology.h b/include/asm-xtensa/topology.h
new file mode 100644
index 00000000000..7309e38a0cc
--- /dev/null
+++ b/include/asm-xtensa/topology.h
@@ -0,0 +1,16 @@
+/*
+ * include/asm-xtensa/topology.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_TOPOLOGY_H
+#define _XTENSA_TOPOLOGY_H
+
+#include <asm-generic/topology.h>
+
+#endif	/* _XTENSA_TOPOLOGY_H */
diff --git a/include/asm-xtensa/types.h b/include/asm-xtensa/types.h
new file mode 100644
index 00000000000..ebac0046985
--- /dev/null
+++ b/include/asm-xtensa/types.h
@@ -0,0 +1,66 @@
+/*
+ * include/asm-xtensa/types.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_TYPES_H
+#define _XTENSA_TYPES_H
+
+#ifndef __ASSEMBLY__
+
+typedef unsigned short umode_t;
+
+/*
+ * __xx is ok: it doesn't pollute the POSIX namespace. Use these in the
+ * header files exported to user space
+ */
+
+typedef __signed__ char __s8;
+typedef unsigned char __u8;
+
+typedef __signed__ short __s16;
+typedef unsigned short __u16;
+
+typedef __signed__ int __s32;
+typedef unsigned int __u32;
+
+#if defined(__GNUC__) && !defined(__STRICT_ANSI__)
+typedef __signed__ long long __s64;
+typedef unsigned long long __u64;
+#endif
+
+/*
+ * These aren't exported outside the kernel to avoid name space clashes
+ */
+#ifdef __KERNEL__
+
+typedef __signed__ char s8;
+typedef unsigned char u8;
+
+typedef __signed__ short s16;
+typedef unsigned short u16;
+
+typedef __signed__ int s32;
+typedef unsigned int u32;
+
+typedef __signed__ long long s64;
+typedef unsigned long long u64;
+
+
+#define BITS_PER_LONG 32
+
+/* Dma addresses are 32-bits wide.  */
+
+typedef u32 dma_addr_t;
+
+typedef unsigned int kmem_bufctl_t;
+
+#endif	/* __KERNEL__ */
+#endif
+
+#endif	/* _XTENSA_TYPES_H */
diff --git a/include/asm-xtensa/uaccess.h b/include/asm-xtensa/uaccess.h
new file mode 100644
index 00000000000..35576b25c7b
--- /dev/null
+++ b/include/asm-xtensa/uaccess.h
@@ -0,0 +1,532 @@
+/*
+ * include/asm-xtensa/uaccess.h
+ *
+ * User space memory access functions
+ *
+ * These routines provide basic accessing functions to the user memory
+ * space for the kernel. This header file provides fuctions such as:
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_UACCESS_H
+#define _XTENSA_UACCESS_H
+
+#include <linux/errno.h>
+
+#define VERIFY_READ    0
+#define VERIFY_WRITE   1
+
+#ifdef __ASSEMBLY__
+
+#define _ASMLANGUAGE
+#include <asm/current.h>
+#include <asm/offsets.h>
+#include <asm/processor.h>
+
+/*
+ * These assembly macros mirror the C macros that follow below.  They
+ * should always have identical functionality.  See
+ * arch/xtensa/kernel/sys.S for usage.
+ */
+
+#define KERNEL_DS	0
+#define USER_DS		1
+
+#define get_ds		(KERNEL_DS)
+
+/*
+ * get_fs reads current->thread.current_ds into a register.
+ * On Entry:
+ * 	<ad>	anything
+ * 	<sp>	stack
+ * On Exit:
+ * 	<ad>	contains current->thread.current_ds
+ */
+	.macro	get_fs	ad, sp
+	GET_CURRENT(\ad,\sp)
+	l32i	\ad, \ad, THREAD_CURRENT_DS
+	.endm
+
+/*
+ * set_fs sets current->thread.current_ds to some value.
+ * On Entry:
+ *	<at>	anything (temp register)
+ *	<av>	value to write
+ *	<sp>	stack
+ * On Exit:
+ *	<at>	destroyed (actually, current)
+ *	<av>	preserved, value to write
+ */
+	.macro	set_fs	at, av, sp
+	GET_CURRENT(\at,\sp)
+	s32i	\av, \at, THREAD_CURRENT_DS
+	.endm
+
+/*
+ * kernel_ok determines whether we should bypass addr/size checking.
+ * See the equivalent C-macro version below for clarity.
+ * On success, kernel_ok branches to a label indicated by parameter
+ * <success>.  This implies that the macro falls through to the next
+ * insruction on an error.
+ *
+ * Note that while this macro can be used independently, we designed
+ * in for optimal use in the access_ok macro below (i.e., we fall
+ * through on error).
+ *
+ * On Entry:
+ * 	<at>		anything (temp register)
+ * 	<success>	label to branch to on success; implies
+ * 			fall-through macro on error
+ * 	<sp>		stack pointer
+ * On Exit:
+ * 	<at>		destroyed (actually, current->thread.current_ds)
+ */
+
+#if ((KERNEL_DS != 0) || (USER_DS == 0))
+# error Assembly macro kernel_ok fails
+#endif
+	.macro	kernel_ok  at, sp, success
+	get_fs	\at, \sp
+	beqz	\at, \success
+	.endm
+
+/*
+ * user_ok determines whether the access to user-space memory is allowed.
+ * See the equivalent C-macro version below for clarity.
+ *
+ * On error, user_ok branches to a label indicated by parameter
+ * <error>.  This implies that the macro falls through to the next
+ * instruction on success.
+ *
+ * Note that while this macro can be used independently, we designed
+ * in for optimal use in the access_ok macro below (i.e., we fall
+ * through on success).
+ *
+ * On Entry:
+ * 	<aa>	register containing memory address
+ * 	<as>	register containing memory size
+ * 	<at>	temp register
+ * 	<error>	label to branch to on error; implies fall-through
+ * 		macro on success
+ * On Exit:
+ * 	<aa>	preserved
+ * 	<as>	preserved
+ * 	<at>	destroyed (actually, (TASK_SIZE + 1 - size))
+ */
+	.macro	user_ok	aa, as, at, error
+	movi	\at, (TASK_SIZE+1)
+	bgeu	\as, \at, \error
+	sub	\at, \at, \as
+	bgeu	\aa, \at, \error
+	.endm
+
+/*
+ * access_ok determines whether a memory access is allowed.  See the
+ * equivalent C-macro version below for clarity.
+ *
+ * On error, access_ok branches to a label indicated by parameter
+ * <error>.  This implies that the macro falls through to the next
+ * instruction on success.
+ *
+ * Note that we assume success is the common case, and we optimize the
+ * branch fall-through case on success.
+ *
+ * On Entry:
+ * 	<aa>	register containing memory address
+ * 	<as>	register containing memory size
+ * 	<at>	temp register
+ * 	<sp>
+ * 	<error>	label to branch to on error; implies fall-through
+ * 		macro on success
+ * On Exit:
+ * 	<aa>	preserved
+ * 	<as>	preserved
+ * 	<at>	destroyed
+ */
+	.macro	access_ok  aa, as, at, sp, error
+	kernel_ok  \at, \sp, .Laccess_ok_\@
+	user_ok    \aa, \as, \at, \error
+.Laccess_ok_\@:
+	.endm
+
+/*
+ * verify_area determines whether a memory access is allowed.  It's
+ * mostly an unnecessary wrapper for access_ok, but we provide it as a
+ * duplicate of the verify_area() C inline function below.  See the
+ * equivalent C version below for clarity.
+ *
+ * On error, verify_area branches to a label indicated by parameter
+ * <error>.  This implies that the macro falls through to the next
+ * instruction on success.
+ *
+ * Note that we assume success is the common case, and we optimize the
+ * branch fall-through case on success.
+ *
+ * On Entry:
+ * 	<aa>	register containing memory address
+ * 	<as>	register containing memory size
+ * 	<at>	temp register
+ * 	<error>	label to branch to on error; implies fall-through
+ * 		macro on success
+ * On Exit:
+ * 	<aa>	preserved
+ * 	<as>	preserved
+ * 	<at>	destroyed
+ */
+	.macro	verify_area	aa, as, at, sp, error
+	access_ok  \at, \aa, \as, \sp, \error
+	.endm
+
+
+#else /* __ASSEMBLY__ not defined */
+
+#include <linux/sched.h>
+#include <asm/types.h>
+
+/*
+ * The fs value determines whether argument validity checking should
+ * be performed or not.  If get_fs() == USER_DS, checking is
+ * performed, with get_fs() == KERNEL_DS, checking is bypassed.
+ *
+ * For historical reasons (Data Segment Register?), these macros are
+ * grossly misnamed.
+ */
+
+#define KERNEL_DS	((mm_segment_t) { 0 })
+#define USER_DS		((mm_segment_t) { 1 })
+
+#define get_ds()	(KERNEL_DS)
+#define get_fs()	(current->thread.current_ds)
+#define set_fs(val)	(current->thread.current_ds = (val))
+
+#define segment_eq(a,b)	((a).seg == (b).seg)
+
+#define __kernel_ok (segment_eq(get_fs(), KERNEL_DS))
+#define __user_ok(addr,size) (((size) <= TASK_SIZE)&&((addr) <= TASK_SIZE-(size)))
+#define __access_ok(addr,size) (__kernel_ok || __user_ok((addr),(size)))
+#define access_ok(type,addr,size) __access_ok((unsigned long)(addr),(size))
+
+extern inline int verify_area(int type, const void * addr, unsigned long size)
+{
+	return access_ok(type,addr,size) ? 0 : -EFAULT;
+}
+
+/*
+ * These are the main single-value transfer routines.  They
+ * automatically use the right size if we just have the right pointer
+ * type.
+ *
+ * This gets kind of ugly. We want to return _two_ values in
+ * "get_user()" and yet we don't want to do any pointers, because that
+ * is too much of a performance impact. Thus we have a few rather ugly
+ * macros here, and hide all the uglyness from the user.
+ *
+ * Careful to not
+ * (a) re-use the arguments for side effects (sizeof is ok)
+ * (b) require any knowledge of processes at this stage
+ */
+#define put_user(x,ptr)	__put_user_check((x),(ptr),sizeof(*(ptr)))
+#define get_user(x,ptr) __get_user_check((x),(ptr),sizeof(*(ptr)))
+
+/*
+ * The "__xxx" versions of the user access functions are versions that
+ * do not verify the address space, that must have been done previously
+ * with a separate "access_ok()" call (this is used when we do multiple
+ * accesses to the same area of user memory).
+ */
+#define __put_user(x,ptr) __put_user_nocheck((x),(ptr),sizeof(*(ptr)))
+#define __get_user(x,ptr) __get_user_nocheck((x),(ptr),sizeof(*(ptr)))
+
+
+extern long __put_user_bad(void);
+
+#define __put_user_nocheck(x,ptr,size)			\
+({							\
+	long __pu_err;					\
+	__put_user_size((x),(ptr),(size),__pu_err);	\
+	__pu_err;					\
+})
+
+#define __put_user_check(x,ptr,size)				\
+({								\
+	long __pu_err = -EFAULT;				\
+	__typeof__(*(ptr)) *__pu_addr = (ptr);			\
+	if (access_ok(VERIFY_WRITE,__pu_addr,size))		\
+		__put_user_size((x),__pu_addr,(size),__pu_err);	\
+	__pu_err;						\
+})
+
+#define __put_user_size(x,ptr,size,retval)			\
+do {								\
+	retval = 0;						\
+	switch (size) {						\
+        case 1: __put_user_asm(x,ptr,retval,1,"s8i");  break;	\
+        case 2: __put_user_asm(x,ptr,retval,2,"s16i"); break;   \
+        case 4: __put_user_asm(x,ptr,retval,4,"s32i"); break;   \
+        case 8: {						\
+		     __typeof__(*ptr) __v64 = x;		\
+		     retval = __copy_to_user(ptr,&__v64,8);	\
+		     break;					\
+	        }						\
+	default: __put_user_bad();				\
+	}							\
+} while (0)
+
+
+/*
+ * Consider a case of a user single load/store would cause both an
+ * unaligned exception and an MMU-related exception (unaligned
+ * exceptions happen first):
+ *
+ * User code passes a bad variable ptr to a system call.
+ * Kernel tries to access the variable.
+ * Unaligned exception occurs.
+ * Unaligned exception handler tries to make aligned accesses.
+ * Double exception occurs for MMU-related cause (e.g., page not mapped).
+ * do_page_fault() thinks the fault address belongs to the kernel, not the
+ * user, and panics.
+ *
+ * The kernel currently prohibits user unaligned accesses.  We use the
+ * __check_align_* macros to check for unaligned addresses before
+ * accessing user space so we don't crash the kernel.  Both
+ * __put_user_asm and __get_user_asm use these alignment macros, so
+ * macro-specific labels such as 0f, 1f, %0, %2, and %3 must stay in
+ * sync.
+ */
+
+#define __check_align_1  ""
+
+#define __check_align_2				\
+	"   _bbci.l %2,  0, 1f		\n"	\
+	"   movi    %0, %3		\n"	\
+	"   _j      2f			\n"
+
+#define __check_align_4				\
+	"   _bbsi.l %2,  0, 0f		\n"	\
+	"   _bbci.l %2,  1, 1f		\n"	\
+	"0: movi    %0, %3		\n"	\
+	"   _j      2f			\n"
+
+
+/*
+ * We don't tell gcc that we are accessing memory, but this is OK
+ * because we do not write to any memory gcc knows about, so there
+ * are no aliasing issues.
+ *
+ * WARNING: If you modify this macro at all, verify that the
+ * __check_align_* macros still work.
+ */
+#define __put_user_asm(x, addr, err, align, insn) \
+   __asm__ __volatile__(			\
+	__check_align_##align			\
+	"1: "insn"  %1, %2, 0		\n"	\
+	"2:				\n"	\
+	"   .section  .fixup,\"ax\"	\n"	\
+	"   .align 4			\n"	\
+	"4:				\n"	\
+	"   .long  2b			\n"	\
+	"5:				\n"	\
+	"   l32r   %2, 4b		\n"	\
+        "   movi   %0, %3		\n"	\
+        "   jx     %2			\n"	\
+	"   .previous			\n"	\
+	"   .section  __ex_table,\"a\"	\n"	\
+	"   .long	1b, 5b		\n"	\
+	"   .previous"				\
+	:"=r" (err)				\
+	:"r" ((int)(x)), "r" (addr), "i" (-EFAULT), "0" (err))
+
+#define __get_user_nocheck(x,ptr,size)				\
+({								\
+	long __gu_err, __gu_val;				\
+	__get_user_size(__gu_val,(ptr),(size),__gu_err);	\
+	(x) = (__typeof__(*(ptr)))__gu_val;			\
+	__gu_err;						\
+})
+
+#define __get_user_check(x,ptr,size)					\
+({									\
+	long __gu_err = -EFAULT, __gu_val = 0;				\
+	const __typeof__(*(ptr)) *__gu_addr = (ptr);			\
+	if (access_ok(VERIFY_READ,__gu_addr,size))			\
+		__get_user_size(__gu_val,__gu_addr,(size),__gu_err);	\
+	(x) = (__typeof__(*(ptr)))__gu_val;				\
+	__gu_err;							\
+})
+
+extern long __get_user_bad(void);
+
+#define __get_user_size(x,ptr,size,retval)				\
+do {									\
+	retval = 0;							\
+        switch (size) {							\
+          case 1: __get_user_asm(x,ptr,retval,1,"l8ui");  break;	\
+          case 2: __get_user_asm(x,ptr,retval,2,"l16ui"); break;	\
+          case 4: __get_user_asm(x,ptr,retval,4,"l32i");  break;	\
+          case 8: retval = __copy_from_user(&x,ptr,8);    break;	\
+          default: (x) = __get_user_bad();				\
+        }								\
+} while (0)
+
+
+/*
+ * WARNING: If you modify this macro at all, verify that the
+ * __check_align_* macros still work.
+ */
+#define __get_user_asm(x, addr, err, align, insn) \
+   __asm__ __volatile__(			\
+	__check_align_##align			\
+	"1: "insn"  %1, %2, 0		\n"	\
+	"2:				\n"	\
+	"   .section  .fixup,\"ax\"	\n"	\
+	"   .align 4			\n"	\
+	"4:				\n"	\
+	"   .long  2b			\n"	\
+	"5:				\n"	\
+	"   l32r   %2, 4b		\n"	\
+	"   movi   %1, 0		\n"	\
+        "   movi   %0, %3		\n"	\
+        "   jx     %2			\n"	\
+	"   .previous			\n"	\
+	"   .section  __ex_table,\"a\"	\n"	\
+	"   .long	1b, 5b		\n"	\
+	"   .previous"				\
+	:"=r" (err), "=r" (x)			\
+	:"r" (addr), "i" (-EFAULT), "0" (err))
+
+
+/*
+ * Copy to/from user space
+ */
+
+/*
+ * We use a generic, arbitrary-sized copy subroutine.  The Xtensa
+ * architecture would cause heavy code bloat if we tried to inline
+ * these functions and provide __constant_copy_* equivalents like the
+ * i386 versions.  __xtensa_copy_user is quite efficient.  See the
+ * .fixup section of __xtensa_copy_user for a discussion on the
+ * X_zeroing equivalents for Xtensa.
+ */
+
+extern unsigned __xtensa_copy_user(void *to, const void *from, unsigned n);
+#define __copy_user(to,from,size) __xtensa_copy_user(to,from,size)
+
+
+static inline unsigned long
+__generic_copy_from_user_nocheck(void *to, const void *from, unsigned long n)
+{
+	return __copy_user(to,from,n);
+}
+
+static inline unsigned long
+__generic_copy_to_user_nocheck(void *to, const void *from, unsigned long n)
+{
+	return __copy_user(to,from,n);
+}
+
+static inline unsigned long
+__generic_copy_to_user(void *to, const void *from, unsigned long n)
+{
+	prefetch(from);
+	if (access_ok(VERIFY_WRITE, to, n))
+		return __copy_user(to,from,n);
+	return n;
+}
+
+static inline unsigned long
+__generic_copy_from_user(void *to, const void *from, unsigned long n)
+{
+	prefetchw(to);
+	if (access_ok(VERIFY_READ, from, n))
+		return __copy_user(to,from,n);
+	else
+		memset(to, 0, n);
+	return n;
+}
+
+#define copy_to_user(to,from,n) __generic_copy_to_user((to),(from),(n))
+#define copy_from_user(to,from,n) __generic_copy_from_user((to),(from),(n))
+#define __copy_to_user(to,from,n) __generic_copy_to_user_nocheck((to),(from),(n))
+#define __copy_from_user(to,from,n) __generic_copy_from_user_nocheck((to),(from),(n))
+#define __copy_to_user_inatomic __copy_to_user
+#define __copy_from_user_inatomic __copy_from_user
+
+
+/*
+ * We need to return the number of bytes not cleared.  Our memset()
+ * returns zero if a problem occurs while accessing user-space memory.
+ * In that event, return no memory cleared.  Otherwise, zero for
+ * success.
+ */
+
+extern inline unsigned long
+__xtensa_clear_user(void *addr, unsigned long size)
+{
+	if ( ! memset(addr, 0, size) )
+		return size;
+	return 0;
+}
+
+extern inline unsigned long
+clear_user(void *addr, unsigned long size)
+{
+	if (access_ok(VERIFY_WRITE, addr, size))
+		return __xtensa_clear_user(addr, size);
+	return size ? -EFAULT : 0;
+}
+
+#define __clear_user  __xtensa_clear_user
+
+
+extern long __strncpy_user(char *, const char *, long);
+#define __strncpy_from_user __strncpy_user
+
+extern inline long
+strncpy_from_user(char *dst, const char *src, long count)
+{
+	if (access_ok(VERIFY_READ, src, 1))
+		return __strncpy_from_user(dst, src, count);
+	return -EFAULT;
+}
+
+
+#define strlen_user(str) strnlen_user((str), TASK_SIZE - 1)
+
+/*
+ * Return the size of a string (including the ending 0!)
+ */
+extern long __strnlen_user(const char *, long);
+
+extern inline long strnlen_user(const char *str, long len)
+{
+	unsigned long top = __kernel_ok ? ~0UL : TASK_SIZE - 1;
+
+	if ((unsigned long)str > top)
+		return 0;
+	return __strnlen_user(str, len);
+}
+
+
+struct exception_table_entry
+{
+	unsigned long insn, fixup;
+};
+
+/* Returns 0 if exception not found and fixup.unit otherwise.  */
+
+extern unsigned long search_exception_table(unsigned long addr);
+extern void sort_exception_table(void);
+
+/* Returns the new pc */
+#define fixup_exception(map_reg, fixup_unit, pc)                \
+({                                                              \
+	fixup_unit;                                             \
+})
+
+#endif	/* __ASSEMBLY__ */
+#endif	/* _XTENSA_UACCESS_H */
diff --git a/include/asm-xtensa/ucontext.h b/include/asm-xtensa/ucontext.h
new file mode 100644
index 00000000000..94c94ed3e00
--- /dev/null
+++ b/include/asm-xtensa/ucontext.h
@@ -0,0 +1,22 @@
+/*
+ * include/asm-xtensa/ucontext.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_UCONTEXT_H
+#define _XTENSA_UCONTEXT_H
+
+struct ucontext {
+	unsigned long	  uc_flags;
+	struct ucontext  *uc_link;
+	stack_t		  uc_stack;
+	struct sigcontext uc_mcontext;
+	sigset_t	  uc_sigmask;	/* mask last for extensibility */
+};
+
+#endif /* _XTENSA_UCONTEXT_H */
diff --git a/include/asm-xtensa/unaligned.h b/include/asm-xtensa/unaligned.h
new file mode 100644
index 00000000000..28220890d0a
--- /dev/null
+++ b/include/asm-xtensa/unaligned.h
@@ -0,0 +1,28 @@
+/*
+ * include/asm-xtensa/unaligned.h
+ *
+ * Xtensa doesn't handle unaligned accesses efficiently.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_UNALIGNED_H
+#define _XTENSA_UNALIGNED_H
+
+#include <linux/string.h>
+
+/* Use memmove here, so gcc does not insert a __builtin_memcpy. */
+
+#define get_unaligned(ptr) \
+  ({ __typeof__(*(ptr)) __tmp; memmove(&__tmp, (ptr), sizeof(*(ptr))); __tmp; })
+
+#define put_unaligned(val, ptr)				\
+  ({ __typeof__(*(ptr)) __tmp = (val);			\
+     memmove((ptr), &__tmp, sizeof(*(ptr)));		\
+     (void)0; })
+
+#endif	/* _XTENSA_UNALIGNED_H */
diff --git a/include/asm-xtensa/unistd.h b/include/asm-xtensa/unistd.h
new file mode 100644
index 00000000000..64c64dd83ba
--- /dev/null
+++ b/include/asm-xtensa/unistd.h
@@ -0,0 +1,537 @@
+/*
+ * include/asm-xtensa/unistd.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_UNISTD_H
+#define _XTENSA_UNISTD_H
+
+#include <linux/linkage.h>
+
+//#define __NR_setup		  0 /* used only by init, to get system going */
+#define __NR_spill		  0
+#define __NR_exit		  1
+#define __NR_fork		  2
+#define __NR_read		  3
+#define __NR_write		  4
+#define __NR_open		  5
+#define __NR_close		  6
+#define __NR_waitpid		  7
+#define __NR_creat		  8
+#define __NR_link		  9
+#define __NR_unlink		 10
+#define __NR_execve		 11
+#define __NR_chdir		 12
+#define __NR_time		 13
+#define __NR_mknod		 14
+#define __NR_chmod		 15
+#define __NR_lchown		 16
+#define __NR_break		 17
+#define __NR_oldstat		 18
+#define __NR_lseek		 19
+#define __NR_getpid		 20
+#define __NR_mount		 21
+#define __NR_oldumount		 22
+#define __NR_setuid		 23
+#define __NR_getuid		 24
+#define __NR_stime		 25
+#define __NR_ptrace		 26
+#define __NR_alarm		 27
+#define __NR_oldfstat		 28
+#define __NR_pause		 29
+#define __NR_utime		 30
+#define __NR_stty		 31
+#define __NR_gtty		 32
+#define __NR_access		 33
+#define __NR_nice		 34
+#define __NR_ftime		 35
+#define __NR_sync		 36
+#define __NR_kill		 37
+#define __NR_rename		 38
+#define __NR_mkdir		 39
+#define __NR_rmdir		 40
+#define __NR_dup		 41
+#define __NR_pipe		 42
+#define __NR_times		 43
+#define __NR_prof		 44
+#define __NR_brk		 45
+#define __NR_setgid		 46
+#define __NR_getgid		 47
+#define __NR_signal		 48
+#define __NR_geteuid		 49
+#define __NR_getegid		 50
+#define __NR_acct		 51
+#define __NR_umount		 52
+#define __NR_lock		 53
+#define __NR_ioctl		 54
+#define __NR_fcntl		 55
+#define __NR_mpx		 56
+#define __NR_setpgid		 57
+#define __NR_ulimit		 58
+#define __NR_oldolduname	 59
+#define __NR_umask		 60
+#define __NR_chroot		 61
+#define __NR_ustat		 62
+#define __NR_dup2		 63
+#define __NR_getppid		 64
+#define __NR_getpgrp		 65
+#define __NR_setsid		 66
+#define __NR_sigaction		 67
+#define __NR_sgetmask		 68
+#define __NR_ssetmask		 69
+#define __NR_setreuid		 70
+#define __NR_setregid		 71
+#define __NR_sigsuspend		 72
+#define __NR_sigpending		 73
+#define __NR_sethostname	 74
+#define __NR_setrlimit		 75
+#define __NR_getrlimit	 	 76	/* Back compatible 2Gig limited rlimit */
+#define __NR_getrusage		 77
+#define __NR_gettimeofday	 78
+#define __NR_settimeofday	 79
+#define __NR_getgroups		 80
+#define __NR_setgroups		 81
+#define __NR_select		 82
+#define __NR_symlink		 83
+#define __NR_oldlstat		 84
+#define __NR_readlink		 85
+#define __NR_uselib		 86
+#define __NR_swapon		 87
+#define __NR_reboot		 88
+#define __NR_readdir		 89
+#define __NR_mmap		 90
+#define __NR_munmap		 91
+#define __NR_truncate		 92
+#define __NR_ftruncate		 93
+#define __NR_fchmod		 94
+#define __NR_fchown		 95
+#define __NR_getpriority	 96
+#define __NR_setpriority	 97
+#define __NR_profil		 98
+#define __NR_statfs		 99
+#define __NR_fstatfs		100
+#define __NR_ioperm		101
+#define __NR_socketcall		102
+#define __NR_syslog		103
+#define __NR_setitimer		104
+#define __NR_getitimer		105
+#define __NR_stat		106
+#define __NR_lstat		107
+#define __NR_fstat		108
+#define __NR_olduname		109
+#define __NR_iopl		110
+#define __NR_vhangup		111
+#define __NR_idle		112
+#define __NR_vm86		113
+#define __NR_wait4		114
+#define __NR_swapoff		115
+#define __NR_sysinfo		116
+#define __NR_ipc		117
+#define __NR_fsync		118
+#define __NR_sigreturn		119
+#define __NR_clone		120
+#define __NR_setdomainname	121
+#define __NR_uname		122
+#define __NR_modify_ldt		123
+#define __NR_adjtimex		124
+#define __NR_mprotect		125
+#define __NR_sigprocmask	126
+#define __NR_create_module	127
+#define __NR_init_module	128
+#define __NR_delete_module	129
+#define __NR_get_kernel_syms	130
+#define __NR_quotactl		131
+#define __NR_getpgid		132
+#define __NR_fchdir		133
+#define __NR_bdflush		134
+#define __NR_sysfs		135
+#define __NR_personality	136
+#define __NR_afs_syscall	137 /* Syscall for Andrew File System */
+#define __NR_setfsuid		138
+#define __NR_setfsgid		139
+#define __NR__llseek		140
+#define __NR_getdents		141
+#define __NR__newselect		142
+#define __NR_flock		143
+#define __NR_msync		144
+#define __NR_readv		145
+#define __NR_writev		146
+#define __NR_cacheflush         147
+#define __NR_cachectl           148
+#define __NR_sysxtensa          149
+#define __NR_sysdummy           150
+#define __NR_getsid		151
+#define __NR_fdatasync		152
+#define __NR__sysctl		153
+#define __NR_mlock		154
+#define __NR_munlock		155
+#define __NR_mlockall		156
+#define __NR_munlockall		157
+#define __NR_sched_setparam		158
+#define __NR_sched_getparam		159
+#define __NR_sched_setscheduler		160
+#define __NR_sched_getscheduler		161
+#define __NR_sched_yield		162
+#define __NR_sched_get_priority_max	163
+#define __NR_sched_get_priority_min	164
+#define __NR_sched_rr_get_interval	165
+#define __NR_nanosleep		166
+#define __NR_mremap		167
+#define __NR_accept             168
+#define __NR_bind               169
+#define __NR_connect            170
+#define __NR_getpeername        171
+#define __NR_getsockname        172
+#define __NR_getsockopt         173
+#define __NR_listen             174
+#define __NR_recv               175
+#define __NR_recvfrom           176
+#define __NR_recvmsg            177
+#define __NR_send               178
+#define __NR_sendmsg            179
+#define __NR_sendto             180
+#define __NR_setsockopt         181
+#define __NR_shutdown           182
+#define __NR_socket             183
+#define __NR_socketpair         184
+#define __NR_setresuid		185
+#define __NR_getresuid		186
+#define __NR_query_module	187
+#define __NR_poll		188
+#define __NR_nfsservctl		189
+#define __NR_setresgid		190
+#define __NR_getresgid		191
+#define __NR_prctl              192
+#define __NR_rt_sigreturn	193
+#define __NR_rt_sigaction	194
+#define __NR_rt_sigprocmask	195
+#define __NR_rt_sigpending	196
+#define __NR_rt_sigtimedwait	197
+#define __NR_rt_sigqueueinfo	198
+#define __NR_rt_sigsuspend	199
+#define __NR_pread		200
+#define __NR_pwrite		201
+#define __NR_chown		202
+#define __NR_getcwd		203
+#define __NR_capget		204
+#define __NR_capset		205
+#define __NR_sigaltstack	206
+#define __NR_sendfile		207
+#define __NR_streams1		208	/* some people actually want it */
+#define __NR_streams2		209	/* some people actually want it */
+#define __NR_mmap2		210
+#define __NR_truncate64		211
+#define __NR_ftruncate64	212
+#define __NR_stat64		213
+#define __NR_lstat64		214
+#define __NR_fstat64		215
+#define __NR_pivot_root		216
+#define __NR_mincore		217
+#define __NR_madvise		218
+#define __NR_getdents64		219
+#define __NR_vfork		220
+
+/* Keep this last; should always equal the last valid call number. */
+#define __NR_Linux_syscalls     220
+
+/* user-visible error numbers are in the range -1 - -125: see
+ * <asm-xtensa/errno.h> */
+
+#define SYSXTENSA_RESERVED	   0	/* don't use this */
+#define SYSXTENSA_ATOMIC_SET	   1	/* set variable */
+#define SYSXTENSA_ATOMIC_EXG_ADD   2	/* exchange memory and add */
+#define SYSXTENSA_ATOMIC_ADD	   3	/* add to memory */
+#define SYSXTENSA_ATOMIC_CMP_SWP   4	/* compare and swap */
+
+#define SYSXTENSA_COUNT		   5	/* count of syscall0 functions*/
+
+#ifdef __KERNEL__
+#define __syscall_return(type, res) return ((type)(res))
+#else
+#define __syscall_return(type, res) \
+do { \
+	if ((unsigned long)(res) >= (unsigned long)(-125)) { \
+	/* Avoid using "res" which is declared to be in register r2; \
+	 * errno might expand to a function call and clobber it.  */ \
+		int __err = -(res); \
+		errno = __err; \
+		res = -1; \
+	} \
+	return (type) (res); \
+} while (0)
+#endif
+
+
+/* Tensilica's xt-xcc compiler is much more agressive at code
+ * optimization than gcc.  Multiple __asm__ statements are
+ * insufficient for xt-xcc because subsequent optimization passes
+ * (beyond the front-end that knows of __asm__ statements and other
+ * such GNU Extensions to C) can modify the register selection for
+ * containment of C variables.
+ *
+ * xt-xcc cannot modify the contents of a single __asm__ statement, so
+ * we create single-asm versions of the syscall macros that are
+ * suitable and optimal for both xt-xcc and gcc.
+ *
+ * Linux takes system-call arguments in registers.  The following
+ * design is optimized for user-land apps (e.g., glibc) which
+ * typically have a function wrapper around the "syscall" assembly
+ * instruction.  It satisfies the Xtensa ABI while minizing argument
+ * shifting.
+ *
+ * The Xtensa ABI and software conventions require the system-call
+ * number in a2.  If an argument exists in a2, we move it to the next
+ * available register.  Note that for improved efficiency, we do NOT
+ * shift all parameters down one register to maintain the original
+ * order.
+ *
+ * At best case (zero arguments), we just write the syscall number to
+ * a2.  At worst case (1 to 6 arguments), we move the argument in a2
+ * to the next available register, then write the syscall number to
+ * a2.
+ *
+ * For clarity, the following truth table enumerates all possibilities.
+ *
+ * arguments	syscall number	arg0, arg1, arg2, arg3, arg4, arg5
+ * ---------	--------------	----------------------------------
+ *	0	      a2
+ *	1	      a2	a3
+ *	2	      a2	a4,   a3
+ *	3	      a2	a5,   a3,   a4
+ *	4	      a2	a6,   a3,   a4,   a5
+ *	5	      a2	a7,   a3,   a4,   a5,   a6
+ *	6	      a2	a8,   a3,   a4,   a5,   a6,   a7
+ */
+
+#define _syscall0(type,name) \
+type name(void) \
+{ \
+long __res; \
+__asm__ __volatile__ ( \
+	"  movi  a2, %1 \n" \
+	"  syscall      \n" \
+	"  mov   %0, a2 \n" \
+	: "=a" (__res) \
+	: "i" (__NR_##name) \
+	: "a2" \
+	); \
+__syscall_return(type,__res); \
+}
+
+#define _syscall1(type,name,type0,arg0) \
+type name(type0 arg0) \
+{ \
+long __res; \
+__asm__ __volatile__ ( \
+	"  mov   a3, %2 \n" \
+	"  movi  a2, %1 \n" \
+	"  syscall      \n" \
+	"  mov   %0, a2 \n" \
+	: "=a" (__res) \
+	: "i" (__NR_##name), "a" (arg0) \
+	: "a2", "a3" \
+	); \
+__syscall_return(type,__res); \
+}
+
+#define _syscall2(type,name,type0,arg0,type1,arg1) \
+type name(type0 arg0,type1 arg1) \
+{ \
+long __res; \
+__asm__ __volatile__ ( \
+	"  mov   a4, %2 \n" \
+	"  mov   a3, %3 \n" \
+	"  movi  a2, %1 \n" \
+	"  syscall      \n" \
+	"  mov   %0, a2 \n" \
+	: "=a" (__res) \
+	: "i" (__NR_##name), "a" (arg0), "a" (arg1) \
+	: "a2", "a3", "a4" \
+	); \
+__syscall_return(type,__res); \
+}
+
+#define _syscall3(type,name,type0,arg0,type1,arg1,type2,arg2) \
+type name(type0 arg0,type1 arg1,type2 arg2) \
+{ \
+long __res; \
+__asm__ __volatile__ ( \
+	"  mov   a5, %2 \n" \
+	"  mov   a4, %4 \n" \
+	"  mov   a3, %3 \n" \
+	"  movi  a2, %1 \n" \
+	"  syscall      \n" \
+	"  mov   %0, a2 \n" \
+	: "=a" (__res) \
+	: "i" (__NR_##name), "a" (arg0), "a" (arg1), "a" (arg2) \
+	: "a2", "a3", "a4", "a5" \
+	); \
+__syscall_return(type,__res); \
+}
+
+#define _syscall4(type,name,type0,arg0,type1,arg1,type2,arg2,type3,arg3) \
+type name(type0 arg0,type1 arg1,type2 arg2,type3 arg3) \
+{ \
+long __res; \
+__asm__ __volatile__ ( \
+	"  mov   a6, %2 \n" \
+	"  mov   a5, %5 \n" \
+	"  mov   a4, %4 \n" \
+	"  mov   a3, %3 \n" \
+	"  movi  a2, %1 \n" \
+	"  syscall      \n" \
+	"  mov   %0, a2 \n" \
+	: "=a" (__res) \
+	: "i" (__NR_##name), "a" (arg0), "a" (arg1), "a" (arg2), "a" (arg3) \
+	: "a2", "a3", "a4", "a5", "a6" \
+	); \
+__syscall_return(type,__res); \
+}
+
+/* Note that we save and restore the a7 frame pointer.
+ * Including a7 in the clobber list doesn't do what you'd expect.
+ */
+#define _syscall5(type,name,type0,arg0,type1,arg1,type2,arg2,type3,arg3,type4,arg4) \
+type name(type0 arg0,type1 arg1,type2 arg2,type3 arg3,type4 arg4) \
+{ \
+long __res; \
+__asm__ __volatile__ ( \
+	"  mov   a9, a7 \n" \
+	"  mov   a7, %2 \n" \
+	"  mov   a6, %6 \n" \
+	"  mov   a5, %5 \n" \
+	"  mov   a4, %4 \n" \
+	"  mov   a3, %3 \n" \
+	"  movi  a2, %1 \n" \
+	"  syscall      \n" \
+	"  mov   a7, a9 \n" \
+	"  mov   %0, a2 \n" \
+	: "=a" (__res) \
+	: "i" (__NR_##name), "a" (arg0), "a" (arg1), "a" (arg2), \
+                             "a" (arg3), "a" (arg4) \
+	: "a2", "a3", "a4", "a5", "a6", "a9" \
+	); \
+__syscall_return(type,__res); \
+}
+
+/* Note that we save and restore the a7 frame pointer.
+ * Including a7 in the clobber list doesn't do what you'd expect.
+ */
+#define _syscall6(type,name,type0,arg0,type1,arg1,type2,arg2,type3,arg3,type4,arg4,type5,arg5) \
+type name(type0 arg0,type1 arg1,type2 arg2,type3 arg3,type4 arg4,type5 arg5) \
+{ \
+long __res; \
+__asm__ __volatile__ ( \
+	"  mov   a9, a7 \n" \
+	"  mov   a8, %2 \n" \
+	"  mov   a7, %7 \n" \
+	"  mov   a6, %6 \n" \
+	"  mov   a5, %5 \n" \
+	"  mov   a4, %4 \n" \
+	"  mov   a3, %3 \n" \
+	"  movi  a2, %1 \n" \
+	"  syscall      \n" \
+	"  mov   a7, a9 \n" \
+	"  mov   %0, a2 \n" \
+	: "=a" (__res) \
+	: "i" (__NR_##name), "a" (arg0), "a" (arg1), "a" (arg2), \
+                             "a" (arg3), "a" (arg4), "a" (arg5)  \
+	: "a2", "a3", "a4", "a5", "a6", "a8", "a9" \
+	); \
+__syscall_return(type,__res); \
+}
+
+
+#ifdef __KERNEL_SYSCALLS__
+
+#include <linux/compiler.h>
+#include <linux/types.h>
+#include <linux/syscalls.h>
+
+/*
+ * we need this inline - forking from kernel space will result
+ * in NO COPY ON WRITE (!!!), until an execve is executed. This
+ * is no problem, but for the stack. This is handled by not letting
+ * main() use the stack at all after fork(). Thus, no function
+ * calls - which means inline code for fork too, as otherwise we
+ * would use the stack upon exit from 'fork()'.
+ *
+ * Actually only pause and fork are needed inline, so that there
+ * won't be any messing with the stack from main(), but we define
+ * some others too.
+ */
+
+#define __NR__exit __NR_exit
+
+static __inline__ _syscall0(int,pause)
+//static __inline__ _syscall1(int,setup,int,magic) FIXME
+static __inline__ _syscall0(int,sync)
+static __inline__ _syscall0(pid_t,setsid)
+static __inline__ _syscall3(int,write,int,fd,const char *,buf,off_t,count)
+static __inline__ _syscall3(int,read,int,fd,char *,buf,off_t,count)
+static __inline__ _syscall3(off_t,lseek,int,fd,off_t,offset,int,count)
+static __inline__ _syscall1(int,dup,int,fd)
+static __inline__ _syscall3(int,execve,const char*,file,char**,argv,char**,envp)
+static __inline__ _syscall3(int,open,const char *,file,int,flag,int,mode)
+static __inline__ _syscall1(int,close,int,fd)
+static __inline__ _syscall1(int,_exit,int,exitcode)
+static __inline__ _syscall3(pid_t,waitpid,pid_t,pid,int *,wait_stat,int,options)
+static __inline__ _syscall1(int,delete_module,const char *,name)
+
+struct stat;
+static __inline__ _syscall2(int,fstat,int,fd,struct stat *,buf)
+static __inline__ _syscall0(pid_t,getpid)
+static __inline__ _syscall2(int,kill,int,pid,int,sig)
+static __inline__ _syscall2(int,stat,const char *, path,struct stat *,buf)
+static __inline__ _syscall1(int,unlink,char *,pathname)
+
+
+
+extern pid_t waitpid(int, int*, int );
+static __inline__ pid_t wait(int * wait_stat)
+{
+	return waitpid(-1,wait_stat,0);
+}
+#endif
+
+/*
+ * "Conditional" syscalls
+ *
+ * What we want is __attribute__((weak,alias("sys_ni_syscall"))),
+ * but it doesn't work on all toolchains, so we just do it by hand
+ */
+#define cond_syscall(x) asm(".weak\t" #x "\n\t.set\t" #x ",sys_ni_syscall");
+
+#ifdef __KERNEL__
+#define __ARCH_WANT_IPC_PARSE_VERSION
+#define __ARCH_WANT_OLD_READDIR
+#define __ARCH_WANT_OLD_STAT
+#define __ARCH_WANT_STAT64
+#define __ARCH_WANT_SYS_ALARM
+#define __ARCH_WANT_SYS_GETHOSTNAME
+#define __ARCH_WANT_SYS_PAUSE
+#define __ARCH_WANT_SYS_SGETMASK
+#define __ARCH_WANT_SYS_SIGNAL
+#define __ARCH_WANT_SYS_TIME
+#define __ARCH_WANT_SYS_UTIME
+#define __ARCH_WANT_SYS_WAITPID
+#define __ARCH_WANT_SYS_SOCKETCALL
+#define __ARCH_WANT_SYS_FADVISE64
+#define __ARCH_WANT_SYS_GETPGRP
+#define __ARCH_WANT_SYS_LLSEEK
+#define __ARCH_WANT_SYS_NICE
+#define __ARCH_WANT_SYS_OLD_GETRLIMIT
+#define __ARCH_WANT_SYS_OLDUMOUNT
+#define __ARCH_WANT_SYS_SIGPENDING
+#define __ARCH_WANT_SYS_SIGPROCMASK
+#define __ARCH_WANT_SYS_RT_SIGACTION
+#endif
+
+
+
+#endif	/* _XTENSA_UNISTD_H */
diff --git a/include/asm-xtensa/user.h b/include/asm-xtensa/user.h
new file mode 100644
index 00000000000..2c3ed23354a
--- /dev/null
+++ b/include/asm-xtensa/user.h
@@ -0,0 +1,20 @@
+/*
+ * include/asm-xtensa/user.h
+ *
+ * Xtensa Processor version.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_USER_H
+#define _XTENSA_USER_H
+
+/* This file usually defines a 'struct user' structure. However, it it only
+ * used for a.out file, which are not supported on Xtensa.
+ */
+
+#endif	/* _XTENSA_USER_H */
diff --git a/include/asm-xtensa/vga.h b/include/asm-xtensa/vga.h
new file mode 100644
index 00000000000..23d82f6acb5
--- /dev/null
+++ b/include/asm-xtensa/vga.h
@@ -0,0 +1,19 @@
+/*
+ * include/asm-xtensa/vga.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_VGA_H
+#define _XTENSA_VGA_H
+
+#define VGA_MAP_MEM(x) (unsigned long)phys_to_virt(x)
+
+#define vga_readb(x)	(*(x))
+#define vga_writeb(x,y)	(*(y) = (x))
+
+#endif
diff --git a/include/asm-xtensa/xor.h b/include/asm-xtensa/xor.h
new file mode 100644
index 00000000000..e7b1f083991
--- /dev/null
+++ b/include/asm-xtensa/xor.h
@@ -0,0 +1,16 @@
+/*
+ * include/asm-xtensa/xor.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_XOR_H
+#define _XTENSA_XOR_H
+
+#include <asm-generic/xor.h>
+
+#endif
-- 
cgit v1.2.3-70-g09d2


From e344b63eeec7850b5e900e10c8a6c61d083fd3a4 Mon Sep 17 00:00:00 2001
From: Chris Zankel <czankel@tensilica.com>
Date: Thu, 23 Jun 2005 22:01:30 -0700
Subject: [PATCH] xtensa: Architecture support for Tensilica Xtensa Part 7

The attached patches provides part 7 of an architecture implementation for the
Tensilica Xtensa CPU series.

Signed-off-by: Chris Zankel <chris@zankel.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-xtensa/xtensa/cacheasm.h               |  708 +++++++++++
 include/asm-xtensa/xtensa/cacheattrasm.h           |  432 +++++++
 include/asm-xtensa/xtensa/config-linux_be/core.h   | 1270 ++++++++++++++++++++
 include/asm-xtensa/xtensa/config-linux_be/defs.h   |  270 +++++
 .../asm-xtensa/xtensa/config-linux_be/specreg.h    |   99 ++
 include/asm-xtensa/xtensa/config-linux_be/system.h |  198 +++
 include/asm-xtensa/xtensa/config-linux_be/tie.h    |  275 +++++
 include/asm-xtensa/xtensa/coreasm.h                |  526 ++++++++
 include/asm-xtensa/xtensa/corebits.h               |   77 ++
 include/asm-xtensa/xtensa/hal.h                    |  822 +++++++++++++
 include/asm-xtensa/xtensa/simcall.h                |  130 ++
 include/asm-xtensa/xtensa/xt2000-uart.h            |  155 +++
 include/asm-xtensa/xtensa/xt2000.h                 |  408 +++++++
 include/asm-xtensa/xtensa/xtboard.h                |  120 ++
 14 files changed, 5490 insertions(+)
 create mode 100644 include/asm-xtensa/xtensa/cacheasm.h
 create mode 100644 include/asm-xtensa/xtensa/cacheattrasm.h
 create mode 100644 include/asm-xtensa/xtensa/config-linux_be/core.h
 create mode 100644 include/asm-xtensa/xtensa/config-linux_be/defs.h
 create mode 100644 include/asm-xtensa/xtensa/config-linux_be/specreg.h
 create mode 100644 include/asm-xtensa/xtensa/config-linux_be/system.h
 create mode 100644 include/asm-xtensa/xtensa/config-linux_be/tie.h
 create mode 100644 include/asm-xtensa/xtensa/coreasm.h
 create mode 100644 include/asm-xtensa/xtensa/corebits.h
 create mode 100644 include/asm-xtensa/xtensa/hal.h
 create mode 100644 include/asm-xtensa/xtensa/simcall.h
 create mode 100644 include/asm-xtensa/xtensa/xt2000-uart.h
 create mode 100644 include/asm-xtensa/xtensa/xt2000.h
 create mode 100644 include/asm-xtensa/xtensa/xtboard.h

(limited to 'include')

diff --git a/include/asm-xtensa/xtensa/cacheasm.h b/include/asm-xtensa/xtensa/cacheasm.h
new file mode 100644
index 00000000000..0cdbb0bf180
--- /dev/null
+++ b/include/asm-xtensa/xtensa/cacheasm.h
@@ -0,0 +1,708 @@
+#ifndef XTENSA_CACHEASM_H
+#define XTENSA_CACHEASM_H
+
+/*
+ * THIS FILE IS GENERATED -- DO NOT MODIFY BY HAND
+ *
+ * include/asm-xtensa/xtensa/cacheasm.h -- assembler-specific cache
+ * related definitions that depend on CORE configuration.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2002 Tensilica Inc.
+ */
+
+
+#include <xtensa/coreasm.h>
+
+
+/*
+ *  This header file defines assembler macros of the form:
+ *	<x>cache_<func>
+ *  where <x> is 'i' or 'd' for instruction and data caches,
+ *  and <func> indicates the function of the macro.
+ *
+ *  The following functions <func> are defined,
+ *  and apply only to the specified cache (I or D):
+ *
+ *  reset
+ *	Resets the cache.
+ *
+ *  sync
+ *	Makes sure any previous cache instructions have been completed;
+ *	ie. makes sure any previous cache control operations
+ *	have had full effect and been synchronized to memory.
+ *	Eg. any invalidate completed [so as not to generate a hit],
+ *	any writebacks or other pipelined writes written to memory, etc.
+ *
+ *  invalidate_line		(single cache line)
+ *  invalidate_region		(specified memory range)
+ *  invalidate_all		(entire cache)
+ *	Invalidates all cache entries that cache
+ *	data from the specified memory range.
+ *	NOTE: locked entries are not invalidated.
+ *
+ *  writeback_line		(single cache line)
+ *  writeback_region		(specified memory range)
+ *  writeback_all		(entire cache)
+ *	Writes back to memory all dirty cache entries
+ *	that cache data from the specified memory range,
+ *	and marks these entries as clean.
+ *	NOTE: on some future implementations, this might
+ *		also invalidate.
+ *	NOTE: locked entries are written back, but never invalidated.
+ *	NOTE: instruction caches never implement writeback.
+ *
+ *  writeback_inv_line		(single cache line)
+ *  writeback_inv_region	(specified memory range)
+ *  writeback_inv_all		(entire cache)
+ *	Writes back to memory all dirty cache entries
+ *	that cache data from the specified memory range,
+ *	and invalidates these entries (including all clean
+ *	cache entries that cache data from that range).
+ *	NOTE: locked entries are written back but not invalidated.
+ *	NOTE: instruction caches never implement writeback.
+ *
+ *  lock_line			(single cache line)
+ *  lock_region			(specified memory range)
+ *	Prefetch and lock the specified memory range into cache.
+ *	NOTE:  if any part of the specified memory range cannot
+ *	be locked, a ??? exception occurs.  These macros don't
+ *	do anything special (yet anyway) to handle this situation.
+ *
+ *  unlock_line			(single cache line)
+ *  unlock_region		(specified memory range)
+ *  unlock_all			(entire cache)
+ *	Unlock cache entries that cache the specified memory range.
+ *	Entries not already locked are unaffected.
+ */
+
+
+
+/***************************   GENERIC -- ALL CACHES   ***************************/
+
+
+/*
+ *  The following macros assume the following cache size/parameter limits
+ *  in the current Xtensa core implementation:
+ *	cache size:	1024 bytes minimum
+ *	line size:	16 - 64 bytes
+ *	way count:	1 - 4
+ *
+ *  Minimum entries per way (ie. per associativity) = 1024 / 64 / 4 = 4
+ *  Hence the assumption that each loop can execute four cache instructions.
+ *
+ *  Correspondingly, the offset range of instructions is assumed able to cover
+ *  four lines, ie. offsets {0,1,2,3} * line_size are assumed valid for
+ *  both hit and indexed cache instructions.  Ie. these offsets are all
+ *  valid:  0, 16, 32, 48, 64, 96, 128, 192 (for line sizes 16, 32, 64).
+ *  This is true of all original cache instructions
+ *  (dhi, ihi, dhwb, dhwbi, dii, iii) which have offsets
+ *  of 0 to 1020 in multiples of 4 (ie. 8 bits shifted by 2).
+ *  This is also true of subsequent cache instructions
+ *  (dhu, ihu, diu, iiu, diwb, diwbi, dpfl, ipfl) which have offsets
+ *  of 0 to 240 in multiples of 16 (ie. 4 bits shifted by 4).
+ *
+ *  (Maximum cache size, currently 32k, doesn't affect the following macros.
+ *  Cache ways > MMU min page size cause aliasing but that's another matter.)
+ */
+
+
+
+/*
+ *  Macro to apply an 'indexed' cache instruction to the entire cache.
+ *
+ *  Parameters:
+ *	cainst		instruction/ that takes an address register parameter
+ *			and an offset parameter (in range 0 .. 3*linesize).
+ *	size		size of cache in bytes
+ *	linesize	size of cache line in bytes
+ *	assoc_or1	number of associativities (ways/sets) in cache
+ *			if all sets affected by cainst,
+ *			or 1 if only one set (or not all sets) of the cache
+ *			is affected by cainst (eg. DIWB or DIWBI [not yet ISA defined]).
+ *	aa, ab		unique address registers (temporaries)
+ */
+
+	.macro	cache_index_all		cainst, size, linesize, assoc_or1, aa, ab
+
+	//  Sanity-check on cache parameters:
+	.ifne	(\size % (\linesize * \assoc_or1 * 4))
+	.err	//  cache configuration outside expected/supported range!
+	.endif
+
+	//  \size byte cache, \linesize byte lines, \assoc_or1 way(s) affected by each \cainst.
+	movi	\aa, (\size / (\linesize * \assoc_or1 * 4))
+	// Possible improvement: need only loop if \aa > 1 ;
+	// however that particular condition is highly unlikely.
+	movi	\ab, 0		// to iterate over cache
+	floop		\aa, cachex\@
+	\cainst		\ab, 0*\linesize
+	\cainst		\ab, 1*\linesize
+	\cainst		\ab, 2*\linesize
+	\cainst		\ab, 3*\linesize
+	addi		\ab, \ab, 4*\linesize	// move to next line
+	floopend	\aa, cachex\@
+
+	.endm
+
+
+/*
+ *  Macro to apply a 'hit' cache instruction to a memory region,
+ *  ie. to any cache entries that cache a specified portion (region) of memory.
+ *  Takes care of the unaligned cases, ie. may apply to one
+ *  more cache line than $asize / lineSize if $aaddr is not aligned.
+ *
+ *
+ *  Parameters are:
+ *	cainst	instruction/macro that takes an address register parameter
+ *		and an offset parameter (currently always zero)
+ *		and generates a cache instruction (eg. "dhi", "dhwb", "ihi", etc.)
+ *	linesize_log2	log2(size of cache line in bytes)
+ *	addr	register containing start address of region (clobbered)
+ *	asize	register containing size of the region in bytes (clobbered)
+ *	askew	unique register used as temporary
+ *
+ * !?!?! 2DO: optimization: iterate max(cache_size and \asize) / linesize
+ */
+
+	.macro	cache_hit_region	cainst, linesize_log2, addr, asize, askew
+
+	//  Make \asize the number of iterations:
+	extui	\askew, \addr, 0, \linesize_log2	// get unalignment amount of \addr
+	add	\asize, \asize, \askew			// ... and add it to \asize
+	addi	\asize, \asize, (1 << \linesize_log2) - 1	// round up!
+	srli	\asize, \asize, \linesize_log2
+
+	//  Iterate over region:
+	floopnez	\asize, cacheh\@
+	\cainst		\addr, 0
+	addi		\addr, \addr, (1 << \linesize_log2)	// move to next line
+	floopend	\asize, cacheh\@
+
+	.endm
+
+
+
+
+
+/***************************   INSTRUCTION CACHE   ***************************/
+
+
+/*
+ *  Reset/initialize the instruction cache by simply invalidating it:
+ *  (need to unlock first also, if cache locking implemented):
+ *
+ *  Parameters:
+ *	aa, ab		unique address registers (temporaries)
+ */
+	.macro	icache_reset	aa, ab
+	icache_unlock_all	\aa, \ab
+	icache_invalidate_all	\aa, \ab
+	.endm
+
+
+/*
+ * Synchronize after an instruction cache operation,
+ * to be sure everything is in sync with memory as to be
+ * expected following any previous instruction cache control operations.
+ *
+ * Parameters are:
+ *	ar	an address register (temporary) (currently unused, but may be used in future)
+ */
+	.macro	icache_sync	ar
+#if XCHAL_ICACHE_SIZE > 0
+	isync
+#endif
+	.endm
+
+
+
+/*
+ *  Invalidate a single line of the instruction cache.
+ *  Parameters are:
+ *	ar	address register that contains (virtual) address to invalidate
+ *		(may get clobbered in a future implementation, but not currently)
+ *	offset	(optional) offset to add to \ar to compute effective address to invalidate
+ *		(note: some number of lsbits are ignored)
+ */
+	.macro	icache_invalidate_line	ar, offset
+#if XCHAL_ICACHE_SIZE > 0
+	ihi	\ar, \offset		// invalidate icache line
+	/*
+	 *  NOTE:  in some version of the silicon [!!!SHOULD HAVE BEEN DOCUMENTED!!!]
+	 *  'ihi' doesn't work, so it had been replaced with 'iii'
+	 *  (which would just invalidate more than it should,
+	 *  which should be okay other than the performance hit
+	 *  because cache locking did not exist in that version,
+	 *  unless user somehow relies on something being cached).
+	 *  [WHAT VERSION IS IT!!?!?
+	 *  IS THERE ANY WAY TO TEST FOR THAT HERE, TO OUTPUT 'III' ONLY IF NEEDED!?!?].
+	 *
+	 *	iii	\ar, \offset
+	 */
+	icache_sync	\ar
+#endif
+	.endm
+
+
+
+
+/*
+ *  Invalidate instruction  cache entries that cache a specified portion of memory.
+ *  Parameters are:
+ *	astart	start address (register gets clobbered)
+ *	asize	size of the region in bytes (register gets clobbered)
+ *	ac	unique register used as temporary
+ */
+	.macro	icache_invalidate_region	astart, asize, ac
+#if XCHAL_ICACHE_SIZE > 0
+	//  Instruction cache region invalidation:
+	cache_hit_region	ihi, XCHAL_ICACHE_LINEWIDTH, \astart, \asize, \ac
+	icache_sync	\ac
+	//  End of instruction cache region invalidation
+#endif
+	.endm
+
+
+
+/*
+ *  Invalidate entire instruction cache.
+ *
+ *  Parameters:
+ *	aa, ab		unique address registers (temporaries)
+ */
+	.macro	icache_invalidate_all	aa, ab
+#if XCHAL_ICACHE_SIZE > 0
+	//  Instruction cache invalidation:
+	cache_index_all		iii, XCHAL_ICACHE_SIZE, XCHAL_ICACHE_LINESIZE, XCHAL_ICACHE_WAYS, \aa, \ab
+	icache_sync	\aa
+	//  End of instruction cache invalidation
+#endif
+	.endm
+
+
+
+/*
+ *  Lock (prefetch & lock) a single line of the instruction cache.
+ *
+ *  Parameters are:
+ *	ar	address register that contains (virtual) address to lock
+ *		(may get clobbered in a future implementation, but not currently)
+ *	offset	offset to add to \ar to compute effective address to lock
+ *		(note: some number of lsbits are ignored)
+ */
+	.macro	icache_lock_line	ar, offset
+#if XCHAL_ICACHE_SIZE > 0 && XCHAL_ICACHE_LINE_LOCKABLE
+	ipfl	\ar, \offset	/* prefetch and lock icache line */
+	icache_sync	\ar
+#endif
+	.endm
+
+
+
+/*
+ *  Lock (prefetch & lock) a specified portion of memory into the instruction cache.
+ *  Parameters are:
+ *	astart	start address (register gets clobbered)
+ *	asize	size of the region in bytes (register gets clobbered)
+ *	ac	unique register used as temporary
+ */
+	.macro	icache_lock_region	astart, asize, ac
+#if XCHAL_ICACHE_SIZE > 0 && XCHAL_ICACHE_LINE_LOCKABLE
+	//  Instruction cache region lock:
+	cache_hit_region	ipfl, XCHAL_ICACHE_LINEWIDTH, \astart, \asize, \ac
+	icache_sync	\ac
+	//  End of instruction cache region lock
+#endif
+	.endm
+
+
+
+/*
+ *  Unlock a single line of the instruction cache.
+ *
+ *  Parameters are:
+ *	ar	address register that contains (virtual) address to unlock
+ *		(may get clobbered in a future implementation, but not currently)
+ *	offset	offset to add to \ar to compute effective address to unlock
+ *		(note: some number of lsbits are ignored)
+ */
+	.macro	icache_unlock_line	ar, offset
+#if XCHAL_ICACHE_SIZE > 0 && XCHAL_ICACHE_LINE_LOCKABLE
+	ihu	\ar, \offset	/* unlock icache line */
+	icache_sync	\ar
+#endif
+	.endm
+
+
+
+/*
+ *  Unlock a specified portion of memory from the instruction cache.
+ *  Parameters are:
+ *	astart	start address (register gets clobbered)
+ *	asize	size of the region in bytes (register gets clobbered)
+ *	ac	unique register used as temporary
+ */
+	.macro	icache_unlock_region	astart, asize, ac
+#if XCHAL_ICACHE_SIZE > 0 && XCHAL_ICACHE_LINE_LOCKABLE
+	//  Instruction cache region unlock:
+	cache_hit_region	ihu, XCHAL_ICACHE_LINEWIDTH, \astart, \asize, \ac
+	icache_sync	\ac
+	//  End of instruction cache region unlock
+#endif
+	.endm
+
+
+
+/*
+ *  Unlock entire instruction cache.
+ *
+ *  Parameters:
+ *	aa, ab		unique address registers (temporaries)
+ */
+	.macro	icache_unlock_all	aa, ab
+#if XCHAL_ICACHE_SIZE > 0 && XCHAL_ICACHE_LINE_LOCKABLE
+	//  Instruction cache unlock:
+	cache_index_all		iiu, XCHAL_ICACHE_SIZE, XCHAL_ICACHE_LINESIZE, 1, \aa, \ab
+	icache_sync	\aa
+	//  End of instruction cache unlock
+#endif
+	.endm
+
+
+
+
+
+/***************************   DATA CACHE   ***************************/
+
+
+
+/*
+ *  Reset/initialize the data cache by simply invalidating it
+ *  (need to unlock first also, if cache locking implemented):
+ *
+ *  Parameters:
+ *	aa, ab		unique address registers (temporaries)
+ */
+	.macro	dcache_reset	aa, ab
+	dcache_unlock_all	\aa, \ab
+	dcache_invalidate_all	\aa, \ab
+	.endm
+
+
+
+
+/*
+ * Synchronize after a data cache operation,
+ * to be sure everything is in sync with memory as to be
+ * expected following any previous data cache control operations.
+ *
+ * Parameters are:
+ *	ar	an address register (temporary) (currently unused, but may be used in future)
+ */
+	.macro	dcache_sync	ar
+#if XCHAL_DCACHE_SIZE > 0
+	//  This previous sequence errs on the conservative side (too much so); a DSYNC should be sufficient:
+	//memw		// synchronize data cache changes relative to subsequent memory accesses
+	//isync		// be conservative and ISYNC as well (just to be sure)
+
+	dsync
+#endif
+	.endm
+
+
+
+/*
+ * Synchronize after a data store operation,
+ * to be sure the stored data is completely off the processor
+ * (and assuming there is no buffering outside the processor,
+ *  that the data is in memory).  This may be required to
+ * ensure that the processor's write buffers are emptied.
+ * A MEMW followed by a read guarantees this, by definition.
+ * We also try to make sure the read itself completes.
+ *
+ * Parameters are:
+ *	ar	an address register (temporary)
+ */
+	.macro	write_sync	ar
+	memw			// ensure previous memory accesses are complete prior to subsequent memory accesses
+	l32i	\ar, sp, 0	// completing this read ensures any previous write has completed, because of MEMW
+	//slot
+	add	\ar, \ar, \ar	// use the result of the read to help ensure the read completes (in future architectures)
+	.endm
+
+
+/*
+ *  Invalidate a single line of the data cache.
+ *  Parameters are:
+ *	ar	address register that contains (virtual) address to invalidate
+ *		(may get clobbered in a future implementation, but not currently)
+ *	offset	(optional) offset to add to \ar to compute effective address to invalidate
+ *		(note: some number of lsbits are ignored)
+ */
+	.macro	dcache_invalidate_line	ar, offset
+#if XCHAL_DCACHE_SIZE > 0
+	dhi	\ar, \offset
+	dcache_sync	\ar
+#endif
+	.endm
+
+
+
+
+
+/*
+ *  Invalidate data cache entries that cache a specified portion of memory.
+ *  Parameters are:
+ *	astart	start address (register gets clobbered)
+ *	asize	size of the region in bytes (register gets clobbered)
+ *	ac	unique register used as temporary
+ */
+	.macro	dcache_invalidate_region	astart, asize, ac
+#if XCHAL_DCACHE_SIZE > 0
+	//  Data cache region invalidation:
+	cache_hit_region	dhi, XCHAL_DCACHE_LINEWIDTH, \astart, \asize, \ac
+	dcache_sync	\ac
+	//  End of data cache region invalidation
+#endif
+	.endm
+
+
+
+#if 0
+/*
+ *  This is a work-around for a bug in SiChip1 (???).
+ *  There should be a proper mechanism for not outputting
+ *  these instructions when not needed.
+ *  To enable work-around, uncomment this and replace 'dii'
+ *  with 'dii_s1' everywhere, eg. in dcache_invalidate_all
+ *  macro below.
+ */
+	.macro	dii_s1	ar, offset
+	dii	\ar, \offset
+	or	\ar, \ar, \ar
+	or	\ar, \ar, \ar
+	or	\ar, \ar, \ar
+	or	\ar, \ar, \ar
+	.endm
+#endif
+
+
+/*
+ *  Invalidate entire data cache.
+ *
+ *  Parameters:
+ *	aa, ab		unique address registers (temporaries)
+ */
+	.macro	dcache_invalidate_all	aa, ab
+#if XCHAL_DCACHE_SIZE > 0
+	//  Data cache invalidation:
+	cache_index_all		dii, XCHAL_DCACHE_SIZE, XCHAL_DCACHE_LINESIZE, XCHAL_DCACHE_WAYS, \aa, \ab
+	dcache_sync	\aa
+	//  End of data cache invalidation
+#endif
+	.endm
+
+
+
+/*
+ *  Writeback a single line of the data cache.
+ *  Parameters are:
+ *	ar	address register that contains (virtual) address to writeback
+ *		(may get clobbered in a future implementation, but not currently)
+ *	offset	offset to add to \ar to compute effective address to writeback
+ *		(note: some number of lsbits are ignored)
+ */
+	.macro	dcache_writeback_line	ar, offset
+#if XCHAL_DCACHE_SIZE > 0 && XCHAL_DCACHE_IS_WRITEBACK
+	dhwb	\ar, \offset
+	dcache_sync	\ar
+#endif
+	.endm
+
+
+
+/*
+ *  Writeback dirty data cache entries that cache a specified portion of memory.
+ *  Parameters are:
+ *	astart	start address (register gets clobbered)
+ *	asize	size of the region in bytes (register gets clobbered)
+ *	ac	unique register used as temporary
+ */
+	.macro	dcache_writeback_region		astart, asize, ac
+#if XCHAL_DCACHE_SIZE > 0 && XCHAL_DCACHE_IS_WRITEBACK
+	//  Data cache region writeback:
+	cache_hit_region	dhwb, XCHAL_DCACHE_LINEWIDTH, \astart, \asize, \ac
+	dcache_sync	\ac
+	//  End of data cache region writeback
+#endif
+	.endm
+
+
+
+/*
+ *  Writeback entire data cache.
+ *  Parameters:
+ *	aa, ab		unique address registers (temporaries)
+ */
+	.macro	dcache_writeback_all	aa, ab
+#if XCHAL_DCACHE_SIZE > 0 && XCHAL_DCACHE_IS_WRITEBACK
+	//  Data cache writeback:
+	cache_index_all		diwb, XCHAL_DCACHE_SIZE, XCHAL_DCACHE_LINESIZE, 1, \aa, \ab
+	dcache_sync	\aa
+	//  End of data cache writeback
+#endif
+	.endm
+
+
+
+/*
+ *  Writeback and invalidate a single line of the data cache.
+ *  Parameters are:
+ *	ar	address register that contains (virtual) address to writeback and invalidate
+ *		(may get clobbered in a future implementation, but not currently)
+ *	offset	offset to add to \ar to compute effective address to writeback and invalidate
+ *		(note: some number of lsbits are ignored)
+ */
+	.macro	dcache_writeback_inv_line	ar, offset
+#if XCHAL_DCACHE_SIZE > 0
+	dhwbi	\ar, \offset	/* writeback and invalidate dcache line */
+	dcache_sync	\ar
+#endif
+	.endm
+
+
+
+/*
+ *  Writeback and invalidate data cache entries that cache a specified portion of memory.
+ *  Parameters are:
+ *	astart	start address (register gets clobbered)
+ *	asize	size of the region in bytes (register gets clobbered)
+ *	ac	unique register used as temporary
+ */
+	.macro	dcache_writeback_inv_region	astart, asize, ac
+#if XCHAL_DCACHE_SIZE > 0
+	//  Data cache region writeback and invalidate:
+	cache_hit_region	dhwbi, XCHAL_DCACHE_LINEWIDTH, \astart, \asize, \ac
+	dcache_sync	\ac
+	//  End of data cache region writeback and invalidate
+#endif
+	.endm
+
+
+
+/*
+ *  Writeback and invalidate entire data cache.
+ *  Parameters:
+ *	aa, ab		unique address registers (temporaries)
+ */
+	.macro	dcache_writeback_inv_all	aa, ab
+#if XCHAL_DCACHE_SIZE > 0
+	//  Data cache writeback and invalidate:
+#if XCHAL_DCACHE_IS_WRITEBACK
+	cache_index_all		diwbi, XCHAL_DCACHE_SIZE, XCHAL_DCACHE_LINESIZE, 1, \aa, \ab
+	dcache_sync	\aa
+#else /*writeback*/
+	//  Data cache does not support writeback, so just invalidate: */
+	dcache_invalidate_all	\aa, \ab
+#endif /*writeback*/
+	//  End of data cache writeback and invalidate
+#endif
+	.endm
+
+
+
+
+/*
+ *  Lock (prefetch & lock) a single line of the data cache.
+ *
+ *  Parameters are:
+ *	ar	address register that contains (virtual) address to lock
+ *		(may get clobbered in a future implementation, but not currently)
+ *	offset	offset to add to \ar to compute effective address to lock
+ *		(note: some number of lsbits are ignored)
+ */
+	.macro	dcache_lock_line	ar, offset
+#if XCHAL_DCACHE_SIZE > 0 && XCHAL_DCACHE_LINE_LOCKABLE
+	dpfl	\ar, \offset	/* prefetch and lock dcache line */
+	dcache_sync	\ar
+#endif
+	.endm
+
+
+
+/*
+ *  Lock (prefetch & lock) a specified portion of memory into the data cache.
+ *  Parameters are:
+ *	astart	start address (register gets clobbered)
+ *	asize	size of the region in bytes (register gets clobbered)
+ *	ac	unique register used as temporary
+ */
+	.macro	dcache_lock_region	astart, asize, ac
+#if XCHAL_DCACHE_SIZE > 0 && XCHAL_DCACHE_LINE_LOCKABLE
+	//  Data cache region lock:
+	cache_hit_region	dpfl, XCHAL_DCACHE_LINEWIDTH, \astart, \asize, \ac
+	dcache_sync	\ac
+	//  End of data cache region lock
+#endif
+	.endm
+
+
+
+/*
+ *  Unlock a single line of the data cache.
+ *
+ *  Parameters are:
+ *	ar	address register that contains (virtual) address to unlock
+ *		(may get clobbered in a future implementation, but not currently)
+ *	offset	offset to add to \ar to compute effective address to unlock
+ *		(note: some number of lsbits are ignored)
+ */
+	.macro	dcache_unlock_line	ar, offset
+#if XCHAL_DCACHE_SIZE > 0 && XCHAL_DCACHE_LINE_LOCKABLE
+	dhu	\ar, \offset	/* unlock dcache line */
+	dcache_sync	\ar
+#endif
+	.endm
+
+
+
+/*
+ *  Unlock a specified portion of memory from the data cache.
+ *  Parameters are:
+ *	astart	start address (register gets clobbered)
+ *	asize	size of the region in bytes (register gets clobbered)
+ *	ac	unique register used as temporary
+ */
+	.macro	dcache_unlock_region	astart, asize, ac
+#if XCHAL_DCACHE_SIZE > 0 && XCHAL_DCACHE_LINE_LOCKABLE
+	//  Data cache region unlock:
+	cache_hit_region	dhu, XCHAL_DCACHE_LINEWIDTH, \astart, \asize, \ac
+	dcache_sync	\ac
+	//  End of data cache region unlock
+#endif
+	.endm
+
+
+
+/*
+ *  Unlock entire data cache.
+ *
+ *  Parameters:
+ *	aa, ab		unique address registers (temporaries)
+ */
+	.macro	dcache_unlock_all	aa, ab
+#if XCHAL_DCACHE_SIZE > 0 && XCHAL_DCACHE_LINE_LOCKABLE
+	//  Data cache unlock:
+	cache_index_all		diu, XCHAL_DCACHE_SIZE, XCHAL_DCACHE_LINESIZE, 1, \aa, \ab
+	dcache_sync	\aa
+	//  End of data cache unlock
+#endif
+	.endm
+
+
+#endif /*XTENSA_CACHEASM_H*/
+
diff --git a/include/asm-xtensa/xtensa/cacheattrasm.h b/include/asm-xtensa/xtensa/cacheattrasm.h
new file mode 100644
index 00000000000..1c3e117b359
--- /dev/null
+++ b/include/asm-xtensa/xtensa/cacheattrasm.h
@@ -0,0 +1,432 @@
+#ifndef XTENSA_CACHEATTRASM_H
+#define XTENSA_CACHEATTRASM_H
+
+/*
+ * THIS FILE IS GENERATED -- DO NOT MODIFY BY HAND
+ *
+ * include/asm-xtensa/xtensa/cacheattrasm.h -- assembler-specific
+ * CACHEATTR register related definitions that depend on CORE
+ * configuration.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2002 Tensilica Inc.
+ */
+
+
+#include <xtensa/coreasm.h>
+
+
+/*
+ *  This header file defines assembler macros of the form:
+ *	<x>cacheattr_<func>
+ *  where:
+ *	<x> is 'i', 'd' or absent for instruction, data
+ *		or both caches; and
+ *	<func> indicates the function of the macro.
+ *
+ *  The following functions are defined:
+ *
+ *  icacheattr_get
+ *	Reads I-cache CACHEATTR into a2 (clobbers a3-a5).
+ *
+ *  dcacheattr_get
+ *	Reads D-cache CACHEATTR into a2 (clobbers a3-a5).
+ *	(Note:  for configs with a real CACHEATTR register, the
+ *	 above two macros are identical.)
+ *
+ *  cacheattr_set
+ *	Writes both I-cache and D-cache CACHEATTRs from a2 (a3-a8 clobbered).
+ *	Works even when changing one's own code's attributes.
+ *
+ *  icacheattr_is_enabled  label
+ *	Branches to \label if I-cache appears to have been enabled
+ *	(eg. if CACHEATTR contains a cache-enabled attribute).
+ *	(clobbers a2-a5,SAR)
+ *
+ *  dcacheattr_is_enabled  label
+ *	Branches to \label if D-cache appears to have been enabled
+ *	(eg. if CACHEATTR contains a cache-enabled attribute).
+ *	(clobbers a2-a5,SAR)
+ *
+ *  cacheattr_is_enabled  label
+ *	Branches to \label if either I-cache or D-cache appears to have been enabled
+ *	(eg. if CACHEATTR contains a cache-enabled attribute).
+ *	(clobbers a2-a5,SAR)
+ *
+ *  The following macros are only defined under certain conditions:
+ *
+ *  icacheattr_set	(if XCHAL_HAVE_MIMIC_CACHEATTR || XCHAL_HAVE_XLT_CACHEATTR)
+ *	Writes I-cache CACHEATTR from a2 (a3-a8 clobbered).
+ *
+ *  dcacheattr_set	(if XCHAL_HAVE_MIMIC_CACHEATTR || XCHAL_HAVE_XLT_CACHEATTR)
+ *	Writes D-cache CACHEATTR from a2 (a3-a8 clobbered).
+ */
+
+
+
+/***************************   GENERIC -- ALL CACHES   ***************************/
+
+/*
+ *  _cacheattr_get
+ *
+ *  (Internal macro.)
+ *  Returns value of CACHEATTR register (or closest equivalent) in a2.
+ *
+ *  Entry:
+ *	(none)
+ *  Exit:
+ *	a2	value read from CACHEATTR
+ *	a3-a5	clobbered (temporaries)
+ */
+	.macro	_cacheattr_get	tlb
+#if XCHAL_HAVE_CACHEATTR
+	rsr	a2, CACHEATTR
+#elif XCHAL_HAVE_MIMIC_CACHEATTR || XCHAL_HAVE_XLT_CACHEATTR
+	//  We have a config that "mimics" CACHEATTR using a simplified
+	//  "MMU" composed of a single statically-mapped way.
+	//  DTLB and ITLB are independent, so there's no single
+	//  cache attribute that can describe both.  So for now
+	//  just return the DTLB state.
+	movi	a5, 0xE0000000
+	movi	a2, 0
+	movi	a3, 0
+1:	add	a3, a3, a5	// next segment
+	r&tlb&1	a4, a3		// get PPN+CA of segment at 0xE0000000, 0xC0000000, ..., 0
+	dsync	// interlock???
+	slli	a2, a2, 4
+	extui	a4, a4, 0, 4	// extract CA
+	or	a2, a2, a4
+	bnez	a3, 1b
+#else
+	//  This macro isn't applicable to arbitrary MMU configurations.
+	//  Just return zero.
+	movi	a2, 0
+#endif
+	.endm
+
+	.macro	icacheattr_get
+	_cacheattr_get	itlb
+	.endm
+
+	.macro	dcacheattr_get
+	_cacheattr_get	dtlb
+	.endm
+
+
+#define XCHAL_CACHEATTR_ALL_BYPASS	0x22222222	/* default (powerup/reset) value of CACHEATTR, all BYPASS
+							   mode (ie. disabled/bypassed caches) */
+
+#if XCHAL_HAVE_CACHEATTR || XCHAL_HAVE_MIMIC_CACHEATTR || XCHAL_HAVE_XLT_CACHEATTR
+
+#define XCHAL_FCA_ENAMASK	0x001A	/* bitmap of fetch attributes that require enabled icache */
+#define XCHAL_LCA_ENAMASK	0x0003	/* bitmap of load  attributes that require enabled dcache */
+#define XCHAL_SCA_ENAMASK	0x0003	/* bitmap of store attributes that require enabled dcache */
+#define XCHAL_LSCA_ENAMASK	(XCHAL_LCA_ENAMASK|XCHAL_SCA_ENAMASK)	/* l/s attrs requiring enabled dcache */
+#define XCHAL_ALLCA_ENAMASK	(XCHAL_FCA_ENAMASK|XCHAL_LSCA_ENAMASK)	/* all attrs requiring enabled caches */
+
+/*
+ *  _cacheattr_is_enabled
+ *
+ *  (Internal macro.)
+ *  Branches to \label if CACHEATTR in a2 indicates an enabled
+ *  cache, using mask in a3.
+ *
+ *  Parameters:
+ *	label	where to branch to if cache is enabled
+ *  Entry:
+ *	a2	contains CACHEATTR value used to determine whether
+ *		caches are enabled
+ *	a3	16-bit constant where each bit correspond to
+ *		one of the 16 possible CA values (in a CACHEATTR mask);
+ *		CA values that indicate the cache is enabled
+ *		have their corresponding bit set in this mask
+ *		(eg. use XCHAL_xCA_ENAMASK , above)
+ *  Exit:
+ *	a2,a4,a5	clobbered
+ *	SAR		clobbered
+ */
+	.macro	_cacheattr_is_enabled	label
+	movi	a4, 8		// loop 8 times
+.Lcaife\@:
+	extui	a5, a2, 0, 4	// get CA nibble
+	ssr	a5		// index into mask according to CA...
+	srl	a5, a3		// ...and get CA's mask bit in a5 bit 0
+	bbsi.l	a5, 0, \label	// if CA indicates cache enabled, jump to label
+	srli	a2, a2, 4	// next nibble
+	addi	a4, a4, -1
+	bnez	a4, .Lcaife\@	// loop for each nibble
+	.endm
+
+#else /* XCHAL_HAVE_CACHEATTR || XCHAL_HAVE_MIMIC_CACHEATTR || XCHAL_HAVE_XLT_CACHEATTR */
+	.macro	_cacheattr_is_enabled	label
+	j	\label		// macro not applicable, assume caches always enabled
+	.endm
+#endif /* XCHAL_HAVE_CACHEATTR || XCHAL_HAVE_MIMIC_CACHEATTR || XCHAL_HAVE_XLT_CACHEATTR */
+
+
+
+/*
+ *  icacheattr_is_enabled
+ *
+ *  Branches to \label if I-cache is enabled.
+ *
+ *  Parameters:
+ *	label	where to branch to if icache is enabled
+ *  Entry:
+ *	(none)
+ *  Exit:
+ *	a2-a5, SAR	clobbered (temporaries)
+ */
+	.macro	icacheattr_is_enabled	label
+#if XCHAL_HAVE_CACHEATTR || XCHAL_HAVE_MIMIC_CACHEATTR || XCHAL_HAVE_XLT_CACHEATTR
+	icacheattr_get
+	movi	a3, XCHAL_FCA_ENAMASK
+#endif
+	_cacheattr_is_enabled	\label
+	.endm
+
+/*
+ *  dcacheattr_is_enabled
+ *
+ *  Branches to \label if D-cache is enabled.
+ *
+ *  Parameters:
+ *	label	where to branch to if dcache is enabled
+ *  Entry:
+ *	(none)
+ *  Exit:
+ *	a2-a5, SAR	clobbered (temporaries)
+ */
+	.macro	dcacheattr_is_enabled	label
+#if XCHAL_HAVE_CACHEATTR || XCHAL_HAVE_MIMIC_CACHEATTR || XCHAL_HAVE_XLT_CACHEATTR
+	dcacheattr_get
+	movi	a3, XCHAL_LSCA_ENAMASK
+#endif
+	_cacheattr_is_enabled	\label
+	.endm
+
+/*
+ *  cacheattr_is_enabled
+ *
+ *  Branches to \label if either I-cache or D-cache is enabled.
+ *
+ *  Parameters:
+ *	label	where to branch to if a cache is enabled
+ *  Entry:
+ *	(none)
+ *  Exit:
+ *	a2-a5, SAR	clobbered (temporaries)
+ */
+	.macro	cacheattr_is_enabled	label
+#if XCHAL_HAVE_CACHEATTR
+	rsr	a2, CACHEATTR
+	movi	a3, XCHAL_ALLCA_ENAMASK
+#elif XCHAL_HAVE_MIMIC_CACHEATTR || XCHAL_HAVE_XLT_CACHEATTR
+	icacheattr_get
+	movi	a3, XCHAL_FCA_ENAMASK
+	_cacheattr_is_enabled	\label
+	dcacheattr_get
+	movi	a3, XCHAL_LSCA_ENAMASK
+#endif
+	_cacheattr_is_enabled	\label
+	.endm
+
+
+
+/*
+ *  The ISA does not have a defined way to change the
+ *  instruction cache attributes of the running code,
+ *  ie. of the memory area that encloses the current PC.
+ *  However, each micro-architecture (or class of
+ *  configurations within a micro-architecture)
+ *  provides a way to deal with this issue.
+ *
+ *  Here are a few macros used to implement the relevant
+ *  approach taken.
+ */
+
+#if XCHAL_HAVE_MIMIC_CACHEATTR || XCHAL_HAVE_XLT_CACHEATTR
+	//  We have a config that "mimics" CACHEATTR using a simplified
+	//  "MMU" composed of a single statically-mapped way.
+
+/*
+ *  icacheattr_set
+ *
+ *  Entry:
+ *	a2		cacheattr value to set
+ *  Exit:
+ *	a2		unchanged
+ *	a3-a8		clobbered (temporaries)
+ */
+	.macro	icacheattr_set
+
+	movi	a5, 0xE0000000	// mask of upper 3 bits
+	movi	a6, 3f		// PC where ITLB is set
+	movi	a3, 0		// start at region 0 (0 .. 7)
+	and	a6, a6, a5	// upper 3 bits of local PC area
+	mov	a7, a2		// copy a2 so it doesn't get clobbered
+	j	3f
+
+# if XCHAL_HAVE_XLT_CACHEATTR
+	//  Can do translations, use generic method:
+1:	sub	a6, a3, a5	// address of some other segment
+	ritlb1	a8, a6		// save its PPN+CA
+	dsync	// interlock??
+	witlb	a4, a6		// make it translate to this code area
+	movi	a6, 5f		// where to jump into it
+	isync
+	sub	a6, a6, a5	// adjust jump address within that other segment
+	jx	a6
+
+	//  Note that in the following code snippet, which runs at a different virtual
+	//  address than it is assembled for, we avoid using literals (eg. via movi/l32r)
+	//  just in case literals end up in a different 512 MB segment, and we avoid
+	//  instructions that rely on the current PC being what is expected.
+	//
+	.align	4
+	_j	6f		// this is at label '5' minus 4 bytes
+	.align	4
+5:	witlb	a4, a3		// we're in other segment, now can write previous segment's CA
+	isync
+	add	a6, a6, a5	// back to previous segment
+	addi	a6, a6, -4	// next jump label
+	jx	a6
+
+6:	sub	a6, a3, a5	// address of some other segment
+	witlb	a8, a6		// restore PPN+CA of other segment
+	mov	a6, a3		// restore a6
+	isync
+# else /* XCHAL_HAVE_XLT_CACHEATTR */
+	//  Use micro-architecture specific method.
+	//  The following 4-instruction sequence is aligned such that
+	//  it all fits within a single I-cache line.  Sixteen byte
+	//  alignment is sufficient for this (using XCHAL_ICACHE_LINESIZE
+	//  actually causes problems because that can be greater than
+	//  the alignment of the reset vector, where this macro is often
+	//  invoked, which would cause the linker to align the reset
+	//  vector code away from the reset vector!!).
+	.align	16 /*XCHAL_ICACHE_LINESIZE*/
+1:	_witlb	a4, a3		// write wired PTE (CA, no PPN) of 512MB segment to ITLB
+	_isync
+	nop
+	nop
+# endif /* XCHAL_HAVE_XLT_CACHEATTR */
+	beq	a3, a5, 4f	// done?
+
+	//  Note that in the WITLB loop, we don't do any load/stores
+	//  (may not be an issue here, but it is important in the DTLB case).
+2:	srli	a7, a7, 4	// next CA
+	sub	a3, a3, a5	// next segment (add 0x20000000)
+3:
+# if XCHAL_HAVE_XLT_CACHEATTR	/* if have translation, preserve it */
+	ritlb1	a8, a3		// get current PPN+CA of segment
+	dsync	// interlock???
+	extui	a4, a7, 0, 4	// extract CA to set
+	srli	a8, a8, 4	// clear CA but keep PPN ...
+	slli	a8, a8, 4	// ...
+	add	a4, a4, a8	// combine new CA with PPN to preserve
+# else
+	extui	a4, a7, 0, 4	// extract CA
+# endif
+	beq	a3, a6, 1b	// current PC's region? if so, do it in a safe way
+	witlb	a4, a3		// write wired PTE (CA [+PPN]) of 512MB segment to ITLB
+	bne	a3, a5, 2b
+	isync			// make sure all ifetch changes take effect
+4:
+	.endm	// icacheattr_set
+
+
+/*
+ *  dcacheattr_set
+ *
+ *  Entry:
+ *	a2		cacheattr value to set
+ *  Exit:
+ *	a2		unchanged
+ *	a3-a8		clobbered (temporaries)
+ */
+
+	.macro	dcacheattr_set
+
+	movi	a5, 0xE0000000	// mask of upper 3 bits
+	movi	a3, 0		// start at region 0 (0 .. 7)
+	mov	a7, a2		// copy a2 so it doesn't get clobbered
+	j	3f
+	//  Note that in the WDTLB loop, we don't do any load/stores
+	//  (including implicit l32r via movi) because it isn't safe.
+2:	srli	a7, a7, 4	// next CA
+	sub	a3, a3, a5	// next segment (add 0x20000000)
+3:
+# if XCHAL_HAVE_XLT_CACHEATTR	/* if have translation, preserve it */
+	rdtlb1	a8, a3		// get current PPN+CA of segment
+	dsync	// interlock???
+	extui	a4, a7, 0, 4	// extract CA to set
+	srli	a8, a8, 4	// clear CA but keep PPN ...
+	slli	a8, a8, 4	// ...
+	add	a4, a4, a8	// combine new CA with PPN to preserve
+# else
+	extui	a4, a7, 0, 4	// extract CA to set
+# endif
+	wdtlb	a4, a3		// write wired PTE (CA [+PPN]) of 512MB segment to DTLB
+	bne	a3, a5, 2b
+	dsync			// make sure all data path changes take effect
+	.endm	// dcacheattr_set
+
+#endif /* XCHAL_HAVE_MIMIC_CACHEATTR || XCHAL_HAVE_XLT_CACHEATTR */
+
+
+
+/*
+ *  cacheattr_set
+ *
+ *  Macro that sets the current CACHEATTR safely
+ *  (both i and d) according to the current contents of a2.
+ *  It works even when changing the cache attributes of
+ *  the currently running code.
+ *
+ *  Entry:
+ *	a2		cacheattr value to set
+ *  Exit:
+ *	a2		unchanged
+ *	a3-a8		clobbered (temporaries)
+ */
+	.macro	cacheattr_set
+
+#if XCHAL_HAVE_CACHEATTR
+# if XCHAL_ICACHE_LINESIZE < 4
+	//  No i-cache, so can always safely write to CACHEATTR:
+	wsr	a2, CACHEATTR
+# else
+	//  The Athens micro-architecture, when using the old
+	//  exception architecture option (ie. with the CACHEATTR register)
+	//  allows changing the cache attributes of the running code
+	//  using the following exact sequence aligned to be within
+	//  an instruction cache line.  (NOTE: using XCHAL_ICACHE_LINESIZE
+	//  alignment actually causes problems because that can be greater
+	//  than the alignment of the reset vector, where this macro is often
+	//  invoked, which would cause the linker to align the reset
+	//  vector code away from the reset vector!!).
+	j	1f
+	.align	16 /*XCHAL_ICACHE_LINESIZE*/	// align to within an I-cache line
+1:	_wsr	a2, CACHEATTR
+	_isync
+	nop
+	nop
+# endif
+#elif XCHAL_HAVE_MIMIC_CACHEATTR || XCHAL_HAVE_XLT_CACHEATTR
+	//  DTLB and ITLB are independent, but to keep semantics
+	//  of this macro we simply write to both.
+	icacheattr_set
+	dcacheattr_set
+#else
+	//  This macro isn't applicable to arbitrary MMU configurations.
+	//  Do nothing in this case.
+#endif
+	.endm
+
+
+#endif /*XTENSA_CACHEATTRASM_H*/
+
diff --git a/include/asm-xtensa/xtensa/config-linux_be/core.h b/include/asm-xtensa/xtensa/config-linux_be/core.h
new file mode 100644
index 00000000000..d54fe5eb106
--- /dev/null
+++ b/include/asm-xtensa/xtensa/config-linux_be/core.h
@@ -0,0 +1,1270 @@
+/*
+ * xtensa/config/core.h -- HAL definitions that are dependent on CORE configuration
+ *
+ *  This header file is sometimes referred to as the "compile-time HAL" or CHAL.
+ *  It was generated for a specific Xtensa processor configuration.
+ *
+ *  Source for configuration-independent binaries (which link in a
+ *  configuration-specific HAL library) must NEVER include this file.
+ *  It is perfectly normal, however, for the HAL source itself to include this file.
+ */
+
+/*
+ * Copyright (c) 2003 Tensilica, Inc.  All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2.1 of the GNU Lesser General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+ *
+ * Further, this software is distributed without any warranty that it is
+ * free of the rightful claim of any third person regarding infringement
+ * or the like.  Any license provided herein, whether implied or
+ * otherwise, applies only to this software file.  Patent licenses, if
+ * any, provided herein do not apply to combinations of this program with
+ * other software, or any other product whatsoever.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this program; if not, write the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston MA 02111-1307,
+ * USA.
+ */
+
+
+#ifndef XTENSA_CONFIG_CORE_H
+#define XTENSA_CONFIG_CORE_H
+
+#include <xtensa/hal.h>
+
+
+/*----------------------------------------------------------------------
+				GENERAL
+  ----------------------------------------------------------------------*/
+
+/*
+ *  Separators for macros that expand into arrays.
+ *  These can be predefined by files that #include this one,
+ *  when different separators are required.
+ */
+/*  Element separator for macros that expand into 1-dimensional arrays:  */
+#ifndef XCHAL_SEP
+#define XCHAL_SEP			,
+#endif
+/*  Array separator for macros that expand into 2-dimensional arrays:  */
+#ifndef XCHAL_SEP2
+#define XCHAL_SEP2			},{
+#endif
+
+
+/*----------------------------------------------------------------------
+				ENDIANNESS
+  ----------------------------------------------------------------------*/
+
+#define XCHAL_HAVE_BE			1
+#define XCHAL_HAVE_LE			0
+#define XCHAL_MEMORY_ORDER		XTHAL_BIGENDIAN
+
+
+/*----------------------------------------------------------------------
+				REGISTER WINDOWS
+  ----------------------------------------------------------------------*/
+
+#define XCHAL_HAVE_WINDOWED		1	/* 1 if windowed registers option configured, 0 otherwise */
+#define XCHAL_NUM_AREGS			64	/* number of physical address regs */
+#define XCHAL_NUM_AREGS_LOG2		6	/* log2(XCHAL_NUM_AREGS) */
+
+
+/*----------------------------------------------------------------------
+				ADDRESS ALIGNMENT
+  ----------------------------------------------------------------------*/
+
+/*  These apply to a selected set of core load and store instructions only (see ISA):  */
+#define XCHAL_UNALIGNED_LOAD_EXCEPTION	1	/* 1 if unaligned loads cause an exception, 0 otherwise */
+#define XCHAL_UNALIGNED_STORE_EXCEPTION	1	/* 1 if unaligned stores cause an exception, 0 otherwise */
+
+
+/*----------------------------------------------------------------------
+				INTERRUPTS
+  ----------------------------------------------------------------------*/
+
+#define XCHAL_HAVE_INTERRUPTS		1	/* 1 if interrupt option configured, 0 otherwise */
+#define XCHAL_HAVE_HIGHPRI_INTERRUPTS	1	/* 1 if high-priority interrupt option configured, 0 otherwise */
+#define XCHAL_HAVE_HIGHLEVEL_INTERRUPTS	XCHAL_HAVE_HIGHPRI_INTERRUPTS
+#define XCHAL_HAVE_NMI			0	/* 1 if NMI option configured, 0 otherwise */
+#define XCHAL_NUM_INTERRUPTS		17	/* number of interrupts */
+#define XCHAL_NUM_INTERRUPTS_LOG2	5	/* number of bits to hold an interrupt number: roundup(log2(number of interrupts)) */
+#define XCHAL_NUM_EXTINTERRUPTS		10	/* number of external interrupts */
+#define XCHAL_NUM_INTLEVELS		4	/* number of interrupt levels (not including level zero!) */
+#define XCHAL_NUM_LOWPRI_LEVELS		1			/* number of low-priority interrupt levels (always 1) */
+#define XCHAL_FIRST_HIGHPRI_LEVEL	(XCHAL_NUM_LOWPRI_LEVELS+1)	/* level of first high-priority interrupt (always 2) */
+#define XCHAL_EXCM_LEVEL		1			/* level of interrupts masked by PS.EXCM (XEA2 only; always 1 in T10xx);
+								   for XEA1, where there is no PS.EXCM, this is always 1;
+								   interrupts at levels FIRST_HIGHPRI <= n <= EXCM_LEVEL, if any,
+								   are termed "medium priority" interrupts (post T10xx only) */
+/*  Note:  1 <= LOWPRI_LEVELS <= EXCM_LEVEL < DEBUGLEVEL <= NUM_INTLEVELS < NMILEVEL <= 15  */
+
+/*  Masks of interrupts at each interrupt level:  */
+#define XCHAL_INTLEVEL0_MASK		0x00000000
+#define XCHAL_INTLEVEL1_MASK		0x000064F9
+#define XCHAL_INTLEVEL2_MASK		0x00008902
+#define XCHAL_INTLEVEL3_MASK		0x00011204
+#define XCHAL_INTLEVEL4_MASK		0x00000000
+#define XCHAL_INTLEVEL5_MASK		0x00000000
+#define XCHAL_INTLEVEL6_MASK		0x00000000
+#define XCHAL_INTLEVEL7_MASK		0x00000000
+#define XCHAL_INTLEVEL8_MASK		0x00000000
+#define XCHAL_INTLEVEL9_MASK		0x00000000
+#define XCHAL_INTLEVEL10_MASK		0x00000000
+#define XCHAL_INTLEVEL11_MASK		0x00000000
+#define XCHAL_INTLEVEL12_MASK		0x00000000
+#define XCHAL_INTLEVEL13_MASK		0x00000000
+#define XCHAL_INTLEVEL14_MASK		0x00000000
+#define XCHAL_INTLEVEL15_MASK		0x00000000
+/*  As an array of entries (eg. for C constant arrays):  */
+#define XCHAL_INTLEVEL_MASKS		0x00000000	XCHAL_SEP \
+					0x000064F9	XCHAL_SEP \
+					0x00008902	XCHAL_SEP \
+					0x00011204	XCHAL_SEP \
+					0x00000000	XCHAL_SEP \
+					0x00000000	XCHAL_SEP \
+					0x00000000	XCHAL_SEP \
+					0x00000000	XCHAL_SEP \
+					0x00000000	XCHAL_SEP \
+					0x00000000	XCHAL_SEP \
+					0x00000000	XCHAL_SEP \
+					0x00000000	XCHAL_SEP \
+					0x00000000	XCHAL_SEP \
+					0x00000000	XCHAL_SEP \
+					0x00000000	XCHAL_SEP \
+					0x00000000
+
+/*  Masks of interrupts at each range 1..n of interrupt levels:  */
+#define XCHAL_INTLEVEL0_ANDBELOW_MASK	0x00000000
+#define XCHAL_INTLEVEL1_ANDBELOW_MASK	0x000064F9
+#define XCHAL_INTLEVEL2_ANDBELOW_MASK	0x0000EDFB
+#define XCHAL_INTLEVEL3_ANDBELOW_MASK	0x0001FFFF
+#define XCHAL_INTLEVEL4_ANDBELOW_MASK	0x0001FFFF
+#define XCHAL_INTLEVEL5_ANDBELOW_MASK	0x0001FFFF
+#define XCHAL_INTLEVEL6_ANDBELOW_MASK	0x0001FFFF
+#define XCHAL_INTLEVEL7_ANDBELOW_MASK	0x0001FFFF
+#define XCHAL_INTLEVEL8_ANDBELOW_MASK	0x0001FFFF
+#define XCHAL_INTLEVEL9_ANDBELOW_MASK	0x0001FFFF
+#define XCHAL_INTLEVEL10_ANDBELOW_MASK	0x0001FFFF
+#define XCHAL_INTLEVEL11_ANDBELOW_MASK	0x0001FFFF
+#define XCHAL_INTLEVEL12_ANDBELOW_MASK	0x0001FFFF
+#define XCHAL_INTLEVEL13_ANDBELOW_MASK	0x0001FFFF
+#define XCHAL_INTLEVEL14_ANDBELOW_MASK	0x0001FFFF
+#define XCHAL_INTLEVEL15_ANDBELOW_MASK	0x0001FFFF
+#define XCHAL_LOWPRI_MASK		XCHAL_INTLEVEL1_ANDBELOW_MASK	/* mask of all low-priority interrupts */
+#define XCHAL_EXCM_MASK			XCHAL_INTLEVEL1_ANDBELOW_MASK	/* mask of all interrupts masked by PS.EXCM (or CEXCM) */
+/*  As an array of entries (eg. for C constant arrays):  */
+#define XCHAL_INTLEVEL_ANDBELOW_MASKS	0x00000000	XCHAL_SEP \
+					0x000064F9	XCHAL_SEP \
+					0x0000EDFB	XCHAL_SEP \
+					0x0001FFFF	XCHAL_SEP \
+					0x0001FFFF	XCHAL_SEP \
+					0x0001FFFF	XCHAL_SEP \
+					0x0001FFFF	XCHAL_SEP \
+					0x0001FFFF	XCHAL_SEP \
+					0x0001FFFF	XCHAL_SEP \
+					0x0001FFFF	XCHAL_SEP \
+					0x0001FFFF	XCHAL_SEP \
+					0x0001FFFF	XCHAL_SEP \
+					0x0001FFFF	XCHAL_SEP \
+					0x0001FFFF	XCHAL_SEP \
+					0x0001FFFF	XCHAL_SEP \
+					0x0001FFFF
+
+/*  Interrupt numbers for each interrupt level at which only one interrupt was configured:  */
+/*#define XCHAL_INTLEVEL1_NUM		...more than one interrupt at this level...*/
+/*#define XCHAL_INTLEVEL2_NUM		...more than one interrupt at this level...*/
+/*#define XCHAL_INTLEVEL3_NUM		...more than one interrupt at this level...*/
+
+/*  Level of each interrupt:  */
+#define XCHAL_INT0_LEVEL		1
+#define XCHAL_INT1_LEVEL		2
+#define XCHAL_INT2_LEVEL		3
+#define XCHAL_INT3_LEVEL		1
+#define XCHAL_INT4_LEVEL		1
+#define XCHAL_INT5_LEVEL		1
+#define XCHAL_INT6_LEVEL		1
+#define XCHAL_INT7_LEVEL		1
+#define XCHAL_INT8_LEVEL		2
+#define XCHAL_INT9_LEVEL		3
+#define XCHAL_INT10_LEVEL		1
+#define XCHAL_INT11_LEVEL		2
+#define XCHAL_INT12_LEVEL		3
+#define XCHAL_INT13_LEVEL		1
+#define XCHAL_INT14_LEVEL		1
+#define XCHAL_INT15_LEVEL		2
+#define XCHAL_INT16_LEVEL		3
+#define XCHAL_INT17_LEVEL		0
+#define XCHAL_INT18_LEVEL		0
+#define XCHAL_INT19_LEVEL		0
+#define XCHAL_INT20_LEVEL		0
+#define XCHAL_INT21_LEVEL		0
+#define XCHAL_INT22_LEVEL		0
+#define XCHAL_INT23_LEVEL		0
+#define XCHAL_INT24_LEVEL		0
+#define XCHAL_INT25_LEVEL		0
+#define XCHAL_INT26_LEVEL		0
+#define XCHAL_INT27_LEVEL		0
+#define XCHAL_INT28_LEVEL		0
+#define XCHAL_INT29_LEVEL		0
+#define XCHAL_INT30_LEVEL		0
+#define XCHAL_INT31_LEVEL		0
+/*  As an array of entries (eg. for C constant arrays):  */
+#define XCHAL_INT_LEVELS		1	XCHAL_SEP \
+					2	XCHAL_SEP \
+					3	XCHAL_SEP \
+					1	XCHAL_SEP \
+					1	XCHAL_SEP \
+					1	XCHAL_SEP \
+					1	XCHAL_SEP \
+					1	XCHAL_SEP \
+					2	XCHAL_SEP \
+					3	XCHAL_SEP \
+					1	XCHAL_SEP \
+					2	XCHAL_SEP \
+					3	XCHAL_SEP \
+					1	XCHAL_SEP \
+					1	XCHAL_SEP \
+					2	XCHAL_SEP \
+					3	XCHAL_SEP \
+					0	XCHAL_SEP \
+					0	XCHAL_SEP \
+					0	XCHAL_SEP \
+					0	XCHAL_SEP \
+					0	XCHAL_SEP \
+					0	XCHAL_SEP \
+					0	XCHAL_SEP \
+					0	XCHAL_SEP \
+					0	XCHAL_SEP \
+					0	XCHAL_SEP \
+					0	XCHAL_SEP \
+					0	XCHAL_SEP \
+					0	XCHAL_SEP \
+					0	XCHAL_SEP \
+					0
+
+/*  Type of each interrupt:  */
+#define XCHAL_INT0_TYPE 	XTHAL_INTTYPE_EXTERN_LEVEL
+#define XCHAL_INT1_TYPE 	XTHAL_INTTYPE_EXTERN_LEVEL
+#define XCHAL_INT2_TYPE 	XTHAL_INTTYPE_EXTERN_LEVEL
+#define XCHAL_INT3_TYPE 	XTHAL_INTTYPE_EXTERN_LEVEL
+#define XCHAL_INT4_TYPE 	XTHAL_INTTYPE_EXTERN_LEVEL
+#define XCHAL_INT5_TYPE 	XTHAL_INTTYPE_EXTERN_LEVEL
+#define XCHAL_INT6_TYPE 	XTHAL_INTTYPE_EXTERN_LEVEL
+#define XCHAL_INT7_TYPE 	XTHAL_INTTYPE_EXTERN_EDGE
+#define XCHAL_INT8_TYPE 	XTHAL_INTTYPE_EXTERN_EDGE
+#define XCHAL_INT9_TYPE 	XTHAL_INTTYPE_EXTERN_EDGE
+#define XCHAL_INT10_TYPE 	XTHAL_INTTYPE_TIMER
+#define XCHAL_INT11_TYPE 	XTHAL_INTTYPE_TIMER
+#define XCHAL_INT12_TYPE 	XTHAL_INTTYPE_TIMER
+#define XCHAL_INT13_TYPE 	XTHAL_INTTYPE_SOFTWARE
+#define XCHAL_INT14_TYPE 	XTHAL_INTTYPE_SOFTWARE
+#define XCHAL_INT15_TYPE 	XTHAL_INTTYPE_SOFTWARE
+#define XCHAL_INT16_TYPE 	XTHAL_INTTYPE_SOFTWARE
+#define XCHAL_INT17_TYPE 	XTHAL_INTTYPE_UNCONFIGURED
+#define XCHAL_INT18_TYPE 	XTHAL_INTTYPE_UNCONFIGURED
+#define XCHAL_INT19_TYPE 	XTHAL_INTTYPE_UNCONFIGURED
+#define XCHAL_INT20_TYPE 	XTHAL_INTTYPE_UNCONFIGURED
+#define XCHAL_INT21_TYPE 	XTHAL_INTTYPE_UNCONFIGURED
+#define XCHAL_INT22_TYPE 	XTHAL_INTTYPE_UNCONFIGURED
+#define XCHAL_INT23_TYPE 	XTHAL_INTTYPE_UNCONFIGURED
+#define XCHAL_INT24_TYPE 	XTHAL_INTTYPE_UNCONFIGURED
+#define XCHAL_INT25_TYPE 	XTHAL_INTTYPE_UNCONFIGURED
+#define XCHAL_INT26_TYPE 	XTHAL_INTTYPE_UNCONFIGURED
+#define XCHAL_INT27_TYPE 	XTHAL_INTTYPE_UNCONFIGURED
+#define XCHAL_INT28_TYPE 	XTHAL_INTTYPE_UNCONFIGURED
+#define XCHAL_INT29_TYPE 	XTHAL_INTTYPE_UNCONFIGURED
+#define XCHAL_INT30_TYPE 	XTHAL_INTTYPE_UNCONFIGURED
+#define XCHAL_INT31_TYPE 	XTHAL_INTTYPE_UNCONFIGURED
+/*  As an array of entries (eg. for C constant arrays):  */
+#define XCHAL_INT_TYPES		XTHAL_INTTYPE_EXTERN_LEVEL     	XCHAL_SEP \
+				XTHAL_INTTYPE_EXTERN_LEVEL     	XCHAL_SEP \
+				XTHAL_INTTYPE_EXTERN_LEVEL     	XCHAL_SEP \
+				XTHAL_INTTYPE_EXTERN_LEVEL     	XCHAL_SEP \
+				XTHAL_INTTYPE_EXTERN_LEVEL     	XCHAL_SEP \
+				XTHAL_INTTYPE_EXTERN_LEVEL     	XCHAL_SEP \
+				XTHAL_INTTYPE_EXTERN_LEVEL     	XCHAL_SEP \
+				XTHAL_INTTYPE_EXTERN_EDGE     	XCHAL_SEP \
+				XTHAL_INTTYPE_EXTERN_EDGE     	XCHAL_SEP \
+				XTHAL_INTTYPE_EXTERN_EDGE     	XCHAL_SEP \
+				XTHAL_INTTYPE_TIMER     	XCHAL_SEP \
+				XTHAL_INTTYPE_TIMER     	XCHAL_SEP \
+				XTHAL_INTTYPE_TIMER     	XCHAL_SEP \
+				XTHAL_INTTYPE_SOFTWARE     	XCHAL_SEP \
+				XTHAL_INTTYPE_SOFTWARE     	XCHAL_SEP \
+				XTHAL_INTTYPE_SOFTWARE     	XCHAL_SEP \
+				XTHAL_INTTYPE_SOFTWARE     	XCHAL_SEP \
+				XTHAL_INTTYPE_UNCONFIGURED     	XCHAL_SEP \
+				XTHAL_INTTYPE_UNCONFIGURED     	XCHAL_SEP \
+				XTHAL_INTTYPE_UNCONFIGURED     	XCHAL_SEP \
+				XTHAL_INTTYPE_UNCONFIGURED     	XCHAL_SEP \
+				XTHAL_INTTYPE_UNCONFIGURED     	XCHAL_SEP \
+				XTHAL_INTTYPE_UNCONFIGURED     	XCHAL_SEP \
+				XTHAL_INTTYPE_UNCONFIGURED     	XCHAL_SEP \
+				XTHAL_INTTYPE_UNCONFIGURED     	XCHAL_SEP \
+				XTHAL_INTTYPE_UNCONFIGURED     	XCHAL_SEP \
+				XTHAL_INTTYPE_UNCONFIGURED     	XCHAL_SEP \
+				XTHAL_INTTYPE_UNCONFIGURED     	XCHAL_SEP \
+				XTHAL_INTTYPE_UNCONFIGURED     	XCHAL_SEP \
+				XTHAL_INTTYPE_UNCONFIGURED     	XCHAL_SEP \
+				XTHAL_INTTYPE_UNCONFIGURED     	XCHAL_SEP \
+				XTHAL_INTTYPE_UNCONFIGURED
+
+/*  Masks of interrupts for each type of interrupt:  */
+#define XCHAL_INTTYPE_MASK_UNCONFIGURED	0xFFFE0000
+#define XCHAL_INTTYPE_MASK_SOFTWARE	0x0001E000
+#define XCHAL_INTTYPE_MASK_EXTERN_EDGE	0x00000380
+#define XCHAL_INTTYPE_MASK_EXTERN_LEVEL	0x0000007F
+#define XCHAL_INTTYPE_MASK_TIMER	0x00001C00
+#define XCHAL_INTTYPE_MASK_NMI		0x00000000
+/*  As an array of entries (eg. for C constant arrays):  */
+#define XCHAL_INTTYPE_MASKS		0xFFFE0000	XCHAL_SEP \
+					0x0001E000	XCHAL_SEP \
+					0x00000380	XCHAL_SEP \
+					0x0000007F	XCHAL_SEP \
+					0x00001C00	XCHAL_SEP \
+					0x00000000
+
+/*  Interrupts assigned to each timer (CCOMPARE0 to CCOMPARE3), -1 if unassigned  */
+#define XCHAL_TIMER0_INTERRUPT	10
+#define XCHAL_TIMER1_INTERRUPT	11
+#define XCHAL_TIMER2_INTERRUPT	12
+#define XCHAL_TIMER3_INTERRUPT	XTHAL_TIMER_UNCONFIGURED
+/*  As an array of entries (eg. for C constant arrays):  */
+#define XCHAL_TIMER_INTERRUPTS	10	XCHAL_SEP \
+				11	XCHAL_SEP \
+				12	XCHAL_SEP \
+				XTHAL_TIMER_UNCONFIGURED
+
+/*  Indexing macros:  */
+#define _XCHAL_INTLEVEL_MASK(n)		XCHAL_INTLEVEL ## n ## _MASK
+#define XCHAL_INTLEVEL_MASK(n)		_XCHAL_INTLEVEL_MASK(n)		/* n = 0 .. 15 */
+#define _XCHAL_INTLEVEL_ANDBELOWMASK(n)	XCHAL_INTLEVEL ## n ## _ANDBELOW_MASK
+#define XCHAL_INTLEVEL_ANDBELOW_MASK(n)	_XCHAL_INTLEVEL_ANDBELOWMASK(n)	/* n = 0 .. 15 */
+#define _XCHAL_INT_LEVEL(n)		XCHAL_INT ## n ## _LEVEL
+#define XCHAL_INT_LEVEL(n)		_XCHAL_INT_LEVEL(n)		/* n = 0 .. 31 */
+#define _XCHAL_INT_TYPE(n)		XCHAL_INT ## n ## _TYPE
+#define XCHAL_INT_TYPE(n)		_XCHAL_INT_TYPE(n)		/* n = 0 .. 31 */
+#define _XCHAL_TIMER_INTERRUPT(n)	XCHAL_TIMER ## n ## _INTERRUPT
+#define XCHAL_TIMER_INTERRUPT(n)	_XCHAL_TIMER_INTERRUPT(n)	/* n = 0 .. 3 */
+
+
+
+/*
+ *  External interrupt vectors/levels.
+ *  These macros describe how Xtensa processor interrupt numbers
+ *  (as numbered internally, eg. in INTERRUPT and INTENABLE registers)
+ *  map to external BInterrupt<n> pins, for those interrupts
+ *  configured as external (level-triggered, edge-triggered, or NMI).
+ *  See the Xtensa processor databook for more details.
+ */
+
+/*  Core interrupt numbers mapped to each EXTERNAL interrupt number:  */
+#define XCHAL_EXTINT0_NUM		0	/* (intlevel 1) */
+#define XCHAL_EXTINT1_NUM		1	/* (intlevel 2) */
+#define XCHAL_EXTINT2_NUM		2	/* (intlevel 3) */
+#define XCHAL_EXTINT3_NUM		3	/* (intlevel 1) */
+#define XCHAL_EXTINT4_NUM		4	/* (intlevel 1) */
+#define XCHAL_EXTINT5_NUM		5	/* (intlevel 1) */
+#define XCHAL_EXTINT6_NUM		6	/* (intlevel 1) */
+#define XCHAL_EXTINT7_NUM		7	/* (intlevel 1) */
+#define XCHAL_EXTINT8_NUM		8	/* (intlevel 2) */
+#define XCHAL_EXTINT9_NUM		9	/* (intlevel 3) */
+
+/*  Corresponding interrupt masks:  */
+#define XCHAL_EXTINT0_MASK		0x00000001
+#define XCHAL_EXTINT1_MASK		0x00000002
+#define XCHAL_EXTINT2_MASK		0x00000004
+#define XCHAL_EXTINT3_MASK		0x00000008
+#define XCHAL_EXTINT4_MASK		0x00000010
+#define XCHAL_EXTINT5_MASK		0x00000020
+#define XCHAL_EXTINT6_MASK		0x00000040
+#define XCHAL_EXTINT7_MASK		0x00000080
+#define XCHAL_EXTINT8_MASK		0x00000100
+#define XCHAL_EXTINT9_MASK		0x00000200
+
+/*  Core config interrupt levels mapped to each external interrupt:  */
+#define XCHAL_EXTINT0_LEVEL		1	/* (int number 0) */
+#define XCHAL_EXTINT1_LEVEL		2	/* (int number 1) */
+#define XCHAL_EXTINT2_LEVEL		3	/* (int number 2) */
+#define XCHAL_EXTINT3_LEVEL		1	/* (int number 3) */
+#define XCHAL_EXTINT4_LEVEL		1	/* (int number 4) */
+#define XCHAL_EXTINT5_LEVEL		1	/* (int number 5) */
+#define XCHAL_EXTINT6_LEVEL		1	/* (int number 6) */
+#define XCHAL_EXTINT7_LEVEL		1	/* (int number 7) */
+#define XCHAL_EXTINT8_LEVEL		2	/* (int number 8) */
+#define XCHAL_EXTINT9_LEVEL		3	/* (int number 9) */
+
+
+/*----------------------------------------------------------------------
+			EXCEPTIONS and VECTORS
+  ----------------------------------------------------------------------*/
+
+#define XCHAL_HAVE_EXCEPTIONS		1	/* 1 if exception option configured, 0 otherwise */
+
+#define XCHAL_XEA_VERSION		2	/* Xtensa Exception Architecture number: 1 for XEA1 (old), 2 for XEA2 (new) */
+#define XCHAL_HAVE_XEA1			0	/* 1 if XEA1, 0 otherwise */
+#define XCHAL_HAVE_XEA2			1	/* 1 if XEA2, 0 otherwise */
+/*  For backward compatibility ONLY -- DO NOT USE (will be removed in future release):  */
+#define XCHAL_HAVE_OLD_EXC_ARCH		XCHAL_HAVE_XEA1	/* (DEPRECATED) 1 if old exception architecture (XEA1), 0 otherwise (eg. XEA2) */
+#define XCHAL_HAVE_EXCM			XCHAL_HAVE_XEA2	/* (DEPRECATED) 1 if PS.EXCM bit exists (currently equals XCHAL_HAVE_TLBS) */
+
+#define XCHAL_RESET_VECTOR_VADDR	0xFE000020
+#define XCHAL_RESET_VECTOR_PADDR	0xFE000020
+#define XCHAL_USER_VECTOR_VADDR		0xD0000220
+#define XCHAL_PROGRAMEXC_VECTOR_VADDR	XCHAL_USER_VECTOR_VADDR		/* for backward compatibility */
+#define XCHAL_USEREXC_VECTOR_VADDR	XCHAL_USER_VECTOR_VADDR		/* for backward compatibility */
+#define XCHAL_USER_VECTOR_PADDR		0x00000220
+#define XCHAL_PROGRAMEXC_VECTOR_PADDR	XCHAL_USER_VECTOR_PADDR		/* for backward compatibility */
+#define XCHAL_USEREXC_VECTOR_PADDR	XCHAL_USER_VECTOR_PADDR		/* for backward compatibility */
+#define XCHAL_KERNEL_VECTOR_VADDR	0xD0000200
+#define XCHAL_STACKEDEXC_VECTOR_VADDR	XCHAL_KERNEL_VECTOR_VADDR	/* for backward compatibility */
+#define XCHAL_KERNELEXC_VECTOR_VADDR	XCHAL_KERNEL_VECTOR_VADDR	/* for backward compatibility */
+#define XCHAL_KERNEL_VECTOR_PADDR	0x00000200
+#define XCHAL_STACKEDEXC_VECTOR_PADDR	XCHAL_KERNEL_VECTOR_PADDR	/* for backward compatibility */
+#define XCHAL_KERNELEXC_VECTOR_PADDR	XCHAL_KERNEL_VECTOR_PADDR	/* for backward compatibility */
+#define XCHAL_DOUBLEEXC_VECTOR_VADDR	0xD0000290
+#define XCHAL_DOUBLEEXC_VECTOR_PADDR	0x00000290
+#define XCHAL_WINDOW_VECTORS_VADDR	0xD0000000
+#define XCHAL_WINDOW_VECTORS_PADDR	0x00000000
+#define XCHAL_INTLEVEL2_VECTOR_VADDR	0xD0000240
+#define XCHAL_INTLEVEL2_VECTOR_PADDR	0x00000240
+#define XCHAL_INTLEVEL3_VECTOR_VADDR	0xD0000250
+#define XCHAL_INTLEVEL3_VECTOR_PADDR	0x00000250
+#define XCHAL_INTLEVEL4_VECTOR_VADDR	0xFE000520
+#define XCHAL_INTLEVEL4_VECTOR_PADDR	0xFE000520
+#define XCHAL_DEBUG_VECTOR_VADDR	XCHAL_INTLEVEL4_VECTOR_VADDR
+#define XCHAL_DEBUG_VECTOR_PADDR	XCHAL_INTLEVEL4_VECTOR_PADDR
+
+/*  Indexing macros:  */
+#define _XCHAL_INTLEVEL_VECTOR_VADDR(n)		XCHAL_INTLEVEL ## n ## _VECTOR_VADDR
+#define XCHAL_INTLEVEL_VECTOR_VADDR(n)		_XCHAL_INTLEVEL_VECTOR_VADDR(n)		/* n = 0 .. 15 */
+
+/*
+ *  General Exception Causes
+ *  (values of EXCCAUSE special register set by general exceptions,
+ *   which vector to the user, kernel, or double-exception vectors):
+ */
+#define XCHAL_EXCCAUSE_ILLEGAL_INSTRUCTION		0	/* Illegal Instruction (IllegalInstruction) */
+#define XCHAL_EXCCAUSE_SYSTEM_CALL			1	/* System Call (SystemCall) */
+#define XCHAL_EXCCAUSE_INSTRUCTION_FETCH_ERROR		2	/* Instruction Fetch Error (InstructionFetchError) */
+#define XCHAL_EXCCAUSE_LOAD_STORE_ERROR			3	/* Load Store Error (LoadStoreError) */
+#define XCHAL_EXCCAUSE_LEVEL1_INTERRUPT			4	/* Level 1 Interrupt (Level1Interrupt) */
+#define XCHAL_EXCCAUSE_ALLOCA				5	/* Stack Extension Assist (Alloca) */
+#define XCHAL_EXCCAUSE_INTEGER_DIVIDE_BY_ZERO		6	/* Integer Divide by Zero (IntegerDivideByZero) */
+#define XCHAL_EXCCAUSE_SPECULATION			7	/* Speculation (Speculation) */
+#define XCHAL_EXCCAUSE_PRIVILEGED			8	/* Privileged Instruction (Privileged) */
+#define XCHAL_EXCCAUSE_UNALIGNED			9	/* Unaligned Load Store (Unaligned) */
+#define XCHAL_EXCCAUSE_ITLB_MISS			16	/* ITlb Miss Exception (ITlbMiss) */
+#define XCHAL_EXCCAUSE_ITLB_MULTIHIT			17	/* ITlb Mutltihit Exception (ITlbMultihit) */
+#define XCHAL_EXCCAUSE_ITLB_PRIVILEGE			18	/* ITlb Privilege Exception (ITlbPrivilege) */
+#define XCHAL_EXCCAUSE_ITLB_SIZE_RESTRICTION		19	/* ITlb Size Restriction Exception (ITlbSizeRestriction) */
+#define XCHAL_EXCCAUSE_FETCH_CACHE_ATTRIBUTE		20	/* Fetch Cache Attribute Exception (FetchCacheAttribute) */
+#define XCHAL_EXCCAUSE_DTLB_MISS			24	/* DTlb Miss Exception (DTlbMiss) */
+#define XCHAL_EXCCAUSE_DTLB_MULTIHIT			25	/* DTlb Multihit Exception (DTlbMultihit) */
+#define XCHAL_EXCCAUSE_DTLB_PRIVILEGE			26	/* DTlb Privilege Exception (DTlbPrivilege) */
+#define XCHAL_EXCCAUSE_DTLB_SIZE_RESTRICTION		27	/* DTlb Size Restriction Exception (DTlbSizeRestriction) */
+#define XCHAL_EXCCAUSE_LOAD_CACHE_ATTRIBUTE		28	/* Load Cache Attribute Exception (LoadCacheAttribute) */
+#define XCHAL_EXCCAUSE_STORE_CACHE_ATTRIBUTE		29	/* Store Cache Attribute Exception (StoreCacheAttribute) */
+#define XCHAL_EXCCAUSE_FLOATING_POINT			40	/* Floating Point Exception (FloatingPoint) */
+
+
+
+/*----------------------------------------------------------------------
+				TIMERS
+  ----------------------------------------------------------------------*/
+
+#define XCHAL_HAVE_CCOUNT		1	/* 1 if have CCOUNT, 0 otherwise */
+/*#define XCHAL_HAVE_TIMERS		XCHAL_HAVE_CCOUNT*/
+#define XCHAL_NUM_TIMERS		3	/* number of CCOMPAREn regs */
+
+
+
+/*----------------------------------------------------------------------
+				DEBUG
+  ----------------------------------------------------------------------*/
+
+#define XCHAL_HAVE_DEBUG		1	/* 1 if debug option configured, 0 otherwise */
+#define XCHAL_HAVE_OCD			1	/* 1 if OnChipDebug option configured, 0 otherwise */
+#define XCHAL_NUM_IBREAK		2	/* number of IBREAKn regs */
+#define XCHAL_NUM_DBREAK		2	/* number of DBREAKn regs */
+#define XCHAL_DEBUGLEVEL		4	/* debug interrupt level */
+/*DebugExternalInterrupt		0		0|1*/
+/*DebugUseDIRArray			0		0|1*/
+
+
+
+
+/*----------------------------------------------------------------------
+			COPROCESSORS and EXTRA STATE
+  ----------------------------------------------------------------------*/
+
+#define XCHAL_HAVE_CP			0	/* 1 if coprocessor option configured (CPENABLE present) */
+#define XCHAL_CP_MAXCFG			0	/* max allowed cp id plus one (per cfg) */
+
+#include <xtensa/config/tie.h>
+
+
+
+
+/*----------------------------------------------------------------------
+			INTERNAL I/D RAM/ROMs and XLMI
+  ----------------------------------------------------------------------*/
+
+#define XCHAL_NUM_INSTROM		0	/* number of core instruction ROMs configured */
+#define XCHAL_NUM_INSTRAM		0	/* number of core instruction RAMs configured */
+#define XCHAL_NUM_DATAROM		0	/* number of core data ROMs configured */
+#define XCHAL_NUM_DATARAM		0	/* number of core data RAMs configured */
+#define XCHAL_NUM_XLMI			0	/* number of core XLMI ports configured */
+#define  XCHAL_NUM_IROM			XCHAL_NUM_INSTROM	/* (DEPRECATED) */
+#define  XCHAL_NUM_IRAM			XCHAL_NUM_INSTRAM	/* (DEPRECATED) */
+#define  XCHAL_NUM_DROM			XCHAL_NUM_DATAROM	/* (DEPRECATED) */
+#define  XCHAL_NUM_DRAM			XCHAL_NUM_DATARAM	/* (DEPRECATED) */
+
+
+
+/*----------------------------------------------------------------------
+				CACHE
+  ----------------------------------------------------------------------*/
+
+/*  Size of the cache lines in log2(bytes):  */
+#define XCHAL_ICACHE_LINEWIDTH		4
+#define XCHAL_DCACHE_LINEWIDTH		4
+/*  Size of the cache lines in bytes:  */
+#define XCHAL_ICACHE_LINESIZE		16
+#define XCHAL_DCACHE_LINESIZE		16
+/*  Max for both I-cache and D-cache (used for general alignment):  */
+#define XCHAL_CACHE_LINEWIDTH_MAX	4
+#define XCHAL_CACHE_LINESIZE_MAX	16
+
+/*  Number of cache sets in log2(lines per way):  */
+#define XCHAL_ICACHE_SETWIDTH		8
+#define XCHAL_DCACHE_SETWIDTH		8
+/*  Max for both I-cache and D-cache (used for general cache-coherency page alignment):  */
+#define XCHAL_CACHE_SETWIDTH_MAX	8
+#define XCHAL_CACHE_SETSIZE_MAX		256
+
+/*  Cache set associativity (number of ways):  */
+#define XCHAL_ICACHE_WAYS		2
+#define XCHAL_DCACHE_WAYS		2
+
+/*  Size of the caches in bytes (ways * 2^(linewidth + setwidth)):  */
+#define XCHAL_ICACHE_SIZE		8192
+#define XCHAL_DCACHE_SIZE		8192
+
+/*  Cache features:  */
+#define XCHAL_DCACHE_IS_WRITEBACK	0
+/*  Whether cache locking feature is available:  */
+#define XCHAL_ICACHE_LINE_LOCKABLE	0
+#define XCHAL_DCACHE_LINE_LOCKABLE	0
+
+/*  Number of (encoded) cache attribute bits:  */
+#define XCHAL_CA_BITS			4	/* number of bits needed to hold cache attribute encoding */
+/*  (The number of access mode bits (decoded cache attribute bits) is defined by the architecture; see xtensa/hal.h?)  */
+
+
+/*  Cache Attribute encodings -- lists of access modes for each cache attribute:  */
+#define XCHAL_FCA_LIST		XTHAL_FAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_FAM_BYPASS	XCHAL_SEP \
+				XTHAL_FAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_FAM_BYPASS	XCHAL_SEP \
+				XTHAL_FAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_FAM_CACHED	XCHAL_SEP \
+				XTHAL_FAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_FAM_CACHED	XCHAL_SEP \
+				XTHAL_FAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_FAM_CACHED	XCHAL_SEP \
+				XTHAL_FAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_FAM_CACHED	XCHAL_SEP \
+				XTHAL_FAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_FAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_FAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_FAM_EXCEPTION
+#define XCHAL_LCA_LIST		XTHAL_LAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_LAM_BYPASSG	XCHAL_SEP \
+				XTHAL_LAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_LAM_BYPASSG	XCHAL_SEP \
+				XTHAL_LAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_LAM_CACHED	XCHAL_SEP \
+				XTHAL_LAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_LAM_CACHED	XCHAL_SEP \
+				XTHAL_LAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_LAM_NACACHED	XCHAL_SEP \
+				XTHAL_LAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_LAM_NACACHED	XCHAL_SEP \
+				XTHAL_LAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_LAM_ISOLATE	XCHAL_SEP \
+				XTHAL_LAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_LAM_CACHED
+#define XCHAL_SCA_LIST		XTHAL_SAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_SAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_SAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_SAM_BYPASS	XCHAL_SEP \
+				XTHAL_SAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_SAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_SAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_SAM_WRITETHRU	XCHAL_SEP \
+				XTHAL_SAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_SAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_SAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_SAM_WRITETHRU	XCHAL_SEP \
+				XTHAL_SAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_SAM_ISOLATE	XCHAL_SEP \
+				XTHAL_SAM_EXCEPTION	XCHAL_SEP \
+				XTHAL_SAM_WRITETHRU
+
+/*  Test:
+	read/only: 0 + 1 + 2 + 4 + 5 + 6 + 8 + 9 + 10 + 12 + 14
+	read/only: 0 + 1 + 2 + 4 + 5 + 6 + 8 + 9 + 10 + 12 + 14
+	all:       0 + 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 + 9 + 10 + 11 + 12 + 13 + 14 + 15
+	fault:     0 + 2 + 4 + 6 + 8 + 10 + 12 + 14
+	r/w/x cached:
+	r/w/x dcached:
+	I-bypass:  1 + 3
+
+	load guard bit set: 1 + 3
+	load guard bit clr: 0 + 2 + 4 + 5 + 6 + 7 + 8 + 9 + 10 + 11 + 12 + 13 + 14 + 15
+	hit-cache r/w/x: 7 + 11
+
+	fams: 5
+	fams: 0 / 6 / 18 / 1 / 2
+	fams: Bypass / Isolate / Cached / Exception / NACached
+
+        MMU okay:  yes
+*/
+
+
+/*----------------------------------------------------------------------
+				MMU
+  ----------------------------------------------------------------------*/
+
+/*
+ *  General notes on MMU parameters.
+ *
+ *  Terminology:
+ *	ASID = address-space ID (acts as an "extension" of virtual addresses)
+ *	VPN  = virtual page number
+ *	PPN  = physical page number
+ *	CA   = encoded cache attribute (access modes)
+ *	TLB  = translation look-aside buffer (term is stretched somewhat here)
+ *	I    = instruction (fetch accesses)
+ *	D    = data (load and store accesses)
+ *	way  = each TLB (ITLB and DTLB) consists of a number of "ways"
+ *		that simultaneously match the virtual address of an access;
+ *		a TLB successfully translates a virtual address if exactly
+ *		one way matches the vaddr; if none match, it is a miss;
+ *		if multiple match, one gets a "multihit" exception;
+ *		each way can be independently configured in terms of number of
+ *		entries, page sizes, which fields are writable or constant, etc.
+ *	set  = group of contiguous ways with exactly identical parameters
+ *	ARF  = auto-refill; hardware services a 1st-level miss by loading a PTE
+ *		from the page table and storing it in one of the auto-refill ways;
+ *		if this PTE load also misses, a miss exception is posted for s/w.
+ *	min-wired = a "min-wired" way can be used to map a single (minimum-sized)
+ * 		page arbitrarily under program control; it has a single entry,
+ *		is non-auto-refill (some other way(s) must be auto-refill),
+ *		all its fields (VPN, PPN, ASID, CA) are all writable, and it
+ *		supports the XCHAL_MMU_MIN_PTE_PAGE_SIZE page size (a current
+ *		restriction is that this be the only page size it supports).
+ *
+ *  TLB way entries are virtually indexed.
+ *  TLB ways that support multiple page sizes:
+ *	- must have all writable VPN and PPN fields;
+ *	- can only use one page size at any given time (eg. setup at startup),
+ *	  selected by the respective ITLBCFG or DTLBCFG special register,
+ *	  whose bits n*4+3 .. n*4 index the list of page sizes for way n
+ *	  (XCHAL_xTLB_SETm_PAGESZ_LOG2_LIST for set m corresponding to way n);
+ *	  this list may be sparse for auto-refill ways because auto-refill
+ *	  ways have independent lists of supported page sizes sharing a
+ *	  common encoding with PTE entries; the encoding is the index into
+ *	  this list; unsupported sizes for a given way are zero in the list;
+ *	  selecting unsupported sizes results in undefined hardware behaviour;
+ *	- is only possible for ways 0 thru 7 (due to ITLBCFG/DTLBCFG definition).
+ */
+
+#define XCHAL_HAVE_CACHEATTR		0	/* 1 if CACHEATTR register present, 0 if TLBs present instead */
+#define XCHAL_HAVE_TLBS			1	/* 1 if TLBs present, 0 if CACHEATTR present instead */
+#define XCHAL_HAVE_MMU			XCHAL_HAVE_TLBS	/* (DEPRECATED; use XCHAL_HAVE_TLBS instead; will be removed in future release) */
+#define XCHAL_HAVE_SPANNING_WAY		0	/* 1 if single way maps entire virtual address space in I+D */
+#define XCHAL_HAVE_IDENTITY_MAP		0	/* 1 if virtual addr == physical addr always, 0 otherwise */
+#define XCHAL_HAVE_MIMIC_CACHEATTR	0	/* 1 if have MMU that mimics a CACHEATTR config (CaMMU) */
+#define XCHAL_HAVE_XLT_CACHEATTR	0	/* 1 if have MMU that mimics a CACHEATTR config, but with translation (CaXltMMU) */
+
+#define XCHAL_MMU_ASID_BITS		8	/* number of bits in ASIDs (address space IDs) */
+#define XCHAL_MMU_ASID_INVALID		0	/* ASID value indicating invalid address space */
+#define XCHAL_MMU_ASID_KERNEL		1	/* ASID value indicating kernel (ring 0) address space */
+#define XCHAL_MMU_RINGS			4	/* number of rings supported (1..4) */
+#define XCHAL_MMU_RING_BITS		2	/* number of bits needed to hold ring number */
+#define XCHAL_MMU_SR_BITS		0	/* number of size-restriction bits supported */
+#define XCHAL_MMU_CA_BITS		4	/* number of bits needed to hold cache attribute encoding */
+#define XCHAL_MMU_MAX_PTE_PAGE_SIZE	12	/* max page size in a PTE structure (log2) */
+#define XCHAL_MMU_MIN_PTE_PAGE_SIZE	12	/* min page size in a PTE structure (log2) */
+
+
+/***  Instruction TLB:  ***/
+
+#define XCHAL_ITLB_WAY_BITS		3	/* number of bits holding the ways */
+#define XCHAL_ITLB_WAYS			7	/* number of ways (n-way set-associative TLB) */
+#define XCHAL_ITLB_ARF_WAYS		4	/* number of auto-refill ways */
+#define XCHAL_ITLB_SETS			4	/* number of sets (groups of ways with identical settings) */
+
+/*  Way set to which each way belongs:  */
+#define XCHAL_ITLB_WAY0_SET		0
+#define XCHAL_ITLB_WAY1_SET		0
+#define XCHAL_ITLB_WAY2_SET		0
+#define XCHAL_ITLB_WAY3_SET		0
+#define XCHAL_ITLB_WAY4_SET		1
+#define XCHAL_ITLB_WAY5_SET		2
+#define XCHAL_ITLB_WAY6_SET		3
+
+/*  Ways sets that are used by hardware auto-refill (ARF):  */
+#define XCHAL_ITLB_ARF_SETS		1	/* number of auto-refill sets */
+#define XCHAL_ITLB_ARF_SET0		0	/* index of n'th auto-refill set */
+
+/*  Way sets that are "min-wired" (see terminology comment above):  */
+#define XCHAL_ITLB_MINWIRED_SETS	0	/* number of "min-wired" sets */
+
+
+/*  ITLB way set 0 (group of ways 0 thru 3):  */
+#define XCHAL_ITLB_SET0_WAY			0	/* index of first way in this way set */
+#define XCHAL_ITLB_SET0_WAYS			4	/* number of (contiguous) ways in this way set */
+#define XCHAL_ITLB_SET0_ENTRIES_LOG2		2	/* log2(number of entries in this way) */
+#define XCHAL_ITLB_SET0_ENTRIES			4	/* number of entries in this way (always a power of 2) */
+#define XCHAL_ITLB_SET0_ARF			1	/* 1=autorefill by h/w, 0=non-autorefill (wired/constant/static) */
+#define XCHAL_ITLB_SET0_PAGESIZES		1	/* number of supported page sizes in this way */
+#define XCHAL_ITLB_SET0_PAGESZ_BITS		0	/* number of bits to encode the page size */
+#define XCHAL_ITLB_SET0_PAGESZ_LOG2_MIN		12	/* log2(minimum supported page size) */
+#define XCHAL_ITLB_SET0_PAGESZ_LOG2_MAX		12	/* log2(maximum supported page size) */
+#define XCHAL_ITLB_SET0_PAGESZ_LOG2_LIST	12	/* list of log2(page size)s, separated by XCHAL_SEP;
+							   2^PAGESZ_BITS entries in list, unsupported entries are zero */
+#define XCHAL_ITLB_SET0_ASID_CONSTMASK		0	/* constant ASID bits; 0 if all writable */
+#define XCHAL_ITLB_SET0_VPN_CONSTMASK		0	/* constant VPN bits, not including entry index bits; 0 if all writable */
+#define XCHAL_ITLB_SET0_PPN_CONSTMASK		0	/* constant PPN bits, including entry index bits; 0 if all writable */
+#define XCHAL_ITLB_SET0_CA_CONSTMASK		0	/* constant CA bits; 0 if all writable */
+#define XCHAL_ITLB_SET0_ASID_RESET		0	/* 1 if ASID reset values defined (and all writable); 0 otherwise */
+#define XCHAL_ITLB_SET0_VPN_RESET		0	/* 1 if VPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_ITLB_SET0_PPN_RESET		0	/* 1 if PPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_ITLB_SET0_CA_RESET		0	/* 1 if CA reset values defined (and all writable); 0 otherwise */
+
+/*  ITLB way set 1 (group of ways 4 thru 4):  */
+#define XCHAL_ITLB_SET1_WAY			4	/* index of first way in this way set */
+#define XCHAL_ITLB_SET1_WAYS			1	/* number of (contiguous) ways in this way set */
+#define XCHAL_ITLB_SET1_ENTRIES_LOG2		2	/* log2(number of entries in this way) */
+#define XCHAL_ITLB_SET1_ENTRIES			4	/* number of entries in this way (always a power of 2) */
+#define XCHAL_ITLB_SET1_ARF			0	/* 1=autorefill by h/w, 0=non-autorefill (wired/constant/static) */
+#define XCHAL_ITLB_SET1_PAGESIZES		4	/* number of supported page sizes in this way */
+#define XCHAL_ITLB_SET1_PAGESZ_BITS		2	/* number of bits to encode the page size */
+#define XCHAL_ITLB_SET1_PAGESZ_LOG2_MIN		20	/* log2(minimum supported page size) */
+#define XCHAL_ITLB_SET1_PAGESZ_LOG2_MAX		26	/* log2(maximum supported page size) */
+#define XCHAL_ITLB_SET1_PAGESZ_LOG2_LIST	20 XCHAL_SEP 22 XCHAL_SEP 24 XCHAL_SEP 26	/* list of log2(page size)s, separated by XCHAL_SEP;
+							   2^PAGESZ_BITS entries in list, unsupported entries are zero */
+#define XCHAL_ITLB_SET1_ASID_CONSTMASK		0	/* constant ASID bits; 0 if all writable */
+#define XCHAL_ITLB_SET1_VPN_CONSTMASK		0	/* constant VPN bits, not including entry index bits; 0 if all writable */
+#define XCHAL_ITLB_SET1_PPN_CONSTMASK		0	/* constant PPN bits, including entry index bits; 0 if all writable */
+#define XCHAL_ITLB_SET1_CA_CONSTMASK		0	/* constant CA bits; 0 if all writable */
+#define XCHAL_ITLB_SET1_ASID_RESET		0	/* 1 if ASID reset values defined (and all writable); 0 otherwise */
+#define XCHAL_ITLB_SET1_VPN_RESET		0	/* 1 if VPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_ITLB_SET1_PPN_RESET		0	/* 1 if PPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_ITLB_SET1_CA_RESET		0	/* 1 if CA reset values defined (and all writable); 0 otherwise */
+
+/*  ITLB way set 2 (group of ways 5 thru 5):  */
+#define XCHAL_ITLB_SET2_WAY			5	/* index of first way in this way set */
+#define XCHAL_ITLB_SET2_WAYS			1	/* number of (contiguous) ways in this way set */
+#define XCHAL_ITLB_SET2_ENTRIES_LOG2		1	/* log2(number of entries in this way) */
+#define XCHAL_ITLB_SET2_ENTRIES			2	/* number of entries in this way (always a power of 2) */
+#define XCHAL_ITLB_SET2_ARF			0	/* 1=autorefill by h/w, 0=non-autorefill (wired/constant/static) */
+#define XCHAL_ITLB_SET2_PAGESIZES		1	/* number of supported page sizes in this way */
+#define XCHAL_ITLB_SET2_PAGESZ_BITS		0	/* number of bits to encode the page size */
+#define XCHAL_ITLB_SET2_PAGESZ_LOG2_MIN		27	/* log2(minimum supported page size) */
+#define XCHAL_ITLB_SET2_PAGESZ_LOG2_MAX		27	/* log2(maximum supported page size) */
+#define XCHAL_ITLB_SET2_PAGESZ_LOG2_LIST	27	/* list of log2(page size)s, separated by XCHAL_SEP;
+							   2^PAGESZ_BITS entries in list, unsupported entries are zero */
+#define XCHAL_ITLB_SET2_ASID_CONSTMASK		0xFF	/* constant ASID bits; 0 if all writable */
+#define XCHAL_ITLB_SET2_VPN_CONSTMASK		0xF0000000	/* constant VPN bits, not including entry index bits; 0 if all writable */
+#define XCHAL_ITLB_SET2_PPN_CONSTMASK		0xF8000000	/* constant PPN bits, including entry index bits; 0 if all writable */
+#define XCHAL_ITLB_SET2_CA_CONSTMASK		0x0000000F	/* constant CA bits; 0 if all writable */
+#define XCHAL_ITLB_SET2_ASID_RESET		0	/* 1 if ASID reset values defined (and all writable); 0 otherwise */
+#define XCHAL_ITLB_SET2_VPN_RESET		0	/* 1 if VPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_ITLB_SET2_PPN_RESET		0	/* 1 if PPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_ITLB_SET2_CA_RESET		0	/* 1 if CA reset values defined (and all writable); 0 otherwise */
+/*  Constant ASID values for each entry of ITLB way set 2 (because ASID_CONSTMASK is non-zero):  */
+#define XCHAL_ITLB_SET2_E0_ASID_CONST		0x01
+#define XCHAL_ITLB_SET2_E1_ASID_CONST		0x01
+/*  Constant VPN values for each entry of ITLB way set 2 (because VPN_CONSTMASK is non-zero):  */
+#define XCHAL_ITLB_SET2_E0_VPN_CONST		0xD0000000
+#define XCHAL_ITLB_SET2_E1_VPN_CONST		0xD8000000
+/*  Constant PPN values for each entry of ITLB way set 2 (because PPN_CONSTMASK is non-zero):  */
+#define XCHAL_ITLB_SET2_E0_PPN_CONST		0x00000000
+#define XCHAL_ITLB_SET2_E1_PPN_CONST		0x00000000
+/*  Constant CA values for each entry of ITLB way set 2 (because CA_CONSTMASK is non-zero):  */
+#define XCHAL_ITLB_SET2_E0_CA_CONST		0x07
+#define XCHAL_ITLB_SET2_E1_CA_CONST		0x03
+
+/*  ITLB way set 3 (group of ways 6 thru 6):  */
+#define XCHAL_ITLB_SET3_WAY			6	/* index of first way in this way set */
+#define XCHAL_ITLB_SET3_WAYS			1	/* number of (contiguous) ways in this way set */
+#define XCHAL_ITLB_SET3_ENTRIES_LOG2		1	/* log2(number of entries in this way) */
+#define XCHAL_ITLB_SET3_ENTRIES			2	/* number of entries in this way (always a power of 2) */
+#define XCHAL_ITLB_SET3_ARF			0	/* 1=autorefill by h/w, 0=non-autorefill (wired/constant/static) */
+#define XCHAL_ITLB_SET3_PAGESIZES		1	/* number of supported page sizes in this way */
+#define XCHAL_ITLB_SET3_PAGESZ_BITS		0	/* number of bits to encode the page size */
+#define XCHAL_ITLB_SET3_PAGESZ_LOG2_MIN		28	/* log2(minimum supported page size) */
+#define XCHAL_ITLB_SET3_PAGESZ_LOG2_MAX		28	/* log2(maximum supported page size) */
+#define XCHAL_ITLB_SET3_PAGESZ_LOG2_LIST	28	/* list of log2(page size)s, separated by XCHAL_SEP;
+							   2^PAGESZ_BITS entries in list, unsupported entries are zero */
+#define XCHAL_ITLB_SET3_ASID_CONSTMASK		0xFF	/* constant ASID bits; 0 if all writable */
+#define XCHAL_ITLB_SET3_VPN_CONSTMASK		0xE0000000	/* constant VPN bits, not including entry index bits; 0 if all writable */
+#define XCHAL_ITLB_SET3_PPN_CONSTMASK		0xF0000000	/* constant PPN bits, including entry index bits; 0 if all writable */
+#define XCHAL_ITLB_SET3_CA_CONSTMASK		0x0000000F	/* constant CA bits; 0 if all writable */
+#define XCHAL_ITLB_SET3_ASID_RESET		0	/* 1 if ASID reset values defined (and all writable); 0 otherwise */
+#define XCHAL_ITLB_SET3_VPN_RESET		0	/* 1 if VPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_ITLB_SET3_PPN_RESET		0	/* 1 if PPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_ITLB_SET3_CA_RESET		0	/* 1 if CA reset values defined (and all writable); 0 otherwise */
+/*  Constant ASID values for each entry of ITLB way set 3 (because ASID_CONSTMASK is non-zero):  */
+#define XCHAL_ITLB_SET3_E0_ASID_CONST		0x01
+#define XCHAL_ITLB_SET3_E1_ASID_CONST		0x01
+/*  Constant VPN values for each entry of ITLB way set 3 (because VPN_CONSTMASK is non-zero):  */
+#define XCHAL_ITLB_SET3_E0_VPN_CONST		0xE0000000
+#define XCHAL_ITLB_SET3_E1_VPN_CONST		0xF0000000
+/*  Constant PPN values for each entry of ITLB way set 3 (because PPN_CONSTMASK is non-zero):  */
+#define XCHAL_ITLB_SET3_E0_PPN_CONST		0xF0000000
+#define XCHAL_ITLB_SET3_E1_PPN_CONST		0xF0000000
+/*  Constant CA values for each entry of ITLB way set 3 (because CA_CONSTMASK is non-zero):  */
+#define XCHAL_ITLB_SET3_E0_CA_CONST		0x07
+#define XCHAL_ITLB_SET3_E1_CA_CONST		0x03
+
+/*  Indexing macros:  */
+#define _XCHAL_ITLB_SET(n,_what)	XCHAL_ITLB_SET ## n ## _what
+#define XCHAL_ITLB_SET(n,what)		_XCHAL_ITLB_SET(n, _ ## what )
+#define _XCHAL_ITLB_SET_E(n,i,_what)	XCHAL_ITLB_SET ## n ## _E ## i ## _what
+#define XCHAL_ITLB_SET_E(n,i,what)	_XCHAL_ITLB_SET_E(n,i, _ ## what )
+/*
+ *  Example use:  XCHAL_ITLB_SET(XCHAL_ITLB_ARF_SET0,ENTRIES)
+ *	to get the value of XCHAL_ITLB_SET<n>_ENTRIES where <n> is the first auto-refill set.
+ */
+
+
+/***  Data TLB:  ***/
+
+#define XCHAL_DTLB_WAY_BITS		4	/* number of bits holding the ways */
+#define XCHAL_DTLB_WAYS			10	/* number of ways (n-way set-associative TLB) */
+#define XCHAL_DTLB_ARF_WAYS		4	/* number of auto-refill ways */
+#define XCHAL_DTLB_SETS			5	/* number of sets (groups of ways with identical settings) */
+
+/*  Way set to which each way belongs:  */
+#define XCHAL_DTLB_WAY0_SET		0
+#define XCHAL_DTLB_WAY1_SET		0
+#define XCHAL_DTLB_WAY2_SET		0
+#define XCHAL_DTLB_WAY3_SET		0
+#define XCHAL_DTLB_WAY4_SET		1
+#define XCHAL_DTLB_WAY5_SET		2
+#define XCHAL_DTLB_WAY6_SET		3
+#define XCHAL_DTLB_WAY7_SET		4
+#define XCHAL_DTLB_WAY8_SET		4
+#define XCHAL_DTLB_WAY9_SET		4
+
+/*  Ways sets that are used by hardware auto-refill (ARF):  */
+#define XCHAL_DTLB_ARF_SETS		1	/* number of auto-refill sets */
+#define XCHAL_DTLB_ARF_SET0		0	/* index of n'th auto-refill set */
+
+/*  Way sets that are "min-wired" (see terminology comment above):  */
+#define XCHAL_DTLB_MINWIRED_SETS	1	/* number of "min-wired" sets */
+#define XCHAL_DTLB_MINWIRED_SET0	4	/* index of n'th "min-wired" set */
+
+
+/*  DTLB way set 0 (group of ways 0 thru 3):  */
+#define XCHAL_DTLB_SET0_WAY			0	/* index of first way in this way set */
+#define XCHAL_DTLB_SET0_WAYS			4	/* number of (contiguous) ways in this way set */
+#define XCHAL_DTLB_SET0_ENTRIES_LOG2		2	/* log2(number of entries in this way) */
+#define XCHAL_DTLB_SET0_ENTRIES			4	/* number of entries in this way (always a power of 2) */
+#define XCHAL_DTLB_SET0_ARF			1	/* 1=autorefill by h/w, 0=non-autorefill (wired/constant/static) */
+#define XCHAL_DTLB_SET0_PAGESIZES		1	/* number of supported page sizes in this way */
+#define XCHAL_DTLB_SET0_PAGESZ_BITS		0	/* number of bits to encode the page size */
+#define XCHAL_DTLB_SET0_PAGESZ_LOG2_MIN		12	/* log2(minimum supported page size) */
+#define XCHAL_DTLB_SET0_PAGESZ_LOG2_MAX		12	/* log2(maximum supported page size) */
+#define XCHAL_DTLB_SET0_PAGESZ_LOG2_LIST	12	/* list of log2(page size)s, separated by XCHAL_SEP;
+							   2^PAGESZ_BITS entries in list, unsupported entries are zero */
+#define XCHAL_DTLB_SET0_ASID_CONSTMASK		0	/* constant ASID bits; 0 if all writable */
+#define XCHAL_DTLB_SET0_VPN_CONSTMASK		0	/* constant VPN bits, not including entry index bits; 0 if all writable */
+#define XCHAL_DTLB_SET0_PPN_CONSTMASK		0	/* constant PPN bits, including entry index bits; 0 if all writable */
+#define XCHAL_DTLB_SET0_CA_CONSTMASK		0	/* constant CA bits; 0 if all writable */
+#define XCHAL_DTLB_SET0_ASID_RESET		0	/* 1 if ASID reset values defined (and all writable); 0 otherwise */
+#define XCHAL_DTLB_SET0_VPN_RESET		0	/* 1 if VPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_DTLB_SET0_PPN_RESET		0	/* 1 if PPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_DTLB_SET0_CA_RESET		0	/* 1 if CA reset values defined (and all writable); 0 otherwise */
+
+/*  DTLB way set 1 (group of ways 4 thru 4):  */
+#define XCHAL_DTLB_SET1_WAY			4	/* index of first way in this way set */
+#define XCHAL_DTLB_SET1_WAYS			1	/* number of (contiguous) ways in this way set */
+#define XCHAL_DTLB_SET1_ENTRIES_LOG2		2	/* log2(number of entries in this way) */
+#define XCHAL_DTLB_SET1_ENTRIES			4	/* number of entries in this way (always a power of 2) */
+#define XCHAL_DTLB_SET1_ARF			0	/* 1=autorefill by h/w, 0=non-autorefill (wired/constant/static) */
+#define XCHAL_DTLB_SET1_PAGESIZES		4	/* number of supported page sizes in this way */
+#define XCHAL_DTLB_SET1_PAGESZ_BITS		2	/* number of bits to encode the page size */
+#define XCHAL_DTLB_SET1_PAGESZ_LOG2_MIN		20	/* log2(minimum supported page size) */
+#define XCHAL_DTLB_SET1_PAGESZ_LOG2_MAX		26	/* log2(maximum supported page size) */
+#define XCHAL_DTLB_SET1_PAGESZ_LOG2_LIST	20 XCHAL_SEP 22 XCHAL_SEP 24 XCHAL_SEP 26	/* list of log2(page size)s, separated by XCHAL_SEP;
+							   2^PAGESZ_BITS entries in list, unsupported entries are zero */
+#define XCHAL_DTLB_SET1_ASID_CONSTMASK		0	/* constant ASID bits; 0 if all writable */
+#define XCHAL_DTLB_SET1_VPN_CONSTMASK		0	/* constant VPN bits, not including entry index bits; 0 if all writable */
+#define XCHAL_DTLB_SET1_PPN_CONSTMASK		0	/* constant PPN bits, including entry index bits; 0 if all writable */
+#define XCHAL_DTLB_SET1_CA_CONSTMASK		0	/* constant CA bits; 0 if all writable */
+#define XCHAL_DTLB_SET1_ASID_RESET		0	/* 1 if ASID reset values defined (and all writable); 0 otherwise */
+#define XCHAL_DTLB_SET1_VPN_RESET		0	/* 1 if VPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_DTLB_SET1_PPN_RESET		0	/* 1 if PPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_DTLB_SET1_CA_RESET		0	/* 1 if CA reset values defined (and all writable); 0 otherwise */
+
+/*  DTLB way set 2 (group of ways 5 thru 5):  */
+#define XCHAL_DTLB_SET2_WAY			5	/* index of first way in this way set */
+#define XCHAL_DTLB_SET2_WAYS			1	/* number of (contiguous) ways in this way set */
+#define XCHAL_DTLB_SET2_ENTRIES_LOG2		1	/* log2(number of entries in this way) */
+#define XCHAL_DTLB_SET2_ENTRIES			2	/* number of entries in this way (always a power of 2) */
+#define XCHAL_DTLB_SET2_ARF			0	/* 1=autorefill by h/w, 0=non-autorefill (wired/constant/static) */
+#define XCHAL_DTLB_SET2_PAGESIZES		1	/* number of supported page sizes in this way */
+#define XCHAL_DTLB_SET2_PAGESZ_BITS		0	/* number of bits to encode the page size */
+#define XCHAL_DTLB_SET2_PAGESZ_LOG2_MIN		27	/* log2(minimum supported page size) */
+#define XCHAL_DTLB_SET2_PAGESZ_LOG2_MAX		27	/* log2(maximum supported page size) */
+#define XCHAL_DTLB_SET2_PAGESZ_LOG2_LIST	27	/* list of log2(page size)s, separated by XCHAL_SEP;
+							   2^PAGESZ_BITS entries in list, unsupported entries are zero */
+#define XCHAL_DTLB_SET2_ASID_CONSTMASK		0xFF	/* constant ASID bits; 0 if all writable */
+#define XCHAL_DTLB_SET2_VPN_CONSTMASK		0xF0000000	/* constant VPN bits, not including entry index bits; 0 if all writable */
+#define XCHAL_DTLB_SET2_PPN_CONSTMASK		0xF8000000	/* constant PPN bits, including entry index bits; 0 if all writable */
+#define XCHAL_DTLB_SET2_CA_CONSTMASK		0x0000000F	/* constant CA bits; 0 if all writable */
+#define XCHAL_DTLB_SET2_ASID_RESET		0	/* 1 if ASID reset values defined (and all writable); 0 otherwise */
+#define XCHAL_DTLB_SET2_VPN_RESET		0	/* 1 if VPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_DTLB_SET2_PPN_RESET		0	/* 1 if PPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_DTLB_SET2_CA_RESET		0	/* 1 if CA reset values defined (and all writable); 0 otherwise */
+/*  Constant ASID values for each entry of DTLB way set 2 (because ASID_CONSTMASK is non-zero):  */
+#define XCHAL_DTLB_SET2_E0_ASID_CONST		0x01
+#define XCHAL_DTLB_SET2_E1_ASID_CONST		0x01
+/*  Constant VPN values for each entry of DTLB way set 2 (because VPN_CONSTMASK is non-zero):  */
+#define XCHAL_DTLB_SET2_E0_VPN_CONST		0xD0000000
+#define XCHAL_DTLB_SET2_E1_VPN_CONST		0xD8000000
+/*  Constant PPN values for each entry of DTLB way set 2 (because PPN_CONSTMASK is non-zero):  */
+#define XCHAL_DTLB_SET2_E0_PPN_CONST		0x00000000
+#define XCHAL_DTLB_SET2_E1_PPN_CONST		0x00000000
+/*  Constant CA values for each entry of DTLB way set 2 (because CA_CONSTMASK is non-zero):  */
+#define XCHAL_DTLB_SET2_E0_CA_CONST		0x07
+#define XCHAL_DTLB_SET2_E1_CA_CONST		0x03
+
+/*  DTLB way set 3 (group of ways 6 thru 6):  */
+#define XCHAL_DTLB_SET3_WAY			6	/* index of first way in this way set */
+#define XCHAL_DTLB_SET3_WAYS			1	/* number of (contiguous) ways in this way set */
+#define XCHAL_DTLB_SET3_ENTRIES_LOG2		1	/* log2(number of entries in this way) */
+#define XCHAL_DTLB_SET3_ENTRIES			2	/* number of entries in this way (always a power of 2) */
+#define XCHAL_DTLB_SET3_ARF			0	/* 1=autorefill by h/w, 0=non-autorefill (wired/constant/static) */
+#define XCHAL_DTLB_SET3_PAGESIZES		1	/* number of supported page sizes in this way */
+#define XCHAL_DTLB_SET3_PAGESZ_BITS		0	/* number of bits to encode the page size */
+#define XCHAL_DTLB_SET3_PAGESZ_LOG2_MIN		28	/* log2(minimum supported page size) */
+#define XCHAL_DTLB_SET3_PAGESZ_LOG2_MAX		28	/* log2(maximum supported page size) */
+#define XCHAL_DTLB_SET3_PAGESZ_LOG2_LIST	28	/* list of log2(page size)s, separated by XCHAL_SEP;
+							   2^PAGESZ_BITS entries in list, unsupported entries are zero */
+#define XCHAL_DTLB_SET3_ASID_CONSTMASK		0xFF	/* constant ASID bits; 0 if all writable */
+#define XCHAL_DTLB_SET3_VPN_CONSTMASK		0xE0000000	/* constant VPN bits, not including entry index bits; 0 if all writable */
+#define XCHAL_DTLB_SET3_PPN_CONSTMASK		0xF0000000	/* constant PPN bits, including entry index bits; 0 if all writable */
+#define XCHAL_DTLB_SET3_CA_CONSTMASK		0x0000000F	/* constant CA bits; 0 if all writable */
+#define XCHAL_DTLB_SET3_ASID_RESET		0	/* 1 if ASID reset values defined (and all writable); 0 otherwise */
+#define XCHAL_DTLB_SET3_VPN_RESET		0	/* 1 if VPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_DTLB_SET3_PPN_RESET		0	/* 1 if PPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_DTLB_SET3_CA_RESET		0	/* 1 if CA reset values defined (and all writable); 0 otherwise */
+/*  Constant ASID values for each entry of DTLB way set 3 (because ASID_CONSTMASK is non-zero):  */
+#define XCHAL_DTLB_SET3_E0_ASID_CONST		0x01
+#define XCHAL_DTLB_SET3_E1_ASID_CONST		0x01
+/*  Constant VPN values for each entry of DTLB way set 3 (because VPN_CONSTMASK is non-zero):  */
+#define XCHAL_DTLB_SET3_E0_VPN_CONST		0xE0000000
+#define XCHAL_DTLB_SET3_E1_VPN_CONST		0xF0000000
+/*  Constant PPN values for each entry of DTLB way set 3 (because PPN_CONSTMASK is non-zero):  */
+#define XCHAL_DTLB_SET3_E0_PPN_CONST		0xF0000000
+#define XCHAL_DTLB_SET3_E1_PPN_CONST		0xF0000000
+/*  Constant CA values for each entry of DTLB way set 3 (because CA_CONSTMASK is non-zero):  */
+#define XCHAL_DTLB_SET3_E0_CA_CONST		0x07
+#define XCHAL_DTLB_SET3_E1_CA_CONST		0x03
+
+/*  DTLB way set 4 (group of ways 7 thru 9):  */
+#define XCHAL_DTLB_SET4_WAY			7	/* index of first way in this way set */
+#define XCHAL_DTLB_SET4_WAYS			3	/* number of (contiguous) ways in this way set */
+#define XCHAL_DTLB_SET4_ENTRIES_LOG2		0	/* log2(number of entries in this way) */
+#define XCHAL_DTLB_SET4_ENTRIES			1	/* number of entries in this way (always a power of 2) */
+#define XCHAL_DTLB_SET4_ARF			0	/* 1=autorefill by h/w, 0=non-autorefill (wired/constant/static) */
+#define XCHAL_DTLB_SET4_PAGESIZES		1	/* number of supported page sizes in this way */
+#define XCHAL_DTLB_SET4_PAGESZ_BITS		0	/* number of bits to encode the page size */
+#define XCHAL_DTLB_SET4_PAGESZ_LOG2_MIN		12	/* log2(minimum supported page size) */
+#define XCHAL_DTLB_SET4_PAGESZ_LOG2_MAX		12	/* log2(maximum supported page size) */
+#define XCHAL_DTLB_SET4_PAGESZ_LOG2_LIST	12	/* list of log2(page size)s, separated by XCHAL_SEP;
+							   2^PAGESZ_BITS entries in list, unsupported entries are zero */
+#define XCHAL_DTLB_SET4_ASID_CONSTMASK		0	/* constant ASID bits; 0 if all writable */
+#define XCHAL_DTLB_SET4_VPN_CONSTMASK		0	/* constant VPN bits, not including entry index bits; 0 if all writable */
+#define XCHAL_DTLB_SET4_PPN_CONSTMASK		0	/* constant PPN bits, including entry index bits; 0 if all writable */
+#define XCHAL_DTLB_SET4_CA_CONSTMASK		0	/* constant CA bits; 0 if all writable */
+#define XCHAL_DTLB_SET4_ASID_RESET		0	/* 1 if ASID reset values defined (and all writable); 0 otherwise */
+#define XCHAL_DTLB_SET4_VPN_RESET		0	/* 1 if VPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_DTLB_SET4_PPN_RESET		0	/* 1 if PPN reset values defined (and all writable); 0 otherwise */
+#define XCHAL_DTLB_SET4_CA_RESET		0	/* 1 if CA reset values defined (and all writable); 0 otherwise */
+
+/*  Indexing macros:  */
+#define _XCHAL_DTLB_SET(n,_what)	XCHAL_DTLB_SET ## n ## _what
+#define XCHAL_DTLB_SET(n,what)		_XCHAL_DTLB_SET(n, _ ## what )
+#define _XCHAL_DTLB_SET_E(n,i,_what)	XCHAL_DTLB_SET ## n ## _E ## i ## _what
+#define XCHAL_DTLB_SET_E(n,i,what)	_XCHAL_DTLB_SET_E(n,i, _ ## what )
+/*
+ *  Example use:  XCHAL_DTLB_SET(XCHAL_DTLB_ARF_SET0,ENTRIES)
+ *	to get the value of XCHAL_DTLB_SET<n>_ENTRIES where <n> is the first auto-refill set.
+ */
+
+
+/*
+ *  Determine whether we have a full MMU (with Page Table and Protection)
+ *  usable for an MMU-based OS:
+ */
+#if XCHAL_HAVE_TLBS && !XCHAL_HAVE_SPANNING_WAY && XCHAL_ITLB_ARF_WAYS > 0 && XCHAL_DTLB_ARF_WAYS > 0 && XCHAL_MMU_RINGS >= 2
+# define XCHAL_HAVE_PTP_MMU		1	/* have full MMU (with page table [autorefill] and protection) */
+#else
+# define XCHAL_HAVE_PTP_MMU		0	/* don't have full MMU */
+#endif
+
+/*
+ *  For full MMUs, report kernel RAM segment and kernel I/O segment static page mappings:
+ */
+#if XCHAL_HAVE_PTP_MMU
+#define XCHAL_KSEG_CACHED_VADDR		0xD0000000	/* virt.addr of kernel RAM cached static map */
+#define XCHAL_KSEG_CACHED_PADDR		0x00000000	/* phys.addr of kseg_cached */
+#define XCHAL_KSEG_CACHED_SIZE		0x08000000	/* size in bytes of kseg_cached (assumed power of 2!!!) */
+#define XCHAL_KSEG_BYPASS_VADDR		0xD8000000	/* virt.addr of kernel RAM bypass (uncached) static map */
+#define XCHAL_KSEG_BYPASS_PADDR		0x00000000	/* phys.addr of kseg_bypass */
+#define XCHAL_KSEG_BYPASS_SIZE		0x08000000	/* size in bytes of kseg_bypass (assumed power of 2!!!) */
+
+#define XCHAL_KIO_CACHED_VADDR		0xE0000000	/* virt.addr of kernel I/O cached static map */
+#define XCHAL_KIO_CACHED_PADDR		0xF0000000	/* phys.addr of kio_cached */
+#define XCHAL_KIO_CACHED_SIZE		0x10000000	/* size in bytes of kio_cached (assumed power of 2!!!) */
+#define XCHAL_KIO_BYPASS_VADDR		0xF0000000	/* virt.addr of kernel I/O bypass (uncached) static map */
+#define XCHAL_KIO_BYPASS_PADDR		0xF0000000	/* phys.addr of kio_bypass */
+#define XCHAL_KIO_BYPASS_SIZE		0x10000000	/* size in bytes of kio_bypass (assumed power of 2!!!) */
+
+#define XCHAL_SEG_MAPPABLE_VADDR	0x00000000	/* start of largest non-static-mapped virtual addr area */
+#define XCHAL_SEG_MAPPABLE_SIZE		0xD0000000	/* size in bytes of  "  */
+/* define XCHAL_SEG_MAPPABLE2_xxx if more areas present, sorted in order of descending size.  */
+#endif
+
+
+/*----------------------------------------------------------------------
+				MISC
+  ----------------------------------------------------------------------*/
+
+#define XCHAL_NUM_WRITEBUFFER_ENTRIES	4	/* number of write buffer entries */
+
+#define XCHAL_CORE_ID			"linux_be"	/* configuration's alphanumeric core identifier
+							   (CoreID) set in the Xtensa Processor Generator */
+
+#define XCHAL_BUILD_UNIQUE_ID		0x00003256	/* software build-unique ID (22-bit) */
+
+/*  These definitions describe the hardware targeted by this software:  */
+#define XCHAL_HW_CONFIGID0		0xC103D1FF	/* config ID reg 0 value (upper 32 of 64 bits) */
+#define XCHAL_HW_CONFIGID1		0x00803256	/* config ID reg 1 value (lower 32 of 64 bits) */
+#define XCHAL_CONFIGID0			XCHAL_HW_CONFIGID0	/* for backward compatibility only -- don't use! */
+#define XCHAL_CONFIGID1			XCHAL_HW_CONFIGID1	/* for backward compatibility only -- don't use! */
+#define XCHAL_HW_RELEASE_MAJOR		1050	/* major release of targeted hardware */
+#define XCHAL_HW_RELEASE_MINOR		1	/* minor release of targeted hardware */
+#define XCHAL_HW_RELEASE_NAME		"T1050.1"	/* full release name of targeted hardware */
+#define XTHAL_HW_REL_T1050	1
+#define XTHAL_HW_REL_T1050_1	1
+#define XCHAL_HW_CONFIGID_RELIABLE	1
+
+
+/*
+ *  Miscellaneous special register fields:
+ */
+
+
+/*  DBREAKC (special register number 160):  */
+#define XCHAL_DBREAKC_VALIDMASK	0xC000003F	/* bits of DBREAKC that are defined */
+/*  MASK field:  */
+#define XCHAL_DBREAKC_MASK_BITS 	6		/* number of bits in MASK field */
+#define XCHAL_DBREAKC_MASK_NUM  	64		/* max number of possible causes (2^bits) */
+#define XCHAL_DBREAKC_MASK_SHIFT	0		/* position of MASK bits in DBREAKC, starting from lsbit */
+#define XCHAL_DBREAKC_MASK_MASK 	0x0000003F	/* mask of bits in MASK field of DBREAKC */
+/*  LOADBREAK field:  */
+#define XCHAL_DBREAKC_LOADBREAK_BITS 	1		/* number of bits in LOADBREAK field */
+#define XCHAL_DBREAKC_LOADBREAK_NUM  	2		/* max number of possible causes (2^bits) */
+#define XCHAL_DBREAKC_LOADBREAK_SHIFT	30		/* position of LOADBREAK bits in DBREAKC, starting from lsbit */
+#define XCHAL_DBREAKC_LOADBREAK_MASK 	0x40000000	/* mask of bits in LOADBREAK field of DBREAKC */
+/*  STOREBREAK field:  */
+#define XCHAL_DBREAKC_STOREBREAK_BITS 	1		/* number of bits in STOREBREAK field */
+#define XCHAL_DBREAKC_STOREBREAK_NUM  	2		/* max number of possible causes (2^bits) */
+#define XCHAL_DBREAKC_STOREBREAK_SHIFT	31		/* position of STOREBREAK bits in DBREAKC, starting from lsbit */
+#define XCHAL_DBREAKC_STOREBREAK_MASK 	0x80000000	/* mask of bits in STOREBREAK field of DBREAKC */
+
+/*  PS (special register number 230):  */
+#define XCHAL_PS_VALIDMASK	0x00070FFF	/* bits of PS that are defined */
+/*  INTLEVEL field:  */
+#define XCHAL_PS_INTLEVEL_BITS 	4		/* number of bits in INTLEVEL field */
+#define XCHAL_PS_INTLEVEL_NUM  	16		/* max number of possible causes (2^bits) */
+#define XCHAL_PS_INTLEVEL_SHIFT	0		/* position of INTLEVEL bits in PS, starting from lsbit */
+#define XCHAL_PS_INTLEVEL_MASK 	0x0000000F	/* mask of bits in INTLEVEL field of PS */
+/*  EXCM field:  */
+#define XCHAL_PS_EXCM_BITS 	1		/* number of bits in EXCM field */
+#define XCHAL_PS_EXCM_NUM  	2		/* max number of possible causes (2^bits) */
+#define XCHAL_PS_EXCM_SHIFT	4		/* position of EXCM bits in PS, starting from lsbit */
+#define XCHAL_PS_EXCM_MASK 	0x00000010	/* mask of bits in EXCM field of PS */
+/*  PROGSTACK field:  */
+#define XCHAL_PS_PROGSTACK_BITS 	1		/* number of bits in PROGSTACK field */
+#define XCHAL_PS_PROGSTACK_NUM  	2		/* max number of possible causes (2^bits) */
+#define XCHAL_PS_PROGSTACK_SHIFT	5		/* position of PROGSTACK bits in PS, starting from lsbit */
+#define XCHAL_PS_PROGSTACK_MASK 	0x00000020	/* mask of bits in PROGSTACK field of PS */
+/*  RING field:  */
+#define XCHAL_PS_RING_BITS 	2		/* number of bits in RING field */
+#define XCHAL_PS_RING_NUM  	4		/* max number of possible causes (2^bits) */
+#define XCHAL_PS_RING_SHIFT	6		/* position of RING bits in PS, starting from lsbit */
+#define XCHAL_PS_RING_MASK 	0x000000C0	/* mask of bits in RING field of PS */
+/*  OWB field:  */
+#define XCHAL_PS_OWB_BITS 	4		/* number of bits in OWB field */
+#define XCHAL_PS_OWB_NUM  	16		/* max number of possible causes (2^bits) */
+#define XCHAL_PS_OWB_SHIFT	8		/* position of OWB bits in PS, starting from lsbit */
+#define XCHAL_PS_OWB_MASK 	0x00000F00	/* mask of bits in OWB field of PS */
+/*  CALLINC field:  */
+#define XCHAL_PS_CALLINC_BITS 	2		/* number of bits in CALLINC field */
+#define XCHAL_PS_CALLINC_NUM  	4		/* max number of possible causes (2^bits) */
+#define XCHAL_PS_CALLINC_SHIFT	16		/* position of CALLINC bits in PS, starting from lsbit */
+#define XCHAL_PS_CALLINC_MASK 	0x00030000	/* mask of bits in CALLINC field of PS */
+/*  WOE field:  */
+#define XCHAL_PS_WOE_BITS 	1		/* number of bits in WOE field */
+#define XCHAL_PS_WOE_NUM  	2		/* max number of possible causes (2^bits) */
+#define XCHAL_PS_WOE_SHIFT	18		/* position of WOE bits in PS, starting from lsbit */
+#define XCHAL_PS_WOE_MASK 	0x00040000	/* mask of bits in WOE field of PS */
+
+/*  EXCCAUSE (special register number 232):  */
+#define XCHAL_EXCCAUSE_VALIDMASK	0x0000003F	/* bits of EXCCAUSE that are defined */
+/*  EXCCAUSE field:  */
+#define XCHAL_EXCCAUSE_BITS 		6		/* number of bits in EXCCAUSE register */
+#define XCHAL_EXCCAUSE_NUM  		64		/* max number of possible causes (2^bits) */
+#define XCHAL_EXCCAUSE_SHIFT		0		/* position of EXCCAUSE bits in register, starting from lsbit */
+#define XCHAL_EXCCAUSE_MASK 		0x0000003F	/* mask of bits in EXCCAUSE register */
+
+/*  DEBUGCAUSE (special register number 233):  */
+#define XCHAL_DEBUGCAUSE_VALIDMASK	0x0000003F	/* bits of DEBUGCAUSE that are defined */
+/*  ICOUNT field:  */
+#define XCHAL_DEBUGCAUSE_ICOUNT_BITS 	1		/* number of bits in ICOUNT field */
+#define XCHAL_DEBUGCAUSE_ICOUNT_NUM  	2		/* max number of possible causes (2^bits) */
+#define XCHAL_DEBUGCAUSE_ICOUNT_SHIFT	0		/* position of ICOUNT bits in DEBUGCAUSE, starting from lsbit */
+#define XCHAL_DEBUGCAUSE_ICOUNT_MASK 	0x00000001	/* mask of bits in ICOUNT field of DEBUGCAUSE */
+/*  IBREAK field:  */
+#define XCHAL_DEBUGCAUSE_IBREAK_BITS 	1		/* number of bits in IBREAK field */
+#define XCHAL_DEBUGCAUSE_IBREAK_NUM  	2		/* max number of possible causes (2^bits) */
+#define XCHAL_DEBUGCAUSE_IBREAK_SHIFT	1		/* position of IBREAK bits in DEBUGCAUSE, starting from lsbit */
+#define XCHAL_DEBUGCAUSE_IBREAK_MASK 	0x00000002	/* mask of bits in IBREAK field of DEBUGCAUSE */
+/*  DBREAK field:  */
+#define XCHAL_DEBUGCAUSE_DBREAK_BITS 	1		/* number of bits in DBREAK field */
+#define XCHAL_DEBUGCAUSE_DBREAK_NUM  	2		/* max number of possible causes (2^bits) */
+#define XCHAL_DEBUGCAUSE_DBREAK_SHIFT	2		/* position of DBREAK bits in DEBUGCAUSE, starting from lsbit */
+#define XCHAL_DEBUGCAUSE_DBREAK_MASK 	0x00000004	/* mask of bits in DBREAK field of DEBUGCAUSE */
+/*  BREAK field:  */
+#define XCHAL_DEBUGCAUSE_BREAK_BITS 	1		/* number of bits in BREAK field */
+#define XCHAL_DEBUGCAUSE_BREAK_NUM  	2		/* max number of possible causes (2^bits) */
+#define XCHAL_DEBUGCAUSE_BREAK_SHIFT	3		/* position of BREAK bits in DEBUGCAUSE, starting from lsbit */
+#define XCHAL_DEBUGCAUSE_BREAK_MASK 	0x00000008	/* mask of bits in BREAK field of DEBUGCAUSE */
+/*  BREAKN field:  */
+#define XCHAL_DEBUGCAUSE_BREAKN_BITS 	1		/* number of bits in BREAKN field */
+#define XCHAL_DEBUGCAUSE_BREAKN_NUM  	2		/* max number of possible causes (2^bits) */
+#define XCHAL_DEBUGCAUSE_BREAKN_SHIFT	4		/* position of BREAKN bits in DEBUGCAUSE, starting from lsbit */
+#define XCHAL_DEBUGCAUSE_BREAKN_MASK 	0x00000010	/* mask of bits in BREAKN field of DEBUGCAUSE */
+/*  DEBUGINT field:  */
+#define XCHAL_DEBUGCAUSE_DEBUGINT_BITS 	1		/* number of bits in DEBUGINT field */
+#define XCHAL_DEBUGCAUSE_DEBUGINT_NUM  	2		/* max number of possible causes (2^bits) */
+#define XCHAL_DEBUGCAUSE_DEBUGINT_SHIFT	5		/* position of DEBUGINT bits in DEBUGCAUSE, starting from lsbit */
+#define XCHAL_DEBUGCAUSE_DEBUGINT_MASK 	0x00000020	/* mask of bits in DEBUGINT field of DEBUGCAUSE */
+
+
+
+/*----------------------------------------------------------------------
+				ISA
+  ----------------------------------------------------------------------*/
+
+#define XCHAL_HAVE_DENSITY		1	/* 1 if density option configured, 0 otherwise */
+#define XCHAL_HAVE_LOOPS		1	/* 1 if zero-overhead loops option configured, 0 otherwise */
+/*  Misc instructions:  */
+#define XCHAL_HAVE_NSA			0	/* 1 if NSA/NSAU instructions option configured, 0 otherwise */
+#define XCHAL_HAVE_MINMAX		0	/* 1 if MIN/MAX instructions option configured, 0 otherwise */
+#define XCHAL_HAVE_SEXT			0	/* 1 if sign-extend instruction option configured, 0 otherwise */
+#define XCHAL_HAVE_CLAMPS		0	/* 1 if CLAMPS instruction option configured, 0 otherwise */
+#define XCHAL_HAVE_MAC16		0	/* 1 if MAC16 option configured, 0 otherwise */
+#define XCHAL_HAVE_MUL16		0	/* 1 if 16-bit integer multiply option configured, 0 otherwise */
+/*#define XCHAL_HAVE_POPC		0*/	/* 1 if CRC instruction option configured, 0 otherwise */
+/*#define XCHAL_HAVE_CRC		0*/	/* 1 if POPC instruction option configured, 0 otherwise */
+
+#define XCHAL_HAVE_SPECULATION		0	/* 1 if speculation option configured, 0 otherwise */
+/*#define XCHAL_HAVE_MP_SYNC		0*/	/* 1 if multiprocessor sync. option configured, 0 otherwise */
+#define XCHAL_HAVE_PRID			0	/* 1 if processor ID register configured, 0 otherwise */
+
+#define XCHAL_NUM_MISC_REGS		2	/* number of miscellaneous registers (0..4) */
+
+/*  These relate a bit more to TIE:  */
+#define XCHAL_HAVE_BOOLEANS		0	/* 1 if booleans option configured, 0 otherwise */
+#define XCHAL_HAVE_MUL32		0	/* 1 if 32-bit integer multiply option configured, 0 otherwise */
+#define XCHAL_HAVE_MUL32_HIGH		0	/* 1 if MUL32 option includes MULUH and MULSH, 0 otherwise */
+#define XCHAL_HAVE_FP			0	/* 1 if floating point option configured, 0 otherwise */
+
+
+/*----------------------------------------------------------------------
+				DERIVED
+  ----------------------------------------------------------------------*/
+
+#if XCHAL_HAVE_BE
+#define XCHAL_INST_ILLN			0xD60F		/* 2-byte illegal instruction, msb-first */
+#define XCHAL_INST_ILLN_BYTE0		0xD6		/* 2-byte illegal instruction, 1st byte */
+#define XCHAL_INST_ILLN_BYTE1		0x0F		/* 2-byte illegal instruction, 2nd byte */
+#else
+#define XCHAL_INST_ILLN			0xF06D		/* 2-byte illegal instruction, lsb-first */
+#define XCHAL_INST_ILLN_BYTE0		0x6D		/* 2-byte illegal instruction, 1st byte */
+#define XCHAL_INST_ILLN_BYTE1		0xF0		/* 2-byte illegal instruction, 2nd byte */
+#endif
+/*  Belongs in xtensa/hal.h:  */
+#define XTHAL_INST_ILL			0x000000	/* 3-byte illegal instruction */
+
+
+/*
+ *  Because information as to exactly which hardware release is targeted
+ *  by a given software build is not always available, compile-time HAL
+ *  Hardware-Release "_AT" macros are fuzzy (return 0, 1, or XCHAL_MAYBE):
+ */
+#ifndef XCHAL_HW_RELEASE_MAJOR
+# define XCHAL_HW_CONFIGID_RELIABLE	0
+#endif
+#if XCHAL_HW_CONFIGID_RELIABLE
+# define XCHAL_HW_RELEASE_AT_OR_BELOW(major,minor)	(XTHAL_REL_LE( XCHAL_HW_RELEASE_MAJOR,XCHAL_HW_RELEASE_MINOR, major,minor ) ? 1 : 0)
+# define XCHAL_HW_RELEASE_AT_OR_ABOVE(major,minor)	(XTHAL_REL_GE( XCHAL_HW_RELEASE_MAJOR,XCHAL_HW_RELEASE_MINOR, major,minor ) ? 1 : 0)
+# define XCHAL_HW_RELEASE_AT(major,minor)		(XTHAL_REL_EQ( XCHAL_HW_RELEASE_MAJOR,XCHAL_HW_RELEASE_MINOR, major,minor ) ? 1 : 0)
+# define XCHAL_HW_RELEASE_MAJOR_AT(major)		((XCHAL_HW_RELEASE_MAJOR == (major)) ? 1 : 0)
+#else
+# define XCHAL_HW_RELEASE_AT_OR_BELOW(major,minor)	( ((major) < 1040 && XCHAL_HAVE_XEA2) ? 0 \
+							: ((major) > 1050 && XCHAL_HAVE_XEA1) ? 1 \
+							: XTHAL_MAYBE )
+# define XCHAL_HW_RELEASE_AT_OR_ABOVE(major,minor)	( ((major) >= 2000 && XCHAL_HAVE_XEA1) ? 0 \
+							: (XTHAL_REL_LE(major,minor, 1040,0) && XCHAL_HAVE_XEA2) ? 1 \
+							: XTHAL_MAYBE )
+# define XCHAL_HW_RELEASE_AT(major,minor)		( (((major) < 1040 && XCHAL_HAVE_XEA2) || \
+							   ((major) >= 2000 && XCHAL_HAVE_XEA1)) ? 0 : XTHAL_MAYBE)
+# define XCHAL_HW_RELEASE_MAJOR_AT(major)		XCHAL_HW_RELEASE_AT(major,0)
+#endif
+
+/*
+ *  Specific errata:
+ */
+
+/*
+ *  Erratum T1020.H13, T1030.H7, T1040.H10, T1050.H4 (fixed in T1040.3 and T1050.1;
+ *  relevant only in XEA1, kernel-vector mode, level-one interrupts and overflows enabled):
+ */
+#define XCHAL_MAYHAVE_ERRATUM_XEA1KWIN	(XCHAL_HAVE_XEA1 && \
+					 (XCHAL_HW_RELEASE_AT_OR_BELOW(1040,2) != 0 \
+					  || XCHAL_HW_RELEASE_AT(1050,0)))
+
+
+
+#endif /*XTENSA_CONFIG_CORE_H*/
+
diff --git a/include/asm-xtensa/xtensa/config-linux_be/defs.h b/include/asm-xtensa/xtensa/config-linux_be/defs.h
new file mode 100644
index 00000000000..f7c58b27337
--- /dev/null
+++ b/include/asm-xtensa/xtensa/config-linux_be/defs.h
@@ -0,0 +1,270 @@
+/* Definitions for Xtensa instructions, types, and protos. */
+
+/*
+ * Copyright (c) 2003 Tensilica, Inc.  All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2.1 of the GNU Lesser General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+ *
+ * Further, this software is distributed without any warranty that it is
+ * free of the rightful claim of any third person regarding infringement
+ * or the like.  Any license provided herein, whether implied or
+ * otherwise, applies only to this software file.  Patent licenses, if
+ * any, provided herein do not apply to combinations of this program with
+ * other software, or any other product whatsoever.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this program; if not, write the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston MA 02111-1307,
+ * USA.
+ */
+
+/* Do not modify. This is automatically generated.*/
+
+#ifndef _XTENSA_BASE_HEADER
+#define _XTENSA_BASE_HEADER
+
+#ifdef __XTENSA__
+#if defined(__GNUC__) && !defined(__XCC__)
+
+#define L8UI_ASM(arr, ars, imm) { \
+  __asm__ volatile("l8ui %0, %1, %2" : "=a" (arr) : "a" (ars) , "i" (imm)); \
+}
+
+#define XT_L8UI(ars, imm) \
+({ \
+  unsigned char _arr; \
+  const unsigned char *_ars = ars; \
+  L8UI_ASM(_arr, _ars, imm); \
+  _arr; \
+})
+
+#define L16UI_ASM(arr, ars, imm) { \
+  __asm__ volatile("l16ui %0, %1, %2" : "=a" (arr) : "a" (ars) , "i" (imm)); \
+}
+
+#define XT_L16UI(ars, imm) \
+({ \
+  unsigned short _arr; \
+  const unsigned short *_ars = ars; \
+  L16UI_ASM(_arr, _ars, imm); \
+  _arr; \
+})
+
+#define L16SI_ASM(arr, ars, imm) {\
+  __asm__ volatile("l16si %0, %1, %2" : "=a" (arr) : "a" (ars) , "i" (imm)); \
+}
+
+#define XT_L16SI(ars, imm) \
+({ \
+  signed short _arr; \
+  const signed short *_ars = ars; \
+  L16SI_ASM(_arr, _ars, imm); \
+  _arr; \
+})
+
+#define L32I_ASM(arr, ars, imm) { \
+  __asm__ volatile("l32i %0, %1, %2" : "=a" (arr) : "a" (ars) , "i" (imm)); \
+}
+
+#define XT_L32I(ars, imm) \
+({ \
+  unsigned _arr; \
+  const unsigned *_ars = ars; \
+  L32I_ASM(_arr, _ars, imm); \
+  _arr; \
+})
+
+#define S8I_ASM(arr, ars, imm) {\
+  __asm__ volatile("s8i %0, %1, %2" : : "a" (arr), "a" (ars) , "i" (imm) : "memory" ); \
+}
+
+#define XT_S8I(arr, ars, imm) \
+({ \
+  signed char _arr = arr; \
+  const signed char *_ars = ars; \
+  S8I_ASM(_arr, _ars, imm); \
+})
+
+#define S16I_ASM(arr, ars, imm) {\
+  __asm__ volatile("s16i %0, %1, %2" : : "a" (arr), "a" (ars) , "i" (imm) : "memory" ); \
+}
+
+#define XT_S16I(arr, ars, imm) \
+({ \
+  signed short _arr = arr; \
+  const signed short *_ars = ars; \
+  S16I_ASM(_arr, _ars, imm); \
+})
+
+#define S32I_ASM(arr, ars, imm) { \
+  __asm__ volatile("s32i %0, %1, %2" : : "a" (arr), "a" (ars) , "i" (imm) : "memory" ); \
+}
+
+#define XT_S32I(arr, ars, imm) \
+({ \
+  signed int _arr = arr; \
+  const signed int *_ars = ars; \
+  S32I_ASM(_arr, _ars, imm); \
+})
+
+#define ADDI_ASM(art, ars, imm) {\
+   __asm__ ("addi %0, %1, %2" : "=a" (art) : "a" (ars), "i" (imm)); \
+}
+
+#define XT_ADDI(ars, imm) \
+({ \
+   unsigned _art; \
+   unsigned _ars = ars; \
+   ADDI_ASM(_art, _ars, imm); \
+   _art; \
+})
+
+#define ABS_ASM(arr, art) {\
+   __asm__ ("abs %0, %1" : "=a" (arr) : "a" (art)); \
+}
+
+#define XT_ABS(art) \
+({ \
+   unsigned _arr; \
+   signed _art = art; \
+   ABS_ASM(_arr, _art); \
+   _arr; \
+})
+
+/* Note: In the following macros that reference SAR, the magic "state"
+   register is used to capture the dependency on SAR.  This is because
+   SAR is a 5-bit register and thus there are no C types that can be
+   used to represent it.  It doesn't appear that the SAR register is
+   even relevant to GCC, but it is marked as "clobbered" just in
+   case.  */
+
+#define SRC_ASM(arr, ars, art) {\
+   register int _xt_sar __asm__ ("state"); \
+   __asm__ ("src %0, %1, %2" \
+	    : "=a" (arr) : "a" (ars), "a" (art), "t" (_xt_sar)); \
+}
+
+#define XT_SRC(ars, art) \
+({ \
+   unsigned _arr; \
+   unsigned _ars = ars; \
+   unsigned _art = art; \
+   SRC_ASM(_arr, _ars, _art); \
+   _arr; \
+})
+
+#define SSR_ASM(ars) {\
+   register int _xt_sar __asm__ ("state"); \
+   __asm__ ("ssr %1" : "=t" (_xt_sar) : "a" (ars) : "sar"); \
+}
+
+#define XT_SSR(ars) \
+({ \
+   unsigned _ars = ars; \
+   SSR_ASM(_ars); \
+})
+
+#define SSL_ASM(ars) {\
+   register int _xt_sar __asm__ ("state"); \
+   __asm__ ("ssl %1" : "=t" (_xt_sar) : "a" (ars) : "sar"); \
+}
+
+#define XT_SSL(ars) \
+({ \
+   unsigned _ars = ars; \
+   SSL_ASM(_ars); \
+})
+
+#define SSA8B_ASM(ars) {\
+   register int _xt_sar __asm__ ("state"); \
+   __asm__ ("ssa8b %1" : "=t" (_xt_sar) : "a" (ars) : "sar"); \
+}
+
+#define XT_SSA8B(ars) \
+({ \
+   unsigned _ars = ars; \
+   SSA8B_ASM(_ars); \
+})
+
+#define SSA8L_ASM(ars) {\
+   register int _xt_sar __asm__ ("state"); \
+   __asm__ ("ssa8l %1" : "=t" (_xt_sar) : "a" (ars) : "sar"); \
+}
+
+#define XT_SSA8L(ars) \
+({ \
+   unsigned _ars = ars; \
+   SSA8L_ASM(_ars); \
+})
+
+#define SSAI_ASM(imm) {\
+   register int _xt_sar __asm__ ("state"); \
+   __asm__ ("ssai %1" : "=t" (_xt_sar) : "i" (imm) : "sar"); \
+}
+
+#define XT_SSAI(imm) \
+({ \
+   SSAI_ASM(imm); \
+})
+
+
+
+
+
+
+
+
+#endif /* __GNUC__ && !__XCC__ */
+
+#ifdef __XCC__
+
+/* Core load/store instructions */
+extern unsigned char _TIE_L8UI(const unsigned char * ars, immediate imm);
+extern unsigned short _TIE_L16UI(const unsigned short * ars, immediate imm);
+extern signed short _TIE_L16SI(const signed short * ars, immediate imm);
+extern unsigned _TIE_L32I(const unsigned * ars, immediate imm);
+extern void _TIE_S8I(unsigned char arr, unsigned char * ars, immediate imm);
+extern void _TIE_S16I(unsigned short arr, unsigned short * ars, immediate imm);
+extern void _TIE_S32I(unsigned arr, unsigned * ars, immediate imm);
+
+#define XT_L8UI  _TIE_L8UI
+#define XT_L16UI _TIE_L16UI
+#define XT_L16SI _TIE_L16SI
+#define XT_L32I  _TIE_L32I
+#define XT_S8I   _TIE_S8I
+#define XT_S16I  _TIE_S16I
+#define XT_S32I  _TIE_S32I
+
+/* Add-immediate instruction */
+extern unsigned _TIE_ADDI(unsigned ars, immediate imm);
+#define XT_ADDI  _TIE_ADDI
+
+/* Absolute value instruction */
+extern unsigned _TIE_ABS(int art);
+#define XT_ABS _TIE_ABS
+
+/* funnel shift instructions */
+extern unsigned _TIE_SRC(unsigned ars, unsigned art);
+#define XT_SRC _TIE_SRC
+extern void _TIE_SSR(unsigned ars);
+#define XT_SSR _TIE_SSR
+extern void _TIE_SSL(unsigned ars);
+#define XT_SSL _TIE_SSL
+extern void _TIE_SSA8B(unsigned ars);
+#define XT_SSA8B _TIE_SSA8B
+extern void _TIE_SSA8L(unsigned ars);
+#define XT_SSA8L _TIE_SSA8L
+extern void _TIE_SSAI(immediate imm);
+#define XT_SSAI _TIE_SSAI
+
+
+#endif /* __XCC__ */
+
+#endif /* __XTENSA__ */
+#endif /* !_XTENSA_BASE_HEADER */
diff --git a/include/asm-xtensa/xtensa/config-linux_be/specreg.h b/include/asm-xtensa/xtensa/config-linux_be/specreg.h
new file mode 100644
index 00000000000..fa4106aa9a0
--- /dev/null
+++ b/include/asm-xtensa/xtensa/config-linux_be/specreg.h
@@ -0,0 +1,99 @@
+/*
+ * Xtensa Special Register symbolic names
+ */
+
+/* $Id: specreg.h,v 1.2 2003/03/07 19:15:18 joetaylor Exp $ */
+
+/*
+ * Copyright (c) 2003 Tensilica, Inc.  All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2.1 of the GNU Lesser General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+ *
+ * Further, this software is distributed without any warranty that it is
+ * free of the rightful claim of any third person regarding infringement
+ * or the like.  Any license provided herein, whether implied or
+ * otherwise, applies only to this software file.  Patent licenses, if
+ * any, provided herein do not apply to combinations of this program with
+ * other software, or any other product whatsoever.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this program; if not, write the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston MA 02111-1307,
+ * USA.
+ */
+
+#ifndef XTENSA_SPECREG_H
+#define XTENSA_SPECREG_H
+
+/*  Include these special register bitfield definitions, for historical reasons:  */
+#include <xtensa/corebits.h>
+
+
+/*  Special registers:  */
+#define LBEG		0
+#define LEND		1
+#define LCOUNT		2
+#define SAR		3
+#define WINDOWBASE	72
+#define WINDOWSTART	73
+#define PTEVADDR	83
+#define RASID		90
+#define ITLBCFG		91
+#define DTLBCFG		92
+#define IBREAKENABLE	96
+#define DDR		104
+#define IBREAKA_0	128
+#define IBREAKA_1	129
+#define DBREAKA_0	144
+#define DBREAKA_1	145
+#define DBREAKC_0	160
+#define DBREAKC_1	161
+#define EPC_1		177
+#define EPC_2		178
+#define EPC_3		179
+#define EPC_4		180
+#define DEPC		192
+#define EPS_2		194
+#define EPS_3		195
+#define EPS_4		196
+#define EXCSAVE_1	209
+#define EXCSAVE_2	210
+#define EXCSAVE_3	211
+#define EXCSAVE_4	212
+#define INTERRUPT	226
+#define INTENABLE	228
+#define PS		230
+#define EXCCAUSE	232
+#define DEBUGCAUSE	233
+#define CCOUNT		234
+#define ICOUNT		236
+#define ICOUNTLEVEL	237
+#define EXCVADDR	238
+#define CCOMPARE_0	240
+#define CCOMPARE_1	241
+#define CCOMPARE_2	242
+#define MISC_REG_0	244
+#define MISC_REG_1	245
+
+/*  Special cases (bases of special register series):  */
+#define IBREAKA		128
+#define DBREAKA		144
+#define DBREAKC		160
+#define EPC		176
+#define EPS		192
+#define EXCSAVE		208
+#define CCOMPARE	240
+
+/*  Special names for read-only and write-only interrupt registers:  */
+#define INTREAD		226
+#define INTSET		226
+#define INTCLEAR	227
+
+#endif /* XTENSA_SPECREG_H */
+
diff --git a/include/asm-xtensa/xtensa/config-linux_be/system.h b/include/asm-xtensa/xtensa/config-linux_be/system.h
new file mode 100644
index 00000000000..cf9d4d308e3
--- /dev/null
+++ b/include/asm-xtensa/xtensa/config-linux_be/system.h
@@ -0,0 +1,198 @@
+/*
+ * xtensa/config/system.h -- HAL definitions that are dependent on SYSTEM configuration
+ *
+ *  NOTE: The location and contents of this file are highly subject to change.
+ *
+ *  Source for configuration-independent binaries (which link in a
+ *  configuration-specific HAL library) must NEVER include this file.
+ *  The HAL itself has historically included this file in some instances,
+ *  but this is not appropriate either, because the HAL is meant to be
+ *  core-specific but system independent.
+ */
+
+/*
+ * Copyright (c) 2003 Tensilica, Inc.  All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2.1 of the GNU Lesser General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+ *
+ * Further, this software is distributed without any warranty that it is
+ * free of the rightful claim of any third person regarding infringement
+ * or the like.  Any license provided herein, whether implied or
+ * otherwise, applies only to this software file.  Patent licenses, if
+ * any, provided herein do not apply to combinations of this program with
+ * other software, or any other product whatsoever.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this program; if not, write the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston MA 02111-1307,
+ * USA.
+ */
+
+
+#ifndef XTENSA_CONFIG_SYSTEM_H
+#define XTENSA_CONFIG_SYSTEM_H
+
+/*#include <xtensa/hal.h>*/
+
+
+
+/*----------------------------------------------------------------------
+				DEVICE ADDRESSES
+  ----------------------------------------------------------------------*/
+
+/*
+ *  Strange place to find these, but the configuration GUI
+ *  allows moving these around to account for various core
+ *  configurations.  Specific boards (and their BSP software)
+ *  will have specific meanings for these components.
+ */
+
+/*  I/O Block areas:  */
+#define XSHAL_IOBLOCK_CACHED_VADDR	0xE0000000
+#define XSHAL_IOBLOCK_CACHED_PADDR	0xF0000000
+#define XSHAL_IOBLOCK_CACHED_SIZE	0x0E000000
+
+#define XSHAL_IOBLOCK_BYPASS_VADDR	0xF0000000
+#define XSHAL_IOBLOCK_BYPASS_PADDR	0xF0000000
+#define XSHAL_IOBLOCK_BYPASS_SIZE	0x0E000000
+
+/*  System ROM:  */
+#define XSHAL_ROM_VADDR		0xEE000000
+#define XSHAL_ROM_PADDR		0xFE000000
+#define XSHAL_ROM_SIZE		0x00400000
+/*  Largest available area (free of vectors):  */
+#define XSHAL_ROM_AVAIL_VADDR	0xEE00052C
+#define XSHAL_ROM_AVAIL_VSIZE	0x003FFAD4
+
+/*  System RAM:  */
+#define XSHAL_RAM_VADDR		0xD0000000
+#define XSHAL_RAM_PADDR		0x00000000
+#define XSHAL_RAM_VSIZE		0x08000000
+#define XSHAL_RAM_PSIZE		0x10000000
+#define XSHAL_RAM_SIZE		XSHAL_RAM_PSIZE
+/*  Largest available area (free of vectors):  */
+#define XSHAL_RAM_AVAIL_VADDR	0xD0000370
+#define XSHAL_RAM_AVAIL_VSIZE	0x07FFFC90
+
+/*
+ *  Shadow system RAM (same device as system RAM, at different address).
+ *  (Emulation boards need this for the SONIC Ethernet driver
+ *   when data caches are configured for writeback mode.)
+ *  NOTE: on full MMU configs, this points to the BYPASS virtual address
+ *  of system RAM, ie. is the same as XSHAL_RAM_* except that virtual
+ *  addresses are viewed through the BYPASS static map rather than
+ *  the CACHED static map.
+ */
+#define XSHAL_RAM_BYPASS_VADDR		0xD8000000
+#define XSHAL_RAM_BYPASS_PADDR		0x00000000
+#define XSHAL_RAM_BYPASS_PSIZE		0x08000000
+
+/*  Alternate system RAM (different device than system RAM):  */
+#define XSHAL_ALTRAM_VADDR		0xCEE00000
+#define XSHAL_ALTRAM_PADDR		0xC0000000
+#define XSHAL_ALTRAM_SIZE		0x00200000
+
+
+/*----------------------------------------------------------------------
+ *			DEVICE-ADDRESS DEPENDENT...
+ *
+ *  Values written to CACHEATTR special register (or its equivalent)
+ *  to enable and disable caches in various modes.
+ *----------------------------------------------------------------------*/
+
+/*----------------------------------------------------------------------
+			BACKWARD COMPATIBILITY ...
+  ----------------------------------------------------------------------*/
+
+/*
+ *  NOTE:  the following two macros are DEPRECATED.  Use the latter
+ *  board-specific macros instead, which are specially tuned for the
+ *  particular target environments' memory maps.
+ */
+#define XSHAL_CACHEATTR_BYPASS		XSHAL_XT2000_CACHEATTR_BYPASS	/* disable caches in bypass mode */
+#define XSHAL_CACHEATTR_DEFAULT		XSHAL_XT2000_CACHEATTR_DEFAULT	/* default setting to enable caches (no writeback!) */
+
+/*----------------------------------------------------------------------
+			ISS (Instruction Set Simulator) SPECIFIC ...
+  ----------------------------------------------------------------------*/
+
+#define XSHAL_ISS_CACHEATTR_WRITEBACK	0x1122222F	/* enable caches in write-back mode */
+#define XSHAL_ISS_CACHEATTR_WRITEALLOC	0x1122222F	/* enable caches in write-allocate mode */
+#define XSHAL_ISS_CACHEATTR_WRITETHRU	0x1122222F	/* enable caches in write-through mode */
+#define XSHAL_ISS_CACHEATTR_BYPASS	0x2222222F	/* disable caches in bypass mode */
+#define XSHAL_ISS_CACHEATTR_DEFAULT	XSHAL_ISS_CACHEATTR_WRITEBACK	/* default setting to enable caches */
+
+/*  For Coware only:  */
+#define XSHAL_COWARE_CACHEATTR_WRITEBACK	0x11222222	/* enable caches in write-back mode */
+#define XSHAL_COWARE_CACHEATTR_WRITEALLOC	0x11222222	/* enable caches in write-allocate mode */
+#define XSHAL_COWARE_CACHEATTR_WRITETHRU	0x11222222	/* enable caches in write-through mode */
+#define XSHAL_COWARE_CACHEATTR_BYPASS		0x22222222	/* disable caches in bypass mode */
+#define XSHAL_COWARE_CACHEATTR_DEFAULT		XSHAL_COWARE_CACHEATTR_WRITEBACK	/* default setting to enable caches */
+
+/*  For BFM and other purposes:  */
+#define XSHAL_ALLVALID_CACHEATTR_WRITEBACK	0x11222222	/* enable caches without any invalid regions */
+#define XSHAL_ALLVALID_CACHEATTR_DEFAULT	XSHAL_ALLVALID_CACHEATTR_WRITEBACK	/* default setting for caches without any invalid regions */
+
+#define XSHAL_ISS_PIPE_REGIONS	0
+#define XSHAL_ISS_SDRAM_REGIONS	0
+
+
+/*----------------------------------------------------------------------
+			XT2000 BOARD SPECIFIC ...
+  ----------------------------------------------------------------------*/
+
+#define XSHAL_XT2000_CACHEATTR_WRITEBACK	0x22FFFFFF	/* enable caches in write-back mode */
+#define XSHAL_XT2000_CACHEATTR_WRITEALLOC	0x22FFFFFF	/* enable caches in write-allocate mode */
+#define XSHAL_XT2000_CACHEATTR_WRITETHRU	0x22FFFFFF	/* enable caches in write-through mode */
+#define XSHAL_XT2000_CACHEATTR_BYPASS		0x22FFFFFF	/* disable caches in bypass mode */
+#define XSHAL_XT2000_CACHEATTR_DEFAULT		XSHAL_XT2000_CACHEATTR_WRITEBACK	/* default setting to enable caches */
+
+#define XSHAL_XT2000_PIPE_REGIONS	0x00001000	/* BusInt pipeline regions */
+#define XSHAL_XT2000_SDRAM_REGIONS	0x00000005	/* BusInt SDRAM regions */
+
+
+/*----------------------------------------------------------------------
+				VECTOR SIZES
+  ----------------------------------------------------------------------*/
+
+/*
+ *  Sizes allocated to vectors by the system (memory map) configuration.
+ *  These sizes are constrained by core configuration (eg. one vector's
+ *  code cannot overflow into another vector) but are dependent on the
+ *  system or board (or LSP) memory map configuration.
+ *
+ *  Whether or not each vector happens to be in a system ROM is also
+ *  a system configuration matter, sometimes useful, included here also:
+ */
+#define XSHAL_RESET_VECTOR_SIZE	0x000004E0
+#define XSHAL_RESET_VECTOR_ISROM	1
+#define XSHAL_USER_VECTOR_SIZE	0x0000001C
+#define XSHAL_USER_VECTOR_ISROM	0
+#define XSHAL_PROGRAMEXC_VECTOR_SIZE	XSHAL_USER_VECTOR_SIZE	/* for backward compatibility */
+#define XSHAL_USEREXC_VECTOR_SIZE	XSHAL_USER_VECTOR_SIZE	/* for backward compatibility */
+#define XSHAL_KERNEL_VECTOR_SIZE	0x0000001C
+#define XSHAL_KERNEL_VECTOR_ISROM	0
+#define XSHAL_STACKEDEXC_VECTOR_SIZE	XSHAL_KERNEL_VECTOR_SIZE	/* for backward compatibility */
+#define XSHAL_KERNELEXC_VECTOR_SIZE	XSHAL_KERNEL_VECTOR_SIZE	/* for backward compatibility */
+#define XSHAL_DOUBLEEXC_VECTOR_SIZE	0x000000E0
+#define XSHAL_DOUBLEEXC_VECTOR_ISROM	0
+#define XSHAL_WINDOW_VECTORS_SIZE	0x00000180
+#define XSHAL_WINDOW_VECTORS_ISROM	0
+#define XSHAL_INTLEVEL2_VECTOR_SIZE	0x0000000C
+#define XSHAL_INTLEVEL2_VECTOR_ISROM	0
+#define XSHAL_INTLEVEL3_VECTOR_SIZE	0x0000000C
+#define XSHAL_INTLEVEL3_VECTOR_ISROM	0
+#define XSHAL_INTLEVEL4_VECTOR_SIZE	0x0000000C
+#define XSHAL_INTLEVEL4_VECTOR_ISROM	1
+#define XSHAL_DEBUG_VECTOR_SIZE		XSHAL_INTLEVEL4_VECTOR_SIZE
+#define XSHAL_DEBUG_VECTOR_ISROM	XSHAL_INTLEVEL4_VECTOR_ISROM
+
+
+#endif /*XTENSA_CONFIG_SYSTEM_H*/
+
diff --git a/include/asm-xtensa/xtensa/config-linux_be/tie.h b/include/asm-xtensa/xtensa/config-linux_be/tie.h
new file mode 100644
index 00000000000..3c2e514602f
--- /dev/null
+++ b/include/asm-xtensa/xtensa/config-linux_be/tie.h
@@ -0,0 +1,275 @@
+/*
+ * xtensa/config/tie.h -- HAL definitions that are dependent on CORE and TIE configuration
+ *
+ *  This header file is sometimes referred to as the "compile-time HAL" or CHAL.
+ *  It was generated for a specific Xtensa processor configuration,
+ *  and furthermore for a specific set of TIE source files that extend
+ *  basic core functionality.
+ *
+ *  Source for configuration-independent binaries (which link in a
+ *  configuration-specific HAL library) must NEVER include this file.
+ *  It is perfectly normal, however, for the HAL source itself to include this file.
+ */
+
+/*
+ * Copyright (c) 2003 Tensilica, Inc.  All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2.1 of the GNU Lesser General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+ *
+ * Further, this software is distributed without any warranty that it is
+ * free of the rightful claim of any third person regarding infringement
+ * or the like.  Any license provided herein, whether implied or
+ * otherwise, applies only to this software file.  Patent licenses, if
+ * any, provided herein do not apply to combinations of this program with
+ * other software, or any other product whatsoever.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this program; if not, write the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston MA 02111-1307,
+ * USA.
+ */
+
+
+#ifndef XTENSA_CONFIG_TIE_H
+#define XTENSA_CONFIG_TIE_H
+
+#include <xtensa/hal.h>
+
+
+/*----------------------------------------------------------------------
+				GENERAL
+  ----------------------------------------------------------------------*/
+
+/*
+ *  Separators for macros that expand into arrays.
+ *  These can be predefined by files that #include this one,
+ *  when different separators are required.
+ */
+/*  Element separator for macros that expand into 1-dimensional arrays:  */
+#ifndef XCHAL_SEP
+#define XCHAL_SEP			,
+#endif
+/*  Array separator for macros that expand into 2-dimensional arrays:  */
+#ifndef XCHAL_SEP2
+#define XCHAL_SEP2			},{
+#endif
+
+
+
+
+
+
+/*----------------------------------------------------------------------
+			COPROCESSORS and EXTRA STATE
+  ----------------------------------------------------------------------*/
+
+#define XCHAL_CP_NUM			0	/* number of coprocessors */
+#define XCHAL_CP_MAX			0	/* max coprocessor id plus one (0 if none) */
+#define XCHAL_CP_MASK			0x00	/* bitmask of coprocessors by id */
+
+/*  Space for coprocessors' state save areas:  */
+#define XCHAL_CP0_SA_SIZE		0
+#define XCHAL_CP1_SA_SIZE		0
+#define XCHAL_CP2_SA_SIZE		0
+#define XCHAL_CP3_SA_SIZE		0
+#define XCHAL_CP4_SA_SIZE		0
+#define XCHAL_CP5_SA_SIZE		0
+#define XCHAL_CP6_SA_SIZE		0
+#define XCHAL_CP7_SA_SIZE		0
+/*  Minimum required alignments of CP state save areas:  */
+#define XCHAL_CP0_SA_ALIGN		1
+#define XCHAL_CP1_SA_ALIGN		1
+#define XCHAL_CP2_SA_ALIGN		1
+#define XCHAL_CP3_SA_ALIGN		1
+#define XCHAL_CP4_SA_ALIGN		1
+#define XCHAL_CP5_SA_ALIGN		1
+#define XCHAL_CP6_SA_ALIGN		1
+#define XCHAL_CP7_SA_ALIGN		1
+
+/*  Indexing macros:  */
+#define _XCHAL_CP_SA_SIZE(n)		XCHAL_CP ## n ## _SA_SIZE
+#define XCHAL_CP_SA_SIZE(n)		_XCHAL_CP_SA_SIZE(n)	/* n = 0 .. 7 */
+#define _XCHAL_CP_SA_ALIGN(n)		XCHAL_CP ## n ## _SA_ALIGN
+#define XCHAL_CP_SA_ALIGN(n)		_XCHAL_CP_SA_ALIGN(n)	/* n = 0 .. 7 */
+
+
+/*  Space for "extra" state (user special registers and non-cp TIE) save area:  */
+#define XCHAL_EXTRA_SA_SIZE		0
+#define XCHAL_EXTRA_SA_ALIGN		1
+
+/*  Total save area size (extra + all coprocessors)  */
+/*  (not useful until xthal_{save,restore}_all_extra() is implemented,  */
+/*   but included for Tor2 beta; doesn't account for alignment!):  */
+#define XCHAL_CPEXTRA_SA_SIZE_TOR2	0	/* Tor2Beta temporary definition -- do not use */
+
+/*  Combined required alignment for all CP and EXTRA state save areas  */
+/*  (does not include required alignment for any base config registers):  */
+#define XCHAL_CPEXTRA_SA_ALIGN		1
+
+/* ... */
+
+
+#ifdef _ASMLANGUAGE
+/*
+ *  Assembly-language specific definitions (assembly macros, etc.).
+ */
+#include <xtensa/config/specreg.h>
+
+/********************
+ *  Macros to save and restore the non-coprocessor TIE portion of EXTRA state.
+ */
+
+/* (none) */
+
+
+/********************
+ *  Macros to create functions that save and restore all EXTRA (non-coprocessor) state
+ *  (does not include zero-overhead loop registers and non-optional registers).
+ */
+
+	/*
+	 *  Macro that expands to the body of a function that
+	 *  stores the extra (non-coprocessor) optional/custom state.
+	 *	Entry:	a2 = ptr to save area in which to save extra state
+	 *	Exit:	any register a2-a15 (?) may have been clobbered.
+	 */
+	.macro	xchal_extra_store_funcbody
+	.endm
+
+
+	/*
+	 *  Macro that expands to the body of a function that
+	 *  loads the extra (non-coprocessor) optional/custom state.
+	 *	Entry:	a2 = ptr to save area from which to restore extra state
+	 *	Exit:	any register a2-a15 (?) may have been clobbered.
+	 */
+	.macro	xchal_extra_load_funcbody
+	.endm
+
+
+/********************
+ *  Macros to save and restore the state of each TIE coprocessor.
+ */
+
+
+
+/********************
+ *  Macros to create functions that save and restore the state of *any* TIE coprocessor.
+ */
+
+	/*
+	 *  Macro that expands to the body of a function
+	 *  that stores the selected coprocessor's state (registers etc).
+	 *	Entry:	a2 = ptr to save area in which to save cp state
+	 *		a3 = coprocessor number
+	 *	Exit:	any register a2-a15 (?) may have been clobbered.
+	 */
+	.macro	xchal_cpi_store_funcbody
+	.endm
+
+
+	/*
+	 *  Macro that expands to the body of a function
+	 *  that loads the selected coprocessor's state (registers etc).
+	 *	Entry:	a2 = ptr to save area from which to restore cp state
+	 *		a3 = coprocessor number
+	 *	Exit:	any register a2-a15 (?) may have been clobbered.
+	 */
+	.macro	xchal_cpi_load_funcbody
+	.endm
+
+#endif /*_ASMLANGUAGE*/
+
+
+/*
+ *  Contents of save areas in terms of libdb register numbers.
+ *  NOTE:  CONTENTS_LIBDB_{UREG,REGF} macros are not defined in this file;
+ *  it is up to the user of this header file to define these macros
+ *  usefully before each expansion of the CONTENTS_LIBDB macros.
+ *  (Fields rsv[123] are reserved for future additions; they are currently
+ *   set to zero but may be set to some useful values in the future.)
+ *
+ *	CONTENTS_LIBDB_SREG(libdbnum, offset, size, align, rsv1, name, sregnum, bitmask, rsv2, rsv3)
+ *	CONTENTS_LIBDB_UREG(libdbnum, offset, size, align, rsv1, name, uregnum, bitmask, rsv2, rsv3)
+ *	CONTENTS_LIBDB_REGF(libdbnum, offset, size, align, rsv1, name, index, numentries, contentsize, regname_base, regfile_name, rsv2, rsv3)
+ */
+
+#define XCHAL_EXTRA_SA_CONTENTS_LIBDB_NUM	0
+#define XCHAL_EXTRA_SA_CONTENTS_LIBDB	/* empty */
+
+#define XCHAL_CP0_SA_CONTENTS_LIBDB_NUM	0
+#define XCHAL_CP0_SA_CONTENTS_LIBDB	/* empty */
+
+#define XCHAL_CP1_SA_CONTENTS_LIBDB_NUM	0
+#define XCHAL_CP1_SA_CONTENTS_LIBDB	/* empty */
+
+#define XCHAL_CP2_SA_CONTENTS_LIBDB_NUM	0
+#define XCHAL_CP2_SA_CONTENTS_LIBDB	/* empty */
+
+#define XCHAL_CP3_SA_CONTENTS_LIBDB_NUM	0
+#define XCHAL_CP3_SA_CONTENTS_LIBDB	/* empty */
+
+#define XCHAL_CP4_SA_CONTENTS_LIBDB_NUM	0
+#define XCHAL_CP4_SA_CONTENTS_LIBDB	/* empty */
+
+#define XCHAL_CP5_SA_CONTENTS_LIBDB_NUM	0
+#define XCHAL_CP5_SA_CONTENTS_LIBDB	/* empty */
+
+#define XCHAL_CP6_SA_CONTENTS_LIBDB_NUM	0
+#define XCHAL_CP6_SA_CONTENTS_LIBDB	/* empty */
+
+#define XCHAL_CP7_SA_CONTENTS_LIBDB_NUM	0
+#define XCHAL_CP7_SA_CONTENTS_LIBDB	/* empty */
+
+
+
+
+
+
+/*----------------------------------------------------------------------
+				MISC
+  ----------------------------------------------------------------------*/
+
+#if 0	/* is there something equivalent for user TIE? */
+#define XCHAL_CORE_ID			"linux_be"	/* configuration's alphanumeric core identifier
+							   (CoreID) set in the Xtensa Processor Generator */
+
+#define XCHAL_BUILD_UNIQUE_ID		0x00003256	/* software build-unique ID (22-bit) */
+
+/*  These definitions describe the hardware targeted by this software:  */
+#define XCHAL_HW_CONFIGID0		0xC103D1FF	/* config ID reg 0 value (upper 32 of 64 bits) */
+#define XCHAL_HW_CONFIGID1		0x00803256	/* config ID reg 1 value (lower 32 of 64 bits) */
+#define XCHAL_CONFIGID0			XCHAL_HW_CONFIGID0	/* for backward compatibility only -- don't use! */
+#define XCHAL_CONFIGID1			XCHAL_HW_CONFIGID1	/* for backward compatibility only -- don't use! */
+#define XCHAL_HW_RELEASE_MAJOR		1050	/* major release of targeted hardware */
+#define XCHAL_HW_RELEASE_MINOR		1	/* minor release of targeted hardware */
+#define XCHAL_HW_RELEASE_NAME		"T1050.1"	/* full release name of targeted hardware */
+#define XTHAL_HW_REL_T1050	1
+#define XTHAL_HW_REL_T1050_1	1
+#define XCHAL_HW_CONFIGID_RELIABLE	1
+#endif /*0*/
+
+
+
+/*----------------------------------------------------------------------
+				ISA
+  ----------------------------------------------------------------------*/
+
+#if 0	/* these probably don't belong here, but are related to or implemented using TIE */
+#define XCHAL_HAVE_BOOLEANS		0	/* 1 if booleans option configured, 0 otherwise */
+/*  Misc instructions:  */
+#define XCHAL_HAVE_MUL32		0	/* 1 if 32-bit integer multiply option configured, 0 otherwise */
+#define XCHAL_HAVE_MUL32_HIGH		0	/* 1 if MUL32 option includes MULUH and MULSH, 0 otherwise */
+
+#define XCHAL_HAVE_FP			0	/* 1 if floating point option configured, 0 otherwise */
+#endif /*0*/
+
+
+#endif /*XTENSA_CONFIG_TIE_H*/
+
diff --git a/include/asm-xtensa/xtensa/coreasm.h b/include/asm-xtensa/xtensa/coreasm.h
new file mode 100644
index 00000000000..a8cfb54c20a
--- /dev/null
+++ b/include/asm-xtensa/xtensa/coreasm.h
@@ -0,0 +1,526 @@
+#ifndef XTENSA_COREASM_H
+#define XTENSA_COREASM_H
+
+/*
+ * THIS FILE IS GENERATED -- DO NOT MODIFY BY HAND
+ *
+ * include/asm-xtensa/xtensa/coreasm.h -- assembler-specific
+ * definitions that depend on CORE configuration.
+ *
+ * Source for configuration-independent binaries (which link in a
+ * configuration-specific HAL library) must NEVER include this file.
+ * It is perfectly normal, however, for the HAL itself to include this
+ * file.
+ *
+ * This file must NOT include xtensa/config/system.h.  Any assembler
+ * header file that depends on system information should likely go in
+ * a new systemasm.h (or sysasm.h) header file.
+ *
+ *  NOTE: macro beqi32 is NOT configuration-dependent, and is placed
+ *        here til we will have configuration-independent header file.
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ *
+ * Copyright (C) 2002 Tensilica Inc.
+ */
+
+
+#include <xtensa/config/core.h>
+#include <xtensa/config/specreg.h>
+
+/*
+ *  Assembly-language specific definitions (assembly macros, etc.).
+ */
+
+/*----------------------------------------------------------------------
+ *  find_ms_setbit
+ *
+ *  This macro finds the most significant bit that is set in <as>
+ *  and return its index + <base> in <ad>, or <base> - 1 if <as> is zero.
+ *  The index counts starting at zero for the lsbit, so the return
+ *  value ranges from <base>-1 (no bit set) to <base>+31 (msbit set).
+ *
+ *  Parameters:
+ *	<ad>	destination address register (any register)
+ *	<as>	source address register
+ *	<at>	temporary address register (must be different than <as>)
+ *	<base>	constant value added to result (usually 0 or 1)
+ *  On entry:
+ *	<ad> = undefined if different than <as>
+ *	<as> = value whose most significant set bit is to be found
+ *	<at> = undefined
+ *	no other registers are used by this macro.
+ *  On exit:
+ *	<ad> = <base> + index of msbit set in original <as>,
+ *	     = <base> - 1 if original <as> was zero.
+ *	<as> clobbered (if not <ad>)
+ *	<at> clobbered (if not <ad>)
+ *  Example:
+ *	find_ms_setbit a0, a4, a0, 0		-- return in a0 index of msbit set in a4
+ */
+
+	.macro	find_ms_setbit ad, as, at, base
+#if XCHAL_HAVE_NSA
+	movi	\at, 31+\base
+	nsau	\as, \as	// get index of \as, numbered from msbit (32 if absent)
+	sub	\ad, \at, \as	// get numbering from lsbit (0..31, -1 if absent)
+#else /* XCHAL_HAVE_NSA */
+	movi	\at, \base	// start with result of 0 (point to lsbit of 32)
+
+	beqz	\as, 2f		// special case for zero argument: return -1
+	bltui	\as, 0x10000, 1f	// is it one of the 16 lsbits? (if so, check lower 16 bits)
+	addi	\at, \at, 16	// no, increment result to upper 16 bits (of 32)
+	//srli	\as, \as, 16	// check upper half (shift right 16 bits)
+	extui	\as, \as, 16, 16	// check upper half (shift right 16 bits)
+1:	bltui	\as, 0x100, 1f	// is it one of the 8 lsbits? (if so, check lower 8 bits)
+	addi	\at, \at, 8	// no, increment result to upper 8 bits (of 16)
+	srli	\as, \as, 8	// shift right to check upper 8 bits
+1:	bltui	\as, 0x10, 1f	// is it one of the 4 lsbits? (if so, check lower 4 bits)
+	addi	\at, \at, 4	// no, increment result to upper 4 bits (of 8)
+	srli	\as, \as, 4	// shift right 4 bits to check upper half
+1:	bltui	\as, 0x4, 1f	// is it one of the 2 lsbits? (if so, check lower 2 bits)
+	addi	\at, \at, 2	// no, increment result to upper 2 bits (of 4)
+	srli	\as, \as, 2	// shift right 2 bits to check upper half
+1:	bltui	\as, 0x2, 1f	// is it the lsbit?
+	addi	\at, \at, 2	// no, increment result to upper bit (of 2)
+2:	addi	\at, \at, -1	// (from just above: add 1;  from beqz: return -1)
+	//srli	\as, \as, 1
+1:				// done! \at contains index of msbit set (or -1 if none set)
+	.if	0x\ad - 0x\at	// destination different than \at ? (works because regs are a0-a15)
+	mov	\ad, \at	// then move result to \ad
+	.endif
+#endif /* XCHAL_HAVE_NSA */
+	.endm	// find_ms_setbit
+
+/*----------------------------------------------------------------------
+ *  find_ls_setbit
+ *
+ *  This macro finds the least significant bit that is set in <as>,
+ *  and return its index in <ad>.
+ *  Usage is the same as for the find_ms_setbit macro.
+ *  Example:
+ *	find_ls_setbit a0, a4, a0, 0	-- return in a0 index of lsbit set in a4
+ */
+
+	.macro	find_ls_setbit ad, as, at, base
+	neg	\at, \as	// keep only the least-significant bit that is set...
+	and	\as, \at, \as	// ... in \as
+	find_ms_setbit	\ad, \as, \at, \base
+	.endm	// find_ls_setbit
+
+/*----------------------------------------------------------------------
+ *  find_ls_one
+ *
+ *  Same as find_ls_setbit with base zero.
+ *  Source (as) and destination (ad) registers must be different.
+ *  Provided for backward compatibility.
+ */
+
+	.macro	find_ls_one ad, as
+	find_ls_setbit	\ad, \as, \ad, 0
+	.endm	// find_ls_one
+
+/*----------------------------------------------------------------------
+ *  floop, floopnez, floopgtz, floopend
+ *
+ *  These macros are used for fast inner loops that
+ *  work whether or not the Loops options is configured.
+ *  If the Loops option is configured, they simply use
+ *  the zero-overhead LOOP instructions; otherwise
+ *  they use explicit decrement and branch instructions.
+ *
+ *  They are used in pairs, with floop, floopnez or floopgtz
+ *  at the beginning of the loop, and floopend at the end.
+ *
+ *  Each pair of loop macro calls must be given the loop count
+ *  address register and a unique label for that loop.
+ *
+ *  Example:
+ *
+ *	movi	 a3, 16     // loop 16 times
+ *	floop    a3, myloop1
+ *	:
+ *	bnez     a7, end1	// exit loop if a7 != 0
+ *	:
+ *	floopend a3, myloop1
+ *  end1:
+ *
+ *  Like the LOOP instructions, these macros cannot be
+ *  nested, must include at least one instruction,
+ *  cannot call functions inside the loop, etc.
+ *  The loop can be exited by jumping to the instruction
+ *  following floopend (or elsewhere outside the loop),
+ *  or continued by jumping to a NOP instruction placed
+ *  immediately before floopend.
+ *
+ *  Unlike LOOP instructions, the register passed to floop*
+ *  cannot be used inside the loop, because it is used as
+ *  the loop counter if the Loops option is not configured.
+ *  And its value is undefined after exiting the loop.
+ *  And because the loop counter register is active inside
+ *  the loop, you can't easily use this construct to loop
+ *  across a register file using ROTW as you might with LOOP
+ *  instructions, unless you copy the loop register along.
+ */
+
+	/*  Named label version of the macros:  */
+
+	.macro	floop		ar, endlabel
+	floop_		\ar, .Lfloopstart_\endlabel, .Lfloopend_\endlabel
+	.endm
+
+	.macro	floopnez	ar, endlabel
+	floopnez_	\ar, .Lfloopstart_\endlabel, .Lfloopend_\endlabel
+	.endm
+
+	.macro	floopgtz	ar, endlabel
+	floopgtz_	\ar, .Lfloopstart_\endlabel, .Lfloopend_\endlabel
+	.endm
+
+	.macro	floopend	ar, endlabel
+	floopend_	\ar, .Lfloopstart_\endlabel, .Lfloopend_\endlabel
+	.endm
+
+	/*  Numbered local label version of the macros:  */
+#if 0 /*UNTESTED*/
+	.macro	floop89		ar
+	floop_		\ar, 8, 9f
+	.endm
+
+	.macro	floopnez89	ar
+	floopnez_	\ar, 8, 9f
+	.endm
+
+	.macro	floopgtz89	ar
+	floopgtz_	\ar, 8, 9f
+	.endm
+
+	.macro	floopend89	ar
+	floopend_	\ar, 8b, 9
+	.endm
+#endif /*0*/
+
+	/*  Underlying version of the macros:  */
+
+	.macro	floop_	ar, startlabel, endlabelref
+	.ifdef	_infloop_
+	.if	_infloop_
+	.err	// Error: floop cannot be nested
+	.endif
+	.endif
+	.set	_infloop_, 1
+#if XCHAL_HAVE_LOOPS
+	loop	\ar, \endlabelref
+#else /* XCHAL_HAVE_LOOPS */
+\startlabel:
+	addi	\ar, \ar, -1
+#endif /* XCHAL_HAVE_LOOPS */
+	.endm	// floop_
+
+	.macro	floopnez_	ar, startlabel, endlabelref
+	.ifdef	_infloop_
+	.if	_infloop_
+	.err	// Error: floopnez cannot be nested
+	.endif
+	.endif
+	.set	_infloop_, 1
+#if XCHAL_HAVE_LOOPS
+	loopnez	\ar, \endlabelref
+#else /* XCHAL_HAVE_LOOPS */
+	beqz	\ar, \endlabelref
+\startlabel:
+	addi	\ar, \ar, -1
+#endif /* XCHAL_HAVE_LOOPS */
+	.endm	// floopnez_
+
+	.macro	floopgtz_	ar, startlabel, endlabelref
+	.ifdef	_infloop_
+	.if	_infloop_
+	.err	// Error: floopgtz cannot be nested
+	.endif
+	.endif
+	.set	_infloop_, 1
+#if XCHAL_HAVE_LOOPS
+	loopgtz	\ar, \endlabelref
+#else /* XCHAL_HAVE_LOOPS */
+	bltz	\ar, \endlabelref
+	beqz	\ar, \endlabelref
+\startlabel:
+	addi	\ar, \ar, -1
+#endif /* XCHAL_HAVE_LOOPS */
+	.endm	// floopgtz_
+
+
+	.macro	floopend_	ar, startlabelref, endlabel
+	.ifndef	_infloop_
+	.err	// Error: floopend without matching floopXXX
+	.endif
+	.ifeq	_infloop_
+	.err	// Error: floopend without matching floopXXX
+	.endif
+	.set	_infloop_, 0
+#if ! XCHAL_HAVE_LOOPS
+	bnez	\ar, \startlabelref
+#endif /* XCHAL_HAVE_LOOPS */
+\endlabel:
+	.endm	// floopend_
+
+/*----------------------------------------------------------------------
+ *  crsil  --  conditional RSIL (read/set interrupt level)
+ *
+ *  Executes the RSIL instruction if it exists, else just reads PS.
+ *  The RSIL instruction does not exist in the new exception architecture
+ *  if the interrupt option is not selected.
+ */
+
+	.macro	crsil	ar, newlevel
+#if XCHAL_HAVE_OLD_EXC_ARCH || XCHAL_HAVE_INTERRUPTS
+	rsil	\ar, \newlevel
+#else
+	rsr	\ar, PS
+#endif
+	.endm	// crsil
+
+/*----------------------------------------------------------------------
+ *  window_spill{4,8,12}
+ *
+ *  These macros spill callers' register windows to the stack.
+ *  They work for both privileged and non-privileged tasks.
+ *  Must be called from a windowed ABI context, eg. within
+ *  a windowed ABI function (ie. valid stack frame, window
+ *  exceptions enabled, not in exception mode, etc).
+ *
+ *  This macro requires a single invocation of the window_spill_common
+ *  macro in the same assembly unit and section.
+ *
+ *  Note that using window_spill{4,8,12} macros is more efficient
+ *  than calling a function implemented using window_spill_function,
+ *  because the latter needs extra code to figure out the size of
+ *  the call to the spilling function.
+ *
+ *  Example usage:
+ *
+ *		.text
+ *		.align	4
+ *		.global	some_function
+ *		.type	some_function,@function
+ *	some_function:
+ *		entry	a1, 16
+ *		:
+ *		:
+ *
+ *		window_spill4	// spill windows of some_function's callers; preserves a0..a3 only;
+ *				// to use window_spill{8,12} in this example function we'd have
+ *				// to increase space allocated by the entry instruction, because
+ *				// 16 bytes only allows call4; 32 or 48 bytes (+locals) are needed
+ *				// for call8/window_spill8 or call12/window_spill12 respectively.
+ *		:
+ *
+ *		retw
+ *
+ *		window_spill_common	// instantiates code used by window_spill4
+ *
+ *
+ *  On entry:
+ *	none (if window_spill4)
+ *	stack frame has enough space allocated for call8 (if window_spill8)
+ *	stack frame has enough space allocated for call12 (if window_spill12)
+ *  On exit:
+ *	 a4..a15 clobbered (if window_spill4)
+ *	 a8..a15 clobbered (if window_spill8)
+ *	a12..a15 clobbered (if window_spill12)
+ *	no caller windows are in live registers
+ */
+
+	.macro	window_spill4
+#if XCHAL_HAVE_WINDOWED
+# if XCHAL_NUM_AREGS == 16
+	movi	a15, 0			// for 16-register files, no need to call to reach the end
+# elif XCHAL_NUM_AREGS == 32
+	call4	.L__wdwspill_assist28	// call deep enough to clear out any live callers
+# elif XCHAL_NUM_AREGS == 64
+	call4	.L__wdwspill_assist60	// call deep enough to clear out any live callers
+# endif
+#endif
+	.endm	// window_spill4
+
+	.macro	window_spill8
+#if XCHAL_HAVE_WINDOWED
+# if XCHAL_NUM_AREGS == 16
+	movi	a15, 0			// for 16-register files, no need to call to reach the end
+# elif XCHAL_NUM_AREGS == 32
+	call8	.L__wdwspill_assist24	// call deep enough to clear out any live callers
+# elif XCHAL_NUM_AREGS == 64
+	call8	.L__wdwspill_assist56	// call deep enough to clear out any live callers
+# endif
+#endif
+	.endm	// window_spill8
+
+	.macro	window_spill12
+#if XCHAL_HAVE_WINDOWED
+# if XCHAL_NUM_AREGS == 16
+	movi	a15, 0			// for 16-register files, no need to call to reach the end
+# elif XCHAL_NUM_AREGS == 32
+	call12	.L__wdwspill_assist20	// call deep enough to clear out any live callers
+# elif XCHAL_NUM_AREGS == 64
+	call12	.L__wdwspill_assist52	// call deep enough to clear out any live callers
+# endif
+#endif
+	.endm	// window_spill12
+
+/*----------------------------------------------------------------------
+ *  window_spill_function
+ *
+ *  This macro outputs a function that will spill its caller's callers'
+ *  register windows to the stack.  Eg. it could be used to implement
+ *  a version of xthal_window_spill() that works in non-privileged tasks.
+ *  This works for both privileged and non-privileged tasks.
+ *
+ *  Typical usage:
+ *
+ *		.text
+ *		.align	4
+ *		.global	my_spill_function
+ *		.type	my_spill_function,@function
+ *	my_spill_function:
+ *		window_spill_function
+ *
+ *  On entry to resulting function:
+ *	none
+ *  On exit from resulting function:
+ *	none (no caller windows are in live registers)
+ */
+
+	.macro	window_spill_function
+#if XCHAL_HAVE_WINDOWED
+# if XCHAL_NUM_AREGS == 32
+	entry	sp, 48
+	bbci.l	a0, 31, 1f		// branch if called with call4
+	bbsi.l	a0, 30, 2f		// branch if called with call12
+	call8	.L__wdwspill_assist16	// called with call8, only need another 8
+	retw
+1:	call12	.L__wdwspill_assist16	// called with call4, only need another 12
+	retw
+2:	call4	.L__wdwspill_assist16	// called with call12, only need another 4
+	retw
+# elif XCHAL_NUM_AREGS == 64
+	entry	sp, 48
+	bbci.l	a0, 31, 1f		// branch if called with call4
+	bbsi.l	a0, 30, 2f		// branch if called with call12
+	call4	.L__wdwspill_assist52	// called with call8, only need a call4
+	retw
+1:	call8	.L__wdwspill_assist52	// called with call4, only need a call8
+	retw
+2:	call12	.L__wdwspill_assist40	// called with call12, can skip a call12
+	retw
+# elif XCHAL_NUM_AREGS == 16
+	entry	sp, 16
+	bbci.l	a0, 31, 1f	// branch if called with call4
+	bbsi.l	a0, 30, 2f	// branch if called with call12
+	movi	a7, 0		// called with call8
+	retw
+1:	movi	a11, 0		// called with call4
+2:	retw			// if called with call12, everything already spilled
+
+//	movi	a15, 0		// trick to spill all but the direct caller
+//	j	1f
+//	//  The entry instruction is magical in the assembler (gets auto-aligned)
+//	//  so we have to jump to it to avoid falling through the padding.
+//	//  We need entry/retw to know where to return.
+//1:	entry	sp, 16
+//	retw
+# else
+#  error "unrecognized address register file size"
+# endif
+#endif /* XCHAL_HAVE_WINDOWED */
+	window_spill_common
+	.endm	// window_spill_function
+
+/*----------------------------------------------------------------------
+ *  window_spill_common
+ *
+ *  Common code used by any number of invocations of the window_spill##
+ *  and window_spill_function macros.
+ *
+ *  Must be instantiated exactly once within a given assembly unit,
+ *  within call/j range of and same section as window_spill##
+ *  macro invocations for that assembly unit.
+ *  (Is automatically instantiated by the window_spill_function macro.)
+ */
+
+	.macro	window_spill_common
+#if XCHAL_HAVE_WINDOWED && (XCHAL_NUM_AREGS == 32 || XCHAL_NUM_AREGS == 64)
+	.ifndef	.L__wdwspill_defined
+# if XCHAL_NUM_AREGS >= 64
+.L__wdwspill_assist60:
+	entry	sp, 32
+	call8	.L__wdwspill_assist52
+	retw
+.L__wdwspill_assist56:
+	entry	sp, 16
+	call4	.L__wdwspill_assist52
+	retw
+.L__wdwspill_assist52:
+	entry	sp, 48
+	call12	.L__wdwspill_assist40
+	retw
+.L__wdwspill_assist40:
+	entry	sp, 48
+	call12	.L__wdwspill_assist28
+	retw
+# endif
+.L__wdwspill_assist28:
+	entry	sp, 48
+	call12	.L__wdwspill_assist16
+	retw
+.L__wdwspill_assist24:
+	entry	sp, 32
+	call8	.L__wdwspill_assist16
+	retw
+.L__wdwspill_assist20:
+	entry	sp, 16
+	call4	.L__wdwspill_assist16
+	retw
+.L__wdwspill_assist16:
+	entry	sp, 16
+	movi	a15, 0
+	retw
+	.set	.L__wdwspill_defined, 1
+	.endif
+#endif /* XCHAL_HAVE_WINDOWED with 32 or 64 aregs */
+	.endm	// window_spill_common
+
+/*----------------------------------------------------------------------
+ *  beqi32
+ *
+ *  macro implements version of beqi for arbitrary 32-bit immidiate value
+ *
+ *     beqi32 ax, ay, imm32, label
+ *
+ *  Compares value in register ax with imm32 value and jumps to label if
+ *  equal. Clobberes register ay if needed
+ *
+ */
+   .macro beqi32	ax, ay, imm, label
+    .ifeq ((\imm-1) & ~7)	// 1..8 ?
+		beqi	\ax, \imm, \label
+    .else
+      .ifeq (\imm+1)		// -1 ?
+		beqi	\ax, \imm, \label
+      .else
+        .ifeq (\imm)		// 0 ?
+		beqz	\ax, \label
+        .else
+		//  We could also handle immediates 10,12,16,32,64,128,256
+		//  but it would be a long macro...
+		movi	\ay, \imm
+		beq	\ax, \ay, \label
+        .endif
+      .endif
+    .endif
+   .endm // beqi32
+
+#endif /*XTENSA_COREASM_H*/
+
diff --git a/include/asm-xtensa/xtensa/corebits.h b/include/asm-xtensa/xtensa/corebits.h
new file mode 100644
index 00000000000..e578ade4163
--- /dev/null
+++ b/include/asm-xtensa/xtensa/corebits.h
@@ -0,0 +1,77 @@
+#ifndef XTENSA_COREBITS_H
+#define XTENSA_COREBITS_H
+
+/*
+ * THIS FILE IS GENERATED -- DO NOT MODIFY BY HAND
+ *
+ * xtensa/corebits.h - Xtensa Special Register field positions and masks.
+ *
+ * (In previous releases, these were defined in specreg.h, a generated file.
+ *  This file is not generated, i.e. it is processor configuration independent.)
+ */
+
+
+/*  EXCCAUSE register fields:  */
+#define EXCCAUSE_EXCCAUSE_SHIFT	0
+#define EXCCAUSE_EXCCAUSE_MASK	0x3F
+/*  Exception causes (mostly incomplete!):  */
+#define EXCCAUSE_ILLEGAL		0
+#define EXCCAUSE_SYSCALL		1
+#define EXCCAUSE_IFETCHERROR		2
+#define EXCCAUSE_LOADSTOREERROR		3
+#define EXCCAUSE_LEVEL1INTERRUPT	4
+#define EXCCAUSE_ALLOCA			5
+
+/*  PS register fields:  */
+#define PS_WOE_SHIFT		18
+#define PS_WOE_MASK		0x00040000
+#define PS_WOE			PS_WOE_MASK
+#define PS_CALLINC_SHIFT	16
+#define PS_CALLINC_MASK		0x00030000
+#define PS_CALLINC(n)		(((n)&3)<<PS_CALLINC_SHIFT)	/* n = 0..3 */
+#define PS_OWB_SHIFT		8
+#define PS_OWB_MASK		0x00000F00
+#define PS_OWB(n)		(((n)&15)<<PS_OWB_SHIFT)	/* n = 0..15 (or 0..7) */
+#define PS_RING_SHIFT		6
+#define PS_RING_MASK		0x000000C0
+#define PS_RING(n)		(((n)&3)<<PS_RING_SHIFT)	/* n = 0..3 */
+#define PS_UM_SHIFT		5
+#define PS_UM_MASK		0x00000020
+#define PS_UM			PS_UM_MASK
+#define PS_EXCM_SHIFT		4
+#define PS_EXCM_MASK		0x00000010
+#define PS_EXCM			PS_EXCM_MASK
+#define PS_INTLEVEL_SHIFT	0
+#define PS_INTLEVEL_MASK	0x0000000F
+#define PS_INTLEVEL(n)		((n)&PS_INTLEVEL_MASK)		/* n = 0..15 */
+/*  Backward compatibility (deprecated):  */
+#define PS_PROGSTACK_SHIFT	PS_UM_SHIFT
+#define PS_PROGSTACK_MASK	PS_UM_MASK
+#define PS_PROG_SHIFT		PS_UM_SHIFT
+#define PS_PROG_MASK		PS_UM_MASK
+#define PS_PROG			PS_UM
+
+/*  DBREAKCn register fields:  */
+#define DBREAKC_MASK_SHIFT		0
+#define DBREAKC_MASK_MASK		0x0000003F
+#define DBREAKC_LOADBREAK_SHIFT		30
+#define DBREAKC_LOADBREAK_MASK		0x40000000
+#define DBREAKC_STOREBREAK_SHIFT	31
+#define DBREAKC_STOREBREAK_MASK		0x80000000
+
+/*  DEBUGCAUSE register fields:  */
+#define DEBUGCAUSE_DEBUGINT_SHIFT	5
+#define DEBUGCAUSE_DEBUGINT_MASK	0x20	/* debug interrupt */
+#define DEBUGCAUSE_BREAKN_SHIFT		4
+#define DEBUGCAUSE_BREAKN_MASK		0x10	/* BREAK.N instruction */
+#define DEBUGCAUSE_BREAK_SHIFT		3
+#define DEBUGCAUSE_BREAK_MASK		0x08	/* BREAK instruction */
+#define DEBUGCAUSE_DBREAK_SHIFT		2
+#define DEBUGCAUSE_DBREAK_MASK		0x04	/* DBREAK match */
+#define DEBUGCAUSE_IBREAK_SHIFT		1
+#define DEBUGCAUSE_IBREAK_MASK		0x02	/* IBREAK match */
+#define DEBUGCAUSE_ICOUNT_SHIFT		0
+#define DEBUGCAUSE_ICOUNT_MASK		0x01	/* ICOUNT would increment to zero */
+
+#endif /*XTENSA_COREBITS_H*/
+
diff --git a/include/asm-xtensa/xtensa/hal.h b/include/asm-xtensa/xtensa/hal.h
new file mode 100644
index 00000000000..d1047250545
--- /dev/null
+++ b/include/asm-xtensa/xtensa/hal.h
@@ -0,0 +1,822 @@
+#ifndef XTENSA_HAL_H
+#define XTENSA_HAL_H
+
+/*
+ * THIS FILE IS GENERATED -- DO NOT MODIFY BY HAND
+ *
+ * include/asm-xtensa/xtensa/hal.h -- contains a definition of the
+ * Core HAL interface.
+ *
+ * All definitions in this header file are independent of any specific
+ * Xtensa processor configuration.  Thus an OS or other software can
+ * include this header file and be compiled into configuration-
+ * independent objects that can be distributed and eventually linked
+ * to the HAL library (libhal.a) to create a configuration-specific
+ * final executable.
+ *
+ * Certain definitions, however, are release-specific -- such as the
+ * XTHAL_RELEASE_xxx macros (or additions made in later releases).
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2002 Tensilica Inc.
+ */
+
+
+/*----------------------------------------------------------------------
+  			 Constant Definitions
+			(shared with assembly)
+  ----------------------------------------------------------------------*/
+
+/*  Software release information (not configuration-specific!):  */
+#define XTHAL_RELEASE_MAJOR	1050
+#define XTHAL_RELEASE_MINOR	0
+#define XTHAL_RELEASE_NAME	"T1050.0-2002-08-06-eng0"
+#define XTHAL_RELEASE_INTERNAL	"2002-08-06-eng0"
+#define XTHAL_REL_T1050	1
+#define XTHAL_REL_T1050_0	1
+#define XTHAL_REL_T1050_0_2002	1
+#define XTHAL_REL_T1050_0_2002_08	1
+#define XTHAL_REL_T1050_0_2002_08_06	1
+#define XTHAL_REL_T1050_0_2002_08_06_ENG0	1
+
+/*  HAL version numbers (these names are for backward compatibility):  */
+#define XTHAL_MAJOR_REV		XTHAL_RELEASE_MAJOR
+#define XTHAL_MINOR_REV		XTHAL_RELEASE_MINOR
+/*
+ *  A bit of software release history on values of XTHAL_{MAJOR,MINOR}_REV:
+ *
+ *	Release		MAJOR	MINOR		Comment
+ *	=======		=====	=====		=======
+ *	T1015.n		n/a	n/a		(HAL not yet available)
+ *	T1020.{0,1,2}	0	1		(HAL beta)
+ *	T1020.{3,4}	0	2		First release.
+ *	T1020.n (n>4)	0	2 or >3		(TBD)
+ *	T1030.0		0	1		(HAL beta)
+ *	T1030.{1,2}	0	3		Equivalent to first release.
+ *	T1030.n (n>=3)	0	>= 3		(TBD)
+ *	T1040.n		1040	n		Full CHAL available from T1040.2
+ *	T1050.n		1050	n		Current release.
+ *
+ *
+ *  Note:  there is a distinction between the software release with
+ *  which something is compiled (accessible using XTHAL_RELEASE_* macros)
+ *  and the software release with which the HAL library was compiled
+ *  (accessible using Xthal_release_* global variables).  This
+ *  distinction is particularly relevant for vendors that distribute
+ *  configuration-independent binaries (eg. an OS), where their customer
+ *  might link it with a HAL of a different Xtensa software release.
+ *  In this case, it may be appropriate for the OS to verify at run-time
+ *  whether XTHAL_RELEASE_* and Xthal_release_* are compatible.
+ *  [Guidelines as to which release is compatible with which are not
+ *  currently provided explicitly, but might be inferred from reading
+ *  OSKit documentation for all releases -- compatibility is also highly
+ *  dependent on which HAL features are used.  Each release is usually
+ *  backward compatible, with very few exceptions if any.]
+ *
+ *  Notes:
+ *	Tornado 2.0 supported in T1020.3+, T1030.1+, and T1040.{0,1} only.
+ *	Tornado 2.0.2 supported in T1040.2+, and T1050.
+ *	Compile-time HAL port of NucleusPlus supported by T1040.2+ and T1050.
+ */
+
+
+/*
+ *  Architectural limits, independent of configuration.
+ *  Note that these are ISA-defined limits, not micro-architecture implementation
+ *  limits enforced by the Xtensa Processor Generator (which may be stricter than
+ *  these below).
+ */
+#define XTHAL_MAX_CPS		8	/* max number of coprocessors (0..7) */
+#define XTHAL_MAX_INTERRUPTS	32	/* max number of interrupts (0..31) */
+#define XTHAL_MAX_INTLEVELS	16	/* max number of interrupt levels (0..15) */
+					/* (as of T1040, implementation limit is 7: 0..6) */
+#define XTHAL_MAX_TIMERS	4	/* max number of timers (CCOMPARE0..CCOMPARE3) */
+					/* (as of T1040, implementation limit is 3: 0..2) */
+
+/*  Misc:  */
+#define XTHAL_LITTLEENDIAN		0
+#define XTHAL_BIGENDIAN			1
+
+
+/*  Interrupt types:  */
+#define XTHAL_INTTYPE_UNCONFIGURED	0
+#define XTHAL_INTTYPE_SOFTWARE		1
+#define XTHAL_INTTYPE_EXTERN_EDGE	2
+#define XTHAL_INTTYPE_EXTERN_LEVEL	3
+#define XTHAL_INTTYPE_TIMER		4
+#define XTHAL_INTTYPE_NMI		5
+#define XTHAL_MAX_INTTYPES		6	/* number of interrupt types */
+
+/*  Timer related:  */
+#define XTHAL_TIMER_UNCONFIGURED	-1	/* Xthal_timer_interrupt[] value for non-existent timers */
+#define XTHAL_TIMER_UNASSIGNED	XTHAL_TIMER_UNCONFIGURED	/* (for backwards compatibility only) */
+
+
+/*  Access Mode bits (tentative):  */	/* bit abbr unit short_name       PPC equ - Description */
+#define XTHAL_AMB_EXCEPTION	0	/* 001 E EX fls: EXception        none    - generate exception on any access (aka "illegal") */
+#define XTHAL_AMB_HITCACHE	1	/* 002 C CH fls: use Cache on Hit ~(I CI) - use cache on hit -- way from tag match [or H HC, or U UC] (ISA: same, except for Isolate case) */
+#define XTHAL_AMB_ALLOCATE	2	/* 004 A AL fl?: ALlocate         none    - refill cache on miss -- way from LRU [or F FI fill] (ISA: Read/Write Miss Refill) */
+#define XTHAL_AMB_WRITETHRU	3	/* 008 W WT --s: WriteThrough     W WT    - store immediately to memory (ISA: same) */
+#define XTHAL_AMB_ISOLATE	4	/* 010 I IS fls: ISolate          none    - use cache regardless of hit-vs-miss -- way from vaddr (ISA: use-cache-on-miss+hit) */
+#define XTHAL_AMB_GUARD		5	/* 020 G GU ?l?: GUard            G *     - non-speculative; spec/replay refs not permitted */
+#if 0
+#define XTHAL_AMB_ORDERED	x	/* 000 O OR fls: ORdered          G *     - mem accesses cannot be out of order */
+#define XTHAL_AMB_FUSEWRITES	x	/* 000 F FW --s: FuseWrites       none    - allow combining/merging multiple writes (to same datapath data unit) into one (implied by writeback) */
+#define XTHAL_AMB_COHERENT	x	/* 000 M MC fl?: Mem/MP Coherent  M       - on reads, other CPUs/bus-masters may need to supply data */
+#define XTHAL_AMB_TRUSTED	x	/* 000 T TR ?l?: TRusted          none    - memory will not bus error (if it does, handle as fatal imprecise interrupt) */
+#define XTHAL_AMB_PREFETCH	x	/* 000 P PR fl?: PRefetch         none    - on refill, read line+1 into prefetch buffers */
+#define XTHAL_AMB_STREAM	x	/* 000 S ST ???: STreaming        none    - access one of N stream buffers */
+#endif /*0*/
+
+#define XTHAL_AM_EXCEPTION	(1<<XTHAL_AMB_EXCEPTION)
+#define XTHAL_AM_HITCACHE	(1<<XTHAL_AMB_HITCACHE)
+#define XTHAL_AM_ALLOCATE	(1<<XTHAL_AMB_ALLOCATE)
+#define XTHAL_AM_WRITETHRU	(1<<XTHAL_AMB_WRITETHRU)
+#define XTHAL_AM_ISOLATE	(1<<XTHAL_AMB_ISOLATE)
+#define XTHAL_AM_GUARD		(1<<XTHAL_AMB_GUARD)
+#if 0
+#define XTHAL_AM_ORDERED	(1<<XTHAL_AMB_ORDERED)
+#define XTHAL_AM_FUSEWRITES	(1<<XTHAL_AMB_FUSEWRITES)
+#define XTHAL_AM_COHERENT	(1<<XTHAL_AMB_COHERENT)
+#define XTHAL_AM_TRUSTED	(1<<XTHAL_AMB_TRUSTED)
+#define XTHAL_AM_PREFETCH	(1<<XTHAL_AMB_PREFETCH)
+#define XTHAL_AM_STREAM		(1<<XTHAL_AMB_STREAM)
+#endif /*0*/
+
+/*
+ *  Allowed Access Modes (bit combinations).
+ *
+ *  Columns are:
+ *  "FOGIWACE"
+ *	Access mode bits (see XTHAL_AMB_xxx above).
+ *	<letter> = bit is set
+ *	'-'      = bit is clear
+ *	'.'      = bit is irrelevant / don't care, as follows:
+ *			E=1 makes all others irrelevant
+ *			W,F relevant only for stores
+ *  "2345"
+ *	Indicates which Xtensa releases support the corresponding
+ *	access mode.  Releases for each character column are:
+ *		2 = prior to T1020.2:   T1015 (V1.5), T1020.0, T1020.1
+ *		3 = T1020.2 and later:  T1020.2+, T1030
+ *		4 = T1040
+ *		5 = T1050 (maybe)
+ *	And the character column contents are:
+ *		<number> = support by release(s)
+ *		"." = unsupported by release(s)
+ *		"?" = support unknown
+ */
+					/* FOGIWACE 2345 */
+/*  For instruction fetch:  */
+#define XTHAL_FAM_EXCEPTION	0x001	/* .......E 2345 exception */
+#define XTHAL_FAM_ISOLATE	0x012	/* .--I.-C- .... isolate */
+#define XTHAL_FAM_BYPASS	0x000	/* .---.--- 2345 bypass */
+#define XTHAL_FAM_NACACHED	0x002	/* .---.-C- .... cached no-allocate (frozen) */
+#define XTHAL_FAM_CACHED	0x006	/* .---.AC- 2345 cached */
+/*  For data load:  */
+#define XTHAL_LAM_EXCEPTION	0x001	/* .......E 2345 exception */
+#define XTHAL_LAM_ISOLATE	0x012	/* .--I.-C- 2345 isolate */
+#define XTHAL_LAM_BYPASS	0x000	/* .O--.--- 2... bypass speculative */
+#define XTHAL_LAM_BYPASSG	0x020	/* .OG-.--- .345 bypass guarded */
+#define XTHAL_LAM_NACACHED	0x002	/* .O--.-C- 2... cached no-allocate speculative */
+#define XTHAL_LAM_NACACHEDG	0x022	/* .OG-.-C- .345 cached no-allocate guarded */
+#define XTHAL_LAM_CACHED	0x006	/* .---.AC- 2345 cached speculative */
+#define XTHAL_LAM_CACHEDG	0x026	/* .?G-.AC- .... cached guarded */
+/*  For data store:  */
+#define XTHAL_SAM_EXCEPTION	0x001	/* .......E 2345 exception */
+#define XTHAL_SAM_ISOLATE	0x032	/* .-GI--C- 2345 isolate */
+#define XTHAL_SAM_BYPASS	0x028	/* -OG-W--- 2345 bypass */
+/*efine XTHAL_SAM_BYPASSF	0x028*/	/* F-G-W--- ...? bypass write-combined */
+#define XTHAL_SAM_WRITETHRU	0x02A	/* -OG-W-C- 234? writethrough */
+/*efine XTHAL_SAM_WRITETHRUF	0x02A*/	/* F-G-W-C- ...5 writethrough write-combined */
+#define XTHAL_SAM_WRITEALLOC	0x02E	/* -OG-WAC- ...? writethrough-allocate */
+/*efine XTHAL_SAM_WRITEALLOCF	0x02E*/	/* F-G-WAC- ...? writethrough-allocate write-combined */
+#define XTHAL_SAM_WRITEBACK	0x026	/* F-G--AC- ...5 writeback */
+
+#if 0
+/*
+    Cache attribute encoding for CACHEATTR (per ISA):
+    (Note:  if this differs from ISA Ref Manual, ISA has precedence)
+
+	Inst-fetches	Loads		Stores
+	-------------	------------	-------------
+0x0	FCA_EXCEPTION  ?LCA_NACACHED_G*	SCA_WRITETHRU	"uncached"
+0x1	FCA_CACHED	LCA_CACHED	SCA_WRITETHRU	cached
+0x2	FCA_BYPASS	LCA_BYPASS_G*	SCA_BYPASS	bypass
+0x3	FCA_CACHED	LCA_CACHED	SCA_WRITEALLOCF	write-allocate
+		     or LCA_EXCEPTION	SCA_EXCEPTION	(if unimplemented)
+0x4	FCA_CACHED	LCA_CACHED	SCA_WRITEBACK	write-back
+		     or LCA_EXCEPTION	SCA_EXCEPTION	(if unimplemented)
+0x5..D	FCA_EXCEPTION	LCA_EXCEPTION	SCA_EXCEPTION	(reserved)
+0xE	FCA_EXCEPTION	LCA_ISOLATE	SCA_ISOLATE	isolate
+0xF	FCA_EXCEPTION	LCA_EXCEPTION	SCA_EXCEPTION	illegal
+     *  Prior to T1020.2?, guard feature not supported, this defaulted to speculative (no _G)
+*/
+#endif /*0*/
+
+
+#if !defined(__ASSEMBLY__) && !defined(_NOCLANGUAGE)
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/*----------------------------------------------------------------------
+  			     HAL
+  ----------------------------------------------------------------------*/
+
+/* Constant to be checked in build = (XTHAL_MAJOR_REV<<16)|XTHAL_MINOR_REV */
+extern const unsigned int Xthal_rev_no;
+
+
+/*----------------------------------------------------------------------
+  			Processor State
+  ----------------------------------------------------------------------*/
+/* save & restore the extra processor state */
+extern void xthal_save_extra(void *base);
+extern void xthal_restore_extra(void *base);
+
+extern void xthal_save_cpregs(void *base, int);
+extern void xthal_restore_cpregs(void *base, int);
+
+/*extern void xthal_save_all_extra(void *base);*/
+/*extern void xthal_restore_all_extra(void *base);*/
+
+/* space for processor state */
+extern const unsigned int Xthal_extra_size;
+extern const unsigned int Xthal_extra_align;
+/* space for TIE register files */
+extern const unsigned int Xthal_cpregs_size[XTHAL_MAX_CPS];
+extern const unsigned int Xthal_cpregs_align[XTHAL_MAX_CPS];
+
+/* total of space for the processor state (for Tor2) */
+extern const unsigned int Xthal_all_extra_size;
+extern const unsigned int Xthal_all_extra_align;
+
+/* initialize the extra processor */
+/*extern void xthal_init_extra(void);*/
+/* initialize the TIE coprocessor */
+/*extern void xthal_init_cp(int);*/
+
+/* initialize the extra processor */
+extern void xthal_init_mem_extra(void *);
+/* initialize the TIE coprocessor */
+extern void xthal_init_mem_cp(void *, int);
+
+/* validate & invalidate the TIE register file */
+extern void xthal_validate_cp(int);
+extern void xthal_invalidate_cp(int);
+
+/* the number of TIE coprocessors contiguous from zero (for Tor2) */
+extern const unsigned int Xthal_num_coprocessors;
+
+/* actual number of coprocessors */
+extern const unsigned char Xthal_cp_num;
+/* index of highest numbered coprocessor, plus one */
+extern const unsigned char Xthal_cp_max;
+/* index of highest allowed coprocessor number, per cfg, plus one */
+/*extern const unsigned char Xthal_cp_maxcfg;*/
+/* bitmask of which coprocessors are present */
+extern const unsigned int  Xthal_cp_mask;
+
+/* read and write cpenable register */
+extern void xthal_set_cpenable(unsigned);
+extern unsigned xthal_get_cpenable(void);
+
+/* read & write extra state register */
+/*extern int xthal_read_extra(void *base, unsigned reg, unsigned *value);*/
+/*extern int xthal_write_extra(void *base, unsigned reg, unsigned value);*/
+
+/* read & write a TIE coprocessor register */
+/*extern int xthal_read_cpreg(void *base, int cp, unsigned reg, unsigned *value);*/
+/*extern int xthal_write_cpreg(void *base, int cp, unsigned reg, unsigned value);*/
+
+/* return coprocessor number based on register */
+/*extern int xthal_which_cp(unsigned reg);*/
+
+/*----------------------------------------------------------------------
+   				Interrupts
+  ----------------------------------------------------------------------*/
+
+/* the number of interrupt levels */
+extern const unsigned char Xthal_num_intlevels;
+/* the number of interrupts */
+extern const unsigned char Xthal_num_interrupts;
+
+/* mask for level of interrupts */
+extern const unsigned int Xthal_intlevel_mask[XTHAL_MAX_INTLEVELS];
+/* mask for level 0 to N interrupts */
+extern const unsigned int Xthal_intlevel_andbelow_mask[XTHAL_MAX_INTLEVELS];
+
+/* level of each interrupt */
+extern const unsigned char Xthal_intlevel[XTHAL_MAX_INTERRUPTS];
+
+/* type per interrupt */
+extern const unsigned char Xthal_inttype[XTHAL_MAX_INTERRUPTS];
+
+/* masks of each type of interrupt */
+extern const unsigned int Xthal_inttype_mask[XTHAL_MAX_INTTYPES];
+
+/* interrupt numbers assigned to each timer interrupt */
+extern const int Xthal_timer_interrupt[XTHAL_MAX_TIMERS];
+
+/***  Virtual interrupt prioritization:  ***/
+
+/*  Convert between interrupt levels (as per PS.INTLEVEL) and virtual interrupt priorities:  */
+extern unsigned	xthal_vpri_to_intlevel(unsigned vpri);
+extern unsigned	xthal_intlevel_to_vpri(unsigned intlevel);
+
+/*  Enables/disables given set (mask) of interrupts; returns previous enabled-mask of all ints:  */
+extern unsigned	xthal_int_enable(unsigned);
+extern unsigned	xthal_int_disable(unsigned);
+
+/*  Set/get virtual priority of an interrupt:  */
+extern int	xthal_set_int_vpri(int intnum, int vpri);
+extern int	xthal_get_int_vpri(int intnum);
+
+/*  Set/get interrupt lockout level for exclusive access to virtual priority data structures:  */
+extern void	xthal_set_vpri_locklevel(unsigned intlevel);
+extern unsigned	xthal_get_vpri_locklevel(void);
+
+/*  Set/get current virtual interrupt priority:  */
+extern unsigned	xthal_set_vpri(unsigned vpri);
+extern unsigned	xthal_get_vpri(unsigned vpri);
+extern unsigned	xthal_set_vpri_intlevel(unsigned intlevel);
+extern unsigned	xthal_set_vpri_lock(void);
+
+
+
+/*----------------------------------------------------------------------
+   			Generic Interrupt Trampolining Support
+  ----------------------------------------------------------------------*/
+
+typedef void (XtHalVoidFunc)(void);
+
+/*
+ *  Bitmask of interrupts currently trampolining down:
+ */
+extern unsigned Xthal_tram_pending;
+
+/*
+ *  Bitmask of which interrupts currently trampolining down
+ *  synchronously are actually enabled; this bitmask is necessary
+ *  because INTENABLE cannot hold that state (sync-trampolining
+ *  interrupts must be kept disabled while trampolining);
+ *  in the current implementation, any bit set here is not set
+ *  in INTENABLE, and vice-versa; once a sync-trampoline is
+ *  handled (at level one), its enable bit must be moved from
+ *  here to INTENABLE:
+ */
+extern unsigned Xthal_tram_enabled;
+
+/*
+ *  Bitmask of interrupts configured for sync trampolining:
+ */
+extern unsigned Xthal_tram_sync;
+
+
+/*  Trampoline support functions:  */
+extern unsigned  xthal_tram_pending_to_service( void );
+extern void      xthal_tram_done( unsigned serviced_mask );
+extern int       xthal_tram_set_sync( int intnum, int sync );
+extern XtHalVoidFunc* xthal_set_tram_trigger_func( XtHalVoidFunc *trigger_fn );
+
+/*  INTENABLE,INTREAD,INTSET,INTCLEAR register access functions:  */
+extern unsigned  xthal_get_intenable( void );
+extern void      xthal_set_intenable( unsigned );
+extern unsigned  xthal_get_intread( void );
+extern void      xthal_set_intset( unsigned );
+extern void      xthal_set_intclear( unsigned );
+
+
+/*----------------------------------------------------------------------
+   				Register Windows
+  ----------------------------------------------------------------------*/
+
+/* number of registers in register window */
+extern const unsigned int  Xthal_num_aregs;
+extern const unsigned char Xthal_num_aregs_log2;
+
+/*  This spill any live register windows (other than the caller's):  */
+extern void      xthal_window_spill( void );
+
+
+/*----------------------------------------------------------------------
+   				Cache
+  ----------------------------------------------------------------------*/
+
+/* size of the cache lines in log2(bytes) */
+extern const unsigned char Xthal_icache_linewidth;
+extern const unsigned char Xthal_dcache_linewidth;
+/* size of the cache lines in bytes */
+extern const unsigned short Xthal_icache_linesize;
+extern const unsigned short Xthal_dcache_linesize;
+/* number of cache sets in log2(lines per way) */
+extern const unsigned char Xthal_icache_setwidth;
+extern const unsigned char Xthal_dcache_setwidth;
+/* cache set associativity (number of ways) */
+extern const unsigned int  Xthal_icache_ways;
+extern const unsigned int  Xthal_dcache_ways;
+/* size of the caches in bytes (ways * 2^(linewidth + setwidth)) */
+extern const unsigned int  Xthal_icache_size;
+extern const unsigned int  Xthal_dcache_size;
+/* cache features */
+extern const unsigned char Xthal_dcache_is_writeback;
+extern const unsigned char Xthal_icache_line_lockable;
+extern const unsigned char Xthal_dcache_line_lockable;
+
+/* cache attribute register control (used by other HAL routines) */
+extern unsigned xthal_get_cacheattr( void );
+extern unsigned xthal_get_icacheattr( void );
+extern unsigned xthal_get_dcacheattr( void );
+extern void     xthal_set_cacheattr( unsigned );
+extern void     xthal_set_icacheattr( unsigned );
+extern void     xthal_set_dcacheattr( unsigned );
+
+/* initialize cache support (must be called once at startup, before all other cache calls) */
+/*extern void xthal_cache_startinit( void );*/
+/* reset caches */
+/*extern void xthal_icache_reset( void );*/
+/*extern void xthal_dcache_reset( void );*/
+/* enable caches */
+extern void xthal_icache_enable( void );	/* DEPRECATED */
+extern void xthal_dcache_enable( void );	/* DEPRECATED */
+/* disable caches */
+extern void xthal_icache_disable( void );	/* DEPRECATED */
+extern void xthal_dcache_disable( void );	/* DEPRECATED */
+
+/* invalidate the caches */
+extern void xthal_icache_all_invalidate( void );
+extern void xthal_dcache_all_invalidate( void );
+extern void xthal_icache_region_invalidate( void *addr, unsigned size );
+extern void xthal_dcache_region_invalidate( void *addr, unsigned size );
+extern void xthal_icache_line_invalidate(void *addr);
+extern void xthal_dcache_line_invalidate(void *addr);
+/* write dirty data back */
+extern void xthal_dcache_all_writeback( void );
+extern void xthal_dcache_region_writeback( void *addr, unsigned size );
+extern void xthal_dcache_line_writeback(void *addr);
+/* write dirty data back and invalidate */
+extern void xthal_dcache_all_writeback_inv( void );
+extern void xthal_dcache_region_writeback_inv( void *addr, unsigned size );
+extern void xthal_dcache_line_writeback_inv(void *addr);
+/* prefetch and lock specified memory range into cache */
+extern void xthal_icache_region_lock( void *addr, unsigned size );
+extern void xthal_dcache_region_lock( void *addr, unsigned size );
+extern void xthal_icache_line_lock(void *addr);
+extern void xthal_dcache_line_lock(void *addr);
+/* unlock from cache */
+extern void xthal_icache_all_unlock( void );
+extern void xthal_dcache_all_unlock( void );
+extern void xthal_icache_region_unlock( void *addr, unsigned size );
+extern void xthal_dcache_region_unlock( void *addr, unsigned size );
+extern void xthal_icache_line_unlock(void *addr);
+extern void xthal_dcache_line_unlock(void *addr);
+
+
+/* sync icache and memory */
+extern void xthal_icache_sync( void );
+/* sync dcache and memory */
+extern void xthal_dcache_sync( void );
+
+/*----------------------------------------------------------------------
+   				Debug
+  ----------------------------------------------------------------------*/
+
+/*  1 if debug option configured, 0 if not:  */
+extern const int Xthal_debug_configured;
+
+/*  Number of instruction and data break registers:  */
+extern const int Xthal_num_ibreak;
+extern const int Xthal_num_dbreak;
+
+/*  Set (plant) and remove software breakpoint, both synchronizing cache:  */
+extern unsigned int xthal_set_soft_break(void *addr);
+extern void         xthal_remove_soft_break(void *addr, unsigned int);
+
+
+/*----------------------------------------------------------------------
+   				Disassembler
+  ----------------------------------------------------------------------*/
+
+/*  Max expected size of the return buffer for a disassembled instruction (hint only):  */
+#define XTHAL_DISASM_BUFSIZE	80
+
+/*  Disassembly option bits for selecting what to return:  */
+#define XTHAL_DISASM_OPT_ADDR	0x0001	/* display address */
+#define XTHAL_DISASM_OPT_OPHEX	0x0002	/* display opcode bytes in hex */
+#define XTHAL_DISASM_OPT_OPCODE	0x0004	/* display opcode name (mnemonic) */
+#define XTHAL_DISASM_OPT_PARMS	0x0008	/* display parameters */
+#define XTHAL_DISASM_OPT_ALL	0x0FFF	/* display everything */
+
+/* routine to get a string for the disassembled instruction */
+extern int xthal_disassemble( unsigned char *instr_buf, void *tgt_addr,
+		       char *buffer, unsigned buflen, unsigned options );
+
+/* routine to get the size of the next instruction. Returns 0 for
+   illegal instruction */
+extern int xthal_disassemble_size( unsigned char *instr_buf );
+
+
+/*----------------------------------------------------------------------
+   				Core Counter
+  ----------------------------------------------------------------------*/
+
+/* counter info */
+extern const unsigned char Xthal_have_ccount;	/* set if CCOUNT register present */
+extern const unsigned char Xthal_num_ccompare;	/* number of CCOMPAREn registers */
+
+/* get CCOUNT register (if not present return 0) */
+extern unsigned xthal_get_ccount(void);
+
+/* set and get CCOMPAREn registers (if not present, get returns 0) */
+extern void     xthal_set_ccompare(int, unsigned);
+extern unsigned xthal_get_ccompare(int);
+
+
+/*----------------------------------------------------------------------
+			Instruction/Data RAM/ROM Access
+  ----------------------------------------------------------------------*/
+
+extern void* xthal_memcpy(void *dst, const void *src, unsigned len);
+extern void* xthal_bcopy(const void *src, void *dst, unsigned len);
+
+/*----------------------------------------------------------------------
+                           MP Synchronization
+  ----------------------------------------------------------------------*/
+extern int      xthal_compare_and_set( int *addr, int test_val, int compare_val );
+extern unsigned xthal_get_prid( void );
+
+/*extern const char  Xthal_have_s32c1i;*/
+extern const unsigned char Xthal_have_prid;
+
+
+/*----------------------------------------------------------------------
+                             Miscellaneous
+  ----------------------------------------------------------------------*/
+
+extern const unsigned int  Xthal_release_major;
+extern const unsigned int  Xthal_release_minor;
+extern const char * const  Xthal_release_name;
+extern const char * const  Xthal_release_internal;
+
+extern const unsigned char Xthal_memory_order;
+extern const unsigned char Xthal_have_windowed;
+extern const unsigned char Xthal_have_density;
+extern const unsigned char Xthal_have_booleans;
+extern const unsigned char Xthal_have_loops;
+extern const unsigned char Xthal_have_nsa;
+extern const unsigned char Xthal_have_minmax;
+extern const unsigned char Xthal_have_sext;
+extern const unsigned char Xthal_have_clamps;
+extern const unsigned char Xthal_have_mac16;
+extern const unsigned char Xthal_have_mul16;
+extern const unsigned char Xthal_have_fp;
+extern const unsigned char Xthal_have_speculation;
+extern const unsigned char Xthal_have_exceptions;
+extern const unsigned char Xthal_xea_version;
+extern const unsigned char Xthal_have_interrupts;
+extern const unsigned char Xthal_have_highlevel_interrupts;
+extern const unsigned char Xthal_have_nmi;
+
+extern const unsigned short Xthal_num_writebuffer_entries;
+
+extern const unsigned int  Xthal_build_unique_id;
+/*  Release info for hardware targeted by software upgrades:  */
+extern const unsigned int  Xthal_hw_configid0;
+extern const unsigned int  Xthal_hw_configid1;
+extern const unsigned int  Xthal_hw_release_major;
+extern const unsigned int  Xthal_hw_release_minor;
+extern const char * const  Xthal_hw_release_name;
+extern const char * const  Xthal_hw_release_internal;
+
+
+/*  Internal memories...  */
+
+extern const unsigned char Xthal_num_instrom;
+extern const unsigned char Xthal_num_instram;
+extern const unsigned char Xthal_num_datarom;
+extern const unsigned char Xthal_num_dataram;
+extern const unsigned char Xthal_num_xlmi;
+extern const unsigned int  Xthal_instrom_vaddr[1];
+extern const unsigned int  Xthal_instrom_paddr[1];
+extern const unsigned int  Xthal_instrom_size [1];
+extern const unsigned int  Xthal_instram_vaddr[1];
+extern const unsigned int  Xthal_instram_paddr[1];
+extern const unsigned int  Xthal_instram_size [1];
+extern const unsigned int  Xthal_datarom_vaddr[1];
+extern const unsigned int  Xthal_datarom_paddr[1];
+extern const unsigned int  Xthal_datarom_size [1];
+extern const unsigned int  Xthal_dataram_vaddr[1];
+extern const unsigned int  Xthal_dataram_paddr[1];
+extern const unsigned int  Xthal_dataram_size [1];
+extern const unsigned int  Xthal_xlmi_vaddr[1];
+extern const unsigned int  Xthal_xlmi_paddr[1];
+extern const unsigned int  Xthal_xlmi_size [1];
+
+
+
+/*----------------------------------------------------------------------
+                         Memory Management Unit
+  ----------------------------------------------------------------------*/
+
+extern const unsigned char Xthal_have_spanning_way;
+extern const unsigned char Xthal_have_identity_map;
+extern const unsigned char Xthal_have_mimic_cacheattr;
+extern const unsigned char Xthal_have_xlt_cacheattr;
+extern const unsigned char Xthal_have_cacheattr;
+extern const unsigned char Xthal_have_tlbs;
+
+extern const unsigned char Xthal_mmu_asid_bits;		/* 0 .. 8 */
+extern const unsigned char Xthal_mmu_asid_kernel;
+extern const unsigned char Xthal_mmu_rings;		/* 1 .. 4 (perhaps 0 if no MMU and/or no protection?) */
+extern const unsigned char Xthal_mmu_ring_bits;
+extern const unsigned char Xthal_mmu_sr_bits;
+extern const unsigned char Xthal_mmu_ca_bits;
+extern const unsigned int  Xthal_mmu_max_pte_page_size;
+extern const unsigned int  Xthal_mmu_min_pte_page_size;
+
+extern const unsigned char Xthal_itlb_way_bits;
+extern const unsigned char Xthal_itlb_ways;
+extern const unsigned char Xthal_itlb_arf_ways;
+extern const unsigned char Xthal_dtlb_way_bits;
+extern const unsigned char Xthal_dtlb_ways;
+extern const unsigned char Xthal_dtlb_arf_ways;
+
+/*  Convert between virtual and physical addresses (through static maps only):  */
+/*** WARNING: these two functions may go away in a future release; don't depend on them! ***/
+extern int  xthal_static_v2p( unsigned vaddr, unsigned *paddrp );
+extern int  xthal_static_p2v( unsigned paddr, unsigned *vaddrp, unsigned cached );
+
+#if 0
+/*******************   EXPERIMENTAL AND TENTATIVE ONLY   ********************/
+
+#define XTHAL_MMU_PAGESZ_COUNT_MAX	8	/* maximum number of different page sizes */
+extern const char	Xthal_mmu_pagesz_count;		/* 0 .. 8		number of different page sizes configured */
+
+/*  Note:  the following table doesn't necessarily have page sizes in increasing order: */
+extern const char	Xthal_mmu_pagesz_log2[XTHAL_MMU_PAGESZ_COUNT_MAX];	/* 10 .. 28 (0 past count) */
+
+/*  Sorted (increasing) table of page sizes, that indexes into the above table: */
+extern const char	Xthal_mmu_pagesz_sorted[XTHAL_MMU_PAGESZ_COUNT_MAX];	/* 0 .. 7 (0 past count) */
+
+/*u32	Xthal_virtual_exceptions;*/	/* bitmask of which exceptions execute in virtual mode... */
+
+extern const char	Xthal_mmu_pte_pagesz_log2_min;	/* ?? minimum page size in PTEs */
+extern const char	Xthal_mmu_pte_pagesz_log2_max;	/* ?? maximum page size in PTEs */
+
+/*  Cache Attribute Bits Implemented by the Cache (part of the cache abstraction) */
+extern const char	Xthal_icache_fca_bits_implemented;	/* ITLB/UTLB only! */
+extern const char	Xthal_dcache_lca_bits_implemented;	/* DTLB/UTLB only! */
+extern const char	Xthal_dcache_sca_bits_implemented; 	/* DTLB/UTLB only! */
+
+/*  Per TLB Parameters (Instruction, Data, Unified)  */
+struct XtHalMmuTlb	Xthal_itlb;	/* description of MMU I-TLB generic features */
+struct XtHalMmuTlb	Xthal_dtlb;	/* description of MMU D-TLB generic features */
+struct XtHalMmuTlb	Xthal_utlb;	/* description of MMU U-TLB generic features */
+
+#define XTHAL_MMU_WAYS_MAX	8	/* maximum number of ways (associativities) for each TLB */
+
+/*  Structure for common information described for each possible TLB (instruction, data and unified): */
+typedef struct XtHalMmuTlb {
+    u8  	va_bits;		/* 32		(number of virtual address bits) */
+    u8  	pa_bits;		/* 32		(number of physical address bits) */
+    bool	tlb_va_indexed;		/* 1	(set if TLB is indexed by virtual address) */
+    bool	tlb_va_tagged;		/* 0	(set if TLB is tagged by virtual address) */
+    bool	cache_va_indexed;	/* 1	(set if cache is indexed by virtual address) */
+    bool	cache_va_tagged;	/* 0	(set if cache is tagged by virtual address) */
+    /*bool	(whether page tables are traversed in vaddr sorted order, paddr sorted order, ...) */
+    /*u8	(set of available page attribute bits, other than cache attribute bits defined above) */
+    /*u32	(various masks for pages, MMU table/TLB entries, etc.) */
+    u8  	way_count;		/* 0 .. 8	(number of ways, a.k.a. associativities, for this TLB) */
+    XtHalMmuTlbWay *	ways[XTHAL_MMU_WAYS_MAX];	/* pointers to per-way parms for each way */
+} XtHalMmuTlb;
+
+/*  Per TLB Way (Per Associativity) Parameters  */
+typedef struct XtHalMmuTlbWay {
+     u32	index_count_log2;	/* 0 .. 4 */
+     u32	pagesz_mask;		/* 0 .. 2^pagesz_count - 1	(each bit corresponds to a size */
+					/*		defined in the Xthal_mmu_pagesz_log2[] table) */
+     u32	vpn_const_mask;
+     u32	vpn_const_value;
+     u64	ppn_const_mask;		/* future may support pa_bits > 32 */
+     u64	ppn_const_value;
+     u32	ppn_id_mask;		/* paddr bits taken directly from vaddr */
+     bool	backgnd_match;		/* 0 or 1 */
+     /*  These are defined in terms of the XTHAL_CACHE_xxx bits: */
+     u8 	fca_const_mask;		/* ITLB/UTLB only! */
+     u8 	fca_const_value;	/* ITLB/UTLB only! */
+     u8 	lca_const_mask;		/* DTLB/UTLB only! */
+     u8 	lca_const_value; 	/* DTLB/UTLB only! */
+     u8 	sca_const_mask; 	/* DTLB/UTLB only! */
+     u8 	sca_const_value; 	/* DTLB/UTLB only! */
+     /*  These define an encoding that map 5 bits in TLB and PTE entries to */
+     /*  8 bits (FCA, ITLB), 16 bits (LCA+SCA, DTLB) or 24 bits (FCA+LCA+SCA, UTLB): */
+     /*  (they may be moved to struct XtHalMmuTlb) */
+     u8		ca_bits;		/* number of bits in TLB/PTE entries for cache attributes */
+     u32 *	ca_map;			/* pointer to array of 2^ca_bits entries of FCA+LCA+SCA bits */
+} XtHalMmuTlbWay;
+
+/*
+ *  The way to determine whether protection support is present in core
+ *  is to [look at Xthal_mmu_rings ???].
+ *  Give info on memory requirements for MMU tables and other in-memory
+ *  data structures (globally, per task, base and per page, etc.) - whatever bounds can be calculated.
+ */
+
+
+/*  Default vectors:  */
+xthal_immu_fetch_miss_vector
+xthal_dmmu_load_miss_vector
+xthal_dmmu_store_miss_vector
+
+/*  Functions called when a fault is detected:  */
+typedef void (XtHalMmuFaultFunc)( unsigned vaddr, ...context... );
+/*  Or, */
+/*	a? = vaddr */
+/*	a? = context... */
+/*	PS.xxx = xxx */
+XtHalMMuFaultFunc *Xthal_immu_fetch_fault_func;
+XtHalMMuFaultFunc *Xthal_dmmu_load_fault_func;
+XtHalMMuFaultFunc *Xthal_dmmu_store_fault_func;
+
+/*  Default Handlers:  */
+/*  The user and/or kernel exception handlers may jump to these handlers to handle the relevant exceptions,
+ *  according to the value of EXCCAUSE.  The exact register state on entry to these handlers is TBD.  */
+/*  When multiple TLB entries match (hit) on the same access:  */
+xthal_immu_fetch_multihit_handler
+xthal_dmmu_load_multihit_handler
+xthal_dmmu_store_multihit_handler
+/*  Protection violations according to cache attributes, and other cache attribute mismatches:  */
+xthal_immu_fetch_attr_handler
+xthal_dmmu_load_attr_handler
+xthal_dmmu_store_attr_handler
+/*  Protection violations due to insufficient ring level:  */
+xthal_immu_fetch_priv_handler
+xthal_dmmu_load_priv_handler
+xthal_dmmu_store_priv_handler
+/*  Alignment exception handlers (if supported by the particular Xtensa MMU configuration):  */
+xthal_dmmu_load_align_handler
+xthal_dmmu_store_align_handler
+
+/*  Or, alternatively, the OS user and/or kernel exception handlers may simply jump to the
+ *  following entry points which will handle any values of EXCCAUSE not handled by the OS:  */
+xthal_user_exc_default_handler
+xthal_kernel_exc_default_handler
+
+#endif /*0*/
+
+#ifdef INCLUDE_DEPRECATED_HAL_CODE
+extern const unsigned char Xthal_have_old_exc_arch;
+extern const unsigned char Xthal_have_mmu;
+extern const unsigned int  Xthal_num_regs;
+extern const unsigned char Xthal_num_iroms;
+extern const unsigned char Xthal_num_irams;
+extern const unsigned char Xthal_num_droms;
+extern const unsigned char Xthal_num_drams;
+extern const unsigned int  Xthal_configid0;
+extern const unsigned int  Xthal_configid1;
+#endif
+
+#ifdef INCLUDE_DEPRECATED_HAL_DEBUG_CODE
+#define XTHAL_24_BIT_BREAK		0x80000000
+#define XTHAL_16_BIT_BREAK		0x40000000
+extern const unsigned short	Xthal_ill_inst_16[16];
+#define XTHAL_DEST_REG		0xf0000000	/* Mask for destination register */
+#define XTHAL_DEST_REG_INST	0x08000000	/* Branch address is in register */
+#define XTHAL_DEST_REL_INST	0x04000000	/* Branch address is relative */
+#define XTHAL_RFW_INST		0x00000800
+#define XTHAL_RFUE_INST		0x00000400
+#define XTHAL_RFI_INST		0x00000200
+#define XTHAL_RFE_INST		0x00000100
+#define XTHAL_RET_INST		0x00000080
+#define XTHAL_BREAK_INST	0x00000040
+#define XTHAL_SYSCALL_INST	0x00000020
+#define XTHAL_LOOP_END		0x00000010	/* Not set by xthal_inst_type */
+#define XTHAL_JUMP_INST		0x00000008	/* Call or jump instruction */
+#define XTHAL_BRANCH_INST	0x00000004	/* Branch instruction */
+#define XTHAL_24_BIT_INST	0x00000002
+#define XTHAL_16_BIT_INST   0x00000001
+typedef struct xthal_state {
+    unsigned	pc;
+    unsigned	ar[16];
+    unsigned	lbeg;
+    unsigned	lend;
+    unsigned	lcount;
+    unsigned	extra_ptr;
+    unsigned	cpregs_ptr[XTHAL_MAX_CPS];
+} XTHAL_STATE;
+extern unsigned int xthal_inst_type(void *addr);
+extern unsigned int xthal_branch_addr(void *addr);
+extern unsigned int xthal_get_npc(XTHAL_STATE *user_state);
+#endif /* INCLUDE_DEPRECATED_HAL_DEBUG_CODE */
+
+#ifdef __cplusplus
+}
+#endif
+#endif /*!__ASSEMBLY__ */
+
+#endif /*XTENSA_HAL_H*/
+
diff --git a/include/asm-xtensa/xtensa/simcall.h b/include/asm-xtensa/xtensa/simcall.h
new file mode 100644
index 00000000000..a2b868929a4
--- /dev/null
+++ b/include/asm-xtensa/xtensa/simcall.h
@@ -0,0 +1,130 @@
+#ifndef SIMCALL_INCLUDED
+#define SIMCALL_INCLUDED
+
+/*
+ * THIS FILE IS GENERATED -- DO NOT MODIFY BY HAND
+ *
+ * include/asm-xtensa/xtensa/simcall.h  -  Simulator call numbers
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ *
+ * Copyright (C) 2002 Tensilica Inc.
+ */
+
+
+/*
+ *  System call like services offered by the simulator host.
+ *  These are modeled after the Linux 2.4 kernel system calls
+ *  for Xtensa processors.  However not all system calls and
+ *  not all functionality of a given system call are implemented,
+ *  or necessarily have well defined or equivalent semantics in
+ *  the context of a simulation (as opposed to a Unix kernel).
+ *
+ *  These services behave largely as if they had been invoked
+ *  as a task in the simulator host's operating system
+ *  (eg. files accessed are those of the simulator host).
+ *  However, these SIMCALLs model a virtual operating system
+ *  so that various definitions, bit assignments etc
+ *  (eg. open mode bits, errno values, etc) are independent
+ *  of the host operating system used to run the simulation.
+ *  Rather these definitions are specific to the Xtensa ISS.
+ *  This way Xtensa ISA code written to use these SIMCALLs
+ *  can (in principle) be simulated on any host.
+ *
+ *  Up to 6 parameters are passed in registers a3 to a8
+ *  (note the 6th parameter isn't passed on the stack,
+ *   unlike windowed function calling conventions).
+ *  The return value is in a2.  A negative value in the
+ *  range -4096 to -1 indicates a negated error code to be
+ *  reported in errno with a return value of -1, otherwise
+ *  the value in a2 is returned as is.
+ */
+
+/* These #defines need to match what's in Xtensa/OS/vxworks/xtiss/simcalls.c */
+
+#define SYS_nop		0	/* n/a - setup; used to flush register windows */
+#define SYS_exit	1	/*x*/
+#define SYS_fork	2
+#define SYS_read	3	/*x*/
+#define SYS_write	4	/*x*/
+#define SYS_open	5	/*x*/
+#define SYS_close	6	/*x*/
+#define SYS_rename	7	/*x 38 - waitpid */
+#define SYS_creat	8	/*x*/
+#define SYS_link	9	/*x (not implemented on WIN32) */
+#define SYS_unlink	10	/*x*/
+#define SYS_execv	11	/* n/a - execve */
+#define SYS_execve	12	/* 11 - chdir */
+#define SYS_pipe	13	/* 42 - time */
+#define SYS_stat	14	/* 106 - mknod */
+#define SYS_chmod	15
+#define SYS_chown	16	/* 202 - lchown */
+#define SYS_utime	17	/* 30 - break */
+#define SYS_wait	18	/* n/a - oldstat */
+#define SYS_lseek	19	/*x*/
+#define SYS_getpid	20
+#define SYS_isatty	21	/* n/a - mount */
+#define SYS_fstat	22	/* 108 - oldumount */
+#define SYS_time	23	/* 13 - setuid */
+#define SYS_gettimeofday 24	/*x 78 - getuid (not implemented on WIN32) */
+#define SYS_times	25	/*X 43 - stime (Xtensa-specific implementation) */
+#define SYS_socket      26
+#define SYS_sendto      27
+#define SYS_recvfrom    28
+#define SYS_select_one  29      /* not compitible select, one file descriptor at the time */
+#define SYS_bind        30
+#define SYS_ioctl	31
+
+/*
+ *  Other...
+ */
+#define SYS_iss_argc      1000	/* returns value of argc */
+#define SYS_iss_argv_size 1001	/* bytes needed for argv & arg strings */
+#define SYS_iss_set_argv  1002	/* saves argv & arg strings at given addr */
+
+/*
+ * SIMCALLs for the ferret memory debugger. All are invoked by
+ * libferret.a ...  ( Xtensa/Target-Libs/ferret )
+ */
+#define SYS_ferret           1010
+#define SYS_malloc           1011
+#define SYS_free             1012
+#define SYS_more_heap        1013
+#define SYS_no_heap          1014
+
+
+/*
+ *  Extra SIMCALLs for GDB:
+ */
+#define SYS_gdb_break         -1	/* invoked by XTOS on user exceptions if EPC points
+					   to a break.n/break, regardless of cause! */
+#define SYS_xmon_out          -2	/* invoked by XMON: ... */
+#define SYS_xmon_in           -3	/* invoked by XMON: ... */
+#define SYS_xmon_flush        -4	/* invoked by XMON: ... */
+#define SYS_gdb_abort         -5	/* invoked by XTOS in _xtos_panic() */
+#define SYS_gdb_illegal_inst  -6	/* invoked by XTOS for illegal instructions (too deeply) */
+#define SYS_xmon_init         -7	/* invoked by XMON: ... */
+#define SYS_gdb_enter_sktloop -8	/* invoked by XTOS on debug exceptions */
+
+/*
+ *  SIMCALLs for vxWorks xtiss BSP:
+ */
+#define SYS_setup_ppp_pipes   -83
+#define SYS_log_msg           -84
+
+/*
+ *  Test SIMCALLs:
+ */
+#define SYS_test_write_state  -100
+#define SYS_test_read_state   -101
+
+/*
+ * SYS_select_one specifiers
+ */
+#define  XTISS_SELECT_ONE_READ    1
+#define  XTISS_SELECT_ONE_WRITE   2
+#define  XTISS_SELECT_ONE_EXCEPT  3
+
+#endif /* !SIMCALL_INCLUDED */
diff --git a/include/asm-xtensa/xtensa/xt2000-uart.h b/include/asm-xtensa/xtensa/xt2000-uart.h
new file mode 100644
index 00000000000..0154460f0ed
--- /dev/null
+++ b/include/asm-xtensa/xtensa/xt2000-uart.h
@@ -0,0 +1,155 @@
+#ifndef _uart_h_included_
+#define _uart_h_included_
+
+/*
+ * THIS FILE IS GENERATED -- DO NOT MODIFY BY HAND
+ *
+ * include/asm-xtensa/xtensa/xt2000-uart.h -- NatSemi PC16552D DUART
+ * definitions
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2002 Tensilica Inc.
+ */
+
+
+#include <xtensa/xt2000.h>
+
+
+/* 16550 UART DEVICE REGISTERS
+   The XT2000 board aligns each register to a 32-bit word but the UART device only uses
+   one byte of the word, which is the least-significant byte regardless of the
+   endianness of the core (ie. byte offset 0 for little-endian and 3 for big-endian).
+   So if using word accesses then endianness doesn't matter.
+   The macros provided here do that.
+*/
+struct uart_dev_s {
+  union {
+    unsigned int rxb;	/* DLAB=0: receive buffer, read-only */
+    unsigned int txb;	/* DLAB=0: transmit buffer, write-only */
+    unsigned int dll;	/* DLAB=1: divisor, least-significant byte latch (was write-only?) */
+  } w0;
+  union {
+    unsigned int ier;	/* DLAB=0: interrupt-enable register (was write-only?) */
+    unsigned int dlm;	/* DLAB=1: divisor, most-significant byte latch (was write-only?) */
+  } w1;
+
+  union {
+    unsigned int isr;	/* DLAB=0: interrupt status register, read-only */
+    unsigned int fcr;	/* DLAB=0: FIFO control register, write-only */
+    unsigned int afr;	/* DLAB=1: alternate function register */
+  } w2;
+
+  unsigned int lcr;	/* line control-register, write-only */
+  unsigned int mcr;	/* modem control-regsiter, write-only */
+  unsigned int lsr;	/* line status register, read-only */
+  unsigned int msr;	/* modem status register, read-only */
+  unsigned int scr;	/* scratch regsiter, read/write */
+};
+
+#define _RXB(u) ((u)->w0.rxb)
+#define _TXB(u) ((u)->w0.txb)
+#define _DLL(u) ((u)->w0.dll)
+#define _IER(u) ((u)->w1.ier)
+#define _DLM(u) ((u)->w1.dlm)
+#define _ISR(u) ((u)->w2.isr)
+#define _FCR(u) ((u)->w2.fcr)
+#define _AFR(u) ((u)->w2.afr)
+#define _LCR(u) ((u)->lcr)
+#define _MCR(u) ((u)->mcr)
+#define _LSR(u) ((u)->lsr)
+#define _MSR(u) ((u)->msr)
+#define _SCR(u) ((u)->scr)
+
+typedef volatile struct uart_dev_s uart_dev_t;
+
+/* IER bits */
+#define RCVR_DATA_REG_INTENABLE 0x01
+#define XMIT_HOLD_REG_INTENABLE    0x02
+#define RCVR_STATUS_INTENABLE   0x04
+#define MODEM_STATUS_INTENABLE     0x08
+
+/* FCR bits */
+#define _FIFO_ENABLE      0x01
+#define RCVR_FIFO_RESET  0x02
+#define XMIT_FIFO_RESET  0x04
+#define DMA_MODE_SELECT  0x08
+#define RCVR_TRIGGER_LSB 0x40
+#define RCVR_TRIGGER_MSB 0x80
+
+/* AFR bits */
+#define AFR_CONC_WRITE	0x01
+#define AFR_BAUDOUT_SEL	0x02
+#define AFR_RXRDY_SEL	0x04
+
+/* ISR bits */
+#define INT_STATUS(r)   ((r)&1)
+#define INT_PRIORITY(r) (((r)>>1)&0x7)
+
+/* LCR bits */
+#define WORD_LENGTH(n)  (((n)-5)&0x3)
+#define STOP_BIT_ENABLE 0x04
+#define PARITY_ENABLE   0x08
+#define EVEN_PARITY     0x10
+#define FORCE_PARITY    0x20
+#define XMIT_BREAK      0x40
+#define DLAB_ENABLE     0x80
+
+/* MCR bits */
+#define _DTR 0x01
+#define _RTS 0x02
+#define _OP1 0x04
+#define _OP2 0x08
+#define LOOP_BACK 0x10
+
+/* LSR Bits */
+#define RCVR_DATA_READY 0x01
+#define OVERRUN_ERROR   0x02
+#define PARITY_ERROR    0x04
+#define FRAMING_ERROR   0x08
+#define BREAK_INTERRUPT 0x10
+#define XMIT_HOLD_EMPTY 0x20
+#define XMIT_EMPTY      0x40
+#define FIFO_ERROR      0x80
+#define RCVR_READY(u)   (_LSR(u)&RCVR_DATA_READY)
+#define XMIT_READY(u)   (_LSR(u)&XMIT_HOLD_EMPTY)
+
+/* MSR bits */
+#define _RDR       0x01
+#define DELTA_DSR 0x02
+#define DELTA_RI  0x04
+#define DELTA_CD  0x08
+#define _CTS       0x10
+#define _DSR       0x20
+#define _RI        0x40
+#define _CD        0x80
+
+/* prototypes */
+void uart_init( uart_dev_t *u, int bitrate );
+void uart_out( uart_dev_t *u, char c );
+void uart_puts( uart_dev_t *u, char *s );
+char uart_in( uart_dev_t *u );
+void uart_enable_rcvr_int( uart_dev_t *u );
+void uart_disable_rcvr_int( uart_dev_t *u );
+
+#ifdef DUART16552_1_VADDR
+/*  DUART present.  */
+#define DUART_1_BASE	(*(uart_dev_t*)DUART16552_1_VADDR)
+#define DUART_2_BASE	(*(uart_dev_t*)DUART16552_2_VADDR)
+#define UART1_PUTS(s)	uart_puts( &DUART_1_BASE, s )
+#define UART2_PUTS(s)	uart_puts( &DUART_2_BASE, s )
+#else
+/*  DUART not configured, use dummy placeholders to allow compiles to work.  */
+#define DUART_1_BASE	(*(uart_dev_t*)0)
+#define DUART_2_BASE	(*(uart_dev_t*)0)
+#define UART1_PUTS(s)
+#define UART2_PUTS(s)
+#endif
+
+/*  Compute 16-bit divisor for baudrate generator, with rounding:  */
+#define DUART_DIVISOR(crystal,speed)	(((crystal)/16 + (speed)/2)/(speed))
+
+#endif /*_uart_h_included_*/
+
diff --git a/include/asm-xtensa/xtensa/xt2000.h b/include/asm-xtensa/xtensa/xt2000.h
new file mode 100644
index 00000000000..703a45002f8
--- /dev/null
+++ b/include/asm-xtensa/xtensa/xt2000.h
@@ -0,0 +1,408 @@
+#ifndef _INC_XT2000_H_
+#define _INC_XT2000_H_
+
+/*
+ * THIS FILE IS GENERATED -- DO NOT MODIFY BY HAND
+ *
+ * include/asm-xtensa/xtensa/xt2000.h - Definitions specific to the
+ * Tensilica XT2000 Emulation Board
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2002 Tensilica Inc.
+ */
+
+
+#include <xtensa/config/core.h>
+#include <xtensa/config/system.h>
+
+
+/*
+ *  Default assignment of XT2000 devices to external interrupts.
+ */
+
+/*  Ethernet interrupt:  */
+#ifdef XCHAL_EXTINT3_NUM
+#define SONIC83934_INTNUM	XCHAL_EXTINT3_NUM
+#define SONIC83934_INTLEVEL	XCHAL_EXTINT3_LEVEL
+#define SONIC83934_INTMASK	XCHAL_EXTINT3_MASK
+#else
+#define SONIC83934_INTMASK	0
+#endif
+
+/*  DUART channel 1 interrupt (P1 - console):  */
+#ifdef XCHAL_EXTINT4_NUM
+#define DUART16552_1_INTNUM	XCHAL_EXTINT4_NUM
+#define DUART16552_1_INTLEVEL	XCHAL_EXTINT4_LEVEL
+#define DUART16552_1_INTMASK	XCHAL_EXTINT4_MASK
+#else
+#define DUART16552_1_INTMASK	0
+#endif
+
+/*  DUART channel 2 interrupt (P2 - 2nd serial port):  */
+#ifdef XCHAL_EXTINT5_NUM
+#define DUART16552_2_INTNUM	XCHAL_EXTINT5_NUM
+#define DUART16552_2_INTLEVEL	XCHAL_EXTINT5_LEVEL
+#define DUART16552_2_INTMASK	XCHAL_EXTINT5_MASK
+#else
+#define DUART16552_2_INTMASK	0
+#endif
+
+/*  FPGA-combined PCI/etc interrupts:  */
+#ifdef XCHAL_EXTINT6_NUM
+#define XT2000_FPGAPCI_INTNUM	XCHAL_EXTINT6_NUM
+#define XT2000_FPGAPCI_INTLEVEL	XCHAL_EXTINT6_LEVEL
+#define XT2000_FPGAPCI_INTMASK	XCHAL_EXTINT6_MASK
+#else
+#define XT2000_FPGAPCI_INTMASK	0
+#endif
+
+
+
+/*
+ *  Device addresses.
+ *
+ *  Note:  for endianness-independence, use 32-bit loads and stores for all
+ *  register accesses to Ethernet, DUART and LED devices.  Undefined bits
+ *  may need to be masked out if needed when reading if the actual register
+ *  size is smaller than 32 bits.
+ *
+ *  Note:  XT2000 bus byte lanes are defined in terms of msbyte and lsbyte
+ *  relative to the processor.  So 32-bit registers are accessed consistently
+ *  from both big and little endian processors.  However, this means byte
+ *  sequences are not consistent between big and little endian processors.
+ *  This is fine for RAM, and for ROM if ROM is created for a specific
+ *  processor (and thus has correct byte sequences).  However this may be
+ *  unexpected for Flash, which might contain a file-system that one wants
+ *  to use for multiple processor configurations (eg. the Flash might contain
+ *  the Ethernet card's address, endianness-independent application data, etc).
+ *  That is, byte sequences written in Flash by a core of a given endianness
+ *  will be byte-swapped when seen by a core of the other endianness.
+ *  Someone implementing an endianness-independent Flash file system will
+ *  likely handle this byte-swapping issue in the Flash driver software.
+ */
+
+#define DUART16552_XTAL_FREQ	18432000	/* crystal frequency in Hz */
+#define XTBOARD_FLASH_MAXSIZE	0x4000000	/* 64 MB (max; depends on what is socketed!) */
+#define XTBOARD_EPROM_MAXSIZE	0x0400000	/* 4 MB (max; depends on what is socketed!) */
+#define XTBOARD_EEPROM_MAXSIZE	0x0080000	/* 512 kB (max; depends on what is socketed!) */
+#define XTBOARD_ASRAM_SIZE	0x0100000	/* 1 MB */
+#define XTBOARD_PCI_MEM_SIZE	0x8000000	/* 128 MB (allocated) */
+#define XTBOARD_PCI_IO_SIZE	0x1000000	/* 16 MB (allocated) */
+
+#ifdef XSHAL_IOBLOCK_BYPASS_PADDR
+/*  PCI memory space:  */
+# define XTBOARD_PCI_MEM_PADDR	(XSHAL_IOBLOCK_BYPASS_PADDR+0x0000000)
+/*  Socketed Flash (eg. 2 x 16-bit devices):  */
+# define XTBOARD_FLASH_PADDR	(XSHAL_IOBLOCK_BYPASS_PADDR+0x8000000)
+/*  PCI I/O space:  */
+# define XTBOARD_PCI_IO_PADDR	(XSHAL_IOBLOCK_BYPASS_PADDR+0xC000000)
+/*  V3 PCI interface chip register/config space:  */
+# define XTBOARD_V3PCI_PADDR	(XSHAL_IOBLOCK_BYPASS_PADDR+0xD000000)
+/*  Bus Interface registers:  */
+# define XTBOARD_BUSINT_PADDR	(XSHAL_IOBLOCK_BYPASS_PADDR+0xD010000)
+/*  FPGA registers:  */
+# define XT2000_FPGAREGS_PADDR	(XSHAL_IOBLOCK_BYPASS_PADDR+0xD020000)
+/*  SONIC SN83934 Ethernet controller/transceiver:  */
+# define SONIC83934_PADDR	(XSHAL_IOBLOCK_BYPASS_PADDR+0xD030000)
+/*  8-character bitmapped LED display:  */
+# define XTBOARD_LED_PADDR	(XSHAL_IOBLOCK_BYPASS_PADDR+0xD040000)
+/*  National-Semi PC16552D DUART:  */
+# define DUART16552_1_PADDR	(XSHAL_IOBLOCK_BYPASS_PADDR+0xD050020)	/* channel 1 (P1 - console) */
+# define DUART16552_2_PADDR	(XSHAL_IOBLOCK_BYPASS_PADDR+0xD050000)	/* channel 2 (P2) */
+/*  Asynchronous Static RAM:  */
+# define XTBOARD_ASRAM_PADDR	(XSHAL_IOBLOCK_BYPASS_PADDR+0xD400000)
+/*  8-bit EEPROM:  */
+# define XTBOARD_EEPROM_PADDR	(XSHAL_IOBLOCK_BYPASS_PADDR+0xD600000)
+/*  2 x 16-bit EPROMs:  */
+# define XTBOARD_EPROM_PADDR	(XSHAL_IOBLOCK_BYPASS_PADDR+0xD800000)
+#endif /* XSHAL_IOBLOCK_BYPASS_PADDR */
+
+/*  These devices might be accessed cached:  */
+#ifdef XSHAL_IOBLOCK_CACHED_PADDR
+# define XTBOARD_PCI_MEM_CACHED_PADDR	(XSHAL_IOBLOCK_CACHED_PADDR+0x0000000)
+# define XTBOARD_FLASH_CACHED_PADDR	(XSHAL_IOBLOCK_CACHED_PADDR+0x8000000)
+# define XTBOARD_ASRAM_CACHED_PADDR	(XSHAL_IOBLOCK_CACHED_PADDR+0xD400000)
+# define XTBOARD_EEPROM_CACHED_PADDR	(XSHAL_IOBLOCK_CACHED_PADDR+0xD600000)
+# define XTBOARD_EPROM_CACHED_PADDR	(XSHAL_IOBLOCK_CACHED_PADDR+0xD800000)
+#endif /* XSHAL_IOBLOCK_CACHED_PADDR */
+
+
+/***  Same thing over again, this time with virtual addresses:  ***/
+
+#ifdef XSHAL_IOBLOCK_BYPASS_VADDR
+/*  PCI memory space:  */
+# define XTBOARD_PCI_MEM_VADDR	(XSHAL_IOBLOCK_BYPASS_VADDR+0x0000000)
+/*  Socketed Flash (eg. 2 x 16-bit devices):  */
+# define XTBOARD_FLASH_VADDR	(XSHAL_IOBLOCK_BYPASS_VADDR+0x8000000)
+/*  PCI I/O space:  */
+# define XTBOARD_PCI_IO_VADDR	(XSHAL_IOBLOCK_BYPASS_VADDR+0xC000000)
+/*  V3 PCI interface chip register/config space:  */
+# define XTBOARD_V3PCI_VADDR	(XSHAL_IOBLOCK_BYPASS_VADDR+0xD000000)
+/*  Bus Interface registers:  */
+# define XTBOARD_BUSINT_VADDR	(XSHAL_IOBLOCK_BYPASS_VADDR+0xD010000)
+/*  FPGA registers:  */
+# define XT2000_FPGAREGS_VADDR	(XSHAL_IOBLOCK_BYPASS_VADDR+0xD020000)
+/*  SONIC SN83934 Ethernet controller/transceiver:  */
+# define SONIC83934_VADDR	(XSHAL_IOBLOCK_BYPASS_VADDR+0xD030000)
+/*  8-character bitmapped LED display:  */
+# define XTBOARD_LED_VADDR	(XSHAL_IOBLOCK_BYPASS_VADDR+0xD040000)
+/*  National-Semi PC16552D DUART:  */
+# define DUART16552_1_VADDR	(XSHAL_IOBLOCK_BYPASS_VADDR+0xD050020)	/* channel 1 (P1 - console) */
+# define DUART16552_2_VADDR	(XSHAL_IOBLOCK_BYPASS_VADDR+0xD050000)	/* channel 2 (P2) */
+/*  Asynchronous Static RAM:  */
+# define XTBOARD_ASRAM_VADDR	(XSHAL_IOBLOCK_BYPASS_VADDR+0xD400000)
+/*  8-bit EEPROM:  */
+# define XTBOARD_EEPROM_VADDR	(XSHAL_IOBLOCK_BYPASS_VADDR+0xD600000)
+/*  2 x 16-bit EPROMs:  */
+# define XTBOARD_EPROM_VADDR	(XSHAL_IOBLOCK_BYPASS_VADDR+0xD800000)
+#endif /* XSHAL_IOBLOCK_BYPASS_VADDR */
+
+/*  These devices might be accessed cached:  */
+#ifdef XSHAL_IOBLOCK_CACHED_VADDR
+# define XTBOARD_PCI_MEM_CACHED_VADDR	(XSHAL_IOBLOCK_CACHED_VADDR+0x0000000)
+# define XTBOARD_FLASH_CACHED_VADDR	(XSHAL_IOBLOCK_CACHED_VADDR+0x8000000)
+# define XTBOARD_ASRAM_CACHED_VADDR	(XSHAL_IOBLOCK_CACHED_VADDR+0xD400000)
+# define XTBOARD_EEPROM_CACHED_VADDR	(XSHAL_IOBLOCK_CACHED_VADDR+0xD600000)
+# define XTBOARD_EPROM_CACHED_VADDR	(XSHAL_IOBLOCK_CACHED_VADDR+0xD800000)
+#endif /* XSHAL_IOBLOCK_CACHED_VADDR */
+
+
+/*  System ROM:  */
+#define XTBOARD_ROM_SIZE		XSHAL_ROM_SIZE
+#ifdef XSHAL_ROM_VADDR
+#define XTBOARD_ROM_VADDR		XSHAL_ROM_VADDR
+#endif
+#ifdef XSHAL_ROM_PADDR
+#define XTBOARD_ROM_PADDR		XSHAL_ROM_PADDR
+#endif
+
+/*  System RAM:  */
+#define XTBOARD_RAM_SIZE		XSHAL_RAM_SIZE
+#ifdef XSHAL_RAM_VADDR
+#define XTBOARD_RAM_VADDR		XSHAL_RAM_VADDR
+#endif
+#ifdef XSHAL_RAM_PADDR
+#define XTBOARD_RAM_PADDR		XSHAL_RAM_PADDR
+#endif
+#define XTBOARD_RAM_BYPASS_VADDR	XSHAL_RAM_BYPASS_VADDR
+#define XTBOARD_RAM_BYPASS_PADDR	XSHAL_RAM_BYPASS_PADDR
+
+
+
+/*
+ *  Things that depend on device addresses.
+ */
+
+
+#define XTBOARD_CACHEATTR_WRITEBACK	XSHAL_XT2000_CACHEATTR_WRITEBACK
+#define XTBOARD_CACHEATTR_WRITEALLOC	XSHAL_XT2000_CACHEATTR_WRITEALLOC
+#define XTBOARD_CACHEATTR_WRITETHRU	XSHAL_XT2000_CACHEATTR_WRITETHRU
+#define XTBOARD_CACHEATTR_BYPASS	XSHAL_XT2000_CACHEATTR_BYPASS
+#define XTBOARD_CACHEATTR_DEFAULT	XSHAL_XT2000_CACHEATTR_DEFAULT
+
+#define XTBOARD_BUSINT_PIPE_REGIONS	XSHAL_XT2000_PIPE_REGIONS
+#define XTBOARD_BUSINT_SDRAM_REGIONS	XSHAL_XT2000_SDRAM_REGIONS
+
+
+
+/*
+ *  BusLogic (FPGA) registers.
+ *  All these registers are normally accessed using 32-bit loads/stores.
+ */
+
+/*  Register offsets:  */
+#define XT2000_DATECD_OFS	0x00	/* date code (read-only) */
+#define XT2000_STSREG_OFS	0x04	/* status (read-only) */
+#define XT2000_SYSLED_OFS	0x08	/* system LED */
+#define XT2000_WRPROT_OFS	0x0C	/* write protect */
+#define XT2000_SWRST_OFS	0x10	/* software reset */
+#define XT2000_SYSRST_OFS	0x14	/* system (peripherals) reset */
+#define XT2000_IMASK_OFS	0x18	/* interrupt mask */
+#define XT2000_ISTAT_OFS	0x1C	/* interrupt status */
+#define XT2000_V3CFG_OFS	0x20	/* V3 config (V320 PCI) */
+
+/*  Physical register addresses:  */
+#ifdef XT2000_FPGAREGS_PADDR
+#define XT2000_DATECD_PADDR	(XT2000_FPGAREGS_PADDR+XT2000_DATECD_OFS)
+#define XT2000_STSREG_PADDR	(XT2000_FPGAREGS_PADDR+XT2000_STSREG_OFS)
+#define XT2000_SYSLED_PADDR	(XT2000_FPGAREGS_PADDR+XT2000_SYSLED_OFS)
+#define XT2000_WRPROT_PADDR	(XT2000_FPGAREGS_PADDR+XT2000_WRPROT_OFS)
+#define XT2000_SWRST_PADDR	(XT2000_FPGAREGS_PADDR+XT2000_SWRST_OFS)
+#define XT2000_SYSRST_PADDR	(XT2000_FPGAREGS_PADDR+XT2000_SYSRST_OFS)
+#define XT2000_IMASK_PADDR	(XT2000_FPGAREGS_PADDR+XT2000_IMASK_OFS)
+#define XT2000_ISTAT_PADDR	(XT2000_FPGAREGS_PADDR+XT2000_ISTAT_OFS)
+#define XT2000_V3CFG_PADDR	(XT2000_FPGAREGS_PADDR+XT2000_V3CFG_OFS)
+#endif
+
+/*  Virtual register addresses:  */
+#ifdef XT2000_FPGAREGS_VADDR
+#define XT2000_DATECD_VADDR	(XT2000_FPGAREGS_VADDR+XT2000_DATECD_OFS)
+#define XT2000_STSREG_VADDR	(XT2000_FPGAREGS_VADDR+XT2000_STSREG_OFS)
+#define XT2000_SYSLED_VADDR	(XT2000_FPGAREGS_VADDR+XT2000_SYSLED_OFS)
+#define XT2000_WRPROT_VADDR	(XT2000_FPGAREGS_VADDR+XT2000_WRPROT_OFS)
+#define XT2000_SWRST_VADDR	(XT2000_FPGAREGS_VADDR+XT2000_SWRST_OFS)
+#define XT2000_SYSRST_VADDR	(XT2000_FPGAREGS_VADDR+XT2000_SYSRST_OFS)
+#define XT2000_IMASK_VADDR	(XT2000_FPGAREGS_VADDR+XT2000_IMASK_OFS)
+#define XT2000_ISTAT_VADDR	(XT2000_FPGAREGS_VADDR+XT2000_ISTAT_OFS)
+#define XT2000_V3CFG_VADDR	(XT2000_FPGAREGS_VADDR+XT2000_V3CFG_OFS)
+/*  Register access (for C code):  */
+#define XT2000_DATECD_REG	(*(volatile unsigned*) XT2000_DATECD_VADDR)
+#define XT2000_STSREG_REG	(*(volatile unsigned*) XT2000_STSREG_VADDR)
+#define XT2000_SYSLED_REG	(*(volatile unsigned*) XT2000_SYSLED_VADDR)
+#define XT2000_WRPROT_REG	(*(volatile unsigned*) XT2000_WRPROT_VADDR)
+#define XT2000_SWRST_REG	(*(volatile unsigned*) XT2000_SWRST_VADDR)
+#define XT2000_SYSRST_REG	(*(volatile unsigned*) XT2000_SYSRST_VADDR)
+#define XT2000_IMASK_REG	(*(volatile unsigned*) XT2000_IMASK_VADDR)
+#define XT2000_ISTAT_REG	(*(volatile unsigned*) XT2000_ISTAT_VADDR)
+#define XT2000_V3CFG_REG	(*(volatile unsigned*) XT2000_V3CFG_VADDR)
+#endif
+
+/*  DATECD (date code) bit fields:  */
+
+/*  BCD-coded month (01..12):  */
+#define XT2000_DATECD_MONTH_SHIFT	24
+#define XT2000_DATECD_MONTH_BITS	8
+#define XT2000_DATECD_MONTH_MASK	0xFF000000
+/*  BCD-coded day (01..31):  */
+#define XT2000_DATECD_DAY_SHIFT		16
+#define XT2000_DATECD_DAY_BITS		8
+#define XT2000_DATECD_DAY_MASK		0x00FF0000
+/*  BCD-coded year (2001..9999):  */
+#define XT2000_DATECD_YEAR_SHIFT	0
+#define XT2000_DATECD_YEAR_BITS		16
+#define XT2000_DATECD_YEAR_MASK		0x0000FFFF
+
+/*  STSREG (status) bit fields:  */
+
+/*  Switch SW3 setting bit fields (0=off/up, 1=on/down):  */
+#define XT2000_STSREG_SW3_SHIFT		0
+#define XT2000_STSREG_SW3_BITS		4
+#define XT2000_STSREG_SW3_MASK		0x0000000F
+/*  Boot-select bits of switch SW3:  */
+#define XT2000_STSREG_BOOTSEL_SHIFT	0
+#define XT2000_STSREG_BOOTSEL_BITS	2
+#define XT2000_STSREG_BOOTSEL_MASK	0x00000003
+/*  Boot-select values:  */
+#define XT2000_STSREG_BOOTSEL_FLASH	0
+#define XT2000_STSREG_BOOTSEL_EPROM16	1
+#define XT2000_STSREG_BOOTSEL_PROM8	2
+#define XT2000_STSREG_BOOTSEL_ASRAM	3
+/*  User-defined bits of switch SW3:  */
+#define XT2000_STSREG_SW3_2_SHIFT	2
+#define XT2000_STSREG_SW3_2_MASK	0x00000004
+#define XT2000_STSREG_SW3_3_SHIFT	3
+#define XT2000_STSREG_SW3_3_MASK	0x00000008
+
+/*  SYSLED (system LED) bit fields:  */
+
+/*  LED control bit (0=off, 1=on):  */
+#define XT2000_SYSLED_LEDON_SHIFT	0
+#define XT2000_SYSLED_LEDON_MASK	0x00000001
+
+/*  WRPROT (write protect) bit fields (0=writable, 1=write-protected [default]):  */
+
+/*  Flash write protect:  */
+#define XT2000_WRPROT_FLWP_SHIFT	0
+#define XT2000_WRPROT_FLWP_MASK		0x00000001
+/*  Reserved but present write protect bits:  */
+#define XT2000_WRPROT_WRP_SHIFT		1
+#define XT2000_WRPROT_WRP_BITS		7
+#define XT2000_WRPROT_WRP_MASK		0x000000FE
+
+/*  SWRST (software reset; allows s/w to generate power-on equivalent reset):  */
+
+/*  Software reset bits:  */
+#define XT2000_SWRST_SWR_SHIFT		0
+#define XT2000_SWRST_SWR_BITS		16
+#define XT2000_SWRST_SWR_MASK		0x0000FFFF
+/*  Software reset value -- writing this value resets the board:  */
+#define XT2000_SWRST_RESETVALUE		0x0000DEAD
+
+/*  SYSRST (system reset; controls reset of individual peripherals):  */
+
+/*  All-device reset:  */
+#define XT2000_SYSRST_ALL_SHIFT		0
+#define XT2000_SYSRST_ALL_BITS		4
+#define XT2000_SYSRST_ALL_MASK		0x0000000F
+/*  HDSP-2534 LED display reset (1=reset, 0=nothing):  */
+#define XT2000_SYSRST_LED_SHIFT		0
+#define XT2000_SYSRST_LED_MASK		0x00000001
+/*  Sonic DP83934 Ethernet controller reset (1=reset, 0=nothing):  */
+#define XT2000_SYSRST_SONIC_SHIFT	1
+#define XT2000_SYSRST_SONIC_MASK	0x00000002
+/*  DP16552 DUART reset (1=reset, 0=nothing):  */
+#define XT2000_SYSRST_DUART_SHIFT	2
+#define XT2000_SYSRST_DUART_MASK	0x00000004
+/*  V3 V320 PCI bridge controller reset (1=reset, 0=nothing):  */
+#define XT2000_SYSRST_V3_SHIFT		3
+#define XT2000_SYSRST_V3_MASK		0x00000008
+
+/*  IMASK (interrupt mask; 0=disable, 1=enable):  */
+/*  ISTAT (interrupt status; 0=inactive, 1=pending):  */
+
+/*  PCI INTP interrupt:  */
+#define XT2000_INTMUX_PCI_INTP_SHIFT	2
+#define XT2000_INTMUX_PCI_INTP_MASK	0x00000004
+/*  PCI INTS interrupt:  */
+#define XT2000_INTMUX_PCI_INTS_SHIFT	3
+#define XT2000_INTMUX_PCI_INTS_MASK	0x00000008
+/*  PCI INTD interrupt:  */
+#define XT2000_INTMUX_PCI_INTD_SHIFT	4
+#define XT2000_INTMUX_PCI_INTD_MASK	0x00000010
+/*  V320 PCI controller interrupt:  */
+#define XT2000_INTMUX_V3_SHIFT		5
+#define XT2000_INTMUX_V3_MASK		0x00000020
+/*  PCI ENUM interrupt:  */
+#define XT2000_INTMUX_PCI_ENUM_SHIFT	6
+#define XT2000_INTMUX_PCI_ENUM_MASK	0x00000040
+/*  PCI DEG interrupt:  */
+#define XT2000_INTMUX_PCI_DEG_SHIFT	7
+#define XT2000_INTMUX_PCI_DEG_MASK	0x00000080
+
+/*  V3CFG (V3 config, V320 PCI controller):  */
+
+/*  V3 address control (0=pass-thru, 1=V3 address bits 31:28 set to 4'b0001 [default]):  */
+#define XT2000_V3CFG_V3ADC_SHIFT	0
+#define XT2000_V3CFG_V3ADC_MASK		0x00000001
+
+/* I2C Devices */
+
+#define	XT2000_I2C_RTC_ID		0x68
+#define	XT2000_I2C_NVRAM0_ID		0x56	/* 1st 256 byte block */
+#define	XT2000_I2C_NVRAM1_ID		0x57	/* 2nd 256 byte block */
+
+/*  NVRAM Board Info structure:  */
+
+#define XT2000_NVRAM_SIZE		512
+
+#define XT2000_NVRAM_BINFO_START	0x100
+#define XT2000_NVRAM_BINFO_SIZE		0x20
+#define XT2000_NVRAM_BINFO_VERSION	0x10	/* version 1.0 */
+#if 0
+#define XT2000_NVRAM_BINFO_VERSION_OFFSET	0x00
+#define XT2000_NVRAM_BINFO_VERSION_SIZE			0x1
+#define XT2000_NVRAM_BINFO_ETH_ADDR_OFFSET	0x02
+#define XT2000_NVRAM_BINFO_ETH_ADDR_SIZE		0x6
+#define XT2000_NVRAM_BINFO_SN_OFFSET		0x10
+#define XT2000_NVRAM_BINFO_SN_SIZE			0xE
+#define	XT2000_NVRAM_BINFO_CRC_OFFSET		0x1E
+#define	XT2000_NVRAM_BINFO_CRC_SIZE			0x2
+#endif /*0*/
+
+#if !defined(__ASSEMBLY__) && !defined(_NOCLANGUAGE)
+typedef struct xt2000_nvram_binfo {
+    unsigned char	version;
+    unsigned char	reserved1;
+    unsigned char	eth_addr[6];
+    unsigned char	reserved8[8];
+    unsigned char	serialno[14];
+    unsigned char	crc[2];		/* 16-bit CRC */
+} xt2000_nvram_binfo;
+#endif /*!__ASSEMBLY__ && !_NOCLANGUAGE*/
+
+
+#endif /*_INC_XT2000_H_*/
+
diff --git a/include/asm-xtensa/xtensa/xtboard.h b/include/asm-xtensa/xtensa/xtboard.h
new file mode 100644
index 00000000000..22469c17530
--- /dev/null
+++ b/include/asm-xtensa/xtensa/xtboard.h
@@ -0,0 +1,120 @@
+#ifndef _xtboard_h_included_
+#define _xtboard_h_included_
+
+/*
+ * THIS FILE IS GENERATED -- DO NOT MODIFY BY HAND
+ *
+ *  xtboard.h  --  Routines for getting useful information from the board.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2002 Tensilica Inc.
+ */
+
+
+#include <xtensa/xt2000.h>
+
+#define	XTBOARD_RTC_ERROR	-1
+#define	XTBOARD_RTC_STOPPED	-2
+
+
+/*  xt2000-i2cdev.c:  */
+typedef void XtboardDelayFunc( unsigned );
+extern XtboardDelayFunc* xtboard_set_nsdelay_func( XtboardDelayFunc *delay_fn );
+extern int xtboard_i2c_read (unsigned id, unsigned char *buf, unsigned addr, unsigned size);
+extern int xtboard_i2c_write(unsigned id, unsigned char *buf, unsigned addr, unsigned size);
+extern int xtboard_i2c_wait_nvram_ack(unsigned id, unsigned swtimer);
+
+/*  xtboard.c:  */
+extern int xtboard_nvram_read (unsigned addr, unsigned len, unsigned char *buf);
+extern int xtboard_nvram_write(unsigned addr, unsigned len, unsigned char *buf);
+extern int xtboard_nvram_binfo_read (xt2000_nvram_binfo *buf);
+extern int xtboard_nvram_binfo_write(xt2000_nvram_binfo *buf);
+extern int xtboard_nvram_binfo_valid(xt2000_nvram_binfo *buf);
+extern int xtboard_ethermac_get(unsigned char *buf);
+extern int xtboard_ethermac_set(unsigned char *buf);
+
+/*+*----------------------------------------------------------------------------
+/ Function: xtboard_get_rtc_time
+/
+/ Description:  Get time stored in real-time clock.
+/
+/ Returns: time in seconds stored in real-time clock.
+/-**----------------------------------------------------------------------------*/
+
+extern unsigned xtboard_get_rtc_time(void);
+
+/*+*----------------------------------------------------------------------------
+/ Function: xtboard_set_rtc_time
+/
+/ Description:  Set time stored in real-time clock.
+/
+/ Parameters: 	time -- time in seconds to store to real-time clock
+/
+/ Returns: 0 on success, xtboard_i2c_write() error code otherwise.
+/-**----------------------------------------------------------------------------*/
+
+extern int xtboard_set_rtc_time(unsigned time);
+
+
+/*  xtfreq.c:  */
+/*+*----------------------------------------------------------------------------
+/ Function: xtboard_measure_sys_clk
+/
+/ Description:  Get frequency of system clock.
+/
+/ Parameters:	none
+/
+/ Returns: frequency of system clock.
+/-**----------------------------------------------------------------------------*/
+
+extern unsigned xtboard_measure_sys_clk(void);
+
+
+#if 0	/* old stuff from xtboard.c: */
+
+/*+*----------------------------------------------------------------------------
+/ Function: xtboard_nvram valid
+/
+/ Description:  Determines if data in NVRAM is valid.
+/
+/ Parameters:	delay -- 10us delay function
+/
+/ Returns: 1 if NVRAM is valid, 0 otherwise
+/-**----------------------------------------------------------------------------*/
+
+extern unsigned xtboard_nvram_valid(void (*delay)( void ));
+
+/*+*----------------------------------------------------------------------------
+/ Function: xtboard_get_nvram_contents
+/
+/ Description:  Returns contents of NVRAM.
+/
+/ Parameters: 	buf -- buffer to NVRAM contents.
+/		delay -- 10us delay function
+/
+/ Returns: 1 if NVRAM is valid, 0 otherwise
+/-**----------------------------------------------------------------------------*/
+
+extern unsigned xtboard_get_nvram_contents(unsigned char *buf, void (*delay)( void ));
+
+/*+*----------------------------------------------------------------------------
+/ Function: xtboard_get_ether_addr
+/
+/ Description:  Returns ethernet address of board.
+/
+/ Parameters: 	buf -- buffer to store ethernet address
+/		delay -- 10us delay function
+/
+/ Returns: nothing.
+/-**----------------------------------------------------------------------------*/
+
+extern void xtboard_get_ether_addr(unsigned char *buf, void (*delay)( void ));
+
+#endif /*0*/
+
+
+#endif /*_xtboard_h_included_*/
+
-- 
cgit v1.2.3-70-g09d2


From 7282bee78798294bb1f0211a842cdb9f4872db3d Mon Sep 17 00:00:00 2001
From: Chris Zankel <czankel@tensilica.com>
Date: Thu, 23 Jun 2005 22:01:33 -0700
Subject: [PATCH] xtensa: Architecture support for Tensilica Xtensa Part 8

The attached patches provides part 8 of an architecture implementation
for the Tensilica Xtensa CPU series.

Signed-off-by: Chris Zankel <chris@zankel.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/xtensa/configs/iss_defconfig          | 531 ++++++++++++++++++
 arch/xtensa/platform-iss/Makefile          |  13 +
 arch/xtensa/platform-iss/console.c         | 303 ++++++++++
 arch/xtensa/platform-iss/io.c              |  32 ++
 arch/xtensa/platform-iss/network.c         | 855 +++++++++++++++++++++++++++++
 arch/xtensa/platform-iss/setup.c           | 112 ++++
 include/asm-xtensa/platform-iss/hardware.h |  29 +
 include/asm-xtensa/platform.h              |  92 ++++
 8 files changed, 1967 insertions(+)
 create mode 100644 arch/xtensa/configs/iss_defconfig
 create mode 100644 arch/xtensa/platform-iss/Makefile
 create mode 100644 arch/xtensa/platform-iss/console.c
 create mode 100644 arch/xtensa/platform-iss/io.c
 create mode 100644 arch/xtensa/platform-iss/network.c
 create mode 100644 arch/xtensa/platform-iss/setup.c
 create mode 100644 include/asm-xtensa/platform-iss/hardware.h
 create mode 100644 include/asm-xtensa/platform.h

(limited to 'include')

diff --git a/arch/xtensa/configs/iss_defconfig b/arch/xtensa/configs/iss_defconfig
new file mode 100644
index 00000000000..802621dd486
--- /dev/null
+++ b/arch/xtensa/configs/iss_defconfig
@@ -0,0 +1,531 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.11-rc2
+# Fri Feb 25 19:21:24 2005
+#
+CONFIG_FRAME_POINTER=y
+CONFIG_XTENSA=y
+# CONFIG_UID16 is not set
+CONFIG_RWSEM_XCHGADD_ALGORITHM=y
+CONFIG_HAVE_DEC_LOCK=y
+CONFIG_GENERIC_HARDIRQS=y
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_CLEAN_COMPILE=y
+CONFIG_BROKEN_ON_SMP=y
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+CONFIG_SWAP=y
+CONFIG_SYSVIPC=y
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_HOTPLUG is not set
+# CONFIG_KOBJECT_UEVENT is not set
+# CONFIG_IKCONFIG is not set
+CONFIG_EMBEDDED=y
+CONFIG_KALLSYMS=y
+# CONFIG_KALLSYMS_ALL is not set
+# CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_FUTEX=y
+CONFIG_EPOLL=y
+# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
+CONFIG_SHMEM=y
+CONFIG_CC_ALIGN_FUNCTIONS=0
+CONFIG_CC_ALIGN_LABELS=0
+CONFIG_CC_ALIGN_LOOPS=0
+CONFIG_CC_ALIGN_JUMPS=0
+# CONFIG_TINY_SHMEM is not set
+
+#
+# Loadable module support
+#
+# CONFIG_MODULES is not set
+
+#
+# Processor type and features
+#
+CONFIG_XTENSA_ARCH_LINUX_BE=y
+# CONFIG_XTENSA_ARCH_LINUX_LE is not set
+# CONFIG_XTENSA_ARCH_LINUX_TEST is not set
+# CONFIG_XTENSA_ARCH_S5 is not set
+# CONFIG_XTENSA_CUSTOM is not set
+CONFIG_MMU=y
+# CONFIG_XTENSA_UNALIGNED_USER is not set
+# CONFIG_PREEMPT is not set
+# CONFIG_MATH_EMULATION is not set
+# CONFIG_HIGHMEM is not set
+
+#
+# Platform options
+#
+CONFIG_XTENSA_PLATFORM_ISS=y
+# CONFIG_XTENSA_PLATFORM_XT2000 is not set
+# CONFIG_XTENSA_PLATFORM_ARUBA is not set
+# CONFIG_XTENSA_CALIBRATE_CCOUNT is not set
+CONFIG_XTENSA_CPU_CLOCK=10
+# CONFIG_GENERIC_CALIBRATE_DELAY is not set
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttyS0,38400 eth0=tuntap,,tap0 ip=192.168.168.5:192.168.168.1 root=nfs nfsroot=192.168.168.1:/opt/montavista/pro/devkit/xtensa/linux_be/target"
+CONFIG_SERIAL_CONSOLE=y
+CONFIG_XTENSA_ISS_NETWORK=y
+
+#
+# Bus options
+#
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+# CONFIG_PCCARD is not set
+
+#
+# PC-card bridges
+#
+
+#
+# PCI Hotplug Support
+#
+
+#
+# Exectuable file formats
+#
+CONFIG_KCORE_ELF=y
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+# CONFIG_STANDALONE is not set
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+# CONFIG_FW_LOADER is not set
+# CONFIG_DEBUG_DRIVER is not set
+
+#
+# Memory Technology Devices (MTD)
+#
+# CONFIG_MTD is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+
+#
+# Block devices
+#
+# CONFIG_BLK_DEV_FD is not set
+# CONFIG_BLK_DEV_COW_COMMON is not set
+# CONFIG_BLK_DEV_LOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+# CONFIG_BLK_DEV_RAM is not set
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_INITRAMFS_SOURCE=""
+# CONFIG_CDROM_PKTCDVD is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+# CONFIG_IOSCHED_CFQ is not set
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_SCSI is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+# CONFIG_MD is not set
+
+#
+# Fusion MPT device support
+#
+
+#
+# IEEE 1394 (FireWire) support
+#
+
+#
+# I2O device support
+#
+
+#
+# Networking support
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+# CONFIG_PACKET_MMAP is not set
+# CONFIG_NETLINK_DEV is not set
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+CONFIG_IP_PNP_BOOTP=y
+CONFIG_IP_PNP_RARP=y
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_TUNNEL is not set
+# CONFIG_IP_TCPDIAG is not set
+# CONFIG_IP_TCPDIAG_IPV6 is not set
+# CONFIG_IPV6 is not set
+# CONFIG_NETFILTER is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+# CONFIG_SCTP_HMAC_NONE is not set
+# CONFIG_SCTP_HMAC_SHA1 is not set
+# CONFIG_SCTP_HMAC_MD5 is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+
+#
+# QoS and/or fair queueing
+#
+# CONFIG_NET_SCHED is not set
+# CONFIG_NET_SCH_CLK_JIFFIES is not set
+# CONFIG_NET_SCH_CLK_GETTIMEOFDAY is not set
+# CONFIG_NET_SCH_CLK_CPU is not set
+# CONFIG_NET_CLS_ROUTE is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_NETDEVICES is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+
+#
+# Userland interfaces
+#
+# CONFIG_INPUT_MOUSEDEV is not set
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_TSDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input I/O drivers
+#
+# CONFIG_GAMEPORT is not set
+CONFIG_SOUND_GAMEPORT=y
+# CONFIG_SERIO is not set
+# CONFIG_SERIO_I8042 is not set
+
+#
+# Input Device Drivers
+#
+# CONFIG_INPUT_KEYBOARD is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Character devices
+#
+CONFIG_VT=y
+CONFIG_VT_CONSOLE=y
+CONFIG_HW_CONSOLE=y
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_UNIX98_PTYS=y
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+CONFIG_WATCHDOG=y
+CONFIG_WATCHDOG_NOWAYOUT=y
+
+#
+# Watchdog Device Drivers
+#
+CONFIG_SOFT_WATCHDOG=y
+# CONFIG_RTC is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_DRM is not set
+# CONFIG_RAW_DRIVER is not set
+
+#
+# I2C support
+#
+# CONFIG_I2C is not set
+
+#
+# Dallas's 1-wire bus
+#
+# CONFIG_W1 is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+# CONFIG_FB is not set
+
+#
+# Console display driver support
+#
+# CONFIG_VGA_CONSOLE is not set
+CONFIG_DUMMY_CONSOLE=y
+# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+
+#
+# USB support
+#
+# CONFIG_USB_ARCH_HAS_HCD is not set
+# CONFIG_USB_ARCH_HAS_OHCI is not set
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support' may also be needed; see USB_STORAGE Help for more information
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# InfiniBand support
+#
+# CONFIG_INFINIBAND is not set
+
+#
+# File systems
+#
+# CONFIG_EXT2_FS is not set
+# CONFIG_EXT3_FS is not set
+# CONFIG_JBD is not set
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_ROMFS_FS is not set
+# CONFIG_QUOTA is not set
+# CONFIG_DNOTIFY is not set
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+# CONFIG_MSDOS_FS is not set
+# CONFIG_VFAT_FS is not set
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_KCORE=y
+CONFIG_SYSFS=y
+CONFIG_DEVFS_FS=y
+CONFIG_DEVFS_MOUNT=y
+# CONFIG_DEVFS_DEBUG is not set
+# CONFIG_DEVPTS_FS_XATTR is not set
+CONFIG_TMPFS=y
+# CONFIG_TMPFS_XATTR is not set
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_RAMFS=y
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+
+#
+# Network File Systems
+#
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V4 is not set
+CONFIG_NFS_DIRECTIO=y
+# CONFIG_NFSD is not set
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+# CONFIG_EXPORTFS is not set
+CONFIG_SUNRPC=y
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+# CONFIG_SMB_FS is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+
+#
+# Native Language Support
+#
+# CONFIG_NLS is not set
+
+#
+# Kernel hacking
+#
+CONFIG_DEBUG_KERNEL=y
+# CONFIG_DEBUG_STACKOVERFLOW is not set
+# CONFIG_DEBUG_SLAB is not set
+# CONFIG_MAGIC_SYSRQ is not set
+# CONFIG_DEBUG_SPINLOCK is not set
+# CONFIG_DEBUG_PAGEALLOC is not set
+# CONFIG_DEBUG_INFO is not set
+# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
+# CONFIG_KGDB is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+
+#
+# Cryptographic options
+#
+# CONFIG_CRYPTO is not set
+
+#
+# Hardware crypto devices
+#
+
+#
+# Library routines
+#
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC32 is not set
+# CONFIG_LIBCRC32C is not set
diff --git a/arch/xtensa/platform-iss/Makefile b/arch/xtensa/platform-iss/Makefile
new file mode 100644
index 00000000000..5b394e9620e
--- /dev/null
+++ b/arch/xtensa/platform-iss/Makefile
@@ -0,0 +1,13 @@
+# $Id: Makefile,v 1.1.1.1 2002/08/28 16:10:14 aroll Exp $
+#
+# Makefile for the Xtensa Instruction Set Simulator (ISS)
+# "prom monitor" library routines under Linux.
+#
+# Note! Dependencies are done automagically by 'make dep', which also
+# removes any old dependencies. DON'T put your own dependencies here
+# unless it's something special (ie not a .c file).
+#
+# Note 2! The CFLAGS definitions are in the main makefile...
+
+obj-y			= io.o console.o setup.o network.o
+
diff --git a/arch/xtensa/platform-iss/console.c b/arch/xtensa/platform-iss/console.c
new file mode 100644
index 00000000000..9e2b53f6a90
--- /dev/null
+++ b/arch/xtensa/platform-iss/console.c
@@ -0,0 +1,303 @@
+/*
+ * arch/xtensa/platform-iss/console.c
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001-2005 Tensilica Inc.
+ *   Authors	Christian Zankel, Joe Taylor
+ */
+
+#include <linux/module.h>
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/console.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/mm.h>
+#include <linux/major.h>
+#include <linux/param.h>
+#include <linux/serial.h>
+#include <linux/serialP.h>
+#include <linux/console.h>
+
+#include <asm/uaccess.h>
+#include <asm/irq.h>
+
+#include <xtensa/simcall.h>
+
+#include <linux/tty.h>
+#include <linux/tty_flip.h>
+
+#ifdef SERIAL_INLINE
+#define _INLINE_ inline
+#endif
+
+#define SERIAL_MAX_NUM_LINES 1
+#define SERIAL_TIMER_VALUE (20 * HZ)
+
+static struct tty_driver *serial_driver;
+static struct timer_list serial_timer;
+
+static DEFINE_SPINLOCK(timer_lock);
+
+int errno;
+
+static int __simc (int a, int b, int c, int d, int e, int f)
+{
+	int ret;
+	__asm__ __volatile__ ("simcall\n"
+			"mov %0, a2\n"
+			"mov %1, a3\n" : "=a" (ret), "=a" (errno)
+			: : "a2", "a3");
+	return ret;
+}
+
+static char *serial_version = "0.1";
+static char *serial_name = "ISS serial driver";
+
+/*
+ * This routine is called whenever a serial port is opened.  It
+ * enables interrupts for a serial port, linking in its async structure into
+ * the IRQ chain.   It also performs the serial-specific
+ * initialization for the tty structure.
+ */
+
+static void rs_poll(unsigned long);
+
+static int rs_open(struct tty_struct *tty, struct file * filp)
+{
+	int line = tty->index;
+
+	if ((line < 0) || (line >= SERIAL_MAX_NUM_LINES))
+		return -ENODEV;
+
+	spin_lock(&timer_lock);
+
+	if (tty->count == 1) {
+		init_timer(&serial_timer);
+		serial_timer.data = (unsigned long) tty;
+		serial_timer.function = rs_poll;
+		mod_timer(&serial_timer, jiffies + SERIAL_TIMER_VALUE);
+	}
+	spin_unlock(&timer_lock);
+
+	return 0;
+}
+
+
+/*
+ * ------------------------------------------------------------
+ * iss_serial_close()
+ *
+ * This routine is called when the serial port gets closed.  First, we
+ * wait for the last remaining data to be sent.  Then, we unlink its
+ * async structure from the interrupt chain if necessary, and we free
+ * that IRQ if nothing is left in the chain.
+ * ------------------------------------------------------------
+ */
+static void rs_close(struct tty_struct *tty, struct file * filp)
+{
+	spin_lock(&timer_lock);
+	if (tty->count == 1)
+		del_timer_sync(&serial_timer);
+	spin_unlock(&timer_lock);
+}
+
+
+static int rs_write(struct tty_struct * tty,
+		    const unsigned char *buf, int count)
+{
+	/* see drivers/char/serialX.c to reference original version */
+
+	__simc (SYS_write, 1, (unsigned long)buf, count, 0, 0);
+	return count;
+}
+
+static void rs_poll(unsigned long priv)
+{
+	struct tty_struct* tty = (struct tty_struct*) priv;
+
+	struct timeval tv = { .tv_sec = 0, .tv_usec = 0 };
+	int i = 0;
+	unsigned char c;
+
+	spin_lock(&timer_lock);
+
+	while (__simc(SYS_select_one, 0, XTISS_SELECT_ONE_READ, (int)&tv,0,0)){
+		__simc (SYS_read, 0, (unsigned long)&c, 1, 0, 0);
+		tty->flip.count++;
+		*tty->flip.char_buf_ptr++ = c;
+		*tty->flip.flag_buf_ptr++ = TTY_NORMAL;
+		i++;
+	}
+
+	if (i)
+		tty_flip_buffer_push(tty);
+
+
+	mod_timer(&serial_timer, jiffies + SERIAL_TIMER_VALUE);
+	spin_unlock(&timer_lock);
+}
+
+
+static void rs_put_char(struct tty_struct *tty, unsigned char ch)
+{
+	char buf[2];
+
+	if (!tty)
+		return;
+
+	buf[0] = ch;
+	buf[1] = '\0';		/* Is this NULL necessary? */
+	__simc (SYS_write, 1, (unsigned long) buf, 1, 0, 0);
+}
+
+static void rs_flush_chars(struct tty_struct *tty)
+{
+}
+
+static int rs_write_room(struct tty_struct *tty)
+{
+	/* Let's say iss can always accept 2K characters.. */
+	return 2 * 1024;
+}
+
+static int rs_chars_in_buffer(struct tty_struct *tty)
+{
+	/* the iss doesn't buffer characters */
+	return 0;
+}
+
+static void rs_hangup(struct tty_struct *tty)
+{
+	/* Stub, once again.. */
+}
+
+static void rs_wait_until_sent(struct tty_struct *tty, int timeout)
+{
+	/* Stub, once again.. */
+}
+
+static int rs_read_proc(char *page, char **start, off_t off, int count,
+			int *eof, void *data)
+{
+	int len = 0;
+	off_t begin = 0;
+
+	len += sprintf(page, "serinfo:1.0 driver:%s\n", serial_version);
+	*eof = 1;
+
+	if (off >= len + begin)
+		return 0;
+
+	*start = page + (off - begin);
+	return ((count < begin + len - off) ? count : begin + len - off);
+}
+
+
+int register_serial(struct serial_struct*);
+void unregister_serial(int);
+
+static struct tty_operations serial_ops = {
+	.open = rs_open,
+	.close = rs_close,
+	.write = rs_write,
+	.put_char = rs_put_char,
+	.flush_chars = rs_flush_chars,
+	.write_room = rs_write_room,
+	.chars_in_buffer = rs_chars_in_buffer,
+	.hangup = rs_hangup,
+	.wait_until_sent = rs_wait_until_sent,
+	.read_proc = rs_read_proc
+};
+
+int __init rs_init(void)
+{
+	serial_driver = alloc_tty_driver(1);
+
+	printk ("%s %s\n", serial_name, serial_version);
+
+	/* Initialize the tty_driver structure */
+
+	serial_driver->owner = THIS_MODULE;
+	serial_driver->driver_name = "iss_serial";
+	serial_driver->name = "ttyS";
+	serial_driver->major = TTY_MAJOR;
+	serial_driver->minor_start = 64;
+	serial_driver->type = TTY_DRIVER_TYPE_SERIAL;
+	serial_driver->subtype = SERIAL_TYPE_NORMAL;
+	serial_driver->init_termios = tty_std_termios;
+	serial_driver->init_termios.c_cflag =
+		B9600 | CS8 | CREAD | HUPCL | CLOCAL;
+	serial_driver->flags = TTY_DRIVER_REAL_RAW;
+
+	tty_set_operations(serial_driver, &serial_ops);
+
+	if (tty_register_driver(serial_driver))
+		panic("Couldn't register serial driver\n");
+	return 0;
+}
+
+
+static __exit void rs_exit(void)
+{
+	int error;
+
+	if ((error = tty_unregister_driver(serial_driver)))
+		printk("ISS_SERIAL: failed to unregister serial driver (%d)\n",
+		       error);
+	put_tty_driver(serial_driver);
+}
+
+
+/* We use `late_initcall' instead of just `__initcall' as a workaround for
+ * the fact that (1) simcons_tty_init can't be called before tty_init,
+ * (2) tty_init is called via `module_init', (3) if statically linked,
+ * module_init == device_init, and (4) there's no ordering of init lists.
+ * We can do this easily because simcons is always statically linked, but
+ * other tty drivers that depend on tty_init and which must use
+ * `module_init' to declare their init routines are likely to be broken.
+ */
+
+late_initcall(rs_init);
+
+
+#ifdef CONFIG_SERIAL_CONSOLE
+
+static void iss_console_write(struct console *co, const char *s, unsigned count)
+{
+	int len = strlen(s);
+
+	if (s != 0 && *s != 0)
+		__simc (SYS_write, 1, (unsigned long)s,
+			count < len ? count : len,0,0);
+}
+
+static struct tty_driver* iss_console_device(struct console *c, int *index)
+{
+	*index = c->index;
+	return serial_driver;
+}
+
+
+static struct console sercons = {
+	.name = "ttyS",
+	.write = iss_console_write,
+	.device = iss_console_device,
+	.flags = CON_PRINTBUFFER,
+	.index = -1
+};
+
+static int __init iss_console_init(void)
+{
+	register_console(&sercons);
+	return 0;
+}
+
+console_initcall(iss_console_init);
+
+#endif /* CONFIG_SERIAL_CONSOLE */
+
diff --git a/arch/xtensa/platform-iss/io.c b/arch/xtensa/platform-iss/io.c
new file mode 100644
index 00000000000..5b161a5cb65
--- /dev/null
+++ b/arch/xtensa/platform-iss/io.c
@@ -0,0 +1,32 @@
+/* This file isn't really needed right now. */
+
+#if 0
+
+#include <asm/io.h>
+#include <xtensa/simcall.h>
+
+extern int __simc ();
+
+
+char iss_serial_getc()
+{
+  char c;
+  __simc( SYS_read, 0, &c, 1 );
+  return c;
+}
+
+void iss_serial_putc( char c )
+{
+  __simc( SYS_write, 1, &c, 1 );
+}
+
+void iss_serial_puts( char *s )
+{
+  if( s != 0 && *s != 0 )
+    __simc( SYS_write, 1, s, strlen(s) );
+}
+
+/*#error Need I/O ports to specific hardware!*/
+
+#endif
+
diff --git a/arch/xtensa/platform-iss/network.c b/arch/xtensa/platform-iss/network.c
new file mode 100644
index 00000000000..498d7dced1f
--- /dev/null
+++ b/arch/xtensa/platform-iss/network.c
@@ -0,0 +1,855 @@
+/*
+ *
+ * arch/xtensa/platform-iss/network.c
+ *
+ * Platform specific initialization.
+ *
+ * Authors: Chris Zankel <chris@zankel.net>
+ * Based on work form the UML team.
+ *
+ * Copyright 2005 Tensilica Inc.
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ */
+
+#include <linux/config.h>
+#include <linux/list.h>
+#include <linux/irq.h>
+#include <linux/spinlock.h>
+#include <linux/slab.h>
+#include <linux/timer.h>
+#include <linux/if_ether.h>
+#include <linux/inetdevice.h>
+#include <linux/init.h>
+#include <linux/if_tun.h>
+#include <linux/etherdevice.h>
+#include <linux/interrupt.h>
+#include <linux/ioctl.h>
+#include <linux/bootmem.h>
+#include <linux/ethtool.h>
+#include <linux/rtnetlink.h>
+#include <linux/timer.h>
+
+#include <xtensa/simcall.h>
+
+#define DRIVER_NAME "iss-netdev"
+#define ETH_MAX_PACKET 1500
+#define ETH_HEADER_OTHER 14
+#define ISS_NET_TIMER_VALUE (2 * HZ)
+
+
+static DEFINE_SPINLOCK(opened_lock);
+static LIST_HEAD(opened);
+
+static DEFINE_SPINLOCK(devices_lock);
+static LIST_HEAD(devices);
+
+/* ------------------------------------------------------------------------- */
+
+/* We currently only support the TUNTAP transport protocol. */
+
+#define TRANSPORT_TUNTAP_NAME "tuntap"
+#define TRANSPORT_TUNTAP_MTU ETH_MAX_PACKET
+
+struct tuntap_info {
+	char dev_name[IFNAMSIZ];
+	int fixed_config;
+	unsigned char gw[ETH_ALEN];
+	int fd;
+};
+
+/* ------------------------------------------------------------------------- */
+
+
+/* This structure contains out private information for the driver. */
+
+struct iss_net_private {
+
+	struct list_head device_list;
+	struct list_head opened_list;
+
+	spinlock_t lock;
+	struct net_device *dev;
+	struct platform_device pdev;
+	struct timer_list tl;
+	struct net_device_stats stats;
+
+	struct timer_list timer;
+	unsigned int timer_val;
+
+	int index;
+	int mtu;
+
+	unsigned char mac[ETH_ALEN];
+	int have_mac;
+
+	struct {
+		union {
+			struct tuntap_info tuntap;
+		} info;
+
+		int (*open)(struct iss_net_private *lp);
+		void (*close)(struct iss_net_private *lp);
+		int (*read)(struct iss_net_private *lp, struct sk_buff **skb);
+		int (*write)(struct iss_net_private *lp, struct sk_buff **skb);
+		unsigned short (*protocol)(struct sk_buff *skb);
+		int (*poll)(struct iss_net_private *lp);
+	} tp;
+
+};
+
+/* ======================= ISS SIMCALL INTERFACE =========================== */
+
+/* Note: __simc must _not_ be declared inline! */
+
+static int errno;
+
+static int __simc (int a, int b, int c, int d, int e, int f)
+{
+	int ret;
+	__asm__ __volatile__ ("simcall\n"
+	    		      "mov %0, a2\n"
+			      "mov %1, a3\n" : "=a" (ret), "=a" (errno)
+			      : : "a2", "a3");
+	return ret;
+}
+
+static int inline simc_open(char *file, int flags, int mode)
+{
+	return __simc(SYS_open, (int) file, flags, mode, 0, 0);
+}
+
+static int inline simc_close(int fd)
+{
+	return __simc(SYS_close, fd, 0, 0, 0, 0);
+}
+
+static int inline simc_ioctl(int fd, int request, void *arg)
+{
+	return __simc(SYS_ioctl, fd, request, (int) arg, 0, 0);
+}
+
+static int inline simc_read(int fd, void *buf, size_t count)
+{
+	return __simc(SYS_read, fd, (int) buf, count, 0, 0);
+}
+
+static int inline simc_write(int fd, void *buf, size_t count)
+{
+	return __simc(SYS_write, fd, (int) buf, count, 0, 0);
+}
+
+static int inline simc_poll(int fd)
+{
+	struct timeval tv = { .tv_sec = 0, .tv_usec = 0 };
+
+	return __simc(SYS_select_one, fd, XTISS_SELECT_ONE_READ, (int)&tv,0,0);
+}
+
+/* ================================ HELPERS ================================ */
+
+
+static char *split_if_spec(char *str, ...)
+{
+	char **arg, *end;
+	va_list ap;
+
+	va_start(ap, str);
+	while ((arg = va_arg(ap, char**)) != NULL) {
+		if (*str == '\0')
+			return NULL;
+		end = strchr(str, ',');
+		if (end != str)
+			*arg = str;
+		if (end == NULL)
+			return NULL;
+		*end ++ = '\0';
+		str = end;
+	}
+	va_end(ap);
+	return str;
+}
+
+
+#if 0
+/* Adjust SKB. */
+
+struct sk_buff *ether_adjust_skb(struct sk_buff *skb, int extra)
+{
+	if ((skb != NULL) && (skb_tailroom(skb) < extra)) {
+		struct sk_buff *skb2;
+
+		skb2 = skb_copy_expand(skb, 0, extra, GFP_ATOMIC);
+		dev_kfree_skb(skb);
+		skb = skb2;
+	}
+	if (skb != NULL)
+		skb_put(skb, extra);
+
+	return skb;
+}
+#endif
+
+/* Return the IP address as a string for a given device. */
+
+static void dev_ip_addr(void *d, char *buf, char *bin_buf)
+{
+	struct net_device *dev = d;
+	struct in_device *ip = dev->ip_ptr;
+	struct in_ifaddr *in;
+	u32 addr;
+
+	if ((ip == NULL) || ((in = ip->ifa_list) == NULL)) {
+		printk(KERN_WARNING "Device not assigned an IP address!\n");
+		return;
+	}
+
+	addr = in->ifa_address;
+	sprintf(buf, "%d.%d.%d.%d", addr & 0xff, (addr >> 8) & 0xff,
+		(addr >> 16) & 0xff, addr >> 24);
+
+	if (bin_buf) {
+		bin_buf[0] = addr & 0xff;
+		bin_buf[1] = (addr >> 8) & 0xff;
+		bin_buf[2] = (addr >> 16) & 0xff;
+		bin_buf[3] = addr >> 24;
+	}
+}
+
+/* Set Ethernet address of the specified device. */
+
+static void inline set_ether_mac(void *d, unsigned char *addr)
+{
+	struct net_device *dev = d;
+	memcpy(dev->dev_addr, addr, ETH_ALEN);
+}
+
+
+/* ======================= TUNTAP TRANSPORT INTERFACE ====================== */
+
+static int tuntap_open(struct iss_net_private *lp)
+{
+	struct ifreq ifr;
+	char *dev_name = lp->tp.info.tuntap.dev_name;
+	int err = -EINVAL;
+	int fd;
+
+	/* We currently only support a fixed configuration. */
+
+	if (!lp->tp.info.tuntap.fixed_config)
+		return -EINVAL;
+
+	if ((fd = simc_open("/dev/net/tun", 02, 0)) < 0) {	/* O_RDWR */
+		printk("Failed to open /dev/net/tun, returned %d "
+		       "(errno = %d)\n", fd, errno);
+		return fd;
+	}
+
+	memset(&ifr, 0, sizeof ifr);
+	ifr.ifr_flags = IFF_TAP | IFF_NO_PI;
+	strlcpy(ifr.ifr_name, dev_name, sizeof ifr.ifr_name - 1);
+
+	if ((err = simc_ioctl(fd, TUNSETIFF, (void*) &ifr)) < 0) {
+		printk("Failed to set interface, returned %d "
+		       "(errno = %d)\n", err, errno);
+		simc_close(fd);
+		return err;
+	}
+
+	lp->tp.info.tuntap.fd = fd;
+	return err;
+}
+
+static void tuntap_close(struct iss_net_private *lp)
+{
+#if 0
+	if (lp->tp.info.tuntap.fixed_config)
+		iter_addresses(lp->tp.info.tuntap.dev, close_addr, lp->host.dev_name);
+#endif
+	simc_close(lp->tp.info.tuntap.fd);
+	lp->tp.info.tuntap.fd = -1;
+}
+
+static int tuntap_read (struct iss_net_private *lp, struct sk_buff **skb)
+{
+#if 0
+	*skb = ether_adjust_skb(*skb, ETH_HEADER_OTHER);
+	if (*skb == NULL)
+		return -ENOMEM;
+#endif
+
+	return simc_read(lp->tp.info.tuntap.fd,
+			(*skb)->data, (*skb)->dev->mtu + ETH_HEADER_OTHER);
+}
+
+static int tuntap_write (struct iss_net_private *lp, struct sk_buff **skb)
+{
+	return simc_write(lp->tp.info.tuntap.fd, (*skb)->data, (*skb)->len);
+}
+
+unsigned short tuntap_protocol(struct sk_buff *skb)
+{
+	return eth_type_trans(skb, skb->dev);
+}
+
+static int tuntap_poll(struct iss_net_private *lp)
+{
+	return simc_poll(lp->tp.info.tuntap.fd);
+}
+
+/*
+ * Currently only a device name is supported.
+ * ethX=tuntap[,[mac address][,[device name]]]
+ */
+
+static int tuntap_probe(struct iss_net_private *lp, int index, char *init)
+{
+	const int len = strlen(TRANSPORT_TUNTAP_NAME);
+	char *dev_name = NULL, *mac_str = NULL, *rem = NULL;
+
+	/* Transport should be 'tuntap': ethX=tuntap,mac,dev_name */
+
+	if (strncmp(init, TRANSPORT_TUNTAP_NAME, len))
+		return 0;
+
+	if (*(init += strlen(TRANSPORT_TUNTAP_NAME)) == ',') {
+		if ((rem=split_if_spec(init+1, &mac_str, &dev_name)) != NULL) {
+			printk("Extra garbage on specification : '%s'\n", rem);
+			return 0;
+		}
+	} else if (*init != '\0') {
+		printk("Invalid argument: %s. Skipping device!\n", init);
+		return 0;
+	}
+
+	if (dev_name) {
+		strncpy(lp->tp.info.tuntap.dev_name, dev_name,
+			 sizeof lp->tp.info.tuntap.dev_name);
+		lp->tp.info.tuntap.fixed_config = 1;
+	} else
+		strcpy(lp->tp.info.tuntap.dev_name, TRANSPORT_TUNTAP_NAME);
+
+
+#if 0
+	if (setup_etheraddr(mac_str, lp->mac))
+		lp->have_mac = 1;
+#endif
+	lp->mtu = TRANSPORT_TUNTAP_MTU;
+
+	//lp->info.tuntap.gate_addr = gate_addr;
+
+	lp->tp.info.tuntap.fd = -1;
+
+	lp->tp.open = tuntap_open;
+	lp->tp.close = tuntap_close;
+	lp->tp.read = tuntap_read;
+	lp->tp.write = tuntap_write;
+	lp->tp.protocol = tuntap_protocol;
+	lp->tp.poll = tuntap_poll;
+
+	printk("TUN/TAP backend - ");
+#if 0
+	if (lp->host.gate_addr != NULL)
+		printk("IP = %s", lp->host.gate_addr);
+#endif
+	printk("\n");
+
+	return 1;
+}
+
+/* ================================ ISS NET ================================ */
+
+static int iss_net_rx(struct net_device *dev)
+{
+	struct iss_net_private *lp = dev->priv;
+	int pkt_len;
+	struct sk_buff *skb;
+
+	/* Check if there is any new data. */
+
+	if (lp->tp.poll(lp) == 0)
+		return 0;
+
+	/* Try to allocate memory, if it fails, try again next round. */
+
+	if ((skb = dev_alloc_skb(dev->mtu + 2 + ETH_HEADER_OTHER)) == NULL) {
+		lp->stats.rx_dropped++;
+		return 0;
+	}
+
+	skb_reserve(skb, 2);
+
+	/* Setup skb */
+
+	skb->dev = dev;
+	skb->mac.raw = skb->data;
+	pkt_len = lp->tp.read(lp, &skb);
+	skb_put(skb, pkt_len);
+
+	if (pkt_len > 0) {
+		skb_trim(skb, pkt_len);
+		skb->protocol = lp->tp.protocol(skb);
+	//	netif_rx(skb);
+		netif_rx_ni(skb);
+
+		lp->stats.rx_bytes += skb->len;
+		lp->stats.rx_packets++;
+		return pkt_len;
+	}
+	kfree_skb(skb);
+	return pkt_len;
+}
+
+static int iss_net_poll(void)
+{
+	struct list_head *ele;
+	int err, ret = 0;
+
+	spin_lock(&opened_lock);
+
+	list_for_each(ele, &opened) {
+		struct iss_net_private *lp;
+
+		lp = list_entry(ele, struct iss_net_private, opened_list);
+
+		if (!netif_running(lp->dev))
+			break;
+
+		spin_lock(&lp->lock);
+
+		while ((err = iss_net_rx(lp->dev)) > 0)
+			ret++;
+
+		spin_unlock(&lp->lock);
+
+		if (err < 0) {
+			printk(KERN_ERR "Device '%s' read returned %d, "
+			       "shutting it down\n", lp->dev->name, err);
+			dev_close(lp->dev);
+		} else {
+			// FIXME reactivate_fd(lp->fd, ISS_ETH_IRQ);
+		}
+	}
+
+	spin_unlock(&opened_lock);
+	return ret;
+}
+
+
+static void iss_net_timer(unsigned long priv)
+{
+	struct iss_net_private* lp = (struct iss_net_private*) priv;
+
+	spin_lock(&lp->lock);
+
+	iss_net_poll();
+
+	mod_timer(&lp->timer, jiffies + lp->timer_val);
+
+	spin_unlock(&lp->lock);
+}
+
+
+static int iss_net_open(struct net_device *dev)
+{
+	struct iss_net_private *lp = dev->priv;
+	char addr[sizeof "255.255.255.255\0"];
+	int err;
+
+	spin_lock(&lp->lock);
+
+	if ((err = lp->tp.open(lp)) < 0)
+		goto out;
+
+	if (!lp->have_mac) {
+		dev_ip_addr(dev, addr, &lp->mac[2]);
+		set_ether_mac(dev, lp->mac);
+	}
+
+	netif_start_queue(dev);
+
+	/* clear buffer - it can happen that the host side of the interface
+	 * is full when we gethere. In this case, new data is never queued,
+	 * SIGIOs never arrive, and the net never works.
+	 */
+	while ((err = iss_net_rx(dev)) > 0)
+		;
+
+	spin_lock(&opened_lock);
+	list_add(&lp->opened_list, &opened);
+	spin_unlock(&opened_lock);
+
+	init_timer(&lp->timer);
+	lp->timer_val = ISS_NET_TIMER_VALUE;
+	lp->timer.data = (unsigned long) lp;
+	lp->timer.function = iss_net_timer;
+	mod_timer(&lp->timer, jiffies + lp->timer_val);
+
+out:
+	spin_unlock(&lp->lock);
+	return err;
+}
+
+static int iss_net_close(struct net_device *dev)
+{
+	struct iss_net_private *lp = dev->priv;
+printk("iss_net_close!\n");
+	netif_stop_queue(dev);
+	spin_lock(&lp->lock);
+
+	spin_lock(&opened_lock);
+	list_del(&opened);
+	spin_unlock(&opened_lock);
+
+	del_timer_sync(&lp->timer);
+
+	lp->tp.close(lp);
+
+	spin_unlock(&lp->lock);
+	return 0;
+}
+
+static int iss_net_start_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct iss_net_private *lp = dev->priv;
+	unsigned long flags;
+	int len;
+
+	netif_stop_queue(dev);
+	spin_lock_irqsave(&lp->lock, flags);
+
+	len = lp->tp.write(lp, &skb);
+
+	if (len == skb->len) {
+		lp->stats.tx_packets++;
+		lp->stats.tx_bytes += skb->len;
+		dev->trans_start = jiffies;
+		netif_start_queue(dev);
+
+		/* this is normally done in the interrupt when tx finishes */
+		netif_wake_queue(dev);
+
+	} else if (len == 0) {
+		netif_start_queue(dev);
+		lp->stats.tx_dropped++;
+
+	} else {
+		netif_start_queue(dev);
+		printk(KERN_ERR "iss_net_start_xmit: failed(%d)\n", len);
+	}
+
+	spin_unlock_irqrestore(&lp->lock, flags);
+
+	dev_kfree_skb(skb);
+	return 0;
+}
+
+
+static struct net_device_stats *iss_net_get_stats(struct net_device *dev)
+{
+	struct iss_net_private *lp = dev->priv;
+	return &lp->stats;
+}
+
+static void iss_net_set_multicast_list(struct net_device *dev)
+{
+#if 0
+	if (dev->flags & IFF_PROMISC)
+		return;
+	else if (dev->mc_count)
+		dev->flags |= IFF_ALLMULTI;
+	else
+		dev->flags &= ~IFF_ALLMULTI;
+#endif
+}
+
+static void iss_net_tx_timeout(struct net_device *dev)
+{
+#if 0
+	dev->trans_start = jiffies;
+	netif_wake_queue(dev);
+#endif
+}
+
+static int iss_net_set_mac(struct net_device *dev, void *addr)
+{
+#if 0
+	struct iss_net_private *lp = dev->priv;
+	struct sockaddr *hwaddr = addr;
+
+	spin_lock(&lp->lock);
+	memcpy(dev->dev_addr, hwaddr->sa_data, ETH_ALEN);
+	spin_unlock(&lp->lock);
+#endif
+
+	return 0;
+}
+
+static int iss_net_change_mtu(struct net_device *dev, int new_mtu)
+{
+#if 0
+	struct iss_net_private *lp = dev->priv;
+	int err = 0;
+
+	spin_lock(&lp->lock);
+
+	// FIXME not needed new_mtu = transport_set_mtu(new_mtu, &lp->user);
+
+	if (new_mtu < 0)
+		err = new_mtu;
+	else
+		dev->mtu = new_mtu;
+
+	spin_unlock(&lp->lock);
+	return err;
+#endif
+	return -EINVAL;
+}
+
+static int iss_net_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+{
+#if 0
+	static const struct ethtool_drvinfo info = {
+		.cmd     = ETHTOOL_GDRVINFO,
+		.driver  = DRIVER_NAME,
+		.version = "42",
+	};
+	void *useraddr;
+	u32 ethcmd;
+
+	switch (cmd) {
+	case SIOCETHTOOL:
+		useraddr = ifr->ifr_data;
+		if (copy_from_user(&ethcmd, useraddr, sizeof(ethcmd)))
+			return -EFAULT;
+
+		switch (ethcmd) {
+			case ETHTOOL_GDRVINFO:
+				if (copy_to_user(useraddr, &info, sizeof(info)))
+					return -EFAULT;
+				return 0;
+			default:
+				return -EOPNOTSUPP;
+		}
+	default:
+		return -EINVAL;
+	}
+#endif
+	return -EINVAL;
+}
+
+void iss_net_user_timer_expire(unsigned long _conn)
+{
+}
+
+
+static struct device_driver iss_net_driver = {
+	.name  = DRIVER_NAME,
+	.bus   = &platform_bus_type,
+};
+
+static int driver_registered;
+
+static int iss_net_configure(int index, char *init)
+{
+	struct net_device *dev;
+	struct iss_net_private *lp;
+	int err;
+
+	if ((dev = alloc_etherdev(sizeof *lp)) == NULL) {
+		printk(KERN_ERR "eth_configure: failed to allocate device\n");
+		return 1;
+	}
+
+	/* Initialize private element. */
+
+	lp = dev->priv;
+	*lp = ((struct iss_net_private) {
+		.device_list		= LIST_HEAD_INIT(lp->device_list),
+		.opened_list		= LIST_HEAD_INIT(lp->opened_list),
+		.lock			= SPIN_LOCK_UNLOCKED,
+		.dev			= dev,
+		.index			= index,
+		//.fd                   = -1,
+		.mac			= { 0xfe, 0xfd, 0x0, 0x0, 0x0, 0x0 },
+		.have_mac		= 0,
+		});
+
+	/*
+	 * Try all transport protocols.
+	 * Note: more protocols can be added by adding '&& !X_init(lp, eth)'.
+	 */
+
+	if (!tuntap_probe(lp, index, init)) {
+		printk("Invalid arguments. Skipping device!\n");
+		goto errout;
+	}
+
+	printk(KERN_INFO "Netdevice %d ", index);
+	if (lp->have_mac)
+		printk("(%02x:%02x:%02x:%02x:%02x:%02x) ",
+				lp->mac[0], lp->mac[1],
+				lp->mac[2], lp->mac[3],
+				lp->mac[4], lp->mac[5]);
+	printk(": ");
+
+	/* sysfs register */
+
+	if (!driver_registered) {
+		driver_register(&iss_net_driver);
+		driver_registered = 1;
+	}
+
+	spin_lock(&devices_lock);
+	list_add(&lp->device_list, &devices);
+	spin_unlock(&devices_lock);
+
+	lp->pdev.id = index;
+	lp->pdev.name = DRIVER_NAME;
+	platform_device_register(&lp->pdev);
+	SET_NETDEV_DEV(dev,&lp->pdev.dev);
+
+	/*
+	 * If this name ends up conflicting with an existing registered
+	 * netdevice, that is OK, register_netdev{,ice}() will notice this
+	 * and fail.
+	 */
+	snprintf(dev->name, sizeof dev->name, "eth%d", index);
+
+	dev->mtu = lp->mtu;
+	dev->open = iss_net_open;
+	dev->hard_start_xmit = iss_net_start_xmit;
+	dev->stop = iss_net_close;
+	dev->get_stats = iss_net_get_stats;
+	dev->set_multicast_list = iss_net_set_multicast_list;
+	dev->tx_timeout = iss_net_tx_timeout;
+	dev->set_mac_address = iss_net_set_mac;
+	dev->change_mtu = iss_net_change_mtu;
+	dev->do_ioctl = iss_net_ioctl;
+	dev->watchdog_timeo = (HZ >> 1);
+	dev->irq = -1;
+
+	rtnl_lock();
+	err = register_netdevice(dev);
+	rtnl_unlock();
+
+	if (err) {
+		printk("Error registering net device!\n");
+		/* XXX: should we call ->remove() here? */
+		free_netdev(dev);
+		return 1;
+	}
+
+	init_timer(&lp->tl);
+	lp->tl.function = iss_net_user_timer_expire;
+
+#if 0
+	if (lp->have_mac)
+		set_ether_mac(dev, lp->mac);
+#endif
+	return 0;
+
+errout:
+	// FIXME: unregister; free, etc..
+	return -EIO;
+
+}
+
+/* ------------------------------------------------------------------------- */
+
+/* Filled in during early boot */
+
+struct list_head eth_cmd_line = LIST_HEAD_INIT(eth_cmd_line);
+
+struct iss_net_init {
+	struct list_head list;
+	char *init;		/* init string */
+	int index;
+};
+
+/*
+ * Parse the command line and look for 'ethX=...' fields, and register all
+ * those fields. They will be later initialized in iss_net_init.
+ */
+
+#define ERR KERN_ERR "iss_net_setup: "
+
+static int iss_net_setup(char *str)
+{
+	struct iss_net_private *device = NULL;
+	struct iss_net_init *new;
+	struct list_head *ele;
+	char *end;
+	int n;
+
+	n = simple_strtoul(str, &end, 0);
+	if (end == str) {
+		printk(ERR "Failed to parse '%s'\n", str);
+		return 1;
+	}
+	if (n < 0) {
+		printk(ERR "Device %d is negative\n", n);
+		return 1;
+	}
+	if (*(str = end) != '=') {
+		printk(ERR "Expected '=' after device number\n");
+		return 1;
+	}
+
+	spin_lock(&devices_lock);
+
+	list_for_each(ele, &devices) {
+		device = list_entry(ele, struct iss_net_private, device_list);
+		if (device->index == n)
+			break;
+	}
+
+	spin_unlock(&devices_lock);
+
+	if (device && device->index == n) {
+		printk(ERR "Device %d already configured\n", n);
+		return 1;
+	}
+
+	if ((new = alloc_bootmem(sizeof new)) == NULL) {
+		printk("Alloc_bootmem failed\n");
+		return 1;
+	}
+
+	INIT_LIST_HEAD(&new->list);
+	new->index = n;
+	new->init = str + 1;
+
+	list_add_tail(&new->list, &eth_cmd_line);
+	return 1;
+}
+
+#undef ERR
+
+__setup("eth", iss_net_setup);
+
+/*
+ * Initialize all ISS Ethernet devices previously registered in iss_net_setup.
+ */
+
+static int iss_net_init(void)
+{
+	struct list_head *ele, *next;
+
+	/* Walk through all Ethernet devices specified in the command line. */
+
+	list_for_each_safe(ele, next, &eth_cmd_line) {
+		struct iss_net_init *eth;
+		eth = list_entry(ele, struct iss_net_init, list);
+		iss_net_configure(eth->index, eth->init);
+	}
+
+	return 1;
+}
+
+module_init(iss_net_init);
+
diff --git a/arch/xtensa/platform-iss/setup.c b/arch/xtensa/platform-iss/setup.c
new file mode 100644
index 00000000000..2e6dcbf0cc0
--- /dev/null
+++ b/arch/xtensa/platform-iss/setup.c
@@ -0,0 +1,112 @@
+/*
+ *
+ * arch/xtensa/platform-iss/setup.c
+ *
+ * Platform specific initialization.
+ *
+ * Authors: Chris Zankel <chris@zankel.net>
+ *          Joe Taylor <joe@tensilica.com>
+ *
+ * Copyright 2001 - 2005 Tensilica Inc.
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ */
+#include <linux/config.h>
+#include <linux/stddef.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/reboot.h>
+#include <linux/pci.h>
+#include <linux/kdev_t.h>
+#include <linux/types.h>
+#include <linux/major.h>
+#include <linux/blkdev.h>
+#include <linux/console.h>
+#include <linux/delay.h>
+#include <linux/stringify.h>
+#include <linux/notifier.h>
+
+#include <asm/platform.h>
+#include <asm/bootparam.h>
+
+
+void __init platform_init(bp_tag_t* bootparam)
+{
+
+}
+
+void platform_halt(void)
+{
+	printk (" ** Called platform_halt(), looping forever! **\n");
+	while (1);
+}
+
+void platform_power_off(void)
+{
+	printk (" ** Called platform_power_off(), looping forever! **\n");
+	while (1);
+}
+void platform_restart(void)
+{
+	/* Flush and reset the mmu, simulate a processor reset, and
+	 * jump to the reset vector. */
+
+	__asm__ __volatile__("movi	a2, 15\n\t"
+			     "wsr	a2, " __stringify(ICOUNTLEVEL) "\n\t"
+			     "movi	a2, 0\n\t"
+			     "wsr	a2, " __stringify(ICOUNT) "\n\t"
+			     "wsr	a2, " __stringify(IBREAKENABLE) "\n\t"
+			     "wsr	a2, " __stringify(LCOUNT) "\n\t"
+			     "movi	a2, 0x1f\n\t"
+			     "wsr	a2, " __stringify(PS) "\n\t"
+			     "isync\n\t"
+			     "jx	%0\n\t"
+			     :
+			     : "a" (XCHAL_RESET_VECTOR_VADDR)
+			     : "a2");
+
+	/* control never gets here */
+}
+
+extern void iss_net_poll(void);
+
+const char twirl[]="|/-\\|/-\\";
+
+void platform_heartbeat(void)
+{
+#if 0
+	static int i = 0, j = 0;
+
+	if (--i < 0) {
+		i = 99;
+		printk("\r%c\r", twirl[j++]);
+		if (j == 8)
+			j = 0;
+	}
+#endif
+}
+
+
+
+static int
+iss_panic_event(struct notifier_block *this, unsigned long event, void *ptr)
+{
+	__asm__ __volatile__("movi a2, -1; simcall\n");
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block iss_panic_block = {
+	iss_panic_event,
+	NULL,
+	0
+};
+
+void __init platform_setup(char **p_cmdline)
+{
+	notifier_chain_register(&panic_notifier_list, &iss_panic_block);
+}
diff --git a/include/asm-xtensa/platform-iss/hardware.h b/include/asm-xtensa/platform-iss/hardware.h
new file mode 100644
index 00000000000..22240f00180
--- /dev/null
+++ b/include/asm-xtensa/platform-iss/hardware.h
@@ -0,0 +1,29 @@
+/*
+ * include/asm-xtensa/platform-iss/hardware.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 Tensilica Inc.
+ */
+
+/*
+ * This file contains the default configuration of ISS.
+ */
+
+#ifndef __ASM_XTENSA_ISS_HARDWARE
+#define __ASM_XTENSA_ISS_HARDWARE
+
+/*
+ * Memory configuration.
+ */
+
+#define PLATFORM_DEFAULT_MEM_START XSHAL_RAM_PADDR
+#define PLATFORM_DEFAULT_MEM_SIZE XSHAL_RAM_VSIZE
+
+/*
+ * Interrupt configuration.
+ */
+
+#endif /* __ASM_XTENSA_ISS_HARDWARE */
diff --git a/include/asm-xtensa/platform.h b/include/asm-xtensa/platform.h
new file mode 100644
index 00000000000..36163894bc2
--- /dev/null
+++ b/include/asm-xtensa/platform.h
@@ -0,0 +1,92 @@
+/*
+ * include/asm-xtensa/platform.h
+ *
+ * Platform specific functions
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_PLATFORM_H
+#define _XTENSA_PLATFORM_H
+
+#include <linux/config.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+
+#include <asm/bootparam.h>
+
+/*
+ * platform_init is called before the mmu is initialized to give the
+ * platform a early hook-up. bp_tag_t is a list of configuration tags
+ * passed from the boot-loader.
+ */
+extern void platform_init(bp_tag_t*);
+
+/*
+ * platform_setup is called from setup_arch with a pointer to the command-line
+ * string.
+ */
+extern void platform_setup (char **);
+
+/*
+ * platform_init_irq is called from init_IRQ.
+ */
+extern void platform_init_irq (void);
+
+/*
+ * platform_restart is called to restart the system.
+ */
+extern void platform_restart (void);
+
+/*
+ * platform_halt is called to stop the system and halt.
+ */
+extern void platform_halt (void);
+
+/*
+ * platform_power_off is called to stop the system and power it off.
+ */
+extern void platform_power_off (void);
+
+/*
+ * platform_idle is called from the idle function.
+ */
+extern void platform_idle (void);
+
+/*
+ * platform_heartbeat is called every HZ
+ */
+extern void platform_heartbeat (void);
+
+/*
+ * platform_pcibios_init is called to allow the platform to setup the pci bus.
+ */
+extern void platform_pcibios_init (void);
+
+/*
+ * platform_pcibios_fixup allows to modify the PCI configuration.
+ */
+extern int platform_pcibios_fixup (void);
+
+/*
+ * platform_calibrate_ccount calibrates cpu clock freq (CONFIG_XTENSA_CALIBRATE)
+ */
+extern void platform_calibrate_ccount (void);
+
+/*
+ * platform_get_rtc_time returns RTC seconds (returns 0 for no error)
+ */
+extern int platform_get_rtc_time(time_t*);
+
+/*
+ * platform_set_rtc_time set RTC seconds (returns 0 for no error)
+ */
+extern int platform_set_rtc_time(time_t);
+
+
+#endif	/* _XTENSA_PLATFORM_H */
+
-- 
cgit v1.2.3-70-g09d2


From 3b6259432dee81f928c22c48c080d5f6325ed92e Mon Sep 17 00:00:00 2001
From: Corey Minyard <minyard@acm.org>
Date: Thu, 23 Jun 2005 22:01:42 -0700
Subject: [PATCH] ipmi: add power cycle capability

This patch to adds "power cycle" functionality to the IPMI power off module
ipmi_poweroff.  It also contains changes to support procfs control of the
feature.

The power cycle action is considered an optional chassis control in the IPMI
specification.  However, it is definitely useful when the hardware supports
it.  A power cycle is usually required in order to reset a firmware in a bad
state.  This action is critical to allow remote management of servers.

The implementation adds power cycle as optional to the ipmi_poweroff module.
It can be modified dynamically through the proc entry mentioned above.  During
a power down and enabled, the power cycle command is sent to the BMC firmware.
 If it fails either due to non-support or some error, it will retry to send
the command as power off.

Signed-off-by: Christopher A. Poblete <Chris_Poblete@dell.com>
Signed-off-by: Corey Minyard <minyard@acm.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/IPMI.txt              |  20 +++++++
 drivers/char/ipmi/ipmi_msghandler.c |  29 +++++++++-
 drivers/char/ipmi/ipmi_poweroff.c   | 112 +++++++++++++++++++++++++++++++++---
 include/linux/ipmi.h                |   5 ++
 4 files changed, 155 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/Documentation/IPMI.txt b/Documentation/IPMI.txt
index 2f99fe6299a..84d3d4d10c1 100644
--- a/Documentation/IPMI.txt
+++ b/Documentation/IPMI.txt
@@ -594,3 +594,23 @@ an event generator, the event receiver from the local management
 controller will be queried and the events sent to the SEL on that
 device.  Otherwise, the events go nowhere since there is nowhere to
 send them.
+
+
+Poweroff
+--------
+
+If the poweroff capability is selected, the IPMI driver will install
+a shutdown function into the standard poweroff function pointer.  This
+is in the ipmi_poweroff module.  When the system requests a powerdown,
+it will send the proper IPMI commands to do this.  This is supported on
+several platforms.
+
+There is a module parameter named "poweroff_control" that may either be zero
+(do a power down) or 2 (do a power cycle, power the system off, then power
+it on in a few seconds).  Setting ipmi_poweroff.poweroff_control=x will do
+the same thing on the kernel command line.  The parameter is also available
+via the proc filesystem in /proc/ipmi/poweroff_control.  Note that if the
+system does not support power cycling, it will always to the power off.
+
+Note that if you have ACPI enabled, the system will prefer using ACPI to
+power off.
diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c
index ed75e96d003..1813d0d198f 100644
--- a/drivers/char/ipmi/ipmi_msghandler.c
+++ b/drivers/char/ipmi/ipmi_msghandler.c
@@ -54,7 +54,9 @@ static int ipmi_init_msghandler(void);
 
 static int initialized = 0;
 
-static struct proc_dir_entry *proc_ipmi_root = NULL;
+#ifdef CONFIG_PROC_FS
+struct proc_dir_entry *proc_ipmi_root = NULL;
+#endif /* CONFIG_PROC_FS */
 
 #define MAX_EVENTS_IN_QUEUE	25
 
@@ -124,11 +126,13 @@ struct ipmi_channel
 	unsigned char protocol;
 };
 
+#ifdef CONFIG_PROC_FS
 struct ipmi_proc_entry
 {
 	char                   *name;
 	struct ipmi_proc_entry *next;
 };
+#endif
 
 #define IPMI_IPMB_NUM_SEQ	64
 #define IPMI_MAX_CHANNELS       8
@@ -156,10 +160,13 @@ struct ipmi_smi
 	struct ipmi_smi_handlers *handlers;
 	void                     *send_info;
 
+#ifdef CONFIG_PROC_FS
 	/* A list of proc entries for this interface.  This does not
 	   need a lock, only one thread creates it and only one thread
 	   destroys it. */
+	spinlock_t             proc_entry_lock;
 	struct ipmi_proc_entry *proc_entries;
+#endif
 
 	/* A table of sequence numbers for this interface.  We use the
            sequence numbers for IPMB messages that go out of the
@@ -1470,8 +1477,9 @@ int ipmi_smi_add_proc_entry(ipmi_smi_t smi, char *name,
 			    read_proc_t *read_proc, write_proc_t *write_proc,
 			    void *data, struct module *owner)
 {
-	struct proc_dir_entry  *file;
 	int                    rv = 0;
+#ifdef CONFIG_PROC_FS
+	struct proc_dir_entry  *file;
 	struct ipmi_proc_entry *entry;
 
 	/* Create a list element. */
@@ -1497,10 +1505,13 @@ int ipmi_smi_add_proc_entry(ipmi_smi_t smi, char *name,
 		file->write_proc = write_proc;
 		file->owner = owner;
 
+		spin_lock(&smi->proc_entry_lock);
 		/* Stick it on the list. */
 		entry->next = smi->proc_entries;
 		smi->proc_entries = entry;
+		spin_unlock(&smi->proc_entry_lock);
 	}
+#endif /* CONFIG_PROC_FS */
 
 	return rv;
 }
@@ -1509,6 +1520,7 @@ static int add_proc_entries(ipmi_smi_t smi, int num)
 {
 	int rv = 0;
 
+#ifdef CONFIG_PROC_FS
 	sprintf(smi->proc_dir_name, "%d", num);
 	smi->proc_dir = proc_mkdir(smi->proc_dir_name, proc_ipmi_root);
 	if (!smi->proc_dir)
@@ -1531,14 +1543,17 @@ static int add_proc_entries(ipmi_smi_t smi, int num)
 		rv = ipmi_smi_add_proc_entry(smi, "version",
 					     version_file_read_proc, NULL,
 					     smi, THIS_MODULE);
+#endif /* CONFIG_PROC_FS */
 
 	return rv;
 }
 
 static void remove_proc_entries(ipmi_smi_t smi)
 {
+#ifdef CONFIG_PROC_FS
 	struct ipmi_proc_entry *entry;
 
+	spin_lock(&smi->proc_entry_lock);
 	while (smi->proc_entries) {
 		entry = smi->proc_entries;
 		smi->proc_entries = entry->next;
@@ -1547,7 +1562,9 @@ static void remove_proc_entries(ipmi_smi_t smi)
 		kfree(entry->name);
 		kfree(entry);
 	}
+	spin_unlock(&smi->proc_entry_lock);
 	remove_proc_entry(smi->proc_dir_name, proc_ipmi_root);
+#endif /* CONFIG_PROC_FS */
 }
 
 static int
@@ -1694,6 +1711,9 @@ int ipmi_register_smi(struct ipmi_smi_handlers *handlers,
 				new_intf->seq_table[j].seqid = 0;
 			}
 			new_intf->curr_seq = 0;
+#ifdef CONFIG_PROC_FS
+			spin_lock_init(&(new_intf->proc_entry_lock));
+#endif
 			spin_lock_init(&(new_intf->waiting_msgs_lock));
 			INIT_LIST_HEAD(&(new_intf->waiting_msgs));
 			spin_lock_init(&(new_intf->events_lock));
@@ -3085,6 +3105,7 @@ static int ipmi_init_msghandler(void)
 		ipmi_interfaces[i] = NULL;
 	}
 
+#ifdef CONFIG_PROC_FS
 	proc_ipmi_root = proc_mkdir("ipmi", NULL);
 	if (!proc_ipmi_root) {
 	    printk(KERN_ERR PFX "Unable to create IPMI proc dir");
@@ -3092,6 +3113,7 @@ static int ipmi_init_msghandler(void)
 	}
 
 	proc_ipmi_root->owner = THIS_MODULE;
+#endif /* CONFIG_PROC_FS */
 
 	init_timer(&ipmi_timer);
 	ipmi_timer.data = 0;
@@ -3129,7 +3151,9 @@ static __exit void cleanup_ipmi(void)
 	atomic_inc(&stop_operation);
 	del_timer_sync(&ipmi_timer);
 
+#ifdef CONFIG_PROC_FS
 	remove_proc_entry(proc_ipmi_root->name, &proc_root);
+#endif /* CONFIG_PROC_FS */
 
 	initialized = 0;
 
@@ -3170,4 +3194,5 @@ EXPORT_SYMBOL(ipmi_get_my_address);
 EXPORT_SYMBOL(ipmi_set_my_LUN);
 EXPORT_SYMBOL(ipmi_get_my_LUN);
 EXPORT_SYMBOL(ipmi_smi_add_proc_entry);
+EXPORT_SYMBOL(proc_ipmi_root);
 EXPORT_SYMBOL(ipmi_user_set_run_to_completion);
diff --git a/drivers/char/ipmi/ipmi_poweroff.c b/drivers/char/ipmi/ipmi_poweroff.c
index cb5cdc6f14b..61329b55c4a 100644
--- a/drivers/char/ipmi/ipmi_poweroff.c
+++ b/drivers/char/ipmi/ipmi_poweroff.c
@@ -34,6 +34,8 @@
 #include <asm/semaphore.h>
 #include <linux/kdev_t.h>
 #include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/proc_fs.h>
 #include <linux/string.h>
 #include <linux/ipmi.h>
 #include <linux/ipmi_smi.h>
@@ -44,6 +46,18 @@
 /* Where to we insert our poweroff function? */
 extern void (*pm_power_off)(void);
 
+/* Definitions for controlling power off (if the system supports it).  It
+ * conveniently matches the IPMI chassis control values. */
+#define IPMI_CHASSIS_POWER_DOWN		0	/* power down, the default. */
+#define IPMI_CHASSIS_POWER_CYCLE	0x02	/* power cycle */
+
+/* the IPMI data command */
+static int poweroff_control = IPMI_CHASSIS_POWER_DOWN;
+
+/* parameter definition to allow user to flag power cycle */
+module_param(poweroff_control, int, IPMI_CHASSIS_POWER_DOWN);
+MODULE_PARM_DESC(poweroff_control, " Set to 2 to enable power cycle instead of power down. Power cycle is contingent on hardware support, otherwise it defaults back to power down.");
+
 /* Stuff from the get device id command. */
 static unsigned int mfg_id;
 static unsigned int prod_id;
@@ -349,26 +363,38 @@ static void ipmi_poweroff_chassis (ipmi_user_t user)
         smi_addr.channel = IPMI_BMC_CHANNEL;
         smi_addr.lun = 0;
 
-	printk(KERN_INFO PFX "Powering down via IPMI chassis control command\n");
+ powercyclefailed:
+	printk(KERN_INFO PFX "Powering %s via IPMI chassis control command\n",
+		((poweroff_control != IPMI_CHASSIS_POWER_CYCLE) ? "down" : "cycle"));
 
 	/*
 	 * Power down
 	 */
 	send_msg.netfn = IPMI_NETFN_CHASSIS_REQUEST;
 	send_msg.cmd = IPMI_CHASSIS_CONTROL_CMD;
-	data[0] = 0; /* Power down */
+	data[0] = poweroff_control;
 	send_msg.data = data;
 	send_msg.data_len = sizeof(data);
 	rv = ipmi_request_in_rc_mode(user,
 				     (struct ipmi_addr *) &smi_addr,
 				     &send_msg);
 	if (rv) {
-		printk(KERN_ERR PFX "Unable to send chassis powerdown message,"
-		       " IPMI error 0x%x\n", rv);
-		goto out;
+		switch (poweroff_control) {
+			case IPMI_CHASSIS_POWER_CYCLE:
+				/* power cycle failed, default to power down */
+				printk(KERN_ERR PFX "Unable to send chassis power " \
+					"cycle message, IPMI error 0x%x\n", rv);
+				poweroff_control = IPMI_CHASSIS_POWER_DOWN;
+				goto powercyclefailed;
+
+			case IPMI_CHASSIS_POWER_DOWN:
+			default:
+				printk(KERN_ERR PFX "Unable to send chassis power " \
+					"down message, IPMI error 0x%x\n", rv);
+				break;
+		}
 	}
 
- out:
 	return;
 }
 
@@ -430,7 +456,8 @@ static void ipmi_po_new_smi(int if_num)
 	if (ready)
 		return;
 
-	rv = ipmi_create_user(if_num, &ipmi_poweroff_handler, NULL, &ipmi_user);
+	rv = ipmi_create_user(if_num, &ipmi_poweroff_handler, NULL,
+			      &ipmi_user);
 	if (rv) {
 		printk(KERN_ERR PFX "could not create IPMI user, error %d\n",
 		       rv);
@@ -509,21 +536,84 @@ static struct ipmi_smi_watcher smi_watcher =
 };
 
 
+#ifdef CONFIG_PROC_FS
+/* displays properties to proc */
+static int proc_read_chassctrl(char *page, char **start, off_t off, int count,
+			       int *eof, void *data)
+{
+	return sprintf(page, "%d\t[ 0=powerdown 2=powercycle ]\n",
+			poweroff_control);
+}
+
+/* process property writes from proc */
+static int proc_write_chassctrl(struct file *file, const char *buffer,
+			        unsigned long count, void *data)
+{
+	int          rv = count;
+	unsigned int newval = 0;
+
+	sscanf(buffer, "%d", &newval);
+	switch (newval) {
+		case IPMI_CHASSIS_POWER_CYCLE:
+			printk(KERN_INFO PFX "power cycle is now enabled\n");
+			poweroff_control = newval;
+			break;
+
+		case IPMI_CHASSIS_POWER_DOWN:
+			poweroff_control = IPMI_CHASSIS_POWER_DOWN;
+			break;
+
+		default:
+			rv = -EINVAL;
+			break;
+	}
+
+	return rv;
+}
+#endif /* CONFIG_PROC_FS */
+
 /*
  * Startup and shutdown functions.
  */
 static int ipmi_poweroff_init (void)
 {
-	int rv;
+	int                   rv;
+	struct proc_dir_entry *file;
 
 	printk ("Copyright (C) 2004 MontaVista Software -"
 		" IPMI Powerdown via sys_reboot version "
 		IPMI_POWEROFF_VERSION ".\n");
 
+	switch (poweroff_control) {
+		case IPMI_CHASSIS_POWER_CYCLE:
+			printk(KERN_INFO PFX "Power cycle is enabled.\n");
+			break;
+
+		case IPMI_CHASSIS_POWER_DOWN:
+		default:
+			poweroff_control = IPMI_CHASSIS_POWER_DOWN;
+			break;
+	}
+
 	rv = ipmi_smi_watcher_register(&smi_watcher);
-	if (rv)
+	if (rv) {
 		printk(KERN_ERR PFX "Unable to register SMI watcher: %d\n", rv);
+		goto out_err;
+	}
+
+#ifdef CONFIG_PROC_FS
+	file = create_proc_entry("poweroff_control", 0, proc_ipmi_root);
+	if (!file) {
+		printk(KERN_ERR PFX "Unable to create proc power control\n");
+	} else {
+		file->nlink = 1;
+		file->read_proc = proc_read_chassctrl;
+		file->write_proc = proc_write_chassctrl;
+		file->owner = THIS_MODULE;
+	}
+#endif
 
+ out_err:
 	return rv;
 }
 
@@ -532,6 +622,10 @@ static __exit void ipmi_poweroff_cleanup(void)
 {
 	int rv;
 
+#ifdef CONFIG_PROC_FS
+	remove_proc_entry("poweroff_control", proc_ipmi_root);
+#endif
+
 	ipmi_smi_watcher_unregister(&smi_watcher);
 
 	if (ready) {
diff --git a/include/linux/ipmi.h b/include/linux/ipmi.h
index 2ec265e1045..596ca613015 100644
--- a/include/linux/ipmi.h
+++ b/include/linux/ipmi.h
@@ -209,6 +209,11 @@ struct kernel_ipmi_msg
 #include <linux/list.h>
 #include <linux/module.h>
 
+#ifdef CONFIG_PROC_FS
+#include <linux/proc_fs.h>
+extern struct proc_dir_entry *proc_ipmi_root;
+#endif /* CONFIG_PROC_FS */
+
 /* Opaque type for a IPMI message user.  One of these is needed to
    send and receive messages. */
 typedef struct ipmi_user *ipmi_user_t;
-- 
cgit v1.2.3-70-g09d2


From a6df7da8f7ee99e6fd1995fad852bacb978a6447 Mon Sep 17 00:00:00 2001
From: Kylene Hall <kjhall@us.ibm.com>
Date: Thu, 23 Jun 2005 22:02:04 -0700
Subject: [PATCH] tpm: TPMs on additional LPC bus

Add support for TPMs on additional LPC buses.

Signed-off-by: Kylene Hall <kjhall@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/tpm/tpm_atmel.c | 1 +
 include/linux/pci_ids.h      | 1 +
 2 files changed, 2 insertions(+)

(limited to 'include')

diff --git a/drivers/char/tpm/tpm_atmel.c b/drivers/char/tpm/tpm_atmel.c
index 68974577a6a..13248400b9c 100644
--- a/drivers/char/tpm/tpm_atmel.c
+++ b/drivers/char/tpm/tpm_atmel.c
@@ -205,6 +205,7 @@ static struct pci_device_id tpm_pci_tbl[] __devinitdata = {
 	{PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801DB_12)},
 	{PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801EB_0)},
 	{PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_8111_LPC)},
+	{PCI_DEVICE(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_CSB6LPC)},
 	{0,}
 };
 
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 63e89e47b8e..bf608808a60 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1568,6 +1568,7 @@
 #define PCI_DEVICE_ID_SERVERWORKS_OSB4USB 0x0220
 #define PCI_DEVICE_ID_SERVERWORKS_CSB5USB PCI_DEVICE_ID_SERVERWORKS_OSB4USB
 #define PCI_DEVICE_ID_SERVERWORKS_CSB6USB 0x0221
+#define PCI_DEVICE_ID_SERVERWORKS_CSB6LPC 0x0227
 #define PCI_DEVICE_ID_SERVERWORKS_GCLE    0x0225
 #define PCI_DEVICE_ID_SERVERWORKS_GCLE2   0x0227
 #define PCI_DEVICE_ID_SERVERWORKS_CSB5ISA 0x0230
-- 
cgit v1.2.3-70-g09d2


From 61fbfa8129c1771061a0e9f47747854293081c5b Mon Sep 17 00:00:00 2001
From: Markus Lidel <Markus.Lidel@shadowconnect.com>
Date: Thu, 23 Jun 2005 22:02:11 -0700
Subject: [PATCH] I2O: bugfixes and compability enhancements

Changes:

 - Fixed sysfs bug where user and parent links where added to the I2O
   device itself
 - Fixed bug when calculating TID for the event handler and cleaned up the
   workflow of i2o_driver_dispatch()
 - Fixed oops when no I2O device could be found for an event delivered to
   Exec-OSM
 - Fixed initialization of spinlock in Exec-OSM
 - Fixed memory leak in i2o_cfg_passthru() and i2o_cfg_passthru()
 - Removed MTRR support
 - Added PCI ID of Promise SX6000 with firmware >= 1.20.x.x
 - Turn of caching for ioremapped memory of in_queue
 - Added initialization sequence for Promise controllers
 - Moved definition of u8 / u16 / u32 for raidutils before first use

Signed-off-by: Markus Lidel <Markus.Lidel@shadowconnect.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/message/i2o/device.c     | 10 +++--
 drivers/message/i2o/driver.c     | 89 +++++++++++++++++++-------------------
 drivers/message/i2o/exec-osm.c   |  9 ++--
 drivers/message/i2o/i2o_config.c | 48 +++++++++++++--------
 drivers/message/i2o/i2o_scsi.c   |  3 +-
 drivers/message/i2o/pci.c        | 93 ++++++++++++++--------------------------
 include/linux/i2o-dev.h          | 16 +++----
 include/linux/i2o.h              |  5 ---
 8 files changed, 124 insertions(+), 149 deletions(-)

(limited to 'include')

diff --git a/drivers/message/i2o/device.c b/drivers/message/i2o/device.c
index eb907e87bc7..280627ae6cf 100644
--- a/drivers/message/i2o/device.c
+++ b/drivers/message/i2o/device.c
@@ -401,25 +401,27 @@ static int i2o_device_class_add(struct class_device *cd)
 
 	/* create user entries for this device */
 	tmp = i2o_iop_find_device(i2o_dev->iop, i2o_dev->lct_data.user_tid);
-	if (tmp)
+	if (tmp && (tmp != i2o_dev))
 		sysfs_create_link(&i2o_dev->device.kobj, &tmp->device.kobj,
 				  "user");
 
 	/* create user entries refering to this device */
 	list_for_each_entry(tmp, &c->devices, list)
-	    if (tmp->lct_data.user_tid == i2o_dev->lct_data.tid)
+	    if ((tmp->lct_data.user_tid == i2o_dev->lct_data.tid)
+		&& (tmp != i2o_dev))
 		sysfs_create_link(&tmp->device.kobj,
 				  &i2o_dev->device.kobj, "user");
 
 	/* create parent entries for this device */
 	tmp = i2o_iop_find_device(i2o_dev->iop, i2o_dev->lct_data.parent_tid);
-	if (tmp)
+	if (tmp && (tmp != i2o_dev))
 		sysfs_create_link(&i2o_dev->device.kobj, &tmp->device.kobj,
 				  "parent");
 
 	/* create parent entries refering to this device */
 	list_for_each_entry(tmp, &c->devices, list)
-	    if (tmp->lct_data.parent_tid == i2o_dev->lct_data.tid)
+	    if ((tmp->lct_data.parent_tid == i2o_dev->lct_data.tid)
+		&& (tmp != i2o_dev))
 		sysfs_create_link(&tmp->device.kobj,
 				  &i2o_dev->device.kobj, "parent");
 
diff --git a/drivers/message/i2o/driver.c b/drivers/message/i2o/driver.c
index 91f4edbb2a2..c71e68f70e7 100644
--- a/drivers/message/i2o/driver.c
+++ b/drivers/message/i2o/driver.c
@@ -18,6 +18,8 @@
 #include <linux/rwsem.h>
 #include <linux/i2o.h>
 
+#define OSM_NAME	"core"
+
 /* max_drivers - Maximum I2O drivers (OSMs) which could be registered */
 unsigned int i2o_max_drivers = I2O_MAX_DRIVERS;
 module_param_named(max_drivers, i2o_max_drivers, uint, 0);
@@ -182,62 +184,59 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m,
 	struct i2o_driver *drv;
 	u32 context = readl(&msg->u.s.icntxt);
 
-	if (likely(context < i2o_max_drivers)) {
-		spin_lock(&i2o_drivers_lock);
-		drv = i2o_drivers[context];
-		spin_unlock(&i2o_drivers_lock);
-
-		if (unlikely(!drv)) {
-			printk(KERN_WARNING "%s: Spurious reply to unknown "
-			       "driver %d\n", c->name, context);
-			return -EIO;
-		}
+	if (unlikely(context >= i2o_max_drivers)) {
+		printk(KERN_WARNING "%s: Spurious reply to unknown driver "
+		       "%d\n", c->name, readl(&msg->u.s.icntxt));
+		return -EIO;
+	}
 
-		if ((readl(&msg->u.head[1]) >> 24) == I2O_CMD_UTIL_EVT_REGISTER) {
-			struct i2o_device *dev, *tmp;
-			struct i2o_event *evt;
-			u16 size;
-			u16 tid;
+	spin_lock(&i2o_drivers_lock);
+	drv = i2o_drivers[context];
+	spin_unlock(&i2o_drivers_lock);
 
-			tid = readl(&msg->u.head[1]) & 0x1fff;
+	if (unlikely(!drv)) {
+		osm_warn("Spurious reply to unknown driver %d\n", context);
+		return -EIO;
+	}
 
-			pr_debug("%s: event received from device %d\n", c->name,
-				 tid);
+	if ((readl(&msg->u.head[1]) >> 24) == I2O_CMD_UTIL_EVT_REGISTER) {
+		struct i2o_device *dev, *tmp;
+		struct i2o_event *evt;
+		u16 size;
+		u16 tid = readl(&msg->u.head[1]) & 0xfff;
 
-			/* cut of header from message size (in 32-bit words) */
-			size = (readl(&msg->u.head[0]) >> 16) - 5;
+		osm_debug("event received from device %d\n", tid);
 
-			evt = kmalloc(size * 4 + sizeof(*evt), GFP_ATOMIC);
-			if (!evt)
-				return -ENOMEM;
-			memset(evt, 0, size * 4 + sizeof(*evt));
+		/* cut of header from message size (in 32-bit words) */
+		size = (readl(&msg->u.head[0]) >> 16) - 5;
 
-			evt->size = size;
-			memcpy_fromio(&evt->tcntxt, &msg->u.s.tcntxt,
-				      (size + 2) * 4);
+		evt = kmalloc(size * 4 + sizeof(*evt), GFP_ATOMIC | __GFP_ZERO);
+		if (!evt)
+			return -ENOMEM;
 
-			list_for_each_entry_safe(dev, tmp, &c->devices, list)
-			    if (dev->lct_data.tid == tid) {
-				evt->i2o_dev = dev;
-				break;
-			}
+		evt->size = size;
+		evt->tcntxt = readl(&msg->u.s.tcntxt);
+		evt->event_indicator = readl(&msg->body[0]);
+		memcpy_fromio(&evt->tcntxt, &msg->u.s.tcntxt, size * 4);
 
-			INIT_WORK(&evt->work, (void (*)(void *))drv->event,
-				  evt);
-			queue_work(drv->event_queue, &evt->work);
-			return 1;
+		list_for_each_entry_safe(dev, tmp, &c->devices, list)
+		    if (dev->lct_data.tid == tid) {
+			evt->i2o_dev = dev;
+			break;
 		}
 
-		if (likely(drv->reply))
-			return drv->reply(c, m, msg);
-		else
-			pr_debug("%s: Reply to driver %s, but no reply function"
-				 " defined!\n", c->name, drv->name);
+		INIT_WORK(&evt->work, (void (*)(void *))drv->event, evt);
+		queue_work(drv->event_queue, &evt->work);
+		return 1;
+	}
+
+	if (unlikely(!drv->reply)) {
+		pr_debug("%s: Reply to driver %s, but no reply function"
+			 " defined!\n", c->name, drv->name);
 		return -EIO;
-	} else
-		printk(KERN_WARNING "%s: Spurious reply to unknown driver "
-		       "%d\n", c->name, readl(&msg->u.s.icntxt));
-	return -EIO;
+	}
+
+	return drv->reply(c, m, msg);
 }
 
 /**
diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c
index 79c1cbfb8f4..1e28e886f1c 100644
--- a/drivers/message/i2o/exec-osm.c
+++ b/drivers/message/i2o/exec-osm.c
@@ -204,12 +204,10 @@ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m,
 				      struct i2o_message __iomem *msg)
 {
 	struct i2o_exec_wait *wait, *tmp;
-	static spinlock_t lock;
+	static spinlock_t lock = SPIN_LOCK_UNLOCKED;
 	int rc = 1;
 	u32 context;
 
-	spin_lock_init(&lock);
-
 	context = readl(&msg->u.s.tcntxt);
 
 	/*
@@ -381,8 +379,9 @@ static int i2o_exec_reply(struct i2o_controller *c, u32 m,
  */
 static void i2o_exec_event(struct i2o_event *evt)
 {
-	osm_info("Event received from device: %d\n",
-		 evt->i2o_dev->lct_data.tid);
+	if(likely(evt->i2o_dev))
+		osm_info("Event received from device: %d\n",
+			 evt->i2o_dev->lct_data.tid);
 	kfree(evt);
 };
 
diff --git a/drivers/message/i2o/i2o_config.c b/drivers/message/i2o/i2o_config.c
index 1fb5cdf67f8..46d373287a3 100644
--- a/drivers/message/i2o/i2o_config.c
+++ b/drivers/message/i2o/i2o_config.c
@@ -555,6 +555,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar
 	u32 sg_offset = 0;
 	u32 sg_count = 0;
 	u32 i = 0;
+	u32 sg_index = 0;
 	i2o_status_block *sb;
 	struct i2o_message *msg;
 	u32 m;
@@ -634,8 +635,8 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar
 		if (sg_count > SG_TABLESIZE) {
 			printk(KERN_DEBUG "%s:IOCTL SG List too large (%u)\n",
 			       c->name, sg_count);
-			kfree(reply);
-			return -EINVAL;
+			rcode = -EINVAL;
+			goto cleanup;
 		}
 
 		for (i = 0; i < sg_count; i++) {
@@ -651,7 +652,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar
 				goto cleanup;
 			}
 			sg_size = sg[i].flag_count & 0xffffff;
-			p = &(sg_list[i]);
+			p = &(sg_list[sg_index++]);
 			/* Allocate memory for the transfer */
 			if (i2o_dma_alloc
 			    (&c->pdev->dev, p, sg_size,
@@ -660,7 +661,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar
 				       "%s: Could not allocate SG buffer - size = %d buffer number %d of %d\n",
 				       c->name, sg_size, i, sg_count);
 				rcode = -ENOMEM;
-				goto cleanup;
+				goto sg_list_cleanup;
 			}
 			/* Copy in the user's SG buffer if necessary */
 			if (sg[i].
@@ -673,7 +674,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar
 					       "%s: Could not copy SG buf %d FROM user\n",
 					       c->name, i);
 					rcode = -EFAULT;
-					goto cleanup;
+					goto sg_list_cleanup;
 				}
 			}
 			//TODO 64bit fix
@@ -683,10 +684,10 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar
 
 	rcode = i2o_msg_post_wait(c, m, 60);
 	if (rcode)
-		goto cleanup;
+		goto sg_list_cleanup;
 
 	if (sg_offset) {
-		u32 msg[128];
+		u32 msg[MSG_FRAME_SIZE];
 		/* Copy back the Scatter Gather buffers back to user space */
 		u32 j;
 		// TODO 64bit fix
@@ -698,14 +699,14 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar
 		// get user msg size in u32s
 		if (get_user(size, &user_msg[0])) {
 			rcode = -EFAULT;
-			goto cleanup;
+			goto sg_list_cleanup;
 		}
 		size = size >> 16;
 		size *= 4;
 		/* Copy in the user's I2O command */
 		if (copy_from_user(msg, user_msg, size)) {
 			rcode = -EFAULT;
-			goto cleanup;
+			goto sg_list_cleanup;
 		}
 		sg_count =
 		    (size - sg_offset * 4) / sizeof(struct sg_simple_element);
@@ -727,7 +728,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar
 					       c->name, sg_list[j].virt,
 					       sg[j].addr_bus);
 					rcode = -EFAULT;
-					goto cleanup;
+					goto sg_list_cleanup;
 				}
 			}
 		}
@@ -741,6 +742,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar
 			       "%s: Could not copy message context FROM user\n",
 			       c->name);
 			rcode = -EFAULT;
+			goto sg_list_cleanup;
 		}
 		if (copy_to_user(user_reply, reply, reply_size)) {
 			printk(KERN_WARNING
@@ -749,6 +751,10 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar
 		}
 	}
 
+      sg_list_cleanup:
+	for (i = 0; i < sg_index; i++)
+		i2o_dma_free(&c->pdev->dev, &sg_list[i]);
+
       cleanup:
 	kfree(reply);
 	return rcode;
@@ -862,8 +868,8 @@ static int i2o_cfg_passthru(unsigned long arg)
 		if (sg_count > SG_TABLESIZE) {
 			printk(KERN_DEBUG "%s:IOCTL SG List too large (%u)\n",
 			       c->name, sg_count);
-			kfree(reply);
-			return -EINVAL;
+			rcode = -EINVAL;
+			goto cleanup;
 		}
 
 		for (i = 0; i < sg_count; i++) {
@@ -875,7 +881,7 @@ static int i2o_cfg_passthru(unsigned long arg)
 				       "%s:Bad SG element %d - not simple (%x)\n",
 				       c->name, i, sg[i].flag_count);
 				rcode = -EINVAL;
-				goto cleanup;
+				goto sg_list_cleanup;
 			}
 			sg_size = sg[i].flag_count & 0xffffff;
 			/* Allocate memory for the transfer */
@@ -885,7 +891,7 @@ static int i2o_cfg_passthru(unsigned long arg)
 				       "%s: Could not allocate SG buffer - size = %d buffer number %d of %d\n",
 				       c->name, sg_size, i, sg_count);
 				rcode = -ENOMEM;
-				goto cleanup;
+				goto sg_list_cleanup;
 			}
 			sg_list[sg_index++] = p;	// sglist indexed with input frame, not our internal frame.
 			/* Copy in the user's SG buffer if necessary */
@@ -899,7 +905,7 @@ static int i2o_cfg_passthru(unsigned long arg)
 					       "%s: Could not copy SG buf %d FROM user\n",
 					       c->name, i);
 					rcode = -EFAULT;
-					goto cleanup;
+					goto sg_list_cleanup;
 				}
 			}
 			//TODO 64bit fix
@@ -909,7 +915,7 @@ static int i2o_cfg_passthru(unsigned long arg)
 
 	rcode = i2o_msg_post_wait(c, m, 60);
 	if (rcode)
-		goto cleanup;
+		goto sg_list_cleanup;
 
 	if (sg_offset) {
 		u32 msg[128];
@@ -924,14 +930,14 @@ static int i2o_cfg_passthru(unsigned long arg)
 		// get user msg size in u32s
 		if (get_user(size, &user_msg[0])) {
 			rcode = -EFAULT;
-			goto cleanup;
+			goto sg_list_cleanup;
 		}
 		size = size >> 16;
 		size *= 4;
 		/* Copy in the user's I2O command */
 		if (copy_from_user(msg, user_msg, size)) {
 			rcode = -EFAULT;
-			goto cleanup;
+			goto sg_list_cleanup;
 		}
 		sg_count =
 		    (size - sg_offset * 4) / sizeof(struct sg_simple_element);
@@ -953,7 +959,7 @@ static int i2o_cfg_passthru(unsigned long arg)
 					       c->name, sg_list[j],
 					       sg[j].addr_bus);
 					rcode = -EFAULT;
-					goto cleanup;
+					goto sg_list_cleanup;
 				}
 			}
 		}
@@ -975,6 +981,10 @@ static int i2o_cfg_passthru(unsigned long arg)
 		}
 	}
 
+      sg_list_cleanup:
+	for (i = 0; i < sg_index; i++)
+		kfree(sg_list[i]);
+
       cleanup:
 	kfree(reply);
 	return rcode;
diff --git a/drivers/message/i2o/i2o_scsi.c b/drivers/message/i2o/i2o_scsi.c
index 43f5875e0be..af40f1c1ec7 100644
--- a/drivers/message/i2o/i2o_scsi.c
+++ b/drivers/message/i2o/i2o_scsi.c
@@ -103,7 +103,8 @@ static struct i2o_scsi_host *i2o_scsi_host_alloc(struct i2o_controller *c)
 
 	list_for_each_entry(i2o_dev, &c->devices, list)
 	    if (i2o_dev->lct_data.class_id == I2O_CLASS_BUS_ADAPTER_PORT) {
-		if (i2o_parm_field_get(i2o_dev, 0x0000, 0, &type, 1) || (type == 1))	/* SCSI bus */
+		if (i2o_parm_field_get(i2o_dev, 0x0000, 0, &type, 1)
+		   && (type == 0x01))	/* SCSI bus */
 			max_channel++;
 	}
 
diff --git a/drivers/message/i2o/pci.c b/drivers/message/i2o/pci.c
index e772752f056..579a8b7a212 100644
--- a/drivers/message/i2o/pci.c
+++ b/drivers/message/i2o/pci.c
@@ -31,10 +31,6 @@
 #include <linux/interrupt.h>
 #include <linux/i2o.h>
 
-#ifdef CONFIG_MTRR
-#include <asm/mtrr.h>
-#endif				// CONFIG_MTRR
-
 /* Module internal functions from other sources */
 extern struct i2o_controller *i2o_iop_alloc(void);
 extern void i2o_iop_free(struct i2o_controller *);
@@ -49,6 +45,8 @@ extern int i2o_driver_dispatch(struct i2o_controller *, u32,
 static struct pci_device_id __devinitdata i2o_pci_ids[] = {
 	{PCI_DEVICE_CLASS(PCI_CLASS_INTELLIGENT_I2O << 8, 0xffff00)},
 	{PCI_DEVICE(PCI_VENDOR_ID_DPT, 0xa511)},
+	{.vendor = PCI_VENDOR_ID_INTEL,.device = 0x1962,
+	 .subvendor = PCI_VENDOR_ID_PROMISE,.subdevice = PCI_ANY_ID},
 	{0}
 };
 
@@ -97,13 +95,6 @@ static void i2o_pci_free(struct i2o_controller *c)
 	i2o_dma_free(dev, &c->hrt);
 	i2o_dma_free(dev, &c->status);
 
-#ifdef CONFIG_MTRR
-	if (c->mtrr_reg0 >= 0)
-		mtrr_del(c->mtrr_reg0, 0, 0);
-	if (c->mtrr_reg1 >= 0)
-		mtrr_del(c->mtrr_reg1, 0, 0);
-#endif
-
 	if (c->raptor && c->in_queue.virt)
 		iounmap(c->in_queue.virt);
 
@@ -178,14 +169,15 @@ static int __devinit i2o_pci_alloc(struct i2o_controller *c)
 		       c->name, (unsigned long)c->base.phys,
 		       (unsigned long)c->base.len);
 
-	c->base.virt = ioremap(c->base.phys, c->base.len);
+	c->base.virt = ioremap_nocache(c->base.phys, c->base.len);
 	if (!c->base.virt) {
 		printk(KERN_ERR "%s: Unable to map controller.\n", c->name);
 		return -ENOMEM;
 	}
 
 	if (c->raptor) {
-		c->in_queue.virt = ioremap(c->in_queue.phys, c->in_queue.len);
+		c->in_queue.virt =
+		    ioremap_nocache(c->in_queue.phys, c->in_queue.len);
 		if (!c->in_queue.virt) {
 			printk(KERN_ERR "%s: Unable to map controller.\n",
 			       c->name);
@@ -199,40 +191,6 @@ static int __devinit i2o_pci_alloc(struct i2o_controller *c)
 	c->post_port = c->base.virt + 0x40;
 	c->reply_port = c->base.virt + 0x44;
 
-#ifdef CONFIG_MTRR
-	/* Enable Write Combining MTRR for IOP's memory region */
-	c->mtrr_reg0 = mtrr_add(c->in_queue.phys, c->in_queue.len,
-				MTRR_TYPE_WRCOMB, 1);
-	c->mtrr_reg1 = -1;
-
-	if (c->mtrr_reg0 < 0)
-		printk(KERN_WARNING "%s: could not enable write combining "
-		       "MTRR\n", c->name);
-	else
-		printk(KERN_INFO "%s: using write combining MTRR\n", c->name);
-
-	/*
-	 * If it is an INTEL i960 I/O processor then set the first 64K to
-	 * Uncacheable since the region contains the messaging unit which
-	 * shouldn't be cached.
-	 */
-	if ((pdev->vendor == PCI_VENDOR_ID_INTEL ||
-	     pdev->vendor == PCI_VENDOR_ID_DPT) && !c->raptor) {
-		printk(KERN_INFO "%s: MTRR workaround for Intel i960 processor"
-		       "\n", c->name);
-		c->mtrr_reg1 = mtrr_add(c->base.phys, 0x10000,
-					MTRR_TYPE_UNCACHABLE, 1);
-
-		if (c->mtrr_reg1 < 0) {
-			printk(KERN_WARNING "%s: Error in setting "
-			       "MTRR_TYPE_UNCACHABLE\n", c->name);
-			mtrr_del(c->mtrr_reg0, c->in_queue.phys,
-				 c->in_queue.len);
-			c->mtrr_reg0 = -1;
-		}
-	}
-#endif
-
 	if (i2o_dma_alloc(dev, &c->status, 8, GFP_KERNEL)) {
 		i2o_pci_free(c);
 		return -ENOMEM;
@@ -385,28 +343,25 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev,
 {
 	struct i2o_controller *c;
 	int rc;
+	struct pci_dev *i960 = NULL;
 
 	printk(KERN_INFO "i2o: Checking for PCI I2O controllers...\n");
 
 	if ((pdev->class & 0xff) > 1) {
-		printk(KERN_WARNING "i2o: I2O controller found but does not "
-		       "support I2O 1.5 (skipping).\n");
+		printk(KERN_WARNING "i2o: %s does not support I2O 1.5 "
+		       "(skipping).\n", pci_name(pdev));
 		return -ENODEV;
 	}
 
 	if ((rc = pci_enable_device(pdev))) {
-		printk(KERN_WARNING "i2o: I2O controller found but could not be"
-		       " enabled.\n");
+		printk(KERN_WARNING "i2o: couldn't enable device %s\n",
+		       pci_name(pdev));
 		return rc;
 	}
 
-	printk(KERN_INFO "i2o: I2O controller found on bus %d at %d.\n",
-	       pdev->bus->number, pdev->devfn);
-
 	if (pci_set_dma_mask(pdev, DMA_32BIT_MASK)) {
-		printk(KERN_WARNING "i2o: I2O controller on bus %d at %d: No "
-		       "suitable DMA available!\n", pdev->bus->number,
-		       pdev->devfn);
+		printk(KERN_WARNING "i2o: no suitable DMA found for %s\n",
+		       pci_name(pdev));
 		rc = -ENODEV;
 		goto disable;
 	}
@@ -415,11 +370,13 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev,
 
 	c = i2o_iop_alloc();
 	if (IS_ERR(c)) {
-		printk(KERN_ERR "i2o: memory for I2O controller could not be "
-		       "allocated\n");
+		printk(KERN_ERR "i2o: couldn't allocate memory for %s\n",
+		       pci_name(pdev));
 		rc = PTR_ERR(c);
 		goto disable;
-	}
+	} else
+		printk(KERN_INFO "%s: controller found (%s)\n", c->name,
+		       pci_name(pdev));
 
 	c->pdev = pdev;
 	c->device = pdev->dev;
@@ -432,9 +389,18 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev,
 	}
 
 	if (pdev->subsystem_vendor == PCI_VENDOR_ID_PROMISE) {
+		/*
+		 * Expose the ship behind i960 for initialization, or it will
+		 * failed
+		 */
+		i960 =
+		    pci_find_slot(c->pdev->bus->number,
+				  PCI_DEVFN(PCI_SLOT(c->pdev->devfn), 0));
+
+		if (i960)
+			pci_write_config_word(i960, 0x42, 0);
+
 		c->promise = 1;
-		printk(KERN_INFO "%s: Promise workarounds activated.\n",
-		       c->name);
 	}
 
 	/* Cards that go bananas if you quiesce them before you reset them. */
@@ -459,6 +425,9 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev,
 	if ((rc = i2o_iop_add(c)))
 		goto uninstall;
 
+	if (i960)
+		pci_write_config_word(i960, 0x42, 0x03ff);
+
 	return 0;
 
       uninstall:
diff --git a/include/linux/i2o-dev.h b/include/linux/i2o-dev.h
index ef7f644dd87..3414325bdcf 100644
--- a/include/linux/i2o-dev.h
+++ b/include/linux/i2o-dev.h
@@ -24,6 +24,14 @@
 #define MAX_I2O_CONTROLLERS	32
 
 //#include <linux/ioctl.h>
+#ifndef __KERNEL__
+
+typedef unsigned char u8;
+typedef unsigned short u16;
+typedef unsigned int u32;
+
+#endif				/* __KERNEL__ */
+
 
 /*
  * I2O Control IOCTLs and structures
@@ -126,14 +134,6 @@ struct i2o_evt_get {
 #define I2O_BUS_CARDBUS 7
 #define I2O_BUS_UNKNOWN 0x80
 
-#ifndef __KERNEL__
-
-typedef unsigned char u8;
-typedef unsigned short u16;
-typedef unsigned int u32;
-
-#endif				/* __KERNEL__ */
-
 typedef struct _i2o_pci_bus {
 	u8 PciFunctionNumber;
 	u8 PciDeviceNumber;
diff --git a/include/linux/i2o.h b/include/linux/i2o.h
index ea9a3ad4b67..40e45a83d3f 100644
--- a/include/linux/i2o.h
+++ b/include/linux/i2o.h
@@ -152,11 +152,6 @@ struct i2o_controller {
 	unsigned int raptor:1;		/* split bar */
 	unsigned int promise:1;		/* Promise controller */
 
-#ifdef CONFIG_MTRR
-	int mtrr_reg0;
-	int mtrr_reg1;
-#endif
-
 	struct list_head devices;	/* list of I2O devices */
 
 	struct notifier_block *event_notifer;	/* Events */
-- 
cgit v1.2.3-70-g09d2


From f88e119c4b824a5017456fa094950d0f4092d96c Mon Sep 17 00:00:00 2001
From: Markus Lidel <Markus.Lidel@shadowconnect.com>
Date: Thu, 23 Jun 2005 22:02:14 -0700
Subject: [PATCH] I2O: first code cleanup of spare warnings and unused
 functions

Changes:

 - Removed unnecessary checking of NULL before calling kfree()
 - Make some functions static
 - Changed pr_debug() into osm_debug()
 - Use i2o_msg_in_to_virt() for getting a pointer to the message frame
 - Cleaned up some comments
 - Changed some le32_to_cpu() into readl() where necessary
 - Make error messages of OSM's look the same
 - Cleaned up error handling in i2o_block_end_request()
 - Removed unused error handling of failed messages in Block-OSM, which
   are not allowed by the I2O spec
 - Corrected the blocksize detection in i2o_block
 - Added hrt and lct sysfs-attribute to controller
 - Call done() function in SCSI-OSM after freeing DMA buffers
 - Removed unneeded variable for message size calculation in
   i2o_scsi_queuecommand()
 - Make some changes to remove sparse warnings
 - Reordered some functions
 - Cleaned up controller initialization
 - Replaced some magic numbers by defines
 - Removed unnecessary dma_sync_single_for_cpu() call on coherent DMA
 - Removed some unused fields in i2o_controller and removed some unused
   functions

Signed-off-by: Markus Lidel <Markus.Lidel@shadowconnect.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/message/i2o/device.c     |   9 +-
 drivers/message/i2o/driver.c     |  46 +++++----
 drivers/message/i2o/exec-osm.c   |  47 +++++----
 drivers/message/i2o/i2o_block.c  | 211 +++++++++++++--------------------------
 drivers/message/i2o/i2o_block.h  |   2 +-
 drivers/message/i2o/i2o_config.c | 118 +++++++++++++++++++++-
 drivers/message/i2o/i2o_scsi.c   |  31 +++---
 drivers/message/i2o/iop.c        |  87 ++++++++++------
 drivers/message/i2o/pci.c        |  67 ++++++-------
 include/linux/i2o.h              |  74 +++-----------
 10 files changed, 356 insertions(+), 336 deletions(-)

(limited to 'include')

diff --git a/drivers/message/i2o/device.c b/drivers/message/i2o/device.c
index 280627ae6cf..f1b7eb63d54 100644
--- a/drivers/message/i2o/device.c
+++ b/drivers/message/i2o/device.c
@@ -282,8 +282,7 @@ int i2o_device_parse_lct(struct i2o_controller *c)
 
 	down(&c->lct_lock);
 
-	if (c->lct)
-		kfree(c->lct);
+	kfree(c->lct);
 
 	lct = c->dlct.virt;
 
@@ -447,8 +446,8 @@ static struct class_interface i2o_device_class_interface = {
  *	ResultCount, ErrorInfoSize, BlockStatus and BlockSize.
  */
 
-int i2o_parm_issue(struct i2o_device *i2o_dev, int cmd, void *oplist,
-		   int oplen, void *reslist, int reslen)
+static int i2o_parm_issue(struct i2o_device *i2o_dev, int cmd, void *oplist,
+			  int oplen, void *reslist, int reslen)
 {
 	struct i2o_message __iomem *msg;
 	u32 m;
@@ -540,7 +539,7 @@ int i2o_parm_field_get(struct i2o_device *i2o_dev, int group, int field,
 		opblk[4] = -1;
 
 	size = i2o_parm_issue(i2o_dev, I2O_CMD_UTIL_PARAMS_GET, opblk,
-			      sizeof(opblk), resblk, sizeof(resblk));
+			      sizeof(opblk), resblk, buflen + 8);
 
 	memcpy(buf, resblk + 8, buflen);	/* cut off header */
 
diff --git a/drivers/message/i2o/driver.c b/drivers/message/i2o/driver.c
index c71e68f70e7..bebdd509b5d 100644
--- a/drivers/message/i2o/driver.c
+++ b/drivers/message/i2o/driver.c
@@ -18,7 +18,7 @@
 #include <linux/rwsem.h>
 #include <linux/i2o.h>
 
-#define OSM_NAME	"core"
+#define OSM_NAME	"i2o"
 
 /* max_drivers - Maximum I2O drivers (OSMs) which could be registered */
 unsigned int i2o_max_drivers = I2O_MAX_DRIVERS;
@@ -78,17 +78,16 @@ int i2o_driver_register(struct i2o_driver *drv)
 	int rc = 0;
 	unsigned long flags;
 
-	pr_debug("i2o: Register driver %s\n", drv->name);
+	osm_debug("Register driver %s\n", drv->name);
 
 	if (drv->event) {
 		drv->event_queue = create_workqueue(drv->name);
 		if (!drv->event_queue) {
-			printk(KERN_ERR "i2o: Could not initialize event queue "
-			       "for driver %s\n", drv->name);
+			osm_err("Could not initialize event queue for driver "
+				"%s\n", drv->name);
 			return -EFAULT;
 		}
-		pr_debug("i2o: Event queue initialized for driver %s\n",
-			 drv->name);
+		osm_debug("Event queue initialized for driver %s\n", drv->name);
 	} else
 		drv->event_queue = NULL;
 
@@ -99,8 +98,8 @@ int i2o_driver_register(struct i2o_driver *drv)
 
 	for (i = 0; i2o_drivers[i]; i++)
 		if (i >= i2o_max_drivers) {
-			printk(KERN_ERR "i2o: too many drivers registered, "
-			       "increase max_drivers\n");
+			osm_err("too many drivers registered, increase "
+				"max_drivers\n");
 			spin_unlock_irqrestore(&i2o_drivers_lock, flags);
 			return -EFAULT;
 		}
@@ -110,8 +109,7 @@ int i2o_driver_register(struct i2o_driver *drv)
 
 	spin_unlock_irqrestore(&i2o_drivers_lock, flags);
 
-	pr_debug("i2o: driver %s gets context id %d\n", drv->name,
-		 drv->context);
+	osm_debug("driver %s gets context id %d\n", drv->name, drv->context);
 
 	list_for_each_entry(c, &i2o_controllers, list) {
 		struct i2o_device *i2o_dev;
@@ -141,7 +139,7 @@ void i2o_driver_unregister(struct i2o_driver *drv)
 	struct i2o_controller *c;
 	unsigned long flags;
 
-	pr_debug("i2o: unregister driver %s\n", drv->name);
+	osm_debug("unregister driver %s\n", drv->name);
 
 	driver_unregister(&drv->driver);
 
@@ -161,7 +159,7 @@ void i2o_driver_unregister(struct i2o_driver *drv)
 	if (drv->event_queue) {
 		destroy_workqueue(drv->event_queue);
 		drv->event_queue = NULL;
-		pr_debug("i2o: event queue removed for %s\n", drv->name);
+		osm_debug("event queue removed for %s\n", drv->name);
 	}
 };
 
@@ -178,15 +176,15 @@ void i2o_driver_unregister(struct i2o_driver *drv)
  *	on success and if the message should be flushed afterwords. Returns
  *	negative error code on failure (the message will be flushed too).
  */
-int i2o_driver_dispatch(struct i2o_controller *c, u32 m,
-			struct i2o_message __iomem *msg)
+int i2o_driver_dispatch(struct i2o_controller *c, u32 m)
 {
 	struct i2o_driver *drv;
+	struct i2o_message __iomem *msg = i2o_msg_out_to_virt(c, m);
 	u32 context = readl(&msg->u.s.icntxt);
 
 	if (unlikely(context >= i2o_max_drivers)) {
-		printk(KERN_WARNING "%s: Spurious reply to unknown driver "
-		       "%d\n", c->name, readl(&msg->u.s.icntxt));
+		osm_warn("%s: Spurious reply to unknown driver %d\n", c->name,
+			 context);
 		return -EIO;
 	}
 
@@ -195,7 +193,8 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m,
 	spin_unlock(&i2o_drivers_lock);
 
 	if (unlikely(!drv)) {
-		osm_warn("Spurious reply to unknown driver %d\n", context);
+		osm_warn("%s: Spurious reply to unknown driver %d\n", c->name,
+			 context);
 		return -EIO;
 	}
 
@@ -207,6 +206,9 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m,
 
 		osm_debug("event received from device %d\n", tid);
 
+		if (!drv->event)
+			return -EIO;
+
 		/* cut of header from message size (in 32-bit words) */
 		size = (readl(&msg->u.head[0]) >> 16) - 5;
 
@@ -231,8 +233,8 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m,
 	}
 
 	if (unlikely(!drv->reply)) {
-		pr_debug("%s: Reply to driver %s, but no reply function"
-			 " defined!\n", c->name, drv->name);
+		osm_debug("%s: Reply to driver %s, but no reply function"
+			  " defined!\n", c->name, drv->name);
 		return -EIO;
 	}
 
@@ -333,11 +335,11 @@ int __init i2o_driver_init(void)
 	if ((i2o_max_drivers < 2) || (i2o_max_drivers > 64) ||
 	    ((i2o_max_drivers ^ (i2o_max_drivers - 1)) !=
 	     (2 * i2o_max_drivers - 1))) {
-		printk(KERN_WARNING "i2o: max_drivers set to %d, but must be "
-		       ">=2 and <= 64 and a power of 2\n", i2o_max_drivers);
+		osm_warn("max_drivers set to %d, but must be >=2 and <= 64 and "
+			 "a power of 2\n", i2o_max_drivers);
 		i2o_max_drivers = I2O_MAX_DRIVERS;
 	}
-	printk(KERN_INFO "i2o: max drivers = %d\n", i2o_max_drivers);
+	osm_info("max drivers = %d\n", i2o_max_drivers);
 
 	i2o_drivers =
 	    kmalloc(i2o_max_drivers * sizeof(*i2o_drivers), GFP_KERNEL);
diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c
index 1e28e886f1c..5581344fbba 100644
--- a/drivers/message/i2o/exec-osm.c
+++ b/drivers/message/i2o/exec-osm.c
@@ -108,7 +108,8 @@ static void i2o_exec_wait_free(struct i2o_exec_wait *wait)
  *	buffer must not be freed. Instead the event completion will free them
  *	for you. In all other cases the buffer are your problem.
  *
- *	Returns 0 on success or negative error code on failure.
+ *	Returns 0 on success, negative error code on timeout or positive error
+ *	code from reply.
  */
 int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long
 			  timeout, struct i2o_dma *dma)
@@ -116,7 +117,7 @@ int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long
 	DECLARE_WAIT_QUEUE_HEAD(wq);
 	struct i2o_exec_wait *wait;
 	static u32 tcntxt = 0x80000000;
-	struct i2o_message __iomem *msg = c->in_queue.virt + m;
+	struct i2o_message __iomem *msg = i2o_msg_in_to_virt(c, m);
 	int rc = 0;
 
 	wait = i2o_exec_wait_alloc();
@@ -161,8 +162,7 @@ int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long
 	barrier();
 
 	if (wait->complete) {
-		if (readl(&wait->msg->body[0]) >> 24)
-			rc = readl(&wait->msg->body[0]) & 0xff;
+		rc = readl(&wait->msg->body[0]) >> 24;
 		i2o_flush_reply(c, wait->m);
 		i2o_exec_wait_free(wait);
 	} else {
@@ -187,6 +187,7 @@ int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long
  *	@c: I2O controller which answers
  *	@m: message id
  *	@msg: pointer to the I2O reply message
+ *	@context: transaction context of request
  *
  *	This function is called in interrupt context only. If the reply reached
  *	before the timeout, the i2o_exec_wait struct is filled with the message
@@ -201,14 +202,12 @@ int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long
  *	message must also be given back to the controller.
  */
 static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m,
-				      struct i2o_message __iomem *msg)
+				      struct i2o_message __iomem *msg,
+				      u32 context)
 {
 	struct i2o_exec_wait *wait, *tmp;
 	static spinlock_t lock = SPIN_LOCK_UNLOCKED;
 	int rc = 1;
-	u32 context;
-
-	context = readl(&msg->u.s.tcntxt);
 
 	/*
 	 * We need to search through the i2o_exec_wait_list to see if the given
@@ -251,7 +250,7 @@ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m,
 
 	spin_unlock(&lock);
 
-	pr_debug("%s: Bogus reply in POST WAIT (tr-context: %08x)!\n", c->name,
+	osm_warn("%s: Bogus reply in POST WAIT (tr-context: %08x)!\n", c->name,
 		 context);
 
 	return -1;
@@ -321,29 +320,35 @@ static void i2o_exec_lct_modified(struct i2o_controller *c)
  *	code on failure and if the reply should be flushed.
  */
 static int i2o_exec_reply(struct i2o_controller *c, u32 m,
-			  struct i2o_message *msg)
+			  struct i2o_message __iomem *msg)
 {
-	if (le32_to_cpu(msg->u.head[0]) & MSG_FAIL) {	// Fail bit is set
-		struct i2o_message __iomem *pmsg;	/* preserved message */
+	u32 context;
+
+	if (readl(&msg->u.head[0]) & MSG_FAIL) {
+		/*
+		 * If Fail bit is set we must take the transaction context of
+		 * the preserved message to find the right request again.
+		 */
+		struct i2o_message __iomem *pmsg;
 		u32 pm;
 
-		pm = le32_to_cpu(msg->body[3]);
+		pm = readl(&msg->body[3]);
 
 		pmsg = i2o_msg_in_to_virt(c, pm);
 
 		i2o_report_status(KERN_INFO, "i2o_core", msg);
 
-		/* Release the preserved msg by resubmitting it as a NOP */
-		i2o_msg_nop(c, pm);
+		context = readl(&pmsg->u.s.tcntxt);
 
-		/* If reply to i2o_post_wait failed, return causes a timeout */
-		return -1;
-	}
+		/* Release the preserved msg */
+		i2o_msg_nop(c, pm);
+	} else
+		context = readl(&msg->u.s.tcntxt);
 
-	if (le32_to_cpu(msg->u.s.tcntxt) & 0x80000000)
-		return i2o_msg_post_wait_complete(c, m, msg);
+	if (context & 0x80000000)
+		return i2o_msg_post_wait_complete(c, m, msg, context);
 
-	if ((le32_to_cpu(msg->u.head[1]) >> 24) == I2O_CMD_LCT_NOTIFY) {
+	if ((readl(&msg->u.head[1]) >> 24) == I2O_CMD_LCT_NOTIFY) {
 		struct work_struct *work;
 
 		pr_debug("%s: LCT notify received\n", c->name);
diff --git a/drivers/message/i2o/i2o_block.c b/drivers/message/i2o/i2o_block.c
index 4830b775906..e69421e36ac 100644
--- a/drivers/message/i2o/i2o_block.c
+++ b/drivers/message/i2o/i2o_block.c
@@ -104,7 +104,8 @@ static int i2o_block_remove(struct device *dev)
 	struct i2o_device *i2o_dev = to_i2o_device(dev);
 	struct i2o_block_device *i2o_blk_dev = dev_get_drvdata(dev);
 
-	osm_info("Device removed %s\n", i2o_blk_dev->gd->disk_name);
+	osm_info("device removed (TID: %03x): %s\n", i2o_dev->lct_data.tid,
+		 i2o_blk_dev->gd->disk_name);
 
 	i2o_event_register(i2o_dev, &i2o_block_driver, 0, 0);
 
@@ -400,71 +401,62 @@ static void i2o_block_delayed_request_fn(void *delayed_request)
 };
 
 /**
- *	i2o_block_reply - Block OSM reply handler.
- *	@c: I2O controller from which the message arrives
- *	@m: message id of reply
- *	qmsg: the actuall I2O message reply
+ *	i2o_block_end_request - Post-processing of completed commands
+ *	@req: request which should be completed
+ *	@uptodate: 1 for success, 0 for I/O error, < 0 for specific error
+ *	@nr_bytes: number of bytes to complete
  *
- *	This function gets all the message replies.
+ *	Mark the request as complete. The lock must not be held when entering.
  *
  */
-static int i2o_block_reply(struct i2o_controller *c, u32 m,
-			   struct i2o_message *msg)
+static void i2o_block_end_request(struct request *req, int uptodate,
+				  int nr_bytes)
 {
-	struct i2o_block_request *ireq;
-	struct request *req;
-	struct i2o_block_device *dev;
-	struct request_queue *q;
-	u8 st;
+	struct i2o_block_request *ireq = req->special;
+	struct i2o_block_device *dev = ireq->i2o_blk_dev;
+	request_queue_t *q = dev->gd->queue;
 	unsigned long flags;
 
-	/* FAILed message */
-	if (unlikely(le32_to_cpu(msg->u.head[0]) & (1 << 13))) {
-		struct i2o_message *pmsg;
-		u32 pm;
-
-		/*
-		 * FAILed message from controller
-		 * We increment the error count and abort it
-		 *
-		 * In theory this will never happen.  The I2O block class
-		 * specification states that block devices never return
-		 * FAILs but instead use the REQ status field...but
-		 * better be on the safe side since no one really follows
-		 * the spec to the book :)
-		 */
-		pm = le32_to_cpu(msg->body[3]);
-		pmsg = i2o_msg_in_to_virt(c, pm);
+	if (end_that_request_chunk(req, uptodate, nr_bytes)) {
+		int leftover = (req->hard_nr_sectors << 9);
 
-		req = i2o_cntxt_list_get(c, le32_to_cpu(pmsg->u.s.tcntxt));
-		if (unlikely(!req)) {
-			osm_err("NULL reply received!\n");
-			return -1;
-		}
+		if (blk_pc_request(req))
+			leftover = req->data_len;
 
-		ireq = req->special;
-		dev = ireq->i2o_blk_dev;
-		q = dev->gd->queue;
+		if (end_io_error(uptodate))
+			end_that_request_chunk(req, 0, leftover);
+	}
 
-		req->errors++;
+	add_disk_randomness(req->rq_disk);
 
-		spin_lock_irqsave(q->queue_lock, flags);
+	spin_lock_irqsave(q->queue_lock, flags);
 
-		while (end_that_request_chunk(req, !req->errors,
-					      le32_to_cpu(pmsg->body[1]))) ;
-		end_that_request_last(req);
+	end_that_request_last(req);
+	dev->open_queue_depth--;
+	list_del(&ireq->queue);
 
-		dev->open_queue_depth--;
-		list_del(&ireq->queue);
-		blk_start_queue(q);
+	blk_start_queue(q);
 
-		spin_unlock_irqrestore(q->queue_lock, flags);
+	spin_unlock_irqrestore(q->queue_lock, flags);
 
-		/* Now flush the message by making it a NOP */
-		i2o_msg_nop(c, pm);
+	i2o_block_sglist_free(ireq);
+	i2o_block_request_free(ireq);
+};
 
-		return -1;
-	}
+/**
+ *	i2o_block_reply - Block OSM reply handler.
+ *	@c: I2O controller from which the message arrives
+ *	@m: message id of reply
+ *	qmsg: the actuall I2O message reply
+ *
+ *	This function gets all the message replies.
+ *
+ */
+static int i2o_block_reply(struct i2o_controller *c, u32 m,
+			   struct i2o_message *msg)
+{
+	struct request *req;
+	int uptodate = 1;
 
 	req = i2o_cntxt_list_get(c, le32_to_cpu(msg->u.s.tcntxt));
 	if (unlikely(!req)) {
@@ -472,61 +464,13 @@ static int i2o_block_reply(struct i2o_controller *c, u32 m,
 		return -1;
 	}
 
-	ireq = req->special;
-	dev = ireq->i2o_blk_dev;
-	q = dev->gd->queue;
-
-	if (unlikely(!dev->i2o_dev)) {
-		/*
-		 * This is HACK, but Intel Integrated RAID allows user
-		 * to delete a volume that is claimed, locked, and in use
-		 * by the OS. We have to check for a reply from a
-		 * non-existent device and flag it as an error or the system
-		 * goes kaput...
-		 */
-		req->errors++;
-		osm_warn("Data transfer to deleted device!\n");
-		spin_lock_irqsave(q->queue_lock, flags);
-		while (end_that_request_chunk
-		       (req, !req->errors, le32_to_cpu(msg->body[1]))) ;
-		end_that_request_last(req);
-
-		dev->open_queue_depth--;
-		list_del(&ireq->queue);
-		blk_start_queue(q);
-
-		spin_unlock_irqrestore(q->queue_lock, flags);
-		return -1;
-	}
-
 	/*
 	 *      Lets see what is cooking. We stuffed the
 	 *      request in the context.
 	 */
 
-	st = le32_to_cpu(msg->body[0]) >> 24;
-
-	if (st != 0) {
-		int err;
-		char *bsa_errors[] = {
-			"Success",
-			"Media Error",
-			"Failure communicating to device",
-			"Device Failure",
-			"Device is not ready",
-			"Media not present",
-			"Media is locked by another user",
-			"Media has failed",
-			"Failure communicating to device",
-			"Device bus failure",
-			"Device is locked by another user",
-			"Device is write protected",
-			"Device has reset",
-			"Volume has changed, waiting for acknowledgement"
-		};
-
-		err = le32_to_cpu(msg->body[0]) & 0xffff;
-
+	if ((le32_to_cpu(msg->body[0]) >> 24) != 0) {
+		u32 status = le32_to_cpu(msg->body[0]);
 		/*
 		 *      Device not ready means two things. One is that the
 		 *      the thing went offline (but not a removal media)
@@ -539,40 +483,23 @@ static int i2o_block_reply(struct i2o_controller *c, u32 m,
 		 *      Don't stick a supertrak100 into cache aggressive modes
 		 */
 
-		osm_err("block-osm: /dev/%s error: %s", dev->gd->disk_name,
-			bsa_errors[le32_to_cpu(msg->body[0]) & 0xffff]);
-		if (le32_to_cpu(msg->body[0]) & 0x00ff0000)
-			printk(KERN_ERR " - DDM attempted %d retries",
-			       (le32_to_cpu(msg->body[0]) >> 16) & 0x00ff);
-		printk(KERN_ERR ".\n");
-		req->errors++;
-	} else
-		req->errors = 0;
-
-	if (!end_that_request_chunk
-	    (req, !req->errors, le32_to_cpu(msg->body[1]))) {
-		add_disk_randomness(req->rq_disk);
-		spin_lock_irqsave(q->queue_lock, flags);
+		osm_err("%03x error status: %02x, detailed status: %04x\n",
+			(le32_to_cpu(msg->u.head[1]) >> 12 & 0xfff),
+			status >> 24, status & 0xffff);
 
-		end_that_request_last(req);
+		req->errors++;
 
-		dev->open_queue_depth--;
-		list_del(&ireq->queue);
-		blk_start_queue(q);
+		uptodate = 0;
+	}
 
-		spin_unlock_irqrestore(q->queue_lock, flags);
-
-		i2o_block_sglist_free(ireq);
-		i2o_block_request_free(ireq);
-	} else
-		osm_err("still remaining chunks\n");
+	i2o_block_end_request(req, uptodate, le32_to_cpu(msg->body[1]));
 
 	return 1;
 };
 
 static void i2o_block_event(struct i2o_event *evt)
 {
-	osm_info("block-osm: event received\n");
+	osm_info("event received\n");
 	kfree(evt);
 };
 
@@ -875,9 +802,7 @@ static int i2o_block_transfer(struct request *req)
 		sg++;
 	}
 
-	writel(I2O_MESSAGE_SIZE
-	       (((unsigned long)mptr -
-		 (unsigned long)&msg->u.head[0]) >> 2) | SGL_OFFSET_8,
+	writel(I2O_MESSAGE_SIZE(mptr - &msg->u.head[0]) | SGL_OFFSET_8,
 	       &msg->u.head[0]);
 
 	list_add_tail(&ireq->queue, &dev->open_queue);
@@ -1048,7 +973,6 @@ static int i2o_block_probe(struct device *dev)
 	int rc;
 	u64 size;
 	u32 blocksize;
-	u16 power;
 	u32 flags, status;
 	int segments;
 
@@ -1058,8 +982,6 @@ static int i2o_block_probe(struct device *dev)
 		return -ENODEV;
 	}
 
-	osm_info("New device detected (TID: %03x)\n", i2o_dev->lct_data.tid);
-
 	if (i2o_device_claim(i2o_dev)) {
 		osm_warn("Unable to claim device. Installation aborted\n");
 		rc = -EFAULT;
@@ -1111,15 +1033,21 @@ static int i2o_block_probe(struct device *dev)
 	 *      Ask for the current media data. If that isn't supported
 	 *      then we ask for the device capacity data
 	 */
-	if (i2o_parm_field_get(i2o_dev, 0x0004, 1, &blocksize, 4) != 0
-	    || i2o_parm_field_get(i2o_dev, 0x0004, 0, &size, 8) != 0) {
-		i2o_parm_field_get(i2o_dev, 0x0000, 3, &blocksize, 4);
-		i2o_parm_field_get(i2o_dev, 0x0000, 4, &size, 8);
-	}
-	osm_debug("blocksize = %d\n", blocksize);
+	if (!i2o_parm_field_get(i2o_dev, 0x0004, 0, &size, 8))
+		if (!i2o_parm_field_get(i2o_dev, 0x0000, 4, &size, 8)) {
+			osm_warn("could not get size of %s\n", gd->disk_name);
+			size = 0;
+		}
 
-	if (i2o_parm_field_get(i2o_dev, 0x0000, 2, &power, 2))
-		power = 0;
+	if (!i2o_parm_field_get(i2o_dev, 0x0004, 1, &blocksize, 4))
+		if (!i2o_parm_field_get(i2o_dev, 0x0000, 3, &blocksize, 4)) {
+			osm_warn("unable to get blocksize of %s\n",
+				 gd->disk_name);
+			blocksize = 0;
+		}
+
+	if (!i2o_parm_field_get(i2o_dev, 0x0000, 2, &i2o_blk_dev->power, 2))
+		i2o_blk_dev->power = 0;
 	i2o_parm_field_get(i2o_dev, 0x0000, 5, &flags, 4);
 	i2o_parm_field_get(i2o_dev, 0x0000, 6, &status, 4);
 
@@ -1131,6 +1059,9 @@ static int i2o_block_probe(struct device *dev)
 
 	unit++;
 
+	osm_info("device added (TID: %03x): %s\n", i2o_dev->lct_data.tid,
+		 i2o_blk_dev->gd->disk_name);
+
 	return 0;
 
       claim_release:
diff --git a/drivers/message/i2o/i2o_block.h b/drivers/message/i2o/i2o_block.h
index ddd9a15679c..712111ffa63 100644
--- a/drivers/message/i2o/i2o_block.h
+++ b/drivers/message/i2o/i2o_block.h
@@ -74,7 +74,7 @@ struct i2o_block_device {
 	int rcache;			/* read cache flags */
 	int wcache;			/* write cache flags */
 	int flags;
-	int power;			/* power state */
+	u16 power;			/* power state */
 	int media_change_flag;		/* media changed flag */
 };
 
diff --git a/drivers/message/i2o/i2o_config.c b/drivers/message/i2o/i2o_config.c
index 46d373287a3..383e89a5c9f 100644
--- a/drivers/message/i2o/i2o_config.c
+++ b/drivers/message/i2o/i2o_config.c
@@ -80,13 +80,123 @@ struct i2o_cfg_info {
 static struct i2o_cfg_info *open_files = NULL;
 static ulong i2o_cfg_info_id = 0;
 
-/*
- *	Each of these describes an i2o message handler. They are
- *	multiplexed by the i2o_core code
+/**
+ *	i2o_config_read_hrt - Returns the HRT of the controller
+ *	@kob: kernel object handle
+ *	@buf: buffer into which the HRT should be copied
+ *	@off: file offset
+ *	@count: number of bytes to read
+ *
+ *	Put @count bytes starting at @off into @buf from the HRT of the I2O
+ *	controller corresponding to @kobj.
+ *
+ *	Returns number of bytes copied into buffer.
+ */
+static ssize_t i2o_config_read_hrt(struct kobject *kobj, char *buf,
+				   loff_t offset, size_t count)
+{
+	struct i2o_controller *c = to_i2o_controller(container_of(kobj,
+								  struct device,
+								  kobj));
+	i2o_hrt *hrt = c->hrt.virt;
+
+	u32 size = (hrt->num_entries * hrt->entry_len + 2) * 4;
+
+	if(offset > size)
+		return 0;
+
+	if(offset + count > size)
+		count = size - offset;
+
+	memcpy(buf, (u8 *) hrt + offset, count);
+
+	return count;
+};
+
+/**
+ *	i2o_config_read_lct - Returns the LCT of the controller
+ *	@kob: kernel object handle
+ *	@buf: buffer into which the LCT should be copied
+ *	@off: file offset
+ *	@count: number of bytes to read
+ *
+ *	Put @count bytes starting at @off into @buf from the LCT of the I2O
+ *	controller corresponding to @kobj.
+ *
+ *	Returns number of bytes copied into buffer.
+ */
+static ssize_t i2o_config_read_lct(struct kobject *kobj, char *buf,
+				   loff_t offset, size_t count)
+{
+	struct i2o_controller *c = to_i2o_controller(container_of(kobj,
+								  struct device,
+								  kobj));
+	u32 size = c->lct->table_size * 4;
+
+	if(offset > size)
+		return 0;
+
+	if(offset + count > size)
+		count = size - offset;
+
+	memcpy(buf, (u8 *) c->lct + offset, count);
+
+	return count;
+};
+
+/* attribute for HRT in sysfs */
+static struct bin_attribute i2o_config_hrt_attr = {
+	.attr = {
+		.name = "hrt",
+		.mode = S_IRUGO,
+		.owner = THIS_MODULE
+	},
+	.size = 0,
+	.read = i2o_config_read_hrt
+};
+
+/* attribute for LCT in sysfs */
+static struct bin_attribute i2o_config_lct_attr = {
+	.attr = {
+		.name = "lct",
+		.mode = S_IRUGO,
+		.owner = THIS_MODULE
+	},
+	.size = 0,
+	.read = i2o_config_read_lct
+};
+
+/**
+ *	i2o_config_notify_controller_add - Notify of added controller
+ *	@c: the controller which was added
+ *
+ *	If a I2O controller is added, we catch the notification to add sysfs
+ *	entries.
+ */
+static void i2o_config_notify_controller_add(struct i2o_controller *c)
+{
+	sysfs_create_bin_file(&(c->device.kobj), &i2o_config_hrt_attr);
+	sysfs_create_bin_file(&(c->device.kobj), &i2o_config_lct_attr);
+};
+
+/**
+ *	i2o_config_notify_controller_remove - Notify of removed controller
+ *	@c: the controller which was removed
+ *
+ *	If a I2O controller is removed, we catch the notification to remove the
+ *	sysfs entries.
  */
+static void i2o_config_notify_controller_remove(struct i2o_controller *c)
+{
+	sysfs_remove_bin_file(&c->device.kobj, &i2o_config_lct_attr);
+	sysfs_remove_bin_file(&c->device.kobj, &i2o_config_hrt_attr);
+};
 
+/* Config OSM driver struct */
 static struct i2o_driver i2o_config_driver = {
-	.name = OSM_NAME
+	.name = OSM_NAME,
+	.notify_controller_add = i2o_config_notify_controller_add,
+	.notify_controller_remove = i2o_config_notify_controller_remove
 };
 
 static int i2o_cfg_getiops(unsigned long arg)
diff --git a/drivers/message/i2o/i2o_scsi.c b/drivers/message/i2o/i2o_scsi.c
index af40f1c1ec7..812c29ec86d 100644
--- a/drivers/message/i2o/i2o_scsi.c
+++ b/drivers/message/i2o/i2o_scsi.c
@@ -40,6 +40,7 @@
  *	Fix the resource management problems.
  */
 
+#define DEBUG 1
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/types.h>
@@ -179,6 +180,8 @@ static int i2o_scsi_remove(struct device *dev)
 	struct i2o_scsi_host *i2o_shost;
 	struct scsi_device *scsi_dev;
 
+	osm_info("device removed (TID: %03x)\n", i2o_dev->lct_data.tid);
+
 	i2o_shost = i2o_scsi_get_host(c);
 
 	shost_for_each_device(scsi_dev, i2o_shost->scsi_host)
@@ -262,8 +265,8 @@ static int i2o_scsi_probe(struct device *dev)
 		return -EFAULT;
 	}
 
-	osm_debug("added new SCSI device %03x (cannel: %d, id: %d, lun: %d)\n",
-		  i2o_dev->lct_data.tid, channel, id, (unsigned int)lun);
+	osm_info("device added (TID: %03x) channel: %d, id: %d, lun: %d\n",
+		 i2o_dev->lct_data.tid, channel, id, (unsigned int)lun);
 
 	return 0;
 };
@@ -439,8 +442,6 @@ static int i2o_scsi_reply(struct i2o_controller *c, u32 m,
 
 	cmd->result = DID_OK << 16 | ds;
 
-	cmd->scsi_done(cmd);
-
 	dev = &c->pdev->dev;
 	if (cmd->use_sg)
 		dma_unmap_sg(dev, (struct scatterlist *)cmd->buffer,
@@ -449,6 +450,8 @@ static int i2o_scsi_reply(struct i2o_controller *c, u32 m,
 		dma_unmap_single(dev, (dma_addr_t) ((long)cmd->SCp.ptr),
 				 cmd->request_bufflen, cmd->sc_data_direction);
 
+	cmd->scsi_done(cmd);
+
 	return 1;
 };
 
@@ -502,7 +505,7 @@ static void i2o_scsi_notify_controller_remove(struct i2o_controller *c)
 
 	scsi_remove_host(i2o_shost->scsi_host);
 	scsi_host_put(i2o_shost->scsi_host);
-	pr_info("I2O SCSI host removed\n");
+	osm_debug("I2O SCSI host removed\n");
 };
 
 /* SCSI OSM driver struct */
@@ -545,7 +548,7 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt,
 	u32 scsi_flags, sg_flags;
 	u32 __iomem *mptr;
 	u32 __iomem *lenptr;
-	u32 len, reqlen;
+	u32 len;
 	int i;
 
 	/*
@@ -580,12 +583,12 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt,
 	if (m == I2O_QUEUE_EMPTY)
 		return SCSI_MLQUEUE_HOST_BUSY;
 
+	mptr = &msg->body[0];
+
 	/*
 	 *      Put together a scsi execscb message
 	 */
 
-	len = SCpnt->request_bufflen;
-
 	switch (SCpnt->sc_data_direction) {
 	case PCI_DMA_NONE:
 		scsi_flags = 0x00000000;	// DATA NO XFER
@@ -637,17 +640,13 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt,
 	 */
 
 	/* Direction, disconnect ok, tag, CDBLen */
-	writel(scsi_flags | 0x20200000 | SCpnt->cmd_len, &msg->body[0]);
-
-	mptr = &msg->body[1];
+	writel(scsi_flags | 0x20200000 | SCpnt->cmd_len, mptr ++);
 
 	/* Write SCSI command into the message - always 16 byte block */
 	memcpy_toio(mptr, SCpnt->cmnd, 16);
 	mptr += 4;
 	lenptr = mptr++;	/* Remember me - fill in when we know */
 
-	reqlen = 12;		// SINGLE SGE
-
 	/* Now fill in the SGList and command */
 	if (SCpnt->use_sg) {
 		struct scatterlist *sg;
@@ -671,7 +670,6 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt,
 			sg++;
 		}
 
-		reqlen = mptr - &msg->u.head[0];
 		writel(len, lenptr);
 	} else {
 		len = SCpnt->request_bufflen;
@@ -691,12 +689,11 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt,
 			sg_flags |= 0xC0000000;
 			writel(sg_flags | SCpnt->request_bufflen, mptr++);
 			writel(dma_addr, mptr++);
-		} else
-			reqlen = 9;
+		}
 	}
 
 	/* Stick the headers on */
-	writel(reqlen << 16 | SGL_OFFSET_10, &msg->u.head[0]);
+	writel((mptr - &msg->u.head[0]) << 16 | SGL_OFFSET_10, &msg->u.head[0]);
 
 	/* Queue the message */
 	i2o_msg_post(c, m);
diff --git a/drivers/message/i2o/iop.c b/drivers/message/i2o/iop.c
index 50c8cedf7a2..62b0d8bed18 100644
--- a/drivers/message/i2o/iop.c
+++ b/drivers/message/i2o/iop.c
@@ -68,7 +68,7 @@ extern void i2o_device_exit(void);
  */
 void i2o_msg_nop(struct i2o_controller *c, u32 m)
 {
-	struct i2o_message __iomem *msg = c->in_queue.virt + m;
+	struct i2o_message __iomem *msg = i2o_msg_in_to_virt(c, m);
 
 	writel(THREE_WORD_MSG_SIZE | SGL_OFFSET_0, &msg->u.head[0]);
 	writel(I2O_CMD_UTIL_NOP << 24 | HOST_TID << 12 | ADAPTER_TID,
@@ -452,8 +452,6 @@ static int i2o_iop_clear(struct i2o_controller *c)
 	/* Enable all IOPs */
 	i2o_iop_enable_all();
 
-	i2o_status_get(c);
-
 	return rc;
 }
 
@@ -591,12 +589,11 @@ static int i2o_iop_init_outbound_queue(struct i2o_controller *c)
 	if (m == I2O_QUEUE_EMPTY)
 		return -ETIMEDOUT;
 
-	writel(EIGHT_WORD_MSG_SIZE | TRL_OFFSET_6, &msg->u.head[0]);
+	writel(EIGHT_WORD_MSG_SIZE | SGL_OFFSET_6, &msg->u.head[0]);
 	writel(I2O_CMD_OUTBOUND_INIT << 24 | HOST_TID << 12 | ADAPTER_TID,
 	       &msg->u.head[1]);
 	writel(i2o_exec_driver.context, &msg->u.s.icntxt);
-	writel(0x0106, &msg->u.s.tcntxt);	/* FIXME: why 0x0106, maybe in
-						   Spec? */
+	writel(0x00000000, &msg->u.s.tcntxt);
 	writel(PAGE_SIZE, &msg->body[0]);
 	writel(MSG_FRAME_SIZE << 16 | 0x80, &msg->body[1]);	/* Outbound msg frame
 								   size in words and Initcode */
@@ -891,8 +888,12 @@ void i2o_iop_remove(struct i2o_controller *c)
 	list_for_each_entry_safe(dev, tmp, &c->devices, list)
 	    i2o_device_remove(dev);
 
+	device_del(&c->device);
+
 	/* Ask the IOP to switch to RESET state */
 	i2o_iop_reset(c);
+
+	put_device(&c->device);
 }
 
 /**
@@ -971,8 +972,10 @@ static int i2o_systab_build(void)
 		systab->iops[count].frame_size = sb->inbound_frame_size;
 		systab->iops[count].last_changed = change_ind;
 		systab->iops[count].iop_capabilities = sb->iop_capabilities;
-		systab->iops[count].inbound_low = i2o_ptr_low(c->post_port);
-		systab->iops[count].inbound_high = i2o_ptr_high(c->post_port);
+		systab->iops[count].inbound_low =
+		    i2o_dma_low(c->base.phys + I2O_IN_PORT);
+		systab->iops[count].inbound_high =
+		    i2o_dma_high(c->base.phys + I2O_IN_PORT);
 
 		count++;
 	}
@@ -1109,6 +1112,30 @@ static int i2o_hrt_get(struct i2o_controller *c)
 	return -EBUSY;
 }
 
+/**
+ *	i2o_iop_free - Free the i2o_controller struct
+ *	@c: I2O controller to free
+ */
+void i2o_iop_free(struct i2o_controller *c)
+{
+	kfree(c);
+};
+
+
+/**
+ *	i2o_iop_release - release the memory for a I2O controller
+ *	@dev: I2O controller which should be released
+ *
+ *	Release the allocated memory. This function is called if refcount of
+ *	device reaches 0 automatically.
+ */
+static void i2o_iop_release(struct device *dev)
+{
+	struct i2o_controller *c = to_i2o_controller(dev);
+
+	i2o_iop_free(c);
+};
+
 /**
  *	i2o_iop_alloc - Allocate and initialize a i2o_controller struct
  *
@@ -1137,6 +1164,10 @@ struct i2o_controller *i2o_iop_alloc(void)
 	c->unit = unit++;
 	sprintf(c->name, "iop%d", c->unit);
 
+	device_initialize(&c->device);
+	c->device.release = &i2o_iop_release;
+	snprintf(c->device.bus_id, BUS_ID_SIZE, "iop%d", c->unit);
+
 #if BITS_PER_LONG == 64
 	spin_lock_init(&c->context_list_lock);
 	atomic_set(&c->context_list_counter, 0);
@@ -1146,15 +1177,6 @@ struct i2o_controller *i2o_iop_alloc(void)
 	return c;
 };
 
-/**
- *	i2o_iop_free - Free the i2o_controller struct
- *	@c: I2O controller to free
- */
-void i2o_iop_free(struct i2o_controller *c)
-{
-	kfree(c);
-};
-
 /**
  *	i2o_iop_add - Initialize the I2O controller and add him to the I2O core
  *	@c: controller
@@ -1168,6 +1190,11 @@ int i2o_iop_add(struct i2o_controller *c)
 {
 	int rc;
 
+	if((rc = device_add(&c->device))) {
+		printk(KERN_ERR "%s: could not register controller\n", c->name);
+		goto iop_reset;
+	}
+
 	printk(KERN_INFO "%s: Activating I2O controller...\n", c->name);
 	printk(KERN_INFO "%s: This may take a few minutes if there are many "
 	       "devices\n", c->name);
@@ -1175,30 +1202,23 @@ int i2o_iop_add(struct i2o_controller *c)
 	if ((rc = i2o_iop_activate(c))) {
 		printk(KERN_ERR "%s: could not activate controller\n",
 		       c->name);
-		i2o_iop_reset(c);
-		return rc;
+		goto iop_reset;
 	}
 
 	pr_debug("%s: building sys table...\n", c->name);
 
-	if ((rc = i2o_systab_build())) {
-		i2o_iop_reset(c);
-		return rc;
-	}
+	if ((rc = i2o_systab_build()))
+		goto iop_reset;
 
 	pr_debug("%s: online controller...\n", c->name);
 
-	if ((rc = i2o_iop_online(c))) {
-		i2o_iop_reset(c);
-		return rc;
-	}
+	if ((rc = i2o_iop_online(c)))
+		goto iop_reset;
 
 	pr_debug("%s: getting LCT...\n", c->name);
 
-	if ((rc = i2o_exec_lct_get(c))) {
-		i2o_iop_reset(c);
-		return rc;
-	}
+	if ((rc = i2o_exec_lct_get(c)))
+		goto iop_reset;
 
 	list_add(&c->list, &i2o_controllers);
 
@@ -1207,6 +1227,11 @@ int i2o_iop_add(struct i2o_controller *c)
 	printk(KERN_INFO "%s: Controller added\n", c->name);
 
 	return 0;
+
+iop_reset:
+	i2o_iop_reset(c);
+
+	return rc;
 };
 
 /**
diff --git a/drivers/message/i2o/pci.c b/drivers/message/i2o/pci.c
index 579a8b7a212..f33fd81f77a 100644
--- a/drivers/message/i2o/pci.c
+++ b/drivers/message/i2o/pci.c
@@ -38,8 +38,7 @@ extern void i2o_iop_free(struct i2o_controller *);
 extern int i2o_iop_add(struct i2o_controller *);
 extern void i2o_iop_remove(struct i2o_controller *);
 
-extern int i2o_driver_dispatch(struct i2o_controller *, u32,
-			       struct i2o_message *);
+extern int i2o_driver_dispatch(struct i2o_controller *, u32);
 
 /* PCI device id table for all I2O controllers */
 static struct pci_device_id __devinitdata i2o_pci_ids[] = {
@@ -89,8 +88,7 @@ static void i2o_pci_free(struct i2o_controller *c)
 
 	i2o_dma_free(dev, &c->out_queue);
 	i2o_dma_free(dev, &c->status_block);
-	if (c->lct)
-		kfree(c->lct);
+	kfree(c->lct);
 	i2o_dma_free(dev, &c->dlct);
 	i2o_dma_free(dev, &c->hrt);
 	i2o_dma_free(dev, &c->status);
@@ -187,9 +185,9 @@ static int __devinit i2o_pci_alloc(struct i2o_controller *c)
 	} else
 		c->in_queue = c->base;
 
-	c->irq_mask = c->base.virt + 0x34;
-	c->post_port = c->base.virt + 0x40;
-	c->reply_port = c->base.virt + 0x44;
+	c->irq_mask = c->base.virt + I2O_IRQ_MASK;
+	c->in_port = c->base.virt + I2O_IN_PORT;
+	c->out_port = c->base.virt + I2O_OUT_PORT;
 
 	if (i2o_dma_alloc(dev, &c->status, 8, GFP_KERNEL)) {
 		i2o_pci_free(c);
@@ -235,49 +233,34 @@ static irqreturn_t i2o_pci_interrupt(int irq, void *dev_id, struct pt_regs *r)
 {
 	struct i2o_controller *c = dev_id;
 	struct device *dev = &c->pdev->dev;
-	struct i2o_message *m;
-	u32 mv;
+	u32 mv = readl(c->out_port);
 
 	/*
 	 * Old 960 steppings had a bug in the I2O unit that caused
 	 * the queue to appear empty when it wasn't.
 	 */
-	mv = I2O_REPLY_READ32(c);
 	if (mv == I2O_QUEUE_EMPTY) {
-		mv = I2O_REPLY_READ32(c);
-		if (unlikely(mv == I2O_QUEUE_EMPTY)) {
+		mv = readl(c->out_port);
+		if (unlikely(mv == I2O_QUEUE_EMPTY))
 			return IRQ_NONE;
-		} else
+		else
 			pr_debug("%s: 960 bug detected\n", c->name);
 	}
 
 	while (mv != I2O_QUEUE_EMPTY) {
-		/*
-		 * Map the message from the page frame map to kernel virtual.
-		 * Because bus_to_virt is deprecated, we have calculate the
-		 * location by ourself!
-		 */
-		m = i2o_msg_out_to_virt(c, mv);
-
-		/*
-		 *      Ensure this message is seen coherently but cachably by
-		 *      the processor
-		 */
-		dma_sync_single_for_cpu(dev, mv, MSG_FRAME_SIZE * 4,
-					PCI_DMA_FROMDEVICE);
-
 		/* dispatch it */
-		if (i2o_driver_dispatch(c, mv, m))
+		if (i2o_driver_dispatch(c, mv))
 			/* flush it if result != 0 */
 			i2o_flush_reply(c, mv);
 
 		/*
 		 * That 960 bug again...
 		 */
-		mv = I2O_REPLY_READ32(c);
+		mv = readl(c->out_port);
 		if (mv == I2O_QUEUE_EMPTY)
-			mv = I2O_REPLY_READ32(c);
+			mv = readl(c->out_port);
 	}
+
 	return IRQ_HANDLED;
 }
 
@@ -294,7 +277,9 @@ static int i2o_pci_irq_enable(struct i2o_controller *c)
 	struct pci_dev *pdev = c->pdev;
 	int rc;
 
-	I2O_IRQ_WRITE32(c, 0xffffffff);
+	wmb();
+	writel(0xffffffff, c->irq_mask);
+	wmb();
 
 	if (pdev->irq) {
 		rc = request_irq(pdev->irq, i2o_pci_interrupt, SA_SHIRQ,
@@ -306,7 +291,8 @@ static int i2o_pci_irq_enable(struct i2o_controller *c)
 		}
 	}
 
-	I2O_IRQ_WRITE32(c, 0x00000000);
+	writel(0x00000000, c->irq_mask);
+	wmb();
 
 	printk(KERN_INFO "%s: Installed at IRQ %d\n", c->name, pdev->irq);
 
@@ -321,7 +307,9 @@ static int i2o_pci_irq_enable(struct i2o_controller *c)
  */
 static void i2o_pci_irq_disable(struct i2o_controller *c)
 {
-	I2O_IRQ_WRITE32(c, 0xffffffff);
+	wmb();
+	writel(0xffffffff, c->irq_mask);
+	wmb();
 
 	if (c->pdev->irq > 0)
 		free_irq(c->pdev->irq, c);
@@ -379,7 +367,7 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev,
 		       pci_name(pdev));
 
 	c->pdev = pdev;
-	c->device = pdev->dev;
+	c->device.parent = get_device(&pdev->dev);
 
 	/* Cards that fall apart if you hit them with large I/O loads... */
 	if (pdev->vendor == PCI_VENDOR_ID_NCR && pdev->device == 0x0630) {
@@ -428,6 +416,8 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev,
 	if (i960)
 		pci_write_config_word(i960, 0x42, 0x03ff);
 
+	get_device(&c->device);
+
 	return 0;
 
       uninstall:
@@ -438,6 +428,7 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev,
 
       free_controller:
 	i2o_iop_free(c);
+	put_device(c->device.parent);
 
       disable:
 	pci_disable_device(pdev);
@@ -461,15 +452,17 @@ static void __devexit i2o_pci_remove(struct pci_dev *pdev)
 	i2o_pci_irq_disable(c);
 	i2o_pci_free(c);
 
+	pci_disable_device(pdev);
+
 	printk(KERN_INFO "%s: Controller removed.\n", c->name);
 
-	i2o_iop_free(c);
-	pci_disable_device(pdev);
+	put_device(c->device.parent);
+	put_device(&c->device);
 };
 
 /* PCI driver for I2O controller */
 static struct pci_driver i2o_pci_driver = {
-	.name = "I2O controller",
+	.name = "PCI_I2O",
 	.id_table = i2o_pci_ids,
 	.probe = i2o_pci_probe,
 	.remove = __devexit_p(i2o_pci_remove),
diff --git a/include/linux/i2o.h b/include/linux/i2o.h
index 40e45a83d3f..e8cd1129001 100644
--- a/include/linux/i2o.h
+++ b/include/linux/i2o.h
@@ -153,12 +153,10 @@ struct i2o_controller {
 	unsigned int promise:1;		/* Promise controller */
 
 	struct list_head devices;	/* list of I2O devices */
-
-	struct notifier_block *event_notifer;	/* Events */
-	atomic_t users;
 	struct list_head list;	/* Controller list */
-	void __iomem *post_port;	/* Inbout port address */
-	void __iomem *reply_port;	/* Outbound port address */
+
+	void __iomem *in_port;	/* Inbout port address */
+	void __iomem *out_port;	/* Outbound port address */
 	void __iomem *irq_mask;		/* Interrupt register address */
 
 	/* Dynamic LCT related data */
@@ -182,9 +180,6 @@ struct i2o_controller {
 	struct resource io_resource;	/* I/O resource allocated to the IOP */
 	struct resource mem_resource;	/* Mem resource allocated to the IOP */
 
-	struct proc_dir_entry *proc_entry;	/* /proc dir */
-
-	struct list_head bus_list;	/* list of busses on IOP */
 	struct device device;
 	struct i2o_device *exec;	/* Executive */
 #if BITS_PER_LONG == 64
@@ -380,49 +375,10 @@ extern int i2o_device_claim_release(struct i2o_device *);
 /* Exec OSM functions */
 extern int i2o_exec_lct_get(struct i2o_controller *);
 
-/* device to i2o_device and driver to i2o_driver convertion functions */
+/* device / driver conversion functions */
 #define to_i2o_driver(drv) container_of(drv,struct i2o_driver, driver)
 #define to_i2o_device(dev) container_of(dev, struct i2o_device, device)
-
-/*
- *	Messenger inlines
- */
-static inline u32 I2O_POST_READ32(struct i2o_controller *c)
-{
-	rmb();
-	return readl(c->post_port);
-};
-
-static inline void I2O_POST_WRITE32(struct i2o_controller *c, u32 val)
-{
-	wmb();
-	writel(val, c->post_port);
-};
-
-static inline u32 I2O_REPLY_READ32(struct i2o_controller *c)
-{
-	rmb();
-	return readl(c->reply_port);
-};
-
-static inline void I2O_REPLY_WRITE32(struct i2o_controller *c, u32 val)
-{
-	wmb();
-	writel(val, c->reply_port);
-};
-
-static inline u32 I2O_IRQ_READ32(struct i2o_controller *c)
-{
-	rmb();
-	return readl(c->irq_mask);
-};
-
-static inline void I2O_IRQ_WRITE32(struct i2o_controller *c, u32 val)
-{
-	wmb();
-	writel(val, c->irq_mask);
-	wmb();
-};
+#define to_i2o_controller(dev) container_of(dev, struct i2o_controller, device)
 
 /**
  *	i2o_msg_get - obtain an I2O message from the IOP
@@ -440,10 +396,12 @@ static inline void I2O_IRQ_WRITE32(struct i2o_controller *c, u32 val)
 static inline u32 i2o_msg_get(struct i2o_controller *c,
 			      struct i2o_message __iomem **msg)
 {
-	u32 m;
+	u32 m = readl(c->in_port);
 
-	if ((m = I2O_POST_READ32(c)) != I2O_QUEUE_EMPTY)
+	if (m != I2O_QUEUE_EMPTY) {
 		*msg = c->in_queue.virt + m;
+		rmb();
+	}
 
 	return m;
 };
@@ -457,7 +415,8 @@ static inline u32 i2o_msg_get(struct i2o_controller *c,
  */
 static inline void i2o_msg_post(struct i2o_controller *c, u32 m)
 {
-	I2O_POST_WRITE32(c, m);
+	wmb();
+	writel(m, c->in_port);
 };
 
 /**
@@ -486,12 +445,10 @@ static inline int i2o_msg_post_wait(struct i2o_controller *c, u32 m,
  *	The I2O controller must be informed that the reply message is not needed
  *	anymore. If you forget to flush the reply, the message frame can't be
  *	used by the controller anymore and is therefore lost.
- *
- *	FIXME: is there a timeout after which the controller reuse the message?
  */
 static inline void i2o_flush_reply(struct i2o_controller *c, u32 m)
 {
-	I2O_REPLY_WRITE32(c, m);
+	writel(m, c->out_port);
 };
 
 /**
@@ -505,8 +462,9 @@ static inline void i2o_flush_reply(struct i2o_controller *c, u32 m)
  *	work for sender side messages as they are ioremap objects
  *	provided by the I2O controller.
  */
-static inline struct i2o_message *i2o_msg_out_to_virt(struct i2o_controller *c,
-						      u32 m)
+static inline struct i2o_message __iomem *i2o_msg_out_to_virt(struct
+							      i2o_controller *c,
+							      u32 m)
 {
 	BUG_ON(m < c->out_queue.phys
 	       || m >= c->out_queue.phys + c->out_queue.len);
@@ -917,7 +875,7 @@ extern void i2o_debug_state(struct i2o_controller *c);
 #define I2OVER15	0x0001
 #define I2OVER20	0x0002
 
-/* Default is 1.5, FIXME: Need support for both 1.5 and 2.0 */
+/* Default is 1.5 */
 #define I2OVERSION	I2OVER15
 
 #define SGL_OFFSET_0    I2OVERSION
-- 
cgit v1.2.3-70-g09d2


From f10378fff658f61307496e0ae00095041725cf07 Mon Sep 17 00:00:00 2001
From: Markus Lidel <Markus.Lidel@shadowconnect.com>
Date: Thu, 23 Jun 2005 22:02:16 -0700
Subject: [PATCH] I2O: new sysfs attributes and Adaptec specific block device
 access and 64-bit DMA support

Changes:
 - Added Bus-OSM which could be used by user space programs to reset a
   channel on the controller
 - Make ioctl's in Config-OSM obsolete in prefer for sysfs attributes and
   move those to its own file
 - Added sysfs attribute for firmware read and write access for I2O
   controllers
 - Added special handling of firmware read and write access for Adaptec
   controllers
 - Added vendor id and product id as sysfs-attribute to Executive classes
 - Added automatic notification of LCT change handling to Exec-OSM
 - Added flushing function to Block-OSM for later barrier implementation
 - Use PRIVATE messages for Block access on Adaptec controllers, which are
   faster then BLOCK class access
 - Cleaned up support for Promise controller
 - New messages are now detected using the IRQ status register as
   suggested by the I2O spec
 - Added i2o_dma_high() and i2o_dma_low() functions
 - Added facility for SG tablesize calculation when using 32-bit and
   64-bit DMA addresses
 - Added i2o_dma_map_single() and i2o_dma_map_sg() which could build the
   SG list for 32-bit as well as 64-bit DMA addresses

Signed-off-by: Markus Lidel <Markus.Lidel@shadowconnect.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/message/i2o/Kconfig      |  18 ++
 drivers/message/i2o/Makefile     |   3 +
 drivers/message/i2o/bus-osm.c    | 164 +++++++++++
 drivers/message/i2o/config-osm.c | 579 +++++++++++++++++++++++++++++++++++++++
 drivers/message/i2o/driver.c     |  12 +-
 drivers/message/i2o/exec-osm.c   |  74 ++++-
 drivers/message/i2o/i2o_block.c  | 277 ++++++++++++-------
 drivers/message/i2o/i2o_block.h  |   4 +-
 drivers/message/i2o/i2o_config.c | 156 +----------
 drivers/message/i2o/i2o_proc.c   |   4 +-
 drivers/message/i2o/i2o_scsi.c   |  30 +-
 drivers/message/i2o/iop.c        | 263 ++++++++----------
 drivers/message/i2o/pci.c        |  67 ++---
 include/linux/i2o-dev.h          |   6 +-
 include/linux/i2o.h              | 321 ++++++++++++++++++----
 15 files changed, 1446 insertions(+), 532 deletions(-)
 create mode 100644 drivers/message/i2o/bus-osm.c
 create mode 100644 drivers/message/i2o/config-osm.c

(limited to 'include')

diff --git a/drivers/message/i2o/Kconfig b/drivers/message/i2o/Kconfig
index 8d132b0d6b1..ce278e060ac 100644
--- a/drivers/message/i2o/Kconfig
+++ b/drivers/message/i2o/Kconfig
@@ -35,6 +35,24 @@ config I2O_CONFIG
 	  To compile this support as a module, choose M here: the
 	  module will be called i2o_config.
 
+config I2O_CONFIG_OLD_IOCTL
+	bool "Enable ioctls (OBSOLETE)"
+	depends on I2O_CONFIG
+	default y
+	---help---
+	  Enables old ioctls.
+
+config I2O_BUS
+	tristate "I2O Bus Adapter OSM"
+	depends on I2O
+	---help---
+	  Include support for the I2O Bus Adapter OSM. The Bus Adapter OSM
+	  provides access to the busses on the I2O controller. The main purpose
+	  is to rescan the bus to find new devices.
+
+	  To compile this support as a module, choose M here: the
+	  module will be called i2o_bus.
+
 config I2O_BLOCK
 	tristate "I2O Block OSM"
 	depends on I2O
diff --git a/drivers/message/i2o/Makefile b/drivers/message/i2o/Makefile
index aabc6cdc3fc..2c2e39aa1ef 100644
--- a/drivers/message/i2o/Makefile
+++ b/drivers/message/i2o/Makefile
@@ -6,8 +6,11 @@
 #
 
 i2o_core-y		+= iop.o driver.o device.o debug.o pci.o exec-osm.o
+i2o_bus-y		+= bus-osm.o
+i2o_config-y		+= config-osm.o
 obj-$(CONFIG_I2O)	+= i2o_core.o
 obj-$(CONFIG_I2O_CONFIG)+= i2o_config.o
+obj-$(CONFIG_I2O_BUS)	+= i2o_bus.o
 obj-$(CONFIG_I2O_BLOCK)	+= i2o_block.o
 obj-$(CONFIG_I2O_SCSI)	+= i2o_scsi.o
 obj-$(CONFIG_I2O_PROC)	+= i2o_proc.o
diff --git a/drivers/message/i2o/bus-osm.c b/drivers/message/i2o/bus-osm.c
new file mode 100644
index 00000000000..d43c35894ae
--- /dev/null
+++ b/drivers/message/i2o/bus-osm.c
@@ -0,0 +1,164 @@
+/*
+ *	Bus Adapter OSM
+ *
+ *	Copyright (C) 2005	Markus Lidel <Markus.Lidel@shadowconnect.com>
+ *
+ *	This program is free software; you can redistribute it and/or modify it
+ *	under the terms of the GNU General Public License as published by the
+ *	Free Software Foundation; either version 2 of the License, or (at your
+ *	option) any later version.
+ *
+ *	Fixes/additions:
+ *		Markus Lidel <Markus.Lidel@shadowconnect.com>
+ *			initial version.
+ */
+
+#include <linux/module.h>
+#include <linux/i2o.h>
+
+#define OSM_NAME	"bus-osm"
+#define OSM_VERSION	"$Rev$"
+#define OSM_DESCRIPTION	"I2O Bus Adapter OSM"
+
+static struct i2o_driver i2o_bus_driver;
+
+/* Bus OSM class handling definition */
+static struct i2o_class_id i2o_bus_class_id[] = {
+	{I2O_CLASS_BUS_ADAPTER},
+	{I2O_CLASS_END}
+};
+
+/**
+ *	i2o_bus_scan - Scan the bus for new devices
+ *	@dev: I2O device of the bus, which should be scanned
+ *
+ *	Scans the bus dev for new / removed devices. After the scan a new LCT
+ *	will be fetched automatically.
+ *
+ *	Returns 0 on success or negative error code on failure.
+ */
+static int i2o_bus_scan(struct i2o_device *dev)
+{
+	struct i2o_message __iomem *msg;
+	u32 m;
+
+	m = i2o_msg_get_wait(dev->iop, &msg, I2O_TIMEOUT_MESSAGE_GET);
+	if (m == I2O_QUEUE_EMPTY)
+		return -ETIMEDOUT;
+
+	writel(FIVE_WORD_MSG_SIZE | SGL_OFFSET_0, &msg->u.head[0]);
+	writel(I2O_CMD_BUS_SCAN << 24 | HOST_TID << 12 | dev->lct_data.tid,
+	       &msg->u.head[1]);
+
+	return i2o_msg_post_wait(dev->iop, m, 60);
+};
+
+/**
+ *	i2o_bus_store_scan - Scan the I2O Bus Adapter
+ *	@d: device which should be scanned
+ *
+ *	Returns count.
+ */
+static ssize_t i2o_bus_store_scan(struct device *d, const char *buf,
+				  size_t count)
+{
+	struct i2o_device *i2o_dev = to_i2o_device(d);
+	int rc;
+
+	if ((rc = i2o_bus_scan(i2o_dev)))
+		osm_warn("bus scan failed %d\n", rc);
+
+	return count;
+}
+
+/* Bus Adapter OSM device attributes */
+static DEVICE_ATTR(scan, S_IWUSR, NULL, i2o_bus_store_scan);
+
+/**
+ *	i2o_bus_probe - verify if dev is a I2O Bus Adapter device and install it
+ *	@dev: device to verify if it is a I2O Bus Adapter device
+ *
+ *	Because we want all Bus Adapters always return 0.
+ *
+ *	Returns 0.
+ */
+static int i2o_bus_probe(struct device *dev)
+{
+	struct i2o_device *i2o_dev = to_i2o_device(get_device(dev));
+
+	device_create_file(dev, &dev_attr_scan);
+
+	osm_info("device added (TID: %03x)\n", i2o_dev->lct_data.tid);
+
+	return 0;
+};
+
+/**
+ *	i2o_bus_remove - remove the I2O Bus Adapter device from the system again
+ *	@dev: I2O Bus Adapter device which should be removed
+ *
+ *	Always returns 0.
+ */
+static int i2o_bus_remove(struct device *dev)
+{
+	struct i2o_device *i2o_dev = to_i2o_device(dev);
+
+	device_remove_file(dev, &dev_attr_scan);
+
+	put_device(dev);
+
+	osm_info("device removed (TID: %03x)\n", i2o_dev->lct_data.tid);
+
+	return 0;
+};
+
+/* Bus Adapter OSM driver struct */
+static struct i2o_driver i2o_bus_driver = {
+	.name = OSM_NAME,
+	.classes = i2o_bus_class_id,
+	.driver = {
+		   .probe = i2o_bus_probe,
+		   .remove = i2o_bus_remove,
+		   },
+};
+
+/**
+ *	i2o_bus_init - Bus Adapter OSM initialization function
+ *
+ *	Only register the Bus Adapter OSM in the I2O core.
+ *
+ *	Returns 0 on success or negative error code on failure.
+ */
+static int __init i2o_bus_init(void)
+{
+	int rc;
+
+	printk(KERN_INFO OSM_DESCRIPTION " v" OSM_VERSION "\n");
+
+	/* Register Bus Adapter OSM into I2O core */
+	rc = i2o_driver_register(&i2o_bus_driver);
+	if (rc) {
+		osm_err("Could not register Bus Adapter OSM\n");
+		return rc;
+	}
+
+	return 0;
+};
+
+/**
+ *	i2o_bus_exit - Bus Adapter OSM exit function
+ *
+ *	Unregisters Bus Adapter OSM from I2O core.
+ */
+static void __exit i2o_bus_exit(void)
+{
+	i2o_driver_unregister(&i2o_bus_driver);
+};
+
+MODULE_AUTHOR("Markus Lidel <Markus.Lidel@shadowconnect.com>");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION(OSM_DESCRIPTION);
+MODULE_VERSION(OSM_VERSION);
+
+module_init(i2o_bus_init);
+module_exit(i2o_bus_exit);
diff --git a/drivers/message/i2o/config-osm.c b/drivers/message/i2o/config-osm.c
new file mode 100644
index 00000000000..d0267609a94
--- /dev/null
+++ b/drivers/message/i2o/config-osm.c
@@ -0,0 +1,579 @@
+/*
+ *	Configuration OSM
+ *
+ *	Copyright (C) 2005	Markus Lidel <Markus.Lidel@shadowconnect.com>
+ *
+ *	This program is free software; you can redistribute it and/or modify it
+ *	under the terms of the GNU General Public License as published by the
+ *	Free Software Foundation; either version 2 of the License, or (at your
+ *	option) any later version.
+ *
+ *	Fixes/additions:
+ *		Markus Lidel <Markus.Lidel@shadowconnect.com>
+ *			initial version.
+ */
+
+#include <linux/module.h>
+#include <linux/i2o.h>
+#include <linux/namei.h>
+
+#include <asm/uaccess.h>
+
+#define OSM_NAME	"config-osm"
+#define OSM_VERSION	"1.248"
+#define OSM_DESCRIPTION	"I2O Configuration OSM"
+
+/* access mode user rw */
+#define S_IWRSR (S_IRUSR | S_IWUSR)
+
+static struct i2o_driver i2o_config_driver;
+
+/* Special file operations for sysfs */
+struct fops_attribute {
+	struct bin_attribute bin;
+	struct file_operations fops;
+};
+
+/**
+ *	sysfs_read_dummy
+ */
+static ssize_t sysfs_read_dummy(struct kobject *kobj, char *buf, loff_t offset,
+				size_t count)
+{
+	return 0;
+};
+
+/**
+ *	sysfs_write_dummy
+ */
+static ssize_t sysfs_write_dummy(struct kobject *kobj, char *buf, loff_t offset,
+				 size_t count)
+{
+	return 0;
+};
+
+/**
+ *	sysfs_create_fops_file - Creates attribute with special file operations
+ *	@kobj: kobject which should contains the attribute
+ *	@attr: attributes which should be used to create file
+ *
+ *	First creates attribute @attr in kobject @kobj. If it is the first time
+ *	this function is called, merge old fops from sysfs with new one and
+ *	write it back. Afterwords the new fops will be set for the created
+ *	attribute.
+ *
+ *	Returns 0 on success or negative error code on failure.
+ */
+static int sysfs_create_fops_file(struct kobject *kobj,
+				  struct fops_attribute *attr)
+{
+	struct file_operations tmp, *fops;
+	struct dentry *d;
+	struct qstr qstr;
+	int rc;
+
+	fops = &attr->fops;
+
+	if (fops->read)
+		attr->bin.read = sysfs_read_dummy;
+
+	if (fops->write)
+		attr->bin.write = sysfs_write_dummy;
+
+	if ((rc = sysfs_create_bin_file(kobj, &attr->bin)))
+		return rc;
+
+	qstr.name = attr->bin.attr.name;
+	qstr.len = strlen(qstr.name);
+	qstr.hash = full_name_hash(qstr.name, qstr.len);
+
+	if ((d = lookup_hash(&qstr, kobj->dentry))) {
+		if (!fops->owner) {
+			memcpy(&tmp, d->d_inode->i_fop, sizeof(tmp));
+			if (fops->read)
+				tmp.read = fops->read;
+			if (fops->write)
+				tmp.write = fops->write;
+			memcpy(fops, &tmp, sizeof(tmp));
+		}
+
+		d->d_inode->i_fop = fops;
+	} else
+		sysfs_remove_bin_file(kobj, &attr->bin);
+
+	return -ENOENT;
+};
+
+/**
+ *	sysfs_remove_fops_file - Remove attribute with special file operations
+ *	@kobj: kobject which contains the attribute
+ *	@attr: attributes which are used to create file
+ *
+ *	Only wrapper arround sysfs_remove_bin_file()
+ *
+ *	Returns 0 on success or negative error code on failure.
+ */
+static inline int sysfs_remove_fops_file(struct kobject *kobj,
+					 struct fops_attribute *attr)
+{
+	return sysfs_remove_bin_file(kobj, &attr->bin);
+};
+
+/**
+ *	i2o_config_read_hrt - Returns the HRT of the controller
+ *	@kob: kernel object handle
+ *	@buf: buffer into which the HRT should be copied
+ *	@off: file offset
+ *	@count: number of bytes to read
+ *
+ *	Put @count bytes starting at @off into @buf from the HRT of the I2O
+ *	controller corresponding to @kobj.
+ *
+ *	Returns number of bytes copied into buffer.
+ */
+static ssize_t i2o_config_read_hrt(struct kobject *kobj, char *buf,
+				   loff_t offset, size_t count)
+{
+	struct i2o_controller *c = kobj_to_i2o_device(kobj)->iop;
+	i2o_hrt *hrt = c->hrt.virt;
+
+	u32 size = (hrt->num_entries * hrt->entry_len + 2) * 4;
+
+	if (offset > size)
+		return 0;
+
+	if (offset + count > size)
+		count = size - offset;
+
+	memcpy(buf, (u8 *) hrt + offset, count);
+
+	return count;
+};
+
+/**
+ *	i2o_config_read_lct - Returns the LCT of the controller
+ *	@kob: kernel object handle
+ *	@buf: buffer into which the LCT should be copied
+ *	@off: file offset
+ *	@count: number of bytes to read
+ *
+ *	Put @count bytes starting at @off into @buf from the LCT of the I2O
+ *	controller corresponding to @kobj.
+ *
+ *	Returns number of bytes copied into buffer.
+ */
+static ssize_t i2o_config_read_lct(struct kobject *kobj, char *buf,
+				   loff_t offset, size_t count)
+{
+	struct i2o_controller *c = kobj_to_i2o_device(kobj)->iop;
+	u32 size = c->lct->table_size * 4;
+
+	if (offset > size)
+		return 0;
+
+	if (offset + count > size)
+		count = size - offset;
+
+	memcpy(buf, (u8 *) c->lct + offset, count);
+
+	return count;
+};
+
+#define I2O_CONFIG_SW_ATTR(_name,_mode,_type,_swid) \
+static ssize_t i2o_config_##_name##_read(struct file *file, char __user *buf, size_t count, loff_t * offset) { \
+	return i2o_config_sw_read(file, buf, count, offset, _type, _swid); \
+};\
+\
+static ssize_t i2o_config_##_name##_write(struct file *file, const char __user *buf, size_t count, loff_t * offset) { \
+	return i2o_config_sw_write(file, buf, count, offset, _type, _swid); \
+}; \
+\
+static struct fops_attribute i2o_config_attr_##_name = { \
+	.bin = { .attr = { .name = __stringify(_name), .mode = _mode, \
+			   .owner = THIS_MODULE }, \
+		 .size = 0, }, \
+	.fops = { .write = i2o_config_##_name##_write, \
+		  .read = i2o_config_##_name##_read} \
+};
+
+#ifdef CONFIG_I2O_EXT_ADAPTEC
+
+/**
+ *	i2o_config_dpt_reagion - Converts type and id to flash region
+ *	@swtype: type of software module reading
+ *	@swid: id of software which should be read
+ *
+ *	Converts type and id from I2O spec to the matching region for DPT /
+ *	Adaptec controllers.
+ *
+ *	Returns region which match type and id or -1 on error.
+ */
+static u32 i2o_config_dpt_region(u8 swtype, u8 swid)
+{
+	switch (swtype) {
+	case I2O_SOFTWARE_MODULE_IRTOS:
+		/*
+		 * content: operation firmware
+		 * region size:
+		 *      0xbc000 for 2554, 3754, 2564, 3757
+		 *      0x170000 for 2865
+		 *      0x17c000 for 3966
+		 */
+		if (!swid)
+			return 0;
+
+		break;
+
+	case I2O_SOFTWARE_MODULE_IOP_PRIVATE:
+		/*
+		 * content: BIOS and SMOR
+		 * BIOS size: first 0x8000 bytes
+		 * region size:
+		 *      0x40000 for 2554, 3754, 2564, 3757
+		 *      0x80000 for 2865, 3966
+		 */
+		if (!swid)
+			return 1;
+
+		break;
+
+	case I2O_SOFTWARE_MODULE_IOP_CONFIG:
+		switch (swid) {
+		case 0:
+			/*
+			 * content: NVRAM defaults
+			 * region size: 0x2000 bytes
+			 */
+			return 2;
+		case 1:
+			/*
+			 * content: serial number
+			 * region size: 0x2000 bytes
+			 */
+			return 3;
+		}
+		break;
+	}
+
+	return -1;
+};
+
+#endif
+
+/**
+ *	i2o_config_sw_read - Read a software module from controller
+ *	@file: file pointer
+ *	@buf: buffer into which the data should be copied
+ *	@count: number of bytes to read
+ *	@off: file offset
+ *	@swtype: type of software module reading
+ *	@swid: id of software which should be read
+ *
+ *	Transfers @count bytes at offset @offset from IOP into buffer using
+ *	type @swtype and id @swid as described in I2O spec.
+ *
+ *	Returns number of bytes copied into buffer or error code on failure.
+ */
+static ssize_t i2o_config_sw_read(struct file *file, char __user * buf,
+				  size_t count, loff_t * offset, u8 swtype,
+				  u32 swid)
+{
+	struct sysfs_dirent *sd = file->f_dentry->d_parent->d_fsdata;
+	struct kobject *kobj = sd->s_element;
+	struct i2o_controller *c = kobj_to_i2o_device(kobj)->iop;
+	u32 m, function = I2O_CMD_SW_UPLOAD;
+	struct i2o_dma buffer;
+	struct i2o_message __iomem *msg;
+	u32 __iomem *mptr;
+	int rc, status;
+
+	m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET);
+	if (m == I2O_QUEUE_EMPTY)
+		return -EBUSY;
+
+	mptr = &msg->body[3];
+
+	if ((rc = i2o_dma_alloc(&c->pdev->dev, &buffer, count, GFP_KERNEL))) {
+		i2o_msg_nop(c, m);
+		return rc;
+	}
+#ifdef CONFIG_I2O_EXT_ADAPTEC
+	if (c->adaptec) {
+		mptr = &msg->body[4];
+		function = I2O_CMD_PRIVATE;
+
+		writel(TEN_WORD_MSG_SIZE | SGL_OFFSET_8, &msg->u.head[0]);
+
+		writel(I2O_VENDOR_DPT << 16 | I2O_DPT_FLASH_READ,
+		       &msg->body[0]);
+		writel(i2o_config_dpt_region(swtype, swid), &msg->body[1]);
+		writel(*offset, &msg->body[2]);
+		writel(count, &msg->body[3]);
+	} else
+#endif
+		writel(NINE_WORD_MSG_SIZE | SGL_OFFSET_7, &msg->u.head[0]);
+
+	writel(0xD0000000 | count, mptr++);
+	writel(buffer.phys, mptr);
+
+	writel(function << 24 | HOST_TID << 12 | ADAPTER_TID, &msg->u.head[1]);
+	writel(i2o_config_driver.context, &msg->u.head[2]);
+	writel(0, &msg->u.head[3]);
+
+#ifdef CONFIG_I2O_EXT_ADAPTEC
+	if (!c->adaptec)
+#endif
+	{
+		writel((u32) swtype << 16 | (u32) 1 << 8, &msg->body[0]);
+		writel(0, &msg->body[1]);
+		writel(swid, &msg->body[2]);
+	}
+
+	status = i2o_msg_post_wait_mem(c, m, 60, &buffer);
+
+	if (status == I2O_POST_WAIT_OK) {
+		if (!(rc = copy_to_user(buf, buffer.virt, count))) {
+			rc = count;
+			*offset += count;
+		}
+	} else
+		rc = -EIO;
+
+	if (status != -ETIMEDOUT)
+		i2o_dma_free(&c->pdev->dev, &buffer);
+
+	return rc;
+};
+
+/**
+ *	i2o_config_sw_write - Write a software module to controller
+ *	@file: file pointer
+ *	@buf: buffer into which the data should be copied
+ *	@count: number of bytes to read
+ *	@off: file offset
+ *	@swtype: type of software module writing
+ *	@swid: id of software which should be written
+ *
+ *	Transfers @count bytes at offset @offset from buffer to IOP using
+ *	type @swtype and id @swid as described in I2O spec.
+ *
+ *	Returns number of bytes copied from buffer or error code on failure.
+ */
+static ssize_t i2o_config_sw_write(struct file *file, const char __user * buf,
+				   size_t count, loff_t * offset, u8 swtype,
+				   u32 swid)
+{
+	struct sysfs_dirent *sd = file->f_dentry->d_parent->d_fsdata;
+	struct kobject *kobj = sd->s_element;
+	struct i2o_controller *c = kobj_to_i2o_device(kobj)->iop;
+	u32 m, function = I2O_CMD_SW_DOWNLOAD;
+	struct i2o_dma buffer;
+	struct i2o_message __iomem *msg;
+	u32 __iomem *mptr;
+	int rc, status;
+
+	m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET);
+	if (m == I2O_QUEUE_EMPTY)
+		return -EBUSY;
+
+	mptr = &msg->body[3];
+
+	if ((rc = i2o_dma_alloc(&c->pdev->dev, &buffer, count, GFP_KERNEL)))
+		goto nop_msg;
+
+	if ((rc = copy_from_user(buffer.virt, buf, count)))
+		goto free_buffer;
+
+#ifdef CONFIG_I2O_EXT_ADAPTEC
+	if (c->adaptec) {
+		mptr = &msg->body[4];
+		function = I2O_CMD_PRIVATE;
+
+		writel(TEN_WORD_MSG_SIZE | SGL_OFFSET_8, &msg->u.head[0]);
+
+		writel(I2O_VENDOR_DPT << 16 | I2O_DPT_FLASH_WRITE,
+		       &msg->body[0]);
+		writel(i2o_config_dpt_region(swtype, swid), &msg->body[1]);
+		writel(*offset, &msg->body[2]);
+		writel(count, &msg->body[3]);
+	} else
+#endif
+		writel(NINE_WORD_MSG_SIZE | SGL_OFFSET_7, &msg->u.head[0]);
+
+	writel(0xD4000000 | count, mptr++);
+	writel(buffer.phys, mptr);
+
+	writel(function << 24 | HOST_TID << 12 | ADAPTER_TID, &msg->u.head[1]);
+	writel(i2o_config_driver.context, &msg->u.head[2]);
+	writel(0, &msg->u.head[3]);
+
+#ifdef CONFIG_I2O_EXT_ADAPTEC
+	if (!c->adaptec)
+#endif
+	{
+		writel((u32) swtype << 16 | (u32) 1 << 8, &msg->body[0]);
+		writel(0, &msg->body[1]);
+		writel(swid, &msg->body[2]);
+	}
+
+	status = i2o_msg_post_wait_mem(c, m, 60, &buffer);
+
+	if (status != -ETIMEDOUT)
+		i2o_dma_free(&c->pdev->dev, &buffer);
+
+	if (status != I2O_POST_WAIT_OK)
+		return -EIO;
+
+	*offset += count;
+
+	return count;
+
+      free_buffer:
+	i2o_dma_free(&c->pdev->dev, &buffer);
+
+      nop_msg:
+	i2o_msg_nop(c, m);
+
+	return rc;
+};
+
+/* attribute for HRT in sysfs */
+static struct bin_attribute i2o_config_hrt_attr = {
+	.attr = {
+		 .name = "hrt",
+		 .mode = S_IRUGO,
+		 .owner = THIS_MODULE},
+	.size = 0,
+	.read = i2o_config_read_hrt
+};
+
+/* attribute for LCT in sysfs */
+static struct bin_attribute i2o_config_lct_attr = {
+	.attr = {
+		 .name = "lct",
+		 .mode = S_IRUGO,
+		 .owner = THIS_MODULE},
+	.size = 0,
+	.read = i2o_config_read_lct
+};
+
+/* IRTOS firmware access */
+I2O_CONFIG_SW_ATTR(irtos, S_IWRSR, I2O_SOFTWARE_MODULE_IRTOS, 0);
+
+#ifdef CONFIG_I2O_EXT_ADAPTEC
+
+/*
+ * attribute for BIOS / SMOR, nvram and serial number access on DPT / Adaptec
+ * controllers
+ */
+I2O_CONFIG_SW_ATTR(bios, S_IWRSR, I2O_SOFTWARE_MODULE_IOP_PRIVATE, 0);
+I2O_CONFIG_SW_ATTR(nvram, S_IWRSR, I2O_SOFTWARE_MODULE_IOP_CONFIG, 0);
+I2O_CONFIG_SW_ATTR(serial, S_IWRSR, I2O_SOFTWARE_MODULE_IOP_CONFIG, 1);
+
+#endif
+
+/**
+ *	i2o_config_notify_controller_add - Notify of added controller
+ *	@c: the controller which was added
+ *
+ *	If a I2O controller is added, we catch the notification to add sysfs
+ *	entries.
+ */
+static void i2o_config_notify_controller_add(struct i2o_controller *c)
+{
+	struct kobject *kobj = &c->exec->device.kobj;
+
+	sysfs_create_bin_file(kobj, &i2o_config_hrt_attr);
+	sysfs_create_bin_file(kobj, &i2o_config_lct_attr);
+
+	sysfs_create_fops_file(kobj, &i2o_config_attr_irtos);
+#ifdef CONFIG_I2O_EXT_ADAPTEC
+	if (c->adaptec) {
+		sysfs_create_fops_file(kobj, &i2o_config_attr_bios);
+		sysfs_create_fops_file(kobj, &i2o_config_attr_nvram);
+		sysfs_create_fops_file(kobj, &i2o_config_attr_serial);
+	}
+#endif
+};
+
+/**
+ *	i2o_config_notify_controller_remove - Notify of removed controller
+ *	@c: the controller which was removed
+ *
+ *	If a I2O controller is removed, we catch the notification to remove the
+ *	sysfs entries.
+ */
+static void i2o_config_notify_controller_remove(struct i2o_controller *c)
+{
+	struct kobject *kobj = &c->exec->device.kobj;
+
+#ifdef CONFIG_I2O_EXT_ADAPTEC
+	if (c->adaptec) {
+		sysfs_remove_fops_file(kobj, &i2o_config_attr_serial);
+		sysfs_remove_fops_file(kobj, &i2o_config_attr_nvram);
+		sysfs_remove_fops_file(kobj, &i2o_config_attr_bios);
+	}
+#endif
+	sysfs_remove_fops_file(kobj, &i2o_config_attr_irtos);
+
+	sysfs_remove_bin_file(kobj, &i2o_config_lct_attr);
+	sysfs_remove_bin_file(kobj, &i2o_config_hrt_attr);
+};
+
+/* Config OSM driver struct */
+static struct i2o_driver i2o_config_driver = {
+	.name = OSM_NAME,
+	.notify_controller_add = i2o_config_notify_controller_add,
+	.notify_controller_remove = i2o_config_notify_controller_remove
+};
+
+#ifdef CONFIG_I2O_CONFIG_OLD_IOCTL
+#include "i2o_config.c"
+#endif
+
+/**
+ *	i2o_config_init - Configuration OSM initialization function
+ *
+ *	Registers Configuration OSM in the I2O core and if old ioctl's are
+ *	compiled in initialize them.
+ *
+ *	Returns 0 on success or negative error code on failure.
+ */
+static int __init i2o_config_init(void)
+{
+	printk(KERN_INFO OSM_DESCRIPTION " v" OSM_VERSION "\n");
+
+	if (i2o_driver_register(&i2o_config_driver)) {
+		osm_err("handler register failed.\n");
+		return -EBUSY;
+	}
+#ifdef CONFIG_I2O_CONFIG_OLD_IOCTL
+	if (i2o_config_old_init())
+		i2o_driver_unregister(&i2o_config_driver);
+#endif
+
+	return 0;
+}
+
+/**
+ *	i2o_config_exit - Configuration OSM exit function
+ *
+ *	If old ioctl's are compiled in exit remove them and unregisters
+ *	Configuration OSM from I2O core.
+ */
+static void i2o_config_exit(void)
+{
+#ifdef CONFIG_I2O_CONFIG_OLD_IOCTL
+	i2o_config_old_exit();
+#endif
+
+	i2o_driver_unregister(&i2o_config_driver);
+}
+
+MODULE_AUTHOR("Markus Lidel <Markus.Lidel@shadowconnect.com>");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION(OSM_DESCRIPTION);
+MODULE_VERSION(OSM_VERSION);
+
+module_init(i2o_config_init);
+module_exit(i2o_config_exit);
diff --git a/drivers/message/i2o/driver.c b/drivers/message/i2o/driver.c
index bebdd509b5d..393be8e2914 100644
--- a/drivers/message/i2o/driver.c
+++ b/drivers/message/i2o/driver.c
@@ -180,7 +180,13 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m)
 {
 	struct i2o_driver *drv;
 	struct i2o_message __iomem *msg = i2o_msg_out_to_virt(c, m);
-	u32 context = readl(&msg->u.s.icntxt);
+	u32 context;
+	unsigned long flags;
+
+	if(unlikely(!msg))
+		return -EIO;
+
+	context = readl(&msg->u.s.icntxt);
 
 	if (unlikely(context >= i2o_max_drivers)) {
 		osm_warn("%s: Spurious reply to unknown driver %d\n", c->name,
@@ -188,9 +194,9 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m)
 		return -EIO;
 	}
 
-	spin_lock(&i2o_drivers_lock);
+	spin_lock_irqsave(&i2o_drivers_lock, flags);
 	drv = i2o_drivers[context];
-	spin_unlock(&i2o_drivers_lock);
+	spin_unlock_irqrestore(&i2o_drivers_lock, flags);
 
 	if (unlikely(!drv)) {
 		osm_warn("%s: Spurious reply to unknown driver %d\n", c->name,
diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c
index 5581344fbba..0160221c802 100644
--- a/drivers/message/i2o/exec-osm.c
+++ b/drivers/message/i2o/exec-osm.c
@@ -206,6 +206,7 @@ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m,
 				      u32 context)
 {
 	struct i2o_exec_wait *wait, *tmp;
+	unsigned long flags;
 	static spinlock_t lock = SPIN_LOCK_UNLOCKED;
 	int rc = 1;
 
@@ -216,11 +217,13 @@ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m,
 	 * already expired. Not much we can do about that except log it for
 	 * debug purposes, increase timeout, and recompile.
 	 */
-	spin_lock(&lock);
+	spin_lock_irqsave(&lock, flags);
 	list_for_each_entry_safe(wait, tmp, &i2o_exec_wait_list, list) {
 		if (wait->tcntxt == context) {
 			list_del(&wait->list);
 
+			spin_unlock_irqrestore(&lock, flags);
+
 			wait->m = m;
 			wait->msg = msg;
 			wait->complete = 1;
@@ -242,13 +245,11 @@ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m,
 				rc = -1;
 			}
 
-			spin_unlock(&lock);
-
 			return rc;
 		}
 	}
 
-	spin_unlock(&lock);
+	spin_unlock_irqrestore(&lock, flags);
 
 	osm_warn("%s: Bogus reply in POST WAIT (tr-context: %08x)!\n", c->name,
 		 context);
@@ -256,6 +257,50 @@ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m,
 	return -1;
 };
 
+/**
+ *	i2o_exec_show_vendor_id - Displays Vendor ID of controller
+ *	@d: device of which the Vendor ID should be displayed
+ *	@buf: buffer into which the Vendor ID should be printed
+ *
+ *	Returns number of bytes printed into buffer.
+ */
+static ssize_t i2o_exec_show_vendor_id(struct device *d, char *buf)
+{
+	struct i2o_device *dev = to_i2o_device(d);
+	u16 id;
+
+	if (i2o_parm_field_get(dev, 0x0000, 0, &id, 2)) {
+		sprintf(buf, "0x%04x", id);
+		return strlen(buf) + 1;
+	}
+
+	return 0;
+};
+
+/**
+ *	i2o_exec_show_product_id - Displays Product ID of controller
+ *	@d: device of which the Product ID should be displayed
+ *	@buf: buffer into which the Product ID should be printed
+ *
+ *	Returns number of bytes printed into buffer.
+ */
+static ssize_t i2o_exec_show_product_id(struct device *d, char *buf)
+{
+	struct i2o_device *dev = to_i2o_device(d);
+	u16 id;
+
+	if (i2o_parm_field_get(dev, 0x0000, 1, &id, 2)) {
+		sprintf(buf, "0x%04x", id);
+		return strlen(buf) + 1;
+	}
+
+	return 0;
+};
+
+/* Exec-OSM device attributes */
+static DEVICE_ATTR(vendor_id, S_IRUGO, i2o_exec_show_vendor_id, NULL);
+static DEVICE_ATTR(product_id, S_IRUGO, i2o_exec_show_product_id, NULL);
+
 /**
  *	i2o_exec_probe - Called if a new I2O device (executive class) appears
  *	@dev: I2O device which should be probed
@@ -268,10 +313,16 @@ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m,
 static int i2o_exec_probe(struct device *dev)
 {
 	struct i2o_device *i2o_dev = to_i2o_device(dev);
+	struct i2o_controller *c = i2o_dev->iop;
 
 	i2o_event_register(i2o_dev, &i2o_exec_driver, 0, 0xffffffff);
 
-	i2o_dev->iop->exec = i2o_dev;
+	c->exec = i2o_dev;
+
+	i2o_exec_lct_notify(c, c->lct->change_ind + 1);
+
+	device_create_file(dev, &dev_attr_vendor_id);
+	device_create_file(dev, &dev_attr_product_id);
 
 	return 0;
 };
@@ -286,6 +337,9 @@ static int i2o_exec_probe(struct device *dev)
  */
 static int i2o_exec_remove(struct device *dev)
 {
+	device_remove_file(dev, &dev_attr_product_id);
+	device_remove_file(dev, &dev_attr_vendor_id);
+
 	i2o_event_register(to_i2o_device(dev), &i2o_exec_driver, 0, 0);
 
 	return 0;
@@ -297,12 +351,16 @@ static int i2o_exec_remove(struct device *dev)
  *
  *	This function handles asynchronus LCT NOTIFY replies. It parses the
  *	new LCT and if the buffer for the LCT was to small sends a LCT NOTIFY
- *	again.
+ *	again, otherwise send LCT NOTIFY to get informed on next LCT change.
  */
 static void i2o_exec_lct_modified(struct i2o_controller *c)
 {
-	if (i2o_device_parse_lct(c) == -EAGAIN)
-		i2o_exec_lct_notify(c, 0);
+	u32 change_ind = 0;
+
+	if (i2o_device_parse_lct(c) != -EAGAIN)
+		change_ind = c->lct->change_ind + 1;
+
+	i2o_exec_lct_notify(c, change_ind);
 };
 
 /**
diff --git a/drivers/message/i2o/i2o_block.c b/drivers/message/i2o/i2o_block.c
index e69421e36ac..1dd2b9dad50 100644
--- a/drivers/message/i2o/i2o_block.c
+++ b/drivers/message/i2o/i2o_block.c
@@ -146,6 +146,29 @@ static int i2o_block_device_flush(struct i2o_device *dev)
 	return i2o_msg_post_wait(dev->iop, m, 60);
 };
 
+/**
+ *	i2o_block_issue_flush - device-flush interface for block-layer
+ *	@queue: the request queue of the device which should be flushed
+ *	@disk: gendisk
+ *	@error_sector: error offset
+ *
+ *	Helper function to provide flush functionality to block-layer.
+ *
+ *	Returns 0 on success or negative error code on failure.
+ */
+
+static int i2o_block_issue_flush(request_queue_t * queue, struct gendisk *disk,
+				 sector_t * error_sector)
+{
+	struct i2o_block_device *i2o_blk_dev = queue->queuedata;
+	int rc = -ENODEV;
+
+	if (likely(i2o_blk_dev))
+		rc = i2o_block_device_flush(i2o_blk_dev->i2o_dev);
+
+	return rc;
+}
+
 /**
  *	i2o_block_device_mount - Mount (load) the media of device dev
  *	@dev: I2O device which should receive the mount request
@@ -299,28 +322,31 @@ static inline void i2o_block_request_free(struct i2o_block_request *ireq)
 
 /**
  *	i2o_block_sglist_alloc - Allocate the SG list and map it
+ *	@c: I2O controller to which the request belongs
  *	@ireq: I2O block request
  *
- *	Builds the SG list and map it into to be accessable by the controller.
+ *	Builds the SG list and map it to be accessable by the controller.
  *
- *	Returns the number of elements in the SG list or 0 on failure.
+ *	Returns 0 on failure or 1 on success.
  */
-static inline int i2o_block_sglist_alloc(struct i2o_block_request *ireq)
+static inline int i2o_block_sglist_alloc(struct i2o_controller *c,
+					 struct i2o_block_request *ireq,
+					 u32 __iomem ** mptr)
 {
-	struct device *dev = &ireq->i2o_blk_dev->i2o_dev->iop->pdev->dev;
 	int nents;
+	enum dma_data_direction direction;
 
+	ireq->dev = &c->pdev->dev;
 	nents = blk_rq_map_sg(ireq->req->q, ireq->req, ireq->sg_table);
 
 	if (rq_data_dir(ireq->req) == READ)
-		ireq->sg_dma_direction = PCI_DMA_FROMDEVICE;
+		direction = PCI_DMA_FROMDEVICE;
 	else
-		ireq->sg_dma_direction = PCI_DMA_TODEVICE;
+		direction = PCI_DMA_TODEVICE;
 
-	ireq->sg_nents = dma_map_sg(dev, ireq->sg_table, nents,
-				    ireq->sg_dma_direction);
+	ireq->sg_nents = nents;
 
-	return ireq->sg_nents;
+	return i2o_dma_map_sg(c, ireq->sg_table, nents, direction, mptr);
 };
 
 /**
@@ -331,10 +357,14 @@ static inline int i2o_block_sglist_alloc(struct i2o_block_request *ireq)
  */
 static inline void i2o_block_sglist_free(struct i2o_block_request *ireq)
 {
-	struct device *dev = &ireq->i2o_blk_dev->i2o_dev->iop->pdev->dev;
+	enum dma_data_direction direction;
 
-	dma_unmap_sg(dev, ireq->sg_table, ireq->sg_nents,
-		     ireq->sg_dma_direction);
+	if (rq_data_dir(ireq->req) == READ)
+		direction = PCI_DMA_FROMDEVICE;
+	else
+		direction = PCI_DMA_TODEVICE;
+
+	dma_unmap_sg(ireq->dev, ireq->sg_table, ireq->sg_nents, direction);
 };
 
 /**
@@ -352,6 +382,11 @@ static int i2o_block_prep_req_fn(struct request_queue *q, struct request *req)
 	struct i2o_block_device *i2o_blk_dev = q->queuedata;
 	struct i2o_block_request *ireq;
 
+	if (unlikely(!i2o_blk_dev)) {
+		osm_err("block device already removed\n");
+		return BLKPREP_KILL;
+	}
+
 	/* request is already processed by us, so return */
 	if (req->flags & REQ_SPECIAL) {
 		osm_debug("REQ_SPECIAL already set!\n");
@@ -414,11 +449,11 @@ static void i2o_block_end_request(struct request *req, int uptodate,
 {
 	struct i2o_block_request *ireq = req->special;
 	struct i2o_block_device *dev = ireq->i2o_blk_dev;
-	request_queue_t *q = dev->gd->queue;
+	request_queue_t *q = req->q;
 	unsigned long flags;
 
 	if (end_that_request_chunk(req, uptodate, nr_bytes)) {
-		int leftover = (req->hard_nr_sectors << 9);
+		int leftover = (req->hard_nr_sectors << KERNEL_SECTOR_SHIFT);
 
 		if (blk_pc_request(req))
 			leftover = req->data_len;
@@ -432,8 +467,11 @@ static void i2o_block_end_request(struct request *req, int uptodate,
 	spin_lock_irqsave(q->queue_lock, flags);
 
 	end_that_request_last(req);
-	dev->open_queue_depth--;
-	list_del(&ireq->queue);
+
+	if (likely(dev)) {
+		dev->open_queue_depth--;
+		list_del(&ireq->queue);
+	}
 
 	blk_start_queue(q);
 
@@ -483,8 +521,8 @@ static int i2o_block_reply(struct i2o_controller *c, u32 m,
 		 *      Don't stick a supertrak100 into cache aggressive modes
 		 */
 
-		osm_err("%03x error status: %02x, detailed status: %04x\n",
-			(le32_to_cpu(msg->u.head[1]) >> 12 & 0xfff),
+		osm_err("TID %03x error status: 0x%02x, detailed status: "
+			"0x%04x\n", (le32_to_cpu(msg->u.head[1]) >> 12 & 0xfff),
 			status >> 24, status & 0xffff);
 
 		req->errors++;
@@ -705,18 +743,25 @@ static int i2o_block_media_changed(struct gendisk *disk)
 static int i2o_block_transfer(struct request *req)
 {
 	struct i2o_block_device *dev = req->rq_disk->private_data;
-	struct i2o_controller *c = dev->i2o_dev->iop;
+	struct i2o_controller *c;
 	int tid = dev->i2o_dev->lct_data.tid;
 	struct i2o_message __iomem *msg;
-	void __iomem *mptr;
+	u32 __iomem *mptr;
 	struct i2o_block_request *ireq = req->special;
-	struct scatterlist *sg;
-	int sgnum;
-	int i;
 	u32 m;
 	u32 tcntxt;
-	u32 sg_flags;
+	u32 sgl_offset = SGL_OFFSET_8;
+	u32 ctl_flags = 0x00000000;
 	int rc;
+	u32 cmd;
+
+	if (unlikely(!dev->i2o_dev)) {
+		osm_err("transfer to removed drive\n");
+		rc = -ENODEV;
+		goto exit;
+	}
+
+	c = dev->i2o_dev->iop;
 
 	m = i2o_msg_get(c, &msg);
 	if (m == I2O_QUEUE_EMPTY) {
@@ -730,80 +775,109 @@ static int i2o_block_transfer(struct request *req)
 		goto nop_msg;
 	}
 
-	if ((sgnum = i2o_block_sglist_alloc(ireq)) <= 0) {
-		rc = -ENOMEM;
-		goto context_remove;
-	}
-
-	/* Build the message based on the request. */
 	writel(i2o_block_driver.context, &msg->u.s.icntxt);
 	writel(tcntxt, &msg->u.s.tcntxt);
-	writel(req->nr_sectors << 9, &msg->body[1]);
 
-	writel((((u64) req->sector) << 9) & 0xffffffff, &msg->body[2]);
-	writel(req->sector >> 23, &msg->body[3]);
-
-	mptr = &msg->body[4];
-
-	sg = ireq->sg_table;
+	mptr = &msg->body[0];
 
 	if (rq_data_dir(req) == READ) {
-		writel(I2O_CMD_BLOCK_READ << 24 | HOST_TID << 12 | tid,
-		       &msg->u.head[1]);
-		sg_flags = 0x10000000;
+		cmd = I2O_CMD_BLOCK_READ << 24;
+
 		switch (dev->rcache) {
-		case CACHE_NULL:
-			writel(0, &msg->body[0]);
-			break;
 		case CACHE_PREFETCH:
-			writel(0x201F0008, &msg->body[0]);
+			ctl_flags = 0x201F0008;
 			break;
+
 		case CACHE_SMARTFETCH:
 			if (req->nr_sectors > 16)
-				writel(0x201F0008, &msg->body[0]);
+				ctl_flags = 0x201F0008;
 			else
-				writel(0x001F0000, &msg->body[0]);
+				ctl_flags = 0x001F0000;
+			break;
+
+		default:
 			break;
 		}
 	} else {
-		writel(I2O_CMD_BLOCK_WRITE << 24 | HOST_TID << 12 | tid,
-		       &msg->u.head[1]);
-		sg_flags = 0x14000000;
+		cmd = I2O_CMD_BLOCK_WRITE << 24;
+
 		switch (dev->wcache) {
-		case CACHE_NULL:
-			writel(0, &msg->body[0]);
-			break;
 		case CACHE_WRITETHROUGH:
-			writel(0x001F0008, &msg->body[0]);
+			ctl_flags = 0x001F0008;
 			break;
 		case CACHE_WRITEBACK:
-			writel(0x001F0010, &msg->body[0]);
+			ctl_flags = 0x001F0010;
 			break;
 		case CACHE_SMARTBACK:
 			if (req->nr_sectors > 16)
-				writel(0x001F0004, &msg->body[0]);
+				ctl_flags = 0x001F0004;
 			else
-				writel(0x001F0010, &msg->body[0]);
+				ctl_flags = 0x001F0010;
 			break;
 		case CACHE_SMARTTHROUGH:
 			if (req->nr_sectors > 16)
-				writel(0x001F0004, &msg->body[0]);
+				ctl_flags = 0x001F0004;
 			else
-				writel(0x001F0010, &msg->body[0]);
+				ctl_flags = 0x001F0010;
+		default:
+			break;
+		}
+	}
+
+#ifdef CONFIG_I2O_EXT_ADAPTEC
+	if (c->adaptec) {
+		u8 cmd[10];
+		u32 scsi_flags;
+		u16 hwsec = queue_hardsect_size(req->q) >> KERNEL_SECTOR_SHIFT;
+
+		memset(cmd, 0, 10);
+
+		sgl_offset = SGL_OFFSET_12;
+
+		writel(I2O_CMD_PRIVATE << 24 | HOST_TID << 12 | tid,
+		       &msg->u.head[1]);
+
+		writel(I2O_VENDOR_DPT << 16 | I2O_CMD_SCSI_EXEC, mptr++);
+		writel(tid, mptr++);
+
+		/*
+		 * ENABLE_DISCONNECT
+		 * SIMPLE_TAG
+		 * RETURN_SENSE_DATA_IN_REPLY_MESSAGE_FRAME
+		 */
+		if (rq_data_dir(req) == READ) {
+			cmd[0] = 0x28;
+			scsi_flags = 0x60a0000a;
+		} else {
+			cmd[0] = 0x2A;
+			scsi_flags = 0xa0a0000a;
 		}
+
+		writel(scsi_flags, mptr++);
+
+		*((u32 *) & cmd[2]) = cpu_to_be32(req->sector * hwsec);
+		*((u16 *) & cmd[7]) = cpu_to_be16(req->nr_sectors * hwsec);
+
+		memcpy_toio(mptr, cmd, 10);
+		mptr += 4;
+		writel(req->nr_sectors << KERNEL_SECTOR_SHIFT, mptr++);
+	} else
+#endif
+	{
+		writel(cmd | HOST_TID << 12 | tid, &msg->u.head[1]);
+		writel(ctl_flags, mptr++);
+		writel(req->nr_sectors << KERNEL_SECTOR_SHIFT, mptr++);
+		writel((u32) (req->sector << KERNEL_SECTOR_SHIFT), mptr++);
+		writel(req->sector >> (32 - KERNEL_SECTOR_SHIFT), mptr++);
 	}
 
-	for (i = sgnum; i > 0; i--) {
-		if (i == 1)
-			sg_flags |= 0x80000000;
-		writel(sg_flags | sg_dma_len(sg), mptr);
-		writel(sg_dma_address(sg), mptr + 4);
-		mptr += 8;
-		sg++;
+	if (!i2o_block_sglist_alloc(c, ireq, &mptr)) {
+		rc = -ENOMEM;
+		goto context_remove;
 	}
 
-	writel(I2O_MESSAGE_SIZE(mptr - &msg->u.head[0]) | SGL_OFFSET_8,
-	       &msg->u.head[0]);
+	writel(I2O_MESSAGE_SIZE(mptr - &msg->u.head[0]) |
+	       sgl_offset, &msg->u.head[0]);
 
 	list_add_tail(&ireq->queue, &dev->open_queue);
 	dev->open_queue_depth++;
@@ -846,11 +920,13 @@ static void i2o_block_request_fn(struct request_queue *q)
 
 			queue_depth = ireq->i2o_blk_dev->open_queue_depth;
 
-			if (queue_depth < I2O_BLOCK_MAX_OPEN_REQUESTS)
+			if (queue_depth < I2O_BLOCK_MAX_OPEN_REQUESTS) {
 				if (!i2o_block_transfer(req)) {
 					blkdev_dequeue_request(req);
 					continue;
-				}
+				} else
+					osm_info("transfer error\n");
+			}
 
 			if (queue_depth)
 				break;
@@ -933,6 +1009,7 @@ static struct i2o_block_device *i2o_block_device_alloc(void)
 	}
 
 	blk_queue_prep_rq(queue, i2o_block_prep_req_fn);
+	blk_queue_issue_flush_fn(queue, i2o_block_issue_flush);
 
 	gd->major = I2O_MAJOR;
 	gd->queue = queue;
@@ -974,7 +1051,18 @@ static int i2o_block_probe(struct device *dev)
 	u64 size;
 	u32 blocksize;
 	u32 flags, status;
-	int segments;
+	u16 body_size = 4;
+	unsigned short max_sectors;
+
+#ifdef CONFIG_I2O_EXT_ADAPTEC
+	if (c->adaptec)
+		body_size = 8;
+#endif
+
+	if (c->limit_sectors)
+		max_sectors = I2O_MAX_SECTORS_LIMITED;
+	else
+		max_sectors = I2O_MAX_SECTORS;
 
 	/* skip devices which are used by IOP */
 	if (i2o_dev->lct_data.user_tid != 0xfff) {
@@ -1009,50 +1097,35 @@ static int i2o_block_probe(struct device *dev)
 	queue = gd->queue;
 	queue->queuedata = i2o_blk_dev;
 
-	blk_queue_max_phys_segments(queue, I2O_MAX_SEGMENTS);
-	blk_queue_max_sectors(queue, I2O_MAX_SECTORS);
-
-	if (c->short_req)
-		segments = 8;
-	else {
-		i2o_status_block *sb;
+	blk_queue_max_phys_segments(queue, I2O_MAX_PHYS_SEGMENTS);
+	blk_queue_max_sectors(queue, max_sectors);
+	blk_queue_max_hw_segments(queue, i2o_sg_tablesize(c, body_size));
 
-		sb = c->status_block.virt;
-
-		segments = (sb->inbound_frame_size -
-			    sizeof(struct i2o_message) / 4 - 4) / 2;
-	}
-
-	blk_queue_max_hw_segments(queue, segments);
-
-	osm_debug("max sectors = %d\n", I2O_MAX_SECTORS);
-	osm_debug("phys segments = %d\n", I2O_MAX_SEGMENTS);
-	osm_debug("hw segments = %d\n", segments);
+	osm_debug("max sectors = %d\n", queue->max_phys_segments);
+	osm_debug("phys segments = %d\n", queue->max_sectors);
+	osm_debug("max hw segments = %d\n", queue->max_hw_segments);
 
 	/*
 	 *      Ask for the current media data. If that isn't supported
 	 *      then we ask for the device capacity data
 	 */
-	if (!i2o_parm_field_get(i2o_dev, 0x0004, 0, &size, 8))
-		if (!i2o_parm_field_get(i2o_dev, 0x0000, 4, &size, 8)) {
-			osm_warn("could not get size of %s\n", gd->disk_name);
-			size = 0;
-		}
+	if (i2o_parm_field_get(i2o_dev, 0x0004, 1, &blocksize, 4) ||
+	    i2o_parm_field_get(i2o_dev, 0x0000, 3, &blocksize, 4)) {
+		blk_queue_hardsect_size(queue, blocksize);
+	} else
+		osm_warn("unable to get blocksize of %s\n", gd->disk_name);
 
-	if (!i2o_parm_field_get(i2o_dev, 0x0004, 1, &blocksize, 4))
-		if (!i2o_parm_field_get(i2o_dev, 0x0000, 3, &blocksize, 4)) {
-			osm_warn("unable to get blocksize of %s\n",
-				 gd->disk_name);
-			blocksize = 0;
-		}
+	if (i2o_parm_field_get(i2o_dev, 0x0004, 0, &size, 8) ||
+	    i2o_parm_field_get(i2o_dev, 0x0000, 4, &size, 8)) {
+		set_capacity(gd, size >> KERNEL_SECTOR_SHIFT);
+	} else
+		osm_warn("could not get size of %s\n", gd->disk_name);
 
 	if (!i2o_parm_field_get(i2o_dev, 0x0000, 2, &i2o_blk_dev->power, 2))
 		i2o_blk_dev->power = 0;
 	i2o_parm_field_get(i2o_dev, 0x0000, 5, &flags, 4);
 	i2o_parm_field_get(i2o_dev, 0x0000, 6, &status, 4);
 
-	set_capacity(gd, size >> 9);
-
 	i2o_event_register(i2o_dev, &i2o_block_driver, 0, 0xffffffff);
 
 	add_disk(gd);
@@ -1109,7 +1182,7 @@ static int __init i2o_block_init(void)
 		goto exit;
 	}
 
-	i2o_blk_req_pool.pool = mempool_create(I2O_REQ_MEMPOOL_SIZE,
+	i2o_blk_req_pool.pool = mempool_create(I2O_BLOCK_REQ_MEMPOOL_SIZE,
 					       mempool_alloc_slab,
 					       mempool_free_slab,
 					       i2o_blk_req_pool.slab);
diff --git a/drivers/message/i2o/i2o_block.h b/drivers/message/i2o/i2o_block.h
index 712111ffa63..9e1a95fb083 100644
--- a/drivers/message/i2o/i2o_block.h
+++ b/drivers/message/i2o/i2o_block.h
@@ -84,9 +84,9 @@ struct i2o_block_request
 	struct list_head queue;
 	struct request *req;		/* corresponding request */
 	struct i2o_block_device *i2o_blk_dev;	/* I2O block device */
-	int sg_dma_direction;		/* direction of DMA buffer read/write */
+	struct device *dev;		/* device used for DMA */
 	int sg_nents;			/* number of SG elements */
-	struct scatterlist sg_table[I2O_MAX_SEGMENTS]; /* SG table */
+	struct scatterlist sg_table[I2O_MAX_PHYS_SEGMENTS]; /* SG table */
 };
 
 /* I2O Block device delayed request */
diff --git a/drivers/message/i2o/i2o_config.c b/drivers/message/i2o/i2o_config.c
index 383e89a5c9f..849d90aad77 100644
--- a/drivers/message/i2o/i2o_config.c
+++ b/drivers/message/i2o/i2o_config.c
@@ -30,27 +30,11 @@
  * 2 of the License, or (at your option) any later version.
  */
 
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/pci.h>
-#include <linux/i2o.h>
-#include <linux/errno.h>
-#include <linux/init.h>
-#include <linux/slab.h>
 #include <linux/miscdevice.h>
-#include <linux/mm.h>
-#include <linux/spinlock.h>
 #include <linux/smp_lock.h>
-#include <linux/ioctl32.h>
 #include <linux/compat.h>
-#include <linux/syscalls.h>
 
 #include <asm/uaccess.h>
-#include <asm/io.h>
-
-#define OSM_NAME	"config-osm"
-#define OSM_VERSION	"$Rev$"
-#define OSM_DESCRIPTION	"I2O Configuration OSM"
 
 extern int i2o_parm_issue(struct i2o_device *, int, void *, int, void *, int);
 
@@ -80,125 +64,6 @@ struct i2o_cfg_info {
 static struct i2o_cfg_info *open_files = NULL;
 static ulong i2o_cfg_info_id = 0;
 
-/**
- *	i2o_config_read_hrt - Returns the HRT of the controller
- *	@kob: kernel object handle
- *	@buf: buffer into which the HRT should be copied
- *	@off: file offset
- *	@count: number of bytes to read
- *
- *	Put @count bytes starting at @off into @buf from the HRT of the I2O
- *	controller corresponding to @kobj.
- *
- *	Returns number of bytes copied into buffer.
- */
-static ssize_t i2o_config_read_hrt(struct kobject *kobj, char *buf,
-				   loff_t offset, size_t count)
-{
-	struct i2o_controller *c = to_i2o_controller(container_of(kobj,
-								  struct device,
-								  kobj));
-	i2o_hrt *hrt = c->hrt.virt;
-
-	u32 size = (hrt->num_entries * hrt->entry_len + 2) * 4;
-
-	if(offset > size)
-		return 0;
-
-	if(offset + count > size)
-		count = size - offset;
-
-	memcpy(buf, (u8 *) hrt + offset, count);
-
-	return count;
-};
-
-/**
- *	i2o_config_read_lct - Returns the LCT of the controller
- *	@kob: kernel object handle
- *	@buf: buffer into which the LCT should be copied
- *	@off: file offset
- *	@count: number of bytes to read
- *
- *	Put @count bytes starting at @off into @buf from the LCT of the I2O
- *	controller corresponding to @kobj.
- *
- *	Returns number of bytes copied into buffer.
- */
-static ssize_t i2o_config_read_lct(struct kobject *kobj, char *buf,
-				   loff_t offset, size_t count)
-{
-	struct i2o_controller *c = to_i2o_controller(container_of(kobj,
-								  struct device,
-								  kobj));
-	u32 size = c->lct->table_size * 4;
-
-	if(offset > size)
-		return 0;
-
-	if(offset + count > size)
-		count = size - offset;
-
-	memcpy(buf, (u8 *) c->lct + offset, count);
-
-	return count;
-};
-
-/* attribute for HRT in sysfs */
-static struct bin_attribute i2o_config_hrt_attr = {
-	.attr = {
-		.name = "hrt",
-		.mode = S_IRUGO,
-		.owner = THIS_MODULE
-	},
-	.size = 0,
-	.read = i2o_config_read_hrt
-};
-
-/* attribute for LCT in sysfs */
-static struct bin_attribute i2o_config_lct_attr = {
-	.attr = {
-		.name = "lct",
-		.mode = S_IRUGO,
-		.owner = THIS_MODULE
-	},
-	.size = 0,
-	.read = i2o_config_read_lct
-};
-
-/**
- *	i2o_config_notify_controller_add - Notify of added controller
- *	@c: the controller which was added
- *
- *	If a I2O controller is added, we catch the notification to add sysfs
- *	entries.
- */
-static void i2o_config_notify_controller_add(struct i2o_controller *c)
-{
-	sysfs_create_bin_file(&(c->device.kobj), &i2o_config_hrt_attr);
-	sysfs_create_bin_file(&(c->device.kobj), &i2o_config_lct_attr);
-};
-
-/**
- *	i2o_config_notify_controller_remove - Notify of removed controller
- *	@c: the controller which was removed
- *
- *	If a I2O controller is removed, we catch the notification to remove the
- *	sysfs entries.
- */
-static void i2o_config_notify_controller_remove(struct i2o_controller *c)
-{
-	sysfs_remove_bin_file(&c->device.kobj, &i2o_config_lct_attr);
-	sysfs_remove_bin_file(&c->device.kobj, &i2o_config_hrt_attr);
-};
-
-/* Config OSM driver struct */
-static struct i2o_driver i2o_config_driver = {
-	.name = OSM_NAME,
-	.notify_controller_add = i2o_config_notify_controller_add,
-	.notify_controller_remove = i2o_config_notify_controller_remove
-};
-
 static int i2o_cfg_getiops(unsigned long arg)
 {
 	struct i2o_controller *c;
@@ -1257,37 +1122,20 @@ static struct miscdevice i2o_miscdev = {
 	&config_fops
 };
 
-static int __init i2o_config_init(void)
+static int __init i2o_config_old_init(void)
 {
-	printk(KERN_INFO OSM_DESCRIPTION " v" OSM_VERSION "\n");
-
 	spin_lock_init(&i2o_config_lock);
 
 	if (misc_register(&i2o_miscdev) < 0) {
 		osm_err("can't register device.\n");
 		return -EBUSY;
 	}
-	/*
-	 *      Install our handler
-	 */
-	if (i2o_driver_register(&i2o_config_driver)) {
-		osm_err("handler register failed.\n");
-		misc_deregister(&i2o_miscdev);
-		return -EBUSY;
-	}
 	return 0;
 }
 
-static void i2o_config_exit(void)
+static void i2o_config_old_exit(void)
 {
 	misc_deregister(&i2o_miscdev);
-	i2o_driver_unregister(&i2o_config_driver);
 }
 
 MODULE_AUTHOR("Red Hat Software");
-MODULE_LICENSE("GPL");
-MODULE_DESCRIPTION(OSM_DESCRIPTION);
-MODULE_VERSION(OSM_VERSION);
-
-module_init(i2o_config_init);
-module_exit(i2o_config_exit);
diff --git a/drivers/message/i2o/i2o_proc.c b/drivers/message/i2o/i2o_proc.c
index b176d0eeff7..e5b74452c49 100644
--- a/drivers/message/i2o/i2o_proc.c
+++ b/drivers/message/i2o/i2o_proc.c
@@ -228,7 +228,7 @@ static const char *i2o_get_class_name(int class)
 	case I2O_CLASS_FLOPPY_DEVICE:
 		idx = 12;
 		break;
-	case I2O_CLASS_BUS_ADAPTER_PORT:
+	case I2O_CLASS_BUS_ADAPTER:
 		idx = 13;
 		break;
 	case I2O_CLASS_PEER_TRANSPORT_AGENT:
@@ -490,7 +490,7 @@ static int i2o_seq_show_lct(struct seq_file *seq, void *v)
 				seq_printf(seq, ", Unknown Device Type");
 			break;
 
-		case I2O_CLASS_BUS_ADAPTER_PORT:
+		case I2O_CLASS_BUS_ADAPTER:
 			if (lct->lct_entry[i].sub_class < BUS_TABLE_SIZE)
 				seq_printf(seq, ", %s",
 					   bus_ports[lct->lct_entry[i].
diff --git a/drivers/message/i2o/i2o_scsi.c b/drivers/message/i2o/i2o_scsi.c
index 812c29ec86d..c3b0c29ac02 100644
--- a/drivers/message/i2o/i2o_scsi.c
+++ b/drivers/message/i2o/i2o_scsi.c
@@ -103,7 +103,7 @@ static struct i2o_scsi_host *i2o_scsi_host_alloc(struct i2o_controller *c)
 	i2o_status_block *sb;
 
 	list_for_each_entry(i2o_dev, &c->devices, list)
-	    if (i2o_dev->lct_data.class_id == I2O_CLASS_BUS_ADAPTER_PORT) {
+	    if (i2o_dev->lct_data.class_id == I2O_CLASS_BUS_ADAPTER) {
 		if (i2o_parm_field_get(i2o_dev, 0x0000, 0, &type, 1)
 		   && (type == 0x01))	/* SCSI bus */
 			max_channel++;
@@ -139,7 +139,7 @@ static struct i2o_scsi_host *i2o_scsi_host_alloc(struct i2o_controller *c)
 
 	i = 0;
 	list_for_each_entry(i2o_dev, &c->devices, list)
-	    if (i2o_dev->lct_data.class_id == I2O_CLASS_BUS_ADAPTER_PORT) {
+	    if (i2o_dev->lct_data.class_id == I2O_CLASS_BUS_ADAPTER) {
 		if (i2o_parm_field_get(i2o_dev, 0x0000, 0, &type, 1) || (type == 1))	/* only SCSI bus */
 			i2o_shost->channel[i++] = i2o_dev;
 
@@ -186,6 +186,7 @@ static int i2o_scsi_remove(struct device *dev)
 
 	shost_for_each_device(scsi_dev, i2o_shost->scsi_host)
 	    if (scsi_dev->hostdata == i2o_dev) {
+		sysfs_remove_link(&i2o_dev->device.kobj, "scsi");
 		scsi_remove_device(scsi_dev);
 		scsi_device_put(scsi_dev);
 		break;
@@ -259,12 +260,14 @@ static int i2o_scsi_probe(struct device *dev)
 	scsi_dev =
 	    __scsi_add_device(i2o_shost->scsi_host, channel, id, lun, i2o_dev);
 
-	if (!scsi_dev) {
+	if (IS_ERR(scsi_dev)) {
 		osm_warn("can not add SCSI device %03x\n",
 			 i2o_dev->lct_data.tid);
-		return -EFAULT;
+		return PTR_ERR(scsi_dev);
 	}
 
+	sysfs_create_link(&i2o_dev->device.kobj, &scsi_dev->sdev_gendev.kobj, "scsi");
+
 	osm_info("device added (TID: %03x) channel: %d, id: %d, lun: %d\n",
 		 i2o_dev->lct_data.tid, channel, id, (unsigned int)lun);
 
@@ -545,7 +548,13 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt,
 	int tid;
 	struct i2o_message __iomem *msg;
 	u32 m;
-	u32 scsi_flags, sg_flags;
+	/*
+	 * ENABLE_DISCONNECT
+	 * SIMPLE_TAG
+	 * RETURN_SENSE_DATA_IN_REPLY_MESSAGE_FRAME
+	 */
+	u32 scsi_flags = 0x20a00000;
+	u32 sg_flags;
 	u32 __iomem *mptr;
 	u32 __iomem *lenptr;
 	u32 len;
@@ -591,17 +600,19 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt,
 
 	switch (SCpnt->sc_data_direction) {
 	case PCI_DMA_NONE:
-		scsi_flags = 0x00000000;	// DATA NO XFER
+		/* DATA NO XFER */
 		sg_flags = 0x00000000;
 		break;
 
 	case PCI_DMA_TODEVICE:
-		scsi_flags = 0x80000000;	// DATA OUT (iop-->dev)
+		/* DATA OUT (iop-->dev) */
+		scsi_flags |= 0x80000000;
 		sg_flags = 0x14000000;
 		break;
 
 	case PCI_DMA_FROMDEVICE:
-		scsi_flags = 0x40000000;	// DATA IN  (iop<--dev)
+		/* DATA IN  (iop<--dev) */
+		scsi_flags |= 0x40000000;
 		sg_flags = 0x10000000;
 		break;
 
@@ -639,8 +650,7 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt,
 	   }
 	 */
 
-	/* Direction, disconnect ok, tag, CDBLen */
-	writel(scsi_flags | 0x20200000 | SCpnt->cmd_len, mptr ++);
+	writel(scsi_flags | SCpnt->cmd_len, mptr++);
 
 	/* Write SCSI command into the message - always 16 byte block */
 	memcpy_toio(mptr, SCpnt->cmnd, 16);
diff --git a/drivers/message/i2o/iop.c b/drivers/message/i2o/iop.c
index 62b0d8bed18..40312053b38 100644
--- a/drivers/message/i2o/iop.c
+++ b/drivers/message/i2o/iop.c
@@ -455,6 +455,70 @@ static int i2o_iop_clear(struct i2o_controller *c)
 	return rc;
 }
 
+/**
+ *	i2o_iop_init_outbound_queue - setup the outbound message queue
+ *	@c: I2O controller
+ *
+ *	Clear and (re)initialize IOP's outbound queue and post the message
+ *	frames to the IOP.
+ *
+ *	Returns 0 on success or a negative errno code on failure.
+ */
+static int i2o_iop_init_outbound_queue(struct i2o_controller *c)
+{
+	u8 *status = c->status.virt;
+	u32 m;
+	struct i2o_message __iomem *msg;
+	ulong timeout;
+	int i;
+
+	osm_debug("%s: Initializing Outbound Queue...\n", c->name);
+
+	memset(status, 0, 4);
+
+	m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET);
+	if (m == I2O_QUEUE_EMPTY)
+		return -ETIMEDOUT;
+
+	writel(EIGHT_WORD_MSG_SIZE | TRL_OFFSET_6, &msg->u.head[0]);
+	writel(I2O_CMD_OUTBOUND_INIT << 24 | HOST_TID << 12 | ADAPTER_TID,
+	       &msg->u.head[1]);
+	writel(i2o_exec_driver.context, &msg->u.s.icntxt);
+	writel(0x0106, &msg->u.s.tcntxt);	/* FIXME: why 0x0106, maybe in
+						   Spec? */
+	writel(PAGE_SIZE, &msg->body[0]);
+	/* Outbound msg frame size in words and Initcode */
+	writel(MSG_FRAME_SIZE << 16 | 0x80, &msg->body[1]);
+	writel(0xd0000004, &msg->body[2]);
+	writel(i2o_dma_low(c->status.phys), &msg->body[3]);
+	writel(i2o_dma_high(c->status.phys), &msg->body[4]);
+
+	i2o_msg_post(c, m);
+
+	timeout = jiffies + I2O_TIMEOUT_INIT_OUTBOUND_QUEUE * HZ;
+	while (*status <= I2O_CMD_IN_PROGRESS) {
+		if (time_after(jiffies, timeout)) {
+			osm_warn("%s: Timeout Initializing\n", c->name);
+			return -ETIMEDOUT;
+		}
+		set_current_state(TASK_UNINTERRUPTIBLE);
+		schedule_timeout(1);
+
+		rmb();
+	}
+
+	m = c->out_queue.phys;
+
+	/* Post frames */
+	for (i = 0; i < NMBR_MSG_FRAMES; i++) {
+		i2o_flush_reply(c, m);
+		udelay(1);	/* Promise */
+		m += MSG_FRAME_SIZE * 4;
+	}
+
+	return 0;
+}
+
 /**
  *	i2o_iop_reset - reset an I2O controller
  *	@c: controller to reset
@@ -491,25 +555,16 @@ static int i2o_iop_reset(struct i2o_controller *c)
 	writel(0, &msg->u.s.tcntxt);	//FIXME: use reasonable transaction context
 	writel(0, &msg->body[0]);
 	writel(0, &msg->body[1]);
-	writel(i2o_ptr_low((void *)c->status.phys), &msg->body[2]);
-	writel(i2o_ptr_high((void *)c->status.phys), &msg->body[3]);
+	writel(i2o_dma_low(c->status.phys), &msg->body[2]);
+	writel(i2o_dma_high(c->status.phys), &msg->body[3]);
 
 	i2o_msg_post(c, m);
 
 	/* Wait for a reply */
 	timeout = jiffies + I2O_TIMEOUT_RESET * HZ;
 	while (!*status) {
-		if (time_after(jiffies, timeout)) {
-			printk(KERN_ERR "%s: IOP reset timeout.\n", c->name);
-			rc = -ETIMEDOUT;
-			goto exit;
-		}
-
-		/* Promise bug */
-		if (status[1] || status[4]) {
-			*status = 0;
+		if (time_after(jiffies, timeout))
 			break;
-		}
 
 		set_current_state(TASK_UNINTERRUPTIBLE);
 		schedule_timeout(1);
@@ -517,14 +572,20 @@ static int i2o_iop_reset(struct i2o_controller *c)
 		rmb();
 	}
 
-	if (*status == I2O_CMD_IN_PROGRESS) {
+	switch (*status) {
+	case I2O_CMD_REJECTED:
+		osm_warn("%s: IOP reset rejected\n", c->name);
+		rc = -EPERM;
+		break;
+
+	case I2O_CMD_IN_PROGRESS:
 		/*
 		 * Once the reset is sent, the IOP goes into the INIT state
-		 * which is indeterminate.  We need to wait until the IOP
-		 * has rebooted before we can let the system talk to
-		 * it. We read the inbound Free_List until a message is
-		 * available. If we can't read one in the given ammount of
-		 * time, we assume the IOP could not reboot properly.
+		 * which is indeterminate. We need to wait until the IOP has
+		 * rebooted before we can let the system talk to it. We read
+		 * the inbound Free_List until a message is available. If we
+		 * can't read one in the given ammount of time, we assume the
+		 * IOP could not reboot properly.
 		 */
 		pr_debug("%s: Reset in progress, waiting for reboot...\n",
 			 c->name);
@@ -543,19 +604,26 @@ static int i2o_iop_reset(struct i2o_controller *c)
 			m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_RESET);
 		}
 		i2o_msg_nop(c, m);
-	}
 
-	/* from here all quiesce commands are safe */
-	c->no_quiesce = 0;
+		/* from here all quiesce commands are safe */
+		c->no_quiesce = 0;
 
-	/* If IopReset was rejected or didn't perform reset, try IopClear */
-	i2o_status_get(c);
-	if (*status == I2O_CMD_REJECTED || sb->iop_state != ADAPTER_STATE_RESET) {
-		printk(KERN_WARNING "%s: Reset rejected, trying to clear\n",
-		       c->name);
-		i2o_iop_clear(c);
-	} else
-		pr_debug("%s: Reset completed.\n", c->name);
+		/* verify if controller is in state RESET */
+		i2o_status_get(c);
+
+		if (!c->promise && (sb->iop_state != ADAPTER_STATE_RESET))
+			osm_warn("%s: reset completed, but adapter not in RESET"
+				 " state.\n", c->name);
+		else
+			osm_debug("%s: reset completed.\n", c->name);
+
+		break;
+
+	default:
+		osm_err("%s: IOP reset timeout.\n", c->name);
+		rc = -ETIMEDOUT;
+		break;
+	}
 
       exit:
 	/* Enable all IOPs */
@@ -564,87 +632,6 @@ static int i2o_iop_reset(struct i2o_controller *c)
 	return rc;
 };
 
-/**
- *	i2o_iop_init_outbound_queue - setup the outbound message queue
- *	@c: I2O controller
- *
- *	Clear and (re)initialize IOP's outbound queue and post the message
- *	frames to the IOP.
- *
- *	Returns 0 on success or a negative errno code on failure.
- */
-static int i2o_iop_init_outbound_queue(struct i2o_controller *c)
-{
-	u8 *status = c->status.virt;
-	u32 m;
-	struct i2o_message __iomem *msg;
-	ulong timeout;
-	int i;
-
-	pr_debug("%s: Initializing Outbound Queue...\n", c->name);
-
-	memset(status, 0, 4);
-
-	m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET);
-	if (m == I2O_QUEUE_EMPTY)
-		return -ETIMEDOUT;
-
-	writel(EIGHT_WORD_MSG_SIZE | SGL_OFFSET_6, &msg->u.head[0]);
-	writel(I2O_CMD_OUTBOUND_INIT << 24 | HOST_TID << 12 | ADAPTER_TID,
-	       &msg->u.head[1]);
-	writel(i2o_exec_driver.context, &msg->u.s.icntxt);
-	writel(0x00000000, &msg->u.s.tcntxt);
-	writel(PAGE_SIZE, &msg->body[0]);
-	writel(MSG_FRAME_SIZE << 16 | 0x80, &msg->body[1]);	/* Outbound msg frame
-								   size in words and Initcode */
-	writel(0xd0000004, &msg->body[2]);
-	writel(i2o_ptr_low((void *)c->status.phys), &msg->body[3]);
-	writel(i2o_ptr_high((void *)c->status.phys), &msg->body[4]);
-
-	i2o_msg_post(c, m);
-
-	timeout = jiffies + I2O_TIMEOUT_INIT_OUTBOUND_QUEUE * HZ;
-	while (*status <= I2O_CMD_IN_PROGRESS) {
-		if (time_after(jiffies, timeout)) {
-			printk(KERN_WARNING "%s: Timeout Initializing\n",
-			       c->name);
-			return -ETIMEDOUT;
-		}
-		set_current_state(TASK_UNINTERRUPTIBLE);
-		schedule_timeout(1);
-
-		rmb();
-	}
-
-	m = c->out_queue.phys;
-
-	/* Post frames */
-	for (i = 0; i < NMBR_MSG_FRAMES; i++) {
-		i2o_flush_reply(c, m);
-		udelay(1);	/* Promise */
-		m += MSG_FRAME_SIZE * 4;
-	}
-
-	return 0;
-}
-
-/**
- *	i2o_iop_send_nop - send a core NOP message
- *	@c: controller
- *
- *	Send a no-operation message with a reply set to cause no
- *	action either. Needed for bringing up promise controllers.
- */
-static int i2o_iop_send_nop(struct i2o_controller *c)
-{
-	struct i2o_message __iomem *msg;
-	u32 m = i2o_msg_get_wait(c, &msg, HZ);
-	if (m == I2O_QUEUE_EMPTY)
-		return -ETIMEDOUT;
-	i2o_msg_nop(c, m);
-	return 0;
-}
-
 /**
  *	i2o_iop_activate - Bring controller up to HOLD
  *	@c: controller
@@ -656,26 +643,9 @@ static int i2o_iop_send_nop(struct i2o_controller *c)
  */
 static int i2o_iop_activate(struct i2o_controller *c)
 {
-	struct pci_dev *i960 = NULL;
 	i2o_status_block *sb = c->status_block.virt;
 	int rc;
-
-	if (c->promise) {
-		/* Beat up the hardware first of all */
-		i960 =
-		    pci_find_slot(c->pdev->bus->number,
-				  PCI_DEVFN(PCI_SLOT(c->pdev->devfn), 0));
-		if (i960)
-			pci_write_config_word(i960, 0x42, 0);
-
-		/* Follow this sequence precisely or the controller
-		   ceases to perform useful functions until reboot */
-		if ((rc = i2o_iop_send_nop(c)))
-			return rc;
-
-		if ((rc = i2o_iop_reset(c)))
-			return rc;
-	}
+	int state;
 
 	/* In INIT state, Wait Inbound Q to initialize (in i2o_status_get) */
 	/* In READY state, Get status */
@@ -684,7 +654,8 @@ static int i2o_iop_activate(struct i2o_controller *c)
 	if (rc) {
 		printk(KERN_INFO "%s: Unable to obtain status, "
 		       "attempting a reset.\n", c->name);
-		if (i2o_iop_reset(c))
+		rc = i2o_iop_reset(c);
+		if (rc)
 			return rc;
 	}
 
@@ -697,37 +668,37 @@ static int i2o_iop_activate(struct i2o_controller *c)
 	switch (sb->iop_state) {
 	case ADAPTER_STATE_FAULTED:
 		printk(KERN_CRIT "%s: hardware fault\n", c->name);
-		return -ENODEV;
+		return -EFAULT;
 
 	case ADAPTER_STATE_READY:
 	case ADAPTER_STATE_OPERATIONAL:
 	case ADAPTER_STATE_HOLD:
 	case ADAPTER_STATE_FAILED:
 		pr_debug("%s: already running, trying to reset...\n", c->name);
-		if (i2o_iop_reset(c))
-			return -ENODEV;
+		rc = i2o_iop_reset(c);
+		if (rc)
+			return rc;
 	}
 
+	/* preserve state */
+	state = sb->iop_state;
+
 	rc = i2o_iop_init_outbound_queue(c);
 	if (rc)
 		return rc;
 
-	if (c->promise) {
-		if ((rc = i2o_iop_send_nop(c)))
-			return rc;
+	/* if adapter was not in RESET state clear now */
+	if (state != ADAPTER_STATE_RESET)
+		i2o_iop_clear(c);
 
-		if ((rc = i2o_status_get(c)))
-			return rc;
+	i2o_status_get(c);
 
-		if (i960)
-			pci_write_config_word(i960, 0x42, 0x3FF);
+	if (sb->iop_state != ADAPTER_STATE_HOLD) {
+		osm_err("%s: failed to bring IOP into HOLD state\n", c->name);
+		return -EIO;
 	}
 
-	/* In HOLD state */
-
-	rc = i2o_hrt_get(c);
-
-	return rc;
+	return i2o_hrt_get(c);
 };
 
 /**
@@ -1030,8 +1001,8 @@ int i2o_status_get(struct i2o_controller *c)
 	writel(0, &msg->u.s.tcntxt);	// FIXME: use resonable transaction context
 	writel(0, &msg->body[0]);
 	writel(0, &msg->body[1]);
-	writel(i2o_ptr_low((void *)c->status_block.phys), &msg->body[2]);
-	writel(i2o_ptr_high((void *)c->status_block.phys), &msg->body[3]);
+	writel(i2o_dma_low(c->status_block.phys), &msg->body[2]);
+	writel(i2o_dma_high(c->status_block.phys), &msg->body[3]);
 	writel(sizeof(i2o_status_block), &msg->body[4]);	/* always 88 bytes */
 
 	i2o_msg_post(c, m);
diff --git a/drivers/message/i2o/pci.c b/drivers/message/i2o/pci.c
index f33fd81f77a..a499af096a6 100644
--- a/drivers/message/i2o/pci.c
+++ b/drivers/message/i2o/pci.c
@@ -49,30 +49,6 @@ static struct pci_device_id __devinitdata i2o_pci_ids[] = {
 	{0}
 };
 
-/**
- *	i2o_dma_realloc - Realloc DMA memory
- *	@dev: struct device pointer to the PCI device of the I2O controller
- *	@addr: pointer to a i2o_dma struct DMA buffer
- *	@len: new length of memory
- *	@gfp_mask: GFP mask
- *
- *	If there was something allocated in the addr, free it first. If len > 0
- *	than try to allocate it and write the addresses back to the addr
- *	structure. If len == 0 set the virtual address to NULL.
- *
- *	Returns the 0 on success or negative error code on failure.
- */
-int i2o_dma_realloc(struct device *dev, struct i2o_dma *addr, size_t len,
-		    unsigned int gfp_mask)
-{
-	i2o_dma_free(dev, addr);
-
-	if (len)
-		return i2o_dma_alloc(dev, addr, len, gfp_mask);
-
-	return 0;
-};
-
 /**
  *	i2o_pci_free - Frees the DMA memory for the I2O controller
  *	@c: I2O controller to free
@@ -185,6 +161,7 @@ static int __devinit i2o_pci_alloc(struct i2o_controller *c)
 	} else
 		c->in_queue = c->base;
 
+	c->irq_status = c->base.virt + I2O_IRQ_STATUS;
 	c->irq_mask = c->base.virt + I2O_IRQ_MASK;
 	c->in_port = c->base.virt + I2O_IN_PORT;
 	c->out_port = c->base.virt + I2O_OUT_PORT;
@@ -232,36 +209,30 @@ static int __devinit i2o_pci_alloc(struct i2o_controller *c)
 static irqreturn_t i2o_pci_interrupt(int irq, void *dev_id, struct pt_regs *r)
 {
 	struct i2o_controller *c = dev_id;
-	struct device *dev = &c->pdev->dev;
-	u32 mv = readl(c->out_port);
-
-	/*
-	 * Old 960 steppings had a bug in the I2O unit that caused
-	 * the queue to appear empty when it wasn't.
-	 */
-	if (mv == I2O_QUEUE_EMPTY) {
-		mv = readl(c->out_port);
-		if (unlikely(mv == I2O_QUEUE_EMPTY))
-			return IRQ_NONE;
-		else
-			pr_debug("%s: 960 bug detected\n", c->name);
-	}
+	u32 m;
+	irqreturn_t rc = IRQ_NONE;
+
+	while (readl(c->irq_status) & I2O_IRQ_OUTBOUND_POST) {
+		m = readl(c->out_port);
+		if (m == I2O_QUEUE_EMPTY) {
+			/*
+			 * Old 960 steppings had a bug in the I2O unit that
+			 * caused the queue to appear empty when it wasn't.
+			 */
+			m = readl(c->out_port);
+			if (unlikely(m == I2O_QUEUE_EMPTY))
+				break;
+		}
 
-	while (mv != I2O_QUEUE_EMPTY) {
 		/* dispatch it */
-		if (i2o_driver_dispatch(c, mv))
+		if (i2o_driver_dispatch(c, m))
 			/* flush it if result != 0 */
-			i2o_flush_reply(c, mv);
+			i2o_flush_reply(c, m);
 
-		/*
-		 * That 960 bug again...
-		 */
-		mv = readl(c->out_port);
-		if (mv == I2O_QUEUE_EMPTY)
-			mv = readl(c->out_port);
+		rc = IRQ_HANDLED;
 	}
 
-	return IRQ_HANDLED;
+	return rc;
 }
 
 /**
diff --git a/include/linux/i2o-dev.h b/include/linux/i2o-dev.h
index 3414325bdcf..90c984ecd52 100644
--- a/include/linux/i2o-dev.h
+++ b/include/linux/i2o-dev.h
@@ -32,6 +32,10 @@ typedef unsigned int u32;
 
 #endif				/* __KERNEL__ */
 
+/*
+ *	Vendors
+ */
+#define I2O_VENDOR_DPT				0x001b
 
 /*
  * I2O Control IOCTLs and structures
@@ -333,7 +337,7 @@ typedef struct _i2o_status_block {
 #define I2O_CLASS_ATE_PERIPHERAL		0x061
 #define I2O_CLASS_FLOPPY_CONTROLLER		0x070
 #define I2O_CLASS_FLOPPY_DEVICE 		0x071
-#define I2O_CLASS_BUS_ADAPTER_PORT		0x080
+#define I2O_CLASS_BUS_ADAPTER			0x080
 #define I2O_CLASS_PEER_TRANSPORT_AGENT		0x090
 #define I2O_CLASS_PEER_TRANSPORT		0x091
 #define	I2O_CLASS_END				0xfff
diff --git a/include/linux/i2o.h b/include/linux/i2o.h
index e8cd1129001..497ea574f96 100644
--- a/include/linux/i2o.h
+++ b/include/linux/i2o.h
@@ -157,7 +157,8 @@ struct i2o_controller {
 
 	void __iomem *in_port;	/* Inbout port address */
 	void __iomem *out_port;	/* Outbound port address */
-	void __iomem *irq_mask;		/* Interrupt register address */
+	void __iomem *irq_status;	/* Interrupt status register address */
+	void __iomem *irq_mask;	/* Interrupt mask register address */
 
 	/* Dynamic LCT related data */
 
@@ -242,15 +243,6 @@ extern int i2o_msg_post_wait_mem(struct i2o_controller *, u32, unsigned long,
 extern void i2o_msg_nop(struct i2o_controller *, u32);
 static inline void i2o_flush_reply(struct i2o_controller *, u32);
 
-/* DMA handling functions */
-static inline int i2o_dma_alloc(struct device *, struct i2o_dma *, size_t,
-				unsigned int);
-static inline void i2o_dma_free(struct device *, struct i2o_dma *);
-int i2o_dma_realloc(struct device *, struct i2o_dma *, size_t, unsigned int);
-
-static inline int i2o_dma_map(struct device *, struct i2o_dma *);
-static inline void i2o_dma_unmap(struct device *, struct i2o_dma *);
-
 /* IOP functions */
 extern int i2o_status_get(struct i2o_controller *);
 
@@ -275,6 +267,16 @@ static inline u32 i2o_ptr_high(void *ptr)
 {
 	return (u32) ((u64) ptr >> 32);
 };
+
+static inline u32 i2o_dma_low(dma_addr_t dma_addr)
+{
+	return (u32) (u64) dma_addr;
+};
+
+static inline u32 i2o_dma_high(dma_addr_t dma_addr)
+{
+	return (u32) ((u64) dma_addr >> 32);
+};
 #else
 static inline u32 i2o_cntxt_list_add(struct i2o_controller *c, void *ptr)
 {
@@ -305,8 +307,246 @@ static inline u32 i2o_ptr_high(void *ptr)
 {
 	return 0;
 };
+
+static inline u32 i2o_dma_low(dma_addr_t dma_addr)
+{
+	return (u32) dma_addr;
+};
+
+static inline u32 i2o_dma_high(dma_addr_t dma_addr)
+{
+	return 0;
+};
+#endif
+
+/**
+ *	i2o_sg_tablesize - Calculate the maximum number of elements in a SGL
+ *	@c: I2O controller for which the calculation should be done
+ *	@body_size: maximum body size used for message in 32-bit words.
+ *
+ *	Return the maximum number of SG elements in a SG list.
+ */
+static inline u16 i2o_sg_tablesize(struct i2o_controller *c, u16 body_size)
+{
+	i2o_status_block *sb = c->status_block.virt;
+	u16 sg_count =
+	    (sb->inbound_frame_size - sizeof(struct i2o_message) / 4) -
+	    body_size;
+
+	if (c->pae_support) {
+		/*
+		 * for 64-bit a SG attribute element must be added and each
+		 * SG element needs 12 bytes instead of 8.
+		 */
+		sg_count -= 2;
+		sg_count /= 3;
+	} else
+		sg_count /= 2;
+
+	if (c->short_req && (sg_count > 8))
+		sg_count = 8;
+
+	return sg_count;
+};
+
+/**
+ *	i2o_dma_map_single - Map pointer to controller and fill in I2O message.
+ *	@c: I2O controller
+ *	@ptr: pointer to the data which should be mapped
+ *	@size: size of data in bytes
+ *	@direction: DMA_TO_DEVICE / DMA_FROM_DEVICE
+ *	@sg_ptr: pointer to the SG list inside the I2O message
+ *
+ *	This function does all necessary DMA handling and also writes the I2O
+ *	SGL elements into the I2O message. For details on DMA handling see also
+ *	dma_map_single(). The pointer sg_ptr will only be set to the end of the
+ *	SG list if the allocation was successful.
+ *
+ *	Returns DMA address which must be checked for failures using
+ *	dma_mapping_error().
+ */
+static inline dma_addr_t i2o_dma_map_single(struct i2o_controller *c, void *ptr,
+					    size_t size,
+					    enum dma_data_direction direction,
+					    u32 __iomem ** sg_ptr)
+{
+	u32 sg_flags;
+	u32 __iomem *mptr = *sg_ptr;
+	dma_addr_t dma_addr;
+
+	switch (direction) {
+	case DMA_TO_DEVICE:
+		sg_flags = 0xd4000000;
+		break;
+	case DMA_FROM_DEVICE:
+		sg_flags = 0xd0000000;
+		break;
+	default:
+		return 0;
+	}
+
+	dma_addr = dma_map_single(&c->pdev->dev, ptr, size, direction);
+	if (!dma_mapping_error(dma_addr)) {
+#ifdef CONFIG_I2O_EXT_ADAPTEC_DMA64
+		if ((sizeof(dma_addr_t) > 4) && c->pae_support) {
+			writel(0x7C020002, mptr++);
+			writel(PAGE_SIZE, mptr++);
+		}
+#endif
+
+		writel(sg_flags | size, mptr++);
+		writel(i2o_dma_low(dma_addr), mptr++);
+#ifdef CONFIG_I2O_EXT_ADAPTEC_DMA64
+		if ((sizeof(dma_addr_t) > 4) && c->pae_support)
+			writel(i2o_dma_high(dma_addr), mptr++);
+#endif
+		*sg_ptr = mptr;
+	}
+	return dma_addr;
+};
+
+/**
+ *	i2o_dma_map_sg - Map a SG List to controller and fill in I2O message.
+ *	@c: I2O controller
+ *	@sg: SG list to be mapped
+ *	@sg_count: number of elements in the SG list
+ *	@direction: DMA_TO_DEVICE / DMA_FROM_DEVICE
+ *	@sg_ptr: pointer to the SG list inside the I2O message
+ *
+ *	This function does all necessary DMA handling and also writes the I2O
+ *	SGL elements into the I2O message. For details on DMA handling see also
+ *	dma_map_sg(). The pointer sg_ptr will only be set to the end of the SG
+ *	list if the allocation was successful.
+ *
+ *	Returns 0 on failure or 1 on success.
+ */
+static inline int i2o_dma_map_sg(struct i2o_controller *c,
+				 struct scatterlist *sg, int sg_count,
+				 enum dma_data_direction direction,
+				 u32 __iomem ** sg_ptr)
+{
+	u32 sg_flags;
+	u32 __iomem *mptr = *sg_ptr;
+
+	switch (direction) {
+	case DMA_TO_DEVICE:
+		sg_flags = 0x14000000;
+		break;
+	case DMA_FROM_DEVICE:
+		sg_flags = 0x10000000;
+		break;
+	default:
+		return 0;
+	}
+
+	sg_count = dma_map_sg(&c->pdev->dev, sg, sg_count, direction);
+	if (!sg_count)
+		return 0;
+
+#ifdef CONFIG_I2O_EXT_ADAPTEC_DMA64
+	if ((sizeof(dma_addr_t) > 4) && c->pae_support) {
+		writel(0x7C020002, mptr++);
+		writel(PAGE_SIZE, mptr++);
+	}
 #endif
 
+	while (sg_count-- > 0) {
+		if (!sg_count)
+			sg_flags |= 0xC0000000;
+		writel(sg_flags | sg_dma_len(sg), mptr++);
+		writel(i2o_dma_low(sg_dma_address(sg)), mptr++);
+#ifdef CONFIG_I2O_EXT_ADAPTEC_DMA64
+		if ((sizeof(dma_addr_t) > 4) && c->pae_support)
+			writel(i2o_dma_high(sg_dma_address(sg)), mptr++);
+#endif
+		sg++;
+	}
+	*sg_ptr = mptr;
+
+	return 1;
+};
+
+/**
+ *	i2o_dma_alloc - Allocate DMA memory
+ *	@dev: struct device pointer to the PCI device of the I2O controller
+ *	@addr: i2o_dma struct which should get the DMA buffer
+ *	@len: length of the new DMA memory
+ *	@gfp_mask: GFP mask
+ *
+ *	Allocate a coherent DMA memory and write the pointers into addr.
+ *
+ *	Returns 0 on success or -ENOMEM on failure.
+ */
+static inline int i2o_dma_alloc(struct device *dev, struct i2o_dma *addr,
+				size_t len, unsigned int gfp_mask)
+{
+	struct pci_dev *pdev = to_pci_dev(dev);
+	int dma_64 = 0;
+
+	if ((sizeof(dma_addr_t) > 4) && (pdev->dma_mask == DMA_64BIT_MASK)) {
+		dma_64 = 1;
+		if (pci_set_dma_mask(pdev, DMA_32BIT_MASK))
+			return -ENOMEM;
+	}
+
+	addr->virt = dma_alloc_coherent(dev, len, &addr->phys, gfp_mask);
+
+	if ((sizeof(dma_addr_t) > 4) && dma_64)
+		if (pci_set_dma_mask(pdev, DMA_64BIT_MASK))
+			printk(KERN_WARNING "i2o: unable to set 64-bit DMA");
+
+	if (!addr->virt)
+		return -ENOMEM;
+
+	memset(addr->virt, 0, len);
+	addr->len = len;
+
+	return 0;
+};
+
+/**
+ *	i2o_dma_free - Free DMA memory
+ *	@dev: struct device pointer to the PCI device of the I2O controller
+ *	@addr: i2o_dma struct which contains the DMA buffer
+ *
+ *	Free a coherent DMA memory and set virtual address of addr to NULL.
+ */
+static inline void i2o_dma_free(struct device *dev, struct i2o_dma *addr)
+{
+	if (addr->virt) {
+		if (addr->phys)
+			dma_free_coherent(dev, addr->len, addr->virt,
+					  addr->phys);
+		else
+			kfree(addr->virt);
+		addr->virt = NULL;
+	}
+};
+
+/**
+ *	i2o_dma_realloc - Realloc DMA memory
+ *	@dev: struct device pointer to the PCI device of the I2O controller
+ *	@addr: pointer to a i2o_dma struct DMA buffer
+ *	@len: new length of memory
+ *	@gfp_mask: GFP mask
+ *
+ *	If there was something allocated in the addr, free it first. If len > 0
+ *	than try to allocate it and write the addresses back to the addr
+ *	structure. If len == 0 set the virtual address to NULL.
+ *
+ *	Returns the 0 on success or negative error code on failure.
+ */
+static inline int i2o_dma_realloc(struct device *dev, struct i2o_dma *addr,
+				  size_t len, unsigned int gfp_mask)
+{
+	i2o_dma_free(dev, addr);
+
+	if (len)
+		return i2o_dma_alloc(dev, addr, len, gfp_mask);
+
+	return 0;
+};
+
 /* I2O driver (OSM) functions */
 extern int i2o_driver_register(struct i2o_driver *);
 extern void i2o_driver_unregister(struct i2o_driver *);
@@ -375,10 +615,11 @@ extern int i2o_device_claim_release(struct i2o_device *);
 /* Exec OSM functions */
 extern int i2o_exec_lct_get(struct i2o_controller *);
 
-/* device / driver conversion functions */
+/* device / driver / kobject conversion functions */
 #define to_i2o_driver(drv) container_of(drv,struct i2o_driver, driver)
 #define to_i2o_device(dev) container_of(dev, struct i2o_device, device)
 #define to_i2o_controller(dev) container_of(dev, struct i2o_controller, device)
+#define kobj_to_i2o_device(kobj) to_i2o_device(container_of(kobj, struct device, kobj))
 
 /**
  *	i2o_msg_get - obtain an I2O message from the IOP
@@ -466,8 +707,10 @@ static inline struct i2o_message __iomem *i2o_msg_out_to_virt(struct
 							      i2o_controller *c,
 							      u32 m)
 {
-	BUG_ON(m < c->out_queue.phys
-	       || m >= c->out_queue.phys + c->out_queue.len);
+	if (unlikely
+	    (m < c->out_queue.phys
+	     || m >= c->out_queue.phys + c->out_queue.len))
+		return NULL;
 
 	return c->out_queue.virt + (m - c->out_queue.phys);
 };
@@ -532,48 +775,6 @@ static inline void i2o_dma_free(struct device *dev, struct i2o_dma *addr)
 	}
 };
 
-/**
- *	i2o_dma_map - Map the memory to DMA
- *	@dev: struct device pointer to the PCI device of the I2O controller
- *	@addr: i2o_dma struct which should be mapped
- *
- *	Map the memory in addr->virt to coherent DMA memory and write the
- *	physical address into addr->phys.
- *
- *	Returns 0 on success or -ENOMEM on failure.
- */
-static inline int i2o_dma_map(struct device *dev, struct i2o_dma *addr)
-{
-	if (!addr->virt)
-		return -EFAULT;
-
-	if (!addr->phys)
-		addr->phys = dma_map_single(dev, addr->virt, addr->len,
-					    DMA_BIDIRECTIONAL);
-	if (!addr->phys)
-		return -ENOMEM;
-
-	return 0;
-};
-
-/**
- *	i2o_dma_unmap - Unmap the DMA memory
- *	@dev: struct device pointer to the PCI device of the I2O controller
- *	@addr: i2o_dma struct which should be unmapped
- *
- *	Unmap the memory in addr->virt from DMA memory.
- */
-static inline void i2o_dma_unmap(struct device *dev, struct i2o_dma *addr)
-{
-	if (!addr->virt)
-		return;
-
-	if (addr->phys) {
-		dma_unmap_single(dev, addr->phys, addr->len, DMA_BIDIRECTIONAL);
-		addr->phys = 0;
-	}
-};
-
 /*
  *	Endian handling wrapped into the macro - keeps the core code
  *	cleaner.
@@ -725,6 +926,14 @@ extern void i2o_debug_state(struct i2o_controller *c);
 #define I2O_CMD_SCSI_ABORT		0x83
 #define I2O_CMD_SCSI_BUSRESET		0x27
 
+/*
+ * Bus Adapter Class
+ */
+#define I2O_CMD_BUS_ADAPTER_RESET	0x85
+#define I2O_CMD_BUS_RESET		0x87
+#define I2O_CMD_BUS_SCAN		0x89
+#define I2O_CMD_BUS_QUIESCE		0x8b
+
 /*
  * Random Block Storage Class
  */
@@ -948,7 +1157,7 @@ extern void i2o_debug_state(struct i2o_controller *c);
 
 /* request queue sizes */
 #define I2O_MAX_SECTORS			1024
-#define I2O_MAX_SEGMENTS		128
+#define I2O_MAX_PHYS_SEGMENTS		MAX_PHYS_SEGMENTS
 
 #define I2O_REQ_MEMPOOL_SIZE		32
 
-- 
cgit v1.2.3-70-g09d2


From b2aaee33fbb354a2f08121aa1c1be55841102761 Mon Sep 17 00:00:00 2001
From: Markus Lidel <Markus.Lidel@shadowconnect.com>
Date: Thu, 23 Jun 2005 22:02:19 -0700
Subject: [PATCH] I2O: Adaptec specific SG_IO access, firmware access through
 sysfs and 2400A workaround

Changes:
 - Provide SG_IO access to BLOCK and EXECUTIVE class on Adaptec
   controllers
 - Use PRIVATE messages in SCSI-OSM because on some controllers normal
   SCSI class commands like READ or READ CAPACITY cause errors
 - Use new DMA and SG list creation function
 - Added workaround to limit sectors per request for Adaptec 2400A
   controllers

Signed-off-by: Markus Lidel <Markus.Lidel@shadowconnect.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/message/i2o/Kconfig      |  18 +++
 drivers/message/i2o/i2o_block.h  |   6 +
 drivers/message/i2o/i2o_config.c |   4 +
 drivers/message/i2o/i2o_scsi.c   | 263 +++++++++++++++++++++++----------------
 drivers/message/i2o/pci.c        |  22 ++++
 include/linux/i2o-dev.h          |  22 ++++
 include/linux/i2o.h              |  37 ++++--
 include/scsi/sg_request.h        |  26 ++++
 8 files changed, 282 insertions(+), 116 deletions(-)
 create mode 100644 include/scsi/sg_request.h

(limited to 'include')

diff --git a/drivers/message/i2o/Kconfig b/drivers/message/i2o/Kconfig
index ce278e060ac..94b6d676c5c 100644
--- a/drivers/message/i2o/Kconfig
+++ b/drivers/message/i2o/Kconfig
@@ -24,6 +24,24 @@ config I2O
 
 	  If unsure, say N.
 
+config I2O_EXT_ADAPTEC
+	bool "Enable Adaptec extensions"
+	depends on I2O
+	default y
+	---help---
+	  Say Y for support of raidutils for Adaptec I2O controllers. You also
+	  have to say Y to "I2O Configuration support", "I2O SCSI OSM" below
+	  and to "SCSI generic support" under "SCSI device configuration".
+
+config I2O_EXT_ADAPTEC_DMA64
+	bool "Enable 64-bit DMA"
+	depends on I2O_EXT_ADAPTEC && ( 64BIT || HIGHMEM64G )
+	default y
+	---help---
+	  Say Y for support of 64-bit DMA transfer mode on Adaptec I2O
+	  controllers.
+	  Note: You need at least firmware version 3709.
+
 config I2O_CONFIG
 	tristate "I2O Configuration support"
 	depends on PCI && I2O
diff --git a/drivers/message/i2o/i2o_block.h b/drivers/message/i2o/i2o_block.h
index 9e1a95fb083..e45cc40ce38 100644
--- a/drivers/message/i2o/i2o_block.h
+++ b/drivers/message/i2o/i2o_block.h
@@ -56,6 +56,12 @@
 #define I2O_BLOCK_RETRY_TIME HZ/4
 #define I2O_BLOCK_MAX_OPEN_REQUESTS 50
 
+/* request queue sizes */
+#define I2O_BLOCK_REQ_MEMPOOL_SIZE		32
+
+#define KERNEL_SECTOR_SHIFT 9
+#define KERNEL_SECTOR_SIZE (1 << KERNEL_SECTOR_SHIFT)
+
 /* I2O Block OSM mempool struct */
 struct i2o_block_mempool {
 	kmem_cache_t	*slab;
diff --git a/drivers/message/i2o/i2o_config.c b/drivers/message/i2o/i2o_config.c
index 849d90aad77..7636833b462 100644
--- a/drivers/message/i2o/i2o_config.c
+++ b/drivers/message/i2o/i2o_config.c
@@ -515,6 +515,7 @@ static int i2o_cfg_evt_get(unsigned long arg, struct file *fp)
 	return 0;
 }
 
+#ifdef CONFIG_I2O_EXT_ADAPTEC
 #ifdef CONFIG_COMPAT
 static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long arg)
 {
@@ -964,6 +965,7 @@ static int i2o_cfg_passthru(unsigned long arg)
 	kfree(reply);
 	return rcode;
 }
+#endif
 
 /*
  * IOCTL Handler
@@ -1018,9 +1020,11 @@ static int i2o_cfg_ioctl(struct inode *inode, struct file *fp, unsigned int cmd,
 		ret = i2o_cfg_evt_get(arg, fp);
 		break;
 
+#ifdef CONFIG_I2O_EXT_ADAPTEC
 	case I2OPASSTHRU:
 		ret = i2o_cfg_passthru(arg);
 		break;
+#endif
 
 	default:
 		osm_debug("unknown ioctl called!\n");
diff --git a/drivers/message/i2o/i2o_scsi.c b/drivers/message/i2o/i2o_scsi.c
index c3b0c29ac02..fef53b509a6 100644
--- a/drivers/message/i2o/i2o_scsi.c
+++ b/drivers/message/i2o/i2o_scsi.c
@@ -55,6 +55,7 @@
 #include <linux/pci.h>
 #include <linux/blkdev.h>
 #include <linux/i2o.h>
+#include <linux/scatterlist.h>
 
 #include <asm/dma.h>
 #include <asm/system.h>
@@ -65,19 +66,23 @@
 #include <scsi/scsi_host.h>
 #include <scsi/scsi_device.h>
 #include <scsi/scsi_cmnd.h>
+#include <scsi/scsi_request.h>
+#include <scsi/sg.h>
+#include <scsi/sg_request.h>
 
 #define OSM_NAME	"scsi-osm"
-#define OSM_VERSION	"$Rev$"
+#define OSM_VERSION	"1.282"
 #define OSM_DESCRIPTION	"I2O SCSI Peripheral OSM"
 
 static struct i2o_driver i2o_scsi_driver;
 
-static int i2o_scsi_max_id = 16;
-static int i2o_scsi_max_lun = 8;
+static unsigned int i2o_scsi_max_id = 16;
+static unsigned int i2o_scsi_max_lun = 255;
 
 struct i2o_scsi_host {
 	struct Scsi_Host *scsi_host;	/* pointer to the SCSI host */
 	struct i2o_controller *iop;	/* pointer to the I2O controller */
+	unsigned int lun;	/* lun's used for block devices */
 	struct i2o_device *channel[0];	/* channel->i2o_dev mapping table */
 };
 
@@ -100,12 +105,17 @@ static struct i2o_scsi_host *i2o_scsi_host_alloc(struct i2o_controller *c)
 	u8 type;
 	int i;
 	size_t size;
-	i2o_status_block *sb;
+	u16 body_size = 6;
+
+#ifdef CONFIG_I2O_EXT_ADAPTEC
+	if (c->adaptec)
+		body_size = 8;
+#endif
 
 	list_for_each_entry(i2o_dev, &c->devices, list)
 	    if (i2o_dev->lct_data.class_id == I2O_CLASS_BUS_ADAPTER) {
 		if (i2o_parm_field_get(i2o_dev, 0x0000, 0, &type, 1)
-		   && (type == 0x01))	/* SCSI bus */
+		    && (type == 0x01))	/* SCSI bus */
 			max_channel++;
 	}
 
@@ -127,20 +137,18 @@ static struct i2o_scsi_host *i2o_scsi_host_alloc(struct i2o_controller *c)
 	scsi_host->max_id = i2o_scsi_max_id;
 	scsi_host->max_lun = i2o_scsi_max_lun;
 	scsi_host->this_id = c->unit;
-
-	sb = c->status_block.virt;
-
-	scsi_host->sg_tablesize = (sb->inbound_frame_size -
-				   sizeof(struct i2o_message) / 4 - 6) / 2;
+	scsi_host->sg_tablesize = i2o_sg_tablesize(c, body_size);
 
 	i2o_shost = (struct i2o_scsi_host *)scsi_host->hostdata;
 	i2o_shost->scsi_host = scsi_host;
 	i2o_shost->iop = c;
+	i2o_shost->lun = 1;
 
 	i = 0;
 	list_for_each_entry(i2o_dev, &c->devices, list)
 	    if (i2o_dev->lct_data.class_id == I2O_CLASS_BUS_ADAPTER) {
-		if (i2o_parm_field_get(i2o_dev, 0x0000, 0, &type, 1) || (type == 1))	/* only SCSI bus */
+		if (i2o_parm_field_get(i2o_dev, 0x0000, 0, &type, 1)
+		    && (type == 0x01))	/* only SCSI bus */
 			i2o_shost->channel[i++] = i2o_dev;
 
 		if (i >= max_channel)
@@ -212,8 +220,8 @@ static int i2o_scsi_probe(struct device *dev)
 	struct Scsi_Host *scsi_host;
 	struct i2o_device *parent;
 	struct scsi_device *scsi_dev;
-	u32 id;
-	u64 lun;
+	u32 id = -1;
+	u64 lun = -1;
 	int channel = -1;
 	int i;
 
@@ -223,8 +231,56 @@ static int i2o_scsi_probe(struct device *dev)
 
 	scsi_host = i2o_shost->scsi_host;
 
-	if (i2o_parm_field_get(i2o_dev, 0, 3, &id, 4) < 0)
+	switch (i2o_dev->lct_data.class_id) {
+	case I2O_CLASS_RANDOM_BLOCK_STORAGE:
+	case I2O_CLASS_EXECUTIVE:
+#ifdef CONFIG_I2O_EXT_ADAPTEC
+		if (c->adaptec) {
+			u8 type;
+			struct i2o_device *d = i2o_shost->channel[0];
+
+			if (i2o_parm_field_get(d, 0x0000, 0, &type, 1)
+			    && (type == 0x01))	/* SCSI bus */
+				if (i2o_parm_field_get(d, 0x0200, 4, &id, 4)) {
+					channel = 0;
+					if (i2o_dev->lct_data.class_id ==
+					    I2O_CLASS_RANDOM_BLOCK_STORAGE)
+						lun = i2o_shost->lun++;
+					else
+						lun = 0;
+				}
+		}
+#endif
+		break;
+
+	case I2O_CLASS_SCSI_PERIPHERAL:
+		if (i2o_parm_field_get(i2o_dev, 0x0000, 3, &id, 4) < 0)
+			return -EFAULT;
+
+		if (i2o_parm_field_get(i2o_dev, 0x0000, 4, &lun, 8) < 0)
+			return -EFAULT;
+
+		parent = i2o_iop_find_device(c, i2o_dev->lct_data.parent_tid);
+		if (!parent) {
+			osm_warn("can not find parent of device %03x\n",
+				 i2o_dev->lct_data.tid);
+			return -EFAULT;
+		}
+
+		for (i = 0; i <= i2o_shost->scsi_host->max_channel; i++)
+			if (i2o_shost->channel[i] == parent)
+				channel = i;
+		break;
+
+	default:
+		return -EFAULT;
+	}
+
+	if (channel == -1) {
+		osm_warn("can not find channel of device %03x\n",
+			 i2o_dev->lct_data.tid);
 		return -EFAULT;
+	}
 
 	if (id >= scsi_host->max_id) {
 		osm_warn("SCSI device id (%d) >= max_id of I2O host (%d)", id,
@@ -232,31 +288,12 @@ static int i2o_scsi_probe(struct device *dev)
 		return -EFAULT;
 	}
 
-	if (i2o_parm_field_get(i2o_dev, 0, 4, &lun, 8) < 0)
-		return -EFAULT;
 	if (lun >= scsi_host->max_lun) {
 		osm_warn("SCSI device id (%d) >= max_lun of I2O host (%d)",
 			 (unsigned int)lun, scsi_host->max_lun);
 		return -EFAULT;
 	}
 
-	parent = i2o_iop_find_device(c, i2o_dev->lct_data.parent_tid);
-	if (!parent) {
-		osm_warn("can not find parent of device %03x\n",
-			 i2o_dev->lct_data.tid);
-		return -EFAULT;
-	}
-
-	for (i = 0; i <= i2o_shost->scsi_host->max_channel; i++)
-		if (i2o_shost->channel[i] == parent)
-			channel = i;
-
-	if (channel == -1) {
-		osm_warn("can not find channel of device %03x\n",
-			 i2o_dev->lct_data.tid);
-		return -EFAULT;
-	}
-
 	scsi_dev =
 	    __scsi_add_device(i2o_shost->scsi_host, channel, id, lun, i2o_dev);
 
@@ -266,7 +303,8 @@ static int i2o_scsi_probe(struct device *dev)
 		return PTR_ERR(scsi_dev);
 	}
 
-	sysfs_create_link(&i2o_dev->device.kobj, &scsi_dev->sdev_gendev.kobj, "scsi");
+	sysfs_create_link(&i2o_dev->device.kobj, &scsi_dev->sdev_gendev.kobj,
+			  "scsi");
 
 	osm_info("device added (TID: %03x) channel: %d, id: %d, lun: %d\n",
 		 i2o_dev->lct_data.tid, channel, id, (unsigned int)lun);
@@ -542,9 +580,7 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt,
 				 void (*done) (struct scsi_cmnd *))
 {
 	struct i2o_controller *c;
-	struct Scsi_Host *host;
 	struct i2o_device *i2o_dev;
-	struct device *dev;
 	int tid;
 	struct i2o_message __iomem *msg;
 	u32 m;
@@ -554,20 +590,16 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt,
 	 * RETURN_SENSE_DATA_IN_REPLY_MESSAGE_FRAME
 	 */
 	u32 scsi_flags = 0x20a00000;
-	u32 sg_flags;
+	u32 sgl_offset;
 	u32 __iomem *mptr;
-	u32 __iomem *lenptr;
-	u32 len;
-	int i;
+	u32 cmd = I2O_CMD_SCSI_EXEC << 24;
+	int rc = 0;
 
 	/*
 	 *      Do the incoming paperwork
 	 */
-
 	i2o_dev = SCpnt->device->hostdata;
-	host = SCpnt->device->host;
 	c = i2o_dev->iop;
-	dev = &c->pdev->dev;
 
 	SCpnt->scsi_done = done;
 
@@ -575,7 +607,7 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt,
 		osm_warn("no I2O device in request\n");
 		SCpnt->result = DID_NO_CONNECT << 16;
 		done(SCpnt);
-		return 0;
+		goto exit;
 	}
 
 	tid = i2o_dev->lct_data.tid;
@@ -583,47 +615,86 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt,
 	osm_debug("qcmd: Tid = %03x\n", tid);
 	osm_debug("Real scsi messages.\n");
 
-	/*
-	 *      Obtain an I2O message. If there are none free then
-	 *      throw it back to the scsi layer
-	 */
-
-	m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET);
-	if (m == I2O_QUEUE_EMPTY)
-		return SCSI_MLQUEUE_HOST_BUSY;
-
-	mptr = &msg->body[0];
-
 	/*
 	 *      Put together a scsi execscb message
 	 */
-
 	switch (SCpnt->sc_data_direction) {
 	case PCI_DMA_NONE:
 		/* DATA NO XFER */
-		sg_flags = 0x00000000;
+		sgl_offset = SGL_OFFSET_0;
 		break;
 
 	case PCI_DMA_TODEVICE:
 		/* DATA OUT (iop-->dev) */
 		scsi_flags |= 0x80000000;
-		sg_flags = 0x14000000;
+		sgl_offset = SGL_OFFSET_10;
 		break;
 
 	case PCI_DMA_FROMDEVICE:
 		/* DATA IN  (iop<--dev) */
 		scsi_flags |= 0x40000000;
-		sg_flags = 0x10000000;
+		sgl_offset = SGL_OFFSET_10;
 		break;
 
 	default:
 		/* Unknown - kill the command */
 		SCpnt->result = DID_NO_CONNECT << 16;
 		done(SCpnt);
-		return 0;
+		goto exit;
 	}
 
-	writel(I2O_CMD_SCSI_EXEC << 24 | HOST_TID << 12 | tid, &msg->u.head[1]);
+	/*
+	 *      Obtain an I2O message. If there are none free then
+	 *      throw it back to the scsi layer
+	 */
+
+	m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET);
+	if (m == I2O_QUEUE_EMPTY) {
+		rc = SCSI_MLQUEUE_HOST_BUSY;
+		goto exit;
+	}
+
+	mptr = &msg->body[0];
+
+#ifdef CONFIG_I2O_EXT_ADAPTEC
+	if (c->adaptec) {
+		u32 adpt_flags = 0;
+
+		if (SCpnt->sc_request && SCpnt->sc_request->upper_private_data) {
+			i2o_sg_io_hdr_t __user *usr_ptr =
+			    ((Sg_request *) (SCpnt->sc_request->
+					     upper_private_data))->header.
+			    usr_ptr;
+
+			if (usr_ptr)
+				get_user(adpt_flags, &usr_ptr->flags);
+		}
+
+		switch (i2o_dev->lct_data.class_id) {
+		case I2O_CLASS_EXECUTIVE:
+		case I2O_CLASS_RANDOM_BLOCK_STORAGE:
+			/* interpret flag has to be set for executive */
+			adpt_flags ^= I2O_DPT_SG_FLAG_INTERPRET;
+			break;
+
+		default:
+			break;
+		}
+
+		/*
+		 * for Adaptec controllers we use the PRIVATE command, because
+		 * the normal SCSI EXEC doesn't support all SCSI commands on
+		 * all controllers (for example READ CAPACITY).
+		 */
+		if (sgl_offset == SGL_OFFSET_10)
+			sgl_offset = SGL_OFFSET_12;
+		cmd = I2O_CMD_PRIVATE << 24;
+		writel(I2O_VENDOR_DPT << 16 | I2O_CMD_SCSI_EXEC, mptr++);
+		writel(adpt_flags | tid, mptr++);
+	}
+#endif
+
+	writel(cmd | HOST_TID << 12 | tid, &msg->u.head[1]);
 	writel(i2o_scsi_driver.context, &msg->u.s.icntxt);
 
 	/* We want the SCSI control block back */
@@ -655,55 +726,30 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt,
 	/* Write SCSI command into the message - always 16 byte block */
 	memcpy_toio(mptr, SCpnt->cmnd, 16);
 	mptr += 4;
-	lenptr = mptr++;	/* Remember me - fill in when we know */
-
-	/* Now fill in the SGList and command */
-	if (SCpnt->use_sg) {
-		struct scatterlist *sg;
-		int sg_count;
-
-		sg = SCpnt->request_buffer;
-		len = 0;
 
-		sg_count = dma_map_sg(dev, sg, SCpnt->use_sg,
-				      SCpnt->sc_data_direction);
-
-		if (unlikely(sg_count <= 0))
-			return -ENOMEM;
-
-		for (i = SCpnt->use_sg; i > 0; i--) {
-			if (i == 1)
-				sg_flags |= 0xC0000000;
-			writel(sg_flags | sg_dma_len(sg), mptr++);
-			writel(sg_dma_address(sg), mptr++);
-			len += sg_dma_len(sg);
-			sg++;
-		}
-
-		writel(len, lenptr);
-	} else {
-		len = SCpnt->request_bufflen;
-
-		writel(len, lenptr);
-
-		if (len > 0) {
-			dma_addr_t dma_addr;
-
-			dma_addr = dma_map_single(dev, SCpnt->request_buffer,
-						  SCpnt->request_bufflen,
-						  SCpnt->sc_data_direction);
-			if (!dma_addr)
-				return -ENOMEM;
-
-			SCpnt->SCp.ptr = (void *)(unsigned long)dma_addr;
-			sg_flags |= 0xC0000000;
-			writel(sg_flags | SCpnt->request_bufflen, mptr++);
-			writel(dma_addr, mptr++);
+	if (sgl_offset != SGL_OFFSET_0) {
+		/* write size of data addressed by SGL */
+		writel(SCpnt->request_bufflen, mptr++);
+
+		/* Now fill in the SGList and command */
+		if (SCpnt->use_sg) {
+			if (!i2o_dma_map_sg(c, SCpnt->request_buffer,
+					    SCpnt->use_sg,
+					    SCpnt->sc_data_direction, &mptr))
+				goto nomem;
+		} else {
+			SCpnt->SCp.dma_handle =
+			    i2o_dma_map_single(c, SCpnt->request_buffer,
+					       SCpnt->request_bufflen,
+					       SCpnt->sc_data_direction, &mptr);
+			if (dma_mapping_error(SCpnt->SCp.dma_handle))
+				goto nomem;
 		}
 	}
 
 	/* Stick the headers on */
-	writel((mptr - &msg->u.head[0]) << 16 | SGL_OFFSET_10, &msg->u.head[0]);
+	writel(I2O_MESSAGE_SIZE(mptr - &msg->u.head[0]) | sgl_offset,
+	       &msg->u.head[0]);
 
 	/* Queue the message */
 	i2o_msg_post(c, m);
@@ -711,6 +757,13 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt,
 	osm_debug("Issued %ld\n", SCpnt->serial_number);
 
 	return 0;
+
+      nomem:
+	rc = -ENOMEM;
+	i2o_msg_nop(c, m);
+
+      exit:
+	return rc;
 };
 
 /**
diff --git a/drivers/message/i2o/pci.c b/drivers/message/i2o/pci.c
index a499af096a6..964fe481849 100644
--- a/drivers/message/i2o/pci.c
+++ b/drivers/message/i2o/pci.c
@@ -362,11 +362,33 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev,
 		c->promise = 1;
 	}
 
+	if (pdev->subsystem_vendor == PCI_VENDOR_ID_DPT)
+		c->adaptec = 1;
+
 	/* Cards that go bananas if you quiesce them before you reset them. */
 	if (pdev->vendor == PCI_VENDOR_ID_DPT) {
 		c->no_quiesce = 1;
 		if (pdev->device == 0xa511)
 			c->raptor = 1;
+
+		if (pdev->subsystem_device == 0xc05a) {
+			c->limit_sectors = 1;
+			printk(KERN_INFO
+			       "%s: limit sectors per request to %d\n", c->name,
+			       I2O_MAX_SECTORS_LIMITED);
+		}
+#ifdef CONFIG_I2O_EXT_ADAPTEC_DMA64
+		if (sizeof(dma_addr_t) > 4) {
+			if (pci_set_dma_mask(pdev, DMA_64BIT_MASK))
+				printk(KERN_INFO "%s: 64-bit DMA unavailable\n",
+				       c->name);
+			else {
+				c->pae_support = 1;
+				printk(KERN_INFO "%s: using 64-bit DMA\n",
+				       c->name);
+			}
+		}
+#endif
 	}
 
 	if ((rc = i2o_pci_alloc(c))) {
diff --git a/include/linux/i2o-dev.h b/include/linux/i2o-dev.h
index 90c984ecd52..d4a08d29e36 100644
--- a/include/linux/i2o-dev.h
+++ b/include/linux/i2o-dev.h
@@ -32,6 +32,13 @@ typedef unsigned int u32;
 
 #endif				/* __KERNEL__ */
 
+/*
+ *	Software module types
+ */
+#define I2O_SOFTWARE_MODULE_IRTOS		0x11
+#define I2O_SOFTWARE_MODULE_IOP_PRIVATE		0x22
+#define I2O_SOFTWARE_MODULE_IOP_CONFIG		0x23
+
 /*
  *	Vendors
  */
@@ -125,6 +132,10 @@ struct i2o_evt_get {
 	int lost;
 };
 
+typedef struct i2o_sg_io_hdr {
+	unsigned int flags;	/* see I2O_DPT_SG_IO_FLAGS */
+} i2o_sg_io_hdr_t;
+
 /**************************************************************************
  * HRT related constants and structures
  **************************************************************************/
@@ -403,4 +414,15 @@ typedef struct _i2o_status_block {
 #define ADAPTER_STATE_FAILED			0x10
 #define ADAPTER_STATE_FAULTED			0x11
 
+
+/*
+ * DPT / Adaptec specific values for i2o_sg_io_hdr flags.
+ */
+#define I2O_DPT_SG_FLAG_INTERPRET		0x00010000
+#define I2O_DPT_SG_FLAG_PHYSICAL		0x00020000
+
+#define I2O_DPT_FLASH_FRAG_SIZE			0x10000
+#define I2O_DPT_FLASH_READ			0x0101
+#define I2O_DPT_FLASH_WRITE			0x0102
+
 #endif				/* _I2O_DEV_H */
diff --git a/include/linux/i2o.h b/include/linux/i2o.h
index 497ea574f96..2039a87c2b9 100644
--- a/include/linux/i2o.h
+++ b/include/linux/i2o.h
@@ -147,10 +147,13 @@ struct i2o_controller {
 
 	struct pci_dev *pdev;	/* PCI device */
 
-	unsigned int short_req:1;	/* use small block sizes */
-	unsigned int no_quiesce:1;	/* dont quiesce before reset */
-	unsigned int raptor:1;		/* split bar */
 	unsigned int promise:1;		/* Promise controller */
+	unsigned int adaptec:1;		/* DPT / Adaptec controller */
+	unsigned int raptor:1;	/* split bar */
+	unsigned int no_quiesce:1;	/* dont quiesce before reset */
+	unsigned int short_req:1;	/* use small block sizes */
+	unsigned int limit_sectors:1;	/* limit number of sectors / request */
+	unsigned int pae_support:1;	/* controller has 64-bit SGL support */
 
 	struct list_head devices;	/* list of I2O devices */
 	struct list_head list;	/* Controller list */
@@ -746,7 +749,21 @@ static inline struct i2o_message __iomem *i2o_msg_in_to_virt(struct i2o_controll
 static inline int i2o_dma_alloc(struct device *dev, struct i2o_dma *addr,
 				size_t len, unsigned int gfp_mask)
 {
+	struct pci_dev *pdev = to_pci_dev(dev);
+	int dma_64 = 0;
+
+	if ((sizeof(dma_addr_t) > 4) && (pdev->dma_mask == DMA_64BIT_MASK)) {
+			dma_64 = 1;
+			if(pci_set_dma_mask(pdev, DMA_32BIT_MASK))
+				return -ENOMEM;
+	}
+
 	addr->virt = dma_alloc_coherent(dev, len, &addr->phys, gfp_mask);
+
+	if ((sizeof(dma_addr_t) > 4) && dma_64)
+		if(pci_set_dma_mask(pdev, DMA_64BIT_MASK))
+			printk(KERN_WARNING "i2o: unable to set 64-bit DMA");
+
 	if (!addr->virt)
 		return -ENOMEM;
 
@@ -946,7 +963,7 @@ extern void i2o_debug_state(struct i2o_controller *c);
 #define I2O_CMD_BLOCK_MEJECT		0x43
 #define I2O_CMD_BLOCK_POWER		0x70
 
-#define I2O_PRIVATE_MSG			0xFF
+#define I2O_CMD_PRIVATE			0xFF
 
 /* Command status values  */
 
@@ -1095,9 +1112,9 @@ extern void i2o_debug_state(struct i2o_controller *c);
 #define SGL_OFFSET_8    (0x0080 | I2OVERSION)
 #define SGL_OFFSET_9    (0x0090 | I2OVERSION)
 #define SGL_OFFSET_10   (0x00A0 | I2OVERSION)
-
-#define TRL_OFFSET_5    (0x0050 | I2OVERSION)
-#define TRL_OFFSET_6    (0x0060 | I2OVERSION)
+#define SGL_OFFSET_11   (0x00B0 | I2OVERSION)
+#define SGL_OFFSET_12   (0x00C0 | I2OVERSION)
+#define SGL_OFFSET(x)   (((x)<<4) | I2OVERSION)
 
 /* Transaction Reply Lists (TRL) Control Word structure */
 #define TRL_SINGLE_FIXED_LENGTH		0x00
@@ -1130,7 +1147,6 @@ extern void i2o_debug_state(struct i2o_controller *c);
 #define HOST_TID		1
 
 #define MSG_FRAME_SIZE		128	/* i2o_scsi assumes >= 32 */
-#define REPLY_FRAME_SIZE	17
 #define SG_TABLESIZE		30
 #define NMBR_MSG_FRAMES		128
 
@@ -1155,11 +1171,10 @@ extern void i2o_debug_state(struct i2o_controller *c);
 #define I2O_HRT_GET_TRIES		3
 #define I2O_LCT_GET_TRIES		3
 
-/* request queue sizes */
+/* defines for max_sectors and max_phys_segments */
 #define I2O_MAX_SECTORS			1024
+#define I2O_MAX_SECTORS_LIMITED		256
 #define I2O_MAX_PHYS_SEGMENTS		MAX_PHYS_SEGMENTS
 
-#define I2O_REQ_MEMPOOL_SIZE		32
-
 #endif				/* __KERNEL__ */
 #endif				/* _I2O_H */
diff --git a/include/scsi/sg_request.h b/include/scsi/sg_request.h
new file mode 100644
index 00000000000..57ff525bdd3
--- /dev/null
+++ b/include/scsi/sg_request.h
@@ -0,0 +1,26 @@
+typedef struct scsi_request Scsi_Request;
+
+static Scsi_Request *dummy_cmdp;	/* only used for sizeof */
+
+typedef struct sg_scatter_hold { /* holding area for scsi scatter gather info */
+	unsigned short k_use_sg; /* Count of kernel scatter-gather pieces */
+	unsigned short sglist_len; /* size of malloc'd scatter-gather list ++ */
+	unsigned bufflen;	/* Size of (aggregate) data buffer */
+	unsigned b_malloc_len;	/* actual len malloc'ed in buffer */
+	void *buffer;		/* Data buffer or scatter list (k_use_sg>0) */
+	char dio_in_use;	/* 0->indirect IO (or mmap), 1->dio */
+	unsigned char cmd_opcode; /* first byte of command */
+} Sg_scatter_hold;
+
+typedef struct sg_request {	/* SG_MAX_QUEUE requests outstanding per file */
+	Scsi_Request *my_cmdp;	/* != 0  when request with lower levels */
+	struct sg_request *nextrp;	/* NULL -> tail request (slist) */
+	struct sg_fd *parentfp;	/* NULL -> not in use */
+	Sg_scatter_hold data;	/* hold buffer, perhaps scatter list */
+	sg_io_hdr_t header;	/* scsi command+info, see <scsi/sg.h> */
+	unsigned char sense_b[sizeof (dummy_cmdp->sr_sense_buffer)];
+	char res_used;		/* 1 -> using reserve buffer, 0 -> not ... */
+	char orphan;		/* 1 -> drop on sight, 0 -> normal */
+	char sg_io_owned;	/* 1 -> packet belongs to SG_IO */
+	volatile char done;	/* 0->before bh, 1->before read, 2->read */
+} Sg_request;
-- 
cgit v1.2.3-70-g09d2


From 9e87545f06930c1d294423a8091d1077e7444a47 Mon Sep 17 00:00:00 2001
From: Markus Lidel <Markus.Lidel@shadowconnect.com>
Date: Thu, 23 Jun 2005 22:02:21 -0700
Subject: [PATCH] I2O: second code cleanup of sparse warnings and unneeded
 syncronization

Changes:
 - Added header "core.h" for i2o_core.ko internal definitions
 - More sparse fixes
 - Changed display of TID's in sysfs attributes from XXX to 0xXXX
 - Use the right functions for accessing I/O and normal memory
 - Removed error handling of SCSI device errors and let the SCSI layer
   take care of it
 - Added new device / removed device handling to SCSI-OSM
 - Make status access volatile
 - Cleaned up activation of I2O controller
 - Removed unnecessary wmb() and rmb() calls
 - Use own struct i2o_io for I/O memory instead of struct i2o_dma

Signed-off-by: Markus Lidel <Markus.Lidel@shadowconnect.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/message/i2o/core.h       |  55 +++++++++++
 drivers/message/i2o/debug.c      |   3 -
 drivers/message/i2o/device.c     |  22 +++--
 drivers/message/i2o/driver.c     |  24 ++---
 drivers/message/i2o/exec-osm.c   |  27 +++---
 drivers/message/i2o/i2o_block.c  |   4 +-
 drivers/message/i2o/i2o_config.c |   8 +-
 drivers/message/i2o/i2o_scsi.c   | 202 +++++++++++++--------------------------
 drivers/message/i2o/iop.c        | 128 +++++++++++++------------
 drivers/message/i2o/pci.c        |  21 +---
 include/linux/i2o-dev.h          |  23 +++--
 include/linux/i2o.h              | 116 ++++++----------------
 12 files changed, 275 insertions(+), 358 deletions(-)
 create mode 100644 drivers/message/i2o/core.h

(limited to 'include')

diff --git a/drivers/message/i2o/core.h b/drivers/message/i2o/core.h
new file mode 100644
index 00000000000..49851cccc48
--- /dev/null
+++ b/drivers/message/i2o/core.h
@@ -0,0 +1,55 @@
+/*
+ *	I2O core internal declarations
+ *
+ *	Copyright (C) 2005	Markus Lidel <Markus.Lidel@shadowconnect.com>
+ *
+ *	This program is free software; you can redistribute it and/or modify it
+ *	under the terms of the GNU General Public License as published by the
+ *	Free Software Foundation; either version 2 of the License, or (at your
+ *	option) any later version.
+ *
+ *	Fixes/additions:
+ *		Markus Lidel <Markus.Lidel@shadowconnect.com>
+ *			initial version.
+ */
+
+/* Exec-OSM */
+extern struct bus_type i2o_bus_type;
+
+extern struct i2o_driver i2o_exec_driver;
+extern int i2o_exec_lct_get(struct i2o_controller *);
+
+extern int __init i2o_exec_init(void);
+extern void __exit i2o_exec_exit(void);
+
+/* driver */
+extern int i2o_driver_dispatch(struct i2o_controller *, u32);
+
+extern int __init i2o_driver_init(void);
+extern void __exit i2o_driver_exit(void);
+
+/* PCI */
+extern int __init i2o_pci_init(void);
+extern void __exit i2o_pci_exit(void);
+
+/* device */
+extern void i2o_device_remove(struct i2o_device *);
+extern int i2o_device_parse_lct(struct i2o_controller *);
+
+extern int i2o_device_init(void);
+extern void i2o_device_exit(void);
+
+/* IOP */
+extern struct i2o_controller *i2o_iop_alloc(void);
+extern void i2o_iop_free(struct i2o_controller *);
+
+extern int i2o_iop_add(struct i2o_controller *);
+extern void i2o_iop_remove(struct i2o_controller *);
+
+/* control registers relative to c->base */
+#define I2O_IRQ_STATUS	0x30
+#define I2O_IRQ_MASK	0x34
+#define I2O_IN_PORT	0x40
+#define I2O_OUT_PORT	0x44
+
+#define I2O_IRQ_OUTBOUND_POST	0x00000008
diff --git a/drivers/message/i2o/debug.c b/drivers/message/i2o/debug.c
index 2a5d478fc60..018ca887ca8 100644
--- a/drivers/message/i2o/debug.c
+++ b/drivers/message/i2o/debug.c
@@ -4,8 +4,6 @@
 #include <linux/pci.h>
 #include <linux/i2o.h>
 
-extern struct i2o_driver **i2o_drivers;
-extern unsigned int i2o_max_drivers;
 static void i2o_report_util_cmd(u8 cmd);
 static void i2o_report_exec_cmd(u8 cmd);
 static void i2o_report_fail_status(u8 req_status, u32 * msg);
@@ -23,7 +21,6 @@ void i2o_report_status(const char *severity, const char *str,
 	u8 cmd = (msg[1] >> 24) & 0xFF;
 	u8 req_status = (msg[4] >> 24) & 0xFF;
 	u16 detailed_status = msg[4] & 0xFFFF;
-	//struct i2o_driver *h = i2o_drivers[msg[2] & (i2o_max_drivers-1)];
 
 	if (cmd == I2O_CMD_UTIL_EVT_REGISTER)
 		return;		// No status in this reply
diff --git a/drivers/message/i2o/device.c b/drivers/message/i2o/device.c
index f1b7eb63d54..0ee342ea29b 100644
--- a/drivers/message/i2o/device.c
+++ b/drivers/message/i2o/device.c
@@ -16,9 +16,7 @@
 #include <linux/module.h>
 #include <linux/i2o.h>
 #include <linux/delay.h>
-
-/* Exec OSM functions */
-extern struct bus_type i2o_bus_type;
+#include "core.h"
 
 /**
  *	i2o_device_issue_claim - claim or release a device
@@ -293,12 +291,12 @@ int i2o_device_parse_lct(struct i2o_controller *c)
 	}
 
 	if (lct->table_size * 4 > c->dlct.len) {
-		memcpy_fromio(c->lct, c->dlct.virt, c->dlct.len);
+		memcpy(c->lct, c->dlct.virt, c->dlct.len);
 		up(&c->lct_lock);
 		return -EAGAIN;
 	}
 
-	memcpy_fromio(c->lct, c->dlct.virt, lct->table_size * 4);
+	memcpy(c->lct, c->dlct.virt, lct->table_size * 4);
 
 	lct = c->lct;
 
@@ -353,7 +351,7 @@ static ssize_t i2o_device_class_show_class_id(struct class_device *cd,
 {
 	struct i2o_device *dev = to_i2o_device(cd->dev);
 
-	sprintf(buf, "%03x\n", dev->lct_data.class_id);
+	sprintf(buf, "0x%03x\n", dev->lct_data.class_id);
 	return strlen(buf) + 1;
 };
 
@@ -368,7 +366,7 @@ static ssize_t i2o_device_class_show_tid(struct class_device *cd, char *buf)
 {
 	struct i2o_device *dev = to_i2o_device(cd->dev);
 
-	sprintf(buf, "%03x\n", dev->lct_data.tid);
+	sprintf(buf, "0x%03x\n", dev->lct_data.tid);
 	return strlen(buf) + 1;
 };
 
@@ -490,7 +488,7 @@ static int i2o_parm_issue(struct i2o_device *i2o_dev, int cmd, void *oplist,
 	if (rc == -ETIMEDOUT)
 		return rc;
 
-	memcpy_fromio(reslist, res.virt, res.len);
+	memcpy(reslist, res.virt, res.len);
 	i2o_dma_free(dev, &res);
 
 	/* Query failed */
@@ -532,17 +530,23 @@ int i2o_parm_field_get(struct i2o_device *i2o_dev, int group, int field,
 		       void *buf, int buflen)
 {
 	u16 opblk[] = { 1, 0, I2O_PARAMS_FIELD_GET, group, 1, field };
-	u8 resblk[8 + buflen];	/* 8 bytes for header */
+	u8 *resblk;		/* 8 bytes for header */
 	int size;
 
 	if (field == -1)	/* whole group */
 		opblk[4] = -1;
 
+	resblk = kmalloc(buflen + 8, GFP_KERNEL | GFP_ATOMIC);
+	if (!resblk)
+		return -ENOMEM;
+
 	size = i2o_parm_issue(i2o_dev, I2O_CMD_UTIL_PARAMS_GET, opblk,
 			      sizeof(opblk), resblk, buflen + 8);
 
 	memcpy(buf, resblk + 8, buflen);	/* cut off header */
 
+	kfree(resblk);
+
 	if (size > buflen)
 		return buflen;
 
diff --git a/drivers/message/i2o/driver.c b/drivers/message/i2o/driver.c
index 393be8e2914..c32f9dbc574 100644
--- a/drivers/message/i2o/driver.c
+++ b/drivers/message/i2o/driver.c
@@ -17,11 +17,12 @@
 #include <linux/module.h>
 #include <linux/rwsem.h>
 #include <linux/i2o.h>
+#include "core.h"
 
 #define OSM_NAME	"i2o"
 
 /* max_drivers - Maximum I2O drivers (OSMs) which could be registered */
-unsigned int i2o_max_drivers = I2O_MAX_DRIVERS;
+static unsigned int i2o_max_drivers = I2O_MAX_DRIVERS;
 module_param_named(max_drivers, i2o_max_drivers, uint, 0);
 MODULE_PARM_DESC(max_drivers, "maximum number of OSM's to support");
 
@@ -179,15 +180,10 @@ void i2o_driver_unregister(struct i2o_driver *drv)
 int i2o_driver_dispatch(struct i2o_controller *c, u32 m)
 {
 	struct i2o_driver *drv;
-	struct i2o_message __iomem *msg = i2o_msg_out_to_virt(c, m);
-	u32 context;
+	struct i2o_message *msg = i2o_msg_out_to_virt(c, m);
+	u32 context = le32_to_cpu(msg->u.s.icntxt);
 	unsigned long flags;
 
-	if(unlikely(!msg))
-		return -EIO;
-
-	context = readl(&msg->u.s.icntxt);
-
 	if (unlikely(context >= i2o_max_drivers)) {
 		osm_warn("%s: Spurious reply to unknown driver %d\n", c->name,
 			 context);
@@ -204,11 +200,11 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m)
 		return -EIO;
 	}
 
-	if ((readl(&msg->u.head[1]) >> 24) == I2O_CMD_UTIL_EVT_REGISTER) {
+	if ((le32_to_cpu(msg->u.head[1]) >> 24) == I2O_CMD_UTIL_EVT_REGISTER) {
 		struct i2o_device *dev, *tmp;
 		struct i2o_event *evt;
 		u16 size;
-		u16 tid = readl(&msg->u.head[1]) & 0xfff;
+		u16 tid = le32_to_cpu(msg->u.head[1]) & 0xfff;
 
 		osm_debug("event received from device %d\n", tid);
 
@@ -216,16 +212,16 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m)
 			return -EIO;
 
 		/* cut of header from message size (in 32-bit words) */
-		size = (readl(&msg->u.head[0]) >> 16) - 5;
+		size = (le32_to_cpu(msg->u.head[0]) >> 16) - 5;
 
 		evt = kmalloc(size * 4 + sizeof(*evt), GFP_ATOMIC | __GFP_ZERO);
 		if (!evt)
 			return -ENOMEM;
 
 		evt->size = size;
-		evt->tcntxt = readl(&msg->u.s.tcntxt);
-		evt->event_indicator = readl(&msg->body[0]);
-		memcpy_fromio(&evt->tcntxt, &msg->u.s.tcntxt, size * 4);
+		evt->tcntxt = le32_to_cpu(msg->u.s.tcntxt);
+		evt->event_indicator = le32_to_cpu(msg->body[0]);
+		memcpy(&evt->tcntxt, &msg->u.s.tcntxt, size * 4);
 
 		list_for_each_entry_safe(dev, tmp, &c->devices, list)
 		    if (dev->lct_data.tid == tid) {
diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c
index 0160221c802..ffe0cecfa06 100644
--- a/drivers/message/i2o/exec-osm.c
+++ b/drivers/message/i2o/exec-osm.c
@@ -30,6 +30,7 @@
 #include <linux/module.h>
 #include <linux/i2o.h>
 #include <linux/delay.h>
+#include "core.h"
 
 #define OSM_NAME "exec-osm"
 
@@ -37,9 +38,6 @@ struct i2o_driver i2o_exec_driver;
 
 static int i2o_exec_lct_notify(struct i2o_controller *c, u32 change_ind);
 
-/* Module internal functions from other sources */
-extern int i2o_device_parse_lct(struct i2o_controller *);
-
 /* global wait list for POST WAIT */
 static LIST_HEAD(i2o_exec_wait_list);
 
@@ -50,7 +48,7 @@ struct i2o_exec_wait {
 	u32 tcntxt;		/* transaction context from reply */
 	int complete;		/* 1 if reply received otherwise 0 */
 	u32 m;			/* message id */
-	struct i2o_message __iomem *msg;	/* pointer to the reply message */
+	struct i2o_message *msg;	/* pointer to the reply message */
 	struct list_head list;	/* node in global wait list */
 };
 
@@ -162,7 +160,7 @@ int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long
 	barrier();
 
 	if (wait->complete) {
-		rc = readl(&wait->msg->body[0]) >> 24;
+		rc = le32_to_cpu(wait->msg->body[0]) >> 24;
 		i2o_flush_reply(c, wait->m);
 		i2o_exec_wait_free(wait);
 	} else {
@@ -202,8 +200,7 @@ int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long
  *	message must also be given back to the controller.
  */
 static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m,
-				      struct i2o_message __iomem *msg,
-				      u32 context)
+				      struct i2o_message *msg, u32 context)
 {
 	struct i2o_exec_wait *wait, *tmp;
 	unsigned long flags;
@@ -378,11 +375,11 @@ static void i2o_exec_lct_modified(struct i2o_controller *c)
  *	code on failure and if the reply should be flushed.
  */
 static int i2o_exec_reply(struct i2o_controller *c, u32 m,
-			  struct i2o_message __iomem *msg)
+			  struct i2o_message *msg)
 {
 	u32 context;
 
-	if (readl(&msg->u.head[0]) & MSG_FAIL) {
+	if (le32_to_cpu(msg->u.head[0]) & MSG_FAIL) {
 		/*
 		 * If Fail bit is set we must take the transaction context of
 		 * the preserved message to find the right request again.
@@ -390,7 +387,7 @@ static int i2o_exec_reply(struct i2o_controller *c, u32 m,
 		struct i2o_message __iomem *pmsg;
 		u32 pm;
 
-		pm = readl(&msg->body[3]);
+		pm = le32_to_cpu(msg->body[3]);
 
 		pmsg = i2o_msg_in_to_virt(c, pm);
 
@@ -401,12 +398,12 @@ static int i2o_exec_reply(struct i2o_controller *c, u32 m,
 		/* Release the preserved msg */
 		i2o_msg_nop(c, pm);
 	} else
-		context = readl(&msg->u.s.tcntxt);
+		context = le32_to_cpu(msg->u.s.tcntxt);
 
 	if (context & 0x80000000)
 		return i2o_msg_post_wait_complete(c, m, msg, context);
 
-	if ((readl(&msg->u.head[1]) >> 24) == I2O_CMD_LCT_NOTIFY) {
+	if ((le32_to_cpu(msg->u.head[1]) >> 24) == I2O_CMD_LCT_NOTIFY) {
 		struct work_struct *work;
 
 		pr_debug("%s: LCT notify received\n", c->name);
@@ -442,9 +439,9 @@ static int i2o_exec_reply(struct i2o_controller *c, u32 m,
  */
 static void i2o_exec_event(struct i2o_event *evt)
 {
-	if(likely(evt->i2o_dev))
-		osm_info("Event received from device: %d\n",
-			 evt->i2o_dev->lct_data.tid);
+	if (likely(evt->i2o_dev))
+		osm_debug("Event received from device: %d\n",
+			  evt->i2o_dev->lct_data.tid);
 	kfree(evt);
 };
 
diff --git a/drivers/message/i2o/i2o_block.c b/drivers/message/i2o/i2o_block.c
index 1dd2b9dad50..28b3918dbc1 100644
--- a/drivers/message/i2o/i2o_block.c
+++ b/drivers/message/i2o/i2o_block.c
@@ -62,7 +62,7 @@
 #include "i2o_block.h"
 
 #define OSM_NAME	"block-osm"
-#define OSM_VERSION	"$Rev$"
+#define OSM_VERSION	"1.287"
 #define OSM_DESCRIPTION	"I2O Block Device OSM"
 
 static struct i2o_driver i2o_block_driver;
@@ -537,7 +537,7 @@ static int i2o_block_reply(struct i2o_controller *c, u32 m,
 
 static void i2o_block_event(struct i2o_event *evt)
 {
-	osm_info("event received\n");
+	osm_debug("event received\n");
 	kfree(evt);
 };
 
diff --git a/drivers/message/i2o/i2o_config.c b/drivers/message/i2o/i2o_config.c
index 7636833b462..8160a1f6c73 100644
--- a/drivers/message/i2o/i2o_config.c
+++ b/drivers/message/i2o/i2o_config.c
@@ -36,6 +36,8 @@
 
 #include <asm/uaccess.h>
 
+#define SG_TABLESIZE		30
+
 extern int i2o_parm_issue(struct i2o_device *, int, void *, int, void *, int);
 
 static int i2o_cfg_ioctl(struct inode *inode, struct file *fp, unsigned int cmd,
@@ -663,7 +665,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar
 		goto sg_list_cleanup;
 
 	if (sg_offset) {
-		u32 msg[MSG_FRAME_SIZE];
+		u32 msg[I2O_OUTBOUND_MSG_FRAME_SIZE];
 		/* Copy back the Scatter Gather buffers back to user space */
 		u32 j;
 		// TODO 64bit fix
@@ -671,7 +673,7 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar
 		int sg_size;
 
 		// re-acquire the original message to handle correctly the sg copy operation
-		memset(&msg, 0, MSG_FRAME_SIZE * 4);
+		memset(&msg, 0, I2O_OUTBOUND_MSG_FRAME_SIZE * 4);
 		// get user msg size in u32s
 		if (get_user(size, &user_msg[0])) {
 			rcode = -EFAULT;
@@ -902,7 +904,7 @@ static int i2o_cfg_passthru(unsigned long arg)
 		int sg_size;
 
 		// re-acquire the original message to handle correctly the sg copy operation
-		memset(&msg, 0, MSG_FRAME_SIZE * 4);
+		memset(&msg, 0, I2O_OUTBOUND_MSG_FRAME_SIZE * 4);
 		// get user msg size in u32s
 		if (get_user(size, &user_msg[0])) {
 			rcode = -EFAULT;
diff --git a/drivers/message/i2o/i2o_scsi.c b/drivers/message/i2o/i2o_scsi.c
index fef53b509a6..9f1744c3933 100644
--- a/drivers/message/i2o/i2o_scsi.c
+++ b/drivers/message/i2o/i2o_scsi.c
@@ -40,7 +40,6 @@
  *	Fix the resource management problems.
  */
 
-#define DEBUG 1
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/types.h>
@@ -338,162 +337,89 @@ static int i2o_scsi_reply(struct i2o_controller *c, u32 m,
 			  struct i2o_message *msg)
 {
 	struct scsi_cmnd *cmd;
+	u32 error;
 	struct device *dev;
-	u8 as, ds, st;
 
 	cmd = i2o_cntxt_list_get(c, le32_to_cpu(msg->u.s.tcntxt));
-
-	if (msg->u.head[0] & (1 << 13)) {
-		struct i2o_message __iomem *pmsg;	/* preserved message */
-		u32 pm;
-		int err = DID_ERROR;
-
-		pm = le32_to_cpu(msg->body[3]);
-
-		pmsg = i2o_msg_in_to_virt(c, pm);
-
-		osm_err("IOP fail.\n");
-		osm_err("From %d To %d Cmd %d.\n",
-			(msg->u.head[1] >> 12) & 0xFFF,
-			msg->u.head[1] & 0xFFF, msg->u.head[1] >> 24);
-		osm_err("Failure Code %d.\n", msg->body[0] >> 24);
-		if (msg->body[0] & (1 << 16))
-			osm_err("Format error.\n");
-		if (msg->body[0] & (1 << 17))
-			osm_err("Path error.\n");
-		if (msg->body[0] & (1 << 18))
-			osm_err("Path State.\n");
-		if (msg->body[0] & (1 << 18))
-		{
-			osm_err("Congestion.\n");
-			err = DID_BUS_BUSY;
-		}
-
-		osm_debug("Failing message is %p.\n", pmsg);
-
-		cmd = i2o_cntxt_list_get(c, readl(&pmsg->u.s.tcntxt));
-		if (!cmd)
-			return 1;
-
-		cmd->result = err << 16;
-		cmd->scsi_done(cmd);
-
-		/* Now flush the message by making it a NOP */
-		i2o_msg_nop(c, pm);
-
-		return 1;
+	if (unlikely(!cmd)) {
+		osm_err("NULL reply received!\n");
+		return -1;
 	}
 
 	/*
 	 *      Low byte is device status, next is adapter status,
 	 *      (then one byte reserved), then request status.
 	 */
-	ds = (u8) le32_to_cpu(msg->body[0]);
-	as = (u8) (le32_to_cpu(msg->body[0]) >> 8);
-	st = (u8) (le32_to_cpu(msg->body[0]) >> 24);
+	error = le32_to_cpu(msg->body[0]);
 
+	osm_debug("Completed %ld\n", cmd->serial_number);
+
+	cmd->result = error & 0xff;
 	/*
-	 *      Is this a control request coming back - eg an abort ?
+	 * if DeviceStatus is not SCSI_SUCCESS copy over the sense data and let
+	 * the SCSI layer handle the error
 	 */
+	if (cmd->result)
+		memcpy(cmd->sense_buffer, &msg->body[3],
+		       min(sizeof(cmd->sense_buffer), (size_t) 40));
 
-	if (!cmd) {
-		if (st)
-			osm_warn("SCSI abort: %08X", le32_to_cpu(msg->body[0]));
-		osm_info("SCSI abort completed.\n");
-		return -EFAULT;
-	}
+	/* only output error code if AdapterStatus is not HBA_SUCCESS */
+	if ((error >> 8) & 0xff)
+		osm_err("SCSI error %08x\n", error);
 
-	osm_debug("Completed %ld\n", cmd->serial_number);
+	dev = &c->pdev->dev;
+	if (cmd->use_sg)
+		dma_unmap_sg(dev, cmd->request_buffer, cmd->use_sg,
+			     cmd->sc_data_direction);
+	else if (cmd->SCp.dma_handle)
+		dma_unmap_single(dev, cmd->SCp.dma_handle, cmd->request_bufflen,
+				 cmd->sc_data_direction);
 
-	if (st) {
-		u32 count, error;
-		/* An error has occurred */
-
-		switch (st) {
-		case 0x06:
-			count = le32_to_cpu(msg->body[1]);
-			if (count < cmd->underflow) {
-				int i;
-
-				osm_err("SCSI underflow 0x%08X 0x%08X\n", count,
-					cmd->underflow);
-				osm_debug("Cmd: ");
-				for (i = 0; i < 15; i++)
-					pr_debug("%02X ", cmd->cmnd[i]);
-				pr_debug(".\n");
-				cmd->result = (DID_ERROR << 16);
-			}
-			break;
+	cmd->scsi_done(cmd);
 
-		default:
-			error = le32_to_cpu(msg->body[0]);
-
-			osm_err("SCSI error %08x\n", error);
-
-			if ((error & 0xff) == 0x02 /*CHECK_CONDITION */ ) {
-				int i;
-				u32 len = sizeof(cmd->sense_buffer);
-				len = (len > 40) ? 40 : len;
-				// Copy over the sense data
-				memcpy(cmd->sense_buffer, (void *)&msg->body[3],
-				       len);
-				for (i = 0; i <= len; i++)
-					osm_info("%02x\n",
-						 cmd->sense_buffer[i]);
-				if (cmd->sense_buffer[0] == 0x70
-				    && cmd->sense_buffer[2] == DATA_PROTECT) {
-					/* This is to handle an array failed */
-					cmd->result = (DID_TIME_OUT << 16);
-					printk(KERN_WARNING "%s: SCSI Data "
-					       "Protect-Device (%d,%d,%d) "
-					       "hba_status=0x%x, dev_status="
-					       "0x%x, cmd=0x%x\n", c->name,
-					       (u32) cmd->device->channel,
-					       (u32) cmd->device->id,
-					       (u32) cmd->device->lun,
-					       (error >> 8) & 0xff,
-					       error & 0xff, cmd->cmnd[0]);
-				} else
-					cmd->result = (DID_ERROR << 16);
-
-				break;
-			}
-
-			switch (as) {
-			case 0x0E:
-				/* SCSI Reset */
-				cmd->result = DID_RESET << 16;
-				break;
-
-			case 0x0F:
-				cmd->result = DID_PARITY << 16;
-				break;
-
-			default:
-				cmd->result = DID_ERROR << 16;
-				break;
-			}
+	return 1;
+};
 
-			break;
-		}
+/**
+ *	i2o_scsi_notify_device_add - Retrieve notifications of added devices
+ *	@i2o_dev: the I2O device which was added
+ *
+ *	If a I2O device is added we catch the notification, because I2O classes
+ *	other then SCSI peripheral will not be received through
+ *	i2o_scsi_probe().
+ */
+static void i2o_scsi_notify_device_add(struct i2o_device *i2o_dev)
+{
+	switch (i2o_dev->lct_data.class_id) {
+	case I2O_CLASS_EXECUTIVE:
+	case I2O_CLASS_RANDOM_BLOCK_STORAGE:
+		i2o_scsi_probe(&i2o_dev->device);
+		break;
 
-		cmd->scsi_done(cmd);
-		return 1;
+	default:
+		break;
 	}
+};
 
-	cmd->result = DID_OK << 16 | ds;
-
-	dev = &c->pdev->dev;
-	if (cmd->use_sg)
-		dma_unmap_sg(dev, (struct scatterlist *)cmd->buffer,
-			     cmd->use_sg, cmd->sc_data_direction);
-	else if (cmd->request_bufflen)
-		dma_unmap_single(dev, (dma_addr_t) ((long)cmd->SCp.ptr),
-				 cmd->request_bufflen, cmd->sc_data_direction);
-
-	cmd->scsi_done(cmd);
+/**
+ *	i2o_scsi_notify_device_remove - Retrieve notifications of removed
+ *				        devices
+ *	@i2o_dev: the I2O device which was removed
+ *
+ *	If a I2O device is removed, we catch the notification to remove the
+ *	corresponding SCSI device.
+ */
+static void i2o_scsi_notify_device_remove(struct i2o_device *i2o_dev)
+{
+	switch (i2o_dev->lct_data.class_id) {
+	case I2O_CLASS_EXECUTIVE:
+	case I2O_CLASS_RANDOM_BLOCK_STORAGE:
+		i2o_scsi_remove(&i2o_dev->device);
+		break;
 
-	return 1;
+	default:
+		break;
+	}
 };
 
 /**
@@ -554,6 +480,8 @@ static struct i2o_driver i2o_scsi_driver = {
 	.name = OSM_NAME,
 	.reply = i2o_scsi_reply,
 	.classes = i2o_scsi_class_id,
+	.notify_device_add = i2o_scsi_notify_device_add,
+	.notify_device_remove = i2o_scsi_notify_device_remove,
 	.notify_controller_add = i2o_scsi_notify_controller_add,
 	.notify_controller_remove = i2o_scsi_notify_controller_remove,
 	.driver = {
@@ -712,7 +640,7 @@ static int i2o_scsi_queuecommand(struct scsi_cmnd *SCpnt,
 	 */
 
 	/* Attach tags to the devices */
-	/*
+	/* FIXME: implement
 	   if(SCpnt->device->tagged_supported) {
 	   if(SCpnt->tag == HEAD_OF_QUEUE_TAG)
 	   scsi_flags |= 0x01000000;
diff --git a/drivers/message/i2o/iop.c b/drivers/message/i2o/iop.c
index 40312053b38..c32022bc2a2 100644
--- a/drivers/message/i2o/iop.c
+++ b/drivers/message/i2o/iop.c
@@ -28,8 +28,10 @@
 #include <linux/module.h>
 #include <linux/i2o.h>
 #include <linux/delay.h>
+#include "core.h"
 
-#define OSM_VERSION	"$Rev$"
+#define OSM_NAME	"i2o"
+#define OSM_VERSION	"1.288"
 #define OSM_DESCRIPTION	"I2O subsystem"
 
 /* global I2O controller list */
@@ -43,20 +45,6 @@ static struct i2o_dma i2o_systab;
 
 static int i2o_hrt_get(struct i2o_controller *c);
 
-/* Module internal functions from other sources */
-extern struct i2o_driver i2o_exec_driver;
-extern int i2o_exec_lct_get(struct i2o_controller *);
-extern void i2o_device_remove(struct i2o_device *);
-
-extern int __init i2o_driver_init(void);
-extern void __exit i2o_driver_exit(void);
-extern int __init i2o_exec_init(void);
-extern void __exit i2o_exec_exit(void);
-extern int __init i2o_pci_init(void);
-extern void __exit i2o_pci_exit(void);
-extern int i2o_device_init(void);
-extern void i2o_device_exit(void);
-
 /**
  *	i2o_msg_nop - Returns a message which is not used
  *	@c: I2O controller from which the message was created
@@ -92,16 +80,16 @@ void i2o_msg_nop(struct i2o_controller *c, u32 m)
  *	address from the read port (see the i2o spec). If no message is
  *	available returns I2O_QUEUE_EMPTY and msg is leaved untouched.
  */
-u32 i2o_msg_get_wait(struct i2o_controller *c, struct i2o_message __iomem **msg,
-		     int wait)
+u32 i2o_msg_get_wait(struct i2o_controller *c,
+		     struct i2o_message __iomem ** msg, int wait)
 {
 	unsigned long timeout = jiffies + wait * HZ;
 	u32 m;
 
 	while ((m = i2o_msg_get(c, msg)) == I2O_QUEUE_EMPTY) {
 		if (time_after(jiffies, timeout)) {
-			pr_debug("%s: Timeout waiting for message frame.\n",
-				 c->name);
+			osm_debug("%s: Timeout waiting for message frame.\n",
+				  c->name);
 			return I2O_QUEUE_EMPTY;
 		}
 		set_current_state(TASK_UNINTERRUPTIBLE);
@@ -466,7 +454,7 @@ static int i2o_iop_clear(struct i2o_controller *c)
  */
 static int i2o_iop_init_outbound_queue(struct i2o_controller *c)
 {
-	u8 *status = c->status.virt;
+	volatile u8 *status = c->status.virt;
 	u32 m;
 	struct i2o_message __iomem *msg;
 	ulong timeout;
@@ -474,21 +462,20 @@ static int i2o_iop_init_outbound_queue(struct i2o_controller *c)
 
 	osm_debug("%s: Initializing Outbound Queue...\n", c->name);
 
-	memset(status, 0, 4);
+	memset(c->status.virt, 0, 4);
 
 	m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET);
 	if (m == I2O_QUEUE_EMPTY)
 		return -ETIMEDOUT;
 
-	writel(EIGHT_WORD_MSG_SIZE | TRL_OFFSET_6, &msg->u.head[0]);
+	writel(EIGHT_WORD_MSG_SIZE | SGL_OFFSET_6, &msg->u.head[0]);
 	writel(I2O_CMD_OUTBOUND_INIT << 24 | HOST_TID << 12 | ADAPTER_TID,
 	       &msg->u.head[1]);
 	writel(i2o_exec_driver.context, &msg->u.s.icntxt);
-	writel(0x0106, &msg->u.s.tcntxt);	/* FIXME: why 0x0106, maybe in
-						   Spec? */
+	writel(0x00000000, &msg->u.s.tcntxt);
 	writel(PAGE_SIZE, &msg->body[0]);
 	/* Outbound msg frame size in words and Initcode */
-	writel(MSG_FRAME_SIZE << 16 | 0x80, &msg->body[1]);
+	writel(I2O_OUTBOUND_MSG_FRAME_SIZE << 16 | 0x80, &msg->body[1]);
 	writel(0xd0000004, &msg->body[2]);
 	writel(i2o_dma_low(c->status.phys), &msg->body[3]);
 	writel(i2o_dma_high(c->status.phys), &msg->body[4]);
@@ -503,17 +490,15 @@ static int i2o_iop_init_outbound_queue(struct i2o_controller *c)
 		}
 		set_current_state(TASK_UNINTERRUPTIBLE);
 		schedule_timeout(1);
-
-		rmb();
 	}
 
 	m = c->out_queue.phys;
 
 	/* Post frames */
-	for (i = 0; i < NMBR_MSG_FRAMES; i++) {
+	for (i = 0; i < I2O_MAX_OUTBOUND_MSG_FRAMES; i++) {
 		i2o_flush_reply(c, m);
 		udelay(1);	/* Promise */
-		m += MSG_FRAME_SIZE * 4;
+		m += I2O_OUTBOUND_MSG_FRAME_SIZE * sizeof(u32);
 	}
 
 	return 0;
@@ -530,20 +515,20 @@ static int i2o_iop_init_outbound_queue(struct i2o_controller *c)
  */
 static int i2o_iop_reset(struct i2o_controller *c)
 {
-	u8 *status = c->status.virt;
+	volatile u8 *status = c->status.virt;
 	struct i2o_message __iomem *msg;
 	u32 m;
 	unsigned long timeout;
 	i2o_status_block *sb = c->status_block.virt;
 	int rc = 0;
 
-	pr_debug("%s: Resetting controller\n", c->name);
+	osm_debug("%s: Resetting controller\n", c->name);
 
 	m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET);
 	if (m == I2O_QUEUE_EMPTY)
 		return -ETIMEDOUT;
 
-	memset(status, 0, 8);
+	memset(c->status_block.virt, 0, 8);
 
 	/* Quiesce all IOPs first */
 	i2o_iop_quiesce_all();
@@ -568,8 +553,6 @@ static int i2o_iop_reset(struct i2o_controller *c)
 
 		set_current_state(TASK_UNINTERRUPTIBLE);
 		schedule_timeout(1);
-
-		rmb();
 	}
 
 	switch (*status) {
@@ -984,11 +967,11 @@ int i2o_status_get(struct i2o_controller *c)
 {
 	struct i2o_message __iomem *msg;
 	u32 m;
-	u8 *status_block;
+	volatile u8 *status_block;
 	unsigned long timeout;
 
 	status_block = (u8 *) c->status_block.virt;
-	memset(status_block, 0, sizeof(i2o_status_block));
+	memset(c->status_block.virt, 0, sizeof(i2o_status_block));
 
 	m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_MESSAGE_GET);
 	if (m == I2O_QUEUE_EMPTY)
@@ -1017,8 +1000,6 @@ int i2o_status_get(struct i2o_controller *c)
 
 		set_current_state(TASK_UNINTERRUPTIBLE);
 		schedule_timeout(1);
-
-		rmb();
 	}
 
 #ifdef DEBUG
@@ -1107,6 +1088,11 @@ static void i2o_iop_release(struct device *dev)
 	i2o_iop_free(c);
 };
 
+/* I2O controller class */
+static struct class i2o_controller_class = {
+	.name = "i2o_controller",
+};
+
 /**
  *	i2o_iop_alloc - Allocate and initialize a i2o_controller struct
  *
@@ -1136,8 +1122,14 @@ struct i2o_controller *i2o_iop_alloc(void)
 	sprintf(c->name, "iop%d", c->unit);
 
 	device_initialize(&c->device);
+	class_device_initialize(&c->classdev);
+
 	c->device.release = &i2o_iop_release;
+	c->classdev.class = &i2o_controller_class;
+	c->classdev.dev = &c->device;
+
 	snprintf(c->device.bus_id, BUS_ID_SIZE, "iop%d", c->unit);
+	snprintf(c->classdev.class_id, BUS_ID_SIZE, "iop%d", c->unit);
 
 #if BITS_PER_LONG == 64
 	spin_lock_init(&c->context_list_lock);
@@ -1161,45 +1153,55 @@ int i2o_iop_add(struct i2o_controller *c)
 {
 	int rc;
 
-	if((rc = device_add(&c->device))) {
-		printk(KERN_ERR "%s: could not register controller\n", c->name);
+	if ((rc = device_add(&c->device))) {
+		osm_err("%s: could not add controller\n", c->name);
 		goto iop_reset;
 	}
 
-	printk(KERN_INFO "%s: Activating I2O controller...\n", c->name);
-	printk(KERN_INFO "%s: This may take a few minutes if there are many "
-	       "devices\n", c->name);
+	if ((rc = class_device_add(&c->classdev))) {
+		osm_err("%s: could not add controller class\n", c->name);
+		goto device_del;
+	}
+
+	osm_info("%s: Activating I2O controller...\n", c->name);
+	osm_info("%s: This may take a few minutes if there are many devices\n",
+		 c->name);
 
 	if ((rc = i2o_iop_activate(c))) {
-		printk(KERN_ERR "%s: could not activate controller\n",
-		       c->name);
-		goto iop_reset;
+		osm_err("%s: could not activate controller\n", c->name);
+		goto class_del;
 	}
 
-	pr_debug("%s: building sys table...\n", c->name);
+	osm_debug("%s: building sys table...\n", c->name);
 
 	if ((rc = i2o_systab_build()))
-		goto iop_reset;
+		goto class_del;
 
-	pr_debug("%s: online controller...\n", c->name);
+	osm_debug("%s: online controller...\n", c->name);
 
 	if ((rc = i2o_iop_online(c)))
-		goto iop_reset;
+		goto class_del;
 
-	pr_debug("%s: getting LCT...\n", c->name);
+	osm_debug("%s: getting LCT...\n", c->name);
 
 	if ((rc = i2o_exec_lct_get(c)))
-		goto iop_reset;
+		goto class_del;
 
 	list_add(&c->list, &i2o_controllers);
 
 	i2o_driver_notify_controller_add_all(c);
 
-	printk(KERN_INFO "%s: Controller added\n", c->name);
+	osm_info("%s: Controller added\n", c->name);
 
 	return 0;
 
-iop_reset:
+      class_del:
+	class_device_del(&c->classdev);
+
+      device_del:
+	device_del(&c->device);
+
+      iop_reset:
 	i2o_iop_reset(c);
 
 	return rc;
@@ -1260,16 +1262,18 @@ static int __init i2o_iop_init(void)
 	if (rc)
 		goto exit;
 
-	rc = i2o_driver_init();
-	if (rc)
+	if ((rc = class_register(&i2o_controller_class))) {
+		osm_err("can't register class i2o_controller\n");
 		goto device_exit;
+	}
 
-	rc = i2o_exec_init();
-	if (rc)
+	if ((rc = i2o_driver_init()))
+		goto class_exit;
+
+	if ((rc = i2o_exec_init()))
 		goto driver_exit;
 
-	rc = i2o_pci_init();
-	if (rc < 0)
+	if ((rc = i2o_pci_init()))
 		goto exec_exit;
 
 	return 0;
@@ -1280,6 +1284,9 @@ static int __init i2o_iop_init(void)
       driver_exit:
 	i2o_driver_exit();
 
+      class_exit:
+	class_unregister(&i2o_controller_class);
+
       device_exit:
 	i2o_device_exit();
 
@@ -1297,6 +1304,7 @@ static void __exit i2o_iop_exit(void)
 	i2o_pci_exit();
 	i2o_exec_exit();
 	i2o_driver_exit();
+	class_unregister(&i2o_controller_class);
 	i2o_device_exit();
 };
 
diff --git a/drivers/message/i2o/pci.c b/drivers/message/i2o/pci.c
index 964fe481849..442e34506b9 100644
--- a/drivers/message/i2o/pci.c
+++ b/drivers/message/i2o/pci.c
@@ -30,15 +30,7 @@
 #include <linux/pci.h>
 #include <linux/interrupt.h>
 #include <linux/i2o.h>
-
-/* Module internal functions from other sources */
-extern struct i2o_controller *i2o_iop_alloc(void);
-extern void i2o_iop_free(struct i2o_controller *);
-
-extern int i2o_iop_add(struct i2o_controller *);
-extern void i2o_iop_remove(struct i2o_controller *);
-
-extern int i2o_driver_dispatch(struct i2o_controller *, u32);
+#include "core.h"
 
 /* PCI device id table for all I2O controllers */
 static struct pci_device_id __devinitdata i2o_pci_ids[] = {
@@ -248,9 +240,7 @@ static int i2o_pci_irq_enable(struct i2o_controller *c)
 	struct pci_dev *pdev = c->pdev;
 	int rc;
 
-	wmb();
 	writel(0xffffffff, c->irq_mask);
-	wmb();
 
 	if (pdev->irq) {
 		rc = request_irq(pdev->irq, i2o_pci_interrupt, SA_SHIRQ,
@@ -263,7 +253,6 @@ static int i2o_pci_irq_enable(struct i2o_controller *c)
 	}
 
 	writel(0x00000000, c->irq_mask);
-	wmb();
 
 	printk(KERN_INFO "%s: Installed at IRQ %d\n", c->name, pdev->irq);
 
@@ -278,9 +267,7 @@ static int i2o_pci_irq_enable(struct i2o_controller *c)
  */
 static void i2o_pci_irq_disable(struct i2o_controller *c)
 {
-	wmb();
 	writel(0xffffffff, c->irq_mask);
-	wmb();
 
 	if (c->pdev->irq > 0)
 		free_irq(c->pdev->irq, c);
@@ -406,11 +393,11 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev,
 	if ((rc = i2o_iop_add(c)))
 		goto uninstall;
 
+	get_device(&c->device);
+
 	if (i960)
 		pci_write_config_word(i960, 0x42, 0x03ff);
 
-	get_device(&c->device);
-
 	return 0;
 
       uninstall:
@@ -478,6 +465,4 @@ void __exit i2o_pci_exit(void)
 {
 	pci_unregister_driver(&i2o_pci_driver);
 };
-
-EXPORT_SYMBOL(i2o_dma_realloc);
 MODULE_DEVICE_TABLE(pci, i2o_pci_ids);
diff --git a/include/linux/i2o-dev.h b/include/linux/i2o-dev.h
index d4a08d29e36..36fd18cdad2 100644
--- a/include/linux/i2o-dev.h
+++ b/include/linux/i2o-dev.h
@@ -32,18 +32,6 @@ typedef unsigned int u32;
 
 #endif				/* __KERNEL__ */
 
-/*
- *	Software module types
- */
-#define I2O_SOFTWARE_MODULE_IRTOS		0x11
-#define I2O_SOFTWARE_MODULE_IOP_PRIVATE		0x22
-#define I2O_SOFTWARE_MODULE_IOP_CONFIG		0x23
-
-/*
- *	Vendors
- */
-#define I2O_VENDOR_DPT				0x001b
-
 /*
  * I2O Control IOCTLs and structures
  */
@@ -414,6 +402,17 @@ typedef struct _i2o_status_block {
 #define ADAPTER_STATE_FAILED			0x10
 #define ADAPTER_STATE_FAULTED			0x11
 
+/*
+ *	Software module types
+ */
+#define I2O_SOFTWARE_MODULE_IRTOS		0x11
+#define I2O_SOFTWARE_MODULE_IOP_PRIVATE		0x22
+#define I2O_SOFTWARE_MODULE_IOP_CONFIG		0x23
+
+/*
+ *	Vendors
+ */
+#define I2O_VENDOR_DPT				0x001b
 
 /*
  * DPT / Adaptec specific values for i2o_sg_io_hdr flags.
diff --git a/include/linux/i2o.h b/include/linux/i2o.h
index 2039a87c2b9..be937d0372a 100644
--- a/include/linux/i2o.h
+++ b/include/linux/i2o.h
@@ -119,12 +119,21 @@ struct i2o_driver {
 };
 
 /*
- *	Contains all information which are necessary for DMA operations
+ *	Contains DMA mapped address information
  */
 struct i2o_dma {
 	void *virt;
 	dma_addr_t phys;
-	u32 len;
+	size_t len;
+};
+
+/*
+ *	Contains IO mapped address information
+ */
+struct i2o_io {
+	void __iomem *virt;
+	unsigned long phys;
+	unsigned long len;
 };
 
 /*
@@ -173,8 +182,8 @@ struct i2o_controller {
 	struct semaphore lct_lock;	/* Lock for LCT updates */
 	struct i2o_dma status_block;	/* IOP status block */
 
-	struct i2o_dma base;	/* controller messaging unit */
-	struct i2o_dma in_queue;	/* inbound message queue Host->IOP */
+	struct i2o_io base;	/* controller messaging unit */
+	struct i2o_io in_queue;	/* inbound message queue Host->IOP */
 	struct i2o_dma out_queue;	/* outbound message queue IOP->Host */
 
 	unsigned int battery:1;		/* Has a battery backup */
@@ -185,6 +194,7 @@ struct i2o_controller {
 	struct resource mem_resource;	/* Mem resource allocated to the IOP */
 
 	struct device device;
+	struct class_device classdev;	/* I2O controller class */
 	struct i2o_device *exec;	/* Executive */
 #if BITS_PER_LONG == 64
 	spinlock_t context_list_lock;	/* lock for context_list */
@@ -235,9 +245,10 @@ struct i2o_sys_tbl {
 extern struct list_head i2o_controllers;
 
 /* Message functions */
-static inline u32 i2o_msg_get(struct i2o_controller *, struct i2o_message __iomem **);
-extern u32 i2o_msg_get_wait(struct i2o_controller *, struct i2o_message __iomem **,
-			    int);
+static inline u32 i2o_msg_get(struct i2o_controller *,
+			      struct i2o_message __iomem **);
+extern u32 i2o_msg_get_wait(struct i2o_controller *,
+			    struct i2o_message __iomem **, int);
 static inline void i2o_msg_post(struct i2o_controller *, u32);
 static inline int i2o_msg_post_wait(struct i2o_controller *, u32,
 				    unsigned long);
@@ -638,14 +649,12 @@ extern int i2o_exec_lct_get(struct i2o_controller *);
  *	available returns I2O_QUEUE_EMPTY and msg is leaved untouched.
  */
 static inline u32 i2o_msg_get(struct i2o_controller *c,
-			      struct i2o_message __iomem **msg)
+			      struct i2o_message __iomem ** msg)
 {
 	u32 m = readl(c->in_port);
 
-	if (m != I2O_QUEUE_EMPTY) {
+	if (m != I2O_QUEUE_EMPTY)
 		*msg = c->in_queue.virt + m;
-		rmb();
-	}
 
 	return m;
 };
@@ -659,7 +668,6 @@ static inline u32 i2o_msg_get(struct i2o_controller *c,
  */
 static inline void i2o_msg_post(struct i2o_controller *c, u32 m)
 {
-	wmb();
 	writel(m, c->in_port);
 };
 
@@ -706,14 +714,11 @@ static inline void i2o_flush_reply(struct i2o_controller *c, u32 m)
  *	work for sender side messages as they are ioremap objects
  *	provided by the I2O controller.
  */
-static inline struct i2o_message __iomem *i2o_msg_out_to_virt(struct
-							      i2o_controller *c,
-							      u32 m)
+static inline struct i2o_message *i2o_msg_out_to_virt(struct i2o_controller *c,
+						      u32 m)
 {
-	if (unlikely
-	    (m < c->out_queue.phys
-	     || m >= c->out_queue.phys + c->out_queue.len))
-		return NULL;
+	BUG_ON(m < c->out_queue.phys
+	       || m >= c->out_queue.phys + c->out_queue.len);
 
 	return c->out_queue.virt + (m - c->out_queue.phys);
 };
@@ -729,69 +734,13 @@ static inline struct i2o_message __iomem *i2o_msg_out_to_virt(struct
  *	work for receive side messages as they are kmalloc objects
  *	in a different pool.
  */
-static inline struct i2o_message __iomem *i2o_msg_in_to_virt(struct i2o_controller *c,
-						     u32 m)
+static inline struct i2o_message __iomem *i2o_msg_in_to_virt(struct
+							     i2o_controller *c,
+							     u32 m)
 {
 	return c->in_queue.virt + m;
 };
 
-/**
- *	i2o_dma_alloc - Allocate DMA memory
- *	@dev: struct device pointer to the PCI device of the I2O controller
- *	@addr: i2o_dma struct which should get the DMA buffer
- *	@len: length of the new DMA memory
- *	@gfp_mask: GFP mask
- *
- *	Allocate a coherent DMA memory and write the pointers into addr.
- *
- *	Returns 0 on success or -ENOMEM on failure.
- */
-static inline int i2o_dma_alloc(struct device *dev, struct i2o_dma *addr,
-				size_t len, unsigned int gfp_mask)
-{
-	struct pci_dev *pdev = to_pci_dev(dev);
-	int dma_64 = 0;
-
-	if ((sizeof(dma_addr_t) > 4) && (pdev->dma_mask == DMA_64BIT_MASK)) {
-			dma_64 = 1;
-			if(pci_set_dma_mask(pdev, DMA_32BIT_MASK))
-				return -ENOMEM;
-	}
-
-	addr->virt = dma_alloc_coherent(dev, len, &addr->phys, gfp_mask);
-
-	if ((sizeof(dma_addr_t) > 4) && dma_64)
-		if(pci_set_dma_mask(pdev, DMA_64BIT_MASK))
-			printk(KERN_WARNING "i2o: unable to set 64-bit DMA");
-
-	if (!addr->virt)
-		return -ENOMEM;
-
-	memset(addr->virt, 0, len);
-	addr->len = len;
-
-	return 0;
-};
-
-/**
- *	i2o_dma_free - Free DMA memory
- *	@dev: struct device pointer to the PCI device of the I2O controller
- *	@addr: i2o_dma struct which contains the DMA buffer
- *
- *	Free a coherent DMA memory and set virtual address of addr to NULL.
- */
-static inline void i2o_dma_free(struct device *dev, struct i2o_dma *addr)
-{
-	if (addr->virt) {
-		if (addr->phys)
-			dma_free_coherent(dev, addr->len, addr->virt,
-					  addr->phys);
-		else
-			kfree(addr->virt);
-		addr->virt = NULL;
-	}
-};
-
 /*
  *	Endian handling wrapped into the macro - keeps the core code
  *	cleaner.
@@ -1141,16 +1090,13 @@ extern void i2o_debug_state(struct i2o_controller *c);
 #define ELEVEN_WORD_MSG_SIZE	0x000B0000
 #define I2O_MESSAGE_SIZE(x)	((x)<<16)
 
-/* Special TID Assignments */
-
+/* special TID assignments */
 #define ADAPTER_TID		0
 #define HOST_TID		1
 
-#define MSG_FRAME_SIZE		128	/* i2o_scsi assumes >= 32 */
-#define SG_TABLESIZE		30
-#define NMBR_MSG_FRAMES		128
-
-#define MSG_POOL_SIZE		(MSG_FRAME_SIZE*NMBR_MSG_FRAMES*sizeof(u32))
+/* outbound queue defines */
+#define I2O_MAX_OUTBOUND_MSG_FRAMES	128
+#define I2O_OUTBOUND_MSG_FRAME_SIZE	128	/* in 32-bit words */
 
 #define I2O_POST_WAIT_OK	0
 #define I2O_POST_WAIT_TIMEOUT	-ETIMEDOUT
-- 
cgit v1.2.3-70-g09d2


From f33213ecf49c98da4e85121b592c3bea8057c2e6 Mon Sep 17 00:00:00 2001
From: Markus Lidel <Markus.Lidel@shadowconnect.com>
Date: Thu, 23 Jun 2005 22:02:23 -0700
Subject: [PATCH] I2O: Lindent run and replacement of printk through osm
 printing functions

Lindent run and replaced printk() through the corresponding osm_*() function

Signed-off-by: Markus Lidel <Markus.Lidel@shadowconnect.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/message/i2o/Kconfig      |  10 +--
 drivers/message/i2o/device.c     |   1 -
 drivers/message/i2o/driver.c     |   3 +-
 drivers/message/i2o/exec-osm.c   |   2 +-
 drivers/message/i2o/i2o_block.c  |   3 +-
 drivers/message/i2o/i2o_block.h  |  28 ++++-----
 drivers/message/i2o/i2o_config.c |  20 +++---
 drivers/message/i2o/i2o_proc.c   |   2 +-
 drivers/message/i2o/iop.c        | 128 ++++++++++++++++++---------------------
 drivers/message/i2o/pci.c        |   5 +-
 include/linux/i2o.h              |   8 +--
 11 files changed, 100 insertions(+), 110 deletions(-)

(limited to 'include')

diff --git a/drivers/message/i2o/Kconfig b/drivers/message/i2o/Kconfig
index 94b6d676c5c..06e8eb19a05 100644
--- a/drivers/message/i2o/Kconfig
+++ b/drivers/message/i2o/Kconfig
@@ -44,8 +44,8 @@ config I2O_EXT_ADAPTEC_DMA64
 
 config I2O_CONFIG
 	tristate "I2O Configuration support"
-	depends on PCI && I2O
-	help
+	depends on I2O
+	---help---
 	  Say Y for support of the configuration interface for the I2O adapters.
 	  If you have a RAID controller from Adaptec and you want to use the
 	  raidutils to manage your RAID array, you have to say Y here.
@@ -74,7 +74,7 @@ config I2O_BUS
 config I2O_BLOCK
 	tristate "I2O Block OSM"
 	depends on I2O
-	help
+	---help---
 	  Include support for the I2O Block OSM. The Block OSM presents disk
 	  and other structured block devices to the operating system. If you
 	  are using an RAID controller, you could access the array only by
@@ -87,7 +87,7 @@ config I2O_BLOCK
 config I2O_SCSI
 	tristate "I2O SCSI OSM"
 	depends on I2O && SCSI
-	help
+	---help---
 	  Allows direct SCSI access to SCSI devices on a SCSI or FibreChannel
 	  I2O controller. You can use both the SCSI and Block OSM together if
 	  you wish. To access a RAID array, you must use the Block OSM driver.
@@ -99,7 +99,7 @@ config I2O_SCSI
 config I2O_PROC
 	tristate "I2O /proc support"
 	depends on I2O
-	help
+	---help---
 	  If you say Y here and to "/proc file system support", you will be
 	  able to read I2O related information from the virtual directory
 	  /proc/i2o.
diff --git a/drivers/message/i2o/device.c b/drivers/message/i2o/device.c
index 0ee342ea29b..d8d6e89a91c 100644
--- a/drivers/message/i2o/device.c
+++ b/drivers/message/i2o/device.c
@@ -443,7 +443,6 @@ static struct class_interface i2o_device_class_interface = {
  *	Note that the minimum sized reslist is 8 bytes and contains
  *	ResultCount, ErrorInfoSize, BlockStatus and BlockSize.
  */
-
 static int i2o_parm_issue(struct i2o_device *i2o_dev, int cmd, void *oplist,
 			  int oplen, void *reslist, int reslen)
 {
diff --git a/drivers/message/i2o/driver.c b/drivers/message/i2o/driver.c
index c32f9dbc574..739bfdef0c6 100644
--- a/drivers/message/i2o/driver.c
+++ b/drivers/message/i2o/driver.c
@@ -117,10 +117,9 @@ int i2o_driver_register(struct i2o_driver *drv)
 
 		i2o_driver_notify_controller_add(drv, c);
 		list_for_each_entry(i2o_dev, &c->devices, list)
-			i2o_driver_notify_device_add(drv, i2o_dev);
+		    i2o_driver_notify_device_add(drv, i2o_dev);
 	}
 
-
 	rc = driver_register(&drv->driver);
 	if (rc)
 		destroy_workqueue(drv->event_queue);
diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c
index ffe0cecfa06..1b7389876e7 100644
--- a/drivers/message/i2o/exec-osm.c
+++ b/drivers/message/i2o/exec-osm.c
@@ -152,7 +152,7 @@ int i2o_msg_post_wait_mem(struct i2o_controller *c, u32 m, unsigned long
 		list_add(&wait->list, &i2o_exec_wait_list);
 
 		wait_event_interruptible_timeout(wq, wait->complete,
-			timeout * HZ);
+						 timeout * HZ);
 
 		wait->wq = NULL;
 	}
diff --git a/drivers/message/i2o/i2o_block.c b/drivers/message/i2o/i2o_block.c
index 28b3918dbc1..f283b5bafdd 100644
--- a/drivers/message/i2o/i2o_block.c
+++ b/drivers/message/i2o/i2o_block.c
@@ -940,7 +940,6 @@ static void i2o_block_request_fn(struct request_queue *q)
 			INIT_WORK(&dreq->work, i2o_block_delayed_request_fn,
 				  dreq);
 
-			osm_info("transfer error\n");
 			if (!queue_delayed_work(i2o_block_driver.event_queue,
 						&dreq->work,
 						I2O_BLOCK_RETRY_TIME))
@@ -1042,8 +1041,8 @@ static struct i2o_block_device *i2o_block_device_alloc(void)
 static int i2o_block_probe(struct device *dev)
 {
 	struct i2o_device *i2o_dev = to_i2o_device(dev);
-	struct i2o_block_device *i2o_blk_dev;
 	struct i2o_controller *c = i2o_dev->iop;
+	struct i2o_block_device *i2o_blk_dev;
 	struct gendisk *gd;
 	struct request_queue *queue;
 	static int unit = 0;
diff --git a/drivers/message/i2o/i2o_block.h b/drivers/message/i2o/i2o_block.h
index e45cc40ce38..4fdaa5bda41 100644
--- a/drivers/message/i2o/i2o_block.h
+++ b/drivers/message/i2o/i2o_block.h
@@ -64,40 +64,38 @@
 
 /* I2O Block OSM mempool struct */
 struct i2o_block_mempool {
-	kmem_cache_t	*slab;
-	mempool_t	*pool;
+	kmem_cache_t *slab;
+	mempool_t *pool;
 };
 
 /* I2O Block device descriptor */
 struct i2o_block_device {
 	struct i2o_device *i2o_dev;	/* pointer to I2O device */
 	struct gendisk *gd;
-	spinlock_t lock;		/* queue lock */
+	spinlock_t lock;	/* queue lock */
 	struct list_head open_queue;	/* list of transfered, but unfinished
 					   requests */
 	unsigned int open_queue_depth;	/* number of requests in the queue */
 
-	int rcache;			/* read cache flags */
-	int wcache;			/* write cache flags */
+	int rcache;		/* read cache flags */
+	int wcache;		/* write cache flags */
 	int flags;
-	u16 power;			/* power state */
-	int media_change_flag;		/* media changed flag */
+	u16 power;		/* power state */
+	int media_change_flag;	/* media changed flag */
 };
 
 /* I2O Block device request */
-struct i2o_block_request
-{
+struct i2o_block_request {
 	struct list_head queue;
-	struct request *req;		/* corresponding request */
+	struct request *req;	/* corresponding request */
 	struct i2o_block_device *i2o_blk_dev;	/* I2O block device */
-	struct device *dev;		/* device used for DMA */
-	int sg_nents;			/* number of SG elements */
-	struct scatterlist sg_table[I2O_MAX_PHYS_SEGMENTS]; /* SG table */
+	struct device *dev;	/* device used for DMA */
+	int sg_nents;		/* number of SG elements */
+	struct scatterlist sg_table[I2O_MAX_PHYS_SEGMENTS];	/* SG table */
 };
 
 /* I2O Block device delayed request */
-struct i2o_block_delayed_request
-{
+struct i2o_block_delayed_request {
 	struct work_struct work;
 	struct request_queue *queue;
 };
diff --git a/drivers/message/i2o/i2o_config.c b/drivers/message/i2o/i2o_config.c
index 8160a1f6c73..8ebc86ff100 100644
--- a/drivers/message/i2o/i2o_config.c
+++ b/drivers/message/i2o/i2o_config.c
@@ -368,9 +368,9 @@ static int i2o_cfg_swul(unsigned long arg)
 
 	i2o_dma_free(&c->pdev->dev, &buffer);
 
-return_ret:
+      return_ret:
 	return ret;
-return_fault:
+      return_fault:
 	ret = -EFAULT;
 	goto return_ret;
 };
@@ -519,7 +519,8 @@ static int i2o_cfg_evt_get(unsigned long arg, struct file *fp)
 
 #ifdef CONFIG_I2O_EXT_ADAPTEC
 #ifdef CONFIG_COMPAT
-static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long arg)
+static int i2o_cfg_passthru32(struct file *file, unsigned cmnd,
+			      unsigned long arg)
 {
 	struct i2o_cmd_passthru32 __user *cmd;
 	struct i2o_controller *c;
@@ -646,8 +647,9 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar
 			    flag_count & 0x04000000 /*I2O_SGL_FLAGS_DIR */ ) {
 				// TODO 64bit fix
 				if (copy_from_user
-				    (p->virt, (void __user *)(unsigned long)sg[i].addr_bus,
-				     sg_size)) {
+				    (p->virt,
+				     (void __user *)(unsigned long)sg[i].
+				     addr_bus, sg_size)) {
 					printk(KERN_DEBUG
 					       "%s: Could not copy SG buf %d FROM user\n",
 					       c->name, i);
@@ -738,11 +740,12 @@ static int i2o_cfg_passthru32(struct file *file, unsigned cmnd, unsigned long ar
 	return rcode;
 }
 
-static long i2o_cfg_compat_ioctl(struct file *file, unsigned cmd, unsigned long arg)
+static long i2o_cfg_compat_ioctl(struct file *file, unsigned cmd,
+				 unsigned long arg)
 {
 	int ret;
-	lock_kernel();		
-	switch (cmd) { 
+	lock_kernel();
+	switch (cmd) {
 	case I2OGETIOPS:
 		ret = i2o_cfg_ioctl(NULL, file, cmd, arg);
 		break;
@@ -1136,6 +1139,7 @@ static int __init i2o_config_old_init(void)
 		osm_err("can't register device.\n");
 		return -EBUSY;
 	}
+
 	return 0;
 }
 
diff --git a/drivers/message/i2o/i2o_proc.c b/drivers/message/i2o/i2o_proc.c
index e5b74452c49..d559a175836 100644
--- a/drivers/message/i2o/i2o_proc.c
+++ b/drivers/message/i2o/i2o_proc.c
@@ -28,7 +28,7 @@
  */
 
 #define OSM_NAME	"proc-osm"
-#define OSM_VERSION	"$Rev$"
+#define OSM_VERSION	"1.145"
 #define OSM_DESCRIPTION	"I2O ProcFS OSM"
 
 #define I2O_MAX_MODULES 4
diff --git a/drivers/message/i2o/iop.c b/drivers/message/i2o/iop.c
index c32022bc2a2..42f8b810d6e 100644
--- a/drivers/message/i2o/iop.c
+++ b/drivers/message/i2o/iop.c
@@ -117,13 +117,13 @@ u32 i2o_cntxt_list_add(struct i2o_controller * c, void *ptr)
 	unsigned long flags;
 
 	if (!ptr)
-		printk(KERN_ERR "%s: couldn't add NULL pointer to context list!"
-		       "\n", c->name);
+		osm_err("%s: couldn't add NULL pointer to context list!\n",
+			c->name);
 
 	entry = kmalloc(sizeof(*entry), GFP_ATOMIC);
 	if (!entry) {
-		printk(KERN_ERR "%s: Could not allocate memory for context "
-		       "list element\n", c->name);
+		osm_err("%s: Could not allocate memory for context list element"
+			"\n", c->name);
 		return 0;
 	}
 
@@ -142,7 +142,7 @@ u32 i2o_cntxt_list_add(struct i2o_controller * c, void *ptr)
 
 	spin_unlock_irqrestore(&c->context_list_lock, flags);
 
-	pr_debug("%s: Add context to list %p -> %d\n", c->name, ptr, context);
+	osm_debug("%s: Add context to list %p -> %d\n", c->name, ptr, context);
 
 	return entry->context;
 };
@@ -174,11 +174,11 @@ u32 i2o_cntxt_list_remove(struct i2o_controller * c, void *ptr)
 	spin_unlock_irqrestore(&c->context_list_lock, flags);
 
 	if (!context)
-		printk(KERN_WARNING "%s: Could not remove nonexistent ptr "
-		       "%p\n", c->name, ptr);
+		osm_warn("%s: Could not remove nonexistent ptr %p\n", c->name,
+			 ptr);
 
-	pr_debug("%s: remove ptr from context list %d -> %p\n", c->name,
-		 context, ptr);
+	osm_debug("%s: remove ptr from context list %d -> %p\n", c->name,
+		  context, ptr);
 
 	return context;
 };
@@ -208,11 +208,10 @@ void *i2o_cntxt_list_get(struct i2o_controller *c, u32 context)
 	spin_unlock_irqrestore(&c->context_list_lock, flags);
 
 	if (!ptr)
-		printk(KERN_WARNING "%s: context id %d not found\n", c->name,
-		       context);
+		osm_warn("%s: context id %d not found\n", c->name, context);
 
-	pr_debug("%s: get ptr from context list %d -> %p\n", c->name, context,
-		 ptr);
+	osm_debug("%s: get ptr from context list %d -> %p\n", c->name, context,
+		  ptr);
 
 	return ptr;
 };
@@ -240,11 +239,11 @@ u32 i2o_cntxt_list_get_ptr(struct i2o_controller * c, void *ptr)
 	spin_unlock_irqrestore(&c->context_list_lock, flags);
 
 	if (!context)
-		printk(KERN_WARNING "%s: Could not find nonexistent ptr "
-		       "%p\n", c->name, ptr);
+		osm_warn("%s: Could not find nonexistent ptr %p\n", c->name,
+			 ptr);
 
-	pr_debug("%s: get context id from context list %p -> %d\n", c->name,
-		 ptr, context);
+	osm_debug("%s: get context id from context list %p -> %d\n", c->name,
+		  ptr, context);
 
 	return context;
 };
@@ -324,10 +323,9 @@ static int i2o_iop_quiesce(struct i2o_controller *c)
 
 	/* Long timeout needed for quiesce if lots of devices */
 	if ((rc = i2o_msg_post_wait(c, m, 240)))
-		printk(KERN_INFO "%s: Unable to quiesce (status=%#x).\n",
-		       c->name, -rc);
+		osm_info("%s: Unable to quiesce (status=%#x).\n", c->name, -rc);
 	else
-		pr_debug("%s: Quiesced.\n", c->name);
+		osm_debug("%s: Quiesced.\n", c->name);
 
 	i2o_status_get(c);	// Entered READY state
 
@@ -365,10 +363,9 @@ static int i2o_iop_enable(struct i2o_controller *c)
 
 	/* How long of a timeout do we need? */
 	if ((rc = i2o_msg_post_wait(c, m, 240)))
-		printk(KERN_ERR "%s: Could not enable (status=%#x).\n",
-		       c->name, -rc);
+		osm_err("%s: Could not enable (status=%#x).\n", c->name, -rc);
 	else
-		pr_debug("%s: Enabled.\n", c->name);
+		osm_debug("%s: Enabled.\n", c->name);
 
 	i2o_status_get(c);	// entered OPERATIONAL state
 
@@ -432,10 +429,9 @@ static int i2o_iop_clear(struct i2o_controller *c)
 	       &msg->u.head[1]);
 
 	if ((rc = i2o_msg_post_wait(c, m, 30)))
-		printk(KERN_INFO "%s: Unable to clear (status=%#x).\n",
-		       c->name, -rc);
+		osm_info("%s: Unable to clear (status=%#x).\n", c->name, -rc);
 	else
-		pr_debug("%s: Cleared.\n", c->name);
+		osm_debug("%s: Cleared.\n", c->name);
 
 	/* Enable all IOPs */
 	i2o_iop_enable_all();
@@ -570,14 +566,13 @@ static int i2o_iop_reset(struct i2o_controller *c)
 		 * can't read one in the given ammount of time, we assume the
 		 * IOP could not reboot properly.
 		 */
-		pr_debug("%s: Reset in progress, waiting for reboot...\n",
-			 c->name);
+		osm_debug("%s: Reset in progress, waiting for reboot...\n",
+			  c->name);
 
 		m = i2o_msg_get_wait(c, &msg, I2O_TIMEOUT_RESET);
 		while (m == I2O_QUEUE_EMPTY) {
 			if (time_after(jiffies, timeout)) {
-				printk(KERN_ERR "%s: IOP reset timeout.\n",
-				       c->name);
+				osm_err("%s: IOP reset timeout.\n", c->name);
 				rc = -ETIMEDOUT;
 				goto exit;
 			}
@@ -635,29 +630,29 @@ static int i2o_iop_activate(struct i2o_controller *c)
 
 	rc = i2o_status_get(c);
 	if (rc) {
-		printk(KERN_INFO "%s: Unable to obtain status, "
-		       "attempting a reset.\n", c->name);
+		osm_info("%s: Unable to obtain status, attempting a reset.\n",
+			 c->name);
 		rc = i2o_iop_reset(c);
 		if (rc)
 			return rc;
 	}
 
 	if (sb->i2o_version > I2OVER15) {
-		printk(KERN_ERR "%s: Not running version 1.5 of the I2O "
-		       "Specification.\n", c->name);
+		osm_err("%s: Not running version 1.5 of the I2O Specification."
+			"\n", c->name);
 		return -ENODEV;
 	}
 
 	switch (sb->iop_state) {
 	case ADAPTER_STATE_FAULTED:
-		printk(KERN_CRIT "%s: hardware fault\n", c->name);
+		osm_err("%s: hardware fault\n", c->name);
 		return -EFAULT;
 
 	case ADAPTER_STATE_READY:
 	case ADAPTER_STATE_OPERATIONAL:
 	case ADAPTER_STATE_HOLD:
 	case ADAPTER_STATE_FAILED:
-		pr_debug("%s: already running, trying to reset...\n", c->name);
+		osm_debug("%s: already running, trying to reset...\n", c->name);
 		rc = i2o_iop_reset(c);
 		if (rc)
 			return rc;
@@ -707,20 +702,18 @@ static int i2o_iop_systab_set(struct i2o_controller *c)
 		res->flags = IORESOURCE_MEM;
 		res->start = 0;
 		res->end = 0;
-		printk(KERN_INFO "%s: requires private memory resources.\n",
-		       c->name);
+		osm_info("%s: requires private memory resources.\n", c->name);
 		root = pci_find_parent_resource(c->pdev, res);
 		if (root == NULL)
-			printk(KERN_WARNING "%s: Can't find parent resource!\n",
-			       c->name);
+			osm_warn("%s: Can't find parent resource!\n", c->name);
 		if (root && allocate_resource(root, res, sb->desired_mem_size, sb->desired_mem_size, sb->desired_mem_size, 1 << 20,	/* Unspecified, so use 1Mb and play safe */
 					      NULL, NULL) >= 0) {
 			c->mem_alloc = 1;
 			sb->current_mem_size = 1 + res->end - res->start;
 			sb->current_mem_base = res->start;
-			printk(KERN_INFO "%s: allocated %ld bytes of PCI memory"
-			       " at 0x%08lX.\n", c->name,
-			       1 + res->end - res->start, res->start);
+			osm_info("%s: allocated %ld bytes of PCI memory at "
+				 "0x%08lX.\n", c->name,
+				 1 + res->end - res->start, res->start);
 		}
 	}
 
@@ -730,20 +723,18 @@ static int i2o_iop_systab_set(struct i2o_controller *c)
 		res->flags = IORESOURCE_IO;
 		res->start = 0;
 		res->end = 0;
-		printk(KERN_INFO "%s: requires private memory resources.\n",
-		       c->name);
+		osm_info("%s: requires private memory resources.\n", c->name);
 		root = pci_find_parent_resource(c->pdev, res);
 		if (root == NULL)
-			printk(KERN_WARNING "%s: Can't find parent resource!\n",
-			       c->name);
+			osm_warn("%s: Can't find parent resource!\n", c->name);
 		if (root && allocate_resource(root, res, sb->desired_io_size, sb->desired_io_size, sb->desired_io_size, 1 << 20,	/* Unspecified, so use 1Mb and play safe */
 					      NULL, NULL) >= 0) {
 			c->io_alloc = 1;
 			sb->current_io_size = 1 + res->end - res->start;
 			sb->current_mem_base = res->start;
-			printk(KERN_INFO "%s: allocated %ld bytes of PCI I/O at"
-			       " 0x%08lX.\n", c->name,
-			       1 + res->end - res->start, res->start);
+			osm_info("%s: allocated %ld bytes of PCI I/O at 0x%08lX"
+				 ".\n", c->name, 1 + res->end - res->start,
+				 res->start);
 		}
 	}
 
@@ -787,10 +778,10 @@ static int i2o_iop_systab_set(struct i2o_controller *c)
 			 PCI_DMA_TODEVICE);
 
 	if (rc < 0)
-		printk(KERN_ERR "%s: Unable to set SysTab (status=%#x).\n",
-		       c->name, -rc);
+		osm_err("%s: Unable to set SysTab (status=%#x).\n", c->name,
+			-rc);
 	else
-		pr_debug("%s: SysTab set.\n", c->name);
+		osm_debug("%s: SysTab set.\n", c->name);
 
 	i2o_status_get(c);	// Entered READY state
 
@@ -814,7 +805,7 @@ static int i2o_iop_online(struct i2o_controller *c)
 		return rc;
 
 	/* In READY state */
-	pr_debug("%s: Attempting to enable...\n", c->name);
+	osm_debug("%s: Attempting to enable...\n", c->name);
 	rc = i2o_iop_enable(c);
 	if (rc)
 		return rc;
@@ -833,7 +824,7 @@ void i2o_iop_remove(struct i2o_controller *c)
 {
 	struct i2o_device *dev, *tmp;
 
-	pr_debug("%s: deleting controller\n", c->name);
+	osm_debug("%s: deleting controller\n", c->name);
 
 	i2o_driver_notify_controller_remove_all(c);
 
@@ -882,8 +873,7 @@ static int i2o_systab_build(void)
 
 	systab = i2o_systab.virt = kmalloc(i2o_systab.len, GFP_KERNEL);
 	if (!systab) {
-		printk(KERN_ERR "i2o: unable to allocate memory for System "
-		       "Table\n");
+		osm_err("unable to allocate memory for System Table\n");
 		return -ENOMEM;
 	}
 	memset(systab, 0, i2o_systab.len);
@@ -895,8 +885,8 @@ static int i2o_systab_build(void)
 		i2o_status_block *sb;
 
 		if (count >= num_controllers) {
-			printk(KERN_ERR "i2o: controller added while building "
-			       "system table\n");
+			osm_err("controller added while building system table"
+				"\n");
 			break;
 		}
 
@@ -910,9 +900,8 @@ static int i2o_systab_build(void)
 		 * it is techninically not part of the I2O subsystem...
 		 */
 		if (unlikely(i2o_status_get(c))) {
-			printk(KERN_ERR "%s: Deleting b/c could not get status"
-			       " while attempting to build system table\n",
-			       c->name);
+			osm_err("%s: Deleting b/c could not get status while "
+				"attempting to build system table\n", c->name);
 			i2o_iop_remove(c);
 			continue;	// try the next one
 		}
@@ -994,7 +983,7 @@ int i2o_status_get(struct i2o_controller *c)
 	timeout = jiffies + I2O_TIMEOUT_STATUS_GET * HZ;
 	while (status_block[87] != 0xFF) {
 		if (time_after(jiffies, timeout)) {
-			printk(KERN_ERR "%s: Get status timeout.\n", c->name);
+			osm_err("%s: Get status timeout.\n", c->name);
 			return -ETIMEDOUT;
 		}
 
@@ -1043,8 +1032,8 @@ static int i2o_hrt_get(struct i2o_controller *c)
 		rc = i2o_msg_post_wait_mem(c, m, 20, &c->hrt);
 
 		if (rc < 0) {
-			printk(KERN_ERR "%s: Unable to get HRT (status=%#x)\n",
-			       c->name, -rc);
+			osm_err("%s: Unable to get HRT (status=%#x)\n", c->name,
+				-rc);
 			return rc;
 		}
 
@@ -1058,8 +1047,8 @@ static int i2o_hrt_get(struct i2o_controller *c)
 			return i2o_parse_hrt(c);
 	}
 
-	printk(KERN_ERR "%s: Unable to get HRT after %d tries, giving up\n",
-	       c->name, I2O_HRT_GET_TRIES);
+	osm_err("%s: Unable to get HRT after %d tries, giving up\n", c->name,
+		I2O_HRT_GET_TRIES);
 
 	return -EBUSY;
 }
@@ -1073,7 +1062,6 @@ void i2o_iop_free(struct i2o_controller *c)
 	kfree(c);
 };
 
-
 /**
  *	i2o_iop_release - release the memory for a I2O controller
  *	@dev: I2O controller which should be released
@@ -1109,8 +1097,8 @@ struct i2o_controller *i2o_iop_alloc(void)
 
 	c = kmalloc(sizeof(*c), GFP_KERNEL);
 	if (!c) {
-		printk(KERN_ERR "i2o: Insufficient memory to allocate a I2O "
-		       "controller.\n");
+		osm_err("i2o: Insufficient memory to allocate a I2O controller."
+			"\n");
 		return ERR_PTR(-ENOMEM);
 	}
 	memset(c, 0, sizeof(*c));
diff --git a/drivers/message/i2o/pci.c b/drivers/message/i2o/pci.c
index 442e34506b9..9971430e518 100644
--- a/drivers/message/i2o/pci.c
+++ b/drivers/message/i2o/pci.c
@@ -179,7 +179,10 @@ static int __devinit i2o_pci_alloc(struct i2o_controller *c)
 		return -ENOMEM;
 	}
 
-	if (i2o_dma_alloc(dev, &c->out_queue, MSG_POOL_SIZE, GFP_KERNEL)) {
+	if (i2o_dma_alloc
+	    (dev, &c->out_queue,
+	     I2O_MAX_OUTBOUND_MSG_FRAMES * I2O_OUTBOUND_MSG_FRAME_SIZE *
+	     sizeof(u32), GFP_KERNEL)) {
 		i2o_pci_free(c);
 		return -ENOMEM;
 	}
diff --git a/include/linux/i2o.h b/include/linux/i2o.h
index be937d0372a..bdc286ec947 100644
--- a/include/linux/i2o.h
+++ b/include/linux/i2o.h
@@ -156,8 +156,8 @@ struct i2o_controller {
 
 	struct pci_dev *pdev;	/* PCI device */
 
-	unsigned int promise:1;		/* Promise controller */
-	unsigned int adaptec:1;		/* DPT / Adaptec controller */
+	unsigned int promise:1;	/* Promise controller */
+	unsigned int adaptec:1;	/* DPT / Adaptec controller */
 	unsigned int raptor:1;	/* split bar */
 	unsigned int no_quiesce:1;	/* dont quiesce before reset */
 	unsigned int short_req:1;	/* use small block sizes */
@@ -174,7 +174,7 @@ struct i2o_controller {
 
 	/* Dynamic LCT related data */
 
-	struct i2o_dma status;	/* status of IOP */
+	struct i2o_dma status;	/* IOP status block */
 
 	struct i2o_dma hrt;	/* HW Resource Table */
 	i2o_lct *lct;		/* Logical Config Table */
@@ -186,7 +186,7 @@ struct i2o_controller {
 	struct i2o_io in_queue;	/* inbound message queue Host->IOP */
 	struct i2o_dma out_queue;	/* outbound message queue IOP->Host */
 
-	unsigned int battery:1;		/* Has a battery backup */
+	unsigned int battery:1;	/* Has a battery backup */
 	unsigned int io_alloc:1;	/* An I/O resource was allocated */
 	unsigned int mem_alloc:1;	/* A memory resource was allocated */
 
-- 
cgit v1.2.3-70-g09d2


From 391cd727eac2e10be7685efd739a3ea9de87393c Mon Sep 17 00:00:00 2001
From: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
Date: Thu, 23 Jun 2005 22:02:43 -0700
Subject: [PATCH] tuner-core.c improvments and Ymec Tvision TVF8533MF support

tuner-core.c, tuner.h:

- tuner-core changed to support multiple I2C devices used on some
  adapters;

- Kconfig now has an option (CONFIG_TUNER_MULTI_I2C) to enable this new
  behavor;

- By default, even enabling CONFIG_TUNER_MULTI_I2C, tuner-core emulates
  the old behavor, using first I2C device for both FM and TV;

- There is a new i2c command (TUNER_SET_ADDR) to allow tuner clients to
  select I2C address for FM or TV tuner;

- Tuner I2C dettach now generates a warning on syslog if failed.

tuner-simple.c:

- TVision TVF-8531MF and TVF-5533 MF tuner included.  It uses, by
  default, I2C on 0xC2 address for TV and on 0xC0 for Radio.  Both TV and
  FM Radio mode are working.

Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/media/video/Kconfig        | 13 +++++++
 drivers/media/video/tuner-core.c   | 79 ++++++++++++++++++++++++++++++++++++--
 drivers/media/video/tuner-simple.c | 28 +++++++++++++-
 include/media/tuner.h              | 10 +++++
 4 files changed, 126 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/drivers/media/video/Kconfig b/drivers/media/video/Kconfig
index 6c05fddb69a..8c349706f85 100644
--- a/drivers/media/video/Kconfig
+++ b/drivers/media/video/Kconfig
@@ -7,6 +7,19 @@ menu "Video For Linux"
 
 comment "Video Adapters"
 
+config CONFIG_TUNER_MULTI_I2C
+	bool "Enable support for multiple I2C devices on Video Adapters (EXPERIMENTAL)"
+	depends on VIDEO_DEV && EXPERIMENTAL
+	---help---
+	  Some video adapters have more than one tuner inside. This patch
+	  enables support for using more than one tuner. This is required
+	  for some cards to allow tunning  both video and radio.
+	  It also improves I2C autodetection for these cards.
+
+	  Only few tuners currently is supporting this. More to come.
+
+	  It is safe to say 'Y' here even if your card has only one I2C tuner.
+
 config VIDEO_BT848
 	tristate "BT848 Video For Linux"
 	depends on VIDEO_DEV && PCI && I2C
diff --git a/drivers/media/video/tuner-core.c b/drivers/media/video/tuner-core.c
index 81882ddab85..71423ae3b4d 100644
--- a/drivers/media/video/tuner-core.c
+++ b/drivers/media/video/tuner-core.c
@@ -1,5 +1,5 @@
 /*
- * $Id: tuner-core.c,v 1.5 2005/02/15 15:59:35 kraxel Exp $
+ * $Id: tuner-core.c,v 1.7 2005/05/30 02:02:47 mchehab Exp $
  *
  * i2c tv tuner chip device driver
  * core core, i.e. kernel interfaces, registering and so on
@@ -23,6 +23,11 @@
 #include <media/tuner.h>
 #include <media/audiochip.h>
 
+/*
+ * comment line bellow to return to old behavor, where only one I2C device is supported
+ */
+/* #define CONFIG_TUNER_MULTI_I2C */
+
 #define UNSET (-1U)
 
 /* standard i2c insmod options */
@@ -53,6 +58,9 @@ MODULE_AUTHOR("Ralph Metzler, Gerd Knorr, Gunther Mayer");
 MODULE_LICENSE("GPL");
 
 static int this_adap;
+#ifdef CONFIG_TUNER_MULTI_I2C
+static unsigned short tv_tuner, radio_tuner;
+#endif
 
 static struct i2c_driver driver;
 static struct i2c_client client_template;
@@ -125,6 +133,28 @@ static void set_freq(struct i2c_client *c, unsigned long freq)
 	t->freq = freq;
 }
 
+#ifdef CONFIG_TUNER_MULTI_I2C
+static void set_addr(struct i2c_client *c, struct tuner_addr *tun_addr)
+{
+	struct tuner *t = i2c_get_clientdata(c);
+
+	switch (tun_addr->type) {
+	case V4L2_TUNER_RADIO:
+ 		radio_tuner=tun_addr->addr;
+		tuner_dbg("radio tuner set to I2C address 0x%02x\n",radio_tuner<<1);
+
+		break;
+	default:
+		tv_tuner=tun_addr->addr;
+		tuner_dbg("TV tuner set to I2C address 0x%02x\n",tv_tuner<<1);
+		break;
+	}
+}
+#else
+#define set_addr(c,tun_addr) \
+		tuner_warn("It is recommended to enable CONFIG_TUNER_MULTI_I2C for this card.\n");
+#endif
+
 static void set_type(struct i2c_client *c, unsigned int type)
 {
 	struct tuner *t = i2c_get_clientdata(c);
@@ -197,8 +227,16 @@ static int tuner_attach(struct i2c_adapter *adap, int addr, int kind)
 {
 	struct tuner *t;
 
+#ifndef CONFIG_TUNER_MULTI_I2C
 	if (this_adap > 0)
 		return -1;
+#else
+	/* by default, first I2C card is both tv and radio tuner */
+	if (this_adap == 0) {
+		tv_tuner = addr;
+		radio_tuner = addr;
+	}
+#endif
 	this_adap++;
 
         client_template.adapter = adap;
@@ -228,6 +266,11 @@ static int tuner_probe(struct i2c_adapter *adap)
 	}
 	this_adap = 0;
 
+#ifdef CONFIG_TUNER_MULTI_I2C
+	tv_tuner = 0;
+	radio_tuner = 0;
+#endif
+
 	if (adap->class & I2C_CLASS_TV_ANALOG)
 		return i2c_probe(adap, &addr_data, tuner_attach);
 	return 0;
@@ -236,8 +279,14 @@ static int tuner_probe(struct i2c_adapter *adap)
 static int tuner_detach(struct i2c_client *client)
 {
 	struct tuner *t = i2c_get_clientdata(client);
+	int err;
+
+	err=i2c_detach_client(&t->i2c);
+	if (err) {
+		tuner_warn ("Client deregistration failed, client not detached.\n");
+		return err;
+	}
 
-	i2c_detach_client(&t->i2c);
 	kfree(t);
 	return 0;
 }
@@ -249,6 +298,17 @@ static int tuner_detach(struct i2c_client *client)
 			  tuner_info("ignore v4l1 call\n"); \
 		          return 0; }
 
+#ifdef CONFIG_TUNER_MULTI_I2C
+#define CHECK_ADDR(tp,cmd)	if (client->addr!=tp) { \
+			  tuner_info ("Cmd %s to addr 0x%02x rejected.\n",cmd,client->addr<<1); \
+			  return 0; }
+#define CHECK_MODE(cmd)	if (t->mode == V4L2_TUNER_RADIO) { \
+			  CHECK_ADDR(radio_tuner,cmd) } else { CHECK_ADDR(tv_tuner,cmd); }
+#else
+#define CHECK_ADDR(tp,cmd)
+#define CHECK_MODE(cmd)
+#endif
+
 static int
 tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 {
@@ -256,18 +316,23 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
         unsigned int *iarg = (int*)arg;
 
         switch (cmd) {
-
 	/* --- configuration --- */
 	case TUNER_SET_TYPE:
 		set_type(client,*iarg);
 		break;
+	case TUNER_SET_ADDR:
+		set_addr(client,(struct tuner_addr *)arg);
+		break;
 	case AUDC_SET_RADIO:
+		CHECK_ADDR(radio_tuner,"AUDC_SET_RADIO");
+
 		if (V4L2_TUNER_RADIO != t->mode) {
 			set_tv_freq(client,400 * 16);
 			t->mode = V4L2_TUNER_RADIO;
 		}
 		break;
 	case AUDC_CONFIG_PINNACLE:
+		CHECK_ADDR(tv_tuner,"AUDC_CONFIG_PINNACLE");
 		switch (*iarg) {
 		case 2:
 			tuner_dbg("pinnacle pal\n");
@@ -295,6 +360,7 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 		};
 		struct video_channel *vc = arg;
 
+		CHECK_ADDR(tv_tuner,"VIDIOCSCHAN");
 		CHECK_V4L2;
 		t->mode = V4L2_TUNER_ANALOG_TV;
 		if (vc->norm < ARRAY_SIZE(map))
@@ -308,6 +374,7 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 	{
 		unsigned long *v = arg;
 
+		CHECK_MODE("VIDIOCSFREQ");
 		CHECK_V4L2;
 		set_freq(client,*v);
 		return 0;
@@ -316,6 +383,7 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 	{
 		struct video_tuner *vt = arg;
 
+		CHECK_ADDR(radio_tuner,"VIDIOCGTUNER:");
 		CHECK_V4L2;
 		if (V4L2_TUNER_RADIO == t->mode  &&  t->has_signal)
 			vt->signal = t->has_signal(client);
@@ -325,6 +393,7 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 	{
 		struct video_audio *va = arg;
 
+		CHECK_ADDR(radio_tuner,"VIDIOCGAUDIO");
 		CHECK_V4L2;
 		if (V4L2_TUNER_RADIO == t->mode  &&  t->is_stereo)
 			va->mode = t->is_stereo(client)
@@ -337,6 +406,7 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 	{
 		v4l2_std_id *id = arg;
 
+		CHECK_ADDR(tv_tuner,"VIDIOC_S_STD");
 		SWITCH_V4L2;
 		t->mode = V4L2_TUNER_ANALOG_TV;
 		t->std = *id;
@@ -349,6 +419,7 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 	{
 		struct v4l2_frequency *f = arg;
 
+		CHECK_MODE("VIDIOC_S_FREQUENCY");
 		SWITCH_V4L2;
 		if (V4L2_TUNER_RADIO == f->type &&
 		    V4L2_TUNER_RADIO != t->mode)
@@ -361,6 +432,7 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 	{
 		struct v4l2_frequency *f = arg;
 
+		CHECK_MODE("VIDIOC_G_FREQUENCY");
 		SWITCH_V4L2;
 		f->type = t->mode;
 		f->frequency = t->freq;
@@ -370,6 +442,7 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 	{
 		struct v4l2_tuner *tuner = arg;
 
+		CHECK_MODE("VIDIOC_G_TUNER");
 		SWITCH_V4L2;
 		if (V4L2_TUNER_RADIO == t->mode  &&  t->has_signal)
 			tuner->signal = t->has_signal(client);
diff --git a/drivers/media/video/tuner-simple.c b/drivers/media/video/tuner-simple.c
index 48c6ceff1dc..f7305c8d53d 100644
--- a/drivers/media/video/tuner-simple.c
+++ b/drivers/media/video/tuner-simple.c
@@ -1,5 +1,5 @@
 /*
- * $Id: tuner-simple.c,v 1.10 2005/03/08 08:38:00 kraxel Exp $
+ * $Id: tuner-simple.c,v 1.14 2005/05/30 02:02:47 mchehab Exp $
  *
  * i2c tv tuner chip device driver
  * controls all those simple 4-control-bytes style tuners.
@@ -212,6 +212,11 @@ static struct tunertype tuners[] = {
         { "Philips FQ1236A MK4", Philips, NTSC,
           16*160.00,16*442.00,0x01,0x02,0x04,0x8e,732 },
 
+	/* Should work for TVF8531MF, TVF8831MF, TVF8731MF */
+	{ "Ymec TVision TVF-8531MF", Philips, NTSC,
+	  16*160.00,16*454.00,0xa0,0x90,0x30,0x8e,732},
+	{ "Ymec TVision TVF-5533MF", Philips, NTSC,
+	  16*160.00,16*454.00,0x01,0x02,0x04,0x8e,732},
 };
 unsigned const int tuner_count = ARRAY_SIZE(tuners);
 
@@ -424,6 +429,13 @@ static void default_set_radio_freq(struct i2c_client *c, unsigned int freq)
 	buffer[2] = tun->config;
 
 	switch (t->type) {
+	case TUNER_YMEC_TVF_5533MF:
+
+		/*These values are empirically determinated */
+		div = (freq*122)/16 - 20;
+		buffer[2] = 0x88; /* could be also 0x80 */
+		buffer[3] = 0x19; /* could be also 0x10, 0x18, 0x99 */
+		break;
 	case TUNER_PHILIPS_FM1216ME_MK3:
 	case TUNER_PHILIPS_FM1236_MK3:
 		buffer[3] = 0x19;
@@ -458,6 +470,20 @@ int default_tuner_init(struct i2c_client *c)
 		   t->type, tuners[t->type].name);
 	strlcpy(c->name, tuners[t->type].name, sizeof(c->name));
 
+	switch (t->type) {
+	case TUNER_YMEC_TVF_5533MF:
+		{
+			struct tuner_addr tun_addr = { V4L2_TUNER_ANALOG_TV, 0xc2>>1 };
+
+			if (c->driver->command) {
+				c->driver->command(c, TUNER_SET_ADDR, &tun_addr);
+			} else {
+				tuner_warn("Couldn't set TV tuner I2C address to 0x%02x\n",tun_addr.addr<<1);
+			}
+			break;
+		}
+	}
+
 	t->tv_freq    = default_set_tv_freq;
 	t->radio_freq = default_set_radio_freq;
 	t->has_signal = tuner_signal;
diff --git a/include/media/tuner.h b/include/media/tuner.h
index 156a9c51ffe..6d29d8db0c8 100644
--- a/include/media/tuner.h
+++ b/include/media/tuner.h
@@ -98,6 +98,9 @@
 #define TUNER_PHILIPS_FQ1216AME_MK4 56 /* Hauppauge PVR-150 PAL */
 #define TUNER_PHILIPS_FQ1236A_MK4 57   /* Hauppauge PVR-500MCE NTSC */
 
+#define TUNER_YMEC_TVF_8531MF 58
+#define TUNER_YMEC_TVF_5533MF 59	/* Pixelview Pro Ultra NTSC */
+
 #define NOTUNER 0
 #define PAL     1	/* PAL_BG */
 #define PAL_I   2
@@ -121,8 +124,10 @@
 
 #define TUNER_SET_TYPE               _IOW('t',1,int)    /* set tuner type */
 #define TUNER_SET_TVFREQ             _IOW('t',2,int)    /* set tv freq */
+#define TUNER_SET_ADDR               _IOW('T',3,int)	/* Chooses tuner I2C address */
 
 #define  TDA9887_SET_CONFIG          _IOW('t',5,int)
+
 /* tv card specific */
 # define TDA9887_PRESENT             (1<<0)
 # define TDA9887_PORT1_INACTIVE      (1<<1)
@@ -143,6 +148,11 @@
 #define I2C_ADDR_TDA8290        0x4b
 #define I2C_ADDR_TDA8275        0x61
 
+struct tuner_addr {
+	enum v4l2_tuner_type type;
+	unsigned short addr;
+};
+
 struct tuner {
 	/* device */
 	struct i2c_client i2c;
-- 
cgit v1.2.3-70-g09d2


From 496400014f22c4dbdbc1e89249a2feba46939708 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Thu, 23 Jun 2005 22:02:58 -0700
Subject: [PATCH] nfsd4: fix fh_expire_type

We're returning NFS4_FH_NOEXPIRE_WITH_OPEN | NFS4_FH_VOL_RENAME for the
fh_expire_type attribute.  This is incorrect:
	1. The spec actually only allows NOEXPIRE_WITH_OPEN when
	   VOLATILE_ANY is also set.
	2. Filehandles for open files can expire, if the file is removed
	   and there is a reboot.
	3. Filehandles are only volatile on rename in the nosubtree check
	   case.

Unfortunately, there's no way to indicate that we only expire on remove.  So
our only choice is FH4_VOLATILE_ANY.  Although it's redundant, we also set
FH4_VOL_RENAME in the subtree check case, since subtreecheck does actually
cause problems in practice and it seems possibly useful to give clients some
way to distinguish that case.

Fix a mispelled #define while we're at it.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfsd/nfs4xdr.c    | 5 ++++-
 include/linux/nfs4.h | 2 +-
 2 files changed, 5 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index 36a058a112d..0ae1467c3bc 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -1366,7 +1366,10 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp,
 	if (bmval0 & FATTR4_WORD0_FH_EXPIRE_TYPE) {
 		if ((buflen -= 4) < 0)
 			goto out_resource;
-		WRITE32( NFS4_FH_NOEXPIRE_WITH_OPEN | NFS4_FH_VOL_RENAME );
+		if (exp->ex_flags & NFSEXP_NOSUBTREECHECK)
+			WRITE32(NFS4_FH_VOLATILE_ANY);
+		else
+			WRITE32(NFS4_FH_VOLATILE_ANY|NFS4_FH_VOL_RENAME);
 	}
 	if (bmval0 & FATTR4_WORD0_CHANGE) {
 		/*
diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index 5bb5b2fd7ba..0c1c306cdae 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -28,7 +28,7 @@
 #define NFS4_ACCESS_DELETE      0x0010
 #define NFS4_ACCESS_EXECUTE     0x0020
 
-#define NFS4_FH_PERISTENT		0x0000
+#define NFS4_FH_PERSISTENT		0x0000
 #define NFS4_FH_NOEXPIRE_WITH_OPEN	0x0001
 #define NFS4_FH_VOLATILE_ANY		0x0002
 #define NFS4_FH_VOL_MIGRATION		0x0004
-- 
cgit v1.2.3-70-g09d2


From 8beefa249371f55432394ac96864c83b0b309c28 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Thu, 23 Jun 2005 22:03:08 -0700
Subject: [PATCH] nfsd4: rename nfs4_file fields

Trivial renaming patch:

I can never remember, while looking at various lists relating the nfsd4 state
structures, which are the "heads" and which are items on other lists, or which
structures are actually on the various lists.  The following convention helps
me: given structures foo and bar, with foo containing the head of a list of
bars, use "bars" for the name of the head of the list contained in the struct
foo, and use "per_foo" for the entries in the struct bars.

Go ahead and do this for struct nfs4_file.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfsd/nfs4state.c        | 22 +++++++++++-----------
 include/linux/nfsd/state.h |  4 ++--
 2 files changed, 13 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index f03a4180fa1..a84a80e8c0c 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -153,7 +153,7 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_f
 		        current_fh->fh_handle.fh_size);
 	dp->dl_time = 0;
 	atomic_set(&dp->dl_count, 1);
-	list_add(&dp->dl_del_perfile, &fp->fi_del_perfile);
+	list_add(&dp->dl_del_perfile, &fp->fi_delegations);
 	list_add(&dp->dl_del_perclnt, &clp->cl_del_perclnt);
 	return dp;
 }
@@ -954,8 +954,8 @@ alloc_init_file(struct inode *ino)
 	fp = kmem_cache_alloc(file_slab, GFP_KERNEL);
 	if (fp) {
 		INIT_LIST_HEAD(&fp->fi_hash);
-		INIT_LIST_HEAD(&fp->fi_perfile);
-		INIT_LIST_HEAD(&fp->fi_del_perfile);
+		INIT_LIST_HEAD(&fp->fi_stateids);
+		INIT_LIST_HEAD(&fp->fi_delegations);
 		list_add(&fp->fi_hash, &file_hashtbl[hashval]);
 		fp->fi_inode = igrab(ino);
 		fp->fi_id = current_fileid++;
@@ -974,7 +974,7 @@ release_all_files(void)
 		while (!list_empty(&file_hashtbl[i])) {
 			fp = list_entry(file_hashtbl[i].next, struct nfs4_file, fi_hash);
 			/* this should never be more than once... */
-			if (!list_empty(&fp->fi_perfile) || !list_empty(&fp->fi_del_perfile)) {
+			if (!list_empty(&fp->fi_stateids) || !list_empty(&fp->fi_delegations)) {
 				printk("ERROR: release_all_files: file %p is open, creating dangling state !!!\n",fp);
 			}
 			release_file(fp);
@@ -1139,7 +1139,7 @@ init_stateid(struct nfs4_stateid *stp, struct nfs4_file *fp, struct nfsd4_open *
 	INIT_LIST_HEAD(&stp->st_perfile);
 	list_add(&stp->st_hash, &stateid_hashtbl[hashval]);
 	list_add(&stp->st_perfilestate, &sop->so_perfilestate);
-	list_add(&stp->st_perfile, &fp->fi_perfile);
+	list_add(&stp->st_perfile, &fp->fi_stateids);
 	stp->st_stateowner = sop;
 	stp->st_file = fp;
 	stp->st_stateid.si_boot = boot_time;
@@ -1204,7 +1204,7 @@ release_state_owner(struct nfs4_stateid *stp, int flag)
 	if (sop->so_confirmed && list_empty(&sop->so_perfilestate))
 		move_to_close_lru(sop);
 	/* unused nfs4_file's are releseed. XXX slab cache? */
-	if (list_empty(&fp->fi_perfile) && list_empty(&fp->fi_del_perfile)) {
+	if (list_empty(&fp->fi_stateids) && list_empty(&fp->fi_delegations)) {
 		release_file(fp);
 	}
 }
@@ -1294,7 +1294,7 @@ nfs4_share_conflict(struct svc_fh *current_fh, unsigned int deny_type)
 	fp = find_file(ino);
 	if (fp) {
 	/* Search for conflicting share reservations */
-		list_for_each_entry(stp, &fp->fi_perfile, st_perfile) {
+		list_for_each_entry(stp, &fp->fi_stateids, st_perfile) {
 			if (test_bit(deny_type, &stp->st_deny_bmap) ||
 			    test_bit(NFS4_SHARE_DENY_BOTH, &stp->st_deny_bmap))
 				return nfserr_share_denied;
@@ -1545,7 +1545,7 @@ find_delegation_file(struct nfs4_file *fp, stateid_t *stid)
 {
 	struct nfs4_delegation *dp;
 
-	list_for_each_entry(dp, &fp->fi_del_perfile, dl_del_perfile) {
+	list_for_each_entry(dp, &fp->fi_delegations, dl_del_perfile) {
 		if (dp->dl_stateid.si_stateownerid == stid->si_stateownerid)
 			return dp;
 	}
@@ -1583,7 +1583,7 @@ nfs4_check_open(struct nfs4_file *fp, struct nfsd4_open *open, struct nfs4_state
 	int status = nfserr_share_denied;
 	struct nfs4_stateowner *sop = open->op_stateowner;
 
-	list_for_each_entry(local, &fp->fi_perfile, st_perfile) {
+	list_for_each_entry(local, &fp->fi_stateids, st_perfile) {
 		/* ignore lock owners */
 		if (local->st_stateowner->so_is_open_owner == 0)
 			continue;
@@ -1830,7 +1830,7 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf
 	            stp->st_stateid.si_fileid, stp->st_stateid.si_generation);
 out:
 	/* take the opportunity to clean up unused state */
-	if (fp && list_empty(&fp->fi_perfile) && list_empty(&fp->fi_del_perfile))
+	if (fp && list_empty(&fp->fi_stateids) && list_empty(&fp->fi_delegations))
 		release_file(fp);
 
 	/* CLAIM_PREVIOUS has different error returns */
@@ -2633,7 +2633,7 @@ alloc_init_lock_stateid(struct nfs4_stateowner *sop, struct nfs4_file *fp, struc
 	INIT_LIST_HEAD(&stp->st_perfilestate);
 	INIT_LIST_HEAD(&stp->st_perlockowner); /* not used */
 	list_add(&stp->st_hash, &lockstateid_hashtbl[hashval]);
-	list_add(&stp->st_perfile, &fp->fi_perfile);
+	list_add(&stp->st_perfile, &fp->fi_stateids);
 	list_add(&stp->st_perfilestate, &sop->so_perfilestate);
 	stp->st_stateowner = sop;
 	stp->st_file = fp;
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index b6b2fe1e7c6..2c3b42674a4 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -218,8 +218,8 @@ struct nfs4_stateowner {
 */
 struct nfs4_file {
 	struct list_head        fi_hash;    /* hash by "struct inode *" */
-	struct list_head        fi_perfile; /* list: nfs4_stateid */
-	struct list_head	fi_del_perfile; /* list: nfs4_delegation */
+	struct list_head        fi_stateids;
+	struct list_head	fi_delegations;
 	struct inode		*fi_inode;
 	u32                     fi_id;      /* used with stateowner->so_id 
 					     * for stateid_hashtbl hash */
-- 
cgit v1.2.3-70-g09d2


From 13cd21845d6a9729ca95e36ae6e8c669623fbfd4 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Thu, 23 Jun 2005 22:03:10 -0700
Subject: [PATCH] nfsd4: reference count struct nfs4_file

Add a struct kref to each nfs4_file and take a reference to it from each
stateid and delegation that refers to it.  The atomicity guarantees are
overkill given that all this stuff is done under the single nfsd4 state lock,
but a) we'd like finer-grained locking some day, and b) this simplifies the
cleanup of the structures a bit, something that has previously been a bit
complicated and bug-prone.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfsd/nfs4state.c        | 100 +++++++++++++++++++++++----------------------
 include/linux/nfsd/state.h |   1 +
 2 files changed, 52 insertions(+), 49 deletions(-)

(limited to 'include')

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index a84a80e8c0c..6ba428afa43 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -111,7 +111,6 @@ opaque_hashval(const void *ptr, int nbytes)
 /* forward declarations */
 static void release_stateowner(struct nfs4_stateowner *sop);
 static void release_stateid(struct nfs4_stateid *stp, int flags);
-static void release_file(struct nfs4_file *fp);
 
 /*
  * Delegation state
@@ -121,6 +120,27 @@ static void release_file(struct nfs4_file *fp);
 spinlock_t recall_lock;
 static struct list_head del_recall_lru;
 
+static void
+free_nfs4_file(struct kref *kref)
+{
+	struct nfs4_file *fp = container_of(kref, struct nfs4_file, fi_ref);
+	list_del(&fp->fi_hash);
+	iput(fp->fi_inode);
+	kmem_cache_free(file_slab, fp);
+}
+
+static inline void
+put_nfs4_file(struct nfs4_file *fi)
+{
+	kref_put(&fi->fi_ref, free_nfs4_file);
+}
+
+static inline void
+get_nfs4_file(struct nfs4_file *fi)
+{
+	kref_get(&fi->fi_ref);
+}
+
 static struct nfs4_delegation *
 alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_fh *current_fh, u32 type)
 {
@@ -136,6 +156,7 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_f
 	INIT_LIST_HEAD(&dp->dl_del_perclnt);
 	INIT_LIST_HEAD(&dp->dl_recall_lru);
 	dp->dl_client = clp;
+	get_nfs4_file(fp);
 	dp->dl_file = fp;
 	dp->dl_flock = NULL;
 	get_file(stp->st_vfs_file);
@@ -163,6 +184,7 @@ nfs4_put_delegation(struct nfs4_delegation *dp)
 {
 	if (atomic_dec_and_test(&dp->dl_count)) {
 		dprintk("NFSD: freeing dp %p\n",dp);
+		put_nfs4_file(dp->dl_file);
 		kmem_cache_free(deleg_slab, dp);
 	}
 }
@@ -953,6 +975,7 @@ alloc_init_file(struct inode *ino)
 
 	fp = kmem_cache_alloc(file_slab, GFP_KERNEL);
 	if (fp) {
+		kref_init(&fp->fi_ref);
 		INIT_LIST_HEAD(&fp->fi_hash);
 		INIT_LIST_HEAD(&fp->fi_stateids);
 		INIT_LIST_HEAD(&fp->fi_delegations);
@@ -964,24 +987,6 @@ alloc_init_file(struct inode *ino)
 	return NULL;
 }
 
-static void
-release_all_files(void)
-{
-	int i;
-	struct nfs4_file *fp;
-
-	for (i=0;i<FILE_HASH_SIZE;i++) {
-		while (!list_empty(&file_hashtbl[i])) {
-			fp = list_entry(file_hashtbl[i].next, struct nfs4_file, fi_hash);
-			/* this should never be more than once... */
-			if (!list_empty(&fp->fi_stateids) || !list_empty(&fp->fi_delegations)) {
-				printk("ERROR: release_all_files: file %p is open, creating dangling state !!!\n",fp);
-			}
-			release_file(fp);
-		}
-	}
-}
-
 static void
 nfsd4_free_slab(kmem_cache_t **slab)
 {
@@ -1141,6 +1146,7 @@ init_stateid(struct nfs4_stateid *stp, struct nfs4_file *fp, struct nfsd4_open *
 	list_add(&stp->st_perfilestate, &sop->so_perfilestate);
 	list_add(&stp->st_perfile, &fp->fi_stateids);
 	stp->st_stateowner = sop;
+	get_nfs4_file(fp);
 	stp->st_file = fp;
 	stp->st_stateid.si_boot = boot_time;
 	stp->st_stateid.si_stateownerid = sop->so_id;
@@ -1166,18 +1172,11 @@ release_stateid(struct nfs4_stateid *stp, int flags)
 		nfsd_close(filp);
 	} else if (flags & LOCK_STATE)
 		locks_remove_posix(filp, (fl_owner_t) stp->st_stateowner);
+	put_nfs4_file(stp->st_file);
 	kmem_cache_free(stateid_slab, stp);
 	stp = NULL;
 }
 
-static void
-release_file(struct nfs4_file *fp)
-{
-	list_del(&fp->fi_hash);
-	iput(fp->fi_inode);
-	kmem_cache_free(file_slab, fp);
-}	
-
 void
 move_to_close_lru(struct nfs4_stateowner *sop)
 {
@@ -1192,7 +1191,6 @@ void
 release_state_owner(struct nfs4_stateid *stp, int flag)
 {
 	struct nfs4_stateowner *sop = stp->st_stateowner;
-	struct nfs4_file *fp = stp->st_file;
 
 	dprintk("NFSD: release_state_owner\n");
 	release_stateid(stp, flag);
@@ -1203,10 +1201,6 @@ release_state_owner(struct nfs4_stateid *stp, int flag)
 	 */
 	if (sop->so_confirmed && list_empty(&sop->so_perfilestate))
 		move_to_close_lru(sop);
-	/* unused nfs4_file's are releseed. XXX slab cache? */
-	if (list_empty(&fp->fi_stateids) && list_empty(&fp->fi_delegations)) {
-		release_file(fp);
-	}
 }
 
 static int
@@ -1236,8 +1230,10 @@ find_file(struct inode *ino)
 	struct nfs4_file *fp;
 
 	list_for_each_entry(fp, &file_hashtbl[hashval], fi_hash) {
-		if (fp->fi_inode == ino)
+		if (fp->fi_inode == ino) {
+			get_nfs4_file(fp);
 			return fp;
+		}
 	}
 	return NULL;
 }
@@ -1288,19 +1284,24 @@ nfs4_share_conflict(struct svc_fh *current_fh, unsigned int deny_type)
 	struct inode *ino = current_fh->fh_dentry->d_inode;
 	struct nfs4_file *fp;
 	struct nfs4_stateid *stp;
+	int ret;
 
 	dprintk("NFSD: nfs4_share_conflict\n");
 
 	fp = find_file(ino);
-	if (fp) {
+	if (!fp)
+		return nfs_ok;
+	ret = nfserr_share_denied;
 	/* Search for conflicting share reservations */
-		list_for_each_entry(stp, &fp->fi_stateids, st_perfile) {
-			if (test_bit(deny_type, &stp->st_deny_bmap) ||
-			    test_bit(NFS4_SHARE_DENY_BOTH, &stp->st_deny_bmap))
-				return nfserr_share_denied;
-		}
+	list_for_each_entry(stp, &fp->fi_stateids, st_perfile) {
+		if (test_bit(deny_type, &stp->st_deny_bmap) ||
+		    test_bit(NFS4_SHARE_DENY_BOTH, &stp->st_deny_bmap))
+			goto out;
 	}
-	return nfs_ok;
+	ret = nfs_ok;
+out:
+	put_nfs4_file(fp);
+	return ret;
 }
 
 static inline void
@@ -1829,10 +1830,8 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf
 	            stp->st_stateid.si_boot, stp->st_stateid.si_stateownerid,
 	            stp->st_stateid.si_fileid, stp->st_stateid.si_generation);
 out:
-	/* take the opportunity to clean up unused state */
-	if (fp && list_empty(&fp->fi_stateids) && list_empty(&fp->fi_delegations))
-		release_file(fp);
-
+	if (fp)
+		put_nfs4_file(fp);
 	/* CLAIM_PREVIOUS has different error returns */
 	nfs4_set_claim_prev(open, &status);
 	/*
@@ -2480,16 +2479,19 @@ find_stateid(stateid_t *stid, int flags)
 static struct nfs4_delegation *
 find_delegation_stateid(struct inode *ino, stateid_t *stid)
 {
-	struct nfs4_file *fp = NULL;
+	struct nfs4_file *fp;
+	struct nfs4_delegation *dl;
 
 	dprintk("NFSD:find_delegation_stateid stateid=(%08x/%08x/%08x/%08x)\n",
                     stid->si_boot, stid->si_stateownerid,
                     stid->si_fileid, stid->si_generation);
 
 	fp = find_file(ino);
-	if (fp)
-		return find_delegation_file(fp, stid);
-	return NULL;
+	if (!fp)
+		return NULL;
+	dl = find_delegation_file(fp, stid);
+	put_nfs4_file(fp);
+	return dl;
 }
 
 /*
@@ -2636,6 +2638,7 @@ alloc_init_lock_stateid(struct nfs4_stateowner *sop, struct nfs4_file *fp, struc
 	list_add(&stp->st_perfile, &fp->fi_stateids);
 	list_add(&stp->st_perfilestate, &sop->so_perfilestate);
 	stp->st_stateowner = sop;
+	get_nfs4_file(fp);
 	stp->st_file = fp;
 	stp->st_stateid.si_boot = boot_time;
 	stp->st_stateid.si_stateownerid = sop->so_id;
@@ -3287,7 +3290,6 @@ __nfs4_state_shutdown(void)
 		unhash_delegation(dp);
 	}
 
-	release_all_files();
 	cancel_delayed_work(&laundromat_work);
 	flush_scheduled_work();
 	nfs4_init = 0;
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index 2c3b42674a4..296e6429fc3 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -217,6 +217,7 @@ struct nfs4_stateowner {
 *      share_acces, share_deny on the file.
 */
 struct nfs4_file {
+	struct kref		fi_ref;
 	struct list_head        fi_hash;    /* hash by "struct inode *" */
 	struct list_head        fi_stateids;
 	struct list_head	fi_delegations;
-- 
cgit v1.2.3-70-g09d2


From 7b190fecfa33d72bcf74c9473134c2ad14ae9545 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Thu, 23 Jun 2005 22:03:23 -0700
Subject: [PATCH] knfsd: nfsd4: delegation recovery

Allow recovery of delegations after reboot.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfsd/nfs4state.c       | 36 ++++++++++++++++++++++++++++--------
 fs/nfsd/nfs4xdr.c         |  2 +-
 include/linux/nfsd/xdr4.h |  1 +
 3 files changed, 30 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 16c9a43218c..0f6119714c8 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -1709,14 +1709,30 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, struct nfs4_sta
 	int status, flag = 0;
 
 	flag = NFS4_OPEN_DELEGATE_NONE;
-	if (open->op_claim_type != NFS4_OPEN_CLAIM_NULL
-	     || !atomic_read(&cb->cb_set) || !sop->so_confirmed)
-		goto out;
-
-	if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE)
-		flag = NFS4_OPEN_DELEGATE_WRITE;
-	else
-		flag = NFS4_OPEN_DELEGATE_READ;
+	open->op_recall = 0;
+	switch (open->op_claim_type) {
+		case NFS4_OPEN_CLAIM_PREVIOUS:
+			if (!atomic_read(&cb->cb_set))
+				open->op_recall = 1;
+			flag = open->op_delegate_type;
+			if (flag == NFS4_OPEN_DELEGATE_NONE)
+				goto out;
+			break;
+		case NFS4_OPEN_CLAIM_NULL:
+			/* Let's not give out any delegations till everyone's
+			 * had the chance to reclaim theirs.... */
+			if (nfs4_in_grace())
+				goto out;
+			if (!atomic_read(&cb->cb_set) || !sop->so_confirmed)
+				goto out;
+			if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE)
+				flag = NFS4_OPEN_DELEGATE_WRITE;
+			else
+				flag = NFS4_OPEN_DELEGATE_READ;
+			break;
+		default:
+			goto out;
+	}
 
 	dp = alloc_init_deleg(sop->so_client, stp, fh, flag);
 	if (dp == NULL) {
@@ -1750,6 +1766,10 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, struct nfs4_sta
 	             dp->dl_stateid.si_fileid,
 	             dp->dl_stateid.si_generation);
 out:
+	if (open->op_claim_type == NFS4_OPEN_CLAIM_PREVIOUS
+			&& flag == NFS4_OPEN_DELEGATE_NONE
+			&& open->op_delegate_type != NFS4_OPEN_DELEGATE_NONE)
+		printk("NFSD: WARNING: refusing delegation reclaim\n");
 	open->op_delegate_type = flag;
 }
 
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index 0ae1467c3bc..cfe978a72ce 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -1972,7 +1972,7 @@ nfsd4_encode_open(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_open
 	case NFS4_OPEN_DELEGATE_READ:
 		RESERVE_SPACE(20 + sizeof(stateid_t));
 		WRITEMEM(&open->op_delegate_stateid, sizeof(stateid_t));
-		WRITE32(0);
+		WRITE32(open->op_recall);
 
 		/*
 		 * TODO: ACE's in delegations
diff --git a/include/linux/nfsd/xdr4.h b/include/linux/nfsd/xdr4.h
index a1f5ad0be1b..4d24d65c0e8 100644
--- a/include/linux/nfsd/xdr4.h
+++ b/include/linux/nfsd/xdr4.h
@@ -210,6 +210,7 @@ struct nfsd4_open {
 	u32		op_share_access;    /* request */
 	u32		op_share_deny;      /* request */
 	stateid_t	op_stateid;         /* response */
+	u32		op_recall;          /* recall */
 	struct nfsd4_change_info  op_cinfo; /* response */
 	u32		op_rflags;          /* response */
 	int		op_truncate;        /* used during processing */
-- 
cgit v1.2.3-70-g09d2


From 76a3550ec50ed86885a10a767ebaebb7c9104721 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Thu, 23 Jun 2005 22:03:26 -0700
Subject: [PATCH] knfsd: nfsd4: rename nfs4_state_init

Somewhat gratuitous rename to simplify following patch.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfsd/nfs4state.c       | 6 +++---
 fs/nfsd/nfssvc.c          | 2 +-
 include/linux/nfsd/nfsd.h | 4 ++--
 3 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 0f6119714c8..e00b3472851 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -3185,7 +3185,7 @@ nfs4_check_open_reclaim(clientid_t *clid)
  */
 
 static void
-__nfs4_state_init(void)
+__nfs4_state_start(void)
 {
 	int i;
 	time_t grace_time;
@@ -3235,7 +3235,7 @@ __nfs4_state_init(void)
 }
 
 int
-nfs4_state_init(void)
+nfs4_state_start(void)
 {
 	int status;
 
@@ -3244,7 +3244,7 @@ nfs4_state_init(void)
 	status = nfsd4_init_slabs();
 	if (status)
 		return status;
-	__nfs4_state_init();
+	__nfs4_state_start();
 	nfs4_init = 1;
 	return 0;
 }
diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index 904df604e86..07b9a065e9d 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -95,7 +95,7 @@ nfsd_svc(unsigned short port, int nrservs)
 	error =	nfsd_racache_init(2*nrservs);
 	if (error<0)
 		goto out;
-	error = nfs4_state_init();
+	error = nfs4_state_start();
 	if (error<0)
 		goto out;
 	if (!nfsd_serv) {
diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h
index 4bf931d5ff5..3855fdc5af7 100644
--- a/include/linux/nfsd/nfsd.h
+++ b/include/linux/nfsd/nfsd.h
@@ -145,12 +145,12 @@ int nfsd_set_posix_acl(struct svc_fh *, int, struct posix_acl *);
  * NFSv4 State
  */
 #ifdef CONFIG_NFSD_V4
-int nfs4_state_init(void);
+int nfs4_state_start(void);
 void nfs4_state_shutdown(void);
 time_t nfs4_lease_time(void);
 void nfs4_reset_lease(time_t leasetime);
 #else
-static inline int nfs4_state_init(void){return 0;}
+static inline int nfs4_state_start(void){return 0;}
 static inline void nfs4_state_shutdown(void){}
 static inline time_t nfs4_lease_time(void){return 0;}
 static inline void nfs4_reset_lease(time_t leasetime){}
-- 
cgit v1.2.3-70-g09d2


From ac4d8ff2a57179de3ef7834c6ab3fac430b0a05d Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Thu, 23 Jun 2005 22:03:30 -0700
Subject: [PATCH] knfsd: nfsd4: clean up state initialization

Separate out stuff that needs initialization on startup from stuff that only
needs initialization on module init from static data.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfsd/nfs4state.c       | 35 +++++++++++++++++------------------
 fs/nfsd/nfsctl.c          |  1 +
 include/linux/nfsd/nfsd.h |  2 ++
 3 files changed, 20 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index e00b3472851..1f68ce36e72 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -117,7 +117,7 @@ static void release_stateid(struct nfs4_stateid *stp, int flags);
  */
 
 /* recall_lock protects the del_recall_lru */
-spinlock_t recall_lock;
+spinlock_t recall_lock = SPIN_LOCK_UNLOCKED;
 static struct list_head del_recall_lru;
 
 static void
@@ -3179,23 +3179,13 @@ nfs4_check_open_reclaim(clientid_t *clid)
 	return nfs4_find_reclaim_client(clid) ? nfs_ok : nfserr_reclaim_bad;
 }
 
+/* initialization to perform at module load time: */
 
-/* 
- * Start and stop routines
- */
-
-static void
-__nfs4_state_start(void)
+void
+nfs4_state_init(void)
 {
 	int i;
-	time_t grace_time;
 
-	if (!nfs4_reclaim_init) {
-		for (i = 0; i < CLIENT_HASH_SIZE; i++)
-			INIT_LIST_HEAD(&reclaim_str_hashtbl[i]);
-		reclaim_str_hashtbl_size = 0;
-		nfs4_reclaim_init = 1;
-	}
 	for (i = 0; i < CLIENT_HASH_SIZE; i++) {
 		INIT_LIST_HEAD(&conf_id_hashtbl[i]);
 		INIT_LIST_HEAD(&conf_str_hashtbl[i]);
@@ -3217,19 +3207,28 @@ __nfs4_state_start(void)
 		INIT_LIST_HEAD(&lock_ownerid_hashtbl[i]);
 		INIT_LIST_HEAD(&lock_ownerstr_hashtbl[i]);
 	}
-	memset(&zerostateid, 0, sizeof(stateid_t));
 	memset(&onestateid, ~0, sizeof(stateid_t));
-
 	INIT_LIST_HEAD(&close_lru);
 	INIT_LIST_HEAD(&client_lru);
 	INIT_LIST_HEAD(&del_recall_lru);
-	spin_lock_init(&recall_lock);
+	for (i = 0; i < CLIENT_HASH_SIZE; i++)
+		INIT_LIST_HEAD(&reclaim_str_hashtbl[i]);
+	reclaim_str_hashtbl_size = 0;
+	nfs4_reclaim_init = 1;
+}
+
+/* initialization to perform when the nfsd service is started: */
+
+static void
+__nfs4_state_start(void)
+{
+	time_t grace_time;
+
 	boot_time = get_seconds();
 	grace_time = max(user_lease_time, lease_time);
 	lease_time = user_lease_time;
 	printk("NFSD: starting %ld-second grace period\n", grace_time);
 	grace_end = boot_time + grace_time;
-	INIT_WORK(&laundromat_work,laundromat_main, NULL);
 	laundry_wq = create_singlethread_workqueue("nfsd4");
 	queue_delayed_work(laundry_wq, &laundromat_work, NFSD_LEASE_TIME*HZ);
 }
diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 161afdcb8f7..3d56531a7a0 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -397,6 +397,7 @@ static int __init init_nfsd(void)
 	nfsd_cache_init();	/* RPC reply cache */
 	nfsd_export_init();	/* Exports table */
 	nfsd_lockd_init();	/* lockd->nfsd callbacks */
+	nfs4_state_init();	/* NFSv4 locking state */
 #ifdef CONFIG_NFSD_V4
 	nfsd_idmap_init();      /* Name to ID mapping */
 #endif /* CONFIG_NFSD_V4 */
diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h
index 3855fdc5af7..21c6e9d86e4 100644
--- a/include/linux/nfsd/nfsd.h
+++ b/include/linux/nfsd/nfsd.h
@@ -145,11 +145,13 @@ int nfsd_set_posix_acl(struct svc_fh *, int, struct posix_acl *);
  * NFSv4 State
  */
 #ifdef CONFIG_NFSD_V4
+void nfs4_state_init(void);
 int nfs4_state_start(void);
 void nfs4_state_shutdown(void);
 time_t nfs4_lease_time(void);
 void nfs4_reset_lease(time_t leasetime);
 #else
+static inline void nfs4_state_init(void){};
 static inline int nfs4_state_start(void){return 0;}
 static inline void nfs4_state_shutdown(void){}
 static inline time_t nfs4_lease_time(void){return 0;}
-- 
cgit v1.2.3-70-g09d2


From bd0b1e954e3ba3e5d2cab941458cf98206471bd2 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Thu, 23 Jun 2005 22:03:35 -0700
Subject: [PATCH] knfsd: nfsd4: idmap initialization

Adopt standard kernel style by defining a no-op function instead of putting
ifdef's in the code where the function is called.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfsd/nfsctl.c           | 4 ----
 include/linux/nfsd_idmap.h | 5 +++++
 2 files changed, 5 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 3d56531a7a0..3da43a3ed32 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -398,9 +398,7 @@ static int __init init_nfsd(void)
 	nfsd_export_init();	/* Exports table */
 	nfsd_lockd_init();	/* lockd->nfsd callbacks */
 	nfs4_state_init();	/* NFSv4 locking state */
-#ifdef CONFIG_NFSD_V4
 	nfsd_idmap_init();      /* Name to ID mapping */
-#endif /* CONFIG_NFSD_V4 */
 	if (proc_mkdir("fs/nfs", NULL)) {
 		struct proc_dir_entry *entry;
 		entry = create_proc_entry("fs/nfs/exports", 0, NULL);
@@ -427,9 +425,7 @@ static void __exit exit_nfsd(void)
 	remove_proc_entry("fs/nfs", NULL);
 	nfsd_stat_shutdown();
 	nfsd_lockd_shutdown();
-#ifdef CONFIG_NFSD_V4
 	nfsd_idmap_shutdown();
-#endif /* CONFIG_NFSD_V4 */
 	unregister_filesystem(&nfsd_fs_type);
 }
 
diff --git a/include/linux/nfsd_idmap.h b/include/linux/nfsd_idmap.h
index 9bb7f30e923..e82746fcad1 100644
--- a/include/linux/nfsd_idmap.h
+++ b/include/linux/nfsd_idmap.h
@@ -43,8 +43,13 @@
 /* XXX from linux/nfs_idmap.h */
 #define IDMAP_NAMESZ 128
 
+#ifdef CONFIG_NFSD_V4
 void nfsd_idmap_init(void);
 void nfsd_idmap_shutdown(void);
+#else
+static inline void nfsd_idmap_init(void) {};
+static inline void nfsd_idmap_shutdown(void) {};
+#endif
 
 int nfsd_map_name_to_uid(struct svc_rqst *, const char *, size_t, __u32 *);
 int nfsd_map_name_to_gid(struct svc_rqst *, const char *, size_t, __u32 *);
-- 
cgit v1.2.3-70-g09d2


From a55370a3c0106106a975c5a09cee800611d0cf50 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Thu, 23 Jun 2005 22:03:52 -0700
Subject: [PATCH] knfsd: nfsd4: reboot hash

For the purposes of reboot recovery we keep a directory with subdirectories
each having a name that is the ascii hex representation of the md5 sum of a
client identifier for an active client.

This adds the code to calculate that name.  We also use it for the purposes of
comparing clients, so if someone ever manages to find two client names that
are md5 collisions, then we'll return clid_inuse to the second.

Signed-off-by: Andy Adamson <andros@citi.umich.edu>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/Kconfig                 |  2 +
 fs/nfsd/Makefile           |  2 +-
 fs/nfsd/nfs4recover.c      | 97 ++++++++++++++++++++++++++++++++++++++++++++++
 fs/nfsd/nfs4state.c        | 80 ++++++++++++++++++--------------------
 include/linux/nfsd/state.h |  6 ++-
 5 files changed, 143 insertions(+), 44 deletions(-)
 create mode 100644 fs/nfsd/nfs4recover.c

(limited to 'include')

diff --git a/fs/Kconfig b/fs/Kconfig
index a7c0cc3203c..5c704d05627 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -1413,6 +1413,8 @@ config NFSD_V4
 	bool "Provide NFSv4 server support (EXPERIMENTAL)"
 	depends on NFSD_V3 && EXPERIMENTAL
 	select NFSD_TCP
+	select CRYPTO_MD5
+	select CRYPTO
 	help
 	  If you would like to include the NFSv4 server as well as the NFSv2
 	  and NFSv3 servers, say Y here.  This feature is experimental, and
diff --git a/fs/nfsd/Makefile b/fs/nfsd/Makefile
index 9f043f44c92..ce341dc76d5 100644
--- a/fs/nfsd/Makefile
+++ b/fs/nfsd/Makefile
@@ -10,5 +10,5 @@ nfsd-$(CONFIG_NFSD_V2_ACL) += nfs2acl.o
 nfsd-$(CONFIG_NFSD_V3)	+= nfs3proc.o nfs3xdr.o
 nfsd-$(CONFIG_NFSD_V3_ACL) += nfs3acl.o
 nfsd-$(CONFIG_NFSD_V4)	+= nfs4proc.o nfs4xdr.o nfs4state.o nfs4idmap.o \
-			   nfs4acl.o nfs4callback.o
+			   nfs4acl.o nfs4callback.o nfs4recover.o
 nfsd-objs		:= $(nfsd-y)
diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c
new file mode 100644
index 00000000000..841a305d794
--- /dev/null
+++ b/fs/nfsd/nfs4recover.c
@@ -0,0 +1,97 @@
+/*
+*  linux/fs/nfsd/nfs4recover.c
+*
+*  Copyright (c) 2004 The Regents of the University of Michigan.
+*  All rights reserved.
+*
+*  Andy Adamson <andros@citi.umich.edu>
+*
+*  Redistribution and use in source and binary forms, with or without
+*  modification, are permitted provided that the following conditions
+*  are met:
+*
+*  1. Redistributions of source code must retain the above copyright
+*     notice, this list of conditions and the following disclaimer.
+*  2. Redistributions in binary form must reproduce the above copyright
+*     notice, this list of conditions and the following disclaimer in the
+*     documentation and/or other materials provided with the distribution.
+*  3. Neither the name of the University nor the names of its
+*     contributors may be used to endorse or promote products derived
+*     from this software without specific prior written permission.
+*
+*  THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
+*  WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+*  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+*  DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+*  FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+*  CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+*  SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
+*  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
+*  LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+*  NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+*  SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+*
+*/
+
+
+#include <linux/sunrpc/svc.h>
+#include <linux/nfsd/nfsd.h>
+#include <linux/nfs4.h>
+#include <linux/nfsd/state.h>
+#include <linux/nfsd/xdr4.h>
+#include <asm/uaccess.h>
+#include <asm/scatterlist.h>
+#include <linux/crypto.h>
+
+
+#define NFSDDBG_FACILITY                NFSDDBG_PROC
+
+static void
+md5_to_hex(char *out, char *md5)
+{
+	int i;
+
+	for (i=0; i<16; i++) {
+		unsigned char c = md5[i];
+
+		*out++ = '0' + ((c&0xf0)>>4) + (c>=0xa0)*('a'-'9'-1);
+		*out++ = '0' + (c&0x0f) + ((c&0x0f)>=0x0a)*('a'-'9'-1);
+	}
+	*out = '\0';
+}
+
+int
+nfs4_make_rec_clidname(char *dname, struct xdr_netobj *clname)
+{
+	struct xdr_netobj cksum;
+	struct crypto_tfm *tfm;
+	struct scatterlist sg[1];
+	int status = nfserr_resource;
+
+	dprintk("NFSD: nfs4_make_rec_clidname for %.*s\n",
+			clname->len, clname->data);
+	tfm = crypto_alloc_tfm("md5", 0);
+	if (tfm == NULL)
+		goto out;
+	cksum.len = crypto_tfm_alg_digestsize(tfm);
+	cksum.data = kmalloc(cksum.len, GFP_KERNEL);
+	if (cksum.data == NULL)
+ 		goto out;
+	crypto_digest_init(tfm);
+
+	sg[0].page = virt_to_page(clname->data);
+	sg[0].offset = offset_in_page(clname->data);
+	sg[0].length = clname->len;
+
+	crypto_digest_update(tfm, sg, 1);
+	crypto_digest_final(tfm, cksum.data);
+
+	md5_to_hex(dname, cksum.data);
+
+	kfree(cksum.data);
+	status = nfs_ok;
+out:
+	if (tfm)
+		crypto_free_tfm(tfm);
+	return status;
+}
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 2a59d176e69..0be0b37c84e 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -231,8 +231,8 @@ unhash_delegation(struct nfs4_delegation *dp)
 
 #define clientid_hashval(id) \
 	((id) & CLIENT_HASH_MASK)
-#define clientstr_hashval(name, namelen) \
-	(opaque_hashval((name), (namelen)) & CLIENT_HASH_MASK)
+#define clientstr_hashval(name) \
+	(opaque_hashval((name), 8) & CLIENT_HASH_MASK)
 /*
  * reclaim_str_hashtbl[] holds known client info from previous reset/reboot
  * used in reboot/reset lease grace period processing
@@ -366,11 +366,12 @@ expire_client(struct nfs4_client *clp)
 }
 
 static struct nfs4_client *
-create_client(struct xdr_netobj name) {
+create_client(struct xdr_netobj name, char *recdir) {
 	struct nfs4_client *clp;
 
 	if (!(clp = alloc_client(name)))
 		goto out;
+	memcpy(clp->cl_recdir, recdir, HEXDIR_LEN);
 	atomic_set(&clp->cl_count, 1);
 	atomic_set(&clp->cl_callback.cb_set, 0);
 	clp->cl_callback.cb_parsed = 0;
@@ -403,11 +404,9 @@ copy_cred(struct svc_cred *target, struct svc_cred *source) {
 	get_group_info(target->cr_group_info);
 }
 
-static int
-cmp_name(struct xdr_netobj *n1, struct xdr_netobj *n2) {
-	if (!n1 || !n2)
-		return 0;
-	return((n1->len == n2->len) && !memcmp(n1->data, n2->data, n2->len));
+static inline int
+same_name(const char *n1, const char *n2) {
+	return 0 == memcmp(n1, n2, HEXDIR_LEN);
 }
 
 static int
@@ -479,8 +478,7 @@ move_to_confirmed(struct nfs4_client *clp)
 	list_del_init(&clp->cl_strhash);
 	list_del_init(&clp->cl_idhash);
 	list_add(&clp->cl_idhash, &conf_id_hashtbl[idhashval]);
-	strhashval = clientstr_hashval(clp->cl_name.data, 
-			clp->cl_name.len);
+	strhashval = clientstr_hashval(clp->cl_recdir);
 	list_add(&clp->cl_strhash, &conf_str_hashtbl[strhashval]);
 	renew_client(clp);
 }
@@ -651,22 +649,27 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid)
 	unsigned int 		strhashval;
 	struct nfs4_client *	conf, * unconf, * new, * clp;
 	int 			status;
+	char                    dname[HEXDIR_LEN];
 	
 	status = nfserr_inval;
 	if (!check_name(clname))
 		goto out;
 
+	status = nfs4_make_rec_clidname(dname, &clname);
+	if (status)
+		goto out;
+
 	/* 
 	 * XXX The Duplicate Request Cache (DRC) has been checked (??)
 	 * We get here on a DRC miss.
 	 */
 
-	strhashval = clientstr_hashval(clname.data, clname.len);
+	strhashval = clientstr_hashval(dname);
 
 	conf = NULL;
 	nfs4_lock_state();
 	list_for_each_entry(clp, &conf_str_hashtbl[strhashval], cl_strhash) {
-		if (!cmp_name(&clp->cl_name, &clname))
+		if (!same_name(clp->cl_recdir, dname))
 			continue;
 		/* 
 		 * CASE 0:
@@ -686,7 +689,7 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid)
 	}
 	unconf = NULL;
 	list_for_each_entry(clp, &unconf_str_hashtbl[strhashval], cl_strhash) {
-		if (!cmp_name(&clp->cl_name, &clname))
+		if (!same_name(clp->cl_recdir, dname))
 			continue;
 		/* cl_name match from a previous SETCLIENTID operation */
 		unconf = clp;
@@ -700,7 +703,8 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid)
 		 */
 		if (unconf)
 			expire_client(unconf);
-		if (!(new = create_client(clname)))
+		new = create_client(clname, dname);
+		if (new == NULL)
 			goto out;
 		copy_verf(new, &clverifier);
 		new->cl_addr = ip_addr;
@@ -728,7 +732,8 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid)
 		     cmp_clid(&unconf->cl_clientid, &conf->cl_clientid)) {
 				expire_client(unconf);
 		}
-		if (!(new = create_client(clname)))
+		new = create_client(clname, dname);
+		if (new == NULL)
 			goto out;
 		copy_verf(new,&conf->cl_verifier);
 		new->cl_addr = ip_addr;
@@ -746,7 +751,8 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid)
 		 * using input clverifier, clname, and callback info
 		 * and generate a new cl_clientid and cl_confirm.
 		 */
-		if (!(new = create_client(clname)))
+		new = create_client(clname, dname);
+		if (new == NULL)
 			goto out;
 		copy_verf(new,&clverifier);
 		new->cl_addr = ip_addr;
@@ -772,7 +778,8 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid)
 		 * new cl_verifier and a new cl_confirm
 		 */
 		expire_client(unconf);
-		if (!(new = create_client(clname)))
+		new = create_client(clname, dname);
+		if (new == NULL)
 			goto out;
 		copy_verf(new,&clverifier);
 		new->cl_addr = ip_addr;
@@ -856,7 +863,7 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confi
 	if ((conf && unconf) && 
 	    (cmp_verf(&unconf->cl_confirm, &confirm)) &&
 	    (cmp_verf(&conf->cl_verifier, &unconf->cl_verifier)) &&
-	    (cmp_name(&conf->cl_name,&unconf->cl_name))  &&
+	    (same_name(conf->cl_recdir,unconf->cl_recdir))  &&
 	    (!cmp_verf(&conf->cl_confirm, &unconf->cl_confirm))) {
 		if (!cmp_creds(&conf->cl_cred, &unconf->cl_cred)) 
 			status = nfserr_clid_inuse;
@@ -876,7 +883,7 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confi
 	if ((conf && !unconf) || 
 	    ((conf && unconf) && 
 	     (!cmp_verf(&conf->cl_verifier, &unconf->cl_verifier) ||
-	      !cmp_name(&conf->cl_name, &unconf->cl_name)))) {
+	      !same_name(conf->cl_recdir, unconf->cl_recdir)))) {
 		if (!cmp_creds(&conf->cl_cred,&rqstp->rq_cred)) {
 			status = nfserr_clid_inuse;
 		} else {
@@ -3074,39 +3081,28 @@ out:
 }
 
 static inline struct nfs4_client_reclaim *
-alloc_reclaim(int namelen)
+alloc_reclaim(void)
 {
-	struct nfs4_client_reclaim *crp = NULL;
-
-	crp = kmalloc(sizeof(struct nfs4_client_reclaim), GFP_KERNEL);
-	if (!crp)
-		return NULL;
-	crp->cr_name.data = kmalloc(namelen, GFP_KERNEL);
-	if (!crp->cr_name.data) {
-		kfree(crp);
-		return NULL;
-	}
-	return crp;
+	return kmalloc(sizeof(struct nfs4_client_reclaim), GFP_KERNEL);
 }
 
 /*
  * failure => all reset bets are off, nfserr_no_grace...
  */
 static int
-nfs4_client_to_reclaim(char *name, int namlen)
+nfs4_client_to_reclaim(char *name)
 {
 	unsigned int strhashval;
 	struct nfs4_client_reclaim *crp = NULL;
 
-	dprintk("NFSD nfs4_client_to_reclaim NAME: %.*s\n", namlen, name);
-	crp = alloc_reclaim(namlen);
+	dprintk("NFSD nfs4_client_to_reclaim NAME: %.*s\n", HEXDIR_LEN, name);
+	crp = alloc_reclaim();
 	if (!crp)
 		return 0;
-	strhashval = clientstr_hashval(name, namlen);
+	strhashval = clientstr_hashval(name);
 	INIT_LIST_HEAD(&crp->cr_strhash);
 	list_add(&crp->cr_strhash, &reclaim_str_hashtbl[strhashval]);
-	memcpy(crp->cr_name.data, name, namlen);
-	crp->cr_name.len = namlen;
+	memcpy(crp->cr_recdir, name, HEXDIR_LEN);
 	reclaim_str_hashtbl_size++;
 	return 1;
 }
@@ -3122,7 +3118,6 @@ nfs4_release_reclaim(void)
 			crp = list_entry(reclaim_str_hashtbl[i].next,
 			                struct nfs4_client_reclaim, cr_strhash);
 			list_del(&crp->cr_strhash);
-			kfree(crp->cr_name.data);
 			kfree(crp);
 			reclaim_str_hashtbl_size--;
 		}
@@ -3145,13 +3140,14 @@ nfs4_find_reclaim_client(clientid_t *clid)
 	if (clp == NULL)
 		return NULL;
 
-	dprintk("NFSD: nfs4_find_reclaim_client for %.*s\n",
-		            clp->cl_name.len, clp->cl_name.data);
+	dprintk("NFSD: nfs4_find_reclaim_client for %.*s with recdir %s\n",
+		            clp->cl_name.len, clp->cl_name.data,
+			    clp->cl_recdir);
 
 	/* find clp->cl_name in reclaim_str_hashtbl */
-	strhashval = clientstr_hashval(clp->cl_name.data, clp->cl_name.len);
+	strhashval = clientstr_hashval(clp->cl_recdir);
 	list_for_each_entry(crp, &reclaim_str_hashtbl[strhashval], cr_strhash) {
-		if (cmp_name(&crp->cr_name, &clp->cl_name)) {
+		if (same_name(crp->cr_recdir, clp->cl_recdir)) {
 			return crp;
 		}
 	}
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index 296e6429fc3..fdaa84addad 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -109,6 +109,8 @@ struct nfs4_callback {
 	struct rpc_clnt *       cb_client;
 };
 
+#define HEXDIR_LEN     33 /* hex version of 16 byte md5 of cl_name plus '\0' */
+
 /*
  * struct nfs4_client - one per client.  Clientids live here.
  * 	o Each nfs4_client is hashed by clientid.
@@ -126,6 +128,7 @@ struct nfs4_client {
 	struct list_head	cl_del_perclnt; /* list: delegations */
 	struct list_head        cl_lru;         /* tail queue */
 	struct xdr_netobj	cl_name; 	/* id generated by client */
+	char                    cl_recdir[HEXDIR_LEN]; /* recovery dir */
 	nfs4_verifier		cl_verifier; 	/* generated by client */
 	time_t                  cl_time;        /* time of last lease renewal */
 	u32			cl_addr; 	/* client ipaddress */
@@ -143,7 +146,7 @@ struct nfs4_client {
  */
 struct nfs4_client_reclaim {
 	struct list_head	cr_strhash;	/* hash by cr_name */
-	struct xdr_netobj 	cr_name; 	/* id generated by client */
+	char			cr_recdir[HEXDIR_LEN]; /* recover dir */
 };
 
 static inline void
@@ -283,6 +286,7 @@ extern void nfs4_free_stateowner(struct kref *kref);
 extern void nfsd4_probe_callback(struct nfs4_client *clp);
 extern void nfsd4_cb_recall(struct nfs4_delegation *dp);
 extern void nfs4_put_delegation(struct nfs4_delegation *dp);
+extern int nfs4_make_rec_clidname(char *clidname, struct xdr_netobj *clname);
 
 static inline void
 nfs4_put_stateowner(struct nfs4_stateowner *so)
-- 
cgit v1.2.3-70-g09d2


From fd39ca9a808c6026989bc2188868a0574eb37108 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Thu, 23 Jun 2005 22:04:03 -0700
Subject: [PATCH] knfsd: nfsd4: make needlessly global code static

This patch contains the following possible cleanups:

- make needlessly global code static

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfsd/nfs4acl.c          |  4 ++--
 fs/nfsd/nfs4callback.c     |  7 +++---
 fs/nfsd/nfs4idmap.c        | 12 +++++-----
 fs/nfsd/nfs4state.c        | 57 ++++++++++++++++++++++++----------------------
 fs/nfsd/nfs4xdr.c          |  4 ++--
 include/linux/nfsd/state.h |  8 -------
 6 files changed, 43 insertions(+), 49 deletions(-)

(limited to 'include')

diff --git a/fs/nfsd/nfs4acl.c b/fs/nfsd/nfs4acl.c
index 11ebf6c4aa5..4a2105552ac 100644
--- a/fs/nfsd/nfs4acl.c
+++ b/fs/nfsd/nfs4acl.c
@@ -125,7 +125,7 @@ static short ace2type(struct nfs4_ace *);
 static int _posix_to_nfsv4_one(struct posix_acl *, struct nfs4_acl *, unsigned int);
 static struct posix_acl *_nfsv4_to_posix_one(struct nfs4_acl *, unsigned int);
 int nfs4_acl_add_ace(struct nfs4_acl *, u32, u32, u32, int, uid_t);
-int nfs4_acl_split(struct nfs4_acl *, struct nfs4_acl *);
+static int nfs4_acl_split(struct nfs4_acl *, struct nfs4_acl *);
 
 struct nfs4_acl *
 nfs4_acl_posix_to_nfsv4(struct posix_acl *pacl, struct posix_acl *dpacl,
@@ -775,7 +775,7 @@ out_err:
 	return pacl;
 }
 
-int
+static int
 nfs4_acl_split(struct nfs4_acl *acl, struct nfs4_acl *dacl)
 {
 	struct list_head *h, *n;
diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 38c3e1c47d8..68bb245491f 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -54,7 +54,6 @@
 
 /* declarations */
 static void nfs4_cb_null(struct rpc_task *task);
-extern spinlock_t recall_lock;
 
 /* Index of predefined Linux callback client operations */
 
@@ -329,12 +328,12 @@ out:
         .p_bufsiz = MAX(NFS4_##argtype##_sz,NFS4_##restype##_sz) << 2,  \
 }
 
-struct rpc_procinfo     nfs4_cb_procedures[] = {
+static struct rpc_procinfo     nfs4_cb_procedures[] = {
     PROC(CB_NULL,      NULL,     enc_cb_null,     dec_cb_null),
     PROC(CB_RECALL,    COMPOUND,   enc_cb_recall,      dec_cb_recall),
 };
 
-struct rpc_version              nfs_cb_version4 = {
+static struct rpc_version       nfs_cb_version4 = {
         .number                 = 1,
         .nrprocs                = sizeof(nfs4_cb_procedures)/sizeof(nfs4_cb_procedures[0]),
         .procs                  = nfs4_cb_procedures
@@ -348,7 +347,7 @@ static struct rpc_version *	nfs_cb_version[] = {
 /*
  * Use the SETCLIENTID credential
  */
-struct rpc_cred *
+static struct rpc_cred *
 nfsd4_lookupcred(struct nfs4_client *clp, int taskflags)
 {
         struct auth_cred acred;
diff --git a/fs/nfsd/nfs4idmap.c b/fs/nfsd/nfs4idmap.c
index 4ba540841cf..5605a26efc5 100644
--- a/fs/nfsd/nfs4idmap.c
+++ b/fs/nfsd/nfs4idmap.c
@@ -104,7 +104,7 @@ ent_update(struct ent *new, struct ent *itm)
 	ent_init(new, itm);
 }
 
-void
+static void
 ent_put(struct cache_head *ch, struct cache_detail *cd)
 {
 	if (cache_put(ch, cd)) {
@@ -186,7 +186,7 @@ warn_no_idmapd(struct cache_detail *detail)
 static int         idtoname_parse(struct cache_detail *, char *, int);
 static struct ent *idtoname_lookup(struct ent *, int);
 
-struct cache_detail idtoname_cache = {
+static struct cache_detail idtoname_cache = {
 	.hash_size	= ENT_HASHMAX,
 	.hash_table	= idtoname_table,
 	.name		= "nfs4.idtoname",
@@ -277,7 +277,7 @@ nametoid_hash(struct ent *ent)
 	return hash_str(ent->name, ENT_HASHBITS);
 }
 
-void
+static void
 nametoid_request(struct cache_detail *cd, struct cache_head *ch, char **bpp,
     int *blen)
 {
@@ -317,9 +317,9 @@ nametoid_show(struct seq_file *m, struct cache_detail *cd, struct cache_head *h)
 }
 
 static struct ent *nametoid_lookup(struct ent *, int);
-int                nametoid_parse(struct cache_detail *, char *, int);
+static int         nametoid_parse(struct cache_detail *, char *, int);
 
-struct cache_detail nametoid_cache = {
+static struct cache_detail nametoid_cache = {
 	.hash_size	= ENT_HASHMAX,
 	.hash_table	= nametoid_table,
 	.name		= "nfs4.nametoid",
@@ -330,7 +330,7 @@ struct cache_detail nametoid_cache = {
 	.warn_no_listener = warn_no_idmapd,
 };
 
-int
+static int
 nametoid_parse(struct cache_detail *cd, char *buf, int buflen)
 {
 	struct ent ent, *res;
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 1b2f67f5eef..8a5f777b1e9 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -54,18 +54,21 @@
 /* Globals */
 static time_t lease_time = 90;     /* default lease time */
 static time_t user_lease_time = 90;
-time_t boot_time;
+static time_t boot_time;
 static int in_grace = 1;
 static u32 current_clientid = 1;
 static u32 current_ownerid = 1;
 static u32 current_fileid = 1;
 static u32 current_delegid = 1;
 static u32 nfs4_init;
-stateid_t zerostateid;             /* bits all 0 */
-stateid_t onestateid;              /* bits all 1 */
+static stateid_t zerostateid;             /* bits all 0 */
+static stateid_t onestateid;              /* bits all 1 */
+
+#define ZERO_STATEID(stateid) (!memcmp((stateid), &zerostateid, sizeof(stateid_t)))
+#define ONE_STATEID(stateid)  (!memcmp((stateid), &onestateid, sizeof(stateid_t)))
 
 /* forward declarations */
-struct nfs4_stateid * find_stateid(stateid_t *stid, int flags);
+static struct nfs4_stateid * find_stateid(stateid_t *stid, int flags);
 static struct nfs4_delegation * find_delegation_stateid(struct inode *ino, stateid_t *stid);
 static void release_stateid_lockowners(struct nfs4_stateid *open_stp);
 
@@ -77,10 +80,10 @@ static void release_stateid_lockowners(struct nfs4_stateid *open_stp);
  */
 static DECLARE_MUTEX(client_sema);
 
-kmem_cache_t *stateowner_slab = NULL;
-kmem_cache_t *file_slab = NULL;
-kmem_cache_t *stateid_slab = NULL;
-kmem_cache_t *deleg_slab = NULL;
+static kmem_cache_t *stateowner_slab = NULL;
+static kmem_cache_t *file_slab = NULL;
+static kmem_cache_t *stateid_slab = NULL;
+static kmem_cache_t *deleg_slab = NULL;
 
 void
 nfs4_lock_state(void)
@@ -116,7 +119,7 @@ static void release_stateid(struct nfs4_stateid *stp, int flags);
  */
 
 /* recall_lock protects the del_recall_lru */
-spinlock_t recall_lock = SPIN_LOCK_UNLOCKED;
+static spinlock_t recall_lock = SPIN_LOCK_UNLOCKED;
 static struct list_head del_recall_lru;
 
 static void
@@ -456,7 +459,7 @@ check_name(struct xdr_netobj name) {
 	return 1;
 }
 
-void
+static void
 add_to_unconfirmed(struct nfs4_client *clp, unsigned int strhashval)
 {
 	unsigned int idhashval;
@@ -468,7 +471,7 @@ add_to_unconfirmed(struct nfs4_client *clp, unsigned int strhashval)
 	clp->cl_time = get_seconds();
 }
 
-void
+static void
 move_to_confirmed(struct nfs4_client *clp)
 {
 	unsigned int idhashval = clientid_hashval(clp->cl_clientid.cl_id);
@@ -567,7 +570,7 @@ parse_octet(unsigned int *lenp, char **addrp)
 }
 
 /* parse and set the setclientid ipv4 callback address */
-int
+static int
 parse_ipv4(unsigned int addr_len, char *addr_val, unsigned int *cbaddrp, unsigned short *cbportp)
 {
 	int temp = 0;
@@ -603,7 +606,7 @@ parse_ipv4(unsigned int addr_len, char *addr_val, unsigned int *cbaddrp, unsigne
 	return 1;
 }
 
-void
+static void
 gen_callback(struct nfs4_client *clp, struct nfsd4_setclientid *se)
 {
 	struct nfs4_callback *cb = &clp->cl_callback;
@@ -1186,7 +1189,7 @@ release_stateid(struct nfs4_stateid *stp, int flags)
 	stp = NULL;
 }
 
-void
+static void
 move_to_close_lru(struct nfs4_stateowner *sop)
 {
 	dprintk("NFSD: move_to_close_lru nfs4_stateowner %p\n", sop);
@@ -1196,7 +1199,7 @@ move_to_close_lru(struct nfs4_stateowner *sop)
 	sop->so_time = get_seconds();
 }
 
-void
+static void
 release_state_owner(struct nfs4_stateid *stp, int flag)
 {
 	struct nfs4_stateowner *sop = stp->st_stateowner;
@@ -1250,7 +1253,7 @@ find_file(struct inode *ino)
 #define TEST_ACCESS(x) ((x > 0 || x < 4)?1:0)
 #define TEST_DENY(x) ((x >= 0 || x < 5)?1:0)
 
-void
+static void
 set_access(unsigned int *access, unsigned long bmap) {
 	int i;
 
@@ -1261,7 +1264,7 @@ set_access(unsigned int *access, unsigned long bmap) {
 	}
 }
 
-void
+static void
 set_deny(unsigned int *deny, unsigned long bmap) {
 	int i;
 
@@ -1287,7 +1290,7 @@ test_share(struct nfs4_stateid *stp, struct nfsd4_open *open) {
  * Called to check deny when READ with all zero stateid or
  * WRITE with all zero or all one stateid
  */
-int
+static int
 nfs4_share_conflict(struct svc_fh *current_fh, unsigned int deny_type)
 {
 	struct inode *ino = current_fh->fh_dentry->d_inode;
@@ -1442,7 +1445,7 @@ int nfsd_change_deleg_cb(struct file_lock **onlist, int arg)
 		return -EAGAIN;
 }
 
-struct lock_manager_operations nfsd_lease_mng_ops = {
+static struct lock_manager_operations nfsd_lease_mng_ops = {
 	.fl_break = nfsd_break_deleg_cb,
 	.fl_release_private = nfsd_release_deleg_cb,
 	.fl_copy_lock = nfsd_copy_lock_deleg_cb,
@@ -1915,7 +1918,7 @@ end_grace(void)
 	in_grace = 0;
 }
 
-time_t
+static time_t
 nfs4_laundromat(void)
 {
 	struct nfs4_client *clp;
@@ -1996,7 +1999,7 @@ laundromat_main(void *not_used)
 /* search ownerid_hashtbl[] and close_lru for stateid owner
  * (stateid->si_stateownerid)
  */
-struct nfs4_stateowner *
+static struct nfs4_stateowner *
 find_openstateowner_id(u32 st_id, int flags) {
 	struct nfs4_stateowner *local = NULL;
 
@@ -2170,7 +2173,7 @@ out:
 /* 
  * Checks for sequence id mutating operations. 
  */
-int
+static int
 nfs4_preprocess_seqid_op(struct svc_fh *current_fh, u32 seqid, stateid_t *stateid, int flags, struct nfs4_stateowner **sopp, struct nfs4_stateid **stpp, clientid_t *lockclid)
 {
 	int status;
@@ -2486,7 +2489,7 @@ static struct list_head lock_ownerid_hashtbl[LOCK_HASH_SIZE];
 static struct list_head	lock_ownerstr_hashtbl[LOCK_HASH_SIZE];
 static struct list_head lockstateid_hashtbl[STATEID_HASH_SIZE];
 
-struct nfs4_stateid *
+static struct nfs4_stateid *
 find_stateid(stateid_t *stid, int flags)
 {
 	struct nfs4_stateid *local = NULL;
@@ -2550,7 +2553,7 @@ nfs4_transform_lock_offset(struct file_lock *lock)
 		lock->fl_end = OFFSET_MAX;
 }
 
-int
+static int
 nfs4_verify_lock_stateowner(struct nfs4_stateowner *sop, unsigned int hashval)
 {
 	struct nfs4_stateowner *local = NULL;
@@ -2660,7 +2663,7 @@ alloc_init_lock_stateowner(unsigned int strhashval, struct nfs4_client *clp, str
 	return sop;
 }
 
-struct nfs4_stateid *
+static struct nfs4_stateid *
 alloc_init_lock_stateid(struct nfs4_stateowner *sop, struct nfs4_file *fp, struct nfs4_stateid *open_stp)
 {
 	struct nfs4_stateid *stp;
@@ -2691,7 +2694,7 @@ out:
 	return stp;
 }
 
-int
+static int
 check_lock_length(u64 offset, u64 length)
 {
 	return ((length == 0)  || ((length != ~(u64)0) &&
@@ -3149,7 +3152,7 @@ nfs4_release_reclaim(void)
 
 /*
  * called from OPEN, CLAIM_PREVIOUS with a new clientid. */
-struct nfs4_client_reclaim *
+static struct nfs4_client_reclaim *
 nfs4_find_reclaim_client(clientid_t *clid)
 {
 	unsigned int strhashval;
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index cfe978a72ce..91fb171d2ac 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -136,7 +136,7 @@ xdr_error:					\
 	}					\
 } while (0)
 
-u32 *read_buf(struct nfsd4_compoundargs *argp, int nbytes)
+static u32 *read_buf(struct nfsd4_compoundargs *argp, int nbytes)
 {
 	/* We want more bytes than seem to be available.
 	 * Maybe we need a new page, maybe we have just run out
@@ -190,7 +190,7 @@ defer_free(struct nfsd4_compoundargs *argp,
 	return 0;
 }
 
-char *savemem(struct nfsd4_compoundargs *argp, u32 *p, int nbytes)
+static char *savemem(struct nfsd4_compoundargs *argp, u32 *p, int nbytes)
 {
 	void *new = NULL;
 	if (p == argp->tmp) {
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index fdaa84addad..0e18ae22127 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -61,11 +61,6 @@ typedef struct {
 #define si_stateownerid   si_opaque.so_stateownerid
 #define si_fileid         si_opaque.so_fileid
 
-extern stateid_t zerostateid;
-extern stateid_t onestateid;
-
-#define ZERO_STATEID(stateid)       (!memcmp((stateid), &zerostateid, sizeof(stateid_t)))
-#define ONE_STATEID(stateid)        (!memcmp((stateid), &onestateid, sizeof(stateid_t)))
 
 struct nfs4_cb_recall {
 	u32			cbr_ident;
@@ -271,12 +266,9 @@ struct nfs4_stateid {
 	((err) != nfserr_stale_stateid) &&      \
 	((err) != nfserr_bad_stateid))
 
-extern time_t nfs4_laundromat(void);
 extern int nfsd4_renew(clientid_t *clid);
 extern int nfs4_preprocess_stateid_op(struct svc_fh *current_fh, 
 		stateid_t *stateid, int flags, struct file **filp);
-extern int nfs4_share_conflict(struct svc_fh *current_fh, 
-		unsigned int deny_type);
 extern void nfs4_lock_state(void);
 extern void nfs4_unlock_state(void);
 extern int nfs4_in_grace(void);
-- 
cgit v1.2.3-70-g09d2


From ea1da636e956ad1591a74904f23d98bbc26a644b Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Thu, 23 Jun 2005 22:04:17 -0700
Subject: [PATCH] knfsd: nfsd4: rename state list fields

Trivial renaming patch:

I can never remember, while looking at various lists relating the nfsd4 state
structures, which are the "heads" and which are items on other lists, or which
structures are actually on the various lists.  The following convention helps
me: given structures foo and bar, with foo containing the head of a list of
bars, use "bars" for the name of the head of the list contained in the struct
foo, and use "per_foo" for the entries in the struct bars.

Already done for struct nfs4_file; go ahead and do it for the other nfsd4
state structures.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfsd/nfs4state.c        | 78 +++++++++++++++++++++++-----------------------
 include/linux/nfsd/state.h | 18 +++++------
 2 files changed, 48 insertions(+), 48 deletions(-)

(limited to 'include')

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 88411519811..22e76e3f06a 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -154,8 +154,8 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_f
 	dp = kmem_cache_alloc(deleg_slab, GFP_KERNEL);
 	if (dp == NULL)
 		return dp;
-	INIT_LIST_HEAD(&dp->dl_del_perfile);
-	INIT_LIST_HEAD(&dp->dl_del_perclnt);
+	INIT_LIST_HEAD(&dp->dl_perfile);
+	INIT_LIST_HEAD(&dp->dl_perclnt);
 	INIT_LIST_HEAD(&dp->dl_recall_lru);
 	dp->dl_client = clp;
 	get_nfs4_file(fp);
@@ -176,8 +176,8 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_f
 		        current_fh->fh_handle.fh_size);
 	dp->dl_time = 0;
 	atomic_set(&dp->dl_count, 1);
-	list_add(&dp->dl_del_perfile, &fp->fi_delegations);
-	list_add(&dp->dl_del_perclnt, &clp->cl_del_perclnt);
+	list_add(&dp->dl_perfile, &fp->fi_delegations);
+	list_add(&dp->dl_perclnt, &clp->cl_delegations);
 	return dp;
 }
 
@@ -214,8 +214,8 @@ nfs4_close_delegation(struct nfs4_delegation *dp)
 static void
 unhash_delegation(struct nfs4_delegation *dp)
 {
-	list_del_init(&dp->dl_del_perfile);
-	list_del_init(&dp->dl_del_perclnt);
+	list_del_init(&dp->dl_perfile);
+	list_del_init(&dp->dl_perclnt);
 	spin_lock(&recall_lock);
 	list_del_init(&dp->dl_recall_lru);
 	spin_unlock(&recall_lock);
@@ -345,11 +345,11 @@ expire_client(struct nfs4_client *clp)
 
 	INIT_LIST_HEAD(&reaplist);
 	spin_lock(&recall_lock);
-	while (!list_empty(&clp->cl_del_perclnt)) {
-		dp = list_entry(clp->cl_del_perclnt.next, struct nfs4_delegation, dl_del_perclnt);
+	while (!list_empty(&clp->cl_delegations)) {
+		dp = list_entry(clp->cl_delegations.next, struct nfs4_delegation, dl_perclnt);
 		dprintk("NFSD: expire client. dp %p, fp %p\n", dp,
 				dp->dl_flock);
-		list_del_init(&dp->dl_del_perclnt);
+		list_del_init(&dp->dl_perclnt);
 		list_move(&dp->dl_recall_lru, &reaplist);
 	}
 	spin_unlock(&recall_lock);
@@ -361,8 +361,8 @@ expire_client(struct nfs4_client *clp)
 	list_del(&clp->cl_idhash);
 	list_del(&clp->cl_strhash);
 	list_del(&clp->cl_lru);
-	while (!list_empty(&clp->cl_perclient)) {
-		sop = list_entry(clp->cl_perclient.next, struct nfs4_stateowner, so_perclient);
+	while (!list_empty(&clp->cl_openowners)) {
+		sop = list_entry(clp->cl_openowners.next, struct nfs4_stateowner, so_perclient);
 		release_stateowner(sop);
 	}
 	put_nfs4_client(clp);
@@ -380,8 +380,8 @@ create_client(struct xdr_netobj name, char *recdir) {
 	clp->cl_callback.cb_parsed = 0;
 	INIT_LIST_HEAD(&clp->cl_idhash);
 	INIT_LIST_HEAD(&clp->cl_strhash);
-	INIT_LIST_HEAD(&clp->cl_perclient);
-	INIT_LIST_HEAD(&clp->cl_del_perclnt);
+	INIT_LIST_HEAD(&clp->cl_openowners);
+	INIT_LIST_HEAD(&clp->cl_delegations);
 	INIT_LIST_HEAD(&clp->cl_lru);
 out:
 	return clp;
@@ -1074,13 +1074,13 @@ alloc_init_open_stateowner(unsigned int strhashval, struct nfs4_client *clp, str
 	INIT_LIST_HEAD(&sop->so_idhash);
 	INIT_LIST_HEAD(&sop->so_strhash);
 	INIT_LIST_HEAD(&sop->so_perclient);
-	INIT_LIST_HEAD(&sop->so_perfilestate);
-	INIT_LIST_HEAD(&sop->so_perlockowner);  /* not used */
+	INIT_LIST_HEAD(&sop->so_stateids);
+	INIT_LIST_HEAD(&sop->so_perstateid);  /* not used */
 	INIT_LIST_HEAD(&sop->so_close_lru);
 	sop->so_time = 0;
 	list_add(&sop->so_idhash, &ownerid_hashtbl[idhashval]);
 	list_add(&sop->so_strhash, &ownerstr_hashtbl[strhashval]);
-	list_add(&sop->so_perclient, &clp->cl_perclient);
+	list_add(&sop->so_perclient, &clp->cl_openowners);
 	sop->so_is_open_owner = 1;
 	sop->so_id = current_ownerid++;
 	sop->so_client = clp;
@@ -1098,10 +1098,10 @@ release_stateid_lockowners(struct nfs4_stateid *open_stp)
 {
 	struct nfs4_stateowner *lock_sop;
 
-	while (!list_empty(&open_stp->st_perlockowner)) {
-		lock_sop = list_entry(open_stp->st_perlockowner.next,
-				struct nfs4_stateowner, so_perlockowner);
-		/* list_del(&open_stp->st_perlockowner);  */
+	while (!list_empty(&open_stp->st_lockowners)) {
+		lock_sop = list_entry(open_stp->st_lockowners.next,
+				struct nfs4_stateowner, so_perstateid);
+		/* list_del(&open_stp->st_lockowners);  */
 		BUG_ON(lock_sop->so_is_open_owner);
 		release_stateowner(lock_sop);
 	}
@@ -1116,10 +1116,10 @@ unhash_stateowner(struct nfs4_stateowner *sop)
 	list_del(&sop->so_strhash);
 	if (sop->so_is_open_owner)
 		list_del(&sop->so_perclient);
-	list_del(&sop->so_perlockowner);
-	while (!list_empty(&sop->so_perfilestate)) {
-		stp = list_entry(sop->so_perfilestate.next, 
-			struct nfs4_stateid, st_perfilestate);
+	list_del(&sop->so_perstateid);
+	while (!list_empty(&sop->so_stateids)) {
+		stp = list_entry(sop->so_stateids.next,
+			struct nfs4_stateid, st_perstateowner);
 		if (sop->so_is_open_owner)
 			release_stateid(stp, OPEN_STATE);
 		else
@@ -1141,11 +1141,11 @@ init_stateid(struct nfs4_stateid *stp, struct nfs4_file *fp, struct nfsd4_open *
 	unsigned int hashval = stateid_hashval(sop->so_id, fp->fi_id);
 
 	INIT_LIST_HEAD(&stp->st_hash);
-	INIT_LIST_HEAD(&stp->st_perfilestate);
-	INIT_LIST_HEAD(&stp->st_perlockowner);
+	INIT_LIST_HEAD(&stp->st_perstateowner);
+	INIT_LIST_HEAD(&stp->st_lockowners);
 	INIT_LIST_HEAD(&stp->st_perfile);
 	list_add(&stp->st_hash, &stateid_hashtbl[hashval]);
-	list_add(&stp->st_perfilestate, &sop->so_perfilestate);
+	list_add(&stp->st_perstateowner, &sop->so_stateids);
 	list_add(&stp->st_perfile, &fp->fi_stateids);
 	stp->st_stateowner = sop;
 	get_nfs4_file(fp);
@@ -1167,7 +1167,7 @@ release_stateid(struct nfs4_stateid *stp, int flags)
 
 	list_del(&stp->st_hash);
 	list_del(&stp->st_perfile);
-	list_del(&stp->st_perfilestate);
+	list_del(&stp->st_perstateowner);
 	if (flags & OPEN_STATE) {
 		release_stateid_lockowners(stp);
 		stp->st_vfs_file = NULL;
@@ -1201,7 +1201,7 @@ release_state_owner(struct nfs4_stateid *stp, int flag)
 	 * released by the laundromat service after the lease period
 	 * to enable us to handle CLOSE replay
 	 */
-	if (sop->so_confirmed && list_empty(&sop->so_perfilestate))
+	if (sop->so_confirmed && list_empty(&sop->so_stateids))
 		move_to_close_lru(sop);
 }
 
@@ -1548,7 +1548,7 @@ find_delegation_file(struct nfs4_file *fp, stateid_t *stid)
 {
 	struct nfs4_delegation *dp;
 
-	list_for_each_entry(dp, &fp->fi_delegations, dl_del_perfile) {
+	list_for_each_entry(dp, &fp->fi_delegations, dl_perfile) {
 		if (dp->dl_stateid.si_stateownerid == stid->si_stateownerid)
 			return dp;
 	}
@@ -1892,7 +1892,7 @@ nfsd4_renew(clientid_t *clid)
 	}
 	renew_client(clp);
 	status = nfserr_cb_path_down;
-	if (!list_empty(&clp->cl_del_perclnt)
+	if (!list_empty(&clp->cl_delegations)
 			&& !atomic_read(&clp->cl_callback.cb_set))
 		goto out;
 	status = nfs_ok;
@@ -2634,13 +2634,13 @@ alloc_init_lock_stateowner(unsigned int strhashval, struct nfs4_client *clp, str
 	INIT_LIST_HEAD(&sop->so_idhash);
 	INIT_LIST_HEAD(&sop->so_strhash);
 	INIT_LIST_HEAD(&sop->so_perclient);
-	INIT_LIST_HEAD(&sop->so_perfilestate);
-	INIT_LIST_HEAD(&sop->so_perlockowner);
+	INIT_LIST_HEAD(&sop->so_stateids);
+	INIT_LIST_HEAD(&sop->so_perstateid);
 	INIT_LIST_HEAD(&sop->so_close_lru); /* not used */
 	sop->so_time = 0;
 	list_add(&sop->so_idhash, &lock_ownerid_hashtbl[idhashval]);
 	list_add(&sop->so_strhash, &lock_ownerstr_hashtbl[strhashval]);
-	list_add(&sop->so_perlockowner, &open_stp->st_perlockowner);
+	list_add(&sop->so_perstateid, &open_stp->st_lockowners);
 	sop->so_is_open_owner = 0;
 	sop->so_id = current_ownerid++;
 	sop->so_client = clp;
@@ -2664,11 +2664,11 @@ alloc_init_lock_stateid(struct nfs4_stateowner *sop, struct nfs4_file *fp, struc
 		goto out;
 	INIT_LIST_HEAD(&stp->st_hash);
 	INIT_LIST_HEAD(&stp->st_perfile);
-	INIT_LIST_HEAD(&stp->st_perfilestate);
-	INIT_LIST_HEAD(&stp->st_perlockowner); /* not used */
+	INIT_LIST_HEAD(&stp->st_perstateowner);
+	INIT_LIST_HEAD(&stp->st_lockowners); /* not used */
 	list_add(&stp->st_hash, &lockstateid_hashtbl[hashval]);
 	list_add(&stp->st_perfile, &fp->fi_stateids);
-	list_add(&stp->st_perfilestate, &sop->so_perfilestate);
+	list_add(&stp->st_perstateowner, &sop->so_stateids);
 	stp->st_stateowner = sop;
 	get_nfs4_file(fp);
 	stp->st_file = fp;
@@ -3081,8 +3081,8 @@ nfsd4_release_lockowner(struct svc_rqst *rqstp, struct nfsd4_release_lockowner *
 		/* check for any locks held by any stateid
 		 * associated with the (lock) stateowner */
 		status = nfserr_locks_held;
-		list_for_each_entry(stp, &local->so_perfilestate,
-				st_perfilestate) {
+		list_for_each_entry(stp, &local->so_stateids,
+				st_perstateowner) {
 			if (check_for_locks(stp->st_vfs_file, local))
 				goto out;
 		}
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index 0e18ae22127..f4f27b76ee6 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -72,8 +72,8 @@ struct nfs4_cb_recall {
 };
 
 struct nfs4_delegation {
-	struct list_head	dl_del_perfile; /* nfs4_file->fi_del_perfile */
-	struct list_head	dl_del_perclnt; /* nfs4_client->cl_del_perclnt*/
+	struct list_head	dl_perfile;
+	struct list_head	dl_perclnt;
 	struct list_head	dl_recall_lru;  /* delegation recalled */
 	atomic_t		dl_count;       /* ref count */
 	struct nfs4_client	*dl_client;
@@ -119,8 +119,8 @@ struct nfs4_callback {
 struct nfs4_client {
 	struct list_head	cl_idhash; 	/* hash by cl_clientid.id */
 	struct list_head	cl_strhash; 	/* hash by cl_name */
-	struct list_head	cl_perclient; 	/* list: stateowners */
-	struct list_head	cl_del_perclnt; /* list: delegations */
+	struct list_head	cl_openowners;
+	struct list_head	cl_delegations;
 	struct list_head        cl_lru;         /* tail queue */
 	struct xdr_netobj	cl_name; 	/* id generated by client */
 	char                    cl_recdir[HEXDIR_LEN]; /* recovery dir */
@@ -195,9 +195,9 @@ struct nfs4_stateowner {
 	struct kref		so_ref;
 	struct list_head        so_idhash;   /* hash by so_id */
 	struct list_head        so_strhash;   /* hash by op_name */
-	struct list_head        so_perclient; /* nfs4_client->cl_perclient */
-	struct list_head        so_perfilestate; /* list: nfs4_stateid */
-	struct list_head        so_perlockowner; /* nfs4_stateid->st_perlockowner */
+	struct list_head        so_perclient;
+	struct list_head        so_stateids;
+	struct list_head        so_perstateid; /* for lockowners only */
 	struct list_head	so_close_lru; /* tail queue */
 	time_t			so_time; /* time of placement on so_close_lru */
 	int			so_is_open_owner; /* 1=openowner,0=lockowner */
@@ -240,8 +240,8 @@ struct nfs4_file {
 struct nfs4_stateid {
 	struct list_head              st_hash; 
 	struct list_head              st_perfile;
-	struct list_head              st_perfilestate; 
-	struct list_head              st_perlockowner;
+	struct list_head              st_perstateowner;
+	struct list_head              st_lockowners;
 	struct nfs4_stateowner      * st_stateowner;
 	struct nfs4_file            * st_file;
 	stateid_t                     st_stateid;
-- 
cgit v1.2.3-70-g09d2


From cb36d6345752fa24827044c68e15f6708a40d9f6 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Thu, 23 Jun 2005 22:04:23 -0700
Subject: [PATCH] knfsd: nfsd4: remove cb_parsed

The cb_parsed field is only used by probe_callback, to determine whether the
callback information has been filled in by setclientid.  But there is no way
that probe_callback() can be called without that having already happened, so
that check is superfluous, as is cb_parsed.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfsd/nfs4callback.c     | 4 +---
 fs/nfsd/nfs4state.c        | 5 +----
 include/linux/nfsd/state.h | 1 -
 3 files changed, 2 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 68bb245491f..583c0710e45 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -386,9 +386,7 @@ nfsd4_probe_callback(struct nfs4_client *clp)
 	char                    hostname[32];
 	int status;
 
-	dprintk("NFSD: probe_callback. cb_parsed %d cb_set %d\n",
-			cb->cb_parsed, atomic_read(&cb->cb_set));
-	if (!cb->cb_parsed || atomic_read(&cb->cb_set))
+	if (atomic_read(&cb->cb_set))
 		return;
 
 	/* Initialize address */
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 26d00465c28..0b47a97e953 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -377,7 +377,6 @@ create_client(struct xdr_netobj name, char *recdir) {
 	memcpy(clp->cl_recdir, recdir, HEXDIR_LEN);
 	atomic_set(&clp->cl_count, 1);
 	atomic_set(&clp->cl_callback.cb_set, 0);
-	clp->cl_callback.cb_parsed = 0;
 	INIT_LIST_HEAD(&clp->cl_idhash);
 	INIT_LIST_HEAD(&clp->cl_strhash);
 	INIT_LIST_HEAD(&clp->cl_openowners);
@@ -620,14 +619,12 @@ gen_callback(struct nfs4_client *clp, struct nfsd4_setclientid *se)
 		goto out_err;
 	cb->cb_prog = se->se_callback_prog;
 	cb->cb_ident = se->se_callback_ident;
-	cb->cb_parsed = 1;
 	return;
 out_err:
 	printk(KERN_INFO "NFSD: this client (clientid %08x/%08x) "
 		"will not receive delegations\n",
 		clp->cl_clientid.cl_boot, clp->cl_clientid.cl_id);
 
-	cb->cb_parsed = 0;
 	return;
 }
 
@@ -872,7 +869,7 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confi
 		else {
 			/* XXX: We just turn off callbacks until we can handle
 			  * change request correctly. */
-			conf->cl_callback.cb_parsed = 0;
+			atomic_set(&conf->cl_callback.cb_set, 0);
 			gen_confirm(conf);
 			expire_client(unconf);
 			status = nfs_ok;
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index f4f27b76ee6..83d29ec03a5 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -92,7 +92,6 @@ struct nfs4_delegation {
 /* client delegation callback info */
 struct nfs4_callback {
 	/* SETCLIENTID info */
-	u32			cb_parsed;  /* addr parsed */
 	u32                     cb_addr;
 	unsigned short          cb_port;
 	u32                     cb_prog;
-- 
cgit v1.2.3-70-g09d2


From 190e4fbf96037e5e526ba3210f2bcc2a3b6fe964 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Thu, 23 Jun 2005 22:04:25 -0700
Subject: [PATCH] knfsd: nfsd4: initialize recovery directory

NFSv4 clients are required to know what state they have on the server so that
they can reclaim it on server reboot.  However, it is possible for
pathalogical combinations of server reboots and network partitions to leave a
client in a state where it cannot know whether it has lost its state on the
server.

For this reason, rfc3530 requires that we store some information about clients
to stable storage.

So we maintain a directory /var/lib/nfs/v4recovery with a subdirectory for
each client with active state.  We leave open the possibility of including
files underneath each such subdirectory with information about the client, but
for now the subdirectories are empty.

We create a client subdirectory whenever a client makes its first non-reclaim
open_confirm.

We remove a client subdirectory whenever either
        a) its lease expires, or
	b) the grace period ends without it reclaiming anything.
When handling reclaims, we allow the reclaim if and only if the client doing
the reclaim has a subdirectory.

This patch adds just the code to scan the recovery directory on nfsd startup.

Signed-off-by: Andy Adamson <andros@citi.umich.edu>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfsd/nfs4recover.c      | 166 +++++++++++++++++++++++++++++++++++++++++++++
 fs/nfsd/nfs4state.c        |  18 ++++-
 include/linux/nfsd/state.h |   4 ++
 3 files changed, 186 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c
index 841a305d794..2dc9851a1d3 100644
--- a/fs/nfsd/nfs4recover.c
+++ b/fs/nfsd/nfs4recover.c
@@ -39,6 +39,9 @@
 #include <linux/nfs4.h>
 #include <linux/nfsd/state.h>
 #include <linux/nfsd/xdr4.h>
+#include <linux/param.h>
+#include <linux/file.h>
+#include <linux/namei.h>
 #include <asm/uaccess.h>
 #include <asm/scatterlist.h>
 #include <linux/crypto.h>
@@ -46,6 +49,27 @@
 
 #define NFSDDBG_FACILITY                NFSDDBG_PROC
 
+/* Globals */
+char recovery_dirname[PATH_MAX] = "/var/lib/nfs/v4recovery";
+static struct nameidata rec_dir;
+static int rec_dir_init = 0;
+
+static void
+nfs4_save_user(uid_t *saveuid, gid_t *savegid)
+{
+	*saveuid = current->fsuid;
+	*savegid = current->fsgid;
+	current->fsuid = 0;
+	current->fsgid = 0;
+}
+
+static void
+nfs4_reset_user(uid_t saveuid, gid_t savegid)
+{
+	current->fsuid = saveuid;
+	current->fsgid = savegid;
+}
+
 static void
 md5_to_hex(char *out, char *md5)
 {
@@ -95,3 +119,145 @@ out:
 		crypto_free_tfm(tfm);
 	return status;
 }
+
+typedef int (recdir_func)(struct dentry *, struct dentry *);
+
+struct dentry_list {
+	struct dentry *dentry;
+	struct list_head list;
+};
+
+struct dentry_list_arg {
+	struct list_head dentries;
+	struct dentry *parent;
+};
+
+static int
+nfsd4_build_dentrylist(void *arg, const char *name, int namlen,
+		loff_t offset, ino_t ino, unsigned int d_type)
+{
+	struct dentry_list_arg *dla = arg;
+	struct list_head *dentries = &dla->dentries;
+	struct dentry *parent = dla->parent;
+	struct dentry *dentry;
+	struct dentry_list *child;
+
+	if (name && isdotent(name, namlen))
+		return nfs_ok;
+	dentry = lookup_one_len(name, parent, namlen);
+	if (IS_ERR(dentry))
+		return PTR_ERR(dentry);
+	child = kmalloc(sizeof(*child), GFP_KERNEL);
+	if (child == NULL)
+		return -ENOMEM;
+	child->dentry = dentry;
+	list_add(&child->list, dentries);
+	return 0;
+}
+
+static int
+nfsd4_list_rec_dir(struct dentry *dir, recdir_func *f)
+{
+	struct file *filp;
+	struct dentry_list_arg dla = {
+		.parent = dir,
+	};
+	struct list_head *dentries = &dla.dentries;
+	struct dentry_list *child;
+	uid_t uid;
+	gid_t gid;
+	int status;
+
+	if (!rec_dir_init)
+		return 0;
+
+	nfs4_save_user(&uid, &gid);
+
+	filp = dentry_open(dget(dir), mntget(rec_dir.mnt),
+			O_RDWR);
+	status = PTR_ERR(filp);
+	if (IS_ERR(filp))
+		goto out;
+	INIT_LIST_HEAD(dentries);
+	status = vfs_readdir(filp, nfsd4_build_dentrylist, &dla);
+	fput(filp);
+	while (!list_empty(dentries)) {
+		child = list_entry(dentries->next, struct dentry_list, list);
+		status = f(dir, child->dentry);
+		if (status)
+			goto out;
+		list_del(&child->list);
+		dput(child->dentry);
+		kfree(child);
+	}
+out:
+	while (!list_empty(dentries)) {
+		child = list_entry(dentries->next, struct dentry_list, list);
+		list_del(&child->list);
+		dput(child->dentry);
+		kfree(child);
+	}
+	nfs4_reset_user(uid, gid);
+	return status;
+}
+
+static int
+load_recdir(struct dentry *parent, struct dentry *child)
+{
+	if (child->d_name.len != HEXDIR_LEN - 1) {
+		printk("nfsd4: illegal name %s in recovery directory\n",
+				child->d_name.name);
+		/* Keep trying; maybe the others are OK: */
+		return nfs_ok;
+	}
+	nfs4_client_to_reclaim(child->d_name.name);
+	return nfs_ok;
+}
+
+int
+nfsd4_recdir_load(void) {
+	int status;
+
+	status = nfsd4_list_rec_dir(rec_dir.dentry, load_recdir);
+	if (status)
+		printk("nfsd4: failed loading clients from recovery"
+			" directory %s\n", rec_dir.dentry->d_name.name);
+	return status;
+}
+
+/*
+ * Hold reference to the recovery directory.
+ */
+
+void
+nfsd4_init_recdir(char *rec_dirname)
+{
+	uid_t			uid = 0;
+	gid_t			gid = 0;
+	int 			status;
+
+	printk("NFSD: Using %s as the NFSv4 state recovery directory\n",
+			rec_dirname);
+
+	BUG_ON(rec_dir_init);
+
+	nfs4_save_user(&uid, &gid);
+
+	status = path_lookup(rec_dirname, LOOKUP_FOLLOW, &rec_dir);
+	if (status == -ENOENT)
+		printk("NFSD: recovery directory %s doesn't exist\n",
+				rec_dirname);
+
+	if (!status)
+		rec_dir_init = 1;
+	nfs4_reset_user(uid, gid);
+}
+
+void
+nfsd4_shutdown_recdir(void)
+{
+	if (!rec_dir_init)
+		return;
+	rec_dir_init = 0;
+	path_release(&rec_dir);
+}
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 0b47a97e953..6b9d23c39af 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -71,6 +71,7 @@ static stateid_t onestateid;              /* bits all 1 */
 static struct nfs4_stateid * find_stateid(stateid_t *stid, int flags);
 static struct nfs4_delegation * find_delegation_stateid(struct inode *ino, stateid_t *stid);
 static void release_stateid_lockowners(struct nfs4_stateid *open_stp);
+extern char recovery_dirname[];
 
 /* Locking:
  *
@@ -3091,8 +3092,8 @@ alloc_reclaim(void)
 /*
  * failure => all reset bets are off, nfserr_no_grace...
  */
-static int
-nfs4_client_to_reclaim(char *name)
+int
+nfs4_client_to_reclaim(const char *name)
 {
 	unsigned int strhashval;
 	struct nfs4_client_reclaim *crp = NULL;
@@ -3202,6 +3203,17 @@ nfs4_state_init(void)
 	reclaim_str_hashtbl_size = 0;
 }
 
+static void
+nfsd4_load_reboot_recovery_data(void)
+{
+	int status;
+
+	nfsd4_init_recdir(recovery_dirname);
+	status = nfsd4_recdir_load();
+	if (status)
+		printk("NFSD: Failure reading reboot recovery data\n");
+}
+
 /* initialization to perform when the nfsd service is started: */
 
 static void
@@ -3228,6 +3240,7 @@ nfs4_state_start(void)
 	status = nfsd4_init_slabs();
 	if (status)
 		return status;
+	nfsd4_load_reboot_recovery_data();
 	__nfs4_state_start();
 	nfs4_init = 1;
 	return 0;
@@ -3286,6 +3299,7 @@ __nfs4_state_shutdown(void)
 	cancel_delayed_work(&laundromat_work);
 	flush_workqueue(laundry_wq);
 	destroy_workqueue(laundry_wq);
+	nfsd4_shutdown_recdir();
 	nfs4_init = 0;
 }
 
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index 83d29ec03a5..19481ab122d 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -278,6 +278,10 @@ extern void nfsd4_probe_callback(struct nfs4_client *clp);
 extern void nfsd4_cb_recall(struct nfs4_delegation *dp);
 extern void nfs4_put_delegation(struct nfs4_delegation *dp);
 extern int nfs4_make_rec_clidname(char *clidname, struct xdr_netobj *clname);
+extern void nfsd4_init_recdir(char *recdir_name);
+extern int nfsd4_recdir_load(void);
+extern void nfsd4_shutdown_recdir(void);
+extern int nfs4_client_to_reclaim(const char *name);
 
 static inline void
 nfs4_put_stateowner(struct nfs4_stateowner *so)
-- 
cgit v1.2.3-70-g09d2


From c7b9a45927e74c81d6562153f7fde9d32da00159 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Thu, 23 Jun 2005 22:04:30 -0700
Subject: [PATCH] knfsd: nfsd4: reboot recovery

This patch adds the code to create and remove client subdirectories from the
recovery directory, as described in the previous patch comment.

Signed-off-by: Andy Adamson <andros@citi.umich.edu>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfsd/nfs4recover.c      | 169 +++++++++++++++++++++++++++++++++++++++++++++
 fs/nfsd/nfs4state.c        |  16 +++++
 include/linux/nfsd/state.h |   5 ++
 3 files changed, 190 insertions(+)

(limited to 'include')

diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c
index 2dc9851a1d3..2805c5245ea 100644
--- a/fs/nfsd/nfs4recover.c
+++ b/fs/nfsd/nfs4recover.c
@@ -120,6 +120,70 @@ out:
 	return status;
 }
 
+static int
+nfsd4_rec_fsync(struct dentry *dentry)
+{
+	struct file *filp;
+	int status = nfs_ok;
+
+	dprintk("NFSD: nfs4_fsync_rec_dir\n");
+	filp = dentry_open(dget(dentry), mntget(rec_dir.mnt), O_RDWR);
+	if (IS_ERR(filp)) {
+		status = PTR_ERR(filp);
+		goto out;
+	}
+	if (filp->f_op && filp->f_op->fsync)
+		status = filp->f_op->fsync(filp, filp->f_dentry, 0);
+	fput(filp);
+out:
+	if (status)
+		printk("nfsd4: unable to sync recovery directory\n");
+	return status;
+}
+
+int
+nfsd4_create_clid_dir(struct nfs4_client *clp)
+{
+	char *dname = clp->cl_recdir;
+	struct dentry *dentry;
+	uid_t uid;
+	gid_t gid;
+	int status;
+
+	dprintk("NFSD: nfsd4_create_clid_dir for \"%s\"\n", dname);
+
+	if (!rec_dir_init || clp->cl_firststate)
+		return 0;
+
+	nfs4_save_user(&uid, &gid);
+
+	/* lock the parent */
+	down(&rec_dir.dentry->d_inode->i_sem);
+
+	dentry = lookup_one_len(dname, rec_dir.dentry, HEXDIR_LEN-1);
+	if (IS_ERR(dentry)) {
+		status = PTR_ERR(dentry);
+		goto out_unlock;
+	}
+	status = -EEXIST;
+	if (dentry->d_inode) {
+		dprintk("NFSD: nfsd4_create_clid_dir: DIRECTORY EXISTS\n");
+		goto out_put;
+	}
+	status = vfs_mkdir(rec_dir.dentry->d_inode, dentry, S_IRWXU);
+out_put:
+	dput(dentry);
+out_unlock:
+	up(&rec_dir.dentry->d_inode->i_sem);
+	if (status == 0) {
+		clp->cl_firststate = 1;
+		status = nfsd4_rec_fsync(rec_dir.dentry);
+	}
+	nfs4_reset_user(uid, gid);
+	dprintk("NFSD: nfsd4_create_clid_dir returns %d\n", status);
+	return status;
+}
+
 typedef int (recdir_func)(struct dentry *, struct dentry *);
 
 struct dentry_list {
@@ -201,6 +265,111 @@ out:
 	return status;
 }
 
+static int
+nfsd4_remove_clid_file(struct dentry *dir, struct dentry *dentry)
+{
+	int status;
+
+	if (!S_ISREG(dir->d_inode->i_mode)) {
+		printk("nfsd4: non-file found in client recovery directory\n");
+		return -EINVAL;
+	}
+	down(&dir->d_inode->i_sem);
+	status = vfs_unlink(dir->d_inode, dentry);
+	up(&dir->d_inode->i_sem);
+	return status;
+}
+
+static int
+nfsd4_clear_clid_dir(struct dentry *dir, struct dentry *dentry)
+{
+	int status;
+
+	/* For now this directory should already be empty, but we empty it of
+	 * any regular files anyway, just in case the directory was created by
+	 * a kernel from the future.... */
+	nfsd4_list_rec_dir(dentry, nfsd4_remove_clid_file);
+	down(&dir->d_inode->i_sem);
+	status = vfs_rmdir(dir->d_inode, dentry);
+	up(&dir->d_inode->i_sem);
+	return status;
+}
+
+static int
+nfsd4_unlink_clid_dir(char *name, int namlen)
+{
+	struct dentry *dentry;
+	int status;
+
+	dprintk("NFSD: nfsd4_unlink_clid_dir. name %.*s\n", namlen, name);
+
+	dentry = lookup_one_len(name, rec_dir.dentry, namlen);
+	if (IS_ERR(dentry)) {
+		status = PTR_ERR(dentry);
+		return status;
+	}
+	status = -ENOENT;
+	if (!dentry->d_inode)
+		goto out;
+
+	status = nfsd4_clear_clid_dir(rec_dir.dentry, dentry);
+out:
+	dput(dentry);
+	return status;
+}
+
+void
+nfsd4_remove_clid_dir(struct nfs4_client *clp)
+{
+	uid_t uid;
+	gid_t gid;
+	int status;
+
+	if (!rec_dir_init || !clp->cl_firststate)
+		return;
+
+	nfs4_save_user(&uid, &gid);
+	status = nfsd4_unlink_clid_dir(clp->cl_recdir, HEXDIR_LEN-1);
+	nfs4_reset_user(uid, gid);
+	if (status == 0)
+		status = nfsd4_rec_fsync(rec_dir.dentry);
+	if (status)
+		printk("NFSD: Failed to remove expired client state directory"
+				" %.*s\n", HEXDIR_LEN, clp->cl_recdir);
+	return;
+}
+
+static int
+purge_old(struct dentry *parent, struct dentry *child)
+{
+	int status;
+
+	if (nfs4_has_reclaimed_state(child->d_name.name))
+		return nfs_ok;
+
+	status = nfsd4_clear_clid_dir(parent, child);
+	if (status)
+		printk("failed to remove client recovery directory %s\n",
+				child->d_name.name);
+	/* Keep trying, success or failure: */
+	return nfs_ok;
+}
+
+void
+nfsd4_recdir_purge_old(void) {
+	int status;
+
+	if (!rec_dir_init)
+		return;
+	status = nfsd4_list_rec_dir(rec_dir.dentry, purge_old);
+	if (status == 0)
+		status = nfsd4_rec_fsync(rec_dir.dentry);
+	if (status)
+		printk("nfsd4: failed to purge old clients from recovery"
+			" directory %s\n", rec_dir.dentry->d_name.name);
+	return;
+}
+
 static int
 load_recdir(struct dentry *parent, struct dentry *child)
 {
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 6b9d23c39af..6cca358cd65 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -905,6 +905,7 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confi
 			conf = find_confirmed_client_by_str(unconf->cl_recdir,
 									hash);
 			if (conf) {
+				nfsd4_remove_clid_dir(conf);
 				expire_client(conf);
 			}
 			move_to_confirmed(unconf);
@@ -1691,6 +1692,7 @@ nfs4_set_claim_prev(struct nfsd4_open *open, int *status)
 			*status = nfserr_reclaim_bad;
 		else {
 			open->op_stateowner->so_confirmed = 1;
+			open->op_stateowner->so_client->cl_firststate = 1;
 			open->op_stateowner->so_seqid--;
 		}
 	}
@@ -1903,6 +1905,7 @@ static void
 end_grace(void)
 {
 	dprintk("NFSD: end of grace period\n");
+	nfsd4_recdir_purge_old();
 	in_grace = 0;
 }
 
@@ -1932,6 +1935,7 @@ nfs4_laundromat(void)
 		}
 		dprintk("NFSD: purging unused client (clientid %08x)\n",
 			clp->cl_clientid.cl_id);
+		nfsd4_remove_clid_dir(clp);
 		expire_client(clp);
 	}
 	INIT_LIST_HEAD(&reaplist);
@@ -2320,6 +2324,8 @@ nfsd4_open_confirm(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfs
 		         stp->st_stateid.si_stateownerid,
 		         stp->st_stateid.si_fileid,
 		         stp->st_stateid.si_generation);
+
+	nfsd4_create_clid_dir(sop->so_client);
 out:
 	if (oc->oc_stateowner)
 		nfs4_get_stateowner(oc->oc_stateowner);
@@ -3089,6 +3095,16 @@ alloc_reclaim(void)
 	return kmalloc(sizeof(struct nfs4_client_reclaim), GFP_KERNEL);
 }
 
+int
+nfs4_has_reclaimed_state(const char *name)
+{
+	unsigned int strhashval = clientstr_hashval(name);
+	struct nfs4_client *clp;
+
+	clp = find_confirmed_client_by_str(name, strhashval);
+	return clp ? 1 : 0;
+}
+
 /*
  * failure => all reset bets are off, nfserr_no_grace...
  */
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index 19481ab122d..a84a3fa99be 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -131,6 +131,7 @@ struct nfs4_client {
 	nfs4_verifier		cl_confirm;	/* generated by server */
 	struct nfs4_callback	cl_callback;    /* callback info */
 	atomic_t		cl_count;	/* ref count */
+	u32			cl_firststate;	/* recovery dir creation */
 };
 
 /* struct nfs4_client_reset
@@ -282,6 +283,10 @@ extern void nfsd4_init_recdir(char *recdir_name);
 extern int nfsd4_recdir_load(void);
 extern void nfsd4_shutdown_recdir(void);
 extern int nfs4_client_to_reclaim(const char *name);
+extern int nfs4_has_reclaimed_state(const char *name);
+extern void nfsd4_recdir_purge_old(void);
+extern int nfsd4_create_clid_dir(struct nfs4_client *clp);
+extern void nfsd4_remove_clid_dir(struct nfs4_client *clp);
 
 static inline void
 nfs4_put_stateowner(struct nfs4_stateowner *so)
-- 
cgit v1.2.3-70-g09d2


From 0964a3d3f1aa96468091924f6b0c391a46dc6d0b Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@cse.unsw.edu.au>
Date: Thu, 23 Jun 2005 22:04:32 -0700
Subject: [PATCH] knfsd: nfsd4 reboot dirname fix

Set the recovery directory via /proc/fs/nfsd/nfs4recoverydir.

It may be changed any time, but is used only on startup.

Signed-off-by: Andy Adamson <andros@citi.umich.edu>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfsd/nfs4recover.c     |  1 -
 fs/nfsd/nfs4state.c       | 37 +++++++++++++++++++++++++++++++++++--
 fs/nfsd/nfsctl.c          | 23 +++++++++++++++++++++++
 include/linux/nfsd/nfsd.h |  2 ++
 4 files changed, 60 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c
index 2805c5245ea..095f1740f3a 100644
--- a/fs/nfsd/nfs4recover.c
+++ b/fs/nfsd/nfs4recover.c
@@ -50,7 +50,6 @@
 #define NFSDDBG_FACILITY                NFSDDBG_PROC
 
 /* Globals */
-char recovery_dirname[PATH_MAX] = "/var/lib/nfs/v4recovery";
 static struct nameidata rec_dir;
 static int rec_dir_init = 0;
 
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 6cca358cd65..89e36526d7f 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -48,6 +48,7 @@
 #include <linux/nfs4.h>
 #include <linux/nfsd/state.h>
 #include <linux/nfsd/xdr4.h>
+#include <linux/namei.h>
 
 #define NFSDDBG_FACILITY                NFSDDBG_PROC
 
@@ -71,7 +72,8 @@ static stateid_t onestateid;              /* bits all 1 */
 static struct nfs4_stateid * find_stateid(stateid_t *stid, int flags);
 static struct nfs4_delegation * find_delegation_stateid(struct inode *ino, stateid_t *stid);
 static void release_stateid_lockowners(struct nfs4_stateid *open_stp);
-extern char recovery_dirname[];
+static char user_recovery_dirname[PATH_MAX] = "/var/lib/nfs/v4recovery";
+static void nfs4_set_recdir(char *recdir);
 
 /* Locking:
  *
@@ -3224,8 +3226,10 @@ nfsd4_load_reboot_recovery_data(void)
 {
 	int status;
 
-	nfsd4_init_recdir(recovery_dirname);
+	nfs4_lock_state();
+	nfsd4_init_recdir(user_recovery_dirname);
 	status = nfsd4_recdir_load();
+	nfs4_unlock_state();
 	if (status)
 		printk("NFSD: Failure reading reboot recovery data\n");
 }
@@ -3329,6 +3333,35 @@ nfs4_state_shutdown(void)
 	nfs4_unlock_state();
 }
 
+static void
+nfs4_set_recdir(char *recdir)
+{
+	nfs4_lock_state();
+	strcpy(user_recovery_dirname, recdir);
+	nfs4_unlock_state();
+}
+
+/*
+ * Change the NFSv4 recovery directory to recdir.
+ */
+int
+nfs4_reset_recoverydir(char *recdir)
+{
+	int status;
+	struct nameidata nd;
+
+	status = path_lookup(recdir, LOOKUP_FOLLOW, &nd);
+	if (status)
+		return status;
+	status = -ENOTDIR;
+	if (S_ISDIR(nd.dentry->d_inode->i_mode)) {
+		nfs4_set_recdir(recdir);
+		status = 0;
+	}
+	path_release(&nd);
+	return status;
+}
+
 /*
  * Called when leasetime is changed.
  *
diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 3da43a3ed32..841c562991e 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -51,6 +51,7 @@ enum {
 	NFSD_Fh,
 	NFSD_Threads,
 	NFSD_Leasetime,
+	NFSD_RecoveryDir,
 };
 
 /*
@@ -66,6 +67,7 @@ static ssize_t write_getfs(struct file *file, char *buf, size_t size);
 static ssize_t write_filehandle(struct file *file, char *buf, size_t size);
 static ssize_t write_threads(struct file *file, char *buf, size_t size);
 static ssize_t write_leasetime(struct file *file, char *buf, size_t size);
+static ssize_t write_recoverydir(struct file *file, char *buf, size_t size);
 
 static ssize_t (*write_op[])(struct file *, char *, size_t) = {
 	[NFSD_Svc] = write_svc,
@@ -78,6 +80,7 @@ static ssize_t (*write_op[])(struct file *, char *, size_t) = {
 	[NFSD_Fh] = write_filehandle,
 	[NFSD_Threads] = write_threads,
 	[NFSD_Leasetime] = write_leasetime,
+	[NFSD_RecoveryDir] = write_recoverydir,
 };
 
 static ssize_t nfsctl_transaction_write(struct file *file, const char __user *buf, size_t size, loff_t *pos)
@@ -349,6 +352,25 @@ static ssize_t write_leasetime(struct file *file, char *buf, size_t size)
 	return strlen(buf);
 }
 
+static ssize_t write_recoverydir(struct file *file, char *buf, size_t size)
+{
+	char *mesg = buf;
+	char *recdir;
+	int len, status;
+
+	if (size > PATH_MAX || buf[size-1] != '\n')
+		return -EINVAL;
+	buf[size-1] = 0;
+
+	recdir = mesg;
+	len = qword_get(&mesg, recdir, size);
+	if (len <= 0)
+		return -EINVAL;
+
+	status = nfs4_reset_recoverydir(recdir);
+	return strlen(buf);
+}
+
 /*----------------------------------------------------------------------------*/
 /*
  *	populating the filesystem.
@@ -369,6 +391,7 @@ static int nfsd_fill_super(struct super_block * sb, void * data, int silent)
 		[NFSD_Threads] = {"threads", &transaction_ops, S_IWUSR|S_IRUSR},
 #ifdef CONFIG_NFSD_V4
 		[NFSD_Leasetime] = {"nfsv4leasetime", &transaction_ops, S_IWUSR|S_IRUSR},
+		[NFSD_RecoveryDir] = {"nfsv4recoverydir", &transaction_ops, S_IWUSR|S_IRUSR},
 #endif
 		/* last one */ {""}
 	};
diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h
index 21c6e9d86e4..5791dfd30dd 100644
--- a/include/linux/nfsd/nfsd.h
+++ b/include/linux/nfsd/nfsd.h
@@ -150,12 +150,14 @@ int nfs4_state_start(void);
 void nfs4_state_shutdown(void);
 time_t nfs4_lease_time(void);
 void nfs4_reset_lease(time_t leasetime);
+int nfs4_reset_recoverydir(char *recdir);
 #else
 static inline void nfs4_state_init(void){};
 static inline int nfs4_state_start(void){return 0;}
 static inline void nfs4_state_shutdown(void){}
 static inline time_t nfs4_lease_time(void){return 0;}
 static inline void nfs4_reset_lease(time_t leasetime){}
+static inline int nfs4_reset_recoverydir(char *recdir) {return 0;}
 #endif
 
 /*
-- 
cgit v1.2.3-70-g09d2


From 239df2e2b0e1f4f69fdf76fb67e865824029e8ab Mon Sep 17 00:00:00 2001
From: Manuel Capinha <mcapinha@gmail.com>
Date: Thu, 23 Jun 2005 22:04:53 -0700
Subject: [PATCH] v4l: add support for PixelView Ultra Pro

The following patch adds support for the PixelView Ultra Pro video capture
card in v4l.

- It removes the remote control key definitions from ir-kbd-gpio.c and
  moves them to ir-common.c so that they can be shared between bt878 and
  cx88 based cards.

- The patch also moves the FUSIONHDTV_3_GOLD_Q card from number 27 to 28
  to regain compatibility with the V4L cvs.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/media/common/ir-common.c      | 33 +++++++++++++++++++++++++++++++++
 drivers/media/video/cx88/cx88-cards.c | 21 +++++++++++++++++++++
 drivers/media/video/cx88/cx88-input.c |  7 +++++++
 drivers/media/video/cx88/cx88.h       |  3 ++-
 drivers/media/video/ir-kbd-gpio.c     | 32 --------------------------------
 include/media/ir-common.h             |  1 +
 6 files changed, 64 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/drivers/media/common/ir-common.c b/drivers/media/common/ir-common.c
index 84a49d2ec91..e5636ef181b 100644
--- a/drivers/media/common/ir-common.c
+++ b/drivers/media/common/ir-common.c
@@ -213,6 +213,39 @@ IR_KEYTAB_TYPE ir_codes_hauppauge_new[IR_KEYTAB_SIZE] = {
 };
 EXPORT_SYMBOL(ir_codes_hauppauge_new);
 
+IR_KEYTAB_TYPE ir_codes_pixelview[IR_KEYTAB_SIZE] = {
+	[  2 ] = KEY_KP0,
+	[  1 ] = KEY_KP1,
+	[ 11 ] = KEY_KP2,
+	[ 27 ] = KEY_KP3,
+	[  5 ] = KEY_KP4,
+	[  9 ] = KEY_KP5,
+	[ 21 ] = KEY_KP6,
+	[  6 ] = KEY_KP7,
+	[ 10 ] = KEY_KP8,
+	[ 18 ] = KEY_KP9,
+
+	[  3 ] = KEY_TUNER,       // TV/FM
+	[  7 ] = KEY_SEARCH,      // scan
+	[ 28 ] = KEY_ZOOM,        // full screen
+	[ 30 ] = KEY_POWER,
+	[ 23 ] = KEY_VOLUMEDOWN,
+	[ 31 ] = KEY_VOLUMEUP,
+	[ 20 ] = KEY_CHANNELDOWN,
+	[ 22 ] = KEY_CHANNELUP,
+	[ 24 ] = KEY_MUTE,
+
+	[  0 ] = KEY_LIST,        // source
+	[ 19 ] = KEY_INFO,        // loop
+	[ 16 ] = KEY_LAST,        // +100
+	[ 13 ] = KEY_CLEAR,       // reset
+	[ 12 ] = BTN_RIGHT,       // fun++
+	[  4 ] = BTN_LEFT,        // fun--
+	[ 14 ] = KEY_GOTO,        // function
+	[ 15 ] = KEY_STOP,         // freeze
+};
+EXPORT_SYMBOL(ir_codes_pixelview);
+
 /* -------------------------------------------------------------------------- */
 
 static void ir_input_key_event(struct input_dev *dev, struct ir_input_state *ir)
diff --git a/drivers/media/video/cx88/cx88-cards.c b/drivers/media/video/cx88/cx88-cards.c
index 98e40026aa4..1c036cc36fe 100644
--- a/drivers/media/video/cx88/cx88-cards.c
+++ b/drivers/media/video/cx88/cx88-cards.c
@@ -628,6 +628,27 @@ struct cx88_board cx88_boards[] = {
 			.gpio1  = 0x0000e07f,
 		}}
 	},
+	[CX88_BOARD_PIXELVIEW_PLAYTV_ULTRA_PRO] = {
+		.name           = "PixelView PlayTV Ultra Pro (Stereo)",
+		.tuner_type     = 38,
+		.input          = {{
+			.type   = CX88_VMUX_TELEVISION,
+			.vmux   = 0,
+			.gpio0  = 0xbf61,  // internal decoder
+		},{
+			.type   = CX88_VMUX_COMPOSITE1,
+			.vmux   = 1,
+			.gpio0  = 0xbf63,
+		},{
+			.type   = CX88_VMUX_SVIDEO,
+			.vmux   = 2,
+			.gpio0  = 0xbf63,
+		}},
+		.radio = {
+			.type  = CX88_RADIO,
+			.gpio0 = 0xbf60,
+		},
+	},
 };
 const unsigned int cx88_bcount = ARRAY_SIZE(cx88_boards);
 
diff --git a/drivers/media/video/cx88/cx88-input.c b/drivers/media/video/cx88/cx88-input.c
index af6ad8cdbdb..fbf21dbe251 100644
--- a/drivers/media/video/cx88/cx88-input.c
+++ b/drivers/media/video/cx88/cx88-input.c
@@ -261,6 +261,13 @@ int cx88_ir_init(struct cx88_core *core, struct pci_dev *pci)
 		ir->mask_keydown = 0x02;
 		ir->polling      = 5; // ms
 		break;
+	case CX88_BOARD_PIXELVIEW_PLAYTV_ULTRA_PRO:
+		ir_codes         = ir_codes_pixelview;
+		ir->gpio_addr    = MO_GP1_IO;
+		ir->mask_keycode = 0x1f;
+		ir->mask_keyup   = 0x80;
+		ir->polling      = 1; // ms
+		break;
 	}
 	if (NULL == ir_codes) {
 		kfree(ir);
diff --git a/drivers/media/video/cx88/cx88.h b/drivers/media/video/cx88/cx88.h
index 0ea24b72d92..7fca1f500c5 100644
--- a/drivers/media/video/cx88/cx88.h
+++ b/drivers/media/video/cx88/cx88.h
@@ -162,7 +162,8 @@ extern struct sram_channel cx88_sram_channels[];
 #define CX88_BOARD_HAUPPAUGE_ROSLYN        24
 #define CX88_BOARD_DIGITALLOGIC_MEC	       25
 #define CX88_BOARD_IODATA_GVBCTV7E         26
-#define CX88_BOARD_DVICO_FUSIONHDTV_3_GOLD_Q   27
+#define CX88_BOARD_PIXELVIEW_PLAYTV_ULTRA_PRO 27
+#define CX88_BOARD_DVICO_FUSIONHDTV_3_GOLD_Q   28
 
 enum cx88_itype {
 	CX88_VMUX_COMPOSITE1 = 1,
diff --git a/drivers/media/video/ir-kbd-gpio.c b/drivers/media/video/ir-kbd-gpio.c
index ab6620de4b3..a9d4b2ad14e 100644
--- a/drivers/media/video/ir-kbd-gpio.c
+++ b/drivers/media/video/ir-kbd-gpio.c
@@ -114,38 +114,6 @@ static IR_KEYTAB_TYPE ir_codes_avermedia_dvbt[IR_KEYTAB_SIZE] = {
 	[ 0x3e ] = KEY_VOLUMEUP,    // 'volume +'
 };
 
-static IR_KEYTAB_TYPE ir_codes_pixelview[IR_KEYTAB_SIZE] = {
-	[  2 ] = KEY_KP0,
-	[  1 ] = KEY_KP1,
-	[ 11 ] = KEY_KP2,
-	[ 27 ] = KEY_KP3,
-	[  5 ] = KEY_KP4,
-	[  9 ] = KEY_KP5,
-	[ 21 ] = KEY_KP6,
-	[  6 ] = KEY_KP7,
-	[ 10 ] = KEY_KP8,
-	[ 18 ] = KEY_KP9,
-
-	[  3 ] = KEY_TUNER,       // TV/FM
-	[  7 ] = KEY_SEARCH,      // scan
-	[ 28 ] = KEY_ZOOM,        // full screen
-	[ 30 ] = KEY_POWER,
-	[ 23 ] = KEY_VOLUMEDOWN,
-	[ 31 ] = KEY_VOLUMEUP,
-	[ 20 ] = KEY_CHANNELDOWN,
-	[ 22 ] = KEY_CHANNELUP,
-	[ 24 ] = KEY_MUTE,
-
-	[  0 ] = KEY_LIST,        // source
-	[ 19 ] = KEY_INFO,        // loop
-	[ 16 ] = KEY_LAST,        // +100
-	[ 13 ] = KEY_CLEAR,       // reset
-	[ 12 ] = BTN_RIGHT,       // fun++
-	[  4 ] = BTN_LEFT,        // fun--
-	[ 14 ] = KEY_GOTO,        // function
-	[ 15 ] = KEY_STOP,         // freeze
-};
-
 /* Attila Kondoros <attila.kondoros@chello.hu> */
 static IR_KEYTAB_TYPE ir_codes_apac_viewcomp[IR_KEYTAB_SIZE] = {
 
diff --git a/include/media/ir-common.h b/include/media/ir-common.h
index 62c963a52d8..b5fa6c585e2 100644
--- a/include/media/ir-common.h
+++ b/include/media/ir-common.h
@@ -50,6 +50,7 @@ extern IR_KEYTAB_TYPE ir_codes_rc5_tv[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_winfast[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_empty[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_hauppauge_new[IR_KEYTAB_SIZE];
+extern IR_KEYTAB_TYPE ir_codes_pixelview[IR_KEYTAB_SIZE];
 
 void ir_input_init(struct input_dev *dev, struct ir_input_state *ir,
 		   int ir_type, IR_KEYTAB_TYPE *ir_codes);
-- 
cgit v1.2.3-70-g09d2


From 3c1d0185db6a44b6304c404f4da1a1a98746ca46 Mon Sep 17 00:00:00 2001
From: Michael Krufky <mkrufky@m1k.net>
Date: Thu, 23 Jun 2005 22:04:56 -0700
Subject: [PATCH] v4l: support tuner for Thomson DDT 7611 (ATSC/NTSC)

Add support for tuner#60: Thomson DDT 7611 (ATSC/NTSC) Change tuner in
card#28 (DViCO FusionHDTV3 Gold-T) from tuner=52 (Tuner Thomson DDT 7610)
to tuner=60 (Tuner Thomson DDT 7611)

Signed-off-by: Michael Krufky <mkrufky@m1k.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/media/video/cx88/cx88-cards.c | 2 +-
 drivers/media/video/tuner-simple.c    | 3 +++
 include/media/tuner.h                 | 2 ++
 3 files changed, 6 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/media/video/cx88/cx88-cards.c b/drivers/media/video/cx88/cx88-cards.c
index c3a561e38e5..a6763f1a44c 100644
--- a/drivers/media/video/cx88/cx88-cards.c
+++ b/drivers/media/video/cx88/cx88-cards.c
@@ -431,7 +431,7 @@ struct cx88_board cx88_boards[] = {
 	},
         [CX88_BOARD_DVICO_FUSIONHDTV_3_GOLD_T] = {
                 .name           = "DViCO - FusionHDTV 3 Gold-T",
-                .tuner_type     = 52, /* Thomson DDT 7611 ATSC/NTSC */
+                .tuner_type     = 60, /* Thomson DDT 7611 ATSC/NTSC */
                /*  See DViCO FusionHDTV 3 Gold for GPIO documentation.  */
                 .input          = {{
                         .type   = CX88_VMUX_TELEVISION,
diff --git a/drivers/media/video/tuner-simple.c b/drivers/media/video/tuner-simple.c
index f7305c8d53d..866f18dc5b5 100644
--- a/drivers/media/video/tuner-simple.c
+++ b/drivers/media/video/tuner-simple.c
@@ -217,6 +217,9 @@ static struct tunertype tuners[] = {
 	  16*160.00,16*454.00,0xa0,0x90,0x30,0x8e,732},
 	{ "Ymec TVision TVF-5533MF", Philips, NTSC,
 	  16*160.00,16*454.00,0x01,0x02,0x04,0x8e,732},
+
+	{ "Thomson DDT 7611 (ATSC/NTSC)", THOMSON, ATSC,
+	  16*157.25,16*454.00,0x39,0x3a,0x3c,0x8e,732},
 };
 unsigned const int tuner_count = ARRAY_SIZE(tuners);
 
diff --git a/include/media/tuner.h b/include/media/tuner.h
index 6d29d8db0c8..cb75f4809c4 100644
--- a/include/media/tuner.h
+++ b/include/media/tuner.h
@@ -101,6 +101,8 @@
 #define TUNER_YMEC_TVF_8531MF 58
 #define TUNER_YMEC_TVF_5533MF 59	/* Pixelview Pro Ultra NTSC */
 
+#define TUNER_THOMSON_DTT7611    60
+
 #define NOTUNER 0
 #define PAL     1	/* PAL_BG */
 #define PAL_I   2
-- 
cgit v1.2.3-70-g09d2


From 56fc08ca375491b965cb76fad65bfb98973e80d8 Mon Sep 17 00:00:00 2001
From: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
Date: Thu, 23 Jun 2005 22:05:07 -0700
Subject: [PATCH] v4l: update for tuner cards and some V4L chips

Tuner improvements and additions.  TEA5767 FM tuner added.  Several small
fixes.

Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
Signed-off-by: Nickolay V Shmyrev <nshmyrev@yandex.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/media/common/ir-common.c   |  39 ++++---
 drivers/media/video/bt832.c        |  50 +++++----
 drivers/media/video/bt832.h        |  36 +++----
 drivers/media/video/msp3400.c      |  13 +--
 drivers/media/video/msp3400.h      |   4 +
 drivers/media/video/tda7432.c      |   1 +
 drivers/media/video/tda9875.c      |   1 +
 drivers/media/video/tda9887.c      |  41 ++++++-
 drivers/media/video/tuner-core.c   | 212 +++++++++++++++++++++++++------------
 drivers/media/video/tuner-simple.c |  52 +++++----
 drivers/media/video/tvaudio.c      |   2 +-
 drivers/media/video/tvmixer.c      |   4 +
 include/media/audiochip.h          |   4 +
 include/media/id.h                 |   4 +
 include/media/ir-common.h          |   2 +-
 include/media/tuner.h              |  23 +++-
 16 files changed, 333 insertions(+), 155 deletions(-)

(limited to 'include')

diff --git a/drivers/media/common/ir-common.c b/drivers/media/common/ir-common.c
index e5636ef181b..4adb2843f8b 100644
--- a/drivers/media/common/ir-common.c
+++ b/drivers/media/common/ir-common.c
@@ -1,5 +1,5 @@
 /*
- * $Id: ir-common.c,v 1.8 2005/02/22 12:28:40 kraxel Exp $
+ * $Id: ir-common.c,v 1.10 2005/05/22 19:23:39 nsh Exp $
  *
  * some common structs and functions to handle infrared remotes via
  * input layer ...
@@ -131,10 +131,10 @@ IR_KEYTAB_TYPE ir_codes_winfast[IR_KEYTAB_SIZE] = {
 	[ 18 ] = KEY_KP0,
 
 	[  0 ] = KEY_POWER,
-	[ 27 ] = KEY_LANGUAGE,  //MTS button
+//      [ 27 ] = MTS button
 	[  2 ] = KEY_TUNER,     // TV/FM
 	[ 30 ] = KEY_VIDEO,
-	[ 22 ] = KEY_INFO,      //display button
+//      [ 22 ] = display button
 	[  4 ] = KEY_VOLUMEUP,
 	[  8 ] = KEY_VOLUMEDOWN,
 	[ 12 ] = KEY_CHANNELUP,
@@ -142,7 +142,7 @@ IR_KEYTAB_TYPE ir_codes_winfast[IR_KEYTAB_SIZE] = {
 	[  3 ] = KEY_ZOOM,      // fullscreen
 	[ 31 ] = KEY_SUBTITLE,  // closed caption/teletext
 	[ 32 ] = KEY_SLEEP,
-	[ 41 ] = KEY_SEARCH,    //boss key
+//      [ 41 ] = boss key
 	[ 20 ] = KEY_MUTE,
 	[ 43 ] = KEY_RED,
 	[ 44 ] = KEY_GREEN,
@@ -150,17 +150,17 @@ IR_KEYTAB_TYPE ir_codes_winfast[IR_KEYTAB_SIZE] = {
 	[ 46 ] = KEY_BLUE,
 	[ 24 ] = KEY_KPPLUS,    //fine tune +
 	[ 25 ] = KEY_KPMINUS,   //fine tune -
-	[ 42 ] = KEY_ANGLE,     //picture in picture
-	[ 33 ] = KEY_KPDOT,
+//      [ 42 ] = picture in picture
+        [ 33 ] = KEY_KPDOT,
 	[ 19 ] = KEY_KPENTER,
-	[ 17 ] = KEY_AGAIN,     //recall
+//      [ 17 ] = recall
 	[ 34 ] = KEY_BACK,
 	[ 35 ] = KEY_PLAYPAUSE,
 	[ 36 ] = KEY_NEXT,
-	[ 37 ] = KEY_T,         //time shifting
+//      [ 37 ] = time shifting
 	[ 38 ] = KEY_STOP,
-	[ 39 ] = KEY_RECORD,
-	[ 40 ] = KEY_SHUFFLE    //snapshot
+	[ 39 ] = KEY_RECORD
+//      [ 40 ] = snapshot
 };
 EXPORT_SYMBOL_GPL(ir_codes_winfast);
 
@@ -184,18 +184,30 @@ IR_KEYTAB_TYPE ir_codes_hauppauge_new[IR_KEYTAB_SIZE] = {
 	[ 0x07 ] = KEY_KP7,             // 7
 	[ 0x08 ] = KEY_KP8,             // 8
 	[ 0x09 ] = KEY_KP9,             // 9
+	[ 0x0a ] = KEY_TEXT,      	// keypad asterisk as well
 	[ 0x0b ] = KEY_RED,             // red button
-	[ 0x0c ] = KEY_OPTION,          // black key without text
+	[ 0x0c ] = KEY_RADIO,           // radio
 	[ 0x0d ] = KEY_MENU,            // menu
+	[ 0x0e ] = KEY_SUBTITLE,	// also the # key
 	[ 0x0f ] = KEY_MUTE,            // mute
 	[ 0x10 ] = KEY_VOLUMEUP,        // volume +
 	[ 0x11 ] = KEY_VOLUMEDOWN,      // volume -
-	[ 0x1e ] = KEY_NEXT,            // skip >|
+	[ 0x12 ] = KEY_PREVIOUS,        // previous channel
+	[ 0x14 ] = KEY_UP,              // up
+	[ 0x15 ] = KEY_DOWN,		// down
+	[ 0x16 ] = KEY_LEFT,		// left
+	[ 0x17 ] = KEY_RIGHT,		// right
+	[ 0x18 ] = KEY_VIDEO,		// Videos
+	[ 0x19 ] = KEY_AUDIO,		// Music
+	[ 0x1a ] = KEY_MHP,		// Pictures - presume this means "Multimedia Home Platform"- no "PICTURES" key in input.h
+	[ 0x1b ] = KEY_EPG,		// Guide
+	[ 0x1c ] = KEY_TV,		// TV
+	[ 0x1e ] = KEY_NEXTSONG,        // skip >|
 	[ 0x1f ] = KEY_EXIT,            // back/exit
 	[ 0x20 ] = KEY_CHANNELUP,       // channel / program +
 	[ 0x21 ] = KEY_CHANNELDOWN,     // channel / program -
 	[ 0x22 ] = KEY_CHANNEL,         // source (old black remote)
-	[ 0x24 ] = KEY_PREVIOUS,        // replay |<
+	[ 0x24 ] = KEY_PREVIOUSSONG,    // replay |<
 	[ 0x25 ] = KEY_ENTER,           // OK
 	[ 0x26 ] = KEY_SLEEP,           // minimize (old black remote)
 	[ 0x29 ] = KEY_BLUE,            // blue key
@@ -412,3 +424,4 @@ EXPORT_SYMBOL_GPL(ir_decode_biphase);
  * c-basic-offset: 8
  * End:
  */
+
diff --git a/drivers/media/video/bt832.c b/drivers/media/video/bt832.c
index 07f72f64c5f..3bb347d93b9 100644
--- a/drivers/media/video/bt832.c
+++ b/drivers/media/video/bt832.c
@@ -6,7 +6,7 @@
   It outputs an 8-bit 4:2:2 YUV or YCrCb video signal which can be directly
   connected to bt848/bt878 GPIO pins on this purpose.
   (see: VLSI Vision Ltd. www.vvl.co.uk for camera datasheets)
-  
+
   Supported Cards:
   -  Pixelview Rev.4E: 0x8a
 		GPIO 0x400000 toggles Bt832 RESET, and the chip changes to i2c 0x88 !
@@ -31,16 +31,16 @@
 #include <linux/errno.h>
 #include <linux/slab.h>
 
-#include "id.h"
-#include "audiochip.h"
+#include <media/audiochip.h>
+#include <media/id.h>
 #include "bttv.h"
 #include "bt832.h"
 
 MODULE_LICENSE("GPL");
 
 /* Addresses to scan */
-static unsigned short normal_i2c[] = { I2C_BT832_ALT1>>1, I2C_BT832_ALT2>>1,
-				       I2C_CLIENT_END };
+static unsigned short normal_i2c[] = {I2C_CLIENT_END};
+static unsigned short normal_i2c_range[] = {I2C_BT832_ALT1>>1,I2C_BT832_ALT2>>1,I2C_CLIENT_END};
 I2C_CLIENT_INSMOD;
 
 /* ---------------------------------------------------------------------- */
@@ -95,7 +95,7 @@ int bt832_init(struct i2c_client *i2c_client_s)
 
 	buf=kmalloc(65,GFP_KERNEL);
 	bt832_hexdump(i2c_client_s,buf);
-	
+
 	if(buf[0x40] != 0x31) {
 		printk("bt832: this i2c chip is no bt832 (id=%02x). Detaching.\n",buf[0x40]);
 		kfree(buf);
@@ -135,7 +135,7 @@ int bt832_init(struct i2c_client *i2c_client_s)
 	buf[1]= 0x27 & (~0x01); // Default | !skip
 	if (2 != (rc = i2c_master_send(i2c_client_s,buf,2)))
                 printk("bt832: i2c i/o error EO: rc == %d (should be 2)\n",rc);
-	
+
         bt832_hexdump(i2c_client_s,buf);
 
 #if 0
@@ -168,8 +168,7 @@ int bt832_init(struct i2c_client *i2c_client_s)
 
 
 
-static int bt832_attach(struct i2c_adapter *adap, int addr,
-			  unsigned short flags, int kind)
+static int bt832_attach(struct i2c_adapter *adap, int addr, int kind)
 {
 	struct bt832 *t;
 
@@ -184,27 +183,32 @@ static int bt832_attach(struct i2c_adapter *adap, int addr,
                 return -ENOMEM;
 	memset(t,0,sizeof(*t));
 	t->client = client_template;
-        t->client.data = t;
+        i2c_set_clientdata(&t->client, t);
         i2c_attach_client(&t->client);
 
 	if(! bt832_init(&t->client)) {
 		bt832_detach(&t->client);
 		return -1;
 	}
-        
+
 	return 0;
 }
 
 static int bt832_probe(struct i2c_adapter *adap)
 {
+#ifdef I2C_CLASS_TV_ANALOG
 	if (adap->class & I2C_CLASS_TV_ANALOG)
 		return i2c_probe(adap, &addr_data, bt832_attach);
+#else
+	if (adap->id == (I2C_ALGO_BIT | I2C_HW_B_BT848))
+		return i2c_probe(adap, &addr_data, bt832_attach);
+#endif
 	return 0;
 }
 
 static int bt832_detach(struct i2c_client *client)
 {
-	struct bt832 *t = (struct bt832*)client->data;
+	struct bt832 *t = i2c_get_clientdata(client);
 
 	printk("bt832: detach.\n");
 	i2c_detach_client(client);
@@ -215,7 +219,7 @@ static int bt832_detach(struct i2c_client *client)
 static int
 bt832_command(struct i2c_client *client, unsigned int cmd, void *arg)
 {
-	struct bt832 *t = (struct bt832*)client->data;
+	struct bt832 *t = i2c_get_clientdata(client);
 
 	printk("bt832: command %x\n",cmd);
 
@@ -249,19 +253,18 @@ static struct i2c_driver driver = {
 };
 static struct i2c_client client_template =
 {
-        .name   = "bt832",
-	.flags  = I2C_CLIENT_ALLOW_USE,
-        .driver = &driver,
+	I2C_DEVNAME("bt832"),
+	.flags      = I2C_CLIENT_ALLOW_USE,
+        .driver     = &driver,
 };
 
 
-int bt832_init_module(void)
+static int __init bt832_init_module(void)
 {
-	i2c_add_driver(&driver);
-	return 0;
+	return i2c_add_driver(&driver);
 }
 
-static void bt832_cleanup_module(void)
+static void __exit bt832_cleanup_module(void)
 {
 	i2c_del_driver(&driver);
 }
@@ -269,3 +272,10 @@ static void bt832_cleanup_module(void)
 module_init(bt832_init_module);
 module_exit(bt832_cleanup_module);
 
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-basic-offset: 8
+ * End:
+ */
diff --git a/drivers/media/video/bt832.h b/drivers/media/video/bt832.h
index 7a98c06e0e3..9b6a8d2c96b 100644
--- a/drivers/media/video/bt832.h
+++ b/drivers/media/video/bt832.h
@@ -1,6 +1,6 @@
 /* Bt832 CMOS Camera Video Processor (VP)
 
- The Bt832 CMOS Camera Video Processor chip connects a Quartsight CMOS 
+ The Bt832 CMOS Camera Video Processor chip connects a Quartsight CMOS
   color digital camera directly to video capture devices via an 8-bit,
   4:2:2 YUV or YCrCb video interface.
 
@@ -85,7 +85,7 @@
 #define BT832_DEVICE_ID		63
 # define BT832_DEVICE_ID__31		0x31 // Bt832 has ID 0x31
 
-/* STMicroelectronivcs VV5404 camera module 
+/* STMicroelectronivcs VV5404 camera module
    i2c: 0x20: sensor address
    i2c: 0xa0: eeprom for ccd defect map
  */
@@ -256,26 +256,26 @@ For the CCIR-601 standards, the sampling is based on a static orthogonal samplin
 //===========================================================================
 // Timing generator SRAM table values for CCIR601 720x480 NTSC
 //===========================================================================
-// For NTSC CCIR656 
+// For NTSC CCIR656
 BYTE BtCard::SRAMTable_NTSC[] =
 {
     // SRAM Timing Table for NTSC
-    0x0c, 0xc0, 0x00, 
-    0x00, 0x90, 0xc2, 
-    0x03, 0x10, 0x03, 
-    0x06, 0x10, 0x34, 
-    0x12, 0x12, 0x65, 
-    0x02, 0x13, 0x24, 
-    0x19, 0x00, 0x24, 
-    0x39, 0x00, 0x96, 
-    0x59, 0x08, 0x93, 
+    0x0c, 0xc0, 0x00,
+    0x00, 0x90, 0xc2,
+    0x03, 0x10, 0x03,
+    0x06, 0x10, 0x34,
+    0x12, 0x12, 0x65,
+    0x02, 0x13, 0x24,
+    0x19, 0x00, 0x24,
+    0x39, 0x00, 0x96,
+    0x59, 0x08, 0x93,
     0x83, 0x08, 0x97,
-    0x03, 0x50, 0x30, 
-    0xc0, 0x40, 0x30, 
-    0x86, 0x01, 0x01, 
-    0xa6, 0x0d, 0x62, 
-    0x03, 0x11, 0x61, 
-    0x05, 0x37, 0x30, 
+    0x03, 0x50, 0x30,
+    0xc0, 0x40, 0x30,
+    0x86, 0x01, 0x01,
+    0xa6, 0x0d, 0x62,
+    0x03, 0x11, 0x61,
+    0x05, 0x37, 0x30,
     0xac, 0x21, 0x50
 };
 
diff --git a/drivers/media/video/msp3400.c b/drivers/media/video/msp3400.c
index 09464d624a6..05b83faa9a0 100644
--- a/drivers/media/video/msp3400.c
+++ b/drivers/media/video/msp3400.c
@@ -147,6 +147,7 @@ static unsigned short normal_i2c[] = {
 	I2C_MSP3400C_ALT  >> 1,
 	I2C_CLIENT_END
 };
+static unsigned short normal_i2c_range[] = {I2C_CLIENT_END,I2C_CLIENT_END};
 I2C_CLIENT_INSMOD;
 
 /* ----------------------------------------------------------------------- */
@@ -735,7 +736,6 @@ static int msp34xx_sleep(struct msp3400c *msp, int timeout)
 {
 	DECLARE_WAITQUEUE(wait, current);
 
-again:
 	add_wait_queue(&msp->wq, &wait);
 	if (!kthread_should_stop()) {
 		if (timeout < 0) {
@@ -751,12 +751,9 @@ again:
 #endif
 		}
 	}
-
+	if (current->flags & PF_FREEZE)
+		refrigerator(PF_FREEZE);
 	remove_wait_queue(&msp->wq, &wait);
-
-	if (try_to_freeze(PF_FREEZE))
-		goto again;
-
 	return msp->restart;
 }
 
@@ -1436,7 +1433,7 @@ static int msp_detach(struct i2c_client *client);
 static int msp_probe(struct i2c_adapter *adap);
 static int msp_command(struct i2c_client *client, unsigned int cmd, void *arg);
 
-static int msp_suspend(struct device * dev, pm_message_t state, u32 level);
+static int msp_suspend(struct device * dev, u32 state, u32 level);
 static int msp_resume(struct device * dev, u32 level);
 
 static void msp_wake_thread(struct i2c_client *client);
@@ -1841,7 +1838,7 @@ static int msp_command(struct i2c_client *client, unsigned int cmd, void *arg)
 	return 0;
 }
 
-static int msp_suspend(struct device * dev, pm_message_t state, u32 level)
+static int msp_suspend(struct device * dev, u32 state, u32 level)
 {
 	struct i2c_client *c = container_of(dev, struct i2c_client, dev);
 
diff --git a/drivers/media/video/msp3400.h b/drivers/media/video/msp3400.h
index d70a954e13a..023f33056a4 100644
--- a/drivers/media/video/msp3400.h
+++ b/drivers/media/video/msp3400.h
@@ -1,3 +1,7 @@
+/*
+ * $Id: msp3400.h,v 1.3 2005/06/12 04:19:19 mchehab Exp $
+ */
+
 #ifndef MSP3400_H
 #define MSP3400_H
 
diff --git a/drivers/media/video/tda7432.c b/drivers/media/video/tda7432.c
index 07ba6d3ed08..376a4a439e9 100644
--- a/drivers/media/video/tda7432.c
+++ b/drivers/media/video/tda7432.c
@@ -74,6 +74,7 @@ static unsigned short normal_i2c[] = {
 	I2C_TDA7432 >> 1,
 	I2C_CLIENT_END,
 };
+static unsigned short normal_i2c_range[] = { I2C_CLIENT_END, I2C_CLIENT_END };
 I2C_CLIENT_INSMOD;
 
 /* Structure of address and subaddresses for the tda7432 */
diff --git a/drivers/media/video/tda9875.c b/drivers/media/video/tda9875.c
index 97b113e070f..4f1114c033a 100644
--- a/drivers/media/video/tda9875.c
+++ b/drivers/media/video/tda9875.c
@@ -44,6 +44,7 @@ static unsigned short normal_i2c[] =  {
     I2C_TDA9875 >> 1,
     I2C_CLIENT_END
 };
+static unsigned short normal_i2c_range[] = {I2C_CLIENT_END};
 I2C_CLIENT_INSMOD;
 
 /* This is a superset of the TDA9875 */
diff --git a/drivers/media/video/tda9887.c b/drivers/media/video/tda9887.c
index 7e6e6dd966a..33d6ee6cde4 100644
--- a/drivers/media/video/tda9887.c
+++ b/drivers/media/video/tda9887.c
@@ -33,6 +33,7 @@ static unsigned short normal_i2c[] = {
 	0x96 >>1,
 	I2C_CLIENT_END,
 };
+static unsigned short normal_i2c_range[] = {I2C_CLIENT_END,I2C_CLIENT_END};
 I2C_CLIENT_INSMOD;
 
 /* insmod options */
@@ -53,6 +54,7 @@ struct tda9887 {
 	unsigned int       config;
 	unsigned int       pinnacle_id;
 	unsigned int       using_v4l2;
+	unsigned int 	   radio_mode;
 };
 
 struct tvnorm {
@@ -212,12 +214,22 @@ static struct tvnorm tvnorms[] = {
 	}
 };
 
-static struct tvnorm radio = {
-	.name = "radio",
+static struct tvnorm radio_stereo = {
+	.name = "Radio Stereo",
+	.b    = ( cFmRadio       |
+		  cQSS           ),
+	.c    = ( cDeemphasisOFF |
+		  cAudioGain6 ),
+	.e    = ( cAudioIF_5_5   |
+		  cRadioIF_38_90 ),
+};
+
+static struct tvnorm radio_mono = {
+	.name = "Radio Mono",
 	.b    = ( cFmRadio       |
 		  cQSS           ),
 	.c    = ( cDeemphasisON  |
-		  cDeemphasis50  ),
+		  cDeemphasis50),
 	.e    = ( cAudioIF_5_5   |
 		  cRadioIF_38_90 ),
 };
@@ -354,7 +366,10 @@ static int tda9887_set_tvnorm(struct tda9887 *t, char *buf)
 	int i;
 
 	if (t->radio) {
-		norm = &radio;
+		if (t->radio_mode == V4L2_TUNER_MODE_MONO)
+			norm = &radio_mono;
+		else
+		        norm = &radio_stereo;
 	} else {
 		for (i = 0; i < ARRAY_SIZE(tvnorms); i++) {
 			if (tvnorms[i].std & t->std) {
@@ -545,11 +560,14 @@ static int tda9887_configure(struct tda9887 *t)
 
 	memset(buf,0,sizeof(buf));
 	tda9887_set_tvnorm(t,buf);
+
 	buf[1] |= cOutputPort1Inactive;
 	buf[1] |= cOutputPort2Inactive;
+
 	if (UNSET != t->pinnacle_id) {
 		tda9887_set_pinnacle(t,buf);
 	}
+
 	tda9887_set_config(t,buf);
 	tda9887_set_insmod(t,buf);
 
@@ -592,9 +610,12 @@ static int tda9887_attach(struct i2c_adapter *adap, int addr, int kind)
         if (NULL == (t = kmalloc(sizeof(*t), GFP_KERNEL)))
                 return -ENOMEM;
 	memset(t,0,sizeof(*t));
+
 	t->client      = client_template;
 	t->std         = 0;
 	t->pinnacle_id = UNSET;
+	t->radio_mode = V4L2_TUNER_MODE_STEREO;
+
         i2c_set_clientdata(&t->client, t);
         i2c_attach_client(&t->client);
 
@@ -733,6 +754,16 @@ tda9887_command(struct i2c_client *client, unsigned int cmd, void *arg)
 		}
 		break;
 	}
+	case VIDIOC_S_TUNER:
+	{
+		struct v4l2_tuner* tuner = arg;
+
+		if (t->radio) {
+			t->radio_mode = tuner->audmode;
+			tda9887_configure (t);
+		}
+		break;
+	}
 	default:
 		/* nothing */
 		break;
@@ -740,7 +771,7 @@ tda9887_command(struct i2c_client *client, unsigned int cmd, void *arg)
 	return 0;
 }
 
-static int tda9887_suspend(struct device * dev, pm_message_t state, u32 level)
+static int tda9887_suspend(struct device * dev, u32 state, u32 level)
 {
 	dprintk("tda9887: suspend\n");
 	return 0;
diff --git a/drivers/media/video/tuner-core.c b/drivers/media/video/tuner-core.c
index 71423ae3b4d..ba13bfadb52 100644
--- a/drivers/media/video/tuner-core.c
+++ b/drivers/media/video/tuner-core.c
@@ -1,5 +1,5 @@
 /*
- * $Id: tuner-core.c,v 1.7 2005/05/30 02:02:47 mchehab Exp $
+ * $Id: tuner-core.c,v 1.15 2005/06/12 01:36:14 mchehab Exp $
  *
  * i2c tv tuner chip device driver
  * core core, i.e. kernel interfaces, registering and so on
@@ -26,15 +26,17 @@
 /*
  * comment line bellow to return to old behavor, where only one I2C device is supported
  */
-/* #define CONFIG_TUNER_MULTI_I2C */
+#define CONFIG_TUNER_MULTI_I2C /**/
 
 #define UNSET (-1U)
 
 /* standard i2c insmod options */
 static unsigned short normal_i2c[] = {
 	0x4b, /* tda8290 */
-	0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67,
-	0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f,
+	I2C_CLIENT_END
+};
+static unsigned short normal_i2c_range[] = {
+	0x60, 0x6f,
 	I2C_CLIENT_END
 };
 I2C_CLIENT_INSMOD;
@@ -59,7 +61,7 @@ MODULE_LICENSE("GPL");
 
 static int this_adap;
 #ifdef CONFIG_TUNER_MULTI_I2C
-static unsigned short tv_tuner, radio_tuner;
+static unsigned short first_tuner, tv_tuner, radio_tuner;
 #endif
 
 static struct i2c_driver driver;
@@ -67,7 +69,7 @@ static struct i2c_client client_template;
 
 /* ---------------------------------------------------------------------- */
 
-// Set tuner frequency,  freq in Units of 62.5kHz = 1/16MHz
+/* Set tuner frequency,  freq in Units of 62.5kHz = 1/16MHz */
 static void set_tv_freq(struct i2c_client *c, unsigned int freq)
 {
 	struct tuner *t = i2c_get_clientdata(c);
@@ -81,14 +83,26 @@ static void set_tv_freq(struct i2c_client *c, unsigned int freq)
 		return;
 	}
 	if (freq < tv_range[0]*16 || freq > tv_range[1]*16) {
-		/* FIXME: better do that chip-specific, but
-		   right now we don't have that in the config
-		   struct and this way is still better than no
-		   check at all */
-		tuner_info("TV freq (%d.%02d) out of range (%d-%d)\n",
-			   freq/16,freq%16*100/16,tv_range[0],tv_range[1]);
-		return;
+
+		if (freq >= tv_range[0]*16364 && freq <= tv_range[1]*16384) {
+			/* V4L2_TUNER_CAP_LOW frequency */
+
+			tuner_dbg("V4L2_TUNER_CAP_LOW freq selected for TV. Tuners yet doesn't support converting it to valid freq.\n");
+
+			t->tv_freq(c,freq>>10);
+
+			return;
+                } else {
+			/* FIXME: better do that chip-specific, but
+			   right now we don't have that in the config
+			   struct and this way is still better than no
+			   check at all */
+			tuner_info("TV freq (%d.%02d) out of range (%d-%d)\n",
+				   freq/16,freq%16*100/16,tv_range[0],tv_range[1]);
+			return;
+		}
 	}
+	tuner_dbg("62.5 Khz freq step selected for TV.\n");
 	t->tv_freq(c,freq);
 }
 
@@ -105,11 +119,29 @@ static void set_radio_freq(struct i2c_client *c, unsigned int freq)
 		return;
 	}
 	if (freq < radio_range[0]*16 || freq > radio_range[1]*16) {
-		tuner_info("radio freq (%d.%02d) out of range (%d-%d)\n",
+		if (freq >= tv_range[0]*16364 && freq <= tv_range[1]*16384) {
+			/* V4L2_TUNER_CAP_LOW frequency */
+			if (t->type == TUNER_TEA5767) {
+				tuner_info("radio freq step 62.5Hz (%d.%06d)\n",(freq>>14),freq%(1<<14)*10000);
+				t->radio_freq(c,freq>>10);
+				return;
+			}
+
+			tuner_dbg("V4L2_TUNER_CAP_LOW freq selected for Radio. Tuners yet doesn't support converting it to valid freq.\n");
+
+			tuner_info("radio freq (%d.%06d)\n",(freq>>14),freq%(1<<14)*10000);
+
+			t->radio_freq(c,freq>>10);
+			return;
+
+                } else {
+			tuner_info("radio freq (%d.%02d) out of range (%d-%d)\n",
 			   freq/16,freq%16*100/16,
-			   radio_range[0],radio_range[1]);
-		return;
+				   radio_range[0],radio_range[1]);
+			return;
+		}
 	}
+	tuner_dbg("62.5 Khz freq step selected for Radio.\n");
 	t->radio_freq(c,freq);
 }
 
@@ -133,34 +165,13 @@ static void set_freq(struct i2c_client *c, unsigned long freq)
 	t->freq = freq;
 }
 
-#ifdef CONFIG_TUNER_MULTI_I2C
-static void set_addr(struct i2c_client *c, struct tuner_addr *tun_addr)
-{
-	struct tuner *t = i2c_get_clientdata(c);
-
-	switch (tun_addr->type) {
-	case V4L2_TUNER_RADIO:
- 		radio_tuner=tun_addr->addr;
-		tuner_dbg("radio tuner set to I2C address 0x%02x\n",radio_tuner<<1);
-
-		break;
-	default:
-		tv_tuner=tun_addr->addr;
-		tuner_dbg("TV tuner set to I2C address 0x%02x\n",tv_tuner<<1);
-		break;
-	}
-}
-#else
-#define set_addr(c,tun_addr) \
-		tuner_warn("It is recommended to enable CONFIG_TUNER_MULTI_I2C for this card.\n");
-#endif
-
 static void set_type(struct i2c_client *c, unsigned int type)
 {
 	struct tuner *t = i2c_get_clientdata(c);
 
+	tuner_dbg ("I2C addr 0x%02x with type %d\n",c->addr<<1,type);
 	/* sanity check */
-	if (type == UNSET  ||  type == TUNER_ABSENT)
+	if (type == UNSET || type == TUNER_ABSENT)
 		return;
 	if (type >= tuner_count)
 		return;
@@ -175,6 +186,7 @@ static void set_type(struct i2c_client *c, unsigned int type)
 		return;
 
 	t->initialized = 1;
+
 	t->type = type;
 	switch (t->type) {
 	case TUNER_MT2032:
@@ -189,6 +201,53 @@ static void set_type(struct i2c_client *c, unsigned int type)
 	}
 }
 
+#ifdef CONFIG_TUNER_MULTI_I2C
+#define CHECK_ADDR(tp,cmd,tun)	if (client->addr!=tp) { \
+			  return 0; } else \
+			  tuner_info ("Cmd %s accepted to "tun"\n",cmd);
+#define CHECK_MODE(cmd)	if (t->mode == V4L2_TUNER_RADIO) { \
+		 	CHECK_ADDR(radio_tuner,cmd,"radio") } else \
+			{ CHECK_ADDR(tv_tuner,cmd,"TV"); }
+#else
+#define CHECK_ADDR(tp,cmd,tun) tuner_info ("Cmd %s accepted to "tun"\n",cmd);
+#define CHECK_MODE(cmd) tuner_info ("Cmd %s accepted\n",cmd);
+#endif
+
+#ifdef CONFIG_TUNER_MULTI_I2C
+
+static void set_addr(struct i2c_client *c, struct tuner_addr *tun_addr)
+{
+	/* ADDR_UNSET defaults to first available tuner */
+	if ( tun_addr->addr == ADDR_UNSET ) {
+		if (first_tuner != c->addr)
+			return;
+		switch (tun_addr->v4l2_tuner) {
+		case V4L2_TUNER_RADIO:
+	 		radio_tuner=c->addr;
+			break;
+		default:
+			tv_tuner=c->addr;
+			break;
+		}
+	} else {
+		/* Sets tuner to its configured value */
+		switch (tun_addr->v4l2_tuner) {
+		case V4L2_TUNER_RADIO:
+ 			radio_tuner=tun_addr->addr;
+			if ( tun_addr->addr == c->addr ) set_type(c,tun_addr->type);
+			return;
+		default:
+			tv_tuner=tun_addr->addr;
+			if ( tun_addr->addr == c->addr ) set_type(c,tun_addr->type);
+			return;
+		}
+	}
+	set_type(c,tun_addr->type);
+}
+#else
+#define set_addr(c,tun_addr) set_type(c,(tun_addr)->type)
+#endif
+
 static char pal[] = "-";
 module_param_string(pal, pal, sizeof(pal), 0644);
 
@@ -233,6 +292,7 @@ static int tuner_attach(struct i2c_adapter *adap, int addr, int kind)
 #else
 	/* by default, first I2C card is both tv and radio tuner */
 	if (this_adap == 0) {
+		first_tuner = addr;
 		tv_tuner = addr;
 		radio_tuner = addr;
 	}
@@ -249,11 +309,12 @@ static int tuner_attach(struct i2c_adapter *adap, int addr, int kind)
         memcpy(&t->i2c,&client_template,sizeof(struct i2c_client));
 	i2c_set_clientdata(&t->i2c, t);
 	t->type       = UNSET;
-	t->radio_if2  = 10700*1000; // 10.7MHz - FM radio
+	t->radio_if2  = 10700*1000; /* 10.7MHz - FM radio */
 
         i2c_attach_client(&t->i2c);
 	tuner_info("chip found @ 0x%x (%s)\n",
 		   addr << 1, adap->name);
+
 	set_type(&t->i2c, t->type);
 	return 0;
 }
@@ -261,12 +322,14 @@ static int tuner_attach(struct i2c_adapter *adap, int addr, int kind)
 static int tuner_probe(struct i2c_adapter *adap)
 {
 	if (0 != addr) {
-		normal_i2c[0] = addr;
-		normal_i2c[1] = I2C_CLIENT_END;
+		normal_i2c[0]       = addr;
+		normal_i2c_range[0] = addr;
+		normal_i2c_range[1] = addr;
 	}
 	this_adap = 0;
 
 #ifdef CONFIG_TUNER_MULTI_I2C
+	first_tuner = 0;
 	tv_tuner = 0;
 	radio_tuner = 0;
 #endif
@@ -298,17 +361,6 @@ static int tuner_detach(struct i2c_client *client)
 			  tuner_info("ignore v4l1 call\n"); \
 		          return 0; }
 
-#ifdef CONFIG_TUNER_MULTI_I2C
-#define CHECK_ADDR(tp,cmd)	if (client->addr!=tp) { \
-			  tuner_info ("Cmd %s to addr 0x%02x rejected.\n",cmd,client->addr<<1); \
-			  return 0; }
-#define CHECK_MODE(cmd)	if (t->mode == V4L2_TUNER_RADIO) { \
-			  CHECK_ADDR(radio_tuner,cmd) } else { CHECK_ADDR(tv_tuner,cmd); }
-#else
-#define CHECK_ADDR(tp,cmd)
-#define CHECK_MODE(cmd)
-#endif
-
 static int
 tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 {
@@ -320,19 +372,19 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 	case TUNER_SET_TYPE:
 		set_type(client,*iarg);
 		break;
-	case TUNER_SET_ADDR:
+	case TUNER_SET_TYPE_ADDR:
 		set_addr(client,(struct tuner_addr *)arg);
 		break;
 	case AUDC_SET_RADIO:
-		CHECK_ADDR(radio_tuner,"AUDC_SET_RADIO");
+		t->mode = V4L2_TUNER_RADIO;
+		CHECK_ADDR(tv_tuner,"AUDC_SET_RADIO","TV");
 
 		if (V4L2_TUNER_RADIO != t->mode) {
 			set_tv_freq(client,400 * 16);
-			t->mode = V4L2_TUNER_RADIO;
 		}
 		break;
 	case AUDC_CONFIG_PINNACLE:
-		CHECK_ADDR(tv_tuner,"AUDC_CONFIG_PINNACLE");
+		CHECK_ADDR(tv_tuner,"AUDC_CONFIG_PINNACLE","TV");
 		switch (*iarg) {
 		case 2:
 			tuner_dbg("pinnacle pal\n");
@@ -360,9 +412,10 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 		};
 		struct video_channel *vc = arg;
 
-		CHECK_ADDR(tv_tuner,"VIDIOCSCHAN");
 		CHECK_V4L2;
 		t->mode = V4L2_TUNER_ANALOG_TV;
+		CHECK_ADDR(tv_tuner,"VIDIOCSCHAN","TV");
+
 		if (vc->norm < ARRAY_SIZE(map))
 			t->std = map[vc->norm];
 		tuner_fixup_std(t);
@@ -383,17 +436,27 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 	{
 		struct video_tuner *vt = arg;
 
-		CHECK_ADDR(radio_tuner,"VIDIOCGTUNER:");
+		CHECK_ADDR(radio_tuner,"VIDIOCGTUNER","radio");
 		CHECK_V4L2;
-		if (V4L2_TUNER_RADIO == t->mode  &&  t->has_signal)
-			vt->signal = t->has_signal(client);
+		if (V4L2_TUNER_RADIO == t->mode) {
+			if (t->has_signal)
+				vt->signal = t->has_signal(client);
+			if (t->is_stereo) {
+				if (t->is_stereo(client))
+					vt-> flags |= VIDEO_TUNER_STEREO_ON;
+				else
+					vt-> flags &= 0xffff ^ VIDEO_TUNER_STEREO_ON;
+			}
+			vt->flags |= V4L2_TUNER_CAP_LOW; /* Allow freqs at 62.5 Hz */
+		}
+
 		return 0;
 	}
 	case VIDIOCGAUDIO:
 	{
 		struct video_audio *va = arg;
 
-		CHECK_ADDR(radio_tuner,"VIDIOCGAUDIO");
+		CHECK_ADDR(radio_tuner,"VIDIOCGAUDIO","radio");
 		CHECK_V4L2;
 		if (V4L2_TUNER_RADIO == t->mode  &&  t->is_stereo)
 			va->mode = t->is_stereo(client)
@@ -406,9 +469,10 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 	{
 		v4l2_std_id *id = arg;
 
-		CHECK_ADDR(tv_tuner,"VIDIOC_S_STD");
 		SWITCH_V4L2;
 		t->mode = V4L2_TUNER_ANALOG_TV;
+		CHECK_ADDR(tv_tuner,"VIDIOC_S_STD","TV");
+
 		t->std = *id;
 		tuner_fixup_std(t);
 		if (t->freq)
@@ -444,13 +508,27 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 
 		CHECK_MODE("VIDIOC_G_TUNER");
 		SWITCH_V4L2;
-		if (V4L2_TUNER_RADIO == t->mode  &&  t->has_signal)
-			tuner->signal = t->has_signal(client);
+		if (V4L2_TUNER_RADIO == t->mode) {
+			if (t->has_signal)
+				tuner -> signal = t->has_signal(client);
+			if (t->is_stereo) {
+				if (t->is_stereo(client)) {
+					tuner -> capability |= V4L2_TUNER_CAP_STEREO;
+					tuner -> rxsubchans |= V4L2_TUNER_SUB_STEREO;
+				} else {
+					tuner -> rxsubchans &= 0xffff ^ V4L2_TUNER_SUB_STEREO;
+				}
+			}
+		}
+		/* Wow to deal with V4L2_TUNER_CAP_LOW ? For now, it accepts from low at 62.5KHz step  to high at 62.5 Hz */
 		tuner->rangelow = tv_range[0] * 16;
-		tuner->rangehigh = tv_range[1] * 16;
+//		tuner->rangehigh = tv_range[1] * 16;
+//		tuner->rangelow = tv_range[0] * 16384;
+		tuner->rangehigh = tv_range[1] * 16384;
 		break;
 	}
 	default:
+		tuner_dbg ("Unimplemented IOCTL 0x%08x called to tuner.\n", cmd);
 		/* nothing */
 		break;
 	}
@@ -458,7 +536,7 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 	return 0;
 }
 
-static int tuner_suspend(struct device * dev, pm_message_t state, u32 level)
+static int tuner_suspend(struct device * dev, u32 state, u32 level)
 {
 	struct i2c_client *c = container_of(dev, struct i2c_client, dev);
 	struct tuner *t = i2c_get_clientdata(c);
diff --git a/drivers/media/video/tuner-simple.c b/drivers/media/video/tuner-simple.c
index 866f18dc5b5..539f3055731 100644
--- a/drivers/media/video/tuner-simple.c
+++ b/drivers/media/video/tuner-simple.c
@@ -1,5 +1,5 @@
 /*
- * $Id: tuner-simple.c,v 1.14 2005/05/30 02:02:47 mchehab Exp $
+ * $Id: tuner-simple.c,v 1.21 2005/06/10 19:53:26 nsh Exp $
  *
  * i2c tv tuner chip device driver
  * controls all those simple 4-control-bytes style tuners.
@@ -220,7 +220,17 @@ static struct tunertype tuners[] = {
 
 	{ "Thomson DDT 7611 (ATSC/NTSC)", THOMSON, ATSC,
 	  16*157.25,16*454.00,0x39,0x3a,0x3c,0x8e,732},
+	{ "Tena TNF9533-D/IF", LGINNOTEK, PAL,
+	  16*160.25, 16*464.25, 0x01,0x02,0x08,0x8e,623},
+
+	/*
+	 * This entry is for TEA5767 FM radio only chip used on several boards
+	 * w/TV tuner
+	 */
+	{ TEA5767_TUNER_NAME, Philips, RADIO,
+	  -1, -1, 0, 0, 0, TEA5767_LOW_LO_32768,0},
 };
+
 unsigned const int tuner_count = ARRAY_SIZE(tuners);
 
 /* ---------------------------------------------------------------------- */
@@ -231,6 +241,7 @@ static int tuner_getstatus(struct i2c_client *c)
 
 	if (1 != i2c_master_recv(c,&byte,1))
 		return 0;
+
 	return byte;
 }
 
@@ -239,17 +250,33 @@ static int tuner_getstatus(struct i2c_client *c)
 #define TUNER_MODE      0x38
 #define TUNER_AFC       0x07
 
-#define TUNER_STEREO    0x10 /* radio mode */
-#define TUNER_SIGNAL    0x07 /* radio mode */
+#define TUNER_STEREO       0x10 /* radio mode */
+#define TUNER_STEREO_MK3   0x04 /* radio mode */
+#define TUNER_SIGNAL       0x07 /* radio mode */
 
 static int tuner_signal(struct i2c_client *c)
 {
-	return (tuner_getstatus(c) & TUNER_SIGNAL)<<13;
+	return (tuner_getstatus(c) & TUNER_SIGNAL) << 13;
 }
 
 static int tuner_stereo(struct i2c_client *c)
 {
-	return (tuner_getstatus (c) & TUNER_STEREO);
+	int stereo, status;
+	struct tuner *t = i2c_get_clientdata(c);
+
+	status = tuner_getstatus (c);
+
+	switch (t->type) {
+        	case TUNER_PHILIPS_FM1216ME_MK3:
+    		case TUNER_PHILIPS_FM1236_MK3:
+		case TUNER_PHILIPS_FM1256_IH3:
+			stereo = ((status & TUNER_SIGNAL) == TUNER_STEREO_MK3);
+			break;
+		default:
+			stereo = status & TUNER_STEREO;
+	}
+
+	return stereo;
 }
 
 #if 0 /* unused */
@@ -432,6 +459,7 @@ static void default_set_radio_freq(struct i2c_client *c, unsigned int freq)
 	buffer[2] = tun->config;
 
 	switch (t->type) {
+	case TUNER_TENA_9533_DI:
 	case TUNER_YMEC_TVF_5533MF:
 
 		/*These values are empirically determinated */
@@ -473,20 +501,6 @@ int default_tuner_init(struct i2c_client *c)
 		   t->type, tuners[t->type].name);
 	strlcpy(c->name, tuners[t->type].name, sizeof(c->name));
 
-	switch (t->type) {
-	case TUNER_YMEC_TVF_5533MF:
-		{
-			struct tuner_addr tun_addr = { V4L2_TUNER_ANALOG_TV, 0xc2>>1 };
-
-			if (c->driver->command) {
-				c->driver->command(c, TUNER_SET_ADDR, &tun_addr);
-			} else {
-				tuner_warn("Couldn't set TV tuner I2C address to 0x%02x\n",tun_addr.addr<<1);
-			}
-			break;
-		}
-	}
-
 	t->tv_freq    = default_set_tv_freq;
 	t->radio_freq = default_set_radio_freq;
 	t->has_signal = tuner_signal;
diff --git a/drivers/media/video/tvaudio.c b/drivers/media/video/tvaudio.c
index 41b635e0d3c..6f5828a9b80 100644
--- a/drivers/media/video/tvaudio.c
+++ b/drivers/media/video/tvaudio.c
@@ -148,6 +148,7 @@ static unsigned short normal_i2c[] = {
 	I2C_TDA9874   >> 1,
 	I2C_PIC16C54  >> 1,
 	I2C_CLIENT_END };
+static unsigned short normal_i2c_range[2] = { I2C_CLIENT_END, I2C_CLIENT_END };
 I2C_CLIENT_INSMOD;
 
 static struct i2c_driver driver;
@@ -285,7 +286,6 @@ static int chip_thread(void *data)
 			schedule();
 		}
 		remove_wait_queue(&chip->wq, &wait);
-		try_to_freeze(PF_FREEZE);
 		if (chip->done || signal_pending(current))
 			break;
 		dprintk("%s: thread wakeup\n", i2c_clientname(&chip->c));
diff --git a/drivers/media/video/tvmixer.c b/drivers/media/video/tvmixer.c
index eafd7061b31..51b99cdbf29 100644
--- a/drivers/media/video/tvmixer.c
+++ b/drivers/media/video/tvmixer.c
@@ -1,3 +1,7 @@
+/*
+ * $Id: tvmixer.c,v 1.8 2005/06/12 04:19:19 mchehab Exp $
+ */
+
 #include <linux/module.h>
 #include <linux/moduleparam.h>
 #include <linux/kernel.h>
diff --git a/include/media/audiochip.h b/include/media/audiochip.h
index d3e9e30608d..f345a61c3bd 100644
--- a/include/media/audiochip.h
+++ b/include/media/audiochip.h
@@ -1,3 +1,7 @@
+/*
+ * $Id: audiochip.h,v 1.3 2005/06/12 04:19:19 mchehab Exp $
+ */
+
 #ifndef AUDIOCHIP_H
 #define AUDIOCHIP_H
 
diff --git a/include/media/id.h b/include/media/id.h
index 1b0320dc8f7..a39a6423914 100644
--- a/include/media/id.h
+++ b/include/media/id.h
@@ -1,3 +1,7 @@
+/*
+ * $Id: id.h,v 1.4 2005/06/12 04:19:19 mchehab Exp $
+ */
+
 /* FIXME: this temporarely, until these are included in linux/i2c-id.h */
 
 /* drivers */
diff --git a/include/media/ir-common.h b/include/media/ir-common.h
index b5fa6c585e2..698670547f1 100644
--- a/include/media/ir-common.h
+++ b/include/media/ir-common.h
@@ -1,5 +1,5 @@
 /*
- * $Id: ir-common.h,v 1.8 2005/02/22 12:28:40 kraxel Exp $
+ * $Id: ir-common.h,v 1.9 2005/05/15 19:01:26 mchehab Exp $
  *
  * some common structs and functions to handle infrared remotes via
  * input layer ...
diff --git a/include/media/tuner.h b/include/media/tuner.h
index cb75f4809c4..2dd8310901e 100644
--- a/include/media/tuner.h
+++ b/include/media/tuner.h
@@ -25,6 +25,8 @@
 
 #include "id.h"
 
+#define ADDR_UNSET (255)
+
 #define TUNER_TEMIC_PAL     0        /* 4002 FH5 (3X 7756, 9483) */
 #define TUNER_PHILIPS_PAL_I 1
 #define TUNER_PHILIPS_NTSC  2
@@ -100,6 +102,11 @@
 
 #define TUNER_YMEC_TVF_8531MF 58
 #define TUNER_YMEC_TVF_5533MF 59	/* Pixelview Pro Ultra NTSC */
+#define TUNER_THOMSON_DTT7611 60
+#define TUNER_TENA_9533_DI    61
+#define TUNER_TEA5767         62	/* Only FM Radio Tuner */
+
+#define TEA5767_TUNER_NAME "Philips TEA5767HN FM Radio"
 
 #define TUNER_THOMSON_DTT7611    60
 
@@ -109,6 +116,7 @@
 #define NTSC    3
 #define SECAM   4
 #define ATSC    5
+#define RADIO	6
 
 #define NoTuner 0
 #define Philips 1
@@ -124,9 +132,17 @@
 #define TCL     11
 #define THOMSON 12
 
+enum v4l_radio_tuner {
+        TEA5767_LOW_LO_32768    = 0,
+        TEA5767_HIGH_LO_32768   = 1,
+        TEA5767_LOW_LO_13MHz    = 2,
+        TEA5767_HIGH_LO_13MHz   = 3,
+};
+
+
 #define TUNER_SET_TYPE               _IOW('t',1,int)    /* set tuner type */
 #define TUNER_SET_TVFREQ             _IOW('t',2,int)    /* set tv freq */
-#define TUNER_SET_ADDR               _IOW('T',3,int)	/* Chooses tuner I2C address */
+#define TUNER_SET_TYPE_ADDR          _IOW('T',3,int)	/* set tuner type and I2C addr */
 
 #define  TDA9887_SET_CONFIG          _IOW('t',5,int)
 
@@ -151,8 +167,9 @@
 #define I2C_ADDR_TDA8275        0x61
 
 struct tuner_addr {
-	enum v4l2_tuner_type type;
-	unsigned short addr;
+	enum v4l2_tuner_type	v4l2_tuner;
+	unsigned int		type;
+	unsigned short		addr;
 };
 
 struct tuner {
-- 
cgit v1.2.3-70-g09d2


From ac19ecc6fa57b0ea320f01831175ff163f47d6a2 Mon Sep 17 00:00:00 2001
From: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
Date: Thu, 23 Jun 2005 22:05:09 -0700
Subject: [PATCH] v4l: update for SAA7134 cards

This patch adds support for various SAA7134 cards and brings some fixes.

Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
Signed-off-by: Fabrice Aeschbacher <fabrice.aeschbacher@laposte.net>
Signed-off-by: Hermann Pitton <hermann.pitton@onlinehome.de>.
Signed-off-by: Nickolay V Shmyrev <nshmyrev@yandex.ru>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/media/video/saa7134/saa6752hs.c       |  21 ++--
 drivers/media/video/saa7134/saa7134-cards.c   | 163 +++++++++++++++++++++++---
 drivers/media/video/saa7134/saa7134-core.c    |   2 +-
 drivers/media/video/saa7134/saa7134-dvb.c     |   2 +-
 drivers/media/video/saa7134/saa7134-empress.c |   2 +-
 drivers/media/video/saa7134/saa7134-i2c.c     |   2 +-
 drivers/media/video/saa7134/saa7134-input.c   | 112 +++++++++++++++++-
 drivers/media/video/saa7134/saa7134-oss.c     |  16 ++-
 drivers/media/video/saa7134/saa7134-tvaudio.c |  21 ++--
 drivers/media/video/saa7134/saa7134-vbi.c     |   2 +-
 drivers/media/video/saa7134/saa7134-video.c   |  61 ++++------
 drivers/media/video/saa7134/saa7134.h         |   8 +-
 drivers/media/video/tveeprom.c                |   3 +-
 drivers/media/video/v4l1-compat.c             |  10 +-
 include/media/tveeprom.h                      |   4 +
 15 files changed, 333 insertions(+), 96 deletions(-)

(limited to 'include')

diff --git a/drivers/media/video/saa7134/saa6752hs.c b/drivers/media/video/saa7134/saa6752hs.c
index 42c2b565c9f..d14158b111b 100644
--- a/drivers/media/video/saa7134/saa6752hs.c
+++ b/drivers/media/video/saa7134/saa6752hs.c
@@ -22,6 +22,7 @@
 
 /* Addresses to scan */
 static unsigned short normal_i2c[] = {0x20, I2C_CLIENT_END};
+static unsigned short normal_i2c_range[] = {I2C_CLIENT_END};
 I2C_CLIENT_INSMOD;
 
 MODULE_DESCRIPTION("device driver for saa6752hs MPEG2 encoder");
@@ -41,16 +42,16 @@ enum saa6752hs_videoformat {
 
 static const struct v4l2_format v4l2_format_table[] =
 {
-	[SAA6752HS_VF_D1] = {
-		.fmt = { .pix = { .width = 720, .height = 576 }, }, },
-	[SAA6752HS_VF_2_3_D1] = {
-		.fmt = { .pix = { .width = 480, .height = 576 }, }, },
-	[SAA6752HS_VF_1_2_D1] = {
-		.fmt = { .pix = { .width = 352, .height = 576 }, }, },
-	[SAA6752HS_VF_SIF] = {
-		.fmt = { .pix = { .width = 352, .height = 288 }, }, },
-	[SAA6752HS_VF_UNKNOWN] = {
-		.fmt = { .pix = { .width = 0, .height = 0 }, }, },
+	[SAA6752HS_VF_D1] =
+		{ .fmt = { .pix = { .width = 720, .height = 576 }}},
+	[SAA6752HS_VF_2_3_D1] =
+		{ .fmt = { .pix = { .width = 480, .height = 576 }}},
+	[SAA6752HS_VF_1_2_D1] =
+		{ .fmt = { .pix = { .width = 352, .height = 576 }}},
+	[SAA6752HS_VF_SIF] =
+		{ .fmt = { .pix = { .width = 352, .height = 288 }}},
+	[SAA6752HS_VF_UNKNOWN] =
+		{ .fmt = { .pix = { .width = 0, .height = 0}}},
 };
 
 struct saa6752hs_state {
diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c
index c51eb7f078d..0c781e24c44 100644
--- a/drivers/media/video/saa7134/saa7134-cards.c
+++ b/drivers/media/video/saa7134/saa7134-cards.c
@@ -1,6 +1,6 @@
 
 /*
- * $Id: saa7134-cards.c,v 1.54 2005/03/07 12:01:51 kraxel Exp $
+ * $Id: saa7134-cards.c,v 1.58 2005/06/07 18:05:00 nsh Exp $
  *
  * device driver for philips saa7134 based TV cards
  * card-specific stuff.
@@ -165,7 +165,7 @@ struct saa7134_board saa7134_boards[] = {
 		.inputs         = {{
 			.name = name_tv,
 			.vmux = 1,
-			.amux = LINE2,
+			.amux = TV,
 			.tv   = 1,
 		},{
 			.name = name_comp1,
@@ -878,7 +878,7 @@ struct saa7134_board saa7134_boards[] = {
         },
 	[SAA7134_BOARD_MANLI_MTV002] = {
 		/* Ognjen Nastic <ognjen@logosoft.ba> */
-		.name           = "Manli MuchTV M-TV002",
+		.name           = "Manli MuchTV M-TV002/Behold TV 403 FM",
 		.audio_clock    = 0x00200000,
 		.tuner_type     = TUNER_PHILIPS_PAL,
 		.inputs         = {{
@@ -899,14 +899,10 @@ struct saa7134_board saa7134_boards[] = {
 			.name = name_radio,
 			.amux = LINE2,
 		},
-		.mute = {
-			.name = name_mute,
-                        .amux = LINE1,
-		},
 	},
 	[SAA7134_BOARD_MANLI_MTV001] = {
 		/* Ognjen Nastic <ognjen@logosoft.ba> UNTESTED */
-		.name           = "Manli MuchTV M-TV001",
+		.name           = "Manli MuchTV M-TV001/Behold TV 401",
 		.audio_clock    = 0x00200000,
 		.tuner_type     = TUNER_PHILIPS_PAL,
 		.inputs         = {{
@@ -923,6 +919,10 @@ struct saa7134_board saa7134_boards[] = {
 			.amux = LINE2,
 			.tv   = 1,
 		}},
+		.mute = {
+			.name = name_mute,
+                        .amux = LINE1,
+		},
         },
 	[SAA7134_BOARD_TG3000TV] = {
 		/* TransGear 3000TV */
@@ -1078,7 +1078,6 @@ struct saa7134_board saa7134_boards[] = {
 		.audio_clock    = 0x00187de7,
 		.tuner_type     = TUNER_PHILIPS_FM1256_IH3,
 		.tda9887_conf   = TDA9887_PRESENT,
-		.gpiomask = 0x3,
 		.inputs         = {{
 			.name = name_tv,
 			.vmux = 1,
@@ -1285,7 +1284,7 @@ struct saa7134_board saa7134_boards[] = {
 			 .gpio =0x8000,
 		 }
 	},
-        [SAA7134_BOARD_AVERMEDIA_307] = {
+        [SAA7134_BOARD_AVERMEDIA_STUDIO_307] = {
 		/*
 		Nickolay V. Shmyrev <nshmyrev@yandex.ru>
 		Lots of thanks to Andrey Zolotarev <zolotarev_andrey@mail.ru>
@@ -1323,6 +1322,35 @@ struct saa7134_board saa7134_boards[] = {
 			.gpio = 0x01,
 		},
         },
+        [SAA7134_BOARD_AVERMEDIA_GO_007_FM] = {
+		.name           = "Avermedia AVerTV GO 007 FM",
+		.audio_clock    = 0x00187de7,
+		.tuner_type     = TUNER_PHILIPS_TDA8290,
+		.gpiomask       = 0x00300003,
+//		.gpiomask       = 0x8c240003,
+		.inputs         = {{
+			.name = name_tv,
+			.vmux = 1,
+			.amux = TV,
+			.tv   = 1,
+			.gpio = 0x01,
+		},{
+			.name = name_comp1,
+			.vmux = 0,
+			.amux = LINE2,
+			.gpio = 0x02,
+		},{
+			.name = name_svideo,
+			.vmux = 6,
+			.amux = LINE2,
+			.gpio = 0x02,
+		}},
+		.radio = {
+			.name = name_radio,
+			.amux = LINE1,
+			.gpio = 0x00300001,
+		},
+        },
 	[SAA7134_BOARD_AVERMEDIA_CARDBUS] = {
 		/* Jon Westgate <oryn@oryn.fsck.tv> */
 		.name           = "AVerMedia Cardbus TV/Radio",
@@ -1492,7 +1520,6 @@ struct saa7134_board saa7134_boards[] = {
 		.audio_clock    = 0x00187de7,
 		.tuner_type     = TUNER_PHILIPS_FQ1216ME,
 		.tda9887_conf   = TDA9887_PRESENT,
-		.gpiomask = 0x3,
 		.inputs         = {{
 			.name = name_tv,
 			.vmux = 1,
@@ -1546,7 +1573,82 @@ struct saa7134_board saa7134_boards[] = {
 //			.gpio = 0x4000,
 		}},
 	},
-};
+        [SAA7134_BOARD_AVERMEDIA_307] = {
+               /*
+                 Davydov Vladimir <vladimir@iqmedia.com>
+               */
+               .name           = "Avermedia AVerTV 307",
+               .audio_clock    = 0x00187de7,
+               .tuner_type     = TUNER_PHILIPS_FQ1216ME,
+               .tda9887_conf   = TDA9887_PRESENT,
+               .inputs         = {{
+                       .name = name_tv,
+                       .vmux = 1,
+                       .amux = TV,
+                       .tv   = 1,
+               },{
+                       .name = name_comp1,
+                       .vmux = 0,
+                       .amux = LINE1,
+               },{
+                       .name = name_comp2,
+                       .vmux = 3,
+                       .amux = LINE1,
+               },{
+                       .name = name_svideo,
+                       .vmux = 8,
+                       .amux = LINE1,
+               }},
+        },
+	[SAA7134_BOARD_ADS_INSTANT_TV] = {
+                .name           = "ADS Tech Instant TV (saa7135)",
+		.audio_clock    = 0x00187de7,
+                .tuner_type     = TUNER_PHILIPS_TDA8290,
+                .inputs         = {{
+                        .name = name_tv,
+                        .vmux = 1,
+                        .amux = TV,
+                        .tv   = 1,
+                },{
+                        .name = name_comp1,
+                        .vmux = 3,
+                        .amux = LINE2,
+                },{
+                        .name = name_svideo,
+                        .vmux = 8,
+                        .amux = LINE2,
+                }},
+	},
+ 	[SAA7134_BOARD_KWORLD_VSTREAM_XPERT] = {
+ 		.name           = "Kworld/Tevion V-Stream Xpert TV PVR7134",
+ 		.audio_clock    = 0x00187de7,
+ 		.tuner_type     = TUNER_PHILIPS_PAL_I,
+ 		.gpiomask	= 0x0700,
+ 		.inputs = {{
+ 			.name   = name_tv,
+ 			.vmux   = 1,
+ 			.amux   = TV,
+ 			.tv     = 1,
+ 			.gpio   = 0x000,
+ 		},{
+ 			.name   = name_comp1,
+ 			.vmux   = 3,
+ 			.amux   = LINE1,
+ 			.gpio   = 0x200,		//gpio by DScaler
+ 		},{
+ 			.name   = name_svideo,
+ 			.vmux   = 0,
+ 			.amux   = LINE1,
+ 			.gpio   = 0x200,
+ 		}},
+ 		.radio = {
+ 			.name   = name_radio,
+ 			.amux   = LINE1,
+ 			.gpio   = 0x100,
+ 		},
+ 	},
+ };
+
 const unsigned int saa7134_bcount = ARRAY_SIZE(saa7134_boards);
 
 /* ------------------------------------------------------------------ */
@@ -1663,7 +1765,7 @@ struct pci_device_id saa7134_pci_tbl[] = {
                 .driver_data  = SAA7134_BOARD_ASUSTeK_TVFM7134,
 	},{
                 .vendor       = PCI_VENDOR_ID_PHILIPS,
-                .device       = PCI_DEVICE_ID_PHILIPS_SAA7133,
+                .device       = PCI_DEVICE_ID_PHILIPS_SAA7135,
                 .subvendor    = PCI_VENDOR_ID_ASUSTEK,
                 .subdevice    = 0x4845,
                 .driver_data  = SAA7135_BOARD_ASUSTeK_TVFM7135,
@@ -1824,6 +1926,12 @@ struct pci_device_id saa7134_pci_tbl[] = {
 		.device       = PCI_DEVICE_ID_PHILIPS_SAA7134,
 		.subvendor    = 0x1461, /* Avermedia Technologies Inc */
 		.subdevice    = 0x9715,
+		.driver_data  = SAA7134_BOARD_AVERMEDIA_STUDIO_307,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7134,
+		.subvendor    = 0x1461, /* Avermedia Technologies Inc */
+		.subdevice    = 0xa70a,
 		.driver_data  = SAA7134_BOARD_AVERMEDIA_307,
 	},{
 		.vendor       = PCI_VENDOR_ID_PHILIPS,
@@ -1844,6 +1952,26 @@ struct pci_device_id saa7134_pci_tbl[] = {
 		.subvendor    = 0x5168,
 		.subdevice    = 0x0306,
 		.driver_data  = SAA7134_BOARD_FLYDVBTDUO,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7133,
+		.subvendor    = 0x1461, /* Avermedia Technologies Inc */
+		.subdevice    = 0xf31f,
+		.driver_data  = SAA7134_BOARD_AVERMEDIA_GO_007_FM,
+
+ 	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7135,
+		.subvendor    = 0x1421,
+		.subdevice    = 0x0350,		/* PCI version */
+		.driver_data  = SAA7134_BOARD_ADS_INSTANT_TV,
+
+ 	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7135,
+		.subvendor    = 0x1421,
+		.subdevice    = 0x0370,		/* cardbus version */
+		.driver_data  = SAA7134_BOARD_ADS_INSTANT_TV,
 
  	},{
 		/* --- boards without eeprom + subsystem ID --- */
@@ -1954,20 +2082,23 @@ int saa7134_board_init1(struct saa7134_dev *dev)
 		dev->has_remote = 1;
 		board_flyvideo(dev);
 		break;
-	case SAA7134_BOARD_FLYTVPLATINUM_FM:
+        case SAA7134_BOARD_FLYTVPLATINUM_FM:
 	case SAA7134_BOARD_CINERGY400:
 	case SAA7134_BOARD_CINERGY600:
 	case SAA7134_BOARD_CINERGY600_MK3:
 	case SAA7134_BOARD_ECS_TVP3XP:
 	case SAA7134_BOARD_ECS_TVP3XP_4CB5:
 	case SAA7134_BOARD_MD2819:
+	case SAA7134_BOARD_KWORLD_VSTREAM_XPERT:
 	case SAA7134_BOARD_AVERMEDIA_STUDIO_305:
 	case SAA7134_BOARD_AVERMEDIA_305:
+	case SAA7134_BOARD_AVERMEDIA_STUDIO_307:
 	case SAA7134_BOARD_AVERMEDIA_307:
+	case SAA7134_BOARD_AVERMEDIA_GO_007_FM:
 //	case SAA7134_BOARD_SABRENT_SBTTVFM:  /* not finished yet */
 	case SAA7134_BOARD_VIDEOMATE_TV_PVR:
-		dev->has_remote = 1;
-		break;
+ 	case SAA7134_BOARD_MANLI_MTV001:
+ 	case SAA7134_BOARD_MANLI_MTV002:
 	case SAA7134_BOARD_AVACSSMARTTV:
 		dev->has_remote = 1;
 		break;
diff --git a/drivers/media/video/saa7134/saa7134-core.c b/drivers/media/video/saa7134/saa7134-core.c
index 634a2d25f2f..f61ed1849a2 100644
--- a/drivers/media/video/saa7134/saa7134-core.c
+++ b/drivers/media/video/saa7134/saa7134-core.c
@@ -1,5 +1,5 @@
 /*
- * $Id: saa7134-core.c,v 1.28 2005/02/22 09:56:29 kraxel Exp $
+ * $Id: saa7134-core.c,v 1.30 2005/05/22 19:23:39 nsh Exp $
  *
  * device driver for philips saa7134 based TV cards
  * driver core
diff --git a/drivers/media/video/saa7134/saa7134-dvb.c b/drivers/media/video/saa7134/saa7134-dvb.c
index c2873ae029f..aa8e2cf62d5 100644
--- a/drivers/media/video/saa7134/saa7134-dvb.c
+++ b/drivers/media/video/saa7134/saa7134-dvb.c
@@ -1,5 +1,5 @@
 /*
- * $Id: saa7134-dvb.c,v 1.12 2005/02/18 12:28:29 kraxel Exp $
+ * $Id: saa7134-dvb.c,v 1.13 2005/06/12 04:19:19 mchehab Exp $
  *
  * (c) 2004 Gerd Knorr <kraxel@bytesex.org> [SuSE Labs]
  *
diff --git a/drivers/media/video/saa7134/saa7134-empress.c b/drivers/media/video/saa7134/saa7134-empress.c
index fa135733690..c85348d0239 100644
--- a/drivers/media/video/saa7134/saa7134-empress.c
+++ b/drivers/media/video/saa7134/saa7134-empress.c
@@ -1,5 +1,5 @@
 /*
- * $Id: saa7134-empress.c,v 1.10 2005/02/03 10:24:33 kraxel Exp $
+ * $Id: saa7134-empress.c,v 1.11 2005/05/22 19:23:39 nsh Exp $
  *
  * (c) 2004 Gerd Knorr <kraxel@bytesex.org> [SuSE Labs]
  *
diff --git a/drivers/media/video/saa7134/saa7134-i2c.c b/drivers/media/video/saa7134/saa7134-i2c.c
index 702bb63d981..b6f002e8421 100644
--- a/drivers/media/video/saa7134/saa7134-i2c.c
+++ b/drivers/media/video/saa7134/saa7134-i2c.c
@@ -1,5 +1,5 @@
 /*
- * $Id: saa7134-i2c.c,v 1.10 2005/01/24 17:37:23 kraxel Exp $
+ * $Id: saa7134-i2c.c,v 1.11 2005/06/12 01:36:14 mchehab Exp $
  *
  * device driver for philips saa7134 based TV cards
  * i2c interface support
diff --git a/drivers/media/video/saa7134/saa7134-input.c b/drivers/media/video/saa7134/saa7134-input.c
index ca50cf531f2..aba2b9de60d 100644
--- a/drivers/media/video/saa7134/saa7134-input.c
+++ b/drivers/media/video/saa7134/saa7134-input.c
@@ -1,5 +1,5 @@
 /*
- * $Id: saa7134-input.c,v 1.16 2004/12/10 12:33:39 kraxel Exp $
+ * $Id: saa7134-input.c,v 1.19 2005/06/07 18:02:26 nsh Exp $
  *
  * handle saa7134 IR remotes via linux kernel input layer.
  *
@@ -308,6 +308,102 @@ static IR_KEYTAB_TYPE videomate_tv_pvr_codes[IR_KEYTAB_SIZE] = {
 	[ 32 ] = KEY_LANGUAGE,
 	[ 33 ] = KEY_SLEEP,
 };
+
+/* Michael Tokarev <mjt@tls.msk.ru>
+   http://www.corpit.ru/mjt/beholdTV/remote_control.jpg
+   keytable is used by MANLI MTV00[12] and BeholdTV 40[13] at
+   least, and probably other cards too.
+   The "ascii-art picture" below (in comments, first row
+   is the keycode in hex, and subsequent row(s) shows
+   the button labels (several variants when appropriate)
+   helps to descide which keycodes to assign to the buttons.
+ */
+static IR_KEYTAB_TYPE manli_codes[IR_KEYTAB_SIZE] = {
+
+	/*  0x1c            0x12  *
+	 * FUNCTION         POWER *
+	 *   FM              (|)  *
+	 *                        */
+	[ 0x1c ] = KEY_RADIO,	/*XXX*/
+	[ 0x12 ] = KEY_POWER,
+
+	/*  0x01    0x02    0x03  *
+	 *   1       2       3    *
+	 *                        *
+	 *  0x04    0x05    0x06  *
+	 *   4       5       6    *
+	 *                        *
+	 *  0x07    0x08    0x09  *
+	 *   7       8       9    *
+	 *                        */
+	[ 0x01 ] = KEY_KP1,
+	[ 0x02 ] = KEY_KP2,
+	[ 0x03 ] = KEY_KP3,
+	[ 0x04 ] = KEY_KP4,
+	[ 0x05 ] = KEY_KP5,
+	[ 0x06 ] = KEY_KP6,
+	[ 0x07 ] = KEY_KP7,
+	[ 0x08 ] = KEY_KP8,
+	[ 0x09 ] = KEY_KP9,
+
+	/*  0x0a    0x00    0x17  *
+	 * RECALL    0      +100  *
+	 *                  PLUS  *
+	 *                        */
+	[ 0x0a ] = KEY_AGAIN,	/*XXX KEY_REWIND? */
+	[ 0x00 ] = KEY_KP0,
+	[ 0x17 ] = KEY_DIGITS,	/*XXX*/
+
+	/*  0x14            0x10  *
+	 *  MENU            INFO  *
+	 *  OSD                   */
+	[ 0x14 ] = KEY_MENU,
+	[ 0x10 ] = KEY_INFO,
+
+	/*          0x0b          *
+	 *           Up           *
+	 *                        *
+	 *  0x18    0x16    0x0c  *
+	 *  Left     Ok     Right *
+	 *                        *
+	 *         0x015          *
+	 *         Down           *
+	 *                        */
+	[ 0x0b ] = KEY_UP,	/*XXX KEY_SCROLLUP? */
+	[ 0x18 ] = KEY_LEFT,	/*XXX KEY_BACK? */
+	[ 0x16 ] = KEY_OK,	/*XXX KEY_SELECT? KEY_ENTER? */
+	[ 0x0c ] = KEY_RIGHT,	/*XXX KEY_FORWARD? */
+	[ 0x15 ] = KEY_DOWN,	/*XXX KEY_SCROLLDOWN? */
+
+	/*  0x11            0x0d  *
+	 *  TV/AV           MODE  *
+	 *  SOURCE         STEREO *
+	 *                        */
+	[ 0x11 ] = KEY_TV,	/*XXX*/
+	[ 0x0d ] = KEY_MODE,	/*XXX there's no KEY_STEREO */
+
+	/*  0x0f    0x1b    0x1a  *
+	 *  AUDIO   Vol+    Chan+ *
+	 *        TIMESHIFT???    *
+	 *                        *
+	 *  0x0e    0x1f    0x1e  *
+	 *  SLEEP   Vol-    Chan- *
+	 *                        */
+	[ 0x0f ] = KEY_AUDIO,
+	[ 0x1b ] = KEY_VOLUMEUP,
+	[ 0x1a ] = KEY_CHANNELUP,
+	[ 0x0e ] = KEY_SLEEP,	/*XXX maybe KEY_PAUSE */
+	[ 0x1f ] = KEY_VOLUMEDOWN,
+	[ 0x1e ] = KEY_CHANNELDOWN,
+
+	/*         0x13     0x19  *
+	 *         MUTE   SNAPSHOT*
+	 *                        */
+	[ 0x13 ] = KEY_MUTE,
+	[ 0x19 ] = KEY_RECORD,	/*XXX*/
+
+	// 0x1d unused ?
+};
 /* ---------------------------------------------------------------------- */
 
 static int build_key(struct saa7134_dev *dev)
@@ -379,7 +475,7 @@ int saa7134_input_init1(struct saa7134_dev *dev)
 	switch (dev->board) {
 	case SAA7134_BOARD_FLYVIDEO2000:
 	case SAA7134_BOARD_FLYVIDEO3000:
-	case SAA7134_BOARD_FLYTVPLATINUM_FM:
+        case SAA7134_BOARD_FLYTVPLATINUM_FM:
 		ir_codes     = flyvideo_codes;
 		mask_keycode = 0xEC00000;
 		mask_keydown = 0x0040000;
@@ -405,8 +501,12 @@ int saa7134_input_init1(struct saa7134_dev *dev)
 		polling      = 50; // ms
 		break;
 	case SAA7134_BOARD_MD2819:
+	case SAA7134_BOARD_KWORLD_VSTREAM_XPERT:
 	case SAA7134_BOARD_AVERMEDIA_305:
 	case SAA7134_BOARD_AVERMEDIA_307:
+	case SAA7134_BOARD_AVERMEDIA_STUDIO_305:
+	case SAA7134_BOARD_AVERMEDIA_STUDIO_307:
+	case SAA7134_BOARD_AVERMEDIA_GO_007_FM:
 		ir_codes     = md2819_codes;
 		mask_keycode = 0x0007C8;
 		mask_keydown = 0x000010;
@@ -415,6 +515,14 @@ int saa7134_input_init1(struct saa7134_dev *dev)
 		saa_setb(SAA7134_GPIO_GPMODE0, 0x4);
 		saa_setb(SAA7134_GPIO_GPSTATUS0, 0x4);
 		break;
+	case SAA7134_BOARD_MANLI_MTV001:
+	case SAA7134_BOARD_MANLI_MTV002:
+		ir_codes     = manli_codes;
+		mask_keycode = 0x001f00;
+		mask_keyup   = 0x004000;
+		mask_keydown = 0x002000;
+		polling      = 50; // ms
+		break;
 	case SAA7134_BOARD_VIDEOMATE_TV_PVR:
 		ir_codes     = videomate_tv_pvr_codes;
 		mask_keycode = 0x00003F;
diff --git a/drivers/media/video/saa7134/saa7134-oss.c b/drivers/media/video/saa7134/saa7134-oss.c
index 6b6a643bf1c..81732904623 100644
--- a/drivers/media/video/saa7134/saa7134-oss.c
+++ b/drivers/media/video/saa7134/saa7134-oss.c
@@ -1,5 +1,5 @@
 /*
- * $Id: saa7134-oss.c,v 1.13 2004/12/10 12:33:39 kraxel Exp $
+ * $Id: saa7134-oss.c,v 1.14 2005/05/18 22:45:16 hhackmann Exp $
  *
  * device driver for philips saa7134 based TV cards
  * oss dsp interface
@@ -49,7 +49,6 @@ MODULE_PARM_DESC(oss_rate,"sample rate (valid are: 32000,48000)");
 
 static int dsp_buffer_conf(struct saa7134_dev *dev, int blksize, int blocks)
 {
-	blksize &= ~0xff;
 	if (blksize < 0x100)
 		blksize = 0x100;
 	if (blksize > 0x10000)
@@ -57,8 +56,6 @@ static int dsp_buffer_conf(struct saa7134_dev *dev, int blksize, int blocks)
 
 	if (blocks < 2)
 		blocks = 2;
-        while ((blksize * blocks) & ~PAGE_MASK)
-		blocks++;
 	if ((blksize * blocks) > 1024*1024)
 		blocks = 1024*1024 / blksize;
 
@@ -79,7 +76,7 @@ static int dsp_buffer_init(struct saa7134_dev *dev)
 		BUG();
 	videobuf_dma_init(&dev->oss.dma);
 	err = videobuf_dma_init_kernel(&dev->oss.dma, PCI_DMA_FROMDEVICE,
-				       dev->oss.bufsize >> PAGE_SHIFT);
+				       (dev->oss.bufsize + PAGE_SIZE) >> PAGE_SHIFT);
 	if (0 != err)
 		return err;
 	return 0;
@@ -163,10 +160,11 @@ static int dsp_rec_start(struct saa7134_dev *dev)
 			fmt |= 0x04;
 		fmt |= (TV == dev->oss.input) ? 0xc0 : 0x80;
 
-		saa_writeb(SAA7134_NUM_SAMPLES0, (dev->oss.blksize & 0x0000ff));
-		saa_writeb(SAA7134_NUM_SAMPLES1, (dev->oss.blksize & 0x00ff00) >>  8);
-		saa_writeb(SAA7134_NUM_SAMPLES2, (dev->oss.blksize & 0xff0000) >> 16);
+		saa_writeb(SAA7134_NUM_SAMPLES0, ((dev->oss.blksize - 1) & 0x0000ff));
+		saa_writeb(SAA7134_NUM_SAMPLES1, ((dev->oss.blksize - 1) & 0x00ff00) >>  8);
+		saa_writeb(SAA7134_NUM_SAMPLES2, ((dev->oss.blksize - 1) & 0xff0000) >> 16);
 		saa_writeb(SAA7134_AUDIO_FORMAT_CTRL, fmt);
+
 		break;
 	case PCI_DEVICE_ID_PHILIPS_SAA7133:
 	case PCI_DEVICE_ID_PHILIPS_SAA7135:
@@ -817,7 +815,7 @@ void saa7134_irq_oss_done(struct saa7134_dev *dev, unsigned long status)
 			reg = SAA7134_RS_BA1(6);
 	} else {
 		/* even */
-		if (0 == (dev->oss.dma_blk & 0x00))
+		if (1 == (dev->oss.dma_blk & 0x01))
 			reg = SAA7134_RS_BA2(6);
 	}
 	if (0 == reg) {
diff --git a/drivers/media/video/saa7134/saa7134-tvaudio.c b/drivers/media/video/saa7134/saa7134-tvaudio.c
index ecac13c006d..3617e7f7a41 100644
--- a/drivers/media/video/saa7134/saa7134-tvaudio.c
+++ b/drivers/media/video/saa7134/saa7134-tvaudio.c
@@ -1,5 +1,5 @@
 /*
- * $Id: saa7134-tvaudio.c,v 1.22 2005/01/07 13:11:19 kraxel Exp $
+ * $Id: saa7134-tvaudio.c,v 1.25 2005/06/07 19:00:38 nsh Exp $
  *
  * device driver for philips saa7134 based TV cards
  * tv audio decoder (fm stereo, nicam, ...)
@@ -181,7 +181,8 @@ static void tvaudio_init(struct saa7134_dev *dev)
 	saa_writeb(SAA7134_AUDIO_CLOCK0,      clock        & 0xff);
 	saa_writeb(SAA7134_AUDIO_CLOCK1,     (clock >>  8) & 0xff);
 	saa_writeb(SAA7134_AUDIO_CLOCK2,     (clock >> 16) & 0xff);
-	saa_writeb(SAA7134_AUDIO_PLL_CTRL,   0x01);
+	// frame locked audio was reported not to be reliable
+	saa_writeb(SAA7134_AUDIO_PLL_CTRL,   0x02);
 
 	saa_writeb(SAA7134_NICAM_ERROR_LOW,  0x14);
 	saa_writeb(SAA7134_NICAM_ERROR_HIGH, 0x50);
@@ -250,6 +251,11 @@ static void mute_input_7134(struct saa7134_dev *dev)
 	saa_andorb(SAA7134_AUDIO_FORMAT_CTRL, 0xc0, ausel);
 	saa_andorb(SAA7134_ANALOG_IO_SELECT, 0x08, ics);
 	saa_andorb(SAA7134_ANALOG_IO_SELECT, 0x07, ocs);
+	// for oss, we need to change the clock configuration
+	if (in->amux == TV)
+		saa_andorb(SAA7134_SIF_SAMPLE_FREQ,   0x03, 0x00);
+	else
+		saa_andorb(SAA7134_SIF_SAMPLE_FREQ,   0x03, 0x01);
 
 	/* switch gpio-connected external audio mux */
 	if (0 == card(dev).gpiomask)
@@ -439,16 +445,15 @@ static int tvaudio_getstereo(struct saa7134_dev *dev, struct saa7134_tvaudio *au
 		nicam = saa_readb(SAA7134_NICAM_STATUS);
 		dprintk("getstereo: nicam=0x%x\n",nicam);
 		switch (nicam & 0x0b) {
+		case 0x08:
+			retval = V4L2_TUNER_SUB_MONO;
+			break;
 		case 0x09:
 			retval = V4L2_TUNER_SUB_LANG1 | V4L2_TUNER_SUB_LANG2;
 			break;
 		case 0x0a:
 			retval = V4L2_TUNER_SUB_MONO | V4L2_TUNER_SUB_STEREO;
 			break;
-		case 0x08:
-		default:
-			retval = V4L2_TUNER_SUB_MONO;
-			break;
 		}
 		break;
 	}
@@ -572,14 +577,14 @@ static int tvaudio_thread(void *data)
 		} else if (0 != dev->last_carrier) {
 			/* no carrier -- try last detected one as fallback */
 			carrier = dev->last_carrier;
-			printk(KERN_WARNING "%s/audio: audio carrier scan failed, "
+			dprintk(KERN_WARNING "%s/audio: audio carrier scan failed, "
 			       "using %d.%03d MHz [last detected]\n",
 			       dev->name, carrier/1000, carrier%1000);
 
 		} else {
 			/* no carrier + no fallback -- use default */
 			carrier = default_carrier;
-			printk(KERN_WARNING "%s/audio: audio carrier scan failed, "
+			dprintk(KERN_WARNING "%s/audio: audio carrier scan failed, "
 			       "using %d.%03d MHz [default]\n",
 			       dev->name, carrier/1000, carrier%1000);
 		}
diff --git a/drivers/media/video/saa7134/saa7134-vbi.c b/drivers/media/video/saa7134/saa7134-vbi.c
index 03c350ffb2d..3c33c591cc8 100644
--- a/drivers/media/video/saa7134/saa7134-vbi.c
+++ b/drivers/media/video/saa7134/saa7134-vbi.c
@@ -1,5 +1,5 @@
 /*
- * $Id: saa7134-vbi.c,v 1.6 2004/12/10 12:33:39 kraxel Exp $
+ * $Id: saa7134-vbi.c,v 1.7 2005/05/24 23:13:06 nsh Exp $
  *
  * device driver for philips saa7134 based TV cards
  * video4linux video interface
diff --git a/drivers/media/video/saa7134/saa7134-video.c b/drivers/media/video/saa7134/saa7134-video.c
index 72f86736a79..c0a2ee52053 100644
--- a/drivers/media/video/saa7134/saa7134-video.c
+++ b/drivers/media/video/saa7134/saa7134-video.c
@@ -1,5 +1,5 @@
 /*
- * $Id: saa7134-video.c,v 1.28 2005/02/15 15:59:35 kraxel Exp $
+ * $Id: saa7134-video.c,v 1.30 2005/06/07 19:00:38 nsh Exp $
  *
  * device driver for philips saa7134 based TV cards
  * video4linux video interface
@@ -31,8 +31,6 @@
 #include "saa7134-reg.h"
 #include "saa7134.h"
 
-#define V4L2_I2C_CLIENTS 1
-
 /* ------------------------------------------------------------------ */
 
 static unsigned int video_debug   = 0;
@@ -276,12 +274,12 @@ static struct saa7134_tvnorm tvnorms[] = {
 
 		.h_start       = 0,
 		.h_stop        = 719,
- 		.video_v_start = 23,
- 		.video_v_stop  = 262,
- 		.vbi_v_start_0 = 10,
- 		.vbi_v_stop_0  = 21,
- 		.vbi_v_start_1 = 273,
- 		.src_timing    = 7,
+		.video_v_start = 23,
+  		.video_v_stop  = 262,
+  		.vbi_v_start_0 = 10,
+  		.vbi_v_stop_0  = 21,
+  		.vbi_v_start_1 = 273,
+  		.src_timing    = 7,
 
 		.sync_control  = 0x18,
 		.luma_control  = 0x40,
@@ -524,22 +522,7 @@ static void set_tvnorm(struct saa7134_dev *dev, struct saa7134_tvnorm *norm)
 	saa_writeb(SAA7134_RAW_DATA_GAIN,         0x40);
 	saa_writeb(SAA7134_RAW_DATA_OFFSET,       0x80);
 
-#ifdef V4L2_I2C_CLIENTS
 	saa7134_i2c_call_clients(dev,VIDIOC_S_STD,&norm->id);
-#else
-	{
-		/* pass down info to the i2c chips (v4l1) */
-		struct video_channel c;
-		memset(&c,0,sizeof(c));
-		c.channel = dev->ctl_input;
-		c.norm = VIDEO_MODE_PAL;
-		if (norm->id & V4L2_STD_NTSC)
-			c.norm = VIDEO_MODE_NTSC;
-		if (norm->id & V4L2_STD_SECAM)
-			c.norm = VIDEO_MODE_SECAM;
-		saa7134_i2c_call_clients(dev,VIDIOCSCHAN,&c);
-	}
-#endif
 }
 
 static void video_mux(struct saa7134_dev *dev, int input)
@@ -1883,11 +1866,9 @@ static int video_do_ioctl(struct inode *inode, struct file *file,
 			return -EINVAL;
 		down(&dev->lock);
 		dev->ctl_freq = f->frequency;
-#ifdef V4L2_I2C_CLIENTS
+
 		saa7134_i2c_call_clients(dev,VIDIOC_S_FREQUENCY,f);
-#else
-		saa7134_i2c_call_clients(dev,VIDIOCSFREQ,&dev->ctl_freq);
-#endif
+
 		saa7134_tvaudio_do_scan(dev);
 		up(&dev->lock);
 		return 0;
@@ -2142,16 +2123,19 @@ static int radio_do_ioctl(struct inode *inode, struct file *file,
                 t->rangelow  = (int)(65*16);
                 t->rangehigh = (int)(108*16);
 
-#ifdef V4L2_I2C_CLIENTS
-		saa7134_i2c_call_clients(dev,VIDIOC_G_TUNER,t);
-#else
-		{
-			struct video_tuner vt;
-			memset(&vt,0,sizeof(vt));
-			saa7134_i2c_call_clients(dev,VIDIOCGTUNER,&vt);
-			t->signal = vt.signal;
-		}
-#endif
+		saa7134_i2c_call_clients(dev, VIDIOC_G_TUNER, t);
+
+		return 0;
+	}
+	case VIDIOC_S_TUNER:
+	{
+		struct v4l2_tuner *t = arg;
+
+		if (0 != t->index)
+			return -EINVAL;
+
+		saa7134_i2c_call_clients(dev,VIDIOC_S_TUNER,t);
+
 		return 0;
 	}
 	case VIDIOC_ENUMINPUT:
@@ -2185,7 +2169,6 @@ static int radio_do_ioctl(struct inode *inode, struct file *file,
 		return 0;
 	}
 	case VIDIOC_S_AUDIO:
-	case VIDIOC_S_TUNER:
 	case VIDIOC_S_INPUT:
 	case VIDIOC_S_STD:
 		return 0;
diff --git a/drivers/media/video/saa7134/saa7134.h b/drivers/media/video/saa7134/saa7134.h
index b808f18890b..d6b1c0d4d0f 100644
--- a/drivers/media/video/saa7134/saa7134.h
+++ b/drivers/media/video/saa7134/saa7134.h
@@ -1,5 +1,5 @@
 /*
- * $Id: saa7134.h,v 1.38 2005/03/07 12:01:51 kraxel Exp $
+ * $Id: saa7134.h,v 1.41 2005/06/07 18:02:26 nsh Exp $
  *
  * v4l2 device driver for philips saa7134 based TV cards
  *
@@ -168,7 +168,7 @@ struct saa7134_format {
 #define SAA7134_BOARD_SABRENT_SBTTVFM  42
 #define SAA7134_BOARD_ZOLID_XPERT_TV7134 43
 #define SAA7134_BOARD_EMPIRE_PCI_TV_RADIO_LE 44
-#define SAA7134_BOARD_AVERMEDIA_307    45
+#define SAA7134_BOARD_AVERMEDIA_STUDIO_307    45
 #define SAA7134_BOARD_AVERMEDIA_CARDBUS 46
 #define SAA7134_BOARD_CINERGY400_CARDBUS 47
 #define SAA7134_BOARD_CINERGY600_MK3   48
@@ -179,6 +179,10 @@ struct saa7134_format {
 #define SAA7135_BOARD_ASUSTeK_TVFM7135 53
 #define SAA7134_BOARD_FLYTVPLATINUM_FM 54
 #define SAA7134_BOARD_FLYDVBTDUO 55
+#define SAA7134_BOARD_AVERMEDIA_307    56
+#define SAA7134_BOARD_AVERMEDIA_GO_007_FM 57
+#define SAA7134_BOARD_ADS_INSTANT_TV 58
+#define SAA7134_BOARD_KWORLD_VSTREAM_XPERT 59
 
 #define SAA7134_MAXBOARDS 8
 #define SAA7134_INPUT_MAX 8
diff --git a/drivers/media/video/tveeprom.c b/drivers/media/video/tveeprom.c
index 3d216973798..cf069bf0d6a 100644
--- a/drivers/media/video/tveeprom.c
+++ b/drivers/media/video/tveeprom.c
@@ -75,7 +75,7 @@ hauppauge_tuner_fmt[] =
 	{ 0x00000007, "PAL(B/G)" },
 	{ 0x00001000, "NTSC(M)" },
 	{ 0x00000010, "PAL(I)" },
-	{ 0x00400000, "SECAM(L/L�)" },
+	{ 0x00400000, "SECAM(L/L´)" },
 	{ 0x00000e00, "PAL(D/K)" },
 	{ 0x03000000, "ATSC Digital" },
 };
@@ -482,6 +482,7 @@ static unsigned short normal_i2c[] = {
 	0xa0 >> 1,
 	I2C_CLIENT_END,
 };
+static unsigned short normal_i2c_range[] = { I2C_CLIENT_END };
 I2C_CLIENT_INSMOD;
 
 struct i2c_driver i2c_driver_tveeprom;
diff --git a/drivers/media/video/v4l1-compat.c b/drivers/media/video/v4l1-compat.c
index b0d4bcb027d..70ecbdb8027 100644
--- a/drivers/media/video/v4l1-compat.c
+++ b/drivers/media/video/v4l1-compat.c
@@ -1,4 +1,6 @@
 /*
+ * $Id: v4l1-compat.c,v 1.9 2005/06/12 04:19:19 mchehab Exp $
+ *
  *	Video for Linux Two
  *	Backward Compatibility Layer
  *
@@ -15,14 +17,11 @@
  *
  */
 
-#ifndef __KERNEL__
-#define __KERNEL__
-#endif
-
 #include <linux/config.h>
 
 #include <linux/init.h>
 #include <linux/module.h>
+#include <linux/moduleparam.h>
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
@@ -787,12 +786,15 @@ v4l_compat_translate_ioctl(struct inode         *inode,
 		    !(qctrl2.flags & V4L2_CTRL_FLAG_DISABLED))
 			aud->step = qctrl2.step;
 		aud->mode = 0;
+
+		memset(&tun2,0,sizeof(tun2));
 		err = drv(inode, file, VIDIOC_G_TUNER, &tun2);
 		if (err < 0) {
 			dprintk("VIDIOCGAUDIO / VIDIOC_G_TUNER: %d\n",err);
 			err = 0;
 			break;
 		}
+
 		if (tun2.rxsubchans & V4L2_TUNER_SUB_LANG2)
 			aud->mode = VIDEO_SOUND_LANG1 | VIDEO_SOUND_LANG2;
 		else if (tun2.rxsubchans & V4L2_TUNER_SUB_STEREO)
diff --git a/include/media/tveeprom.h b/include/media/tveeprom.h
index 627603e561a..5c4fe30e8d1 100644
--- a/include/media/tveeprom.h
+++ b/include/media/tveeprom.h
@@ -1,3 +1,7 @@
+/*
+ * $Id: tveeprom.h,v 1.2 2005/06/12 04:19:19 mchehab Exp $
+ */
+
 struct tveeprom {
 	u32 has_radio;
 
-- 
cgit v1.2.3-70-g09d2


From c988d2b2845495373f666a381d354a7f80981d62 Mon Sep 17 00:00:00 2001
From: Matt Domsch <Matt_Domsch@dell.com>
Date: Thu, 23 Jun 2005 22:05:15 -0700
Subject: [PATCH] modules: add version and srcversion to sysfs

This patch adds version and srcversion files to
/sys/module/${modulename} containing the version and srcversion fields
of the module's modinfo section (if present).

/sys/module/e1000
|-- srcversion
`-- version

This patch differs slightly from the version posted in January, as it
now uses the new kstrdup() call in -mm.

Why put this in sysfs?

a) Tools like DKMS, which deal with changing out individual kernel
   modules without replacing the whole kernel, can behave smarter if they
   can tell the version of a given module.  The autoinstaller feature, for
   example, which determines if your system has a "good" version of a
   driver (i.e.  if the one provided by DKMS has a newer verson than that
   provided by the kernel package installed), and to automatically compile
   and install a newer version if DKMS has it but your kernel doesn't yet
   have that version.

b) Because sysadmins manually, or with tools like DKMS, can switch out
   modules on the file system, you can't count on 'modinfo foo.ko', which
   looks at /lib/modules/${kernelver}/...  actually matching what is loaded
   into the kernel already.  Hence asking sysfs for this.

c) as the unbind-driver-from-device work takes shape, it will be
   possible to rebind a driver that's built-in (no .ko to modinfo for the
   version) to a newly loaded module.  sysfs will have the
   currently-built-in version info, for comparison.

d) tech support scripts can then easily grab the version info for what's
   running presently - a question I get often.

There has been renewed interest in this patch on linux-scsi by driver
authors.

As the idea originated from GregKH, I leave his Signed-off-by: intact,
though the implementation is nearly completely new.  Compiled and run on
x86 and x86_64.

From: Matthew Dobson <colpatch@us.ibm.com>

      build fix

From: Thierry Vignaud <tvignaud@mandriva.com>

      build fix

From: Matthew Dobson <colpatch@us.ibm.com>

      warning fix

Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>
Signed-off-by: Matt Domsch <Matt_Domsch@dell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/module.h |  5 +++
 kernel/module.c        | 95 ++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 100 insertions(+)

(limited to 'include')

diff --git a/include/linux/module.h b/include/linux/module.h
index 0e432a0f4ae..f05372b7fe7 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -51,6 +51,9 @@ struct module_attribute {
         ssize_t (*show)(struct module_attribute *, struct module *, char *);
         ssize_t (*store)(struct module_attribute *, struct module *,
 			 const char *, size_t count);
+	void (*setup)(struct module *, const char *);
+	int (*test)(struct module *);
+	void (*free)(struct module *);
 };
 
 struct module_kobject
@@ -239,6 +242,8 @@ struct module
 	/* Sysfs stuff. */
 	struct module_kobject mkobj;
 	struct module_param_attrs *param_attrs;
+	const char *version;
+	const char *srcversion;
 
 	/* Exported symbols */
 	const struct kernel_symbol *syms;
diff --git a/kernel/module.c b/kernel/module.c
index a566745dde6..0494c89a0d2 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -35,6 +35,7 @@
 #include <linux/notifier.h>
 #include <linux/stop_machine.h>
 #include <linux/device.h>
+#include <linux/string.h>
 #include <asm/uaccess.h>
 #include <asm/semaphore.h>
 #include <asm/cacheflush.h>
@@ -370,6 +371,43 @@ static inline void percpu_modcopy(void *pcpudst, const void *src,
 #endif /* CONFIG_SMP */
 
 #ifdef CONFIG_MODULE_UNLOAD
+#define MODINFO_ATTR(field)	\
+static void setup_modinfo_##field(struct module *mod, const char *s)  \
+{                                                                     \
+	mod->field = kstrdup(s, GFP_KERNEL);                          \
+}                                                                     \
+static ssize_t show_modinfo_##field(struct module_attribute *mattr,   \
+	                struct module *mod, char *buffer)             \
+{                                                                     \
+	return sprintf(buffer, "%s\n", mod->field);                   \
+}                                                                     \
+static int modinfo_##field##_exists(struct module *mod)               \
+{                                                                     \
+	return mod->field != NULL;                                    \
+}                                                                     \
+static void free_modinfo_##field(struct module *mod)                  \
+{                                                                     \
+        kfree(mod->field);                                            \
+        mod->field = NULL;                                            \
+}                                                                     \
+static struct module_attribute modinfo_##field = {                    \
+	.attr = { .name = __stringify(field), .mode = 0444,           \
+		  .owner = THIS_MODULE },                             \
+	.show = show_modinfo_##field,                                 \
+	.setup = setup_modinfo_##field,                               \
+	.test = modinfo_##field##_exists,                             \
+	.free = free_modinfo_##field,                                 \
+};
+
+MODINFO_ATTR(version);
+MODINFO_ATTR(srcversion);
+
+static struct module_attribute *modinfo_attrs[] = {
+	&modinfo_version,
+	&modinfo_srcversion,
+	NULL,
+};
+
 /* Init the unload section of the module. */
 static void module_unload_init(struct module *mod)
 {
@@ -1031,6 +1069,32 @@ static void module_remove_refcnt_attr(struct module *mod)
 }
 #endif
 
+#ifdef CONFIG_MODULE_UNLOAD
+static int module_add_modinfo_attrs(struct module *mod)
+{
+	struct module_attribute *attr;
+	int error = 0;
+	int i;
+
+	for (i = 0; (attr = modinfo_attrs[i]) && !error; i++) {
+		if (!attr->test ||
+		    (attr->test && attr->test(mod)))
+			error = sysfs_create_file(&mod->mkobj.kobj,&attr->attr);
+	}
+	return error;
+}
+
+static void module_remove_modinfo_attrs(struct module *mod)
+{
+	struct module_attribute *attr;
+	int i;
+
+	for (i = 0; (attr = modinfo_attrs[i]); i++) {
+		sysfs_remove_file(&mod->mkobj.kobj,&attr->attr);
+		attr->free(mod);
+	}
+}
+#endif
 
 static int mod_sysfs_setup(struct module *mod,
 			   struct kernel_param *kparam,
@@ -1056,6 +1120,12 @@ static int mod_sysfs_setup(struct module *mod,
 	if (err)
 		goto out_unreg;
 
+#ifdef CONFIG_MODULE_UNLOAD
+	err = module_add_modinfo_attrs(mod);
+	if (err)
+		goto out_unreg;
+#endif
+
 	return 0;
 
 out_unreg:
@@ -1066,6 +1136,9 @@ out:
 
 static void mod_kobject_remove(struct module *mod)
 {
+#ifdef CONFIG_MODULE_UNLOAD
+	module_remove_modinfo_attrs(mod);
+#endif
 	module_remove_refcnt_attr(mod);
 	module_param_sysfs_remove(mod);
 
@@ -1311,6 +1384,23 @@ static char *get_modinfo(Elf_Shdr *sechdrs,
 	return NULL;
 }
 
+#ifdef CONFIG_MODULE_UNLOAD
+static void setup_modinfo(struct module *mod, Elf_Shdr *sechdrs,
+			  unsigned int infoindex)
+{
+	struct module_attribute *attr;
+	int i;
+
+	for (i = 0; (attr = modinfo_attrs[i]); i++) {
+		if (attr->setup)
+			attr->setup(mod,
+				    get_modinfo(sechdrs,
+						infoindex,
+						attr->attr.name));
+	}
+}
+#endif
+
 #ifdef CONFIG_KALLSYMS
 int is_exported(const char *name, const struct module *mod)
 {
@@ -1615,6 +1705,11 @@ static struct module *load_module(void __user *umod,
 	/* Set up license info based on the info section */
 	set_license(mod, get_modinfo(sechdrs, infoindex, "license"));
 
+#ifdef CONFIG_MODULE_UNLOAD
+	/* Set up MODINFO_ATTR fields */
+	setup_modinfo(mod, sechdrs, infoindex);
+#endif
+
 	/* Fix up syms, so that st_value is a pointer to location. */
 	err = simplify_symbols(sechdrs, symindex, strtab, versindex, pcpuindex,
 			       mod);
-- 
cgit v1.2.3-70-g09d2


From 420edbcc09008342c7b2665453f6b370739aadb0 Mon Sep 17 00:00:00 2001
From: Carsten Otte <cotte@de.ibm.com>
Date: Thu, 23 Jun 2005 22:05:23 -0700
Subject: [PATCH] xip: bdev: execute in place

This is the block device related part.  The block device operation
direct_access now has a struct block_device as first parameter.

Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/s390/block/dcssblk.c | 44 ++++++++++++++++++++++++++++++++++++++++----
 include/linux/fs.h           |  1 +
 2 files changed, 41 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/drivers/s390/block/dcssblk.c b/drivers/s390/block/dcssblk.c
index 16ab8d363ac..6bc27d52326 100644
--- a/drivers/s390/block/dcssblk.c
+++ b/drivers/s390/block/dcssblk.c
@@ -35,14 +35,17 @@
 static int dcssblk_open(struct inode *inode, struct file *filp);
 static int dcssblk_release(struct inode *inode, struct file *filp);
 static int dcssblk_make_request(struct request_queue *q, struct bio *bio);
+static int dcssblk_direct_access(struct block_device *bdev, sector_t secnum,
+				 unsigned long *data);
 
 static char dcssblk_segments[DCSSBLK_PARM_LEN] = "\0";
 
 static int dcssblk_major;
 static struct block_device_operations dcssblk_devops = {
-	.owner   = THIS_MODULE,
-	.open    = dcssblk_open,
-	.release = dcssblk_release,
+	.owner   	= THIS_MODULE,
+	.open    	= dcssblk_open,
+	.release 	= dcssblk_release,
+	.direct_access 	= dcssblk_direct_access,
 };
 
 static ssize_t dcssblk_add_store(struct device * dev, struct device_attribute *attr, const char * buf,
@@ -641,6 +644,20 @@ dcssblk_make_request(request_queue_t *q, struct bio *bio)
 		/* Request beyond end of DCSS segment. */
 		goto fail;
 	}
+	/* verify data transfer direction */
+	if (dev_info->is_shared) {
+		switch (dev_info->segment_type) {
+		case SEG_TYPE_SR:
+		case SEG_TYPE_ER:
+		case SEG_TYPE_SC:
+			/* cannot write to these segments */
+			if (bio_data_dir(bio) == WRITE) {
+				PRINT_WARN("rejecting write to ro segment %s\n", dev_info->dev.bus_id);
+				goto fail;
+			}
+		}
+	}
+
 	index = (bio->bi_sector >> 3);
 	bio_for_each_segment(bvec, bio, i) {
 		page_addr = (unsigned long)
@@ -661,7 +678,26 @@ dcssblk_make_request(request_queue_t *q, struct bio *bio)
 	bio_endio(bio, bytes_done, 0);
 	return 0;
 fail:
-	bio_io_error(bio, bytes_done);
+	bio_io_error(bio, bio->bi_size);
+	return 0;
+}
+
+static int
+dcssblk_direct_access (struct block_device *bdev, sector_t secnum,
+			unsigned long *data)
+{
+	struct dcssblk_dev_info *dev_info;
+	unsigned long pgoff;
+
+	dev_info = bdev->bd_disk->private_data;
+	if (!dev_info)
+		return -ENODEV;
+	if (secnum % (PAGE_SIZE/512))
+		return -EINVAL;
+	pgoff = secnum / (PAGE_SIZE / 512);
+	if ((pgoff+1)*PAGE_SIZE-1 > dev_info->end - dev_info->start)
+		return -ERANGE;
+	*data = (unsigned long) (dev_info->start+pgoff*PAGE_SIZE);
 	return 0;
 }
 
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 83857d8070d..929bf8d20c8 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -885,6 +885,7 @@ struct block_device_operations {
 	int (*ioctl) (struct inode *, struct file *, unsigned, unsigned long);
 	long (*unlocked_ioctl) (struct file *, unsigned, unsigned long);
 	long (*compat_ioctl) (struct file *, unsigned, unsigned long);
+	int (*direct_access) (struct block_device *, sector_t, unsigned long *);
 	int (*media_changed) (struct gendisk *);
 	int (*revalidate_disk) (struct gendisk *);
 	struct module *owner;
-- 
cgit v1.2.3-70-g09d2


From ceffc078528befc008c6f2c2c4decda79eabd534 Mon Sep 17 00:00:00 2001
From: Carsten Otte <cotte@de.ibm.com>
Date: Thu, 23 Jun 2005 22:05:25 -0700
Subject: [PATCH] xip: fs/mm: execute in place

- generic_file* file operations do no longer have a xip/non-xip split
- filemap_xip.c implements a new set of fops that require get_xip_page
  aop to work proper. all new fops are exported GPL-only (don't like to
  see whatever code use those except GPL modules)
- __xip_unmap now uses page_check_address, which is no longer static
  in rmap.c, and defined in linux/rmap.h
- mm/filemap.h is now much more clean, plainly having just Linus'
  inline funcs moved here from filemap.c
- fix includes in filemap_xip to make it build cleanly on i386

Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/open.c            |   4 +-
 include/linux/fs.h   |  18 ++
 include/linux/rmap.h |   6 +
 mm/Makefile          |   1 +
 mm/filemap.c         |  74 +------
 mm/filemap.h         |  94 +++++++++
 mm/filemap_xip.c     | 581 +++++++++++++++++++++++++++++++++++++++++++++++++++
 mm/rmap.c            |   4 +-
 8 files changed, 707 insertions(+), 75 deletions(-)
 create mode 100644 mm/filemap.h
 create mode 100644 mm/filemap_xip.c

(limited to 'include')

diff --git a/fs/open.c b/fs/open.c
index 8ec63f73591..3f4a4286fdc 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -808,7 +808,9 @@ struct file *dentry_open(struct dentry *dentry, struct vfsmount *mnt, int flags)
 
 	/* NB: we're sure to have correct a_ops only after f_op->open */
 	if (f->f_flags & O_DIRECT) {
-		if (!f->f_mapping->a_ops || !f->f_mapping->a_ops->direct_IO) {
+		if (!f->f_mapping->a_ops ||
+		    ((!f->f_mapping->a_ops->direct_IO) &&
+		    (!f->f_mapping->a_ops->get_xip_page))) {
 			fput(f);
 			f = ERR_PTR(-EINVAL);
 		}
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 929bf8d20c8..79c0fafc021 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -330,6 +330,8 @@ struct address_space_operations {
 	int (*releasepage) (struct page *, int);
 	ssize_t (*direct_IO)(int, struct kiocb *, const struct iovec *iov,
 			loff_t offset, unsigned long nr_segs);
+	struct page* (*get_xip_page)(struct address_space *, sector_t,
+			int);
 };
 
 struct backing_dev_info;
@@ -1497,6 +1499,22 @@ extern loff_t remote_llseek(struct file *file, loff_t offset, int origin);
 extern int generic_file_open(struct inode * inode, struct file * filp);
 extern int nonseekable_open(struct inode * inode, struct file * filp);
 
+#ifdef CONFIG_FS_XIP
+extern ssize_t xip_file_aio_read(struct kiocb *iocb, char __user *buf,
+				 size_t count, loff_t pos);
+extern ssize_t xip_file_readv(struct file *filp, const struct iovec *iov,
+			      unsigned long nr_segs, loff_t *ppos);
+extern ssize_t xip_file_sendfile(struct file *in_file, loff_t *ppos,
+				 size_t count, read_actor_t actor,
+				 void *target);
+extern int xip_file_mmap(struct file * file, struct vm_area_struct * vma);
+extern ssize_t xip_file_aio_write(struct kiocb *iocb, const char __user *buf,
+				  size_t count, loff_t pos);
+extern ssize_t xip_file_writev(struct file *file, const struct iovec *iov,
+			       unsigned long nr_segs, loff_t *ppos);
+extern int xip_truncate_page(struct address_space *mapping, loff_t from);
+#endif
+
 static inline void do_generic_file_read(struct file * filp, loff_t *ppos,
 					read_descriptor_t * desc,
 					read_actor_t actor)
diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index 11b484e37ac..e80fb7ee6ef 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -92,6 +92,12 @@ static inline void page_dup_rmap(struct page *page)
 int page_referenced(struct page *, int is_locked, int ignore_token);
 int try_to_unmap(struct page *);
 
+/*
+ * Called from mm/filemap_xip.c to unmap empty zero page
+ */
+pte_t *page_check_address(struct page *, struct mm_struct *, unsigned long);
+
+
 /*
  * Used by swapoff to help locate where page is expected in vma.
  */
diff --git a/mm/Makefile b/mm/Makefile
index 8f70ffd763c..4cd69e3ce42 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -19,3 +19,4 @@ obj-$(CONFIG_SPARSEMEM)	+= sparse.o
 obj-$(CONFIG_SHMEM) += shmem.o
 obj-$(CONFIG_TINY_SHMEM) += tiny-shmem.o
 
+obj-$(CONFIG_FS_XIP) += filemap_xip.o
diff --git a/mm/filemap.c b/mm/filemap.c
index a3598b542a3..7332194d7af 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -28,6 +28,7 @@
 #include <linux/blkdev.h>
 #include <linux/security.h>
 #include <linux/syscalls.h>
+#include "filemap.h"
 /*
  * FIXME: remove all knowledge of the buffer layer from the core VM
  */
@@ -1714,32 +1715,7 @@ int remove_suid(struct dentry *dentry)
 }
 EXPORT_SYMBOL(remove_suid);
 
-/*
- * Copy as much as we can into the page and return the number of bytes which
- * were sucessfully copied.  If a fault is encountered then clear the page
- * out to (offset+bytes) and return the number of bytes which were copied.
- */
-static inline size_t
-filemap_copy_from_user(struct page *page, unsigned long offset,
-			const char __user *buf, unsigned bytes)
-{
-	char *kaddr;
-	int left;
-
-	kaddr = kmap_atomic(page, KM_USER0);
-	left = __copy_from_user_inatomic(kaddr + offset, buf, bytes);
-	kunmap_atomic(kaddr, KM_USER0);
-
-	if (left != 0) {
-		/* Do it the slow way */
-		kaddr = kmap(page);
-		left = __copy_from_user(kaddr + offset, buf, bytes);
-		kunmap(page);
-	}
-	return bytes - left;
-}
-
-static size_t
+size_t
 __filemap_copy_from_user_iovec(char *vaddr, 
 			const struct iovec *iov, size_t base, size_t bytes)
 {
@@ -1766,52 +1742,6 @@ __filemap_copy_from_user_iovec(char *vaddr,
 	return copied - left;
 }
 
-/*
- * This has the same sideeffects and return value as filemap_copy_from_user().
- * The difference is that on a fault we need to memset the remainder of the
- * page (out to offset+bytes), to emulate filemap_copy_from_user()'s
- * single-segment behaviour.
- */
-static inline size_t
-filemap_copy_from_user_iovec(struct page *page, unsigned long offset,
-			const struct iovec *iov, size_t base, size_t bytes)
-{
-	char *kaddr;
-	size_t copied;
-
-	kaddr = kmap_atomic(page, KM_USER0);
-	copied = __filemap_copy_from_user_iovec(kaddr + offset, iov,
-						base, bytes);
-	kunmap_atomic(kaddr, KM_USER0);
-	if (copied != bytes) {
-		kaddr = kmap(page);
-		copied = __filemap_copy_from_user_iovec(kaddr + offset, iov,
-							base, bytes);
-		kunmap(page);
-	}
-	return copied;
-}
-
-static inline void
-filemap_set_next_iovec(const struct iovec **iovp, size_t *basep, size_t bytes)
-{
-	const struct iovec *iov = *iovp;
-	size_t base = *basep;
-
-	while (bytes) {
-		int copy = min(bytes, iov->iov_len - base);
-
-		bytes -= copy;
-		base += copy;
-		if (iov->iov_len == base) {
-			iov++;
-			base = 0;
-		}
-	}
-	*iovp = iov;
-	*basep = base;
-}
-
 /*
  * Performs necessary checks before doing a write
  *
diff --git a/mm/filemap.h b/mm/filemap.h
new file mode 100644
index 00000000000..c2d0546a57e
--- /dev/null
+++ b/mm/filemap.h
@@ -0,0 +1,94 @@
+/*
+ *	linux/mm/filemap.h
+ *
+ * Copyright (C) 1994-1999  Linus Torvalds
+ */
+
+#ifndef __FILEMAP_H
+#define __FILEMAP_H
+
+#include <linux/types.h>
+#include <linux/fs.h>
+#include <linux/mm.h>
+#include <linux/highmem.h>
+#include <linux/uio.h>
+#include <linux/config.h>
+#include <asm/uaccess.h>
+
+extern size_t
+__filemap_copy_from_user_iovec(char *vaddr,
+			       const struct iovec *iov,
+			       size_t base,
+			       size_t bytes);
+
+/*
+ * Copy as much as we can into the page and return the number of bytes which
+ * were sucessfully copied.  If a fault is encountered then clear the page
+ * out to (offset+bytes) and return the number of bytes which were copied.
+ */
+static inline size_t
+filemap_copy_from_user(struct page *page, unsigned long offset,
+			const char __user *buf, unsigned bytes)
+{
+	char *kaddr;
+	int left;
+
+	kaddr = kmap_atomic(page, KM_USER0);
+	left = __copy_from_user_inatomic(kaddr + offset, buf, bytes);
+	kunmap_atomic(kaddr, KM_USER0);
+
+	if (left != 0) {
+		/* Do it the slow way */
+		kaddr = kmap(page);
+		left = __copy_from_user(kaddr + offset, buf, bytes);
+		kunmap(page);
+	}
+	return bytes - left;
+}
+
+/*
+ * This has the same sideeffects and return value as filemap_copy_from_user().
+ * The difference is that on a fault we need to memset the remainder of the
+ * page (out to offset+bytes), to emulate filemap_copy_from_user()'s
+ * single-segment behaviour.
+ */
+static inline size_t
+filemap_copy_from_user_iovec(struct page *page, unsigned long offset,
+			const struct iovec *iov, size_t base, size_t bytes)
+{
+	char *kaddr;
+	size_t copied;
+
+	kaddr = kmap_atomic(page, KM_USER0);
+	copied = __filemap_copy_from_user_iovec(kaddr + offset, iov,
+						base, bytes);
+	kunmap_atomic(kaddr, KM_USER0);
+	if (copied != bytes) {
+		kaddr = kmap(page);
+		copied = __filemap_copy_from_user_iovec(kaddr + offset, iov,
+							base, bytes);
+		kunmap(page);
+	}
+	return copied;
+}
+
+static inline void
+filemap_set_next_iovec(const struct iovec **iovp, size_t *basep, size_t bytes)
+{
+	const struct iovec *iov = *iovp;
+	size_t base = *basep;
+
+	while (bytes) {
+		int copy = min(bytes, iov->iov_len - base);
+
+		bytes -= copy;
+		base += copy;
+		if (iov->iov_len == base) {
+			iov++;
+			base = 0;
+		}
+	}
+	*iovp = iov;
+	*basep = base;
+}
+#endif
diff --git a/mm/filemap_xip.c b/mm/filemap_xip.c
new file mode 100644
index 00000000000..7d63acd4881
--- /dev/null
+++ b/mm/filemap_xip.c
@@ -0,0 +1,581 @@
+/*
+ *	linux/mm/filemap_xip.c
+ *
+ * Copyright (C) 2005 IBM Corporation
+ * Author: Carsten Otte <cotte@de.ibm.com>
+ *
+ * derived from linux/mm/filemap.c - Copyright (C) Linus Torvalds
+ *
+ */
+
+#include <linux/fs.h>
+#include <linux/pagemap.h>
+#include <linux/module.h>
+#include <linux/uio.h>
+#include <linux/rmap.h>
+#include <asm/tlbflush.h>
+#include "filemap.h"
+
+/*
+ * This is a file read routine for execute in place files, and uses
+ * the mapping->a_ops->get_xip_page() function for the actual low-level
+ * stuff.
+ *
+ * Note the struct file* is not used at all.  It may be NULL.
+ */
+static void
+do_xip_mapping_read(struct address_space *mapping,
+		    struct file_ra_state *_ra,
+		    struct file *filp,
+		    loff_t *ppos,
+		    read_descriptor_t *desc,
+		    read_actor_t actor)
+{
+	struct inode *inode = mapping->host;
+	unsigned long index, end_index, offset;
+	loff_t isize;
+
+	BUG_ON(!mapping->a_ops->get_xip_page);
+
+	index = *ppos >> PAGE_CACHE_SHIFT;
+	offset = *ppos & ~PAGE_CACHE_MASK;
+
+	isize = i_size_read(inode);
+	if (!isize)
+		goto out;
+
+	end_index = (isize - 1) >> PAGE_CACHE_SHIFT;
+	for (;;) {
+		struct page *page;
+		unsigned long nr, ret;
+
+		/* nr is the maximum number of bytes to copy from this page */
+		nr = PAGE_CACHE_SIZE;
+		if (index >= end_index) {
+			if (index > end_index)
+				goto out;
+			nr = ((isize - 1) & ~PAGE_CACHE_MASK) + 1;
+			if (nr <= offset) {
+				goto out;
+			}
+		}
+		nr = nr - offset;
+
+		page = mapping->a_ops->get_xip_page(mapping,
+			index*(PAGE_SIZE/512), 0);
+		if (!page)
+			goto no_xip_page;
+		if (unlikely(IS_ERR(page))) {
+			if (PTR_ERR(page) == -ENODATA) {
+				/* sparse */
+				page = virt_to_page(empty_zero_page);
+			} else {
+				desc->error = PTR_ERR(page);
+				goto out;
+			}
+		} else
+			BUG_ON(!PageUptodate(page));
+
+		/* If users can be writing to this page using arbitrary
+		 * virtual addresses, take care about potential aliasing
+		 * before reading the page on the kernel side.
+		 */
+		if (mapping_writably_mapped(mapping))
+			flush_dcache_page(page);
+
+		/*
+		 * Ok, we have the page, and it's up-to-date, so
+		 * now we can copy it to user space...
+		 *
+		 * The actor routine returns how many bytes were actually used..
+		 * NOTE! This may not be the same as how much of a user buffer
+		 * we filled up (we may be padding etc), so we can only update
+		 * "pos" here (the actor routine has to update the user buffer
+		 * pointers and the remaining count).
+		 */
+		ret = actor(desc, page, offset, nr);
+		offset += ret;
+		index += offset >> PAGE_CACHE_SHIFT;
+		offset &= ~PAGE_CACHE_MASK;
+
+		if (ret == nr && desc->count)
+			continue;
+		goto out;
+
+no_xip_page:
+		/* Did not get the page. Report it */
+		desc->error = -EIO;
+		goto out;
+	}
+
+out:
+	*ppos = ((loff_t) index << PAGE_CACHE_SHIFT) + offset;
+	if (filp)
+		file_accessed(filp);
+}
+
+/*
+ * This is the "read()" routine for all filesystems
+ * that uses the get_xip_page address space operation.
+ */
+static ssize_t
+__xip_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
+		unsigned long nr_segs, loff_t *ppos)
+{
+	struct file *filp = iocb->ki_filp;
+	ssize_t retval;
+	unsigned long seg;
+	size_t count;
+
+	count = 0;
+	for (seg = 0; seg < nr_segs; seg++) {
+		const struct iovec *iv = &iov[seg];
+
+		/*
+		 * If any segment has a negative length, or the cumulative
+		 * length ever wraps negative then return -EINVAL.
+		 */
+		count += iv->iov_len;
+		if (unlikely((ssize_t)(count|iv->iov_len) < 0))
+			return -EINVAL;
+		if (access_ok(VERIFY_WRITE, iv->iov_base, iv->iov_len))
+			continue;
+		if (seg == 0)
+			return -EFAULT;
+		nr_segs = seg;
+		count -= iv->iov_len;	/* This segment is no good */
+		break;
+	}
+
+	retval = 0;
+	if (count) {
+		for (seg = 0; seg < nr_segs; seg++) {
+			read_descriptor_t desc;
+
+			desc.written = 0;
+			desc.arg.buf = iov[seg].iov_base;
+			desc.count = iov[seg].iov_len;
+			if (desc.count == 0)
+				continue;
+			desc.error = 0;
+			do_xip_mapping_read(filp->f_mapping, &filp->f_ra, filp,
+					    ppos, &desc, file_read_actor);
+			retval += desc.written;
+			if (!retval) {
+				retval = desc.error;
+				break;
+			}
+		}
+	}
+	return retval;
+}
+
+ssize_t
+xip_file_aio_read(struct kiocb *iocb, char __user *buf, size_t count,
+		  loff_t pos)
+{
+	struct iovec local_iov = { .iov_base = buf, .iov_len = count };
+
+	BUG_ON(iocb->ki_pos != pos);
+	return __xip_file_aio_read(iocb, &local_iov, 1, &iocb->ki_pos);
+}
+EXPORT_SYMBOL_GPL(xip_file_aio_read);
+
+ssize_t
+xip_file_readv(struct file *filp, const struct iovec *iov,
+	       unsigned long nr_segs, loff_t *ppos)
+{
+	struct kiocb kiocb;
+
+	init_sync_kiocb(&kiocb, filp);
+	return __xip_file_aio_read(&kiocb, iov, nr_segs, ppos);
+}
+EXPORT_SYMBOL_GPL(xip_file_readv);
+
+ssize_t
+xip_file_sendfile(struct file *in_file, loff_t *ppos,
+	     size_t count, read_actor_t actor, void *target)
+{
+	read_descriptor_t desc;
+
+	if (!count)
+		return 0;
+
+	desc.written = 0;
+	desc.count = count;
+	desc.arg.data = target;
+	desc.error = 0;
+
+	do_xip_mapping_read(in_file->f_mapping, &in_file->f_ra, in_file,
+			    ppos, &desc, actor);
+	if (desc.written)
+		return desc.written;
+	return desc.error;
+}
+EXPORT_SYMBOL_GPL(xip_file_sendfile);
+
+/*
+ * __xip_unmap is invoked from xip_unmap and
+ * xip_write
+ *
+ * This function walks all vmas of the address_space and unmaps the
+ * empty_zero_page when found at pgoff. Should it go in rmap.c?
+ */
+static void
+__xip_unmap (struct address_space * mapping,
+		     unsigned long pgoff)
+{
+	struct vm_area_struct *vma;
+	struct mm_struct *mm;
+	struct prio_tree_iter iter;
+	unsigned long address;
+	pte_t *pte;
+	pte_t pteval;
+
+	spin_lock(&mapping->i_mmap_lock);
+	vma_prio_tree_foreach(vma, &iter, &mapping->i_mmap, pgoff, pgoff) {
+		mm = vma->vm_mm;
+		address = vma->vm_start +
+			((pgoff - vma->vm_pgoff) << PAGE_SHIFT);
+		BUG_ON(address < vma->vm_start || address >= vma->vm_end);
+		/*
+		 * We need the page_table_lock to protect us from page faults,
+		 * munmap, fork, etc...
+		 */
+		pte = page_check_address(virt_to_page(empty_zero_page), mm,
+					 address);
+		if (!IS_ERR(pte)) {
+			/* Nuke the page table entry. */
+			flush_cache_page(vma, address, pte_pfn(pte));
+			pteval = ptep_clear_flush(vma, address, pte);
+			BUG_ON(pte_dirty(pteval));
+			pte_unmap(pte);
+			spin_unlock(&mm->page_table_lock);
+		}
+	}
+	spin_unlock(&mapping->i_mmap_lock);
+}
+
+/*
+ * xip_nopage() is invoked via the vma operations vector for a
+ * mapped memory region to read in file data during a page fault.
+ *
+ * This function is derived from filemap_nopage, but used for execute in place
+ */
+static struct page *
+xip_file_nopage(struct vm_area_struct * area,
+		   unsigned long address,
+		   int *type)
+{
+	struct file *file = area->vm_file;
+	struct address_space *mapping = file->f_mapping;
+	struct inode *inode = mapping->host;
+	struct page *page;
+	unsigned long size, pgoff, endoff;
+
+	pgoff = ((address - area->vm_start) >> PAGE_CACHE_SHIFT)
+		+ area->vm_pgoff;
+	endoff = ((area->vm_end - area->vm_start) >> PAGE_CACHE_SHIFT)
+		+ area->vm_pgoff;
+
+	size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+	if (pgoff >= size) {
+		return NULL;
+	}
+
+	page = mapping->a_ops->get_xip_page(mapping, pgoff*(PAGE_SIZE/512), 0);
+	if (!IS_ERR(page)) {
+		BUG_ON(!PageUptodate(page));
+		return page;
+	}
+	if (PTR_ERR(page) != -ENODATA)
+		return NULL;
+
+	/* sparse block */
+	if ((area->vm_flags & (VM_WRITE | VM_MAYWRITE)) &&
+	    (area->vm_flags & (VM_SHARED| VM_MAYSHARE)) &&
+	    (!(mapping->host->i_sb->s_flags & MS_RDONLY))) {
+		/* maybe shared writable, allocate new block */
+		page = mapping->a_ops->get_xip_page (mapping,
+			pgoff*(PAGE_SIZE/512), 1);
+		if (IS_ERR(page))
+			return NULL;
+		BUG_ON(!PageUptodate(page));
+		/* unmap page at pgoff from all other vmas */
+		__xip_unmap(mapping, pgoff);
+	} else {
+		/* not shared and writable, use empty_zero_page */
+		page = virt_to_page(empty_zero_page);
+	}
+
+	return page;
+}
+
+static struct vm_operations_struct xip_file_vm_ops = {
+	.nopage         = xip_file_nopage,
+};
+
+int xip_file_mmap(struct file * file, struct vm_area_struct * vma)
+{
+	BUG_ON(!file->f_mapping->a_ops->get_xip_page);
+
+	file_accessed(file);
+	vma->vm_ops = &xip_file_vm_ops;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(xip_file_mmap);
+
+static ssize_t
+do_xip_file_write(struct kiocb *iocb, const struct iovec *iov,
+		  unsigned long nr_segs, loff_t pos, loff_t *ppos,
+		  size_t count)
+{
+	struct file *file = iocb->ki_filp;
+	struct address_space * mapping = file->f_mapping;
+	struct address_space_operations *a_ops = mapping->a_ops;
+	struct inode 	*inode = mapping->host;
+	long		status = 0;
+	struct page	*page;
+	size_t		bytes;
+	const struct iovec *cur_iov = iov; /* current iovec */
+	size_t		iov_base = 0;	   /* offset in the current iovec */
+	char __user	*buf;
+	ssize_t		written = 0;
+
+	BUG_ON(!mapping->a_ops->get_xip_page);
+
+	buf = iov->iov_base;
+	do {
+		unsigned long index;
+		unsigned long offset;
+		size_t copied;
+
+		offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */
+		index = pos >> PAGE_CACHE_SHIFT;
+		bytes = PAGE_CACHE_SIZE - offset;
+		if (bytes > count)
+			bytes = count;
+
+		/*
+		 * Bring in the user page that we will copy from _first_.
+		 * Otherwise there's a nasty deadlock on copying from the
+		 * same page as we're writing to, without it being marked
+		 * up-to-date.
+		 */
+		fault_in_pages_readable(buf, bytes);
+
+		page = a_ops->get_xip_page(mapping,
+						    index*(PAGE_SIZE/512), 0);
+		if (IS_ERR(page) && (PTR_ERR(page) == -ENODATA)) {
+			/* we allocate a new page unmap it */
+			page = a_ops->get_xip_page(mapping,
+				index*(PAGE_SIZE/512), 1);
+			if (!IS_ERR(page))
+			/* unmap page at pgoff from all other vmas */
+			__xip_unmap(mapping, index);
+
+		}
+
+		if (IS_ERR(page)) {
+			status = PTR_ERR(page);
+			break;
+		}
+
+		BUG_ON(!PageUptodate(page));
+
+		if (likely(nr_segs == 1))
+			copied = filemap_copy_from_user(page, offset,
+							buf, bytes);
+		else
+			copied = filemap_copy_from_user_iovec(page, offset,
+						cur_iov, iov_base, bytes);
+		flush_dcache_page(page);
+		if (likely(copied > 0)) {
+			status = copied;
+
+			if (status >= 0) {
+				written += status;
+				count -= status;
+				pos += status;
+				buf += status;
+				if (unlikely(nr_segs > 1))
+					filemap_set_next_iovec(&cur_iov,
+							&iov_base, status);
+			}
+		}
+		if (unlikely(copied != bytes))
+			if (status >= 0)
+				status = -EFAULT;
+		if (status < 0)
+			break;
+	} while (count);
+	*ppos = pos;
+	/*
+	 * No need to use i_size_read() here, the i_size
+	 * cannot change under us because we hold i_sem.
+	 */
+	if (pos > inode->i_size) {
+		i_size_write(inode, pos);
+		mark_inode_dirty(inode);
+	}
+
+	return written ? written : status;
+}
+
+static ssize_t
+xip_file_aio_write_nolock(struct kiocb *iocb, const struct iovec *iov,
+				unsigned long nr_segs, loff_t *ppos)
+{
+	struct file *file = iocb->ki_filp;
+	struct address_space * mapping = file->f_mapping;
+	size_t ocount;		/* original count */
+	size_t count;		/* after file limit checks */
+	struct inode 	*inode = mapping->host;
+	unsigned long	seg;
+	loff_t		pos;
+	ssize_t		written;
+	ssize_t		err;
+
+	ocount = 0;
+	for (seg = 0; seg < nr_segs; seg++) {
+		const struct iovec *iv = &iov[seg];
+
+		/*
+		 * If any segment has a negative length, or the cumulative
+		 * length ever wraps negative then return -EINVAL.
+		 */
+		ocount += iv->iov_len;
+		if (unlikely((ssize_t)(ocount|iv->iov_len) < 0))
+			return -EINVAL;
+		if (access_ok(VERIFY_READ, iv->iov_base, iv->iov_len))
+			continue;
+		if (seg == 0)
+			return -EFAULT;
+		nr_segs = seg;
+		ocount -= iv->iov_len;	/* This segment is no good */
+		break;
+	}
+
+	count = ocount;
+	pos = *ppos;
+
+	vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);
+
+	written = 0;
+
+	err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode));
+	if (err)
+		goto out;
+
+	if (count == 0)
+		goto out;
+
+	err = remove_suid(file->f_dentry);
+	if (err)
+		goto out;
+
+	inode_update_time(inode, 1);
+
+	/* use execute in place to copy directly to disk */
+	written = do_xip_file_write (iocb, iov,
+				  nr_segs, pos, ppos, count);
+ out:
+	return written ? written : err;
+}
+
+static ssize_t
+__xip_file_write_nolock(struct file *file, const struct iovec *iov,
+			unsigned long nr_segs, loff_t *ppos)
+{
+	struct kiocb kiocb;
+
+	init_sync_kiocb(&kiocb, file);
+	return xip_file_aio_write_nolock(&kiocb, iov, nr_segs, ppos);
+}
+
+ssize_t
+xip_file_aio_write(struct kiocb *iocb, const char __user *buf,
+		       size_t count, loff_t pos)
+{
+	struct file *file = iocb->ki_filp;
+	struct address_space *mapping = file->f_mapping;
+	struct inode *inode = mapping->host;
+	ssize_t ret;
+	struct iovec local_iov = { .iov_base = (void __user *)buf,
+				   .iov_len = count };
+
+	BUG_ON(iocb->ki_pos != pos);
+
+	down(&inode->i_sem);
+	ret = xip_file_aio_write_nolock(iocb, &local_iov, 1, &iocb->ki_pos);
+	up(&inode->i_sem);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(xip_file_aio_write);
+
+ssize_t xip_file_writev(struct file *file, const struct iovec *iov,
+			unsigned long nr_segs, loff_t *ppos)
+{
+	struct address_space *mapping = file->f_mapping;
+	struct inode *inode = mapping->host;
+	ssize_t ret;
+
+	down(&inode->i_sem);
+	ret = __xip_file_write_nolock(file, iov, nr_segs, ppos);
+	up(&inode->i_sem);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(xip_file_writev);
+
+/*
+ * truncate a page used for execute in place
+ * functionality is analog to block_truncate_page but does use get_xip_page
+ * to get the page instead of page cache
+ */
+int
+xip_truncate_page(struct address_space *mapping, loff_t from)
+{
+	pgoff_t index = from >> PAGE_CACHE_SHIFT;
+	unsigned offset = from & (PAGE_CACHE_SIZE-1);
+	unsigned blocksize;
+	unsigned length;
+	struct page *page;
+	void *kaddr;
+	int err;
+
+	BUG_ON(!mapping->a_ops->get_xip_page);
+
+	blocksize = 1 << mapping->host->i_blkbits;
+	length = offset & (blocksize - 1);
+
+	/* Block boundary? Nothing to do */
+	if (!length)
+		return 0;
+
+	length = blocksize - length;
+
+	page = mapping->a_ops->get_xip_page(mapping,
+					    index*(PAGE_SIZE/512), 0);
+	err = -ENOMEM;
+	if (!page)
+		goto out;
+	if (unlikely(IS_ERR(page))) {
+		if (PTR_ERR(page) == -ENODATA) {
+			/* Hole? No need to truncate */
+			return 0;
+		} else {
+			err = PTR_ERR(page);
+			goto out;
+		}
+	} else
+		BUG_ON(!PageUptodate(page));
+	kaddr = kmap_atomic(page, KM_USER0);
+	memset(kaddr + offset, 0, length);
+	kunmap_atomic(kaddr, KM_USER0);
+
+	flush_dcache_page(page);
+	err = 0;
+out:
+	return err;
+}
+EXPORT_SYMBOL_GPL(xip_truncate_page);
diff --git a/mm/rmap.c b/mm/rmap.c
index 89770bd25f3..08ac5c7fa91 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -247,8 +247,8 @@ unsigned long page_address_in_vma(struct page *page, struct vm_area_struct *vma)
  *
  * On success returns with mapped pte and locked mm->page_table_lock.
  */
-static pte_t *page_check_address(struct page *page, struct mm_struct *mm,
-					unsigned long address)
+pte_t *page_check_address(struct page *page, struct mm_struct *mm,
+			  unsigned long address)
 {
 	pgd_t *pgd;
 	pud_t *pud;
-- 
cgit v1.2.3-70-g09d2


From 6d79125bba55ee82701f1c7d4ebbc1aa20ecbe4e Mon Sep 17 00:00:00 2001
From: Carsten Otte <cotte@de.ibm.com>
Date: Thu, 23 Jun 2005 22:05:26 -0700
Subject: [PATCH] xip: ext2: execute in place

These are the ext2 related parts.  Ext2 now uses the xip_* file operations
along with the get_xip_page aop when mounted with -o xip.

Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/Kconfig              | 17 +++++++++++
 fs/ext2/Makefile        |  1 +
 fs/ext2/ext2.h          |  2 ++
 fs/ext2/file.c          | 18 +++++++++++
 fs/ext2/inode.c         | 31 ++++++++++++++++---
 fs/ext2/namei.c         | 12 ++++++--
 fs/ext2/super.c         | 27 ++++++++++++++++-
 fs/ext2/xip.c           | 80 +++++++++++++++++++++++++++++++++++++++++++++++++
 fs/ext2/xip.h           | 25 ++++++++++++++++
 include/linux/ext2_fs.h | 25 ++++++++--------
 include/linux/fs.h      |  5 ++++
 11 files changed, 223 insertions(+), 20 deletions(-)
 create mode 100644 fs/ext2/xip.c
 create mode 100644 fs/ext2/xip.h

(limited to 'include')

diff --git a/fs/Kconfig b/fs/Kconfig
index 5c704d05627..8157f2e2d51 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -50,6 +50,23 @@ config EXT2_FS_SECURITY
 	  If you are not using a security module that requires using
 	  extended attributes for file security labels, say N.
 
+config EXT2_FS_XIP
+	bool "Ext2 execute in place support"
+	depends on EXT2_FS
+	help
+	  Execute in place can be used on memory-backed block devices. If you
+	  enable this option, you can select to mount block devices which are
+	  capable of this feature without using the page cache.
+
+	  If you do not use a block device that is capable of using this,
+	  or if unsure, say N.
+
+config FS_XIP
+# execute in place
+	bool
+	depends on EXT2_FS_XIP
+	default y
+
 config EXT3_FS
 	tristate "Ext3 journalling file system support"
 	help
diff --git a/fs/ext2/Makefile b/fs/ext2/Makefile
index ee240a14e70..c5d02da73bc 100644
--- a/fs/ext2/Makefile
+++ b/fs/ext2/Makefile
@@ -10,3 +10,4 @@ ext2-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \
 ext2-$(CONFIG_EXT2_FS_XATTR)	 += xattr.o xattr_user.o xattr_trusted.o
 ext2-$(CONFIG_EXT2_FS_POSIX_ACL) += acl.o
 ext2-$(CONFIG_EXT2_FS_SECURITY)	 += xattr_security.o
+ext2-$(CONFIG_EXT2_FS_XIP)	 += xip.o
diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h
index 8f0fd726c3f..eed521d22cf 100644
--- a/fs/ext2/ext2.h
+++ b/fs/ext2/ext2.h
@@ -147,9 +147,11 @@ extern struct file_operations ext2_dir_operations;
 /* file.c */
 extern struct inode_operations ext2_file_inode_operations;
 extern struct file_operations ext2_file_operations;
+extern struct file_operations ext2_xip_file_operations;
 
 /* inode.c */
 extern struct address_space_operations ext2_aops;
+extern struct address_space_operations ext2_aops_xip;
 extern struct address_space_operations ext2_nobh_aops;
 
 /* namei.c */
diff --git a/fs/ext2/file.c b/fs/ext2/file.c
index f5e86141ec5..2b3d572365a 100644
--- a/fs/ext2/file.c
+++ b/fs/ext2/file.c
@@ -55,6 +55,24 @@ struct file_operations ext2_file_operations = {
 	.sendfile	= generic_file_sendfile,
 };
 
+#ifdef CONFIG_EXT2_FS_XIP
+struct file_operations ext2_xip_file_operations = {
+	.llseek		= generic_file_llseek,
+	.read		= do_sync_read,
+	.write		= do_sync_write,
+	.aio_read	= xip_file_aio_read,
+	.aio_write	= xip_file_aio_write,
+	.ioctl		= ext2_ioctl,
+	.mmap		= xip_file_mmap,
+	.open		= generic_file_open,
+	.release	= ext2_release_file,
+	.fsync		= ext2_sync_file,
+	.readv		= xip_file_readv,
+	.writev		= xip_file_writev,
+	.sendfile	= xip_file_sendfile,
+};
+#endif
+
 struct inode_operations ext2_file_inode_operations = {
 	.truncate	= ext2_truncate,
 #ifdef CONFIG_EXT2_FS_XATTR
diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index a50d9db4b6e..53dceb0c659 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -33,6 +33,7 @@
 #include <linux/mpage.h>
 #include "ext2.h"
 #include "acl.h"
+#include "xip.h"
 
 MODULE_AUTHOR("Remy Card and others");
 MODULE_DESCRIPTION("Second Extended Filesystem");
@@ -594,6 +595,16 @@ out:
 	if (err)
 		goto cleanup;
 
+	if (ext2_use_xip(inode->i_sb)) {
+		/*
+		 * we need to clear the block
+		 */
+		err = ext2_clear_xip_target (inode,
+			le32_to_cpu(chain[depth-1].key));
+		if (err)
+			goto cleanup;
+	}
+
 	if (ext2_splice_branch(inode, iblock, chain, partial, left) < 0)
 		goto changed;
 
@@ -691,6 +702,11 @@ struct address_space_operations ext2_aops = {
 	.writepages		= ext2_writepages,
 };
 
+struct address_space_operations ext2_aops_xip = {
+	.bmap			= ext2_bmap,
+	.get_xip_page		= ext2_get_xip_page,
+};
+
 struct address_space_operations ext2_nobh_aops = {
 	.readpage		= ext2_readpage,
 	.readpages		= ext2_readpages,
@@ -910,7 +926,9 @@ void ext2_truncate (struct inode * inode)
 	iblock = (inode->i_size + blocksize-1)
 					>> EXT2_BLOCK_SIZE_BITS(inode->i_sb);
 
-	if (test_opt(inode->i_sb, NOBH))
+	if (mapping_is_xip(inode->i_mapping))
+		xip_truncate_page(inode->i_mapping, inode->i_size);
+	else if (test_opt(inode->i_sb, NOBH))
 		nobh_truncate_page(inode->i_mapping, inode->i_size);
 	else
 		block_truncate_page(inode->i_mapping,
@@ -1110,11 +1128,16 @@ void ext2_read_inode (struct inode * inode)
 
 	if (S_ISREG(inode->i_mode)) {
 		inode->i_op = &ext2_file_inode_operations;
-		inode->i_fop = &ext2_file_operations;
-		if (test_opt(inode->i_sb, NOBH))
+		if (ext2_use_xip(inode->i_sb)) {
+			inode->i_mapping->a_ops = &ext2_aops_xip;
+			inode->i_fop = &ext2_xip_file_operations;
+		} else if (test_opt(inode->i_sb, NOBH)) {
 			inode->i_mapping->a_ops = &ext2_nobh_aops;
-		else
+			inode->i_fop = &ext2_file_operations;
+		} else {
 			inode->i_mapping->a_ops = &ext2_aops;
+			inode->i_fop = &ext2_file_operations;
+		}
 	} else if (S_ISDIR(inode->i_mode)) {
 		inode->i_op = &ext2_dir_inode_operations;
 		inode->i_fop = &ext2_dir_operations;
diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c
index 3176b3d3ffa..c5513953c82 100644
--- a/fs/ext2/namei.c
+++ b/fs/ext2/namei.c
@@ -34,6 +34,7 @@
 #include "ext2.h"
 #include "xattr.h"
 #include "acl.h"
+#include "xip.h"
 
 /*
  * Couple of helper functions - make the code slightly cleaner.
@@ -127,11 +128,16 @@ static int ext2_create (struct inode * dir, struct dentry * dentry, int mode, st
 	int err = PTR_ERR(inode);
 	if (!IS_ERR(inode)) {
 		inode->i_op = &ext2_file_inode_operations;
-		inode->i_fop = &ext2_file_operations;
-		if (test_opt(inode->i_sb, NOBH))
+		if (ext2_use_xip(inode->i_sb)) {
+			inode->i_mapping->a_ops = &ext2_aops_xip;
+			inode->i_fop = &ext2_xip_file_operations;
+		} else if (test_opt(inode->i_sb, NOBH)) {
 			inode->i_mapping->a_ops = &ext2_nobh_aops;
-		else
+			inode->i_fop = &ext2_file_operations;
+		} else {
 			inode->i_mapping->a_ops = &ext2_aops;
+			inode->i_fop = &ext2_file_operations;
+		}
 		mark_inode_dirty(inode);
 		err = ext2_add_nondir(dentry, inode);
 	}
diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index 661c3d98d94..876e391f287 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -31,6 +31,7 @@
 #include "ext2.h"
 #include "xattr.h"
 #include "acl.h"
+#include "xip.h"
 
 static void ext2_sync_super(struct super_block *sb,
 			    struct ext2_super_block *es);
@@ -257,7 +258,7 @@ enum {
 	Opt_bsd_df, Opt_minix_df, Opt_grpid, Opt_nogrpid,
 	Opt_resgid, Opt_resuid, Opt_sb, Opt_err_cont, Opt_err_panic, Opt_err_ro,
 	Opt_nouid32, Opt_check, Opt_nocheck, Opt_debug, Opt_oldalloc, Opt_orlov, Opt_nobh,
-	Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl,
+	Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl, Opt_xip,
 	Opt_ignore, Opt_err,
 };
 
@@ -286,6 +287,7 @@ static match_table_t tokens = {
 	{Opt_nouser_xattr, "nouser_xattr"},
 	{Opt_acl, "acl"},
 	{Opt_noacl, "noacl"},
+	{Opt_xip, "xip"},
 	{Opt_ignore, "grpquota"},
 	{Opt_ignore, "noquota"},
 	{Opt_ignore, "quota"},
@@ -397,6 +399,13 @@ static int parse_options (char * options,
 			printk("EXT2 (no)acl options not supported\n");
 			break;
 #endif
+		case Opt_xip:
+#ifdef CONFIG_EXT2_FS_XIP
+			set_opt (sbi->s_mount_opt, XIP);
+#else
+			printk("EXT2 xip option not supported\n");
+#endif
+			break;
 		case Opt_ignore:
 			break;
 		default:
@@ -640,6 +649,9 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 		((EXT2_SB(sb)->s_mount_opt & EXT2_MOUNT_POSIX_ACL) ?
 		 MS_POSIXACL : 0);
 
+	ext2_xip_verify_sb(sb); /* see if bdev supports xip, unset
+				    EXT2_MOUNT_XIP if not */
+
 	if (le32_to_cpu(es->s_rev_level) == EXT2_GOOD_OLD_REV &&
 	    (EXT2_HAS_COMPAT_FEATURE(sb, ~0U) ||
 	     EXT2_HAS_RO_COMPAT_FEATURE(sb, ~0U) ||
@@ -668,6 +680,13 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 
 	blocksize = BLOCK_SIZE << le32_to_cpu(sbi->s_es->s_log_block_size);
 
+	if ((ext2_use_xip(sb)) && ((blocksize != PAGE_SIZE) ||
+				  (sb->s_blocksize != blocksize))) {
+		if (!silent)
+			printk("XIP: Unsupported blocksize\n");
+		goto failed_mount;
+	}
+
 	/* If the blocksize doesn't match, re-read the thing.. */
 	if (sb->s_blocksize != blocksize) {
 		brelse(bh);
@@ -916,6 +935,7 @@ static int ext2_remount (struct super_block * sb, int * flags, char * data)
 {
 	struct ext2_sb_info * sbi = EXT2_SB(sb);
 	struct ext2_super_block * es;
+	unsigned long old_mount_opt = sbi->s_mount_opt;
 
 	/*
 	 * Allow the "check" option to be passed as a remount option.
@@ -927,6 +947,11 @@ static int ext2_remount (struct super_block * sb, int * flags, char * data)
 		((sbi->s_mount_opt & EXT2_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0);
 
 	es = sbi->s_es;
+	if (((sbi->s_mount_opt & EXT2_MOUNT_XIP) !=
+	    (old_mount_opt & EXT2_MOUNT_XIP)) &&
+	    invalidate_inodes(sb))
+		ext2_warning(sb, __FUNCTION__, "busy inodes while remounting "\
+			     "xip remain in cache (no functional problem)");
 	if ((*flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY))
 		return 0;
 	if (*flags & MS_RDONLY) {
diff --git a/fs/ext2/xip.c b/fs/ext2/xip.c
new file mode 100644
index 00000000000..d44431d1a33
--- /dev/null
+++ b/fs/ext2/xip.c
@@ -0,0 +1,80 @@
+/*
+ *  linux/fs/ext2/xip.c
+ *
+ * Copyright (C) 2005 IBM Corporation
+ * Author: Carsten Otte (cotte@de.ibm.com)
+ */
+
+#include <linux/mm.h>
+#include <linux/fs.h>
+#include <linux/genhd.h>
+#include <linux/buffer_head.h>
+#include <linux/ext2_fs_sb.h>
+#include <linux/ext2_fs.h>
+#include "ext2.h"
+#include "xip.h"
+
+static inline int
+__inode_direct_access(struct inode *inode, sector_t sector, unsigned long *data) {
+	BUG_ON(!inode->i_sb->s_bdev->bd_disk->fops->direct_access);
+	return inode->i_sb->s_bdev->bd_disk->fops
+		->direct_access(inode->i_sb->s_bdev,sector,data);
+}
+
+int
+ext2_clear_xip_target(struct inode *inode, int block) {
+	sector_t sector = block*(PAGE_SIZE/512);
+	unsigned long data;
+	int rc;
+
+	rc = __inode_direct_access(inode, sector, &data);
+	if (rc)
+		return rc;
+	clear_page((void*)data);
+	return 0;
+}
+
+void ext2_xip_verify_sb(struct super_block *sb)
+{
+	struct ext2_sb_info *sbi = EXT2_SB(sb);
+
+	if ((sbi->s_mount_opt & EXT2_MOUNT_XIP)) {
+		if ((sb->s_bdev == NULL) ||
+			sb->s_bdev->bd_disk == NULL ||
+			sb->s_bdev->bd_disk->fops == NULL ||
+			sb->s_bdev->bd_disk->fops->direct_access == NULL) {
+			sbi->s_mount_opt &= (~EXT2_MOUNT_XIP);
+			ext2_warning(sb, __FUNCTION__,
+				"ignoring xip option - not supported by bdev");
+		}
+	}
+}
+
+struct page*
+ext2_get_xip_page(struct address_space *mapping, sector_t blockno,
+		   int create)
+{
+	int rc;
+	unsigned long data;
+	struct buffer_head tmp;
+
+	tmp.b_state = 0;
+	tmp.b_blocknr = 0;
+	rc = ext2_get_block(mapping->host, blockno/(PAGE_SIZE/512) , &tmp,
+				create);
+	if (rc)
+		return ERR_PTR(rc);
+	if (tmp.b_blocknr == 0) {
+		/* SPARSE block */
+		BUG_ON(create);
+		return ERR_PTR(-ENODATA);
+	}
+
+	rc = __inode_direct_access
+		(mapping->host,tmp.b_blocknr*(PAGE_SIZE/512) ,&data);
+	if (rc)
+		return ERR_PTR(rc);
+
+	SetPageUptodate(virt_to_page(data));
+	return virt_to_page(data);
+}
diff --git a/fs/ext2/xip.h b/fs/ext2/xip.h
new file mode 100644
index 00000000000..aa85331d6c5
--- /dev/null
+++ b/fs/ext2/xip.h
@@ -0,0 +1,25 @@
+/*
+ *  linux/fs/ext2/xip.h
+ *
+ * Copyright (C) 2005 IBM Corporation
+ * Author: Carsten Otte (cotte@de.ibm.com)
+ */
+
+#ifdef CONFIG_EXT2_FS_XIP
+extern void ext2_xip_verify_sb (struct super_block *);
+extern int ext2_clear_xip_target (struct inode *, int);
+
+static inline int ext2_use_xip (struct super_block *sb)
+{
+	struct ext2_sb_info *sbi = EXT2_SB(sb);
+	return (sbi->s_mount_opt & EXT2_MOUNT_XIP);
+}
+struct page* ext2_get_xip_page (struct address_space *, sector_t, int);
+#define mapping_is_xip(map) unlikely(map->a_ops->get_xip_page)
+#else
+#define mapping_is_xip(map)			0
+#define ext2_xip_verify_sb(sb)			do { } while (0)
+#define ext2_use_xip(sb)			0
+#define ext2_clear_xip_target(inode, chain)	0
+#define ext2_get_xip_page			NULL
+#endif
diff --git a/include/linux/ext2_fs.h b/include/linux/ext2_fs.h
index fab43527e59..a657130ba03 100644
--- a/include/linux/ext2_fs.h
+++ b/include/linux/ext2_fs.h
@@ -300,18 +300,19 @@ struct ext2_inode {
 /*
  * Mount flags
  */
-#define EXT2_MOUNT_CHECK		0x0001	/* Do mount-time checks */
-#define EXT2_MOUNT_OLDALLOC		0x0002  /* Don't use the new Orlov allocator */
-#define EXT2_MOUNT_GRPID		0x0004	/* Create files with directory's group */
-#define EXT2_MOUNT_DEBUG		0x0008	/* Some debugging messages */
-#define EXT2_MOUNT_ERRORS_CONT		0x0010	/* Continue on errors */
-#define EXT2_MOUNT_ERRORS_RO		0x0020	/* Remount fs ro on errors */
-#define EXT2_MOUNT_ERRORS_PANIC		0x0040	/* Panic on errors */
-#define EXT2_MOUNT_MINIX_DF		0x0080	/* Mimics the Minix statfs */
-#define EXT2_MOUNT_NOBH			0x0100	/* No buffer_heads */
-#define EXT2_MOUNT_NO_UID32		0x0200  /* Disable 32-bit UIDs */
-#define EXT2_MOUNT_XATTR_USER		0x4000	/* Extended user attributes */
-#define EXT2_MOUNT_POSIX_ACL		0x8000	/* POSIX Access Control Lists */
+#define EXT2_MOUNT_CHECK		0x000001  /* Do mount-time checks */
+#define EXT2_MOUNT_OLDALLOC		0x000002  /* Don't use the new Orlov allocator */
+#define EXT2_MOUNT_GRPID		0x000004  /* Create files with directory's group */
+#define EXT2_MOUNT_DEBUG		0x000008  /* Some debugging messages */
+#define EXT2_MOUNT_ERRORS_CONT		0x000010  /* Continue on errors */
+#define EXT2_MOUNT_ERRORS_RO		0x000020  /* Remount fs ro on errors */
+#define EXT2_MOUNT_ERRORS_PANIC		0x000040  /* Panic on errors */
+#define EXT2_MOUNT_MINIX_DF		0x000080  /* Mimics the Minix statfs */
+#define EXT2_MOUNT_NOBH			0x000100  /* No buffer_heads */
+#define EXT2_MOUNT_NO_UID32		0x000200  /* Disable 32-bit UIDs */
+#define EXT2_MOUNT_XATTR_USER		0x004000  /* Extended user attributes */
+#define EXT2_MOUNT_POSIX_ACL		0x008000  /* POSIX Access Control Lists */
+#define EXT2_MOUNT_XIP			0x010000  /* Execute in place */
 
 #define clear_opt(o, opt)		o &= ~EXT2_MOUNT_##opt
 #define set_opt(o, opt)			o |= EXT2_MOUNT_##opt
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 79c0fafc021..7e0501895f3 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1513,6 +1513,11 @@ extern ssize_t xip_file_aio_write(struct kiocb *iocb, const char __user *buf,
 extern ssize_t xip_file_writev(struct file *file, const struct iovec *iov,
 			       unsigned long nr_segs, loff_t *ppos);
 extern int xip_truncate_page(struct address_space *mapping, loff_t from);
+#else
+static inline int xip_truncate_page(struct address_space *mapping, loff_t from)
+{
+	return 0;
+}
 #endif
 
 static inline void do_generic_file_read(struct file * filp, loff_t *ppos,
-- 
cgit v1.2.3-70-g09d2


From eb6fe0c388e43b02e261f0fdee60e42f6298d7f7 Mon Sep 17 00:00:00 2001
From: Carsten Otte <cotte@de.ibm.com>
Date: Thu, 23 Jun 2005 22:05:28 -0700
Subject: [PATCH] xip: reduce code duplication

This patch reworks filemap_xip.c with the goal to reduce code duplication
from mm/filemap.c.  It applies agains 2.6.12-rc6-mm1.  Instead of
implementing the aio functions, this one implements the synchronous
read/write functions only.  For readv and writev, the generic fallback is
used.  For aio, we rely on the application doing the fallback.  Since our
"synchronous" function does memcpy immediately anyway, there is no
performance difference between using the fallbacks or implementing each
operation.

Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext2/file.c     |   8 +-
 include/linux/fs.h |  12 +--
 mm/filemap.h       |   2 +-
 mm/filemap_xip.c   | 246 ++++++++++++-----------------------------------------
 4 files changed, 63 insertions(+), 205 deletions(-)

(limited to 'include')

diff --git a/fs/ext2/file.c b/fs/ext2/file.c
index 2b3d572365a..a484412fc78 100644
--- a/fs/ext2/file.c
+++ b/fs/ext2/file.c
@@ -58,17 +58,13 @@ struct file_operations ext2_file_operations = {
 #ifdef CONFIG_EXT2_FS_XIP
 struct file_operations ext2_xip_file_operations = {
 	.llseek		= generic_file_llseek,
-	.read		= do_sync_read,
-	.write		= do_sync_write,
-	.aio_read	= xip_file_aio_read,
-	.aio_write	= xip_file_aio_write,
+	.read		= xip_file_read,
+	.write		= xip_file_write,
 	.ioctl		= ext2_ioctl,
 	.mmap		= xip_file_mmap,
 	.open		= generic_file_open,
 	.release	= ext2_release_file,
 	.fsync		= ext2_sync_file,
-	.readv		= xip_file_readv,
-	.writev		= xip_file_writev,
 	.sendfile	= xip_file_sendfile,
 };
 #endif
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 7e0501895f3..3ae8e37bdfc 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1500,18 +1500,14 @@ extern int generic_file_open(struct inode * inode, struct file * filp);
 extern int nonseekable_open(struct inode * inode, struct file * filp);
 
 #ifdef CONFIG_FS_XIP
-extern ssize_t xip_file_aio_read(struct kiocb *iocb, char __user *buf,
-				 size_t count, loff_t pos);
-extern ssize_t xip_file_readv(struct file *filp, const struct iovec *iov,
-			      unsigned long nr_segs, loff_t *ppos);
+extern ssize_t xip_file_read(struct file *filp, char __user *buf, size_t len,
+			     loff_t *ppos);
 extern ssize_t xip_file_sendfile(struct file *in_file, loff_t *ppos,
 				 size_t count, read_actor_t actor,
 				 void *target);
 extern int xip_file_mmap(struct file * file, struct vm_area_struct * vma);
-extern ssize_t xip_file_aio_write(struct kiocb *iocb, const char __user *buf,
-				  size_t count, loff_t pos);
-extern ssize_t xip_file_writev(struct file *file, const struct iovec *iov,
-			       unsigned long nr_segs, loff_t *ppos);
+extern ssize_t xip_file_write(struct file *filp, const char __user *buf,
+			      size_t len, loff_t *ppos);
 extern int xip_truncate_page(struct address_space *mapping, loff_t from);
 #else
 static inline int xip_truncate_page(struct address_space *mapping, loff_t from)
diff --git a/mm/filemap.h b/mm/filemap.h
index c2d0546a57e..13793ba0ce1 100644
--- a/mm/filemap.h
+++ b/mm/filemap.h
@@ -15,7 +15,7 @@
 #include <linux/config.h>
 #include <asm/uaccess.h>
 
-extern size_t
+size_t
 __filemap_copy_from_user_iovec(char *vaddr,
 			       const struct iovec *iov,
 			       size_t base,
diff --git a/mm/filemap_xip.c b/mm/filemap_xip.c
index 7d63acd4881..3b6e384b98a 100644
--- a/mm/filemap_xip.c
+++ b/mm/filemap_xip.c
@@ -114,83 +114,28 @@ out:
 		file_accessed(filp);
 }
 
-/*
- * This is the "read()" routine for all filesystems
- * that uses the get_xip_page address space operation.
- */
-static ssize_t
-__xip_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
-		unsigned long nr_segs, loff_t *ppos)
-{
-	struct file *filp = iocb->ki_filp;
-	ssize_t retval;
-	unsigned long seg;
-	size_t count;
-
-	count = 0;
-	for (seg = 0; seg < nr_segs; seg++) {
-		const struct iovec *iv = &iov[seg];
-
-		/*
-		 * If any segment has a negative length, or the cumulative
-		 * length ever wraps negative then return -EINVAL.
-		 */
-		count += iv->iov_len;
-		if (unlikely((ssize_t)(count|iv->iov_len) < 0))
-			return -EINVAL;
-		if (access_ok(VERIFY_WRITE, iv->iov_base, iv->iov_len))
-			continue;
-		if (seg == 0)
-			return -EFAULT;
-		nr_segs = seg;
-		count -= iv->iov_len;	/* This segment is no good */
-		break;
-	}
-
-	retval = 0;
-	if (count) {
-		for (seg = 0; seg < nr_segs; seg++) {
-			read_descriptor_t desc;
-
-			desc.written = 0;
-			desc.arg.buf = iov[seg].iov_base;
-			desc.count = iov[seg].iov_len;
-			if (desc.count == 0)
-				continue;
-			desc.error = 0;
-			do_xip_mapping_read(filp->f_mapping, &filp->f_ra, filp,
-					    ppos, &desc, file_read_actor);
-			retval += desc.written;
-			if (!retval) {
-				retval = desc.error;
-				break;
-			}
-		}
-	}
-	return retval;
-}
-
 ssize_t
-xip_file_aio_read(struct kiocb *iocb, char __user *buf, size_t count,
-		  loff_t pos)
+xip_file_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos)
 {
-	struct iovec local_iov = { .iov_base = buf, .iov_len = count };
+	read_descriptor_t desc;
 
-	BUG_ON(iocb->ki_pos != pos);
-	return __xip_file_aio_read(iocb, &local_iov, 1, &iocb->ki_pos);
-}
-EXPORT_SYMBOL_GPL(xip_file_aio_read);
+	if (!access_ok(VERIFY_WRITE, buf, len))
+		return -EFAULT;
 
-ssize_t
-xip_file_readv(struct file *filp, const struct iovec *iov,
-	       unsigned long nr_segs, loff_t *ppos)
-{
-	struct kiocb kiocb;
+	desc.written = 0;
+	desc.arg.buf = buf;
+	desc.count = len;
+	desc.error = 0;
 
-	init_sync_kiocb(&kiocb, filp);
-	return __xip_file_aio_read(&kiocb, iov, nr_segs, ppos);
+	do_xip_mapping_read(filp->f_mapping, &filp->f_ra, filp,
+			    ppos, &desc, file_read_actor);
+
+	if (desc.written)
+		return desc.written;
+	else
+		return desc.error;
 }
-EXPORT_SYMBOL_GPL(xip_file_readv);
+EXPORT_SYMBOL_GPL(xip_file_read);
 
 ssize_t
 xip_file_sendfile(struct file *in_file, loff_t *ppos,
@@ -326,25 +271,19 @@ int xip_file_mmap(struct file * file, struct vm_area_struct * vma)
 EXPORT_SYMBOL_GPL(xip_file_mmap);
 
 static ssize_t
-do_xip_file_write(struct kiocb *iocb, const struct iovec *iov,
-		  unsigned long nr_segs, loff_t pos, loff_t *ppos,
-		  size_t count)
+__xip_file_write(struct file *filp, const char __user *buf,
+		  size_t count, loff_t pos, loff_t *ppos)
 {
-	struct file *file = iocb->ki_filp;
-	struct address_space * mapping = file->f_mapping;
+	struct address_space * mapping = filp->f_mapping;
 	struct address_space_operations *a_ops = mapping->a_ops;
 	struct inode 	*inode = mapping->host;
 	long		status = 0;
 	struct page	*page;
 	size_t		bytes;
-	const struct iovec *cur_iov = iov; /* current iovec */
-	size_t		iov_base = 0;	   /* offset in the current iovec */
-	char __user	*buf;
 	ssize_t		written = 0;
 
 	BUG_ON(!mapping->a_ops->get_xip_page);
 
-	buf = iov->iov_base;
 	do {
 		unsigned long index;
 		unsigned long offset;
@@ -365,15 +304,14 @@ do_xip_file_write(struct kiocb *iocb, const struct iovec *iov,
 		fault_in_pages_readable(buf, bytes);
 
 		page = a_ops->get_xip_page(mapping,
-						    index*(PAGE_SIZE/512), 0);
+					   index*(PAGE_SIZE/512), 0);
 		if (IS_ERR(page) && (PTR_ERR(page) == -ENODATA)) {
 			/* we allocate a new page unmap it */
 			page = a_ops->get_xip_page(mapping,
-				index*(PAGE_SIZE/512), 1);
+						   index*(PAGE_SIZE/512), 1);
 			if (!IS_ERR(page))
-			/* unmap page at pgoff from all other vmas */
-			__xip_unmap(mapping, index);
-
+				/* unmap page at pgoff from all other vmas */
+				__xip_unmap(mapping, index);
 		}
 
 		if (IS_ERR(page)) {
@@ -383,12 +321,7 @@ do_xip_file_write(struct kiocb *iocb, const struct iovec *iov,
 
 		BUG_ON(!PageUptodate(page));
 
-		if (likely(nr_segs == 1))
-			copied = filemap_copy_from_user(page, offset,
-							buf, bytes);
-		else
-			copied = filemap_copy_from_user_iovec(page, offset,
-						cur_iov, iov_base, bytes);
+		copied = filemap_copy_from_user(page, offset, buf, bytes);
 		flush_dcache_page(page);
 		if (likely(copied > 0)) {
 			status = copied;
@@ -398,9 +331,6 @@ do_xip_file_write(struct kiocb *iocb, const struct iovec *iov,
 				count -= status;
 				pos += status;
 				buf += status;
-				if (unlikely(nr_segs > 1))
-					filemap_set_next_iovec(&cur_iov,
-							&iov_base, status);
 			}
 		}
 		if (unlikely(copied != bytes))
@@ -422,110 +352,52 @@ do_xip_file_write(struct kiocb *iocb, const struct iovec *iov,
 	return written ? written : status;
 }
 
-static ssize_t
-xip_file_aio_write_nolock(struct kiocb *iocb, const struct iovec *iov,
-				unsigned long nr_segs, loff_t *ppos)
+ssize_t
+xip_file_write(struct file *filp, const char __user *buf, size_t len,
+	       loff_t *ppos)
 {
-	struct file *file = iocb->ki_filp;
-	struct address_space * mapping = file->f_mapping;
-	size_t ocount;		/* original count */
-	size_t count;		/* after file limit checks */
-	struct inode 	*inode = mapping->host;
-	unsigned long	seg;
-	loff_t		pos;
-	ssize_t		written;
-	ssize_t		err;
+	struct address_space *mapping = filp->f_mapping;
+	struct inode *inode = mapping->host;
+	size_t count;
+	loff_t pos;
+	ssize_t ret;
 
-	ocount = 0;
-	for (seg = 0; seg < nr_segs; seg++) {
-		const struct iovec *iv = &iov[seg];
+	down(&inode->i_sem);
 
-		/*
-		 * If any segment has a negative length, or the cumulative
-		 * length ever wraps negative then return -EINVAL.
-		 */
-		ocount += iv->iov_len;
-		if (unlikely((ssize_t)(ocount|iv->iov_len) < 0))
-			return -EINVAL;
-		if (access_ok(VERIFY_READ, iv->iov_base, iv->iov_len))
-			continue;
-		if (seg == 0)
-			return -EFAULT;
-		nr_segs = seg;
-		ocount -= iv->iov_len;	/* This segment is no good */
-		break;
+	if (!access_ok(VERIFY_READ, buf, len)) {
+		ret=-EFAULT;
+		goto out_up;
 	}
 
-	count = ocount;
 	pos = *ppos;
+	count = len;
 
 	vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);
 
-	written = 0;
-
-	err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode));
-	if (err)
-		goto out;
+	/* We can write back this queue in page reclaim */
+	current->backing_dev_info = mapping->backing_dev_info;
 
+	ret = generic_write_checks(filp, &pos, &count, S_ISBLK(inode->i_mode));
+	if (ret)
+		goto out_backing;
 	if (count == 0)
-		goto out;
+		goto out_backing;
 
-	err = remove_suid(file->f_dentry);
-	if (err)
-		goto out;
+	ret = remove_suid(filp->f_dentry);
+	if (ret)
+		goto out_backing;
 
 	inode_update_time(inode, 1);
 
-	/* use execute in place to copy directly to disk */
-	written = do_xip_file_write (iocb, iov,
-				  nr_segs, pos, ppos, count);
- out:
-	return written ? written : err;
-}
-
-static ssize_t
-__xip_file_write_nolock(struct file *file, const struct iovec *iov,
-			unsigned long nr_segs, loff_t *ppos)
-{
-	struct kiocb kiocb;
-
-	init_sync_kiocb(&kiocb, file);
-	return xip_file_aio_write_nolock(&kiocb, iov, nr_segs, ppos);
-}
-
-ssize_t
-xip_file_aio_write(struct kiocb *iocb, const char __user *buf,
-		       size_t count, loff_t pos)
-{
-	struct file *file = iocb->ki_filp;
-	struct address_space *mapping = file->f_mapping;
-	struct inode *inode = mapping->host;
-	ssize_t ret;
-	struct iovec local_iov = { .iov_base = (void __user *)buf,
-				   .iov_len = count };
+	ret = __xip_file_write (filp, buf, count, pos, ppos);
 
-	BUG_ON(iocb->ki_pos != pos);
-
-	down(&inode->i_sem);
-	ret = xip_file_aio_write_nolock(iocb, &local_iov, 1, &iocb->ki_pos);
+ out_backing:
+	current->backing_dev_info = NULL;
+ out_up:
 	up(&inode->i_sem);
 	return ret;
 }
-EXPORT_SYMBOL_GPL(xip_file_aio_write);
-
-ssize_t xip_file_writev(struct file *file, const struct iovec *iov,
-			unsigned long nr_segs, loff_t *ppos)
-{
-	struct address_space *mapping = file->f_mapping;
-	struct inode *inode = mapping->host;
-	ssize_t ret;
-
-	down(&inode->i_sem);
-	ret = __xip_file_write_nolock(file, iov, nr_segs, ppos);
-	up(&inode->i_sem);
-	return ret;
-}
-EXPORT_SYMBOL_GPL(xip_file_writev);
+EXPORT_SYMBOL_GPL(xip_file_write);
 
 /*
  * truncate a page used for execute in place
@@ -541,7 +413,6 @@ xip_truncate_page(struct address_space *mapping, loff_t from)
 	unsigned length;
 	struct page *page;
 	void *kaddr;
-	int err;
 
 	BUG_ON(!mapping->a_ops->get_xip_page);
 
@@ -556,17 +427,14 @@ xip_truncate_page(struct address_space *mapping, loff_t from)
 
 	page = mapping->a_ops->get_xip_page(mapping,
 					    index*(PAGE_SIZE/512), 0);
-	err = -ENOMEM;
 	if (!page)
-		goto out;
+		return -ENOMEM;
 	if (unlikely(IS_ERR(page))) {
-		if (PTR_ERR(page) == -ENODATA) {
+		if (PTR_ERR(page) == -ENODATA)
 			/* Hole? No need to truncate */
 			return 0;
-		} else {
-			err = PTR_ERR(page);
-			goto out;
-		}
+		else
+			return PTR_ERR(page);
 	} else
 		BUG_ON(!PageUptodate(page));
 	kaddr = kmap_atomic(page, KM_USER0);
@@ -574,8 +442,6 @@ xip_truncate_page(struct address_space *mapping, loff_t from)
 	kunmap_atomic(kaddr, KM_USER0);
 
 	flush_dcache_page(page);
-	err = 0;
-out:
-	return err;
+	return 0;
 }
 EXPORT_SYMBOL_GPL(xip_truncate_page);
-- 
cgit v1.2.3-70-g09d2


From 52c1da39534fb382c061de58b65f678ad74b59f5 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Thu, 23 Jun 2005 22:05:33 -0700
Subject: [PATCH] make various thing static

Another rollup of patches which give various symbols static scope

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/media/common/saa7146_fops.c |  2 +-
 drivers/media/video/tvaudio.c       | 22 +++++++++++-----------
 drivers/scsi/hosts.c                |  2 +-
 drivers/scsi/scsi.c                 |  6 ++++--
 drivers/scsi/scsi_debug.c           |  2 +-
 drivers/scsi/scsi_lib.c             |  2 +-
 drivers/scsi/scsi_priv.h            |  4 ----
 drivers/scsi/scsi_sysfs.c           |  4 ++--
 fs/namespace.c                      |  2 +-
 fs/reiserfs/stree.c                 |  2 +-
 include/linux/irq.h                 |  1 -
 include/linux/namespace.h           |  1 -
 include/net/sctp/sm.h               |  6 ------
 kernel/irq/spurious.c               |  2 +-
 kernel/module.c                     |  2 +-
 kernel/power/swsusp.c               |  2 +-
 net/sctp/sm_statefuns.c             | 16 ++++++++++++++--
 17 files changed, 40 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/drivers/media/common/saa7146_fops.c b/drivers/media/common/saa7146_fops.c
index cb826c9adfe..c04fd11526e 100644
--- a/drivers/media/common/saa7146_fops.c
+++ b/drivers/media/common/saa7146_fops.c
@@ -403,7 +403,7 @@ static struct file_operations video_fops =
 	.llseek		= no_llseek,
 };
 
-void vv_callback(struct saa7146_dev *dev, unsigned long status)
+static void vv_callback(struct saa7146_dev *dev, unsigned long status)
 {
 	u32 isr = status;
 	
diff --git a/drivers/media/video/tvaudio.c b/drivers/media/video/tvaudio.c
index 5430b25b910..9a493bea76d 100644
--- a/drivers/media/video/tvaudio.c
+++ b/drivers/media/video/tvaudio.c
@@ -1236,17 +1236,17 @@ static int ta8874z_checkit(struct CHIPSTATE *chip)
 /* audio chip descriptions - struct CHIPDESC                              */
 
 /* insmod options to enable/disable individual audio chips */
-int tda8425  = 1;
-int tda9840  = 1;
-int tda9850  = 1;
-int tda9855  = 1;
-int tda9873  = 1;
-int tda9874a = 1;
-int tea6300  = 0;  // address clash with msp34xx
-int tea6320  = 0;  // address clash with msp34xx
-int tea6420  = 1;
-int pic16c54 = 1;
-int ta8874z  = 0;  // address clash with tda9840
+static int tda8425  = 1;
+static int tda9840  = 1;
+static int tda9850  = 1;
+static int tda9855  = 1;
+static int tda9873  = 1;
+static int tda9874a = 1;
+static int tea6300  = 0;  // address clash with msp34xx
+static int tea6320  = 0;  // address clash with msp34xx
+static int tea6420  = 1;
+static int pic16c54 = 1;
+static int ta8874z  = 0;  // address clash with tda9840
 
 module_param(tda8425, int, 0444);
 module_param(tda9840, int, 0444);
diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c
index ba347576d99..d7a38b6713f 100644
--- a/drivers/scsi/hosts.c
+++ b/drivers/scsi/hosts.c
@@ -56,7 +56,7 @@ static struct class shost_class = {
  * @shost:	pointer to struct Scsi_Host
  * recovery:	recovery requested to run.
  **/
-void scsi_host_cancel(struct Scsi_Host *shost, int recovery)
+static void scsi_host_cancel(struct Scsi_Host *shost, int recovery)
 {
 	struct scsi_device *sdev;
 
diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
index 5578ae9a9e4..1cb5f7d4f27 100644
--- a/drivers/scsi/scsi.c
+++ b/drivers/scsi/scsi.c
@@ -68,6 +68,8 @@
 #include "scsi_priv.h"
 #include "scsi_logging.h"
 
+static void scsi_done(struct scsi_cmnd *cmd);
+static int scsi_retry_command(struct scsi_cmnd *cmd);
 
 /*
  * Definitions and constants.
@@ -741,7 +743,7 @@ static DEFINE_PER_CPU(struct list_head, scsi_done_q);
  *
  * This function is interrupt context safe.
  */
-void scsi_done(struct scsi_cmnd *cmd)
+static void scsi_done(struct scsi_cmnd *cmd)
 {
 	/*
 	 * We don't have to worry about this one timing out any more.
@@ -836,7 +838,7 @@ static void scsi_softirq(struct softirq_action *h)
  *              level drivers should not become re-entrant as a result of
  *              this.
  */
-int scsi_retry_command(struct scsi_cmnd *cmd)
+static int scsi_retry_command(struct scsi_cmnd *cmd)
 {
 	/*
 	 * Restore the SCSI command state.
diff --git a/drivers/scsi/scsi_debug.c b/drivers/scsi/scsi_debug.c
index e0208886b45..322b5a41a36 100644
--- a/drivers/scsi/scsi_debug.c
+++ b/drivers/scsi/scsi_debug.c
@@ -1783,7 +1783,7 @@ static void __exit scsi_debug_exit(void)
 device_initcall(scsi_debug_init);
 module_exit(scsi_debug_exit);
 
-void pseudo_0_release(struct device * dev)
+static void pseudo_0_release(struct device * dev)
 {
 	if (SCSI_DEBUG_OPT_NOISE & scsi_debug_opts)
 		printk(KERN_INFO "scsi_debug: pseudo_0_release() called\n");
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 9f996499fa9..621dee8b8cb 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -44,7 +44,7 @@ struct scsi_host_sg_pool {
 #endif
 
 #define SP(x) { x, "sgpool-" #x } 
-struct scsi_host_sg_pool scsi_sg_pools[] = { 
+static struct scsi_host_sg_pool scsi_sg_pools[] = {
 	SP(8),
 	SP(16),
 	SP(32),
diff --git a/drivers/scsi/scsi_priv.h b/drivers/scsi/scsi_priv.h
index c01580df447..96d4f745975 100644
--- a/drivers/scsi/scsi_priv.h
+++ b/drivers/scsi/scsi_priv.h
@@ -61,8 +61,6 @@ extern void scsi_exit_hosts(void);
 extern int scsi_dispatch_cmd(struct scsi_cmnd *cmd);
 extern int scsi_setup_command_freelist(struct Scsi_Host *shost);
 extern void scsi_destroy_command_freelist(struct Scsi_Host *shost);
-extern void scsi_done(struct scsi_cmnd *cmd);
-extern int scsi_retry_command(struct scsi_cmnd *cmd);
 extern int scsi_insert_special_req(struct scsi_request *sreq, int);
 extern void scsi_init_cmd_from_req(struct scsi_cmnd *cmd,
 		struct scsi_request *sreq);
@@ -136,7 +134,6 @@ extern void scsi_exit_sysctl(void);
 #endif /* CONFIG_SYSCTL */
 
 /* scsi_sysfs.c */
-extern void scsi_device_dev_release(struct device *);
 extern int scsi_sysfs_add_sdev(struct scsi_device *);
 extern int scsi_sysfs_add_host(struct Scsi_Host *);
 extern int scsi_sysfs_register(void);
@@ -145,7 +142,6 @@ extern void scsi_sysfs_device_initialize(struct scsi_device *);
 extern int scsi_sysfs_target_initialize(struct scsi_device *);
 extern struct scsi_transport_template blank_transport_template;
 
-extern struct class sdev_class;
 extern struct bus_type scsi_bus_type;
 
 /* 
diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c
index 93b41100a6d..beed7fbe1cb 100644
--- a/drivers/scsi/scsi_sysfs.c
+++ b/drivers/scsi/scsi_sysfs.c
@@ -150,7 +150,7 @@ static void scsi_device_cls_release(struct class_device *class_dev)
 	put_device(&sdev->sdev_gendev);
 }
 
-void scsi_device_dev_release(struct device *dev)
+static void scsi_device_dev_release(struct device *dev)
 {
 	struct scsi_device *sdev;
 	struct device *parent;
@@ -185,7 +185,7 @@ void scsi_device_dev_release(struct device *dev)
 		put_device(parent);
 }
 
-struct class sdev_class = {
+static struct class sdev_class = {
 	.name		= "scsi_device",
 	.release	= scsi_device_cls_release,
 };
diff --git a/fs/namespace.c b/fs/namespace.c
index 3b93e5d750e..208c079e9fd 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -337,7 +337,7 @@ int may_umount(struct vfsmount *mnt)
 
 EXPORT_SYMBOL(may_umount);
 
-void umount_tree(struct vfsmount *mnt)
+static void umount_tree(struct vfsmount *mnt)
 {
 	struct vfsmount *p;
 	LIST_HEAD(kill);
diff --git a/fs/reiserfs/stree.c b/fs/reiserfs/stree.c
index c47f8fd31a2..63158491e15 100644
--- a/fs/reiserfs/stree.c
+++ b/fs/reiserfs/stree.c
@@ -223,7 +223,7 @@ extern struct tree_balance * cur_tb;
 const struct reiserfs_key  MIN_KEY = {0, 0, {{0, 0},}};
 
 /* Maximal possible key. It is never in the tree. */
-const struct reiserfs_key  MAX_KEY = {
+static const struct reiserfs_key  MAX_KEY = {
 	__constant_cpu_to_le32(0xffffffff),
 	__constant_cpu_to_le32(0xffffffff),
 	{{__constant_cpu_to_le32(0xffffffff),
diff --git a/include/linux/irq.h b/include/linux/irq.h
index 7fc1022be9e..12277799c00 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -88,7 +88,6 @@ extern fastcall int handle_IRQ_event(unsigned int irq, struct pt_regs *regs,
 				       struct irqaction *action);
 extern fastcall unsigned int __do_IRQ(unsigned int irq, struct pt_regs *regs);
 extern void note_interrupt(unsigned int irq, irq_desc_t *desc, int action_ret);
-extern void report_bad_irq(unsigned int irq, irq_desc_t *desc, int action_ret);
 extern int can_request_irq(unsigned int irq, unsigned long irqflags);
 
 extern void init_irq_proc(void);
diff --git a/include/linux/namespace.h b/include/linux/namespace.h
index 9eca1558d72..697991b69f9 100644
--- a/include/linux/namespace.h
+++ b/include/linux/namespace.h
@@ -12,7 +12,6 @@ struct namespace {
 	struct rw_semaphore	sem;
 };
 
-extern void umount_tree(struct vfsmount *);
 extern int copy_namespace(int, struct task_struct *);
 extern void __put_namespace(struct namespace *namespace);
 
diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h
index a53e08a45e3..88d9fe5975d 100644
--- a/include/net/sctp/sm.h
+++ b/include/net/sctp/sm.h
@@ -131,7 +131,6 @@ sctp_state_fn_t sctp_sf_do_ecne;
 sctp_state_fn_t sctp_sf_ootb;
 sctp_state_fn_t sctp_sf_pdiscard;
 sctp_state_fn_t sctp_sf_violation;
-sctp_state_fn_t sctp_sf_violation_chunklen;
 sctp_state_fn_t sctp_sf_discard_chunk;
 sctp_state_fn_t sctp_sf_do_5_2_1_siminit;
 sctp_state_fn_t sctp_sf_do_5_2_2_dupinit;
@@ -259,11 +258,6 @@ struct sctp_chunk *sctp_make_fwdtsn(const struct sctp_association *asoc,
 void sctp_chunk_assign_tsn(struct sctp_chunk *);
 void sctp_chunk_assign_ssn(struct sctp_chunk *);
 
-sctp_disposition_t  sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands,
-					   __u16 error,
-					   const struct sctp_association *asoc,
-					   struct sctp_transport *transport);
-
 /* Prototypes for statetable processing. */
 
 int sctp_do_sm(sctp_event_t event_type, sctp_subtype_t subtype,
diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c
index f6297c30690..ba039e827d5 100644
--- a/kernel/irq/spurious.c
+++ b/kernel/irq/spurious.c
@@ -45,7 +45,7 @@ __report_bad_irq(unsigned int irq, irq_desc_t *desc, irqreturn_t action_ret)
 	}
 }
 
-void report_bad_irq(unsigned int irq, irq_desc_t *desc, irqreturn_t action_ret)
+static void report_bad_irq(unsigned int irq, irq_desc_t *desc, irqreturn_t action_ret)
 {
 	static int count = 100;
 
diff --git a/kernel/module.c b/kernel/module.c
index 0494c89a0d2..068e271ab3a 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -730,7 +730,7 @@ static int obsparm_copy_string(const char *val, struct kernel_param *kp)
 	return 0;
 }
 
-int set_obsolete(const char *val, struct kernel_param *kp)
+static int set_obsolete(const char *val, struct kernel_param *kp)
 {
 	unsigned int min, max;
 	unsigned int size, maxsize;
diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c
index 90b3b68dee3..53f9f8720ee 100644
--- a/kernel/power/swsusp.c
+++ b/kernel/power/swsusp.c
@@ -81,7 +81,7 @@ static int nr_copy_pages_check;
 extern char resume_file[];
 
 /* Local variables that should not be affected by save */
-unsigned int nr_copy_pages __nosavedata = 0;
+static unsigned int nr_copy_pages __nosavedata = 0;
 
 /* Suspend pagedir is allocated before final copy, therefore it
    must be freed after resume 
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 058189684c7..86073df418f 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -92,6 +92,17 @@ static sctp_disposition_t sctp_sf_shut_8_4_5(const struct sctp_endpoint *ep,
 					     sctp_cmd_seq_t *commands);
 static struct sctp_sackhdr *sctp_sm_pull_sack(struct sctp_chunk *chunk);
 
+static sctp_disposition_t sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands,
+					   __u16 error,
+					   const struct sctp_association *asoc,
+					   struct sctp_transport *transport);
+
+static sctp_disposition_t sctp_sf_violation_chunklen(
+				     const struct sctp_endpoint *ep,
+				     const struct sctp_association *asoc,
+				     const sctp_subtype_t type,
+				     void *arg,
+				     sctp_cmd_seq_t *commands);
 
 /* Small helper function that checks if the chunk length
  * is of the appropriate length.  The 'required_length' argument
@@ -2328,7 +2339,7 @@ sctp_disposition_t sctp_sf_cookie_echoed_abort(const struct sctp_endpoint *ep,
  *
  * This is common code called by several sctp_sf_*_abort() functions above.
  */
-sctp_disposition_t  sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands,
+static sctp_disposition_t sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands,
 					   __u16 error,
 					   const struct sctp_association *asoc,
 					   struct sctp_transport *transport)
@@ -3687,7 +3698,8 @@ sctp_disposition_t sctp_sf_violation(const struct sctp_endpoint *ep,
  *
  * Generate an  ABORT chunk and terminate the association.
  */
-sctp_disposition_t sctp_sf_violation_chunklen(const struct sctp_endpoint *ep,
+static sctp_disposition_t sctp_sf_violation_chunklen(
+				     const struct sctp_endpoint *ep,
 				     const struct sctp_association *asoc,
 				     const sctp_subtype_t type,
 				     void *arg,
-- 
cgit v1.2.3-70-g09d2


From 793ae77469121227cd910c4b99f24be1de34bcca Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@ppc970.osdl.org>
Date: Fri, 24 Jun 2005 10:39:17 -0700
Subject: Add "memory" clobbers to the x86 inline asm of strncmp and friends

They don't actually clobber memory, but gcc doesn't even know they
_read_ memory, so can apparently re-order memory accesses around them.

Which obviously does the wrong thing if the memory access happens to
change the memory that the compare function is accessing..

Verified to fix a strange boot problem by Jens Axboe.
---
 include/asm-i386/string.h | 32 ++++++++++++++++++++++----------
 1 file changed, 22 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/string.h b/include/asm-i386/string.h
index 6a78ac58c19..02c8f5d2206 100644
--- a/include/asm-i386/string.h
+++ b/include/asm-i386/string.h
@@ -116,7 +116,8 @@ __asm__ __volatile__(
 	"orb $1,%%al\n"
 	"3:"
 	:"=a" (__res), "=&S" (d0), "=&D" (d1)
-		     :"1" (cs),"2" (ct));
+	:"1" (cs),"2" (ct)
+	:"memory");
 return __res;
 }
 
@@ -138,8 +139,9 @@ __asm__ __volatile__(
 	"3:\tsbbl %%eax,%%eax\n\t"
 	"orb $1,%%al\n"
 	"4:"
-		     :"=a" (__res), "=&S" (d0), "=&D" (d1), "=&c" (d2)
-		     :"1" (cs),"2" (ct),"3" (count));
+	:"=a" (__res), "=&S" (d0), "=&D" (d1), "=&c" (d2)
+	:"1" (cs),"2" (ct),"3" (count)
+	:"memory");
 return __res;
 }
 
@@ -158,7 +160,9 @@ __asm__ __volatile__(
 	"movl $1,%1\n"
 	"2:\tmovl %1,%0\n\t"
 	"decl %0"
-	:"=a" (__res), "=&S" (d0) : "1" (s),"0" (c));
+	:"=a" (__res), "=&S" (d0)
+	:"1" (s),"0" (c)
+	:"memory");
 return __res;
 }
 
@@ -175,7 +179,9 @@ __asm__ __volatile__(
 	"leal -1(%%esi),%0\n"
 	"2:\ttestb %%al,%%al\n\t"
 	"jne 1b"
-	:"=g" (__res), "=&S" (d0), "=&a" (d1) :"0" (0),"1" (s),"2" (c));
+	:"=g" (__res), "=&S" (d0), "=&a" (d1)
+	:"0" (0),"1" (s),"2" (c)
+	:"memory");
 return __res;
 }
 
@@ -189,7 +195,9 @@ __asm__ __volatile__(
 	"scasb\n\t"
 	"notl %0\n\t"
 	"decl %0"
-	:"=c" (__res), "=&D" (d0) :"1" (s),"a" (0), "0" (0xffffffffu));
+	:"=c" (__res), "=&D" (d0)
+	:"1" (s),"a" (0), "0" (0xffffffffu)
+	:"memory");
 return __res;
 }
 
@@ -333,7 +341,9 @@ __asm__ __volatile__(
 	"je 1f\n\t"
 	"movl $1,%0\n"
 	"1:\tdecl %0"
-	:"=D" (__res), "=&c" (d0) : "a" (c),"0" (cs),"1" (count));
+	:"=D" (__res), "=&c" (d0)
+	:"a" (c),"0" (cs),"1" (count)
+	:"memory");
 return __res;
 }
 
@@ -369,7 +379,7 @@ __asm__ __volatile__(
 	"je 2f\n\t"
 	"stosb\n"
 	"2:"
-	: "=&c" (d0), "=&D" (d1)
+	:"=&c" (d0), "=&D" (d1)
 	:"a" (c), "q" (count), "0" (count/4), "1" ((long) s)
 	:"memory");
 return (s);	
@@ -392,7 +402,8 @@ __asm__ __volatile__(
 	"jne 1b\n"
 	"3:\tsubl %2,%0"
 	:"=a" (__res), "=&d" (d0)
-	:"c" (s),"1" (count));
+	:"c" (s),"1" (count)
+	:"memory");
 return __res;
 }
 /* end of additional stuff */
@@ -473,7 +484,8 @@ static inline void * memscan(void * addr, int c, size_t size)
 		"dec %%edi\n"
 		"1:"
 		: "=D" (addr), "=c" (size)
-		: "0" (addr), "1" (size), "a" (c));
+		: "0" (addr), "1" (size), "a" (c)
+		: "memory");
 	return addr;
 }
 
-- 
cgit v1.2.3-70-g09d2


From c6b56949de86694d837750a0a89c766b9871e81c Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Fri, 24 Jun 2005 20:54:34 +0100
Subject: [PATCH] ARM: 2740/1: ixp2000 align{b,w} need to parenthesize their
 arguments

Patch from Lennert Buytenhek

Two macros that are used on the ixp2000 to fixup byte lane enables
for I/O space accesses, align{b,w}, use their arguments without
parenthesizing them.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-ixp2000/io.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-ixp2000/io.h b/include/asm-arm/arch-ixp2000/io.h
index 083462668e1..e5c742bc233 100644
--- a/include/asm-arm/arch-ixp2000/io.h
+++ b/include/asm-arm/arch-ixp2000/io.h
@@ -27,8 +27,8 @@
  * since that isn't available on the A? revisions we just keep doing
  * things manually.
  */
-#define alignb(addr)		(void __iomem *)((unsigned long)addr ^ 3)
-#define alignw(addr)		(void __iomem *)((unsigned long)addr ^ 2)
+#define alignb(addr)		(void __iomem *)((unsigned long)(addr) ^ 3)
+#define alignw(addr)		(void __iomem *)((unsigned long)(addr) ^ 2)
 
 #define outb(v,p)		__raw_writeb((v),alignb(___io(p)))
 #define outw(v,p)		__raw_writew((v),alignw(___io(p)))
-- 
cgit v1.2.3-70-g09d2


From c4982887cacf2122bc256e901598b58caf4a34be Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Fri, 24 Jun 2005 20:54:35 +0100
Subject: [PATCH] ARM: 2744/1: ixp2000 gpio irq support

Patch from Lennert Buytenhek

This patch cleans up the ixp2000 gpio irq code and implements the
set_irq_type method for gpio irqs so that users can select for which
events (falling edge/rising edge/level low/level high) on the gpio
pin they want the corresponding gpio irq to be triggered.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Deepak Saxena
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-ixp2000/core.c            | 85 ++++++++++++++++++++++++++-------
 drivers/i2c/busses/i2c-ixp2000.c        |  3 +-
 include/asm-arm/arch-ixp2000/gpio.h     | 31 +++++-------
 include/asm-arm/arch-ixp2000/platform.h | 22 +--------
 4 files changed, 82 insertions(+), 59 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-ixp2000/core.c b/arch/arm/mach-ixp2000/core.c
index fc0555596d6..0ee34acb8d7 100644
--- a/arch/arm/mach-ixp2000/core.c
+++ b/arch/arm/mach-ixp2000/core.c
@@ -40,6 +40,8 @@
 #include <asm/mach/time.h>
 #include <asm/mach/irq.h>
 
+#include <asm/arch/gpio.h>
+
 static DEFINE_SPINLOCK(ixp2000_slowport_lock);
 static unsigned long ixp2000_slowport_irq_flags;
 
@@ -179,7 +181,7 @@ static int ixp2000_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 	/* clear timer 1 */
 	ixp2000_reg_write(IXP2000_T1_CLR, 1);
-	
+
 	while ((next_jiffy_time - *missing_jiffy_timer_csr) > ticks_per_jiffy) {
 		timer_tick(regs);
 		next_jiffy_time -= ticks_per_jiffy;
@@ -238,35 +240,40 @@ void __init ixp2000_init_time(unsigned long tick_rate)
 /*************************************************************************
  * GPIO helpers
  *************************************************************************/
-static unsigned long GPIO_IRQ_rising_edge;
 static unsigned long GPIO_IRQ_falling_edge;
+static unsigned long GPIO_IRQ_rising_edge;
 static unsigned long GPIO_IRQ_level_low;
 static unsigned long GPIO_IRQ_level_high;
 
-void gpio_line_config(int line, int style)
+static void update_gpio_int_csrs(void)
+{
+	ixp2000_reg_write(IXP2000_GPIO_FEDR, GPIO_IRQ_falling_edge);
+	ixp2000_reg_write(IXP2000_GPIO_REDR, GPIO_IRQ_rising_edge);
+	ixp2000_reg_write(IXP2000_GPIO_LSLR, GPIO_IRQ_level_low);
+	ixp2000_reg_write(IXP2000_GPIO_LSHR, GPIO_IRQ_level_high);
+}
+
+void gpio_line_config(int line, int direction)
 {
 	unsigned long flags;
 
 	local_irq_save(flags);
+	if (direction == GPIO_OUT) {
+		irq_desc[line + IRQ_IXP2000_GPIO0].valid = 0;
 
-	if(style == GPIO_OUT) {
 		/* if it's an output, it ain't an interrupt anymore */
-		ixp2000_reg_write(IXP2000_GPIO_PDSR, (1 << line));
 		GPIO_IRQ_falling_edge &= ~(1 << line);
 		GPIO_IRQ_rising_edge &= ~(1 << line);
 		GPIO_IRQ_level_low &= ~(1 << line);
 		GPIO_IRQ_level_high &= ~(1 << line);
-		ixp2000_reg_write(IXP2000_GPIO_FEDR, GPIO_IRQ_falling_edge);
-		ixp2000_reg_write(IXP2000_GPIO_REDR, GPIO_IRQ_rising_edge);
-		ixp2000_reg_write(IXP2000_GPIO_LSHR, GPIO_IRQ_level_high);
-		ixp2000_reg_write(IXP2000_GPIO_LSLR, GPIO_IRQ_level_low);
-		irq_desc[line+IRQ_IXP2000_GPIO0].valid = 0;
-	} else if(style == GPIO_IN) {
-		ixp2000_reg_write(IXP2000_GPIO_PDCR, (1 << line));
+		update_gpio_int_csrs();
+
+		ixp2000_reg_write(IXP2000_GPIO_PDSR, 1 << line);
+	} else if (direction == GPIO_IN) {
+		ixp2000_reg_write(IXP2000_GPIO_PDCR, 1 << line);
 	}
-		
 	local_irq_restore(flags);
-}	
+}
 
 
 /*************************************************************************
@@ -285,9 +292,50 @@ static void ixp2000_GPIO_irq_handler(unsigned int irq, struct irqdesc *desc, str
 	}
 }
 
+static int ixp2000_GPIO_irq_type(unsigned int irq, unsigned int type)
+{
+	int line = irq - IRQ_IXP2000_GPIO0;
+
+	/*
+	 * First, configure this GPIO line as an input.
+	 */
+	ixp2000_reg_write(IXP2000_GPIO_PDCR, 1 << line);
+
+	/*
+	 * Then, set the proper trigger type.
+	 */
+	if (type & IRQT_FALLING)
+		GPIO_IRQ_falling_edge |= 1 << line;
+	else
+		GPIO_IRQ_falling_edge &= ~(1 << line);
+	if (type & IRQT_RISING)
+		GPIO_IRQ_rising_edge |= 1 << line;
+	else
+		GPIO_IRQ_rising_edge &= ~(1 << line);
+	if (type & IRQT_LOW)
+		GPIO_IRQ_level_low |= 1 << line;
+	else
+		GPIO_IRQ_level_low &= ~(1 << line);
+	if (type & IRQT_HIGH)
+		GPIO_IRQ_level_high |= 1 << line;
+	else
+		GPIO_IRQ_level_high &= ~(1 << line);
+	update_gpio_int_csrs();
+
+	/*
+	 * Finally, mark the corresponding IRQ as valid.
+	 */
+	irq_desc[irq].valid = 1;
+
+	return 0;
+}
+
 static void ixp2000_GPIO_irq_mask_ack(unsigned int irq)
 {
 	ixp2000_reg_write(IXP2000_GPIO_INCR, (1 << (irq - IRQ_IXP2000_GPIO0)));
+
+	ixp2000_reg_write(IXP2000_GPIO_EDSR, (1 << (irq - IRQ_IXP2000_GPIO0)));
+	ixp2000_reg_write(IXP2000_GPIO_LDSR, (1 << (irq - IRQ_IXP2000_GPIO0)));
 	ixp2000_reg_write(IXP2000_GPIO_INST, (1 << (irq - IRQ_IXP2000_GPIO0)));
 }
 
@@ -302,6 +350,7 @@ static void ixp2000_GPIO_irq_unmask(unsigned int irq)
 }
 
 static struct irqchip ixp2000_GPIO_irq_chip = {
+	.type	= ixp2000_GPIO_irq_type,
 	.ack	= ixp2000_GPIO_irq_mask_ack,
 	.mask	= ixp2000_GPIO_irq_mask,
 	.unmask	= ixp2000_GPIO_irq_unmask
@@ -338,7 +387,7 @@ static void ixp2000_irq_mask(unsigned int irq)
 
 static void ixp2000_irq_unmask(unsigned int irq)
 {
-	ixp2000_reg_write(IXP2000_IRQ_ENABLE_SET,  (1 << irq));
+	ixp2000_reg_write(IXP2000_IRQ_ENABLE_SET, (1 << irq));
 }
 
 static struct irqchip ixp2000_irq_chip = {
@@ -375,16 +424,16 @@ void __init ixp2000_init_irq(void)
 	 * our mask/unmask code much simpler.
 	 */
 	for (irq = IRQ_IXP2000_SOFT_INT; irq <= IRQ_IXP2000_THDB3; irq++) {
-		if((1 << irq) & IXP2000_VALID_IRQ_MASK) {
+		if ((1 << irq) & IXP2000_VALID_IRQ_MASK) {
 			set_irq_chip(irq, &ixp2000_irq_chip);
 			set_irq_handler(irq, do_level_IRQ);
 			set_irq_flags(irq, IRQF_VALID);
 		} else set_irq_flags(irq, 0);
 	}
-	
+
 	/*
 	 * GPIO IRQs are invalid until someone sets the interrupt mode
-	 * by calling gpio_line_set();
+	 * by calling set_irq_type().
 	 */
 	for (irq = IRQ_IXP2000_GPIO0; irq <= IRQ_IXP2000_GPIO7; irq++) {
 		set_irq_chip(irq, &ixp2000_GPIO_irq_chip);
diff --git a/drivers/i2c/busses/i2c-ixp2000.c b/drivers/i2c/busses/i2c-ixp2000.c
index ec943cad231..1956af382cd 100644
--- a/drivers/i2c/busses/i2c-ixp2000.c
+++ b/drivers/i2c/busses/i2c-ixp2000.c
@@ -33,7 +33,8 @@
 #include <linux/i2c.h>
 #include <linux/i2c-algo-bit.h>
 
-#include <asm/hardware.h>	/* Pick up IXP42000-specific bits */
+#include <asm/hardware.h>	/* Pick up IXP2000-specific bits */
+#include <asm/arch/gpio.h>
 
 static inline int ixp2000_scl_pin(void *data)
 {
diff --git a/include/asm-arm/arch-ixp2000/gpio.h b/include/asm-arm/arch-ixp2000/gpio.h
index 84634af5cc6..03cbbe1fd9d 100644
--- a/include/asm-arm/arch-ixp2000/gpio.h
+++ b/include/asm-arm/arch-ixp2000/gpio.h
@@ -1,5 +1,5 @@
 /*
- * include/asm-arm/arch-ixp2000/ixp2000-gpio.h
+ * include/asm-arm/arch-ixp2000/gpio.h
  *
  * Copyright (C) 2002 Intel Corporation.
  *
@@ -16,26 +16,18 @@
  * Use this instead of directly setting the GPIO registers.
  * GPIOs may also be used as GPIOs (e.g. for emulating i2c/smb)
  */
-#ifndef _ASM_ARCH_IXP2000_GPIO_H_
-#define _ASM_ARCH_IXP2000_GPIO_H_
+#ifndef __ASM_ARCH_GPIO_H
+#define __ASM_ARCH_GPIO_H
 
 #ifndef __ASSEMBLY__
-#define GPIO_OUT			0x0
-#define GPIO_IN				0x80
+
+#define GPIO_IN				0
+#define GPIO_OUT			1
 
 #define IXP2000_GPIO_LOW		0
 #define IXP2000_GPIO_HIGH		1
 
-#define GPIO_NO_EDGES           	0
-#define GPIO_FALLING_EDGE       	1
-#define GPIO_RISING_EDGE        	2
-#define GPIO_BOTH_EDGES         	3
-#define GPIO_LEVEL_LOW          	4
-#define GPIO_LEVEL_HIGH         	8
-
-extern void set_GPIO_IRQ_edge(int gpio_nr, int edge);
-extern void set_GPIO_IRQ_level(int gpio_nr, int level);
-extern void gpio_line_config(int line, int style);
+extern void gpio_line_config(int line, int direction);
 
 static inline int gpio_line_get(int line)
 {
@@ -45,11 +37,12 @@ static inline int gpio_line_get(int line)
 static inline void gpio_line_set(int line, int value)
 {
 	if (value == IXP2000_GPIO_HIGH) {
-		ixp_reg_write(IXP2000_GPIO_POSR, BIT(line));
-	} else if (value == IXP2000_GPIO_LOW)
-		ixp_reg_write(IXP2000_GPIO_POCR, BIT(line));
+		ixp2000_reg_write(IXP2000_GPIO_POSR, 1 << line);
+	} else if (value == IXP2000_GPIO_LOW) {
+		ixp2000_reg_write(IXP2000_GPIO_POCR, 1 << line);
+	}
 }
 
 #endif /* !__ASSEMBLY__ */
-#endif /* ASM_ARCH_IXP2000_GPIO_H_ */
 
+#endif /* ASM_ARCH_IXP2000_GPIO_H_ */
diff --git a/include/asm-arm/arch-ixp2000/platform.h b/include/asm-arm/arch-ixp2000/platform.h
index 901bba6d02b..52ded516ea5 100644
--- a/include/asm-arm/arch-ixp2000/platform.h
+++ b/include/asm-arm/arch-ixp2000/platform.h
@@ -138,30 +138,10 @@ struct ixp2000_flash_data {
 	unsigned long (*bank_setup)(unsigned long);
 };
 
-/*
- * GPIO helper functions
- */
-#define	GPIO_IN		0
-#define	GPIO_OUT	1
-
-extern void gpio_line_config(int line, int style);
-
-static inline int gpio_line_get(int line)
-{
-	return (((*IXP2000_GPIO_PLR) >> line) & 1);
-}
-
-static inline void gpio_line_set(int line, int value)
-{
-	if (value) 
-		ixp2000_reg_write(IXP2000_GPIO_POSR, (1 << line));
-	else 
-		ixp2000_reg_write(IXP2000_GPIO_POCR, (1 << line));
-}
-
 struct ixp2000_i2c_pins {
 	unsigned long sda_pin;
 	unsigned long scl_pin;
 };
 
+
 #endif /*  !__ASSEMBLY__ */
-- 
cgit v1.2.3-70-g09d2


From 5932ae3f5d610fd8d047ef4693bab9f084e5c56d Mon Sep 17 00:00:00 2001
From: Deepak Saxena <dsaxena@plexity.net>
Date: Fri, 24 Jun 2005 20:54:35 +0100
Subject: [PATCH] ARM: 2745/1: Fix IXP4xx debug macros

Patch from Deepak Saxena

Current IXP4xx debug macros do not work in the small window between
the MMU being enabled and the call to map_io() b/c the standard
peripheral mapping is not properly setup for use with the low-level
debug code. This patch creates a new section-aligned mapping for the
UART specifically for use with the debug macros.

Signed-off-by: Deepak Saxena
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-ixp4xx/common.c             |  8 ++++++++
 include/asm-arm/arch-ixp4xx/debug-macro.S |  1 +
 include/asm-arm/arch-ixp4xx/ixp4xx-regs.h | 10 ++++++++++
 3 files changed, 19 insertions(+)

(limited to 'include')

diff --git a/arch/arm/mach-ixp4xx/common.c b/arch/arm/mach-ixp4xx/common.c
index 267ba02d77d..f39e8408488 100644
--- a/arch/arm/mach-ixp4xx/common.c
+++ b/arch/arm/mach-ixp4xx/common.c
@@ -141,7 +141,15 @@ static struct map_desc ixp4xx_io_desc[] __initdata = {
 		.physical	= IXP4XX_PCI_CFG_BASE_PHYS,
 		.length		= IXP4XX_PCI_CFG_REGION_SIZE,
 		.type		= MT_DEVICE
+	},
+#ifdef CONFIG_DEBUG_LL
+	{	/* Debug UART mapping */
+		.virtual	= IXP4XX_DEBUG_UART_BASE_VIRT,
+		.physical	= IXP4XX_DEBUG_UART_BASE_PHYS,
+		.length		= IXP4XX_DEBUG_UART_REGION_SIZE,
+		.type		= MT_DEVICE
 	}
+#endif
 };
 
 void __init ixp4xx_map_io(void)
diff --git a/include/asm-arm/arch-ixp4xx/debug-macro.S b/include/asm-arm/arch-ixp4xx/debug-macro.S
index 4499ae8e4b4..45a6c6cc29d 100644
--- a/include/asm-arm/arch-ixp4xx/debug-macro.S
+++ b/include/asm-arm/arch-ixp4xx/debug-macro.S
@@ -14,6 +14,7 @@
                 mrc     p15, 0, \rx, c1, c0
                 tst     \rx, #1                 @ MMU enabled?
                 moveq   \rx, #0xc8000000
+		orrne	\rx, \rx, #0x00b00000
                 movne   \rx, #0xff000000
                 add     \rx,\rx,#3              @ Uart regs are at off set of 3 if
 						@ byte writes used - Big Endian.
diff --git a/include/asm-arm/arch-ixp4xx/ixp4xx-regs.h b/include/asm-arm/arch-ixp4xx/ixp4xx-regs.h
index 8eeb1db6309..004696a95bd 100644
--- a/include/asm-arm/arch-ixp4xx/ixp4xx-regs.h
+++ b/include/asm-arm/arch-ixp4xx/ixp4xx-regs.h
@@ -69,6 +69,16 @@
 #define IXP4XX_PERIPHERAL_BASE_VIRT	(0xFFBF2000)
 #define IXP4XX_PERIPHERAL_REGION_SIZE	(0x0000C000)
 
+/*
+ * Debug UART
+ *
+ * This is basically a remap of UART1 into a region that is section
+ * aligned so that it * can be used with the low-level debug code.
+ */
+#define	IXP4XX_DEBUG_UART_BASE_PHYS	(0xC8000000)
+#define	IXP4XX_DEBUG_UART_BASE_VIRT	(0xffb00000)
+#define	IXP4XX_DEBUG_UART_REGION_SIZE	(0x00001000)
+
 #define IXP4XX_EXP_CS0_OFFSET	0x00
 #define IXP4XX_EXP_CS1_OFFSET   0x04
 #define IXP4XX_EXP_CS2_OFFSET   0x08
-- 
cgit v1.2.3-70-g09d2


From 75043cb5b386e5a01fd03b88f647dd992de02f97 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@mail.ru>
Date: Fri, 24 Jun 2005 20:52:52 +0000
Subject: [PATCH] fs/qnx4/*: fix sparse warnings

This patch fixes sparse warnings in the qnx4fs (and might even make
qnx4fs work on big-endian boxes)

Signed-off-by: Alexey Dobriyan <adobriyan@mail.ru>
Signed-off-by: Domen Puncer <domen@coderock.org>
Signed-off-by: Anders Larsen <al@alarsen.net>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/qnx4/dir.c            |  2 +-
 fs/qnx4/inode.c          |  4 ++--
 include/linux/qnx4_fs.h  | 18 +++++++++---------
 include/linux/qnxtypes.h | 16 ++++++++--------
 4 files changed, 20 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/fs/qnx4/dir.c b/fs/qnx4/dir.c
index cd66147cca0..7a8f5595c26 100644
--- a/fs/qnx4/dir.c
+++ b/fs/qnx4/dir.c
@@ -61,7 +61,7 @@ static int qnx4_readdir(struct file *filp, void *dirent, filldir_t filldir)
 						ino = blknum * QNX4_INODES_PER_BLOCK + ix - 1;
 					else {
 						le  = (struct qnx4_link_info*)de;
-						ino = ( le->dl_inode_blk - 1 ) *
+						ino = ( le32_to_cpu(le->dl_inode_blk) - 1 ) *
 							QNX4_INODES_PER_BLOCK +
 							le->dl_inode_ndx;
 					}
diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c
index aa92d6b76a9..b79162a3547 100644
--- a/fs/qnx4/inode.c
+++ b/fs/qnx4/inode.c
@@ -236,7 +236,7 @@ unsigned long qnx4_block_map( struct inode *inode, long iblock )
 	struct buffer_head *bh = NULL;
 	struct qnx4_xblk *xblk = NULL;
 	struct qnx4_inode_entry *qnx4_inode = qnx4_raw_inode(inode);
-	qnx4_nxtnt_t nxtnt = le16_to_cpu(qnx4_inode->di_num_xtnts);
+	u16 nxtnt = le16_to_cpu(qnx4_inode->di_num_xtnts);
 
 	if ( iblock < le32_to_cpu(qnx4_inode->di_first_xtnt.xtnt_size) ) {
 		// iblock is in the first extent. This is easy.
@@ -372,7 +372,7 @@ static int qnx4_fill_super(struct super_block *s, void *data, int silent)
 		printk("qnx4: unable to read the superblock\n");
 		goto outnobh;
 	}
-	if ( le32_to_cpu( *(__u32*)bh->b_data ) != QNX4_SUPER_MAGIC ) {
+	if ( le32_to_cpup((__le32*) bh->b_data) != QNX4_SUPER_MAGIC ) {
 		if (!silent)
 			printk("qnx4: wrong fsid in superblock.\n");
 		goto out;
diff --git a/include/linux/qnx4_fs.h b/include/linux/qnx4_fs.h
index 22ba580b0ae..fc610bb0f73 100644
--- a/include/linux/qnx4_fs.h
+++ b/include/linux/qnx4_fs.h
@@ -46,11 +46,11 @@ struct qnx4_inode_entry {
 	char		di_fname[QNX4_SHORT_NAME_MAX];
 	qnx4_off_t	di_size;
 	qnx4_xtnt_t	di_first_xtnt;
-	__u32		di_xblk;
-	__s32		di_ftime;
-	__s32		di_mtime;
-	__s32		di_atime;
-	__s32		di_ctime;
+	__le32		di_xblk;
+	__le32		di_ftime;
+	__le32		di_mtime;
+	__le32		di_atime;
+	__le32		di_ctime;
 	qnx4_nxtnt_t	di_num_xtnts;
 	qnx4_mode_t	di_mode;
 	qnx4_muid_t	di_uid;
@@ -63,18 +63,18 @@ struct qnx4_inode_entry {
 
 struct qnx4_link_info {
 	char		dl_fname[QNX4_NAME_MAX];
-	__u32		dl_inode_blk;
+	__le32		dl_inode_blk;
 	__u8		dl_inode_ndx;
 	__u8		dl_spare[10];
 	__u8		dl_status;
 };
 
 struct qnx4_xblk {
-	__u32		xblk_next_xblk;
-	__u32		xblk_prev_xblk;
+	__le32		xblk_next_xblk;
+	__le32		xblk_prev_xblk;
 	__u8		xblk_num_xtnts;
 	__u8		xblk_spare[3];
-	__s32		xblk_num_blocks;
+	__le32		xblk_num_blocks;
 	qnx4_xtnt_t	xblk_xtnts[QNX4_MAX_XTNTS_PER_XBLK];
 	char		xblk_signature[8];
 	qnx4_xtnt_t	xblk_first_xtnt;
diff --git a/include/linux/qnxtypes.h b/include/linux/qnxtypes.h
index fb518e318c7..a3eb1137857 100644
--- a/include/linux/qnxtypes.h
+++ b/include/linux/qnxtypes.h
@@ -12,18 +12,18 @@
 #ifndef _QNX4TYPES_H
 #define _QNX4TYPES_H
 
-typedef __u16 qnx4_nxtnt_t;
+typedef __le16 qnx4_nxtnt_t;
 typedef __u8  qnx4_ftype_t;
 
 typedef struct {
-	__u32 xtnt_blk;
-	__u32 xtnt_size;
+	__le32 xtnt_blk;
+	__le32 xtnt_size;
 } qnx4_xtnt_t;
 
-typedef __u16 qnx4_mode_t;
-typedef __u16 qnx4_muid_t;
-typedef __u16 qnx4_mgid_t;
-typedef __u32 qnx4_off_t;
-typedef __u16 qnx4_nlink_t;
+typedef __le16 qnx4_mode_t;
+typedef __le16 qnx4_muid_t;
+typedef __le16 qnx4_mgid_t;
+typedef __le32 qnx4_off_t;
+typedef __le16 qnx4_nlink_t;
 
 #endif
-- 
cgit v1.2.3-70-g09d2


From 7533fca8e866ee7355ca53f1216e3fa4c718f991 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Fri, 24 Jun 2005 23:11:31 +0100
Subject: [PATCH] ARM: 2747/1: allow platforms to provide their own iomap
 implementation

Patch from Lennert Buytenhek

This patch conditionalises the io{read,write}{8,16,32} defines and the
prototypes for ioport_map/ioport_unmap in asm-arm/io.h on ioread8 not
already having been defined.  This is done so that platforms can provide
their own implementation of the iomap API, ixp2000 for example needs
this.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Deepak Saxena
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/io.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/asm-arm/io.h b/include/asm-arm/io.h
index 08a46302d26..cc4b5f5dbfc 100644
--- a/include/asm-arm/io.h
+++ b/include/asm-arm/io.h
@@ -275,6 +275,7 @@ extern void __iounmap(void __iomem *addr);
 /*
  * io{read,write}{8,16,32} macros
  */
+#ifndef ioread8
 #define ioread8(p)	({ unsigned int __v = __raw_readb(p); __v; })
 #define ioread16(p)	({ unsigned int __v = le16_to_cpu(__raw_readw(p)); __v; })
 #define ioread32(p)	({ unsigned int __v = le32_to_cpu(__raw_readl(p)); __v; })
@@ -293,6 +294,7 @@ extern void __iounmap(void __iomem *addr);
 
 extern void __iomem *ioport_map(unsigned long port, unsigned int nr);
 extern void ioport_unmap(void __iomem *addr);
+#endif
 
 struct pci_dev;
 
-- 
cgit v1.2.3-70-g09d2


From 2966207c7e5945947c4db3a48aa4fa819807c5be Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Fri, 24 Jun 2005 23:11:31 +0100
Subject: [PATCH] ARM: 2748/1: ixp2000 implementation of the iomap api

Patch from Lennert Buytenhek

A number of ixp2000 models have a bug where the byte lanes for PCI I/O
transactions are swapped.  We already work around this in our versions
of {in,out}{b,w,l}, but we also need to perform these workarounds in a
custom implementation of the new iomap API, provided in this patch.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Deepak Saxena
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-ixp2000/io.h | 72 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 72 insertions(+)

(limited to 'include')

diff --git a/include/asm-arm/arch-ixp2000/io.h b/include/asm-arm/arch-ixp2000/io.h
index e5c742bc233..5e56b47446e 100644
--- a/include/asm-arm/arch-ixp2000/io.h
+++ b/include/asm-arm/arch-ixp2000/io.h
@@ -48,6 +48,78 @@
 #define insw(p,d,l)		__raw_readsw(alignw(___io(p)),d,l)
 #define insl(p,d,l)		__raw_readsl(___io(p),d,l)
 
+#define __is_io_address(p)	((((unsigned long)(p)) & ~(IXP2000_PCI_IO_SIZE - 1)) == IXP2000_PCI_IO_VIRT_BASE)
+
+#define ioread8(p)						\
+	({							\
+		unsigned int __v;				\
+								\
+		if (__is_io_address(p)) {			\
+			__v = __raw_readb(alignb(p));		\
+		} else {					\
+			__v = __raw_readb(p);			\
+		}						\
+								\
+		__v;						\
+	})							\
+
+#define ioread16(p)						\
+	({							\
+		unsigned int __v;				\
+								\
+		if (__is_io_address(p)) {			\
+			__v = __raw_readw(alignw(p));		\
+		} else {					\
+			__v = le16_to_cpu(__raw_readw(p));	\
+		}						\
+								\
+		__v;						\
+	})
+
+#define ioread32(p)						\
+	({							\
+		unsigned int __v;				\
+								\
+		if (__is_io_address(p)) {			\
+			__v = __raw_readl(p);			\
+		} else {					\
+			__v = le32_to_cpu(__raw_readl(p));	\
+		}						\
+								\
+		 __v;						\
+	})
+
+#define iowrite8(v,p)						\
+	({							\
+		if (__is_io_address(p)) {			\
+			__raw_writeb((v), alignb(p));		\
+		} else {					\
+			__raw_writeb((v), p);			\
+		}						\
+	})
+
+#define iowrite16(v,p)						\
+	({							\
+		if (__is_io_address(p)) {			\
+			__raw_writew((v), alignw(p));		\
+		} else {					\
+			__raw_writew(cpu_to_le16(v), p);	\
+		}						\
+	})
+
+#define iowrite32(v,p)						\
+	({							\
+		if (__is_io_address(p)) {			\
+			__raw_writel((v), p);			\
+		} else {					\
+			__raw_writel(cpu_to_le32(v), p);	\
+		}						\
+	})
+
+#define ioport_map(port, nr)	___io(port)
+
+#define ioport_unmap(addr)
+
 
 #ifdef CONFIG_ARCH_IXDP2X01
 /*
-- 
cgit v1.2.3-70-g09d2


From e55c57e0b51c68d78845549505057169c6c3cba6 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Fri, 24 Jun 2005 20:11:03 -0700
Subject: [SPARC64]: Report any user access faults in termios accessors.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/asm-sparc64/termios.h | 78 +++++++++++++++++++++++--------------------
 1 file changed, 41 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/include/asm-sparc64/termios.h b/include/asm-sparc64/termios.h
index 8effce0da08..9777a9cca88 100644
--- a/include/asm-sparc64/termios.h
+++ b/include/asm-sparc64/termios.h
@@ -100,16 +100,17 @@ struct winsize {
 #define user_termio_to_kernel_termios(termios, termio) \
 ({ \
 	unsigned short tmp; \
-	get_user(tmp, &(termio)->c_iflag); \
+	int err; \
+	err = get_user(tmp, &(termio)->c_iflag); \
 	(termios)->c_iflag = (0xffff0000 & ((termios)->c_iflag)) | tmp; \
-	get_user(tmp, &(termio)->c_oflag); \
+	err |= get_user(tmp, &(termio)->c_oflag); \
 	(termios)->c_oflag = (0xffff0000 & ((termios)->c_oflag)) | tmp; \
-	get_user(tmp, &(termio)->c_cflag); \
+	err |= get_user(tmp, &(termio)->c_cflag); \
 	(termios)->c_cflag = (0xffff0000 & ((termios)->c_cflag)) | tmp; \
-	get_user(tmp, &(termio)->c_lflag); \
+	err |= get_user(tmp, &(termio)->c_lflag); \
 	(termios)->c_lflag = (0xffff0000 & ((termios)->c_lflag)) | tmp; \
-	copy_from_user((termios)->c_cc, (termio)->c_cc, NCC); \
-	0; \
+	err |= copy_from_user((termios)->c_cc, (termio)->c_cc, NCC); \
+	err; \
 })
 
 /*
@@ -119,53 +120,56 @@ struct winsize {
  */
 #define kernel_termios_to_user_termio(termio, termios) \
 ({ \
-	put_user((termios)->c_iflag, &(termio)->c_iflag); \
-	put_user((termios)->c_oflag, &(termio)->c_oflag); \
-	put_user((termios)->c_cflag, &(termio)->c_cflag); \
-	put_user((termios)->c_lflag, &(termio)->c_lflag); \
-	put_user((termios)->c_line,  &(termio)->c_line); \
-	copy_to_user((termio)->c_cc, (termios)->c_cc, NCC); \
+	int err; \
+	err  = put_user((termios)->c_iflag, &(termio)->c_iflag); \
+	err |= put_user((termios)->c_oflag, &(termio)->c_oflag); \
+	err |= put_user((termios)->c_cflag, &(termio)->c_cflag); \
+	err |= put_user((termios)->c_lflag, &(termio)->c_lflag); \
+	err |= put_user((termios)->c_line,  &(termio)->c_line); \
+	err |= copy_to_user((termio)->c_cc, (termios)->c_cc, NCC); \
 	if (!((termios)->c_lflag & ICANON)) { \
-		put_user((termios)->c_cc[VMIN], &(termio)->c_cc[_VMIN]); \
-		put_user((termios)->c_cc[VTIME], &(termio)->c_cc[_VTIME]); \
+		err |= put_user((termios)->c_cc[VMIN], &(termio)->c_cc[_VMIN]); \
+		err |= put_user((termios)->c_cc[VTIME], &(termio)->c_cc[_VTIME]); \
 	} \
-	0; \
+	err; \
 })
 
 #define user_termios_to_kernel_termios(k, u) \
 ({ \
-	get_user((k)->c_iflag, &(u)->c_iflag); \
-	get_user((k)->c_oflag, &(u)->c_oflag); \
-	get_user((k)->c_cflag, &(u)->c_cflag); \
-	get_user((k)->c_lflag, &(u)->c_lflag); \
-	get_user((k)->c_line,  &(u)->c_line); \
-	copy_from_user((k)->c_cc, (u)->c_cc, NCCS); \
+	int err; \
+	err  = get_user((k)->c_iflag, &(u)->c_iflag); \
+	err |= get_user((k)->c_oflag, &(u)->c_oflag); \
+	err |= get_user((k)->c_cflag, &(u)->c_cflag); \
+	err |= get_user((k)->c_lflag, &(u)->c_lflag); \
+	err |= get_user((k)->c_line,  &(u)->c_line); \
+	err |= copy_from_user((k)->c_cc, (u)->c_cc, NCCS); \
 	if((k)->c_lflag & ICANON) { \
-		get_user((k)->c_cc[VEOF], &(u)->c_cc[VEOF]); \
-		get_user((k)->c_cc[VEOL], &(u)->c_cc[VEOL]); \
+		err |= get_user((k)->c_cc[VEOF], &(u)->c_cc[VEOF]); \
+		err |= get_user((k)->c_cc[VEOL], &(u)->c_cc[VEOL]); \
 	} else { \
-		get_user((k)->c_cc[VMIN],  &(u)->c_cc[_VMIN]); \
-		get_user((k)->c_cc[VTIME], &(u)->c_cc[_VTIME]); \
+		err |= get_user((k)->c_cc[VMIN],  &(u)->c_cc[_VMIN]); \
+		err |= get_user((k)->c_cc[VTIME], &(u)->c_cc[_VTIME]); \
 	} \
-	0; \
+	err; \
 })
 
 #define kernel_termios_to_user_termios(u, k) \
 ({ \
-	put_user((k)->c_iflag, &(u)->c_iflag); \
-	put_user((k)->c_oflag, &(u)->c_oflag); \
-	put_user((k)->c_cflag, &(u)->c_cflag); \
-	put_user((k)->c_lflag, &(u)->c_lflag); \
-	put_user((k)->c_line, &(u)->c_line); \
-	copy_to_user((u)->c_cc, (k)->c_cc, NCCS); \
+	int err; \
+	err  = put_user((k)->c_iflag, &(u)->c_iflag); \
+	err |= put_user((k)->c_oflag, &(u)->c_oflag); \
+	err |= put_user((k)->c_cflag, &(u)->c_cflag); \
+	err |= put_user((k)->c_lflag, &(u)->c_lflag); \
+	err |= put_user((k)->c_line, &(u)->c_line); \
+	err |= copy_to_user((u)->c_cc, (k)->c_cc, NCCS); \
 	if(!((k)->c_lflag & ICANON)) { \
-		put_user((k)->c_cc[VMIN],  &(u)->c_cc[_VMIN]); \
-		put_user((k)->c_cc[VTIME], &(u)->c_cc[_VTIME]); \
+		err |= put_user((k)->c_cc[VMIN],  &(u)->c_cc[_VMIN]); \
+		err |= put_user((k)->c_cc[VTIME], &(u)->c_cc[_VTIME]); \
 	} else { \
-		put_user((k)->c_cc[VEOF], &(u)->c_cc[VEOF]); \
-		put_user((k)->c_cc[VEOL], &(u)->c_cc[VEOL]); \
+		err |= put_user((k)->c_cc[VEOF], &(u)->c_cc[VEOF]); \
+		err |= put_user((k)->c_cc[VEOL], &(u)->c_cc[VEOL]); \
 	} \
-	0; \
+	err; \
 })
 
 #endif	/* __KERNEL__ */
-- 
cgit v1.2.3-70-g09d2


From 321ab6a5fab812658626aee6bce2617f8cfb3a55 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Sat, 25 Jun 2005 19:30:04 +0100
Subject: [PATCH] ARM: 2752/1: disable ixp2000 PCI I/O software workaround on
 chips that don't need it

Patch from Lennert Buytenhek

The later ixp2000 models don't need the PCI I/O workaround that we
currently perform.  Add a config option to disable the workaround,
and panic on boot if a kernel without the workaround is booted on a
buggy chip.  As only pre-production ixp2000s need the workaround,
the default is for it not to be configured in.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Deepak Saxena
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/configs/enp2611_defconfig          |  1 +
 arch/arm/configs/ixdp2400_defconfig         |  1 +
 arch/arm/configs/ixdp2401_defconfig         |  1 +
 arch/arm/configs/ixdp2800_defconfig         |  1 +
 arch/arm/configs/ixdp2801_defconfig         |  1 +
 arch/arm/mach-ixp2000/Kconfig               |  8 ++++++++
 arch/arm/mach-ixp2000/pci.c                 | 13 +++++++++++++
 include/asm-arm/arch-ixp2000/io.h           | 20 ++++++++++++++------
 include/asm-arm/arch-ixp2000/ixp2000-regs.h |  2 +-
 9 files changed, 41 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/arm/configs/enp2611_defconfig b/arch/arm/configs/enp2611_defconfig
index af215342456..b8c51ee7f1b 100644
--- a/arch/arm/configs/enp2611_defconfig
+++ b/arch/arm/configs/enp2611_defconfig
@@ -99,6 +99,7 @@ CONFIG_ARCH_ENP2611=y
 # CONFIG_ARCH_IXDP2800 is not set
 # CONFIG_ARCH_IXDP2401 is not set
 # CONFIG_ARCH_IXDP2801 is not set
+# CONFIG_IXP2000_SUPPORT_BROKEN_PCI_IO is not set
 
 #
 # Processor Type
diff --git a/arch/arm/configs/ixdp2400_defconfig b/arch/arm/configs/ixdp2400_defconfig
index a7ee1a442f3..3cfbe2ec29c 100644
--- a/arch/arm/configs/ixdp2400_defconfig
+++ b/arch/arm/configs/ixdp2400_defconfig
@@ -100,6 +100,7 @@ CONFIG_ARCH_IXDP2400=y
 CONFIG_ARCH_IXDP2X00=y
 # CONFIG_ARCH_IXDP2401 is not set
 # CONFIG_ARCH_IXDP2801 is not set
+# CONFIG_IXP2000_SUPPORT_BROKEN_PCI_IO is not set
 
 #
 # Processor Type
diff --git a/arch/arm/configs/ixdp2401_defconfig b/arch/arm/configs/ixdp2401_defconfig
index 2da48fca5c1..5c87e8e6969 100644
--- a/arch/arm/configs/ixdp2401_defconfig
+++ b/arch/arm/configs/ixdp2401_defconfig
@@ -100,6 +100,7 @@ CONFIG_ARCH_SUPPORTS_BIG_ENDIAN=y
 CONFIG_ARCH_IXDP2401=y
 # CONFIG_ARCH_IXDP2801 is not set
 CONFIG_ARCH_IXDP2X01=y
+# CONFIG_IXP2000_SUPPORT_BROKEN_PCI_IO is not set
 
 #
 # Processor Type
diff --git a/arch/arm/configs/ixdp2800_defconfig b/arch/arm/configs/ixdp2800_defconfig
index 7ba867f751f..3cb561a551c 100644
--- a/arch/arm/configs/ixdp2800_defconfig
+++ b/arch/arm/configs/ixdp2800_defconfig
@@ -100,6 +100,7 @@ CONFIG_ARCH_IXDP2800=y
 CONFIG_ARCH_IXDP2X00=y
 # CONFIG_ARCH_IXDP2401 is not set
 # CONFIG_ARCH_IXDP2801 is not set
+# CONFIG_IXP2000_SUPPORT_BROKEN_PCI_IO is not set
 
 #
 # Processor Type
diff --git a/arch/arm/configs/ixdp2801_defconfig b/arch/arm/configs/ixdp2801_defconfig
index c4df0ec34b0..b1e162f29cb 100644
--- a/arch/arm/configs/ixdp2801_defconfig
+++ b/arch/arm/configs/ixdp2801_defconfig
@@ -100,6 +100,7 @@ CONFIG_ARCH_SUPPORTS_BIG_ENDIAN=y
 # CONFIG_ARCH_IXDP2401 is not set
 CONFIG_ARCH_IXDP2801=y
 CONFIG_ARCH_IXDP2X01=y
+# CONFIG_IXP2000_SUPPORT_BROKEN_PCI_IO is not set
 
 #
 # Processor Type
diff --git a/arch/arm/mach-ixp2000/Kconfig b/arch/arm/mach-ixp2000/Kconfig
index 9361e05f6fa..ecb58d83478 100644
--- a/arch/arm/mach-ixp2000/Kconfig
+++ b/arch/arm/mach-ixp2000/Kconfig
@@ -54,6 +54,14 @@ config ARCH_IXDP2X01
 	depends on ARCH_IXDP2401 || ARCH_IXDP2801
 	default y	
 
+config IXP2000_SUPPORT_BROKEN_PCI_IO
+	bool "Support broken PCI I/O on older IXP2000s"
+	default y
+	help
+	  Say 'N' here if you only intend to run your kernel on an
+	  IXP2000 B0 or later model and do not need the PCI I/O
+	  byteswap workaround.  Say 'Y' otherwise.
+
 endmenu
 
 endif
diff --git a/arch/arm/mach-ixp2000/pci.c b/arch/arm/mach-ixp2000/pci.c
index 5ff2f2718c5..0788fb2b5c1 100644
--- a/arch/arm/mach-ixp2000/pci.c
+++ b/arch/arm/mach-ixp2000/pci.c
@@ -198,6 +198,19 @@ clear_master_aborts(void)
 void __init
 ixp2000_pci_preinit(void)
 {
+#ifndef CONFIG_IXP2000_SUPPORT_BROKEN_PCI_IO
+	/*
+	 * Configure the PCI unit to properly byteswap I/O transactions,
+	 * and verify that it worked.
+	 */
+	ixp2000_reg_write(IXP2000_PCI_CONTROL,
+			  (*IXP2000_PCI_CONTROL | PCI_CONTROL_IEE));
+
+	if ((*IXP2000_PCI_CONTROL & PCI_CONTROL_IEE) == 0)
+		panic("IXP2000: PCI I/O is broken on this ixp model, and "
+			"the needed workaround has not been configured in");
+#endif
+
 	hook_fault_code(16+6, ixp2000_pci_abort_handler, SIGBUS,
 				"PCI config cycle to non-existent device");
 }
diff --git a/include/asm-arm/arch-ixp2000/io.h b/include/asm-arm/arch-ixp2000/io.h
index 5e56b47446e..3241cd6f077 100644
--- a/include/asm-arm/arch-ixp2000/io.h
+++ b/include/asm-arm/arch-ixp2000/io.h
@@ -17,16 +17,21 @@
 
 #define IO_SPACE_LIMIT		0xffffffff
 #define __mem_pci(a)		(a)
-#define ___io(p)		((void __iomem *)((p)+IXP2000_PCI_IO_VIRT_BASE))
 
 /*
- * The IXP2400 before revision B0 asserts byte lanes for PCI I/O
+ * The A? revisions of the IXP2000s assert byte lanes for PCI I/O
  * transactions the other way round (MEM transactions don't have this
- * issue), so we need to override the standard functions.  B0 and later
- * have a bit that can be set to 1 to get the 'proper' behavior, but
- * since that isn't available on the A? revisions we just keep doing
- * things manually.
+ * issue), so if we want to support those models, we need to override
+ * the standard I/O functions.
+ *
+ * B0 and later have a bit that can be set to 1 to get the proper
+ * behavior for I/O transactions, which then allows us to use the
+ * standard I/O functions.  This is what we do if the user does not
+ * explicitly ask for support for pre-B0.
  */
+#ifdef CONFIG_IXP2000_SUPPORT_BROKEN_PCI_IO
+#define ___io(p)		((void __iomem *)((p)+IXP2000_PCI_IO_VIRT_BASE))
+
 #define alignb(addr)		(void __iomem *)((unsigned long)(addr) ^ 3)
 #define alignw(addr)		(void __iomem *)((unsigned long)(addr) ^ 2)
 
@@ -119,6 +124,9 @@
 #define ioport_map(port, nr)	___io(port)
 
 #define ioport_unmap(addr)
+#else
+#define __io(p)			((void __iomem *)((p)+IXP2000_PCI_IO_VIRT_BASE))
+#endif
 
 
 #ifdef CONFIG_ARCH_IXDP2X01
diff --git a/include/asm-arm/arch-ixp2000/ixp2000-regs.h b/include/asm-arm/arch-ixp2000/ixp2000-regs.h
index a1d9e181b10..5eb47d4bfbf 100644
--- a/include/asm-arm/arch-ixp2000/ixp2000-regs.h
+++ b/include/asm-arm/arch-ixp2000/ixp2000-regs.h
@@ -241,7 +241,7 @@
 #define PCI_CONTROL_BE_DEI		(1 << 21)	/* Big Endian Data Enable In  */
 #define PCI_CONTROL_BE_BEO		(1 << 20)	/* Big Endian Byte Enable Out */
 #define PCI_CONTROL_BE_BEI		(1 << 19)	/* Big Endian Byte Enable In  */
-#define PCI_CONTROL_PNR			(1 << 17)	/* PCI Not Reset bit */
+#define PCI_CONTROL_IEE			(1 << 17)	/* I/O cycle Endian swap Enable */
 
 #define IXP2000_PCI_RST_REL		(1 << 2)
 #define CFG_RST_DIR			(*IXP2000_PCI_CONTROL & IXP2000_PCICNTL_PCF)
-- 
cgit v1.2.3-70-g09d2


From 8749af68216e1ebf6460992fce548f400ecf63a4 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Sat, 25 Jun 2005 19:39:45 +0100
Subject: [PATCH] ARM: Generic Dynamic Tick Timer support for ARM, take 4

This patch adds support for Dynamic Tick Timer for ARM. Dynamic Tick is
also known as VST (Variable Scheduling Timeouts).

Dynamic Tick has been in use in the OMAP tree since last October.  The
patch is not intrusive, and does not do anything unless CONFIG_NO_IDLE_HZ
is defined.  This patch has the following fixed based on comments from
RMK:
- Time is updated before calling interrupt handlers.
- Added new interrupt flag SA_TIMER to avoid duplicate timer interrupts
- Moved struct dyn_tick_timer to time.h until we at some point probably
  have an arch independent dyn-tick.h
- Cleaned up testing for DYN_TICK_ENABLED in irq.c

 I've cleaned up this patch to fix some remaining issues:
 - Call the timer tick handler with irqs disabled, as it would be from
   a normal interrupt
 - if we have a dyn_tick, we better implement all methods.
 - generic timer_dyn_reprogram() call, to be called before sleeping
 - added command line option - "dyntick=" to allow boot-time control
   of this feature
    -- rmk

Signed-off-by: Tony Lindgren
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig            |  15 +++++++
 arch/arm/kernel/irq.c       |  14 ++++++
 arch/arm/kernel/time.c      | 103 ++++++++++++++++++++++++++++++++++++++++++++
 include/asm-arm/mach/time.h |  21 +++++++++
 include/asm-arm/signal.h    |   1 +
 5 files changed, 154 insertions(+)

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index bfcf4228036..c8d94dcd8ef 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -346,6 +346,21 @@ config PREEMPT
 	  Say Y here if you are building a kernel for a desktop, embedded
 	  or real-time system.  Say N if you are unsure.
 
+config NO_IDLE_HZ
+	bool "Dynamic tick timer"
+	help
+	  Select this option if you want to disable continuous timer ticks
+	  and have them programmed to occur as required. This option saves
+	  power as the system can remain in idle state for longer.
+
+	  By default dynamic tick is disabled during the boot, and can be
+	  manually enabled with:
+
+	    echo 1 > /sys/devices/system/timer/timer0/dyn_tick
+
+	  Alternatively, if you want dynamic tick automatically enabled
+	  during boot, pass "dyntick=enable" via the kernel command string.
+
 config ARCH_DISCONTIGMEM_ENABLE
 	bool
 	default (ARCH_LH7A40X && !LH7A40X_CONTIGMEM)
diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c
index ff187f4308f..395137a8fad 100644
--- a/arch/arm/kernel/irq.c
+++ b/arch/arm/kernel/irq.c
@@ -4,6 +4,10 @@
  *  Copyright (C) 1992 Linus Torvalds
  *  Modifications for ARM processor Copyright (C) 1995-2000 Russell King.
  *
+ *  Support for Dynamic Tick Timer Copyright (C) 2004-2005 Nokia Corporation.
+ *  Dynamic Tick Timer written by Tony Lindgren <tony@atomide.com> and
+ *  Tuukka Tikkanen <tuukka.tikkanen@elektrobit.com>.
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
@@ -37,6 +41,7 @@
 #include <asm/irq.h>
 #include <asm/system.h>
 #include <asm/mach/irq.h>
+#include <asm/mach/time.h>
 
 /*
  * Maximum IRQ count.  Currently, this is arbitary.  However, it should
@@ -329,6 +334,15 @@ __do_irq(unsigned int irq, struct irqaction *action, struct pt_regs *regs)
 
 	spin_unlock(&irq_controller_lock);
 
+#ifdef CONFIG_NO_IDLE_HZ
+	if (!(action->flags & SA_TIMER) && system_timer->dyn_tick != NULL) {
+		write_seqlock(&xtime_lock);
+		if (system_timer->dyn_tick->state & DYN_TICK_ENABLED)
+			system_timer->dyn_tick->handler(irq, 0, regs);
+		write_sequnlock(&xtime_lock);
+	}
+#endif
+
 	if (!(action->flags & SA_INTERRUPT))
 		local_irq_enable();
 
diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c
index c232f24f4a6..06054c9ba07 100644
--- a/arch/arm/kernel/time.c
+++ b/arch/arm/kernel/time.c
@@ -381,6 +381,95 @@ static struct sysdev_class timer_sysclass = {
 	.resume		= timer_resume,
 };
 
+#ifdef CONFIG_NO_IDLE_HZ
+static int timer_dyn_tick_enable(void)
+{
+	struct dyn_tick_timer *dyn_tick = system_timer->dyn_tick;
+	unsigned long flags;
+	int ret = -ENODEV;
+
+	if (dyn_tick) {
+		write_seqlock_irqsave(&xtime_lock, flags);
+		ret = 0;
+		if (!(dyn_tick->state & DYN_TICK_ENABLED)) {
+			ret = dyn_tick->enable();
+
+			if (ret == 0)
+				dyn_tick->state |= DYN_TICK_ENABLED;
+		}
+		write_sequnlock_irqrestore(&xtime_lock, flags);
+	}
+
+	return ret;
+}
+
+static int timer_dyn_tick_disable(void)
+{
+	struct dyn_tick_timer *dyn_tick = system_timer->dyn_tick;
+	unsigned long flags;
+	int ret = -ENODEV;
+
+	if (dyn_tick) {
+		write_seqlock_irqsave(&xtime_lock, flags);
+		ret = 0;
+		if (dyn_tick->state & DYN_TICK_ENABLED) {
+			ret = dyn_tick->disable();
+
+			if (ret == 0)
+				dyn_tick->state &= ~DYN_TICK_ENABLED;
+		}
+		write_sequnlock_irqrestore(&xtime_lock, flags);
+	}
+
+	return ret;
+}
+
+void timer_dyn_reprogram(void)
+{
+	struct dyn_tick_timer *dyn_tick = system_timer->dyn_tick;
+	unsigned long flags;
+
+	write_seqlock_irqsave(&xtime_lock, flags);
+	if (dyn_tick->state & DYN_TICK_ENABLED)
+		dyn_tick->reprogram(next_timer_interrupt() - jiffies);
+	write_sequnlock_irqrestore(&xtime_lock, flags);
+}
+
+static ssize_t timer_show_dyn_tick(struct sys_device *dev, char *buf)
+{
+	return sprintf(buf, "%i\n",
+		       (system_timer->dyn_tick->state & DYN_TICK_ENABLED) >> 1);
+}
+
+static ssize_t timer_set_dyn_tick(struct sys_device *dev, const char *buf,
+				  size_t count)
+{
+	unsigned int enable = simple_strtoul(buf, NULL, 2);
+
+	if (enable)
+		timer_dyn_tick_enable();
+	else
+		timer_dyn_tick_disable();
+
+	return count;
+}
+static SYSDEV_ATTR(dyn_tick, 0644, timer_show_dyn_tick, timer_set_dyn_tick);
+
+/*
+ * dyntick=enable|disable
+ */
+static char dyntick_str[4] __initdata = "";
+
+static int __init dyntick_setup(char *str)
+{
+	if (str)
+		strlcpy(dyntick_str, str, sizeof(dyntick_str));
+	return 1;
+}
+
+__setup("dyntick=", dyntick_setup);
+#endif
+
 static int __init timer_init_sysfs(void)
 {
 	int ret = sysdev_class_register(&timer_sysclass);
@@ -388,6 +477,20 @@ static int __init timer_init_sysfs(void)
 		system_timer->dev.cls = &timer_sysclass;
 		ret = sysdev_register(&system_timer->dev);
 	}
+
+#ifdef CONFIG_NO_IDLE_HZ
+	if (ret == 0 && system_timer->dyn_tick) {
+		ret = sysdev_create_file(&system_timer->dev, &attr_dyn_tick);
+
+		/*
+		 * Turn on dynamic tick after calibrate delay
+		 * for correct bogomips
+		 */
+		if (ret == 0 && dyntick_str[0] == 'e')
+			ret = timer_dyn_tick_enable();
+	}
+#endif
+
 	return ret;
 }
 
diff --git a/include/asm-arm/mach/time.h b/include/asm-arm/mach/time.h
index 5cf4fd659fd..047980ad18d 100644
--- a/include/asm-arm/mach/time.h
+++ b/include/asm-arm/mach/time.h
@@ -39,8 +39,29 @@ struct sys_timer {
 	void			(*suspend)(void);
 	void			(*resume)(void);
 	unsigned long		(*offset)(void);
+
+#ifdef CONFIG_NO_IDLE_HZ
+	struct dyn_tick_timer	*dyn_tick;
+#endif
+};
+
+#ifdef CONFIG_NO_IDLE_HZ
+
+#define DYN_TICK_SKIPPING	(1 << 2)
+#define DYN_TICK_ENABLED	(1 << 1)
+#define DYN_TICK_SUITABLE	(1 << 0)
+
+struct dyn_tick_timer {
+	unsigned int	state;			/* Current state */
+	int		(*enable)(void);	/* Enables dynamic tick */
+	int		(*disable)(void);	/* Disables dynamic tick */
+	void		(*reprogram)(unsigned long); /* Reprograms the timer */
+	int		(*handler)(int, void *, struct pt_regs *);
 };
 
+void timer_dyn_reprogram(void);
+#endif
+
 extern struct sys_timer *system_timer;
 extern void timer_tick(struct pt_regs *);
 
diff --git a/include/asm-arm/signal.h b/include/asm-arm/signal.h
index 46e69ae395a..760f6e65af0 100644
--- a/include/asm-arm/signal.h
+++ b/include/asm-arm/signal.h
@@ -114,6 +114,7 @@ typedef unsigned long sigset_t;
 #define SIGSTKSZ	8192
 
 #ifdef __KERNEL__
+#define SA_TIMER		0x40000000
 #define SA_IRQNOMASK		0x08000000
 #endif
 
-- 
cgit v1.2.3-70-g09d2


From e70c9d5e61c6cb2272c866fc1303e62975006752 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor_core@ameritech.net>
Date: Sat, 25 Jun 2005 14:54:25 -0700
Subject: [PATCH] I8K: use standard DMI interface

I8K: Change to use stock dmi infrastructure instead of homegrown
     parsing code. The driver now requires box's DMI data to match
     list of supported models so driver can be safely compiled-in
     by default without fear of it poking into random SMM BIOS
     code. DMI checks can be ignored with i8k.ignore_dmi option.

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/kernel-parameters.txt |   3 +
 arch/i386/kernel/dmi_scan.c         |   6 +-
 drivers/char/i8k.c                  | 302 +++++++-----------------------------
 include/linux/dmi.h                 |   1 +
 4 files changed, 60 insertions(+), 252 deletions(-)

(limited to 'include')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 4924d387a65..86db43fd6b0 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -548,6 +548,9 @@ running once the system is up.
 
 	i810=		[HW,DRM]
 
+	i8k.ignore_dmi	[HW] Continue probing hardware even if DMI data
+			indicates that the driver is running on unsupported
+			hardware.
 	i8k.force	[HW] Activate i8k driver even if SMM BIOS signature
 			does not match list of supported models.
 	i8k.power_status
diff --git a/arch/i386/kernel/dmi_scan.c b/arch/i386/kernel/dmi_scan.c
index 6ed7e28f306..3facd20212b 100644
--- a/arch/i386/kernel/dmi_scan.c
+++ b/arch/i386/kernel/dmi_scan.c
@@ -414,6 +414,7 @@ static void __init dmi_decode(struct dmi_header *dm)
 			dmi_save_ident(dm, DMI_PRODUCT_VERSION, 6);
 			dmi_printk(("Serial Number: %s\n",
 				dmi_string(dm, data[7])));
+			dmi_save_ident(dm, DMI_PRODUCT_SERIAL, 7);
 			break;
 		case 2:
 			dmi_printk(("Board Vendor: %s\n",
@@ -470,7 +471,6 @@ fail:		d++;
 
 	return count;
 }
-
 EXPORT_SYMBOL(dmi_check_system);
 
 /**
@@ -480,8 +480,8 @@ EXPORT_SYMBOL(dmi_check_system);
  *	Returns one DMI data value, can be used to perform
  *	complex DMI data checks.
  */
-char * dmi_get_system_info(int field)
+char *dmi_get_system_info(int field)
 {
 	return dmi_ident[field];
 }
-
+EXPORT_SYMBOL(dmi_get_system_info);
diff --git a/drivers/char/i8k.c b/drivers/char/i8k.c
index bf5e43beca6..81d2f675fb7 100644
--- a/drivers/char/i8k.c
+++ b/drivers/char/i8k.c
@@ -20,7 +20,7 @@
 #include <linux/types.h>
 #include <linux/init.h>
 #include <linux/proc_fs.h>
-#include <linux/apm_bios.h>
+#include <linux/dmi.h>
 #include <asm/uaccess.h>
 #include <asm/io.h>
 
@@ -52,18 +52,7 @@
 
 #define I8K_TEMPERATURE_BUG	1
 
-#define DELL_SIGNATURE		"Dell Computer"
-
-static char *supported_models[] = {
-	"Inspiron",
-	"Latitude",
-	NULL
-};
-
-static char system_vendor[48] = "?";
-static char product_name[48] = "?";
-static char bios_version[4] = "?";
-static char serial_number[16] = "?";
+static char bios_version[4];
 
 MODULE_AUTHOR("Massimo Dal Zotto (dz@debian.org)");
 MODULE_DESCRIPTION("Driver for accessing SMM BIOS on Dell laptops");
@@ -73,6 +62,10 @@ static int force;
 module_param(force, bool, 0);
 MODULE_PARM_DESC(force, "Force loading without checking for supported models");
 
+static int ignore_dmi;
+module_param(ignore_dmi, bool, 0);
+MODULE_PARM_DESC(ignore_dmi, "Continue probing hardware even if DMI data does not match");
+
 static int restricted;
 module_param(restricted, bool, 0);
 MODULE_PARM_DESC(restricted, "Allow fan control if SYS_ADMIN capability set");
@@ -99,11 +92,10 @@ typedef struct {
 	unsigned int edi __attribute__ ((packed));
 } SMMRegisters;
 
-typedef struct {
-	u8 type;
-	u8 length;
-	u16 handle;
-} DMIHeader;
+static inline char *i8k_get_dmi_data(int field)
+{
+	return dmi_get_system_info(field) ? : "N/A";
+}
 
 /*
  * Call the System Management Mode BIOS. Code provided by Jonathan Buzzard.
@@ -162,15 +154,6 @@ static int i8k_get_bios_version(void)
 	return regs.eax;
 }
 
-/*
- * Read the machine id.
- */
-static int i8k_get_serial_number(unsigned char *buff)
-{
-	strlcpy(buff, serial_number, sizeof(serial_number));
-	return 0;
-}
-
 /*
  * Read the Fn key status.
  */
@@ -328,7 +311,7 @@ static int i8k_get_dell_signature(void)
 static int i8k_ioctl(struct inode *ip, struct file *fp, unsigned int cmd,
 		     unsigned long arg)
 {
-	int val;
+	int val = 0;
 	int speed;
 	unsigned char buff[16];
 	int __user *argp = (int __user *)arg;
@@ -343,7 +326,7 @@ static int i8k_ioctl(struct inode *ip, struct file *fp, unsigned int cmd,
 
 	case I8K_MACHINE_ID:
 		memset(buff, 0, 16);
-		val = i8k_get_serial_number(buff);
+		strlcpy(buff, i8k_get_dmi_data(DMI_PRODUCT_SERIAL), sizeof(buff));
 		break;
 
 	case I8K_FN_STATUS:
@@ -451,10 +434,10 @@ static int i8k_get_info(char *buffer, char **start, off_t fpos, int length)
 	n = sprintf(buffer, "%s %s %s %d %d %d %d %d %d %d\n",
 		    I8K_PROC_FMT,
 		    bios_version,
-		    serial_number,
+		    dmi_get_system_info(DMI_PRODUCT_SERIAL) ? : "N/A",
 		    cpu_temp,
-		    left_fan,
-		    right_fan, left_speed, right_speed, ac_power, fn_key);
+		    left_fan, right_fan, left_speed, right_speed,
+		    ac_power, fn_key);
 
 	return n;
 }
@@ -486,201 +469,23 @@ static ssize_t i8k_read(struct file *f, char __user * buffer, size_t len,
 	return len;
 }
 
-static char *__init string_trim(char *s, int size)
-{
-	int len;
-	char *p;
-
-	if ((len = strlen(s)) > size) {
-		len = size;
-	}
-
-	for (p = s + len - 1; len && (*p == ' '); len--, p--) {
-		*p = '\0';
-	}
-
-	return s;
-}
-
-/* DMI code, stolen from arch/i386/kernel/dmi_scan.c */
-
-/*
- * |<-- dmi->length -->|
- * |                   |
- * |dmi header    s=N  | string1,\0, ..., stringN,\0, ..., \0
- *                |                       |
- *                +-----------------------+
- */
-static char *__init dmi_string(DMIHeader * dmi, u8 s)
-{
-	u8 *p;
-
-	if (!s) {
-		return "";
-	}
-	s--;
-
-	p = (u8 *) dmi + dmi->length;
-	while (s > 0) {
-		p += strlen(p);
-		p++;
-		s--;
-	}
-
-	return p;
-}
-
-static void __init dmi_decode(DMIHeader * dmi)
-{
-	u8 *data = (u8 *) dmi;
-	char *p;
-
-#ifdef I8K_DEBUG
-	int i;
-	printk("%08x ", (int)data);
-	for (i = 0; i < data[1] && i < 64; i++) {
-		printk("%02x ", data[i]);
-	}
-	printk("\n");
-#endif
-
-	switch (dmi->type) {
-	case 0:		/* BIOS Information */
-		p = dmi_string(dmi, data[5]);
-		if (*p) {
-			strlcpy(bios_version, p, sizeof(bios_version));
-			string_trim(bios_version, sizeof(bios_version));
-		}
-		break;
-	case 1:		/* System Information */
-		p = dmi_string(dmi, data[4]);
-		if (*p) {
-			strlcpy(system_vendor, p, sizeof(system_vendor));
-			string_trim(system_vendor, sizeof(system_vendor));
-		}
-		p = dmi_string(dmi, data[5]);
-		if (*p) {
-			strlcpy(product_name, p, sizeof(product_name));
-			string_trim(product_name, sizeof(product_name));
-		}
-		p = dmi_string(dmi, data[7]);
-		if (*p) {
-			strlcpy(serial_number, p, sizeof(serial_number));
-			string_trim(serial_number, sizeof(serial_number));
-		}
-		break;
-	}
-}
-
-static int __init dmi_table(u32 base, int len, int num,
-			    void (*fn) (DMIHeader *))
-{
-	u8 *buf;
-	u8 *data;
-	DMIHeader *dmi;
-	int i = 1;
-
-	buf = ioremap(base, len);
-	if (buf == NULL) {
-		return -1;
-	}
-	data = buf;
-
-	/*
-	 * Stop when we see al the items the table claimed to have
-	 * or we run off the end of the table (also happens)
-	 */
-	while ((i < num) && ((data - buf) < len)) {
-		dmi = (DMIHeader *) data;
-		/*
-		 * Avoid misparsing crud if the length of the last
-		 * record is crap
-		 */
-		if ((data - buf + dmi->length) >= len) {
-			break;
-		}
-		fn(dmi);
-		data += dmi->length;
-		/*
-		 * Don't go off the end of the data if there is
-		 * stuff looking like string fill past the end
-		 */
-		while (((data - buf) < len) && (*data || data[1])) {
-			data++;
-		}
-		data += 2;
-		i++;
-	}
-	iounmap(buf);
-
-	return 0;
-}
-
-static int __init dmi_iterate(void (*decode) (DMIHeader *))
-{
-	unsigned char buf[20];
-	void __iomem *p = ioremap(0xe0000, 0x20000), *q;
-
-	if (!p)
-		return -1;
-
-	for (q = p; q < p + 0x20000; q += 16) {
-		memcpy_fromio(buf, q, 20);
-		if (memcmp(buf, "_DMI_", 5) == 0) {
-			u16 num  = buf[13] << 8 | buf[12];
-			u16 len  = buf[7] << 8 | buf[6];
-			u32 base = buf[11] << 24 | buf[10] << 16 | buf[9] << 8 | buf[8];
-#ifdef I8K_DEBUG
-			printk(KERN_INFO "DMI %d.%d present.\n",
-			       buf[14] >> 4, buf[14] & 0x0F);
-			printk(KERN_INFO "%d structures occupying %d bytes.\n",
-			       buf[13] << 8 | buf[12], buf[7] << 8 | buf[6]);
-			printk(KERN_INFO "DMI table at 0x%08X.\n",
-			       buf[11] << 24 | buf[10] << 16 | buf[9] << 8 |
-			       buf[8]);
-#endif
-			if (dmi_table(base, len, num, decode) == 0) {
-				iounmap(p);
-				return 0;
-			}
-		}
-	}
-	iounmap(p);
-	return -1;
-}
-
-/* end of DMI code */
-
-/*
- * Get DMI information.
- */
-static int __init i8k_dmi_probe(void)
-{
-	char **p;
-
-	if (dmi_iterate(dmi_decode) != 0) {
-		printk(KERN_INFO "i8k: unable to get DMI information\n");
-		return -ENODEV;
-	}
-
-	if (strncmp(system_vendor, DELL_SIGNATURE, strlen(DELL_SIGNATURE)) != 0) {
-		printk(KERN_INFO "i8k: not running on a Dell system\n");
-		return -ENODEV;
-	}
-
-	for (p = supported_models;; p++) {
-		if (!*p) {
-			printk(KERN_INFO "i8k: unsupported model: %s\n",
-			       product_name);
-			return -ENODEV;
-		}
-		if (strncmp(product_name, *p, strlen(*p)) == 0) {
-			break;
-		}
-	}
-
-	return 0;
-}
+static struct dmi_system_id __initdata i8k_dmi_table[] = {
+	{
+		.ident = "Dell Inspiron",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Computer"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Inspiron"),
+		},
+	},
+	{
+		.ident = "Dell Latitude",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Computer"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Latitude"),
+		},
+	},
+	{ }
+};
 
 /*
  * Probe for the presence of a supported laptop.
@@ -689,23 +494,30 @@ static int __init i8k_probe(void)
 {
 	char buff[4];
 	int version;
-	int smm_found = 0;
 
 	/*
 	 * Get DMI information
 	 */
-	if (i8k_dmi_probe() != 0) {
+	if (!dmi_check_system(i8k_dmi_table)) {
+		if (!ignore_dmi && !force)
+			return -ENODEV;
+
+		printk(KERN_INFO "i8k: not running on a supported Dell system.\n");
 		printk(KERN_INFO "i8k: vendor=%s, model=%s, version=%s\n",
-		       system_vendor, product_name, bios_version);
+			i8k_get_dmi_data(DMI_SYS_VENDOR),
+			i8k_get_dmi_data(DMI_PRODUCT_NAME),
+			i8k_get_dmi_data(DMI_BIOS_VERSION));
 	}
 
+	strlcpy(bios_version, i8k_get_dmi_data(DMI_BIOS_VERSION), sizeof(bios_version));
+
 	/*
 	 * Get SMM Dell signature
 	 */
 	if (i8k_get_dell_signature() != 0) {
-		printk(KERN_INFO "i8k: unable to get SMM Dell signature\n");
-	} else {
-		smm_found = 1;
+		printk(KERN_ERR "i8k: unable to get SMM Dell signature\n");
+		if (!force)
+			return -ENODEV;
 	}
 
 	/*
@@ -713,9 +525,8 @@ static int __init i8k_probe(void)
 	 */
 	version = i8k_get_bios_version();
 	if (version <= 0) {
-		printk(KERN_INFO "i8k: unable to get SMM BIOS version\n");
+		printk(KERN_WARNING "i8k: unable to get SMM BIOS version\n");
 	} else {
-		smm_found = 1;
 		buff[0] = (version >> 16) & 0xff;
 		buff[1] = (version >> 8) & 0xff;
 		buff[2] = (version) & 0xff;
@@ -723,21 +534,15 @@ static int __init i8k_probe(void)
 		/*
 		 * If DMI BIOS version is unknown use SMM BIOS version.
 		 */
-		if (bios_version[0] == '?') {
-			strcpy(bios_version, buff);
-		}
+		if (!dmi_get_system_info(DMI_BIOS_VERSION))
+			strlcpy(bios_version, buff, sizeof(bios_version));
+
 		/*
 		 * Check if the two versions match.
 		 */
-		if (strncmp(buff, bios_version, sizeof(bios_version)) != 0) {
-			printk(KERN_INFO
-			       "i8k: BIOS version mismatch: %s != %s\n", buff,
-			       bios_version);
-		}
-	}
-
-	if (!smm_found && !force) {
-		return -ENODEV;
+		if (strncmp(buff, bios_version, sizeof(bios_version)) != 0)
+			printk(KERN_WARNING "i8k: BIOS version mismatch: %s != %s\n",
+				buff, bios_version);
 	}
 
 	return 0;
@@ -751,9 +556,8 @@ int __init i8k_init(void)
 	struct proc_dir_entry *proc_i8k;
 
 	/* Are we running on an supported laptop? */
-	if (i8k_probe() != 0) {
+	if (i8k_probe())
 		return -ENODEV;
-	}
 
 	/* Register the proc entry */
 	proc_i8k = create_proc_info_entry("i8k", 0, NULL, i8k_get_info);
diff --git a/include/linux/dmi.h b/include/linux/dmi.h
index d2bcf556088..5e93e6dce9a 100644
--- a/include/linux/dmi.h
+++ b/include/linux/dmi.h
@@ -9,6 +9,7 @@ enum dmi_field {
 	DMI_SYS_VENDOR,
 	DMI_PRODUCT_NAME,
 	DMI_PRODUCT_VERSION,
+	DMI_PRODUCT_SERIAL,
 	DMI_BOARD_VENDOR,
 	DMI_BOARD_NAME,
 	DMI_BOARD_VERSION,
-- 
cgit v1.2.3-70-g09d2


From b4819b593740a6d11db07b52e0fe35975b29a185 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
Date: Sat, 25 Jun 2005 14:54:31 -0700
Subject: [PATCH] mips: add MIPS-specific support for flatmem/discontigmem

2.6.12-git6 doesn't boot on some MIPS machines.  They need the support of flat
memory and discontig memory.

Signed-off-by: Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/mips/Kconfig          | 6 ++++++
 arch/mips/kernel/setup.c   | 4 ++++
 arch/mips/mm/init.c        | 4 ++--
 arch/mips/mm/pgtable.c     | 2 +-
 include/asm-mips/mmzone.h  | 4 ++++
 include/asm-mips/page.h    | 2 +-
 include/asm-mips/pgtable.h | 2 +-
 7 files changed, 19 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 94f5a8eb2c2..bd9de7b00c0 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -1416,6 +1416,12 @@ config HIGHMEM
 	bool "High Memory Support"
 	depends on MIPS32 && (CPU_R3000 || CPU_SB1 || CPU_R7000 || CPU_RM9000 || CPU_R10000) && !(MACH_DECSTATION || MOMENCO_JAGUAR_ATX)
 
+config ARCH_FLATMEM_ENABLE
+	def_bool y
+	depends on !NUMA
+
+source "mm/Kconfig"
+
 config SMP
 	bool "Multi-Processing support"
 	depends on CPU_RM9000 || (SIBYTE_SB1250 && !SIBYTE_STANDALONE) || SGI_IP27
diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c
index 6018ca25ace..3a240e3e004 100644
--- a/arch/mips/kernel/setup.c
+++ b/arch/mips/kernel/setup.c
@@ -33,6 +33,7 @@
 #include <linux/root_dev.h>
 #include <linux/highmem.h>
 #include <linux/console.h>
+#include <linux/mmzone.h>
 
 #include <asm/addrspace.h>
 #include <asm/bootinfo.h>
@@ -356,6 +357,8 @@ static inline void bootmem_init(void)
 	}
 #endif
 
+	memory_present(0, first_usable_pfn, max_low_pfn);
+
 	/* Initialize the boot-time allocator with low memory only.  */
 	bootmap_size = init_bootmem(first_usable_pfn, max_low_pfn);
 
@@ -557,6 +560,7 @@ void __init setup_arch(char **cmdline_p)
 
 	parse_cmdline_early();
 	bootmem_init();
+	sparse_init();
 	paging_init();
 	resource_init();
 }
diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c
index 73843c52877..9c9a271c8a3 100644
--- a/arch/mips/mm/init.c
+++ b/arch/mips/mm/init.c
@@ -128,7 +128,7 @@ static void __init fixrange_init(unsigned long start, unsigned long end,
 #endif /* CONFIG_MIPS64 */
 #endif /* CONFIG_HIGHMEM */
 
-#ifndef CONFIG_DISCONTIGMEM
+#ifndef CONFIG_NEED_MULTIPLE_NODES
 extern void pagetable_init(void);
 
 void __init paging_init(void)
@@ -253,7 +253,7 @@ void __init mem_init(void)
 	       initsize >> 10,
 	       (unsigned long) (totalhigh_pages << (PAGE_SHIFT-10)));
 }
-#endif /* !CONFIG_DISCONTIGMEM */
+#endif /* !CONFIG_NEED_MULTIPLE_NODES */
 
 #ifdef CONFIG_BLK_DEV_INITRD
 void free_initrd_mem(unsigned long start, unsigned long end)
diff --git a/arch/mips/mm/pgtable.c b/arch/mips/mm/pgtable.c
index 3b88fdeef32..3fe94202da8 100644
--- a/arch/mips/mm/pgtable.c
+++ b/arch/mips/mm/pgtable.c
@@ -5,7 +5,7 @@
 
 void show_mem(void)
 {
-#ifndef CONFIG_DISCONTIGMEM  /* XXX(hch): later.. */
+#ifndef CONFIG_NEED_MULTIPLE_NODES  /* XXX(hch): later.. */
 	int pfn, total = 0, reserved = 0;
 	int shared = 0, cached = 0;
 	int highmem = 0;
diff --git a/include/asm-mips/mmzone.h b/include/asm-mips/mmzone.h
index 29ee13be0b2..d721143dbd4 100644
--- a/include/asm-mips/mmzone.h
+++ b/include/asm-mips/mmzone.h
@@ -8,6 +8,8 @@
 #include <asm/page.h>
 #include <mmzone.h>
 
+#ifdef CONFIG_DISCONTIGMEM
+
 #define kvaddr_to_nid(kvaddr)	pa_to_nid(__pa(kvaddr))
 #define pfn_to_nid(pfn)		pa_to_nid((pfn) << PAGE_SHIFT)
 
@@ -36,4 +38,6 @@
 /* XXX: FIXME -- wli */
 #define kern_addr_valid(addr)	(0)
 
+#endif /* CONFIG_DISCONTIGMEM */
+
 #endif /* _ASM_MMZONE_H_ */
diff --git a/include/asm-mips/page.h b/include/asm-mips/page.h
index d1bf8240e73..5cae35cd9ba 100644
--- a/include/asm-mips/page.h
+++ b/include/asm-mips/page.h
@@ -127,7 +127,7 @@ static __inline__ int get_order(unsigned long size)
 
 #define pfn_to_kaddr(pfn)	__va((pfn) << PAGE_SHIFT)
 
-#ifndef CONFIG_DISCONTIGMEM
+#ifndef CONFIG_NEED_MULTIPLE_NODES
 #define pfn_to_page(pfn)	(mem_map + (pfn))
 #define page_to_pfn(page)	((unsigned long)((page) - mem_map))
 #define pfn_valid(pfn)		((pfn) < max_mapnr)
diff --git a/include/asm-mips/pgtable.h b/include/asm-mips/pgtable.h
index 878843203d6..e76ccd6e3a5 100644
--- a/include/asm-mips/pgtable.h
+++ b/include/asm-mips/pgtable.h
@@ -350,7 +350,7 @@ static inline void update_mmu_cache(struct vm_area_struct *vma,
 	__update_cache(vma, address, pte);
 }
 
-#ifndef CONFIG_DISCONTIGMEM
+#ifndef CONFIG_NEED_MULTIPLE_NODES
 #define kern_addr_valid(addr)	(1)
 #endif
 
-- 
cgit v1.2.3-70-g09d2


From 33d9e9b56d5ccd7776fdfe3ecce4a2793dee6fd3 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Sat, 25 Jun 2005 14:54:37 -0700
Subject: [PATCH] ppc32: Add support for Freescale e200 (Book-E) core

The e200 core is a Book-E core (similar to e500) that has a unified L1 cache
and is not cache coherent on the bus.  The e200 core also adds a separate
exception level for debug exceptions.  Part of this patch helps to cleanup a
few cases that are true for all Freescale Book-E parts, not just e500.

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/Kconfig                 | 17 ++++++-----
 arch/ppc/Makefile                |  3 +-
 arch/ppc/kernel/Makefile         |  2 ++
 arch/ppc/kernel/cputable.c       | 25 +++++++++++++++-
 arch/ppc/kernel/entry.S          |  9 ++++++
 arch/ppc/kernel/head_booke.h     | 64 +++++++++++++++++++++++++++++++++++++++-
 arch/ppc/kernel/head_fsl_booke.S | 51 ++++++++++++++++++++++++++++++++
 arch/ppc/kernel/misc.S           |  8 +++++
 arch/ppc/kernel/perfmon.c        |  2 +-
 arch/ppc/kernel/traps.c          | 24 +++++++++++++--
 arch/ppc/mm/fsl_booke_mmu.c      |  2 +-
 include/asm-ppc/mmu.h            |  2 +-
 include/asm-ppc/mmu_context.h    |  2 +-
 include/asm-ppc/ppc_asm.h        |  2 ++
 include/asm-ppc/reg.h            |  1 +
 include/asm-ppc/reg_booke.h      | 18 ++++++++++-
 16 files changed, 214 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/arch/ppc/Kconfig b/arch/ppc/Kconfig
index 848f43970a4..979590a9baf 100644
--- a/arch/ppc/Kconfig
+++ b/arch/ppc/Kconfig
@@ -88,6 +88,9 @@ config 8xx
 	depends on BROKEN
 	bool "8xx"
 
+config E200
+	bool "e200"
+
 config E500
 	bool "e500"
 
@@ -98,12 +101,12 @@ config PPC_FPU
 
 config BOOKE
 	bool
-	depends on E500
+	depends on E200 || E500
 	default y
 
 config FSL_BOOKE
 	bool
-	depends on E500
+	depends on E200 || E500
 	default y
 
 config PTE_64BIT
@@ -141,16 +144,16 @@ config ALTIVEC
 
 config SPE
 	bool "SPE Support"
-	depends on E500
+	depends on E200 || E500
 	---help---
 	  This option enables kernel support for the Signal Processing
 	  Extensions (SPE) to the PowerPC processor. The kernel currently
 	  supports saving and restoring SPE registers, and turning on the
 	  'spe enable' bit so user processes can execute SPE instructions.
 
-	  This option is only usefully if you have a processor that supports
+	  This option is only useful if you have a processor that supports
 	  SPE (e500, otherwise known as 85xx series), but does not have any
-	  affect on a non-spe cpu (it does, however add code to the kernel).
+	  effect on a non-spe cpu (it does, however add code to the kernel).
 
 	  If in doubt, say Y here.
 
@@ -200,7 +203,7 @@ config TAU_AVERAGE
 
 config MATH_EMULATION
 	bool "Math emulation"
-	depends on 4xx || 8xx || E500
+	depends on 4xx || 8xx || E200 || E500
 	---help---
 	  Some PowerPC chips designed for embedded applications do not have
 	  a floating-point unit and therefore do not implement the
@@ -254,7 +257,7 @@ config PPC_STD_MMU
 
 config NOT_COHERENT_CACHE
 	bool
-	depends on 4xx || 8xx
+	depends on 4xx || 8xx || E200
 	default y
 
 endmenu
diff --git a/arch/ppc/Makefile b/arch/ppc/Makefile
index 0432a25b473..f9b0d778dd8 100644
--- a/arch/ppc/Makefile
+++ b/arch/ppc/Makefile
@@ -29,7 +29,7 @@ CPP		= $(CC) -E $(CFLAGS)
 
 CHECKFLAGS	+= -D__powerpc__
 
-ifndef CONFIG_E500
+ifndef CONFIG_FSL_BOOKE
 CFLAGS		+= -mstring
 endif
 
@@ -38,6 +38,7 @@ cpu-as-$(CONFIG_4xx)		+= -Wa,-m405
 cpu-as-$(CONFIG_6xx)		+= -Wa,-maltivec
 cpu-as-$(CONFIG_POWER4)		+= -Wa,-maltivec
 cpu-as-$(CONFIG_E500)		+= -Wa,-me500
+cpu-as-$(CONFIG_E200)		+= -Wa,-me200
 
 AFLAGS += $(cpu-as-y)
 CFLAGS += $(cpu-as-y)
diff --git a/arch/ppc/kernel/Makefile b/arch/ppc/kernel/Makefile
index b284451802c..8276c0b551d 100644
--- a/arch/ppc/kernel/Makefile
+++ b/arch/ppc/kernel/Makefile
@@ -26,7 +26,9 @@ obj-$(CONFIG_KGDB)		+= ppc-stub.o
 obj-$(CONFIG_SMP)		+= smp.o smp-tbsync.o
 obj-$(CONFIG_TAU)		+= temp.o
 obj-$(CONFIG_ALTIVEC)		+= vecemu.o vector.o
+ifndef CONFIG_E200
 obj-$(CONFIG_FSL_BOOKE)		+= perfmon_fsl_booke.o
+endif
 
 ifndef CONFIG_MATH_EMULATION
 obj-$(CONFIG_8xx)		+= softemu8xx.o
diff --git a/arch/ppc/kernel/cputable.c b/arch/ppc/kernel/cputable.c
index 01c226008db..50936cda0af 100644
--- a/arch/ppc/kernel/cputable.c
+++ b/arch/ppc/kernel/cputable.c
@@ -903,7 +903,30 @@ struct cpu_spec	cpu_specs[] = {
 		.dcache_bsize		= 32,
 	},
 #endif /* CONFIG_44x */
-#ifdef CONFIG_E500
+#ifdef CONFIG_FSL_BOOKE
+	{ 	/* e200z5 */
+		.pvr_mask		= 0xfff00000,
+		.pvr_value		= 0x81000000,
+		.cpu_name		= "e200z5",
+		/* xxx - galak: add CPU_FTR_MAYBE_CAN_DOZE */
+		.cpu_features		= CPU_FTR_USE_TB,
+		.cpu_user_features	= PPC_FEATURE_32 |
+			PPC_FEATURE_HAS_MMU | PPC_FEATURE_HAS_EFP_SINGLE |
+			PPC_FEATURE_UNIFIED_CACHE,
+		.dcache_bsize		= 32,
+	},
+	{ 	/* e200z6 */
+		.pvr_mask		= 0xfff00000,
+		.pvr_value		= 0x81100000,
+		.cpu_name		= "e200z6",
+		/* xxx - galak: add CPU_FTR_MAYBE_CAN_DOZE */
+		.cpu_features		= CPU_FTR_USE_TB,
+		.cpu_user_features	= PPC_FEATURE_32 |
+			PPC_FEATURE_HAS_MMU | PPC_FEATURE_SPE_COMP |
+			PPC_FEATURE_HAS_EFP_SINGLE |
+			PPC_FEATURE_UNIFIED_CACHE,
+		.dcache_bsize		= 32,
+	},
 	{ 	/* e500 */
 		.pvr_mask		= 0xffff0000,
 		.pvr_value		= 0x80200000,
diff --git a/arch/ppc/kernel/entry.S b/arch/ppc/kernel/entry.S
index 8377b6ca26d..d4df68629cc 100644
--- a/arch/ppc/kernel/entry.S
+++ b/arch/ppc/kernel/entry.S
@@ -60,6 +60,11 @@ mcheck_transfer_to_handler:
 	TRANSFER_TO_HANDLER_EXC_LEVEL(MCHECK)
 	b	transfer_to_handler_full
 
+	.globl	debug_transfer_to_handler
+debug_transfer_to_handler:
+	TRANSFER_TO_HANDLER_EXC_LEVEL(DEBUG)
+	b	transfer_to_handler_full
+
 	.globl	crit_transfer_to_handler
 crit_transfer_to_handler:
 	TRANSFER_TO_HANDLER_EXC_LEVEL(CRIT)
@@ -835,6 +840,10 @@ ret_from_crit_exc:
 	RET_FROM_EXC_LEVEL(SPRN_CSRR0, SPRN_CSRR1, RFCI)
 
 #ifdef CONFIG_BOOKE
+	.globl	ret_from_debug_exc
+ret_from_debug_exc:
+	RET_FROM_EXC_LEVEL(SPRN_DSRR0, SPRN_DSRR1, RFDI)
+
 	.globl	ret_from_mcheck_exc
 ret_from_mcheck_exc:
 	RET_FROM_EXC_LEVEL(SPRN_MCSRR0, SPRN_MCSRR1, RFMCI)
diff --git a/arch/ppc/kernel/head_booke.h b/arch/ppc/kernel/head_booke.h
index 9c50f9d2657..9342acf12e7 100644
--- a/arch/ppc/kernel/head_booke.h
+++ b/arch/ppc/kernel/head_booke.h
@@ -49,6 +49,7 @@
  *
  * On 40x critical is the only additional level
  * On 44x/e500 we have critical and machine check
+ * On e200 we have critical and debug (machine check occurs via critical)
  *
  * Additionally we reserve a SPRG for each priority level so we can free up a
  * GPR to use as the base for indirect access to the exception stacks.  This
@@ -60,12 +61,16 @@
 
 /* CRIT_SPRG only used in critical exception handling */
 #define CRIT_SPRG	SPRN_SPRG2
-/* MCHECK_SPRG only used in critical exception handling */
+/* MCHECK_SPRG only used in machine check exception handling */
 #define MCHECK_SPRG	SPRN_SPRG6W
 
 #define MCHECK_STACK_TOP	(exception_stack_top - 4096)
 #define CRIT_STACK_TOP		(exception_stack_top)
 
+/* only on e200 for now */
+#define DEBUG_STACK_TOP		(exception_stack_top - 4096)
+#define DEBUG_SPRG		SPRN_SPRG6W
+
 #ifdef CONFIG_SMP
 #define BOOKE_LOAD_EXC_LEVEL_STACK(level)		\
 	mfspr	r8,SPRN_PIR;				\
@@ -124,6 +129,8 @@
 
 #define CRITICAL_EXCEPTION_PROLOG \
 		EXC_LEVEL_EXCEPTION_PROLOG(CRIT, SPRN_CSRR0, SPRN_CSRR1)
+#define DEBUG_EXCEPTION_PROLOG \
+		EXC_LEVEL_EXCEPTION_PROLOG(DEBUG, SPRN_DSRR0, SPRN_DSRR1)
 #define MCHECK_EXCEPTION_PROLOG \
 		EXC_LEVEL_EXCEPTION_PROLOG(MCHECK, SPRN_MCSRR0, SPRN_MCSRR1)
 
@@ -205,6 +212,60 @@ label:
  * save (and later restore) the MSR via SPRN_CSRR1, which will still have
  * the MSR_DE bit set.
  */
+#ifdef CONFIG_E200
+#define DEBUG_EXCEPTION							      \
+	START_EXCEPTION(Debug);						      \
+	DEBUG_EXCEPTION_PROLOG;						      \
+									      \
+	/*								      \
+	 * If there is a single step or branch-taken exception in an	      \
+	 * exception entry sequence, it was probably meant to apply to	      \
+	 * the code where the exception occurred (since exception entry	      \
+	 * doesn't turn off DE automatically).  We simulate the effect	      \
+	 * of turning off DE on entry to an exception handler by turning      \
+	 * off DE in the CSRR1 value and clearing the debug status.	      \
+	 */								      \
+	mfspr	r10,SPRN_DBSR;		/* check single-step/branch taken */  \
+	andis.	r10,r10,DBSR_IC@h;					      \
+	beq+	2f;							      \
+									      \
+	lis	r10,KERNELBASE@h;	/* check if exception in vectors */   \
+	ori	r10,r10,KERNELBASE@l;					      \
+	cmplw	r12,r10;						      \
+	blt+	2f;			/* addr below exception vectors */    \
+									      \
+	lis	r10,Debug@h;						      \
+	ori	r10,r10,Debug@l;					      \
+	cmplw	r12,r10;						      \
+	bgt+	2f;			/* addr above exception vectors */    \
+									      \
+	/* here it looks like we got an inappropriate debug exception. */     \
+1:	rlwinm	r9,r9,0,~MSR_DE;	/* clear DE in the CDRR1 value */     \
+	lis	r10,DBSR_IC@h;		/* clear the IC event */	      \
+	mtspr	SPRN_DBSR,r10;						      \
+	/* restore state and get out */					      \
+	lwz	r10,_CCR(r11);						      \
+	lwz	r0,GPR0(r11);						      \
+	lwz	r1,GPR1(r11);						      \
+	mtcrf	0x80,r10;						      \
+	mtspr	SPRN_DSRR0,r12;						      \
+	mtspr	SPRN_DSRR1,r9;						      \
+	lwz	r9,GPR9(r11);						      \
+	lwz	r12,GPR12(r11);						      \
+	mtspr	DEBUG_SPRG,r8;						      \
+	BOOKE_LOAD_EXC_LEVEL_STACK(DEBUG); /* r8 points to the debug stack */ \
+	lwz	r10,GPR10-INT_FRAME_SIZE(r8);				      \
+	lwz	r11,GPR11-INT_FRAME_SIZE(r8);				      \
+	mfspr	r8,DEBUG_SPRG;						      \
+									      \
+	RFDI;								      \
+	b	.;							      \
+									      \
+	/* continue normal handling for a critical exception... */	      \
+2:	mfspr	r4,SPRN_DBSR;						      \
+	addi	r3,r1,STACK_FRAME_OVERHEAD;				      \
+	EXC_XFER_TEMPLATE(DebugException, 0x2002, (MSR_KERNEL & ~(MSR_ME|MSR_DE|MSR_CE)), NOCOPY, debug_transfer_to_handler, ret_from_debug_exc)
+#else
 #define DEBUG_EXCEPTION							      \
 	START_EXCEPTION(Debug);						      \
 	CRITICAL_EXCEPTION_PROLOG;					      \
@@ -257,6 +318,7 @@ label:
 2:	mfspr	r4,SPRN_DBSR;						      \
 	addi	r3,r1,STACK_FRAME_OVERHEAD;				      \
 	EXC_XFER_TEMPLATE(DebugException, 0x2002, (MSR_KERNEL & ~(MSR_ME|MSR_DE|MSR_CE)), NOCOPY, crit_transfer_to_handler, ret_from_crit_exc)
+#endif
 
 #define INSTRUCTION_STORAGE_EXCEPTION					      \
 	START_EXCEPTION(InstructionStorage)				      \
diff --git a/arch/ppc/kernel/head_fsl_booke.S b/arch/ppc/kernel/head_fsl_booke.S
index ce36e88ba62..eb804b7a3cb 100644
--- a/arch/ppc/kernel/head_fsl_booke.S
+++ b/arch/ppc/kernel/head_fsl_booke.S
@@ -102,6 +102,7 @@ invstr:	mflr	r6				/* Make it accessible */
 	or	r7,r7,r4
 	mtspr	SPRN_MAS6,r7
 	tlbsx	0,r6				/* search MSR[IS], SPID=PID0 */
+#ifndef CONFIG_E200
 	mfspr	r7,SPRN_MAS1
 	andis.	r7,r7,MAS1_VALID@h
 	bne	match_TLB
@@ -118,6 +119,7 @@ invstr:	mflr	r6				/* Make it accessible */
 	or	r7,r7,r4
 	mtspr	SPRN_MAS6,r7
 	tlbsx	0,r6				/* Fall through, we had to match */
+#endif
 match_TLB:
 	mfspr	r7,SPRN_MAS0
 	rlwinm	r3,r7,16,20,31			/* Extract MAS0(Entry) */
@@ -196,8 +198,10 @@ skpinv:	addi	r6,r6,1				/* Increment */
 /* 4. Clear out PIDs & Search info */
 	li	r6,0
 	mtspr	SPRN_PID0,r6
+#ifndef CONFIG_E200
 	mtspr	SPRN_PID1,r6
 	mtspr	SPRN_PID2,r6
+#endif
 	mtspr	SPRN_MAS6,r6
 
 /* 5. Invalidate mapping we started in */
@@ -277,7 +281,9 @@ skpinv:	addi	r6,r6,1				/* Increment */
 	SET_IVOR(32, SPEUnavailable);
 	SET_IVOR(33, SPEFloatingPointData);
 	SET_IVOR(34, SPEFloatingPointRound);
+#ifndef CONFIG_E200
 	SET_IVOR(35, PerformanceMonitor);
+#endif
 
 	/* Establish the interrupt vector base */
 	lis	r4,interrupt_base@h	/* IVPR only uses the high 16-bits */
@@ -285,6 +291,9 @@ skpinv:	addi	r6,r6,1				/* Increment */
 
 	/* Setup the defaults for TLB entries */
 	li	r2,(MAS4_TSIZED(BOOKE_PAGESZ_4K))@l
+#ifdef CONFIG_E200
+	oris	r2,r2,MAS4_TLBSELD(1)@h
+#endif
    	mtspr	SPRN_MAS4, r2
 
 #if 0
@@ -293,6 +302,12 @@ skpinv:	addi	r6,r6,1				/* Increment */
 	oris	r2,r2,HID0_DOZE@h
 	mtspr	SPRN_HID0, r2
 #endif
+#ifdef CONFIG_E200
+	/* enable dedicated debug exception handling resources (Debug APU) */
+	mfspr	r2,SPRN_HID0
+	ori 	r2,r2,HID0_DAPUEN@l
+	mtspr	SPRN_HID0,r2
+#endif
 
 #if !defined(CONFIG_BDI_SWITCH)
 	/*
@@ -414,7 +429,12 @@ interrupt_base:
 	CRITICAL_EXCEPTION(0x0100, CriticalInput, UnknownException)
 
 	/* Machine Check Interrupt */
+#ifdef CONFIG_E200
+	/* no RFMCI, MCSRRs on E200 */
+	CRITICAL_EXCEPTION(0x0200, MachineCheck, MachineCheckException)
+#else
 	MCHECK_EXCEPTION(0x0200, MachineCheck, MachineCheckException)
+#endif
 
 	/* Data Storage Interrupt */
 	START_EXCEPTION(DataStorage)
@@ -519,8 +539,13 @@ interrupt_base:
 	/* Floating Point Unavailable Interrupt */
 #ifdef CONFIG_PPC_FPU
 	FP_UNAVAILABLE_EXCEPTION
+#else
+#ifdef CONFIG_E200
+	/* E200 treats 'normal' floating point instructions as FP Unavail exception */
+	EXCEPTION(0x0800, FloatingPointUnavailable, ProgramCheckException, EXC_XFER_EE)
 #else
 	EXCEPTION(0x0800, FloatingPointUnavailable, UnknownException, EXC_XFER_EE)
+#endif
 #endif
 
 	/* System Call Interrupt */
@@ -691,6 +716,7 @@ interrupt_base:
 /*
  * Local functions
  */
+
 	/*
 	 * Data TLB exceptions will bail out to this point
 	 * if they can't resolve the lightweight TLB fault.
@@ -761,6 +787,31 @@ END_FTR_SECTION_IFSET(CPU_FTR_BIG_PHYS)
 2:	rlwimi	r11, r12, 0, 20, 31	/* Extract RPN from PTE and merge with perms */
 	mtspr	SPRN_MAS3, r11
 #endif
+#ifdef CONFIG_E200
+	/* Round robin TLB1 entries assignment */
+	mfspr	r12, SPRN_MAS0
+
+	/* Extract TLB1CFG(NENTRY) */
+	mfspr	r11, SPRN_TLB1CFG
+	andi.	r11, r11, 0xfff
+
+	/* Extract MAS0(NV) */
+	andi.	r13, r12, 0xfff
+	addi	r13, r13, 1
+	cmpw	0, r13, r11
+	addi	r12, r12, 1
+
+	/* check if we need to wrap */
+	blt	7f
+
+	/* wrap back to first free tlbcam entry */
+	lis	r13, tlbcam_index@ha
+	lwz	r13, tlbcam_index@l(r13)
+	rlwimi	r12, r13, 0, 20, 31
+7:
+	mtspr   SPRN_MAS0,r12
+#endif /* CONFIG_E200 */
+
 	tlbwe
 
 	/* Done...restore registers and get out of here.  */
diff --git a/arch/ppc/kernel/misc.S b/arch/ppc/kernel/misc.S
index 7329ef177a1..a3132f8e799 100644
--- a/arch/ppc/kernel/misc.S
+++ b/arch/ppc/kernel/misc.S
@@ -593,6 +593,14 @@ _GLOBAL(flush_instruction_cache)
 	iccci	0,r3
 #endif
 #elif CONFIG_FSL_BOOKE
+BEGIN_FTR_SECTION
+	mfspr   r3,SPRN_L1CSR0
+	ori     r3,r3,L1CSR0_CFI|L1CSR0_CLFC
+	/* msync; isync recommended here */
+	mtspr   SPRN_L1CSR0,r3
+	isync
+	blr
+END_FTR_SECTION_IFCLR(CPU_FTR_SPLIT_ID_CACHE)
 	mfspr	r3,SPRN_L1CSR1
 	ori	r3,r3,L1CSR1_ICFI|L1CSR1_ICLFR
 	mtspr	SPRN_L1CSR1,r3
diff --git a/arch/ppc/kernel/perfmon.c b/arch/ppc/kernel/perfmon.c
index 918f6b252e4..fa1dad96b83 100644
--- a/arch/ppc/kernel/perfmon.c
+++ b/arch/ppc/kernel/perfmon.c
@@ -36,7 +36,7 @@
 /* A lock to regulate grabbing the interrupt */
 DEFINE_SPINLOCK(perfmon_lock);
 
-#ifdef CONFIG_FSL_BOOKE
+#if defined (CONFIG_FSL_BOOKE) && !defined (CONFIG_E200)
 static void dummy_perf(struct pt_regs *regs)
 {
 	unsigned int pmgc0 = mfpmr(PMRN_PMGC0);
diff --git a/arch/ppc/kernel/traps.c b/arch/ppc/kernel/traps.c
index 2ca8ecfeefd..9e6ae569665 100644
--- a/arch/ppc/kernel/traps.c
+++ b/arch/ppc/kernel/traps.c
@@ -173,13 +173,13 @@ static inline int check_io_access(struct pt_regs *regs)
 /* On 4xx, the reason for the machine check or program exception
    is in the ESR. */
 #define get_reason(regs)	((regs)->dsisr)
-#ifndef CONFIG_E500
+#ifndef CONFIG_FSL_BOOKE
 #define get_mc_reason(regs)	((regs)->dsisr)
 #else
 #define get_mc_reason(regs)	(mfspr(SPRN_MCSR))
 #endif
 #define REASON_FP		ESR_FP
-#define REASON_ILLEGAL		ESR_PIL
+#define REASON_ILLEGAL		(ESR_PIL | ESR_PUO)
 #define REASON_PRIVILEGED	ESR_PPR
 #define REASON_TRAP		ESR_PTR
 
@@ -302,7 +302,25 @@ void MachineCheckException(struct pt_regs *regs)
 		printk("Bus - Instruction Parity Error\n");
 	if (reason & MCSR_BUS_RPERR)
 		printk("Bus - Read Parity Error\n");
-#else /* !CONFIG_4xx && !CONFIG_E500 */
+#elif defined (CONFIG_E200)
+	printk("Machine check in kernel mode.\n");
+	printk("Caused by (from MCSR=%lx): ", reason);
+
+	if (reason & MCSR_MCP)
+		printk("Machine Check Signal\n");
+	if (reason & MCSR_CP_PERR)
+		printk("Cache Push Parity Error\n");
+	if (reason & MCSR_CPERR)
+		printk("Cache Parity Error\n");
+	if (reason & MCSR_EXCP_ERR)
+		printk("ISI, ITLB, or Bus Error on first instruction fetch for an exception handler\n");
+	if (reason & MCSR_BUS_IRERR)
+		printk("Bus - Read Bus Error on instruction fetch\n");
+	if (reason & MCSR_BUS_DRERR)
+		printk("Bus - Read Bus Error on data load\n");
+	if (reason & MCSR_BUS_WRERR)
+		printk("Bus - Write Bus Error on buffered store or cache line push\n");
+#else /* !CONFIG_4xx && !CONFIG_E500 && !CONFIG_E200 */
 	printk("Machine check in kernel mode.\n");
 	printk("Caused by (from SRR1=%lx): ", reason);
 	switch (reason & 0x601F0000) {
diff --git a/arch/ppc/mm/fsl_booke_mmu.c b/arch/ppc/mm/fsl_booke_mmu.c
index e07990efa04..9fa884de5c7 100644
--- a/arch/ppc/mm/fsl_booke_mmu.c
+++ b/arch/ppc/mm/fsl_booke_mmu.c
@@ -126,7 +126,7 @@ void settlbcam(int index, unsigned long virt, phys_addr_t phys,
 		flags |= _PAGE_COHERENT;
 #endif
 
-	TLBCAM[index].MAS0 = MAS0_TLBSEL(1) | MAS0_ESEL(index);
+	TLBCAM[index].MAS0 = MAS0_TLBSEL(1) | MAS0_ESEL(index) | MAS0_NV(index+1);
 	TLBCAM[index].MAS1 = MAS1_VALID | MAS1_IPROT | MAS1_TSIZE(tsize) | MAS1_TID(pid);
 	TLBCAM[index].MAS2 = virt & PAGE_MASK;
 
diff --git a/include/asm-ppc/mmu.h b/include/asm-ppc/mmu.h
index d465aee1c82..9205db404c7 100644
--- a/include/asm-ppc/mmu.h
+++ b/include/asm-ppc/mmu.h
@@ -405,7 +405,7 @@ typedef struct _P601_BAT {
 
 #define MAS0_TLBSEL(x)	((x << 28) & 0x30000000)
 #define MAS0_ESEL(x)	((x << 16) & 0x0FFF0000)
-#define MAS0_NV		0x00000FFF
+#define MAS0_NV(x)	((x) & 0x00000FFF)
 
 #define MAS1_VALID 	0x80000000
 #define MAS1_IPROT	0x40000000
diff --git a/include/asm-ppc/mmu_context.h b/include/asm-ppc/mmu_context.h
index 9222fa6ca17..ccabbce39d8 100644
--- a/include/asm-ppc/mmu_context.h
+++ b/include/asm-ppc/mmu_context.h
@@ -63,7 +63,7 @@ static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
 #define LAST_CONTEXT    	255
 #define FIRST_CONTEXT    	1
 
-#elif defined(CONFIG_E500)
+#elif defined(CONFIG_E200) || defined(CONFIG_E500)
 #define NO_CONTEXT      	256
 #define LAST_CONTEXT    	255
 #define FIRST_CONTEXT    	1
diff --git a/include/asm-ppc/ppc_asm.h b/include/asm-ppc/ppc_asm.h
index 13fa8e7483c..f76221def48 100644
--- a/include/asm-ppc/ppc_asm.h
+++ b/include/asm-ppc/ppc_asm.h
@@ -174,6 +174,8 @@ END_FTR_SECTION_IFCLR(CPU_FTR_601)
 #define CLR_TOP32(r)
 #endif /* CONFIG_PPC64BRIDGE */
 
+#define RFCI		.long 0x4c000066	/* rfci instruction */
+#define RFDI		.long 0x4c00004e	/* rfdi instruction */
 #define RFMCI		.long 0x4c00004c	/* rfmci instruction */
 
 #ifdef CONFIG_IBM405_ERR77
diff --git a/include/asm-ppc/reg.h b/include/asm-ppc/reg.h
index c418aab7cd3..88b4222154d 100644
--- a/include/asm-ppc/reg.h
+++ b/include/asm-ppc/reg.h
@@ -160,6 +160,7 @@
 #define HID0_ICFI	(1<<11)		/* Instr. Cache Flash Invalidate */
 #define HID0_DCI	(1<<10)		/* Data Cache Invalidate */
 #define HID0_SPD	(1<<9)		/* Speculative disable */
+#define HID0_DAPUEN	(1<<8)		/* Debug APU enable */
 #define HID0_SGE	(1<<7)		/* Store Gathering Enable */
 #define HID0_SIED	(1<<7)		/* Serial Instr. Execution [Disable] */
 #define HID0_DFCA	(1<<6)		/* Data Cache Flush Assist */
diff --git a/include/asm-ppc/reg_booke.h b/include/asm-ppc/reg_booke.h
index 45c5e6f2b7a..00ad9c754c7 100644
--- a/include/asm-ppc/reg_booke.h
+++ b/include/asm-ppc/reg_booke.h
@@ -165,6 +165,8 @@ do {						\
 #define SPRN_MCSRR1	0x23B	/* Machine Check Save and Restore Register 1 */
 #define SPRN_MCSR	0x23C	/* Machine Check Status Register */
 #define SPRN_MCAR	0x23D	/* Machine Check Address Register */
+#define SPRN_DSRR0	0x23E	/* Debug Save and Restore Register 0 */
+#define SPRN_DSRR1	0x23F	/* Debug Save and Restore Register 1 */
 #define SPRN_MAS0	0x270	/* MMU Assist Register 0 */
 #define SPRN_MAS1	0x271	/* MMU Assist Register 1 */
 #define SPRN_MAS2	0x272	/* MMU Assist Register 2 */
@@ -264,6 +266,17 @@ do {						\
 #define MCSR_BUS_IPERR 	0x00000002UL /* Instruction parity Error */
 #define MCSR_BUS_RPERR 	0x00000001UL /* Read parity Error */
 #endif
+#ifdef CONFIG_E200
+#define MCSR_MCP 	0x80000000UL /* Machine Check Input Pin */
+#define MCSR_CP_PERR 	0x20000000UL /* Cache Push Parity Error */
+#define MCSR_CPERR 	0x10000000UL /* Cache Parity Error */
+#define MCSR_EXCP_ERR 	0x08000000UL /* ISI, ITLB, or Bus Error on 1st insn
+					fetch for an exception handler */
+#define MCSR_BUS_IRERR 	0x00000010UL /* Read Bus Error on instruction fetch*/
+#define MCSR_BUS_DRERR 	0x00000008UL /* Read Bus Error on data load */
+#define MCSR_BUS_WRERR 	0x00000004UL /* Write Bus Error on buffered
+					store or cache line push */
+#endif
 
 /* Bit definitions for the DBSR. */
 /*
@@ -311,6 +324,7 @@ do {						\
 #define ESR_ST		0x00800000	/* Store Operation */
 #define ESR_DLK		0x00200000	/* Data Cache Locking */
 #define ESR_ILK		0x00100000	/* Instr. Cache Locking */
+#define ESR_PUO		0x00040000	/* Unimplemented Operation exception */
 #define ESR_BO		0x00020000	/* Byte Ordering */
 
 /* Bit definitions related to the DBCR0. */
@@ -387,10 +401,12 @@ do {						\
 #define ICCR_CACHE	1		/* Cacheable */
 
 /* Bit definitions for L1CSR0. */
+#define L1CSR0_CLFC	0x00000100	/* Cache Lock Bits Flash Clear */
 #define L1CSR0_DCFI	0x00000002	/* Data Cache Flash Invalidate */
+#define L1CSR0_CFI	0x00000002	/* Cache Flash Invalidate */
 #define L1CSR0_DCE	0x00000001	/* Data Cache Enable */
 
-/* Bit definitions for L1CSR0. */
+/* Bit definitions for L1CSR1. */
 #define L1CSR1_ICLFR	0x00000100	/* Instr Cache Lock Bits Flash Reset */
 #define L1CSR1_ICFI	0x00000002	/* Instr Cache Flash Invalidate */
 #define L1CSR1_ICE	0x00000001	/* Instr Cache Enable */
-- 
cgit v1.2.3-70-g09d2


From 3d9077afea4927e488282da7189de9159db20c17 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Sat, 25 Jun 2005 14:54:39 -0700
Subject: [PATCH] ppc32: Remove FSL OCP support

Support for the OCP device model on Freescale (FSL) PPC's is no longer used.
All FSL PPC's that were using OCP have be converted to using the platform
device model.

Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/Kconfig.debug    |  2 +-
 include/asm-ppc/fsl_ocp.h | 54 -----------------------------------------------
 include/asm-ppc/ocp.h     |  4 ----
 3 files changed, 1 insertion(+), 59 deletions(-)
 delete mode 100644 include/asm-ppc/fsl_ocp.h

(limited to 'include')

diff --git a/arch/ppc/Kconfig.debug b/arch/ppc/Kconfig.debug
index d2e1eea8e8e..e16c7710d4b 100644
--- a/arch/ppc/Kconfig.debug
+++ b/arch/ppc/Kconfig.debug
@@ -66,7 +66,7 @@ config SERIAL_TEXT_DEBUG
 
 config PPC_OCP
 	bool
-	depends on IBM_OCP || FSL_OCP || XILINX_OCP
+	depends on IBM_OCP || XILINX_OCP
 	default y
 
 endmenu
diff --git a/include/asm-ppc/fsl_ocp.h b/include/asm-ppc/fsl_ocp.h
deleted file mode 100644
index 050fbba8d04..00000000000
--- a/include/asm-ppc/fsl_ocp.h
+++ /dev/null
@@ -1,54 +0,0 @@
-/*
- * include/asm-ppc/fsl_ocp.h
- *
- * Definitions for the on-chip peripherals on Freescale PPC processors
- *
- * Maintainer: Kumar Gala (kumar.gala@freescale.com)
- *
- * Copyright 2004 Freescale Semiconductor, Inc
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- */
-
-#ifdef __KERNEL__
-#ifndef __ASM_FS_OCP_H__
-#define __ASM_FS_OCP_H__
-
-/* A table of information for supporting the Gianfar Ethernet Controller
- * This helps identify which enet controller we are dealing with,
- * and what type of enet controller it is
- */
-struct ocp_gfar_data {
-	uint interruptTransmit;
-	uint interruptError;
-	uint interruptReceive;
-	uint interruptPHY;
-	uint flags;
-	uint phyid;
-	uint phyregidx;
-	unsigned char mac_addr[6];
-};
-
-/* Flags in the flags field */
-#define GFAR_HAS_COALESCE		0x20
-#define GFAR_HAS_RMON			0x10
-#define GFAR_HAS_MULTI_INTR		0x08
-#define GFAR_FIRM_SET_MACADDR		0x04
-#define GFAR_HAS_PHY_INTR		0x02	/* if not set use a timer */
-#define GFAR_HAS_GIGABIT		0x01
-
-/* Data structure for I2C support.  Just contains a couple flags
- * to distinguish various I2C implementations*/
-struct ocp_fs_i2c_data {
-	uint flags;
-};
-
-/* Flags for I2C */
-#define FS_I2C_SEPARATE_DFSRR	0x02
-#define FS_I2C_CLOCK_5200	0x01
-
-#endif	/* __ASM_FS_OCP_H__ */
-#endif	/* __KERNEL__ */
diff --git a/include/asm-ppc/ocp.h b/include/asm-ppc/ocp.h
index c726f184519..983116f59d9 100644
--- a/include/asm-ppc/ocp.h
+++ b/include/asm-ppc/ocp.h
@@ -202,10 +202,6 @@ static DEVICE_ATTR(name##_##field, S_IRUGO, show_##name##_##field, NULL);
 #include <asm/ibm_ocp.h>
 #endif
 
-#ifdef CONFIG_FSL_OCP
-#include <asm/fsl_ocp.h>
-#endif
-
 #endif				/* CONFIG_PPC_OCP */
 #endif				/* __OCP_H__ */
 #endif				/* __KERNEL__ */
-- 
cgit v1.2.3-70-g09d2


From f370513640492641b4046bfd9a6e4714f6ae530d Mon Sep 17 00:00:00 2001
From: Zwane Mwaikambo <zwane@linuxpower.ca>
Date: Sat, 25 Jun 2005 14:54:50 -0700
Subject: [PATCH] i386 CPU hotplug

(The i386 CPU hotplug patch provides infrastructure for some work which Pavel
is doing as well as for ACPI S3 (suspend-to-RAM) work which Li Shaohua
<shaohua.li@intel.com> is doing)

The following provides i386 architecture support for safely unregistering and
registering processors during runtime, updated for the current -mm tree.  In
order to avoid dumping cpu hotplug code into kernel/irq/* i dropped the
cpu_online check in do_IRQ() by modifying fixup_irqs().  The difference being
that on cpu offline, fixup_irqs() is called before we clear the cpu from
cpu_online_map and a long delay in order to ensure that we never have any
queued external interrupts on the APICs.  There are additional changes to s390
and ppc64 to account for this change.

1) Add CONFIG_HOTPLUG_CPU
2) disable local APIC timer on dead cpus.
3) Disable preempt around irq balancing to prevent CPUs going down.
4) Print irq stats for all possible cpus.
5) Debugging check for interrupts on offline cpus.
6) Hacky fixup_irqs() to redirect irqs when cpus go off/online.
7) play_dead() for offline cpus to spin inside.
8) Handle offline cpus set in flush_tlb_others().
9) Grab lock earlier in smp_call_function() to prevent CPUs going down.
10) Implement __cpu_disable() and __cpu_die().
11) Enable local interrupts in cpu_enable() after fixup_irqs()
12) Don't fiddle with NMI on dead cpu, but leave intact on other cpus.
13) Program IRQ affinity whilst cpu is still in cpu_online_map on offline.

Signed-off-by: Zwane Mwaikambo <zwane@linuxpower.ca>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig               |  9 ++++
 arch/i386/kernel/apic.c         |  3 +-
 arch/i386/kernel/io_apic.c      |  2 +
 arch/i386/kernel/irq.c          | 67 ++++++++++++++++++++++------
 arch/i386/kernel/process.c      | 39 +++++++++++++++-
 arch/i386/kernel/smp.c          | 24 ++++++----
 arch/i386/kernel/smpboot.c      | 98 ++++++++++++++++++++++++++++++++++++++---
 arch/i386/kernel/traps.c        |  8 ++++
 arch/ia64/kernel/smpboot.c      |  1 +
 arch/ppc64/kernel/pSeries_smp.c |  5 ++-
 arch/s390/kernel/smp.c          |  4 +-
 include/asm-i386/cpu.h          |  2 +
 include/asm-i386/irq.h          |  4 ++
 include/asm-i386/smp.h          |  3 ++
 kernel/cpu.c                    | 14 +++---
 15 files changed, 243 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index d4ae5f9ceae..b4cd11e5845 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -1250,6 +1250,15 @@ config SCx200
 	  This support is also available as a module.  If compiled as a
 	  module, it will be called scx200.
 
+config HOTPLUG_CPU
+	bool "Support for hot-pluggable CPUs (EXPERIMENTAL)"
+	depends on SMP && HOTPLUG && EXPERIMENTAL
+	---help---
+	  Say Y here to experiment with turning CPUs off and on.  CPUs
+	  can be controlled through /sys/devices/system/cpu.
+
+	  Say N.
+
 source "drivers/pcmcia/Kconfig"
 
 source "drivers/pci/hotplug/Kconfig"
diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index 8d993fa7175..a28a088f3e7 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -26,6 +26,7 @@
 #include <linux/mc146818rtc.h>
 #include <linux/kernel_stat.h>
 #include <linux/sysdev.h>
+#include <linux/cpu.h>
 
 #include <asm/atomic.h>
 #include <asm/smp.h>
@@ -1048,7 +1049,7 @@ void __init setup_secondary_APIC_clock(void)
 	setup_APIC_timer(calibration_result);
 }
 
-void __init disable_APIC_timer(void)
+void __devinit disable_APIC_timer(void)
 {
 	if (using_apic_timer) {
 		unsigned long v;
diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index 08540bc4ba3..3c2b3bdfc80 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -576,9 +576,11 @@ static int balanced_irq(void *unused)
 		try_to_freeze(PF_FREEZE);
 		if (time_after(jiffies,
 				prev_balance_time+balanced_irq_interval)) {
+			preempt_disable();
 			do_irq_balance();
 			prev_balance_time = jiffies;
 			time_remaining = balanced_irq_interval;
+			preempt_enable();
 		}
 	}
 	return 0;
diff --git a/arch/i386/kernel/irq.c b/arch/i386/kernel/irq.c
index 73945a3c53c..af115004aec 100644
--- a/arch/i386/kernel/irq.c
+++ b/arch/i386/kernel/irq.c
@@ -15,6 +15,9 @@
 #include <linux/seq_file.h>
 #include <linux/interrupt.h>
 #include <linux/kernel_stat.h>
+#include <linux/notifier.h>
+#include <linux/cpu.h>
+#include <linux/delay.h>
 
 DEFINE_PER_CPU(irq_cpustat_t, irq_stat) ____cacheline_maxaligned_in_smp;
 EXPORT_PER_CPU_SYMBOL(irq_stat);
@@ -210,9 +213,8 @@ int show_interrupts(struct seq_file *p, void *v)
 
 	if (i == 0) {
 		seq_printf(p, "           ");
-		for (j=0; j<NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "CPU%d       ",j);
+		for_each_cpu(j)
+			seq_printf(p, "CPU%d       ",j);
 		seq_putc(p, '\n');
 	}
 
@@ -225,9 +227,8 @@ int show_interrupts(struct seq_file *p, void *v)
 #ifndef CONFIG_SMP
 		seq_printf(p, "%10u ", kstat_irqs(i));
 #else
-		for (j = 0; j < NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
+		for_each_cpu(j)
+			seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
 #endif
 		seq_printf(p, " %14s", irq_desc[i].handler->typename);
 		seq_printf(p, "  %s", action->name);
@@ -240,16 +241,14 @@ skip:
 		spin_unlock_irqrestore(&irq_desc[i].lock, flags);
 	} else if (i == NR_IRQS) {
 		seq_printf(p, "NMI: ");
-		for (j = 0; j < NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "%10u ", nmi_count(j));
+		for_each_cpu(j)
+			seq_printf(p, "%10u ", nmi_count(j));
 		seq_putc(p, '\n');
 #ifdef CONFIG_X86_LOCAL_APIC
 		seq_printf(p, "LOC: ");
-		for (j = 0; j < NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "%10u ",
-					per_cpu(irq_stat,j).apic_timer_irqs);
+		for_each_cpu(j)
+			seq_printf(p, "%10u ",
+				per_cpu(irq_stat,j).apic_timer_irqs);
 		seq_putc(p, '\n');
 #endif
 		seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count));
@@ -259,3 +258,45 @@ skip:
 	}
 	return 0;
 }
+
+#ifdef CONFIG_HOTPLUG_CPU
+#include <mach_apic.h>
+
+void fixup_irqs(cpumask_t map)
+{
+	unsigned int irq;
+	static int warned;
+
+	for (irq = 0; irq < NR_IRQS; irq++) {
+		cpumask_t mask;
+		if (irq == 2)
+			continue;
+
+		cpus_and(mask, irq_affinity[irq], map);
+		if (any_online_cpu(mask) == NR_CPUS) {
+			printk("Breaking affinity for irq %i\n", irq);
+			mask = map;
+		}
+		if (irq_desc[irq].handler->set_affinity)
+			irq_desc[irq].handler->set_affinity(irq, mask);
+		else if (irq_desc[irq].action && !(warned++))
+			printk("Cannot set affinity for irq %i\n", irq);
+	}
+
+#if 0
+	barrier();
+	/* Ingo Molnar says: "after the IO-APIC masks have been redirected
+	   [note the nop - the interrupt-enable boundary on x86 is two
+	   instructions from sti] - to flush out pending hardirqs and
+	   IPIs. After this point nothing is supposed to reach this CPU." */
+	__asm__ __volatile__("sti; nop; cli");
+	barrier();
+#else
+	/* That doesn't seem sufficient.  Give it 1ms. */
+	local_irq_enable();
+	mdelay(1);
+	local_irq_disable();
+#endif
+}
+#endif
+
diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index aea2ce1145d..c1b11e8df60 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -13,6 +13,7 @@
 
 #include <stdarg.h>
 
+#include <linux/cpu.h>
 #include <linux/errno.h>
 #include <linux/sched.h>
 #include <linux/fs.h>
@@ -55,6 +56,9 @@
 #include <linux/irq.h>
 #include <linux/err.h>
 
+#include <asm/tlbflush.h>
+#include <asm/cpu.h>
+
 asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
 
 static int hlt_counter;
@@ -143,14 +147,44 @@ static void poll_idle (void)
 	}
 }
 
+#ifdef CONFIG_HOTPLUG_CPU
+#include <asm/nmi.h>
+/* We don't actually take CPU down, just spin without interrupts. */
+static inline void play_dead(void)
+{
+	/* Ack it */
+	__get_cpu_var(cpu_state) = CPU_DEAD;
+
+	/* We shouldn't have to disable interrupts while dead, but
+	 * some interrupts just don't seem to go away, and this makes
+	 * it "work" for testing purposes. */
+	/* Death loop */
+	while (__get_cpu_var(cpu_state) != CPU_UP_PREPARE)
+		cpu_relax();
+
+	local_irq_disable();
+	__flush_tlb_all();
+	cpu_set(smp_processor_id(), cpu_online_map);
+	enable_APIC_timer();
+	local_irq_enable();
+}
+#else
+static inline void play_dead(void)
+{
+	BUG();
+}
+#endif /* CONFIG_HOTPLUG_CPU */
+
 /*
  * The idle thread. There's no useful work to be
  * done, so just try to conserve power and have a
  * low exit latency (ie sit in a loop waiting for
  * somebody to say that they'd like to reschedule)
  */
-void cpu_idle (void)
+void cpu_idle(void)
 {
+	int cpu = raw_smp_processor_id();
+
 	/* endless idle loop with no priority at all */
 	while (1) {
 		while (!need_resched()) {
@@ -165,6 +199,9 @@ void cpu_idle (void)
 			if (!idle)
 				idle = default_idle;
 
+			if (cpu_is_offline(cpu))
+				play_dead();
+
 			__get_cpu_var(irq_stat).idle_timestamp = jiffies;
 			idle();
 		}
diff --git a/arch/i386/kernel/smp.c b/arch/i386/kernel/smp.c
index 68be7d0c723..35f521612b2 100644
--- a/arch/i386/kernel/smp.c
+++ b/arch/i386/kernel/smp.c
@@ -19,6 +19,7 @@
 #include <linux/mc146818rtc.h>
 #include <linux/cache.h>
 #include <linux/interrupt.h>
+#include <linux/cpu.h>
 #include <linux/module.h>
 
 #include <asm/mtrr.h>
@@ -164,7 +165,7 @@ void send_IPI_mask_bitmask(cpumask_t cpumask, int vector)
 	unsigned long flags;
 
 	local_irq_save(flags);
-		
+	WARN_ON(mask & ~cpus_addr(cpu_online_map)[0]);
 	/*
 	 * Wait for idle.
 	 */
@@ -346,21 +347,21 @@ out:
 static void flush_tlb_others(cpumask_t cpumask, struct mm_struct *mm,
 						unsigned long va)
 {
-	cpumask_t tmp;
 	/*
 	 * A couple of (to be removed) sanity checks:
 	 *
-	 * - we do not send IPIs to not-yet booted CPUs.
 	 * - current CPU must not be in mask
 	 * - mask must exist :)
 	 */
 	BUG_ON(cpus_empty(cpumask));
-
-	cpus_and(tmp, cpumask, cpu_online_map);
-	BUG_ON(!cpus_equal(cpumask, tmp));
 	BUG_ON(cpu_isset(smp_processor_id(), cpumask));
 	BUG_ON(!mm);
 
+	/* If a CPU which we ran on has gone down, OK. */
+	cpus_and(cpumask, cpumask, cpu_online_map);
+	if (cpus_empty(cpumask))
+		return;
+
 	/*
 	 * i'm not happy about this global shared spinlock in the
 	 * MM hot path, but we'll see how contended it is.
@@ -476,6 +477,7 @@ void flush_tlb_all(void)
  */
 void smp_send_reschedule(int cpu)
 {
+	WARN_ON(cpu_is_offline(cpu));
 	send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR);
 }
 
@@ -516,10 +518,15 @@ int smp_call_function (void (*func) (void *info), void *info, int nonatomic,
  */
 {
 	struct call_data_struct data;
-	int cpus = num_online_cpus()-1;
+	int cpus;
 
-	if (!cpus)
+	/* Holding any lock stops cpus from going down. */
+	spin_lock(&call_lock);
+	cpus = num_online_cpus() - 1;
+	if (!cpus) {
+		spin_unlock(&call_lock);
 		return 0;
+	}
 
 	/* Can deadlock when called with interrupts disabled */
 	WARN_ON(irqs_disabled());
@@ -531,7 +538,6 @@ int smp_call_function (void (*func) (void *info), void *info, int nonatomic,
 	if (wait)
 		atomic_set(&data.finished, 0);
 
-	spin_lock(&call_lock);
 	call_data = &data;
 	mb();
 	
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index c20d96d5c15..ad74a46e9ef 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -44,6 +44,9 @@
 #include <linux/smp_lock.h>
 #include <linux/irq.h>
 #include <linux/bootmem.h>
+#include <linux/notifier.h>
+#include <linux/cpu.h>
+#include <linux/percpu.h>
 
 #include <linux/delay.h>
 #include <linux/mc146818rtc.h>
@@ -96,6 +99,9 @@ static int trampoline_exec;
 
 static void map_cpu_to_logical_apicid(void);
 
+/* State of each CPU. */
+DEFINE_PER_CPU(int, cpu_state) = { 0 };
+
 /*
  * Currently trivial. Write the real->protected mode
  * bootstrap into the page concerned. The caller
@@ -1119,6 +1125,9 @@ static void __init smp_boot_cpus(unsigned int max_cpus)
    who understands all this stuff should rewrite it properly. --RR 15/Jul/02 */
 void __init smp_prepare_cpus(unsigned int max_cpus)
 {
+	smp_commenced_mask = cpumask_of_cpu(0);
+	cpu_callin_map = cpumask_of_cpu(0);
+	mb();
 	smp_boot_cpus(max_cpus);
 }
 
@@ -1128,20 +1137,99 @@ void __devinit smp_prepare_boot_cpu(void)
 	cpu_set(smp_processor_id(), cpu_callout_map);
 }
 
-int __devinit __cpu_up(unsigned int cpu)
+#ifdef CONFIG_HOTPLUG_CPU
+
+/* must be called with the cpucontrol mutex held */
+static int __devinit cpu_enable(unsigned int cpu)
 {
-	/* This only works at boot for x86.  See "rewrite" above. */
-	if (cpu_isset(cpu, smp_commenced_mask)) {
-		local_irq_enable();
-		return -ENOSYS;
+	/* get the target out of its holding state */
+	per_cpu(cpu_state, cpu) = CPU_UP_PREPARE;
+	wmb();
+
+	/* wait for the processor to ack it. timeout? */
+	while (!cpu_online(cpu))
+		cpu_relax();
+
+	fixup_irqs(cpu_online_map);
+	/* counter the disable in fixup_irqs() */
+	local_irq_enable();
+	return 0;
+}
+
+int __cpu_disable(void)
+{
+	cpumask_t map = cpu_online_map;
+	int cpu = smp_processor_id();
+
+	/*
+	 * Perhaps use cpufreq to drop frequency, but that could go
+	 * into generic code.
+ 	 *
+	 * We won't take down the boot processor on i386 due to some
+	 * interrupts only being able to be serviced by the BSP.
+	 * Especially so if we're not using an IOAPIC	-zwane
+	 */
+	if (cpu == 0)
+		return -EBUSY;
+
+	/* We enable the timer again on the exit path of the death loop */
+	disable_APIC_timer();
+	/* Allow any queued timer interrupts to get serviced */
+	local_irq_enable();
+	mdelay(1);
+	local_irq_disable();
+
+	cpu_clear(cpu, map);
+	fixup_irqs(map);
+	/* It's now safe to remove this processor from the online map */
+	cpu_clear(cpu, cpu_online_map);
+	return 0;
+}
+
+void __cpu_die(unsigned int cpu)
+{
+	/* We don't do anything here: idle task is faking death itself. */
+	unsigned int i;
+
+	for (i = 0; i < 10; i++) {
+		/* They ack this in play_dead by setting CPU_DEAD */
+		if (per_cpu(cpu_state, cpu) == CPU_DEAD)
+			return;
+		current->state = TASK_UNINTERRUPTIBLE;
+		schedule_timeout(HZ/10);
 	}
+ 	printk(KERN_ERR "CPU %u didn't die...\n", cpu);
+}
+#else /* ... !CONFIG_HOTPLUG_CPU */
+int __cpu_disable(void)
+{
+	return -ENOSYS;
+}
 
+void __cpu_die(unsigned int cpu)
+{
+	/* We said "no" in __cpu_disable */
+	BUG();
+}
+#endif /* CONFIG_HOTPLUG_CPU */
+
+int __devinit __cpu_up(unsigned int cpu)
+{
 	/* In case one didn't come up */
 	if (!cpu_isset(cpu, cpu_callin_map)) {
+		printk(KERN_DEBUG "skipping cpu%d, didn't come online\n", cpu);
 		local_irq_enable();
 		return -EIO;
 	}
 
+#ifdef CONFIG_HOTPLUG_CPU
+	/* Already up, and in cpu_quiescent now? */
+	if (cpu_isset(cpu, smp_commenced_mask)) {
+		cpu_enable(cpu);
+		return 0;
+	}
+#endif
+
 	local_irq_enable();
 	/* Unleash the CPU! */
 	cpu_set(cpu, smp_commenced_mask);
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index e4d4e2162c7..207ea8ba716 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -625,6 +625,14 @@ fastcall void do_nmi(struct pt_regs * regs, long error_code)
 	nmi_enter();
 
 	cpu = smp_processor_id();
+
+#ifdef CONFIG_HOTPLUG_CPU
+	if (!cpu_online(cpu)) {
+		nmi_exit();
+		return;
+	}
+#endif
+
 	++nmi_count(cpu);
 
 	if (!nmi_callback(regs, cpu))
diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c
index 3865f088ffa..a888ddc10f7 100644
--- a/arch/ia64/kernel/smpboot.c
+++ b/arch/ia64/kernel/smpboot.c
@@ -688,6 +688,7 @@ int __cpu_disable(void)
 		return -EBUSY;
 
 	remove_siblinginfo(cpu);
+	cpu_clear(cpu, cpu_online_map);
 	fixup_irqs();
 	local_flush_tlb_all();
 	cpu_clear(cpu, cpu_callin_map);
diff --git a/arch/ppc64/kernel/pSeries_smp.c b/arch/ppc64/kernel/pSeries_smp.c
index 30154140f7e..62c55a12356 100644
--- a/arch/ppc64/kernel/pSeries_smp.c
+++ b/arch/ppc64/kernel/pSeries_smp.c
@@ -93,10 +93,13 @@ static int query_cpu_stopped(unsigned int pcpu)
 
 int pSeries_cpu_disable(void)
 {
+	int cpu = smp_processor_id();
+
+	cpu_clear(cpu, cpu_online_map);
 	systemcfg->processorCount--;
 
 	/*fix boot_cpuid here*/
-	if (smp_processor_id() == boot_cpuid)
+	if (cpu == boot_cpuid)
 		boot_cpuid = any_online_cpu(cpu_online_map);
 
 	/* FIXME: abstract this to not be platform specific later on */
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index fdfcf0488b4..93c71fef99d 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -679,12 +679,14 @@ __cpu_disable(void)
 {
 	unsigned long flags;
 	ec_creg_mask_parms cr_parms;
+	int cpu = smp_processor_id();
 
 	spin_lock_irqsave(&smp_reserve_lock, flags);
-	if (smp_cpu_reserved[smp_processor_id()] != 0) {
+	if (smp_cpu_reserved[cpu] != 0) {
 		spin_unlock_irqrestore(&smp_reserve_lock, flags);
 		return -EBUSY;
 	}
+	cpu_clear(cpu, cpu_online_map);
 
 #ifdef CONFIG_PFAULT
 	/* Disable pfault pseudo page faults on this cpu. */
diff --git a/include/asm-i386/cpu.h b/include/asm-i386/cpu.h
index 002740b2195..e7252c216ca 100644
--- a/include/asm-i386/cpu.h
+++ b/include/asm-i386/cpu.h
@@ -5,6 +5,7 @@
 #include <linux/cpu.h>
 #include <linux/topology.h>
 #include <linux/nodemask.h>
+#include <linux/percpu.h>
 
 #include <asm/node.h>
 
@@ -16,4 +17,5 @@ extern int arch_register_cpu(int num);
 extern void arch_unregister_cpu(int);
 #endif
 
+DECLARE_PER_CPU(int, cpu_state);
 #endif /* _ASM_I386_CPU_H_ */
diff --git a/include/asm-i386/irq.h b/include/asm-i386/irq.h
index 05b9e61b0a7..e2d8bf23ad7 100644
--- a/include/asm-i386/irq.h
+++ b/include/asm-i386/irq.h
@@ -38,4 +38,8 @@ extern void release_vm86_irqs(struct task_struct *);
 extern int irqbalance_disable(char *str);
 #endif
 
+#ifdef CONFIG_HOTPLUG_CPU
+extern void fixup_irqs(cpumask_t map);
+#endif
+
 #endif /* _ASM_IRQ_H */
diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h
index 55ef31f66bb..507f2fd39a6 100644
--- a/include/asm-i386/smp.h
+++ b/include/asm-i386/smp.h
@@ -83,6 +83,9 @@ static __inline int logical_smp_processor_id(void)
 }
 
 #endif
+
+extern int __cpu_disable(void);
+extern void __cpu_die(unsigned int cpu);
 #endif /* !__ASSEMBLY__ */
 
 #define NO_PROC_ID		0xFF		/* No processor magic marker */
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 628f4ccda12..53d8263ae12 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -63,19 +63,15 @@ static int take_cpu_down(void *unused)
 {
 	int err;
 
-	/* Take offline: makes arch_cpu_down somewhat easier. */
-	cpu_clear(smp_processor_id(), cpu_online_map);
-
 	/* Ensure this CPU doesn't handle any more interrupts. */
 	err = __cpu_disable();
 	if (err < 0)
-		cpu_set(smp_processor_id(), cpu_online_map);
-	else
-		/* Force idle task to run as soon as we yield: it should
-		   immediately notice cpu is offline and die quickly. */
-		sched_idle_next();
+		return err;
 
-	return err;
+	/* Force idle task to run as soon as we yield: it should
+	   immediately notice cpu is offline and die quickly. */
+	sched_idle_next();
+	return 0;
 }
 
 int cpu_down(unsigned int cpu)
-- 
cgit v1.2.3-70-g09d2


From 67664c8f7e74def5adf66298a1245d82af72db2c Mon Sep 17 00:00:00 2001
From: Ashok Raj <ashok.raj@intel.com>
Date: Sat, 25 Jun 2005 14:54:52 -0700
Subject: [PATCH] i386: Dont use IPI broadcast when using cpu hotplug.

This patch introduces a startup parameter no_broadcast.  When we enable
CONFIG_HOTPLUG_CPU, we dont want to use broadcast shortcut as it has ill
effects on a offline cpu.  If we issue broadcast, the IPI is also delivered
to offline cpus, or partially up cpu causing stale IPI's to be handled,
which is a problem and can cause undesirable effects.

Introduces a new startup cmdline option no_ipi_broadcast, that can be
switched at cmdline if necessary.

Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Acked-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mach-default/setup.c           | 27 +++++++++++++++++++++++++++
 include/asm-i386/mach-default/mach_ipi.h | 27 +++++++++++++++++++++++++--
 2 files changed, 52 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/i386/mach-default/setup.c b/arch/i386/mach-default/setup.c
index 0aa08eaa893..e5a1a83d09e 100644
--- a/arch/i386/mach-default/setup.c
+++ b/arch/i386/mach-default/setup.c
@@ -10,6 +10,14 @@
 #include <asm/acpi.h>
 #include <asm/arch_hooks.h>
 
+#ifdef CONFIG_HOTPLUG_CPU
+#define DEFAULT_SEND_IPI	(1)
+#else
+#define DEFAULT_SEND_IPI	(0)
+#endif
+
+int no_broadcast=DEFAULT_SEND_IPI;
+
 /**
  * pre_intr_init_hook - initialisation prior to setting up interrupt vectors
  *
@@ -104,3 +112,22 @@ void __init mca_nmi_hook(void)
 	printk("NMI generated from unknown source!\n");
 }
 #endif
+
+static __init int no_ipi_broadcast(char *str)
+{
+	get_option(&str, &no_broadcast);
+	printk ("Using %s mode\n", no_broadcast ? "No IPI Broadcast" :
+											"IPI Broadcast");
+	return 1;
+}
+
+__setup("no_ipi_broadcast", no_ipi_broadcast);
+
+static int __init print_ipi_mode(void)
+{
+	printk ("Using IPI %s mode\n", no_broadcast ? "No-Shortcut" :
+											"Shortcut");
+	return 0;
+}
+
+late_initcall(print_ipi_mode);
diff --git a/include/asm-i386/mach-default/mach_ipi.h b/include/asm-i386/mach-default/mach_ipi.h
index 6f2b17a2008..cc756a67cd6 100644
--- a/include/asm-i386/mach-default/mach_ipi.h
+++ b/include/asm-i386/mach-default/mach_ipi.h
@@ -4,11 +4,34 @@
 void send_IPI_mask_bitmask(cpumask_t mask, int vector);
 void __send_IPI_shortcut(unsigned int shortcut, int vector);
 
+extern int no_broadcast;
+
 static inline void send_IPI_mask(cpumask_t mask, int vector)
 {
 	send_IPI_mask_bitmask(mask, vector);
 }
 
+static inline void __local_send_IPI_allbutself(int vector)
+{
+	if (no_broadcast) {
+		cpumask_t mask = cpu_online_map;
+		int this_cpu = get_cpu();
+
+		cpu_clear(this_cpu, mask);
+		send_IPI_mask(mask, vector);
+		put_cpu();
+	} else
+		__send_IPI_shortcut(APIC_DEST_ALLBUT, vector);
+}
+
+static inline void __local_send_IPI_all(int vector)
+{
+	if (no_broadcast)
+		send_IPI_mask(cpu_online_map, vector);
+	else
+		__send_IPI_shortcut(APIC_DEST_ALLINC, vector);
+}
+
 static inline void send_IPI_allbutself(int vector)
 {
 	/*
@@ -18,13 +41,13 @@ static inline void send_IPI_allbutself(int vector)
 	if (!(num_online_cpus() > 1))
 		return;
 
-	__send_IPI_shortcut(APIC_DEST_ALLBUT, vector);
+	__local_send_IPI_allbutself(vector);
 	return;
 }
 
 static inline void send_IPI_all(int vector)
 {
-	__send_IPI_shortcut(APIC_DEST_ALLINC, vector);
+	__local_send_IPI_all(vector);
 }
 
 #endif /* __ASM_MACH_IPI_H */
-- 
cgit v1.2.3-70-g09d2


From 6fe940d6c300886de4ff1454d8ffd363172af433 Mon Sep 17 00:00:00 2001
From: Li Shaohua <shaohua.li@intel.com>
Date: Sat, 25 Jun 2005 14:54:53 -0700
Subject: [PATCH] sep initializing rework

Make SEP init per-cpu, so it is hotplug safe.

Signed-off-by: Li Shaohua<shaohua.li@intel.com>
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/cpu/common.c |  3 +++
 arch/i386/kernel/smp.c        | 10 ++++++++++
 arch/i386/kernel/smpboot.c    | 11 +++++++++++
 arch/i386/kernel/sysenter.c   | 12 +++++++-----
 arch/i386/power/cpu.c         |  6 +++---
 include/asm-i386/processor.h  |  2 ++
 include/asm-i386/smp.h        |  2 ++
 7 files changed, 38 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index b9954248d0a..d58e169fbdb 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -432,6 +432,9 @@ void __init identify_cpu(struct cpuinfo_x86 *c)
 #ifdef CONFIG_X86_MCE
 	mcheck_init(c);
 #endif
+	if (c == &boot_cpu_data)
+		sysenter_setup();
+	enable_sep_cpu();
 }
 
 #ifdef CONFIG_X86_HT
diff --git a/arch/i386/kernel/smp.c b/arch/i386/kernel/smp.c
index 35f521612b2..cec4bde6716 100644
--- a/arch/i386/kernel/smp.c
+++ b/arch/i386/kernel/smp.c
@@ -495,6 +495,16 @@ struct call_data_struct {
 	int wait;
 };
 
+void lock_ipi_call_lock(void)
+{
+	spin_lock_irq(&call_lock);
+}
+
+void unlock_ipi_call_lock(void)
+{
+	spin_unlock_irq(&call_lock);
+}
+
 static struct call_data_struct * call_data;
 
 /*
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index ad74a46e9ef..c5517f33230 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -449,7 +449,18 @@ static void __init start_secondary(void *unused)
 	 * the local TLBs too.
 	 */
 	local_flush_tlb();
+
+	/*
+	 * We need to hold call_lock, so there is no inconsistency
+	 * between the time smp_call_function() determines number of
+	 * IPI receipients, and the time when the determination is made
+	 * for which cpus receive the IPI. Holding this
+	 * lock helps us to not include this cpu in a currently in progress
+	 * smp_call_function().
+	 */
+	lock_ipi_call_lock();
 	cpu_set(smp_processor_id(), cpu_online_map);
+	unlock_ipi_call_lock();
 
 	/* We can take interrupts now: we're officially "up". */
 	local_irq_enable();
diff --git a/arch/i386/kernel/sysenter.c b/arch/i386/kernel/sysenter.c
index 960d8bd137d..0bada1870bd 100644
--- a/arch/i386/kernel/sysenter.c
+++ b/arch/i386/kernel/sysenter.c
@@ -21,11 +21,16 @@
 
 extern asmlinkage void sysenter_entry(void);
 
-void enable_sep_cpu(void *info)
+void enable_sep_cpu(void)
 {
 	int cpu = get_cpu();
 	struct tss_struct *tss = &per_cpu(init_tss, cpu);
 
+	if (!boot_cpu_has(X86_FEATURE_SEP)) {
+		put_cpu();
+		return;
+	}
+
 	tss->ss1 = __KERNEL_CS;
 	tss->esp1 = sizeof(struct tss_struct) + (unsigned long) tss;
 	wrmsr(MSR_IA32_SYSENTER_CS, __KERNEL_CS, 0);
@@ -41,7 +46,7 @@ void enable_sep_cpu(void *info)
 extern const char vsyscall_int80_start, vsyscall_int80_end;
 extern const char vsyscall_sysenter_start, vsyscall_sysenter_end;
 
-static int __init sysenter_setup(void)
+int __init sysenter_setup(void)
 {
 	void *page = (void *)get_zeroed_page(GFP_ATOMIC);
 
@@ -58,8 +63,5 @@ static int __init sysenter_setup(void)
 	       &vsyscall_sysenter_start,
 	       &vsyscall_sysenter_end - &vsyscall_sysenter_start);
 
-	on_each_cpu(enable_sep_cpu, NULL, 1, 1);
 	return 0;
 }
-
-__initcall(sysenter_setup);
diff --git a/arch/i386/power/cpu.c b/arch/i386/power/cpu.c
index 6f521cf19a1..d099d01461f 100644
--- a/arch/i386/power/cpu.c
+++ b/arch/i386/power/cpu.c
@@ -22,9 +22,11 @@
 #include <linux/device.h>
 #include <linux/suspend.h>
 #include <linux/acpi.h>
+
 #include <asm/uaccess.h>
 #include <asm/acpi.h>
 #include <asm/tlbflush.h>
+#include <asm/processor.h>
 
 static struct saved_context saved_context;
 
@@ -33,8 +35,6 @@ unsigned long saved_context_esp, saved_context_ebp;
 unsigned long saved_context_esi, saved_context_edi;
 unsigned long saved_context_eflags;
 
-extern void enable_sep_cpu(void *);
-
 void __save_processor_state(struct saved_context *ctxt)
 {
 	kernel_fpu_begin();
@@ -136,7 +136,7 @@ void __restore_processor_state(struct saved_context *ctxt)
 	 * sysenter MSRs
 	 */
 	if (boot_cpu_has(X86_FEATURE_SEP))
-		enable_sep_cpu(NULL);
+		enable_sep_cpu();
 
 	fix_processor_context();
 	do_fpu_end();
diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h
index c76c50e9622..6f0f93d0d41 100644
--- a/include/asm-i386/processor.h
+++ b/include/asm-i386/processor.h
@@ -691,5 +691,7 @@ extern void select_idle_routine(const struct cpuinfo_x86 *c);
 #define cache_line_size() (boot_cpu_data.x86_cache_alignment)
 
 extern unsigned long boot_option_idle_override;
+extern void enable_sep_cpu(void);
+extern int sysenter_setup(void);
 
 #endif /* __ASM_I386_PROCESSOR_H */
diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h
index 507f2fd39a6..2451ead0ca5 100644
--- a/include/asm-i386/smp.h
+++ b/include/asm-i386/smp.h
@@ -42,6 +42,8 @@ extern void smp_message_irq(int cpl, void *dev_id, struct pt_regs *regs);
 extern void smp_invalidate_rcv(void);		/* Process an NMI */
 extern void (*mtrr_hook) (void);
 extern void zap_low_mappings (void);
+extern void lock_ipi_call_lock(void);
+extern void unlock_ipi_call_lock(void);
 
 #define MAX_APICID 256
 extern u8 x86_cpu_to_apicid[];
-- 
cgit v1.2.3-70-g09d2


From e1367daf3eed5cd619ee88c9907e1e6ddaa58406 Mon Sep 17 00:00:00 2001
From: Li Shaohua <shaohua.li@intel.com>
Date: Sat, 25 Jun 2005 14:54:56 -0700
Subject: [PATCH] cpu state clean after hot remove

Clean CPU states in order to reuse smp boot code for CPU hotplug.

Signed-off-by: Li Shaohua<shaohua.li@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/cpu/common.c |  12 +++
 arch/i386/kernel/irq.c        |   5 ++
 arch/i386/kernel/process.c    |  20 +++--
 arch/i386/kernel/smpboot.c    | 175 ++++++++++++++++++++++++++++++++++--------
 drivers/base/cpu.c            |   8 +-
 include/asm-i386/irq.h        |   2 +
 include/asm-i386/smp.h        |   8 ++
 7 files changed, 187 insertions(+), 43 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index aac74758caf..2203a9d2021 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -651,3 +651,15 @@ void __devinit cpu_init(void)
 	clear_used_math();
 	mxcsr_feature_mask_init();
 }
+
+#ifdef CONFIG_HOTPLUG_CPU
+void __devinit cpu_uninit(void)
+{
+	int cpu = raw_smp_processor_id();
+	cpu_clear(cpu, cpu_initialized);
+
+	/* lazy TLB state */
+	per_cpu(cpu_tlbstate, cpu).state = 0;
+	per_cpu(cpu_tlbstate, cpu).active_mm = &init_mm;
+}
+#endif
diff --git a/arch/i386/kernel/irq.c b/arch/i386/kernel/irq.c
index af115004aec..ce66dcc26d9 100644
--- a/arch/i386/kernel/irq.c
+++ b/arch/i386/kernel/irq.c
@@ -156,6 +156,11 @@ void irq_ctx_init(int cpu)
 		cpu,hardirq_ctx[cpu],softirq_ctx[cpu]);
 }
 
+void irq_ctx_exit(int cpu)
+{
+	hardirq_ctx[cpu] = NULL;
+}
+
 extern asmlinkage void __do_softirq(void);
 
 asmlinkage void do_softirq(void)
diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index e06f2dc7123..5f8cfa6b794 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -152,21 +152,19 @@ static void poll_idle (void)
 /* We don't actually take CPU down, just spin without interrupts. */
 static inline void play_dead(void)
 {
+	/* This must be done before dead CPU ack */
+	cpu_exit_clear();
+	wbinvd();
+	mb();
 	/* Ack it */
 	__get_cpu_var(cpu_state) = CPU_DEAD;
 
-	/* We shouldn't have to disable interrupts while dead, but
-	 * some interrupts just don't seem to go away, and this makes
-	 * it "work" for testing purposes. */
-	/* Death loop */
-	while (__get_cpu_var(cpu_state) != CPU_UP_PREPARE)
-		cpu_relax();
-
+	/*
+	 * With physical CPU hotplug, we should halt the cpu
+	 */
 	local_irq_disable();
-	__flush_tlb_all();
-	cpu_set(smp_processor_id(), cpu_online_map);
-	enable_APIC_timer();
-	local_irq_enable();
+	while (1)
+		__asm__ __volatile__("hlt":::"memory");
 }
 #else
 static inline void play_dead(void)
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index fb0b200d1d8..d66bf489a2e 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -90,6 +90,12 @@ cpumask_t cpu_callout_map;
 EXPORT_SYMBOL(cpu_callout_map);
 static cpumask_t smp_commenced_mask;
 
+/* TSC's upper 32 bits can't be written in eariler CPU (before prescott), there
+ * is no way to resync one AP against BP. TBD: for prescott and above, we
+ * should use IA64's algorithm
+ */
+static int __devinitdata tsc_sync_disabled;
+
 /* Per CPU bogomips and other parameters */
 struct cpuinfo_x86 cpu_data[NR_CPUS] __cacheline_aligned;
 EXPORT_SYMBOL(cpu_data);
@@ -427,7 +433,7 @@ static void __devinit smp_callin(void)
 	/*
 	 *      Synchronize the TSC with the BP
 	 */
-	if (cpu_has_tsc && cpu_khz)
+	if (cpu_has_tsc && cpu_khz && !tsc_sync_disabled)
 		synchronize_tsc_ap();
 }
 
@@ -507,6 +513,7 @@ static void __devinit start_secondary(void *unused)
 	lock_ipi_call_lock();
 	cpu_set(smp_processor_id(), cpu_online_map);
 	unlock_ipi_call_lock();
+	per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE;
 
 	/* We can take interrupts now: we're officially "up". */
 	local_irq_enable();
@@ -816,8 +823,43 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip)
 #endif	/* WAKE_SECONDARY_VIA_INIT */
 
 extern cpumask_t cpu_initialized;
+static inline int alloc_cpu_id(void)
+{
+	cpumask_t	tmp_map;
+	int cpu;
+	cpus_complement(tmp_map, cpu_present_map);
+	cpu = first_cpu(tmp_map);
+	if (cpu >= NR_CPUS)
+		return -ENODEV;
+	return cpu;
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+static struct task_struct * __devinitdata cpu_idle_tasks[NR_CPUS];
+static inline struct task_struct * alloc_idle_task(int cpu)
+{
+	struct task_struct *idle;
+
+	if ((idle = cpu_idle_tasks[cpu]) != NULL) {
+		/* initialize thread_struct.  we really want to avoid destroy
+		 * idle tread
+		 */
+		idle->thread.esp = (unsigned long)(((struct pt_regs *)
+			(THREAD_SIZE + (unsigned long) idle->thread_info)) - 1);
+		init_idle(idle, cpu);
+		return idle;
+	}
+	idle = fork_idle(cpu);
+
+	if (!IS_ERR(idle))
+		cpu_idle_tasks[cpu] = idle;
+	return idle;
+}
+#else
+#define alloc_idle_task(cpu) fork_idle(cpu)
+#endif
 
-static int __devinit do_boot_cpu(int apicid)
+static int __devinit do_boot_cpu(int apicid, int cpu)
 /*
  * NOTE - on most systems this is a PHYSICAL apic ID, but on multiquad
  * (ie clustered apic addressing mode), this is a LOGICAL apic ID.
@@ -826,16 +868,17 @@ static int __devinit do_boot_cpu(int apicid)
 {
 	struct task_struct *idle;
 	unsigned long boot_error;
-	int timeout, cpu;
+	int timeout;
 	unsigned long start_eip;
 	unsigned short nmi_high = 0, nmi_low = 0;
 
-	cpu = ++cpucount;
+	++cpucount;
+
 	/*
 	 * We can't use kernel_thread since we must avoid to
 	 * reschedule the child.
 	 */
-	idle = fork_idle(cpu);
+	idle = alloc_idle_task(cpu);
 	if (IS_ERR(idle))
 		panic("failed fork for CPU %d", cpu);
 	idle->thread.eip = (unsigned long) start_secondary;
@@ -902,13 +945,16 @@ static int __devinit do_boot_cpu(int apicid)
 			inquire_remote_apic(apicid);
 		}
 	}
-	x86_cpu_to_apicid[cpu] = apicid;
+
 	if (boot_error) {
 		/* Try to put things back the way they were before ... */
 		unmap_cpu_to_logical_apicid(cpu);
 		cpu_clear(cpu, cpu_callout_map); /* was set here (do_boot_cpu()) */
 		cpu_clear(cpu, cpu_initialized); /* was set by cpu_init() */
 		cpucount--;
+	} else {
+		x86_cpu_to_apicid[cpu] = apicid;
+		cpu_set(cpu, cpu_present_map);
 	}
 
 	/* mark "stuck" area as not stuck */
@@ -917,6 +963,75 @@ static int __devinit do_boot_cpu(int apicid)
 	return boot_error;
 }
 
+#ifdef CONFIG_HOTPLUG_CPU
+void cpu_exit_clear(void)
+{
+	int cpu = raw_smp_processor_id();
+
+	idle_task_exit();
+
+	cpucount --;
+	cpu_uninit();
+	irq_ctx_exit(cpu);
+
+	cpu_clear(cpu, cpu_callout_map);
+	cpu_clear(cpu, cpu_callin_map);
+	cpu_clear(cpu, cpu_present_map);
+
+	cpu_clear(cpu, smp_commenced_mask);
+	unmap_cpu_to_logical_apicid(cpu);
+}
+
+struct warm_boot_cpu_info {
+	struct completion *complete;
+	int apicid;
+	int cpu;
+};
+
+static void __devinit do_warm_boot_cpu(void *p)
+{
+	struct warm_boot_cpu_info *info = p;
+	do_boot_cpu(info->apicid, info->cpu);
+	complete(info->complete);
+}
+
+int __devinit smp_prepare_cpu(int cpu)
+{
+	DECLARE_COMPLETION(done);
+	struct warm_boot_cpu_info info;
+	struct work_struct task;
+	int	apicid, ret;
+
+	lock_cpu_hotplug();
+	apicid = x86_cpu_to_apicid[cpu];
+	if (apicid == BAD_APICID) {
+		ret = -ENODEV;
+		goto exit;
+	}
+
+	info.complete = &done;
+	info.apicid = apicid;
+	info.cpu = cpu;
+	INIT_WORK(&task, do_warm_boot_cpu, &info);
+
+	tsc_sync_disabled = 1;
+
+	/* init low mem mapping */
+	memcpy(swapper_pg_dir, swapper_pg_dir + USER_PGD_PTRS,
+			sizeof(swapper_pg_dir[0]) * KERNEL_PGD_PTRS);
+	flush_tlb_all();
+	schedule_work(&task);
+	wait_for_completion(&done);
+
+	tsc_sync_disabled = 0;
+	zap_low_mappings();
+	ret = 0;
+exit:
+	unlock_cpu_hotplug();
+	return ret;
+}
+#endif
+
 static void smp_tune_scheduling (void)
 {
 	unsigned long cachesize;       /* kB   */
@@ -1069,7 +1184,7 @@ static void __init smp_boot_cpus(unsigned int max_cpus)
 		if (max_cpus <= cpucount+1)
 			continue;
 
-		if (do_boot_cpu(apicid))
+		if (((cpu = alloc_cpu_id()) <= 0) || do_boot_cpu(apicid, cpu))
 			printk("CPU #%d not responding - cannot use it.\n",
 								apicid);
 		else
@@ -1149,25 +1264,24 @@ void __devinit smp_prepare_boot_cpu(void)
 {
 	cpu_set(smp_processor_id(), cpu_online_map);
 	cpu_set(smp_processor_id(), cpu_callout_map);
+	cpu_set(smp_processor_id(), cpu_present_map);
+	per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE;
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
-
-/* must be called with the cpucontrol mutex held */
-static int __devinit cpu_enable(unsigned int cpu)
+static void
+remove_siblinginfo(int cpu)
 {
-	/* get the target out of its holding state */
-	per_cpu(cpu_state, cpu) = CPU_UP_PREPARE;
-	wmb();
-
-	/* wait for the processor to ack it. timeout? */
-	while (!cpu_online(cpu))
-		cpu_relax();
-
-	fixup_irqs(cpu_online_map);
-	/* counter the disable in fixup_irqs() */
-	local_irq_enable();
-	return 0;
+	int sibling;
+
+	for_each_cpu_mask(sibling, cpu_sibling_map[cpu])
+		cpu_clear(cpu, cpu_sibling_map[sibling]);
+	for_each_cpu_mask(sibling, cpu_core_map[cpu])
+		cpu_clear(cpu, cpu_core_map[sibling]);
+	cpus_clear(cpu_sibling_map[cpu]);
+	cpus_clear(cpu_core_map[cpu]);
+	phys_proc_id[cpu] = BAD_APICID;
+	cpu_core_id[cpu] = BAD_APICID;
 }
 
 int __cpu_disable(void)
@@ -1193,6 +1307,8 @@ int __cpu_disable(void)
 	mdelay(1);
 	local_irq_disable();
 
+	remove_siblinginfo(cpu);
+
 	cpu_clear(cpu, map);
 	fixup_irqs(map);
 	/* It's now safe to remove this processor from the online map */
@@ -1207,8 +1323,10 @@ void __cpu_die(unsigned int cpu)
 
 	for (i = 0; i < 10; i++) {
 		/* They ack this in play_dead by setting CPU_DEAD */
-		if (per_cpu(cpu_state, cpu) == CPU_DEAD)
+		if (per_cpu(cpu_state, cpu) == CPU_DEAD) {
+			printk ("CPU %d is now offline\n", cpu);
 			return;
+		}
 		current->state = TASK_UNINTERRUPTIBLE;
 		schedule_timeout(HZ/10);
 	}
@@ -1236,15 +1354,8 @@ int __devinit __cpu_up(unsigned int cpu)
 		return -EIO;
 	}
 
-#ifdef CONFIG_HOTPLUG_CPU
-	/* Already up, and in cpu_quiescent now? */
-	if (cpu_isset(cpu, smp_commenced_mask)) {
-		cpu_enable(cpu);
-		return 0;
-	}
-#endif
-
 	local_irq_enable();
+	per_cpu(cpu_state, cpu) = CPU_UP_PREPARE;
 	/* Unleash the CPU! */
 	cpu_set(cpu, smp_commenced_mask);
 	while (!cpu_isset(cpu, cpu_online_map))
@@ -1258,10 +1369,12 @@ void __init smp_cpus_done(unsigned int max_cpus)
 	setup_ioapic_dest();
 #endif
 	zap_low_mappings();
+#ifndef CONFIG_HOTPLUG_CPU
 	/*
 	 * Disable executability of the SMP trampoline:
 	 */
 	set_kernel_exec((unsigned long)trampoline_base, trampoline_exec);
+#endif
 }
 
 void __init smp_intr_init(void)
diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c
index 6ef3069b571..bdd7e9f55c8 100644
--- a/drivers/base/cpu.c
+++ b/drivers/base/cpu.c
@@ -16,6 +16,10 @@ struct sysdev_class cpu_sysdev_class = {
 EXPORT_SYMBOL(cpu_sysdev_class);
 
 #ifdef CONFIG_HOTPLUG_CPU
+#ifndef __HAVE_ARCH_SMP_PREPARE_CPU
+#define smp_prepare_cpu(cpu) (0)
+#endif
+
 static ssize_t show_online(struct sys_device *dev, char *buf)
 {
 	struct cpu *cpu = container_of(dev, struct cpu, sysdev);
@@ -36,7 +40,9 @@ static ssize_t store_online(struct sys_device *dev, const char *buf,
 			kobject_hotplug(&dev->kobj, KOBJ_OFFLINE);
 		break;
 	case '1':
-		ret = cpu_up(cpu->sysdev.id);
+		ret = smp_prepare_cpu(cpu->sysdev.id);
+		if (ret == 0)
+			ret = cpu_up(cpu->sysdev.id);
 		break;
 	default:
 		ret = -EINVAL;
diff --git a/include/asm-i386/irq.h b/include/asm-i386/irq.h
index e2d8bf23ad7..270f1986b19 100644
--- a/include/asm-i386/irq.h
+++ b/include/asm-i386/irq.h
@@ -29,9 +29,11 @@ extern void release_vm86_irqs(struct task_struct *);
 
 #ifdef CONFIG_4KSTACKS
   extern void irq_ctx_init(int cpu);
+  extern void irq_ctx_exit(int cpu);
 # define __ARCH_HAS_DO_SOFTIRQ
 #else
 # define irq_ctx_init(cpu) do { } while (0)
+# define irq_ctx_exit(cpu) do { } while (0)
 #endif
 
 #ifdef CONFIG_IRQBALANCE
diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h
index 2451ead0ca5..c9996eda540 100644
--- a/include/asm-i386/smp.h
+++ b/include/asm-i386/smp.h
@@ -48,6 +48,14 @@ extern void unlock_ipi_call_lock(void);
 #define MAX_APICID 256
 extern u8 x86_cpu_to_apicid[];
 
+#ifdef CONFIG_HOTPLUG_CPU
+extern void cpu_exit_clear(void);
+extern void cpu_uninit(void);
+
+#define __HAVE_ARCH_SMP_PREPARE_CPU
+extern int smp_prepare_cpu(int cpu);
+#endif
+
 /*
  * This function is needed by all SMP systems. It must _always_ be valid
  * from the initial startup. We map APIC_BASE very early in page_setup(),
-- 
cgit v1.2.3-70-g09d2


From 52a119feaad92d44a0e97d01b22afbcbaf3fc079 Mon Sep 17 00:00:00 2001
From: Ashok Raj <ashok.raj@intel.com>
Date: Sat, 25 Jun 2005 14:54:57 -0700
Subject: [PATCH] make smp_prepare_cpu to a weak function

I really wish smp_prepare_cpu() would disappear eventually.  In the interim
this is ideally a weak function, so we dont end up changing several places
to define this dummy in headers.

Today since the dummy declaration is done only in drivers/base/cpu.c but
the function is called in kernel/power/smp.c i get undefined reference in
my cpu hotplug code for x86_64 under development.

Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/base/cpu.c     | 9 +++++----
 include/asm-i386/smp.h | 3 ---
 include/linux/cpu.h    | 1 +
 3 files changed, 6 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c
index bdd7e9f55c8..0bf2dc11cdb 100644
--- a/drivers/base/cpu.c
+++ b/drivers/base/cpu.c
@@ -16,9 +16,10 @@ struct sysdev_class cpu_sysdev_class = {
 EXPORT_SYMBOL(cpu_sysdev_class);
 
 #ifdef CONFIG_HOTPLUG_CPU
-#ifndef __HAVE_ARCH_SMP_PREPARE_CPU
-#define smp_prepare_cpu(cpu) (0)
-#endif
+int __attribute__((weak)) smp_prepare_cpu (int cpu)
+{
+	return 0;
+}
 
 static ssize_t show_online(struct sys_device *dev, char *buf)
 {
@@ -41,7 +42,7 @@ static ssize_t store_online(struct sys_device *dev, const char *buf,
 		break;
 	case '1':
 		ret = smp_prepare_cpu(cpu->sysdev.id);
-		if (ret == 0)
+		if (!ret)
 			ret = cpu_up(cpu->sysdev.id);
 		break;
 	default:
diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h
index c9996eda540..edad9b4712f 100644
--- a/include/asm-i386/smp.h
+++ b/include/asm-i386/smp.h
@@ -51,9 +51,6 @@ extern u8 x86_cpu_to_apicid[];
 #ifdef CONFIG_HOTPLUG_CPU
 extern void cpu_exit_clear(void);
 extern void cpu_uninit(void);
-
-#define __HAVE_ARCH_SMP_PREPARE_CPU
-extern int smp_prepare_cpu(int cpu);
 #endif
 
 /*
diff --git a/include/linux/cpu.h b/include/linux/cpu.h
index fe0298e5dae..e8904c0da68 100644
--- a/include/linux/cpu.h
+++ b/include/linux/cpu.h
@@ -69,6 +69,7 @@ extern struct semaphore cpucontrol;
 	register_cpu_notifier(&fn##_nb);			\
 }
 int cpu_down(unsigned int cpu);
+extern int __attribute__((weak)) smp_prepare_cpu(int cpu);
 #define cpu_is_offline(cpu) unlikely(!cpu_online(cpu))
 #else
 #define lock_cpu_hotplug()	do { } while (0)
-- 
cgit v1.2.3-70-g09d2


From e6982c671c560da4a0bc5c908cbcbec12bd5991d Mon Sep 17 00:00:00 2001
From: Ashok Raj <ashok.raj@intel.com>
Date: Sat, 25 Jun 2005 14:54:58 -0700
Subject: [PATCH] x86_64: Change init sections for CPU hotplug support

This patch adds __cpuinit and __cpuinitdata sections that need to exist past
boot to support cpu hotplug.

Caveat: This is done *only* for EM64T CPU Hotplug support, on request from
Andi Kleen.  Much of the generic hotplug code in kernel, and none of the other
archs that support CPU hotplug today, i386, ia64, ppc64, s390 and parisc dont
mark sections with __cpuinit, but only mark them as __devinit, and
__devinitdata.

If someone is motivated to change generic code, we need to make sure all
existing hotplug code does not break, on other arch's that dont use __cpuinit,
and __cpudevinit.

Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Acked-by: Andi Kleen <ak@muc.de>
Acked-by: Zwane Mwaikambo <zwane@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/apic.c      |  8 ++++----
 arch/x86_64/kernel/i387.c      |  2 +-
 arch/x86_64/kernel/mce.c       |  8 ++++----
 arch/x86_64/kernel/mce_intel.c |  4 ++--
 arch/x86_64/kernel/nmi.c       |  4 ++--
 arch/x86_64/kernel/process.c   |  2 +-
 arch/x86_64/kernel/setup.c     | 18 +++++++++---------
 arch/x86_64/kernel/setup64.c   |  6 +++---
 arch/x86_64/kernel/smpboot.c   | 15 +++++----------
 arch/x86_64/mm/numa.c          |  2 +-
 include/linux/init.h           | 12 ++++++++++++
 11 files changed, 44 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index f8e6cc4fecd..2a6c893ccf6 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -285,7 +285,7 @@ void __init init_bsp_APIC(void)
 	apic_write_around(APIC_LVT1, value);
 }
 
-void __init setup_local_APIC (void)
+void __cpuinit setup_local_APIC (void)
 {
 	unsigned int value, ver, maxlvt;
 
@@ -534,7 +534,7 @@ static struct sys_device device_lapic = {
 	.cls		= &lapic_sysclass,
 };
 
-static void __init apic_pm_activate(void)
+static void __cpuinit apic_pm_activate(void)
 {
 	apic_pm_state.active = 1;
 }
@@ -774,14 +774,14 @@ void __init setup_boot_APIC_clock (void)
 	local_irq_enable();
 }
 
-void __init setup_secondary_APIC_clock(void)
+void __cpuinit setup_secondary_APIC_clock(void)
 {
 	local_irq_disable(); /* FIXME: Do we need this? --RR */
 	setup_APIC_timer(calibration_result);
 	local_irq_enable();
 }
 
-void __init disable_APIC_timer(void)
+void __cpuinit disable_APIC_timer(void)
 {
 	if (using_apic_timer) {
 		unsigned long v;
diff --git a/arch/x86_64/kernel/i387.c b/arch/x86_64/kernel/i387.c
index ba139cac57c..d9b22b633e3 100644
--- a/arch/x86_64/kernel/i387.c
+++ b/arch/x86_64/kernel/i387.c
@@ -42,7 +42,7 @@ void mxcsr_feature_mask_init(void)
  * Called at bootup to set up the initial FPU state that is later cloned
  * into all processes.
  */
-void __init fpu_init(void)
+void __cpuinit fpu_init(void)
 {
 	unsigned long oldcr0 = read_cr0();
 	extern void __bad_fxsave_alignment(void);
diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c
index 3a89d735a4f..7ab15c8ab95 100644
--- a/arch/x86_64/kernel/mce.c
+++ b/arch/x86_64/kernel/mce.c
@@ -327,7 +327,7 @@ static void mce_init(void *dummy)
 }
 
 /* Add per CPU specific workarounds here */
-static void __init mce_cpu_quirks(struct cpuinfo_x86 *c) 
+static void __cpuinit mce_cpu_quirks(struct cpuinfo_x86 *c)
 { 
 	/* This should be disabled by the BIOS, but isn't always */
 	if (c->x86_vendor == X86_VENDOR_AMD && c->x86 == 15) {
@@ -337,7 +337,7 @@ static void __init mce_cpu_quirks(struct cpuinfo_x86 *c)
 	}
 }			
 
-static void __init mce_cpu_features(struct cpuinfo_x86 *c)
+static void __cpuinit mce_cpu_features(struct cpuinfo_x86 *c)
 {
 	switch (c->x86_vendor) {
 	case X86_VENDOR_INTEL:
@@ -352,7 +352,7 @@ static void __init mce_cpu_features(struct cpuinfo_x86 *c)
  * Called for each booted CPU to set up machine checks.
  * Must be called with preempt off. 
  */
-void __init mcheck_init(struct cpuinfo_x86 *c)
+void __cpuinit mcheck_init(struct cpuinfo_x86 *c)
 {
 	static cpumask_t mce_cpus __initdata = CPU_MASK_NONE;
 
@@ -542,7 +542,7 @@ ACCESSOR(bank4ctl,bank[4],mce_restart())
 ACCESSOR(tolerant,tolerant,)
 ACCESSOR(check_interval,check_interval,mce_restart())
 
-static __init int mce_init_device(void)
+static __cpuinit int mce_init_device(void)
 {
 	int err;
 	if (!mce_available(&boot_cpu_data))
diff --git a/arch/x86_64/kernel/mce_intel.c b/arch/x86_64/kernel/mce_intel.c
index 4db9a640069..0be0a795981 100644
--- a/arch/x86_64/kernel/mce_intel.c
+++ b/arch/x86_64/kernel/mce_intel.c
@@ -42,7 +42,7 @@ done:
 	irq_exit();
 }
 
-static void __init intel_init_thermal(struct cpuinfo_x86 *c)
+static void __cpuinit intel_init_thermal(struct cpuinfo_x86 *c)
 {
 	u32 l, h;
 	int tm2 = 0;
@@ -93,7 +93,7 @@ static void __init intel_init_thermal(struct cpuinfo_x86 *c)
 	return;
 }
 
-void __init mce_intel_feature_init(struct cpuinfo_x86 *c)
+void __cpuinit mce_intel_feature_init(struct cpuinfo_x86 *c)
 {
 	intel_init_thermal(c);
 }
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 31c0f2e6ac9..4e44d6e6b7e 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -98,7 +98,7 @@ static unsigned int nmi_p4_cccr_val;
 	(P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT|	\
 	 P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE)
 
-static __init inline int nmi_known_cpu(void)
+static __cpuinit inline int nmi_known_cpu(void)
 {
 	switch (boot_cpu_data.x86_vendor) {
 	case X86_VENDOR_AMD:
@@ -110,7 +110,7 @@ static __init inline int nmi_known_cpu(void)
 }
 
 /* Run after command line and cpu_init init, but before all other checks */
-void __init nmi_watchdog_default(void)
+void __cpuinit nmi_watchdog_default(void)
 {
 	if (nmi_watchdog != NMI_DEFAULT)
 		return;
diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index e59d1f9d616..15fa35a81a7 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -204,7 +204,7 @@ static void mwait_idle(void)
 	}
 }
 
-void __init select_idle_routine(const struct cpuinfo_x86 *c)
+void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c)
 {
 	static int printed;
 	if (cpu_has(c, X86_FEATURE_MWAIT)) {
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 000015dd5a8..ca9a915a6d4 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -676,7 +676,7 @@ void __init setup_arch(char **cmdline_p)
 #endif
 }
 
-static int __init get_model_name(struct cpuinfo_x86 *c)
+static int __cpuinit get_model_name(struct cpuinfo_x86 *c)
 {
 	unsigned int *v;
 
@@ -692,7 +692,7 @@ static int __init get_model_name(struct cpuinfo_x86 *c)
 }
 
 
-static void __init display_cacheinfo(struct cpuinfo_x86 *c)
+static void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
 {
 	unsigned int n, dummy, eax, ebx, ecx, edx;
 
@@ -803,7 +803,7 @@ static int __init init_amd(struct cpuinfo_x86 *c)
 	return r;
 }
 
-static void __init detect_ht(struct cpuinfo_x86 *c)
+static void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 {
 #ifdef CONFIG_SMP
 	u32 	eax, ebx, ecx, edx;
@@ -864,7 +864,7 @@ static void __init detect_ht(struct cpuinfo_x86 *c)
 /*
  * find out the number of processor cores on the die
  */
-static int __init intel_num_cpu_cores(struct cpuinfo_x86 *c)
+static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c)
 {
 	unsigned int eax;
 
@@ -882,7 +882,7 @@ static int __init intel_num_cpu_cores(struct cpuinfo_x86 *c)
 		return 1;
 }
 
-static void __init init_intel(struct cpuinfo_x86 *c)
+static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 {
 	/* Cache sizes */
 	unsigned n;
@@ -902,7 +902,7 @@ static void __init init_intel(struct cpuinfo_x86 *c)
  	c->x86_num_cores = intel_num_cpu_cores(c);
 }
 
-void __init get_cpu_vendor(struct cpuinfo_x86 *c)
+void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c)
 {
 	char *v = c->x86_vendor_id;
 
@@ -923,7 +923,7 @@ struct cpu_model_info {
 /* Do some early cpuid on the boot CPU to get some parameter that are
    needed before check_bugs. Everything advanced is in identify_cpu
    below. */
-void __init early_identify_cpu(struct cpuinfo_x86 *c)
+void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
 {
 	u32 tfms;
 
@@ -977,7 +977,7 @@ void __init early_identify_cpu(struct cpuinfo_x86 *c)
 /*
  * This does the hard work of actually picking apart the CPU stuff...
  */
-void __init identify_cpu(struct cpuinfo_x86 *c)
+void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 {
 	int i;
 	u32 xlvl;
@@ -1054,7 +1054,7 @@ void __init identify_cpu(struct cpuinfo_x86 *c)
 }
  
 
-void __init print_cpu_info(struct cpuinfo_x86 *c)
+void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
 {
 	if (c->x86_model_id[0])
 		printk("%s", c->x86_model_id);
diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c
index 678b7ac33b8..f2b9c6bc999 100644
--- a/arch/x86_64/kernel/setup64.c
+++ b/arch/x86_64/kernel/setup64.c
@@ -29,7 +29,7 @@
 
 char x86_boot_params[BOOT_PARAM_SIZE] __initdata = {0,};
 
-cpumask_t cpu_initialized __initdata = CPU_MASK_NONE;
+cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
 
 struct x8664_pda cpu_pda[NR_CPUS] __cacheline_aligned; 
 
@@ -171,7 +171,7 @@ void syscall_init(void)
 	wrmsrl(MSR_SYSCALL_MASK, EF_TF|EF_DF|EF_IE|0x3000); 
 }
 
-void __init check_efer(void)
+void __cpuinit check_efer(void)
 {
 	unsigned long efer;
 
@@ -188,7 +188,7 @@ void __init check_efer(void)
  * 'CPU state barrier', nothing should get across.
  * A lot of state is already set up in PDA init.
  */
-void __init cpu_init (void)
+void __cpuinit cpu_init (void)
 {
 #ifdef CONFIG_SMP
 	int cpu = stack_smp_processor_id();
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index f1ec0f34594..bc98a6722cb 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -58,11 +58,6 @@
 #include <asm/proto.h>
 #include <asm/nmi.h>
 
-/* Change for real CPU hotplug. Note other files need to be fixed
-   first too. */
-#define __cpuinit __init
-#define __cpuinitdata __initdata
-
 /* Number of siblings per CPU package */
 int smp_num_siblings = 1;
 /* Package ID of each logical CPU */
@@ -823,7 +818,7 @@ static __cpuinit void smp_cleanup_boot(void)
  *
  * RED-PEN audit/test this more. I bet there is more state messed up here.
  */
-static __cpuinit void disable_smp(void)
+static __init void disable_smp(void)
 {
 	cpu_present_map = cpumask_of_cpu(0);
 	cpu_possible_map = cpumask_of_cpu(0);
@@ -838,7 +833,7 @@ static __cpuinit void disable_smp(void)
 /*
  * Handle user cpus=... parameter.
  */
-static __cpuinit void enforce_max_cpus(unsigned max_cpus)
+static __init void enforce_max_cpus(unsigned max_cpus)
 {
 	int i, k;
 	k = 0;
@@ -855,7 +850,7 @@ static __cpuinit void enforce_max_cpus(unsigned max_cpus)
 /*
  * Various sanity checks.
  */
-static int __cpuinit smp_sanity_check(unsigned max_cpus)
+static int __init smp_sanity_check(unsigned max_cpus)
 {
 	if (!physid_isset(hard_smp_processor_id(), phys_cpu_present_map)) {
 		printk("weird, boot CPU (#%d) not listed by the BIOS.\n",
@@ -913,7 +908,7 @@ static int __cpuinit smp_sanity_check(unsigned max_cpus)
  * Prepare for SMP bootup.  The MP table or ACPI has been read
  * earlier.  Just do some sanity checking here and enable APIC mode.
  */
-void __cpuinit smp_prepare_cpus(unsigned int max_cpus)
+void __init smp_prepare_cpus(unsigned int max_cpus)
 {
 	int i;
 
@@ -1019,7 +1014,7 @@ int __cpuinit __cpu_up(unsigned int cpu)
 /*
  * Finish the SMP boot.
  */
-void __cpuinit smp_cpus_done(unsigned int max_cpus)
+void __init smp_cpus_done(unsigned int max_cpus)
 {
 	zap_low_mappings();
 	smp_cleanup_boot();
diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c
index 84cde796ecb..ac61c186eb0 100644
--- a/arch/x86_64/mm/numa.c
+++ b/arch/x86_64/mm/numa.c
@@ -251,7 +251,7 @@ void __init numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
 	setup_node_bootmem(0, start_pfn << PAGE_SHIFT, end_pfn << PAGE_SHIFT);
 }
 
-__init void numa_add_cpu(int cpu)
+__cpuinit void numa_add_cpu(int cpu)
 {
 	/* BP is initialized elsewhere */
 	if (cpu) 
diff --git a/include/linux/init.h b/include/linux/init.h
index 05c83e0521c..59008c3826c 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -229,6 +229,18 @@ void __init parse_early_param(void);
 #define __devexitdata __exitdata
 #endif
 
+#ifdef CONFIG_HOTPLUG_CPU
+#define __cpuinit
+#define __cpuinitdata
+#define __cpuexit
+#define __cpuexitdata
+#else
+#define __cpuinit	__init
+#define __cpuinitdata __initdata
+#define __cpuexit __exit
+#define __cpuexitdata	__exitdata
+#endif
+
 /* Functions marked as __devexit may be discarded at kernel link time, depending
    on config options.  Newer versions of binutils detect references from
    retained sections to discarded sections and flag an error.  Pointers to
-- 
cgit v1.2.3-70-g09d2


From 76e4f660d9f4c6d1bb473f72be2988c35eaca948 Mon Sep 17 00:00:00 2001
From: Ashok Raj <ashok.raj@intel.com>
Date: Sat, 25 Jun 2005 14:55:00 -0700
Subject: [PATCH] x86_64: CPU hotplug support

  Experimental CPU hotplug patch for x86_64
  -----------------------------------------
This supports logical CPU online and offline.
- Test with maxcpus=1, and then kick other cpu's off to test if init code
  is all cleaned up. CONFIG_SCHED_SMT works as well.
- idle threads are forked on demand from keventd threads for clean startup

TBD:
1. Not tested on a real NUMA machine (tested with numa=fake=2)
2. Handle ACPI pieces for physical hotplug support.

Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Acked-by: Andi Kleen <ak@muc.de>
Acked-by: Zwane Mwaikambo <zwane@arm.linux.org.uk>
Signed-off-by: Shaohua.li<shaohua.li@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mach-default/topology.c |  15 ++-
 arch/x86_64/Kconfig               |   9 ++
 arch/x86_64/kernel/irq.c          |  29 ++++++
 arch/x86_64/kernel/process.c      |  29 +++++-
 arch/x86_64/kernel/smpboot.c      | 205 +++++++++++++++++++++++++++++++++++---
 arch/x86_64/kernel/traps.c        |   8 +-
 include/asm-x86_64/irq.h          |   5 +
 include/asm-x86_64/smp.h          |   2 +
 8 files changed, 277 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/arch/i386/mach-default/topology.c b/arch/i386/mach-default/topology.c
index 5b3e8817dae..23395fff35d 100644
--- a/arch/i386/mach-default/topology.c
+++ b/arch/i386/mach-default/topology.c
@@ -73,12 +73,11 @@ static int __init topology_init(void)
 {
 	int i;
 
-	for (i = 0; i < MAX_NUMNODES; i++) {
-		if (node_online(i))
-			arch_register_node(i);
-	}
-	for (i = 0; i < NR_CPUS; i++)
-		if (cpu_possible(i)) arch_register_cpu(i);
+	for_each_online_node(i)
+		arch_register_node(i);
+
+	for_each_cpu(i)
+		arch_register_cpu(i);
 	return 0;
 }
 
@@ -88,8 +87,8 @@ static int __init topology_init(void)
 {
 	int i;
 
-	for (i = 0; i < NR_CPUS; i++)
-		if (cpu_possible(i)) arch_register_cpu(i);
+	for_each_cpu(i)
+		arch_register_cpu(i);
 	return 0;
 }
 
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index db259757dc8..92f5a526602 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -313,6 +313,15 @@ config NR_CPUS
 	  This is purely to save memory - each supported CPU requires
 	  memory in the static kernel configuration.
 
+config HOTPLUG_CPU
+	bool "Support for hot-pluggable CPUs (EXPERIMENTAL)"
+	depends on SMP && HOTPLUG && EXPERIMENTAL
+	help
+		Say Y here to experiment with turning CPUs off and on.  CPUs
+		can be controlled through /sys/devices/system/cpu/cpu#.
+		Say N if you want to disable CPU hotplug.
+
+
 config HPET_TIMER
 	bool
 	default y
diff --git a/arch/x86_64/kernel/irq.c b/arch/x86_64/kernel/irq.c
index 62b112e4deb..cc3fb85f514 100644
--- a/arch/x86_64/kernel/irq.c
+++ b/arch/x86_64/kernel/irq.c
@@ -14,6 +14,7 @@
 #include <linux/interrupt.h>
 #include <linux/seq_file.h>
 #include <linux/module.h>
+#include <linux/delay.h>
 #include <asm/uaccess.h>
 #include <asm/io_apic.h>
 
@@ -106,3 +107,31 @@ asmlinkage unsigned int do_IRQ(struct pt_regs *regs)
 	return 1;
 }
 
+#ifdef CONFIG_HOTPLUG_CPU
+void fixup_irqs(cpumask_t map)
+{
+	unsigned int irq;
+	static int warned;
+
+	for (irq = 0; irq < NR_IRQS; irq++) {
+		cpumask_t mask;
+		if (irq == 2)
+			continue;
+
+		cpus_and(mask, irq_affinity[irq], map);
+		if (any_online_cpu(mask) == NR_CPUS) {
+			printk("Breaking affinity for irq %i\n", irq);
+			mask = map;
+		}
+		if (irq_desc[irq].handler->set_affinity)
+			irq_desc[irq].handler->set_affinity(irq, mask);
+		else if (irq_desc[irq].action && !(warned++))
+			printk("Cannot set affinity for irq %i\n", irq);
+	}
+
+	/* That doesn't seem sufficient.  Give it 1ms. */
+	local_irq_enable();
+	mdelay(1);
+	local_irq_disable();
+}
+#endif
diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 15fa35a81a7..1d91271796e 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -8,7 +8,8 @@
  * 
  *  X86-64 port
  *	Andi Kleen.
- * 
+ *
+ *	CPU hotplug support - ashok.raj@intel.com
  *  $Id: process.c,v 1.38 2002/01/15 10:08:03 ak Exp $
  */
 
@@ -18,6 +19,7 @@
 
 #include <stdarg.h>
 
+#include <linux/cpu.h>
 #include <linux/errno.h>
 #include <linux/sched.h>
 #include <linux/kernel.h>
@@ -154,6 +156,29 @@ void cpu_idle_wait(void)
 }
 EXPORT_SYMBOL_GPL(cpu_idle_wait);
 
+#ifdef CONFIG_HOTPLUG_CPU
+DECLARE_PER_CPU(int, cpu_state);
+
+#include <asm/nmi.h>
+/* We don't actually take CPU down, just spin without interrupts. */
+static inline void play_dead(void)
+{
+	idle_task_exit();
+	wbinvd();
+	mb();
+	/* Ack it */
+	__get_cpu_var(cpu_state) = CPU_DEAD;
+
+	while (1)
+		safe_halt();
+}
+#else
+static inline void play_dead(void)
+{
+	BUG();
+}
+#endif /* CONFIG_HOTPLUG_CPU */
+
 /*
  * The idle thread. There's no useful work to be
  * done, so just try to conserve power and have a
@@ -174,6 +199,8 @@ void cpu_idle (void)
 			idle = pm_idle;
 			if (!idle)
 				idle = default_idle;
+			if (cpu_is_offline(smp_processor_id()))
+				play_dead();
 			idle();
 		}
 
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index bc98a6722cb..5a3f955b657 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -34,6 +34,7 @@
  *      Andi Kleen              :       Converted to new state machine.
  *					Various cleanups.
  *					Probably mostly hotplug CPU ready now.
+ *	Ashok Raj			: CPU hotplug support
  */
 
 
@@ -98,6 +99,37 @@ EXPORT_SYMBOL(cpu_core_map);
 extern unsigned char trampoline_data[];
 extern unsigned char trampoline_end[];
 
+/* State of each CPU */
+DEFINE_PER_CPU(int, cpu_state) = { 0 };
+
+/*
+ * Store all idle threads, this can be reused instead of creating
+ * a new thread. Also avoids complicated thread destroy functionality
+ * for idle threads.
+ */
+struct task_struct *idle_thread_array[NR_CPUS] __cpuinitdata ;
+
+#define get_idle_for_cpu(x)     (idle_thread_array[(x)])
+#define set_idle_for_cpu(x,p)   (idle_thread_array[(x)] = (p))
+
+/*
+ * cpu_possible_map should be static, it cannot change as cpu's
+ * are onlined, or offlined. The reason is per-cpu data-structures
+ * are allocated by some modules at init time, and dont expect to
+ * do this dynamically on cpu arrival/departure.
+ * cpu_present_map on the other hand can change dynamically.
+ * In case when cpu_hotplug is not compiled, then we resort to current
+ * behaviour, which is cpu_possible == cpu_present.
+ * If cpu-hotplug is supported, then we need to preallocate for all
+ * those NR_CPUS, hence cpu_possible_map represents entire NR_CPUS range.
+ * - Ashok Raj
+ */
+#ifdef CONFIG_HOTPLUG_CPU
+#define fixup_cpu_possible_map(x)	cpu_set((x), cpu_possible_map)
+#else
+#define fixup_cpu_possible_map(x)
+#endif
+
 /*
  * Currently trivial. Write the real->protected mode
  * bootstrap into the page concerned. The caller
@@ -623,33 +655,77 @@ static int __cpuinit wakeup_secondary_via_INIT(int phys_apicid, unsigned int sta
 	return (send_status | accept_status);
 }
 
+struct create_idle {
+	struct task_struct *idle;
+	struct completion done;
+	int cpu;
+};
+
+void do_fork_idle(void *_c_idle)
+{
+	struct create_idle *c_idle = _c_idle;
+
+	c_idle->idle = fork_idle(c_idle->cpu);
+	complete(&c_idle->done);
+}
+
 /*
  * Boot one CPU.
  */
 static int __cpuinit do_boot_cpu(int cpu, int apicid)
 {
-	struct task_struct *idle;
 	unsigned long boot_error;
 	int timeout;
 	unsigned long start_rip;
+	struct create_idle c_idle = {
+		.cpu = cpu,
+		.done = COMPLETION_INITIALIZER(c_idle.done),
+	};
+	DECLARE_WORK(work, do_fork_idle, &c_idle);
+
+	c_idle.idle = get_idle_for_cpu(cpu);
+
+	if (c_idle.idle) {
+		c_idle.idle->thread.rsp = (unsigned long) (((struct pt_regs *)
+			(THREAD_SIZE + (unsigned long) c_idle.idle->thread_info)) - 1);
+		init_idle(c_idle.idle, cpu);
+		goto do_rest;
+	}
+
 	/*
-	 * We can't use kernel_thread since we must avoid to
-	 * reschedule the child.
+	 * During cold boot process, keventd thread is not spun up yet.
+	 * When we do cpu hot-add, we create idle threads on the fly, we should
+	 * not acquire any attributes from the calling context. Hence the clean
+	 * way to create kernel_threads() is to do that from keventd().
+	 * We do the current_is_keventd() due to the fact that ACPI notifier
+	 * was also queuing to keventd() and when the caller is already running
+	 * in context of keventd(), we would end up with locking up the keventd
+	 * thread.
 	 */
-	idle = fork_idle(cpu);
-	if (IS_ERR(idle)) {
+	if (!keventd_up() || current_is_keventd())
+		work.func(work.data);
+	else {
+		schedule_work(&work);
+		wait_for_completion(&c_idle.done);
+	}
+
+	if (IS_ERR(c_idle.idle)) {
 		printk("failed fork for CPU %d\n", cpu);
-		return PTR_ERR(idle);
+		return PTR_ERR(c_idle.idle);
 	}
 
-	cpu_pda[cpu].pcurrent = idle;
+	set_idle_for_cpu(cpu, c_idle.idle);
+
+do_rest:
+
+	cpu_pda[cpu].pcurrent = c_idle.idle;
 
 	start_rip = setup_trampoline();
 
-	init_rsp = idle->thread.rsp;
+	init_rsp = c_idle.idle->thread.rsp;
 	per_cpu(init_tss,cpu).rsp0 = init_rsp;
 	initial_code = start_secondary;
-	clear_ti_thread_flag(idle->thread_info, TIF_FORK);
+	clear_ti_thread_flag(c_idle.idle->thread_info, TIF_FORK);
 
 	printk(KERN_INFO "Booting processor %d/%d rip %lx rsp %lx\n", cpu, apicid,
 	       start_rip, init_rsp);
@@ -925,10 +1001,9 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 		int apicid = cpu_present_to_apicid(i);
 		if (physid_isset(apicid, phys_cpu_present_map)) {
 			cpu_set(i, cpu_present_map);
-			/* possible map would be different if we supported real
-			   CPU hotplug. */
 			cpu_set(i, cpu_possible_map);
 		}
+		fixup_cpu_possible_map(i);
 	}
 
 	if (smp_sanity_check(max_cpus) < 0) {
@@ -977,9 +1052,6 @@ void __init smp_prepare_boot_cpu(void)
 
 /*
  * Entry point to boot a CPU.
- *
- * This is all __cpuinit, not __devinit for now because we don't support
- * CPU hotplug (yet).
  */
 int __cpuinit __cpu_up(unsigned int cpu)
 {
@@ -996,6 +1068,14 @@ int __cpuinit __cpu_up(unsigned int cpu)
 		return -EINVAL;
 	}
 
+	/*
+	 * Already booted CPU?
+	 */
+ 	if (cpu_isset(cpu, cpu_callin_map)) {
+		Dprintk("do_boot_cpu %d Already started\n", cpu);
+ 		return -ENOSYS;
+	}
+
 	/* Boot it! */
 	err = do_boot_cpu(cpu, apicid);
 	if (err < 0) {
@@ -1008,7 +1088,9 @@ int __cpuinit __cpu_up(unsigned int cpu)
 
 	while (!cpu_isset(cpu, cpu_online_map))
 		cpu_relax();
-	return 0;
+	err = 0;
+
+	return err;
 }
 
 /*
@@ -1016,7 +1098,9 @@ int __cpuinit __cpu_up(unsigned int cpu)
  */
 void __init smp_cpus_done(unsigned int max_cpus)
 {
+#ifndef CONFIG_HOTPLUG_CPU
 	zap_low_mappings();
+#endif
 	smp_cleanup_boot();
 
 #ifdef CONFIG_X86_IO_APIC
@@ -1028,3 +1112,94 @@ void __init smp_cpus_done(unsigned int max_cpus)
 
 	check_nmi_watchdog();
 }
+
+#ifdef CONFIG_HOTPLUG_CPU
+
+static void
+remove_siblinginfo(int cpu)
+{
+	int sibling;
+
+	for_each_cpu_mask(sibling, cpu_sibling_map[cpu])
+		cpu_clear(cpu, cpu_sibling_map[sibling]);
+	for_each_cpu_mask(sibling, cpu_core_map[cpu])
+		cpu_clear(cpu, cpu_core_map[sibling]);
+	cpus_clear(cpu_sibling_map[cpu]);
+	cpus_clear(cpu_core_map[cpu]);
+	phys_proc_id[cpu] = BAD_APICID;
+	cpu_core_id[cpu] = BAD_APICID;
+}
+
+void remove_cpu_from_maps(void)
+{
+	int cpu = smp_processor_id();
+
+	cpu_clear(cpu, cpu_callout_map);
+	cpu_clear(cpu, cpu_callin_map);
+	clear_bit(cpu, &cpu_initialized); /* was set by cpu_init() */
+}
+
+int __cpu_disable(void)
+{
+	int cpu = smp_processor_id();
+
+	/*
+	 * Perhaps use cpufreq to drop frequency, but that could go
+	 * into generic code.
+ 	 *
+	 * We won't take down the boot processor on i386 due to some
+	 * interrupts only being able to be serviced by the BSP.
+	 * Especially so if we're not using an IOAPIC	-zwane
+	 */
+	if (cpu == 0)
+		return -EBUSY;
+
+	disable_APIC_timer();
+
+	/*
+	 * HACK:
+	 * Allow any queued timer interrupts to get serviced
+	 * This is only a temporary solution until we cleanup
+	 * fixup_irqs as we do for IA64.
+	 */
+	local_irq_enable();
+	mdelay(1);
+
+	local_irq_disable();
+	remove_siblinginfo(cpu);
+
+	/* It's now safe to remove this processor from the online map */
+	cpu_clear(cpu, cpu_online_map);
+	remove_cpu_from_maps();
+	fixup_irqs(cpu_online_map);
+	return 0;
+}
+
+void __cpu_die(unsigned int cpu)
+{
+	/* We don't do anything here: idle task is faking death itself. */
+	unsigned int i;
+
+	for (i = 0; i < 10; i++) {
+		/* They ack this in play_dead by setting CPU_DEAD */
+		if (per_cpu(cpu_state, cpu) == CPU_DEAD)
+			return;
+		current->state = TASK_UNINTERRUPTIBLE;
+		schedule_timeout(HZ/10);
+	}
+ 	printk(KERN_ERR "CPU %u didn't die...\n", cpu);
+}
+
+#else /* ... !CONFIG_HOTPLUG_CPU */
+
+int __cpu_disable(void)
+{
+	return -ENOSYS;
+}
+
+void __cpu_die(unsigned int cpu)
+{
+	/* We said "no" in __cpu_disable */
+	BUG();
+}
+#endif /* CONFIG_HOTPLUG_CPU */
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 121646fc43f..10273663000 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -586,11 +586,17 @@ static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 asmlinkage void default_do_nmi(struct pt_regs *regs)
 {
 	unsigned char reason = 0;
+	int cpu;
+
+	cpu = smp_processor_id();
 
 	/* Only the BSP gets external NMIs from the system.  */
-	if (!smp_processor_id())
+	if (!cpu)
 		reason = get_nmi_reason();
 
+	if (!cpu_online(cpu))
+		return;
+
 	if (!(reason & 0xc0)) {
 		if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 0, SIGINT)
 								== NOTIFY_STOP)
diff --git a/include/asm-x86_64/irq.h b/include/asm-x86_64/irq.h
index 3af50b3c3b0..eb3b7aa9eb9 100644
--- a/include/asm-x86_64/irq.h
+++ b/include/asm-x86_64/irq.h
@@ -52,4 +52,9 @@ struct irqaction;
 struct pt_regs;
 int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
 
+#ifdef CONFIG_HOTPLUG_CPU
+#include <linux/cpumask.h>
+extern void fixup_irqs(cpumask_t map);
+#endif
+
 #endif /* _ASM_IRQ_H */
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index a7425aa5a3b..9c6242fb99d 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -77,6 +77,8 @@ extern __inline int hard_smp_processor_id(void)
 }
 
 extern int safe_smp_processor_id(void);
+extern int __cpu_disable(void);
+extern void __cpu_die(unsigned int cpu);
 
 #endif /* !ASSEMBLY */
 
-- 
cgit v1.2.3-70-g09d2


From 884d9e40b4089014f40c49e86ac6505842db2b53 Mon Sep 17 00:00:00 2001
From: Ashok Raj <ashok.raj@intel.com>
Date: Sat, 25 Jun 2005 14:55:02 -0700
Subject: [PATCH] x86_64: Dont use broadcast shortcut to make it cpu hotplug
 safe.

Broadcast IPI's provide un-expected behaviour for cpu hotplug.  CPU's in
offline state also end up receiving the IPI.  Once the cpus become online they
receive these stale IPI's which are bad and introduce unexpected behaviour.

This is easily avoided by not sending a broadcast and addressing just the
CPU's in online map.  Doing prelim cycle counts it appears there is no big
overhead and numbers seem around 0x3000-0x3900 on an average on x86 and x86_64
systems with CPUS running 3G, both for broadcast and mask version of the
API's.

The shortcuts are useful only for flat mode (where the perf shows no
degradation), and in cluster mode, its unicast anyway.  Its simpler to just
not use broadcast anymore.

Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Acked-by: Andi Kleen <ak@muc.de>
Acked-by: Zwane Mwaikambo <zwane@arm.linux.org.uk>
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/genapic_flat.c | 42 ++++++++++++++++++++++++---------------
 arch/x86_64/kernel/smp.c          | 10 ++++++++++
 arch/x86_64/kernel/smpboot.c      | 19 +++++++++++++++++-
 include/asm-x86_64/smp.h          |  2 ++
 4 files changed, 56 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/genapic_flat.c b/arch/x86_64/kernel/genapic_flat.c
index b4cbbad0422..00f3fa6df71 100644
--- a/arch/x86_64/kernel/genapic_flat.c
+++ b/arch/x86_64/kernel/genapic_flat.c
@@ -7,6 +7,8 @@
  * Hacked for x86-64 by James Cleverdon from i386 architecture code by
  * Martin Bligh, Andi Kleen, James Bottomley, John Stultz, and
  * James Cleverdon.
+ * Ashok Raj <ashok.raj@intel.com>
+ * 	Removed IPI broadcast shortcut to support CPU hotplug
  */
 #include <linux/config.h>
 #include <linux/threads.h>
@@ -45,22 +47,6 @@ static void flat_init_apic_ldr(void)
 	apic_write_around(APIC_LDR, val);
 }
 
-static void flat_send_IPI_allbutself(int vector)
-{
-	/*
-	 * if there are no other CPUs in the system then
-	 * we get an APIC send error if we try to broadcast.
-	 * thus we have to avoid sending IPIs in this case.
-	 */
-	if (num_online_cpus() > 1)
-		__send_IPI_shortcut(APIC_DEST_ALLBUT, vector, APIC_DEST_LOGICAL);
-}
-
-static void flat_send_IPI_all(int vector)
-{
-	__send_IPI_shortcut(APIC_DEST_ALLINC, vector, APIC_DEST_LOGICAL);
-}
-
 static void flat_send_IPI_mask(cpumask_t cpumask, int vector)
 {
 	unsigned long mask = cpus_addr(cpumask)[0];
@@ -93,6 +79,30 @@ static void flat_send_IPI_mask(cpumask_t cpumask, int vector)
 	local_irq_restore(flags);
 }
 
+static void flat_send_IPI_allbutself(int vector)
+{
+	cpumask_t mask;
+	/*
+	 * if there are no other CPUs in the system then
+	 * we get an APIC send error if we try to broadcast.
+	 * thus we have to avoid sending IPIs in this case.
+	 */
+	int this_cpu = get_cpu();
+
+	mask = cpu_online_map;
+	cpu_clear(this_cpu, mask);
+
+	if (cpus_weight(mask) >= 1)
+		flat_send_IPI_mask(mask, vector);
+
+	put_cpu();
+}
+
+static void flat_send_IPI_all(int vector)
+{
+	flat_send_IPI_mask(cpu_online_map, vector);
+}
+
 static int flat_apic_id_registered(void)
 {
 	return physid_isset(GET_APIC_ID(apic_read(APIC_ID)), phys_cpu_present_map);
diff --git a/arch/x86_64/kernel/smp.c b/arch/x86_64/kernel/smp.c
index 1e379ed17b1..ccae392886a 100644
--- a/arch/x86_64/kernel/smp.c
+++ b/arch/x86_64/kernel/smp.c
@@ -283,6 +283,16 @@ struct call_data_struct {
 
 static struct call_data_struct * call_data;
 
+void lock_ipi_call_lock(void)
+{
+	spin_lock_irq(&call_lock);
+}
+
+void unlock_ipi_call_lock(void)
+{
+	spin_unlock_irq(&call_lock);
+}
+
 /*
  * this function sends a 'generic call function' IPI to all other CPUs
  * in the system.
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index 571a55462fa..b969ee12872 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -507,10 +507,23 @@ void __cpuinit start_secondary(void)
 	 */
 	set_cpu_sibling_map(smp_processor_id());
 
+	/*
+	 * We need to hold call_lock, so there is no inconsistency
+	 * between the time smp_call_function() determines number of
+	 * IPI receipients, and the time when the determination is made
+	 * for which cpus receive the IPI in genapic_flat.c. Holding this
+	 * lock helps us to not include this cpu in a currently in progress
+	 * smp_call_function().
+	 */
+	lock_ipi_call_lock();
+
 	/*
 	 * Allow the master to continue.
 	 */
 	cpu_set(smp_processor_id(), cpu_online_map);
+	per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE;
+	unlock_ipi_call_lock();
+
 	mb();
 
 	/* Wait for TSC sync to not schedule things before.
@@ -1038,6 +1051,7 @@ void __init smp_prepare_boot_cpu(void)
 	cpu_set(me, cpu_callout_map);
 	cpu_set(0, cpu_sibling_map[0]);
 	cpu_set(0, cpu_core_map[0]);
+	per_cpu(cpu_state, me) = CPU_ONLINE;
 }
 
 /*
@@ -1066,6 +1080,7 @@ int __cpuinit __cpu_up(unsigned int cpu)
  		return -ENOSYS;
 	}
 
+	per_cpu(cpu_state, cpu) = CPU_UP_PREPARE;
 	/* Boot it! */
 	err = do_boot_cpu(cpu, apicid);
 	if (err < 0) {
@@ -1170,8 +1185,10 @@ void __cpu_die(unsigned int cpu)
 
 	for (i = 0; i < 10; i++) {
 		/* They ack this in play_dead by setting CPU_DEAD */
-		if (per_cpu(cpu_state, cpu) == CPU_DEAD)
+		if (per_cpu(cpu_state, cpu) == CPU_DEAD) {
+			printk ("CPU %d is now offline\n", cpu);
 			return;
+		}
 		current->state = TASK_UNINTERRUPTIBLE;
 		schedule_timeout(HZ/10);
 	}
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index 9c6242fb99d..aeb1b73e21e 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -43,6 +43,8 @@ extern cpumask_t cpu_callout_map;
 extern void smp_alloc_memory(void);
 extern volatile unsigned long smp_invalidate_needed;
 extern int pic_mode;
+extern void lock_ipi_call_lock(void);
+extern void unlock_ipi_call_lock(void);
 extern int smp_num_siblings;
 extern void smp_flush_tlb(void);
 extern void smp_message_irq(int cpl, void *dev_id, struct pt_regs *regs);
-- 
cgit v1.2.3-70-g09d2


From 5a72e04df5470df0ec646029d31e5528167ab1a7 Mon Sep 17 00:00:00 2001
From: Li Shaohua <shaohua.li@intel.com>
Date: Sat, 25 Jun 2005 14:55:06 -0700
Subject: [PATCH] suspend/resume SMP support

Using CPU hotplug to support suspend/resume SMP.  Both S3 and S4 use
disable/enable_nonboot_cpus API.  The S4 part is based on Pavel's original S4
SMP patch.

Signed-off-by: Li Shaohua<shaohua.li@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/cpu/mcheck/k7.c      |  2 +-
 arch/i386/kernel/cpu/mcheck/mce.c     |  2 +-
 arch/i386/kernel/cpu/mcheck/p4.c      |  4 +-
 arch/i386/kernel/cpu/mcheck/p6.c      |  2 +-
 arch/i386/kernel/cpu/mcheck/winchip.c |  2 +-
 drivers/acpi/Kconfig                  |  2 +-
 include/linux/suspend.h               |  2 +-
 kernel/power/Kconfig                  |  6 ++-
 kernel/power/Makefile                 |  6 +--
 kernel/power/disk.c                   | 35 +++++++-------
 kernel/power/main.c                   | 16 ++++---
 kernel/power/smp.c                    | 89 +++++++++++++----------------------
 kernel/power/swsusp.c                 |  2 +
 13 files changed, 80 insertions(+), 90 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/cpu/mcheck/k7.c b/arch/i386/kernel/cpu/mcheck/k7.c
index 8df52e86c4d..c4abe765739 100644
--- a/arch/i386/kernel/cpu/mcheck/k7.c
+++ b/arch/i386/kernel/cpu/mcheck/k7.c
@@ -69,7 +69,7 @@ static fastcall void k7_machine_check(struct pt_regs * regs, long error_code)
 
 
 /* AMD K7 machine check is Intel like */
-void __init amd_mcheck_init(struct cpuinfo_x86 *c)
+void __devinit amd_mcheck_init(struct cpuinfo_x86 *c)
 {
 	u32 l, h;
 	int i;
diff --git a/arch/i386/kernel/cpu/mcheck/mce.c b/arch/i386/kernel/cpu/mcheck/mce.c
index 7218a7341fb..2cf25d2ba0f 100644
--- a/arch/i386/kernel/cpu/mcheck/mce.c
+++ b/arch/i386/kernel/cpu/mcheck/mce.c
@@ -16,7 +16,7 @@
 
 #include "mce.h"
 
-int mce_disabled __initdata = 0;
+int mce_disabled __devinitdata = 0;
 int nr_mce_banks;
 
 EXPORT_SYMBOL_GPL(nr_mce_banks);	/* non-fatal.o */
diff --git a/arch/i386/kernel/cpu/mcheck/p4.c b/arch/i386/kernel/cpu/mcheck/p4.c
index 8b16ceb929b..0abccb6fdf9 100644
--- a/arch/i386/kernel/cpu/mcheck/p4.c
+++ b/arch/i386/kernel/cpu/mcheck/p4.c
@@ -78,7 +78,7 @@ fastcall void smp_thermal_interrupt(struct pt_regs *regs)
 }
 
 /* P4/Xeon Thermal regulation detect and init */
-static void __init intel_init_thermal(struct cpuinfo_x86 *c)
+static void __devinit intel_init_thermal(struct cpuinfo_x86 *c)
 {
 	u32 l, h;
 	unsigned int cpu = smp_processor_id();
@@ -232,7 +232,7 @@ static fastcall void intel_machine_check(struct pt_regs * regs, long error_code)
 }
 
 
-void __init intel_p4_mcheck_init(struct cpuinfo_x86 *c)
+void __devinit intel_p4_mcheck_init(struct cpuinfo_x86 *c)
 {
 	u32 l, h;
 	int i;
diff --git a/arch/i386/kernel/cpu/mcheck/p6.c b/arch/i386/kernel/cpu/mcheck/p6.c
index 46640f8c249..f01b73f947e 100644
--- a/arch/i386/kernel/cpu/mcheck/p6.c
+++ b/arch/i386/kernel/cpu/mcheck/p6.c
@@ -80,7 +80,7 @@ static fastcall void intel_machine_check(struct pt_regs * regs, long error_code)
 }
 
 /* Set up machine check reporting for processors with Intel style MCE */
-void __init intel_p6_mcheck_init(struct cpuinfo_x86 *c)
+void __devinit intel_p6_mcheck_init(struct cpuinfo_x86 *c)
 {
 	u32 l, h;
 	int i;
diff --git a/arch/i386/kernel/cpu/mcheck/winchip.c b/arch/i386/kernel/cpu/mcheck/winchip.c
index 753fa7acb98..7bae68fa168 100644
--- a/arch/i386/kernel/cpu/mcheck/winchip.c
+++ b/arch/i386/kernel/cpu/mcheck/winchip.c
@@ -23,7 +23,7 @@ static fastcall void winchip_machine_check(struct pt_regs * regs, long error_cod
 }
 
 /* Set up machine check reporting on the Winchip C6 series */
-void __init winchip_mcheck_init(struct cpuinfo_x86 *c)
+void __devinit winchip_mcheck_init(struct cpuinfo_x86 *c)
 {
 	u32 lo, hi;
 	machine_check_vector = winchip_machine_check;
diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig
index 670fdb5142d..86c52520ed3 100644
--- a/drivers/acpi/Kconfig
+++ b/drivers/acpi/Kconfig
@@ -55,7 +55,7 @@ if ACPI_INTERPRETER
 
 config ACPI_SLEEP
 	bool "Sleep States (EXPERIMENTAL)"
-	depends on X86
+	depends on X86 && (!SMP || SUSPEND_SMP)
 	depends on EXPERIMENTAL && PM
 	default y
 	---help---
diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index 2bf0d5fabcd..f2e96fdfaae 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -58,7 +58,7 @@ static inline int software_suspend(void)
 }
 #endif
 
-#ifdef CONFIG_SMP
+#ifdef CONFIG_SUSPEND_SMP
 extern void disable_nonboot_cpus(void);
 extern void enable_nonboot_cpus(void);
 #else
diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig
index 696387ffe49..fdb37763650 100644
--- a/kernel/power/Kconfig
+++ b/kernel/power/Kconfig
@@ -28,7 +28,7 @@ config PM_DEBUG
 
 config SOFTWARE_SUSPEND
 	bool "Software Suspend (EXPERIMENTAL)"
-	depends on EXPERIMENTAL && PM && SWAP
+	depends on EXPERIMENTAL && PM && SWAP && (SUSPEND_SMP || !SMP)
 	---help---
 	  Enable the possibility of suspending the machine.
 	  It doesn't need APM.
@@ -72,3 +72,7 @@ config PM_STD_PARTITION
 	  suspended image to. It will simply pick the first available swap 
 	  device.
 
+config SUSPEND_SMP
+	bool
+	depends on HOTPLUG_CPU && X86 && PM
+	default y
diff --git a/kernel/power/Makefile b/kernel/power/Makefile
index fbdc634135a..2f438d0eaa1 100644
--- a/kernel/power/Makefile
+++ b/kernel/power/Makefile
@@ -3,9 +3,9 @@ ifeq ($(CONFIG_PM_DEBUG),y)
 EXTRA_CFLAGS	+=	-DDEBUG
 endif
 
-swsusp-smp-$(CONFIG_SMP)	+= smp.o
-
 obj-y				:= main.o process.o console.o pm.o
-obj-$(CONFIG_SOFTWARE_SUSPEND)	+= swsusp.o $(swsusp-smp-y) disk.o
+obj-$(CONFIG_SOFTWARE_SUSPEND)	+= swsusp.o disk.o
+
+obj-$(CONFIG_SUSPEND_SMP)	+= smp.o
 
 obj-$(CONFIG_MAGIC_SYSRQ)	+= poweroff.o
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index 02b6764034d..fb8de63c291 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -117,8 +117,8 @@ static void finish(void)
 {
 	device_resume();
 	platform_finish();
-	enable_nonboot_cpus();
 	thaw_processes();
+	enable_nonboot_cpus();
 	pm_restore_console();
 }
 
@@ -131,28 +131,35 @@ static int prepare_processes(void)
 
 	sys_sync();
 
+	disable_nonboot_cpus();
+
 	if (freeze_processes()) {
 		error = -EBUSY;
-		return error;
+		goto thaw;
 	}
 
 	if (pm_disk_mode == PM_DISK_PLATFORM) {
 		if (pm_ops && pm_ops->prepare) {
 			if ((error = pm_ops->prepare(PM_SUSPEND_DISK)))
-				return error;
+				goto thaw;
 		}
 	}
 
 	/* Free memory before shutting down devices. */
 	free_some_memory();
-
 	return 0;
+thaw:
+	thaw_processes();
+	enable_nonboot_cpus();
+	pm_restore_console();
+	return error;
 }
 
 static void unprepare_processes(void)
 {
-	enable_nonboot_cpus();
+	platform_finish();
 	thaw_processes();
+	enable_nonboot_cpus();
 	pm_restore_console();
 }
 
@@ -160,15 +167,9 @@ static int prepare_devices(void)
 {
 	int error;
 
-	disable_nonboot_cpus();
-	if ((error = device_suspend(PMSG_FREEZE))) {
+	if ((error = device_suspend(PMSG_FREEZE)))
 		printk("Some devices failed to suspend\n");
-		platform_finish();
-		enable_nonboot_cpus();
-		return error;
-	}
-
-	return 0;
+	return error;
 }
 
 /**
@@ -185,9 +186,9 @@ int pm_suspend_disk(void)
 	int error;
 
 	error = prepare_processes();
-	if (!error) {
-		error = prepare_devices();
-	}
+	if (error)
+		return error;
+	error = prepare_devices();
 
 	if (error) {
 		unprepare_processes();
@@ -250,7 +251,7 @@ static int software_resume(void)
 
 	if ((error = prepare_processes())) {
 		swsusp_close();
-		goto Cleanup;
+		goto Done;
 	}
 
 	pr_debug("PM: Reading swsusp image.\n");
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 4cdebc972ff..c94cb9e9509 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -55,6 +55,13 @@ static int suspend_prepare(suspend_state_t state)
 
 	pm_prepare_console();
 
+	disable_nonboot_cpus();
+
+	if (num_online_cpus() != 1) {
+		error = -EPERM;
+		goto Enable_cpu;
+	}
+
 	if (freeze_processes()) {
 		error = -EAGAIN;
 		goto Thaw;
@@ -75,6 +82,8 @@ static int suspend_prepare(suspend_state_t state)
 		pm_ops->finish(state);
  Thaw:
 	thaw_processes();
+ Enable_cpu:
+	enable_nonboot_cpus();
 	pm_restore_console();
 	return error;
 }
@@ -113,6 +122,7 @@ static void suspend_finish(suspend_state_t state)
 	if (pm_ops && pm_ops->finish)
 		pm_ops->finish(state);
 	thaw_processes();
+	enable_nonboot_cpus();
 	pm_restore_console();
 }
 
@@ -150,12 +160,6 @@ static int enter_state(suspend_state_t state)
 		goto Unlock;
 	}
 
-	/* Suspend is hard to get right on SMP. */
-	if (num_online_cpus() != 1) {
-		error = -EPERM;
-		goto Unlock;
-	}
-
 	pr_debug("PM: Preparing system for %s sleep\n", pm_states[state]);
 	if ((error = suspend_prepare(state)))
 		goto Unlock;
diff --git a/kernel/power/smp.c b/kernel/power/smp.c
index 457c2302ed4..bbe23079c62 100644
--- a/kernel/power/smp.c
+++ b/kernel/power/smp.c
@@ -13,73 +13,52 @@
 #include <linux/interrupt.h>
 #include <linux/suspend.h>
 #include <linux/module.h>
+#include <linux/cpu.h>
 #include <asm/atomic.h>
 #include <asm/tlbflush.h>
 
-static atomic_t cpu_counter, freeze;
-
-
-static void smp_pause(void * data)
-{
-	struct saved_context ctxt;
-	__save_processor_state(&ctxt);
-	printk("Sleeping in:\n");
-	dump_stack();
-	atomic_inc(&cpu_counter);
-	while (atomic_read(&freeze)) {
-		/* FIXME: restore takes place at random piece inside this.
-		   This should probably be written in assembly, and
-		   preserve general-purpose registers, too
-
-		   What about stack? We may need to move to new stack here.
-
-		   This should better be ran with interrupts disabled.
-		 */
-		cpu_relax();
-		barrier();
-	}
-	atomic_dec(&cpu_counter);
-	__restore_processor_state(&ctxt);
-}
-
-static cpumask_t oldmask;
+/* This is protected by pm_sem semaphore */
+static cpumask_t frozen_cpus;
 
 void disable_nonboot_cpus(void)
 {
-	oldmask = current->cpus_allowed;
-	set_cpus_allowed(current, cpumask_of_cpu(0));
-	printk("Freezing CPUs (at %d)", raw_smp_processor_id());
-	current->state = TASK_INTERRUPTIBLE;
-	schedule_timeout(HZ);
-	printk("...");
-	BUG_ON(raw_smp_processor_id() != 0);
-
-	/* FIXME: for this to work, all the CPUs must be running
-	 * "idle" thread (or we deadlock). Is that guaranteed? */
+	int cpu, error;
 
-	atomic_set(&cpu_counter, 0);
-	atomic_set(&freeze, 1);
-	smp_call_function(smp_pause, NULL, 0, 0);
-	while (atomic_read(&cpu_counter) < (num_online_cpus() - 1)) {
-		cpu_relax();
-		barrier();
+	error = 0;
+	cpus_clear(frozen_cpus);
+	printk("Freezing cpus ...\n");
+	for_each_online_cpu(cpu) {
+		if (cpu == 0)
+			continue;
+		error = cpu_down(cpu);
+		if (!error) {
+			cpu_set(cpu, frozen_cpus);
+			printk("CPU%d is down\n", cpu);
+			continue;
+		}
+		printk("Error taking cpu %d down: %d\n", cpu, error);
 	}
-	printk("ok\n");
+	BUG_ON(smp_processor_id() != 0);
+	if (error)
+		panic("cpus not sleeping");
 }
 
 void enable_nonboot_cpus(void)
 {
-	printk("Restarting CPUs");
-	atomic_set(&freeze, 0);
-	while (atomic_read(&cpu_counter)) {
-		cpu_relax();
-		barrier();
-	}
-	printk("...");
-	set_cpus_allowed(current, oldmask);
-	schedule();
-	printk("ok\n");
+	int cpu, error;
 
+	printk("Thawing cpus ...\n");
+	for_each_cpu_mask(cpu, frozen_cpus) {
+		error = smp_prepare_cpu(cpu);
+		if (!error)
+			error = cpu_up(cpu);
+		if (!error) {
+			printk("CPU%d is up\n", cpu);
+			continue;
+		}
+		printk("Error taking cpu %d up: %d\n", cpu, error);
+		panic("Not enough cpus");
+	}
+	cpus_clear(frozen_cpus);
 }
 
-
diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c
index 53f9f8720ee..339b5c3735b 100644
--- a/kernel/power/swsusp.c
+++ b/kernel/power/swsusp.c
@@ -1193,8 +1193,10 @@ static const char * sanity_check(void)
 		return "version";
 	if (strcmp(swsusp_info.uts.machine,system_utsname.machine))
 		return "machine";
+#if 0
 	if(swsusp_info.cpus != num_online_cpus())
 		return "number of cpus";
+#endif
 	return NULL;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 620b03276488c3cf103caf1e326bd21f00d3df84 Mon Sep 17 00:00:00 2001
From: Pavel Machek <pavel@ucw.cz>
Date: Sat, 25 Jun 2005 14:55:11 -0700
Subject: [PATCH] properly stop devices before poweroff

Without this patch, Linux provokes emergency disk shutdowns and
similar nastiness. It was in SuSE kernels for some time, IIRC.

Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/pm.h | 33 +++++++++++++++++++++------------
 kernel/sys.c       |  3 +++
 2 files changed, 24 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/linux/pm.h b/include/linux/pm.h
index ed2b76e7519..14479325e3f 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -103,7 +103,8 @@ extern int pm_active;
 /*
  * Register a device with power management
  */
-struct pm_dev __deprecated *pm_register(pm_dev_t type, unsigned long id, pm_callback callback);
+struct pm_dev __deprecated *
+pm_register(pm_dev_t type, unsigned long id, pm_callback callback);
 
 /*
  * Unregister a device with power management
@@ -190,17 +191,18 @@ typedef u32 __bitwise pm_message_t;
 /*
  * There are 4 important states driver can be in:
  * ON     -- driver is working
- * FREEZE -- stop operations and apply whatever policy is applicable to a suspended driver
- *           of that class, freeze queues for block like IDE does, drop packets for
- *           ethernet, etc... stop DMA engine too etc... so a consistent image can be
- *           saved; but do not power any hardware down.
- * SUSPEND - like FREEZE, but hardware is doing as much powersaving as possible. Roughly
- *           pci D3.
+ * FREEZE -- stop operations and apply whatever policy is applicable to a
+ *           suspended driver of that class, freeze queues for block like IDE
+ *           does, drop packets for ethernet, etc... stop DMA engine too etc...
+ *           so a consistent image can be saved; but do not power any hardware
+ *           down.
+ * SUSPEND - like FREEZE, but hardware is doing as much powersaving as
+ *           possible. Roughly pci D3.
  *
- * Unfortunately, current drivers only recognize numeric values 0 (ON) and 3 (SUSPEND).
- * We'll need to fix the drivers. So yes, putting 3 to all diferent defines is intentional,
- * and will go away as soon as drivers are fixed. Also note that typedef is neccessary,
- * we'll probably want to switch to
+ * Unfortunately, current drivers only recognize numeric values 0 (ON) and 3
+ * (SUSPEND).  We'll need to fix the drivers. So yes, putting 3 to all different
+ * defines is intentional, and will go away as soon as drivers are fixed.  Also
+ * note that typedef is neccessary, we'll probably want to switch to
  *   typedef struct pm_message_t { int event; int flags; } pm_message_t
  * or something similar soon.
  */
@@ -222,11 +224,18 @@ struct dev_pm_info {
 
 extern void device_pm_set_parent(struct device * dev, struct device * parent);
 
-extern int device_suspend(pm_message_t state);
 extern int device_power_down(pm_message_t state);
 extern void device_power_up(void);
 extern void device_resume(void);
 
+#ifdef CONFIG_PM
+extern int device_suspend(pm_message_t state);
+#else
+static inline int device_suspend(pm_message_t state)
+{
+	return 0;
+}
+#endif
 
 #endif /* __KERNEL__ */
 
diff --git a/kernel/sys.c b/kernel/sys.c
index da24bc1292d..dac10161ca2 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -405,6 +405,7 @@ asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user
 	case LINUX_REBOOT_CMD_HALT:
 		notifier_call_chain(&reboot_notifier_list, SYS_HALT, NULL);
 		system_state = SYSTEM_HALT;
+		device_suspend(PMSG_SUSPEND);
 		device_shutdown();
 		printk(KERN_EMERG "System halted.\n");
 		machine_halt();
@@ -415,6 +416,7 @@ asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user
 	case LINUX_REBOOT_CMD_POWER_OFF:
 		notifier_call_chain(&reboot_notifier_list, SYS_POWER_OFF, NULL);
 		system_state = SYSTEM_POWER_OFF;
+		device_suspend(PMSG_SUSPEND);
 		device_shutdown();
 		printk(KERN_EMERG "Power down.\n");
 		machine_power_off();
@@ -431,6 +433,7 @@ asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user
 
 		notifier_call_chain(&reboot_notifier_list, SYS_RESTART, buffer);
 		system_state = SYSTEM_RESTART;
+		device_suspend(PMSG_FREEZE);
 		device_shutdown();
 		printk(KERN_EMERG "Restarting system with command '%s'.\n", buffer);
 		machine_restart(buffer);
-- 
cgit v1.2.3-70-g09d2


From 8d783b3e02002bce8cf9d4e4a82922ee7e59b1e5 Mon Sep 17 00:00:00 2001
From: Pavel Machek <pavel@ucw.cz>
Date: Sat, 25 Jun 2005 14:55:14 -0700
Subject: [PATCH] swsusp: clean assembly parts

This patch fixes register saving so that each register is only saved once,
and adds missing saving of %cr8 on x86-64.  Some reordering so that
save/restore is more logical/safer (segment registers should be restored
after gdt).

Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/power/cpu.c        | 17 +++++++----------
 arch/x86_64/kernel/suspend.c | 18 +++++++++---------
 include/asm-x86_64/suspend.h |  2 +-
 3 files changed, 17 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/arch/i386/power/cpu.c b/arch/i386/power/cpu.c
index d099d01461f..0e6b45b6125 100644
--- a/arch/i386/power/cpu.c
+++ b/arch/i386/power/cpu.c
@@ -44,7 +44,6 @@ void __save_processor_state(struct saved_context *ctxt)
 	 */
 	asm volatile ("sgdt %0" : "=m" (ctxt->gdt_limit));
 	asm volatile ("sidt %0" : "=m" (ctxt->idt_limit));
-	asm volatile ("sldt %0" : "=m" (ctxt->ldt));
 	asm volatile ("str %0"  : "=m" (ctxt->tr));
 
 	/*
@@ -107,7 +106,6 @@ static void fix_processor_context(void)
 
 void __restore_processor_state(struct saved_context *ctxt)
 {
-
 	/*
 	 * control registers
 	 */
@@ -116,6 +114,13 @@ void __restore_processor_state(struct saved_context *ctxt)
 	asm volatile ("movl %0, %%cr2" :: "r" (ctxt->cr2));
 	asm volatile ("movl %0, %%cr0" :: "r" (ctxt->cr0));
 
+	/*
+	 * now restore the descriptor tables to their proper values
+	 * ltr is done i fix_processor_context().
+	 */
+	asm volatile ("lgdt %0" :: "m" (ctxt->gdt_limit));
+	asm volatile ("lidt %0" :: "m" (ctxt->idt_limit));
+
 	/*
 	 * segment registers
 	 */
@@ -124,14 +129,6 @@ void __restore_processor_state(struct saved_context *ctxt)
 	asm volatile ("movw %0, %%gs" :: "r" (ctxt->gs));
 	asm volatile ("movw %0, %%ss" :: "r" (ctxt->ss));
 
-	/*
-	 * now restore the descriptor tables to their proper values
-	 * ltr is done i fix_processor_context().
-	 */
-	asm volatile ("lgdt %0" :: "m" (ctxt->gdt_limit));
-	asm volatile ("lidt %0" :: "m" (ctxt->idt_limit));
-	asm volatile ("lldt %0" :: "m" (ctxt->ldt));
-
 	/*
 	 * sysenter MSRs
 	 */
diff --git a/arch/x86_64/kernel/suspend.c b/arch/x86_64/kernel/suspend.c
index ebaa1e37d65..6c0f402e3a8 100644
--- a/arch/x86_64/kernel/suspend.c
+++ b/arch/x86_64/kernel/suspend.c
@@ -44,7 +44,6 @@ void __save_processor_state(struct saved_context *ctxt)
 	 */
 	asm volatile ("sgdt %0" : "=m" (ctxt->gdt_limit));
 	asm volatile ("sidt %0" : "=m" (ctxt->idt_limit));
-	asm volatile ("sldt %0" : "=m" (ctxt->ldt));
 	asm volatile ("str %0"  : "=m" (ctxt->tr));
 
 	/* XMM0..XMM15 should be handled by kernel_fpu_begin(). */
@@ -69,6 +68,7 @@ void __save_processor_state(struct saved_context *ctxt)
 	asm volatile ("movq %%cr2, %0" : "=r" (ctxt->cr2));
 	asm volatile ("movq %%cr3, %0" : "=r" (ctxt->cr3));
 	asm volatile ("movq %%cr4, %0" : "=r" (ctxt->cr4));
+	asm volatile ("movq %%cr8, %0" : "=r" (ctxt->cr8));
 }
 
 void save_processor_state(void)
@@ -90,11 +90,19 @@ void __restore_processor_state(struct saved_context *ctxt)
 	/*
 	 * control registers
 	 */
+	asm volatile ("movq %0, %%cr8" :: "r" (ctxt->cr8));
 	asm volatile ("movq %0, %%cr4" :: "r" (ctxt->cr4));
 	asm volatile ("movq %0, %%cr3" :: "r" (ctxt->cr3));
 	asm volatile ("movq %0, %%cr2" :: "r" (ctxt->cr2));
 	asm volatile ("movq %0, %%cr0" :: "r" (ctxt->cr0));
 
+	/*
+	 * now restore the descriptor tables to their proper values
+	 * ltr is done i fix_processor_context().
+	 */
+	asm volatile ("lgdt %0" :: "m" (ctxt->gdt_limit));
+	asm volatile ("lidt %0" :: "m" (ctxt->idt_limit));
+
 	/*
 	 * segment registers
 	 */
@@ -108,14 +116,6 @@ void __restore_processor_state(struct saved_context *ctxt)
 	wrmsrl(MSR_GS_BASE, ctxt->gs_base);
 	wrmsrl(MSR_KERNEL_GS_BASE, ctxt->gs_kernel_base);
 
-	/*
-	 * now restore the descriptor tables to their proper values
-	 * ltr is done i fix_processor_context().
-	 */
-	asm volatile ("lgdt %0" :: "m" (ctxt->gdt_limit));
-	asm volatile ("lidt %0" :: "m" (ctxt->idt_limit));
-	asm volatile ("lldt %0" :: "m" (ctxt->ldt));
-
 	fix_processor_context();
 
 	do_fpu_end();
diff --git a/include/asm-x86_64/suspend.h b/include/asm-x86_64/suspend.h
index ec745807fea..bb9f40597d0 100644
--- a/include/asm-x86_64/suspend.h
+++ b/include/asm-x86_64/suspend.h
@@ -16,7 +16,7 @@ arch_prepare_suspend(void)
 struct saved_context {
   	u16 ds, es, fs, gs, ss;
 	unsigned long gs_base, gs_kernel_base, fs_base;
-	unsigned long cr0, cr2, cr3, cr4;
+	unsigned long cr0, cr2, cr3, cr4, cr8;
 	u16 gdt_pad;
 	u16 gdt_limit;
 	unsigned long gdt_base;
-- 
cgit v1.2.3-70-g09d2


From 84dd8d7e9c080b4db66b00b8bc36ccf09a90f824 Mon Sep 17 00:00:00 2001
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Date: Sat, 25 Jun 2005 14:55:26 -0700
Subject: [PATCH] uml: add profile_pc for i386

Cope with a conditional i386 definition, which is wrong for UML.  Before we
just used that one, but it wasn't defined for CONFIG_SMP, so in that case
we got link errors.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Cc: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-um/ptrace-i386.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/asm-um/ptrace-i386.h b/include/asm-um/ptrace-i386.h
index 04222f35c43..fe882b9d917 100644
--- a/include/asm-um/ptrace-i386.h
+++ b/include/asm-um/ptrace-i386.h
@@ -32,6 +32,10 @@
 #define PT_REGS_SYSCALL_RET(r) PT_REGS_EAX(r)
 #define PT_FIX_EXEC_STACK(sp) do ; while(0)
 
+/* Cope with a conditional i386 definition. */
+#undef profile_pc
+#define profile_pc(regs) PT_REGS_IP(regs)
+
 #define user_mode(r) UPT_IS_USER(&(r)->regs)
 
 #endif
-- 
cgit v1.2.3-70-g09d2


From 77fa22450de00d535de2cc8be653983560828000 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sat, 25 Jun 2005 14:55:30 -0700
Subject: [PATCH] s390: improved machine check handling

Improved machine check handling.  Kernel is now able to receive machine checks
while in kernel mode (system call, interrupt and program check handling).
Also register validation is now performed.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/kernel/entry.S       | 102 +++++++++++--
 arch/s390/kernel/entry64.S     |  97 +++++++++++--
 arch/s390/kernel/process.c     |  46 ++----
 arch/s390/kernel/setup.c       |  13 +-
 arch/s390/kernel/smp.c         |  13 +-
 drivers/s390/s390mach.c        | 321 ++++++++++++++++++++++++++++++++++++-----
 drivers/s390/s390mach.h        |  35 ++++-
 include/asm-s390/lowcore.h     |   7 +-
 include/asm-s390/processor.h   |  52 +++----
 include/asm-s390/ptrace.h      |   2 +-
 include/asm-s390/system.h      |  21 ++-
 include/asm-s390/thread_info.h |   2 +
 12 files changed, 576 insertions(+), 135 deletions(-)

(limited to 'include')

diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index c0e09b33feb..5b262b5d869 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -7,6 +7,7 @@
  *    Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
  *               Hartmut Penner (hp@de.ibm.com),
  *               Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com),
+ *		 Heiko Carstens <heiko.carstens@de.ibm.com>
  */
 
 #include <linux/sys.h>
@@ -49,9 +50,9 @@ SP_ILC       =  STACK_FRAME_OVERHEAD + __PT_ILC
 SP_TRAP      =  STACK_FRAME_OVERHEAD + __PT_TRAP
 SP_SIZE      =  STACK_FRAME_OVERHEAD + __PT_SIZE
 
-_TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | \
+_TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | _TIF_MCCK_PENDING | \
 		 _TIF_RESTART_SVC | _TIF_SINGLE_STEP )
-_TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NEED_RESCHED)
+_TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | _TIF_MCCK_PENDING)
 
 STACK_SHIFT = PAGE_SHIFT + THREAD_ORDER
 STACK_SIZE  = 1 << STACK_SHIFT
@@ -121,7 +122,11 @@ STACK_SIZE  = 1 << STACK_SHIFT
 	bz	BASED(stack_overflow)
 3:
 #endif
-2:	s	%r15,BASED(.Lc_spsize)	# make room for registers & psw
+2:
+	.endm
+
+	.macro  CREATE_STACK_FRAME psworg,savearea
+	s	%r15,BASED(.Lc_spsize)	# make room for registers & psw
 	mvc	SP_PSW(8,%r15),0(%r12)	# move user PSW to stack
 	la	%r12,\psworg
 	st	%r2,SP_ORIG_R2(%r15)	# store original content of gpr 2
@@ -161,6 +166,13 @@ __switch_to_base:
         be      __switch_to_noper-__switch_to_base(%r1)	# we got away w/o bashing TLB's
         lctl    %c9,%c11,__THREAD_per(%r3)	# Nope we didn't
 __switch_to_noper:
+	l	%r4,__THREAD_info(%r2)		# get thread_info of prev
+	tm	__TI_flags+3(%r4),_TIF_MCCK_PENDING # machine check pending?
+	bz	__switch_to_no_mcck-__switch_to_base(%r1)
+	ni	__TI_flags+3(%r4),255-_TIF_MCCK_PENDING # clear flag in prev
+	l	%r4,__THREAD_info(%r3)		# get thread_info of next
+	oi	__TI_flags+3(%r4),_TIF_MCCK_PENDING # set it in next
+__switch_to_no_mcck:
         stm     %r6,%r15,__SF_GPRS(%r15)# store __switch_to registers of prev task
 	st	%r15,__THREAD_ksp(%r2)	# store kernel stack to prev->tss.ksp
 	l	%r15,__THREAD_ksp(%r3)	# load kernel stack from next->tss.ksp
@@ -185,6 +197,7 @@ system_call:
 sysc_saveall:
 	SAVE_ALL_BASE __LC_SAVE_AREA
         SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1
+	CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA
 	lh	%r7,0x8a	  # get svc number from lowcore
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
 sysc_vtime:
@@ -234,6 +247,8 @@ sysc_work_loop:
 # One of the work bits is on. Find out which one.
 #
 sysc_work:
+	tm	__TI_flags+3(%r9),_TIF_MCCK_PENDING
+	bo	BASED(sysc_mcck_pending)
 	tm	__TI_flags+3(%r9),_TIF_NEED_RESCHED
 	bo	BASED(sysc_reschedule)
 	tm	__TI_flags+3(%r9),_TIF_SIGPENDING
@@ -252,6 +267,14 @@ sysc_reschedule:
 	la      %r14,BASED(sysc_work_loop)
 	br      %r1		       # call scheduler
 
+#
+# _TIF_MCCK_PENDING is set, call handler
+#
+sysc_mcck_pending:
+	l	%r1,BASED(.Ls390_handle_mcck)
+	la	%r14,BASED(sysc_work_loop)
+	br	%r1			# TIF bit will be cleared by handler
+
 #
 # _TIF_SIGPENDING is set, call do_signal
 #
@@ -430,6 +453,7 @@ pgm_check_handler:
         tm      __LC_PGM_INT_CODE+1,0x80 # check whether we got a per exception
         bnz     BASED(pgm_per)           # got per exception -> special case
 	SAVE_ALL __LC_PGM_OLD_PSW,__LC_SAVE_AREA,1
+	CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
 	tm	SP_PSW+1(%r15),0x01	# interrupting from user ?
 	bz	BASED(pgm_no_vtime)
@@ -468,6 +492,7 @@ pgm_per:
 #
 pgm_per_std:
 	SAVE_ALL __LC_PGM_OLD_PSW,__LC_SAVE_AREA,1
+	CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
 	tm	SP_PSW+1(%r15),0x01	# interrupting from user ?
 	bz	BASED(pgm_no_vtime2)
@@ -493,6 +518,7 @@ pgm_no_vtime2:
 #
 pgm_svcper:
 	SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1
+	CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
 	tm	SP_PSW+1(%r15),0x01	# interrupting from user ?
 	bz	BASED(pgm_no_vtime3)
@@ -521,6 +547,7 @@ io_int_handler:
 	stck	__LC_INT_CLOCK
 	SAVE_ALL_BASE __LC_SAVE_AREA+16
         SAVE_ALL __LC_IO_OLD_PSW,__LC_SAVE_AREA+16,0
+	CREATE_STACK_FRAME __LC_IO_OLD_PSW,__LC_SAVE_AREA+16
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
 	tm	SP_PSW+1(%r15),0x01	# interrupting from user ?
 	bz	BASED(io_no_vtime)
@@ -578,15 +605,25 @@ io_work:
 	lr	%r15,%r1
 #
 # One of the work bits is on. Find out which one.
-# Checked are: _TIF_SIGPENDING and _TIF_NEED_RESCHED
+# Checked are: _TIF_SIGPENDING, _TIF_NEED_RESCHED and _TIF_MCCK_PENDING
 #
 io_work_loop:
+	tm	__TI_flags+3(%r9),_TIF_MCCK_PENDING
+	bo      BASED(io_mcck_pending)
 	tm	__TI_flags+3(%r9),_TIF_NEED_RESCHED
 	bo	BASED(io_reschedule)
 	tm	__TI_flags+3(%r9),_TIF_SIGPENDING
 	bo	BASED(io_sigpending)
 	b	BASED(io_leave)
 
+#
+# _TIF_MCCK_PENDING is set, call handler
+#
+io_mcck_pending:
+	l	%r1,BASED(.Ls390_handle_mcck)
+	l	%r14,BASED(io_work_loop)
+	br	%r1		       # TIF bit will be cleared by handler
+
 #
 # _TIF_NEED_RESCHED is set, call schedule
 #	
@@ -621,6 +658,7 @@ ext_int_handler:
 	stck	__LC_INT_CLOCK
 	SAVE_ALL_BASE __LC_SAVE_AREA+16
         SAVE_ALL __LC_EXT_OLD_PSW,__LC_SAVE_AREA+16,0
+	CREATE_STACK_FRAME __LC_EXT_OLD_PSW,__LC_SAVE_AREA+16
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
 	tm	SP_PSW+1(%r15),0x01	# interrupting from user ?
 	bz	BASED(ext_no_vtime)
@@ -642,19 +680,62 @@ ext_no_vtime:
 
         .globl mcck_int_handler
 mcck_int_handler:
-	STORE_TIMER __LC_ASYNC_ENTER_TIMER
+	spt	__LC_CPU_TIMER_SAVE_AREA	# revalidate cpu timer
+	lm	%r0,%r15,__LC_GPREGS_SAVE_AREA	# revalidate gprs
 	SAVE_ALL_BASE __LC_SAVE_AREA+32
-        SAVE_ALL __LC_MCK_OLD_PSW,__LC_SAVE_AREA+32,0
+	la	%r12,__LC_MCK_OLD_PSW
+	tm	__LC_MCCK_CODE,0x80     # system damage?
+	bo	BASED(mcck_int_main)	# yes -> rest of mcck code invalid
+	tm	__LC_MCCK_CODE+5,0x02   # stored cpu timer value valid?
+	bo	BASED(0f)
+	spt	__LC_LAST_UPDATE_TIMER	# revalidate cpu timer
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
-	tm	SP_PSW+1(%r15),0x01	# interrupting from user ?
+	mvc	__LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER
+	mvc	__LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER
+	mvc	__LC_LAST_UPDATE_TIMER(8),__LC_EXIT_TIMER
+0:	tm	__LC_MCCK_CODE+2,0x08   # mwp of old psw valid?
+	bno	BASED(mcck_no_vtime)	# no -> skip cleanup critical
+	tm	__LC_MCK_OLD_PSW+1,0x01 # interrupting from user ?
 	bz	BASED(mcck_no_vtime)
 	UPDATE_VTIME __LC_EXIT_TIMER,__LC_ASYNC_ENTER_TIMER,__LC_USER_TIMER
 	UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER
 	mvc	__LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER
 mcck_no_vtime:
 #endif
+0:
+	tm	__LC_MCCK_CODE+2,0x09   # mwp + ia of old psw valid?
+	bno	BASED(mcck_int_main)	# no -> skip cleanup critical
+	tm	__LC_MCK_OLD_PSW+1,0x01	# test problem state bit
+	bnz	BASED(mcck_int_main)	# from user -> load async stack
+	clc	__LC_MCK_OLD_PSW+4(4),BASED(.Lcritical_end)
+	bhe	BASED(mcck_int_main)
+	clc	__LC_MCK_OLD_PSW+4(4),BASED(.Lcritical_start)
+	bl	BASED(mcck_int_main)
+	l	%r14,BASED(.Lcleanup_critical)
+	basr	%r14,%r14
+mcck_int_main:
+	l	%r14,__LC_PANIC_STACK	# are we already on the panic stack?
+	slr	%r14,%r15
+	sra	%r14,PAGE_SHIFT
+	be	BASED(0f)
+	l	%r15,__LC_PANIC_STACK	# load panic stack
+0:	CREATE_STACK_FRAME __LC_MCK_OLD_PSW,__LC_SAVE_AREA+32
+	l	%r9,__LC_THREAD_INFO	# load pointer to thread_info struct
+	la	%r2,SP_PTREGS(%r15)	# load pt_regs
 	l       %r1,BASED(.Ls390_mcck)
-	basr    %r14,%r1	  # call machine check handler
+	basr    %r14,%r1		# call machine check handler
+	tm      SP_PSW+1(%r15),0x01	# returning to user ?
+	bno	BASED(mcck_return)
+	l	%r1,__LC_KERNEL_STACK   # switch to kernel stack
+	s	%r1,BASED(.Lc_spsize)
+	mvc	SP_PTREGS(__PT_SIZE,%r1),SP_PTREGS(%r15)
+	xc      __SF_BACKCHAIN(4,%r1),__SF_BACKCHAIN(%r1) # clear back chain
+	lr	%r15,%r1
+	stosm	__SF_EMPTY(%r15),0x04	# turn dat on
+	tm	__TI_flags+3(%r9),_TIF_MCCK_PENDING
+	bno	BASED(mcck_return)
+	l	%r1,BASED(.Ls390_handle_mcck)
+	basr	%r14,%r1		# call machine check handler
 mcck_return:
         RESTORE_ALL 0
 
@@ -742,7 +823,7 @@ cleanup_critical:
 	clc	4(4,%r12),BASED(cleanup_table_sysc_work_loop)
 	bl	BASED(0f)
 	clc	4(4,%r12),BASED(cleanup_table_sysc_work_loop+4)
-	bl	BASED(cleanup_sysc_leave)
+	bl	BASED(cleanup_sysc_return)
 0:
 	br	%r14
 
@@ -760,6 +841,7 @@ cleanup_system_call:
 	mvc	__LC_SAVE_AREA(16),__LC_SAVE_AREA+16
 0:	st	%r13,__LC_SAVE_AREA+20
 	SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1
+	CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA
 	st	%r15,__LC_SAVE_AREA+28
 	lh	%r7,0x8a
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
@@ -834,6 +916,8 @@ cleanup_sysc_leave_insn:
  * Symbol constants
  */
 .Ls390_mcck:   .long  s390_do_machine_check
+.Ls390_handle_mcck:
+	       .long  s390_handle_mcck
 .Ldo_IRQ:      .long  do_IRQ
 .Ldo_extint:   .long  do_extint
 .Ldo_signal:   .long  do_signal
diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S
index 51527ab8c8f..57ca75d0ad7 100644
--- a/arch/s390/kernel/entry64.S
+++ b/arch/s390/kernel/entry64.S
@@ -7,6 +7,7 @@
  *    Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
  *               Hartmut Penner (hp@de.ibm.com),
  *               Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com),
+ *		 Heiko Carstens <heiko.carstens@de.ibm.com>
  */
 
 #include <linux/sys.h>
@@ -52,9 +53,9 @@ SP_SIZE      =  STACK_FRAME_OVERHEAD + __PT_SIZE
 STACK_SHIFT = PAGE_SHIFT + THREAD_ORDER
 STACK_SIZE  = 1 << STACK_SHIFT
 
-_TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | \
+_TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | _TIF_MCCK_PENDING | \
 		 _TIF_RESTART_SVC | _TIF_SINGLE_STEP )
-_TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NEED_RESCHED)
+_TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | _TIF_MCCK_PENDING)
 
 #define BASED(name) name-system_call(%r13)
 
@@ -114,7 +115,11 @@ _TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NEED_RESCHED)
 	jz	stack_overflow
 3:
 #endif
-2:	aghi    %r15,-SP_SIZE		# make room for registers & psw
+2:
+	.endm
+
+	.macro	CREATE_STACK_FRAME psworg,savearea
+	aghi    %r15,-SP_SIZE		# make room for registers & psw
 	mvc     SP_PSW(16,%r15),0(%r12)	# move user PSW to stack
 	la	%r12,\psworg
 	stg	%r2,SP_ORIG_R2(%r15)	# store original content of gpr 2
@@ -152,6 +157,13 @@ __switch_to:
         je      __switch_to_noper            # we got away without bashing TLB's
         lctlg   %c9,%c11,__THREAD_per(%r3)	# Nope we didn't
 __switch_to_noper:
+	lg	%r4,__THREAD_info(%r2)              # get thread_info of prev
+	tm	__TI_flags+7(%r4),_TIF_MCCK_PENDING # machine check pending?
+	jz	__switch_to_no_mcck
+	ni	__TI_flags+7(%r4),255-_TIF_MCCK_PENDING # clear flag in prev
+	lg	%r4,__THREAD_info(%r3)		    # get thread_info of next
+	oi	__TI_flags+7(%r4),_TIF_MCCK_PENDING # set it in next
+__switch_to_no_mcck:
         stmg    %r6,%r15,__SF_GPRS(%r15)# store __switch_to registers of prev task
 	stg	%r15,__THREAD_ksp(%r2)	# store kernel stack to prev->tss.ksp
 	lg	%r15,__THREAD_ksp(%r3)	# load kernel stack from next->tss.ksp
@@ -176,6 +188,7 @@ system_call:
 sysc_saveall:
 	SAVE_ALL_BASE __LC_SAVE_AREA
         SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1
+        CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA
 	llgh    %r7,__LC_SVC_INT_CODE # get svc number from lowcore
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
 sysc_vtime:
@@ -232,6 +245,8 @@ sysc_work_loop:
 # One of the work bits is on. Find out which one.
 #
 sysc_work:
+	tm	__TI_flags+7(%r9),_TIF_MCCK_PENDING
+	jo	sysc_mcck_pending
 	tm	__TI_flags+7(%r9),_TIF_NEED_RESCHED
 	jo	sysc_reschedule
 	tm	__TI_flags+7(%r9),_TIF_SIGPENDING
@@ -249,6 +264,13 @@ sysc_reschedule:
 	larl    %r14,sysc_work_loop
         jg      schedule            # return point is sysc_return
 
+#
+# _TIF_MCCK_PENDING is set, call handler
+#
+sysc_mcck_pending:
+	larl	%r14,sysc_work_loop
+	jg	s390_handle_mcck    # TIF bit will be cleared by handler
+
 #
 # _TIF_SIGPENDING is set, call do_signal
 #
@@ -474,6 +496,7 @@ pgm_check_handler:
         tm      __LC_PGM_INT_CODE+1,0x80 # check whether we got a per exception
         jnz     pgm_per                  # got per exception -> special case
 	SAVE_ALL __LC_PGM_OLD_PSW,__LC_SAVE_AREA,1
+	CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
 	tm	SP_PSW+1(%r15),0x01	# interrupting from user ?
 	jz	pgm_no_vtime
@@ -512,6 +535,7 @@ pgm_per:
 #
 pgm_per_std:
 	SAVE_ALL __LC_PGM_OLD_PSW,__LC_SAVE_AREA,1
+	CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
 	tm	SP_PSW+1(%r15),0x01	# interrupting from user ?
 	jz	pgm_no_vtime2
@@ -537,6 +561,7 @@ pgm_no_vtime2:
 #
 pgm_svcper:
 	SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1
+	CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
 	tm	SP_PSW+1(%r15),0x01	# interrupting from user ?
 	jz	pgm_no_vtime3
@@ -564,6 +589,7 @@ io_int_handler:
 	stck	__LC_INT_CLOCK
 	SAVE_ALL_BASE __LC_SAVE_AREA+32
         SAVE_ALL __LC_IO_OLD_PSW,__LC_SAVE_AREA+32,0
+	CREATE_STACK_FRAME __LC_IO_OLD_PSW,__LC_SAVE_AREA+32
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
 	tm	SP_PSW+1(%r15),0x01	# interrupting from user ?
 	jz	io_no_vtime
@@ -621,15 +647,24 @@ io_work:
 	lgr	%r15,%r1
 #
 # One of the work bits is on. Find out which one.
-# Checked are: _TIF_SIGPENDING and _TIF_NEED_RESCHED
+# Checked are: _TIF_SIGPENDING, _TIF_NEED_RESCHED and _TIF_MCCK_PENDING
 #
 io_work_loop:
+	tm	__TI_flags+7(%r9),_TIF_MCCK_PENDING
+	jo	io_mcck_pending
 	tm	__TI_flags+7(%r9),_TIF_NEED_RESCHED
 	jo	io_reschedule
 	tm	__TI_flags+7(%r9),_TIF_SIGPENDING
 	jo	io_sigpending
 	j	io_leave
 
+#
+# _TIF_MCCK_PENDING is set, call handler
+#
+io_mcck_pending:
+	larl	%r14,io_work_loop
+	jg	s390_handle_mcck	# TIF bit will be cleared by handler
+
 #
 # _TIF_NEED_RESCHED is set, call schedule
 #	
@@ -661,6 +696,7 @@ ext_int_handler:
 	stck	__LC_INT_CLOCK
 	SAVE_ALL_BASE __LC_SAVE_AREA+32
         SAVE_ALL __LC_EXT_OLD_PSW,__LC_SAVE_AREA+32,0
+	CREATE_STACK_FRAME __LC_EXT_OLD_PSW,__LC_SAVE_AREA+32
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
 	tm	SP_PSW+1(%r15),0x01	# interrupting from user ?
 	jz	ext_no_vtime
@@ -680,18 +716,60 @@ ext_no_vtime:
  */
         .globl mcck_int_handler
 mcck_int_handler:
-	STORE_TIMER __LC_ASYNC_ENTER_TIMER
+	la	%r1,4095		# revalidate r1
+	spt	__LC_CPU_TIMER_SAVE_AREA-4095(%r1)	# revalidate cpu timer
+  	lmg     %r0,%r15,__LC_GPREGS_SAVE_AREA-4095(%r1)# revalidate gprs
 	SAVE_ALL_BASE __LC_SAVE_AREA+64
-        SAVE_ALL __LC_MCK_OLD_PSW,__LC_SAVE_AREA+64,0
+	la	%r12,__LC_MCK_OLD_PSW
+	tm	__LC_MCCK_CODE,0x80     # system damage?
+	jo	mcck_int_main		# yes -> rest of mcck code invalid
+	tm	__LC_MCCK_CODE+5,0x02   # stored cpu timer value valid?
+	jo	0f
+	spt	__LC_LAST_UPDATE_TIMER
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
-	tm	SP_PSW+1(%r15),0x01	# interrupting from user ?
+	mvc	__LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER
+	mvc	__LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER
+	mvc	__LC_LAST_UPDATE_TIMER(8),__LC_EXIT_TIMER
+0:	tm	__LC_MCCK_CODE+2,0x08	# mwp of old psw valid?
+	jno	mcck_no_vtime		# no -> no timer update
+	tm      __LC_MCK_OLD_PSW+1,0x01 # interrupting from user ?
 	jz	mcck_no_vtime
 	UPDATE_VTIME __LC_EXIT_TIMER,__LC_ASYNC_ENTER_TIMER,__LC_USER_TIMER
 	UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER
 	mvc	__LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER
 mcck_no_vtime:
 #endif
-	brasl   %r14,s390_do_machine_check
+0:
+	tm	__LC_MCCK_CODE+2,0x09   # mwp + ia of old psw valid?
+	jno	mcck_int_main		# no -> skip cleanup critical
+	tm      __LC_MCK_OLD_PSW+1,0x01 # test problem state bit
+	jnz	mcck_int_main		# from user -> load kernel stack
+	clc	__LC_MCK_OLD_PSW+8(8),BASED(.Lcritical_end)
+	jhe	mcck_int_main
+	clc     __LC_MCK_OLD_PSW+8(8),BASED(.Lcritical_start)
+	jl	mcck_int_main
+	brasl   %r14,cleanup_critical
+mcck_int_main:
+	lg      %r14,__LC_PANIC_STACK   # are we already on the panic stack?
+	slgr	%r14,%r15
+	srag	%r14,%r14,PAGE_SHIFT
+	jz	0f
+	lg      %r15,__LC_PANIC_STACK   # load panic stack
+0:	CREATE_STACK_FRAME __LC_MCK_OLD_PSW,__LC_SAVE_AREA+64
+	lg	%r9,__LC_THREAD_INFO	# load pointer to thread_info struct
+	la	%r2,SP_PTREGS(%r15)	# load pt_regs
+	brasl	%r14,s390_do_machine_check
+	tm	SP_PSW+1(%r15),0x01     # returning to user ?
+	jno	mcck_return
+	lg	%r1,__LC_KERNEL_STACK	# switch to kernel stack
+	aghi	%r1,-SP_SIZE
+	mvc	SP_PTREGS(__PT_SIZE,%r1),SP_PTREGS(%r15)
+	xc	__SF_BACKCHAIN(8,%r1),__SF_BACKCHAIN(%r1) # clear back chain
+	lgr	%r15,%r1
+	stosm	__SF_EMPTY(%r15),0x04	# turn dat on
+	tm	__TI_flags+7(%r9),_TIF_MCCK_PENDING
+	jno	mcck_return
+	brasl	%r14,s390_handle_mcck
 mcck_return:
         RESTORE_ALL 0
 
@@ -775,7 +853,7 @@ cleanup_critical:
 	clc	8(8,%r12),BASED(cleanup_table_sysc_work_loop)
 	jl	0f
 	clc	8(8,%r12),BASED(cleanup_table_sysc_work_loop+8)
-	jl	cleanup_sysc_leave
+	jl	cleanup_sysc_return
 0:
 	br	%r14
 
@@ -793,6 +871,7 @@ cleanup_system_call:
 	mvc	__LC_SAVE_AREA(32),__LC_SAVE_AREA+32
 0:	stg	%r13,__LC_SAVE_AREA+40
 	SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1
+	CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA
 	stg	%r15,__LC_SAVE_AREA+56
 	llgh	%r7,__LC_SVC_INT_CODE
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index 7aea25d6e30..9f3dff6c0b7 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -91,13 +91,12 @@ void do_monitor_call(struct pt_regs *regs, long interruption_code)
 			    (void *)(long) smp_processor_id());
 }
 
+extern void s390_handle_mcck(void);
 /*
  * The idle loop on a S390...
  */
 void default_idle(void)
 {
-	psw_t wait_psw;
-	unsigned long reg;
 	int cpu, rc;
 
 	local_irq_disable();
@@ -125,38 +124,17 @@ void default_idle(void)
 		cpu_die();
 #endif
 
-	/* 
-	 * Wait for external, I/O or machine check interrupt and
-	 * switch off machine check bit after the wait has ended.
-	 */
-	wait_psw.mask = PSW_KERNEL_BITS | PSW_MASK_MCHECK | PSW_MASK_WAIT |
-		PSW_MASK_IO | PSW_MASK_EXT;
-#ifndef CONFIG_ARCH_S390X
-	asm volatile (
-		"    basr %0,0\n"
-		"0:  la   %0,1f-0b(%0)\n"
-		"    st   %0,4(%1)\n"
-		"    oi   4(%1),0x80\n"
-		"    lpsw 0(%1)\n"
-		"1:  la   %0,2f-1b(%0)\n"
-		"    st   %0,4(%1)\n"
-		"    oi   4(%1),0x80\n"
-		"    ni   1(%1),0xf9\n"
-		"    lpsw 0(%1)\n"
-		"2:"
-		: "=&a" (reg) : "a" (&wait_psw) : "memory", "cc" );
-#else /* CONFIG_ARCH_S390X */
-	asm volatile (
-		"    larl  %0,0f\n"
-		"    stg   %0,8(%1)\n"
-		"    lpswe 0(%1)\n"
-		"0:  larl  %0,1f\n"
-		"    stg   %0,8(%1)\n"
-		"    ni    1(%1),0xf9\n"
-		"    lpswe 0(%1)\n"
-		"1:"
-		: "=&a" (reg) : "a" (&wait_psw) : "memory", "cc" );
-#endif /* CONFIG_ARCH_S390X */
+	local_mcck_disable();
+	if (test_thread_flag(TIF_MCCK_PENDING)) {
+		local_mcck_enable();
+		local_irq_enable();
+		s390_handle_mcck();
+		return;
+	}
+
+	/* Wait for external, I/O or machine check interrupt. */
+	__load_psw_mask(PSW_KERNEL_BITS | PSW_MASK_WAIT |
+			PSW_MASK_IO | PSW_MASK_EXT);
 }
 
 void cpu_idle(void)
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index df83215beac..eb7be0ad717 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -414,7 +414,8 @@ setup_lowcore(void)
 	lc->program_new_psw.mask = PSW_KERNEL_BITS;
 	lc->program_new_psw.addr =
 		PSW_ADDR_AMODE | (unsigned long)pgm_check_handler;
-	lc->mcck_new_psw.mask = PSW_KERNEL_BITS;
+	lc->mcck_new_psw.mask =
+		PSW_KERNEL_BITS & ~PSW_MASK_MCHECK & ~PSW_MASK_DAT;
 	lc->mcck_new_psw.addr =
 		PSW_ADDR_AMODE | (unsigned long) mcck_int_handler;
 	lc->io_new_psw.mask = PSW_KERNEL_BITS;
@@ -424,12 +425,18 @@ setup_lowcore(void)
 	lc->kernel_stack = ((unsigned long) &init_thread_union) + THREAD_SIZE;
 	lc->async_stack = (unsigned long)
 		__alloc_bootmem(ASYNC_SIZE, ASYNC_SIZE, 0) + ASYNC_SIZE;
-#ifdef CONFIG_CHECK_STACK
 	lc->panic_stack = (unsigned long)
 		__alloc_bootmem(PAGE_SIZE, PAGE_SIZE, 0) + PAGE_SIZE;
-#endif
 	lc->current_task = (unsigned long) init_thread_union.thread_info.task;
 	lc->thread_info = (unsigned long) &init_thread_union;
+#ifndef CONFIG_ARCH_S390X
+	if (MACHINE_HAS_IEEE) {
+		lc->extended_save_area_addr = (__u32)
+			__alloc_bootmem(PAGE_SIZE, PAGE_SIZE, 0);
+		/* enable extended save area */
+		ctl_set_bit(14, 29);
+	}
+#endif
 #ifdef CONFIG_ARCH_S390X
 	if (MACHINE_HAS_DIAG44)
 		lc->diag44_opcode = 0x83000044;
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 93c71fef99d..50c335067cf 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -773,13 +773,24 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 
 		*(lowcore_ptr[i]) = S390_lowcore;
 		lowcore_ptr[i]->async_stack = stack + (ASYNC_SIZE);
-#ifdef CONFIG_CHECK_STACK
 		stack = __get_free_pages(GFP_KERNEL,0);
 		if (stack == 0ULL)
 			panic("smp_boot_cpus failed to allocate memory\n");
 		lowcore_ptr[i]->panic_stack = stack + (PAGE_SIZE);
+#ifndef __s390x__
+		if (MACHINE_HAS_IEEE) {
+			lowcore_ptr[i]->extended_save_area_addr =
+				(__u32) __get_free_pages(GFP_KERNEL,0);
+			if (lowcore_ptr[i]->extended_save_area_addr == 0)
+				panic("smp_boot_cpus failed to "
+				      "allocate memory\n");
+		}
 #endif
 	}
+#ifndef __s390x__
+	if (MACHINE_HAS_IEEE)
+		ctl_set_bit(14, 29); /* enable extended save area */
+#endif
 	set_prefix((u32)(unsigned long) lowcore_ptr[smp_processor_id()]);
 
 	for_each_cpu(cpu)
diff --git a/drivers/s390/s390mach.c b/drivers/s390/s390mach.c
index ffa996c8a90..5bb255e02ac 100644
--- a/drivers/s390/s390mach.c
+++ b/drivers/s390/s390mach.c
@@ -31,14 +31,14 @@ extern void css_reiterate_subchannels(void);
 extern struct workqueue_struct *slow_path_wq;
 extern struct work_struct slow_path_work;
 
-static void
+static NORET_TYPE void
 s390_handle_damage(char *msg)
 {
-	printk(KERN_EMERG "%s\n", msg);
 #ifdef CONFIG_SMP
 	smp_send_stop();
 #endif
 	disabled_wait((unsigned long) __builtin_return_address(0));
+	for(;;);
 }
 
 /*
@@ -122,40 +122,39 @@ repeat:
 	return 0;
 }
 
+struct mcck_struct {
+	int kill_task;
+	int channel_report;
+	int warning;
+	unsigned long long mcck_code;
+};
+
+static DEFINE_PER_CPU(struct mcck_struct, cpu_mcck);
+
 /*
- * machine check handler.
+ * Main machine check handler function. Will be called with interrupts enabled
+ * or disabled and machine checks enabled or disabled.
  */
 void
-s390_do_machine_check(void)
+s390_handle_mcck(void)
 {
-	struct mci *mci;
-
-	mci = (struct mci *) &S390_lowcore.mcck_interruption_code;
+	unsigned long flags;
+	struct mcck_struct mcck;
 
-	if (mci->sd)		/* system damage */
-		s390_handle_damage("received system damage machine check\n");
+	/*
+	 * Disable machine checks and get the current state of accumulated
+	 * machine checks. Afterwards delete the old state and enable machine
+	 * checks again.
+	 */
+	local_irq_save(flags);
+	local_mcck_disable();
+	mcck = __get_cpu_var(cpu_mcck);
+	memset(&__get_cpu_var(cpu_mcck), 0, sizeof(struct mcck_struct));
+	clear_thread_flag(TIF_MCCK_PENDING);
+	local_mcck_enable();
+	local_irq_restore(flags);
 
-	if (mci->pd)		/* instruction processing damage */
-		s390_handle_damage("received instruction processing "
-				   "damage machine check\n");
-
-	if (mci->se)		/* storage error uncorrected */
-		s390_handle_damage("received storage error uncorrected "
-				   "machine check\n");
-
-	if (mci->sc)		/* storage error corrected */
-		printk(KERN_WARNING
-		       "received storage error corrected machine check\n");
-
-	if (mci->ke)		/* storage key-error uncorrected */
-		s390_handle_damage("received storage key-error uncorrected "
-				   "machine check\n");
-
-	if (mci->ds && mci->fa)	/* storage degradation */
-		s390_handle_damage("received storage degradation machine "
-				   "check\n");
-
-	if (mci->cp)		/* channel report word pending */
+	if (mcck.channel_report)
 		up(&m_sem);
 
 #ifdef CONFIG_MACHCHK_WARNING
@@ -168,7 +167,7 @@ s390_do_machine_check(void)
  * On VM we only get one interrupt per virtally presented machinecheck.
  * Though one suffices, we may get one interrupt per (virtual) processor.
  */
-	if (mci->w) {	/* WARNING pending ? */
+	if (mcck.warning) {	/* WARNING pending ? */
 		static int mchchk_wng_posted = 0;
 		/*
 		 * Use single machine clear, as we cannot handle smp right now
@@ -178,6 +177,261 @@ s390_do_machine_check(void)
 			kill_proc(1, SIGPWR, 1);
 	}
 #endif
+
+	if (mcck.kill_task) {
+		local_irq_enable();
+		printk(KERN_EMERG "mcck: Terminating task because of machine "
+		       "malfunction (code 0x%016llx).\n", mcck.mcck_code);
+		printk(KERN_EMERG "mcck: task: %s, pid: %d.\n",
+		       current->comm, current->pid);
+		do_exit(SIGSEGV);
+	}
+}
+
+/*
+ * returns 0 if all registers could be validated
+ * returns 1 otherwise
+ */
+static int
+s390_revalidate_registers(struct mci *mci)
+{
+	int kill_task;
+	u64 tmpclock;
+	u64 zero;
+	void *fpt_save_area, *fpt_creg_save_area;
+
+	kill_task = 0;
+	zero = 0;
+	/* General purpose registers */
+	if (!mci->gr)
+		/*
+		 * General purpose registers couldn't be restored and have
+		 * unknown contents. Process needs to be terminated.
+		 */
+		kill_task = 1;
+
+	/* Revalidate floating point registers */
+	if (!mci->fp)
+		/*
+		 * Floating point registers can't be restored and
+		 * therefore the process needs to be terminated.
+		 */
+		kill_task = 1;
+
+#ifndef __s390x__
+	asm volatile("ld 0,0(%0)\n"
+		     "ld 2,8(%0)\n"
+		     "ld 4,16(%0)\n"
+		     "ld 6,24(%0)"
+		     : : "a" (&S390_lowcore.floating_pt_save_area));
+#endif
+
+	if (MACHINE_HAS_IEEE) {
+#ifdef __s390x__
+		fpt_save_area = &S390_lowcore.floating_pt_save_area;
+		fpt_creg_save_area = &S390_lowcore.fpt_creg_save_area;
+#else
+		fpt_save_area = (void *) S390_lowcore.extended_save_area_addr;
+		fpt_creg_save_area = fpt_save_area+128;
+#endif
+		/* Floating point control register */
+		if (!mci->fc) {
+			/*
+			 * Floating point control register can't be restored.
+			 * Task will be terminated.
+			 */
+			asm volatile ("lfpc 0(%0)" : : "a" (&zero));
+			kill_task = 1;
+
+		}
+		else
+			asm volatile (
+				"lfpc 0(%0)"
+				: : "a" (fpt_creg_save_area));
+
+		asm volatile("ld  0,0(%0)\n"
+			     "ld  1,8(%0)\n"
+			     "ld  2,16(%0)\n"
+			     "ld  3,24(%0)\n"
+			     "ld  4,32(%0)\n"
+			     "ld  5,40(%0)\n"
+			     "ld  6,48(%0)\n"
+			     "ld  7,56(%0)\n"
+			     "ld  8,64(%0)\n"
+			     "ld  9,72(%0)\n"
+			     "ld 10,80(%0)\n"
+			     "ld 11,88(%0)\n"
+			     "ld 12,96(%0)\n"
+			     "ld 13,104(%0)\n"
+			     "ld 14,112(%0)\n"
+			     "ld 15,120(%0)\n"
+			     : : "a" (fpt_save_area));
+	}
+
+	/* Revalidate access registers */
+	asm volatile("lam 0,15,0(%0)"
+		     : : "a" (&S390_lowcore.access_regs_save_area));
+	if (!mci->ar)
+		/*
+		 * Access registers have unknown contents.
+		 * Terminating task.
+		 */
+		kill_task = 1;
+
+	/* Revalidate control registers */
+	if (!mci->cr)
+		/*
+		 * Control registers have unknown contents.
+		 * Can't recover and therefore stopping machine.
+		 */
+		s390_handle_damage("invalid control registers.");
+	else
+#ifdef __s390x__
+		asm volatile("lctlg 0,15,0(%0)"
+			     : : "a" (&S390_lowcore.cregs_save_area));
+#else
+		asm volatile("lctl 0,15,0(%0)"
+			     : : "a" (&S390_lowcore.cregs_save_area));
+#endif
+
+	/*
+	 * We don't even try to revalidate the TOD register, since we simply
+	 * can't write something sensible into that register.
+	 */
+
+#ifdef __s390x__
+	/*
+	 * See if we can revalidate the TOD programmable register with its
+	 * old contents (should be zero) otherwise set it to zero.
+	 */
+	if (!mci->pr)
+		asm volatile("sr 0,0\n"
+			     "sckpf"
+			     : : : "0", "cc");
+	else
+		asm volatile(
+			"l 0,0(%0)\n"
+			"sckpf"
+			: : "a" (&S390_lowcore.tod_progreg_save_area) : "0", "cc");
+#endif
+
+	/* Revalidate clock comparator register */
+	asm volatile ("stck 0(%1)\n"
+		      "sckc 0(%1)"
+		      : "=m" (tmpclock) : "a" (&(tmpclock)) : "cc", "memory");
+
+	/* Check if old PSW is valid */
+	if (!mci->wp)
+		/*
+		 * Can't tell if we come from user or kernel mode
+		 * -> stopping machine.
+		 */
+		s390_handle_damage("old psw invalid.");
+
+	if (!mci->ms || !mci->pm || !mci->ia)
+		kill_task = 1;
+
+	return kill_task;
+}
+
+/*
+ * machine check handler.
+ */
+void
+s390_do_machine_check(struct pt_regs *regs)
+{
+	struct mci *mci;
+	struct mcck_struct *mcck;
+	int umode;
+
+	mci = (struct mci *) &S390_lowcore.mcck_interruption_code;
+	mcck = &__get_cpu_var(cpu_mcck);
+	umode = user_mode(regs);
+
+	if (mci->sd)
+		/* System damage -> stopping machine */
+		s390_handle_damage("received system damage machine check.");
+
+	if (mci->pd) {
+		if (mci->b) {
+			/* Processing backup -> verify if we can survive this */
+			u64 z_mcic, o_mcic, t_mcic;
+#ifdef __s390x__
+			z_mcic = (1ULL<<63 | 1ULL<<59 | 1ULL<<29);
+			o_mcic = (1ULL<<43 | 1ULL<<42 | 1ULL<<41 | 1ULL<<40 |
+				  1ULL<<36 | 1ULL<<35 | 1ULL<<34 | 1ULL<<32 |
+				  1ULL<<30 | 1ULL<<21 | 1ULL<<20 | 1ULL<<17 |
+				  1ULL<<16);
+#else
+			z_mcic = (1ULL<<63 | 1ULL<<59 | 1ULL<<57 | 1ULL<<50 |
+				  1ULL<<29);
+			o_mcic = (1ULL<<43 | 1ULL<<42 | 1ULL<<41 | 1ULL<<40 |
+				  1ULL<<36 | 1ULL<<35 | 1ULL<<34 | 1ULL<<32 |
+				  1ULL<<30 | 1ULL<<20 | 1ULL<<17 | 1ULL<<16);
+#endif
+			t_mcic = *(u64 *)mci;
+
+			if (((t_mcic & z_mcic) != 0) ||
+			    ((t_mcic & o_mcic) != o_mcic)) {
+				s390_handle_damage("processing backup machine "
+						   "check with damage.");
+			}
+			if (!umode)
+				s390_handle_damage("processing backup machine "
+						   "check in kernel mode.");
+			mcck->kill_task = 1;
+			mcck->mcck_code = *(unsigned long long *) mci;
+		}
+		else {
+			/* Processing damage -> stopping machine */
+			s390_handle_damage("received instruction processing "
+					   "damage machine check.");
+		}
+	}
+	if (s390_revalidate_registers(mci)) {
+		if (umode) {
+			/*
+			 * Couldn't restore all register contents while in
+			 * user mode -> mark task for termination.
+			 */
+			mcck->kill_task = 1;
+			mcck->mcck_code = *(unsigned long long *) mci;
+			set_thread_flag(TIF_MCCK_PENDING);
+		}
+		else
+			/*
+			 * Couldn't restore all register contents while in
+			 * kernel mode -> stopping machine.
+			 */
+			s390_handle_damage("unable to revalidate registers.");
+	}
+
+	if (mci->se)
+		/* Storage error uncorrected */
+		s390_handle_damage("received storage error uncorrected "
+				   "machine check.");
+
+	if (mci->ke)
+		/* Storage key-error uncorrected */
+		s390_handle_damage("received storage key-error uncorrected "
+				   "machine check.");
+
+	if (mci->ds && mci->fa)
+		/* Storage degradation */
+		s390_handle_damage("received storage degradation machine "
+				   "check.");
+
+	if (mci->cp) {
+		/* Channel report word pending */
+		mcck->channel_report = 1;
+		set_thread_flag(TIF_MCCK_PENDING);
+	}
+
+	if (mci->w) {
+		/* Warning pending */
+		mcck->warning = 1;
+		set_thread_flag(TIF_MCCK_PENDING);
+	}
 }
 
 /*
@@ -189,9 +443,8 @@ static int
 machine_check_init(void)
 {
 	init_MUTEX_LOCKED(&m_sem);
-	ctl_clear_bit(14, 25);	/* disable damage MCH */
-	ctl_set_bit(14, 26);	/* enable degradation MCH */
-	ctl_set_bit(14, 27);	/* enable system recovery MCH */
+	ctl_clear_bit(14, 25);	/* disable external damage MCH */
+	ctl_set_bit(14, 27);    /* enable system recovery MCH */
 #ifdef CONFIG_MACHCHK_WARNING
 	ctl_set_bit(14, 24);	/* enable warning MCH */
 #endif
diff --git a/drivers/s390/s390mach.h b/drivers/s390/s390mach.h
index 7e26f0f1b0d..4eaa7017918 100644
--- a/drivers/s390/s390mach.h
+++ b/drivers/s390/s390mach.h
@@ -16,20 +16,45 @@ struct mci {
 	__u32   sd              :  1; /* 00 system damage */
 	__u32   pd              :  1; /* 01 instruction-processing damage */
 	__u32   sr              :  1; /* 02 system recovery */
-	__u32   to_be_defined_1 :  4; /* 03-06 */
+	__u32   to_be_defined_1 :  1; /* 03 */
+	__u32   cd              :  1; /* 04 timing-facility damage */
+	__u32   ed              :  1; /* 05 external damage */
+	__u32   to_be_defined_2 :  1; /* 06 */
 	__u32   dg              :  1; /* 07 degradation */
 	__u32   w               :  1; /* 08 warning pending */
 	__u32   cp              :  1; /* 09 channel-report pending */
-	__u32   to_be_defined_2 :  6; /* 10-15 */
+	__u32   sp              :  1; /* 10 service-processor damage */
+	__u32   ck              :  1; /* 11 channel-subsystem damage */
+	__u32   to_be_defined_3 :  2; /* 12-13 */
+	__u32   b               :  1; /* 14 backed up */
+	__u32   to_be_defined_4 :  1; /* 15 */
 	__u32   se              :  1; /* 16 storage error uncorrected */
 	__u32   sc              :  1; /* 17 storage error corrected */
 	__u32   ke              :  1; /* 18 storage-key error uncorrected */
 	__u32   ds              :  1; /* 19 storage degradation */
-	__u32	to_be_defined_3 :  4; /* 20-23 */
+	__u32   wp              :  1; /* 20 psw mwp validity */
+	__u32   ms              :  1; /* 21 psw mask and key validity */
+	__u32   pm              :  1; /* 22 psw program mask and cc validity */
+	__u32   ia              :  1; /* 23 psw instruction address validity */
 	__u32   fa              :  1; /* 24 failing storage address validity */
-	__u32   to_be_defined_4 :  7; /* 25-31 */
+	__u32   to_be_defined_5 :  1; /* 25 */
+	__u32   ec              :  1; /* 26 external damage code validity */
+	__u32   fp              :  1; /* 27 floating point register validity */
+	__u32   gr              :  1; /* 28 general register validity */
+	__u32   cr              :  1; /* 29 control register validity */
+	__u32   to_be_defined_6 :  1; /* 30 */
+	__u32   st              :  1; /* 31 storage logical validity */
 	__u32   ie              :  1; /* 32 indirect storage error */
-	__u32	to_be_defined_5 : 31; /* 33-63 */
+	__u32   ar              :  1; /* 33 access register validity */
+	__u32   da              :  1; /* 34 delayed access exception */
+	__u32   to_be_defined_7 :  7; /* 35-41 */
+	__u32   pr              :  1; /* 42 tod programmable register validity */
+	__u32   fc              :  1; /* 43 fp control register validity */
+	__u32   ap              :  1; /* 44 ancillary report */
+	__u32   to_be_defined_8 :  1; /* 45 */
+	__u32   ct              :  1; /* 46 cpu timer validity */
+	__u32   cc              :  1; /* 47 clock comparator validity */
+	__u32	to_be_defined_9 : 16; /* 47-63 */
 };
 
 /*
diff --git a/include/asm-s390/lowcore.h b/include/asm-s390/lowcore.h
index df5172fc589..76b5b19c0ae 100644
--- a/include/asm-s390/lowcore.h
+++ b/include/asm-s390/lowcore.h
@@ -109,10 +109,14 @@
 
 #ifndef __s390x__
 #define __LC_PFAULT_INTPARM             0x080
+#define __LC_CPU_TIMER_SAVE_AREA        0x0D8
 #define __LC_AREGS_SAVE_AREA            0x120
+#define __LC_GPREGS_SAVE_AREA           0x180
 #define __LC_CREGS_SAVE_AREA            0x1C0
 #else /* __s390x__ */
 #define __LC_PFAULT_INTPARM             0x11B8
+#define __LC_GPREGS_SAVE_AREA           0x1280
+#define __LC_CPU_TIMER_SAVE_AREA        0x1328
 #define __LC_AREGS_SAVE_AREA            0x1340
 #define __LC_CREGS_SAVE_AREA            0x1380
 #endif /* __s390x__ */
@@ -167,7 +171,8 @@ struct _lowcore
 	__u16        subchannel_nr;            /* 0x0ba */
 	__u32        io_int_parm;              /* 0x0bc */
 	__u32        io_int_word;              /* 0x0c0 */
-        __u8         pad3[0xD8-0xC4];          /* 0x0c4 */
+        __u8         pad3[0xD4-0xC4];          /* 0x0c4 */
+	__u32        extended_save_area_addr;  /* 0x0d4 */
 	__u32        cpu_timer_save_area[2];   /* 0x0d8 */
 	__u32        clock_comp_save_area[2];  /* 0x0e0 */
 	__u32        mcck_interruption_code[2]; /* 0x0e8 */
diff --git a/include/asm-s390/processor.h b/include/asm-s390/processor.h
index fb46e9090b5..8bd14de69e3 100644
--- a/include/asm-s390/processor.h
+++ b/include/asm-s390/processor.h
@@ -206,6 +206,18 @@ unsigned long get_wchan(struct task_struct *p);
 	asm volatile ("ex 0,%0" : : "i" (__LC_DIAG44_OPCODE) : "memory")
 #endif /* __s390x__ */
 
+/*
+ * Set PSW to specified value.
+ */
+static inline void __load_psw(psw_t psw)
+{
+#ifndef __s390x__
+	asm volatile ("lpsw  0(%0)" : : "a" (&psw), "m" (psw) : "cc" );
+#else
+	asm volatile ("lpswe 0(%0)" : : "a" (&psw), "m" (psw) : "cc" );
+#endif
+}
+
 /*
  * Set PSW mask to specified value, while leaving the
  * PSW addr pointing to the next instruction.
@@ -214,8 +226,8 @@ unsigned long get_wchan(struct task_struct *p);
 static inline void __load_psw_mask (unsigned long mask)
 {
 	unsigned long addr;
-
 	psw_t psw;
+
 	psw.mask = mask;
 
 #ifndef __s390x__
@@ -241,30 +253,8 @@ static inline void __load_psw_mask (unsigned long mask)
  */
 static inline void enabled_wait(void)
 {
-	unsigned long reg;
-	psw_t wait_psw;
-
-	wait_psw.mask = PSW_BASE_BITS | PSW_MASK_IO | PSW_MASK_EXT |
-		PSW_MASK_MCHECK | PSW_MASK_WAIT | PSW_DEFAULT_KEY;
-#ifndef __s390x__
-	asm volatile (
-		"    basr %0,0\n"
-		"0:  la   %0,1f-0b(%0)\n"
-		"    st   %0,4(%1)\n"
-		"    oi   4(%1),0x80\n"
-		"    lpsw 0(%1)\n"
-		"1:"
-		: "=&a" (reg) : "a" (&wait_psw), "m" (wait_psw)
-		: "memory", "cc" );
-#else /* __s390x__ */
-	asm volatile (
-		"    larl  %0,0f\n"
-		"    stg   %0,8(%1)\n"
-		"    lpswe 0(%1)\n"
-		"0:"
-		: "=&a" (reg) : "a" (&wait_psw), "m" (wait_psw)
-		: "memory", "cc" );
-#endif /* __s390x__ */
+	__load_psw_mask(PSW_BASE_BITS | PSW_MASK_IO | PSW_MASK_EXT |
+			PSW_MASK_MCHECK | PSW_MASK_WAIT | PSW_DEFAULT_KEY);
 }
 
 /*
@@ -273,13 +263,11 @@ static inline void enabled_wait(void)
 
 static inline void disabled_wait(unsigned long code)
 {
-        char psw_buffer[2*sizeof(psw_t)];
         unsigned long ctl_buf;
-        psw_t *dw_psw = (psw_t *)(((unsigned long) &psw_buffer+sizeof(psw_t)-1)
-                                  & -sizeof(psw_t));
+        psw_t dw_psw;
 
-        dw_psw->mask = PSW_BASE_BITS | PSW_MASK_WAIT;
-        dw_psw->addr = code;
+        dw_psw.mask = PSW_BASE_BITS | PSW_MASK_WAIT;
+        dw_psw.addr = code;
         /* 
          * Store status and then load disabled wait psw,
          * the processor is dead afterwards
@@ -301,7 +289,7 @@ static inline void disabled_wait(unsigned long code)
                       "    oi    0x1c0,0x10\n" /* fake protection bit */
                       "    lpsw 0(%1)"
                       : "=m" (ctl_buf)
-		      : "a" (dw_psw), "a" (&ctl_buf), "m" (dw_psw) : "cc" );
+		      : "a" (&dw_psw), "a" (&ctl_buf), "m" (dw_psw) : "cc" );
 #else /* __s390x__ */
         asm volatile ("    stctg 0,0,0(%2)\n"
                       "    ni    4(%2),0xef\n" /* switch off protection */
@@ -333,7 +321,7 @@ static inline void disabled_wait(unsigned long code)
                       "    oi    0x384(1),0x10\n" /* fake protection bit */
                       "    lpswe 0(%1)"
                       : "=m" (ctl_buf)
-		      : "a" (dw_psw), "a" (&ctl_buf),
+		      : "a" (&dw_psw), "a" (&ctl_buf),
 		        "m" (dw_psw) : "cc", "0", "1");
 #endif /* __s390x__ */
 }
diff --git a/include/asm-s390/ptrace.h b/include/asm-s390/ptrace.h
index 4eff8f2e3bf..fc7c96edc69 100644
--- a/include/asm-s390/ptrace.h
+++ b/include/asm-s390/ptrace.h
@@ -276,7 +276,7 @@ typedef struct
 #endif /* __s390x__ */
 
 #define PSW_KERNEL_BITS	(PSW_BASE_BITS | PSW_MASK_DAT | PSW_ASC_PRIMARY | \
-			 PSW_DEFAULT_KEY)
+			 PSW_MASK_MCHECK | PSW_DEFAULT_KEY)
 #define PSW_USER_BITS	(PSW_BASE_BITS | PSW_MASK_DAT | PSW_ASC_HOME | \
 			 PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK | \
 			 PSW_MASK_PSTATE | PSW_DEFAULT_KEY)
diff --git a/include/asm-s390/system.h b/include/asm-s390/system.h
index 81514d76edc..e3cb3ce1d24 100644
--- a/include/asm-s390/system.h
+++ b/include/asm-s390/system.h
@@ -16,6 +16,7 @@
 #include <asm/types.h>
 #include <asm/ptrace.h>
 #include <asm/setup.h>
+#include <asm/processor.h>
 
 #ifdef __KERNEL__
 
@@ -331,9 +332,6 @@ __cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size)
 
 #ifdef __s390x__
 
-#define __load_psw(psw) \
-        __asm__ __volatile__("lpswe 0(%0)" : : "a" (&psw), "m" (psw) : "cc" );
-
 #define __ctl_load(array, low, high) ({ \
 	typedef struct { char _[sizeof(array)]; } addrtype; \
 	__asm__ __volatile__ ( \
@@ -390,9 +388,6 @@ __cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size)
 
 #else /* __s390x__ */
 
-#define __load_psw(psw) \
-	__asm__ __volatile__("lpsw 0(%0)" : : "a" (&psw) : "cc" );
-
 #define __ctl_load(array, low, high) ({ \
 	typedef struct { char _[sizeof(array)]; } addrtype; \
 	__asm__ __volatile__ ( \
@@ -451,6 +446,20 @@ __cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size)
 /* For spinlocks etc */
 #define local_irq_save(x)	((x) = local_irq_disable())
 
+/*
+ * Use to set psw mask except for the first byte which
+ * won't be changed by this function.
+ */
+static inline void
+__set_psw_mask(unsigned long mask)
+{
+	local_save_flags(mask);
+	__load_psw_mask(mask);
+}
+
+#define local_mcck_enable()  __set_psw_mask(PSW_KERNEL_BITS)
+#define local_mcck_disable() __set_psw_mask(PSW_KERNEL_BITS & ~PSW_MASK_MCHECK)
+
 #ifdef CONFIG_SMP
 
 extern void smp_ctl_set_bit(int cr, int bit);
diff --git a/include/asm-s390/thread_info.h b/include/asm-s390/thread_info.h
index fe101d41e84..6c18a3f2431 100644
--- a/include/asm-s390/thread_info.h
+++ b/include/asm-s390/thread_info.h
@@ -96,6 +96,7 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_RESTART_SVC		4	/* restart svc with new svc number */
 #define TIF_SYSCALL_AUDIT	5	/* syscall auditing active */
 #define TIF_SINGLE_STEP		6	/* deliver sigtrap on return to user */
+#define TIF_MCCK_PENDING	7	/* machine check handling is pending */
 #define TIF_USEDFPU		16	/* FPU was used by this task this quantum (SMP) */
 #define TIF_POLLING_NRFLAG	17	/* true if poll_idle() is polling 
 					   TIF_NEED_RESCHED */
@@ -109,6 +110,7 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_RESTART_SVC	(1<<TIF_RESTART_SVC)
 #define _TIF_SYSCALL_AUDIT	(1<<TIF_SYSCALL_AUDIT)
 #define _TIF_SINGLE_STEP	(1<<TIF_SINGLE_STEP)
+#define _TIF_MCCK_PENDING	(1<<TIF_MCCK_PENDING)
 #define _TIF_USEDFPU		(1<<TIF_USEDFPU)
 #define _TIF_POLLING_NRFLAG	(1<<TIF_POLLING_NRFLAG)
 #define _TIF_31BIT		(1<<TIF_31BIT)
-- 
cgit v1.2.3-70-g09d2


From 6b979de395c7e1b7e59f74a870e1d1911853eccb Mon Sep 17 00:00:00 2001
From: Christian Borntraeger <cborntra@de.ibm.com>
Date: Sat, 25 Jun 2005 14:55:32 -0700
Subject: [PATCH] s390: add vmcp interface

Add interface to issue VM control program commands.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/defconfig          |   1 +
 arch/s390/kernel/cpcmd.c     | 109 ++++++++++++---------
 arch/s390/kernel/setup.c     |   6 +-
 arch/s390/kernel/smp.c       |   6 +-
 arch/s390/kernel/traps.c     |   2 +-
 arch/s390/mm/extmem.c        |   4 +-
 drivers/s390/Kconfig         |   7 ++
 drivers/s390/char/Makefile   |   1 +
 drivers/s390/char/con3215.c  |   4 +-
 drivers/s390/char/con3270.c  |   4 +-
 drivers/s390/char/vmcp.c     | 219 +++++++++++++++++++++++++++++++++++++++++++
 drivers/s390/char/vmcp.h     |  30 ++++++
 drivers/s390/char/vmlogrdr.c |  10 +-
 drivers/s390/net/smsgiucv.c  |   4 +-
 include/asm-s390/cpcmd.h     |  18 +++-
 15 files changed, 359 insertions(+), 66 deletions(-)
 create mode 100644 drivers/s390/char/vmcp.c
 create mode 100644 drivers/s390/char/vmcp.h

(limited to 'include')

diff --git a/arch/s390/defconfig b/arch/s390/defconfig
index 07fd0414a4b..89850b2c27e 100644
--- a/arch/s390/defconfig
+++ b/arch/s390/defconfig
@@ -245,6 +245,7 @@ CONFIG_S390_TAPE_BLOCK=y
 #
 CONFIG_S390_TAPE_34XX=m
 # CONFIG_VMLOGRDR is not set
+# CONFIG_VMCP is not set
 # CONFIG_MONREADER is not set
 # CONFIG_DCSS_SHM is not set
 
diff --git a/arch/s390/kernel/cpcmd.c b/arch/s390/kernel/cpcmd.c
index 44df8dc07c5..20062145e84 100644
--- a/arch/s390/kernel/cpcmd.c
+++ b/arch/s390/kernel/cpcmd.c
@@ -2,7 +2,7 @@
  *  arch/s390/kernel/cpcmd.c
  *
  *  S390 version
- *    Copyright (C) 1999,2000 IBM Deutschland Entwicklung GmbH, IBM Corporation
+ *    Copyright (C) 1999,2005 IBM Deutschland Entwicklung GmbH, IBM Corporation
  *    Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
  *               Christian Borntraeger (cborntra@de.ibm.com),
  */
@@ -18,93 +18,114 @@
 #include <asm/system.h>
 
 static DEFINE_SPINLOCK(cpcmd_lock);
-static char cpcmd_buf[240];
+static char cpcmd_buf[241];
 
 /*
  * the caller of __cpcmd has to ensure that the response buffer is below 2 GB
  */
-void __cpcmd(char *cmd, char *response, int rlen)
+int  __cpcmd(const char *cmd, char *response, int rlen, int *response_code)
 {
 	const int mask = 0x40000000L;
 	unsigned long flags;
+	int return_code;
+	int return_len;
 	int cmdlen;
 
 	spin_lock_irqsave(&cpcmd_lock, flags);
 	cmdlen = strlen(cmd);
 	BUG_ON(cmdlen > 240);
-	strcpy(cpcmd_buf, cmd);
+	memcpy(cpcmd_buf, cmd, cmdlen);
 	ASCEBC(cpcmd_buf, cmdlen);
 
 	if (response != NULL && rlen > 0) {
 		memset(response, 0, rlen);
 #ifndef CONFIG_ARCH_S390X
-		asm volatile ("LRA   2,0(%0)\n\t"
-                              "LR    4,%1\n\t"
-                              "O     4,%4\n\t"
-                              "LRA   3,0(%2)\n\t"
-                              "LR    5,%3\n\t"
-                              ".long 0x83240008 # Diagnose X'08'\n\t"
-                              : /* no output */
-                              : "a" (cpcmd_buf), "d" (cmdlen),
-                                "a" (response), "d" (rlen), "m" (mask)
-                              : "cc", "2", "3", "4", "5" );
+		asm volatile (	"lra	2,0(%2)\n"
+				"lr	4,%3\n"
+				"o	4,%6\n"
+				"lra	3,0(%4)\n"
+				"lr	5,%5\n"
+				"diag	2,4,0x8\n"
+				"brc	8, .Litfits\n"
+				"ar	5, %5\n"
+				".Litfits: \n"
+				"lr	%0,4\n"
+				"lr	%1,5\n"
+				: "=d" (return_code), "=d" (return_len)
+				: "a" (cpcmd_buf), "d" (cmdlen),
+				"a" (response), "d" (rlen), "m" (mask)
+				: "cc", "2", "3", "4", "5" );
 #else /* CONFIG_ARCH_S390X */
-                asm volatile ("   lrag  2,0(%0)\n"
-                              "   lgr   4,%1\n"
-                              "   o     4,%4\n"
-                              "   lrag  3,0(%2)\n"
-                              "   lgr   5,%3\n"
-                              "   sam31\n"
-                              "   .long 0x83240008 # Diagnose X'08'\n"
-                              "   sam64"
-                              : /* no output */
-                              : "a" (cpcmd_buf), "d" (cmdlen),
-                                "a" (response), "d" (rlen), "m" (mask)
-                              : "cc", "2", "3", "4", "5" );
+                asm volatile (	"lrag	2,0(%2)\n"
+				"lgr	4,%3\n"
+				"o	4,%6\n"
+				"lrag	3,0(%4)\n"
+				"lgr	5,%5\n"
+				"sam31\n"
+				"diag	2,4,0x8\n"
+				"sam64\n"
+				"brc	8, .Litfits\n"
+				"agr	5, %5\n"
+				".Litfits: \n"
+				"lgr	%0,4\n"
+				"lgr	%1,5\n"
+				: "=d" (return_code), "=d" (return_len)
+				: "a" (cpcmd_buf), "d" (cmdlen),
+				"a" (response), "d" (rlen), "m" (mask)
+				: "cc", "2", "3", "4", "5" );
 #endif /* CONFIG_ARCH_S390X */
                 EBCASC(response, rlen);
         } else {
+		return_len = 0;
 #ifndef CONFIG_ARCH_S390X
-                asm volatile ("LRA   2,0(%0)\n\t"
-                              "LR    3,%1\n\t"
-                              ".long 0x83230008 # Diagnose X'08'\n\t"
-                              : /* no output */
-                              : "a" (cpcmd_buf), "d" (cmdlen)
-                              : "2", "3"  );
+                asm volatile (	"lra	2,0(%1)\n"
+				"lr	3,%2\n"
+				"diag	2,3,0x8\n"
+				"lr	%0,3\n"
+				: "=d" (return_code)
+				: "a" (cpcmd_buf), "d" (cmdlen)
+				: "2", "3"  );
 #else /* CONFIG_ARCH_S390X */
-                asm volatile ("   lrag  2,0(%0)\n"
-                              "   lgr   3,%1\n"
-                              "   sam31\n"
-                              "   .long 0x83230008 # Diagnose X'08'\n"
-                              "   sam64"
-                              : /* no output */
-                              : "a" (cpcmd_buf), "d" (cmdlen)
-                              : "2", "3"  );
+                asm volatile (	"lrag	2,0(%1)\n"
+				"lgr	3,%2\n"
+				"sam31\n"
+				"diag	2,3,0x8\n"
+				"sam64\n"
+				"lgr	%0,3\n"
+				: "=d" (return_code)
+				: "a" (cpcmd_buf), "d" (cmdlen)
+				: "2", "3" );
 #endif /* CONFIG_ARCH_S390X */
         }
 	spin_unlock_irqrestore(&cpcmd_lock, flags);
+	if (response_code != NULL)
+		*response_code = return_code;
+	return return_len;
 }
 
 EXPORT_SYMBOL(__cpcmd);
 
 #ifdef CONFIG_ARCH_S390X
-void cpcmd(char *cmd, char *response, int rlen)
+int cpcmd(const char *cmd, char *response, int rlen, int *response_code)
 {
 	char *lowbuf;
+	int len;
+
 	if ((rlen == 0) || (response == NULL)
 	    || !((unsigned long)response >> 31))
-		__cpcmd(cmd, response, rlen);
+		len = __cpcmd(cmd, response, rlen, response_code);
 	else {
 		lowbuf = kmalloc(rlen, GFP_KERNEL | GFP_DMA);
 		if (!lowbuf) {
 			printk(KERN_WARNING
 				"cpcmd: could not allocate response buffer\n");
-			return;
+			return -ENOMEM;
 		}
-		__cpcmd(cmd, lowbuf, rlen);
+		len = __cpcmd(cmd, lowbuf, rlen, response_code);
 		memcpy(response, lowbuf, rlen);
 		kfree(lowbuf);
 	}
+	return len;
 }
 
 EXPORT_SYMBOL(cpcmd);
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index eb7be0ad717..b6d740ac0e6 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -198,11 +198,11 @@ static void __init conmode_default(void)
 	char *ptr;
 
         if (MACHINE_IS_VM) {
-		__cpcmd("QUERY CONSOLE", query_buffer, 1024);
+		__cpcmd("QUERY CONSOLE", query_buffer, 1024, NULL);
 		console_devno = simple_strtoul(query_buffer + 5, NULL, 16);
 		ptr = strstr(query_buffer, "SUBCHANNEL =");
 		console_irq = simple_strtoul(ptr + 13, NULL, 16);
-		__cpcmd("QUERY TERM", query_buffer, 1024);
+		__cpcmd("QUERY TERM", query_buffer, 1024, NULL);
 		ptr = strstr(query_buffer, "CONMODE");
 		/*
 		 * Set the conmode to 3215 so that the device recognition 
@@ -211,7 +211,7 @@ static void __init conmode_default(void)
 		 * 3215 and the 3270 driver will try to access the console
 		 * device (3215 as console and 3270 as normal tty).
 		 */
-		__cpcmd("TERM CONMODE 3215", NULL, 0);
+		__cpcmd("TERM CONMODE 3215", NULL, 0, NULL);
 		if (ptr == NULL) {
 #if defined(CONFIG_SCLP_CONSOLE)
 			SET_CONSOLE_SCLP;
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 50c335067cf..642572a8e33 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -284,7 +284,7 @@ static void do_machine_restart(void * __unused)
 	 * locks are always held disabled).
 	 */
 	if (MACHINE_IS_VM)
-		cpcmd ("IPL", NULL, 0);
+		cpcmd ("IPL", NULL, 0, NULL);
 	else
 		reipl (0x10000 | S390_lowcore.ipl_device);
 }
@@ -313,7 +313,7 @@ static void do_machine_halt(void * __unused)
 	if (atomic_compare_and_swap(-1, smp_processor_id(), &cpuid) == 0) {
 		smp_send_stop();
 		if (MACHINE_IS_VM && strlen(vmhalt_cmd) > 0)
-			cpcmd(vmhalt_cmd, NULL, 0);
+			cpcmd(vmhalt_cmd, NULL, 0, NULL);
 		signal_processor(smp_processor_id(),
 				 sigp_stop_and_store_status);
 	}
@@ -332,7 +332,7 @@ static void do_machine_power_off(void * __unused)
 	if (atomic_compare_and_swap(-1, smp_processor_id(), &cpuid) == 0) {
 		smp_send_stop();
 		if (MACHINE_IS_VM && strlen(vmpoff_cmd) > 0)
-			cpcmd(vmpoff_cmd, NULL, 0);
+			cpcmd(vmpoff_cmd, NULL, 0, NULL);
 		signal_processor(smp_processor_id(),
 				 sigp_stop_and_store_status);
 	}
diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
index ca34b6f34b3..bc7b7be7acb 100644
--- a/arch/s390/kernel/traps.c
+++ b/arch/s390/kernel/traps.c
@@ -735,7 +735,7 @@ void __init trap_init(void)
 						    &ext_int_pfault);
 #endif
 #ifndef CONFIG_ARCH_S390X
-		cpcmd("SET PAGEX ON", NULL, 0);
+		cpcmd("SET PAGEX ON", NULL, 0, NULL);
 #endif
 	}
 }
diff --git a/arch/s390/mm/extmem.c b/arch/s390/mm/extmem.c
index 648deed17e2..c5348108ca3 100644
--- a/arch/s390/mm/extmem.c
+++ b/arch/s390/mm/extmem.c
@@ -576,8 +576,8 @@ segment_save(char *name)
 			segtype_string[seg->range[i].start & 0xff]);
 	}
 	sprintf(cmd2, "SAVESEG %s", name);
-	cpcmd(cmd1, NULL, 0);
-	cpcmd(cmd2, NULL, 0);
+	cpcmd(cmd1, NULL, 0, NULL);
+	cpcmd(cmd2, NULL, 0, NULL);
 	spin_unlock(&dcss_lock);
 }
 
diff --git a/drivers/s390/Kconfig b/drivers/s390/Kconfig
index 96413c2cd1a..a86a650f3d6 100644
--- a/drivers/s390/Kconfig
+++ b/drivers/s390/Kconfig
@@ -187,6 +187,13 @@ config VMLOGRDR
 	  *SYMPTOM.
 	  This driver depends on the IUCV support driver.
 
+config VMCP
+	tristate "Support for the z/VM CP interface (VM only)"
+	help
+	  Select this option if you want to be able to interact with the control
+	  program on z/VM
+
+
 config MONREADER
 	tristate "API for reading z/VM monitor service records"
 	depends on IUCV
diff --git a/drivers/s390/char/Makefile b/drivers/s390/char/Makefile
index 14e8cce9f86..6377a96735d 100644
--- a/drivers/s390/char/Makefile
+++ b/drivers/s390/char/Makefile
@@ -19,6 +19,7 @@ obj-$(CONFIG_SCLP_CPI) += sclp_cpi.o
 
 obj-$(CONFIG_ZVM_WATCHDOG) += vmwatchdog.o
 obj-$(CONFIG_VMLOGRDR) += vmlogrdr.o
+obj-$(CONFIG_VMCP) += vmcp.o
 
 tape-$(CONFIG_S390_TAPE_BLOCK) += tape_block.o
 tape-$(CONFIG_PROC_FS) += tape_proc.o
diff --git a/drivers/s390/char/con3215.c b/drivers/s390/char/con3215.c
index 022f17bff73..f11a67fda40 100644
--- a/drivers/s390/char/con3215.c
+++ b/drivers/s390/char/con3215.c
@@ -860,8 +860,8 @@ con3215_init(void)
 
 	/* Set the console mode for VM */
 	if (MACHINE_IS_VM) {
-		cpcmd("TERM CONMODE 3215", NULL, 0);
-		cpcmd("TERM AUTOCR OFF", NULL, 0);
+		cpcmd("TERM CONMODE 3215", NULL, 0, NULL);
+		cpcmd("TERM AUTOCR OFF", NULL, 0, NULL);
 	}
 
 	/* allocate 3215 request structures */
diff --git a/drivers/s390/char/con3270.c b/drivers/s390/char/con3270.c
index d52fb57a6b1..fc7a213e591 100644
--- a/drivers/s390/char/con3270.c
+++ b/drivers/s390/char/con3270.c
@@ -591,8 +591,8 @@ con3270_init(void)
 
 	/* Set the console mode for VM */
 	if (MACHINE_IS_VM) {
-		cpcmd("TERM CONMODE 3270", 0, 0);
-		cpcmd("TERM AUTOCR OFF", 0, 0);
+		cpcmd("TERM CONMODE 3270", NULL, 0, NULL);
+		cpcmd("TERM AUTOCR OFF", NULL, 0, NULL);
 	}
 
 	cdev = ccw_device_probe_console();
diff --git a/drivers/s390/char/vmcp.c b/drivers/s390/char/vmcp.c
new file mode 100644
index 00000000000..dfbbf235ca2
--- /dev/null
+++ b/drivers/s390/char/vmcp.c
@@ -0,0 +1,219 @@
+/*
+ * Copyright (C) 2004,2005 IBM Corporation
+ * Interface implementation for communication with the v/VM control program
+ * Author(s): Christian Borntraeger <cborntra@de.ibm.com>
+ *
+ *
+ * z/VMs CP offers the possibility to issue commands via the diagnose code 8
+ * this driver implements a character device that issues these commands and
+ * returns the answer of CP.
+
+ * The idea of this driver is based on cpint from Neale Ferguson and #CP in CMS
+ */
+
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+#include <asm/cpcmd.h>
+#include <asm/debug.h>
+#include <asm/uaccess.h>
+#include "vmcp.h"
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Christian Borntraeger <cborntra@de.ibm.com>");
+MODULE_DESCRIPTION("z/VM CP interface");
+
+static debug_info_t *vmcp_debug;
+
+static int vmcp_open(struct inode *inode, struct file *file)
+{
+	struct vmcp_session *session;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	session = kmalloc(sizeof(*session), GFP_KERNEL);
+	if (!session)
+		return -ENOMEM;
+	session->bufsize = PAGE_SIZE;
+	session->response = NULL;
+	session->resp_size = 0;
+	init_MUTEX(&session->mutex);
+	file->private_data = session;
+	return nonseekable_open(inode, file);
+}
+
+static int vmcp_release(struct inode *inode, struct file *file)
+{
+	struct vmcp_session *session;
+
+	session = (struct vmcp_session *)file->private_data;
+	file->private_data = NULL;
+	free_pages((unsigned long)session->response, get_order(session->bufsize));
+	kfree(session);
+	return 0;
+}
+
+static ssize_t
+vmcp_read(struct file *file, char __user * buff, size_t count, loff_t * ppos)
+{
+	size_t tocopy;
+	struct vmcp_session *session;
+
+	session = (struct vmcp_session *)file->private_data;
+	if (down_interruptible(&session->mutex))
+		return -ERESTARTSYS;
+	if (!session->response) {
+		up(&session->mutex);
+		return 0;
+	}
+	if (*ppos > session->resp_size) {
+		up(&session->mutex);
+		return 0;
+	}
+	tocopy = min(session->resp_size - (size_t) (*ppos), count);
+	tocopy = min(tocopy,session->bufsize - (size_t) (*ppos));
+
+	if (copy_to_user(buff, session->response + (*ppos), tocopy)) {
+		up(&session->mutex);
+		return -EFAULT;
+	}
+	up(&session->mutex);
+	*ppos += tocopy;
+	return tocopy;
+}
+
+static ssize_t
+vmcp_write(struct file *file, const char __user * buff, size_t count,
+	   loff_t * ppos)
+{
+	char *cmd;
+	struct vmcp_session *session;
+
+	if (count > 240)
+		return -EINVAL;
+	cmd = kmalloc(count + 1, GFP_KERNEL);
+	if (!cmd)
+		return -ENOMEM;
+	if (copy_from_user(cmd, buff, count)) {
+		kfree(cmd);
+		return -EFAULT;
+	}
+	cmd[count] = '\0';
+	session = (struct vmcp_session *)file->private_data;
+	if (down_interruptible(&session->mutex))
+		return -ERESTARTSYS;
+	if (!session->response)
+		session->response = (char *)__get_free_pages(GFP_KERNEL
+						| __GFP_REPEAT 	| GFP_DMA,
+						get_order(session->bufsize));
+	if (!session->response) {
+		up(&session->mutex);
+		kfree(cmd);
+		return -ENOMEM;
+	}
+	debug_text_event(vmcp_debug, 1, cmd);
+	session->resp_size = cpcmd(cmd, session->response,
+				   session->bufsize,
+				   &session->resp_code);
+	up(&session->mutex);
+	kfree(cmd);
+	*ppos = 0;		/* reset the file pointer after a command */
+	return count;
+}
+
+
+/*
+ * These ioctls are available, as the semantics of the diagnose 8 call
+ * does not fit very well into a Linux call. Diagnose X'08' is described in
+ * CP Programming Services SC24-6084-00
+ *
+ * VMCP_GETCODE: gives the CP return code back to user space
+ * VMCP_SETBUF: sets the response buffer for the next write call. diagnose 8
+ * expects adjacent pages in real storage and to make matters worse, we
+ * dont know the size of the response. Therefore we default to PAGESIZE and
+ * let userspace to change the response size, if userspace expects a bigger
+ * response
+ */
+static long vmcp_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	struct vmcp_session *session;
+	int temp;
+
+	session = (struct vmcp_session *)file->private_data;
+	if (down_interruptible(&session->mutex))
+		return -ERESTARTSYS;
+	switch (cmd) {
+	case VMCP_GETCODE:
+		temp = session->resp_code;
+		up(&session->mutex);
+		return put_user(temp, (int __user *)arg);
+	case VMCP_SETBUF:
+		free_pages((unsigned long)session->response,
+				get_order(session->bufsize));
+		session->response=NULL;
+		temp = get_user(session->bufsize, (int __user *)arg);
+		if (get_order(session->bufsize) > 8) {
+			session->bufsize = PAGE_SIZE;
+			temp = -EINVAL;
+		}
+		up(&session->mutex);
+		return temp;
+	case VMCP_GETSIZE:
+		temp = session->resp_size;
+		up(&session->mutex);
+		return put_user(temp, (int __user *)arg);
+	default:
+		up(&session->mutex);
+		return -ENOIOCTLCMD;
+	}
+}
+
+static struct file_operations vmcp_fops = {
+	.owner		= THIS_MODULE,
+	.open		= &vmcp_open,
+	.release	= &vmcp_release,
+	.read		= &vmcp_read,
+	.llseek		= &no_llseek,
+	.write		= &vmcp_write,
+	.unlocked_ioctl	= &vmcp_ioctl,
+	.compat_ioctl	= &vmcp_ioctl
+};
+
+static struct miscdevice vmcp_dev = {
+	.name	= "vmcp",
+	.minor	= MISC_DYNAMIC_MINOR,
+	.fops	= &vmcp_fops,
+};
+
+static int __init vmcp_init(void)
+{
+	int ret;
+
+	if (!MACHINE_IS_VM) {
+		printk(KERN_WARNING
+		       "z/VM CP interface is only available under z/VM\n");
+		return -ENODEV;
+	}
+	ret = misc_register(&vmcp_dev);
+	if (!ret)
+		printk(KERN_INFO "z/VM CP interface loaded\n");
+	else
+		printk(KERN_WARNING
+		       "z/VM CP interface not loaded. Could not register misc device.\n");
+	vmcp_debug = debug_register("vmcp", 0, 1, 240);
+	debug_register_view(vmcp_debug, &debug_hex_ascii_view);
+	return ret;
+}
+
+static void __exit vmcp_exit(void)
+{
+	WARN_ON(misc_deregister(&vmcp_dev) != 0);
+	debug_unregister(vmcp_debug);
+	printk(KERN_INFO "z/VM CP interface unloaded.\n");
+}
+
+module_init(vmcp_init);
+module_exit(vmcp_exit);
diff --git a/drivers/s390/char/vmcp.h b/drivers/s390/char/vmcp.h
new file mode 100644
index 00000000000..87389e73046
--- /dev/null
+++ b/drivers/s390/char/vmcp.h
@@ -0,0 +1,30 @@
+/*
+ * Copyright (C) 2004, 2005 IBM Corporation
+ * Interface implementation for communication with the v/VM control program
+ * Version 1.0
+ * Author(s): Christian Borntraeger <cborntra@de.ibm.com>
+ *
+ *
+ * z/VMs CP offers the possibility to issue commands via the diagnose code 8
+ * this driver implements a character device that issues these commands and
+ * returns the answer of CP.
+ *
+ * The idea of this driver is based on cpint from Neale Ferguson
+ */
+
+#include <asm/semaphore.h>
+#include <linux/ioctl.h>
+
+#define VMCP_GETCODE _IOR(0x10, 1, int)
+#define VMCP_SETBUF _IOW(0x10, 2, int)
+#define VMCP_GETSIZE _IOR(0x10, 3, int)
+
+struct vmcp_session {
+	unsigned int bufsize;
+	char *response;
+	int resp_size;
+	int resp_code;
+	/* As we use copy_from/to_user, which might     *
+	 * sleep and cannot use a spinlock              */
+	struct semaphore mutex;
+};
diff --git a/drivers/s390/char/vmlogrdr.c b/drivers/s390/char/vmlogrdr.c
index f7717327d15..491f00c032e 100644
--- a/drivers/s390/char/vmlogrdr.c
+++ b/drivers/s390/char/vmlogrdr.c
@@ -236,7 +236,7 @@ vmlogrdr_get_recording_class_AB(void) {
 	int len,i;
 
 	printk (KERN_DEBUG "vmlogrdr: query command: %s\n", cp_command);
-	cpcmd(cp_command, cp_response, sizeof(cp_response));
+	cpcmd(cp_command, cp_response, sizeof(cp_response), NULL);
 	printk (KERN_DEBUG "vmlogrdr: response: %s", cp_response);
 	len = strnlen(cp_response,sizeof(cp_response));
 	// now the parsing
@@ -288,7 +288,7 @@ vmlogrdr_recording(struct vmlogrdr_priv_t * logptr, int action, int purge) {
 
 		printk (KERN_DEBUG "vmlogrdr: recording command: %s\n",
 			cp_command);
-		cpcmd(cp_command, cp_response, sizeof(cp_response));
+		cpcmd(cp_command, cp_response, sizeof(cp_response), NULL);
 		printk (KERN_DEBUG "vmlogrdr: recording response: %s",
 			cp_response);
 	}
@@ -301,7 +301,7 @@ vmlogrdr_recording(struct vmlogrdr_priv_t * logptr, int action, int purge) {
 		qid_string);
 
 	printk (KERN_DEBUG "vmlogrdr: recording command: %s\n", cp_command);
-	cpcmd(cp_command, cp_response, sizeof(cp_response));
+	cpcmd(cp_command, cp_response, sizeof(cp_response), NULL);
 	printk (KERN_DEBUG "vmlogrdr: recording response: %s",
 		cp_response);
 	/* The recording command will usually answer with 'Command complete'
@@ -607,7 +607,7 @@ vmlogrdr_purge_store(struct device * dev, struct device_attribute *attr, const c
 			 priv->recording_name);
 
 	printk (KERN_DEBUG "vmlogrdr: recording command: %s\n", cp_command);
-	cpcmd(cp_command, cp_response, sizeof(cp_response));
+	cpcmd(cp_command, cp_response, sizeof(cp_response), NULL);
 	printk (KERN_DEBUG "vmlogrdr: recording response: %s",
 		cp_response);
 
@@ -682,7 +682,7 @@ vmlogrdr_recording_status_show(struct device_driver *driver, char *buf) {
 	char cp_command[] = "QUERY RECORDING ";
 	int len;
 
-	cpcmd(cp_command, buf, 4096);
+	cpcmd(cp_command, buf, 4096, NULL);
 	len = strlen(buf);
 	return len;
 }
diff --git a/drivers/s390/net/smsgiucv.c b/drivers/s390/net/smsgiucv.c
index 1e3f7f3c662..d6469baa7e1 100644
--- a/drivers/s390/net/smsgiucv.c
+++ b/drivers/s390/net/smsgiucv.c
@@ -138,7 +138,7 @@ static void __exit
 smsg_exit(void)
 {
 	if (smsg_handle > 0) {
-		cpcmd("SET SMSG OFF", 0, 0);
+		cpcmd("SET SMSG OFF", NULL, 0, NULL);
 		iucv_sever(smsg_pathid, 0);
 		iucv_unregister_program(smsg_handle);
 		driver_unregister(&smsg_driver);
@@ -177,7 +177,7 @@ smsg_init(void)
 		smsg_handle = 0;
 		return -EIO;
 	}
-	cpcmd("SET SMSG IUCV", 0, 0);
+	cpcmd("SET SMSG IUCV", NULL, 0, NULL);
 	return 0;
 }
 
diff --git a/include/asm-s390/cpcmd.h b/include/asm-s390/cpcmd.h
index 1d33c5da083..1fcf65be7a2 100644
--- a/include/asm-s390/cpcmd.h
+++ b/include/asm-s390/cpcmd.h
@@ -11,14 +11,28 @@
 #define __CPCMD__
 
 /*
+ * the lowlevel function for cpcmd
  * the caller of __cpcmd has to ensure that the response buffer is below 2 GB
  */
-extern void __cpcmd(char *cmd, char *response, int rlen);
+extern int __cpcmd(const char *cmd, char *response, int rlen, int *response_code);
 
 #ifndef __s390x__
 #define cpcmd __cpcmd
 #else
-extern void cpcmd(char *cmd, char *response, int rlen);
+/*
+ * cpcmd is the in-kernel interface for issuing CP commands
+ *
+ * cmd:		null-terminated command string, max 240 characters
+ * response:	response buffer for VM's textual response
+ * rlen:	size of the response buffer, cpcmd will not exceed this size
+ *		but will cap the output, if its too large. Everything that
+ *		did not fit into the buffer will be silently dropped
+ * response_code: return pointer for VM's error code
+ * return value: the size of the response. The caller can check if the buffer
+ *		was large enough by comparing the return value and rlen
+ * NOTE: If the response buffer is not below 2 GB, cpcmd can sleep
+ */
+extern int cpcmd(const char *cmd, char *response, int rlen, int *response_code);
 #endif /*__s390x__*/
 
 #endif
-- 
cgit v1.2.3-70-g09d2


From 66a464dbc8e0345b6f972b92bf1118e043d7c987 Mon Sep 17 00:00:00 2001
From: Michael Holzheu <holzheu@de.ibm.com>
Date: Sat, 25 Jun 2005 14:55:33 -0700
Subject: [PATCH] s390: debug feature changes

This patch changes the memory allocation method for the s390 debug feature.
Trace buffers had been allocated using the get_free_pages() function before.
Therefore it was not possible to get big memory areas in a running system due
to memory fragmentation.  Now the trace buffers are subdivided into several
subbuffers with pagesize.  Therefore it is now possible to allocate more
memory for the trace buffers and more trace records can be written.

In addition to that, dynamic specification of the size of the trace buffers is
implemented.  It is now possible to change the size of a trace buffer using a
new debugfs file instance.  When writing a number into this file, the trace
buffer size is changed to 'number * pagesize'.

In the past all the traces could be obtained from userspace by accessing files
in the "proc" filesystem.  Now with debugfs we have a new filesystem which
should be used for debugging purposes.  This patch moves the debug feature
from procfs to debugfs.

Since the interface of debug_register() changed, all device drivers, which use
the debug feature had to be adjusted.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/s390/s390dbf.txt |  79 ++--
 arch/s390/kernel/debug.c       | 820 +++++++++++++++++++++++++++--------------
 drivers/s390/block/dasd.c      |   4 +-
 drivers/s390/block/dasd_proc.c |   3 +-
 drivers/s390/char/tape_34xx.c  |   6 +-
 drivers/s390/char/tape_core.c  |   2 +-
 drivers/s390/char/tape_proc.c  |   1 +
 drivers/s390/char/vmcp.c       |   2 +-
 drivers/s390/cio/cio.c         |   8 +-
 drivers/s390/cio/qdio.c        |  14 +-
 drivers/s390/cio/qdio.h        |  16 +-
 drivers/s390/net/claw.c        |   4 +-
 drivers/s390/net/ctcdbug.c     |  10 +-
 drivers/s390/net/ctcdbug.h     |  10 +-
 drivers/s390/net/iucv.h        |   6 +-
 drivers/s390/net/lcs.c         |   8 +-
 drivers/s390/net/netiucv.c     |  12 +-
 drivers/s390/net/qeth.h        |  14 +-
 drivers/s390/net/qeth_main.c   |  14 +-
 include/asm-s390/debug.h       |  48 ++-
 20 files changed, 678 insertions(+), 403 deletions(-)

(limited to 'include')

diff --git a/Documentation/s390/s390dbf.txt b/Documentation/s390/s390dbf.txt
index 2d1cd939b4d..e24fdeada97 100644
--- a/Documentation/s390/s390dbf.txt
+++ b/Documentation/s390/s390dbf.txt
@@ -12,8 +12,8 @@ where log records can be stored efficiently in memory, where each component
 One purpose of this is to inspect the debug logs after a production system crash
 in order to analyze the reason for the crash.
 If the system still runs but only a subcomponent which uses dbf failes,
-it is possible to look at the debug logs on a live system via the Linux proc
-filesystem.
+it is possible to look at the debug logs on a live system via the Linux
+debugfs filesystem.
 The debug feature may also very useful for kernel and driver development.
 
 Design:
@@ -52,16 +52,18 @@ Each debug entry contains the following data:
 - Flag, if entry is an exception or not
 
 The debug logs can be inspected in a live system through entries in
-the proc-filesystem. Under the path /proc/s390dbf there is 
+the debugfs-filesystem. Under the toplevel directory "s390dbf" there is
 a directory for each registered component, which is named like the
-corresponding component.
+corresponding component. The debugfs normally should be mounted to
+/sys/kernel/debug therefore the debug feature can be accessed unter
+/sys/kernel/debug/s390dbf.
 
 The content of the directories are files which represent different views
 to the debug log. Each component can decide which views should be
 used through registering them with the function debug_register_view().
 Predefined views for hex/ascii, sprintf and raw binary data are provided.
 It is also possible to define other views. The content of
-a view can be inspected simply by reading the corresponding proc file.
+a view can be inspected simply by reading the corresponding debugfs file.
 
 All debug logs have an an actual debug level (range from 0 to 6).
 The default level is 3. Event and Exception functions have a 'level'
@@ -69,14 +71,14 @@ parameter. Only debug entries with a level that is lower or equal
 than the actual level are written to the log. This means, when
 writing events, high priority log entries should have a low level
 value whereas low priority entries should have a high one.
-The actual debug level can be changed with the help of the proc-filesystem 
-through writing a number string "x" to the 'level' proc file which is
+The actual debug level can be changed with the help of the debugfs-filesystem
+through writing a number string "x" to the 'level' debugfs file which is
 provided for every debug log. Debugging can be switched off completely
-by using "-" on the 'level' proc file.
+by using "-" on the 'level' debugfs file.
 
 Example:
 
-> echo "-" > /proc/s390dbf/dasd/level
+> echo "-" > /sys/kernel/debug/s390dbf/dasd/level
 
 It is also possible to deactivate the debug feature globally for every
 debug log. You can change the behavior using  2 sysctl parameters in
@@ -99,11 +101,11 @@ Kernel Interfaces:
 ------------------
 
 ----------------------------------------------------------------------------
-debug_info_t *debug_register(char *name, int pages_index, int nr_areas,
+debug_info_t *debug_register(char *name, int pages, int nr_areas,
                              int buf_size);
 
-Parameter:    name:        Name of debug log (e.g. used for proc entry) 
-              pages_index: 2^pages_index pages will be allocated per area
+Parameter:    name:        Name of debug log (e.g. used for debugfs entry)
+              pages:       number of pages, which will be allocated per area
               nr_areas:    number of debug areas
               buf_size:    size of data area in each debug entry
 
@@ -134,7 +136,7 @@ Return Value:  none
 Description:   Sets new actual debug level if new_level is valid. 
 
 ---------------------------------------------------------------------------
-+void debug_stop_all(void);
+void debug_stop_all(void);
 
 Parameter:     none
 
@@ -270,7 +272,7 @@ Parameter:     id:    handle for debug log
 Return Value:  0  : ok 
                < 0: Error 
 
-Description:   registers new debug view and creates proc dir entry 
+Description:   registers new debug view and creates debugfs dir entry
 
 ---------------------------------------------------------------------------
 int debug_unregister_view (debug_info_t * id, struct debug_view *view); 
@@ -281,7 +283,7 @@ Parameter:     id:    handle for debug log
 Return Value:  0  : ok 
                < 0: Error 
 
-Description:   unregisters debug view and removes proc dir entry 
+Description:   unregisters debug view and removes debugfs dir entry
 
 
 
@@ -308,7 +310,7 @@ static int init(void)
 {
     /* register 4 debug areas with one page each and 4 byte data field */
 
-    debug_info = debug_register ("test", 0, 4, 4 );
+    debug_info = debug_register ("test", 1, 4, 4 );
     debug_register_view(debug_info,&debug_hex_ascii_view);
     debug_register_view(debug_info,&debug_raw_view);
 
@@ -343,7 +345,7 @@ static int init(void)
     /* register 4 debug areas with one page each and data field for */
     /* format string pointer + 2 varargs (= 3 * sizeof(long))       */
 
-    debug_info = debug_register ("test", 0, 4, sizeof(long) * 3);
+    debug_info = debug_register ("test", 1, 4, sizeof(long) * 3);
     debug_register_view(debug_info,&debug_sprintf_view);
 
     debug_sprintf_event(debug_info, 2 , "first event in %s:%i\n",__FILE__,__LINE__);
@@ -362,16 +364,16 @@ module_exit(cleanup);
 
 
 
-ProcFS Interface
+Debugfs Interface
 ----------------
 Views to the debug logs can be investigated through reading the corresponding 
-proc-files:
+debugfs-files:
 
 Example:
 
-> ls /proc/s390dbf/dasd
-flush  hex_ascii  level      raw 
-> cat /proc/s390dbf/dasd/hex_ascii | sort +1
+> ls /sys/kernel/debug/s390dbf/dasd
+flush  hex_ascii  level pages raw
+> cat /sys/kernel/debug/s390dbf/dasd/hex_ascii | sort +1
 00 00974733272:680099 2 - 02 0006ad7e  07 ea 4a 90 | ....
 00 00974733272:682210 2 - 02 0006ade6  46 52 45 45 | FREE
 00 00974733272:682213 2 - 02 0006adf6  07 ea 4a 90 | ....
@@ -391,25 +393,36 @@ Changing the debug level
 Example:
 
 
-> cat /proc/s390dbf/dasd/level
+> cat /sys/kernel/debug/s390dbf/dasd/level
 3
-> echo "5" > /proc/s390dbf/dasd/level
-> cat /proc/s390dbf/dasd/level
+> echo "5" > /sys/kernel/debug/s390dbf/dasd/level
+> cat /sys/kernel/debug/s390dbf/dasd/level
 5
 
 Flushing debug areas
 --------------------
 Debug areas can be flushed with piping the number of the desired
-area (0...n) to the proc file "flush". When using "-" all debug areas
+area (0...n) to the debugfs file "flush". When using "-" all debug areas
 are flushed.
 
 Examples:
 
 1. Flush debug area 0:
-> echo "0" > /proc/s390dbf/dasd/flush  
+> echo "0" > /sys/kernel/debug/s390dbf/dasd/flush
 
 2. Flush all debug areas:
-> echo "-" > /proc/s390dbf/dasd/flush
+> echo "-" > /sys/kernel/debug/s390dbf/dasd/flush
+
+Changing the size of debug areas
+------------------------------------
+It is possible the change the size of debug areas through piping
+the number of pages to the debugfs file "pages". The resize request will
+also flush the debug areas.
+
+Example:
+
+Define 4 pages for the debug areas of debug feature "dasd":
+> echo "4" > /sys/kernel/debug/s390dbf/dasd/pages
 
 Stooping the debug feature
 --------------------------
@@ -491,7 +504,7 @@ Defining views
 --------------
 
 Views are specified with the 'debug_view' structure. There are defined
-callback functions which are used for reading and writing the proc files:
+callback functions which are used for reading and writing the debugfs files:
 
 struct debug_view {
         char name[DEBUG_MAX_PROCF_LEN];  
@@ -525,7 +538,7 @@ typedef int (debug_input_proc_t) (debug_info_t* id,
 The "private_data" member can be used as pointer to view specific data.
 It is not used by the debug feature itself.
 
-The output when reading a debug-proc file is structured like this:
+The output when reading a debugfs file is structured like this:
 
 "prolog_proc output"
 
@@ -534,13 +547,13 @@ The output when reading a debug-proc file is structured like this:
 "header_proc output 3"  "format_proc output 3"
 ...
 
-When a view is read from the proc fs, the Debug Feature calls the 
+When a view is read from the debugfs, the Debug Feature calls the
 'prolog_proc' once for writing the prolog.
 Then 'header_proc' and 'format_proc' are called for each 
 existing debug entry.
 
 The input_proc can be used to implement functionality when it is written to 
-the view (e.g. like with 'echo "0" > /proc/s390dbf/dasd/level).
+the view (e.g. like with 'echo "0" > /sys/kernel/debug/s390dbf/dasd/level).
 
 For header_proc there can be used the default function
 debug_dflt_header_fn() which is defined in in debug.h.
@@ -602,7 +615,7 @@ debug_info = debug_register ("test", 0, 4, 4 ));
 debug_register_view(debug_info, &debug_test_view);
 for(i = 0; i < 10; i ++) debug_int_event(debug_info, 1, i);
 
-> cat /proc/s390dbf/test/myview
+> cat /sys/kernel/debug/s390dbf/test/myview
 00 00964419734:611402 1 - 00 88042ca   This error...........
 00 00964419734:611405 1 - 00 88042ca   That error...........
 00 00964419734:611408 1 - 00 88042ca   Problem..............
diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c
index 91f8ce5543d..960ba6029c3 100644
--- a/arch/s390/kernel/debug.c
+++ b/arch/s390/kernel/debug.c
@@ -19,22 +19,27 @@
 #include <linux/sysctl.h>
 #include <asm/uaccess.h>
 #include <asm/semaphore.h>
-
 #include <linux/module.h>
 #include <linux/init.h>
+#include <linux/fs.h>
+#include <linux/debugfs.h>
 
 #include <asm/debug.h>
 
 #define DEBUG_PROLOG_ENTRY -1
 
+#define ALL_AREAS 0 /* copy all debug areas */
+#define NO_AREAS  1 /* copy no debug areas */
+
 /* typedefs */
 
 typedef struct file_private_info {
 	loff_t offset;			/* offset of last read in file */
 	int    act_area;                /* number of last formated area */
+	int    act_page;                /* act page in given area */
 	int    act_entry;               /* last formated entry (offset */
                                         /* relative to beginning of last */
-                                        /* formated area) */ 
+                                        /* formated page) */
 	size_t act_entry_offset;        /* up to this offset we copied */
 					/* in last read the last formated */
 					/* entry to userland */
@@ -51,8 +56,8 @@ typedef struct
 	 * This assumes that all args are converted into longs 
 	 * on L/390 this is the case for all types of parameter 
 	 * except of floats, and long long (32 bit) 
-         *
-         */
+	 *
+	 */
 	long args[0];
 } debug_sprintf_entry_t;
 
@@ -63,32 +68,38 @@ extern void tod_to_timeval(uint64_t todval, struct timeval *xtime);
 
 static int debug_init(void);
 static ssize_t debug_output(struct file *file, char __user *user_buf,
-			    size_t user_len, loff_t * offset);
+			size_t user_len, loff_t * offset);
 static ssize_t debug_input(struct file *file, const char __user *user_buf,
-			   size_t user_len, loff_t * offset);
+			size_t user_len, loff_t * offset);
 static int debug_open(struct inode *inode, struct file *file);
 static int debug_close(struct inode *inode, struct file *file);
-static debug_info_t*  debug_info_create(char *name, int page_order, int nr_areas, int buf_size);
+static debug_info_t*  debug_info_create(char *name, int pages_per_area,
+			int nr_areas, int buf_size);
 static void debug_info_get(debug_info_t *);
 static void debug_info_put(debug_info_t *);
 static int debug_prolog_level_fn(debug_info_t * id,
-				 struct debug_view *view, char *out_buf);
+			struct debug_view *view, char *out_buf);
 static int debug_input_level_fn(debug_info_t * id, struct debug_view *view,
-				struct file *file, const char __user *user_buf,
-				size_t user_buf_size, loff_t * offset);
+			struct file *file, const char __user *user_buf,
+			size_t user_buf_size, loff_t * offset);
+static int debug_prolog_pages_fn(debug_info_t * id,
+			struct debug_view *view, char *out_buf);
+static int debug_input_pages_fn(debug_info_t * id, struct debug_view *view,
+			struct file *file, const char __user *user_buf,
+			size_t user_buf_size, loff_t * offset);
 static int debug_input_flush_fn(debug_info_t * id, struct debug_view *view,
-                                struct file *file, const char __user *user_buf,
-                                size_t user_buf_size, loff_t * offset);
+			struct file *file, const char __user *user_buf,
+			size_t user_buf_size, loff_t * offset);
 static int debug_hex_ascii_format_fn(debug_info_t * id, struct debug_view *view,
-                                char *out_buf, const char *in_buf);
+			char *out_buf, const char *in_buf);
 static int debug_raw_format_fn(debug_info_t * id,
-				 struct debug_view *view, char *out_buf,
-				 const char *in_buf);
+			struct debug_view *view, char *out_buf,
+			const char *in_buf);
 static int debug_raw_header_fn(debug_info_t * id, struct debug_view *view,
-                         int area, debug_entry_t * entry, char *out_buf);
+			int area, debug_entry_t * entry, char *out_buf);
 
 static int debug_sprintf_format_fn(debug_info_t * id, struct debug_view *view,
-				   char *out_buf, debug_sprintf_entry_t *curr_event);
+			char *out_buf, debug_sprintf_entry_t *curr_event);
 
 /* globals */
 
@@ -119,6 +130,15 @@ struct debug_view debug_level_view = {
 	NULL
 };
 
+struct debug_view debug_pages_view = {
+	"pages",
+	&debug_prolog_pages_fn,
+	NULL,
+	NULL,
+	&debug_input_pages_fn,
+	NULL
+};
+
 struct debug_view debug_flush_view = {
         "flush",
         NULL,
@@ -149,98 +169,161 @@ DECLARE_MUTEX(debug_lock);
 static int initialized;
 
 static struct file_operations debug_file_ops = {
-	.owner	 = THIS_MODULE,
+	.owner   = THIS_MODULE,
 	.read    = debug_output,
-	.write   = debug_input,	
+	.write   = debug_input,
 	.open    = debug_open,
 	.release = debug_close,
 };
 
-static struct proc_dir_entry *debug_proc_root_entry;
+static struct dentry *debug_debugfs_root_entry;
 
 /* functions */
 
+/*
+ * debug_areas_alloc
+ * - Debug areas are implemented as a threedimensonal array:
+ *   areas[areanumber][pagenumber][pageoffset]
+ */
+
+static debug_entry_t***
+debug_areas_alloc(int pages_per_area, int nr_areas)
+{
+	debug_entry_t*** areas;
+	int i,j;
+
+	areas = (debug_entry_t ***) kmalloc(nr_areas *
+					sizeof(debug_entry_t**),
+					GFP_KERNEL);
+	if (!areas)
+		goto fail_malloc_areas;
+	for (i = 0; i < nr_areas; i++) {
+		areas[i] = (debug_entry_t**) kmalloc(pages_per_area *
+				sizeof(debug_entry_t*),GFP_KERNEL);
+		if (!areas[i]) {
+			goto fail_malloc_areas2;
+		}
+		for(j = 0; j < pages_per_area; j++) {
+			areas[i][j] = (debug_entry_t*)kmalloc(PAGE_SIZE,
+						GFP_KERNEL);
+			if(!areas[i][j]) {
+				for(j--; j >=0 ; j--) {
+					kfree(areas[i][j]);
+				}
+				kfree(areas[i]);
+				goto fail_malloc_areas2;
+			} else {
+				memset(areas[i][j],0,PAGE_SIZE);
+			}
+		}
+	}
+	return areas;
+
+fail_malloc_areas2:
+	for(i--; i >= 0; i--){
+		for(j=0; j < pages_per_area;j++){
+			kfree(areas[i][j]);
+		}
+		kfree(areas[i]);
+	}
+	kfree(areas);
+fail_malloc_areas:
+	return NULL;
+
+}
+
+
 /*
  * debug_info_alloc
  * - alloc new debug-info
  */
 
-static debug_info_t*  debug_info_alloc(char *name, int page_order,
-                                        int nr_areas, int buf_size)
+static debug_info_t*
+debug_info_alloc(char *name, int pages_per_area, int nr_areas, int buf_size,
+		int level, int mode)
 {
 	debug_info_t* rc;
-	int i;
 
 	/* alloc everything */
 
-	rc = (debug_info_t*) kmalloc(sizeof(debug_info_t), GFP_ATOMIC);
+	rc = (debug_info_t*) kmalloc(sizeof(debug_info_t), GFP_KERNEL);
 	if(!rc)
 		goto fail_malloc_rc;
-	rc->active_entry = (int*)kmalloc(nr_areas * sizeof(int), GFP_ATOMIC);
-	if(!rc->active_entry)
-		goto fail_malloc_active_entry;
-	memset(rc->active_entry, 0, nr_areas * sizeof(int));
-	rc->areas = (debug_entry_t **) kmalloc(nr_areas *
-						sizeof(debug_entry_t *),
-						GFP_ATOMIC);
-	if (!rc->areas)
-		goto fail_malloc_areas;
-	for (i = 0; i < nr_areas; i++) {
-		rc->areas[i] = (debug_entry_t *) __get_free_pages(GFP_ATOMIC,
-								page_order);
-		if (!rc->areas[i]) {
-			for (i--; i >= 0; i--) {
-				free_pages((unsigned long) rc->areas[i],
-						page_order);
-			}
-			goto fail_malloc_areas2;
-		} else {
-			memset(rc->areas[i], 0, PAGE_SIZE << page_order);
-		}
+	rc->active_entries = (int*)kmalloc(nr_areas * sizeof(int), GFP_KERNEL);
+	if(!rc->active_entries)
+		goto fail_malloc_active_entries;
+	memset(rc->active_entries, 0, nr_areas * sizeof(int));
+	rc->active_pages = (int*)kmalloc(nr_areas * sizeof(int), GFP_KERNEL);
+	if(!rc->active_pages)
+		goto fail_malloc_active_pages;
+	memset(rc->active_pages, 0, nr_areas * sizeof(int));
+	if((mode == ALL_AREAS) && (pages_per_area != 0)){
+		rc->areas = debug_areas_alloc(pages_per_area, nr_areas);
+		if(!rc->areas)
+			goto fail_malloc_areas;
+	} else {
+		rc->areas = NULL;
 	}
 
 	/* initialize members */
 
 	spin_lock_init(&rc->lock);
-	rc->page_order  = page_order;
-	rc->nr_areas    = nr_areas;
-	rc->active_area = 0;
-	rc->level       = DEBUG_DEFAULT_LEVEL;
-	rc->buf_size    = buf_size;
-	rc->entry_size  = sizeof(debug_entry_t) + buf_size;
-	strlcpy(rc->name, name, sizeof(rc->name));
+	rc->pages_per_area = pages_per_area;
+	rc->nr_areas       = nr_areas;
+	rc->active_area    = 0;
+	rc->level          = level;
+	rc->buf_size       = buf_size;
+	rc->entry_size     = sizeof(debug_entry_t) + buf_size;
+	strlcpy(rc->name, name, sizeof(rc->name)-1);
 	memset(rc->views, 0, DEBUG_MAX_VIEWS * sizeof(struct debug_view *));
-#ifdef CONFIG_PROC_FS
-	memset(rc->proc_entries, 0 ,DEBUG_MAX_VIEWS *
-		sizeof(struct proc_dir_entry*));
-#endif /* CONFIG_PROC_FS */
+	memset(rc->debugfs_entries, 0 ,DEBUG_MAX_VIEWS *
+		sizeof(struct dentry*));
 	atomic_set(&(rc->ref_count), 0);
 
 	return rc;
 
-fail_malloc_areas2:
-	kfree(rc->areas);
 fail_malloc_areas:
-	kfree(rc->active_entry);
-fail_malloc_active_entry:
+	kfree(rc->active_pages);
+fail_malloc_active_pages:
+	kfree(rc->active_entries);
+fail_malloc_active_entries:
 	kfree(rc);
 fail_malloc_rc:
 	return NULL;
 }
 
 /*
- * debug_info_free
- * - free memory debug-info
+ * debug_areas_free
+ * - free all debug areas
  */
 
-static void debug_info_free(debug_info_t* db_info){
-	int i;
+static void
+debug_areas_free(debug_info_t* db_info)
+{
+	int i,j;
+
+	if(!db_info->areas)
+		return;
 	for (i = 0; i < db_info->nr_areas; i++) {
-		free_pages((unsigned long) db_info->areas[i],
-		db_info->page_order);
+		for(j = 0; j < db_info->pages_per_area; j++) {
+			kfree(db_info->areas[i][j]);
+		}
+		kfree(db_info->areas[i]);
 	}
 	kfree(db_info->areas);
-	kfree(db_info->active_entry);
+	db_info->areas = NULL;
+}
+
+/*
+ * debug_info_free
+ * - free memory debug-info
+ */
+
+static void
+debug_info_free(debug_info_t* db_info){
+	debug_areas_free(db_info);
+	kfree(db_info->active_entries);
+	kfree(db_info->active_pages);
 	kfree(db_info);
 }
 
@@ -249,21 +332,22 @@ static void debug_info_free(debug_info_t* db_info){
  * - create new debug-info
  */
 
-static debug_info_t*  debug_info_create(char *name, int page_order, 
-                                        int nr_areas, int buf_size)
+static debug_info_t*
+debug_info_create(char *name, int pages_per_area, int nr_areas, int buf_size)
 {
 	debug_info_t* rc;
 
-        rc = debug_info_alloc(name, page_order, nr_areas, buf_size);
+        rc = debug_info_alloc(name, pages_per_area, nr_areas, buf_size,
+				DEBUG_DEFAULT_LEVEL, ALL_AREAS);
         if(!rc) 
 		goto out;
 
-
-	/* create proc rood directory */
-        rc->proc_root_entry = proc_mkdir(rc->name, debug_proc_root_entry);
+	/* create root directory */
+        rc->debugfs_root_entry = debugfs_create_dir(rc->name,
+					debug_debugfs_root_entry);
 
 	/* append new element to linked list */
-        if (debug_area_first == NULL) {
+        if (!debug_area_first) {
                 /* first element in list */
                 debug_area_first = rc;
                 rc->prev = NULL;
@@ -285,17 +369,21 @@ out:
  * - copy debug-info
  */
 
-static debug_info_t* debug_info_copy(debug_info_t* in)
+static debug_info_t*
+debug_info_copy(debug_info_t* in, int mode)
 {
-        int i;
+        int i,j;
         debug_info_t* rc;
-        rc = debug_info_alloc(in->name, in->page_order, 
-                                in->nr_areas, in->buf_size);
-        if(!rc)
+
+        rc = debug_info_alloc(in->name, in->pages_per_area, in->nr_areas,
+				in->buf_size, in->level, mode);
+        if(!rc || (mode == NO_AREAS))
                 goto out;
 
         for(i = 0; i < in->nr_areas; i++){
-                memcpy(rc->areas[i],in->areas[i], PAGE_SIZE << in->page_order);
+		for(j = 0; j < in->pages_per_area; j++) {
+			memcpy(rc->areas[i][j], in->areas[i][j],PAGE_SIZE);
+		}
         }
 out:
         return rc;
@@ -306,7 +394,8 @@ out:
  * - increments reference count for debug-info
  */
 
-static void debug_info_get(debug_info_t * db_info)
+static void
+debug_info_get(debug_info_t * db_info)
 {
 	if (db_info)
 		atomic_inc(&db_info->ref_count);
@@ -317,29 +406,20 @@ static void debug_info_get(debug_info_t * db_info)
  * - decreases reference count for debug-info and frees it if necessary
  */
 
-static void debug_info_put(debug_info_t *db_info)
+static void
+debug_info_put(debug_info_t *db_info)
 {
 	int i;
 
 	if (!db_info)
 		return;
 	if (atomic_dec_and_test(&db_info->ref_count)) {
-#ifdef DEBUG
-		printk(KERN_INFO "debug: freeing debug area %p (%s)\n",
-		       db_info, db_info->name);
-#endif
 		for (i = 0; i < DEBUG_MAX_VIEWS; i++) {
-			if (db_info->views[i] == NULL)
+			if (!db_info->views[i])
 				continue;
-#ifdef CONFIG_PROC_FS
-			remove_proc_entry(db_info->proc_entries[i]->name,
-					  db_info->proc_root_entry);
-#endif
+			debugfs_remove(db_info->debugfs_entries[i]);
 		}
-#ifdef CONFIG_PROC_FS
-		remove_proc_entry(db_info->proc_root_entry->name,
-				  debug_proc_root_entry);
-#endif
+		debugfs_remove(db_info->debugfs_root_entry);
 		if(db_info == debug_area_first)
 			debug_area_first = db_info->next;
 		if(db_info == debug_area_last)
@@ -355,9 +435,9 @@ static void debug_info_put(debug_info_t *db_info)
  * - format one debug entry and return size of formated data
  */
 
-static int debug_format_entry(file_private_info_t *p_info)
+static int
+debug_format_entry(file_private_info_t *p_info)
 {
-	debug_info_t *id_org    = p_info->debug_info_org;
 	debug_info_t *id_snap   = p_info->debug_info_snap;
 	struct debug_view *view = p_info->view;
 	debug_entry_t *act_entry;
@@ -365,22 +445,23 @@ static int debug_format_entry(file_private_info_t *p_info)
 	if(p_info->act_entry == DEBUG_PROLOG_ENTRY){
 		/* print prolog */
         	if (view->prolog_proc)
-                	len += view->prolog_proc(id_org, view,p_info->temp_buf);
+                	len += view->prolog_proc(id_snap,view,p_info->temp_buf);
 		goto out;
 	}
-
-	act_entry = (debug_entry_t *) ((char*)id_snap->areas[p_info->act_area] +
-					p_info->act_entry);
+	if (!id_snap->areas) /* this is true, if we have a prolog only view */
+		goto out;    /* or if 'pages_per_area' is 0 */
+	act_entry = (debug_entry_t *) ((char*)id_snap->areas[p_info->act_area]
+				[p_info->act_page] + p_info->act_entry);
                         
 	if (act_entry->id.stck == 0LL)
 			goto out;  /* empty entry */
 	if (view->header_proc)
-		len += view->header_proc(id_org, view, p_info->act_area, 
+		len += view->header_proc(id_snap, view, p_info->act_area,
 					act_entry, p_info->temp_buf + len);
 	if (view->format_proc)
-		len += view->format_proc(id_org, view, p_info->temp_buf + len,
+		len += view->format_proc(id_snap, view, p_info->temp_buf + len,
 						DEBUG_DATA(act_entry));
-      out:
+out:
         return len;
 }
 
@@ -389,20 +470,30 @@ static int debug_format_entry(file_private_info_t *p_info)
  * - goto next entry in p_info
  */
 
-extern inline int debug_next_entry(file_private_info_t *p_info)
+extern inline int
+debug_next_entry(file_private_info_t *p_info)
 {
-	debug_info_t *id = p_info->debug_info_snap;
+	debug_info_t *id;
+
+	id = p_info->debug_info_snap;
 	if(p_info->act_entry == DEBUG_PROLOG_ENTRY){
 		p_info->act_entry = 0;
+		p_info->act_page  = 0;
 		goto out;
 	}
-	if ((p_info->act_entry += id->entry_size)
-		> ((PAGE_SIZE << (id->page_order)) 
-		- id->entry_size)){
-
-		/* next area */
+	if(!id->areas)
+		return 1;
+	p_info->act_entry += id->entry_size;
+	/* switch to next page, if we reached the end of the page  */
+	if (p_info->act_entry > (PAGE_SIZE - id->entry_size)){
+		/* next page */
 		p_info->act_entry = 0;
-        	p_info->act_area++;
+		p_info->act_page += 1;
+		if((p_info->act_page % id->pages_per_area) == 0) {
+			/* next area */
+        		p_info->act_area++;
+			p_info->act_page=0;
+		}
         	if(p_info->act_area >= id->nr_areas)
 			return 1;
 	}
@@ -416,13 +507,14 @@ out:
  * - copies formated debug entries to the user buffer
  */
 
-static ssize_t debug_output(struct file *file,		/* file descriptor */
-			    char __user *user_buf,	/* user buffer */
-			    size_t  len,		/* length of buffer */
-			    loff_t *offset)	      /* offset in the file */
+static ssize_t
+debug_output(struct file *file,		/* file descriptor */
+	    char __user *user_buf,	/* user buffer */
+	    size_t  len,		/* length of buffer */
+	    loff_t *offset)		/* offset in the file */
 {
 	size_t count = 0;
-	size_t entry_offset, size = 0;
+	size_t entry_offset;
 	file_private_info_t *p_info;
 
 	p_info = ((file_private_info_t *) file->private_data);
@@ -430,27 +522,33 @@ static ssize_t debug_output(struct file *file,		/* file descriptor */
 		return -EPIPE;
 	if(p_info->act_area >= p_info->debug_info_snap->nr_areas)
 		return 0;
-
 	entry_offset = p_info->act_entry_offset;
-
 	while(count < len){
-		size = debug_format_entry(p_info);
-		size = min((len - count), (size - entry_offset));
-
-		if(size){
-			if (copy_to_user(user_buf + count, 
-					p_info->temp_buf + entry_offset, size))
-			return -EFAULT;
+		int formatted_line_size;
+		int formatted_line_residue;
+		int user_buf_residue;
+		size_t copy_size;
+
+		formatted_line_size = debug_format_entry(p_info);
+		formatted_line_residue = formatted_line_size - entry_offset;
+		user_buf_residue = len-count;
+		copy_size = min(user_buf_residue, formatted_line_residue);
+		if(copy_size){
+			if (copy_to_user(user_buf + count, p_info->temp_buf
+					+ entry_offset, copy_size))
+				return -EFAULT;
+			count += copy_size;
+			entry_offset += copy_size;
 		}
-		count += size;
-		entry_offset = 0;
-		if(count != len)
-			if(debug_next_entry(p_info)) 
+		if(copy_size == formatted_line_residue){
+			entry_offset = 0;
+			if(debug_next_entry(p_info))
 				goto out;
+		}
 	}
 out:
 	p_info->offset           = *offset + count;
-	p_info->act_entry_offset = size;	
+	p_info->act_entry_offset = entry_offset;
 	*offset = p_info->offset;
 	return count;
 }
@@ -461,9 +559,9 @@ out:
  * - calls input function of view
  */
 
-static ssize_t debug_input(struct file *file,
-			   const char __user *user_buf, size_t length,
-			   loff_t *offset)
+static ssize_t
+debug_input(struct file *file, const char __user *user_buf, size_t length,
+		loff_t *offset)
 {
 	int rc = 0;
 	file_private_info_t *p_info;
@@ -487,26 +585,23 @@ static ssize_t debug_input(struct file *file,
  *   handle
  */
 
-static int debug_open(struct inode *inode, struct file *file)
+static int
+debug_open(struct inode *inode, struct file *file)
 {
 	int i = 0, rc = 0;
 	file_private_info_t *p_info;
 	debug_info_t *debug_info, *debug_info_snapshot;
 
-#ifdef DEBUG
-	printk("debug_open\n");
-#endif
 	down(&debug_lock);
 
 	/* find debug log and view */
-
 	debug_info = debug_area_first;
 	while(debug_info != NULL){
 		for (i = 0; i < DEBUG_MAX_VIEWS; i++) {
-			if (debug_info->views[i] == NULL)
+			if (!debug_info->views[i])
 				continue;
-			else if (debug_info->proc_entries[i] ==
-				 PDE(file->f_dentry->d_inode)) {
+			else if (debug_info->debugfs_entries[i] ==
+				 file->f_dentry) {
 				goto found;	/* found view ! */
 			}
 		}
@@ -516,41 +611,42 @@ static int debug_open(struct inode *inode, struct file *file)
 	rc = -EINVAL;
 	goto out;
 
-      found:
+found:
 
-	/* make snapshot of current debug areas to get it consistent */
+	/* Make snapshot of current debug areas to get it consistent.     */
+	/* To copy all the areas is only needed, if we have a view which  */
+	/* formats the debug areas. */
 
-	debug_info_snapshot = debug_info_copy(debug_info);
+	if(!debug_info->views[i]->format_proc &&
+		!debug_info->views[i]->header_proc){
+		debug_info_snapshot = debug_info_copy(debug_info, NO_AREAS);
+	} else {
+		debug_info_snapshot = debug_info_copy(debug_info, ALL_AREAS);
+	}
 
 	if(!debug_info_snapshot){
-#ifdef DEBUG
-		printk(KERN_ERR "debug_open: debug_info_copy failed (out of mem)\n");
-#endif
 		rc = -ENOMEM;
 		goto out;
 	}
-
-	if ((file->private_data =
-	     kmalloc(sizeof(file_private_info_t), GFP_ATOMIC)) == 0) {
-#ifdef DEBUG
-		printk(KERN_ERR "debug_open: kmalloc failed\n");
-#endif
-		debug_info_free(debug_info_snapshot);	
+	p_info = (file_private_info_t *) kmalloc(sizeof(file_private_info_t),
+						GFP_KERNEL);
+	if(!p_info){
+		if(debug_info_snapshot)
+			debug_info_free(debug_info_snapshot);
 		rc = -ENOMEM;
 		goto out;
 	}
-	p_info = (file_private_info_t *) file->private_data;
 	p_info->offset = 0;
 	p_info->debug_info_snap = debug_info_snapshot;
 	p_info->debug_info_org  = debug_info;
 	p_info->view = debug_info->views[i];
 	p_info->act_area = 0;
+	p_info->act_page = 0;
 	p_info->act_entry = DEBUG_PROLOG_ENTRY;
 	p_info->act_entry_offset = 0;
-
+	file->private_data = p_info;
 	debug_info_get(debug_info);
-
-      out:
+out:
 	up(&debug_lock);
 	return rc;
 }
@@ -561,14 +657,13 @@ static int debug_open(struct inode *inode, struct file *file)
  * - deletes  private_data area of the file handle
  */
 
-static int debug_close(struct inode *inode, struct file *file)
+static int
+debug_close(struct inode *inode, struct file *file)
 {
 	file_private_info_t *p_info;
-#ifdef DEBUG
-	printk("debug_close\n");
-#endif
 	p_info = (file_private_info_t *) file->private_data;
-	debug_info_free(p_info->debug_info_snap);
+	if(p_info->debug_info_snap)
+		debug_info_free(p_info->debug_info_snap);
 	debug_info_put(p_info->debug_info_org);
 	kfree(file->private_data);
 	return 0;		/* success */
@@ -580,8 +675,8 @@ static int debug_close(struct inode *inode, struct file *file)
  * - returns handle for debug area
  */
 
-debug_info_t *debug_register
-    (char *name, int page_order, int nr_areas, int buf_size) 
+debug_info_t*
+debug_register (char *name, int pages_per_area, int nr_areas, int buf_size)
 {
 	debug_info_t *rc = NULL;
 
@@ -591,18 +686,14 @@ debug_info_t *debug_register
 
         /* create new debug_info */
 
-	rc = debug_info_create(name, page_order, nr_areas, buf_size);
+	rc = debug_info_create(name, pages_per_area, nr_areas, buf_size);
 	if(!rc) 
 		goto out;
 	debug_register_view(rc, &debug_level_view);
         debug_register_view(rc, &debug_flush_view);
-#ifdef DEBUG
-	printk(KERN_INFO
-	       "debug: reserved %d areas of %d pages for debugging %s\n",
-	       nr_areas, 1 << page_order, rc->name);
-#endif
-      out:
-        if (rc == NULL){
+	debug_register_view(rc, &debug_pages_view);
+out:
+        if (!rc){
 		printk(KERN_ERR "debug: debug_register failed for %s\n",name);
         }
 	up(&debug_lock);
@@ -614,27 +705,65 @@ debug_info_t *debug_register
  * - give back debug area
  */
 
-void debug_unregister(debug_info_t * id)
+void
+debug_unregister(debug_info_t * id)
 {
 	if (!id)
 		goto out;
 	down(&debug_lock);
-#ifdef DEBUG
-	printk(KERN_INFO "debug: unregistering %s\n", id->name);
-#endif
 	debug_info_put(id);
 	up(&debug_lock);
 
-      out:
+out:
 	return;
 }
 
+/*
+ * debug_set_size:
+ * - set area size (number of pages) and number of areas
+ */
+static int
+debug_set_size(debug_info_t* id, int nr_areas, int pages_per_area)
+{
+	unsigned long flags;
+	debug_entry_t *** new_areas;
+	int rc=0;
+
+	if(!id || (nr_areas <= 0) || (pages_per_area < 0))
+		return -EINVAL;
+	if(pages_per_area > 0){
+		new_areas = debug_areas_alloc(pages_per_area, nr_areas);
+		if(!new_areas) {
+			printk(KERN_WARNING "debug: could not allocate memory "\
+					 "for pagenumber: %i\n",pages_per_area);
+			rc = -ENOMEM;
+			goto out;
+		}
+	} else {
+		new_areas = NULL;
+	}
+	spin_lock_irqsave(&id->lock,flags);
+	debug_areas_free(id);
+	id->areas = new_areas;
+	id->nr_areas = nr_areas;
+	id->pages_per_area = pages_per_area;
+	id->active_area = 0;
+	memset(id->active_entries,0,sizeof(int)*id->nr_areas);
+	memset(id->active_pages, 0, sizeof(int)*id->nr_areas);
+	spin_unlock_irqrestore(&id->lock,flags);
+	printk(KERN_INFO "debug: %s: set new size (%i pages)\n"\
+			 ,id->name, pages_per_area);
+out:
+	return rc;
+}
+
 /*
  * debug_set_level:
  * - set actual debug level
  */
 
-void debug_set_level(debug_info_t* id, int new_level)
+void
+debug_set_level(debug_info_t* id, int new_level)
 {
 	unsigned long flags;
 	if(!id)
@@ -649,10 +778,6 @@ void debug_set_level(debug_info_t* id, int new_level)
                         id->name, new_level, 0, DEBUG_MAX_LEVEL);
         } else {
                 id->level = new_level;
-#ifdef DEBUG
-                printk(KERN_INFO 
-			"debug: %s: new level %i\n",id->name,id->level);
-#endif
         }
 	spin_unlock_irqrestore(&id->lock,flags);
 }
@@ -663,11 +788,16 @@ void debug_set_level(debug_info_t* id, int new_level)
  * - set active entry to next in the ring buffer
  */
 
-extern inline void proceed_active_entry(debug_info_t * id)
+extern inline void
+proceed_active_entry(debug_info_t * id)
 {
-	if ((id->active_entry[id->active_area] += id->entry_size)
-	    > ((PAGE_SIZE << (id->page_order)) - id->entry_size))
-		id->active_entry[id->active_area] = 0;
+	if ((id->active_entries[id->active_area] += id->entry_size)
+	    > (PAGE_SIZE - id->entry_size)){
+		id->active_entries[id->active_area] = 0;
+		id->active_pages[id->active_area] =
+			(id->active_pages[id->active_area] + 1) %
+			id->pages_per_area;
+	}
 }
 
 /*
@@ -675,7 +805,8 @@ extern inline void proceed_active_entry(debug_info_t * id)
  * - set active area to next in the ring buffer
  */
 
-extern inline void proceed_active_area(debug_info_t * id)
+extern inline void
+proceed_active_area(debug_info_t * id)
 {
 	id->active_area++;
 	id->active_area = id->active_area % id->nr_areas;
@@ -685,10 +816,12 @@ extern inline void proceed_active_area(debug_info_t * id)
  * get_active_entry:
  */
 
-extern inline debug_entry_t *get_active_entry(debug_info_t * id)
+extern inline debug_entry_t*
+get_active_entry(debug_info_t * id)
 {
-	return (debug_entry_t *) ((char *) id->areas[id->active_area] +
-				  id->active_entry[id->active_area]);
+	return (debug_entry_t *) (((char *) id->areas[id->active_area]
+					[id->active_pages[id->active_area]]) +
+					id->active_entries[id->active_area]);
 }
 
 /*
@@ -696,8 +829,9 @@ extern inline debug_entry_t *get_active_entry(debug_info_t * id)
  * - set timestamp, caller address, cpu number etc.
  */
 
-extern inline void debug_finish_entry(debug_info_t * id, debug_entry_t* active,
-		int level, int exception)
+extern inline void
+debug_finish_entry(debug_info_t * id, debug_entry_t* active, int level,
+			int exception)
 {
 	STCK(active->id.stck);
 	active->id.fields.cpuid = smp_processor_id();
@@ -721,7 +855,8 @@ static int debug_active=1;
  * always allow read, allow write only if debug_stoppable is set or
  * if debug_active is already off
  */
-static int s390dbf_procactive(ctl_table *table, int write, struct file *filp,
+static int
+s390dbf_procactive(ctl_table *table, int write, struct file *filp,
                      void __user *buffer, size_t *lenp, loff_t *ppos)
 {
 	if (!write || debug_stoppable || !debug_active)
@@ -766,7 +901,8 @@ static struct ctl_table s390dbf_dir_table[] = {
 
 struct ctl_table_header *s390dbf_sysctl_header;
 
-void debug_stop_all(void)
+void
+debug_stop_all(void)
 {
 	if (debug_stoppable)
 		debug_active = 0;
@@ -778,13 +914,13 @@ void debug_stop_all(void)
  * - write debug entry with given size
  */
 
-debug_entry_t *debug_event_common(debug_info_t * id, int level, const void *buf,
-			          int len)
+debug_entry_t*
+debug_event_common(debug_info_t * id, int level, const void *buf, int len)
 {
 	unsigned long flags;
 	debug_entry_t *active;
 
-	if (!debug_active)
+	if (!debug_active || !id->areas)
 		return NULL;
 	spin_lock_irqsave(&id->lock, flags);
 	active = get_active_entry(id);
@@ -801,13 +937,13 @@ debug_entry_t *debug_event_common(debug_info_t * id, int level, const void *buf,
  * - write debug entry with given size and switch to next debug area
  */
 
-debug_entry_t *debug_exception_common(debug_info_t * id, int level, 
-                                      const void *buf, int len)
+debug_entry_t
+*debug_exception_common(debug_info_t * id, int level, const void *buf, int len)
 {
 	unsigned long flags;
 	debug_entry_t *active;
 
-	if (!debug_active)
+	if (!debug_active || !id->areas)
 		return NULL;
 	spin_lock_irqsave(&id->lock, flags);
 	active = get_active_entry(id);
@@ -823,7 +959,8 @@ debug_entry_t *debug_exception_common(debug_info_t * id, int level,
  * counts arguments in format string for sprintf view
  */
 
-extern inline int debug_count_numargs(char *string)
+extern inline int
+debug_count_numargs(char *string)
 {
 	int numargs=0;
 
@@ -838,8 +975,8 @@ extern inline int debug_count_numargs(char *string)
  * debug_sprintf_event:
  */
 
-debug_entry_t *debug_sprintf_event(debug_info_t* id,
-                                   int level,char *string,...)
+debug_entry_t*
+debug_sprintf_event(debug_info_t* id, int level,char *string,...)
 {
 	va_list   ap;
 	int numargs,idx;
@@ -849,7 +986,7 @@ debug_entry_t *debug_sprintf_event(debug_info_t* id,
 
 	if((!id) || (level > id->level))
 		return NULL;
-	if (!debug_active)
+	if (!debug_active || !id->areas)
 		return NULL;
 	numargs=debug_count_numargs(string);
 
@@ -871,8 +1008,8 @@ debug_entry_t *debug_sprintf_event(debug_info_t* id,
  * debug_sprintf_exception:
  */
 
-debug_entry_t *debug_sprintf_exception(debug_info_t* id,
-                                       int level,char *string,...)
+debug_entry_t*
+debug_sprintf_exception(debug_info_t* id, int level,char *string,...)
 {
 	va_list   ap;
 	int numargs,idx;
@@ -882,7 +1019,7 @@ debug_entry_t *debug_sprintf_exception(debug_info_t* id,
 
 	if((!id) || (level > id->level))
 		return NULL;
-	if (!debug_active)
+	if (!debug_active || !id->areas)
 		return NULL;
 
 	numargs=debug_count_numargs(string);
@@ -906,15 +1043,14 @@ debug_entry_t *debug_sprintf_exception(debug_info_t* id,
  * - is called exactly once to initialize the debug feature
  */
 
-static int __init debug_init(void)
+static int
+__init debug_init(void)
 {
 	int rc = 0;
 
 	s390dbf_sysctl_header = register_sysctl_table(s390dbf_dir_table, 1);
 	down(&debug_lock);
-#ifdef CONFIG_PROC_FS
-	debug_proc_root_entry = proc_mkdir(DEBUG_DIR_ROOT, NULL);
-#endif /* CONFIG_PROC_FS */
+	debug_debugfs_root_entry = debugfs_create_dir(DEBUG_DIR_ROOT,NULL);
 	printk(KERN_INFO "debug: Initialization complete\n");
 	initialized = 1;
 	up(&debug_lock);
@@ -926,13 +1062,14 @@ static int __init debug_init(void)
  * debug_register_view:
  */
 
-int debug_register_view(debug_info_t * id, struct debug_view *view)
+int
+debug_register_view(debug_info_t * id, struct debug_view *view)
 {
 	int rc = 0;
 	int i;
 	unsigned long flags;
 	mode_t mode = S_IFREG;
-	struct proc_dir_entry *pde;
+	struct dentry *pde;
 
 	if (!id)
 		goto out;
@@ -940,16 +1077,17 @@ int debug_register_view(debug_info_t * id, struct debug_view *view)
 		mode |= S_IRUSR;
 	if (view->input_proc)
 		mode |= S_IWUSR;
-	pde = create_proc_entry(view->name, mode, id->proc_root_entry);
+	pde = debugfs_create_file(view->name, mode, id->debugfs_root_entry,
+				NULL, &debug_file_ops);
 	if (!pde){
-		printk(KERN_WARNING "debug: create_proc_entry() failed! Cannot register view %s/%s\n", id->name,view->name);
+		printk(KERN_WARNING "debug: debugfs_create_file() failed!"\
+			" Cannot register view %s/%s\n", id->name,view->name);
 		rc = -1;
 		goto out;
 	}
-
 	spin_lock_irqsave(&id->lock, flags);
 	for (i = 0; i < DEBUG_MAX_VIEWS; i++) {
-		if (id->views[i] == NULL)
+		if (!id->views[i])
 			break;
 	}
 	if (i == DEBUG_MAX_VIEWS) {
@@ -957,16 +1095,14 @@ int debug_register_view(debug_info_t * id, struct debug_view *view)
 			id->name,view->name);
 		printk(KERN_WARNING 
 			"debug: maximum number of views reached (%i)!\n", i);
-		remove_proc_entry(pde->name, id->proc_root_entry);
+		debugfs_remove(pde);
 		rc = -1;
-	}
-	else {
+	} else {
 		id->views[i] = view;
-		pde->proc_fops = &debug_file_ops;
-		id->proc_entries[i] = pde;
+		id->debugfs_entries[i] = pde;
 	}
 	spin_unlock_irqrestore(&id->lock, flags);
-      out:
+out:
 	return rc;
 }
 
@@ -974,7 +1110,8 @@ int debug_register_view(debug_info_t * id, struct debug_view *view)
  * debug_unregister_view:
  */
 
-int debug_unregister_view(debug_info_t * id, struct debug_view *view)
+int
+debug_unregister_view(debug_info_t * id, struct debug_view *view)
 {
 	int rc = 0;
 	int i;
@@ -990,15 +1127,46 @@ int debug_unregister_view(debug_info_t * id, struct debug_view *view)
 	if (i == DEBUG_MAX_VIEWS)
 		rc = -1;
 	else {
-#ifdef CONFIG_PROC_FS
-		remove_proc_entry(id->proc_entries[i]->name,
-				  id->proc_root_entry);
-#endif
+		debugfs_remove(id->debugfs_entries[i]);
 		id->views[i] = NULL;
 		rc = 0;
 	}
 	spin_unlock_irqrestore(&id->lock, flags);
-      out:
+out:
+	return rc;
+}
+
+static inline char *
+debug_get_user_string(const char __user *user_buf, size_t user_len)
+{
+	char* buffer;
+
+	buffer = kmalloc(user_len + 1, GFP_KERNEL);
+	if (!buffer)
+		return ERR_PTR(-ENOMEM);
+	if (copy_from_user(buffer, user_buf, user_len) != 0) {
+		kfree(buffer);
+		return ERR_PTR(-EFAULT);
+	}
+	/* got the string, now strip linefeed. */
+	if (buffer[user_len - 1] == '\n')
+		buffer[user_len - 1] = 0;
+	else
+		buffer[user_len] = 0;
+        return buffer;
+}
+
+static inline int
+debug_get_uint(char *buf)
+{
+	int rc;
+
+	for(; isspace(*buf); buf++);
+	rc = simple_strtoul(buf, &buf, 10);
+	if(*buf){
+		printk("debug: no integer specified!\n");
+		rc = -EINVAL;
+	}
 	return rc;
 }
 
@@ -1011,13 +1179,69 @@ int debug_unregister_view(debug_info_t * id, struct debug_view *view)
  * prints out actual debug level
  */
 
-static int debug_prolog_level_fn(debug_info_t * id,
+static int
+debug_prolog_pages_fn(debug_info_t * id,
 				 struct debug_view *view, char *out_buf)
+{
+	return sprintf(out_buf, "%i\n", id->pages_per_area);
+}
+
+/*
+ * reads new size (number of pages per debug area)
+ */
+
+static int
+debug_input_pages_fn(debug_info_t * id, struct debug_view *view,
+			struct file *file, const char __user *user_buf,
+			size_t user_len, loff_t * offset)
+{
+	char *str;
+	int rc,new_pages;
+
+	if (user_len > 0x10000)
+                user_len = 0x10000;
+	if (*offset != 0){
+		rc = -EPIPE;
+		goto out;
+	}
+	str = debug_get_user_string(user_buf,user_len);
+	if(IS_ERR(str)){
+		rc = PTR_ERR(str);
+		goto out;
+	}
+	new_pages = debug_get_uint(str);
+	if(new_pages < 0){
+		rc = -EINVAL;
+		goto free_str;
+	}
+	rc = debug_set_size(id,id->nr_areas, new_pages);
+	if(rc != 0){
+		rc = -EINVAL;
+		goto free_str;
+	}
+	rc = user_len;
+free_str:
+	kfree(str);
+out:
+	*offset += user_len;
+	return rc;		/* number of input characters */
+}
+
+/*
+ * prints out actual debug level
+ */
+
+static int
+debug_prolog_level_fn(debug_info_t * id, struct debug_view *view, char *out_buf)
 {
 	int rc = 0;
 
-	if(id->level == -1) rc = sprintf(out_buf,"-\n");
-	else rc = sprintf(out_buf, "%i\n", id->level);
+	if(id->level == DEBUG_OFF_LEVEL) {
+		rc = sprintf(out_buf,"-\n");
+	}
+	else {
+		rc = sprintf(out_buf, "%i\n", id->level);
+	}
 	return rc;
 }
 
@@ -1025,30 +1249,43 @@ static int debug_prolog_level_fn(debug_info_t * id,
  * reads new debug level
  */
 
-static int debug_input_level_fn(debug_info_t * id, struct debug_view *view,
-				struct file *file, const char __user *user_buf,
-				size_t in_buf_size, loff_t * offset)
+static int
+debug_input_level_fn(debug_info_t * id, struct debug_view *view,
+			struct file *file, const char __user *user_buf,
+			size_t user_len, loff_t * offset)
 {
-	char input_buf[1];
-	int rc = in_buf_size;
+	char *str;
+	int rc,new_level;
 
-	if (*offset != 0)
+	if (user_len > 0x10000)
+                user_len = 0x10000;
+	if (*offset != 0){
+		rc = -EPIPE;
 		goto out;
-	if (copy_from_user(input_buf, user_buf, 1)){
-		rc = -EFAULT;
+	}
+	str = debug_get_user_string(user_buf,user_len);
+	if(IS_ERR(str)){
+		rc = PTR_ERR(str);
 		goto out;
 	}
-	if (isdigit(input_buf[0])) {
-		int new_level = ((int) input_buf[0] - (int) '0');
-		debug_set_level(id, new_level);
-	} else if(input_buf[0] == '-') {
+	if(str[0] == '-'){
 		debug_set_level(id, DEBUG_OFF_LEVEL);
+		rc = user_len;
+		goto free_str;
 	} else {
-		printk(KERN_INFO "debug: level `%c` is not valid\n",
-		       input_buf[0]);
+		new_level = debug_get_uint(str);
 	}
-      out:
-	*offset += in_buf_size;
+	if(new_level < 0) {
+		printk(KERN_INFO "debug: level `%s` is not valid\n", str);
+		rc = -EINVAL;
+	} else {
+		debug_set_level(id, new_level);
+		rc = user_len;
+	}
+free_str:
+	kfree(str);
+out:
+	*offset += user_len;
 	return rc;		/* number of input characters */
 }
 
@@ -1057,29 +1294,36 @@ static int debug_input_level_fn(debug_info_t * id, struct debug_view *view,
  * flushes debug areas
  */
  
-void debug_flush(debug_info_t* id, int area)
+void
+debug_flush(debug_info_t* id, int area)
 {
         unsigned long flags;
-        int i;
+        int i,j;
 
-        if(!id)
+        if(!id || !id->areas)
                 return;
         spin_lock_irqsave(&id->lock,flags);
         if(area == DEBUG_FLUSH_ALL){
                 id->active_area = 0;
-                memset(id->active_entry, 0, id->nr_areas * sizeof(int));
-                for (i = 0; i < id->nr_areas; i++) 
-                        memset(id->areas[i], 0, PAGE_SIZE << id->page_order);
+                memset(id->active_entries, 0, id->nr_areas * sizeof(int));
+                for (i = 0; i < id->nr_areas; i++) {
+			id->active_pages[i] = 0;
+			for(j = 0; j < id->pages_per_area; j++) {
+                        	memset(id->areas[i][j], 0, PAGE_SIZE);
+			}
+		}
                 printk(KERN_INFO "debug: %s: all areas flushed\n",id->name);
         } else if(area >= 0 && area < id->nr_areas) {
-                id->active_entry[area] = 0;
-                memset(id->areas[area], 0, PAGE_SIZE << id->page_order);
-                printk(KERN_INFO
-                        "debug: %s: area %i has been flushed\n",
+                id->active_entries[area] = 0;
+		id->active_pages[area] = 0;
+		for(i = 0; i < id->pages_per_area; i++) {
+                	memset(id->areas[area][i],0,PAGE_SIZE);
+		}
+                printk(KERN_INFO "debug: %s: area %i has been flushed\n",
                         id->name, area);
         } else {
                 printk(KERN_INFO
-                        "debug: %s: area %i cannot be flushed (range: %i - %i)\n",
+                      "debug: %s: area %i cannot be flushed (range: %i - %i)\n",
                         id->name, area, 0, id->nr_areas-1);
         }
         spin_unlock_irqrestore(&id->lock,flags);
@@ -1089,15 +1333,20 @@ void debug_flush(debug_info_t* id, int area)
  * view function: flushes debug areas 
  */
 
-static int debug_input_flush_fn(debug_info_t * id, struct debug_view *view,
-                                struct file *file, const char __user *user_buf,
-                                size_t in_buf_size, loff_t * offset)
+static int
+debug_input_flush_fn(debug_info_t * id, struct debug_view *view,
+			struct file *file, const char __user *user_buf,
+			size_t user_len, loff_t * offset)
 {
         char input_buf[1];
-        int rc = in_buf_size;
- 
-        if (*offset != 0)
+        int rc = user_len;
+
+	if (user_len > 0x10000)
+                user_len = 0x10000;
+        if (*offset != 0){
+		rc = -EPIPE;
                 goto out;
+	}
         if (copy_from_user(input_buf, user_buf, 1)){
                 rc = -EFAULT;
                 goto out;
@@ -1114,8 +1363,8 @@ static int debug_input_flush_fn(debug_info_t * id, struct debug_view *view,
 
         printk(KERN_INFO "debug: area `%c` is not valid\n", input_buf[0]);
 
-      out:
-        *offset += in_buf_size;
+out:
+        *offset += user_len;
         return rc;              /* number of input characters */
 }
 
@@ -1123,8 +1372,9 @@ static int debug_input_flush_fn(debug_info_t * id, struct debug_view *view,
  * prints debug header in raw format
  */
 
-int debug_raw_header_fn(debug_info_t * id, struct debug_view *view,
-                         int area, debug_entry_t * entry, char *out_buf)
+static int
+debug_raw_header_fn(debug_info_t * id, struct debug_view *view,
+			int area, debug_entry_t * entry, char *out_buf)
 {
         int rc;
 
@@ -1137,7 +1387,8 @@ int debug_raw_header_fn(debug_info_t * id, struct debug_view *view,
  * prints debug data in raw format
  */
 
-static int debug_raw_format_fn(debug_info_t * id, struct debug_view *view,
+static int
+debug_raw_format_fn(debug_info_t * id, struct debug_view *view,
 			       char *out_buf, const char *in_buf)
 {
 	int rc;
@@ -1151,8 +1402,9 @@ static int debug_raw_format_fn(debug_info_t * id, struct debug_view *view,
  * prints debug data in hex/ascii format
  */
 
-static int debug_hex_ascii_format_fn(debug_info_t * id, struct debug_view *view,
-		    		  char *out_buf, const char *in_buf)
+static int
+debug_hex_ascii_format_fn(debug_info_t * id, struct debug_view *view,
+	    		  char *out_buf, const char *in_buf)
 {
 	int i, rc = 0;
 
@@ -1176,7 +1428,8 @@ static int debug_hex_ascii_format_fn(debug_info_t * id, struct debug_view *view,
  * prints header for debug entry
  */
 
-int debug_dflt_header_fn(debug_info_t * id, struct debug_view *view,
+int
+debug_dflt_header_fn(debug_info_t * id, struct debug_view *view,
 			 int area, debug_entry_t * entry, char *out_buf)
 {
 	struct timeval time_val;
@@ -1210,8 +1463,9 @@ int debug_dflt_header_fn(debug_info_t * id, struct debug_view *view,
 
 #define DEBUG_SPRINTF_MAX_ARGS 10
 
-int debug_sprintf_format_fn(debug_info_t * id, struct debug_view *view,
-                            char *out_buf, debug_sprintf_entry_t *curr_event)
+static int
+debug_sprintf_format_fn(debug_info_t * id, struct debug_view *view,
+                        char *out_buf, debug_sprintf_entry_t *curr_event)
 {
 	int num_longs, num_used_args = 0,i, rc = 0;
 	int index[DEBUG_SPRINTF_MAX_ARGS];
@@ -1251,14 +1505,10 @@ out:
 /*
  * clean up module
  */
-void __exit debug_exit(void)
+void
+__exit debug_exit(void)
 {
-#ifdef DEBUG
-	printk("debug_cleanup_module: \n");
-#endif
-#ifdef CONFIG_PROC_FS
-	remove_proc_entry(debug_proc_root_entry->name, NULL);
-#endif /* CONFIG_PROC_FS */
+	debugfs_remove(debug_debugfs_root_entry);
 	unregister_sysctl_table(s390dbf_sysctl_header);
 	return;
 }
@@ -1266,7 +1516,7 @@ void __exit debug_exit(void)
 /*
  * module definitions
  */
-core_initcall(debug_init);
+postcore_initcall(debug_init);
 module_exit(debug_exit);
 MODULE_LICENSE("GPL");
 
diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c
index 3e39508bd92..6527ff6f470 100644
--- a/drivers/s390/block/dasd.c
+++ b/drivers/s390/block/dasd.c
@@ -176,7 +176,7 @@ dasd_state_known_to_basic(struct dasd_device * device)
 		return rc;
 
 	/* register 'device' debug area, used for all DBF_DEV_XXX calls */
-	device->debug_area = debug_register(device->cdev->dev.bus_id, 0, 2,
+	device->debug_area = debug_register(device->cdev->dev.bus_id, 1, 2,
 					    8 * sizeof (long));
 	debug_register_view(device->debug_area, &debug_sprintf_view);
 	debug_set_level(device->debug_area, DBF_EMERG);
@@ -1981,7 +1981,7 @@ dasd_init(void)
 	init_waitqueue_head(&dasd_init_waitq);
 
 	/* register 'common' DASD debug area, used for all DBF_XXX calls */
-	dasd_debug_area = debug_register("dasd", 0, 2, 8 * sizeof (long));
+	dasd_debug_area = debug_register("dasd", 1, 2, 8 * sizeof (long));
 	if (dasd_debug_area == NULL) {
 		rc = -ENOMEM;
 		goto failed;
diff --git a/drivers/s390/block/dasd_proc.c b/drivers/s390/block/dasd_proc.c
index d7f19745911..43c34f8c5e6 100644
--- a/drivers/s390/block/dasd_proc.c
+++ b/drivers/s390/block/dasd_proc.c
@@ -9,13 +9,14 @@
  *
  * /proc interface for the dasd driver.
  *
- * $Revision: 1.31 $
+ * $Revision: 1.32 $
  */
 
 #include <linux/config.h>
 #include <linux/ctype.h>
 #include <linux/seq_file.h>
 #include <linux/vmalloc.h>
+#include <linux/proc_fs.h>
 
 #include <asm/debug.h>
 #include <asm/uaccess.h>
diff --git a/drivers/s390/char/tape_34xx.c b/drivers/s390/char/tape_34xx.c
index 480ec87976f..20be88e91fa 100644
--- a/drivers/s390/char/tape_34xx.c
+++ b/drivers/s390/char/tape_34xx.c
@@ -1351,13 +1351,13 @@ tape_34xx_init (void)
 {
 	int rc;
 
-	TAPE_DBF_AREA = debug_register ( "tape_34xx", 1, 2, 4*sizeof(long));
+	TAPE_DBF_AREA = debug_register ( "tape_34xx", 2, 2, 4*sizeof(long));
 	debug_register_view(TAPE_DBF_AREA, &debug_sprintf_view);
 #ifdef DBF_LIKE_HELL
 	debug_set_level(TAPE_DBF_AREA, 6);
 #endif
 
-	DBF_EVENT(3, "34xx init: $Revision: 1.21 $\n");
+	DBF_EVENT(3, "34xx init: $Revision: 1.23 $\n");
 	/* Register driver for 3480/3490 tapes. */
 	rc = ccw_driver_register(&tape_34xx_driver);
 	if (rc)
@@ -1378,7 +1378,7 @@ tape_34xx_exit(void)
 MODULE_DEVICE_TABLE(ccw, tape_34xx_ids);
 MODULE_AUTHOR("(C) 2001-2002 IBM Deutschland Entwicklung GmbH");
 MODULE_DESCRIPTION("Linux on zSeries channel attached 3480 tape "
-		   "device driver ($Revision: 1.21 $)");
+		   "device driver ($Revision: 1.23 $)");
 MODULE_LICENSE("GPL");
 
 module_init(tape_34xx_init);
diff --git a/drivers/s390/char/tape_core.c b/drivers/s390/char/tape_core.c
index b4df4a515b1..0597aa0e27e 100644
--- a/drivers/s390/char/tape_core.c
+++ b/drivers/s390/char/tape_core.c
@@ -1186,7 +1186,7 @@ tape_mtop(struct tape_device *device, int mt_op, int mt_count)
 static int
 tape_init (void)
 {
-	TAPE_DBF_AREA = debug_register ( "tape", 1, 2, 4*sizeof(long));
+	TAPE_DBF_AREA = debug_register ( "tape", 2, 2, 4*sizeof(long));
 	debug_register_view(TAPE_DBF_AREA, &debug_sprintf_view);
 #ifdef DBF_LIKE_HELL
 	debug_set_level(TAPE_DBF_AREA, 6);
diff --git a/drivers/s390/char/tape_proc.c b/drivers/s390/char/tape_proc.c
index 801d17cca34..5fec0a10cc3 100644
--- a/drivers/s390/char/tape_proc.c
+++ b/drivers/s390/char/tape_proc.c
@@ -15,6 +15,7 @@
 #include <linux/module.h>
 #include <linux/vmalloc.h>
 #include <linux/seq_file.h>
+#include <linux/proc_fs.h>
 
 #define TAPE_DBF_AREA	tape_core_dbf
 
diff --git a/drivers/s390/char/vmcp.c b/drivers/s390/char/vmcp.c
index dfbbf235ca2..7f11a608a63 100644
--- a/drivers/s390/char/vmcp.c
+++ b/drivers/s390/char/vmcp.c
@@ -203,7 +203,7 @@ static int __init vmcp_init(void)
 	else
 		printk(KERN_WARNING
 		       "z/VM CP interface not loaded. Could not register misc device.\n");
-	vmcp_debug = debug_register("vmcp", 0, 1, 240);
+	vmcp_debug = debug_register("vmcp", 1, 1, 240);
 	debug_register_view(vmcp_debug, &debug_hex_ascii_view);
 	return ret;
 }
diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c
index 1d9b3f18d8d..ea813bdce1d 100644
--- a/drivers/s390/cio/cio.c
+++ b/drivers/s390/cio/cio.c
@@ -1,7 +1,7 @@
 /*
  *  drivers/s390/cio/cio.c
  *   S/390 common I/O routines -- low level i/o calls
- *   $Revision: 1.133 $
+ *   $Revision: 1.134 $
  *
  *    Copyright (C) 1999-2002 IBM Deutschland Entwicklung GmbH,
  *			      IBM Corporation
@@ -63,17 +63,17 @@ __setup ("cio_msg=", cio_setup);
 static int __init
 cio_debug_init (void)
 {
-	cio_debug_msg_id = debug_register ("cio_msg", 4, 4, 16*sizeof (long));
+	cio_debug_msg_id = debug_register ("cio_msg", 16, 4, 16*sizeof (long));
 	if (!cio_debug_msg_id)
 		goto out_unregister;
 	debug_register_view (cio_debug_msg_id, &debug_sprintf_view);
 	debug_set_level (cio_debug_msg_id, 2);
-	cio_debug_trace_id = debug_register ("cio_trace", 4, 4, 8);
+	cio_debug_trace_id = debug_register ("cio_trace", 16, 4, 8);
 	if (!cio_debug_trace_id)
 		goto out_unregister;
 	debug_register_view (cio_debug_trace_id, &debug_hex_ascii_view);
 	debug_set_level (cio_debug_trace_id, 2);
-	cio_debug_crw_id = debug_register ("cio_crw", 2, 4, 16*sizeof (long));
+	cio_debug_crw_id = debug_register ("cio_crw", 4, 4, 16*sizeof (long));
 	if (!cio_debug_crw_id)
 		goto out_unregister;
 	debug_register_view (cio_debug_crw_id, &debug_sprintf_view);
diff --git a/drivers/s390/cio/qdio.c b/drivers/s390/cio/qdio.c
index bbe9f45d143..82194c4eadf 100644
--- a/drivers/s390/cio/qdio.c
+++ b/drivers/s390/cio/qdio.c
@@ -56,7 +56,7 @@
 #include "ioasm.h"
 #include "chsc.h"
 
-#define VERSION_QDIO_C "$Revision: 1.98 $"
+#define VERSION_QDIO_C "$Revision: 1.101 $"
 
 /****************** MODULE PARAMETER VARIABLES ********************/
 MODULE_AUTHOR("Utz Bacher <utz.bacher@de.ibm.com>");
@@ -3342,7 +3342,7 @@ static int
 qdio_register_dbf_views(void)
 {
 	qdio_dbf_setup=debug_register(QDIO_DBF_SETUP_NAME,
-				      QDIO_DBF_SETUP_INDEX,
+				      QDIO_DBF_SETUP_PAGES,
 				      QDIO_DBF_SETUP_NR_AREAS,
 				      QDIO_DBF_SETUP_LEN);
 	if (!qdio_dbf_setup)
@@ -3351,7 +3351,7 @@ qdio_register_dbf_views(void)
 	debug_set_level(qdio_dbf_setup,QDIO_DBF_SETUP_LEVEL);
 
 	qdio_dbf_sbal=debug_register(QDIO_DBF_SBAL_NAME,
-				     QDIO_DBF_SBAL_INDEX,
+				     QDIO_DBF_SBAL_PAGES,
 				     QDIO_DBF_SBAL_NR_AREAS,
 				     QDIO_DBF_SBAL_LEN);
 	if (!qdio_dbf_sbal)
@@ -3361,7 +3361,7 @@ qdio_register_dbf_views(void)
 	debug_set_level(qdio_dbf_sbal,QDIO_DBF_SBAL_LEVEL);
 
 	qdio_dbf_sense=debug_register(QDIO_DBF_SENSE_NAME,
-				      QDIO_DBF_SENSE_INDEX,
+				      QDIO_DBF_SENSE_PAGES,
 				      QDIO_DBF_SENSE_NR_AREAS,
 				      QDIO_DBF_SENSE_LEN);
 	if (!qdio_dbf_sense)
@@ -3371,7 +3371,7 @@ qdio_register_dbf_views(void)
 	debug_set_level(qdio_dbf_sense,QDIO_DBF_SENSE_LEVEL);
 
 	qdio_dbf_trace=debug_register(QDIO_DBF_TRACE_NAME,
-				      QDIO_DBF_TRACE_INDEX,
+				      QDIO_DBF_TRACE_PAGES,
 				      QDIO_DBF_TRACE_NR_AREAS,
 				      QDIO_DBF_TRACE_LEN);
 	if (!qdio_dbf_trace)
@@ -3382,7 +3382,7 @@ qdio_register_dbf_views(void)
 
 #ifdef CONFIG_QDIO_DEBUG
         qdio_dbf_slsb_out=debug_register(QDIO_DBF_SLSB_OUT_NAME,
-                                         QDIO_DBF_SLSB_OUT_INDEX,
+                                         QDIO_DBF_SLSB_OUT_PAGES,
                                          QDIO_DBF_SLSB_OUT_NR_AREAS,
                                          QDIO_DBF_SLSB_OUT_LEN);
         if (!qdio_dbf_slsb_out)
@@ -3391,7 +3391,7 @@ qdio_register_dbf_views(void)
         debug_set_level(qdio_dbf_slsb_out,QDIO_DBF_SLSB_OUT_LEVEL);
 
         qdio_dbf_slsb_in=debug_register(QDIO_DBF_SLSB_IN_NAME,
-                                        QDIO_DBF_SLSB_IN_INDEX,
+                                        QDIO_DBF_SLSB_IN_PAGES,
                                         QDIO_DBF_SLSB_IN_NR_AREAS,
                                         QDIO_DBF_SLSB_IN_LEN);
         if (!qdio_dbf_slsb_in)
diff --git a/drivers/s390/cio/qdio.h b/drivers/s390/cio/qdio.h
index b6daadac4e8..6b8aa6a852b 100644
--- a/drivers/s390/cio/qdio.h
+++ b/drivers/s390/cio/qdio.h
@@ -3,7 +3,7 @@
 
 #include <asm/page.h>
 
-#define VERSION_CIO_QDIO_H "$Revision: 1.32 $"
+#define VERSION_CIO_QDIO_H "$Revision: 1.33 $"
 
 #ifdef CONFIG_QDIO_DEBUG
 #define QDIO_VERBOSE_LEVEL 9
@@ -132,7 +132,7 @@ enum qdio_irq_states {
 
 #define QDIO_DBF_SETUP_NAME "qdio_setup"
 #define QDIO_DBF_SETUP_LEN 8
-#define QDIO_DBF_SETUP_INDEX 2
+#define QDIO_DBF_SETUP_PAGES 4
 #define QDIO_DBF_SETUP_NR_AREAS 1
 #ifdef CONFIG_QDIO_DEBUG
 #define QDIO_DBF_SETUP_LEVEL 6
@@ -142,7 +142,7 @@ enum qdio_irq_states {
 
 #define QDIO_DBF_SBAL_NAME "qdio_labs" /* sbal */
 #define QDIO_DBF_SBAL_LEN 256
-#define QDIO_DBF_SBAL_INDEX 2
+#define QDIO_DBF_SBAL_PAGES 4
 #define QDIO_DBF_SBAL_NR_AREAS 2
 #ifdef CONFIG_QDIO_DEBUG
 #define QDIO_DBF_SBAL_LEVEL 6
@@ -154,16 +154,16 @@ enum qdio_irq_states {
 #define QDIO_DBF_TRACE_LEN 8
 #define QDIO_DBF_TRACE_NR_AREAS 2
 #ifdef CONFIG_QDIO_DEBUG
-#define QDIO_DBF_TRACE_INDEX 4
+#define QDIO_DBF_TRACE_PAGES 16
 #define QDIO_DBF_TRACE_LEVEL 4 /* -------- could be even more verbose here */
 #else /* CONFIG_QDIO_DEBUG */
-#define QDIO_DBF_TRACE_INDEX 2
+#define QDIO_DBF_TRACE_PAGES 4
 #define QDIO_DBF_TRACE_LEVEL 2
 #endif /* CONFIG_QDIO_DEBUG */
 
 #define QDIO_DBF_SENSE_NAME "qdio_sense"
 #define QDIO_DBF_SENSE_LEN 64
-#define QDIO_DBF_SENSE_INDEX 1
+#define QDIO_DBF_SENSE_PAGES 2
 #define QDIO_DBF_SENSE_NR_AREAS 1
 #ifdef CONFIG_QDIO_DEBUG
 #define QDIO_DBF_SENSE_LEVEL 6
@@ -176,13 +176,13 @@ enum qdio_irq_states {
 
 #define QDIO_DBF_SLSB_OUT_NAME "qdio_slsb_out"
 #define QDIO_DBF_SLSB_OUT_LEN QDIO_MAX_BUFFERS_PER_Q
-#define QDIO_DBF_SLSB_OUT_INDEX 8
+#define QDIO_DBF_SLSB_OUT_PAGES 256
 #define QDIO_DBF_SLSB_OUT_NR_AREAS 1
 #define QDIO_DBF_SLSB_OUT_LEVEL 6
 
 #define QDIO_DBF_SLSB_IN_NAME "qdio_slsb_in"
 #define QDIO_DBF_SLSB_IN_LEN QDIO_MAX_BUFFERS_PER_Q
-#define QDIO_DBF_SLSB_IN_INDEX 8
+#define QDIO_DBF_SLSB_IN_PAGES 256
 #define QDIO_DBF_SLSB_IN_NR_AREAS 1
 #define QDIO_DBF_SLSB_IN_LEVEL 6
 #endif /* CONFIG_QDIO_DEBUG */
diff --git a/drivers/s390/net/claw.c b/drivers/s390/net/claw.c
index a99927d54eb..60440dbe3a2 100644
--- a/drivers/s390/net/claw.c
+++ b/drivers/s390/net/claw.c
@@ -146,8 +146,8 @@ claw_unregister_debug_facility(void)
 static int
 claw_register_debug_facility(void)
 {
-	claw_dbf_setup = debug_register("claw_setup", 1, 1, 8);
-	claw_dbf_trace = debug_register("claw_trace", 1, 2, 8);
+	claw_dbf_setup = debug_register("claw_setup", 2, 1, 8);
+	claw_dbf_trace = debug_register("claw_trace", 2, 2, 8);
 	if (claw_dbf_setup == NULL || claw_dbf_trace == NULL) {
 		printk(KERN_WARNING "Not enough memory for debug facility.\n");
 		claw_unregister_debug_facility();
diff --git a/drivers/s390/net/ctcdbug.c b/drivers/s390/net/ctcdbug.c
index 2c86bfa11b2..0e2a8bb9303 100644
--- a/drivers/s390/net/ctcdbug.c
+++ b/drivers/s390/net/ctcdbug.c
@@ -1,6 +1,6 @@
 /*
  *
- * linux/drivers/s390/net/ctcdbug.c ($Revision: 1.4 $)
+ * linux/drivers/s390/net/ctcdbug.c ($Revision: 1.6 $)
  *
  * CTC / ESCON network driver - s390 dbf exploit.
  *
@@ -9,7 +9,7 @@
  *    Author(s): Original Code written by
  *			  Peter Tiedemann (ptiedem@de.ibm.com)
  *
- *    $Revision: 1.4 $	 $Date: 2004/08/04 10:11:59 $
+ *    $Revision: 1.6 $	 $Date: 2005/05/11 08:10:17 $
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -51,15 +51,15 @@ int
 ctc_register_dbf_views(void)
 {
 	ctc_dbf_setup = debug_register(CTC_DBF_SETUP_NAME,
-					CTC_DBF_SETUP_INDEX,
+					CTC_DBF_SETUP_PAGES,
 					CTC_DBF_SETUP_NR_AREAS,
 					CTC_DBF_SETUP_LEN);
 	ctc_dbf_data = debug_register(CTC_DBF_DATA_NAME,
-				       CTC_DBF_DATA_INDEX,
+				       CTC_DBF_DATA_PAGES,
 				       CTC_DBF_DATA_NR_AREAS,
 				       CTC_DBF_DATA_LEN);
 	ctc_dbf_trace = debug_register(CTC_DBF_TRACE_NAME,
-					CTC_DBF_TRACE_INDEX,
+					CTC_DBF_TRACE_PAGES,
 					CTC_DBF_TRACE_NR_AREAS,
 					CTC_DBF_TRACE_LEN);
 
diff --git a/drivers/s390/net/ctcdbug.h b/drivers/s390/net/ctcdbug.h
index 7fe2ebd1792..7d6afa1627c 100644
--- a/drivers/s390/net/ctcdbug.h
+++ b/drivers/s390/net/ctcdbug.h
@@ -1,6 +1,6 @@
 /*
  *
- * linux/drivers/s390/net/ctcdbug.h ($Revision: 1.5 $)
+ * linux/drivers/s390/net/ctcdbug.h ($Revision: 1.6 $)
  *
  * CTC / ESCON network driver - s390 dbf exploit.
  *
@@ -9,7 +9,7 @@
  *    Author(s): Original Code written by
  *			  Peter Tiedemann (ptiedem@de.ibm.com)
  *
- *    $Revision: 1.5 $	 $Date: 2005/02/27 19:46:44 $
+ *    $Revision: 1.6 $	 $Date: 2005/05/11 08:10:17 $
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -35,19 +35,19 @@
  */
 #define CTC_DBF_SETUP_NAME "ctc_setup"
 #define CTC_DBF_SETUP_LEN 16
-#define CTC_DBF_SETUP_INDEX 3
+#define CTC_DBF_SETUP_PAGES 8
 #define CTC_DBF_SETUP_NR_AREAS 1
 #define CTC_DBF_SETUP_LEVEL 3
 
 #define CTC_DBF_DATA_NAME "ctc_data"
 #define CTC_DBF_DATA_LEN 128
-#define CTC_DBF_DATA_INDEX 3
+#define CTC_DBF_DATA_PAGES 8
 #define CTC_DBF_DATA_NR_AREAS 1
 #define CTC_DBF_DATA_LEVEL 3
 
 #define CTC_DBF_TRACE_NAME "ctc_trace"
 #define CTC_DBF_TRACE_LEN 16
-#define CTC_DBF_TRACE_INDEX 2
+#define CTC_DBF_TRACE_PAGES 4
 #define CTC_DBF_TRACE_NR_AREAS 2
 #define CTC_DBF_TRACE_LEVEL 3
 
diff --git a/drivers/s390/net/iucv.h b/drivers/s390/net/iucv.h
index 198330217ef..0c4644d3d2f 100644
--- a/drivers/s390/net/iucv.h
+++ b/drivers/s390/net/iucv.h
@@ -37,19 +37,19 @@
  */
 #define IUCV_DBF_SETUP_NAME "iucv_setup"
 #define IUCV_DBF_SETUP_LEN 32
-#define IUCV_DBF_SETUP_INDEX 1
+#define IUCV_DBF_SETUP_PAGES 2
 #define IUCV_DBF_SETUP_NR_AREAS 1
 #define IUCV_DBF_SETUP_LEVEL 3
 
 #define IUCV_DBF_DATA_NAME "iucv_data"
 #define IUCV_DBF_DATA_LEN 128
-#define IUCV_DBF_DATA_INDEX 1
+#define IUCV_DBF_DATA_PAGES 2
 #define IUCV_DBF_DATA_NR_AREAS 1
 #define IUCV_DBF_DATA_LEVEL 2
 
 #define IUCV_DBF_TRACE_NAME "iucv_trace"
 #define IUCV_DBF_TRACE_LEN 16
-#define IUCV_DBF_TRACE_INDEX 2
+#define IUCV_DBF_TRACE_PAGES 4
 #define IUCV_DBF_TRACE_NR_AREAS 1
 #define IUCV_DBF_TRACE_LEVEL 3
 
diff --git a/drivers/s390/net/lcs.c b/drivers/s390/net/lcs.c
index ab086242d30..46f34ba93ac 100644
--- a/drivers/s390/net/lcs.c
+++ b/drivers/s390/net/lcs.c
@@ -11,7 +11,7 @@
  *			  Frank Pavlic (pavlic@de.ibm.com) and
  *		 	  Martin Schwidefsky <schwidefsky@de.ibm.com>
  *
- *    $Revision: 1.98 $	 $Date: 2005/04/18 13:41:29 $
+ *    $Revision: 1.99 $	 $Date: 2005/05/11 08:10:17 $
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -59,7 +59,7 @@
 /**
  * initialization string for output
  */
-#define VERSION_LCS_C  "$Revision: 1.98 $"
+#define VERSION_LCS_C  "$Revision: 1.99 $"
 
 static char version[] __initdata = "LCS driver ("VERSION_LCS_C "/" VERSION_LCS_H ")";
 static char debug_buffer[255];
@@ -93,8 +93,8 @@ lcs_unregister_debug_facility(void)
 static int
 lcs_register_debug_facility(void)
 {
-	lcs_dbf_setup = debug_register("lcs_setup", 1, 1, 8);
-	lcs_dbf_trace = debug_register("lcs_trace", 1, 2, 8);
+	lcs_dbf_setup = debug_register("lcs_setup", 2, 1, 8);
+	lcs_dbf_trace = debug_register("lcs_trace", 2, 2, 8);
 	if (lcs_dbf_setup == NULL || lcs_dbf_trace == NULL) {
 		PRINT_ERR("Not enough memory for debug facility.\n");
 		lcs_unregister_debug_facility();
diff --git a/drivers/s390/net/netiucv.c b/drivers/s390/net/netiucv.c
index 3fd4fb754b2..69425a7a6e9 100644
--- a/drivers/s390/net/netiucv.c
+++ b/drivers/s390/net/netiucv.c
@@ -1,5 +1,5 @@
 /*
- * $Id: netiucv.c,v 1.63 2004/07/27 13:36:05 mschwide Exp $
+ * $Id: netiucv.c,v 1.66 2005/05/11 08:10:17 holzheu Exp $
  *
  * IUCV network driver
  *
@@ -30,7 +30,7 @@
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  *
- * RELEASE-TAG: IUCV network driver $Revision: 1.63 $
+ * RELEASE-TAG: IUCV network driver $Revision: 1.66 $
  *
  */
 
@@ -391,15 +391,15 @@ static int
 iucv_register_dbf_views(void)
 {
 	iucv_dbf_setup = debug_register(IUCV_DBF_SETUP_NAME,
-					IUCV_DBF_SETUP_INDEX,
+					IUCV_DBF_SETUP_PAGES,
 					IUCV_DBF_SETUP_NR_AREAS,
 					IUCV_DBF_SETUP_LEN);
 	iucv_dbf_data = debug_register(IUCV_DBF_DATA_NAME,
-				       IUCV_DBF_DATA_INDEX,
+				       IUCV_DBF_DATA_PAGES,
 				       IUCV_DBF_DATA_NR_AREAS,
 				       IUCV_DBF_DATA_LEN);
 	iucv_dbf_trace = debug_register(IUCV_DBF_TRACE_NAME,
-					IUCV_DBF_TRACE_INDEX,
+					IUCV_DBF_TRACE_PAGES,
 					IUCV_DBF_TRACE_NR_AREAS,
 					IUCV_DBF_TRACE_LEN);
 
@@ -2076,7 +2076,7 @@ DRIVER_ATTR(remove, 0200, NULL, remove_write);
 static void
 netiucv_banner(void)
 {
-	char vbuf[] = "$Revision: 1.63 $";
+	char vbuf[] = "$Revision: 1.66 $";
 	char *version = vbuf;
 
 	if ((version = strchr(version, ':'))) {
diff --git a/drivers/s390/net/qeth.h b/drivers/s390/net/qeth.h
index a755b57db46..008e0a5d2eb 100644
--- a/drivers/s390/net/qeth.h
+++ b/drivers/s390/net/qeth.h
@@ -42,44 +42,44 @@
  */
 #define QETH_DBF_SETUP_NAME "qeth_setup"
 #define QETH_DBF_SETUP_LEN 8
-#define QETH_DBF_SETUP_INDEX 3
+#define QETH_DBF_SETUP_PAGES 8
 #define QETH_DBF_SETUP_NR_AREAS 1
 #define QETH_DBF_SETUP_LEVEL 5
 
 #define QETH_DBF_MISC_NAME "qeth_misc"
 #define QETH_DBF_MISC_LEN 128
-#define QETH_DBF_MISC_INDEX 1
+#define QETH_DBF_MISC_PAGES 2
 #define QETH_DBF_MISC_NR_AREAS 1
 #define QETH_DBF_MISC_LEVEL 2
 
 #define QETH_DBF_DATA_NAME "qeth_data"
 #define QETH_DBF_DATA_LEN 96
-#define QETH_DBF_DATA_INDEX 3
+#define QETH_DBF_DATA_PAGES 8
 #define QETH_DBF_DATA_NR_AREAS 1
 #define QETH_DBF_DATA_LEVEL 2
 
 #define QETH_DBF_CONTROL_NAME "qeth_control"
 #define QETH_DBF_CONTROL_LEN 256
-#define QETH_DBF_CONTROL_INDEX 3
+#define QETH_DBF_CONTROL_PAGES 8
 #define QETH_DBF_CONTROL_NR_AREAS 2
 #define QETH_DBF_CONTROL_LEVEL 5
 
 #define QETH_DBF_TRACE_NAME "qeth_trace"
 #define QETH_DBF_TRACE_LEN 8
-#define QETH_DBF_TRACE_INDEX 2
+#define QETH_DBF_TRACE_PAGES 4
 #define QETH_DBF_TRACE_NR_AREAS 2
 #define QETH_DBF_TRACE_LEVEL 3
 extern debug_info_t *qeth_dbf_trace;
 
 #define QETH_DBF_SENSE_NAME "qeth_sense"
 #define QETH_DBF_SENSE_LEN 64
-#define QETH_DBF_SENSE_INDEX 1
+#define QETH_DBF_SENSE_PAGES 2
 #define QETH_DBF_SENSE_NR_AREAS 1
 #define QETH_DBF_SENSE_LEVEL 2
 
 #define QETH_DBF_QERR_NAME "qeth_qerr"
 #define QETH_DBF_QERR_LEN 8
-#define QETH_DBF_QERR_INDEX 1
+#define QETH_DBF_QERR_PAGES 2
 #define QETH_DBF_QERR_NR_AREAS 2
 #define QETH_DBF_QERR_LEVEL 2
 
diff --git a/drivers/s390/net/qeth_main.c b/drivers/s390/net/qeth_main.c
index 208127a5033..3cb88c77003 100644
--- a/drivers/s390/net/qeth_main.c
+++ b/drivers/s390/net/qeth_main.c
@@ -7639,31 +7639,31 @@ static int
 qeth_register_dbf_views(void)
 {
 	qeth_dbf_setup = debug_register(QETH_DBF_SETUP_NAME,
-					QETH_DBF_SETUP_INDEX,
+					QETH_DBF_SETUP_PAGES,
 					QETH_DBF_SETUP_NR_AREAS,
 					QETH_DBF_SETUP_LEN);
 	qeth_dbf_misc = debug_register(QETH_DBF_MISC_NAME,
-				       QETH_DBF_MISC_INDEX,
+				       QETH_DBF_MISC_PAGES,
 				       QETH_DBF_MISC_NR_AREAS,
 				       QETH_DBF_MISC_LEN);
 	qeth_dbf_data = debug_register(QETH_DBF_DATA_NAME,
-				       QETH_DBF_DATA_INDEX,
+				       QETH_DBF_DATA_PAGES,
 				       QETH_DBF_DATA_NR_AREAS,
 				       QETH_DBF_DATA_LEN);
 	qeth_dbf_control = debug_register(QETH_DBF_CONTROL_NAME,
-					  QETH_DBF_CONTROL_INDEX,
+					  QETH_DBF_CONTROL_PAGES,
 					  QETH_DBF_CONTROL_NR_AREAS,
 					  QETH_DBF_CONTROL_LEN);
 	qeth_dbf_sense = debug_register(QETH_DBF_SENSE_NAME,
-					QETH_DBF_SENSE_INDEX,
+					QETH_DBF_SENSE_PAGES,
 					QETH_DBF_SENSE_NR_AREAS,
 					QETH_DBF_SENSE_LEN);
 	qeth_dbf_qerr = debug_register(QETH_DBF_QERR_NAME,
-				       QETH_DBF_QERR_INDEX,
+				       QETH_DBF_QERR_PAGES,
 				       QETH_DBF_QERR_NR_AREAS,
 				       QETH_DBF_QERR_LEN);
 	qeth_dbf_trace = debug_register(QETH_DBF_TRACE_NAME,
-					QETH_DBF_TRACE_INDEX,
+					QETH_DBF_TRACE_PAGES,
 					QETH_DBF_TRACE_NR_AREAS,
 					QETH_DBF_TRACE_LEN);
 
diff --git a/include/asm-s390/debug.h b/include/asm-s390/debug.h
index 6bbcdea42a8..92360d90144 100644
--- a/include/asm-s390/debug.h
+++ b/include/asm-s390/debug.h
@@ -9,6 +9,8 @@
 #ifndef DEBUG_H
 #define DEBUG_H
 
+#include <linux/config.h>
+#include <linux/fs.h>
 #include <linux/string.h>
 
 /* Note:
@@ -31,19 +33,18 @@ struct __debug_entry{
 } __attribute__((packed));
 
 
-#define __DEBUG_FEATURE_VERSION      1  /* version of debug feature */
+#define __DEBUG_FEATURE_VERSION      2  /* version of debug feature */
 
 #ifdef __KERNEL__
 #include <linux/spinlock.h>
 #include <linux/kernel.h>
 #include <linux/time.h>
-#include <linux/proc_fs.h>
 
 #define DEBUG_MAX_LEVEL            6  /* debug levels range from 0 to 6 */
 #define DEBUG_OFF_LEVEL            -1 /* level where debug is switched off */
 #define DEBUG_FLUSH_ALL            -1 /* parameter to flush all areas */
 #define DEBUG_MAX_VIEWS            10 /* max number of views in proc fs */
-#define DEBUG_MAX_PROCF_LEN        64 /* max length for a proc file name */
+#define DEBUG_MAX_NAME_LEN         64 /* max length for a debugfs file name */
 #define DEBUG_DEFAULT_LEVEL        3  /* initial debug level */
 
 #define DEBUG_DIR_ROOT "s390dbf" /* name of debug root directory in proc fs */
@@ -64,16 +65,17 @@ typedef struct debug_info {
 	spinlock_t lock;			
 	int level;
 	int nr_areas;
-	int page_order;
+	int pages_per_area;
 	int buf_size;
 	int entry_size;	
-	debug_entry_t** areas;
+	debug_entry_t*** areas;
 	int active_area;
-	int *active_entry;
-	struct proc_dir_entry* proc_root_entry;
-	struct proc_dir_entry* proc_entries[DEBUG_MAX_VIEWS];
+	int *active_pages;
+	int *active_entries;
+	struct dentry* debugfs_root_entry;
+	struct dentry* debugfs_entries[DEBUG_MAX_VIEWS];
 	struct debug_view* views[DEBUG_MAX_VIEWS];	
-	char name[DEBUG_MAX_PROCF_LEN];
+	char name[DEBUG_MAX_NAME_LEN];
 } debug_info_t;
 
 typedef int (debug_header_proc_t) (debug_info_t* id,
@@ -98,7 +100,7 @@ int debug_dflt_header_fn(debug_info_t* id, struct debug_view* view,
 		         int area, debug_entry_t* entry, char* out_buf);						
 				
 struct debug_view {
-	char name[DEBUG_MAX_PROCF_LEN];
+	char name[DEBUG_MAX_NAME_LEN];
 	debug_prolog_proc_t* prolog_proc;
 	debug_header_proc_t* header_proc;
 	debug_format_proc_t* format_proc;
@@ -120,7 +122,7 @@ debug_entry_t* debug_exception_common(debug_info_t* id, int level,
 
 /* Debug Feature API: */
 
-debug_info_t* debug_register(char* name, int pages_index, int nr_areas,
+debug_info_t* debug_register(char* name, int pages, int nr_areas,
                              int buf_size);
 
 void debug_unregister(debug_info_t* id);
@@ -132,7 +134,8 @@ void debug_stop_all(void);
 extern inline debug_entry_t* 
 debug_event(debug_info_t* id, int level, void* data, int length)
 {
-	if ((!id) || (level > id->level)) return NULL;
+	if ((!id) || (level > id->level) || (id->pages_per_area == 0))
+		return NULL;
         return debug_event_common(id,level,data,length);
 }
 
@@ -140,7 +143,8 @@ extern inline debug_entry_t*
 debug_int_event(debug_info_t* id, int level, unsigned int tag)
 {
         unsigned int t=tag;
-	if ((!id) || (level > id->level)) return NULL;
+	if ((!id) || (level > id->level) || (id->pages_per_area == 0))
+		return NULL;
         return debug_event_common(id,level,&t,sizeof(unsigned int));
 }
 
@@ -148,14 +152,16 @@ extern inline debug_entry_t *
 debug_long_event (debug_info_t* id, int level, unsigned long tag)
 {
         unsigned long t=tag;
-	if ((!id) || (level > id->level)) return NULL;
+	if ((!id) || (level > id->level) || (id->pages_per_area == 0))
+		return NULL;
         return debug_event_common(id,level,&t,sizeof(unsigned long));
 }
 
 extern inline debug_entry_t* 
 debug_text_event(debug_info_t* id, int level, const char* txt)
 {
-	if ((!id) || (level > id->level)) return NULL;
+	if ((!id) || (level > id->level) || (id->pages_per_area == 0))
+		return NULL;
         return debug_event_common(id,level,txt,strlen(txt));
 }
 
@@ -167,7 +173,8 @@ debug_sprintf_event(debug_info_t* id,int level,char *string,...)
 extern inline debug_entry_t* 
 debug_exception(debug_info_t* id, int level, void* data, int length)
 {
-	if ((!id) || (level > id->level)) return NULL;
+	if ((!id) || (level > id->level) || (id->pages_per_area == 0))
+		return NULL;
         return debug_exception_common(id,level,data,length);
 }
 
@@ -175,7 +182,8 @@ extern inline debug_entry_t*
 debug_int_exception(debug_info_t* id, int level, unsigned int tag)
 {
         unsigned int t=tag;
-	if ((!id) || (level > id->level)) return NULL;
+	if ((!id) || (level > id->level) || (id->pages_per_area == 0))
+		return NULL;
         return debug_exception_common(id,level,&t,sizeof(unsigned int));
 }
 
@@ -183,14 +191,16 @@ extern inline debug_entry_t *
 debug_long_exception (debug_info_t* id, int level, unsigned long tag)
 {
         unsigned long t=tag;
-	if ((!id) || (level > id->level)) return NULL;
+	if ((!id) || (level > id->level) || (id->pages_per_area == 0))
+		return NULL;
         return debug_exception_common(id,level,&t,sizeof(unsigned long));
 }
 
 extern inline debug_entry_t* 
 debug_text_exception(debug_info_t* id, int level, const char* txt)
 {
-	if ((!id) || (level > id->level)) return NULL;
+	if ((!id) || (level > id->level) || (id->pages_per_area == 0))
+		return NULL;
         return debug_exception_common(id,level,txt,strlen(txt));
 }
 
-- 
cgit v1.2.3-70-g09d2


From b2b18660066997420b716c1881a6be8b82700d97 Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@us.ibm.com>
Date: Sat, 25 Jun 2005 14:55:38 -0700
Subject: [PATCH] RCU: clean up a few remaining synchronize_kernel() calls

2.6.12-rc6-mm1 has a few remaining synchronize_kernel()s, some (but not
all) in comments.  This patch changes these synchronize_kernel() calls (and
comments) to synchronize_rcu() or synchronize_sched() as follows:

- arch/x86_64/kernel/mce.c mce_read(): change to synchronize_sched() to
  handle races with machine-check exceptions (synchronize_rcu() would not cut
  it given RCU implementations intended for hardcore realtime use.

- drivers/input/serio/i8042.c i8042_stop(): change to synchronize_sched() to
  handle races with i8042_interrupt() interrupt handler.  Again,
  synchronize_rcu() would not cut it given RCU implementations intended for
  hardcore realtime use.

- include/*/kdebug.h comments: change to synchronize_sched() to handle races
  with NMIs.  As before, synchronize_rcu() would not cut it...

- include/linux/list.h comment: change to synchronize_rcu(), since this
  comment is for list_del_rcu().

- security/keys/key.c unregister_key_type(): change to synchronize_rcu(),
  since this is interacting with RCU read side.

- security/keys/process_keys.c install_session_keyring(): change to
  synchronize_rcu(), since this is interacting with RCU read side.

Signed-off-by: "Paul E. McKenney" <paulmck@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/mce.c     | 2 +-
 drivers/input/serio/i8042.c  | 2 +-
 include/asm-i386/kdebug.h    | 2 +-
 include/asm-ppc64/kdebug.h   | 2 +-
 include/asm-sparc64/kdebug.h | 2 +-
 include/asm-x86_64/kdebug.h  | 2 +-
 include/linux/list.h         | 2 +-
 security/keys/key.c          | 2 +-
 security/keys/process_keys.c | 2 +-
 9 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c
index 7ab15c8ab95..21e70625a49 100644
--- a/arch/x86_64/kernel/mce.c
+++ b/arch/x86_64/kernel/mce.c
@@ -411,7 +411,7 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize, loff
 	memset(mcelog.entry, 0, next * sizeof(struct mce));
 	mcelog.next = 0;
 
-	synchronize_kernel();	
+	synchronize_sched();
 
 	/* Collect entries that were still getting written before the synchronize. */
 
diff --git a/drivers/input/serio/i8042.c b/drivers/input/serio/i8042.c
index 5900de3c3f4..a9bf549c8dc 100644
--- a/drivers/input/serio/i8042.c
+++ b/drivers/input/serio/i8042.c
@@ -396,7 +396,7 @@ static void i8042_stop(struct serio *serio)
 	struct i8042_port *port = serio->port_data;
 
 	port->exists = 0;
-	synchronize_kernel();
+	synchronize_sched();
 	port->serio = NULL;
 }
 
diff --git a/include/asm-i386/kdebug.h b/include/asm-i386/kdebug.h
index de6498b0d49..b3f8d5f59d5 100644
--- a/include/asm-i386/kdebug.h
+++ b/include/asm-i386/kdebug.h
@@ -18,7 +18,7 @@ struct die_args {
 };
 
 /* Note - you should never unregister because that can race with NMIs.
-   If you really want to do it first unregister - then synchronize_kernel - then free.
+   If you really want to do it first unregister - then synchronize_sched - then free.
   */
 int register_die_notifier(struct notifier_block *nb);
 extern struct notifier_block *i386die_chain;
diff --git a/include/asm-ppc64/kdebug.h b/include/asm-ppc64/kdebug.h
index 488634258a7..d383d161cf8 100644
--- a/include/asm-ppc64/kdebug.h
+++ b/include/asm-ppc64/kdebug.h
@@ -17,7 +17,7 @@ struct die_args {
 
 /*
    Note - you should never unregister because that can race with NMIs.
-   If you really want to do it first unregister - then synchronize_kernel -
+   If you really want to do it first unregister - then synchronize_sched -
    then free.
  */
 int register_die_notifier(struct notifier_block *nb);
diff --git a/include/asm-sparc64/kdebug.h b/include/asm-sparc64/kdebug.h
index f70d3dad01f..6321f5a0198 100644
--- a/include/asm-sparc64/kdebug.h
+++ b/include/asm-sparc64/kdebug.h
@@ -16,7 +16,7 @@ struct die_args {
 };
 
 /* Note - you should never unregister because that can race with NMIs.
- * If you really want to do it first unregister - then synchronize_kernel
+ * If you really want to do it first unregister - then synchronize_sched
  * - then free.
  */
 int register_die_notifier(struct notifier_block *nb);
diff --git a/include/asm-x86_64/kdebug.h b/include/asm-x86_64/kdebug.h
index 6277f75cbb4..b90341994d8 100644
--- a/include/asm-x86_64/kdebug.h
+++ b/include/asm-x86_64/kdebug.h
@@ -14,7 +14,7 @@ struct die_args {
 }; 
 
 /* Note - you should never unregister because that can race with NMIs.
-   If you really want to do it first unregister - then synchronize_kernel - then free. 
+   If you really want to do it first unregister - then synchronize_sched - then free.
   */
 int register_die_notifier(struct notifier_block *nb);
 extern struct notifier_block *die_chain;
diff --git a/include/linux/list.h b/include/linux/list.h
index 399b51d1721..aab2db21b01 100644
--- a/include/linux/list.h
+++ b/include/linux/list.h
@@ -185,7 +185,7 @@ static inline void list_del(struct list_head *entry)
  * list_for_each_entry_rcu().
  *
  * Note that the caller is not permitted to immediately free
- * the newly deleted entry.  Instead, either synchronize_kernel()
+ * the newly deleted entry.  Instead, either synchronize_rcu()
  * or call_rcu() must be used to defer freeing until an RCU
  * grace period has elapsed.
  */
diff --git a/security/keys/key.c b/security/keys/key.c
index 3304d37bb37..fb89f984446 100644
--- a/security/keys/key.c
+++ b/security/keys/key.c
@@ -980,7 +980,7 @@ void unregister_key_type(struct key_type *ktype)
 	spin_unlock(&key_serial_lock);
 
 	/* make sure everyone revalidates their keys */
-	synchronize_kernel();
+	synchronize_rcu();
 
 	/* we should now be able to destroy the payloads of all the keys of
 	 * this type with impunity */
diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c
index 34db087bbcc..9b0369c5a22 100644
--- a/security/keys/process_keys.c
+++ b/security/keys/process_keys.c
@@ -234,7 +234,7 @@ static int install_session_keyring(struct task_struct *tsk,
 	ret = 0;
 
 	/* we're using RCU on the pointer */
-	synchronize_kernel();
+	synchronize_rcu();
 	key_put(old);
  error:
 	return ret;
-- 
cgit v1.2.3-70-g09d2


From 7897986bad8f6cd50d6149345aca7f6480f49464 Mon Sep 17 00:00:00 2001
From: Nick Piggin <nickpiggin@yahoo.com.au>
Date: Sat, 25 Jun 2005 14:57:13 -0700
Subject: [PATCH] sched: balance timers

Do CPU load averaging over a number of different intervals.  Allow each
interval to be chosen by sending a parameter to source_load and target_load.
0 is instantaneous, idx > 0 returns a decaying average with the most recent
sample weighted at 2^(idx-1).  To a maximum of 3 (could be easily increased).

So generally a higher number will result in more conservative balancing.

Signed-off-by: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/topology.h   |   4 ++
 include/asm-x86_64/topology.h |   6 +-
 include/linux/sched.h         |   4 ++
 include/linux/topology.h      |   8 +++
 kernel/sched.c                | 138 ++++++++++++++++++++++--------------------
 5 files changed, 95 insertions(+), 65 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/topology.h b/include/asm-i386/topology.h
index 6d0f67507b2..0055fbfeec7 100644
--- a/include/asm-i386/topology.h
+++ b/include/asm-i386/topology.h
@@ -74,6 +74,10 @@ static inline int node_to_first_cpu(int node)
 	.imbalance_pct		= 125,			\
 	.cache_hot_time		= (10*1000000),		\
 	.cache_nice_tries	= 1,			\
+	.busy_idx		= 3,			\
+	.idle_idx		= 1,			\
+	.newidle_idx		= 2,			\
+	.wake_idx		= 1,			\
 	.per_cpu_gain		= 100,			\
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_EXEC	\
diff --git a/include/asm-x86_64/topology.h b/include/asm-x86_64/topology.h
index 8f77e9f6bc2..fe8d80a1575 100644
--- a/include/asm-x86_64/topology.h
+++ b/include/asm-x86_64/topology.h
@@ -39,7 +39,11 @@ extern int __node_distance(int, int);
 	.busy_factor		= 32,			\
 	.imbalance_pct		= 125,			\
 	.cache_hot_time		= (10*1000000),		\
-	.cache_nice_tries	= 1,			\
+	.cache_nice_tries	= 2,			\
+	.busy_idx		= 3,			\
+	.idle_idx		= 2,			\
+	.newidle_idx		= 1, 			\
+	.wake_idx		= 1,			\
 	.per_cpu_gain		= 100,			\
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_NEWIDLE	\
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2c69682b044..664981ac1fb 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -488,6 +488,10 @@ struct sched_domain {
 	unsigned long long cache_hot_time; /* Task considered cache hot (ns) */
 	unsigned int cache_nice_tries;	/* Leave cache hot tasks for # tries */
 	unsigned int per_cpu_gain;	/* CPU % gained by adding domain cpus */
+	unsigned int busy_idx;
+	unsigned int idle_idx;
+	unsigned int newidle_idx;
+	unsigned int wake_idx;
 	int flags;			/* See SD_* */
 
 	/* Runtime fields. */
diff --git a/include/linux/topology.h b/include/linux/topology.h
index d70e8972c67..ae9c2216dfa 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -89,6 +89,10 @@
 	.cache_hot_time		= 0,			\
 	.cache_nice_tries	= 0,			\
 	.per_cpu_gain		= 25,			\
+	.busy_idx		= 0,			\
+	.idle_idx		= 0,			\
+	.newidle_idx		= 0,			\
+	.wake_idx		= 0,			\
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_NEWIDLE	\
 				| SD_BALANCE_EXEC	\
@@ -115,6 +119,10 @@
 	.cache_hot_time		= (5*1000000/2),	\
 	.cache_nice_tries	= 1,			\
 	.per_cpu_gain		= 100,			\
+	.busy_idx		= 2,			\
+	.idle_idx		= 0,			\
+	.newidle_idx		= 1,			\
+	.wake_idx		= 1,			\
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_NEWIDLE	\
 				| SD_BALANCE_EXEC	\
diff --git a/kernel/sched.c b/kernel/sched.c
index f665de34ed8..b597b07e791 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -206,7 +206,7 @@ struct runqueue {
 	 */
 	unsigned long nr_running;
 #ifdef CONFIG_SMP
-	unsigned long cpu_load;
+	unsigned long cpu_load[3];
 #endif
 	unsigned long long nr_switches;
 
@@ -886,23 +886,27 @@ void kick_process(task_t *p)
  * We want to under-estimate the load of migration sources, to
  * balance conservatively.
  */
-static inline unsigned long source_load(int cpu)
+static inline unsigned long source_load(int cpu, int type)
 {
 	runqueue_t *rq = cpu_rq(cpu);
 	unsigned long load_now = rq->nr_running * SCHED_LOAD_SCALE;
+	if (type == 0)
+		return load_now;
 
-	return min(rq->cpu_load, load_now);
+	return min(rq->cpu_load[type-1], load_now);
 }
 
 /*
  * Return a high guess at the load of a migration-target cpu
  */
-static inline unsigned long target_load(int cpu)
+static inline unsigned long target_load(int cpu, int type)
 {
 	runqueue_t *rq = cpu_rq(cpu);
 	unsigned long load_now = rq->nr_running * SCHED_LOAD_SCALE;
+	if (type == 0)
+		return load_now;
 
-	return max(rq->cpu_load, load_now);
+	return max(rq->cpu_load[type-1], load_now);
 }
 
 #endif
@@ -967,7 +971,7 @@ static int try_to_wake_up(task_t * p, unsigned int state, int sync)
 	runqueue_t *rq;
 #ifdef CONFIG_SMP
 	unsigned long load, this_load;
-	struct sched_domain *sd;
+	struct sched_domain *sd, *this_sd = NULL;
 	int new_cpu;
 #endif
 
@@ -986,72 +990,64 @@ static int try_to_wake_up(task_t * p, unsigned int state, int sync)
 	if (unlikely(task_running(rq, p)))
 		goto out_activate;
 
-#ifdef CONFIG_SCHEDSTATS
+	new_cpu = cpu;
+
 	schedstat_inc(rq, ttwu_cnt);
 	if (cpu == this_cpu) {
 		schedstat_inc(rq, ttwu_local);
-	} else {
-		for_each_domain(this_cpu, sd) {
-			if (cpu_isset(cpu, sd->span)) {
-				schedstat_inc(sd, ttwu_wake_remote);
-				break;
-			}
+		goto out_set_cpu;
+	}
+
+	for_each_domain(this_cpu, sd) {
+		if (cpu_isset(cpu, sd->span)) {
+			schedstat_inc(sd, ttwu_wake_remote);
+			this_sd = sd;
+			break;
 		}
 	}
-#endif
 
-	new_cpu = cpu;
-	if (cpu == this_cpu || unlikely(!cpu_isset(this_cpu, p->cpus_allowed)))
+	if (unlikely(!cpu_isset(this_cpu, p->cpus_allowed)))
 		goto out_set_cpu;
 
-	load = source_load(cpu);
-	this_load = target_load(this_cpu);
-
 	/*
-	 * If sync wakeup then subtract the (maximum possible) effect of
-	 * the currently running task from the load of the current CPU:
+	 * Check for affine wakeup and passive balancing possibilities.
 	 */
-	if (sync)
-		this_load -= SCHED_LOAD_SCALE;
-
-	/* Don't pull the task off an idle CPU to a busy one */
-	if (load < SCHED_LOAD_SCALE/2 && this_load > SCHED_LOAD_SCALE/2)
-		goto out_set_cpu;
+	if (this_sd) {
+		int idx = this_sd->wake_idx;
+		unsigned int imbalance;
 
-	new_cpu = this_cpu; /* Wake to this CPU if we can */
+		load = source_load(cpu, idx);
+		this_load = target_load(this_cpu, idx);
 
-	/*
-	 * Scan domains for affine wakeup and passive balancing
-	 * possibilities.
-	 */
-	for_each_domain(this_cpu, sd) {
-		unsigned int imbalance;
 		/*
-		 * Start passive balancing when half the imbalance_pct
-		 * limit is reached.
+		 * If sync wakeup then subtract the (maximum possible) effect of
+		 * the currently running task from the load of the current CPU:
 		 */
-		imbalance = sd->imbalance_pct + (sd->imbalance_pct - 100) / 2;
+		if (sync)
+			this_load -= SCHED_LOAD_SCALE;
+
+		 /* Don't pull the task off an idle CPU to a busy one */
+		if (load < SCHED_LOAD_SCALE/2 && this_load > SCHED_LOAD_SCALE/2)
+			goto out_set_cpu;
 
-		if ((sd->flags & SD_WAKE_AFFINE) &&
-				!task_hot(p, rq->timestamp_last_tick, sd)) {
+		new_cpu = this_cpu; /* Wake to this CPU if we can */
+
+		if ((this_sd->flags & SD_WAKE_AFFINE) &&
+			!task_hot(p, rq->timestamp_last_tick, this_sd)) {
 			/*
 			 * This domain has SD_WAKE_AFFINE and p is cache cold
 			 * in this domain.
 			 */
-			if (cpu_isset(cpu, sd->span)) {
-				schedstat_inc(sd, ttwu_move_affine);
-				goto out_set_cpu;
-			}
-		} else if ((sd->flags & SD_WAKE_BALANCE) &&
+			schedstat_inc(this_sd, ttwu_move_affine);
+			goto out_set_cpu;
+		} else if ((this_sd->flags & SD_WAKE_BALANCE) &&
 				imbalance*this_load <= 100*load) {
 			/*
 			 * This domain has SD_WAKE_BALANCE and there is
 			 * an imbalance.
 			 */
-			if (cpu_isset(cpu, sd->span)) {
-				schedstat_inc(sd, ttwu_move_balance);
-				goto out_set_cpu;
-			}
+			schedstat_inc(this_sd, ttwu_move_balance);
+			goto out_set_cpu;
 		}
 	}
 
@@ -1509,7 +1505,7 @@ static int find_idlest_cpu(struct task_struct *p, int this_cpu,
 	cpus_and(mask, sd->span, p->cpus_allowed);
 
 	for_each_cpu_mask(i, mask) {
-		load = target_load(i);
+		load = target_load(i, sd->wake_idx);
 
 		if (load < min_load) {
 			min_cpu = i;
@@ -1522,7 +1518,7 @@ static int find_idlest_cpu(struct task_struct *p, int this_cpu,
 	}
 
 	/* add +1 to account for the new task */
-	this_load = source_load(this_cpu) + SCHED_LOAD_SCALE;
+	this_load = source_load(this_cpu, sd->wake_idx) + SCHED_LOAD_SCALE;
 
 	/*
 	 * Would with the addition of the new task to the
@@ -1767,8 +1763,15 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 {
 	struct sched_group *busiest = NULL, *this = NULL, *group = sd->groups;
 	unsigned long max_load, avg_load, total_load, this_load, total_pwr;
+	int load_idx;
 
 	max_load = this_load = total_load = total_pwr = 0;
+	if (idle == NOT_IDLE)
+		load_idx = sd->busy_idx;
+	else if (idle == NEWLY_IDLE)
+		load_idx = sd->newidle_idx;
+	else
+		load_idx = sd->idle_idx;
 
 	do {
 		unsigned long load;
@@ -1783,9 +1786,9 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 		for_each_cpu_mask(i, group->cpumask) {
 			/* Bias balancing toward cpus of our domain */
 			if (local_group)
-				load = target_load(i);
+				load = target_load(i, load_idx);
 			else
-				load = source_load(i);
+				load = source_load(i, load_idx);
 
 			avg_load += load;
 		}
@@ -1895,7 +1898,7 @@ static runqueue_t *find_busiest_queue(struct sched_group *group)
 	int i;
 
 	for_each_cpu_mask(i, group->cpumask) {
-		load = source_load(i);
+		load = source_load(i, 0);
 
 		if (load > max_load) {
 			max_load = load;
@@ -2150,18 +2153,23 @@ static void rebalance_tick(int this_cpu, runqueue_t *this_rq,
 	unsigned long old_load, this_load;
 	unsigned long j = jiffies + CPU_OFFSET(this_cpu);
 	struct sched_domain *sd;
+	int i;
 
-	/* Update our load */
-	old_load = this_rq->cpu_load;
 	this_load = this_rq->nr_running * SCHED_LOAD_SCALE;
-	/*
-	 * Round up the averaging division if load is increasing. This
-	 * prevents us from getting stuck on 9 if the load is 10, for
-	 * example.
-	 */
-	if (this_load > old_load)
-		old_load++;
-	this_rq->cpu_load = (old_load + this_load) / 2;
+	/* Update our load */
+	for (i = 0; i < 3; i++) {
+		unsigned long new_load = this_load;
+		int scale = 1 << i;
+		old_load = this_rq->cpu_load[i];
+		/*
+		 * Round up the averaging division if load is increasing. This
+		 * prevents us from getting stuck on 9 if the load is 10, for
+		 * example.
+		 */
+		if (new_load > old_load)
+			new_load += scale-1;
+		this_rq->cpu_load[i] = (old_load*(scale-1) + new_load) / scale;
+	}
 
 	for_each_domain(this_cpu, sd) {
 		unsigned long interval;
@@ -4921,13 +4929,15 @@ void __init sched_init(void)
 
 		rq = cpu_rq(i);
 		spin_lock_init(&rq->lock);
+		rq->nr_running = 0;
 		rq->active = rq->arrays;
 		rq->expired = rq->arrays + 1;
 		rq->best_expired_prio = MAX_PRIO;
 
 #ifdef CONFIG_SMP
 		rq->sd = &sched_domain_dummy;
-		rq->cpu_load = 0;
+		for (j = 1; j < 3; j++)
+			rq->cpu_load[j] = 0;
 		rq->active_balance = 0;
 		rq->push_cpu = 0;
 		rq->migration_thread = NULL;
-- 
cgit v1.2.3-70-g09d2


From cafb20c1f9976a70d633bb1e1c8c24eab00e4e80 Mon Sep 17 00:00:00 2001
From: Nick Piggin <nickpiggin@yahoo.com.au>
Date: Sat, 25 Jun 2005 14:57:17 -0700
Subject: [PATCH] sched: no aggressive idle balancing

Remove the very aggressive idle stuff that has recently gone into 2.6 - it is
going against the direction we are trying to go.  Hopefully we can regain
performance through other methods.

Signed-off-by: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/topology.h   |  1 -
 include/asm-x86_64/topology.h |  1 -
 include/linux/topology.h      |  1 -
 kernel/sched.c                | 21 ++-------------------
 4 files changed, 2 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/topology.h b/include/asm-i386/topology.h
index 0055fbfeec7..5eb6f61dcef 100644
--- a/include/asm-i386/topology.h
+++ b/include/asm-i386/topology.h
@@ -82,7 +82,6 @@ static inline int node_to_first_cpu(int node)
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_EXEC	\
 				| SD_BALANCE_NEWIDLE	\
-				| SD_WAKE_IDLE		\
 				| SD_WAKE_BALANCE,	\
 	.last_balance		= jiffies,		\
 	.balance_interval	= 1,			\
diff --git a/include/asm-x86_64/topology.h b/include/asm-x86_64/topology.h
index fe8d80a1575..9cb7459ce72 100644
--- a/include/asm-x86_64/topology.h
+++ b/include/asm-x86_64/topology.h
@@ -48,7 +48,6 @@ extern int __node_distance(int, int);
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_NEWIDLE	\
 				| SD_BALANCE_EXEC	\
-				| SD_WAKE_IDLE		\
 				| SD_WAKE_BALANCE,	\
 	.last_balance		= jiffies,		\
 	.balance_interval	= 1,			\
diff --git a/include/linux/topology.h b/include/linux/topology.h
index ae9c2216dfa..b23ec64df7f 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -127,7 +127,6 @@
 				| SD_BALANCE_NEWIDLE	\
 				| SD_BALANCE_EXEC	\
 				| SD_WAKE_AFFINE	\
-				| SD_WAKE_IDLE		\
 				| SD_WAKE_BALANCE,	\
 	.last_balance		= jiffies,		\
 	.balance_interval	= 1,			\
diff --git a/kernel/sched.c b/kernel/sched.c
index 5ae3568eed0..396724a2519 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -414,22 +414,6 @@ static inline runqueue_t *this_rq_lock(void)
 	return rq;
 }
 
-#ifdef CONFIG_SCHED_SMT
-static int cpu_and_siblings_are_idle(int cpu)
-{
-	int sib;
-	for_each_cpu_mask(sib, cpu_sibling_map[cpu]) {
-		if (idle_cpu(sib))
-			continue;
-		return 0;
-	}
-
-	return 1;
-}
-#else
-#define cpu_and_siblings_are_idle(A) idle_cpu(A)
-#endif
-
 #ifdef CONFIG_SCHEDSTATS
 /*
  * Called when a process is dequeued from the active array and given
@@ -1652,12 +1636,11 @@ int can_migrate_task(task_t *p, runqueue_t *rq, int this_cpu,
 
 	/*
 	 * Aggressive migration if:
-	 * 1) the [whole] cpu is idle, or
+	 * 1) task is cache cold, or
 	 * 2) too many balance attempts have failed.
 	 */
 
-	if (cpu_and_siblings_are_idle(this_cpu) || \
-			sd->nr_balance_failed > sd->cache_nice_tries)
+	if (sd->nr_balance_failed > sd->cache_nice_tries)
 		return 1;
 
 	if (task_hot(p, rq->timestamp_last_tick, sd))
-- 
cgit v1.2.3-70-g09d2


From 147cbb4bbe991452698f0772d8292f22825710ba Mon Sep 17 00:00:00 2001
From: Nick Piggin <nickpiggin@yahoo.com.au>
Date: Sat, 25 Jun 2005 14:57:19 -0700
Subject: [PATCH] sched: balance on fork

Reimplement the balance on exec balancing to be sched-domains aware.  Use this
to also do balance on fork balancing.  Make x86_64 do balance on fork over the
NUMA domain.

The problem that the non sched domains aware blancing became apparent on dual
core, multi socket opterons.  What we want is for the new tasks to be sent to
a different socket, but more often than not, we would first load up our
sibling core, or fill two cores of a single remote socket before selecting a
new one.

This gives large improvements to STREAM on such systems.

Signed-off-by: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-x86_64/topology.h |   2 +
 include/linux/sched.h         |  10 +--
 include/linux/topology.h      |   2 +
 kernel/sched.c                | 164 ++++++++++++++++++++++++++++--------------
 4 files changed, 119 insertions(+), 59 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86_64/topology.h b/include/asm-x86_64/topology.h
index 9cb7459ce72..802d09b9c99 100644
--- a/include/asm-x86_64/topology.h
+++ b/include/asm-x86_64/topology.h
@@ -44,9 +44,11 @@ extern int __node_distance(int, int);
 	.idle_idx		= 2,			\
 	.newidle_idx		= 1, 			\
 	.wake_idx		= 1,			\
+	.forkexec_idx		= 1,			\
 	.per_cpu_gain		= 100,			\
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_NEWIDLE	\
+				| SD_BALANCE_FORK	\
 				| SD_BALANCE_EXEC	\
 				| SD_WAKE_BALANCE,	\
 	.last_balance		= jiffies,		\
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 664981ac1fb..613491d3a87 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -460,10 +460,11 @@ enum idle_type
 #define SD_LOAD_BALANCE		1	/* Do load balancing on this domain. */
 #define SD_BALANCE_NEWIDLE	2	/* Balance when about to become idle */
 #define SD_BALANCE_EXEC		4	/* Balance on exec */
-#define SD_WAKE_IDLE		8	/* Wake to idle CPU on task wakeup */
-#define SD_WAKE_AFFINE		16	/* Wake task to waking CPU */
-#define SD_WAKE_BALANCE		32	/* Perform balancing at task wakeup */
-#define SD_SHARE_CPUPOWER	64	/* Domain members share cpu power */
+#define SD_BALANCE_FORK		8	/* Balance on fork, clone */
+#define SD_WAKE_IDLE		16	/* Wake to idle CPU on task wakeup */
+#define SD_WAKE_AFFINE		32	/* Wake task to waking CPU */
+#define SD_WAKE_BALANCE		64	/* Perform balancing at task wakeup */
+#define SD_SHARE_CPUPOWER	128	/* Domain members share cpu power */
 
 struct sched_group {
 	struct sched_group *next;	/* Must be a circular list */
@@ -492,6 +493,7 @@ struct sched_domain {
 	unsigned int idle_idx;
 	unsigned int newidle_idx;
 	unsigned int wake_idx;
+	unsigned int forkexec_idx;
 	int flags;			/* See SD_* */
 
 	/* Runtime fields. */
diff --git a/include/linux/topology.h b/include/linux/topology.h
index b23ec64df7f..665597207de 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -93,6 +93,7 @@
 	.idle_idx		= 0,			\
 	.newidle_idx		= 0,			\
 	.wake_idx		= 0,			\
+	.forkexec_idx		= 0,			\
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_NEWIDLE	\
 				| SD_BALANCE_EXEC	\
@@ -123,6 +124,7 @@
 	.idle_idx		= 0,			\
 	.newidle_idx		= 1,			\
 	.wake_idx		= 1,			\
+	.forkexec_idx		= 0,			\
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_NEWIDLE	\
 				| SD_BALANCE_EXEC	\
diff --git a/kernel/sched.c b/kernel/sched.c
index 396724a2519..7ecc237e2aa 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -893,6 +893,79 @@ static inline unsigned long target_load(int cpu, int type)
 	return max(rq->cpu_load[type-1], load_now);
 }
 
+/*
+ * find_idlest_group finds and returns the least busy CPU group within the
+ * domain.
+ */
+static struct sched_group *
+find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)
+{
+	struct sched_group *idlest = NULL, *this = NULL, *group = sd->groups;
+	unsigned long min_load = ULONG_MAX, this_load = 0;
+	int load_idx = sd->forkexec_idx;
+	int imbalance = 100 + (sd->imbalance_pct-100)/2;
+
+	do {
+		unsigned long load, avg_load;
+		int local_group;
+		int i;
+
+		local_group = cpu_isset(this_cpu, group->cpumask);
+		/* XXX: put a cpus allowed check */
+
+		/* Tally up the load of all CPUs in the group */
+		avg_load = 0;
+
+		for_each_cpu_mask(i, group->cpumask) {
+			/* Bias balancing toward cpus of our domain */
+			if (local_group)
+				load = source_load(i, load_idx);
+			else
+				load = target_load(i, load_idx);
+
+			avg_load += load;
+		}
+
+		/* Adjust by relative CPU power of the group */
+		avg_load = (avg_load * SCHED_LOAD_SCALE) / group->cpu_power;
+
+		if (local_group) {
+			this_load = avg_load;
+			this = group;
+		} else if (avg_load < min_load) {
+			min_load = avg_load;
+			idlest = group;
+		}
+		group = group->next;
+	} while (group != sd->groups);
+
+	if (!idlest || 100*this_load < imbalance*min_load)
+		return NULL;
+	return idlest;
+}
+
+/*
+ * find_idlest_queue - find the idlest runqueue among the cpus in group.
+ */
+static int find_idlest_cpu(struct sched_group *group, int this_cpu)
+{
+	unsigned long load, min_load = ULONG_MAX;
+	int idlest = -1;
+	int i;
+
+	for_each_cpu_mask(i, group->cpumask) {
+		load = source_load(i, 0);
+
+		if (load < min_load || (load == min_load && i == this_cpu)) {
+			min_load = load;
+			idlest = i;
+		}
+	}
+
+	return idlest;
+}
+
+
 #endif
 
 /*
@@ -1107,11 +1180,6 @@ int fastcall wake_up_state(task_t *p, unsigned int state)
 	return try_to_wake_up(p, state, 0);
 }
 
-#ifdef CONFIG_SMP
-static int find_idlest_cpu(struct task_struct *p, int this_cpu,
-			   struct sched_domain *sd);
-#endif
-
 /*
  * Perform scheduler related setup for a newly forked process p.
  * p is forked by current.
@@ -1181,12 +1249,38 @@ void fastcall wake_up_new_task(task_t * p, unsigned long clone_flags)
 	unsigned long flags;
 	int this_cpu, cpu;
 	runqueue_t *rq, *this_rq;
+#ifdef CONFIG_SMP
+	struct sched_domain *tmp, *sd = NULL;
+#endif
 
 	rq = task_rq_lock(p, &flags);
-	cpu = task_cpu(p);
+	BUG_ON(p->state != TASK_RUNNING);
 	this_cpu = smp_processor_id();
+	cpu = task_cpu(p);
 
-	BUG_ON(p->state != TASK_RUNNING);
+#ifdef CONFIG_SMP
+	for_each_domain(cpu, tmp)
+		if (tmp->flags & SD_BALANCE_FORK)
+			sd = tmp;
+
+	if (sd) {
+		struct sched_group *group;
+
+		cpu = task_cpu(p);
+		group = find_idlest_group(sd, p, cpu);
+		if (group) {
+			int new_cpu;
+			new_cpu = find_idlest_cpu(group, cpu);
+			if (new_cpu != -1 && new_cpu != cpu &&
+					cpu_isset(new_cpu, p->cpus_allowed)) {
+				set_task_cpu(p, new_cpu);
+				task_rq_unlock(rq, &flags);
+				rq = task_rq_lock(p, &flags);
+				cpu = task_cpu(p);
+			}
+		}
+	}
+#endif
 
 	/*
 	 * We decrease the sleep average of forking parents
@@ -1480,51 +1574,6 @@ static void double_lock_balance(runqueue_t *this_rq, runqueue_t *busiest)
 	}
 }
 
-/*
- * find_idlest_cpu - find the least busy runqueue.
- */
-static int find_idlest_cpu(struct task_struct *p, int this_cpu,
-			   struct sched_domain *sd)
-{
-	unsigned long load, min_load, this_load;
-	int i, min_cpu;
-	cpumask_t mask;
-
-	min_cpu = UINT_MAX;
-	min_load = ULONG_MAX;
-
-	cpus_and(mask, sd->span, p->cpus_allowed);
-
-	for_each_cpu_mask(i, mask) {
-		load = target_load(i, sd->wake_idx);
-
-		if (load < min_load) {
-			min_cpu = i;
-			min_load = load;
-
-			/* break out early on an idle CPU: */
-			if (!min_load)
-				break;
-		}
-	}
-
-	/* add +1 to account for the new task */
-	this_load = source_load(this_cpu, sd->wake_idx) + SCHED_LOAD_SCALE;
-
-	/*
-	 * Would with the addition of the new task to the
-	 * current CPU there be an imbalance between this
-	 * CPU and the idlest CPU?
-	 *
-	 * Use half of the balancing threshold - new-context is
-	 * a good opportunity to balance.
-	 */
-	if (min_load*(100 + (sd->imbalance_pct-100)/2) < this_load*100)
-		return min_cpu;
-
-	return this_cpu;
-}
-
 /*
  * If dest_cpu is allowed for this process, migrate the task to it.
  * This is accomplished by forcing the cpu_allowed mask to only
@@ -1578,8 +1627,15 @@ void sched_exec(void)
 			sd = tmp;
 
 	if (sd) {
+		struct sched_group *group;
 		schedstat_inc(sd, sbe_attempts);
-		new_cpu = find_idlest_cpu(current, this_cpu, sd);
+		group = find_idlest_group(sd, current, this_cpu);
+		if (!group)
+			goto out;
+		new_cpu = find_idlest_cpu(group, this_cpu);
+		if (new_cpu == -1)
+			goto out;
+
 		if (new_cpu != this_cpu) {
 			schedstat_inc(sd, sbe_pushed);
 			put_cpu();
@@ -1792,12 +1848,10 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 		if (local_group) {
 			this_load = avg_load;
 			this = group;
-			goto nextgroup;
 		} else if (avg_load > max_load) {
 			max_load = avg_load;
 			busiest = group;
 		}
-nextgroup:
 		group = group->next;
 	} while (group != sd->groups);
 
-- 
cgit v1.2.3-70-g09d2


From 68767a0ae428801649d510d9a65bb71feed44dd1 Mon Sep 17 00:00:00 2001
From: Nick Piggin <nickpiggin@yahoo.com.au>
Date: Sat, 25 Jun 2005 14:57:20 -0700
Subject: [PATCH] sched: schedstats update for balance on fork

Add SCHEDSTAT statistics for sched-balance-fork.

Signed-off-by: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/sched.h | 10 ++++++--
 kernel/sched.c        | 63 +++++++++++++++++++++++++++++----------------------
 2 files changed, 44 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 613491d3a87..36a10781c3f 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -517,10 +517,16 @@ struct sched_domain {
 	unsigned long alb_failed;
 	unsigned long alb_pushed;
 
-	/* sched_balance_exec() stats */
-	unsigned long sbe_attempts;
+	/* SD_BALANCE_EXEC stats */
+	unsigned long sbe_cnt;
+	unsigned long sbe_balanced;
 	unsigned long sbe_pushed;
 
+	/* SD_BALANCE_FORK stats */
+	unsigned long sbf_cnt;
+	unsigned long sbf_balanced;
+	unsigned long sbf_pushed;
+
 	/* try_to_wake_up() stats */
 	unsigned long ttwu_wake_remote;
 	unsigned long ttwu_move_affine;
diff --git a/kernel/sched.c b/kernel/sched.c
index 7ecc237e2aa..2711130cd97 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -309,7 +309,7 @@ static inline void task_rq_unlock(runqueue_t *rq, unsigned long *flags)
  * bump this up when changing the output format or the meaning of an existing
  * format, so that tools can adapt (or abort)
  */
-#define SCHEDSTAT_VERSION 11
+#define SCHEDSTAT_VERSION 12
 
 static int show_schedstat(struct seq_file *seq, void *v)
 {
@@ -356,9 +356,10 @@ static int show_schedstat(struct seq_file *seq, void *v)
 				    sd->lb_nobusyq[itype],
 				    sd->lb_nobusyg[itype]);
 			}
-			seq_printf(seq, " %lu %lu %lu %lu %lu %lu %lu %lu\n",
+			seq_printf(seq, " %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu\n",
 			    sd->alb_cnt, sd->alb_failed, sd->alb_pushed,
-			    sd->sbe_pushed, sd->sbe_attempts,
+			    sd->sbe_cnt, sd->sbe_balanced, sd->sbe_pushed,
+			    sd->sbf_cnt, sd->sbf_balanced, sd->sbf_pushed,
 			    sd->ttwu_wake_remote, sd->ttwu_move_affine, sd->ttwu_move_balance);
 		}
 #endif
@@ -1264,24 +1265,34 @@ void fastcall wake_up_new_task(task_t * p, unsigned long clone_flags)
 			sd = tmp;
 
 	if (sd) {
+		int new_cpu;
 		struct sched_group *group;
 
+		schedstat_inc(sd, sbf_cnt);
 		cpu = task_cpu(p);
 		group = find_idlest_group(sd, p, cpu);
-		if (group) {
-			int new_cpu;
-			new_cpu = find_idlest_cpu(group, cpu);
-			if (new_cpu != -1 && new_cpu != cpu &&
-					cpu_isset(new_cpu, p->cpus_allowed)) {
-				set_task_cpu(p, new_cpu);
-				task_rq_unlock(rq, &flags);
-				rq = task_rq_lock(p, &flags);
-				cpu = task_cpu(p);
-			}
+		if (!group) {
+			schedstat_inc(sd, sbf_balanced);
+			goto no_forkbalance;
+		}
+
+		new_cpu = find_idlest_cpu(group, cpu);
+		if (new_cpu == -1 || new_cpu == cpu) {
+			schedstat_inc(sd, sbf_balanced);
+			goto no_forkbalance;
+		}
+
+		if (cpu_isset(new_cpu, p->cpus_allowed)) {
+			schedstat_inc(sd, sbf_pushed);
+			set_task_cpu(p, new_cpu);
+			task_rq_unlock(rq, &flags);
+			rq = task_rq_lock(p, &flags);
+			cpu = task_cpu(p);
 		}
 	}
-#endif
 
+no_forkbalance:
+#endif
 	/*
 	 * We decrease the sleep average of forking parents
 	 * and children as well, to keep max-interactive tasks
@@ -1618,30 +1629,28 @@ void sched_exec(void)
 	struct sched_domain *tmp, *sd = NULL;
 	int new_cpu, this_cpu = get_cpu();
 
-	/* Prefer the current CPU if there's only this task running */
-	if (this_rq()->nr_running <= 1)
-		goto out;
-
 	for_each_domain(this_cpu, tmp)
 		if (tmp->flags & SD_BALANCE_EXEC)
 			sd = tmp;
 
 	if (sd) {
 		struct sched_group *group;
-		schedstat_inc(sd, sbe_attempts);
+		schedstat_inc(sd, sbe_cnt);
 		group = find_idlest_group(sd, current, this_cpu);
-		if (!group)
+		if (!group) {
+			schedstat_inc(sd, sbe_balanced);
 			goto out;
+		}
 		new_cpu = find_idlest_cpu(group, this_cpu);
-		if (new_cpu == -1)
+		if (new_cpu == -1 || new_cpu == this_cpu) {
+			schedstat_inc(sd, sbe_balanced);
 			goto out;
-
-		if (new_cpu != this_cpu) {
-			schedstat_inc(sd, sbe_pushed);
-			put_cpu();
-			sched_migrate_task(current, new_cpu);
-			return;
 		}
+
+		schedstat_inc(sd, sbe_pushed);
+		put_cpu();
+		sched_migrate_task(current, new_cpu);
+		return;
 	}
 out:
 	put_cpu();
-- 
cgit v1.2.3-70-g09d2


From 687f1661d302bc70ce906594a6d3f615ef075a50 Mon Sep 17 00:00:00 2001
From: Nick Piggin <nickpiggin@yahoo.com.au>
Date: Sat, 25 Jun 2005 14:57:21 -0700
Subject: [PATCH] sched: sched tuning

Do some basic initial tuning.

Signed-off-by: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/domain.c     |  2 +-
 include/asm-i386/topology.h   |  2 +-
 include/asm-ia64/topology.h   | 61 +++++++++++++++++++++++++++++++++----------
 include/asm-x86_64/topology.h |  3 +--
 include/linux/topology.h      | 11 ++++----
 5 files changed, 55 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/domain.c b/arch/ia64/kernel/domain.c
index fe532c97043..afbde79c3b3 100644
--- a/arch/ia64/kernel/domain.c
+++ b/arch/ia64/kernel/domain.c
@@ -14,7 +14,7 @@
 #include <linux/topology.h>
 #include <linux/nodemask.h>
 
-#define SD_NODES_PER_DOMAIN 6
+#define SD_NODES_PER_DOMAIN 16
 
 #ifdef CONFIG_NUMA
 /**
diff --git a/include/asm-i386/topology.h b/include/asm-i386/topology.h
index 5eb6f61dcef..2461b731781 100644
--- a/include/asm-i386/topology.h
+++ b/include/asm-i386/topology.h
@@ -81,7 +81,7 @@ static inline int node_to_first_cpu(int node)
 	.per_cpu_gain		= 100,			\
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_EXEC	\
-				| SD_BALANCE_NEWIDLE	\
+				| SD_BALANCE_FORK	\
 				| SD_WAKE_BALANCE,	\
 	.last_balance		= jiffies,		\
 	.balance_interval	= 1,			\
diff --git a/include/asm-ia64/topology.h b/include/asm-ia64/topology.h
index 21cf351fd05..4e64c2a6b36 100644
--- a/include/asm-ia64/topology.h
+++ b/include/asm-ia64/topology.h
@@ -42,25 +42,54 @@
 
 void build_cpu_to_node_map(void);
 
+#define SD_CPU_INIT (struct sched_domain) {		\
+	.span			= CPU_MASK_NONE,	\
+	.parent			= NULL,			\
+	.groups			= NULL,			\
+	.min_interval		= 1,			\
+	.max_interval		= 4,			\
+	.busy_factor		= 64,			\
+	.imbalance_pct		= 125,			\
+	.cache_hot_time		= (10*1000000),		\
+	.per_cpu_gain		= 100,			\
+	.cache_nice_tries	= 2,			\
+	.busy_idx		= 2,			\
+	.idle_idx		= 1,			\
+	.newidle_idx		= 2,			\
+	.wake_idx		= 1,			\
+	.forkexec_idx		= 1,			\
+	.flags			= SD_LOAD_BALANCE	\
+				| SD_BALANCE_NEWIDLE	\
+				| SD_BALANCE_EXEC	\
+				| SD_WAKE_AFFINE,	\
+	.last_balance		= jiffies,		\
+	.balance_interval	= 1,			\
+	.nr_balance_failed	= 0,			\
+}
+
 /* sched_domains SD_NODE_INIT for IA64 NUMA machines */
 #define SD_NODE_INIT (struct sched_domain) {		\
 	.span			= CPU_MASK_NONE,	\
 	.parent			= NULL,			\
 	.groups			= NULL,			\
-	.min_interval		= 80,			\
-	.max_interval		= 320,			\
-	.busy_factor		= 320,			\
+	.min_interval		= 8,			\
+	.max_interval		= 8*(min(num_online_cpus(), 32)), \
+	.busy_factor		= 64,			\
 	.imbalance_pct		= 125,			\
 	.cache_hot_time		= (10*1000000),		\
-	.cache_nice_tries	= 1,			\
+	.cache_nice_tries	= 2,			\
+	.busy_idx		= 3,			\
+	.idle_idx		= 2,			\
+	.newidle_idx		= 0, /* unused */	\
+	.wake_idx		= 1,			\
+	.forkexec_idx		= 1,			\
 	.per_cpu_gain		= 100,			\
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_EXEC	\
-				| SD_BALANCE_NEWIDLE	\
-				| SD_WAKE_IDLE		\
+				| SD_BALANCE_FORK	\
 				| SD_WAKE_BALANCE,	\
 	.last_balance		= jiffies,		\
-	.balance_interval	= 1,			\
+	.balance_interval	= 64,			\
 	.nr_balance_failed	= 0,			\
 }
 
@@ -69,17 +98,21 @@ void build_cpu_to_node_map(void);
 	.span			= CPU_MASK_NONE,	\
 	.parent			= NULL,			\
 	.groups			= NULL,			\
-	.min_interval		= 80,			\
-	.max_interval		= 320,			\
-	.busy_factor		= 320,			\
-	.imbalance_pct		= 125,			\
+	.min_interval		= 64,			\
+	.max_interval		= 64*num_online_cpus(),	\
+	.busy_factor		= 128,			\
+	.imbalance_pct		= 133,			\
 	.cache_hot_time		= (10*1000000),		\
 	.cache_nice_tries	= 1,			\
+	.busy_idx		= 3,			\
+	.idle_idx		= 3,			\
+	.newidle_idx		= 0, /* unused */	\
+	.wake_idx		= 0, /* unused */	\
+	.forkexec_idx		= 0, /* unused */	\
 	.per_cpu_gain		= 100,			\
-	.flags			= SD_LOAD_BALANCE	\
-				| SD_BALANCE_EXEC,	\
+	.flags			= SD_LOAD_BALANCE,	\
 	.last_balance		= jiffies,		\
-	.balance_interval	= 100*(63+num_online_cpus())/64,   \
+	.balance_interval	= 64,			\
 	.nr_balance_failed	= 0,			\
 }
 
diff --git a/include/asm-x86_64/topology.h b/include/asm-x86_64/topology.h
index 802d09b9c99..c1bc3fad482 100644
--- a/include/asm-x86_64/topology.h
+++ b/include/asm-x86_64/topology.h
@@ -42,12 +42,11 @@ extern int __node_distance(int, int);
 	.cache_nice_tries	= 2,			\
 	.busy_idx		= 3,			\
 	.idle_idx		= 2,			\
-	.newidle_idx		= 1, 			\
+	.newidle_idx		= 0, 			\
 	.wake_idx		= 1,			\
 	.forkexec_idx		= 1,			\
 	.per_cpu_gain		= 100,			\
 	.flags			= SD_LOAD_BALANCE	\
-				| SD_BALANCE_NEWIDLE	\
 				| SD_BALANCE_FORK	\
 				| SD_BALANCE_EXEC	\
 				| SD_WAKE_BALANCE,	\
diff --git a/include/linux/topology.h b/include/linux/topology.h
index 665597207de..0320225e96d 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -91,7 +91,7 @@
 	.per_cpu_gain		= 25,			\
 	.busy_idx		= 0,			\
 	.idle_idx		= 0,			\
-	.newidle_idx		= 0,			\
+	.newidle_idx		= 1,			\
 	.wake_idx		= 0,			\
 	.forkexec_idx		= 0,			\
 	.flags			= SD_LOAD_BALANCE	\
@@ -121,15 +121,14 @@
 	.cache_nice_tries	= 1,			\
 	.per_cpu_gain		= 100,			\
 	.busy_idx		= 2,			\
-	.idle_idx		= 0,			\
-	.newidle_idx		= 1,			\
+	.idle_idx		= 1,			\
+	.newidle_idx		= 2,			\
 	.wake_idx		= 1,			\
-	.forkexec_idx		= 0,			\
+	.forkexec_idx		= 1,			\
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_NEWIDLE	\
 				| SD_BALANCE_EXEC	\
-				| SD_WAKE_AFFINE	\
-				| SD_WAKE_BALANCE,	\
+				| SD_WAKE_AFFINE,	\
 	.last_balance		= jiffies,		\
 	.balance_interval	= 1,			\
 	.nr_balance_failed	= 0,			\
-- 
cgit v1.2.3-70-g09d2


From 4866cde064afbb6c2a488c265e696879de616daa Mon Sep 17 00:00:00 2001
From: Nick Piggin <nickpiggin@yahoo.com.au>
Date: Sat, 25 Jun 2005 14:57:23 -0700
Subject: [PATCH] sched: cleanup context switch locking

Instead of requiring architecture code to interact with the scheduler's
locking implementation, provide a couple of defines that can be used by the
architecture to request runqueue unlocked context switches, and ask for
interrupts to be enabled over the context switch.

Also replaces the "switch_lock" used by these architectures with an oncpu
flag (note, not a potentially slow bitflag).  This eliminates one bus
locked memory operation when context switching, and simplifies the
task_running function.

Signed-off-by: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-arm/system.h     |  30 ++--------
 include/asm-ia64/system.h    |  10 +---
 include/asm-mips/system.h    |  10 +---
 include/asm-s390/system.h    |  17 +-----
 include/asm-sparc/system.h   |   4 +-
 include/asm-sparc64/system.h |  14 ++---
 include/linux/init_task.h    |   1 -
 include/linux/sched.h        |  10 +++-
 kernel/sched.c               | 132 +++++++++++++++++++++++++++++++++++--------
 9 files changed, 131 insertions(+), 97 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/system.h b/include/asm-arm/system.h
index 39dd7008013..3d0d2860b6d 100644
--- a/include/asm-arm/system.h
+++ b/include/asm-arm/system.h
@@ -145,34 +145,12 @@ extern unsigned int user_debug;
 #define set_wmb(var, value) do { var = value; wmb(); } while (0)
 #define nop() __asm__ __volatile__("mov\tr0,r0\t@ nop\n\t");
 
-#ifdef CONFIG_SMP
 /*
- * Define our own context switch locking.  This allows us to enable
- * interrupts over the context switch, otherwise we end up with high
- * interrupt latency.  The real problem area is switch_mm() which may
- * do a full cache flush.
+ * switch_mm() may do a full cache flush over the context switch,
+ * so enable interrupts over the context switch to avoid high
+ * latency.
  */
-#define prepare_arch_switch(rq,next)					\
-do {									\
-	spin_lock(&(next)->switch_lock);				\
-	spin_unlock_irq(&(rq)->lock);					\
-} while (0)
-
-#define finish_arch_switch(rq,prev)					\
-	spin_unlock(&(prev)->switch_lock)
-
-#define task_running(rq,p)						\
-	((rq)->curr == (p) || spin_is_locked(&(p)->switch_lock))
-#else
-/*
- * Our UP-case is more simple, but we assume knowledge of how
- * spin_unlock_irq() and friends are implemented.  This avoids
- * us needlessly decrementing and incrementing the preempt count.
- */
-#define prepare_arch_switch(rq,next)	local_irq_enable()
-#define finish_arch_switch(rq,prev)	spin_unlock(&(rq)->lock)
-#define task_running(rq,p)		((rq)->curr == (p))
-#endif
+#define __ARCH_WANT_INTERRUPTS_ON_CTXSW
 
 /*
  * switch_to(prev, next) should switch from task `prev' to `next'
diff --git a/include/asm-ia64/system.h b/include/asm-ia64/system.h
index 6f516e76d1f..cd2cf76b2db 100644
--- a/include/asm-ia64/system.h
+++ b/include/asm-ia64/system.h
@@ -183,8 +183,6 @@ do {								\
 
 #ifdef __KERNEL__
 
-#define prepare_to_switch()    do { } while(0)
-
 #ifdef CONFIG_IA32_SUPPORT
 # define IS_IA32_PROCESS(regs)	(ia64_psr(regs)->is != 0)
 #else
@@ -274,13 +272,7 @@ extern void ia64_load_extra (struct task_struct *task);
  * of that CPU which will not be released, because there we wait for the
  * tasklist_lock to become available.
  */
-#define prepare_arch_switch(rq, next)		\
-do {						\
-	spin_lock(&(next)->switch_lock);	\
-	spin_unlock(&(rq)->lock);		\
-} while (0)
-#define finish_arch_switch(rq, prev)	spin_unlock_irq(&(prev)->switch_lock)
-#define task_running(rq, p) 		((rq)->curr == (p) || spin_is_locked(&(p)->switch_lock))
+#define __ARCH_WANT_UNLOCKED_CTXSW
 
 #define ia64_platform_is(x) (strcmp(x, platform_name) == 0)
 
diff --git a/include/asm-mips/system.h b/include/asm-mips/system.h
index 888fd890846..169f3d4265b 100644
--- a/include/asm-mips/system.h
+++ b/include/asm-mips/system.h
@@ -422,16 +422,10 @@ extern void __die_if_kernel(const char *, struct pt_regs *, const char *file,
 extern int stop_a_enabled;
 
 /*
- * Taken from include/asm-ia64/system.h; prevents deadlock on SMP
+ * See include/asm-ia64/system.h; prevents deadlock on SMP
  * systems.
  */
-#define prepare_arch_switch(rq, next)		\
-do {						\
-	spin_lock(&(next)->switch_lock);	\
-	spin_unlock(&(rq)->lock);		\
-} while (0)
-#define finish_arch_switch(rq, prev)	spin_unlock_irq(&(prev)->switch_lock)
-#define task_running(rq, p) 		((rq)->curr == (p) || spin_is_locked(&(p)->switch_lock))
+#define __ARCH_WANT_UNLOCKED_CTXSW
 
 #define arch_align_stack(x) (x)
 
diff --git a/include/asm-s390/system.h b/include/asm-s390/system.h
index e3cb3ce1d24..b4a9f05a93d 100644
--- a/include/asm-s390/system.h
+++ b/include/asm-s390/system.h
@@ -104,29 +104,18 @@ static inline void restore_access_regs(unsigned int *acrs)
 	prev = __switch_to(prev,next);					     \
 } while (0)
 
-#define prepare_arch_switch(rq, next)	do { } while(0)
-#define task_running(rq, p)		((rq)->curr == (p))
-
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
 extern void account_user_vtime(struct task_struct *);
 extern void account_system_vtime(struct task_struct *);
-
-#define finish_arch_switch(rq, prev) do {				     \
-	set_fs(current->thread.mm_segment);				     \
-	spin_unlock(&(rq)->lock);					     \
-	account_system_vtime(prev);					     \
-	local_irq_enable();						     \
-} while (0)
-
 #else
+#define account_system_vtime(prev) do { } while (0)
+#endif
 
 #define finish_arch_switch(rq, prev) do {				     \
 	set_fs(current->thread.mm_segment);				     \
-	spin_unlock_irq(&(rq)->lock);					     \
+	account_system_vtime(prev);					     \
 } while (0)
 
-#endif
-
 #define nop() __asm__ __volatile__ ("nop")
 
 #define xchg(ptr,x) \
diff --git a/include/asm-sparc/system.h b/include/asm-sparc/system.h
index 80cf20cfaee..898562ebe94 100644
--- a/include/asm-sparc/system.h
+++ b/include/asm-sparc/system.h
@@ -101,7 +101,7 @@ extern void fpsave(unsigned long *fpregs, unsigned long *fsr,
  * SWITCH_ENTER and SWITH_DO_LAZY_FPU do not work yet (e.g. SMP does not work)
  * XXX WTF is the above comment? Found in late teen 2.4.x.
  */
-#define prepare_arch_switch(rq, next) do { \
+#define prepare_arch_switch(next) do { \
 	__asm__ __volatile__( \
 	".globl\tflush_patch_switch\nflush_patch_switch:\n\t" \
 	"save %sp, -0x40, %sp; save %sp, -0x40, %sp; save %sp, -0x40, %sp\n\t" \
@@ -109,8 +109,6 @@ extern void fpsave(unsigned long *fpregs, unsigned long *fsr,
 	"save %sp, -0x40, %sp\n\t" \
 	"restore; restore; restore; restore; restore; restore; restore"); \
 } while(0)
-#define finish_arch_switch(rq, next)	spin_unlock_irq(&(rq)->lock)
-#define task_running(rq, p)		((rq)->curr == (p))
 
 	/* Much care has gone into this code, do not touch it.
 	 *
diff --git a/include/asm-sparc64/system.h b/include/asm-sparc64/system.h
index fd12ca386f4..f9be2c5b4dc 100644
--- a/include/asm-sparc64/system.h
+++ b/include/asm-sparc64/system.h
@@ -139,19 +139,13 @@ extern void __flushw_user(void);
 #define flush_user_windows flushw_user
 #define flush_register_windows flushw_all
 
-#define prepare_arch_switch(rq, next)		\
-do {	spin_lock(&(next)->switch_lock);	\
-	spin_unlock(&(rq)->lock);		\
+/* Don't hold the runqueue lock over context switch */
+#define __ARCH_WANT_UNLOCKED_CTXSW
+#define prepare_arch_switch(next)		\
+do {						\
 	flushw_all();				\
 } while (0)
 
-#define finish_arch_switch(rq, prev)		\
-do {	spin_unlock_irq(&(prev)->switch_lock);	\
-} while (0)
-
-#define task_running(rq, p) \
-	((rq)->curr == (p) || spin_is_locked(&(p)->switch_lock))
-
 	/* See what happens when you design the chip correctly?
 	 *
 	 * We tell gcc we clobber all non-fixed-usage registers except
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index a6a8c1a38d5..03206a425d7 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -108,7 +108,6 @@ extern struct group_info init_groups;
 	.blocked	= {{0}},					\
 	.alloc_lock	= SPIN_LOCK_UNLOCKED,				\
 	.proc_lock	= SPIN_LOCK_UNLOCKED,				\
-	.switch_lock	= SPIN_LOCK_UNLOCKED,				\
 	.journal_info	= NULL,						\
 	.cpu_timers	= INIT_CPU_TIMERS(tsk.cpu_timers),		\
 }
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 36a10781c3f..d27be933742 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -368,6 +368,11 @@ struct signal_struct {
 #endif
 };
 
+/* Context switch must be unlocked if interrupts are to be enabled */
+#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW
+# define __ARCH_WANT_UNLOCKED_CTXSW
+#endif
+
 /*
  * Bits in flags field of signal_struct.
  */
@@ -594,6 +599,9 @@ struct task_struct {
 
 	int lock_depth;		/* BKL lock depth */
 
+#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW)
+	int oncpu;
+#endif
 	int prio, static_prio;
 	struct list_head run_list;
 	prio_array_t *array;
@@ -716,8 +724,6 @@ struct task_struct {
 	spinlock_t alloc_lock;
 /* Protection of proc_dentry: nesting proc_lock, dcache_lock, write_lock_irq(&tasklist_lock); */
 	spinlock_t proc_lock;
-/* context-switch lock */
-	spinlock_t switch_lock;
 
 /* journalling filesystem info */
 	void *journal_info;
diff --git a/kernel/sched.c b/kernel/sched.c
index 98bf1c091da..b1410577f9a 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -268,14 +268,71 @@ static DEFINE_PER_CPU(struct runqueue, runqueues);
 #define task_rq(p)		cpu_rq(task_cpu(p))
 #define cpu_curr(cpu)		(cpu_rq(cpu)->curr)
 
-/*
- * Default context-switch locking:
- */
 #ifndef prepare_arch_switch
-# define prepare_arch_switch(rq, next)	do { } while (0)
-# define finish_arch_switch(rq, next)	spin_unlock_irq(&(rq)->lock)
-# define task_running(rq, p)		((rq)->curr == (p))
+# define prepare_arch_switch(next)	do { } while (0)
+#endif
+#ifndef finish_arch_switch
+# define finish_arch_switch(prev)	do { } while (0)
+#endif
+
+#ifndef __ARCH_WANT_UNLOCKED_CTXSW
+static inline int task_running(runqueue_t *rq, task_t *p)
+{
+	return rq->curr == p;
+}
+
+static inline void prepare_lock_switch(runqueue_t *rq, task_t *next)
+{
+}
+
+static inline void finish_lock_switch(runqueue_t *rq, task_t *prev)
+{
+	spin_unlock_irq(&rq->lock);
+}
+
+#else /* __ARCH_WANT_UNLOCKED_CTXSW */
+static inline int task_running(runqueue_t *rq, task_t *p)
+{
+#ifdef CONFIG_SMP
+	return p->oncpu;
+#else
+	return rq->curr == p;
+#endif
+}
+
+static inline void prepare_lock_switch(runqueue_t *rq, task_t *next)
+{
+#ifdef CONFIG_SMP
+	/*
+	 * We can optimise this out completely for !SMP, because the
+	 * SMP rebalancing from interrupt is the only thing that cares
+	 * here.
+	 */
+	next->oncpu = 1;
+#endif
+#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW
+	spin_unlock_irq(&rq->lock);
+#else
+	spin_unlock(&rq->lock);
 #endif
+}
+
+static inline void finish_lock_switch(runqueue_t *rq, task_t *prev)
+{
+#ifdef CONFIG_SMP
+	/*
+	 * After ->oncpu is cleared, the task can be moved to a different CPU.
+	 * We must ensure this doesn't happen until the switch is completely
+	 * finished.
+	 */
+	smp_wmb();
+	prev->oncpu = 0;
+#endif
+#ifndef __ARCH_WANT_INTERRUPTS_ON_CTXSW
+	local_irq_enable();
+#endif
+}
+#endif /* __ARCH_WANT_UNLOCKED_CTXSW */
 
 /*
  * task_rq_lock - lock the runqueue a given task resides on and disable
@@ -1196,17 +1253,14 @@ void fastcall sched_fork(task_t *p)
 	p->state = TASK_RUNNING;
 	INIT_LIST_HEAD(&p->run_list);
 	p->array = NULL;
-	spin_lock_init(&p->switch_lock);
 #ifdef CONFIG_SCHEDSTATS
 	memset(&p->sched_info, 0, sizeof(p->sched_info));
 #endif
+#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW)
+	p->oncpu = 0;
+#endif
 #ifdef CONFIG_PREEMPT
-	/*
-	 * During context-switch we hold precisely one spinlock, which
-	 * schedule_tail drops. (in the common case it's this_rq()->lock,
-	 * but it also can be p->switch_lock.) So we compensate with a count
-	 * of 1. Also, we want to start with kernel preemption disabled.
-	 */
+	/* Want to start with kernel preemption disabled. */
 	p->thread_info->preempt_count = 1;
 #endif
 	/*
@@ -1387,23 +1441,41 @@ void fastcall sched_exit(task_t * p)
 	task_rq_unlock(rq, &flags);
 }
 
+/**
+ * prepare_task_switch - prepare to switch tasks
+ * @rq: the runqueue preparing to switch
+ * @next: the task we are going to switch to.
+ *
+ * This is called with the rq lock held and interrupts off. It must
+ * be paired with a subsequent finish_task_switch after the context
+ * switch.
+ *
+ * prepare_task_switch sets up locking and calls architecture specific
+ * hooks.
+ */
+static inline void prepare_task_switch(runqueue_t *rq, task_t *next)
+{
+	prepare_lock_switch(rq, next);
+	prepare_arch_switch(next);
+}
+
 /**
  * finish_task_switch - clean up after a task-switch
  * @prev: the thread we just switched away from.
  *
- * We enter this with the runqueue still locked, and finish_arch_switch()
- * will unlock it along with doing any other architecture-specific cleanup
- * actions.
+ * finish_task_switch must be called after the context switch, paired
+ * with a prepare_task_switch call before the context switch.
+ * finish_task_switch will reconcile locking set up by prepare_task_switch,
+ * and do any other architecture-specific cleanup actions.
  *
  * Note that we may have delayed dropping an mm in context_switch(). If
  * so, we finish that here outside of the runqueue lock.  (Doing it
  * with the lock held can cause deadlocks; see schedule() for
  * details.)
  */
-static inline void finish_task_switch(task_t *prev)
+static inline void finish_task_switch(runqueue_t *rq, task_t *prev)
 	__releases(rq->lock)
 {
-	runqueue_t *rq = this_rq();
 	struct mm_struct *mm = rq->prev_mm;
 	unsigned long prev_task_flags;
 
@@ -1421,7 +1493,8 @@ static inline void finish_task_switch(task_t *prev)
 	 *		Manfred Spraul <manfred@colorfullife.com>
 	 */
 	prev_task_flags = prev->flags;
-	finish_arch_switch(rq, prev);
+	finish_arch_switch(prev);
+	finish_lock_switch(rq, prev);
 	if (mm)
 		mmdrop(mm);
 	if (unlikely(prev_task_flags & PF_DEAD))
@@ -1435,8 +1508,12 @@ static inline void finish_task_switch(task_t *prev)
 asmlinkage void schedule_tail(task_t *prev)
 	__releases(rq->lock)
 {
-	finish_task_switch(prev);
-
+	runqueue_t *rq = this_rq();
+	finish_task_switch(rq, prev);
+#ifdef __ARCH_WANT_UNLOCKED_CTXSW
+	/* In this case, finish_task_switch does not reenable preemption */
+	preempt_enable();
+#endif
 	if (current->set_child_tid)
 		put_user(current->pid, current->set_child_tid);
 }
@@ -2816,11 +2893,15 @@ switch_tasks:
 		rq->curr = next;
 		++*switch_count;
 
-		prepare_arch_switch(rq, next);
+		prepare_task_switch(rq, next);
 		prev = context_switch(rq, prev, next);
 		barrier();
-
-		finish_task_switch(prev);
+		/*
+		 * this_rq must be evaluated again because prev may have moved
+		 * CPUs since it called schedule(), thus the 'rq' on its stack
+		 * frame will be invalid.
+		 */
+		finish_task_switch(this_rq(), prev);
 	} else
 		spin_unlock_irq(&rq->lock);
 
@@ -4085,6 +4166,9 @@ void __devinit init_idle(task_t *idle, int cpu)
 
 	spin_lock_irqsave(&rq->lock, flags);
 	rq->curr = rq->idle = idle;
+#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW)
+	idle->oncpu = 1;
+#endif
 	set_tsk_need_resched(idle);
 	spin_unlock_irqrestore(&rq->lock, flags);
 
-- 
cgit v1.2.3-70-g09d2


From 476d139c218e44e045e4bc6d4cc02b010b343939 Mon Sep 17 00:00:00 2001
From: Nick Piggin <nickpiggin@yahoo.com.au>
Date: Sat, 25 Jun 2005 14:57:29 -0700
Subject: [PATCH] sched: consolidate sbe sbf

Consolidate balance-on-exec with balance-on-fork.  This is made easy by the
sched-domains RCU patches.

As well as the general goodness of code reduction, this allows the runqueues
to be unlocked during balance-on-fork.

schedstats is a problem.  Maybe just have balance-on-event instead of
distinguishing fork and exec?

Signed-off-by: Nick Piggin <nickpiggin@yahoo.com.au>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/sched.h |   2 +-
 kernel/fork.c         |  21 +++---
 kernel/sched.c        | 174 ++++++++++++++++++++------------------------------
 3 files changed, 81 insertions(+), 116 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index d27be933742..edb2c69a887 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -930,7 +930,7 @@ extern void FASTCALL(wake_up_new_task(struct task_struct * tsk,
 #else
  static inline void kick_process(struct task_struct *tsk) { }
 #endif
-extern void FASTCALL(sched_fork(task_t * p));
+extern void FASTCALL(sched_fork(task_t * p, int clone_flags));
 extern void FASTCALL(sched_exit(task_t * p));
 
 extern int in_group_p(gid_t);
diff --git a/kernel/fork.c b/kernel/fork.c
index a28d11e1087..2c7806873bf 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1003,9 +1003,6 @@ static task_t *copy_process(unsigned long clone_flags,
 	p->pdeath_signal = 0;
 	p->exit_state = 0;
 
-	/* Perform scheduler related setup */
-	sched_fork(p);
-
 	/*
 	 * Ok, make it visible to the rest of the system.
 	 * We dont wake it up yet.
@@ -1014,18 +1011,24 @@ static task_t *copy_process(unsigned long clone_flags,
 	INIT_LIST_HEAD(&p->ptrace_children);
 	INIT_LIST_HEAD(&p->ptrace_list);
 
+	/* Perform scheduler related setup. Assign this task to a CPU. */
+	sched_fork(p, clone_flags);
+
 	/* Need tasklist lock for parent etc handling! */
 	write_lock_irq(&tasklist_lock);
 
 	/*
-	 * The task hasn't been attached yet, so cpus_allowed mask cannot
-	 * have changed. The cpus_allowed mask of the parent may have
-	 * changed after it was copied first time, and it may then move to
-	 * another CPU - so we re-copy it here and set the child's CPU to
-	 * the parent's CPU. This avoids alot of nasty races.
+	 * The task hasn't been attached yet, so its cpus_allowed mask will
+	 * not be changed, nor will its assigned CPU.
+	 *
+	 * The cpus_allowed mask of the parent may have changed after it was
+	 * copied first time - so re-copy it here, then check the child's CPU
+	 * to ensure it is on a valid CPU (and if not, just force it back to
+	 * parent's CPU). This avoids alot of nasty races.
 	 */
 	p->cpus_allowed = current->cpus_allowed;
-	set_task_cpu(p, smp_processor_id());
+	if (unlikely(!cpu_isset(task_cpu(p), p->cpus_allowed)))
+		set_task_cpu(p, smp_processor_id());
 
 	/*
 	 * Check for pending SIGKILL! The new thread should not be allowed
diff --git a/kernel/sched.c b/kernel/sched.c
index 54ce787b620..579da278e72 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1021,8 +1021,59 @@ static int find_idlest_cpu(struct sched_group *group, int this_cpu)
 	return idlest;
 }
 
+/*
+ * sched_balance_self: balance the current task (running on cpu) in domains
+ * that have the 'flag' flag set. In practice, this is SD_BALANCE_FORK and
+ * SD_BALANCE_EXEC.
+ *
+ * Balance, ie. select the least loaded group.
+ *
+ * Returns the target CPU number, or the same CPU if no balancing is needed.
+ *
+ * preempt must be disabled.
+ */
+static int sched_balance_self(int cpu, int flag)
+{
+	struct task_struct *t = current;
+	struct sched_domain *tmp, *sd = NULL;
 
-#endif
+	for_each_domain(cpu, tmp)
+		if (tmp->flags & flag)
+			sd = tmp;
+
+	while (sd) {
+		cpumask_t span;
+		struct sched_group *group;
+		int new_cpu;
+		int weight;
+
+		span = sd->span;
+		group = find_idlest_group(sd, t, cpu);
+		if (!group)
+			goto nextlevel;
+
+		new_cpu = find_idlest_cpu(group, cpu);
+		if (new_cpu == -1 || new_cpu == cpu)
+			goto nextlevel;
+
+		/* Now try balancing at a lower domain level */
+		cpu = new_cpu;
+nextlevel:
+		sd = NULL;
+		weight = cpus_weight(span);
+		for_each_domain(cpu, tmp) {
+			if (weight <= cpus_weight(tmp->span))
+				break;
+			if (tmp->flags & flag)
+				sd = tmp;
+		}
+		/* while loop will break here if sd == NULL */
+	}
+
+	return cpu;
+}
+
+#endif /* CONFIG_SMP */
 
 /*
  * wake_idle() will wake a task on an idle cpu if task->cpu is
@@ -1240,8 +1291,15 @@ int fastcall wake_up_state(task_t *p, unsigned int state)
  * Perform scheduler related setup for a newly forked process p.
  * p is forked by current.
  */
-void fastcall sched_fork(task_t *p)
+void fastcall sched_fork(task_t *p, int clone_flags)
 {
+	int cpu = get_cpu();
+
+#ifdef CONFIG_SMP
+	cpu = sched_balance_self(cpu, SD_BALANCE_FORK);
+#endif
+	set_task_cpu(p, cpu);
+
 	/*
 	 * We mark the process as running here, but have not actually
 	 * inserted it onto the runqueue yet. This guarantees that
@@ -1282,12 +1340,10 @@ void fastcall sched_fork(task_t *p)
 		 * runqueue lock is not a problem.
 		 */
 		current->time_slice = 1;
-		preempt_disable();
 		scheduler_tick();
-		local_irq_enable();
-		preempt_enable();
-	} else
-		local_irq_enable();
+	}
+	local_irq_enable();
+	put_cpu();
 }
 
 /*
@@ -1302,64 +1358,12 @@ void fastcall wake_up_new_task(task_t * p, unsigned long clone_flags)
 	unsigned long flags;
 	int this_cpu, cpu;
 	runqueue_t *rq, *this_rq;
-#ifdef CONFIG_SMP
-	struct sched_domain *tmp, *sd = NULL;
-#endif
 
 	rq = task_rq_lock(p, &flags);
 	BUG_ON(p->state != TASK_RUNNING);
 	this_cpu = smp_processor_id();
 	cpu = task_cpu(p);
 
-#ifdef CONFIG_SMP
-	for_each_domain(cpu, tmp)
-		if (tmp->flags & SD_BALANCE_FORK)
-			sd = tmp;
-
-	if (sd) {
-		cpumask_t span;
-		int new_cpu;
-		struct sched_group *group;
-
-again:
-		schedstat_inc(sd, sbf_cnt);
-		span = sd->span;
-		cpu = task_cpu(p);
-		group = find_idlest_group(sd, p, cpu);
-		if (!group) {
-			schedstat_inc(sd, sbf_balanced);
-			goto nextlevel;
-		}
-
-		new_cpu = find_idlest_cpu(group, cpu);
-		if (new_cpu == -1 || new_cpu == cpu) {
-			schedstat_inc(sd, sbf_balanced);
-			goto nextlevel;
-		}
-
-		if (cpu_isset(new_cpu, p->cpus_allowed)) {
-			schedstat_inc(sd, sbf_pushed);
-			set_task_cpu(p, new_cpu);
-			task_rq_unlock(rq, &flags);
-			rq = task_rq_lock(p, &flags);
-			cpu = task_cpu(p);
-		}
-
-		/* Now try balancing at a lower domain level */
-nextlevel:
-		sd = NULL;
-		for_each_domain(cpu, tmp) {
-			if (cpus_subset(span, tmp->span))
-				break;
-			if (tmp->flags & SD_BALANCE_FORK)
-				sd = tmp;
-		}
-
-		if (sd)
-			goto again;
-	}
-
-#endif
 	/*
 	 * We decrease the sleep average of forking parents
 	 * and children as well, to keep max-interactive tasks
@@ -1708,58 +1712,16 @@ out:
 }
 
 /*
- * sched_exec(): find the highest-level, exec-balance-capable
- * domain and try to migrate the task to the least loaded CPU.
- *
- * execve() is a valuable balancing opportunity, because at this point
- * the task has the smallest effective memory and cache footprint.
+ * sched_exec - execve() is a valuable balancing opportunity, because at
+ * this point the task has the smallest effective memory and cache footprint.
  */
 void sched_exec(void)
 {
-	struct sched_domain *tmp, *sd = NULL;
 	int new_cpu, this_cpu = get_cpu();
-
-	for_each_domain(this_cpu, tmp)
-		if (tmp->flags & SD_BALANCE_EXEC)
-			sd = tmp;
-
-	if (sd) {
-		cpumask_t span;
-		struct sched_group *group;
-again:
-		schedstat_inc(sd, sbe_cnt);
-		span = sd->span;
-		group = find_idlest_group(sd, current, this_cpu);
-		if (!group) {
-			schedstat_inc(sd, sbe_balanced);
-			goto nextlevel;
-		}
-		new_cpu = find_idlest_cpu(group, this_cpu);
-		if (new_cpu == -1 || new_cpu == this_cpu) {
-			schedstat_inc(sd, sbe_balanced);
-			goto nextlevel;
-		}
-
-		schedstat_inc(sd, sbe_pushed);
-		put_cpu();
-		sched_migrate_task(current, new_cpu);
-
-		/* Now try balancing at a lower domain level */
-		this_cpu = get_cpu();
-nextlevel:
-		sd = NULL;
-		for_each_domain(this_cpu, tmp) {
-			if (cpus_subset(span, tmp->span))
-				break;
-			if (tmp->flags & SD_BALANCE_EXEC)
-				sd = tmp;
-		}
-
-		if (sd)
-			goto again;
-	}
-
+	new_cpu = sched_balance_self(this_cpu, SD_BALANCE_EXEC);
 	put_cpu();
+	if (new_cpu != this_cpu)
+		sched_migrate_task(current, new_cpu);
 }
 
 /*
-- 
cgit v1.2.3-70-g09d2


From 1a20ff27ef75d866730ee796acd811a925af762f Mon Sep 17 00:00:00 2001
From: Dinakar Guniguntala <dino@in.ibm.com>
Date: Sat, 25 Jun 2005 14:57:33 -0700
Subject: [PATCH] Dynamic sched domains: sched changes

The following patches add dynamic sched domains functionality that was
extensively discussed on lkml and lse-tech.  I would like to see this added to
-mm

o The main advantage with this feature is that it ensures that the scheduler
  load balacing code only balances against the cpus that are in the sched
  domain as defined by an exclusive cpuset and not all of the cpus in the
  system. This removes any overhead due to load balancing code trying to
  pull tasks outside of the cpu exclusive cpuset only to be prevented by
  the tasks' cpus_allowed mask.
o cpu exclusive cpusets are useful for servers running orthogonal
  workloads such as RT applications requiring low latency and HPC
  applications that are throughput sensitive

o It provides a new API partition_sched_domains in sched.c
  that makes dynamic sched domains possible.
o cpu_exclusive cpusets sets are now associated with a sched domain.
  Which means that the users can dynamically modify the sched domains
  through the cpuset file system interface
o ia64 sched domain code has been updated to support this feature as well
o Currently, this does not support hotplug. (However some of my tests
  indicate hotplug+preempt is currently broken)
o I have tested it extensively on x86.
o This should have very minimal impact on performance as none of
  the fast paths are affected

Signed-off-by: Dinakar Guniguntala <dino@in.ibm.com>
Acked-by: Paul Jackson <pj@sgi.com>
Acked-by: Nick Piggin <nickpiggin@yahoo.com.au>
Acked-by: Matthew Dobson <colpatch@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/sched.h |   2 +
 kernel/sched.c        | 132 ++++++++++++++++++++++++++++++++------------------
 2 files changed, 88 insertions(+), 46 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index edb2c69a887..98c109e4f43 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -539,6 +539,8 @@ struct sched_domain {
 #endif
 };
 
+extern void partition_sched_domains(cpumask_t *partition1,
+				    cpumask_t *partition2);
 #ifdef ARCH_HAS_SCHED_DOMAIN
 /* Useful helpers that arch setup code may use. Defined in kernel/sched.c */
 extern cpumask_t cpu_isolated_map;
diff --git a/kernel/sched.c b/kernel/sched.c
index d3d81b82e37..dee96b22635 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -262,7 +262,7 @@ static DEFINE_PER_CPU(struct runqueue, runqueues);
 
 /*
  * The domain tree (rq->sd) is protected by RCU's quiescent state transition.
- * See update_sched_domains: synchronize_kernel for details.
+ * See detach_destroy_domains: synchronize_sched for details.
  *
  * The domain tree of any CPU may only be accessed from within
  * preempt-disabled sections.
@@ -4624,7 +4624,7 @@ int __init migration_init(void)
 #endif
 
 #ifdef CONFIG_SMP
-#define SCHED_DOMAIN_DEBUG
+#undef SCHED_DOMAIN_DEBUG
 #ifdef SCHED_DOMAIN_DEBUG
 static void sched_domain_debug(struct sched_domain *sd, int cpu)
 {
@@ -4717,7 +4717,7 @@ static void sched_domain_debug(struct sched_domain *sd, int cpu)
 #define sched_domain_debug(sd, cpu) {}
 #endif
 
-static int __devinit sd_degenerate(struct sched_domain *sd)
+static int sd_degenerate(struct sched_domain *sd)
 {
 	if (cpus_weight(sd->span) == 1)
 		return 1;
@@ -4740,7 +4740,7 @@ static int __devinit sd_degenerate(struct sched_domain *sd)
 	return 1;
 }
 
-static int __devinit sd_parent_degenerate(struct sched_domain *sd,
+static int sd_parent_degenerate(struct sched_domain *sd,
 						struct sched_domain *parent)
 {
 	unsigned long cflags = sd->flags, pflags = parent->flags;
@@ -4772,7 +4772,7 @@ static int __devinit sd_parent_degenerate(struct sched_domain *sd,
  * Attach the domain 'sd' to 'cpu' as its base domain.  Callers must
  * hold the hotplug lock.
  */
-void __devinit cpu_attach_domain(struct sched_domain *sd, int cpu)
+void cpu_attach_domain(struct sched_domain *sd, int cpu)
 {
 	runqueue_t *rq = cpu_rq(cpu);
 	struct sched_domain *tmp;
@@ -4823,7 +4823,7 @@ __setup ("isolcpus=", isolated_cpu_setup);
  * covered by the given span, and will set each group's ->cpumask correctly,
  * and ->cpu_power to 0.
  */
-void __devinit init_sched_build_groups(struct sched_group groups[],
+void init_sched_build_groups(struct sched_group groups[],
 			cpumask_t span, int (*group_fn)(int cpu))
 {
 	struct sched_group *first = NULL, *last = NULL;
@@ -4859,13 +4859,14 @@ void __devinit init_sched_build_groups(struct sched_group groups[],
 
 
 #ifdef ARCH_HAS_SCHED_DOMAIN
-extern void __devinit arch_init_sched_domains(void);
-extern void __devinit arch_destroy_sched_domains(void);
+extern void build_sched_domains(const cpumask_t *cpu_map);
+extern void arch_init_sched_domains(const cpumask_t *cpu_map);
+extern void arch_destroy_sched_domains(const cpumask_t *cpu_map);
 #else
 #ifdef CONFIG_SCHED_SMT
 static DEFINE_PER_CPU(struct sched_domain, cpu_domains);
 static struct sched_group sched_group_cpus[NR_CPUS];
-static int __devinit cpu_to_cpu_group(int cpu)
+static int cpu_to_cpu_group(int cpu)
 {
 	return cpu;
 }
@@ -4873,7 +4874,7 @@ static int __devinit cpu_to_cpu_group(int cpu)
 
 static DEFINE_PER_CPU(struct sched_domain, phys_domains);
 static struct sched_group sched_group_phys[NR_CPUS];
-static int __devinit cpu_to_phys_group(int cpu)
+static int cpu_to_phys_group(int cpu)
 {
 #ifdef CONFIG_SCHED_SMT
 	return first_cpu(cpu_sibling_map[cpu]);
@@ -4886,7 +4887,7 @@ static int __devinit cpu_to_phys_group(int cpu)
 
 static DEFINE_PER_CPU(struct sched_domain, node_domains);
 static struct sched_group sched_group_nodes[MAX_NUMNODES];
-static int __devinit cpu_to_node_group(int cpu)
+static int cpu_to_node_group(int cpu)
 {
 	return cpu_to_node(cpu);
 }
@@ -4917,39 +4918,28 @@ static void check_sibling_maps(void)
 #endif
 
 /*
- * Set up scheduler domains and groups.  Callers must hold the hotplug lock.
+ * Build sched domains for a given set of cpus and attach the sched domains
+ * to the individual cpus
  */
-static void __devinit arch_init_sched_domains(void)
+static void build_sched_domains(const cpumask_t *cpu_map)
 {
 	int i;
-	cpumask_t cpu_default_map;
-
-#if defined(CONFIG_SCHED_SMT) && defined(CONFIG_NUMA)
-	check_sibling_maps();
-#endif
-	/*
-	 * Setup mask for cpus without special case scheduling requirements.
-	 * For now this just excludes isolated cpus, but could be used to
-	 * exclude other special cases in the future.
-	 */
-	cpus_complement(cpu_default_map, cpu_isolated_map);
-	cpus_and(cpu_default_map, cpu_default_map, cpu_online_map);
 
 	/*
-	 * Set up domains. Isolated domains just stay on the NULL domain.
+	 * Set up domains for cpus specified by the cpu_map.
 	 */
-	for_each_cpu_mask(i, cpu_default_map) {
+	for_each_cpu_mask(i, *cpu_map) {
 		int group;
 		struct sched_domain *sd = NULL, *p;
 		cpumask_t nodemask = node_to_cpumask(cpu_to_node(i));
 
-		cpus_and(nodemask, nodemask, cpu_default_map);
+		cpus_and(nodemask, nodemask, *cpu_map);
 
 #ifdef CONFIG_NUMA
 		sd = &per_cpu(node_domains, i);
 		group = cpu_to_node_group(i);
 		*sd = SD_NODE_INIT;
-		sd->span = cpu_default_map;
+		sd->span = *cpu_map;
 		sd->groups = &sched_group_nodes[group];
 #endif
 
@@ -4967,7 +4957,7 @@ static void __devinit arch_init_sched_domains(void)
 		group = cpu_to_cpu_group(i);
 		*sd = SD_SIBLING_INIT;
 		sd->span = cpu_sibling_map[i];
-		cpus_and(sd->span, sd->span, cpu_default_map);
+		cpus_and(sd->span, sd->span, *cpu_map);
 		sd->parent = p;
 		sd->groups = &sched_group_cpus[group];
 #endif
@@ -4977,7 +4967,7 @@ static void __devinit arch_init_sched_domains(void)
 	/* Set up CPU (sibling) groups */
 	for_each_online_cpu(i) {
 		cpumask_t this_sibling_map = cpu_sibling_map[i];
-		cpus_and(this_sibling_map, this_sibling_map, cpu_default_map);
+		cpus_and(this_sibling_map, this_sibling_map, *cpu_map);
 		if (i != first_cpu(this_sibling_map))
 			continue;
 
@@ -4990,7 +4980,7 @@ static void __devinit arch_init_sched_domains(void)
 	for (i = 0; i < MAX_NUMNODES; i++) {
 		cpumask_t nodemask = node_to_cpumask(i);
 
-		cpus_and(nodemask, nodemask, cpu_default_map);
+		cpus_and(nodemask, nodemask, *cpu_map);
 		if (cpus_empty(nodemask))
 			continue;
 
@@ -5000,12 +4990,12 @@ static void __devinit arch_init_sched_domains(void)
 
 #ifdef CONFIG_NUMA
 	/* Set up node groups */
-	init_sched_build_groups(sched_group_nodes, cpu_default_map,
+	init_sched_build_groups(sched_group_nodes, *cpu_map,
 					&cpu_to_node_group);
 #endif
 
 	/* Calculate CPU power for physical packages and nodes */
-	for_each_cpu_mask(i, cpu_default_map) {
+	for_each_cpu_mask(i, *cpu_map) {
 		int power;
 		struct sched_domain *sd;
 #ifdef CONFIG_SCHED_SMT
@@ -5029,7 +5019,7 @@ static void __devinit arch_init_sched_domains(void)
 	}
 
 	/* Attach the domains */
-	for_each_online_cpu(i) {
+	for_each_cpu_mask(i, *cpu_map) {
 		struct sched_domain *sd;
 #ifdef CONFIG_SCHED_SMT
 		sd = &per_cpu(cpu_domains, i);
@@ -5039,16 +5029,71 @@ static void __devinit arch_init_sched_domains(void)
 		cpu_attach_domain(sd, i);
 	}
 }
+/*
+ * Set up scheduler domains and groups.  Callers must hold the hotplug lock.
+ */
+static void arch_init_sched_domains(cpumask_t *cpu_map)
+{
+	cpumask_t cpu_default_map;
 
-#ifdef CONFIG_HOTPLUG_CPU
-static void __devinit arch_destroy_sched_domains(void)
+#if defined(CONFIG_SCHED_SMT) && defined(CONFIG_NUMA)
+	check_sibling_maps();
+#endif
+	/*
+	 * Setup mask for cpus without special case scheduling requirements.
+	 * For now this just excludes isolated cpus, but could be used to
+	 * exclude other special cases in the future.
+	 */
+	cpus_andnot(cpu_default_map, *cpu_map, cpu_isolated_map);
+
+	build_sched_domains(&cpu_default_map);
+}
+
+static void arch_destroy_sched_domains(const cpumask_t *cpu_map)
 {
 	/* Do nothing: everything is statically allocated. */
 }
-#endif
 
 #endif /* ARCH_HAS_SCHED_DOMAIN */
 
+/*
+ * Detach sched domains from a group of cpus specified in cpu_map
+ * These cpus will now be attached to the NULL domain
+ */
+static inline void detach_destroy_domains(const cpumask_t *cpu_map)
+{
+	int i;
+
+	for_each_cpu_mask(i, *cpu_map)
+		cpu_attach_domain(NULL, i);
+	synchronize_sched();
+	arch_destroy_sched_domains(cpu_map);
+}
+
+/*
+ * Partition sched domains as specified by the cpumasks below.
+ * This attaches all cpus from the cpumasks to the NULL domain,
+ * waits for a RCU quiescent period, recalculates sched
+ * domain information and then attaches them back to the
+ * correct sched domains
+ * Call with hotplug lock held
+ */
+void partition_sched_domains(cpumask_t *partition1, cpumask_t *partition2)
+{
+	cpumask_t change_map;
+
+	cpus_and(*partition1, *partition1, cpu_online_map);
+	cpus_and(*partition2, *partition2, cpu_online_map);
+	cpus_or(change_map, *partition1, *partition2);
+
+	/* Detach sched domains from all of the affected cpus */
+	detach_destroy_domains(&change_map);
+	if (!cpus_empty(*partition1))
+		build_sched_domains(partition1);
+	if (!cpus_empty(*partition2))
+		build_sched_domains(partition2);
+}
+
 #ifdef CONFIG_HOTPLUG_CPU
 /*
  * Force a reinitialization of the sched domains hierarchy.  The domains
@@ -5059,15 +5104,10 @@ static void __devinit arch_destroy_sched_domains(void)
 static int update_sched_domains(struct notifier_block *nfb,
 				unsigned long action, void *hcpu)
 {
-	int i;
-
 	switch (action) {
 	case CPU_UP_PREPARE:
 	case CPU_DOWN_PREPARE:
-		for_each_online_cpu(i)
-			cpu_attach_domain(NULL, i);
-		synchronize_kernel();
-		arch_destroy_sched_domains();
+		detach_destroy_domains(&cpu_online_map);
 		return NOTIFY_OK;
 
 	case CPU_UP_CANCELED:
@@ -5083,7 +5123,7 @@ static int update_sched_domains(struct notifier_block *nfb,
 	}
 
 	/* The hotplug lock is already held by cpu_up/cpu_down */
-	arch_init_sched_domains();
+	arch_init_sched_domains(&cpu_online_map);
 
 	return NOTIFY_OK;
 }
@@ -5092,7 +5132,7 @@ static int update_sched_domains(struct notifier_block *nfb,
 void __init sched_init_smp(void)
 {
 	lock_cpu_hotplug();
-	arch_init_sched_domains();
+	arch_init_sched_domains(&cpu_online_map);
 	unlock_cpu_hotplug();
 	/* XXX: Theoretical race here - CPU may be hotplugged now */
 	hotcpu_notifier(update_sched_domains, 0);
-- 
cgit v1.2.3-70-g09d2


From f8cbd99bd3a023db8d6356d19a5f6f539d367327 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sat, 25 Jun 2005 14:57:39 -0700
Subject: [PATCH] sched: voluntary kernel preemption

This patch adds a new preemption model: 'Voluntary Kernel Preemption'.  The
3 models can be selected from a new menu:

            (X) No Forced Preemption (Server)
            ( ) Voluntary Kernel Preemption (Desktop)
            ( ) Preemptible Kernel (Low-Latency Desktop)

we still default to the stock (Server) preemption model.

Voluntary preemption works by adding a cond_resched()
(reschedule-if-needed) call to every might_sleep() check.  It is lighter
than CONFIG_PREEMPT - at the cost of not having as tight latencies.  It
represents a different latency/complexity/overhead tradeoff.

It has no runtime impact at all if disabled.  Here are size stats that show
how the various preemption models impact the kernel's size:

    text    data     bss     dec     hex filename
 3618774  547184  179896 4345854  424ffe vmlinux.stock
 3626406  547184  179896 4353486  426dce vmlinux.voluntary   +0.2%
 3748414  548640  179896 4476950  445016 vmlinux.preempt     +3.5%

voluntary-preempt is +0.2% of .text, preempt is +3.5%.

This feature has been tested for many months by lots of people (and it's
also included in the RHEL4 distribution and earlier variants were in Fedora
as well), and it's intended for users and distributions who dont want to
use full-blown CONFIG_PREEMPT for one reason or another.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/kernel.h | 18 +++++++++++-----
 kernel/Kconfig.preempt | 57 +++++++++++++++++++++++++++++++++++++++++++-------
 2 files changed, 62 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index e25b97062ce..687ba8c9973 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -58,15 +58,23 @@ struct completion;
  * be biten later when the calling function happens to sleep when it is not
  * supposed to.
  */
+#ifdef CONFIG_PREEMPT_VOLUNTARY
+extern int cond_resched(void);
+# define might_resched() cond_resched()
+#else
+# define might_resched() do { } while (0)
+#endif
+
 #ifdef CONFIG_DEBUG_SPINLOCK_SLEEP
-#define might_sleep() __might_sleep(__FILE__, __LINE__)
-#define might_sleep_if(cond) do { if (unlikely(cond)) might_sleep(); } while (0)
-void __might_sleep(char *file, int line);
+  void __might_sleep(char *file, int line);
+# define might_sleep() \
+	do { __might_sleep(__FILE__, __LINE__); might_resched(); } while (0)
 #else
-#define might_sleep() do {} while(0)
-#define might_sleep_if(cond) do {} while (0)
+# define might_sleep() do { might_resched(); } while (0)
 #endif
 
+#define might_sleep_if(cond) do { if (unlikely(cond)) might_sleep(); } while (0)
+
 #define abs(x) ({				\
 		int __x = (x);			\
 		(__x < 0) ? -__x : __x;		\
diff --git a/kernel/Kconfig.preempt b/kernel/Kconfig.preempt
index 34c631221aa..0b46a5dff4c 100644
--- a/kernel/Kconfig.preempt
+++ b/kernel/Kconfig.preempt
@@ -1,15 +1,56 @@
 
-config PREEMPT
-	bool "Preemptible Kernel"
+choice
+	prompt "Preemption Model"
+	default PREEMPT_NONE
+
+config PREEMPT_NONE
+	bool "No Forced Preemption (Server)"
+	help
+	  This is the traditional Linux preemption model, geared towards
+	  throughput. It will still provide good latencies most of the
+	  time, but there are no guarantees and occasional longer delays
+	  are possible.
+
+	  Select this option if you are building a kernel for a server or
+	  scientific/computation system, or if you want to maximize the
+	  raw processing power of the kernel, irrespective of scheduling
+	  latencies.
+
+config PREEMPT_VOLUNTARY
+	bool "Voluntary Kernel Preemption (Desktop)"
 	help
-	  This option reduces the latency of the kernel when reacting to
-	  real-time or interactive events by allowing a low priority process to
-	  be preempted even if it is in kernel mode executing a system call.
-	  This allows applications to run more reliably even when the system is
+	  This option reduces the latency of the kernel by adding more
+	  "explicit preemption points" to the kernel code. These new
+	  preemption points have been selected to reduce the maximum
+	  latency of rescheduling, providing faster application reactions,
+	  at the cost of slighly lower throughput.
+
+	  This allows reaction to interactive events by allowing a
+	  low priority process to voluntarily preempt itself even if it
+	  is in kernel mode executing a system call. This allows
+	  applications to run more 'smoothly' even when the system is
 	  under load.
 
-	  Say Y here if you are building a kernel for a desktop, embedded
-	  or real-time system.  Say N if you are unsure.
+	  Select this if you are building a kernel for a desktop system.
+
+config PREEMPT
+	bool "Preemptible Kernel (Low-Latency Desktop)"
+	help
+	  This option reduces the latency of the kernel by making
+	  all kernel code (that is not executing in a critical section)
+	  preemptible.  This allows reaction to interactive events by
+	  permitting a low priority process to be preempted involuntarily
+	  even if it is in kernel mode executing a system call and would
+	  otherwise not be about to reach a natural preemption point.
+	  This allows applications to run more 'smoothly' even when the
+	  system is under load, at the cost of slighly lower throughput
+	  and a slight runtime overhead to kernel code.
+
+	  Select this if you are building a kernel for a desktop or
+	  embedded system with latency requirements in the milliseconds
+	  range.
+
+endchoice
 
 config PREEMPT_BKL
 	bool "Preempt The Big Kernel Lock"
-- 
cgit v1.2.3-70-g09d2


From 8f43d03fe2c4962c11d8227ac9505e590bad758b Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:40 -0700
Subject: [PATCH] kexec: x86: rename APIC_MODE_EXINT

From: "Maciej W. Rozycki" <macro@linux-mips.org>

Rename APIC_MODE_EXINT to APIC_MODE_EXTINT - I think it should be named
after what the mode is called in documentation.

From: "Eric W. Biederman" <ebiederm@lnxi.com>

I have reduced this patch to just the name change in the header.  And
integrated the changes into the patches that add those
lines. Otherwise I ran into some ugly dependencies.

Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org
Signed-off-by: Eric Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/apicdef.h   | 2 +-
 include/asm-x86_64/apicdef.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/apicdef.h b/include/asm-i386/apicdef.h
index c689554ad5b..41e8d2d918e 100644
--- a/include/asm-i386/apicdef.h
+++ b/include/asm-i386/apicdef.h
@@ -90,7 +90,7 @@
 #define			SET_APIC_DELIVERY_MODE(x,y)	(((x)&~0x700)|((y)<<8))
 #define				APIC_MODE_FIXED		0x0
 #define				APIC_MODE_NMI		0x4
-#define				APIC_MODE_EXINT		0x7
+#define				APIC_MODE_EXTINT	0x7
 #define 	APIC_LVT1	0x360
 #define		APIC_LVTERR	0x370
 #define		APIC_TMICT	0x380
diff --git a/include/asm-x86_64/apicdef.h b/include/asm-x86_64/apicdef.h
index bfebdb69065..9388062c4f6 100644
--- a/include/asm-x86_64/apicdef.h
+++ b/include/asm-x86_64/apicdef.h
@@ -94,7 +94,7 @@
 #define			SET_APIC_DELIVERY_MODE(x,y)	(((x)&~0x700)|((y)<<8))
 #define				APIC_MODE_FIXED		0x0
 #define				APIC_MODE_NMI		0x4
-#define				APIC_MODE_EXINT		0x7
+#define				APIC_MODE_EXTINT	0x7
 #define 	APIC_LVT1	0x360
 #define		APIC_LVTERR	0x370
 #define		APIC_TMICT	0x380
-- 
cgit v1.2.3-70-g09d2


From 9635b47d910223745258768418003580ef7dba17 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:41 -0700
Subject: [PATCH] kexec: x86: local apic fix

From: "Maciej W. Rozycki" <macro@linux-mips.org>

Fix a kexec problem whcih causes local APIC detection failure.

The problem is detect_init_APIC() is called early, before the command line
have been processed.  Therefore "lapic" (and "nolapic") have not been seen,
yet.

Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
Signed-off-by: Eric Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/apic.c  | 25 +++++--------------------
 arch/i386/kernel/setup.c | 11 +++++++++++
 include/asm-i386/apic.h  | 13 +++++++++++++
 3 files changed, 29 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index b905d7bb9a0..cf45bed96d0 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -40,6 +40,11 @@
 
 #include "io_ports.h"
 
+/*
+ * Knob to control our willingness to enable the local APIC.
+ */
+int enable_local_apic __initdata = 0; /* -1=force-disable, +1=force-enable */
+
 /*
  * Debug level
  */
@@ -666,26 +671,6 @@ static void apic_pm_activate(void) { }
  * Original code written by Keir Fraser.
  */
 
-/*
- * Knob to control our willingness to enable the local APIC.
- */
-int enable_local_apic __initdata = 0; /* -1=force-disable, +1=force-enable */
-
-static int __init lapic_disable(char *str)
-{
-	enable_local_apic = -1;
-	clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
-	return 0;
-}
-__setup("nolapic", lapic_disable);
-
-static int __init lapic_enable(char *str)
-{
-	enable_local_apic = 1;
-	return 0;
-}
-__setup("lapic", lapic_enable);
-
 static int __init apic_set_verbosity(char *str)
 {
 	if (strcmp("debug", str) == 0)
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index cba67e4ba0a..f1ad9fdeaad 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -44,6 +44,7 @@
 #include <linux/edd.h>
 #include <linux/nodemask.h>
 #include <video/edid.h>
+#include <asm/apic.h>
 #include <asm/e820.h>
 #include <asm/mpspec.h>
 #include <asm/setup.h>
@@ -835,6 +836,16 @@ static void __init parse_cmdline_early (char ** cmdline_p)
 #endif /* CONFIG_X86_LOCAL_APIC */
 #endif /* CONFIG_ACPI_BOOT */
 
+#ifdef CONFIG_X86_LOCAL_APIC
+		/* enable local APIC */
+		else if (!memcmp(from, "lapic", 5))
+			lapic_enable();
+
+		/* disable local APIC */
+		else if (!memcmp(from, "nolapic", 6))
+			lapic_disable();
+#endif /* CONFIG_X86_LOCAL_APIC */
+
 		/*
 		 * highmem=size forces highmem to be exactly 'size' bytes.
 		 * This works even on boxes that have no highmem otherwise.
diff --git a/include/asm-i386/apic.h b/include/asm-i386/apic.h
index a5810cf7b57..53268cd9306 100644
--- a/include/asm-i386/apic.h
+++ b/include/asm-i386/apic.h
@@ -5,6 +5,7 @@
 #include <linux/pm.h>
 #include <asm/fixmap.h>
 #include <asm/apicdef.h>
+#include <asm/processor.h>
 #include <asm/system.h>
 
 #define Dprintk(x...)
@@ -16,8 +17,20 @@
 #define APIC_VERBOSE 1
 #define APIC_DEBUG   2
 
+extern int enable_local_apic;
 extern int apic_verbosity;
 
+static inline void lapic_disable(void)
+{
+	enable_local_apic = -1;
+	clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
+}
+
+static inline void lapic_enable(void)
+{
+	enable_local_apic = 1;
+}
+
 /*
  * Define the default level of output to be very little
  * This can be turned up by using apic=verbose for more
-- 
cgit v1.2.3-70-g09d2


From 650927ef8ab1e9b05b77a3f32ca7adcedaae9306 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:44 -0700
Subject: [PATCH] kexec: x86: resture apic virtual wire mode on shutdown

When coming out of apic mode attempt to set the appropriate
apic back into virtual wire mode.  This improves on previous versions
of this patch by by never setting bot the local apic and the ioapic
into veritual wire mode.

This code looks at data from the mptable to see if an ioapic has
an ExtInt input to make this decision.  A future improvement
is to figure out which apic or ioapic was in virtual wire mode
at boot time and to remember it.  That is potentially a more accurate
method, of selecting which apic to place in virutal wire mode.

Signed-off-by: Eric Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/apic.c    | 38 +++++++++++++++++++++++++++++++++++++-
 arch/i386/kernel/io_apic.c | 33 ++++++++++++++++++++++++++++++++-
 include/asm-i386/apic.h    |  2 +-
 include/asm-i386/apicdef.h |  1 +
 4 files changed, 71 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index cf45bed96d0..93df90bbb87 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -211,7 +211,7 @@ void __init connect_bsp_APIC(void)
 	enable_apic_mode();
 }
 
-void disconnect_bsp_APIC(void)
+void disconnect_bsp_APIC(int virt_wire_setup)
 {
 	if (pic_mode) {
 		/*
@@ -225,6 +225,42 @@ void disconnect_bsp_APIC(void)
 		outb(0x70, 0x22);
 		outb(0x00, 0x23);
 	}
+	else {
+		/* Go back to Virtual Wire compatibility mode */
+		unsigned long value;
+
+		/* For the spurious interrupt use vector F, and enable it */
+		value = apic_read(APIC_SPIV);
+		value &= ~APIC_VECTOR_MASK;
+		value |= APIC_SPIV_APIC_ENABLED;
+		value |= 0xf;
+		apic_write_around(APIC_SPIV, value);
+
+		if (!virt_wire_setup) {
+			/* For LVT0 make it edge triggered, active high, external and enabled */
+			value = apic_read(APIC_LVT0);
+			value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
+				APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
+				APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED );
+			value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
+			value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_EXTINT);
+			apic_write_around(APIC_LVT0, value);
+		}
+		else {
+			/* Disable LVT0 */
+			apic_write_around(APIC_LVT0, APIC_LVT_MASKED);
+		}
+
+		/* For LVT1 make it edge triggered, active high, nmi and enabled */
+		value = apic_read(APIC_LVT1);
+		value &= ~(
+			APIC_MODE_MASK | APIC_SEND_PENDING |
+			APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
+			APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED);
+		value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
+		value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_NMI);
+		apic_write_around(APIC_LVT1, value);
+	}
 }
 
 void disable_local_APIC(void)
diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index 3c2b3bdfc80..fc81c8cddd2 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -1636,12 +1636,43 @@ static void __init enable_IO_APIC(void)
  */
 void disable_IO_APIC(void)
 {
+	int pin;
 	/*
 	 * Clear the IO-APIC before rebooting:
 	 */
 	clear_IO_APIC();
 
-	disconnect_bsp_APIC();
+	/*
+	 * If the i82559 is routed through an IOAPIC
+	 * Put that IOAPIC in virtual wire mode
+	 * so legacy interrups can be delivered.
+	 */
+	pin = find_isa_irq_pin(0, mp_ExtINT);
+	if (pin != -1) {
+		struct IO_APIC_route_entry entry;
+		unsigned long flags;
+
+		memset(&entry, 0, sizeof(entry));
+		entry.mask            = 0; /* Enabled */
+		entry.trigger         = 0; /* Edge */
+		entry.irr             = 0;
+		entry.polarity        = 0; /* High */
+		entry.delivery_status = 0;
+		entry.dest_mode       = 0; /* Physical */
+		entry.delivery_mode   = 7; /* ExtInt */
+		entry.vector          = 0;
+		entry.dest.physical.physical_dest = 0;
+
+
+		/*
+		 * Add it to the IO-APIC irq-routing table:
+		 */
+		spin_lock_irqsave(&ioapic_lock, flags);
+		io_apic_write(0, 0x11+2*pin, *(((int *)&entry)+1));
+		io_apic_write(0, 0x10+2*pin, *(((int *)&entry)+0));
+		spin_unlock_irqrestore(&ioapic_lock, flags);
+	}
+	disconnect_bsp_APIC(pin != -1);
 }
 
 /*
diff --git a/include/asm-i386/apic.h b/include/asm-i386/apic.h
index 53268cd9306..6a1b1882285 100644
--- a/include/asm-i386/apic.h
+++ b/include/asm-i386/apic.h
@@ -100,7 +100,7 @@ extern void (*wait_timer_tick)(void);
 extern int get_maxlvt(void);
 extern void clear_local_APIC(void);
 extern void connect_bsp_APIC (void);
-extern void disconnect_bsp_APIC (void);
+extern void disconnect_bsp_APIC (int virt_wire_setup);
 extern void disable_local_APIC (void);
 extern void lapic_shutdown (void);
 extern int verify_local_APIC (void);
diff --git a/include/asm-i386/apicdef.h b/include/asm-i386/apicdef.h
index 41e8d2d918e..0fed5e3c699 100644
--- a/include/asm-i386/apicdef.h
+++ b/include/asm-i386/apicdef.h
@@ -86,6 +86,7 @@
 #define			APIC_LVT_REMOTE_IRR		(1<<14)
 #define			APIC_INPUT_POLARITY		(1<<13)
 #define			APIC_SEND_PENDING		(1<<12)
+#define			APIC_MODE_MASK			0x700
 #define			GET_APIC_DELIVERY_MODE(x)	(((x)>>8)&0x7)
 #define			SET_APIC_DELIVERY_MODE(x,y)	(((x)&~0x700)|((y)<<8))
 #define				APIC_MODE_FIXED		0x0
-- 
cgit v1.2.3-70-g09d2


From 208fb93162d51faa69b9774fa7809858d84fd9dc Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:45 -0700
Subject: [PATCH] kexec: x86_64: restore apic virtual wire mode on shutdown

When coming out of apic mode attempt to set the appropriate
apic back into virtual wire mode.  This improves on previous versions
of this patch by by never setting bot the local apic and the ioapic
into veritual wire mode.

This code looks at data from the mptable to see if an ioapic has
an ExtInt input to make this decision.  A future improvement
is to figure out which apic or ioapic was in virtual wire mode
at boot time and to remember it.  That is potentially a more accurate
method, of selecting which apic to place in virutal wire mode.

Signed-off-by: Eric Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/apic.c    | 38 +++++++++++++++++++++++++++++++++++++-
 arch/x86_64/kernel/io_apic.c | 36 ++++++++++++++++++++++++++++++++++--
 include/asm-x86_64/apic.h    |  2 +-
 3 files changed, 72 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 2a6c893ccf6..375d369570c 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -133,7 +133,7 @@ void __init connect_bsp_APIC(void)
 	}
 }
 
-void disconnect_bsp_APIC(void)
+void disconnect_bsp_APIC(int virt_wire_setup)
 {
 	if (pic_mode) {
 		/*
@@ -146,6 +146,42 @@ void disconnect_bsp_APIC(void)
 		outb(0x70, 0x22);
 		outb(0x00, 0x23);
 	}
+	else {
+		/* Go back to Virtual Wire compatibility mode */
+		unsigned long value;
+
+		/* For the spurious interrupt use vector F, and enable it */
+		value = apic_read(APIC_SPIV);
+		value &= ~APIC_VECTOR_MASK;
+		value |= APIC_SPIV_APIC_ENABLED;
+		value |= 0xf;
+		apic_write_around(APIC_SPIV, value);
+
+		if (!virt_wire_setup) {
+			/* For LVT0 make it edge triggered, active high, external and enabled */
+			value = apic_read(APIC_LVT0);
+			value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
+				APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
+				APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED );
+			value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
+			value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_EXTINT);
+			apic_write_around(APIC_LVT0, value);
+		}
+		else {
+			/* Disable LVT0 */
+			apic_write_around(APIC_LVT0, APIC_LVT_MASKED);
+		}
+
+		/* For LVT1 make it edge triggered, active high, nmi and enabled */
+		value = apic_read(APIC_LVT1);
+		value &= ~(
+			APIC_MODE_MASK | APIC_SEND_PENDING |
+			APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
+			APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED);
+		value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
+		value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_NMI);
+		apic_write_around(APIC_LVT1, value);
+	}
 }
 
 void disable_local_APIC(void)
diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index afd87e64d0a..157190d986b 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -330,7 +330,7 @@ static int find_irq_entry(int apic, int pin, int type)
 /*
  * Find the pin to which IRQ[irq] (ISA) is connected
  */
-static int __init find_isa_irq_pin(int irq, int type)
+static int find_isa_irq_pin(int irq, int type)
 {
 	int i;
 
@@ -1132,12 +1132,44 @@ static void __init enable_IO_APIC(void)
  */
 void disable_IO_APIC(void)
 {
+	int pin;
 	/*
 	 * Clear the IO-APIC before rebooting:
 	 */
 	clear_IO_APIC();
 
-	disconnect_bsp_APIC();
+	/*
+	 * If the i82559 is routed through an IOAPIC
+	 * Put that IOAPIC in virtual wire mode
+	 * so legacy interrups can be delivered.
+	 */
+	pin = find_isa_irq_pin(0, mp_ExtINT);
+	if (pin != -1) {
+		struct IO_APIC_route_entry entry;
+		unsigned long flags;
+
+		memset(&entry, 0, sizeof(entry));
+		entry.mask            = 0; /* Enabled */
+		entry.trigger         = 0; /* Edge */
+		entry.irr             = 0;
+		entry.polarity        = 0; /* High */
+		entry.delivery_status = 0;
+		entry.dest_mode       = 0; /* Physical */
+		entry.delivery_mode   = 7; /* ExtInt */
+		entry.vector          = 0;
+		entry.dest.physical.physical_dest = 0;
+
+
+		/*
+		 * Add it to the IO-APIC irq-routing table:
+		 */
+		spin_lock_irqsave(&ioapic_lock, flags);
+		io_apic_write(0, 0x11+2*pin, *(((int *)&entry)+1));
+		io_apic_write(0, 0x10+2*pin, *(((int *)&entry)+0));
+		spin_unlock_irqrestore(&ioapic_lock, flags);
+	}
+
+	disconnect_bsp_APIC(pin != -1);
 }
 
 /*
diff --git a/include/asm-x86_64/apic.h b/include/asm-x86_64/apic.h
index e4b1017b8b2..16ec82e16b2 100644
--- a/include/asm-x86_64/apic.h
+++ b/include/asm-x86_64/apic.h
@@ -77,7 +77,7 @@ static inline void ack_APIC_irq(void)
 extern int get_maxlvt (void);
 extern void clear_local_APIC (void);
 extern void connect_bsp_APIC (void);
-extern void disconnect_bsp_APIC (void);
+extern void disconnect_bsp_APIC (int virt_wire_setup);
 extern void disable_local_APIC (void);
 extern int verify_local_APIC (void);
 extern void cache_APIC_registers (void);
-- 
cgit v1.2.3-70-g09d2


From 60bad7fadf59313a6359f8828bb0087884ad001a Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:46 -0700
Subject: [PATCH] kexec: vmlinux: fix physical addresses

In vmlinux.lds.h the code is carefull to define every section so vmlinux
properly reports the correct physical load address of code, as well as
it's virtual address.

The new SECURITY_INIT definition fails to follow that convention and
and causes incorrect physical address to appear in the vmlinux if
there are any security initcalls.

This patch updates the SECURITY_INIT to follow the convention in the rest of
the file.

Signed-off-by: Eric Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-generic/vmlinux.lds.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 99cef06a364..b3bb326ae5b 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -73,7 +73,7 @@
 	}
 
 #define SECURITY_INIT							\
-	.security_initcall.init : {					\
+	.security_initcall.init : AT(ADDR(.security_initcall.init) - LOAD_OFFSET) { \
 		VMLINUX_SYMBOL(__security_initcall_start) = .;		\
 		*(.security_initcall.init) 				\
 		VMLINUX_SYMBOL(__security_initcall_end) = .;		\
-- 
cgit v1.2.3-70-g09d2


From 3d345e3fc9e9177deb7c82e5c79e32d77eb63cce Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:49 -0700
Subject: [PATCH] kexec: x86: add CONFIG_PYSICAL_START

For one kernel to report a crash another kernel has created we need
to have 2 kernels loaded simultaneously in memory.  To accomplish this
the two kernels need to built to run at different physical addresses.

This patch adds the CONFIG_PHYSICAL_START option to the x86 kernel
so we can do just that.  You need to know what you are doing and
the ramifications are before changing this value, and most users
won't care so I have made it depend on CONFIG_EMBEDDED

bzImage kernels will work and run at a different address when compiled
with this option but they will still load at 1MB.  If you need a kernel
loaded at a different address as well you need to boot a vmlinux.

Signed-off-by: Eric Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig                | 11 +++++++++++
 arch/i386/boot/compressed/head.S |  7 ++++---
 arch/i386/boot/compressed/misc.c |  7 ++++---
 arch/i386/kernel/vmlinux.lds.S   |  2 +-
 include/asm-i386/page.h          |  3 +++
 5 files changed, 23 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index 961ab20032f..31567f4d333 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -942,6 +942,17 @@ config SECCOMP
 
 source kernel/Kconfig.hz
 
+config PHYSICAL_START
+	hex "Physical address where the kernel is loaded" if EMBEDDED
+	default "0x100000"
+	help
+	  This gives the physical address where the kernel is loaded.
+	  Primarily used in the case of kexec on panic where the
+	  fail safe kernel needs to run at a different address than
+	  the panic-ed kernel.
+
+	  Don't change this unless you know what you are doing.
+
 endmenu
 
 
diff --git a/arch/i386/boot/compressed/head.S b/arch/i386/boot/compressed/head.S
index c5e80b69e7d..b5893e4ecd3 100644
--- a/arch/i386/boot/compressed/head.S
+++ b/arch/i386/boot/compressed/head.S
@@ -25,6 +25,7 @@
 
 #include <linux/linkage.h>
 #include <asm/segment.h>
+#include <asm/page.h>
 
 	.globl startup_32
 	
@@ -74,7 +75,7 @@ startup_32:
 	popl %esi	# discard address
 	popl %esi	# real mode pointer
 	xorl %ebx,%ebx
-	ljmp $(__BOOT_CS), $0x100000
+	ljmp $(__BOOT_CS), $__PHYSICAL_START
 
 /*
  * We come here, if we were loaded high.
@@ -99,7 +100,7 @@ startup_32:
 	popl %ecx	# lcount
 	popl %edx	# high_buffer_start
 	popl %eax	# hcount
-	movl $0x100000,%edi
+	movl $__PHYSICAL_START,%edi
 	cli		# make sure we don't get interrupted
 	ljmp $(__BOOT_CS), $0x1000 # and jump to the move routine
 
@@ -124,5 +125,5 @@ move_routine_start:
 	movsl
 	movl %ebx,%esi	# Restore setup pointer
 	xorl %ebx,%ebx
-	ljmp $(__BOOT_CS), $0x100000
+	ljmp $(__BOOT_CS), $__PHYSICAL_START
 move_routine_end:
diff --git a/arch/i386/boot/compressed/misc.c b/arch/i386/boot/compressed/misc.c
index cedc55cc47d..e1330766965 100644
--- a/arch/i386/boot/compressed/misc.c
+++ b/arch/i386/boot/compressed/misc.c
@@ -13,6 +13,7 @@
 #include <linux/vmalloc.h>
 #include <linux/tty.h>
 #include <asm/io.h>
+#include <asm/page.h>
 
 /*
  * gzip declarations
@@ -308,7 +309,7 @@ static void setup_normal_output_buffer(void)
 #else
 	if ((RM_ALT_MEM_K > RM_EXT_MEM_K ? RM_ALT_MEM_K : RM_EXT_MEM_K) < 1024) error("Less than 2MB of memory");
 #endif
-	output_data = (char *)0x100000; /* Points to 1M */
+	output_data = (char *)__PHYSICAL_START; /* Normally Points to 1M */
 	free_mem_end_ptr = (long)real_mode;
 }
 
@@ -333,8 +334,8 @@ static void setup_output_buffer_if_we_run_high(struct moveparams *mv)
 	low_buffer_size = low_buffer_end - LOW_BUFFER_START;
 	high_loaded = 1;
 	free_mem_end_ptr = (long)high_buffer_start;
-	if ( (0x100000 + low_buffer_size) > ((ulg)high_buffer_start)) {
-		high_buffer_start = (uch *)(0x100000 + low_buffer_size);
+	if ( (__PHYSICAL_START + low_buffer_size) > ((ulg)high_buffer_start)) {
+		high_buffer_start = (uch *)(__PHYSICAL_START + low_buffer_size);
 		mv->hcount = 0; /* say: we need not to move high_buffer */
 	}
 	else mv->hcount = -1;
diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S
index e17ab69c1f0..7e01a528a83 100644
--- a/arch/i386/kernel/vmlinux.lds.S
+++ b/arch/i386/kernel/vmlinux.lds.S
@@ -14,7 +14,7 @@ ENTRY(phys_startup_32)
 jiffies = jiffies_64;
 SECTIONS
 {
-  . = LOAD_OFFSET + 0x100000;
+  . = __KERNEL_START;
   phys_startup_32 = startup_32 - LOAD_OFFSET;
   /* read-only */
   _text = .;			/* Text and read-only data */
diff --git a/include/asm-i386/page.h b/include/asm-i386/page.h
index dea8f8e6d86..8d93f732d72 100644
--- a/include/asm-i386/page.h
+++ b/include/asm-i386/page.h
@@ -126,9 +126,12 @@ extern int page_is_ram(unsigned long pagenr);
 
 #ifdef __ASSEMBLY__
 #define __PAGE_OFFSET		(0xC0000000)
+#define __PHYSICAL_START	CONFIG_PHYSICAL_START
 #else
 #define __PAGE_OFFSET		(0xC0000000UL)
+#define __PHYSICAL_START	((unsigned long)CONFIG_PHYSICAL_START)
 #endif
+#define __KERNEL_START		(__PAGE_OFFSET + __PHYSICAL_START)
 
 
 #define PAGE_OFFSET		((unsigned long)__PAGE_OFFSET)
-- 
cgit v1.2.3-70-g09d2


From d0537508a9921efced238b20967e50e519ac34af Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:52 -0700
Subject: [PATCH] kexec: x86_64: add CONFIG_PHYSICAL_START

For one kernel to report a crash another kernel has created we need
to have 2 kernels loaded simultaneously in memory.  To accomplish this
the two kernels need to built to run at different physical addresses.

This patch adds the CONFIG_PHYSICAL_START option to the x86_64 kernel
so we can do just that.  You need to know what you are doing and
the ramifications are before changing this value, and most users
won't care so I have made it depend on CONFIG_EMBEDDED

bzImage kernels will work and run at a different address when compiled
with this option but they will still load at 1MB.  If you need a kernel
loaded at a different address as well you need to boot a vmlinux.

Signed-off-by: Eric Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/Kconfig                | 11 +++++++++++
 arch/x86_64/boot/compressed/head.S |  7 ++++---
 arch/x86_64/boot/compressed/misc.c |  7 ++++---
 arch/x86_64/kernel/head.S          | 18 +++++++++---------
 include/asm-x86_64/page.h          |  6 ++++--
 5 files changed, 32 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index a853d87ff7e..56c02cf6397 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -369,6 +369,17 @@ config X86_MCE_INTEL
 	   Additional support for intel specific MCE features such as
 	   the thermal monitor.
 
+config PHYSICAL_START
+	hex "Physical address where the kernel is loaded" if EMBEDDED
+	default "0x100000"
+	help
+	  This gives the physical address where the kernel is loaded.
+	  Primarily used in the case of kexec on panic where the
+	  fail safe kernel needs to run at a different address than
+	  the panic-ed kernel.
+
+	  Don't change this unless you know what you are doing.
+
 config SECCOMP
 	bool "Enable seccomp to safely compute untrusted bytecode"
 	depends on PROC_FS
diff --git a/arch/x86_64/boot/compressed/head.S b/arch/x86_64/boot/compressed/head.S
index 27264dbd575..b0df2aeff5d 100644
--- a/arch/x86_64/boot/compressed/head.S
+++ b/arch/x86_64/boot/compressed/head.S
@@ -28,6 +28,7 @@
 
 #include <linux/linkage.h>
 #include <asm/segment.h>
+#include <asm/page.h>
 
 	.code32
 	.globl startup_32
@@ -77,7 +78,7 @@ startup_32:
 	jnz  3f
 	addl $8,%esp
 	xorl %ebx,%ebx
-	ljmp $(__KERNEL_CS), $0x100000
+	ljmp $(__KERNEL_CS), $__PHYSICAL_START
 
 /*
  * We come here, if we were loaded high.
@@ -103,7 +104,7 @@ startup_32:
 	popl %ecx	# lcount
 	popl %edx	# high_buffer_start
 	popl %eax	# hcount
-	movl $0x100000,%edi
+	movl $__PHYSICAL_START,%edi
 	cli		# make sure we don't get interrupted
 	ljmp $(__KERNEL_CS), $0x1000 # and jump to the move routine
 
@@ -128,7 +129,7 @@ move_routine_start:
 	movsl
 	movl %ebx,%esi	# Restore setup pointer
 	xorl %ebx,%ebx
-	ljmp $(__KERNEL_CS), $0x100000
+	ljmp $(__KERNEL_CS), $__PHYSICAL_START
 move_routine_end:
 
 
diff --git a/arch/x86_64/boot/compressed/misc.c b/arch/x86_64/boot/compressed/misc.c
index c8b9216f9e6..c43cedb5c1d 100644
--- a/arch/x86_64/boot/compressed/misc.c
+++ b/arch/x86_64/boot/compressed/misc.c
@@ -11,6 +11,7 @@
 
 #include "miscsetup.h"
 #include <asm/io.h>
+#include <asm/page.h>
 
 /*
  * gzip declarations
@@ -284,7 +285,7 @@ void setup_normal_output_buffer(void)
 #else
 	if ((ALT_MEM_K > EXT_MEM_K ? ALT_MEM_K : EXT_MEM_K) < 1024) error("Less than 2MB of memory");
 #endif
-	output_data = (char *)0x100000; /* Points to 1M */
+	output_data = (char *)__PHYSICAL_START; /* Normally Points to 1M */
 	free_mem_end_ptr = (long)real_mode;
 }
 
@@ -307,8 +308,8 @@ void setup_output_buffer_if_we_run_high(struct moveparams *mv)
 	low_buffer_size = low_buffer_end - LOW_BUFFER_START;
 	high_loaded = 1;
 	free_mem_end_ptr = (long)high_buffer_start;
-	if ( (0x100000 + low_buffer_size) > ((ulg)high_buffer_start)) {
-		high_buffer_start = (uch *)(0x100000 + low_buffer_size);
+	if ( (__PHYSICAL_START + low_buffer_size) > ((ulg)high_buffer_start)) {
+		high_buffer_start = (uch *)(__PHYSICAL_START + low_buffer_size);
 		mv->hcount = 0; /* say: we need not to move high_buffer */
 	}
 	else mv->hcount = -1;
diff --git a/arch/x86_64/kernel/head.S b/arch/x86_64/kernel/head.S
index 9bd2e7a4b81..8d765aa77a2 100644
--- a/arch/x86_64/kernel/head.S
+++ b/arch/x86_64/kernel/head.S
@@ -248,23 +248,23 @@ ENTRY(_stext)
 	 */
 .org 0x1000
 ENTRY(init_level4_pgt)
-	.quad	0x0000000000102007		/* -> level3_ident_pgt */
+	.quad	0x0000000000002007 + __PHYSICAL_START	/* -> level3_ident_pgt */
 	.fill	255,8,0
-	.quad	0x000000000010a007
+	.quad	0x000000000000a007 + __PHYSICAL_START
 	.fill	254,8,0
 	/* (2^48-(2*1024*1024*1024))/(2^39) = 511 */
-	.quad	0x0000000000103007		/* -> level3_kernel_pgt */
+	.quad	0x0000000000003007 + __PHYSICAL_START	/* -> level3_kernel_pgt */
 
 .org 0x2000
 ENTRY(level3_ident_pgt)
-	.quad	0x0000000000104007
+	.quad	0x0000000000004007 + __PHYSICAL_START
 	.fill	511,8,0
 
 .org 0x3000
 ENTRY(level3_kernel_pgt)
 	.fill	510,8,0
 	/* (2^48-(2*1024*1024*1024)-((2^39)*511))/(2^30) = 510 */
-	.quad	0x0000000000105007		/* -> level2_kernel_pgt */
+	.quad	0x0000000000005007 + __PHYSICAL_START	/* -> level2_kernel_pgt */
 	.fill	1,8,0
 
 .org 0x4000
@@ -337,17 +337,17 @@ ENTRY(empty_bad_pmd_table)
 
 .org 0xa000
 ENTRY(level3_physmem_pgt)
-	.quad	0x0000000000105007		/* -> level2_kernel_pgt (so that __va works even before pagetable_init) */
+	.quad	0x0000000000005007 + __PHYSICAL_START	/* -> level2_kernel_pgt (so that __va works even before pagetable_init) */
 
 	.org 0xb000
 #ifdef CONFIG_ACPI_SLEEP
 ENTRY(wakeup_level4_pgt)
-	.quad	0x0000000000102007		/* -> level3_ident_pgt */
+	.quad	0x0000000000002007 + __PHYSICAL_START	/* -> level3_ident_pgt */
 	.fill	255,8,0
-	.quad	0x000000000010a007
+	.quad	0x000000000000a007 + __PHYSICAL_START
 	.fill	254,8,0
 	/* (2^48-(2*1024*1024*1024))/(2^39) = 511 */
-	.quad	0x0000000000103007		/* -> level3_kernel_pgt */
+	.quad	0x0000000000003007 + __PHYSICAL_START	/* -> level3_kernel_pgt */
 #endif
 
 	.data
diff --git a/include/asm-x86_64/page.h b/include/asm-x86_64/page.h
index 60130f4ca98..431318764af 100644
--- a/include/asm-x86_64/page.h
+++ b/include/asm-x86_64/page.h
@@ -64,12 +64,14 @@ typedef struct { unsigned long pgprot; } pgprot_t;
 #define __pgd(x) ((pgd_t) { (x) } )
 #define __pgprot(x)	((pgprot_t) { (x) } )
 
-#define __START_KERNEL		0xffffffff80100000UL
+#define __PHYSICAL_START	((unsigned long)CONFIG_PHYSICAL_START)
+#define __START_KERNEL		(__START_KERNEL_map + __PHYSICAL_START)
 #define __START_KERNEL_map	0xffffffff80000000UL
 #define __PAGE_OFFSET           0xffff810000000000UL
 
 #else
-#define __START_KERNEL		0xffffffff80100000
+#define __PHYSICAL_START	CONFIG_PHYSICAL_START
+#define __START_KERNEL		(__START_KERNEL_map + __PHYSICAL_START)
 #define __START_KERNEL_map	0xffffffff80000000
 #define __PAGE_OFFSET           0xffff810000000000
 #endif /* !__ASSEMBLY__ */
-- 
cgit v1.2.3-70-g09d2


From dc009d92435f99498cbc579ce76bf28e837e2c14 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:52 -0700
Subject: [PATCH] kexec: add kexec syscalls

This patch introduces the architecture independent implementation the
sys_kexec_load, the compat_sys_kexec_load system calls.

Kexec on panic support has been integrated into the core patch and is
relatively clean.

In addition the hopefully architecture independent option
crashkernel=size@location has been docuemented.  It's purpose is to reserve
space for the panic kernel to live, and where no DMA transfer will ever be
setup to access.

Signed-off-by: Eric Biederman <ebiederm@xmission.com>
Signed-off-by: Alexander Nyberg <alexn@telia.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/kernel-parameters.txt |    4 +
 MAINTAINERS                         |   10 +
 include/linux/kexec.h               |  127 +++++
 include/linux/reboot.h              |    3 +
 include/linux/syscalls.h            |    5 +-
 kernel/Makefile                     |    1 +
 kernel/kexec.c                      | 1036 +++++++++++++++++++++++++++++++++++
 kernel/panic.c                      |   23 +-
 kernel/sys.c                        |   20 +
 kernel/sys_ni.c                     |    2 +
 10 files changed, 1227 insertions(+), 4 deletions(-)
 create mode 100644 include/linux/kexec.h
 create mode 100644 kernel/kexec.c

(limited to 'include')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 86db43fd6b0..560ff5ae3fd 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -358,6 +358,10 @@ running once the system is up.
 	cpia_pp=	[HW,PPT]
 			Format: { parport<nr> | auto | none }
 
+	crashkernel=nn[KMG]@ss[KMG]
+			[KNL] Reserve a chunk of physical memory to
+			hold a kernel to switch to with kexec on panic.
+
 	cs4232=		[HW,OSS]
 			Format: <io>,<irq>,<dma>,<dma2>,<mpuio>,<mpuirq>
 
diff --git a/MAINTAINERS b/MAINTAINERS
index dbdd8494b2e..81728572799 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1330,6 +1330,16 @@ M:	rml@novell.com
 L:	linux-kernel@vger.kernel.org
 S:	Maintained
 
+KEXEC
+P:	Eric Biederman
+P:	Randy Dunlap
+M:	ebiederm@xmission.com
+M:	rddunlap@osdl.org
+W:	http://www.xmission.com/~ebiederm/files/kexec/
+L:	linux-kernel@vger.kernel.org
+L:	fastboot@osdl.org
+S:	Maintained
+
 LANMEDIA WAN CARD DRIVER
 P:	Andrew Stanley-Jones
 M:	asj@lanmedia.com
diff --git a/include/linux/kexec.h b/include/linux/kexec.h
new file mode 100644
index 00000000000..e3fc35f4e35
--- /dev/null
+++ b/include/linux/kexec.h
@@ -0,0 +1,127 @@
+#ifndef LINUX_KEXEC_H
+#define LINUX_KEXEC_H
+
+#ifdef CONFIG_KEXEC
+#include <linux/types.h>
+#include <linux/list.h>
+#include <linux/linkage.h>
+#include <linux/compat.h>
+#include <asm/kexec.h>
+
+/* Verify architecture specific macros are defined */
+
+#ifndef KEXEC_SOURCE_MEMORY_LIMIT
+#error KEXEC_SOURCE_MEMORY_LIMIT not defined
+#endif
+
+#ifndef KEXEC_DESTINATION_MEMORY_LIMIT
+#error KEXEC_DESTINATION_MEMORY_LIMIT not defined
+#endif
+
+#ifndef KEXEC_CONTROL_MEMORY_LIMIT
+#error KEXEC_CONTROL_MEMORY_LIMIT not defined
+#endif
+
+#ifndef KEXEC_CONTROL_CODE_SIZE
+#error KEXEC_CONTROL_CODE_SIZE not defined
+#endif
+
+#ifndef KEXEC_ARCH
+#error KEXEC_ARCH not defined
+#endif
+
+/*
+ * This structure is used to hold the arguments that are used when loading
+ * kernel binaries.
+ */
+
+typedef unsigned long kimage_entry_t;
+#define IND_DESTINATION  0x1
+#define IND_INDIRECTION  0x2
+#define IND_DONE         0x4
+#define IND_SOURCE       0x8
+
+#define KEXEC_SEGMENT_MAX 8
+struct kexec_segment {
+	void __user *buf;
+	size_t bufsz;
+	unsigned long mem;	/* User space sees this as a (void *) ... */
+	size_t memsz;
+};
+
+#ifdef CONFIG_COMPAT
+struct compat_kexec_segment {
+	compat_uptr_t buf;
+	compat_size_t bufsz;
+	compat_ulong_t mem;	/* User space sees this as a (void *) ... */
+	compat_size_t memsz;
+};
+#endif
+
+struct kimage {
+	kimage_entry_t head;
+	kimage_entry_t *entry;
+	kimage_entry_t *last_entry;
+
+	unsigned long destination;
+
+	unsigned long start;
+	struct page *control_code_page;
+
+	unsigned long nr_segments;
+	struct kexec_segment segment[KEXEC_SEGMENT_MAX];
+
+	struct list_head control_pages;
+	struct list_head dest_pages;
+	struct list_head unuseable_pages;
+
+	/* Address of next control page to allocate for crash kernels. */
+	unsigned long control_page;
+
+	/* Flags to indicate special processing */
+	unsigned int type : 1;
+#define KEXEC_TYPE_DEFAULT 0
+#define KEXEC_TYPE_CRASH   1
+};
+
+
+
+/* kexec interface functions */
+extern NORET_TYPE void machine_kexec(struct kimage *image) ATTRIB_NORET;
+extern int machine_kexec_prepare(struct kimage *image);
+extern void machine_kexec_cleanup(struct kimage *image);
+extern asmlinkage long sys_kexec_load(unsigned long entry,
+	unsigned long nr_segments, struct kexec_segment __user *segments,
+	unsigned long flags);
+#ifdef CONFIG_COMPAT
+extern asmlinkage long compat_sys_kexec_load(unsigned long entry,
+	unsigned long nr_segments, struct compat_kexec_segment __user *segments,
+	unsigned long flags);
+#endif
+extern struct page *kimage_alloc_control_pages(struct kimage *image, unsigned int order);
+extern void crash_kexec(void);
+extern struct kimage *kexec_image;
+
+#define KEXEC_ON_CRASH  0x00000001
+#define KEXEC_ARCH_MASK 0xffff0000
+
+/* These values match the ELF architecture values.
+ * Unless there is a good reason that should continue to be the case.
+ */
+#define KEXEC_ARCH_DEFAULT ( 0 << 16)
+#define KEXEC_ARCH_386     ( 3 << 16)
+#define KEXEC_ARCH_X86_64  (62 << 16)
+#define KEXEC_ARCH_PPC     (20 << 16)
+#define KEXEC_ARCH_PPC64   (21 << 16)
+#define KEXEC_ARCH_IA_64   (50 << 16)
+
+#define KEXEC_FLAGS    (KEXEC_ON_CRASH)  /* List of defined/legal kexec flags */
+
+/* Location of a reserved region to hold the crash kernel.
+ */
+extern struct resource crashk_res;
+
+#else /* !CONFIG_KEXEC */
+static inline void crash_kexec(void) { }
+#endif /* CONFIG_KEXEC */
+#endif /* LINUX_KEXEC_H */
diff --git a/include/linux/reboot.h b/include/linux/reboot.h
index d60fafc8bdc..c5a05e16edb 100644
--- a/include/linux/reboot.h
+++ b/include/linux/reboot.h
@@ -51,6 +51,9 @@ extern void machine_restart(char *cmd);
 extern void machine_halt(void);
 extern void machine_power_off(void);
 
+extern void machine_shutdown(void);
+extern void machine_crash_shutdown(void);
+
 #endif
 
 #endif /* _LINUX_REBOOT_H */
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index c39f6f72cbb..7ba8f8f747a 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -159,8 +159,9 @@ asmlinkage long sys_shutdown(int, int);
 asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd,
 				void __user *arg);
 asmlinkage long sys_restart_syscall(void);
-asmlinkage long sys_kexec_load(void *entry, unsigned long nr_segments,
-			struct kexec_segment *segments, unsigned long flags);
+asmlinkage long sys_kexec_load(unsigned long entry,
+	unsigned long nr_segments, struct kexec_segment __user *segments,
+	unsigned long flags);
 
 asmlinkage long sys_exit(int error_code);
 asmlinkage void sys_exit_group(int error_code);
diff --git a/kernel/Makefile b/kernel/Makefile
index b01d26fe8db..cfc8b0dea95 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -17,6 +17,7 @@ obj-$(CONFIG_MODULES) += module.o
 obj-$(CONFIG_KALLSYMS) += kallsyms.o
 obj-$(CONFIG_PM) += power/
 obj-$(CONFIG_BSD_PROCESS_ACCT) += acct.o
+obj-$(CONFIG_KEXEC) += kexec.o
 obj-$(CONFIG_COMPAT) += compat.o
 obj-$(CONFIG_CPUSETS) += cpuset.o
 obj-$(CONFIG_IKCONFIG) += configs.o
diff --git a/kernel/kexec.c b/kernel/kexec.c
new file mode 100644
index 00000000000..def9c73ec9a
--- /dev/null
+++ b/kernel/kexec.c
@@ -0,0 +1,1036 @@
+/*
+ * kexec.c - kexec system call
+ * Copyright (C) 2002-2004 Eric Biederman  <ebiederm@xmission.com>
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2.  See the file COPYING for more details.
+ */
+
+#include <linux/mm.h>
+#include <linux/file.h>
+#include <linux/slab.h>
+#include <linux/fs.h>
+#include <linux/kexec.h>
+#include <linux/spinlock.h>
+#include <linux/list.h>
+#include <linux/highmem.h>
+#include <linux/syscalls.h>
+#include <linux/reboot.h>
+#include <linux/syscalls.h>
+#include <linux/ioport.h>
+#include <asm/page.h>
+#include <asm/uaccess.h>
+#include <asm/io.h>
+#include <asm/system.h>
+#include <asm/semaphore.h>
+
+/* Location of the reserved area for the crash kernel */
+struct resource crashk_res = {
+	.name  = "Crash kernel",
+	.start = 0,
+	.end   = 0,
+	.flags = IORESOURCE_BUSY | IORESOURCE_MEM
+};
+
+/*
+ * When kexec transitions to the new kernel there is a one-to-one
+ * mapping between physical and virtual addresses.  On processors
+ * where you can disable the MMU this is trivial, and easy.  For
+ * others it is still a simple predictable page table to setup.
+ *
+ * In that environment kexec copies the new kernel to its final
+ * resting place.  This means I can only support memory whose
+ * physical address can fit in an unsigned long.  In particular
+ * addresses where (pfn << PAGE_SHIFT) > ULONG_MAX cannot be handled.
+ * If the assembly stub has more restrictive requirements
+ * KEXEC_SOURCE_MEMORY_LIMIT and KEXEC_DEST_MEMORY_LIMIT can be
+ * defined more restrictively in <asm/kexec.h>.
+ *
+ * The code for the transition from the current kernel to the
+ * the new kernel is placed in the control_code_buffer, whose size
+ * is given by KEXEC_CONTROL_CODE_SIZE.  In the best case only a single
+ * page of memory is necessary, but some architectures require more.
+ * Because this memory must be identity mapped in the transition from
+ * virtual to physical addresses it must live in the range
+ * 0 - TASK_SIZE, as only the user space mappings are arbitrarily
+ * modifiable.
+ *
+ * The assembly stub in the control code buffer is passed a linked list
+ * of descriptor pages detailing the source pages of the new kernel,
+ * and the destination addresses of those source pages.  As this data
+ * structure is not used in the context of the current OS, it must
+ * be self-contained.
+ *
+ * The code has been made to work with highmem pages and will use a
+ * destination page in its final resting place (if it happens
+ * to allocate it).  The end product of this is that most of the
+ * physical address space, and most of RAM can be used.
+ *
+ * Future directions include:
+ *  - allocating a page table with the control code buffer identity
+ *    mapped, to simplify machine_kexec and make kexec_on_panic more
+ *    reliable.
+ */
+
+/*
+ * KIMAGE_NO_DEST is an impossible destination address..., for
+ * allocating pages whose destination address we do not care about.
+ */
+#define KIMAGE_NO_DEST (-1UL)
+
+static int kimage_is_destination_range(
+	struct kimage *image, unsigned long start, unsigned long end);
+static struct page *kimage_alloc_page(struct kimage *image, unsigned int gfp_mask, unsigned long dest);
+
+static int do_kimage_alloc(struct kimage **rimage, unsigned long entry,
+	unsigned long nr_segments, struct kexec_segment __user *segments)
+{
+	size_t segment_bytes;
+	struct kimage *image;
+	unsigned long i;
+	int result;
+
+	/* Allocate a controlling structure */
+	result = -ENOMEM;
+	image = kmalloc(sizeof(*image), GFP_KERNEL);
+	if (!image) {
+		goto out;
+	}
+	memset(image, 0, sizeof(*image));
+	image->head = 0;
+	image->entry = &image->head;
+	image->last_entry = &image->head;
+	image->control_page = ~0; /* By default this does not apply */
+	image->start = entry;
+	image->type = KEXEC_TYPE_DEFAULT;
+
+	/* Initialize the list of control pages */
+	INIT_LIST_HEAD(&image->control_pages);
+
+	/* Initialize the list of destination pages */
+	INIT_LIST_HEAD(&image->dest_pages);
+
+	/* Initialize the list of unuseable pages */
+	INIT_LIST_HEAD(&image->unuseable_pages);
+
+	/* Read in the segments */
+	image->nr_segments = nr_segments;
+	segment_bytes = nr_segments * sizeof(*segments);
+	result = copy_from_user(image->segment, segments, segment_bytes);
+	if (result)
+		goto out;
+
+	/*
+	 * Verify we have good destination addresses.  The caller is
+	 * responsible for making certain we don't attempt to load
+	 * the new image into invalid or reserved areas of RAM.  This
+	 * just verifies it is an address we can use.
+	 *
+	 * Since the kernel does everything in page size chunks ensure
+	 * the destination addreses are page aligned.  Too many
+	 * special cases crop of when we don't do this.  The most
+	 * insidious is getting overlapping destination addresses
+	 * simply because addresses are changed to page size
+	 * granularity.
+	 */
+	result = -EADDRNOTAVAIL;
+	for (i = 0; i < nr_segments; i++) {
+		unsigned long mstart, mend;
+		mstart = image->segment[i].mem;
+		mend   = mstart + image->segment[i].memsz;
+		if ((mstart & ~PAGE_MASK) || (mend & ~PAGE_MASK))
+			goto out;
+		if (mend >= KEXEC_DESTINATION_MEMORY_LIMIT)
+			goto out;
+	}
+
+	/* Verify our destination addresses do not overlap.
+	 * If we alloed overlapping destination addresses
+	 * through very weird things can happen with no
+	 * easy explanation as one segment stops on another.
+	 */
+	result = -EINVAL;
+	for(i = 0; i < nr_segments; i++) {
+		unsigned long mstart, mend;
+		unsigned long j;
+		mstart = image->segment[i].mem;
+		mend   = mstart + image->segment[i].memsz;
+		for(j = 0; j < i; j++) {
+			unsigned long pstart, pend;
+			pstart = image->segment[j].mem;
+			pend   = pstart + image->segment[j].memsz;
+			/* Do the segments overlap ? */
+			if ((mend > pstart) && (mstart < pend))
+				goto out;
+		}
+	}
+
+	/* Ensure our buffer sizes are strictly less than
+	 * our memory sizes.  This should always be the case,
+	 * and it is easier to check up front than to be surprised
+	 * later on.
+	 */
+	result = -EINVAL;
+	for(i = 0; i < nr_segments; i++) {
+		if (image->segment[i].bufsz > image->segment[i].memsz)
+			goto out;
+	}
+
+
+	result = 0;
+ out:
+	if (result == 0) {
+		*rimage = image;
+	} else {
+		kfree(image);
+	}
+	return result;
+
+}
+
+static int kimage_normal_alloc(struct kimage **rimage, unsigned long entry,
+	unsigned long nr_segments, struct kexec_segment __user *segments)
+{
+	int result;
+	struct kimage *image;
+
+	/* Allocate and initialize a controlling structure */
+	image = NULL;
+	result = do_kimage_alloc(&image, entry, nr_segments, segments);
+	if (result) {
+		goto out;
+	}
+	*rimage = image;
+
+	/*
+	 * Find a location for the control code buffer, and add it
+	 * the vector of segments so that it's pages will also be
+	 * counted as destination pages.
+	 */
+	result = -ENOMEM;
+	image->control_code_page = kimage_alloc_control_pages(image,
+		get_order(KEXEC_CONTROL_CODE_SIZE));
+	if (!image->control_code_page) {
+		printk(KERN_ERR "Could not allocate control_code_buffer\n");
+		goto out;
+	}
+
+	result = 0;
+ out:
+	if (result == 0) {
+		*rimage = image;
+	} else {
+		kfree(image);
+	}
+	return result;
+}
+
+static int kimage_crash_alloc(struct kimage **rimage, unsigned long entry,
+	unsigned long nr_segments, struct kexec_segment *segments)
+{
+	int result;
+	struct kimage *image;
+	unsigned long i;
+
+	image = NULL;
+	/* Verify we have a valid entry point */
+	if ((entry < crashk_res.start) || (entry > crashk_res.end)) {
+		result = -EADDRNOTAVAIL;
+		goto out;
+	}
+
+	/* Allocate and initialize a controlling structure */
+	result = do_kimage_alloc(&image, entry, nr_segments, segments);
+	if (result) {
+		goto out;
+	}
+
+	/* Enable the special crash kernel control page
+	 * allocation policy.
+	 */
+	image->control_page = crashk_res.start;
+	image->type = KEXEC_TYPE_CRASH;
+
+	/*
+	 * Verify we have good destination addresses.  Normally
+	 * the caller is responsible for making certain we don't
+	 * attempt to load the new image into invalid or reserved
+	 * areas of RAM.  But crash kernels are preloaded into a
+	 * reserved area of ram.  We must ensure the addresses
+	 * are in the reserved area otherwise preloading the
+	 * kernel could corrupt things.
+	 */
+	result = -EADDRNOTAVAIL;
+	for (i = 0; i < nr_segments; i++) {
+		unsigned long mstart, mend;
+		mstart = image->segment[i].mem;
+		mend = mstart + image->segment[i].memsz;
+		/* Ensure we are within the crash kernel limits */
+		if ((mstart < crashk_res.start) || (mend > crashk_res.end))
+			goto out;
+	}
+
+
+	/*
+	 * Find a location for the control code buffer, and add
+	 * the vector of segments so that it's pages will also be
+	 * counted as destination pages.
+	 */
+	result = -ENOMEM;
+	image->control_code_page = kimage_alloc_control_pages(image,
+		get_order(KEXEC_CONTROL_CODE_SIZE));
+	if (!image->control_code_page) {
+		printk(KERN_ERR "Could not allocate control_code_buffer\n");
+		goto out;
+	}
+
+	result = 0;
+ out:
+	if (result == 0) {
+		*rimage = image;
+	} else {
+		kfree(image);
+	}
+	return result;
+}
+
+static int kimage_is_destination_range(
+	struct kimage *image, unsigned long start, unsigned long end)
+{
+	unsigned long i;
+
+	for (i = 0; i < image->nr_segments; i++) {
+		unsigned long mstart, mend;
+		mstart = image->segment[i].mem;
+		mend   = mstart + image->segment[i].memsz;
+		if ((end > mstart) && (start < mend)) {
+			return 1;
+		}
+	}
+	return 0;
+}
+
+static struct page *kimage_alloc_pages(unsigned int gfp_mask, unsigned int order)
+{
+	struct page *pages;
+	pages = alloc_pages(gfp_mask, order);
+	if (pages) {
+		unsigned int count, i;
+		pages->mapping = NULL;
+		pages->private = order;
+		count = 1 << order;
+		for(i = 0; i < count; i++) {
+			SetPageReserved(pages + i);
+		}
+	}
+	return pages;
+}
+
+static void kimage_free_pages(struct page *page)
+{
+	unsigned int order, count, i;
+	order = page->private;
+	count = 1 << order;
+	for(i = 0; i < count; i++) {
+		ClearPageReserved(page + i);
+	}
+	__free_pages(page, order);
+}
+
+static void kimage_free_page_list(struct list_head *list)
+{
+	struct list_head *pos, *next;
+	list_for_each_safe(pos, next, list) {
+		struct page *page;
+
+		page = list_entry(pos, struct page, lru);
+		list_del(&page->lru);
+
+		kimage_free_pages(page);
+	}
+}
+
+static struct page *kimage_alloc_normal_control_pages(
+	struct kimage *image, unsigned int order)
+{
+	/* Control pages are special, they are the intermediaries
+	 * that are needed while we copy the rest of the pages
+	 * to their final resting place.  As such they must
+	 * not conflict with either the destination addresses
+	 * or memory the kernel is already using.
+	 *
+	 * The only case where we really need more than one of
+	 * these are for architectures where we cannot disable
+	 * the MMU and must instead generate an identity mapped
+	 * page table for all of the memory.
+	 *
+	 * At worst this runs in O(N) of the image size.
+	 */
+	struct list_head extra_pages;
+	struct page *pages;
+	unsigned int count;
+
+	count = 1 << order;
+	INIT_LIST_HEAD(&extra_pages);
+
+	/* Loop while I can allocate a page and the page allocated
+	 * is a destination page.
+	 */
+	do {
+		unsigned long pfn, epfn, addr, eaddr;
+		pages = kimage_alloc_pages(GFP_KERNEL, order);
+		if (!pages)
+			break;
+		pfn   = page_to_pfn(pages);
+		epfn  = pfn + count;
+		addr  = pfn << PAGE_SHIFT;
+		eaddr = epfn << PAGE_SHIFT;
+		if ((epfn >= (KEXEC_CONTROL_MEMORY_LIMIT >> PAGE_SHIFT)) ||
+			kimage_is_destination_range(image, addr, eaddr))
+		{
+			list_add(&pages->lru, &extra_pages);
+			pages = NULL;
+		}
+	} while(!pages);
+	if (pages) {
+		/* Remember the allocated page... */
+		list_add(&pages->lru, &image->control_pages);
+
+		/* Because the page is already in it's destination
+		 * location we will never allocate another page at
+		 * that address.  Therefore kimage_alloc_pages
+		 * will not return it (again) and we don't need
+		 * to give it an entry in image->segment[].
+		 */
+	}
+	/* Deal with the destination pages I have inadvertently allocated.
+	 *
+	 * Ideally I would convert multi-page allocations into single
+	 * page allocations, and add everyting to image->dest_pages.
+	 *
+	 * For now it is simpler to just free the pages.
+	 */
+	kimage_free_page_list(&extra_pages);
+	return pages;
+
+}
+
+static struct page *kimage_alloc_crash_control_pages(
+	struct kimage *image, unsigned int order)
+{
+	/* Control pages are special, they are the intermediaries
+	 * that are needed while we copy the rest of the pages
+	 * to their final resting place.  As such they must
+	 * not conflict with either the destination addresses
+	 * or memory the kernel is already using.
+	 *
+	 * Control pages are also the only pags we must allocate
+	 * when loading a crash kernel.  All of the other pages
+	 * are specified by the segments and we just memcpy
+	 * into them directly.
+	 *
+	 * The only case where we really need more than one of
+	 * these are for architectures where we cannot disable
+	 * the MMU and must instead generate an identity mapped
+	 * page table for all of the memory.
+	 *
+	 * Given the low demand this implements a very simple
+	 * allocator that finds the first hole of the appropriate
+	 * size in the reserved memory region, and allocates all
+	 * of the memory up to and including the hole.
+	 */
+	unsigned long hole_start, hole_end, size;
+	struct page *pages;
+	pages = NULL;
+	size = (1 << order) << PAGE_SHIFT;
+	hole_start = (image->control_page + (size - 1)) & ~(size - 1);
+	hole_end   = hole_start + size - 1;
+	while(hole_end <= crashk_res.end) {
+		unsigned long i;
+		if (hole_end > KEXEC_CONTROL_MEMORY_LIMIT) {
+			break;
+		}
+		if (hole_end > crashk_res.end) {
+			break;
+		}
+		/* See if I overlap any of the segments */
+		for(i = 0; i < image->nr_segments; i++) {
+			unsigned long mstart, mend;
+			mstart = image->segment[i].mem;
+			mend   = mstart + image->segment[i].memsz - 1;
+			if ((hole_end >= mstart) && (hole_start <= mend)) {
+				/* Advance the hole to the end of the segment */
+				hole_start = (mend + (size - 1)) & ~(size - 1);
+				hole_end   = hole_start + size - 1;
+				break;
+			}
+		}
+		/* If I don't overlap any segments I have found my hole! */
+		if (i == image->nr_segments) {
+			pages = pfn_to_page(hole_start >> PAGE_SHIFT);
+			break;
+		}
+	}
+	if (pages) {
+		image->control_page = hole_end;
+	}
+	return pages;
+}
+
+
+struct page *kimage_alloc_control_pages(
+	struct kimage *image, unsigned int order)
+{
+	struct page *pages = NULL;
+	switch(image->type) {
+	case KEXEC_TYPE_DEFAULT:
+		pages = kimage_alloc_normal_control_pages(image, order);
+		break;
+	case KEXEC_TYPE_CRASH:
+		pages = kimage_alloc_crash_control_pages(image, order);
+		break;
+	}
+	return pages;
+}
+
+static int kimage_add_entry(struct kimage *image, kimage_entry_t entry)
+{
+	if (*image->entry != 0) {
+		image->entry++;
+	}
+	if (image->entry == image->last_entry) {
+		kimage_entry_t *ind_page;
+		struct page *page;
+		page = kimage_alloc_page(image, GFP_KERNEL, KIMAGE_NO_DEST);
+		if (!page) {
+			return -ENOMEM;
+		}
+		ind_page = page_address(page);
+		*image->entry = virt_to_phys(ind_page) | IND_INDIRECTION;
+		image->entry = ind_page;
+		image->last_entry =
+			ind_page + ((PAGE_SIZE/sizeof(kimage_entry_t)) - 1);
+	}
+	*image->entry = entry;
+	image->entry++;
+	*image->entry = 0;
+	return 0;
+}
+
+static int kimage_set_destination(
+	struct kimage *image, unsigned long destination)
+{
+	int result;
+
+	destination &= PAGE_MASK;
+	result = kimage_add_entry(image, destination | IND_DESTINATION);
+	if (result == 0) {
+		image->destination = destination;
+	}
+	return result;
+}
+
+
+static int kimage_add_page(struct kimage *image, unsigned long page)
+{
+	int result;
+
+	page &= PAGE_MASK;
+	result = kimage_add_entry(image, page | IND_SOURCE);
+	if (result == 0) {
+		image->destination += PAGE_SIZE;
+	}
+	return result;
+}
+
+
+static void kimage_free_extra_pages(struct kimage *image)
+{
+	/* Walk through and free any extra destination pages I may have */
+	kimage_free_page_list(&image->dest_pages);
+
+	/* Walk through and free any unuseable pages I have cached */
+	kimage_free_page_list(&image->unuseable_pages);
+
+}
+static int kimage_terminate(struct kimage *image)
+{
+	if (*image->entry != 0) {
+		image->entry++;
+	}
+	*image->entry = IND_DONE;
+	return 0;
+}
+
+#define for_each_kimage_entry(image, ptr, entry) \
+	for (ptr = &image->head; (entry = *ptr) && !(entry & IND_DONE); \
+		ptr = (entry & IND_INDIRECTION)? \
+			phys_to_virt((entry & PAGE_MASK)): ptr +1)
+
+static void kimage_free_entry(kimage_entry_t entry)
+{
+	struct page *page;
+
+	page = pfn_to_page(entry >> PAGE_SHIFT);
+	kimage_free_pages(page);
+}
+
+static void kimage_free(struct kimage *image)
+{
+	kimage_entry_t *ptr, entry;
+	kimage_entry_t ind = 0;
+
+	if (!image)
+		return;
+	kimage_free_extra_pages(image);
+	for_each_kimage_entry(image, ptr, entry) {
+		if (entry & IND_INDIRECTION) {
+			/* Free the previous indirection page */
+			if (ind & IND_INDIRECTION) {
+				kimage_free_entry(ind);
+			}
+			/* Save this indirection page until we are
+			 * done with it.
+			 */
+			ind = entry;
+		}
+		else if (entry & IND_SOURCE) {
+			kimage_free_entry(entry);
+		}
+	}
+	/* Free the final indirection page */
+	if (ind & IND_INDIRECTION) {
+		kimage_free_entry(ind);
+	}
+
+	/* Handle any machine specific cleanup */
+	machine_kexec_cleanup(image);
+
+	/* Free the kexec control pages... */
+	kimage_free_page_list(&image->control_pages);
+	kfree(image);
+}
+
+static kimage_entry_t *kimage_dst_used(struct kimage *image, unsigned long page)
+{
+	kimage_entry_t *ptr, entry;
+	unsigned long destination = 0;
+
+	for_each_kimage_entry(image, ptr, entry) {
+		if (entry & IND_DESTINATION) {
+			destination = entry & PAGE_MASK;
+		}
+		else if (entry & IND_SOURCE) {
+			if (page == destination) {
+				return ptr;
+			}
+			destination += PAGE_SIZE;
+		}
+	}
+	return 0;
+}
+
+static struct page *kimage_alloc_page(struct kimage *image, unsigned int gfp_mask, unsigned long destination)
+{
+	/*
+	 * Here we implement safeguards to ensure that a source page
+	 * is not copied to its destination page before the data on
+	 * the destination page is no longer useful.
+	 *
+	 * To do this we maintain the invariant that a source page is
+	 * either its own destination page, or it is not a
+	 * destination page at all.
+	 *
+	 * That is slightly stronger than required, but the proof
+	 * that no problems will not occur is trivial, and the
+	 * implementation is simply to verify.
+	 *
+	 * When allocating all pages normally this algorithm will run
+	 * in O(N) time, but in the worst case it will run in O(N^2)
+	 * time.   If the runtime is a problem the data structures can
+	 * be fixed.
+	 */
+	struct page *page;
+	unsigned long addr;
+
+	/*
+	 * Walk through the list of destination pages, and see if I
+	 * have a match.
+	 */
+	list_for_each_entry(page, &image->dest_pages, lru) {
+		addr = page_to_pfn(page) << PAGE_SHIFT;
+		if (addr == destination) {
+			list_del(&page->lru);
+			return page;
+		}
+	}
+	page = NULL;
+	while (1) {
+		kimage_entry_t *old;
+
+		/* Allocate a page, if we run out of memory give up */
+		page = kimage_alloc_pages(gfp_mask, 0);
+		if (!page) {
+			return 0;
+		}
+		/* If the page cannot be used file it away */
+		if (page_to_pfn(page) > (KEXEC_SOURCE_MEMORY_LIMIT >> PAGE_SHIFT)) {
+			list_add(&page->lru, &image->unuseable_pages);
+			continue;
+		}
+		addr = page_to_pfn(page) << PAGE_SHIFT;
+
+		/* If it is the destination page we want use it */
+		if (addr == destination)
+			break;
+
+		/* If the page is not a destination page use it */
+		if (!kimage_is_destination_range(image, addr, addr + PAGE_SIZE))
+			break;
+
+		/*
+		 * I know that the page is someones destination page.
+		 * See if there is already a source page for this
+		 * destination page.  And if so swap the source pages.
+		 */
+		old = kimage_dst_used(image, addr);
+		if (old) {
+			/* If so move it */
+			unsigned long old_addr;
+			struct page *old_page;
+
+			old_addr = *old & PAGE_MASK;
+			old_page = pfn_to_page(old_addr >> PAGE_SHIFT);
+			copy_highpage(page, old_page);
+			*old = addr | (*old & ~PAGE_MASK);
+
+			/* The old page I have found cannot be a
+			 * destination page, so return it.
+			 */
+			addr = old_addr;
+			page = old_page;
+			break;
+		}
+		else {
+			/* Place the page on the destination list I
+			 * will use it later.
+			 */
+			list_add(&page->lru, &image->dest_pages);
+		}
+	}
+	return page;
+}
+
+static int kimage_load_normal_segment(struct kimage *image,
+	struct kexec_segment *segment)
+{
+	unsigned long maddr;
+	unsigned long ubytes, mbytes;
+	int result;
+	unsigned char *buf;
+
+	result = 0;
+	buf = segment->buf;
+	ubytes = segment->bufsz;
+	mbytes = segment->memsz;
+	maddr = segment->mem;
+
+	result = kimage_set_destination(image, maddr);
+	if (result < 0) {
+		goto out;
+	}
+	while(mbytes) {
+		struct page *page;
+		char *ptr;
+		size_t uchunk, mchunk;
+		page = kimage_alloc_page(image, GFP_HIGHUSER, maddr);
+		if (page == 0) {
+			result  = -ENOMEM;
+			goto out;
+		}
+		result = kimage_add_page(image, page_to_pfn(page) << PAGE_SHIFT);
+		if (result < 0) {
+			goto out;
+		}
+		ptr = kmap(page);
+		/* Start with a clear page */
+		memset(ptr, 0, PAGE_SIZE);
+		ptr += maddr & ~PAGE_MASK;
+		mchunk = PAGE_SIZE - (maddr & ~PAGE_MASK);
+		if (mchunk > mbytes) {
+			mchunk = mbytes;
+		}
+		uchunk = mchunk;
+		if (uchunk > ubytes) {
+			uchunk = ubytes;
+		}
+		result = copy_from_user(ptr, buf, uchunk);
+		kunmap(page);
+		if (result) {
+			result = (result < 0) ? result : -EIO;
+			goto out;
+		}
+		ubytes -= uchunk;
+		maddr  += mchunk;
+		buf    += mchunk;
+		mbytes -= mchunk;
+	}
+ out:
+	return result;
+}
+
+static int kimage_load_crash_segment(struct kimage *image,
+	struct kexec_segment *segment)
+{
+	/* For crash dumps kernels we simply copy the data from
+	 * user space to it's destination.
+	 * We do things a page at a time for the sake of kmap.
+	 */
+	unsigned long maddr;
+	unsigned long ubytes, mbytes;
+	int result;
+	unsigned char *buf;
+
+	result = 0;
+	buf = segment->buf;
+	ubytes = segment->bufsz;
+	mbytes = segment->memsz;
+	maddr = segment->mem;
+	while(mbytes) {
+		struct page *page;
+		char *ptr;
+		size_t uchunk, mchunk;
+		page = pfn_to_page(maddr >> PAGE_SHIFT);
+		if (page == 0) {
+			result  = -ENOMEM;
+			goto out;
+		}
+		ptr = kmap(page);
+		ptr += maddr & ~PAGE_MASK;
+		mchunk = PAGE_SIZE - (maddr & ~PAGE_MASK);
+		if (mchunk > mbytes) {
+			mchunk = mbytes;
+		}
+		uchunk = mchunk;
+		if (uchunk > ubytes) {
+			uchunk = ubytes;
+			/* Zero the trailing part of the page */
+			memset(ptr + uchunk, 0, mchunk - uchunk);
+		}
+		result = copy_from_user(ptr, buf, uchunk);
+		kunmap(page);
+		if (result) {
+			result = (result < 0) ? result : -EIO;
+			goto out;
+		}
+		ubytes -= uchunk;
+		maddr  += mchunk;
+		buf    += mchunk;
+		mbytes -= mchunk;
+	}
+ out:
+	return result;
+}
+
+static int kimage_load_segment(struct kimage *image,
+	struct kexec_segment *segment)
+{
+	int result = -ENOMEM;
+	switch(image->type) {
+	case KEXEC_TYPE_DEFAULT:
+		result = kimage_load_normal_segment(image, segment);
+		break;
+	case KEXEC_TYPE_CRASH:
+		result = kimage_load_crash_segment(image, segment);
+		break;
+	}
+	return result;
+}
+
+/*
+ * Exec Kernel system call: for obvious reasons only root may call it.
+ *
+ * This call breaks up into three pieces.
+ * - A generic part which loads the new kernel from the current
+ *   address space, and very carefully places the data in the
+ *   allocated pages.
+ *
+ * - A generic part that interacts with the kernel and tells all of
+ *   the devices to shut down.  Preventing on-going dmas, and placing
+ *   the devices in a consistent state so a later kernel can
+ *   reinitialize them.
+ *
+ * - A machine specific part that includes the syscall number
+ *   and the copies the image to it's final destination.  And
+ *   jumps into the image at entry.
+ *
+ * kexec does not sync, or unmount filesystems so if you need
+ * that to happen you need to do that yourself.
+ */
+struct kimage *kexec_image = NULL;
+static struct kimage *kexec_crash_image = NULL;
+/*
+ * A home grown binary mutex.
+ * Nothing can wait so this mutex is safe to use
+ * in interrupt context :)
+ */
+static int kexec_lock = 0;
+
+asmlinkage long sys_kexec_load(unsigned long entry,
+	unsigned long nr_segments, struct kexec_segment __user *segments,
+	unsigned long flags)
+{
+	struct kimage **dest_image, *image;
+	int locked;
+	int result;
+
+	/* We only trust the superuser with rebooting the system. */
+	if (!capable(CAP_SYS_BOOT))
+		return -EPERM;
+
+	/*
+	 * Verify we have a legal set of flags
+	 * This leaves us room for future extensions.
+	 */
+	if ((flags & KEXEC_FLAGS) != (flags & ~KEXEC_ARCH_MASK))
+		return -EINVAL;
+
+	/* Verify we are on the appropriate architecture */
+	if (((flags & KEXEC_ARCH_MASK) != KEXEC_ARCH) &&
+		((flags & KEXEC_ARCH_MASK) != KEXEC_ARCH_DEFAULT))
+	{
+		return -EINVAL;
+	}
+
+	/* Put an artificial cap on the number
+	 * of segments passed to kexec_load.
+	 */
+	if (nr_segments > KEXEC_SEGMENT_MAX)
+		return -EINVAL;
+
+	image = NULL;
+	result = 0;
+
+	/* Because we write directly to the reserved memory
+	 * region when loading crash kernels we need a mutex here to
+	 * prevent multiple crash  kernels from attempting to load
+	 * simultaneously, and to prevent a crash kernel from loading
+	 * over the top of a in use crash kernel.
+	 *
+	 * KISS: always take the mutex.
+	 */
+	locked = xchg(&kexec_lock, 1);
+	if (locked) {
+		return -EBUSY;
+	}
+	dest_image = &kexec_image;
+	if (flags & KEXEC_ON_CRASH) {
+		dest_image = &kexec_crash_image;
+	}
+	if (nr_segments > 0) {
+		unsigned long i;
+		/* Loading another kernel to reboot into */
+		if ((flags & KEXEC_ON_CRASH) == 0) {
+			result = kimage_normal_alloc(&image, entry, nr_segments, segments);
+		}
+		/* Loading another kernel to switch to if this one crashes */
+		else if (flags & KEXEC_ON_CRASH) {
+			/* Free any current crash dump kernel before
+			 * we corrupt it.
+			 */
+			kimage_free(xchg(&kexec_crash_image, NULL));
+			result = kimage_crash_alloc(&image, entry, nr_segments, segments);
+		}
+		if (result) {
+			goto out;
+		}
+		result = machine_kexec_prepare(image);
+		if (result) {
+			goto out;
+		}
+		for(i = 0; i < nr_segments; i++) {
+			result = kimage_load_segment(image, &image->segment[i]);
+			if (result) {
+				goto out;
+			}
+		}
+		result = kimage_terminate(image);
+		if (result) {
+			goto out;
+		}
+	}
+	/* Install the new kernel, and  Uninstall the old */
+	image = xchg(dest_image, image);
+
+ out:
+	xchg(&kexec_lock, 0); /* Release the mutex */
+	kimage_free(image);
+	return result;
+}
+
+#ifdef CONFIG_COMPAT
+asmlinkage long compat_sys_kexec_load(unsigned long entry,
+	unsigned long nr_segments, struct compat_kexec_segment __user *segments,
+	unsigned long flags)
+{
+	struct compat_kexec_segment in;
+	struct kexec_segment out, __user *ksegments;
+	unsigned long i, result;
+
+	/* Don't allow clients that don't understand the native
+	 * architecture to do anything.
+	 */
+	if ((flags & KEXEC_ARCH_MASK) == KEXEC_ARCH_DEFAULT) {
+		return -EINVAL;
+	}
+
+	if (nr_segments > KEXEC_SEGMENT_MAX) {
+		return -EINVAL;
+	}
+
+	ksegments = compat_alloc_user_space(nr_segments * sizeof(out));
+	for (i=0; i < nr_segments; i++) {
+		result = copy_from_user(&in, &segments[i], sizeof(in));
+		if (result) {
+			return -EFAULT;
+		}
+
+		out.buf   = compat_ptr(in.buf);
+		out.bufsz = in.bufsz;
+		out.mem   = in.mem;
+		out.memsz = in.memsz;
+
+		result = copy_to_user(&ksegments[i], &out, sizeof(out));
+		if (result) {
+			return -EFAULT;
+		}
+	}
+
+	return sys_kexec_load(entry, nr_segments, ksegments, flags);
+}
+#endif
+
+void crash_kexec(void)
+{
+	struct kimage *image;
+	int locked;
+
+
+	/* Take the kexec_lock here to prevent sys_kexec_load
+	 * running on one cpu from replacing the crash kernel
+	 * we are using after a panic on a different cpu.
+	 *
+	 * If the crash kernel was not located in a fixed area
+	 * of memory the xchg(&kexec_crash_image) would be
+	 * sufficient.  But since I reuse the memory...
+	 */
+	locked = xchg(&kexec_lock, 1);
+	if (!locked) {
+		image = xchg(&kexec_crash_image, NULL);
+		if (image) {
+			machine_crash_shutdown();
+			machine_kexec(image);
+		}
+		xchg(&kexec_lock, 0);
+	}
+}
diff --git a/kernel/panic.c b/kernel/panic.c
index 081f7465fc8..66f43d33cd8 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -18,6 +18,7 @@
 #include <linux/sysrq.h>
 #include <linux/interrupt.h>
 #include <linux/nmi.h>
+#include <linux/kexec.h>
 
 int panic_timeout;
 int panic_on_oops;
@@ -63,6 +64,13 @@ NORET_TYPE void panic(const char * fmt, ...)
         unsigned long caller = (unsigned long) __builtin_return_address(0);
 #endif
 
+	/*
+	 * It's possible to come here directly from a panic-assertion and not
+	 * have preempt disabled. Some functions called from here want
+	 * preempt to be disabled. No point enabling it later though...
+	 */
+	preempt_disable();
+
 	bust_spinlocks(1);
 	va_start(args, fmt);
 	vsnprintf(buf, sizeof(buf), fmt, args);
@@ -70,7 +78,19 @@ NORET_TYPE void panic(const char * fmt, ...)
 	printk(KERN_EMERG "Kernel panic - not syncing: %s\n",buf);
 	bust_spinlocks(0);
 
+	/*
+	 * If we have crashed and we have a crash kernel loaded let it handle
+	 * everything else.
+	 * Do we want to call this before we try to display a message?
+	 */
+	crash_kexec();
+
 #ifdef CONFIG_SMP
+	/*
+	 * Note smp_send_stop is the usual smp shutdown function, which
+	 * unfortunately means it may not be hardened to work in a panic
+	 * situation.
+	 */
 	smp_send_stop();
 #endif
 
@@ -79,8 +99,7 @@ NORET_TYPE void panic(const char * fmt, ...)
 	if (!panic_blink)
 		panic_blink = no_blink;
 
-	if (panic_timeout > 0)
-	{
+	if (panic_timeout > 0) {
 		/*
 	 	 * Delay timeout seconds before rebooting the machine. 
 		 * We can't use the "normal" timers since we just panicked..
diff --git a/kernel/sys.c b/kernel/sys.c
index dac10161ca2..9a24374c23b 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -16,6 +16,8 @@
 #include <linux/init.h>
 #include <linux/highuid.h>
 #include <linux/fs.h>
+#include <linux/kernel.h>
+#include <linux/kexec.h>
 #include <linux/workqueue.h>
 #include <linux/device.h>
 #include <linux/key.h>
@@ -439,6 +441,24 @@ asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user
 		machine_restart(buffer);
 		break;
 
+#ifdef CONFIG_KEXEC
+	case LINUX_REBOOT_CMD_KEXEC:
+	{
+		struct kimage *image;
+		image = xchg(&kexec_image, 0);
+		if (!image) {
+			unlock_kernel();
+			return -EINVAL;
+		}
+		notifier_call_chain(&reboot_notifier_list, SYS_RESTART, NULL);
+		system_state = SYSTEM_RESTART;
+		device_shutdown();
+		printk(KERN_EMERG "Starting new kernel\n");
+		machine_shutdown();
+		machine_kexec(image);
+		break;
+	}
+#endif
 #ifdef CONFIG_SOFTWARE_SUSPEND
 	case LINUX_REBOOT_CMD_SW_SUSPEND:
 		{
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 6f15bea7d1a..29196ce9b40 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -18,6 +18,8 @@ cond_syscall(sys_acct);
 cond_syscall(sys_lookup_dcookie);
 cond_syscall(sys_swapon);
 cond_syscall(sys_swapoff);
+cond_syscall(sys_kexec_load);
+cond_syscall(compat_sys_kexec_load);
 cond_syscall(sys_init_module);
 cond_syscall(sys_delete_module);
 cond_syscall(sys_socketpair);
-- 
cgit v1.2.3-70-g09d2


From 5033cba087f6ac773002123aafbea1aad4267682 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:56 -0700
Subject: [PATCH] kexec: x86 kexec core

This is the i386 implementation of kexec.

Signed-off-by: Eric Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig                  |  17 +++
 arch/i386/kernel/Makefile          |   1 +
 arch/i386/kernel/crash.c           |  42 +++++++
 arch/i386/kernel/machine_kexec.c   | 220 +++++++++++++++++++++++++++++++++++++
 arch/i386/kernel/relocate_kernel.S | 120 ++++++++++++++++++++
 arch/i386/kernel/syscall_table.S   |   2 +-
 include/asm-i386/kexec.h           |  28 +++++
 7 files changed, 429 insertions(+), 1 deletion(-)
 create mode 100644 arch/i386/kernel/crash.c
 create mode 100644 arch/i386/kernel/machine_kexec.c
 create mode 100644 arch/i386/kernel/relocate_kernel.S
 create mode 100644 include/asm-i386/kexec.h

(limited to 'include')

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index 31567f4d333..0f391cbf116 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -953,6 +953,23 @@ config PHYSICAL_START
 
 	  Don't change this unless you know what you are doing.
 
+config KEXEC
+	bool "kexec system call (EXPERIMENTAL)"
+	depends on EXPERIMENTAL
+	help
+	  kexec is a system call that implements the ability to shutdown your
+	  current kernel, and to start another kernel.  It is like a reboot
+	  but it is indepedent of the system firmware.   And like a reboot
+	  you can start any kernel with it, not just Linux.
+
+	  The name comes from the similiarity to the exec system call.
+
+	  It is an ongoing process to be certain the hardware in a machine
+	  is properly shutdown, so do not be surprised if this code does not
+	  initially work for you.  It may help to enable device hotplugging
+	  support.  As of this writing the exact hardware interface is
+	  strongly in flux, so no good recommendation can be made.
+
 endmenu
 
 
diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile
index 51ecd512603..4cc83b322b3 100644
--- a/arch/i386/kernel/Makefile
+++ b/arch/i386/kernel/Makefile
@@ -24,6 +24,7 @@ obj-$(CONFIG_X86_MPPARSE)	+= mpparse.o
 obj-$(CONFIG_X86_LOCAL_APIC)	+= apic.o nmi.o
 obj-$(CONFIG_X86_IO_APIC)	+= io_apic.o
 obj-$(CONFIG_X86_REBOOTFIXUPS)	+= reboot_fixups.o
+obj-$(CONFIG_KEXEC)		+= machine_kexec.o relocate_kernel.o crash.o
 obj-$(CONFIG_X86_NUMAQ)		+= numaq.o
 obj-$(CONFIG_X86_SUMMIT_NUMA)	+= summit.o
 obj-$(CONFIG_KPROBES)		+= kprobes.o
diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c
new file mode 100644
index 00000000000..fa27a6c2abb
--- /dev/null
+++ b/arch/i386/kernel/crash.c
@@ -0,0 +1,42 @@
+/*
+ * Architecture specific (i386) functions for kexec based crash dumps.
+ *
+ * Created by: Hariprasad Nellitheertha (hari@in.ibm.com)
+ *
+ * Copyright (C) IBM Corporation, 2004. All rights reserved.
+ *
+ */
+
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/smp.h>
+#include <linux/irq.h>
+#include <linux/reboot.h>
+#include <linux/kexec.h>
+#include <linux/irq.h>
+#include <linux/delay.h>
+#include <linux/elf.h>
+#include <linux/elfcore.h>
+
+#include <asm/processor.h>
+#include <asm/hardirq.h>
+#include <asm/nmi.h>
+#include <asm/hw_irq.h>
+
+#define MAX_NOTE_BYTES 1024
+typedef u32 note_buf_t[MAX_NOTE_BYTES/4];
+
+note_buf_t crash_notes[NR_CPUS];
+
+void machine_crash_shutdown(void)
+{
+	/* This function is only called after the system
+	 * has paniced or is otherwise in a critical state.
+	 * The minimum amount of code to allow a kexec'd kernel
+	 * to run successfully needs to happen here.
+	 *
+	 * In practice this means shooting down the other cpus in
+	 * an SMP system.
+	 */
+}
diff --git a/arch/i386/kernel/machine_kexec.c b/arch/i386/kernel/machine_kexec.c
new file mode 100644
index 00000000000..671880415d1
--- /dev/null
+++ b/arch/i386/kernel/machine_kexec.c
@@ -0,0 +1,220 @@
+/*
+ * machine_kexec.c - handle transition of Linux booting another kernel
+ * Copyright (C) 2002-2005 Eric Biederman  <ebiederm@xmission.com>
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2.  See the file COPYING for more details.
+ */
+
+#include <linux/mm.h>
+#include <linux/kexec.h>
+#include <linux/delay.h>
+#include <asm/pgtable.h>
+#include <asm/pgalloc.h>
+#include <asm/tlbflush.h>
+#include <asm/mmu_context.h>
+#include <asm/io.h>
+#include <asm/apic.h>
+#include <asm/cpufeature.h>
+
+static inline unsigned long read_cr3(void)
+{
+	unsigned long cr3;
+	asm volatile("movl %%cr3,%0": "=r"(cr3));
+	return cr3;
+}
+
+#define PAGE_ALIGNED __attribute__ ((__aligned__(PAGE_SIZE)))
+
+#define L0_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
+#define L1_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
+#define L2_ATTR (_PAGE_PRESENT)
+
+#define LEVEL0_SIZE (1UL << 12UL)
+
+#ifndef CONFIG_X86_PAE
+#define LEVEL1_SIZE (1UL << 22UL)
+static u32 pgtable_level1[1024] PAGE_ALIGNED;
+
+static void identity_map_page(unsigned long address)
+{
+	unsigned long level1_index, level2_index;
+	u32 *pgtable_level2;
+
+	/* Find the current page table */
+	pgtable_level2 = __va(read_cr3());
+
+	/* Find the indexes of the physical address to identity map */
+	level1_index = (address % LEVEL1_SIZE)/LEVEL0_SIZE;
+	level2_index = address / LEVEL1_SIZE;
+
+	/* Identity map the page table entry */
+	pgtable_level1[level1_index] = address | L0_ATTR;
+	pgtable_level2[level2_index] = __pa(pgtable_level1) | L1_ATTR;
+
+	/* Flush the tlb so the new mapping takes effect.
+	 * Global tlb entries are not flushed but that is not an issue.
+	 */
+	load_cr3(pgtable_level2);
+}
+
+#else
+#define LEVEL1_SIZE (1UL << 21UL)
+#define LEVEL2_SIZE (1UL << 30UL)
+static u64 pgtable_level1[512] PAGE_ALIGNED;
+static u64 pgtable_level2[512] PAGE_ALIGNED;
+
+static void identity_map_page(unsigned long address)
+{
+	unsigned long level1_index, level2_index, level3_index;
+	u64 *pgtable_level3;
+
+	/* Find the current page table */
+	pgtable_level3 = __va(read_cr3());
+
+	/* Find the indexes of the physical address to identity map */
+	level1_index = (address % LEVEL1_SIZE)/LEVEL0_SIZE;
+	level2_index = (address % LEVEL2_SIZE)/LEVEL1_SIZE;
+	level3_index = address / LEVEL2_SIZE;
+
+	/* Identity map the page table entry */
+	pgtable_level1[level1_index] = address | L0_ATTR;
+	pgtable_level2[level2_index] = __pa(pgtable_level1) | L1_ATTR;
+	set_64bit(&pgtable_level3[level3_index], __pa(pgtable_level2) | L2_ATTR);
+
+	/* Flush the tlb so the new mapping takes effect.
+	 * Global tlb entries are not flushed but that is not an issue.
+	 */
+	load_cr3(pgtable_level3);
+}
+#endif
+
+
+static void set_idt(void *newidt, __u16 limit)
+{
+	unsigned char curidt[6];
+
+	/* ia32 supports unaliged loads & stores */
+	(*(__u16 *)(curidt)) = limit;
+	(*(__u32 *)(curidt +2)) = (unsigned long)(newidt);
+
+	__asm__ __volatile__ (
+		"lidt %0\n"
+		: "=m" (curidt)
+		);
+};
+
+
+static void set_gdt(void *newgdt, __u16 limit)
+{
+	unsigned char curgdt[6];
+
+	/* ia32 supports unaligned loads & stores */
+	(*(__u16 *)(curgdt)) = limit;
+	(*(__u32 *)(curgdt +2)) = (unsigned long)(newgdt);
+
+	__asm__ __volatile__ (
+		"lgdt %0\n"
+		: "=m" (curgdt)
+		);
+};
+
+static void load_segments(void)
+{
+#define __STR(X) #X
+#define STR(X) __STR(X)
+
+	__asm__ __volatile__ (
+		"\tljmp $"STR(__KERNEL_CS)",$1f\n"
+		"\t1:\n"
+		"\tmovl $"STR(__KERNEL_DS)",%eax\n"
+		"\tmovl %eax,%ds\n"
+		"\tmovl %eax,%es\n"
+		"\tmovl %eax,%fs\n"
+		"\tmovl %eax,%gs\n"
+		"\tmovl %eax,%ss\n"
+		);
+#undef STR
+#undef __STR
+}
+
+typedef asmlinkage NORET_TYPE void (*relocate_new_kernel_t)(
+	unsigned long indirection_page, unsigned long reboot_code_buffer,
+	unsigned long start_address, unsigned int has_pae) ATTRIB_NORET;
+
+const extern unsigned char relocate_new_kernel[];
+extern void relocate_new_kernel_end(void);
+const extern unsigned int relocate_new_kernel_size;
+
+/*
+ * A architecture hook called to validate the
+ * proposed image and prepare the control pages
+ * as needed.  The pages for KEXEC_CONTROL_CODE_SIZE
+ * have been allocated, but the segments have yet
+ * been copied into the kernel.
+ *
+ * Do what every setup is needed on image and the
+ * reboot code buffer to allow us to avoid allocations
+ * later.
+ *
+ * Currently nothing.
+ */
+int machine_kexec_prepare(struct kimage *image)
+{
+	return 0;
+}
+
+/*
+ * Undo anything leftover by machine_kexec_prepare
+ * when an image is freed.
+ */
+void machine_kexec_cleanup(struct kimage *image)
+{
+}
+
+/*
+ * Do not allocate memory (or fail in any way) in machine_kexec().
+ * We are past the point of no return, committed to rebooting now.
+ */
+NORET_TYPE void machine_kexec(struct kimage *image)
+{
+	unsigned long page_list;
+	unsigned long reboot_code_buffer;
+	relocate_new_kernel_t rnk;
+
+	/* Interrupts aren't acceptable while we reboot */
+	local_irq_disable();
+
+	/* Compute some offsets */
+	reboot_code_buffer = page_to_pfn(image->control_code_page) << PAGE_SHIFT;
+	page_list = image->head;
+
+	/* Set up an identity mapping for the reboot_code_buffer */
+	identity_map_page(reboot_code_buffer);
+
+	/* copy it out */
+	memcpy((void *)reboot_code_buffer, relocate_new_kernel, relocate_new_kernel_size);
+
+	/* The segment registers are funny things, they are
+	 * automatically loaded from a table, in memory wherever you
+	 * set them to a specific selector, but this table is never
+	 * accessed again you set the segment to a different selector.
+	 *
+	 * The more common model is are caches where the behide
+	 * the scenes work is done, but is also dropped at arbitrary
+	 * times.
+	 *
+	 * I take advantage of this here by force loading the
+	 * segments, before I zap the gdt with an invalid value.
+	 */
+	load_segments();
+	/* The gdt & idt are now invalid.
+	 * If you want to load them you must set up your own idt & gdt.
+	 */
+	set_gdt(phys_to_virt(0),0);
+	set_idt(phys_to_virt(0),0);
+
+	/* now call it */
+	rnk = (relocate_new_kernel_t) reboot_code_buffer;
+	(*rnk)(page_list, reboot_code_buffer, image->start, cpu_has_pae);
+}
diff --git a/arch/i386/kernel/relocate_kernel.S b/arch/i386/kernel/relocate_kernel.S
new file mode 100644
index 00000000000..d312616effa
--- /dev/null
+++ b/arch/i386/kernel/relocate_kernel.S
@@ -0,0 +1,120 @@
+/*
+ * relocate_kernel.S - put the kernel image in place to boot
+ * Copyright (C) 2002-2004 Eric Biederman  <ebiederm@xmission.com>
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2.  See the file COPYING for more details.
+ */
+
+#include <linux/linkage.h>
+
+	/*
+	 * Must be relocatable PIC code callable as a C function, that once
+	 * it starts can not use the previous processes stack.
+	 */
+	.globl relocate_new_kernel
+relocate_new_kernel:
+	/* read the arguments and say goodbye to the stack */
+	movl  4(%esp), %ebx /* page_list */
+	movl  8(%esp), %ebp /* reboot_code_buffer */
+	movl  12(%esp), %edx /* start address */
+	movl  16(%esp), %ecx /* cpu_has_pae */
+
+	/* zero out flags, and disable interrupts */
+	pushl $0
+	popfl
+
+	/* set a new stack at the bottom of our page... */
+	lea   4096(%ebp), %esp
+
+	/* store the parameters back on the stack */
+	pushl   %edx /* store the start address */
+
+	/* Set cr0 to a known state:
+	 * 31 0 == Paging disabled
+	 * 18 0 == Alignment check disabled
+	 * 16 0 == Write protect disabled
+	 * 3  0 == No task switch
+	 * 2  0 == Don't do FP software emulation.
+	 * 0  1 == Proctected mode enabled
+	 */
+	movl	%cr0, %eax
+	andl	$~((1<<31)|(1<<18)|(1<<16)|(1<<3)|(1<<2)), %eax
+	orl	$(1<<0), %eax
+	movl	%eax, %cr0
+
+	/* clear cr4 if applicable */
+	testl	%ecx, %ecx
+	jz	1f
+	/* Set cr4 to a known state:
+	 * Setting everything to zero seems safe.
+	 */
+	movl	%cr4, %eax
+	andl	$0, %eax
+	movl	%eax, %cr4
+
+	jmp 1f
+1:
+
+	/* Flush the TLB (needed?) */
+	xorl	%eax, %eax
+	movl	%eax, %cr3
+
+	/* Do the copies */
+	movl	%ebx, %ecx
+	jmp	1f
+
+0:	/* top, read another word from the indirection page */
+	movl	(%ebx), %ecx
+	addl	$4, %ebx
+1:
+	testl	$0x1,   %ecx  /* is it a destination page */
+	jz	2f
+	movl	%ecx,	%edi
+	andl	$0xfffff000, %edi
+	jmp     0b
+2:
+	testl	$0x2,	%ecx  /* is it an indirection page */
+	jz	2f
+	movl	%ecx,	%ebx
+	andl	$0xfffff000, %ebx
+	jmp     0b
+2:
+	testl   $0x4,   %ecx /* is it the done indicator */
+	jz      2f
+	jmp     3f
+2:
+	testl   $0x8,   %ecx /* is it the source indicator */
+	jz      0b	     /* Ignore it otherwise */
+	movl    %ecx,   %esi /* For every source page do a copy */
+	andl    $0xfffff000, %esi
+
+	movl    $1024, %ecx
+	rep ; movsl
+	jmp     0b
+
+3:
+
+	/* To be certain of avoiding problems with self-modifying code
+	 * I need to execute a serializing instruction here.
+	 * So I flush the TLB, it's handy, and not processor dependent.
+	 */
+	xorl	%eax, %eax
+	movl	%eax, %cr3
+
+	/* set all of the registers to known values */
+	/* leave %esp alone */
+
+	xorl	%eax, %eax
+	xorl	%ebx, %ebx
+	xorl    %ecx, %ecx
+	xorl    %edx, %edx
+	xorl    %esi, %esi
+	xorl    %edi, %edi
+	xorl    %ebp, %ebp
+	ret
+relocate_new_kernel_end:
+
+	.globl relocate_new_kernel_size
+relocate_new_kernel_size:
+	.long relocate_new_kernel_end - relocate_new_kernel
diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S
index d408afaf649..442a6e937b1 100644
--- a/arch/i386/kernel/syscall_table.S
+++ b/arch/i386/kernel/syscall_table.S
@@ -283,7 +283,7 @@ ENTRY(sys_call_table)
 	.long sys_mq_timedreceive	/* 280 */
 	.long sys_mq_notify
 	.long sys_mq_getsetattr
-	.long sys_ni_syscall		/* reserved for kexec */
+	.long sys_kexec_load
 	.long sys_waitid
 	.long sys_ni_syscall		/* 285 */ /* available */
 	.long sys_add_key
diff --git a/include/asm-i386/kexec.h b/include/asm-i386/kexec.h
new file mode 100644
index 00000000000..a1599b55d62
--- /dev/null
+++ b/include/asm-i386/kexec.h
@@ -0,0 +1,28 @@
+#ifndef _I386_KEXEC_H
+#define _I386_KEXEC_H
+
+#include <asm/fixmap.h>
+
+/*
+ * KEXEC_SOURCE_MEMORY_LIMIT maximum page get_free_page can return.
+ * I.e. Maximum page that is mapped directly into kernel memory,
+ * and kmap is not required.
+ *
+ * Someone correct me if FIXADDR_START - PAGEOFFSET is not the correct
+ * calculation for the amount of memory directly mappable into the
+ * kernel memory space.
+ */
+
+/* Maximum physical address we can use pages from */
+#define KEXEC_SOURCE_MEMORY_LIMIT (-1UL)
+/* Maximum address we can reach in physical address mode */
+#define KEXEC_DESTINATION_MEMORY_LIMIT (-1UL)
+/* Maximum address we can use for the control code buffer */
+#define KEXEC_CONTROL_MEMORY_LIMIT TASK_SIZE
+
+#define KEXEC_CONTROL_CODE_SIZE	4096
+
+/* The native architecture */
+#define KEXEC_ARCH KEXEC_ARCH_386
+
+#endif /* _I386_KEXEC_H */
-- 
cgit v1.2.3-70-g09d2


From 5234f5eb04abbbfa306ccfbc2ccbb6e73f515b15 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:58:02 -0700
Subject: [PATCH] kexec: x86_64 kexec implementation

This is the x86_64 implementation of machine kexec.  32bit compatibility
support has been implemented, and machine_kexec has been enhanced to not care
about the changing internal kernel paget table structures.

From: Alexander Nyberg <alexn@dsv.su.se>

      build fix

Signed-off-by: Eric Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/Kconfig                  |  17 +++
 arch/x86_64/ia32/ia32entry.S         |   2 +-
 arch/x86_64/kernel/Makefile          |   1 +
 arch/x86_64/kernel/crash.c           |  38 ++++++
 arch/x86_64/kernel/machine_kexec.c   | 245 +++++++++++++++++++++++++++++++++++
 arch/x86_64/kernel/relocate_kernel.S | 143 ++++++++++++++++++++
 include/asm-x86_64/kexec.h           |  28 ++++
 include/asm-x86_64/unistd.h          |   2 +-
 8 files changed, 474 insertions(+), 2 deletions(-)
 create mode 100644 arch/x86_64/kernel/crash.c
 create mode 100644 arch/x86_64/kernel/machine_kexec.c
 create mode 100644 arch/x86_64/kernel/relocate_kernel.S
 create mode 100644 include/asm-x86_64/kexec.h

(limited to 'include')

diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 56c02cf6397..d09437b5c48 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -380,6 +380,23 @@ config PHYSICAL_START
 
 	  Don't change this unless you know what you are doing.
 
+config KEXEC
+	bool "kexec system call (EXPERIMENTAL)"
+	depends on EXPERIMENTAL
+	help
+	  kexec is a system call that implements the ability to shutdown your
+	  current kernel, and to start another kernel.  It is like a reboot
+	  but it is indepedent of the system firmware.   And like a reboot
+	  you can start any kernel with it, not just Linux.
+
+	  The name comes from the similiarity to the exec system call.
+
+	  It is an ongoing process to be certain the hardware in a machine
+	  is properly shutdown, so do not be surprised if this code does not
+	  initially work for you.  It may help to enable device hotplugging
+	  support.  As of this writing the exact hardware interface is
+	  strongly in flux, so no good recommendation can be made.
+
 config SECCOMP
 	bool "Enable seccomp to safely compute untrusted bytecode"
 	depends on PROC_FS
diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S
index f3ca0db85b5..cc935427d53 100644
--- a/arch/x86_64/ia32/ia32entry.S
+++ b/arch/x86_64/ia32/ia32entry.S
@@ -589,7 +589,7 @@ ia32_sys_call_table:
 	.quad compat_sys_mq_timedreceive	/* 280 */
 	.quad compat_sys_mq_notify
 	.quad compat_sys_mq_getsetattr
-	.quad quiet_ni_syscall		/* reserved for kexec */
+	.quad compat_sys_kexec_load	/* reserved for kexec */
 	.quad compat_sys_waitid
 	.quad quiet_ni_syscall		/* sys_altroot */
 	.quad sys_add_key
diff --git a/arch/x86_64/kernel/Makefile b/arch/x86_64/kernel/Makefile
index 5ca4a4598fd..48f9e2c19cd 100644
--- a/arch/x86_64/kernel/Makefile
+++ b/arch/x86_64/kernel/Makefile
@@ -20,6 +20,7 @@ obj-$(CONFIG_SMP)		+= smp.o smpboot.o trampoline.o
 obj-$(CONFIG_X86_LOCAL_APIC)	+= apic.o  nmi.o
 obj-$(CONFIG_X86_IO_APIC)	+= io_apic.o mpparse.o \
 		genapic.o genapic_cluster.o genapic_flat.o
+obj-$(CONFIG_KEXEC)		+= machine_kexec.o relocate_kernel.o crash.o
 obj-$(CONFIG_PM)		+= suspend.o
 obj-$(CONFIG_SOFTWARE_SUSPEND)	+= suspend_asm.o
 obj-$(CONFIG_CPU_FREQ)		+= cpufreq/
diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c
new file mode 100644
index 00000000000..7caf8a49d0c
--- /dev/null
+++ b/arch/x86_64/kernel/crash.c
@@ -0,0 +1,38 @@
+/*
+ * Architecture specific (x86_64) functions for kexec based crash dumps.
+ *
+ * Created by: Hariprasad Nellitheertha (hari@in.ibm.com)
+ *
+ * Copyright (C) IBM Corporation, 2004. All rights reserved.
+ *
+ */
+
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/smp.h>
+#include <linux/irq.h>
+#include <linux/reboot.h>
+#include <linux/kexec.h>
+
+#include <asm/processor.h>
+#include <asm/hardirq.h>
+#include <asm/nmi.h>
+#include <asm/hw_irq.h>
+
+#define MAX_NOTE_BYTES 1024
+typedef u32 note_buf_t[MAX_NOTE_BYTES/4];
+
+note_buf_t crash_notes[NR_CPUS];
+
+void machine_crash_shutdown(void)
+{
+	/* This function is only called after the system
+	 * has paniced or is otherwise in a critical state.
+	 * The minimum amount of code to allow a kexec'd kernel
+	 * to run successfully needs to happen here.
+	 *
+	 * In practice this means shooting down the other cpus in
+	 * an SMP system.
+	 */
+}
diff --git a/arch/x86_64/kernel/machine_kexec.c b/arch/x86_64/kernel/machine_kexec.c
new file mode 100644
index 00000000000..200b5993f8d
--- /dev/null
+++ b/arch/x86_64/kernel/machine_kexec.c
@@ -0,0 +1,245 @@
+/*
+ * machine_kexec.c - handle transition of Linux booting another kernel
+ * Copyright (C) 2002-2005 Eric Biederman  <ebiederm@xmission.com>
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2.  See the file COPYING for more details.
+ */
+
+#include <linux/mm.h>
+#include <linux/kexec.h>
+#include <linux/delay.h>
+#include <linux/string.h>
+#include <linux/reboot.h>
+#include <asm/pda.h>
+#include <asm/pgtable.h>
+#include <asm/pgalloc.h>
+#include <asm/tlbflush.h>
+#include <asm/mmu_context.h>
+#include <asm/io.h>
+#include <asm/apic.h>
+#include <asm/cpufeature.h>
+#include <asm/hw_irq.h>
+
+#define LEVEL0_SIZE (1UL << 12UL)
+#define LEVEL1_SIZE (1UL << 21UL)
+#define LEVEL2_SIZE (1UL << 30UL)
+#define LEVEL3_SIZE (1UL << 39UL)
+#define LEVEL4_SIZE (1UL << 48UL)
+
+#define L0_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
+#define L1_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_PSE)
+#define L2_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
+#define L3_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
+
+static void init_level2_page(
+	u64 *level2p, unsigned long addr)
+{
+	unsigned long end_addr;
+	addr &= PAGE_MASK;
+	end_addr = addr + LEVEL2_SIZE;
+	while(addr < end_addr) {
+		*(level2p++) = addr | L1_ATTR;
+		addr += LEVEL1_SIZE;
+	}
+}
+
+static int init_level3_page(struct kimage *image,
+	u64 *level3p, unsigned long addr, unsigned long last_addr)
+{
+	unsigned long end_addr;
+	int result;
+	result = 0;
+	addr &= PAGE_MASK;
+	end_addr = addr + LEVEL3_SIZE;
+	while((addr < last_addr) && (addr < end_addr)) {
+		struct page *page;
+		u64 *level2p;
+		page = kimage_alloc_control_pages(image, 0);
+		if (!page) {
+			result = -ENOMEM;
+			goto out;
+		}
+		level2p = (u64 *)page_address(page);
+		init_level2_page(level2p, addr);
+		*(level3p++) = __pa(level2p) | L2_ATTR;
+		addr += LEVEL2_SIZE;
+	}
+	/* clear the unused entries */
+	while(addr < end_addr) {
+		*(level3p++) = 0;
+		addr += LEVEL2_SIZE;
+	}
+out:
+	return result;
+}
+
+
+static int init_level4_page(struct kimage *image,
+	u64 *level4p, unsigned long addr, unsigned long last_addr)
+{
+	unsigned long end_addr;
+	int result;
+	result = 0;
+	addr &= PAGE_MASK;
+	end_addr = addr + LEVEL4_SIZE;
+	while((addr < last_addr) && (addr < end_addr)) {
+		struct page *page;
+		u64 *level3p;
+		page = kimage_alloc_control_pages(image, 0);
+		if (!page) {
+			result = -ENOMEM;
+			goto out;
+		}
+		level3p = (u64 *)page_address(page);
+		result = init_level3_page(image, level3p, addr, last_addr);
+		if (result) {
+			goto out;
+		}
+		*(level4p++) = __pa(level3p) | L3_ATTR;
+		addr += LEVEL3_SIZE;
+	}
+	/* clear the unused entries */
+	while(addr < end_addr) {
+		*(level4p++) = 0;
+		addr += LEVEL3_SIZE;
+	}
+ out:
+	return result;
+}
+
+
+static int init_pgtable(struct kimage *image, unsigned long start_pgtable)
+{
+	u64 *level4p;
+	level4p = (u64 *)__va(start_pgtable);
+	return init_level4_page(image, level4p, 0, end_pfn << PAGE_SHIFT);
+}
+
+static void set_idt(void *newidt, u16 limit)
+{
+	unsigned char curidt[10];
+
+	/* x86-64 supports unaliged loads & stores */
+	(*(u16 *)(curidt)) = limit;
+	(*(u64 *)(curidt +2)) = (unsigned long)(newidt);
+
+	__asm__ __volatile__ (
+		"lidt %0\n"
+		: "=m" (curidt)
+		);
+};
+
+
+static void set_gdt(void *newgdt, u16 limit)
+{
+	unsigned char curgdt[10];
+
+	/* x86-64 supports unaligned loads & stores */
+	(*(u16 *)(curgdt)) = limit;
+	(*(u64 *)(curgdt +2)) = (unsigned long)(newgdt);
+
+	__asm__ __volatile__ (
+		"lgdt %0\n"
+		: "=m" (curgdt)
+		);
+};
+
+static void load_segments(void)
+{
+	__asm__ __volatile__ (
+		"\tmovl $"STR(__KERNEL_DS)",%eax\n"
+		"\tmovl %eax,%ds\n"
+		"\tmovl %eax,%es\n"
+		"\tmovl %eax,%ss\n"
+		"\tmovl %eax,%fs\n"
+		"\tmovl %eax,%gs\n"
+		);
+#undef STR
+#undef __STR
+}
+
+typedef NORET_TYPE void (*relocate_new_kernel_t)(
+	unsigned long indirection_page, unsigned long control_code_buffer,
+	unsigned long start_address, unsigned long pgtable) ATTRIB_NORET;
+
+const extern unsigned char relocate_new_kernel[];
+const extern unsigned long relocate_new_kernel_size;
+
+int machine_kexec_prepare(struct kimage *image)
+{
+	unsigned long start_pgtable, control_code_buffer;
+	int result;
+
+	/* Calculate the offsets */
+	start_pgtable       = page_to_pfn(image->control_code_page) << PAGE_SHIFT;
+	control_code_buffer = start_pgtable + 4096UL;
+
+	/* Setup the identity mapped 64bit page table */
+	result = init_pgtable(image, start_pgtable);
+	if (result) {
+		return result;
+	}
+
+	/* Place the code in the reboot code buffer */
+	memcpy(__va(control_code_buffer), relocate_new_kernel, relocate_new_kernel_size);
+
+	return 0;
+}
+
+void machine_kexec_cleanup(struct kimage *image)
+{
+	return;
+}
+
+/*
+ * Do not allocate memory (or fail in any way) in machine_kexec().
+ * We are past the point of no return, committed to rebooting now.
+ */
+NORET_TYPE void machine_kexec(struct kimage *image)
+{
+	unsigned long page_list;
+	unsigned long control_code_buffer;
+	unsigned long start_pgtable;
+	relocate_new_kernel_t rnk;
+
+	/* Interrupts aren't acceptable while we reboot */
+	local_irq_disable();
+
+	/* Calculate the offsets */
+	page_list           = image->head;
+	start_pgtable       = page_to_pfn(image->control_code_page) << PAGE_SHIFT;
+	control_code_buffer = start_pgtable + 4096UL;
+
+	/* Set the low half of the page table to my identity mapped
+	 * page table for kexec.  Leave the high half pointing at the
+	 * kernel pages.   Don't bother to flush the global pages
+	 * as that will happen when I fully switch to my identity mapped
+	 * page table anyway.
+	 */
+	memcpy(__va(read_cr3()), __va(start_pgtable), PAGE_SIZE/2);
+	__flush_tlb();
+
+
+	/* The segment registers are funny things, they are
+	 * automatically loaded from a table, in memory wherever you
+	 * set them to a specific selector, but this table is never
+	 * accessed again unless you set the segment to a different selector.
+	 *
+	 * The more common model are caches where the behide
+	 * the scenes work is done, but is also dropped at arbitrary
+	 * times.
+	 *
+	 * I take advantage of this here by force loading the
+	 * segments, before I zap the gdt with an invalid value.
+	 */
+	load_segments();
+	/* The gdt & idt are now invalid.
+	 * If you want to load them you must set up your own idt & gdt.
+	 */
+	set_gdt(phys_to_virt(0),0);
+	set_idt(phys_to_virt(0),0);
+	/* now call it */
+	rnk = (relocate_new_kernel_t) control_code_buffer;
+	(*rnk)(page_list, control_code_buffer, image->start, start_pgtable);
+}
diff --git a/arch/x86_64/kernel/relocate_kernel.S b/arch/x86_64/kernel/relocate_kernel.S
new file mode 100644
index 00000000000..d24fa9b72a2
--- /dev/null
+++ b/arch/x86_64/kernel/relocate_kernel.S
@@ -0,0 +1,143 @@
+/*
+ * relocate_kernel.S - put the kernel image in place to boot
+ * Copyright (C) 2002-2005 Eric Biederman  <ebiederm@xmission.com>
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2.  See the file COPYING for more details.
+ */
+
+#include <linux/linkage.h>
+
+	/*
+	 * Must be relocatable PIC code callable as a C function, that once
+	 * it starts can not use the previous processes stack.
+	 */
+	.globl relocate_new_kernel
+	.code64
+relocate_new_kernel:
+	/* %rdi page_list
+	 * %rsi reboot_code_buffer
+	 * %rdx start address
+	 * %rcx page_table
+	 * %r8  arg5
+	 * %r9  arg6
+	 */
+
+	/* zero out flags, and disable interrupts */
+	pushq $0
+	popfq
+
+	/* set a new stack at the bottom of our page... */
+	lea   4096(%rsi), %rsp
+
+	/* store the parameters back on the stack */
+	pushq	%rdx /* store the start address */
+
+	/* Set cr0 to a known state:
+	 * 31 1 == Paging enabled
+	 * 18 0 == Alignment check disabled
+	 * 16 0 == Write protect disabled
+	 * 3  0 == No task switch
+	 * 2  0 == Don't do FP software emulation.
+	 * 0  1 == Proctected mode enabled
+	 */
+	movq	%cr0, %rax
+	andq	$~((1<<18)|(1<<16)|(1<<3)|(1<<2)), %rax
+	orl	$((1<<31)|(1<<0)), %eax
+	movq	%rax, %cr0
+
+	/* Set cr4 to a known state:
+	 * 10 0 == xmm exceptions disabled
+	 * 9  0 == xmm registers instructions disabled
+	 * 8  0 == performance monitoring counter disabled
+	 * 7  0 == page global disabled
+	 * 6  0 == machine check exceptions disabled
+	 * 5  1 == physical address extension enabled
+	 * 4  0 == page size extensions	disabled
+	 * 3  0 == Debug extensions disabled
+	 * 2  0 == Time stamp disable (disabled)
+	 * 1  0 == Protected mode virtual interrupts disabled
+	 * 0  0 == VME disabled
+	 */
+
+	movq	$((1<<5)), %rax
+	movq	%rax, %cr4
+
+	jmp 1f
+1:
+
+	/* Switch to the identity mapped page tables,
+	 * and flush the TLB.
+	*/
+	movq	%rcx, %cr3
+
+	/* Do the copies */
+	movq	%rdi, %rcx 	/* Put the page_list in %rcx */
+	xorq	%rdi, %rdi
+	xorq	%rsi, %rsi
+	jmp	1f
+
+0:	/* top, read another word for the indirection page */
+
+	movq	(%rbx), %rcx
+	addq	$8,	%rbx
+1:
+	testq	$0x1,	%rcx  /* is it a destination page? */
+	jz	2f
+	movq	%rcx,	%rdi
+	andq	$0xfffffffffffff000, %rdi
+	jmp	0b
+2:
+	testq	$0x2,	%rcx  /* is it an indirection page? */
+	jz	2f
+	movq	%rcx,   %rbx
+	andq	$0xfffffffffffff000, %rbx
+	jmp	0b
+2:
+	testq	$0x4,	%rcx  /* is it the done indicator? */
+	jz	2f
+	jmp	3f
+2:
+	testq	$0x8,	%rcx  /* is it the source indicator? */
+	jz	0b	      /* Ignore it otherwise */
+	movq	%rcx,   %rsi  /* For ever source page do a copy */
+	andq	$0xfffffffffffff000, %rsi
+
+	movq	$512,   %rcx
+	rep ; movsq
+	jmp	0b
+3:
+
+	/* To be certain of avoiding problems with self-modifying code
+	 * I need to execute a serializing instruction here.
+	 * So I flush the TLB by reloading %cr3 here, it's handy,
+	 * and not processor dependent.
+	 */
+	movq	%cr3, %rax
+	movq	%rax, %cr3
+
+	/* set all of the registers to known values */
+	/* leave %rsp alone */
+
+	xorq	%rax, %rax
+	xorq	%rbx, %rbx
+	xorq    %rcx, %rcx
+	xorq    %rdx, %rdx
+	xorq    %rsi, %rsi
+	xorq    %rdi, %rdi
+	xorq    %rbp, %rbp
+	xorq	%r8,  %r8
+	xorq	%r9,  %r9
+	xorq	%r10, %r9
+	xorq	%r11, %r11
+	xorq	%r12, %r12
+	xorq	%r13, %r13
+	xorq	%r14, %r14
+	xorq	%r15, %r15
+
+	ret
+relocate_new_kernel_end:
+
+	.globl relocate_new_kernel_size
+relocate_new_kernel_size:
+	.quad relocate_new_kernel_end - relocate_new_kernel
diff --git a/include/asm-x86_64/kexec.h b/include/asm-x86_64/kexec.h
new file mode 100644
index 00000000000..dc33646dc7d
--- /dev/null
+++ b/include/asm-x86_64/kexec.h
@@ -0,0 +1,28 @@
+#ifndef _X86_64_KEXEC_H
+#define _X86_64_KEXEC_H
+
+#include <asm/page.h>
+#include <asm/proto.h>
+
+/*
+ * KEXEC_SOURCE_MEMORY_LIMIT maximum page get_free_page can return.
+ * I.e. Maximum page that is mapped directly into kernel memory,
+ * and kmap is not required.
+ *
+ * So far x86_64 is limited to 40 physical address bits.
+ */
+
+/* Maximum physical address we can use pages from */
+#define KEXEC_SOURCE_MEMORY_LIMIT      (0xFFFFFFFFFFUL)
+/* Maximum address we can reach in physical address mode */
+#define KEXEC_DESTINATION_MEMORY_LIMIT (0xFFFFFFFFFFUL)
+/* Maximum address we can use for the control pages */
+#define KEXEC_CONTROL_MEMORY_LIMIT     (0xFFFFFFFFFFUL)
+
+/* Allocate one page for the pdp and the second for the code */
+#define KEXEC_CONTROL_CODE_SIZE  (4096UL + 4096UL)
+
+/* The native architecture */
+#define KEXEC_ARCH KEXEC_ARCH_X86_64
+
+#endif /* _X86_64_KEXEC_H */
diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h
index 3c9af6fd433..d767adcbf0f 100644
--- a/include/asm-x86_64/unistd.h
+++ b/include/asm-x86_64/unistd.h
@@ -552,7 +552,7 @@ __SYSCALL(__NR_mq_notify, sys_mq_notify)
 #define __NR_mq_getsetattr 	245
 __SYSCALL(__NR_mq_getsetattr, sys_mq_getsetattr)
 #define __NR_kexec_load 	246
-__SYSCALL(__NR_kexec_load, sys_ni_syscall)
+__SYSCALL(__NR_kexec_load, sys_kexec_load)
 #define __NR_waitid		247
 __SYSCALL(__NR_waitid, sys_waitid)
 #define __NR_add_key		248
-- 
cgit v1.2.3-70-g09d2


From 70765aa4bdb8862a49fcf5b28f3deaf561cf5ae7 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:58:07 -0700
Subject: [PATCH] kexec: kexec ppc support

I have tweaked this patch slightly to handle an empty list
of pages to relocate passed to relocate_new_kernel.  And
I have added ppc_md.machine_crash_shutdown.  To keep up with
the changes in the generic kexec infrastructure.

From: Albert Herranz <albert_herranz@yahoo.es>

The following patch adds support for kexec on the ppc32 platform.

Non-OpenFirmware based platforms are likely to work directly without
additional changes on the kernel side.  The kexec-tools userland package
may need to be slightly updated, though.

For OpenFirmware based machines, additional work is still needed on the
kernel side before kexec support is ready.  Benjamin Herrenschmidt is
kindly working on that part.

In order for a ppc platform to use the kexec kernel services it must
implement some ppc_md hooks.  Otherwise, kexec will be explicitly disabled,
as suggested by benh.

There are 3+1 new ppc_md hooks that a platform supporting kexec may
implement.  Two of them are mandatory for kexec to work.  See
include/asm-ppc/machdep.h for details.

- machine_kexec_prepare(image)

  This function is called to make any arrangements to the image before it
  is loaded.

  This hook _MUST_ be provided by a platform in order to activate kexec
  support for that platform.  Otherwise, the platform is considered to not
  support kexec and the kexec_load system call will fail (that makes all
  existing platforms by default non-kexec'able).

- machine_kexec_cleanup(image)

  This function is called to make any cleanups on image after the loaded
  image data it is freed.  This hook is optional.  A platform may or may
  not provide this hook.

- machine_kexec(image)

  This function is called to perform the _actual_ kexec.  This hook
  _MUST_ be provided by a platform in order to activate kexec support for
  that platform.

  If a platform provides machine_kexec_prepare but forgets to provide
  machine_kexec, a kexec will fall back to a reboot.

  A ready-to-use machine_kexec_simple() generic function is provided to,
  hopefully, simplify kexec adoption for embedded platforms.  A platform
  may call this function from its specific machine_kexec hook, like this:

void myplatform_kexec(struct kimage *image)
{
        machine_kexec_simple(image);
}

- machine_shutdown()

  This function is called to perform any machine specific shutdowns, not
  already done by drivers.  This hook is optional.  A platform may or may
  not provide this hook.

An example (trimmed) platform specific module for a platform supporting
kexec through the existing machine_kexec_simple follows:

/* ... */

#ifdef CONFIG_KEXEC
int myplatform_kexec_prepare(struct kimage *image)
{
        /* here, we can place additional preparations
*/
        return 0; /* yes, we support kexec */
}

void myplatform_kexec(struct kimage *image)
{
        machine_kexec_simple(image);
}
#endif /* CONFIG_KEXEC */

/* ... */

void __init
platform_init(unsigned long r3, unsigned long r4,
unsigned long r5,
              unsigned long r6, unsigned long r7)
{

/* ... */

#ifdef CONFIG_KEXEC
        ppc_md.machine_kexec_prepare =
myplatform_kexec_prepare;
        ppc_md.machine_kexec         =
myplatform_kexec;
#endif /* CONFIG_KEXEC */

/* ... */

}

The kexec ppc kernel support has been heavily tested on the GameCube Linux
port, and, as reported in the fastboot mailing list, it has been tested too
on a Moto 82xx ppc by Rick Richardson.

Signed-off-by: Albert Herranz <albert_herranz@yahoo.es>
Signed-off-by: Eric Biederman <ebiederm@xmission.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/Kconfig                  |  20 +++++++
 arch/ppc/kernel/Makefile          |   1 +
 arch/ppc/kernel/machine_kexec.c   | 122 +++++++++++++++++++++++++++++++++++++
 arch/ppc/kernel/misc.S            |   2 +-
 arch/ppc/kernel/relocate_kernel.S | 123 ++++++++++++++++++++++++++++++++++++++
 include/asm-ppc/kexec.h           |  38 ++++++++++++
 include/asm-ppc/machdep.h         |  31 ++++++++++
 7 files changed, 336 insertions(+), 1 deletion(-)
 create mode 100644 arch/ppc/kernel/machine_kexec.c
 create mode 100644 arch/ppc/kernel/relocate_kernel.S
 create mode 100644 include/asm-ppc/kexec.h

(limited to 'include')

diff --git a/arch/ppc/Kconfig b/arch/ppc/Kconfig
index 979590a9baf..a7835cd3f51 100644
--- a/arch/ppc/Kconfig
+++ b/arch/ppc/Kconfig
@@ -217,6 +217,26 @@ config MATH_EMULATION
 	  here.  Saying Y here will not hurt performance (on any machine) but
 	  will increase the size of the kernel.
 
+config KEXEC
+	bool "kexec system call (EXPERIMENTAL)"
+	depends on EXPERIMENTAL
+	help
+	  kexec is a system call that implements the ability to shutdown your
+	  current kernel, and to start another kernel.  It is like a reboot
+	  but it is indepedent of the system firmware.   And like a reboot
+	  you can start any kernel with it, not just Linux.
+
+	  The name comes from the similiarity to the exec system call.
+
+	  It is an ongoing process to be certain the hardware in a machine
+	  is properly shutdown, so do not be surprised if this code does not
+	  initially work for you.  It may help to enable device hotplugging
+	  support.  As of this writing the exact hardware interface is
+	  strongly in flux, so no good recommendation can be made.
+
+	  In the GameCube implementation, kexec allows you to load and
+	  run DOL files, including kernel and homebrew DOLs.
+
 source "drivers/cpufreq/Kconfig"
 
 config CPU_FREQ_PMAC
diff --git a/arch/ppc/kernel/Makefile b/arch/ppc/kernel/Makefile
index 8276c0b551d..b1457a8a9c0 100644
--- a/arch/ppc/kernel/Makefile
+++ b/arch/ppc/kernel/Makefile
@@ -29,6 +29,7 @@ obj-$(CONFIG_ALTIVEC)		+= vecemu.o vector.o
 ifndef CONFIG_E200
 obj-$(CONFIG_FSL_BOOKE)		+= perfmon_fsl_booke.o
 endif
+obj-$(CONFIG_KEXEC)		+= machine_kexec.o relocate_kernel.o
 
 ifndef CONFIG_MATH_EMULATION
 obj-$(CONFIG_8xx)		+= softemu8xx.o
diff --git a/arch/ppc/kernel/machine_kexec.c b/arch/ppc/kernel/machine_kexec.c
new file mode 100644
index 00000000000..435ad9ea0a8
--- /dev/null
+++ b/arch/ppc/kernel/machine_kexec.c
@@ -0,0 +1,122 @@
+/*
+ * machine_kexec.c - handle transition of Linux booting another kernel
+ * Copyright (C) 2002-2003 Eric Biederman  <ebiederm@xmission.com>
+ *
+ * GameCube/ppc32 port Copyright (C) 2004 Albert Herranz
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2.  See the file COPYING for more details.
+ */
+
+#include <linux/mm.h>
+#include <linux/kexec.h>
+#include <linux/delay.h>
+#include <linux/reboot.h>
+#include <asm/pgtable.h>
+#include <asm/pgalloc.h>
+#include <asm/mmu_context.h>
+#include <asm/io.h>
+#include <asm/hw_irq.h>
+#include <asm/cacheflush.h>
+#include <asm/machdep.h>
+
+typedef NORET_TYPE void (*relocate_new_kernel_t)(
+	unsigned long indirection_page, unsigned long reboot_code_buffer,
+	unsigned long start_address) ATTRIB_NORET;
+
+const extern unsigned char relocate_new_kernel[];
+const extern unsigned int relocate_new_kernel_size;
+
+void machine_shutdown(void)
+{
+	if (ppc_md.machine_shutdown) {
+		ppc_md.machine_shutdown();
+	}
+}
+
+void machine_crash_shutdown(void)
+{
+	if (ppc_md.machine_crash_shutdown) {
+		ppc_md.machine_crash_shutdown();
+	}
+}
+
+/*
+ * Do what every setup is needed on image and the
+ * reboot code buffer to allow us to avoid allocations
+ * later.
+ */
+int machine_kexec_prepare(struct kimage *image)
+{
+	if (ppc_md.machine_kexec_prepare) {
+		return ppc_md.machine_kexec_prepare(image);
+	}
+	/*
+	 * Fail if platform doesn't provide its own machine_kexec_prepare
+	 * implementation.
+	 */
+	return -ENOSYS;
+}
+
+void machine_kexec_cleanup(struct kimage *image)
+{
+	if (ppc_md.machine_kexec_cleanup) {
+		ppc_md.machine_kexec_cleanup(image);
+	}
+}
+
+/*
+ * Do not allocate memory (or fail in any way) in machine_kexec().
+ * We are past the point of no return, committed to rebooting now.
+ */
+NORET_TYPE void machine_kexec(struct kimage *image)
+{
+	if (ppc_md.machine_kexec) {
+		ppc_md.machine_kexec(image);
+	} else {
+		/*
+		 * Fall back to normal restart if platform doesn't provide
+		 * its own kexec function, and user insist to kexec...
+		 */
+		machine_restart(NULL);
+	}
+	for(;;);
+}
+
+
+/*
+ * This is a generic machine_kexec function suitable at least for
+ * non-OpenFirmware embedded platforms.
+ * It merely copies the image relocation code to the control page and
+ * jumps to it.
+ * A platform specific function may just call this one.
+ */
+void machine_kexec_simple(struct kimage *image)
+{
+	unsigned long page_list;
+	unsigned long reboot_code_buffer, reboot_code_buffer_phys;
+	relocate_new_kernel_t rnk;
+
+	/* Interrupts aren't acceptable while we reboot */
+	local_irq_disable();
+
+	page_list = image->head;
+
+	/* we need both effective and real address here */
+	reboot_code_buffer =
+		(unsigned long)page_address(image->control_code_page);
+	reboot_code_buffer_phys = virt_to_phys((void *)reboot_code_buffer);
+
+	/* copy our kernel relocation code to the control code page */
+	memcpy((void *)reboot_code_buffer,
+		relocate_new_kernel, relocate_new_kernel_size);
+
+	flush_icache_range(reboot_code_buffer,
+		reboot_code_buffer + KEXEC_CONTROL_CODE_SIZE);
+	printk(KERN_INFO "Bye!\n");
+
+	/* now call it */
+	rnk = (relocate_new_kernel_t) reboot_code_buffer;
+	(*rnk)(page_list, reboot_code_buffer_phys, image->start);
+}
+
diff --git a/arch/ppc/kernel/misc.S b/arch/ppc/kernel/misc.S
index a3132f8e799..b6a63a49a23 100644
--- a/arch/ppc/kernel/misc.S
+++ b/arch/ppc/kernel/misc.S
@@ -1444,7 +1444,7 @@ _GLOBAL(sys_call_table)
 	.long sys_mq_timedreceive	/* 265 */
 	.long sys_mq_notify
 	.long sys_mq_getsetattr
-	.long sys_ni_syscall		/* 268 reserved for sys_kexec_load */
+	.long sys_kexec_load
 	.long sys_add_key
 	.long sys_request_key		/* 270 */
 	.long sys_keyctl
diff --git a/arch/ppc/kernel/relocate_kernel.S b/arch/ppc/kernel/relocate_kernel.S
new file mode 100644
index 00000000000..7ff69c4af92
--- /dev/null
+++ b/arch/ppc/kernel/relocate_kernel.S
@@ -0,0 +1,123 @@
+/*
+ * relocate_kernel.S - put the kernel image in place to boot
+ * Copyright (C) 2002-2003 Eric Biederman  <ebiederm@xmission.com>
+ *
+ * GameCube/ppc32 port Copyright (C) 2004 Albert Herranz
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2.  See the file COPYING for more details.
+ */
+
+#include <asm/reg.h>
+#include <asm/ppc_asm.h>
+#include <asm/processor.h>
+
+#include <asm/kexec.h>
+
+#define PAGE_SIZE      4096 /* must be same value as in <asm/page.h> */
+
+	/*
+	 * Must be relocatable PIC code callable as a C function.
+	 */
+	.globl relocate_new_kernel
+relocate_new_kernel:
+	/* r3 = page_list   */
+	/* r4 = reboot_code_buffer */
+	/* r5 = start_address      */
+
+	li	r0, 0
+
+	/*
+	 * Set Machine Status Register to a known status,
+	 * switch the MMU off and jump to 1: in a single step.
+	 */
+
+	mr	r8, r0
+	ori     r8, r8, MSR_RI|MSR_ME
+	mtspr	SRR1, r8
+	addi	r8, r4, 1f - relocate_new_kernel
+	mtspr	SRR0, r8
+	sync
+	rfi
+
+1:
+	/* from this point address translation is turned off */
+	/* and interrupts are disabled */
+
+	/* set a new stack at the bottom of our page... */
+	/* (not really needed now) */
+	addi	r1, r4, KEXEC_CONTROL_CODE_SIZE - 8 /* for LR Save+Back Chain */
+	stw	r0, 0(r1)
+
+	/* Do the copies */
+	li	r6, 0 /* checksum */
+	mr	r0, r3
+	b	1f
+
+0:	/* top, read another word for the indirection page */
+	lwzu	r0, 4(r3)
+
+1:
+	/* is it a destination page? (r8) */
+	rlwinm.	r7, r0, 0, 31, 31 /* IND_DESTINATION (1<<0) */
+	beq	2f
+
+	rlwinm	r8, r0, 0, 0, 19 /* clear kexec flags, page align */
+	b	0b
+
+2:	/* is it an indirection page? (r3) */
+	rlwinm.	r7, r0, 0, 30, 30 /* IND_INDIRECTION (1<<1) */
+	beq	2f
+
+	rlwinm	r3, r0, 0, 0, 19 /* clear kexec flags, page align */
+	subi	r3, r3, 4
+	b	0b
+
+2:	/* are we done? */
+	rlwinm.	r7, r0, 0, 29, 29 /* IND_DONE (1<<2) */
+	beq	2f
+	b	3f
+
+2:	/* is it a source page? (r9) */
+	rlwinm.	r7, r0, 0, 28, 28 /* IND_SOURCE (1<<3) */
+	beq	0b
+
+	rlwinm	r9, r0, 0, 0, 19 /* clear kexec flags, page align */
+
+	li	r7, PAGE_SIZE / 4
+	mtctr   r7
+	subi    r9, r9, 4
+	subi    r8, r8, 4
+9:
+	lwzu    r0, 4(r9)  /* do the copy */
+	xor	r6, r6, r0
+	stwu    r0, 4(r8)
+	dcbst	0, r8
+	sync
+	icbi	0, r8
+	bdnz    9b
+
+	addi    r9, r9, 4
+	addi    r8, r8, 4
+	b	0b
+
+3:
+
+	/* To be certain of avoiding problems with self-modifying code
+	 * execute a serializing instruction here.
+	 */
+	isync
+	sync
+
+	/* jump to the entry point, usually the setup routine */
+	mtlr	r5
+	blrl
+
+1:	b	1b
+
+relocate_new_kernel_end:
+
+	.globl relocate_new_kernel_size
+relocate_new_kernel_size:
+	.long relocate_new_kernel_end - relocate_new_kernel
+
diff --git a/include/asm-ppc/kexec.h b/include/asm-ppc/kexec.h
new file mode 100644
index 00000000000..73191310d8d
--- /dev/null
+++ b/include/asm-ppc/kexec.h
@@ -0,0 +1,38 @@
+#ifndef _PPC_KEXEC_H
+#define _PPC_KEXEC_H
+
+#ifdef CONFIG_KEXEC
+
+/*
+ * KEXEC_SOURCE_MEMORY_LIMIT maximum page get_free_page can return.
+ * I.e. Maximum page that is mapped directly into kernel memory,
+ * and kmap is not required.
+ *
+ * Someone correct me if FIXADDR_START - PAGEOFFSET is not the correct
+ * calculation for the amount of memory directly mappable into the
+ * kernel memory space.
+ */
+
+/* Maximum physical address we can use pages from */
+#define KEXEC_SOURCE_MEMORY_LIMIT (-1UL)
+/* Maximum address we can reach in physical address mode */
+#define KEXEC_DESTINATION_MEMORY_LIMIT (-1UL)
+/* Maximum address we can use for the control code buffer */
+#define KEXEC_CONTROL_MEMORY_LIMIT TASK_SIZE
+
+#define KEXEC_CONTROL_CODE_SIZE	4096
+
+/* The native architecture */
+#define KEXEC_ARCH KEXEC_ARCH_PPC
+
+#ifndef __ASSEMBLY__
+
+struct kimage;
+
+extern void machine_kexec_simple(struct kimage *image);
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* CONFIG_KEXEC */
+
+#endif /* _PPC_KEXEC_H */
diff --git a/include/asm-ppc/machdep.h b/include/asm-ppc/machdep.h
index b78d40870c9..1d4ab70a56f 100644
--- a/include/asm-ppc/machdep.h
+++ b/include/asm-ppc/machdep.h
@@ -4,6 +4,7 @@
 
 #include <linux/config.h>
 #include <linux/init.h>
+#include <linux/kexec.h>
 
 #include <asm/setup.h>
 #include <asm/page.h>
@@ -114,6 +115,36 @@ struct machdep_calls {
 	/* functions for dealing with other cpus */
 	struct smp_ops_t *smp_ops;
 #endif /* CONFIG_SMP */
+
+#ifdef CONFIG_KEXEC
+	/* Called to shutdown machine specific hardware not already controlled
+	 * by other drivers.
+	 * XXX Should we move this one out of kexec scope?
+	 */
+	void (*machine_shutdown)(void);
+
+	/* Called to do the minimal shutdown needed to run a kexec'd kernel
+	 * to run successfully.
+	 * XXX Should we move this one out of kexec scope?
+	 */
+	void (*machine_crash_shutdown)(void);
+
+	/* Called to do what every setup is needed on image and the
+	 * reboot code buffer. Returns 0 on success.
+	 * Provide your own (maybe dummy) implementation if your platform
+	 * claims to support kexec.
+	 */
+	int (*machine_kexec_prepare)(struct kimage *image);
+
+	/* Called to handle any machine specific cleanup on image */
+	void (*machine_kexec_cleanup)(struct kimage *image);
+
+	/* Called to perform the _real_ kexec.
+	 * Do NOT allocate memory or fail here. We are past the point of
+	 * no return.
+	 */
+	void (*machine_kexec)(struct kimage *image);
+#endif /* CONFIG_KEXEC */
 };
 
 extern struct machdep_calls ppc_md;
-- 
cgit v1.2.3-70-g09d2


From f4c82d5132b0592f5d6befc5b652cbd4b08f12ff Mon Sep 17 00:00:00 2001
From: R Sharada <sharada@in.ibm.com>
Date: Sat, 25 Jun 2005 14:58:08 -0700
Subject: [PATCH] ppc64 kexec: native hash clear

Add code to clear the hash table and invalidate the tlb for native (SMP,
non-LPAR) mode.  Supports 16M and 4k pages.

Signed-off-by: Milton Miller <miltonm@bga.com>
Signed-off-by: R Sharada <sharada@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/mm/hash_native.c | 47 ++++++++++++++++++++++++++++++++++++++++++++-
 include/asm-ppc64/mmu.h     | 22 +++++++++++++++++++++
 2 files changed, 68 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/ppc64/mm/hash_native.c b/arch/ppc64/mm/hash_native.c
index 52b6b930534..4fec05817d6 100644
--- a/arch/ppc64/mm/hash_native.c
+++ b/arch/ppc64/mm/hash_native.c
@@ -304,6 +304,50 @@ static void native_hpte_invalidate(unsigned long slot, unsigned long va,
 	local_irq_restore(flags);
 }
 
+/*
+ * clear all mappings on kexec.  All cpus are in real mode (or they will
+ * be when they isi), and we are the only one left.  We rely on our kernel
+ * mapping being 0xC0's and the hardware ignoring those two real bits.
+ *
+ * TODO: add batching support when enabled.  remember, no dynamic memory here,
+ * athough there is the control page available...
+ */
+static void native_hpte_clear(void)
+{
+	unsigned long slot, slots, flags;
+	HPTE *hptep = htab_address;
+	Hpte_dword0 dw0;
+	unsigned long pteg_count;
+
+	pteg_count = htab_hash_mask + 1;
+
+	local_irq_save(flags);
+
+	/* we take the tlbie lock and hold it.  Some hardware will
+	 * deadlock if we try to tlbie from two processors at once.
+	 */
+	spin_lock(&native_tlbie_lock);
+
+	slots = pteg_count * HPTES_PER_GROUP;
+
+	for (slot = 0; slot < slots; slot++, hptep++) {
+		/*
+		 * we could lock the pte here, but we are the only cpu
+		 * running,  right?  and for crash dump, we probably
+		 * don't want to wait for a maybe bad cpu.
+		 */
+		dw0 = hptep->dw0.dw0;
+
+		if (dw0.v) {
+			hptep->dw0.dword0 = 0;
+			tlbie(slot2va(dw0.avpn, dw0.l, dw0.h, slot), dw0.l);
+		}
+	}
+
+	spin_unlock(&native_tlbie_lock);
+	local_irq_restore(flags);
+}
+
 static void native_flush_hash_range(unsigned long context,
 				    unsigned long number, int local)
 {
@@ -415,7 +459,8 @@ void hpte_init_native(void)
 	ppc_md.hpte_updatepp	= native_hpte_updatepp;
 	ppc_md.hpte_updateboltedpp = native_hpte_updateboltedpp;
 	ppc_md.hpte_insert	= native_hpte_insert;
-	ppc_md.hpte_remove     	= native_hpte_remove;
+	ppc_md.hpte_remove	= native_hpte_remove;
+	ppc_md.hpte_clear_all	= native_hpte_clear;
 	if (tlb_batching_enabled())
 		ppc_md.flush_hash_range = native_flush_hash_range;
 	htab_finish_init();
diff --git a/include/asm-ppc64/mmu.h b/include/asm-ppc64/mmu.h
index 9d03a98a4fa..f373de5e3dd 100644
--- a/include/asm-ppc64/mmu.h
+++ b/include/asm-ppc64/mmu.h
@@ -181,6 +181,28 @@ static inline void tlbiel(unsigned long va)
 	asm volatile("ptesync": : :"memory");
 }
 
+static inline unsigned long slot2va(unsigned long avpn, unsigned long large,
+		unsigned long secondary, unsigned long slot)
+{
+	unsigned long va;
+
+	va = avpn << 23;
+
+	if (!large) {
+		unsigned long vpi, pteg;
+
+		pteg = slot / HPTES_PER_GROUP;
+		if (secondary)
+			pteg = ~pteg;
+
+		vpi = ((va >> 28) ^ pteg) & htab_hash_mask;
+
+		va |= vpi << PAGE_SHIFT;
+	}
+
+	return va;
+}
+
 /*
  * Handle a fault by adding an HPTE. If the address can't be determined
  * to be valid via Linux page tables, return 1. If handled return 0
-- 
cgit v1.2.3-70-g09d2


From fce0d5740322b98b863f9e609f5a9bd4c06703af Mon Sep 17 00:00:00 2001
From: R Sharada <sharada@in.ibm.com>
Date: Sat, 25 Jun 2005 14:58:10 -0700
Subject: [PATCH] ppc64: kexec support for ppc64

This patch implements the kexec support for ppc64 platforms.

A couple of notes:

1)  We copy the pages in virtual mode, using the full base kernel
    and a statically allocated stack.   At kexec_prepare time we
    scan the pages and if any overlap our (0, _end[]) range we
    return -ETXTBSY.

    On PowerPC 64 systems running in LPAR (logical partitioning)
    mode, only a small region of memory, referred to as the RMO,
    can be accessed in real mode.  Since Linux runs with only one
    zone of memory in the memory allocator, and it can be orders of
    magnitude more memory than the RMO, looping until we allocate
    pages in the source region is not feasible.  Copying in virtual
    means we don't have to write a hash table generation and call
    hypervisor to insert translations, instead we rely on the pinned
    kernel linear mapping.  The kernel already has move to linked
    location built in, so there is no requirement to load it at 0.

    If we want to load something other than a kernel, then a stub
    can be written to copy a linear chunk in real mode.

2)  The start entry point gets passed parameters from the kernel.
    Slaves are started at a fixed address after copying code from
    the entry point.

    All CPUs get passed their firmware assigned physical id in r3
    (most calling conventions use this register for the first
    argument).

    This is used to distinguish each CPU from all other CPUs.
    Since firmware is not around, there is no other way to obtain
    this information other than to pass it somewhere.

    A single CPU, referred to here as the master and the one executing
    the kexec call, branches to start with the address of start in r4.
    While this can be calculated, we have to load it through a gpr to
    branch to this point so defining the register this is contained
    in is free.  A stack of unspecified size is available at r1
    (also common calling convention).

    All remaining running CPUs are sent to start at absolute address
    0x60 after copying the first 0x100 bytes from start to address 0.
    This convention was chosen because it matches what the kernel
    has been doing itself.  (only gpr3 is defined).

    Note: This is not quite the convention of the kexec bootblock v2
    in the kernel.  A stub has been written to convert between them,
    and we may adjust the kernel in the future to allow this directly
    without any stub.

3)  Destination pages can be placed anywhere, even where they
    would not be accessible in real mode.  This will allow us to
    place ram disks above the RMO if we choose.

Signed-off-by: Milton Miller <miltonm@bga.com>
Signed-off-by: R Sharada <sharada@in.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/Kconfig                |  17 +++
 arch/ppc64/kernel/Makefile        |   1 +
 arch/ppc64/kernel/head.S          |   6 +-
 arch/ppc64/kernel/machine_kexec.c | 301 ++++++++++++++++++++++++++++++++++++++
 arch/ppc64/kernel/misc.S          | 175 +++++++++++++++++++++-
 arch/ppc64/kernel/mpic.c          |  29 ++++
 arch/ppc64/kernel/mpic.h          |   3 +
 arch/ppc64/kernel/pSeries_setup.c |   4 +-
 arch/ppc64/kernel/setup.c         |  22 +--
 arch/ppc64/kernel/xics.c          |  25 ++++
 include/asm-ppc64/kexec.h         |  41 ++++++
 include/asm-ppc64/machdep.h       |   1 +
 include/asm-ppc64/xics.h          |   1 +
 13 files changed, 609 insertions(+), 17 deletions(-)
 create mode 100644 arch/ppc64/kernel/machine_kexec.c
 create mode 100644 include/asm-ppc64/kexec.h

(limited to 'include')

diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig
index 5f40b438b58..f804f25232a 100644
--- a/arch/ppc64/Kconfig
+++ b/arch/ppc64/Kconfig
@@ -142,6 +142,23 @@ config PPC_SPLPAR
 	  processors, that is, which share physical processors between
 	  two or more partitions.
 
+config KEXEC
+	bool "kexec system call (EXPERIMENTAL)"
+	depends on PPC_MULTIPLATFORM && EXPERIMENTAL
+	help
+	  kexec is a system call that implements the ability to shutdown your
+	  current kernel, and to start another kernel.  It is like a reboot
+	  but it is indepedent of the system firmware.  And like a reboot
+	  you can start any kernel with it, not just Linux.
+
+	  The name comes from the similiarity to the exec system call.
+
+	  It is an ongoing process to be certain the hardware in a machine
+	  is properly shutdown, so do not be surprised if this code does not
+	  initially work for you.  It may help to enable device hotplugging
+	  support.  As of this writing the exact hardware interface is
+	  strongly in flux, so no good recommendation can be made.
+
 config IBMVIO
 	depends on PPC_PSERIES || PPC_ISERIES
 	bool
diff --git a/arch/ppc64/kernel/Makefile b/arch/ppc64/kernel/Makefile
index dffbfb7ac8d..d9b2660ef22 100644
--- a/arch/ppc64/kernel/Makefile
+++ b/arch/ppc64/kernel/Makefile
@@ -36,6 +36,7 @@ obj-$(CONFIG_PPC_PSERIES) += pSeries_pci.o pSeries_lpar.o pSeries_hvCall.o \
 obj-$(CONFIG_PPC_BPA) += bpa_setup.o bpa_iommu.o bpa_nvram.o \
 			 bpa_iic.o spider-pic.o
 
+obj-$(CONFIG_KEXEC)		+= machine_kexec.o
 obj-$(CONFIG_EEH)		+= eeh.o
 obj-$(CONFIG_PROC_FS)		+= proc_ppc64.o
 obj-$(CONFIG_RTAS_FLASH)	+= rtas_flash.o
diff --git a/arch/ppc64/kernel/head.S b/arch/ppc64/kernel/head.S
index 02c8f4e3e4b..675c2708588 100644
--- a/arch/ppc64/kernel/head.S
+++ b/arch/ppc64/kernel/head.S
@@ -1194,7 +1194,7 @@ _GLOBAL(pSeries_secondary_smp_init)
 	bl	.__restore_cpu_setup
 
 	/* Set up a paca value for this processor. Since we have the
-	 * physical cpu id in r3, we need to search the pacas to find
+	 * physical cpu id in r24, we need to search the pacas to find
 	 * which logical id maps to our physical one.
 	 */
 	LOADADDR(r13, paca) 		/* Get base vaddr of paca array	 */
@@ -1207,8 +1207,8 @@ _GLOBAL(pSeries_secondary_smp_init)
 	cmpwi	r5,NR_CPUS
 	blt	1b
 
-99:	HMT_LOW				/* Couldn't find our CPU id      */
-	b	99b
+	mr	r3,r24			/* not found, copy phys to r3	 */
+	b	.kexec_wait		/* next kernel might do better	 */
 
 2:	mtspr	SPRG3,r13		/* Save vaddr of paca in SPRG3	 */
 	/* From now on, r24 is expected to be logica cpuid */
diff --git a/arch/ppc64/kernel/machine_kexec.c b/arch/ppc64/kernel/machine_kexec.c
new file mode 100644
index 00000000000..217965d60a4
--- /dev/null
+++ b/arch/ppc64/kernel/machine_kexec.c
@@ -0,0 +1,301 @@
+/*
+ * machine_kexec.c - handle transition of Linux booting another kernel
+ *
+ * Copyright (C) 2004-2005, IBM Corp.
+ *
+ * Created by: Milton D Miller II
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2.  See the file COPYING for more details.
+ */
+
+
+#include <linux/cpumask.h>
+#include <linux/kexec.h>
+#include <linux/smp.h>
+#include <linux/thread_info.h>
+#include <linux/errno.h>
+
+#include <asm/page.h>
+#include <asm/current.h>
+#include <asm/machdep.h>
+#include <asm/cacheflush.h>
+#include <asm/paca.h>
+#include <asm/mmu.h>
+#include <asm/sections.h>	/* _end */
+#include <asm/prom.h>
+
+#define HASH_GROUP_SIZE 0x80	/* size of each hash group, asm/mmu.h */
+
+/* Have this around till we move it into crash specific file */
+note_buf_t crash_notes[NR_CPUS];
+
+/* Dummy for now. Not sure if we need to have a crash shutdown in here
+ * and if what it will achieve. Letting it be now to compile the code
+ * in generic kexec environment
+ */
+void machine_crash_shutdown(void)
+{
+	/* do nothing right now */
+	/* smp_relase_cpus() if we want smp on panic kernel */
+	/* cpu_irq_down to isolate us until we are ready */
+}
+
+int machine_kexec_prepare(struct kimage *image)
+{
+	int i;
+	unsigned long begin, end;	/* limits of segment */
+	unsigned long low, high;	/* limits of blocked memory range */
+	struct device_node *node;
+	unsigned long *basep;
+	unsigned int *sizep;
+
+	if (!ppc_md.hpte_clear_all)
+		return -ENOENT;
+
+	/*
+	 * Since we use the kernel fault handlers and paging code to
+	 * handle the virtual mode, we must make sure no destination
+	 * overlaps kernel static data or bss.
+	 */
+	for(i = 0; i < image->nr_segments; i++)
+		if (image->segment[i].mem < __pa(_end))
+			return -ETXTBSY;
+
+	/*
+	 * For non-LPAR, we absolutely can not overwrite the mmu hash
+	 * table, since we are still using the bolted entries in it to
+	 * do the copy.  Check that here.
+	 *
+	 * It is safe if the end is below the start of the blocked
+	 * region (end <= low), or if the beginning is after the
+	 * end of the blocked region (begin >= high).  Use the
+	 * boolean identity !(a || b)  === (!a && !b).
+	 */
+	if (htab_address) {
+		low = __pa(htab_address);
+		high = low + (htab_hash_mask + 1) * HASH_GROUP_SIZE;
+
+		for(i = 0; i < image->nr_segments; i++) {
+			begin = image->segment[i].mem;
+			end = begin + image->segment[i].memsz;
+
+			if ((begin < high) && (end > low))
+				return -ETXTBSY;
+		}
+	}
+
+	/* We also should not overwrite the tce tables */
+	for (node = of_find_node_by_type(NULL, "pci"); node != NULL;
+			node = of_find_node_by_type(node, "pci")) {
+		basep = (unsigned long *)get_property(node, "linux,tce-base",
+							NULL);
+		sizep = (unsigned int *)get_property(node, "linux,tce-size",
+							NULL);
+		if (basep == NULL || sizep == NULL)
+			continue;
+
+		low = *basep;
+		high = low + (*sizep);
+
+		for(i = 0; i < image->nr_segments; i++) {
+			begin = image->segment[i].mem;
+			end = begin + image->segment[i].memsz;
+
+			if ((begin < high) && (end > low))
+				return -ETXTBSY;
+		}
+	}
+
+	return 0;
+}
+
+void machine_kexec_cleanup(struct kimage *image)
+{
+	/* we do nothing in prepare that needs to be undone */
+}
+
+#define IND_FLAGS (IND_DESTINATION | IND_INDIRECTION | IND_DONE | IND_SOURCE)
+
+static void copy_segments(unsigned long ind)
+{
+	unsigned long entry;
+	unsigned long *ptr;
+	void *dest;
+	void *addr;
+
+	/*
+	 * We rely on kexec_load to create a lists that properly
+	 * initializes these pointers before they are used.
+	 * We will still crash if the list is wrong, but at least
+	 * the compiler will be quiet.
+	 */
+	ptr = NULL;
+	dest = NULL;
+
+	for (entry = ind; !(entry & IND_DONE); entry = *ptr++) {
+		addr = __va(entry & PAGE_MASK);
+
+		switch (entry & IND_FLAGS) {
+		case IND_DESTINATION:
+			dest = addr;
+			break;
+		case IND_INDIRECTION:
+			ptr = addr;
+			break;
+		case IND_SOURCE:
+			copy_page(dest, addr);
+			dest += PAGE_SIZE;
+		}
+	}
+}
+
+void kexec_copy_flush(struct kimage *image)
+{
+	long i, nr_segments = image->nr_segments;
+	struct  kexec_segment ranges[KEXEC_SEGMENT_MAX];
+
+	/* save the ranges on the stack to efficiently flush the icache */
+	memcpy(ranges, image->segment, sizeof(ranges));
+
+	/*
+	 * After this call we may not use anything allocated in dynamic
+	 * memory, including *image.
+	 *
+	 * Only globals and the stack are allowed.
+	 */
+	copy_segments(image->head);
+
+	/*
+	 * we need to clear the icache for all dest pages sometime,
+	 * including ones that were in place on the original copy
+	 */
+	for (i = 0; i < nr_segments; i++)
+		flush_icache_range(ranges[i].mem + KERNELBASE,
+				ranges[i].mem + KERNELBASE +
+				ranges[i].memsz);
+}
+
+#ifdef CONFIG_SMP
+
+/* FIXME: we should schedule this function to be called on all cpus based
+ * on calling the interrupts, but we would like to call it off irq level
+ * so that the interrupt controller is clean.
+ */
+void kexec_smp_down(void *arg)
+{
+	if (ppc_md.cpu_irq_down)
+		ppc_md.cpu_irq_down();
+
+	local_irq_disable();
+	kexec_smp_wait();
+	/* NOTREACHED */
+}
+
+static void kexec_prepare_cpus(void)
+{
+	int my_cpu, i, notified=-1;
+
+	smp_call_function(kexec_smp_down, NULL, 0, /* wait */0);
+	my_cpu = get_cpu();
+
+	/* check the others cpus are now down (via paca hw cpu id == -1) */
+	for (i=0; i < NR_CPUS; i++) {
+		if (i == my_cpu)
+			continue;
+
+		while (paca[i].hw_cpu_id != -1) {
+			if (!cpu_possible(i)) {
+				printk("kexec: cpu %d hw_cpu_id %d is not"
+						" possible, ignoring\n",
+						i, paca[i].hw_cpu_id);
+				break;
+			}
+			if (!cpu_online(i)) {
+				/* Fixme: this can be spinning in
+				 * pSeries_secondary_wait with a paca
+				 * waiting for it to go online.
+				 */
+				printk("kexec: cpu %d hw_cpu_id %d is not"
+						" online, ignoring\n",
+						i, paca[i].hw_cpu_id);
+				break;
+			}
+			if (i != notified) {
+				printk( "kexec: waiting for cpu %d (physical"
+						" %d) to go down\n",
+						i, paca[i].hw_cpu_id);
+				notified = i;
+			}
+		}
+	}
+
+	/* after we tell the others to go down */
+	if (ppc_md.cpu_irq_down)
+		ppc_md.cpu_irq_down();
+
+	put_cpu();
+
+	local_irq_disable();
+}
+
+#else /* ! SMP */
+
+static void kexec_prepare_cpus(void)
+{
+	/*
+	 * move the secondarys to us so that we can copy
+	 * the new kernel 0-0x100 safely
+	 *
+	 * do this if kexec in setup.c ?
+	 */
+	smp_relase_cpus();
+	if (ppc_md.cpu_irq_down)
+		ppc_md.cpu_irq_down();
+	local_irq_disable();
+}
+
+#endif /* SMP */
+
+/*
+ * kexec thread structure and stack.
+ *
+ * We need to make sure that this is 16384-byte aligned due to the
+ * way process stacks are handled.  It also must be statically allocated
+ * or allocated as part of the kimage, because everything else may be
+ * overwritten when we copy the kexec image.  We piggyback on the
+ * "init_task" linker section here to statically allocate a stack.
+ *
+ * We could use a smaller stack if we don't care about anything using
+ * current, but that audit has not been performed.
+ */
+union thread_union kexec_stack
+	__attribute__((__section__(".data.init_task"))) = { };
+
+/* Our assembly helper, in kexec_stub.S */
+extern NORET_TYPE void kexec_sequence(void *newstack, unsigned long start,
+	void *image, void *control, void (*clear_all)(void)) ATTRIB_NORET;
+
+/* too late to fail here */
+void machine_kexec(struct kimage *image)
+{
+
+	/* prepare control code if any */
+
+	/* shutdown other cpus into our wait loop and quiesce interrupts */
+	kexec_prepare_cpus();
+
+	/* switch to a staticly allocated stack.  Based on irq stack code.
+	 * XXX: the task struct will likely be invalid once we do the copy!
+	 */
+	kexec_stack.thread_info.task = current_thread_info()->task;
+	kexec_stack.thread_info.flags = 0;
+
+	/* Some things are best done in assembly.  Finding globals with
+	 * a toc is easier in C, so pass in what we can.
+	 */
+	kexec_sequence(&kexec_stack, image->start, image,
+			page_address(image->control_code_page),
+			ppc_md.hpte_clear_all);
+	/* NOTREACHED */
+}
diff --git a/arch/ppc64/kernel/misc.S b/arch/ppc64/kernel/misc.S
index e3c73b3425d..f3dea0c5a88 100644
--- a/arch/ppc64/kernel/misc.S
+++ b/arch/ppc64/kernel/misc.S
@@ -680,6 +680,177 @@ _GLOBAL(kernel_thread)
 	ld	r30,-16(r1)
 	blr
 
+/* kexec_wait(phys_cpu)
+ *
+ * wait for the flag to change, indicating this kernel is going away but
+ * the slave code for the next one is at addresses 0 to 100.
+ *
+ * This is used by all slaves.
+ *
+ * Physical (hardware) cpu id should be in r3.
+ */
+_GLOBAL(kexec_wait)
+	bl	1f
+1:	mflr	r5
+	addi	r5,r5,kexec_flag-1b
+
+99:	HMT_LOW
+#ifdef CONFIG_KEXEC		/* use no memory without kexec */
+	lwz	r4,0(r5)
+	cmpwi	0,r4,0
+	bnea	0x60
+#endif
+	b	99b
+
+/* this can be in text because we won't change it until we are
+ * running in real anyways
+ */
+kexec_flag:
+	.long	0
+
+
+#ifdef CONFIG_KEXEC
+
+/* kexec_smp_wait(void)
+ *
+ * call with interrupts off
+ * note: this is a terminal routine, it does not save lr
+ *
+ * get phys id from paca
+ * set paca id to -1 to say we got here
+ * switch to real mode
+ * join other cpus in kexec_wait(phys_id)
+ */
+_GLOBAL(kexec_smp_wait)
+	lhz	r3,PACAHWCPUID(r13)
+	li	r4,-1
+	sth	r4,PACAHWCPUID(r13)	/* let others know we left */
+	bl	real_mode
+	b	.kexec_wait
+
+/*
+ * switch to real mode (turn mmu off)
+ * we use the early kernel trick that the hardware ignores bits
+ * 0 and 1 (big endian) of the effective address in real mode
+ *
+ * don't overwrite r3 here, it is live for kexec_wait above.
+ */
+real_mode:	/* assume normal blr return */
+1:	li	r9,MSR_RI
+	li	r10,MSR_DR|MSR_IR
+	mflr	r11		/* return address to SRR0 */
+	mfmsr	r12
+	andc	r9,r12,r9
+	andc	r10,r12,r10
+
+	mtmsrd	r9,1
+	mtspr	SPRN_SRR1,r10
+	mtspr	SPRN_SRR0,r11
+	rfid
+
+
+/*
+ * kexec_sequence(newstack, start, image, control, clear_all())
+ *
+ * does the grungy work with stack switching and real mode switches
+ * also does simple calls to other code
+ */
+
+_GLOBAL(kexec_sequence)
+	mflr	r0
+	std	r0,16(r1)
+
+	/* switch stacks to newstack -- &kexec_stack.stack */
+	stdu	r1,THREAD_SIZE-112(r3)
+	mr	r1,r3
+
+	li	r0,0
+	std	r0,16(r1)
+
+	/* save regs for local vars on new stack.
+	 * yes, we won't go back, but ...
+	 */
+	std	r31,-8(r1)
+	std	r30,-16(r1)
+	std	r29,-24(r1)
+	std	r28,-32(r1)
+	std	r27,-40(r1)
+	std	r26,-48(r1)
+	std	r25,-56(r1)
+
+	stdu	r1,-112-64(r1)
+
+	/* save args into preserved regs */
+	mr	r31,r3			/* newstack (both) */
+	mr	r30,r4			/* start (real) */
+	mr	r29,r5			/* image (virt) */
+	mr	r28,r6			/* control, unused */
+	mr	r27,r7			/* clear_all() fn desc */
+	mr	r26,r8			/* spare */
+	lhz	r25,PACAHWCPUID(r13)	/* get our phys cpu from paca */
+
+	/* disable interrupts, we are overwriting kernel data next */
+	mfmsr	r3
+	rlwinm	r3,r3,0,17,15
+	mtmsrd	r3,1
+
+	/* copy dest pages, flush whole dest image */
+	mr	r3,r29
+	bl	.kexec_copy_flush	/* (image) */
+
+	/* turn off mmu */
+	bl	real_mode
+
+	/* clear out hardware hash page table and tlb */
+	ld	r5,0(r27)		/* deref function descriptor */
+	mtctr	r5
+	bctrl				/* ppc_md.hash_clear_all(void); */
+
+/*
+ *   kexec image calling is:
+ *      the first 0x100 bytes of the entry point are copied to 0
+ *
+ *      all slaves branch to slave = 0x60 (absolute)
+ *              slave(phys_cpu_id);
+ *
+ *      master goes to start = entry point
+ *              start(phys_cpu_id, start, 0);
+ *
+ *
+ *   a wrapper is needed to call existing kernels, here is an approximate
+ *   description of one method:
+ *
+ * v2: (2.6.10)
+ *   start will be near the boot_block (maybe 0x100 bytes before it?)
+ *   it will have a 0x60, which will b to boot_block, where it will wait
+ *   and 0 will store phys into struct boot-block and load r3 from there,
+ *   copy kernel 0-0x100 and tell slaves to back down to 0x60 again
+ *
+ * v1: (2.6.9)
+ *    boot block will have all cpus scanning device tree to see if they
+ *    are the boot cpu ?????
+ *    other device tree differences (prop sizes, va vs pa, etc)...
+ */
+
+	/* copy  0x100 bytes starting at start to 0 */
+	li	r3,0
+	mr	r4,r30
+	li	r5,0x100
+	li	r6,0
+	bl	.copy_and_flush	/* (dest, src, copy limit, start offset) */
+1:	/* assume normal blr return */
+
+	/* release other cpus to the new kernel secondary start at 0x60 */
+	mflr	r5
+	li	r6,1
+	stw	r6,kexec_flag-1b(5)
+	mr	r3,r25	# my phys cpu
+	mr	r4,r30	# start, aka phys mem offset
+	mtlr	4
+	li	r5,0
+	blr	/* image->start(physid, image->start, 0); */
+#endif /* CONFIG_KEXEC */
+
 /* Why isn't this a) automatic, b) written in 'C'? */	
 	.balign 8
 _GLOBAL(sys_call_table32)
@@ -951,7 +1122,7 @@ _GLOBAL(sys_call_table32)
 	.llong .compat_sys_mq_timedreceive /* 265 */
 	.llong .compat_sys_mq_notify
 	.llong .compat_sys_mq_getsetattr
-	.llong .sys_ni_syscall		/* 268 reserved for sys_kexec_load */
+	.llong .compat_sys_kexec_load
 	.llong .sys32_add_key
 	.llong .sys32_request_key
 	.llong .compat_sys_keyctl
@@ -1227,7 +1398,7 @@ _GLOBAL(sys_call_table)
 	.llong .sys_mq_timedreceive	/* 265 */
 	.llong .sys_mq_notify
 	.llong .sys_mq_getsetattr
-	.llong .sys_ni_syscall		/* 268 reserved for sys_kexec_load */
+	.llong .sys_kexec_load
 	.llong .sys_add_key
 	.llong .sys_request_key		/* 270 */
 	.llong .sys_keyctl
diff --git a/arch/ppc64/kernel/mpic.c b/arch/ppc64/kernel/mpic.c
index 593ea5b82af..e8fbab1df37 100644
--- a/arch/ppc64/kernel/mpic.c
+++ b/arch/ppc64/kernel/mpic.c
@@ -792,6 +792,35 @@ void mpic_setup_this_cpu(void)
 #endif /* CONFIG_SMP */
 }
 
+/*
+ * XXX: someone who knows mpic should check this.
+ * do we need to eoi the ipi here (see xics comments)?
+ * or can we reset the mpic in the new kernel?
+ */
+void mpic_teardown_this_cpu(void)
+{
+	struct mpic *mpic = mpic_primary;
+	unsigned long flags;
+	u32 msk = 1 << hard_smp_processor_id();
+	unsigned int i;
+
+	BUG_ON(mpic == NULL);
+
+	DBG("%s: teardown_this_cpu(%d)\n", mpic->name, hard_smp_processor_id());
+	spin_lock_irqsave(&mpic_lock, flags);
+
+	/* let the mpic know we don't want intrs.  */
+	for (i = 0; i < mpic->num_sources ; i++)
+		mpic_irq_write(i, MPIC_IRQ_DESTINATION,
+			mpic_irq_read(i, MPIC_IRQ_DESTINATION) & ~msk);
+
+	/* Set current processor priority to max */
+	mpic_cpu_write(MPIC_CPU_CURRENT_TASK_PRI, 0xf);
+
+	spin_unlock_irqrestore(&mpic_lock, flags);
+}
+
+
 void mpic_send_ipi(unsigned int ipi_no, unsigned int cpu_mask)
 {
 	struct mpic *mpic = mpic_primary;
diff --git a/arch/ppc64/kernel/mpic.h b/arch/ppc64/kernel/mpic.h
index 63e177143ea..99fbbc9a084 100644
--- a/arch/ppc64/kernel/mpic.h
+++ b/arch/ppc64/kernel/mpic.h
@@ -255,6 +255,9 @@ extern unsigned int mpic_irq_get_priority(unsigned int irq);
 /* Setup a non-boot CPU */
 extern void mpic_setup_this_cpu(void);
 
+/* Clean up for kexec (or cpu offline or ...) */
+extern void mpic_teardown_this_cpu(void);
+
 /* Request IPIs on primary mpic */
 extern void mpic_request_ipis(void);
 
diff --git a/arch/ppc64/kernel/pSeries_setup.c b/arch/ppc64/kernel/pSeries_setup.c
index f2b41243342..44d9af72d22 100644
--- a/arch/ppc64/kernel/pSeries_setup.c
+++ b/arch/ppc64/kernel/pSeries_setup.c
@@ -187,14 +187,16 @@ static void __init pSeries_setup_arch(void)
 {
 	/* Fixup ppc_md depending on the type of interrupt controller */
 	if (ppc64_interrupt_controller == IC_OPEN_PIC) {
-		ppc_md.init_IRQ       = pSeries_init_mpic; 
+		ppc_md.init_IRQ       = pSeries_init_mpic;
 		ppc_md.get_irq        = mpic_get_irq;
+	 	ppc_md.cpu_irq_down   = mpic_teardown_this_cpu;
 		/* Allocate the mpic now, so that find_and_init_phbs() can
 		 * fill the ISUs */
 		pSeries_setup_mpic();
 	} else {
 		ppc_md.init_IRQ       = xics_init_IRQ;
 		ppc_md.get_irq        = xics_get_irq;
+		ppc_md.cpu_irq_down   = xics_teardown_cpu;
 	}
 
 #ifdef CONFIG_SMP
diff --git a/arch/ppc64/kernel/setup.c b/arch/ppc64/kernel/setup.c
index 0a47a5ef428..d5e4866e9ac 100644
--- a/arch/ppc64/kernel/setup.c
+++ b/arch/ppc64/kernel/setup.c
@@ -677,11 +677,16 @@ void __init setup_system(void)
 	DBG(" <- setup_system()\n");
 }
 
-
-void machine_restart(char *cmd)
+/* also used by kexec */
+void machine_shutdown(void)
 {
 	if (ppc_md.nvram_sync)
 		ppc_md.nvram_sync();
+}
+
+void machine_restart(char *cmd)
+{
+	machine_shutdown();
 	ppc_md.restart(cmd);
 #ifdef CONFIG_SMP
 	smp_send_stop();
@@ -690,13 +695,11 @@ void machine_restart(char *cmd)
 	local_irq_disable();
 	while (1) ;
 }
-
 EXPORT_SYMBOL(machine_restart);
-  
+
 void machine_power_off(void)
 {
-	if (ppc_md.nvram_sync)
-		ppc_md.nvram_sync();
+	machine_shutdown();
 	ppc_md.power_off();
 #ifdef CONFIG_SMP
 	smp_send_stop();
@@ -705,13 +708,11 @@ void machine_power_off(void)
 	local_irq_disable();
 	while (1) ;
 }
-
 EXPORT_SYMBOL(machine_power_off);
-  
+
 void machine_halt(void)
 {
-	if (ppc_md.nvram_sync)
-		ppc_md.nvram_sync();
+	machine_shutdown();
 	ppc_md.halt();
 #ifdef CONFIG_SMP
 	smp_send_stop();
@@ -720,7 +721,6 @@ void machine_halt(void)
 	local_irq_disable();
 	while (1) ;
 }
-
 EXPORT_SYMBOL(machine_halt);
 
 static int ppc64_panic_event(struct notifier_block *this,
diff --git a/arch/ppc64/kernel/xics.c b/arch/ppc64/kernel/xics.c
index 879f39b90a3..677c4450984 100644
--- a/arch/ppc64/kernel/xics.c
+++ b/arch/ppc64/kernel/xics.c
@@ -647,6 +647,31 @@ static void xics_set_affinity(unsigned int virq, cpumask_t cpumask)
 	}
 }
 
+void xics_teardown_cpu(void)
+{
+	int cpu = smp_processor_id();
+	int status;
+
+	ops->cppr_info(cpu, 0x00);
+	iosync();
+
+	/*
+	 * we need to EOI the IPI if we got here from kexec down IPI
+	 *
+	 * xics doesn't care if we duplicate an EOI as long as we
+	 * don't EOI and raise priority.
+	 *
+	 * probably need to check all the other interrupts too
+	 * should we be flagging idle loop instead?
+	 * or creating some task to be scheduled?
+	 */
+	ops->xirr_info_set(cpu, XICS_IPI);
+
+	status = rtas_set_indicator(GLOBAL_INTERRUPT_QUEUE,
+		(1UL << interrupt_server_size) - 1 - default_distrib_server, 0);
+	WARN_ON(status != 0);
+}
+
 #ifdef CONFIG_HOTPLUG_CPU
 
 /* Interrupts are disabled. */
diff --git a/include/asm-ppc64/kexec.h b/include/asm-ppc64/kexec.h
new file mode 100644
index 00000000000..511908afaee
--- /dev/null
+++ b/include/asm-ppc64/kexec.h
@@ -0,0 +1,41 @@
+#ifndef _PPC64_KEXEC_H
+#define _PPC64_KEXEC_H
+
+/*
+ * KEXEC_SOURCE_MEMORY_LIMIT maximum page get_free_page can return.
+ * I.e. Maximum page that is mapped directly into kernel memory,
+ * and kmap is not required.
+ */
+
+/* Maximum physical address we can use pages from */
+/* XXX: since we copy virt we can use any page we allocate */
+#define KEXEC_SOURCE_MEMORY_LIMIT (-1UL)
+
+/* Maximum address we can reach in physical address mode */
+/* XXX: I want to allow initrd in highmem.  otherwise set to rmo on lpar */
+#define KEXEC_DESTINATION_MEMORY_LIMIT (-1UL)
+
+/* Maximum address we can use for the control code buffer */
+/* XXX: unused today, ppc32 uses TASK_SIZE, probably left over from use_mm  */
+#define KEXEC_CONTROL_MEMORY_LIMIT (-1UL)
+
+/* XXX: today we don't use this at all, althogh we have a static stack */
+#define KEXEC_CONTROL_CODE_SIZE 4096
+
+/* The native architecture */
+#define KEXEC_ARCH KEXEC_ARCH_PPC64
+
+#define MAX_NOTE_BYTES 1024
+
+#ifndef __ASSEMBLY__
+
+typedef u32 note_buf_t[MAX_NOTE_BYTES/4];
+
+extern note_buf_t crash_notes[];
+
+extern void kexec_smp_wait(void);	/* get and clear naca physid, wait for
+					  master to copy new code to 0 */
+
+#endif /* __ASSEMBLY__ */
+#endif /* _PPC_KEXEC_H */
+
diff --git a/include/asm-ppc64/machdep.h b/include/asm-ppc64/machdep.h
index 553b2ea23be..9cdad3ed152 100644
--- a/include/asm-ppc64/machdep.h
+++ b/include/asm-ppc64/machdep.h
@@ -86,6 +86,7 @@ struct machdep_calls {
 
 	void		(*init_IRQ)(void);
 	int		(*get_irq)(struct pt_regs *);
+	void		(*cpu_irq_down)(void);
 
 	/* PCI stuff */
 	void		(*pcibios_fixup)(void);
diff --git a/include/asm-ppc64/xics.h b/include/asm-ppc64/xics.h
index fdec5e7a7af..0c45e14e26c 100644
--- a/include/asm-ppc64/xics.h
+++ b/include/asm-ppc64/xics.h
@@ -17,6 +17,7 @@
 void xics_init_IRQ(void);
 int xics_get_irq(struct pt_regs *);
 void xics_setup_cpu(void);
+void xics_teardown_cpu(void);
 void xics_cause_IPI(int cpu);
 void xics_request_IPIs(void);
 void xics_migrate_irqs_away(void);
-- 
cgit v1.2.3-70-g09d2


From cf13f0eaffa31bf6a145c53c589654b11c72ddc7 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sat, 25 Jun 2005 14:58:11 -0700
Subject: [PATCH] kexec: s390 support

Add kexec support for s390 architecture.

From: Milton Miller <miltonm@bga.com>

- Fix passing of first argument to relocate_kernel assembly.
- Fix Kconfig description.
- Remove wrong comment and comments that describe obvious things.
- Allow only KEXEC_TYPE_DEFAULT as image type -> dump not supported.

Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/Kconfig                    |  8 +++
 arch/s390/kernel/Makefile            | 10 ++++
 arch/s390/kernel/compat_wrapper.S    |  8 +++
 arch/s390/kernel/crash.c             | 17 +++++++
 arch/s390/kernel/machine_kexec.c     | 98 ++++++++++++++++++++++++++++++++++++
 arch/s390/kernel/relocate_kernel.S   | 81 +++++++++++++++++++++++++++++
 arch/s390/kernel/relocate_kernel64.S | 82 ++++++++++++++++++++++++++++++
 arch/s390/kernel/syscalls.S          |  2 +-
 include/asm-s390/kexec.h             | 42 ++++++++++++++++
 include/asm-s390/unistd.h            |  2 +-
 include/linux/kexec.h                |  1 +
 11 files changed, 349 insertions(+), 2 deletions(-)
 create mode 100644 arch/s390/kernel/crash.c
 create mode 100644 arch/s390/kernel/machine_kexec.c
 create mode 100644 arch/s390/kernel/relocate_kernel.S
 create mode 100644 arch/s390/kernel/relocate_kernel64.S
 create mode 100644 include/asm-s390/kexec.h

(limited to 'include')

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 32696c1d928..6600ee87f89 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -455,6 +455,14 @@ config NO_IDLE_HZ_INIT
 	  The HZ timer is switched off in idle by default. That means the
 	  HZ timer is already disabled at boot time.
 
+config KEXEC
+	bool "kexec system call (EXPERIMENTAL)"
+	depends on EXPERIMENTAL
+	help
+	  kexec is a system call that implements the ability to shutdown your
+	  current kernel, and to start another kernel.  It is like a reboot
+	  but is independent of hardware/microcode support.
+
 endmenu
 
 config PCMCIA
diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile
index b41e0e199a7..ab1e49d2e51 100644
--- a/arch/s390/kernel/Makefile
+++ b/arch/s390/kernel/Makefile
@@ -25,6 +25,16 @@ obj-$(CONFIG_ARCH_S390X)	+= entry64.o reipl64.o
 
 obj-$(CONFIG_VIRT_TIMER)	+= vtime.o
 
+# Kexec part
+S390_KEXEC_OBJS := machine_kexec.o crash.o
+ifeq ($(CONFIG_ARCH_S390X),y)
+S390_KEXEC_OBJS += relocate_kernel64.o
+else
+S390_KEXEC_OBJS += relocate_kernel.o
+endif
+obj-$(CONFIG_KEXEC) += $(S390_KEXEC_OBJS)
+
+
 #
 # This is just to get the dependencies...
 #
diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S
index 7a607b1d038..bf529739c8a 100644
--- a/arch/s390/kernel/compat_wrapper.S
+++ b/arch/s390/kernel/compat_wrapper.S
@@ -1441,3 +1441,11 @@ compat_sys_waitid_wrapper:
 	lgfr	%r5,%r5			# int
 	llgtr	%r6,%r6			# struct rusage_emu31 *
 	jg	compat_sys_waitid
+
+	.globl	compat_sys_kexec_load_wrapper
+compat_sys_kexec_load_wrapper:
+	llgfr	%r2,%r2			# unsigned long
+	llgfr	%r3,%r3			# unsigned long
+	llgtr	%r4,%r4			# struct kexec_segment *
+	llgfr	%r5,%r5			# unsigned long
+	jg	compat_sys_kexec_load
diff --git a/arch/s390/kernel/crash.c b/arch/s390/kernel/crash.c
new file mode 100644
index 00000000000..db38283c1f2
--- /dev/null
+++ b/arch/s390/kernel/crash.c
@@ -0,0 +1,17 @@
+/*
+ * arch/s390/kernel/crash.c
+ *
+ * (C) Copyright IBM Corp. 2005
+ *
+ * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>
+ *
+ */
+
+#include <linux/threads.h>
+#include <linux/kexec.h>
+
+note_buf_t crash_notes[NR_CPUS];
+
+void machine_crash_shutdown(void)
+{
+}
diff --git a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c
new file mode 100644
index 00000000000..7a94db76df4
--- /dev/null
+++ b/arch/s390/kernel/machine_kexec.c
@@ -0,0 +1,98 @@
+/*
+ * arch/s390/kernel/machine_kexec.c
+ *
+ * (C) Copyright IBM Corp. 2005
+ *
+ * Author(s): Rolf Adelsberger <adelsberger@de.ibm.com>
+ *
+ */
+
+/*
+ * s390_machine_kexec.c - handle the transition of Linux booting another kernel
+ * on the S390 architecture.
+ */
+
+#include <asm/cio.h>
+#include <asm/setup.h>
+#include <linux/device.h>
+#include <linux/mm.h>
+#include <linux/kexec.h>
+#include <linux/delay.h>
+#include <asm/pgtable.h>
+#include <asm/pgalloc.h>
+#include <asm/system.h>
+
+static void kexec_halt_all_cpus(void *);
+
+typedef void (*relocate_kernel_t) (kimage_entry_t *, unsigned long);
+
+const extern unsigned char relocate_kernel[];
+const extern unsigned long long relocate_kernel_len;
+
+int
+machine_kexec_prepare(struct kimage *image)
+{
+	unsigned long reboot_code_buffer;
+
+	/* We don't support anything but the default image type for now. */
+	if (image->type != KEXEC_TYPE_DEFAULT)
+		return -EINVAL;
+
+	/* Get the destination where the assembler code should be copied to.*/
+	reboot_code_buffer = page_to_pfn(image->control_code_page)<<PAGE_SHIFT;
+
+	/* Then copy it */
+	memcpy((void *) reboot_code_buffer, relocate_kernel,
+	       relocate_kernel_len);
+	return 0;
+}
+
+void
+machine_kexec_cleanup(struct kimage *image)
+{
+}
+
+void
+machine_shutdown(void)
+{
+	printk(KERN_INFO "kexec: machine_shutdown called\n");
+}
+
+NORET_TYPE void
+machine_kexec(struct kimage *image)
+{
+	clear_all_subchannels();
+
+	/* Disable lowcore protection */
+	ctl_clear_bit(0,28);
+
+	on_each_cpu(kexec_halt_all_cpus, image, 0, 0);
+	for(;;);
+}
+
+static void
+kexec_halt_all_cpus(void *kernel_image)
+{
+	static atomic_t cpuid = ATOMIC_INIT(-1);
+	int cpu;
+	struct kimage *image;
+	relocate_kernel_t data_mover;
+
+	if (atomic_compare_and_swap(-1, smp_processor_id(), &cpuid))
+		signal_processor(smp_processor_id(), sigp_stop);
+
+	/* Wait for all other cpus to enter stopped state */
+	for_each_online_cpu(cpu) {
+		if (cpu == smp_processor_id())
+			continue;
+		while(!smp_cpu_not_running(cpu))
+			cpu_relax();
+	}
+
+	image = (struct kimage *) kernel_image;
+	data_mover = (relocate_kernel_t)
+		(page_to_pfn(image->control_code_page) << PAGE_SHIFT);
+
+	/* Call the moving routine */
+	(*data_mover) (&image->head, image->start);
+}
diff --git a/arch/s390/kernel/relocate_kernel.S b/arch/s390/kernel/relocate_kernel.S
new file mode 100644
index 00000000000..d5e4a62fbb7
--- /dev/null
+++ b/arch/s390/kernel/relocate_kernel.S
@@ -0,0 +1,81 @@
+/*
+ * arch/s390/kernel/relocate_kernel.S
+ *
+ * (C) Copyright IBM Corp. 2005
+ *
+ * Author(s): Rolf Adelsberger <adelsberger@de.ibm.com>
+ *
+ */
+
+/*
+ * moves the new kernel to its destination...
+ * %r2 = pointer to first kimage_entry_t
+ * %r3 = start address - where to jump to after the job is done...
+ *
+ * %r5 will be used as temp. storage
+ * %r6 holds the destination address
+ * %r7 = PAGE_SIZE
+ * %r8 holds the source address
+ * %r9 = PAGE_SIZE
+ * %r10 is a page mask
+ */
+
+	.text
+	.globl		relocate_kernel
+	relocate_kernel:
+		basr	%r13,0		#base address
+	.base:
+		spx	zero64-.base(%r13)	#absolute addressing mode
+		stnsm	sys_msk-.base(%r13),0xf8	#disable DAT and IRQ (external)
+		lhi	%r10,-1		#preparing the mask
+		sll	%r10,12		#shift it such that it becomes 0xf000
+	.top:
+		lhi	%r7,4096	#load PAGE_SIZE in r7
+		lhi	%r9,4096	#load PAGE_SIZE in r9
+		l	%r5,0(%r2)	#read another word for indirection page
+		ahi	%r2,4		#increment pointer
+		tml	%r5,0x1		#is it a destination page?
+		je	.indir_check	#NO, goto "indir_check"
+		lr	%r6,%r5		#r6 = r5
+		nr	%r6,%r10	#mask it out and...
+		j	.top		#...next iteration
+	.indir_check:
+		tml	%r5,0x2		#is it a indirection page?
+		je	.done_test	#NO, goto "done_test"
+		nr	%r5,%r10	#YES, mask out,
+		lr	%r2,%r5		#move it into the right register,
+		j	.top		#and read next...
+	.done_test:
+		tml	%r5,0x4		#is it the done indicator?
+		je	.source_test	#NO! Well, then it should be the source indicator...
+		j	.done		#ok, lets finish it here...
+	.source_test:
+		tml	%r5,0x8		#it should be a source indicator...
+		je	.top		#NO, ignore it...
+		lr	%r8,%r5		#r8 = r5
+		nr	%r8,%r10	#masking
+	0:	mvcle	%r6,%r8,0x0	#copy PAGE_SIZE bytes from r8 to r6 - pad with 0
+		jo	0b
+		j	.top
+	.done:
+		sr	%r0,%r0		#clear register r0
+		la	%r4,load_psw-.base(%r13)	#load psw-address into the register
+		o	%r3,4(%r4)	#or load address into psw
+		st	%r3,4(%r4)
+		mvc	0(8,%r0),0(%r4)	#copy psw to absolute address 0
+		sr	%r1,%r1		#clear %r1
+		sr	%r2,%r2		#clear %r2
+		sigp	%r1,%r2,0x12	#set cpuid to zero
+		lpsw	0		#hopefully start new kernel...
+
+		.align	8
+	zero64:
+		.quad	0
+	load_psw:
+		.long	0x00080000,0x80000000
+	sys_msk:
+		.quad	0
+	relocate_kernel_end:
+	.globl	relocate_kernel_len
+	relocate_kernel_len:
+		.quad	relocate_kernel_end - relocate_kernel
diff --git a/arch/s390/kernel/relocate_kernel64.S b/arch/s390/kernel/relocate_kernel64.S
new file mode 100644
index 00000000000..96290cc4eb3
--- /dev/null
+++ b/arch/s390/kernel/relocate_kernel64.S
@@ -0,0 +1,82 @@
+/*
+ * arch/s390/kernel/relocate_kernel64.S
+ *
+ * (C) Copyright IBM Corp. 2005
+ *
+ * Author(s): Rolf Adelsberger <adelsberger@de.ibm.com>
+ *
+ */
+
+/*
+ * moves the new kernel to its destination...
+ * %r2 = pointer to first kimage_entry_t
+ * %r3 = start address - where to jump to after the job is done...
+ *
+ * %r5 will be used as temp. storage
+ * %r6 holds the destination address
+ * %r7 = PAGE_SIZE
+ * %r8 holds the source address
+ * %r9 = PAGE_SIZE
+ *
+ * 0xf000 is a page_mask
+ */
+
+	.text
+	.globl		relocate_kernel
+	relocate_kernel:
+		basr	%r13,0		#base address
+	.base:
+		spx	zero64-.base(%r13)	#absolute addressing mode
+		stnsm	sys_msk-.base(%r13),0xf8	#disable DAT and IRQ (external)
+	.top:
+		lghi	%r7,4096	#load PAGE_SIZE in r7
+		lghi	%r9,4096	#load PAGE_SIZE in r9
+		lg	%r5,0(%r2)	#read another word for indirection page
+		aghi	%r2,8		#increment pointer
+		tml	%r5,0x1		#is it a destination page?
+		je	.indir_check	#NO, goto "indir_check"
+		lgr	%r6,%r5		#r6 = r5
+		nill	%r6,0xf000	#mask it out and...
+		j	.top		#...next iteration
+	.indir_check:
+		tml     %r5,0x2		#is it a indirection page?
+		je      .done_test	#NO, goto "done_test"
+		nill    %r5,0xf000	#YES, mask out,
+		lgr     %r2,%r5		#move it into the right register,
+		j       .top		#and read next...
+	.done_test:
+		tml     %r5,0x4		#is it the done indicator?
+		je      .source_test	#NO! Well, then it should be the source indicator...
+		j       .done		#ok, lets finish it here...
+	.source_test:
+		tml     %r5,0x8		#it should be a source indicator...
+		je      .top		#NO, ignore it...
+		lgr     %r8,%r5		#r8 = r5
+		nill    %r8,0xf000	#masking
+	0:	mvcle   %r6,%r8,0x0	#copy PAGE_SIZE bytes from r8 to r6 - pad with 0
+		jo	0b
+		j       .top
+	.done:
+		sgr     %r0,%r0		#clear register r0
+		la      %r4,load_psw-.base(%r13)	#load psw-address into the register
+		o	%r3,4(%r4)	#or load address into psw
+		st	%r3,4(%r4)
+		mvc     0(8,%r0),0(%r4)	#copy psw to absolute address 0
+		sam31			#31 bit mode
+		sr      %r1,%r1		#erase register r1
+		sr      %r2,%r2		#erase register r2
+		sigp    %r1,%r2,0x12	#set cpuid to zero
+		lpsw    0		#hopefully start new kernel...
+
+	        .align	8
+	zero64:
+		.quad	0
+	load_psw:
+		.long	0x00080000,0x80000000
+	sys_msk:
+		.quad	0
+	relocate_kernel_end:
+	.globl	relocate_kernel_len
+	relocate_kernel_len:
+		.quad	relocate_kernel_end - relocate_kernel
+
diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S
index 515938628f8..a8668afb5f8 100644
--- a/arch/s390/kernel/syscalls.S
+++ b/arch/s390/kernel/syscalls.S
@@ -285,7 +285,7 @@ SYSCALL(sys_mq_timedsend,sys_mq_timedsend,compat_sys_mq_timedsend_wrapper)
 SYSCALL(sys_mq_timedreceive,sys_mq_timedreceive,compat_sys_mq_timedreceive_wrapper)
 SYSCALL(sys_mq_notify,sys_mq_notify,compat_sys_mq_notify_wrapper) /* 275 */
 SYSCALL(sys_mq_getsetattr,sys_mq_getsetattr,compat_sys_mq_getsetattr_wrapper)
-NI_SYSCALL							/* reserved for kexec */
+SYSCALL(sys_kexec_load,sys_kexec_load,compat_sys_kexec_load_wrapper)
 SYSCALL(sys_add_key,sys_add_key,compat_sys_add_key_wrapper)
 SYSCALL(sys_request_key,sys_request_key,compat_sys_request_key_wrapper)
 SYSCALL(sys_keyctl,sys_keyctl,compat_sys_keyctl)		/* 280 */
diff --git a/include/asm-s390/kexec.h b/include/asm-s390/kexec.h
new file mode 100644
index 00000000000..54cf7d9f251
--- /dev/null
+++ b/include/asm-s390/kexec.h
@@ -0,0 +1,42 @@
+/*
+ * include/asm-s390/kexec.h
+ *
+ * (C) Copyright IBM Corp. 2005
+ *
+ * Author(s): Rolf Adelsberger <adelsberger@de.ibm.com>
+ *
+ */
+
+#ifndef _S390_KEXEC_H
+#define _S390_KEXEC_H
+
+#include <asm/page.h>
+#include <asm/processor.h>
+/*
+ * KEXEC_SOURCE_MEMORY_LIMIT maximum page get_free_page can return.
+ * I.e. Maximum page that is mapped directly into kernel memory,
+ * and kmap is not required.
+ */
+
+/* Maximum physical address we can use pages from */
+#define KEXEC_SOURCE_MEMORY_LIMIT (-1UL)
+
+/* Maximum address we can reach in physical address mode */
+#define KEXEC_DESTINATION_MEMORY_LIMIT (-1UL)
+
+/* Maximum address we can use for the control pages */
+/* Not more than 2GB */
+#define KEXEC_CONTROL_MEMORY_LIMIT (1<<31)
+
+/* Allocate one page for the pdp and the second for the code */
+#define KEXEC_CONTROL_CODE_SIZE 4096
+
+/* The native architecture */
+#define KEXEC_ARCH KEXEC_ARCH_S390
+
+#define MAX_NOTE_BYTES 1024
+typedef u32 note_buf_t[MAX_NOTE_BYTES/4];
+
+extern note_buf_t crash_notes[];
+
+#endif /*_S390_KEXEC_H */
diff --git a/include/asm-s390/unistd.h b/include/asm-s390/unistd.h
index f1a204f7c0f..363db45f8d0 100644
--- a/include/asm-s390/unistd.h
+++ b/include/asm-s390/unistd.h
@@ -269,7 +269,7 @@
 #define __NR_mq_timedreceive	274
 #define __NR_mq_notify		275
 #define __NR_mq_getsetattr	276
-/* Number 277 is reserved for new sys_kexec_load */
+#define __NR_kexec_load		277
 #define __NR_add_key		278
 #define __NR_request_key	279
 #define __NR_keyctl		280
diff --git a/include/linux/kexec.h b/include/linux/kexec.h
index e3fc35f4e35..0653a27c3d7 100644
--- a/include/linux/kexec.h
+++ b/include/linux/kexec.h
@@ -114,6 +114,7 @@ extern struct kimage *kexec_image;
 #define KEXEC_ARCH_PPC     (20 << 16)
 #define KEXEC_ARCH_PPC64   (21 << 16)
 #define KEXEC_ARCH_IA_64   (50 << 16)
+#define KEXEC_ARCH_S390    (22 << 16)
 
 #define KEXEC_FLAGS    (KEXEC_ON_CRASH)  /* List of defined/legal kexec flags */
 
-- 
cgit v1.2.3-70-g09d2


From 625f1c8219d95300ed32e4c67eb62a50ded095ba Mon Sep 17 00:00:00 2001
From: Vivek Goyal <vgoyal@in.ibm.com>
Date: Sat, 25 Jun 2005 14:58:12 -0700
Subject: [PATCH] Kdump: Export crash notes section address through sysfs

o Following patch exports kexec global variable "crash_notes" to user space
  through sysfs as kernel attribute in /sys/kernel.

Signed-off-by: Maneesh Soni <maneesh@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/crash.c   |  2 --
 arch/x86_64/kernel/crash.c |  3 ---
 include/asm-i386/kexec.h   |  5 +++++
 include/asm-x86_64/kexec.h |  5 +++++
 kernel/ksysfs.c            | 13 +++++++++++++
 5 files changed, 23 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c
index 59b92d21746..3645ad7ac20 100644
--- a/arch/i386/kernel/crash.c
+++ b/arch/i386/kernel/crash.c
@@ -26,8 +26,6 @@
 #include <asm/apic.h>
 #include <mach_ipi.h>
 
-#define MAX_NOTE_BYTES 1024
-typedef u32 note_buf_t[MAX_NOTE_BYTES/4];
 
 note_buf_t crash_notes[NR_CPUS];
 
diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c
index 7caf8a49d0c..6183bcb8525 100644
--- a/arch/x86_64/kernel/crash.c
+++ b/arch/x86_64/kernel/crash.c
@@ -20,9 +20,6 @@
 #include <asm/nmi.h>
 #include <asm/hw_irq.h>
 
-#define MAX_NOTE_BYTES 1024
-typedef u32 note_buf_t[MAX_NOTE_BYTES/4];
-
 note_buf_t crash_notes[NR_CPUS];
 
 void machine_crash_shutdown(void)
diff --git a/include/asm-i386/kexec.h b/include/asm-i386/kexec.h
index a1599b55d62..6ed2a03e37b 100644
--- a/include/asm-i386/kexec.h
+++ b/include/asm-i386/kexec.h
@@ -25,4 +25,9 @@
 /* The native architecture */
 #define KEXEC_ARCH KEXEC_ARCH_386
 
+#define MAX_NOTE_BYTES 1024
+typedef u32 note_buf_t[MAX_NOTE_BYTES/4];
+
+extern note_buf_t crash_notes[];
+
 #endif /* _I386_KEXEC_H */
diff --git a/include/asm-x86_64/kexec.h b/include/asm-x86_64/kexec.h
index dc33646dc7d..42d2ff15c59 100644
--- a/include/asm-x86_64/kexec.h
+++ b/include/asm-x86_64/kexec.h
@@ -25,4 +25,9 @@
 /* The native architecture */
 #define KEXEC_ARCH KEXEC_ARCH_X86_64
 
+#define MAX_NOTE_BYTES 1024
+typedef u32 note_buf_t[MAX_NOTE_BYTES/4];
+
+extern note_buf_t crash_notes[];
+
 #endif /* _X86_64_KEXEC_H */
diff --git a/kernel/ksysfs.c b/kernel/ksysfs.c
index 1f064a63f8c..015fb69ad94 100644
--- a/kernel/ksysfs.c
+++ b/kernel/ksysfs.c
@@ -30,12 +30,25 @@ static ssize_t hotplug_seqnum_show(struct subsystem *subsys, char *page)
 KERNEL_ATTR_RO(hotplug_seqnum);
 #endif
 
+#ifdef CONFIG_KEXEC
+#include <asm/kexec.h>
+
+static ssize_t crash_notes_show(struct subsystem *subsys, char *page)
+{
+	return sprintf(page, "%p\n", (void *)crash_notes);
+}
+KERNEL_ATTR_RO(crash_notes);
+#endif
+
 decl_subsys(kernel, NULL, NULL);
 EXPORT_SYMBOL_GPL(kernel_subsys);
 
 static struct attribute * kernel_attrs[] = {
 #ifdef CONFIG_HOTPLUG
 	&hotplug_seqnum_attr.attr,
+#endif
+#ifdef CONFIG_KEXEC
+	&crash_notes_attr.attr,
 #endif
 	NULL
 };
-- 
cgit v1.2.3-70-g09d2


From 92aa63a5a1bf2e7b0c79e6716d24b76dbbdcf951 Mon Sep 17 00:00:00 2001
From: Vivek Goyal <vgoyal@in.ibm.com>
Date: Sat, 25 Jun 2005 14:58:18 -0700
Subject: [PATCH] kdump: Retrieve saved max pfn

This patch retrieves the max_pfn being used by previous kernel and stores it
in a safe location (saved_max_pfn) before it is overwritten due to user
defined memory map.  This pfn is used to make sure that user does not try to
read the physical memory beyond saved_max_pfn.

Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/setup.c | 12 ++++++++++++
 include/linux/bootmem.h  |  4 ++++
 mm/bootmem.c             |  8 ++++++++
 3 files changed, 24 insertions(+)

(limited to 'include')

diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index d88ebdfa6cc..8d58a053e12 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -59,6 +59,9 @@
 #include "setup_arch_pre.h"
 #include <bios_ebda.h>
 
+/* Forward Declaration. */
+void __init find_max_pfn(void);
+
 /* This value is set up by the early boot code to point to the value
    immediately after the boot time page tables.  It contains a *physical*
    address, and must not be in the .bss segment! */
@@ -736,6 +739,15 @@ static void __init parse_cmdline_early (char ** cmdline_p)
 			if (to != command_line)
 				to--;
 			if (!memcmp(from+7, "exactmap", 8)) {
+#ifdef CONFIG_CRASH_DUMP
+				/* If we are doing a crash dump, we
+				 * still need to know the real mem
+				 * size before original memory map is
+				 * reset.
+				 */
+				find_max_pfn();
+				saved_max_pfn = max_pfn;
+#endif
 				from += 8+7;
 				e820.nr_map = 0;
 				userdef = 1;
diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index 500f451ce0c..82bd8842d11 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -22,6 +22,10 @@ extern unsigned long min_low_pfn;
  */
 extern unsigned long max_pfn;
 
+#ifdef CONFIG_CRASH_DUMP
+extern unsigned long saved_max_pfn;
+#endif
+
 /*
  * node_bootmem_map is a map pointer - the bits represent all physical 
  * memory pages (including holes) on the node.
diff --git a/mm/bootmem.c b/mm/bootmem.c
index f82f7aebbee..45275f1f894 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -33,6 +33,14 @@ EXPORT_SYMBOL(max_pfn);		/* This is exported so
 				 * dma_get_required_mask(), which uses
 				 * it, can be an inline function */
 
+#ifdef CONFIG_CRASH_DUMP
+/*
+ * If we have booted due to a crash, max_pfn will be a very low value. We need
+ * to know the amount of memory that the previous kernel used.
+ */
+unsigned long saved_max_pfn;
+#endif
+
 /* return the number of _pages_ that will be allocated for the boot bitmap */
 unsigned long __init bootmem_bootmap_pages (unsigned long pages)
 {
-- 
cgit v1.2.3-70-g09d2


From 60e64d46a58236e3c718074372cab6a5b56a3b15 Mon Sep 17 00:00:00 2001
From: Vivek Goyal <vgoyal@in.ibm.com>
Date: Sat, 25 Jun 2005 14:58:19 -0700
Subject: [PATCH] kdump: Routines for copying dump pages

This patch provides the interfaces necessary to read the dump contents,
treating it as a high memory device.

Signed off by Hariprasad Nellitheertha <hari@in.ibm.com>
Signed-off-by: Eric Biederman <ebiederm@xmission.com>
Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mm/highmem.c     | 18 +++++++++++++++++
 include/asm-i386/highmem.h |  1 +
 include/linux/crash_dump.h | 13 ++++++++++++
 include/linux/highmem.h    |  1 +
 kernel/Makefile            |  1 +
 kernel/crash_dump.c        | 49 ++++++++++++++++++++++++++++++++++++++++++++++
 6 files changed, 83 insertions(+)
 create mode 100644 include/linux/crash_dump.h
 create mode 100644 kernel/crash_dump.c

(limited to 'include')

diff --git a/arch/i386/mm/highmem.c b/arch/i386/mm/highmem.c
index 4b7aaf99d7e..b6eb4dcb877 100644
--- a/arch/i386/mm/highmem.c
+++ b/arch/i386/mm/highmem.c
@@ -75,6 +75,24 @@ void kunmap_atomic(void *kvaddr, enum km_type type)
 	preempt_check_resched();
 }
 
+/* This is the same as kmap_atomic() but can map memory that doesn't
+ * have a struct page associated with it.
+ */
+void *kmap_atomic_pfn(unsigned long pfn, enum km_type type)
+{
+	enum fixed_addresses idx;
+	unsigned long vaddr;
+
+	inc_preempt_count();
+
+	idx = type + KM_TYPE_NR*smp_processor_id();
+	vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
+	set_pte(kmap_pte-idx, pfn_pte(pfn, kmap_prot));
+	__flush_tlb_one(vaddr);
+
+	return (void*) vaddr;
+}
+
 struct page *kmap_atomic_to_page(void *ptr)
 {
 	unsigned long idx, vaddr = (unsigned long)ptr;
diff --git a/include/asm-i386/highmem.h b/include/asm-i386/highmem.h
index 1df42bf347d..0fd331306b6 100644
--- a/include/asm-i386/highmem.h
+++ b/include/asm-i386/highmem.h
@@ -70,6 +70,7 @@ void *kmap(struct page *page);
 void kunmap(struct page *page);
 void *kmap_atomic(struct page *page, enum km_type type);
 void kunmap_atomic(void *kvaddr, enum km_type type);
+void *kmap_atomic_pfn(unsigned long pfn, enum km_type type);
 struct page *kmap_atomic_to_page(void *ptr);
 
 #define flush_cache_kmaps()	do { } while (0)
diff --git a/include/linux/crash_dump.h b/include/linux/crash_dump.h
new file mode 100644
index 00000000000..7d983b81742
--- /dev/null
+++ b/include/linux/crash_dump.h
@@ -0,0 +1,13 @@
+#ifndef LINUX_CRASH_DUMP_H
+#define LINUX_CRASH_DUMP_H
+
+#ifdef CONFIG_CRASH_DUMP
+#include <linux/kexec.h>
+#include <linux/smp_lock.h>
+#include <linux/device.h>
+#include <linux/proc_fs.h>
+
+extern ssize_t copy_oldmem_page(unsigned long, char *, size_t,
+						unsigned long, int);
+#endif /* CONFIG_CRASH_DUMP */
+#endif /* LINUX_CRASHDUMP_H */
diff --git a/include/linux/highmem.h b/include/linux/highmem.h
index 2a7e6c65c88..6bece9280eb 100644
--- a/include/linux/highmem.h
+++ b/include/linux/highmem.h
@@ -28,6 +28,7 @@ static inline void *kmap(struct page *page)
 
 #define kmap_atomic(page, idx)		page_address(page)
 #define kunmap_atomic(addr, idx)	do { } while (0)
+#define kmap_atomic_pfn(pfn, idx)	page_address(pfn_to_page(pfn))
 #define kmap_atomic_to_page(ptr)	virt_to_page(ptr)
 
 #endif /* CONFIG_HIGHMEM */
diff --git a/kernel/Makefile b/kernel/Makefile
index cfc8b0dea95..cb05cd05d23 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -28,6 +28,7 @@ obj-$(CONFIG_AUDITSYSCALL) += auditsc.o
 obj-$(CONFIG_KPROBES) += kprobes.o
 obj-$(CONFIG_SYSFS) += ksysfs.o
 obj-$(CONFIG_GENERIC_HARDIRQS) += irq/
+obj-$(CONFIG_CRASH_DUMP) += crash_dump.o
 obj-$(CONFIG_SECCOMP) += seccomp.o
 
 ifneq ($(CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER),y)
diff --git a/kernel/crash_dump.c b/kernel/crash_dump.c
new file mode 100644
index 00000000000..5a1e6d5d203
--- /dev/null
+++ b/kernel/crash_dump.c
@@ -0,0 +1,49 @@
+/*
+ *	kernel/crash_dump.c - Memory preserving reboot related code.
+ *
+ *	Created by: Hariprasad Nellitheertha (hari@in.ibm.com)
+ *	Copyright (C) IBM Corporation, 2004. All rights reserved
+ */
+
+#include <linux/smp_lock.h>
+#include <linux/errno.h>
+#include <linux/proc_fs.h>
+#include <linux/bootmem.h>
+#include <linux/highmem.h>
+#include <linux/crash_dump.h>
+
+#include <asm/io.h>
+#include <asm/uaccess.h>
+
+/*
+ * Copy a page from "oldmem". For this page, there is no pte mapped
+ * in the current kernel. We stitch up a pte, similar to kmap_atomic.
+ */
+ssize_t copy_oldmem_page(unsigned long pfn, char *buf,
+				size_t csize, unsigned long offset, int userbuf)
+{
+	void *page, *vaddr;
+
+	if (!csize)
+		return 0;
+
+	page = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!page)
+		return -ENOMEM;
+
+	vaddr = kmap_atomic_pfn(pfn, KM_PTE0);
+	copy_page(page, vaddr);
+	kunmap_atomic(vaddr, KM_PTE0);
+
+	if (userbuf) {
+		if (copy_to_user(buf, (page + offset), csize)) {
+			kfree(page);
+			return -EFAULT;
+		}
+	} else {
+		memcpy(buf, (page + offset), csize);
+	}
+
+	kfree(page);
+	return csize;
+}
-- 
cgit v1.2.3-70-g09d2


From 2030eae52b416a9a9f0ffda74c982b7f1e19496d Mon Sep 17 00:00:00 2001
From: Vivek Goyal <vgoyal@in.ibm.com>
Date: Sat, 25 Jun 2005 14:58:20 -0700
Subject: [PATCH] Retrieve elfcorehdr address from command line

This patch adds support for retrieving the address of elf core header if one
is passed in command line.

Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/kernel-parameters.txt | 4 ++++
 arch/i386/kernel/setup.c            | 8 ++++++++
 include/linux/crash_dump.h          | 1 +
 kernel/crash_dump.c                 | 3 +++
 4 files changed, 16 insertions(+)

(limited to 'include')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 560ff5ae3fd..f44bb5567c5 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -451,6 +451,10 @@ running once the system is up.
 			Format: {"as"|"cfq"|"deadline"|"noop"}
 			See Documentation/block/as-iosched.txt
 			and Documentation/block/deadline-iosched.txt for details.
+	elfcorehdr=	[IA-32]
+			Specifies physical address of start of kernel core image
+			elf header.
+			See Documentation/kdump.txt for details.
 
 	enforcing	[SELINUX] Set initial enforcing status.
 			Format: {"0" | "1"}
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index 8d58a053e12..7306353c520 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -44,6 +44,7 @@
 #include <linux/edd.h>
 #include <linux/nodemask.h>
 #include <linux/kexec.h>
+#include <linux/crash_dump.h>
 
 #include <video/edid.h>
 
@@ -881,6 +882,13 @@ static void __init parse_cmdline_early (char ** cmdline_p)
 			}
 		}
 #endif
+#ifdef CONFIG_CRASH_DUMP
+		/* elfcorehdr= specifies the location of elf core header
+		 * stored by the crashed kernel.
+		 */
+		else if (!memcmp(from, "elfcorehdr=", 11))
+			elfcorehdr_addr = memparse(from+11, &from);
+#endif
 
 		/*
 		 * highmem=size forces highmem to be exactly 'size' bytes.
diff --git a/include/linux/crash_dump.h b/include/linux/crash_dump.h
index 7d983b81742..3f25fd1eaa4 100644
--- a/include/linux/crash_dump.h
+++ b/include/linux/crash_dump.h
@@ -7,6 +7,7 @@
 #include <linux/device.h>
 #include <linux/proc_fs.h>
 
+extern unsigned long long elfcorehdr_addr;
 extern ssize_t copy_oldmem_page(unsigned long, char *, size_t,
 						unsigned long, int);
 #endif /* CONFIG_CRASH_DUMP */
diff --git a/kernel/crash_dump.c b/kernel/crash_dump.c
index 5a1e6d5d203..10b966c3744 100644
--- a/kernel/crash_dump.c
+++ b/kernel/crash_dump.c
@@ -15,6 +15,9 @@
 #include <asm/io.h>
 #include <asm/uaccess.h>
 
+/* Stores the physical address of elf header of crash image. */
+unsigned long long elfcorehdr_addr;
+
 /*
  * Copy a page from "oldmem". For this page, there is no pte mapped
  * in the current kernel. We stitch up a pte, similar to kmap_atomic.
-- 
cgit v1.2.3-70-g09d2


From 666bfddbe8b8fd4fd44617d6c55193d5ac7edb29 Mon Sep 17 00:00:00 2001
From: Vivek Goyal <vgoyal@in.ibm.com>
Date: Sat, 25 Jun 2005 14:58:21 -0700
Subject: [PATCH] kdump: Access dump file in elf format (/proc/vmcore)

From: "Vivek Goyal" <vgoyal@in.ibm.com>

o Support for /proc/vmcore interface. This interface exports elf core image
  either in ELF32 or ELF64 format, depending on the format in which elf headers
  have been stored by crashed kernel.
o Added support for CONFIG_VMCORE config option.
o Removed the dependency on /proc/kcore.

From: "Eric W. Biederman" <ebiederm@xmission.com>

This patch has been refactored to more closely match the prevailing style in
the affected files.  And to clearly indicate the dependency between
/proc/kcore and proc/vmcore.c

From: Hariprasad Nellitheertha <hari@in.ibm.com>

This patch contains the code that provides an ELF format interface to the
previous kernel's memory post kexec reboot.

Signed off by Hariprasad Nellitheertha <hari@in.ibm.com>
Signed-off-by: Eric Biederman <ebiederm@xmission.com>
Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/Kconfig                 |   6 +
 fs/proc/Makefile           |   1 +
 fs/proc/proc_misc.c        |   6 +
 fs/proc/vmcore.c           | 451 +++++++++++++++++++++++++++++++++++++++++++++
 include/linux/crash_dump.h |   4 +
 include/linux/proc_fs.h    |   7 +
 kernel/crash_dump.c        |   2 +-
 7 files changed, 476 insertions(+), 1 deletion(-)
 create mode 100644 fs/proc/vmcore.c

(limited to 'include')

diff --git a/fs/Kconfig b/fs/Kconfig
index 8157f2e2d51..06217795623 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -734,6 +734,12 @@ config PROC_KCORE
 	bool "/proc/kcore support" if !ARM
 	depends on PROC_FS && MMU
 
+config PROC_VMCORE
+        bool "/proc/vmcore support (EXPERIMENTAL)"
+        depends on PROC_FS && EMBEDDED && EXPERIMENTAL && CRASH_DUMP
+        help
+        Exports the dump image of crashed kernel in ELF format.
+
 config SYSFS
 	bool "sysfs file system support" if EMBEDDED
 	default y
diff --git a/fs/proc/Makefile b/fs/proc/Makefile
index 738b9b60293..7431d7ba2d0 100644
--- a/fs/proc/Makefile
+++ b/fs/proc/Makefile
@@ -11,4 +11,5 @@ proc-y       += inode.o root.o base.o generic.o array.o \
 		kmsg.o proc_tty.o proc_misc.o
 
 proc-$(CONFIG_PROC_KCORE)	+= kcore.o
+proc-$(CONFIG_PROC_VMCORE)	+= vmcore.o
 proc-$(CONFIG_PROC_DEVICETREE)	+= proc_devtree.o
diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c
index 94b570ad037..a3453555a94 100644
--- a/fs/proc/proc_misc.c
+++ b/fs/proc/proc_misc.c
@@ -44,6 +44,7 @@
 #include <linux/jiffies.h>
 #include <linux/sysrq.h>
 #include <linux/vmalloc.h>
+#include <linux/crash_dump.h>
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 #include <asm/io.h>
@@ -618,6 +619,11 @@ void __init proc_misc_init(void)
 				(size_t)high_memory - PAGE_OFFSET + PAGE_SIZE;
 	}
 #endif
+#ifdef CONFIG_PROC_VMCORE
+	proc_vmcore = create_proc_entry("vmcore", S_IRUSR, NULL);
+	if (proc_vmcore)
+		proc_vmcore->proc_fops = &proc_vmcore_operations;
+#endif
 #ifdef CONFIG_MAGIC_SYSRQ
 	entry = create_proc_entry("sysrq-trigger", S_IWUSR, NULL);
 	if (entry)
diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c
new file mode 100644
index 00000000000..8ad46785584
--- /dev/null
+++ b/fs/proc/vmcore.c
@@ -0,0 +1,451 @@
+/*
+ *	fs/proc/vmcore.c Interface for accessing the crash
+ * 				 dump from the system's previous life.
+ * 	Heavily borrowed from fs/proc/kcore.c
+ *	Created by: Hariprasad Nellitheertha (hari@in.ibm.com)
+ *	Copyright (C) IBM Corporation, 2004. All rights reserved
+ *
+ */
+
+#include <linux/config.h>
+#include <linux/mm.h>
+#include <linux/proc_fs.h>
+#include <linux/user.h>
+#include <linux/a.out.h>
+#include <linux/elf.h>
+#include <linux/elfcore.h>
+#include <linux/proc_fs.h>
+#include <linux/highmem.h>
+#include <linux/bootmem.h>
+#include <linux/init.h>
+#include <linux/crash_dump.h>
+#include <linux/list.h>
+#include <asm/uaccess.h>
+#include <asm/io.h>
+
+/* List representing chunks of contiguous memory areas and their offsets in
+ * vmcore file.
+ */
+static LIST_HEAD(vmcore_list);
+
+/* Stores the pointer to the buffer containing kernel elf core headers. */
+static char *elfcorebuf;
+static size_t elfcorebuf_sz;
+
+/* Total size of vmcore file. */
+static u64 vmcore_size;
+
+struct proc_dir_entry *proc_vmcore = NULL;
+
+/* Reads a page from the oldmem device from given offset. */
+static ssize_t read_from_oldmem(char *buf, size_t count,
+			     loff_t *ppos, int userbuf)
+{
+	unsigned long pfn, offset;
+	size_t nr_bytes;
+	ssize_t read = 0, tmp;
+
+	if (!count)
+		return 0;
+
+	offset = (unsigned long)(*ppos % PAGE_SIZE);
+	pfn = (unsigned long)(*ppos / PAGE_SIZE);
+	if (pfn > saved_max_pfn)
+		return -EINVAL;
+
+	do {
+		if (count > (PAGE_SIZE - offset))
+			nr_bytes = PAGE_SIZE - offset;
+		else
+			nr_bytes = count;
+
+		tmp = copy_oldmem_page(pfn, buf, nr_bytes, offset, userbuf);
+		if (tmp < 0)
+			return tmp;
+		*ppos += nr_bytes;
+		count -= nr_bytes;
+		buf += nr_bytes;
+		read += nr_bytes;
+		++pfn;
+		offset = 0;
+	} while (count);
+
+	return read;
+}
+
+/* Maps vmcore file offset to respective physical address in memroy. */
+static u64 map_offset_to_paddr(loff_t offset, struct list_head *vc_list,
+					struct vmcore **m_ptr)
+{
+	struct vmcore *m;
+	u64 paddr;
+
+	list_for_each_entry(m, vc_list, list) {
+		u64 start, end;
+		start = m->offset;
+		end = m->offset + m->size - 1;
+		if (offset >= start && offset <= end) {
+			paddr = m->paddr + offset - start;
+			*m_ptr = m;
+			return paddr;
+		}
+	}
+	*m_ptr = NULL;
+	return 0;
+}
+
+/* Read from the ELF header and then the crash dump. On error, negative value is
+ * returned otherwise number of bytes read are returned.
+ */
+static ssize_t read_vmcore(struct file *file, char __user *buffer,
+				size_t buflen, loff_t *fpos)
+{
+	ssize_t acc = 0, tmp;
+	size_t tsz, nr_bytes;
+	u64 start;
+	struct vmcore *curr_m = NULL;
+
+	if (buflen == 0 || *fpos >= vmcore_size)
+		return 0;
+
+	/* trim buflen to not go beyond EOF */
+	if (buflen > vmcore_size - *fpos)
+		buflen = vmcore_size - *fpos;
+
+	/* Read ELF core header */
+	if (*fpos < elfcorebuf_sz) {
+		tsz = elfcorebuf_sz - *fpos;
+		if (buflen < tsz)
+			tsz = buflen;
+		if (copy_to_user(buffer, elfcorebuf + *fpos, tsz))
+			return -EFAULT;
+		buflen -= tsz;
+		*fpos += tsz;
+		buffer += tsz;
+		acc += tsz;
+
+		/* leave now if filled buffer already */
+		if (buflen == 0)
+			return acc;
+	}
+
+	start = map_offset_to_paddr(*fpos, &vmcore_list, &curr_m);
+	if (!curr_m)
+        	return -EINVAL;
+	if ((tsz = (PAGE_SIZE - (start & ~PAGE_MASK))) > buflen)
+		tsz = buflen;
+
+	/* Calculate left bytes in current memory segment. */
+	nr_bytes = (curr_m->size - (start - curr_m->paddr));
+	if (tsz > nr_bytes)
+		tsz = nr_bytes;
+
+	while (buflen) {
+		tmp = read_from_oldmem(buffer, tsz, &start, 1);
+		if (tmp < 0)
+			return tmp;
+		buflen -= tsz;
+		*fpos += tsz;
+		buffer += tsz;
+		acc += tsz;
+		if (start >= (curr_m->paddr + curr_m->size)) {
+			if (curr_m->list.next == &vmcore_list)
+				return acc;	/*EOF*/
+			curr_m = list_entry(curr_m->list.next,
+						struct vmcore, list);
+			start = curr_m->paddr;
+		}
+		if ((tsz = (PAGE_SIZE - (start & ~PAGE_MASK))) > buflen)
+			tsz = buflen;
+		/* Calculate left bytes in current memory segment. */
+		nr_bytes = (curr_m->size - (start - curr_m->paddr));
+		if (tsz > nr_bytes)
+			tsz = nr_bytes;
+	}
+	return acc;
+}
+
+static int open_vmcore(struct inode *inode, struct file *filp)
+{
+	return 0;
+}
+
+struct file_operations proc_vmcore_operations = {
+	.read		= read_vmcore,
+	.open		= open_vmcore,
+};
+
+static struct vmcore* __init get_new_element(void)
+{
+	struct vmcore *p;
+
+	p = kmalloc(sizeof(*p), GFP_KERNEL);
+	if (p)
+		memset(p, 0, sizeof(*p));
+	return p;
+}
+
+static u64 __init get_vmcore_size_elf64(char *elfptr)
+{
+	int i;
+	u64 size;
+	Elf64_Ehdr *ehdr_ptr;
+	Elf64_Phdr *phdr_ptr;
+
+	ehdr_ptr = (Elf64_Ehdr *)elfptr;
+	phdr_ptr = (Elf64_Phdr*)(elfptr + sizeof(Elf64_Ehdr));
+	size = sizeof(Elf64_Ehdr) + ((ehdr_ptr->e_phnum) * sizeof(Elf64_Phdr));
+	for (i = 0; i < ehdr_ptr->e_phnum; i++) {
+		size += phdr_ptr->p_memsz;
+		phdr_ptr++;
+	}
+	return size;
+}
+
+/* Merges all the PT_NOTE headers into one. */
+static int __init merge_note_headers_elf64(char *elfptr, size_t *elfsz,
+						struct list_head *vc_list)
+{
+	int i, nr_ptnote=0, rc=0;
+	char *tmp;
+	Elf64_Ehdr *ehdr_ptr;
+	Elf64_Phdr phdr, *phdr_ptr;
+	Elf64_Nhdr *nhdr_ptr;
+	u64 phdr_sz = 0, note_off;
+
+	ehdr_ptr = (Elf64_Ehdr *)elfptr;
+	phdr_ptr = (Elf64_Phdr*)(elfptr + sizeof(Elf64_Ehdr));
+	for (i = 0; i < ehdr_ptr->e_phnum; i++, phdr_ptr++) {
+		int j;
+		void *notes_section;
+		struct vmcore *new;
+		u64 offset, max_sz, sz, real_sz = 0;
+		if (phdr_ptr->p_type != PT_NOTE)
+			continue;
+		nr_ptnote++;
+		max_sz = phdr_ptr->p_memsz;
+		offset = phdr_ptr->p_offset;
+		notes_section = kmalloc(max_sz, GFP_KERNEL);
+		if (!notes_section)
+			return -ENOMEM;
+		rc = read_from_oldmem(notes_section, max_sz, &offset, 0);
+		if (rc < 0) {
+			kfree(notes_section);
+			return rc;
+		}
+		nhdr_ptr = notes_section;
+		for (j = 0; j < max_sz; j += sz) {
+			if (nhdr_ptr->n_namesz == 0)
+				break;
+			sz = sizeof(Elf64_Nhdr) +
+				((nhdr_ptr->n_namesz + 3) & ~3) +
+				((nhdr_ptr->n_descsz + 3) & ~3);
+			real_sz += sz;
+			nhdr_ptr = (Elf64_Nhdr*)((char*)nhdr_ptr + sz);
+		}
+
+		/* Add this contiguous chunk of notes section to vmcore list.*/
+		new = get_new_element();
+		if (!new) {
+			kfree(notes_section);
+			return -ENOMEM;
+		}
+		new->paddr = phdr_ptr->p_offset;
+		new->size = real_sz;
+		list_add_tail(&new->list, vc_list);
+		phdr_sz += real_sz;
+		kfree(notes_section);
+	}
+
+	/* Prepare merged PT_NOTE program header. */
+	phdr.p_type    = PT_NOTE;
+	phdr.p_flags   = 0;
+	note_off = sizeof(Elf64_Ehdr) +
+			(ehdr_ptr->e_phnum - nr_ptnote +1) * sizeof(Elf64_Phdr);
+	phdr.p_offset  = note_off;
+	phdr.p_vaddr   = phdr.p_paddr = 0;
+	phdr.p_filesz  = phdr.p_memsz = phdr_sz;
+	phdr.p_align   = 0;
+
+	/* Add merged PT_NOTE program header*/
+	tmp = elfptr + sizeof(Elf64_Ehdr);
+	memcpy(tmp, &phdr, sizeof(phdr));
+	tmp += sizeof(phdr);
+
+	/* Remove unwanted PT_NOTE program headers. */
+	i = (nr_ptnote - 1) * sizeof(Elf64_Phdr);
+	*elfsz = *elfsz - i;
+	memmove(tmp, tmp+i, ((*elfsz)-sizeof(Elf64_Ehdr)-sizeof(Elf64_Phdr)));
+
+	/* Modify e_phnum to reflect merged headers. */
+	ehdr_ptr->e_phnum = ehdr_ptr->e_phnum - nr_ptnote + 1;
+
+	return 0;
+}
+
+/* Add memory chunks represented by program headers to vmcore list. Also update
+ * the new offset fields of exported program headers. */
+static int __init process_ptload_program_headers_elf64(char *elfptr,
+						size_t elfsz,
+						struct list_head *vc_list)
+{
+	int i;
+	Elf64_Ehdr *ehdr_ptr;
+	Elf64_Phdr *phdr_ptr;
+	loff_t vmcore_off;
+	struct vmcore *new;
+
+	ehdr_ptr = (Elf64_Ehdr *)elfptr;
+	phdr_ptr = (Elf64_Phdr*)(elfptr + sizeof(Elf64_Ehdr)); /* PT_NOTE hdr */
+
+	/* First program header is PT_NOTE header. */
+	vmcore_off = sizeof(Elf64_Ehdr) +
+			(ehdr_ptr->e_phnum) * sizeof(Elf64_Phdr) +
+			phdr_ptr->p_memsz; /* Note sections */
+
+	for (i = 0; i < ehdr_ptr->e_phnum; i++, phdr_ptr++) {
+		if (phdr_ptr->p_type != PT_LOAD)
+			continue;
+
+		/* Add this contiguous chunk of memory to vmcore list.*/
+		new = get_new_element();
+		if (!new)
+			return -ENOMEM;
+		new->paddr = phdr_ptr->p_offset;
+		new->size = phdr_ptr->p_memsz;
+		list_add_tail(&new->list, vc_list);
+
+		/* Update the program header offset. */
+		phdr_ptr->p_offset = vmcore_off;
+		vmcore_off = vmcore_off + phdr_ptr->p_memsz;
+	}
+	return 0;
+}
+
+/* Sets offset fields of vmcore elements. */
+static void __init set_vmcore_list_offsets_elf64(char *elfptr,
+						struct list_head *vc_list)
+{
+	loff_t vmcore_off;
+	Elf64_Ehdr *ehdr_ptr;
+	struct vmcore *m;
+
+	ehdr_ptr = (Elf64_Ehdr *)elfptr;
+
+	/* Skip Elf header and program headers. */
+	vmcore_off = sizeof(Elf64_Ehdr) +
+			(ehdr_ptr->e_phnum) * sizeof(Elf64_Phdr);
+
+	list_for_each_entry(m, vc_list, list) {
+		m->offset = vmcore_off;
+		vmcore_off += m->size;
+	}
+}
+
+static int __init parse_crash_elf64_headers(void)
+{
+	int rc=0;
+	Elf64_Ehdr ehdr;
+	u64 addr;
+
+	addr = elfcorehdr_addr;
+
+	/* Read Elf header */
+	rc = read_from_oldmem((char*)&ehdr, sizeof(Elf64_Ehdr), &addr, 0);
+	if (rc < 0)
+		return rc;
+
+	/* Do some basic Verification. */
+	if (memcmp(ehdr.e_ident, ELFMAG, SELFMAG) != 0 ||
+		(ehdr.e_type != ET_CORE) ||
+		!elf_check_arch(&ehdr) ||
+		ehdr.e_ident[EI_CLASS] != ELFCLASS64 ||
+		ehdr.e_ident[EI_VERSION] != EV_CURRENT ||
+		ehdr.e_version != EV_CURRENT ||
+		ehdr.e_ehsize != sizeof(Elf64_Ehdr) ||
+		ehdr.e_phentsize != sizeof(Elf64_Phdr) ||
+		ehdr.e_phnum == 0) {
+		printk(KERN_WARNING "Warning: Core image elf header is not"
+					"sane\n");
+		return -EINVAL;
+	}
+
+	/* Read in all elf headers. */
+	elfcorebuf_sz = sizeof(Elf64_Ehdr) + ehdr.e_phnum * sizeof(Elf64_Phdr);
+	elfcorebuf = kmalloc(elfcorebuf_sz, GFP_KERNEL);
+	if (!elfcorebuf)
+		return -ENOMEM;
+	addr = elfcorehdr_addr;
+	rc = read_from_oldmem(elfcorebuf, elfcorebuf_sz, &addr, 0);
+	if (rc < 0) {
+		kfree(elfcorebuf);
+		return rc;
+	}
+
+	/* Merge all PT_NOTE headers into one. */
+	rc = merge_note_headers_elf64(elfcorebuf, &elfcorebuf_sz, &vmcore_list);
+	if (rc) {
+		kfree(elfcorebuf);
+		return rc;
+	}
+	rc = process_ptload_program_headers_elf64(elfcorebuf, elfcorebuf_sz,
+							&vmcore_list);
+	if (rc) {
+		kfree(elfcorebuf);
+		return rc;
+	}
+	set_vmcore_list_offsets_elf64(elfcorebuf, &vmcore_list);
+	return 0;
+}
+
+static int __init parse_crash_elf_headers(void)
+{
+	unsigned char e_ident[EI_NIDENT];
+	u64 addr;
+	int rc=0;
+
+	addr = elfcorehdr_addr;
+	rc = read_from_oldmem(e_ident, EI_NIDENT, &addr, 0);
+	if (rc < 0)
+		return rc;
+	if (memcmp(e_ident, ELFMAG, SELFMAG) != 0) {
+		printk(KERN_WARNING "Warning: Core image elf header"
+					" not found\n");
+		return -EINVAL;
+	}
+
+	if (e_ident[EI_CLASS] == ELFCLASS64) {
+		rc = parse_crash_elf64_headers();
+		if (rc)
+			return rc;
+
+		/* Determine vmcore size. */
+		vmcore_size = get_vmcore_size_elf64(elfcorebuf);
+	} else {
+		printk(KERN_WARNING "Warning: Core image elf header is not"
+					" sane\n");
+		return -EINVAL;
+	}
+	return 0;
+}
+
+/* Init function for vmcore module. */
+static int __init vmcore_init(void)
+{
+	int rc = 0;
+
+	/* If elfcorehdr= has been passed in cmdline, then capture the dump.*/
+	if (!(elfcorehdr_addr < ELFCORE_ADDR_MAX))
+		return rc;
+	rc = parse_crash_elf_headers();
+	if (rc) {
+		printk(KERN_WARNING "Kdump: vmcore not initialized\n");
+		return rc;
+	}
+
+	/* Initialize /proc/vmcore size if proc is already up. */
+	if (proc_vmcore)
+		proc_vmcore->size = vmcore_size;
+	return 0;
+}
+module_init(vmcore_init)
diff --git a/include/linux/crash_dump.h b/include/linux/crash_dump.h
index 3f25fd1eaa4..534d750d922 100644
--- a/include/linux/crash_dump.h
+++ b/include/linux/crash_dump.h
@@ -7,8 +7,12 @@
 #include <linux/device.h>
 #include <linux/proc_fs.h>
 
+#define ELFCORE_ADDR_MAX	(-1ULL)
 extern unsigned long long elfcorehdr_addr;
 extern ssize_t copy_oldmem_page(unsigned long, char *, size_t,
 						unsigned long, int);
+extern struct file_operations proc_vmcore_operations;
+extern struct proc_dir_entry *proc_vmcore;
+
 #endif /* CONFIG_CRASH_DUMP */
 #endif /* LINUX_CRASHDUMP_H */
diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h
index 59e505261fd..0563581e3a0 100644
--- a/include/linux/proc_fs.h
+++ b/include/linux/proc_fs.h
@@ -74,6 +74,13 @@ struct kcore_list {
 	size_t size;
 };
 
+struct vmcore {
+	struct list_head list;
+	unsigned long long paddr;
+	unsigned long size;
+	loff_t offset;
+};
+
 #ifdef CONFIG_PROC_FS
 
 extern struct proc_dir_entry proc_root;
diff --git a/kernel/crash_dump.c b/kernel/crash_dump.c
index 10b966c3744..459ba49e376 100644
--- a/kernel/crash_dump.c
+++ b/kernel/crash_dump.c
@@ -16,7 +16,7 @@
 #include <asm/uaccess.h>
 
 /* Stores the physical address of elf header of crash image. */
-unsigned long long elfcorehdr_addr;
+unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX;
 
 /*
  * Copy a page from "oldmem". For this page, there is no pte mapped
-- 
cgit v1.2.3-70-g09d2


From 6e274d144302068a00794ec22e73520c0615cb6f Mon Sep 17 00:00:00 2001
From: Alexander Nyberg <alexn@telia.com>
Date: Sat, 25 Jun 2005 14:58:26 -0700
Subject: [PATCH] kdump: Use real pt_regs from exception

Makes kexec_crashdump() take a pt_regs * as an argument.  This allows to
get exact register state at the point of the crash.  If we come from direct
panic assertion NULL will be passed and the current registers saved before
crashdump.

This hooks into two places:
die(): check the conditions under which we will panic when calling
do_exit and go there directly with the pt_regs that caused the fatal
fault.

die_nmi(): If we receive an NMI lockup while in the kernel use the
pt_regs and go directly to crash_kexec(). We're probably nested up badly
at this point so this might be the only chance to escape with proper
information.

Signed-off-by: Alexander Nyberg <alexn@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/crash.c          | 36 ++++++++++++++++++++++++------------
 arch/i386/kernel/traps.c          | 17 +++++++++++++++++
 arch/ppc/kernel/machine_kexec.c   |  2 +-
 arch/ppc64/kernel/machine_kexec.c |  2 +-
 arch/s390/kernel/crash.c          |  2 +-
 arch/x86_64/kernel/crash.c        |  2 +-
 drivers/char/sysrq.c              |  2 +-
 include/linux/kexec.h             |  8 ++++++--
 include/linux/reboot.h            |  3 ++-
 kernel/kexec.c                    | 13 +++++++++++--
 kernel/panic.c                    |  2 +-
 11 files changed, 66 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c
index a021681d21f..8bdb4b6af0f 100644
--- a/arch/i386/kernel/crash.c
+++ b/arch/i386/kernel/crash.c
@@ -100,12 +100,31 @@ static void crash_get_current_regs(struct pt_regs *regs)
 	regs->eip = (unsigned long)current_text_addr();
 }
 
-static void crash_save_self(void)
+/* CPU does not save ss and esp on stack if execution is already
+ * running in kernel mode at the time of NMI occurrence. This code
+ * fixes it.
+ */
+static void crash_setup_regs(struct pt_regs *newregs, struct pt_regs *oldregs)
+{
+	memcpy(newregs, oldregs, sizeof(*newregs));
+	newregs->esp = (unsigned long)&(oldregs->esp);
+	__asm__ __volatile__("xorl %eax, %eax;");
+	__asm__ __volatile__ ("movw %%ss, %%ax;" :"=a"(newregs->xss));
+}
+
+/* We may have saved_regs from where the error came from
+ * or it is NULL if via a direct panic().
+ */
+static void crash_save_self(struct pt_regs *saved_regs)
 {
 	struct pt_regs regs;
 	int cpu;
 	cpu = smp_processor_id();
-	crash_get_current_regs(&regs);
+
+	if (saved_regs)
+		crash_setup_regs(&regs, saved_regs);
+	else
+		crash_get_current_regs(&regs);
 	crash_save_this_cpu(&regs, cpu);
 }
 
@@ -124,15 +143,8 @@ static int crash_nmi_callback(struct pt_regs *regs, int cpu)
 		return 1;
 	local_irq_disable();
 
-	/* CPU does not save ss and esp on stack if execution is already
-	 * running in kernel mode at the time of NMI occurrence. This code
-	 * fixes it.
-	 */
 	if (!user_mode(regs)) {
-		memcpy(&fixed_regs, regs, sizeof(*regs));
-		fixed_regs.esp = (unsigned long)&(regs->esp);
-		__asm__ __volatile__("xorl %eax, %eax;");
-		__asm__ __volatile__ ("movw %%ss, %%ax;" :"=a"(fixed_regs.xss));
+		crash_setup_regs(&fixed_regs, regs);
 		regs = &fixed_regs;
 	}
 	crash_save_this_cpu(regs, cpu);
@@ -184,7 +196,7 @@ static void nmi_shootdown_cpus(void)
 }
 #endif
 
-void machine_crash_shutdown(void)
+void machine_crash_shutdown(struct pt_regs *regs)
 {
 	/* This function is only called after the system
 	 * has paniced or is otherwise in a critical state.
@@ -204,5 +216,5 @@ void machine_crash_shutdown(void)
 #if defined(CONFIG_X86_IO_APIC)
 	disable_IO_APIC();
 #endif
-	crash_save_self();
+	crash_save_self(regs);
 }
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 207ea8ba716..e458463ebc0 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -27,6 +27,7 @@
 #include <linux/ptrace.h>
 #include <linux/utsname.h>
 #include <linux/kprobes.h>
+#include <linux/kexec.h>
 
 #ifdef CONFIG_EISA
 #include <linux/ioport.h>
@@ -294,6 +295,9 @@ bug:
 	printk("Kernel BUG\n");
 }
 
+/* This is gone through when something in the kernel
+ * has done something bad and is about to be terminated.
+*/
 void die(const char * str, struct pt_regs * regs, long err)
 {
 	static struct {
@@ -341,6 +345,10 @@ void die(const char * str, struct pt_regs * regs, long err)
 	bust_spinlocks(0);
 	die.lock_owner = -1;
 	spin_unlock_irq(&die.lock);
+
+	if (kexec_should_crash(current))
+		crash_kexec(regs);
+
 	if (in_interrupt())
 		panic("Fatal exception in interrupt");
 
@@ -570,6 +578,15 @@ void die_nmi (struct pt_regs *regs, const char *msg)
 	console_silent();
 	spin_unlock(&nmi_print_lock);
 	bust_spinlocks(0);
+
+	/* If we are in kernel we are probably nested up pretty bad
+	 * and might aswell get out now while we still can.
+	*/
+	if (!user_mode(regs)) {
+		current->thread.trap_no = 2;
+		crash_kexec(regs);
+	}
+
 	do_exit(SIGSEGV);
 }
 
diff --git a/arch/ppc/kernel/machine_kexec.c b/arch/ppc/kernel/machine_kexec.c
index 435ad9ea0a8..b82535357d6 100644
--- a/arch/ppc/kernel/machine_kexec.c
+++ b/arch/ppc/kernel/machine_kexec.c
@@ -34,7 +34,7 @@ void machine_shutdown(void)
 	}
 }
 
-void machine_crash_shutdown(void)
+void machine_crash_shutdown(struct pt_regs *regs)
 {
 	if (ppc_md.machine_crash_shutdown) {
 		ppc_md.machine_crash_shutdown();
diff --git a/arch/ppc64/kernel/machine_kexec.c b/arch/ppc64/kernel/machine_kexec.c
index 217965d60a4..06b25b59c8a 100644
--- a/arch/ppc64/kernel/machine_kexec.c
+++ b/arch/ppc64/kernel/machine_kexec.c
@@ -34,7 +34,7 @@ note_buf_t crash_notes[NR_CPUS];
  * and if what it will achieve. Letting it be now to compile the code
  * in generic kexec environment
  */
-void machine_crash_shutdown(void)
+void machine_crash_shutdown(struct pt_regs *regs)
 {
 	/* do nothing right now */
 	/* smp_relase_cpus() if we want smp on panic kernel */
diff --git a/arch/s390/kernel/crash.c b/arch/s390/kernel/crash.c
index db38283c1f2..7bd169c58b0 100644
--- a/arch/s390/kernel/crash.c
+++ b/arch/s390/kernel/crash.c
@@ -12,6 +12,6 @@
 
 note_buf_t crash_notes[NR_CPUS];
 
-void machine_crash_shutdown(void)
+void machine_crash_shutdown(struct pt_regs *regs)
 {
 }
diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c
index 6183bcb8525..d7fa4248501 100644
--- a/arch/x86_64/kernel/crash.c
+++ b/arch/x86_64/kernel/crash.c
@@ -22,7 +22,7 @@
 
 note_buf_t crash_notes[NR_CPUS];
 
-void machine_crash_shutdown(void)
+void machine_crash_shutdown(struct pt_regs *regs)
 {
 	/* This function is only called after the system
 	 * has paniced or is otherwise in a critical state.
diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c
index 53b2c8fab00..af79805b557 100644
--- a/drivers/char/sysrq.c
+++ b/drivers/char/sysrq.c
@@ -100,7 +100,7 @@ static struct sysrq_key_op sysrq_unraw_op = {
 static void sysrq_handle_crashdump(int key, struct pt_regs *pt_regs,
 				struct tty_struct *tty)
 {
-	crash_kexec();
+	crash_kexec(pt_regs);
 }
 static struct sysrq_key_op sysrq_crashdump_op = {
 	.handler	= sysrq_handle_crashdump,
diff --git a/include/linux/kexec.h b/include/linux/kexec.h
index 0653a27c3d7..7383173a3a9 100644
--- a/include/linux/kexec.h
+++ b/include/linux/kexec.h
@@ -99,7 +99,8 @@ extern asmlinkage long compat_sys_kexec_load(unsigned long entry,
 	unsigned long flags);
 #endif
 extern struct page *kimage_alloc_control_pages(struct kimage *image, unsigned int order);
-extern void crash_kexec(void);
+extern void crash_kexec(struct pt_regs *);
+int kexec_should_crash(struct task_struct *);
 extern struct kimage *kexec_image;
 
 #define KEXEC_ON_CRASH  0x00000001
@@ -123,6 +124,9 @@ extern struct kimage *kexec_image;
 extern struct resource crashk_res;
 
 #else /* !CONFIG_KEXEC */
-static inline void crash_kexec(void) { }
+struct pt_regs;
+struct task_struct;
+static inline void crash_kexec(struct pt_regs *regs) { }
+static inline int kexec_should_crash(struct task_struct *p) { return 0; }
 #endif /* CONFIG_KEXEC */
 #endif /* LINUX_KEXEC_H */
diff --git a/include/linux/reboot.h b/include/linux/reboot.h
index c5a05e16edb..2d4dd23168d 100644
--- a/include/linux/reboot.h
+++ b/include/linux/reboot.h
@@ -52,7 +52,8 @@ extern void machine_halt(void);
 extern void machine_power_off(void);
 
 extern void machine_shutdown(void);
-extern void machine_crash_shutdown(void);
+struct pt_regs;
+extern void machine_crash_shutdown(struct pt_regs *);
 
 #endif
 
diff --git a/kernel/kexec.c b/kernel/kexec.c
index a0411b3bd54..277f22afe74 100644
--- a/kernel/kexec.c
+++ b/kernel/kexec.c
@@ -18,6 +18,8 @@
 #include <linux/reboot.h>
 #include <linux/syscalls.h>
 #include <linux/ioport.h>
+#include <linux/hardirq.h>
+
 #include <asm/page.h>
 #include <asm/uaccess.h>
 #include <asm/io.h>
@@ -32,6 +34,13 @@ struct resource crashk_res = {
 	.flags = IORESOURCE_BUSY | IORESOURCE_MEM
 };
 
+int kexec_should_crash(struct task_struct *p)
+{
+	if (in_interrupt() || !p->pid || p->pid == 1 || panic_on_oops)
+		return 1;
+	return 0;
+}
+
 /*
  * When kexec transitions to the new kernel there is a one-to-one
  * mapping between physical and virtual addresses.  On processors
@@ -1010,7 +1019,7 @@ asmlinkage long compat_sys_kexec_load(unsigned long entry,
 }
 #endif
 
-void crash_kexec(void)
+void crash_kexec(struct pt_regs *regs)
 {
 	struct kimage *image;
 	int locked;
@@ -1028,7 +1037,7 @@ void crash_kexec(void)
 	if (!locked) {
 		image = xchg(&kexec_crash_image, NULL);
 		if (image) {
-			machine_crash_shutdown();
+			machine_crash_shutdown(regs);
 			machine_kexec(image);
 		}
 		xchg(&kexec_lock, 0);
diff --git a/kernel/panic.c b/kernel/panic.c
index 66f43d33cd8..74ba5f3e46c 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -83,7 +83,7 @@ NORET_TYPE void panic(const char * fmt, ...)
 	 * everything else.
 	 * Do we want to call this before we try to display a message?
 	 */
-	crash_kexec();
+	crash_kexec(NULL);
 
 #ifdef CONFIG_SMP
 	/*
-- 
cgit v1.2.3-70-g09d2


From 72414d3f1d22fc3e311b162fca95c430048d38ce Mon Sep 17 00:00:00 2001
From: Maneesh Soni <maneesh@in.ibm.com>
Date: Sat, 25 Jun 2005 14:58:28 -0700
Subject: [PATCH] kexec code cleanup

o Following patch provides purely cosmetic changes and corrects CodingStyle
  guide lines related certain issues like below in kexec related files

  o braces for one line "if" statements, "for" loops,
  o more than 80 column wide lines,
  o No space after "while", "for" and "switch" key words

o Changes:
  o take-2: Removed the extra tab before "case" key words.
  o take-3: Put operator at the end of line and space before "*/"

Signed-off-by: Maneesh Soni <maneesh@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/crash.c           |  23 +--
 arch/i386/kernel/machine_kexec.c   |  16 +-
 arch/ppc/kernel/machine_kexec.c    |  30 ++--
 arch/ppc64/kernel/machine_kexec.c  |   9 +-
 arch/s390/kernel/machine_kexec.c   |   4 +-
 arch/x86_64/kernel/machine_kexec.c |  49 +++---
 drivers/char/mem.c                 |   2 +-
 include/linux/kexec.h              |  13 +-
 include/linux/syscalls.h           |   6 +-
 kernel/kexec.c                     | 302 ++++++++++++++++++++-----------------
 10 files changed, 243 insertions(+), 211 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c
index 8bdb4b6af0f..e5fab12f792 100644
--- a/arch/i386/kernel/crash.c
+++ b/arch/i386/kernel/crash.c
@@ -31,10 +31,11 @@ note_buf_t crash_notes[NR_CPUS];
 /* This keeps a track of which one is crashing cpu. */
 static int crashing_cpu;
 
-static u32 *append_elf_note(u32 *buf,
-	char *name, unsigned type, void *data, size_t data_len)
+static u32 *append_elf_note(u32 *buf, char *name, unsigned type, void *data,
+							       size_t data_len)
 {
 	struct elf_note note;
+
 	note.n_namesz = strlen(name) + 1;
 	note.n_descsz = data_len;
 	note.n_type   = type;
@@ -44,26 +45,28 @@ static u32 *append_elf_note(u32 *buf,
 	buf += (note.n_namesz + 3)/4;
 	memcpy(buf, data, note.n_descsz);
 	buf += (note.n_descsz + 3)/4;
+
 	return buf;
 }
 
 static void final_note(u32 *buf)
 {
 	struct elf_note note;
+
 	note.n_namesz = 0;
 	note.n_descsz = 0;
 	note.n_type   = 0;
 	memcpy(buf, &note, sizeof(note));
 }
 
-
 static void crash_save_this_cpu(struct pt_regs *regs, int cpu)
 {
 	struct elf_prstatus prstatus;
 	u32 *buf;
-	if ((cpu < 0) || (cpu >= NR_CPUS)) {
+
+	if ((cpu < 0) || (cpu >= NR_CPUS))
 		return;
-	}
+
 	/* Using ELF notes here is opportunistic.
 	 * I need a well defined structure format
 	 * for the data I pass, and I need tags
@@ -75,9 +78,8 @@ static void crash_save_this_cpu(struct pt_regs *regs, int cpu)
 	memset(&prstatus, 0, sizeof(prstatus));
 	prstatus.pr_pid = current->pid;
 	elf_core_copy_regs(&prstatus.pr_reg, regs);
-	buf = append_elf_note(buf, "CORE", NT_PRSTATUS,
-		&prstatus, sizeof(prstatus));
-
+	buf = append_elf_note(buf, "CORE", NT_PRSTATUS, &prstatus,
+				sizeof(prstatus));
 	final_note(buf);
 }
 
@@ -119,8 +121,8 @@ static void crash_save_self(struct pt_regs *saved_regs)
 {
 	struct pt_regs regs;
 	int cpu;
-	cpu = smp_processor_id();
 
+	cpu = smp_processor_id();
 	if (saved_regs)
 		crash_setup_regs(&regs, saved_regs);
 	else
@@ -153,6 +155,7 @@ static int crash_nmi_callback(struct pt_regs *regs, int cpu)
 	/* Assume hlt works */
 	__asm__("hlt");
 	for(;;);
+
 	return 1;
 }
 
@@ -169,8 +172,8 @@ static void smp_send_nmi_allbutself(void)
 static void nmi_shootdown_cpus(void)
 {
 	unsigned long msecs;
-	atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1);
 
+	atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1);
 	/* Would it be better to replace the trap vector here? */
 	set_nmi_callback(crash_nmi_callback);
 	/* Ensure the new callback function is set before sending
diff --git a/arch/i386/kernel/machine_kexec.c b/arch/i386/kernel/machine_kexec.c
index 671880415d1..52ed18d8b51 100644
--- a/arch/i386/kernel/machine_kexec.c
+++ b/arch/i386/kernel/machine_kexec.c
@@ -80,7 +80,8 @@ static void identity_map_page(unsigned long address)
 	/* Identity map the page table entry */
 	pgtable_level1[level1_index] = address | L0_ATTR;
 	pgtable_level2[level2_index] = __pa(pgtable_level1) | L1_ATTR;
-	set_64bit(&pgtable_level3[level3_index], __pa(pgtable_level2) | L2_ATTR);
+	set_64bit(&pgtable_level3[level3_index],
+					       __pa(pgtable_level2) | L2_ATTR);
 
 	/* Flush the tlb so the new mapping takes effect.
 	 * Global tlb entries are not flushed but that is not an issue.
@@ -139,8 +140,10 @@ static void load_segments(void)
 }
 
 typedef asmlinkage NORET_TYPE void (*relocate_new_kernel_t)(
-	unsigned long indirection_page, unsigned long reboot_code_buffer,
-	unsigned long start_address, unsigned int has_pae) ATTRIB_NORET;
+					unsigned long indirection_page,
+					unsigned long reboot_code_buffer,
+					unsigned long start_address,
+					unsigned int has_pae) ATTRIB_NORET;
 
 const extern unsigned char relocate_new_kernel[];
 extern void relocate_new_kernel_end(void);
@@ -180,20 +183,23 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 {
 	unsigned long page_list;
 	unsigned long reboot_code_buffer;
+
 	relocate_new_kernel_t rnk;
 
 	/* Interrupts aren't acceptable while we reboot */
 	local_irq_disable();
 
 	/* Compute some offsets */
-	reboot_code_buffer = page_to_pfn(image->control_code_page) << PAGE_SHIFT;
+	reboot_code_buffer = page_to_pfn(image->control_code_page)
+								<< PAGE_SHIFT;
 	page_list = image->head;
 
 	/* Set up an identity mapping for the reboot_code_buffer */
 	identity_map_page(reboot_code_buffer);
 
 	/* copy it out */
-	memcpy((void *)reboot_code_buffer, relocate_new_kernel, relocate_new_kernel_size);
+	memcpy((void *)reboot_code_buffer, relocate_new_kernel,
+						relocate_new_kernel_size);
 
 	/* The segment registers are funny things, they are
 	 * automatically loaded from a table, in memory wherever you
diff --git a/arch/ppc/kernel/machine_kexec.c b/arch/ppc/kernel/machine_kexec.c
index b82535357d6..84d65a87191 100644
--- a/arch/ppc/kernel/machine_kexec.c
+++ b/arch/ppc/kernel/machine_kexec.c
@@ -21,24 +21,23 @@
 #include <asm/machdep.h>
 
 typedef NORET_TYPE void (*relocate_new_kernel_t)(
-	unsigned long indirection_page, unsigned long reboot_code_buffer,
-	unsigned long start_address) ATTRIB_NORET;
+				unsigned long indirection_page,
+				unsigned long reboot_code_buffer,
+				unsigned long start_address) ATTRIB_NORET;
 
 const extern unsigned char relocate_new_kernel[];
 const extern unsigned int relocate_new_kernel_size;
 
 void machine_shutdown(void)
 {
-	if (ppc_md.machine_shutdown) {
+	if (ppc_md.machine_shutdown)
 		ppc_md.machine_shutdown();
-	}
 }
 
 void machine_crash_shutdown(struct pt_regs *regs)
 {
-	if (ppc_md.machine_crash_shutdown) {
+	if (ppc_md.machine_crash_shutdown)
 		ppc_md.machine_crash_shutdown();
-	}
 }
 
 /*
@@ -48,9 +47,8 @@ void machine_crash_shutdown(struct pt_regs *regs)
  */
 int machine_kexec_prepare(struct kimage *image)
 {
-	if (ppc_md.machine_kexec_prepare) {
+	if (ppc_md.machine_kexec_prepare)
 		return ppc_md.machine_kexec_prepare(image);
-	}
 	/*
 	 * Fail if platform doesn't provide its own machine_kexec_prepare
 	 * implementation.
@@ -60,9 +58,8 @@ int machine_kexec_prepare(struct kimage *image)
 
 void machine_kexec_cleanup(struct kimage *image)
 {
-	if (ppc_md.machine_kexec_cleanup) {
+	if (ppc_md.machine_kexec_cleanup)
 		ppc_md.machine_kexec_cleanup(image);
-	}
 }
 
 /*
@@ -71,9 +68,9 @@ void machine_kexec_cleanup(struct kimage *image)
  */
 NORET_TYPE void machine_kexec(struct kimage *image)
 {
-	if (ppc_md.machine_kexec) {
+	if (ppc_md.machine_kexec)
 		ppc_md.machine_kexec(image);
-	} else {
+	else {
 		/*
 		 * Fall back to normal restart if platform doesn't provide
 		 * its own kexec function, and user insist to kexec...
@@ -83,7 +80,6 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	for(;;);
 }
 
-
 /*
  * This is a generic machine_kexec function suitable at least for
  * non-OpenFirmware embedded platforms.
@@ -104,15 +100,15 @@ void machine_kexec_simple(struct kimage *image)
 
 	/* we need both effective and real address here */
 	reboot_code_buffer =
-		(unsigned long)page_address(image->control_code_page);
+			(unsigned long)page_address(image->control_code_page);
 	reboot_code_buffer_phys = virt_to_phys((void *)reboot_code_buffer);
 
 	/* copy our kernel relocation code to the control code page */
-	memcpy((void *)reboot_code_buffer,
-		relocate_new_kernel, relocate_new_kernel_size);
+	memcpy((void *)reboot_code_buffer, relocate_new_kernel,
+						relocate_new_kernel_size);
 
 	flush_icache_range(reboot_code_buffer,
-		reboot_code_buffer + KEXEC_CONTROL_CODE_SIZE);
+				reboot_code_buffer + KEXEC_CONTROL_CODE_SIZE);
 	printk(KERN_INFO "Bye!\n");
 
 	/* now call it */
diff --git a/arch/ppc64/kernel/machine_kexec.c b/arch/ppc64/kernel/machine_kexec.c
index 06b25b59c8a..fdb2fc649d7 100644
--- a/arch/ppc64/kernel/machine_kexec.c
+++ b/arch/ppc64/kernel/machine_kexec.c
@@ -58,7 +58,7 @@ int machine_kexec_prepare(struct kimage *image)
 	 * handle the virtual mode, we must make sure no destination
 	 * overlaps kernel static data or bss.
 	 */
-	for(i = 0; i < image->nr_segments; i++)
+	for (i = 0; i < image->nr_segments; i++)
 		if (image->segment[i].mem < __pa(_end))
 			return -ETXTBSY;
 
@@ -76,7 +76,7 @@ int machine_kexec_prepare(struct kimage *image)
 		low = __pa(htab_address);
 		high = low + (htab_hash_mask + 1) * HASH_GROUP_SIZE;
 
-		for(i = 0; i < image->nr_segments; i++) {
+		for (i = 0; i < image->nr_segments; i++) {
 			begin = image->segment[i].mem;
 			end = begin + image->segment[i].memsz;
 
@@ -98,7 +98,7 @@ int machine_kexec_prepare(struct kimage *image)
 		low = *basep;
 		high = low + (*sizep);
 
-		for(i = 0; i < image->nr_segments; i++) {
+		for (i = 0; i < image->nr_segments; i++) {
 			begin = image->segment[i].mem;
 			end = begin + image->segment[i].memsz;
 
@@ -274,7 +274,8 @@ union thread_union kexec_stack
 
 /* Our assembly helper, in kexec_stub.S */
 extern NORET_TYPE void kexec_sequence(void *newstack, unsigned long start,
-	void *image, void *control, void (*clear_all)(void)) ATTRIB_NORET;
+					void *image, void *control,
+					void (*clear_all)(void)) ATTRIB_NORET;
 
 /* too late to fail here */
 void machine_kexec(struct kimage *image)
diff --git a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c
index 7a94db76df4..2721c3a32b8 100644
--- a/arch/s390/kernel/machine_kexec.c
+++ b/arch/s390/kernel/machine_kexec.c
@@ -67,7 +67,7 @@ machine_kexec(struct kimage *image)
 	ctl_clear_bit(0,28);
 
 	on_each_cpu(kexec_halt_all_cpus, image, 0, 0);
-	for(;;);
+	for (;;);
 }
 
 static void
@@ -85,7 +85,7 @@ kexec_halt_all_cpus(void *kernel_image)
 	for_each_online_cpu(cpu) {
 		if (cpu == smp_processor_id())
 			continue;
-		while(!smp_cpu_not_running(cpu))
+		while (!smp_cpu_not_running(cpu))
 			cpu_relax();
 	}
 
diff --git a/arch/x86_64/kernel/machine_kexec.c b/arch/x86_64/kernel/machine_kexec.c
index 200b5993f8d..60d1eff4156 100644
--- a/arch/x86_64/kernel/machine_kexec.c
+++ b/arch/x86_64/kernel/machine_kexec.c
@@ -32,29 +32,31 @@
 #define L2_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
 #define L3_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
 
-static void init_level2_page(
-	u64 *level2p, unsigned long addr)
+static void init_level2_page(u64 *level2p, unsigned long addr)
 {
 	unsigned long end_addr;
+
 	addr &= PAGE_MASK;
 	end_addr = addr + LEVEL2_SIZE;
-	while(addr < end_addr) {
+	while (addr < end_addr) {
 		*(level2p++) = addr | L1_ATTR;
 		addr += LEVEL1_SIZE;
 	}
 }
 
-static int init_level3_page(struct kimage *image,
-	u64 *level3p, unsigned long addr, unsigned long last_addr)
+static int init_level3_page(struct kimage *image, u64 *level3p,
+				unsigned long addr, unsigned long last_addr)
 {
 	unsigned long end_addr;
 	int result;
+
 	result = 0;
 	addr &= PAGE_MASK;
 	end_addr = addr + LEVEL3_SIZE;
-	while((addr < last_addr) && (addr < end_addr)) {
+	while ((addr < last_addr) && (addr < end_addr)) {
 		struct page *page;
 		u64 *level2p;
+
 		page = kimage_alloc_control_pages(image, 0);
 		if (!page) {
 			result = -ENOMEM;
@@ -66,7 +68,7 @@ static int init_level3_page(struct kimage *image,
 		addr += LEVEL2_SIZE;
 	}
 	/* clear the unused entries */
-	while(addr < end_addr) {
+	while (addr < end_addr) {
 		*(level3p++) = 0;
 		addr += LEVEL2_SIZE;
 	}
@@ -75,17 +77,19 @@ out:
 }
 
 
-static int init_level4_page(struct kimage *image,
-	u64 *level4p, unsigned long addr, unsigned long last_addr)
+static int init_level4_page(struct kimage *image, u64 *level4p,
+				unsigned long addr, unsigned long last_addr)
 {
 	unsigned long end_addr;
 	int result;
+
 	result = 0;
 	addr &= PAGE_MASK;
 	end_addr = addr + LEVEL4_SIZE;
-	while((addr < last_addr) && (addr < end_addr)) {
+	while ((addr < last_addr) && (addr < end_addr)) {
 		struct page *page;
 		u64 *level3p;
+
 		page = kimage_alloc_control_pages(image, 0);
 		if (!page) {
 			result = -ENOMEM;
@@ -100,11 +104,11 @@ static int init_level4_page(struct kimage *image,
 		addr += LEVEL3_SIZE;
 	}
 	/* clear the unused entries */
-	while(addr < end_addr) {
+	while (addr < end_addr) {
 		*(level4p++) = 0;
 		addr += LEVEL3_SIZE;
 	}
- out:
+out:
 	return result;
 }
 
@@ -113,7 +117,7 @@ static int init_pgtable(struct kimage *image, unsigned long start_pgtable)
 {
 	u64 *level4p;
 	level4p = (u64 *)__va(start_pgtable);
-	return init_level4_page(image, level4p, 0, end_pfn << PAGE_SHIFT);
+ 	return init_level4_page(image, level4p, 0, end_pfn << PAGE_SHIFT);
 }
 
 static void set_idt(void *newidt, u16 limit)
@@ -159,9 +163,10 @@ static void load_segments(void)
 #undef __STR
 }
 
-typedef NORET_TYPE void (*relocate_new_kernel_t)(
-	unsigned long indirection_page, unsigned long control_code_buffer,
-	unsigned long start_address, unsigned long pgtable) ATTRIB_NORET;
+typedef NORET_TYPE void (*relocate_new_kernel_t)(unsigned long indirection_page,
+					unsigned long control_code_buffer,
+					unsigned long start_address,
+					unsigned long pgtable) ATTRIB_NORET;
 
 const extern unsigned char relocate_new_kernel[];
 const extern unsigned long relocate_new_kernel_size;
@@ -172,17 +177,17 @@ int machine_kexec_prepare(struct kimage *image)
 	int result;
 
 	/* Calculate the offsets */
-	start_pgtable       = page_to_pfn(image->control_code_page) << PAGE_SHIFT;
+	start_pgtable = page_to_pfn(image->control_code_page) << PAGE_SHIFT;
 	control_code_buffer = start_pgtable + 4096UL;
 
 	/* Setup the identity mapped 64bit page table */
 	result = init_pgtable(image, start_pgtable);
-	if (result) {
+	if (result)
 		return result;
-	}
 
 	/* Place the code in the reboot code buffer */
-	memcpy(__va(control_code_buffer), relocate_new_kernel, relocate_new_kernel_size);
+	memcpy(__va(control_code_buffer), relocate_new_kernel,
+						relocate_new_kernel_size);
 
 	return 0;
 }
@@ -207,8 +212,8 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	local_irq_disable();
 
 	/* Calculate the offsets */
-	page_list           = image->head;
-	start_pgtable       = page_to_pfn(image->control_code_page) << PAGE_SHIFT;
+	page_list = image->head;
+	start_pgtable = page_to_pfn(image->control_code_page) << PAGE_SHIFT;
 	control_code_buffer = start_pgtable + 4096UL;
 
 	/* Set the low half of the page table to my identity mapped
diff --git a/drivers/char/mem.c b/drivers/char/mem.c
index b64108dd765..42187381506 100644
--- a/drivers/char/mem.c
+++ b/drivers/char/mem.c
@@ -287,7 +287,7 @@ static ssize_t read_oldmem(struct file *file, char __user *buf,
 	size_t read = 0, csize;
 	int rc = 0;
 
-	while(count) {
+	while (count) {
 		pfn = *ppos / PAGE_SIZE;
 		if (pfn > saved_max_pfn)
 			return read;
diff --git a/include/linux/kexec.h b/include/linux/kexec.h
index 7383173a3a9..c8468472aec 100644
--- a/include/linux/kexec.h
+++ b/include/linux/kexec.h
@@ -91,14 +91,17 @@ extern NORET_TYPE void machine_kexec(struct kimage *image) ATTRIB_NORET;
 extern int machine_kexec_prepare(struct kimage *image);
 extern void machine_kexec_cleanup(struct kimage *image);
 extern asmlinkage long sys_kexec_load(unsigned long entry,
-	unsigned long nr_segments, struct kexec_segment __user *segments,
-	unsigned long flags);
+					unsigned long nr_segments,
+					struct kexec_segment __user *segments,
+					unsigned long flags);
 #ifdef CONFIG_COMPAT
 extern asmlinkage long compat_sys_kexec_load(unsigned long entry,
-	unsigned long nr_segments, struct compat_kexec_segment __user *segments,
-	unsigned long flags);
+				unsigned long nr_segments,
+				struct compat_kexec_segment __user *segments,
+				unsigned long flags);
 #endif
-extern struct page *kimage_alloc_control_pages(struct kimage *image, unsigned int order);
+extern struct page *kimage_alloc_control_pages(struct kimage *image,
+						unsigned int order);
 extern void crash_kexec(struct pt_regs *);
 int kexec_should_crash(struct task_struct *);
 extern struct kimage *kexec_image;
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 7ba8f8f747a..52830b6d94e 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -159,9 +159,9 @@ asmlinkage long sys_shutdown(int, int);
 asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd,
 				void __user *arg);
 asmlinkage long sys_restart_syscall(void);
-asmlinkage long sys_kexec_load(unsigned long entry,
-	unsigned long nr_segments, struct kexec_segment __user *segments,
-	unsigned long flags);
+asmlinkage long sys_kexec_load(unsigned long entry, unsigned long nr_segments,
+				struct kexec_segment __user *segments,
+				unsigned long flags);
 
 asmlinkage long sys_exit(int error_code);
 asmlinkage void sys_exit_group(int error_code);
diff --git a/kernel/kexec.c b/kernel/kexec.c
index 277f22afe74..7843548cf2d 100644
--- a/kernel/kexec.c
+++ b/kernel/kexec.c
@@ -87,12 +87,15 @@ int kexec_should_crash(struct task_struct *p)
  */
 #define KIMAGE_NO_DEST (-1UL)
 
-static int kimage_is_destination_range(
-	struct kimage *image, unsigned long start, unsigned long end);
-static struct page *kimage_alloc_page(struct kimage *image, unsigned int gfp_mask, unsigned long dest);
+static int kimage_is_destination_range(struct kimage *image,
+				       unsigned long start, unsigned long end);
+static struct page *kimage_alloc_page(struct kimage *image,
+				       unsigned int gfp_mask,
+				       unsigned long dest);
 
 static int do_kimage_alloc(struct kimage **rimage, unsigned long entry,
-	unsigned long nr_segments, struct kexec_segment __user *segments)
+	                    unsigned long nr_segments,
+                            struct kexec_segment __user *segments)
 {
 	size_t segment_bytes;
 	struct kimage *image;
@@ -102,9 +105,9 @@ static int do_kimage_alloc(struct kimage **rimage, unsigned long entry,
 	/* Allocate a controlling structure */
 	result = -ENOMEM;
 	image = kmalloc(sizeof(*image), GFP_KERNEL);
-	if (!image) {
+	if (!image)
 		goto out;
-	}
+
 	memset(image, 0, sizeof(*image));
 	image->head = 0;
 	image->entry = &image->head;
@@ -145,6 +148,7 @@ static int do_kimage_alloc(struct kimage **rimage, unsigned long entry,
 	result = -EADDRNOTAVAIL;
 	for (i = 0; i < nr_segments; i++) {
 		unsigned long mstart, mend;
+
 		mstart = image->segment[i].mem;
 		mend   = mstart + image->segment[i].memsz;
 		if ((mstart & ~PAGE_MASK) || (mend & ~PAGE_MASK))
@@ -159,12 +163,13 @@ static int do_kimage_alloc(struct kimage **rimage, unsigned long entry,
 	 * easy explanation as one segment stops on another.
 	 */
 	result = -EINVAL;
-	for(i = 0; i < nr_segments; i++) {
+	for (i = 0; i < nr_segments; i++) {
 		unsigned long mstart, mend;
 		unsigned long j;
+
 		mstart = image->segment[i].mem;
 		mend   = mstart + image->segment[i].memsz;
-		for(j = 0; j < i; j++) {
+		for (j = 0; j < i; j++) {
 			unsigned long pstart, pend;
 			pstart = image->segment[j].mem;
 			pend   = pstart + image->segment[j].memsz;
@@ -180,25 +185,25 @@ static int do_kimage_alloc(struct kimage **rimage, unsigned long entry,
 	 * later on.
 	 */
 	result = -EINVAL;
-	for(i = 0; i < nr_segments; i++) {
+	for (i = 0; i < nr_segments; i++) {
 		if (image->segment[i].bufsz > image->segment[i].memsz)
 			goto out;
 	}
 
-
 	result = 0;
- out:
-	if (result == 0) {
+out:
+	if (result == 0)
 		*rimage = image;
-	} else {
+	else
 		kfree(image);
-	}
+
 	return result;
 
 }
 
 static int kimage_normal_alloc(struct kimage **rimage, unsigned long entry,
-	unsigned long nr_segments, struct kexec_segment __user *segments)
+				unsigned long nr_segments,
+				struct kexec_segment __user *segments)
 {
 	int result;
 	struct kimage *image;
@@ -206,9 +211,9 @@ static int kimage_normal_alloc(struct kimage **rimage, unsigned long entry,
 	/* Allocate and initialize a controlling structure */
 	image = NULL;
 	result = do_kimage_alloc(&image, entry, nr_segments, segments);
-	if (result) {
+	if (result)
 		goto out;
-	}
+
 	*rimage = image;
 
 	/*
@@ -218,7 +223,7 @@ static int kimage_normal_alloc(struct kimage **rimage, unsigned long entry,
 	 */
 	result = -ENOMEM;
 	image->control_code_page = kimage_alloc_control_pages(image,
-		get_order(KEXEC_CONTROL_CODE_SIZE));
+					   get_order(KEXEC_CONTROL_CODE_SIZE));
 	if (!image->control_code_page) {
 		printk(KERN_ERR "Could not allocate control_code_buffer\n");
 		goto out;
@@ -226,16 +231,17 @@ static int kimage_normal_alloc(struct kimage **rimage, unsigned long entry,
 
 	result = 0;
  out:
-	if (result == 0) {
+	if (result == 0)
 		*rimage = image;
-	} else {
+	else
 		kfree(image);
-	}
+
 	return result;
 }
 
 static int kimage_crash_alloc(struct kimage **rimage, unsigned long entry,
-	unsigned long nr_segments, struct kexec_segment *segments)
+				unsigned long nr_segments,
+				struct kexec_segment *segments)
 {
 	int result;
 	struct kimage *image;
@@ -250,9 +256,8 @@ static int kimage_crash_alloc(struct kimage **rimage, unsigned long entry,
 
 	/* Allocate and initialize a controlling structure */
 	result = do_kimage_alloc(&image, entry, nr_segments, segments);
-	if (result) {
+	if (result)
 		goto out;
-	}
 
 	/* Enable the special crash kernel control page
 	 * allocation policy.
@@ -272,6 +277,7 @@ static int kimage_crash_alloc(struct kimage **rimage, unsigned long entry,
 	result = -EADDRNOTAVAIL;
 	for (i = 0; i < nr_segments; i++) {
 		unsigned long mstart, mend;
+
 		mstart = image->segment[i].mem;
 		mend = mstart + image->segment[i].memsz - 1;
 		/* Ensure we are within the crash kernel limits */
@@ -279,7 +285,6 @@ static int kimage_crash_alloc(struct kimage **rimage, unsigned long entry,
 			goto out;
 	}
 
-
 	/*
 	 * Find a location for the control code buffer, and add
 	 * the vector of segments so that it's pages will also be
@@ -287,80 +292,84 @@ static int kimage_crash_alloc(struct kimage **rimage, unsigned long entry,
 	 */
 	result = -ENOMEM;
 	image->control_code_page = kimage_alloc_control_pages(image,
-		get_order(KEXEC_CONTROL_CODE_SIZE));
+					   get_order(KEXEC_CONTROL_CODE_SIZE));
 	if (!image->control_code_page) {
 		printk(KERN_ERR "Could not allocate control_code_buffer\n");
 		goto out;
 	}
 
 	result = 0;
- out:
-	if (result == 0) {
+out:
+	if (result == 0)
 		*rimage = image;
-	} else {
+	else
 		kfree(image);
-	}
+
 	return result;
 }
 
-static int kimage_is_destination_range(
-	struct kimage *image, unsigned long start, unsigned long end)
+static int kimage_is_destination_range(struct kimage *image,
+					unsigned long start,
+					unsigned long end)
 {
 	unsigned long i;
 
 	for (i = 0; i < image->nr_segments; i++) {
 		unsigned long mstart, mend;
+
 		mstart = image->segment[i].mem;
-		mend   = mstart + image->segment[i].memsz;
-		if ((end > mstart) && (start < mend)) {
+		mend = mstart + image->segment[i].memsz;
+		if ((end > mstart) && (start < mend))
 			return 1;
-		}
 	}
+
 	return 0;
 }
 
-static struct page *kimage_alloc_pages(unsigned int gfp_mask, unsigned int order)
+static struct page *kimage_alloc_pages(unsigned int gfp_mask,
+					unsigned int order)
 {
 	struct page *pages;
+
 	pages = alloc_pages(gfp_mask, order);
 	if (pages) {
 		unsigned int count, i;
 		pages->mapping = NULL;
 		pages->private = order;
 		count = 1 << order;
-		for(i = 0; i < count; i++) {
+		for (i = 0; i < count; i++)
 			SetPageReserved(pages + i);
-		}
 	}
+
 	return pages;
 }
 
 static void kimage_free_pages(struct page *page)
 {
 	unsigned int order, count, i;
+
 	order = page->private;
 	count = 1 << order;
-	for(i = 0; i < count; i++) {
+	for (i = 0; i < count; i++)
 		ClearPageReserved(page + i);
-	}
 	__free_pages(page, order);
 }
 
 static void kimage_free_page_list(struct list_head *list)
 {
 	struct list_head *pos, *next;
+
 	list_for_each_safe(pos, next, list) {
 		struct page *page;
 
 		page = list_entry(pos, struct page, lru);
 		list_del(&page->lru);
-
 		kimage_free_pages(page);
 	}
 }
 
-static struct page *kimage_alloc_normal_control_pages(
-	struct kimage *image, unsigned int order)
+static struct page *kimage_alloc_normal_control_pages(struct kimage *image,
+							unsigned int order)
 {
 	/* Control pages are special, they are the intermediaries
 	 * that are needed while we copy the rest of the pages
@@ -387,6 +396,7 @@ static struct page *kimage_alloc_normal_control_pages(
 	 */
 	do {
 		unsigned long pfn, epfn, addr, eaddr;
+
 		pages = kimage_alloc_pages(GFP_KERNEL, order);
 		if (!pages)
 			break;
@@ -395,12 +405,12 @@ static struct page *kimage_alloc_normal_control_pages(
 		addr  = pfn << PAGE_SHIFT;
 		eaddr = epfn << PAGE_SHIFT;
 		if ((epfn >= (KEXEC_CONTROL_MEMORY_LIMIT >> PAGE_SHIFT)) ||
-			kimage_is_destination_range(image, addr, eaddr))
-		{
+			      kimage_is_destination_range(image, addr, eaddr)) {
 			list_add(&pages->lru, &extra_pages);
 			pages = NULL;
 		}
-	} while(!pages);
+	} while (!pages);
+
 	if (pages) {
 		/* Remember the allocated page... */
 		list_add(&pages->lru, &image->control_pages);
@@ -420,12 +430,12 @@ static struct page *kimage_alloc_normal_control_pages(
 	 * For now it is simpler to just free the pages.
 	 */
 	kimage_free_page_list(&extra_pages);
-	return pages;
 
+	return pages;
 }
 
-static struct page *kimage_alloc_crash_control_pages(
-	struct kimage *image, unsigned int order)
+static struct page *kimage_alloc_crash_control_pages(struct kimage *image,
+						      unsigned int order)
 {
 	/* Control pages are special, they are the intermediaries
 	 * that are needed while we copy the rest of the pages
@@ -450,21 +460,22 @@ static struct page *kimage_alloc_crash_control_pages(
 	 */
 	unsigned long hole_start, hole_end, size;
 	struct page *pages;
+
 	pages = NULL;
 	size = (1 << order) << PAGE_SHIFT;
 	hole_start = (image->control_page + (size - 1)) & ~(size - 1);
 	hole_end   = hole_start + size - 1;
-	while(hole_end <= crashk_res.end) {
+	while (hole_end <= crashk_res.end) {
 		unsigned long i;
-		if (hole_end > KEXEC_CONTROL_MEMORY_LIMIT) {
+
+		if (hole_end > KEXEC_CONTROL_MEMORY_LIMIT)
 			break;
-		}
-		if (hole_end > crashk_res.end) {
+		if (hole_end > crashk_res.end)
 			break;
-		}
 		/* See if I overlap any of the segments */
-		for(i = 0; i < image->nr_segments; i++) {
+		for (i = 0; i < image->nr_segments; i++) {
 			unsigned long mstart, mend;
+
 			mstart = image->segment[i].mem;
 			mend   = mstart + image->segment[i].memsz - 1;
 			if ((hole_end >= mstart) && (hole_start <= mend)) {
@@ -480,18 +491,19 @@ static struct page *kimage_alloc_crash_control_pages(
 			break;
 		}
 	}
-	if (pages) {
+	if (pages)
 		image->control_page = hole_end;
-	}
+
 	return pages;
 }
 
 
-struct page *kimage_alloc_control_pages(
-	struct kimage *image, unsigned int order)
+struct page *kimage_alloc_control_pages(struct kimage *image,
+					 unsigned int order)
 {
 	struct page *pages = NULL;
-	switch(image->type) {
+
+	switch (image->type) {
 	case KEXEC_TYPE_DEFAULT:
 		pages = kimage_alloc_normal_control_pages(image, order);
 		break;
@@ -499,43 +511,46 @@ struct page *kimage_alloc_control_pages(
 		pages = kimage_alloc_crash_control_pages(image, order);
 		break;
 	}
+
 	return pages;
 }
 
 static int kimage_add_entry(struct kimage *image, kimage_entry_t entry)
 {
-	if (*image->entry != 0) {
+	if (*image->entry != 0)
 		image->entry++;
-	}
+
 	if (image->entry == image->last_entry) {
 		kimage_entry_t *ind_page;
 		struct page *page;
+
 		page = kimage_alloc_page(image, GFP_KERNEL, KIMAGE_NO_DEST);
-		if (!page) {
+		if (!page)
 			return -ENOMEM;
-		}
+
 		ind_page = page_address(page);
 		*image->entry = virt_to_phys(ind_page) | IND_INDIRECTION;
 		image->entry = ind_page;
-		image->last_entry =
-			ind_page + ((PAGE_SIZE/sizeof(kimage_entry_t)) - 1);
+		image->last_entry = ind_page +
+				      ((PAGE_SIZE/sizeof(kimage_entry_t)) - 1);
 	}
 	*image->entry = entry;
 	image->entry++;
 	*image->entry = 0;
+
 	return 0;
 }
 
-static int kimage_set_destination(
-	struct kimage *image, unsigned long destination)
+static int kimage_set_destination(struct kimage *image,
+				   unsigned long destination)
 {
 	int result;
 
 	destination &= PAGE_MASK;
 	result = kimage_add_entry(image, destination | IND_DESTINATION);
-	if (result == 0) {
+	if (result == 0)
 		image->destination = destination;
-	}
+
 	return result;
 }
 
@@ -546,9 +561,9 @@ static int kimage_add_page(struct kimage *image, unsigned long page)
 
 	page &= PAGE_MASK;
 	result = kimage_add_entry(image, page | IND_SOURCE);
-	if (result == 0) {
+	if (result == 0)
 		image->destination += PAGE_SIZE;
-	}
+
 	return result;
 }
 
@@ -564,10 +579,11 @@ static void kimage_free_extra_pages(struct kimage *image)
 }
 static int kimage_terminate(struct kimage *image)
 {
-	if (*image->entry != 0) {
+	if (*image->entry != 0)
 		image->entry++;
-	}
+
 	*image->entry = IND_DONE;
+
 	return 0;
 }
 
@@ -591,26 +607,24 @@ static void kimage_free(struct kimage *image)
 
 	if (!image)
 		return;
+
 	kimage_free_extra_pages(image);
 	for_each_kimage_entry(image, ptr, entry) {
 		if (entry & IND_INDIRECTION) {
 			/* Free the previous indirection page */
-			if (ind & IND_INDIRECTION) {
+			if (ind & IND_INDIRECTION)
 				kimage_free_entry(ind);
-			}
 			/* Save this indirection page until we are
 			 * done with it.
 			 */
 			ind = entry;
 		}
-		else if (entry & IND_SOURCE) {
+		else if (entry & IND_SOURCE)
 			kimage_free_entry(entry);
-		}
 	}
 	/* Free the final indirection page */
-	if (ind & IND_INDIRECTION) {
+	if (ind & IND_INDIRECTION)
 		kimage_free_entry(ind);
-	}
 
 	/* Handle any machine specific cleanup */
 	machine_kexec_cleanup(image);
@@ -620,26 +634,28 @@ static void kimage_free(struct kimage *image)
 	kfree(image);
 }
 
-static kimage_entry_t *kimage_dst_used(struct kimage *image, unsigned long page)
+static kimage_entry_t *kimage_dst_used(struct kimage *image,
+					unsigned long page)
 {
 	kimage_entry_t *ptr, entry;
 	unsigned long destination = 0;
 
 	for_each_kimage_entry(image, ptr, entry) {
-		if (entry & IND_DESTINATION) {
+		if (entry & IND_DESTINATION)
 			destination = entry & PAGE_MASK;
-		}
 		else if (entry & IND_SOURCE) {
-			if (page == destination) {
+			if (page == destination)
 				return ptr;
-			}
 			destination += PAGE_SIZE;
 		}
 	}
+
 	return 0;
 }
 
-static struct page *kimage_alloc_page(struct kimage *image, unsigned int gfp_mask, unsigned long destination)
+static struct page *kimage_alloc_page(struct kimage *image,
+					unsigned int gfp_mask,
+					unsigned long destination)
 {
 	/*
 	 * Here we implement safeguards to ensure that a source page
@@ -679,11 +695,11 @@ static struct page *kimage_alloc_page(struct kimage *image, unsigned int gfp_mas
 
 		/* Allocate a page, if we run out of memory give up */
 		page = kimage_alloc_pages(gfp_mask, 0);
-		if (!page) {
+		if (!page)
 			return 0;
-		}
 		/* If the page cannot be used file it away */
-		if (page_to_pfn(page) > (KEXEC_SOURCE_MEMORY_LIMIT >> PAGE_SHIFT)) {
+		if (page_to_pfn(page) >
+				(KEXEC_SOURCE_MEMORY_LIMIT >> PAGE_SHIFT)) {
 			list_add(&page->lru, &image->unuseable_pages);
 			continue;
 		}
@@ -694,7 +710,8 @@ static struct page *kimage_alloc_page(struct kimage *image, unsigned int gfp_mas
 			break;
 
 		/* If the page is not a destination page use it */
-		if (!kimage_is_destination_range(image, addr, addr + PAGE_SIZE))
+		if (!kimage_is_destination_range(image, addr,
+						  addr + PAGE_SIZE))
 			break;
 
 		/*
@@ -727,11 +744,12 @@ static struct page *kimage_alloc_page(struct kimage *image, unsigned int gfp_mas
 			list_add(&page->lru, &image->dest_pages);
 		}
 	}
+
 	return page;
 }
 
 static int kimage_load_normal_segment(struct kimage *image,
-	struct kexec_segment *segment)
+					 struct kexec_segment *segment)
 {
 	unsigned long maddr;
 	unsigned long ubytes, mbytes;
@@ -745,34 +763,36 @@ static int kimage_load_normal_segment(struct kimage *image,
 	maddr = segment->mem;
 
 	result = kimage_set_destination(image, maddr);
-	if (result < 0) {
+	if (result < 0)
 		goto out;
-	}
-	while(mbytes) {
+
+	while (mbytes) {
 		struct page *page;
 		char *ptr;
 		size_t uchunk, mchunk;
+
 		page = kimage_alloc_page(image, GFP_HIGHUSER, maddr);
 		if (page == 0) {
 			result  = -ENOMEM;
 			goto out;
 		}
-		result = kimage_add_page(image, page_to_pfn(page) << PAGE_SHIFT);
-		if (result < 0) {
+		result = kimage_add_page(image, page_to_pfn(page)
+								<< PAGE_SHIFT);
+		if (result < 0)
 			goto out;
-		}
+
 		ptr = kmap(page);
 		/* Start with a clear page */
 		memset(ptr, 0, PAGE_SIZE);
 		ptr += maddr & ~PAGE_MASK;
 		mchunk = PAGE_SIZE - (maddr & ~PAGE_MASK);
-		if (mchunk > mbytes) {
+		if (mchunk > mbytes)
 			mchunk = mbytes;
-		}
+
 		uchunk = mchunk;
-		if (uchunk > ubytes) {
+		if (uchunk > ubytes)
 			uchunk = ubytes;
-		}
+
 		result = copy_from_user(ptr, buf, uchunk);
 		kunmap(page);
 		if (result) {
@@ -784,12 +804,12 @@ static int kimage_load_normal_segment(struct kimage *image,
 		buf    += mchunk;
 		mbytes -= mchunk;
 	}
- out:
+out:
 	return result;
 }
 
 static int kimage_load_crash_segment(struct kimage *image,
-	struct kexec_segment *segment)
+					struct kexec_segment *segment)
 {
 	/* For crash dumps kernels we simply copy the data from
 	 * user space to it's destination.
@@ -805,10 +825,11 @@ static int kimage_load_crash_segment(struct kimage *image,
 	ubytes = segment->bufsz;
 	mbytes = segment->memsz;
 	maddr = segment->mem;
-	while(mbytes) {
+	while (mbytes) {
 		struct page *page;
 		char *ptr;
 		size_t uchunk, mchunk;
+
 		page = pfn_to_page(maddr >> PAGE_SHIFT);
 		if (page == 0) {
 			result  = -ENOMEM;
@@ -817,9 +838,9 @@ static int kimage_load_crash_segment(struct kimage *image,
 		ptr = kmap(page);
 		ptr += maddr & ~PAGE_MASK;
 		mchunk = PAGE_SIZE - (maddr & ~PAGE_MASK);
-		if (mchunk > mbytes) {
+		if (mchunk > mbytes)
 			mchunk = mbytes;
-		}
+
 		uchunk = mchunk;
 		if (uchunk > ubytes) {
 			uchunk = ubytes;
@@ -837,15 +858,16 @@ static int kimage_load_crash_segment(struct kimage *image,
 		buf    += mchunk;
 		mbytes -= mchunk;
 	}
- out:
+out:
 	return result;
 }
 
 static int kimage_load_segment(struct kimage *image,
-	struct kexec_segment *segment)
+				struct kexec_segment *segment)
 {
 	int result = -ENOMEM;
-	switch(image->type) {
+
+	switch (image->type) {
 	case KEXEC_TYPE_DEFAULT:
 		result = kimage_load_normal_segment(image, segment);
 		break;
@@ -853,6 +875,7 @@ static int kimage_load_segment(struct kimage *image,
 		result = kimage_load_crash_segment(image, segment);
 		break;
 	}
+
 	return result;
 }
 
@@ -885,9 +908,9 @@ static struct kimage *kexec_crash_image = NULL;
  */
 static int kexec_lock = 0;
 
-asmlinkage long sys_kexec_load(unsigned long entry,
-	unsigned long nr_segments, struct kexec_segment __user *segments,
-	unsigned long flags)
+asmlinkage long sys_kexec_load(unsigned long entry, unsigned long nr_segments,
+				struct kexec_segment __user *segments,
+				unsigned long flags)
 {
 	struct kimage **dest_image, *image;
 	int locked;
@@ -907,9 +930,7 @@ asmlinkage long sys_kexec_load(unsigned long entry,
 	/* Verify we are on the appropriate architecture */
 	if (((flags & KEXEC_ARCH_MASK) != KEXEC_ARCH) &&
 		((flags & KEXEC_ARCH_MASK) != KEXEC_ARCH_DEFAULT))
-	{
 		return -EINVAL;
-	}
 
 	/* Put an artificial cap on the number
 	 * of segments passed to kexec_load.
@@ -929,58 +950,59 @@ asmlinkage long sys_kexec_load(unsigned long entry,
 	 * KISS: always take the mutex.
 	 */
 	locked = xchg(&kexec_lock, 1);
-	if (locked) {
+	if (locked)
 		return -EBUSY;
-	}
+
 	dest_image = &kexec_image;
-	if (flags & KEXEC_ON_CRASH) {
+	if (flags & KEXEC_ON_CRASH)
 		dest_image = &kexec_crash_image;
-	}
 	if (nr_segments > 0) {
 		unsigned long i;
+
 		/* Loading another kernel to reboot into */
-		if ((flags & KEXEC_ON_CRASH) == 0) {
-			result = kimage_normal_alloc(&image, entry, nr_segments, segments);
-		}
+		if ((flags & KEXEC_ON_CRASH) == 0)
+			result = kimage_normal_alloc(&image, entry,
+							nr_segments, segments);
 		/* Loading another kernel to switch to if this one crashes */
 		else if (flags & KEXEC_ON_CRASH) {
 			/* Free any current crash dump kernel before
 			 * we corrupt it.
 			 */
 			kimage_free(xchg(&kexec_crash_image, NULL));
-			result = kimage_crash_alloc(&image, entry, nr_segments, segments);
+			result = kimage_crash_alloc(&image, entry,
+						     nr_segments, segments);
 		}
-		if (result) {
+		if (result)
 			goto out;
-		}
+
 		result = machine_kexec_prepare(image);
-		if (result) {
+		if (result)
 			goto out;
-		}
-		for(i = 0; i < nr_segments; i++) {
+
+		for (i = 0; i < nr_segments; i++) {
 			result = kimage_load_segment(image, &image->segment[i]);
-			if (result) {
+			if (result)
 				goto out;
-			}
 		}
 		result = kimage_terminate(image);
-		if (result) {
+		if (result)
 			goto out;
-		}
 	}
 	/* Install the new kernel, and  Uninstall the old */
 	image = xchg(dest_image, image);
 
- out:
+out:
 	xchg(&kexec_lock, 0); /* Release the mutex */
 	kimage_free(image);
+
 	return result;
 }
 
 #ifdef CONFIG_COMPAT
 asmlinkage long compat_sys_kexec_load(unsigned long entry,
-	unsigned long nr_segments, struct compat_kexec_segment __user *segments,
-	unsigned long flags)
+				unsigned long nr_segments,
+				struct compat_kexec_segment __user *segments,
+				unsigned long flags)
 {
 	struct compat_kexec_segment in;
 	struct kexec_segment out, __user *ksegments;
@@ -989,20 +1011,17 @@ asmlinkage long compat_sys_kexec_load(unsigned long entry,
 	/* Don't allow clients that don't understand the native
 	 * architecture to do anything.
 	 */
-	if ((flags & KEXEC_ARCH_MASK) == KEXEC_ARCH_DEFAULT) {
+	if ((flags & KEXEC_ARCH_MASK) == KEXEC_ARCH_DEFAULT)
 		return -EINVAL;
-	}
 
-	if (nr_segments > KEXEC_SEGMENT_MAX) {
+	if (nr_segments > KEXEC_SEGMENT_MAX)
 		return -EINVAL;
-	}
 
 	ksegments = compat_alloc_user_space(nr_segments * sizeof(out));
 	for (i=0; i < nr_segments; i++) {
 		result = copy_from_user(&in, &segments[i], sizeof(in));
-		if (result) {
+		if (result)
 			return -EFAULT;
-		}
 
 		out.buf   = compat_ptr(in.buf);
 		out.bufsz = in.bufsz;
@@ -1010,9 +1029,8 @@ asmlinkage long compat_sys_kexec_load(unsigned long entry,
 		out.memsz = in.memsz;
 
 		result = copy_to_user(&ksegments[i], &out, sizeof(out));
-		if (result) {
+		if (result)
 			return -EFAULT;
-		}
 	}
 
 	return sys_kexec_load(entry, nr_segments, ksegments, flags);
-- 
cgit v1.2.3-70-g09d2


From 8c0e33c133021ee241e9d51255b9fb18eb34ef0e Mon Sep 17 00:00:00 2001
From: Nick Wilson <njw@osdl.org>
Date: Sat, 25 Jun 2005 14:59:00 -0700
Subject: [PATCH] Use ALIGN to remove duplicate code

This patch makes use of ALIGN() to remove duplicate round-up code.

Signed-off-by: Nick Wilson <njw@osdl.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/a.out.h | 2 +-
 kernel/resource.c     | 2 +-
 lib/bitmap.c          | 3 +--
 mm/bootmem.c          | 6 +++---
 4 files changed, 6 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/a.out.h b/include/linux/a.out.h
index af8a1dfa5c3..f913cc3e1b0 100644
--- a/include/linux/a.out.h
+++ b/include/linux/a.out.h
@@ -138,7 +138,7 @@ enum machine_type {
 #endif
 #endif
 
-#define _N_SEGMENT_ROUND(x) (((x) + SEGMENT_SIZE - 1) & ~(SEGMENT_SIZE - 1))
+#define _N_SEGMENT_ROUND(x) ALIGN(x, SEGMENT_SIZE)
 
 #define _N_TXTENDADDR(x) (N_TXTADDR(x)+(x).a_text)
 
diff --git a/kernel/resource.c b/kernel/resource.c
index 52f696f11ad..26967e04220 100644
--- a/kernel/resource.c
+++ b/kernel/resource.c
@@ -263,7 +263,7 @@ static int find_resource(struct resource *root, struct resource *new,
 			new->start = min;
 		if (new->end > max)
 			new->end = max;
-		new->start = (new->start + align - 1) & ~(align - 1);
+		new->start = ALIGN(new->start, align);
 		if (alignf)
 			alignf(alignf_data, new, size, align);
 		if (new->start < new->end && new->end - new->start >= size - 1) {
diff --git a/lib/bitmap.c b/lib/bitmap.c
index d1388a5ce89..fb9371fdd44 100644
--- a/lib/bitmap.c
+++ b/lib/bitmap.c
@@ -289,7 +289,6 @@ EXPORT_SYMBOL(__bitmap_weight);
 
 #define CHUNKSZ				32
 #define nbits_to_hold_value(val)	fls(val)
-#define roundup_power2(val,modulus)	(((val) + (modulus) - 1) & ~((modulus) - 1))
 #define unhex(c)			(isdigit(c) ? (c - '0') : (toupper(c) - 'A' + 10))
 #define BASEDEC 10		/* fancier cpuset lists input in decimal */
 
@@ -316,7 +315,7 @@ int bitmap_scnprintf(char *buf, unsigned int buflen,
 	if (chunksz == 0)
 		chunksz = CHUNKSZ;
 
-	i = roundup_power2(nmaskbits, CHUNKSZ) - CHUNKSZ;
+	i = ALIGN(nmaskbits, CHUNKSZ) - CHUNKSZ;
 	for (; i >= 0; i -= CHUNKSZ) {
 		chunkmask = ((1ULL << chunksz) - 1);
 		word = i / BITS_PER_LONG;
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 45275f1f894..c1330cc1978 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -65,7 +65,7 @@ static unsigned long __init init_bootmem_core (pg_data_t *pgdat,
 	pgdat->pgdat_next = pgdat_list;
 	pgdat_list = pgdat;
 
-	mapsize = (mapsize + (sizeof(long) - 1UL)) & ~(sizeof(long) - 1UL);
+	mapsize = ALIGN(mapsize, sizeof(long));
 	bdata->node_bootmem_map = phys_to_virt(mapstart << PAGE_SHIFT);
 	bdata->node_boot_start = (start << PAGE_SHIFT);
 	bdata->node_low_pfn = end;
@@ -186,7 +186,7 @@ __alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size,
 	} else
 		preferred = 0;
 
-	preferred = ((preferred + align - 1) & ~(align - 1)) >> PAGE_SHIFT;
+	preferred = ALIGN(preferred, align) >> PAGE_SHIFT;
 	preferred += offset;
 	areasize = (size+PAGE_SIZE-1)/PAGE_SIZE;
 	incr = align >> PAGE_SHIFT ? : 1;
@@ -227,7 +227,7 @@ found:
 	 */
 	if (align < PAGE_SIZE &&
 	    bdata->last_offset && bdata->last_pos+1 == start) {
-		offset = (bdata->last_offset+align-1) & ~(align-1);
+		offset = ALIGN(bdata->last_offset, align);
 		BUG_ON(offset > PAGE_SIZE);
 		remaining_size = PAGE_SIZE-offset;
 		if (size < remaining_size) {
-- 
cgit v1.2.3-70-g09d2


From 681ea4b930768444e9d88651c1362b0bf6d2a42b Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Sat, 25 Jun 2005 14:59:04 -0700
Subject: [PATCH] drivers/char/nvram.c: possible cleanups

This patch contains the following possible cleanups:
- make the needlessly global function __nvram_set_checksum static
- #if 0 the unused global function nvram_set_checksum
- remove the EXPORT_SYMBOL's for both functions

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/nvram.c  | 6 +++---
 include/linux/nvram.h | 2 --
 2 files changed, 3 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/drivers/char/nvram.c b/drivers/char/nvram.c
index f63a3fd7ca6..1af733d0732 100644
--- a/drivers/char/nvram.c
+++ b/drivers/char/nvram.c
@@ -211,12 +211,13 @@ nvram_check_checksum(void)
 	return rv;
 }
 
-void
+static void
 __nvram_set_checksum(void)
 {
 	mach_set_checksum();
 }
 
+#if 0
 void
 nvram_set_checksum(void)
 {
@@ -226,6 +227,7 @@ nvram_set_checksum(void)
 	__nvram_set_checksum();
 	spin_unlock_irqrestore(&rtc_lock, flags);
 }
+#endif  /*  0  */
 
 /*
  * The are the file operation function for user access to /dev/nvram
@@ -921,6 +923,4 @@ EXPORT_SYMBOL(__nvram_write_byte);
 EXPORT_SYMBOL(nvram_write_byte);
 EXPORT_SYMBOL(__nvram_check_checksum);
 EXPORT_SYMBOL(nvram_check_checksum);
-EXPORT_SYMBOL(__nvram_set_checksum);
-EXPORT_SYMBOL(nvram_set_checksum);
 MODULE_ALIAS_MISCDEV(NVRAM_MINOR);
diff --git a/include/linux/nvram.h b/include/linux/nvram.h
index b031e41b5e0..9189829c131 100644
--- a/include/linux/nvram.h
+++ b/include/linux/nvram.h
@@ -20,8 +20,6 @@ extern void __nvram_write_byte(unsigned char c, int i);
 extern void nvram_write_byte(unsigned char c, int i);
 extern int __nvram_check_checksum(void);
 extern int nvram_check_checksum(void);
-extern void __nvram_set_checksum(void);
-extern void nvram_set_checksum(void);
 #endif
 
 #endif  /* _LINUX_NVRAM_H */
-- 
cgit v1.2.3-70-g09d2


From 93d17d3d84b7147e8f07aeeb15ec01aa92c6b564 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Sat, 25 Jun 2005 14:59:10 -0700
Subject: [PATCH] drivers/block/ll_rw_blk.c: cleanups

This patch contains the following cleanups:
- make needlessly global code static
- remove the following unused global functions:
  - blkdev_scsi_issue_flush_fn
  - __blk_attempt_remerge
- remove the following unused EXPORT_SYMBOL's:
  - blk_phys_contig_segment
  - blk_hw_contig_segment
  - blkdev_scsi_issue_flush_fn
  - __blk_attempt_remerge

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/ll_rw_blk.c | 67 ++++++-----------------------------------------
 include/linux/blkdev.h    |  6 -----
 2 files changed, 8 insertions(+), 65 deletions(-)

(limited to 'include')

diff --git a/drivers/block/ll_rw_blk.c b/drivers/block/ll_rw_blk.c
index fd94ea27d59..fc86d53fe78 100644
--- a/drivers/block/ll_rw_blk.c
+++ b/drivers/block/ll_rw_blk.c
@@ -37,6 +37,7 @@
 
 static void blk_unplug_work(void *data);
 static void blk_unplug_timeout(unsigned long data);
+static void drive_stat_acct(struct request *rq, int nr_sectors, int new_io);
 
 /*
  * For the allocated request tables
@@ -1137,7 +1138,7 @@ new_hw_segment:
 }
 
 
-int blk_phys_contig_segment(request_queue_t *q, struct bio *bio,
+static int blk_phys_contig_segment(request_queue_t *q, struct bio *bio,
 				   struct bio *nxt)
 {
 	if (!(q->queue_flags & (1 << QUEUE_FLAG_CLUSTER)))
@@ -1158,9 +1159,7 @@ int blk_phys_contig_segment(request_queue_t *q, struct bio *bio,
 	return 0;
 }
 
-EXPORT_SYMBOL(blk_phys_contig_segment);
-
-int blk_hw_contig_segment(request_queue_t *q, struct bio *bio,
+static int blk_hw_contig_segment(request_queue_t *q, struct bio *bio,
 				 struct bio *nxt)
 {
 	if (unlikely(!bio_flagged(bio, BIO_SEG_VALID)))
@@ -1176,8 +1175,6 @@ int blk_hw_contig_segment(request_queue_t *q, struct bio *bio,
 	return 1;
 }
 
-EXPORT_SYMBOL(blk_hw_contig_segment);
-
 /*
  * map a request to scatterlist, return number of sg entries setup. Caller
  * must make sure sg can hold rq->nr_phys_segments entries
@@ -1825,7 +1822,7 @@ static inline int ioc_batching(request_queue_t *q, struct io_context *ioc)
  * is the behaviour we want though - once it gets a wakeup it should be given
  * a nice run.
  */
-void ioc_set_batching(request_queue_t *q, struct io_context *ioc)
+static void ioc_set_batching(request_queue_t *q, struct io_context *ioc)
 {
 	if (!ioc || ioc_batching(q, ioc))
 		return;
@@ -2254,45 +2251,7 @@ int blkdev_issue_flush(struct block_device *bdev, sector_t *error_sector)
 
 EXPORT_SYMBOL(blkdev_issue_flush);
 
-/**
- * blkdev_scsi_issue_flush_fn - issue flush for SCSI devices
- * @q:		device queue
- * @disk:	gendisk
- * @error_sector:	error offset
- *
- * Description:
- *    Devices understanding the SCSI command set, can use this function as
- *    a helper for issuing a cache flush. Note: driver is required to store
- *    the error offset (in case of error flushing) in ->sector of struct
- *    request.
- */
-int blkdev_scsi_issue_flush_fn(request_queue_t *q, struct gendisk *disk,
-			       sector_t *error_sector)
-{
-	struct request *rq = blk_get_request(q, WRITE, __GFP_WAIT);
-	int ret;
-
-	rq->flags |= REQ_BLOCK_PC | REQ_SOFTBARRIER;
-	rq->sector = 0;
-	memset(rq->cmd, 0, sizeof(rq->cmd));
-	rq->cmd[0] = 0x35;
-	rq->cmd_len = 12;
-	rq->data = NULL;
-	rq->data_len = 0;
-	rq->timeout = 60 * HZ;
-
-	ret = blk_execute_rq(q, disk, rq);
-
-	if (ret && error_sector)
-		*error_sector = rq->sector;
-
-	blk_put_request(rq);
-	return ret;
-}
-
-EXPORT_SYMBOL(blkdev_scsi_issue_flush_fn);
-
-void drive_stat_acct(struct request *rq, int nr_sectors, int new_io)
+static void drive_stat_acct(struct request *rq, int nr_sectors, int new_io)
 {
 	int rw = rq_data_dir(rq);
 
@@ -2551,16 +2510,6 @@ void blk_attempt_remerge(request_queue_t *q, struct request *rq)
 
 EXPORT_SYMBOL(blk_attempt_remerge);
 
-/*
- * Non-locking blk_attempt_remerge variant.
- */
-void __blk_attempt_remerge(request_queue_t *q, struct request *rq)
-{
-	attempt_back_merge(q, rq);
-}
-
-EXPORT_SYMBOL(__blk_attempt_remerge);
-
 static int __make_request(request_queue_t *q, struct bio *bio)
 {
 	struct request *req, *freereq = NULL;
@@ -2971,7 +2920,7 @@ void submit_bio(int rw, struct bio *bio)
 
 EXPORT_SYMBOL(submit_bio);
 
-void blk_recalc_rq_segments(struct request *rq)
+static void blk_recalc_rq_segments(struct request *rq)
 {
 	struct bio *bio, *prevbio = NULL;
 	int nr_phys_segs, nr_hw_segs;
@@ -3013,7 +2962,7 @@ void blk_recalc_rq_segments(struct request *rq)
 	rq->nr_hw_segments = nr_hw_segs;
 }
 
-void blk_recalc_rq_sectors(struct request *rq, int nsect)
+static void blk_recalc_rq_sectors(struct request *rq, int nsect)
 {
 	if (blk_fs_request(rq)) {
 		rq->hard_sector += nsect;
@@ -3601,7 +3550,7 @@ static struct sysfs_ops queue_sysfs_ops = {
 	.store	= queue_attr_store,
 };
 
-struct kobj_type queue_ktype = {
+static struct kobj_type queue_ktype = {
 	.sysfs_ops	= &queue_sysfs_ops,
 	.default_attrs	= default_attrs,
 };
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 60272141ff1..b54a0348a89 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -539,15 +539,12 @@ extern void generic_make_request(struct bio *bio);
 extern void blk_put_request(struct request *);
 extern void blk_end_sync_rq(struct request *rq);
 extern void blk_attempt_remerge(request_queue_t *, struct request *);
-extern void __blk_attempt_remerge(request_queue_t *, struct request *);
 extern struct request *blk_get_request(request_queue_t *, int, int);
 extern void blk_insert_request(request_queue_t *, struct request *, int, void *);
 extern void blk_requeue_request(request_queue_t *, struct request *);
 extern void blk_plug_device(request_queue_t *);
 extern int blk_remove_plug(request_queue_t *);
 extern void blk_recount_segments(request_queue_t *, struct bio *);
-extern int blk_phys_contig_segment(request_queue_t *q, struct bio *, struct bio *);
-extern int blk_hw_contig_segment(request_queue_t *q, struct bio *, struct bio *);
 extern int scsi_cmd_ioctl(struct file *, struct gendisk *, unsigned int, void __user *);
 extern void blk_start_queue(request_queue_t *q);
 extern void blk_stop_queue(request_queue_t *q);
@@ -631,7 +628,6 @@ extern void blk_queue_dma_alignment(request_queue_t *, int);
 extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev);
 extern void blk_queue_ordered(request_queue_t *, int);
 extern void blk_queue_issue_flush_fn(request_queue_t *, issue_flush_fn *);
-extern int blkdev_scsi_issue_flush_fn(request_queue_t *, struct gendisk *, sector_t *);
 extern struct request *blk_start_pre_flush(request_queue_t *,struct request *);
 extern int blk_complete_barrier_rq(request_queue_t *, struct request *, int);
 extern int blk_complete_barrier_rq_locked(request_queue_t *, struct request *, int);
@@ -675,8 +671,6 @@ extern int blkdev_issue_flush(struct block_device *, sector_t *);
 
 #define blkdev_entry_to_request(entry) list_entry((entry), struct request, queuelist)
 
-extern void drive_stat_acct(struct request *, int, int);
-
 static inline int queue_hardsect_size(request_queue_t *q)
 {
 	int retval = 512;
-- 
cgit v1.2.3-70-g09d2


From 3e1d1d28d99dabe63c64f7f40f1ca1d646de1f73 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <christoph@lameter.com>
Date: Fri, 24 Jun 2005 23:13:50 -0700
Subject: [PATCH] Cleanup patch for process freezing

1. Establish a simple API for process freezing defined in linux/include/sched.h:

   frozen(process)		Check for frozen process
   freezing(process)		Check if a process is being frozen
   freeze(process)		Tell a process to freeze (go to refrigerator)
   thaw_process(process)	Restart process
   frozen_process(process)	Process is frozen now

2. Remove all references to PF_FREEZE and PF_FROZEN from all
   kernel sources except sched.h

3. Fix numerous locations where try_to_freeze is manually done by a driver

4. Remove the argument that is no longer necessary from two function calls.

5. Some whitespace cleanup

6. Clear potential race in refrigerator (provides an open window of PF_FREEZE
   cleared before setting PF_FROZEN, recalc_sigpending does not check
   PF_FROZEN).

This patch does not address the problem of freeze_processes() violating the rule
that a task may only modify its own flags by setting PF_FREEZE. This is not clean
in an SMP environment. freeze(process) is therefore not SMP safe!

Signed-off-by: Christoph Lameter <christoph@lameter.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/power/kernel_threads.txt    |  3 +-
 Documentation/power/swsusp.txt            |  3 +-
 arch/frv/kernel/signal.c                  |  4 +-
 arch/h8300/kernel/signal.c                |  4 +-
 arch/i386/kernel/io_apic.c                |  2 +-
 arch/i386/kernel/signal.c                 |  4 +-
 arch/m32r/kernel/signal.c                 |  4 +-
 arch/ppc/kernel/signal.c                  |  3 +-
 arch/x86_64/kernel/signal.c               |  2 +-
 drivers/block/pktcdvd.c                   |  3 +-
 drivers/ieee1394/ieee1394_core.c          |  4 +-
 drivers/ieee1394/nodemgr.c                |  2 +-
 drivers/input/gameport/gameport.c         |  2 +-
 drivers/input/serio/serio.c               |  2 +-
 drivers/macintosh/therm_adt746x.c         |  4 +-
 drivers/md/md.c                           |  3 +-
 drivers/media/dvb/dvb-core/dvb_frontend.c |  3 +-
 drivers/media/video/msp3400.c             |  3 +-
 drivers/media/video/video-buf-dvb.c       |  3 +-
 drivers/net/8139too.c                     |  2 +-
 drivers/net/irda/sir_kthread.c            |  3 +-
 drivers/net/irda/stir4200.c               |  4 +-
 drivers/net/wireless/airo.c               |  2 +-
 drivers/pcmcia/cs.c                       |  2 +-
 drivers/pnp/pnpbios/core.c                |  2 +-
 drivers/usb/core/hub.c                    |  2 +-
 drivers/usb/gadget/file_storage.c         |  3 +-
 drivers/usb/storage/usb.c                 |  4 +-
 drivers/w1/w1.c                           |  4 +-
 fs/afs/kafsasyncd.c                       |  2 +-
 fs/afs/kafstimod.c                        |  2 +-
 fs/jbd/journal.c                          |  4 +-
 fs/jfs/jfs_logmgr.c                       |  4 +-
 fs/jfs/jfs_txnmgr.c                       |  8 ++--
 fs/lockd/clntproc.c                       |  2 +-
 fs/xfs/linux-2.6/xfs_buf.c                |  4 +-
 fs/xfs/linux-2.6/xfs_super.c              |  2 +-
 include/linux/sched.h                     | 73 +++++++++++++++++++++++++------
 kernel/power/process.c                    | 26 +++++------
 kernel/sched.c                            |  3 +-
 kernel/signal.c                           |  5 +--
 mm/pdflush.c                              |  2 +-
 mm/vmscan.c                               |  4 +-
 net/rxrpc/krxiod.c                        |  2 +-
 net/rxrpc/krxsecd.c                       |  2 +-
 net/rxrpc/krxtimod.c                      |  2 +-
 net/sunrpc/svcsock.c                      |  6 +--
 47 files changed, 126 insertions(+), 113 deletions(-)

(limited to 'include')

diff --git a/Documentation/power/kernel_threads.txt b/Documentation/power/kernel_threads.txt
index 60b548105ed..fb57784986b 100644
--- a/Documentation/power/kernel_threads.txt
+++ b/Documentation/power/kernel_threads.txt
@@ -12,8 +12,7 @@ refrigerator. Code to do this looks like this:
 	do {
 		hub_events();
 		wait_event_interruptible(khubd_wait, !list_empty(&hub_event_list));
-		if (current->flags & PF_FREEZE)
-			refrigerator(PF_FREEZE);
+		try_to_freeze();
 	} while (!signal_pending(current));
 
 from drivers/usb/core/hub.c::hub_thread()
diff --git a/Documentation/power/swsusp.txt b/Documentation/power/swsusp.txt
index c7c3459fde4..4e1627cc5b5 100644
--- a/Documentation/power/swsusp.txt
+++ b/Documentation/power/swsusp.txt
@@ -164,8 +164,7 @@ place where the thread is safe to be frozen (no kernel semaphores
 should be held at that point and it must be safe to sleep there), and
 add:
 
-            if (current->flags & PF_FREEZE)
-                    refrigerator(PF_FREEZE);
+            try_to_freeze();
 
 If the thread is needed for writing the image to storage, you should
 instead set the PF_NOFREEZE process flag when creating the thread.
diff --git a/arch/frv/kernel/signal.c b/arch/frv/kernel/signal.c
index d8d8f3d4304..36a2dffc8eb 100644
--- a/arch/frv/kernel/signal.c
+++ b/arch/frv/kernel/signal.c
@@ -536,10 +536,8 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
 	if (!user_mode(regs))
 		return 1;
 
-	if (current->flags & PF_FREEZE) {
-		refrigerator(0);
+	if (try_to_freeze())
 		goto no_signal;
-	}
 
 	if (!oldset)
 		oldset = &current->blocked;
diff --git a/arch/h8300/kernel/signal.c b/arch/h8300/kernel/signal.c
index a4799d633ef..5aab87eae1f 100644
--- a/arch/h8300/kernel/signal.c
+++ b/arch/h8300/kernel/signal.c
@@ -517,10 +517,8 @@ asmlinkage int do_signal(struct pt_regs *regs, sigset_t *oldset)
 	if ((regs->ccr & 0x10))
 		return 1;
 
-	if (current->flags & PF_FREEZE) {
-		refrigerator(0);
+	if (try_to_freeze())
 		goto no_signal;
-	}
 
 	current->thread.esp0 = (unsigned long) regs;
 
diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index 08540bc4ba3..2451a3a9944 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -573,7 +573,7 @@ static int balanced_irq(void *unused)
 	for ( ; ; ) {
 		set_current_state(TASK_INTERRUPTIBLE);
 		time_remaining = schedule_timeout(time_remaining);
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 		if (time_after(jiffies,
 				prev_balance_time+balanced_irq_interval)) {
 			do_irq_balance();
diff --git a/arch/i386/kernel/signal.c b/arch/i386/kernel/signal.c
index b9b8f4e20fa..ac5b1e975c5 100644
--- a/arch/i386/kernel/signal.c
+++ b/arch/i386/kernel/signal.c
@@ -608,10 +608,8 @@ int fastcall do_signal(struct pt_regs *regs, sigset_t *oldset)
 	if (!user_mode(regs))
 		return 1;
 
-	if (current->flags & PF_FREEZE) {
-		refrigerator(0);
+	if (try_to_freeze)
 		goto no_signal;
-	}
 
 	if (!oldset)
 		oldset = &current->blocked;
diff --git a/arch/m32r/kernel/signal.c b/arch/m32r/kernel/signal.c
index 50311eb07a2..5aef7e406ef 100644
--- a/arch/m32r/kernel/signal.c
+++ b/arch/m32r/kernel/signal.c
@@ -371,10 +371,8 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
 	if (!user_mode(regs))
 		return 1;
 
-	if (current->flags & PF_FREEZE) {
-		refrigerator(0);
+	if (try_to_freeze()) 
 		goto no_signal;
-	}
 
 	if (!oldset)
 		oldset = &current->blocked;
diff --git a/arch/ppc/kernel/signal.c b/arch/ppc/kernel/signal.c
index 7c8437da09d..8aaeb6f4e75 100644
--- a/arch/ppc/kernel/signal.c
+++ b/arch/ppc/kernel/signal.c
@@ -705,8 +705,7 @@ int do_signal(sigset_t *oldset, struct pt_regs *regs)
 	unsigned long frame, newsp;
 	int signr, ret;
 
-	if (current->flags & PF_FREEZE) {
-		refrigerator(PF_FREEZE);
+	if (try_to_freeze()) {
 		signr = 0;
 		if (!signal_pending(current))
 			goto no_signal;
diff --git a/arch/x86_64/kernel/signal.c b/arch/x86_64/kernel/signal.c
index 98b7ba95d58..98590a989f3 100644
--- a/arch/x86_64/kernel/signal.c
+++ b/arch/x86_64/kernel/signal.c
@@ -425,7 +425,7 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
 	if (!user_mode(regs))
 		return 1;
 
-	if (try_to_freeze(0))
+	if (try_to_freeze())
 		goto no_signal;
 
 	if (!oldset)
diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index 7f3d78de265..7b838342f0a 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -1251,8 +1251,7 @@ static int kcdrwd(void *foobar)
 			VPRINTK("kcdrwd: wake up\n");
 
 			/* make swsusp happy with our thread */
-			if (current->flags & PF_FREEZE)
-				refrigerator(PF_FREEZE);
+			try_to_freeze();
 
 			list_for_each_entry(pkt, &pd->cdrw.pkt_active_list, list) {
 				if (!pkt->sleep_time)
diff --git a/drivers/ieee1394/ieee1394_core.c b/drivers/ieee1394/ieee1394_core.c
index 2d9a9b74e68..629070b83a3 100644
--- a/drivers/ieee1394/ieee1394_core.c
+++ b/drivers/ieee1394/ieee1394_core.c
@@ -1041,10 +1041,8 @@ static int hpsbpkt_thread(void *__hi)
 
 	while (1) {
 		if (down_interruptible(&khpsbpkt_sig)) {
-			if (current->flags & PF_FREEZE) {
-				refrigerator(0);
+			if (try_to_freeze())
 				continue;
-			}
 			printk("khpsbpkt: received unexpected signal?!\n" );
 			break;
 		}
diff --git a/drivers/ieee1394/nodemgr.c b/drivers/ieee1394/nodemgr.c
index 32abb6dda88..9a46c3b44bf 100644
--- a/drivers/ieee1394/nodemgr.c
+++ b/drivers/ieee1394/nodemgr.c
@@ -1510,7 +1510,7 @@ static int nodemgr_host_thread(void *__hi)
 
 		if (down_interruptible(&hi->reset_sem) ||
 		    down_interruptible(&nodemgr_serialize)) {
-			if (try_to_freeze(PF_FREEZE))
+			if (try_to_freeze())
 				continue;
 			printk("NodeMgr: received unexpected signal?!\n" );
 			break;
diff --git a/drivers/input/gameport/gameport.c b/drivers/input/gameport/gameport.c
index e152d0fa0cd..c77a82e4605 100644
--- a/drivers/input/gameport/gameport.c
+++ b/drivers/input/gameport/gameport.c
@@ -439,7 +439,7 @@ static int gameport_thread(void *nothing)
 	do {
 		gameport_handle_events();
 		wait_event_interruptible(gameport_wait, !list_empty(&gameport_event_list));
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 	} while (!signal_pending(current));
 
 	printk(KERN_DEBUG "gameport: kgameportd exiting\n");
diff --git a/drivers/input/serio/serio.c b/drivers/input/serio/serio.c
index feab4970406..341824c4852 100644
--- a/drivers/input/serio/serio.c
+++ b/drivers/input/serio/serio.c
@@ -344,7 +344,7 @@ static int serio_thread(void *nothing)
 	do {
 		serio_handle_events();
 		wait_event_interruptible(serio_wait, !list_empty(&serio_event_list));
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 	} while (!signal_pending(current));
 
 	printk(KERN_DEBUG "serio: kseriod exiting\n");
diff --git a/drivers/macintosh/therm_adt746x.c b/drivers/macintosh/therm_adt746x.c
index 5ba190ce14a..c9ca1118e44 100644
--- a/drivers/macintosh/therm_adt746x.c
+++ b/drivers/macintosh/therm_adt746x.c
@@ -328,9 +328,7 @@ static int monitor_task(void *arg)
 	struct thermostat* th = arg;
 
 	while(!kthread_should_stop()) {
-		if (current->flags & PF_FREEZE)
-			refrigerator(PF_FREEZE);
-
+		try_to_freeze();
 		msleep_interruptible(2000);
 
 #ifndef DEBUG
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 0c6b5b6baff..3802f7a17f1 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -2976,8 +2976,7 @@ static int md_thread(void * arg)
 		wait_event_interruptible_timeout(thread->wqueue,
 						 test_bit(THREAD_WAKEUP, &thread->flags),
 						 thread->timeout);
-		if (current->flags & PF_FREEZE)
-			refrigerator(PF_FREEZE);
+		try_to_freeze();
 
 		clear_bit(THREAD_WAKEUP, &thread->flags);
 
diff --git a/drivers/media/dvb/dvb-core/dvb_frontend.c b/drivers/media/dvb/dvb-core/dvb_frontend.c
index d6b7a9de471..f11daae91cd 100644
--- a/drivers/media/dvb/dvb-core/dvb_frontend.c
+++ b/drivers/media/dvb/dvb-core/dvb_frontend.c
@@ -391,8 +391,7 @@ static int dvb_frontend_thread(void *data)
 			break;
 		}
 
-		if (current->flags & PF_FREEZE)
-			refrigerator(PF_FREEZE);
+		try_to_freeze();
 
 		if (down_interruptible(&fepriv->sem))
 			break;
diff --git a/drivers/media/video/msp3400.c b/drivers/media/video/msp3400.c
index 1b7d38e96f1..b4ee9dfe6d4 100644
--- a/drivers/media/video/msp3400.c
+++ b/drivers/media/video/msp3400.c
@@ -750,8 +750,7 @@ static int msp34xx_sleep(struct msp3400c *msp, int timeout)
 #endif
 		}
 	}
-	if (current->flags & PF_FREEZE)
-		refrigerator(PF_FREEZE);
+	try_to_freeze();
 	remove_wait_queue(&msp->wq, &wait);
 	return msp->restart;
 }
diff --git a/drivers/media/video/video-buf-dvb.c b/drivers/media/video/video-buf-dvb.c
index 5f870075b55..15f5bb48696 100644
--- a/drivers/media/video/video-buf-dvb.c
+++ b/drivers/media/video/video-buf-dvb.c
@@ -62,8 +62,7 @@ static int videobuf_dvb_thread(void *data)
 			break;
 		if (kthread_should_stop())
 			break;
-		if (current->flags & PF_FREEZE)
-			refrigerator(PF_FREEZE);
+		try_to_freeze();
 
 		/* feed buffer data to demux */
 		if (buf->state == STATE_DONE)
diff --git a/drivers/net/8139too.c b/drivers/net/8139too.c
index 047202c4d9a..5a4a08a7c95 100644
--- a/drivers/net/8139too.c
+++ b/drivers/net/8139too.c
@@ -1606,7 +1606,7 @@ static int rtl8139_thread (void *data)
 		do {
 			timeout = interruptible_sleep_on_timeout (&tp->thr_wait, timeout);
 			/* make swsusp happy with our thread */
-			try_to_freeze(PF_FREEZE);
+			try_to_freeze();
 		} while (!signal_pending (current) && (timeout > 0));
 
 		if (signal_pending (current)) {
diff --git a/drivers/net/irda/sir_kthread.c b/drivers/net/irda/sir_kthread.c
index 18cea109953..c65054364bc 100644
--- a/drivers/net/irda/sir_kthread.c
+++ b/drivers/net/irda/sir_kthread.c
@@ -135,8 +135,7 @@ static int irda_thread(void *startup)
 		remove_wait_queue(&irda_rq_queue.kick, &wait);
 
 		/* make swsusp happy with our thread */
-		if (current->flags & PF_FREEZE)
-			refrigerator(PF_FREEZE);
+		try_to_freeze();
 
 		run_irda_queue();
 	}
diff --git a/drivers/net/irda/stir4200.c b/drivers/net/irda/stir4200.c
index 66f488c1371..15f207323d9 100644
--- a/drivers/net/irda/stir4200.c
+++ b/drivers/net/irda/stir4200.c
@@ -763,7 +763,7 @@ static int stir_transmit_thread(void *arg)
 	{
 #ifdef CONFIG_PM
 		/* if suspending, then power off and wait */
-		if (unlikely(current->flags & PF_FREEZE)) {
+		if (unlikely(freezing(current))) {
 			if (stir->receiving)
 				receive_stop(stir);
 			else
@@ -771,7 +771,7 @@ static int stir_transmit_thread(void *arg)
 
 			write_reg(stir, REG_CTRL1, CTRL1_TXPWD|CTRL1_RXPWD);
 
-			refrigerator(PF_FREEZE);
+			refrigerator();
 
 			if (change_speed(stir, stir->speed))
 				break;
diff --git a/drivers/net/wireless/airo.c b/drivers/net/wireless/airo.c
index fb10a2db63a..d72e0385e4f 100644
--- a/drivers/net/wireless/airo.c
+++ b/drivers/net/wireless/airo.c
@@ -2918,7 +2918,7 @@ static int airo_thread(void *data) {
 			flush_signals(current);
 
 		/* make swsusp happy with our thread */
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 
 		if (test_bit(JOB_DIE, &ai->flags))
 			break;
diff --git a/drivers/pcmcia/cs.c b/drivers/pcmcia/cs.c
index d136b3c8fac..48e4f04530d 100644
--- a/drivers/pcmcia/cs.c
+++ b/drivers/pcmcia/cs.c
@@ -718,7 +718,7 @@ static int pccardd(void *__skt)
 		}
 
 		schedule();
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 
 		if (!skt->thread)
 			break;
diff --git a/drivers/pnp/pnpbios/core.c b/drivers/pnp/pnpbios/core.c
index e939c93a931..778a324028f 100644
--- a/drivers/pnp/pnpbios/core.c
+++ b/drivers/pnp/pnpbios/core.c
@@ -182,7 +182,7 @@ static int pnp_dock_thread(void * unused)
 		msleep_interruptible(2000);
 
 		if(signal_pending(current)) {
-			if (try_to_freeze(PF_FREEZE))
+			if (try_to_freeze())
 				continue;
 			break;
 		}
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index d2d648ee864..a8d879a85d0 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -2808,7 +2808,7 @@ static int hub_thread(void *__unused)
 	do {
 		hub_events();
 		wait_event_interruptible(khubd_wait, !list_empty(&hub_event_list)); 
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 	} while (!signal_pending(current));
 
 	pr_debug ("%s: khubd exiting\n", usbcore_name);
diff --git a/drivers/usb/gadget/file_storage.c b/drivers/usb/gadget/file_storage.c
index 037a7f16382..a9be85103d2 100644
--- a/drivers/usb/gadget/file_storage.c
+++ b/drivers/usb/gadget/file_storage.c
@@ -1554,8 +1554,7 @@ static int sleep_thread(struct fsg_dev *fsg)
 	rc = wait_event_interruptible(fsg->thread_wqh,
 			fsg->thread_wakeup_needed);
 	fsg->thread_wakeup_needed = 0;
-	if (current->flags & PF_FREEZE)
-		refrigerator(PF_FREEZE);
+	try_to_freeze();
 	return (rc ? -EINTR : 0);
 }
 
diff --git a/drivers/usb/storage/usb.c b/drivers/usb/storage/usb.c
index 35c1ca6b5a8..77e7fc258aa 100644
--- a/drivers/usb/storage/usb.c
+++ b/drivers/usb/storage/usb.c
@@ -847,10 +847,8 @@ retry:
 		wait_event_interruptible_timeout(us->delay_wait,
 				test_bit(US_FLIDX_DISCONNECTING, &us->flags),
 				delay_use * HZ);
-		if (current->flags & PF_FREEZE) {
-			refrigerator(PF_FREEZE);
+		if (try_to_freeze())
 			goto retry;
-		}
 	}
 
 	/* If the device is still connected, perform the scanning */
diff --git a/drivers/w1/w1.c b/drivers/w1/w1.c
index b460927ec32..312cf3220f1 100644
--- a/drivers/w1/w1.c
+++ b/drivers/w1/w1.c
@@ -646,7 +646,7 @@ static int w1_control(void *data)
 	while (!control_needs_exit || have_to_wait) {
 		have_to_wait = 0;
 
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 		msleep_interruptible(w1_timeout * 1000);
 
 		if (signal_pending(current))
@@ -725,7 +725,7 @@ int w1_process(void *data)
 	allow_signal(SIGTERM);
 
 	while (!test_bit(W1_MASTER_NEED_EXIT, &dev->flags)) {
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 		msleep_interruptible(w1_timeout * 1000);
 
 		if (signal_pending(current))
diff --git a/fs/afs/kafsasyncd.c b/fs/afs/kafsasyncd.c
index 6fc88ae8ad9..7ac07d0d47b 100644
--- a/fs/afs/kafsasyncd.c
+++ b/fs/afs/kafsasyncd.c
@@ -116,7 +116,7 @@ static int kafsasyncd(void *arg)
 		remove_wait_queue(&kafsasyncd_sleepq, &myself);
 		set_current_state(TASK_RUNNING);
 
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 
 		/* discard pending signals */
 		afs_discard_my_signals();
diff --git a/fs/afs/kafstimod.c b/fs/afs/kafstimod.c
index 86e710dd057..65bc05ab818 100644
--- a/fs/afs/kafstimod.c
+++ b/fs/afs/kafstimod.c
@@ -91,7 +91,7 @@ static int kafstimod(void *arg)
 			complete_and_exit(&kafstimod_dead, 0);
 		}
 
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 
 		/* discard pending signals */
 		afs_discard_my_signals();
diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index 1e6f2e2ad4a..5e7b4394951 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -167,7 +167,7 @@ loop:
 	}
 
 	wake_up(&journal->j_wait_done_commit);
-	if (current->flags & PF_FREEZE) {
+	if (freezing(current)) {
 		/*
 		 * The simpler the better. Flushing journal isn't a
 		 * good idea, because that depends on threads that may
@@ -175,7 +175,7 @@ loop:
 		 */
 		jbd_debug(1, "Now suspending kjournald\n");
 		spin_unlock(&journal->j_state_lock);
-		refrigerator(PF_FREEZE);
+		refrigerator();
 		spin_lock(&journal->j_state_lock);
 	} else {
 		/*
diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c
index 7c8387ed419..79d07624bfe 100644
--- a/fs/jfs/jfs_logmgr.c
+++ b/fs/jfs/jfs_logmgr.c
@@ -2359,9 +2359,9 @@ int jfsIOWait(void *arg)
 			lbmStartIO(bp);
 			spin_lock_irq(&log_redrive_lock);
 		}
-		if (current->flags & PF_FREEZE) {
+		if (freezing(current)) {
 			spin_unlock_irq(&log_redrive_lock);
-			refrigerator(PF_FREEZE);
+			refrigerator();
 		} else {
 			add_wait_queue(&jfs_IO_thread_wait, &wq);
 			set_current_state(TASK_INTERRUPTIBLE);
diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c
index 8cbaaff1d5f..121c981ff45 100644
--- a/fs/jfs/jfs_txnmgr.c
+++ b/fs/jfs/jfs_txnmgr.c
@@ -2788,9 +2788,9 @@ int jfs_lazycommit(void *arg)
 		/* In case a wakeup came while all threads were active */
 		jfs_commit_thread_waking = 0;
 
-		if (current->flags & PF_FREEZE) {
+		if (freezing(current)) {
 			LAZY_UNLOCK(flags);
-			refrigerator(PF_FREEZE);
+			refrigerator();
 		} else {
 			DECLARE_WAITQUEUE(wq, current);
 
@@ -2987,9 +2987,9 @@ int jfs_sync(void *arg)
 		/* Add anon_list2 back to anon_list */
 		list_splice_init(&TxAnchor.anon_list2, &TxAnchor.anon_list);
 
-		if (current->flags & PF_FREEZE) {
+		if (freezing(current)) {
 			TXN_UNLOCK();
-			refrigerator(PF_FREEZE);
+			refrigerator();
 		} else {
 			DECLARE_WAITQUEUE(wq, current);
 
diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c
index fd77ed1d710..14b3ce87fa2 100644
--- a/fs/lockd/clntproc.c
+++ b/fs/lockd/clntproc.c
@@ -313,7 +313,7 @@ static int nlm_wait_on_grace(wait_queue_head_t *queue)
 	prepare_to_wait(queue, &wait, TASK_INTERRUPTIBLE);
 	if (!signalled ()) {
 		schedule_timeout(NLMCLNT_GRACE_WAIT);
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 		if (!signalled ())
 			status = 0;
 	}
diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index c60e69431e1..df0cba239dd 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -1771,9 +1771,9 @@ xfsbufd(
 
 	INIT_LIST_HEAD(&tmp);
 	do {
-		if (unlikely(current->flags & PF_FREEZE)) {
+		if (unlikely(freezing(current))) {
 			xfsbufd_force_sleep = 1;
-			refrigerator(PF_FREEZE);
+			refrigerator();
 		} else {
 			xfsbufd_force_sleep = 0;
 		}
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 5fe9af38aa2..f6dd7de2592 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -483,7 +483,7 @@ xfssyncd(
 		set_current_state(TASK_INTERRUPTIBLE);
 		timeleft = schedule_timeout(timeleft);
 		/* swsusp */
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 		if (vfsp->vfs_flag & VFS_UMOUNT)
 			break;
 
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2c69682b044..e7fd09b0557 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1245,33 +1245,78 @@ extern void normalize_rt_tasks(void);
 
 #endif
 
-/* try_to_freeze
- *
- * Checks whether we need to enter the refrigerator
- * and returns 1 if we did so.
- */
 #ifdef CONFIG_PM
-extern void refrigerator(unsigned long);
+/*
+ * Check if a process has been frozen
+ */
+static inline int frozen(struct task_struct *p)
+{
+	return p->flags & PF_FROZEN;
+}
+
+/*
+ * Check if there is a request to freeze a process
+ */
+static inline int freezing(struct task_struct *p)
+{
+	return p->flags & PF_FREEZE;
+}
+
+/*
+ * Request that a process be frozen
+ * FIXME: SMP problem. We may not modify other process' flags!
+ */
+static inline void freeze(struct task_struct *p)
+{
+	p->flags |= PF_FREEZE;
+}
+
+/*
+ * Wake up a frozen process
+ */
+static inline int thaw_process(struct task_struct *p)
+{
+	if (frozen(p)) {
+		p->flags &= ~PF_FROZEN;
+		wake_up_process(p);
+		return 1;
+	}
+	return 0;
+}
+
+/*
+ * freezing is complete, mark process as frozen
+ */
+static inline void frozen_process(struct task_struct *p)
+{
+	p->flags = (p->flags & ~PF_FREEZE) | PF_FROZEN;
+}
+
+extern void refrigerator(void);
 extern int freeze_processes(void);
 extern void thaw_processes(void);
 
-static inline int try_to_freeze(unsigned long refrigerator_flags)
+static inline int try_to_freeze(void)
 {
-	if (unlikely(current->flags & PF_FREEZE)) {
-		refrigerator(refrigerator_flags);
+	if (freezing(current)) {
+		refrigerator();
 		return 1;
 	} else
 		return 0;
 }
 #else
-static inline void refrigerator(unsigned long flag) {}
+static inline int frozen(struct task_struct *p) { return 0; }
+static inline int freezing(struct task_struct *p) { return 0; }
+static inline void freeze(struct task_struct *p) { BUG(); }
+static inline int thaw_process(struct task_struct *p) { return 1; }
+static inline void frozen_process(struct task_struct *p) { BUG(); }
+
+static inline void refrigerator(void) {}
 static inline int freeze_processes(void) { BUG(); return 0; }
 static inline void thaw_processes(void) {}
 
-static inline int try_to_freeze(unsigned long refrigerator_flags)
-{
-	return 0;
-}
+static inline int try_to_freeze(void) { return 0; }
+
 #endif /* CONFIG_PM */
 #endif /* __KERNEL__ */
 
diff --git a/kernel/power/process.c b/kernel/power/process.c
index 78d92dc6a1e..0a086640bcf 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -32,7 +32,7 @@ static inline int freezeable(struct task_struct * p)
 }
 
 /* Refrigerator is place where frozen processes are stored :-). */
-void refrigerator(unsigned long flag)
+void refrigerator(void)
 {
 	/* Hmm, should we be allowed to suspend when there are realtime
 	   processes around? */
@@ -41,14 +41,13 @@ void refrigerator(unsigned long flag)
 	current->state = TASK_UNINTERRUPTIBLE;
 	pr_debug("%s entered refrigerator\n", current->comm);
 	printk("=");
-	current->flags &= ~PF_FREEZE;
 
+	frozen_process(current);
 	spin_lock_irq(&current->sighand->siglock);
 	recalc_sigpending(); /* We sent fake signal, clean it up */
 	spin_unlock_irq(&current->sighand->siglock);
 
-	current->flags |= PF_FROZEN;
-	while (current->flags & PF_FROZEN)
+	while (frozen(current))
 		schedule();
 	pr_debug("%s left refrigerator\n", current->comm);
 	current->state = save;
@@ -57,10 +56,10 @@ void refrigerator(unsigned long flag)
 /* 0 = success, else # of processes that we failed to stop */
 int freeze_processes(void)
 {
-       int todo;
-       unsigned long start_time;
+	int todo;
+	unsigned long start_time;
 	struct task_struct *g, *p;
-	
+
 	printk( "Stopping tasks: " );
 	start_time = jiffies;
 	do {
@@ -70,14 +69,12 @@ int freeze_processes(void)
 			unsigned long flags;
 			if (!freezeable(p))
 				continue;
-			if ((p->flags & PF_FROZEN) ||
+			if ((frozen(p)) ||
 			    (p->state == TASK_TRACED) ||
 			    (p->state == TASK_STOPPED))
 				continue;
 
-			/* FIXME: smp problem here: we may not access other process' flags
-			   without locking */
-			p->flags |= PF_FREEZE;
+			freeze(p);
 			spin_lock_irqsave(&p->sighand->siglock, flags);
 			signal_wake_up(p, 0);
 			spin_unlock_irqrestore(&p->sighand->siglock, flags);
@@ -91,7 +88,7 @@ int freeze_processes(void)
 			return todo;
 		}
 	} while(todo);
-	
+
 	printk( "|\n" );
 	BUG_ON(in_atomic());
 	return 0;
@@ -106,10 +103,7 @@ void thaw_processes(void)
 	do_each_thread(g, p) {
 		if (!freezeable(p))
 			continue;
-		if (p->flags & PF_FROZEN) {
-			p->flags &= ~PF_FROZEN;
-			wake_up_process(p);
-		} else
+		if (!thaw_process(p))
 			printk(KERN_INFO " Strange, %s not stopped\n", p->comm );
 	} while_each_thread(g, p);
 
diff --git a/kernel/sched.c b/kernel/sched.c
index 76080d142e3..6fa9ea4ae44 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4174,8 +4174,7 @@ static int migration_thread(void * data)
 		struct list_head *head;
 		migration_req_t *req;
 
-		if (current->flags & PF_FREEZE)
-			refrigerator(PF_FREEZE);
+		try_to_freeze();
 
 		spin_lock_irq(&rq->lock);
 
diff --git a/kernel/signal.c b/kernel/signal.c
index d1258729a5f..ca1186eef93 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -213,7 +213,7 @@ static inline int has_pending_signals(sigset_t *signal, sigset_t *blocked)
 fastcall void recalc_sigpending_tsk(struct task_struct *t)
 {
 	if (t->signal->group_stop_count > 0 ||
-	    (t->flags & PF_FREEZE) ||
+	    (freezing(t)) ||
 	    PENDING(&t->pending, &t->blocked) ||
 	    PENDING(&t->signal->shared_pending, &t->blocked))
 		set_tsk_thread_flag(t, TIF_SIGPENDING);
@@ -2231,8 +2231,7 @@ sys_rt_sigtimedwait(const sigset_t __user *uthese,
 			current->state = TASK_INTERRUPTIBLE;
 			timeout = schedule_timeout(timeout);
 
-			if (current->flags & PF_FREEZE)
-				refrigerator(PF_FREEZE);
+			try_to_freeze();
 			spin_lock_irq(&current->sighand->siglock);
 			sig = dequeue_signal(current, &these, &info);
 			current->blocked = current->real_blocked;
diff --git a/mm/pdflush.c b/mm/pdflush.c
index 38ce279cc8c..d6781951267 100644
--- a/mm/pdflush.c
+++ b/mm/pdflush.c
@@ -105,7 +105,7 @@ static int __pdflush(struct pdflush_work *my_work)
 		spin_unlock_irq(&pdflush_lock);
 
 		schedule();
-		if (try_to_freeze(PF_FREEZE)) {
+		if (try_to_freeze()) {
 			spin_lock_irq(&pdflush_lock);
 			continue;
 		}
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 4b8e62a1937..1fa312a8db7 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1216,8 +1216,8 @@ static int kswapd(void *p)
 	order = 0;
 	for ( ; ; ) {
 		unsigned long new_order;
-		if (current->flags & PF_FREEZE)
-			refrigerator(PF_FREEZE);
+
+		try_to_freeze();
 
 		prepare_to_wait(&pgdat->kswapd_wait, &wait, TASK_INTERRUPTIBLE);
 		new_order = pgdat->kswapd_max_order;
diff --git a/net/rxrpc/krxiod.c b/net/rxrpc/krxiod.c
index 2b537f425a1..dada34a77b2 100644
--- a/net/rxrpc/krxiod.c
+++ b/net/rxrpc/krxiod.c
@@ -138,7 +138,7 @@ static int rxrpc_krxiod(void *arg)
 
 		_debug("### End Work");
 
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 
                 /* discard pending signals */
 		rxrpc_discard_my_signals();
diff --git a/net/rxrpc/krxsecd.c b/net/rxrpc/krxsecd.c
index 6020c89d922..1aadd026d35 100644
--- a/net/rxrpc/krxsecd.c
+++ b/net/rxrpc/krxsecd.c
@@ -107,7 +107,7 @@ static int rxrpc_krxsecd(void *arg)
 
 		_debug("### End Inbound Calls");
 
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 
                 /* discard pending signals */
 		rxrpc_discard_my_signals();
diff --git a/net/rxrpc/krxtimod.c b/net/rxrpc/krxtimod.c
index 249c2b0290b..3ac81cdd121 100644
--- a/net/rxrpc/krxtimod.c
+++ b/net/rxrpc/krxtimod.c
@@ -90,7 +90,7 @@ static int krxtimod(void *arg)
 			complete_and_exit(&krxtimod_dead, 0);
 		}
 
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 
 		/* discard pending signals */
 		rxrpc_discard_my_signals();
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 05907035bc9..56db8f13e6c 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -1185,8 +1185,8 @@ svc_recv(struct svc_serv *serv, struct svc_rqst *rqstp, long timeout)
 	arg->page_len = (pages-2)*PAGE_SIZE;
 	arg->len = (pages-1)*PAGE_SIZE;
 	arg->tail[0].iov_len = 0;
-	
-	try_to_freeze(PF_FREEZE);
+
+	try_to_freeze();
 	if (signalled())
 		return -EINTR;
 
@@ -1227,7 +1227,7 @@ svc_recv(struct svc_serv *serv, struct svc_rqst *rqstp, long timeout)
 
 		schedule_timeout(timeout);
 
-		try_to_freeze(PF_FREEZE);
+		try_to_freeze();
 
 		spin_lock_bh(&serv->sv_lock);
 		remove_wait_queue(&rqstp->rq_wait, &wait);
-- 
cgit v1.2.3-70-g09d2


From 91f6a589fb6bb4ce6b6b196e910d3b907e1d0a40 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Sun, 26 Jun 2005 22:24:13 +0100
Subject: [PATCH] ARM: 2753/1: move ixdp* cpld mappings

Patch from Lennert Buytenhek

All ixdp platforms currently have a cpld mapped in at 0xfafff000.
Since this address is not 1M-aligned, a regular page mapping will be
used instead of a section mapping, which opens up the possibility of
triggering ixp2400 erratum #66 as we only do the XCB=101 workaround
thing for section mappings.
There is still a lot of space higher up in the virtual memory map
for 1M mappings, so move the cpld mapping to 0xfe000000 and make it
1M big so that a section mapping will be used for it.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Deepak Saxena
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-ixp2000/ixdp2x00.h | 4 ++--
 include/asm-arm/arch-ixp2000/ixdp2x01.h | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-ixp2000/ixdp2x00.h b/include/asm-arm/arch-ixp2000/ixdp2x00.h
index 3a398dfbf12..229381c6428 100644
--- a/include/asm-arm/arch-ixp2000/ixdp2x00.h
+++ b/include/asm-arm/arch-ixp2000/ixdp2x00.h
@@ -21,8 +21,8 @@
  * On board CPLD memory map
  */
 #define IXDP2X00_PHYS_CPLD_BASE		0xc7000000
-#define IXDP2X00_VIRT_CPLD_BASE		0xfafff000
-#define IXDP2X00_CPLD_SIZE		0x00001000
+#define IXDP2X00_VIRT_CPLD_BASE		0xfe000000
+#define IXDP2X00_CPLD_SIZE		0x00100000
 
 
 #define IXDP2X00_CPLD_REG(x)  	\
diff --git a/include/asm-arm/arch-ixp2000/ixdp2x01.h b/include/asm-arm/arch-ixp2000/ixdp2x01.h
index b3a1bcda8d0..b768009c3a5 100644
--- a/include/asm-arm/arch-ixp2000/ixdp2x01.h
+++ b/include/asm-arm/arch-ixp2000/ixdp2x01.h
@@ -18,8 +18,8 @@
 #define __IXDP2X01_H__
 
 #define	IXDP2X01_PHYS_CPLD_BASE		0xc6024000
-#define	IXDP2X01_VIRT_CPLD_BASE		0xfafff000
-#define	IXDP2X01_CPLD_REGION_SIZE	0x00001000
+#define	IXDP2X01_VIRT_CPLD_BASE		0xfe000000
+#define	IXDP2X01_CPLD_REGION_SIZE	0x00100000
 
 #define IXDP2X01_CPLD_VIRT_REG(reg) (volatile unsigned long*)(IXDP2X01_VIRT_CPLD_BASE | reg)
 #define IXDP2X01_CPLD_PHYS_REG(reg) (volatile u32*)(IXDP2X01_PHYS_CPLD_BASE | reg)
-- 
cgit v1.2.3-70-g09d2


From 15aca90363719f11f200f510a13d18bda9dea789 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Sun, 26 Jun 2005 22:24:14 +0100
Subject: [PATCH] ARM: 2754/1: move ixp2000 VMALLOC_END up

Patch from Lennert Buytenhek

As the ixdp cpld mappings now live at 0xfe000000, we can push
VMALLOC_END upwards to 0xfb000000, where the first iotable mapping
begins.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Deepak Saxena
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-ixp2000/vmalloc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-ixp2000/vmalloc.h b/include/asm-arm/arch-ixp2000/vmalloc.h
index 473dff4ec56..275136963a0 100644
--- a/include/asm-arm/arch-ixp2000/vmalloc.h
+++ b/include/asm-arm/arch-ixp2000/vmalloc.h
@@ -17,4 +17,4 @@
  * The vmalloc() routines leaves a hole of 4kB between each vmalloced
  * area for the same reason. ;)
  */
-#define VMALLOC_END	    0xfaffefff
+#define VMALLOC_END	    0xfb000000
-- 
cgit v1.2.3-70-g09d2


From 67603be482a0361a13c2238b79172849ee2e307c Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Sun, 26 Jun 2005 22:24:16 +0100
Subject: [PATCH] ARM: 2755/1: describe ixp2000 virtual memory map layout

Patch from Lennert Buytenhek

Add a comment to asm/arch-ixp2000/ixp2000-regs.h describing the
ixp2000 virtual memory map layout.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Deepak Saxena
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-ixp2000/ixp2000-regs.h | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

(limited to 'include')

diff --git a/include/asm-arm/arch-ixp2000/ixp2000-regs.h b/include/asm-arm/arch-ixp2000/ixp2000-regs.h
index 5eb47d4bfbf..213c0cfbb15 100644
--- a/include/asm-arm/arch-ixp2000/ixp2000-regs.h
+++ b/include/asm-arm/arch-ixp2000/ixp2000-regs.h
@@ -18,6 +18,21 @@
 #ifndef _IXP2000_REGS_H_
 #define _IXP2000_REGS_H_
 
+/*
+ * IXP2000 linux memory map:
+ *
+ * virt		phys		size
+ * fb000000	db000000	16M		PCI CFG1
+ * fc000000	da000000	16M		PCI CFG0
+ * fd000000	d8000000	16M		PCI I/O
+ * fe[0-7]00000			8M		per-platform mappings
+ * feb00000	c8000000	1M		MSF
+ * fec00000	df000000	1M		PCI CSRs
+ * fed00000	de000000	1M		PCI CREG
+ * fee00000	d6000000	1M		INTCTL
+ * fef00000	c0000000	1M		CAP
+ */
+
 /* 
  * Static I/O regions.
  *
-- 
cgit v1.2.3-70-g09d2


From baaf7ed1797de1defce3e3a03f3c9c49cdf3b534 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Sun, 26 Jun 2005 22:24:17 +0100
Subject: [PATCH] ARM: 2756/1: add ixp2000 msf mapping

Patch from Lennert Buytenhek

Add a mapping for the ixp2400 and ixp2800 msf unit.  The msf is the
ixp2000's 'media and switch fabric' unit, which handles the networking
part of the chip.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Deepak Saxena
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-ixp2000/core.c                | 5 +++++
 include/asm-arm/arch-ixp2000/ixp2000-regs.h | 4 ++++
 2 files changed, 9 insertions(+)

(limited to 'include')

diff --git a/arch/arm/mach-ixp2000/core.c b/arch/arm/mach-ixp2000/core.c
index efd77eee065..4b9d841e04c 100644
--- a/arch/arm/mach-ixp2000/core.c
+++ b/arch/arm/mach-ixp2000/core.c
@@ -102,6 +102,11 @@ static struct map_desc ixp2000_io_desc[] __initdata = {
 		.physical	= IXP2000_PCI_CSR_PHYS_BASE,
 		.length		= IXP2000_PCI_CSR_SIZE,
 		.type		= MT_DEVICE
+	}, {
+		.virtual	= IXP2000_MSF_VIRT_BASE,
+		.physical	= IXP2000_MSF_PHYS_BASE,
+		.length		= IXP2000_MSF_SIZE,
+		.type		= MT_DEVICE
 	}, {
 		.virtual	= IXP2000_PCI_IO_VIRT_BASE,
 		.physical	= IXP2000_PCI_IO_PHYS_BASE,
diff --git a/include/asm-arm/arch-ixp2000/ixp2000-regs.h b/include/asm-arm/arch-ixp2000/ixp2000-regs.h
index 213c0cfbb15..75623f81ef7 100644
--- a/include/asm-arm/arch-ixp2000/ixp2000-regs.h
+++ b/include/asm-arm/arch-ixp2000/ixp2000-regs.h
@@ -86,6 +86,10 @@
 #define	IXP2000_PCI_CSR_VIRT_BASE	0xfec00000
 #define	IXP2000_PCI_CSR_SIZE		0x00100000
 
+#define IXP2000_MSF_PHYS_BASE		0xc8000000
+#define IXP2000_MSF_VIRT_BASE		0xfeb00000
+#define IXP2000_MSF_SIZE		0x00100000
+
 #define IXP2000_PCI_IO_PHYS_BASE	0xd8000000
 #define	IXP2000_PCI_IO_VIRT_BASE	0xfd000000
 #define IXP2000_PCI_IO_SIZE     	0x01000000
-- 
cgit v1.2.3-70-g09d2


From 169a3e66637c667b43dab7c319ffd5c99804cad8 Mon Sep 17 00:00:00 2001
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Sun, 26 Jun 2005 17:54:11 -0400
Subject: bonding: xor/802.3ad improved slave hash

Add support for alternate slave selection algorithms to bonding
balance-xor and 802.3ad modes.  Default mode (what we have now: xor of
MAC addresses) is "layer2", new choice is "layer3+4", using IP and port
information for hashing to select peer.

Originally submitted by Jason Gabler for balance-xor mode;
modified by Jay Vosburgh to additionally support 802.3ad mode.  Jason's
original comment is as follows:

The attached patch to the Linux Etherchannel Bonding driver modifies the
driver's "balance-xor" mode as follows:

      - alternate hashing policy support for mode 2
        * Added kernel parameter "xmit_policy" to allow the specification
          of different hashing policies for mode 2.  The original mode 2
          policy is the default, now found in xmit_hash_policy_layer2().
        * Added xmit_hash_policy_layer34()

This patch was inspired by hashing policies implemented by Cisco,
Foundry and IBM, which are explained in
Foundry documentation found at:
http://www.foundrynet.com/services/documentation/sribcg/Trunking.html#112750

Signed-off-by: Jason Gabler <jygabler@lbl.gov>
Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
---
 drivers/net/bonding/bond_3ad.c  |   3 +-
 drivers/net/bonding/bond_main.c | 107 ++++++++++++++++++++++++++++++++++++----
 drivers/net/bonding/bonding.h   |  10 +++-
 include/linux/if_bonding.h      |   7 +++
 4 files changed, 114 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index 6233c4ffb80..a2e8dda5afa 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -2346,7 +2346,6 @@ int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev)
 {
 	struct slave *slave, *start_at;
 	struct bonding *bond = dev->priv;
-	struct ethhdr *data = (struct ethhdr *)skb->data;
 	int slave_agg_no;
 	int slaves_in_agg;
 	int agg_id;
@@ -2377,7 +2376,7 @@ int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev)
 		goto out;
 	}
 
-	slave_agg_no = (data->h_dest[5]^bond->dev->dev_addr[5]) % slaves_in_agg;
+	slave_agg_no = bond->xmit_hash_policy(skb, dev, slaves_in_agg);
 
 	bond_for_each_slave(bond, slave, i) {
 		struct aggregator *agg = SLAVE_AD_INFO(slave).port.aggregator;
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 545f6fe025a..2c930da90a8 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -479,6 +479,14 @@
  * 	- Support for generating gratuitous ARPs in active-backup mode.
  * 	  Includes support for VLAN tagging all bonding-generated ARPs
  * 	  as needed.  Set version to 2.6.2.
+ * 2005/06/08 - Jason Gabler <jygabler at lbl dot gov>
+ *	- alternate hashing policy support for mode 2
+ *	  * Added kernel parameter "xmit_hash_policy" to allow the selection
+ *	    of different hashing policies for mode 2.  The original mode 2
+ *	    policy is the default, now found in xmit_hash_policy_layer2().
+ *	  * Added xmit_hash_policy_layer34()
+ *	- Modified by Jay Vosburgh <fubar@us.ibm.com> to also support mode 4.
+ *	  Set version to 2.6.3.
  */
 
 //#define BONDING_DEBUG 1
@@ -493,7 +501,10 @@
 #include <linux/ptrace.h>
 #include <linux/ioport.h>
 #include <linux/in.h>
+#include <net/ip.h>
 #include <linux/ip.h>
+#include <linux/tcp.h>
+#include <linux/udp.h>
 #include <linux/slab.h>
 #include <linux/string.h>
 #include <linux/init.h>
@@ -541,6 +552,7 @@ static int use_carrier	= 1;
 static char *mode	= NULL;
 static char *primary	= NULL;
 static char *lacp_rate	= NULL;
+static char *xmit_hash_policy = NULL;
 static int arp_interval = BOND_LINK_ARP_INTERV;
 static char *arp_ip_target[BOND_MAX_ARP_TARGETS] = { NULL, };
 
@@ -560,6 +572,8 @@ module_param(primary, charp, 0);
 MODULE_PARM_DESC(primary, "Primary network device to use");
 module_param(lacp_rate, charp, 0);
 MODULE_PARM_DESC(lacp_rate, "LACPDU tx rate to request from 802.3ad partner (slow/fast)");
+module_param(xmit_hash_policy, charp, 0);
+MODULE_PARM_DESC(xmit_hash_policy, "XOR hashing method : 0 for layer 2 (default), 1 for layer 3+4");
 module_param(arp_interval, int, 0);
 MODULE_PARM_DESC(arp_interval, "arp interval in milliseconds");
 module_param_array(arp_ip_target, charp, NULL, 0);
@@ -579,6 +593,7 @@ static struct proc_dir_entry *bond_proc_dir = NULL;
 static u32 arp_target[BOND_MAX_ARP_TARGETS] = { 0, } ;
 static int arp_ip_count	= 0;
 static int bond_mode	= BOND_MODE_ROUNDROBIN;
+static int xmit_hashtype= BOND_XMIT_POLICY_LAYER2;
 static int lacp_fast	= 0;
 static int app_abi_ver	= 0;
 static int orig_app_abi_ver = -1; /* This is used to save the first ABI version
@@ -588,7 +603,6 @@ static int orig_app_abi_ver = -1; /* This is used to save the first ABI version
 				   * command comes from an application using
 				   * another ABI version.
 				   */
-
 struct bond_parm_tbl {
 	char *modename;
 	int mode;
@@ -611,9 +625,15 @@ static struct bond_parm_tbl bond_mode_tbl[] = {
 {	NULL,			-1},
 };
 
+static struct bond_parm_tbl xmit_hashtype_tbl[] = {
+{	"layer2",		BOND_XMIT_POLICY_LAYER2},
+{	"layer3+4",		BOND_XMIT_POLICY_LAYER34},
+{	NULL,			-1},
+};
+
 /*-------------------------- Forward declarations ---------------------------*/
 
-static inline void bond_set_mode_ops(struct net_device *bond_dev, int mode);
+static inline void bond_set_mode_ops(struct bonding *bond, int mode);
 static void bond_send_gratuitous_arp(struct bonding *bond);
 
 /*---------------------------- General routines -----------------------------*/
@@ -3724,6 +3744,46 @@ static void bond_unregister_lacpdu(struct bonding *bond)
 	dev_remove_pack(&(BOND_AD_INFO(bond).ad_pkt_type));
 }
 
+/*---------------------------- Hashing Policies -----------------------------*/
+
+/*
+ * Hash for the the output device based upon layer 3 and layer 4 data. If
+ * the packet is a frag or not TCP or UDP, just use layer 3 data.  If it is
+ * altogether not IP, mimic bond_xmit_hash_policy_l2()
+ */
+static int bond_xmit_hash_policy_l34(struct sk_buff *skb,
+				    struct net_device *bond_dev, int count)
+{
+	struct ethhdr *data = (struct ethhdr *)skb->data;
+	struct iphdr *iph = skb->nh.iph;
+	u16 *layer4hdr = (u16 *)((u32 *)iph + iph->ihl);
+	int layer4_xor = 0;
+
+	if (skb->protocol == __constant_htons(ETH_P_IP)) {
+		if (!(iph->frag_off & __constant_htons(IP_MF|IP_OFFSET)) &&
+		    (iph->protocol == IPPROTO_TCP ||
+		     iph->protocol == IPPROTO_UDP)) {
+			layer4_xor = htons((*layer4hdr ^ *(layer4hdr + 1)));
+		}
+		return (layer4_xor ^
+			((ntohl(iph->saddr ^ iph->daddr)) & 0xffff)) % count;
+
+	}
+
+	return (data->h_dest[5] ^ bond_dev->dev_addr[5]) % count;
+}
+
+/*
+ * Hash for the output device based upon layer 2 data
+ */
+static int bond_xmit_hash_policy_l2(struct sk_buff *skb,
+				   struct net_device *bond_dev, int count)
+{
+	struct ethhdr *data = (struct ethhdr *)skb->data;
+
+	return (data->h_dest[5] ^ bond_dev->dev_addr[5]) % count;
+}
+
 /*-------------------------- Device entry points ----------------------------*/
 
 static int bond_open(struct net_device *bond_dev)
@@ -4310,14 +4370,13 @@ out:
 }
 
 /*
- * in XOR mode, we determine the output device by performing xor on
- * the source and destination hw adresses.  If this device is not
- * enabled, find the next slave following this xor slave.
+ * In bond_xmit_xor() , we determine the output device by using a pre-
+ * determined xmit_hash_policy(), If the selected device is not enabled,
+ * find the next active slave.
  */
 static int bond_xmit_xor(struct sk_buff *skb, struct net_device *bond_dev)
 {
 	struct bonding *bond = bond_dev->priv;
-	struct ethhdr *data = (struct ethhdr *)skb->data;
 	struct slave *slave, *start_at;
 	int slave_no;
 	int i;
@@ -4329,7 +4388,7 @@ static int bond_xmit_xor(struct sk_buff *skb, struct net_device *bond_dev)
 		goto out;
 	}
 
-	slave_no = (data->h_dest[5]^bond_dev->dev_addr[5]) % bond->slave_cnt;
+	slave_no = bond->xmit_hash_policy(skb, bond_dev, bond->slave_cnt);
 
 	bond_for_each_slave(bond, slave, i) {
 		slave_no--;
@@ -4425,8 +4484,10 @@ out:
 /*
  * set bond mode specific net device operations
  */
-static inline void bond_set_mode_ops(struct net_device *bond_dev, int mode)
+static inline void bond_set_mode_ops(struct bonding *bond, int mode)
 {
+	struct net_device *bond_dev = bond->dev;
+
 	switch (mode) {
 	case BOND_MODE_ROUNDROBIN:
 		bond_dev->hard_start_xmit = bond_xmit_roundrobin;
@@ -4436,12 +4497,20 @@ static inline void bond_set_mode_ops(struct net_device *bond_dev, int mode)
 		break;
 	case BOND_MODE_XOR:
 		bond_dev->hard_start_xmit = bond_xmit_xor;
+		if (bond->params.xmit_policy == BOND_XMIT_POLICY_LAYER34)
+			bond->xmit_hash_policy = bond_xmit_hash_policy_l34;
+		else
+			bond->xmit_hash_policy = bond_xmit_hash_policy_l2;
 		break;
 	case BOND_MODE_BROADCAST:
 		bond_dev->hard_start_xmit = bond_xmit_broadcast;
 		break;
 	case BOND_MODE_8023AD:
 		bond_dev->hard_start_xmit = bond_3ad_xmit_xor;
+		if (bond->params.xmit_policy == BOND_XMIT_POLICY_LAYER34)
+			bond->xmit_hash_policy = bond_xmit_hash_policy_l34;
+		else
+			bond->xmit_hash_policy = bond_xmit_hash_policy_l2;
 		break;
 	case BOND_MODE_TLB:
 	case BOND_MODE_ALB:
@@ -4490,7 +4559,7 @@ static int __init bond_init(struct net_device *bond_dev, struct bond_params *par
 	bond_dev->change_mtu = bond_change_mtu;
 	bond_dev->set_mac_address = bond_set_mac_address;
 
-	bond_set_mode_ops(bond_dev, bond->params.mode);
+	bond_set_mode_ops(bond, bond->params.mode);
 
 	bond_dev->destructor = free_netdev;
 
@@ -4601,6 +4670,25 @@ static int bond_check_params(struct bond_params *params)
 		}
 	}
 
+	if (xmit_hash_policy) {
+		if ((bond_mode != BOND_MODE_XOR) &&
+		    (bond_mode != BOND_MODE_8023AD)) {
+			printk(KERN_INFO DRV_NAME
+			       ": xor_mode param is irrelevant in mode %s\n",
+			       bond_mode_name(bond_mode));
+		} else {
+			xmit_hashtype = bond_parse_parm(xmit_hash_policy,
+							xmit_hashtype_tbl);
+			if (xmit_hashtype == -1) {
+				printk(KERN_ERR DRV_NAME
+			       	": Error: Invalid xmit_hash_policy \"%s\"\n",
+			       	xmit_hash_policy == NULL ? "NULL" :
+				       xmit_hash_policy);
+				return -EINVAL;
+			}
+		}
+	}
+
 	if (lacp_rate) {
 		if (bond_mode != BOND_MODE_8023AD) {
 			printk(KERN_INFO DRV_NAME
@@ -4812,6 +4900,7 @@ static int bond_check_params(struct bond_params *params)
 
 	/* fill params struct with the proper values */
 	params->mode = bond_mode;
+	params->xmit_policy = xmit_hashtype;
 	params->miimon = miimon;
 	params->arp_interval = arp_interval;
 	params->updelay = updelay;
diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index 6558af22eda..d27f377b3ee 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -25,6 +25,10 @@
  *
  * 2003/12/01 - Shmulik Hen <shmulik.hen at intel dot com>
  *	- Code cleanup and style changes
+ *
+ * 2005/05/05 - Jason Gabler <jygabler at lbl dot gov>
+ *      - added "xmit_policy" kernel parameter for alternate hashing policy
+ *	  support for mode 2
  */
 
 #ifndef _LINUX_BONDING_H
@@ -36,8 +40,8 @@
 #include "bond_3ad.h"
 #include "bond_alb.h"
 
-#define DRV_VERSION	"2.6.2"
-#define DRV_RELDATE	"June 5, 2005"
+#define DRV_VERSION	"2.6.3"
+#define DRV_RELDATE	"June 8, 2005"
 #define DRV_NAME	"bonding"
 #define DRV_DESCRIPTION	"Ethernet Channel Bonding Driver"
 
@@ -137,6 +141,7 @@
 
 struct bond_params {
 	int mode;
+	int xmit_policy;
 	int miimon;
 	int arp_interval;
 	int use_carrier;
@@ -198,6 +203,7 @@ struct bonding {
 #endif /* CONFIG_PROC_FS */
 	struct   list_head bond_list;
 	struct   dev_mc_list *mc_list;
+	int      (*xmit_hash_policy)(struct sk_buff *, struct net_device *, int);
 	u32      master_ip;
 	u16      flags;
 	struct   ad_bond_info ad_info;
diff --git a/include/linux/if_bonding.h b/include/linux/if_bonding.h
index 57024ce2c74..84598fa2e9d 100644
--- a/include/linux/if_bonding.h
+++ b/include/linux/if_bonding.h
@@ -35,6 +35,9 @@
  *
  * 2003/12/01 - Shmulik Hen <shmulik.hen at intel dot com>
  *	- Code cleanup and style changes
+ *
+ * 2005/05/05 - Jason Gabler <jygabler at lbl dot gov>
+ *      - added definitions for various XOR hashing policies
  */
 
 #ifndef _LINUX_IF_BONDING_H
@@ -80,6 +83,10 @@
 
 #define BOND_DEFAULT_MAX_BONDS  1   /* Default maximum number of devices to support */
 
+/* hashing types */
+#define BOND_XMIT_POLICY_LAYER2		0 /* layer 2 (MAC only), default */
+#define BOND_XMIT_POLICY_LAYER34	1 /* layer 3+4 (IP ^ MAC) */
+
 typedef struct ifbond {
 	__s32 bond_mode;
 	__s32 num_slaves;
-- 
cgit v1.2.3-70-g09d2


From 32e9e25ef20789c24ffa1f41489a13932cf82c77 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Sun, 26 Jun 2005 15:28:10 -0700
Subject: [ATALK]: Include asm/byteorder.h in linux/atalk.h

We're using __be16 in userland visible types, so we
have to include asm/byteorder.h so that works.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/atalk.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/linux/atalk.h b/include/linux/atalk.h
index 09a1451c115..911c09cb9bf 100644
--- a/include/linux/atalk.h
+++ b/include/linux/atalk.h
@@ -1,6 +1,8 @@
 #ifndef __LINUX_ATALK_H__
 #define __LINUX_ATALK_H__
 
+#include <asm/byteorder.h>
+
 /*
  * AppleTalk networking structures
  *
-- 
cgit v1.2.3-70-g09d2


From f49d16ef2d6f008119d4ee2c895781fb229bad68 Mon Sep 17 00:00:00 2001
From: Manfred Spraul <manfred@colorfullife.com>
Date: Sun, 26 Jun 2005 11:36:52 +0200
Subject: [PATCH] forcedeth: Add support for new device id

This is a multi-part message in MIME format.
---
 drivers/net/forcedeth.c | 17 ++++++++++++++++-
 include/linux/pci_ids.h |  2 ++
 2 files changed, 18 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/net/forcedeth.c b/drivers/net/forcedeth.c
index b471d1a8ffd..64f0f697c95 100644
--- a/drivers/net/forcedeth.c
+++ b/drivers/net/forcedeth.c
@@ -84,6 +84,7 @@
  *	0.32: 16 Apr 2005: RX_ERROR4 handling added.
  *	0.33: 16 May 2005: Support for MCP51 added.
  *	0.34: 18 Jun 2005: Add DEV_NEED_LINKTIMER to all nForce nics.
+ *	0.35: 26 Jun 2005: Support for MCP55 added.
  *
  * Known bugs:
  * We suspect that on some hardware no TX done interrupts are generated.
@@ -95,7 +96,7 @@
  * DEV_NEED_TIMERIRQ will not harm you on sane hardware, only generating a few
  * superfluous timer interrupts from the nic.
  */
-#define FORCEDETH_VERSION		"0.34"
+#define FORCEDETH_VERSION		"0.35"
 #define DRV_NAME			"forcedeth"
 
 #include <linux/module.h>
@@ -2284,6 +2285,20 @@ static struct pci_device_id pci_tbl[] = {
 		.subdevice = PCI_ANY_ID,
 		.driver_data = DEV_NEED_LASTPACKET1|DEV_IRQMASK_2|DEV_NEED_TIMERIRQ|DEV_NEED_LINKTIMER,
 	},
+	{	/* MCP55 Ethernet Controller */
+		.vendor = PCI_VENDOR_ID_NVIDIA,
+		.device = PCI_DEVICE_ID_NVIDIA_NVENET_14,
+		.subvendor = PCI_ANY_ID,
+		.subdevice = PCI_ANY_ID,
+		.driver_data = DEV_NEED_LASTPACKET1|DEV_IRQMASK_2|DEV_NEED_TIMERIRQ|DEV_NEED_LINKTIMER,
+	},
+	{	/* MCP55 Ethernet Controller */
+		.vendor = PCI_VENDOR_ID_NVIDIA,
+		.device = PCI_DEVICE_ID_NVIDIA_NVENET_15,
+		.subvendor = PCI_ANY_ID,
+		.subdevice = PCI_ANY_ID,
+		.driver_data = DEV_NEED_LASTPACKET1|DEV_IRQMASK_2|DEV_NEED_TIMERIRQ|DEV_NEED_LINKTIMER,
+	},
 	{0,},
 };
 
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index bf608808a60..3af7450278b 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1284,6 +1284,8 @@
 #define PCI_DEVICE_ID_NVIDIA_GEFORCE_FX_GO5700_2    0x0348
 #define PCI_DEVICE_ID_NVIDIA_QUADRO_FX_GO1000       0x034C
 #define PCI_DEVICE_ID_NVIDIA_QUADRO_FX_1100         0x034E
+#define PCI_DEVICE_ID_NVIDIA_NVENET_14              0x0372
+#define PCI_DEVICE_ID_NVIDIA_NVENET_15              0x0373
 
 #define PCI_VENDOR_ID_IMS		0x10e0
 #define PCI_DEVICE_ID_IMS_8849		0x8849
-- 
cgit v1.2.3-70-g09d2


From 279385949ebb41ad166fd37505fe552cdb74ed59 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Sun, 19 Jun 2005 01:27:20 +0200
Subject: [PATCH] bring over ieee80211.h from mainline

the prototypes and inlines aren't actually needed, but let's not diverge
from -mm too far.
---
 include/net/ieee80211.h | 882 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 882 insertions(+)
 create mode 100644 include/net/ieee80211.h

(limited to 'include')

diff --git a/include/net/ieee80211.h b/include/net/ieee80211.h
new file mode 100644
index 00000000000..7fe57f957a5
--- /dev/null
+++ b/include/net/ieee80211.h
@@ -0,0 +1,882 @@
+/*
+ * Merged with mainline ieee80211.h in Aug 2004.  Original ieee802_11
+ * remains copyright by the original authors
+ *
+ * Portions of the merged code are based on Host AP (software wireless
+ * LAN access point) driver for Intersil Prism2/2.5/3.
+ *
+ * Copyright (c) 2001-2002, SSH Communications Security Corp and Jouni Malinen
+ * <jkmaline@cc.hut.fi>
+ * Copyright (c) 2002-2003, Jouni Malinen <jkmaline@cc.hut.fi>
+ *
+ * Adaption to a generic IEEE 802.11 stack by James Ketrenos
+ * <jketreno@linux.intel.com>
+ * Copyright (c) 2004, Intel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation. See README and COPYING for
+ * more details.
+ */
+#ifndef IEEE80211_H
+#define IEEE80211_H
+
+#include <linux/if_ether.h> /* ETH_ALEN */
+#include <linux/kernel.h>   /* ARRAY_SIZE */
+
+#if WIRELESS_EXT < 17
+#define IW_QUAL_QUAL_INVALID   0x10
+#define IW_QUAL_LEVEL_INVALID  0x20
+#define IW_QUAL_NOISE_INVALID  0x40
+#define IW_QUAL_QUAL_UPDATED   0x1
+#define IW_QUAL_LEVEL_UPDATED  0x2
+#define IW_QUAL_NOISE_UPDATED  0x4
+#endif
+
+#define IEEE80211_DATA_LEN		2304
+/* Maximum size for the MA-UNITDATA primitive, 802.11 standard section
+   6.2.1.1.2.
+
+   The figure in section 7.1.2 suggests a body size of up to 2312
+   bytes is allowed, which is a bit confusing, I suspect this
+   represents the 2304 bytes of real data, plus a possible 8 bytes of
+   WEP IV and ICV. (this interpretation suggested by Ramiro Barreiro) */
+
+
+#define IEEE80211_HLEN			30
+#define IEEE80211_FRAME_LEN		(IEEE80211_DATA_LEN + IEEE80211_HLEN)
+
+struct ieee80211_hdr {
+	u16 frame_ctl;
+	u16 duration_id;
+	u8 addr1[ETH_ALEN];
+	u8 addr2[ETH_ALEN];
+	u8 addr3[ETH_ALEN];
+	u16 seq_ctl;
+	u8 addr4[ETH_ALEN];
+} __attribute__ ((packed));
+
+struct ieee80211_hdr_3addr {
+	u16 frame_ctl;
+	u16 duration_id;
+	u8 addr1[ETH_ALEN];
+	u8 addr2[ETH_ALEN];
+	u8 addr3[ETH_ALEN];
+	u16 seq_ctl;
+} __attribute__ ((packed));
+
+enum eap_type {
+	EAP_PACKET = 0,
+	EAPOL_START,
+	EAPOL_LOGOFF,
+	EAPOL_KEY,
+	EAPOL_ENCAP_ASF_ALERT
+};
+
+static const char *eap_types[] = {
+	[EAP_PACKET]		= "EAP-Packet",
+	[EAPOL_START]		= "EAPOL-Start",
+	[EAPOL_LOGOFF]		= "EAPOL-Logoff",
+	[EAPOL_KEY]		= "EAPOL-Key",
+	[EAPOL_ENCAP_ASF_ALERT]	= "EAPOL-Encap-ASF-Alert"
+};
+
+static inline const char *eap_get_type(int type)
+{
+	return (type >= ARRAY_SIZE(eap_types)) ? "Unknown" : eap_types[type];
+}
+
+struct eapol {
+	u8 snap[6];
+	u16 ethertype;
+	u8 version;
+	u8 type;
+	u16 length;
+} __attribute__ ((packed));
+
+#define IEEE80211_3ADDR_LEN 24
+#define IEEE80211_4ADDR_LEN 30
+#define IEEE80211_FCS_LEN    4
+
+#define MIN_FRAG_THRESHOLD     256U
+#define	MAX_FRAG_THRESHOLD     2346U
+
+/* Frame control field constants */
+#define IEEE80211_FCTL_VERS		0x0002
+#define IEEE80211_FCTL_FTYPE		0x000c
+#define IEEE80211_FCTL_STYPE		0x00f0
+#define IEEE80211_FCTL_TODS		0x0100
+#define IEEE80211_FCTL_FROMDS		0x0200
+#define IEEE80211_FCTL_MOREFRAGS	0x0400
+#define IEEE80211_FCTL_RETRY		0x0800
+#define IEEE80211_FCTL_PM		0x1000
+#define IEEE80211_FCTL_MOREDATA	0x2000
+#define IEEE80211_FCTL_WEP		0x4000
+#define IEEE80211_FCTL_ORDER		0x8000
+
+#define IEEE80211_FTYPE_MGMT		0x0000
+#define IEEE80211_FTYPE_CTL		0x0004
+#define IEEE80211_FTYPE_DATA		0x0008
+
+/* management */
+#define IEEE80211_STYPE_ASSOC_REQ	0x0000
+#define IEEE80211_STYPE_ASSOC_RESP 	0x0010
+#define IEEE80211_STYPE_REASSOC_REQ	0x0020
+#define IEEE80211_STYPE_REASSOC_RESP	0x0030
+#define IEEE80211_STYPE_PROBE_REQ	0x0040
+#define IEEE80211_STYPE_PROBE_RESP	0x0050
+#define IEEE80211_STYPE_BEACON		0x0080
+#define IEEE80211_STYPE_ATIM		0x0090
+#define IEEE80211_STYPE_DISASSOC	0x00A0
+#define IEEE80211_STYPE_AUTH		0x00B0
+#define IEEE80211_STYPE_DEAUTH		0x00C0
+
+/* control */
+#define IEEE80211_STYPE_PSPOLL		0x00A0
+#define IEEE80211_STYPE_RTS		0x00B0
+#define IEEE80211_STYPE_CTS		0x00C0
+#define IEEE80211_STYPE_ACK		0x00D0
+#define IEEE80211_STYPE_CFEND		0x00E0
+#define IEEE80211_STYPE_CFENDACK	0x00F0
+
+/* data */
+#define IEEE80211_STYPE_DATA		0x0000
+#define IEEE80211_STYPE_DATA_CFACK	0x0010
+#define IEEE80211_STYPE_DATA_CFPOLL	0x0020
+#define IEEE80211_STYPE_DATA_CFACKPOLL	0x0030
+#define IEEE80211_STYPE_NULLFUNC	0x0040
+#define IEEE80211_STYPE_CFACK		0x0050
+#define IEEE80211_STYPE_CFPOLL		0x0060
+#define IEEE80211_STYPE_CFACKPOLL	0x0070
+
+#define IEEE80211_SCTL_FRAG		0x000F
+#define IEEE80211_SCTL_SEQ		0xFFF0
+
+
+/* debug macros */
+
+#ifdef CONFIG_IEEE80211_DEBUG
+extern u32 ieee80211_debug_level;
+#define IEEE80211_DEBUG(level, fmt, args...) \
+do { if (ieee80211_debug_level & (level)) \
+  printk(KERN_DEBUG "ieee80211: %c %s " fmt, \
+         in_interrupt() ? 'I' : 'U', __FUNCTION__ , ## args); } while (0)
+#else
+#define IEEE80211_DEBUG(level, fmt, args...) do {} while (0)
+#endif	/* CONFIG_IEEE80211_DEBUG */
+
+/*
+ * To use the debug system;
+ *
+ * If you are defining a new debug classification, simply add it to the #define
+ * list here in the form of:
+ *
+ * #define IEEE80211_DL_xxxx VALUE
+ *
+ * shifting value to the left one bit from the previous entry.  xxxx should be
+ * the name of the classification (for example, WEP)
+ *
+ * You then need to either add a IEEE80211_xxxx_DEBUG() macro definition for your
+ * classification, or use IEEE80211_DEBUG(IEEE80211_DL_xxxx, ...) whenever you want
+ * to send output to that classification.
+ *
+ * To add your debug level to the list of levels seen when you perform
+ *
+ * % cat /proc/net/ipw/debug_level
+ *
+ * you simply need to add your entry to the ipw_debug_levels array.
+ *
+ * If you do not see debug_level in /proc/net/ipw then you do not have
+ * CONFIG_IEEE80211_DEBUG defined in your kernel configuration
+ *
+ */
+
+#define IEEE80211_DL_INFO          (1<<0)
+#define IEEE80211_DL_WX            (1<<1)
+#define IEEE80211_DL_SCAN          (1<<2)
+#define IEEE80211_DL_STATE         (1<<3)
+#define IEEE80211_DL_MGMT          (1<<4)
+#define IEEE80211_DL_FRAG          (1<<5)
+#define IEEE80211_DL_EAP           (1<<6)
+#define IEEE80211_DL_DROP          (1<<7)
+
+#define IEEE80211_DL_TX            (1<<8)
+#define IEEE80211_DL_RX            (1<<9)
+
+#define IEEE80211_ERROR(f, a...) printk(KERN_ERR "ieee80211: " f, ## a)
+#define IEEE80211_WARNING(f, a...) printk(KERN_WARNING "ieee80211: " f, ## a)
+#define IEEE80211_DEBUG_INFO(f, a...)   IEEE80211_DEBUG(IEEE80211_DL_INFO, f, ## a)
+
+#define IEEE80211_DEBUG_WX(f, a...)     IEEE80211_DEBUG(IEEE80211_DL_WX, f, ## a)
+#define IEEE80211_DEBUG_SCAN(f, a...)   IEEE80211_DEBUG(IEEE80211_DL_SCAN, f, ## a)
+#define IEEE80211_DEBUG_STATE(f, a...)  IEEE80211_DEBUG(IEEE80211_DL_STATE, f, ## a)
+#define IEEE80211_DEBUG_MGMT(f, a...)  IEEE80211_DEBUG(IEEE80211_DL_MGMT, f, ## a)
+#define IEEE80211_DEBUG_FRAG(f, a...)  IEEE80211_DEBUG(IEEE80211_DL_FRAG, f, ## a)
+#define IEEE80211_DEBUG_EAP(f, a...)  IEEE80211_DEBUG(IEEE80211_DL_EAP, f, ## a)
+#define IEEE80211_DEBUG_DROP(f, a...)  IEEE80211_DEBUG(IEEE80211_DL_DROP, f, ## a)
+#define IEEE80211_DEBUG_TX(f, a...)  IEEE80211_DEBUG(IEEE80211_DL_TX, f, ## a)
+#define IEEE80211_DEBUG_RX(f, a...)  IEEE80211_DEBUG(IEEE80211_DL_RX, f, ## a)
+#include <linux/netdevice.h>
+#include <linux/wireless.h>
+#include <linux/if_arp.h> /* ARPHRD_ETHER */
+
+#ifndef WIRELESS_SPY
+#define WIRELESS_SPY		// enable iwspy support
+#endif
+#include <net/iw_handler.h>	// new driver API
+
+#ifndef ETH_P_PAE
+#define ETH_P_PAE 0x888E /* Port Access Entity (IEEE 802.1X) */
+#endif /* ETH_P_PAE */
+
+#define ETH_P_PREAUTH 0x88C7 /* IEEE 802.11i pre-authentication */
+
+#ifndef ETH_P_80211_RAW
+#define ETH_P_80211_RAW (ETH_P_ECONET + 1)
+#endif
+
+/* IEEE 802.11 defines */
+
+#define P80211_OUI_LEN 3
+
+struct ieee80211_snap_hdr {
+
+        u8    dsap;   /* always 0xAA */
+        u8    ssap;   /* always 0xAA */
+        u8    ctrl;   /* always 0x03 */
+        u8    oui[P80211_OUI_LEN];    /* organizational universal id */
+
+} __attribute__ ((packed));
+
+#define SNAP_SIZE sizeof(struct ieee80211_snap_hdr)
+
+#define WLAN_FC_GET_TYPE(fc) ((fc) & IEEE80211_FCTL_FTYPE)
+#define WLAN_FC_GET_STYPE(fc) ((fc) & IEEE80211_FCTL_STYPE)
+
+#define WLAN_GET_SEQ_FRAG(seq) ((seq) & IEEE80211_SCTL_FRAG)
+#define WLAN_GET_SEQ_SEQ(seq)  ((seq) & IEEE80211_SCTL_SEQ)
+
+/* Authentication algorithms */
+#define WLAN_AUTH_OPEN 0
+#define WLAN_AUTH_SHARED_KEY 1
+
+#define WLAN_AUTH_CHALLENGE_LEN 128
+
+#define WLAN_CAPABILITY_BSS (1<<0)
+#define WLAN_CAPABILITY_IBSS (1<<1)
+#define WLAN_CAPABILITY_CF_POLLABLE (1<<2)
+#define WLAN_CAPABILITY_CF_POLL_REQUEST (1<<3)
+#define WLAN_CAPABILITY_PRIVACY (1<<4)
+#define WLAN_CAPABILITY_SHORT_PREAMBLE (1<<5)
+#define WLAN_CAPABILITY_PBCC (1<<6)
+#define WLAN_CAPABILITY_CHANNEL_AGILITY (1<<7)
+
+/* Status codes */
+#define WLAN_STATUS_SUCCESS 0
+#define WLAN_STATUS_UNSPECIFIED_FAILURE 1
+#define WLAN_STATUS_CAPS_UNSUPPORTED 10
+#define WLAN_STATUS_REASSOC_NO_ASSOC 11
+#define WLAN_STATUS_ASSOC_DENIED_UNSPEC 12
+#define WLAN_STATUS_NOT_SUPPORTED_AUTH_ALG 13
+#define WLAN_STATUS_UNKNOWN_AUTH_TRANSACTION 14
+#define WLAN_STATUS_CHALLENGE_FAIL 15
+#define WLAN_STATUS_AUTH_TIMEOUT 16
+#define WLAN_STATUS_AP_UNABLE_TO_HANDLE_NEW_STA 17
+#define WLAN_STATUS_ASSOC_DENIED_RATES 18
+/* 802.11b */
+#define WLAN_STATUS_ASSOC_DENIED_NOSHORT 19
+#define WLAN_STATUS_ASSOC_DENIED_NOPBCC 20
+#define WLAN_STATUS_ASSOC_DENIED_NOAGILITY 21
+
+/* Reason codes */
+#define WLAN_REASON_UNSPECIFIED 1
+#define WLAN_REASON_PREV_AUTH_NOT_VALID 2
+#define WLAN_REASON_DEAUTH_LEAVING 3
+#define WLAN_REASON_DISASSOC_DUE_TO_INACTIVITY 4
+#define WLAN_REASON_DISASSOC_AP_BUSY 5
+#define WLAN_REASON_CLASS2_FRAME_FROM_NONAUTH_STA 6
+#define WLAN_REASON_CLASS3_FRAME_FROM_NONASSOC_STA 7
+#define WLAN_REASON_DISASSOC_STA_HAS_LEFT 8
+#define WLAN_REASON_STA_REQ_ASSOC_WITHOUT_AUTH 9
+
+
+/* Information Element IDs */
+#define WLAN_EID_SSID 0
+#define WLAN_EID_SUPP_RATES 1
+#define WLAN_EID_FH_PARAMS 2
+#define WLAN_EID_DS_PARAMS 3
+#define WLAN_EID_CF_PARAMS 4
+#define WLAN_EID_TIM 5
+#define WLAN_EID_IBSS_PARAMS 6
+#define WLAN_EID_CHALLENGE 16
+#define WLAN_EID_RSN 48
+#define WLAN_EID_GENERIC 221
+
+#define IEEE80211_MGMT_HDR_LEN 24
+#define IEEE80211_DATA_HDR3_LEN 24
+#define IEEE80211_DATA_HDR4_LEN 30
+
+
+#define IEEE80211_STATMASK_SIGNAL (1<<0)
+#define IEEE80211_STATMASK_RSSI (1<<1)
+#define IEEE80211_STATMASK_NOISE (1<<2)
+#define IEEE80211_STATMASK_RATE (1<<3)
+#define IEEE80211_STATMASK_WEMASK 0x7
+
+
+#define IEEE80211_CCK_MODULATION    (1<<0)
+#define IEEE80211_OFDM_MODULATION   (1<<1)
+
+#define IEEE80211_24GHZ_BAND     (1<<0)
+#define IEEE80211_52GHZ_BAND     (1<<1)
+
+#define IEEE80211_CCK_RATE_1MB		        0x02
+#define IEEE80211_CCK_RATE_2MB		        0x04
+#define IEEE80211_CCK_RATE_5MB		        0x0B
+#define IEEE80211_CCK_RATE_11MB		        0x16
+#define IEEE80211_OFDM_RATE_6MB		        0x0C
+#define IEEE80211_OFDM_RATE_9MB		        0x12
+#define IEEE80211_OFDM_RATE_12MB		0x18
+#define IEEE80211_OFDM_RATE_18MB		0x24
+#define IEEE80211_OFDM_RATE_24MB		0x30
+#define IEEE80211_OFDM_RATE_36MB		0x48
+#define IEEE80211_OFDM_RATE_48MB		0x60
+#define IEEE80211_OFDM_RATE_54MB		0x6C
+#define IEEE80211_BASIC_RATE_MASK		0x80
+
+#define IEEE80211_CCK_RATE_1MB_MASK		(1<<0)
+#define IEEE80211_CCK_RATE_2MB_MASK		(1<<1)
+#define IEEE80211_CCK_RATE_5MB_MASK		(1<<2)
+#define IEEE80211_CCK_RATE_11MB_MASK		(1<<3)
+#define IEEE80211_OFDM_RATE_6MB_MASK		(1<<4)
+#define IEEE80211_OFDM_RATE_9MB_MASK		(1<<5)
+#define IEEE80211_OFDM_RATE_12MB_MASK		(1<<6)
+#define IEEE80211_OFDM_RATE_18MB_MASK		(1<<7)
+#define IEEE80211_OFDM_RATE_24MB_MASK		(1<<8)
+#define IEEE80211_OFDM_RATE_36MB_MASK		(1<<9)
+#define IEEE80211_OFDM_RATE_48MB_MASK		(1<<10)
+#define IEEE80211_OFDM_RATE_54MB_MASK		(1<<11)
+
+#define IEEE80211_CCK_RATES_MASK	        0x0000000F
+#define IEEE80211_CCK_BASIC_RATES_MASK	(IEEE80211_CCK_RATE_1MB_MASK | \
+	IEEE80211_CCK_RATE_2MB_MASK)
+#define IEEE80211_CCK_DEFAULT_RATES_MASK	(IEEE80211_CCK_BASIC_RATES_MASK | \
+        IEEE80211_CCK_RATE_5MB_MASK | \
+        IEEE80211_CCK_RATE_11MB_MASK)
+
+#define IEEE80211_OFDM_RATES_MASK		0x00000FF0
+#define IEEE80211_OFDM_BASIC_RATES_MASK	(IEEE80211_OFDM_RATE_6MB_MASK | \
+	IEEE80211_OFDM_RATE_12MB_MASK | \
+	IEEE80211_OFDM_RATE_24MB_MASK)
+#define IEEE80211_OFDM_DEFAULT_RATES_MASK	(IEEE80211_OFDM_BASIC_RATES_MASK | \
+	IEEE80211_OFDM_RATE_9MB_MASK  | \
+	IEEE80211_OFDM_RATE_18MB_MASK | \
+	IEEE80211_OFDM_RATE_36MB_MASK | \
+	IEEE80211_OFDM_RATE_48MB_MASK | \
+	IEEE80211_OFDM_RATE_54MB_MASK)
+#define IEEE80211_DEFAULT_RATES_MASK (IEEE80211_OFDM_DEFAULT_RATES_MASK | \
+                                IEEE80211_CCK_DEFAULT_RATES_MASK)
+
+#define IEEE80211_NUM_OFDM_RATES	    8
+#define IEEE80211_NUM_CCK_RATES	            4
+#define IEEE80211_OFDM_SHIFT_MASK_A         4
+
+
+
+
+/* NOTE: This data is for statistical purposes; not all hardware provides this
+ *       information for frames received.  Not setting these will not cause
+ *       any adverse affects. */
+struct ieee80211_rx_stats {
+	u32 mac_time;
+	s8 rssi;
+	u8 signal;
+	u8 noise;
+	u16 rate; /* in 100 kbps */
+	u8 received_channel;
+	u8 control;
+	u8 mask;
+	u8 freq;
+	u16 len;
+};
+
+/* IEEE 802.11 requires that STA supports concurrent reception of at least
+ * three fragmented frames. This define can be increased to support more
+ * concurrent frames, but it should be noted that each entry can consume about
+ * 2 kB of RAM and increasing cache size will slow down frame reassembly. */
+#define IEEE80211_FRAG_CACHE_LEN 4
+
+struct ieee80211_frag_entry {
+	unsigned long first_frag_time;
+	unsigned int seq;
+	unsigned int last_frag;
+	struct sk_buff *skb;
+	u8 src_addr[ETH_ALEN];
+	u8 dst_addr[ETH_ALEN];
+};
+
+struct ieee80211_stats {
+	unsigned int tx_unicast_frames;
+	unsigned int tx_multicast_frames;
+	unsigned int tx_fragments;
+	unsigned int tx_unicast_octets;
+	unsigned int tx_multicast_octets;
+	unsigned int tx_deferred_transmissions;
+	unsigned int tx_single_retry_frames;
+	unsigned int tx_multiple_retry_frames;
+	unsigned int tx_retry_limit_exceeded;
+	unsigned int tx_discards;
+	unsigned int rx_unicast_frames;
+	unsigned int rx_multicast_frames;
+	unsigned int rx_fragments;
+	unsigned int rx_unicast_octets;
+	unsigned int rx_multicast_octets;
+	unsigned int rx_fcs_errors;
+	unsigned int rx_discards_no_buffer;
+	unsigned int tx_discards_wrong_sa;
+	unsigned int rx_discards_undecryptable;
+	unsigned int rx_message_in_msg_fragments;
+	unsigned int rx_message_in_bad_msg_fragments;
+};
+
+struct ieee80211_device;
+
+#define SEC_KEY_1         (1<<0)
+#define SEC_KEY_2         (1<<1)
+#define SEC_KEY_3         (1<<2)
+#define SEC_KEY_4         (1<<3)
+#define SEC_ACTIVE_KEY    (1<<4)
+#define SEC_AUTH_MODE     (1<<5)
+#define SEC_UNICAST_GROUP (1<<6)
+#define SEC_LEVEL         (1<<7)
+#define SEC_ENABLED       (1<<8)
+
+#define SEC_LEVEL_0      0 /* None */
+#define SEC_LEVEL_1      1 /* WEP 40 and 104 bit */
+#define SEC_LEVEL_2      2 /* Level 1 + TKIP */
+#define SEC_LEVEL_2_CKIP 3 /* Level 1 + CKIP */
+#define SEC_LEVEL_3      4 /* Level 2 + CCMP */
+
+#define WEP_KEYS 4
+#define WEP_KEY_LEN 13
+
+struct ieee80211_security {
+	u16 active_key:2,
+            enabled:1,
+	    auth_mode:2,
+            auth_algo:4,
+            unicast_uses_group:1;
+	u8 key_sizes[WEP_KEYS];
+	u8 keys[WEP_KEYS][WEP_KEY_LEN];
+	u8 level;
+	u16 flags;
+} __attribute__ ((packed));
+
+
+/*
+
+ 802.11 data frame from AP
+
+      ,-------------------------------------------------------------------.
+Bytes |  2   |  2   |    6    |    6    |    6    |  2   | 0..2312 |   4  |
+      |------|------|---------|---------|---------|------|---------|------|
+Desc. | ctrl | dura |  DA/RA  |   TA    |    SA   | Sequ |  frame  |  fcs |
+      |      | tion | (BSSID) |         |         | ence |  data   |      |
+      `-------------------------------------------------------------------'
+
+Total: 28-2340 bytes
+
+*/
+
+struct ieee80211_header_data {
+	u16 frame_ctl;
+	u16 duration_id;
+	u8 addr1[6];
+	u8 addr2[6];
+	u8 addr3[6];
+	u16 seq_ctrl;
+};
+
+#define BEACON_PROBE_SSID_ID_POSITION 12
+
+/* Management Frame Information Element Types */
+#define MFIE_TYPE_SSID       0
+#define MFIE_TYPE_RATES      1
+#define MFIE_TYPE_FH_SET     2
+#define MFIE_TYPE_DS_SET     3
+#define MFIE_TYPE_CF_SET     4
+#define MFIE_TYPE_TIM        5
+#define MFIE_TYPE_IBSS_SET   6
+#define MFIE_TYPE_CHALLENGE  16
+#define MFIE_TYPE_RSN	     48
+#define MFIE_TYPE_RATES_EX   50
+#define MFIE_TYPE_GENERIC    221
+
+struct ieee80211_info_element_hdr {
+	u8 id;
+	u8 len;
+} __attribute__ ((packed));
+
+struct ieee80211_info_element {
+	u8 id;
+	u8 len;
+	u8 data[0];
+} __attribute__ ((packed));
+
+/*
+ * These are the data types that can make up management packets
+ *
+	u16 auth_algorithm;
+	u16 auth_sequence;
+	u16 beacon_interval;
+	u16 capability;
+	u8 current_ap[ETH_ALEN];
+	u16 listen_interval;
+	struct {
+		u16 association_id:14, reserved:2;
+	} __attribute__ ((packed));
+	u32 time_stamp[2];
+	u16 reason;
+	u16 status;
+*/
+
+struct ieee80211_authentication {
+	struct ieee80211_header_data header;
+	u16 algorithm;
+	u16 transaction;
+	u16 status;
+	struct ieee80211_info_element info_element;
+} __attribute__ ((packed));
+
+
+struct ieee80211_probe_response {
+	struct ieee80211_header_data header;
+	u32 time_stamp[2];
+	u16 beacon_interval;
+	u16 capability;
+	struct ieee80211_info_element info_element;
+} __attribute__ ((packed));
+
+struct ieee80211_assoc_request_frame {
+	u16 capability;
+	u16 listen_interval;
+	u8 current_ap[ETH_ALEN];
+	struct ieee80211_info_element info_element;
+} __attribute__ ((packed));
+
+struct ieee80211_assoc_response_frame {
+	struct ieee80211_hdr_3addr header;
+	u16 capability;
+	u16 status;
+	u16 aid;
+	struct ieee80211_info_element info_element; /* supported rates */
+} __attribute__ ((packed));
+
+
+struct ieee80211_txb {
+	u8 nr_frags;
+	u8 encrypted;
+	u16 reserved;
+	u16 frag_size;
+	u16 payload_size;
+	struct sk_buff *fragments[0];
+};
+
+
+/* SWEEP TABLE ENTRIES NUMBER*/
+#define MAX_SWEEP_TAB_ENTRIES		  42
+#define MAX_SWEEP_TAB_ENTRIES_PER_PACKET  7
+/* MAX_RATES_LENGTH needs to be 12.  The spec says 8, and many APs
+ * only use 8, and then use extended rates for the remaining supported
+ * rates.  Other APs, however, stick all of their supported rates on the
+ * main rates information element... */
+#define MAX_RATES_LENGTH                  ((u8)12)
+#define MAX_RATES_EX_LENGTH               ((u8)16)
+#define MAX_NETWORK_COUNT                  128
+
+#define CRC_LENGTH                 4U
+
+#define MAX_WPA_IE_LEN 64
+
+#define NETWORK_EMPTY_ESSID (1<<0)
+#define NETWORK_HAS_OFDM    (1<<1)
+#define NETWORK_HAS_CCK     (1<<2)
+
+struct ieee80211_network {
+	/* These entries are used to identify a unique network */
+	u8 bssid[ETH_ALEN];
+	u8 channel;
+	/* Ensure null-terminated for any debug msgs */
+	u8 ssid[IW_ESSID_MAX_SIZE + 1];
+	u8 ssid_len;
+
+	/* These are network statistics */
+	struct ieee80211_rx_stats stats;
+	u16 capability;
+	u8 rates[MAX_RATES_LENGTH];
+	u8 rates_len;
+	u8 rates_ex[MAX_RATES_EX_LENGTH];
+	u8 rates_ex_len;
+	unsigned long last_scanned;
+	u8 mode;
+	u8 flags;
+	u32 last_associate;
+	u32 time_stamp[2];
+	u16 beacon_interval;
+	u16 listen_interval;
+	u16 atim_window;
+	u8 wpa_ie[MAX_WPA_IE_LEN];
+	size_t wpa_ie_len;
+	u8 rsn_ie[MAX_WPA_IE_LEN];
+	size_t rsn_ie_len;
+	struct list_head list;
+};
+
+enum ieee80211_state {
+	IEEE80211_UNINITIALIZED = 0,
+	IEEE80211_INITIALIZED,
+	IEEE80211_ASSOCIATING,
+	IEEE80211_ASSOCIATED,
+	IEEE80211_AUTHENTICATING,
+	IEEE80211_AUTHENTICATED,
+	IEEE80211_SHUTDOWN
+};
+
+#define DEFAULT_MAX_SCAN_AGE (15 * HZ)
+#define DEFAULT_FTS 2346
+#define MAC_FMT "%02x:%02x:%02x:%02x:%02x:%02x"
+#define MAC_ARG(x) ((u8*)(x))[0],((u8*)(x))[1],((u8*)(x))[2],((u8*)(x))[3],((u8*)(x))[4],((u8*)(x))[5]
+
+
+extern inline int is_broadcast_ether_addr(const u8 *addr)
+{
+	return ((addr[0] == 0xff) && (addr[1] == 0xff) && (addr[2] == 0xff) &&   \
+		(addr[3] == 0xff) && (addr[4] == 0xff) && (addr[5] == 0xff));
+}
+
+#define CFG_IEEE80211_RESERVE_FCS (1<<0)
+#define CFG_IEEE80211_COMPUTE_FCS (1<<1)
+
+struct ieee80211_device {
+	struct net_device *dev;
+
+	/* Bookkeeping structures */
+	struct net_device_stats stats;
+	struct ieee80211_stats ieee_stats;
+
+	/* Probe / Beacon management */
+	struct list_head network_free_list;
+	struct list_head network_list;
+	struct ieee80211_network *networks;
+	int scans;
+	int scan_age;
+
+	int iw_mode; /* operating mode (IW_MODE_*) */
+
+	spinlock_t lock;
+
+	int tx_headroom; /* Set to size of any additional room needed at front
+			  * of allocated Tx SKBs */
+	u32 config;
+
+	/* WEP and other encryption related settings at the device level */
+	int open_wep; /* Set to 1 to allow unencrypted frames */
+
+	int reset_on_keychange; /* Set to 1 if the HW needs to be reset on
+				 * WEP key changes */
+
+	/* If the host performs {en,de}cryption, then set to 1 */
+	int host_encrypt;
+	int host_decrypt;
+	int ieee802_1x; /* is IEEE 802.1X used */
+
+	/* WPA data */
+	int wpa_enabled;
+	int drop_unencrypted;
+	int tkip_countermeasures;
+	int privacy_invoked;
+	size_t wpa_ie_len;
+	u8 *wpa_ie;
+
+	struct list_head crypt_deinit_list;
+	struct ieee80211_crypt_data *crypt[WEP_KEYS];
+	int tx_keyidx; /* default TX key index (crypt[tx_keyidx]) */
+	struct timer_list crypt_deinit_timer;
+
+	int bcrx_sta_key; /* use individual keys to override default keys even
+			   * with RX of broad/multicast frames */
+
+	/* Fragmentation structures */
+	struct ieee80211_frag_entry frag_cache[IEEE80211_FRAG_CACHE_LEN];
+	unsigned int frag_next_idx;
+	u16 fts; /* Fragmentation Threshold */
+
+	/* Association info */
+	u8 bssid[ETH_ALEN];
+
+	enum ieee80211_state state;
+
+	int mode;       /* A, B, G */
+	int modulation; /* CCK, OFDM */
+	int freq_band;  /* 2.4Ghz, 5.2Ghz, Mixed */
+	int abg_ture;   /* ABG flag              */
+
+	/* Callback functions */
+	void (*set_security)(struct net_device *dev,
+			     struct ieee80211_security *sec);
+	int (*hard_start_xmit)(struct ieee80211_txb *txb,
+			       struct net_device *dev);
+	int (*reset_port)(struct net_device *dev);
+
+	/* This must be the last item so that it points to the data
+	 * allocated beyond this structure by alloc_ieee80211 */
+	u8 priv[0];
+};
+
+#define IEEE_A            (1<<0)
+#define IEEE_B            (1<<1)
+#define IEEE_G            (1<<2)
+#define IEEE_MODE_MASK    (IEEE_A|IEEE_B|IEEE_G)
+
+extern inline void *ieee80211_priv(struct net_device *dev)
+{
+	return ((struct ieee80211_device *)netdev_priv(dev))->priv;
+}
+
+extern inline int ieee80211_is_empty_essid(const char *essid, int essid_len)
+{
+	/* Single white space is for Linksys APs */
+	if (essid_len == 1 && essid[0] == ' ')
+		return 1;
+
+	/* Otherwise, if the entire essid is 0, we assume it is hidden */
+	while (essid_len) {
+		essid_len--;
+		if (essid[essid_len] != '\0')
+			return 0;
+	}
+
+	return 1;
+}
+
+extern inline int ieee80211_is_valid_mode(struct ieee80211_device *ieee, int mode)
+{
+	/*
+	 * It is possible for both access points and our device to support
+	 * combinations of modes, so as long as there is one valid combination
+	 * of ap/device supported modes, then return success
+	 *
+	 */
+	if ((mode & IEEE_A) &&
+	    (ieee->modulation & IEEE80211_OFDM_MODULATION) &&
+	    (ieee->freq_band & IEEE80211_52GHZ_BAND))
+		return 1;
+
+	if ((mode & IEEE_G) &&
+	    (ieee->modulation & IEEE80211_OFDM_MODULATION) &&
+	    (ieee->freq_band & IEEE80211_24GHZ_BAND))
+		return 1;
+
+	if ((mode & IEEE_B) &&
+	    (ieee->modulation & IEEE80211_CCK_MODULATION) &&
+	    (ieee->freq_band & IEEE80211_24GHZ_BAND))
+		return 1;
+
+	return 0;
+}
+
+extern inline int ieee80211_get_hdrlen(u16 fc)
+{
+	int hdrlen = 24;
+
+	switch (WLAN_FC_GET_TYPE(fc)) {
+	case IEEE80211_FTYPE_DATA:
+		if ((fc & IEEE80211_FCTL_FROMDS) && (fc & IEEE80211_FCTL_TODS))
+			hdrlen = 30; /* Addr4 */
+		break;
+	case IEEE80211_FTYPE_CTL:
+		switch (WLAN_FC_GET_STYPE(fc)) {
+		case IEEE80211_STYPE_CTS:
+		case IEEE80211_STYPE_ACK:
+			hdrlen = 10;
+			break;
+		default:
+			hdrlen = 16;
+			break;
+		}
+		break;
+	}
+
+	return hdrlen;
+}
+
+
+
+/* ieee80211.c */
+extern void free_ieee80211(struct net_device *dev);
+extern struct net_device *alloc_ieee80211(int sizeof_priv);
+
+extern int ieee80211_set_encryption(struct ieee80211_device *ieee);
+
+/* ieee80211_tx.c */
+
+
+extern int ieee80211_xmit(struct sk_buff *skb,
+			  struct net_device *dev);
+extern void ieee80211_txb_free(struct ieee80211_txb *);
+
+
+/* ieee80211_rx.c */
+extern int ieee80211_rx(struct ieee80211_device *ieee, struct sk_buff *skb,
+			struct ieee80211_rx_stats *rx_stats);
+extern void ieee80211_rx_mgt(struct ieee80211_device *ieee,
+			     struct ieee80211_hdr *header,
+			     struct ieee80211_rx_stats *stats);
+
+/* iee80211_wx.c */
+extern int ieee80211_wx_get_scan(struct ieee80211_device *ieee,
+				 struct iw_request_info *info,
+				 union iwreq_data *wrqu, char *key);
+extern int ieee80211_wx_set_encode(struct ieee80211_device *ieee,
+				   struct iw_request_info *info,
+				   union iwreq_data *wrqu, char *key);
+extern int ieee80211_wx_get_encode(struct ieee80211_device *ieee,
+				   struct iw_request_info *info,
+				   union iwreq_data *wrqu, char *key);
+
+
+extern inline void ieee80211_increment_scans(struct ieee80211_device *ieee)
+{
+	ieee->scans++;
+}
+
+extern inline int ieee80211_get_scans(struct ieee80211_device *ieee)
+{
+	return ieee->scans;
+}
+
+static inline const char *escape_essid(const char *essid, u8 essid_len) {
+	static char escaped[IW_ESSID_MAX_SIZE * 2 + 1];
+	const char *s = essid;
+	char *d = escaped;
+
+	if (ieee80211_is_empty_essid(essid, essid_len)) {
+		memcpy(escaped, "<hidden>", sizeof("<hidden>"));
+		return escaped;
+	}
+
+	essid_len = min(essid_len, (u8)IW_ESSID_MAX_SIZE);
+	while (essid_len--) {
+		if (*s == '\0') {
+			*d++ = '\\';
+			*d++ = '0';
+			s++;
+		} else {
+			*d++ = *s++;
+		}
+	}
+	*d = '\0';
+	return escaped;
+}
+
+#endif /* IEEE80211_H */
-- 
cgit v1.2.3-70-g09d2


From ec9f47cd6a14ca069bb7552a984c0a338fc7262b Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 27 Jun 2005 11:12:54 +0100
Subject: [PATCH] Serial: Split 8250 port table

Add separate files for the different 8250 ISA-based serial boards.

Looking across all the various architectures, it seems reasonable that
we can key the availability of the configuration options for these
beasts to the bus-related symbols (iow, CONFIG_ISA).  We also standardise
the base baud/uart clock rate for these boards - I'm sure that isn't
architecture specific, but is solely dependent on the crystal fitted
on the board (which should be the same no matter what type of machine
its fitted into.)

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/serial/8250.c          | 25 +++++---------
 drivers/serial/8250_accent.c   | 47 ++++++++++++++++++++++++++
 drivers/serial/8250_boca.c     | 61 ++++++++++++++++++++++++++++++++++
 drivers/serial/8250_fourport.c | 53 +++++++++++++++++++++++++++++
 drivers/serial/8250_hub6.c     | 58 ++++++++++++++++++++++++++++++++
 drivers/serial/8250_mca.c      | 64 +++++++++++++++++++++++++++++++++++
 drivers/serial/Kconfig         | 75 +++++++++++++++++++++++++++++++++---------
 drivers/serial/Makefile        |  5 +++
 include/linux/serial_8250.h    |  1 +
 9 files changed, 357 insertions(+), 32 deletions(-)
 create mode 100644 drivers/serial/8250_accent.c
 create mode 100644 drivers/serial/8250_boca.c
 create mode 100644 drivers/serial/8250_fourport.c
 create mode 100644 drivers/serial/8250_hub6.c
 create mode 100644 drivers/serial/8250_mca.c

(limited to 'include')

diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index d8b9d2b8c20..34e75bc8f4c 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -77,23 +77,9 @@ static unsigned int share_irqs = SERIAL8250_SHARE_IRQS;
  */
 #define is_real_interrupt(irq)	((irq) != 0)
 
-/*
- * This converts from our new CONFIG_ symbols to the symbols
- * that asm/serial.h expects.  You _NEED_ to comment out the
- * linux/config.h include contained inside asm/serial.h for
- * this to work.
- */
-#undef CONFIG_SERIAL_MANY_PORTS
-#undef CONFIG_SERIAL_DETECT_IRQ
-#undef CONFIG_SERIAL_MULTIPORT
-#undef CONFIG_HUB6
-
 #ifdef CONFIG_SERIAL_8250_DETECT_IRQ
 #define CONFIG_SERIAL_DETECT_IRQ 1
 #endif
-#ifdef CONFIG_SERIAL_8250_MULTIPORT
-#define CONFIG_SERIAL_MULTIPORT 1
-#endif
 #ifdef CONFIG_SERIAL_8250_MANY_PORTS
 #define CONFIG_SERIAL_MANY_PORTS 1
 #endif
@@ -2323,10 +2309,11 @@ static int __devinit serial8250_probe(struct device *dev)
 {
 	struct plat_serial8250_port *p = dev->platform_data;
 	struct uart_port port;
+	int ret, i;
 
 	memset(&port, 0, sizeof(struct uart_port));
 
-	for (; p && p->flags != 0; p++) {
+	for (i = 0; p && p->flags != 0; p++, i++) {
 		port.iobase	= p->iobase;
 		port.membase	= p->membase;
 		port.irq	= p->irq;
@@ -2335,10 +2322,16 @@ static int __devinit serial8250_probe(struct device *dev)
 		port.iotype	= p->iotype;
 		port.flags	= p->flags;
 		port.mapbase	= p->mapbase;
+		port.hub6	= p->hub6;
 		port.dev	= dev;
 		if (share_irqs)
 			port.flags |= UPF_SHARE_IRQ;
-		serial8250_register_port(&port);
+		ret = serial8250_register_port(&port);
+		if (ret < 0) {
+			dev_err(dev, "unable to register port at index %d "
+				"(IO%lx MEM%lx IRQ%d): %d\n", i,
+				p->iobase, p->mapbase, p->irq, ret);
+		}
 	}
 	return 0;
 }
diff --git a/drivers/serial/8250_accent.c b/drivers/serial/8250_accent.c
new file mode 100644
index 00000000000..1f2c276063e
--- /dev/null
+++ b/drivers/serial/8250_accent.c
@@ -0,0 +1,47 @@
+/*
+ *  linux/drivers/serial/8250_accent.c
+ *
+ *  Copyright (C) 2005 Russell King.
+ *  Data taken from include/asm-i386/serial.h
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/serial_8250.h>
+
+#define PORT(_base,_irq)				\
+	{						\
+		.iobase		= _base,		\
+		.irq		= _irq,			\
+		.uartclk	= 1843200,		\
+		.iotype		= UPIO_PORT,		\
+		.flags		= UPF_BOOT_AUTOCONF,	\
+	}
+
+static struct plat_serial8250_port accent_data[] = {
+	PORT(0x330, 4),
+	PORT(0x338, 4),
+	{ },
+};
+
+static struct platform_device accent_device = {
+	.name			= "serial8250",
+	.id			= 2,
+	.dev			= {
+		.platform_data	= accent_data,
+	},
+};
+
+static int __init accent_init(void)
+{
+	return platform_device_register(&accent_device);
+}
+
+module_init(accent_init);
+
+MODULE_AUTHOR("Russell King");
+MODULE_DESCRIPTION("8250 serial probe module for Accent Async cards");
+MODULE_LICENSE("GPL");
diff --git a/drivers/serial/8250_boca.c b/drivers/serial/8250_boca.c
new file mode 100644
index 00000000000..465c9ea1e7a
--- /dev/null
+++ b/drivers/serial/8250_boca.c
@@ -0,0 +1,61 @@
+/*
+ *  linux/drivers/serial/8250_boca.c
+ *
+ *  Copyright (C) 2005 Russell King.
+ *  Data taken from include/asm-i386/serial.h
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/serial_8250.h>
+
+#define PORT(_base,_irq)				\
+	{						\
+		.iobase		= _base,		\
+		.irq		= _irq,			\
+		.uartclk	= 1843200,		\
+		.iotype		= UPIO_PORT,		\
+		.flags		= UPF_BOOT_AUTOCONF,	\
+	}
+
+static struct plat_serial8250_port boca_data[] = {
+	PORT(0x100, 12),
+	PORT(0x108, 12),
+	PORT(0x110, 12),
+	PORT(0x118, 12),
+	PORT(0x120, 12),
+	PORT(0x128, 12),
+	PORT(0x130, 12),
+	PORT(0x138, 12),
+	PORT(0x140, 12),
+	PORT(0x148, 12),
+	PORT(0x150, 12),
+	PORT(0x158, 12),
+	PORT(0x160, 12),
+	PORT(0x168, 12),
+	PORT(0x170, 12),
+	PORT(0x178, 12),
+	{ },
+};
+
+static struct platform_device boca_device = {
+	.name			= "serial8250",
+	.id			= 3,
+	.dev			= {
+		.platform_data	= boca_data,
+	},
+};
+
+static int __init boca_init(void)
+{
+	return platform_device_register(&boca_device);
+}
+
+module_init(boca_init);
+
+MODULE_AUTHOR("Russell King");
+MODULE_DESCRIPTION("8250 serial probe module for Boca cards");
+MODULE_LICENSE("GPL");
diff --git a/drivers/serial/8250_fourport.c b/drivers/serial/8250_fourport.c
new file mode 100644
index 00000000000..e9b4d908ef4
--- /dev/null
+++ b/drivers/serial/8250_fourport.c
@@ -0,0 +1,53 @@
+/*
+ *  linux/drivers/serial/8250_fourport.c
+ *
+ *  Copyright (C) 2005 Russell King.
+ *  Data taken from include/asm-i386/serial.h
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/serial_8250.h>
+
+#define PORT(_base,_irq)						\
+	{								\
+		.iobase		= _base,				\
+		.irq		= _irq,					\
+		.uartclk	= 1843200,				\
+		.iotype		= UPIO_PORT,				\
+		.flags		= UPF_BOOT_AUTOCONF | UPF_FOURPORT,	\
+	}
+
+static struct plat_serial8250_port fourport_data[] = {
+	PORT(0x1a0, 9),
+	PORT(0x1a8, 9),
+	PORT(0x1b0, 9),
+	PORT(0x1b8, 9),
+	PORT(0x2a0, 5),
+	PORT(0x2a8, 5),
+	PORT(0x2b0, 5),
+	PORT(0x2b8, 5),
+	{ },
+};
+
+static struct platform_device fourport_device = {
+	.name			= "serial8250",
+	.id			= 1,
+	.dev			= {
+		.platform_data	= fourport_data,
+	},
+};
+
+static int __init fourport_init(void)
+{
+	return platform_device_register(&fourport_device);
+}
+
+module_init(fourport_init);
+
+MODULE_AUTHOR("Russell King");
+MODULE_DESCRIPTION("8250 serial probe module for AST Fourport cards");
+MODULE_LICENSE("GPL");
diff --git a/drivers/serial/8250_hub6.c b/drivers/serial/8250_hub6.c
new file mode 100644
index 00000000000..77f396f84b4
--- /dev/null
+++ b/drivers/serial/8250_hub6.c
@@ -0,0 +1,58 @@
+/*
+ *  linux/drivers/serial/8250_hub6.c
+ *
+ *  Copyright (C) 2005 Russell King.
+ *  Data taken from include/asm-i386/serial.h
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/serial_8250.h>
+
+#define HUB6(card,port)							\
+	{								\
+		.iobase		= 0x302,				\
+		.irq		= 3,					\
+		.uartclk	= 1843200,				\
+		.iotype		= UPIO_HUB6,				\
+		.flags		= UPF_BOOT_AUTOCONF,			\
+		.hub6		= (card) << 6 | (port) << 3 | 1,	\
+	}
+
+static struct plat_serial8250_port hub6_data[] = {
+	HUB6(0,0),
+	HUB6(0,1),
+	HUB6(0,2),
+	HUB6(0,3),
+	HUB6(0,4),
+	HUB6(0,5),
+	HUB6(1,0),
+	HUB6(1,1),
+	HUB6(1,2),
+	HUB6(1,3),
+	HUB6(1,4),
+	HUB6(1,5),
+	{ },
+};
+
+static struct platform_device hub6_device = {
+	.name			= "serial8250",
+	.id			= 4,
+	.dev			= {
+		.platform_data	= hub6_data,
+	},
+};
+
+static int __init hub6_init(void)
+{
+	return platform_device_register(&hub6_device);
+}
+
+module_init(hub6_init);
+
+MODULE_AUTHOR("Russell King");
+MODULE_DESCRIPTION("8250 serial probe module for Hub6 cards");
+MODULE_LICENSE("GPL");
diff --git a/drivers/serial/8250_mca.c b/drivers/serial/8250_mca.c
new file mode 100644
index 00000000000..f0c40d68b8c
--- /dev/null
+++ b/drivers/serial/8250_mca.c
@@ -0,0 +1,64 @@
+/*
+ *  linux/drivers/serial/8250_mca.c
+ *
+ *  Copyright (C) 2005 Russell King.
+ *  Data taken from include/asm-i386/serial.h
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/mca.h>
+#include <linux/serial_8250.h>
+
+/*
+ * FIXME: Should we be doing AUTO_IRQ here?
+ */
+#ifdef CONFIG_SERIAL_8250_DETECT_IRQ
+#define MCA_FLAGS	UPF_BOOT_AUTOCONF | UPF_SKIP_TEST | UPF_AUTO_IRQ
+#else
+#define MCA_FLAGS	UPF_BOOT_AUTOCONF | UPF_SKIP_TEST
+#endif
+
+#define PORT(_base,_irq)			\
+	{					\
+		.iobase		= _base,	\
+		.irq		= _irq,		\
+		.uartclk	= 1843200,	\
+		.iotype		= UPIO_PORT,	\
+		.flags		= MCA_FLAGS,	\
+	}
+
+static struct plat_serial8250_port mca_data[] = {
+	PORT(0x3220, 3),
+	PORT(0x3228, 3),
+	PORT(0x4220, 3),
+	PORT(0x4228, 3),
+	PORT(0x5220, 3),
+	PORT(0x5228, 3),
+	{ },
+};
+
+static struct platform_device mca_device = {
+	.name			= "serial8250",
+	.id			= 5,
+	.dev			= {
+		.platform_data	= mca_data,
+	},
+};
+
+static int __init mca_init(void)
+{
+	if (!MCA_bus)
+		return -ENODEV;
+	return platform_device_register(&mca_device);
+}
+
+module_init(mca_init);
+
+MODULE_AUTHOR("Russell King");
+MODULE_DESCRIPTION("8250 serial probe module for MCA ports");
+MODULE_LICENSE("GPL");
diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig
index 25fcef2c42d..e879bce160d 100644
--- a/drivers/serial/Kconfig
+++ b/drivers/serial/Kconfig
@@ -89,11 +89,11 @@ config SERIAL_8250_NR_UARTS
 	int "Maximum number of non-legacy 8250/16550 serial ports"
 	depends on SERIAL_8250
 	default "4"
-	---help---
-	  Set this to the number of non-legacy serial ports you want
-	  the driver to support.  This includes any ports discovered
-	  via ACPI or PCI enumeration and any ports that may be added
-	  at run-time via hot-plug.
+	help
+	  Set this to the number of serial ports you want the driver
+	  to support.  This includes any ports discovered via ACPI or
+	  PCI enumeration and any ports that may be added at run-time
+	  via hot-plug, or any ISA multi-port serial cards.
 
 config SERIAL_8250_EXTENDED
 	bool "Extended 8250/16550 serial driver options"
@@ -141,31 +141,74 @@ config SERIAL_8250_DETECT_IRQ
 
 	  If unsure, say N.
 
-config SERIAL_8250_MULTIPORT
-	bool "Support special multiport boards"
-	depends on SERIAL_8250_EXTENDED
-	help
-	  Some multiport serial ports have special ports which are used to
-	  signal when there are any serial ports on the board which need
-	  servicing. Say Y here to enable the serial driver to take advantage
-	  of those special I/O ports.
-
 config SERIAL_8250_RSA
 	bool "Support RSA serial ports"
 	depends on SERIAL_8250_EXTENDED
 	help
 	  ::: To be written :::
 
-comment "Non-8250 serial port support"
+#
+# Multi-port serial cards
+#
+
+config SERIAL_8250_FOURPORT
+	tristate "Support Fourport cards"
+	depends on SERIAL_8250 != n && ISA && SERIAL_8250_MANY_PORTS
+	help
+	  Say Y here if you have an AST FourPort serial board.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called 8250_fourport.
+
+config SERIAL_8250_ACCENT
+	tristate "Support Accent cards"
+	depends on SERIAL_8250 != n && ISA && SERIAL_8250_MANY_PORTS
+	help
+	  Say Y here if you have an Accent Async serial board.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called 8250_accent.
+
+
+config SERIAL_8250_BOCA
+	tristate "Support Boca cards"
+	depends on SERIAL_8250 != n && ISA && SERIAL_8250_MANY_PORTS
+	help
+	  Say Y here if you have a Boca serial board.  Please read the Boca
+	  mini-HOWTO, avaialble from <http://www.tldp.org/docs.html#howto>
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called 8250_boca.
+
+
+config SERIAL_8250_HUB6
+	tristate "Support Hub6 cards"
+	depends on SERIAL_8250 != n && ISA && SERIAL_8250_MANY_PORTS
+	help
+	  Say Y here if you have a HUB6 serial board.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called 8250_hub6.
+
+config SERIAL_8250_MCA
+	tristate "Support 8250-type ports on MCA buses"
+	depends on SERIAL_8250 != n && MCA
+	help
+	  Say Y here if you have a MCA serial ports.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called 8250_mca.
 
 config SERIAL_8250_ACORN
 	tristate "Acorn expansion card serial port support"
-	depends on ARM && ARCH_ACORN && SERIAL_8250
+	depends on ARCH_ACORN && SERIAL_8250
 	help
 	  If you have an Atomwide Serial card or Serial Port card for an Acorn
 	  system, say Y to this option.  The driver can handle 1, 2, or 3 port
 	  cards.  If unsure, say N.
 
+comment "Non-8250 serial port support"
+
 config SERIAL_AMBA_PL010
 	tristate "ARM AMBA PL010 serial port support"
 	depends on ARM_AMBA
diff --git a/drivers/serial/Makefile b/drivers/serial/Makefile
index 8f1cdde7dbe..65bd4381685 100644
--- a/drivers/serial/Makefile
+++ b/drivers/serial/Makefile
@@ -17,6 +17,11 @@ obj-$(CONFIG_SERIAL_8250) += 8250.o $(serial-8250-y)
 obj-$(CONFIG_SERIAL_8250_CS) += serial_cs.o
 obj-$(CONFIG_SERIAL_8250_ACORN) += 8250_acorn.o
 obj-$(CONFIG_SERIAL_8250_CONSOLE) += 8250_early.o
+obj-$(CONFIG_SERIAL_8250_FOURPORT) += 8250_fourport.o
+obj-$(CONFIG_SERIAL_8250_ACCENT) += 8250_accent.o
+obj-$(CONFIG_SERIAL_8250_BOCA) += 8250_boca.o
+obj-$(CONFIG_SERIAL_8250_HUB6) += 8250_hub6.o
+obj-$(CONFIG_SERIAL_8250_MCA) += 8250_mca.o
 obj-$(CONFIG_SERIAL_AMBA_PL010) += amba-pl010.o
 obj-$(CONFIG_SERIAL_AMBA_PL011) += amba-pl011.o
 obj-$(CONFIG_SERIAL_CLPS711X) += clps711x.o
diff --git a/include/linux/serial_8250.h b/include/linux/serial_8250.h
index 823181af6dd..3e3c1fa35b0 100644
--- a/include/linux/serial_8250.h
+++ b/include/linux/serial_8250.h
@@ -22,6 +22,7 @@ struct plat_serial8250_port {
 	unsigned int	uartclk;	/* UART clock rate */
 	unsigned char	regshift;	/* register shift */
 	unsigned char	iotype;		/* UPIO_* */
+	unsigned char	hub6;
 	unsigned int	flags;		/* UPF_* flags */
 };
 
-- 
cgit v1.2.3-70-g09d2


From 2ea83398b75309d8fdc999c4bb252e72d7e4fd9d Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 27 Jun 2005 14:04:05 +0100
Subject: [PATCH] ARM: Add VST idle loop call

This call allows the dynamic tick support to reprogram the timer
immediately before the CPU idles.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/process.c   |  5 ++++-
 arch/arm/kernel/time.c      | 10 +++++++---
 include/asm-arm/mach/time.h |  2 ++
 3 files changed, 13 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
index 8f146a4b475..bbea636ff68 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
@@ -32,6 +32,7 @@
 #include <asm/leds.h>
 #include <asm/processor.h>
 #include <asm/uaccess.h>
+#include <asm/mach/time.h>
 
 extern const char *processor_modes[];
 extern void setup_mm_for_reboot(char mode);
@@ -85,8 +86,10 @@ EXPORT_SYMBOL(pm_power_off);
 void default_idle(void)
 {
 	local_irq_disable();
-	if (!need_resched() && !hlt_counter)
+	if (!need_resched() && !hlt_counter) {
+		timer_dyn_reprogram();
 		arch_idle();
+	}
 	local_irq_enable();
 }
 
diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c
index 06054c9ba07..1b7fcd50c3e 100644
--- a/arch/arm/kernel/time.c
+++ b/arch/arm/kernel/time.c
@@ -424,15 +424,19 @@ static int timer_dyn_tick_disable(void)
 	return ret;
 }
 
+/*
+ * Reprogram the system timer for at least the calculated time interval.
+ * This function should be called from the idle thread with IRQs disabled,
+ * immediately before sleeping.
+ */
 void timer_dyn_reprogram(void)
 {
 	struct dyn_tick_timer *dyn_tick = system_timer->dyn_tick;
-	unsigned long flags;
 
-	write_seqlock_irqsave(&xtime_lock, flags);
+	write_seqlock(&xtime_lock);
 	if (dyn_tick->state & DYN_TICK_ENABLED)
 		dyn_tick->reprogram(next_timer_interrupt() - jiffies);
-	write_sequnlock_irqrestore(&xtime_lock, flags);
+	write_sequnlock(&xtime_lock);
 }
 
 static ssize_t timer_show_dyn_tick(struct sys_device *dev, char *buf)
diff --git a/include/asm-arm/mach/time.h b/include/asm-arm/mach/time.h
index 047980ad18d..2cf279a4401 100644
--- a/include/asm-arm/mach/time.h
+++ b/include/asm-arm/mach/time.h
@@ -60,6 +60,8 @@ struct dyn_tick_timer {
 };
 
 void timer_dyn_reprogram(void);
+#else
+#define timer_dyn_reprogram()	do { } while (0)
 #endif
 
 extern struct sys_timer *system_timer;
-- 
cgit v1.2.3-70-g09d2


From c79ebfa8d2c126815e67cbf0dc2cb185dcae59f9 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 27 Jun 2005 14:23:38 +0100
Subject: [PATCH] ARM: Fix speeling eroor in io.h

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/io.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-arm/io.h b/include/asm-arm/io.h
index cc4b5f5dbfc..cfa71a0dffb 100644
--- a/include/asm-arm/io.h
+++ b/include/asm-arm/io.h
@@ -82,7 +82,7 @@ extern void __readwrite_bug(const char *fn);
  * only.  Their primary purpose is to access PCI and ISA peripherals.
  *
  * Note that for a big endian machine, this implies that the following
- * big endian mode connectivity is in place, as described by numerious
+ * big endian mode connectivity is in place, as described by numerous
  * ARM documents:
  *
  *    PCI:  D0-D7   D8-D15 D16-D23 D24-D31
-- 
cgit v1.2.3-70-g09d2


From eb4e10d091b369027ec205163e06f7eec335996f Mon Sep 17 00:00:00 2001
From: Michael Burian <dynmail1@gassner-waagen.at>
Date: Mon, 27 Jun 2005 15:58:39 +0100
Subject: [PATCH] ARM: 2758/1: Fix comment in file header to read "ARM" instead
 i386

Patch from Michael Burian

This does not look like an include file for "i386", so use "ARM" instead.

Signed-off-by: Michael Burian <dynmail1@gassner-waagen.at>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/ide.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-arm/ide.h b/include/asm-arm/ide.h
index 2114acb3d23..4f68c8a5a19 100644
--- a/include/asm-arm/ide.h
+++ b/include/asm-arm/ide.h
@@ -5,7 +5,7 @@
  */
 
 /*
- *  This file contains the i386 architecture specific IDE code.
+ *  This file contains the ARM architecture specific IDE code.
  */
 
 #ifndef __ASMARM_IDE_H
-- 
cgit v1.2.3-70-g09d2


From b46ffaefe3c251897de23e5a534dd922c969d67b Mon Sep 17 00:00:00 2001
From: Deepak Saxena <dsaxena@plexity.net>
Date: Mon, 27 Jun 2005 21:48:48 +0100
Subject: [PATCH] ARM: 2759/1: Fix IXP4xx debug code (again)

Patch from Deepak Saxena

Accidently swapped the order of movne and orrne. Bad.

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-ixp4xx/debug-macro.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-ixp4xx/debug-macro.S b/include/asm-arm/arch-ixp4xx/debug-macro.S
index 45a6c6cc29d..2e23651e217 100644
--- a/include/asm-arm/arch-ixp4xx/debug-macro.S
+++ b/include/asm-arm/arch-ixp4xx/debug-macro.S
@@ -14,8 +14,8 @@
                 mrc     p15, 0, \rx, c1, c0
                 tst     \rx, #1                 @ MMU enabled?
                 moveq   \rx, #0xc8000000
-		orrne	\rx, \rx, #0x00b00000
                 movne   \rx, #0xff000000
+		orrne	\rx, \rx, #0x00b00000
                 add     \rx,\rx,#3              @ Uart regs are at off set of 3 if
 						@ byte writes used - Big Endian.
                 .endm
-- 
cgit v1.2.3-70-g09d2


From 22e2c507c301c3dbbcf91b4948b88f78842ee6c9 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Mon, 27 Jun 2005 10:55:12 +0200
Subject: [PATCH] Update cfq io scheduler to time sliced design

This updates the CFQ io scheduler to the new time sliced design (cfq
v3).  It provides full process fairness, while giving excellent
aggregate system throughput even for many competing processes.  It
supports io priorities, either inherited from the cpu nice value or set
directly with the ioprio_get/set syscalls.  The latter closely mimic
set/getpriority.

This import is based on my latest from -mm.

Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/syscall_table.S |    2 +
 arch/ia64/kernel/entry.S         |    4 +-
 arch/ppc/kernel/misc.S           |    2 +
 drivers/block/as-iosched.c       |    5 +-
 drivers/block/cfq-iosched.c      | 1910 +++++++++++++++++++++++++-------------
 drivers/block/deadline-iosched.c |    3 +-
 drivers/block/elevator.c         |    9 +-
 drivers/block/ll_rw_blk.c        |   59 +-
 fs/Makefile                      |    1 +
 fs/ioprio.c                      |  172 ++++
 fs/reiserfs/journal.c            |   12 +
 include/asm-i386/unistd.h        |    4 +-
 include/asm-ia64/unistd.h        |    2 +
 include/asm-ppc/unistd.h         |    4 +-
 include/asm-x86_64/unistd.h      |    6 +-
 include/linux/bio.h              |   14 +
 include/linux/blkdev.h           |   25 +-
 include/linux/elevator.h         |    8 +-
 include/linux/fs.h               |   19 +
 include/linux/init_task.h        |    2 +
 include/linux/ioprio.h           |   87 ++
 include/linux/sched.h            |    6 +-
 include/linux/writeback.h        |    6 +-
 kernel/exit.c                    |    2 +
 kernel/fork.c                    |    5 +
 kernel/sched.c                   |    8 -
 26 files changed, 1685 insertions(+), 692 deletions(-)
 create mode 100644 fs/ioprio.c
 create mode 100644 include/linux/ioprio.h

(limited to 'include')

diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S
index 442a6e937b1..3db9a04aec6 100644
--- a/arch/i386/kernel/syscall_table.S
+++ b/arch/i386/kernel/syscall_table.S
@@ -289,3 +289,5 @@ ENTRY(sys_call_table)
 	.long sys_add_key
 	.long sys_request_key
 	.long sys_keyctl
+	.long sys_ioprio_set
+	.long sys_ioprio_get		/* 290 */
diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index b1d5d3d5276..785a51b0ad8 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -1577,8 +1577,8 @@ sys_call_table:
 	data8 sys_add_key
 	data8 sys_request_key
 	data8 sys_keyctl
-	data8 sys_ni_syscall
-	data8 sys_ni_syscall			// 1275
+	data8 sys_ioprio_set
+	data8 sys_ioprio_get			// 1275
 	data8 sys_set_zone_reclaim
 	data8 sys_ni_syscall
 	data8 sys_ni_syscall
diff --git a/arch/ppc/kernel/misc.S b/arch/ppc/kernel/misc.S
index b6a63a49a23..191a8def3bd 100644
--- a/arch/ppc/kernel/misc.S
+++ b/arch/ppc/kernel/misc.S
@@ -1449,3 +1449,5 @@ _GLOBAL(sys_call_table)
 	.long sys_request_key		/* 270 */
 	.long sys_keyctl
 	.long sys_waitid
+	.long sys_ioprio_set
+	.long sys_ioprio_get
diff --git a/drivers/block/as-iosched.c b/drivers/block/as-iosched.c
index 3410b4d294b..91aeb678135 100644
--- a/drivers/block/as-iosched.c
+++ b/drivers/block/as-iosched.c
@@ -1806,7 +1806,8 @@ static void as_put_request(request_queue_t *q, struct request *rq)
 	rq->elevator_private = NULL;
 }
 
-static int as_set_request(request_queue_t *q, struct request *rq, int gfp_mask)
+static int as_set_request(request_queue_t *q, struct request *rq,
+			  struct bio *bio, int gfp_mask)
 {
 	struct as_data *ad = q->elevator->elevator_data;
 	struct as_rq *arq = mempool_alloc(ad->arq_pool, gfp_mask);
@@ -1827,7 +1828,7 @@ static int as_set_request(request_queue_t *q, struct request *rq, int gfp_mask)
 	return 1;
 }
 
-static int as_may_queue(request_queue_t *q, int rw)
+static int as_may_queue(request_queue_t *q, int rw, struct bio *bio)
 {
 	int ret = ELV_MQUEUE_MAY;
 	struct as_data *ad = q->elevator->elevator_data;
diff --git a/drivers/block/cfq-iosched.c b/drivers/block/cfq-iosched.c
index 3ac47dde64d..35f6e569d5e 100644
--- a/drivers/block/cfq-iosched.c
+++ b/drivers/block/cfq-iosched.c
@@ -21,22 +21,33 @@
 #include <linux/hash.h>
 #include <linux/rbtree.h>
 #include <linux/mempool.h>
-
-static unsigned long max_elapsed_crq;
-static unsigned long max_elapsed_dispatch;
+#include <linux/ioprio.h>
+#include <linux/writeback.h>
 
 /*
  * tunables
  */
 static int cfq_quantum = 4;		/* max queue in one round of service */
 static int cfq_queued = 8;		/* minimum rq allocate limit per-queue*/
-static int cfq_service = HZ;		/* period over which service is avg */
-static int cfq_fifo_expire_r = HZ / 2;	/* fifo timeout for sync requests */
-static int cfq_fifo_expire_w = 5 * HZ;	/* fifo timeout for async requests */
-static int cfq_fifo_rate = HZ / 8;	/* fifo expiry rate */
+static int cfq_fifo_expire[2] = { HZ / 4, HZ / 8 };
 static int cfq_back_max = 16 * 1024;	/* maximum backwards seek, in KiB */
 static int cfq_back_penalty = 2;	/* penalty of a backwards seek */
 
+static int cfq_slice_sync = HZ / 10;
+static int cfq_slice_async = HZ / 50;
+static int cfq_slice_async_rq = 2;
+static int cfq_slice_idle = HZ / 50;
+
+#define CFQ_IDLE_GRACE		(HZ / 10)
+#define CFQ_SLICE_SCALE		(5)
+
+#define CFQ_KEY_ASYNC		(0)
+
+/*
+ * disable queueing at the driver/hardware level
+ */
+static int cfq_max_depth = 1;
+
 /*
  * for the hash of cfqq inside the cfqd
  */
@@ -55,6 +66,7 @@ static int cfq_back_penalty = 2;	/* penalty of a backwards seek */
 #define list_entry_hash(ptr)	hlist_entry((ptr), struct cfq_rq, hash)
 
 #define list_entry_cfqq(ptr)	list_entry((ptr), struct cfq_queue, cfq_list)
+#define list_entry_fifo(ptr)	list_entry((ptr), struct request, queuelist)
 
 #define RQ_DATA(rq)		(rq)->elevator_private
 
@@ -75,78 +87,101 @@ static int cfq_back_penalty = 2;	/* penalty of a backwards seek */
 #define rb_entry_crq(node)	rb_entry((node), struct cfq_rq, rb_node)
 #define rq_rb_key(rq)		(rq)->sector
 
-/*
- * threshold for switching off non-tag accounting
- */
-#define CFQ_MAX_TAG		(4)
-
-/*
- * sort key types and names
- */
-enum {
-	CFQ_KEY_PGID,
-	CFQ_KEY_TGID,
-	CFQ_KEY_UID,
-	CFQ_KEY_GID,
-	CFQ_KEY_LAST,
-};
-
-static char *cfq_key_types[] = { "pgid", "tgid", "uid", "gid", NULL };
-
 static kmem_cache_t *crq_pool;
 static kmem_cache_t *cfq_pool;
 static kmem_cache_t *cfq_ioc_pool;
 
+#define CFQ_PRIO_LISTS		IOPRIO_BE_NR
+#define cfq_class_idle(cfqq)	((cfqq)->ioprio_class == IOPRIO_CLASS_IDLE)
+#define cfq_class_be(cfqq)	((cfqq)->ioprio_class == IOPRIO_CLASS_BE)
+#define cfq_class_rt(cfqq)	((cfqq)->ioprio_class == IOPRIO_CLASS_RT)
+
+#define cfq_cfqq_sync(cfqq)	((cfqq)->key != CFQ_KEY_ASYNC)
+
+/*
+ * Per block device queue structure
+ */
 struct cfq_data {
-	struct list_head rr_list;
+	atomic_t ref;
+	request_queue_t *queue;
+
+	/*
+	 * rr list of queues with requests and the count of them
+	 */
+	struct list_head rr_list[CFQ_PRIO_LISTS];
+	struct list_head busy_rr;
+	struct list_head cur_rr;
+	struct list_head idle_rr;
+	unsigned int busy_queues;
+
+	/*
+	 * non-ordered list of empty cfqq's
+	 */
 	struct list_head empty_list;
 
+	/*
+	 * cfqq lookup hash
+	 */
 	struct hlist_head *cfq_hash;
-	struct hlist_head *crq_hash;
 
-	/* queues on rr_list (ie they have pending requests */
-	unsigned int busy_queues;
+	/*
+	 * global crq hash for all queues
+	 */
+	struct hlist_head *crq_hash;
 
 	unsigned int max_queued;
 
-	atomic_t ref;
+	mempool_t *crq_pool;
 
-	int key_type;
+	int rq_in_driver;
 
-	mempool_t *crq_pool;
+	/*
+	 * schedule slice state info
+	 */
+	/*
+	 * idle window management
+	 */
+	struct timer_list idle_slice_timer;
+	struct work_struct unplug_work;
 
-	request_queue_t *queue;
+	struct cfq_queue *active_queue;
+	struct cfq_io_context *active_cic;
+	int cur_prio, cur_end_prio;
+	unsigned int dispatch_slice;
+
+	struct timer_list idle_class_timer;
 
 	sector_t last_sector;
+	unsigned long last_end_request;
 
-	int rq_in_driver;
+	unsigned int rq_starved;
 
 	/*
 	 * tunables, see top of file
 	 */
 	unsigned int cfq_quantum;
 	unsigned int cfq_queued;
-	unsigned int cfq_fifo_expire_r;
-	unsigned int cfq_fifo_expire_w;
-	unsigned int cfq_fifo_batch_expire;
+	unsigned int cfq_fifo_expire[2];
 	unsigned int cfq_back_penalty;
 	unsigned int cfq_back_max;
-	unsigned int find_best_crq;
-
-	unsigned int cfq_tagged;
+	unsigned int cfq_slice[2];
+	unsigned int cfq_slice_async_rq;
+	unsigned int cfq_slice_idle;
+	unsigned int cfq_max_depth;
 };
 
+/*
+ * Per process-grouping structure
+ */
 struct cfq_queue {
 	/* reference count */
 	atomic_t ref;
 	/* parent cfq_data */
 	struct cfq_data *cfqd;
-	/* hash of mergeable requests */
+	/* cfqq lookup hash */
 	struct hlist_node cfq_hash;
 	/* hash key */
-	unsigned long key;
-	/* whether queue is on rr (or empty) list */
-	int on_rr;
+	unsigned int key;
 	/* on either rr or empty list of cfqd */
 	struct list_head cfq_list;
 	/* sorted list of pending requests */
@@ -158,21 +193,35 @@ struct cfq_queue {
 	/* currently allocated requests */
 	int allocated[2];
 	/* fifo list of requests in sort_list */
-	struct list_head fifo[2];
-	/* last time fifo expired */
-	unsigned long last_fifo_expire;
-
-	int key_type;
-
-	unsigned long service_start;
-	unsigned long service_used;
+	struct list_head fifo;
 
-	unsigned int max_rate;
+	unsigned long slice_start;
+	unsigned long slice_end;
+	unsigned long slice_left;
+	unsigned long service_last;
 
 	/* number of requests that have been handed to the driver */
 	int in_flight;
-	/* number of currently allocated requests */
-	int alloc_limit[2];
+
+	/* io prio of this group */
+	unsigned short ioprio, org_ioprio;
+	unsigned short ioprio_class, org_ioprio_class;
+
+	/* whether queue is on rr (or empty) list */
+	unsigned on_rr : 1;
+	/* idle slice, waiting for new request submission */
+	unsigned wait_request : 1;
+	/* set when wait_request gets set, reset on first rq alloc */
+	unsigned must_alloc : 1;
+	/* only gets one must_alloc per slice */
+	unsigned must_alloc_slice : 1;
+	/* idle slice, request added, now waiting to dispatch it */
+	unsigned must_dispatch : 1;
+	/* fifo expire per-slice */
+	unsigned fifo_expire : 1;
+
+	unsigned idle_window : 1;
+	unsigned prio_changed : 1;
 };
 
 struct cfq_rq {
@@ -184,42 +233,17 @@ struct cfq_rq {
 	struct cfq_queue *cfq_queue;
 	struct cfq_io_context *io_context;
 
-	unsigned long service_start;
-	unsigned long queue_start;
-
-	unsigned int in_flight : 1;
-	unsigned int accounted : 1;
-	unsigned int is_sync   : 1;
-	unsigned int is_write  : 1;
+	unsigned in_flight : 1;
+	unsigned accounted : 1;
+	unsigned is_sync   : 1;
+	unsigned requeued  : 1;
 };
 
-static struct cfq_queue *cfq_find_cfq_hash(struct cfq_data *, unsigned long);
+static struct cfq_queue *cfq_find_cfq_hash(struct cfq_data *, unsigned int);
 static void cfq_dispatch_sort(request_queue_t *, struct cfq_rq *);
-static void cfq_update_next_crq(struct cfq_rq *);
 static void cfq_put_cfqd(struct cfq_data *cfqd);
 
-/*
- * what the fairness is based on (ie how processes are grouped and
- * differentiated)
- */
-static inline unsigned long
-cfq_hash_key(struct cfq_data *cfqd, struct task_struct *tsk)
-{
-	/*
-	 * optimize this so that ->key_type is the offset into the struct
-	 */
-	switch (cfqd->key_type) {
-		case CFQ_KEY_PGID:
-			return process_group(tsk);
-		default:
-		case CFQ_KEY_TGID:
-			return tsk->tgid;
-		case CFQ_KEY_UID:
-			return tsk->uid;
-		case CFQ_KEY_GID:
-			return tsk->gid;
-	}
-}
+#define process_sync(tsk)	((tsk)->flags & PF_SYNCWRITE)
 
 /*
  * lots of deadline iosched dupes, can be abstracted later...
@@ -235,16 +259,12 @@ static void cfq_remove_merge_hints(request_queue_t *q, struct cfq_rq *crq)
 
 	if (q->last_merge == crq->request)
 		q->last_merge = NULL;
-
-	cfq_update_next_crq(crq);
 }
 
 static inline void cfq_add_crq_hash(struct cfq_data *cfqd, struct cfq_rq *crq)
 {
 	const int hash_idx = CFQ_MHASH_FN(rq_hash_key(crq->request));
 
-	BUG_ON(!hlist_unhashed(&crq->hash));
-
 	hlist_add_head(&crq->hash, &cfqd->crq_hash[hash_idx]);
 }
 
@@ -257,8 +277,6 @@ static struct request *cfq_find_rq_hash(struct cfq_data *cfqd, sector_t offset)
 		struct cfq_rq *crq = list_entry_hash(entry);
 		struct request *__rq = crq->request;
 
-		BUG_ON(hlist_unhashed(&crq->hash));
-
 		if (!rq_mergeable(__rq)) {
 			cfq_del_crq_hash(crq);
 			continue;
@@ -287,36 +305,16 @@ cfq_choose_req(struct cfq_data *cfqd, struct cfq_rq *crq1, struct cfq_rq *crq2)
 		return crq2;
 	if (crq2 == NULL)
 		return crq1;
+	if (crq1->requeued)
+		return crq1;
+	if (crq2->requeued)
+		return crq2;
 
 	s1 = crq1->request->sector;
 	s2 = crq2->request->sector;
 
 	last = cfqd->last_sector;
 
-#if 0
-	if (!list_empty(&cfqd->queue->queue_head)) {
-		struct list_head *entry = &cfqd->queue->queue_head;
-		unsigned long distance = ~0UL;
-		struct request *rq;
-
-		while ((entry = entry->prev) != &cfqd->queue->queue_head) {
-			rq = list_entry_rq(entry);
-
-			if (blk_barrier_rq(rq))
-				break;
-
-			if (distance < abs(s1 - rq->sector + rq->nr_sectors)) {
-				distance = abs(s1 - rq->sector +rq->nr_sectors);
-				last = rq->sector + rq->nr_sectors;
-			}
-			if (distance < abs(s2 - rq->sector + rq->nr_sectors)) {
-				distance = abs(s2 - rq->sector +rq->nr_sectors);
-				last = rq->sector + rq->nr_sectors;
-			}
-		}
-	}
-#endif
-
 	/*
 	 * by definition, 1KiB is 2 sectors
 	 */
@@ -377,11 +375,13 @@ cfq_find_next_crq(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 	struct cfq_rq *crq_next = NULL, *crq_prev = NULL;
 	struct rb_node *rbnext, *rbprev;
 
-	if (!ON_RB(&last->rb_node))
-		return NULL;
-
-	if ((rbnext = rb_next(&last->rb_node)) == NULL)
+	if (ON_RB(&last->rb_node))
+		rbnext = rb_next(&last->rb_node);
+	else {
 		rbnext = rb_first(&cfqq->sort_list);
+		if (rbnext == &last->rb_node)
+			rbnext = NULL;
+	}
 
 	rbprev = rb_prev(&last->rb_node);
 
@@ -401,67 +401,53 @@ static void cfq_update_next_crq(struct cfq_rq *crq)
 		cfqq->next_crq = cfq_find_next_crq(cfqq->cfqd, cfqq, crq);
 }
 
-static int cfq_check_sort_rr_list(struct cfq_queue *cfqq)
+static void cfq_resort_rr_list(struct cfq_queue *cfqq, int preempted)
 {
-	struct list_head *head = &cfqq->cfqd->rr_list;
-	struct list_head *next, *prev;
-
-	/*
-	 * list might still be ordered
-	 */
-	next = cfqq->cfq_list.next;
-	if (next != head) {
-		struct cfq_queue *cnext = list_entry_cfqq(next);
+	struct cfq_data *cfqd = cfqq->cfqd;
+	struct list_head *list, *entry;
 
-		if (cfqq->service_used > cnext->service_used)
-			return 1;
-	}
+	BUG_ON(!cfqq->on_rr);
 
-	prev = cfqq->cfq_list.prev;
-	if (prev != head) {
-		struct cfq_queue *cprev = list_entry_cfqq(prev);
+	list_del(&cfqq->cfq_list);
 
-		if (cfqq->service_used < cprev->service_used)
-			return 1;
+	if (cfq_class_rt(cfqq))
+		list = &cfqd->cur_rr;
+	else if (cfq_class_idle(cfqq))
+		list = &cfqd->idle_rr;
+	else {
+		/*
+		 * if cfqq has requests in flight, don't allow it to be
+		 * found in cfq_set_active_queue before it has finished them.
+		 * this is done to increase fairness between a process that
+		 * has lots of io pending vs one that only generates one
+		 * sporadically or synchronously
+		 */
+		if (cfqq->in_flight)
+			list = &cfqd->busy_rr;
+		else
+			list = &cfqd->rr_list[cfqq->ioprio];
 	}
 
-	return 0;
-}
-
-static void cfq_sort_rr_list(struct cfq_queue *cfqq, int new_queue)
-{
-	struct list_head *entry = &cfqq->cfqd->rr_list;
-
-	if (!cfqq->on_rr)
-		return;
-	if (!new_queue && !cfq_check_sort_rr_list(cfqq))
+	/*
+	 * if queue was preempted, just add to front to be fair. busy_rr
+	 * isn't sorted.
+	 */
+	if (preempted || list == &cfqd->busy_rr) {
+		list_add(&cfqq->cfq_list, list);
 		return;
-
-	list_del(&cfqq->cfq_list);
+	}
 
 	/*
-	 * sort by our mean service_used, sub-sort by in-flight requests
+	 * sort by when queue was last serviced
 	 */
-	while ((entry = entry->prev) != &cfqq->cfqd->rr_list) {
+	entry = list;
+	while ((entry = entry->prev) != list) {
 		struct cfq_queue *__cfqq = list_entry_cfqq(entry);
 
-		if (cfqq->service_used > __cfqq->service_used)
+		if (!__cfqq->service_last)
+			break;
+		if (time_before(__cfqq->service_last, cfqq->service_last))
 			break;
-		else if (cfqq->service_used == __cfqq->service_used) {
-			struct list_head *prv;
-
-			while ((prv = entry->prev) != &cfqq->cfqd->rr_list) {
-				__cfqq = list_entry_cfqq(prv);
-
-				WARN_ON(__cfqq->service_used > cfqq->service_used);
-				if (cfqq->service_used != __cfqq->service_used)
-					break;
-				if (cfqq->in_flight > __cfqq->in_flight)
-					break;
-
-				entry = prv;
-			}
-		}
 	}
 
 	list_add(&cfqq->cfq_list, entry);
@@ -469,28 +455,24 @@ static void cfq_sort_rr_list(struct cfq_queue *cfqq, int new_queue)
 
 /*
  * add to busy list of queues for service, trying to be fair in ordering
- * the pending list according to requests serviced
+ * the pending list according to last request service
  */
 static inline void
-cfq_add_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+cfq_add_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq, int requeue)
 {
-	/*
-	 * it's currently on the empty list
-	 */
+	BUG_ON(cfqq->on_rr);
 	cfqq->on_rr = 1;
 	cfqd->busy_queues++;
 
-	if (time_after(jiffies, cfqq->service_start + cfq_service))
-		cfqq->service_used >>= 3;
-
-	cfq_sort_rr_list(cfqq, 1);
+	cfq_resort_rr_list(cfqq, requeue);
 }
 
 static inline void
 cfq_del_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq)
 {
-	list_move(&cfqq->cfq_list, &cfqd->empty_list);
+	BUG_ON(!cfqq->on_rr);
 	cfqq->on_rr = 0;
+	list_move(&cfqq->cfq_list, &cfqd->empty_list);
 
 	BUG_ON(!cfqd->busy_queues);
 	cfqd->busy_queues--;
@@ -505,16 +487,17 @@ static inline void cfq_del_crq_rb(struct cfq_rq *crq)
 
 	if (ON_RB(&crq->rb_node)) {
 		struct cfq_data *cfqd = cfqq->cfqd;
+		const int sync = crq->is_sync;
 
-		BUG_ON(!cfqq->queued[crq->is_sync]);
+		BUG_ON(!cfqq->queued[sync]);
+		cfqq->queued[sync]--;
 
 		cfq_update_next_crq(crq);
 
-		cfqq->queued[crq->is_sync]--;
 		rb_erase(&crq->rb_node, &cfqq->sort_list);
 		RB_CLEAR_COLOR(&crq->rb_node);
 
-		if (RB_EMPTY(&cfqq->sort_list) && cfqq->on_rr)
+		if (cfqq->on_rr && RB_EMPTY(&cfqq->sort_list))
 			cfq_del_cfqq_rr(cfqd, cfqq);
 	}
 }
@@ -562,7 +545,7 @@ static void cfq_add_crq_rb(struct cfq_rq *crq)
 	rb_insert_color(&crq->rb_node, &cfqq->sort_list);
 
 	if (!cfqq->on_rr)
-		cfq_add_cfqq_rr(cfqd, cfqq);
+		cfq_add_cfqq_rr(cfqd, cfqq, crq->requeued);
 
 	/*
 	 * check if this request is a better next-serve candidate
@@ -581,11 +564,10 @@ cfq_reposition_crq_rb(struct cfq_queue *cfqq, struct cfq_rq *crq)
 	cfq_add_crq_rb(crq);
 }
 
-static struct request *
-cfq_find_rq_rb(struct cfq_data *cfqd, sector_t sector)
+static struct request *cfq_find_rq_rb(struct cfq_data *cfqd, sector_t sector)
+
 {
-	const unsigned long key = cfq_hash_key(cfqd, current);
-	struct cfq_queue *cfqq = cfq_find_cfq_hash(cfqd, key);
+	struct cfq_queue *cfqq = cfq_find_cfq_hash(cfqd, current->pid);
 	struct rb_node *n;
 
 	if (!cfqq)
@@ -609,20 +591,23 @@ out:
 
 static void cfq_deactivate_request(request_queue_t *q, struct request *rq)
 {
+	struct cfq_data *cfqd = q->elevator->elevator_data;
 	struct cfq_rq *crq = RQ_DATA(rq);
 
 	if (crq) {
 		struct cfq_queue *cfqq = crq->cfq_queue;
 
-		if (cfqq->cfqd->cfq_tagged) {
-			cfqq->service_used--;
-			cfq_sort_rr_list(cfqq, 0);
-		}
-
 		if (crq->accounted) {
 			crq->accounted = 0;
-			cfqq->cfqd->rq_in_driver--;
+			WARN_ON(!cfqd->rq_in_driver);
+			cfqd->rq_in_driver--;
 		}
+		if (crq->in_flight) {
+			crq->in_flight = 0;
+			WARN_ON(!cfqq->in_flight);
+			cfqq->in_flight--;
+		}
+		crq->requeued = 1;
 	}
 }
 
@@ -640,11 +625,10 @@ static void cfq_remove_request(request_queue_t *q, struct request *rq)
 	struct cfq_rq *crq = RQ_DATA(rq);
 
 	if (crq) {
-		cfq_remove_merge_hints(q, crq);
 		list_del_init(&rq->queuelist);
+		cfq_del_crq_rb(crq);
+		cfq_remove_merge_hints(q, crq);
 
-		if (crq->cfq_queue)
-			cfq_del_crq_rb(crq);
 	}
 }
 
@@ -662,21 +646,15 @@ cfq_merge(request_queue_t *q, struct request **req, struct bio *bio)
 	}
 
 	__rq = cfq_find_rq_hash(cfqd, bio->bi_sector);
-	if (__rq) {
-		BUG_ON(__rq->sector + __rq->nr_sectors != bio->bi_sector);
-
-		if (elv_rq_merge_ok(__rq, bio)) {
-			ret = ELEVATOR_BACK_MERGE;
-			goto out;
-		}
+	if (__rq && elv_rq_merge_ok(__rq, bio)) {
+		ret = ELEVATOR_BACK_MERGE;
+		goto out;
 	}
 
 	__rq = cfq_find_rq_rb(cfqd, bio->bi_sector + bio_sectors(bio));
-	if (__rq) {
-		if (elv_rq_merge_ok(__rq, bio)) {
-			ret = ELEVATOR_FRONT_MERGE;
-			goto out;
-		}
+	if (__rq && elv_rq_merge_ok(__rq, bio)) {
+		ret = ELEVATOR_FRONT_MERGE;
+		goto out;
 	}
 
 	return ELEVATOR_NO_MERGE;
@@ -709,20 +687,194 @@ static void
 cfq_merged_requests(request_queue_t *q, struct request *rq,
 		    struct request *next)
 {
-	struct cfq_rq *crq = RQ_DATA(rq);
-	struct cfq_rq *cnext = RQ_DATA(next);
-
 	cfq_merged_request(q, rq);
 
-	if (!list_empty(&rq->queuelist) && !list_empty(&next->queuelist)) {
-		if (time_before(cnext->queue_start, crq->queue_start)) {
-			list_move(&rq->queuelist, &next->queuelist);
-			crq->queue_start = cnext->queue_start;
+	/*
+	 * reposition in fifo if next is older than rq
+	 */
+	if (!list_empty(&rq->queuelist) && !list_empty(&next->queuelist) &&
+	    time_before(next->start_time, rq->start_time))
+		list_move(&rq->queuelist, &next->queuelist);
+
+	cfq_remove_request(q, next);
+}
+
+static inline void
+__cfq_set_active_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+{
+	if (cfqq) {
+		/*
+		 * stop potential idle class queues waiting service
+		 */
+		del_timer(&cfqd->idle_class_timer);
+
+		cfqq->slice_start = jiffies;
+		cfqq->slice_end = 0;
+		cfqq->slice_left = 0;
+		cfqq->must_alloc_slice = 0;
+		cfqq->fifo_expire = 0;
+	}
+
+	cfqd->active_queue = cfqq;
+}
+
+/*
+ * 0
+ * 0,1
+ * 0,1,2
+ * 0,1,2,3
+ * 0,1,2,3,4
+ * 0,1,2,3,4,5
+ * 0,1,2,3,4,5,6
+ * 0,1,2,3,4,5,6,7
+ */
+static int cfq_get_next_prio_level(struct cfq_data *cfqd)
+{
+	int prio, wrap;
+
+	prio = -1;
+	wrap = 0;
+	do {
+		int p;
+
+		for (p = cfqd->cur_prio; p <= cfqd->cur_end_prio; p++) {
+			if (!list_empty(&cfqd->rr_list[p])) {
+				prio = p;
+				break;
+			}
+		}
+
+		if (prio != -1)
+			break;
+		cfqd->cur_prio = 0;
+		if (++cfqd->cur_end_prio == CFQ_PRIO_LISTS) {
+			cfqd->cur_end_prio = 0;
+			if (wrap)
+				break;
+			wrap = 1;
 		}
+	} while (1);
+
+	if (unlikely(prio == -1))
+		return -1;
+
+	BUG_ON(prio >= CFQ_PRIO_LISTS);
+
+	list_splice_init(&cfqd->rr_list[prio], &cfqd->cur_rr);
+
+	cfqd->cur_prio = prio + 1;
+	if (cfqd->cur_prio > cfqd->cur_end_prio) {
+		cfqd->cur_end_prio = cfqd->cur_prio;
+		cfqd->cur_prio = 0;
+	}
+	if (cfqd->cur_end_prio == CFQ_PRIO_LISTS) {
+		cfqd->cur_prio = 0;
+		cfqd->cur_end_prio = 0;
 	}
 
-	cfq_update_next_crq(cnext);
-	cfq_remove_request(q, next);
+	return prio;
+}
+
+static void cfq_set_active_queue(struct cfq_data *cfqd)
+{
+	struct cfq_queue *cfqq = NULL;
+
+	/*
+	 * if current list is non-empty, grab first entry. if it is empty,
+	 * get next prio level and grab first entry then if any are spliced
+	 */
+	if (!list_empty(&cfqd->cur_rr) || cfq_get_next_prio_level(cfqd) != -1)
+		cfqq = list_entry_cfqq(cfqd->cur_rr.next);
+
+	/*
+	 * if we have idle queues and no rt or be queues had pending
+	 * requests, either allow immediate service if the grace period
+	 * has passed or arm the idle grace timer
+	 */
+	if (!cfqq && !list_empty(&cfqd->idle_rr)) {
+		unsigned long end = cfqd->last_end_request + CFQ_IDLE_GRACE;
+
+		if (time_after_eq(jiffies, end))
+			cfqq = list_entry_cfqq(cfqd->idle_rr.next);
+		else
+			mod_timer(&cfqd->idle_class_timer, end);
+	}
+
+	__cfq_set_active_queue(cfqd, cfqq);
+}
+
+/*
+ * current cfqq expired its slice (or was too idle), select new one
+ */
+static inline void cfq_slice_expired(struct cfq_data *cfqd, int preempted)
+{
+	struct cfq_queue *cfqq = cfqd->active_queue;
+
+	if (cfqq) {
+		unsigned long now = jiffies;
+
+		if (cfqq->wait_request)
+			del_timer(&cfqd->idle_slice_timer);
+
+		if (!preempted && !cfqq->in_flight)
+			cfqq->service_last = now;
+
+		cfqq->must_dispatch = 0;
+		cfqq->wait_request = 0;
+
+		/*
+		 * store what was left of this slice, if the queue idled out
+		 * or was preempted
+		 */
+		if (time_after(now, cfqq->slice_end))
+			cfqq->slice_left = now - cfqq->slice_end;
+		else
+			cfqq->slice_left = 0;
+
+		if (cfqq->on_rr)
+			cfq_resort_rr_list(cfqq, preempted);
+
+		cfqd->active_queue = NULL;
+
+		if (cfqd->active_cic) {
+			put_io_context(cfqd->active_cic->ioc);
+			cfqd->active_cic = NULL;
+		}
+	}
+
+	cfqd->dispatch_slice = 0;
+}
+
+static int cfq_arm_slice_timer(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+
+{
+	WARN_ON(!RB_EMPTY(&cfqq->sort_list));
+	WARN_ON(cfqq != cfqd->active_queue);
+
+	/*
+	 * idle is disabled, either manually or by past process history
+	 */
+	if (!cfqd->cfq_slice_idle)
+		return 0;
+	if (!cfqq->idle_window)
+		return 0;
+	/*
+	 * task has exited, don't wait
+	 */
+	if (cfqd->active_cic && !cfqd->active_cic->ioc->task)
+		return 0;
+
+	cfqq->wait_request = 1;
+	cfqq->must_alloc = 1;
+
+	if (!timer_pending(&cfqd->idle_slice_timer)) {
+		unsigned long slice_left = cfqq->slice_end - 1;
+
+		cfqd->idle_slice_timer.expires = min(jiffies + cfqd->cfq_slice_idle, slice_left);
+		add_timer(&cfqd->idle_slice_timer);
+	}
+
+	return 1;
 }
 
 /*
@@ -738,31 +890,39 @@ static void cfq_dispatch_sort(request_queue_t *q, struct cfq_rq *crq)
 	struct request *__rq;
 	sector_t last;
 
-	cfq_del_crq_rb(crq);
-	cfq_remove_merge_hints(q, crq);
 	list_del(&crq->request->queuelist);
 
 	last = cfqd->last_sector;
-	while ((entry = entry->prev) != head) {
-		__rq = list_entry_rq(entry);
+	list_for_each_entry_reverse(__rq, head, queuelist) {
+		struct cfq_rq *__crq = RQ_DATA(__rq);
 
-		if (blk_barrier_rq(crq->request))
+		if (blk_barrier_rq(__rq))
+			break;
+		if (!blk_fs_request(__rq))
 			break;
-		if (!blk_fs_request(crq->request))
+		if (__crq->requeued)
 			break;
 
-		if (crq->request->sector > __rq->sector)
+		if (__rq->sector <= crq->request->sector)
 			break;
 		if (__rq->sector > last && crq->request->sector < last) {
-			last = crq->request->sector;
+			last = crq->request->sector + crq->request->nr_sectors;
 			break;
 		}
+		entry = &__rq->queuelist;
 	}
 
 	cfqd->last_sector = last;
+
+	cfqq->next_crq = cfq_find_next_crq(cfqd, cfqq, crq);
+
+	cfq_del_crq_rb(crq);
+	cfq_remove_merge_hints(q, crq);
+
 	crq->in_flight = 1;
+	crq->requeued = 0;
 	cfqq->in_flight++;
-	list_add(&crq->request->queuelist, entry);
+	list_add_tail(&crq->request->queuelist, entry);
 }
 
 /*
@@ -771,105 +931,176 @@ static void cfq_dispatch_sort(request_queue_t *q, struct cfq_rq *crq)
 static inline struct cfq_rq *cfq_check_fifo(struct cfq_queue *cfqq)
 {
 	struct cfq_data *cfqd = cfqq->cfqd;
-	const int reads = !list_empty(&cfqq->fifo[0]);
-	const int writes = !list_empty(&cfqq->fifo[1]);
-	unsigned long now = jiffies;
+	struct request *rq;
 	struct cfq_rq *crq;
 
-	if (time_before(now, cfqq->last_fifo_expire + cfqd->cfq_fifo_batch_expire))
+	if (cfqq->fifo_expire)
 		return NULL;
 
-	crq = RQ_DATA(list_entry(cfqq->fifo[0].next, struct request, queuelist));
-	if (reads && time_after(now, crq->queue_start + cfqd->cfq_fifo_expire_r)) {
-		cfqq->last_fifo_expire = now;
-		return crq;
-	}
+	if (!list_empty(&cfqq->fifo)) {
+		int fifo = cfq_cfqq_sync(cfqq);
 
-	crq = RQ_DATA(list_entry(cfqq->fifo[1].next, struct request, queuelist));
-	if (writes && time_after(now, crq->queue_start + cfqd->cfq_fifo_expire_w)) {
-		cfqq->last_fifo_expire = now;
-		return crq;
+		crq = RQ_DATA(list_entry_fifo(cfqq->fifo.next));
+		rq = crq->request;
+		if (time_after(jiffies, rq->start_time + cfqd->cfq_fifo_expire[fifo])) {
+			cfqq->fifo_expire = 1;
+			return crq;
+		}
 	}
 
 	return NULL;
 }
 
 /*
- * dispatch a single request from given queue
+ * Scale schedule slice based on io priority
  */
+static inline int
+cfq_prio_to_slice(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+{
+	const int base_slice = cfqd->cfq_slice[cfq_cfqq_sync(cfqq)];
+
+	WARN_ON(cfqq->ioprio >= IOPRIO_BE_NR);
+
+	return base_slice + (base_slice/CFQ_SLICE_SCALE * (4 - cfqq->ioprio));
+}
+
 static inline void
-cfq_dispatch_request(request_queue_t *q, struct cfq_data *cfqd,
-		     struct cfq_queue *cfqq)
+cfq_set_prio_slice(struct cfq_data *cfqd, struct cfq_queue *cfqq)
 {
-	struct cfq_rq *crq;
+	cfqq->slice_end = cfq_prio_to_slice(cfqd, cfqq) + jiffies;
+}
 
-	/*
-	 * follow expired path, else get first next available
-	 */
-	if ((crq = cfq_check_fifo(cfqq)) == NULL) {
-		if (cfqd->find_best_crq)
-			crq = cfqq->next_crq;
-		else
-			crq = rb_entry_crq(rb_first(&cfqq->sort_list));
-	}
+static inline int
+cfq_prio_to_maxrq(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+{
+	const int base_rq = cfqd->cfq_slice_async_rq;
 
-	cfqd->last_sector = crq->request->sector + crq->request->nr_sectors;
+	WARN_ON(cfqq->ioprio >= IOPRIO_BE_NR);
 
-	/*
-	 * finally, insert request into driver list
-	 */
-	cfq_dispatch_sort(q, crq);
+	return 2 * (base_rq + base_rq * (CFQ_PRIO_LISTS - 1 - cfqq->ioprio));
 }
 
-static int cfq_dispatch_requests(request_queue_t *q, int max_dispatch)
+/*
+ * get next queue for service
+ */
+static struct cfq_queue *cfq_select_queue(struct cfq_data *cfqd, int force)
 {
-	struct cfq_data *cfqd = q->elevator->elevator_data;
+	unsigned long now = jiffies;
 	struct cfq_queue *cfqq;
-	struct list_head *entry, *tmp;
-	int queued, busy_queues, first_round;
 
-	if (list_empty(&cfqd->rr_list))
-		return 0;
+	cfqq = cfqd->active_queue;
+	if (!cfqq)
+		goto new_queue;
 
-	queued = 0;
-	first_round = 1;
-restart:
-	busy_queues = 0;
-	list_for_each_safe(entry, tmp, &cfqd->rr_list) {
-		cfqq = list_entry_cfqq(entry);
+	/*
+	 * slice has expired
+	 */
+	if (!cfqq->must_dispatch && time_after(jiffies, cfqq->slice_end))
+		goto new_queue;
 
-		BUG_ON(RB_EMPTY(&cfqq->sort_list));
+	/*
+	 * if queue has requests, dispatch one. if not, check if
+	 * enough slice is left to wait for one
+	 */
+	if (!RB_EMPTY(&cfqq->sort_list))
+		goto keep_queue;
+	else if (!force && cfq_cfqq_sync(cfqq) &&
+		 time_before(now, cfqq->slice_end)) {
+		if (cfq_arm_slice_timer(cfqd, cfqq))
+			return NULL;
+	}
+
+new_queue:
+	cfq_slice_expired(cfqd, 0);
+	cfq_set_active_queue(cfqd);
+keep_queue:
+	return cfqd->active_queue;
+}
+
+static int
+__cfq_dispatch_requests(struct cfq_data *cfqd, struct cfq_queue *cfqq,
+			int max_dispatch)
+{
+	int dispatched = 0;
+
+	BUG_ON(RB_EMPTY(&cfqq->sort_list));
+
+	do {
+		struct cfq_rq *crq;
 
 		/*
-		 * first round of queueing, only select from queues that
-		 * don't already have io in-flight
+		 * follow expired path, else get first next available
 		 */
-		if (first_round && cfqq->in_flight)
-			continue;
+		if ((crq = cfq_check_fifo(cfqq)) == NULL)
+			crq = cfqq->next_crq;
+
+		/*
+		 * finally, insert request into driver dispatch list
+		 */
+		cfq_dispatch_sort(cfqd->queue, crq);
 
-		cfq_dispatch_request(q, cfqd, cfqq);
+		cfqd->dispatch_slice++;
+		dispatched++;
 
-		if (!RB_EMPTY(&cfqq->sort_list))
-			busy_queues++;
+		if (!cfqd->active_cic) {
+			atomic_inc(&crq->io_context->ioc->refcount);
+			cfqd->active_cic = crq->io_context;
+		}
 
-		queued++;
-	}
+		if (RB_EMPTY(&cfqq->sort_list))
+			break;
+
+	} while (dispatched < max_dispatch);
+
+	/*
+	 * if slice end isn't set yet, set it. if at least one request was
+	 * sync, use the sync time slice value
+	 */
+	if (!cfqq->slice_end)
+		cfq_set_prio_slice(cfqd, cfqq);
+
+	/*
+	 * expire an async queue immediately if it has used up its slice. idle
+	 * queue always expire after 1 dispatch round.
+	 */
+	if ((!cfq_cfqq_sync(cfqq) &&
+	    cfqd->dispatch_slice >= cfq_prio_to_maxrq(cfqd, cfqq)) ||
+	    cfq_class_idle(cfqq))
+		cfq_slice_expired(cfqd, 0);
+
+	return dispatched;
+}
+
+static int
+cfq_dispatch_requests(request_queue_t *q, int max_dispatch, int force)
+{
+	struct cfq_data *cfqd = q->elevator->elevator_data;
+	struct cfq_queue *cfqq;
+
+	if (!cfqd->busy_queues)
+		return 0;
+
+	cfqq = cfq_select_queue(cfqd, force);
+	if (cfqq) {
+		cfqq->wait_request = 0;
+		cfqq->must_dispatch = 0;
+		del_timer(&cfqd->idle_slice_timer);
+
+		if (cfq_class_idle(cfqq))
+			max_dispatch = 1;
 
-	if ((queued < max_dispatch) && (busy_queues || first_round)) {
-		first_round = 0;
-		goto restart;
+		return __cfq_dispatch_requests(cfqd, cfqq, max_dispatch);
 	}
 
-	return queued;
+	return 0;
 }
 
 static inline void cfq_account_dispatch(struct cfq_rq *crq)
 {
 	struct cfq_queue *cfqq = crq->cfq_queue;
 	struct cfq_data *cfqd = cfqq->cfqd;
-	unsigned long now, elapsed;
 
-	if (!blk_fs_request(crq->request))
+	if (unlikely(!blk_fs_request(crq->request)))
 		return;
 
 	/*
@@ -879,65 +1110,34 @@ static inline void cfq_account_dispatch(struct cfq_rq *crq)
 	if (crq->accounted)
 		return;
 
-	now = jiffies;
-	if (cfqq->service_start == ~0UL)
-		cfqq->service_start = now;
-
-	/*
-	 * on drives with tagged command queueing, command turn-around time
-	 * doesn't necessarily reflect the time spent processing this very
-	 * command inside the drive. so do the accounting differently there,
-	 * by just sorting on the number of requests
-	 */
-	if (cfqd->cfq_tagged) {
-		if (time_after(now, cfqq->service_start + cfq_service)) {
-			cfqq->service_start = now;
-			cfqq->service_used /= 10;
-		}
-
-		cfqq->service_used++;
-		cfq_sort_rr_list(cfqq, 0);
-	}
-
-	elapsed = now - crq->queue_start;
-	if (elapsed > max_elapsed_dispatch)
-		max_elapsed_dispatch = elapsed;
-
 	crq->accounted = 1;
-	crq->service_start = now;
-
-	if (++cfqd->rq_in_driver >= CFQ_MAX_TAG && !cfqd->cfq_tagged) {
-		cfqq->cfqd->cfq_tagged = 1;
-		printk("cfq: depth %d reached, tagging now on\n", CFQ_MAX_TAG);
-	}
+	cfqd->rq_in_driver++;
 }
 
 static inline void
 cfq_account_completion(struct cfq_queue *cfqq, struct cfq_rq *crq)
 {
 	struct cfq_data *cfqd = cfqq->cfqd;
+	unsigned long now;
 
 	if (!crq->accounted)
 		return;
 
+	now = jiffies;
+
 	WARN_ON(!cfqd->rq_in_driver);
 	cfqd->rq_in_driver--;
 
-	if (!cfqd->cfq_tagged) {
-		unsigned long now = jiffies;
-		unsigned long duration = now - crq->service_start;
+	if (!cfq_class_idle(cfqq))
+		cfqd->last_end_request = now;
 
-		if (time_after(now, cfqq->service_start + cfq_service)) {
-			cfqq->service_start = now;
-			cfqq->service_used >>= 3;
-		}
-
-		cfqq->service_used += duration;
-		cfq_sort_rr_list(cfqq, 0);
-
-		if (duration > max_elapsed_crq)
-			max_elapsed_crq = duration;
+	if (!cfqq->in_flight && cfqq->on_rr) {
+		cfqq->service_last = now;
+		cfq_resort_rr_list(cfqq, 0);
 	}
+
+	if (crq->is_sync)
+		crq->io_context->last_end_request = now;
 }
 
 static struct request *cfq_next_request(request_queue_t *q)
@@ -950,7 +1150,15 @@ static struct request *cfq_next_request(request_queue_t *q)
 dispatch:
 		rq = list_entry_rq(q->queue_head.next);
 
-		if ((crq = RQ_DATA(rq)) != NULL) {
+		crq = RQ_DATA(rq);
+		if (crq) {
+			/*
+			 * if idle window is disabled, allow queue buildup
+			 */
+			if (!crq->in_flight && !crq->cfq_queue->idle_window &&
+			    cfqd->rq_in_driver >= cfqd->cfq_max_depth)
+				return NULL;
+
 			cfq_remove_merge_hints(q, crq);
 			cfq_account_dispatch(crq);
 		}
@@ -958,7 +1166,7 @@ dispatch:
 		return rq;
 	}
 
-	if (cfq_dispatch_requests(q, cfqd->cfq_quantum))
+	if (cfq_dispatch_requests(q, cfqd->cfq_quantum, 0))
 		goto dispatch;
 
 	return NULL;
@@ -972,14 +1180,22 @@ dispatch:
  */
 static void cfq_put_queue(struct cfq_queue *cfqq)
 {
-	BUG_ON(!atomic_read(&cfqq->ref));
+	struct cfq_data *cfqd = cfqq->cfqd;
+
+	BUG_ON(atomic_read(&cfqq->ref) <= 0);
 
 	if (!atomic_dec_and_test(&cfqq->ref))
 		return;
 
 	BUG_ON(rb_first(&cfqq->sort_list));
+	BUG_ON(cfqq->allocated[READ] + cfqq->allocated[WRITE]);
 	BUG_ON(cfqq->on_rr);
 
+	if (unlikely(cfqd->active_queue == cfqq)) {
+		cfq_slice_expired(cfqd, 0);
+		kblockd_schedule_work(&cfqd->unplug_work);
+	}
+
 	cfq_put_cfqd(cfqq->cfqd);
 
 	/*
@@ -991,7 +1207,7 @@ static void cfq_put_queue(struct cfq_queue *cfqq)
 }
 
 static inline struct cfq_queue *
-__cfq_find_cfq_hash(struct cfq_data *cfqd, unsigned long key, const int hashval)
+__cfq_find_cfq_hash(struct cfq_data *cfqd, unsigned int key, const int hashval)
 {
 	struct hlist_head *hash_list = &cfqd->cfq_hash[hashval];
 	struct hlist_node *entry, *next;
@@ -1007,94 +1223,220 @@ __cfq_find_cfq_hash(struct cfq_data *cfqd, unsigned long key, const int hashval)
 }
 
 static struct cfq_queue *
-cfq_find_cfq_hash(struct cfq_data *cfqd, unsigned long key)
+cfq_find_cfq_hash(struct cfq_data *cfqd, unsigned int key)
 {
 	return __cfq_find_cfq_hash(cfqd, key, hash_long(key, CFQ_QHASH_SHIFT));
 }
 
-static inline void
-cfq_rehash_cfqq(struct cfq_data *cfqd, struct cfq_queue **cfqq,
-		struct cfq_io_context *cic)
+static void cfq_free_io_context(struct cfq_io_context *cic)
 {
-	unsigned long hashkey = cfq_hash_key(cfqd, current);
-	unsigned long hashval = hash_long(hashkey, CFQ_QHASH_SHIFT);
-	struct cfq_queue *__cfqq;
-	unsigned long flags;
+	struct cfq_io_context *__cic;
+	struct list_head *entry, *next;
 
-	spin_lock_irqsave(cfqd->queue->queue_lock, flags);
-
-	hlist_del(&(*cfqq)->cfq_hash);
-
-	__cfqq = __cfq_find_cfq_hash(cfqd, hashkey, hashval);
-	if (!__cfqq || __cfqq == *cfqq) {
-		__cfqq = *cfqq;
-		hlist_add_head(&__cfqq->cfq_hash, &cfqd->cfq_hash[hashval]);
-		__cfqq->key_type = cfqd->key_type;
-	} else {
-		atomic_inc(&__cfqq->ref);
-		cic->cfqq = __cfqq;
-		cfq_put_queue(*cfqq);
-		*cfqq = __cfqq;
+	list_for_each_safe(entry, next, &cic->list) {
+		__cic = list_entry(entry, struct cfq_io_context, list);
+		kmem_cache_free(cfq_ioc_pool, __cic);
 	}
 
-	cic->cfqq = __cfqq;
-	spin_unlock_irqrestore(cfqd->queue->queue_lock, flags);
+	kmem_cache_free(cfq_ioc_pool, cic);
 }
 
-static void cfq_free_io_context(struct cfq_io_context *cic)
+/*
+ * Called with interrupts disabled
+ */
+static void cfq_exit_single_io_context(struct cfq_io_context *cic)
 {
-	kmem_cache_free(cfq_ioc_pool, cic);
+	struct cfq_data *cfqd = cic->cfqq->cfqd;
+	request_queue_t *q = cfqd->queue;
+
+	WARN_ON(!irqs_disabled());
+
+	spin_lock(q->queue_lock);
+
+	if (unlikely(cic->cfqq == cfqd->active_queue)) {
+		cfq_slice_expired(cfqd, 0);
+		kblockd_schedule_work(&cfqd->unplug_work);
+	}
+
+	cfq_put_queue(cic->cfqq);
+	cic->cfqq = NULL;
+	spin_unlock(q->queue_lock);
 }
 
 /*
- * locking hierarchy is: io_context lock -> queue locks
+ * Another task may update the task cic list, if it is doing a queue lookup
+ * on its behalf. cfq_cic_lock excludes such concurrent updates
  */
 static void cfq_exit_io_context(struct cfq_io_context *cic)
 {
-	struct cfq_queue *cfqq = cic->cfqq;
-	struct list_head *entry = &cic->list;
-	request_queue_t *q;
+	struct cfq_io_context *__cic;
+	struct list_head *entry;
 	unsigned long flags;
 
+	local_irq_save(flags);
+
 	/*
 	 * put the reference this task is holding to the various queues
 	 */
-	spin_lock_irqsave(&cic->ioc->lock, flags);
-	while ((entry = cic->list.next) != &cic->list) {
-		struct cfq_io_context *__cic;
-
+	list_for_each(entry, &cic->list) {
 		__cic = list_entry(entry, struct cfq_io_context, list);
-		list_del(entry);
-
-		q = __cic->cfqq->cfqd->queue;
-		spin_lock(q->queue_lock);
-		cfq_put_queue(__cic->cfqq);
-		spin_unlock(q->queue_lock);
+		cfq_exit_single_io_context(__cic);
 	}
 
-	q = cfqq->cfqd->queue;
-	spin_lock(q->queue_lock);
-	cfq_put_queue(cfqq);
-	spin_unlock(q->queue_lock);
-
-	cic->cfqq = NULL;
-	spin_unlock_irqrestore(&cic->ioc->lock, flags);
+	cfq_exit_single_io_context(cic);
+	local_irq_restore(flags);
 }
 
-static struct cfq_io_context *cfq_alloc_io_context(int gfp_flags)
+static struct cfq_io_context *
+cfq_alloc_io_context(struct cfq_data *cfqd, int gfp_mask)
 {
-	struct cfq_io_context *cic = kmem_cache_alloc(cfq_ioc_pool, gfp_flags);
+	struct cfq_io_context *cic = kmem_cache_alloc(cfq_ioc_pool, gfp_mask);
 
 	if (cic) {
-		cic->dtor = cfq_free_io_context;
-		cic->exit = cfq_exit_io_context;
 		INIT_LIST_HEAD(&cic->list);
 		cic->cfqq = NULL;
+		cic->key = NULL;
+		cic->last_end_request = jiffies;
+		cic->ttime_total = 0;
+		cic->ttime_samples = 0;
+		cic->ttime_mean = 0;
+		cic->dtor = cfq_free_io_context;
+		cic->exit = cfq_exit_io_context;
 	}
 
 	return cic;
 }
 
+static void cfq_init_prio_data(struct cfq_queue *cfqq)
+{
+	struct task_struct *tsk = current;
+	int ioprio_class;
+
+	if (!cfqq->prio_changed)
+		return;
+
+	ioprio_class = IOPRIO_PRIO_CLASS(tsk->ioprio);
+	switch (ioprio_class) {
+		default:
+			printk(KERN_ERR "cfq: bad prio %x\n", ioprio_class);
+		case IOPRIO_CLASS_NONE:
+			/*
+			 * no prio set, place us in the middle of the BE classes
+			 */
+			cfqq->ioprio = task_nice_ioprio(tsk);
+			cfqq->ioprio_class = IOPRIO_CLASS_BE;
+			break;
+		case IOPRIO_CLASS_RT:
+			cfqq->ioprio = task_ioprio(tsk);
+			cfqq->ioprio_class = IOPRIO_CLASS_RT;
+			break;
+		case IOPRIO_CLASS_BE:
+			cfqq->ioprio = task_ioprio(tsk);
+			cfqq->ioprio_class = IOPRIO_CLASS_BE;
+			break;
+		case IOPRIO_CLASS_IDLE:
+			cfqq->ioprio_class = IOPRIO_CLASS_IDLE;
+			cfqq->ioprio = 7;
+			cfqq->idle_window = 0;
+			break;
+	}
+
+	/*
+	 * keep track of original prio settings in case we have to temporarily
+	 * elevate the priority of this queue
+	 */
+	cfqq->org_ioprio = cfqq->ioprio;
+	cfqq->org_ioprio_class = cfqq->ioprio_class;
+
+	if (cfqq->on_rr)
+		cfq_resort_rr_list(cfqq, 0);
+
+	cfqq->prio_changed = 0;
+}
+
+static inline void changed_ioprio(struct cfq_queue *cfqq)
+{
+	if (cfqq) {
+		struct cfq_data *cfqd = cfqq->cfqd;
+
+		spin_lock(cfqd->queue->queue_lock);
+		cfqq->prio_changed = 1;
+		cfq_init_prio_data(cfqq);
+		spin_unlock(cfqd->queue->queue_lock);
+	}
+}
+
+/*
+ * callback from sys_ioprio_set, irqs are disabled
+ */
+static int cfq_ioc_set_ioprio(struct io_context *ioc, unsigned int ioprio)
+{
+	struct cfq_io_context *cic = ioc->cic;
+
+	changed_ioprio(cic->cfqq);
+
+	list_for_each_entry(cic, &cic->list, list)
+		changed_ioprio(cic->cfqq);
+
+	return 0;
+}
+
+static struct cfq_queue *
+cfq_get_queue(struct cfq_data *cfqd, unsigned int key, int gfp_mask)
+{
+	const int hashval = hash_long(key, CFQ_QHASH_SHIFT);
+	struct cfq_queue *cfqq, *new_cfqq = NULL;
+
+retry:
+	cfqq = __cfq_find_cfq_hash(cfqd, key, hashval);
+
+	if (!cfqq) {
+		if (new_cfqq) {
+			cfqq = new_cfqq;
+			new_cfqq = NULL;
+		} else if (gfp_mask & __GFP_WAIT) {
+			spin_unlock_irq(cfqd->queue->queue_lock);
+			new_cfqq = kmem_cache_alloc(cfq_pool, gfp_mask);
+			spin_lock_irq(cfqd->queue->queue_lock);
+			goto retry;
+		} else {
+			cfqq = kmem_cache_alloc(cfq_pool, gfp_mask);
+			if (!cfqq)
+				goto out;
+		}
+
+		memset(cfqq, 0, sizeof(*cfqq));
+
+		INIT_HLIST_NODE(&cfqq->cfq_hash);
+		INIT_LIST_HEAD(&cfqq->cfq_list);
+		RB_CLEAR_ROOT(&cfqq->sort_list);
+		INIT_LIST_HEAD(&cfqq->fifo);
+
+		cfqq->key = key;
+		hlist_add_head(&cfqq->cfq_hash, &cfqd->cfq_hash[hashval]);
+		atomic_set(&cfqq->ref, 0);
+		cfqq->cfqd = cfqd;
+		atomic_inc(&cfqd->ref);
+		cfqq->service_last = 0;
+		/*
+		 * set ->slice_left to allow preemption for a new process
+		 */
+		cfqq->slice_left = 2 * cfqd->cfq_slice_idle;
+		cfqq->idle_window = 1;
+		cfqq->ioprio = -1;
+		cfqq->ioprio_class = -1;
+		cfqq->prio_changed = 1;
+	}
+
+	if (new_cfqq)
+		kmem_cache_free(cfq_pool, new_cfqq);
+
+	atomic_inc(&cfqq->ref);
+out:
+	WARN_ON((gfp_mask & __GFP_WAIT) && !cfqq);
+	return cfqq;
+}
+
 /*
  * Setup general io context and cfq io context. There can be several cfq
  * io contexts per general io context, if this process is doing io to more
@@ -1102,39 +1444,39 @@ static struct cfq_io_context *cfq_alloc_io_context(int gfp_flags)
  * cfqq, so we don't need to worry about it disappearing
  */
 static struct cfq_io_context *
-cfq_get_io_context(struct cfq_queue **cfqq, int gfp_flags)
+cfq_get_io_context(struct cfq_data *cfqd, pid_t pid, int gfp_mask)
 {
-	struct cfq_data *cfqd = (*cfqq)->cfqd;
-	struct cfq_queue *__cfqq = *cfqq;
+	struct io_context *ioc = NULL;
 	struct cfq_io_context *cic;
-	struct io_context *ioc;
 
-	might_sleep_if(gfp_flags & __GFP_WAIT);
+	might_sleep_if(gfp_mask & __GFP_WAIT);
 
-	ioc = get_io_context(gfp_flags);
+	ioc = get_io_context(gfp_mask);
 	if (!ioc)
 		return NULL;
 
 	if ((cic = ioc->cic) == NULL) {
-		cic = cfq_alloc_io_context(gfp_flags);
+		cic = cfq_alloc_io_context(cfqd, gfp_mask);
 
 		if (cic == NULL)
 			goto err;
 
+		/*
+		 * manually increment generic io_context usage count, it
+		 * cannot go away since we are already holding one ref to it
+		 */
 		ioc->cic = cic;
+		ioc->set_ioprio = cfq_ioc_set_ioprio;
 		cic->ioc = ioc;
-		cic->cfqq = __cfqq;
-		atomic_inc(&__cfqq->ref);
+		cic->key = cfqd;
+		atomic_inc(&cfqd->ref);
 	} else {
 		struct cfq_io_context *__cic;
-		unsigned long flags;
 
 		/*
-		 * since the first cic on the list is actually the head
-		 * itself, need to check this here or we'll duplicate an
-		 * cic per ioc for no reason
+		 * the first cic on the list is actually the head itself
 		 */
-		if (cic->cfqq == __cfqq)
+		if (cic->key == cfqd)
 			goto out;
 
 		/*
@@ -1142,152 +1484,259 @@ cfq_get_io_context(struct cfq_queue **cfqq, int gfp_flags)
 		 * should be ok here, the list will usually not be more than
 		 * 1 or a few entries long
 		 */
-		spin_lock_irqsave(&ioc->lock, flags);
 		list_for_each_entry(__cic, &cic->list, list) {
 			/*
 			 * this process is already holding a reference to
 			 * this queue, so no need to get one more
 			 */
-			if (__cic->cfqq == __cfqq) {
+			if (__cic->key == cfqd) {
 				cic = __cic;
-				spin_unlock_irqrestore(&ioc->lock, flags);
 				goto out;
 			}
 		}
-		spin_unlock_irqrestore(&ioc->lock, flags);
 
 		/*
 		 * nope, process doesn't have a cic assoicated with this
 		 * cfqq yet. get a new one and add to list
 		 */
-		__cic = cfq_alloc_io_context(gfp_flags);
+		__cic = cfq_alloc_io_context(cfqd, gfp_mask);
 		if (__cic == NULL)
 			goto err;
 
 		__cic->ioc = ioc;
-		__cic->cfqq = __cfqq;
-		atomic_inc(&__cfqq->ref);
-		spin_lock_irqsave(&ioc->lock, flags);
+		__cic->key = cfqd;
+		atomic_inc(&cfqd->ref);
 		list_add(&__cic->list, &cic->list);
-		spin_unlock_irqrestore(&ioc->lock, flags);
-
 		cic = __cic;
-		*cfqq = __cfqq;
 	}
 
 out:
-	/*
-	 * if key_type has been changed on the fly, we lazily rehash
-	 * each queue at lookup time
-	 */
-	if ((*cfqq)->key_type != cfqd->key_type)
-		cfq_rehash_cfqq(cfqd, cfqq, cic);
-
 	return cic;
 err:
 	put_io_context(ioc);
 	return NULL;
 }
 
-static struct cfq_queue *
-__cfq_get_queue(struct cfq_data *cfqd, unsigned long key, int gfp_mask)
+static void
+cfq_update_io_thinktime(struct cfq_data *cfqd, struct cfq_io_context *cic)
 {
-	const int hashval = hash_long(key, CFQ_QHASH_SHIFT);
-	struct cfq_queue *cfqq, *new_cfqq = NULL;
+	unsigned long elapsed, ttime;
 
-retry:
-	cfqq = __cfq_find_cfq_hash(cfqd, key, hashval);
+	/*
+	 * if this context already has stuff queued, thinktime is from
+	 * last queue not last end
+	 */
+#if 0
+	if (time_after(cic->last_end_request, cic->last_queue))
+		elapsed = jiffies - cic->last_end_request;
+	else
+		elapsed = jiffies - cic->last_queue;
+#else
+		elapsed = jiffies - cic->last_end_request;
+#endif
 
-	if (!cfqq) {
-		if (new_cfqq) {
-			cfqq = new_cfqq;
-			new_cfqq = NULL;
-		} else {
-			spin_unlock_irq(cfqd->queue->queue_lock);
-			new_cfqq = kmem_cache_alloc(cfq_pool, gfp_mask);
-			spin_lock_irq(cfqd->queue->queue_lock);
+	ttime = min(elapsed, 2UL * cfqd->cfq_slice_idle);
 
-			if (!new_cfqq && !(gfp_mask & __GFP_WAIT))
-				goto out;
-
-			goto retry;
-		}
+	cic->ttime_samples = (7*cic->ttime_samples + 256) / 8;
+	cic->ttime_total = (7*cic->ttime_total + 256*ttime) / 8;
+	cic->ttime_mean = (cic->ttime_total + 128) / cic->ttime_samples;
+}
 
-		memset(cfqq, 0, sizeof(*cfqq));
+#define sample_valid(samples)	((samples) > 80)
 
-		INIT_HLIST_NODE(&cfqq->cfq_hash);
-		INIT_LIST_HEAD(&cfqq->cfq_list);
-		RB_CLEAR_ROOT(&cfqq->sort_list);
-		INIT_LIST_HEAD(&cfqq->fifo[0]);
-		INIT_LIST_HEAD(&cfqq->fifo[1]);
+/*
+ * Disable idle window if the process thinks too long or seeks so much that
+ * it doesn't matter
+ */
+static void
+cfq_update_idle_window(struct cfq_data *cfqd, struct cfq_queue *cfqq,
+		       struct cfq_io_context *cic)
+{
+	int enable_idle = cfqq->idle_window;
 
-		cfqq->key = key;
-		hlist_add_head(&cfqq->cfq_hash, &cfqd->cfq_hash[hashval]);
-		atomic_set(&cfqq->ref, 0);
-		cfqq->cfqd = cfqd;
-		atomic_inc(&cfqd->ref);
-		cfqq->key_type = cfqd->key_type;
-		cfqq->service_start = ~0UL;
+	if (!cic->ioc->task || !cfqd->cfq_slice_idle)
+		enable_idle = 0;
+	else if (sample_valid(cic->ttime_samples)) {
+		if (cic->ttime_mean > cfqd->cfq_slice_idle)
+			enable_idle = 0;
+		else
+			enable_idle = 1;
 	}
 
-	if (new_cfqq)
-		kmem_cache_free(cfq_pool, new_cfqq);
+	cfqq->idle_window = enable_idle;
+}
 
-	atomic_inc(&cfqq->ref);
-out:
-	WARN_ON((gfp_mask & __GFP_WAIT) && !cfqq);
-	return cfqq;
+
+/*
+ * Check if new_cfqq should preempt the currently active queue. Return 0 for
+ * no or if we aren't sure, a 1 will cause a preempt.
+ */
+static int
+cfq_should_preempt(struct cfq_data *cfqd, struct cfq_queue *new_cfqq,
+		   struct cfq_rq *crq)
+{
+	struct cfq_queue *cfqq = cfqd->active_queue;
+
+	if (cfq_class_idle(new_cfqq))
+		return 0;
+
+	if (!cfqq)
+		return 1;
+
+	if (cfq_class_idle(cfqq))
+		return 1;
+	if (!new_cfqq->wait_request)
+		return 0;
+	/*
+	 * if it doesn't have slice left, forget it
+	 */
+	if (new_cfqq->slice_left < cfqd->cfq_slice_idle)
+		return 0;
+	if (crq->is_sync && !cfq_cfqq_sync(cfqq))
+		return 1;
+
+	return 0;
+}
+
+/*
+ * cfqq preempts the active queue. if we allowed preempt with no slice left,
+ * let it have half of its nominal slice.
+ */
+static void cfq_preempt_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+{
+	struct cfq_queue *__cfqq, *next;
+
+	list_for_each_entry_safe(__cfqq, next, &cfqd->cur_rr, cfq_list)
+		cfq_resort_rr_list(__cfqq, 1);
+
+	if (!cfqq->slice_left)
+		cfqq->slice_left = cfq_prio_to_slice(cfqd, cfqq) / 2;
+
+	cfqq->slice_end = cfqq->slice_left + jiffies;
+	cfq_slice_expired(cfqd, 1);
+	__cfq_set_active_queue(cfqd, cfqq);
+}
+
+/*
+ * should really be a ll_rw_blk.c helper
+ */
+static void cfq_start_queueing(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+{
+	request_queue_t *q = cfqd->queue;
+
+	if (!blk_queue_plugged(q))
+		q->request_fn(q);
+	else
+		__generic_unplug_device(q);
+}
+
+/*
+ * Called when a new fs request (crq) is added (to cfqq). Check if there's
+ * something we should do about it
+ */
+static void
+cfq_crq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq,
+		 struct cfq_rq *crq)
+{
+	const int sync = crq->is_sync;
+
+	cfqq->next_crq = cfq_choose_req(cfqd, cfqq->next_crq, crq);
+
+	if (sync) {
+		struct cfq_io_context *cic = crq->io_context;
+
+		cfq_update_io_thinktime(cfqd, cic);
+		cfq_update_idle_window(cfqd, cfqq, cic);
+
+		cic->last_queue = jiffies;
+	}
+
+	if (cfqq == cfqd->active_queue) {
+		/*
+		 * if we are waiting for a request for this queue, let it rip
+		 * immediately and flag that we must not expire this queue
+		 * just now
+		 */
+		if (cfqq->wait_request) {
+			cfqq->must_dispatch = 1;
+			del_timer(&cfqd->idle_slice_timer);
+			cfq_start_queueing(cfqd, cfqq);
+		}
+	} else if (cfq_should_preempt(cfqd, cfqq, crq)) {
+		/*
+		 * not the active queue - expire current slice if it is
+		 * idle and has expired it's mean thinktime or this new queue
+		 * has some old slice time left and is of higher priority
+		 */
+		cfq_preempt_queue(cfqd, cfqq);
+		cfqq->must_dispatch = 1;
+		cfq_start_queueing(cfqd, cfqq);
+	}
 }
 
-static void cfq_enqueue(struct cfq_data *cfqd, struct cfq_rq *crq)
+static void cfq_enqueue(struct cfq_data *cfqd, struct request *rq)
 {
-	crq->is_sync = 0;
-	if (rq_data_dir(crq->request) == READ || current->flags & PF_SYNCWRITE)
-		crq->is_sync = 1;
+	struct cfq_rq *crq = RQ_DATA(rq);
+	struct cfq_queue *cfqq = crq->cfq_queue;
+
+	cfq_init_prio_data(cfqq);
 
 	cfq_add_crq_rb(crq);
-	crq->queue_start = jiffies;
 
-	list_add_tail(&crq->request->queuelist, &crq->cfq_queue->fifo[crq->is_sync]);
+	list_add_tail(&rq->queuelist, &cfqq->fifo);
+
+	if (rq_mergeable(rq)) {
+		cfq_add_crq_hash(cfqd, crq);
+
+		if (!cfqd->queue->last_merge)
+			cfqd->queue->last_merge = rq;
+	}
+
+	cfq_crq_enqueued(cfqd, cfqq, crq);
 }
 
 static void
 cfq_insert_request(request_queue_t *q, struct request *rq, int where)
 {
 	struct cfq_data *cfqd = q->elevator->elevator_data;
-	struct cfq_rq *crq = RQ_DATA(rq);
 
 	switch (where) {
 		case ELEVATOR_INSERT_BACK:
-			while (cfq_dispatch_requests(q, cfqd->cfq_quantum))
+			while (cfq_dispatch_requests(q, INT_MAX, 1))
 				;
 			list_add_tail(&rq->queuelist, &q->queue_head);
+			/*
+			 * If we were idling with pending requests on
+			 * inactive cfqqs, force dispatching will
+			 * remove the idle timer and the queue won't
+			 * be kicked by __make_request() afterward.
+			 * Kick it here.
+			 */
+			kblockd_schedule_work(&cfqd->unplug_work);
 			break;
 		case ELEVATOR_INSERT_FRONT:
 			list_add(&rq->queuelist, &q->queue_head);
 			break;
 		case ELEVATOR_INSERT_SORT:
 			BUG_ON(!blk_fs_request(rq));
-			cfq_enqueue(cfqd, crq);
+			cfq_enqueue(cfqd, rq);
 			break;
 		default:
 			printk("%s: bad insert point %d\n", __FUNCTION__,where);
 			return;
 	}
+}
 
-	if (rq_mergeable(rq)) {
-		cfq_add_crq_hash(cfqd, crq);
-
-		if (!q->last_merge)
-			q->last_merge = rq;
-	}
+static inline int cfq_pending_requests(struct cfq_data *cfqd)
+{
+	return !list_empty(&cfqd->queue->queue_head) || cfqd->busy_queues;
 }
 
 static int cfq_queue_empty(request_queue_t *q)
 {
 	struct cfq_data *cfqd = q->elevator->elevator_data;
 
-	return list_empty(&q->queue_head) && list_empty(&cfqd->rr_list);
+	return !cfq_pending_requests(cfqd);
 }
 
 static void cfq_completed_request(request_queue_t *q, struct request *rq)
@@ -1332,51 +1781,132 @@ cfq_latter_request(request_queue_t *q, struct request *rq)
 	return NULL;
 }
 
-static int cfq_may_queue(request_queue_t *q, int rw)
+/*
+ * we temporarily boost lower priority queues if they are holding fs exclusive
+ * resources. they are boosted to normal prio (CLASS_BE/4)
+ */
+static void cfq_prio_boost(struct cfq_queue *cfqq)
 {
-	struct cfq_data *cfqd = q->elevator->elevator_data;
-	struct cfq_queue *cfqq;
-	int ret = ELV_MQUEUE_MAY;
+	const int ioprio_class = cfqq->ioprio_class;
+	const int ioprio = cfqq->ioprio;
 
-	if (current->flags & PF_MEMALLOC)
-		return ELV_MQUEUE_MAY;
+	if (has_fs_excl()) {
+		/*
+		 * boost idle prio on transactions that would lock out other
+		 * users of the filesystem
+		 */
+		if (cfq_class_idle(cfqq))
+			cfqq->ioprio_class = IOPRIO_CLASS_BE;
+		if (cfqq->ioprio > IOPRIO_NORM)
+			cfqq->ioprio = IOPRIO_NORM;
+	} else {
+		/*
+		 * check if we need to unboost the queue
+		 */
+		if (cfqq->ioprio_class != cfqq->org_ioprio_class)
+			cfqq->ioprio_class = cfqq->org_ioprio_class;
+		if (cfqq->ioprio != cfqq->org_ioprio)
+			cfqq->ioprio = cfqq->org_ioprio;
+	}
 
-	cfqq = cfq_find_cfq_hash(cfqd, cfq_hash_key(cfqd, current));
-	if (cfqq) {
-		int limit = cfqd->max_queued;
+	/*
+	 * refile between round-robin lists if we moved the priority class
+	 */
+	if ((ioprio_class != cfqq->ioprio_class || ioprio != cfqq->ioprio) &&
+	    cfqq->on_rr)
+		cfq_resort_rr_list(cfqq, 0);
+}
 
-		if (cfqq->allocated[rw] < cfqd->cfq_queued)
-			return ELV_MQUEUE_MUST;
+static inline pid_t cfq_queue_pid(struct task_struct *task, int rw)
+{
+	if (rw == READ || process_sync(task))
+		return task->pid;
 
-		if (cfqd->busy_queues)
-			limit = q->nr_requests / cfqd->busy_queues;
+	return CFQ_KEY_ASYNC;
+}
 
-		if (limit < cfqd->cfq_queued)
-			limit = cfqd->cfq_queued;
-		else if (limit > cfqd->max_queued)
-			limit = cfqd->max_queued;
+static inline int
+__cfq_may_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq,
+		struct task_struct *task, int rw)
+{
+	if (cfqq->wait_request && cfqq->must_alloc)
+		return ELV_MQUEUE_MUST;
 
-		if (cfqq->allocated[rw] >= limit) {
-			if (limit > cfqq->alloc_limit[rw])
-				cfqq->alloc_limit[rw] = limit;
+	return ELV_MQUEUE_MAY;
+#if 0
+	if (!cfqq || task->flags & PF_MEMALLOC)
+		return ELV_MQUEUE_MAY;
+	if (!cfqq->allocated[rw] || cfqq->must_alloc) {
+		if (cfqq->wait_request)
+			return ELV_MQUEUE_MUST;
 
-			ret = ELV_MQUEUE_NO;
+		/*
+		 * only allow 1 ELV_MQUEUE_MUST per slice, otherwise we
+		 * can quickly flood the queue with writes from a single task
+		 */
+		if (rw == READ || !cfqq->must_alloc_slice) {
+			cfqq->must_alloc_slice = 1;
+			return ELV_MQUEUE_MUST;
 		}
+
+		return ELV_MQUEUE_MAY;
 	}
+	if (cfq_class_idle(cfqq))
+		return ELV_MQUEUE_NO;
+	if (cfqq->allocated[rw] >= cfqd->max_queued) {
+		struct io_context *ioc = get_io_context(GFP_ATOMIC);
+		int ret = ELV_MQUEUE_NO;
 
-	return ret;
+		if (ioc && ioc->nr_batch_requests)
+			ret = ELV_MQUEUE_MAY;
+
+		put_io_context(ioc);
+		return ret;
+	}
+
+	return ELV_MQUEUE_MAY;
+#endif
+}
+
+static int cfq_may_queue(request_queue_t *q, int rw, struct bio *bio)
+{
+	struct cfq_data *cfqd = q->elevator->elevator_data;
+	struct task_struct *tsk = current;
+	struct cfq_queue *cfqq;
+
+	/*
+	 * don't force setup of a queue from here, as a call to may_queue
+	 * does not necessarily imply that a request actually will be queued.
+	 * so just lookup a possibly existing queue, or return 'may queue'
+	 * if that fails
+	 */
+	cfqq = cfq_find_cfq_hash(cfqd, cfq_queue_pid(tsk, rw));
+	if (cfqq) {
+		cfq_init_prio_data(cfqq);
+		cfq_prio_boost(cfqq);
+
+		return __cfq_may_queue(cfqd, cfqq, tsk, rw);
+	}
+
+	return ELV_MQUEUE_MAY;
 }
 
 static void cfq_check_waiters(request_queue_t *q, struct cfq_queue *cfqq)
 {
+	struct cfq_data *cfqd = q->elevator->elevator_data;
 	struct request_list *rl = &q->rq;
-	const int write = waitqueue_active(&rl->wait[WRITE]);
-	const int read = waitqueue_active(&rl->wait[READ]);
 
-	if (read && cfqq->allocated[READ] < cfqq->alloc_limit[READ])
-		wake_up(&rl->wait[READ]);
-	if (write && cfqq->allocated[WRITE] < cfqq->alloc_limit[WRITE])
-		wake_up(&rl->wait[WRITE]);
+	if (cfqq->allocated[READ] <= cfqd->max_queued || cfqd->rq_starved) {
+		smp_mb();
+		if (waitqueue_active(&rl->wait[READ]))
+			wake_up(&rl->wait[READ]);
+	}
+
+	if (cfqq->allocated[WRITE] <= cfqd->max_queued || cfqd->rq_starved) {
+		smp_mb();
+		if (waitqueue_active(&rl->wait[WRITE]))
+			wake_up(&rl->wait[WRITE]);
+	}
 }
 
 /*
@@ -1389,69 +1919,59 @@ static void cfq_put_request(request_queue_t *q, struct request *rq)
 
 	if (crq) {
 		struct cfq_queue *cfqq = crq->cfq_queue;
+		const int rw = rq_data_dir(rq);
 
-		BUG_ON(q->last_merge == rq);
-		BUG_ON(!hlist_unhashed(&crq->hash));
-
-		if (crq->io_context)
-			put_io_context(crq->io_context->ioc);
+		BUG_ON(!cfqq->allocated[rw]);
+		cfqq->allocated[rw]--;
 
-		BUG_ON(!cfqq->allocated[crq->is_write]);
-		cfqq->allocated[crq->is_write]--;
+		put_io_context(crq->io_context->ioc);
 
 		mempool_free(crq, cfqd->crq_pool);
 		rq->elevator_private = NULL;
 
-		smp_mb();
 		cfq_check_waiters(q, cfqq);
 		cfq_put_queue(cfqq);
 	}
 }
 
 /*
- * Allocate cfq data structures associated with this request. A queue and
+ * Allocate cfq data structures associated with this request.
  */
-static int cfq_set_request(request_queue_t *q, struct request *rq, int gfp_mask)
+static int
+cfq_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
+		int gfp_mask)
 {
 	struct cfq_data *cfqd = q->elevator->elevator_data;
 	struct cfq_io_context *cic;
 	const int rw = rq_data_dir(rq);
-	struct cfq_queue *cfqq, *saved_cfqq;
+	struct cfq_queue *cfqq;
 	struct cfq_rq *crq;
 	unsigned long flags;
 
 	might_sleep_if(gfp_mask & __GFP_WAIT);
 
+	cic = cfq_get_io_context(cfqd, cfq_queue_pid(current, rw), gfp_mask);
+
 	spin_lock_irqsave(q->queue_lock, flags);
 
-	cfqq = __cfq_get_queue(cfqd, cfq_hash_key(cfqd, current), gfp_mask);
-	if (!cfqq)
-		goto out_lock;
+	if (!cic)
+		goto queue_fail;
+
+	if (!cic->cfqq) {
+		cfqq = cfq_get_queue(cfqd, current->pid, gfp_mask);
+		if (!cfqq)
+			goto queue_fail;
 
-repeat:
-	if (cfqq->allocated[rw] >= cfqd->max_queued)
-		goto out_lock;
+		cic->cfqq = cfqq;
+	} else
+		cfqq = cic->cfqq;
 
 	cfqq->allocated[rw]++;
+	cfqq->must_alloc = 0;
+	cfqd->rq_starved = 0;
+	atomic_inc(&cfqq->ref);
 	spin_unlock_irqrestore(q->queue_lock, flags);
 
-	/*
-	 * if hashing type has changed, the cfq_queue might change here.
-	 */
-	saved_cfqq = cfqq;
-	cic = cfq_get_io_context(&cfqq, gfp_mask);
-	if (!cic)
-		goto err;
-
-	/*
-	 * repeat allocation checks on queue change
-	 */
-	if (unlikely(saved_cfqq != cfqq)) {
-		spin_lock_irqsave(q->queue_lock, flags);
-		saved_cfqq->allocated[rw]--;
-		goto repeat;
-	}
-
 	crq = mempool_alloc(cfqd->crq_pool, gfp_mask);
 	if (crq) {
 		RB_CLEAR(&crq->rb_node);
@@ -1460,24 +1980,130 @@ repeat:
 		INIT_HLIST_NODE(&crq->hash);
 		crq->cfq_queue = cfqq;
 		crq->io_context = cic;
-		crq->service_start = crq->queue_start = 0;
-		crq->in_flight = crq->accounted = crq->is_sync = 0;
-		crq->is_write = rw;
+		crq->in_flight = crq->accounted = 0;
+		crq->is_sync = (rw == READ || process_sync(current));
+		crq->requeued = 0;
 		rq->elevator_private = crq;
-		cfqq->alloc_limit[rw] = 0;
 		return 0;
 	}
 
-	put_io_context(cic->ioc);
-err:
 	spin_lock_irqsave(q->queue_lock, flags);
 	cfqq->allocated[rw]--;
+	if (!(cfqq->allocated[0] + cfqq->allocated[1]))
+		cfqq->must_alloc = 1;
 	cfq_put_queue(cfqq);
-out_lock:
+queue_fail:
+	if (cic)
+		put_io_context(cic->ioc);
+	/*
+	 * mark us rq allocation starved. we need to kickstart the process
+	 * ourselves if there are no pending requests that can do it for us.
+	 * that would be an extremely rare OOM situation
+	 */
+	cfqd->rq_starved = 1;
+	kblockd_schedule_work(&cfqd->unplug_work);
 	spin_unlock_irqrestore(q->queue_lock, flags);
 	return 1;
 }
 
+static void cfq_kick_queue(void *data)
+{
+	request_queue_t *q = data;
+	struct cfq_data *cfqd = q->elevator->elevator_data;
+	unsigned long flags;
+
+	spin_lock_irqsave(q->queue_lock, flags);
+
+	if (cfqd->rq_starved) {
+		struct request_list *rl = &q->rq;
+
+		/*
+		 * we aren't guaranteed to get a request after this, but we
+		 * have to be opportunistic
+		 */
+		smp_mb();
+		if (waitqueue_active(&rl->wait[READ]))
+			wake_up(&rl->wait[READ]);
+		if (waitqueue_active(&rl->wait[WRITE]))
+			wake_up(&rl->wait[WRITE]);
+	}
+
+	blk_remove_plug(q);
+	q->request_fn(q);
+	spin_unlock_irqrestore(q->queue_lock, flags);
+}
+
+/*
+ * Timer running if the active_queue is currently idling inside its time slice
+ */
+static void cfq_idle_slice_timer(unsigned long data)
+{
+	struct cfq_data *cfqd = (struct cfq_data *) data;
+	struct cfq_queue *cfqq;
+	unsigned long flags;
+
+	spin_lock_irqsave(cfqd->queue->queue_lock, flags);
+
+	if ((cfqq = cfqd->active_queue) != NULL) {
+		unsigned long now = jiffies;
+
+		/*
+		 * expired
+		 */
+		if (time_after(now, cfqq->slice_end))
+			goto expire;
+
+		/*
+		 * only expire and reinvoke request handler, if there are
+		 * other queues with pending requests
+		 */
+		if (!cfq_pending_requests(cfqd)) {
+			cfqd->idle_slice_timer.expires = min(now + cfqd->cfq_slice_idle, cfqq->slice_end);
+			add_timer(&cfqd->idle_slice_timer);
+			goto out_cont;
+		}
+
+		/*
+		 * not expired and it has a request pending, let it dispatch
+		 */
+		if (!RB_EMPTY(&cfqq->sort_list)) {
+			cfqq->must_dispatch = 1;
+			goto out_kick;
+		}
+	}
+expire:
+	cfq_slice_expired(cfqd, 0);
+out_kick:
+	if (cfq_pending_requests(cfqd))
+		kblockd_schedule_work(&cfqd->unplug_work);
+out_cont:
+	spin_unlock_irqrestore(cfqd->queue->queue_lock, flags);
+}
+
+/*
+ * Timer running if an idle class queue is waiting for service
+ */
+static void cfq_idle_class_timer(unsigned long data)
+{
+	struct cfq_data *cfqd = (struct cfq_data *) data;
+	unsigned long flags, end;
+
+	spin_lock_irqsave(cfqd->queue->queue_lock, flags);
+
+	/*
+	 * race with a non-idle queue, reset timer
+	 */
+	end = cfqd->last_end_request + CFQ_IDLE_GRACE;
+	if (!time_after_eq(jiffies, end)) {
+		cfqd->idle_class_timer.expires = end;
+		add_timer(&cfqd->idle_class_timer);
+	} else
+		kblockd_schedule_work(&cfqd->unplug_work);
+
+	spin_unlock_irqrestore(cfqd->queue->queue_lock, flags);
+}
+
+
 static void cfq_put_cfqd(struct cfq_data *cfqd)
 {
 	request_queue_t *q = cfqd->queue;
@@ -1485,6 +2111,8 @@ static void cfq_put_cfqd(struct cfq_data *cfqd)
 	if (!atomic_dec_and_test(&cfqd->ref))
 		return;
 
+	blk_sync_queue(q);
+
 	blk_put_queue(q);
 
 	mempool_destroy(cfqd->crq_pool);
@@ -1495,7 +2123,11 @@ static void cfq_put_cfqd(struct cfq_data *cfqd)
 
 static void cfq_exit_queue(elevator_t *e)
 {
-	cfq_put_cfqd(e->elevator_data);
+	struct cfq_data *cfqd = e->elevator_data;
+
+	del_timer_sync(&cfqd->idle_slice_timer);
+	del_timer_sync(&cfqd->idle_class_timer);
+	cfq_put_cfqd(cfqd);
 }
 
 static int cfq_init_queue(request_queue_t *q, elevator_t *e)
@@ -1508,7 +2140,13 @@ static int cfq_init_queue(request_queue_t *q, elevator_t *e)
 		return -ENOMEM;
 
 	memset(cfqd, 0, sizeof(*cfqd));
-	INIT_LIST_HEAD(&cfqd->rr_list);
+
+	for (i = 0; i < CFQ_PRIO_LISTS; i++)
+		INIT_LIST_HEAD(&cfqd->rr_list[i]);
+
+	INIT_LIST_HEAD(&cfqd->busy_rr);
+	INIT_LIST_HEAD(&cfqd->cur_rr);
+	INIT_LIST_HEAD(&cfqd->idle_rr);
 	INIT_LIST_HEAD(&cfqd->empty_list);
 
 	cfqd->crq_hash = kmalloc(sizeof(struct hlist_head) * CFQ_MHASH_ENTRIES, GFP_KERNEL);
@@ -1533,25 +2171,32 @@ static int cfq_init_queue(request_queue_t *q, elevator_t *e)
 	cfqd->queue = q;
 	atomic_inc(&q->refcnt);
 
-	/*
-	 * just set it to some high value, we want anyone to be able to queue
-	 * some requests. fairness is handled differently
-	 */
-	q->nr_requests = 1024;
-	cfqd->max_queued = q->nr_requests / 16;
+	cfqd->max_queued = q->nr_requests / 4;
 	q->nr_batching = cfq_queued;
-	cfqd->key_type = CFQ_KEY_TGID;
-	cfqd->find_best_crq = 1;
+
+	init_timer(&cfqd->idle_slice_timer);
+	cfqd->idle_slice_timer.function = cfq_idle_slice_timer;
+	cfqd->idle_slice_timer.data = (unsigned long) cfqd;
+
+	init_timer(&cfqd->idle_class_timer);
+	cfqd->idle_class_timer.function = cfq_idle_class_timer;
+	cfqd->idle_class_timer.data = (unsigned long) cfqd;
+
+	INIT_WORK(&cfqd->unplug_work, cfq_kick_queue, q);
+
 	atomic_set(&cfqd->ref, 1);
 
 	cfqd->cfq_queued = cfq_queued;
 	cfqd->cfq_quantum = cfq_quantum;
-	cfqd->cfq_fifo_expire_r = cfq_fifo_expire_r;
-	cfqd->cfq_fifo_expire_w = cfq_fifo_expire_w;
-	cfqd->cfq_fifo_batch_expire = cfq_fifo_rate;
+	cfqd->cfq_fifo_expire[0] = cfq_fifo_expire[0];
+	cfqd->cfq_fifo_expire[1] = cfq_fifo_expire[1];
 	cfqd->cfq_back_max = cfq_back_max;
 	cfqd->cfq_back_penalty = cfq_back_penalty;
-
+	cfqd->cfq_slice[0] = cfq_slice_async;
+	cfqd->cfq_slice[1] = cfq_slice_sync;
+	cfqd->cfq_slice_async_rq = cfq_slice_async_rq;
+	cfqd->cfq_slice_idle = cfq_slice_idle;
+	cfqd->cfq_max_depth = cfq_max_depth;
 	return 0;
 out_crqpool:
 	kfree(cfqd->cfq_hash);
@@ -1595,7 +2240,6 @@ fail:
 	return -ENOMEM;
 }
 
-
 /*
  * sysfs parts below -->
  */
@@ -1620,45 +2264,6 @@ cfq_var_store(unsigned int *var, const char *page, size_t count)
 	return count;
 }
 
-static ssize_t
-cfq_clear_elapsed(struct cfq_data *cfqd, const char *page, size_t count)
-{
-	max_elapsed_dispatch = max_elapsed_crq = 0;
-	return count;
-}
-
-static ssize_t
-cfq_set_key_type(struct cfq_data *cfqd, const char *page, size_t count)
-{
-	spin_lock_irq(cfqd->queue->queue_lock);
-	if (!strncmp(page, "pgid", 4))
-		cfqd->key_type = CFQ_KEY_PGID;
-	else if (!strncmp(page, "tgid", 4))
-		cfqd->key_type = CFQ_KEY_TGID;
-	else if (!strncmp(page, "uid", 3))
-		cfqd->key_type = CFQ_KEY_UID;
-	else if (!strncmp(page, "gid", 3))
-		cfqd->key_type = CFQ_KEY_GID;
-	spin_unlock_irq(cfqd->queue->queue_lock);
-	return count;
-}
-
-static ssize_t
-cfq_read_key_type(struct cfq_data *cfqd, char *page)
-{
-	ssize_t len = 0;
-	int i;
-
-	for (i = CFQ_KEY_PGID; i < CFQ_KEY_LAST; i++) {
-		if (cfqd->key_type == i)
-			len += sprintf(page+len, "[%s] ", cfq_key_types[i]);
-		else
-			len += sprintf(page+len, "%s ", cfq_key_types[i]);
-	}
-	len += sprintf(page+len, "\n");
-	return len;
-}
-
 #define SHOW_FUNCTION(__FUNC, __VAR, __CONV)				\
 static ssize_t __FUNC(struct cfq_data *cfqd, char *page)		\
 {									\
@@ -1669,12 +2274,15 @@ static ssize_t __FUNC(struct cfq_data *cfqd, char *page)		\
 }
 SHOW_FUNCTION(cfq_quantum_show, cfqd->cfq_quantum, 0);
 SHOW_FUNCTION(cfq_queued_show, cfqd->cfq_queued, 0);
-SHOW_FUNCTION(cfq_fifo_expire_r_show, cfqd->cfq_fifo_expire_r, 1);
-SHOW_FUNCTION(cfq_fifo_expire_w_show, cfqd->cfq_fifo_expire_w, 1);
-SHOW_FUNCTION(cfq_fifo_batch_expire_show, cfqd->cfq_fifo_batch_expire, 1);
-SHOW_FUNCTION(cfq_find_best_show, cfqd->find_best_crq, 0);
+SHOW_FUNCTION(cfq_fifo_expire_sync_show, cfqd->cfq_fifo_expire[1], 1);
+SHOW_FUNCTION(cfq_fifo_expire_async_show, cfqd->cfq_fifo_expire[0], 1);
 SHOW_FUNCTION(cfq_back_max_show, cfqd->cfq_back_max, 0);
 SHOW_FUNCTION(cfq_back_penalty_show, cfqd->cfq_back_penalty, 0);
+SHOW_FUNCTION(cfq_slice_idle_show, cfqd->cfq_slice_idle, 1);
+SHOW_FUNCTION(cfq_slice_sync_show, cfqd->cfq_slice[1], 1);
+SHOW_FUNCTION(cfq_slice_async_show, cfqd->cfq_slice[0], 1);
+SHOW_FUNCTION(cfq_slice_async_rq_show, cfqd->cfq_slice_async_rq, 0);
+SHOW_FUNCTION(cfq_max_depth_show, cfqd->cfq_max_depth, 0);
 #undef SHOW_FUNCTION
 
 #define STORE_FUNCTION(__FUNC, __PTR, MIN, MAX, __CONV)			\
@@ -1694,12 +2302,15 @@ static ssize_t __FUNC(struct cfq_data *cfqd, const char *page, size_t count)	\
 }
 STORE_FUNCTION(cfq_quantum_store, &cfqd->cfq_quantum, 1, UINT_MAX, 0);
 STORE_FUNCTION(cfq_queued_store, &cfqd->cfq_queued, 1, UINT_MAX, 0);
-STORE_FUNCTION(cfq_fifo_expire_r_store, &cfqd->cfq_fifo_expire_r, 1, UINT_MAX, 1);
-STORE_FUNCTION(cfq_fifo_expire_w_store, &cfqd->cfq_fifo_expire_w, 1, UINT_MAX, 1);
-STORE_FUNCTION(cfq_fifo_batch_expire_store, &cfqd->cfq_fifo_batch_expire, 0, UINT_MAX, 1);
-STORE_FUNCTION(cfq_find_best_store, &cfqd->find_best_crq, 0, 1, 0);
+STORE_FUNCTION(cfq_fifo_expire_sync_store, &cfqd->cfq_fifo_expire[1], 1, UINT_MAX, 1);
+STORE_FUNCTION(cfq_fifo_expire_async_store, &cfqd->cfq_fifo_expire[0], 1, UINT_MAX, 1);
 STORE_FUNCTION(cfq_back_max_store, &cfqd->cfq_back_max, 0, UINT_MAX, 0);
 STORE_FUNCTION(cfq_back_penalty_store, &cfqd->cfq_back_penalty, 1, UINT_MAX, 0);
+STORE_FUNCTION(cfq_slice_idle_store, &cfqd->cfq_slice_idle, 0, UINT_MAX, 1);
+STORE_FUNCTION(cfq_slice_sync_store, &cfqd->cfq_slice[1], 1, UINT_MAX, 1);
+STORE_FUNCTION(cfq_slice_async_store, &cfqd->cfq_slice[0], 1, UINT_MAX, 1);
+STORE_FUNCTION(cfq_slice_async_rq_store, &cfqd->cfq_slice_async_rq, 1, UINT_MAX, 0);
+STORE_FUNCTION(cfq_max_depth_store, &cfqd->cfq_max_depth, 1, UINT_MAX, 0);
 #undef STORE_FUNCTION
 
 static struct cfq_fs_entry cfq_quantum_entry = {
@@ -1712,25 +2323,15 @@ static struct cfq_fs_entry cfq_queued_entry = {
 	.show = cfq_queued_show,
 	.store = cfq_queued_store,
 };
-static struct cfq_fs_entry cfq_fifo_expire_r_entry = {
+static struct cfq_fs_entry cfq_fifo_expire_sync_entry = {
 	.attr = {.name = "fifo_expire_sync", .mode = S_IRUGO | S_IWUSR },
-	.show = cfq_fifo_expire_r_show,
-	.store = cfq_fifo_expire_r_store,
+	.show = cfq_fifo_expire_sync_show,
+	.store = cfq_fifo_expire_sync_store,
 };
-static struct cfq_fs_entry cfq_fifo_expire_w_entry = {
+static struct cfq_fs_entry cfq_fifo_expire_async_entry = {
 	.attr = {.name = "fifo_expire_async", .mode = S_IRUGO | S_IWUSR },
-	.show = cfq_fifo_expire_w_show,
-	.store = cfq_fifo_expire_w_store,
-};
-static struct cfq_fs_entry cfq_fifo_batch_expire_entry = {
-	.attr = {.name = "fifo_batch_expire", .mode = S_IRUGO | S_IWUSR },
-	.show = cfq_fifo_batch_expire_show,
-	.store = cfq_fifo_batch_expire_store,
-};
-static struct cfq_fs_entry cfq_find_best_entry = {
-	.attr = {.name = "find_best_crq", .mode = S_IRUGO | S_IWUSR },
-	.show = cfq_find_best_show,
-	.store = cfq_find_best_store,
+	.show = cfq_fifo_expire_async_show,
+	.store = cfq_fifo_expire_async_store,
 };
 static struct cfq_fs_entry cfq_back_max_entry = {
 	.attr = {.name = "back_seek_max", .mode = S_IRUGO | S_IWUSR },
@@ -1742,27 +2343,43 @@ static struct cfq_fs_entry cfq_back_penalty_entry = {
 	.show = cfq_back_penalty_show,
 	.store = cfq_back_penalty_store,
 };
-static struct cfq_fs_entry cfq_clear_elapsed_entry = {
-	.attr = {.name = "clear_elapsed", .mode = S_IWUSR },
-	.store = cfq_clear_elapsed,
+static struct cfq_fs_entry cfq_slice_sync_entry = {
+	.attr = {.name = "slice_sync", .mode = S_IRUGO | S_IWUSR },
+	.show = cfq_slice_sync_show,
+	.store = cfq_slice_sync_store,
 };
-static struct cfq_fs_entry cfq_key_type_entry = {
-	.attr = {.name = "key_type", .mode = S_IRUGO | S_IWUSR },
-	.show = cfq_read_key_type,
-	.store = cfq_set_key_type,
+static struct cfq_fs_entry cfq_slice_async_entry = {
+	.attr = {.name = "slice_async", .mode = S_IRUGO | S_IWUSR },
+	.show = cfq_slice_async_show,
+	.store = cfq_slice_async_store,
+};
+static struct cfq_fs_entry cfq_slice_async_rq_entry = {
+	.attr = {.name = "slice_async_rq", .mode = S_IRUGO | S_IWUSR },
+	.show = cfq_slice_async_rq_show,
+	.store = cfq_slice_async_rq_store,
+};
+static struct cfq_fs_entry cfq_slice_idle_entry = {
+	.attr = {.name = "slice_idle", .mode = S_IRUGO | S_IWUSR },
+	.show = cfq_slice_idle_show,
+	.store = cfq_slice_idle_store,
+};
+static struct cfq_fs_entry cfq_max_depth_entry = {
+	.attr = {.name = "max_depth", .mode = S_IRUGO | S_IWUSR },
+	.show = cfq_max_depth_show,
+	.store = cfq_max_depth_store,
 };
-
 static struct attribute *default_attrs[] = {
 	&cfq_quantum_entry.attr,
 	&cfq_queued_entry.attr,
-	&cfq_fifo_expire_r_entry.attr,
-	&cfq_fifo_expire_w_entry.attr,
-	&cfq_fifo_batch_expire_entry.attr,
-	&cfq_key_type_entry.attr,
-	&cfq_find_best_entry.attr,
+	&cfq_fifo_expire_sync_entry.attr,
+	&cfq_fifo_expire_async_entry.attr,
 	&cfq_back_max_entry.attr,
 	&cfq_back_penalty_entry.attr,
-	&cfq_clear_elapsed_entry.attr,
+	&cfq_slice_sync_entry.attr,
+	&cfq_slice_async_entry.attr,
+	&cfq_slice_async_rq_entry.attr,
+	&cfq_slice_idle_entry.attr,
+	&cfq_max_depth_entry.attr,
 	NULL,
 };
 
@@ -1832,21 +2449,46 @@ static int __init cfq_init(void)
 {
 	int ret;
 
+	/*
+	 * could be 0 on HZ < 1000 setups
+	 */
+	if (!cfq_slice_async)
+		cfq_slice_async = 1;
+	if (!cfq_slice_idle)
+		cfq_slice_idle = 1;
+
 	if (cfq_slab_setup())
 		return -ENOMEM;
 
 	ret = elv_register(&iosched_cfq);
-	if (!ret) {
-		__module_get(THIS_MODULE);
-		return 0;
-	}
+	if (ret)
+		cfq_slab_kill();
 
-	cfq_slab_kill();
 	return ret;
 }
 
 static void __exit cfq_exit(void)
 {
+	struct task_struct *g, *p;
+	unsigned long flags;
+
+	read_lock_irqsave(&tasklist_lock, flags);
+
+	/*
+	 * iterate each process in the system, removing our io_context
+	 */
+	do_each_thread(g, p) {
+		struct io_context *ioc = p->io_context;
+
+		if (ioc && ioc->cic) {
+			ioc->cic->exit(ioc->cic);
+			cfq_free_io_context(ioc->cic);
+			ioc->cic = NULL;
+		}
+	} while_each_thread(g, p);
+
+	read_unlock_irqrestore(&tasklist_lock, flags);
+
 	cfq_slab_kill();
 	elv_unregister(&iosched_cfq);
 }
diff --git a/drivers/block/deadline-iosched.c b/drivers/block/deadline-iosched.c
index 4bc2fea7327..ff5201e0215 100644
--- a/drivers/block/deadline-iosched.c
+++ b/drivers/block/deadline-iosched.c
@@ -760,7 +760,8 @@ static void deadline_put_request(request_queue_t *q, struct request *rq)
 }
 
 static int
-deadline_set_request(request_queue_t *q, struct request *rq, int gfp_mask)
+deadline_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
+		     int gfp_mask)
 {
 	struct deadline_data *dd = q->elevator->elevator_data;
 	struct deadline_rq *drq;
diff --git a/drivers/block/elevator.c b/drivers/block/elevator.c
index f831f08f839..98f0126a2de 100644
--- a/drivers/block/elevator.c
+++ b/drivers/block/elevator.c
@@ -486,12 +486,13 @@ struct request *elv_former_request(request_queue_t *q, struct request *rq)
 	return NULL;
 }
 
-int elv_set_request(request_queue_t *q, struct request *rq, int gfp_mask)
+int elv_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
+		    int gfp_mask)
 {
 	elevator_t *e = q->elevator;
 
 	if (e->ops->elevator_set_req_fn)
-		return e->ops->elevator_set_req_fn(q, rq, gfp_mask);
+		return e->ops->elevator_set_req_fn(q, rq, bio, gfp_mask);
 
 	rq->elevator_private = NULL;
 	return 0;
@@ -505,12 +506,12 @@ void elv_put_request(request_queue_t *q, struct request *rq)
 		e->ops->elevator_put_req_fn(q, rq);
 }
 
-int elv_may_queue(request_queue_t *q, int rw)
+int elv_may_queue(request_queue_t *q, int rw, struct bio *bio)
 {
 	elevator_t *e = q->elevator;
 
 	if (e->ops->elevator_may_queue_fn)
-		return e->ops->elevator_may_queue_fn(q, rw);
+		return e->ops->elevator_may_queue_fn(q, rw, bio);
 
 	return ELV_MQUEUE_MAY;
 }
diff --git a/drivers/block/ll_rw_blk.c b/drivers/block/ll_rw_blk.c
index 60e64091de1..234fdcfbdf0 100644
--- a/drivers/block/ll_rw_blk.c
+++ b/drivers/block/ll_rw_blk.c
@@ -276,6 +276,7 @@ static inline void rq_init(request_queue_t *q, struct request *rq)
 	rq->errors = 0;
 	rq->rq_status = RQ_ACTIVE;
 	rq->bio = rq->biotail = NULL;
+	rq->ioprio = 0;
 	rq->buffer = NULL;
 	rq->ref_count = 1;
 	rq->q = q;
@@ -1442,11 +1443,7 @@ void __generic_unplug_device(request_queue_t *q)
 	if (!blk_remove_plug(q))
 		return;
 
-	/*
-	 * was plugged, fire request_fn if queue has stuff to do
-	 */
-	if (elv_next_request(q))
-		q->request_fn(q);
+	q->request_fn(q);
 }
 EXPORT_SYMBOL(__generic_unplug_device);
 
@@ -1776,8 +1773,8 @@ static inline void blk_free_request(request_queue_t *q, struct request *rq)
 	mempool_free(rq, q->rq.rq_pool);
 }
 
-static inline struct request *blk_alloc_request(request_queue_t *q, int rw,
-						int gfp_mask)
+static inline struct request *
+blk_alloc_request(request_queue_t *q, int rw, struct bio *bio, int gfp_mask)
 {
 	struct request *rq = mempool_alloc(q->rq.rq_pool, gfp_mask);
 
@@ -1790,7 +1787,7 @@ static inline struct request *blk_alloc_request(request_queue_t *q, int rw,
 	 */
 	rq->flags = rw;
 
-	if (!elv_set_request(q, rq, gfp_mask))
+	if (!elv_set_request(q, rq, bio, gfp_mask))
 		return rq;
 
 	mempool_free(rq, q->rq.rq_pool);
@@ -1872,7 +1869,8 @@ static void freed_request(request_queue_t *q, int rw)
 /*
  * Get a free request, queue_lock must not be held
  */
-static struct request *get_request(request_queue_t *q, int rw, int gfp_mask)
+static struct request *get_request(request_queue_t *q, int rw, struct bio *bio,
+				   int gfp_mask)
 {
 	struct request *rq = NULL;
 	struct request_list *rl = &q->rq;
@@ -1895,7 +1893,7 @@ static struct request *get_request(request_queue_t *q, int rw, int gfp_mask)
 		}
 	}
 
-	switch (elv_may_queue(q, rw)) {
+	switch (elv_may_queue(q, rw, bio)) {
 		case ELV_MQUEUE_NO:
 			goto rq_starved;
 		case ELV_MQUEUE_MAY:
@@ -1920,7 +1918,7 @@ get_rq:
 		set_queue_congested(q, rw);
 	spin_unlock_irq(q->queue_lock);
 
-	rq = blk_alloc_request(q, rw, gfp_mask);
+	rq = blk_alloc_request(q, rw, bio, gfp_mask);
 	if (!rq) {
 		/*
 		 * Allocation failed presumably due to memory. Undo anything
@@ -1961,7 +1959,8 @@ out:
  * No available requests for this queue, unplug the device and wait for some
  * requests to become available.
  */
-static struct request *get_request_wait(request_queue_t *q, int rw)
+static struct request *get_request_wait(request_queue_t *q, int rw,
+					struct bio *bio)
 {
 	DEFINE_WAIT(wait);
 	struct request *rq;
@@ -1972,7 +1971,7 @@ static struct request *get_request_wait(request_queue_t *q, int rw)
 		prepare_to_wait_exclusive(&rl->wait[rw], &wait,
 				TASK_UNINTERRUPTIBLE);
 
-		rq = get_request(q, rw, GFP_NOIO);
+		rq = get_request(q, rw, bio, GFP_NOIO);
 
 		if (!rq) {
 			struct io_context *ioc;
@@ -2003,9 +2002,9 @@ struct request *blk_get_request(request_queue_t *q, int rw, int gfp_mask)
 	BUG_ON(rw != READ && rw != WRITE);
 
 	if (gfp_mask & __GFP_WAIT)
-		rq = get_request_wait(q, rw);
+		rq = get_request_wait(q, rw, NULL);
 	else
-		rq = get_request(q, rw, gfp_mask);
+		rq = get_request(q, rw, NULL, gfp_mask);
 
 	return rq;
 }
@@ -2333,7 +2332,6 @@ static void __blk_put_request(request_queue_t *q, struct request *req)
 		return;
 
 	req->rq_status = RQ_INACTIVE;
-	req->q = NULL;
 	req->rl = NULL;
 
 	/*
@@ -2462,6 +2460,8 @@ static int attempt_merge(request_queue_t *q, struct request *req,
 		req->rq_disk->in_flight--;
 	}
 
+	req->ioprio = ioprio_best(req->ioprio, next->ioprio);
+
 	__blk_put_request(q, next);
 	return 1;
 }
@@ -2514,11 +2514,13 @@ static int __make_request(request_queue_t *q, struct bio *bio)
 {
 	struct request *req, *freereq = NULL;
 	int el_ret, rw, nr_sectors, cur_nr_sectors, barrier, err, sync;
+	unsigned short prio;
 	sector_t sector;
 
 	sector = bio->bi_sector;
 	nr_sectors = bio_sectors(bio);
 	cur_nr_sectors = bio_cur_sectors(bio);
+	prio = bio_prio(bio);
 
 	rw = bio_data_dir(bio);
 	sync = bio_sync(bio);
@@ -2559,6 +2561,7 @@ again:
 			req->biotail->bi_next = bio;
 			req->biotail = bio;
 			req->nr_sectors = req->hard_nr_sectors += nr_sectors;
+			req->ioprio = ioprio_best(req->ioprio, prio);
 			drive_stat_acct(req, nr_sectors, 0);
 			if (!attempt_back_merge(q, req))
 				elv_merged_request(q, req);
@@ -2583,6 +2586,7 @@ again:
 			req->hard_cur_sectors = cur_nr_sectors;
 			req->sector = req->hard_sector = sector;
 			req->nr_sectors = req->hard_nr_sectors += nr_sectors;
+			req->ioprio = ioprio_best(req->ioprio, prio);
 			drive_stat_acct(req, nr_sectors, 0);
 			if (!attempt_front_merge(q, req))
 				elv_merged_request(q, req);
@@ -2610,7 +2614,7 @@ get_rq:
 		freereq = NULL;
 	} else {
 		spin_unlock_irq(q->queue_lock);
-		if ((freereq = get_request(q, rw, GFP_ATOMIC)) == NULL) {
+		if ((freereq = get_request(q, rw, bio, GFP_ATOMIC)) == NULL) {
 			/*
 			 * READA bit set
 			 */
@@ -2618,7 +2622,7 @@ get_rq:
 			if (bio_rw_ahead(bio))
 				goto end_io;
 	
-			freereq = get_request_wait(q, rw);
+			freereq = get_request_wait(q, rw, bio);
 		}
 		goto again;
 	}
@@ -2646,6 +2650,7 @@ get_rq:
 	req->buffer = bio_data(bio);	/* see ->buffer comment above */
 	req->waiting = NULL;
 	req->bio = req->biotail = bio;
+	req->ioprio = prio;
 	req->rq_disk = bio->bi_bdev->bd_disk;
 	req->start_time = jiffies;
 
@@ -2674,7 +2679,7 @@ static inline void blk_partition_remap(struct bio *bio)
 	if (bdev != bdev->bd_contains) {
 		struct hd_struct *p = bdev->bd_part;
 
-		switch (bio->bi_rw) {
+		switch (bio_data_dir(bio)) {
 		case READ:
 			p->read_sectors += bio_sectors(bio);
 			p->reads++;
@@ -2693,6 +2698,7 @@ void blk_finish_queue_drain(request_queue_t *q)
 {
 	struct request_list *rl = &q->rq;
 	struct request *rq;
+	int requeued = 0;
 
 	spin_lock_irq(q->queue_lock);
 	clear_bit(QUEUE_FLAG_DRAIN, &q->queue_flags);
@@ -2701,9 +2707,13 @@ void blk_finish_queue_drain(request_queue_t *q)
 		rq = list_entry_rq(q->drain_list.next);
 
 		list_del_init(&rq->queuelist);
-		__elv_add_request(q, rq, ELEVATOR_INSERT_BACK, 1);
+		elv_requeue_request(q, rq);
+		requeued++;
 	}
 
+	if (requeued)
+		q->request_fn(q);
+
 	spin_unlock_irq(q->queue_lock);
 
 	wake_up(&rl->wait[0]);
@@ -2900,7 +2910,7 @@ void submit_bio(int rw, struct bio *bio)
 
 	BIO_BUG_ON(!bio->bi_size);
 	BIO_BUG_ON(!bio->bi_io_vec);
-	bio->bi_rw = rw;
+	bio->bi_rw |= rw;
 	if (rw & WRITE)
 		mod_page_state(pgpgout, count);
 	else
@@ -3257,8 +3267,11 @@ void exit_io_context(void)
 	struct io_context *ioc;
 
 	local_irq_save(flags);
+	task_lock(current);
 	ioc = current->io_context;
 	current->io_context = NULL;
+	ioc->task = NULL;
+	task_unlock(current);
 	local_irq_restore(flags);
 
 	if (ioc->aic && ioc->aic->exit)
@@ -3293,12 +3306,12 @@ struct io_context *get_io_context(int gfp_flags)
 	ret = kmem_cache_alloc(iocontext_cachep, gfp_flags);
 	if (ret) {
 		atomic_set(&ret->refcount, 1);
-		ret->pid = tsk->pid;
+		ret->task = current;
+		ret->set_ioprio = NULL;
 		ret->last_waited = jiffies; /* doesn't matter... */
 		ret->nr_batch_requests = 0; /* because this is 0 */
 		ret->aic = NULL;
 		ret->cic = NULL;
-		spin_lock_init(&ret->lock);
 
 		local_irq_save(flags);
 
diff --git a/fs/Makefile b/fs/Makefile
index fc92e59e9fa..20edcf28bfd 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -10,6 +10,7 @@ obj-y :=	open.o read_write.o file_table.o buffer.o  bio.o super.o \
 		ioctl.o readdir.o select.o fifo.o locks.o dcache.o inode.o \
 		attr.o bad_inode.o file.o filesystems.o namespace.o aio.o \
 		seq_file.o xattr.o libfs.o fs-writeback.o mpage.o direct-io.o \
+		ioprio.o
 
 obj-$(CONFIG_EPOLL)		+= eventpoll.o
 obj-$(CONFIG_COMPAT)		+= compat.o
diff --git a/fs/ioprio.c b/fs/ioprio.c
new file mode 100644
index 00000000000..663e420636d
--- /dev/null
+++ b/fs/ioprio.c
@@ -0,0 +1,172 @@
+/*
+ * fs/ioprio.c
+ *
+ * Copyright (C) 2004 Jens Axboe <axboe@suse.de>
+ *
+ * Helper functions for setting/querying io priorities of processes. The
+ * system calls closely mimmick getpriority/setpriority, see the man page for
+ * those. The prio argument is a composite of prio class and prio data, where
+ * the data argument has meaning within that class. The standard scheduling
+ * classes have 8 distinct prio levels, with 0 being the highest prio and 7
+ * being the lowest.
+ *
+ * IOW, setting BE scheduling class with prio 2 is done ala:
+ *
+ * unsigned int prio = (IOPRIO_CLASS_BE << IOPRIO_CLASS_SHIFT) | 2;
+ *
+ * ioprio_set(PRIO_PROCESS, pid, prio);
+ *
+ * See also Documentation/block/ioprio.txt
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/ioprio.h>
+#include <linux/blkdev.h>
+
+static int set_task_ioprio(struct task_struct *task, int ioprio)
+{
+	struct io_context *ioc;
+
+	if (task->uid != current->euid &&
+	    task->uid != current->uid && !capable(CAP_SYS_NICE))
+		return -EPERM;
+
+	task_lock(task);
+
+	task->ioprio = ioprio;
+
+	ioc = task->io_context;
+	if (ioc && ioc->set_ioprio)
+		ioc->set_ioprio(ioc, ioprio);
+
+	task_unlock(task);
+	return 0;
+}
+
+asmlinkage int sys_ioprio_set(int which, int who, int ioprio)
+{
+	int class = IOPRIO_PRIO_CLASS(ioprio);
+	int data = IOPRIO_PRIO_DATA(ioprio);
+	struct task_struct *p, *g;
+	struct user_struct *user;
+	int ret;
+
+	switch (class) {
+		case IOPRIO_CLASS_RT:
+			if (!capable(CAP_SYS_ADMIN))
+				return -EPERM;
+			/* fall through, rt has prio field too */
+		case IOPRIO_CLASS_BE:
+			if (data >= IOPRIO_BE_NR || data < 0)
+				return -EINVAL;
+
+			break;
+		case IOPRIO_CLASS_IDLE:
+			break;
+		default:
+			return -EINVAL;
+	}
+
+	ret = -ESRCH;
+	read_lock_irq(&tasklist_lock);
+	switch (which) {
+		case IOPRIO_WHO_PROCESS:
+			if (!who)
+				p = current;
+			else
+				p = find_task_by_pid(who);
+			if (p)
+				ret = set_task_ioprio(p, ioprio);
+			break;
+		case IOPRIO_WHO_PGRP:
+			if (!who)
+				who = process_group(current);
+			do_each_task_pid(who, PIDTYPE_PGID, p) {
+				ret = set_task_ioprio(p, ioprio);
+				if (ret)
+					break;
+			} while_each_task_pid(who, PIDTYPE_PGID, p);
+			break;
+		case IOPRIO_WHO_USER:
+			if (!who)
+				user = current->user;
+			else
+				user = find_user(who);
+
+			if (!user)
+				break;
+
+			do_each_thread(g, p) {
+				if (p->uid != who)
+					continue;
+				ret = set_task_ioprio(p, ioprio);
+				if (ret)
+					break;
+			} while_each_thread(g, p);
+
+			if (who)
+				free_uid(user);
+			break;
+		default:
+			ret = -EINVAL;
+	}
+
+	read_unlock_irq(&tasklist_lock);
+	return ret;
+}
+
+asmlinkage int sys_ioprio_get(int which, int who)
+{
+	struct task_struct *g, *p;
+	struct user_struct *user;
+	int ret = -ESRCH;
+
+	read_lock_irq(&tasklist_lock);
+	switch (which) {
+		case IOPRIO_WHO_PROCESS:
+			if (!who)
+				p = current;
+			else
+				p = find_task_by_pid(who);
+			if (p)
+				ret = p->ioprio;
+			break;
+		case IOPRIO_WHO_PGRP:
+			if (!who)
+				who = process_group(current);
+			do_each_task_pid(who, PIDTYPE_PGID, p) {
+				if (ret == -ESRCH)
+					ret = p->ioprio;
+				else
+					ret = ioprio_best(ret, p->ioprio);
+			} while_each_task_pid(who, PIDTYPE_PGID, p);
+			break;
+		case IOPRIO_WHO_USER:
+			if (!who)
+				user = current->user;
+			else
+				user = find_user(who);
+
+			if (!user)
+				break;
+
+			do_each_thread(g, p) {
+				if (p->uid != user->uid)
+					continue;
+				if (ret == -ESRCH)
+					ret = p->ioprio;
+				else
+					ret = ioprio_best(ret, p->ioprio);
+			} while_each_thread(g, p);
+
+			if (who)
+				free_uid(user);
+			break;
+		default:
+			ret = -EINVAL;
+	}
+
+	read_unlock_irq(&tasklist_lock);
+	return ret;
+}
+
diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c
index 7b87707acc3..d1bcf0da672 100644
--- a/fs/reiserfs/journal.c
+++ b/fs/reiserfs/journal.c
@@ -645,18 +645,22 @@ struct buffer_chunk {
 
 static void write_chunk(struct buffer_chunk *chunk) {
     int i;
+    get_fs_excl();
     for (i = 0; i < chunk->nr ; i++) {
 	submit_logged_buffer(chunk->bh[i]) ;
     }
     chunk->nr = 0;
+    put_fs_excl();
 }
 
 static void write_ordered_chunk(struct buffer_chunk *chunk) {
     int i;
+    get_fs_excl();
     for (i = 0; i < chunk->nr ; i++) {
 	submit_ordered_buffer(chunk->bh[i]) ;
     }
     chunk->nr = 0;
+    put_fs_excl();
 }
 
 static int add_to_chunk(struct buffer_chunk *chunk, struct buffer_head *bh,
@@ -918,6 +922,8 @@ static int flush_commit_list(struct super_block *s, struct reiserfs_journal_list
     return 0 ;
   }
 
+  get_fs_excl();
+
   /* before we can put our commit blocks on disk, we have to make sure everyone older than
   ** us is on disk too
   */
@@ -1055,6 +1061,7 @@ put_jl:
 
   if (retval)
     reiserfs_abort (s, retval, "Journal write error in %s", __FUNCTION__);
+  put_fs_excl();
   return retval;
 }
 
@@ -1251,6 +1258,8 @@ static int flush_journal_list(struct super_block *s,
     return 0 ;
   }
 
+  get_fs_excl();
+
   /* if all the work is already done, get out of here */
   if (atomic_read(&(jl->j_nonzerolen)) <= 0 && 
       atomic_read(&(jl->j_commit_left)) <= 0) {
@@ -1450,6 +1459,7 @@ flush_older_and_return:
   put_journal_list(s, jl);
   if (flushall)
     up(&journal->j_flush_sem);
+  put_fs_excl();
   return err ;
 } 
 
@@ -2719,6 +2729,7 @@ relock:
   th->t_trans_id = journal->j_trans_id ;
   unlock_journal(p_s_sb) ;
   INIT_LIST_HEAD (&th->t_list);
+  get_fs_excl();
   return 0 ;
 
 out_fail:
@@ -3526,6 +3537,7 @@ static int do_journal_end(struct reiserfs_transaction_handle *th, struct super_b
   BUG_ON (th->t_refcount > 1);
   BUG_ON (!th->t_trans_id);
 
+  put_fs_excl();
   current->journal_info = th->t_handle_save;
   reiserfs_check_lock_depth(p_s_sb, "journal end");
   if (journal->j_len == 0) {
diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h
index 176413fb9ae..e25e4c71a87 100644
--- a/include/asm-i386/unistd.h
+++ b/include/asm-i386/unistd.h
@@ -294,8 +294,10 @@
 #define __NR_add_key		286
 #define __NR_request_key	287
 #define __NR_keyctl		288
+#define __NR_ioprio_set		289
+#define __NR_ioprio_get		290
 
-#define NR_syscalls 289
+#define NR_syscalls 291
 
 /*
  * user-visible error numbers are in the range -1 - -128: see
diff --git a/include/asm-ia64/unistd.h b/include/asm-ia64/unistd.h
index f7f43ec2483..517f1649ee6 100644
--- a/include/asm-ia64/unistd.h
+++ b/include/asm-ia64/unistd.h
@@ -263,6 +263,8 @@
 #define __NR_add_key			1271
 #define __NR_request_key		1272
 #define __NR_keyctl			1273
+#define __NR_ioprio_set			1274
+#define __NR_ioprio_get			1275
 #define __NR_set_zone_reclaim		1276
 
 #ifdef __KERNEL__
diff --git a/include/asm-ppc/unistd.h b/include/asm-ppc/unistd.h
index cc51e5c9acc..e8b79220b29 100644
--- a/include/asm-ppc/unistd.h
+++ b/include/asm-ppc/unistd.h
@@ -277,8 +277,10 @@
 #define __NR_request_key	270
 #define __NR_keyctl		271
 #define __NR_waitid		272
+#define __NR_ioprio_set		273
+#define __NR_ioprio_get		274
 
-#define __NR_syscalls		273
+#define __NR_syscalls		275
 
 #define __NR(n)	#n
 
diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h
index d767adcbf0f..6560439a83e 100644
--- a/include/asm-x86_64/unistd.h
+++ b/include/asm-x86_64/unistd.h
@@ -561,8 +561,12 @@ __SYSCALL(__NR_add_key, sys_add_key)
 __SYSCALL(__NR_request_key, sys_request_key)
 #define __NR_keyctl		250
 __SYSCALL(__NR_keyctl, sys_keyctl)
+#define __NR_ioprio_set		251
+__SYSCALL(__NR_ioprio_set, sys_ioprio_set)
+#define __NR_ioprio_get		252
+__SYSCALL(__NR_ioprio_get, sys_ioprio_get)
 
-#define __NR_syscall_max __NR_keyctl
+#define __NR_syscall_max __NR_ioprio_get
 #ifndef __NO_STUBS
 
 /* user-visible error numbers are in the range -1 - -4095 */
diff --git a/include/linux/bio.h b/include/linux/bio.h
index 038022763f0..36ef29fa0d8 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -22,6 +22,7 @@
 
 #include <linux/highmem.h>
 #include <linux/mempool.h>
+#include <linux/ioprio.h>
 
 /* Platforms may set this to teach the BIO layer about IOMMU hardware. */
 #include <asm/io.h>
@@ -149,6 +150,19 @@ struct bio {
 #define BIO_RW_FAILFAST	3
 #define BIO_RW_SYNC	4
 
+/*
+ * upper 16 bits of bi_rw define the io priority of this bio
+ */
+#define BIO_PRIO_SHIFT	(8 * sizeof(unsigned long) - IOPRIO_BITS)
+#define bio_prio(bio)	((bio)->bi_rw >> BIO_PRIO_SHIFT)
+#define bio_prio_valid(bio)	ioprio_valid(bio_prio(bio))
+
+#define bio_set_prio(bio, prio)		do {			\
+	WARN_ON(prio >= (1 << IOPRIO_BITS));			\
+	(bio)->bi_rw &= ((1UL << BIO_PRIO_SHIFT) - 1);		\
+	(bio)->bi_rw |= ((unsigned long) (prio) << BIO_PRIO_SHIFT);	\
+} while (0)
+
 /*
  * various member access, note that bio_data should of course not be used
  * on highmem page vectors
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index b54a0348a89..21a8674cd14 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -54,16 +54,23 @@ struct as_io_context {
 
 struct cfq_queue;
 struct cfq_io_context {
-	void (*dtor)(struct cfq_io_context *);
-	void (*exit)(struct cfq_io_context *);
-
-	struct io_context *ioc;
-
 	/*
 	 * circular list of cfq_io_contexts belonging to a process io context
 	 */
 	struct list_head list;
 	struct cfq_queue *cfqq;
+	void *key;
+
+	struct io_context *ioc;
+
+	unsigned long last_end_request;
+	unsigned long last_queue;
+	unsigned long ttime_total;
+	unsigned long ttime_samples;
+	unsigned long ttime_mean;
+
+	void (*dtor)(struct cfq_io_context *);
+	void (*exit)(struct cfq_io_context *);
 };
 
 /*
@@ -73,7 +80,9 @@ struct cfq_io_context {
  */
 struct io_context {
 	atomic_t refcount;
-	pid_t pid;
+	struct task_struct *task;
+
+	int (*set_ioprio)(struct io_context *, unsigned int);
 
 	/*
 	 * For request batching
@@ -81,8 +90,6 @@ struct io_context {
 	unsigned long last_waited; /* Time last woken after wait for request */
 	int nr_batch_requests;     /* Number of requests left in the batch */
 
-	spinlock_t lock;
-
 	struct as_io_context *aic;
 	struct cfq_io_context *cic;
 };
@@ -134,6 +141,8 @@ struct request {
 
 	void *elevator_private;
 
+	unsigned short ioprio;
+
 	int rq_status;	/* should split this into a few status bits */
 	struct gendisk *rq_disk;
 	int errors;
diff --git a/include/linux/elevator.h b/include/linux/elevator.h
index ee54f81faad..ea6bbc2d740 100644
--- a/include/linux/elevator.h
+++ b/include/linux/elevator.h
@@ -16,9 +16,9 @@ typedef void (elevator_remove_req_fn) (request_queue_t *, struct request *);
 typedef void (elevator_requeue_req_fn) (request_queue_t *, struct request *);
 typedef struct request *(elevator_request_list_fn) (request_queue_t *, struct request *);
 typedef void (elevator_completed_req_fn) (request_queue_t *, struct request *);
-typedef int (elevator_may_queue_fn) (request_queue_t *, int);
+typedef int (elevator_may_queue_fn) (request_queue_t *, int, struct bio *);
 
-typedef int (elevator_set_req_fn) (request_queue_t *, struct request *, int);
+typedef int (elevator_set_req_fn) (request_queue_t *, struct request *, struct bio *, int);
 typedef void (elevator_put_req_fn) (request_queue_t *, struct request *);
 typedef void (elevator_deactivate_req_fn) (request_queue_t *, struct request *);
 
@@ -96,9 +96,9 @@ extern struct request *elv_former_request(request_queue_t *, struct request *);
 extern struct request *elv_latter_request(request_queue_t *, struct request *);
 extern int elv_register_queue(request_queue_t *q);
 extern void elv_unregister_queue(request_queue_t *q);
-extern int elv_may_queue(request_queue_t *, int);
+extern int elv_may_queue(request_queue_t *, int, struct bio *);
 extern void elv_completed_request(request_queue_t *, struct request *);
-extern int elv_set_request(request_queue_t *, struct request *, int);
+extern int elv_set_request(request_queue_t *, struct request *, struct bio *, int);
 extern void elv_put_request(request_queue_t *, struct request *);
 
 /*
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 3ae8e37bdfc..047bde30836 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -213,6 +213,7 @@ extern int dir_notify_enable;
 #include <linux/radix-tree.h>
 #include <linux/prio_tree.h>
 #include <linux/init.h>
+#include <linux/sched.h>
 
 #include <asm/atomic.h>
 #include <asm/semaphore.h>
@@ -822,16 +823,34 @@ enum {
 #define vfs_check_frozen(sb, level) \
 	wait_event((sb)->s_wait_unfrozen, ((sb)->s_frozen < (level)))
 
+static inline void get_fs_excl(void)
+{
+	atomic_inc(&current->fs_excl);
+}
+
+static inline void put_fs_excl(void)
+{
+	atomic_dec(&current->fs_excl);
+}
+
+static inline int has_fs_excl(void)
+{
+	return atomic_read(&current->fs_excl);
+}
+
+
 /*
  * Superblock locking.
  */
 static inline void lock_super(struct super_block * sb)
 {
+	get_fs_excl();
 	down(&sb->s_lock);
 }
 
 static inline void unlock_super(struct super_block * sb)
 {
+	put_fs_excl();
 	up(&sb->s_lock);
 }
 
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 03206a425d7..c727c195a91 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -81,6 +81,7 @@ extern struct group_info init_groups;
 	.mm		= NULL,						\
 	.active_mm	= &init_mm,					\
 	.run_list	= LIST_HEAD_INIT(tsk.run_list),			\
+	.ioprio		= 0,						\
 	.time_slice	= HZ,						\
 	.tasks		= LIST_HEAD_INIT(tsk.tasks),			\
 	.ptrace_children= LIST_HEAD_INIT(tsk.ptrace_children),		\
@@ -110,6 +111,7 @@ extern struct group_info init_groups;
 	.proc_lock	= SPIN_LOCK_UNLOCKED,				\
 	.journal_info	= NULL,						\
 	.cpu_timers	= INIT_CPU_TIMERS(tsk.cpu_timers),		\
+	.fs_excl	= ATOMIC_INIT(0),				\
 }
 
 
diff --git a/include/linux/ioprio.h b/include/linux/ioprio.h
new file mode 100644
index 00000000000..7811300d88e
--- /dev/null
+++ b/include/linux/ioprio.h
@@ -0,0 +1,87 @@
+#ifndef IOPRIO_H
+#define IOPRIO_H
+
+#include <linux/sched.h>
+
+/*
+ * Gives us 8 prio classes with 13-bits of data for each class
+ */
+#define IOPRIO_BITS		(16)
+#define IOPRIO_CLASS_SHIFT	(13)
+#define IOPRIO_PRIO_MASK	((1UL << IOPRIO_CLASS_SHIFT) - 1)
+
+#define IOPRIO_PRIO_CLASS(mask)	((mask) >> IOPRIO_CLASS_SHIFT)
+#define IOPRIO_PRIO_DATA(mask)	((mask) & IOPRIO_PRIO_MASK)
+
+#define ioprio_valid(mask)	(IOPRIO_PRIO_CLASS((mask)) != IOPRIO_CLASS_NONE)
+
+/*
+ * These are the io priority groups as implemented by CFQ. RT is the realtime
+ * class, it always gets premium service. BE is the best-effort scheduling
+ * class, the default for any process. IDLE is the idle scheduling class, it
+ * is only served when no one else is using the disk.
+ */
+enum {
+	IOPRIO_CLASS_NONE,
+	IOPRIO_CLASS_RT,
+	IOPRIO_CLASS_BE,
+	IOPRIO_CLASS_IDLE,
+};
+
+/*
+ * 8 best effort priority levels are supported
+ */
+#define IOPRIO_BE_NR	(8)
+
+asmlinkage int sys_ioprio_set(int, int, int);
+asmlinkage int sys_ioprio_get(int, int);
+
+enum {
+	IOPRIO_WHO_PROCESS = 1,
+	IOPRIO_WHO_PGRP,
+	IOPRIO_WHO_USER,
+};
+
+/*
+ * if process has set io priority explicitly, use that. if not, convert
+ * the cpu scheduler nice value to an io priority
+ */
+#define IOPRIO_NORM	(4)
+static inline int task_ioprio(struct task_struct *task)
+{
+	WARN_ON(!ioprio_valid(task->ioprio));
+	return IOPRIO_PRIO_DATA(task->ioprio);
+}
+
+static inline int task_nice_ioprio(struct task_struct *task)
+{
+	return (task_nice(task) + 20) / 5;
+}
+
+/*
+ * For inheritance, return the highest of the two given priorities
+ */
+static inline int ioprio_best(unsigned short aprio, unsigned short bprio)
+{
+	unsigned short aclass = IOPRIO_PRIO_CLASS(aprio);
+	unsigned short bclass = IOPRIO_PRIO_CLASS(bprio);
+
+	if (!ioprio_valid(aprio))
+		return bprio;
+	if (!ioprio_valid(bprio))
+		return aprio;
+
+	if (aclass == IOPRIO_CLASS_NONE)
+		aclass = IOPRIO_CLASS_BE;
+	if (bclass == IOPRIO_CLASS_NONE)
+		bclass = IOPRIO_CLASS_BE;
+
+	if (aclass == bclass)
+		return min(aprio, bprio);
+	if (aclass > bclass)
+		return bprio;
+	else
+		return aprio;
+}
+
+#endif
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 9530b190316..ff48815bd3a 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -608,6 +608,8 @@ struct task_struct {
 	struct list_head run_list;
 	prio_array_t *array;
 
+	unsigned short ioprio;
+
 	unsigned long sleep_avg;
 	unsigned long long timestamp, last_ran;
 	unsigned long long sched_time; /* sched_clock time spent running */
@@ -763,6 +765,7 @@ struct task_struct {
 	nodemask_t mems_allowed;
 	int cpuset_mems_generation;
 #endif
+	atomic_t fs_excl;	/* holding fs exclusive resources */
 };
 
 static inline pid_t process_group(struct task_struct *tsk)
@@ -1112,7 +1115,8 @@ extern void unhash_process(struct task_struct *p);
 
 /*
  * Protects ->fs, ->files, ->mm, ->ptrace, ->group_info, ->comm, keyring
- * subscriptions and synchronises with wait4().  Also used in procfs.
+ * subscriptions and synchronises with wait4().  Also used in procfs.  Also
+ * pins the final release of task.io_context.
  *
  * Nests both inside and outside of read_lock(&tasklist_lock).
  * It must not be nested with write_lock_irq(&tasklist_lock),
diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index 1262cb43c3a..d5c3fe1bf33 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -14,11 +14,13 @@ extern struct list_head inode_unused;
  * Yes, writeback.h requires sched.h
  * No, sched.h is not included from here.
  */
-static inline int current_is_pdflush(void)
+static inline int task_is_pdflush(struct task_struct *task)
 {
-	return current->flags & PF_FLUSHER;
+	return task->flags & PF_FLUSHER;
 }
 
+#define current_is_pdflush()	task_is_pdflush(current)
+
 /*
  * fs/fs-writeback.c
  */
diff --git a/kernel/exit.c b/kernel/exit.c
index 3ebcd60a19c..9d1b10ed013 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -784,6 +784,8 @@ fastcall NORET_TYPE void do_exit(long code)
 
 	profile_task_exit(tsk);
 
+	WARN_ON(atomic_read(&tsk->fs_excl));
+
 	if (unlikely(in_interrupt()))
 		panic("Aiee, killing interrupt handler!");
 	if (unlikely(!tsk->pid))
diff --git a/kernel/fork.c b/kernel/fork.c
index 2c7806873bf..cdef6cea890 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1090,6 +1090,11 @@ static task_t *copy_process(unsigned long clone_flags,
 		spin_unlock(&current->sighand->siglock);
 	}
 
+	/*
+	 * inherit ioprio
+	 */
+	p->ioprio = current->ioprio;
+
 	SET_LINKS(p);
 	if (unlikely(p->ptrace & PT_PTRACED))
 		__ptrace_link(p, current->parent);
diff --git a/kernel/sched.c b/kernel/sched.c
index a07cff90d84..e2b0d3e4dd0 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -3448,15 +3448,7 @@ int task_nice(const task_t *p)
 {
 	return TASK_NICE(p);
 }
-
-/*
- * The only users of task_nice are binfmt_elf and binfmt_elf32.
- * binfmt_elf is no longer modular, but binfmt_elf32 still is.
- * Therefore, task_nice is needed if there is a compat_mode.
- */
-#ifdef CONFIG_COMPAT
 EXPORT_SYMBOL_GPL(task_nice);
-#endif
 
 /**
  * idle_cpu - is a given cpu idle currently?
-- 
cgit v1.2.3-70-g09d2


From 3b18152c327707ae6a2eeba4cfb66457143753bc Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Mon, 27 Jun 2005 10:56:24 +0200
Subject: [PATCH] CFQ io scheduler updates

- Adjust slice values

- Instead of one async queue, one is defined per priority level. This
  prevents kernel threads (such as reiserfs/x and others) that run at
  higher io priority from conflicting with others. Previously, it was a
  coin toss what io prio the async queue got, it was defined by who
  first set up the queue.

- Let a time slice only begin, when the previous slice is completely
  done. Previously we could be somewhat unfair to a new sync slice, if
  the previous slice was async and had several ios queued. This might
  need a little tweaking if throughput suffers a little due to this,
  allowing perhaps an overlap of a single request or so.

- Optimize the calling of kblockd_schedule_work() by doing it only when
  it is strictly necessary (no requests in driver and work left to do).

- Correct sync vs async logic. A 'normal' process can be purely async as
  well, and a flusher can be purely sync as well. Sync or async is now a
  property of the class defined and requests pending. Previously writers
  could be considered sync, when they were really async.

- Get rid of the bit fields in cfqq and crq, use flags instead.

- Various other cleanups and fixes

Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/cfq-iosched.c | 460 ++++++++++++++++++++++++++++----------------
 include/linux/ioprio.h      |   1 +
 2 files changed, 300 insertions(+), 161 deletions(-)

(limited to 'include')

diff --git a/drivers/block/cfq-iosched.c b/drivers/block/cfq-iosched.c
index baa3e268250..1ecb179b860 100644
--- a/drivers/block/cfq-iosched.c
+++ b/drivers/block/cfq-iosched.c
@@ -34,14 +34,15 @@ static int cfq_back_max = 16 * 1024;	/* maximum backwards seek, in KiB */
 static int cfq_back_penalty = 2;	/* penalty of a backwards seek */
 
 static int cfq_slice_sync = HZ / 10;
-static int cfq_slice_async = HZ / 50;
+static int cfq_slice_async = HZ / 25;
 static int cfq_slice_async_rq = 2;
-static int cfq_slice_idle = HZ / 50;
+static int cfq_slice_idle = HZ / 100;
 
 #define CFQ_IDLE_GRACE		(HZ / 10)
 #define CFQ_SLICE_SCALE		(5)
 
 #define CFQ_KEY_ASYNC		(0)
+#define CFQ_KEY_ANY		(0xffff)
 
 /*
  * disable queueing at the driver/hardware level
@@ -96,7 +97,16 @@ static kmem_cache_t *cfq_ioc_pool;
 #define cfq_class_be(cfqq)	((cfqq)->ioprio_class == IOPRIO_CLASS_BE)
 #define cfq_class_rt(cfqq)	((cfqq)->ioprio_class == IOPRIO_CLASS_RT)
 
-#define cfq_cfqq_sync(cfqq)	((cfqq)->key != CFQ_KEY_ASYNC)
+#define ASYNC			(0)
+#define SYNC			(1)
+
+#define cfq_cfqq_dispatched(cfqq)	\
+	((cfqq)->on_dispatch[ASYNC] + (cfqq)->on_dispatch[SYNC])
+
+#define cfq_cfqq_class_sync(cfqq)	((cfqq)->key != CFQ_KEY_ASYNC)
+
+#define cfq_cfqq_sync(cfqq)		\
+	(cfq_cfqq_class_sync(cfqq) || (cfqq)->on_dispatch[SYNC])
 
 /*
  * Per block device queue structure
@@ -200,28 +210,15 @@ struct cfq_queue {
 	unsigned long slice_left;
 	unsigned long service_last;
 
-	/* number of requests that have been handed to the driver */
-	int in_flight;
+	/* number of requests that are on the dispatch list */
+	int on_dispatch[2];
 
 	/* io prio of this group */
 	unsigned short ioprio, org_ioprio;
 	unsigned short ioprio_class, org_ioprio_class;
 
-	/* whether queue is on rr (or empty) list */
-	unsigned on_rr : 1;
-	/* idle slice, waiting for new request submission */
-	unsigned wait_request : 1;
-	/* set when wait_request gets set, reset on first rq alloc */
-	unsigned must_alloc : 1;
-	/* only gets one must_alloc per slice */
-	unsigned must_alloc_slice : 1;
-	/* idle slice, request added, now waiting to dispatch it */
-	unsigned must_dispatch : 1;
-	/* fifo expire per-slice */
-	unsigned fifo_expire : 1;
-
-	unsigned idle_window : 1;
-	unsigned prio_changed : 1;
+	/* various state flags, see below */
+	unsigned int flags;
 };
 
 struct cfq_rq {
@@ -233,15 +230,77 @@ struct cfq_rq {
 	struct cfq_queue *cfq_queue;
 	struct cfq_io_context *io_context;
 
-	unsigned in_flight : 1;
-	unsigned accounted : 1;
-	unsigned is_sync   : 1;
-	unsigned requeued  : 1;
+	unsigned int crq_flags;
 };
 
-static struct cfq_queue *cfq_find_cfq_hash(struct cfq_data *, unsigned int);
+enum cfqq_state_flags {
+	CFQ_CFQQ_FLAG_on_rr = 0,
+	CFQ_CFQQ_FLAG_wait_request,
+	CFQ_CFQQ_FLAG_must_alloc,
+	CFQ_CFQQ_FLAG_must_alloc_slice,
+	CFQ_CFQQ_FLAG_must_dispatch,
+	CFQ_CFQQ_FLAG_fifo_expire,
+	CFQ_CFQQ_FLAG_idle_window,
+	CFQ_CFQQ_FLAG_prio_changed,
+	CFQ_CFQQ_FLAG_expired,
+};
+
+#define CFQ_CFQQ_FNS(name)						\
+static inline void cfq_mark_cfqq_##name(struct cfq_queue *cfqq)		\
+{									\
+	cfqq->flags |= (1 << CFQ_CFQQ_FLAG_##name);			\
+}									\
+static inline void cfq_clear_cfqq_##name(struct cfq_queue *cfqq)	\
+{									\
+	cfqq->flags &= ~(1 << CFQ_CFQQ_FLAG_##name);			\
+}									\
+static inline int cfq_cfqq_##name(const struct cfq_queue *cfqq)		\
+{									\
+	return (cfqq->flags & (1 << CFQ_CFQQ_FLAG_##name)) != 0;	\
+}
+
+CFQ_CFQQ_FNS(on_rr);
+CFQ_CFQQ_FNS(wait_request);
+CFQ_CFQQ_FNS(must_alloc);
+CFQ_CFQQ_FNS(must_alloc_slice);
+CFQ_CFQQ_FNS(must_dispatch);
+CFQ_CFQQ_FNS(fifo_expire);
+CFQ_CFQQ_FNS(idle_window);
+CFQ_CFQQ_FNS(prio_changed);
+CFQ_CFQQ_FNS(expired);
+#undef CFQ_CFQQ_FNS
+
+enum cfq_rq_state_flags {
+	CFQ_CRQ_FLAG_in_flight = 0,
+	CFQ_CRQ_FLAG_in_driver,
+	CFQ_CRQ_FLAG_is_sync,
+	CFQ_CRQ_FLAG_requeued,
+};
+
+#define CFQ_CRQ_FNS(name)						\
+static inline void cfq_mark_crq_##name(struct cfq_rq *crq)		\
+{									\
+	crq->crq_flags |= (1 << CFQ_CRQ_FLAG_##name);			\
+}									\
+static inline void cfq_clear_crq_##name(struct cfq_rq *crq)		\
+{									\
+	crq->crq_flags &= ~(1 << CFQ_CRQ_FLAG_##name);			\
+}									\
+static inline int cfq_crq_##name(const struct cfq_rq *crq)		\
+{									\
+	return (crq->crq_flags & (1 << CFQ_CRQ_FLAG_##name)) != 0;	\
+}
+
+CFQ_CRQ_FNS(in_flight);
+CFQ_CRQ_FNS(in_driver);
+CFQ_CRQ_FNS(is_sync);
+CFQ_CRQ_FNS(requeued);
+#undef CFQ_CRQ_FNS
+
+static struct cfq_queue *cfq_find_cfq_hash(struct cfq_data *, unsigned int, unsigned short);
 static void cfq_dispatch_sort(request_queue_t *, struct cfq_rq *);
 static void cfq_put_cfqd(struct cfq_data *cfqd);
+static inline int cfq_pending_requests(struct cfq_data *cfqd);
 
 #define process_sync(tsk)	((tsk)->flags & PF_SYNCWRITE)
 
@@ -305,9 +364,9 @@ cfq_choose_req(struct cfq_data *cfqd, struct cfq_rq *crq1, struct cfq_rq *crq2)
 		return crq2;
 	if (crq2 == NULL)
 		return crq1;
-	if (crq1->requeued)
+	if (cfq_crq_requeued(crq1))
 		return crq1;
-	if (crq2->requeued)
+	if (cfq_crq_requeued(crq2))
 		return crq2;
 
 	s1 = crq1->request->sector;
@@ -407,7 +466,7 @@ static void cfq_resort_rr_list(struct cfq_queue *cfqq, int preempted)
 	struct cfq_data *cfqd = cfqq->cfqd;
 	struct list_head *list, *entry;
 
-	BUG_ON(!cfqq->on_rr);
+	BUG_ON(!cfq_cfqq_on_rr(cfqq));
 
 	list_del(&cfqq->cfq_list);
 
@@ -423,7 +482,7 @@ static void cfq_resort_rr_list(struct cfq_queue *cfqq, int preempted)
 		 * has lots of io pending vs one that only generates one
 		 * sporadically or synchronously
 		 */
-		if (cfqq->in_flight)
+		if (cfq_cfqq_dispatched(cfqq))
 			list = &cfqd->busy_rr;
 		else
 			list = &cfqd->rr_list[cfqq->ioprio];
@@ -461,8 +520,8 @@ static void cfq_resort_rr_list(struct cfq_queue *cfqq, int preempted)
 static inline void
 cfq_add_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq, int requeue)
 {
-	BUG_ON(cfqq->on_rr);
-	cfqq->on_rr = 1;
+	BUG_ON(cfq_cfqq_on_rr(cfqq));
+	cfq_mark_cfqq_on_rr(cfqq);
 	cfqd->busy_queues++;
 
 	cfq_resort_rr_list(cfqq, requeue);
@@ -471,8 +530,8 @@ cfq_add_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq, int requeue)
 static inline void
 cfq_del_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq)
 {
-	BUG_ON(!cfqq->on_rr);
-	cfqq->on_rr = 0;
+	BUG_ON(!cfq_cfqq_on_rr(cfqq));
+	cfq_clear_cfqq_on_rr(cfqq);
 	list_move(&cfqq->cfq_list, &cfqd->empty_list);
 
 	BUG_ON(!cfqd->busy_queues);
@@ -488,7 +547,7 @@ static inline void cfq_del_crq_rb(struct cfq_rq *crq)
 
 	if (ON_RB(&crq->rb_node)) {
 		struct cfq_data *cfqd = cfqq->cfqd;
-		const int sync = crq->is_sync;
+		const int sync = cfq_crq_is_sync(crq);
 
 		BUG_ON(!cfqq->queued[sync]);
 		cfqq->queued[sync]--;
@@ -498,7 +557,7 @@ static inline void cfq_del_crq_rb(struct cfq_rq *crq)
 		rb_erase(&crq->rb_node, &cfqq->sort_list);
 		RB_CLEAR_COLOR(&crq->rb_node);
 
-		if (cfqq->on_rr && RB_EMPTY(&cfqq->sort_list))
+		if (cfq_cfqq_on_rr(cfqq) && RB_EMPTY(&cfqq->sort_list))
 			cfq_del_cfqq_rr(cfqd, cfqq);
 	}
 }
@@ -534,7 +593,7 @@ static void cfq_add_crq_rb(struct cfq_rq *crq)
 	struct cfq_rq *__alias;
 
 	crq->rb_key = rq_rb_key(rq);
-	cfqq->queued[crq->is_sync]++;
+	cfqq->queued[cfq_crq_is_sync(crq)]++;
 
 	/*
 	 * looks a little odd, but the first insert might return an alias.
@@ -545,8 +604,8 @@ static void cfq_add_crq_rb(struct cfq_rq *crq)
 
 	rb_insert_color(&crq->rb_node, &cfqq->sort_list);
 
-	if (!cfqq->on_rr)
-		cfq_add_cfqq_rr(cfqd, cfqq, crq->requeued);
+	if (!cfq_cfqq_on_rr(cfqq))
+		cfq_add_cfqq_rr(cfqd, cfqq, cfq_crq_requeued(crq));
 
 	/*
 	 * check if this request is a better next-serve candidate
@@ -559,7 +618,7 @@ cfq_reposition_crq_rb(struct cfq_queue *cfqq, struct cfq_rq *crq)
 {
 	if (ON_RB(&crq->rb_node)) {
 		rb_erase(&crq->rb_node, &cfqq->sort_list);
-		cfqq->queued[crq->is_sync]--;
+		cfqq->queued[cfq_crq_is_sync(crq)]--;
 	}
 
 	cfq_add_crq_rb(crq);
@@ -568,7 +627,7 @@ cfq_reposition_crq_rb(struct cfq_queue *cfqq, struct cfq_rq *crq)
 static struct request *cfq_find_rq_rb(struct cfq_data *cfqd, sector_t sector)
 
 {
-	struct cfq_queue *cfqq = cfq_find_cfq_hash(cfqd, current->pid);
+	struct cfq_queue *cfqq = cfq_find_cfq_hash(cfqd, current->pid, CFQ_KEY_ANY);
 	struct rb_node *n;
 
 	if (!cfqq)
@@ -598,17 +657,19 @@ static void cfq_deactivate_request(request_queue_t *q, struct request *rq)
 	if (crq) {
 		struct cfq_queue *cfqq = crq->cfq_queue;
 
-		if (crq->accounted) {
-			crq->accounted = 0;
+		if (cfq_crq_in_driver(crq)) {
+			cfq_clear_crq_in_driver(crq);
 			WARN_ON(!cfqd->rq_in_driver);
 			cfqd->rq_in_driver--;
 		}
-		if (crq->in_flight) {
-			crq->in_flight = 0;
-			WARN_ON(!cfqq->in_flight);
-			cfqq->in_flight--;
+		if (cfq_crq_in_flight(crq)) {
+			const int sync = cfq_crq_is_sync(crq);
+
+			cfq_clear_crq_in_flight(crq);
+			WARN_ON(!cfqq->on_dispatch[sync]);
+			cfqq->on_dispatch[sync]--;
 		}
-		crq->requeued = 1;
+		cfq_mark_crq_requeued(crq);
 	}
 }
 
@@ -712,8 +773,9 @@ __cfq_set_active_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq)
 		cfqq->slice_start = jiffies;
 		cfqq->slice_end = 0;
 		cfqq->slice_left = 0;
-		cfqq->must_alloc_slice = 0;
-		cfqq->fifo_expire = 0;
+		cfq_clear_cfqq_must_alloc_slice(cfqq);
+		cfq_clear_cfqq_fifo_expire(cfqq);
+		cfq_clear_cfqq_expired(cfqq);
 	}
 
 	cfqd->active_queue = cfqq;
@@ -776,9 +838,18 @@ static int cfq_get_next_prio_level(struct cfq_data *cfqd)
 	return prio;
 }
 
-static void cfq_set_active_queue(struct cfq_data *cfqd)
+static struct cfq_queue *cfq_set_active_queue(struct cfq_data *cfqd)
 {
-	struct cfq_queue *cfqq = NULL;
+	struct cfq_queue *cfqq;
+
+	/*
+	 * if current queue is expired but not done with its requests yet,
+	 * wait for that to happen
+	 */
+	if ((cfqq = cfqd->active_queue) != NULL) {
+		if (cfq_cfqq_expired(cfqq) && cfq_cfqq_dispatched(cfqq))
+			return NULL;
+	}
 
 	/*
 	 * if current list is non-empty, grab first entry. if it is empty,
@@ -802,50 +873,66 @@ static void cfq_set_active_queue(struct cfq_data *cfqd)
 	}
 
 	__cfq_set_active_queue(cfqd, cfqq);
+	return cfqq;
 }
 
 /*
  * current cfqq expired its slice (or was too idle), select new one
  */
-static inline void cfq_slice_expired(struct cfq_data *cfqd, int preempted)
+static void
+__cfq_slice_expired(struct cfq_data *cfqd, struct cfq_queue *cfqq,
+		    int preempted)
 {
-	struct cfq_queue *cfqq = cfqd->active_queue;
-
-	if (cfqq) {
-		unsigned long now = jiffies;
+	unsigned long now = jiffies;
 
-		if (cfqq->wait_request)
-			del_timer(&cfqd->idle_slice_timer);
+	if (cfq_cfqq_wait_request(cfqq))
+		del_timer(&cfqd->idle_slice_timer);
 
-		if (!preempted && !cfqq->in_flight)
-			cfqq->service_last = now;
+	if (!preempted && !cfq_cfqq_dispatched(cfqq))
+		cfqq->service_last = now;
 
-		cfqq->must_dispatch = 0;
-		cfqq->wait_request = 0;
+	cfq_clear_cfqq_must_dispatch(cfqq);
+	cfq_clear_cfqq_wait_request(cfqq);
 
-		/*
-		 * store what was left of this slice, if the queue idled out
-		 * or was preempted
-		 */
-		if (time_after(now, cfqq->slice_end))
-			cfqq->slice_left = now - cfqq->slice_end;
-		else
-			cfqq->slice_left = 0;
+	/*
+	 * store what was left of this slice, if the queue idled out
+	 * or was preempted
+	 */
+	if (time_after(now, cfqq->slice_end))
+		cfqq->slice_left = now - cfqq->slice_end;
+	else
+		cfqq->slice_left = 0;
 
-		if (cfqq->on_rr)
-			cfq_resort_rr_list(cfqq, preempted);
+	if (cfq_cfqq_on_rr(cfqq))
+		cfq_resort_rr_list(cfqq, preempted);
 
+	if (cfqq == cfqd->active_queue)
 		cfqd->active_queue = NULL;
 
-		if (cfqd->active_cic) {
-			put_io_context(cfqd->active_cic->ioc);
-			cfqd->active_cic = NULL;
-		}
+	if (cfqd->active_cic) {
+		put_io_context(cfqd->active_cic->ioc);
+		cfqd->active_cic = NULL;
 	}
 
 	cfqd->dispatch_slice = 0;
 }
 
+static inline void cfq_slice_expired(struct cfq_data *cfqd, int preempted)
+{
+	struct cfq_queue *cfqq = cfqd->active_queue;
+
+	if (cfqq) {
+		/*
+		 * use deferred expiry, if there are requests in progress as
+		 * not to disturb the slice of the next queue
+		 */
+		if (cfq_cfqq_dispatched(cfqq))
+			cfq_mark_cfqq_expired(cfqq);
+		else
+			__cfq_slice_expired(cfqd, cfqq, preempted);
+	}
+}
+
 static int cfq_arm_slice_timer(struct cfq_data *cfqd, struct cfq_queue *cfqq)
 
 {
@@ -857,7 +944,7 @@ static int cfq_arm_slice_timer(struct cfq_data *cfqd, struct cfq_queue *cfqq)
 	 */
 	if (!cfqd->cfq_slice_idle)
 		return 0;
-	if (!cfqq->idle_window)
+	if (!cfq_cfqq_idle_window(cfqq))
 		return 0;
 	/*
 	 * task has exited, don't wait
@@ -865,13 +952,13 @@ static int cfq_arm_slice_timer(struct cfq_data *cfqd, struct cfq_queue *cfqq)
 	if (cfqd->active_cic && !cfqd->active_cic->ioc->task)
 		return 0;
 
-	cfqq->wait_request = 1;
-	cfqq->must_alloc = 1;
+	cfq_mark_cfqq_must_dispatch(cfqq);
+	cfq_mark_cfqq_wait_request(cfqq);
 
 	if (!timer_pending(&cfqd->idle_slice_timer)) {
-		unsigned long slice_left = cfqq->slice_end - 1;
+		unsigned long slice_left = min(cfqq->slice_end - 1, (unsigned long) cfqd->cfq_slice_idle);
 
-		cfqd->idle_slice_timer.expires = min(jiffies + cfqd->cfq_slice_idle, slice_left);
+		cfqd->idle_slice_timer.expires = jiffies + slice_left;
 		add_timer(&cfqd->idle_slice_timer);
 	}
 
@@ -901,7 +988,7 @@ static void cfq_dispatch_sort(request_queue_t *q, struct cfq_rq *crq)
 			break;
 		if (!blk_fs_request(__rq))
 			break;
-		if (__crq->requeued)
+		if (cfq_crq_requeued(__crq))
 			break;
 
 		if (__rq->sector <= crq->request->sector)
@@ -920,9 +1007,10 @@ static void cfq_dispatch_sort(request_queue_t *q, struct cfq_rq *crq)
 	cfq_del_crq_rb(crq);
 	cfq_remove_merge_hints(q, crq);
 
-	crq->in_flight = 1;
-	crq->requeued = 0;
-	cfqq->in_flight++;
+	cfq_mark_crq_in_flight(crq);
+	cfq_clear_crq_requeued(crq);
+
+	cfqq->on_dispatch[cfq_crq_is_sync(crq)]++;
 	list_add_tail(&crq->request->queuelist, entry);
 }
 
@@ -935,16 +1023,16 @@ static inline struct cfq_rq *cfq_check_fifo(struct cfq_queue *cfqq)
 	struct request *rq;
 	struct cfq_rq *crq;
 
-	if (cfqq->fifo_expire)
+	if (cfq_cfqq_fifo_expire(cfqq))
 		return NULL;
 
 	if (!list_empty(&cfqq->fifo)) {
-		int fifo = cfq_cfqq_sync(cfqq);
+		int fifo = cfq_cfqq_class_sync(cfqq);
 
 		crq = RQ_DATA(list_entry_fifo(cfqq->fifo.next));
 		rq = crq->request;
 		if (time_after(jiffies, rq->start_time + cfqd->cfq_fifo_expire[fifo])) {
-			cfqq->fifo_expire = 1;
+			cfq_mark_cfqq_fifo_expire(cfqq);
 			return crq;
 		}
 	}
@@ -953,7 +1041,9 @@ static inline struct cfq_rq *cfq_check_fifo(struct cfq_queue *cfqq)
 }
 
 /*
- * Scale schedule slice based on io priority
+ * Scale schedule slice based on io priority. Use the sync time slice only
+ * if a queue is marked sync and has sync io queued. A sync queue with async
+ * io only, should not get full sync slice length.
  */
 static inline int
 cfq_prio_to_slice(struct cfq_data *cfqd, struct cfq_queue *cfqq)
@@ -981,6 +1071,16 @@ cfq_prio_to_maxrq(struct cfq_data *cfqd, struct cfq_queue *cfqq)
 	return 2 * (base_rq + base_rq * (CFQ_PRIO_LISTS - 1 - cfqq->ioprio));
 }
 
+/*
+ * scheduler run of queue, if there are requests pending and no one in the
+ * driver that will restart queueing
+ */
+static inline void cfq_schedule_dispatch(struct cfq_data *cfqd)
+{
+	if (!cfqd->rq_in_driver && cfq_pending_requests(cfqd))
+		kblockd_schedule_work(&cfqd->unplug_work);
+}
+
 /*
  * get next queue for service
  */
@@ -993,11 +1093,14 @@ static struct cfq_queue *cfq_select_queue(struct cfq_data *cfqd, int force)
 	if (!cfqq)
 		goto new_queue;
 
+	if (cfq_cfqq_expired(cfqq))
+		goto new_queue;
+
 	/*
 	 * slice has expired
 	 */
-	if (!cfqq->must_dispatch && time_after(jiffies, cfqq->slice_end))
-		goto new_queue;
+	if (!cfq_cfqq_must_dispatch(cfqq) && time_after(now, cfqq->slice_end))
+		goto expire;
 
 	/*
 	 * if queue has requests, dispatch one. if not, check if
@@ -1005,17 +1108,18 @@ static struct cfq_queue *cfq_select_queue(struct cfq_data *cfqd, int force)
 	 */
 	if (!RB_EMPTY(&cfqq->sort_list))
 		goto keep_queue;
-	else if (!force && cfq_cfqq_sync(cfqq) &&
+	else if (!force && cfq_cfqq_class_sync(cfqq) &&
 		 time_before(now, cfqq->slice_end)) {
 		if (cfq_arm_slice_timer(cfqd, cfqq))
 			return NULL;
 	}
 
-new_queue:
+expire:
 	cfq_slice_expired(cfqd, 0);
-	cfq_set_active_queue(cfqd);
+new_queue:
+	cfqq = cfq_set_active_queue(cfqd);
 keep_queue:
-	return cfqd->active_queue;
+	return cfqq;
 }
 
 static int
@@ -1083,8 +1187,8 @@ cfq_dispatch_requests(request_queue_t *q, int max_dispatch, int force)
 
 	cfqq = cfq_select_queue(cfqd, force);
 	if (cfqq) {
-		cfqq->wait_request = 0;
-		cfqq->must_dispatch = 0;
+		cfq_clear_cfqq_must_dispatch(cfqq);
+		cfq_clear_cfqq_wait_request(cfqq);
 		del_timer(&cfqd->idle_slice_timer);
 
 		if (cfq_class_idle(cfqq))
@@ -1108,10 +1212,10 @@ static inline void cfq_account_dispatch(struct cfq_rq *crq)
 	 * accounted bit is necessary since some drivers will call
 	 * elv_next_request() many times for the same request (eg ide)
 	 */
-	if (crq->accounted)
+	if (cfq_crq_in_driver(crq))
 		return;
 
-	crq->accounted = 1;
+	cfq_mark_crq_in_driver(crq);
 	cfqd->rq_in_driver++;
 }
 
@@ -1121,7 +1225,7 @@ cfq_account_completion(struct cfq_queue *cfqq, struct cfq_rq *crq)
 	struct cfq_data *cfqd = cfqq->cfqd;
 	unsigned long now;
 
-	if (!crq->accounted)
+	if (!cfq_crq_in_driver(crq))
 		return;
 
 	now = jiffies;
@@ -1132,12 +1236,18 @@ cfq_account_completion(struct cfq_queue *cfqq, struct cfq_rq *crq)
 	if (!cfq_class_idle(cfqq))
 		cfqd->last_end_request = now;
 
-	if (!cfqq->in_flight && cfqq->on_rr) {
-		cfqq->service_last = now;
-		cfq_resort_rr_list(cfqq, 0);
+	if (!cfq_cfqq_dispatched(cfqq)) {
+		if (cfq_cfqq_on_rr(cfqq)) {
+			cfqq->service_last = now;
+			cfq_resort_rr_list(cfqq, 0);
+		}
+		if (cfq_cfqq_expired(cfqq)) {
+			__cfq_slice_expired(cfqd, cfqq, 0);
+			cfq_schedule_dispatch(cfqd);
+		}
 	}
 
-	if (crq->is_sync)
+	if (cfq_crq_is_sync(crq))
 		crq->io_context->last_end_request = now;
 }
 
@@ -1153,10 +1263,13 @@ dispatch:
 
 		crq = RQ_DATA(rq);
 		if (crq) {
+			struct cfq_queue *cfqq = crq->cfq_queue;
+
 			/*
 			 * if idle window is disabled, allow queue buildup
 			 */
-			if (!crq->in_flight && !crq->cfq_queue->idle_window &&
+			if (!cfq_crq_in_driver(crq) &&
+			    !cfq_cfqq_idle_window(cfqq) &&
 			    cfqd->rq_in_driver >= cfqd->cfq_max_depth)
 				return NULL;
 
@@ -1190,11 +1303,11 @@ static void cfq_put_queue(struct cfq_queue *cfqq)
 
 	BUG_ON(rb_first(&cfqq->sort_list));
 	BUG_ON(cfqq->allocated[READ] + cfqq->allocated[WRITE]);
-	BUG_ON(cfqq->on_rr);
+	BUG_ON(cfq_cfqq_on_rr(cfqq));
 
 	if (unlikely(cfqd->active_queue == cfqq)) {
-		cfq_slice_expired(cfqd, 0);
-		kblockd_schedule_work(&cfqd->unplug_work);
+		__cfq_slice_expired(cfqd, cfqq, 0);
+		cfq_schedule_dispatch(cfqd);
 	}
 
 	cfq_put_cfqd(cfqq->cfqd);
@@ -1208,15 +1321,17 @@ static void cfq_put_queue(struct cfq_queue *cfqq)
 }
 
 static inline struct cfq_queue *
-__cfq_find_cfq_hash(struct cfq_data *cfqd, unsigned int key, const int hashval)
+__cfq_find_cfq_hash(struct cfq_data *cfqd, unsigned int key, unsigned int prio,
+		    const int hashval)
 {
 	struct hlist_head *hash_list = &cfqd->cfq_hash[hashval];
 	struct hlist_node *entry, *next;
 
 	hlist_for_each_safe(entry, next, hash_list) {
 		struct cfq_queue *__cfqq = list_entry_qhash(entry);
+		const unsigned short __p = IOPRIO_PRIO_VALUE(__cfqq->ioprio_class, __cfqq->ioprio);
 
-		if (__cfqq->key == key)
+		if (__cfqq->key == key && (__p == prio || prio == CFQ_KEY_ANY))
 			return __cfqq;
 	}
 
@@ -1224,9 +1339,9 @@ __cfq_find_cfq_hash(struct cfq_data *cfqd, unsigned int key, const int hashval)
 }
 
 static struct cfq_queue *
-cfq_find_cfq_hash(struct cfq_data *cfqd, unsigned int key)
+cfq_find_cfq_hash(struct cfq_data *cfqd, unsigned int key, unsigned short prio)
 {
-	return __cfq_find_cfq_hash(cfqd, key, hash_long(key, CFQ_QHASH_SHIFT));
+	return __cfq_find_cfq_hash(cfqd, key, prio, hash_long(key, CFQ_QHASH_SHIFT));
 }
 
 static void cfq_free_io_context(struct cfq_io_context *cic)
@@ -1255,8 +1370,8 @@ static void cfq_exit_single_io_context(struct cfq_io_context *cic)
 	spin_lock(q->queue_lock);
 
 	if (unlikely(cic->cfqq == cfqd->active_queue)) {
-		cfq_slice_expired(cfqd, 0);
-		kblockd_schedule_work(&cfqd->unplug_work);
+		__cfq_slice_expired(cfqd, cic->cfqq, 0);
+		cfq_schedule_dispatch(cfqd);
 	}
 
 	cfq_put_queue(cic->cfqq);
@@ -1313,7 +1428,7 @@ static void cfq_init_prio_data(struct cfq_queue *cfqq)
 	struct task_struct *tsk = current;
 	int ioprio_class;
 
-	if (!cfqq->prio_changed)
+	if (!cfq_cfqq_prio_changed(cfqq))
 		return;
 
 	ioprio_class = IOPRIO_PRIO_CLASS(tsk->ioprio);
@@ -1338,7 +1453,7 @@ static void cfq_init_prio_data(struct cfq_queue *cfqq)
 		case IOPRIO_CLASS_IDLE:
 			cfqq->ioprio_class = IOPRIO_CLASS_IDLE;
 			cfqq->ioprio = 7;
-			cfqq->idle_window = 0;
+			cfq_clear_cfqq_idle_window(cfqq);
 			break;
 	}
 
@@ -1349,10 +1464,10 @@ static void cfq_init_prio_data(struct cfq_queue *cfqq)
 	cfqq->org_ioprio = cfqq->ioprio;
 	cfqq->org_ioprio_class = cfqq->ioprio_class;
 
-	if (cfqq->on_rr)
+	if (cfq_cfqq_on_rr(cfqq))
 		cfq_resort_rr_list(cfqq, 0);
 
-	cfqq->prio_changed = 0;
+	cfq_clear_cfqq_prio_changed(cfqq);
 }
 
 static inline void changed_ioprio(struct cfq_queue *cfqq)
@@ -1361,7 +1476,7 @@ static inline void changed_ioprio(struct cfq_queue *cfqq)
 		struct cfq_data *cfqd = cfqq->cfqd;
 
 		spin_lock(cfqd->queue->queue_lock);
-		cfqq->prio_changed = 1;
+		cfq_mark_cfqq_prio_changed(cfqq);
 		cfq_init_prio_data(cfqq);
 		spin_unlock(cfqd->queue->queue_lock);
 	}
@@ -1383,13 +1498,14 @@ static int cfq_ioc_set_ioprio(struct io_context *ioc, unsigned int ioprio)
 }
 
 static struct cfq_queue *
-cfq_get_queue(struct cfq_data *cfqd, unsigned int key, int gfp_mask)
+cfq_get_queue(struct cfq_data *cfqd, unsigned int key, unsigned short ioprio,
+	      int gfp_mask)
 {
 	const int hashval = hash_long(key, CFQ_QHASH_SHIFT);
 	struct cfq_queue *cfqq, *new_cfqq = NULL;
 
 retry:
-	cfqq = __cfq_find_cfq_hash(cfqd, key, hashval);
+	cfqq = __cfq_find_cfq_hash(cfqd, key, ioprio, hashval);
 
 	if (!cfqq) {
 		if (new_cfqq) {
@@ -1423,10 +1539,9 @@ retry:
 		 * set ->slice_left to allow preemption for a new process
 		 */
 		cfqq->slice_left = 2 * cfqd->cfq_slice_idle;
-		cfqq->idle_window = 1;
-		cfqq->ioprio = -1;
-		cfqq->ioprio_class = -1;
-		cfqq->prio_changed = 1;
+		cfq_mark_cfqq_idle_window(cfqq);
+		cfq_mark_cfqq_prio_changed(cfqq);
+		cfq_init_prio_data(cfqq);
 	}
 
 	if (new_cfqq)
@@ -1553,7 +1668,7 @@ static void
 cfq_update_idle_window(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 		       struct cfq_io_context *cic)
 {
-	int enable_idle = cfqq->idle_window;
+	int enable_idle = cfq_cfqq_idle_window(cfqq);
 
 	if (!cic->ioc->task || !cfqd->cfq_slice_idle)
 		enable_idle = 0;
@@ -1564,7 +1679,10 @@ cfq_update_idle_window(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 			enable_idle = 1;
 	}
 
-	cfqq->idle_window = enable_idle;
+	if (enable_idle)
+		cfq_mark_cfqq_idle_window(cfqq);
+	else
+		cfq_clear_cfqq_idle_window(cfqq);
 }
 
 
@@ -1586,14 +1704,14 @@ cfq_should_preempt(struct cfq_data *cfqd, struct cfq_queue *new_cfqq,
 
 	if (cfq_class_idle(cfqq))
 		return 1;
-	if (!new_cfqq->wait_request)
+	if (!cfq_cfqq_wait_request(new_cfqq))
 		return 0;
 	/*
 	 * if it doesn't have slice left, forget it
 	 */
 	if (new_cfqq->slice_left < cfqd->cfq_slice_idle)
 		return 0;
-	if (crq->is_sync && !cfq_cfqq_sync(cfqq))
+	if (cfq_crq_is_sync(crq) && !cfq_cfqq_sync(cfqq))
 		return 1;
 
 	return 0;
@@ -1614,7 +1732,7 @@ static void cfq_preempt_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq)
 		cfqq->slice_left = cfq_prio_to_slice(cfqd, cfqq) / 2;
 
 	cfqq->slice_end = cfqq->slice_left + jiffies;
-	cfq_slice_expired(cfqd, 1);
+	__cfq_slice_expired(cfqd, cfqq, 1);
 	__cfq_set_active_queue(cfqd, cfqq);
 }
 
@@ -1639,7 +1757,7 @@ static void
 cfq_crq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 		 struct cfq_rq *crq)
 {
-	const int sync = crq->is_sync;
+	const int sync = cfq_crq_is_sync(crq);
 
 	cfqq->next_crq = cfq_choose_req(cfqd, cfqq->next_crq, crq);
 
@@ -1658,8 +1776,8 @@ cfq_crq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 		 * immediately and flag that we must not expire this queue
 		 * just now
 		 */
-		if (cfqq->wait_request) {
-			cfqq->must_dispatch = 1;
+		if (cfq_cfqq_wait_request(cfqq)) {
+			cfq_mark_cfqq_must_dispatch(cfqq);
 			del_timer(&cfqd->idle_slice_timer);
 			cfq_start_queueing(cfqd, cfqq);
 		}
@@ -1670,7 +1788,7 @@ cfq_crq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 		 * has some old slice time left and is of higher priority
 		 */
 		cfq_preempt_queue(cfqd, cfqq);
-		cfqq->must_dispatch = 1;
+		cfq_mark_cfqq_must_dispatch(cfqq);
 		cfq_start_queueing(cfqd, cfqq);
 	}
 }
@@ -1713,7 +1831,7 @@ cfq_insert_request(request_queue_t *q, struct request *rq, int where)
 			 * be kicked by __make_request() afterward.
 			 * Kick it here.
 			 */
-			kblockd_schedule_work(&cfqd->unplug_work);
+			cfq_schedule_dispatch(cfqd);
 			break;
 		case ELEVATOR_INSERT_FRONT:
 			list_add(&rq->queuelist, &q->queue_head);
@@ -1750,9 +1868,11 @@ static void cfq_completed_request(request_queue_t *q, struct request *rq)
 
 	cfqq = crq->cfq_queue;
 
-	if (crq->in_flight) {
-		WARN_ON(!cfqq->in_flight);
-		cfqq->in_flight--;
+	if (cfq_crq_in_flight(crq)) {
+		const int sync = cfq_crq_is_sync(crq);
+
+		WARN_ON(!cfqq->on_dispatch[sync]);
+		cfqq->on_dispatch[sync]--;
 	}
 
 	cfq_account_completion(cfqq, crq);
@@ -1814,7 +1934,7 @@ static void cfq_prio_boost(struct cfq_queue *cfqq)
 	 * refile between round-robin lists if we moved the priority class
 	 */
 	if ((ioprio_class != cfqq->ioprio_class || ioprio != cfqq->ioprio) &&
-	    cfqq->on_rr)
+	    cfq_cfqq_on_rr(cfqq))
 		cfq_resort_rr_list(cfqq, 0);
 }
 
@@ -1830,23 +1950,27 @@ static inline int
 __cfq_may_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 		struct task_struct *task, int rw)
 {
-	if (cfqq->wait_request && cfqq->must_alloc)
+#if 1
+	if ((cfq_cfqq_wait_request(cfqq) || cfq_cfqq_must_alloc(cfqq)) &&
+	    !cfq_cfqq_must_alloc_slice) {
+		cfq_mark_cfqq_must_alloc_slice(cfqq);
 		return ELV_MQUEUE_MUST;
+	}
 
 	return ELV_MQUEUE_MAY;
-#if 0
+#else
 	if (!cfqq || task->flags & PF_MEMALLOC)
 		return ELV_MQUEUE_MAY;
-	if (!cfqq->allocated[rw] || cfqq->must_alloc) {
-		if (cfqq->wait_request)
+	if (!cfqq->allocated[rw] || cfq_cfqq_must_alloc(cfqq)) {
+		if (cfq_cfqq_wait_request(cfqq))
 			return ELV_MQUEUE_MUST;
 
 		/*
 		 * only allow 1 ELV_MQUEUE_MUST per slice, otherwise we
 		 * can quickly flood the queue with writes from a single task
 		 */
-		if (rw == READ || !cfqq->must_alloc_slice) {
-			cfqq->must_alloc_slice = 1;
+		if (rw == READ || !cfq_cfqq_must_alloc_slice) {
+			cfq_mark_cfqq_must_alloc_slice(cfqq);
 			return ELV_MQUEUE_MUST;
 		}
 
@@ -1881,7 +2005,7 @@ static int cfq_may_queue(request_queue_t *q, int rw, struct bio *bio)
 	 * so just lookup a possibly existing queue, or return 'may queue'
 	 * if that fails
 	 */
-	cfqq = cfq_find_cfq_hash(cfqd, cfq_queue_pid(tsk, rw));
+	cfqq = cfq_find_cfq_hash(cfqd, cfq_queue_pid(tsk, rw), tsk->ioprio);
 	if (cfqq) {
 		cfq_init_prio_data(cfqq);
 		cfq_prio_boost(cfqq);
@@ -1943,15 +2067,17 @@ cfq_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
 		int gfp_mask)
 {
 	struct cfq_data *cfqd = q->elevator->elevator_data;
+	struct task_struct *tsk = current;
 	struct cfq_io_context *cic;
 	const int rw = rq_data_dir(rq);
+	pid_t key = cfq_queue_pid(tsk, rw);
 	struct cfq_queue *cfqq;
 	struct cfq_rq *crq;
 	unsigned long flags;
 
 	might_sleep_if(gfp_mask & __GFP_WAIT);
 
-	cic = cfq_get_io_context(cfqd, cfq_queue_pid(current, rw), gfp_mask);
+	cic = cfq_get_io_context(cfqd, key, gfp_mask);
 
 	spin_lock_irqsave(q->queue_lock, flags);
 
@@ -1959,7 +2085,7 @@ cfq_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
 		goto queue_fail;
 
 	if (!cic->cfqq) {
-		cfqq = cfq_get_queue(cfqd, current->pid, gfp_mask);
+		cfqq = cfq_get_queue(cfqd, key, tsk->ioprio, gfp_mask);
 		if (!cfqq)
 			goto queue_fail;
 
@@ -1968,7 +2094,7 @@ cfq_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
 		cfqq = cic->cfqq;
 
 	cfqq->allocated[rw]++;
-	cfqq->must_alloc = 0;
+	cfq_clear_cfqq_must_alloc(cfqq);
 	cfqd->rq_starved = 0;
 	atomic_inc(&cfqq->ref);
 	spin_unlock_irqrestore(q->queue_lock, flags);
@@ -1981,9 +2107,15 @@ cfq_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
 		INIT_HLIST_NODE(&crq->hash);
 		crq->cfq_queue = cfqq;
 		crq->io_context = cic;
-		crq->in_flight = crq->accounted = 0;
-		crq->is_sync = (rw == READ || process_sync(current));
-		crq->requeued = 0;
+		cfq_clear_crq_in_flight(crq);
+		cfq_clear_crq_in_driver(crq);
+		cfq_clear_crq_requeued(crq);
+
+		if (rw == READ || process_sync(tsk))
+			cfq_mark_crq_is_sync(crq);
+		else
+			cfq_clear_crq_is_sync(crq);
+
 		rq->elevator_private = crq;
 		return 0;
 	}
@@ -1991,7 +2123,7 @@ cfq_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
 	spin_lock_irqsave(q->queue_lock, flags);
 	cfqq->allocated[rw]--;
 	if (!(cfqq->allocated[0] + cfqq->allocated[1]))
-		cfqq->must_alloc = 1;
+		cfq_mark_cfqq_must_alloc(cfqq);
 	cfq_put_queue(cfqq);
 queue_fail:
 	if (cic)
@@ -2002,7 +2134,7 @@ queue_fail:
 	 * that would be an extremely rare OOM situation
 	 */
 	cfqd->rq_starved = 1;
-	kblockd_schedule_work(&cfqd->unplug_work);
+	cfq_schedule_dispatch(cfqd);
 	spin_unlock_irqrestore(q->queue_lock, flags);
 	return 1;
 }
@@ -2068,15 +2200,14 @@ static void cfq_idle_slice_timer(unsigned long data)
 		 * not expired and it has a request pending, let it dispatch
 		 */
 		if (!RB_EMPTY(&cfqq->sort_list)) {
-			cfqq->must_dispatch = 1;
+			cfq_mark_cfqq_must_dispatch(cfqq);
 			goto out_kick;
 		}
 	}
 expire:
 	cfq_slice_expired(cfqd, 0);
 out_kick:
-	if (cfq_pending_requests(cfqd))
-		kblockd_schedule_work(&cfqd->unplug_work);
+	cfq_schedule_dispatch(cfqd);
 out_cont:
 	spin_unlock_irqrestore(cfqd->queue->queue_lock, flags);
 }
@@ -2099,11 +2230,17 @@ static void cfq_idle_class_timer(unsigned long data)
 		cfqd->idle_class_timer.expires = end;
 		add_timer(&cfqd->idle_class_timer);
 	} else
-		kblockd_schedule_work(&cfqd->unplug_work);
+		cfq_schedule_dispatch(cfqd);
 
 	spin_unlock_irqrestore(cfqd->queue->queue_lock, flags);
 }
 
+static void cfq_shutdown_timer_wq(struct cfq_data *cfqd)
+{
+	del_timer_sync(&cfqd->idle_slice_timer);
+	del_timer_sync(&cfqd->idle_class_timer);
+	blk_sync_queue(cfqd->queue);
+}
 
 static void cfq_put_cfqd(struct cfq_data *cfqd)
 {
@@ -2112,7 +2249,7 @@ static void cfq_put_cfqd(struct cfq_data *cfqd)
 	if (!atomic_dec_and_test(&cfqd->ref))
 		return;
 
-	blk_sync_queue(q);
+	cfq_shutdown_timer_wq(cfqd);
 
 	blk_put_queue(q);
 
@@ -2126,8 +2263,7 @@ static void cfq_exit_queue(elevator_t *e)
 {
 	struct cfq_data *cfqd = e->elevator_data;
 
-	del_timer_sync(&cfqd->idle_slice_timer);
-	del_timer_sync(&cfqd->idle_class_timer);
+	cfq_shutdown_timer_wq(cfqd);
 	cfq_put_cfqd(cfqd);
 }
 
@@ -2198,6 +2334,7 @@ static int cfq_init_queue(request_queue_t *q, elevator_t *e)
 	cfqd->cfq_slice_async_rq = cfq_slice_async_rq;
 	cfqd->cfq_slice_idle = cfq_slice_idle;
 	cfqd->cfq_max_depth = cfq_max_depth;
+
 	return 0;
 out_crqpool:
 	kfree(cfqd->cfq_hash);
@@ -2369,6 +2506,7 @@ static struct cfq_fs_entry cfq_max_depth_entry = {
 	.show = cfq_max_depth_show,
 	.store = cfq_max_depth_store,
 };
+
 static struct attribute *default_attrs[] = {
 	&cfq_quantum_entry.attr,
 	&cfq_queued_entry.attr,
diff --git a/include/linux/ioprio.h b/include/linux/ioprio.h
index 7811300d88e..8a453a0b5e4 100644
--- a/include/linux/ioprio.h
+++ b/include/linux/ioprio.h
@@ -12,6 +12,7 @@
 
 #define IOPRIO_PRIO_CLASS(mask)	((mask) >> IOPRIO_CLASS_SHIFT)
 #define IOPRIO_PRIO_DATA(mask)	((mask) & IOPRIO_PRIO_MASK)
+#define IOPRIO_PRIO_VALUE(class, data)	(((class) << IOPRIO_CLASS_SHIFT) | data)
 
 #define ioprio_valid(mask)	(IOPRIO_PRIO_CLASS((mask)) != IOPRIO_CLASS_NONE)
 
-- 
cgit v1.2.3-70-g09d2


From 4808a1c0261176f9c7e28e7f108d41a381a7d0fc Mon Sep 17 00:00:00 2001
From: Olav Kongas <ok@artecdesign.ee>
Date: Sat, 9 Apr 2005 22:57:39 +0300
Subject: [PATCH] USB: Add isp116x-hcd USB host controller driver

This patch provides an "isp116x-hcd" driver for Philips'
ISP1160/ISP1161 USB host controllers.

The driver:
 - is relatively small, meant for use on embedded platforms.
 - runs usbtests 1-14 without problems for days.
 - has been in use by 6-7 different people on ARM and PPC platforms,
   running a range of devices including USB hubs.
 - supports suspend/resume of both the platform device and the root hub;
   supports remote wakeup of the root hub (but NOT the platform device)
   by USB devices.
 - does NOT support ISO transfers (nobody has asked for them).
 - is PIO-only.

Signed-off-by: Olav Kongas <ok@artecdesign.ee>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/Makefile           |    1 +
 drivers/usb/host/Kconfig       |   13 +
 drivers/usb/host/Makefile      |    1 +
 drivers/usb/host/isp116x-hcd.c | 1882 ++++++++++++++++++++++++++++++++++++++++
 drivers/usb/host/isp116x.h     |  583 +++++++++++++
 include/linux/usb_isp116x.h    |   47 +
 6 files changed, 2527 insertions(+)
 create mode 100644 drivers/usb/host/isp116x-hcd.c
 create mode 100644 drivers/usb/host/isp116x.h
 create mode 100644 include/linux/usb_isp116x.h

(limited to 'include')

diff --git a/drivers/usb/Makefile b/drivers/usb/Makefile
index a61d4433a98..c149c06388b 100644
--- a/drivers/usb/Makefile
+++ b/drivers/usb/Makefile
@@ -9,6 +9,7 @@ obj-$(CONFIG_USB)		+= core/
 obj-$(CONFIG_USB_MON)		+= mon/
 
 obj-$(CONFIG_USB_EHCI_HCD)	+= host/
+obj-$(CONFIG_USB_ISP116X_HCD)	+= host/
 obj-$(CONFIG_USB_OHCI_HCD)	+= host/
 obj-$(CONFIG_USB_UHCI_HCD)	+= host/
 obj-$(CONFIG_USB_SL811_HCD)	+= host/
diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
index 19e598c9641..ed1899d307d 100644
--- a/drivers/usb/host/Kconfig
+++ b/drivers/usb/host/Kconfig
@@ -49,6 +49,19 @@ config USB_EHCI_ROOT_HUB_TT
 
 	  This supports the EHCI implementation from TransDimension Inc.
 
+config USB_ISP116X_HCD
+	tristate "ISP116X HCD support"
+	depends on USB
+	default N
+	---help---
+	  The ISP1160 and ISP1161 chips are USB host controllers. Enable this
+	  option if your board has this chip. If unsure, say N.
+
+	  This driver does not support isochronous transfers.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called isp116x-hcd.
+
 config USB_OHCI_HCD
 	tristate "OHCI HCD support"
 	depends on USB && USB_ARCH_HAS_OHCI
diff --git a/drivers/usb/host/Makefile b/drivers/usb/host/Makefile
index 5dbd3e7a27c..350d14fc1cc 100644
--- a/drivers/usb/host/Makefile
+++ b/drivers/usb/host/Makefile
@@ -4,6 +4,7 @@
 #
 
 obj-$(CONFIG_USB_EHCI_HCD)	+= ehci-hcd.o
+obj-$(CONFIG_USB_ISP116X_HCD)	+= isp116x-hcd.o
 obj-$(CONFIG_USB_OHCI_HCD)	+= ohci-hcd.o
 obj-$(CONFIG_USB_UHCI_HCD)	+= uhci-hcd.o
 obj-$(CONFIG_USB_SL811_HCD)	+= sl811-hcd.o
diff --git a/drivers/usb/host/isp116x-hcd.c b/drivers/usb/host/isp116x-hcd.c
new file mode 100644
index 00000000000..69e7433d9ce
--- /dev/null
+++ b/drivers/usb/host/isp116x-hcd.c
@@ -0,0 +1,1882 @@
+/*
+ * ISP116x HCD (Host Controller Driver) for USB.
+ *
+ * Derived from the SL811 HCD, rewritten for ISP116x.
+ * Copyright (C) 2005 Olav Kongas <ok@artecdesign.ee>
+ *
+ * Portions:
+ * Copyright (C) 2004 Psion Teklogix (for NetBook PRO)
+ * Copyright (C) 2004 David Brownell
+ *
+ * Periodic scheduling is based on Roman's OHCI code
+ * Copyright (C) 1999 Roman Weissgaerber
+ *
+ */
+
+/*
+ * The driver basically works. A number of people have used it with a range
+ * of devices.
+ *
+ *The driver passes all usbtests 1-14.
+ *
+ * Suspending/resuming of root hub via sysfs works. Remote wakeup works too.
+ * And suspending/resuming of platform device works too. Suspend/resume
+ * via HCD operations vector is not implemented.
+ *
+ * Iso transfer support is not implemented. Adding this would include
+ * implementing recovery from the failure to service the processed ITL
+ * fifo ram in time, which will involve chip reset.
+ *
+ * TODO:
+ + More testing of suspend/resume.
+*/
+
+/*
+  ISP116x chips require certain delays between accesses to its
+  registers. The following timing options exist.
+
+  1. Configure your memory controller (the best)
+  2. Implement platform-specific delay function possibly
+  combined with configuring the memory controller; see
+  include/linux/usb-isp116x.h for more info. Some broken
+  memory controllers line LH7A400 SMC need this. Also,
+  uncomment for that to work the following
+  USE_PLATFORM_DELAY macro.
+  3. Use ndelay (easiest, poorest). For that, uncomment
+  the following USE_NDELAY macro.
+*/
+#define USE_PLATFORM_DELAY
+//#define USE_NDELAY
+
+//#define DEBUG
+//#define VERBOSE
+/* Transfer descriptors. See dump_ptd() for printout format  */
+//#define PTD_TRACE
+/* enqueuing/finishing log of urbs */
+//#define URB_TRACE
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/kernel.h>
+#include <linux/delay.h>
+#include <linux/ioport.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/smp_lock.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/list.h>
+#include <linux/interrupt.h>
+#include <linux/usb.h>
+#include <linux/usb_isp116x.h>
+
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/system.h>
+#include <asm/byteorder.h>
+
+#ifndef DEBUG
+#	define	STUB_DEBUG_FILE
+#endif
+
+#include "../core/hcd.h"
+#include "isp116x.h"
+
+#define DRIVER_VERSION	"08 Apr 2005"
+#define DRIVER_DESC	"ISP116x USB Host Controller Driver"
+
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE("GPL");
+
+static const char hcd_name[] = "isp116x-hcd";
+
+/*-----------------------------------------------------------------*/
+
+/*
+  Write len bytes to fifo, pad till 32-bit boundary
+ */
+static void write_ptddata_to_fifo(struct isp116x *isp116x, void *buf, int len)
+{
+	u8 *dp = (u8 *) buf;
+	u16 *dp2 = (u16 *) buf;
+	u16 w;
+	int quot = len % 4;
+
+	if ((unsigned long)dp2 & 1) {
+		/* not aligned */
+		for (; len > 1; len -= 2) {
+			w = *dp++;
+			w |= *dp++ << 8;
+			isp116x_raw_write_data16(isp116x, w);
+		}
+		if (len)
+			isp116x_write_data16(isp116x, (u16) * dp);
+	} else {
+		/* aligned */
+		for (; len > 1; len -= 2)
+			isp116x_raw_write_data16(isp116x, *dp2++);
+		if (len)
+			isp116x_write_data16(isp116x, 0xff & *((u8 *) dp2));
+	}
+	if (quot == 1 || quot == 2)
+		isp116x_raw_write_data16(isp116x, 0);
+}
+
+/*
+  Read len bytes from fifo and then read till 32-bit boundary.
+ */
+static void read_ptddata_from_fifo(struct isp116x *isp116x, void *buf, int len)
+{
+	u8 *dp = (u8 *) buf;
+	u16 *dp2 = (u16 *) buf;
+	u16 w;
+	int quot = len % 4;
+
+	if ((unsigned long)dp2 & 1) {
+		/* not aligned */
+		for (; len > 1; len -= 2) {
+			w = isp116x_raw_read_data16(isp116x);
+			*dp++ = w & 0xff;
+			*dp++ = (w >> 8) & 0xff;
+		}
+		if (len)
+			*dp = 0xff & isp116x_read_data16(isp116x);
+	} else {
+		/* aligned */
+		for (; len > 1; len -= 2)
+			*dp2++ = isp116x_raw_read_data16(isp116x);
+		if (len)
+			*(u8 *) dp2 = 0xff & isp116x_read_data16(isp116x);
+	}
+	if (quot == 1 || quot == 2)
+		isp116x_raw_read_data16(isp116x);
+}
+
+/*
+  Write ptd's and data for scheduled transfers into
+  the fifo ram. Fifo must be empty and ready.
+*/
+static void pack_fifo(struct isp116x *isp116x)
+{
+	struct isp116x_ep *ep;
+	struct ptd *ptd;
+	int buflen = isp116x->atl_last_dir == PTD_DIR_IN
+	    ? isp116x->atl_bufshrt : isp116x->atl_buflen;
+	int ptd_count = 0;
+
+	isp116x_write_reg16(isp116x, HCuPINT, HCuPINT_AIIEOT);
+	isp116x_write_reg16(isp116x, HCXFERCTR, buflen);
+	isp116x_write_addr(isp116x, HCATLPORT | ISP116x_WRITE_OFFSET);
+	for (ep = isp116x->atl_active; ep; ep = ep->active) {
+		++ptd_count;
+		ptd = &ep->ptd;
+		dump_ptd(ptd);
+		dump_ptd_out_data(ptd, ep->data);
+		isp116x_write_data16(isp116x, ptd->count);
+		isp116x_write_data16(isp116x, ptd->mps);
+		isp116x_write_data16(isp116x, ptd->len);
+		isp116x_write_data16(isp116x, ptd->faddr);
+		buflen -= sizeof(struct ptd);
+		/* Skip writing data for last IN PTD */
+		if (ep->active || (isp116x->atl_last_dir != PTD_DIR_IN)) {
+			write_ptddata_to_fifo(isp116x, ep->data, ep->length);
+			buflen -= ALIGN(ep->length, 4);
+		}
+	}
+	BUG_ON(buflen);
+}
+
+/*
+  Read the processed ptd's and data from fifo ram back to
+  URBs' buffers. Fifo must be full and done
+*/
+static void unpack_fifo(struct isp116x *isp116x)
+{
+	struct isp116x_ep *ep;
+	struct ptd *ptd;
+	int buflen = isp116x->atl_last_dir == PTD_DIR_IN
+	    ? isp116x->atl_buflen : isp116x->atl_bufshrt;
+
+	isp116x_write_reg16(isp116x, HCuPINT, HCuPINT_AIIEOT);
+	isp116x_write_reg16(isp116x, HCXFERCTR, buflen);
+	isp116x_write_addr(isp116x, HCATLPORT);
+	for (ep = isp116x->atl_active; ep; ep = ep->active) {
+		ptd = &ep->ptd;
+		ptd->count = isp116x_read_data16(isp116x);
+		ptd->mps = isp116x_read_data16(isp116x);
+		ptd->len = isp116x_read_data16(isp116x);
+		ptd->faddr = isp116x_read_data16(isp116x);
+		buflen -= sizeof(struct ptd);
+		/* Skip reading data for last Setup or Out PTD */
+		if (ep->active || (isp116x->atl_last_dir == PTD_DIR_IN)) {
+			read_ptddata_from_fifo(isp116x, ep->data, ep->length);
+			buflen -= ALIGN(ep->length, 4);
+		}
+		dump_ptd(ptd);
+		dump_ptd_in_data(ptd, ep->data);
+	}
+	BUG_ON(buflen);
+}
+
+/*---------------------------------------------------------------*/
+
+/*
+  Set up PTD's.
+*/
+static void preproc_atl_queue(struct isp116x *isp116x)
+{
+	struct isp116x_ep *ep;
+	struct urb *urb;
+	struct ptd *ptd;
+	u16 toggle, dir, len;
+
+	for (ep = isp116x->atl_active; ep; ep = ep->active) {
+		BUG_ON(list_empty(&ep->hep->urb_list));
+		urb = container_of(ep->hep->urb_list.next,
+				   struct urb, urb_list);
+		ptd = &ep->ptd;
+		len = ep->length;
+		spin_lock(&urb->lock);
+		ep->data = (unsigned char *)urb->transfer_buffer
+		    + urb->actual_length;
+
+		switch (ep->nextpid) {
+		case USB_PID_IN:
+			toggle = usb_gettoggle(urb->dev, ep->epnum, 0);
+			dir = PTD_DIR_IN;
+			break;
+		case USB_PID_OUT:
+			toggle = usb_gettoggle(urb->dev, ep->epnum, 1);
+			dir = PTD_DIR_OUT;
+			break;
+		case USB_PID_SETUP:
+			toggle = 0;
+			dir = PTD_DIR_SETUP;
+			len = sizeof(struct usb_ctrlrequest);
+			ep->data = urb->setup_packet;
+			break;
+		case USB_PID_ACK:
+			toggle = 1;
+			len = 0;
+			dir = (urb->transfer_buffer_length
+			       && usb_pipein(urb->pipe))
+			    ? PTD_DIR_OUT : PTD_DIR_IN;
+			break;
+		default:
+			/* To please gcc */
+			toggle = dir = 0;
+			ERR("%s %d: ep->nextpid %d\n", __func__, __LINE__,
+			    ep->nextpid);
+			BUG_ON(1);
+		}
+
+		ptd->count = PTD_CC_MSK | PTD_ACTIVE_MSK | PTD_TOGGLE(toggle);
+		ptd->mps = PTD_MPS(ep->maxpacket)
+		    | PTD_SPD(urb->dev->speed == USB_SPEED_LOW)
+		    | PTD_EP(ep->epnum);
+		ptd->len = PTD_LEN(len) | PTD_DIR(dir);
+		ptd->faddr = PTD_FA(usb_pipedevice(urb->pipe));
+		spin_unlock(&urb->lock);
+		if (!ep->active) {
+			ptd->mps |= PTD_LAST_MSK;
+			isp116x->atl_last_dir = dir;
+		}
+		isp116x->atl_bufshrt = sizeof(struct ptd) + isp116x->atl_buflen;
+		isp116x->atl_buflen = isp116x->atl_bufshrt + ALIGN(len, 4);
+	}
+}
+
+/*
+  Analyze transfer results, handle partial transfers and errors
+*/
+static void postproc_atl_queue(struct isp116x *isp116x)
+{
+	struct isp116x_ep *ep;
+	struct urb *urb;
+	struct usb_device *udev;
+	struct ptd *ptd;
+	int short_not_ok;
+	u8 cc;
+
+	for (ep = isp116x->atl_active; ep; ep = ep->active) {
+		BUG_ON(list_empty(&ep->hep->urb_list));
+		urb =
+		    container_of(ep->hep->urb_list.next, struct urb, urb_list);
+		udev = urb->dev;
+		ptd = &ep->ptd;
+		cc = PTD_GET_CC(ptd);
+
+		spin_lock(&urb->lock);
+		short_not_ok = 1;
+
+		/* Data underrun is special. For allowed underrun
+		   we clear the error and continue as normal. For
+		   forbidden underrun we finish the DATA stage
+		   immediately while for control transfer,
+		   we do a STATUS stage. */
+		if (cc == TD_DATAUNDERRUN) {
+			if (!(urb->transfer_flags & URB_SHORT_NOT_OK)) {
+				DBG("Allowed data underrun\n");
+				cc = TD_CC_NOERROR;
+				short_not_ok = 0;
+			} else {
+				ep->error_count = 1;
+				if (usb_pipecontrol(urb->pipe))
+					ep->nextpid = USB_PID_ACK;
+				else
+					usb_settoggle(udev, ep->epnum,
+						      ep->nextpid ==
+						      USB_PID_OUT,
+						      PTD_GET_TOGGLE(ptd) ^ 1);
+				urb->status = cc_to_error[TD_DATAUNDERRUN];
+				spin_unlock(&urb->lock);
+				continue;
+			}
+		}
+		/* Keep underrun error through the STATUS stage */
+		if (urb->status == cc_to_error[TD_DATAUNDERRUN])
+			cc = TD_DATAUNDERRUN;
+
+		if (cc != TD_CC_NOERROR && cc != TD_NOTACCESSED
+		    && (++ep->error_count >= 3 || cc == TD_CC_STALL
+			|| cc == TD_DATAOVERRUN)) {
+			if (urb->status == -EINPROGRESS)
+				urb->status = cc_to_error[cc];
+			if (ep->nextpid == USB_PID_ACK)
+				ep->nextpid = 0;
+			spin_unlock(&urb->lock);
+			continue;
+		}
+		/* According to usb spec, zero-length Int transfer signals
+		   finishing of the urb. Hey, does this apply only
+		   for IN endpoints? */
+		if (usb_pipeint(urb->pipe) && !PTD_GET_LEN(ptd)) {
+			if (urb->status == -EINPROGRESS)
+				urb->status = 0;
+			spin_unlock(&urb->lock);
+			continue;
+		}
+
+		/* Relax after previously failed, but later succeeded
+		   or correctly NAK'ed retransmission attempt */
+		if (ep->error_count
+		    && (cc == TD_CC_NOERROR || cc == TD_NOTACCESSED))
+			ep->error_count = 0;
+
+		/* Take into account idiosyncracies of the isp116x chip
+		   regarding toggle bit for failed transfers */
+		if (ep->nextpid == USB_PID_OUT)
+			usb_settoggle(udev, ep->epnum, 1, PTD_GET_TOGGLE(ptd)
+				      ^ (ep->error_count > 0));
+		else if (ep->nextpid == USB_PID_IN)
+			usb_settoggle(udev, ep->epnum, 0, PTD_GET_TOGGLE(ptd)
+				      ^ (ep->error_count > 0));
+
+		switch (ep->nextpid) {
+		case USB_PID_IN:
+		case USB_PID_OUT:
+			urb->actual_length += PTD_GET_COUNT(ptd);
+			if (PTD_GET_ACTIVE(ptd)
+			    || (cc != TD_CC_NOERROR && cc < 0x0E))
+				break;
+			if (urb->transfer_buffer_length != urb->actual_length) {
+				if (short_not_ok)
+					break;
+			} else {
+				if (urb->transfer_flags & URB_ZERO_PACKET
+				    && ep->nextpid == USB_PID_OUT
+				    && !(PTD_GET_COUNT(ptd) % ep->maxpacket)) {
+					DBG("Zero packet requested\n");
+					break;
+				}
+			}
+			/* All data for this URB is transferred, let's finish */
+			if (usb_pipecontrol(urb->pipe))
+				ep->nextpid = USB_PID_ACK;
+			else if (urb->status == -EINPROGRESS)
+				urb->status = 0;
+			break;
+		case USB_PID_SETUP:
+			if (PTD_GET_ACTIVE(ptd)
+			    || (cc != TD_CC_NOERROR && cc < 0x0E))
+				break;
+			if (urb->transfer_buffer_length == urb->actual_length)
+				ep->nextpid = USB_PID_ACK;
+			else if (usb_pipeout(urb->pipe)) {
+				usb_settoggle(udev, 0, 1, 1);
+				ep->nextpid = USB_PID_OUT;
+			} else {
+				usb_settoggle(udev, 0, 0, 1);
+				ep->nextpid = USB_PID_IN;
+			}
+			break;
+		case USB_PID_ACK:
+			if (PTD_GET_ACTIVE(ptd)
+			    || (cc != TD_CC_NOERROR && cc < 0x0E))
+				break;
+			if (urb->status == -EINPROGRESS)
+				urb->status = 0;
+			ep->nextpid = 0;
+			break;
+		default:
+			BUG_ON(1);
+		}
+		spin_unlock(&urb->lock);
+	}
+}
+
+/*
+  Take done or failed requests out of schedule. Give back
+  processed urbs.
+*/
+static void finish_request(struct isp116x *isp116x, struct isp116x_ep *ep,
+			   struct urb *urb, struct pt_regs *regs)
+__releases(isp116x->lock) __acquires(isp116x->lock)
+{
+	unsigned i;
+
+	urb->hcpriv = NULL;
+	ep->error_count = 0;
+
+	if (usb_pipecontrol(urb->pipe))
+		ep->nextpid = USB_PID_SETUP;
+
+	urb_dbg(urb, "Finish");
+
+	spin_unlock(&isp116x->lock);
+	usb_hcd_giveback_urb(isp116x_to_hcd(isp116x), urb, regs);
+	spin_lock(&isp116x->lock);
+
+	/* take idle endpoints out of the schedule */
+	if (!list_empty(&ep->hep->urb_list))
+		return;
+
+	/* async deschedule */
+	if (!list_empty(&ep->schedule)) {
+		list_del_init(&ep->schedule);
+		return;
+	}
+
+	/* periodic deschedule */
+	DBG("deschedule qh%d/%p branch %d\n", ep->period, ep, ep->branch);
+	for (i = ep->branch; i < PERIODIC_SIZE; i += ep->period) {
+		struct isp116x_ep *temp;
+		struct isp116x_ep **prev = &isp116x->periodic[i];
+
+		while (*prev && ((temp = *prev) != ep))
+			prev = &temp->next;
+		if (*prev)
+			*prev = ep->next;
+		isp116x->load[i] -= ep->load;
+	}
+	ep->branch = PERIODIC_SIZE;
+	isp116x_to_hcd(isp116x)->self.bandwidth_allocated -=
+	    ep->load / ep->period;
+
+	/* switch irq type? */
+	if (!--isp116x->periodic_count) {
+		isp116x->irqenb &= ~HCuPINT_SOF;
+		isp116x->irqenb |= HCuPINT_ATL;
+	}
+}
+
+/*
+  Scan transfer lists, schedule transfers, send data off
+  to chip.
+ */
+static void start_atl_transfers(struct isp116x *isp116x)
+{
+	struct isp116x_ep *last_ep = NULL, *ep;
+	struct urb *urb;
+	u16 load = 0;
+	int len, index, speed, byte_time;
+
+	if (atomic_read(&isp116x->atl_finishing))
+		return;
+
+	if (!HC_IS_RUNNING(isp116x_to_hcd(isp116x)->state))
+		return;
+
+	/* FIFO not empty? */
+	if (isp116x_read_reg16(isp116x, HCBUFSTAT) & HCBUFSTAT_ATL_FULL)
+		return;
+
+	isp116x->atl_active = NULL;
+	isp116x->atl_buflen = isp116x->atl_bufshrt = 0;
+
+	/* Schedule int transfers */
+	if (isp116x->periodic_count) {
+		isp116x->fmindex = index =
+		    (isp116x->fmindex + 1) & (PERIODIC_SIZE - 1);
+		if ((load = isp116x->load[index])) {
+			/* Bring all int transfers for this frame
+			   into the active queue */
+			isp116x->atl_active = last_ep =
+			    isp116x->periodic[index];
+			while (last_ep->next)
+				last_ep = (last_ep->active = last_ep->next);
+			last_ep->active = NULL;
+		}
+	}
+
+	/* Schedule control/bulk transfers */
+	list_for_each_entry(ep, &isp116x->async, schedule) {
+		urb = container_of(ep->hep->urb_list.next,
+				   struct urb, urb_list);
+		speed = urb->dev->speed;
+		byte_time = speed == USB_SPEED_LOW
+		    ? BYTE_TIME_LOWSPEED : BYTE_TIME_FULLSPEED;
+
+		if (ep->nextpid == USB_PID_SETUP) {
+			len = sizeof(struct usb_ctrlrequest);
+		} else if (ep->nextpid == USB_PID_ACK) {
+			len = 0;
+		} else {
+			/* Find current free length ... */
+			len = (MAX_LOAD_LIMIT - load) / byte_time;
+
+			/* ... then limit it to configured max size ... */
+			len = min(len, speed == USB_SPEED_LOW ?
+				  MAX_TRANSFER_SIZE_LOWSPEED :
+				  MAX_TRANSFER_SIZE_FULLSPEED);
+
+			/* ... and finally cut to the multiple of MaxPacketSize,
+			   or to the real length if there's enough room. */
+			if (len <
+			    (urb->transfer_buffer_length -
+			     urb->actual_length)) {
+				len -= len % ep->maxpacket;
+				if (!len)
+					continue;
+			} else
+				len = urb->transfer_buffer_length -
+				    urb->actual_length;
+			BUG_ON(len < 0);
+		}
+
+		load += len * byte_time;
+		if (load > MAX_LOAD_LIMIT)
+			break;
+
+		ep->active = NULL;
+		ep->length = len;
+		if (last_ep)
+			last_ep->active = ep;
+		else
+			isp116x->atl_active = ep;
+		last_ep = ep;
+	}
+
+	/* Avoid starving of endpoints */
+	if ((&isp116x->async)->next != (&isp116x->async)->prev)
+		list_move(&isp116x->async, (&isp116x->async)->next);
+
+	if (isp116x->atl_active) {
+		preproc_atl_queue(isp116x);
+		pack_fifo(isp116x);
+	}
+}
+
+/*
+  Finish the processed transfers
+*/
+static void finish_atl_transfers(struct isp116x *isp116x, struct pt_regs *regs)
+{
+	struct isp116x_ep *ep;
+	struct urb *urb;
+
+	if (!isp116x->atl_active)
+		return;
+	/* Fifo not ready? */
+	if (!(isp116x_read_reg16(isp116x, HCBUFSTAT) & HCBUFSTAT_ATL_DONE))
+		return;
+
+	atomic_inc(&isp116x->atl_finishing);
+	unpack_fifo(isp116x);
+	postproc_atl_queue(isp116x);
+	for (ep = isp116x->atl_active; ep; ep = ep->active) {
+		urb =
+		    container_of(ep->hep->urb_list.next, struct urb, urb_list);
+		/* USB_PID_ACK check here avoids finishing of
+		   control transfers, for which TD_DATAUNDERRUN
+		   occured, while URB_SHORT_NOT_OK was set */
+		if (urb && urb->status != -EINPROGRESS
+		    && ep->nextpid != USB_PID_ACK)
+			finish_request(isp116x, ep, urb, regs);
+	}
+	atomic_dec(&isp116x->atl_finishing);
+}
+
+static irqreturn_t isp116x_irq(struct usb_hcd *hcd, struct pt_regs *regs)
+{
+	struct isp116x *isp116x = hcd_to_isp116x(hcd);
+	u16 irqstat;
+	irqreturn_t ret = IRQ_NONE;
+
+	spin_lock(&isp116x->lock);
+	isp116x_write_reg16(isp116x, HCuPINTENB, 0);
+	irqstat = isp116x_read_reg16(isp116x, HCuPINT);
+	isp116x_write_reg16(isp116x, HCuPINT, irqstat);
+
+	if (irqstat & (HCuPINT_ATL | HCuPINT_SOF)) {
+		ret = IRQ_HANDLED;
+		finish_atl_transfers(isp116x, regs);
+	}
+
+	if (irqstat & HCuPINT_OPR) {
+		u32 intstat = isp116x_read_reg32(isp116x, HCINTSTAT);
+		isp116x_write_reg32(isp116x, HCINTSTAT, intstat);
+		if (intstat & HCINT_UE) {
+			ERR("Unrecoverable error\n");
+			/* What should we do here? Reset?  */
+		}
+		if (intstat & HCINT_RHSC) {
+			isp116x->rhstatus =
+			    isp116x_read_reg32(isp116x, HCRHSTATUS);
+			isp116x->rhport[0] =
+			    isp116x_read_reg32(isp116x, HCRHPORT1);
+			isp116x->rhport[1] =
+			    isp116x_read_reg32(isp116x, HCRHPORT2);
+		}
+		if (intstat & HCINT_RD) {
+			DBG("---- remote wakeup\n");
+			schedule_work(&isp116x->rh_resume);
+			ret = IRQ_HANDLED;
+		}
+		irqstat &= ~HCuPINT_OPR;
+		ret = IRQ_HANDLED;
+	}
+
+	if (irqstat & (HCuPINT_ATL | HCuPINT_SOF)) {
+		start_atl_transfers(isp116x);
+	}
+
+	isp116x_write_reg16(isp116x, HCuPINTENB, isp116x->irqenb);
+	spin_unlock(&isp116x->lock);
+	return ret;
+}
+
+/*-----------------------------------------------------------------*/
+
+/* usb 1.1 says max 90% of a frame is available for periodic transfers.
+ * this driver doesn't promise that much since it's got to handle an
+ * IRQ per packet; irq handling latencies also use up that time.
+ */
+
+/* out of 1000 us */
+#define	MAX_PERIODIC_LOAD	600
+static int balance(struct isp116x *isp116x, u16 period, u16 load)
+{
+	int i, branch = -ENOSPC;
+
+	/* search for the least loaded schedule branch of that period
+	   which has enough bandwidth left unreserved. */
+	for (i = 0; i < period; i++) {
+		if (branch < 0 || isp116x->load[branch] > isp116x->load[i]) {
+			int j;
+
+			for (j = i; j < PERIODIC_SIZE; j += period) {
+				if ((isp116x->load[j] + load)
+				    > MAX_PERIODIC_LOAD)
+					break;
+			}
+			if (j < PERIODIC_SIZE)
+				continue;
+			branch = i;
+		}
+	}
+	return branch;
+}
+
+/* NB! ALL the code above this point runs with isp116x->lock
+   held, irqs off
+*/
+
+/*-----------------------------------------------------------------*/
+
+static int isp116x_urb_enqueue(struct usb_hcd *hcd,
+			       struct usb_host_endpoint *hep, struct urb *urb,
+			       int mem_flags)
+{
+	struct isp116x *isp116x = hcd_to_isp116x(hcd);
+	struct usb_device *udev = urb->dev;
+	unsigned int pipe = urb->pipe;
+	int is_out = !usb_pipein(pipe);
+	int type = usb_pipetype(pipe);
+	int epnum = usb_pipeendpoint(pipe);
+	struct isp116x_ep *ep = NULL;
+	unsigned long flags;
+	int i;
+	int ret = 0;
+
+	urb_dbg(urb, "Enqueue");
+
+	if (type == PIPE_ISOCHRONOUS) {
+		ERR("Isochronous transfers not supported\n");
+		urb_dbg(urb, "Refused to enqueue");
+		return -ENXIO;
+	}
+	/* avoid all allocations within spinlocks: request or endpoint */
+	if (!hep->hcpriv) {
+		ep = kcalloc(1, sizeof *ep, (__force unsigned)mem_flags);
+		if (!ep)
+			return -ENOMEM;
+	}
+
+	spin_lock_irqsave(&isp116x->lock, flags);
+	if (!HC_IS_RUNNING(hcd->state)) {
+		ret = -ENODEV;
+		goto fail;
+	}
+
+	if (hep->hcpriv)
+		ep = hep->hcpriv;
+	else {
+		INIT_LIST_HEAD(&ep->schedule);
+		ep->udev = usb_get_dev(udev);
+		ep->epnum = epnum;
+		ep->maxpacket = usb_maxpacket(udev, urb->pipe, is_out);
+		usb_settoggle(udev, epnum, is_out, 0);
+
+		if (type == PIPE_CONTROL) {
+			ep->nextpid = USB_PID_SETUP;
+		} else if (is_out) {
+			ep->nextpid = USB_PID_OUT;
+		} else {
+			ep->nextpid = USB_PID_IN;
+		}
+
+		if (urb->interval) {
+			/*
+			   With INT URBs submitted, the driver works with SOF
+			   interrupt enabled and ATL interrupt disabled. After
+			   the PTDs are written to fifo ram, the chip starts
+			   fifo processing and usb transfers after the next
+			   SOF and continues until the transfers are finished
+			   (succeeded or failed) or the frame ends. Therefore,
+			   the transfers occur only in every second frame,
+			   while fifo reading/writing and data processing
+			   occur in every other second frame. */
+			if (urb->interval < 2)
+				urb->interval = 2;
+			if (urb->interval > 2 * PERIODIC_SIZE)
+				urb->interval = 2 * PERIODIC_SIZE;
+			ep->period = urb->interval >> 1;
+			ep->branch = PERIODIC_SIZE;
+			ep->load = usb_calc_bus_time(udev->speed,
+						     !is_out,
+						     (type == PIPE_ISOCHRONOUS),
+						     usb_maxpacket(udev, pipe,
+								   is_out)) /
+			    1000;
+		}
+		hep->hcpriv = ep;
+		ep->hep = hep;
+	}
+
+	/* maybe put endpoint into schedule */
+	switch (type) {
+	case PIPE_CONTROL:
+	case PIPE_BULK:
+		if (list_empty(&ep->schedule))
+			list_add_tail(&ep->schedule, &isp116x->async);
+		break;
+	case PIPE_INTERRUPT:
+		urb->interval = ep->period;
+		ep->length = min((int)ep->maxpacket,
+				 urb->transfer_buffer_length);
+
+		/* urb submitted for already existing endpoint */
+		if (ep->branch < PERIODIC_SIZE)
+			break;
+
+		ret = ep->branch = balance(isp116x, ep->period, ep->load);
+		if (ret < 0)
+			goto fail;
+		ret = 0;
+
+		urb->start_frame = (isp116x->fmindex & (PERIODIC_SIZE - 1))
+		    + ep->branch;
+
+		/* sort each schedule branch by period (slow before fast)
+		   to share the faster parts of the tree without needing
+		   dummy/placeholder nodes */
+		DBG("schedule qh%d/%p branch %d\n", ep->period, ep, ep->branch);
+		for (i = ep->branch; i < PERIODIC_SIZE; i += ep->period) {
+			struct isp116x_ep **prev = &isp116x->periodic[i];
+			struct isp116x_ep *here = *prev;
+
+			while (here && ep != here) {
+				if (ep->period > here->period)
+					break;
+				prev = &here->next;
+				here = *prev;
+			}
+			if (ep != here) {
+				ep->next = here;
+				*prev = ep;
+			}
+			isp116x->load[i] += ep->load;
+		}
+		hcd->self.bandwidth_allocated += ep->load / ep->period;
+
+		/* switch over to SOFint */
+		if (!isp116x->periodic_count++) {
+			isp116x->irqenb &= ~HCuPINT_ATL;
+			isp116x->irqenb |= HCuPINT_SOF;
+			isp116x_write_reg16(isp116x, HCuPINTENB,
+					    isp116x->irqenb);
+		}
+	}
+
+	/* in case of unlink-during-submit */
+	spin_lock(&urb->lock);
+	if (urb->status != -EINPROGRESS) {
+		spin_unlock(&urb->lock);
+		finish_request(isp116x, ep, urb, NULL);
+		ret = 0;
+		goto fail;
+	}
+	urb->hcpriv = hep;
+	spin_unlock(&urb->lock);
+	start_atl_transfers(isp116x);
+
+      fail:
+	spin_unlock_irqrestore(&isp116x->lock, flags);
+	return ret;
+}
+
+/*
+   Dequeue URBs.
+*/
+static int isp116x_urb_dequeue(struct usb_hcd *hcd, struct urb *urb)
+{
+	struct isp116x *isp116x = hcd_to_isp116x(hcd);
+	struct usb_host_endpoint *hep;
+	struct isp116x_ep *ep, *ep_act;
+	unsigned long flags;
+
+	spin_lock_irqsave(&isp116x->lock, flags);
+	hep = urb->hcpriv;
+	/* URB already unlinked (or never linked)? */
+	if (!hep) {
+		spin_unlock_irqrestore(&isp116x->lock, flags);
+		return 0;
+	}
+	ep = hep->hcpriv;
+	WARN_ON(hep != ep->hep);
+
+	/* In front of queue? */
+	if (ep->hep->urb_list.next == &urb->urb_list)
+		/* active? */
+		for (ep_act = isp116x->atl_active; ep_act;
+		     ep_act = ep_act->active)
+			if (ep_act == ep) {
+				VDBG("dequeue, urb %p active; wait for irq\n",
+				     urb);
+				urb = NULL;
+				break;
+			}
+
+	if (urb)
+		finish_request(isp116x, ep, urb, NULL);
+
+	spin_unlock_irqrestore(&isp116x->lock, flags);
+	return 0;
+}
+
+static void isp116x_endpoint_disable(struct usb_hcd *hcd,
+				     struct usb_host_endpoint *hep)
+{
+	int i;
+	struct isp116x_ep *ep = hep->hcpriv;;
+
+	if (!ep)
+		return;
+
+	/* assume we'd just wait for the irq */
+	for (i = 0; i < 100 && !list_empty(&hep->urb_list); i++)
+		msleep(3);
+	if (!list_empty(&hep->urb_list))
+		WARN("ep %p not empty?\n", ep);
+
+	usb_put_dev(ep->udev);
+	kfree(ep);
+	hep->hcpriv = NULL;
+}
+
+static int isp116x_get_frame(struct usb_hcd *hcd)
+{
+	struct isp116x *isp116x = hcd_to_isp116x(hcd);
+	u32 fmnum;
+	unsigned long flags;
+
+	spin_lock_irqsave(&isp116x->lock, flags);
+	fmnum = isp116x_read_reg32(isp116x, HCFMNUM);
+	spin_unlock_irqrestore(&isp116x->lock, flags);
+	return (int)fmnum;
+}
+
+/*----------------------------------------------------------------*/
+
+/*
+  Adapted from ohci-hub.c. Currently we don't support autosuspend.
+*/
+static int isp116x_hub_status_data(struct usb_hcd *hcd, char *buf)
+{
+	struct isp116x *isp116x = hcd_to_isp116x(hcd);
+	int ports, i, changed = 0;
+
+	if (!HC_IS_RUNNING(hcd->state))
+		return -ESHUTDOWN;
+
+	ports = isp116x->rhdesca & RH_A_NDP;
+
+	/* init status */
+	if (isp116x->rhstatus & (RH_HS_LPSC | RH_HS_OCIC))
+		buf[0] = changed = 1;
+	else
+		buf[0] = 0;
+
+	for (i = 0; i < ports; i++) {
+		u32 status = isp116x->rhport[i];
+
+		if (status & (RH_PS_CSC | RH_PS_PESC | RH_PS_PSSC
+			      | RH_PS_OCIC | RH_PS_PRSC)) {
+			changed = 1;
+			buf[0] |= 1 << (i + 1);
+			continue;
+		}
+	}
+	return changed;
+}
+
+static void isp116x_hub_descriptor(struct isp116x *isp116x,
+				   struct usb_hub_descriptor *desc)
+{
+	u32 reg = isp116x->rhdesca;
+
+	desc->bDescriptorType = 0x29;
+	desc->bDescLength = 9;
+	desc->bHubContrCurrent = 0;
+	desc->bNbrPorts = (u8) (reg & 0x3);
+	/* Power switching, device type, overcurrent. */
+	desc->wHubCharacteristics =
+	    (__force __u16) cpu_to_le16((u16) ((reg >> 8) & 0x1f));
+	desc->bPwrOn2PwrGood = (u8) ((reg >> 24) & 0xff);
+	/* two bitmaps:  ports removable, and legacy PortPwrCtrlMask */
+	desc->bitmap[0] = desc->bNbrPorts == 1 ? 1 << 1 : 3 << 1;
+	desc->bitmap[1] = ~0;
+}
+
+/* Perform reset of a given port.
+   It would be great to just start the reset and let the
+   USB core to clear the reset in due time. However,
+   root hub ports should be reset for at least 50 ms, while
+   our chip stays in reset for about 10 ms. I.e., we must
+   repeatedly reset it ourself here.
+*/
+static inline void root_port_reset(struct isp116x *isp116x, unsigned port)
+{
+	u32 tmp;
+	unsigned long flags, t;
+
+	/* Root hub reset should be 50 ms, but some devices
+	   want it even longer. */
+	t = jiffies + msecs_to_jiffies(100);
+
+	while (time_before(jiffies, t)) {
+		spin_lock_irqsave(&isp116x->lock, flags);
+		/* spin until any current reset finishes */
+		for (;;) {
+			tmp = isp116x_read_reg32(isp116x, port ?
+						 HCRHPORT2 : HCRHPORT1);
+			if (!(tmp & RH_PS_PRS))
+				break;
+			udelay(500);
+		}
+		/* Don't reset a disconnected port */
+		if (!(tmp & RH_PS_CCS)) {
+			spin_unlock_irqrestore(&isp116x->lock, flags);
+			break;
+		}
+		/* Reset lasts 10ms (claims datasheet) */
+		isp116x_write_reg32(isp116x, port ? HCRHPORT2 :
+				    HCRHPORT1, (RH_PS_PRS));
+		spin_unlock_irqrestore(&isp116x->lock, flags);
+		msleep(10);
+	}
+}
+
+/* Adapted from ohci-hub.c */
+static int isp116x_hub_control(struct usb_hcd *hcd,
+			       u16 typeReq,
+			       u16 wValue, u16 wIndex, char *buf, u16 wLength)
+{
+	struct isp116x *isp116x = hcd_to_isp116x(hcd);
+	int ret = 0;
+	unsigned long flags;
+	int ports = isp116x->rhdesca & RH_A_NDP;
+	u32 tmp = 0;
+
+	switch (typeReq) {
+	case ClearHubFeature:
+		DBG("ClearHubFeature: ");
+		switch (wValue) {
+		case C_HUB_OVER_CURRENT:
+			DBG("C_HUB_OVER_CURRENT\n");
+			spin_lock_irqsave(&isp116x->lock, flags);
+			isp116x_write_reg32(isp116x, HCRHSTATUS, RH_HS_OCIC);
+			spin_unlock_irqrestore(&isp116x->lock, flags);
+		case C_HUB_LOCAL_POWER:
+			DBG("C_HUB_LOCAL_POWER\n");
+			break;
+		default:
+			goto error;
+		}
+		break;
+	case SetHubFeature:
+		DBG("SetHubFeature: ");
+		switch (wValue) {
+		case C_HUB_OVER_CURRENT:
+		case C_HUB_LOCAL_POWER:
+			DBG("C_HUB_OVER_CURRENT or C_HUB_LOCAL_POWER\n");
+			break;
+		default:
+			goto error;
+		}
+		break;
+	case GetHubDescriptor:
+		DBG("GetHubDescriptor\n");
+		isp116x_hub_descriptor(isp116x,
+				       (struct usb_hub_descriptor *)buf);
+		break;
+	case GetHubStatus:
+		DBG("GetHubStatus\n");
+		*(__le32 *) buf = cpu_to_le32(0);
+		break;
+	case GetPortStatus:
+		DBG("GetPortStatus\n");
+		if (!wIndex || wIndex > ports)
+			goto error;
+		tmp = isp116x->rhport[--wIndex];
+		*(__le32 *) buf = cpu_to_le32(tmp);
+		DBG("GetPortStatus: port[%d]  %08x\n", wIndex + 1, tmp);
+		break;
+	case ClearPortFeature:
+		DBG("ClearPortFeature: ");
+		if (!wIndex || wIndex > ports)
+			goto error;
+		wIndex--;
+
+		switch (wValue) {
+		case USB_PORT_FEAT_ENABLE:
+			DBG("USB_PORT_FEAT_ENABLE\n");
+			tmp = RH_PS_CCS;
+			break;
+		case USB_PORT_FEAT_C_ENABLE:
+			DBG("USB_PORT_FEAT_C_ENABLE\n");
+			tmp = RH_PS_PESC;
+			break;
+		case USB_PORT_FEAT_SUSPEND:
+			DBG("USB_PORT_FEAT_SUSPEND\n");
+			tmp = RH_PS_POCI;
+			break;
+		case USB_PORT_FEAT_C_SUSPEND:
+			DBG("USB_PORT_FEAT_C_SUSPEND\n");
+			tmp = RH_PS_PSSC;
+			break;
+		case USB_PORT_FEAT_POWER:
+			DBG("USB_PORT_FEAT_POWER\n");
+			tmp = RH_PS_LSDA;
+			break;
+		case USB_PORT_FEAT_C_CONNECTION:
+			DBG("USB_PORT_FEAT_C_CONNECTION\n");
+			tmp = RH_PS_CSC;
+			break;
+		case USB_PORT_FEAT_C_OVER_CURRENT:
+			DBG("USB_PORT_FEAT_C_OVER_CURRENT\n");
+			tmp = RH_PS_OCIC;
+			break;
+		case USB_PORT_FEAT_C_RESET:
+			DBG("USB_PORT_FEAT_C_RESET\n");
+			tmp = RH_PS_PRSC;
+			break;
+		default:
+			goto error;
+		}
+		spin_lock_irqsave(&isp116x->lock, flags);
+		isp116x_write_reg32(isp116x, wIndex
+				    ? HCRHPORT2 : HCRHPORT1, tmp);
+		isp116x->rhport[wIndex] =
+		    isp116x_read_reg32(isp116x, wIndex ? HCRHPORT2 : HCRHPORT1);
+		spin_unlock_irqrestore(&isp116x->lock, flags);
+		break;
+	case SetPortFeature:
+		DBG("SetPortFeature: ");
+		if (!wIndex || wIndex > ports)
+			goto error;
+		wIndex--;
+		switch (wValue) {
+		case USB_PORT_FEAT_SUSPEND:
+			DBG("USB_PORT_FEAT_SUSPEND\n");
+			spin_lock_irqsave(&isp116x->lock, flags);
+			isp116x_write_reg32(isp116x, wIndex
+					    ? HCRHPORT2 : HCRHPORT1, RH_PS_PSS);
+			break;
+		case USB_PORT_FEAT_POWER:
+			DBG("USB_PORT_FEAT_POWER\n");
+			spin_lock_irqsave(&isp116x->lock, flags);
+			isp116x_write_reg32(isp116x, wIndex
+					    ? HCRHPORT2 : HCRHPORT1, RH_PS_PPS);
+			break;
+		case USB_PORT_FEAT_RESET:
+			DBG("USB_PORT_FEAT_RESET\n");
+			root_port_reset(isp116x, wIndex);
+			spin_lock_irqsave(&isp116x->lock, flags);
+			break;
+		default:
+			goto error;
+		}
+		isp116x->rhport[wIndex] =
+		    isp116x_read_reg32(isp116x, wIndex ? HCRHPORT2 : HCRHPORT1);
+		spin_unlock_irqrestore(&isp116x->lock, flags);
+		break;
+
+	default:
+	      error:
+		/* "protocol stall" on error */
+		DBG("PROTOCOL STALL\n");
+		ret = -EPIPE;
+	}
+	return ret;
+}
+
+#ifdef	CONFIG_PM
+
+static int isp116x_hub_suspend(struct usb_hcd *hcd)
+{
+	struct isp116x *isp116x = hcd_to_isp116x(hcd);
+	unsigned long flags;
+	u32 val;
+	int ret = 0;
+
+	spin_lock_irqsave(&isp116x->lock, flags);
+
+	val = isp116x_read_reg32(isp116x, HCCONTROL);
+	switch (val & HCCONTROL_HCFS) {
+	case HCCONTROL_USB_OPER:
+		hcd->state = HC_STATE_QUIESCING;
+		val &= (~HCCONTROL_HCFS & ~HCCONTROL_RWE);
+		val |= HCCONTROL_USB_SUSPEND;
+		if (hcd->remote_wakeup)
+			val |= HCCONTROL_RWE;
+		/* Wait for usb transfers to finish */
+		mdelay(2);
+		isp116x_write_reg32(isp116x, HCCONTROL, val);
+		hcd->state = HC_STATE_SUSPENDED;
+		/* Wait for devices to suspend */
+		mdelay(5);
+	case HCCONTROL_USB_SUSPEND:
+		break;
+	case HCCONTROL_USB_RESUME:
+		isp116x_write_reg32(isp116x, HCCONTROL,
+				    (val & ~HCCONTROL_HCFS) |
+				    HCCONTROL_USB_RESET);
+	case HCCONTROL_USB_RESET:
+		ret = -EBUSY;
+		break;
+	default:
+		ret = -EINVAL;
+	}
+
+	spin_unlock_irqrestore(&isp116x->lock, flags);
+	return ret;
+}
+
+static int isp116x_hub_resume(struct usb_hcd *hcd)
+{
+	struct isp116x *isp116x = hcd_to_isp116x(hcd);
+	u32 val;
+	int ret = -EINPROGRESS;
+
+	msleep(5);
+	spin_lock_irq(&isp116x->lock);
+
+	val = isp116x_read_reg32(isp116x, HCCONTROL);
+	switch (val & HCCONTROL_HCFS) {
+	case HCCONTROL_USB_SUSPEND:
+		val &= ~HCCONTROL_HCFS;
+		val |= HCCONTROL_USB_RESUME;
+		isp116x_write_reg32(isp116x, HCCONTROL, val);
+	case HCCONTROL_USB_RESUME:
+		break;
+	case HCCONTROL_USB_OPER:
+		/* Without setting power_state here the
+		   SUSPENDED state won't be removed from
+		   sysfs/usbN/power.state as a response to remote
+		   wakeup. Maybe in the future. */
+		hcd->self.root_hub->dev.power.power_state = PMSG_ON;
+		ret = 0;
+		break;
+	default:
+		ret = -EBUSY;
+	}
+
+	if (ret != -EINPROGRESS) {
+		spin_unlock_irq(&isp116x->lock);
+		return ret;
+	}
+
+	val = isp116x->rhdesca & RH_A_NDP;
+	while (val--) {
+		u32 stat =
+		    isp116x_read_reg32(isp116x, val ? HCRHPORT2 : HCRHPORT1);
+		/* force global, not selective, resume */
+		if (!(stat & RH_PS_PSS))
+			continue;
+		DBG("%s: Resuming port %d\n", __func__, val);
+		isp116x_write_reg32(isp116x, RH_PS_POCI, val
+				    ? HCRHPORT2 : HCRHPORT1);
+	}
+	spin_unlock_irq(&isp116x->lock);
+
+	hcd->state = HC_STATE_RESUMING;
+	mdelay(20);
+
+	/* Go operational */
+	spin_lock_irq(&isp116x->lock);
+	val = isp116x_read_reg32(isp116x, HCCONTROL);
+	isp116x_write_reg32(isp116x, HCCONTROL,
+			    (val & ~HCCONTROL_HCFS) | HCCONTROL_USB_OPER);
+	spin_unlock_irq(&isp116x->lock);
+	/* see analogous comment above */
+	hcd->self.root_hub->dev.power.power_state = PMSG_ON;
+	hcd->state = HC_STATE_RUNNING;
+
+	return 0;
+}
+
+static void isp116x_rh_resume(void *_hcd)
+{
+	struct usb_hcd *hcd = _hcd;
+
+	usb_resume_device(hcd->self.root_hub);
+}
+
+#else
+
+#define	isp116x_hub_suspend	NULL
+#define	isp116x_hub_resume	NULL
+
+static void isp116x_rh_resume(void *_hcd)
+{
+}
+
+#endif
+
+/*-----------------------------------------------------------------*/
+
+#ifdef STUB_DEBUG_FILE
+
+static inline void create_debug_file(struct isp116x *isp116x)
+{
+}
+
+static inline void remove_debug_file(struct isp116x *isp116x)
+{
+}
+
+#else
+
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+
+static void dump_irq(struct seq_file *s, char *label, u16 mask)
+{
+	seq_printf(s, "%s %04x%s%s%s%s%s%s\n", label, mask,
+		   mask & HCuPINT_CLKRDY ? " clkrdy" : "",
+		   mask & HCuPINT_SUSP ? " susp" : "",
+		   mask & HCuPINT_OPR ? " opr" : "",
+		   mask & HCuPINT_AIIEOT ? " eot" : "",
+		   mask & HCuPINT_ATL ? " atl" : "",
+		   mask & HCuPINT_SOF ? " sof" : "");
+}
+
+static void dump_int(struct seq_file *s, char *label, u32 mask)
+{
+	seq_printf(s, "%s %08x%s%s%s%s%s%s%s\n", label, mask,
+		   mask & HCINT_MIE ? " MIE" : "",
+		   mask & HCINT_RHSC ? " rhsc" : "",
+		   mask & HCINT_FNO ? " fno" : "",
+		   mask & HCINT_UE ? " ue" : "",
+		   mask & HCINT_RD ? " rd" : "",
+		   mask & HCINT_SF ? " sof" : "", mask & HCINT_SO ? " so" : "");
+}
+
+static int proc_isp116x_show(struct seq_file *s, void *unused)
+{
+	struct isp116x *isp116x = s->private;
+	struct isp116x_ep *ep;
+	struct urb *urb;
+	unsigned i;
+	char *str;
+
+	seq_printf(s, "%s\n%s version %s\n",
+		   isp116x_to_hcd(isp116x)->product_desc, hcd_name,
+		   DRIVER_VERSION);
+
+	if (HC_IS_SUSPENDED(isp116x_to_hcd(isp116x)->state)) {
+		seq_printf(s, "HCD is suspended\n");
+		return 0;
+	}
+	if (!HC_IS_RUNNING(isp116x_to_hcd(isp116x)->state)) {
+		seq_printf(s, "HCD not running\n");
+		return 0;
+	}
+
+	spin_lock_irq(&isp116x->lock);
+
+	dump_irq(s, "hc_irq_enable", isp116x_read_reg16(isp116x, HCuPINTENB));
+	dump_irq(s, "hc_irq_status", isp116x_read_reg16(isp116x, HCuPINT));
+	dump_int(s, "hc_int_enable", isp116x_read_reg32(isp116x, HCINTENB));
+	dump_int(s, "hc_int_status", isp116x_read_reg32(isp116x, HCINTSTAT));
+
+	list_for_each_entry(ep, &isp116x->async, schedule) {
+
+		switch (ep->nextpid) {
+		case USB_PID_IN:
+			str = "in";
+			break;
+		case USB_PID_OUT:
+			str = "out";
+			break;
+		case USB_PID_SETUP:
+			str = "setup";
+			break;
+		case USB_PID_ACK:
+			str = "status";
+			break;
+		default:
+			str = "?";
+			break;
+		};
+		seq_printf(s, "%p, ep%d%s, maxpacket %d:\n", ep,
+			   ep->epnum, str, ep->maxpacket);
+		list_for_each_entry(urb, &ep->hep->urb_list, urb_list) {
+			seq_printf(s, "  urb%p, %d/%d\n", urb,
+				   urb->actual_length,
+				   urb->transfer_buffer_length);
+		}
+	}
+	if (!list_empty(&isp116x->async))
+		seq_printf(s, "\n");
+
+	seq_printf(s, "periodic size= %d\n", PERIODIC_SIZE);
+
+	for (i = 0; i < PERIODIC_SIZE; i++) {
+		ep = isp116x->periodic[i];
+		if (!ep)
+			continue;
+		seq_printf(s, "%2d [%3d]:\n", i, isp116x->load[i]);
+
+		/* DUMB: prints shared entries multiple times */
+		do {
+			seq_printf(s, "   %d/%p (%sdev%d ep%d%s max %d)\n",
+				   ep->period, ep,
+				   (ep->udev->speed ==
+				    USB_SPEED_FULL) ? "" : "ls ",
+				   ep->udev->devnum, ep->epnum,
+				   (ep->epnum ==
+				    0) ? "" : ((ep->nextpid ==
+						USB_PID_IN) ? "in" : "out"),
+				   ep->maxpacket);
+			ep = ep->next;
+		} while (ep);
+	}
+	spin_unlock_irq(&isp116x->lock);
+	seq_printf(s, "\n");
+
+	return 0;
+}
+
+static int proc_isp116x_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, proc_isp116x_show, PDE(inode)->data);
+}
+
+static struct file_operations proc_ops = {
+	.open = proc_isp116x_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
+
+/* expect just one isp116x per system */
+static const char proc_filename[] = "driver/isp116x";
+
+static void create_debug_file(struct isp116x *isp116x)
+{
+	struct proc_dir_entry *pde;
+
+	pde = create_proc_entry(proc_filename, 0, NULL);
+	if (pde == NULL)
+		return;
+
+	pde->proc_fops = &proc_ops;
+	pde->data = isp116x;
+	isp116x->pde = pde;
+}
+
+static void remove_debug_file(struct isp116x *isp116x)
+{
+	if (isp116x->pde)
+		remove_proc_entry(proc_filename, NULL);
+}
+
+#endif
+
+/*-----------------------------------------------------------------*/
+
+/*
+  Software reset - can be called from any contect.
+*/
+static int isp116x_sw_reset(struct isp116x *isp116x)
+{
+	int retries = 15;
+	unsigned long flags;
+	int ret = 0;
+
+	spin_lock_irqsave(&isp116x->lock, flags);
+	isp116x_write_reg16(isp116x, HCSWRES, HCSWRES_MAGIC);
+	isp116x_write_reg32(isp116x, HCCMDSTAT, HCCMDSTAT_HCR);
+	while (--retries) {
+		/* It usually resets within 1 ms */
+		mdelay(1);
+		if (!(isp116x_read_reg32(isp116x, HCCMDSTAT) & HCCMDSTAT_HCR))
+			break;
+	}
+	if (!retries) {
+		ERR("Software reset timeout\n");
+		ret = -ETIME;
+	}
+	spin_unlock_irqrestore(&isp116x->lock, flags);
+	return ret;
+}
+
+/*
+  Reset. Tries to perform platform-specific hardware
+  reset first; falls back to software reset.
+*/
+static int isp116x_reset(struct usb_hcd *hcd)
+{
+	struct isp116x *isp116x = hcd_to_isp116x(hcd);
+	unsigned long t;
+	u16 clkrdy = 0;
+	int ret = 0, timeout = 15 /* ms */ ;
+
+	if (isp116x->board && isp116x->board->reset) {
+		/* Hardware reset */
+		isp116x->board->reset(hcd->self.controller, 1);
+		msleep(10);
+		if (isp116x->board->clock)
+			isp116x->board->clock(hcd->self.controller, 1);
+		msleep(1);
+		isp116x->board->reset(hcd->self.controller, 0);
+	} else
+		ret = isp116x_sw_reset(isp116x);
+
+	if (ret)
+		return ret;
+
+	t = jiffies + msecs_to_jiffies(timeout);
+	while (time_before_eq(jiffies, t)) {
+		msleep(4);
+		spin_lock_irq(&isp116x->lock);
+		clkrdy = isp116x_read_reg16(isp116x, HCuPINT) & HCuPINT_CLKRDY;
+		spin_unlock_irq(&isp116x->lock);
+		if (clkrdy)
+			break;
+	}
+	if (!clkrdy) {
+		ERR("Clock not ready after 20ms\n");
+		ret = -ENODEV;
+	}
+	return ret;
+}
+
+static void isp116x_stop(struct usb_hcd *hcd)
+{
+	struct isp116x *isp116x = hcd_to_isp116x(hcd);
+	unsigned long flags;
+	u32 val;
+
+	spin_lock_irqsave(&isp116x->lock, flags);
+	isp116x_write_reg16(isp116x, HCuPINTENB, 0);
+
+	/* Switch off ports' power, some devices don't come up
+	   after next 'insmod' without this */
+	val = isp116x_read_reg32(isp116x, HCRHDESCA);
+	val &= ~(RH_A_NPS | RH_A_PSM);
+	isp116x_write_reg32(isp116x, HCRHDESCA, val);
+	isp116x_write_reg32(isp116x, HCRHSTATUS, RH_HS_LPS);
+	spin_unlock_irqrestore(&isp116x->lock, flags);
+
+	/* Put the chip into reset state */
+	if (isp116x->board && isp116x->board->reset)
+		isp116x->board->reset(hcd->self.controller, 0);
+	else
+		isp116x_sw_reset(isp116x);
+
+	/* Stop the clock */
+	if (isp116x->board && isp116x->board->clock)
+		isp116x->board->clock(hcd->self.controller, 0);
+}
+
+/*
+  Configure the chip. The chip must be successfully reset by now.
+*/
+static int isp116x_start(struct usb_hcd *hcd)
+{
+	struct isp116x *isp116x = hcd_to_isp116x(hcd);
+	struct isp116x_platform_data *board = isp116x->board;
+	struct usb_device *udev;
+	u32 val;
+	unsigned long flags;
+
+	spin_lock_irqsave(&isp116x->lock, flags);
+
+	/* clear interrupt status and disable all interrupt sources */
+	isp116x_write_reg16(isp116x, HCuPINT, 0xff);
+	isp116x_write_reg16(isp116x, HCuPINTENB, 0);
+
+	val = isp116x_read_reg16(isp116x, HCCHIPID);
+	if ((val & HCCHIPID_MASK) != HCCHIPID_MAGIC) {
+		ERR("Invalid chip ID %04x\n", val);
+		spin_unlock_irqrestore(&isp116x->lock, flags);
+		return -ENODEV;
+	}
+
+	isp116x_write_reg16(isp116x, HCITLBUFLEN, ISP116x_ITL_BUFSIZE);
+	isp116x_write_reg16(isp116x, HCATLBUFLEN, ISP116x_ATL_BUFSIZE);
+
+	/* ----- HW conf */
+	val = HCHWCFG_INT_ENABLE | HCHWCFG_DBWIDTH(1);
+	if (board->sel15Kres)
+		val |= HCHWCFG_15KRSEL;
+	/* Remote wakeup won't work without working clock */
+	if (board->clknotstop || board->remote_wakeup_enable)
+		val |= HCHWCFG_CLKNOTSTOP;
+	if (board->oc_enable)
+		val |= HCHWCFG_ANALOG_OC;
+	if (board->int_act_high)
+		val |= HCHWCFG_INT_POL;
+	if (board->int_edge_triggered)
+		val |= HCHWCFG_INT_TRIGGER;
+	isp116x_write_reg16(isp116x, HCHWCFG, val);
+
+	/* ----- Root hub conf */
+	val = 0;
+	/* AN10003_1.pdf recommends NPS to be always 1 */
+	if (board->no_power_switching)
+		val |= RH_A_NPS;
+	if (board->power_switching_mode)
+		val |= RH_A_PSM;
+	if (board->potpg)
+		val |= (board->potpg << 24) & RH_A_POTPGT;
+	else
+		val |= (25 << 24) & RH_A_POTPGT;
+	isp116x_write_reg32(isp116x, HCRHDESCA, val);
+	isp116x->rhdesca = isp116x_read_reg32(isp116x, HCRHDESCA);
+
+	val = RH_B_PPCM;
+	isp116x_write_reg32(isp116x, HCRHDESCB, val);
+	isp116x->rhdescb = isp116x_read_reg32(isp116x, HCRHDESCB);
+
+	val = 0;
+	if (board->remote_wakeup_enable) {
+		hcd->can_wakeup = 1;
+		val |= RH_HS_DRWE;
+	}
+	isp116x_write_reg32(isp116x, HCRHSTATUS, val);
+	isp116x->rhstatus = isp116x_read_reg32(isp116x, HCRHSTATUS);
+
+	isp116x_write_reg32(isp116x, HCFMINTVL, 0x27782edf);
+	spin_unlock_irqrestore(&isp116x->lock, flags);
+
+	udev = usb_alloc_dev(NULL, &hcd->self, 0);
+	if (!udev) {
+		isp116x_stop(hcd);
+		return -ENOMEM;
+	}
+
+	udev->speed = USB_SPEED_FULL;
+	hcd->state = HC_STATE_RUNNING;
+
+	if (usb_hcd_register_root_hub(udev, hcd) != 0) {
+		isp116x_stop(hcd);
+		usb_put_dev(udev);
+		return -ENODEV;
+	}
+
+	spin_lock_irqsave(&isp116x->lock, flags);
+	/* Set up interrupts */
+	isp116x->intenb = HCINT_MIE | HCINT_RHSC | HCINT_UE;
+	if (board->remote_wakeup_enable)
+		isp116x->intenb |= HCINT_RD;
+	isp116x->irqenb = HCuPINT_ATL | HCuPINT_OPR;	/* | HCuPINT_SUSP; */
+	isp116x_write_reg32(isp116x, HCINTENB, isp116x->intenb);
+	isp116x_write_reg16(isp116x, HCuPINTENB, isp116x->irqenb);
+
+	/* Go operational */
+	val = HCCONTROL_USB_OPER;
+	/* Remote wakeup connected - NOT SUPPORTED */
+	/*  if (board->remote_wakeup_connected)
+	   val |= HCCONTROL_RWC;  */
+	if (board->remote_wakeup_enable)
+		val |= HCCONTROL_RWE;
+	isp116x_write_reg32(isp116x, HCCONTROL, val);
+
+	/* Disable ports to avoid race in device enumeration */
+	isp116x_write_reg32(isp116x, HCRHPORT1, RH_PS_CCS);
+	isp116x_write_reg32(isp116x, HCRHPORT2, RH_PS_CCS);
+
+	isp116x_show_regs(isp116x);
+	spin_unlock_irqrestore(&isp116x->lock, flags);
+	return 0;
+}
+
+/*-----------------------------------------------------------------*/
+
+static struct hc_driver isp116x_hc_driver = {
+	.description = hcd_name,
+	.product_desc = "ISP116x Host Controller",
+	.hcd_priv_size = sizeof(struct isp116x),
+
+	.irq = isp116x_irq,
+	.flags = HCD_USB11,
+
+	.reset = isp116x_reset,
+	.start = isp116x_start,
+	.stop = isp116x_stop,
+
+	.urb_enqueue = isp116x_urb_enqueue,
+	.urb_dequeue = isp116x_urb_dequeue,
+	.endpoint_disable = isp116x_endpoint_disable,
+
+	.get_frame_number = isp116x_get_frame,
+
+	.hub_status_data = isp116x_hub_status_data,
+	.hub_control = isp116x_hub_control,
+	.hub_suspend = isp116x_hub_suspend,
+	.hub_resume = isp116x_hub_resume,
+};
+
+/*----------------------------------------------------------------*/
+
+static int __init_or_module isp116x_remove(struct device *dev)
+{
+	struct usb_hcd *hcd = dev_get_drvdata(dev);
+	struct isp116x *isp116x = hcd_to_isp116x(hcd);
+	struct platform_device *pdev;
+	struct resource *res;
+
+	pdev = container_of(dev, struct platform_device, dev);
+	remove_debug_file(isp116x);
+	usb_remove_hcd(hcd);
+
+	iounmap(isp116x->data_reg);
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	release_mem_region(res->start, 2);
+	iounmap(isp116x->addr_reg);
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	release_mem_region(res->start, 2);
+
+	usb_put_hcd(hcd);
+	return 0;
+}
+
+#define resource_len(r) (((r)->end - (r)->start) + 1)
+
+static int __init isp116x_probe(struct device *dev)
+{
+	struct usb_hcd *hcd;
+	struct isp116x *isp116x;
+	struct platform_device *pdev;
+	struct resource *addr, *data;
+	void __iomem *addr_reg;
+	void __iomem *data_reg;
+	int irq;
+	int ret = 0;
+
+	pdev = container_of(dev, struct platform_device, dev);
+	if (pdev->num_resources < 3) {
+		ret = -ENODEV;
+		goto err1;
+	}
+
+	data = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	addr = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	irq = platform_get_irq(pdev, 0);
+	if (!addr || !data || irq < 0) {
+		ret = -ENODEV;
+		goto err1;
+	}
+
+	if (dev->dma_mask) {
+		DBG("DMA not supported\n");
+		ret = -EINVAL;
+		goto err1;
+	}
+
+	if (!request_mem_region(addr->start, 2, hcd_name)) {
+		ret = -EBUSY;
+		goto err1;
+	}
+	addr_reg = ioremap(addr->start, resource_len(addr));
+	if (addr_reg == NULL) {
+		ret = -ENOMEM;
+		goto err2;
+	}
+	if (!request_mem_region(data->start, 2, hcd_name)) {
+		ret = -EBUSY;
+		goto err3;
+	}
+	data_reg = ioremap(data->start, resource_len(data));
+	if (data_reg == NULL) {
+		ret = -ENOMEM;
+		goto err4;
+	}
+
+	/* allocate and initialize hcd */
+	hcd = usb_create_hcd(&isp116x_hc_driver, dev, dev->bus_id);
+	if (!hcd) {
+		ret = -ENOMEM;
+		goto err5;
+	}
+	/* this rsrc_start is bogus */
+	hcd->rsrc_start = addr->start;
+	isp116x = hcd_to_isp116x(hcd);
+	isp116x->data_reg = data_reg;
+	isp116x->addr_reg = addr_reg;
+	spin_lock_init(&isp116x->lock);
+	INIT_LIST_HEAD(&isp116x->async);
+	INIT_WORK(&isp116x->rh_resume, isp116x_rh_resume, hcd);
+	isp116x->board = dev->platform_data;
+
+	if (!isp116x->board) {
+		ERR("Platform data structure not initialized\n");
+		ret = -ENODEV;
+		goto err6;
+	}
+	if (isp116x_check_platform_delay(isp116x)) {
+		ERR("USE_PLATFORM_DELAY defined, but delay function not "
+		    "implemented.\n");
+		ERR("See comments in drivers/usb/host/isp116x-hcd.c\n");
+		ret = -ENODEV;
+		goto err6;
+	}
+
+	ret = usb_add_hcd(hcd, irq, SA_INTERRUPT);
+	if (ret != 0)
+		goto err6;
+
+	create_debug_file(isp116x);
+	return 0;
+
+      err6:
+	usb_put_hcd(hcd);
+      err5:
+	iounmap(data_reg);
+      err4:
+	release_mem_region(data->start, 2);
+      err3:
+	iounmap(addr_reg);
+      err2:
+	release_mem_region(addr->start, 2);
+      err1:
+	ERR("init error, %d\n", ret);
+	return ret;
+}
+
+#ifdef	CONFIG_PM
+/*
+  Suspend of platform device
+*/
+static int isp116x_suspend(struct device *dev, pm_message_t state, u32 phase)
+{
+	int ret = 0;
+	struct usb_hcd *hcd = dev_get_drvdata(dev);
+
+	VDBG("%s: state %x, phase %x\n", __func__, state, phase);
+
+	if (phase != SUSPEND_DISABLE && phase != SUSPEND_POWER_DOWN)
+		return 0;
+
+	ret = usb_suspend_device(hcd->self.root_hub, state);
+	if (!ret) {
+		dev->power.power_state = state;
+		INFO("%s suspended\n", (char *)hcd_name);
+	} else
+		ERR("%s suspend failed\n", (char *)hcd_name);
+
+	return ret;
+}
+
+/*
+  Resume platform device
+*/
+static int isp116x_resume(struct device *dev, u32 phase)
+{
+	int ret = 0;
+	struct usb_hcd *hcd = dev_get_drvdata(dev);
+
+	VDBG("%s:  state %x, phase %x\n", __func__, dev->power.power_state,
+	     phase);
+	if (phase != RESUME_POWER_ON)
+		return 0;
+
+	ret = usb_resume_device(hcd->self.root_hub);
+	if (!ret) {
+		dev->power.power_state = PMSG_ON;
+		VDBG("%s resumed\n", (char *)hcd_name);
+	}
+	return ret;
+}
+
+#else
+
+#define	isp116x_suspend    NULL
+#define	isp116x_resume     NULL
+
+#endif
+
+static struct device_driver isp116x_driver = {
+	.name = (char *)hcd_name,
+	.bus = &platform_bus_type,
+	.probe = isp116x_probe,
+	.remove = isp116x_remove,
+	.suspend = isp116x_suspend,
+	.resume = isp116x_resume,
+};
+
+/*-----------------------------------------------------------------*/
+
+static int __init isp116x_init(void)
+{
+	if (usb_disabled())
+		return -ENODEV;
+
+	INFO("driver %s, %s\n", hcd_name, DRIVER_VERSION);
+	return driver_register(&isp116x_driver);
+}
+
+module_init(isp116x_init);
+
+static void __exit isp116x_cleanup(void)
+{
+	driver_unregister(&isp116x_driver);
+}
+
+module_exit(isp116x_cleanup);
diff --git a/drivers/usb/host/isp116x.h b/drivers/usb/host/isp116x.h
new file mode 100644
index 00000000000..58873470dcf
--- /dev/null
+++ b/drivers/usb/host/isp116x.h
@@ -0,0 +1,583 @@
+/*
+ * ISP116x register declarations and HCD data structures
+ *
+ * Copyright (C) 2005 Olav Kongas <ok@artecdesign.ee>
+ * Portions:
+ * Copyright (C) 2004 Lothar Wassmann
+ * Copyright (C) 2004 Psion Teklogix
+ * Copyright (C) 2004 David Brownell
+ */
+
+/* us of 1ms frame */
+#define  MAX_LOAD_LIMIT		850
+
+/* Full speed: max # of bytes to transfer for a single urb
+   at a time must be < 1024 && must be multiple of 64.
+   832 allows transfering 4kiB within 5 frames. */
+#define MAX_TRANSFER_SIZE_FULLSPEED	832
+
+/* Low speed: there is no reason to schedule in very big
+   chunks; often the requested long transfers are for
+   string descriptors containing short strings. */
+#define MAX_TRANSFER_SIZE_LOWSPEED	64
+
+/* Bytetime (us), a rough indication of how much time it
+   would take to transfer a byte of useful data over USB */
+#define BYTE_TIME_FULLSPEED	1
+#define BYTE_TIME_LOWSPEED	20
+
+/* Buffer sizes */
+#define ISP116x_BUF_SIZE	4096
+#define ISP116x_ITL_BUFSIZE	0
+#define ISP116x_ATL_BUFSIZE	((ISP116x_BUF_SIZE) - 2*(ISP116x_ITL_BUFSIZE))
+
+#define ISP116x_WRITE_OFFSET	0x80
+
+/*------------ ISP116x registers/bits ------------*/
+#define	HCREVISION	0x00
+#define	HCCONTROL	0x01
+#define		HCCONTROL_HCFS	(3 << 6)	/* host controller
+						   functional state */
+#define		HCCONTROL_USB_RESET	(0 << 6)
+#define		HCCONTROL_USB_RESUME	(1 << 6)
+#define		HCCONTROL_USB_OPER	(2 << 6)
+#define		HCCONTROL_USB_SUSPEND	(3 << 6)
+#define		HCCONTROL_RWC	(1 << 9)	/* remote wakeup connected */
+#define		HCCONTROL_RWE	(1 << 10)	/* remote wakeup enable */
+#define	HCCMDSTAT	0x02
+#define		HCCMDSTAT_HCR	(1 << 0)	/* host controller reset */
+#define		HCCMDSTAT_SOC	(3 << 16)	/* scheduling overrun count */
+#define	HCINTSTAT	0x03
+#define		HCINT_SO	(1 << 0)	/* scheduling overrun */
+#define		HCINT_WDH	(1 << 1)	/* writeback of done_head */
+#define		HCINT_SF	(1 << 2)	/* start frame */
+#define		HCINT_RD	(1 << 3)	/* resume detect */
+#define		HCINT_UE	(1 << 4)	/* unrecoverable error */
+#define		HCINT_FNO	(1 << 5)	/* frame number overflow */
+#define		HCINT_RHSC	(1 << 6)	/* root hub status change */
+#define		HCINT_OC	(1 << 30)	/* ownership change */
+#define		HCINT_MIE	(1 << 31)	/* master interrupt enable */
+#define	HCINTENB	0x04
+#define	HCINTDIS	0x05
+#define	HCFMINTVL	0x0d
+#define	HCFMREM		0x0e
+#define	HCFMNUM		0x0f
+#define	HCLSTHRESH	0x11
+#define	HCRHDESCA	0x12
+#define		RH_A_NDP	(0x3 << 0)	/* # downstream ports */
+#define		RH_A_PSM	(1 << 8)	/* power switching mode */
+#define		RH_A_NPS	(1 << 9)	/* no power switching */
+#define		RH_A_DT		(1 << 10)	/* device type (mbz) */
+#define		RH_A_OCPM	(1 << 11)	/* overcurrent protection
+						   mode */
+#define		RH_A_NOCP	(1 << 12)	/* no overcurrent protection */
+#define		RH_A_POTPGT	(0xff << 24)	/* power on -> power good
+						   time */
+#define	HCRHDESCB	0x13
+#define		RH_B_DR		(0xffff << 0)	/* device removable flags */
+#define		RH_B_PPCM	(0xffff << 16)	/* port power control mask */
+#define	HCRHSTATUS	0x14
+#define		RH_HS_LPS	(1 << 0)	/* local power status */
+#define		RH_HS_OCI	(1 << 1)	/* over current indicator */
+#define		RH_HS_DRWE	(1 << 15)	/* device remote wakeup
+						   enable */
+#define		RH_HS_LPSC	(1 << 16)	/* local power status change */
+#define		RH_HS_OCIC	(1 << 17)	/* over current indicator
+						   change */
+#define		RH_HS_CRWE	(1 << 31)	/* clear remote wakeup
+						   enable */
+#define	HCRHPORT1	0x15
+#define		RH_PS_CCS	(1 << 0)	/* current connect status */
+#define		RH_PS_PES	(1 << 1)	/* port enable status */
+#define		RH_PS_PSS	(1 << 2)	/* port suspend status */
+#define		RH_PS_POCI	(1 << 3)	/* port over current
+						   indicator */
+#define		RH_PS_PRS	(1 << 4)	/* port reset status */
+#define		RH_PS_PPS	(1 << 8)	/* port power status */
+#define		RH_PS_LSDA	(1 << 9)	/* low speed device attached */
+#define		RH_PS_CSC	(1 << 16)	/* connect status change */
+#define		RH_PS_PESC	(1 << 17)	/* port enable status change */
+#define		RH_PS_PSSC	(1 << 18)	/* port suspend status
+						   change */
+#define		RH_PS_OCIC	(1 << 19)	/* over current indicator
+						   change */
+#define		RH_PS_PRSC	(1 << 20)	/* port reset status change */
+#define		HCRHPORT_CLRMASK	(0x1f << 16)
+#define	HCRHPORT2	0x16
+#define	HCHWCFG		0x20
+#define		HCHWCFG_15KRSEL		(1 << 12)
+#define		HCHWCFG_CLKNOTSTOP	(1 << 11)
+#define		HCHWCFG_ANALOG_OC	(1 << 10)
+#define		HCHWCFG_DACK_MODE	(1 << 8)
+#define		HCHWCFG_EOT_POL		(1 << 7)
+#define		HCHWCFG_DACK_POL	(1 << 6)
+#define		HCHWCFG_DREQ_POL	(1 << 5)
+#define		HCHWCFG_DBWIDTH_MASK	(0x03 << 3)
+#define		HCHWCFG_DBWIDTH(n)	(((n) << 3) & HCHWCFG_DBWIDTH_MASK)
+#define		HCHWCFG_INT_POL		(1 << 2)
+#define		HCHWCFG_INT_TRIGGER	(1 << 1)
+#define		HCHWCFG_INT_ENABLE	(1 << 0)
+#define	HCDMACFG	0x21
+#define		HCDMACFG_BURST_LEN_MASK	(0x03 << 5)
+#define		HCDMACFG_BURST_LEN(n)	(((n) << 5) & HCDMACFG_BURST_LEN_MASK)
+#define		HCDMACFG_BURST_LEN_1	HCDMACFG_BURST_LEN(0)
+#define		HCDMACFG_BURST_LEN_4	HCDMACFG_BURST_LEN(1)
+#define		HCDMACFG_BURST_LEN_8	HCDMACFG_BURST_LEN(2)
+#define		HCDMACFG_DMA_ENABLE	(1 << 4)
+#define		HCDMACFG_BUF_TYPE_MASK	(0x07 << 1)
+#define		HCDMACFG_CTR_SEL	(1 << 2)
+#define		HCDMACFG_ITLATL_SEL	(1 << 1)
+#define		HCDMACFG_DMA_RW_SELECT	(1 << 0)
+#define	HCXFERCTR	0x22
+#define	HCuPINT		0x24
+#define		HCuPINT_SOF		(1 << 0)
+#define		HCuPINT_ATL		(1 << 1)
+#define		HCuPINT_AIIEOT		(1 << 2)
+#define		HCuPINT_OPR		(1 << 4)
+#define		HCuPINT_SUSP		(1 << 5)
+#define		HCuPINT_CLKRDY		(1 << 6)
+#define	HCuPINTENB	0x25
+#define	HCCHIPID	0x27
+#define		HCCHIPID_MASK		0xff00
+#define		HCCHIPID_MAGIC		0x6100
+#define	HCSCRATCH	0x28
+#define	HCSWRES		0x29
+#define		HCSWRES_MAGIC		0x00f6
+#define	HCITLBUFLEN	0x2a
+#define	HCATLBUFLEN	0x2b
+#define	HCBUFSTAT	0x2c
+#define		HCBUFSTAT_ITL0_FULL	(1 << 0)
+#define		HCBUFSTAT_ITL1_FULL	(1 << 1)
+#define		HCBUFSTAT_ATL_FULL	(1 << 2)
+#define		HCBUFSTAT_ITL0_DONE	(1 << 3)
+#define		HCBUFSTAT_ITL1_DONE	(1 << 4)
+#define		HCBUFSTAT_ATL_DONE	(1 << 5)
+#define	HCRDITL0LEN	0x2d
+#define	HCRDITL1LEN	0x2e
+#define	HCITLPORT	0x40
+#define	HCATLPORT	0x41
+
+/* Philips transfer descriptor */
+struct ptd {
+	u16 count;
+#define	PTD_COUNT_MSK	(0x3ff << 0)
+#define	PTD_TOGGLE_MSK	(1 << 10)
+#define	PTD_ACTIVE_MSK	(1 << 11)
+#define	PTD_CC_MSK	(0xf << 12)
+	u16 mps;
+#define	PTD_MPS_MSK	(0x3ff << 0)
+#define	PTD_SPD_MSK	(1 << 10)
+#define	PTD_LAST_MSK	(1 << 11)
+#define	PTD_EP_MSK	(0xf << 12)
+	u16 len;
+#define	PTD_LEN_MSK	(0x3ff << 0)
+#define	PTD_DIR_MSK	(3 << 10)
+#define	PTD_DIR_SETUP	(0)
+#define	PTD_DIR_OUT	(1)
+#define	PTD_DIR_IN	(2)
+#define	PTD_B5_5_MSK	(1 << 13)
+	u16 faddr;
+#define	PTD_FA_MSK	(0x7f << 0)
+#define	PTD_FMT_MSK	(1 << 7)
+} __attribute__ ((packed, aligned(2)));
+
+/* PTD accessor macros. */
+#define PTD_GET_COUNT(p)	(((p)->count & PTD_COUNT_MSK) >> 0)
+#define PTD_COUNT(v)		(((v) << 0) & PTD_COUNT_MSK)
+#define PTD_GET_TOGGLE(p)	(((p)->count & PTD_TOGGLE_MSK) >> 10)
+#define PTD_TOGGLE(v)		(((v) << 10) & PTD_TOGGLE_MSK)
+#define PTD_GET_ACTIVE(p)	(((p)->count & PTD_ACTIVE_MSK) >> 11)
+#define PTD_ACTIVE(v)		(((v) << 11) & PTD_ACTIVE_MSK)
+#define PTD_GET_CC(p)		(((p)->count & PTD_CC_MSK) >> 12)
+#define PTD_CC(v)		(((v) << 12) & PTD_CC_MSK)
+#define PTD_GET_MPS(p)		(((p)->mps & PTD_MPS_MSK) >> 0)
+#define PTD_MPS(v)		(((v) << 0) & PTD_MPS_MSK)
+#define PTD_GET_SPD(p)		(((p)->mps & PTD_SPD_MSK) >> 10)
+#define PTD_SPD(v)		(((v) << 10) & PTD_SPD_MSK)
+#define PTD_GET_LAST(p)		(((p)->mps & PTD_LAST_MSK) >> 11)
+#define PTD_LAST(v)		(((v) << 11) & PTD_LAST_MSK)
+#define PTD_GET_EP(p)		(((p)->mps & PTD_EP_MSK) >> 12)
+#define PTD_EP(v)		(((v) << 12) & PTD_EP_MSK)
+#define PTD_GET_LEN(p)		(((p)->len & PTD_LEN_MSK) >> 0)
+#define PTD_LEN(v)		(((v) << 0) & PTD_LEN_MSK)
+#define PTD_GET_DIR(p)		(((p)->len & PTD_DIR_MSK) >> 10)
+#define PTD_DIR(v)		(((v) << 10) & PTD_DIR_MSK)
+#define PTD_GET_B5_5(p)		(((p)->len & PTD_B5_5_MSK) >> 13)
+#define PTD_B5_5(v)		(((v) << 13) & PTD_B5_5_MSK)
+#define PTD_GET_FA(p)		(((p)->faddr & PTD_FA_MSK) >> 0)
+#define PTD_FA(v)		(((v) << 0) & PTD_FA_MSK)
+#define PTD_GET_FMT(p)		(((p)->faddr & PTD_FMT_MSK) >> 7)
+#define PTD_FMT(v)		(((v) << 7) & PTD_FMT_MSK)
+
+/*  Hardware transfer status codes -- CC from ptd->count */
+#define TD_CC_NOERROR      0x00
+#define TD_CC_CRC          0x01
+#define TD_CC_BITSTUFFING  0x02
+#define TD_CC_DATATOGGLEM  0x03
+#define TD_CC_STALL        0x04
+#define TD_DEVNOTRESP      0x05
+#define TD_PIDCHECKFAIL    0x06
+#define TD_UNEXPECTEDPID   0x07
+#define TD_DATAOVERRUN     0x08
+#define TD_DATAUNDERRUN    0x09
+    /* 0x0A, 0x0B reserved for hardware */
+#define TD_BUFFEROVERRUN   0x0C
+#define TD_BUFFERUNDERRUN  0x0D
+    /* 0x0E, 0x0F reserved for HCD */
+#define TD_NOTACCESSED     0x0F
+
+/* map PTD status codes (CC) to errno values */
+static const int cc_to_error[16] = {
+	/* No  Error  */ 0,
+	/* CRC Error  */ -EILSEQ,
+	/* Bit Stuff  */ -EPROTO,
+	/* Data Togg  */ -EILSEQ,
+	/* Stall      */ -EPIPE,
+	/* DevNotResp */ -ETIMEDOUT,
+	/* PIDCheck   */ -EPROTO,
+	/* UnExpPID   */ -EPROTO,
+	/* DataOver   */ -EOVERFLOW,
+	/* DataUnder  */ -EREMOTEIO,
+	/* (for hw)   */ -EIO,
+	/* (for hw)   */ -EIO,
+	/* BufferOver */ -ECOMM,
+	/* BuffUnder  */ -ENOSR,
+	/* (for HCD)  */ -EALREADY,
+	/* (for HCD)  */ -EALREADY
+};
+
+/*--------------------------------------------------------------*/
+
+#define	LOG2_PERIODIC_SIZE	5	/* arbitrary; this matches OHCI */
+#define	PERIODIC_SIZE		(1 << LOG2_PERIODIC_SIZE)
+
+struct isp116x {
+	spinlock_t lock;
+	struct work_struct rh_resume;
+
+	void __iomem *addr_reg;
+	void __iomem *data_reg;
+
+	struct isp116x_platform_data *board;
+
+	struct proc_dir_entry *pde;
+	unsigned long stat1, stat2, stat4, stat8, stat16;
+
+	/* HC registers */
+	u32 intenb;		/* "OHCI" interrupts */
+	u16 irqenb;		/* uP interrupts */
+
+	/* Root hub registers */
+	u32 rhdesca;
+	u32 rhdescb;
+	u32 rhstatus;
+	u32 rhport[2];
+
+	/* async schedule: control, bulk */
+	struct list_head async;
+
+	/* periodic schedule: int */
+	u16 load[PERIODIC_SIZE];
+	struct isp116x_ep *periodic[PERIODIC_SIZE];
+	unsigned periodic_count;
+	u16 fmindex;
+
+	/* Schedule for the current frame */
+	struct isp116x_ep *atl_active;
+	int atl_buflen;
+	int atl_bufshrt;
+	int atl_last_dir;
+	atomic_t atl_finishing;
+};
+
+static inline struct isp116x *hcd_to_isp116x(struct usb_hcd *hcd)
+{
+	return (struct isp116x *)(hcd->hcd_priv);
+}
+
+static inline struct usb_hcd *isp116x_to_hcd(struct isp116x *isp116x)
+{
+	return container_of((void *)isp116x, struct usb_hcd, hcd_priv);
+}
+
+struct isp116x_ep {
+	struct usb_host_endpoint *hep;
+	struct usb_device *udev;
+	struct ptd ptd;
+
+	u8 maxpacket;
+	u8 epnum;
+	u8 nextpid;
+	u16 error_count;
+	u16 length;		/* of current packet */
+	unsigned char *data;	/* to databuf */
+	/* queue of active EP's (the ones scheduled for the
+	   current frame) */
+	struct isp116x_ep *active;
+
+	/* periodic schedule */
+	u16 period;
+	u16 branch;
+	u16 load;
+	struct isp116x_ep *next;
+
+	/* async schedule */
+	struct list_head schedule;
+};
+
+/*-------------------------------------------------------------------------*/
+
+#ifdef DEBUG
+#define DBG(stuff...)		printk(KERN_DEBUG "116x: " stuff)
+#else
+#define DBG(stuff...)		do{}while(0)
+#endif
+
+#ifdef VERBOSE
+#    define VDBG		DBG
+#else
+#    define VDBG(stuff...)	do{}while(0)
+#endif
+
+#define ERR(stuff...)		printk(KERN_ERR "116x: " stuff)
+#define WARN(stuff...)		printk(KERN_WARNING "116x: " stuff)
+#define INFO(stuff...)		printk(KERN_INFO "116x: " stuff)
+
+/* ------------------------------------------------- */
+
+#if defined(USE_PLATFORM_DELAY)
+#if defined(USE_NDELAY)
+#error USE_PLATFORM_DELAY and USE_NDELAY simultaneously defined.
+#endif
+#define	isp116x_delay(h,d)	(h)->board->delay(	\
+				isp116x_to_hcd(h)->self.controller,d)
+#define isp116x_check_platform_delay(h)	((h)->board->delay == NULL)
+#elif defined(USE_NDELAY)
+#define	isp116x_delay(h,d)	ndelay(d)
+#define isp116x_check_platform_delay(h)	0
+#else
+#define	isp116x_delay(h,d)	do{}while(0)
+#define isp116x_check_platform_delay(h)	0
+#endif
+
+#if defined(DEBUG)
+#define	IRQ_TEST()	BUG_ON(!irqs_disabled())
+#else
+#define	IRQ_TEST()	do{}while(0)
+#endif
+
+static inline void isp116x_write_addr(struct isp116x *isp116x, unsigned reg)
+{
+	IRQ_TEST();
+	writew(reg & 0xff, isp116x->addr_reg);
+	isp116x_delay(isp116x, 300);
+}
+
+static inline void isp116x_write_data16(struct isp116x *isp116x, u16 val)
+{
+	writew(val, isp116x->data_reg);
+	isp116x_delay(isp116x, 150);
+}
+
+static inline void isp116x_raw_write_data16(struct isp116x *isp116x, u16 val)
+{
+	__raw_writew(val, isp116x->data_reg);
+	isp116x_delay(isp116x, 150);
+}
+
+static inline u16 isp116x_read_data16(struct isp116x *isp116x)
+{
+	u16 val;
+
+	val = readw(isp116x->data_reg);
+	isp116x_delay(isp116x, 150);
+	return val;
+}
+
+static inline u16 isp116x_raw_read_data16(struct isp116x *isp116x)
+{
+	u16 val;
+
+	val = __raw_readw(isp116x->data_reg);
+	isp116x_delay(isp116x, 150);
+	return val;
+}
+
+static inline void isp116x_write_data32(struct isp116x *isp116x, u32 val)
+{
+	writew(val & 0xffff, isp116x->data_reg);
+	isp116x_delay(isp116x, 150);
+	writew(val >> 16, isp116x->data_reg);
+	isp116x_delay(isp116x, 150);
+}
+
+static inline u32 isp116x_read_data32(struct isp116x *isp116x)
+{
+	u32 val;
+
+	val = (u32) readw(isp116x->data_reg);
+	isp116x_delay(isp116x, 150);
+	val |= ((u32) readw(isp116x->data_reg)) << 16;
+	isp116x_delay(isp116x, 150);
+	return val;
+}
+
+/* Let's keep register access functions out of line. Hint:
+   we wait at least 150 ns at every access.
+*/
+static u16 isp116x_read_reg16(struct isp116x *isp116x, unsigned reg)
+{
+	isp116x_write_addr(isp116x, reg);
+	return isp116x_read_data16(isp116x);
+}
+
+static u32 isp116x_read_reg32(struct isp116x *isp116x, unsigned reg)
+{
+	isp116x_write_addr(isp116x, reg);
+	return isp116x_read_data32(isp116x);
+}
+
+static void isp116x_write_reg16(struct isp116x *isp116x, unsigned reg,
+				unsigned val)
+{
+	isp116x_write_addr(isp116x, reg | ISP116x_WRITE_OFFSET);
+	isp116x_write_data16(isp116x, (u16) (val & 0xffff));
+}
+
+static void isp116x_write_reg32(struct isp116x *isp116x, unsigned reg,
+				unsigned val)
+{
+	isp116x_write_addr(isp116x, reg | ISP116x_WRITE_OFFSET);
+	isp116x_write_data32(isp116x, (u32) val);
+}
+
+#define isp116x_show_reg(d,r) {					\
+	if ((r) < 0x20) {			                \
+		DBG("%-12s[%02x]: %08x\n", #r,			\
+			r, isp116x_read_reg32(d, r));		\
+	} else {						\
+		DBG("%-12s[%02x]:     %04x\n", #r,		\
+			r, isp116x_read_reg16(d, r));	    	\
+	}							\
+}
+
+static inline void isp116x_show_regs(struct isp116x *isp116x)
+{
+	isp116x_show_reg(isp116x, HCREVISION);
+	isp116x_show_reg(isp116x, HCCONTROL);
+	isp116x_show_reg(isp116x, HCCMDSTAT);
+	isp116x_show_reg(isp116x, HCINTSTAT);
+	isp116x_show_reg(isp116x, HCINTENB);
+	isp116x_show_reg(isp116x, HCFMINTVL);
+	isp116x_show_reg(isp116x, HCFMREM);
+	isp116x_show_reg(isp116x, HCFMNUM);
+	isp116x_show_reg(isp116x, HCLSTHRESH);
+	isp116x_show_reg(isp116x, HCRHDESCA);
+	isp116x_show_reg(isp116x, HCRHDESCB);
+	isp116x_show_reg(isp116x, HCRHSTATUS);
+	isp116x_show_reg(isp116x, HCRHPORT1);
+	isp116x_show_reg(isp116x, HCRHPORT2);
+	isp116x_show_reg(isp116x, HCHWCFG);
+	isp116x_show_reg(isp116x, HCDMACFG);
+	isp116x_show_reg(isp116x, HCXFERCTR);
+	isp116x_show_reg(isp116x, HCuPINT);
+	isp116x_show_reg(isp116x, HCuPINTENB);
+	isp116x_show_reg(isp116x, HCCHIPID);
+	isp116x_show_reg(isp116x, HCSCRATCH);
+	isp116x_show_reg(isp116x, HCITLBUFLEN);
+	isp116x_show_reg(isp116x, HCATLBUFLEN);
+	isp116x_show_reg(isp116x, HCBUFSTAT);
+	isp116x_show_reg(isp116x, HCRDITL0LEN);
+	isp116x_show_reg(isp116x, HCRDITL1LEN);
+}
+
+#if defined(URB_TRACE)
+
+#define PIPETYPE(pipe)  ({ char *__s;			\
+	if (usb_pipecontrol(pipe))	__s = "ctrl";	\
+	else if (usb_pipeint(pipe))	__s = "int";	\
+	else if (usb_pipebulk(pipe))	__s = "bulk";	\
+	else				__s = "iso";	\
+	__s;})
+#define PIPEDIR(pipe)   ({ usb_pipein(pipe) ? "in" : "out"; })
+#define URB_NOTSHORT(urb) ({ (urb)->transfer_flags & URB_SHORT_NOT_OK ? \
+	"short_not_ok" : ""; })
+
+/* print debug info about the URB */
+static void urb_dbg(struct urb *urb, char *msg)
+{
+	unsigned int pipe;
+
+	if (!urb) {
+		DBG("%s: zero urb\n", msg);
+		return;
+	}
+	pipe = urb->pipe;
+	DBG("%s: FA %d ep%d%s %s: len %d/%d %s\n", msg,
+	    usb_pipedevice(pipe), usb_pipeendpoint(pipe),
+	    PIPEDIR(pipe), PIPETYPE(pipe),
+	    urb->transfer_buffer_length, urb->actual_length, URB_NOTSHORT(urb));
+}
+
+#else
+
+#define  urb_dbg(urb,msg)   do{}while(0)
+
+#endif				/* ! defined(URB_TRACE) */
+
+#if defined(PTD_TRACE)
+
+#define PTD_DIR_STR(ptd)  ({char __c;		\
+	switch(PTD_GET_DIR(ptd)){		\
+	case 0:  __c = 's'; break;		\
+	case 1:  __c = 'o'; break;		\
+	default: __c = 'i'; break;		\
+	}; __c;})
+
+/*
+  Dump PTD info. The code documents the format
+  perfectly, right :)
+*/
+static inline void dump_ptd(struct ptd *ptd)
+{
+	printk("td: %x %d%c%d %d,%d,%d  %x %x%x%x\n",
+	       PTD_GET_CC(ptd), PTD_GET_FA(ptd),
+	       PTD_DIR_STR(ptd), PTD_GET_EP(ptd),
+	       PTD_GET_COUNT(ptd), PTD_GET_LEN(ptd), PTD_GET_MPS(ptd),
+	       PTD_GET_TOGGLE(ptd), PTD_GET_ACTIVE(ptd),
+	       PTD_GET_SPD(ptd), PTD_GET_LAST(ptd));
+}
+
+static inline void dump_ptd_out_data(struct ptd *ptd, u8 * buf)
+{
+	int k;
+
+	if (PTD_GET_DIR(ptd) != PTD_DIR_IN && PTD_GET_LEN(ptd)) {
+		printk("-> ");
+		for (k = 0; k < PTD_GET_LEN(ptd); ++k)
+			printk("%02x ", ((u8 *) buf)[k]);
+		printk("\n");
+	}
+}
+
+static inline void dump_ptd_in_data(struct ptd *ptd, u8 * buf)
+{
+	int k;
+
+	if (PTD_GET_DIR(ptd) == PTD_DIR_IN && PTD_GET_COUNT(ptd)) {
+		printk("<- ");
+		for (k = 0; k < PTD_GET_COUNT(ptd); ++k)
+			printk("%02x ", ((u8 *) buf)[k]);
+		printk("\n");
+	}
+	if (PTD_GET_LAST(ptd))
+		printk("-\n");
+}
+
+#else
+
+#define dump_ptd(ptd)               do{}while(0)
+#define dump_ptd_in_data(ptd,buf)   do{}while(0)
+#define dump_ptd_out_data(ptd,buf)  do{}while(0)
+
+#endif				/* ! defined(PTD_TRACE) */
diff --git a/include/linux/usb_isp116x.h b/include/linux/usb_isp116x.h
new file mode 100644
index 00000000000..5f5a9d9bd6c
--- /dev/null
+++ b/include/linux/usb_isp116x.h
@@ -0,0 +1,47 @@
+
+/*
+ * Board initialization code should put one of these into dev->platform_data
+ * and place the isp116x onto platform_bus.
+ */
+
+struct isp116x_platform_data {
+	/* Enable internal resistors on downstream ports */
+	unsigned sel15Kres:1;
+	/* Chip's internal clock won't be stopped in suspended state.
+	   Setting/unsetting this bit takes effect only if
+	   'remote_wakeup_enable' below is not set. */
+	unsigned clknotstop:1;
+	/* On-chip overcurrent protection */
+	unsigned oc_enable:1;
+	/* INT output polarity */
+	unsigned int_act_high:1;
+	/* INT edge or level triggered */
+	unsigned int_edge_triggered:1;
+	/* WAKEUP pin connected - NOT SUPPORTED  */
+	/* unsigned remote_wakeup_connected:1; */
+	/* Wakeup by devices on usb bus enabled */
+	unsigned remote_wakeup_enable:1;
+	/* Switch or not to switch (keep always powered) */
+	unsigned no_power_switching:1;
+	/* Ganged port power switching (0) or individual port
+	   power switching (1) */
+	unsigned power_switching_mode:1;
+	/* Given port_power, msec/2 after power on till power good */
+	u8 potpg;
+	/* Hardware reset set/clear. If implemented, this function must:
+	   if set == 0,   deassert chip's HW reset pin
+	   otherwise,     assert chip's HW reset pin       */
+	void (*reset) (struct device * dev, int set);
+	/* Hardware clock start/stop. If implemented, this function must:
+	   if start == 0,    stop the external clock
+	   otherwise,        start the external clock
+	 */
+	void (*clock) (struct device * dev, int start);
+	/* Inter-io delay (ns). The chip is picky about access timings; it
+	   expects at least:
+	   150ns delay between consecutive accesses to DATA_REG,
+	   300ns delay between access to ADDR_REG and DATA_REG
+	   OE, WE MUST NOT be changed during these intervals
+	 */
+	void (*delay) (struct device * dev, int delay);
+};
-- 
cgit v1.2.3-70-g09d2


From 65111084c63d7674dc37833e8eb59cfdaa4d0bda Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Thu, 28 Apr 2005 13:52:31 -0700
Subject: [PATCH] USB: more omap_udc updates (dma and omap1710)

More omap_udc updates:

  * OMAP 1710 updates
      - new UDC bit for clearing endpoint toggle, affecting CLEAR_HALT
      - new OTG bits affecting wakeup
  * Fix the bug Vladimir noted, that IN-DMA transfer code path kicks in
    for under 1024 bytes (not "up to 1024 bytes")
  * Handle transceiver setup more intelligently
      - use transceiver whenever one's available; this can be handy
        for GPIO based, loopback, or transceiverless configs
      - cleanup correctly after the "unrecognized HMC" case
  * DMA performance tweaks
      - allow burst/pack for memory access
      - use 16 bit DMA access most of the time on TIPB
  * Add workarounds for some DMA errata (not observed "in the wild"):
      - DMA CSAC/CDAC reads returning zero
      - RX/TX DMA config registers bit 12 always reads as zero (TI patch)
  * More "sparse" warnings removed, notably "changing" the SETUP packet
    to return data in USB byteorder (an API change, null effect on OMAP
    except for these warnings).

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/omap_udc.c   | 153 +++++++++++++++++++++++++++++-----------
 drivers/usb/gadget/omap_udc.h   |   4 +-
 include/asm-arm/arch-omap/usb.h |   9 +++
 3 files changed, 125 insertions(+), 41 deletions(-)

(limited to 'include')

diff --git a/drivers/usb/gadget/omap_udc.c b/drivers/usb/gadget/omap_udc.c
index 4ec91a68f96..a2b812af6e6 100644
--- a/drivers/usb/gadget/omap_udc.c
+++ b/drivers/usb/gadget/omap_udc.c
@@ -166,7 +166,7 @@ static int omap_ep_enable(struct usb_ep *_ep,
 	maxp = le16_to_cpu (desc->wMaxPacketSize);
 	if ((desc->bmAttributes == USB_ENDPOINT_XFER_BULK
 				&& maxp != ep->maxpacket)
-			|| desc->wMaxPacketSize > ep->maxpacket
+			|| le16_to_cpu(desc->wMaxPacketSize) > ep->maxpacket
 			|| !desc->wMaxPacketSize) {
 		DBG("%s, bad %s maxpacket\n", __FUNCTION__, _ep->name);
 		return -ERANGE;
@@ -213,7 +213,7 @@ static int omap_ep_enable(struct usb_ep *_ep,
 	ep->has_dma = 0;
 	ep->lch = -1;
 	use_ep(ep, UDC_EP_SEL);
-	UDC_CTRL_REG = UDC_RESET_EP;
+	UDC_CTRL_REG = udc->clr_halt;
 	ep->ackwait = 0;
 	deselect_ep();
 
@@ -537,6 +537,32 @@ static int read_fifo(struct omap_ep *ep, struct omap_req *req)
 
 /*-------------------------------------------------------------------------*/
 
+static inline dma_addr_t dma_csac(unsigned lch)
+{
+	dma_addr_t	csac;
+
+	/* omap 3.2/3.3 erratum: sometimes 0 is returned if CSAC/CDAC is
+	 * read before the DMA controller finished disabling the channel.
+	 */
+	csac = omap_readw(OMAP_DMA_CSAC(lch));
+	if (csac == 0)
+		csac = omap_readw(OMAP_DMA_CSAC(lch));
+	return csac;
+}
+
+static inline dma_addr_t dma_cdac(unsigned lch)
+{
+	dma_addr_t	cdac;
+
+	/* omap 3.2/3.3 erratum: sometimes 0 is returned if CSAC/CDAC is
+	 * read before the DMA controller finished disabling the channel.
+	 */
+	cdac = omap_readw(OMAP_DMA_CDAC(lch));
+	if (cdac == 0)
+		cdac = omap_readw(OMAP_DMA_CDAC(lch));
+	return cdac;
+}
+
 static u16 dma_src_len(struct omap_ep *ep, dma_addr_t start)
 {
 	dma_addr_t	end;
@@ -547,7 +573,7 @@ static u16 dma_src_len(struct omap_ep *ep, dma_addr_t start)
 	if (cpu_is_omap15xx())
 		return 0;
 
-	end = omap_readw(OMAP_DMA_CSAC(ep->lch));
+	end = dma_csac(ep->lch);
 	if (end == ep->dma_counter)
 		return 0;
 
@@ -558,14 +584,14 @@ static u16 dma_src_len(struct omap_ep *ep, dma_addr_t start)
 }
 
 #define DMA_DEST_LAST(x) (cpu_is_omap15xx() \
-		? OMAP_DMA_CSAC(x) /* really: CPC */ \
-		: OMAP_DMA_CDAC(x))
+		? omap_readw(OMAP_DMA_CSAC(x)) /* really: CPC */ \
+		: dma_cdac(x))
 
 static u16 dma_dest_len(struct omap_ep *ep, dma_addr_t start)
 {
 	dma_addr_t	end;
 
-	end = omap_readw(DMA_DEST_LAST(ep->lch));
+	end = DMA_DEST_LAST(ep->lch);
 	if (end == ep->dma_counter)
 		return 0;
 
@@ -592,7 +618,7 @@ static void next_in_dma(struct omap_ep *ep, struct omap_req *req)
 				: OMAP_DMA_SYNC_ELEMENT;
 
 	/* measure length in either bytes or packets */
-	if ((cpu_is_omap16xx() && length <= (UDC_TXN_TSC + 1))
+	if ((cpu_is_omap16xx() && length <= UDC_TXN_TSC)
 			|| (cpu_is_omap15xx() && length < ep->maxpacket)) {
 		txdma_ctrl = UDC_TXN_EOT | length;
 		omap_set_dma_transfer_params(ep->lch, OMAP_DMA_DATA_TYPE_S8,
@@ -601,15 +627,15 @@ static void next_in_dma(struct omap_ep *ep, struct omap_req *req)
 		length = min(length / ep->maxpacket,
 				(unsigned) UDC_TXN_TSC + 1);
  		txdma_ctrl = length;
-		omap_set_dma_transfer_params(ep->lch, OMAP_DMA_DATA_TYPE_S8,
-				ep->ep.maxpacket, length, sync_mode);
+		omap_set_dma_transfer_params(ep->lch, OMAP_DMA_DATA_TYPE_S16,
+				ep->ep.maxpacket >> 1, length, sync_mode);
 		length *= ep->maxpacket;
 	}
 	omap_set_dma_src_params(ep->lch, OMAP_DMA_PORT_EMIFF,
 		OMAP_DMA_AMODE_POST_INC, req->req.dma + req->req.actual);
 
 	omap_start_dma(ep->lch);
-	ep->dma_counter = omap_readw(OMAP_DMA_CSAC(ep->lch));
+	ep->dma_counter = dma_csac(ep->lch);
 	UDC_DMA_IRQ_EN_REG |= UDC_TX_DONE_IE(ep->dma_channel);
 	UDC_TXDMA_REG(ep->dma_channel) = UDC_TXN_START | txdma_ctrl;
 	req->dma_bytes = length;
@@ -649,12 +675,12 @@ static void next_out_dma(struct omap_ep *ep, struct omap_req *req)
 	packets = (req->req.length - req->req.actual) / ep->ep.maxpacket;
 	packets = min(packets, (unsigned)UDC_RXN_TC + 1);
 	req->dma_bytes = packets * ep->ep.maxpacket;
-	omap_set_dma_transfer_params(ep->lch, OMAP_DMA_DATA_TYPE_S8,
-			ep->ep.maxpacket, packets,
+	omap_set_dma_transfer_params(ep->lch, OMAP_DMA_DATA_TYPE_S16,
+			ep->ep.maxpacket >> 1, packets,
 			OMAP_DMA_SYNC_ELEMENT);
 	omap_set_dma_dest_params(ep->lch, OMAP_DMA_PORT_EMIFF,
 		OMAP_DMA_AMODE_POST_INC, req->req.dma + req->req.actual);
-	ep->dma_counter = omap_readw(DMA_DEST_LAST(ep->lch));
+	ep->dma_counter = DMA_DEST_LAST(ep->lch);
 
 	UDC_RXDMA_REG(ep->dma_channel) = UDC_RXN_STOP | (packets - 1);
 	UDC_DMA_IRQ_EN_REG |= UDC_RX_EOT_IE(ep->dma_channel);
@@ -762,7 +788,7 @@ static void dma_channel_claim(struct omap_ep *ep, unsigned channel)
 		reg = UDC_TXDMA_CFG_REG;
 	else
 		reg = UDC_RXDMA_CFG_REG;
-	reg |= 1 << 12;		/* "pulse" activated */
+	reg |= UDC_DMA_REQ;		/* "pulse" activated */
 
 	ep->dma_channel = 0;
 	ep->lch = -1;
@@ -786,6 +812,11 @@ static void dma_channel_claim(struct omap_ep *ep, unsigned channel)
 			ep->ep.name, dma_error, ep, &ep->lch);
 		if (status == 0) {
 			UDC_TXDMA_CFG_REG = reg;
+			/* EMIFF */
+			omap_set_dma_src_burst_mode(ep->lch,
+						OMAP_DMA_DATA_BURST_4);
+			omap_set_dma_src_data_pack(ep->lch, 1);
+			/* TIPB */
 			omap_set_dma_dest_params(ep->lch,
 				OMAP_DMA_PORT_TIPB,
 				OMAP_DMA_AMODE_CONSTANT,
@@ -796,10 +827,15 @@ static void dma_channel_claim(struct omap_ep *ep, unsigned channel)
 			ep->ep.name, dma_error, ep, &ep->lch);
 		if (status == 0) {
 			UDC_RXDMA_CFG_REG = reg;
+			/* TIPB */
 			omap_set_dma_src_params(ep->lch,
 				OMAP_DMA_PORT_TIPB,
 				OMAP_DMA_AMODE_CONSTANT,
 				(unsigned long) io_v2p((u32)&UDC_DATA_DMA_REG));
+			/* EMIFF */
+			omap_set_dma_dest_burst_mode(ep->lch,
+						OMAP_DMA_DATA_BURST_4);
+			omap_set_dma_dest_data_pack(ep->lch, 1);
 		}
 	}
 	if (status)
@@ -864,9 +900,13 @@ static void dma_channel_release(struct omap_ep *ep)
 			(ep->bEndpointAddress & USB_DIR_IN) ? 't' : 'r',
 			ep->dma_channel - 1, req);
 
+	/* NOTE: re-setting RX_REQ/TX_REQ because of a chip bug (before
+	 * OMAP 1710 ES2.0) where reading the DMA_CFG can clear them.
+	 */
+
 	/* wait till current packet DMA finishes, and fifo empties */
 	if (ep->bEndpointAddress & USB_DIR_IN) {
-		UDC_TXDMA_CFG_REG &= ~mask;
+		UDC_TXDMA_CFG_REG = (UDC_TXDMA_CFG_REG & ~mask) | UDC_DMA_REQ;
 
 		if (req) {
 			finish_in_dma(ep, req, -ECONNRESET);
@@ -879,7 +919,7 @@ static void dma_channel_release(struct omap_ep *ep)
 		while (UDC_TXDMA_CFG_REG & mask)
 			udelay(10);
 	} else {
-		UDC_RXDMA_CFG_REG &= ~mask;
+		UDC_RXDMA_CFG_REG = (UDC_RXDMA_CFG_REG & ~mask) | UDC_DMA_REQ;
 
 		/* dma empties the fifo */
 		while (UDC_RXDMA_CFG_REG & mask)
@@ -1140,7 +1180,7 @@ static int omap_ep_set_halt(struct usb_ep *_ep, int value)
 				dma_channel_claim(ep, channel);
 		} else {
 			use_ep(ep, 0);
-			UDC_CTRL_REG = UDC_RESET_EP;
+			UDC_CTRL_REG = ep->udc->clr_halt;
 			ep->ackwait = 0;
 			if (!(ep->bEndpointAddress & USB_DIR_IN)) {
 				UDC_CTRL_REG = UDC_SET_FIFO_EN;
@@ -1514,6 +1554,10 @@ static void ep0_irq(struct omap_udc *udc, u16 irq_src)
 			UDC_EP_NUM_REG = 0;
 		} while (UDC_IRQ_SRC_REG & UDC_SETUP);
 
+#define	w_value		le16_to_cpup (&u.r.wValue)
+#define	w_index		le16_to_cpup (&u.r.wIndex)
+#define	w_length	le16_to_cpup (&u.r.wLength)
+
 		/* Delegate almost all control requests to the gadget driver,
 		 * except for a handful of ch9 status/feature requests that
 		 * hardware doesn't autodecode _and_ the gadget API hides.
@@ -1528,11 +1572,11 @@ static void ep0_irq(struct omap_udc *udc, u16 irq_src)
 			/* udc needs to know when ep != 0 is valid */
 			if (u.r.bRequestType != USB_RECIP_DEVICE)
 				goto delegate;
-			if (u.r.wLength != 0)
+			if (w_length != 0)
 				goto do_stall;
 			udc->ep0_set_config = 1;
-			udc->ep0_reset_config = (u.r.wValue == 0);
-			VDBG("set config %d\n", u.r.wValue);
+			udc->ep0_reset_config = (w_value == 0);
+			VDBG("set config %d\n", w_value);
 
 			/* update udc NOW since gadget driver may start
 			 * queueing requests immediately; clear config
@@ -1548,18 +1592,18 @@ static void ep0_irq(struct omap_udc *udc, u16 irq_src)
 			/* clear endpoint halt */
 			if (u.r.bRequestType != USB_RECIP_ENDPOINT)
 				goto delegate;
-			if (u.r.wValue != USB_ENDPOINT_HALT
-					|| u.r.wLength != 0)
+			if (w_value != USB_ENDPOINT_HALT
+					|| w_length != 0)
 				goto do_stall;
-			ep = &udc->ep[u.r.wIndex & 0xf];
+			ep = &udc->ep[w_index & 0xf];
 			if (ep != ep0) {
-				if (u.r.wIndex & USB_DIR_IN)
+				if (w_index & USB_DIR_IN)
 					ep += 16;
 				if (ep->bmAttributes == USB_ENDPOINT_XFER_ISOC
 						|| !ep->desc)
 					goto do_stall;
 				use_ep(ep, 0);
-				UDC_CTRL_REG = UDC_RESET_EP;
+				UDC_CTRL_REG = udc->clr_halt;
 				ep->ackwait = 0;
 				if (!(ep->bEndpointAddress & USB_DIR_IN)) {
 					UDC_CTRL_REG = UDC_SET_FIFO_EN;
@@ -1577,11 +1621,11 @@ static void ep0_irq(struct omap_udc *udc, u16 irq_src)
 			/* set endpoint halt */
 			if (u.r.bRequestType != USB_RECIP_ENDPOINT)
 				goto delegate;
-			if (u.r.wValue != USB_ENDPOINT_HALT
-					|| u.r.wLength != 0)
+			if (w_value != USB_ENDPOINT_HALT
+					|| w_length != 0)
 				goto do_stall;
-			ep = &udc->ep[u.r.wIndex & 0xf];
-			if (u.r.wIndex & USB_DIR_IN)
+			ep = &udc->ep[w_index & 0xf];
+			if (w_index & USB_DIR_IN)
 				ep += 16;
 			if (ep->bmAttributes == USB_ENDPOINT_XFER_ISOC
 					|| ep == ep0 || !ep->desc)
@@ -1619,13 +1663,13 @@ ep0out_status_stage:
 			UDC_CTRL_REG = UDC_SET_FIFO_EN;
 			UDC_EP_NUM_REG = UDC_EP_DIR;
 			status = 0;
-			VDBG("GET_STATUS, interface %d\n", u.r.wIndex);
+			VDBG("GET_STATUS, interface %d\n", w_index);
 			/* next, status stage */
 			break;
 		default:
 delegate:
 			/* activate the ep0out fifo right away */
-			if (!udc->ep0_in && u.r.wLength) {
+			if (!udc->ep0_in && w_length) {
 				UDC_EP_NUM_REG = 0;
 				UDC_CTRL_REG = UDC_SET_FIFO_EN;
 			}
@@ -1636,7 +1680,11 @@ delegate:
 			 */
 			VDBG("SETUP %02x.%02x v%04x i%04x l%04x\n",
 				u.r.bRequestType, u.r.bRequest,
-				u.r.wValue, u.r.wIndex, u.r.wLength);
+				w_value, w_index, w_length);
+
+#undef	w_value
+#undef	w_index
+#undef	w_length
 
 			/* The gadget driver may return an error here,
 			 * causing an immediate protocol stall.
@@ -2181,14 +2229,14 @@ static int proc_otg_show(struct seq_file *s)
 
 	tmp = OTG_REV_REG;
 	trans = USB_TRANSCEIVER_CTRL_REG;
-	seq_printf(s, "\nOTG rev %d.%d, transceiver_ctrl %03x\n",
+	seq_printf(s, "\nOTG rev %d.%d, transceiver_ctrl %05x\n",
 		tmp >> 4, tmp & 0xf, trans);
 	tmp = OTG_SYSCON_1_REG;
 	seq_printf(s, "otg_syscon1 %08x usb2 %s, usb1 %s, usb0 %s,"
 			FOURBITS "\n", tmp,
 		trx_mode(USB2_TRX_MODE(tmp), trans & CONF_USB2_UNI_R),
 		trx_mode(USB1_TRX_MODE(tmp), trans & CONF_USB1_UNI_R),
-		(USB0_TRX_MODE(tmp) == 0)
+		(USB0_TRX_MODE(tmp) == 0 && !cpu_is_omap1710())
 			? "internal"
 			: trx_mode(USB0_TRX_MODE(tmp), 1),
 		(tmp & OTG_IDLE_EN) ? " !otg" : "",
@@ -2418,6 +2466,10 @@ static inline void remove_proc_file(void) {}
 /* Before this controller can enumerate, we need to pick an endpoint
  * configuration, or "fifo_mode"  That involves allocating 2KB of packet
  * buffer space among the endpoints we'll be operating.
+ *
+ * NOTE: as of OMAP 1710 ES2.0, writing a new endpoint config when
+ * UDC_SYSCON_1_REG.CFG_LOCK is set can now work.  We won't use that
+ * capability yet though.
  */
 static unsigned __init
 omap_ep_setup(char *name, u8 addr, u8 type,
@@ -2690,6 +2742,19 @@ static int __init omap_udc_probe(struct device *dev)
 			FUNC_MUX_CTRL_0_REG = tmp;
 		}
 	} else {
+		/* The transceiver may package some GPIO logic or handle
+		 * loopback and/or transceiverless setup; if we find one,
+		 * use it.  Except for OTG, we don't _need_ to talk to one;
+		 * but not having one probably means no VBUS detection.
+		 */
+		xceiv = otg_get_transceiver();
+		if (xceiv)
+			type = xceiv->label;
+		else if (config->otg) {
+			DBG("OTG requires external transceiver!\n");
+			goto cleanup0;
+		}
+
 		hmc = HMC_1610;
 		switch (hmc) {
 		case 0:			/* POWERUP DEFAULT == 0 */
@@ -2706,25 +2771,27 @@ static int __init omap_udc_probe(struct device *dev)
 		case 16:
 		case 19:
 		case 25:
-			xceiv = otg_get_transceiver();
 			if (!xceiv) {
 				DBG("external transceiver not registered!\n");
-				if (config->otg)
-					goto cleanup0;
 				type = "unknown";
-			} else
-				type = xceiv->label;
+			}
 			break;
 		case 21:			/* internal loopback */
 			type = "loopback";
 			break;
 		case 14:			/* transceiverless */
+			if (cpu_is_omap1710())
+				goto bad_on_1710;
+			/* FALL THROUGH */
+		case 13:
+		case 15:
 			type = "no";
 			break;
 
 		default:
+bad_on_1710:
 			ERR("unrecognized UDC HMC mode %d\n", hmc);
-			return -ENODEV;
+			goto cleanup0;
 		}
 	}
 	INFO("hmc mode %d, %s transceiver\n", hmc, type);
@@ -2741,6 +2808,12 @@ static int __init omap_udc_probe(struct device *dev)
 	udc->gadget.is_otg = (config->otg != 0);
 #endif
 
+	/* starting with omap1710 es2.0, clear toggle is a separate bit */
+	if (UDC_REV_REG >= 0x61)
+		udc->clr_halt = UDC_RESET_EP | UDC_CLRDATA_TOGGLE;
+	else
+		udc->clr_halt = UDC_RESET_EP;
+
 	/* USB general purpose IRQ:  ep0, state changes, dma, etc */
 	status = request_irq(odev->resource[1].start, omap_udc_irq,
 			SA_SAMPLE_RANDOM, driver_name, udc);
diff --git a/drivers/usb/gadget/omap_udc.h b/drivers/usb/gadget/omap_udc.h
index c9e68541622..652ee462734 100644
--- a/drivers/usb/gadget/omap_udc.h
+++ b/drivers/usb/gadget/omap_udc.h
@@ -20,6 +20,7 @@
 #define	UDC_CTRL_REG			UDC_REG(0x0C)	/* Endpoint control */
 #	define	UDC_CLR_HALT		(1 << 7)
 #	define	UDC_SET_HALT		(1 << 6)
+#	define	UDC_CLRDATA_TOGGLE	(1 << 3)
 #	define	UDC_SET_FIFO_EN		(1 << 2)
 #	define	UDC_CLR_EP		(1 << 1)
 #	define	UDC_RESET_EP		(1 << 0)
@@ -99,6 +100,7 @@
 
 /* DMA configuration registers:  up to three channels in each direction.  */
 #define	UDC_RXDMA_CFG_REG		UDC_REG(0x40)	/* 3 eps for RX DMA */
+#	define	UDC_DMA_REQ		(1 << 12)
 #define	UDC_TXDMA_CFG_REG		UDC_REG(0x44)	/* 3 eps for TX DMA */
 #define	UDC_DATA_DMA_REG		UDC_REG(0x48)	/* rx/tx fifo addr */
 
@@ -162,6 +164,7 @@ struct omap_udc {
 	spinlock_t			lock;
 	struct omap_ep			ep[32];
 	u16				devstat;
+	u16				clr_halt;
 	struct otg_transceiver		*transceiver;
 	struct list_head		iso;
 	unsigned			softconnect:1;
@@ -171,7 +174,6 @@ struct omap_udc {
 	unsigned			ep0_set_config:1;
 	unsigned			ep0_reset_config:1;
 	unsigned			ep0_setup:1;
-
 	struct completion		*done;
 };
 
diff --git a/include/asm-arm/arch-omap/usb.h b/include/asm-arm/arch-omap/usb.h
index 1438c6cef0c..054fb9a8e0c 100644
--- a/include/asm-arm/arch-omap/usb.h
+++ b/include/asm-arm/arch-omap/usb.h
@@ -47,6 +47,15 @@
 #	define	 HMC_TLLATTACH		(1 << 6)
 #	define	 OTG_HMC(w)		(((w)>>0)&0x3f)
 #define OTG_CTRL_REG			OTG_REG32(0x0c)
+#	define	 OTG_USB2_EN		(1 << 29)
+#	define	 OTG_USB2_DP		(1 << 28)
+#	define	 OTG_USB2_DM		(1 << 27)
+#	define	 OTG_USB1_EN		(1 << 26)
+#	define	 OTG_USB1_DP		(1 << 25)
+#	define	 OTG_USB1_DM		(1 << 24)
+#	define	 OTG_USB0_EN		(1 << 23)
+#	define	 OTG_USB0_DP		(1 << 22)
+#	define	 OTG_USB0_DM		(1 << 21)
 #	define	 OTG_ASESSVLD		(1 << 20)
 #	define	 OTG_BSESSEND		(1 << 19)
 #	define	 OTG_BSESSVLD		(1 << 18)
-- 
cgit v1.2.3-70-g09d2


From 1bbc169621cbe502b9143a27eb12802a0f1d43a0 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Sat, 7 May 2005 13:05:13 -0700
Subject: [PATCH] USB: gadget driver updates (SETUP api change)

This updates most of the gadget framework to expect SETUP packets use
USB byteorder (matching the annotation in <linux/usb_ch9.h> and usage
in the host side stack):

  - definition in <linux/usb_gadget.h>
  - gadget drivers:  Ethernet/RNDIS, serial/ACM, file_storage, gadgetfs.
  - dummy_hcd

It also includes some other similar changes as suggested by "sparse",
which was used to detect byteorder bugs.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/dummy_hcd.c    |  3 ---
 drivers/usb/gadget/ether.c        |  6 +++---
 drivers/usb/gadget/file_storage.c | 19 +++++++++----------
 drivers/usb/gadget/inode.c        | 12 ++++++------
 drivers/usb/gadget/serial.c       | 36 +++++++++++++++++++-----------------
 drivers/usb/gadget/zero.c         |  6 +++---
 include/linux/usb_gadget.h        |  2 +-
 7 files changed, 41 insertions(+), 43 deletions(-)

(limited to 'include')

diff --git a/drivers/usb/gadget/dummy_hcd.c b/drivers/usb/gadget/dummy_hcd.c
index 73d2f24050a..f9540adf2a4 100644
--- a/drivers/usb/gadget/dummy_hcd.c
+++ b/drivers/usb/gadget/dummy_hcd.c
@@ -1267,9 +1267,6 @@ restart:
 			struct dummy_ep			*ep2;
 
 			setup = *(struct usb_ctrlrequest*) urb->setup_packet;
-			le16_to_cpus (&setup.wIndex);
-			le16_to_cpus (&setup.wValue);
-			le16_to_cpus (&setup.wLength);
 			if (setup.wLength != urb->transfer_buffer_length) {
 				maybe_set_status (urb, -EOVERFLOW);
 				goto return_urb;
diff --git a/drivers/usb/gadget/ether.c b/drivers/usb/gadget/ether.c
index 3830a0a0fd5..9f8413e3c10 100644
--- a/drivers/usb/gadget/ether.c
+++ b/drivers/usb/gadget/ether.c
@@ -1277,9 +1277,9 @@ eth_setup (struct usb_gadget *gadget, const struct usb_ctrlrequest *ctrl)
 	struct eth_dev		*dev = get_gadget_data (gadget);
 	struct usb_request	*req = dev->req;
 	int			value = -EOPNOTSUPP;
-	u16			wIndex = (__force u16) ctrl->wIndex;
-	u16			wValue = (__force u16) ctrl->wValue;
-	u16			wLength = (__force u16) ctrl->wLength;
+	u16			wIndex = le16_to_cpu(ctrl->wIndex);
+	u16			wValue = le16_to_cpu(ctrl->wValue);
+	u16			wLength = le16_to_cpu(ctrl->wLength);
 
 	/* descriptors just go into the pre-allocated ep0 buffer,
 	 * while config change events may enable network traffic.
diff --git a/drivers/usb/gadget/file_storage.c b/drivers/usb/gadget/file_storage.c
index f5ce45c4b2a..4f57085619b 100644
--- a/drivers/usb/gadget/file_storage.c
+++ b/drivers/usb/gadget/file_storage.c
@@ -819,7 +819,7 @@ static void inline put_be32(u8 *buf, u32 val)
 	buf[0] = val >> 24;
 	buf[1] = val >> 16;
 	buf[2] = val >> 8;
-	buf[3] = val;
+	buf[3] = val & 0xff;
 }
 
 
@@ -1277,8 +1277,8 @@ static int class_setup_req(struct fsg_dev *fsg,
 {
 	struct usb_request	*req = fsg->ep0req;
 	int			value = -EOPNOTSUPP;
-	u16			w_index = ctrl->wIndex;
-	u16			w_length = ctrl->wLength;
+	u16			w_index = le16_to_cpu(ctrl->wIndex);
+	u16			w_length = le16_to_cpu(ctrl->wLength);
 
 	if (!fsg->config)
 		return value;
@@ -1345,7 +1345,7 @@ static int class_setup_req(struct fsg_dev *fsg,
 			"unknown class-specific control req "
 			"%02x.%02x v%04x i%04x l%u\n",
 			ctrl->bRequestType, ctrl->bRequest,
-			ctrl->wValue, w_index, w_length);
+			le16_to_cpu(ctrl->wValue), w_index, w_length);
 	return value;
 }
 
@@ -1359,8 +1359,8 @@ static int standard_setup_req(struct fsg_dev *fsg,
 {
 	struct usb_request	*req = fsg->ep0req;
 	int			value = -EOPNOTSUPP;
-	u16			w_index = ctrl->wIndex;
-	u16			w_value = ctrl->wValue;
+	u16			w_index = le16_to_cpu(ctrl->wIndex);
+	u16			w_value = le16_to_cpu(ctrl->wValue);
 
 	/* Usually this just stores reply data in the pre-allocated ep0 buffer,
 	 * but config change events will also reconfigure hardware. */
@@ -1469,7 +1469,7 @@ static int standard_setup_req(struct fsg_dev *fsg,
 		VDBG(fsg,
 			"unknown control req %02x.%02x v%04x i%04x l%u\n",
 			ctrl->bRequestType, ctrl->bRequest,
-			w_value, w_index, ctrl->wLength);
+			w_value, w_index, le16_to_cpu(ctrl->wLength));
 	}
 
 	return value;
@@ -1481,7 +1481,7 @@ static int fsg_setup(struct usb_gadget *gadget,
 {
 	struct fsg_dev		*fsg = get_gadget_data(gadget);
 	int			rc;
-	int			w_length = ctrl->wLength;
+	int			w_length = le16_to_cpu(ctrl->wLength);
 
 	++fsg->ep0_req_tag;		// Record arrival of a new request
 	fsg->ep0req->context = NULL;
@@ -1497,8 +1497,7 @@ static int fsg_setup(struct usb_gadget *gadget,
 	if (rc >= 0 && rc != DELAYED_STATUS) {
 		rc = min(rc, w_length);
 		fsg->ep0req->length = rc;
-		fsg->ep0req->zero = (rc < w_length &&
-				(rc % gadget->ep0->maxpacket) == 0);
+		fsg->ep0req->zero = rc < w_length;
 		fsg->ep0req_name = (ctrl->bRequestType & USB_DIR_IN ?
 				"ep0-in" : "ep0-out");
 		rc = ep0_queue(fsg);
diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index 1e5e6ddef78..020815397a4 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -417,8 +417,8 @@ ep_read (struct file *fd, char __user *buf, size_t len, loff_t *ptr)
 		goto free1;
 
 	value = ep_io (data, kbuf, len);
-	VDEBUG (data->dev, "%s read %d OUT, status %d\n",
-		data->name, len, value);
+	VDEBUG (data->dev, "%s read %zu OUT, status %d\n",
+		data->name, len, (int) value);
 	if (value >= 0 && copy_to_user (buf, kbuf, value))
 		value = -EFAULT;
 
@@ -465,8 +465,8 @@ ep_write (struct file *fd, const char __user *buf, size_t len, loff_t *ptr)
 	}
 
 	value = ep_io (data, kbuf, len);
-	VDEBUG (data->dev, "%s write %d IN, status %d\n",
-		data->name, len, value);
+	VDEBUG (data->dev, "%s write %zu IN, status %d\n",
+		data->name, len, (int) value);
 free1:
 	up (&data->lock);
 	kfree (kbuf);
@@ -1318,8 +1318,8 @@ gadgetfs_setup (struct usb_gadget *gadget, const struct usb_ctrlrequest *ctrl)
 	struct usb_request		*req = dev->req;
 	int				value = -EOPNOTSUPP;
 	struct usb_gadgetfs_event	*event;
-	u16				w_value = ctrl->wValue;
-	u16				w_length = ctrl->wLength;
+	u16				w_value = le16_to_cpu(ctrl->wValue);
+	u16				w_length = le16_to_cpu(ctrl->wLength);
 
 	spin_lock (&dev->lock);
 	dev->setup_abort = 0;
diff --git a/drivers/usb/gadget/serial.c b/drivers/usb/gadget/serial.c
index 4d591c764e3..9e4f1c6935a 100644
--- a/drivers/usb/gadget/serial.c
+++ b/drivers/usb/gadget/serial.c
@@ -300,18 +300,18 @@ static int gs_build_config_buf(u8 *buf, enum usb_device_speed speed,
 		u8 type, unsigned int index, int is_otg);
 
 static struct usb_request *gs_alloc_req(struct usb_ep *ep, unsigned int len,
-	int kmalloc_flags);
+	unsigned kmalloc_flags);
 static void gs_free_req(struct usb_ep *ep, struct usb_request *req);
 
 static struct gs_req_entry *gs_alloc_req_entry(struct usb_ep *ep, unsigned len,
-	int kmalloc_flags);
+	unsigned kmalloc_flags);
 static void gs_free_req_entry(struct usb_ep *ep, struct gs_req_entry *req);
 
-static int gs_alloc_ports(struct gs_dev *dev, int kmalloc_flags);
+static int gs_alloc_ports(struct gs_dev *dev, unsigned kmalloc_flags);
 static void gs_free_ports(struct gs_dev *dev);
 
 /* circular buffer */
-static struct gs_buf *gs_buf_alloc(unsigned int size, int kmalloc_flags);
+static struct gs_buf *gs_buf_alloc(unsigned int size, unsigned kmalloc_flags);
 static void gs_buf_free(struct gs_buf *gb);
 static void gs_buf_clear(struct gs_buf *gb);
 static unsigned int gs_buf_data_avail(struct gs_buf *gb);
@@ -1607,9 +1607,9 @@ static int gs_setup(struct usb_gadget *gadget,
 	int ret = -EOPNOTSUPP;
 	struct gs_dev *dev = get_gadget_data(gadget);
 	struct usb_request *req = dev->dev_ctrl_req;
-	u16 wIndex = ctrl->wIndex;
-	u16 wValue = ctrl->wValue;
-	u16 wLength = ctrl->wLength;
+	u16 wIndex = le16_to_cpu(ctrl->wIndex);
+	u16 wValue = le16_to_cpu(ctrl->wValue);
+	u16 wLength = le16_to_cpu(ctrl->wLength);
 
 	switch (ctrl->bRequestType & USB_TYPE_MASK) {
 	case USB_TYPE_STANDARD:
@@ -1651,9 +1651,9 @@ static int gs_setup_standard(struct usb_gadget *gadget,
 	int ret = -EOPNOTSUPP;
 	struct gs_dev *dev = get_gadget_data(gadget);
 	struct usb_request *req = dev->dev_ctrl_req;
-	u16 wIndex = ctrl->wIndex;
-	u16 wValue = ctrl->wValue;
-	u16 wLength = ctrl->wLength;
+	u16 wIndex = le16_to_cpu(ctrl->wIndex);
+	u16 wValue = le16_to_cpu(ctrl->wValue);
+	u16 wLength = le16_to_cpu(ctrl->wLength);
 
 	switch (ctrl->bRequest) {
 	case USB_REQ_GET_DESCRIPTOR:
@@ -1782,9 +1782,9 @@ static int gs_setup_class(struct usb_gadget *gadget,
 	struct gs_dev *dev = get_gadget_data(gadget);
 	struct gs_port *port = dev->dev_port[0];	/* ACM only has one port */
 	struct usb_request *req = dev->dev_ctrl_req;
-	u16 wIndex = ctrl->wIndex;
-	u16 wValue = ctrl->wValue;
-	u16 wLength = ctrl->wLength;
+	u16 wIndex = le16_to_cpu(ctrl->wIndex);
+	u16 wValue = le16_to_cpu(ctrl->wValue);
+	u16 wLength = le16_to_cpu(ctrl->wLength);
 
 	switch (ctrl->bRequest) {
 	case USB_CDC_REQ_SET_LINE_CODING:
@@ -2119,7 +2119,8 @@ static int gs_build_config_buf(u8 *buf, enum usb_device_speed speed,
  * Allocate a usb_request and its buffer.  Returns a pointer to the
  * usb_request or NULL if there is an error.
  */
-static struct usb_request *gs_alloc_req(struct usb_ep *ep, unsigned int len, int kmalloc_flags)
+static struct usb_request *
+gs_alloc_req(struct usb_ep *ep, unsigned int len, unsigned kmalloc_flags)
 {
 	struct usb_request *req;
 
@@ -2159,7 +2160,8 @@ static void gs_free_req(struct usb_ep *ep, struct usb_request *req)
  * Allocates a request and its buffer, using the given
  * endpoint, buffer len, and kmalloc flags.
  */
-static struct gs_req_entry *gs_alloc_req_entry(struct usb_ep *ep, unsigned len, int kmalloc_flags)
+static struct gs_req_entry *
+gs_alloc_req_entry(struct usb_ep *ep, unsigned len, unsigned kmalloc_flags)
 {
 	struct gs_req_entry	*req;
 
@@ -2200,7 +2202,7 @@ static void gs_free_req_entry(struct usb_ep *ep, struct gs_req_entry *req)
  *
  * The device lock is normally held when calling this function.
  */
-static int gs_alloc_ports(struct gs_dev *dev, int kmalloc_flags)
+static int gs_alloc_ports(struct gs_dev *dev, unsigned kmalloc_flags)
 {
 	int i;
 	struct gs_port *port;
@@ -2282,7 +2284,7 @@ static void gs_free_ports(struct gs_dev *dev)
  *
  * Allocate a circular buffer and all associated memory.
  */
-static struct gs_buf *gs_buf_alloc(unsigned int size, int kmalloc_flags)
+static struct gs_buf *gs_buf_alloc(unsigned int size, unsigned kmalloc_flags)
 {
 	struct gs_buf *gb;
 
diff --git a/drivers/usb/gadget/zero.c b/drivers/usb/gadget/zero.c
index 6e49432071a..a6e035e2447 100644
--- a/drivers/usb/gadget/zero.c
+++ b/drivers/usb/gadget/zero.c
@@ -919,9 +919,9 @@ zero_setup (struct usb_gadget *gadget, const struct usb_ctrlrequest *ctrl)
 	struct zero_dev		*dev = get_gadget_data (gadget);
 	struct usb_request	*req = dev->req;
 	int			value = -EOPNOTSUPP;
-	u16			w_index = ctrl->wIndex;
-	u16			w_value = ctrl->wValue;
-	u16			w_length = ctrl->wLength;
+	u16			w_index = le16_to_cpu(ctrl->wIndex);
+	u16			w_value = le16_to_cpu(ctrl->wValue);
+	u16			w_length = le16_to_cpu(ctrl->wLength);
 
 	/* usually this stores reply data in the pre-allocated ep0 buffer,
 	 * but config change events will reconfigure hardware.
diff --git a/include/linux/usb_gadget.h b/include/linux/usb_gadget.h
index 9bba9997947..b00f127cb44 100644
--- a/include/linux/usb_gadget.h
+++ b/include/linux/usb_gadget.h
@@ -711,7 +711,7 @@ usb_gadget_disconnect (struct usb_gadget *gadget)
  * 	the hardware level driver. Most calls must be handled by
  * 	the gadget driver, including descriptor and configuration
  * 	management.  The 16 bit members of the setup data are in
- * 	cpu order. Called in_interrupt; this may not sleep.  Driver
+ * 	USB byte order. Called in_interrupt; this may not sleep.  Driver
  *	queues a response to ep0, or returns negative to stall.
  * @disconnect: Invoked after all transfers have been stopped,
  * 	when the host is disconnected.  May be called in_interrupt; this
-- 
cgit v1.2.3-70-g09d2


From 5da0106f0b9b13afa4a902c01d4c98b002df55ff Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Tue, 31 May 2005 10:21:11 -0700
Subject: [PATCH] USB: wireless usb <linux/usb_ch9.h> declarations

This provides declarations for new requests, descriptors, and bitfields as
defined in the Wireless USB 1.0 spec.  Device support will involve a new
"Wire Adapter" device class, connecting a USB Host to a cluster of wireless
USB devices.  There will be two adapter types:

  * Host Wireless Adapter (HWA):  the downstream link is wireless, which
    connects a wireless USB host to wireless USB devices (not unlike like
    a hub) including to the second type of adapter.

  * Device Wireless Adapter (DWA): the upstream link is wireless, for
    connecting existing USB devices through wired links into the cluser.

All wireless USB devices will need persistent (and secure!) key storage, and
it's probable that Linux -- or device firmware -- will need to be involved
with that to bootstrap the initial secure key exchange.

Some user interface is required in that initial key exchange, and since the
most "hands-off" one is a wired USB link, I suspect wireless operation will
usually not be the only mode for wireless USB devices.  (Plus, devices can
recharge batteries using wired USB...)  All other key exchange protocols need
error prone user interactions, like copying and/or verifying keys.

It'll likely be a while before we have commercial Wireless USB hardware,
much less Linux implementations that know how to use it.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/usb_ch9.h | 183 ++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 176 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/usb_ch9.h b/include/linux/usb_ch9.h
index f5fe94e09a0..39e7ff4ffd2 100644
--- a/include/linux/usb_ch9.h
+++ b/include/linux/usb_ch9.h
@@ -6,11 +6,14 @@
  *
  * - the master/host side Linux-USB kernel driver API;
  * - the "usbfs" user space API; and
- * - (eventually) a Linux "gadget" slave/device side driver API.
+ * - the Linux "gadget" slave/device/peripheral side driver API.
  *
  * USB 2.0 adds an additional "On The Go" (OTG) mode, which lets systems
  * act either as a USB master/host or as a USB slave/device.  That means
- * the master and slave side APIs will benefit from working well together.
+ * the master and slave side APIs benefit from working well together.
+ *
+ * There's also "Wireless USB", using low power short range radios for
+ * peripheral interconnection but otherwise building on the USB framework.
  */
 
 #ifndef __LINUX_USB_CH9_H
@@ -68,6 +71,18 @@
 #define USB_REQ_SET_INTERFACE		0x0B
 #define USB_REQ_SYNCH_FRAME		0x0C
 
+#define USB_REQ_SET_ENCRYPTION		0x0D	/* Wireless USB */
+#define USB_REQ_GET_ENCRYPTION		0x0E
+#define USB_REQ_SET_HANDSHAKE		0x0F
+#define USB_REQ_GET_HANDSHAKE		0x10
+#define USB_REQ_SET_CONNECTION		0x11
+#define USB_REQ_SET_SECURITY_DATA	0x12
+#define USB_REQ_GET_SECURITY_DATA	0x13
+#define USB_REQ_SET_WUSB_DATA		0x14
+#define USB_REQ_LOOPBACK_DATA_WRITE	0x15
+#define USB_REQ_LOOPBACK_DATA_READ	0x16
+#define USB_REQ_SET_INTERFACE_DS	0x17
+
 /*
  * USB feature flags are written using USB_REQ_{CLEAR,SET}_FEATURE, and
  * are read as a bit array returned by USB_REQ_GET_STATUS.  (So there
@@ -75,10 +90,12 @@
  */
 #define USB_DEVICE_SELF_POWERED		0	/* (read only) */
 #define USB_DEVICE_REMOTE_WAKEUP	1	/* dev may initiate wakeup */
-#define USB_DEVICE_TEST_MODE		2	/* (high speed only) */
-#define USB_DEVICE_B_HNP_ENABLE		3	/* dev may initiate HNP */
-#define USB_DEVICE_A_HNP_SUPPORT	4	/* RH port supports HNP */
-#define USB_DEVICE_A_ALT_HNP_SUPPORT	5	/* other RH port does */
+#define USB_DEVICE_TEST_MODE		2	/* (wired high speed only) */
+#define USB_DEVICE_BATTERY		2	/* (wireless) */
+#define USB_DEVICE_B_HNP_ENABLE		3	/* (otg) dev may initiate HNP */
+#define USB_DEVICE_WUSB_DEVICE		3	/* (wireless)*/
+#define USB_DEVICE_A_HNP_SUPPORT	4	/* (otg) RH port supports HNP */
+#define USB_DEVICE_A_ALT_HNP_SUPPORT	5	/* (otg) other RH port does */
 #define USB_DEVICE_DEBUG_MODE		6	/* (special devices only) */
 
 #define USB_ENDPOINT_HALT		0	/* IN/OUT will STALL */
@@ -135,6 +152,13 @@ struct usb_ctrlrequest {
 #define USB_DT_OTG			0x09
 #define USB_DT_DEBUG			0x0a
 #define USB_DT_INTERFACE_ASSOCIATION	0x0b
+/* these are from the Wireless USB spec */
+#define USB_DT_SECURITY			0x0c
+#define USB_DT_KEY			0x0d
+#define USB_DT_ENCRYPTION_TYPE		0x0e
+#define USB_DT_BOS			0x0f
+#define USB_DT_DEVICE_CAPABILITY	0x10
+#define USB_DT_WIRELESS_ENDPOINT_COMP	0x11
 
 /* conventional codes for class-specific descriptors */
 #define USB_DT_CS_DEVICE		0x21
@@ -192,6 +216,7 @@ struct usb_device_descriptor {
 #define USB_CLASS_CSCID			0x0b	/* chip+ smart card */
 #define USB_CLASS_CONTENT_SEC		0x0d	/* content security */
 #define USB_CLASS_VIDEO			0x0e
+#define USB_CLASS_WIRELESS_CONTROLLER	0xe0
 #define USB_CLASS_APP_SPEC		0xfe
 #define USB_CLASS_VENDOR_SPEC		0xff
 
@@ -223,6 +248,7 @@ struct usb_config_descriptor {
 #define USB_CONFIG_ATT_ONE		(1 << 7)	/* must be set */
 #define USB_CONFIG_ATT_SELFPOWER	(1 << 6)	/* self powered */
 #define USB_CONFIG_ATT_WAKEUP		(1 << 5)	/* can wakeup */
+#define USB_CONFIG_ATT_BATTERY		(1 << 4)	/* battery powered */
 
 /*-------------------------------------------------------------------------*/
 
@@ -289,6 +315,7 @@ struct usb_endpoint_descriptor {
 #define USB_ENDPOINT_XFER_ISOC		1
 #define USB_ENDPOINT_XFER_BULK		2
 #define USB_ENDPOINT_XFER_INT		3
+#define USB_ENDPOINT_MAX_ADJUSTABLE	0x80
 
 
 /*-------------------------------------------------------------------------*/
@@ -350,6 +377,147 @@ struct usb_interface_assoc_descriptor {
 } __attribute__ ((packed));
 
 
+/*-------------------------------------------------------------------------*/
+
+/* USB_DT_SECURITY:  group of wireless security descriptors, including
+ * encryption types available for setting up a CC/association.
+ */
+struct usb_security_descriptor {
+	__u8  bLength;
+	__u8  bDescriptorType;
+
+	__le16 wTotalLength;
+	__u8  bNumEncryptionTypes;
+};
+
+/*-------------------------------------------------------------------------*/
+
+/* USB_DT_KEY:  used with {GET,SET}_SECURITY_DATA; only public keys
+ * may be retrieved.
+ */
+struct usb_key_descriptor {
+	__u8  bLength;
+	__u8  bDescriptorType;
+
+	__u8  tTKID[3];
+	__u8  bReserved;
+	__u8  bKeyData[0];
+};
+
+/*-------------------------------------------------------------------------*/
+
+/* USB_DT_ENCRYPTION_TYPE:  bundled in DT_SECURITY groups */
+struct usb_encryption_descriptor {
+	__u8  bLength;
+	__u8  bDescriptorType;
+
+	__u8  bEncryptionType;
+#define	USB_ENC_TYPE_UNSECURE		0
+#define	USB_ENC_TYPE_WIRED		1	/* non-wireless mode */
+#define	USB_ENC_TYPE_CCM_1		2	/* aes128/cbc session */
+#define	USB_ENC_TYPE_RSA_1		3	/* rsa3072/sha1 auth */
+	__u8  bEncryptionValue;		/* use in SET_ENCRYPTION */
+	__u8  bAuthKeyIndex;
+};
+
+
+/*-------------------------------------------------------------------------*/
+
+/* USB_DT_BOS:  group of wireless capabilities */
+struct usb_bos_descriptor {
+	__u8  bLength;
+	__u8  bDescriptorType;
+
+	__le16 wTotalLength;
+	__u8  bNumDeviceCaps;
+};
+
+/*-------------------------------------------------------------------------*/
+
+/* USB_DT_DEVICE_CAPABILITY:  grouped with BOS */
+struct usb_dev_cap_header {
+	__u8  bLength;
+	__u8  bDescriptorType;
+	__u8  bDevCapabilityType;
+};
+
+#define	USB_CAP_TYPE_WIRELESS_USB	1
+
+struct usb_wireless_cap_descriptor {	/* Ultra Wide Band */
+	__u8  bLength;
+	__u8  bDescriptorType;
+	__u8  bDevCapabilityType;
+
+	__u8  bmAttributes;
+#define	USB_WIRELESS_P2P_DRD		(1 << 1)
+#define	USB_WIRELESS_BEACON_MASK	(3 << 2)
+#define	USB_WIRELESS_BEACON_SELF	(1 << 2)
+#define	USB_WIRELESS_BEACON_DIRECTED	(2 << 2)
+#define	USB_WIRELESS_BEACON_NONE	(3 << 2)
+	__le16 wPHYRates;	/* bit rates, Mbps */
+#define	USB_WIRELESS_PHY_53		(1 << 0)	/* always set */
+#define	USB_WIRELESS_PHY_80		(1 << 1)
+#define	USB_WIRELESS_PHY_107		(1 << 2)	/* always set */
+#define	USB_WIRELESS_PHY_160		(1 << 3)
+#define	USB_WIRELESS_PHY_200		(1 << 4)	/* always set */
+#define	USB_WIRELESS_PHY_320		(1 << 5)
+#define	USB_WIRELESS_PHY_400		(1 << 6)
+#define	USB_WIRELESS_PHY_480		(1 << 7)
+	__u8  bmTFITXPowerInfo;	/* TFI power levels */
+	__u8  bmFFITXPowerInfo;	/* FFI power levels */
+	__le16 bmBandGroup;
+	__u8  bReserved;
+};
+
+/*-------------------------------------------------------------------------*/
+
+/* USB_DT_WIRELESS_ENDPOINT_COMP:  companion descriptor associated with
+ * each endpoint descriptor for a wireless device
+ */
+struct usb_wireless_ep_comp_descriptor {
+	__u8  bLength;
+	__u8  bDescriptorType;
+
+	__u8  bMaxBurst;
+	__u8  bMaxSequence;
+	__le16 wMaxStreamDelay;
+	__le16 wOverTheAirPacketSize;
+	__u8  bOverTheAirInterval;
+	__u8  bmCompAttributes;
+#define USB_ENDPOINT_SWITCH_MASK	0x03	/* in bmCompAttributes */
+#define USB_ENDPOINT_SWITCH_NO		0
+#define USB_ENDPOINT_SWITCH_SWITCH	1
+#define USB_ENDPOINT_SWITCH_SCALE	2
+};
+
+/*-------------------------------------------------------------------------*/
+
+/* USB_REQ_SET_HANDSHAKE is a four-way handshake used between a wireless
+ * host and a device for connection set up, mutual authentication, and
+ * exchanging short lived session keys.  The handshake depends on a CC.
+ */
+struct usb_handshake {
+	__u8 bMessageNumber;
+	__u8 bStatus;
+	__u8 tTKID[3];
+	__u8 bReserved;
+	__u8 CDID[16];
+	__u8 nonce[16];
+	__u8 MIC[8];
+};
+
+/*-------------------------------------------------------------------------*/
+
+/* USB_REQ_SET_CONNECTION modifies or revokes a connection context (CC).
+ * A CC may also be set up using non-wireless secure channels (including
+ * wired USB!), and some devices may support CCs with multiple hosts.
+ */
+struct usb_connection_context {
+	__u8 CHID[16];		/* persistent host id */
+	__u8 CDID[16];		/* device id (unique w/in host context) */
+	__u8 CK[16];		/* connection key */
+};
+
 /*-------------------------------------------------------------------------*/
 
 /* USB 2.0 defines three speeds, here's how Linux identifies them */
@@ -357,7 +525,8 @@ struct usb_interface_assoc_descriptor {
 enum usb_device_speed {
 	USB_SPEED_UNKNOWN = 0,			/* enumerating */
 	USB_SPEED_LOW, USB_SPEED_FULL,		/* usb 1.1 */
-	USB_SPEED_HIGH				/* usb 2.0 */
+	USB_SPEED_HIGH,				/* usb 2.0 */
+	USB_SPEED_VARIABLE,			/* wireless (usb 2.5) */
 };
 
 enum usb_device_state {
-- 
cgit v1.2.3-70-g09d2


From e4ee69c8c1e7ff9790fbce29c7be50db57323a6f Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Mon, 27 Jun 2005 14:36:32 -0700
Subject: [PATCH] ppc32: Bump PMU interrupt priority

The Power Management Unit on PowerMacs is very sensitive to timeouts during
async message exchanges.  It uses rather crude protocol based on a shift
register with an interrupt and is almost continuously exchanging messages with
the host CPU on laptops.

This patch adds a routine to the open_pic driver to be able to select a PMU
driver so that it bumps it's interrupt priority to above the normal level.

This will allow PMU interrupts to occur while another interrupt is pending,
and thus reduce the risk of machine beeing abruptly shutdown by the PMU due to
a timeout in PMU communication caused by excessive interrupt latency.  The
problem is very rare, and usually just doesn't happen, but it is still useful
to make things even more robust.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/syslib/open_pic.c  | 26 +++++++++++++++++++-------
 drivers/macintosh/via-pmu.c | 10 ++++++++++
 include/asm-ppc/open_pic.h  |  6 ++++++
 3 files changed, 35 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/ppc/syslib/open_pic.c b/arch/ppc/syslib/open_pic.c
index b45d8268bf9..ad39b86ca92 100644
--- a/arch/ppc/syslib/open_pic.c
+++ b/arch/ppc/syslib/open_pic.c
@@ -370,8 +370,9 @@ void __init openpic_init(int offset)
 	/* Initialize IPI interrupts */
 	if ( ppc_md.progress ) ppc_md.progress("openpic: ipi",0x3bb);
 	for (i = 0; i < OPENPIC_NUM_IPI; i++) {
-		/* Disabled, Priority 10..13 */
-		openpic_initipi(i, 10+i, OPENPIC_VEC_IPI+i+offset);
+		/* Disabled, increased priorities 10..13 */
+		openpic_initipi(i, OPENPIC_PRIORITY_IPI_BASE+i,
+				OPENPIC_VEC_IPI+i+offset);
 		/* IPIs are per-CPU */
 		irq_desc[OPENPIC_VEC_IPI+i+offset].status |= IRQ_PER_CPU;
 		irq_desc[OPENPIC_VEC_IPI+i+offset].handler = &open_pic_ipi;
@@ -399,8 +400,9 @@ void __init openpic_init(int offset)
 		if (sense & IRQ_SENSE_MASK)
 			irq_desc[i+offset].status = IRQ_LEVEL;
 
-		/* Enabled, Priority 8 */
-		openpic_initirq(i, 8, i+offset, (sense & IRQ_POLARITY_MASK),
+		/* Enabled, Default priority */
+		openpic_initirq(i, OPENPIC_PRIORITY_DEFAULT, i+offset,
+				(sense & IRQ_POLARITY_MASK),
 				(sense & IRQ_SENSE_MASK));
 		/* Processor 0 */
 		openpic_mapirq(i, CPU_MASK_CPU0, CPU_MASK_NONE);
@@ -655,6 +657,18 @@ static void __init openpic_maptimer(u_int timer, cpumask_t cpumask)
 		      cpus_addr(phys)[0]);
 }
 
+/*
+ * Change the priority of an interrupt
+ */
+void __init
+openpic_set_irq_priority(u_int irq, u_int pri)
+{
+	check_arg_irq(irq);
+	openpic_safe_writefield(&ISR[irq - open_pic_irq_offset]->Vector_Priority,
+				OPENPIC_PRIORITY_MASK,
+				pri << OPENPIC_PRIORITY_SHIFT);
+}
+
 /*
  * Initalize the interrupt source which will generate an NMI.
  * This raises the interrupt's priority from 8 to 9.
@@ -665,9 +679,7 @@ void __init
 openpic_init_nmi_irq(u_int irq)
 {
 	check_arg_irq(irq);
-	openpic_safe_writefield(&ISR[irq - open_pic_irq_offset]->Vector_Priority,
-				OPENPIC_PRIORITY_MASK,
-				9 << OPENPIC_PRIORITY_SHIFT);
+	openpic_set_irq_priority(irq, OPENPIC_PRIORITY_NMI);
 }
 
 /*
diff --git a/drivers/macintosh/via-pmu.c b/drivers/macintosh/via-pmu.c
index b941ee22099..5375df03c6f 100644
--- a/drivers/macintosh/via-pmu.c
+++ b/drivers/macintosh/via-pmu.c
@@ -63,6 +63,10 @@
 #include <asm/backlight.h>
 #endif
 
+#ifdef CONFIG_PPC32
+#include <asm/open_pic.h>
+#endif
+
 /* Some compile options */
 #undef SUSPEND_USES_PMU
 #define DEBUG_SLEEP
@@ -407,6 +411,12 @@ static int __init via_pmu_start(void)
 	batt_req.complete = 1;
 #endif
 
+#ifdef CONFIG_PPC32
+	if (pmu_kind == PMU_KEYLARGO_BASED)
+		openpic_set_irq_priority(vias->intrs[0].line,
+					 OPENPIC_PRIORITY_DEFAULT + 1);
+#endif
+
 	if (request_irq(vias->intrs[0].line, via_pmu_interrupt, 0, "VIA-PMU",
 			(void *)0)) {
 		printk(KERN_ERR "VIA-PMU: can't get irq %d\n",
diff --git a/include/asm-ppc/open_pic.h b/include/asm-ppc/open_pic.h
index dbe85331974..7848aa610c0 100644
--- a/include/asm-ppc/open_pic.h
+++ b/include/asm-ppc/open_pic.h
@@ -25,6 +25,11 @@
 #define OPENPIC_VEC_IPI		118	/* and up */
 #define OPENPIC_VEC_SPURIOUS	255
 
+/* Priorities */
+#define OPENPIC_PRIORITY_IPI_BASE	10
+#define OPENPIC_PRIORITY_DEFAULT	4
+#define OPENPIC_PRIORITY_NMI		9
+
 /* OpenPIC IRQ controller structure */
 extern struct hw_interrupt_type open_pic;
 
@@ -42,6 +47,7 @@ extern int epic_serial_mode;
 extern void openpic_set_sources(int first_irq, int num_irqs, void __iomem *isr);
 extern void openpic_init(int linux_irq_offset);
 extern void openpic_init_nmi_irq(u_int irq);
+extern void openpic_set_irq_priority(u_int irq, u_int pri);
 extern void openpic_hookup_cascade(u_int irq, char *name,
 				   int (*cascade_fn)(struct pt_regs *));
 extern u_int openpic_irq(void);
-- 
cgit v1.2.3-70-g09d2


From 8c8709334cec803368a432a33e0f2e116d48fe07 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Mon, 27 Jun 2005 14:36:34 -0700
Subject: [PATCH] ppc32: Remove CONFIG_PMAC_PBOOK

This patch removes CONFIG_PMAC_PBOOK (PowerBook support).  This is now
split into CONFIG_PMAC_MEDIABAY for the actual hotswap bay that some
powerbooks have, CONFIG_PM for power management related code, and just left
out of any CONFIG_* option for some generally useful stuff that can be used
on non-laptops as well.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/platforms/pmac_sleep.S     |   4 +-
 arch/ppc/platforms/pmac_time.c      |   8 +-
 drivers/block/swim3.c               |  10 +-
 drivers/char/misc.c                 |   3 -
 drivers/ide/ppc/pmac.c              |   8 +-
 drivers/ieee1394/ohci1394.c         |  10 +-
 drivers/macintosh/Kconfig           |  35 ++-----
 drivers/macintosh/Makefile          |   2 +-
 drivers/macintosh/adb.c             |  10 +-
 drivers/macintosh/via-pmu.c         |  70 ++++++--------
 drivers/usb/host/ohci-pci.c         |  13 +--
 drivers/video/aty/aty128fb.c        |  14 ---
 drivers/video/chipsfb.c             | 176 +++++++++++++++++++-----------------
 include/linux/pmu.h                 |   6 +-
 sound/oss/dmasound/dmasound_awacs.c |  14 +--
 sound/ppc/awacs.c                   |   8 +-
 sound/ppc/daca.c                    |   6 +-
 sound/ppc/pmac.c                    |  11 ++-
 sound/ppc/pmac.h                    |   2 +-
 sound/ppc/tumbler.c                 |   4 +-
 20 files changed, 193 insertions(+), 221 deletions(-)

(limited to 'include')

diff --git a/arch/ppc/platforms/pmac_sleep.S b/arch/ppc/platforms/pmac_sleep.S
index f459ade1bd6..016a7464915 100644
--- a/arch/ppc/platforms/pmac_sleep.S
+++ b/arch/ppc/platforms/pmac_sleep.S
@@ -46,7 +46,7 @@
 	.section .text
 	.align	5
 
-#if defined(CONFIG_PMAC_PBOOK) || defined(CONFIG_CPU_FREQ_PMAC)
+#if defined(CONFIG_PM) || defined(CONFIG_CPU_FREQ_PMAC)
 
 /* This gets called by via-pmu.c late during the sleep process.
  * The PMU was already send the sleep command and will shut us down
@@ -382,7 +382,7 @@ turn_on_mmu:
 	isync
 	rfi
 
-#endif /* defined(CONFIG_PMAC_PBOOK) || defined(CONFIG_CPU_FREQ) */
+#endif /* defined(CONFIG_PM) || defined(CONFIG_CPU_FREQ) */
 
 	.section .data
 	.balign	L1_CACHE_LINE_SIZE
diff --git a/arch/ppc/platforms/pmac_time.c b/arch/ppc/platforms/pmac_time.c
index de60ccc7db9..778ce4fec36 100644
--- a/arch/ppc/platforms/pmac_time.c
+++ b/arch/ppc/platforms/pmac_time.c
@@ -206,7 +206,7 @@ via_calibrate_decr(void)
 	return 1;
 }
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 /*
  * Reset the time after a sleep.
  */
@@ -238,7 +238,7 @@ time_sleep_notify(struct pmu_sleep_notifier *self, int when)
 static struct pmu_sleep_notifier time_sleep_notifier __pmacdata = {
 	time_sleep_notify, SLEEP_LEVEL_MISC,
 };
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
 
 /*
  * Query the OF and get the decr frequency.
@@ -251,9 +251,9 @@ pmac_calibrate_decr(void)
 	struct device_node *cpu;
 	unsigned int freq, *fp;
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 	pmu_register_sleep_notifier(&time_sleep_notifier);
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
 
 	/* We assume MacRISC2 machines have correct device-tree
 	 * calibration. That's better since the VIA itself seems
diff --git a/drivers/block/swim3.c b/drivers/block/swim3.c
index 5b09cf154ac..e5f7494c00e 100644
--- a/drivers/block/swim3.c
+++ b/drivers/block/swim3.c
@@ -253,7 +253,7 @@ static int floppy_revalidate(struct gendisk *disk);
 static int swim3_add_device(struct device_node *swims);
 int swim3_init(void);
 
-#ifndef CONFIG_PMAC_PBOOK
+#ifndef CONFIG_PMAC_MEDIABAY
 #define check_media_bay(which, what)	1
 #endif
 
@@ -297,9 +297,11 @@ static void do_fd_request(request_queue_t * q)
 	int i;
 	for(i=0;i<floppy_count;i++)
 	{
+#ifdef CONFIG_PMAC_MEDIABAY
 		if (floppy_states[i].media_bay &&
 			check_media_bay(floppy_states[i].media_bay, MB_FD))
 			continue;
+#endif /* CONFIG_PMAC_MEDIABAY */
 		start_request(&floppy_states[i]);
 	}
 	sti();
@@ -856,8 +858,10 @@ static int floppy_ioctl(struct inode *inode, struct file *filp,
 	if ((cmd & 0x80) && !capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
+#ifdef CONFIG_PMAC_MEDIABAY
 	if (fs->media_bay && check_media_bay(fs->media_bay, MB_FD))
 		return -ENXIO;
+#endif
 
 	switch (cmd) {
 	case FDEJECT:
@@ -881,8 +885,10 @@ static int floppy_open(struct inode *inode, struct file *filp)
 	int n, err = 0;
 
 	if (fs->ref_count == 0) {
+#ifdef CONFIG_PMAC_MEDIABAY
 		if (fs->media_bay && check_media_bay(fs->media_bay, MB_FD))
 			return -ENXIO;
+#endif
 		out_8(&sw->setup, S_IBM_DRIVE | S_FCLK_DIV2);
 		out_8(&sw->control_bic, 0xff);
 		out_8(&sw->mode, 0x95);
@@ -967,8 +973,10 @@ static int floppy_revalidate(struct gendisk *disk)
 	struct swim3 __iomem *sw;
 	int ret, n;
 
+#ifdef CONFIG_PMAC_MEDIABAY
 	if (fs->media_bay && check_media_bay(fs->media_bay, MB_FD))
 		return -ENXIO;
+#endif
 
 	sw = fs->swim3;
 	grab_drive(fs, revalidating, 0);
diff --git a/drivers/char/misc.c b/drivers/char/misc.c
index 31cf84d6902..931efd58f87 100644
--- a/drivers/char/misc.c
+++ b/drivers/char/misc.c
@@ -308,9 +308,6 @@ static int __init misc_init(void)
 #endif
 #ifdef CONFIG_BVME6000
 	rtc_DP8570A_init();
-#endif
-#ifdef CONFIG_PMAC_PBOOK
-	pmu_device_init();
 #endif
 	if (register_chrdev(MISC_MAJOR,"misc",&misc_fops)) {
 		printk("unable to get major %d for misc devices\n",
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index 569f1676744..818380b5fd2 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -1324,9 +1324,9 @@ pmac_ide_setup_device(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif)
 	/* XXX FIXME: Media bay stuff need re-organizing */
 	if (np->parent && np->parent->name
 	    && strcasecmp(np->parent->name, "media-bay") == 0) {
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PMAC_MEDIABAY
 		media_bay_set_ide_infos(np->parent, pmif->regbase, pmif->irq, hwif->index);
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PMAC_MEDIABAY */
 		pmif->mediabay = 1;
 		if (!bidp)
 			pmif->aapl_bus_id = 1;
@@ -1382,10 +1382,10 @@ pmac_ide_setup_device(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif)
 	       hwif->index, model_name[pmif->kind], pmif->aapl_bus_id,
 	       pmif->mediabay ? " (mediabay)" : "", hwif->irq);
 			
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PMAC_MEDIABAY
 	if (pmif->mediabay && check_media_bay_by_base(pmif->regbase, MB_CD) == 0)
 		hwif->noprobe = 0;
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PMAC_MEDIABAY */
 
 	hwif->sg_max_nents = MAX_DCMDS;
 
diff --git a/drivers/ieee1394/ohci1394.c b/drivers/ieee1394/ohci1394.c
index 36e25ac823d..b3d3d22fde6 100644
--- a/drivers/ieee1394/ohci1394.c
+++ b/drivers/ieee1394/ohci1394.c
@@ -3538,8 +3538,8 @@ static void ohci1394_pci_remove(struct pci_dev *pdev)
 
 static int ohci1394_pci_resume (struct pci_dev *pdev)
 {
-#ifdef CONFIG_PMAC_PBOOK
-	{
+#ifdef CONFIG_PPC_PMAC
+	if (_machine == _MACH_Pmac) {
 		struct device_node *of_node;
 
 		/* Re-enable 1394 */
@@ -3547,7 +3547,7 @@ static int ohci1394_pci_resume (struct pci_dev *pdev)
 		if (of_node)
 			pmac_call_feature (PMAC_FTR_1394_ENABLE, of_node, 0, 1);
 	}
-#endif
+#endif /* CONFIG_PPC_PMAC */
 
 	pci_enable_device(pdev);
 
@@ -3557,8 +3557,8 @@ static int ohci1394_pci_resume (struct pci_dev *pdev)
 
 static int ohci1394_pci_suspend (struct pci_dev *pdev, pm_message_t state)
 {
-#ifdef CONFIG_PMAC_PBOOK
-	{
+#ifdef CONFIG_PPC_PMAC
+	if (_machine == _MACH_Pmac) {
 		struct device_node *of_node;
 
 		/* Disable 1394 */
diff --git a/drivers/macintosh/Kconfig b/drivers/macintosh/Kconfig
index b0ace5bc950..91691a6c004 100644
--- a/drivers/macintosh/Kconfig
+++ b/drivers/macintosh/Kconfig
@@ -86,33 +86,18 @@ config PMAC_SMU
 	  on the "SMU" system control chip which replaces the old PMU.
 	  If you don't know, say Y.
 
-config PMAC_PBOOK
-	bool "Power management support for PowerBooks"
-	depends on ADB_PMU
-	---help---
-	  This provides support for putting a PowerBook to sleep; it also
-	  enables media bay support.  Power management works on the
-	  PB2400/3400/3500, Wallstreet, Lombard, and Bronze PowerBook G3 and
-	  the Titanium Powerbook G4, as well as the iBooks.  You should get
-	  the power management daemon, pmud, to make it work and you must have
-	  the /dev/pmu device (see the pmud README).
-
-	  Get pmud from <ftp://ftp.samba.org/pub/ppclinux/pmud/>.
-
-	  If you have a PowerBook, you should say Y here.
-
-	  You may also want to compile the dma sound driver as a module and
-	  have it autoloaded. The act of removing the module shuts down the
-	  sound hardware for more power savings.
-
-config PM
-	bool
-	depends on PPC_PMAC && ADB_PMU && PMAC_PBOOK
-	default y
-
 config PMAC_APM_EMU
 	tristate "APM emulation"
-	depends on PMAC_PBOOK
+	depends on PPC_PMAC && PPC32 && PM
+
+config PMAC_MEDIABAY
+	bool "Support PowerBook hotswap media bay"
+	depends on PPC_PMAC && PPC32
+	help
+	  This option adds support for older PowerBook's hotswap media bay
+	  that can contains batteries, floppy drives, or IDE devices. PCI
+	  devices are not fully supported in the bay as I never had one to
+	  try with
 
 # made a separate option since backlight may end up beeing used
 # on non-powerbook machines (but only on PMU based ones AFAIK)
diff --git a/drivers/macintosh/Makefile b/drivers/macintosh/Makefile
index b3f88a4fcef..f5ae171dbfe 100644
--- a/drivers/macintosh/Makefile
+++ b/drivers/macintosh/Makefile
@@ -6,7 +6,7 @@
 
 obj-$(CONFIG_PPC_PMAC)		+= macio_asic.o
 
-obj-$(CONFIG_PMAC_PBOOK)	+= mediabay.o
+obj-$(CONFIG_PMAC_MEDIABAY)	+= mediabay.o
 obj-$(CONFIG_MAC_EMUMOUSEBTN)	+= mac_hid.o
 obj-$(CONFIG_INPUT_ADBHID)	+= adbhid.o
 obj-$(CONFIG_ANSLCD)		+= ans-lcd.o
diff --git a/drivers/macintosh/adb.c b/drivers/macintosh/adb.c
index 493e2afa191..c0dc1e3fa58 100644
--- a/drivers/macintosh/adb.c
+++ b/drivers/macintosh/adb.c
@@ -90,7 +90,7 @@ static int sleepy_trackpad;
 static int autopoll_devs;
 int __adb_probe_sync;
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 static int adb_notify_sleep(struct pmu_sleep_notifier *self, int when);
 static struct pmu_sleep_notifier adb_sleep_notifier = {
 	adb_notify_sleep,
@@ -320,9 +320,9 @@ int __init adb_init(void)
 		printk(KERN_WARNING "Warning: no ADB interface detected\n");
 		adb_controller = NULL;
 	} else {
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 		pmu_register_sleep_notifier(&adb_sleep_notifier);
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
 #ifdef CONFIG_PPC
 		if (machine_is_compatible("AAPL,PowerBook1998") ||
 			machine_is_compatible("PowerBook1,1"))
@@ -337,7 +337,7 @@ int __init adb_init(void)
 
 __initcall(adb_init);
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 /*
  * notify clients before sleep and reset bus afterwards
  */
@@ -378,7 +378,7 @@ adb_notify_sleep(struct pmu_sleep_notifier *self, int when)
 	}
 	return PBOOK_SLEEP_OK;
 }
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
 
 static int
 do_adb_reset_bus(void)
diff --git a/drivers/macintosh/via-pmu.c b/drivers/macintosh/via-pmu.c
index 5375df03c6f..4a0a0ad2d03 100644
--- a/drivers/macintosh/via-pmu.c
+++ b/drivers/macintosh/via-pmu.c
@@ -155,10 +155,10 @@ static spinlock_t pmu_lock;
 static u8 pmu_intr_mask;
 static int pmu_version;
 static int drop_interrupts;
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 static int option_lid_wakeup = 1;
 static int sleep_in_progress;
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
 static unsigned long async_req_locks;
 static unsigned int pmu_irq_stats[11];
 
@@ -168,7 +168,6 @@ static struct proc_dir_entry *proc_pmu_irqstats;
 static struct proc_dir_entry *proc_pmu_options;
 static int option_server_mode;
 
-#ifdef CONFIG_PMAC_PBOOK
 int pmu_battery_count;
 int pmu_cur_battery;
 unsigned int pmu_power_flags;
@@ -176,7 +175,6 @@ struct pmu_battery_info pmu_batteries[PMU_MAX_BATTERIES];
 static int query_batt_timer = BATTERY_POLLING_COUNT;
 static struct adb_request batt_req;
 static struct proc_dir_entry *proc_pmu_batt[PMU_MAX_BATTERIES];
-#endif /* CONFIG_PMAC_PBOOK */
 
 #if defined(CONFIG_INPUT_ADBHID) && defined(CONFIG_PMAC_BACKLIGHT)
 extern int disable_kernel_backlight;
@@ -210,11 +208,9 @@ static int proc_get_irqstats(char *page, char **start, off_t off,
 static int pmu_set_backlight_level(int level, void* data);
 static int pmu_set_backlight_enable(int on, int level, void* data);
 #endif /* CONFIG_PMAC_BACKLIGHT */
-#ifdef CONFIG_PMAC_PBOOK
 static void pmu_pass_intr(unsigned char *data, int len);
 static int proc_get_batt(char *page, char **start, off_t off,
 			int count, int *eof, void *data);
-#endif /* CONFIG_PMAC_PBOOK */
 static int proc_read_options(char *page, char **start, off_t off,
 			int count, int *eof, void *data);
 static int proc_write_options(struct file *file, const char __user *buffer,
@@ -407,9 +403,7 @@ static int __init via_pmu_start(void)
 
 	bright_req_1.complete = 1;
 	bright_req_2.complete = 1;
-#ifdef CONFIG_PMAC_PBOOK
 	batt_req.complete = 1;
-#endif
 
 #ifdef CONFIG_PPC32
 	if (pmu_kind == PMU_KEYLARGO_BASED)
@@ -468,7 +462,7 @@ static int __init via_pmu_dev_init(void)
 	register_backlight_controller(&pmu_backlight_controller, NULL, "pmu");
 #endif /* CONFIG_PMAC_BACKLIGHT */
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PPC32
   	if (machine_is_compatible("AAPL,3400/2400") ||
   		machine_is_compatible("AAPL,3500")) {
 		int mb = pmac_call_feature(PMAC_FTR_GET_MB_INFO,
@@ -496,20 +490,19 @@ static int __init via_pmu_dev_init(void)
 				pmu_batteries[1].flags |= PMU_BATT_TYPE_SMART;
 		}
 	}
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PPC32 */
+
 	/* Create /proc/pmu */
 	proc_pmu_root = proc_mkdir("pmu", NULL);
 	if (proc_pmu_root) {
-#ifdef CONFIG_PMAC_PBOOK
-		int i;
+		long i;
 
 		for (i=0; i<pmu_battery_count; i++) {
 			char title[16];
-			sprintf(title, "battery_%d", i);
+			sprintf(title, "battery_%ld", i);
 			proc_pmu_batt[i] = create_proc_read_entry(title, 0, proc_pmu_root,
 						proc_get_batt, (void *)i);
 		}
-#endif /* CONFIG_PMAC_PBOOK */
 
 		proc_pmu_info = create_proc_read_entry("info", 0, proc_pmu_root,
 					proc_get_info, NULL);
@@ -629,8 +622,6 @@ static void pmu_set_server_mode(int server_mode)
 	pmu_wait_complete(&req);
 }
 
-#ifdef CONFIG_PMAC_PBOOK
-
 /* This new version of the code for 2400/3400/3500 powerbooks
  * is inspired from the implementation in gkrellm-pmu
  */
@@ -813,8 +804,6 @@ query_battery_state(void)
 			2, PMU_SMART_BATTERY_STATE, pmu_cur_battery+1);
 }
 
-#endif /* CONFIG_PMAC_PBOOK */
-
 static int __pmac
 proc_get_info(char *page, char **start, off_t off,
 		int count, int *eof, void *data)
@@ -823,11 +812,9 @@ proc_get_info(char *page, char **start, off_t off,
 
 	p += sprintf(p, "PMU driver version     : %d\n", PMU_DRIVER_VERSION);
 	p += sprintf(p, "PMU firmware version   : %02x\n", pmu_version);
-#ifdef CONFIG_PMAC_PBOOK
 	p += sprintf(p, "AC Power               : %d\n",
 		((pmu_power_flags & PMU_PWR_AC_PRESENT) != 0));
 	p += sprintf(p, "Battery count          : %d\n", pmu_battery_count);
-#endif /* CONFIG_PMAC_PBOOK */
 
 	return p - page;
 }
@@ -859,12 +846,11 @@ proc_get_irqstats(char *page, char **start, off_t off,
 	return p - page;
 }
 
-#ifdef CONFIG_PMAC_PBOOK
 static int __pmac
 proc_get_batt(char *page, char **start, off_t off,
 		int count, int *eof, void *data)
 {
-	int batnum = (int)data;
+	long batnum = (long)data;
 	char *p = page;
 	
 	p += sprintf(p, "\n");
@@ -883,7 +869,6 @@ proc_get_batt(char *page, char **start, off_t off,
 
 	return p - page;
 }
-#endif /* CONFIG_PMAC_PBOOK */
 
 static int __pmac
 proc_read_options(char *page, char **start, off_t off,
@@ -891,11 +876,11 @@ proc_read_options(char *page, char **start, off_t off,
 {
 	char *p = page;
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 	if (pmu_kind == PMU_KEYLARGO_BASED &&
 	    pmac_call_feature(PMAC_FTR_SLEEP_STATE,NULL,0,-1) >= 0)
 		p += sprintf(p, "lid_wakeup=%d\n", option_lid_wakeup);
-#endif /* CONFIG_PMAC_PBOOK */
+#endif
 	if (pmu_kind == PMU_KEYLARGO_BASED)
 		p += sprintf(p, "server_mode=%d\n", option_server_mode);
 
@@ -932,12 +917,12 @@ proc_write_options(struct file *file, const char __user *buffer,
 	*(val++) = 0;
 	while(*val == ' ')
 		val++;
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 	if (pmu_kind == PMU_KEYLARGO_BASED &&
 	    pmac_call_feature(PMAC_FTR_SLEEP_STATE,NULL,0,-1) >= 0)
 		if (!strcmp(label, "lid_wakeup"))
 			option_lid_wakeup = ((*val) == '1');
-#endif /* CONFIG_PMAC_PBOOK */
+#endif
 	if (pmu_kind == PMU_KEYLARGO_BASED && !strcmp(label, "server_mode")) {
 		int new_value;
 		new_value = ((*val) == '1');
@@ -1432,7 +1417,6 @@ next:
 	}
 	/* Tick interrupt */
 	else if ((1 << pirq) & PMU_INT_TICK) {
-#ifdef CONFIG_PMAC_PBOOK
 		/* Environement or tick interrupt, query batteries */
 		if (pmu_battery_count) {
 			if ((--query_batt_timer) == 0) {
@@ -1447,7 +1431,6 @@ next:
 		pmu_pass_intr(data, len);
 	} else {
 	       pmu_pass_intr(data, len);
-#endif /* CONFIG_PMAC_PBOOK */
 	}
 	goto next;
 }
@@ -2062,7 +2045,7 @@ pmu_i2c_simple_write(int bus, int addr,  u8* data, int len)
 	return -1;
 }
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 
 static LIST_HEAD(sleep_notifiers);
 
@@ -2715,6 +2698,8 @@ powerbook_sleep_3400(void)
 	return 0;
 }
 
+#endif /* CONFIG_PM */
+
 /*
  * Support for /dev/pmu device
  */
@@ -2894,11 +2879,11 @@ static int __pmac
 pmu_ioctl(struct inode * inode, struct file *filp,
 		     u_int cmd, u_long arg)
 {
-	struct pmu_private *pp = filp->private_data;
 	__u32 __user *argp = (__u32 __user *)arg;
-	int error;
+	int error = -EINVAL;
 
 	switch (cmd) {
+#ifdef CONFIG_PM
 	case PMU_IOC_SLEEP:
 		if (!capable(CAP_SYS_ADMIN))
 			return -EACCES;
@@ -2920,12 +2905,13 @@ pmu_ioctl(struct inode * inode, struct file *filp,
 			error = -ENOSYS;
 		}
 		sleep_in_progress = 0;
-		return error;
+		break;
 	case PMU_IOC_CAN_SLEEP:
 		if (pmac_call_feature(PMAC_FTR_SLEEP_STATE,NULL,0,-1) < 0)
 			return put_user(0, argp);
 		else
 			return put_user(1, argp);
+#endif /* CONFIG_PM */
 
 #ifdef CONFIG_PMAC_BACKLIGHT
 	/* Backlight should have its own device or go via
@@ -2946,11 +2932,13 @@ pmu_ioctl(struct inode * inode, struct file *filp,
 		error = get_user(value, argp);
 		if (!error)
 			error = set_backlight_level(value);
-		return error;
+		break;
 	}
 #ifdef CONFIG_INPUT_ADBHID
 	case PMU_IOC_GRAB_BACKLIGHT: {
+		struct pmu_private *pp = filp->private_data;
 		unsigned long flags;
+
 		if (pp->backlight_locker)
 			return 0;
 		pp->backlight_locker = 1;
@@ -2966,7 +2954,7 @@ pmu_ioctl(struct inode * inode, struct file *filp,
 	case PMU_IOC_HAS_ADB:
 		return put_user(pmu_has_adb, argp);
 	}
-	return -EINVAL;
+	return error;
 }
 
 static struct file_operations pmu_device_fops __pmacdata = {
@@ -2982,14 +2970,16 @@ static struct miscdevice pmu_device __pmacdata = {
 	PMU_MINOR, "pmu", &pmu_device_fops
 };
 
-void pmu_device_init(void)
+static int pmu_device_init(void)
 {
 	if (!via)
-		return;
+		return 0;
 	if (misc_register(&pmu_device) < 0)
 		printk(KERN_ERR "via-pmu: cannot register misc device.\n");
+	return 0;
 }
-#endif /* CONFIG_PMAC_PBOOK */
+device_initcall(pmu_device_init);
+
 
 #ifdef DEBUG_SLEEP
 static inline void  __pmac
@@ -3157,12 +3147,12 @@ EXPORT_SYMBOL(pmu_i2c_combined_read);
 EXPORT_SYMBOL(pmu_i2c_stdsub_write);
 EXPORT_SYMBOL(pmu_i2c_simple_read);
 EXPORT_SYMBOL(pmu_i2c_simple_write);
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 EXPORT_SYMBOL(pmu_register_sleep_notifier);
 EXPORT_SYMBOL(pmu_unregister_sleep_notifier);
 EXPORT_SYMBOL(pmu_enable_irled);
 EXPORT_SYMBOL(pmu_battery_count);
 EXPORT_SYMBOL(pmu_batteries);
 EXPORT_SYMBOL(pmu_power_flags);
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
 
diff --git a/drivers/usb/host/ohci-pci.c b/drivers/usb/host/ohci-pci.c
index 57fd07d0054..eede6be098d 100644
--- a/drivers/usb/host/ohci-pci.c
+++ b/drivers/usb/host/ohci-pci.c
@@ -14,14 +14,11 @@
  * This file is licenced under the GPL.
  */
  
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PPC_PMAC
 #include <asm/machdep.h>
 #include <asm/pmac_feature.h>
 #include <asm/pci-bridge.h>
 #include <asm/prom.h>
-#ifndef CONFIG_PM
-#	define CONFIG_PM
-#endif
 #endif
 
 #ifndef CONFIG_PCI
@@ -132,7 +129,7 @@ static int ohci_pci_suspend (struct usb_hcd *hcd, pm_message_t message)
 	/* let things settle down a bit */
 	msleep (100);
 	
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PPC_PMAC
 	if (_machine == _MACH_Pmac) {
 	   	struct device_node	*of_node;
  
@@ -141,7 +138,7 @@ static int ohci_pci_suspend (struct usb_hcd *hcd, pm_message_t message)
 		if (of_node)
 			pmac_call_feature(PMAC_FTR_USB_ENABLE, of_node, 0, 0);
 	}
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PPC_PMAC */
 	return 0;
 }
 
@@ -151,7 +148,7 @@ static int ohci_pci_resume (struct usb_hcd *hcd)
 	struct ohci_hcd		*ohci = hcd_to_ohci (hcd);
 	int			retval = 0;
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PPC_PMAC
 	if (_machine == _MACH_Pmac) {
 		struct device_node *of_node;
 
@@ -160,7 +157,7 @@ static int ohci_pci_resume (struct usb_hcd *hcd)
 		if (of_node)
 			pmac_call_feature (PMAC_FTR_USB_ENABLE, of_node, 0, 1);
 	}
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PPC_PMAC */
 
 	/* resume root hub */
 	if (time_before (jiffies, ohci->next_statechange))
diff --git a/drivers/video/aty/aty128fb.c b/drivers/video/aty/aty128fb.c
index 9789115980a..7bc1d44d881 100644
--- a/drivers/video/aty/aty128fb.c
+++ b/drivers/video/aty/aty128fb.c
@@ -350,10 +350,8 @@ static int default_vmode __initdata = VMODE_1024_768_60;
 static int default_cmode __initdata = CMODE_8;
 #endif
 
-#ifdef CONFIG_PMAC_PBOOK
 static int default_crt_on __initdata = 0;
 static int default_lcd_on __initdata = 1;
-#endif
 
 #ifdef CONFIG_MTRR
 static int mtrr = 1;
@@ -1249,7 +1247,6 @@ static int aty128_crtc_to_var(const struct aty128_crtc *crtc,
 	return 0;
 }
 
-#ifdef CONFIG_PMAC_PBOOK
 static void aty128_set_crt_enable(struct aty128fb_par *par, int on)
 {
 	if (on) {
@@ -1284,7 +1281,6 @@ static void aty128_set_lcd_enable(struct aty128fb_par *par, int on)
 		aty_st_le32(LVDS_GEN_CNTL, reg);
 	}
 }
-#endif /* CONFIG_PMAC_PBOOK */
 
 static void aty128_set_pll(struct aty128_pll *pll, const struct aty128fb_par *par)
 {
@@ -1491,12 +1487,10 @@ static int aty128fb_set_par(struct fb_info *info)
 	info->fix.visual = par->crtc.bpp == 8 ? FB_VISUAL_PSEUDOCOLOR
 		: FB_VISUAL_DIRECTCOLOR;
 
-#ifdef CONFIG_PMAC_PBOOK
 	if (par->chip_gen == rage_M3) {
 		aty128_set_crt_enable(par, par->crt_on);
 		aty128_set_lcd_enable(par, par->lcd_on);
 	}
-#endif
 	if (par->accel_flags & FB_ACCELF_TEXT)
 		aty128_init_engine(par);
 
@@ -1652,7 +1646,6 @@ static int __init aty128fb_setup(char *options)
 		return 0;
 
 	while ((this_opt = strsep(&options, ",")) != NULL) {
-#ifdef CONFIG_PMAC_PBOOK
 		if (!strncmp(this_opt, "lcd:", 4)) {
 			default_lcd_on = simple_strtoul(this_opt+4, NULL, 0);
 			continue;
@@ -1660,7 +1653,6 @@ static int __init aty128fb_setup(char *options)
 			default_crt_on = simple_strtoul(this_opt+4, NULL, 0);
 			continue;
 		}
-#endif
 #ifdef CONFIG_MTRR
 		if(!strncmp(this_opt, "nomtrr", 6)) {
 			mtrr = 0;
@@ -1752,10 +1744,8 @@ static int __init aty128_init(struct pci_dev *pdev, const struct pci_device_id *
 	info->fbops = &aty128fb_ops;
 	info->flags = FBINFO_FLAG_DEFAULT;
 
-#ifdef CONFIG_PMAC_PBOOK
 	par->lcd_on = default_lcd_on;
 	par->crt_on = default_crt_on;
-#endif
 
 	var = default_var;
 #ifdef CONFIG_PPC_PMAC
@@ -2035,12 +2025,10 @@ static int aty128fb_blank(int blank, struct fb_info *fb)
 
 	aty_st_8(CRTC_EXT_CNTL+1, state);
 
-#ifdef CONFIG_PMAC_PBOOK
 	if (par->chip_gen == rage_M3) {
 		aty128_set_crt_enable(par, par->crt_on && !blank);
 		aty128_set_lcd_enable(par, par->lcd_on && !blank);
 	}
-#endif	
 #ifdef CONFIG_PMAC_BACKLIGHT
 	if ((_machine == _MACH_Pmac) && !blank)
 		set_backlight_enable(1);
@@ -2124,7 +2112,6 @@ static int aty128fb_setcolreg(u_int regno, u_int red, u_int green, u_int blue,
 static int aty128fb_ioctl(struct inode *inode, struct file *file, u_int cmd,
 			  u_long arg, struct fb_info *info)
 {
-#ifdef CONFIG_PMAC_PBOOK
 	struct aty128fb_par *par = info->par;
 	u32 value;
 	int rc;
@@ -2149,7 +2136,6 @@ static int aty128fb_ioctl(struct inode *inode, struct file *file, u_int cmd,
 		value = (par->crt_on << 1) | par->lcd_on;
 		return put_user(value, (__u32 __user *)arg);
 	}
-#endif
 	return -EINVAL;
 }
 
diff --git a/drivers/video/chipsfb.c b/drivers/video/chipsfb.c
index 95e72550d43..e75a965ec76 100644
--- a/drivers/video/chipsfb.c
+++ b/drivers/video/chipsfb.c
@@ -28,22 +28,17 @@
 #include <linux/fb.h>
 #include <linux/init.h>
 #include <linux/pci.h>
+#include <linux/console.h>
 #include <asm/io.h>
 
 #ifdef CONFIG_PMAC_BACKLIGHT
 #include <asm/backlight.h>
 #endif
-#ifdef CONFIG_PMAC_PBOOK
-#include <linux/adb.h>
-#include <linux/pmu.h>
-#endif
 
 /*
  * Since we access the display with inb/outb to fixed port numbers,
  * we can only handle one 6555x chip.  -- paulus
  */
-static struct fb_info chipsfb_info;
-
 #define write_ind(num, val, ap, dp)	do { \
 	outb((num), (ap)); outb((val), (dp)); \
 } while (0)
@@ -74,14 +69,6 @@ static struct fb_info chipsfb_info;
 	inb(0x3da); read_ind(num, var, 0x3c0, 0x3c1); \
 } while (0)
 
-#ifdef CONFIG_PMAC_PBOOK
-static unsigned char *save_framebuffer;
-int chips_sleep_notify(struct pmu_sleep_notifier *self, int when);
-static struct pmu_sleep_notifier chips_sleep_notifier = {
-	chips_sleep_notify, SLEEP_LEVEL_VIDEO,
-};
-#endif
-
 /*
  * Exported functions
  */
@@ -356,6 +343,8 @@ static struct fb_var_screeninfo chipsfb_var __initdata = {
 
 static void __init init_chips(struct fb_info *p, unsigned long addr)
 {
+	memset(p->screen_base, 0, 0x100000);
+
 	p->fix = chipsfb_fix;
 	p->fix.smem_start = addr;
 
@@ -366,34 +355,41 @@ static void __init init_chips(struct fb_info *p, unsigned long addr)
 
 	fb_alloc_cmap(&p->cmap, 256, 0);
 
-	if (register_framebuffer(p) < 0) {
-		printk(KERN_ERR "C&T 65550 framebuffer failed to register\n");
-		return;
-	}
-
-	printk(KERN_INFO "fb%d: Chips 65550 frame buffer (%dK RAM detected)\n",
-		p->node, p->fix.smem_len / 1024);
-
 	chips_hw_init();
 }
 
 static int __devinit
 chipsfb_pci_init(struct pci_dev *dp, const struct pci_device_id *ent)
 {
-	struct fb_info *p = &chipsfb_info;
+	struct fb_info *p;
 	unsigned long addr, size;
 	unsigned short cmd;
+	int rc = -ENODEV;
+
+	if (pci_enable_device(dp) < 0) {
+		dev_err(&dp->dev, "Cannot enable PCI device\n");
+		goto err_out;
+	}
 
 	if ((dp->resource[0].flags & IORESOURCE_MEM) == 0)
-		return -ENODEV;
+		goto err_disable;
 	addr = pci_resource_start(dp, 0);
 	size = pci_resource_len(dp, 0);
 	if (addr == 0)
-		return -ENODEV;
-	if (p->screen_base != 0)
-		return -EBUSY;
-	if (!request_mem_region(addr, size, "chipsfb"))
-		return -EBUSY;
+		goto err_disable;
+
+	p = framebuffer_alloc(0, &dp->dev);
+	if (p == NULL) {
+		dev_err(&dp->dev, "Cannot allocate framebuffer structure\n");
+		rc = -ENOMEM;
+		goto err_disable;
+	}
+
+	if (pci_request_region(dp, 0, "chipsfb") != 0) {
+		dev_err(&dp->dev, "Cannot request framebuffer\n");
+		rc = -EBUSY;
+		goto err_release_fb;
+	}
 
 #ifdef __BIG_ENDIAN
 	addr += 0x800000;	// Use big-endian aperture
@@ -411,37 +407,89 @@ chipsfb_pci_init(struct pci_dev *dp, const struct pci_device_id *ent)
 	set_backlight_enable(1);
 #endif /* CONFIG_PMAC_BACKLIGHT */
 
+#ifdef CONFIG_PPC
 	p->screen_base = __ioremap(addr, 0x200000, _PAGE_NO_CACHE);
+#else
+	p->screen_base = ioremap(addr, 0x200000);
+#endif
 	if (p->screen_base == NULL) {
-		release_mem_region(addr, size);
-		return -ENOMEM;
+		dev_err(&dp->dev, "Cannot map framebuffer\n");
+		rc = -ENOMEM;
+		goto err_release_pci;
 	}
+
+	pci_set_drvdata(dp, p);
 	p->device = &dp->dev;
+
 	init_chips(p, addr);
 
-#ifdef CONFIG_PMAC_PBOOK
-	pmu_register_sleep_notifier(&chips_sleep_notifier);
-#endif /* CONFIG_PMAC_PBOOK */
+	if (register_framebuffer(p) < 0) {
+		dev_err(&dp->dev,"C&T 65550 framebuffer failed to register\n");
+		goto err_unmap;
+	}
+
+	dev_info(&dp->dev,"fb%d: Chips 65550 frame buffer"
+		 " (%dK RAM detected)\n",
+		 p->node, p->fix.smem_len / 1024);
 
-	pci_set_drvdata(dp, p);
 	return 0;
+
+ err_unmap:
+	iounmap(p->screen_base);
+ err_release_pci:
+	pci_release_region(dp, 0);
+ err_release_fb:
+	framebuffer_release(p);
+ err_disable:
+ err_out:
+	return rc;
 }
 
 static void __devexit chipsfb_remove(struct pci_dev *dp)
 {
 	struct fb_info *p = pci_get_drvdata(dp);
 
-	if (p != &chipsfb_info || p->screen_base == NULL)
+	if (p->screen_base == NULL)
 		return;
 	unregister_framebuffer(p);
 	iounmap(p->screen_base);
 	p->screen_base = NULL;
-	release_mem_region(pci_resource_start(dp, 0), pci_resource_len(dp, 0));
+	pci_release_region(dp, 0);
+}
+
+#ifdef CONFIG_PM
+static int chipsfb_pci_suspend(struct pci_dev *pdev, pm_message_t state)
+{
+        struct fb_info *p = pci_get_drvdata(pdev);
+
+	if (state == pdev->dev.power.power_state)
+		return 0;
+	if (state != PM_SUSPEND_MEM)
+		goto done;
+
+	acquire_console_sem();
+	chipsfb_blank(1, p);
+	fb_set_suspend(p, 1);
+	release_console_sem();
+ done:
+	pdev->dev.power.power_state = state;
+	return 0;
+}
+
+static int chipsfb_pci_resume(struct pci_dev *pdev)
+{
+        struct fb_info *p = pci_get_drvdata(pdev);
 
-#ifdef CONFIG_PMAC_PBOOK
-	pmu_unregister_sleep_notifier(&chips_sleep_notifier);
-#endif /* CONFIG_PMAC_PBOOK */
+	acquire_console_sem();
+	fb_set_suspend(p, 0);
+	chipsfb_blank(0, p);
+	release_console_sem();
+
+	pdev->dev.power.power_state = PMSG_ON;
+	return 0;
 }
+#endif /* CONFIG_PM */
+
 
 static struct pci_device_id chipsfb_pci_tbl[] = {
 	{ PCI_VENDOR_ID_CT, PCI_DEVICE_ID_CT_65550, PCI_ANY_ID, PCI_ANY_ID },
@@ -455,6 +503,10 @@ static struct pci_driver chipsfb_driver = {
 	.id_table =	chipsfb_pci_tbl,
 	.probe =	chipsfb_pci_init,
 	.remove =	__devexit_p(chipsfb_remove),
+#ifdef CONFIG_PM
+	.suspend =	chipsfb_pci_suspend,
+	.resume =	chipsfb_pci_resume,
+#endif
 };
 
 int __init chips_init(void)
@@ -472,48 +524,4 @@ static void __exit chipsfb_exit(void)
 	pci_unregister_driver(&chipsfb_driver);
 }
 
-#ifdef CONFIG_PMAC_PBOOK
-/*
- * Save the contents of the frame buffer when we go to sleep,
- * and restore it when we wake up again.
- */
-int
-chips_sleep_notify(struct pmu_sleep_notifier *self, int when)
-{
-	struct fb_info *p = &chipsfb_info;
-	int nb = p->var.yres * p->fix.line_length;
-
-	if (p->screen_base == NULL)
-		return PBOOK_SLEEP_OK;
-
-	switch (when) {
-	case PBOOK_SLEEP_REQUEST:
-		save_framebuffer = vmalloc(nb);
-		if (save_framebuffer == NULL)
-			return PBOOK_SLEEP_REFUSE;
-		break;
-	case PBOOK_SLEEP_REJECT:
-		if (save_framebuffer) {
-			vfree(save_framebuffer);
-			save_framebuffer = NULL;
-		}
-		break;
-	case PBOOK_SLEEP_NOW:
-		chipsfb_blank(1, p);
-		if (save_framebuffer)
-			memcpy(save_framebuffer, p->screen_base, nb);
-		break;
-	case PBOOK_WAKE:
-		if (save_framebuffer) {
-			memcpy(p->screen_base, save_framebuffer, nb);
-			vfree(save_framebuffer);
-			save_framebuffer = NULL;
-		}
-		chipsfb_blank(0, p);
-		break;
-	}
-	return PBOOK_SLEEP_OK;
-}
-#endif /* CONFIG_PMAC_PBOOK */
-
 MODULE_LICENSE("GPL");
diff --git a/include/linux/pmu.h b/include/linux/pmu.h
index 6d73eada277..373bd3b9b33 100644
--- a/include/linux/pmu.h
+++ b/include/linux/pmu.h
@@ -166,7 +166,7 @@ extern int pmu_i2c_simple_read(int bus, int addr,  u8* data, int len);
 extern int pmu_i2c_simple_write(int bus, int addr,  u8* data, int len);
 
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 /*
  * Stuff for putting the powerbook to sleep and waking it again.
  *
@@ -208,6 +208,8 @@ struct pmu_sleep_notifier
 int pmu_register_sleep_notifier(struct pmu_sleep_notifier* notifier);
 int pmu_unregister_sleep_notifier(struct pmu_sleep_notifier* notifier);
 
+#endif /* CONFIG_PM */
+
 #define PMU_MAX_BATTERIES	2
 
 /* values for pmu_power_flags */
@@ -235,6 +237,4 @@ extern int pmu_battery_count;
 extern struct pmu_battery_info pmu_batteries[PMU_MAX_BATTERIES];
 extern unsigned int pmu_power_flags;
 
-#endif /* CONFIG_PMAC_PBOOK */
-
 #endif	/* __KERNEL__ */
diff --git a/sound/oss/dmasound/dmasound_awacs.c b/sound/oss/dmasound/dmasound_awacs.c
index 33108661e67..2704e1598ad 100644
--- a/sound/oss/dmasound/dmasound_awacs.c
+++ b/sound/oss/dmasound/dmasound_awacs.c
@@ -255,7 +255,7 @@ static int awacs_burgundy_read_mvolume(unsigned address);
 
 static volatile struct dbdma_cmd *emergency_dbdma_cmd;
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 /*
  * Stuff for restoring after a sleep.
  */
@@ -263,7 +263,7 @@ static int awacs_sleep_notify(struct pmu_sleep_notifier *self, int when);
 struct pmu_sleep_notifier awacs_sleep_notifier = {
 	awacs_sleep_notify, SLEEP_LEVEL_SOUND,
 };
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
 
 /* for (soft) sample rate translations */
 int expand_bal;		/* Balance factor for expanding (not volume!) */
@@ -675,7 +675,7 @@ static void PMacIrqCleanup(void)
 	kfree(awacs_rx_cmd_space);
 	kfree(beep_dbdma_cmd_space);
 	kfree(beep_buf);
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 	pmu_unregister_sleep_notifier(&awacs_sleep_notifier);
 #endif
 }
@@ -1415,7 +1415,7 @@ load_awacs(void)
 	}
 }
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 /*
  * Save state when going to sleep, restore it afterwards.
  */
@@ -1551,7 +1551,7 @@ static int awacs_sleep_notify(struct pmu_sleep_notifier *self, int when)
 	}
 	return PBOOK_SLEEP_OK;
 }
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
 
 
 /* All the burgundy functions: */
@@ -3053,9 +3053,9 @@ printk("dmasound_pmac: Awacs/Screamer Codec Mfct: %d Rev %d\n", mfg, rev);
 	if ((res=setup_beep()))
 		return res ;
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 	pmu_register_sleep_notifier(&awacs_sleep_notifier);
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
 
 	/* Powerbooks have odd ways of enabling inputs such as
 	   an expansion-bay CD or sound from an internal modem
diff --git a/sound/ppc/awacs.c b/sound/ppc/awacs.c
index e052bd071e5..061e52d3d77 100644
--- a/sound/ppc/awacs.c
+++ b/sound/ppc/awacs.c
@@ -90,7 +90,7 @@ snd_pmac_awacs_write_noreg(pmac_t *chip, int reg, int val)
 	snd_pmac_awacs_write(chip, val | (reg << 12));
 }
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 /* Recalibrate chip */
 static void screamer_recalibrate(pmac_t *chip)
 {
@@ -642,7 +642,7 @@ static void awacs_restore_all_regs(pmac_t *chip)
 	}
 }
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 static void snd_pmac_awacs_suspend(pmac_t *chip)
 {
 	snd_pmac_awacs_write_noreg(chip, 1, (chip->awacs_reg[1]
@@ -676,7 +676,7 @@ static void snd_pmac_awacs_resume(pmac_t *chip)
 	}
 #endif
 }
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
 
 #ifdef PMAC_SUPPORT_AUTOMUTE
 /*
@@ -883,7 +883,7 @@ snd_pmac_awacs_init(pmac_t *chip)
 	 * set lowlevel callbacks
 	 */
 	chip->set_format = snd_pmac_awacs_set_format;
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 	chip->suspend = snd_pmac_awacs_suspend;
 	chip->resume = snd_pmac_awacs_resume;
 #endif
diff --git a/sound/ppc/daca.c b/sound/ppc/daca.c
index f24a9169361..a737f298e77 100644
--- a/sound/ppc/daca.c
+++ b/sound/ppc/daca.c
@@ -218,7 +218,7 @@ static snd_kcontrol_new_t daca_mixers[] = {
 };
 
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 static void daca_resume(pmac_t *chip)
 {
 	pmac_daca_t *mix = chip->mixer_data;
@@ -227,7 +227,7 @@ static void daca_resume(pmac_t *chip)
 				  mix->amp_on ? 0x05 : 0x04);
 	daca_set_volume(mix);
 }
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
 
 
 static void daca_cleanup(pmac_t *chip)
@@ -275,7 +275,7 @@ int __init snd_pmac_daca_init(pmac_t *chip)
 			return err;
 	}
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 	chip->resume = daca_resume;
 #endif
 
diff --git a/sound/ppc/pmac.c b/sound/ppc/pmac.c
index 080ef392846..75b8b742303 100644
--- a/sound/ppc/pmac.c
+++ b/sound/ppc/pmac.c
@@ -36,7 +36,7 @@
 #include <asm/pci-bridge.h>
 
 
-#if defined(CONFIG_PM) && defined(CONFIG_PMAC_PBOOK)
+#ifdef CONFIG_PM
 static int snd_pmac_register_sleep_notifier(pmac_t *chip);
 static int snd_pmac_unregister_sleep_notifier(pmac_t *chip);
 static int snd_pmac_suspend(snd_card_t *card, pm_message_t state);
@@ -782,7 +782,7 @@ static int snd_pmac_free(pmac_t *chip)
 	}
 
 	snd_pmac_sound_feature(chip, 0);
-#if defined(CONFIG_PM) && defined(CONFIG_PMAC_PBOOK)
+#ifdef CONFIG_PM
 	snd_pmac_unregister_sleep_notifier(chip);
 #endif
 
@@ -1292,7 +1292,7 @@ int __init snd_pmac_new(snd_card_t *card, pmac_t **chip_return)
 	/* Reset dbdma channels */
 	snd_pmac_dbdma_reset(chip);
 
-#if defined(CONFIG_PM) && defined(CONFIG_PMAC_PBOOK)
+#ifdef CONFIG_PM
 	/* add sleep notifier */
 	if (! snd_pmac_register_sleep_notifier(chip))
 		snd_card_set_pm_callback(chip->card, snd_pmac_suspend, snd_pmac_resume, chip);
@@ -1316,7 +1316,7 @@ int __init snd_pmac_new(snd_card_t *card, pmac_t **chip_return)
  * sleep notify for powerbook
  */
 
-#if defined(CONFIG_PM) && defined(CONFIG_PMAC_PBOOK)
+#ifdef CONFIG_PM
 
 /*
  * Save state when going to sleep, restore it afterwards.
@@ -1414,4 +1414,5 @@ static int snd_pmac_unregister_sleep_notifier(pmac_t *chip)
 	return 0;
 }
 
-#endif /* CONFIG_PM && CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
+
diff --git a/sound/ppc/pmac.h b/sound/ppc/pmac.h
index 0a84c05f714..582db522011 100644
--- a/sound/ppc/pmac.h
+++ b/sound/ppc/pmac.h
@@ -167,7 +167,7 @@ struct snd_pmac {
 	void (*set_format)(pmac_t *chip);
 	void (*update_automute)(pmac_t *chip, int do_notify);
 	int (*detect_headphone)(pmac_t *chip);
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 	void (*suspend)(pmac_t *chip);
 	void (*resume)(pmac_t *chip);
 #endif
diff --git a/sound/ppc/tumbler.c b/sound/ppc/tumbler.c
index 9332237cb6a..36c5d5d45bb 100644
--- a/sound/ppc/tumbler.c
+++ b/sound/ppc/tumbler.c
@@ -1128,7 +1128,7 @@ static void tumbler_reset_audio(pmac_t *chip)
 	}
 }
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 /* suspend mixer */
 static void tumbler_suspend(pmac_t *chip)
 {
@@ -1370,7 +1370,7 @@ int __init snd_pmac_tumbler_init(pmac_t *chip)
 	if ((err = snd_ctl_add(chip->card, chip->drc_sw_ctl)) < 0)
 		return err;
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 	chip->suspend = tumbler_suspend;
 	chip->resume = tumbler_resume;
 #endif
-- 
cgit v1.2.3-70-g09d2


From ffaa8bd6c904d1ab79b677905067349a5ff51d84 Mon Sep 17 00:00:00 2001
From: Andrea Arcangeli <andrea@suse.de>
Date: Mon, 27 Jun 2005 14:36:36 -0700
Subject: [PATCH] seccomp: tsc disable

I believe at least for seccomp it's worth to turn off the tsc, not just for
HT but for the L2 cache too.  So it's up to you, either you turn it off
completely (which isn't very nice IMHO) or I recommend to apply this below
patch.

This has been tested successfully on x86-64 against current cogito
repository (i686 compiles so I didn't bother testing ;).  People selling
the cpu through cpushare may appreciate this bit for a peace of mind.

There's no way to get any timing info anymore with this applied
(gettimeofday is forbidden of course).  The seccomp environment is
completely deterministic so it can't be allowed to get timing info, it has
to be deterministic so in the future I can enable a computing mode that
does a parallel computing for each task with server side transparent
checkpointing and verification that the output is the same from all the 2/3
seller computers for each task, without the buyer even noticing (for now
the verification is left to the buyer client side and there's no
checkpointing, since that would require more kernel changes to track the
dirty bits but it'll be easy to extend once the basic mode is finished).

Eliminating a cold-cache read of the cr4 global variable will save one
cacheline during the tlb flush while making the code per-cpu-safe at the
same time.  Thanks to Mikael Pettersson for noticing the tlb flush wasn't
per-cpu-safe.

The global tlb flush can run from irq (IPI calling do_flush_tlb_all) but
it'll be transparent to the switch_to code since the IPI won't make any
change to the cr4 contents from the point of view of the interrupted code
and since it's now all per-cpu stuff, it will not race.  So no need to
disable irqs in switch_to slow path.

Signed-off-by: Andrea Arcangeli <andrea@cpushare.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/process.c    | 29 +++++++++++++++++++++++++++++
 arch/x86_64/kernel/process.c  | 29 +++++++++++++++++++++++++++++
 include/asm-i386/tlbflush.h   | 12 +++++++-----
 include/asm-x86_64/tlbflush.h | 12 +++++++-----
 include/linux/seccomp.h       | 10 ++++++++++
 5 files changed, 82 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index 5f8cfa6b794..ba243a4cc11 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -616,6 +616,33 @@ handle_io_bitmap(struct thread_struct *next, struct tss_struct *tss)
 	tss->io_bitmap_base = INVALID_IO_BITMAP_OFFSET_LAZY;
 }
 
+/*
+ * This function selects if the context switch from prev to next
+ * has to tweak the TSC disable bit in the cr4.
+ */
+static inline void disable_tsc(struct task_struct *prev_p,
+			       struct task_struct *next_p)
+{
+	struct thread_info *prev, *next;
+
+	/*
+	 * gcc should eliminate the ->thread_info dereference if
+	 * has_secure_computing returns 0 at compile time (SECCOMP=n).
+	 */
+	prev = prev_p->thread_info;
+	next = next_p->thread_info;
+
+	if (has_secure_computing(prev) || has_secure_computing(next)) {
+		/* slow path here */
+		if (has_secure_computing(prev) &&
+		    !has_secure_computing(next)) {
+			write_cr4(read_cr4() & ~X86_CR4_TSD);
+		} else if (!has_secure_computing(prev) &&
+			   has_secure_computing(next))
+			write_cr4(read_cr4() | X86_CR4_TSD);
+	}
+}
+
 /*
  *	switch_to(x,yn) should switch tasks from x to y.
  *
@@ -695,6 +722,8 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas
 	if (unlikely(prev->io_bitmap_ptr || next->io_bitmap_ptr))
 		handle_io_bitmap(next, tss);
 
+	disable_tsc(prev_p, next_p);
+
 	return prev_p;
 }
 
diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 1d91271796e..7577f9d7a75 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -481,6 +481,33 @@ out:
 	return err;
 }
 
+/*
+ * This function selects if the context switch from prev to next
+ * has to tweak the TSC disable bit in the cr4.
+ */
+static inline void disable_tsc(struct task_struct *prev_p,
+			       struct task_struct *next_p)
+{
+	struct thread_info *prev, *next;
+
+	/*
+	 * gcc should eliminate the ->thread_info dereference if
+	 * has_secure_computing returns 0 at compile time (SECCOMP=n).
+	 */
+	prev = prev_p->thread_info;
+	next = next_p->thread_info;
+
+	if (has_secure_computing(prev) || has_secure_computing(next)) {
+		/* slow path here */
+		if (has_secure_computing(prev) &&
+		    !has_secure_computing(next)) {
+			write_cr4(read_cr4() & ~X86_CR4_TSD);
+		} else if (!has_secure_computing(prev) &&
+			   has_secure_computing(next))
+			write_cr4(read_cr4() | X86_CR4_TSD);
+	}
+}
+
 /*
  * This special macro can be used to load a debugging register
  */
@@ -599,6 +626,8 @@ struct task_struct *__switch_to(struct task_struct *prev_p, struct task_struct *
 		}
 	}
 
+	disable_tsc(prev_p, next_p);
+
 	return prev_p;
 }
 
diff --git a/include/asm-i386/tlbflush.h b/include/asm-i386/tlbflush.h
index f22fab0cea2..ab216e1370e 100644
--- a/include/asm-i386/tlbflush.h
+++ b/include/asm-i386/tlbflush.h
@@ -22,16 +22,18 @@
  */
 #define __flush_tlb_global()						\
 	do {								\
-		unsigned int tmpreg;					\
+		unsigned int tmpreg, cr4, cr4_orig;			\
 									\
 		__asm__ __volatile__(					\
-			"movl %1, %%cr4;  # turn off PGE     \n"	\
+			"movl %%cr4, %2;  # turn off PGE     \n"	\
+			"movl %2, %1;                        \n"	\
+			"andl %3, %1;                        \n"	\
+			"movl %1, %%cr4;                     \n"	\
 			"movl %%cr3, %0;                     \n"	\
 			"movl %0, %%cr3;  # flush TLB        \n"	\
 			"movl %2, %%cr4;  # turn PGE back on \n"	\
-			: "=&r" (tmpreg)				\
-			: "r" (mmu_cr4_features & ~X86_CR4_PGE),	\
-			  "r" (mmu_cr4_features)			\
+			: "=&r" (tmpreg), "=&r" (cr4), "=&r" (cr4_orig)	\
+			: "i" (~X86_CR4_PGE)				\
 			: "memory");					\
 	} while (0)
 
diff --git a/include/asm-x86_64/tlbflush.h b/include/asm-x86_64/tlbflush.h
index 2e811ac262a..06174238252 100644
--- a/include/asm-x86_64/tlbflush.h
+++ b/include/asm-x86_64/tlbflush.h
@@ -22,16 +22,18 @@
  */
 #define __flush_tlb_global()						\
 	do {								\
-		unsigned long tmpreg;					\
+		unsigned long tmpreg, cr4, cr4_orig;			\
 									\
 		__asm__ __volatile__(					\
-			"movq %1, %%cr4;  # turn off PGE     \n"	\
+			"movq %%cr4, %2;  # turn off PGE     \n"	\
+			"movq %2, %1;                        \n"	\
+			"andq %3, %1;                        \n"	\
+			"movq %1, %%cr4;                     \n"	\
 			"movq %%cr3, %0;  # flush TLB        \n"	\
 			"movq %0, %%cr3;                     \n"	\
 			"movq %2, %%cr4;  # turn PGE back on \n"	\
-			: "=&r" (tmpreg)				\
-			: "r" (mmu_cr4_features & ~X86_CR4_PGE),	\
-			  "r" (mmu_cr4_features)			\
+			: "=&r" (tmpreg), "=&r" (cr4), "=&r" (cr4_orig)	\
+			: "i" (~X86_CR4_PGE)				\
 			: "memory");					\
 	} while (0)
 
diff --git a/include/linux/seccomp.h b/include/linux/seccomp.h
index 3a2702bbb1d..dc89116bb1c 100644
--- a/include/linux/seccomp.h
+++ b/include/linux/seccomp.h
@@ -19,6 +19,11 @@ static inline void secure_computing(int this_syscall)
 		__secure_computing(this_syscall);
 }
 
+static inline int has_secure_computing(struct thread_info *ti)
+{
+	return unlikely(test_ti_thread_flag(ti, TIF_SECCOMP));
+}
+
 #else /* CONFIG_SECCOMP */
 
 #if (__GNUC__ > 2)
@@ -28,6 +33,11 @@ static inline void secure_computing(int this_syscall)
 #endif
 
 #define secure_computing(x) do { } while (0)
+/* static inline to preserve typechecking */
+static inline int has_secure_computing(struct thread_info *ti)
+{
+	return 0;
+}
 
 #endif /* CONFIG_SECCOMP */
 
-- 
cgit v1.2.3-70-g09d2


From 3de0a70bd926ff974adb27a38d4fd1049f05e54e Mon Sep 17 00:00:00 2001
From: Mike Miller <mike.miller@hp.com>
Date: Mon, 27 Jun 2005 14:36:48 -0700
Subject: [PATCH] cciss: pci id fix

This patch fixes a PCI ID I got wrong before.  It also adds support for
another new SAS controller due out this summer.  I didn't have a marketing
name prior to my last submission.  Also modifies the copyright date range.

Signed-off-by: Mike Miller <mike.miller@hp.com>
Acked-by: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/cciss.txt | 1 +
 drivers/block/cciss.c   | 9 ++++++---
 include/linux/pci_ids.h | 3 ++-
 3 files changed, 9 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/Documentation/cciss.txt b/Documentation/cciss.txt
index d599beb9df8..c8f9a73111d 100644
--- a/Documentation/cciss.txt
+++ b/Documentation/cciss.txt
@@ -17,6 +17,7 @@ This driver is known to work with the following cards:
 	* SA P600
 	* SA P800
 	* SA E400
+	* SA E300
 
 If nodes are not already created in the /dev/cciss directory, run as root:
 
diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c
index abde27027c0..0cd606ce222 100644
--- a/drivers/block/cciss.c
+++ b/drivers/block/cciss.c
@@ -1,6 +1,6 @@
 /*
  *    Disk Array driver for HP SA 5xxx and 6xxx Controllers
- *    Copyright 2000, 2002 Hewlett-Packard Development Company, L.P.
+ *    Copyright 2000, 2005 Hewlett-Packard Development Company, L.P.
  *
  *    This program is free software; you can redistribute it and/or modify
  *    it under the terms of the GNU General Public License as published by
@@ -54,7 +54,7 @@
 MODULE_AUTHOR("Hewlett-Packard Company");
 MODULE_DESCRIPTION("Driver for HP Controller SA5xxx SA6xxx version 2.6.6");
 MODULE_SUPPORTED_DEVICE("HP SA5i SA5i+ SA532 SA5300 SA5312 SA641 SA642 SA6400"
-			" SA6i P600 P800 E400");
+			" SA6i P600 P800 E400 E300");
 MODULE_LICENSE("GPL");
 
 #include "cciss_cmd.h"
@@ -85,8 +85,10 @@ static const struct pci_device_id cciss_pci_device_id[] = {
 		0x103C, 0x3225, 0, 0, 0},
 	{ PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSB,
 		0x103c, 0x3223, 0, 0, 0},
-	{ PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSB,
+	{ PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSC,
 		0x103c, 0x3231, 0, 0, 0},
+	{ PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSC,
+		0x103c, 0x3233, 0, 0, 0},
 	{0,}
 };
 MODULE_DEVICE_TABLE(pci, cciss_pci_device_id);
@@ -110,6 +112,7 @@ static struct board_type products[] = {
 	{ 0x3225103C, "Smart Array P600", &SA5_access},
 	{ 0x3223103C, "Smart Array P800", &SA5_access},
 	{ 0x3231103C, "Smart Array E400", &SA5_access},
+	{ 0x3233103C, "Smart Array E300", &SA5_access},
 };
 
 /* How long to wait (in millesconds) for board to go into simple mode */
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 3af7450278b..a3961e1d518 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -712,8 +712,9 @@
 #define PCI_DEVICE_ID_HP_DIVA_AUX	0x1290
 #define PCI_DEVICE_ID_HP_DIVA_RMP3	0x1301
 #define PCI_DEVICE_ID_HP_CISSA		0x3220
-#define PCI_DEVICE_ID_HP_CISSB		0x3230
+#define PCI_DEVICE_ID_HP_CISSB		0x3222
 #define PCI_DEVICE_ID_HP_ZX2_IOC	0x4031
+#define PCI_DEVICE_ID_HP_CISSC		0x3230
 
 #define PCI_VENDOR_ID_PCTECH		0x1042
 #define PCI_DEVICE_ID_PCTECH_RZ1000	0x1000
-- 
cgit v1.2.3-70-g09d2


From cd6fb584cf7f18ec6b221192b57d712ecc8c1859 Mon Sep 17 00:00:00 2001
From: Mike Miller <mike.miller@hp.com>
Date: Mon, 27 Jun 2005 14:36:49 -0700
Subject: [PATCH] cciss: pci domain info pass 2

This is pass 2 of my patch to add pci domain info to an existing ioctl.  This
time I insert the domain between dev_fn and board_id as Willy suggested and
change the var to unsigned short to ease Christoph's concerns.  Although I
thought unsigned int was the correct var type for this.  I also thought it
didn't matter where I inserted it in the structure.

Signed-off-by: Mike Miller <mike.miller@hp.com>
Acked-by: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/cciss.c       | 1 +
 include/linux/cciss_ioctl.h | 1 +
 2 files changed, 2 insertions(+)

(limited to 'include')

diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c
index 0cd606ce222..d5d0fa538f1 100644
--- a/drivers/block/cciss.c
+++ b/drivers/block/cciss.c
@@ -638,6 +638,7 @@ static int cciss_ioctl(struct inode *inode, struct file *filep,
 		cciss_pci_info_struct pciinfo;
 
 		if (!arg) return -EINVAL;
+		pciinfo.domain = pci_domain_nr(host->pdev->bus);
 		pciinfo.bus = host->pdev->bus->number;
 		pciinfo.dev_fn = host->pdev->devfn;
 		pciinfo.board_id = host->board_id;
diff --git a/include/linux/cciss_ioctl.h b/include/linux/cciss_ioctl.h
index ee0c6e8995d..424d5e622b4 100644
--- a/include/linux/cciss_ioctl.h
+++ b/include/linux/cciss_ioctl.h
@@ -10,6 +10,7 @@
 typedef struct _cciss_pci_info_struct
 {
 	unsigned char 	bus;
+	unsigned short	domain;
 	unsigned char 	dev_fn;
 	__u32 		board_id;
 } cciss_pci_info_struct; 
-- 
cgit v1.2.3-70-g09d2


From 9ec4b1f356b3bad928ae8e2aa9caebfa737d52df Mon Sep 17 00:00:00 2001
From: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Date: Mon, 27 Jun 2005 15:17:01 -0700
Subject: [PATCH] kprobes: fix single-step out of line - take2

Now that PPC64 has no-execute support, here is a second try to fix the
single step out of line during kprobe execution.  Kprobes on x86_64 already
solved this problem by allocating an executable page and using it as the
scratch area for stepping out of line.  Reuse that.

Signed-off-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/kprobes.c  |  26 ++++++++--
 arch/x86_64/kernel/kprobes.c | 113 +------------------------------------------
 include/asm-ia64/kprobes.h   |   1 +
 include/asm-ppc64/kprobes.h  |   2 +-
 include/linux/kprobes.h      |   2 +
 kernel/kprobes.c             | 101 ++++++++++++++++++++++++++++++++++++++
 6 files changed, 128 insertions(+), 117 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/kprobes.c b/arch/ppc64/kernel/kprobes.c
index 782ce3efa2c..86cc5496db9 100644
--- a/arch/ppc64/kernel/kprobes.c
+++ b/arch/ppc64/kernel/kprobes.c
@@ -36,6 +36,8 @@
 #include <asm/kdebug.h>
 #include <asm/sstep.h>
 
+static DECLARE_MUTEX(kprobe_mutex);
+
 static struct kprobe *current_kprobe;
 static unsigned long kprobe_status, kprobe_saved_msr;
 static struct kprobe *kprobe_prev;
@@ -54,6 +56,15 @@ int arch_prepare_kprobe(struct kprobe *p)
 		printk("Cannot register a kprobe on rfid or mtmsrd\n");
 		ret = -EINVAL;
 	}
+
+	/* insn must be on a special executable page on ppc64 */
+	if (!ret) {
+		up(&kprobe_mutex);
+		p->ainsn.insn = get_insn_slot();
+		down(&kprobe_mutex);
+		if (!p->ainsn.insn)
+			ret = -ENOMEM;
+	}
 	return ret;
 }
 
@@ -79,16 +90,22 @@ void arch_disarm_kprobe(struct kprobe *p)
 
 void arch_remove_kprobe(struct kprobe *p)
 {
+	up(&kprobe_mutex);
+	free_insn_slot(p->ainsn.insn);
+	down(&kprobe_mutex);
 }
 
 static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 {
+	kprobe_opcode_t insn = *p->ainsn.insn;
+
 	regs->msr |= MSR_SE;
-	/*single step inline if it a breakpoint instruction*/
-	if (p->opcode == BREAKPOINT_INSTRUCTION)
+
+	/* single step inline if it is a trap variant */
+	if (IS_TW(insn) || IS_TD(insn) || IS_TWI(insn) || IS_TDI(insn))
 		regs->nip = (unsigned long)p->addr;
 	else
-		regs->nip = (unsigned long)&p->ainsn.insn;
+		regs->nip = (unsigned long)p->ainsn.insn;
 }
 
 static inline void save_previous_kprobe(void)
@@ -205,9 +222,10 @@ no_kprobe:
 static void resume_execution(struct kprobe *p, struct pt_regs *regs)
 {
 	int ret;
+	unsigned int insn = *p->ainsn.insn;
 
 	regs->nip = (unsigned long)p->addr;
-	ret = emulate_step(regs, p->ainsn.insn[0]);
+	ret = emulate_step(regs, insn);
 	if (ret == 0)
 		regs->nip = (unsigned long)p->addr + 4;
 }
diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c
index 4e680f87a75..6a1c88376be 100644
--- a/arch/x86_64/kernel/kprobes.c
+++ b/arch/x86_64/kernel/kprobes.c
@@ -38,7 +38,7 @@
 #include <linux/string.h>
 #include <linux/slab.h>
 #include <linux/preempt.h>
-#include <linux/moduleloader.h>
+
 #include <asm/cacheflush.h>
 #include <asm/pgtable.h>
 #include <asm/kdebug.h>
@@ -51,8 +51,6 @@ static struct kprobe *kprobe_prev;
 static unsigned long kprobe_status_prev, kprobe_old_rflags_prev, kprobe_saved_rflags_prev;
 static struct pt_regs jprobe_saved_regs;
 static long *jprobe_saved_rsp;
-static kprobe_opcode_t *get_insn_slot(void);
-static void free_insn_slot(kprobe_opcode_t *slot);
 void jprobe_return_end(void);
 
 /* copy of the kernel stack at the probe fire time */
@@ -681,112 +679,3 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 	}
 	return 0;
 }
-
-/*
- * kprobe->ainsn.insn points to the copy of the instruction to be single-stepped.
- * By default on x86_64, pages we get from kmalloc or vmalloc are not
- * executable.  Single-stepping an instruction on such a page yields an
- * oops.  So instead of storing the instruction copies in their respective
- * kprobe objects, we allocate a page, map it executable, and store all the
- * instruction copies there.  (We can allocate additional pages if somebody
- * inserts a huge number of probes.)  Each page can hold up to INSNS_PER_PAGE
- * instruction slots, each of which is MAX_INSN_SIZE*sizeof(kprobe_opcode_t)
- * bytes.
- */
-#define INSNS_PER_PAGE (PAGE_SIZE/(MAX_INSN_SIZE*sizeof(kprobe_opcode_t)))
-struct kprobe_insn_page {
-	struct hlist_node hlist;
-	kprobe_opcode_t *insns;		/* page of instruction slots */
-	char slot_used[INSNS_PER_PAGE];
-	int nused;
-};
-
-static struct hlist_head kprobe_insn_pages;
-
-/**
- * get_insn_slot() - Find a slot on an executable page for an instruction.
- * We allocate an executable page if there's no room on existing ones.
- */
-static kprobe_opcode_t *get_insn_slot(void)
-{
-	struct kprobe_insn_page *kip;
-	struct hlist_node *pos;
-
-	hlist_for_each(pos, &kprobe_insn_pages) {
-		kip = hlist_entry(pos, struct kprobe_insn_page, hlist);
-		if (kip->nused < INSNS_PER_PAGE) {
-			int i;
-			for (i = 0; i < INSNS_PER_PAGE; i++) {
-				if (!kip->slot_used[i]) {
-					kip->slot_used[i] = 1;
-					kip->nused++;
-					return kip->insns + (i*MAX_INSN_SIZE);
-				}
-			}
-			/* Surprise!  No unused slots.  Fix kip->nused. */
-			kip->nused = INSNS_PER_PAGE;
-		}
-	}
-
-	/* All out of space.  Need to allocate a new page. Use slot 0.*/
-	kip = kmalloc(sizeof(struct kprobe_insn_page), GFP_KERNEL);
-	if (!kip) {
-		return NULL;
-	}
-
-	/*
-	 * For the %rip-relative displacement fixups to be doable, we
-	 * need our instruction copy to be within +/- 2GB of any data it
-	 * might access via %rip.  That is, within 2GB of where the
-	 * kernel image and loaded module images reside.  So we allocate
-	 * a page in the module loading area.
-	 */
-	kip->insns = module_alloc(PAGE_SIZE);
-	if (!kip->insns) {
-		kfree(kip);
-		return NULL;
-	}
-	INIT_HLIST_NODE(&kip->hlist);
-	hlist_add_head(&kip->hlist, &kprobe_insn_pages);
-	memset(kip->slot_used, 0, INSNS_PER_PAGE);
-	kip->slot_used[0] = 1;
-	kip->nused = 1;
-	return kip->insns;
-}
-
-/**
- * free_insn_slot() - Free instruction slot obtained from get_insn_slot().
- */
-static void free_insn_slot(kprobe_opcode_t *slot)
-{
-	struct kprobe_insn_page *kip;
-	struct hlist_node *pos;
-
-	hlist_for_each(pos, &kprobe_insn_pages) {
-		kip = hlist_entry(pos, struct kprobe_insn_page, hlist);
-		if (kip->insns <= slot
-		    && slot < kip->insns+(INSNS_PER_PAGE*MAX_INSN_SIZE)) {
-			int i = (slot - kip->insns) / MAX_INSN_SIZE;
-			kip->slot_used[i] = 0;
-			kip->nused--;
-			if (kip->nused == 0) {
-				/*
-				 * Page is no longer in use.  Free it unless
-				 * it's the last one.  We keep the last one
-				 * so as not to have to set it up again the
-				 * next time somebody inserts a probe.
-				 */
-				hlist_del(&kip->hlist);
-				if (hlist_empty(&kprobe_insn_pages)) {
-					INIT_HLIST_NODE(&kip->hlist);
-					hlist_add_head(&kip->hlist,
-						&kprobe_insn_pages);
-				} else {
-					module_free(NULL, kip->insns);
-					kfree(kip);
-				}
-			}
-			return;
-		}
-	}
-}
diff --git a/include/asm-ia64/kprobes.h b/include/asm-ia64/kprobes.h
index 7b700035e36..25d8b1edfcb 100644
--- a/include/asm-ia64/kprobes.h
+++ b/include/asm-ia64/kprobes.h
@@ -28,6 +28,7 @@
 #include <linux/ptrace.h>
 #include <asm/break.h>
 
+#define MAX_INSN_SIZE   16
 #define BREAK_INST	(long)(__IA64_BREAK_KPROBE << 6)
 
 typedef union cmp_inst {
diff --git a/include/asm-ppc64/kprobes.h b/include/asm-ppc64/kprobes.h
index 19b468bed05..790cf7c5277 100644
--- a/include/asm-ppc64/kprobes.h
+++ b/include/asm-ppc64/kprobes.h
@@ -45,7 +45,7 @@ typedef unsigned int kprobe_opcode_t;
 /* Architecture specific copy of original instruction */
 struct arch_specific_insn {
 	/* copy of original instruction */
-	kprobe_opcode_t insn[MAX_INSN_SIZE];
+	kprobe_opcode_t *insn;
 };
 
 #ifdef CONFIG_KPROBES
diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h
index 5e1a7b0d7b3..d304d457985 100644
--- a/include/linux/kprobes.h
+++ b/include/linux/kprobes.h
@@ -177,6 +177,8 @@ extern void arch_arm_kprobe(struct kprobe *p);
 extern void arch_disarm_kprobe(struct kprobe *p);
 extern void arch_remove_kprobe(struct kprobe *p);
 extern void show_registers(struct pt_regs *regs);
+extern kprobe_opcode_t *get_insn_slot(void);
+extern void free_insn_slot(kprobe_opcode_t *slot);
 
 /* Get the kprobe at this addr (if any).  Must have called lock_kprobes */
 struct kprobe *get_kprobe(void *addr);
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 334f37472c5..65242529a75 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -36,6 +36,7 @@
 #include <linux/hash.h>
 #include <linux/init.h>
 #include <linux/module.h>
+#include <linux/moduleloader.h>
 #include <asm/cacheflush.h>
 #include <asm/errno.h>
 #include <asm/kdebug.h>
@@ -50,6 +51,106 @@ unsigned int kprobe_cpu = NR_CPUS;
 static DEFINE_SPINLOCK(kprobe_lock);
 static struct kprobe *curr_kprobe;
 
+/*
+ * kprobe->ainsn.insn points to the copy of the instruction to be
+ * single-stepped. x86_64, POWER4 and above have no-exec support and
+ * stepping on the instruction on a vmalloced/kmalloced/data page
+ * is a recipe for disaster
+ */
+#define INSNS_PER_PAGE	(PAGE_SIZE/(MAX_INSN_SIZE * sizeof(kprobe_opcode_t)))
+
+struct kprobe_insn_page {
+	struct hlist_node hlist;
+	kprobe_opcode_t *insns;		/* Page of instruction slots */
+	char slot_used[INSNS_PER_PAGE];
+	int nused;
+};
+
+static struct hlist_head kprobe_insn_pages;
+
+/**
+ * get_insn_slot() - Find a slot on an executable page for an instruction.
+ * We allocate an executable page if there's no room on existing ones.
+ */
+kprobe_opcode_t *get_insn_slot(void)
+{
+	struct kprobe_insn_page *kip;
+	struct hlist_node *pos;
+
+	hlist_for_each(pos, &kprobe_insn_pages) {
+		kip = hlist_entry(pos, struct kprobe_insn_page, hlist);
+		if (kip->nused < INSNS_PER_PAGE) {
+			int i;
+			for (i = 0; i < INSNS_PER_PAGE; i++) {
+				if (!kip->slot_used[i]) {
+					kip->slot_used[i] = 1;
+					kip->nused++;
+					return kip->insns + (i * MAX_INSN_SIZE);
+				}
+			}
+			/* Surprise!  No unused slots.  Fix kip->nused. */
+			kip->nused = INSNS_PER_PAGE;
+		}
+	}
+
+	/* All out of space.  Need to allocate a new page. Use slot 0.*/
+	kip = kmalloc(sizeof(struct kprobe_insn_page), GFP_KERNEL);
+	if (!kip) {
+		return NULL;
+	}
+
+	/*
+	 * Use module_alloc so this page is within +/- 2GB of where the
+	 * kernel image and loaded module images reside. This is required
+	 * so x86_64 can correctly handle the %rip-relative fixups.
+	 */
+	kip->insns = module_alloc(PAGE_SIZE);
+	if (!kip->insns) {
+		kfree(kip);
+		return NULL;
+	}
+	INIT_HLIST_NODE(&kip->hlist);
+	hlist_add_head(&kip->hlist, &kprobe_insn_pages);
+	memset(kip->slot_used, 0, INSNS_PER_PAGE);
+	kip->slot_used[0] = 1;
+	kip->nused = 1;
+	return kip->insns;
+}
+
+void free_insn_slot(kprobe_opcode_t *slot)
+{
+	struct kprobe_insn_page *kip;
+	struct hlist_node *pos;
+
+	hlist_for_each(pos, &kprobe_insn_pages) {
+		kip = hlist_entry(pos, struct kprobe_insn_page, hlist);
+		if (kip->insns <= slot &&
+		    slot < kip->insns + (INSNS_PER_PAGE * MAX_INSN_SIZE)) {
+			int i = (slot - kip->insns) / MAX_INSN_SIZE;
+			kip->slot_used[i] = 0;
+			kip->nused--;
+			if (kip->nused == 0) {
+				/*
+				 * Page is no longer in use.  Free it unless
+				 * it's the last one.  We keep the last one
+				 * so as not to have to set it up again the
+				 * next time somebody inserts a probe.
+				 */
+				hlist_del(&kip->hlist);
+				if (hlist_empty(&kprobe_insn_pages)) {
+					INIT_HLIST_NODE(&kip->hlist);
+					hlist_add_head(&kip->hlist,
+						&kprobe_insn_pages);
+				} else {
+					module_free(NULL, kip->insns);
+					kfree(kip);
+				}
+			}
+			return;
+		}
+	}
+}
+
 /* Locks kprobe: irqs must be disabled */
 void lock_kprobes(void)
 {
-- 
cgit v1.2.3-70-g09d2


From 802eae7c800fb7f583e6c06afa363585af2bef00 Mon Sep 17 00:00:00 2001
From: Rusty Lynch <rusty.lynch@intel.com>
Date: Mon, 27 Jun 2005 15:17:08 -0700
Subject: [PATCH] Return probe redesign: architecture independent changes

The following is the second version of the function return probe patches
I sent out earlier this week.  Changes since my last submission include:

* Fix in ppc64 code removing an unneeded call to re-enable preemption
* Fix a build problem in ia64 when kprobes was turned off
* Added another BUG_ON check to each of the architecture trampoline
  handlers

My initial patch description ==>

 From my experiences with adding return probes to x86_64 and ia64, and the
feedback on LKML to those patches, I think we can simplify the design
for return probes.

The following patch tweaks the original design such that:

* Instead of storing the stack address in the return probe instance, the
  task pointer is stored.  This gives us all we need in order to:
    - find the correct return probe instance when we enter the trampoline
      (even if we are recursing)
    - find all left-over return probe instances when the task is going away

  This has the side effect of simplifying the implementation since more
  work can be done in kernel/kprobes.c since architecture specific knowledge
  of the stack layout is no longer required.  Specifically, we no longer have:
	- arch_get_kprobe_task()
	- arch_kprobe_flush_task()
	- get_rp_inst_tsk()
	- get_rp_inst()
	- trampoline_post_handler() <see next bullet>

* Instead of splitting the return probe handling and cleanup logic across
  the pre and post trampoline handlers, all the work is pushed into the
  pre function (trampoline_probe_handler), and then we skip single stepping
  the original function.  In this case the original instruction to be single
  stepped was just a NOP, and we can do without the extra interruption.

The new flow of events to having a return probe handler execute when a target
function exits is:

* At system initialization time, a kprobe is inserted at the beginning of
  kretprobe_trampoline.  kernel/kprobes.c use to handle this on it's own,
  but ia64 needed to do this a little differently (i.e. a function pointer
  is really a pointer to a structure containing the instruction pointer and
  a global pointer), so I added the notion of arch_init(), so that
  kernel/kprobes.c:init_kprobes() now allows architecture specific
  initialization by calling arch_init() before exiting.  Each architecture
  now registers a kprobe on it's own trampoline function.

* register_kretprobe() will insert a kprobe at the beginning of the targeted
  function with the kprobe pre_handler set to arch_prepare_kretprobe
  (still no change)

* When the target function is entered, the kprobe is fired, calling
  arch_prepare_kretprobe (still no change)

* In arch_prepare_kretprobe() we try to get a free instance and if one is
  available then we fill out the instance with a pointer to the return probe,
  the original return address, and a pointer to the task structure (instead
  of the stack address.)  Just like before we change the return address
  to the trampoline function and mark the instance as used.

  If multiple return probes are registered for a given target function,
  then arch_prepare_kretprobe() will get called multiple times for the same
  task (since our kprobe implementation is able to handle multiple kprobes
  at the same address.)  Past the first call to arch_prepare_kretprobe,
  we end up with the original address stored in the return probe instance
  pointing to our trampoline function. (This is a significant difference
  from the original arch_prepare_kretprobe design.)

* Target function executes like normal and then returns to kretprobe_trampoline.

* kprobe inserted on the first instruction of kretprobe_trampoline is fired
  and calls trampoline_probe_handler() (no change here)

* trampoline_probe_handler() consumes each of the instances associated with
  the current task by calling the registered handler function and marking
  the instance as unused until an instance is found that has a return address
  different then the trampoline function.

  (change similar to my previous ia64 RFC)

* If the task is killed with some left-over return probe instances (meaning
  that a target function was entered, but never returned), then we just
  free any instances associated with the task.  (Not much different other
  then we can handle this without calling architecture specific functions.)

  There is a known problem that this patch does not yet solve where
  registering a return probe flush_old_exec or flush_thread will put us
  in a bad state.  Most likely the best way to handle this is to not allow
  registering return probes on these two functions.

  (Significant change)

This patch series applies to the 2.6.12-rc6-mm1 kernel, and provides:
  * kernel/kprobes.c changes
  * i386 patch of existing return probes implementation
  * x86_64 patch of existing return probe implementation
  * ia64 implementation
  * ppc64 implementation (provided by Ananth)

This patch implements the architecture independant changes for a reworking
of the kprobes based function return probes design. Changes include:

  * Removing functions for querying a return probe instance off a stack address
  * Removing the stack_addr field from the kretprobe_instance definition,
    and adding a task pointer
  * Adding architecture specific initialization via arch_init()
  * Removing extern definitions for the architecture trampoline functions
    (this isn't needed anymore since the architecture handles the
     initialization of the kprobe in the return probe trampoline function.)

Signed-off-by: Rusty Lynch <rusty.lynch@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/kprobes.h | 28 +++-----------------
 kernel/kprobes.c        | 69 ++++++++++++++-----------------------------------
 2 files changed, 22 insertions(+), 75 deletions(-)

(limited to 'include')

diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h
index d304d457985..b7a194c4362 100644
--- a/include/linux/kprobes.h
+++ b/include/linux/kprobes.h
@@ -104,33 +104,12 @@ struct jprobe {
 };
 
 #ifdef ARCH_SUPPORTS_KRETPROBES
-extern int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs);
-extern void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs,
-							unsigned long flags);
-extern struct task_struct *arch_get_kprobe_task(void *ptr);
 extern void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs);
-extern void arch_kprobe_flush_task(struct task_struct *tk);
 #else /* ARCH_SUPPORTS_KRETPROBES */
-static inline void kretprobe_trampoline(void)
-{
-}
-static inline int trampoline_probe_handler(struct kprobe *p,
-						struct pt_regs *regs)
-{
-	return 0;
-}
-static inline void trampoline_post_handler(struct kprobe *p,
-				struct pt_regs *regs, unsigned long flags)
-{
-}
 static inline void arch_prepare_kretprobe(struct kretprobe *rp,
 					struct pt_regs *regs)
 {
 }
-static inline void arch_kprobe_flush_task(struct task_struct *tk)
-{
-}
-#define arch_get_kprobe_task(ptr) ((struct task_struct *)NULL)
 #endif /* ARCH_SUPPORTS_KRETPROBES */
 /*
  * Function-return probe -
@@ -155,8 +134,8 @@ struct kretprobe_instance {
 	struct hlist_node uflist; /* either on free list or used list */
 	struct hlist_node hlist;
 	struct kretprobe *rp;
-	void *ret_addr;
-	void *stack_addr;
+	kprobe_opcode_t *ret_addr;
+	struct task_struct *task;
 };
 
 #ifdef CONFIG_KPROBES
@@ -176,6 +155,7 @@ extern void arch_copy_kprobe(struct kprobe *p);
 extern void arch_arm_kprobe(struct kprobe *p);
 extern void arch_disarm_kprobe(struct kprobe *p);
 extern void arch_remove_kprobe(struct kprobe *p);
+extern int arch_init(void);
 extern void show_registers(struct pt_regs *regs);
 extern kprobe_opcode_t *get_insn_slot(void);
 extern void free_insn_slot(kprobe_opcode_t *slot);
@@ -196,8 +176,6 @@ int register_kretprobe(struct kretprobe *rp);
 void unregister_kretprobe(struct kretprobe *rp);
 
 struct kretprobe_instance *get_free_rp_inst(struct kretprobe *rp);
-struct kretprobe_instance *get_rp_inst(void *sara);
-struct kretprobe_instance *get_rp_inst_tsk(struct task_struct *tk);
 void add_rp_inst(struct kretprobe_instance *ri);
 void kprobe_flush_task(struct task_struct *tk);
 void recycle_rp_inst(struct kretprobe_instance *ri);
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 65242529a75..90c0e82b650 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -240,12 +240,6 @@ static int aggr_break_handler(struct kprobe *p, struct pt_regs *regs)
 	return 0;
 }
 
-struct kprobe trampoline_p = {
-		.addr = (kprobe_opcode_t *) &kretprobe_trampoline,
-		.pre_handler = trampoline_probe_handler,
-		.post_handler = trampoline_post_handler
-};
-
 struct kretprobe_instance *get_free_rp_inst(struct kretprobe *rp)
 {
 	struct hlist_node *node;
@@ -264,35 +258,18 @@ static struct kretprobe_instance *get_used_rp_inst(struct kretprobe *rp)
 	return NULL;
 }
 
-struct kretprobe_instance *get_rp_inst(void *sara)
-{
-	struct hlist_head *head;
-	struct hlist_node *node;
-	struct task_struct *tsk;
-	struct kretprobe_instance *ri;
-
-	tsk = arch_get_kprobe_task(sara);
-	head = &kretprobe_inst_table[hash_ptr(tsk, KPROBE_HASH_BITS)];
-	hlist_for_each_entry(ri, node, head, hlist) {
-		if (ri->stack_addr == sara)
-			return ri;
-	}
-	return NULL;
-}
-
 void add_rp_inst(struct kretprobe_instance *ri)
 {
-	struct task_struct *tsk;
 	/*
 	 * Remove rp inst off the free list -
 	 * Add it back when probed function returns
 	 */
 	hlist_del(&ri->uflist);
-	tsk = arch_get_kprobe_task(ri->stack_addr);
+
 	/* Add rp inst onto table */
 	INIT_HLIST_NODE(&ri->hlist);
 	hlist_add_head(&ri->hlist,
-			&kretprobe_inst_table[hash_ptr(tsk, KPROBE_HASH_BITS)]);
+			&kretprobe_inst_table[hash_ptr(ri->task, KPROBE_HASH_BITS)]);
 
 	/* Also add this rp inst to the used list. */
 	INIT_HLIST_NODE(&ri->uflist);
@@ -319,34 +296,25 @@ struct hlist_head * kretprobe_inst_table_head(struct task_struct *tsk)
 	return &kretprobe_inst_table[hash_ptr(tsk, KPROBE_HASH_BITS)];
 }
 
-struct kretprobe_instance *get_rp_inst_tsk(struct task_struct *tk)
-{
-	struct task_struct *tsk;
-	struct hlist_head *head;
-	struct hlist_node *node;
-	struct kretprobe_instance *ri;
-
-	head = &kretprobe_inst_table[hash_ptr(tk, KPROBE_HASH_BITS)];
-
-	hlist_for_each_entry(ri, node, head, hlist) {
-		tsk = arch_get_kprobe_task(ri->stack_addr);
-		if (tsk == tk)
-			return ri;
-	}
-	return NULL;
-}
-
 /*
- * This function is called from do_exit or do_execv when task tk's stack is
- * about to be recycled. Recycle any function-return probe instances
- * associated with this task. These represent probed functions that have
- * been called but may never return.
+ * This function is called from exit_thread or flush_thread when task tk's
+ * stack is being recycled so that we can recycle any function-return probe
+ * instances associated with this task. These left over instances represent
+ * probed functions that have been called but will never return.
  */
 void kprobe_flush_task(struct task_struct *tk)
 {
+        struct kretprobe_instance *ri;
+        struct hlist_head *head;
+	struct hlist_node *node, *tmp;
 	unsigned long flags = 0;
+
 	spin_lock_irqsave(&kprobe_lock, flags);
-	arch_kprobe_flush_task(tk);
+        head = kretprobe_inst_table_head(current);
+        hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+                if (ri->task == tk)
+                        recycle_rp_inst(ri);
+        }
 	spin_unlock_irqrestore(&kprobe_lock, flags);
 }
 
@@ -606,9 +574,10 @@ static int __init init_kprobes(void)
 		INIT_HLIST_HEAD(&kretprobe_inst_table[i]);
 	}
 
-	err = register_die_notifier(&kprobe_exceptions_nb);
-	/* Register the trampoline probe for return probe */
-	register_kprobe(&trampoline_p);
+	err = arch_init();
+	if (!err)
+		err = register_die_notifier(&kprobe_exceptions_nb);
+
 	return err;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 9508dbfe39112813612085c00d55bacd398eddc6 Mon Sep 17 00:00:00 2001
From: Rusty Lynch <rusty.lynch@intel.com>
Date: Mon, 27 Jun 2005 15:17:12 -0700
Subject: [PATCH] Return probe redesign: ia64 specific implementation

The following patch implements function return probes for ia64 using
the revised design.  With this new design we no longer need to do some
of the odd hacks previous required on the last ia64 return probe port
that I sent out for comments.

Note that this new implementation still does not resolve the problem noted
by Keith Owens where backtrace data is lost after a return probe is hit.

Changes include:
 * Addition of kretprobe_trampoline to act as a dummy function for instrumented
   functions to return to, and for the return probe infrastructure to place
   a kprobe on on, gaining control so that the return probe handler
   can be called, and so that the instruction pointer can be moved back
   to the original return address.
 * Addition of arch_init(), allowing a kprobe to be registered on
   kretprobe_trampoline
 * Addition of trampoline_probe_handler() which is used as the pre_handler
   for the kprobe inserted on kretprobe_implementation.  This is the function
   that handles the details for calling the return probe handler function
   and returning control back at the original return address
 * Addition of arch_prepare_kretprobe() which is setup as the pre_handler
   for a kprobe registered at the beginning of the target function by
   kernel/kprobes.c so that a return probe instance can be setup when
   a caller enters the target function.  (A return probe instance contains
   all the needed information for trampoline_probe_handler to do it's job.)
 * Hooks added to the exit path of a task so that we can cleanup any left-over
   return probe instances (i.e. if a task dies while inside a targeted function
   then the return probe instance was reserved at the beginning of the function
   but the function never returns so we need to mark the instance as unused.)

Signed-off-by: Rusty Lynch <rusty.lynch@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/kprobes.c | 103 ++++++++++++++++++++++++++++++++++++++++++++-
 arch/ia64/kernel/process.c |  16 +++++++
 include/asm-ia64/kprobes.h |  13 +++---
 3 files changed, 125 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index 5978823d5c6..c97e18e634c 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -290,6 +290,94 @@ static inline void set_current_kprobe(struct kprobe *p)
 	current_kprobe = p;
 }
 
+static void kretprobe_trampoline(void)
+{
+}
+
+/*
+ * At this point the target function has been tricked into
+ * returning into our trampoline.  Lookup the associated instance
+ * and then:
+ *    - call the handler function
+ *    - cleanup by marking the instance as unused
+ *    - long jump back to the original return address
+ */
+int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct kretprobe_instance *ri = NULL;
+	struct hlist_head *head;
+	struct hlist_node *node, *tmp;
+	unsigned long orig_ret_address = 0;
+	unsigned long trampoline_address =
+		((struct fnptr *)kretprobe_trampoline)->ip;
+
+        head = kretprobe_inst_table_head(current);
+
+	/*
+	 * It is possible to have multiple instances associated with a given
+	 * task either because an multiple functions in the call path
+	 * have a return probe installed on them, and/or more then one return
+	 * return probe was registered for a target function.
+	 *
+	 * We can handle this because:
+	 *     - instances are always inserted at the head of the list
+	 *     - when multiple return probes are registered for the same
+	 *       function, the first instance's ret_addr will point to the
+	 *       real return address, and all the rest will point to
+	 *       kretprobe_trampoline
+	 */
+	hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+                if (ri->task != current)
+			/* another task is sharing our hash bucket */
+                        continue;
+
+		if (ri->rp && ri->rp->handler)
+			ri->rp->handler(ri, regs);
+
+		orig_ret_address = (unsigned long)ri->ret_addr;
+		recycle_rp_inst(ri);
+
+		if (orig_ret_address != trampoline_address)
+			/*
+			 * This is the real return address. Any other
+			 * instances associated with this task are for
+			 * other calls deeper on the call stack
+			 */
+			break;
+	}
+
+	BUG_ON(!orig_ret_address || (orig_ret_address == trampoline_address));
+	regs->cr_iip = orig_ret_address;
+
+	unlock_kprobes();
+	preempt_enable_no_resched();
+
+        /*
+         * By returning a non-zero value, we are telling
+         * kprobe_handler() that we have handled unlocking
+         * and re-enabling preemption.
+         */
+        return 1;
+}
+
+void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs)
+{
+	struct kretprobe_instance *ri;
+
+	if ((ri = get_free_rp_inst(rp)) != NULL) {
+		ri->rp = rp;
+		ri->task = current;
+		ri->ret_addr = (kprobe_opcode_t *)regs->b0;
+
+		/* Replace the return addr with trampoline addr */
+		regs->b0 = ((struct fnptr *)kretprobe_trampoline)->ip;
+
+		add_rp_inst(ri);
+	} else {
+		rp->nmissed++;
+	}
+}
+
 int arch_prepare_kprobe(struct kprobe *p)
 {
 	unsigned long addr = (unsigned long) p->addr;
@@ -492,8 +580,8 @@ static int pre_kprobes_handler(struct die_args *args)
 	if (p->pre_handler && p->pre_handler(p, regs))
 		/*
 		 * Our pre-handler is specifically requesting that we just
-		 * do a return.  This is handling the case where the
-		 * pre-handler is really our special jprobe pre-handler.
+		 * do a return.  This is used for both the jprobe pre-handler
+		 * and the kretprobe trampoline
 		 */
 		return 1;
 
@@ -599,3 +687,14 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 	*regs = jprobe_saved_regs;
 	return 1;
 }
+
+static struct kprobe trampoline_p = {
+	.pre_handler = trampoline_probe_handler
+};
+
+int __init arch_init(void)
+{
+	trampoline_p.addr =
+		(kprobe_opcode_t *)((struct fnptr *)kretprobe_trampoline)->ip;
+	return register_kprobe(&trampoline_p);
+}
diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c
index ebb71f3d6d1..6e35bff05d5 100644
--- a/arch/ia64/kernel/process.c
+++ b/arch/ia64/kernel/process.c
@@ -27,6 +27,7 @@
 #include <linux/efi.h>
 #include <linux/interrupt.h>
 #include <linux/delay.h>
+#include <linux/kprobes.h>
 
 #include <asm/cpu.h>
 #include <asm/delay.h>
@@ -707,6 +708,13 @@ kernel_thread_helper (int (*fn)(void *), void *arg)
 void
 flush_thread (void)
 {
+	/*
+	 * Remove function-return probe instances associated with this task
+	 * and put them back on the free list. Do not insert an exit probe for
+	 * this function, it will be disabled by kprobe_flush_task if you do.
+	 */
+	kprobe_flush_task(current);
+
 	/* drop floating-point and debug-register state if it exists: */
 	current->thread.flags &= ~(IA64_THREAD_FPH_VALID | IA64_THREAD_DBG_VALID);
 	ia64_drop_fpu(current);
@@ -721,6 +729,14 @@ flush_thread (void)
 void
 exit_thread (void)
 {
+
+	/*
+	 * Remove function-return probe instances associated with this task
+	 * and put them back on the free list. Do not insert an exit probe for
+	 * this function, it will be disabled by kprobe_flush_task if you do.
+	 */
+	kprobe_flush_task(current);
+
 	ia64_drop_fpu(current);
 #ifdef CONFIG_PERFMON
        /* if needed, stop monitoring and flush state to perfmon context */
diff --git a/include/asm-ia64/kprobes.h b/include/asm-ia64/kprobes.h
index 25d8b1edfcb..bf36a32e37e 100644
--- a/include/asm-ia64/kprobes.h
+++ b/include/asm-ia64/kprobes.h
@@ -64,6 +64,8 @@ typedef struct _bundle {
 
 #define JPROBE_ENTRY(pentry)	(kprobe_opcode_t *)pentry
 
+#define ARCH_SUPPORTS_KRETPROBES
+
 #define SLOT0_OPCODE_SHIFT	(37)
 #define SLOT1_p1_OPCODE_SHIFT	(37 - (64-46))
 #define SLOT2_OPCODE_SHIFT 	(37)
@@ -94,11 +96,6 @@ struct arch_specific_insn {
  	unsigned short target_br_reg;
 };
 
-/* ia64 does not need this */
-static inline void jprobe_return(void)
-{
-}
-
 /* ia64 does not need this */
 static inline void arch_copy_kprobe(struct kprobe *p)
 {
@@ -107,6 +104,12 @@ static inline void arch_copy_kprobe(struct kprobe *p)
 #ifdef CONFIG_KPROBES
 extern int kprobe_exceptions_notify(struct notifier_block *self,
 				    unsigned long val, void *data);
+
+/* ia64 does not need this */
+static inline void jprobe_return(void)
+{
+}
+
 #else				/* !CONFIG_KPROBES */
 static inline int kprobe_exceptions_notify(struct notifier_block *self,
 					   unsigned long val, void *data)
-- 
cgit v1.2.3-70-g09d2


From 97f7943d70ff0e1e92ea627c44cfacfdae65dbc4 Mon Sep 17 00:00:00 2001
From: Rusty Lynch <rusty.lynch@intel.com>
Date: Mon, 27 Jun 2005 15:17:15 -0700
Subject: [PATCH] Return probe redesign: ppc64 specific implementation

The following is a patch provided by Ananth Mavinakayanahalli that implements
the new PPC64 specific parts of the new function return probe design.

NOTE: Since getting Ananth's patch, I changed trampoline_probe_handler()
      to consume each of the outstanding return probem instances (feedback
      on my original RFC after Ananth cut a patch), and also added the
      arch_init() function (adding arch specific initialization.) I have
      cross compiled but have not testing this on a PPC64 machine.

Changes include:
 * Addition of kretprobe_trampoline to act as a dummy function for instrumented
   functions to return to, and for the return probe infrastructure to place
   a kprobe on on, gaining control so that the return probe handler
   can be called, and so that the instruction pointer can be moved back
   to the original return address.
 * Addition of arch_init(), allowing a kprobe to be registered on
   kretprobe_trampoline
 * Addition of trampoline_probe_handler() which is used as the pre_handler
   for the kprobe inserted on kretprobe_implementation.  This is the function
   that handles the details for calling the return probe handler function
   and returning control back at the original return address
 * Addition of arch_prepare_kretprobe() which is setup as the pre_handler
   for a kprobe registered at the beginning of the target function by
   kernel/kprobes.c so that a return probe instance can be setup when
   a caller enters the target function.  (A return probe instance contains
   all the needed information for trampoline_probe_handler to do it's job.)
 * Hooks added to the exit path of a task so that we can cleanup any left-over
   return probe instances (i.e. if a task dies while inside a targeted function
   then the return probe instance was reserved at the beginning of the function
   but the function never returns so we need to mark the instance as unused.)

Signed-off-by: Rusty Lynch <rusty.lynch@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/kprobes.c | 99 +++++++++++++++++++++++++++++++++++++++++++++
 arch/ppc64/kernel/process.c |  4 ++
 include/asm-ppc64/kprobes.h |  3 ++
 3 files changed, 106 insertions(+)

(limited to 'include')

diff --git a/arch/ppc64/kernel/kprobes.c b/arch/ppc64/kernel/kprobes.c
index 86cc5496db9..1d2ff6d6b0b 100644
--- a/arch/ppc64/kernel/kprobes.c
+++ b/arch/ppc64/kernel/kprobes.c
@@ -122,6 +122,23 @@ static inline void restore_previous_kprobe(void)
 	kprobe_saved_msr = kprobe_saved_msr_prev;
 }
 
+void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs)
+{
+	struct kretprobe_instance *ri;
+
+	if ((ri = get_free_rp_inst(rp)) != NULL) {
+		ri->rp = rp;
+		ri->task = current;
+		ri->ret_addr = (kprobe_opcode_t *)regs->link;
+
+		/* Replace the return addr with trampoline addr */
+		regs->link = (unsigned long)kretprobe_trampoline;
+		add_rp_inst(ri);
+	} else {
+		rp->nmissed++;
+	}
+}
+
 static inline int kprobe_handler(struct pt_regs *regs)
 {
 	struct kprobe *p;
@@ -211,6 +228,78 @@ no_kprobe:
 	return ret;
 }
 
+/*
+ * Function return probe trampoline:
+ * 	- init_kprobes() establishes a probepoint here
+ * 	- When the probed function returns, this probe
+ * 		causes the handlers to fire
+ */
+void kretprobe_trampoline_holder(void)
+{
+	asm volatile(".global kretprobe_trampoline\n"
+			"kretprobe_trampoline:\n"
+			"nop\n");
+}
+
+/*
+ * Called when the probe at kretprobe trampoline is hit
+ */
+int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
+{
+        struct kretprobe_instance *ri = NULL;
+        struct hlist_head *head;
+        struct hlist_node *node, *tmp;
+	unsigned long orig_ret_address = 0;
+	unsigned long trampoline_address =(unsigned long)&kretprobe_trampoline;
+
+        head = kretprobe_inst_table_head(current);
+
+	/*
+	 * It is possible to have multiple instances associated with a given
+	 * task either because an multiple functions in the call path
+	 * have a return probe installed on them, and/or more then one return
+	 * return probe was registered for a target function.
+	 *
+	 * We can handle this because:
+	 *     - instances are always inserted at the head of the list
+	 *     - when multiple return probes are registered for the same
+         *       function, the first instance's ret_addr will point to the
+	 *       real return address, and all the rest will point to
+	 *       kretprobe_trampoline
+	 */
+	hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+                if (ri->task != current)
+			/* another task is sharing our hash bucket */
+                        continue;
+
+		if (ri->rp && ri->rp->handler)
+			ri->rp->handler(ri, regs);
+
+		orig_ret_address = (unsigned long)ri->ret_addr;
+		recycle_rp_inst(ri);
+
+		if (orig_ret_address != trampoline_address)
+			/*
+			 * This is the real return address. Any other
+			 * instances associated with this task are for
+			 * other calls deeper on the call stack
+			 */
+			break;
+	}
+
+	BUG_ON(!orig_ret_address || (orig_ret_address == trampoline_address));
+	regs->nip = orig_ret_address;
+
+	unlock_kprobes();
+
+        /*
+         * By returning a non-zero value, we are telling
+         * kprobe_handler() that we have handled unlocking
+         * and re-enabling preemption.
+         */
+        return 1;
+}
+
 /*
  * Called after single-stepping.  p->addr is the address of the
  * instruction whose first byte has been replaced by the "breakpoint"
@@ -349,3 +438,13 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 	memcpy(regs, &jprobe_saved_regs, sizeof(struct pt_regs));
 	return 1;
 }
+
+static struct kprobe trampoline_p = {
+	.addr = (kprobe_opcode_t *) &kretprobe_trampoline,
+	.pre_handler = trampoline_probe_handler
+};
+
+int __init arch_init(void)
+{
+	return register_kprobe(&trampoline_p);
+}
diff --git a/arch/ppc64/kernel/process.c b/arch/ppc64/kernel/process.c
index aba89554d89..f7cae05e40f 100644
--- a/arch/ppc64/kernel/process.c
+++ b/arch/ppc64/kernel/process.c
@@ -36,6 +36,7 @@
 #include <linux/kallsyms.h>
 #include <linux/interrupt.h>
 #include <linux/utsname.h>
+#include <linux/kprobes.h>
 
 #include <asm/pgtable.h>
 #include <asm/uaccess.h>
@@ -307,6 +308,8 @@ void show_regs(struct pt_regs * regs)
 
 void exit_thread(void)
 {
+	kprobe_flush_task(current);
+
 #ifndef CONFIG_SMP
 	if (last_task_used_math == current)
 		last_task_used_math = NULL;
@@ -321,6 +324,7 @@ void flush_thread(void)
 {
 	struct thread_info *t = current_thread_info();
 
+	kprobe_flush_task(current);
 	if (t->flags & _TIF_ABI_PENDING)
 		t->flags ^= (_TIF_ABI_PENDING | _TIF_32BIT);
 
diff --git a/include/asm-ppc64/kprobes.h b/include/asm-ppc64/kprobes.h
index 790cf7c5277..0802919c323 100644
--- a/include/asm-ppc64/kprobes.h
+++ b/include/asm-ppc64/kprobes.h
@@ -42,6 +42,9 @@ typedef unsigned int kprobe_opcode_t;
 
 #define JPROBE_ENTRY(pentry)	(kprobe_opcode_t *)((func_descr_t *)pentry)
 
+#define ARCH_SUPPORTS_KRETPROBES
+void kretprobe_trampoline(void);
+
 /* Architecture specific copy of original instruction */
 struct arch_specific_insn {
 	/* copy of original instruction */
-- 
cgit v1.2.3-70-g09d2


From c7b645f934e52a54af58142d91fb51f881f8ce26 Mon Sep 17 00:00:00 2001
From: Keshavamurthy Anil S <anil.s.keshavamurthy@intel.com>
Date: Mon, 27 Jun 2005 15:17:16 -0700
Subject: [PATCH] kprobes/ia64: refuse kprobe on ivt code

Not safe to insert kprobes on IVT code.

This patch checks to see if the address on which Kprobes is being inserted is
in ivt code and if it is in ivt code then refuse to register kprobe.

Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Acked-by: David Mosberger <davidm@napali.hpl.hp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/kprobes.c     | 18 ++++++++++++++++--
 arch/ia64/kernel/vmlinux.lds.S |  7 ++++++-
 include/asm-ia64/sections.h    |  1 +
 3 files changed, 23 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index ec2ceade12b..3aa3167edbe 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -34,6 +34,7 @@
 
 #include <asm/pgtable.h>
 #include <asm/kdebug.h>
+#include <asm/sections.h>
 
 extern void jprobe_inst_return(void);
 
@@ -263,14 +264,27 @@ static inline void get_kprobe_inst(bundle_t *bundle, uint slot,
 	}
 }
 
+/* Returns non-zero if the addr is in the Interrupt Vector Table */
+static inline int in_ivt_functions(unsigned long addr)
+{
+	return (addr >= (unsigned long)__start_ivt_text
+		&& addr < (unsigned long)__end_ivt_text);
+}
+
 static int valid_kprobe_addr(int template, int slot, unsigned long addr)
 {
 	if ((slot > 2) || ((bundle_encoding[template][1] == L) && slot > 1)) {
-		printk(KERN_WARNING "Attempting to insert unaligned kprobe at 0x%lx\n",
-				addr);
+		printk(KERN_WARNING "Attempting to insert unaligned kprobe "
+				"at 0x%lx\n", addr);
 		return -EINVAL;
 	}
 
+ 	if (in_ivt_functions(addr)) {
+ 		printk(KERN_WARNING "Kprobes can't be inserted inside "
+				"IVT functions at 0x%lx\n", addr);
+ 		return -EINVAL;
+ 	}
+
 	if (slot == 1 && bundle_encoding[template][1] != L) {
 		printk(KERN_WARNING "Inserting kprobes on slot #1 "
 		       "is not supported\n");
diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S
index b9f0db4c1b0..a676e79e068 100644
--- a/arch/ia64/kernel/vmlinux.lds.S
+++ b/arch/ia64/kernel/vmlinux.lds.S
@@ -8,6 +8,11 @@
 #define LOAD_OFFSET	(KERNEL_START - KERNEL_TR_PAGE_SIZE)
 #include <asm-generic/vmlinux.lds.h>
 
+#define IVT_TEXT							\
+		VMLINUX_SYMBOL(__start_ivt_text) = .;			\
+		*(.text.ivt)						\
+		VMLINUX_SYMBOL(__end_ivt_text) = .;
+
 OUTPUT_FORMAT("elf64-ia64-little")
 OUTPUT_ARCH(ia64)
 ENTRY(phys_start)
@@ -39,7 +44,7 @@ SECTIONS
 
   .text : AT(ADDR(.text) - LOAD_OFFSET)
     {
-	*(.text.ivt)
+	IVT_TEXT
 	*(.text)
 	SCHED_TEXT
 	LOCK_TEXT
diff --git a/include/asm-ia64/sections.h b/include/asm-ia64/sections.h
index 8e3dbde1b42..e9eb7f62d32 100644
--- a/include/asm-ia64/sections.h
+++ b/include/asm-ia64/sections.h
@@ -17,6 +17,7 @@ extern char __start_gate_vtop_patchlist[], __end_gate_vtop_patchlist[];
 extern char __start_gate_fsyscall_patchlist[], __end_gate_fsyscall_patchlist[];
 extern char __start_gate_brl_fsys_bubble_down_patchlist[], __end_gate_brl_fsys_bubble_down_patchlist[];
 extern char __start_unwind[], __end_unwind[];
+extern char __start_ivt_text[], __end_ivt_text[];
 
 #endif /* _ASM_IA64_SECTIONS_H */
 
-- 
cgit v1.2.3-70-g09d2


From b445e26cbf784cdba10f2b6c3e2cd3ee7bab360a Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Mon, 27 Jun 2005 15:42:04 -0700
Subject: [SPARC64]: Avoid membar instructions in delay slots.

In particular, avoid membar instructions in the delay
slot of a jmpl instruction.

UltraSPARC-I, II, IIi, and IIe have a bug, documented in
the UltraSPARC-IIi User's Manual, Appendix K, Erratum 51

The long and short of it is that if the IMU unit misses
on a branch or jmpl, and there is a store buffer synchronizing
membar in the delay slot, the chip can stop fetching instructions.

If interrupts are enabled or some other trap is enabled, the
chip will unwedge itself, but performance will suffer.

We already had a workaround for this bug in a few spots, but
it's better to have the entire tree sanitized for this rule.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc64/kernel/entry.S      |   6 ++-
 arch/sparc64/kernel/semaphore.c  |  12 +++--
 arch/sparc64/kernel/trampoline.S |   3 +-
 arch/sparc64/lib/U1memcpy.S      | 103 ++++++++++++++++++++-------------------
 arch/sparc64/lib/VISsave.S       |  15 +++++-
 arch/sparc64/lib/atomic.S        |  42 ++++++++++------
 arch/sparc64/lib/bitops.S        |  31 +++++++-----
 arch/sparc64/lib/debuglocks.c    |   6 ++-
 arch/sparc64/lib/dec_and_lock.S  |   6 ++-
 arch/sparc64/lib/rwsem.S         |  15 ++++--
 arch/sparc64/mm/init.c           |   6 ++-
 arch/sparc64/mm/ultra.S          |   3 +-
 include/asm-sparc64/rwsem.h      |   3 +-
 include/asm-sparc64/spinlock.h   |  29 +++++++----
 include/asm-sparc64/spitfire.h   |   1 -
 15 files changed, 171 insertions(+), 110 deletions(-)

(limited to 'include')

diff --git a/arch/sparc64/kernel/entry.S b/arch/sparc64/kernel/entry.S
index a47f2d0b1a2..ffe717ab7f8 100644
--- a/arch/sparc64/kernel/entry.S
+++ b/arch/sparc64/kernel/entry.S
@@ -271,8 +271,9 @@ cplus_fptrap_insn_1:
 	fmuld		%f0, %f2, %f26
 	faddd		%f0, %f2, %f28
 	fmuld		%f0, %f2, %f30
+	membar		#Sync
 	b,pt		%xcc, fpdis_exit
-	 membar		#Sync
+	 nop
 2:	andcc		%g5, FPRS_DU, %g0
 	bne,pt		%icc, 3f
 	 fzero		%f32
@@ -301,8 +302,9 @@ cplus_fptrap_insn_2:
 	fmuld		%f32, %f34, %f58
 	faddd		%f32, %f34, %f60
 	fmuld		%f32, %f34, %f62
+	membar		#Sync
 	ba,pt		%xcc, fpdis_exit
-	 membar		#Sync
+	 nop
 3:	mov		SECONDARY_CONTEXT, %g3
 	add		%g6, TI_FPREGS, %g1
 	ldxa		[%g3] ASI_DMMU, %g5
diff --git a/arch/sparc64/kernel/semaphore.c b/arch/sparc64/kernel/semaphore.c
index 63496c43fe1..a809e63f03e 100644
--- a/arch/sparc64/kernel/semaphore.c
+++ b/arch/sparc64/kernel/semaphore.c
@@ -32,8 +32,9 @@ static __inline__ int __sem_update_count(struct semaphore *sem, int incr)
 "	add	%1, %4, %1\n"
 "	cas	[%3], %0, %1\n"
 "	cmp	%0, %1\n"
+"	membar	#StoreLoad | #StoreStore\n"
 "	bne,pn	%%icc, 1b\n"
-"	 membar #StoreLoad | #StoreStore\n"
+"	 nop\n"
 	: "=&r" (old_count), "=&r" (tmp), "=m" (sem->count)
 	: "r" (&sem->count), "r" (incr), "m" (sem->count)
 	: "cc");
@@ -71,8 +72,9 @@ void up(struct semaphore *sem)
 "	cmp	%%g1, %%g7\n"
 "	bne,pn	%%icc, 1b\n"
 "	 addcc	%%g7, 1, %%g0\n"
+"	membar	#StoreLoad | #StoreStore\n"
 "	ble,pn	%%icc, 3f\n"
-"	 membar	#StoreLoad | #StoreStore\n"
+"	 nop\n"
 "2:\n"
 "	.subsection 2\n"
 "3:	mov	%0, %%g1\n"
@@ -128,8 +130,9 @@ void __sched down(struct semaphore *sem)
 "	cmp	%%g1, %%g7\n"
 "	bne,pn	%%icc, 1b\n"
 "	 cmp	%%g7, 1\n"
+"	membar	#StoreLoad | #StoreStore\n"
 "	bl,pn	%%icc, 3f\n"
-"	 membar	#StoreLoad | #StoreStore\n"
+"	 nop\n"
 "2:\n"
 "	.subsection 2\n"
 "3:	mov	%0, %%g1\n"
@@ -233,8 +236,9 @@ int __sched down_interruptible(struct semaphore *sem)
 "	cmp	%%g1, %%g7\n"
 "	bne,pn	%%icc, 1b\n"
 "	 cmp	%%g7, 1\n"
+"	membar	#StoreLoad | #StoreStore\n"
 "	bl,pn	%%icc, 3f\n"
-"	 membar	#StoreLoad | #StoreStore\n"
+"	 nop\n"
 "2:\n"
 "	.subsection 2\n"
 "3:	mov	%2, %%g1\n"
diff --git a/arch/sparc64/kernel/trampoline.S b/arch/sparc64/kernel/trampoline.S
index 2c8f9344b4e..3a145fc39cf 100644
--- a/arch/sparc64/kernel/trampoline.S
+++ b/arch/sparc64/kernel/trampoline.S
@@ -98,8 +98,9 @@ startup_continue:
 
 	sethi		%hi(prom_entry_lock), %g2
 1:	ldstub		[%g2 + %lo(prom_entry_lock)], %g1
+	membar		#StoreLoad | #StoreStore
 	brnz,pn		%g1, 1b
-	 membar		#StoreLoad | #StoreStore
+	 nop
 
 	sethi		%hi(p1275buf), %g2
 	or		%g2, %lo(p1275buf), %g2
diff --git a/arch/sparc64/lib/U1memcpy.S b/arch/sparc64/lib/U1memcpy.S
index da9b520c718..bafd2fc07ac 100644
--- a/arch/sparc64/lib/U1memcpy.S
+++ b/arch/sparc64/lib/U1memcpy.S
@@ -87,14 +87,17 @@
 #define LOOP_CHUNK3(src, dest, len, branch_dest)		\
 	MAIN_LOOP_CHUNK(src, dest, f32, f48, len, branch_dest)
 
+#define DO_SYNC			membar	#Sync;
 #define STORE_SYNC(dest, fsrc)				\
 	EX_ST(STORE_BLK(%fsrc, %dest));			\
-	add			%dest, 0x40, %dest;
+	add			%dest, 0x40, %dest;	\
+	DO_SYNC
 
 #define STORE_JUMP(dest, fsrc, target)			\
 	EX_ST(STORE_BLK(%fsrc, %dest));			\
 	add			%dest, 0x40, %dest;	\
-	ba,pt			%xcc, target;
+	ba,pt			%xcc, target;		\
+	 nop;
 
 #define FINISH_VISCHUNK(dest, f0, f1, left)	\
 	subcc			%left, 8, %left;\
@@ -239,17 +242,17 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f0, %f2, %f48
 1:	FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f32,f34,f36,f38,f40,f42,f44,f46,f0)
-	STORE_JUMP(o0, f48, 40f) membar #Sync
+	STORE_JUMP(o0, f48, 40f)
 2:	FREG_FROB(f32,f34,f36,f38,f40,f42,f44,f46,f0)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f0, f2, f4, f6, f8, f10,f12,f14,f16)
-	STORE_JUMP(o0, f48, 48f) membar #Sync
+	STORE_JUMP(o0, f48, 48f)
 3:	FREG_FROB(f0, f2, f4, f6, f8, f10,f12,f14,f16)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32)
-	STORE_JUMP(o0, f48, 56f) membar #Sync
+	STORE_JUMP(o0, f48, 56f)
 
 1:	FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18)
 	LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
@@ -260,17 +263,17 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f2, %f4, %f48
 1:	FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f34,f36,f38,f40,f42,f44,f46,f0, f2)
-	STORE_JUMP(o0, f48, 41f) membar #Sync
+	STORE_JUMP(o0, f48, 41f)
 2:	FREG_FROB(f34,f36,f38,f40,f42,f44,f46,f0, f2)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18)
-	STORE_JUMP(o0, f48, 49f) membar #Sync
+	STORE_JUMP(o0, f48, 49f)
 3:	FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34)
-	STORE_JUMP(o0, f48, 57f) membar #Sync
+	STORE_JUMP(o0, f48, 57f)
 
 1:	FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20)
 	LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
@@ -281,17 +284,17 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f4, %f6, %f48
 1:	FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f36,f38,f40,f42,f44,f46,f0, f2, f4)
-	STORE_JUMP(o0, f48, 42f) membar #Sync
+	STORE_JUMP(o0, f48, 42f)
 2:	FREG_FROB(f36,f38,f40,f42,f44,f46,f0, f2, f4)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20)
-	STORE_JUMP(o0, f48, 50f) membar #Sync
+	STORE_JUMP(o0, f48, 50f)
 3:	FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36)
-	STORE_JUMP(o0, f48, 58f) membar #Sync
+	STORE_JUMP(o0, f48, 58f)
 
 1:	FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22)
 	LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
@@ -302,17 +305,17 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f6, %f8, %f48
 1:	FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f38,f40,f42,f44,f46,f0, f2, f4, f6)
-	STORE_JUMP(o0, f48, 43f) membar #Sync
+	STORE_JUMP(o0, f48, 43f)
 2:	FREG_FROB(f38,f40,f42,f44,f46,f0, f2, f4, f6)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22)
-	STORE_JUMP(o0, f48, 51f) membar #Sync
+	STORE_JUMP(o0, f48, 51f)
 3:	FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38)
-	STORE_JUMP(o0, f48, 59f) membar #Sync
+	STORE_JUMP(o0, f48, 59f)
 
 1:	FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24)
 	LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
@@ -323,17 +326,17 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f8, %f10, %f48
 1:	FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f40,f42,f44,f46,f0, f2, f4, f6, f8)
-	STORE_JUMP(o0, f48, 44f) membar #Sync
+	STORE_JUMP(o0, f48, 44f)
 2:	FREG_FROB(f40,f42,f44,f46,f0, f2, f4, f6, f8)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24)
-	STORE_JUMP(o0, f48, 52f) membar #Sync
+	STORE_JUMP(o0, f48, 52f)
 3:	FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40)
-	STORE_JUMP(o0, f48, 60f) membar #Sync
+	STORE_JUMP(o0, f48, 60f)
 
 1:	FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26)
 	LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
@@ -344,17 +347,17 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f10, %f12, %f48
 1:	FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f42,f44,f46,f0, f2, f4, f6, f8, f10)
-	STORE_JUMP(o0, f48, 45f) membar #Sync
+	STORE_JUMP(o0, f48, 45f)
 2:	FREG_FROB(f42,f44,f46,f0, f2, f4, f6, f8, f10)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26)
-	STORE_JUMP(o0, f48, 53f) membar #Sync
+	STORE_JUMP(o0, f48, 53f)
 3:	FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42)
-	STORE_JUMP(o0, f48, 61f) membar #Sync
+	STORE_JUMP(o0, f48, 61f)
 
 1:	FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28)
 	LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
@@ -365,17 +368,17 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f12, %f14, %f48
 1:	FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f44,f46,f0, f2, f4, f6, f8, f10,f12)
-	STORE_JUMP(o0, f48, 46f) membar #Sync
+	STORE_JUMP(o0, f48, 46f)
 2:	FREG_FROB(f44,f46,f0, f2, f4, f6, f8, f10,f12)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28)
-	STORE_JUMP(o0, f48, 54f) membar #Sync
+	STORE_JUMP(o0, f48, 54f)
 3:	FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44)
-	STORE_JUMP(o0, f48, 62f) membar #Sync
+	STORE_JUMP(o0, f48, 62f)
 
 1:	FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30)
 	LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
@@ -386,17 +389,17 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f14, %f16, %f48
 1:	FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f46,f0, f2, f4, f6, f8, f10,f12,f14)
-	STORE_JUMP(o0, f48, 47f) membar #Sync
+	STORE_JUMP(o0, f48, 47f)
 2:	FREG_FROB(f46,f0, f2, f4, f6, f8, f10,f12,f14)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30)
-	STORE_JUMP(o0, f48, 55f) membar #Sync
+	STORE_JUMP(o0, f48, 55f)
 3:	FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30)
-	STORE_SYNC(o0, f48) membar #Sync
+	STORE_SYNC(o0, f48)
 	FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46)
-	STORE_JUMP(o0, f48, 63f) membar #Sync
+	STORE_JUMP(o0, f48, 63f)
 
 40:	FINISH_VISCHUNK(o0, f0,  f2,  g3)
 41:	FINISH_VISCHUNK(o0, f2,  f4,  g3)
diff --git a/arch/sparc64/lib/VISsave.S b/arch/sparc64/lib/VISsave.S
index 65e328d600a..4e18989bd60 100644
--- a/arch/sparc64/lib/VISsave.S
+++ b/arch/sparc64/lib/VISsave.S
@@ -72,7 +72,11 @@ vis1:	ldub		[%g6 + TI_FPSAVED], %g3
 
 	stda		%f48, [%g3 + %g1] ASI_BLK_P
 5:	membar		#Sync
-	jmpl		%g7 + %g0, %g0
+	ba,pt		%xcc, 80f
+	 nop
+
+	.align		32
+80:	jmpl		%g7 + %g0, %g0
 	 nop
 
 6:	ldub		[%g3 + TI_FPSAVED], %o5
@@ -87,8 +91,11 @@ vis1:	ldub		[%g6 + TI_FPSAVED], %g3
 	stda		%f32, [%g2 + %g1] ASI_BLK_P
 	stda		%f48, [%g3 + %g1] ASI_BLK_P
 	membar		#Sync
-	jmpl		%g7 + %g0, %g0
+	ba,pt		%xcc, 80f
+	 nop
 
+	.align		32
+80:	jmpl		%g7 + %g0, %g0
 	 nop
 
 	.align		32
@@ -126,6 +133,10 @@ VISenterhalf:
 	stda		%f0, [%g2 + %g1] ASI_BLK_P
 	stda		%f16, [%g3 + %g1] ASI_BLK_P
 	membar		#Sync
+	ba,pt		%xcc, 4f
+	 nop
+
+	.align		32
 4:	and		%o5, FPRS_DU, %o5
 	jmpl		%g7 + %g0, %g0
 	 wr		%o5, FPRS_FEF, %fprs
diff --git a/arch/sparc64/lib/atomic.S b/arch/sparc64/lib/atomic.S
index e528b8d1a3e..faf87c31598 100644
--- a/arch/sparc64/lib/atomic.S
+++ b/arch/sparc64/lib/atomic.S
@@ -7,18 +7,6 @@
 #include <linux/config.h>
 #include <asm/asi.h>
 
-	/* On SMP we need to use memory barriers to ensure
-	 * correct memory operation ordering, nop these out
-	 * for uniprocessor.
-	 */
-#ifdef CONFIG_SMP
-#define ATOMIC_PRE_BARRIER	membar #StoreLoad | #LoadLoad
-#define ATOMIC_POST_BARRIER	membar #StoreLoad | #StoreStore
-#else
-#define ATOMIC_PRE_BARRIER	nop
-#define ATOMIC_POST_BARRIER	nop
-#endif
-
 	.text
 
 	/* Two versions of the atomic routines, one that
@@ -52,6 +40,24 @@ atomic_sub: /* %o0 = decrement, %o1 = atomic_ptr */
 	 nop
 	.size	atomic_sub, .-atomic_sub
 
+	/* On SMP we need to use memory barriers to ensure
+	 * correct memory operation ordering, nop these out
+	 * for uniprocessor.
+	 */
+#ifdef CONFIG_SMP
+
+#define ATOMIC_PRE_BARRIER	membar #StoreLoad | #LoadLoad;
+#define ATOMIC_POST_BARRIER	\
+	ba,pt %xcc, 80b;	\
+	membar #StoreLoad | #StoreStore
+
+80:	retl
+	 nop
+#else
+#define ATOMIC_PRE_BARRIER
+#define ATOMIC_POST_BARRIER
+#endif
+
 	.globl	atomic_add_ret
 	.type	atomic_add_ret,#function
 atomic_add_ret: /* %o0 = increment, %o1 = atomic_ptr */
@@ -62,9 +68,10 @@ atomic_add_ret: /* %o0 = increment, %o1 = atomic_ptr */
 	cmp	%g1, %g7
 	bne,pn	%icc, 1b
 	 add	%g7, %o0, %g7
+	sra	%g7, 0, %o0
 	ATOMIC_POST_BARRIER
 	retl
-	 sra	%g7, 0, %o0
+	 nop
 	.size	atomic_add_ret, .-atomic_add_ret
 
 	.globl	atomic_sub_ret
@@ -77,9 +84,10 @@ atomic_sub_ret: /* %o0 = decrement, %o1 = atomic_ptr */
 	cmp	%g1, %g7
 	bne,pn	%icc, 1b
 	 sub	%g7, %o0, %g7
+	sra	%g7, 0, %o0
 	ATOMIC_POST_BARRIER
 	retl
-	 sra	%g7, 0, %o0
+	 nop
 	.size	atomic_sub_ret, .-atomic_sub_ret
 
 	.globl	atomic64_add
@@ -118,9 +126,10 @@ atomic64_add_ret: /* %o0 = increment, %o1 = atomic_ptr */
 	cmp	%g1, %g7
 	bne,pn	%xcc, 1b
 	 add	%g7, %o0, %g7
+	mov	%g7, %o0
 	ATOMIC_POST_BARRIER
 	retl
-	 mov	%g7, %o0
+	 nop
 	.size	atomic64_add_ret, .-atomic64_add_ret
 
 	.globl	atomic64_sub_ret
@@ -133,7 +142,8 @@ atomic64_sub_ret: /* %o0 = decrement, %o1 = atomic_ptr */
 	cmp	%g1, %g7
 	bne,pn	%xcc, 1b
 	 sub	%g7, %o0, %g7
+	mov	%g7, %o0
 	ATOMIC_POST_BARRIER
 	retl
-	 mov	%g7, %o0
+	 nop
 	.size	atomic64_sub_ret, .-atomic64_sub_ret
diff --git a/arch/sparc64/lib/bitops.S b/arch/sparc64/lib/bitops.S
index 886dcd2b376..31afbfe6c1e 100644
--- a/arch/sparc64/lib/bitops.S
+++ b/arch/sparc64/lib/bitops.S
@@ -7,20 +7,26 @@
 #include <linux/config.h>
 #include <asm/asi.h>
 
+	.text
+
 	/* On SMP we need to use memory barriers to ensure
 	 * correct memory operation ordering, nop these out
 	 * for uniprocessor.
 	 */
+
 #ifdef CONFIG_SMP
 #define BITOP_PRE_BARRIER	membar #StoreLoad | #LoadLoad
-#define BITOP_POST_BARRIER	membar #StoreLoad | #StoreStore
+#define BITOP_POST_BARRIER	\
+	ba,pt	%xcc, 80b;	\
+	membar #StoreLoad | #StoreStore
+
+80:	retl
+	 nop
 #else
-#define BITOP_PRE_BARRIER	nop
-#define BITOP_POST_BARRIER	nop
+#define BITOP_PRE_BARRIER
+#define BITOP_POST_BARRIER
 #endif
 
-	.text
-
 	.globl	test_and_set_bit
 	.type	test_and_set_bit,#function
 test_and_set_bit:	/* %o0=nr, %o1=addr */
@@ -37,10 +43,11 @@ test_and_set_bit:	/* %o0=nr, %o1=addr */
 	cmp	%g7, %g1
 	bne,pn	%xcc, 1b
 	 and	%g7, %o2, %g2
-	BITOP_POST_BARRIER
 	clr	%o0
+	movrne	%g2, 1, %o0
+	BITOP_POST_BARRIER
 	retl
-	 movrne	%g2, 1, %o0
+	 nop
 	.size	test_and_set_bit, .-test_and_set_bit
 
 	.globl	test_and_clear_bit
@@ -59,10 +66,11 @@ test_and_clear_bit:	/* %o0=nr, %o1=addr */
 	cmp	%g7, %g1
 	bne,pn	%xcc, 1b
 	 and	%g7, %o2, %g2
-	BITOP_POST_BARRIER
 	clr	%o0
+	movrne	%g2, 1, %o0
+	BITOP_POST_BARRIER
 	retl
-	 movrne	%g2, 1, %o0
+	 nop
 	.size	test_and_clear_bit, .-test_and_clear_bit
 
 	.globl	test_and_change_bit
@@ -81,10 +89,11 @@ test_and_change_bit:	/* %o0=nr, %o1=addr */
 	cmp	%g7, %g1
 	bne,pn	%xcc, 1b
 	 and	%g7, %o2, %g2
-	BITOP_POST_BARRIER
 	clr	%o0
+	movrne	%g2, 1, %o0
+	BITOP_POST_BARRIER
 	retl
-	 movrne	%g2, 1, %o0
+	 nop
 	.size	test_and_change_bit, .-test_and_change_bit
 
 	.globl	set_bit
diff --git a/arch/sparc64/lib/debuglocks.c b/arch/sparc64/lib/debuglocks.c
index c421e0c6532..f03344cf784 100644
--- a/arch/sparc64/lib/debuglocks.c
+++ b/arch/sparc64/lib/debuglocks.c
@@ -252,8 +252,9 @@ wlock_again:
 "		andn	%%g1, %%g3, %%g7\n"
 "		casx	[%0], %%g1, %%g7\n"
 "		cmp	%%g1, %%g7\n"
+"		membar	#StoreLoad | #StoreStore\n"
 "		bne,pn	%%xcc, 1b\n"
-"		 membar	#StoreLoad | #StoreStore"
+"		 nop"
 		: /* no outputs */
 		: "r" (&(rw->lock))
 		: "g3", "g1", "g7", "cc", "memory");
@@ -351,8 +352,9 @@ int _do_write_trylock (rwlock_t *rw, char *str)
 "		andn	%%g1, %%g3, %%g7\n"
 "		casx	[%0], %%g1, %%g7\n"
 "		cmp	%%g1, %%g7\n"
+"		membar	#StoreLoad | #StoreStore\n"
 "		bne,pn	%%xcc, 1b\n"
-"		 membar	#StoreLoad | #StoreStore"
+"		 nop"
 		: /* no outputs */
 		: "r" (&(rw->lock))
 		: "g3", "g1", "g7", "cc", "memory");
diff --git a/arch/sparc64/lib/dec_and_lock.S b/arch/sparc64/lib/dec_and_lock.S
index 7e6fdaebedb..8ee288dd0af 100644
--- a/arch/sparc64/lib/dec_and_lock.S
+++ b/arch/sparc64/lib/dec_and_lock.S
@@ -48,8 +48,9 @@ start_to_zero:
 #endif
 to_zero:
 	ldstub	[%o1], %g3
+	membar	#StoreLoad | #StoreStore
 	brnz,pn	%g3, spin_on_lock
-	 membar	#StoreLoad | #StoreStore
+	 nop
 loop2:	cas	[%o0], %g2, %g7		/* ASSERT(g7 == 0) */
 	cmp	%g2, %g7
 
@@ -71,8 +72,9 @@ loop2:	cas	[%o0], %g2, %g7		/* ASSERT(g7 == 0) */
 	 nop
 spin_on_lock:
 	ldub	[%o1], %g3
+	membar	#LoadLoad
 	brnz,pt	%g3, spin_on_lock
-	 membar	#LoadLoad
+	 nop
 	ba,pt	%xcc, to_zero
 	 nop
 	nop
diff --git a/arch/sparc64/lib/rwsem.S b/arch/sparc64/lib/rwsem.S
index 174ff7b9164..75f0e6b951d 100644
--- a/arch/sparc64/lib/rwsem.S
+++ b/arch/sparc64/lib/rwsem.S
@@ -17,8 +17,9 @@ __down_read:
 	bne,pn		%icc, 1b
 	 add		%g7, 1, %g7
 	cmp		%g7, 0
+	membar		#StoreLoad | #StoreStore
 	bl,pn		%icc, 3f
-	 membar		#StoreLoad | #StoreStore
+	 nop
 2:
 	retl
 	 nop
@@ -57,8 +58,9 @@ __down_write:
 	cmp		%g3, %g7
 	bne,pn		%icc, 1b
 	 cmp		%g7, 0
+	membar		#StoreLoad | #StoreStore
 	bne,pn		%icc, 3f
-	 membar		#StoreLoad | #StoreStore
+	 nop
 2:	retl
 	 nop
 3:
@@ -97,8 +99,9 @@ __up_read:
 	cmp		%g1, %g7
 	bne,pn		%icc, 1b
 	 cmp		%g7, 0
+	membar		#StoreLoad | #StoreStore
 	bl,pn		%icc, 3f
-	 membar		#StoreLoad | #StoreStore
+	 nop
 2:	retl
 	 nop
 3:	sethi		%hi(RWSEM_ACTIVE_MASK), %g1
@@ -126,8 +129,9 @@ __up_write:
 	bne,pn		%icc, 1b
 	 sub		%g7, %g1, %g7
 	cmp		%g7, 0
+	membar		#StoreLoad | #StoreStore
 	bl,pn		%icc, 3f
-	 membar		#StoreLoad | #StoreStore
+	 nop
 2:
 	retl
 	 nop
@@ -151,8 +155,9 @@ __downgrade_write:
 	bne,pn		%icc, 1b
 	 sub		%g7, %g1, %g7
 	cmp		%g7, 0
+	membar		#StoreLoad | #StoreStore
 	bl,pn		%icc, 3f
-	 membar		#StoreLoad | #StoreStore
+	 nop
 2:
 	retl
 	 nop
diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c
index 9c5222075da..8fc413cb6ac 100644
--- a/arch/sparc64/mm/init.c
+++ b/arch/sparc64/mm/init.c
@@ -136,8 +136,9 @@ static __inline__ void set_dcache_dirty(struct page *page, int this_cpu)
 			     "or	%%g1, %0, %%g1\n\t"
 			     "casx	[%2], %%g7, %%g1\n\t"
 			     "cmp	%%g7, %%g1\n\t"
+			     "membar	#StoreLoad | #StoreStore\n\t"
 			     "bne,pn	%%xcc, 1b\n\t"
-			     " membar	#StoreLoad | #StoreStore"
+			     " nop"
 			     : /* no outputs */
 			     : "r" (mask), "r" (non_cpu_bits), "r" (&page->flags)
 			     : "g1", "g7");
@@ -157,8 +158,9 @@ static __inline__ void clear_dcache_dirty_cpu(struct page *page, unsigned long c
 			     " andn	%%g7, %1, %%g1\n\t"
 			     "casx	[%2], %%g7, %%g1\n\t"
 			     "cmp	%%g7, %%g1\n\t"
+			     "membar	#StoreLoad | #StoreStore\n\t"
 			     "bne,pn	%%xcc, 1b\n\t"
-			     " membar	#StoreLoad | #StoreStore\n"
+			     " nop\n"
 			     "2:"
 			     : /* no outputs */
 			     : "r" (cpu), "r" (mask), "r" (&page->flags),
diff --git a/arch/sparc64/mm/ultra.S b/arch/sparc64/mm/ultra.S
index 7a093432101..7a2431d3abc 100644
--- a/arch/sparc64/mm/ultra.S
+++ b/arch/sparc64/mm/ultra.S
@@ -266,8 +266,9 @@ __cheetah_flush_tlb_pending:	/* 22 insns */
 	 andn		%o3, 1, %o3
 	stxa		%g0, [%o3] ASI_IMMU_DEMAP
 2:	stxa		%g0, [%o3] ASI_DMMU_DEMAP	
+	membar		#Sync
 	brnz,pt		%o1, 1b
-	 membar		#Sync
+	 nop
 	stxa		%g2, [%o4] ASI_DMMU
 	flush		%g6
 	wrpr		%g0, 0, %tl
diff --git a/include/asm-sparc64/rwsem.h b/include/asm-sparc64/rwsem.h
index bf2ae90ed3d..a1cc94f9598 100644
--- a/include/asm-sparc64/rwsem.h
+++ b/include/asm-sparc64/rwsem.h
@@ -55,8 +55,9 @@ static __inline__ int rwsem_atomic_update(int delta, struct rw_semaphore *sem)
 		"add		%%g1, %1, %%g7\n\t"
 		"cas		[%2], %%g1, %%g7\n\t"
 		"cmp		%%g1, %%g7\n\t"
+		"membar		#StoreLoad | #StoreStore\n\t"
 		"bne,pn		%%icc, 1b\n\t"
-		" membar	#StoreLoad | #StoreStore\n\t"
+		" nop\n\t"
 		"mov		%%g7, %0\n\t"
 		: "=&r" (tmp)
 		: "0" (tmp), "r" (sem)
diff --git a/include/asm-sparc64/spinlock.h b/include/asm-sparc64/spinlock.h
index db7581bdb53..9cb93a5c2b4 100644
--- a/include/asm-sparc64/spinlock.h
+++ b/include/asm-sparc64/spinlock.h
@@ -52,12 +52,14 @@ static inline void _raw_spin_lock(spinlock_t *lock)
 
 	__asm__ __volatile__(
 "1:	ldstub		[%1], %0\n"
+"	membar		#StoreLoad | #StoreStore\n"
 "	brnz,pn		%0, 2f\n"
-"	 membar		#StoreLoad | #StoreStore\n"
+"	 nop\n"
 "	.subsection	2\n"
 "2:	ldub		[%1], %0\n"
+"	membar		#LoadLoad\n"
 "	brnz,pt		%0, 2b\n"
-"	 membar		#LoadLoad\n"
+"	 nop\n"
 "	ba,a,pt		%%xcc, 1b\n"
 "	.previous"
 	: "=&r" (tmp)
@@ -95,16 +97,18 @@ static inline void _raw_spin_lock_flags(spinlock_t *lock, unsigned long flags)
 
 	__asm__ __volatile__(
 "1:	ldstub		[%2], %0\n"
-"	brnz,pn		%0, 2f\n"
 "	membar		#StoreLoad | #StoreStore\n"
+"	brnz,pn		%0, 2f\n"
+"	 nop\n"
 "	.subsection	2\n"
 "2:	rdpr		%%pil, %1\n"
 "	wrpr		%3, %%pil\n"
 "3:	ldub		[%2], %0\n"
-"	brnz,pt		%0, 3b\n"
 "	membar		#LoadLoad\n"
+"	brnz,pt		%0, 3b\n"
+"	 nop\n"
 "	ba,pt		%%xcc, 1b\n"
-"	wrpr		%1, %%pil\n"
+"	 wrpr		%1, %%pil\n"
 "	.previous"
 	: "=&r" (tmp1), "=&r" (tmp2)
 	: "r"(lock), "r"(flags)
@@ -162,12 +166,14 @@ static void inline __read_lock(rwlock_t *lock)
 "4:	 add		%0, 1, %1\n"
 "	cas		[%2], %0, %1\n"
 "	cmp		%0, %1\n"
+"	membar		#StoreLoad | #StoreStore\n"
 "	bne,pn		%%icc, 1b\n"
-"	 membar		#StoreLoad | #StoreStore\n"
+"	 nop\n"
 "	.subsection	2\n"
 "2:	ldsw		[%2], %0\n"
+"	membar		#LoadLoad\n"
 "	brlz,pt		%0, 2b\n"
-"	 membar		#LoadLoad\n"
+"	 nop\n"
 "	ba,a,pt		%%xcc, 4b\n"
 "	.previous"
 	: "=&r" (tmp1), "=&r" (tmp2)
@@ -204,12 +210,14 @@ static void inline __write_lock(rwlock_t *lock)
 "4:	 or		%0, %3, %1\n"
 "	cas		[%2], %0, %1\n"
 "	cmp		%0, %1\n"
+"	membar		#StoreLoad | #StoreStore\n"
 "	bne,pn		%%icc, 1b\n"
-"	 membar		#StoreLoad | #StoreStore\n"
+"	 nop\n"
 "	.subsection	2\n"
 "2:	lduw		[%2], %0\n"
+"	membar		#LoadLoad\n"
 "	brnz,pt		%0, 2b\n"
-"	 membar		#LoadLoad\n"
+"	 nop\n"
 "	ba,a,pt		%%xcc, 4b\n"
 "	.previous"
 	: "=&r" (tmp1), "=&r" (tmp2)
@@ -240,8 +248,9 @@ static int inline __write_trylock(rwlock_t *lock)
 "	 or		%0, %4, %1\n"
 "	cas		[%3], %0, %1\n"
 "	cmp		%0, %1\n"
+"	membar		#StoreLoad | #StoreStore\n"
 "	bne,pn		%%icc, 1b\n"
-"	 membar		#StoreLoad | #StoreStore\n"
+"	 nop\n"
 "	mov		1, %2\n"
 "2:"
 	: "=&r" (tmp1), "=&r" (tmp2), "=&r" (result)
diff --git a/include/asm-sparc64/spitfire.h b/include/asm-sparc64/spitfire.h
index 9d7613eea81..1aa932773af 100644
--- a/include/asm-sparc64/spitfire.h
+++ b/include/asm-sparc64/spitfire.h
@@ -111,7 +111,6 @@ static __inline__ void spitfire_put_dcache_tag(unsigned long addr, unsigned long
 			     "membar	#Sync"
 			     : /* No outputs */
 			     : "r" (tag), "r" (addr), "i" (ASI_DCACHE_TAG));
-	__asm__ __volatile__ ("membar #Sync" : : : "memory");
 }
 
 /* The instruction cache lines are flushed with this, but note that
-- 
cgit v1.2.3-70-g09d2


From 63b614522cba5a015923c0e8f284be6e01c13f1a Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Mon, 27 Jun 2005 17:04:45 -0700
Subject: [SPARC64]: Get rid of fast IRQ feature.

The only real user was the assembler floppy interrupt
handler, which does not need to be in assembly.

This makes it so that there are less pieces of code which
know about the internal layout of ivector_table[] and
friends.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc64/kernel/auxio.c         |   2 +-
 arch/sparc64/kernel/entry.S         | 110 -----------------------
 arch/sparc64/kernel/irq.c           | 171 +++++++++++-------------------------
 arch/sparc64/kernel/sparc64_ksyms.c |   1 -
 include/asm-sparc64/auxio.h         |   2 +
 include/asm-sparc64/floppy.h        |  16 ++--
 include/asm-sparc64/irq.h           |   7 +-
 7 files changed, 62 insertions(+), 247 deletions(-)

(limited to 'include')

diff --git a/arch/sparc64/kernel/auxio.c b/arch/sparc64/kernel/auxio.c
index a0716ccc2f4..8852c20c8d9 100644
--- a/arch/sparc64/kernel/auxio.c
+++ b/arch/sparc64/kernel/auxio.c
@@ -16,7 +16,7 @@
 #include <asm/ebus.h>
 #include <asm/auxio.h>
 
-/* This cannot be static, as it is referenced in entry.S */
+/* This cannot be static, as it is referenced in irq.c */
 void __iomem *auxio_register = NULL;
 
 enum auxio_type {
diff --git a/arch/sparc64/kernel/entry.S b/arch/sparc64/kernel/entry.S
index ffe717ab7f8..eee516a71c1 100644
--- a/arch/sparc64/kernel/entry.S
+++ b/arch/sparc64/kernel/entry.S
@@ -701,116 +701,6 @@ utrap_ill:
 	ba,pt		%xcc, rtrap
 	 clr		%l6
 
-#ifdef CONFIG_BLK_DEV_FD
-	.globl		floppy_hardint
-floppy_hardint:
-	wr		%g0, (1 << 11), %clear_softint
-	sethi		%hi(doing_pdma), %g1
-	ld		[%g1 + %lo(doing_pdma)], %g2
-	brz,pn		%g2, floppy_dosoftint
-	 sethi		%hi(fdc_status), %g3
-	ldx		[%g3 + %lo(fdc_status)], %g3
-	sethi		%hi(pdma_vaddr), %g5
-	ldx		[%g5 + %lo(pdma_vaddr)], %g4
-	sethi		%hi(pdma_size), %g5
-	ldx		[%g5 + %lo(pdma_size)], %g5
-
-next_byte:
-	lduba		[%g3] ASI_PHYS_BYPASS_EC_E, %g7
-	andcc		%g7, 0x80, %g0
-	be,pn		%icc, floppy_fifo_emptied
-	 andcc		%g7, 0x20, %g0
-	be,pn		%icc, floppy_overrun
-	 andcc		%g7, 0x40, %g0
-	be,pn		%icc, floppy_write
-	 sub		%g5, 1, %g5
-
-	inc		%g3
-	lduba		[%g3] ASI_PHYS_BYPASS_EC_E, %g7
-	dec		%g3
-	orcc		%g0, %g5, %g0
-	stb		%g7, [%g4]
-	bne,pn		%xcc, next_byte
-	 add		%g4, 1, %g4
-
-	b,pt		%xcc, floppy_tdone
-	 nop
-
-floppy_write:
-	ldub		[%g4], %g7
-	orcc		%g0, %g5, %g0
-	inc		%g3
-	stba		%g7, [%g3] ASI_PHYS_BYPASS_EC_E
-	dec		%g3
-	bne,pn		%xcc, next_byte
-	 add		%g4, 1, %g4
-
-floppy_tdone:
-	sethi		%hi(pdma_vaddr), %g1
-	stx		%g4, [%g1 + %lo(pdma_vaddr)]
-	sethi		%hi(pdma_size), %g1
-	stx		%g5, [%g1 + %lo(pdma_size)]
-	sethi		%hi(auxio_register), %g1
-	ldx		[%g1 + %lo(auxio_register)], %g7
-	lduba		[%g7] ASI_PHYS_BYPASS_EC_E, %g5
-	or		%g5, AUXIO_AUX1_FTCNT, %g5
-/*	andn		%g5, AUXIO_AUX1_MASK, %g5 */
-	stba		%g5, [%g7] ASI_PHYS_BYPASS_EC_E
-	andn		%g5, AUXIO_AUX1_FTCNT, %g5
-/*	andn		%g5, AUXIO_AUX1_MASK, %g5 */
-
-	nop; nop;  nop; nop;  nop; nop;
-	nop; nop;  nop; nop;  nop; nop;
-
-	stba		%g5, [%g7] ASI_PHYS_BYPASS_EC_E
-	sethi		%hi(doing_pdma), %g1
-	b,pt		%xcc, floppy_dosoftint
-	 st		%g0, [%g1 + %lo(doing_pdma)]
-
-floppy_fifo_emptied:
-	sethi		%hi(pdma_vaddr), %g1
-	stx		%g4, [%g1 + %lo(pdma_vaddr)]
-	sethi		%hi(pdma_size), %g1
-	stx		%g5, [%g1 + %lo(pdma_size)]
-	sethi		%hi(irq_action), %g1
-	or		%g1, %lo(irq_action), %g1
-	ldx		[%g1 + (11 << 3)], %g3		! irqaction[floppy_irq]
-	ldx		[%g3 + 0x08], %g4		! action->flags>>48==ino
-	sethi		%hi(ivector_table), %g3
-	srlx		%g4, 48, %g4
-	or		%g3, %lo(ivector_table), %g3
-	sllx		%g4, 5, %g4
-	ldx		[%g3 + %g4], %g4		! &ivector_table[ino]
-	ldx		[%g4 + 0x10], %g4		! bucket->iclr
-	stwa		%g0, [%g4] ASI_PHYS_BYPASS_EC_E	! ICLR_IDLE
-	membar		#Sync				! probably not needed...
-	retry
-
-floppy_overrun:
-	sethi		%hi(pdma_vaddr), %g1
-	stx		%g4, [%g1 + %lo(pdma_vaddr)]
-	sethi		%hi(pdma_size), %g1
-	stx		%g5, [%g1 + %lo(pdma_size)]
-	sethi		%hi(doing_pdma), %g1
-	st		%g0, [%g1 + %lo(doing_pdma)]
-
-floppy_dosoftint:
-	rdpr		%pil, %g2
-	wrpr		%g0, 15, %pil
-	sethi		%hi(109f), %g7
-	b,pt		%xcc, etrap_irq
-109:	 or		%g7, %lo(109b), %g7
-
-	mov		11, %o0
-	mov		0, %o1
-	call		sparc_floppy_irq
-	 add		%sp, PTREGS_OFF, %o2
-
-	b,pt		%xcc, rtrap_irq
-	 nop
-
-#endif /* CONFIG_BLK_DEV_FD */
-
 	/* XXX Here is stuff we still need to write... -DaveM XXX */
 	.globl		netbsd_syscall
 netbsd_syscall:
diff --git a/arch/sparc64/kernel/irq.c b/arch/sparc64/kernel/irq.c
index 4dcb8af9409..42471257730 100644
--- a/arch/sparc64/kernel/irq.c
+++ b/arch/sparc64/kernel/irq.c
@@ -37,6 +37,7 @@
 #include <asm/uaccess.h>
 #include <asm/cache.h>
 #include <asm/cpudata.h>
+#include <asm/auxio.h>
 
 #ifdef CONFIG_SMP
 static void distribute_irqs(void);
@@ -834,137 +835,65 @@ void handler_irq(int irq, struct pt_regs *regs)
 }
 
 #ifdef CONFIG_BLK_DEV_FD
-extern void floppy_interrupt(int irq, void *dev_cookie, struct pt_regs *regs);
+extern irqreturn_t floppy_interrupt(int, void *, struct pt_regs *);;
 
-void sparc_floppy_irq(int irq, void *dev_cookie, struct pt_regs *regs)
-{
-	struct irqaction *action = *(irq + irq_action);
-	struct ino_bucket *bucket;
-	int cpu = smp_processor_id();
-
-	irq_enter();
-	kstat_this_cpu.irqs[irq]++;
-
-	*(irq_work(cpu, irq)) = 0;
-	bucket = get_ino_in_irqaction(action) + ivector_table;
-
-	bucket->flags |= IBF_INPROGRESS;
-
-	floppy_interrupt(irq, dev_cookie, regs);
-	upa_writel(ICLR_IDLE, bucket->iclr);
-
-	bucket->flags &= ~IBF_INPROGRESS;
-
-	irq_exit();
-}
-#endif
-
-/* The following assumes that the branch lies before the place we
- * are branching to.  This is the case for a trap vector...
- * You have been warned.
- */
-#define SPARC_BRANCH(dest_addr, inst_addr) \
-          (0x10800000 | ((((dest_addr)-(inst_addr))>>2)&0x3fffff))
-
-#define SPARC_NOP (0x01000000)
+/* XXX No easy way to include asm/floppy.h XXX */
+extern unsigned char *pdma_vaddr;
+extern unsigned long pdma_size;
+extern volatile int doing_pdma;
+extern unsigned long fdc_status;
 
-static void install_fast_irq(unsigned int cpu_irq,
-			     irqreturn_t (*handler)(int, void *, struct pt_regs *))
+irqreturn_t sparc_floppy_irq(int irq, void *dev_cookie, struct pt_regs *regs)
 {
-	extern unsigned long sparc64_ttable_tl0;
-	unsigned long ttent = (unsigned long) &sparc64_ttable_tl0;
-	unsigned int *insns;
-
-	ttent += 0x820;
-	ttent += (cpu_irq - 1) << 5;
-	insns = (unsigned int *) ttent;
-	insns[0] = SPARC_BRANCH(((unsigned long) handler),
-				((unsigned long)&insns[0]));
-	insns[1] = SPARC_NOP;
-	__asm__ __volatile__("membar #StoreStore; flush %0" : : "r" (ttent));
-}
-
-int request_fast_irq(unsigned int irq,
-		     irqreturn_t (*handler)(int, void *, struct pt_regs *),
-		     unsigned long irqflags, const char *name, void *dev_id)
-{
-	struct irqaction *action;
-	struct ino_bucket *bucket = __bucket(irq);
-	unsigned long flags;
-
-	/* No pil0 dummy buckets allowed here. */
-	if (bucket < &ivector_table[0] ||
-	    bucket >= &ivector_table[NUM_IVECS]) {
-		unsigned int *caller;
-
-		__asm__ __volatile__("mov %%i7, %0" : "=r" (caller));
-		printk(KERN_CRIT "request_fast_irq: Old style IRQ registry attempt "
-		       "from %p, irq %08x.\n", caller, irq);
-		return -EINVAL;
-	}	
-	
-	if (!handler)
-		return -EINVAL;
+	if (likely(doing_pdma)) {
+		void __iomem *stat = (void __iomem *) fdc_status;
+		unsigned char *vaddr = pdma_vaddr;
+		unsigned long size = pdma_size;
+		u8 val;
+
+		while (size) {
+			val = readb(stat);
+			if (unlikely(!(val & 0x80))) {
+				pdma_vaddr = vaddr;
+				pdma_size = size;
+				return IRQ_HANDLED;
+			}
+			if (unlikely(!(val & 0x20))) {
+				pdma_vaddr = vaddr;
+				pdma_size = size;
+				doing_pdma = 0;
+				goto main_interrupt;
+			}
+			if (val & 0x40) {
+				/* read */
+				*vaddr++ = readb(stat + 1);
+			} else {
+				unsigned char data = *vaddr++;
 
-	if ((bucket->pil == 0) || (bucket->pil == 14)) {
-		printk("request_fast_irq: Trying to register shared IRQ 0 or 14.\n");
-		return -EBUSY;
-	}
+				/* write */
+				writeb(data, stat + 1);
+			}
+			size--;
+		}
 
-	spin_lock_irqsave(&irq_action_lock, flags);
+		pdma_vaddr = vaddr;
+		pdma_size = size;
 
-	action = *(bucket->pil + irq_action);
-	if (action) {
-		if (action->flags & SA_SHIRQ)
-			panic("Trying to register fast irq when already shared.\n");
-		if (irqflags & SA_SHIRQ)
-			panic("Trying to register fast irq as shared.\n");
-		printk("request_fast_irq: Trying to register yet already owned.\n");
-		spin_unlock_irqrestore(&irq_action_lock, flags);
-		return -EBUSY;
-	}
+		/* Send Terminal Count pulse to floppy controller. */
+		val = readb(auxio_register);
+		val |= AUXIO_AUX1_FTCNT;
+		writeb(val, auxio_register);
+		val &= AUXIO_AUX1_FTCNT;
+		writeb(val, auxio_register);
 
-	/*
-	 * We do not check for SA_SAMPLE_RANDOM in this path. Neither do we
-	 * support smp intr affinity in this path.
-	 */
-	if (irqflags & SA_STATIC_ALLOC) {
-		if (static_irq_count < MAX_STATIC_ALLOC)
-			action = &static_irqaction[static_irq_count++];
-		else
-			printk("Request for IRQ%d (%s) SA_STATIC_ALLOC failed "
-			       "using kmalloc\n", bucket->pil, name);
-	}
-	if (action == NULL)
-		action = (struct irqaction *)kmalloc(sizeof(struct irqaction),
-						     GFP_ATOMIC);
-	if (!action) {
-		spin_unlock_irqrestore(&irq_action_lock, flags);
-		return -ENOMEM;
+		doing_pdma = 0;
 	}
-	install_fast_irq(bucket->pil, handler);
 
-	bucket->irq_info = action;
-	bucket->flags |= IBF_ACTIVE;
-
-	action->handler = handler;
-	action->flags = irqflags;
-	action->dev_id = NULL;
-	action->name = name;
-	action->next = NULL;
-	put_ino_in_irqaction(action, irq);
-	put_smpaff_in_irqaction(action, CPU_MASK_NONE);
-
-	*(bucket->pil + irq_action) = action;
-	enable_irq(irq);
-
-	spin_unlock_irqrestore(&irq_action_lock, flags);
-
-#ifdef CONFIG_SMP
-	distribute_irqs();
-#endif
-	return 0;
+main_interrupt:
+	return floppy_interrupt(irq, dev_cookie, regs);
 }
+EXPORT_SYMBOL(sparc_floppy_irq);
+#endif
 
 /* We really don't need these at all on the Sparc.  We only have
  * stubs here because they are exported to modules.
diff --git a/arch/sparc64/kernel/sparc64_ksyms.c b/arch/sparc64/kernel/sparc64_ksyms.c
index e78cc53594f..56cd96f4a5c 100644
--- a/arch/sparc64/kernel/sparc64_ksyms.c
+++ b/arch/sparc64/kernel/sparc64_ksyms.c
@@ -227,7 +227,6 @@ EXPORT_SYMBOL(__flush_dcache_range);
 
 EXPORT_SYMBOL(mostek_lock);
 EXPORT_SYMBOL(mstk48t02_regs);
-EXPORT_SYMBOL(request_fast_irq);
 #ifdef CONFIG_SUN_AUXIO
 EXPORT_SYMBOL(auxio_set_led);
 EXPORT_SYMBOL(auxio_set_lte);
diff --git a/include/asm-sparc64/auxio.h b/include/asm-sparc64/auxio.h
index 5eb01dd4715..81a590a50a1 100644
--- a/include/asm-sparc64/auxio.h
+++ b/include/asm-sparc64/auxio.h
@@ -75,6 +75,8 @@
 
 #ifndef __ASSEMBLY__
 
+extern void __iomem *auxio_register;
+
 #define AUXIO_LTE_ON	1
 #define AUXIO_LTE_OFF	0
 
diff --git a/include/asm-sparc64/floppy.h b/include/asm-sparc64/floppy.h
index e071b4b4edf..49d49a28594 100644
--- a/include/asm-sparc64/floppy.h
+++ b/include/asm-sparc64/floppy.h
@@ -159,7 +159,7 @@ static void sun_82077_fd_outb(unsigned char value, unsigned long port)
  * underruns.  If non-zero, doing_pdma encodes the direction of
  * the transfer for debugging.  1=read 2=write
  */
-char *pdma_vaddr;
+unsigned char *pdma_vaddr;
 unsigned long pdma_size;
 volatile int doing_pdma = 0;
 
@@ -209,8 +209,7 @@ static void sun_fd_enable_dma(void)
 	pdma_areasize = pdma_size;
 }
 
-/* Our low-level entry point in arch/sparc/kernel/entry.S */
-extern irqreturn_t floppy_hardint(int irq, void *unused, struct pt_regs *regs);
+extern irqreturn_t sparc_floppy_irq(int, void *, struct pt_regs *);
 
 static int sun_fd_request_irq(void)
 {
@@ -220,8 +219,8 @@ static int sun_fd_request_irq(void)
 	if(!once) {
 		once = 1;
 
-		error = request_fast_irq(FLOPPY_IRQ, floppy_hardint, 
-					 SA_INTERRUPT, "floppy", NULL);
+		error = request_irq(FLOPPY_IRQ, sparc_floppy_irq, 
+				    SA_INTERRUPT, "floppy", NULL);
 
 		return ((error == 0) ? 0 : -1);
 	}
@@ -615,7 +614,7 @@ static unsigned long __init sun_floppy_init(void)
 		struct linux_ebus *ebus;
 		struct linux_ebus_device *edev = NULL;
 		unsigned long config = 0;
-		unsigned long auxio_reg;
+		void __iomem *auxio_reg;
 
 		for_each_ebus(ebus) {
 			for_each_ebusdev(edev, ebus) {
@@ -642,7 +641,7 @@ static unsigned long __init sun_floppy_init(void)
 		/* Make sure the high density bit is set, some systems
 		 * (most notably Ultra5/Ultra10) come up with it clear.
 		 */
-		auxio_reg = edev->resource[2].start;
+		auxio_reg = (void __iomem *) edev->resource[2].start;
 		writel(readl(auxio_reg)|0x2, auxio_reg);
 
 		sun_pci_ebus_dev = ebus->self;
@@ -650,7 +649,8 @@ static unsigned long __init sun_floppy_init(void)
 		spin_lock_init(&sun_pci_fd_ebus_dma.lock);
 
 		/* XXX ioremap */
-		sun_pci_fd_ebus_dma.regs = edev->resource[1].start;
+		sun_pci_fd_ebus_dma.regs = (void __iomem *)
+			edev->resource[1].start;
 		if (!sun_pci_fd_ebus_dma.regs)
 			return 0;
 
diff --git a/include/asm-sparc64/irq.h b/include/asm-sparc64/irq.h
index 3aef0ca6775..018e2e46082 100644
--- a/include/asm-sparc64/irq.h
+++ b/include/asm-sparc64/irq.h
@@ -19,7 +19,7 @@
 /* You should not mess with this directly. That's the job of irq.c.
  *
  * If you make changes here, please update hand coded assembler of
- * SBUS/floppy interrupt handler in entry.S -DaveM
+ * the vectored interrupt trap handler in entry.S -DaveM
  *
  * This is currently one DCACHE line, two buckets per L2 cache
  * line.  Keep this in mind please.
@@ -122,11 +122,6 @@ extern void enable_irq(unsigned int);
 extern unsigned int build_irq(int pil, int inofixup, unsigned long iclr, unsigned long imap);
 extern unsigned int sbus_build_irq(void *sbus, unsigned int ino);
 
-extern int request_fast_irq(unsigned int irq,
-			    irqreturn_t (*handler)(int, void *, struct pt_regs *),
-			    unsigned long flags, __const__ char *devname,
-			    void *dev_id);
-
 static __inline__ void set_softint(unsigned long bits)
 {
 	__asm__ __volatile__("wr	%0, 0x0, %%set_softint"
-- 
cgit v1.2.3-70-g09d2


From da9091ee3b5f9808c64abb925cefe7b100018614 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Mon, 27 Jun 2005 15:24:30 -0700
Subject: [PATCH] ide: it8212 backport for Bartlomiej IDE

This lets you throw out the iteraid stuff that has ended up back in due
to stupid goings on in the IDE world. Its the same heavily tested code
shipped in Fedora/Red Hat products but without the other dependancies on
the Bartlomiej IDE layer.

Pre-requisite: the ide-disk patch I sent to handle pure LBA devices.

Obviously you lose things like hot unplug with the Bartlomiej IDE layer
at the moment but that won't matter to most users.

The patch does the following
- Add IT8211/12 to pci_ids.h
- Add Makefile/Kconfig entry
- Add it8212 driver

No core IDE code is touched by this diff

Embedded system testing and the ability to force raid mode off by David
Howells

Made possible by the ite reference code, documentation and also several
clarifications and pieces of assistance provided by ITE themselves

Signed-off-by: Alan Cox <alan@redhat.com>
Acked-by: Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/ide/Kconfig      |   6 +
 drivers/ide/pci/Makefile |   1 +
 drivers/ide/pci/it821x.c | 812 +++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/pci_ids.h  |   2 +
 4 files changed, 821 insertions(+)
 create mode 100644 drivers/ide/pci/it821x.c

(limited to 'include')

diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig
index 0273f124a4f..5f33df47aa7 100644
--- a/drivers/ide/Kconfig
+++ b/drivers/ide/Kconfig
@@ -606,6 +606,12 @@ config BLK_DEV_IT8172
 	  <http://www.ite.com.tw/ia/brief_it8172bsp.htm>; picture of the
 	  board at <http://www.mvista.com/partners/semiconductor/ite.html>.
 
+config BLK_DEV_IT821X
+	tristate "IT821X IDE support"
+	help
+	  This driver adds support for the ITE 8211 IDE controller and the
+	  IT 8212 IDE RAID controller in both RAID and pass-through mode.
+
 config BLK_DEV_NS87415
 	tristate "NS87415 chipset support"
 	help
diff --git a/drivers/ide/pci/Makefile b/drivers/ide/pci/Makefile
index 55e6e553e49..af46226c179 100644
--- a/drivers/ide/pci/Makefile
+++ b/drivers/ide/pci/Makefile
@@ -12,6 +12,7 @@ obj-$(CONFIG_BLK_DEV_HPT34X)		+= hpt34x.o
 obj-$(CONFIG_BLK_DEV_HPT366)		+= hpt366.o
 #obj-$(CONFIG_BLK_DEV_HPT37X)		+= hpt37x.o
 obj-$(CONFIG_BLK_DEV_IT8172)		+= it8172.o
+obj-$(CONFIG_BLK_DEV_IT821X)		+= it821x.o
 obj-$(CONFIG_BLK_DEV_NS87415)		+= ns87415.o
 obj-$(CONFIG_BLK_DEV_OPTI621)		+= opti621.o
 obj-$(CONFIG_BLK_DEV_PDC202XX_OLD)	+= pdc202xx_old.o
diff --git a/drivers/ide/pci/it821x.c b/drivers/ide/pci/it821x.c
new file mode 100644
index 00000000000..e440036e651
--- /dev/null
+++ b/drivers/ide/pci/it821x.c
@@ -0,0 +1,812 @@
+
+/*
+ * linux/drivers/ide/pci/it821x.c		Version 0.09	December 2004
+ *
+ * Copyright (C) 2004		Red Hat <alan@redhat.com>
+ *
+ *  May be copied or modified under the terms of the GNU General Public License
+ *  Based in part on the ITE vendor provided SCSI driver.
+ *
+ *  Documentation available from
+ * 	http://www.ite.com.tw/pc/IT8212F_V04.pdf
+ *  Some other documents are NDA.
+ *
+ *  The ITE8212 isn't exactly a standard IDE controller. It has two
+ *  modes. In pass through mode then it is an IDE controller. In its smart
+ *  mode its actually quite a capable hardware raid controller disguised
+ *  as an IDE controller. Smart mode only understands DMA read/write and
+ *  identify, none of the fancier commands apply. The IT8211 is identical
+ *  in other respects but lacks the raid mode.
+ *
+ *  Errata:
+ *  o	Rev 0x10 also requires master/slave hold the same DMA timings and
+ *	cannot do ATAPI MWDMA.
+ *  o	The identify data for raid volumes lacks CHS info (technically ok)
+ *	but also fails to set the LBA28 and other bits. We fix these in
+ *	the IDE probe quirk code.
+ *  o	If you write LBA48 sized I/O's (ie > 256 sector) in smart mode
+ *	raid then the controller firmware dies
+ *  o	Smart mode without RAID doesn't clear all the necessary identify
+ *	bits to reduce the command set to the one used
+ *
+ *  This has a few impacts on the driver
+ *  - In pass through mode we do all the work you would expect
+ *  - In smart mode the clocking set up is done by the controller generally
+ *    but we must watch the other limits and filter.
+ *  - There are a few extra vendor commands that actually talk to the
+ *    controller but only work PIO with no IRQ.
+ *
+ *  Vendor areas of the identify block in smart mode are used for the
+ *  timing and policy set up. Each HDD in raid mode also has a serial
+ *  block on the disk. The hardware extra commands are get/set chip status,
+ *  rebuild, get rebuild status.
+ *
+ *  In Linux the driver supports pass through mode as if the device was
+ *  just another IDE controller. If the smart mode is running then
+ *  volumes are managed by the controller firmware and each IDE "disk"
+ *  is a raid volume. Even more cute - the controller can do automated
+ *  hotplug and rebuild.
+ *
+ *  The pass through controller itself is a little demented. It has a
+ *  flaw that it has a single set of PIO/MWDMA timings per channel so
+ *  non UDMA devices restrict each others performance. It also has a
+ *  single clock source per channel so mixed UDMA100/133 performance
+ *  isn't perfect and we have to pick a clock. Thankfully none of this
+ *  matters in smart mode. ATAPI DMA is not currently supported.
+ *
+ *  It seems the smart mode is a win for RAID1/RAID10 but otherwise not.
+ *
+ *  TODO
+ *	-	ATAPI UDMA is ok but not MWDMA it seems
+ *	-	RAID configuration ioctls
+ *	-	Move to libata once it grows up
+ */
+
+#include <linux/config.h>
+#include <linux/types.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/hdreg.h>
+#include <linux/ide.h>
+#include <linux/init.h>
+
+#include <asm/io.h>
+
+struct it821x_dev
+{
+	unsigned int smart:1,		/* Are we in smart raid mode */
+		timing10:1;		/* Rev 0x10 */
+	u8	clock_mode;		/* 0, ATA_50 or ATA_66 */
+	u8	want[2][2];		/* Mode/Pri log for master slave */
+	/* We need these for switching the clock when DMA goes on/off
+	   The high byte is the 66Mhz timing */
+	u16	pio[2];			/* Cached PIO values */
+	u16	mwdma[2];		/* Cached MWDMA values */
+	u16	udma[2];		/* Cached UDMA values (per drive) */
+};
+
+#define ATA_66		0
+#define ATA_50		1
+#define ATA_ANY		2
+
+#define UDMA_OFF	0
+#define MWDMA_OFF	0
+
+/*
+ *	We allow users to force the card into non raid mode without
+ *	flashing the alternative BIOS. This is also neccessary right now
+ *	for embedded platforms that cannot run a PC BIOS but are using this
+ *	device.
+ */
+
+static int it8212_noraid;
+
+/**
+ *	it821x_program	-	program the PIO/MWDMA registers
+ *	@drive: drive to tune
+ *
+ *	Program the PIO/MWDMA timing for this channel according to the
+ *	current clock.
+ */
+
+static void it821x_program(ide_drive_t *drive, u16 timing)
+{
+	ide_hwif_t *hwif	= drive->hwif;
+	struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+	int channel = hwif->channel;
+	u8 conf;
+
+	/* Program PIO/MWDMA timing bits */
+	if(itdev->clock_mode == ATA_66)
+		conf = timing >> 8;
+	else
+		conf = timing & 0xFF;
+	pci_write_config_byte(hwif->pci_dev, 0x54 + 4 * channel, conf);
+}
+
+/**
+ *	it821x_program_udma	-	program the UDMA registers
+ *	@drive: drive to tune
+ *
+ *	Program the UDMA timing for this drive according to the
+ *	current clock.
+ */
+
+static void it821x_program_udma(ide_drive_t *drive, u16 timing)
+{
+	ide_hwif_t *hwif	= drive->hwif;
+	struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+	int channel = hwif->channel;
+	int unit = drive->select.b.unit;
+	u8 conf;
+
+	/* Program UDMA timing bits */
+	if(itdev->clock_mode == ATA_66)
+		conf = timing >> 8;
+	else
+		conf = timing & 0xFF;
+	if(itdev->timing10 == 0)
+		pci_write_config_byte(hwif->pci_dev, 0x56 + 4 * channel + unit, conf);
+	else {
+		pci_write_config_byte(hwif->pci_dev, 0x56 + 4 * channel, conf);
+		pci_write_config_byte(hwif->pci_dev, 0x56 + 4 * channel + 1, conf);
+	}
+}
+
+
+/**
+ *	it821x_clock_strategy
+ *	@hwif: hardware interface
+ *
+ *	Select between the 50 and 66Mhz base clocks to get the best
+ *	results for this interface.
+ */
+
+static void it821x_clock_strategy(ide_drive_t *drive)
+{
+	ide_hwif_t *hwif = drive->hwif;
+	struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+
+	u8 unit = drive->select.b.unit;
+	ide_drive_t *pair = &hwif->drives[1-unit];
+
+	int clock, altclock;
+	u8 v;
+	int sel = 0;
+
+	if(itdev->want[0][0] > itdev->want[1][0]) {
+		clock = itdev->want[0][1];
+		altclock = itdev->want[1][1];
+	} else {
+		clock = itdev->want[1][1];
+		altclock = itdev->want[0][1];
+	}
+
+	/* Master doesn't care does the slave ? */
+	if(clock == ATA_ANY)
+		clock = altclock;
+
+	/* Nobody cares - keep the same clock */
+	if(clock == ATA_ANY)
+		return;
+	/* No change */
+	if(clock == itdev->clock_mode)
+		return;
+
+	/* Load this into the controller ? */
+	if(clock == ATA_66)
+		itdev->clock_mode = ATA_66;
+	else {
+		itdev->clock_mode = ATA_50;
+		sel = 1;
+	}
+	pci_read_config_byte(hwif->pci_dev, 0x50, &v);
+	v &= ~(1 << (1 + hwif->channel));
+	v |= sel << (1 + hwif->channel);
+	pci_write_config_byte(hwif->pci_dev, 0x50, v);
+
+	/*
+	 *	Reprogram the UDMA/PIO of the pair drive for the switch
+	 *	MWDMA will be dealt with by the dma switcher
+	 */
+	if(pair && itdev->udma[1-unit] != UDMA_OFF) {
+		it821x_program_udma(pair, itdev->udma[1-unit]);
+		it821x_program(pair, itdev->pio[1-unit]);
+	}
+	/*
+	 *	Reprogram the UDMA/PIO of our drive for the switch.
+	 *	MWDMA will be dealt with by the dma switcher
+	 */
+	if(itdev->udma[unit] != UDMA_OFF) {
+		it821x_program_udma(drive, itdev->udma[unit]);
+		it821x_program(drive, itdev->pio[unit]);
+	}
+}
+
+/**
+ *	it821x_ratemask	-	Compute available modes
+ *	@drive: IDE drive
+ *
+ *	Compute the available speeds for the devices on the interface. This
+ *	is all modes to ATA133 clipped by drive cable setup.
+ */
+
+static u8 it821x_ratemask (ide_drive_t *drive)
+{
+	u8 mode	= 4;
+	if (!eighty_ninty_three(drive))
+		mode = min(mode, (u8)1);
+	return mode;
+}
+
+/**
+ *	it821x_tuneproc	-	tune a drive
+ *	@drive: drive to tune
+ *	@mode_wanted: the target operating mode
+ *
+ *	Load the timing settings for this device mode into the
+ *	controller. By the time we are called the mode has been
+ *	modified as neccessary to handle the absence of seperate
+ *	master/slave timers for MWDMA/PIO.
+ *
+ *	This code is only used in pass through mode.
+ */
+
+static void it821x_tuneproc (ide_drive_t *drive, byte mode_wanted)
+{
+	ide_hwif_t *hwif	= drive->hwif;
+	struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+	int unit = drive->select.b.unit;
+
+	/* Spec says 89 ref driver uses 88 */
+	static u16 pio[]	= { 0xAA88, 0xA382, 0xA181, 0x3332, 0x3121 };
+	static u8 pio_want[]    = { ATA_66, ATA_66, ATA_66, ATA_66, ATA_ANY };
+
+	if(itdev->smart)
+		return;
+
+	/* We prefer 66Mhz clock for PIO 0-3, don't care for PIO4 */
+	itdev->want[unit][1] = pio_want[mode_wanted];
+	itdev->want[unit][0] = 1;	/* PIO is lowest priority */
+	itdev->pio[unit] = pio[mode_wanted];
+	it821x_clock_strategy(drive);
+	it821x_program(drive, itdev->pio[unit]);
+}
+
+/**
+ *	it821x_tune_mwdma	-	tune a channel for MWDMA
+ *	@drive: drive to set up
+ *	@mode_wanted: the target operating mode
+ *
+ *	Load the timing settings for this device mode into the
+ *	controller when doing MWDMA in pass through mode. The caller
+ *	must manage the whole lack of per device MWDMA/PIO timings and
+ *	the shared MWDMA/PIO timing register.
+ */
+
+static void it821x_tune_mwdma (ide_drive_t *drive, byte mode_wanted)
+{
+	ide_hwif_t *hwif	= drive->hwif;
+	struct it821x_dev *itdev = (void *)ide_get_hwifdata(hwif);
+	int unit = drive->select.b.unit;
+	int channel = hwif->channel;
+	u8 conf;
+
+	static u16 dma[]	= { 0x8866, 0x3222, 0x3121 };
+	static u8 mwdma_want[]	= { ATA_ANY, ATA_66, ATA_ANY };
+
+	itdev->want[unit][1] = mwdma_want[mode_wanted];
+	itdev->want[unit][0] = 2;	/* MWDMA is low priority */
+	itdev->mwdma[unit] = dma[mode_wanted];
+	itdev->udma[unit] = UDMA_OFF;
+
+	/* UDMA bits off - Revision 0x10 do them in pairs */
+	pci_read_config_byte(hwif->pci_dev, 0x50, &conf);
+	if(itdev->timing10)
+		conf |= channel ? 0x60: 0x18;
+	else
+		conf |= 1 << (3 + 2 * channel + unit);
+	pci_write_config_byte(hwif->pci_dev, 0x50, conf);
+
+	it821x_clock_strategy(drive);
+	/* FIXME: do we need to program this ? */
+	/* it821x_program(drive, itdev->mwdma[unit]); */
+}
+
+/**
+ *	it821x_tune_udma	-	tune a channel for UDMA
+ *	@drive: drive to set up
+ *	@mode_wanted: the target operating mode
+ *
+ *	Load the timing settings for this device mode into the
+ *	controller when doing UDMA modes in pass through.
+ */
+
+static void it821x_tune_udma (ide_drive_t *drive, byte mode_wanted)
+{
+	ide_hwif_t *hwif	= drive->hwif;
+	struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+	int unit = drive->select.b.unit;
+	int channel = hwif->channel;
+	u8 conf;
+
+	static u16 udma[]	= { 0x4433, 0x4231, 0x3121, 0x2121, 0x1111, 0x2211, 0x1111 };
+	static u8 udma_want[]	= { ATA_ANY, ATA_50, ATA_ANY, ATA_66, ATA_66, ATA_50, ATA_66 };
+
+	itdev->want[unit][1] = udma_want[mode_wanted];
+	itdev->want[unit][0] = 3;	/* UDMA is high priority */
+	itdev->mwdma[unit] = MWDMA_OFF;
+	itdev->udma[unit] = udma[mode_wanted];
+	if(mode_wanted >= 5)
+		itdev->udma[unit] |= 0x8080;	/* UDMA 5/6 select on */
+
+	/* UDMA on. Again revision 0x10 must do the pair */
+	pci_read_config_byte(hwif->pci_dev, 0x50, &conf);
+	if(itdev->timing10)
+		conf &= channel ? 0x9F: 0xE7;
+	else
+		conf &= ~ (1 << (3 + 2 * channel + unit));
+	pci_write_config_byte(hwif->pci_dev, 0x50, conf);
+
+	it821x_clock_strategy(drive);
+	it821x_program_udma(drive, itdev->udma[unit]);
+
+}
+
+/**
+ *	config_it821x_chipset_for_pio	-	set drive timings
+ *	@drive: drive to tune
+ *	@speed we want
+ *
+ *	Compute the best pio mode we can for a given device. We must
+ *	pick a speed that does not cause problems with the other device
+ *	on the cable.
+ */
+
+static void config_it821x_chipset_for_pio (ide_drive_t *drive, byte set_speed)
+{
+	u8 unit = drive->select.b.unit;
+	ide_hwif_t *hwif = drive->hwif;
+	ide_drive_t *pair = &hwif->drives[1-unit];
+	u8 speed = 0, set_pio	= ide_get_best_pio_mode(drive, 255, 5, NULL);
+	u8 pair_pio;
+
+	/* We have to deal with this mess in pairs */
+	if(pair != NULL) {
+		pair_pio = ide_get_best_pio_mode(pair, 255, 5, NULL);
+		/* Trim PIO to the slowest of the master/slave */
+		if(pair_pio < set_pio)
+			set_pio = pair_pio;
+	}
+	it821x_tuneproc(drive, set_pio);
+	speed = XFER_PIO_0 + set_pio;
+	/* XXX - We trim to the lowest of the pair so the other drive
+	   will always be fine at this point until we do hotplug passthru */
+
+	if (set_speed)
+		(void) ide_config_drive_speed(drive, speed);
+}
+
+/**
+ *	it821x_dma_read	-	DMA hook
+ *	@drive: drive for DMA
+ *
+ *	The IT821x has a single timing register for MWDMA and for PIO
+ *	operations. As we flip back and forth we have to reload the
+ *	clock. In addition the rev 0x10 device only works if the same
+ *	timing value is loaded into the master and slave UDMA clock
+ * 	so we must also reload that.
+ *
+ *	FIXME: we could figure out in advance if we need to do reloads
+ */
+
+static void it821x_dma_start(ide_drive_t *drive)
+{
+	ide_hwif_t *hwif = drive->hwif;
+	struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+	int unit = drive->select.b.unit;
+	if(itdev->mwdma[unit] != MWDMA_OFF)
+		it821x_program(drive, itdev->mwdma[unit]);
+	else if(itdev->udma[unit] != UDMA_OFF && itdev->timing10)
+		it821x_program_udma(drive, itdev->udma[unit]);
+	ide_dma_start(drive);
+}
+
+/**
+ *	it821x_dma_write	-	DMA hook
+ *	@drive: drive for DMA stop
+ *
+ *	The IT821x has a single timing register for MWDMA and for PIO
+ *	operations. As we flip back and forth we have to reload the
+ *	clock.
+ */
+
+static int it821x_dma_end(ide_drive_t *drive)
+{
+	ide_hwif_t *hwif = drive->hwif;
+	int unit = drive->select.b.unit;
+	struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+	int ret = __ide_dma_end(drive);
+	if(itdev->mwdma[unit] != MWDMA_OFF)
+		it821x_program(drive, itdev->pio[unit]);
+	return ret;
+}
+
+
+/**
+ *	it821x_tune_chipset	-	set controller timings
+ *	@drive: Drive to set up
+ *	@xferspeed: speed we want to achieve
+ *
+ *	Tune the ITE chipset for the desired mode. If we can't achieve
+ *	the desired mode then tune for a lower one, but ultimately
+ *	make the thing work.
+ */
+
+static int it821x_tune_chipset (ide_drive_t *drive, byte xferspeed)
+{
+
+	ide_hwif_t *hwif	= drive->hwif;
+	struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+	u8 speed		= ide_rate_filter(it821x_ratemask(drive), xferspeed);
+
+	if(!itdev->smart) {
+		switch(speed) {
+			case XFER_PIO_4:
+			case XFER_PIO_3:
+			case XFER_PIO_2:
+			case XFER_PIO_1:
+			case XFER_PIO_0:
+				it821x_tuneproc(drive, (speed - XFER_PIO_0));
+				break;
+			/* MWDMA tuning is really hard because our MWDMA and PIO
+			   timings are kept in the same place. We can switch in the
+			   host dma on/off callbacks */
+			case XFER_MW_DMA_2:
+			case XFER_MW_DMA_1:
+			case XFER_MW_DMA_0:
+				it821x_tune_mwdma(drive, (speed - XFER_MW_DMA_0));
+				break;
+			case XFER_UDMA_6:
+			case XFER_UDMA_5:
+			case XFER_UDMA_4:
+			case XFER_UDMA_3:
+			case XFER_UDMA_2:
+			case XFER_UDMA_1:
+			case XFER_UDMA_0:
+				it821x_tune_udma(drive, (speed - XFER_UDMA_0));
+				break;
+			default:
+				return 1;
+		}
+	}
+	/*
+	 *	In smart mode the clocking is done by the host controller
+	 * 	snooping the mode we picked. The rest of it is not our problem
+	 */
+	return ide_config_drive_speed(drive, speed);
+}
+
+/**
+ *	config_chipset_for_dma	-	configure for DMA
+ *	@drive: drive to configure
+ *
+ *	Called by the IDE layer when it wants the timings set up.
+ */
+
+static int config_chipset_for_dma (ide_drive_t *drive)
+{
+	u8 speed	= ide_dma_speed(drive, it821x_ratemask(drive));
+
+	config_it821x_chipset_for_pio(drive, !speed);
+	it821x_tune_chipset(drive, speed);
+	return ide_dma_enable(drive);
+}
+
+/**
+ *	it821x_configure_drive_for_dma	-	set up for DMA transfers
+ *	@drive: drive we are going to set up
+ *
+ *	Set up the drive for DMA, tune the controller and drive as
+ *	required. If the drive isn't suitable for DMA or we hit
+ *	other problems then we will drop down to PIO and set up
+ *	PIO appropriately
+ */
+
+static int it821x_config_drive_for_dma (ide_drive_t *drive)
+{
+	ide_hwif_t *hwif	= drive->hwif;
+
+	if (ide_use_dma(drive)) {
+		if (config_chipset_for_dma(drive))
+			return hwif->ide_dma_on(drive);
+	}
+	config_it821x_chipset_for_pio(drive, 1);
+	return hwif->ide_dma_off_quietly(drive);
+}
+
+/**
+ *	ata66_it821x	-	check for 80 pin cable
+ *	@hwif: interface to check
+ *
+ *	Check for the presence of an ATA66 capable cable on the
+ *	interface. Problematic as it seems some cards don't have
+ *	the needed logic onboard.
+ */
+
+static unsigned int __devinit ata66_it821x(ide_hwif_t *hwif)
+{
+	/* The reference driver also only does disk side */
+	return 1;
+}
+
+/**
+ *	it821x_fixup	-	post init callback
+ *	@hwif: interface
+ *
+ *	This callback is run after the drives have been probed but
+ *	before anything gets attached. It allows drivers to do any
+ *	final tuning that is needed, or fixups to work around bugs.
+ */
+
+static void __devinit it821x_fixups(ide_hwif_t *hwif)
+{
+	struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+	int i;
+
+	if(!itdev->smart) {
+		/*
+		 *	If we are in pass through mode then not much
+		 *	needs to be done, but we do bother to clear the
+		 *	IRQ mask as we may well be in PIO (eg rev 0x10)
+		 *	for now and we know unmasking is safe on this chipset.
+		 */
+		for (i = 0; i < 2; i++) {
+			ide_drive_t *drive = &hwif->drives[i];
+			if(drive->present)
+				drive->unmask = 1;
+		}
+		return;
+	}
+	/*
+	 *	Perform fixups on smart mode. We need to "lose" some
+	 *	capabilities the firmware lacks but does not filter, and
+	 *	also patch up some capability bits that it forgets to set
+	 *	in RAID mode.
+	 */
+
+	for(i = 0; i < 2; i++) {
+		ide_drive_t *drive = &hwif->drives[i];
+		struct hd_driveid *id;
+		u16 *idbits;
+
+		if(!drive->present)
+			continue;
+		id = drive->id;
+		idbits = (u16 *)drive->id;
+
+		/* Check for RAID v native */
+		if(strstr(id->model, "Integrated Technology Express")) {
+			/* In raid mode the ident block is slightly buggy
+			   We need to set the bits so that the IDE layer knows
+			   LBA28. LBA48 and DMA ar valid */
+			id->capability |= 3;		/* LBA28, DMA */
+			id->command_set_2 |= 0x0400;	/* LBA48 valid */
+			id->cfs_enable_2 |= 0x0400;	/* LBA48 on */
+			/* Reporting logic */
+			printk(KERN_INFO "%s: IT8212 %sRAID %d volume",
+				drive->name,
+				idbits[147] ? "Bootable ":"",
+				idbits[129]);
+				if(idbits[129] != 1)
+					printk("(%dK stripe)", idbits[146]);
+				printk(".\n");
+			/* Now the core code will have wrongly decided no DMA
+			   so we need to fix this */
+			hwif->ide_dma_off_quietly(drive);
+#ifdef CONFIG_IDEDMA_ONLYDISK
+			if (drive->media == ide_disk)
+#endif
+				hwif->ide_dma_check(drive);
+		} else {
+			/* Non RAID volume. Fixups to stop the core code
+			   doing unsupported things */
+			id->field_valid &= 1;
+			id->queue_depth = 0;
+			id->command_set_1 = 0;
+			id->command_set_2 &= 0xC400;
+			id->cfsse &= 0xC000;
+			id->cfs_enable_1 = 0;
+			id->cfs_enable_2 &= 0xC400;
+			id->csf_default &= 0xC000;
+			id->word127 = 0;
+			id->dlf = 0;
+			id->csfo = 0;
+			id->cfa_power = 0;
+			printk(KERN_INFO "%s: Performing identify fixups.\n",
+				drive->name);
+		}
+	}
+
+}
+
+/**
+ *	init_hwif_it821x	-	set up hwif structs
+ *	@hwif: interface to set up
+ *
+ *	We do the basic set up of the interface structure. The IT8212
+ *	requires several custom handlers so we override the default
+ *	ide DMA handlers appropriately
+ */
+
+static void __devinit init_hwif_it821x(ide_hwif_t *hwif)
+{
+	struct it821x_dev *idev = kmalloc(sizeof(struct it821x_dev), GFP_KERNEL);
+	u8 conf;
+
+	if(idev == NULL) {
+		printk(KERN_ERR "it821x: out of memory, falling back to legacy behaviour.\n");
+		goto fallback;
+	}
+	memset(idev, 0, sizeof(struct it821x_dev));
+	ide_set_hwifdata(hwif, idev);
+
+	pci_read_config_byte(hwif->pci_dev, 0x50, &conf);
+	if(conf & 1) {
+		idev->smart = 1;
+		hwif->atapi_dma = 0;
+		/* Long I/O's although allowed in LBA48 space cause the
+		   onboard firmware to enter the twighlight zone */
+		hwif->rqsize = 256;
+	}
+
+	/* Pull the current clocks from 0x50 also */
+	if (conf & (1 << (1 + hwif->channel)))
+		idev->clock_mode = ATA_50;
+	else
+		idev->clock_mode = ATA_66;
+
+	idev->want[0][1] = ATA_ANY;
+	idev->want[1][1] = ATA_ANY;
+
+	/*
+	 *	Not in the docs but according to the reference driver
+	 *	this is neccessary.
+	 */
+
+	pci_read_config_byte(hwif->pci_dev, 0x08, &conf);
+	if(conf == 0x10) {
+		idev->timing10 = 1;
+		hwif->atapi_dma = 0;
+		if(!idev->smart)
+			printk(KERN_WARNING "it821x: Revision 0x10, workarounds activated.\n");
+	}
+
+	hwif->speedproc = &it821x_tune_chipset;
+	hwif->tuneproc	= &it821x_tuneproc;
+
+	/* MWDMA/PIO clock switching for pass through mode */
+	if(!idev->smart) {
+		hwif->dma_start = &it821x_dma_start;
+		hwif->ide_dma_end = &it821x_dma_end;
+	}
+
+	hwif->drives[0].autotune = 1;
+	hwif->drives[1].autotune = 1;
+
+	if (!hwif->dma_base)
+		goto fallback;
+
+	hwif->ultra_mask = 0x7f;
+	hwif->mwdma_mask = 0x07;
+	hwif->swdma_mask = 0x07;
+
+	hwif->ide_dma_check = &it821x_config_drive_for_dma;
+	if (!(hwif->udma_four))
+		hwif->udma_four = ata66_it821x(hwif);
+
+	/*
+	 *	The BIOS often doesn't set up DMA on this controller
+	 *	so we always do it.
+	 */
+
+	hwif->autodma = 1;
+	hwif->drives[0].autodma = hwif->autodma;
+	hwif->drives[1].autodma = hwif->autodma;
+	return;
+fallback:
+	hwif->autodma = 0;
+	return;
+}
+
+static void __devinit it8212_disable_raid(struct pci_dev *dev)
+{
+	/* Reset local CPU, and set BIOS not ready */
+	pci_write_config_byte(dev, 0x5E, 0x01);
+
+	/* Set to bypass mode, and reset PCI bus */
+	pci_write_config_byte(dev, 0x50, 0x00);
+	pci_write_config_word(dev, PCI_COMMAND,
+			      PCI_COMMAND_PARITY | PCI_COMMAND_IO |
+			      PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER);
+	pci_write_config_word(dev, 0x40, 0xA0F3);
+
+	pci_write_config_dword(dev,0x4C, 0x02040204);
+	pci_write_config_byte(dev, 0x42, 0x36);
+	pci_write_config_byte(dev, PCI_LATENCY_TIMER, 0);
+}
+
+static unsigned int __devinit init_chipset_it821x(struct pci_dev *dev, const char *name)
+{
+	u8 conf;
+	static char *mode[2] = { "pass through", "smart" };
+
+	/* Force the card into bypass mode if so requested */
+	if (it8212_noraid) {
+		printk(KERN_INFO "it8212: forcing bypass mode.\n");
+		it8212_disable_raid(dev);
+	}
+	pci_read_config_byte(dev, 0x50, &conf);
+	printk(KERN_INFO "it821x: controller in %s mode.\n", mode[conf & 1]);
+	return 0;
+}
+
+
+#define DECLARE_ITE_DEV(name_str)			\
+	{						\
+		.name		= name_str,		\
+		.init_chipset	= init_chipset_it821x,	\
+		.init_hwif	= init_hwif_it821x,	\
+		.channels	= 2,			\
+		.autodma	= AUTODMA,		\
+		.bootable	= ON_BOARD,		\
+		.fixup	 	= it821x_fixups		\
+	}
+
+static ide_pci_device_t it821x_chipsets[] __devinitdata = {
+	/* 0 */ DECLARE_ITE_DEV("IT8212"),
+};
+
+/**
+ *	it821x_init_one	-	pci layer discovery entry
+ *	@dev: PCI device
+ *	@id: ident table entry
+ *
+ *	Called by the PCI code when it finds an ITE821x controller.
+ *	We then use the IDE PCI generic helper to do most of the work.
+ */
+
+static int __devinit it821x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	ide_setup_pci_device(dev, &it821x_chipsets[id->driver_data]);
+	return 0;
+}
+
+static struct pci_device_id it821x_pci_tbl[] = {
+	{ PCI_VENDOR_ID_ITE, PCI_DEVICE_ID_ITE_8211,  PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
+	{ PCI_VENDOR_ID_ITE, PCI_DEVICE_ID_ITE_8212,  PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
+	{ 0, },
+};
+
+MODULE_DEVICE_TABLE(pci, it821x_pci_tbl);
+
+static struct pci_driver driver = {
+	.name		= "ITE821x IDE",
+	.id_table	= it821x_pci_tbl,
+	.probe		= it821x_init_one,
+};
+
+static int __init it821x_ide_init(void)
+{
+	return ide_pci_register_driver(&driver);
+}
+
+module_init(it821x_ide_init);
+
+module_param_named(noraid, it8212_noraid, int, S_IRUGO);
+MODULE_PARM_DESC(it8212_noraid, "Force card into bypass mode");
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("PCI driver module for the ITE 821x");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index a3961e1d518..1e0bc6a8d65 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1815,6 +1815,8 @@
 #define PCI_VENDOR_ID_ITE		0x1283
 #define PCI_DEVICE_ID_ITE_IT8172G	0x8172
 #define PCI_DEVICE_ID_ITE_IT8172G_AUDIO 0x0801
+#define PCI_DEVICE_ID_ITE_8211		0x8211
+#define PCI_DEVICE_ID_ITE_8212		0x8212
 #define PCI_DEVICE_ID_ITE_8872		0x8872
 #define PCI_DEVICE_ID_ITE_IT8330G_0	0xe886
 
-- 
cgit v1.2.3-70-g09d2


From c47abbbffd17a7e774ec1ef952a1c3621a9cb13b Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Mon, 27 Jun 2005 15:24:31 -0700
Subject: [PATCH] ide: sensible probing for PCI systems

Old ISA/VESA systems sometimes put tertiary IDE controllers at addresses
0x1e8, 0x168, 0x1e0 or 0x160.  Linux thus probes these addresses on x86
systems.  Unfortunately some PCI systems now use these addresses for other
purposes which leads to users seeing minute plus hangs during boot or even
crashes.

The following patch (again has been in Fedora for a while) only probes the
obscure legacy ISA ports on machinea that are pre-PCI.  This seems to keep
everyone happy and if there is someone with that utterly weird corner case
the ide= command line still provides a get out of jail card.
Unsurprisingly we've not found anyone so affected.

Signed-off-by: Alan Cox <alan@redhat.com>
Acked-by: Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/ide.h | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/ide.h b/include/asm-i386/ide.h
index 859ebf4da63..79dfab87135 100644
--- a/include/asm-i386/ide.h
+++ b/include/asm-i386/ide.h
@@ -41,13 +41,17 @@ static __inline__ int ide_default_irq(unsigned long base)
 
 static __inline__ unsigned long ide_default_io_base(int index)
 {
+	if (pci_find_device(PCI_ANY_ID, PCI_ANY_ID, NULL) == NULL) {
+		switch(index) {
+			case 2: return 0x1e8;
+			case 3: return 0x168;
+			case 4: return 0x1e0;
+			case 5: return 0x160;
+		}
+	}
 	switch (index) {
 		case 0:	return 0x1f0;
 		case 1:	return 0x170;
-		case 2: return 0x1e8;
-		case 3: return 0x168;
-		case 4: return 0x1e0;
-		case 5: return 0x160;
 		default:
 			return 0;
 	}
-- 
cgit v1.2.3-70-g09d2


From 1ad275e3e7d253d44f03868e85977c908e334fed Mon Sep 17 00:00:00 2001
From: Dominik Brodowski <linux@dominikbrodowski.net>
Date: Mon, 27 Jun 2005 16:28:06 -0700
Subject: [PATCH] pcmcia: device and driver matching

The actual matching of pcmcia drivers and pcmcia devices.  The original
version of this was written by David Woodhouse.

Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/pcmcia/ds.c             | 123 +++++++++++++++++++++++++++-
 include/linux/mod_devicetable.h |  33 ++++++++
 include/pcmcia/device_id.h      | 175 ++++++++++++++++++++++++++++++++++++++++
 include/pcmcia/ds.h             |   7 +-
 4 files changed, 336 insertions(+), 2 deletions(-)
 create mode 100644 include/pcmcia/device_id.h

(limited to 'include')

diff --git a/drivers/pcmcia/ds.c b/drivers/pcmcia/ds.c
index 35d479b0df6..5701b93b2dd 100644
--- a/drivers/pcmcia/ds.c
+++ b/drivers/pcmcia/ds.c
@@ -101,6 +101,9 @@ struct pcmcia_bus_socket {
 	u8			device_count; /* the number of devices, used
 					       * only internally and subject
 					       * to incorrectness and change */
+
+	u8			device_add_pending;
+	struct work_struct	device_add;
 };
 static spinlock_t pcmcia_dev_list_lock;
 
@@ -512,6 +515,10 @@ static struct pcmcia_device * pcmcia_device_add(struct pcmcia_bus_socket *s, uns
 
 	down(&device_add_lock);
 
+	/* max of 2 devices per card */
+	if (s->device_count == 2)
+		goto err_put;
+
 	p_dev = kmalloc(sizeof(struct pcmcia_device), GFP_KERNEL);
 	if (!p_dev)
 		goto err_put;
@@ -537,6 +544,8 @@ static struct pcmcia_device * pcmcia_device_add(struct pcmcia_bus_socket *s, uns
 	list_add_tail(&p_dev->socket_device_list, &s->devices_list);
 	spin_unlock_irqrestore(&pcmcia_dev_list_lock, flags);
 
+	pcmcia_device_query(p_dev);
+
 	if (device_register(&p_dev->dev)) {
 		spin_lock_irqsave(&pcmcia_dev_list_lock, flags);
 		list_del(&p_dev->socket_device_list);
@@ -591,14 +600,123 @@ static int pcmcia_card_add(struct pcmcia_socket *s)
 }
 
 
+static void pcmcia_delayed_add_pseudo_device(void *data)
+{
+	struct pcmcia_bus_socket *s = data;
+	pcmcia_device_add(s, 0);
+	s->device_add_pending = 0;
+}
+
+static inline void pcmcia_add_pseudo_device(struct pcmcia_bus_socket *s)
+{
+	if (!s->device_add_pending) {
+		schedule_work(&s->device_add);
+		s->device_add_pending = 1;
+	}
+	return;
+}
+
+
+static inline int pcmcia_devmatch(struct pcmcia_device *dev,
+				  struct pcmcia_device_id *did)
+{
+	if (did->match_flags & PCMCIA_DEV_ID_MATCH_MANF_ID) {
+		if ((!dev->has_manf_id) || (dev->manf_id != did->manf_id))
+			return 0;
+	}
+
+	if (did->match_flags & PCMCIA_DEV_ID_MATCH_CARD_ID) {
+		if ((!dev->has_card_id) || (dev->card_id != did->card_id))
+			return 0;
+	}
+
+	if (did->match_flags & PCMCIA_DEV_ID_MATCH_FUNCTION) {
+		if (dev->func != did->function)
+			return 0;
+	}
+
+	if (did->match_flags & PCMCIA_DEV_ID_MATCH_PROD_ID1) {
+		if (!dev->prod_id[0])
+			return 0;
+		if (strcmp(did->prod_id[0], dev->prod_id[0]))
+			return 0;
+	}
+
+	if (did->match_flags & PCMCIA_DEV_ID_MATCH_PROD_ID2) {
+		if (!dev->prod_id[1])
+			return 0;
+		if (strcmp(did->prod_id[1], dev->prod_id[1]))
+			return 0;
+	}
+
+	if (did->match_flags & PCMCIA_DEV_ID_MATCH_PROD_ID3) {
+		if (!dev->prod_id[2])
+			return 0;
+		if (strcmp(did->prod_id[2], dev->prod_id[2]))
+			return 0;
+	}
+
+	if (did->match_flags & PCMCIA_DEV_ID_MATCH_PROD_ID4) {
+		if (!dev->prod_id[3])
+			return 0;
+		if (strcmp(did->prod_id[3], dev->prod_id[3]))
+			return 0;
+	}
+
+	if (did->match_flags & PCMCIA_DEV_ID_MATCH_DEVICE_NO) {
+		/* handle pseudo multifunction devices:
+		 * there are at most two pseudo multifunction devices.
+		 * if we're matching against the first, schedule a
+		 * call which will then check whether there are two
+		 * pseudo devices, and if not, add the second one.
+		 */
+		if (dev->device_no == 0)
+			pcmcia_add_pseudo_device(dev->socket->pcmcia);
+
+		if (dev->device_no != did->device_no)
+			return 0;
+	}
+
+	if (did->match_flags & PCMCIA_DEV_ID_MATCH_FUNC_ID) {
+		if ((!dev->has_func_id) || (dev->func_id != did->func_id))
+			return 0;
+
+		/* if this is a pseudo-multi-function device,
+		 * we need explicit matches */
+		if (did->match_flags & PCMCIA_DEV_ID_MATCH_DEVICE_NO)
+			return 0;
+		if (dev->device_no)
+			return 0;
+
+		/* also, FUNC_ID matching needs to be activated by userspace
+		 * after it has re-checked that there is no possible module
+		 * with a prod_id/manf_id/card_id match.
+		 */
+		if (!dev->allow_func_id_match)
+			return 0;
+	}
+
+	dev->dev.driver_data = (void *) did;
+
+	return 1;
+}
+
+
 static int pcmcia_bus_match(struct device * dev, struct device_driver * drv) {
 	struct pcmcia_device * p_dev = to_pcmcia_dev(dev);
 	struct pcmcia_driver * p_drv = to_pcmcia_drv(drv);
+	struct pcmcia_device_id *did = p_drv->id_table;
 
 	/* matching by cardmgr */
 	if (p_dev->cardmgr == p_drv)
 		return 1;
 
+	while (did && did->match_flags) {
+		if (pcmcia_devmatch(p_dev, did))
+			return 1;
+		did++;
+	}
+
 	return 0;
 }
 
@@ -922,7 +1040,9 @@ static int bind_request(struct pcmcia_bus_socket *s, bind_info_t *bind_info)
 rescan:
 	p_dev->cardmgr = p_drv;
 
-	pcmcia_device_query(p_dev);
+	/* if a driver is already running, we can abort */
+	if (p_dev->dev.driver)
+		goto err_put_module;
 
 	/*
 	 * Prevent this racing with a card insertion.
@@ -1595,6 +1715,7 @@ static int __devinit pcmcia_bus_add_socket(struct class_device *class_dev)
 
 	init_waitqueue_head(&s->queue);
 	INIT_LIST_HEAD(&s->devices_list);
+	INIT_WORK(&s->device_add, pcmcia_delayed_add_pseudo_device, s);
 
 	/* Set up hotline to Card Services */
 	s->callback.owner = THIS_MODULE;
diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h
index d6eb7b2efc0..e9651cd8310 100644
--- a/include/linux/mod_devicetable.h
+++ b/include/linux/mod_devicetable.h
@@ -175,4 +175,37 @@ struct serio_device_id {
 };
 
 
+/* PCMCIA */
+
+struct pcmcia_device_id {
+	__u16		match_flags;
+
+	__u16		manf_id;
+	__u16 		card_id;
+
+	__u8  		func_id;
+
+	/* for real multi-function devices */
+	__u8  		function;
+
+	/* for pseude multi-function devices */
+	__u8  		device_no;
+
+	const char *	prod_id[4];
+	__u32 		prod_id_hash[4];
+
+	/* not matched against */
+	kernel_ulong_t	driver_info;
+};
+
+#define PCMCIA_DEV_ID_MATCH_MANF_ID	0x0001
+#define PCMCIA_DEV_ID_MATCH_CARD_ID	0x0002
+#define PCMCIA_DEV_ID_MATCH_FUNC_ID	0x0004
+#define PCMCIA_DEV_ID_MATCH_FUNCTION	0x0008
+#define PCMCIA_DEV_ID_MATCH_PROD_ID1	0x0010
+#define PCMCIA_DEV_ID_MATCH_PROD_ID2	0x0020
+#define PCMCIA_DEV_ID_MATCH_PROD_ID3	0x0040
+#define PCMCIA_DEV_ID_MATCH_PROD_ID4	0x0080
+#define PCMCIA_DEV_ID_MATCH_DEVICE_NO	0x0100
+
 #endif /* LINUX_MOD_DEVICETABLE_H */
diff --git a/include/pcmcia/device_id.h b/include/pcmcia/device_id.h
new file mode 100644
index 00000000000..acf68656de3
--- /dev/null
+++ b/include/pcmcia/device_id.h
@@ -0,0 +1,175 @@
+/*
+ * Copyright (2003-2004) 	Dominik Brodowski <linux@brodo.de>
+ *				David Woodhouse
+ *
+ * License: GPL v2
+ */
+
+#define PCMCIA_DEVICE_MANF_CARD(manf, card) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_MANF_ID| \
+			PCMCIA_DEV_ID_MATCH_CARD_ID, \
+	.manf_id = (manf), \
+	.card_id = (card), }
+
+#define PCMCIA_DEVICE_FUNC_ID(func) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_FUNC_ID, \
+	.func_id = (func), }
+
+#define PCMCIA_DEVICE_PROD_ID1(v1, vh1) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1, \
+	.prod_id = { (v1), NULL, NULL, NULL }, \
+	.prod_id_hash = { (vh1), 0, 0, 0 }, }
+
+#define PCMCIA_DEVICE_PROD_ID2(v2, vh2) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID2, \
+	.prod_id = { NULL, (v2), NULL, NULL },  \
+	.prod_id_hash = { 0, (vh2), 0, 0 }, }
+
+#define PCMCIA_DEVICE_PROD_ID12(v1, v2, vh1, vh2) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID2, \
+	.prod_id = { (v1), (v2), NULL, NULL }, \
+	.prod_id_hash = { (vh1), (vh2), 0, 0 }, }
+
+#define PCMCIA_DEVICE_PROD_ID13(v1, v3, vh1, vh3) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID3, \
+	.prod_id = { (v1), NULL, (v3), NULL }, \
+	.prod_id_hash = { (vh1), 0, (vh3), 0 }, }
+
+#define PCMCIA_DEVICE_PROD_ID14(v1, v4, vh1, vh4) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID4, \
+	.prod_id = { (v1), NULL, NULL, (v4) }, \
+	.prod_id_hash = { (vh1), 0, 0, (vh4) }, }
+
+#define PCMCIA_DEVICE_PROD_ID123(v1, v2, v3, vh1, vh2, vh3) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID2| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID3, \
+	.prod_id = { (v1), (v2), (v3), NULL },\
+	.prod_id_hash = { (vh1), (vh2), (vh3), 0 }, }
+
+#define PCMCIA_DEVICE_PROD_ID124(v1, v2, v4, vh1, vh2, vh4) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID2| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID4, \
+	.prod_id = { (v1), (v2), NULL, (v4) }, \
+	.prod_id_hash = { (vh1), (vh2), 0, (vh4) }, }
+
+#define PCMCIA_DEVICE_PROD_ID134(v1, v3, v4, vh1, vh3, vh4) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID3| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID4, \
+	.prod_id = { (v1), NULL, (v3), (v4) }, \
+	.prod_id_hash = { (vh1), 0, (vh3), (vh4) }, }
+
+#define PCMCIA_DEVICE_PROD_ID1234(v1, v2, v3, v4, vh1, vh2, vh3, vh4) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID2| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID3| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID4, \
+	.prod_id = { (v1), (v2), (v3), (v4) }, \
+	.prod_id_hash = { (vh1), (vh2), (vh3), (vh4) }, }
+
+
+/* multi-function devices */
+
+#define PCMCIA_MFC_DEVICE_MANF_CARD(mfc, manf, card) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_MANF_ID| \
+			PCMCIA_DEV_ID_MATCH_CARD_ID| \
+			PCMCIA_DEV_ID_MATCH_FUNCTION, \
+	.manf_id = (manf), \
+	.card_id = (card), \
+	.function = (mfc), }
+
+#define PCMCIA_MFC_DEVICE_PROD_ID1(mfc, v1, vh1) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_FUNCTION, \
+	.prod_id = { (v1), NULL, NULL, NULL }, \
+	.prod_id_hash = { (vh1), 0, 0, 0 }, \
+	.function = (mfc), }
+
+#define PCMCIA_MFC_DEVICE_PROD_ID2(mfc, v2, vh2) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID2| \
+			PCMCIA_DEV_ID_MATCH_FUNCTION, \
+	.prod_id = { NULL, (v2), NULL, NULL },  \
+	.prod_id_hash = { 0, (vh2), 0, 0 }, \
+	.function = (mfc), }
+
+#define PCMCIA_MFC_DEVICE_PROD_ID12(mfc, v1, v2, vh1, vh2) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID2| \
+			PCMCIA_DEV_ID_MATCH_FUNCTION, \
+	.prod_id = { (v1), (v2), NULL, NULL }, \
+	.prod_id_hash = { (vh1), (vh2), 0, 0 }, \
+	.function = (mfc), }
+
+#define PCMCIA_MFC_DEVICE_PROD_ID13(mfc, v1, v3, vh1, vh3) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID3| \
+			PCMCIA_DEV_ID_MATCH_FUNCTION, \
+	.prod_id = { (v1), NULL, (v3), NULL }, \
+	.prod_id_hash = { (vh1), 0, (vh3), 0 }, \
+	.function = (mfc), }
+
+#define PCMCIA_MFC_DEVICE_PROD_ID123(mfc, v1, v2, v3, vh1, vh2, vh3) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID2| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID3| \
+			PCMCIA_DEV_ID_MATCH_FUNCTION, \
+	.prod_id = { (v1), (v2), (v3), NULL },\
+	.prod_id_hash = { (vh1), (vh2), (vh3), 0 }, \
+	.function = (mfc), }
+
+/* pseudo multi-function devices */
+
+#define PCMCIA_PFC_DEVICE_MANF_CARD(mfc, manf, card) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_MANF_ID| \
+			PCMCIA_DEV_ID_MATCH_CARD_ID| \
+			PCMCIA_DEV_ID_MATCH_DEVICE_NO, \
+	.manf_id = (manf), \
+	.card_id = (card), \
+	.device_no = (mfc), }
+
+#define PCMCIA_PFC_DEVICE_PROD_ID1(mfc, v1, vh1) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_DEVICE_NO, \
+	.prod_id = { (v1), NULL, NULL, NULL }, \
+	.prod_id_hash = { (vh1), 0, 0, 0 }, \
+	.device_no = (mfc), }
+
+#define PCMCIA_PFC_DEVICE_PROD_ID2(mfc, v2, vh2) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID2| \
+			PCMCIA_DEV_ID_MATCH_DEVICE_NO, \
+	.prod_id = { NULL, (v2), NULL, NULL },  \
+	.prod_id_hash = { 0, (vh2), 0, 0 }, \
+	.device_no = (mfc), }
+
+#define PCMCIA_PFC_DEVICE_PROD_ID12(mfc, v1, v2, vh1, vh2) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID2| \
+			PCMCIA_DEV_ID_MATCH_DEVICE_NO, \
+	.prod_id = { (v1), (v2), NULL, NULL }, \
+	.prod_id_hash = { (vh1), (vh2), 0, 0 }, \
+	.device_no = (mfc), }
+
+#define PCMCIA_PFC_DEVICE_PROD_ID13(mfc, v1, v3, vh1, vh3) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID3| \
+			PCMCIA_DEV_ID_MATCH_DEVICE_NO, \
+	.prod_id = { (v1), NULL, (v3), NULL }, \
+	.prod_id_hash = { (vh1), 0, (vh3), 0 }, \
+	.device_no = (mfc), }
+
+#define PCMCIA_PFC_DEVICE_PROD_ID123(mfc, v1, v2, v3, vh1, vh2, vh3) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID2| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID3| \
+			PCMCIA_DEV_ID_MATCH_DEVICE_NO, \
+	.prod_id = { (v1), (v2), (v3), NULL },\
+	.prod_id_hash = { (vh1), (vh2), (vh3), 0 }, \
+	.device_no = (mfc), }
+
+
+#define PCMCIA_DEVICE_NULL { .match_flags = 0, }
diff --git a/include/pcmcia/ds.h b/include/pcmcia/ds.h
index 312fd958c90..c267edde9d0 100644
--- a/include/pcmcia/ds.h
+++ b/include/pcmcia/ds.h
@@ -18,6 +18,8 @@
 
 #include <pcmcia/bulkmem.h>
 #include <pcmcia/cs_types.h>
+#include <pcmcia/device_id.h>
+#include <linux/mod_devicetable.h>
 
 typedef struct tuple_parse_t {
     tuple_t		tuple;
@@ -135,6 +137,7 @@ struct pcmcia_driver {
 	dev_link_t		*(*attach)(void);
 	void			(*detach)(dev_link_t *);
 	struct module		*owner;
+	struct pcmcia_device_id	*id_table;
 	struct device_driver	drv;
 };
 
@@ -173,7 +176,9 @@ struct pcmcia_device {
 	u8			has_manf_id:1;
 	u8			has_card_id:1;
 	u8			has_func_id:1;
-	u8			reserved:5;
+
+	u8			allow_func_id_match:1;
+	u8			reserved:4;
 
 	u8			func_id;
 	u16			manf_id;
-- 
cgit v1.2.3-70-g09d2


From ea7b38825bba66a81745a706da70a1c81adc95bd Mon Sep 17 00:00:00 2001
From: Dominik Brodowski <linux@dominikbrodowski.net>
Date: Mon, 27 Jun 2005 16:28:07 -0700
Subject: [PATCH] pcmcia: match for fake CIS

Add another match flag for devices needing a CIS override.  The driver will
only probe/attach if the CIS has been replaced before.

Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/pcmcia/ds.c             |  8 +++++
 include/linux/mod_devicetable.h |  2 ++
 include/pcmcia/device_id.h      | 74 +++++++++++++++++++++++++++++++++++++++++
 3 files changed, 84 insertions(+)

(limited to 'include')

diff --git a/drivers/pcmcia/ds.c b/drivers/pcmcia/ds.c
index 3ac7a443f66..c0611d56eab 100644
--- a/drivers/pcmcia/ds.c
+++ b/drivers/pcmcia/ds.c
@@ -733,6 +733,14 @@ static inline int pcmcia_devmatch(struct pcmcia_device *dev,
 			return 0;
 	}
 
+	if (did->match_flags & PCMCIA_DEV_ID_MATCH_FAKE_CIS) {
+		if (!dev->socket->fake_cis) {
+			/* FIXME: evaluate using firmware helpers to
+			 * automagically load it from userspace */
+			return 0;
+		}
+	}
+
 	dev->dev.driver_data = (void *) did;
 
 	return 1;
diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h
index e9651cd8310..c0106d68bb6 100644
--- a/include/linux/mod_devicetable.h
+++ b/include/linux/mod_devicetable.h
@@ -196,6 +196,7 @@ struct pcmcia_device_id {
 
 	/* not matched against */
 	kernel_ulong_t	driver_info;
+	char *		cisfile;
 };
 
 #define PCMCIA_DEV_ID_MATCH_MANF_ID	0x0001
@@ -207,5 +208,6 @@ struct pcmcia_device_id {
 #define PCMCIA_DEV_ID_MATCH_PROD_ID3	0x0040
 #define PCMCIA_DEV_ID_MATCH_PROD_ID4	0x0080
 #define PCMCIA_DEV_ID_MATCH_DEVICE_NO	0x0100
+#define PCMCIA_DEV_ID_MATCH_FAKE_CIS	0x0200
 
 #endif /* LINUX_MOD_DEVICETABLE_H */
diff --git a/include/pcmcia/device_id.h b/include/pcmcia/device_id.h
index acf68656de3..346d81ece28 100644
--- a/include/pcmcia/device_id.h
+++ b/include/pcmcia/device_id.h
@@ -171,5 +171,79 @@
 	.prod_id_hash = { (vh1), (vh2), (vh3), 0 }, \
 	.device_no = (mfc), }
 
+/* cards needing a CIS override */
+
+#define PCMCIA_DEVICE_CIS_MANF_CARD(manf, card, _cisfile) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_FAKE_CIS | \
+			PCMCIA_DEV_ID_MATCH_MANF_ID| \
+			PCMCIA_DEV_ID_MATCH_CARD_ID, \
+	.manf_id = (manf), \
+	.card_id = (card), \
+	.cisfile = (_cisfile)}
+
+#define PCMCIA_DEVICE_CIS_PROD_ID12(v1, v2, vh1, vh2, _cisfile) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_FAKE_CIS | \
+			PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID2, \
+	.prod_id = { (v1), (v2), NULL, NULL }, \
+	.prod_id_hash = { (vh1), (vh2), 0, 0 }, \
+	.cisfile = (_cisfile)}
+
+#define PCMCIA_DEVICE_CIS_PROD_ID123(v1, v2, v3, vh1, vh2, vh3, _cisfile) { \
+	.match_flags = PCMCIA_DEV_ID_MATCH_FAKE_CIS | \
+			PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID2| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID3, \
+	.prod_id = { (v1), (v2), (v3), NULL },\
+	.prod_id_hash = { (vh1), (vh2), (vh3), 0 }, \
+	.cisfile = (_cisfile)}
+
+
+#define PCMCIA_DEVICE_CIS_PROD_ID2(v2, vh2, _cisfile) { \
+	.match_flags =  PCMCIA_DEV_ID_MATCH_FAKE_CIS | \
+			PCMCIA_DEV_ID_MATCH_PROD_ID2, \
+	.prod_id = { NULL, (v2), NULL, NULL },  \
+	.prod_id_hash = { 0, (vh2), 0, 0 }, \
+	.cisfile = (_cisfile)}
+
+#define PCMCIA_PFC_DEVICE_CIS_PROD_ID12(mfc, v1, v2, vh1, vh2, _cisfile) { \
+	.match_flags =  PCMCIA_DEV_ID_MATCH_FAKE_CIS | \
+			PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID2| \
+			PCMCIA_DEV_ID_MATCH_DEVICE_NO, \
+	.prod_id = { (v1), (v2), NULL, NULL }, \
+	.prod_id_hash = { (vh1), (vh2), 0, 0 },\
+	.device_no = (mfc), \
+	.cisfile = (_cisfile)}
+
+#define PCMCIA_MFC_DEVICE_CIS_MANF_CARD(mfc, manf, card, _cisfile) { \
+	.match_flags =  PCMCIA_DEV_ID_MATCH_FAKE_CIS | \
+			PCMCIA_DEV_ID_MATCH_MANF_ID| \
+			PCMCIA_DEV_ID_MATCH_CARD_ID| \
+			PCMCIA_DEV_ID_MATCH_FUNCTION, \
+	.manf_id = (manf), \
+	.card_id = (card), \
+	.function = (mfc), \
+	.cisfile = (_cisfile)}
+
+#define PCMCIA_MFC_DEVICE_CIS_PROD_ID12(mfc, v1, v2, vh1, vh2, _cisfile) { \
+	.match_flags =  PCMCIA_DEV_ID_MATCH_FAKE_CIS | \
+			PCMCIA_DEV_ID_MATCH_PROD_ID1| \
+			PCMCIA_DEV_ID_MATCH_PROD_ID2| \
+			PCMCIA_DEV_ID_MATCH_FUNCTION, \
+	.prod_id = { (v1), (v2), NULL, NULL }, \
+	.prod_id_hash = { (vh1), (vh2), 0, 0 }, \
+	.function = (mfc), \
+	.cisfile = (_cisfile)}
+
+#define PCMCIA_MFC_DEVICE_CIS_PROD_ID4(mfc, v4, vh4, _cisfile) { \
+	.match_flags =  PCMCIA_DEV_ID_MATCH_FAKE_CIS | \
+			PCMCIA_DEV_ID_MATCH_PROD_ID4| \
+			PCMCIA_DEV_ID_MATCH_FUNCTION, \
+	.prod_id = { NULL, NULL, NULL, (v4) }, \
+	.prod_id_hash = { 0, 0, 0, (vh4) }, \
+	.function = (mfc), \
+	.cisfile = (_cisfile)}
+
 
 #define PCMCIA_DEVICE_NULL { .match_flags = 0, }
-- 
cgit v1.2.3-70-g09d2


From f602ff7eb4e44e7245bfeeba4d078144703fcd76 Mon Sep 17 00:00:00 2001
From: Dominik Brodowski <linux@dominikbrodowski.net>
Date: Mon, 27 Jun 2005 16:28:09 -0700
Subject: [PATCH] pcmcia: match "anonymous" cards

If a card doesn't provide _any_ information about itself, assume it is a
so-called "anonymous" card.  pcmciamtd will bind to it if it is configured to
do so.

Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/pcmcia/ds.c             | 9 +++++++++
 include/linux/mod_devicetable.h | 1 +
 2 files changed, 10 insertions(+)

(limited to 'include')

diff --git a/drivers/pcmcia/ds.c b/drivers/pcmcia/ds.c
index f657a2a77b2..66680699e91 100644
--- a/drivers/pcmcia/ds.c
+++ b/drivers/pcmcia/ds.c
@@ -746,6 +746,15 @@ static inline int pcmcia_devmatch(struct pcmcia_device *dev,
 		}
 	}
 
+	if (did->match_flags & PCMCIA_DEV_ID_MATCH_ANONYMOUS) {
+		int i;
+		for (i=0; i<4; i++)
+			if (dev->prod_id[i])
+				return 0;
+		if (dev->has_manf_id || dev->has_card_id || dev->has_func_id)
+			return 0;
+	}
+
 	dev->dev.driver_data = (void *) did;
 
 	return 1;
diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h
index c0106d68bb6..8a8dc82a941 100644
--- a/include/linux/mod_devicetable.h
+++ b/include/linux/mod_devicetable.h
@@ -209,5 +209,6 @@ struct pcmcia_device_id {
 #define PCMCIA_DEV_ID_MATCH_PROD_ID4	0x0080
 #define PCMCIA_DEV_ID_MATCH_DEVICE_NO	0x0100
 #define PCMCIA_DEV_ID_MATCH_FAKE_CIS	0x0200
+#define PCMCIA_DEV_ID_MATCH_ANONYMOUS	0x0400
 
 #endif /* LINUX_MOD_DEVICETABLE_H */
-- 
cgit v1.2.3-70-g09d2


From e94e15f70559962f60915a09d44658b73a453fe2 Mon Sep 17 00:00:00 2001
From: Dominik Brodowski <linux@dominikbrodowski.net>
Date: Mon, 27 Jun 2005 16:28:15 -0700
Subject: [PATCH] pcmcia: cleanups

From: Adrian Bunk <bunk@stusta.de>

This patch contains the following cleanups:
- make needlessly global code static
- remove the following unneeded EXPORT_SYMBOL's:
  - ds.c: pcmcia_report_error
  - ds.c: pcmcia_bus_type

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/pcmcia/ds.c             | 10 +++++-----
 drivers/pcmcia/rsrc_nonstatic.c |  6 +++---
 include/pcmcia/cs.h             |  1 -
 include/pcmcia/ds.h             |  2 --
 4 files changed, 8 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/drivers/pcmcia/ds.c b/drivers/pcmcia/ds.c
index 5c88ff8e7d7..0de0f230045 100644
--- a/drivers/pcmcia/ds.c
+++ b/drivers/pcmcia/ds.c
@@ -60,7 +60,7 @@ MODULE_DESCRIPTION("PCMCIA Driver Services");
 MODULE_LICENSE("GPL");
 
 #ifdef DEBUG
-int ds_pc_debug;
+static int ds_pc_debug;
 
 module_param_named(pc_debug, ds_pc_debug, int, 0644);
 
@@ -108,6 +108,8 @@ struct pcmcia_bus_socket {
 };
 static spinlock_t pcmcia_dev_list_lock;
 
+static struct bus_type pcmcia_bus_type;
+
 #define DS_SOCKET_PRESENT		0x01
 #define DS_SOCKET_BUSY			0x02
 #define DS_SOCKET_REMOVAL_PENDING	0x10
@@ -218,7 +220,7 @@ static const lookup_t service_table[] = {
 };
 
 
-int pcmcia_report_error(client_handle_t handle, error_info_t *err)
+static int pcmcia_report_error(client_handle_t handle, error_info_t *err)
 {
 	int i;
 	char *serv;
@@ -248,7 +250,6 @@ int pcmcia_report_error(client_handle_t handle, error_info_t *err)
 
 	return CS_SUCCESS;
 } /* report_error */
-EXPORT_SYMBOL(pcmcia_report_error);
 
 /* end of code which was in cs.c before */
 
@@ -1901,13 +1902,12 @@ static struct class_interface pcmcia_bus_interface = {
 };
 
 
-struct bus_type pcmcia_bus_type = {
+static struct bus_type pcmcia_bus_type = {
 	.name = "pcmcia",
 	.hotplug = pcmcia_bus_hotplug,
 	.match = pcmcia_bus_match,
 	.dev_attrs = pcmcia_dev_attrs,
 };
-EXPORT_SYMBOL(pcmcia_bus_type);
 
 
 static int __init init_pcmcia_bus(void)
diff --git a/drivers/pcmcia/rsrc_nonstatic.c b/drivers/pcmcia/rsrc_nonstatic.c
index 5876bab7c14..9a0b835d612 100644
--- a/drivers/pcmcia/rsrc_nonstatic.c
+++ b/drivers/pcmcia/rsrc_nonstatic.c
@@ -601,7 +601,7 @@ static int nonstatic_adjust_io_region(struct resource *res, unsigned long r_star
 
 ======================================================================*/
 
-struct resource *nonstatic_find_io_region(unsigned long base, int num,
+static struct resource *nonstatic_find_io_region(unsigned long base, int num,
 		   unsigned long align, struct pcmcia_socket *s)
 {
 	struct resource *res = make_resource(0, num, IORESOURCE_IO, s->dev.class_id);
@@ -635,8 +635,8 @@ struct resource *nonstatic_find_io_region(unsigned long base, int num,
 	return res;
 }
 
-struct resource * nonstatic_find_mem_region(u_long base, u_long num, u_long align,
-				 int low, struct pcmcia_socket *s)
+static struct resource * nonstatic_find_mem_region(u_long base, u_long num,
+		u_long align, int low, struct pcmcia_socket *s)
 {
 	struct resource *res = make_resource(0, num, IORESOURCE_MEM, s->dev.class_id);
 	struct socket_data *s_data = s->resource_data;
diff --git a/include/pcmcia/cs.h b/include/pcmcia/cs.h
index 8d8643adc78..60a3decea82 100644
--- a/include/pcmcia/cs.h
+++ b/include/pcmcia/cs.h
@@ -417,7 +417,6 @@ int pcmcia_suspend_card(struct pcmcia_socket *skt);
 int pcmcia_resume_card(struct pcmcia_socket *skt);
 int pcmcia_eject_card(struct pcmcia_socket *skt);
 int pcmcia_insert_card(struct pcmcia_socket *skt);
-int pcmcia_report_error(client_handle_t handle, error_info_t *err);
 
 struct pcmcia_socket * pcmcia_get_socket(struct pcmcia_socket *skt);
 void pcmcia_put_socket(struct pcmcia_socket *skt);
diff --git a/include/pcmcia/ds.h b/include/pcmcia/ds.h
index c267edde9d0..2b52553f2d9 100644
--- a/include/pcmcia/ds.h
+++ b/include/pcmcia/ds.h
@@ -131,8 +131,6 @@ typedef struct dev_link_t {
 
 struct pcmcia_socket;
 
-extern struct bus_type pcmcia_bus_type;
-
 struct pcmcia_driver {
 	dev_link_t		*(*attach)(void);
 	void			(*detach)(dev_link_t *);
-- 
cgit v1.2.3-70-g09d2


From f4d7510d3d3b4501c94b4b00cf42fd58d49aeddd Mon Sep 17 00:00:00 2001
From: Jun Komuro <komurojun-mbn@nifty.com>
Date: Mon, 27 Jun 2005 16:28:44 -0700
Subject: [PATCH] pcmcia: more IDs for TDK multifunction cards

Add new pcmcia id_table for fmvj18x_cs and serial_cs.
(TDK multi-function card (NetPartner9610 and MobileNetworker3200))

Signed-off-by: Jun Komuro <komurojun-mbn@nifty.com>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/net/pcmcia/fmvj18x_cs.c | 6 +++++-
 drivers/serial/serial_cs.c      | 2 ++
 include/pcmcia/ciscode.h        | 2 ++
 3 files changed, 9 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/net/pcmcia/fmvj18x_cs.c b/drivers/net/pcmcia/fmvj18x_cs.c
index b17aedd5e5a..917adbbf0b5 100644
--- a/drivers/net/pcmcia/fmvj18x_cs.c
+++ b/drivers/net/pcmcia/fmvj18x_cs.c
@@ -435,7 +435,9 @@ static void fmvj18x_config(dev_link_t *link)
 		pcmcia_get_status(handle, &status);
 		if (status.CardState & CS_EVENT_3VCARD)
 		    link->conf.Vcc = 33; /* inserted in 3.3V slot */
-	    } else if (le16_to_cpu(buf[1]) == PRODID_TDK_GN3410) {
+	    } else if (le16_to_cpu(buf[1]) == PRODID_TDK_GN3410
+			|| le16_to_cpu(buf[1]) == PRODID_TDK_NP9610
+			|| le16_to_cpu(buf[1]) == PRODID_TDK_MN3200) {
 		/* MultiFunction Card */
 		link->conf.ConfigBase = 0x800;
 		link->conf.ConfigIndex = 0x47;
@@ -783,6 +785,8 @@ static struct pcmcia_device_id fmvj18x_ids[] = {
 	PCMCIA_DEVICE_PROD_ID1("PCMCIA MBH10302", 0x8f4005da),
 	PCMCIA_DEVICE_PROD_ID1("UBKK,V2.0", 0x90888080),
 	PCMCIA_PFC_DEVICE_PROD_ID12(0, "TDK", "GlobalNetworker 3410/3412", 0x1eae9475, 0xd9a93bed),
+	PCMCIA_PFC_DEVICE_MANF_CARD(0, 0x0105, 0x0d0a),
+	PCMCIA_PFC_DEVICE_MANF_CARD(0, 0x0105, 0x0e0a),
 	PCMCIA_DEVICE_NULL,
 };
 MODULE_DEVICE_TABLE(pcmcia, fmvj18x_ids);
diff --git a/drivers/serial/serial_cs.c b/drivers/serial/serial_cs.c
index ca86f1f668f..73a34b18866 100644
--- a/drivers/serial/serial_cs.c
+++ b/drivers/serial/serial_cs.c
@@ -785,6 +785,8 @@ static struct pcmcia_device_id serial_ids[] = {
 	PCMCIA_PFC_DEVICE_MANF_CARD(1, 0x016c, 0x0081),
 	PCMCIA_PFC_DEVICE_MANF_CARD(1, 0x021b, 0x0101),
 	PCMCIA_PFC_DEVICE_MANF_CARD(1, 0x08a1, 0xc0ab),
+	PCMCIA_PFC_DEVICE_MANF_CARD(1, 0x0105, 0x0d0a),
+	PCMCIA_PFC_DEVICE_MANF_CARD(1, 0x0105, 0x0e0a),
 	PCMCIA_PFC_DEVICE_PROD_ID123(1, "MEGAHERTZ", "CC/XJEM3288", "DATA/FAX/CELL ETHERNET MODEM", 0xf510db04, 0x04cd2988, 0x46a52d63),
 	PCMCIA_PFC_DEVICE_PROD_ID123(1, "MEGAHERTZ", "CC/XJEM3336", "DATA/FAX/CELL ETHERNET MODEM", 0xf510db04, 0x0143b773, 0x46a52d63),
 	PCMCIA_PFC_DEVICE_PROD_ID123(1, "MEGAHERTZ", "EM1144T", "PCMCIA MODEM", 0xf510db04, 0x856d66c8, 0xbd6c43ef),
diff --git a/include/pcmcia/ciscode.h b/include/pcmcia/ciscode.h
index 2000b43ece9..da19c297dd6 100644
--- a/include/pcmcia/ciscode.h
+++ b/include/pcmcia/ciscode.h
@@ -112,6 +112,8 @@
 
 #define MANFID_TDK			0x0105
 #define PRODID_TDK_CF010		0x0900
+#define PRODID_TDK_NP9610		0x0d0a
+#define PRODID_TDK_MN3200		0x0e0a
 #define PRODID_TDK_GN3410		0x4815
 
 #define MANFID_TOSHIBA			0x0098
-- 
cgit v1.2.3-70-g09d2


From 3c29976a6469b81a7858812dc2d4b8430d74004a Mon Sep 17 00:00:00 2001
From: Dominik Brodowski <linux@dominikbrodowski.net>
Date: Mon, 27 Jun 2005 16:28:46 -0700
Subject: [PATCH] pcmcia: mark parent bridge windows as resources available for
 PCMCIA devices

Automatically mark the parent PCI-PCI bridge windows as resources available
for PCMCIA usage.

Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/pcmcia/rsrc_mgr.c       |  2 +-
 drivers/pcmcia/rsrc_nonstatic.c | 58 +++++++++++++++++++++++++++++++++++++++++
 include/pcmcia/ss.h             |  3 ++-
 3 files changed, 61 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/pcmcia/rsrc_mgr.c b/drivers/pcmcia/rsrc_mgr.c
index b6843f8d300..b9269e66281 100644
--- a/drivers/pcmcia/rsrc_mgr.c
+++ b/drivers/pcmcia/rsrc_mgr.c
@@ -72,7 +72,7 @@ int pcmcia_adjust_resource_info(adjust_t *adj)
 			/* you can't use the old interface if the new
 			 * one was used before */
 			spin_lock_irqsave(&s->lock, flags);
-			if ((s->resource_setup_done) &&
+			if ((s->resource_setup_new) &&
 			    !(s->resource_setup_old)) {
 				spin_unlock_irqrestore(&s->lock, flags);
 				continue;
diff --git a/drivers/pcmcia/rsrc_nonstatic.c b/drivers/pcmcia/rsrc_nonstatic.c
index 6b463609a3a..c8f21796c59 100644
--- a/drivers/pcmcia/rsrc_nonstatic.c
+++ b/drivers/pcmcia/rsrc_nonstatic.c
@@ -768,6 +768,58 @@ static int nonstatic_adjust_resource_info(struct pcmcia_socket *s, adjust_t *adj
 	return CS_UNSUPPORTED_FUNCTION;
 }
 
+#ifdef CONFIG_PCI
+static int nonstatic_autoadd_resources(struct pcmcia_socket *s)
+{
+	struct resource *res;
+	int i, done = 0;
+
+	if (!s->cb_dev || !s->cb_dev->bus)
+		return -ENODEV;
+
+	for (i=0; i < PCI_BUS_NUM_RESOURCES; i++) {
+		res = s->cb_dev->bus->resource[i];
+		if (!res)
+			continue;
+
+		if (res->flags & IORESOURCE_IO) {
+			if (res == &ioport_resource)
+				continue;
+			printk(KERN_INFO "pcmcia: parent PCI bridge I/O window: 0x%lx - 0x%lx\n",
+			       res->start, res->end);
+			if (!adjust_io(s, ADD_MANAGED_RESOURCE, res->start, res->end))
+				done |= IORESOURCE_IO;
+
+		}
+
+		if (res->flags & IORESOURCE_MEM) {
+			if (res == &iomem_resource)
+				continue;
+			printk(KERN_INFO "pcmcia: parent PCI bridge Memory window: 0x%lx - 0x%lx\n",
+			       res->start, res->end);
+			if (!adjust_memory(s, ADD_MANAGED_RESOURCE, res->start, res->end))
+				done |= IORESOURCE_MEM;
+		}
+	}
+
+	/* if we got at least one of IO, and one of MEM, we can be glad and
+	 * activate the PCMCIA subsystem */
+	if (done & (IORESOURCE_MEM | IORESOURCE_IO))
+		s->resource_setup_done = 1;
+
+	return 0;
+}
+
+#else
+
+static inline int nonstatic_autoadd_resources(struct pcmcia_socket *s)
+{
+	return -ENODEV;
+}
+
+#endif
+
+
 static int nonstatic_init(struct pcmcia_socket *s)
 {
 	struct socket_data *data;
@@ -782,6 +834,8 @@ static int nonstatic_init(struct pcmcia_socket *s)
 
 	s->resource_data = (void *) data;
 
+	nonstatic_autoadd_resources(s);
+
 	return 0;
 }
 
@@ -862,6 +916,8 @@ static ssize_t store_io_db(struct class_device *class_dev, const char *buf, size
 		return -EINVAL;
 
 	ret = adjust_io(s, add, start_addr, end_addr);
+	if (!ret)
+		s->resource_setup_new = 1;
 
 	return ret ? ret : count;
 }
@@ -912,6 +968,8 @@ static ssize_t store_mem_db(struct class_device *class_dev, const char *buf, siz
 		return -EINVAL;
 
 	ret = adjust_memory(s, add, start_addr, end_addr);
+	if (!ret)
+		s->resource_setup_new = 1;
 
 	return ret ? ret : count;
 }
diff --git a/include/pcmcia/ss.h b/include/pcmcia/ss.h
index 67b867f31fe..f8797767051 100644
--- a/include/pcmcia/ss.h
+++ b/include/pcmcia/ss.h
@@ -216,8 +216,9 @@ struct pcmcia_socket {
 
 	/* is set to one if resource setup is done using adjust_resource_info() */
 	u8				resource_setup_old:1;
+	u8				resource_setup_new:1;
 
-	u8				reserved:6;
+	u8				reserved:5;
 
 	/* socket operations */
 	struct pccard_operations *	ops;
-- 
cgit v1.2.3-70-g09d2


From dc109497bd5799770fedfc6503119808497b3677 Mon Sep 17 00:00:00 2001
From: Dominik Brodowski <linux@dominikbrodowski.net>
Date: Mon, 27 Jun 2005 16:28:50 -0700
Subject: [PATCH] pcmcia: merge struct pcmcia_bus_socket into struct
 pcmcia_socket

Merge struct pcmcia_bus_socket into struct pcmcia_socket.

Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/pcmcia/ds.c           | 123 ++++++++++++++----------------------------
 drivers/pcmcia/ds_internal.h  |  45 ++--------------
 drivers/pcmcia/pcmcia_ioctl.c |  97 +++++++++++++++------------------
 include/pcmcia/ss.h           |  31 +++++++++--
 4 files changed, 118 insertions(+), 178 deletions(-)

(limited to 'include')

diff --git a/drivers/pcmcia/ds.c b/drivers/pcmcia/ds.c
index 85e2cb2b9e9..f796dcd64dc 100644
--- a/drivers/pcmcia/ds.c
+++ b/drivers/pcmcia/ds.c
@@ -75,7 +75,7 @@ module_param_named(pc_debug, ds_pc_debug, int, 0644);
 
 spinlock_t pcmcia_dev_list_lock;
 
-static int unbind_request(struct pcmcia_bus_socket *s);
+static int unbind_request(struct pcmcia_socket *s);
 
 /*====================================================================*/
 
@@ -313,24 +313,6 @@ static inline int pcmcia_load_firmware(struct pcmcia_device *dev, char * filenam
 /*======================================================================*/
 
 
-void pcmcia_release_bus_socket(struct kref *refcount)
-{
-	struct pcmcia_bus_socket *s = container_of(refcount, struct pcmcia_bus_socket, refcount);
-	pcmcia_put_socket(s->parent);
-	kfree(s);
-}
-
-void pcmcia_put_bus_socket(struct pcmcia_bus_socket *s)
-{
-	kref_put(&s->refcount, pcmcia_release_bus_socket);
-}
-
-struct pcmcia_bus_socket *pcmcia_get_bus_socket(struct pcmcia_bus_socket *s)
-{
-	kref_get(&s->refcount);
-	return (s);
-}
-
 /**
  * pcmcia_register_driver - register a PCMCIA driver with the bus core
  *
@@ -387,7 +369,7 @@ static void pcmcia_release_dev(struct device *dev)
 {
 	struct pcmcia_device *p_dev = to_pcmcia_dev(dev);
 	ds_dbg(1, "releasing dev %p\n", p_dev);
-	pcmcia_put_bus_socket(p_dev->socket->pcmcia);
+	pcmcia_put_socket(p_dev->socket);
 	kfree(p_dev);
 }
 
@@ -522,12 +504,12 @@ static int pcmcia_device_query(struct pcmcia_device *p_dev)
  */
 static DECLARE_MUTEX(device_add_lock);
 
-struct pcmcia_device * pcmcia_device_add(struct pcmcia_bus_socket *s, unsigned int function)
+struct pcmcia_device * pcmcia_device_add(struct pcmcia_socket *s, unsigned int function)
 {
 	struct pcmcia_device *p_dev;
 	unsigned long flags;
 
-	s = pcmcia_get_bus_socket(s);
+	s = pcmcia_get_socket(s);
 	if (!s)
 		return NULL;
 
@@ -542,18 +524,18 @@ struct pcmcia_device * pcmcia_device_add(struct pcmcia_bus_socket *s, unsigned i
 		goto err_put;
 	memset(p_dev, 0, sizeof(struct pcmcia_device));
 
-	p_dev->socket = s->parent;
+	p_dev->socket = s;
 	p_dev->device_no = (s->device_count++);
 	p_dev->func   = function;
 
 	p_dev->dev.bus = &pcmcia_bus_type;
-	p_dev->dev.parent = s->parent->dev.dev;
+	p_dev->dev.parent = s->dev.dev;
 	p_dev->dev.release = pcmcia_release_dev;
 	sprintf (p_dev->dev.bus_id, "%d.%d", p_dev->socket->sock, p_dev->device_no);
 
 	/* compat */
 	p_dev->client.client_magic = CLIENT_MAGIC;
-	p_dev->client.Socket = s->parent;
+	p_dev->client.Socket = s;
 	p_dev->client.Function = function;
 	p_dev->client.state = CLIENT_UNBOUND;
 
@@ -581,7 +563,7 @@ struct pcmcia_device * pcmcia_device_add(struct pcmcia_bus_socket *s, unsigned i
 	s->device_count--;
  err_put:
 	up(&device_add_lock);
-	pcmcia_put_bus_socket(s);
+	pcmcia_put_socket(s);
 
 	return NULL;
 }
@@ -612,7 +594,7 @@ static int pcmcia_card_add(struct pcmcia_socket *s)
 	/* this doesn't handle multifunction devices on one pcmcia function
 	 * yet. */
 	for (i=0; i < no_funcs; i++)
-		pcmcia_device_add(s->pcmcia, i);
+		pcmcia_device_add(s, i);
 
 	return (ret);
 }
@@ -620,12 +602,12 @@ static int pcmcia_card_add(struct pcmcia_socket *s)
 
 static void pcmcia_delayed_add_pseudo_device(void *data)
 {
-	struct pcmcia_bus_socket *s = data;
+	struct pcmcia_socket *s = data;
 	pcmcia_device_add(s, 0);
 	s->pcmcia_state.device_add_pending = 0;
 }
 
-static inline void pcmcia_add_pseudo_device(struct pcmcia_bus_socket *s)
+static inline void pcmcia_add_pseudo_device(struct pcmcia_socket *s)
 {
 	if (!s->pcmcia_state.device_add_pending) {
 		schedule_work(&s->device_add);
@@ -650,7 +632,7 @@ static void pcmcia_bus_rescan(struct pcmcia_socket *skt)
 
 	/* must be called with skt_sem held */
 	spin_lock_irqsave(&pcmcia_dev_list_lock, flags);
-	if (list_empty(&skt->pcmcia->devices_list))
+	if (list_empty(&skt->devices_list))
 		no_devices=1;
 	spin_unlock_irqrestore(&pcmcia_dev_list_lock, flags);
 
@@ -727,7 +709,7 @@ static inline int pcmcia_devmatch(struct pcmcia_device *dev,
 		 * pseudo devices, and if not, add the second one.
 		 */
 		if (dev->device_no == 0)
-			pcmcia_add_pseudo_device(dev->socket->pcmcia);
+			pcmcia_add_pseudo_device(dev->socket);
 
 		if (dev->device_no != did->device_no)
 			return 0;
@@ -947,21 +929,13 @@ static int send_event_callback(struct device *dev, void * _data)
 
 static int send_event(struct pcmcia_socket *s, event_t event, int priority)
 {
-	int ret = 0;
 	struct send_event_data private;
-	struct pcmcia_bus_socket *skt = pcmcia_get_bus_socket(s->pcmcia);
-
-	if (!skt)
-		return 0;
 
 	private.skt = s;
 	private.event = event;
 	private.priority = priority;
 
-	ret = bus_for_each_dev(&pcmcia_bus_type, NULL, &private, send_event_callback);
-
-	pcmcia_put_bus_socket(skt);
-	return ret;
+	return bus_for_each_dev(&pcmcia_bus_type, NULL, &private, send_event_callback);
 } /* send_event */
 
 
@@ -972,25 +946,25 @@ static int send_event(struct pcmcia_socket *s, event_t event, int priority)
 
 static int ds_event(struct pcmcia_socket *skt, event_t event, int priority)
 {
-	struct pcmcia_bus_socket *s = skt->pcmcia;
+	struct pcmcia_socket *s = pcmcia_get_socket(skt);
 	int ret = 0;
 
 	ds_dbg(1, "ds_event(0x%06x, %d, 0x%p)\n",
-	       event, priority, s);
+	       event, priority, skt);
     
 	switch (event) {
 
 	case CS_EVENT_CARD_REMOVAL:
 		s->pcmcia_state.present = 0;
 	    	send_event(skt, event, priority);
-		unbind_request(s);
-		handle_event(s, event);
+		unbind_request(skt);
+		handle_event(skt, event);
 		break;
 	
 	case CS_EVENT_CARD_INSERTION:
 		s->pcmcia_state.present = 1;
 		pcmcia_card_add(skt);
-		handle_event(s, event);
+		handle_event(skt, event);
 		break;
 
 	case CS_EVENT_EJECTION_REQUEST:
@@ -998,11 +972,13 @@ static int ds_event(struct pcmcia_socket *skt, event_t event, int priority)
 		break;
 
 	default:
-		handle_event(s, event);
+		handle_event(skt, event);
 		send_event(skt, event, priority);
 		break;
     }
 
+    pcmcia_put_socket(s);
+
     return 0;
 } /* ds_event */
 
@@ -1011,8 +987,7 @@ static int ds_event(struct pcmcia_socket *skt, event_t event, int priority)
 int pcmcia_register_client(client_handle_t *handle, client_reg_t *req)
 {
 	client_t *client = NULL;
-	struct pcmcia_socket *s;
-	struct pcmcia_bus_socket *skt = NULL;
+	struct pcmcia_socket *s = NULL;
 	struct pcmcia_device *p_dev = NULL;
 
 	/* Look for unbound client with matching dev_info */
@@ -1023,14 +998,11 @@ int pcmcia_register_client(client_handle_t *handle, client_reg_t *req)
 		if (s->state & SOCKET_CARDBUS)
 			continue;
 
-		skt = s->pcmcia;
-		if (!skt)
-			continue;
-		skt = pcmcia_get_bus_socket(skt);
-		if (!skt)
+		s = pcmcia_get_socket(s);
+		if (!s)
 			continue;
 		spin_lock_irqsave(&pcmcia_dev_list_lock, flags);
-		list_for_each_entry(p_dev, &skt->devices_list, socket_device_list) {
+		list_for_each_entry(p_dev, &s->devices_list, socket_device_list) {
 			struct pcmcia_driver *p_drv;
 			p_dev = pcmcia_get_dev(p_dev);
 			if (!p_dev)
@@ -1049,14 +1021,14 @@ int pcmcia_register_client(client_handle_t *handle, client_reg_t *req)
 			pcmcia_put_dev(p_dev);
 		}
 		spin_unlock_irqrestore(&pcmcia_dev_list_lock, flags);
-		pcmcia_put_bus_socket(skt);
+		pcmcia_put_socket(s);
 	}
  found:
 	up_read(&pcmcia_socket_list_rwsem);
 	if (!p_dev || !client)
 		return -ENODEV;
 
-	pcmcia_put_bus_socket(skt); /* safe, as we already hold a reference from bind_device */
+	pcmcia_put_socket(s); /* safe, as we already hold a reference from bind_device */
 
 	*handle = client;
 	client->state &= ~CLIENT_UNBOUND;
@@ -1106,12 +1078,12 @@ EXPORT_SYMBOL(pcmcia_register_client);
 /* unbind _all_ devices attached to a given pcmcia_bus_socket. The
  * drivers have been called with EVENT_CARD_REMOVAL before.
  */
-static int unbind_request(struct pcmcia_bus_socket *s)
+static int unbind_request(struct pcmcia_socket *s)
 {
 	struct pcmcia_device	*p_dev;
 	unsigned long		flags;
 
-	ds_dbg(2, "unbind_request(%d)\n", s->parent->sock);
+	ds_dbg(2, "unbind_request(%d)\n", s->sock);
 
 	s->device_count = 0;
 
@@ -1176,24 +1148,14 @@ static struct pcmcia_callback pcmcia_bus_callback = {
 static int __devinit pcmcia_bus_add_socket(struct class_device *class_dev)
 {
 	struct pcmcia_socket *socket = class_get_devdata(class_dev);
-	struct pcmcia_bus_socket *s;
 	int ret;
 
-	s = kmalloc(sizeof(struct pcmcia_bus_socket), GFP_KERNEL);
-	if(!s)
-		return -ENOMEM;
-	memset(s, 0, sizeof(struct pcmcia_bus_socket));
-
-	/* get reference to parent socket */
-	s->parent = pcmcia_get_socket(socket);
-	if (!s->parent) {
+	socket = pcmcia_get_socket(socket);
+	if (!socket) {
 		printk(KERN_ERR "PCMCIA obtaining reference to socket %p failed\n", socket);
-		kfree (s);
 		return -ENODEV;
 	}
 
-	kref_init(&s->refcount);
-    
 	/*
 	 * Ugly. But we want to wait for the socket threads to have started up.
 	 * We really should let the drivers themselves drive some of this..
@@ -1201,19 +1163,17 @@ static int __devinit pcmcia_bus_add_socket(struct class_device *class_dev)
 	msleep(250);
 
 #ifdef CONFIG_PCMCIA_IOCTL
-	init_waitqueue_head(&s->queue);
+	init_waitqueue_head(&socket->queue);
 #endif
-	INIT_LIST_HEAD(&s->devices_list);
-	INIT_WORK(&s->device_add, pcmcia_delayed_add_pseudo_device, s);
-
-	/* Set up hotline to Card Services */
-	socket->pcmcia = s;
+	INIT_LIST_HEAD(&socket->devices_list);
+	INIT_WORK(&socket->device_add, pcmcia_delayed_add_pseudo_device, socket);
+	memset(&socket->pcmcia_state, 0, sizeof(u8));
+	socket->device_count = 0;
 
 	ret = pccard_register_pcmcia(socket, &pcmcia_bus_callback);
 	if (ret) {
 		printk(KERN_ERR "PCMCIA registration PCCard core failed for socket %p\n", socket);
-		pcmcia_put_bus_socket(s);
-		socket->pcmcia = NULL;
+		pcmcia_put_socket(socket);
 		return (ret);
 	}
 
@@ -1224,14 +1184,13 @@ static void pcmcia_bus_remove_socket(struct class_device *class_dev)
 {
 	struct pcmcia_socket *socket = class_get_devdata(class_dev);
 
-	if (!socket || !socket->pcmcia)
+	if (!socket)
 		return;
 
+	socket->pcmcia_state.dead = 1;
 	pccard_register_pcmcia(socket, NULL);
 
-	socket->pcmcia->pcmcia_state.dead = 1;
-	pcmcia_put_bus_socket(socket->pcmcia);
-	socket->pcmcia = NULL;
+	pcmcia_put_socket(socket);
 
 	return;
 }
diff --git a/drivers/pcmcia/ds_internal.h b/drivers/pcmcia/ds_internal.h
index 2c3bb189dec..d359bd25a51 100644
--- a/drivers/pcmcia/ds_internal.h
+++ b/drivers/pcmcia/ds_internal.h
@@ -1,56 +1,21 @@
 /* ds_internal.h - internal header for 16-bit PCMCIA devices management */
 
-struct user_info_t;
-
-/* Socket state information */
-struct pcmcia_bus_socket {
-	struct kref		refcount;
-	struct pcmcia_socket	*parent;
-
-	/* the PCMCIA devices connected to this socket (normally one, more
-	 * for multifunction devices: */
-	struct list_head	devices_list;
-	u8			device_count; /* the number of devices, used
-					       * only internally and subject
-					       * to incorrectness and change */
-
-	struct {
-		u8		present:1,
-				busy:1,
-				dead:1,
-				device_add_pending:1,
-				reserved:4;
-	} 			pcmcia_state;
-
-	struct work_struct	device_add;
-
-
-#ifdef CONFIG_PCMCIA_IOCTL
-	struct user_info_t	*user;
-	wait_queue_head_t	queue;
-#endif
-};
 extern spinlock_t pcmcia_dev_list_lock;
-
 extern struct bus_type pcmcia_bus_type;
 
-
 extern struct pcmcia_device * pcmcia_get_dev(struct pcmcia_device *p_dev);
 extern void pcmcia_put_dev(struct pcmcia_device *p_dev);
 
-struct pcmcia_bus_socket *pcmcia_get_bus_socket(struct pcmcia_bus_socket *s);
-void pcmcia_put_bus_socket(struct pcmcia_bus_socket *s);
-
-struct pcmcia_device * pcmcia_device_add(struct pcmcia_bus_socket *s, unsigned int function);
+struct pcmcia_device * pcmcia_device_add(struct pcmcia_socket *s, unsigned int function);
 
 #ifdef CONFIG_PCMCIA_IOCTL
 extern void __init pcmcia_setup_ioctl(void);
 extern void __exit pcmcia_cleanup_ioctl(void);
-extern void handle_event(struct pcmcia_bus_socket *s, event_t event);
-extern int handle_request(struct pcmcia_bus_socket *s, event_t event);
+extern void handle_event(struct pcmcia_socket *s, event_t event);
+extern int handle_request(struct pcmcia_socket *s, event_t event);
 #else
 static inline void __init pcmcia_setup_ioctl(void) { return; }
 static inline void __init pcmcia_cleanup_ioctl(void) { return; }
-static inline void handle_event(struct pcmcia_bus_socket *s, event_t event) { return; }
-static inline int handle_request(struct pcmcia_bus_socket *s, event_t event) { return CS_SUCCESS; }
+static inline void handle_event(struct pcmcia_socket *s, event_t event) { return; }
+static inline int handle_request(struct pcmcia_socket *s, event_t event) { return CS_SUCCESS; }
 #endif
diff --git a/drivers/pcmcia/pcmcia_ioctl.c b/drivers/pcmcia/pcmcia_ioctl.c
index b223f5235de..0bebfdbdc27 100644
--- a/drivers/pcmcia/pcmcia_ioctl.c
+++ b/drivers/pcmcia/pcmcia_ioctl.c
@@ -70,7 +70,7 @@ typedef struct user_info_t {
 	int			event_head, event_tail;
 	event_t			event[MAX_EVENTS];
 	struct user_info_t	*next;
-	struct pcmcia_bus_socket *socket;
+	struct pcmcia_socket	*socket;
 } user_info_t;
 
 
@@ -87,15 +87,6 @@ extern int ds_pc_debug;
 #endif
 
 
-static struct pcmcia_bus_socket * get_socket_info_by_nr(unsigned int nr)
-{
-	struct pcmcia_socket * s = pcmcia_get_socket_by_nr(nr);
-	if (s && s->pcmcia)
-		return s->pcmcia;
-	else
-		return NULL;
-}
-
 /* backwards-compatible accessing of driver --- by name! */
 
 static struct pcmcia_driver * get_pcmcia_driver (dev_info_t *dev_info)
@@ -172,7 +163,7 @@ static void queue_event(user_info_t *user, event_t event)
     user->event[user->event_head] = event;
 }
 
-void handle_event(struct pcmcia_bus_socket *s, event_t event)
+void handle_event(struct pcmcia_socket *s, event_t event)
 {
     user_info_t *user;
     for (user = s->user; user; user = user->next)
@@ -204,18 +195,18 @@ void handle_event(struct pcmcia_bus_socket *s, event_t event)
 
 ======================================================================*/
 
-static int bind_request(struct pcmcia_bus_socket *s, bind_info_t *bind_info)
+static int bind_request(struct pcmcia_socket *s, bind_info_t *bind_info)
 {
 	struct pcmcia_driver *p_drv;
 	struct pcmcia_device *p_dev;
 	int ret = 0;
 	unsigned long flags;
 
-	s = pcmcia_get_bus_socket(s);
+	s = pcmcia_get_socket(s);
 	if (!s)
 		return -EINVAL;
 
-	ds_dbg(2, "bind_request(%d, '%s')\n", s->parent->sock,
+	ds_dbg(2, "bind_request(%d, '%s')\n", s->sock,
 	       (char *)bind_info->dev_info);
 
 	p_drv = get_pcmcia_driver(&bind_info->dev_info);
@@ -278,9 +269,9 @@ rescan:
 	/*
 	 * Prevent this racing with a card insertion.
 	 */
-	down(&s->parent->skt_sem);
+	down(&s->skt_sem);
 	bus_rescan_devices(&pcmcia_bus_type);
-	up(&s->parent->skt_sem);
+	up(&s->skt_sem);
 
 	/* check whether the driver indeed matched. I don't care if this
 	 * is racy or not, because it can only happen on cardmgr access
@@ -294,7 +285,7 @@ rescan:
  err_put_driver:
 	put_driver(&p_drv->drv);
  err_put:
-	pcmcia_put_bus_socket(s);
+	pcmcia_put_socket(s);
 
 	return (ret);
 } /* bind_request */
@@ -302,7 +293,7 @@ rescan:
 
 extern struct pci_bus *pcmcia_lookup_bus(struct pcmcia_socket *s);
 
-static int get_device_info(struct pcmcia_bus_socket *s, bind_info_t *bind_info, int first)
+static int get_device_info(struct pcmcia_socket *s, bind_info_t *bind_info, int first)
 {
 	dev_node_t *node;
 	struct pcmcia_device *p_dev;
@@ -317,7 +308,7 @@ static int get_device_info(struct pcmcia_bus_socket *s, bind_info_t *bind_info,
 	{
 		struct pci_bus *bus;
 
-		bus = pcmcia_lookup_bus(s->parent);
+		bus = pcmcia_lookup_bus(s);
 		if (bus) {
 			struct list_head *list;
 			struct pci_dev *dev = NULL;
@@ -391,21 +382,21 @@ static int get_device_info(struct pcmcia_bus_socket *s, bind_info_t *bind_info,
 static int ds_open(struct inode *inode, struct file *file)
 {
     socket_t i = iminor(inode);
-    struct pcmcia_bus_socket *s;
+    struct pcmcia_socket *s;
     user_info_t *user;
 
     ds_dbg(0, "ds_open(socket %d)\n", i);
 
-    s = get_socket_info_by_nr(i);
+    s = pcmcia_get_socket_by_nr(i);
     if (!s)
 	    return -ENODEV;
-    s = pcmcia_get_bus_socket(s);
+    s = pcmcia_get_socket(s);
     if (!s)
 	    return -ENODEV;
 
     if ((file->f_flags & O_ACCMODE) != O_RDONLY) {
 	    if (s->pcmcia_state.busy) {
-		    pcmcia_put_bus_socket(s);
+		    pcmcia_put_socket(s);
 		    return -EBUSY;
 	    }
 	else
@@ -414,7 +405,7 @@ static int ds_open(struct inode *inode, struct file *file)
 
     user = kmalloc(sizeof(user_info_t), GFP_KERNEL);
     if (!user) {
-	    pcmcia_put_bus_socket(s);
+	    pcmcia_put_socket(s);
 	    return -ENOMEM;
     }
     user->event_tail = user->event_head = 0;
@@ -433,7 +424,7 @@ static int ds_open(struct inode *inode, struct file *file)
 
 static int ds_release(struct inode *inode, struct file *file)
 {
-    struct pcmcia_bus_socket *s;
+    struct pcmcia_socket *s;
     user_info_t *user, **link;
 
     ds_dbg(0, "ds_release(socket %d)\n", iminor(inode));
@@ -456,7 +447,7 @@ static int ds_release(struct inode *inode, struct file *file)
     *link = user->next;
     user->user_magic = 0;
     kfree(user);
-    pcmcia_put_bus_socket(s);
+    pcmcia_put_socket(s);
 out:
     return 0;
 } /* ds_release */
@@ -466,7 +457,7 @@ out:
 static ssize_t ds_read(struct file *file, char __user *buf,
 		       size_t count, loff_t *ppos)
 {
-    struct pcmcia_bus_socket *s;
+    struct pcmcia_socket *s;
     user_info_t *user;
     int ret;
 
@@ -510,7 +501,7 @@ static ssize_t ds_write(struct file *file, const char __user *buf,
 /* No kernel lock - fine */
 static u_int ds_poll(struct file *file, poll_table *wait)
 {
-    struct pcmcia_bus_socket *s;
+    struct pcmcia_socket *s;
     user_info_t *user;
 
     ds_dbg(2, "ds_poll(socket %d)\n", iminor(file->f_dentry->d_inode));
@@ -536,7 +527,7 @@ extern int pcmcia_adjust_resource_info(adjust_t *adj);
 static int ds_ioctl(struct inode * inode, struct file * file,
 		    u_int cmd, u_long arg)
 {
-    struct pcmcia_bus_socket *s;
+    struct pcmcia_socket *s;
     void __user *uarg = (char __user *)arg;
     u_int size;
     int ret, err;
@@ -589,57 +580,57 @@ static int ds_ioctl(struct inode * inode, struct file * file,
 	break;
     case DS_GET_CONFIGURATION_INFO:
 	if (buf->config.Function &&
-	   (buf->config.Function >= s->parent->functions))
+	   (buf->config.Function >= s->functions))
 	    ret = CS_BAD_ARGS;
 	else
-	    ret = pccard_get_configuration_info(s->parent,
+	    ret = pccard_get_configuration_info(s,
 			buf->config.Function, &buf->config);
 	break;
     case DS_GET_FIRST_TUPLE:
-	down(&s->parent->skt_sem);
-	pcmcia_validate_mem(s->parent);
-	up(&s->parent->skt_sem);
-	ret = pccard_get_first_tuple(s->parent, BIND_FN_ALL, &buf->tuple);
+	down(&s->skt_sem);
+	pcmcia_validate_mem(s);
+	up(&s->skt_sem);
+	ret = pccard_get_first_tuple(s, BIND_FN_ALL, &buf->tuple);
 	break;
     case DS_GET_NEXT_TUPLE:
-	ret = pccard_get_next_tuple(s->parent, BIND_FN_ALL, &buf->tuple);
+	ret = pccard_get_next_tuple(s, BIND_FN_ALL, &buf->tuple);
 	break;
     case DS_GET_TUPLE_DATA:
 	buf->tuple.TupleData = buf->tuple_parse.data;
 	buf->tuple.TupleDataMax = sizeof(buf->tuple_parse.data);
-	ret = pccard_get_tuple_data(s->parent, &buf->tuple);
+	ret = pccard_get_tuple_data(s, &buf->tuple);
 	break;
     case DS_PARSE_TUPLE:
 	buf->tuple.TupleData = buf->tuple_parse.data;
 	ret = pccard_parse_tuple(&buf->tuple, &buf->tuple_parse.parse);
 	break;
     case DS_RESET_CARD:
-	ret = pccard_reset_card(s->parent);
+	ret = pccard_reset_card(s);
 	break;
     case DS_GET_STATUS:
 	if (buf->status.Function &&
-	   (buf->status.Function >= s->parent->functions))
+	   (buf->status.Function >= s->functions))
 	    ret = CS_BAD_ARGS;
 	else
-	ret = pccard_get_status(s->parent, buf->status.Function, &buf->status);
+	ret = pccard_get_status(s, buf->status.Function, &buf->status);
 	break;
     case DS_VALIDATE_CIS:
-	down(&s->parent->skt_sem);
-	pcmcia_validate_mem(s->parent);
-	up(&s->parent->skt_sem);
-	ret = pccard_validate_cis(s->parent, BIND_FN_ALL, &buf->cisinfo);
+	down(&s->skt_sem);
+	pcmcia_validate_mem(s);
+	up(&s->skt_sem);
+	ret = pccard_validate_cis(s, BIND_FN_ALL, &buf->cisinfo);
 	break;
     case DS_SUSPEND_CARD:
-	ret = pcmcia_suspend_card(s->parent);
+	ret = pcmcia_suspend_card(s);
 	break;
     case DS_RESUME_CARD:
-	ret = pcmcia_resume_card(s->parent);
+	ret = pcmcia_resume_card(s);
 	break;
     case DS_EJECT_CARD:
-	err = pcmcia_eject_card(s->parent);
+	err = pcmcia_eject_card(s);
 	break;
     case DS_INSERT_CARD:
-	err = pcmcia_insert_card(s->parent);
+	err = pcmcia_insert_card(s);
 	break;
     case DS_ACCESS_CONFIGURATION_REGISTER:
 	if ((buf->conf_reg.Action == CS_WRITE) && !capable(CAP_SYS_ADMIN)) {
@@ -647,10 +638,10 @@ static int ds_ioctl(struct inode * inode, struct file * file,
 	    goto free_out;
 	}
 	if (buf->conf_reg.Function &&
-	   (buf->conf_reg.Function >= s->parent->functions))
+	   (buf->conf_reg.Function >= s->functions))
 	    ret = CS_BAD_ARGS;
 	else
-	    ret = pccard_access_configuration_register(s->parent,
+	    ret = pccard_access_configuration_register(s,
 			buf->conf_reg.Function, &buf->conf_reg);
 	break;
     case DS_GET_FIRST_REGION:
@@ -671,11 +662,11 @@ static int ds_ioctl(struct inode * inode, struct file * file,
 	goto free_out;
 	break;
     case DS_GET_FIRST_WINDOW:
-	ret = pcmcia_get_window(s->parent, &buf->win_info.handle, 0,
+	ret = pcmcia_get_window(s, &buf->win_info.handle, 0,
 			&buf->win_info.window);
 	break;
     case DS_GET_NEXT_WINDOW:
-	ret = pcmcia_get_window(s->parent, &buf->win_info.handle,
+	ret = pcmcia_get_window(s, &buf->win_info.handle,
 			buf->win_info.handle->index + 1, &buf->win_info.window);
 	break;
     case DS_GET_MEM_PAGE:
@@ -683,7 +674,7 @@ static int ds_ioctl(struct inode * inode, struct file * file,
 			   &buf->win_info.map);
 	break;
     case DS_REPLACE_CIS:
-	ret = pcmcia_replace_cis(s->parent, &buf->cisdump);
+	ret = pcmcia_replace_cis(s, &buf->cisdump);
 	break;
     case DS_BIND_REQUEST:
 	if (!capable(CAP_SYS_ADMIN)) {
diff --git a/include/pcmcia/ss.h b/include/pcmcia/ss.h
index f8797767051..0f7aacc33fe 100644
--- a/include/pcmcia/ss.h
+++ b/include/pcmcia/ss.h
@@ -15,10 +15,12 @@
 #ifndef _LINUX_SS_H
 #define _LINUX_SS_H
 
+#include <linux/config.h>
+#include <linux/device.h>
+
 #include <pcmcia/cs_types.h>
 #include <pcmcia/cs.h>
 #include <pcmcia/bulkmem.h>
-#include <linux/device.h>
 
 /* Definitions for card status flags for GetStatus */
 #define SS_WRPROT	0x0001
@@ -171,7 +173,7 @@ typedef struct window_t {
 
 struct config_t;
 struct pcmcia_callback;
-
+struct user_info_t;
 
 struct pcmcia_socket {
 	struct module			*owner;
@@ -242,9 +244,32 @@ struct pcmcia_socket {
 	unsigned int			thread_events;
 
 	/* pcmcia (16-bit) */
-	struct pcmcia_bus_socket	*pcmcia;
 	struct pcmcia_callback		*callback;
 
+#if defined(CONFIG_PCMCIA) || defined(CONFIG_PCMCIA_MODULE)
+	struct list_head		devices_list;	/*  PCMCIA devices */
+	u8				device_count;	/* the number of devices, used
+							 * only internally and subject
+							 * to incorrectness and change */
+
+	struct {
+		u8			present:1,	/* PCMCIA card is present in socket */
+					busy:1,		/* "master" ioctl is used */
+					dead:1,		/* pcmcia module is being unloaded */
+					device_add_pending:1, /* a pseudo-multifunction-device
+							       * add event is pending */
+					reserved:4;
+	} 				pcmcia_state;
+
+	struct work_struct		device_add;	/* for adding further pseudo-multifunction
+							 * devices */
+
+#ifdef CONFIG_PCMCIA_IOCTL
+	struct user_info_t		*user;
+	wait_queue_head_t		queue;
+#endif
+#endif
+
 	/* cardbus (32-bit) */
 #ifdef CONFIG_CARDBUS
 	struct resource *		cb_cis_res;
-- 
cgit v1.2.3-70-g09d2


From 3448139b41b9e3b8799eed7d427cd50789dadc3e Mon Sep 17 00:00:00 2001
From: Dominik Brodowski <linux@dominikbrodowski.net>
Date: Mon, 27 Jun 2005 16:28:54 -0700
Subject: [PATCH] pcmcia: ds.c cleanup

Clean up ds.c

Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/pcmcia/pcmcia_compat.c   | 34 ----------------------
 drivers/pcmcia/pcmcia_ioctl.c    | 23 +++++++++++++++
 drivers/pcmcia/pcmcia_resource.c | 62 +++++++++++++++++++++++-----------------
 include/pcmcia/cs.h              |  1 -
 4 files changed, 58 insertions(+), 62 deletions(-)

(limited to 'include')

diff --git a/drivers/pcmcia/pcmcia_compat.c b/drivers/pcmcia/pcmcia_compat.c
index 68b80084f83..1cc83317e7e 100644
--- a/drivers/pcmcia/pcmcia_compat.c
+++ b/drivers/pcmcia/pcmcia_compat.c
@@ -74,19 +74,6 @@ int pcmcia_validate_cis(client_handle_t handle, cisinfo_t *info)
 }
 EXPORT_SYMBOL(pcmcia_validate_cis);
 
-int pcmcia_get_configuration_info(client_handle_t handle,
-				  config_info_t *config)
-{
-	struct pcmcia_socket *s;
-
-	if ((CHECK_HANDLE(handle)) || !config)
-		return CS_BAD_HANDLE;
-	s = SOCKET(handle);
-	if (!s)
-		return CS_BAD_HANDLE;
-	return pccard_get_configuration_info(s, handle->Function, config);
-}
-EXPORT_SYMBOL(pcmcia_get_configuration_info);
 
 int pcmcia_reset_card(client_handle_t handle, client_req_t *req)
 {
@@ -102,24 +89,3 @@ int pcmcia_reset_card(client_handle_t handle, client_req_t *req)
 }
 EXPORT_SYMBOL(pcmcia_reset_card);
 
-int pcmcia_get_status(client_handle_t handle, cs_status_t *status)
-{
-	struct pcmcia_socket *s;
-	if (CHECK_HANDLE(handle))
-		return CS_BAD_HANDLE;
-	s = SOCKET(handle);
-	return pccard_get_status(s, handle->Function, status);
-}
-EXPORT_SYMBOL(pcmcia_get_status);
-
-int pcmcia_access_configuration_register(client_handle_t handle,
-					 conf_reg_t *reg)
-{
-	struct pcmcia_socket *s;
-	if (CHECK_HANDLE(handle))
-		return CS_BAD_HANDLE;
-	s = SOCKET(handle);
-	return pccard_access_configuration_register(s, handle->Function, reg);
-}
-EXPORT_SYMBOL(pcmcia_access_configuration_register);
-
diff --git a/drivers/pcmcia/pcmcia_ioctl.c b/drivers/pcmcia/pcmcia_ioctl.c
index fa2a79d29e1..3084d8a3ba4 100644
--- a/drivers/pcmcia/pcmcia_ioctl.c
+++ b/drivers/pcmcia/pcmcia_ioctl.c
@@ -71,6 +71,29 @@ extern int ds_pc_debug;
 #define ds_dbg(lvl, fmt, arg...) do { } while (0)
 #endif
 
+static const char *release = "Linux Kernel Card Services";
+
+/** pcmcia_get_card_services_info
+ *
+ * Return information about this version of Card Services
+ */
+static int pcmcia_get_card_services_info(servinfo_t *info)
+{
+	unsigned int socket_count = 0;
+	struct list_head *tmp;
+	info->Signature[0] = 'C';
+	info->Signature[1] = 'S';
+	down_read(&pcmcia_socket_list_rwsem);
+	list_for_each(tmp, &pcmcia_socket_list)
+		socket_count++;
+	up_read(&pcmcia_socket_list_rwsem);
+	info->Count = socket_count;
+	info->Revision = CS_RELEASE_CODE;
+	info->CSLevel = 0x0210;
+	info->VendorString = (char *)release;
+	return CS_SUCCESS;
+} /* get_card_services_info */
+
 
 /* backwards-compatible accessing of driver --- by name! */
 
diff --git a/drivers/pcmcia/pcmcia_resource.c b/drivers/pcmcia/pcmcia_resource.c
index cf66b997894..9ed3d4a9b99 100644
--- a/drivers/pcmcia/pcmcia_resource.c
+++ b/drivers/pcmcia/pcmcia_resource.c
@@ -36,8 +36,6 @@
 #include "ds_internal.h"
 
 
-static const char *release = "Linux Kernel Card Services";
-
 /* Access speed for IO windows */
 static int io_speed = 0;
 module_param(io_speed, int, 0444);
@@ -202,7 +200,18 @@ int pccard_access_configuration_register(struct pcmcia_socket *s,
 	}
 	return CS_SUCCESS;
 } /* pccard_access_configuration_register */
-EXPORT_SYMBOL(pccard_access_configuration_register);
+
+int pcmcia_access_configuration_register(client_handle_t handle,
+					 conf_reg_t *reg)
+{
+	struct pcmcia_socket *s;
+	if (CHECK_HANDLE(handle))
+		return CS_BAD_HANDLE;
+	s = SOCKET(handle);
+	return pccard_access_configuration_register(s, handle->Function, reg);
+}
+EXPORT_SYMBOL(pcmcia_access_configuration_register);
+
 
 
 int pccard_get_configuration_info(struct pcmcia_socket *s,
@@ -260,31 +269,20 @@ int pccard_get_configuration_info(struct pcmcia_socket *s,
 
 	return CS_SUCCESS;
 } /* pccard_get_configuration_info */
-EXPORT_SYMBOL(pccard_get_configuration_info);
 
-
-/** pcmcia_get_card_services_info
- *
- * Return information about this version of Card Services
- */
-
-int pcmcia_get_card_services_info(servinfo_t *info)
+int pcmcia_get_configuration_info(client_handle_t handle,
+				  config_info_t *config)
 {
-	unsigned int socket_count = 0;
-	struct list_head *tmp;
-	info->Signature[0] = 'C';
-	info->Signature[1] = 'S';
-	down_read(&pcmcia_socket_list_rwsem);
-	list_for_each(tmp, &pcmcia_socket_list)
-		socket_count++;
-	up_read(&pcmcia_socket_list_rwsem);
-	info->Count = socket_count;
-	info->Revision = CS_RELEASE_CODE;
-	info->CSLevel = 0x0210;
-	info->VendorString = (char *)release;
-	return CS_SUCCESS;
-} /* get_card_services_info */
-EXPORT_SYMBOL(pcmcia_get_card_services_info);
+	struct pcmcia_socket *s;
+
+	if ((CHECK_HANDLE(handle)) || !config)
+		return CS_BAD_HANDLE;
+	s = SOCKET(handle);
+	if (!s)
+		return CS_BAD_HANDLE;
+	return pccard_get_configuration_info(s, handle->Function, config);
+}
+EXPORT_SYMBOL(pcmcia_get_configuration_info);
 
 
 /** pcmcia_get_window
@@ -379,7 +377,17 @@ int pccard_get_status(struct pcmcia_socket *s, unsigned int function,
 		(val & SS_READY) ? CS_EVENT_READY_CHANGE : 0;
 	return CS_SUCCESS;
 } /* pccard_get_status */
-EXPORT_SYMBOL(pccard_get_status);
+
+int pcmcia_get_status(client_handle_t handle, cs_status_t *status)
+{
+	struct pcmcia_socket *s;
+	if (CHECK_HANDLE(handle))
+		return CS_BAD_HANDLE;
+	s = SOCKET(handle);
+	return pccard_get_status(s, handle->Function, status);
+}
+EXPORT_SYMBOL(pcmcia_get_status);
+
 
 
 /** pcmcia_get_mem_page
diff --git a/include/pcmcia/cs.h b/include/pcmcia/cs.h
index 60a3decea82..b42ddc0c114 100644
--- a/include/pcmcia/cs.h
+++ b/include/pcmcia/cs.h
@@ -396,7 +396,6 @@ struct pcmcia_socket;
 int pcmcia_access_configuration_register(client_handle_t handle, conf_reg_t *reg);
 int pcmcia_deregister_client(client_handle_t handle);
 int pcmcia_get_configuration_info(client_handle_t handle, config_info_t *config);
-int pcmcia_get_card_services_info(servinfo_t *info);
 int pcmcia_get_first_window(window_handle_t *win, win_req_t *req);
 int pcmcia_get_next_window(window_handle_t *win, win_req_t *req);
 int pcmcia_get_status(client_handle_t handle, cs_status_t *status);
-- 
cgit v1.2.3-70-g09d2


From aecab27aeabaa897d69fc082686df314329830de Mon Sep 17 00:00:00 2001
From: Dominik Brodowski <linux@dominikbrodowski.net>
Date: Mon, 27 Jun 2005 16:28:56 -0700
Subject: [PATCH] pcmcia: mod_devicetable.h fix for different sizes in kernel-
 and userspace

The size of pointers may differ between (userspace) modpost and (kernelspace)
modules -- so fix mod_devicetable.h to reflect this possibility.

Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mod_devicetable.h | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h
index 8a8dc82a941..9b6d05172ed 100644
--- a/include/linux/mod_devicetable.h
+++ b/include/linux/mod_devicetable.h
@@ -191,12 +191,22 @@ struct pcmcia_device_id {
 	/* for pseude multi-function devices */
 	__u8  		device_no;
 
-	const char *	prod_id[4];
 	__u32 		prod_id_hash[4];
 
+	/* not matched against in kernelspace*/
+#ifdef __KERNEL__
+	const char *	prod_id[4];
+#else
+	kernel_ulong_t	prod_id[4];
+#endif
+
 	/* not matched against */
 	kernel_ulong_t	driver_info;
+#ifdef __KERNEL__
 	char *		cisfile;
+#else
+	kernel_ulong_t	cisfile;
+#endif
 };
 
 #define PCMCIA_DEV_ID_MATCH_MANF_ID	0x0001
-- 
cgit v1.2.3-70-g09d2


From a5fe736eaf9bae1b45317313de04b564441b94f2 Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jgarzik@pobox.com>
Date: Mon, 27 Jun 2005 22:47:18 -0400
Subject: Update is_multicast_ether_addr() definition; net/ieee80211.h
 cleanups.

---
 include/linux/etherdevice.h |  2 +-
 include/net/ieee80211.h     | 48 ++++++++++-----------------------------------
 2 files changed, 11 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h
index a1478258d00..8a2df4dfbc5 100644
--- a/include/linux/etherdevice.h
+++ b/include/linux/etherdevice.h
@@ -65,7 +65,7 @@ static inline int is_zero_ether_addr(const u8 *addr)
  */
 static inline int is_multicast_ether_addr(const u8 *addr)
 {
-	return addr[0] & 0x01;
+	return ((addr[0] != 0xff) && (0x01 & addr[0]));
 }
 
 /**
diff --git a/include/net/ieee80211.h b/include/net/ieee80211.h
index 7fe57f957a5..151c4f20355 100644
--- a/include/net/ieee80211.h
+++ b/include/net/ieee80211.h
@@ -94,6 +94,8 @@ struct eapol {
 	u16 length;
 } __attribute__ ((packed));
 
+#define IEEE80211_1ADDR_LEN 10
+#define IEEE80211_2ADDR_LEN 16
 #define IEEE80211_3ADDR_LEN 24
 #define IEEE80211_4ADDR_LEN 30
 #define IEEE80211_FCS_LEN    4
@@ -300,23 +302,6 @@ struct ieee80211_snap_hdr {
 #define WLAN_REASON_STA_REQ_ASSOC_WITHOUT_AUTH 9
 
 
-/* Information Element IDs */
-#define WLAN_EID_SSID 0
-#define WLAN_EID_SUPP_RATES 1
-#define WLAN_EID_FH_PARAMS 2
-#define WLAN_EID_DS_PARAMS 3
-#define WLAN_EID_CF_PARAMS 4
-#define WLAN_EID_TIM 5
-#define WLAN_EID_IBSS_PARAMS 6
-#define WLAN_EID_CHALLENGE 16
-#define WLAN_EID_RSN 48
-#define WLAN_EID_GENERIC 221
-
-#define IEEE80211_MGMT_HDR_LEN 24
-#define IEEE80211_DATA_HDR3_LEN 24
-#define IEEE80211_DATA_HDR4_LEN 30
-
-
 #define IEEE80211_STATMASK_SIGNAL (1<<0)
 #define IEEE80211_STATMASK_RSSI (1<<1)
 #define IEEE80211_STATMASK_NOISE (1<<2)
@@ -441,6 +426,8 @@ struct ieee80211_stats {
 
 struct ieee80211_device;
 
+#include "ieee80211_crypt.h"
+
 #define SEC_KEY_1         (1<<0)
 #define SEC_KEY_2         (1<<1)
 #define SEC_KEY_3         (1<<2)
@@ -488,15 +475,6 @@ Total: 28-2340 bytes
 
 */
 
-struct ieee80211_header_data {
-	u16 frame_ctl;
-	u16 duration_id;
-	u8 addr1[6];
-	u8 addr2[6];
-	u8 addr3[6];
-	u16 seq_ctrl;
-};
-
 #define BEACON_PROBE_SSID_ID_POSITION 12
 
 /* Management Frame Information Element Types */
@@ -541,7 +519,7 @@ struct ieee80211_info_element {
 */
 
 struct ieee80211_authentication {
-	struct ieee80211_header_data header;
+	struct ieee80211_hdr_3addr header;
 	u16 algorithm;
 	u16 transaction;
 	u16 status;
@@ -550,7 +528,7 @@ struct ieee80211_authentication {
 
 
 struct ieee80211_probe_response {
-	struct ieee80211_header_data header;
+	struct ieee80211_hdr_3addr header;
 	u32 time_stamp[2];
 	u16 beacon_interval;
 	u16 capability;
@@ -648,12 +626,6 @@ enum ieee80211_state {
 #define MAC_ARG(x) ((u8*)(x))[0],((u8*)(x))[1],((u8*)(x))[2],((u8*)(x))[3],((u8*)(x))[4],((u8*)(x))[5]
 
 
-extern inline int is_broadcast_ether_addr(const u8 *addr)
-{
-	return ((addr[0] == 0xff) && (addr[1] == 0xff) && (addr[2] == 0xff) &&   \
-		(addr[3] == 0xff) && (addr[4] == 0xff) && (addr[5] == 0xff));
-}
-
 #define CFG_IEEE80211_RESERVE_FCS (1<<0)
 #define CFG_IEEE80211_COMPUTE_FCS (1<<1)
 
@@ -787,21 +759,21 @@ extern inline int ieee80211_is_valid_mode(struct ieee80211_device *ieee, int mod
 
 extern inline int ieee80211_get_hdrlen(u16 fc)
 {
-	int hdrlen = 24;
+	int hdrlen = IEEE80211_3ADDR_LEN;
 
 	switch (WLAN_FC_GET_TYPE(fc)) {
 	case IEEE80211_FTYPE_DATA:
 		if ((fc & IEEE80211_FCTL_FROMDS) && (fc & IEEE80211_FCTL_TODS))
-			hdrlen = 30; /* Addr4 */
+			hdrlen = IEEE80211_4ADDR_LEN;
 		break;
 	case IEEE80211_FTYPE_CTL:
 		switch (WLAN_FC_GET_STYPE(fc)) {
 		case IEEE80211_STYPE_CTS:
 		case IEEE80211_STYPE_ACK:
-			hdrlen = 10;
+			hdrlen = IEEE80211_1ADDR_LEN;
 			break;
 		default:
-			hdrlen = 16;
+			hdrlen = IEEE80211_2ADDR_LEN;
 			break;
 		}
 		break;
-- 
cgit v1.2.3-70-g09d2


From c431ada45d65b305a6aab4557067e564b23ce5a5 Mon Sep 17 00:00:00 2001
From: Rajesh Shah <rajesh.shah@intel.com>
Date: Thu, 28 Apr 2005 00:25:45 -0700
Subject: [PATCH] acpi bridge hotadd: ACPI based root bridge hot-add

When you hot-plug a (root) bridge hierarchy, it may have p2p bridges and
devices attached to it that have not been configured by firmware.  In this
case, we need to configure the devices before starting them.  This patch
separates device start from device scan so that we can introduce the
configuration step in the middle.

I kept the existing semantics for pci_scan_bus() since there are a huge number
of callers to that function.

Also, I have no way of testing the changes I made to the parisc files, so this
needs review by those folks.  Sorry for the massive cross-post, this touches
files in many different places.

Signed-off-by: Rajesh Shah <rajesh.shah@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/i386/pci/common.c   |  2 +-
 arch/i386/pci/legacy.c   |  2 ++
 arch/i386/pci/numa.c     |  2 ++
 arch/ia64/pci/pci.c      |  2 +-
 drivers/acpi/pci_bind.c  | 16 +++++++++++++++-
 drivers/acpi/pci_root.c  | 24 +++++++++++++++++++++++-
 drivers/parisc/dino.c    |  1 +
 drivers/parisc/lba_pci.c |  2 ++
 drivers/pci/probe.c      |  2 --
 include/linux/pci.h      |  8 ++++++--
 10 files changed, 53 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/i386/pci/common.c b/arch/i386/pci/common.c
index 2a2e79fbfef..87325263cd4 100644
--- a/arch/i386/pci/common.c
+++ b/arch/i386/pci/common.c
@@ -134,7 +134,7 @@ struct pci_bus * __devinit pcibios_scan_root(int busnum)
 
 	printk("PCI: Probing PCI hardware (bus %02x)\n", busnum);
 
-	return pci_scan_bus(busnum, &pci_root_ops, NULL);
+	return pci_scan_bus_parented(NULL, busnum, &pci_root_ops, NULL);
 }
 
 extern u8 pci_cache_line_size;
diff --git a/arch/i386/pci/legacy.c b/arch/i386/pci/legacy.c
index 1492e375386..149a9588c25 100644
--- a/arch/i386/pci/legacy.c
+++ b/arch/i386/pci/legacy.c
@@ -45,6 +45,8 @@ static int __init pci_legacy_init(void)
 
 	printk("PCI: Probing PCI hardware\n");
 	pci_root_bus = pcibios_scan_root(0);
+	if (pci_root_bus)
+		pci_bus_add_devices(pci_root_bus);
 
 	pcibios_fixup_peer_bridges();
 
diff --git a/arch/i386/pci/numa.c b/arch/i386/pci/numa.c
index 9e369546189..adbe17a38f6 100644
--- a/arch/i386/pci/numa.c
+++ b/arch/i386/pci/numa.c
@@ -115,6 +115,8 @@ static int __init pci_numa_init(void)
 		return 0;
 
 	pci_root_bus = pcibios_scan_root(0);
+	if (pci_root_bus)
+		pci_bus_add_devices(pci_root_bus);
 	if (num_online_nodes() > 1)
 		for_each_online_node(quad) {
 			if (quad == 0)
diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c
index e3fc4edea11..c0661d3382e 100644
--- a/arch/ia64/pci/pci.c
+++ b/arch/ia64/pci/pci.c
@@ -312,7 +312,7 @@ pci_acpi_scan_root(struct acpi_device *device, int domain, int bus)
 	acpi_walk_resources(device->handle, METHOD_NAME__CRS, add_window,
 			&info);
 
-	pbus = pci_scan_bus(bus, &pci_root_ops, controller);
+	pbus = pci_scan_bus_parented(NULL, bus, &pci_root_ops, controller);
 	if (pbus)
 		pcibios_setup_root_windows(pbus, controller);
 
diff --git a/drivers/acpi/pci_bind.c b/drivers/acpi/pci_bind.c
index 5d19b39e9e2..7753df1f9fb 100644
--- a/drivers/acpi/pci_bind.c
+++ b/drivers/acpi/pci_bind.c
@@ -129,6 +129,8 @@ acpi_pci_bind (
 	char			*pathname = NULL;
 	struct acpi_buffer	buffer = {0, NULL};
 	acpi_handle		handle = NULL;
+	struct pci_dev		*dev;
+	struct pci_bus 		*bus;
 
 	ACPI_FUNCTION_TRACE("acpi_pci_bind");
 
@@ -193,8 +195,20 @@ acpi_pci_bind (
 	 * Locate matching device in PCI namespace.  If it doesn't exist
 	 * this typically means that the device isn't currently inserted
 	 * (e.g. docking station, port replicator, etc.).
+	 * We cannot simply search the global pci device list, since
+	 * PCI devices are added to the global pci list when the root
+	 * bridge start ops are run, which may not have happened yet.
 	 */
-	data->dev = pci_find_slot(data->id.bus, PCI_DEVFN(data->id.device, data->id.function));
+	bus = pci_find_bus(data->id.segment, data->id.bus);
+	if (bus) {
+		list_for_each_entry(dev, &bus->devices, bus_list) {
+			if (dev->devfn == PCI_DEVFN(data->id.device,
+						data->id.function)) {
+				data->dev = dev;
+				break;
+			}
+		}
+	}
 	if (!data->dev) {
 		ACPI_DEBUG_PRINT((ACPI_DB_INFO, 
 			"Device %02x:%02x:%02x.%02x not present in PCI namespace\n",
diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index 7e6b8e3b2ed..5d2f77fcd50 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -46,6 +46,7 @@ ACPI_MODULE_NAME		("pci_root")
 
 static int acpi_pci_root_add (struct acpi_device *device);
 static int acpi_pci_root_remove (struct acpi_device *device, int type);
+static int acpi_pci_root_start (struct acpi_device *device);
 
 static struct acpi_driver acpi_pci_root_driver = {
 	.name =		ACPI_PCI_ROOT_DRIVER_NAME,
@@ -54,6 +55,7 @@ static struct acpi_driver acpi_pci_root_driver = {
 	.ops =		{
 				.add =    acpi_pci_root_add,
 				.remove = acpi_pci_root_remove,
+				.start =  acpi_pci_root_start,
 			},
 };
 
@@ -169,6 +171,7 @@ acpi_pci_root_add (
 	if (!root)
 		return_VALUE(-ENOMEM);
 	memset(root, 0, sizeof(struct acpi_pci_root));
+	INIT_LIST_HEAD(&root->node);
 
 	root->handle = device->handle;
 	strcpy(acpi_device_name(device), ACPI_PCI_ROOT_DEVICE_NAME);
@@ -298,12 +301,31 @@ acpi_pci_root_add (
 			root->id.bus);
 
 end:
-	if (result)
+	if (result) {
+		if (!list_empty(&root->node))
+			list_del(&root->node);
 		kfree(root);
+	}
 
 	return_VALUE(result);
 }
 
+static int
+acpi_pci_root_start (
+	struct acpi_device	*device)
+{
+	struct acpi_pci_root	*root;
+
+	ACPI_FUNCTION_TRACE("acpi_pci_root_start");
+
+	list_for_each_entry(root, &acpi_pci_roots, node) {
+		if (root->handle == device->handle) {
+			pci_bus_add_devices(root->bus);
+			return_VALUE(0);
+		}
+	}
+	return_VALUE(-ENODEV);
+}
 
 static int
 acpi_pci_root_remove (
diff --git a/drivers/parisc/dino.c b/drivers/parisc/dino.c
index b0d2a73d1d4..2f2dbef2c3b 100644
--- a/drivers/parisc/dino.c
+++ b/drivers/parisc/dino.c
@@ -993,6 +993,7 @@ dino_driver_callback(struct parisc_device *dev)
 	bus = pci_scan_bus_parented(&dev->dev, dino_current_bus,
 				    &dino_cfg_ops, NULL);
 	if(bus) {
+		pci_bus_add_devices(bus);
 		/* This code *depends* on scanning being single threaded
 		 * if it isn't, this global bus number count will fail
 		 */
diff --git a/drivers/parisc/lba_pci.c b/drivers/parisc/lba_pci.c
index dc838804c0d..7fdd80b7eb4 100644
--- a/drivers/parisc/lba_pci.c
+++ b/drivers/parisc/lba_pci.c
@@ -1570,6 +1570,8 @@ lba_driver_probe(struct parisc_device *dev)
 	lba_bus = lba_dev->hba.hba_bus =
 		pci_scan_bus_parented(&dev->dev, lba_dev->hba.bus_num.start,
 				cfg_ops, NULL);
+	if (lba_bus)
+		pci_bus_add_devices(lba_bus);
 
 	/* This is in lieu of calling pci_assign_unassigned_resources() */
 	if (is_pdc_pat()) {
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index fd48b201eb5..3dc00f0ca8a 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -911,8 +911,6 @@ struct pci_bus * __devinit pci_scan_bus_parented(struct device *parent, int bus,
 
 	b->subordinate = pci_scan_child_bus(b);
 
-	pci_bus_add_devices(b);
-
 	return b;
 
 sys_create_link_err:
diff --git a/include/linux/pci.h b/include/linux/pci.h
index b5238bd1883..0e9844929fe 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -734,16 +734,20 @@ void pcibios_update_irq(struct pci_dev *, int irq);
 /* Generic PCI functions used internally */
 
 extern struct pci_bus *pci_find_bus(int domain, int busnr);
+void pci_bus_add_devices(struct pci_bus *bus);
 struct pci_bus *pci_scan_bus_parented(struct device *parent, int bus, struct pci_ops *ops, void *sysdata);
 static inline struct pci_bus *pci_scan_bus(int bus, struct pci_ops *ops, void *sysdata)
 {
-	return pci_scan_bus_parented(NULL, bus, ops, sysdata);
+	struct pci_bus *root_bus;
+	root_bus = pci_scan_bus_parented(NULL, bus, ops, sysdata);
+	if (root_bus)
+		pci_bus_add_devices(root_bus);
+	return root_bus;
 }
 int pci_scan_slot(struct pci_bus *bus, int devfn);
 struct pci_dev * pci_scan_single_device(struct pci_bus *bus, int devfn);
 unsigned int pci_scan_child_bus(struct pci_bus *bus);
 void pci_bus_add_device(struct pci_dev *dev);
-void pci_bus_add_devices(struct pci_bus *bus);
 void pci_name_device(struct pci_dev *dev);
 char *pci_class_name(u32 class);
 void pci_read_bridge_bases(struct pci_bus *child);
-- 
cgit v1.2.3-70-g09d2


From 3fb02738b0fd36f47710a2bf207129efd2f5daa2 Mon Sep 17 00:00:00 2001
From: Rajesh Shah <rajesh.shah@intel.com>
Date: Thu, 28 Apr 2005 00:25:52 -0700
Subject: [PATCH] acpi bridge hotadd: Allow ACPI .add and .start operations to
 be done independently

Create new interfaces to recursively add an acpi namespace object to the acpi
device list, and recursively start the namespace object.  This is needed for
ACPI based hotplug of a root bridge hierarchy where the add operation must be
performed first and the start operation must be performed separately after the
hot-plugged devices have been properly configured.

Signed-off-by: Rajesh Shah <rajesh.shah@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/acpi/container.c      |   2 +-
 drivers/acpi/processor_core.c |   2 +-
 drivers/acpi/scan.c           | 126 +++++++++++++++++++++++++++++++++---------
 include/acpi/acpi_bus.h       |  17 +++++-
 4 files changed, 119 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/drivers/acpi/container.c b/drivers/acpi/container.c
index 5a0adbf8bc0..97013ddfa20 100644
--- a/drivers/acpi/container.c
+++ b/drivers/acpi/container.c
@@ -153,7 +153,7 @@ container_device_add(struct acpi_device **device, acpi_handle handle)
 		return_VALUE(-ENODEV);
 	}
 
-	result = acpi_bus_scan(*device);
+	result = acpi_bus_start(*device);
 
 	return_VALUE(result);
 }
diff --git a/drivers/acpi/processor_core.c b/drivers/acpi/processor_core.c
index f4778747e88..76156ac91bd 100644
--- a/drivers/acpi/processor_core.c
+++ b/drivers/acpi/processor_core.c
@@ -723,7 +723,7 @@ int acpi_processor_device_add(
 		return_VALUE(-ENODEV);
 	}
 
-	acpi_bus_scan(*device);
+	acpi_bus_start(*device);
 
 	pr = acpi_driver_data(*device);
 	if (!pr)
diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
index e8588559328..337d49b5564 100644
--- a/drivers/acpi/scan.c
+++ b/drivers/acpi/scan.c
@@ -553,20 +553,29 @@ acpi_bus_driver_init (
 	 * upon possible configuration and currently allocated resources.
 	 */
 
+	ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Driver successfully bound to device\n"));
+	return_VALUE(0);
+}
+
+int
+acpi_start_single_object (
+		struct acpi_device *device)
+{
+	int result = 0;
+	struct acpi_driver *driver;
+
+	ACPI_FUNCTION_TRACE("acpi_start_single_object");
+
+	if (!(driver = device->driver))
+		return_VALUE(0);
+
 	if (driver->ops.start) {
 		result = driver->ops.start(device);
 		if (result && driver->ops.remove)
 			driver->ops.remove(device, ACPI_BUS_REMOVAL_NORMAL);
-		return_VALUE(result);
 	}
 
-	ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Driver successfully bound to device\n"));
-
-	if (driver->ops.scan) {
-		driver->ops.scan(device);
-	}
-
-	return_VALUE(0);
+	return_VALUE(result);
 }
 
 static int acpi_driver_attach(struct acpi_driver * drv)
@@ -586,6 +595,7 @@ static int acpi_driver_attach(struct acpi_driver * drv)
 
 		if (!acpi_bus_match(dev, drv)) {
 			if (!acpi_bus_driver_init(dev, drv)) {
+				acpi_start_single_object(dev);
 				atomic_inc(&drv->references);
 				count++;
 				ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Found driver [%s] for device [%s]\n",
@@ -1009,8 +1019,8 @@ acpi_bus_remove (
 }
 
 
-int
-acpi_bus_add (
+static int
+acpi_add_single_object (
 	struct acpi_device	**child,
 	struct acpi_device	*parent,
 	acpi_handle		handle,
@@ -1019,7 +1029,7 @@ acpi_bus_add (
 	int			result = 0;
 	struct acpi_device	*device = NULL;
 
-	ACPI_FUNCTION_TRACE("acpi_bus_add");
+	ACPI_FUNCTION_TRACE("acpi_add_single_object");
 
 	if (!child)
 		return_VALUE(-EINVAL);
@@ -1140,7 +1150,7 @@ acpi_bus_add (
 	 *
 	 * TBD: Assumes LDM provides driver hot-plug capability.
 	 */
-	acpi_bus_find_driver(device);
+	result = acpi_bus_find_driver(device);
 
 end:
 	if (!result)
@@ -1153,10 +1163,10 @@ end:
 
 	return_VALUE(result);
 }
-EXPORT_SYMBOL(acpi_bus_add);
 
 
-int acpi_bus_scan (struct acpi_device	*start)
+static int acpi_bus_scan (struct acpi_device	*start,
+		struct acpi_bus_ops *ops)
 {
 	acpi_status		status = AE_OK;
 	struct acpi_device	*parent = NULL;
@@ -1229,9 +1239,20 @@ int acpi_bus_scan (struct acpi_device	*start)
 			continue;
 		}
 
-		status = acpi_bus_add(&child, parent, chandle, type);
-		if (ACPI_FAILURE(status))
-			continue;
+		if (ops->acpi_op_add)
+			status = acpi_add_single_object(&child, parent,
+					chandle, type);
+		 else
+			status = acpi_bus_get_device(chandle, &child);
+
+		 if (ACPI_FAILURE(status))
+			 continue;
+
+		if (ops->acpi_op_start) {
+			status = acpi_start_single_object(child);
+			if (ACPI_FAILURE(status))
+				continue;
+		}
 
 		/*
 		 * If the device is present, enabled, and functioning then
@@ -1257,8 +1278,50 @@ int acpi_bus_scan (struct acpi_device	*start)
 
 	return_VALUE(0);
 }
-EXPORT_SYMBOL(acpi_bus_scan);
 
+int
+acpi_bus_add (
+	struct acpi_device	**child,
+	struct acpi_device	*parent,
+	acpi_handle		handle,
+	int			type)
+{
+	int result;
+	struct acpi_bus_ops ops;
+
+	ACPI_FUNCTION_TRACE("acpi_bus_add");
+
+	result = acpi_add_single_object(child, parent, handle, type);
+	if (!result) {
+		memset(&ops, 0, sizeof(ops));
+		ops.acpi_op_add = 1;
+		result = acpi_bus_scan(*child, &ops);
+	}
+	return_VALUE(result);
+}
+EXPORT_SYMBOL(acpi_bus_add);
+
+int
+acpi_bus_start (
+	struct acpi_device *device)
+{
+	int result;
+	struct acpi_bus_ops ops;
+
+	ACPI_FUNCTION_TRACE("acpi_bus_start");
+
+	if (!device)
+		return_VALUE(-EINVAL);
+
+	result = acpi_start_single_object(device);
+	if (!result) {
+		memset(&ops, 0, sizeof(ops));
+		ops.acpi_op_start = 1;
+		result = acpi_bus_scan(device, &ops);
+	}
+	return_VALUE(result);
+}
+EXPORT_SYMBOL(acpi_bus_start);
 
 static int
 acpi_bus_trim(struct acpi_device	*start,
@@ -1331,13 +1394,19 @@ acpi_bus_scan_fixed (
 	/*
 	 * Enumerate all fixed-feature devices.
 	 */
-	if (acpi_fadt.pwr_button == 0)
-		result = acpi_bus_add(&device, acpi_root, 
+	if (acpi_fadt.pwr_button == 0) {
+		result = acpi_add_single_object(&device, acpi_root,
 			NULL, ACPI_BUS_TYPE_POWER_BUTTON);
+		if (!result)
+			result = acpi_start_single_object(device);
+	}
 
-	if (acpi_fadt.sleep_button == 0)
-		result = acpi_bus_add(&device, acpi_root, 
+	if (acpi_fadt.sleep_button == 0) {
+		result = acpi_add_single_object(&device, acpi_root,
 			NULL, ACPI_BUS_TYPE_SLEEP_BUTTON);
+		if (!result)
+			result = acpi_start_single_object(device);
+	}
 
 	return_VALUE(result);
 }
@@ -1346,6 +1415,7 @@ acpi_bus_scan_fixed (
 static int __init acpi_scan_init(void)
 {
 	int result;
+	struct acpi_bus_ops ops;
 
 	ACPI_FUNCTION_TRACE("acpi_scan_init");
 
@@ -1357,17 +1427,23 @@ static int __init acpi_scan_init(void)
 	/*
 	 * Create the root device in the bus's device tree
 	 */
-	result = acpi_bus_add(&acpi_root, NULL, ACPI_ROOT_OBJECT, 
+	result = acpi_add_single_object(&acpi_root, NULL, ACPI_ROOT_OBJECT,
 		ACPI_BUS_TYPE_SYSTEM);
 	if (result)
 		goto Done;
 
+	result = acpi_start_single_object(acpi_root);
+
 	/*
 	 * Enumerate devices in the ACPI namespace.
 	 */
 	result = acpi_bus_scan_fixed(acpi_root);
-	if (!result) 
-		result = acpi_bus_scan(acpi_root);
+	if (!result) {
+		memset(&ops, 0, sizeof(ops));
+		ops.acpi_op_add = 1;
+		ops.acpi_op_start = 1;
+		result = acpi_bus_scan(acpi_root, &ops);
+	}
 
 	if (result)
 		acpi_device_unregister(acpi_root, ACPI_BUS_REMOVAL_NORMAL);
diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h
index c627bc408a6..9ad142476f3 100644
--- a/include/acpi/acpi_bus.h
+++ b/include/acpi/acpi_bus.h
@@ -108,6 +108,21 @@ typedef int (*acpi_op_unbind)	(struct acpi_device *device);
 typedef int (*acpi_op_match)	(struct acpi_device *device,
 				 struct acpi_driver *driver);
 
+struct acpi_bus_ops {
+	u32 			acpi_op_add:1;
+	u32			acpi_op_remove:1;
+	u32			acpi_op_lock:1;
+	u32			acpi_op_start:1;
+	u32			acpi_op_stop:1;
+	u32			acpi_op_suspend:1;
+	u32			acpi_op_resume:1;
+	u32			acpi_op_scan:1;
+	u32			acpi_op_bind:1;
+	u32			acpi_op_unbind:1;
+	u32			acpi_op_match:1;
+	u32			reserved:21;
+};
+
 struct acpi_device_ops {
 	acpi_op_add		add;
 	acpi_op_remove		remove;
@@ -327,9 +342,9 @@ int acpi_bus_generate_event (struct acpi_device *device, u8 type, int data);
 int acpi_bus_receive_event (struct acpi_bus_event *event);
 int acpi_bus_register_driver (struct acpi_driver *driver);
 int acpi_bus_unregister_driver (struct acpi_driver *driver);
-int acpi_bus_scan (struct acpi_device *start);
 int acpi_bus_add (struct acpi_device **child, struct acpi_device *parent,
 		acpi_handle handle, int type);
+int acpi_bus_start (struct acpi_device *device);
 
 
 int acpi_match_ids (struct acpi_device	*device, char	*ids);
-- 
cgit v1.2.3-70-g09d2


From 4ce448e5fae62689b06027b46f470b944e5c2193 Mon Sep 17 00:00:00 2001
From: Rajesh Shah <rajesh.shah@intel.com>
Date: Thu, 28 Apr 2005 00:25:53 -0700
Subject: [PATCH] acpi bridge hotadd: Export the interface to get PCI id for an
 ACPI handle

Export an acpi interface to get PCI domain/bus/devfn information from the
corresponding namespace handle.  Used by acpiphp code to transpate the device
handle of the hot-plugged root bridge to the corresponding pci location
information.

Signed-off-by: Rajesh Shah <rajesh.shah@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/acpi/pci_bind.c     | 11 +++++------
 include/acpi/acpi_drivers.h |  1 +
 2 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/drivers/acpi/pci_bind.c b/drivers/acpi/pci_bind.c
index 7753df1f9fb..5148f3c10b5 100644
--- a/drivers/acpi/pci_bind.c
+++ b/drivers/acpi/pci_bind.c
@@ -61,15 +61,14 @@ acpi_pci_data_handler (
 
 
 /**
- * acpi_os_get_pci_id
+ * acpi_get_pci_id
  * ------------------
  * This function is used by the ACPI Interpreter (a.k.a. Core Subsystem)
  * to resolve PCI information for ACPI-PCI devices defined in the namespace.
  * This typically occurs when resolving PCI operation region information.
  */
-#ifdef ACPI_FUTURE_USAGE
 acpi_status
-acpi_os_get_pci_id (
+acpi_get_pci_id (
 	acpi_handle		handle,
 	struct acpi_pci_id	*id)
 {
@@ -78,7 +77,7 @@ acpi_os_get_pci_id (
 	struct acpi_device	*device = NULL;
 	struct acpi_pci_data	*data = NULL;
 
-	ACPI_FUNCTION_TRACE("acpi_os_get_pci_id");
+	ACPI_FUNCTION_TRACE("acpi_get_pci_id");
 
 	if (!id)
 		return_ACPI_STATUS(AE_BAD_PARAMETER);
@@ -92,7 +91,7 @@ acpi_os_get_pci_id (
 	}
 
 	status = acpi_get_data(handle, acpi_pci_data_handler, (void**) &data);
-	if (ACPI_FAILURE(status) || !data || !data->dev) {
+	if (ACPI_FAILURE(status) || !data) {
 		ACPI_DEBUG_PRINT((ACPI_DB_ERROR, 
 			"Invalid ACPI-PCI context for device %s\n",
 			acpi_device_bid(device)));
@@ -115,7 +114,7 @@ acpi_os_get_pci_id (
 
 	return_ACPI_STATUS(AE_OK);
 }
-#endif  /*  ACPI_FUTURE_USAGE  */
+EXPORT_SYMBOL(acpi_get_pci_id);
 
 	
 int
diff --git a/include/acpi/acpi_drivers.h b/include/acpi/acpi_drivers.h
index c62e92ec43b..4ec722d7338 100644
--- a/include/acpi/acpi_drivers.h
+++ b/include/acpi/acpi_drivers.h
@@ -68,6 +68,7 @@ void acpi_pci_irq_del_prt (int segment, int bus);
 
 struct pci_bus;
 
+acpi_status acpi_get_pci_id (acpi_handle handle, struct acpi_pci_id *id);
 int acpi_pci_bind (struct acpi_device *device);
 int acpi_pci_unbind (struct acpi_device *device);
 int acpi_pci_bind_root (struct acpi_device *device, struct acpi_pci_id *id, struct pci_bus *bus);
-- 
cgit v1.2.3-70-g09d2


From b1bb248a5d2230a3d8ef42199c742194a8580b15 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Thu, 28 Apr 2005 00:25:58 -0700
Subject: [PATCH] ACPI based I/O APIC hot-plug: add interfaces

This patch adds the following new interfaces for I/O xAPIC
hotplug. The implementation of these interfaces depends on each
architecture.

    o int acpi_register_ioapic(acpi_handle handle, u64 phys_addr,
			       u32 gsi_base);

        This new interface is to add a new I/O xAPIC specified by
        phys_addr and gsi_base pair. phys_addr is the physical address
        to which the I/O xAPIC is mapped and gsi_base is global system
        interrupt base of the I/O xAPIC. acpi_register_ioapic returns
        0 on success, or negative value on error.

    o int acpi_unregister_ioapic(acpi_handle handle, u32 gsi_base);

        This new interface is to remove a I/O xAPIC specified by
        gsi_base. acpi_unregister_ioapic returns 0 on success, or
        negative value on error.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/i386/kernel/acpi/boot.c | 16 ++++++++++++++++
 arch/ia64/kernel/acpi.c      | 17 +++++++++++++++++
 include/linux/acpi.h         |  3 +++
 3 files changed, 36 insertions(+)

(limited to 'include')

diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index 9f63ae0f404..f5360d88bcf 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -507,6 +507,22 @@ acpi_unmap_lsapic(int cpu)
 EXPORT_SYMBOL(acpi_unmap_lsapic);
 #endif /* CONFIG_ACPI_HOTPLUG_CPU */
 
+int
+acpi_register_ioapic(acpi_handle handle, u64 phys_addr, u32 gsi_base)
+{
+	/* TBD */
+	return -EINVAL;
+}
+EXPORT_SYMBOL(acpi_register_ioapic);
+
+int
+acpi_unregister_ioapic(acpi_handle handle, u32 gsi_base)
+{
+	/* TBD */
+	return -EINVAL;
+}
+EXPORT_SYMBOL(acpi_unregister_ioapic);
+
 static unsigned long __init
 acpi_scan_rsdp (
 	unsigned long		start,
diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c
index 72dfd9e7de0..ab798867acd 100644
--- a/arch/ia64/kernel/acpi.c
+++ b/arch/ia64/kernel/acpi.c
@@ -825,4 +825,21 @@ acpi_map_iosapic (acpi_handle handle, u32 depth, void *context, void **ret)
 	return AE_OK;
 }
 #endif /* CONFIG_NUMA */
+
+int
+acpi_register_ioapic (acpi_handle handle, u64 phys_addr, u32 gsi_base)
+{
+	/* TBD */
+	return -EINVAL;
+}
+EXPORT_SYMBOL(acpi_register_ioapic);
+
+int
+acpi_unregister_ioapic (acpi_handle handle, u32 gsi_base)
+{
+	/* TBD */
+	return -EINVAL;
+}
+EXPORT_SYMBOL(acpi_unregister_ioapic);
+
 #endif /* CONFIG_ACPI_BOOT */
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index b123cc08773..f5bc298707e 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -407,6 +407,9 @@ int acpi_map_lsapic(acpi_handle handle, int *pcpu);
 int acpi_unmap_lsapic(int cpu);
 #endif /* CONFIG_ACPI_HOTPLUG_CPU */
 
+int acpi_register_ioapic(acpi_handle handle, u64 phys_addr, u32 gsi_base);
+int acpi_unregister_ioapic(acpi_handle handle, u32 gsi_base);
+
 extern int acpi_mp_config;
 
 extern u32 pci_mmcfg_base_addr;
-- 
cgit v1.2.3-70-g09d2


From 0e888adc41ffc02b700ade715c182a17e766af84 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Thu, 28 Apr 2005 00:25:58 -0700
Subject: [PATCH] ACPI based I/O APIC hot-plug: ia64 support

This is an ia64 implementation of acpi_register_ioapic() and
acpi_unregister_ioapic() interfaces.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/ia64/kernel/acpi.c    |  21 ++++---
 arch/ia64/kernel/iosapic.c | 134 ++++++++++++++++++++++++++++++++++++++-------
 include/asm-ia64/iosapic.h |  12 +++-
 3 files changed, 135 insertions(+), 32 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c
index ab798867acd..cda06f88c66 100644
--- a/arch/ia64/kernel/acpi.c
+++ b/arch/ia64/kernel/acpi.c
@@ -236,9 +236,7 @@ acpi_parse_iosapic (acpi_table_entry_header *header, const unsigned long end)
 	if (BAD_MADT_ENTRY(iosapic, end))
 		return -EINVAL;
 
-	iosapic_init(iosapic->address, iosapic->global_irq_base);
-
-	return 0;
+	return iosapic_init(iosapic->address, iosapic->global_irq_base);
 }
 
 
@@ -772,7 +770,7 @@ EXPORT_SYMBOL(acpi_unmap_lsapic);
  
 
 #ifdef CONFIG_ACPI_NUMA
-acpi_status __init
+acpi_status __devinit
 acpi_map_iosapic (acpi_handle handle, u32 depth, void *context, void **ret)
 {
 	struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL};
@@ -829,16 +827,23 @@ acpi_map_iosapic (acpi_handle handle, u32 depth, void *context, void **ret)
 int
 acpi_register_ioapic (acpi_handle handle, u64 phys_addr, u32 gsi_base)
 {
-	/* TBD */
-	return -EINVAL;
+	int err;
+
+	if ((err = iosapic_init(phys_addr, gsi_base)))
+		return err;
+
+#if CONFIG_ACPI_NUMA
+	acpi_map_iosapic(handle, 0, NULL, NULL);
+#endif /* CONFIG_ACPI_NUMA */
+
+	return 0;
 }
 EXPORT_SYMBOL(acpi_register_ioapic);
 
 int
 acpi_unregister_ioapic (acpi_handle handle, u32 gsi_base)
 {
-	/* TBD */
-	return -EINVAL;
+	return iosapic_remove(gsi_base);
 }
 EXPORT_SYMBOL(acpi_unregister_ioapic);
 
diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c
index 88b014381df..c170be095cc 100644
--- a/arch/ia64/kernel/iosapic.c
+++ b/arch/ia64/kernel/iosapic.c
@@ -129,14 +129,13 @@ static struct iosapic {
 	char __iomem	*addr;		/* base address of IOSAPIC */
 	unsigned int 	gsi_base;	/* first GSI assigned to this IOSAPIC */
 	unsigned short 	num_rte;	/* number of RTE in this IOSAPIC */
+	int		rtes_inuse;	/* # of RTEs in use on this IOSAPIC */
 #ifdef CONFIG_NUMA
 	unsigned short	node;		/* numa node association via pxm */
 #endif
 } iosapic_lists[NR_IOSAPICS];
 
-static int num_iosapic;
-
-static unsigned char pcat_compat __initdata;	/* 8259 compatibility flag */
+static unsigned char pcat_compat __devinitdata;	/* 8259 compatibility flag */
 
 static int iosapic_kmalloc_ok;
 static LIST_HEAD(free_rte_list);
@@ -149,7 +148,7 @@ find_iosapic (unsigned int gsi)
 {
 	int i;
 
-	for (i = 0; i < num_iosapic; i++) {
+	for (i = 0; i < NR_IOSAPICS; i++) {
 		if ((unsigned) (gsi - iosapic_lists[i].gsi_base) < iosapic_lists[i].num_rte)
 			return i;
 	}
@@ -598,6 +597,7 @@ register_intr (unsigned int gsi, int vector, unsigned char delivery,
 		rte->refcnt++;
 		list_add_tail(&rte->rte_list, &iosapic_intr_info[vector].rtes);
 		iosapic_intr_info[vector].count++;
+		iosapic_lists[index].rtes_inuse++;
 	}
 	else if (vector_is_shared(vector)) {
 		struct iosapic_intr_info *info = &iosapic_intr_info[vector];
@@ -778,7 +778,7 @@ void
 iosapic_unregister_intr (unsigned int gsi)
 {
 	unsigned long flags;
-	int irq, vector;
+	int irq, vector, index;
 	irq_desc_t *idesc;
 	u32 low32;
 	unsigned long trigger, polarity;
@@ -819,6 +819,9 @@ iosapic_unregister_intr (unsigned int gsi)
 		list_del(&rte->rte_list);
 		iosapic_intr_info[vector].count--;
 		iosapic_free_rte(rte);
+		index = find_iosapic(gsi);
+		iosapic_lists[index].rtes_inuse--;
+		WARN_ON(iosapic_lists[index].rtes_inuse < 0);
 
 		trigger	 = iosapic_intr_info[vector].trigger;
 		polarity = iosapic_intr_info[vector].polarity;
@@ -952,30 +955,86 @@ iosapic_system_init (int system_pcat_compat)
 	}
 }
 
-void __init
+static inline int
+iosapic_alloc (void)
+{
+	int index;
+
+	for (index = 0; index < NR_IOSAPICS; index++)
+		if (!iosapic_lists[index].addr)
+			return index;
+
+	printk(KERN_WARNING "%s: failed to allocate iosapic\n", __FUNCTION__);
+	return -1;
+}
+
+static inline void
+iosapic_free (int index)
+{
+	memset(&iosapic_lists[index], 0, sizeof(iosapic_lists[0]));
+}
+
+static inline int
+iosapic_check_gsi_range (unsigned int gsi_base, unsigned int ver)
+{
+	int index;
+	unsigned int gsi_end, base, end;
+
+	/* check gsi range */
+	gsi_end = gsi_base + ((ver >> 16) & 0xff);
+	for (index = 0; index < NR_IOSAPICS; index++) {
+		if (!iosapic_lists[index].addr)
+			continue;
+
+		base = iosapic_lists[index].gsi_base;
+		end  = base + iosapic_lists[index].num_rte - 1;
+
+		if (gsi_base < base && gsi_end < base)
+			continue;/* OK */
+
+		if (gsi_base > end && gsi_end > end)
+			continue; /* OK */
+
+		return -EBUSY;
+	}
+	return 0;
+}
+
+int __devinit
 iosapic_init (unsigned long phys_addr, unsigned int gsi_base)
 {
-	int num_rte;
+	int num_rte, err, index;
 	unsigned int isa_irq, ver;
 	char __iomem *addr;
+	unsigned long flags;
+
+	spin_lock_irqsave(&iosapic_lock, flags);
+	{
+		addr = ioremap(phys_addr, 0);
+		ver = iosapic_version(addr);
 
-	addr = ioremap(phys_addr, 0);
-	ver = iosapic_version(addr);
+		if ((err = iosapic_check_gsi_range(gsi_base, ver))) {
+			iounmap(addr);
+			spin_unlock_irqrestore(&iosapic_lock, flags);
+			return err;
+		}
 
-	/*
-	 * The MAX_REDIR register holds the highest input pin
-	 * number (starting from 0).
-	 * We add 1 so that we can use it for number of pins (= RTEs)
-	 */
-	num_rte = ((ver >> 16) & 0xff) + 1;
+		/*
+		 * The MAX_REDIR register holds the highest input pin
+		 * number (starting from 0).
+		 * We add 1 so that we can use it for number of pins (= RTEs)
+		 */
+		num_rte = ((ver >> 16) & 0xff) + 1;
 
-	iosapic_lists[num_iosapic].addr = addr;
-	iosapic_lists[num_iosapic].gsi_base = gsi_base;
-	iosapic_lists[num_iosapic].num_rte = num_rte;
+		index = iosapic_alloc();
+		iosapic_lists[index].addr = addr;
+		iosapic_lists[index].gsi_base = gsi_base;
+		iosapic_lists[index].num_rte = num_rte;
 #ifdef CONFIG_NUMA
-	iosapic_lists[num_iosapic].node = MAX_NUMNODES;
+		iosapic_lists[index].node = MAX_NUMNODES;
 #endif
-	num_iosapic++;
+	}
+	spin_unlock_irqrestore(&iosapic_lock, flags);
 
 	if ((gsi_base == 0) && pcat_compat) {
 		/*
@@ -986,10 +1045,43 @@ iosapic_init (unsigned long phys_addr, unsigned int gsi_base)
 		for (isa_irq = 0; isa_irq < 16; ++isa_irq)
 			iosapic_override_isa_irq(isa_irq, isa_irq, IOSAPIC_POL_HIGH, IOSAPIC_EDGE);
 	}
+	return 0;
+}
+
+#ifdef CONFIG_HOTPLUG
+int
+iosapic_remove (unsigned int gsi_base)
+{
+	int index, err = 0;
+	unsigned long flags;
+
+	spin_lock_irqsave(&iosapic_lock, flags);
+	{
+		index = find_iosapic(gsi_base);
+		if (index < 0) {
+			printk(KERN_WARNING "%s: No IOSAPIC for GSI base %u\n",
+			       __FUNCTION__, gsi_base);
+			goto out;
+		}
+
+		if (iosapic_lists[index].rtes_inuse) {
+			err = -EBUSY;
+			printk(KERN_WARNING "%s: IOSAPIC for GSI base %u is busy\n",
+			       __FUNCTION__, gsi_base);
+			goto out;
+		}
+
+		iounmap(iosapic_lists[index].addr);
+		iosapic_free(index);
+	}
+ out:
+	spin_unlock_irqrestore(&iosapic_lock, flags);
+	return err;
 }
+#endif /* CONFIG_HOTPLUG */
 
 #ifdef CONFIG_NUMA
-void __init
+void __devinit
 map_iosapic_to_node(unsigned int gsi_base, int node)
 {
 	int index;
diff --git a/include/asm-ia64/iosapic.h b/include/asm-ia64/iosapic.h
index 38a7a72791c..1093f35b3b9 100644
--- a/include/asm-ia64/iosapic.h
+++ b/include/asm-ia64/iosapic.h
@@ -71,8 +71,11 @@ static inline void iosapic_eoi(char __iomem *iosapic, u32 vector)
 }
 
 extern void __init iosapic_system_init (int pcat_compat);
-extern void __init iosapic_init (unsigned long address,
+extern int __devinit iosapic_init (unsigned long address,
 				    unsigned int gsi_base);
+#ifdef CONFIG_HOTPLUG
+extern int iosapic_remove (unsigned int gsi_base);
+#endif /* CONFIG_HOTPLUG */
 extern int gsi_to_vector (unsigned int gsi);
 extern int gsi_to_irq (unsigned int gsi);
 extern void iosapic_enable_intr (unsigned int vector);
@@ -94,11 +97,14 @@ extern unsigned int iosapic_version (char __iomem *addr);
 
 extern void iosapic_pci_fixup (int);
 #ifdef CONFIG_NUMA
-extern void __init map_iosapic_to_node (unsigned int, int);
+extern void __devinit map_iosapic_to_node (unsigned int, int);
 #endif
 #else
 #define iosapic_system_init(pcat_compat)			do { } while (0)
-#define iosapic_init(address,gsi_base)				do { } while (0)
+#define iosapic_init(address,gsi_base)				(-EINVAL)
+#ifdef CONFIG_HOTPLUG
+#define iosapic_remove(gsi_base)				(-ENODEV)
+#endif /* CONFIG_HOTPLUG */
 #define iosapic_register_intr(gsi,polarity,trigger)		(gsi)
 #define iosapic_unregister_intr(irq)				do { } while (0)
 #define iosapic_override_isa_irq(isa_irq,gsi,polarity,trigger)	do { } while (0)
-- 
cgit v1.2.3-70-g09d2


From a0d399a808916d22c1c222c6b5ca4e8edd6d91a9 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Thu, 28 Apr 2005 00:25:59 -0700
Subject: [PATCH] ACPI based I/O APIC hot-plug: acpiphp support

This patch adds PCI based I/O xAPIC hot-add support to ACPIPHP
driver. When PCI root bridge is hot-added, all PCI based I/O xAPICs
under the root bridge are hot-added by this patch. Hot-remove support
is TBD.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/hotplug/acpiphp_glue.c | 127 +++++++++++++++++++++++++++++++++++++
 include/linux/pci_ids.h            |   2 +
 2 files changed, 129 insertions(+)

(limited to 'include')

diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c
index b4a92123625..424e7de181a 100644
--- a/drivers/pci/hotplug/acpiphp_glue.c
+++ b/drivers/pci/hotplug/acpiphp_glue.c
@@ -552,6 +552,132 @@ static void remove_bridge(acpi_handle handle)
 	}
 }
 
+static struct pci_dev * get_apic_pci_info(acpi_handle handle)
+{
+	struct acpi_pci_id id;
+	struct pci_bus *bus;
+	struct pci_dev *dev;
+
+	if (ACPI_FAILURE(acpi_get_pci_id(handle, &id)))
+		return NULL;
+
+	bus = pci_find_bus(id.segment, id.bus);
+	if (!bus)
+		return NULL;
+
+	dev = pci_get_slot(bus, PCI_DEVFN(id.device, id.function));
+	if (!dev)
+		return NULL;
+
+	if ((dev->class != PCI_CLASS_SYSTEM_PIC_IOAPIC) &&
+	    (dev->class != PCI_CLASS_SYSTEM_PIC_IOXAPIC))
+	{
+		pci_dev_put(dev);
+		return NULL;
+	}
+
+	return dev;
+}
+
+static int get_gsi_base(acpi_handle handle, u32 *gsi_base)
+{
+	acpi_status status;
+	int result = -1;
+	unsigned long gsb;
+	struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL};
+	union acpi_object *obj;
+	void *table;
+
+	status = acpi_evaluate_integer(handle, "_GSB", NULL, &gsb);
+	if (ACPI_SUCCESS(status)) {
+		*gsi_base = (u32)gsb;
+		return 0;
+	}
+
+	status = acpi_evaluate_object(handle, "_MAT", NULL, &buffer);
+	if (ACPI_FAILURE(status) || !buffer.length || !buffer.pointer)
+		return -1;
+
+	obj = buffer.pointer;
+	if (obj->type != ACPI_TYPE_BUFFER)
+		goto out;
+
+	table = obj->buffer.pointer;
+	switch (((acpi_table_entry_header *)table)->type) {
+	case ACPI_MADT_IOSAPIC:
+		*gsi_base = ((struct acpi_table_iosapic *)table)->global_irq_base;
+		result = 0;
+		break;
+	case ACPI_MADT_IOAPIC:
+		*gsi_base = ((struct acpi_table_ioapic *)table)->global_irq_base;
+		result = 0;
+		break;
+	default:
+		break;
+	}
+ out:
+	acpi_os_free(buffer.pointer);
+	return result;
+}
+
+static acpi_status
+ioapic_add(acpi_handle handle, u32 lvl, void *context, void **rv)
+{
+	acpi_status status;
+	unsigned long sta;
+	acpi_handle tmp;
+	struct pci_dev *pdev;
+	u32 gsi_base;
+	u64 phys_addr;
+
+	/* Evaluate _STA if present */
+	status = acpi_evaluate_integer(handle, "_STA", NULL, &sta);
+	if (ACPI_SUCCESS(status) && sta != ACPI_STA_ALL)
+		return AE_CTRL_DEPTH;
+
+	/* Scan only PCI bus scope */
+	status = acpi_get_handle(handle, "_HID", &tmp);
+	if (ACPI_SUCCESS(status))
+		return AE_CTRL_DEPTH;
+
+	if (get_gsi_base(handle, &gsi_base))
+		return AE_OK;
+
+	pdev = get_apic_pci_info(handle);
+	if (!pdev)
+		return AE_OK;
+
+	if (pci_enable_device(pdev)) {
+		pci_dev_put(pdev);
+		return AE_OK;
+	}
+
+	pci_set_master(pdev);
+
+	if (pci_request_region(pdev, 0, "I/O APIC(acpiphp)")) {
+		pci_disable_device(pdev);
+		pci_dev_put(pdev);
+		return AE_OK;
+	}
+
+	phys_addr = pci_resource_start(pdev, 0);
+	if (acpi_register_ioapic(handle, phys_addr, gsi_base)) {
+		pci_release_region(pdev, 0);
+		pci_disable_device(pdev);
+		pci_dev_put(pdev);
+		return AE_OK;
+	}
+
+	return AE_OK;
+}
+
+static int acpiphp_configure_ioapics(acpi_handle handle)
+{
+	acpi_walk_namespace(ACPI_TYPE_DEVICE, handle,
+			    ACPI_UINT32_MAX, ioapic_add, NULL, NULL);
+	return 0;
+}
+
 static int power_on_slot(struct acpiphp_slot *slot)
 {
 	acpi_status status;
@@ -942,6 +1068,7 @@ static int acpiphp_configure_bridge (acpi_handle handle)
 	acpiphp_sanitize_bus(bus);
 	acpiphp_set_hpp_values(handle, bus);
 	pci_enable_bridges(bus);
+	acpiphp_configure_ioapics(handle);
 	return 0;
 }
 
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index bf608808a60..810bbbcee40 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -62,6 +62,8 @@
 
 #define PCI_BASE_CLASS_SYSTEM		0x08
 #define PCI_CLASS_SYSTEM_PIC		0x0800
+#define PCI_CLASS_SYSTEM_PIC_IOAPIC	0x080010
+#define PCI_CLASS_SYSTEM_PIC_IOXAPIC	0x080020
 #define PCI_CLASS_SYSTEM_DMA		0x0801
 #define PCI_CLASS_SYSTEM_TIMER		0x0802
 #define PCI_CLASS_SYSTEM_RTC		0x0803
-- 
cgit v1.2.3-70-g09d2


From 2311b1f2bbd36fa5f366a7448c718b2556e0f02c Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Fri, 13 May 2005 17:44:10 +1000
Subject: [PATCH] PCI: fix-pci-mmap-on-ppc-and-ppc64.patch

This is an updated version of Ben's fix-pci-mmap-on-ppc-and-ppc64.patch
which is in 2.6.12-rc4-mm1.

It fixes the patch to work on PPC iSeries, removes some debug printks
at Ben's request, and incorporates your
fix-pci-mmap-on-ppc-and-ppc64-fix.patch also.

Originally from Benjamin Herrenschmidt <benh@kernel.crashing.org>

This patch was discussed at length on linux-pci and so far, the last
iteration of it didn't raise any comment.  It's effect is a nop on
architecture that don't define the new pci_resource_to_user() callback
anyway.  It allows architecture like ppc who put weird things inside of
PCI resource structures to convert to some different value for user
visible ones.  It also fixes mmap'ing of IO space on those archs.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/ppc/kernel/pci.c   | 21 +++++++++++++++++++--
 arch/ppc64/kernel/pci.c | 22 ++++++++++++++++++++--
 drivers/pci/pci-sysfs.c | 26 +++++++++++++++++++++-----
 drivers/pci/proc.c      | 14 ++++++++++----
 include/asm-ppc/pci.h   |  6 ++++++
 include/asm-ppc64/pci.h |  7 +++++++
 include/linux/pci.h     | 14 ++++++++++++++
 7 files changed, 97 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/arch/ppc/kernel/pci.c b/arch/ppc/kernel/pci.c
index 6d7b92d7245..70cfb6ffd87 100644
--- a/arch/ppc/kernel/pci.c
+++ b/arch/ppc/kernel/pci.c
@@ -1495,7 +1495,7 @@ static struct resource *__pci_mmap_make_offset(struct pci_dev *dev,
 		*offset += hose->pci_mem_offset;
 		res_bit = IORESOURCE_MEM;
 	} else {
-		io_offset = (unsigned long)hose->io_base_virt;
+		io_offset = hose->io_base_virt - ___IO_BASE;
 		*offset += io_offset;
 		res_bit = IORESOURCE_IO;
 	}
@@ -1522,7 +1522,7 @@ static struct resource *__pci_mmap_make_offset(struct pci_dev *dev,
 
 		/* found it! construct the final physical address */
 		if (mmap_state == pci_mmap_io)
-			*offset += hose->io_base_phys - _IO_BASE;
+			*offset += hose->io_base_phys - io_offset;
 		return rp;
 	}
 
@@ -1739,6 +1739,23 @@ long sys_pciconfig_iobase(long which, unsigned long bus, unsigned long devfn)
 	return result;
 }
 
+void pci_resource_to_user(const struct pci_dev *dev, int bar,
+			  const struct resource *rsrc,
+			  u64 *start, u64 *end)
+{
+	struct pci_controller *hose = pci_bus_to_hose(dev->bus->number);
+	unsigned long offset = 0;
+
+	if (hose == NULL)
+		return;
+
+	if (rsrc->flags & IORESOURCE_IO)
+		offset = ___IO_BASE - hose->io_base_virt + hose->io_base_phys;
+
+	*start = rsrc->start + offset;
+	*end = rsrc->end + offset;
+}
+
 void __init
 pci_init_resource(struct resource *res, unsigned long start, unsigned long end,
 		  int flags, char *name)
diff --git a/arch/ppc64/kernel/pci.c b/arch/ppc64/kernel/pci.c
index 580676f87d2..ae6f579d3fa 100644
--- a/arch/ppc64/kernel/pci.c
+++ b/arch/ppc64/kernel/pci.c
@@ -351,7 +351,7 @@ static struct resource *__pci_mmap_make_offset(struct pci_dev *dev,
 		*offset += hose->pci_mem_offset;
 		res_bit = IORESOURCE_MEM;
 	} else {
-		io_offset = (unsigned long)hose->io_base_virt;
+		io_offset = (unsigned long)hose->io_base_virt - pci_io_base;
 		*offset += io_offset;
 		res_bit = IORESOURCE_IO;
 	}
@@ -378,7 +378,7 @@ static struct resource *__pci_mmap_make_offset(struct pci_dev *dev,
 
 		/* found it! construct the final physical address */
 		if (mmap_state == pci_mmap_io)
-			*offset += hose->io_base_phys - io_offset;
+		       	*offset += hose->io_base_phys - io_offset;
 		return rp;
 	}
 
@@ -944,4 +944,22 @@ int pci_read_irq_line(struct pci_dev *pci_dev)
 }
 EXPORT_SYMBOL(pci_read_irq_line);
 
+void pci_resource_to_user(const struct pci_dev *dev, int bar,
+			  const struct resource *rsrc,
+			  u64 *start, u64 *end)
+{
+	struct pci_controller *hose = pci_bus_to_host(dev->bus);
+	unsigned long offset = 0;
+
+	if (hose == NULL)
+		return;
+
+	if (rsrc->flags & IORESOURCE_IO)
+		offset = pci_io_base - (unsigned long)hose->io_base_virt +
+			hose->io_base_phys;
+
+	*start = rsrc->start + offset;
+	*end = rsrc->end + offset;
+}
+
 #endif /* CONFIG_PPC_MULTIPLATFORM */
diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index a15f94072a6..cc9d65388e6 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -60,15 +60,18 @@ resource_show(struct device * dev, struct device_attribute *attr, char * buf)
 	char * str = buf;
 	int i;
 	int max = 7;
+	u64 start, end;
 
 	if (pci_dev->subordinate)
 		max = DEVICE_COUNT_RESOURCE;
 
 	for (i = 0; i < max; i++) {
-		str += sprintf(str,"0x%016lx 0x%016lx 0x%016lx\n",
-			       pci_resource_start(pci_dev,i),
-			       pci_resource_end(pci_dev,i),
-			       pci_resource_flags(pci_dev,i));
+		struct resource *res =  &pci_dev->resource[i];
+		pci_resource_to_user(pci_dev, i, res, &start, &end);
+		str += sprintf(str,"0x%016llx 0x%016llx 0x%016llx\n",
+			       (unsigned long long)start,
+			       (unsigned long long)end,
+			       (unsigned long long)res->flags);
 	}
 	return (str - buf);
 }
@@ -313,8 +316,21 @@ pci_mmap_resource(struct kobject *kobj, struct bin_attribute *attr,
 						       struct device, kobj));
 	struct resource *res = (struct resource *)attr->private;
 	enum pci_mmap_state mmap_type;
+	u64 start, end;
+	int i;
 
-	vma->vm_pgoff += res->start >> PAGE_SHIFT;
+	for (i = 0; i < PCI_ROM_RESOURCE; i++)
+		if (res == &pdev->resource[i])
+			break;
+	if (i >= PCI_ROM_RESOURCE)
+		return -ENODEV;
+
+	/* pci_mmap_page_range() expects the same kind of entry as coming
+	 * from /proc/bus/pci/ which is a "user visible" value. If this is
+	 * different from the resource itself, arch will do necessary fixup.
+	 */
+	pci_resource_to_user(pdev, i, res, &start, &end);
+	vma->vm_pgoff += start >> PAGE_SHIFT;
 	mmap_type = res->flags & IORESOURCE_MEM ? pci_mmap_mem : pci_mmap_io;
 
 	return pci_mmap_page_range(pdev, vma, mmap_type, 0);
diff --git a/drivers/pci/proc.c b/drivers/pci/proc.c
index e68bbfb1e7c..7988fc8df3f 100644
--- a/drivers/pci/proc.c
+++ b/drivers/pci/proc.c
@@ -355,14 +355,20 @@ static int show_device(struct seq_file *m, void *v)
 			dev->device,
 			dev->irq);
 	/* Here should be 7 and not PCI_NUM_RESOURCES as we need to preserve compatibility */
-	for(i=0; i<7; i++)
+	for (i=0; i<7; i++) {
+		u64 start, end;
+		pci_resource_to_user(dev, i, &dev->resource[i], &start, &end);
 		seq_printf(m, LONG_FORMAT,
-			dev->resource[i].start |
+			((unsigned long)start) |
 			(dev->resource[i].flags & PCI_REGION_FLAG_MASK));
-	for(i=0; i<7; i++)
+	}
+	for (i=0; i<7; i++) {
+		u64 start, end;
+		pci_resource_to_user(dev, i, &dev->resource[i], &start, &end);
 		seq_printf(m, LONG_FORMAT,
 			dev->resource[i].start < dev->resource[i].end ?
-			dev->resource[i].end - dev->resource[i].start + 1 : 0);
+			(unsigned long)(end - start) + 1 : 0);
+	}
 	seq_putc(m, '\t');
 	if (drv)
 		seq_printf(m, "%s", drv->name);
diff --git a/include/asm-ppc/pci.h b/include/asm-ppc/pci.h
index ce5ae6d048f..002e7b30577 100644
--- a/include/asm-ppc/pci.h
+++ b/include/asm-ppc/pci.h
@@ -103,6 +103,12 @@ extern pgprot_t	pci_phys_mem_access_prot(struct file *file,
 					 unsigned long size,
 					 pgprot_t prot);
 
+#define HAVE_ARCH_PCI_RESOURCE_TO_USER
+extern void pci_resource_to_user(const struct pci_dev *dev, int bar,
+				 const struct resource *rsrc,
+				 u64 *start, u64 *end);
+
+
 #endif	/* __KERNEL__ */
 
 #endif /* __PPC_PCI_H */
diff --git a/include/asm-ppc64/pci.h b/include/asm-ppc64/pci.h
index 6cd593f660a..411bf5dee39 100644
--- a/include/asm-ppc64/pci.h
+++ b/include/asm-ppc64/pci.h
@@ -136,6 +136,13 @@ extern pgprot_t	pci_phys_mem_access_prot(struct file *file,
 					 unsigned long size,
 					 pgprot_t prot);
 
+#ifdef CONFIG_PPC_MULTIPLATFORM
+#define HAVE_ARCH_PCI_RESOURCE_TO_USER
+extern void pci_resource_to_user(const struct pci_dev *dev, int bar,
+				 const struct resource *rsrc,
+				 u64 *start, u64 *end);
+#endif /* CONFIG_PPC_MULTIPLATFORM */
+
 
 #endif	/* __KERNEL__ */
 
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 0e9844929fe..cfa1455848f 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -1020,6 +1020,20 @@ static inline char *pci_name(struct pci_dev *pdev)
 #define pci_pretty_name(dev) ""
 #endif
 
+
+/* Some archs don't want to expose struct resource to userland as-is
+ * in sysfs and /proc
+ */
+#ifndef HAVE_ARCH_PCI_RESOURCE_TO_USER
+static inline void pci_resource_to_user(const struct pci_dev *dev, int bar,
+                const struct resource *rsrc, u64 *start, u64 *end)
+{
+	*start = rsrc->start;
+	*end = rsrc->end;
+}
+#endif /* HAVE_ARCH_PCI_RESOURCE_TO_USER */
+
+
 /*
  *  The world is not perfect and supplies us with broken PCI devices.
  *  For at least a part of these bugs we need a work-around, so both
-- 
cgit v1.2.3-70-g09d2


From e24c2d963a604d9eaa560c90371fa387d3eec8f1 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 2 Jun 2005 12:55:50 -0700
Subject: [PATCH] PCI: DMA bursting advice

After seeing, at best, "guesses" as to the following kind
of information in several drivers, I decided that we really
need a way for platforms to specifically give advice in this
area for what works best with their PCI controller implementation.

Basically, this new interface gives DMA bursting advice on
PCI.  There are three forms of the advice:

1) Burst as much as possible, it is not necessary to end bursts
   on some particular boundary for best performance.

2) Burst on some byte count multiple.  A DMA burst to some multiple of
   number of bytes may be done, but it is important to end the burst
   on an exact multiple for best performance.

   The best example of this I am aware of are the PPC64 PCI
   controllers, where if you end a burst mid-cacheline then
   chip has to refetch the data and the IOMMU translations
   which hurts performance a lot.

3) Burst on a single byte count multiple.  Bursts shall end
   exactly on the next multiple boundary for best performance.

   Sparc64 and Alpha's PCI controllers operate this way.  They
   disconnect any device which tries to burst across a cacheline
   boundary.

   Actually, newer sparc64 PCI controllers do not have this behavior.
   That is why the "pdev" is passed into the interface, so I can
   add code later to check which PCI controller the system is using
   and give advice accordingly.

Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/asm-alpha/pci.h   | 17 +++++++++++++++++
 include/asm-arm/pci.h     |  8 ++++++++
 include/asm-frv/pci.h     |  8 ++++++++
 include/asm-i386/pci.h    |  8 ++++++++
 include/asm-ia64/pci.h    | 17 +++++++++++++++++
 include/asm-mips/pci.h    |  8 ++++++++
 include/asm-parisc/pci.h  | 17 +++++++++++++++++
 include/asm-ppc/pci.h     |  8 ++++++++
 include/asm-ppc64/pci.h   | 17 +++++++++++++++++
 include/asm-sh/pci.h      |  8 ++++++++
 include/asm-sh64/pci.h    |  8 ++++++++
 include/asm-sparc/pci.h   |  8 ++++++++
 include/asm-sparc64/pci.h | 17 +++++++++++++++++
 include/asm-v850/pci.h    |  8 ++++++++
 include/asm-x86_64/pci.h  |  8 ++++++++
 include/linux/pci.h       |  9 +++++++++
 16 files changed, 174 insertions(+)

(limited to 'include')

diff --git a/include/asm-alpha/pci.h b/include/asm-alpha/pci.h
index 0c7b57bc043..6c71dc1ad4c 100644
--- a/include/asm-alpha/pci.h
+++ b/include/asm-alpha/pci.h
@@ -223,6 +223,23 @@ pci_dac_dma_sync_single_for_device(struct pci_dev *pdev, dma64_addr_t dma_addr,
 	/* Nothing to do. */
 }
 
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	unsigned long cacheline_size;
+	u8 byte;
+
+	pci_read_config_byte(pdev, PCI_CACHE_LINE_SIZE, &byte);
+	if (byte == 0)
+		cacheline_size = 1024;
+	else
+		cacheline_size = (int) byte * 4;
+
+	*strat = PCI_DMA_BURST_BOUNDARY;
+	*strategy_parameter = cacheline_size;
+}
+
 /* TODO: integrate with include/asm-generic/pci.h ? */
 static inline int pci_get_legacy_ide_irq(struct pci_dev *dev, int channel)
 {
diff --git a/include/asm-arm/pci.h b/include/asm-arm/pci.h
index 40ffaefbeb1..bc2ec425aca 100644
--- a/include/asm-arm/pci.h
+++ b/include/asm-arm/pci.h
@@ -42,6 +42,14 @@ static inline void pcibios_penalize_isa_irq(int irq)
 #define pci_unmap_len(PTR, LEN_NAME)		((PTR)->LEN_NAME)
 #define pci_unmap_len_set(PTR, LEN_NAME, VAL)	(((PTR)->LEN_NAME) = (VAL))
 
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	*strat = PCI_DMA_BURST_INFINITY;
+	*strategy_parameter = ~0UL;
+}
+
 #define HAVE_PCI_MMAP
 extern int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
                                enum pci_mmap_state mmap_state, int write_combine);
diff --git a/include/asm-frv/pci.h b/include/asm-frv/pci.h
index a6a469231f6..13427240664 100644
--- a/include/asm-frv/pci.h
+++ b/include/asm-frv/pci.h
@@ -57,6 +57,14 @@ extern void pci_free_consistent(struct pci_dev *hwdev, size_t size,
  */
 #define PCI_DMA_BUS_IS_PHYS	(1)
 
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	*strat = PCI_DMA_BURST_INFINITY;
+	*strategy_parameter = ~0UL;
+}
+
 /*
  *	These are pretty much arbitary with the CoMEM implementation.
  *	We have the whole address space to ourselves.
diff --git a/include/asm-i386/pci.h b/include/asm-i386/pci.h
index fb749b85a73..bf07b3af85e 100644
--- a/include/asm-i386/pci.h
+++ b/include/asm-i386/pci.h
@@ -99,6 +99,14 @@ static inline void pcibios_add_platform_entries(struct pci_dev *dev)
 {
 }
 
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	*strat = PCI_DMA_BURST_INFINITY;
+	*strategy_parameter = ~0UL;
+}
+
 #endif /* __KERNEL__ */
 
 /* implement the pci_ DMA API in terms of the generic device dma_ one */
diff --git a/include/asm-ia64/pci.h b/include/asm-ia64/pci.h
index a8314ee4e7d..c9f1ab4e477 100644
--- a/include/asm-ia64/pci.h
+++ b/include/asm-ia64/pci.h
@@ -82,6 +82,23 @@ extern int pcibios_prep_mwi (struct pci_dev *);
 #define sg_dma_len(sg)		((sg)->dma_length)
 #define sg_dma_address(sg)	((sg)->dma_address)
 
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	unsigned long cacheline_size;
+	u8 byte;
+
+	pci_read_config_byte(pdev, PCI_CACHE_LINE_SIZE, &byte);
+	if (byte == 0)
+		cacheline_size = 1024;
+	else
+		cacheline_size = (int) byte * 4;
+
+	*strat = PCI_DMA_BURST_MULTIPLE;
+	*strategy_parameter = cacheline_size;
+}
+
 #define HAVE_PCI_MMAP
 extern int pci_mmap_page_range (struct pci_dev *dev, struct vm_area_struct *vma,
 				enum pci_mmap_state mmap_state, int write_combine);
diff --git a/include/asm-mips/pci.h b/include/asm-mips/pci.h
index c9c576b4855..20b93bfa456 100644
--- a/include/asm-mips/pci.h
+++ b/include/asm-mips/pci.h
@@ -130,6 +130,14 @@ extern void pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev,
 extern void pci_dac_dma_sync_single_for_device(struct pci_dev *pdev,
 	dma64_addr_t dma_addr, size_t len, int direction);
 
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	*strat = PCI_DMA_BURST_INFINITY;
+	*strategy_parameter = ~0UL;
+}
+
 extern void pcibios_resource_to_bus(struct pci_dev *dev,
 	struct pci_bus_region *region, struct resource *res);
 
diff --git a/include/asm-parisc/pci.h b/include/asm-parisc/pci.h
index 0763c2982fb..f9f5bf90111 100644
--- a/include/asm-parisc/pci.h
+++ b/include/asm-parisc/pci.h
@@ -230,6 +230,23 @@ extern inline void pcibios_register_hba(struct pci_hba_data *x)
 /* export the pci_ DMA API in terms of the dma_ one */
 #include <asm-generic/pci-dma-compat.h>
 
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	unsigned long cacheline_size;
+	u8 byte;
+
+	pci_read_config_byte(pdev, PCI_CACHE_LINE_SIZE, &byte);
+	if (byte == 0)
+		cacheline_size = 1024;
+	else
+		cacheline_size = (int) byte * 4;
+
+	*strat = PCI_DMA_BURST_MULTIPLE;
+	*strategy_parameter = cacheline_size;
+}
+
 extern void
 pcibios_resource_to_bus(struct pci_dev *dev, struct pci_bus_region *region,
 			 struct resource *res);
diff --git a/include/asm-ppc/pci.h b/include/asm-ppc/pci.h
index 002e7b30577..669e9de7a52 100644
--- a/include/asm-ppc/pci.h
+++ b/include/asm-ppc/pci.h
@@ -69,6 +69,14 @@ extern unsigned long pci_bus_to_phys(unsigned int ba, int busnr);
 #define pci_unmap_len(PTR, LEN_NAME)		(0)
 #define pci_unmap_len_set(PTR, LEN_NAME, VAL)	do { } while (0)
 
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	*strat = PCI_DMA_BURST_INFINITY;
+	*strategy_parameter = ~0UL;
+}
+
 /*
  * At present there are very few 32-bit PPC machines that can have
  * memory above the 4GB point, and we don't support that.
diff --git a/include/asm-ppc64/pci.h b/include/asm-ppc64/pci.h
index 411bf5dee39..20beb10c090 100644
--- a/include/asm-ppc64/pci.h
+++ b/include/asm-ppc64/pci.h
@@ -78,6 +78,23 @@ static inline int pci_dac_dma_supported(struct pci_dev *hwdev,u64 mask)
 	return 0;
 }
 
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	unsigned long cacheline_size;
+	u8 byte;
+
+	pci_read_config_byte(pdev, PCI_CACHE_LINE_SIZE, &byte);
+	if (byte == 0)
+		cacheline_size = 1024;
+	else
+		cacheline_size = (int) byte * 4;
+
+	*strat = PCI_DMA_BURST_MULTIPLE;
+	*strategy_parameter = cacheline_size;
+}
+
 extern int pci_domain_nr(struct pci_bus *bus);
 
 /* Decide whether to display the domain number in /proc */
diff --git a/include/asm-sh/pci.h b/include/asm-sh/pci.h
index 9c3b63d0105..7237bc6a728 100644
--- a/include/asm-sh/pci.h
+++ b/include/asm-sh/pci.h
@@ -96,6 +96,14 @@ static inline void pcibios_penalize_isa_irq(int irq)
 #define sg_dma_address(sg)	(virt_to_bus((sg)->dma_address))
 #define sg_dma_len(sg)		((sg)->length)
 
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	*strat = PCI_DMA_BURST_INFINITY;
+	*strategy_parameter = ~0UL;
+}
+
 /* Board-specific fixup routines. */
 extern void pcibios_fixup(void);
 extern void pcibios_fixup_irqs(void);
diff --git a/include/asm-sh64/pci.h b/include/asm-sh64/pci.h
index 8cc14e13975..0ac15ab01cc 100644
--- a/include/asm-sh64/pci.h
+++ b/include/asm-sh64/pci.h
@@ -86,6 +86,14 @@ static inline void pcibios_penalize_isa_irq(int irq)
 #define sg_dma_address(sg)	((sg)->dma_address)
 #define sg_dma_len(sg)		((sg)->length)
 
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	*strat = PCI_DMA_BURST_INFINITY;
+	*strategy_parameter = ~0UL;
+}
+
 /* Board-specific fixup routines. */
 extern void pcibios_fixup(void);
 extern void pcibios_fixup_irqs(void);
diff --git a/include/asm-sparc/pci.h b/include/asm-sparc/pci.h
index d200a25a737..2fd65db95e9 100644
--- a/include/asm-sparc/pci.h
+++ b/include/asm-sparc/pci.h
@@ -144,6 +144,14 @@ extern inline int pci_dma_supported(struct pci_dev *hwdev, u64 mask)
 
 #define pci_dac_dma_supported(dev, mask)	(0)
 
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	*strat = PCI_DMA_BURST_INFINITY;
+	*strategy_parameter = ~0UL;
+}
+
 static inline void pcibios_add_platform_entries(struct pci_dev *dev)
 {
 }
diff --git a/include/asm-sparc64/pci.h b/include/asm-sparc64/pci.h
index 2a0c85cd1c1..402667300d0 100644
--- a/include/asm-sparc64/pci.h
+++ b/include/asm-sparc64/pci.h
@@ -220,6 +220,23 @@ static inline int pci_dma_mapping_error(dma_addr_t dma_addr)
 	return (dma_addr == PCI_DMA_ERROR_CODE);
 }
 
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	unsigned long cacheline_size;
+	u8 byte;
+
+	pci_read_config_byte(pdev, PCI_CACHE_LINE_SIZE, &byte);
+	if (byte == 0)
+		cacheline_size = 1024;
+	else
+		cacheline_size = (int) byte * 4;
+
+	*strat = PCI_DMA_BURST_BOUNDARY;
+	*strategy_parameter = cacheline_size;
+}
+
 /* Return the index of the PCI controller for device PDEV. */
 
 extern int pci_domain_nr(struct pci_bus *bus);
diff --git a/include/asm-v850/pci.h b/include/asm-v850/pci.h
index e41941447b4..d26eb8d6731 100644
--- a/include/asm-v850/pci.h
+++ b/include/asm-v850/pci.h
@@ -81,6 +81,14 @@ extern void
 pci_free_consistent (struct pci_dev *pdev, size_t size, void *cpu_addr,
 		     dma_addr_t dma_addr);
 
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	*strat = PCI_DMA_BURST_INFINITY;
+	*strategy_parameter = ~0UL;
+}
+
 static inline void pcibios_add_platform_entries(struct pci_dev *dev)
 {
 }
diff --git a/include/asm-x86_64/pci.h b/include/asm-x86_64/pci.h
index 8712520ca47..8461d6af102 100644
--- a/include/asm-x86_64/pci.h
+++ b/include/asm-x86_64/pci.h
@@ -123,6 +123,14 @@ pci_dac_dma_sync_single_for_device(struct pci_dev *pdev, dma64_addr_t dma_addr,
 	flush_write_buffers();
 }
 
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	*strat = PCI_DMA_BURST_INFINITY;
+	*strategy_parameter = ~0UL;
+}
+
 #define HAVE_PCI_MMAP
 extern int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
 			       enum pci_mmap_state mmap_state, int write_combine);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index cfa1455848f..9ce4f1be093 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -874,6 +874,15 @@ int pci_scan_bridge(struct pci_bus *bus, struct pci_dev * dev, int max, int pass
 #define	pci_pool_alloc(pool, flags, handle) dma_pool_alloc(pool, flags, handle)
 #define	pci_pool_free(pool, vaddr, addr) dma_pool_free(pool, vaddr, addr)
 
+enum pci_dma_burst_strategy {
+	PCI_DMA_BURST_INFINITY,	/* make bursts as large as possible,
+				   strategy_parameter is N/A */
+	PCI_DMA_BURST_BOUNDARY, /* disconnect at every strategy_parameter
+				   byte boundaries */
+	PCI_DMA_BURST_MULTIPLE, /* disconnect at some multiple of
+				   strategy_parameter byte boundaries */
+};
+
 #if defined(CONFIG_ISA) || defined(CONFIG_EISA)
 extern struct pci_dev *isa_bridge;
 #endif
-- 
cgit v1.2.3-70-g09d2


From bb4a61b6eaee01707f24deeefc5d7136f25f75c5 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 6 Jun 2005 23:07:46 -0700
Subject: [PATCH] PCI: fix up errors after dma bursting patch and CONFIG_PCI=n

With CONFIG_PCI=n:

In file included from include/linux/pci.h:917,
                 from lib/iomap.c:6:
include/asm/pci.h:104: warning: `enum pci_dma_burst_strategy' declared inside parameter list
include/asm/pci.h:104: warning: its scope is only this definition or declaration, which is probably not what you want.
include/asm/pci.h: In function `pci_dma_burst_advice':
include/asm/pci.h:106: dereferencing pointer to incomplete type
include/asm/pci.h:106: `PCI_DMA_BURST_INFINITY' undeclared (first use in this function)
include/asm/pci.h:106: (Each undeclared identifier is reported only once
include/asm/pci.h:106: for each function it appears in.)
make[1]: *** [lib/iomap.o] Error 1

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/asm-alpha/pci.h   | 2 ++
 include/asm-arm/pci.h     | 2 ++
 include/asm-frv/pci.h     | 2 ++
 include/asm-i386/pci.h    | 2 ++
 include/asm-ia64/pci.h    | 2 ++
 include/asm-mips/pci.h    | 2 ++
 include/asm-parisc/pci.h  | 2 ++
 include/asm-ppc/pci.h     | 2 ++
 include/asm-ppc64/pci.h   | 2 ++
 include/asm-sh/pci.h      | 2 ++
 include/asm-sh64/pci.h    | 2 ++
 include/asm-sparc/pci.h   | 2 ++
 include/asm-sparc64/pci.h | 2 ++
 include/asm-v850/pci.h    | 2 ++
 include/asm-x86_64/pci.h  | 2 ++
 include/linux/pci.h       | 2 ++
 16 files changed, 32 insertions(+)

(limited to 'include')

diff --git a/include/asm-alpha/pci.h b/include/asm-alpha/pci.h
index 6c71dc1ad4c..b7806aa3785 100644
--- a/include/asm-alpha/pci.h
+++ b/include/asm-alpha/pci.h
@@ -223,6 +223,7 @@ pci_dac_dma_sync_single_for_device(struct pci_dev *pdev, dma64_addr_t dma_addr,
 	/* Nothing to do. */
 }
 
+#ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -239,6 +240,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_BOUNDARY;
 	*strategy_parameter = cacheline_size;
 }
+#endif
 
 /* TODO: integrate with include/asm-generic/pci.h ? */
 static inline int pci_get_legacy_ide_irq(struct pci_dev *dev, int channel)
diff --git a/include/asm-arm/pci.h b/include/asm-arm/pci.h
index bc2ec425aca..e300646fe65 100644
--- a/include/asm-arm/pci.h
+++ b/include/asm-arm/pci.h
@@ -42,6 +42,7 @@ static inline void pcibios_penalize_isa_irq(int irq)
 #define pci_unmap_len(PTR, LEN_NAME)		((PTR)->LEN_NAME)
 #define pci_unmap_len_set(PTR, LEN_NAME, VAL)	(((PTR)->LEN_NAME) = (VAL))
 
+#ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -49,6 +50,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_INFINITY;
 	*strategy_parameter = ~0UL;
 }
+#endif
 
 #define HAVE_PCI_MMAP
 extern int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
diff --git a/include/asm-frv/pci.h b/include/asm-frv/pci.h
index 13427240664..b4efe5e3591 100644
--- a/include/asm-frv/pci.h
+++ b/include/asm-frv/pci.h
@@ -57,6 +57,7 @@ extern void pci_free_consistent(struct pci_dev *hwdev, size_t size,
  */
 #define PCI_DMA_BUS_IS_PHYS	(1)
 
+#ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -64,6 +65,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_INFINITY;
 	*strategy_parameter = ~0UL;
 }
+#endif
 
 /*
  *	These are pretty much arbitary with the CoMEM implementation.
diff --git a/include/asm-i386/pci.h b/include/asm-i386/pci.h
index bf07b3af85e..3561899eb82 100644
--- a/include/asm-i386/pci.h
+++ b/include/asm-i386/pci.h
@@ -99,6 +99,7 @@ static inline void pcibios_add_platform_entries(struct pci_dev *dev)
 {
 }
 
+#ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -106,6 +107,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_INFINITY;
 	*strategy_parameter = ~0UL;
 }
+#endif
 
 #endif /* __KERNEL__ */
 
diff --git a/include/asm-ia64/pci.h b/include/asm-ia64/pci.h
index c9f1ab4e477..0c4c5d801d3 100644
--- a/include/asm-ia64/pci.h
+++ b/include/asm-ia64/pci.h
@@ -82,6 +82,7 @@ extern int pcibios_prep_mwi (struct pci_dev *);
 #define sg_dma_len(sg)		((sg)->dma_length)
 #define sg_dma_address(sg)	((sg)->dma_address)
 
+#ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -98,6 +99,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_MULTIPLE;
 	*strategy_parameter = cacheline_size;
 }
+#endif
 
 #define HAVE_PCI_MMAP
 extern int pci_mmap_page_range (struct pci_dev *dev, struct vm_area_struct *vma,
diff --git a/include/asm-mips/pci.h b/include/asm-mips/pci.h
index 20b93bfa456..2d323b6e147 100644
--- a/include/asm-mips/pci.h
+++ b/include/asm-mips/pci.h
@@ -130,6 +130,7 @@ extern void pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev,
 extern void pci_dac_dma_sync_single_for_device(struct pci_dev *pdev,
 	dma64_addr_t dma_addr, size_t len, int direction);
 
+#ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -137,6 +138,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_INFINITY;
 	*strategy_parameter = ~0UL;
 }
+#endif
 
 extern void pcibios_resource_to_bus(struct pci_dev *dev,
 	struct pci_bus_region *region, struct resource *res);
diff --git a/include/asm-parisc/pci.h b/include/asm-parisc/pci.h
index f9f5bf90111..ee741c15017 100644
--- a/include/asm-parisc/pci.h
+++ b/include/asm-parisc/pci.h
@@ -230,6 +230,7 @@ extern inline void pcibios_register_hba(struct pci_hba_data *x)
 /* export the pci_ DMA API in terms of the dma_ one */
 #include <asm-generic/pci-dma-compat.h>
 
+#ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -246,6 +247,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_MULTIPLE;
 	*strategy_parameter = cacheline_size;
 }
+#endif
 
 extern void
 pcibios_resource_to_bus(struct pci_dev *dev, struct pci_bus_region *region,
diff --git a/include/asm-ppc/pci.h b/include/asm-ppc/pci.h
index 669e9de7a52..db0a2a0ec74 100644
--- a/include/asm-ppc/pci.h
+++ b/include/asm-ppc/pci.h
@@ -69,6 +69,7 @@ extern unsigned long pci_bus_to_phys(unsigned int ba, int busnr);
 #define pci_unmap_len(PTR, LEN_NAME)		(0)
 #define pci_unmap_len_set(PTR, LEN_NAME, VAL)	do { } while (0)
 
+#ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -76,6 +77,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_INFINITY;
 	*strategy_parameter = ~0UL;
 }
+#endif
 
 /*
  * At present there are very few 32-bit PPC machines that can have
diff --git a/include/asm-ppc64/pci.h b/include/asm-ppc64/pci.h
index 20beb10c090..d12dfce21e2 100644
--- a/include/asm-ppc64/pci.h
+++ b/include/asm-ppc64/pci.h
@@ -78,6 +78,7 @@ static inline int pci_dac_dma_supported(struct pci_dev *hwdev,u64 mask)
 	return 0;
 }
 
+#ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -94,6 +95,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_MULTIPLE;
 	*strategy_parameter = cacheline_size;
 }
+#endif
 
 extern int pci_domain_nr(struct pci_bus *bus);
 
diff --git a/include/asm-sh/pci.h b/include/asm-sh/pci.h
index 7237bc6a728..26044889c77 100644
--- a/include/asm-sh/pci.h
+++ b/include/asm-sh/pci.h
@@ -96,6 +96,7 @@ static inline void pcibios_penalize_isa_irq(int irq)
 #define sg_dma_address(sg)	(virt_to_bus((sg)->dma_address))
 #define sg_dma_len(sg)		((sg)->length)
 
+#ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -103,6 +104,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_INFINITY;
 	*strategy_parameter = ~0UL;
 }
+#endif
 
 /* Board-specific fixup routines. */
 extern void pcibios_fixup(void);
diff --git a/include/asm-sh64/pci.h b/include/asm-sh64/pci.h
index 0ac15ab01cc..c68870e02d9 100644
--- a/include/asm-sh64/pci.h
+++ b/include/asm-sh64/pci.h
@@ -86,6 +86,7 @@ static inline void pcibios_penalize_isa_irq(int irq)
 #define sg_dma_address(sg)	((sg)->dma_address)
 #define sg_dma_len(sg)		((sg)->length)
 
+#ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -93,6 +94,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_INFINITY;
 	*strategy_parameter = ~0UL;
 }
+#endif
 
 /* Board-specific fixup routines. */
 extern void pcibios_fixup(void);
diff --git a/include/asm-sparc/pci.h b/include/asm-sparc/pci.h
index 2fd65db95e9..44bb38758c9 100644
--- a/include/asm-sparc/pci.h
+++ b/include/asm-sparc/pci.h
@@ -144,6 +144,7 @@ extern inline int pci_dma_supported(struct pci_dev *hwdev, u64 mask)
 
 #define pci_dac_dma_supported(dev, mask)	(0)
 
+#ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -151,6 +152,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_INFINITY;
 	*strategy_parameter = ~0UL;
 }
+#endif
 
 static inline void pcibios_add_platform_entries(struct pci_dev *dev)
 {
diff --git a/include/asm-sparc64/pci.h b/include/asm-sparc64/pci.h
index 402667300d0..84e41c1ef3f 100644
--- a/include/asm-sparc64/pci.h
+++ b/include/asm-sparc64/pci.h
@@ -220,6 +220,7 @@ static inline int pci_dma_mapping_error(dma_addr_t dma_addr)
 	return (dma_addr == PCI_DMA_ERROR_CODE);
 }
 
+#ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -236,6 +237,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_BOUNDARY;
 	*strategy_parameter = cacheline_size;
 }
+#endif
 
 /* Return the index of the PCI controller for device PDEV. */
 
diff --git a/include/asm-v850/pci.h b/include/asm-v850/pci.h
index d26eb8d6731..8e79be0fe99 100644
--- a/include/asm-v850/pci.h
+++ b/include/asm-v850/pci.h
@@ -81,6 +81,7 @@ extern void
 pci_free_consistent (struct pci_dev *pdev, size_t size, void *cpu_addr,
 		     dma_addr_t dma_addr);
 
+#ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -88,6 +89,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_INFINITY;
 	*strategy_parameter = ~0UL;
 }
+#endif
 
 static inline void pcibios_add_platform_entries(struct pci_dev *dev)
 {
diff --git a/include/asm-x86_64/pci.h b/include/asm-x86_64/pci.h
index 8461d6af102..c1961db88fa 100644
--- a/include/asm-x86_64/pci.h
+++ b/include/asm-x86_64/pci.h
@@ -123,6 +123,7 @@ pci_dac_dma_sync_single_for_device(struct pci_dev *pdev, dma64_addr_t dma_addr,
 	flush_write_buffers();
 }
 
+#ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -130,6 +131,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_INFINITY;
 	*strategy_parameter = ~0UL;
 }
+#endif
 
 #define HAVE_PCI_MMAP
 extern int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 9ce4f1be093..66798b46f30 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -985,6 +985,8 @@ static inline int pci_proc_domain(struct pci_bus *bus)
 }
 #endif
 
+#define pci_dma_burst_advice(pdev, strat, strategy_parameter) do { } while (0)
+
 #endif /* !CONFIG_PCI */
 
 /* these helpers provide future and backwards compatibility
-- 
cgit v1.2.3-70-g09d2


From 545493917dc90298e1c38f018ad893f5518928e7 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 23 Jun 2005 17:35:56 -0700
Subject: [PATCH] PCI: add proper MCFG table parsing to ACPI core.

This patch is the first step in properly handling the MCFG PCI table.
It defines the structures properly, and saves off the table so that the
pci mmconfig code can access it.  It moves the parsing of the table a
little later in the boot process, but still before the information is
needed.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/i386/kernel/acpi/boot.c | 41 +++++++++++++++++++++++++++++++++--------
 arch/i386/pci/mmconfig.c     | 12 +++++++-----
 arch/x86_64/pci/mmconfig.c   | 16 +++++++++-------
 include/linux/acpi.h         | 16 +++++++++++++---
 4 files changed, 62 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index f5360d88bcf..b7808a89d94 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -159,9 +159,15 @@ char *__acpi_map_table(unsigned long phys, unsigned long size)
 #endif
 
 #ifdef CONFIG_PCI_MMCONFIG
-static int __init acpi_parse_mcfg(unsigned long phys_addr, unsigned long size)
+/* The physical address of the MMCONFIG aperture.  Set from ACPI tables. */
+struct acpi_table_mcfg_config *pci_mmcfg_config;
+int pci_mmcfg_config_num;
+
+int __init acpi_parse_mcfg(unsigned long phys_addr, unsigned long size)
 {
 	struct acpi_table_mcfg *mcfg;
+	unsigned long i;
+	int config_size;
 
 	if (!phys_addr || !size)
 		return -EINVAL;
@@ -172,18 +178,38 @@ static int __init acpi_parse_mcfg(unsigned long phys_addr, unsigned long size)
 		return -ENODEV;
 	}
 
-	if (mcfg->base_reserved) {
-		printk(KERN_ERR PREFIX "MMCONFIG not in low 4GB of memory\n");
+	/* how many config structures do we have */
+	pci_mmcfg_config_num = 0;
+	i = size - sizeof(struct acpi_table_mcfg);
+	while (i >= sizeof(struct acpi_table_mcfg_config)) {
+		++pci_mmcfg_config_num;
+		i -= sizeof(struct acpi_table_mcfg_config);
+	};
+	if (pci_mmcfg_config_num == 0) {
+		printk(KERN_ERR PREFIX "MMCONFIG has no entries\n");
 		return -ENODEV;
 	}
 
-	pci_mmcfg_base_addr = mcfg->base_address;
+	config_size = pci_mmcfg_config_num * sizeof(*pci_mmcfg_config);
+	pci_mmcfg_config = kmalloc(config_size, GFP_KERNEL);
+	if (!pci_mmcfg_config) {
+		printk(KERN_WARNING PREFIX
+		       "No memory for MCFG config tables\n");
+		return -ENOMEM;
+	}
+
+	memcpy(pci_mmcfg_config, &mcfg->config, config_size);
+	for (i = 0; i < pci_mmcfg_config_num; ++i) {
+		if (mcfg->config[i].base_reserved) {
+			printk(KERN_ERR PREFIX
+			       "MMCONFIG not in low 4GB of memory\n");
+			return -ENODEV;
+		}
+	}
 
 	return 0;
 }
-#else
-#define	acpi_parse_mcfg NULL
-#endif /* !CONFIG_PCI_MMCONFIG */
+#endif /* CONFIG_PCI_MMCONFIG */
 
 #ifdef CONFIG_X86_LOCAL_APIC
 static int __init
@@ -1139,7 +1165,6 @@ int __init acpi_boot_init(void)
 	acpi_process_madt();
 
 	acpi_table_parse(ACPI_HPET, acpi_parse_hpet);
-	acpi_table_parse(ACPI_MCFG, acpi_parse_mcfg);
 
 	return 0;
 }
diff --git a/arch/i386/pci/mmconfig.c b/arch/i386/pci/mmconfig.c
index 021a50aa51f..5fbaa913225 100644
--- a/arch/i386/pci/mmconfig.c
+++ b/arch/i386/pci/mmconfig.c
@@ -11,11 +11,9 @@
 
 #include <linux/pci.h>
 #include <linux/init.h>
+#include <linux/acpi.h>
 #include "pci.h"
 
-/* The physical address of the MMCONFIG aperture.  Set from ACPI tables. */
-u32 pci_mmcfg_base_addr;
-
 #define mmcfg_virt_addr ((void __iomem *) fix_to_virt(FIX_PCIE_MCFG))
 
 /* The base address of the last MMCONFIG device accessed */
@@ -27,7 +25,7 @@ static u32 mmcfg_last_accessed_device;
 
 static inline void pci_exp_set_dev_base(int bus, int devfn)
 {
-	u32 dev_base = pci_mmcfg_base_addr | (bus << 20) | (devfn << 12);
+	u32 dev_base = pci_mmcfg_config[0].base_address | (bus << 20) | (devfn << 12);
 	if (dev_base != mmcfg_last_accessed_device) {
 		mmcfg_last_accessed_device = dev_base;
 		set_fixmap_nocache(FIX_PCIE_MCFG, dev_base);
@@ -101,7 +99,11 @@ static int __init pci_mmcfg_init(void)
 {
 	if ((pci_probe & PCI_PROBE_MMCONF) == 0)
 		goto out;
-	if (!pci_mmcfg_base_addr)
+
+	acpi_table_parse(ACPI_MCFG, acpi_parse_mcfg);
+	if ((pci_mmcfg_config_num == 0) ||
+	    (pci_mmcfg_config == NULL) ||
+	    (pci_mmcfg_config[0].base_address == 0))
 		goto out;
 
 	/* Kludge for now. Don't use mmconfig on AMD systems because
diff --git a/arch/x86_64/pci/mmconfig.c b/arch/x86_64/pci/mmconfig.c
index b693c232fd0..09cfcc1234b 100644
--- a/arch/x86_64/pci/mmconfig.c
+++ b/arch/x86_64/pci/mmconfig.c
@@ -7,15 +7,13 @@
 
 #include <linux/pci.h>
 #include <linux/init.h>
+#include <linux/acpi.h>
 #include "pci.h"
 
 #define MMCONFIG_APER_SIZE (256*1024*1024)
 
-/* The physical address of the MMCONFIG aperture.  Set from ACPI tables. */
-u32 pci_mmcfg_base_addr;
-
 /* Static virtual mapping of the MMCONFIG aperture */
-char *pci_mmcfg_virt;
+static char *pci_mmcfg_virt;
 
 static inline char *pci_dev_base(unsigned int bus, unsigned int devfn)
 {
@@ -77,7 +75,11 @@ static int __init pci_mmcfg_init(void)
 {
 	if ((pci_probe & PCI_PROBE_MMCONF) == 0)
 		return 0;
-	if (!pci_mmcfg_base_addr)
+
+	acpi_table_parse(ACPI_MCFG, acpi_parse_mcfg);
+	if ((pci_mmcfg_config_num == 0) ||
+	    (pci_mmcfg_config == NULL) ||
+	    (pci_mmcfg_config[0].base_address == 0))
 		return 0;
 
 	/* Kludge for now. Don't use mmconfig on AMD systems because
@@ -88,13 +90,13 @@ static int __init pci_mmcfg_init(void)
 		return 0; 
 
 	/* RED-PEN i386 doesn't do _nocache right now */
-	pci_mmcfg_virt = ioremap_nocache(pci_mmcfg_base_addr, MMCONFIG_APER_SIZE);
+	pci_mmcfg_virt = ioremap_nocache(pci_mmcfg_config[0].base_address, MMCONFIG_APER_SIZE);
 	if (!pci_mmcfg_virt) { 
 		printk("PCI: Cannot map mmconfig aperture\n");
 		return 0;
 	}	
 
-	printk(KERN_INFO "PCI: Using MMCONFIG at %x\n", pci_mmcfg_base_addr);
+	printk(KERN_INFO "PCI: Using MMCONFIG at %x\n", pci_mmcfg_config[0].base_address);
 	raw_pci_ops = &pci_mmcfg;
 	pci_probe = (pci_probe & ~PCI_PROBE_MASK) | PCI_PROBE_MMCONF;
 
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index f5bc298707e..ef8483673aa 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -342,11 +342,19 @@ struct acpi_table_ecdt {
 
 /* PCI MMCONFIG */
 
+/* Defined in PCI Firmware Specification 3.0 */
+struct acpi_table_mcfg_config {
+	u32				base_address;
+	u32				base_reserved;
+	u16				pci_segment_group_number;
+	u8				start_bus_number;
+	u8				end_bus_number;
+	u8				reserved[4];
+} __attribute__ ((packed));
 struct acpi_table_mcfg {
 	struct acpi_table_header	header;
 	u8				reserved[8];
-	u32				base_address;
-	u32				base_reserved;
+	struct acpi_table_mcfg_config	config[0];
 } __attribute__ ((packed));
 
 /* Table Handlers */
@@ -391,6 +399,7 @@ int acpi_table_parse (enum acpi_table_id id, acpi_table_handler handler);
 int acpi_get_table_header_early (enum acpi_table_id id, struct acpi_table_header **header);
 int acpi_table_parse_madt (enum acpi_madt_entry_id id, acpi_madt_entry_handler handler, unsigned int max_entries);
 int acpi_table_parse_srat (enum acpi_srat_entry_id id, acpi_madt_entry_handler handler, unsigned int max_entries);
+int acpi_parse_mcfg (unsigned long phys_addr, unsigned long size);
 void acpi_table_print (struct acpi_table_header *header, unsigned long phys_addr);
 void acpi_table_print_madt_entry (acpi_table_entry_header *madt);
 void acpi_table_print_srat_entry (acpi_table_entry_header *srat);
@@ -412,7 +421,8 @@ int acpi_unregister_ioapic(acpi_handle handle, u32 gsi_base);
 
 extern int acpi_mp_config;
 
-extern u32 pci_mmcfg_base_addr;
+extern struct acpi_table_mcfg_config *pci_mmcfg_config;
+extern int pci_mmcfg_config_num;
 
 extern int sbf_port ;
 
-- 
cgit v1.2.3-70-g09d2


From 66b7f8a30437b8639e798f7db8e9be1da5711efa Mon Sep 17 00:00:00 2001
From: Mark Maule <maule@sgi.com>
Date: Mon, 25 Apr 2005 13:51:00 -0700
Subject: [IA64-SGI] pcdp: add PCDP pci interface support

Resend 2 with changes per Bjorn Helgaas comments.  Changes from original:

+ Change globals to vga_console_iobase/vga_console_membase and make them
  unconditional.
+ Address style-related comments.

Patch to extend the PCDP vga setup code to support PCI io/mem translations
for the legacy vga ioport and ram spaces on architectures (e.g. altix) which
need them.

Summary of the changes:

drivers/firmware/pcdp.c
drivers/firmware/pcdp.h
-----------------------
+ add declaration for the spec-defined PCI interface struct (pcdp_if_pci)
  as well as support macros.

+ extend setup_vga_console() to know about pcdp_if_pci and add a couple of
  globals to hold the io and mem translation offsets if present.

arch/ia64/kernel/setup.c
------------------------
+ tweek early_console_setup() to allow multiple early console setup routines
  to be called.

include/asm-ia64/vga.h
----------------------
+ make VGA_MAP_MEM vga_console_membase aware

Signed-off-by: Mark Maule <maule@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/setup.c | 12 ++++++++----
 drivers/firmware/pcdp.c  | 24 +++++++++++++++++++++---
 drivers/firmware/pcdp.h  | 33 +++++++++++++++++++++++++++++----
 include/asm-ia64/vga.h   |  5 ++++-
 4 files changed, 62 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index d14692e0920..2693e1522d7 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -72,6 +72,8 @@ DEFINE_PER_CPU(unsigned long, ia64_phys_stacked_size_p8);
 unsigned long ia64_cycles_per_usec;
 struct ia64_boot_param *ia64_boot_param;
 struct screen_info screen_info;
+unsigned long vga_console_iobase;
+unsigned long vga_console_membase;
 
 unsigned long ia64_max_cacheline_size;
 unsigned long ia64_iobase;	/* virtual address for I/O accesses */
@@ -273,23 +275,25 @@ io_port_init (void)
 static inline int __init
 early_console_setup (char *cmdline)
 {
+	int earlycons = 0;
+
 #ifdef CONFIG_SERIAL_SGI_L1_CONSOLE
 	{
 		extern int sn_serial_console_early_setup(void);
 		if (!sn_serial_console_early_setup())
-			return 0;
+			earlycons++;
 	}
 #endif
 #ifdef CONFIG_EFI_PCDP
 	if (!efi_setup_pcdp_console(cmdline))
-		return 0;
+		earlycons++;
 #endif
 #ifdef CONFIG_SERIAL_8250_CONSOLE
 	if (!early_serial_console_init(cmdline))
-		return 0;
+		earlycons++;
 #endif
 
-	return -1;
+	return (earlycons) ? 0 : -1;
 }
 
 static inline void
diff --git a/drivers/firmware/pcdp.c b/drivers/firmware/pcdp.c
index 839b44a7e08..53c95c0bbf4 100644
--- a/drivers/firmware/pcdp.c
+++ b/drivers/firmware/pcdp.c
@@ -16,6 +16,7 @@
 #include <linux/console.h>
 #include <linux/efi.h>
 #include <linux/serial.h>
+#include <asm/vga.h>
 #include "pcdp.h"
 
 static int __init
@@ -40,10 +41,27 @@ setup_serial_console(struct pcdp_uart *uart)
 }
 
 static int __init
-setup_vga_console(struct pcdp_vga *vga)
+setup_vga_console(struct pcdp_device *dev)
 {
 #if defined(CONFIG_VT) && defined(CONFIG_VGA_CONSOLE)
-	if (efi_mem_type(0xA0000) == EFI_CONVENTIONAL_MEMORY) {
+	u8 *if_ptr;
+
+	if_ptr = ((u8 *)dev + sizeof(struct pcdp_device));
+	if (if_ptr[0] == PCDP_IF_PCI) {
+		struct pcdp_if_pci if_pci;
+
+		/* struct copy since ifptr might not be correctly aligned */
+
+		memcpy(&if_pci, if_ptr, sizeof(if_pci));
+
+		if (if_pci.trans & PCDP_PCI_TRANS_IOPORT)
+			vga_console_iobase = if_pci.ioport_tra;
+
+		if (if_pci.trans & PCDP_PCI_TRANS_MMIO)
+			vga_console_membase = if_pci.mmio_tra;
+	}
+
+	if (efi_mem_type(vga_console_membase + 0xA0000) == EFI_CONVENTIONAL_MEMORY) {
 		printk(KERN_ERR "PCDP: VGA selected, but frame buffer is not MMIO!\n");
 		return -ENODEV;
 	}
@@ -95,7 +113,7 @@ efi_setup_pcdp_console(char *cmdline)
 	     dev = (struct pcdp_device *) ((u8 *) dev + dev->length)) {
 		if (dev->flags & PCDP_PRIMARY_CONSOLE) {
 			if (dev->type == PCDP_CONSOLE_VGA) {
-				return setup_vga_console((struct pcdp_vga *) dev);
+				return setup_vga_console(dev);
 			}
 		}
 	}
diff --git a/drivers/firmware/pcdp.h b/drivers/firmware/pcdp.h
index 1dc7c88b7b4..e72cc47de33 100644
--- a/drivers/firmware/pcdp.h
+++ b/drivers/firmware/pcdp.h
@@ -52,11 +52,34 @@ struct pcdp_uart {
 	u32				clock_rate;
 	u8				pci_prog_intfc;
 	u8				flags;
-};
+} __attribute__((packed));
+
+#define PCDP_IF_PCI	1
+
+/* pcdp_if_pci.trans */
+#define PCDP_PCI_TRANS_IOPORT	0x02
+#define PCDP_PCI_TRANS_MMIO	0x01
+
+struct pcdp_if_pci {
+	u8			interconnect;
+	u8			reserved;
+	u16			length;
+	u8			segment;
+	u8			bus;
+	u8			dev;
+	u8			fun;
+	u16			dev_id;
+	u16			vendor_id;
+	u32			acpi_interrupt;
+	u64			mmio_tra;
+	u64			ioport_tra;
+	u8			flags;
+	u8			trans;
+} __attribute__((packed));
 
 struct pcdp_vga {
 	u8			count;		/* address space descriptors */
-};
+} __attribute__((packed));
 
 /* pcdp_device.flags */
 #define PCDP_PRIMARY_CONSOLE	1
@@ -66,7 +89,9 @@ struct pcdp_device {
 	u8			flags;
 	u16			length;
 	u16			efi_index;
-};
+	/* next data is pcdp_if_pci or pcdp_if_acpi (not yet supported) */
+	/* next data is device specific type (currently only pcdp_vga) */
+} __attribute__((packed));
 
 struct pcdp {
 	u8			signature[4];
@@ -81,4 +106,4 @@ struct pcdp {
 	u32			num_uarts;
 	struct pcdp_uart	uart[0];	/* actual size is num_uarts */
 	/* remainder of table is pcdp_device structures */
-};
+} __attribute__((packed));
diff --git a/include/asm-ia64/vga.h b/include/asm-ia64/vga.h
index 1f446d6841f..bc3349ffc50 100644
--- a/include/asm-ia64/vga.h
+++ b/include/asm-ia64/vga.h
@@ -14,7 +14,10 @@
  * videoram directly without any black magic.
  */
 
-#define VGA_MAP_MEM(x)	((unsigned long) ioremap((x), 0))
+extern unsigned long vga_console_iobase;
+extern unsigned long vga_console_membase;
+
+#define VGA_MAP_MEM(x)	((unsigned long) ioremap(vga_console_membase + (x), 0))
 
 #define vga_readb(x)	(*(x))
 #define vga_writeb(x,y)	(*(y) = (x))
-- 
cgit v1.2.3-70-g09d2


From 8e4641b3ee8073a46995cf97e0875df25b88d776 Mon Sep 17 00:00:00 2001
From: Prarit Bhargava <prarit@sgi.com>
Date: Mon, 9 May 2005 05:07:00 -0700
Subject: [IA64] sparse cleanup of shub_mmr.h

This patch is a sparse compile cleanup of shub_mmr.h using both the defconfig
and the sn2_defconfig config files.

The issue with this file was the missing usage of __IA64_UL_CONST wrapper.
This wrapper is defined in include/asm-ia64/types.h and wraps a long
constant definition with UL or with nothing depending on its usage in the
kernel.  The missing wrapper caused many sparse compile errors like

        warning: constant 0x0x0000000010000380 so big it is long

Signed-off-by: Prarit Bhargava <prarit@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/kernel/sn2/ptc_deadlock.S |   1 +
 include/asm-ia64/sn/shub_mmr.h         | 346 +++++++++++++++++----------------
 2 files changed, 175 insertions(+), 172 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/sn/kernel/sn2/ptc_deadlock.S b/arch/ia64/sn/kernel/sn2/ptc_deadlock.S
index 7947312801e..96cb71d1568 100644
--- a/arch/ia64/sn/kernel/sn2/ptc_deadlock.S
+++ b/arch/ia64/sn/kernel/sn2/ptc_deadlock.S
@@ -6,6 +6,7 @@
  * Copyright (C) 2000-2004 Silicon Graphics, Inc. All rights reserved.
  */
 
+#include <asm/types.h>
 #include <asm/sn/shub_mmr.h>
 
 #define DEADLOCKBIT	SH_PIO_WRITE_STATUS_WRITE_DEADLOCK_SHFT
diff --git a/include/asm-ia64/sn/shub_mmr.h b/include/asm-ia64/sn/shub_mmr.h
index 323fa0cd8d8..7de1d1d4b71 100644
--- a/include/asm-ia64/sn/shub_mmr.h
+++ b/include/asm-ia64/sn/shub_mmr.h
@@ -14,96 +14,98 @@
 /*                        Register "SH_IPI_INT"                         */
 /*               SHub Inter-Processor Interrupt Registers               */
 /* ==================================================================== */
-#define SH1_IPI_INT                               0x0000000110000380
-#define SH2_IPI_INT                               0x0000000010000380
+#define SH1_IPI_INT			__IA64_UL_CONST(0x0000000110000380)
+#define SH2_IPI_INT			__IA64_UL_CONST(0x0000000010000380)
 
 /*   SH_IPI_INT_TYPE                                                    */
 /*   Description:  Type of Interrupt: 0=INT, 2=PMI, 4=NMI, 5=INIT       */
-#define SH_IPI_INT_TYPE_SHFT                     0
-#define SH_IPI_INT_TYPE_MASK                     0x0000000000000007
+#define SH_IPI_INT_TYPE_SHFT				0
+#define SH_IPI_INT_TYPE_MASK		__IA64_UL_CONST(0x0000000000000007)
 
 /*   SH_IPI_INT_AGT                                                     */
 /*   Description:  Agent, must be 0 for SHub                            */
-#define SH_IPI_INT_AGT_SHFT                      3
-#define SH_IPI_INT_AGT_MASK                      0x0000000000000008
+#define SH_IPI_INT_AGT_SHFT				3
+#define SH_IPI_INT_AGT_MASK		__IA64_UL_CONST(0x0000000000000008)
 
 /*   SH_IPI_INT_PID                                                     */
 /*   Description:  Processor ID, same setting as on targeted McKinley  */
-#define SH_IPI_INT_PID_SHFT                      4
-#define SH_IPI_INT_PID_MASK                      0x00000000000ffff0
+#define SH_IPI_INT_PID_SHFT                      	4
+#define SH_IPI_INT_PID_MASK		__IA64_UL_CONST(0x00000000000ffff0)
 
 /*   SH_IPI_INT_BASE                                                    */
 /*   Description:  Optional interrupt vector area, 2MB aligned          */
-#define SH_IPI_INT_BASE_SHFT                     21
-#define SH_IPI_INT_BASE_MASK                     0x0003ffffffe00000
+#define SH_IPI_INT_BASE_SHFT				21
+#define SH_IPI_INT_BASE_MASK 		__IA64_UL_CONST(0x0003ffffffe00000)
 
 /*   SH_IPI_INT_IDX                                                     */
 /*   Description:  Targeted McKinley interrupt vector                   */
-#define SH_IPI_INT_IDX_SHFT                      52
-#define SH_IPI_INT_IDX_MASK                      0x0ff0000000000000
+#define SH_IPI_INT_IDX_SHFT				52
+#define SH_IPI_INT_IDX_MASK		__IA64_UL_CONST(0x0ff0000000000000)
 
 /*   SH_IPI_INT_SEND                                                    */
 /*   Description:  Send Interrupt Message to PI, This generates a puls  */
-#define SH_IPI_INT_SEND_SHFT                     63
-#define SH_IPI_INT_SEND_MASK                     0x8000000000000000
+#define SH_IPI_INT_SEND_SHFT				63
+#define SH_IPI_INT_SEND_MASK		__IA64_UL_CONST(0x8000000000000000)
 
 /* ==================================================================== */
 /*                     Register "SH_EVENT_OCCURRED"                     */
 /*                    SHub Interrupt Event Occurred                     */
 /* ==================================================================== */
-#define SH1_EVENT_OCCURRED                        0x0000000110010000
-#define SH1_EVENT_OCCURRED_ALIAS                  0x0000000110010008
-#define SH2_EVENT_OCCURRED                        0x0000000010010000
-#define SH2_EVENT_OCCURRED_ALIAS                  0x0000000010010008
+#define SH1_EVENT_OCCURRED		__IA64_UL_CONST(0x0000000110010000)
+#define SH1_EVENT_OCCURRED_ALIAS	__IA64_UL_CONST(0x0000000110010008)
+#define SH2_EVENT_OCCURRED		__IA64_UL_CONST(0x0000000010010000)
+#define SH2_EVENT_OCCURRED_ALIAS 	__IA64_UL_CONST(0x0000000010010008)
 
 /* ==================================================================== */
 /*                     Register "SH_PI_CAM_CONTROL"                     */
 /*                      CRB CAM MMR Access Control                      */
 /* ==================================================================== */
-#define SH1_PI_CAM_CONTROL                        0x0000000120050300
+#define SH1_PI_CAM_CONTROL		__IA64_UL_CONST(0x0000000120050300)
 
 /* ==================================================================== */
 /*                        Register "SH_SHUB_ID"                         */
 /*                            SHub ID Number                            */
 /* ==================================================================== */
-#define SH1_SHUB_ID                               0x0000000110060580
-#define SH1_SHUB_ID_REVISION_SHFT                 28
-#define SH1_SHUB_ID_REVISION_MASK                 0x00000000f0000000
+#define SH1_SHUB_ID			__IA64_UL_CONST(0x0000000110060580)
+#define SH1_SHUB_ID_REVISION_SHFT			28
+#define SH1_SHUB_ID_REVISION_MASK	__IA64_UL_CONST(0x00000000f0000000)
 
 /* ==================================================================== */
 /*                          Register "SH_RTC"                           */
 /*                           Real-time Clock                            */
 /* ==================================================================== */
-#define SH1_RTC                                   0x00000001101c0000
-#define SH2_RTC					  0x00000002101c0000
-#define SH_RTC_MASK                               0x007fffffffffffff
+#define SH1_RTC				__IA64_UL_CONST(0x00000001101c0000)
+#define SH2_RTC				__IA64_UL_CONST(0x00000002101c0000)
+#define SH_RTC_MASK			__IA64_UL_CONST(0x007fffffffffffff)
 
 /* ==================================================================== */
 /*                   Register "SH_PIO_WRITE_STATUS_0|1"                 */
 /*                      PIO Write Status for CPU 0 & 1                  */
 /* ==================================================================== */
-#define SH1_PIO_WRITE_STATUS_0                    0x0000000120070200
-#define SH1_PIO_WRITE_STATUS_1                    0x0000000120070280
-#define SH2_PIO_WRITE_STATUS_0                    0x0000000020070200
-#define SH2_PIO_WRITE_STATUS_1                    0x0000000020070280
-#define SH2_PIO_WRITE_STATUS_2                    0x0000000020070300
-#define SH2_PIO_WRITE_STATUS_3                    0x0000000020070380
+#define SH1_PIO_WRITE_STATUS_0		__IA64_UL_CONST(0x0000000120070200)
+#define SH1_PIO_WRITE_STATUS_1		__IA64_UL_CONST(0x0000000120070280)
+#define SH2_PIO_WRITE_STATUS_0		__IA64_UL_CONST(0x0000000020070200)
+#define SH2_PIO_WRITE_STATUS_1		__IA64_UL_CONST(0x0000000020070280)
+#define SH2_PIO_WRITE_STATUS_2		__IA64_UL_CONST(0x0000000020070300)
+#define SH2_PIO_WRITE_STATUS_3		__IA64_UL_CONST(0x0000000020070380)
 
 /*   SH_PIO_WRITE_STATUS_0_WRITE_DEADLOCK                               */
 /*   Description:  Deadlock response detected                           */
-#define SH_PIO_WRITE_STATUS_WRITE_DEADLOCK_SHFT 1
-#define SH_PIO_WRITE_STATUS_WRITE_DEADLOCK_MASK 0x0000000000000002
+#define SH_PIO_WRITE_STATUS_WRITE_DEADLOCK_SHFT		1
+#define SH_PIO_WRITE_STATUS_WRITE_DEADLOCK_MASK \
+					__IA64_UL_CONST(0x0000000000000002)
 
 /*   SH_PIO_WRITE_STATUS_0_PENDING_WRITE_COUNT                          */
 /*   Description:  Count of currently pending PIO writes                */
-#define SH_PIO_WRITE_STATUS_PENDING_WRITE_COUNT_SHFT 56
-#define SH_PIO_WRITE_STATUS_PENDING_WRITE_COUNT_MASK 0x3f00000000000000
+#define SH_PIO_WRITE_STATUS_PENDING_WRITE_COUNT_SHFT	56
+#define SH_PIO_WRITE_STATUS_PENDING_WRITE_COUNT_MASK \
+					__IA64_UL_CONST(0x3f00000000000000)
 
 /* ==================================================================== */
 /*                Register "SH_PIO_WRITE_STATUS_0_ALIAS"                */
 /* ==================================================================== */
-#define SH1_PIO_WRITE_STATUS_0_ALIAS              0x0000000120070208
-#define SH2_PIO_WRITE_STATUS_0_ALIAS              0x0000000020070208
+#define SH1_PIO_WRITE_STATUS_0_ALIAS	__IA64_UL_CONST(0x0000000120070208)
+#define SH2_PIO_WRITE_STATUS_0_ALIAS	__IA64_UL_CONST(0x0000000020070208)
 
 /* ==================================================================== */
 /*                     Register "SH_EVENT_OCCURRED"                     */
@@ -111,33 +113,33 @@
 /* ==================================================================== */
 /*   SH_EVENT_OCCURRED_UART_INT                                         */
 /*   Description:  Pending Junk Bus UART Interrupt                      */
-#define SH_EVENT_OCCURRED_UART_INT_SHFT          20
-#define SH_EVENT_OCCURRED_UART_INT_MASK          0x0000000000100000
+#define SH_EVENT_OCCURRED_UART_INT_SHFT			20
+#define SH_EVENT_OCCURRED_UART_INT_MASK	__IA64_UL_CONST(0x0000000000100000)
 
 /*   SH_EVENT_OCCURRED_IPI_INT                                          */
 /*   Description:  Pending IPI Interrupt                                */
-#define SH_EVENT_OCCURRED_IPI_INT_SHFT           28
-#define SH_EVENT_OCCURRED_IPI_INT_MASK           0x0000000010000000
+#define SH_EVENT_OCCURRED_IPI_INT_SHFT			28
+#define SH_EVENT_OCCURRED_IPI_INT_MASK	__IA64_UL_CONST(0x0000000010000000)
 
 /*   SH_EVENT_OCCURRED_II_INT0                                          */
 /*   Description:  Pending II 0 Interrupt                               */
-#define SH_EVENT_OCCURRED_II_INT0_SHFT           29
-#define SH_EVENT_OCCURRED_II_INT0_MASK           0x0000000020000000
+#define SH_EVENT_OCCURRED_II_INT0_SHFT			29
+#define SH_EVENT_OCCURRED_II_INT0_MASK	__IA64_UL_CONST(0x0000000020000000)
 
 /*   SH_EVENT_OCCURRED_II_INT1                                          */
 /*   Description:  Pending II 1 Interrupt                               */
-#define SH_EVENT_OCCURRED_II_INT1_SHFT           30
-#define SH_EVENT_OCCURRED_II_INT1_MASK           0x0000000040000000
+#define SH_EVENT_OCCURRED_II_INT1_SHFT			30
+#define SH_EVENT_OCCURRED_II_INT1_MASK	__IA64_UL_CONST(0x0000000040000000)
 
 /*   SH2_EVENT_OCCURRED_EXTIO_INT2                                      */
 /*   Description:  Pending SHUB 2 EXT IO INT2                           */
-#define SH2_EVENT_OCCURRED_EXTIO_INT2_SHFT       33
-#define SH2_EVENT_OCCURRED_EXTIO_INT2_MASK       0x0000000200000000
+#define SH2_EVENT_OCCURRED_EXTIO_INT2_SHFT		33
+#define SH2_EVENT_OCCURRED_EXTIO_INT2_MASK __IA64_UL_CONST(0x0000000200000000)
 
 /*   SH2_EVENT_OCCURRED_EXTIO_INT3                                      */
 /*   Description:  Pending SHUB 2 EXT IO INT3                           */
-#define SH2_EVENT_OCCURRED_EXTIO_INT3_SHFT       34
-#define SH2_EVENT_OCCURRED_EXTIO_INT3_MASK       0x0000000400000000
+#define SH2_EVENT_OCCURRED_EXTIO_INT3_SHFT		34
+#define SH2_EVENT_OCCURRED_EXTIO_INT3_MASK __IA64_UL_CONST(0x0000000400000000)
 
 #define SH_ALL_INT_MASK \
 	(SH_EVENT_OCCURRED_UART_INT_MASK | SH_EVENT_OCCURRED_IPI_INT_MASK | \
@@ -149,310 +151,310 @@
 /* ==================================================================== */
 /*                         LEDS                                         */
 /* ==================================================================== */
-#define SH1_REAL_JUNK_BUS_LED0			 0x7fed00000UL
-#define SH1_REAL_JUNK_BUS_LED1			 0x7fed10000UL
-#define SH1_REAL_JUNK_BUS_LED2			 0x7fed20000UL
-#define SH1_REAL_JUNK_BUS_LED3			 0x7fed30000UL
+#define SH1_REAL_JUNK_BUS_LED0			0x7fed00000UL
+#define SH1_REAL_JUNK_BUS_LED1			0x7fed10000UL
+#define SH1_REAL_JUNK_BUS_LED2			0x7fed20000UL
+#define SH1_REAL_JUNK_BUS_LED3			0x7fed30000UL
 
-#define SH2_REAL_JUNK_BUS_LED0			 0xf0000000UL
-#define SH2_REAL_JUNK_BUS_LED1			 0xf0010000UL
-#define SH2_REAL_JUNK_BUS_LED2			 0xf0020000UL
-#define SH2_REAL_JUNK_BUS_LED3			 0xf0030000UL
+#define SH2_REAL_JUNK_BUS_LED0			0xf0000000UL
+#define SH2_REAL_JUNK_BUS_LED1			0xf0010000UL
+#define SH2_REAL_JUNK_BUS_LED2			0xf0020000UL
+#define SH2_REAL_JUNK_BUS_LED3			0xf0030000UL
 
 /* ==================================================================== */
 /*                         Register "SH1_PTC_0"                         */
 /*       Puge Translation Cache Message Configuration Information       */
 /* ==================================================================== */
-#define SH1_PTC_0                                 0x00000001101a0000
+#define SH1_PTC_0			__IA64_UL_CONST(0x00000001101a0000)
 
 /*   SH1_PTC_0_A                                                        */
 /*   Description:  Type                                                 */
-#define SH1_PTC_0_A_SHFT                          0
+#define SH1_PTC_0_A_SHFT				0
 
 /*   SH1_PTC_0_PS                                                       */
 /*   Description:  Page Size                                            */
-#define SH1_PTC_0_PS_SHFT                         2
+#define SH1_PTC_0_PS_SHFT				2
 
 /*   SH1_PTC_0_RID                                                      */
 /*   Description:  Region ID                                            */
-#define SH1_PTC_0_RID_SHFT                        8
+#define SH1_PTC_0_RID_SHFT				8
 
 /*   SH1_PTC_0_START                                                    */
 /*   Description:  Start                                                */
-#define SH1_PTC_0_START_SHFT                      63
+#define SH1_PTC_0_START_SHFT				63
 
 /* ==================================================================== */
 /*                         Register "SH1_PTC_1"                         */
 /*       Puge Translation Cache Message Configuration Information       */
 /* ==================================================================== */
-#define SH1_PTC_1                                 0x00000001101a0080
+#define SH1_PTC_1			__IA64_UL_CONST(0x00000001101a0080)
 
 /*   SH1_PTC_1_START                                                    */
 /*   Description:  PTC_1 Start                                          */
-#define SH1_PTC_1_START_SHFT                      63
-
+#define SH1_PTC_1_START_SHFT				63
 
 /* ==================================================================== */
 /*                         Register "SH2_PTC"                           */
 /*       Puge Translation Cache Message Configuration Information       */
 /* ==================================================================== */
-#define SH2_PTC                                   0x0000000170000000
+#define SH2_PTC				__IA64_UL_CONST(0x0000000170000000)
 
 /*   SH2_PTC_A                                                          */
 /*   Description:  Type                                                 */
-#define SH2_PTC_A_SHFT                            0
+#define SH2_PTC_A_SHFT					0
 
 /*   SH2_PTC_PS                                                         */
 /*   Description:  Page Size                                            */
-#define SH2_PTC_PS_SHFT                           2
+#define SH2_PTC_PS_SHFT					2
 
 /*   SH2_PTC_RID                                                      */
 /*   Description:  Region ID                                            */
-#define SH2_PTC_RID_SHFT                          4
+#define SH2_PTC_RID_SHFT				4
 
 /*   SH2_PTC_START                                                      */
 /*   Description:  Start                                                */
-#define SH2_PTC_START_SHFT                        63
+#define SH2_PTC_START_SHFT				63
 
 /*   SH2_PTC_ADDR_RID                                                   */
 /*   Description:  Region ID                                            */
-#define SH2_PTC_ADDR_SHFT                         4
-#define SH2_PTC_ADDR_MASK                         0x1ffffffffffff000
+#define SH2_PTC_ADDR_SHFT				4
+#define SH2_PTC_ADDR_MASK		__IA64_UL_CONST(0x1ffffffffffff000)
 
 /* ==================================================================== */
 /*                    Register "SH_RTC1_INT_CONFIG"                     */
 /*                SHub RTC 1 Interrupt Config Registers                 */
 /* ==================================================================== */
 
-#define SH1_RTC1_INT_CONFIG                      0x0000000110001480
-#define SH2_RTC1_INT_CONFIG                      0x0000000010001480
-#define SH_RTC1_INT_CONFIG_MASK                  0x0ff3ffffffefffff
-#define SH_RTC1_INT_CONFIG_INIT                  0x0000000000000000
+#define SH1_RTC1_INT_CONFIG		__IA64_UL_CONST(0x0000000110001480)
+#define SH2_RTC1_INT_CONFIG		__IA64_UL_CONST(0x0000000010001480)
+#define SH_RTC1_INT_CONFIG_MASK		__IA64_UL_CONST(0x0ff3ffffffefffff)
+#define SH_RTC1_INT_CONFIG_INIT		__IA64_UL_CONST(0x0000000000000000)
 
 /*   SH_RTC1_INT_CONFIG_TYPE                                            */
 /*   Description:  Type of Interrupt: 0=INT, 2=PMI, 4=NMI, 5=INIT       */
-#define SH_RTC1_INT_CONFIG_TYPE_SHFT             0
-#define SH_RTC1_INT_CONFIG_TYPE_MASK             0x0000000000000007
+#define SH_RTC1_INT_CONFIG_TYPE_SHFT			0
+#define SH_RTC1_INT_CONFIG_TYPE_MASK	__IA64_UL_CONST(0x0000000000000007)
 
 /*   SH_RTC1_INT_CONFIG_AGT                                             */
 /*   Description:  Agent, must be 0 for SHub                            */
-#define SH_RTC1_INT_CONFIG_AGT_SHFT              3
-#define SH_RTC1_INT_CONFIG_AGT_MASK              0x0000000000000008
+#define SH_RTC1_INT_CONFIG_AGT_SHFT			3
+#define SH_RTC1_INT_CONFIG_AGT_MASK	__IA64_UL_CONST(0x0000000000000008)
 
 /*   SH_RTC1_INT_CONFIG_PID                                             */
 /*   Description:  Processor ID, same setting as on targeted McKinley  */
-#define SH_RTC1_INT_CONFIG_PID_SHFT              4
-#define SH_RTC1_INT_CONFIG_PID_MASK              0x00000000000ffff0
+#define SH_RTC1_INT_CONFIG_PID_SHFT			4
+#define SH_RTC1_INT_CONFIG_PID_MASK	__IA64_UL_CONST(0x00000000000ffff0)
 
 /*   SH_RTC1_INT_CONFIG_BASE                                            */
 /*   Description:  Optional interrupt vector area, 2MB aligned          */
-#define SH_RTC1_INT_CONFIG_BASE_SHFT             21
-#define SH_RTC1_INT_CONFIG_BASE_MASK             0x0003ffffffe00000
+#define SH_RTC1_INT_CONFIG_BASE_SHFT			21
+#define SH_RTC1_INT_CONFIG_BASE_MASK	__IA64_UL_CONST(0x0003ffffffe00000)
 
 /*   SH_RTC1_INT_CONFIG_IDX                                             */
 /*   Description:  Targeted McKinley interrupt vector                   */
-#define SH_RTC1_INT_CONFIG_IDX_SHFT              52
-#define SH_RTC1_INT_CONFIG_IDX_MASK              0x0ff0000000000000
+#define SH_RTC1_INT_CONFIG_IDX_SHFT			52
+#define SH_RTC1_INT_CONFIG_IDX_MASK	__IA64_UL_CONST(0x0ff0000000000000)
 
 /* ==================================================================== */
 /*                    Register "SH_RTC1_INT_ENABLE"                     */
 /*                SHub RTC 1 Interrupt Enable Registers                 */
 /* ==================================================================== */
 
-#define SH1_RTC1_INT_ENABLE                      0x0000000110001500
-#define SH2_RTC1_INT_ENABLE                      0x0000000010001500
-#define SH_RTC1_INT_ENABLE_MASK                  0x0000000000000001
-#define SH_RTC1_INT_ENABLE_INIT                  0x0000000000000000
+#define SH1_RTC1_INT_ENABLE		__IA64_UL_CONST(0x0000000110001500)
+#define SH2_RTC1_INT_ENABLE		__IA64_UL_CONST(0x0000000010001500)
+#define SH_RTC1_INT_ENABLE_MASK		__IA64_UL_CONST(0x0000000000000001)
+#define SH_RTC1_INT_ENABLE_INIT		__IA64_UL_CONST(0x0000000000000000)
 
 /*   SH_RTC1_INT_ENABLE_RTC1_ENABLE                                     */
 /*   Description:  Enable RTC 1 Interrupt                               */
-#define SH_RTC1_INT_ENABLE_RTC1_ENABLE_SHFT      0
-#define SH_RTC1_INT_ENABLE_RTC1_ENABLE_MASK      0x0000000000000001
+#define SH_RTC1_INT_ENABLE_RTC1_ENABLE_SHFT		0
+#define SH_RTC1_INT_ENABLE_RTC1_ENABLE_MASK \
+					__IA64_UL_CONST(0x0000000000000001)
 
 /* ==================================================================== */
 /*                    Register "SH_RTC2_INT_CONFIG"                     */
 /*                SHub RTC 2 Interrupt Config Registers                 */
 /* ==================================================================== */
 
-#define SH1_RTC2_INT_CONFIG                      0x0000000110001580
-#define SH2_RTC2_INT_CONFIG                      0x0000000010001580
-#define SH_RTC2_INT_CONFIG_MASK                  0x0ff3ffffffefffff
-#define SH_RTC2_INT_CONFIG_INIT                  0x0000000000000000
+#define SH1_RTC2_INT_CONFIG		__IA64_UL_CONST(0x0000000110001580)
+#define SH2_RTC2_INT_CONFIG		__IA64_UL_CONST(0x0000000010001580)
+#define SH_RTC2_INT_CONFIG_MASK		__IA64_UL_CONST(0x0ff3ffffffefffff)
+#define SH_RTC2_INT_CONFIG_INIT		__IA64_UL_CONST(0x0000000000000000)
 
 /*   SH_RTC2_INT_CONFIG_TYPE                                            */
 /*   Description:  Type of Interrupt: 0=INT, 2=PMI, 4=NMI, 5=INIT       */
-#define SH_RTC2_INT_CONFIG_TYPE_SHFT             0
-#define SH_RTC2_INT_CONFIG_TYPE_MASK             0x0000000000000007
+#define SH_RTC2_INT_CONFIG_TYPE_SHFT			0
+#define SH_RTC2_INT_CONFIG_TYPE_MASK	__IA64_UL_CONST(0x0000000000000007)
 
 /*   SH_RTC2_INT_CONFIG_AGT                                             */
 /*   Description:  Agent, must be 0 for SHub                            */
-#define SH_RTC2_INT_CONFIG_AGT_SHFT              3
-#define SH_RTC2_INT_CONFIG_AGT_MASK              0x0000000000000008
+#define SH_RTC2_INT_CONFIG_AGT_SHFT			3
+#define SH_RTC2_INT_CONFIG_AGT_MASK	__IA64_UL_CONST(0x0000000000000008)
 
 /*   SH_RTC2_INT_CONFIG_PID                                             */
 /*   Description:  Processor ID, same setting as on targeted McKinley  */
-#define SH_RTC2_INT_CONFIG_PID_SHFT              4
-#define SH_RTC2_INT_CONFIG_PID_MASK              0x00000000000ffff0
+#define SH_RTC2_INT_CONFIG_PID_SHFT			4
+#define SH_RTC2_INT_CONFIG_PID_MASK	__IA64_UL_CONST(0x00000000000ffff0)
 
 /*   SH_RTC2_INT_CONFIG_BASE                                            */
 /*   Description:  Optional interrupt vector area, 2MB aligned          */
-#define SH_RTC2_INT_CONFIG_BASE_SHFT             21
-#define SH_RTC2_INT_CONFIG_BASE_MASK             0x0003ffffffe00000
+#define SH_RTC2_INT_CONFIG_BASE_SHFT			21
+#define SH_RTC2_INT_CONFIG_BASE_MASK	__IA64_UL_CONST(0x0003ffffffe00000)
 
 /*   SH_RTC2_INT_CONFIG_IDX                                             */
 /*   Description:  Targeted McKinley interrupt vector                   */
-#define SH_RTC2_INT_CONFIG_IDX_SHFT              52
-#define SH_RTC2_INT_CONFIG_IDX_MASK              0x0ff0000000000000
+#define SH_RTC2_INT_CONFIG_IDX_SHFT			52
+#define SH_RTC2_INT_CONFIG_IDX_MASK	__IA64_UL_CONST(0x0ff0000000000000)
 
 /* ==================================================================== */
 /*                    Register "SH_RTC2_INT_ENABLE"                     */
 /*                SHub RTC 2 Interrupt Enable Registers                 */
 /* ==================================================================== */
 
-#define SH1_RTC2_INT_ENABLE                      0x0000000110001600
-#define SH2_RTC2_INT_ENABLE                      0x0000000010001600
-#define SH_RTC2_INT_ENABLE_MASK                  0x0000000000000001
-#define SH_RTC2_INT_ENABLE_INIT                  0x0000000000000000
+#define SH1_RTC2_INT_ENABLE		__IA64_UL_CONST(0x0000000110001600)
+#define SH2_RTC2_INT_ENABLE		__IA64_UL_CONST(0x0000000010001600)
+#define SH_RTC2_INT_ENABLE_MASK		__IA64_UL_CONST(0x0000000000000001)
+#define SH_RTC2_INT_ENABLE_INIT		__IA64_UL_CONST(0x0000000000000000)
 
 /*   SH_RTC2_INT_ENABLE_RTC2_ENABLE                                     */
 /*   Description:  Enable RTC 2 Interrupt                               */
-#define SH_RTC2_INT_ENABLE_RTC2_ENABLE_SHFT      0
-#define SH_RTC2_INT_ENABLE_RTC2_ENABLE_MASK      0x0000000000000001
+#define SH_RTC2_INT_ENABLE_RTC2_ENABLE_SHFT		0
+#define SH_RTC2_INT_ENABLE_RTC2_ENABLE_MASK \
+					__IA64_UL_CONST(0x0000000000000001)
 
 /* ==================================================================== */
 /*                    Register "SH_RTC3_INT_CONFIG"                     */
 /*                SHub RTC 3 Interrupt Config Registers                 */
 /* ==================================================================== */
 
-#define SH1_RTC3_INT_CONFIG                      0x0000000110001680
-#define SH2_RTC3_INT_CONFIG                      0x0000000010001680
-#define SH_RTC3_INT_CONFIG_MASK                  0x0ff3ffffffefffff
-#define SH_RTC3_INT_CONFIG_INIT                  0x0000000000000000
+#define SH1_RTC3_INT_CONFIG		__IA64_UL_CONST(0x0000000110001680)
+#define SH2_RTC3_INT_CONFIG		__IA64_UL_CONST(0x0000000010001680)
+#define SH_RTC3_INT_CONFIG_MASK		__IA64_UL_CONST(0x0ff3ffffffefffff)
+#define SH_RTC3_INT_CONFIG_INIT		__IA64_UL_CONST(0x0000000000000000)
 
 /*   SH_RTC3_INT_CONFIG_TYPE                                            */
 /*   Description:  Type of Interrupt: 0=INT, 2=PMI, 4=NMI, 5=INIT       */
-#define SH_RTC3_INT_CONFIG_TYPE_SHFT             0
-#define SH_RTC3_INT_CONFIG_TYPE_MASK             0x0000000000000007
+#define SH_RTC3_INT_CONFIG_TYPE_SHFT			0
+#define SH_RTC3_INT_CONFIG_TYPE_MASK	__IA64_UL_CONST(0x0000000000000007)
 
 /*   SH_RTC3_INT_CONFIG_AGT                                             */
 /*   Description:  Agent, must be 0 for SHub                            */
-#define SH_RTC3_INT_CONFIG_AGT_SHFT              3
-#define SH_RTC3_INT_CONFIG_AGT_MASK              0x0000000000000008
+#define SH_RTC3_INT_CONFIG_AGT_SHFT			3
+#define SH_RTC3_INT_CONFIG_AGT_MASK	__IA64_UL_CONST(0x0000000000000008)
 
 /*   SH_RTC3_INT_CONFIG_PID                                             */
 /*   Description:  Processor ID, same setting as on targeted McKinley  */
-#define SH_RTC3_INT_CONFIG_PID_SHFT              4
-#define SH_RTC3_INT_CONFIG_PID_MASK              0x00000000000ffff0
+#define SH_RTC3_INT_CONFIG_PID_SHFT			4
+#define SH_RTC3_INT_CONFIG_PID_MASK	__IA64_UL_CONST(0x00000000000ffff0)
 
 /*   SH_RTC3_INT_CONFIG_BASE                                            */
 /*   Description:  Optional interrupt vector area, 2MB aligned          */
-#define SH_RTC3_INT_CONFIG_BASE_SHFT             21
-#define SH_RTC3_INT_CONFIG_BASE_MASK             0x0003ffffffe00000
+#define SH_RTC3_INT_CONFIG_BASE_SHFT			21
+#define SH_RTC3_INT_CONFIG_BASE_MASK	__IA64_UL_CONST(0x0003ffffffe00000)
 
 /*   SH_RTC3_INT_CONFIG_IDX                                             */
 /*   Description:  Targeted McKinley interrupt vector                   */
-#define SH_RTC3_INT_CONFIG_IDX_SHFT              52
-#define SH_RTC3_INT_CONFIG_IDX_MASK              0x0ff0000000000000
+#define SH_RTC3_INT_CONFIG_IDX_SHFT			52
+#define SH_RTC3_INT_CONFIG_IDX_MASK	__IA64_UL_CONST(0x0ff0000000000000)
 
 /* ==================================================================== */
 /*                    Register "SH_RTC3_INT_ENABLE"                     */
 /*                SHub RTC 3 Interrupt Enable Registers                 */
 /* ==================================================================== */
 
-#define SH1_RTC3_INT_ENABLE                      0x0000000110001700
-#define SH2_RTC3_INT_ENABLE                      0x0000000010001700
-#define SH_RTC3_INT_ENABLE_MASK                  0x0000000000000001
-#define SH_RTC3_INT_ENABLE_INIT                  0x0000000000000000
+#define SH1_RTC3_INT_ENABLE		__IA64_UL_CONST(0x0000000110001700)
+#define SH2_RTC3_INT_ENABLE		__IA64_UL_CONST(0x0000000010001700)
+#define SH_RTC3_INT_ENABLE_MASK		__IA64_UL_CONST(0x0000000000000001)
+#define SH_RTC3_INT_ENABLE_INIT		__IA64_UL_CONST(0x0000000000000000)
 
 /*   SH_RTC3_INT_ENABLE_RTC3_ENABLE                                     */
 /*   Description:  Enable RTC 3 Interrupt                               */
-#define SH_RTC3_INT_ENABLE_RTC3_ENABLE_SHFT      0
-#define SH_RTC3_INT_ENABLE_RTC3_ENABLE_MASK      0x0000000000000001
+#define SH_RTC3_INT_ENABLE_RTC3_ENABLE_SHFT		0
+#define SH_RTC3_INT_ENABLE_RTC3_ENABLE_MASK \
+					__IA64_UL_CONST(0x0000000000000001)
 
 /*   SH_EVENT_OCCURRED_RTC1_INT                                         */
 /*   Description:  Pending RTC 1 Interrupt                              */
-#define SH_EVENT_OCCURRED_RTC1_INT_SHFT          24
-#define SH_EVENT_OCCURRED_RTC1_INT_MASK          0x0000000001000000
+#define SH_EVENT_OCCURRED_RTC1_INT_SHFT			24
+#define SH_EVENT_OCCURRED_RTC1_INT_MASK	__IA64_UL_CONST(0x0000000001000000)
 
 /*   SH_EVENT_OCCURRED_RTC2_INT                                         */
 /*   Description:  Pending RTC 2 Interrupt                              */
-#define SH_EVENT_OCCURRED_RTC2_INT_SHFT          25
-#define SH_EVENT_OCCURRED_RTC2_INT_MASK          0x0000000002000000
+#define SH_EVENT_OCCURRED_RTC2_INT_SHFT			25
+#define SH_EVENT_OCCURRED_RTC2_INT_MASK	__IA64_UL_CONST(0x0000000002000000)
 
 /*   SH_EVENT_OCCURRED_RTC3_INT                                         */
 /*   Description:  Pending RTC 3 Interrupt                              */
-#define SH_EVENT_OCCURRED_RTC3_INT_SHFT          26
-#define SH_EVENT_OCCURRED_RTC3_INT_MASK          0x0000000004000000
+#define SH_EVENT_OCCURRED_RTC3_INT_SHFT			26
+#define SH_EVENT_OCCURRED_RTC3_INT_MASK	__IA64_UL_CONST(0x0000000004000000)
 
 /* ==================================================================== */
 /*                       Register "SH_IPI_ACCESS"                       */
 /*                 CPU interrupt Access Permission Bits                 */
 /* ==================================================================== */
 
-#define SH1_IPI_ACCESS                           0x0000000110060480
-#define SH2_IPI_ACCESS0                          0x0000000010060c00
-#define SH2_IPI_ACCESS1                          0x0000000010060c80
-#define SH2_IPI_ACCESS2                          0x0000000010060d00
-#define SH2_IPI_ACCESS3                          0x0000000010060d80
+#define SH1_IPI_ACCESS			__IA64_UL_CONST(0x0000000110060480)
+#define SH2_IPI_ACCESS0			__IA64_UL_CONST(0x0000000010060c00)
+#define SH2_IPI_ACCESS1			__IA64_UL_CONST(0x0000000010060c80)
+#define SH2_IPI_ACCESS2			__IA64_UL_CONST(0x0000000010060d00)
+#define SH2_IPI_ACCESS3			__IA64_UL_CONST(0x0000000010060d80)
 
 /* ==================================================================== */
 /*                        Register "SH_INT_CMPB"                        */
 /*                  RTC Compare Value for Processor B                   */
 /* ==================================================================== */
 
-#define SH1_INT_CMPB                             0x00000001101b0080
-#define SH2_INT_CMPB                             0x00000000101b0080
-#define SH_INT_CMPB_MASK                         0x007fffffffffffff
-#define SH_INT_CMPB_INIT                         0x0000000000000000
+#define SH1_INT_CMPB			__IA64_UL_CONST(0x00000001101b0080)
+#define SH2_INT_CMPB			__IA64_UL_CONST(0x00000000101b0080)
+#define SH_INT_CMPB_MASK		__IA64_UL_CONST(0x007fffffffffffff)
+#define SH_INT_CMPB_INIT		__IA64_UL_CONST(0x0000000000000000)
 
 /*   SH_INT_CMPB_REAL_TIME_CMPB                                         */
 /*   Description:  Real Time Clock Compare                              */
-#define SH_INT_CMPB_REAL_TIME_CMPB_SHFT          0
-#define SH_INT_CMPB_REAL_TIME_CMPB_MASK          0x007fffffffffffff
+#define SH_INT_CMPB_REAL_TIME_CMPB_SHFT			0
+#define SH_INT_CMPB_REAL_TIME_CMPB_MASK	__IA64_UL_CONST(0x007fffffffffffff)
 
 /* ==================================================================== */
 /*                        Register "SH_INT_CMPC"                        */
 /*                  RTC Compare Value for Processor C                   */
 /* ==================================================================== */
 
-#define SH1_INT_CMPC                             0x00000001101b0100
-#define SH2_INT_CMPC                             0x00000000101b0100
-#define SH_INT_CMPC_MASK                         0x007fffffffffffff
-#define SH_INT_CMPC_INIT                         0x0000000000000000
+#define SH1_INT_CMPC			__IA64_UL_CONST(0x00000001101b0100)
+#define SH2_INT_CMPC			__IA64_UL_CONST(0x00000000101b0100)
+#define SH_INT_CMPC_MASK		__IA64_UL_CONST(0x007fffffffffffff)
+#define SH_INT_CMPC_INIT		__IA64_UL_CONST(0x0000000000000000)
 
 /*   SH_INT_CMPC_REAL_TIME_CMPC                                         */
 /*   Description:  Real Time Clock Compare                              */
-#define SH_INT_CMPC_REAL_TIME_CMPC_SHFT          0
-#define SH_INT_CMPC_REAL_TIME_CMPC_MASK          0x007fffffffffffff
+#define SH_INT_CMPC_REAL_TIME_CMPC_SHFT			0
+#define SH_INT_CMPC_REAL_TIME_CMPC_MASK	__IA64_UL_CONST(0x007fffffffffffff)
 
 /* ==================================================================== */
 /*                        Register "SH_INT_CMPD"                        */
 /*                  RTC Compare Value for Processor D                   */
 /* ==================================================================== */
 
-#define SH1_INT_CMPD                             0x00000001101b0180
-#define SH2_INT_CMPD                             0x00000000101b0180
-#define SH_INT_CMPD_MASK                         0x007fffffffffffff
-#define SH_INT_CMPD_INIT                         0x0000000000000000
+#define SH1_INT_CMPD			__IA64_UL_CONST(0x00000001101b0180)
+#define SH2_INT_CMPD			__IA64_UL_CONST(0x00000000101b0180)
+#define SH_INT_CMPD_MASK		__IA64_UL_CONST(0x007fffffffffffff)
+#define SH_INT_CMPD_INIT		__IA64_UL_CONST(0x0000000000000000)
 
 /*   SH_INT_CMPD_REAL_TIME_CMPD                                         */
 /*   Description:  Real Time Clock Compare                              */
-#define SH_INT_CMPD_REAL_TIME_CMPD_SHFT          0
-#define SH_INT_CMPD_REAL_TIME_CMPD_MASK          0x007fffffffffffff
+#define SH_INT_CMPD_REAL_TIME_CMPD_SHFT			0
+#define SH_INT_CMPD_REAL_TIME_CMPD_MASK	__IA64_UL_CONST(0x007fffffffffffff)
 
 /* ==================================================================== */
 /*                Register "SH_MD_DQLP_MMR_DIR_PRIVEC0"                 */
 /*                      privilege vector for acc=0                      */
 /* ==================================================================== */
-
-#define SH1_MD_DQLP_MMR_DIR_PRIVEC0              0x0000000100030300
+#define SH1_MD_DQLP_MMR_DIR_PRIVEC0	__IA64_UL_CONST(0x0000000100030300)
 
 /* ==================================================================== */
 /*                Register "SH_MD_DQRP_MMR_DIR_PRIVEC0"                 */
 /*                      privilege vector for acc=0                      */
 /* ==================================================================== */
-
-#define SH1_MD_DQRP_MMR_DIR_PRIVEC0              0x0000000100050300
+#define SH1_MD_DQRP_MMR_DIR_PRIVEC0	__IA64_UL_CONST(0x0000000100050300)
 
 /* ==================================================================== */
 /* Some MMRs are functionally identical (or close enough) on both SHUB1 */
@@ -484,17 +486,17 @@
 /*                    Engine 0 Control and Status Register                    */
 /* ========================================================================== */
 
-#define SH2_BT_ENG_CSR_0                         0x0000000030040000
-#define SH2_BT_ENG_SRC_ADDR_0                    0x0000000030040080
-#define SH2_BT_ENG_DEST_ADDR_0                   0x0000000030040100
-#define SH2_BT_ENG_NOTIF_ADDR_0                  0x0000000030040180
+#define SH2_BT_ENG_CSR_0		__IA64_UL_CONST(0x0000000030040000)
+#define SH2_BT_ENG_SRC_ADDR_0		__IA64_UL_CONST(0x0000000030040080)
+#define SH2_BT_ENG_DEST_ADDR_0		__IA64_UL_CONST(0x0000000030040100)
+#define SH2_BT_ENG_NOTIF_ADDR_0		__IA64_UL_CONST(0x0000000030040180)
 
 /* ========================================================================== */
 /*                       BTE interfaces 1-3                                   */
 /* ========================================================================== */
 
-#define SH2_BT_ENG_CSR_1                         0x0000000030050000
-#define SH2_BT_ENG_CSR_2                         0x0000000030060000
-#define SH2_BT_ENG_CSR_3                         0x0000000030070000
+#define SH2_BT_ENG_CSR_1		__IA64_UL_CONST(0x0000000030050000)
+#define SH2_BT_ENG_CSR_2		__IA64_UL_CONST(0x0000000030060000)
+#define SH2_BT_ENG_CSR_3		__IA64_UL_CONST(0x0000000030070000)
 
 #endif /* _ASM_IA64_SN_SHUB_MMR_H */
-- 
cgit v1.2.3-70-g09d2


From 71a5d027c90dd7e14ddaaf22358ff5138ba89ec9 Mon Sep 17 00:00:00 2001
From: Jack Steiner <steiner@sgi.com>
Date: Tue, 10 May 2005 08:01:00 -0700
Subject: [IA64-SGI] - new macros for SGI SN simulator

This patch changes some macros that are used when running kernel on the
SGI simulator.

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/kernel/io_init.c   |  2 +-
 arch/ia64/sn/kernel/setup.c     | 27 +++++++++++++++++++--------
 include/asm-ia64/sn/simulator.h | 13 +++++++------
 include/asm-ia64/sn/sn_sal.h    | 10 ++++++++++
 4 files changed, 37 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c
index 9e07f5463f2..783eb432384 100644
--- a/arch/ia64/sn/kernel/io_init.c
+++ b/arch/ia64/sn/kernel/io_init.c
@@ -384,7 +384,7 @@ static int __init sn_pci_init(void)
 	extern void register_sn_procfs(void);
 #endif
 
-	if (!ia64_platform_is("sn2") || IS_RUNNING_ON_SIMULATOR())
+	if (!ia64_platform_is("sn2") || IS_RUNNING_ON_FAKE_PROM())
 		return 0;
 
 	/*
diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c
index 3e5ba34ac2e..22e10d282c7 100644
--- a/arch/ia64/sn/kernel/setup.c
+++ b/arch/ia64/sn/kernel/setup.c
@@ -96,6 +96,7 @@ u8 sn_coherency_id;
 EXPORT_SYMBOL(sn_coherency_id);
 u8 sn_region_size;
 EXPORT_SYMBOL(sn_region_size);
+int sn_prom_type;	/* 0=hardware, 1=medusa/realprom, 2=medusa/fakeprom */
 
 short physical_node_map[MAX_PHYSNODE_ID];
 
@@ -354,7 +355,7 @@ void __init sn_setup(char **cmdline_p)
 
 	ia64_mark_idle = &snidle;
 
-	/* 
+	/*
 	 * For the bootcpu, we do this here. All other cpus will make the
 	 * call as part of cpu_init in slave cpu initialization.
 	 */
@@ -401,7 +402,7 @@ static void __init sn_init_pdas(char **cmdline_p)
 		nodepdaindr[cnode] =
 		    alloc_bootmem_node(NODE_DATA(cnode), sizeof(nodepda_t));
 		memset(nodepdaindr[cnode], 0, sizeof(nodepda_t));
-		memset(nodepdaindr[cnode]->phys_cpuid, -1, 
+		memset(nodepdaindr[cnode]->phys_cpuid, -1,
 		    sizeof(nodepdaindr[cnode]->phys_cpuid));
 	}
 
@@ -431,7 +432,7 @@ static void __init sn_init_pdas(char **cmdline_p)
 	}
 
 	/*
-	 * Initialize the per node hubdev.  This includes IO Nodes and 
+	 * Initialize the per node hubdev.  This includes IO Nodes and
 	 * headless/memless nodes.
 	 */
 	for (cnode = 0; cnode < numionodes; cnode++) {
@@ -459,6 +460,14 @@ void __init sn_cpu_init(void)
 	int i;
 	static int wars_have_been_checked;
 
+	if (smp_processor_id() == 0 && IS_MEDUSA()) {
+		if (ia64_sn_is_fake_prom())
+			sn_prom_type = 2;
+		else
+			sn_prom_type = 1;
+		printk("Running on medusa with %s PROM\n", (sn_prom_type == 1) ? "real" : "fake");
+	}
+
 	memset(pda, 0, sizeof(pda));
 	if (ia64_sn_get_sn_info(0, &sn_hub_info->shub2, &sn_hub_info->nasid_bitmask, &sn_hub_info->nasid_shift,
 				&sn_system_size, &sn_sharing_domain_size, &sn_partition_id,
@@ -524,7 +533,7 @@ void __init sn_cpu_init(void)
 	 */
 	{
 		u64 pio1[] = {SH1_PIO_WRITE_STATUS_0, 0, SH1_PIO_WRITE_STATUS_1, 0};
-		u64 pio2[] = {SH2_PIO_WRITE_STATUS_0, SH2_PIO_WRITE_STATUS_1, 
+		u64 pio2[] = {SH2_PIO_WRITE_STATUS_0, SH2_PIO_WRITE_STATUS_1,
 			SH2_PIO_WRITE_STATUS_2, SH2_PIO_WRITE_STATUS_3};
 		u64 *pio;
 		pio = is_shub1() ? pio1 : pio2;
@@ -556,6 +565,10 @@ static void __init scan_for_ionodes(void)
 	int nasid = 0;
 	lboard_t *brd;
 
+	/* fakeprom does not support klgraph */
+	if (IS_RUNNING_ON_FAKE_PROM())
+		return;
+
 	/* Setup ionodes with memory */
 	for (nasid = 0; nasid < MAX_PHYSNODE_ID; nasid += 2) {
 		char *klgraph_header;
@@ -567,8 +580,6 @@ static void __init scan_for_ionodes(void)
 		cnodeid = -1;
 		klgraph_header = __va(ia64_sn_get_klconfig_addr(nasid));
 		if (!klgraph_header) {
-			if (IS_RUNNING_ON_SIMULATOR())
-				continue;
 			BUG();	/* All nodes must have klconfig tables! */
 		}
 		cnodeid = nasid_to_cnodeid(nasid);
@@ -634,8 +645,8 @@ int
 nasid_slice_to_cpuid(int nasid, int slice)
 {
 	long cpu;
-	
-	for (cpu=0; cpu < NR_CPUS; cpu++) 
+
+	for (cpu=0; cpu < NR_CPUS; cpu++)
 		if (cpuid_to_nasid(cpu) == nasid &&
 					cpuid_to_slice(cpu) == slice)
 			return cpu;
diff --git a/include/asm-ia64/sn/simulator.h b/include/asm-ia64/sn/simulator.h
index 78eb4f869c8..cf770e246af 100644
--- a/include/asm-ia64/sn/simulator.h
+++ b/include/asm-ia64/sn/simulator.h
@@ -10,16 +10,17 @@
 
 #include <linux/config.h>
 
-#ifdef CONFIG_IA64_SGI_SN_SIM
-
 #define SNMAGIC 0xaeeeeeee8badbeefL
-#define IS_RUNNING_ON_SIMULATOR() ({long sn; asm("mov %0=cpuid[%1]" : "=r"(sn) : "r"(2)); sn == SNMAGIC;})
-
-#define SIMULATOR_SLEEP()	asm("nop.i 0x8beef")
+#define IS_MEDUSA()			({long sn; asm("mov %0=cpuid[%1]" : "=r"(sn) : "r"(2)); sn == SNMAGIC;})
 
+#ifdef CONFIG_IA64_SGI_SN_SIM
+#define SIMULATOR_SLEEP()		asm("nop.i 0x8beef")
+#define IS_RUNNING_ON_SIMULATOR() 	(sn_prom_type)
+#define IS_RUNNING_ON_FAKE_PROM()	(sn_prom_type == 2)
+extern int sn_prom_type;		/* 0=hardware, 1=medusa/realprom, 2=medusa/fakeprom */
 #else
-
 #define IS_RUNNING_ON_SIMULATOR()	(0)
+#define IS_RUNNING_ON_FAKE_PROM()	(0)
 #define SIMULATOR_SLEEP()
 
 #endif
diff --git a/include/asm-ia64/sn/sn_sal.h b/include/asm-ia64/sn/sn_sal.h
index eb0395ad0d6..1455375d2ce 100644
--- a/include/asm-ia64/sn/sn_sal.h
+++ b/include/asm-ia64/sn/sn_sal.h
@@ -132,6 +132,8 @@
 #define SALRET_INVALID_ARG	(-2)
 #define SALRET_ERROR		(-3)
 
+#define SN_SAL_FAKE_PROM			   0x02009999
+
 
 /**
  * sn_sal_rev_major - get the major SGI SAL revision number
@@ -1105,4 +1107,12 @@ ia64_sn_bte_recovery(nasid_t nasid)
 	return (int) rv.status;
 }
 
+static inline int
+ia64_sn_is_fake_prom(void)
+{
+	struct ia64_sal_retval rv;
+	SAL_CALL_NOLOCK(rv, SN_SAL_FAKE_PROM, 0, 0, 0, 0, 0, 0, 0);
+	return (rv.status == 0);
+}
+
 #endif /* _ASM_IA64_SN_SN_SAL_H */
-- 
cgit v1.2.3-70-g09d2


From 71030994a77e7689a7a792128b131ef7a10b9f74 Mon Sep 17 00:00:00 2001
From: Colin Ngam <cngam@sgi.com>
Date: Tue, 10 May 2005 09:12:00 -0700
Subject: [IA64-SGI] Fix TIO IOSPACE MMR Addres

This patches provides support on Shub2 for the separate TIO IOSPACE MMR.  This
patch is SN specific.

Signed-off-by: Colin Ngam <cngam@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 include/asm-ia64/sn/addrs.h | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-ia64/sn/addrs.h b/include/asm-ia64/sn/addrs.h
index 1bfdfb4d7b0..103d745dc5f 100644
--- a/include/asm-ia64/sn/addrs.h
+++ b/include/asm-ia64/sn/addrs.h
@@ -216,6 +216,10 @@
 #define TIO_SWIN_WIDGETNUM(x)		(((x)  >> TIO_SWIN_SIZE_BITS) & TIO_SWIN_WIDGET_MASK)
 
 
+#define TIO_IOSPACE_ADDR(n,x)					\
+	/* Move in the Chiplet ID for TIO Local Block MMR */	\
+	(REMOTE_ADDR(n,x) | 1UL << (NASID_SHIFT - 2))
+
 /*
  * The following macros produce the correct base virtual address for
  * the hub registers. The REMOTE_HUB_* macro produce
@@ -233,13 +237,16 @@
 #define REMOTE_HUB_ADDR(n,x)						\
 	((n & 1) ?							\
 	/* TIO: */							\
-	((volatile u64 *)(GLOBAL_MMR_ADDR(n,x)))			\
-	: /* SHUB: */							\
-	(((x) & BWIN_TOP) ? ((volatile u64 *)(GLOBAL_MMR_ADDR(n,x)))\
+	(is_shub2() ?							\
+	/* TIO on Shub2 */						\
+	(volatile u64 *)(TIO_IOSPACE_ADDR(n,x))				\
+	: /* TIO on shub1 */						\
+	(volatile u64 *)(GLOBAL_MMR_ADDR(n,x)))				\
+									\
+	: /* SHUB1 and SHUB2 MMRs: */					\
+	(((x) & BWIN_TOP) ? ((volatile u64 *)(GLOBAL_MMR_ADDR(n,x)))	\
 	: ((volatile u64 *)(NODE_SWIN_BASE(n,1) + 0x800000 + (x)))))
 
-
-
 #define HUB_L(x)			(*((volatile typeof(*x) *)x))
 #define	HUB_S(x,d)			(*((volatile typeof(*x) *)x) = (d))
 
-- 
cgit v1.2.3-70-g09d2


From 92a582ed2757456ca9599f8b4ea2064f2154eb02 Mon Sep 17 00:00:00 2001
From: Prarit Bhargava <prarit@sgi.com>
Date: Tue, 10 May 2005 12:40:00 -0700
Subject: [IA64] sparse cleanup of TIOCA files

This patch is a sparse compile cleanup of tioca_provider.c, sn_hwperf.h, and
tioca_provider.h.  Each of these files had sparse warnings when
compiled.

Signed-off-by: Prarit Bhargava <prarit@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/pci/tioca_provider.c    | 8 ++++----
 include/asm-ia64/sn/sn2/sn_hwperf.h  | 2 ++
 include/asm-ia64/sn/tioca_provider.h | 1 +
 3 files changed, 7 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/sn/pci/tioca_provider.c b/arch/ia64/sn/pci/tioca_provider.c
index 8dae9eb4545..05aa8c2fe9b 100644
--- a/arch/ia64/sn/pci/tioca_provider.c
+++ b/arch/ia64/sn/pci/tioca_provider.c
@@ -336,7 +336,7 @@ tioca_dma_d48(struct pci_dev *pdev, uint64_t paddr)
 	if (!ct_addr)
 		return 0;
 
-	bus_addr = (dma_addr_t) (ct_addr & 0xffffffffffff);
+	bus_addr = (dma_addr_t) (ct_addr & 0xffffffffffffUL);
 	node_upper = ct_addr >> 48;
 
 	if (node_upper > 64) {
@@ -464,7 +464,7 @@ map_return:
  * For mappings created using the direct modes (64 or 48) there are no
  * resources to release.
  */
-void
+static void
 tioca_dma_unmap(struct pci_dev *pdev, dma_addr_t bus_addr, int dir)
 {
 	int i, entry;
@@ -514,7 +514,7 @@ tioca_dma_unmap(struct pci_dev *pdev, dma_addr_t bus_addr, int dir)
  * The mapping mode used is based on the devices dma_mask.  As a last resort
  * use the GART mapped mode.
  */
-uint64_t
+static uint64_t
 tioca_dma_map(struct pci_dev *pdev, uint64_t paddr, size_t byte_count)
 {
 	uint64_t mapaddr;
@@ -580,7 +580,7 @@ tioca_error_intr_handler(int irq, void *arg, struct pt_regs *pt)
  * On successful setup, returns the kernel version of tioca_common back to
  * the caller.
  */
-void *
+static void *
 tioca_bus_fixup(struct pcibus_bussoft *prom_bussoft)
 {
 	struct tioca_common *tioca_common;
diff --git a/include/asm-ia64/sn/sn2/sn_hwperf.h b/include/asm-ia64/sn/sn2/sn_hwperf.h
index b0c4d6dd77b..df75f4c4aec 100644
--- a/include/asm-ia64/sn/sn2/sn_hwperf.h
+++ b/include/asm-ia64/sn/sn2/sn_hwperf.h
@@ -223,4 +223,6 @@ struct sn_hwperf_ioctl_args {
 #define SN_HWPERF_OP_RECONFIGURE	253
 #define SN_HWPERF_OP_INVAL		254
 
+int sn_topology_open(struct inode *inode, struct file *file);
+int sn_topology_release(struct inode *inode, struct file *file);
 #endif				/* SN_HWPERF_H */
diff --git a/include/asm-ia64/sn/tioca_provider.h b/include/asm-ia64/sn/tioca_provider.h
index b6acc22ab23..5ccec608d32 100644
--- a/include/asm-ia64/sn/tioca_provider.h
+++ b/include/asm-ia64/sn/tioca_provider.h
@@ -201,6 +201,7 @@ tioca_tlbflush(struct tioca_kernel *tioca_kernel)
 }
 
 extern uint32_t	tioca_gart_found;
+extern struct list_head tioca_list;
 extern int tioca_init_provider(void);
 extern void tioca_fastwrite_enable(struct tioca_kernel *tioca_kern);
 #endif /* _ASM_IA64_SN_TIO_CA_AGP_PROVIDER_H */
-- 
cgit v1.2.3-70-g09d2


From c1ffb6a8aaed45ba2dc0a8f09241b0a96f9955ba Mon Sep 17 00:00:00 2001
From: Bruce Losure <blosure@sgi.com>
Date: Tue, 24 May 2005 08:30:00 -0700
Subject: [IA64-SGI] Altix patch to tiocx, add subsys_initcall

This patch fixes an ordering issue between the init code for the
tiocx bus driver and tiocx-related device drivers.   Also adds
a new brick to the list of known FPGA bricks.

Signed-off-by: Bruce Losure <blosure@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/kernel/tiocx.c | 14 +++++++++-----
 include/asm-ia64/sn/l1.h    |  1 +
 2 files changed, 10 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/sn/kernel/tiocx.c b/arch/ia64/sn/kernel/tiocx.c
index a087b274847..8716f4d5314 100644
--- a/arch/ia64/sn/kernel/tiocx.c
+++ b/arch/ia64/sn/kernel/tiocx.c
@@ -204,8 +204,8 @@ cx_device_register(nasid_t nasid, int part_num, int mfg_num,
 	cx_dev->dev.parent = NULL;
 	cx_dev->dev.bus = &tiocx_bus_type;
 	cx_dev->dev.release = tiocx_bus_release;
-	snprintf(cx_dev->dev.bus_id, BUS_ID_SIZE, "%d.0x%x",
-		 cx_dev->cx_id.nasid, cx_dev->cx_id.part_num);
+	snprintf(cx_dev->dev.bus_id, BUS_ID_SIZE, "%d",
+		 cx_dev->cx_id.nasid);
 	device_register(&cx_dev->dev);
 	get_device(&cx_dev->dev);
 
@@ -236,7 +236,6 @@ int cx_device_unregister(struct cx_dev *cx_dev)
  */
 static int cx_device_reload(struct cx_dev *cx_dev)
 {
-	device_remove_file(&cx_dev->dev, &dev_attr_cxdev_control);
 	cx_device_unregister(cx_dev);
 	return cx_device_register(cx_dev->cx_id.nasid, cx_dev->cx_id.part_num,
 				  cx_dev->cx_id.mfg_num, cx_dev->hubdev);
@@ -383,6 +382,7 @@ static int is_fpga_brick(int nasid)
 	switch (tiocx_btchar_get(nasid)) {
 	case L1_BRICKTYPE_SA:
 	case L1_BRICKTYPE_ATHENA:
+	case L1_BRICKTYPE_DAYTONA:
 		return 1;
 	}
 	return 0;
@@ -409,7 +409,7 @@ static int tiocx_reload(struct cx_dev *cx_dev)
 		uint64_t cx_id;
 
 		cx_id =
-		    *(volatile int32_t *)(TIO_SWIN_BASE(nasid, TIOCX_CORELET) +
+		    *(volatile uint64_t *)(TIO_SWIN_BASE(nasid, TIOCX_CORELET) +
 					  WIDGET_ID);
 		part_num = XWIDGET_PART_NUM(cx_id);
 		mfg_num = XWIDGET_MFG_NUM(cx_id);
@@ -458,6 +458,10 @@ static ssize_t store_cxdev_control(struct device *dev, struct device_attribute *
 
 	switch (n) {
 	case 1:
+		tio_corelet_reset(cx_dev->cx_id.nasid, TIOCX_CORELET);
+		tiocx_reload(cx_dev);
+		break;
+	case 2:
 		tiocx_reload(cx_dev);
 		break;
 	case 3:
@@ -537,7 +541,7 @@ static void __exit tiocx_exit(void)
 	bus_unregister(&tiocx_bus_type);
 }
 
-module_init(tiocx_init);
+subsys_initcall(tiocx_init);
 module_exit(tiocx_exit);
 
 /************************************************************************
diff --git a/include/asm-ia64/sn/l1.h b/include/asm-ia64/sn/l1.h
index 08050d37b66..2e5f0aa3888 100644
--- a/include/asm-ia64/sn/l1.h
+++ b/include/asm-ia64/sn/l1.h
@@ -33,5 +33,6 @@
 #define L1_BRICKTYPE_PA		0x6a            /* j */
 #define L1_BRICKTYPE_IA		0x6b            /* k */
 #define L1_BRICKTYPE_ATHENA	0x2b            /* + */
+#define L1_BRICKTYPE_DAYTONA	0x7a            /* z */
 
 #endif /* _ASM_IA64_SN_L1_H */
-- 
cgit v1.2.3-70-g09d2


From a68db763af9b676590c3fe9ec3f17bf18015eb2f Mon Sep 17 00:00:00 2001
From: Peter Chubb <peterc@gelato.unsw.edu.au>
Date: Thu, 23 Jun 2005 21:14:00 -0700
Subject: [IA64] Fix another IA64 preemption problem

There's another problem shown up by Ingo's recent patch to make
smp_processor_id() complain if it's called with preemption enabled.
local_finish_flush_tlb_mm() calls activate_context() in a situation
where it could be rescheduled to another processor.  This patch
disables preemption around the call.

Signed-off-by: Peter Chubb <peterc@gelato.unsw.edu.au>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/smp.c         | 3 +++
 include/asm-ia64/mmu_context.h | 3 +++
 2 files changed, 6 insertions(+)

(limited to 'include')

diff --git a/arch/ia64/kernel/smp.c b/arch/ia64/kernel/smp.c
index b49d4ddaab9..0166a984709 100644
--- a/arch/ia64/kernel/smp.c
+++ b/arch/ia64/kernel/smp.c
@@ -231,13 +231,16 @@ smp_flush_tlb_all (void)
 void
 smp_flush_tlb_mm (struct mm_struct *mm)
 {
+	preempt_disable();
 	/* this happens for the common case of a single-threaded fork():  */
 	if (likely(mm == current->active_mm && atomic_read(&mm->mm_users) == 1))
 	{
 		local_finish_flush_tlb_mm(mm);
+		preempt_enable();
 		return;
 	}
 
+	preempt_enable();
 	/*
 	 * We could optimize this further by using mm->cpu_vm_mask to track which CPUs
 	 * have been running in the address space.  It's not clear that this is worth the
diff --git a/include/asm-ia64/mmu_context.h b/include/asm-ia64/mmu_context.h
index 0096e7e0501..e3e5fededb0 100644
--- a/include/asm-ia64/mmu_context.h
+++ b/include/asm-ia64/mmu_context.h
@@ -132,6 +132,9 @@ reload_context (mm_context_t context)
 	ia64_srlz_i();			/* srlz.i implies srlz.d */
 }
 
+/*
+ * Must be called with preemption off
+ */
 static inline void
 activate_context (struct mm_struct *mm)
 {
-- 
cgit v1.2.3-70-g09d2


From 053a7b5b7617a72d7c61b6f84196d1c0f79b9849 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Tue, 28 Jun 2005 19:22:25 +0100
Subject: [PATCH] ARM SMP: __xchg support

This enables the existing __xchg implementation to be used on SMP.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/system.h | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/system.h b/include/asm-arm/system.h
index 3d0d2860b6d..cdf49f442fd 100644
--- a/include/asm-arm/system.h
+++ b/include/asm-arm/system.h
@@ -290,7 +290,6 @@ do {									\
 })
 
 #ifdef CONFIG_SMP
-#error SMP not supported
 
 #define smp_mb()		mb()
 #define smp_rmb()		rmb()
@@ -304,6 +303,8 @@ do {									\
 #define smp_wmb()		barrier()
 #define smp_read_barrier_depends()		do { } while(0)
 
+#endif /* CONFIG_SMP */
+
 #if defined(CONFIG_CPU_SA1100) || defined(CONFIG_CPU_SA110)
 /*
  * On the StrongARM, "swp" is terminally broken since it bypasses the
@@ -316,9 +317,16 @@ do {									\
  *
  * We choose (1) since its the "easiest" to achieve here and is not
  * dependent on the processor type.
+ *
+ * NOTE that this solution won't work on an SMP system, so explcitly
+ * forbid it here.
  */
+#ifdef CONFIG_SMP
+#error SMP is not supported on SA1100/SA110
+#else
 #define swp_is_buggy
 #endif
+#endif
 
 static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size)
 {
@@ -361,8 +369,6 @@ static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size
 	return ret;
 }
 
-#endif /* CONFIG_SMP */
-
 #endif /* __ASSEMBLY__ */
 
 #define arch_align_stack(x) (x)
-- 
cgit v1.2.3-70-g09d2


From b3563c4fbff906991a1b4ef4609f99cca2a0de6a Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 28 Jun 2005 12:54:43 -0700
Subject: [NETLINK]: Clear padding in netlink messages

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netlink.h   | 1 +
 include/linux/rtnetlink.h | 5 ++++-
 net/core/rtnetlink.c      | 1 +
 3 files changed, 6 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/netlink.h b/include/linux/netlink.h
index 3029cad63a0..27e4d164a10 100644
--- a/include/linux/netlink.h
+++ b/include/linux/netlink.h
@@ -168,6 +168,7 @@ __nlmsg_put(struct sk_buff *skb, u32 pid, u32 seq, int type, int len, int flags)
 	nlh->nlmsg_flags = flags;
 	nlh->nlmsg_pid = pid;
 	nlh->nlmsg_seq = seq;
+	memset(NLMSG_DATA(nlh) + len, 0, NLMSG_ALIGN(size) - size);
 	return nlh;
 }
 
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index d021888b58f..dc26e82ba0f 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -898,7 +898,9 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi
 	memcpy(skb_put(skb, attrlen), data, attrlen); })
 
 #define RTA_PUT_NOHDR(skb, attrlen, data) \
-	RTA_APPEND(skb, RTA_ALIGN(attrlen), data)
+({	RTA_APPEND(skb, RTA_ALIGN(attrlen), data); \
+	memset(skb->tail - (RTA_ALIGN(attrlen) - attrlen), 0, \
+	       RTA_ALIGN(attrlen) - attrlen); })
 
 #define RTA_PUT_U8(skb, attrtype, value) \
 ({	u8 _tmp = (value); \
@@ -978,6 +980,7 @@ __rta_reserve(struct sk_buff *skb, int attrtype, int attrlen)
 	rta = (struct rtattr*)skb_put(skb, RTA_ALIGN(size));
 	rta->rta_type = attrtype;
 	rta->rta_len = size;
+	memset(RTA_DATA(rta) + attrlen, 0, RTA_ALIGN(size) - size);
 	return rta;
 }
 
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index e013d836a7a..879237c378f 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -126,6 +126,7 @@ void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const void *data
 	rta->rta_type = attrtype;
 	rta->rta_len = size;
 	memcpy(RTA_DATA(rta), data, attrlen);
+	memset(RTA_DATA(rta) + attrlen, 0, RTA_ALIGN(size) - size);
 }
 
 size_t rtattr_strlcpy(char *dest, const struct rtattr *rta, size_t size)
-- 
cgit v1.2.3-70-g09d2


From 8a47077a0b5aa2649751c46e7a27884e6686ccbf Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 28 Jun 2005 12:56:45 -0700
Subject: [NETLINK]: Missing padding fields in dumped structures

Plug holes with padding fields and initialized them to zero.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pkt_cls.h   | 1 +
 include/linux/pkt_sched.h | 9 ++++++---
 include/linux/rtnetlink.h | 5 +++++
 net/ipv6/addrconf.c       | 3 +++
 net/sched/cls_rsvp.h      | 1 +
 net/sched/sch_cbq.c       | 1 +
 6 files changed, 17 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/pkt_cls.h b/include/linux/pkt_cls.h
index 25d2d67c1fa..bd2c5a2bbbf 100644
--- a/include/linux/pkt_cls.h
+++ b/include/linux/pkt_cls.h
@@ -276,6 +276,7 @@ struct tc_rsvp_pinfo
 	__u8	protocol;
 	__u8	tunnelid;
 	__u8	tunnelhdr;
+	__u8	pad;
 };
 
 /* ROUTE filter */
diff --git a/include/linux/pkt_sched.h b/include/linux/pkt_sched.h
index 1d9da36eb9d..60ffcb9c579 100644
--- a/include/linux/pkt_sched.h
+++ b/include/linux/pkt_sched.h
@@ -221,9 +221,11 @@ struct tc_gred_qopt
 /* gred setup */
 struct tc_gred_sopt
 {
-       __u32           DPs;
-       __u32           def_DP;
-       __u8            grio;
+       __u32		DPs;
+       __u32		def_DP;
+       __u8		grio;
+       __u8		pad1;
+       __u16		pad2;
 };
 
 /* HTB section */
@@ -351,6 +353,7 @@ struct tc_cbq_ovl
 #define	TC_CBQ_OVL_DROP		3
 #define	TC_CBQ_OVL_RCLASSIC	4
 	unsigned char	priority2;
+	__u16		pad;
 	__u32		penalty;
 };
 
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index dc26e82ba0f..657c05ab8f9 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -363,6 +363,8 @@ enum
 struct rta_session
 {
 	__u8	proto;
+	__u8	pad1;
+	__u16	pad2;
 
 	union {
 		struct {
@@ -635,10 +637,13 @@ struct ifinfomsg
 struct prefixmsg
 {
 	unsigned char	prefix_family;
+	unsigned char	prefix_pad1;
+	unsigned short	prefix_pad2;
 	int		prefix_ifindex;
 	unsigned char	prefix_type;
 	unsigned char	prefix_len;
 	unsigned char	prefix_flags;
+	unsigned char	prefix_pad3;
 };
 
 enum 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 8140bed78a2..1b2902d8eb9 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -3031,9 +3031,12 @@ static int inet6_fill_prefix(struct sk_buff *skb, struct inet6_dev *idev,
 	nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*pmsg), flags);
 	pmsg = NLMSG_DATA(nlh);
 	pmsg->prefix_family = AF_INET6;
+	pmsg->prefix_pad1 = 0;
+	pmsg->prefix_pad2 = 0;
 	pmsg->prefix_ifindex = idev->dev->ifindex;
 	pmsg->prefix_len = pinfo->prefix_len;
 	pmsg->prefix_type = pinfo->type;
+	pmsg->prefix_pad3 = 0;
 	
 	pmsg->prefix_flags = 0;
 	if (pinfo->onlink)
diff --git a/net/sched/cls_rsvp.h b/net/sched/cls_rsvp.h
index 232fb919681..006168d6937 100644
--- a/net/sched/cls_rsvp.h
+++ b/net/sched/cls_rsvp.h
@@ -618,6 +618,7 @@ static int rsvp_dump(struct tcf_proto *tp, unsigned long fh,
 	pinfo.protocol = s->protocol;
 	pinfo.tunnelid = s->tunnelid;
 	pinfo.tunnelhdr = f->tunnelhdr;
+	pinfo.pad = 0;
 	RTA_PUT(skb, TCA_RSVP_PINFO, sizeof(pinfo), &pinfo);
 	if (f->res.classid)
 		RTA_PUT(skb, TCA_RSVP_CLASSID, 4, &f->res.classid);
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index baeb3111f75..09453f997d8 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1528,6 +1528,7 @@ static __inline__ int cbq_dump_ovl(struct sk_buff *skb, struct cbq_class *cl)
 
 	opt.strategy = cl->ovl_strategy;
 	opt.priority2 = cl->priority2+1;
+	opt.pad = 0;
 	opt.penalty = (cl->penalty*1000)/HZ;
 	RTA_PUT(skb, TCA_CBQ_OVL_STRATEGY, sizeof(opt), &opt);
 	return skb->len;
-- 
cgit v1.2.3-70-g09d2


From 072cd703b6a2a3adc14fbc2d969424c7022a7892 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Tue, 28 Jun 2005 21:01:03 +0100
Subject: [PATCH] ARM: 2762/1: S3C24XX Audio platform data

Patch from Ben Dooks

This provides the s3c24xx audio platform data which can be
supplied from any of the board specific drivers.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-s3c2410/audio.h | 49 ++++++++++++++++++++++++++++++++++++
 1 file changed, 49 insertions(+)
 create mode 100644 include/asm-arm/arch-s3c2410/audio.h

(limited to 'include')

diff --git a/include/asm-arm/arch-s3c2410/audio.h b/include/asm-arm/arch-s3c2410/audio.h
new file mode 100644
index 00000000000..0d276e67f2f
--- /dev/null
+++ b/include/asm-arm/arch-s3c2410/audio.h
@@ -0,0 +1,49 @@
+/* linux/include/asm-arm/arch-s3c2410/audio.h
+ *
+ * (c) 2004-2005 Simtec Electronics
+ *	http://www.simtec.co.uk/products/SWLINUX/
+ *	Ben Dooks <ben@simtec.co.uk>
+ *
+ * S3C24XX - Audio platfrom_device info
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Changelog:
+ *	20-Nov-2004 BJD  Created file
+ *	07-Mar-2005 BJD  Added suspend/resume calls
+*/
+
+#ifndef __ASM_ARCH_AUDIO_H
+#define __ASM_ARCH_AUDIO_H __FILE__
+
+/* struct s3c24xx_iis_ops
+ *
+ * called from the s3c24xx audio core to deal with the architecture
+ * or the codec's setup and control.
+ *
+ * the pointer to itself is passed through in case the caller wants to
+ * embed this in an larger structure for easy reference to it's context.
+*/
+
+struct s3c24xx_iis_ops {
+	struct module *owner;
+
+	int	(*startup)(struct s3c24xx_iis_ops *me);
+	void	(*shutdown)(struct s3c24xx_iis_ops *me);
+	int	(*suspend)(struct s3c24xx_iis_ops *me);
+	int	(*resume)(struct s3c24xx_iis_ops *me);
+
+	int	(*open)(struct s3c24xx_iis_ops *me, snd_pcm_substream_t *strm);
+	int	(*close)(struct s3c24xx_iis_ops *me, snd_pcm_substream_t *strm);
+	int	(*prepare)(struct s3c24xx_iis_ops *me, snd_pcm_substream_t *strm, snd_pcm_runtime_t *rt);
+};
+
+struct s3c24xx_platdata_iis {
+	const char		*codec_clk;
+	struct s3c24xx_iis_ops	*ops;
+	int			(*match_dev)(struct device *dev);
+};
+
+#endif /* __ASM_ARCH_AUDIO_H */
-- 
cgit v1.2.3-70-g09d2


From 2f85a42964dd43fed3a339701db046bee5a8b903 Mon Sep 17 00:00:00 2001
From: Vlad Yasevich <vladislav.yasevich@hp.com>
Date: Tue, 28 Jun 2005 13:24:23 -0700
Subject: [SCTP] Make init & delayed sack timeouts configurable by user.

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/sysctl.h       |  1 +
 include/net/sctp/constants.h | 18 +++---------------
 include/net/sctp/structs.h   |  4 ++++
 net/sctp/endpointola.c       | 13 +++++--------
 net/sctp/protocol.c          |  5 ++++-
 net/sctp/sysctl.c            | 13 +++++++++++++
 net/sctp/transport.c         |  1 -
 7 files changed, 30 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index ebfe1250f0a..5b5f434ac9a 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -641,6 +641,7 @@ enum {
 	NET_SCTP_ADDIP_ENABLE		 = 13,
 	NET_SCTP_PRSCTP_ENABLE		 = 14,
 	NET_SCTP_SNDBUF_POLICY		 = 15,
+	NET_SCTP_SACK_TIMEOUT		 = 16,
 };
 
 /* /proc/sys/net/bridge */
diff --git a/include/net/sctp/constants.h b/include/net/sctp/constants.h
index 4868c7f7749..5999e5684bb 100644
--- a/include/net/sctp/constants.h
+++ b/include/net/sctp/constants.h
@@ -263,23 +263,11 @@ enum { SCTP_MIN_PMTU = 576 };
 enum { SCTP_MAX_DUP_TSNS = 16 };
 enum { SCTP_MAX_GABS = 16 };
 
-/* Here we define the default timers.  */
+/* Heartbeat interval - 30 secs */
+#define SCTP_DEFAULT_TIMEOUT_HEARTBEAT	(30 * HZ)
 
-/* cookie timer def = ? seconds */
-#define SCTP_DEFAULT_TIMEOUT_T1_COOKIE	(3 * HZ)
-
-/* init timer def = 3 seconds  */
-#define SCTP_DEFAULT_TIMEOUT_T1_INIT	(3 * HZ)
-
-/* shutdown timer def = 300 ms */
-#define SCTP_DEFAULT_TIMEOUT_T2_SHUTDOWN ((300 * HZ) / 1000)
-
-/* 0 seconds + RTO */
-#define SCTP_DEFAULT_TIMEOUT_HEARTBEAT	(10 * HZ)
-
-/* recv timer def = 200ms (in usec) */
+/* Delayed sack timer - 200ms */
 #define SCTP_DEFAULT_TIMEOUT_SACK	((200 * HZ) / 1000)
-#define SCTP_DEFAULT_TIMEOUT_SACK_MAX	((500 * HZ) / 1000) /* 500 ms */
 
 /* RTO.Initial              - 3  seconds
  * RTO.Min                  - 1  second
diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index dfad4d3c581..47727c7cc62 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -161,6 +161,9 @@ extern struct sctp_globals {
 	 */
 	int sndbuf_policy;
 
+	/* Delayed SACK timeout  200ms default*/
+	int sack_timeout;
+
 	/* HB.interval		    - 30 seconds  */
 	int hb_interval;
 
@@ -217,6 +220,7 @@ extern struct sctp_globals {
 #define sctp_sndbuf_policy	 	(sctp_globals.sndbuf_policy)
 #define sctp_max_retrans_path		(sctp_globals.max_retrans_path)
 #define sctp_max_retrans_init		(sctp_globals.max_retrans_init)
+#define sctp_sack_timeout		(sctp_globals.sack_timeout)
 #define sctp_hb_interval		(sctp_globals.hb_interval)
 #define sctp_max_instreams		(sctp_globals.max_instreams)
 #define sctp_max_outstreams		(sctp_globals.max_outstreams)
diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c
index 2ec0320fac3..c44bf4165c6 100644
--- a/net/sctp/endpointola.c
+++ b/net/sctp/endpointola.c
@@ -102,9 +102,9 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep,
 	/* Set up the base timeout information.  */
 	ep->timeouts[SCTP_EVENT_TIMEOUT_NONE] = 0;
 	ep->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE] =
-		SCTP_DEFAULT_TIMEOUT_T1_COOKIE;
+		msecs_to_jiffies(sp->rtoinfo.srto_initial);
 	ep->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT] =
-		SCTP_DEFAULT_TIMEOUT_T1_INIT;
+		msecs_to_jiffies(sp->rtoinfo.srto_initial);
 	ep->timeouts[SCTP_EVENT_TIMEOUT_T2_SHUTDOWN] =
 		msecs_to_jiffies(sp->rtoinfo.srto_initial);
 	ep->timeouts[SCTP_EVENT_TIMEOUT_T3_RTX] = 0;
@@ -117,12 +117,9 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep,
         ep->timeouts[SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD]
 		= 5 * msecs_to_jiffies(sp->rtoinfo.srto_max);
 
-	ep->timeouts[SCTP_EVENT_TIMEOUT_HEARTBEAT] =
-		SCTP_DEFAULT_TIMEOUT_HEARTBEAT;
-	ep->timeouts[SCTP_EVENT_TIMEOUT_SACK] =
-		SCTP_DEFAULT_TIMEOUT_SACK;
-	ep->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE] =
-		sp->autoclose * HZ;
+	ep->timeouts[SCTP_EVENT_TIMEOUT_HEARTBEAT] = 0;
+	ep->timeouts[SCTP_EVENT_TIMEOUT_SACK] = sctp_sack_timeout;
+	ep->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE] = sp->autoclose * HZ;
 
 	/* Use SCTP specific send buffer space queues.  */
 	ep->sndbuf_policy = sctp_sndbuf_policy;
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index 5135e1a25d2..e7f37faba7c 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -1050,7 +1050,10 @@ SCTP_STATIC __init int sctp_init(void)
 	sctp_sndbuf_policy		= 0;
 
 	/* HB.interval              - 30 seconds */
-	sctp_hb_interval		= 30 * HZ;
+	sctp_hb_interval		= SCTP_DEFAULT_TIMEOUT_HEARTBEAT;
+
+	/* delayed SACK timeout */
+	sctp_sack_timeout		= SCTP_DEFAULT_TIMEOUT_SACK;
 
 	/* Implementation specific variables. */
 
diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c
index 7fc31849312..dc4893474f1 100644
--- a/net/sctp/sysctl.c
+++ b/net/sctp/sysctl.c
@@ -47,6 +47,8 @@
 static ctl_handler sctp_sysctl_jiffies_ms;
 static long rto_timer_min = 1;
 static long rto_timer_max = 86400000; /* One day */
+static long sack_timer_min = 1;
+static long sack_timer_max = 500;
 
 static ctl_table sctp_table[] = {
 	{
@@ -187,6 +189,17 @@ static ctl_table sctp_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
 	},
+	{
+		.ctl_name	= NET_SCTP_SACK_TIMEOUT,
+		.procname	= "sack_timeout",
+		.data		= &sctp_sack_timeout,
+		.maxlen		= sizeof(long),
+		.mode		= 0644,
+		.proc_handler	= &proc_doulongvec_ms_jiffies_minmax,
+		.strategy	= &sctp_sysctl_jiffies_ms,
+		.extra1         = &sack_timer_min,
+		.extra2         = &sack_timer_max,
+	},
 	{ .ctl_name = 0 }
 };
 
diff --git a/net/sctp/transport.c b/net/sctp/transport.c
index 0ec0fde6e6c..a63b6917960 100644
--- a/net/sctp/transport.c
+++ b/net/sctp/transport.c
@@ -103,7 +103,6 @@ static struct sctp_transport *sctp_transport_init(struct sctp_transport *peer,
 
 	/* Set up the heartbeat timer. */
 	init_timer(&peer->hb_timer);
-	peer->hb_interval = SCTP_DEFAULT_TIMEOUT_HEARTBEAT;
 	peer->hb_timer.function = sctp_generate_heartbeat_event;
 	peer->hb_timer.data = (unsigned long)peer;
 
-- 
cgit v1.2.3-70-g09d2


From 7fe40f73d7591b38f129fe6a9c0fa46e0b192d09 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Tue, 28 Jun 2005 15:46:24 -0700
Subject: [IPV6]: remove more unused IPV6_AUTHHDR things.

Remove two more unused IPV6_AUTHHDR option things,
which I failed to remove them last time,
plus, mark IPV6_AUTHHDR obsolete.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/in6.h      | 2 +-
 include/net/ipv6.h       | 1 -
 net/ipv6/ip6_flowlabel.c | 1 -
 3 files changed, 1 insertion(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/in6.h b/include/linux/in6.h
index f8256c58284..dcf5720ffcb 100644
--- a/include/linux/in6.h
+++ b/include/linux/in6.h
@@ -156,7 +156,7 @@ struct in6_flowlabel_req
 #define IPV6_CHECKSUM		7
 #define IPV6_HOPLIMIT		8
 #define IPV6_NEXTHOP		9
-#define IPV6_AUTHHDR		10
+#define IPV6_AUTHHDR		10	/* obsolete */
 #define IPV6_FLOWINFO		11
 
 #define IPV6_UNICAST_HOPS	16
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 771b47e30f8..69324465e8b 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -183,7 +183,6 @@ struct ipv6_txoptions
 	struct ipv6_opt_hdr	*hopopt;
 	struct ipv6_opt_hdr	*dst0opt;
 	struct ipv6_rt_hdr	*srcrt;	/* Routing Header */
-	struct ipv6_opt_hdr	*auth;
 	struct ipv6_opt_hdr	*dst1opt;
 
 	/* Option buffer, as read by IPV6_PKTOPTIONS, starts here. */
diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c
index 0e5f7499deb..b6c73da5ff3 100644
--- a/net/ipv6/ip6_flowlabel.c
+++ b/net/ipv6/ip6_flowlabel.c
@@ -244,7 +244,6 @@ struct ipv6_txoptions *fl6_merge_options(struct ipv6_txoptions * opt_space,
 		opt_space->opt_nflen = 0;
 	}
 	opt_space->dst1opt = fopt->dst1opt;
-	opt_space->auth = fopt->auth;
 	opt_space->opt_flen = fopt->opt_flen;
 	return opt_space;
 }
-- 
cgit v1.2.3-70-g09d2


From bcd61272db5e643b6d9c01c9d5085b914d9f19df Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arndb@de.ibm.com>
Date: Tue, 28 Jun 2005 15:58:50 -0700
Subject: [NET]: Add missing include to linux/netdevice.h

linux/etherdevice.h can't be included standalone at the moment, which
is required in order to sort the header files in the recommended
alphabetic order. This patch fixes that and is needed to build spider_net.

Signed-off-by: Arnd Bergmann <arndb@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/etherdevice.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h
index 8a2df4dfbc5..cf3847edc50 100644
--- a/include/linux/etherdevice.h
+++ b/include/linux/etherdevice.h
@@ -25,6 +25,7 @@
 #define _LINUX_ETHERDEVICE_H
 
 #include <linux/if_ether.h>
+#include <linux/netdevice.h>
 #include <linux/random.h>
 
 #ifdef __KERNEL__
-- 
cgit v1.2.3-70-g09d2


From 05133fc498e788e1c1ca4e906f9e05d9779fd63b Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Tue, 28 Jun 2005 20:44:54 -0700
Subject: [PATCH] swabb.h warning fixes

In file included from drivers/media/dvb/ttpci/av7110_hw.c:38:
include/linux/byteorder/swabb.h:96: warning: type qualifiers ignored on function return type
include/linux/byteorder/swabb.h:110: warning: type qualifiers ignored on function return type
In file included from drivers/media/dvb/ttpci/av7110_v4l.c:36:
include/linux/byteorder/swabb.h:96: warning: type qualifiers ignored on function return type
include/linux/byteorder/swabb.h:110: warning: type qualifiers ignored on function return type
In file included from drivers/media/dvb/ttpci/av7110_av.c:37:
include/linux/byteorder/swabb.h:96: warning: type qualifiers ignored on function return type
include/linux/byteorder/swabb.h:110: warning: type qualifiers ignored on function return type
drivers/isdn/icn/icn.c:719:4: warning: #warning TODO test headroom or use skb->nb to flag ACK
In file included from drivers/media/dvb/ttpci/av7110_ca.c:39:
include/linux/byteorder/swabb.h:96: warning: type qualifiers ignored on function return type
include/linux/byteorder/swabb.h:110: warning: type qualifiers ignored on function return type
In file included from drivers/media/dvb/ttpci/av7110.c:41:
include/linux/byteorder/swabb.h:96: warning: type qualifiers ignored on function return type
include/linux/byteorder/swabb.h:110: warning: type qualifiers ignored on function return type

Does declaring a function to return a const value actually mean something to
gcc?

Dunno.  Kill it and replace sone `__inline__'s with `inline' too.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/byteorder/swabb.h | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/byteorder/swabb.h b/include/linux/byteorder/swabb.h
index d28d9a804d3..d5f2a320510 100644
--- a/include/linux/byteorder/swabb.h
+++ b/include/linux/byteorder/swabb.h
@@ -92,29 +92,32 @@
 #endif /* OPTIMIZE */
 
 
-static __inline__ __const__ __u32 __fswahw32(__u32 x)
+static inline __u32 __fswahw32(__u32 x)
 {
 	return __arch__swahw32(x);
 }
-static __inline__ __u32 __swahw32p(__u32 *x)
+
+static inline __u32 __swahw32p(__u32 *x)
 {
 	return __arch__swahw32p(x);
 }
-static __inline__ void __swahw32s(__u32 *addr)
+
+static inline void __swahw32s(__u32 *addr)
 {
 	__arch__swahw32s(addr);
 }
 
-
-static __inline__ __const__ __u32 __fswahb32(__u32 x)
+static inline __u32 __fswahb32(__u32 x)
 {
 	return __arch__swahb32(x);
 }
-static __inline__ __u32 __swahb32p(__u32 *x)
+
+static inline __u32 __swahb32p(__u32 *x)
 {
 	return __arch__swahb32p(x);
 }
-static __inline__ void __swahb32s(__u32 *addr)
+
+static inline void __swahb32s(__u32 *addr)
 {
 	__arch__swahb32s(addr);
 }
-- 
cgit v1.2.3-70-g09d2


From 687a21cee17000177b1935896b9b475acf136678 Mon Sep 17 00:00:00 2001
From: Pekka J Enberg <penberg@cs.Helsinki.FI>
Date: Tue, 28 Jun 2005 20:44:55 -0700
Subject: [PATCH] rename wakeup_bdflush to wakeup_pdflush

Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/buffer.c               | 4 ++--
 include/linux/writeback.h | 2 +-
 mm/page-writeback.c       | 2 +-
 mm/vmscan.c               | 2 +-
 4 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/fs/buffer.c b/fs/buffer.c
index 13e5938a64f..561e63a1496 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -278,7 +278,7 @@ EXPORT_SYMBOL(thaw_bdev);
  */
 static void do_sync(unsigned long wait)
 {
-	wakeup_bdflush(0);
+	wakeup_pdflush(0);
 	sync_inodes(0);		/* All mappings, inodes and their blockdevs */
 	DQUOT_SYNC(NULL);
 	sync_supers();		/* Write the superblocks */
@@ -497,7 +497,7 @@ static void free_more_memory(void)
 	struct zone **zones;
 	pg_data_t *pgdat;
 
-	wakeup_bdflush(1024);
+	wakeup_pdflush(1024);
 	yield();
 
 	for_each_pgdat(pgdat) {
diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index d5c3fe1bf33..542dbaee651 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -85,7 +85,7 @@ static inline void wait_on_inode(struct inode *inode)
 /*
  * mm/page-writeback.c
  */
-int wakeup_bdflush(long nr_pages);
+int wakeup_pdflush(long nr_pages);
 void laptop_io_completion(void);
 void laptop_sync_completion(void);
 void throttle_vm_writeout(void);
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 613b99a5591..a6329fa8f86 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -354,7 +354,7 @@ static void background_writeout(unsigned long _min_pages)
  * the whole world.  Returns 0 if a pdflush thread was dispatched.  Returns
  * -1 if all pdflush threads were busy.
  */
-int wakeup_bdflush(long nr_pages)
+int wakeup_pdflush(long nr_pages)
 {
 	if (nr_pages == 0) {
 		struct writeback_state wbs;
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 1fa312a8db7..cfffe5098d5 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -972,7 +972,7 @@ int try_to_free_pages(struct zone **zones, unsigned int gfp_mask)
 		 * writeout.  So in laptop mode, write out the whole world.
 		 */
 		if (total_scanned > sc.swap_cluster_max + sc.swap_cluster_max/2) {
-			wakeup_bdflush(laptop_mode ? 0 : total_scanned);
+			wakeup_pdflush(laptop_mode ? 0 : total_scanned);
 			sc.may_writepage = 1;
 		}
 
-- 
cgit v1.2.3-70-g09d2


From 334a13ec3d01a1a4b4f2249735b793105cb4a519 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Tue, 28 Jun 2005 20:44:58 -0700
Subject: [PATCH] really remove xattr_acl.h

Looks like it sneaked back with the NFS ACL merge..

Signed-off-by: Christoph Hellwig <hch@lst.de>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfs/nfs3acl.c          | 14 ++++++-------
 fs/nfsd/vfs.c             | 13 ++++++------
 include/linux/xattr_acl.h | 50 -----------------------------------------------
 3 files changed, 13 insertions(+), 64 deletions(-)
 delete mode 100644 include/linux/xattr_acl.h

(limited to 'include')

diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c
index ee3536fc84a..1b7a3ef2f81 100644
--- a/fs/nfs/nfs3acl.c
+++ b/fs/nfs/nfs3acl.c
@@ -2,7 +2,7 @@
 #include <linux/nfs.h>
 #include <linux/nfs3.h>
 #include <linux/nfs_fs.h>
-#include <linux/xattr_acl.h>
+#include <linux/posix_acl_xattr.h>
 #include <linux/nfsacl.h>
 
 #define NFSDBG_FACILITY	NFSDBG_PROC
@@ -53,9 +53,9 @@ ssize_t nfs3_getxattr(struct dentry *dentry, const char *name,
 	struct posix_acl *acl;
 	int type, error = 0;
 
-	if (strcmp(name, XATTR_NAME_ACL_ACCESS) == 0)
+	if (strcmp(name, POSIX_ACL_XATTR_ACCESS) == 0)
 		type = ACL_TYPE_ACCESS;
-	else if (strcmp(name, XATTR_NAME_ACL_DEFAULT) == 0)
+	else if (strcmp(name, POSIX_ACL_XATTR_DEFAULT) == 0)
 		type = ACL_TYPE_DEFAULT;
 	else
 		return -EOPNOTSUPP;
@@ -82,9 +82,9 @@ int nfs3_setxattr(struct dentry *dentry, const char *name,
 	struct posix_acl *acl;
 	int type, error;
 
-	if (strcmp(name, XATTR_NAME_ACL_ACCESS) == 0)
+	if (strcmp(name, POSIX_ACL_XATTR_ACCESS) == 0)
 		type = ACL_TYPE_ACCESS;
-	else if (strcmp(name, XATTR_NAME_ACL_DEFAULT) == 0)
+	else if (strcmp(name, POSIX_ACL_XATTR_DEFAULT) == 0)
 		type = ACL_TYPE_DEFAULT;
 	else
 		return -EOPNOTSUPP;
@@ -103,9 +103,9 @@ int nfs3_removexattr(struct dentry *dentry, const char *name)
 	struct inode *inode = dentry->d_inode;
 	int type;
 
-	if (strcmp(name, XATTR_NAME_ACL_ACCESS) == 0)
+	if (strcmp(name, POSIX_ACL_XATTR_ACCESS) == 0)
 		type = ACL_TYPE_ACCESS;
-	else if (strcmp(name, XATTR_NAME_ACL_DEFAULT) == 0)
+	else if (strcmp(name, POSIX_ACL_XATTR_DEFAULT) == 0)
 		type = ACL_TYPE_DEFAULT;
 	else
 		return -EOPNOTSUPP;
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index de340ffd33c..be24ead89d9 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -46,10 +46,9 @@
 #include <linux/nfsd/nfsfh.h>
 #include <linux/quotaops.h>
 #include <linux/dnotify.h>
-#include <linux/xattr_acl.h>
 #include <linux/posix_acl.h>
-#ifdef CONFIG_NFSD_V4
 #include <linux/posix_acl_xattr.h>
+#ifdef CONFIG_NFSD_V4
 #include <linux/xattr.h>
 #include <linux/nfs4.h>
 #include <linux/nfs4_acl.h>
@@ -1872,10 +1871,10 @@ nfsd_get_posix_acl(struct svc_fh *fhp, int type)
 		return ERR_PTR(-EOPNOTSUPP);
 	switch(type) {
 		case ACL_TYPE_ACCESS:
-			name = XATTR_NAME_ACL_ACCESS;
+			name = POSIX_ACL_XATTR_ACCESS;
 			break;
 		case ACL_TYPE_DEFAULT:
-			name = XATTR_NAME_ACL_DEFAULT;
+			name = POSIX_ACL_XATTR_DEFAULT;
 			break;
 		default:
 			return ERR_PTR(-EOPNOTSUPP);
@@ -1919,17 +1918,17 @@ nfsd_set_posix_acl(struct svc_fh *fhp, int type, struct posix_acl *acl)
 		return -EOPNOTSUPP;
 	switch(type) {
 		case ACL_TYPE_ACCESS:
-			name = XATTR_NAME_ACL_ACCESS;
+			name = POSIX_ACL_XATTR_ACCESS;
 			break;
 		case ACL_TYPE_DEFAULT:
-			name = XATTR_NAME_ACL_DEFAULT;
+			name = POSIX_ACL_XATTR_DEFAULT;
 			break;
 		default:
 			return -EOPNOTSUPP;
 	}
 
 	if (acl && acl->a_count) {
-		size = xattr_acl_size(acl->a_count);
+		size = posix_acl_xattr_size(acl->a_count);
 		value = kmalloc(size, GFP_KERNEL);
 		if (!value)
 			return -ENOMEM;
diff --git a/include/linux/xattr_acl.h b/include/linux/xattr_acl.h
deleted file mode 100644
index 7a1f9b93a45..00000000000
--- a/include/linux/xattr_acl.h
+++ /dev/null
@@ -1,50 +0,0 @@
-/*
-  File: linux/xattr_acl.h
-
-  (extended attribute representation of access control lists)
-
-  (C) 2000 Andreas Gruenbacher, <a.gruenbacher@computer.org>
-*/
-
-#ifndef _LINUX_XATTR_ACL_H
-#define _LINUX_XATTR_ACL_H
-
-#include <linux/posix_acl.h>
-
-#define XATTR_NAME_ACL_ACCESS	"system.posix_acl_access"
-#define XATTR_NAME_ACL_DEFAULT	"system.posix_acl_default"
-
-#define XATTR_ACL_VERSION	0x0002
-
-typedef struct {
-	__u16		e_tag;
-	__u16		e_perm;
-	__u32		e_id;
-} xattr_acl_entry;
-
-typedef struct {
-	__u32		a_version;
-	xattr_acl_entry	a_entries[0];
-} xattr_acl_header;
-
-static inline size_t xattr_acl_size(int count)
-{
-	return sizeof(xattr_acl_header) + count * sizeof(xattr_acl_entry);
-}
-
-static inline int xattr_acl_count(size_t size)
-{
-	if (size < sizeof(xattr_acl_header))
-		return -1;
-	size -= sizeof(xattr_acl_header);
-	if (size % sizeof(xattr_acl_entry))
-		return -1;
-	return size / sizeof(xattr_acl_entry);
-}
-
-struct posix_acl * posix_acl_from_xattr(const void *value, size_t size);
-int posix_acl_to_xattr(const struct posix_acl *acl, void *buffer, size_t size);
-
-
-
-#endif /* _LINUX_XATTR_ACL_H */
-- 
cgit v1.2.3-70-g09d2


From 12822bc272e857544476cef9175462711899008b Mon Sep 17 00:00:00 2001
From: GOTO Masanori <gotom@debian.or.jp>
Date: Tue, 28 Jun 2005 20:45:03 -0700
Subject: [PATCH] headers: enable ppc64 ___arch__swab16 and ___arch__swab32

This patch cleans up asm-ppc64/byteorder.h to enable ___arch__swab16 and
___arch__swab32 which are marked TODO currently.  It removes ___arch__swab64
because ppc64 does not have short instruction combinations for swab64, the
recent gcc generates enough smart code that is equivalent to hand assembled
code under my tests.

Signed-off-by: GOTO Masanori <gotom@debian.or.jp>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-ppc64/byteorder.h | 10 ----------
 1 file changed, 10 deletions(-)

(limited to 'include')

diff --git a/include/asm-ppc64/byteorder.h b/include/asm-ppc64/byteorder.h
index 80327532de6..8b57da62b67 100644
--- a/include/asm-ppc64/byteorder.h
+++ b/include/asm-ppc64/byteorder.h
@@ -40,7 +40,6 @@ static __inline__ void st_le32(volatile __u32 *addr, const __u32 val)
 	__asm__ __volatile__ ("stwbrx %1,0,%2" : "=m" (*addr) : "r" (val), "r" (addr));
 }
 
-#if 0
 static __inline__ __attribute_const__ __u16 ___arch__swab16(__u16 value)
 {
 	__u16 result;
@@ -63,17 +62,8 @@ static __inline__ __attribute_const__ __u32 ___arch__swab32(__u32 value)
 	return result;
 }
 
-static __inline__ __attribute_const__ __u64 ___arch__swab64(__u64 value)
-{
-	__u64 result;
-#error implement me
-}
-
 #define __arch__swab16(x) ___arch__swab16(x)
 #define __arch__swab32(x) ___arch__swab32(x)
-#define __arch__swab64(x) ___arch__swab64(x)
-
-#endif
 
 /* The same, but returns converted value from the location pointer by addr. */
 #define __arch__swab16p(addr) ld_le16(addr)
-- 
cgit v1.2.3-70-g09d2


From 3607d1dfc80dcfbd3a6f236c70aa0d8eb7292278 Mon Sep 17 00:00:00 2001
From: GOTO Masanori <gotom@debian.or.jp>
Date: Tue, 28 Jun 2005 20:45:04 -0700
Subject: [PATCH] headers: include linux/compiler.h for __user

This patch lets i2c-dev.h include linux/compiler.h so that __user is defined.

Signed-off-by: GOTO Masanori <gotom@debian.or.jp>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/i2c-dev.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/linux/i2c-dev.h b/include/linux/i2c-dev.h
index d228230ffe5..54169567976 100644
--- a/include/linux/i2c-dev.h
+++ b/include/linux/i2c-dev.h
@@ -25,6 +25,7 @@
 #define _LINUX_I2C_DEV_H
 
 #include <linux/types.h>
+#include <linux/compiler.h>
 
 /* Some IOCTL commands are defined in <linux/i2c.h> */
 /* Note: 10-bit addresses are NOT supported! */
-- 
cgit v1.2.3-70-g09d2


From 4cceb4d13abaedbd52e54053367c793ed4aedb6b Mon Sep 17 00:00:00 2001
From: GOTO Masanori <gotom@debian.or.jp>
Date: Tue, 28 Jun 2005 20:45:05 -0700
Subject: [PATCH] headers: include linux/types.h for usb_ch9.h

This patch for usb_ch9.h includes linux/types.h instead of asm/types.h so that
__le16 and so on is explicitly defined.  It also cleans up non standard //
comment.

Signed-off-by: GOTO Masanori <gotom@debian.or.jp>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/usb_ch9.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/usb_ch9.h b/include/linux/usb_ch9.h
index 39e7ff4ffd2..ee21e6bf386 100644
--- a/include/linux/usb_ch9.h
+++ b/include/linux/usb_ch9.h
@@ -19,7 +19,7 @@
 #ifndef __LINUX_USB_CH9_H
 #define __LINUX_USB_CH9_H
 
-#include <asm/types.h>		/* __u8 etc */
+#include <linux/types.h>	/* __u8 etc */
 
 /*-------------------------------------------------------------------------*/
 
@@ -294,8 +294,8 @@ struct usb_endpoint_descriptor {
 	__le16 wMaxPacketSize;
 	__u8  bInterval;
 
-	// NOTE:  these two are _only_ in audio endpoints.
-	// use USB_DT_ENDPOINT*_SIZE in bLength, not sizeof.
+	/* NOTE:  these two are _only_ in audio endpoints. */
+	/* use USB_DT_ENDPOINT*_SIZE in bLength, not sizeof. */
 	__u8  bRefresh;
 	__u8  bSynchAddress;
 } __attribute__ ((packed));
-- 
cgit v1.2.3-70-g09d2


From fb3cc4320e1fd87143683b540e459a2e20fdc9bb Mon Sep 17 00:00:00 2001
From: Nick Piggin <nickpiggin@yahoo.com.au>
Date: Tue, 28 Jun 2005 20:45:15 -0700
Subject: [PATCH] blk: light iocontext ops

get_io_context needlessly turned off interrupts and checked for racing io
context creations.  Both of which aren't needed, because the io context can
only be created while in process context of the current process.

Also, split the function in 2.  A light version, current_io_context does not
elevate the reference count specifically, but can be used when in process
context, because the process holds a reference itself.

Signed-off-by: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Jens Axboe <axboe@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/ll_rw_blk.c | 56 +++++++++++++++++++++--------------------------
 include/linux/blkdev.h    |  1 +
 2 files changed, 26 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/drivers/block/ll_rw_blk.c b/drivers/block/ll_rw_blk.c
index 5caebe2cf0a..1197462bb6b 100644
--- a/drivers/block/ll_rw_blk.c
+++ b/drivers/block/ll_rw_blk.c
@@ -1876,7 +1876,7 @@ static struct request *get_request(request_queue_t *q, int rw, struct bio *bio,
 {
 	struct request *rq = NULL;
 	struct request_list *rl = &q->rq;
-	struct io_context *ioc = get_io_context(GFP_ATOMIC);
+	struct io_context *ioc = current_io_context(GFP_ATOMIC);
 
 	if (unlikely(test_bit(QUEUE_FLAG_DRAIN, &q->queue_flags)))
 		goto out;
@@ -1959,7 +1959,6 @@ rq_starved:
 	rq_init(q, rq);
 	rq->rl = rl;
 out:
-	put_io_context(ioc);
 	return rq;
 }
 
@@ -1997,9 +1996,8 @@ static struct request *get_request_wait(request_queue_t *q, int rw,
 			 * up to a big batch of them for a small period time.
 			 * See ioc_batching, ioc_set_batching
 			 */
-			ioc = get_io_context(GFP_NOIO);
+			ioc = current_io_context(GFP_NOIO);
 			ioc_set_batching(q, ioc);
-			put_io_context(ioc);
 
 			spin_lock_irq(q->queue_lock);
 		}
@@ -3282,24 +3280,20 @@ void exit_io_context(void)
 
 /*
  * If the current task has no IO context then create one and initialise it.
- * If it does have a context, take a ref on it.
+ * Otherwise, return its existing IO context.
  *
- * This is always called in the context of the task which submitted the I/O.
- * But weird things happen, so we disable local interrupts to ensure exclusive
- * access to *current.
+ * This returned IO context doesn't have a specifically elevated refcount,
+ * but since the current task itself holds a reference, the context can be
+ * used in general code, so long as it stays within `current` context.
  */
-struct io_context *get_io_context(int gfp_flags)
+struct io_context *current_io_context(int gfp_flags)
 {
 	struct task_struct *tsk = current;
-	unsigned long flags;
 	struct io_context *ret;
 
-	local_irq_save(flags);
 	ret = tsk->io_context;
-	if (ret)
-		goto out;
-
-	local_irq_restore(flags);
+	if (likely(ret))
+		return ret;
 
 	ret = kmem_cache_alloc(iocontext_cachep, gfp_flags);
 	if (ret) {
@@ -3310,25 +3304,25 @@ struct io_context *get_io_context(int gfp_flags)
 		ret->nr_batch_requests = 0; /* because this is 0 */
 		ret->aic = NULL;
 		ret->cic = NULL;
+		tsk->io_context = ret;
+	}
 
-		local_irq_save(flags);
-
-		/*
-		 * very unlikely, someone raced with us in setting up the task
-		 * io context. free new context and just grab a reference.
-		 */
-		if (!tsk->io_context)
-			tsk->io_context = ret;
-		else {
-			kmem_cache_free(iocontext_cachep, ret);
-			ret = tsk->io_context;
-		}
+	return ret;
+}
+EXPORT_SYMBOL(current_io_context);
 
-out:
+/*
+ * If the current task has no IO context then create one and initialise it.
+ * If it does have a context, take a ref on it.
+ *
+ * This is always called in the context of the task which submitted the I/O.
+ */
+struct io_context *get_io_context(int gfp_flags)
+{
+	struct io_context *ret;
+	ret = current_io_context(gfp_flags);
+	if (likely(ret))
 		atomic_inc(&ret->refcount);
-		local_irq_restore(flags);
-	}
-
 	return ret;
 }
 EXPORT_SYMBOL(get_io_context);
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 21a8674cd14..0881b5cdee3 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -96,6 +96,7 @@ struct io_context {
 
 void put_io_context(struct io_context *ioc);
 void exit_io_context(void);
+struct io_context *current_io_context(int gfp_flags);
 struct io_context *get_io_context(int gfp_flags);
 void copy_io_context(struct io_context **pdst, struct io_context **psrc);
 void swap_io_context(struct io_context **ioc1, struct io_context **ioc2);
-- 
cgit v1.2.3-70-g09d2


From 200803dfe4ff772740d63db725ab2f1b185ccf92 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Tue, 28 Jun 2005 20:45:18 -0700
Subject: [PATCH] irqpoll

Anyone reporting a stuck IRQ should try these options.  Its effectiveness
varies we've found in the Fedora case.  Quite a few systems with misdescribed
IRQ routing just work when you use irqpoll.  It also fixes up the VIA systems
although thats now fixed with the VIA quirk (which we could just make default
as its what Redmond OS does but Linus didn't like it historically).

A small number of systems have jammed IRQ sources or misdescribes that cause
an IRQ that we have no handler registered anywhere for.  In those cases it
doesn't help.

Signed-off-by: Alan Cox <number6@the-village.bc.nu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/kernel-parameters.txt |  15 +++++
 arch/ppc64/kernel/irq.c             |   2 +-
 include/linux/irq.h                 |   5 +-
 kernel/irq/handle.c                 |   2 +-
 kernel/irq/spurious.c               | 113 +++++++++++++++++++++++++++++++++++-
 5 files changed, 131 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 0f71251f12b..67e99f14419 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -437,6 +437,10 @@ running once the system is up.
 			Format: {"of[f]" | "sk[ipmbr]"}
 			See comment in arch/i386/boot/edd.S
 
+	edd		[EDD]
+			Format: {"of[f]" | "sk[ipmbr]"}
+			See comment in arch/i386/boot/edd.S
+
 	eicon=		[HW,ISDN] 
 			Format: <id>,<membase>,<irq>
 
@@ -622,6 +626,17 @@ running once the system is up.
 	ips=		[HW,SCSI] Adaptec / IBM ServeRAID controller
 			See header of drivers/scsi/ips.c.
 
+	irqfixup	[HW]
+			When an interrupt is not handled search all handlers
+			for it. Intended to get systems with badly broken
+			firmware running.
+
+	irqpoll		[HW]
+			When an interrupt is not handled search all handlers
+			for it. Also check all handlers each timer
+			interrupt. Intended to get systems with badly broken
+			firmware running.
+
 	isapnp=		[ISAPNP]
 			Format: <RDP>, <reset>, <pci_scan>, <verbosity>
 
diff --git a/arch/ppc64/kernel/irq.c b/arch/ppc64/kernel/irq.c
index 3defc8c33ad..ffe300611f0 100644
--- a/arch/ppc64/kernel/irq.c
+++ b/arch/ppc64/kernel/irq.c
@@ -245,7 +245,7 @@ void ppc_irq_dispatch_handler(struct pt_regs *regs, int irq)
 
 		spin_lock(&desc->lock);
 		if (!noirqdebug)
-			note_interrupt(irq, desc, action_ret);
+			note_interrupt(irq, desc, action_ret, regs);
 		if (likely(!(desc->status & IRQ_PENDING)))
 			break;
 		desc->status &= ~IRQ_PENDING;
diff --git a/include/linux/irq.h b/include/linux/irq.h
index 12277799c00..069d3b84d31 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -85,9 +85,10 @@ extern int no_irq_affinity;
 extern int noirqdebug_setup(char *str);
 
 extern fastcall int handle_IRQ_event(unsigned int irq, struct pt_regs *regs,
-				       struct irqaction *action);
+					struct irqaction *action);
 extern fastcall unsigned int __do_IRQ(unsigned int irq, struct pt_regs *regs);
-extern void note_interrupt(unsigned int irq, irq_desc_t *desc, int action_ret);
+extern void note_interrupt(unsigned int irq, irq_desc_t *desc,
+					int action_ret, struct pt_regs *regs);
 extern int can_request_irq(unsigned int irq, unsigned long irqflags);
 
 extern void init_irq_proc(void);
diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
index 436c7d93c00..c29f83c1649 100644
--- a/kernel/irq/handle.c
+++ b/kernel/irq/handle.c
@@ -172,7 +172,7 @@ fastcall unsigned int __do_IRQ(unsigned int irq, struct pt_regs *regs)
 
 		spin_lock(&desc->lock);
 		if (!noirqdebug)
-			note_interrupt(irq, desc, action_ret);
+			note_interrupt(irq, desc, action_ret, regs);
 		if (likely(!(desc->status & IRQ_PENDING)))
 			break;
 		desc->status &= ~IRQ_PENDING;
diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c
index ba039e827d5..7df9abd5ec8 100644
--- a/kernel/irq/spurious.c
+++ b/kernel/irq/spurious.c
@@ -11,6 +11,83 @@
 #include <linux/kallsyms.h>
 #include <linux/interrupt.h>
 
+static int irqfixup;
+
+/*
+ * Recovery handler for misrouted interrupts.
+ */
+
+static int misrouted_irq(int irq, struct pt_regs *regs)
+{
+	int i;
+	irq_desc_t *desc;
+	int ok = 0;
+	int work = 0;	/* Did we do work for a real IRQ */
+
+	for(i = 1; i < NR_IRQS; i++) {
+		struct irqaction *action;
+
+		if (i == irq)	/* Already tried */
+			continue;
+		desc = &irq_desc[i];
+		spin_lock(&desc->lock);
+		action = desc->action;
+		/* Already running on another processor */
+		if (desc->status & IRQ_INPROGRESS) {
+			/*
+			 * Already running: If it is shared get the other
+			 * CPU to go looking for our mystery interrupt too
+			 */
+			if (desc->action && (desc->action->flags & SA_SHIRQ))
+				desc->status |= IRQ_PENDING;
+			spin_unlock(&desc->lock);
+			continue;
+		}
+		/* Honour the normal IRQ locking */
+		desc->status |= IRQ_INPROGRESS;
+		spin_unlock(&desc->lock);
+		while (action) {
+			/* Only shared IRQ handlers are safe to call */
+			if (action->flags & SA_SHIRQ) {
+				if (action->handler(i, action->dev_id, regs) ==
+						IRQ_HANDLED)
+					ok = 1;
+			}
+			action = action->next;
+		}
+		local_irq_disable();
+		/* Now clean up the flags */
+		spin_lock(&desc->lock);
+		action = desc->action;
+
+		/*
+		 * While we were looking for a fixup someone queued a real
+		 * IRQ clashing with our walk
+		 */
+
+		while ((desc->status & IRQ_PENDING) && action) {
+			/*
+			 * Perform real IRQ processing for the IRQ we deferred
+			 */
+			work = 1;
+			spin_unlock(&desc->lock);
+			handle_IRQ_event(i, regs, action);
+			spin_lock(&desc->lock);
+			desc->status &= ~IRQ_PENDING;
+		}
+		desc->status &= ~IRQ_INPROGRESS;
+		/*
+		 * If we did actual work for the real IRQ line we must let the
+		 * IRQ controller clean up too
+		 */
+		if(work)
+			desc->handler->end(i);
+		spin_unlock(&desc->lock);
+	}
+	/* So the caller can adjust the irq error counts */
+	return ok;
+}
+
 /*
  * If 99,900 of the previous 100,000 interrupts have not been handled
  * then assume that the IRQ is stuck in some manner. Drop a diagnostic
@@ -31,7 +108,8 @@ __report_bad_irq(unsigned int irq, irq_desc_t *desc, irqreturn_t action_ret)
 		printk(KERN_ERR "irq event %d: bogus return value %x\n",
 				irq, action_ret);
 	} else {
-		printk(KERN_ERR "irq %d: nobody cared!\n", irq);
+		printk(KERN_ERR "irq %d: nobody cared (try booting with "
+				"the \"irqpoll\" option)\n", irq);
 	}
 	dump_stack();
 	printk(KERN_ERR "handlers:\n");
@@ -55,7 +133,8 @@ static void report_bad_irq(unsigned int irq, irq_desc_t *desc, irqreturn_t actio
 	}
 }
 
-void note_interrupt(unsigned int irq, irq_desc_t *desc, irqreturn_t action_ret)
+void note_interrupt(unsigned int irq, irq_desc_t *desc, irqreturn_t action_ret,
+			struct pt_regs *regs)
 {
 	if (action_ret != IRQ_HANDLED) {
 		desc->irqs_unhandled++;
@@ -63,6 +142,15 @@ void note_interrupt(unsigned int irq, irq_desc_t *desc, irqreturn_t action_ret)
 			report_bad_irq(irq, desc, action_ret);
 	}
 
+	if (unlikely(irqfixup)) {
+		/* Don't punish working computers */
+		if ((irqfixup == 2 && irq == 0) || action_ret == IRQ_NONE) {
+			int ok = misrouted_irq(irq, regs);
+			if (action_ret == IRQ_NONE)
+				desc->irqs_unhandled -= ok;
+		}
+	}
+
 	desc->irq_count++;
 	if (desc->irq_count < 100000)
 		return;
@@ -94,3 +182,24 @@ int __init noirqdebug_setup(char *str)
 
 __setup("noirqdebug", noirqdebug_setup);
 
+static int __init irqfixup_setup(char *str)
+{
+	irqfixup = 1;
+	printk(KERN_WARNING "Misrouted IRQ fixup support enabled.\n");
+	printk(KERN_WARNING "This may impact system performance.\n");
+	return 1;
+}
+
+__setup("irqfixup", irqfixup_setup);
+
+static int __init irqpoll_setup(char *str)
+{
+	irqfixup = 2;
+	printk(KERN_WARNING "Misrouted IRQ fixup and polling support "
+				"enabled\n");
+	printk(KERN_WARNING "This may significantly impact system "
+				"performance\n");
+	return 1;
+}
+
+__setup("irqpoll", irqpoll_setup);
-- 
cgit v1.2.3-70-g09d2


From 586b0cab2516640fec4dffc3049c4d8bca188f89 Mon Sep 17 00:00:00 2001
From: Mauro Carvalho Chehab <maurochehab@gmail.com>
Date: Tue, 28 Jun 2005 20:45:21 -0700
Subject: [PATCH] v4l: tuner improvements

*tuner-core.c:
- some tuner_info msgs will be generated only if insmod opt
        tuner_debug enabled.
- Implemented tuner-core support for VIDIO_S_TUNER to allow
        changing mono/stereo mode
- Remove unneeded config options.
- I2C_CLIENT_MULTI option removed.
- support for Philips FMD12ME hybrid tuner
- allow to initialize with another tuner
- Move PHILIPS_FMD initialization code to set_type function,

* tda8290:

- Fix dumb error in tda8290 tunning.
- Radio tuner uses high-precision step instead of 62.5 KHz.

*tea5767.c:
- tuner_info msgs will be generated only if insmod tuner option
        tuner_debug enabled.
- some cleanups for better reading.
- Radio tuner uses high-precision step instead of 62.5 KHz.
- Changing radio mode stereo/mono for tea5767 working.

*tuner-simple.c:
- TNF9533-D/IF UHF fixup.
- Radio tuners now uses high-precision step instead of 62.5 KHz.

*mt20xx.c:
        - Radio tuner uses high-precision step instead of 62.5 KHz.

*tda9887.c:
        - tab and blank spaces corrections.

Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
Signed-off-by: Gerd Knorr <kraxel@bytesex.org>
Signed-off-by: Nickolay V Shmyrev <nshmyrev@yandex.ru>
Signed-off-by: Hartmut Hackmann <hartmut.hackmann@t-online.de>
Signed-off-by: Michael Krufky <mkrufky@m1k.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/media/video/Makefile       |   3 +-
 drivers/media/video/mt20xx.c       |   6 +-
 drivers/media/video/tda8290.c      |  20 ++-
 drivers/media/video/tda9887.c      |   7 +-
 drivers/media/video/tea5767.c      | 334 +++++++++++++++++++++++++++++++++++++
 drivers/media/video/tuner-core.c   | 163 +++++++++---------
 drivers/media/video/tuner-simple.c |  34 ++--
 include/media/tuner.h              |  20 ++-
 8 files changed, 471 insertions(+), 116 deletions(-)
 create mode 100644 drivers/media/video/tea5767.c

(limited to 'include')

diff --git a/drivers/media/video/Makefile b/drivers/media/video/Makefile
index 2dc906fdfa5..810e7aac0a5 100644
--- a/drivers/media/video/Makefile
+++ b/drivers/media/video/Makefile
@@ -7,8 +7,7 @@ bttv-objs	:=	bttv-driver.o bttv-cards.o bttv-if.o \
 zoran-objs      :=	zr36120.o zr36120_i2c.o zr36120_mem.o
 zr36067-objs	:=	zoran_procfs.o zoran_device.o \
 			zoran_driver.o zoran_card.o
-tuner-objs	:=	tuner-core.o tuner-simple.o mt20xx.o tda8290.o
-
+tuner-objs	:=	tuner-core.o tuner-simple.o mt20xx.o tda8290.o tea5767.o
 obj-$(CONFIG_VIDEO_DEV) += videodev.o v4l2-common.o v4l1-compat.o
 
 obj-$(CONFIG_VIDEO_BT848) += bttv.o msp3400.o tvaudio.o \
diff --git a/drivers/media/video/mt20xx.c b/drivers/media/video/mt20xx.c
index 95ad17b7f38..9c005cb128d 100644
--- a/drivers/media/video/mt20xx.c
+++ b/drivers/media/video/mt20xx.c
@@ -1,5 +1,5 @@
 /*
- * $Id: mt20xx.c,v 1.4 2005/03/04 09:24:56 kraxel Exp $
+ * $Id: mt20xx.c,v 1.5 2005/06/16 08:29:49 nsh Exp $
  *
  * i2c tv tuner chip device driver
  * controls microtune tuners, mt2032 + mt2050 at the moment.
@@ -295,8 +295,8 @@ static void mt2032_set_radio_freq(struct i2c_client *c, unsigned int freq)
 	int if2 = t->radio_if2;
 
 	// per Manual for FM tuning: first if center freq. 1085 MHz
-        mt2032_set_if_freq(c, freq*62500 /* freq*1000*1000/16 */,
-			   1085*1000*1000,if2,if2,if2);
+        mt2032_set_if_freq(c, freq * 1000 / 16,
+			      1085*1000*1000,if2,if2,if2);
 }
 
 // Initalization as described in "MT203x Programming Procedures", Rev 1.2, Feb.2001
diff --git a/drivers/media/video/tda8290.c b/drivers/media/video/tda8290.c
index b27cc348d95..f59d4601cc6 100644
--- a/drivers/media/video/tda8290.c
+++ b/drivers/media/video/tda8290.c
@@ -1,5 +1,5 @@
 /*
- * $Id: tda8290.c,v 1.7 2005/03/07 12:01:51 kraxel Exp $
+ * $Id: tda8290.c,v 1.11 2005/06/18 06:09:06 nsh Exp $
  *
  * i2c tv tuner chip device driver
  * controls the philips tda8290+75 tuner chip combo.
@@ -69,7 +69,7 @@ static __u8 get_freq_entry( struct freq_entry* table, __u16 freq)
 static unsigned char i2c_enable_bridge[2] = 	{ 0x21, 0xC0 };
 static unsigned char i2c_disable_bridge[2] = 	{ 0x21, 0x80 };
 static unsigned char i2c_init_tda8275[14] = 	{ 0x00, 0x00, 0x00, 0x00,
-						  0x7C, 0x04, 0xA3, 0x3F,
+						  0xfC, 0x04, 0xA3, 0x3F,
 						  0x2A, 0x04, 0xFF, 0x00,
 						  0x00, 0x40 };
 static unsigned char i2c_set_VS[2] = 		{ 0x30, 0x6F };
@@ -138,16 +138,24 @@ static int tda8290_tune(struct i2c_client *c)
 
 static void set_frequency(struct tuner *t, u16 ifc)
 {
-	u32 N = (((t->freq<<3)+ifc)&0x3fffc);
+	u32 freq;
+	u32 N;
 
-	N = N >> get_freq_entry(div_table, t->freq);
+	if (t->mode == V4L2_TUNER_RADIO)
+		freq = t->freq / 1000;
+	else
+		freq = t->freq;
+
+	N = (((freq<<3)+ifc)&0x3fffc);
+
+	N = N >> get_freq_entry(div_table, freq);
 	t->i2c_set_freq[0] = 0;
 	t->i2c_set_freq[1] = (unsigned char)(N>>8);
 	t->i2c_set_freq[2] = (unsigned char) N;
 	t->i2c_set_freq[3] = 0x40;
 	t->i2c_set_freq[4] = 0x52;
-	t->i2c_set_freq[5] = get_freq_entry(band_table, t->freq);
-	t->i2c_set_freq[6] = get_freq_entry(agc_table,  t->freq);
+	t->i2c_set_freq[5] = get_freq_entry(band_table, freq);
+	t->i2c_set_freq[6] = get_freq_entry(agc_table,  freq);
 	t->i2c_set_freq[7] = 0x8f;
 }
 
diff --git a/drivers/media/video/tda9887.c b/drivers/media/video/tda9887.c
index 39773633cc3..ee35562f4d1 100644
--- a/drivers/media/video/tda9887.c
+++ b/drivers/media/video/tda9887.c
@@ -368,7 +368,7 @@ static int tda9887_set_tvnorm(struct tda9887 *t, char *buf)
 		if (t->radio_mode == V4L2_TUNER_MODE_MONO)
 			norm = &radio_mono;
 		else
-		        norm = &radio_stereo;
+			norm = &radio_stereo;
 	} else {
 		for (i = 0; i < ARRAY_SIZE(tvnorms); i++) {
 			if (tvnorms[i].std & t->std) {
@@ -566,7 +566,6 @@ static int tda9887_configure(struct tda9887 *t)
 	if (UNSET != t->pinnacle_id) {
 		tda9887_set_pinnacle(t,buf);
 	}
-
 	tda9887_set_config(t,buf);
 	tda9887_set_insmod(t,buf);
 
@@ -615,8 +614,8 @@ static int tda9887_attach(struct i2c_adapter *adap, int addr, int kind)
 	t->pinnacle_id = UNSET;
 	t->radio_mode = V4L2_TUNER_MODE_STEREO;
 
-        i2c_set_clientdata(&t->client, t);
-        i2c_attach_client(&t->client);
+	i2c_set_clientdata(&t->client, t);
+	i2c_attach_client(&t->client);
 
 	return 0;
 }
diff --git a/drivers/media/video/tea5767.c b/drivers/media/video/tea5767.c
new file mode 100644
index 00000000000..a29f08f81f6
--- /dev/null
+++ b/drivers/media/video/tea5767.c
@@ -0,0 +1,334 @@
+/*
+ * For Philips TEA5767 FM Chip used on some TV Cards like Prolink Pixelview
+ * I2C address is allways 0xC0.
+ *
+ * $Id: tea5767.c,v 1.11 2005/06/21 15:40:33 mchehab Exp $
+ *
+ * Copyright (c) 2005 Mauro Carvalho Chehab (mchehab@brturbo.com.br)
+ * This code is placed under the terms of the GNU General Public License
+ *
+ * tea5767 autodetection thanks to Torsten Seeboth and Atsushi Nakagawa
+ * from their contributions on DScaler.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/string.h>
+#include <linux/timer.h>
+#include <linux/delay.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/videodev.h>
+#include <linux/i2c.h>
+#include <linux/i2c-algo-bit.h>
+
+#include <media/tuner.h>
+
+/* Declared at tuner-core.c */
+extern unsigned int tuner_debug;
+
+#define PREFIX "TEA5767 "
+
+/*****************************************************************************/
+
+/******************************
+ * Write mode register values *
+ ******************************/
+
+/* First register */
+#define TEA5767_MUTE		0x80 /* Mutes output */
+#define TEA5767_SEARCH		0x40 /* Activates station search */
+/* Bits 0-5 for divider MSB */
+
+/* Second register */
+/* Bits 0-7 for divider LSB */
+
+/* Third register */
+
+/* Station search from botton to up */
+#define TEA5767_SEARCH_UP	0x80
+
+/* Searches with ADC output = 10 */
+#define TEA5767_SRCH_HIGH_LVL	0x60
+
+/* Searches with ADC output = 10 */
+#define TEA5767_SRCH_MID_LVL	0x40
+
+/* Searches with ADC output = 5 */
+#define TEA5767_SRCH_LOW_LVL	0x20
+
+/* if on, div=4*(Frf+Fif)/Fref otherwise, div=4*(Frf-Fif)/Freq) */
+#define TEA5767_HIGH_LO_INJECT	0x10
+
+/* Disable stereo */
+#define TEA5767_MONO		0x08
+
+/* Disable right channel and turns to mono */
+#define TEA5767_MUTE_RIGHT	0x04
+
+/* Disable left channel and turns to mono */
+#define TEA5767_MUTE_LEFT	0x02
+
+#define TEA5767_PORT1_HIGH	0x01
+
+/* Forth register */
+#define TEA5767_PORT2_HIGH	0x80
+/* Chips stops working. Only I2C bus remains on */
+#define TEA5767_STDBY		0x40
+
+/* Japan freq (76-108 MHz. If disabled, 87.5-108 MHz */
+#define TEA5767_JAPAN_BAND	0x20
+
+/* Unselected means 32.768 KHz freq as reference. Otherwise Xtal at 13 MHz */
+#define TEA5767_XTAL_32768	0x10
+
+/* Cuts weak signals */
+#define TEA5767_SOFT_MUTE	0x08
+
+/* Activates high cut control */
+#define TEA5767_HIGH_CUT_CTRL	0x04
+
+/* Activates stereo noise control */
+#define TEA5767_ST_NOISE_CTL	0x02
+
+/* If activate PORT 1 indicates SEARCH or else it is used as PORT1 */
+#define TEA5767_SRCH_IND	0x01
+
+/* Fiveth register */
+
+/* By activating, it will use Xtal at 13 MHz as reference for divider */
+#define TEA5767_PLLREF_ENABLE	0x80
+
+/* By activating, deemphasis=50, or else, deemphasis of 50us */
+#define TEA5767_DEEMPH_75	0X40
+
+/*****************************
+ * Read mode register values *
+ *****************************/
+
+/* First register */
+#define TEA5767_READY_FLAG_MASK	0x80
+#define TEA5767_BAND_LIMIT_MASK	0X40
+/* Bits 0-5 for divider MSB after search or preset */
+
+/* Second register */
+/* Bits 0-7 for divider LSB after search or preset */
+
+/* Third register */
+#define TEA5767_STEREO_MASK	0x80
+#define TEA5767_IF_CNTR_MASK	0x7f
+
+/* Four register */
+#define TEA5767_ADC_LEVEL_MASK	0xf0
+
+/* should be 0 */
+#define TEA5767_CHIP_ID_MASK	0x0f
+
+/* Fiveth register */
+/* Reserved for future extensions */
+#define TEA5767_RESERVED_MASK	0xff
+
+/*****************************************************************************/
+
+static void set_tv_freq(struct i2c_client *c, unsigned int freq)
+{
+	struct tuner *t = i2c_get_clientdata(c);
+
+	tuner_warn("This tuner doesn't support TV freq.\n");
+}
+
+static void tea5767_status_dump(unsigned char *buffer)
+{
+	unsigned int div, frq;
+
+	if (TEA5767_READY_FLAG_MASK & buffer[0])
+		printk(PREFIX "Ready Flag ON\n");
+	else
+		printk(PREFIX "Ready Flag OFF\n");
+
+	if (TEA5767_BAND_LIMIT_MASK & buffer[0])
+		printk(PREFIX "Tuner at band limit\n");
+	else
+		printk(PREFIX "Tuner not at band limit\n");
+
+	div=((buffer[0]&0x3f)<<8) | buffer[1];
+
+	switch (TEA5767_HIGH_LO_32768) {
+	case TEA5767_HIGH_LO_13MHz:
+		frq = 1000*(div*50-700-225)/4; /* Freq in KHz */
+		break;
+	case TEA5767_LOW_LO_13MHz:
+		frq = 1000*(div*50+700+225)/4; /* Freq in KHz */
+		break;
+	case TEA5767_LOW_LO_32768:
+		frq = 1000*(div*32768/1000+700+225)/4; /* Freq in KHz */
+		break;
+	case TEA5767_HIGH_LO_32768:
+	default:
+		frq = 1000*(div*32768/1000-700-225)/4; /* Freq in KHz */
+		break;
+	}
+        buffer[0] = (div>>8) & 0x3f;
+        buffer[1] = div      & 0xff;
+
+	printk(PREFIX "Frequency %d.%03d KHz (divider = 0x%04x)\n",
+						frq/1000,frq%1000,div);
+
+	if (TEA5767_STEREO_MASK & buffer[2])
+		printk(PREFIX "Stereo\n");
+	else
+		printk(PREFIX "Mono\n");
+
+	printk(PREFIX "IF Counter = %d\n",buffer[2] & TEA5767_IF_CNTR_MASK);
+
+	printk(PREFIX "ADC Level = %d\n",(buffer[3] & TEA5767_ADC_LEVEL_MASK)>>4);
+
+	printk(PREFIX "Chip ID = %d\n",(buffer[3] & TEA5767_CHIP_ID_MASK));
+
+	printk(PREFIX "Reserved = 0x%02x\n",(buffer[4] & TEA5767_RESERVED_MASK));
+}
+
+/* Freq should be specifyed at 62.5 Hz */
+static void set_radio_freq(struct i2c_client *c, unsigned int frq)
+{
+	struct tuner *t = i2c_get_clientdata(c);
+        unsigned char buffer[5];
+	unsigned div;
+	int rc;
+
+	if ( tuner_debug )
+		printk(PREFIX "radio freq counter %d\n",frq);
+
+	/* Rounds freq to next decimal value - for 62.5 KHz step */
+	/* frq = 20*(frq/16)+radio_frq[frq%16]; */
+
+	buffer[2] = TEA5767_PORT1_HIGH;
+	buffer[3] = TEA5767_PORT2_HIGH | TEA5767_HIGH_CUT_CTRL | TEA5767_ST_NOISE_CTL | TEA5767_JAPAN_BAND;
+	buffer[4]=0;
+
+	if (t->audmode == V4L2_TUNER_MODE_MONO) {
+		tuner_dbg("TEA5767 set to mono\n");
+		buffer[2] |= TEA5767_MONO;
+	} else
+ 		tuner_dbg("TEA5767 set to stereo\n");
+
+	switch (t->type) {
+	case TEA5767_HIGH_LO_13MHz:
+		tuner_dbg("TEA5767 radio HIGH LO inject xtal @ 13 MHz\n");
+		buffer[2] |= TEA5767_HIGH_LO_INJECT;
+		buffer[4] |= TEA5767_PLLREF_ENABLE;
+		div = (frq*4/16+700+225+25)/50;
+		break;
+	case TEA5767_LOW_LO_13MHz:
+		tuner_dbg("TEA5767 radio LOW LO inject xtal @ 13 MHz\n");
+
+		buffer[4] |= TEA5767_PLLREF_ENABLE;
+		div = (frq*4/16-700-225+25)/50;
+		break;
+	case TEA5767_LOW_LO_32768:
+		tuner_dbg("TEA5767 radio LOW LO inject xtal @ 32,768 MHz\n");
+		buffer[3] |= TEA5767_XTAL_32768;
+		/* const 700=4000*175 Khz - to adjust freq to right value */
+		div = (1000*(frq*4/16-700-225)+16384)>>15;
+		break;
+	case TEA5767_HIGH_LO_32768:
+	default:
+		tuner_dbg("TEA5767 radio HIGH LO inject xtal @ 32,768 MHz\n");
+
+		buffer[2] |= TEA5767_HIGH_LO_INJECT;
+		buffer[3] |= TEA5767_XTAL_32768;
+		div = (1000*(frq*4/16+700+225)+16384)>>15;
+		break;
+	}
+        buffer[0] = (div>>8) & 0x3f;
+        buffer[1] = div      & 0xff;
+
+	if ( tuner_debug )
+		tea5767_status_dump(buffer);
+
+        if (5 != (rc = i2c_master_send(c,buffer,5)))
+		tuner_warn("i2c i/o error: rc == %d (should be 5)\n",rc);
+}
+
+static int tea5767_signal(struct i2c_client *c)
+{
+	unsigned char buffer[5];
+	int rc;
+	struct tuner *t = i2c_get_clientdata(c);
+
+	memset(buffer,0,sizeof(buffer));
+        if (5 != (rc = i2c_master_recv(c,buffer,5)))
+                tuner_warn ( "i2c i/o error: rc == %d (should be 5)\n",rc);
+
+	return ((buffer[3] & TEA5767_ADC_LEVEL_MASK) <<(13-4));
+}
+
+static int tea5767_stereo(struct i2c_client *c)
+{
+	unsigned char buffer[5];
+	int rc;
+	struct tuner *t = i2c_get_clientdata(c);
+
+	memset(buffer,0,sizeof(buffer));
+        if (5 != (rc = i2c_master_recv(c,buffer,5)))
+                tuner_warn ( "i2c i/o error: rc == %d (should be 5)\n",rc);
+
+	rc = buffer[2] & TEA5767_STEREO_MASK;
+
+	if ( tuner_debug )
+		tuner_dbg("TEA5767 radio ST GET = %02x\n", rc);
+
+	return ( (buffer[2] & TEA5767_STEREO_MASK) ? V4L2_TUNER_SUB_STEREO: 0);
+}
+
+int tea_detection(struct i2c_client *c)
+{
+	unsigned char buffer[5]= { 0xff, 0xff, 0xff, 0xff, 0xff };
+	int rc;
+	struct tuner *t = i2c_get_clientdata(c);
+
+        if (5 != (rc = i2c_master_recv(c,buffer,5))) {
+                tuner_warn ( "it is not a TEA5767. Received %i chars.\n",rc );
+		return EINVAL;
+	}
+
+	/* If all bytes are the same then it's a TV tuner and not a tea5767 chip. */
+	if (buffer[0] == buffer[1] &&  buffer[0] == buffer[2] &&
+	    buffer[0] == buffer[3] &&  buffer[0] == buffer[4]) {
+                tuner_warn ( "All bytes are equal. It is not a TEA5767\n" );
+		return EINVAL;
+	}
+
+	/*  Status bytes:
+	 *  Byte 4: bit 3:1 : CI (Chip Identification) == 0
+	 *	    bit 0   : internally set to 0
+	 *  Byte 5: bit 7:0 : == 0
+	 */
+
+	if (!((buffer[3] & 0x0f) == 0x00) && (buffer[4] == 0x00)) {
+                tuner_warn ( "Chip ID is not zero. It is not a TEA5767\n" );
+		return EINVAL;
+	}
+	tuner_warn ( "TEA5767 detected.\n" );
+	return 0;
+}
+
+int tea5767_tuner_init(struct i2c_client *c)
+{
+	struct tuner *t = i2c_get_clientdata(c);
+
+	if (tea_detection(c)==EINVAL) return EINVAL;
+
+        tuner_info("type set to %d (%s)\n",
+                   t->type, TEA5767_TUNER_NAME);
+        strlcpy(c->name, TEA5767_TUNER_NAME, sizeof(c->name));
+
+	t->tv_freq    = set_tv_freq;
+	t->radio_freq = set_radio_freq;
+	t->has_signal = tea5767_signal;
+	t->is_stereo  = tea5767_stereo;
+
+	return (0);
+}
diff --git a/drivers/media/video/tuner-core.c b/drivers/media/video/tuner-core.c
index eaabfc85870..6f6bf4a633f 100644
--- a/drivers/media/video/tuner-core.c
+++ b/drivers/media/video/tuner-core.c
@@ -1,5 +1,5 @@
 /*
- * $Id: tuner-core.c,v 1.15 2005/06/12 01:36:14 mchehab Exp $
+ * $Id: tuner-core.c,v 1.29 2005/06/21 15:40:33 mchehab Exp $
  *
  * i2c tv tuner chip device driver
  * core core, i.e. kernel interfaces, registering and so on
@@ -26,7 +26,6 @@
 /*
  * comment line bellow to return to old behavor, where only one I2C device is supported
  */
-#define CONFIG_TUNER_MULTI_I2C /**/
 
 #define UNSET (-1U)
 
@@ -58,9 +57,7 @@ MODULE_AUTHOR("Ralph Metzler, Gerd Knorr, Gunther Mayer");
 MODULE_LICENSE("GPL");
 
 static int this_adap;
-#ifdef CONFIG_TUNER_MULTI_I2C
 static unsigned short first_tuner, tv_tuner, radio_tuner;
-#endif
 
 static struct i2c_driver driver;
 static struct i2c_client client_template;
@@ -81,26 +78,9 @@ static void set_tv_freq(struct i2c_client *c, unsigned int freq)
 		return;
 	}
 	if (freq < tv_range[0]*16 || freq > tv_range[1]*16) {
-
-		if (freq >= tv_range[0]*16364 && freq <= tv_range[1]*16384) {
-			/* V4L2_TUNER_CAP_LOW frequency */
-
-			tuner_dbg("V4L2_TUNER_CAP_LOW freq selected for TV. Tuners yet doesn't support converting it to valid freq.\n");
-
-			t->tv_freq(c,freq>>10);
-
-			return;
-                } else {
-			/* FIXME: better do that chip-specific, but
-			   right now we don't have that in the config
-			   struct and this way is still better than no
-			   check at all */
 			tuner_info("TV freq (%d.%02d) out of range (%d-%d)\n",
 				   freq/16,freq%16*100/16,tv_range[0],tv_range[1]);
-			return;
-		}
 	}
-	tuner_dbg("62.5 Khz freq step selected for TV.\n");
 	t->tv_freq(c,freq);
 }
 
@@ -116,31 +96,18 @@ static void set_radio_freq(struct i2c_client *c, unsigned int freq)
 		tuner_info("no radio tuning for this one, sorry.\n");
 		return;
 	}
-	if (freq < radio_range[0]*16 || freq > radio_range[1]*16) {
-		if (freq >= tv_range[0]*16364 && freq <= tv_range[1]*16384) {
-			/* V4L2_TUNER_CAP_LOW frequency */
-			if (t->type == TUNER_TEA5767) {
-				tuner_info("radio freq step 62.5Hz (%d.%06d)\n",(freq>>14),freq%(1<<14)*10000);
-				t->radio_freq(c,freq>>10);
-				return;
-			}
-
-			tuner_dbg("V4L2_TUNER_CAP_LOW freq selected for Radio. Tuners yet doesn't support converting it to valid freq.\n");
-
-			tuner_info("radio freq (%d.%06d)\n",(freq>>14),freq%(1<<14)*10000);
-
-			t->radio_freq(c,freq>>10);
-			return;
-
-                } else {
-			tuner_info("radio freq (%d.%02d) out of range (%d-%d)\n",
-			   freq/16,freq%16*100/16,
-				   radio_range[0],radio_range[1]);
-			return;
-		}
+	if (freq >= radio_range[0]*16000 && freq <= radio_range[1]*16000) {
+		if (tuner_debug)
+			tuner_info("radio freq step 62.5Hz (%d.%06d)\n",
+			 		    freq/16000,freq%16000*1000/16);
+		t->radio_freq(c,freq);
+        } else {
+		tuner_info("radio freq (%d.%02d) out of range (%d-%d)\n",
+			    freq/16,freq%16*100/16,
+			    radio_range[0],radio_range[1]);
 	}
-	tuner_dbg("62.5 Khz freq step selected for Radio.\n");
-	t->radio_freq(c,freq);
+
+	return;
 }
 
 static void set_freq(struct i2c_client *c, unsigned long freq)
@@ -166,8 +133,8 @@ static void set_freq(struct i2c_client *c, unsigned long freq)
 static void set_type(struct i2c_client *c, unsigned int type)
 {
 	struct tuner *t = i2c_get_clientdata(c);
+	unsigned char buffer[4];
 
-	tuner_dbg ("I2C addr 0x%02x with type %d\n",c->addr<<1,type);
 	/* sanity check */
 	if (type == UNSET || type == TUNER_ABSENT)
 		return;
@@ -179,8 +146,8 @@ static void set_type(struct i2c_client *c, unsigned int type)
 		t->type = type;
 		return;
 	}
-	if (t->initialized)
-		/* run only once */
+	if ((t->initialized) && (t->type == type))
+		/* run only once except type change  Hac 04/05*/
 		return;
 
 	t->initialized = 1;
@@ -193,25 +160,42 @@ static void set_type(struct i2c_client *c, unsigned int type)
 	case TUNER_PHILIPS_TDA8290:
 		tda8290_init(c);
 		break;
+	case TUNER_TEA5767:
+		if (tea5767_tuner_init(c)==EINVAL) t->type=TUNER_ABSENT;
+		break;
+	case TUNER_PHILIPS_FMD1216ME_MK3:
+		buffer[0] = 0x0b;
+		buffer[1] = 0xdc;
+		buffer[2] = 0x9c;
+		buffer[3] = 0x60;
+    		i2c_master_send(c,buffer,4);
+		mdelay(1);
+		buffer[2] = 0x86;
+		buffer[3] = 0x54;
+    		i2c_master_send(c,buffer,4);
+		default_tuner_init(c);
+		break;
 	default:
+		/* TEA5767 autodetection code */
+			if (tea5767_tuner_init(c)!=EINVAL) {
+				t->type = TUNER_TEA5767;
+				if (first_tuner == 0x60)
+					first_tuner++;
+				break;
+			}
+
 		default_tuner_init(c);
 		break;
 	}
+	tuner_dbg ("I2C addr 0x%02x with type %d\n",c->addr<<1,type);
 }
 
-#ifdef CONFIG_TUNER_MULTI_I2C
 #define CHECK_ADDR(tp,cmd,tun)	if (client->addr!=tp) { \
-			  return 0; } else \
+			  return 0; } else if (tuner_debug) \
 			  tuner_info ("Cmd %s accepted to "tun"\n",cmd);
 #define CHECK_MODE(cmd)	if (t->mode == V4L2_TUNER_RADIO) { \
 		 	CHECK_ADDR(radio_tuner,cmd,"radio") } else \
 			{ CHECK_ADDR(tv_tuner,cmd,"TV"); }
-#else
-#define CHECK_ADDR(tp,cmd,tun) tuner_info ("Cmd %s accepted to "tun"\n",cmd);
-#define CHECK_MODE(cmd) tuner_info ("Cmd %s accepted\n",cmd);
-#endif
-
-#ifdef CONFIG_TUNER_MULTI_I2C
 
 static void set_addr(struct i2c_client *c, struct tuner_addr *tun_addr)
 {
@@ -242,9 +226,6 @@ static void set_addr(struct i2c_client *c, struct tuner_addr *tun_addr)
 	}
 	set_type(c,tun_addr->type);
 }
-#else
-#define set_addr(c,tun_addr) set_type(c,(tun_addr)->type)
-#endif
 
 static char pal[] = "-";
 module_param_string(pal, pal, sizeof(pal), 0644);
@@ -284,17 +265,12 @@ static int tuner_attach(struct i2c_adapter *adap, int addr, int kind)
 {
 	struct tuner *t;
 
-#ifndef CONFIG_TUNER_MULTI_I2C
-	if (this_adap > 0)
-		return -1;
-#else
 	/* by default, first I2C card is both tv and radio tuner */
 	if (this_adap == 0) {
 		first_tuner = addr;
 		tv_tuner = addr;
 		radio_tuner = addr;
 	}
-#endif
 	this_adap++;
 
         client_template.adapter = adap;
@@ -308,6 +284,7 @@ static int tuner_attach(struct i2c_adapter *adap, int addr, int kind)
 	i2c_set_clientdata(&t->i2c, t);
 	t->type       = UNSET;
 	t->radio_if2  = 10700*1000; /* 10.7MHz - FM radio */
+	t->audmode    = V4L2_TUNER_MODE_STEREO;
 
         i2c_attach_client(&t->i2c);
 	tuner_info("chip found @ 0x%x (%s)\n",
@@ -325,11 +302,9 @@ static int tuner_probe(struct i2c_adapter *adap)
 	}
 	this_adap = 0;
 
-#ifdef CONFIG_TUNER_MULTI_I2C
 	first_tuner = 0;
 	tv_tuner = 0;
 	radio_tuner = 0;
-#endif
 
 	if (adap->class & I2C_CLASS_TV_ANALOG)
 		return i2c_probe(adap, &addr_data, tuner_attach);
@@ -392,8 +367,7 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 			t->radio_if2 = 41300 * 1000;
 			break;
 		}
-                break;
-
+		break;
 	/* --- v4l ioctls --- */
 	/* take care: bttv does userspace copying, we'll get a
 	   kernel pointer here... */
@@ -440,11 +414,18 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 				vt->signal = t->has_signal(client);
 			if (t->is_stereo) {
 				if (t->is_stereo(client))
-					vt-> flags |= VIDEO_TUNER_STEREO_ON;
+					vt->flags |= VIDEO_TUNER_STEREO_ON;
 				else
-					vt-> flags &= 0xffff ^ VIDEO_TUNER_STEREO_ON;
+					vt->flags &= ~VIDEO_TUNER_STEREO_ON;
 			}
 			vt->flags |= V4L2_TUNER_CAP_LOW; /* Allow freqs at 62.5 Hz */
+
+			vt->rangelow = radio_range[0] * 16000;
+			vt->rangehigh = radio_range[1] * 16000;
+
+		} else {
+			vt->rangelow = tv_range[0] * 16;
+			vt->rangehigh = tv_range[1] * 16;
 		}
 
 		return 0;
@@ -510,20 +491,46 @@ tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 				tuner -> signal = t->has_signal(client);
 			if (t->is_stereo) {
 				if (t->is_stereo(client)) {
-					tuner -> capability |= V4L2_TUNER_CAP_STEREO;
-					tuner -> rxsubchans |= V4L2_TUNER_SUB_STEREO;
+					tuner -> rxsubchans = V4L2_TUNER_SUB_STEREO | V4L2_TUNER_SUB_MONO;
 				} else {
-					tuner -> rxsubchans &= 0xffff ^ V4L2_TUNER_SUB_STEREO;
+					tuner -> rxsubchans = V4L2_TUNER_SUB_MONO;
 				}
 			}
+			tuner->capability |= V4L2_TUNER_CAP_LOW | V4L2_TUNER_CAP_STEREO;
+			tuner->audmode = t->audmode;
+
+			tuner->rangelow = radio_range[0] * 16000;
+			tuner->rangehigh = radio_range[1] * 16000;
+		} else {
+			tuner->rangelow = tv_range[0] * 16;
+			tuner->rangehigh = tv_range[1] * 16;
 		}
-		/* Wow to deal with V4L2_TUNER_CAP_LOW ? For now, it accepts from low at 62.5KHz step  to high at 62.5 Hz */
-		tuner->rangelow = tv_range[0] * 16;
-//		tuner->rangehigh = tv_range[1] * 16;
-//		tuner->rangelow = tv_range[0] * 16384;
-		tuner->rangehigh = tv_range[1] * 16384;
 		break;
 	}
+	case VIDIOC_S_TUNER: /* Allow changing radio range and audio mode */
+	{
+		struct v4l2_tuner *tuner = arg;
+
+		CHECK_ADDR(radio_tuner,"VIDIOC_S_TUNER","radio");
+		SWITCH_V4L2;
+
+		/* To switch the audio mode, applications initialize the
+		   index and audmode fields and the reserved array and
+		   call the VIDIOC_S_TUNER ioctl. */
+		/* rxsubchannels: V4L2_TUNER_MODE_MONO, V4L2_TUNER_MODE_STEREO,
+		   V4L2_TUNER_MODE_LANG1, V4L2_TUNER_MODE_LANG2,
+		   V4L2_TUNER_MODE_SAP */
+
+		if (tuner->audmode == V4L2_TUNER_MODE_MONO)
+			t->audmode = V4L2_TUNER_MODE_MONO;
+		else
+			t->audmode = V4L2_TUNER_MODE_STEREO;
+
+		set_radio_freq(client, t->freq);
+		break;
+	}
+	case TDA9887_SET_CONFIG: /* Nothing to do on tuner-core */
+		break;
 	default:
 		tuner_dbg ("Unimplemented IOCTL 0x%08x called to tuner.\n", cmd);
 		/* nothing */
diff --git a/drivers/media/video/tuner-simple.c b/drivers/media/video/tuner-simple.c
index 539f3055731..c39ed6226ee 100644
--- a/drivers/media/video/tuner-simple.c
+++ b/drivers/media/video/tuner-simple.c
@@ -1,5 +1,5 @@
 /*
- * $Id: tuner-simple.c,v 1.21 2005/06/10 19:53:26 nsh Exp $
+ * $Id: tuner-simple.c,v 1.31 2005/06/21 16:02:25 mkrufky Exp $
  *
  * i2c tv tuner chip device driver
  * controls all those simple 4-control-bytes style tuners.
@@ -207,28 +207,27 @@ static struct tunertype tuners[] = {
 	{ "LG PAL (TAPE series)", LGINNOTEK, PAL,
           16*170.00, 16*450.00, 0x01,0x02,0x08,0xce,623},
 
-        { "Philips PAL/SECAM multi (FQ1216AME MK4)", Philips, PAL,
-          16*160.00,16*442.00,0x01,0x02,0x04,0xce,623 },
-        { "Philips FQ1236A MK4", Philips, NTSC,
-          16*160.00,16*442.00,0x01,0x02,0x04,0x8e,732 },
+	{ "Philips PAL/SECAM multi (FQ1216AME MK4)", Philips, PAL,
+	  16*160.00,16*442.00,0x01,0x02,0x04,0xce,623 },
+	{ "Philips FQ1236A MK4", Philips, NTSC,
+	  16*160.00,16*442.00,0x01,0x02,0x04,0x8e,732 },
 
 	/* Should work for TVF8531MF, TVF8831MF, TVF8731MF */
 	{ "Ymec TVision TVF-8531MF", Philips, NTSC,
 	  16*160.00,16*454.00,0xa0,0x90,0x30,0x8e,732},
 	{ "Ymec TVision TVF-5533MF", Philips, NTSC,
 	  16*160.00,16*454.00,0x01,0x02,0x04,0x8e,732},
-
 	{ "Thomson DDT 7611 (ATSC/NTSC)", THOMSON, ATSC,
 	  16*157.25,16*454.00,0x39,0x3a,0x3c,0x8e,732},
-	{ "Tena TNF9533-D/IF", LGINNOTEK, PAL,
-	  16*160.25, 16*464.25, 0x01,0x02,0x08,0x8e,623},
+	/* Should work for TNF9533-D/IF, TNF9533-B/DF */
+	{ "Tena TNF9533-D/IF", Philips, PAL,
+          16*160.25,16*464.25,0x01,0x02,0x04,0x8e,623},
 
-	/*
-	 * This entry is for TEA5767 FM radio only chip used on several boards
-	 * w/TV tuner
-	 */
+	/* This entry is for TEA5767 FM radio only chip used on several boards w/TV tuner */
 	{ TEA5767_TUNER_NAME, Philips, RADIO,
-	  -1, -1, 0, 0, 0, TEA5767_LOW_LO_32768,0},
+          -1, -1, 0, 0, 0, TEA5767_LOW_LO_32768,0},
+	{ "Philips FMD1216ME MK3 Hybrid Tuner", Philips, PAL,
+	  16*160.00,16*442.00,0x51,0x52,0x54,0x86,623 },
 };
 
 unsigned const int tuner_count = ARRAY_SIZE(tuners);
@@ -455,24 +454,24 @@ static void default_set_radio_freq(struct i2c_client *c, unsigned int freq)
 	int rc;
 
 	tun=&tuners[t->type];
-	div = freq + (int)(16*10.7);
+	div = (freq / 1000) + (int)(16*10.7);
 	buffer[2] = tun->config;
 
 	switch (t->type) {
 	case TUNER_TENA_9533_DI:
 	case TUNER_YMEC_TVF_5533MF:
-
 		/*These values are empirically determinated */
-		div = (freq*122)/16 - 20;
+		div = (freq * 122) / 16000 - 20;
 		buffer[2] = 0x88; /* could be also 0x80 */
 		buffer[3] = 0x19; /* could be also 0x10, 0x18, 0x99 */
 		break;
 	case TUNER_PHILIPS_FM1216ME_MK3:
 	case TUNER_PHILIPS_FM1236_MK3:
+	case TUNER_PHILIPS_FMD1216ME_MK3:
 		buffer[3] = 0x19;
 		break;
 	case TUNER_PHILIPS_FM1256_IH3:
-		div = (20 * freq)/16 + 333 * 2;
+		div = (20 * freq) / 16000 + 333 * 2;
 	        buffer[2] = 0x80;
 		buffer[3] = 0x19;
 		break;
@@ -505,6 +504,7 @@ int default_tuner_init(struct i2c_client *c)
 	t->radio_freq = default_set_radio_freq;
 	t->has_signal = tuner_signal;
 	t->is_stereo  = tuner_stereo;
+
 	return 0;
 }
 
diff --git a/include/media/tuner.h b/include/media/tuner.h
index 2dd8310901e..4794c563236 100644
--- a/include/media/tuner.h
+++ b/include/media/tuner.h
@@ -1,5 +1,6 @@
 
-/*
+/* $Id: tuner.h,v 1.33 2005/06/21 14:58:08 mkrufky Exp $
+ *
     tuner.h - definition for different tuners
 
     Copyright (C) 1997 Markus Schroeder (schroedm@uni-duesseldorf.de)
@@ -23,6 +24,8 @@
 #ifndef _TUNER_H
 #define _TUNER_H
 
+#include <linux/videodev2.h>
+
 #include "id.h"
 
 #define ADDR_UNSET (255)
@@ -88,7 +91,7 @@
 #define TUNER_LG_NTSC_TAPE       47
 
 #define TUNER_TNF_8831BGFF       48
-#define TUNER_MICROTUNE_4042FI5  49	/* FusionHDTV 3 Gold - 4042 FI5 (3X 8147) */
+#define TUNER_MICROTUNE_4042FI5  49	/* DViCO FusionHDTV 3 Gold-Q - 4042 FI5 (3X 8147) */
 #define TUNER_TCL_2002N          50
 #define TUNER_PHILIPS_FM1256_IH3   51
 
@@ -98,18 +101,18 @@
 #define TUNER_LG_PAL_TAPE        55    /* Hauppauge PVR-150 PAL */
 
 #define TUNER_PHILIPS_FQ1216AME_MK4 56 /* Hauppauge PVR-150 PAL */
-#define TUNER_PHILIPS_FQ1236A_MK4 57   /* Hauppauge PVR-500MCE NTSC */
+#define TUNER_PHILIPS_FQ1236A_MK4   57   /* Hauppauge PVR-500MCE NTSC */
 
 #define TUNER_YMEC_TVF_8531MF 58
 #define TUNER_YMEC_TVF_5533MF 59	/* Pixelview Pro Ultra NTSC */
-#define TUNER_THOMSON_DTT7611 60
+#define TUNER_THOMSON_DTT7611 60	/* DViCO FusionHDTV 3 Gold-T */
 #define TUNER_TENA_9533_DI    61
+
 #define TUNER_TEA5767         62	/* Only FM Radio Tuner */
+#define TUNER_PHILIPS_FMD1216ME_MK3 63
 
 #define TEA5767_TUNER_NAME "Philips TEA5767HN FM Radio"
 
-#define TUNER_THOMSON_DTT7611    60
-
 #define NOTUNER 0
 #define PAL     1	/* PAL_BG */
 #define PAL_I   2
@@ -194,11 +197,15 @@ struct tuner {
 	unsigned char i2c_easy_mode[2];
 	unsigned char i2c_set_freq[8];
 
+	/* used to keep track of audmode */
+	unsigned int audmode;
+
 	/* function ptrs */
 	void (*tv_freq)(struct i2c_client *c, unsigned int freq);
 	void (*radio_freq)(struct i2c_client *c, unsigned int freq);
 	int  (*has_signal)(struct i2c_client *c);
 	int  (*is_stereo)(struct i2c_client *c);
+	int  (*set_tuner)(struct i2c_client *c, struct v4l2_tuner *v);
 };
 
 extern unsigned int tuner_debug;
@@ -206,6 +213,7 @@ extern unsigned const int tuner_count;
 
 extern int microtune_init(struct i2c_client *c);
 extern int tda8290_init(struct i2c_client *c);
+extern int tea5767_tuner_init(struct i2c_client *c);
 extern int default_tuner_init(struct i2c_client *c);
 
 #define tuner_warn(fmt, arg...) \
-- 
cgit v1.2.3-70-g09d2


From 115d6f3fd25991f2a7de1ff4d758086209b1ed12 Mon Sep 17 00:00:00 2001
From: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
Date: Tue, 28 Jun 2005 20:45:27 -0700
Subject: [PATCH] V4L: API new webcam formats included

Add Philips Webcam format.

Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
Signed-off-by: Luc Saillard <luc@saillard.org>.
Signed-off-by: Nickolay V Shmyrev <nshmyrev@yandex.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/videodev2.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/linux/videodev2.h b/include/linux/videodev2.h
index 4e0edce5376..acbfc525576 100644
--- a/include/linux/videodev2.h
+++ b/include/linux/videodev2.h
@@ -221,6 +221,8 @@ struct v4l2_pix_format
 /*  Vendor-specific formats   */
 #define V4L2_PIX_FMT_WNVA     v4l2_fourcc('W','N','V','A') /* Winnov hw compress */
 #define V4L2_PIX_FMT_SN9C10X  v4l2_fourcc('S','9','1','0') /* SN9C10x compression */
+#define V4L2_PIX_FMT_PWC1     v4l2_fourcc('P','W','C','1') /* pwc older webcam */
+#define V4L2_PIX_FMT_PWC2     v4l2_fourcc('P','W','C','2') /* pwc newer webcam */
 
 /*
  *	F O R M A T   E N U M E R A T I O N
-- 
cgit v1.2.3-70-g09d2


From b9a05d1d51e27681e8a1162d8f11c1820915e921 Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jgarzik@pobox.com>
Date: Wed, 29 Jun 2005 01:17:55 -0400
Subject: [PATCH] ieee80211.h build fix

This crept in with the resync-to-mainline.  Nothing uses 802.11-crypt in
mainline, so we can safely comment it out for now.

Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/net/ieee80211.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/net/ieee80211.h b/include/net/ieee80211.h
index 151c4f20355..db09580ad14 100644
--- a/include/net/ieee80211.h
+++ b/include/net/ieee80211.h
@@ -426,7 +426,9 @@ struct ieee80211_stats {
 
 struct ieee80211_device;
 
+#if 0 /* for later */
 #include "ieee80211_crypt.h"
+#endif
 
 #define SEC_KEY_1         (1<<0)
 #define SEC_KEY_2         (1<<1)
-- 
cgit v1.2.3-70-g09d2


From b720f73296916e87c744193c57bc8687d35888fe Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 29 Jun 2005 15:15:54 +0100
Subject: [PATCH] ARM: Convert ARM timer implementations to use readl/writel

Convert ARMs timer implementations to use readl/writel instead of accessing
the registers via a struct.

People have recently asked if accessing timers via a structure is the
"right way" and its not the Linux way.  So fix this code to conform to
"The Linux Way"(tm).

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
---
 arch/arm/mach-integrator/core.c      | 42 +++++++++----------------
 arch/arm/mach-versatile/core.c       | 61 ++++++++++++++----------------------
 include/asm-arm/hardware/arm_timer.h | 21 +++++++++++++
 3 files changed, 59 insertions(+), 65 deletions(-)
 create mode 100644 include/asm-arm/hardware/arm_timer.h

(limited to 'include')

diff --git a/arch/arm/mach-integrator/core.c b/arch/arm/mach-integrator/core.c
index 9222e57bd87..797c7fddaa0 100644
--- a/arch/arm/mach-integrator/core.c
+++ b/arch/arm/mach-integrator/core.c
@@ -20,6 +20,7 @@
 #include <asm/irq.h>
 #include <asm/io.h>
 #include <asm/hardware/amba.h>
+#include <asm/hardware/arm_timer.h>
 #include <asm/arch/cm.h>
 #include <asm/system.h>
 #include <asm/leds.h>
@@ -156,16 +157,6 @@ EXPORT_SYMBOL(cm_control);
 #define TICKS2USECS(x)	((x) / TICKS_PER_uSEC)
 #endif
 
-/*
- * What does it look like?
- */
-typedef struct TimerStruct {
-	unsigned long TimerLoad;
-	unsigned long TimerValue;
-	unsigned long TimerControl;
-	unsigned long TimerClear;
-} TimerStruct_t;
-
 static unsigned long timer_reload;
 
 /*
@@ -174,7 +165,6 @@ static unsigned long timer_reload;
  */
 unsigned long integrator_gettimeoffset(void)
 {
-	volatile TimerStruct_t *timer1 = (TimerStruct_t *)TIMER1_VA_BASE;
 	unsigned long ticks1, ticks2, status;
 
 	/*
@@ -183,11 +173,11 @@ unsigned long integrator_gettimeoffset(void)
 	 * an interrupt.  We get around this by ensuring that the
 	 * counter has not reloaded between our two reads.
 	 */
-	ticks2 = timer1->TimerValue & 0xffff;
+	ticks2 = readl(TIMER1_VA_BASE + TIMER_VALUE) & 0xffff;
 	do {
 		ticks1 = ticks2;
 		status = __raw_readl(VA_IC_BASE + IRQ_RAW_STATUS);
-		ticks2 = timer1->TimerValue & 0xffff;
+		ticks2 = readl(TIMER1_VA_BASE + TIMER_VALUE) & 0xffff;
 	} while (ticks2 > ticks1);
 
 	/*
@@ -213,20 +203,19 @@ unsigned long integrator_gettimeoffset(void)
 static irqreturn_t
 integrator_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
-	volatile TimerStruct_t *timer1 = (volatile TimerStruct_t *)TIMER1_VA_BASE;
-
 	write_seqlock(&xtime_lock);
 
 	/*
 	 * clear the interrupt
 	 */
-	timer1->TimerClear = 1;
+	writel(1, TIMER1_VA_BASE + TIMER_INTCLR);
 
 	/*
 	 * the clock tick routines are only processed on the
 	 * primary CPU
 	 */
 	if (hard_smp_processor_id() == 0) {
+		nmi_tick();
 		timer_tick(regs);
 #ifdef CONFIG_SMP
 		smp_send_timer();
@@ -256,32 +245,29 @@ static struct irqaction integrator_timer_irq = {
  */
 void __init integrator_time_init(unsigned long reload, unsigned int ctrl)
 {
-	volatile TimerStruct_t *timer0 = (volatile TimerStruct_t *)TIMER0_VA_BASE;
-	volatile TimerStruct_t *timer1 = (volatile TimerStruct_t *)TIMER1_VA_BASE;
-	volatile TimerStruct_t *timer2 = (volatile TimerStruct_t *)TIMER2_VA_BASE;
-	unsigned int timer_ctrl = 0x80 | 0x40;	/* periodic */
+	unsigned int timer_ctrl = TIMER_CTRL_ENABLE | TIMER_CTRL_PERIODIC;
 
 	timer_reload = reload;
 	timer_ctrl |= ctrl;
 
 	if (timer_reload > 0x100000) {
 		timer_reload >>= 8;
-		timer_ctrl |= 0x08; /* /256 */
+		timer_ctrl |= TIMER_CTRL_DIV256;
 	} else if (timer_reload > 0x010000) {
 		timer_reload >>= 4;
-		timer_ctrl |= 0x04; /* /16 */
+		timer_ctrl |= TIMER_CTRL_DIV16;
 	}
 
 	/*
 	 * Initialise to a known state (all timers off)
 	 */
-	timer0->TimerControl = 0;
-	timer1->TimerControl = 0;
-	timer2->TimerControl = 0;
+	writel(0, TIMER0_VA_BASE + TIMER_CTRL);
+	writel(0, TIMER1_VA_BASE + TIMER_CTRL);
+	writel(0, TIMER2_VA_BASE + TIMER_CTRL);
 
-	timer1->TimerLoad    = timer_reload;
-	timer1->TimerValue   = timer_reload;
-	timer1->TimerControl = timer_ctrl;
+	writel(timer_reload, TIMER1_VA_BASE + TIMER_LOAD);
+	writel(timer_reload, TIMER1_VA_BASE + TIMER_VALUE);
+	writel(timer_ctrl, TIMER1_VA_BASE + TIMER_CTRL);
 
 	/*
 	 * Make irqs happen for the system timer
diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c
index 9d1f2253e98..f01c0f8a2bb 100644
--- a/arch/arm/mach-versatile/core.c
+++ b/arch/arm/mach-versatile/core.c
@@ -33,6 +33,7 @@
 #include <asm/mach-types.h>
 #include <asm/hardware/amba.h>
 #include <asm/hardware/amba_clcd.h>
+#include <asm/hardware/arm_timer.h>
 #include <asm/hardware/icst307.h>
 
 #include <asm/mach/arch.h>
@@ -788,38 +789,25 @@ void __init versatile_init(void)
  */
 #define TIMER_INTERVAL	(TICKS_PER_uSEC * mSEC_10)
 #if TIMER_INTERVAL >= 0x100000
-#define TIMER_RELOAD	(TIMER_INTERVAL >> 8)		/* Divide by 256 */
-#define TIMER_CTRL	0x88				/* Enable, Clock / 256 */
+#define TIMER_RELOAD	(TIMER_INTERVAL >> 8)
+#define TIMER_DIVISOR	(TIMER_CTRL_DIV256)
 #define TICKS2USECS(x)	(256 * (x) / TICKS_PER_uSEC)
 #elif TIMER_INTERVAL >= 0x10000
 #define TIMER_RELOAD	(TIMER_INTERVAL >> 4)		/* Divide by 16 */
-#define TIMER_CTRL	0x84				/* Enable, Clock / 16 */
+#define TIMER_DIVISOR	(TIMER_CTRL_DIV16)
 #define TICKS2USECS(x)	(16 * (x) / TICKS_PER_uSEC)
 #else
 #define TIMER_RELOAD	(TIMER_INTERVAL)
-#define TIMER_CTRL	0x80				/* Enable */
+#define TIMER_DIVISOR	(TIMER_CTRL_DIV1)
 #define TICKS2USECS(x)	((x) / TICKS_PER_uSEC)
 #endif
 
-#define TIMER_CTRL_IE	(1 << 5)	/* Interrupt Enable */
-
-/*
- * What does it look like?
- */
-typedef struct TimerStruct {
-	unsigned long TimerLoad;
-	unsigned long TimerValue;
-	unsigned long TimerControl;
-	unsigned long TimerClear;
-} TimerStruct_t;
-
 /*
  * Returns number of ms since last clock interrupt.  Note that interrupts
  * will have been disabled by do_gettimeoffset()
  */
 static unsigned long versatile_gettimeoffset(void)
 {
-	volatile TimerStruct_t *timer0 = (TimerStruct_t *)TIMER0_VA_BASE;
 	unsigned long ticks1, ticks2, status;
 
 	/*
@@ -828,11 +816,11 @@ static unsigned long versatile_gettimeoffset(void)
 	 * an interrupt.  We get around this by ensuring that the
 	 * counter has not reloaded between our two reads.
 	 */
-	ticks2 = timer0->TimerValue & 0xffff;
+	ticks2 = readl(TIMER0_VA_BASE + TIMER_VALUE) & 0xffff;
 	do {
 		ticks1 = ticks2;
 		status = __raw_readl(VA_IC_BASE + VIC_IRQ_RAW_STATUS);
-		ticks2 = timer0->TimerValue & 0xffff;
+		ticks2 = readl(TIMER0_VA_BASE + TIMER_VALUE) & 0xffff;
 	} while (ticks2 > ticks1);
 
 	/*
@@ -859,12 +847,10 @@ static unsigned long versatile_gettimeoffset(void)
  */
 static irqreturn_t versatile_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
-	volatile TimerStruct_t *timer0 = (volatile TimerStruct_t *)TIMER0_VA_BASE;
-
 	write_seqlock(&xtime_lock);
 
 	// ...clear the interrupt
-	timer0->TimerClear = 1;
+	writel(1, TIMER0_VA_BASE + TIMER_INTCLR);
 
 	timer_tick(regs);
 
@@ -884,31 +870,32 @@ static struct irqaction versatile_timer_irq = {
  */
 static void __init versatile_timer_init(void)
 {
-	volatile TimerStruct_t *timer0 = (volatile TimerStruct_t *)TIMER0_VA_BASE;
-	volatile TimerStruct_t *timer1 = (volatile TimerStruct_t *)TIMER1_VA_BASE;
-	volatile TimerStruct_t *timer2 = (volatile TimerStruct_t *)TIMER2_VA_BASE;
-	volatile TimerStruct_t *timer3 = (volatile TimerStruct_t *)TIMER3_VA_BASE;
+	u32 val;
 
 	/* 
 	 * set clock frequency: 
 	 *	VERSATILE_REFCLK is 32KHz
 	 *	VERSATILE_TIMCLK is 1MHz
 	 */
-	*(volatile unsigned int *)IO_ADDRESS(VERSATILE_SCTL_BASE) |= 
-	  ((VERSATILE_TIMCLK << VERSATILE_TIMER1_EnSel) | (VERSATILE_TIMCLK << VERSATILE_TIMER2_EnSel) | 
-	   (VERSATILE_TIMCLK << VERSATILE_TIMER3_EnSel) | (VERSATILE_TIMCLK << VERSATILE_TIMER4_EnSel));
+	val = readl(IO_ADDRESS(VERSATILE_SCTL_BASE));
+	writel((VERSATILE_TIMCLK << VERSATILE_TIMER1_EnSel) |
+	       (VERSATILE_TIMCLK << VERSATILE_TIMER2_EnSel) | 
+	       (VERSATILE_TIMCLK << VERSATILE_TIMER3_EnSel) |
+	       (VERSATILE_TIMCLK << VERSATILE_TIMER4_EnSel) | val,
+	       IO_ADDRESS(VERSATILE_SCTL_BASE));
 
 	/*
 	 * Initialise to a known state (all timers off)
 	 */
-	timer0->TimerControl = 0;
-	timer1->TimerControl = 0;
-	timer2->TimerControl = 0;
-	timer3->TimerControl = 0;
-
-	timer0->TimerLoad    = TIMER_RELOAD;
-	timer0->TimerValue   = TIMER_RELOAD;
-	timer0->TimerControl = TIMER_CTRL | 0x40 | TIMER_CTRL_IE;  /* periodic + IE */
+	writel(0, TIMER0_VA_BASE + TIMER_CTRL);
+	writel(0, TIMER1_VA_BASE + TIMER_CTRL);
+	writel(0, TIMER2_VA_BASE + TIMER_CTRL);
+	writel(0, TIMER3_VA_BASE + TIMER_CTRL);
+
+	writel(TIMER_RELOAD, TIMER0_VA_BASE + TIMER_LOAD);
+	writel(TIMER_RELOAD, TIMER0_VA_BASE + TIMER_VALUE);
+	writel(TIMER_DIVISOR | TIMER_CTRL_ENABLE | TIMER_CTRL_PERIODIC |
+	       TIMER_CTRL_IE, TIMER0_VA_BASE + TIMER_CTRL);
 
 	/* 
 	 * Make irqs happen for the system timer
diff --git a/include/asm-arm/hardware/arm_timer.h b/include/asm-arm/hardware/arm_timer.h
new file mode 100644
index 00000000000..04be3bdf46b
--- /dev/null
+++ b/include/asm-arm/hardware/arm_timer.h
@@ -0,0 +1,21 @@
+#ifndef __ASM_ARM_HARDWARE_ARM_TIMER_H
+#define __ASM_ARM_HARDWARE_ARM_TIMER_H
+
+#define TIMER_LOAD	0x00
+#define TIMER_VALUE	0x04
+#define TIMER_CTRL	0x08
+#define TIMER_CTRL_ONESHOT	(1 << 0)
+#define TIMER_CTRL_32BIT	(1 << 1)
+#define TIMER_CTRL_DIV1		(0 << 2)
+#define TIMER_CTRL_DIV16	(1 << 2)
+#define TIMER_CTRL_DIV256	(2 << 2)
+#define TIMER_CTRL_IE		(1 << 5)	/* Interrupt Enable (versatile only) */
+#define TIMER_CTRL_PERIODIC	(1 << 6)
+#define TIMER_CTRL_ENABLE	(1 << 7)
+
+#define TIMER_INTCLR	0x0c
+#define TIMER_RIS	0x10
+#define TIMER_MIS	0x14
+#define TIMER_BGLOAD	0x18
+
+#endif
-- 
cgit v1.2.3-70-g09d2


From 9ec3c75cdeb57fa3c12a813c63706ece7fb63648 Mon Sep 17 00:00:00 2001
From: Richard Purdie <rpurdie@rpsys.net>
Date: Wed, 29 Jun 2005 15:17:49 +0100
Subject: [PATCH] ARM: 2768/1: PXA: Add a required header file for LL_DEBUG

Patch from Richard Purdie

With DEBUG enabled, head.S includes arch/debug-macro.S. On the PXA, this
contains references to the macro io_p2v() so hardware.h needs to be
included.

Signed-off-by: Richard Purdie <rpurdie@openedhand.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-pxa/debug-macro.S | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/asm-arm/arch-pxa/debug-macro.S b/include/asm-arm/arch-pxa/debug-macro.S
index f288e74b67c..b6ec6887917 100644
--- a/include/asm-arm/arch-pxa/debug-macro.S
+++ b/include/asm-arm/arch-pxa/debug-macro.S
@@ -11,6 +11,8 @@
  *
 */
 
+#include "hardware.h"
+
 		.macro	addruart,rx
 		mrc	p15, 0, \rx, c1, c0
 		tst	\rx, #1			@ MMU enabled?
-- 
cgit v1.2.3-70-g09d2


From 026d02a236f429eb61a1277166bd425f8514c431 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 29 Jun 2005 18:45:19 +0100
Subject: [PATCH] Serial: Split 8250 port table (part 2)

Remove legacy ISA serial ports for Accent, Boca, Fourport, Hub6 and MCA
from the architecture specific serial.h include.

The only ports which remain in asm-*/serial.h are the platform specific
entries.  These should really be converted by platform maintainers to
use a platform device, such as can be found in
arch/arm/mach-footbridge/isa.c

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/parisc/configs/712_defconfig   |   2 +-
 arch/parisc/configs/a500_defconfig  |   2 +-
 arch/parisc/configs/b180_defconfig  |   2 +-
 arch/parisc/configs/c3000_defconfig |   2 +-
 arch/parisc/defconfig               |   2 +-
 drivers/serial/8250.c               |   2 +-
 drivers/serial/Kconfig              |   2 +-
 include/asm-alpha/serial.h          |  47 +----------------
 include/asm-arm26/serial.h          |  22 +-------
 include/asm-i386/serial.h           | 102 +-----------------------------------
 include/asm-m68k/serial.h           |  47 +----------------
 include/asm-mips/serial.h           |  84 -----------------------------
 include/asm-parisc/serial.h         |  16 +-----
 include/asm-ppc/pc_serial.h         |  86 +-----------------------------
 include/asm-sh/bigsur/serial.h      |   5 +-
 include/asm-sh/ec3104/serial.h      |   4 +-
 include/asm-sh/serial.h             |   6 +--
 include/asm-sh64/serial.h           |   4 +-
 include/asm-x86_64/serial.h         | 102 +-----------------------------------
 19 files changed, 20 insertions(+), 519 deletions(-)

(limited to 'include')

diff --git a/arch/parisc/configs/712_defconfig b/arch/parisc/configs/712_defconfig
index 872085dea8a..6efaa9293ee 100644
--- a/arch/parisc/configs/712_defconfig
+++ b/arch/parisc/configs/712_defconfig
@@ -506,7 +506,7 @@ CONFIG_HW_CONSOLE=y
 #
 CONFIG_SERIAL_8250=y
 CONFIG_SERIAL_8250_CONSOLE=y
-CONFIG_SERIAL_8250_NR_UARTS=8
+CONFIG_SERIAL_8250_NR_UARTS=17
 CONFIG_SERIAL_8250_EXTENDED=y
 CONFIG_SERIAL_8250_MANY_PORTS=y
 CONFIG_SERIAL_8250_SHARE_IRQ=y
diff --git a/arch/parisc/configs/a500_defconfig b/arch/parisc/configs/a500_defconfig
index d28ebfa1070..30fc03ed0cf 100644
--- a/arch/parisc/configs/a500_defconfig
+++ b/arch/parisc/configs/a500_defconfig
@@ -662,7 +662,7 @@ CONFIG_HW_CONSOLE=y
 CONFIG_SERIAL_8250=y
 CONFIG_SERIAL_8250_CONSOLE=y
 CONFIG_SERIAL_8250_CS=m
-CONFIG_SERIAL_8250_NR_UARTS=8
+CONFIG_SERIAL_8250_NR_UARTS=17
 CONFIG_SERIAL_8250_EXTENDED=y
 CONFIG_SERIAL_8250_MANY_PORTS=y
 CONFIG_SERIAL_8250_SHARE_IRQ=y
diff --git a/arch/parisc/configs/b180_defconfig b/arch/parisc/configs/b180_defconfig
index 1700d7aec68..46c9511f322 100644
--- a/arch/parisc/configs/b180_defconfig
+++ b/arch/parisc/configs/b180_defconfig
@@ -514,7 +514,7 @@ CONFIG_HW_CONSOLE=y
 #
 CONFIG_SERIAL_8250=y
 CONFIG_SERIAL_8250_CONSOLE=y
-CONFIG_SERIAL_8250_NR_UARTS=4
+CONFIG_SERIAL_8250_NR_UARTS=13
 CONFIG_SERIAL_8250_EXTENDED=y
 CONFIG_SERIAL_8250_MANY_PORTS=y
 CONFIG_SERIAL_8250_SHARE_IRQ=y
diff --git a/arch/parisc/configs/c3000_defconfig b/arch/parisc/configs/c3000_defconfig
index b27980161c3..67aca6ccc9b 100644
--- a/arch/parisc/configs/c3000_defconfig
+++ b/arch/parisc/configs/c3000_defconfig
@@ -661,7 +661,7 @@ CONFIG_HW_CONSOLE=y
 #
 CONFIG_SERIAL_8250=y
 CONFIG_SERIAL_8250_CONSOLE=y
-CONFIG_SERIAL_8250_NR_UARTS=4
+CONFIG_SERIAL_8250_NR_UARTS=13
 CONFIG_SERIAL_8250_EXTENDED=y
 CONFIG_SERIAL_8250_MANY_PORTS=y
 CONFIG_SERIAL_8250_SHARE_IRQ=y
diff --git a/arch/parisc/defconfig b/arch/parisc/defconfig
index ebd6301aa59..fdae21c503d 100644
--- a/arch/parisc/defconfig
+++ b/arch/parisc/defconfig
@@ -517,7 +517,7 @@ CONFIG_HW_CONSOLE=y
 #
 CONFIG_SERIAL_8250=y
 CONFIG_SERIAL_8250_CONSOLE=y
-CONFIG_SERIAL_8250_NR_UARTS=4
+CONFIG_SERIAL_8250_NR_UARTS=13
 CONFIG_SERIAL_8250_EXTENDED=y
 CONFIG_SERIAL_8250_MANY_PORTS=y
 CONFIG_SERIAL_8250_SHARE_IRQ=y
diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index 429994d5bc8..9224fc3184e 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -105,7 +105,7 @@ static struct old_serial_port old_serial_port[] = {
 	SERIAL_PORT_DFNS /* defined in asm/serial.h */
 };
 
-#define UART_NR	(ARRAY_SIZE(old_serial_port) + CONFIG_SERIAL_8250_NR_UARTS)
+#define UART_NR	CONFIG_SERIAL_8250_NR_UARTS
 
 #ifdef CONFIG_SERIAL_8250_RSA
 
diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig
index e879bce160d..e0d0a470ddf 100644
--- a/drivers/serial/Kconfig
+++ b/drivers/serial/Kconfig
@@ -86,7 +86,7 @@ config SERIAL_8250_ACPI
 	  namespace, say Y here.  If unsure, say N.
 
 config SERIAL_8250_NR_UARTS
-	int "Maximum number of non-legacy 8250/16550 serial ports"
+	int "Maximum number of 8250/16550 serial ports"
 	depends on SERIAL_8250
 	default "4"
 	help
diff --git a/include/asm-alpha/serial.h b/include/asm-alpha/serial.h
index 7b2d9ee95a4..7e4b2987d45 100644
--- a/include/asm-alpha/serial.h
+++ b/include/asm-alpha/serial.h
@@ -22,54 +22,9 @@
 #define STD_COM4_FLAGS ASYNC_BOOT_AUTOCONF
 #endif
 
-#ifdef CONFIG_SERIAL_MANY_PORTS
-#define FOURPORT_FLAGS ASYNC_FOURPORT
-#define ACCENT_FLAGS 0
-#define BOCA_FLAGS 0
-#endif
-	
-#define STD_SERIAL_PORT_DEFNS			\
+#define SERIAL_PORT_DFNS			\
 	/* UART CLK   PORT IRQ     FLAGS        */			\
 	{ 0, BASE_BAUD, 0x3F8, 4, STD_COM_FLAGS },	/* ttyS0 */	\
 	{ 0, BASE_BAUD, 0x2F8, 3, STD_COM_FLAGS },	/* ttyS1 */	\
 	{ 0, BASE_BAUD, 0x3E8, 4, STD_COM_FLAGS },	/* ttyS2 */	\
 	{ 0, BASE_BAUD, 0x2E8, 3, STD_COM4_FLAGS },	/* ttyS3 */
-
-
-#ifdef CONFIG_SERIAL_MANY_PORTS
-#define EXTRA_SERIAL_PORT_DEFNS			\
-	{ 0, BASE_BAUD, 0x1A0, 9, FOURPORT_FLAGS }, 	/* ttyS4 */	\
-	{ 0, BASE_BAUD, 0x1A8, 9, FOURPORT_FLAGS },	/* ttyS5 */	\
-	{ 0, BASE_BAUD, 0x1B0, 9, FOURPORT_FLAGS },	/* ttyS6 */	\
-	{ 0, BASE_BAUD, 0x1B8, 9, FOURPORT_FLAGS },	/* ttyS7 */	\
-	{ 0, BASE_BAUD, 0x2A0, 5, FOURPORT_FLAGS },	/* ttyS8 */	\
-	{ 0, BASE_BAUD, 0x2A8, 5, FOURPORT_FLAGS },	/* ttyS9 */	\
-	{ 0, BASE_BAUD, 0x2B0, 5, FOURPORT_FLAGS },	/* ttyS10 */	\
-	{ 0, BASE_BAUD, 0x2B8, 5, FOURPORT_FLAGS },	/* ttyS11 */	\
-	{ 0, BASE_BAUD, 0x330, 4, ACCENT_FLAGS },	/* ttyS12 */	\
-	{ 0, BASE_BAUD, 0x338, 4, ACCENT_FLAGS },	/* ttyS13 */	\
-	{ 0, BASE_BAUD, 0x000, 0, 0 },	/* ttyS14 (spare) */		\
-	{ 0, BASE_BAUD, 0x000, 0, 0 },	/* ttyS15 (spare) */		\
-	{ 0, BASE_BAUD, 0x100, 12, BOCA_FLAGS },	/* ttyS16 */	\
-	{ 0, BASE_BAUD, 0x108, 12, BOCA_FLAGS },	/* ttyS17 */	\
-	{ 0, BASE_BAUD, 0x110, 12, BOCA_FLAGS },	/* ttyS18 */	\
-	{ 0, BASE_BAUD, 0x118, 12, BOCA_FLAGS },	/* ttyS19 */	\
-	{ 0, BASE_BAUD, 0x120, 12, BOCA_FLAGS },	/* ttyS20 */	\
-	{ 0, BASE_BAUD, 0x128, 12, BOCA_FLAGS },	/* ttyS21 */	\
-	{ 0, BASE_BAUD, 0x130, 12, BOCA_FLAGS },	/* ttyS22 */	\
-	{ 0, BASE_BAUD, 0x138, 12, BOCA_FLAGS },	/* ttyS23 */	\
-	{ 0, BASE_BAUD, 0x140, 12, BOCA_FLAGS },	/* ttyS24 */	\
-	{ 0, BASE_BAUD, 0x148, 12, BOCA_FLAGS },	/* ttyS25 */	\
-	{ 0, BASE_BAUD, 0x150, 12, BOCA_FLAGS },	/* ttyS26 */	\
-	{ 0, BASE_BAUD, 0x158, 12, BOCA_FLAGS },	/* ttyS27 */	\
-	{ 0, BASE_BAUD, 0x160, 12, BOCA_FLAGS },	/* ttyS28 */	\
-	{ 0, BASE_BAUD, 0x168, 12, BOCA_FLAGS },	/* ttyS29 */	\
-	{ 0, BASE_BAUD, 0x170, 12, BOCA_FLAGS },	/* ttyS30 */	\
-	{ 0, BASE_BAUD, 0x178, 12, BOCA_FLAGS },	/* ttyS31 */
-#else
-#define EXTRA_SERIAL_PORT_DEFNS
-#endif
-
-#define SERIAL_PORT_DFNS		\
-	STD_SERIAL_PORT_DEFNS		\
-	EXTRA_SERIAL_PORT_DEFNS
diff --git a/include/asm-arm26/serial.h b/include/asm-arm26/serial.h
index 21e1df31f08..5fc747d1b50 100644
--- a/include/asm-arm26/serial.h
+++ b/include/asm-arm26/serial.h
@@ -30,34 +30,16 @@
 #if defined(CONFIG_ARCH_A5K)
      /* UART CLK        PORT  IRQ     FLAGS        */
 
-#define STD_SERIAL_PORT_DEFNS                                           \
+#define SERIAL_PORT_DFNS                                                \
         { 0, BASE_BAUD, 0x3F8, 10, STD_COM_FLAGS },     /* ttyS0 */     \
         { 0, BASE_BAUD, 0x2F8, 10, STD_COM_FLAGS },     /* ttyS1 */
 
 #else
 
-#define STD_SERIAL_PORT_DEFNS                                           \
+#define SERIAL_PORT_DFNS                                                \
         { 0, BASE_BAUD, 0    ,  0, STD_COM_FLAGS },     /* ttyS0 */     \
         { 0, BASE_BAUD, 0    ,  0, STD_COM_FLAGS },     /* ttyS1 */
 
 #endif
 
-#define EXTRA_SERIAL_PORT_DEFNS \
-        { 0, BASE_BAUD, 0    ,  0, STD_COM_FLAGS },     /* ttyS2 */     \
-        { 0, BASE_BAUD, 0    ,  0, STD_COM_FLAGS },     /* ttyS3 */     \
-        { 0, BASE_BAUD, 0    ,  0, STD_COM_FLAGS },     /* ttyS4 */     \
-        { 0, BASE_BAUD, 0    ,  0, STD_COM_FLAGS },     /* ttyS5 */     \
-        { 0, BASE_BAUD, 0    ,  0, STD_COM_FLAGS },     /* ttyS6 */     \
-        { 0, BASE_BAUD, 0    ,  0, STD_COM_FLAGS },     /* ttyS7 */     \
-        { 0, BASE_BAUD, 0    ,  0, STD_COM_FLAGS },     /* ttyS8 */     \
-        { 0, BASE_BAUD, 0    ,  0, STD_COM_FLAGS },     /* ttyS9 */     \
-        { 0, BASE_BAUD, 0    ,  0, STD_COM_FLAGS },     /* ttyS10 */    \
-        { 0, BASE_BAUD, 0    ,  0, STD_COM_FLAGS },     /* ttyS11 */    \
-        { 0, BASE_BAUD, 0    ,  0, STD_COM_FLAGS },     /* ttyS12 */    \
-        { 0, BASE_BAUD, 0    ,  0, STD_COM_FLAGS },     /* ttyS13 */
-
-#define SERIAL_PORT_DFNS		\
-	STD_SERIAL_PORT_DEFNS		\
-	EXTRA_SERIAL_PORT_DEFNS
-
 #endif
diff --git a/include/asm-i386/serial.h b/include/asm-i386/serial.h
index 21ddecc77c7..e1ecfccb743 100644
--- a/include/asm-i386/serial.h
+++ b/include/asm-i386/serial.h
@@ -22,109 +22,9 @@
 #define STD_COM4_FLAGS ASYNC_BOOT_AUTOCONF
 #endif
 
-#ifdef CONFIG_SERIAL_MANY_PORTS
-#define FOURPORT_FLAGS ASYNC_FOURPORT
-#define ACCENT_FLAGS 0
-#define BOCA_FLAGS 0
-#define HUB6_FLAGS 0
-#endif
-
-#define MCA_COM_FLAGS	(STD_COM_FLAGS|ASYNC_BOOT_ONLYMCA)
-
-/*
- * The following define the access methods for the HUB6 card. All
- * access is through two ports for all 24 possible chips. The card is
- * selected through the high 2 bits, the port on that card with the
- * "middle" 3 bits, and the register on that port with the bottom
- * 3 bits.
- *
- * While the access port and interrupt is configurable, the default
- * port locations are 0x302 for the port control register, and 0x303
- * for the data read/write register. Normally, the interrupt is at irq3
- * but can be anything from 3 to 7 inclusive. Note that using 3 will
- * require disabling com2.
- */
-
-#define C_P(card,port) (((card)<<6|(port)<<3) + 1)
-
-#define STD_SERIAL_PORT_DEFNS			\
+#define SERIAL_PORT_DFNS			\
 	/* UART CLK   PORT IRQ     FLAGS        */			\
 	{ 0, BASE_BAUD, 0x3F8, 4, STD_COM_FLAGS },	/* ttyS0 */	\
 	{ 0, BASE_BAUD, 0x2F8, 3, STD_COM_FLAGS },	/* ttyS1 */	\
 	{ 0, BASE_BAUD, 0x3E8, 4, STD_COM_FLAGS },	/* ttyS2 */	\
 	{ 0, BASE_BAUD, 0x2E8, 3, STD_COM4_FLAGS },	/* ttyS3 */
-
-
-#ifdef CONFIG_SERIAL_MANY_PORTS
-#define EXTRA_SERIAL_PORT_DEFNS			\
-	{ 0, BASE_BAUD, 0x1A0, 9, FOURPORT_FLAGS }, 	/* ttyS4 */	\
-	{ 0, BASE_BAUD, 0x1A8, 9, FOURPORT_FLAGS },	/* ttyS5 */	\
-	{ 0, BASE_BAUD, 0x1B0, 9, FOURPORT_FLAGS },	/* ttyS6 */	\
-	{ 0, BASE_BAUD, 0x1B8, 9, FOURPORT_FLAGS },	/* ttyS7 */	\
-	{ 0, BASE_BAUD, 0x2A0, 5, FOURPORT_FLAGS },	/* ttyS8 */	\
-	{ 0, BASE_BAUD, 0x2A8, 5, FOURPORT_FLAGS },	/* ttyS9 */	\
-	{ 0, BASE_BAUD, 0x2B0, 5, FOURPORT_FLAGS },	/* ttyS10 */	\
-	{ 0, BASE_BAUD, 0x2B8, 5, FOURPORT_FLAGS },	/* ttyS11 */	\
-	{ 0, BASE_BAUD, 0x330, 4, ACCENT_FLAGS },	/* ttyS12 */	\
-	{ 0, BASE_BAUD, 0x338, 4, ACCENT_FLAGS },	/* ttyS13 */	\
-	{ 0, BASE_BAUD, 0x000, 0, 0 },	/* ttyS14 (spare) */		\
-	{ 0, BASE_BAUD, 0x000, 0, 0 },	/* ttyS15 (spare) */		\
-	{ 0, BASE_BAUD, 0x100, 12, BOCA_FLAGS },	/* ttyS16 */	\
-	{ 0, BASE_BAUD, 0x108, 12, BOCA_FLAGS },	/* ttyS17 */	\
-	{ 0, BASE_BAUD, 0x110, 12, BOCA_FLAGS },	/* ttyS18 */	\
-	{ 0, BASE_BAUD, 0x118, 12, BOCA_FLAGS },	/* ttyS19 */	\
-	{ 0, BASE_BAUD, 0x120, 12, BOCA_FLAGS },	/* ttyS20 */	\
-	{ 0, BASE_BAUD, 0x128, 12, BOCA_FLAGS },	/* ttyS21 */	\
-	{ 0, BASE_BAUD, 0x130, 12, BOCA_FLAGS },	/* ttyS22 */	\
-	{ 0, BASE_BAUD, 0x138, 12, BOCA_FLAGS },	/* ttyS23 */	\
-	{ 0, BASE_BAUD, 0x140, 12, BOCA_FLAGS },	/* ttyS24 */	\
-	{ 0, BASE_BAUD, 0x148, 12, BOCA_FLAGS },	/* ttyS25 */	\
-	{ 0, BASE_BAUD, 0x150, 12, BOCA_FLAGS },	/* ttyS26 */	\
-	{ 0, BASE_BAUD, 0x158, 12, BOCA_FLAGS },	/* ttyS27 */	\
-	{ 0, BASE_BAUD, 0x160, 12, BOCA_FLAGS },	/* ttyS28 */	\
-	{ 0, BASE_BAUD, 0x168, 12, BOCA_FLAGS },	/* ttyS29 */	\
-	{ 0, BASE_BAUD, 0x170, 12, BOCA_FLAGS },	/* ttyS30 */	\
-	{ 0, BASE_BAUD, 0x178, 12, BOCA_FLAGS },	/* ttyS31 */
-#else
-#define EXTRA_SERIAL_PORT_DEFNS
-#endif
-
-/* You can have up to four HUB6's in the system, but I've only
- * included two cards here for a total of twelve ports.
- */
-#if (defined(CONFIG_HUB6) && defined(CONFIG_SERIAL_MANY_PORTS))
-#define HUB6_SERIAL_PORT_DFNS		\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,0) },  /* ttyS32 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,1) },  /* ttyS33 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,2) },  /* ttyS34 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,3) },  /* ttyS35 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,4) },  /* ttyS36 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,5) },  /* ttyS37 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,0) },  /* ttyS38 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,1) },  /* ttyS39 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,2) },  /* ttyS40 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,3) },  /* ttyS41 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,4) },  /* ttyS42 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,5) },  /* ttyS43 */
-#else
-#define HUB6_SERIAL_PORT_DFNS
-#endif
-
-#ifdef CONFIG_MCA
-#define MCA_SERIAL_PORT_DFNS			\
-	{ 0, BASE_BAUD, 0x3220, 3, MCA_COM_FLAGS },	\
-	{ 0, BASE_BAUD, 0x3228, 3, MCA_COM_FLAGS },	\
-	{ 0, BASE_BAUD, 0x4220, 3, MCA_COM_FLAGS },	\
-	{ 0, BASE_BAUD, 0x4228, 3, MCA_COM_FLAGS },	\
-	{ 0, BASE_BAUD, 0x5220, 3, MCA_COM_FLAGS },	\
-	{ 0, BASE_BAUD, 0x5228, 3, MCA_COM_FLAGS },
-#else
-#define MCA_SERIAL_PORT_DFNS
-#endif
-
-#define SERIAL_PORT_DFNS		\
-	STD_SERIAL_PORT_DEFNS		\
-	EXTRA_SERIAL_PORT_DEFNS		\
-	HUB6_SERIAL_PORT_DFNS		\
-	MCA_SERIAL_PORT_DFNS
-
diff --git a/include/asm-m68k/serial.h b/include/asm-m68k/serial.h
index 9f5bcdc105f..3fe29f8b019 100644
--- a/include/asm-m68k/serial.h
+++ b/include/asm-m68k/serial.h
@@ -26,54 +26,9 @@
 #define STD_COM4_FLAGS ASYNC_BOOT_AUTOCONF
 #endif
 
-#ifdef CONFIG_SERIAL_MANY_PORTS
-#define FOURPORT_FLAGS ASYNC_FOURPORT
-#define ACCENT_FLAGS 0
-#define BOCA_FLAGS 0
-#endif
-
-#define STD_SERIAL_PORT_DEFNS			\
+#define SERIAL_PORT_DFNS			\
 	/* UART CLK   PORT IRQ     FLAGS        */			\
 	{ 0, BASE_BAUD, 0x3F8, 4, STD_COM_FLAGS },	/* ttyS0 */	\
 	{ 0, BASE_BAUD, 0x2F8, 3, STD_COM_FLAGS },	/* ttyS1 */	\
 	{ 0, BASE_BAUD, 0x3E8, 4, STD_COM_FLAGS },	/* ttyS2 */	\
 	{ 0, BASE_BAUD, 0x2E8, 3, STD_COM4_FLAGS },	/* ttyS3 */
-
-
-#ifdef CONFIG_SERIAL_MANY_PORTS
-#define EXTRA_SERIAL_PORT_DEFNS			\
-	{ 0, BASE_BAUD, 0x1A0, 9, FOURPORT_FLAGS },	/* ttyS4 */	\
-	{ 0, BASE_BAUD, 0x1A8, 9, FOURPORT_FLAGS },	/* ttyS5 */	\
-	{ 0, BASE_BAUD, 0x1B0, 9, FOURPORT_FLAGS },	/* ttyS6 */	\
-	{ 0, BASE_BAUD, 0x1B8, 9, FOURPORT_FLAGS },	/* ttyS7 */	\
-	{ 0, BASE_BAUD, 0x2A0, 5, FOURPORT_FLAGS },	/* ttyS8 */	\
-	{ 0, BASE_BAUD, 0x2A8, 5, FOURPORT_FLAGS },	/* ttyS9 */	\
-	{ 0, BASE_BAUD, 0x2B0, 5, FOURPORT_FLAGS },	/* ttyS10 */	\
-	{ 0, BASE_BAUD, 0x2B8, 5, FOURPORT_FLAGS },	/* ttyS11 */	\
-	{ 0, BASE_BAUD, 0x330, 4, ACCENT_FLAGS },	/* ttyS12 */	\
-	{ 0, BASE_BAUD, 0x338, 4, ACCENT_FLAGS },	/* ttyS13 */	\
-	{ 0, BASE_BAUD, 0x000, 0, 0 },	/* ttyS14 (spare) */		\
-	{ 0, BASE_BAUD, 0x000, 0, 0 },	/* ttyS15 (spare) */		\
-	{ 0, BASE_BAUD, 0x100, 12, BOCA_FLAGS },	/* ttyS16 */	\
-	{ 0, BASE_BAUD, 0x108, 12, BOCA_FLAGS },	/* ttyS17 */	\
-	{ 0, BASE_BAUD, 0x110, 12, BOCA_FLAGS },	/* ttyS18 */	\
-	{ 0, BASE_BAUD, 0x118, 12, BOCA_FLAGS },	/* ttyS19 */	\
-	{ 0, BASE_BAUD, 0x120, 12, BOCA_FLAGS },	/* ttyS20 */	\
-	{ 0, BASE_BAUD, 0x128, 12, BOCA_FLAGS },	/* ttyS21 */	\
-	{ 0, BASE_BAUD, 0x130, 12, BOCA_FLAGS },	/* ttyS22 */	\
-	{ 0, BASE_BAUD, 0x138, 12, BOCA_FLAGS },	/* ttyS23 */	\
-	{ 0, BASE_BAUD, 0x140, 12, BOCA_FLAGS },	/* ttyS24 */	\
-	{ 0, BASE_BAUD, 0x148, 12, BOCA_FLAGS },	/* ttyS25 */	\
-	{ 0, BASE_BAUD, 0x150, 12, BOCA_FLAGS },	/* ttyS26 */	\
-	{ 0, BASE_BAUD, 0x158, 12, BOCA_FLAGS },	/* ttyS27 */	\
-	{ 0, BASE_BAUD, 0x160, 12, BOCA_FLAGS },	/* ttyS28 */	\
-	{ 0, BASE_BAUD, 0x168, 12, BOCA_FLAGS },	/* ttyS29 */	\
-	{ 0, BASE_BAUD, 0x170, 12, BOCA_FLAGS },	/* ttyS30 */	\
-	{ 0, BASE_BAUD, 0x178, 12, BOCA_FLAGS },	/* ttyS31 */
-#else
-#define EXTRA_SERIAL_PORT_DEFNS
-#endif
-
-#define SERIAL_PORT_DFNS		\
-	STD_SERIAL_PORT_DEFNS		\
-	EXTRA_SERIAL_PORT_DEFNS
diff --git a/include/asm-mips/serial.h b/include/asm-mips/serial.h
index 8a70ff58f76..4eed8e2acdc 100644
--- a/include/asm-mips/serial.h
+++ b/include/asm-mips/serial.h
@@ -29,32 +29,6 @@
 #define STD_COM4_FLAGS ASYNC_BOOT_AUTOCONF
 #endif
 
-#ifdef CONFIG_SERIAL_MANY_PORTS
-#define FOURPORT_FLAGS ASYNC_FOURPORT
-#define ACCENT_FLAGS 0
-#define BOCA_FLAGS 0
-#define HUB6_FLAGS 0
-#define RS_TABLE_SIZE	64
-#else
-#define RS_TABLE_SIZE
-#endif
-
-/*
- * The following define the access methods for the HUB6 card. All
- * access is through two ports for all 24 possible chips. The card is
- * selected through the high 2 bits, the port on that card with the
- * "middle" 3 bits, and the register on that port with the bottom
- * 3 bits.
- *
- * While the access port and interrupt is configurable, the default
- * port locations are 0x302 for the port control register, and 0x303
- * for the data read/write register. Normally, the interrupt is at irq3
- * but can be anything from 3 to 7 inclusive. Note that using 3 will
- * require disabling com2.
- */
-
-#define C_P(card,port) (((card)<<6|(port)<<3) + 1)
-
 #ifdef CONFIG_MACH_JAZZ
 #include <asm/jazz.h>
 
@@ -240,66 +214,10 @@
 	{ 0, BASE_BAUD, 0x3E8, 4, STD_COM_FLAGS },	/* ttyS2 */	\
 	{ 0, BASE_BAUD, 0x2E8, 3, STD_COM4_FLAGS },	/* ttyS3 */
 
-#ifdef CONFIG_SERIAL_MANY_PORTS
-#define EXTRA_SERIAL_PORT_DEFNS			\
-	{ 0, BASE_BAUD, 0x1A0, 9, FOURPORT_FLAGS }, 	/* ttyS4 */	\
-	{ 0, BASE_BAUD, 0x1A8, 9, FOURPORT_FLAGS },	/* ttyS5 */	\
-	{ 0, BASE_BAUD, 0x1B0, 9, FOURPORT_FLAGS },	/* ttyS6 */	\
-	{ 0, BASE_BAUD, 0x1B8, 9, FOURPORT_FLAGS },	/* ttyS7 */	\
-	{ 0, BASE_BAUD, 0x2A0, 5, FOURPORT_FLAGS },	/* ttyS8 */	\
-	{ 0, BASE_BAUD, 0x2A8, 5, FOURPORT_FLAGS },	/* ttyS9 */	\
-	{ 0, BASE_BAUD, 0x2B0, 5, FOURPORT_FLAGS },	/* ttyS10 */	\
-	{ 0, BASE_BAUD, 0x2B8, 5, FOURPORT_FLAGS },	/* ttyS11 */	\
-	{ 0, BASE_BAUD, 0x330, 4, ACCENT_FLAGS },	/* ttyS12 */	\
-	{ 0, BASE_BAUD, 0x338, 4, ACCENT_FLAGS },	/* ttyS13 */	\
-	{ 0, BASE_BAUD, 0x000, 0, 0 },			/* ttyS14 (spare) */ \
-	{ 0, BASE_BAUD, 0x000, 0, 0 },			/* ttyS15 (spare) */ \
-	{ 0, BASE_BAUD, 0x100, 12, BOCA_FLAGS },	/* ttyS16 */	\
-	{ 0, BASE_BAUD, 0x108, 12, BOCA_FLAGS },	/* ttyS17 */	\
-	{ 0, BASE_BAUD, 0x110, 12, BOCA_FLAGS },	/* ttyS18 */	\
-	{ 0, BASE_BAUD, 0x118, 12, BOCA_FLAGS },	/* ttyS19 */	\
-	{ 0, BASE_BAUD, 0x120, 12, BOCA_FLAGS },	/* ttyS20 */	\
-	{ 0, BASE_BAUD, 0x128, 12, BOCA_FLAGS },	/* ttyS21 */	\
-	{ 0, BASE_BAUD, 0x130, 12, BOCA_FLAGS },	/* ttyS22 */	\
-	{ 0, BASE_BAUD, 0x138, 12, BOCA_FLAGS },	/* ttyS23 */	\
-	{ 0, BASE_BAUD, 0x140, 12, BOCA_FLAGS },	/* ttyS24 */	\
-	{ 0, BASE_BAUD, 0x148, 12, BOCA_FLAGS },	/* ttyS25 */	\
-	{ 0, BASE_BAUD, 0x150, 12, BOCA_FLAGS },	/* ttyS26 */	\
-	{ 0, BASE_BAUD, 0x158, 12, BOCA_FLAGS },	/* ttyS27 */	\
-	{ 0, BASE_BAUD, 0x160, 12, BOCA_FLAGS },	/* ttyS28 */	\
-	{ 0, BASE_BAUD, 0x168, 12, BOCA_FLAGS },	/* ttyS29 */	\
-	{ 0, BASE_BAUD, 0x170, 12, BOCA_FLAGS },	/* ttyS30 */	\
-	{ 0, BASE_BAUD, 0x178, 12, BOCA_FLAGS },	/* ttyS31 */
-#else /* CONFIG_SERIAL_MANY_PORTS */
-#define EXTRA_SERIAL_PORT_DEFNS
-#endif /* CONFIG_SERIAL_MANY_PORTS */
-
 #else /* CONFIG_HAVE_STD_PC_SERIAL_PORTS */
 #define STD_SERIAL_PORT_DEFNS
-#define EXTRA_SERIAL_PORT_DEFNS
 #endif /* CONFIG_HAVE_STD_PC_SERIAL_PORTS */
 
-/* You can have up to four HUB6's in the system, but I've only
- * included two cards here for a total of twelve ports.
- */
-#if (defined(CONFIG_HUB6) && defined(CONFIG_SERIAL_MANY_PORTS))
-#define HUB6_SERIAL_PORT_DFNS		\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,0) },  /* ttyS32 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,1) },  /* ttyS33 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,2) },  /* ttyS34 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,3) },  /* ttyS35 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,4) },  /* ttyS36 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,5) },  /* ttyS37 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,0) },  /* ttyS38 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,1) },  /* ttyS39 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,2) },  /* ttyS40 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,3) },  /* ttyS41 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,4) },  /* ttyS42 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,5) },  /* ttyS43 */
-#else
-#define HUB6_SERIAL_PORT_DFNS
-#endif
-
 #ifdef CONFIG_MOMENCO_JAGUAR_ATX
 /* Ordinary NS16552 duart with a 20MHz crystal.  */
 #define JAGUAR_ATX_UART_CLK	20000000
@@ -427,8 +345,6 @@
 	COBALT_SERIAL_PORT_DEFNS			\
 	DDB5477_SERIAL_PORT_DEFNS			\
 	EV96100_SERIAL_PORT_DEFNS			\
-	EXTRA_SERIAL_PORT_DEFNS				\
-	HUB6_SERIAL_PORT_DFNS				\
 	IP32_SERIAL_PORT_DEFNS                          \
 	ITE_SERIAL_PORT_DEFNS           		\
 	IVR_SERIAL_PORT_DEFNS           		\
diff --git a/include/asm-parisc/serial.h b/include/asm-parisc/serial.h
index 239c5dcab7e..82fd820d684 100644
--- a/include/asm-parisc/serial.h
+++ b/include/asm-parisc/serial.h
@@ -19,18 +19,4 @@
  * A500 w/ PCI serial cards: 5 + 4 * card ~= 17
  */
  
-#define STD_SERIAL_PORT_DEFNS			\
-	{ 0, },		/* ttyS0 */	\
-	{ 0, },		/* ttyS1 */	\
-	{ 0, },		/* ttyS2 */	\
-	{ 0, },		/* ttyS3 */	\
-	{ 0, },		/* ttyS4 */	\
-	{ 0, },		/* ttyS5 */	\
-	{ 0, },		/* ttyS6 */	\
-	{ 0, },		/* ttyS7 */	\
-	{ 0, },		/* ttyS8 */
-
-
-#define SERIAL_PORT_DFNS		\
-	STD_SERIAL_PORT_DEFNS
-
+#define SERIAL_PORT_DFNS
diff --git a/include/asm-ppc/pc_serial.h b/include/asm-ppc/pc_serial.h
index fa9cbb67ce3..8f994f9f885 100644
--- a/include/asm-ppc/pc_serial.h
+++ b/include/asm-ppc/pc_serial.h
@@ -35,93 +35,9 @@
 #define STD_COM4_FLAGS ASYNC_BOOT_AUTOCONF
 #endif
 
-#ifdef CONFIG_SERIAL_MANY_PORTS
-#define FOURPORT_FLAGS ASYNC_FOURPORT
-#define ACCENT_FLAGS 0
-#define BOCA_FLAGS 0
-#define HUB6_FLAGS 0
-#endif
-	
-/*
- * The following define the access methods for the HUB6 card. All
- * access is through two ports for all 24 possible chips. The card is
- * selected through the high 2 bits, the port on that card with the
- * "middle" 3 bits, and the register on that port with the bottom
- * 3 bits.
- *
- * While the access port and interrupt is configurable, the default
- * port locations are 0x302 for the port control register, and 0x303
- * for the data read/write register. Normally, the interrupt is at irq3
- * but can be anything from 3 to 7 inclusive. Note that using 3 will
- * require disabling com2.
- */
-
-#define C_P(card,port) (((card)<<6|(port)<<3) + 1)
-
-#define STD_SERIAL_PORT_DEFNS			\
+#define SERIAL_PORT_DFNS			\
 	/* UART CLK   PORT IRQ     FLAGS        */			\
 	{ 0, BASE_BAUD, 0x3F8, 4, STD_COM_FLAGS },	/* ttyS0 */	\
 	{ 0, BASE_BAUD, 0x2F8, 3, STD_COM_FLAGS },	/* ttyS1 */	\
 	{ 0, BASE_BAUD, 0x3E8, 4, STD_COM_FLAGS },	/* ttyS2 */	\
 	{ 0, BASE_BAUD, 0x2E8, 3, STD_COM4_FLAGS },	/* ttyS3 */
-
-
-#ifdef CONFIG_SERIAL_MANY_PORTS
-#define EXTRA_SERIAL_PORT_DEFNS			\
-	{ 0, BASE_BAUD, 0x1A0, 9, FOURPORT_FLAGS }, 	/* ttyS4 */	\
-	{ 0, BASE_BAUD, 0x1A8, 9, FOURPORT_FLAGS },	/* ttyS5 */	\
-	{ 0, BASE_BAUD, 0x1B0, 9, FOURPORT_FLAGS },	/* ttyS6 */	\
-	{ 0, BASE_BAUD, 0x1B8, 9, FOURPORT_FLAGS },	/* ttyS7 */	\
-	{ 0, BASE_BAUD, 0x2A0, 5, FOURPORT_FLAGS },	/* ttyS8 */	\
-	{ 0, BASE_BAUD, 0x2A8, 5, FOURPORT_FLAGS },	/* ttyS9 */	\
-	{ 0, BASE_BAUD, 0x2B0, 5, FOURPORT_FLAGS },	/* ttyS10 */	\
-	{ 0, BASE_BAUD, 0x2B8, 5, FOURPORT_FLAGS },	/* ttyS11 */	\
-	{ 0, BASE_BAUD, 0x330, 4, ACCENT_FLAGS },	/* ttyS12 */	\
-	{ 0, BASE_BAUD, 0x338, 4, ACCENT_FLAGS },	/* ttyS13 */	\
-	{ 0, BASE_BAUD, 0x000, 0, 0 },	/* ttyS14 (spare) */		\
-	{ 0, BASE_BAUD, 0x000, 0, 0 },	/* ttyS15 (spare) */		\
-	{ 0, BASE_BAUD, 0x100, 12, BOCA_FLAGS },	/* ttyS16 */	\
-	{ 0, BASE_BAUD, 0x108, 12, BOCA_FLAGS },	/* ttyS17 */	\
-	{ 0, BASE_BAUD, 0x110, 12, BOCA_FLAGS },	/* ttyS18 */	\
-	{ 0, BASE_BAUD, 0x118, 12, BOCA_FLAGS },	/* ttyS19 */	\
-	{ 0, BASE_BAUD, 0x120, 12, BOCA_FLAGS },	/* ttyS20 */	\
-	{ 0, BASE_BAUD, 0x128, 12, BOCA_FLAGS },	/* ttyS21 */	\
-	{ 0, BASE_BAUD, 0x130, 12, BOCA_FLAGS },	/* ttyS22 */	\
-	{ 0, BASE_BAUD, 0x138, 12, BOCA_FLAGS },	/* ttyS23 */	\
-	{ 0, BASE_BAUD, 0x140, 12, BOCA_FLAGS },	/* ttyS24 */	\
-	{ 0, BASE_BAUD, 0x148, 12, BOCA_FLAGS },	/* ttyS25 */	\
-	{ 0, BASE_BAUD, 0x150, 12, BOCA_FLAGS },	/* ttyS26 */	\
-	{ 0, BASE_BAUD, 0x158, 12, BOCA_FLAGS },	/* ttyS27 */	\
-	{ 0, BASE_BAUD, 0x160, 12, BOCA_FLAGS },	/* ttyS28 */	\
-	{ 0, BASE_BAUD, 0x168, 12, BOCA_FLAGS },	/* ttyS29 */	\
-	{ 0, BASE_BAUD, 0x170, 12, BOCA_FLAGS },	/* ttyS30 */	\
-	{ 0, BASE_BAUD, 0x178, 12, BOCA_FLAGS },	/* ttyS31 */
-#else
-#define EXTRA_SERIAL_PORT_DEFNS
-#endif
-
-/* You can have up to four HUB6's in the system, but I've only
- * included two cards here for a total of twelve ports.
- */
-#if (defined(CONFIG_HUB6) && defined(CONFIG_SERIAL_MANY_PORTS))
-#define HUB6_SERIAL_PORT_DFNS		\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,0) },  /* ttyS32 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,1) },  /* ttyS33 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,2) },  /* ttyS34 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,3) },  /* ttyS35 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,4) },  /* ttyS36 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,5) },  /* ttyS37 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,0) },  /* ttyS38 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,1) },  /* ttyS39 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,2) },  /* ttyS40 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,3) },  /* ttyS41 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,4) },  /* ttyS42 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,5) },  /* ttyS43 */
-#else
-#define HUB6_SERIAL_PORT_DFNS
-#endif
-
-#define SERIAL_PORT_DFNS		\
-	STD_SERIAL_PORT_DEFNS		\
-	EXTRA_SERIAL_PORT_DEFNS		\
-	HUB6_SERIAL_PORT_DFNS
diff --git a/include/asm-sh/bigsur/serial.h b/include/asm-sh/bigsur/serial.h
index 540f1220592..7233af42f75 100644
--- a/include/asm-sh/bigsur/serial.h
+++ b/include/asm-sh/bigsur/serial.h
@@ -14,13 +14,10 @@
 #define STD_COM_FLAGS (ASYNC_BOOT_AUTOCONF | ASYNC_SKIP_TEST)
 
 
-#define STD_SERIAL_PORT_DEFNS                   \
+#define SERIAL_PORT_DFNS                   \
         /* UART CLK   PORT IRQ     FLAGS        */                      \
         { 0, BASE_BAUD, 0x3F8, HD64465_IRQ_UART, STD_COM_FLAGS } /* ttyS0 */ 
 
-
-#define SERIAL_PORT_DFNS STD_SERIAL_PORT_DEFNS
-
 /* XXX: This should be moved ino irq.h */
 #define irq_cannonicalize(x) (x)
 
diff --git a/include/asm-sh/ec3104/serial.h b/include/asm-sh/ec3104/serial.h
index f8eb16312ed..cfe4d78ec1e 100644
--- a/include/asm-sh/ec3104/serial.h
+++ b/include/asm-sh/ec3104/serial.h
@@ -10,13 +10,11 @@
  * it's got the keyboard controller behind it so we can't really use it
  * (without moving the keyboard driver to userspace, which doesn't sound
  * like a very good idea) */
-#define STD_SERIAL_PORT_DEFNS			\
+#define SERIAL_PORT_DFNS			\
 	/* UART CLK   PORT IRQ     FLAGS        */			\
 	{ 0, BASE_BAUD, 0x11C00, EC3104_IRQBASE+7, STD_COM_FLAGS }, /* ttyS0 */	\
 	{ 0, BASE_BAUD, 0x12000, EC3104_IRQBASE+8, STD_COM_FLAGS }, /* ttyS1 */	\
 	{ 0, BASE_BAUD, 0x12400, EC3104_IRQBASE+9, STD_COM_FLAGS }, /* ttyS2 */
 
-#define SERIAL_PORT_DFNS STD_SERIAL_PORT_DEFNS
-
 /* XXX: This should be moved ino irq.h */
 #define irq_cannonicalize(x) (x)
diff --git a/include/asm-sh/serial.h b/include/asm-sh/serial.h
index 5474dbdbaa8..f51e232d5cd 100644
--- a/include/asm-sh/serial.h
+++ b/include/asm-sh/serial.h
@@ -29,20 +29,18 @@
 #ifdef CONFIG_HD64465
 #include <asm/hd64465.h>
 
-#define STD_SERIAL_PORT_DEFNS                   \
+#define SERIAL_PORT_DFNS                   \
         /* UART CLK   PORT IRQ     FLAGS        */                      \
         { 0, BASE_BAUD, 0x3F8, HD64465_IRQ_UART, STD_COM_FLAGS }  /* ttyS0 */
 
 #else
 
-#define STD_SERIAL_PORT_DEFNS			\
+#define SERIAL_PORT_DFNS			\
 	/* UART CLK   PORT IRQ     FLAGS        */			\
 	{ 0, BASE_BAUD, 0x3F8, 4, STD_COM_FLAGS },	/* ttyS0 */	\
 	{ 0, BASE_BAUD, 0x2F8, 3, STD_COM_FLAGS }	/* ttyS1 */
 
 #endif
 
-#define SERIAL_PORT_DFNS STD_SERIAL_PORT_DEFNS
-
 #endif
 #endif /* _ASM_SERIAL_H */
diff --git a/include/asm-sh64/serial.h b/include/asm-sh64/serial.h
index 8e39b4e90c7..29c9be15112 100644
--- a/include/asm-sh64/serial.h
+++ b/include/asm-sh64/serial.h
@@ -20,13 +20,11 @@
 
 #define STD_COM_FLAGS (ASYNC_BOOT_AUTOCONF | ASYNC_SKIP_TEST)
 
-#define STD_SERIAL_PORT_DEFNS			\
+#define SERIAL_PORT_DFNS			\
 	/* UART CLK   PORT IRQ     FLAGS        */			\
 	{ 0, BASE_BAUD, 0x3F8, 4, STD_COM_FLAGS },	/* ttyS0 */	\
 	{ 0, BASE_BAUD, 0x2F8, 3, STD_COM_FLAGS }	/* ttyS1 */
 
-#define SERIAL_PORT_DFNS STD_SERIAL_PORT_DEFNS
-
 /* XXX: This should be moved ino irq.h */
 #define irq_cannonicalize(x) (x)
 
diff --git a/include/asm-x86_64/serial.h b/include/asm-x86_64/serial.h
index dbab232044c..dc752eafa68 100644
--- a/include/asm-x86_64/serial.h
+++ b/include/asm-x86_64/serial.h
@@ -22,109 +22,9 @@
 #define STD_COM4_FLAGS ASYNC_BOOT_AUTOCONF
 #endif
 
-#ifdef CONFIG_SERIAL_MANY_PORTS
-#define FOURPORT_FLAGS ASYNC_FOURPORT
-#define ACCENT_FLAGS 0
-#define BOCA_FLAGS 0
-#define HUB6_FLAGS 0
-#endif
-
-#define MCA_COM_FLAGS	(STD_COM_FLAGS|ASYNC_BOOT_ONLYMCA)
-
-/*
- * The following define the access methods for the HUB6 card. All
- * access is through two ports for all 24 possible chips. The card is
- * selected through the high 2 bits, the port on that card with the
- * "middle" 3 bits, and the register on that port with the bottom
- * 3 bits.
- *
- * While the access port and interrupt is configurable, the default
- * port locations are 0x302 for the port control register, and 0x303
- * for the data read/write register. Normally, the interrupt is at irq3
- * but can be anything from 3 to 7 inclusive. Note that using 3 will
- * require disabling com2.
- */
-
-#define C_P(card,port) (((card)<<6|(port)<<3) + 1)
-
-#define STD_SERIAL_PORT_DEFNS			\
+#define SERIAL_PORT_DFNS			\
 	/* UART CLK   PORT IRQ     FLAGS        */			\
 	{ 0, BASE_BAUD, 0x3F8, 4, STD_COM_FLAGS },	/* ttyS0 */	\
 	{ 0, BASE_BAUD, 0x2F8, 3, STD_COM_FLAGS },	/* ttyS1 */	\
 	{ 0, BASE_BAUD, 0x3E8, 4, STD_COM_FLAGS },	/* ttyS2 */	\
 	{ 0, BASE_BAUD, 0x2E8, 3, STD_COM4_FLAGS },	/* ttyS3 */
-
-
-#ifdef CONFIG_SERIAL_MANY_PORTS
-#define EXTRA_SERIAL_PORT_DEFNS			\
-	{ 0, BASE_BAUD, 0x1A0, 9, FOURPORT_FLAGS }, 	/* ttyS4 */	\
-	{ 0, BASE_BAUD, 0x1A8, 9, FOURPORT_FLAGS },	/* ttyS5 */	\
-	{ 0, BASE_BAUD, 0x1B0, 9, FOURPORT_FLAGS },	/* ttyS6 */	\
-	{ 0, BASE_BAUD, 0x1B8, 9, FOURPORT_FLAGS },	/* ttyS7 */	\
-	{ 0, BASE_BAUD, 0x2A0, 5, FOURPORT_FLAGS },	/* ttyS8 */	\
-	{ 0, BASE_BAUD, 0x2A8, 5, FOURPORT_FLAGS },	/* ttyS9 */	\
-	{ 0, BASE_BAUD, 0x2B0, 5, FOURPORT_FLAGS },	/* ttyS10 */	\
-	{ 0, BASE_BAUD, 0x2B8, 5, FOURPORT_FLAGS },	/* ttyS11 */	\
-	{ 0, BASE_BAUD, 0x330, 4, ACCENT_FLAGS },	/* ttyS12 */	\
-	{ 0, BASE_BAUD, 0x338, 4, ACCENT_FLAGS },	/* ttyS13 */	\
-	{ 0, BASE_BAUD, 0x000, 0, 0 },	/* ttyS14 (spare) */		\
-	{ 0, BASE_BAUD, 0x000, 0, 0 },	/* ttyS15 (spare) */		\
-	{ 0, BASE_BAUD, 0x100, 12, BOCA_FLAGS },	/* ttyS16 */	\
-	{ 0, BASE_BAUD, 0x108, 12, BOCA_FLAGS },	/* ttyS17 */	\
-	{ 0, BASE_BAUD, 0x110, 12, BOCA_FLAGS },	/* ttyS18 */	\
-	{ 0, BASE_BAUD, 0x118, 12, BOCA_FLAGS },	/* ttyS19 */	\
-	{ 0, BASE_BAUD, 0x120, 12, BOCA_FLAGS },	/* ttyS20 */	\
-	{ 0, BASE_BAUD, 0x128, 12, BOCA_FLAGS },	/* ttyS21 */	\
-	{ 0, BASE_BAUD, 0x130, 12, BOCA_FLAGS },	/* ttyS22 */	\
-	{ 0, BASE_BAUD, 0x138, 12, BOCA_FLAGS },	/* ttyS23 */	\
-	{ 0, BASE_BAUD, 0x140, 12, BOCA_FLAGS },	/* ttyS24 */	\
-	{ 0, BASE_BAUD, 0x148, 12, BOCA_FLAGS },	/* ttyS25 */	\
-	{ 0, BASE_BAUD, 0x150, 12, BOCA_FLAGS },	/* ttyS26 */	\
-	{ 0, BASE_BAUD, 0x158, 12, BOCA_FLAGS },	/* ttyS27 */	\
-	{ 0, BASE_BAUD, 0x160, 12, BOCA_FLAGS },	/* ttyS28 */	\
-	{ 0, BASE_BAUD, 0x168, 12, BOCA_FLAGS },	/* ttyS29 */	\
-	{ 0, BASE_BAUD, 0x170, 12, BOCA_FLAGS },	/* ttyS30 */	\
-	{ 0, BASE_BAUD, 0x178, 12, BOCA_FLAGS },	/* ttyS31 */
-#else
-#define EXTRA_SERIAL_PORT_DEFNS
-#endif
-
-/* You can have up to four HUB6's in the system, but I've only
- * included two cards here for a total of twelve ports.
- */
-#if (defined(CONFIG_HUB6) && defined(CONFIG_SERIAL_MANY_PORTS))
-#define HUB6_SERIAL_PORT_DFNS		\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,0) },  /* ttyS32 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,1) },  /* ttyS33 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,2) },  /* ttyS34 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,3) },  /* ttyS35 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,4) },  /* ttyS36 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(0,5) },  /* ttyS37 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,0) },  /* ttyS38 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,1) },  /* ttyS39 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,2) },  /* ttyS40 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,3) },  /* ttyS41 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,4) },  /* ttyS42 */	\
-	{ 0, BASE_BAUD, 0x302, 3, HUB6_FLAGS, C_P(1,5) },  /* ttyS43 */
-#else
-#define HUB6_SERIAL_PORT_DFNS
-#endif
-
-#ifdef CONFIG_MCA
-#define MCA_SERIAL_PORT_DFNS			\
-	{ 0, BASE_BAUD, 0x3220, 3, MCA_COM_FLAGS },	\
-	{ 0, BASE_BAUD, 0x3228, 3, MCA_COM_FLAGS },	\
-	{ 0, BASE_BAUD, 0x4220, 3, MCA_COM_FLAGS },	\
-	{ 0, BASE_BAUD, 0x4228, 3, MCA_COM_FLAGS },	\
-	{ 0, BASE_BAUD, 0x5220, 3, MCA_COM_FLAGS },	\
-	{ 0, BASE_BAUD, 0x5228, 3, MCA_COM_FLAGS },
-#else
-#define MCA_SERIAL_PORT_DFNS
-#endif
-
-#define SERIAL_PORT_DFNS		\
-	STD_SERIAL_PORT_DEFNS		\
-	EXTRA_SERIAL_PORT_DEFNS		\
-	HUB6_SERIAL_PORT_DFNS		\
-	MCA_SERIAL_PORT_DFNS
-
-- 
cgit v1.2.3-70-g09d2


From bea248fb30c3122ece8c34798527fac431c1d7b0 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:07:09 +1000
Subject: [PATCH] ppc64: Remove lpqueue pointer from the paca on iSeries

The iSeries code keeps a pointer to the ItLpQueue in its paca struct. But
all these pointers end up pointing to the one place, ie. xItLpQueue.

So remove the pointer from the paca struct and just refer to xItLpQueue
directly where needed.

The only complication is that the spread_lpevents logic was implemented by
having a NULL lpqueue pointer in the paca on CPUs that weren't supposed to
process events. Instead we just compare the spread_lpevents value to the
processor id to get the same behaviour.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/ItLpQueue.c     | 16 +++++++++-------
 arch/ppc64/kernel/iSeries_setup.c |  6 ++----
 arch/ppc64/kernel/idle.c          |  4 ++--
 arch/ppc64/kernel/irq.c           |  6 ++----
 arch/ppc64/kernel/mf.c            |  5 ++---
 arch/ppc64/kernel/pacaData.c      |  1 -
 arch/ppc64/kernel/time.c          |  7 ++-----
 include/asm-ppc64/paca.h          |  2 --
 8 files changed, 19 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index cdea00d7707..e90dca8bd13 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -69,15 +69,17 @@ struct HvLpEvent * ItLpQueue_getNextLpEvent( struct ItLpQueue * lpQueue )
 	return nextLpEvent;
 }
 
+unsigned long spread_lpevents = 1;
+
 int ItLpQueue_isLpIntPending( struct ItLpQueue * lpQueue )
 {
-	int retval = 0;
-	struct HvLpEvent * nextLpEvent;
-	if ( lpQueue ) {
-		nextLpEvent = (struct HvLpEvent *)lpQueue->xSlicCurEventPtr;
-		retval = nextLpEvent->xFlags.xValid | lpQueue->xPlicOverflowIntPending;
-	}
-	return retval;
+	struct HvLpEvent *next_event;
+
+	if (smp_processor_id() >= spread_lpevents)
+		return 0;
+
+	next_event = (struct HvLpEvent *)lpQueue->xSlicCurEventPtr;
+	return next_event->xFlags.xValid | lpQueue->xPlicOverflowIntPending;
 }
 
 void ItLpQueue_clearValid( struct HvLpEvent * event )
diff --git a/arch/ppc64/kernel/iSeries_setup.c b/arch/ppc64/kernel/iSeries_setup.c
index 86966ce76b5..2049b6dbafc 100644
--- a/arch/ppc64/kernel/iSeries_setup.c
+++ b/arch/ppc64/kernel/iSeries_setup.c
@@ -855,17 +855,15 @@ late_initcall(iSeries_src_init);
 
 static int set_spread_lpevents(char *str)
 {
-	unsigned long i;
 	unsigned long val = simple_strtoul(str, NULL, 0);
+	extern unsigned long spread_lpevents;
 
 	/*
 	 * The parameter is the number of processors to share in processing
 	 * lp events.
 	 */
 	if (( val > 0) && (val <= NR_CPUS)) {
-		for (i = 1; i < val; ++i)
-			paca[i].lpqueue_ptr = paca[0].lpqueue_ptr;
-
+		spread_lpevents = val;
 		printk("lpevent processing spread over %ld processors\n", val);
 	} else {
 		printk("invalid spread_lpevents %ld\n", val);
diff --git a/arch/ppc64/kernel/idle.c b/arch/ppc64/kernel/idle.c
index bdf13b4dc1c..63977a7a309 100644
--- a/arch/ppc64/kernel/idle.c
+++ b/arch/ppc64/kernel/idle.c
@@ -88,7 +88,7 @@ static int iSeries_idle(void)
 
 	while (1) {
 		if (lpaca->lppaca.shared_proc) {
-			if (ItLpQueue_isLpIntPending(lpaca->lpqueue_ptr))
+			if (ItLpQueue_isLpIntPending(&xItLpQueue))
 				process_iSeries_events();
 			if (!need_resched())
 				yield_shared_processor();
@@ -100,7 +100,7 @@ static int iSeries_idle(void)
 
 				while (!need_resched()) {
 					HMT_medium();
-					if (ItLpQueue_isLpIntPending(lpaca->lpqueue_ptr))
+					if (ItLpQueue_isLpIntPending(&xItLpQueue))
 						process_iSeries_events();
 					HMT_low();
 				}
diff --git a/arch/ppc64/kernel/irq.c b/arch/ppc64/kernel/irq.c
index 3defc8c33ad..b1e6acb02a9 100644
--- a/arch/ppc64/kernel/irq.c
+++ b/arch/ppc64/kernel/irq.c
@@ -269,7 +269,6 @@ out:
 void do_IRQ(struct pt_regs *regs)
 {
 	struct paca_struct *lpaca;
-	struct ItLpQueue *lpq;
 
 	irq_enter();
 
@@ -295,9 +294,8 @@ void do_IRQ(struct pt_regs *regs)
 		iSeries_smp_message_recv(regs);
 	}
 #endif /* CONFIG_SMP */
-	lpq = lpaca->lpqueue_ptr;
-	if (lpq && ItLpQueue_isLpIntPending(lpq))
-		lpevent_count += ItLpQueue_process(lpq, regs);
+	if (ItLpQueue_isLpIntPending(&xItLpQueue))
+		lpevent_count += ItLpQueue_process(&xItLpQueue, regs);
 
 	irq_exit();
 
diff --git a/arch/ppc64/kernel/mf.c b/arch/ppc64/kernel/mf.c
index d98bebf7042..d6a297a4feb 100644
--- a/arch/ppc64/kernel/mf.c
+++ b/arch/ppc64/kernel/mf.c
@@ -802,9 +802,8 @@ int mf_get_boot_rtc(struct rtc_time *tm)
 	/* We need to poll here as we are not yet taking interrupts */
 	while (rtc_data.busy) {
 		extern unsigned long lpevent_count;
-		struct ItLpQueue *lpq = get_paca()->lpqueue_ptr;
-		if (lpq && ItLpQueue_isLpIntPending(lpq))
-			lpevent_count += ItLpQueue_process(lpq, NULL);
+		if (ItLpQueue_isLpIntPending(&xItLpQueue))
+			lpevent_count += ItLpQueue_process(&xItLpQueue, NULL);
 	}
 	return rtc_set_tm(rtc_data.rc, rtc_data.ce_msg.ce_msg, tm);
 }
diff --git a/arch/ppc64/kernel/pacaData.c b/arch/ppc64/kernel/pacaData.c
index a3e0975c26c..ebfb517019e 100644
--- a/arch/ppc64/kernel/pacaData.c
+++ b/arch/ppc64/kernel/pacaData.c
@@ -45,7 +45,6 @@ extern unsigned long __toc_start;
 #ifdef CONFIG_PPC_ISERIES
 #define EXTRA_INITS(number, lpq)					    \
 	.lppaca_ptr = &paca[number].lppaca,				    \
-	.lpqueue_ptr = (lpq),		/* &xItLpQueue, */		    \
 	.reg_save_ptr = &paca[number].reg_save,				    \
 	.reg_save = {							    \
 		.xDesc = 0xd397d9e2,	/* "LpRS" */			    \
diff --git a/arch/ppc64/kernel/time.c b/arch/ppc64/kernel/time.c
index 2a532db9138..cdc43afb563 100644
--- a/arch/ppc64/kernel/time.c
+++ b/arch/ppc64/kernel/time.c
@@ -367,11 +367,8 @@ int timer_interrupt(struct pt_regs * regs)
 	set_dec(next_dec);
 
 #ifdef CONFIG_PPC_ISERIES
-	{
-		struct ItLpQueue *lpq = lpaca->lpqueue_ptr;
-		if (lpq && ItLpQueue_isLpIntPending(lpq))
-			lpevent_count += ItLpQueue_process(lpq, regs);
-	}
+	if (ItLpQueue_isLpIntPending(&xItLpQueue))
+		lpevent_count += ItLpQueue_process(&xItLpQueue, regs);
 #endif
 
 /* collect purr register values often, for accurate calculations */
diff --git a/include/asm-ppc64/paca.h b/include/asm-ppc64/paca.h
index ae76cae1483..0146f51684e 100644
--- a/include/asm-ppc64/paca.h
+++ b/include/asm-ppc64/paca.h
@@ -20,7 +20,6 @@
 #include	<asm/types.h>
 #include	<asm/lppaca.h>
 #include	<asm/iSeries/ItLpRegSave.h>
-#include	<asm/iSeries/ItLpQueue.h>
 #include	<asm/mmu.h>
 
 register struct paca_struct *local_paca asm("r13");
@@ -62,7 +61,6 @@ struct paca_struct {
 	u16 paca_index;			/* Logical processor number */
 
 	u32 default_decr;		/* Default decrementer value */
-	struct ItLpQueue *lpqueue_ptr;	/* LpQueue handled by this CPU */
 	u64 kernel_toc;			/* Kernel TOC address */
 	u64 stab_real;			/* Absolute address of segment table */
 	u64 stab_addr;			/* Virtual address of segment table */
-- 
cgit v1.2.3-70-g09d2


From 1b19bc721416ae5bc813521d9e010a89f4816120 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:07:57 +1000
Subject: [PATCH] ppc64: Don't pass the pointers to xItLpQueue around

Because there's only one ItLpQueue and we know where it is, ie. xItLpQueue,
there's no point passing pointers to it it around all over the place.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/ItLpQueue.c         | 46 +++++++++++++++++------------------
 arch/ppc64/kernel/idle.c              |  4 +--
 arch/ppc64/kernel/irq.c               |  4 +--
 arch/ppc64/kernel/mf.c                |  4 +--
 arch/ppc64/kernel/time.c              |  4 +--
 include/asm-ppc64/iSeries/ItLpQueue.h |  6 ++---
 6 files changed, 34 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index 61be23ed500..35f6deac0b9 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -17,10 +17,10 @@
 #include <asm/iSeries/HvLpEvent.h>
 #include <asm/iSeries/HvCallEvent.h>
 
-static __inline__ int set_inUse( struct ItLpQueue * lpQueue )
+static __inline__ int set_inUse(void)
 {
 	int t;
-	u32 * inUseP = &(lpQueue->xInUseWord);
+	u32 * inUseP = &xItLpQueue.xInUseWord;
 
 	__asm__ __volatile__("\n\
 1:	lwarx	%0,0,%2		\n\
@@ -31,37 +31,37 @@ static __inline__ int set_inUse( struct ItLpQueue * lpQueue )
 	stwcx.	%0,0,%2		\n\
 	bne-	1b		\n\
 2:	eieio"
-	: "=&r" (t), "=m" (lpQueue->xInUseWord)
-	: "r" (inUseP), "m" (lpQueue->xInUseWord)
+	: "=&r" (t), "=m" (xItLpQueue.xInUseWord)
+	: "r" (inUseP), "m" (xItLpQueue.xInUseWord)
 	: "cc");
 
 	return t;
 }
 
-static __inline__ void clear_inUse( struct ItLpQueue * lpQueue )
+static __inline__ void clear_inUse(void)
 {
-	lpQueue->xInUseWord = 0;
+	xItLpQueue.xInUseWord = 0;
 }
 
 /* Array of LpEvent handler functions */
 extern LpEventHandler lpEventHandler[HvLpEvent_Type_NumTypes];
 unsigned long ItLpQueueInProcess = 0;
 
-struct HvLpEvent * ItLpQueue_getNextLpEvent( struct ItLpQueue * lpQueue )
+struct HvLpEvent * ItLpQueue_getNextLpEvent(void)
 {
 	struct HvLpEvent * nextLpEvent = 
-		(struct HvLpEvent *)lpQueue->xSlicCurEventPtr;
+		(struct HvLpEvent *)xItLpQueue.xSlicCurEventPtr;
 	if ( nextLpEvent->xFlags.xValid ) {
 		/* rmb() needed only for weakly consistent machines (regatta) */
 		rmb();
 		/* Set pointer to next potential event */
-		lpQueue->xSlicCurEventPtr += ((nextLpEvent->xSizeMinus1 +
+		xItLpQueue.xSlicCurEventPtr += ((nextLpEvent->xSizeMinus1 +
 				      LpEventAlign ) /
 				      LpEventAlign ) *
 				      LpEventAlign;
 		/* Wrap to beginning if no room at end */
-		if (lpQueue->xSlicCurEventPtr > lpQueue->xSlicLastValidEventPtr)
-			lpQueue->xSlicCurEventPtr = lpQueue->xSlicEventStackPtr;
+		if (xItLpQueue.xSlicCurEventPtr > xItLpQueue.xSlicLastValidEventPtr)
+			xItLpQueue.xSlicCurEventPtr = xItLpQueue.xSlicEventStackPtr;
 	}
 	else 
 		nextLpEvent = NULL;
@@ -71,15 +71,15 @@ struct HvLpEvent * ItLpQueue_getNextLpEvent( struct ItLpQueue * lpQueue )
 
 static unsigned long spread_lpevents = NR_CPUS;
 
-int ItLpQueue_isLpIntPending( struct ItLpQueue * lpQueue )
+int ItLpQueue_isLpIntPending(void)
 {
 	struct HvLpEvent *next_event;
 
 	if (smp_processor_id() >= spread_lpevents)
 		return 0;
 
-	next_event = (struct HvLpEvent *)lpQueue->xSlicCurEventPtr;
-	return next_event->xFlags.xValid | lpQueue->xPlicOverflowIntPending;
+	next_event = (struct HvLpEvent *)xItLpQueue.xSlicCurEventPtr;
+	return next_event->xFlags.xValid | xItLpQueue.xPlicOverflowIntPending;
 }
 
 void ItLpQueue_clearValid( struct HvLpEvent * event )
@@ -104,13 +104,13 @@ void ItLpQueue_clearValid( struct HvLpEvent * event )
 	event->xFlags.xValid = 0;
 }
 
-unsigned ItLpQueue_process( struct ItLpQueue * lpQueue, struct pt_regs *regs )
+unsigned ItLpQueue_process(struct pt_regs *regs)
 {
 	unsigned numIntsProcessed = 0;
 	struct HvLpEvent * nextLpEvent;
 
 	/* If we have recursed, just return */
-	if ( !set_inUse( lpQueue ) )
+	if ( !set_inUse() )
 		return 0;
 	
 	if (ItLpQueueInProcess == 0)
@@ -119,13 +119,13 @@ unsigned ItLpQueue_process( struct ItLpQueue * lpQueue, struct pt_regs *regs )
 		BUG();
 
 	for (;;) {
-		nextLpEvent = ItLpQueue_getNextLpEvent( lpQueue );
+		nextLpEvent = ItLpQueue_getNextLpEvent();
 		if ( nextLpEvent ) {
 			/* Count events to return to caller
-			 * and count processed events in lpQueue
+			 * and count processed events in xItLpQueue
  			 */
 			++numIntsProcessed;
-			lpQueue->xLpIntCount++;		
+			xItLpQueue.xLpIntCount++;
 			/* Call appropriate handler here, passing 
 			 * a pointer to the LpEvent.  The handler
 			 * must make a copy of the LpEvent if it
@@ -140,7 +140,7 @@ unsigned ItLpQueue_process( struct ItLpQueue * lpQueue, struct pt_regs *regs )
 			 * here!
   			 */
 			if ( nextLpEvent->xType < HvLpEvent_Type_NumTypes )
-				lpQueue->xLpIntCountByType[nextLpEvent->xType]++;
+				xItLpQueue.xLpIntCountByType[nextLpEvent->xType]++;
 			if ( nextLpEvent->xType < HvLpEvent_Type_NumTypes &&
 			     lpEventHandler[nextLpEvent->xType] ) 
 				lpEventHandler[nextLpEvent->xType](nextLpEvent, regs);
@@ -148,19 +148,19 @@ unsigned ItLpQueue_process( struct ItLpQueue * lpQueue, struct pt_regs *regs )
 				printk(KERN_INFO "Unexpected Lp Event type=%d\n", nextLpEvent->xType );
 			
 			ItLpQueue_clearValid( nextLpEvent );
-		} else if ( lpQueue->xPlicOverflowIntPending )
+		} else if ( xItLpQueue.xPlicOverflowIntPending )
 			/*
 			 * No more valid events. If overflow events are
 			 * pending process them
 			 */
-			HvCallEvent_getOverflowLpEvents( lpQueue->xIndex);
+			HvCallEvent_getOverflowLpEvents( xItLpQueue.xIndex);
 		else
 			break;
 	}
 
 	ItLpQueueInProcess = 0;
 	mb();
-	clear_inUse( lpQueue );
+	clear_inUse();
 
 	get_paca()->lpevent_count += numIntsProcessed;
 
diff --git a/arch/ppc64/kernel/idle.c b/arch/ppc64/kernel/idle.c
index 63977a7a309..a7ebd0238d3 100644
--- a/arch/ppc64/kernel/idle.c
+++ b/arch/ppc64/kernel/idle.c
@@ -88,7 +88,7 @@ static int iSeries_idle(void)
 
 	while (1) {
 		if (lpaca->lppaca.shared_proc) {
-			if (ItLpQueue_isLpIntPending(&xItLpQueue))
+			if (ItLpQueue_isLpIntPending())
 				process_iSeries_events();
 			if (!need_resched())
 				yield_shared_processor();
@@ -100,7 +100,7 @@ static int iSeries_idle(void)
 
 				while (!need_resched()) {
 					HMT_medium();
-					if (ItLpQueue_isLpIntPending(&xItLpQueue))
+					if (ItLpQueue_isLpIntPending())
 						process_iSeries_events();
 					HMT_low();
 				}
diff --git a/arch/ppc64/kernel/irq.c b/arch/ppc64/kernel/irq.c
index b1e6acb02a9..46a7151ad6d 100644
--- a/arch/ppc64/kernel/irq.c
+++ b/arch/ppc64/kernel/irq.c
@@ -294,8 +294,8 @@ void do_IRQ(struct pt_regs *regs)
 		iSeries_smp_message_recv(regs);
 	}
 #endif /* CONFIG_SMP */
-	if (ItLpQueue_isLpIntPending(&xItLpQueue))
-		lpevent_count += ItLpQueue_process(&xItLpQueue, regs);
+	if (ItLpQueue_isLpIntPending())
+		lpevent_count += ItLpQueue_process(regs);
 
 	irq_exit();
 
diff --git a/arch/ppc64/kernel/mf.c b/arch/ppc64/kernel/mf.c
index d6a297a4feb..ef920699802 100644
--- a/arch/ppc64/kernel/mf.c
+++ b/arch/ppc64/kernel/mf.c
@@ -802,8 +802,8 @@ int mf_get_boot_rtc(struct rtc_time *tm)
 	/* We need to poll here as we are not yet taking interrupts */
 	while (rtc_data.busy) {
 		extern unsigned long lpevent_count;
-		if (ItLpQueue_isLpIntPending(&xItLpQueue))
-			lpevent_count += ItLpQueue_process(&xItLpQueue, NULL);
+		if (ItLpQueue_isLpIntPending())
+			lpevent_count += ItLpQueue_process(NULL);
 	}
 	return rtc_set_tm(rtc_data.rc, rtc_data.ce_msg.ce_msg, tm);
 }
diff --git a/arch/ppc64/kernel/time.c b/arch/ppc64/kernel/time.c
index cdc43afb563..c133f9c28c5 100644
--- a/arch/ppc64/kernel/time.c
+++ b/arch/ppc64/kernel/time.c
@@ -367,8 +367,8 @@ int timer_interrupt(struct pt_regs * regs)
 	set_dec(next_dec);
 
 #ifdef CONFIG_PPC_ISERIES
-	if (ItLpQueue_isLpIntPending(&xItLpQueue))
-		lpevent_count += ItLpQueue_process(&xItLpQueue, regs);
+	if (ItLpQueue_isLpIntPending())
+		lpevent_count += ItLpQueue_process(regs);
 #endif
 
 /* collect purr register values often, for accurate calculations */
diff --git a/include/asm-ppc64/iSeries/ItLpQueue.h b/include/asm-ppc64/iSeries/ItLpQueue.h
index 393299e04d7..832497e3259 100644
--- a/include/asm-ppc64/iSeries/ItLpQueue.h
+++ b/include/asm-ppc64/iSeries/ItLpQueue.h
@@ -76,9 +76,9 @@ struct ItLpQueue {
 
 extern struct ItLpQueue xItLpQueue;
 
-extern struct HvLpEvent *ItLpQueue_getNextLpEvent(struct ItLpQueue *);
-extern int ItLpQueue_isLpIntPending(struct ItLpQueue *);
-extern unsigned ItLpQueue_process(struct ItLpQueue *, struct pt_regs *);
+extern struct HvLpEvent *ItLpQueue_getNextLpEvent(void);
+extern int ItLpQueue_isLpIntPending(void);
+extern unsigned ItLpQueue_process(struct pt_regs *);
 extern void ItLpQueue_clearValid(struct HvLpEvent *);
 
 #endif /* _ITLPQUEUE_H */
-- 
cgit v1.2.3-70-g09d2


From 512d31d6a824a961f39b418f11480f678320e4f3 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:08:27 +1000
Subject: [PATCH] ppc64: Move initialisation of xItLpQueue into ItLpQueue.c

The xItLpQueue is initalised manually in iSeries_setup_arch().  Move
this code into ItLpQueue.c for a cleaner separation.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/ItLpQueue.c         | 22 ++++++++++++++++++++++
 arch/ppc64/kernel/iSeries_setup.c     | 21 +--------------------
 include/asm-ppc64/iSeries/ItLpQueue.h |  1 +
 3 files changed, 24 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index 35f6deac0b9..091aaed4842 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -11,6 +11,7 @@
 #include <linux/stddef.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
+#include <linux/bootmem.h>
 #include <asm/system.h>
 #include <asm/paca.h>
 #include <asm/iSeries/ItLpQueue.h>
@@ -186,3 +187,24 @@ static int set_spread_lpevents(char *str)
 }
 __setup("spread_lpevents=", set_spread_lpevents);
 
+void setup_hvlpevent_queue(void)
+{
+	void *eventStack;
+
+	/*
+	 * Allocate a page for the Event Stack. The Hypervisor needs the
+	 * absolute real address, so we subtract out the KERNELBASE and add
+	 * in the absolute real address of the kernel load area.
+	 */
+	eventStack = alloc_bootmem_pages(LpEventStackSize);
+	memset(eventStack, 0, LpEventStackSize);
+
+	/* Invoke the hypervisor to initialize the event stack */
+	HvCallEvent_setLpEventStack(0, eventStack, LpEventStackSize);
+
+	xItLpQueue.xSlicEventStackPtr = (char *)eventStack;
+	xItLpQueue.xSlicCurEventPtr = (char *)eventStack;
+	xItLpQueue.xSlicLastValidEventPtr = (char *)eventStack +
+					(LpEventStackSize - LpEventMaxSize);
+	xItLpQueue.xIndex = 0;
+}
diff --git a/arch/ppc64/kernel/iSeries_setup.c b/arch/ppc64/kernel/iSeries_setup.c
index 78f28356025..b3f770f6d40 100644
--- a/arch/ppc64/kernel/iSeries_setup.c
+++ b/arch/ppc64/kernel/iSeries_setup.c
@@ -24,7 +24,6 @@
 #include <linux/smp.h>
 #include <linux/param.h>
 #include <linux/string.h>
-#include <linux/bootmem.h>
 #include <linux/initrd.h>
 #include <linux/seq_file.h>
 #include <linux/kdev_t.h>
@@ -676,7 +675,6 @@ static void __init iSeries_bolt_kernel(unsigned long saddr, unsigned long eaddr)
  */
 static void __init iSeries_setup_arch(void)
 {
-	void *eventStack;
 	unsigned procIx = get_paca()->lppaca.dyn_hv_phys_proc_index;
 
 	/* Add an eye catcher and the systemcfg layout version number */
@@ -685,24 +683,7 @@ static void __init iSeries_setup_arch(void)
 	systemcfg->version.minor = SYSTEMCFG_MINOR;
 
 	/* Setup the Lp Event Queue */
-
-	/* Allocate a page for the Event Stack
-	 * The hypervisor wants the absolute real address, so
-	 * we subtract out the KERNELBASE and add in the
-	 * absolute real address of the kernel load area
-	 */
-	eventStack = alloc_bootmem_pages(LpEventStackSize);
-	memset(eventStack, 0, LpEventStackSize);
-
-	/* Invoke the hypervisor to initialize the event stack */
-	HvCallEvent_setLpEventStack(0, eventStack, LpEventStackSize);
-
-	/* Initialize fields in our Lp Event Queue */
-	xItLpQueue.xSlicEventStackPtr = (char *)eventStack;
-	xItLpQueue.xSlicCurEventPtr = (char *)eventStack;
-	xItLpQueue.xSlicLastValidEventPtr = (char *)eventStack +
-					(LpEventStackSize - LpEventMaxSize);
-	xItLpQueue.xIndex = 0;
+	setup_hvlpevent_queue();
 
 	/* Compute processor frequency */
 	procFreqHz = ((1UL << 34) * 1000000) /
diff --git a/include/asm-ppc64/iSeries/ItLpQueue.h b/include/asm-ppc64/iSeries/ItLpQueue.h
index 832497e3259..de90feedbe1 100644
--- a/include/asm-ppc64/iSeries/ItLpQueue.h
+++ b/include/asm-ppc64/iSeries/ItLpQueue.h
@@ -80,5 +80,6 @@ extern struct HvLpEvent *ItLpQueue_getNextLpEvent(void);
 extern int ItLpQueue_isLpIntPending(void);
 extern unsigned ItLpQueue_process(struct pt_regs *);
 extern void ItLpQueue_clearValid(struct HvLpEvent *);
+extern void setup_hvlpevent_queue(void);
 
 #endif /* _ITLPQUEUE_H */
-- 
cgit v1.2.3-70-g09d2


From 0f6014b37e25e50724867c0a4127615427ec2a75 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:08:56 +1000
Subject: [PATCH] ppc64: Make two ItLpQueue related functions static

External parties don't need to use ItLpQueue_getNextLpEvent() or
ItLpQueue_clearValid(), they're internal to ItLpQueue.c

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/ItLpQueue.c         | 4 ++--
 include/asm-ppc64/iSeries/ItLpQueue.h | 2 --
 2 files changed, 2 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index 11cd31dfcef..ac0b05a6e06 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -62,7 +62,7 @@ static __inline__ void clear_inUse(void)
 extern LpEventHandler lpEventHandler[HvLpEvent_Type_NumTypes];
 unsigned long ItLpQueueInProcess = 0;
 
-struct HvLpEvent * ItLpQueue_getNextLpEvent(void)
+static struct HvLpEvent * ItLpQueue_getNextLpEvent(void)
 {
 	struct HvLpEvent * nextLpEvent = 
 		(struct HvLpEvent *)xItLpQueue.xSlicCurEventPtr;
@@ -97,7 +97,7 @@ int ItLpQueue_isLpIntPending(void)
 	return next_event->xFlags.xValid | xItLpQueue.xPlicOverflowIntPending;
 }
 
-void ItLpQueue_clearValid( struct HvLpEvent * event )
+static void ItLpQueue_clearValid( struct HvLpEvent * event )
 {
 	/* Clear the valid bit of the event
 	 * Also clear bits within this event that might
diff --git a/include/asm-ppc64/iSeries/ItLpQueue.h b/include/asm-ppc64/iSeries/ItLpQueue.h
index de90feedbe1..be1cb7f3f88 100644
--- a/include/asm-ppc64/iSeries/ItLpQueue.h
+++ b/include/asm-ppc64/iSeries/ItLpQueue.h
@@ -76,10 +76,8 @@ struct ItLpQueue {
 
 extern struct ItLpQueue xItLpQueue;
 
-extern struct HvLpEvent *ItLpQueue_getNextLpEvent(void);
 extern int ItLpQueue_isLpIntPending(void);
 extern unsigned ItLpQueue_process(struct pt_regs *);
-extern void ItLpQueue_clearValid(struct HvLpEvent *);
 extern void setup_hvlpevent_queue(void);
 
 #endif /* _ITLPQUEUE_H */
-- 
cgit v1.2.3-70-g09d2


From a61874648d14450f4d397489527998e3dd1119de Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:15:32 +1000
Subject: [PATCH] ppc64: Rename xItLpQueue to hvlpevent_queue

The xItLpQueue is a queue of HvLpEvents that we're given by the Hypervisor.
Rename xItLpQueue to hvlpevent_queue and make the type struct hvlpevent_queue.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/ItLpQueue.c         | 44 +++++++++++++++++------------------
 arch/ppc64/kernel/LparData.c          |  4 ++--
 include/asm-ppc64/iSeries/ItLpQueue.h |  4 ++--
 3 files changed, 26 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index 7ddbfb9750d..e55fe1a2725 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -26,7 +26,7 @@
  *
  * It is written to by the hypervisor so cannot end up in the BSS.
  */
-struct ItLpQueue xItLpQueue __attribute__((__section__(".data")));
+struct hvlpevent_queue hvlpevent_queue __attribute__((__section__(".data")));
 
 static char *event_types[9] = {
 	"Hypervisor\t\t",
@@ -43,7 +43,7 @@ static char *event_types[9] = {
 static __inline__ int set_inUse(void)
 {
 	int t;
-	u32 * inUseP = &xItLpQueue.xInUseWord;
+	u32 * inUseP = &hvlpevent_queue.xInUseWord;
 
 	__asm__ __volatile__("\n\
 1:	lwarx	%0,0,%2		\n\
@@ -54,8 +54,8 @@ static __inline__ int set_inUse(void)
 	stwcx.	%0,0,%2		\n\
 	bne-	1b		\n\
 2:	eieio"
-	: "=&r" (t), "=m" (xItLpQueue.xInUseWord)
-	: "r" (inUseP), "m" (xItLpQueue.xInUseWord)
+	: "=&r" (t), "=m" (hvlpevent_queue.xInUseWord)
+	: "r" (inUseP), "m" (hvlpevent_queue.xInUseWord)
 	: "cc");
 
 	return t;
@@ -63,7 +63,7 @@ static __inline__ int set_inUse(void)
 
 static __inline__ void clear_inUse(void)
 {
-	xItLpQueue.xInUseWord = 0;
+	hvlpevent_queue.xInUseWord = 0;
 }
 
 /* Array of LpEvent handler functions */
@@ -73,18 +73,18 @@ unsigned long ItLpQueueInProcess = 0;
 static struct HvLpEvent * ItLpQueue_getNextLpEvent(void)
 {
 	struct HvLpEvent * nextLpEvent = 
-		(struct HvLpEvent *)xItLpQueue.xSlicCurEventPtr;
+		(struct HvLpEvent *)hvlpevent_queue.xSlicCurEventPtr;
 	if ( nextLpEvent->xFlags.xValid ) {
 		/* rmb() needed only for weakly consistent machines (regatta) */
 		rmb();
 		/* Set pointer to next potential event */
-		xItLpQueue.xSlicCurEventPtr += ((nextLpEvent->xSizeMinus1 +
+		hvlpevent_queue.xSlicCurEventPtr += ((nextLpEvent->xSizeMinus1 +
 				      LpEventAlign ) /
 				      LpEventAlign ) *
 				      LpEventAlign;
 		/* Wrap to beginning if no room at end */
-		if (xItLpQueue.xSlicCurEventPtr > xItLpQueue.xSlicLastValidEventPtr)
-			xItLpQueue.xSlicCurEventPtr = xItLpQueue.xSlicEventStackPtr;
+		if (hvlpevent_queue.xSlicCurEventPtr > hvlpevent_queue.xSlicLastValidEventPtr)
+			hvlpevent_queue.xSlicCurEventPtr = hvlpevent_queue.xSlicEventStackPtr;
 	}
 	else 
 		nextLpEvent = NULL;
@@ -101,8 +101,8 @@ int ItLpQueue_isLpIntPending(void)
 	if (smp_processor_id() >= spread_lpevents)
 		return 0;
 
-	next_event = (struct HvLpEvent *)xItLpQueue.xSlicCurEventPtr;
-	return next_event->xFlags.xValid | xItLpQueue.xPlicOverflowIntPending;
+	next_event = (struct HvLpEvent *)hvlpevent_queue.xSlicCurEventPtr;
+	return next_event->xFlags.xValid | hvlpevent_queue.xPlicOverflowIntPending;
 }
 
 static void ItLpQueue_clearValid( struct HvLpEvent * event )
@@ -145,10 +145,10 @@ unsigned ItLpQueue_process(struct pt_regs *regs)
 		nextLpEvent = ItLpQueue_getNextLpEvent();
 		if ( nextLpEvent ) {
 			/* Count events to return to caller
-			 * and count processed events in xItLpQueue
+			 * and count processed events in hvlpevent_queue
  			 */
 			++numIntsProcessed;
-			xItLpQueue.xLpIntCount++;
+			hvlpevent_queue.xLpIntCount++;
 			/* Call appropriate handler here, passing 
 			 * a pointer to the LpEvent.  The handler
 			 * must make a copy of the LpEvent if it
@@ -163,7 +163,7 @@ unsigned ItLpQueue_process(struct pt_regs *regs)
 			 * here!
   			 */
 			if ( nextLpEvent->xType < HvLpEvent_Type_NumTypes )
-				xItLpQueue.xLpIntCountByType[nextLpEvent->xType]++;
+				hvlpevent_queue.xLpIntCountByType[nextLpEvent->xType]++;
 			if ( nextLpEvent->xType < HvLpEvent_Type_NumTypes &&
 			     lpEventHandler[nextLpEvent->xType] ) 
 				lpEventHandler[nextLpEvent->xType](nextLpEvent, regs);
@@ -171,12 +171,12 @@ unsigned ItLpQueue_process(struct pt_regs *regs)
 				printk(KERN_INFO "Unexpected Lp Event type=%d\n", nextLpEvent->xType );
 			
 			ItLpQueue_clearValid( nextLpEvent );
-		} else if ( xItLpQueue.xPlicOverflowIntPending )
+		} else if ( hvlpevent_queue.xPlicOverflowIntPending )
 			/*
 			 * No more valid events. If overflow events are
 			 * pending process them
 			 */
-			HvCallEvent_getOverflowLpEvents( xItLpQueue.xIndex);
+			HvCallEvent_getOverflowLpEvents( hvlpevent_queue.xIndex);
 		else
 			break;
 	}
@@ -224,11 +224,11 @@ void setup_hvlpevent_queue(void)
 	/* Invoke the hypervisor to initialize the event stack */
 	HvCallEvent_setLpEventStack(0, eventStack, LpEventStackSize);
 
-	xItLpQueue.xSlicEventStackPtr = (char *)eventStack;
-	xItLpQueue.xSlicCurEventPtr = (char *)eventStack;
-	xItLpQueue.xSlicLastValidEventPtr = (char *)eventStack +
+	hvlpevent_queue.xSlicEventStackPtr = (char *)eventStack;
+	hvlpevent_queue.xSlicCurEventPtr = (char *)eventStack;
+	hvlpevent_queue.xSlicLastValidEventPtr = (char *)eventStack +
 					(LpEventStackSize - LpEventMaxSize);
-	xItLpQueue.xIndex = 0;
+	hvlpevent_queue.xIndex = 0;
 }
 
 static int proc_lpevents_show(struct seq_file *m, void *v)
@@ -237,11 +237,11 @@ static int proc_lpevents_show(struct seq_file *m, void *v)
 
 	seq_printf(m, "LpEventQueue 0\n");
 	seq_printf(m, "  events processed:\t%lu\n",
-		   (unsigned long)xItLpQueue.xLpIntCount);
+		   (unsigned long)hvlpevent_queue.xLpIntCount);
 
 	for (i = 0; i < 9; ++i)
 		seq_printf(m, "    %s %10lu\n", event_types[i],
-			   (unsigned long)xItLpQueue.xLpIntCountByType[i]);
+			   (unsigned long)hvlpevent_queue.xLpIntCountByType[i]);
 
 	seq_printf(m, "\n  events processed by processor:\n");
 
diff --git a/arch/ppc64/kernel/LparData.c b/arch/ppc64/kernel/LparData.c
index f42ee35f927..6ffcf67dd50 100644
--- a/arch/ppc64/kernel/LparData.c
+++ b/arch/ppc64/kernel/LparData.c
@@ -193,7 +193,7 @@ struct ItVpdAreas itVpdAreas = {
 		0,0,0,			/* 13 - 15 */
 		sizeof(struct IoHriProcessorVpd),/* 16 length of Proc Vpd */
 		0,0,0,0,0,0,		/* 17 - 22  */
-		sizeof(struct ItLpQueue),/*     23 length of Lp Queue */
+		sizeof(struct hvlpevent_queue),	/* 23 length of Lp Queue */
 		0,0			/* 24 - 25 */
 		},
 	.xSlicVpdAdrs = {			/* VPD addresses */
@@ -211,7 +211,7 @@ struct ItVpdAreas itVpdAreas = {
 		0,0,0,			/* 13 - 15 */
 		&xIoHriProcessorVpd,	/*      16 Proc Vpd */
 		0,0,0,0,0,0,		/* 17 - 22 */
-		&xItLpQueue,		/*      23 Lp Queue */
+		&hvlpevent_queue,	/*      23 Lp Queue */
 		0,0
 	}
 };
diff --git a/include/asm-ppc64/iSeries/ItLpQueue.h b/include/asm-ppc64/iSeries/ItLpQueue.h
index be1cb7f3f88..bebfb364f55 100644
--- a/include/asm-ppc64/iSeries/ItLpQueue.h
+++ b/include/asm-ppc64/iSeries/ItLpQueue.h
@@ -41,7 +41,7 @@ struct HvLpEvent;
 #define LpEventMaxSize		256
 #define LpEventAlign		64
 
-struct ItLpQueue {
+struct hvlpevent_queue {
 /*
  * The xSlicCurEventPtr is the pointer to the next event stack entry
  * that will become valid.  The OS must peek at this entry to determine
@@ -74,7 +74,7 @@ struct ItLpQueue {
 	u64	xLpIntCountByType[9];	// 0x38-0x7F Event counts by type
 };
 
-extern struct ItLpQueue xItLpQueue;
+extern struct hvlpevent_queue hvlpevent_queue;
 
 extern int ItLpQueue_isLpIntPending(void);
 extern unsigned ItLpQueue_process(struct pt_regs *);
-- 
cgit v1.2.3-70-g09d2


From 937b31b114b5540f456ce1566aae67e02db41f2c Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:15:42 +1000
Subject: [PATCH] ppc64: Rename ItLpQueue_* functions to hvlpevent_queue_*

Now that we've renamed the xItLpQueue structure, rename the functions that
operate on it also.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/ItLpQueue.c         | 12 ++++++------
 arch/ppc64/kernel/idle.c              |  4 ++--
 arch/ppc64/kernel/irq.c               |  4 ++--
 arch/ppc64/kernel/mf.c                |  4 ++--
 arch/ppc64/kernel/time.c              |  4 ++--
 include/asm-ppc64/iSeries/ItLpQueue.h |  4 ++--
 6 files changed, 16 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index e55fe1a2725..a81e49b91af 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -70,7 +70,7 @@ static __inline__ void clear_inUse(void)
 extern LpEventHandler lpEventHandler[HvLpEvent_Type_NumTypes];
 unsigned long ItLpQueueInProcess = 0;
 
-static struct HvLpEvent * ItLpQueue_getNextLpEvent(void)
+static struct HvLpEvent * get_next_hvlpevent(void)
 {
 	struct HvLpEvent * nextLpEvent = 
 		(struct HvLpEvent *)hvlpevent_queue.xSlicCurEventPtr;
@@ -94,7 +94,7 @@ static struct HvLpEvent * ItLpQueue_getNextLpEvent(void)
 
 static unsigned long spread_lpevents = NR_CPUS;
 
-int ItLpQueue_isLpIntPending(void)
+int hvlpevent_is_pending(void)
 {
 	struct HvLpEvent *next_event;
 
@@ -105,7 +105,7 @@ int ItLpQueue_isLpIntPending(void)
 	return next_event->xFlags.xValid | hvlpevent_queue.xPlicOverflowIntPending;
 }
 
-static void ItLpQueue_clearValid( struct HvLpEvent * event )
+static void hvlpevent_clear_valid( struct HvLpEvent * event )
 {
 	/* Clear the valid bit of the event
 	 * Also clear bits within this event that might
@@ -127,7 +127,7 @@ static void ItLpQueue_clearValid( struct HvLpEvent * event )
 	event->xFlags.xValid = 0;
 }
 
-unsigned ItLpQueue_process(struct pt_regs *regs)
+unsigned process_hvlpevents(struct pt_regs *regs)
 {
 	unsigned numIntsProcessed = 0;
 	struct HvLpEvent * nextLpEvent;
@@ -142,7 +142,7 @@ unsigned ItLpQueue_process(struct pt_regs *regs)
 		BUG();
 
 	for (;;) {
-		nextLpEvent = ItLpQueue_getNextLpEvent();
+		nextLpEvent = get_next_hvlpevent();
 		if ( nextLpEvent ) {
 			/* Count events to return to caller
 			 * and count processed events in hvlpevent_queue
@@ -170,7 +170,7 @@ unsigned ItLpQueue_process(struct pt_regs *regs)
 			else
 				printk(KERN_INFO "Unexpected Lp Event type=%d\n", nextLpEvent->xType );
 			
-			ItLpQueue_clearValid( nextLpEvent );
+			hvlpevent_clear_valid( nextLpEvent );
 		} else if ( hvlpevent_queue.xPlicOverflowIntPending )
 			/*
 			 * No more valid events. If overflow events are
diff --git a/arch/ppc64/kernel/idle.c b/arch/ppc64/kernel/idle.c
index a7ebd0238d3..08952c7e621 100644
--- a/arch/ppc64/kernel/idle.c
+++ b/arch/ppc64/kernel/idle.c
@@ -88,7 +88,7 @@ static int iSeries_idle(void)
 
 	while (1) {
 		if (lpaca->lppaca.shared_proc) {
-			if (ItLpQueue_isLpIntPending())
+			if (hvlpevent_is_pending())
 				process_iSeries_events();
 			if (!need_resched())
 				yield_shared_processor();
@@ -100,7 +100,7 @@ static int iSeries_idle(void)
 
 				while (!need_resched()) {
 					HMT_medium();
-					if (ItLpQueue_isLpIntPending())
+					if (hvlpevent_is_pending())
 						process_iSeries_events();
 					HMT_low();
 				}
diff --git a/arch/ppc64/kernel/irq.c b/arch/ppc64/kernel/irq.c
index 46a7151ad6d..bd4b035af62 100644
--- a/arch/ppc64/kernel/irq.c
+++ b/arch/ppc64/kernel/irq.c
@@ -294,8 +294,8 @@ void do_IRQ(struct pt_regs *regs)
 		iSeries_smp_message_recv(regs);
 	}
 #endif /* CONFIG_SMP */
-	if (ItLpQueue_isLpIntPending())
-		lpevent_count += ItLpQueue_process(regs);
+	if (hvlpevent_is_pending())
+		lpevent_count += process_hvlpevents(regs);
 
 	irq_exit();
 
diff --git a/arch/ppc64/kernel/mf.c b/arch/ppc64/kernel/mf.c
index ef920699802..609bf1bda7b 100644
--- a/arch/ppc64/kernel/mf.c
+++ b/arch/ppc64/kernel/mf.c
@@ -802,8 +802,8 @@ int mf_get_boot_rtc(struct rtc_time *tm)
 	/* We need to poll here as we are not yet taking interrupts */
 	while (rtc_data.busy) {
 		extern unsigned long lpevent_count;
-		if (ItLpQueue_isLpIntPending())
-			lpevent_count += ItLpQueue_process(NULL);
+		if (hvlpevent_is_pending())
+			lpevent_count += process_hvlpevents(NULL);
 	}
 	return rtc_set_tm(rtc_data.rc, rtc_data.ce_msg.ce_msg, tm);
 }
diff --git a/arch/ppc64/kernel/time.c b/arch/ppc64/kernel/time.c
index c133f9c28c5..f9c18400ebf 100644
--- a/arch/ppc64/kernel/time.c
+++ b/arch/ppc64/kernel/time.c
@@ -367,8 +367,8 @@ int timer_interrupt(struct pt_regs * regs)
 	set_dec(next_dec);
 
 #ifdef CONFIG_PPC_ISERIES
-	if (ItLpQueue_isLpIntPending())
-		lpevent_count += ItLpQueue_process(regs);
+	if (hvlpevent_is_pending())
+		lpevent_count += process_hvlpevents(regs);
 #endif
 
 /* collect purr register values often, for accurate calculations */
diff --git a/include/asm-ppc64/iSeries/ItLpQueue.h b/include/asm-ppc64/iSeries/ItLpQueue.h
index bebfb364f55..f0f24a39ff6 100644
--- a/include/asm-ppc64/iSeries/ItLpQueue.h
+++ b/include/asm-ppc64/iSeries/ItLpQueue.h
@@ -76,8 +76,8 @@ struct hvlpevent_queue {
 
 extern struct hvlpevent_queue hvlpevent_queue;
 
-extern int ItLpQueue_isLpIntPending(void);
-extern unsigned ItLpQueue_process(struct pt_regs *);
+extern int hvlpevent_is_pending(void);
+extern unsigned process_hvlpevents(struct pt_regs *);
 extern void setup_hvlpevent_queue(void);
 
 #endif /* _ITLPQUEUE_H */
-- 
cgit v1.2.3-70-g09d2


From 74889802a1585af4e1652f0cb853ac22a65816a4 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:15:53 +1000
Subject: [PATCH] ppc64: Don't count number of events processed for caller

Currently we count the number of lpevents processed in 3 seperate places.

One of these counters is never read, so just remove it. This means
hvlpevent_queue_process() no longer needs to return the number of events
processed.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/ItLpQueue.c         | 9 ++-------
 arch/ppc64/kernel/irq.c               | 3 +--
 arch/ppc64/kernel/mf.c                | 3 +--
 arch/ppc64/kernel/time.c              | 3 +--
 include/asm-ppc64/iSeries/ItLpQueue.h | 2 +-
 5 files changed, 6 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index a81e49b91af..a849f6775ad 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -127,14 +127,14 @@ static void hvlpevent_clear_valid( struct HvLpEvent * event )
 	event->xFlags.xValid = 0;
 }
 
-unsigned process_hvlpevents(struct pt_regs *regs)
+void process_hvlpevents(struct pt_regs *regs)
 {
 	unsigned numIntsProcessed = 0;
 	struct HvLpEvent * nextLpEvent;
 
 	/* If we have recursed, just return */
 	if ( !set_inUse() )
-		return 0;
+		return;
 	
 	if (ItLpQueueInProcess == 0)
 		ItLpQueueInProcess = 1;
@@ -144,9 +144,6 @@ unsigned process_hvlpevents(struct pt_regs *regs)
 	for (;;) {
 		nextLpEvent = get_next_hvlpevent();
 		if ( nextLpEvent ) {
-			/* Count events to return to caller
-			 * and count processed events in hvlpevent_queue
- 			 */
 			++numIntsProcessed;
 			hvlpevent_queue.xLpIntCount++;
 			/* Call appropriate handler here, passing 
@@ -186,8 +183,6 @@ unsigned process_hvlpevents(struct pt_regs *regs)
 	clear_inUse();
 
 	get_paca()->lpevent_count += numIntsProcessed;
-
-	return numIntsProcessed;
 }
 
 static int set_spread_lpevents(char *str)
diff --git a/arch/ppc64/kernel/irq.c b/arch/ppc64/kernel/irq.c
index bd4b035af62..e79420bb72a 100644
--- a/arch/ppc64/kernel/irq.c
+++ b/arch/ppc64/kernel/irq.c
@@ -66,7 +66,6 @@ EXPORT_SYMBOL(irq_desc);
 int distribute_irqs = 1;
 int __irq_offset_value;
 int ppc_spurious_interrupts;
-unsigned long lpevent_count;
 u64 ppc64_interrupt_controller;
 
 int show_interrupts(struct seq_file *p, void *v)
@@ -295,7 +294,7 @@ void do_IRQ(struct pt_regs *regs)
 	}
 #endif /* CONFIG_SMP */
 	if (hvlpevent_is_pending())
-		lpevent_count += process_hvlpevents(regs);
+		process_hvlpevents(regs);
 
 	irq_exit();
 
diff --git a/arch/ppc64/kernel/mf.c b/arch/ppc64/kernel/mf.c
index 609bf1bda7b..ef4a338ebd0 100644
--- a/arch/ppc64/kernel/mf.c
+++ b/arch/ppc64/kernel/mf.c
@@ -801,9 +801,8 @@ int mf_get_boot_rtc(struct rtc_time *tm)
 		return rc;
 	/* We need to poll here as we are not yet taking interrupts */
 	while (rtc_data.busy) {
-		extern unsigned long lpevent_count;
 		if (hvlpevent_is_pending())
-			lpevent_count += process_hvlpevents(NULL);
+			process_hvlpevents(NULL);
 	}
 	return rtc_set_tm(rtc_data.rc, rtc_data.ce_msg.ce_msg, tm);
 }
diff --git a/arch/ppc64/kernel/time.c b/arch/ppc64/kernel/time.c
index f9c18400ebf..909462e1ade 100644
--- a/arch/ppc64/kernel/time.c
+++ b/arch/ppc64/kernel/time.c
@@ -99,7 +99,6 @@ unsigned long tb_to_ns_shift;
 struct gettimeofday_struct do_gtod;
 
 extern unsigned long wall_jiffies;
-extern unsigned long lpevent_count;
 extern int smp_tb_synchronized;
 
 extern struct timezone sys_tz;
@@ -368,7 +367,7 @@ int timer_interrupt(struct pt_regs * regs)
 
 #ifdef CONFIG_PPC_ISERIES
 	if (hvlpevent_is_pending())
-		lpevent_count += process_hvlpevents(regs);
+		process_hvlpevents(regs);
 #endif
 
 /* collect purr register values often, for accurate calculations */
diff --git a/include/asm-ppc64/iSeries/ItLpQueue.h b/include/asm-ppc64/iSeries/ItLpQueue.h
index f0f24a39ff6..6ba74c0d910 100644
--- a/include/asm-ppc64/iSeries/ItLpQueue.h
+++ b/include/asm-ppc64/iSeries/ItLpQueue.h
@@ -77,7 +77,7 @@ struct hvlpevent_queue {
 extern struct hvlpevent_queue hvlpevent_queue;
 
 extern int hvlpevent_is_pending(void);
-extern unsigned process_hvlpevents(struct pt_regs *);
+extern void process_hvlpevents(struct pt_regs *);
 extern void setup_hvlpevent_queue(void);
 
 #endif /* _ITLPQUEUE_H */
-- 
cgit v1.2.3-70-g09d2


From ed094150bdeb6eca691f1b7a05bd34d4d1a95488 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:16:09 +1000
Subject: [PATCH] ppc64: Simplify counting of lpevents, remove lpevent_count
 from paca

Currently there's a per-cpu count of lpevents processed, a per-queue (ie.
global) total count, and a count by event type.

Replace all that with a count by event for each cpu. We only need to add
it up int the proc code.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/ItLpQueue.c         | 44 +++++++++++++++++++++++------------
 include/asm-ppc64/iSeries/ItLpQueue.h |  2 --
 include/asm-ppc64/paca.h              |  1 -
 3 files changed, 29 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index a849f6775ad..4a6ab9de629 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -28,7 +28,9 @@
  */
 struct hvlpevent_queue hvlpevent_queue __attribute__((__section__(".data")));
 
-static char *event_types[9] = {
+DEFINE_PER_CPU(unsigned long[HvLpEvent_Type_NumTypes], hvlpevent_counts);
+
+static char *event_types[HvLpEvent_Type_NumTypes] = {
 	"Hypervisor\t\t",
 	"Machine Facilities\t",
 	"Session Manager\t",
@@ -129,7 +131,6 @@ static void hvlpevent_clear_valid( struct HvLpEvent * event )
 
 void process_hvlpevents(struct pt_regs *regs)
 {
-	unsigned numIntsProcessed = 0;
 	struct HvLpEvent * nextLpEvent;
 
 	/* If we have recursed, just return */
@@ -144,8 +145,6 @@ void process_hvlpevents(struct pt_regs *regs)
 	for (;;) {
 		nextLpEvent = get_next_hvlpevent();
 		if ( nextLpEvent ) {
-			++numIntsProcessed;
-			hvlpevent_queue.xLpIntCount++;
 			/* Call appropriate handler here, passing 
 			 * a pointer to the LpEvent.  The handler
 			 * must make a copy of the LpEvent if it
@@ -160,7 +159,7 @@ void process_hvlpevents(struct pt_regs *regs)
 			 * here!
   			 */
 			if ( nextLpEvent->xType < HvLpEvent_Type_NumTypes )
-				hvlpevent_queue.xLpIntCountByType[nextLpEvent->xType]++;
+				__get_cpu_var(hvlpevent_counts)[nextLpEvent->xType]++;
 			if ( nextLpEvent->xType < HvLpEvent_Type_NumTypes &&
 			     lpEventHandler[nextLpEvent->xType] ) 
 				lpEventHandler[nextLpEvent->xType](nextLpEvent, regs);
@@ -181,8 +180,6 @@ void process_hvlpevents(struct pt_regs *regs)
 	ItLpQueueInProcess = 0;
 	mb();
 	clear_inUse();
-
-	get_paca()->lpevent_count += numIntsProcessed;
 }
 
 static int set_spread_lpevents(char *str)
@@ -228,20 +225,37 @@ void setup_hvlpevent_queue(void)
 
 static int proc_lpevents_show(struct seq_file *m, void *v)
 {
-	unsigned int i;
+	int cpu, i;
+	unsigned long sum;
+	static unsigned long cpu_totals[NR_CPUS];
+
+	/* FIXME: do we care that there's no locking here? */
+	sum = 0;
+	for_each_online_cpu(cpu) {
+		cpu_totals[cpu] = 0;
+		for (i = 0; i < HvLpEvent_Type_NumTypes; i++) {
+			cpu_totals[cpu] += per_cpu(hvlpevent_counts, cpu)[i];
+		}
+		sum += cpu_totals[cpu];
+	}
 
 	seq_printf(m, "LpEventQueue 0\n");
-	seq_printf(m, "  events processed:\t%lu\n",
-		   (unsigned long)hvlpevent_queue.xLpIntCount);
+	seq_printf(m, "  events processed:\t%lu\n", sum);
 
-	for (i = 0; i < 9; ++i)
-		seq_printf(m, "    %s %10lu\n", event_types[i],
-			   (unsigned long)hvlpevent_queue.xLpIntCountByType[i]);
+	for (i = 0; i < HvLpEvent_Type_NumTypes; ++i) {
+		sum = 0;
+		for_each_online_cpu(cpu) {
+			sum += per_cpu(hvlpevent_counts, cpu)[i];
+		}
+
+		seq_printf(m, "    %s %10lu\n", event_types[i], sum);
+	}
 
 	seq_printf(m, "\n  events processed by processor:\n");
 
-	for_each_online_cpu(i)
-		seq_printf(m, "    CPU%02d  %10u\n", i, paca[i].lpevent_count);
+	for_each_online_cpu(cpu) {
+		seq_printf(m, "    CPU%02d  %10lu\n", cpu, cpu_totals[cpu]);
+	}
 
 	return 0;
 }
diff --git a/include/asm-ppc64/iSeries/ItLpQueue.h b/include/asm-ppc64/iSeries/ItLpQueue.h
index 6ba74c0d910..51db08852db 100644
--- a/include/asm-ppc64/iSeries/ItLpQueue.h
+++ b/include/asm-ppc64/iSeries/ItLpQueue.h
@@ -70,8 +70,6 @@ struct hvlpevent_queue {
 	u8	xIndex;			// 0x28 unique sequential index.
 	u8	xSlicRsvd[3];		// 0x29-2b
 	u32	xInUseWord;		// 0x2C
-	u64	xLpIntCount;		// 0x30 Total Lp Int msgs processed
-	u64	xLpIntCountByType[9];	// 0x38-0x7F Event counts by type
 };
 
 extern struct hvlpevent_queue hvlpevent_queue;
diff --git a/include/asm-ppc64/paca.h b/include/asm-ppc64/paca.h
index 0146f51684e..2f0f36f73d3 100644
--- a/include/asm-ppc64/paca.h
+++ b/include/asm-ppc64/paca.h
@@ -89,7 +89,6 @@ struct paca_struct {
 	u64 next_jiffy_update_tb;	/* TB value for next jiffy update */
 	u64 saved_r1;			/* r1 save for RTAS calls */
 	u64 saved_msr;			/* MSR saved here by enter_rtas */
-	u32 lpevent_count;		/* lpevents processed  */
 	u8 proc_enabled;		/* irq soft-enable flag */
 
 	/* not yet used */
-- 
cgit v1.2.3-70-g09d2


From 719d1cd86780c156f954fc34f34481adac197aec Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:17:02 +1000
Subject: [PATCH] ppc64: Replace custom locking code with a spinlock

The hvlpevent_queue (formally ItLpQueue) has a member called xInUseWord
which is used for serialising access to the queue. Because it's a word
(ie. 32 bit) there's a custom 32-bit version of test_and_set_bit() or
thereabouts in ItLpQueue.c.

The xInUseWord is not shared with they hypervisor, so we can replace it
with a spinlock and remove the custom code.

There is also another locking mechanism (ItLpQueueInProcess). This is
redundant because it's only manipulated while the lock's held. Remove it.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/ItLpQueue.c         | 38 ++---------------------------------
 include/asm-ppc64/iSeries/ItLpQueue.h |  2 +-
 2 files changed, 3 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index a4f32cbf529..4231861288a 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -42,35 +42,8 @@ static char *event_types[HvLpEvent_Type_NumTypes] = {
 	"Virtual I/O"
 };
 
-static __inline__ int set_inUse(void)
-{
-	int t;
-	u32 * inUseP = &hvlpevent_queue.xInUseWord;
-
-	__asm__ __volatile__("\n\
-1:	lwarx	%0,0,%2		\n\
-	cmpwi	0,%0,0		\n\
-	li	%0,0		\n\
-	bne-	2f		\n\
-	addi	%0,%0,1		\n\
-	stwcx.	%0,0,%2		\n\
-	bne-	1b		\n\
-2:	eieio"
-	: "=&r" (t), "=m" (hvlpevent_queue.xInUseWord)
-	: "r" (inUseP), "m" (hvlpevent_queue.xInUseWord)
-	: "cc");
-
-	return t;
-}
-
-static __inline__ void clear_inUse(void)
-{
-	hvlpevent_queue.xInUseWord = 0;
-}
-
 /* Array of LpEvent handler functions */
 extern LpEventHandler lpEventHandler[HvLpEvent_Type_NumTypes];
-unsigned long ItLpQueueInProcess = 0;
 
 static struct HvLpEvent * get_next_hvlpevent(void)
 {
@@ -144,14 +117,9 @@ void process_hvlpevents(struct pt_regs *regs)
 	struct HvLpEvent * event;
 
 	/* If we have recursed, just return */
-	if ( !set_inUse() )
+	if (!spin_trylock(&hvlpevent_queue.lock))
 		return;
 
-	if (ItLpQueueInProcess == 0)
-		ItLpQueueInProcess = 1;
-	else
-		BUG();
-
 	for (;;) {
 		event = get_next_hvlpevent();
 		if (event) {
@@ -187,9 +155,7 @@ void process_hvlpevents(struct pt_regs *regs)
 			break;
 	}
 
-	ItLpQueueInProcess = 0;
-	mb();
-	clear_inUse();
+	spin_unlock(&hvlpevent_queue.lock);
 }
 
 static int set_spread_lpevents(char *str)
diff --git a/include/asm-ppc64/iSeries/ItLpQueue.h b/include/asm-ppc64/iSeries/ItLpQueue.h
index 51db08852db..69b26ad7413 100644
--- a/include/asm-ppc64/iSeries/ItLpQueue.h
+++ b/include/asm-ppc64/iSeries/ItLpQueue.h
@@ -69,7 +69,7 @@ struct hvlpevent_queue {
 	char	*xSlicEventStackPtr;	// 0x20
 	u8	xIndex;			// 0x28 unique sequential index.
 	u8	xSlicRsvd[3];		// 0x29-2b
-	u32	xInUseWord;		// 0x2C
+	spinlock_t	lock;
 };
 
 extern struct hvlpevent_queue hvlpevent_queue;
-- 
cgit v1.2.3-70-g09d2


From 0edb586049e57c56e625536476931117a57671e9 Mon Sep 17 00:00:00 2001
From: Cornelia Huck <cohuck@de.ibm.com>
Date: Wed, 22 Jun 2005 16:59:51 +0200
Subject: [PATCH] driver core: add bus_find_device & driver_find_device
 functions

Add bus_find_device() and driver_find_device() which allow searching for a
device in the bus's resp. the driver's klist and obtain a reference on it.

Signed-off-by: Cornelia Huck <cohuck@de.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/bus.c     | 34 ++++++++++++++++++++++++++++++++++
 drivers/base/driver.c  | 35 +++++++++++++++++++++++++++++++++++
 include/linux/device.h |  5 +++++
 3 files changed, 74 insertions(+)

(limited to 'include')

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index c3fac7fd555..2c64b792d07 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -177,6 +177,39 @@ int bus_for_each_dev(struct bus_type * bus, struct device * start,
 	return error;
 }
 
+/**
+ * bus_find_device - device iterator for locating a particular device.
+ * @bus: bus type
+ * @start: Device to begin with
+ * @data: Data to pass to match function
+ * @match: Callback function to check device
+ *
+ * This is similar to the bus_for_each_dev() function above, but it
+ * returns a reference to a device that is 'found' for later use, as
+ * determined by the @match callback.
+ *
+ * The callback should return 0 if the device doesn't match and non-zero
+ * if it does.  If the callback returns non-zero, this function will
+ * return to the caller and not iterate over any more devices.
+ */
+struct device * bus_find_device(struct bus_type *bus,
+				struct device *start, void *data,
+				int (*match)(struct device *, void *))
+{
+	struct klist_iter i;
+	struct device *dev;
+
+	if (!bus)
+		return NULL;
+
+	klist_iter_init_node(&bus->klist_devices, &i,
+			     (start ? &start->knode_bus : NULL));
+	while ((dev = next_device(&i)))
+		if (match(dev, data) && get_device(dev))
+			break;
+	klist_iter_exit(&i);
+	return dev;
+}
 
 
 static struct device_driver * next_driver(struct klist_iter * i)
@@ -557,6 +590,7 @@ int __init buses_init(void)
 
 
 EXPORT_SYMBOL_GPL(bus_for_each_dev);
+EXPORT_SYMBOL_GPL(bus_find_device);
 EXPORT_SYMBOL_GPL(bus_for_each_drv);
 
 EXPORT_SYMBOL_GPL(bus_add_device);
diff --git a/drivers/base/driver.c b/drivers/base/driver.c
index 1b645886e9e..291c5954a3a 100644
--- a/drivers/base/driver.c
+++ b/drivers/base/driver.c
@@ -55,6 +55,41 @@ int driver_for_each_device(struct device_driver * drv, struct device * start,
 EXPORT_SYMBOL_GPL(driver_for_each_device);
 
 
+/**
+ * driver_find_device - device iterator for locating a particular device.
+ * @driver: The device's driver
+ * @start: Device to begin with
+ * @data: Data to pass to match function
+ * @match: Callback function to check device
+ *
+ * This is similar to the driver_for_each_device() function above, but
+ * it returns a reference to a device that is 'found' for later use, as
+ * determined by the @match callback.
+ *
+ * The callback should return 0 if the device doesn't match and non-zero
+ * if it does.  If the callback returns non-zero, this function will
+ * return to the caller and not iterate over any more devices.
+ */
+struct device * driver_find_device(struct device_driver *drv,
+				   struct device * start, void * data,
+				   int (*match)(struct device *, void *))
+{
+	struct klist_iter i;
+	struct device *dev;
+
+	if (!drv)
+		return NULL;
+
+	klist_iter_init_node(&drv->klist_devices, &i,
+			     (start ? &start->knode_driver : NULL));
+	while ((dev = next_device(&i)))
+		if (match(dev, data) && get_device(dev))
+			break;
+	klist_iter_exit(&i);
+	return dev;
+}
+EXPORT_SYMBOL_GPL(driver_find_device);
+
 /**
  *	driver_create_file - create sysfs file for driver.
  *	@drv:	driver.
diff --git a/include/linux/device.h b/include/linux/device.h
index 7b781a72b29..07222c531d3 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -80,6 +80,8 @@ extern struct bus_type * find_bus(char * name);
 
 int bus_for_each_dev(struct bus_type * bus, struct device * start, void * data,
 		     int (*fn)(struct device *, void *));
+struct device * bus_find_device(struct bus_type *bus, struct device *start,
+				void *data, int (*match)(struct device *, void *));
 
 int bus_for_each_drv(struct bus_type * bus, struct device_driver * start, 
 		     void * data, int (*fn)(struct device_driver *, void *));
@@ -142,6 +144,9 @@ extern void driver_remove_file(struct device_driver *, struct driver_attribute *
 
 extern int driver_for_each_device(struct device_driver * drv, struct device * start,
 				  void * data, int (*fn)(struct device *, void *));
+struct device * driver_find_device(struct device_driver *drv,
+				   struct device *start, void *data,
+				   int (*match)(struct device *, void *));
 
 
 /*
-- 
cgit v1.2.3-70-g09d2


From 23d3d602cb96addd3c1158424fb01a49ea5e81b1 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Wed, 22 Jun 2005 16:09:05 -0700
Subject: [PATCH] driver core: change bus_rescan_devices to return void

No one was looking at the return value of bus_rescan_devices, and it
really wasn't anything that anyone in the kernel would ever care about.
So change it which enabled some counting code to be removed also.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/bus.c     | 27 +++++++++------------------
 include/linux/device.h |  2 +-
 2 files changed, 10 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 7e17488271a..96fe2f95675 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -483,31 +483,22 @@ void bus_remove_driver(struct device_driver * drv)
 /* Helper for bus_rescan_devices's iter */
 static int bus_rescan_devices_helper(struct device *dev, void *data)
 {
-	int *count = data;
-
-	if (!dev->driver && (device_attach(dev) > 0))
-		(*count)++;
-
+	if (!dev->driver)
+		device_attach(dev);
 	return 0;
 }
 
-
 /**
- *	bus_rescan_devices - rescan devices on the bus for possible drivers
- *	@bus:	the bus to scan.
+ * bus_rescan_devices - rescan devices on the bus for possible drivers
+ * @bus: the bus to scan.
  *
- *	This function will look for devices on the bus with no driver
- *	attached and rescan it against existing drivers to see if it
- *	matches any. Calls device_attach(). Returns the number of devices
- *	that were sucessfully bound to a driver.
+ * This function will look for devices on the bus with no driver
+ * attached and rescan it against existing drivers to see if it matches
+ * any by calling device_attach() for the unbound devices.
  */
-int bus_rescan_devices(struct bus_type * bus)
+void bus_rescan_devices(struct bus_type * bus)
 {
-	int count = 0;
-
-	bus_for_each_dev(bus, NULL, &count, bus_rescan_devices_helper);
-
-	return count;
+	bus_for_each_dev(bus, NULL, NULL, bus_rescan_devices_helper);
 }
 
 
diff --git a/include/linux/device.h b/include/linux/device.h
index 07222c531d3..f378c846e6d 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -69,7 +69,7 @@ struct bus_type {
 extern int bus_register(struct bus_type * bus);
 extern void bus_unregister(struct bus_type * bus);
 
-extern int bus_rescan_devices(struct bus_type * bus);
+extern void bus_rescan_devices(struct bus_type * bus);
 
 extern struct bus_type * get_bus(struct bus_type * bus);
 extern void put_bus(struct bus_type * bus);
-- 
cgit v1.2.3-70-g09d2


From cfb0810eab39d1162f45b73fc96f45ab1cbcbe8b Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Thu, 30 Jun 2005 11:06:49 +0100
Subject: [PATCH] ARM: Don't try to send a signal to pid0

If we receive an unrecognised abort during boot, don't try to
send a signal to pid0, but instead report the current state.
This leads to less confusing debug reports.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/traps.c  | 12 ++------
 arch/arm/mm/fault.c      | 75 ++++++++++++++++++++++++++----------------------
 include/asm-arm/system.h |  4 ++-
 3 files changed, 45 insertions(+), 46 deletions(-)

(limited to 'include')

diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c
index 2fb0a4cfb37..df2cb06ce42 100644
--- a/arch/arm/kernel/traps.c
+++ b/arch/arm/kernel/traps.c
@@ -230,16 +230,8 @@ NORET_TYPE void die(const char *str, struct pt_regs *regs, int err)
 	do_exit(SIGSEGV);
 }
 
-void die_if_kernel(const char *str, struct pt_regs *regs, int err)
-{
-	if (user_mode(regs))
-    		return;
-
-    	die(str, regs, err);
-}
-
-static void notify_die(const char *str, struct pt_regs *regs, siginfo_t *info,
-		       unsigned long err, unsigned long trap)
+void notify_die(const char *str, struct pt_regs *regs, struct siginfo *info,
+		unsigned long err, unsigned long trap)
 {
 	if (user_mode(regs)) {
 		current->thread.error_code = err;
diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c
index e25b4fd8412..65bfe84b6d6 100644
--- a/arch/arm/mm/fault.c
+++ b/arch/arm/mm/fault.c
@@ -372,49 +372,50 @@ do_bad(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 static struct fsr_info {
 	int	(*fn)(unsigned long addr, unsigned int fsr, struct pt_regs *regs);
 	int	sig;
+	int	code;
 	const char *name;
 } fsr_info[] = {
 	/*
 	 * The following are the standard ARMv3 and ARMv4 aborts.  ARMv5
 	 * defines these to be "precise" aborts.
 	 */
-	{ do_bad,		SIGSEGV, "vector exception"		   },
-	{ do_bad,		SIGILL,	 "alignment exception"		   },
-	{ do_bad,		SIGKILL, "terminal exception"		   },
-	{ do_bad,		SIGILL,	 "alignment exception"		   },
-	{ do_bad,		SIGBUS,	 "external abort on linefetch"	   },
-	{ do_translation_fault,	SIGSEGV, "section translation fault"	   },
-	{ do_bad,		SIGBUS,	 "external abort on linefetch"	   },
-	{ do_page_fault,	SIGSEGV, "page translation fault"	   },
-	{ do_bad,		SIGBUS,	 "external abort on non-linefetch" },
-	{ do_bad,		SIGSEGV, "section domain fault"		   },
-	{ do_bad,		SIGBUS,	 "external abort on non-linefetch" },
-	{ do_bad,		SIGSEGV, "page domain fault"		   },
-	{ do_bad,		SIGBUS,	 "external abort on translation"   },
-	{ do_sect_fault,	SIGSEGV, "section permission fault"	   },
-	{ do_bad,		SIGBUS,	 "external abort on translation"   },
-	{ do_page_fault,	SIGSEGV, "page permission fault"	   },
+	{ do_bad,		SIGSEGV, 0,		"vector exception"		   },
+	{ do_bad,		SIGILL,	 BUS_ADRALN,	"alignment exception"		   },
+	{ do_bad,		SIGKILL, 0,		"terminal exception"		   },
+	{ do_bad,		SIGILL,	 BUS_ADRALN,	"alignment exception"		   },
+	{ do_bad,		SIGBUS,	 0,		"external abort on linefetch"	   },
+	{ do_translation_fault,	SIGSEGV, SEGV_MAPERR,	"section translation fault"	   },
+	{ do_bad,		SIGBUS,	 0,		"external abort on linefetch"	   },
+	{ do_page_fault,	SIGSEGV, SEGV_MAPERR,	"page translation fault"	   },
+	{ do_bad,		SIGBUS,	 0,		"external abort on non-linefetch"  },
+	{ do_bad,		SIGSEGV, SEGV_ACCERR,	"section domain fault"		   },
+	{ do_bad,		SIGBUS,	 0,		"external abort on non-linefetch"  },
+	{ do_bad,		SIGSEGV, SEGV_ACCERR,	"page domain fault"		   },
+	{ do_bad,		SIGBUS,	 0,		"external abort on translation"	   },
+	{ do_sect_fault,	SIGSEGV, SEGV_ACCERR,	"section permission fault"	   },
+	{ do_bad,		SIGBUS,	 0,		"external abort on translation"	   },
+	{ do_page_fault,	SIGSEGV, SEGV_ACCERR,	"page permission fault"		   },
 	/*
 	 * The following are "imprecise" aborts, which are signalled by bit
 	 * 10 of the FSR, and may not be recoverable.  These are only
 	 * supported if the CPU abort handler supports bit 10.
 	 */
-	{ do_bad,		SIGBUS,  "unknown 16"			   },
-	{ do_bad,		SIGBUS,  "unknown 17"			   },
-	{ do_bad,		SIGBUS,  "unknown 18"			   },
-	{ do_bad,		SIGBUS,  "unknown 19"			   },
-	{ do_bad,		SIGBUS,  "lock abort"			   }, /* xscale */
-	{ do_bad,		SIGBUS,  "unknown 21"			   },
-	{ do_bad,		SIGBUS,  "imprecise external abort"	   }, /* xscale */
-	{ do_bad,		SIGBUS,  "unknown 23"			   },
-	{ do_bad,		SIGBUS,  "dcache parity error"		   }, /* xscale */
-	{ do_bad,		SIGBUS,  "unknown 25"			   },
-	{ do_bad,		SIGBUS,  "unknown 26"			   },
-	{ do_bad,		SIGBUS,  "unknown 27"			   },
-	{ do_bad,		SIGBUS,  "unknown 28"			   },
-	{ do_bad,		SIGBUS,  "unknown 29"			   },
-	{ do_bad,		SIGBUS,  "unknown 30"			   },
-	{ do_bad,		SIGBUS,  "unknown 31"			   }
+	{ do_bad,		SIGBUS,  0,		"unknown 16"			   },
+	{ do_bad,		SIGBUS,  0,		"unknown 17"			   },
+	{ do_bad,		SIGBUS,  0,		"unknown 18"			   },
+	{ do_bad,		SIGBUS,  0,		"unknown 19"			   },
+	{ do_bad,		SIGBUS,  0,		"lock abort"			   }, /* xscale */
+	{ do_bad,		SIGBUS,  0,		"unknown 21"			   },
+	{ do_bad,		SIGBUS,  BUS_OBJERR,	"imprecise external abort"	   }, /* xscale */
+	{ do_bad,		SIGBUS,  0,		"unknown 23"			   },
+	{ do_bad,		SIGBUS,  0,		"dcache parity error"		   }, /* xscale */
+	{ do_bad,		SIGBUS,  0,		"unknown 25"			   },
+	{ do_bad,		SIGBUS,  0,		"unknown 26"			   },
+	{ do_bad,		SIGBUS,  0,		"unknown 27"			   },
+	{ do_bad,		SIGBUS,  0,		"unknown 28"			   },
+	{ do_bad,		SIGBUS,  0,		"unknown 29"			   },
+	{ do_bad,		SIGBUS,  0,		"unknown 30"			   },
+	{ do_bad,		SIGBUS,  0,		"unknown 31"			   }
 };
 
 void __init
@@ -435,15 +436,19 @@ asmlinkage void
 do_DataAbort(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 {
 	const struct fsr_info *inf = fsr_info + (fsr & 15) + ((fsr & (1 << 10)) >> 6);
+	struct siginfo info;
 
 	if (!inf->fn(addr, fsr, regs))
 		return;
 
 	printk(KERN_ALERT "Unhandled fault: %s (0x%03x) at 0x%08lx\n",
 		inf->name, fsr, addr);
-	force_sig(inf->sig, current);
-	show_pte(current->mm, addr);
-	die_if_kernel("Oops", regs, 0);
+
+	info.si_signo = inf->sig;
+	info.si_errno = 0;
+	info.si_code  = inf->code;
+	info.si_addr  = (void __user *)addr;
+	notify_die("", regs, &info, fsr, 0);
 }
 
 asmlinkage void
diff --git a/include/asm-arm/system.h b/include/asm-arm/system.h
index cdf49f442fd..2f44b204421 100644
--- a/include/asm-arm/system.h
+++ b/include/asm-arm/system.h
@@ -85,7 +85,9 @@ struct pt_regs;
 void die(const char *msg, struct pt_regs *regs, int err)
 		__attribute__((noreturn));
 
-void die_if_kernel(const char *str, struct pt_regs *regs, int err);
+struct siginfo;
+void notify_die(const char *str, struct pt_regs *regs, struct siginfo *info,
+		unsigned long err, unsigned long trap);
 
 void hook_fault_code(int nr, int (*fn)(unsigned long, unsigned int,
 				       struct pt_regs *),
-- 
cgit v1.2.3-70-g09d2


From 5ee24d9594ffb070261b70461f71c42913c663bb Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Thu, 30 Jun 2005 02:58:48 -0700
Subject: [PATCH] s390: fix finish_arch_switch

Commit 4866cde064afbb6c2a488c265e696879de616daa requires finish_arch_switch
to have only one parameter instead of two.

Also fix another compile error (double declaration of account_system_vtime)
if CONFIG_VIRT_CPU_ACCOUNTING is not defined.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-s390/system.h | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-s390/system.h b/include/asm-s390/system.h
index b4a9f05a93d..864cae7e1fd 100644
--- a/include/asm-s390/system.h
+++ b/include/asm-s390/system.h
@@ -107,11 +107,9 @@ static inline void restore_access_regs(unsigned int *acrs)
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
 extern void account_user_vtime(struct task_struct *);
 extern void account_system_vtime(struct task_struct *);
-#else
-#define account_system_vtime(prev) do { } while (0)
 #endif
 
-#define finish_arch_switch(rq, prev) do {				     \
+#define finish_arch_switch(prev) do {					     \
 	set_fs(current->thread.mm_segment);				     \
 	account_system_vtime(prev);					     \
 } while (0)
-- 
cgit v1.2.3-70-g09d2


From 306e440daf5f40b195afd83d05dee89fa63189e7 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Thu, 30 Jun 2005 02:58:55 -0700
Subject: [PATCH] x86: i8253/i8259A lock cleanup

Introduce proper declarations for i8253_lock and i8259A_lock.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/apic.c                  | 2 +-
 arch/i386/kernel/apm.c                   | 5 ++---
 arch/i386/kernel/io_apic.c               | 2 +-
 arch/i386/kernel/time.c                  | 5 ++++-
 arch/i386/kernel/timers/timer_cyclone.c  | 4 ++--
 arch/i386/kernel/timers/timer_pit.c      | 4 +---
 arch/i386/kernel/timers/timer_tsc.c      | 3 +--
 arch/i386/mach-voyager/voyager_basic.c   | 2 +-
 arch/x86_64/kernel/io_apic.c             | 1 -
 drivers/ide/legacy/hd.c                  | 4 +++-
 drivers/input/gameport/gameport.c        | 3 ++-
 drivers/input/joystick/analog.c          | 4 +++-
 include/asm-i386/i8253.h                 | 6 ++++++
 include/asm-i386/mach-default/do_timer.h | 1 +
 include/asm-x86_64/io_apic.h             | 2 ++
 15 files changed, 30 insertions(+), 18 deletions(-)
 create mode 100644 include/asm-i386/i8253.h

(limited to 'include')

diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index 93df90bbb87..bd1dbf3bd22 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -35,6 +35,7 @@
 #include <asm/desc.h>
 #include <asm/arch_hooks.h>
 #include <asm/hpet.h>
+#include <asm/i8253.h>
 
 #include <mach_apic.h>
 
@@ -879,7 +880,6 @@ fake_ioapic_page:
  */
 static unsigned int __devinit get_8254_timer_count(void)
 {
-	extern spinlock_t i8253_lock;
 	unsigned long flags;
 
 	unsigned int count;
diff --git a/arch/i386/kernel/apm.c b/arch/i386/kernel/apm.c
index d48ce929096..064211d5f41 100644
--- a/arch/i386/kernel/apm.c
+++ b/arch/i386/kernel/apm.c
@@ -228,10 +228,10 @@
 #include <asm/system.h>
 #include <asm/uaccess.h>
 #include <asm/desc.h>
+#include <asm/i8253.h>
 
 #include "io_ports.h"
 
-extern spinlock_t i8253_lock;
 extern unsigned long get_cmos_time(void);
 extern void machine_real_restart(unsigned char *, int);
 
@@ -1168,8 +1168,7 @@ static void get_time_diff(void)
 static void reinit_timer(void)
 {
 #ifdef INIT_TIMER_AFTER_SUSPEND
-	unsigned long	flags;
-	extern spinlock_t i8253_lock;
+	unsigned long flags;
 
 	spin_lock_irqsave(&i8253_lock, flags);
 	/* set the clock to 100 Hz */
diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index 35eb8e29c48..6578f40bd50 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -37,6 +37,7 @@
 #include <asm/smp.h>
 #include <asm/desc.h>
 #include <asm/timer.h>
+#include <asm/i8259.h>
 
 #include <mach_apic.h>
 
@@ -1566,7 +1567,6 @@ void print_all_local_APICs (void)
 
 void /*__init*/ print_PIC(void)
 {
-	extern spinlock_t i8259A_lock;
 	unsigned int v;
 	unsigned long flags;
 
diff --git a/arch/i386/kernel/time.c b/arch/i386/kernel/time.c
index e68d9fdb075..2854c357377 100644
--- a/arch/i386/kernel/time.c
+++ b/arch/i386/kernel/time.c
@@ -68,7 +68,8 @@
 
 #include "io_ports.h"
 
-extern spinlock_t i8259A_lock;
+#include <asm/i8259.h>
+
 int pit_latch_buggy;              /* extern */
 
 #include "do_timer.h"
@@ -85,6 +86,8 @@ extern unsigned long wall_jiffies;
 DEFINE_SPINLOCK(rtc_lock);
 EXPORT_SYMBOL(rtc_lock);
 
+#include <asm/i8253.h>
+
 DEFINE_SPINLOCK(i8253_lock);
 EXPORT_SYMBOL(i8253_lock);
 
diff --git a/arch/i386/kernel/timers/timer_cyclone.c b/arch/i386/kernel/timers/timer_cyclone.c
index f6f1206a11b..13892a65c94 100644
--- a/arch/i386/kernel/timers/timer_cyclone.c
+++ b/arch/i386/kernel/timers/timer_cyclone.c
@@ -17,9 +17,9 @@
 #include <asm/io.h>
 #include <asm/pgtable.h>
 #include <asm/fixmap.h>
-#include "io_ports.h"
+#include <asm/i8253.h>
 
-extern spinlock_t i8253_lock;
+#include "io_ports.h"
 
 /* Number of usecs that the last interrupt was delayed */
 static int delay_at_last_interrupt;
diff --git a/arch/i386/kernel/timers/timer_pit.c b/arch/i386/kernel/timers/timer_pit.c
index 967d5453cd0..06de036a820 100644
--- a/arch/i386/kernel/timers/timer_pit.c
+++ b/arch/i386/kernel/timers/timer_pit.c
@@ -15,9 +15,8 @@
 #include <asm/smp.h>
 #include <asm/io.h>
 #include <asm/arch_hooks.h>
+#include <asm/i8253.h>
 
-extern spinlock_t i8259A_lock;
-extern spinlock_t i8253_lock;
 #include "do_timer.h"
 #include "io_ports.h"
 
@@ -166,7 +165,6 @@ struct init_timer_opts __initdata timer_pit_init = {
 
 void setup_pit_timer(void)
 {
-	extern spinlock_t i8253_lock;
 	unsigned long flags;
 
 	spin_lock_irqsave(&i8253_lock, flags);
diff --git a/arch/i386/kernel/timers/timer_tsc.c b/arch/i386/kernel/timers/timer_tsc.c
index f46e625bab6..8f4e4d5bc56 100644
--- a/arch/i386/kernel/timers/timer_tsc.c
+++ b/arch/i386/kernel/timers/timer_tsc.c
@@ -24,6 +24,7 @@
 #include "mach_timer.h"
 
 #include <asm/hpet.h>
+#include <asm/i8253.h>
 
 #ifdef CONFIG_HPET_TIMER
 static unsigned long hpet_usec_quotient;
@@ -35,8 +36,6 @@ static inline void cpufreq_delayed_get(void);
 
 int tsc_disable __devinitdata = 0;
 
-extern spinlock_t i8253_lock;
-
 static int use_tsc;
 /* Number of usecs that the last interrupt was delayed */
 static int delay_at_last_interrupt;
diff --git a/arch/i386/mach-voyager/voyager_basic.c b/arch/i386/mach-voyager/voyager_basic.c
index 602aea240e9..3e439ce5e1b 100644
--- a/arch/i386/mach-voyager/voyager_basic.c
+++ b/arch/i386/mach-voyager/voyager_basic.c
@@ -30,6 +30,7 @@
 #include <linux/irq.h>
 #include <asm/tlbflush.h>
 #include <asm/arch_hooks.h>
+#include <asm/i8253.h>
 
 /*
  * Power off function, if any
@@ -182,7 +183,6 @@ voyager_timer_interrupt(struct pt_regs *regs)
 		 * and swiftly introduce it to something sharp and
 		 * pointy.  */
 		__u16 val;
-		extern spinlock_t i8253_lock;
 
 		spin_lock(&i8253_lock);
 		
diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index 157190d986b..d206d7e49cf 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -1064,7 +1064,6 @@ void print_all_local_APICs (void)
 
 void __apicdebuginit print_PIC(void)
 {
-	extern spinlock_t i8259A_lock;
 	unsigned int v;
 	unsigned long flags;
 
diff --git a/drivers/ide/legacy/hd.c b/drivers/ide/legacy/hd.c
index e884cd4b22f..242029c9c0c 100644
--- a/drivers/ide/legacy/hd.c
+++ b/drivers/ide/legacy/hd.c
@@ -156,11 +156,13 @@ else \
 
 
 #if (HD_DELAY > 0)
+
+#include <asm/i8253.h>
+
 unsigned long last_req;
 
 unsigned long read_timer(void)
 {
-        extern spinlock_t i8253_lock;
 	unsigned long t, flags;
 	int i;
 
diff --git a/drivers/input/gameport/gameport.c b/drivers/input/gameport/gameport.c
index 3e72c9b1461..ab09cf4093e 100644
--- a/drivers/input/gameport/gameport.c
+++ b/drivers/input/gameport/gameport.c
@@ -60,12 +60,13 @@ static void gameport_disconnect_port(struct gameport *gameport);
 
 #if defined(__i386__)
 
+#include <asm/i8253.h>
+
 #define DELTA(x,y)      ((y)-(x)+((y)<(x)?1193182/HZ:0))
 #define GET_TIME(x)     do { x = get_time_pit(); } while (0)
 
 static unsigned int get_time_pit(void)
 {
-	extern spinlock_t i8253_lock;
 	unsigned long flags;
 	unsigned int count;
 
diff --git a/drivers/input/joystick/analog.c b/drivers/input/joystick/analog.c
index 504b7d55056..c3a5739030c 100644
--- a/drivers/input/joystick/analog.c
+++ b/drivers/input/joystick/analog.c
@@ -140,12 +140,14 @@ struct analog_port {
  */
 
 #ifdef __i386__
+
+#include <asm/i8253.h>
+
 #define GET_TIME(x)	do { if (cpu_has_tsc) rdtscl(x); else x = get_time_pit(); } while (0)
 #define DELTA(x,y)	(cpu_has_tsc ? ((y) - (x)) : ((x) - (y) + ((x) < (y) ? CLOCK_TICK_RATE / HZ : 0)))
 #define TIME_NAME	(cpu_has_tsc?"TSC":"PIT")
 static unsigned int get_time_pit(void)
 {
-        extern spinlock_t i8253_lock;
         unsigned long flags;
         unsigned int count;
 
diff --git a/include/asm-i386/i8253.h b/include/asm-i386/i8253.h
new file mode 100644
index 00000000000..015d8df0769
--- /dev/null
+++ b/include/asm-i386/i8253.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_I8253_H__
+#define __ASM_I8253_H__
+
+extern spinlock_t i8253_lock;
+
+#endif	/* __ASM_I8253_H__ */
diff --git a/include/asm-i386/mach-default/do_timer.h b/include/asm-i386/mach-default/do_timer.h
index 03dd13a48a8..56211414fc9 100644
--- a/include/asm-i386/mach-default/do_timer.h
+++ b/include/asm-i386/mach-default/do_timer.h
@@ -1,6 +1,7 @@
 /* defines for inline arch setup functions */
 
 #include <asm/apic.h>
+#include <asm/i8259.h>
 
 /**
  * do_timer_interrupt_hook - hook into timer tick
diff --git a/include/asm-x86_64/io_apic.h b/include/asm-x86_64/io_apic.h
index 32573749004..a8babd2bbe8 100644
--- a/include/asm-x86_64/io_apic.h
+++ b/include/asm-x86_64/io_apic.h
@@ -217,4 +217,6 @@ extern int assign_irq_vector(int irq);
 
 void enable_NMI_through_LVT0 (void * dummy);
 
+extern spinlock_t i8259A_lock;
+
 #endif
-- 
cgit v1.2.3-70-g09d2


From 5b0de927d9c9a72e42a4b581a897710f9ae5a6d1 Mon Sep 17 00:00:00 2001
From: Chris Zankel <chris@zankel.net>
Date: Thu, 30 Jun 2005 02:58:57 -0700
Subject: [PATCH] xtensa: cleanups for errno and ipc.

I noticed this because I was doing some more ipc cleanups and I did the
original errno and ipc cleanups for other architectures, so it stuck out.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Chris Zankel <chris@zankel.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-xtensa/errno.h | 128 +--------------------------------------------
 include/asm-xtensa/ipc.h   |  20 +------
 2 files changed, 2 insertions(+), 146 deletions(-)

(limited to 'include')

diff --git a/include/asm-xtensa/errno.h b/include/asm-xtensa/errno.h
index ced5194d275..a0f3b96b79b 100644
--- a/include/asm-xtensa/errno.h
+++ b/include/asm-xtensa/errno.h
@@ -11,132 +11,6 @@
 #ifndef _XTENSA_ERRNO_H
 #define _XTENSA_ERRNO_H
 
-#define	EPERM		 1	/* Operation not permitted */
-#define	ENOENT		 2	/* No such file or directory */
-#define	ESRCH		 3	/* No such process */
-#define	EINTR		 4	/* Interrupted system call */
-#define	EIO		 5	/* I/O error */
-#define	ENXIO		 6	/* No such device or address */
-#define	E2BIG		 7	/* Arg list too long */
-#define	ENOEXEC		 8	/* Exec format error */
-#define	EBADF		 9	/* Bad file number */
-#define	ECHILD		10	/* No child processes */
-#define	EAGAIN		11	/* Try again */
-#define	ENOMEM		12	/* Out of memory */
-#define	EACCES		13	/* Permission denied */
-#define	EFAULT		14	/* Bad address */
-#define	ENOTBLK		15	/* Block device required */
-#define	EBUSY		16	/* Device or resource busy */
-#define	EEXIST		17	/* File exists */
-#define	EXDEV		18	/* Cross-device link */
-#define	ENODEV		19	/* No such device */
-#define	ENOTDIR		20	/* Not a directory */
-#define	EISDIR		21	/* Is a directory */
-#define	EINVAL		22	/* Invalid argument */
-#define	ENFILE		23	/* File table overflow */
-#define	EMFILE		24	/* Too many open files */
-#define	ENOTTY		25	/* Not a typewriter */
-#define	ETXTBSY		26	/* Text file busy */
-#define	EFBIG		27	/* File too large */
-#define	ENOSPC		28	/* No space left on device */
-#define	ESPIPE		29	/* Illegal seek */
-#define	EROFS		30	/* Read-only file system */
-#define	EMLINK		31	/* Too many links */
-#define	EPIPE		32	/* Broken pipe */
-#define	EDOM		33	/* Math argument out of domain of func */
-#define	ERANGE		34	/* Math result not representable */
-#define	EDEADLK		35	/* Resource deadlock would occur */
-#define	ENAMETOOLONG	36	/* File name too long */
-#define	ENOLCK		37	/* No record locks available */
-#define	ENOSYS		38	/* Function not implemented */
-#define	ENOTEMPTY	39	/* Directory not empty */
-#define	ELOOP		40	/* Too many symbolic links encountered */
-#define	EWOULDBLOCK	EAGAIN	/* Operation would block */
-#define	ENOMSG		42	/* No message of desired type */
-#define	EIDRM		43	/* Identifier removed */
-#define	ECHRNG		44	/* Channel number out of range */
-#define	EL2NSYNC	45	/* Level 2 not synchronized */
-#define	EL3HLT		46	/* Level 3 halted */
-#define	EL3RST		47	/* Level 3 reset */
-#define	ELNRNG		48	/* Link number out of range */
-#define	EUNATCH		49	/* Protocol driver not attached */
-#define	ENOCSI		50	/* No CSI structure available */
-#define	EL2HLT		51	/* Level 2 halted */
-#define	EBADE		52	/* Invalid exchange */
-#define	EBADR		53	/* Invalid request descriptor */
-#define	EXFULL		54	/* Exchange full */
-#define	ENOANO		55	/* No anode */
-#define	EBADRQC		56	/* Invalid request code */
-#define	EBADSLT		57	/* Invalid slot */
-
-#define	EDEADLOCK	EDEADLK
-
-#define	EBFONT		59	/* Bad font file format */
-#define	ENOSTR		60	/* Device not a stream */
-#define	ENODATA		61	/* No data available */
-#define	ETIME		62	/* Timer expired */
-#define	ENOSR		63	/* Out of streams resources */
-#define	ENONET		64	/* Machine is not on the network */
-#define	ENOPKG		65	/* Package not installed */
-#define	EREMOTE		66	/* Object is remote */
-#define	ENOLINK		67	/* Link has been severed */
-#define	EADV		68	/* Advertise error */
-#define	ESRMNT		69	/* Srmount error */
-#define	ECOMM		70	/* Communication error on send */
-#define	EPROTO		71	/* Protocol error */
-#define	EMULTIHOP	72	/* Multihop attempted */
-#define	EDOTDOT		73	/* RFS specific error */
-#define	EBADMSG		74	/* Not a data message */
-#define	EOVERFLOW	75	/* Value too large for defined data type */
-#define	ENOTUNIQ	76	/* Name not unique on network */
-#define	EBADFD		77	/* File descriptor in bad state */
-#define	EREMCHG		78	/* Remote address changed */
-#define	ELIBACC		79	/* Can not access a needed shared library */
-#define	ELIBBAD		80	/* Accessing a corrupted shared library */
-#define	ELIBSCN		81	/* .lib section in a.out corrupted */
-#define	ELIBMAX		82	/* Attempting to link in too many shared libraries */
-#define	ELIBEXEC	83	/* Cannot exec a shared library directly */
-#define	EILSEQ		84	/* Illegal byte sequence */
-#define	ERESTART	85	/* Interrupted system call should be restarted */
-#define	ESTRPIPE	86	/* Streams pipe error */
-#define	EUSERS		87	/* Too many users */
-#define	ENOTSOCK	88	/* Socket operation on non-socket */
-#define	EDESTADDRREQ	89	/* Destination address required */
-#define	EMSGSIZE	90	/* Message too long */
-#define	EPROTOTYPE	91	/* Protocol wrong type for socket */
-#define	ENOPROTOOPT	92	/* Protocol not available */
-#define	EPROTONOSUPPORT	93	/* Protocol not supported */
-#define	ESOCKTNOSUPPORT	94	/* Socket type not supported */
-#define	EOPNOTSUPP	95	/* Operation not supported on transport endpoint */
-#define	EPFNOSUPPORT	96	/* Protocol family not supported */
-#define	EAFNOSUPPORT	97	/* Address family not supported by protocol */
-#define	EADDRINUSE	98	/* Address already in use */
-#define	EADDRNOTAVAIL	99	/* Cannot assign requested address */
-#define	ENETDOWN	100	/* Network is down */
-#define	ENETUNREACH	101	/* Network is unreachable */
-#define	ENETRESET	102	/* Network dropped connection because of reset */
-#define	ECONNABORTED	103	/* Software caused connection abort */
-#define	ECONNRESET	104	/* Connection reset by peer */
-#define	ENOBUFS		105	/* No buffer space available */
-#define	EISCONN		106	/* Transport endpoint is already connected */
-#define	ENOTCONN	107	/* Transport endpoint is not connected */
-#define	ESHUTDOWN	108	/* Cannot send after transport endpoint shutdown */
-#define	ETOOMANYREFS	109	/* Too many references: cannot splice */
-#define	ETIMEDOUT	110	/* Connection timed out */
-#define	ECONNREFUSED	111	/* Connection refused */
-#define	EHOSTDOWN	112	/* Host is down */
-#define	EHOSTUNREACH	113	/* No route to host */
-#define	EALREADY	114	/* Operation already in progress */
-#define	EINPROGRESS	115	/* Operation now in progress */
-#define	ESTALE		116	/* Stale NFS file handle */
-#define	EUCLEAN		117	/* Structure needs cleaning */
-#define	ENOTNAM		118	/* Not a XENIX named type file */
-#define	ENAVAIL		119	/* No XENIX semaphores available */
-#define	EISNAM		120	/* Is a named type file */
-#define	EREMOTEIO	121	/* Remote I/O error */
-#define	EDQUOT		122	/* Quota exceeded */
-
-#define	ENOMEDIUM	123	/* No medium found */
-#define	EMEDIUMTYPE	124	/* Wrong medium type */
+#include <asm-generic/errno.h>
 
 #endif	/* _XTENSA_ERRNO_H */
diff --git a/include/asm-xtensa/ipc.h b/include/asm-xtensa/ipc.h
index d37bdb4d4c9..a9eed4e21cb 100644
--- a/include/asm-xtensa/ipc.h
+++ b/include/asm-xtensa/ipc.h
@@ -11,24 +11,6 @@
 #ifndef _XTENSA_IPC_H
 #define _XTENSA_IPC_H
 
-struct ipc_kludge {
-	struct msgbuf __user *msgp;
-	long msgtyp;
-};
-
-#define SEMOP		 1
-#define SEMGET		 2
-#define SEMCTL		 3
-#define SEMTIMEDOP	 4
-#define MSGSND		11
-#define MSGRCV		12
-#define MSGGET		13
-#define MSGCTL		14
-#define SHMAT		21
-#define SHMDT		22
-#define SHMGET		23
-#define SHMCTL		24
-
-#define IPCCALL(version,op)	((version)<<16 | (op))
+#include <asm-generic/ipc.h>
 
 #endif	/* _XTENSA_IPC_H */
-- 
cgit v1.2.3-70-g09d2


From 9ec55a9bd365dfc78945bb8e6bf5d0fdf1d75ad0 Mon Sep 17 00:00:00 2001
From: Chris Zankel <czankel@tensilica.com>
Date: Thu, 30 Jun 2005 02:59:00 -0700
Subject: [PATCH] xtensa: Fix asm macro

Removed dead code in arch/xtensa/kernel/pci.c and use the pci_name() macro.
 Fixed an error in the delay asm macro: '1' is an invalid immediate value.

Signed-off-by: Chris Zankel <chris@zankel.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/xtensa/kernel/pci.c   | 95 +++-------------------------------------------
 include/asm-xtensa/delay.h |  2 +-
 2 files changed, 6 insertions(+), 91 deletions(-)

(limited to 'include')

diff --git a/arch/xtensa/kernel/pci.c b/arch/xtensa/kernel/pci.c
index d29a8164863..09887c96e9a 100644
--- a/arch/xtensa/kernel/pci.c
+++ b/arch/xtensa/kernel/pci.c
@@ -57,50 +57,6 @@ struct pci_controller** pci_ctrl_tail = &pci_ctrl_head;
 
 static int pci_bus_count;
 
-static void pcibios_fixup_resources(struct pci_dev* dev);
-
-#if 0 // FIXME
-struct pci_fixup pcibios_fixups[] = {
-	{ DECLARE_PCI_FIXUP_HEADER, PCI_ANY_ID, PCI_ANY_ID, pcibios_fixup_resources },
-	{ 0 }
-};
-#endif
-
-void
-pcibios_update_resource(struct pci_dev *dev, struct resource *root,
-			struct resource *res, int resource)
-{
-	u32 new, check, mask;
-	int reg;
-	struct pci_controller* pci_ctrl = dev->sysdata;
-
-	new = res->start;
-	if (pci_ctrl && res->flags & IORESOURCE_IO) {
-		new -= pci_ctrl->io_space.base;
-	}
-	new |= (res->flags & PCI_REGION_FLAG_MASK);
-	if (resource < 6) {
-		reg = PCI_BASE_ADDRESS_0 + 4*resource;
-	} else if (resource == PCI_ROM_RESOURCE) {
-		res->flags |= PCI_ROM_ADDRESS_ENABLE;
-		reg = dev->rom_base_reg;
-	} else {
-	/* Somebody might have asked allocation of a non-standard resource */
-		return;
-	}
-
-	pci_write_config_dword(dev, reg, new);
-	pci_read_config_dword(dev, reg, &check);
-	mask = (new & PCI_BASE_ADDRESS_SPACE_IO) ?
-		PCI_BASE_ADDRESS_IO_MASK : PCI_BASE_ADDRESS_MEM_MASK;
-
-	if ((new ^ check) & mask) {
-		printk(KERN_ERR "PCI: Error while updating region "
-		       "%s/%d (%08x != %08x)\n", dev->slot_name, resource,
-		       new, check);
-	}
-}
-
 /*
  * We need to avoid collisions with `mirrored' VGA ports
  * and other strange ISA hardware, so we always want the
@@ -125,7 +81,7 @@ pcibios_align_resource(void *data, struct resource *res, unsigned long size,
 
 		if (size > 0x100) {
 			printk(KERN_ERR "PCI: I/O Region %s/%d too large"
-			       " (%ld bytes)\n", dev->slot_name,
+			       " (%ld bytes)\n", pci_name(dev),
 			       dev->resource - res, size);
 		}
 
@@ -149,7 +105,7 @@ pcibios_enable_resources(struct pci_dev *dev, int mask)
 		r = &dev->resource[idx];
 		if (!r->start && r->end) {
 			printk (KERN_ERR "PCI: Device %s not available because "
-				"of resource collisions\n", dev->slot_name);
+				"of resource collisions\n", pci_name(dev));
 			return -EINVAL;
 		}
 		if (r->flags & IORESOURCE_IO)
@@ -161,7 +117,7 @@ pcibios_enable_resources(struct pci_dev *dev, int mask)
 		cmd |= PCI_COMMAND_MEMORY;
 	if (cmd != old_cmd) {
 		printk("PCI: Enabling device %s (%04x -> %04x)\n",
-			dev->slot_name, old_cmd, cmd);
+			pci_name(dev), old_cmd, cmd);
 		pci_write_config_word(dev, PCI_COMMAND, cmd);
 	}
 	return 0;
@@ -293,7 +249,7 @@ int pcibios_enable_device(struct pci_dev *dev, int mask)
 		r = &dev->resource[idx];
 		if (!r->start && r->end) {
 			printk(KERN_ERR "PCI: Device %s not available because "
-			       "of resource collisions\n", dev->slot_name);
+			       "of resource collisions\n", pci_name(dev));
 			return -EINVAL;
 		}
 		if (r->flags & IORESOURCE_IO)
@@ -303,7 +259,7 @@ int pcibios_enable_device(struct pci_dev *dev, int mask)
 	}
 	if (cmd != old_cmd) {
 		printk("PCI: Enabling device %s (%04x -> %04x)\n",
-		       dev->slot_name, old_cmd, cmd);
+		       pci_name(dev), old_cmd, cmd);
 		pci_write_config_word(dev, PCI_COMMAND, cmd);
 	}
 
@@ -325,47 +281,6 @@ pci_controller_num(struct pci_dev *dev)
 
 #endif /* CONFIG_PROC_FS */
 
-
-static void
-pcibios_fixup_resources(struct pci_dev *dev)
-{
-	struct pci_controller* pci_ctrl = (struct pci_controller *)dev->sysdata;
-	int i;
-	unsigned long offset;
-
-	if (!pci_ctrl) {
-		printk(KERN_ERR "No pci_ctrl for PCI dev %s!\n",dev->slot_name);
-		return;
-	}
-	for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) {
-		struct resource *res = dev->resource + i;
-		if (!res->start || !res->flags)
-			continue;
-		if (res->end == 0xffffffff) {
-			DBG("PCI:%s Resource %d [%08lx-%08lx] is unassigned\n",
-			    dev->slot_name, i, res->start, res->end);
-			res->end -= res->start;
-			res->start = 0;
-			continue;
-		}
-		offset = 0;
-		if (res->flags & IORESOURCE_IO)
-			offset = (unsigned long) pci_ctrl->io_space.base;
-		else if (res->flags & IORESOURCE_MEM)
-			offset = (unsigned long) pci_ctrl->mem_space.base;
-
-		if (offset != 0) {
-			res->start += offset;
-			res->end += offset;
-#ifdef DEBUG
-			printk("Fixup res %d (%lx) of dev %s: %lx -> %lx\n",
-			       i, res->flags, dev->slot_name,
-			       res->start - offset, res->start);
-#endif
-		}
-	}
-}
-
 /*
  * Platform support for /proc/bus/pci/X/Y mmap()s,
  * modelled on the sparc64 implementation by Dave Miller.
diff --git a/include/asm-xtensa/delay.h b/include/asm-xtensa/delay.h
index 6359c55e77a..0a123d53a63 100644
--- a/include/asm-xtensa/delay.h
+++ b/include/asm-xtensa/delay.h
@@ -21,7 +21,7 @@ extern unsigned long loops_per_jiffy;
 extern __inline__ void __delay(unsigned long loops)
 {
   /* 2 cycles per loop. */
-  __asm__ __volatile__ ("1: addi %0, %0, -2; bgeui %0, 1, 1b"
+  __asm__ __volatile__ ("1: addi %0, %0, -2; bgeui %0, 2, 1b"
 			: "=r" (loops) : "0" (loops));
 }
 
-- 
cgit v1.2.3-70-g09d2