From 06ef66081620e94fe35a518f98624b83a140096e Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:10:45 -0700
Subject: [IA64] __ia64_syscall() is no longer used anywhere in the kernel. 
 Remove it.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S      | 12 ------------
 arch/ia64/kernel/ia64_ksyms.c |  3 ---
 2 files changed, 15 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index bd86fea49a0..5b469db2326 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -470,18 +470,6 @@ ENTRY(load_switch_stack)
 	br.cond.sptk.many b7
 END(load_switch_stack)
 
-GLOBAL_ENTRY(__ia64_syscall)
-	.regstk 6,0,0,0
-	mov r15=in5				// put syscall number in place
-	break __BREAK_SYSCALL
-	movl r2=errno
-	cmp.eq p6,p7=-1,r10
-	;;
-(p6)	st4 [r2]=r8
-(p6)	mov r8=-1
-	br.ret.sptk.many rp
-END(__ia64_syscall)
-
 GLOBAL_ENTRY(execve)
 	mov r15=__NR_execve			// put syscall number in place
 	break __BREAK_SYSCALL
diff --git a/arch/ia64/kernel/ia64_ksyms.c b/arch/ia64/kernel/ia64_ksyms.c
index 7bbf019c986..01572814abe 100644
--- a/arch/ia64/kernel/ia64_ksyms.c
+++ b/arch/ia64/kernel/ia64_ksyms.c
@@ -58,9 +58,6 @@ EXPORT_SYMBOL(__strlen_user);
 EXPORT_SYMBOL(__strncpy_from_user);
 EXPORT_SYMBOL(__strnlen_user);
 
-#include <asm/unistd.h>
-EXPORT_SYMBOL(__ia64_syscall);
-
 /* from arch/ia64/lib */
 extern void __divsi3(void);
 extern void __udivsi3(void);
-- 
cgit v1.2.3-70-g09d2


From 9ec1a7ad434fa951ff845dbee3397cf6ad4f76df Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:13:33 -0700
Subject: [IA64] Use dynamic prediction for RSE-clearing branches.

This by itself is good for a 1-2 cycle speed up.  Effect is bigger
when combined with the later patches.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index 5b469db2326..c0f28339d58 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -1012,14 +1012,14 @@ rse_clear_invalid:
 	mov loc5=0
 	mov loc6=0
 	mov loc7=0
-(pRecurse) br.call.sptk.few b0=rse_clear_invalid
+(pRecurse) br.call.dptk.few b0=rse_clear_invalid
 	;;
 	mov loc8=0
 	mov loc9=0
 	cmp.ne pReturn,p0=r0,in1	// if recursion count != 0, we need to do a br.ret
 	mov loc10=0
 	mov loc11=0
-(pReturn) br.ret.sptk.many b0
+(pReturn) br.ret.dptk.many b0
 #endif /* !CONFIG_ITANIUM */
 #	undef pRecurse
 #	undef pReturn
-- 
cgit v1.2.3-70-g09d2


From 3c79c8b1d92a9ae3edf3cbcd2c43c553ee0f1d83 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:15:13 -0700
Subject: [IA64] Schedule fp-clearing insns at least 6 cycles after reading
 ar.bsp.

Decreases syscall overhead by approximately 6 cycles.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S | 25 +++++++++++++++----------
 1 file changed, 15 insertions(+), 10 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index c0f28339d58..0c84bed1bda 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -705,15 +705,15 @@ ENTRY(ia64_leave_syscall)
 	// start restoring the state saved on the kernel stack (struct pt_regs):
 	ld8 r9=[r2],PT(CR_IPSR)-PT(R9)
 	ld8 r11=[r3],PT(CR_IIP)-PT(R11)
-	mov f6=f0		// clear f6
+	nop.i 0
 	;;
 	invala			// M0|1 invalidate ALAT
 	rsm psr.i | psr.ic	// M2 initiate turning off of interrupt and interruption collection
-	mov f9=f0		// clear f9
+	nop.i 0
 
 	ld8 r29=[r2],16		// load cr.ipsr
 	ld8 r28=[r3],16			// load cr.iip
-	mov f8=f0		// clear f8
+	mov r22=r0		// clear r22
 	;;
 	ld8 r30=[r2],16		// M0|1 load cr.ifs
 	ld8 r25=[r3],16		// M0|1 load ar.unat
@@ -721,15 +721,15 @@ ENTRY(ia64_leave_syscall)
 	;;
 	ld8 r26=[r2],PT(B0)-PT(AR_PFS)	// M0|1 load ar.pfs
 (pKStk)	mov r22=psr		// M2 read PSR now that interrupts are disabled
-	mov f10=f0		// clear f10
+(pUStk) add r14=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13
 	;;
 	ld8 r21=[r2],PT(AR_RNAT)-PT(B0) // load b0
 	ld8 r27=[r3],PT(PR)-PT(AR_RSC)	// load ar.rsc
-	mov f11=f0		// clear f11
+	mov f6=f0		// clear f6
 	;;
 	ld8 r24=[r2],PT(AR_FPSR)-PT(AR_RNAT)	// load ar.rnat (may be garbage)
 	ld8 r31=[r3],PT(R1)-PT(PR)		// load predicates
-(pUStk) add r14=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13
+	mov f7=f0		// clear f7
 	;;
 	ld8 r20=[r2],PT(R12)-PT(AR_FPSR)	// load ar.fpsr
 	ld8.fill r1=[r3],16	// load r1
@@ -737,24 +737,29 @@ ENTRY(ia64_leave_syscall)
 	;;
 	srlz.d			// M0  ensure interruption collection is off
 	ld8.fill r13=[r3],16
-	mov f7=f0		// clear f7
+	mov f8=f0		// clear f8
 	;;
 	ld8.fill r12=[r2]	// restore r12 (sp)
 	mov.m ar.ssd=r0		// M2 clear ar.ssd
-	mov r22=r0		// clear r22
+	mov b6=r18		// I0  restore b6
 
+	nop.m 0
+	mov f9=f0		// clear f9
+	shr.u r18=r19,16	// I0|1 get byte size of existing "dirty" partition
+	;;
 	ld8.fill r15=[r3]	// restore r15
 (pUStk) st1 [r14]=r17
 	addl r3=THIS_CPU(ia64_phys_stacked_size_p8),r0
 	;;
 (pUStk)	ld4 r17=[r3]		// r17 = cpu_data->phys_stacked_size_p8
 	mov.m ar.csd=r0		// M2 clear ar.csd
-	mov b6=r18		// I0  restore b6
+	mov f10=f0		// clear f10
 	;;
 	mov r14=r0		// clear r14
-	shr.u r18=r19,16	// I0|1 get byte size of existing "dirty" partition
+	mov f11=f0		// clear f11
 (pKStk) br.cond.dpnt.many skip_rbs_switch
 
+
 	mov.m ar.ccv=r0		// clear ar.ccv
 (pNonSys) br.cond.dpnt.many dont_preserve_current_frame
 	br.cond.sptk.many rbs_switch
-- 
cgit v1.2.3-70-g09d2


From 96e017495e6833adcbad84783e3c5eb685836bbf Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:16:07 -0700
Subject: [IA64] On return from syscall, hint b7 with
 __kernel_syscall_via_epc().

Why is this a good idea?  Clearing b7 to 0 is guaranteed to do us no
good and writing it with __kernel_syscall_via_epc() yields a 6 cycle
improvement _if_ the application performs another EPC-based system-
call without overwriting b7, which is not all that uncommon.  Well
worth the minimal cost of 1 bundle of code.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index 0c84bed1bda..6359d7ffbb7 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -692,7 +692,7 @@ ENTRY(ia64_leave_syscall)
 	;;
 (p6)	ld4 r31=[r18]				// load current_thread_info()->flags
 	ld8 r19=[r2],PT(B6)-PT(LOADRS)		// load ar.rsc value for "loadrs"
-	mov b7=r0		// clear b7
+	nop.i 0
 	;;
 	ld8 r23=[r3],PT(R11)-PT(AR_BSPSTORE)	// load ar.bspstore (may be garbage)
 	ld8 r18=[r2],PT(R9)-PT(B6)		// load b6
@@ -754,7 +754,14 @@ ENTRY(ia64_leave_syscall)
 (pUStk)	ld4 r17=[r3]		// r17 = cpu_data->phys_stacked_size_p8
 	mov.m ar.csd=r0		// M2 clear ar.csd
 	mov f10=f0		// clear f10
+
+	nop.m 0
+	movl r14=__kernel_syscall_via_epc	// X
 	;;
+	nop.m 0
+	nop.m 0
+	mov b7=r14		// I0   clear b7 (hint with __kernel_syscall_via_epc)
+
 	mov r14=r0		// clear r14
 	mov f11=f0		// clear f11
 (pKStk) br.cond.dpnt.many skip_rbs_switch
-- 
cgit v1.2.3-70-g09d2


From 060561ff79b01eea58e6d72abfb8e7580ff21f2a Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:17:03 -0700
Subject: [IA64] In syscall-entry, use st8 instead of stf8 to clear pt_regs.r8

Using stf8 seemed like a clever idea at the time, but stf8 forces
the cache-line to be invalidated in the L1D (if it happens to be
there already).  This patch eliminates a guaranteed L1D cache-miss
and, by itself, is good for a 1-2 cycle improvement for heavy-weight
syscalls.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/ivt.S | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/ivt.S b/arch/ia64/kernel/ivt.S
index d9c05d53435..dc78c356ddd 100644
--- a/arch/ia64/kernel/ivt.S
+++ b/arch/ia64/kernel/ivt.S
@@ -1,7 +1,7 @@
 /*
  * arch/ia64/kernel/ivt.S
  *
- * Copyright (C) 1998-2001, 2003 Hewlett-Packard Co
+ * Copyright (C) 1998-2001, 2003, 2005 Hewlett-Packard Co
  *	Stephane Eranian <eranian@hpl.hp.com>
  *	David Mosberger <davidm@hpl.hp.com>
  * Copyright (C) 2000, 2002-2003 Intel Co
@@ -918,7 +918,7 @@ GLOBAL_ENTRY(ia64_syscall_setup)
 	tnat.nz p14,p0=in6
 	cmp.lt p10,p9=r11,r8	// frame size can't be more than local+8
 	;;
-	stf8 [r16]=f1		// ensure pt_regs.r8 != 0 (see handle_syscall_error)
+	mov r8=1
 (p9)	tnat.nz p10,p0=r15
 	adds r12=-16,r1		// switch to kernel memory stack (with 16 bytes of scratch)
 
@@ -929,9 +929,9 @@ GLOBAL_ENTRY(ia64_syscall_setup)
 	mov r13=r2				// establish `current'
 	movl r1=__gp				// establish kernel global pointer
 	;;
+	st8 [r16]=r8		// ensure pt_regs.r8 != 0 (see handle_syscall_error)
 (p14)	mov in6=-1
 (p8)	mov in7=-1
-	nop.i 0
 
 	cmp.eq pSys,pNonSys=r0,r0		// set pSys=1, pNonSys=0
 	movl r17=FPSR_DEFAULT
-- 
cgit v1.2.3-70-g09d2


From 87e522a0f7f8a7a5a1a880517989835c4f09c63e Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:17:44 -0700
Subject: [IA64] Schedule ia64_leave_syscall() to read ar.bsp earlier

Reschedule code to read ar.bsp as early as possible.  To enable this,
don't bother clearing some of the registers when we're returning to
kernel stacks.  Also, instead of trying to support the pNonSys case
(which makes no sense), do a bugcheck instead (with break 0).  Finally,
remove a clear of r14 which is a left-over from the previous patch.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S | 43 +++++++++++++++++++------------------------
 1 file changed, 19 insertions(+), 24 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index 6359d7ffbb7..d97a07c77a4 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -694,22 +694,22 @@ ENTRY(ia64_leave_syscall)
 	ld8 r19=[r2],PT(B6)-PT(LOADRS)		// load ar.rsc value for "loadrs"
 	nop.i 0
 	;;
-	ld8 r23=[r3],PT(R11)-PT(AR_BSPSTORE)	// load ar.bspstore (may be garbage)
+	mov r16=ar.bsp				// M2  get existing backing store pointer
 	ld8 r18=[r2],PT(R9)-PT(B6)		// load b6
 (p6)	and r15=TIF_WORK_MASK,r31		// any work other than TIF_SYSCALL_TRACE?
 	;;
-	mov r16=ar.bsp				// M2  get existing backing store pointer
+	ld8 r23=[r3],PT(R11)-PT(AR_BSPSTORE)	// load ar.bspstore (may be garbage)
 (p6)	cmp4.ne.unc p6,p0=r15, r0		// any special work pending?
 (p6)	br.cond.spnt .work_pending_syscall
 	;;
 	// start restoring the state saved on the kernel stack (struct pt_regs):
 	ld8 r9=[r2],PT(CR_IPSR)-PT(R9)
 	ld8 r11=[r3],PT(CR_IIP)-PT(R11)
-	nop.i 0
+(pNonSys) break 0		//      bug check: we shouldn't be here if pNonSys is TRUE!
 	;;
 	invala			// M0|1 invalidate ALAT
 	rsm psr.i | psr.ic	// M2 initiate turning off of interrupt and interruption collection
-	nop.i 0
+	cmp.eq p9,p0=r0,r0	// set p9 to indicate that we should restore cr.ifs
 
 	ld8 r29=[r2],16		// load cr.ipsr
 	ld8 r28=[r3],16			// load cr.iip
@@ -717,11 +717,11 @@ ENTRY(ia64_leave_syscall)
 	;;
 	ld8 r30=[r2],16		// M0|1 load cr.ifs
 	ld8 r25=[r3],16		// M0|1 load ar.unat
-	cmp.eq p9,p0=r0,r0	// set p9 to indicate that we should restore cr.ifs
+(pUStk) add r14=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13
 	;;
 	ld8 r26=[r2],PT(B0)-PT(AR_PFS)	// M0|1 load ar.pfs
 (pKStk)	mov r22=psr		// M2 read PSR now that interrupts are disabled
-(pUStk) add r14=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13
+	nop 0
 	;;
 	ld8 r21=[r2],PT(AR_RNAT)-PT(B0) // load b0
 	ld8 r27=[r3],PT(PR)-PT(AR_RSC)	// load ar.rsc
@@ -735,40 +735,35 @@ ENTRY(ia64_leave_syscall)
 	ld8.fill r1=[r3],16	// load r1
 (pUStk) mov r17=1
 	;;
-	srlz.d			// M0  ensure interruption collection is off
+(pUStk) st1 [r14]=r17
 	ld8.fill r13=[r3],16
 	mov f8=f0		// clear f8
 	;;
 	ld8.fill r12=[r2]	// restore r12 (sp)
-	mov.m ar.ssd=r0		// M2 clear ar.ssd
+	ld8.fill r15=[r3]	// restore r15
 	mov b6=r18		// I0  restore b6
 
-	nop.m 0
+	addl r17=THIS_CPU(ia64_phys_stacked_size_p8),r0
 	mov f9=f0		// clear f9
+(pKStk) br.cond.dpnt.many skip_rbs_switch
+
+	srlz.d			// M0  ensure interruption collection is off
 	shr.u r18=r19,16	// I0|1 get byte size of existing "dirty" partition
+	cover			// B    add current frame into dirty partition and set cr.ifs
 	;;
-	ld8.fill r15=[r3]	// restore r15
-(pUStk) st1 [r14]=r17
-	addl r3=THIS_CPU(ia64_phys_stacked_size_p8),r0
-	;;
-(pUStk)	ld4 r17=[r3]		// r17 = cpu_data->phys_stacked_size_p8
-	mov.m ar.csd=r0		// M2 clear ar.csd
+(pUStk)	ld4 r17=[r17]		// r17 = cpu_data->phys_stacked_size_p8
+	mov r19=ar.bsp		// M2   get new backing store pointer
 	mov f10=f0		// clear f10
 
 	nop.m 0
 	movl r14=__kernel_syscall_via_epc	// X
 	;;
-	nop.m 0
-	nop.m 0
+	mov.m ar.csd=r0		// M2 clear ar.csd
+	mov.m ar.ccv=r0		// clear ar.ccv
 	mov b7=r14		// I0   clear b7 (hint with __kernel_syscall_via_epc)
 
-	mov r14=r0		// clear r14
+	mov.m ar.ssd=r0		// M2 clear ar.ssd
 	mov f11=f0		// clear f11
-(pKStk) br.cond.dpnt.many skip_rbs_switch
-
-
-	mov.m ar.ccv=r0		// clear ar.ccv
-(pNonSys) br.cond.dpnt.many dont_preserve_current_frame
 	br.cond.sptk.many rbs_switch
 END(ia64_leave_syscall)
 
@@ -946,10 +941,10 @@ GLOBAL_ENTRY(ia64_leave_kernel)
 	 */
 (pNonSys) br.cond.dpnt dont_preserve_current_frame
 
-rbs_switch:
 	cover				// add current frame into dirty partition and set cr.ifs
 	;;
 	mov r19=ar.bsp			// get new backing store pointer
+rbs_switch:
 	sub r16=r16,r18			// krbs = old bsp - size of dirty partition
 	cmp.ne p9,p0=r0,r0		// clear p9 to skip restore of cr.ifs
 	;;
-- 
cgit v1.2.3-70-g09d2


From c03f058fbf685f2ff630095d2c1e98d331b81e82 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:18:22 -0700
Subject: [IA64] In ia64_leave_syscall(), fix comments and whitespace only.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S | 79 ++++++++++++++++++++++++------------------------
 1 file changed, 39 insertions(+), 40 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index d97a07c77a4..a7542c3d3b3 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -625,7 +625,7 @@ END(ia64_ret_from_syscall)
  *	      r8-r11: restored (syscall return value(s))
  *		 r12: restored (user-level stack pointer)
  *		 r13: restored (user-level thread pointer)
- *		 r14: cleared
+ *		 r14: set to __kernel_syscall_via_epc
  *		 r15: restored (syscall #)
  *	     r16-r17: cleared
  *		 r18: user-level b6
@@ -646,7 +646,7 @@ END(ia64_ret_from_syscall)
  *		  pr: restored (user-level pr)
  *		  b0: restored (user-level rp)
  *	          b6: restored
- *		  b7: cleared
+ *		  b7: set to __kernel_syscall_via_epc
  *	     ar.unat: restored (user-level ar.unat)
  *	      ar.pfs: restored (user-level ar.pfs)
  *	      ar.rsc: restored (user-level ar.rsc)
@@ -708,63 +708,63 @@ ENTRY(ia64_leave_syscall)
 (pNonSys) break 0		//      bug check: we shouldn't be here if pNonSys is TRUE!
 	;;
 	invala			// M0|1 invalidate ALAT
-	rsm psr.i | psr.ic	// M2 initiate turning off of interrupt and interruption collection
-	cmp.eq p9,p0=r0,r0	// set p9 to indicate that we should restore cr.ifs
+	rsm psr.i | psr.ic	// M2   turn off interrupts and interruption collection
+	cmp.eq p9,p0=r0,r0	// A    set p9 to indicate that we should restore cr.ifs
 
-	ld8 r29=[r2],16		// load cr.ipsr
-	ld8 r28=[r3],16			// load cr.iip
-	mov r22=r0		// clear r22
+	ld8 r29=[r2],16		// M0|1 load cr.ipsr
+	ld8 r28=[r3],16		// M0|1 load cr.iip
+	mov r22=r0		// A    clear r22
 	;;
 	ld8 r30=[r2],16		// M0|1 load cr.ifs
 	ld8 r25=[r3],16		// M0|1 load ar.unat
 (pUStk) add r14=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13
 	;;
 	ld8 r26=[r2],PT(B0)-PT(AR_PFS)	// M0|1 load ar.pfs
-(pKStk)	mov r22=psr		// M2 read PSR now that interrupts are disabled
+(pKStk)	mov r22=psr			// M2   read PSR now that interrupts are disabled
 	nop 0
 	;;
-	ld8 r21=[r2],PT(AR_RNAT)-PT(B0) // load b0
-	ld8 r27=[r3],PT(PR)-PT(AR_RSC)	// load ar.rsc
-	mov f6=f0		// clear f6
+	ld8 r21=[r2],PT(AR_RNAT)-PT(B0) // M0|1 load b0
+	ld8 r27=[r3],PT(PR)-PT(AR_RSC)	// M0|1 load ar.rsc
+	mov f6=f0			// F    clear f6
 	;;
-	ld8 r24=[r2],PT(AR_FPSR)-PT(AR_RNAT)	// load ar.rnat (may be garbage)
-	ld8 r31=[r3],PT(R1)-PT(PR)		// load predicates
-	mov f7=f0		// clear f7
+	ld8 r24=[r2],PT(AR_FPSR)-PT(AR_RNAT)	// M0|1 load ar.rnat (may be garbage)
+	ld8 r31=[r3],PT(R1)-PT(PR)		// M0|1 load predicates
+	mov f7=f0				// F    clear f7
 	;;
-	ld8 r20=[r2],PT(R12)-PT(AR_FPSR)	// load ar.fpsr
-	ld8.fill r1=[r3],16	// load r1
-(pUStk) mov r17=1
+	ld8 r20=[r2],PT(R12)-PT(AR_FPSR)	// M0|1 load ar.fpsr
+	ld8.fill r1=[r3],16			// M0|1 load r1
+(pUStk) mov r17=1				// A
 	;;
-(pUStk) st1 [r14]=r17
-	ld8.fill r13=[r3],16
-	mov f8=f0		// clear f8
+(pUStk) st1 [r14]=r17				// M2|3
+	ld8.fill r13=[r3],16			// M0|1
+	mov f8=f0				// F    clear f8
 	;;
-	ld8.fill r12=[r2]	// restore r12 (sp)
-	ld8.fill r15=[r3]	// restore r15
-	mov b6=r18		// I0  restore b6
+	ld8.fill r12=[r2]			// M0|1 restore r12 (sp)
+	ld8.fill r15=[r3]			// M0|1 restore r15
+	mov b6=r18				// I0   restore b6
 
-	addl r17=THIS_CPU(ia64_phys_stacked_size_p8),r0
-	mov f9=f0		// clear f9
-(pKStk) br.cond.dpnt.many skip_rbs_switch
+	addl r17=THIS_CPU(ia64_phys_stacked_size_p8),r0 // A
+	mov f9=f0					// F    clear f9
+(pKStk) br.cond.dpnt.many skip_rbs_switch		// B
 
-	srlz.d			// M0  ensure interruption collection is off
-	shr.u r18=r19,16	// I0|1 get byte size of existing "dirty" partition
-	cover			// B    add current frame into dirty partition and set cr.ifs
+	srlz.d				// M0   ensure interruption collection is off (for cover)
+	shr.u r18=r19,16		// I0|1 get byte size of existing "dirty" partition
+	cover				// B    add current frame into dirty partition & set cr.ifs
 	;;
-(pUStk)	ld4 r17=[r17]		// r17 = cpu_data->phys_stacked_size_p8
-	mov r19=ar.bsp		// M2   get new backing store pointer
-	mov f10=f0		// clear f10
+(pUStk) ld4 r17=[r17]			// M0|1 r17 = cpu_data->phys_stacked_size_p8
+	mov r19=ar.bsp			// M2   get new backing store pointer
+	mov f10=f0			// F    clear f10
 
 	nop.m 0
-	movl r14=__kernel_syscall_via_epc	// X
+	movl r14=__kernel_syscall_via_epc // X
 	;;
-	mov.m ar.csd=r0		// M2 clear ar.csd
-	mov.m ar.ccv=r0		// clear ar.ccv
-	mov b7=r14		// I0   clear b7 (hint with __kernel_syscall_via_epc)
+	mov.m ar.csd=r0			// M2   clear ar.csd
+	mov.m ar.ccv=r0			// M2   clear ar.ccv
+	mov b7=r14			// I0   clear b7 (hint with __kernel_syscall_via_epc)
 
-	mov.m ar.ssd=r0		// M2 clear ar.ssd
-	mov f11=f0		// clear f11
-	br.cond.sptk.many rbs_switch
+	mov.m ar.ssd=r0			// M2   clear ar.ssd
+	mov f11=f0			// F    clear f11
+	br.cond.sptk.many rbs_switch	// B
 END(ia64_leave_syscall)
 
 #ifdef CONFIG_IA32_SUPPORT
@@ -940,7 +940,6 @@ GLOBAL_ENTRY(ia64_leave_kernel)
 	 * NOTE: alloc, loadrs, and cover can't be predicated.
 	 */
 (pNonSys) br.cond.dpnt dont_preserve_current_frame
-
 	cover				// add current frame into dirty partition and set cr.ifs
 	;;
 	mov r19=ar.bsp			// get new backing store pointer
-- 
cgit v1.2.3-70-g09d2


From f8fa5448fc9b4a7806b1297a0b57808f12fe4d43 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:19:04 -0700
Subject: [IA64] Reschedule break_fault() for better performance.

This patch reorganizes break_fault() to optimistically assume that a
system-call is being performed from user-space (which is almost always
the case).  If it turns out that (a) we're not being called due to a
system call or (b) we're being called from within the kernel, we fixup
the no-longer-valid assumptions in non_syscall() and .break_fixup(),
respectively.

With this approach, there are 3 major phases:

 - Phase 1: Read various control & application registers, in
	    particular the current task pointer from AR.K6.
 - Phase 2: Do all memory loads (load system-call entry,
	    load current_thread_info()->flags, prefetch
	    kernel register-backing store) and switch
	    to kernel register-stack.
 - Phase 3: Call ia64_syscall_setup() and invoke
	    syscall-handler.

Good for 26-30 cycles of improvement on break-based syscall-path.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/ivt.S | 173 ++++++++++++++++++++++++++++++-------------------
 1 file changed, 106 insertions(+), 67 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/ivt.S b/arch/ia64/kernel/ivt.S
index dc78c356ddd..386087edabf 100644
--- a/arch/ia64/kernel/ivt.S
+++ b/arch/ia64/kernel/ivt.S
@@ -687,82 +687,118 @@ ENTRY(break_fault)
 	 * to prevent leaking bits from kernel to user level.
 	 */
 	DBG_FAULT(11)
-	mov r16=IA64_KR(CURRENT)		// r16 = current task; 12 cycle read lat.
-	mov r17=cr.iim
-	mov r18=__IA64_BREAK_SYSCALL
-	mov r21=ar.fpsr
-	mov r29=cr.ipsr
-	mov r19=b6
-	mov r25=ar.unat
-	mov r27=ar.rsc
-	mov r26=ar.pfs
-	mov r28=cr.iip
-	mov r31=pr				// prepare to save predicates
-	mov r20=r1
-	;;
+	mov.m r16=IA64_KR(CURRENT)		// M2 r16 <- current task (12 cyc)
+	mov r29=cr.ipsr				// M2 (12 cyc)
+	mov r31=pr				// I0 (2 cyc)
+
+	mov r17=cr.iim				// M2 (2 cyc)
+	mov.m r27=ar.rsc			// M2 (12 cyc)
+	mov r18=__IA64_BREAK_SYSCALL		// A
+
+	mov.m ar.rsc=0				// M2
+	mov.m r21=ar.fpsr			// M2 (12 cyc)
+	mov r19=b6				// I0 (2 cyc)
+	;;
+	mov.m r23=ar.bspstore			// M2 (12 cyc)
+	mov.m r24=ar.rnat			// M2 (5 cyc)
+	mov.i r26=ar.pfs			// I0 (2 cyc)
+
+	invala					// M0|1
+	nop.m 0					// M
+	mov r20=r1				// A			save r1
+
+	nop.m 0
+	movl r30=sys_call_table			// X
+
+	mov r28=cr.iip				// M2 (2 cyc)
+	cmp.eq p0,p7=r18,r17			// I0 is this a system call?
+(p7)	br.cond.spnt non_syscall		// B  no ->
+	//
+	// From this point on, we are definitely on the syscall-path
+	// and we can use (non-banked) scratch registers.
+	//
+///////////////////////////////////////////////////////////////////////
+	mov r1=r16				// A    move task-pointer to "addl"-addressable reg
+	mov r2=r16				// A    setup r2 for ia64_syscall_setup
+	add r9=TI_FLAGS+IA64_TASK_SIZE,r16	// A	r9 = &current_thread_info()->flags
+
 	adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r16
-	cmp.eq p0,p7=r18,r17			// is this a system call? (p7 <- false, if so)
-(p7)	br.cond.spnt non_syscall
+	adds r15=-1024,r15			// A    subtract 1024 from syscall number
+	mov r3=NR_syscalls - 1
 	;;
-	ld1 r17=[r16]				// load current->thread.on_ustack flag
-	st1 [r16]=r0				// clear current->thread.on_ustack flag
-	add r1=-IA64_TASK_THREAD_ON_USTACK_OFFSET,r16	// set r1 for MINSTATE_START_SAVE_MIN_VIRT
+	ld1.bias r17=[r16]			// M0|1 r17 = current->thread.on_ustack flag
+	ld4 r9=[r9]				// M0|1 r9 = current_thread_info()->flags
+	extr.u r8=r29,41,2			// I0   extract ei field from cr.ipsr
+
+	shladd r30=r15,3,r30			// A    r30 = sys_call_table + 8*(syscall-1024)
+	addl r22=IA64_RBS_OFFSET,r1		// A    compute base of RBS
+	cmp.leu p6,p7=r15,r3			// A    syscall number in range?
 	;;
-	invala
 
-	/* adjust return address so we skip over the break instruction: */
+	lfetch.fault.excl.nt1 [r22]		// M0|1 prefetch RBS
+(p6)	ld8 r30=[r30]				// M0|1 load address of syscall entry point
+	tnat.nz.or p7,p0=r15			// I0	is syscall nr a NaT?
 
-	extr.u r8=r29,41,2			// extract ei field from cr.ipsr
-	;;
-	cmp.eq p6,p7=2,r8			// isr.ei==2?
-	mov r2=r1				// setup r2 for ia64_syscall_setup
-	;;
-(p6)	mov r8=0				// clear ei to 0
-(p6)	adds r28=16,r28				// switch cr.iip to next bundle cr.ipsr.ei wrapped
-(p7)	adds r8=1,r8				// increment ei to next slot
-	;;
-	cmp.eq pKStk,pUStk=r0,r17		// are we in kernel mode already?
-	dep r29=r8,r29,41,2			// insert new ei into cr.ipsr
+	mov.m ar.bspstore=r22			// M2   switch to kernel RBS
+	cmp.eq p8,p9=2,r8			// A    isr.ei==2?
 	;;
 
-	// switch from user to kernel RBS:
-	MINSTATE_START_SAVE_MIN_VIRT
-	br.call.sptk.many b7=ia64_syscall_setup
-	;;
-	MINSTATE_END_SAVE_MIN_VIRT		// switch to bank 1
-	ssm psr.ic | PSR_DEFAULT_BITS
-	;;
-	srlz.i					// guarantee that interruption collection is on
-	mov r3=NR_syscalls - 1
-	;;
-(p15)	ssm psr.i				// restore psr.i
-	// p10==true means out registers are more than 8 or r15's Nat is true
-(p10)	br.cond.spnt.many ia64_ret_from_syscall
-	;;
-	movl r16=sys_call_table
+(p8)	mov r8=0				// A    clear ei to 0
+(p7)	movl r30=sys_ni_syscall			// X
 
-	adds r15=-1024,r15			// r15 contains the syscall number---subtract 1024
-	movl r2=ia64_ret_from_syscall
-	;;
-	shladd r20=r15,3,r16			// r20 = sys_call_table + 8*(syscall-1024)
-	cmp.leu p6,p7=r15,r3			// (syscall > 0 && syscall < 1024 + NR_syscalls) ?
-	mov rp=r2				// set the real return addr
+(p8)	adds r28=16,r28				// A    switch cr.iip to next bundle
+(p9)	adds r8=1,r8				// A    increment ei to next slot
+	nop.i 0
 	;;
-(p6)	ld8 r20=[r20]				// load address of syscall entry point
-(p7)	movl r20=sys_ni_syscall
 
-	add r2=TI_FLAGS+IA64_TASK_SIZE,r13
-	;;
-	ld4 r2=[r2]				// r2 = current_thread_info()->flags
-	;;
-	and r2=_TIF_SYSCALL_TRACEAUDIT,r2	// mask trace or audit
+	mov.m r25=ar.unat			// M2 (5 cyc)
+	dep r29=r8,r29,41,2			// I0   insert new ei into cr.ipsr
+	adds r15=1024,r15			// A    restore original syscall number
+	//
+	// If any of the above loads miss in L1D, we'll stall here until
+	// the data arrives.
+	//
+///////////////////////////////////////////////////////////////////////
+	st1 [r16]=r0				// M2|3 clear current->thread.on_ustack flag
+	mov b6=r30				// I0   setup syscall handler branch reg early
+	cmp.eq pKStk,pUStk=r0,r17		// A    were we on kernel stacks already?
+
+	and r9=_TIF_SYSCALL_TRACEAUDIT,r9	// A    mask trace or audit
+	mov r18=ar.bsp				// M2 (12 cyc)
+(pKStk)	br.cond.spnt .break_fixup		// B	we're already in kernel-mode -- fix up RBS
+	;;
+.back_from_break_fixup:
+(pUStk)	addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r1 // A    compute base of memory stack
+	cmp.eq p14,p0=r9,r0			// A    are syscalls being traced/audited?
+	br.call.sptk.many b7=ia64_syscall_setup	// B
+1:
+	mov ar.rsc=0x3				// M2   set eager mode, pl 0, LE, loadrs=0
+	nop 0
+	bsw.1					// B (6 cyc) regs are saved, switch to bank 1
 	;;
-	cmp.eq p8,p0=r2,r0
-	mov b6=r20
+
+	ssm psr.ic | PSR_DEFAULT_BITS		// M2	now it's safe to re-enable intr.-collection
+	movl r3=ia64_ret_from_syscall		// X
 	;;
-(p8)	br.call.sptk.many b6=b6			// ignore this return addr
-	br.cond.sptk ia64_trace_syscall
+
+	srlz.i					// M0   ensure interruption collection is on
+	mov rp=r3				// I0   set the real return addr
+(p10)	br.cond.spnt.many ia64_ret_from_syscall	// B    return if bad call-frame or r15 is a NaT
+
+(p15)	ssm psr.i				// M2   restore psr.i
+(p14)	br.call.sptk.many b6=b6			// B    invoke syscall-handker (ignore return addr)
+	br.cond.spnt.many ia64_trace_syscall	// B	do syscall-tracing thingamagic
 	// NOT REACHED
+///////////////////////////////////////////////////////////////////////
+	// On entry, we optimistically assumed that we're coming from user-space.
+	// For the rare cases where a system-call is done from within the kernel,
+	// we fix things up at this point:
+.break_fixup:
+	add r1=-IA64_PT_REGS_SIZE,sp		// A    allocate space for pt_regs structure
+	mov ar.rnat=r24				// M2	restore kernel's AR.RNAT
+	;;
+	mov ar.bspstore=r23			// M2	restore kernel's AR.BSPSTORE
+	br.cond.sptk .back_from_break_fixup
 END(break_fault)
 
 	.org ia64_ivt+0x3000
@@ -837,8 +873,6 @@ END(interrupt)
 	 *	- r31: saved pr
 	 *	-  b0: original contents (to be saved)
 	 * On exit:
-	 *	- executing on bank 1 registers
-	 *	- psr.ic enabled, interrupts restored
 	 *	-  p10: TRUE if syscall is invoked with more than 8 out
 	 *		registers or r15's Nat is true
 	 *	-  r1: kernel's gp
@@ -846,8 +880,11 @@ END(interrupt)
 	 *	-  r8: -EINVAL if p10 is true
 	 *	- r12: points to kernel stack
 	 *	- r13: points to current task
+	 *	- r14: preserved (same as on entry)
+	 *	- p13: preserved
 	 *	- p15: TRUE if interrupts need to be re-enabled
 	 *	- ar.fpsr: set to kernel settings
+	 *	-  b6: preserved (same as on entry)
 	 */
 GLOBAL_ENTRY(ia64_syscall_setup)
 #if PT(B6) != 0
@@ -915,7 +952,7 @@ GLOBAL_ENTRY(ia64_syscall_setup)
 (p13)	mov in5=-1
 	;;
 	st8 [r16]=r21,PT(R8)-PT(AR_FPSR)	// save ar.fpsr
-	tnat.nz p14,p0=in6
+	tnat.nz p13,p0=in6
 	cmp.lt p10,p9=r11,r8	// frame size can't be more than local+8
 	;;
 	mov r8=1
@@ -930,7 +967,7 @@ GLOBAL_ENTRY(ia64_syscall_setup)
 	movl r1=__gp				// establish kernel global pointer
 	;;
 	st8 [r16]=r8		// ensure pt_regs.r8 != 0 (see handle_syscall_error)
-(p14)	mov in6=-1
+(p13)	mov in6=-1
 (p8)	mov in7=-1
 
 	cmp.eq pSys,pNonSys=r0,r0		// set pSys=1, pNonSys=0
@@ -1002,6 +1039,8 @@ END(dispatch_illegal_op_fault)
 	FAULT(17)
 
 ENTRY(non_syscall)
+	mov ar.rsc=r27			// restore ar.rsc before SAVE_MIN_WITH_COVER
+	;;
 	SAVE_MIN_WITH_COVER
 
 	// There is no particular reason for this code to be here, other than that
-- 
cgit v1.2.3-70-g09d2


From 70929a57cfea8c18de13fcea9ae6771018a98949 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:19:37 -0700
Subject: [IA64] Reschedule __kernel_syscall_via_epc().

Avoid some stalls, which is good for about 2 cycles when invoking a
light-weight handler.  When invoking a heavy-weight handler, this
helps by about 7 cycles, with most of the improvement coming from the
improved branch-prediction achieved by splitting the BBB bundle into
two MIB bundles.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/gate.S | 31 ++++++++++++++++++-------------
 1 file changed, 18 insertions(+), 13 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/gate.S b/arch/ia64/kernel/gate.S
index facf75acdc8..3cd3f2e971f 100644
--- a/arch/ia64/kernel/gate.S
+++ b/arch/ia64/kernel/gate.S
@@ -79,31 +79,34 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
 	;;
 	rsm psr.be // note: on McKinley "rsm psr.be/srlz.d" is slightly faster than "rum psr.be"
 	LOAD_FSYSCALL_TABLE(r14)
-
+	;;
 	mov r16=IA64_KR(CURRENT)		// 12 cycle read latency
-	tnat.nz p10,p9=r15
+	shladd r18=r17,3,r14
 	mov r19=NR_syscalls-1
 	;;
-	shladd r18=r17,3,r14
-
-	srlz.d
-	cmp.ne p8,p0=r0,r0			// p8 <- FALSE
+	lfetch [r18]				// M0|1
+	mov r29=psr				// read psr (12 cyc load latency)
 	/* Note: if r17 is a NaT, p6 will be set to zero.  */
 	cmp.geu p6,p7=r19,r17			// (syscall > 0 && syscall < 1024+NR_syscalls)?
 	;;
-(p6)	ld8 r18=[r18]
 	mov r21=ar.fpsr
-	add r14=-8,r14				// r14 <- addr of fsys_bubble_down entry
+	tnat.nz p10,p9=r15
+	mov r26=ar.pfs
 	;;
+	srlz.d
+(p6)	ld8 r18=[r18]
+	nop.i 0
+	;;
+	nop.m 0
 (p6)	mov b7=r18
-(p6)	tbit.z p8,p0=r18,0
+(p6)	tbit.z.unc p8,p0=r18,0
+
+	nop.m 0
+	nop.i 0
 (p8)	br.dptk.many b7
 
-(p6)	rsm psr.i
 	mov r27=ar.rsc
-	mov r26=ar.pfs
-	;;
-	mov r29=psr				// read psr (12 cyc load latency)
+(p6)	rsm psr.i
 /*
  * brl.cond doesn't work as intended because the linker would convert this branch
  * into a branch to a PLT.  Perhaps there will be a way to avoid this with some
@@ -111,6 +114,8 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
  * instead.
  */
 #ifdef CONFIG_ITANIUM
+	add r14=-8,r14				// r14 <- addr of fsys_bubble_down entry
+	;;
 (p6)	ld8 r14=[r14]				// r14 <- fsys_bubble_down
 	;;
 (p6)	mov b7=r14
-- 
cgit v1.2.3-70-g09d2


From 21bc4f9b34cc1eab3610955207f72c52495ae8ed Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:20:11 -0700
Subject: [IA64] Annotate __kernel_syscall_via_epc() with McKinley dispatch
 info.

Two other very minor changes: use "mov.i" instead of "mov" for reading
ar.pfs (for clarity; doesn't affect the code at all).  Also, predicate
the load of r14 for consistency.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/gate.S | 44 ++++++++++++++++++++++----------------------
 1 file changed, 22 insertions(+), 22 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/gate.S b/arch/ia64/kernel/gate.S
index 3cd3f2e971f..272e64c0e21 100644
--- a/arch/ia64/kernel/gate.S
+++ b/arch/ia64/kernel/gate.S
@@ -72,41 +72,41 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
 	 * bundle get executed.  The remaining code must be safe even if
 	 * they do not get executed.
 	 */
-	adds r17=-1024,r15
-	mov r10=0				// default to successful syscall execution
-	epc
+	adds r17=-1024,r15			// A
+	mov r10=0				// A    default to successful syscall execution
+	epc					// B	causes split-issue
 }
 	;;
-	rsm psr.be // note: on McKinley "rsm psr.be/srlz.d" is slightly faster than "rum psr.be"
-	LOAD_FSYSCALL_TABLE(r14)
+	rsm psr.be				// M2 (5 cyc to srlz.d)
+	LOAD_FSYSCALL_TABLE(r14)		// X
 	;;
-	mov r16=IA64_KR(CURRENT)		// 12 cycle read latency
-	shladd r18=r17,3,r14
-	mov r19=NR_syscalls-1
+	mov r16=IA64_KR(CURRENT)		// M2 (12 cyc)
+	shladd r18=r17,3,r14			// A
+	mov r19=NR_syscalls-1			// A
 	;;
 	lfetch [r18]				// M0|1
-	mov r29=psr				// read psr (12 cyc load latency)
-	/* Note: if r17 is a NaT, p6 will be set to zero.  */
-	cmp.geu p6,p7=r19,r17			// (syscall > 0 && syscall < 1024+NR_syscalls)?
+	mov r29=psr				// M2 (12 cyc)
+	// If r17 is a NaT, p6 will be zero
+	cmp.geu p6,p7=r19,r17			// A    (sysnr > 0 && sysnr < 1024+NR_syscalls)?
 	;;
-	mov r21=ar.fpsr
-	tnat.nz p10,p9=r15
-	mov r26=ar.pfs
+	mov r21=ar.fpsr				// M2 (12 cyc)
+	tnat.nz p10,p9=r15			// I0
+	mov.i r26=ar.pfs			// I0 (would stall anyhow due to srlz.d...)
 	;;
-	srlz.d
-(p6)	ld8 r18=[r18]
+	srlz.d					// M0 (forces split-issue) ensure PSR.BE==0
+(p6)	ld8 r18=[r18]				// M0|1
 	nop.i 0
 	;;
 	nop.m 0
-(p6)	mov b7=r18
-(p6)	tbit.z.unc p8,p0=r18,0
+(p6)	mov b7=r18				// I0
+(p6)	tbit.z.unc p8,p0=r18,0			// I0 (dual-issues with "mov b7=r18"!)
 
 	nop.m 0
 	nop.i 0
-(p8)	br.dptk.many b7
+(p8)	br.dptk.many b7				// B
 
-	mov r27=ar.rsc
-(p6)	rsm psr.i
+	mov r27=ar.rsc				// M2 (12 cyc)
+(p6)	rsm psr.i				// M2
 /*
  * brl.cond doesn't work as intended because the linker would convert this branch
  * into a branch to a PLT.  Perhaps there will be a way to avoid this with some
@@ -114,7 +114,7 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
  * instead.
  */
 #ifdef CONFIG_ITANIUM
-	add r14=-8,r14				// r14 <- addr of fsys_bubble_down entry
+(p6)	add r14=-8,r14				// r14 <- addr of fsys_bubble_down entry
 	;;
 (p6)	ld8 r14=[r14]				// r14 <- fsys_bubble_down
 	;;
-- 
cgit v1.2.3-70-g09d2


From 1ba7be7d691f6df2557d39c5b1a2e14c32e5dd20 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:20:51 -0700
Subject: [IA64] Reschedule fsys_bubble_down().

Improvements come from eliminating srlz.i, not scheduling AR/CR-reads
too early (while there are others still pending), scheduling the
backing-store switch as well as possible, splitting the BBB bundle
into a MIB/MBB pair.

Why is it safe to eliminate the srlz.i?  Observe
that we used to clear bits ~PSR_PRESERVED_BITS in PSR.L.  Since
PSR_PRESERVED_BITS==PSR.{UP,MFL,MFH,PK,DT,PP,SP,RT,IC}, we
ended up clearing PSR.{BE,AC,I,DFL,DFH,DI,DB,SI,TB}.  However,

 PSR.BE : already is turned off in __kernel_syscall_via_epc()
 PSR.AC : don't care (kernel normally turns PSR.AC on)
 PSR.I  : already turned off by the time fsys_bubble_down gets invoked
 PSR.DFL: always 0 (kernel never turns it on)
 PSR.DFH: don't care --- kernel never touches f32-f127 on its own
	  initiative
 PSR.DI : always 0 (kernel never turns it on)
 PSR.SI : always 0 (kernel never turns it on)
 PSR.DB : don't care --- kernel never enables kernel-level breakpoints
 PSR.TB : must be 0 already; if it wasn't zero on entry to
	  __kernel_syscall_via_epc, the branch to fsys_bubble_down
	  will trigger a taken branch; the taken-trap-handler then
	  converts the syscall into a break-based system-call.

In other words: all the bits we're clearying are either 0 already or
are don't cares!  Thus, we don't have to write PSR.L at all and we
don't have to do a srlz.i either.

Good for another ~20 cycle improvement for EPC-based heavy-weight
syscalls.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/fsys.S | 73 ++++++++++++++++++++++---------------------------
 1 file changed, 33 insertions(+), 40 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/fsys.S b/arch/ia64/kernel/fsys.S
index 0d8650f7fce..57c6556b1e0 100644
--- a/arch/ia64/kernel/fsys.S
+++ b/arch/ia64/kernel/fsys.S
@@ -549,9 +549,6 @@ GLOBAL_ENTRY(fsys_bubble_down)
 	 *	- r27: ar.rsc
 	 *	- r29: psr
 	 */
-#	define PSR_PRESERVED_BITS	(IA64_PSR_UP | IA64_PSR_MFL | IA64_PSR_MFH | IA64_PSR_PK \
-					 | IA64_PSR_DT | IA64_PSR_PP | IA64_PSR_SP | IA64_PSR_RT \
-					 | IA64_PSR_IC)
 	/*
 	 * Reading psr.l gives us only bits 0-31, psr.it, and psr.mc.  The rest we have
 	 * to synthesize.
@@ -560,62 +557,58 @@ GLOBAL_ENTRY(fsys_bubble_down)
 					 | IA64_PSR_BN | IA64_PSR_I)
 
 	invala
-	movl r8=PSR_ONE_BITS
+	movl r14=ia64_ret_from_syscall
 
-	mov r25=ar.unat			// save ar.unat (5 cyc)
-	movl r9=PSR_PRESERVED_BITS
-
-	mov ar.rsc=0			// set enforced lazy mode, pl 0, little-endian, loadrs=0
+	nop.m 0
 	movl r28=__kernel_syscall_via_break
 	;;
-	mov r23=ar.bspstore		// save ar.bspstore (12 cyc)
-	mov r31=pr			// save pr (2 cyc)
-	mov r20=r1			// save caller's gp in r20
-	;;
+
 	mov r2=r16			// copy current task addr to addl-addressable register
-	and r9=r9,r29
-	mov r19=b6			// save b6 (2 cyc)
+	adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r16
+	mov r31=pr			// save pr (2 cyc)
 	;;
-	mov psr.l=r9			// slam the door (17 cyc to srlz.i)
-	or r29=r8,r29			// construct cr.ipsr value to save
+	st1 [r16]=r0			// clear current->thread.on_ustack flag
 	addl r22=IA64_RBS_OFFSET,r2	// compute base of RBS
+	add r3=TI_FLAGS+IA64_TASK_SIZE,r2
 	;;
-	// GAS reports a spurious RAW hazard on the read of ar.rnat because it thinks
-	// we may be reading ar.itc after writing to psr.l.  Avoid that message with
-	// this directive:
-	dv_serialize_data
-	mov.m r24=ar.rnat		// read ar.rnat (5 cyc lat)
+	ld4 r3=[r3]				// r2 = current_thread_info()->flags
 	lfetch.fault.excl.nt1 [r22]
-	adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r2
-
-	// ensure previous insn group is issued before we stall for srlz.i:
+	nop.i 0
 	;;
-	srlz.i				// ensure new psr.l has been established
-	/////////////////////////////////////////////////////////////////////////////
-	////////// from this point on, execution is not interruptible anymore
-	/////////////////////////////////////////////////////////////////////////////
-	addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r2	// compute base of memory stack
-	cmp.ne pKStk,pUStk=r0,r0	// set pKStk <- 0, pUStk <- 1
+	mov ar.rsc=0			// set enforced lazy mode, pl 0, little-endian, loadrs=0
+	nop.m 0
+	nop.i 0
+	;;
+	mov r23=ar.bspstore		// save ar.bspstore (12 cyc)
+	mov.m r24=ar.rnat		// read ar.rnat (5 cyc lat)
+	nop.i 0
 	;;
-	st1 [r16]=r0			// clear current->thread.on_ustack flag
 	mov ar.bspstore=r22		// switch to kernel RBS
-	mov b6=r18			// copy syscall entry-point to b6 (7 cyc)
-	add r3=TI_FLAGS+IA64_TASK_SIZE,r2
+	movl r8=PSR_ONE_BITS			// X
 	;;
-	ld4 r3=[r3]				// r2 = current_thread_info()->flags
+	mov r25=ar.unat			// save ar.unat (5 cyc)
+	mov r19=b6			// save b6 (2 cyc)
+	mov r20=r1			// save caller's gp in r20
+	;;
+	or r29=r8,r29			// construct cr.ipsr value to save
+	mov b6=r18			// copy syscall entry-point to b6 (7 cyc)
+	addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r2	// compute base of memory stack
+
 	mov r18=ar.bsp			// save (kernel) ar.bsp (12 cyc)
-	mov ar.rsc=0x3			// set eager mode, pl 0, little-endian, loadrs=0
+	cmp.ne pKStk,pUStk=r0,r0	// set pKStk <- 0, pUStk <- 1
 	br.call.sptk.many b7=ia64_syscall_setup
 	;;
-	ssm psr.i
-	movl r2=ia64_ret_from_syscall
+	mov ar.rsc=0x3			// set eager mode, pl 0, little-endian, loadrs=0
+	mov rp=r14				// set the real return addr
+	nop.i 0
 	;;
-	mov rp=r2				// set the real return addr
+	ssm psr.i
 	tbit.z p8,p0=r3,TIF_SYSCALL_TRACE
-	;;
 (p10)	br.cond.spnt.many ia64_ret_from_syscall	// p10==true means out registers are more than 8
+
+	nop.m 0
 (p8)	br.call.sptk.many b6=b6		// ignore this return addr
-	br.cond.sptk ia64_trace_syscall
+	br.cond.spnt ia64_trace_syscall
 END(fsys_bubble_down)
 
 	.rodata
-- 
cgit v1.2.3-70-g09d2


From fbf7192ba06e4b55022231e5fdb9db8ce3ad6307 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:21:26 -0700
Subject: [IA64] Annotate fsys_bubble_down() with McKinley dispatch info.

This patch changes comments & formatting only.  There is no code
change.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/fsys.S | 118 ++++++++++++++++++++++++++++++------------------
 1 file changed, 74 insertions(+), 44 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/fsys.S b/arch/ia64/kernel/fsys.S
index 57c6556b1e0..d09a5b8a098 100644
--- a/arch/ia64/kernel/fsys.S
+++ b/arch/ia64/kernel/fsys.S
@@ -531,84 +531,114 @@ GLOBAL_ENTRY(fsys_bubble_down)
 	.altrp b6
 	.body
 	/*
-	 * We get here for syscalls that don't have a lightweight handler.  For those, we
-	 * need to bubble down into the kernel and that requires setting up a minimal
-	 * pt_regs structure, and initializing the CPU state more or less as if an
-	 * interruption had occurred.  To make syscall-restarts work, we setup pt_regs
-	 * such that cr_iip points to the second instruction in syscall_via_break.
-	 * Decrementing the IP hence will restart the syscall via break and not
-	 * decrementing IP will return us to the caller, as usual.  Note that we preserve
-	 * the value of psr.pp rather than initializing it from dcr.pp.  This makes it
-	 * possible to distinguish fsyscall execution from other privileged execution.
+	 * We get here for syscalls that don't have a lightweight
+	 * handler.  For those, we need to bubble down into the kernel
+	 * and that requires setting up a minimal pt_regs structure,
+	 * and initializing the CPU state more or less as if an
+	 * interruption had occurred.  To make syscall-restarts work,
+	 * we setup pt_regs such that cr_iip points to the second
+	 * instruction in syscall_via_break.  Decrementing the IP
+	 * hence will restart the syscall via break and not
+	 * decrementing IP will return us to the caller, as usual.
+	 * Note that we preserve the value of psr.pp rather than
+	 * initializing it from dcr.pp.  This makes it possible to
+	 * distinguish fsyscall execution from other privileged
+	 * execution.
 	 *
 	 * On entry:
-	 *	- normal fsyscall handler register usage, except that we also have:
+	 *	- normal fsyscall handler register usage, except
+	 *	  that we also have:
 	 *	- r18: address of syscall entry point
 	 *	- r21: ar.fpsr
 	 *	- r26: ar.pfs
 	 *	- r27: ar.rsc
 	 *	- r29: psr
+	 *
+	 * We used to clear some PSR bits here but that requires slow
+	 * serialization.  Fortuntely, that isn't really necessary.
+	 * The rationale is as follows: we used to clear bits
+	 * ~PSR_PRESERVED_BITS in PSR.L.  Since
+	 * PSR_PRESERVED_BITS==PSR.{UP,MFL,MFH,PK,DT,PP,SP,RT,IC}, we
+	 * ended up clearing PSR.{BE,AC,I,DFL,DFH,DI,DB,SI,TB}.
+	 * However,
+	 *
+	 * PSR.BE : already is turned off in __kernel_syscall_via_epc()
+	 * PSR.AC : don't care (kernel normally turns PSR.AC on)
+	 * PSR.I  : already turned off by the time fsys_bubble_down gets
+	 *	    invoked
+	 * PSR.DFL: always 0 (kernel never turns it on)
+	 * PSR.DFH: don't care --- kernel never touches f32-f127 on its own
+	 *	    initiative
+	 * PSR.DI : always 0 (kernel never turns it on)
+	 * PSR.SI : always 0 (kernel never turns it on)
+	 * PSR.DB : don't care --- kernel never enables kernel-level
+	 *	    breakpoints
+	 * PSR.TB : must be 0 already; if it wasn't zero on entry to
+	 *          __kernel_syscall_via_epc, the branch to fsys_bubble_down
+	 *          will trigger a taken branch; the taken-trap-handler then
+	 *          converts the syscall into a break-based system-call.
 	 */
 	/*
-	 * Reading psr.l gives us only bits 0-31, psr.it, and psr.mc.  The rest we have
-	 * to synthesize.
+	 * Reading psr.l gives us only bits 0-31, psr.it, and psr.mc.
+	 * The rest we have to synthesize.
 	 */
-#	define PSR_ONE_BITS		((3 << IA64_PSR_CPL0_BIT) | (0x1 << IA64_PSR_RI_BIT) \
+#	define PSR_ONE_BITS		((3 << IA64_PSR_CPL0_BIT)	\
+					 | (0x1 << IA64_PSR_RI_BIT)	\
 					 | IA64_PSR_BN | IA64_PSR_I)
 
-	invala
-	movl r14=ia64_ret_from_syscall
+	invala					// M0|1
+	movl r14=ia64_ret_from_syscall		// X
 
 	nop.m 0
-	movl r28=__kernel_syscall_via_break
+	movl r28=__kernel_syscall_via_break	// X	create cr.iip
 	;;
 
-	mov r2=r16			// copy current task addr to addl-addressable register
-	adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r16
-	mov r31=pr			// save pr (2 cyc)
+	mov r2=r16				// A    get task addr to addl-addressable register
+	adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r16 // A
+	mov r31=pr				// I0   save pr (2 cyc)
 	;;
-	st1 [r16]=r0			// clear current->thread.on_ustack flag
-	addl r22=IA64_RBS_OFFSET,r2	// compute base of RBS
-	add r3=TI_FLAGS+IA64_TASK_SIZE,r2
+	st1 [r16]=r0				// M2|3 clear current->thread.on_ustack flag
+	addl r22=IA64_RBS_OFFSET,r2		// A    compute base of RBS
+	add r3=TI_FLAGS+IA64_TASK_SIZE,r2	// A
 	;;
-	ld4 r3=[r3]				// r2 = current_thread_info()->flags
-	lfetch.fault.excl.nt1 [r22]
+	ld4 r3=[r3]				// M0|1 r3 = current_thread_info()->flags
+	lfetch.fault.excl.nt1 [r22]		// M0|1 prefetch register backing-store
 	nop.i 0
 	;;
-	mov ar.rsc=0			// set enforced lazy mode, pl 0, little-endian, loadrs=0
+	mov ar.rsc=0				// M2   set enforced lazy mode, pl 0, LE, loadrs=0
 	nop.m 0
 	nop.i 0
 	;;
-	mov r23=ar.bspstore		// save ar.bspstore (12 cyc)
-	mov.m r24=ar.rnat		// read ar.rnat (5 cyc lat)
+	mov r23=ar.bspstore			// M2 (12 cyc) save ar.bspstore
+	mov.m r24=ar.rnat			// M2 (5 cyc) read ar.rnat (dual-issues!)
 	nop.i 0
 	;;
-	mov ar.bspstore=r22		// switch to kernel RBS
+	mov ar.bspstore=r22			// M2 (6 cyc) switch to kernel RBS
 	movl r8=PSR_ONE_BITS			// X
 	;;
-	mov r25=ar.unat			// save ar.unat (5 cyc)
-	mov r19=b6			// save b6 (2 cyc)
-	mov r20=r1			// save caller's gp in r20
+	mov r25=ar.unat				// M2 (5 cyc) save ar.unat
+	mov r19=b6				// I0   save b6 (2 cyc)
+	mov r20=r1				// A    save caller's gp in r20
 	;;
-	or r29=r8,r29			// construct cr.ipsr value to save
-	mov b6=r18			// copy syscall entry-point to b6 (7 cyc)
-	addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r2	// compute base of memory stack
+	or r29=r8,r29				// A    construct cr.ipsr value to save
+	mov b6=r18				// I0   copy syscall entry-point to b6 (7 cyc)
+	addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r2 // A compute base of memory stack
 
-	mov r18=ar.bsp			// save (kernel) ar.bsp (12 cyc)
-	cmp.ne pKStk,pUStk=r0,r0	// set pKStk <- 0, pUStk <- 1
-	br.call.sptk.many b7=ia64_syscall_setup
+	mov r18=ar.bsp				// M2   save (kernel) ar.bsp (12 cyc)
+	cmp.ne pKStk,pUStk=r0,r0		// A    set pKStk <- 0, pUStk <- 1
+	br.call.sptk.many b7=ia64_syscall_setup	// B
 	;;
-	mov ar.rsc=0x3			// set eager mode, pl 0, little-endian, loadrs=0
-	mov rp=r14				// set the real return addr
+	mov ar.rsc=0x3				// M2   set eager mode, pl 0, LE, loadrs=0
+	mov rp=r14				// I0   set the real return addr
 	nop.i 0
 	;;
-	ssm psr.i
-	tbit.z p8,p0=r3,TIF_SYSCALL_TRACE
-(p10)	br.cond.spnt.many ia64_ret_from_syscall	// p10==true means out registers are more than 8
+	ssm psr.i				// M2   we're on kernel stacks now, reenable irqs
+	tbit.z p8,p0=r3,TIF_SYSCALL_TRACE	// I0
+(p10)	br.cond.spnt.many ia64_ret_from_syscall	// B    return if bad call-frame or r15 is a NaT
 
 	nop.m 0
-(p8)	br.call.sptk.many b6=b6		// ignore this return addr
-	br.cond.spnt ia64_trace_syscall
+(p8)	br.call.sptk.many b6=b6			// B    (ignore return address)
+	br.cond.spnt ia64_trace_syscall		// B
 END(fsys_bubble_down)
 
 	.rodata
-- 
cgit v1.2.3-70-g09d2


From e7e965fa1961a8ce32cbbb1bd436c655ad03973e Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:22:08 -0700
Subject: [IA64] use srlz.d instead of srlz.i in ia64_leave_kernel()

This patch switches the srlz.i in ia64_leave_kernel() to srlz.d.  As
per architecture manual, the former is needed only to ensure that the
clearing of PSR.IC is seen by the VHPT for subsequent instruction
fetches.  However, since the remainder of the code (up to and
including the RFI instruction) is mapped by a pinned TLB entry, there
is no chance of an iTLB miss and we don't care whether or not the VHPT
sees PSR.IC cleared.  Since srlz.d is substantially cheaper than
srlz.i, this should shave off a few cycles off the interrupt path
(unverified though; I'm not setup to measure this at the moment).

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index a7542c3d3b3..4517d4ab5ef 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -880,7 +880,7 @@ GLOBAL_ENTRY(ia64_leave_kernel)
 	ldf.fill f7=[r2],PT(F11)-PT(F7)
 	ldf.fill f8=[r3],32
 	;;
-	srlz.i			// ensure interruption collection is off
+	srlz.d	// ensure that inter. collection is off (VHPT is don't care, since text is pinned)
 	mov ar.ccv=r15
 	;;
 	ldf.fill f11=[r2]
-- 
cgit v1.2.3-70-g09d2


From 8e3e50168c8537807b7a6f78588cd72e21363262 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:22:40 -0700
Subject: [IA64] need r29=psr *after* rsm psr.i

Yanmin Zhang pointed out a sequence problem when saving the psr.  David
Mosberger provided this patch (which gave up a cycle).

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/gate.S | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/gate.S b/arch/ia64/kernel/gate.S
index 272e64c0e21..86948ce63e4 100644
--- a/arch/ia64/kernel/gate.S
+++ b/arch/ia64/kernel/gate.S
@@ -77,7 +77,7 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
 	epc					// B	causes split-issue
 }
 	;;
-	rsm psr.be				// M2 (5 cyc to srlz.d)
+	rsm psr.be | psr.i			// M2 (5 cyc to srlz.d)
 	LOAD_FSYSCALL_TABLE(r14)		// X
 	;;
 	mov r16=IA64_KR(CURRENT)		// M2 (12 cyc)
@@ -98,15 +98,14 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
 	nop.i 0
 	;;
 	nop.m 0
-(p6)	mov b7=r18				// I0
 (p6)	tbit.z.unc p8,p0=r18,0			// I0 (dual-issues with "mov b7=r18"!)
-
-	nop.m 0
 	nop.i 0
+	;;
+(p8)	ssm psr.i
+(p6)	mov b7=r18				// I0
 (p8)	br.dptk.many b7				// B
 
 	mov r27=ar.rsc				// M2 (12 cyc)
-(p6)	rsm psr.i				// M2
 /*
  * brl.cond doesn't work as intended because the linker would convert this branch
  * into a branch to a PLT.  Perhaps there will be a way to avoid this with some
@@ -123,7 +122,7 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
 #else
 	BRL_COND_FSYS_BUBBLE_DOWN(p6)
 #endif
-
+	ssm psr.i
 	mov r10=-1
 (p10)	mov r8=EINVAL
 (p9)	mov r8=ENOSYS
-- 
cgit v1.2.3-70-g09d2


From ebcc80c1b6629a445f7471cc1ddb48faf8a84e70 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Thu, 5 May 2005 06:40:00 -0700
Subject: [IA64] Merge audit fix for fsyscalls with syscall-optimizations

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/fsys.S | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/fsys.S b/arch/ia64/kernel/fsys.S
index d09a5b8a098..f566ff43a38 100644
--- a/arch/ia64/kernel/fsys.S
+++ b/arch/ia64/kernel/fsys.S
@@ -630,10 +630,10 @@ GLOBAL_ENTRY(fsys_bubble_down)
 	;;
 	mov ar.rsc=0x3				// M2   set eager mode, pl 0, LE, loadrs=0
 	mov rp=r14				// I0   set the real return addr
-	nop.i 0
+	and r3=_TIF_SYSCALL_TRACEAUDIT,r3	// A
 	;;
 	ssm psr.i				// M2   we're on kernel stacks now, reenable irqs
-	tbit.z p8,p0=r3,TIF_SYSCALL_TRACE	// I0
+	cmp.eq p8,p0=r3,r0			// A
 (p10)	br.cond.spnt.many ia64_ret_from_syscall	// B    return if bad call-frame or r15 is a NaT
 
 	nop.m 0
-- 
cgit v1.2.3-70-g09d2


From 7f9eaedf894dbaa08c157832e9a6c9c03ffed1ed Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Tue, 10 May 2005 12:49:00 -0700
Subject: [IA64] Fix convert_to_non_syscall() so gdb inferior calls work again

Fix convert_to_non_syscall() so it arranges for the kernel to be left
via ia64_leave_kernel() rather than ia64_leave_syscall().  The latter
no longer tolerates being called with pSys=0 and pNonSys=1.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/ptrace.c | 22 +++++++++++++++++++++-
 1 file changed, 21 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/ptrace.c b/arch/ia64/kernel/ptrace.c
index c253fd5914f..9e730c7bf0c 100644
--- a/arch/ia64/kernel/ptrace.c
+++ b/arch/ia64/kernel/ptrace.c
@@ -705,12 +705,32 @@ convert_to_non_syscall (struct task_struct *child, struct pt_regs  *pt,
 			break;
 	}
 
+	/*
+	 * Note: at the time of this call, the target task is blocked
+	 * in notify_resume_user() and by clearling PRED_LEAVE_SYSCALL
+	 * (aka, "pLvSys") we redirect execution from
+	 * .work_pending_syscall_end to .work_processed_kernel.
+	 */
 	unw_get_pr(&prev_info, &pr);
-	pr &= ~(1UL << PRED_SYSCALL);
+	pr &= ~((1UL << PRED_SYSCALL) | (1UL << PRED_LEAVE_SYSCALL));
 	pr |=  (1UL << PRED_NON_SYSCALL);
 	unw_set_pr(&prev_info, pr);
 
 	pt->cr_ifs = (1UL << 63) | cfm;
+	/*
+	 * Clear the memory that is NOT written on syscall-entry to
+	 * ensure we do not leak kernel-state to user when execution
+	 * resumes.
+	 */
+	pt->r2 = 0;
+	pt->r3 = 0;
+	pt->r14 = 0;
+	memset(&pt->r16, 0, 16*8);	/* clear r16-r31 */
+	memset(&pt->f6, 0, 6*16);	/* clear f6-f11 */
+	pt->b7 = 0;
+	pt->ar_ccv = 0;
+	pt->ar_csd = 0;
+	pt->ar_ssd = 0;
 }
 
 static int
-- 
cgit v1.2.3-70-g09d2


From 0908db22b189b28664cba3965ebb7e0df59c749a Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Sun, 19 Jun 2005 19:48:16 +0100
Subject: [PATCH] ARM SMP: Messages about CPUs should be prefixed by CPU%u

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/smp.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 45ed036336e..34892758f09 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -145,7 +145,8 @@ int __init __cpu_up(unsigned int cpu)
 	pgd_free(pgd);
 
 	if (ret) {
-		printk(KERN_CRIT "cpu_up: processor %d failed to boot\n", cpu);
+		printk(KERN_CRIT "CPU%u: processor failed to boot\n", cpu);
+
 		/*
 		 * FIXME: We need to clean up the new idle thread. --rmk
 		 */
-- 
cgit v1.2.3-70-g09d2


From 8830f04a092b47f3d246271b24685cd9eab82027 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 20 Jun 2005 09:51:03 +0100
Subject: [PATCH] ARM: Fix delayed dcache flush for ARMv6 non-aliasing caches

flush_dcache_page() did nothing for these caches, but since they
suffer from I/D cache coherency issues, we need to ensure that data
is written back to RAM.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/fault-armv.c | 31 +++++++++++--------------------
 arch/arm/mm/flush.c      | 44 ++++++++++++++++++--------------------------
 2 files changed, 29 insertions(+), 46 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/mm/fault-armv.c b/arch/arm/mm/fault-armv.c
index 01967ddeef5..be4ab3d73c9 100644
--- a/arch/arm/mm/fault-armv.c
+++ b/arch/arm/mm/fault-armv.c
@@ -77,9 +77,8 @@ no_pmd:
 }
 
 static void
-make_coherent(struct vm_area_struct *vma, unsigned long addr, struct page *page, int dirty)
+make_coherent(struct address_space *mapping, struct vm_area_struct *vma, unsigned long addr, unsigned long pfn)
 {
-	struct address_space *mapping = page_mapping(page);
 	struct mm_struct *mm = vma->vm_mm;
 	struct vm_area_struct *mpnt;
 	struct prio_tree_iter iter;
@@ -87,9 +86,6 @@ make_coherent(struct vm_area_struct *vma, unsigned long addr, struct page *page,
 	pgoff_t pgoff;
 	int aliases = 0;
 
-	if (!mapping)
-		return;
-
 	pgoff = vma->vm_pgoff + ((addr - vma->vm_start) >> PAGE_SHIFT);
 
 	/*
@@ -115,9 +111,11 @@ make_coherent(struct vm_area_struct *vma, unsigned long addr, struct page *page,
 	if (aliases)
 		adjust_pte(vma, addr);
 	else
-		flush_cache_page(vma, addr, page_to_pfn(page));
+		flush_cache_page(vma, addr, pfn);
 }
 
+void __flush_dcache_page(struct address_space *mapping, struct page *page);
+
 /*
  * Take care of architecture specific things when placing a new PTE into
  * a page table, or changing an existing PTE.  Basically, there are two
@@ -134,29 +132,22 @@ make_coherent(struct vm_area_struct *vma, unsigned long addr, struct page *page,
 void update_mmu_cache(struct vm_area_struct *vma, unsigned long addr, pte_t pte)
 {
 	unsigned long pfn = pte_pfn(pte);
+	struct address_space *mapping;
 	struct page *page;
 
 	if (!pfn_valid(pfn))
 		return;
+
 	page = pfn_to_page(pfn);
-	if (page_mapping(page)) {
+	mapping = page_mapping(page);
+	if (mapping) {
 		int dirty = test_and_clear_bit(PG_dcache_dirty, &page->flags);
 
-		if (dirty) {
-			/*
-			 * This is our first userspace mapping of this page.
-			 * Ensure that the physical page is coherent with
-			 * the kernel mapping.
-			 *
-			 * FIXME: only need to do this on VIVT and aliasing
-			 *        VIPT cache architectures.  We can do that
-			 *	  by choosing whether to set this bit...
-			 */
-			__cpuc_flush_dcache_page(page_address(page));
-		}
+		if (dirty)
+			__flush_dcache_page(mapping, page);
 
 		if (cache_is_vivt())
-			make_coherent(vma, addr, page, dirty);
+			make_coherent(mapping, vma, addr, pfn);
 	}
 }
 
diff --git a/arch/arm/mm/flush.c b/arch/arm/mm/flush.c
index 4085ed983e4..191788fb18d 100644
--- a/arch/arm/mm/flush.c
+++ b/arch/arm/mm/flush.c
@@ -37,13 +37,8 @@ static void flush_pfn_alias(unsigned long pfn, unsigned long vaddr)
 #define flush_pfn_alias(pfn,vaddr)	do { } while (0)
 #endif
 
-static void __flush_dcache_page(struct address_space *mapping, struct page *page)
+void __flush_dcache_page(struct address_space *mapping, struct page *page)
 {
-	struct mm_struct *mm = current->active_mm;
-	struct vm_area_struct *mpnt;
-	struct prio_tree_iter iter;
-	pgoff_t pgoff;
-
 	/*
 	 * Writeback any data associated with the kernel mapping of this
 	 * page.  This ensures that data in the physical page is mutually
@@ -52,24 +47,21 @@ static void __flush_dcache_page(struct address_space *mapping, struct page *page
 	__cpuc_flush_dcache_page(page_address(page));
 
 	/*
-	 * If there's no mapping pointer here, then this page isn't
-	 * visible to userspace yet, so there are no cache lines
-	 * associated with any other aliases.
-	 */
-	if (!mapping)
-		return;
-
-	/*
-	 * This is a page cache page.  If we have a VIPT cache, we
-	 * only need to do one flush - which would be at the relevant
+	 * If this is a page cache page, and we have an aliasing VIPT cache,
+	 * we only need to do one flush - which would be at the relevant
 	 * userspace colour, which is congruent with page->index.
 	 */
-	if (cache_is_vipt()) {
-		if (cache_is_vipt_aliasing())
-			flush_pfn_alias(page_to_pfn(page),
-					page->index << PAGE_CACHE_SHIFT);
-		return;
-	}
+	if (mapping && cache_is_vipt_aliasing())
+		flush_pfn_alias(page_to_pfn(page),
+				page->index << PAGE_CACHE_SHIFT);
+}
+
+static void __flush_dcache_aliases(struct address_space *mapping, struct page *page)
+{
+	struct mm_struct *mm = current->active_mm;
+	struct vm_area_struct *mpnt;
+	struct prio_tree_iter iter;
+	pgoff_t pgoff;
 
 	/*
 	 * There are possible user space mappings of this page:
@@ -116,12 +108,12 @@ void flush_dcache_page(struct page *page)
 {
 	struct address_space *mapping = page_mapping(page);
 
-	if (cache_is_vipt_nonaliasing())
-		return;
-
 	if (mapping && !mapping_mapped(mapping))
 		set_bit(PG_dcache_dirty, &page->flags);
-	else
+	else {
 		__flush_dcache_page(mapping, page);
+		if (mapping && cache_is_vivt())
+			__flush_dcache_aliases(mapping, page);
+	}
 }
 EXPORT_SYMBOL(flush_dcache_page);
-- 
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 'arch')

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 5abc100e882a979f8f5be8184938305b3fb2df2b Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 20 Jun 2005 12:31:14 +0100
Subject: [PATCH] ARM: Ensure DMA-bounced buffers are properly written to RAM

When DMA bounce buffers were unmapped and the data was memcpy'd to
the original buffer, we were not ensuring that the data was written
to RAM.  This means that there was the potential for page cache
pages to have different cache states depending whether they've been
bounced or not.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/common/dmabounce.c | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/common/dmabounce.c b/arch/arm/common/dmabounce.c
index 5797b1b100a..39dd33ff6f3 100644
--- a/arch/arm/common/dmabounce.c
+++ b/arch/arm/common/dmabounce.c
@@ -302,12 +302,24 @@ unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
 
 		DO_STATS ( device_info->bounce_count++ );
 
-		if ((dir == DMA_FROM_DEVICE) ||
-		    (dir == DMA_BIDIRECTIONAL)) {
+		if (dir == DMA_FROM_DEVICE || dir == DMA_BIDIRECTIONAL) {
+			unsigned long ptr;
+
 			dev_dbg(dev,
 				"%s: copy back safe %p to unsafe %p size %d\n",
 				__func__, buf->safe, buf->ptr, size);
 			memcpy(buf->ptr, buf->safe, size);
+
+			/*
+			 * DMA buffers must have the same cache properties
+			 * as if they were really used for DMA - which means
+			 * data must be written back to RAM.  Note that
+			 * we don't use dmac_flush_range() here for the
+			 * bidirectional case because we know the cache
+			 * lines will be coherent with the data written.
+			 */
+			ptr = (unsigned long)buf->ptr;
+			dmac_clean_range(ptr, ptr + size);
 		}
 		free_safe_buffer(device_info, buf);
 	}
-- 
cgit v1.2.3-70-g09d2


From ad21798e0ecc3b54720a2420b341f51dfb254706 Mon Sep 17 00:00:00 2001
From: Anton Blanchard <anton@samba.org>
Date: Mon, 20 Jun 2005 21:43:07 +1000
Subject: [PATCH] ppc64: quieten RTAS printks

Some rtasd printks were too loud. They would appear on a quiet boot
even though they were only informational.

Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/rtasd.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/rtasd.c b/arch/ppc64/kernel/rtasd.c
index ff65dc33320..b0c3b829fe4 100644
--- a/arch/ppc64/kernel/rtasd.c
+++ b/arch/ppc64/kernel/rtasd.c
@@ -440,7 +440,7 @@ static int rtasd(void *unused)
 		goto error;
 	}
 
-	printk(KERN_ERR "RTAS daemon started\n");
+	printk(KERN_INFO "RTAS daemon started\n");
 
 	DEBUG("will sleep for %d jiffies\n", (HZ*60/rtas_event_scan_rate) / 2);
 
@@ -485,7 +485,7 @@ static int __init rtas_init(void)
 	/* No RTAS, only warn if we are on a pSeries box  */
 	if (rtas_token("event-scan") == RTAS_UNKNOWN_SERVICE) {
 		if (systemcfg->platform & PLATFORM_PSERIES)
-			printk(KERN_ERR "rtasd: no event-scan on system\n");
+			printk(KERN_INFO "rtasd: no event-scan on system\n");
 		return 1;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 0231c290d817513d448963520a97eadd51a370b7 Mon Sep 17 00:00:00 2001
From: Anton Blanchard <anton@samba.org>
Date: Mon, 20 Jun 2005 21:43:15 +1000
Subject: [PATCH] ppc64: use cpu_has_feature macro

Use the new cpu_has_feature macros instead of open coding it.

Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/pSeries_smp.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/pSeries_smp.c b/arch/ppc64/kernel/pSeries_smp.c
index fbad349ec58..4203bd020c8 100644
--- a/arch/ppc64/kernel/pSeries_smp.c
+++ b/arch/ppc64/kernel/pSeries_smp.c
@@ -375,7 +375,7 @@ static int smp_pSeries_cpu_bootable(unsigned int nr)
 	 * cpus are assumed to be secondary threads.
 	 */
 	if (system_state < SYSTEM_RUNNING &&
-	    cur_cpu_spec->cpu_features & CPU_FTR_SMT &&
+	    cpu_has_feature(CPU_FTR_SMT) &&
 	    !smt_enabled_at_boot && nr % 2 != 0)
 		return 0;
 
@@ -419,8 +419,8 @@ void __init smp_init_pSeries(void)
 #endif
 
 	/* Mark threads which are still spinning in hold loops. */
-	if (cur_cpu_spec->cpu_features & CPU_FTR_SMT)
-		for_each_present_cpu(i) {
+	if (cpu_has_feature(CPU_FTR_SMT)) {
+		for_each_present_cpu(i) { 
 			if (i % 2 == 0)
 				/*
 				 * Even-numbered logical cpus correspond to
@@ -428,8 +428,9 @@ void __init smp_init_pSeries(void)
 				 */
 				cpu_set(i, of_spin_map);
 		}
-	else
+	} else {
 		of_spin_map = cpu_present_map;
+	}
 
 	cpu_clear(boot_cpuid, of_spin_map);
 
-- 
cgit v1.2.3-70-g09d2


From d3588ba9bbade7f1fde59726c9af43eeeb08e092 Mon Sep 17 00:00:00 2001
From: John Rose <johnrose@austin.ibm.com>
Date: Mon, 20 Jun 2005 21:43:48 +1000
Subject: [PATCH] initialize TCE tables

A fairly recent platform requirement states that the OS must clear the
whole TCE table at setup time, in case firmware left any active
mappings in it.  Without this initialization, dynamic bus removes can
fail.  Firmware rejects these requests if active mappings still exist
for a slot that has been deallocated by the OS.

Signed-off-by: John Rose <johnrose@austin.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/iommu.c | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/iommu.c b/arch/ppc64/kernel/iommu.c
index 344164681d2..8316426ccaf 100644
--- a/arch/ppc64/kernel/iommu.c
+++ b/arch/ppc64/kernel/iommu.c
@@ -423,6 +423,9 @@ struct iommu_table *iommu_init_table(struct iommu_table *tbl)
 	tbl->it_largehint = tbl->it_halfpoint;
 	spin_lock_init(&tbl->it_lock);
 
+	/* Clear the hardware table in case firmware left allocations in it */
+	ppc_md.tce_free(tbl, tbl->it_offset, tbl->it_size);
+
 	if (!welcomed) {
 		printk(KERN_INFO "IOMMU table initialized, virtual merging %s\n",
 		       novmerge ? "disabled" : "enabled");
-- 
cgit v1.2.3-70-g09d2


From 34c8eacab670e578a2aaafdf1061efd214b2f639 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 20 Jun 2005 12:56:40 +0100
Subject: [PATCH] ARM: Remove obsolete arch/arm/kernel/arch.c

This is not used anymore - RiscPC now contains the necessary
supporting code.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/Makefile |  2 +-
 arch/arm/kernel/arch.c   | 46 ----------------------------------------------
 2 files changed, 1 insertion(+), 47 deletions(-)
 delete mode 100644 arch/arm/kernel/arch.c

(limited to 'arch')

diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile
index 4a2af55e134..3e1b0327e4d 100644
--- a/arch/arm/kernel/Makefile
+++ b/arch/arm/kernel/Makefile
@@ -6,7 +6,7 @@ AFLAGS_head.o := -DTEXTADDR=$(TEXTADDR) -DDATAADDR=$(DATAADDR)
 
 # Object file lists.
 
-obj-y		:= arch.o compat.o dma.o entry-armv.o entry-common.o irq.o   \
+obj-y		:= compat.o dma.o entry-armv.o entry-common.o irq.o \
 		   process.o ptrace.o semaphore.o setup.o signal.o sys_arm.o \
 		   time.o traps.o
 
diff --git a/arch/arm/kernel/arch.c b/arch/arm/kernel/arch.c
deleted file mode 100644
index 4e02fbeb10a..00000000000
--- a/arch/arm/kernel/arch.c
+++ /dev/null
@@ -1,46 +0,0 @@
-/*
- *  linux/arch/arm/kernel/arch.c
- *
- *  Architecture specific fixups.
- */
-#include <linux/config.h>
-#include <linux/init.h>
-#include <linux/types.h>
-
-#include <asm/elf.h>
-#include <asm/page.h>
-#include <asm/setup.h>
-#include <asm/mach/arch.h>
-
-unsigned int vram_size;
-
-#ifdef CONFIG_ARCH_ACORN
-
-unsigned int memc_ctrl_reg;
-unsigned int number_mfm_drives;
-
-static int __init parse_tag_acorn(const struct tag *tag)
-{
-	memc_ctrl_reg = tag->u.acorn.memc_control_reg;
-	number_mfm_drives = tag->u.acorn.adfsdrives;
-
-	switch (tag->u.acorn.vram_pages) {
-	case 512:
-		vram_size += PAGE_SIZE * 256;
-	case 256:
-		vram_size += PAGE_SIZE * 256;
-	default:
-		break;
-	}
-#if 0
-	if (vram_size) {
-		desc->video_start = 0x02000000;
-		desc->video_end   = 0x02000000 + vram_size;
-	}
-#endif
-	return 0;
-}
-
-__tagtable(ATAG_ACORN, parse_tag_acorn);
-
-#endif
-- 
cgit v1.2.3-70-g09d2


From f29481c0e7e55efc25598c1a6c503015cfe45245 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 20 Jun 2005 15:49:59 +0100
Subject: [PATCH] ARM: Remove gcc type-isms from GCC helper functions

Convert ugly GCC types to Linux types:

	UQImode -> u8
	SImode -> s32
	USImode -> u32
	DImode -> s64
	UDImode -> u64
	word_type -> int

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/lib/ashldi3.c  | 16 ++++++------
 arch/arm/lib/ashrdi3.c  | 14 +++++------
 arch/arm/lib/gcclib.h   | 17 +++++--------
 arch/arm/lib/longlong.h | 66 ++++++++++++++++++++++++-------------------------
 arch/arm/lib/lshrdi3.c  | 16 ++++++------
 arch/arm/lib/muldi3.c   | 18 +++++++-------
 arch/arm/lib/ucmpdi2.c  | 12 ++++-----
 arch/arm/lib/udivdi3.c  | 26 +++++++++----------
 8 files changed, 90 insertions(+), 95 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/lib/ashldi3.c b/arch/arm/lib/ashldi3.c
index 130f5a83966..02d2b628ebc 100644
--- a/arch/arm/lib/ashldi3.c
+++ b/arch/arm/lib/ashldi3.c
@@ -31,11 +31,11 @@ Boston, MA 02111-1307, USA.  */
 
 #include "gcclib.h"
 
-DItype
-__ashldi3 (DItype u, word_type b)
+s64
+__ashldi3 (s64 u, int b)
 {
   DIunion w;
-  word_type bm;
+  int bm;
   DIunion uu;
 
   if (b == 0)
@@ -43,17 +43,17 @@ __ashldi3 (DItype u, word_type b)
 
   uu.ll = u;
 
-  bm = (sizeof (SItype) * BITS_PER_UNIT) - b;
+  bm = (sizeof (s32) * BITS_PER_UNIT) - b;
   if (bm <= 0)
     {
       w.s.low = 0;
-      w.s.high = (USItype)uu.s.low << -bm;
+      w.s.high = (u32)uu.s.low << -bm;
     }
   else
     {
-      USItype carries = (USItype)uu.s.low >> bm;
-      w.s.low = (USItype)uu.s.low << b;
-      w.s.high = ((USItype)uu.s.high << b) | carries;
+      u32 carries = (u32)uu.s.low >> bm;
+      w.s.low = (u32)uu.s.low << b;
+      w.s.high = ((u32)uu.s.high << b) | carries;
     }
 
   return w.ll;
diff --git a/arch/arm/lib/ashrdi3.c b/arch/arm/lib/ashrdi3.c
index 71625d218f8..89f6fb7ed8f 100644
--- a/arch/arm/lib/ashrdi3.c
+++ b/arch/arm/lib/ashrdi3.c
@@ -31,11 +31,11 @@ Boston, MA 02111-1307, USA.  */
 
 #include "gcclib.h"
 
-DItype
-__ashrdi3 (DItype u, word_type b)
+s64
+__ashrdi3 (s64 u, int b)
 {
   DIunion w;
-  word_type bm;
+  int bm;
   DIunion uu;
 
   if (b == 0)
@@ -43,18 +43,18 @@ __ashrdi3 (DItype u, word_type b)
 
   uu.ll = u;
 
-  bm = (sizeof (SItype) * BITS_PER_UNIT) - b;
+  bm = (sizeof (s32) * BITS_PER_UNIT) - b;
   if (bm <= 0)
     {
       /* w.s.high = 1..1 or 0..0 */
-      w.s.high = uu.s.high >> (sizeof (SItype) * BITS_PER_UNIT - 1);
+      w.s.high = uu.s.high >> (sizeof (s32) * BITS_PER_UNIT - 1);
       w.s.low = uu.s.high >> -bm;
     }
   else
     {
-      USItype carries = (USItype)uu.s.high << bm;
+      u32 carries = (u32)uu.s.high << bm;
       w.s.high = uu.s.high >> b;
-      w.s.low = ((USItype)uu.s.low >> b) | carries;
+      w.s.low = ((u32)uu.s.low >> b) | carries;
     }
 
   return w.ll;
diff --git a/arch/arm/lib/gcclib.h b/arch/arm/lib/gcclib.h
index 65314a3d9e2..283f5294cc3 100644
--- a/arch/arm/lib/gcclib.h
+++ b/arch/arm/lib/gcclib.h
@@ -1,25 +1,20 @@
 /* gcclib.h -- definitions for various functions 'borrowed' from gcc-2.95.3 */
 /* I Molton     29/07/01 */
 
-#define BITS_PER_UNIT  8
-#define SI_TYPE_SIZE (sizeof (SItype) * BITS_PER_UNIT)
+#include <linux/types.h>
 
-typedef unsigned int UQItype    __attribute__ ((mode (QI)));
-typedef          int SItype     __attribute__ ((mode (SI)));
-typedef unsigned int USItype    __attribute__ ((mode (SI)));
-typedef          int DItype     __attribute__ ((mode (DI)));
-typedef          int word_type 	__attribute__ ((mode (__word__)));
-typedef unsigned int UDItype    __attribute__ ((mode (DI)));
+#define BITS_PER_UNIT  8
+#define SI_TYPE_SIZE (sizeof(s32) * BITS_PER_UNIT)
 
 #ifdef __ARMEB__
-  struct DIstruct {SItype high, low;};
+  struct DIstruct {s32 high, low;};
 #else
-  struct DIstruct {SItype low, high;};
+  struct DIstruct {s32 low, high;};
 #endif
 
 typedef union
 {
   struct DIstruct s;
-  DItype ll;
+  s64 ll;
 } DIunion;
 
diff --git a/arch/arm/lib/longlong.h b/arch/arm/lib/longlong.h
index 179eea4edc3..3ef76957481 100644
--- a/arch/arm/lib/longlong.h
+++ b/arch/arm/lib/longlong.h
@@ -26,18 +26,18 @@
 
 #define __BITS4 (SI_TYPE_SIZE / 4)
 #define __ll_B (1L << (SI_TYPE_SIZE / 2))
-#define __ll_lowpart(t) ((USItype) (t) % __ll_B)
-#define __ll_highpart(t) ((USItype) (t) / __ll_B)
+#define __ll_lowpart(t) ((u32) (t) % __ll_B)
+#define __ll_highpart(t) ((u32) (t) / __ll_B)
 
 /* Define auxiliary asm macros.
 
    1) umul_ppmm(high_prod, low_prod, multipler, multiplicand)
-   multiplies two USItype integers MULTIPLER and MULTIPLICAND,
-   and generates a two-part USItype product in HIGH_PROD and
+   multiplies two u32 integers MULTIPLER and MULTIPLICAND,
+   and generates a two-part u32 product in HIGH_PROD and
    LOW_PROD.
 
-   2) __umulsidi3(a,b) multiplies two USItype integers A and B,
-   and returns a UDItype product.  This is just a variant of umul_ppmm.
+   2) __umulsidi3(a,b) multiplies two u32 integers A and B,
+   and returns a u64 product.  This is just a variant of umul_ppmm.
 
    3) udiv_qrnnd(quotient, remainder, high_numerator, low_numerator,
    denominator) divides a two-word unsigned integer, composed by the
@@ -77,23 +77,23 @@
 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \
   __asm__ ("adds	%1, %4, %5					\n\
 	adc	%0, %2, %3"						\
-	   : "=r" ((USItype) (sh)),					\
-	     "=&r" ((USItype) (sl))					\
-	   : "%r" ((USItype) (ah)),					\
-	     "rI" ((USItype) (bh)),					\
-	     "%r" ((USItype) (al)),					\
-	     "rI" ((USItype) (bl)))
+	   : "=r" ((u32) (sh)),					\
+	     "=&r" ((u32) (sl))					\
+	   : "%r" ((u32) (ah)),					\
+	     "rI" ((u32) (bh)),					\
+	     "%r" ((u32) (al)),					\
+	     "rI" ((u32) (bl)))
 #define sub_ddmmss(sh, sl, ah, al, bh, bl) \
   __asm__ ("subs	%1, %4, %5					\n\
 	sbc	%0, %2, %3"						\
-	   : "=r" ((USItype) (sh)),					\
-	     "=&r" ((USItype) (sl))					\
-	   : "r" ((USItype) (ah)),					\
-	     "rI" ((USItype) (bh)),					\
-	     "r" ((USItype) (al)),					\
-	     "rI" ((USItype) (bl)))
+	   : "=r" ((u32) (sh)),					\
+	     "=&r" ((u32) (sl))					\
+	   : "r" ((u32) (ah)),					\
+	     "rI" ((u32) (bh)),					\
+	     "r" ((u32) (al)),					\
+	     "rI" ((u32) (bl)))
 #define umul_ppmm(xh, xl, a, b) \
-{register USItype __t0, __t1, __t2;					\
+{register u32 __t0, __t1, __t2;					\
   __asm__ ("%@ Inlined umul_ppmm					\n\
 	mov	%2, %5, lsr #16						\n\
 	mov	%0, %6, lsr #16						\n\
@@ -107,11 +107,11 @@
 	addcs	%0, %0, #65536						\n\
 	adds	%1, %1, %3, lsl #16					\n\
 	adc	%0, %0, %3, lsr #16"					\
-	   : "=&r" ((USItype) (xh)),					\
-	     "=r" ((USItype) (xl)),					\
+	   : "=&r" ((u32) (xh)),					\
+	     "=r" ((u32) (xl)),					\
 	     "=&r" (__t0), "=&r" (__t1), "=r" (__t2)			\
-	   : "r" ((USItype) (a)),					\
-	     "r" ((USItype) (b)));}
+	   : "r" ((u32) (a)),					\
+	     "r" ((u32) (b)));}
 #define UMUL_TIME 20
 #define UDIV_TIME 100
 #endif /* __arm__ */
@@ -123,14 +123,14 @@
 
 #define __udiv_qrnnd_c(q, r, n1, n0, d) \
   do {									\
-    USItype __d1, __d0, __q1, __q0;					\
-    USItype __r1, __r0, __m;						\
+    u32 __d1, __d0, __q1, __q0;					\
+    u32 __r1, __r0, __m;						\
     __d1 = __ll_highpart (d);						\
     __d0 = __ll_lowpart (d);						\
 									\
     __r1 = (n1) % __d1;							\
     __q1 = (n1) / __d1;							\
-    __m = (USItype) __q1 * __d0;					\
+    __m = (u32) __q1 * __d0;					\
     __r1 = __r1 * __ll_B | __ll_highpart (n0);				\
     if (__r1 < __m)							\
       {									\
@@ -143,7 +143,7 @@
 									\
     __r0 = __r1 % __d1;							\
     __q0 = __r1 / __d1;							\
-    __m = (USItype) __q0 * __d0;					\
+    __m = (u32) __q0 * __d0;					\
     __r0 = __r0 * __ll_B | __ll_lowpart (n0);				\
     if (__r0 < __m)							\
       {									\
@@ -154,7 +154,7 @@
       }									\
     __r0 -= __m;							\
 									\
-    (q) = (USItype) __q1 * __ll_B | __q0;				\
+    (q) = (u32) __q1 * __ll_B | __q0;				\
     (r) = __r0;								\
   } while (0)
 
@@ -163,14 +163,14 @@
 
 #define count_leading_zeros(count, x) \
   do {									\
-    USItype __xr = (x);							\
-    USItype __a;							\
+    u32 __xr = (x);							\
+    u32 __a;							\
 									\
     if (SI_TYPE_SIZE <= 32)						\
       {									\
-	__a = __xr < ((USItype)1<<2*__BITS4)				\
-	  ? (__xr < ((USItype)1<<__BITS4) ? 0 : __BITS4)		\
-	  : (__xr < ((USItype)1<<3*__BITS4) ?  2*__BITS4 : 3*__BITS4);	\
+	__a = __xr < ((u32)1<<2*__BITS4)				\
+	  ? (__xr < ((u32)1<<__BITS4) ? 0 : __BITS4)		\
+	  : (__xr < ((u32)1<<3*__BITS4) ?  2*__BITS4 : 3*__BITS4);	\
       }									\
     else								\
       {									\
diff --git a/arch/arm/lib/lshrdi3.c b/arch/arm/lib/lshrdi3.c
index b666f1bad45..5c2385acdec 100644
--- a/arch/arm/lib/lshrdi3.c
+++ b/arch/arm/lib/lshrdi3.c
@@ -31,11 +31,11 @@ Boston, MA 02111-1307, USA.  */
 
 #include "gcclib.h"
 
-DItype
-__lshrdi3 (DItype u, word_type b)
+s64
+__lshrdi3 (s64 u, int b)
 {
   DIunion w;
-  word_type bm;
+  int bm;
   DIunion uu;
 
   if (b == 0)
@@ -43,17 +43,17 @@ __lshrdi3 (DItype u, word_type b)
 
   uu.ll = u;
 
-  bm = (sizeof (SItype) * BITS_PER_UNIT) - b;
+  bm = (sizeof (s32) * BITS_PER_UNIT) - b;
   if (bm <= 0)
     {
       w.s.high = 0;
-      w.s.low = (USItype)uu.s.high >> -bm;
+      w.s.low = (u32)uu.s.high >> -bm;
     }
   else
     {
-      USItype carries = (USItype)uu.s.high << bm;
-      w.s.high = (USItype)uu.s.high >> b;
-      w.s.low = ((USItype)uu.s.low >> b) | carries;
+      u32 carries = (u32)uu.s.high << bm;
+      w.s.high = (u32)uu.s.high >> b;
+      w.s.low = ((u32)uu.s.low >> b) | carries;
     }
 
   return w.ll;
diff --git a/arch/arm/lib/muldi3.c b/arch/arm/lib/muldi3.c
index 44d611b1cfd..5b649fa9e2f 100644
--- a/arch/arm/lib/muldi3.c
+++ b/arch/arm/lib/muldi3.c
@@ -32,7 +32,7 @@ Boston, MA 02111-1307, USA.  */
 #include "gcclib.h"
 
 #define umul_ppmm(xh, xl, a, b) \
-{register USItype __t0, __t1, __t2;                                     \
+{register u32 __t0, __t1, __t2;                                     \
   __asm__ ("%@ Inlined umul_ppmm					\n\
         mov     %2, %5, lsr #16						\n\
         mov     %0, %6, lsr #16						\n\
@@ -46,11 +46,11 @@ Boston, MA 02111-1307, USA.  */
         addcs   %0, %0, #65536						\n\
         adds    %1, %1, %3, lsl #16					\n\
         adc     %0, %0, %3, lsr #16"                                    \
-           : "=&r" ((USItype) (xh)),                                    \
-             "=r" ((USItype) (xl)),                                     \
+           : "=&r" ((u32) (xh)),                                    \
+             "=r" ((u32) (xl)),                                     \
              "=&r" (__t0), "=&r" (__t1), "=r" (__t2)                    \
-           : "r" ((USItype) (a)),                                       \
-             "r" ((USItype) (b)));}
+           : "r" ((u32) (a)),                                       \
+             "r" ((u32) (b)));}
 
 
 #define __umulsidi3(u, v) \
@@ -59,8 +59,8 @@ Boston, MA 02111-1307, USA.  */
     __w.ll; })
 
 
-DItype
-__muldi3 (DItype u, DItype v)
+s64
+__muldi3 (s64 u, s64 v)
 {
   DIunion w;
   DIunion uu, vv;
@@ -69,8 +69,8 @@ __muldi3 (DItype u, DItype v)
   vv.ll = v;
 
   w.ll = __umulsidi3 (uu.s.low, vv.s.low);
-  w.s.high += ((USItype) uu.s.low * (USItype) vv.s.high
-               + (USItype) uu.s.high * (USItype) vv.s.low);
+  w.s.high += ((u32) uu.s.low * (u32) vv.s.high
+               + (u32) uu.s.high * (u32) vv.s.low);
 
   return w.ll;
 }
diff --git a/arch/arm/lib/ucmpdi2.c b/arch/arm/lib/ucmpdi2.c
index 6c6ae63efa0..3eb438c065b 100644
--- a/arch/arm/lib/ucmpdi2.c
+++ b/arch/arm/lib/ucmpdi2.c
@@ -31,20 +31,20 @@ Boston, MA 02111-1307, USA.  */
 
 #include "gcclib.h"
 
-word_type
-__ucmpdi2 (DItype a, DItype b)
+int
+__ucmpdi2 (s64 a, s64 b)
 {
   DIunion au, bu;
 
   au.ll = a, bu.ll = b;
 
-  if ((USItype) au.s.high < (USItype) bu.s.high)
+  if ((u32) au.s.high < (u32) bu.s.high)
     return 0;
-  else if ((USItype) au.s.high > (USItype) bu.s.high)
+  else if ((u32) au.s.high > (u32) bu.s.high)
     return 2;
-  if ((USItype) au.s.low < (USItype) bu.s.low)
+  if ((u32) au.s.low < (u32) bu.s.low)
     return 0;
-  else if ((USItype) au.s.low > (USItype) bu.s.low)
+  else if ((u32) au.s.low > (u32) bu.s.low)
     return 2;
   return 1;
 }
diff --git a/arch/arm/lib/udivdi3.c b/arch/arm/lib/udivdi3.c
index d25195f673f..df1d5ef62bc 100644
--- a/arch/arm/lib/udivdi3.c
+++ b/arch/arm/lib/udivdi3.c
@@ -32,7 +32,7 @@ Boston, MA 02111-1307, USA.  */
 #include "gcclib.h"
 #include "longlong.h"
 
-static const UQItype __clz_tab[] =
+static const u8 __clz_tab[] =
 {
   0,1,2,2,3,3,3,3,4,4,4,4,4,4,4,4,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,
   6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,
@@ -44,15 +44,15 @@ static const UQItype __clz_tab[] =
   8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
 };
 
-UDItype
-__udivmoddi4 (UDItype n, UDItype d, UDItype *rp)
+u64
+__udivmoddi4 (u64 n, u64 d, u64 *rp)
 {
   DIunion ww;
   DIunion nn, dd;
   DIunion rr;
-  USItype d0, d1, n0, n1, n2;
-  USItype q0, q1;
-  USItype b, bm;
+  u32 d0, d1, n0, n1, n2;
+  u32 q0, q1;
+  u32 b, bm;
 
   nn.ll = n;
   dd.ll = d;
@@ -185,7 +185,7 @@ __udivmoddi4 (UDItype n, UDItype d, UDItype *rp)
             }
           else
             {
-              USItype m1, m0;
+              u32 m1, m0;
               /* Normalize.  */
 
               b = SI_TYPE_SIZE - bm;
@@ -224,16 +224,16 @@ __udivmoddi4 (UDItype n, UDItype d, UDItype *rp)
   return ww.ll;
 }
 
-UDItype
-__udivdi3 (UDItype n, UDItype d)
+u64
+__udivdi3 (u64 n, u64 d)
 {
-  return __udivmoddi4 (n, d, (UDItype *) 0);
+  return __udivmoddi4 (n, d, (u64 *) 0);
 }
 
-UDItype
-__umoddi3 (UDItype u, UDItype v)
+u64
+__umoddi3 (u64 u, u64 v)
 {
-  UDItype w;
+  u64 w;
 
   (void) __udivmoddi4 (u ,v, &w);
 
-- 
cgit v1.2.3-70-g09d2


From 3ade2fe0fd0238d68938b8f5f73ebd0561d2d2e5 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 20 Jun 2005 16:45:32 +0100
Subject: [PATCH] ARM: Lindent GCC helper functions

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/lib/ashldi3.c  |  47 +++---
 arch/arm/lib/ashrdi3.c  |  48 +++----
 arch/arm/lib/gcclib.h   |  20 +--
 arch/arm/lib/longlong.h |   2 +-
 arch/arm/lib/lshrdi3.c  |  47 +++---
 arch/arm/lib/muldi3.c   |  21 ++-
 arch/arm/lib/ucmpdi2.c  |  30 ++--
 arch/arm/lib/udivdi3.c  | 372 +++++++++++++++++++++++-------------------------
 8 files changed, 274 insertions(+), 313 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/lib/ashldi3.c b/arch/arm/lib/ashldi3.c
index 02d2b628ebc..b62875cfd8f 100644
--- a/arch/arm/lib/ashldi3.c
+++ b/arch/arm/lib/ashldi3.c
@@ -31,31 +31,26 @@ Boston, MA 02111-1307, USA.  */
 
 #include "gcclib.h"
 
-s64
-__ashldi3 (s64 u, int b)
+s64 __ashldi3(s64 u, int b)
 {
-  DIunion w;
-  int bm;
-  DIunion uu;
-
-  if (b == 0)
-    return u;
-
-  uu.ll = u;
-
-  bm = (sizeof (s32) * BITS_PER_UNIT) - b;
-  if (bm <= 0)
-    {
-      w.s.low = 0;
-      w.s.high = (u32)uu.s.low << -bm;
-    }
-  else
-    {
-      u32 carries = (u32)uu.s.low >> bm;
-      w.s.low = (u32)uu.s.low << b;
-      w.s.high = ((u32)uu.s.high << b) | carries;
-    }
-
-  return w.ll;
+	DIunion w;
+	int bm;
+	DIunion uu;
+
+	if (b == 0)
+		return u;
+
+	uu.ll = u;
+
+	bm = (sizeof(s32) * BITS_PER_UNIT) - b;
+	if (bm <= 0) {
+		w.s.low = 0;
+		w.s.high = (u32) uu.s.low << -bm;
+	} else {
+		u32 carries = (u32) uu.s.low >> bm;
+		w.s.low = (u32) uu.s.low << b;
+		w.s.high = ((u32) uu.s.high << b) | carries;
+	}
+
+	return w.ll;
 }
-
diff --git a/arch/arm/lib/ashrdi3.c b/arch/arm/lib/ashrdi3.c
index 89f6fb7ed8f..9a8600a7543 100644
--- a/arch/arm/lib/ashrdi3.c
+++ b/arch/arm/lib/ashrdi3.c
@@ -31,31 +31,27 @@ Boston, MA 02111-1307, USA.  */
 
 #include "gcclib.h"
 
-s64
-__ashrdi3 (s64 u, int b)
+s64 __ashrdi3(s64 u, int b)
 {
-  DIunion w;
-  int bm;
-  DIunion uu;
-
-  if (b == 0)
-    return u;
-
-  uu.ll = u;
-
-  bm = (sizeof (s32) * BITS_PER_UNIT) - b;
-  if (bm <= 0)
-    {
-      /* w.s.high = 1..1 or 0..0 */
-      w.s.high = uu.s.high >> (sizeof (s32) * BITS_PER_UNIT - 1);
-      w.s.low = uu.s.high >> -bm;
-    }
-  else
-    {
-      u32 carries = (u32)uu.s.high << bm;
-      w.s.high = uu.s.high >> b;
-      w.s.low = ((u32)uu.s.low >> b) | carries;
-    }
-
-  return w.ll;
+	DIunion w;
+	int bm;
+	DIunion uu;
+
+	if (b == 0)
+		return u;
+
+	uu.ll = u;
+
+	bm = (sizeof(s32) * BITS_PER_UNIT) - b;
+	if (bm <= 0) {
+		/* w.s.high = 1..1 or 0..0 */
+		w.s.high = uu.s.high >> (sizeof(s32) * BITS_PER_UNIT - 1);
+		w.s.low = uu.s.high >> -bm;
+	} else {
+		u32 carries = (u32) uu.s.high << bm;
+		w.s.high = uu.s.high >> b;
+		w.s.low = ((u32) uu.s.low >> b) | carries;
+	}
+
+	return w.ll;
 }
diff --git a/arch/arm/lib/gcclib.h b/arch/arm/lib/gcclib.h
index 283f5294cc3..8b6dcc656de 100644
--- a/arch/arm/lib/gcclib.h
+++ b/arch/arm/lib/gcclib.h
@@ -3,18 +3,20 @@
 
 #include <linux/types.h>
 
-#define BITS_PER_UNIT  8
-#define SI_TYPE_SIZE (sizeof(s32) * BITS_PER_UNIT)
+#define BITS_PER_UNIT	8
+#define SI_TYPE_SIZE	(sizeof(s32) * BITS_PER_UNIT)
 
 #ifdef __ARMEB__
-  struct DIstruct {s32 high, low;};
+struct DIstruct {
+	s32 high, low;
+};
 #else
-  struct DIstruct {s32 low, high;};
+struct DIstruct {
+	s32 low, high;
+};
 #endif
 
-typedef union
-{
-  struct DIstruct s;
-  s64 ll;
+typedef union {
+	struct DIstruct s;
+	s64 ll;
 } DIunion;
-
diff --git a/arch/arm/lib/longlong.h b/arch/arm/lib/longlong.h
index 3ef76957481..90ae647e4d7 100644
--- a/arch/arm/lib/longlong.h
+++ b/arch/arm/lib/longlong.h
@@ -114,7 +114,7 @@
 	     "r" ((u32) (b)));}
 #define UMUL_TIME 20
 #define UDIV_TIME 100
-#endif /* __arm__ */
+#endif				/* __arm__ */
 
 #define __umulsidi3(u, v) \
   ({DIunion __w;							\
diff --git a/arch/arm/lib/lshrdi3.c b/arch/arm/lib/lshrdi3.c
index 5c2385acdec..3681f49d2b6 100644
--- a/arch/arm/lib/lshrdi3.c
+++ b/arch/arm/lib/lshrdi3.c
@@ -31,31 +31,26 @@ Boston, MA 02111-1307, USA.  */
 
 #include "gcclib.h"
 
-s64
-__lshrdi3 (s64 u, int b)
+s64 __lshrdi3(s64 u, int b)
 {
-  DIunion w;
-  int bm;
-  DIunion uu;
-
-  if (b == 0)
-    return u;
-
-  uu.ll = u;
-
-  bm = (sizeof (s32) * BITS_PER_UNIT) - b;
-  if (bm <= 0)
-    {
-      w.s.high = 0;
-      w.s.low = (u32)uu.s.high >> -bm;
-    }
-  else
-    {
-      u32 carries = (u32)uu.s.high << bm;
-      w.s.high = (u32)uu.s.high >> b;
-      w.s.low = ((u32)uu.s.low >> b) | carries;
-    }
-
-  return w.ll;
+	DIunion w;
+	int bm;
+	DIunion uu;
+
+	if (b == 0)
+		return u;
+
+	uu.ll = u;
+
+	bm = (sizeof(s32) * BITS_PER_UNIT) - b;
+	if (bm <= 0) {
+		w.s.high = 0;
+		w.s.low = (u32) uu.s.high >> -bm;
+	} else {
+		u32 carries = (u32) uu.s.high << bm;
+		w.s.high = (u32) uu.s.high >> b;
+		w.s.low = ((u32) uu.s.low >> b) | carries;
+	}
+
+	return w.ll;
 }
-
diff --git a/arch/arm/lib/muldi3.c b/arch/arm/lib/muldi3.c
index 5b649fa9e2f..0a3b93313f1 100644
--- a/arch/arm/lib/muldi3.c
+++ b/arch/arm/lib/muldi3.c
@@ -52,26 +52,21 @@ Boston, MA 02111-1307, USA.  */
            : "r" ((u32) (a)),                                       \
              "r" ((u32) (b)));}
 
-
 #define __umulsidi3(u, v) \
   ({DIunion __w;                                                        \
     umul_ppmm (__w.s.high, __w.s.low, u, v);                            \
     __w.ll; })
 
-
-s64
-__muldi3 (s64 u, s64 v)
+s64 __muldi3(s64 u, s64 v)
 {
-  DIunion w;
-  DIunion uu, vv;
+	DIunion w;
+	DIunion uu, vv;
 
-  uu.ll = u,
-  vv.ll = v;
+	uu.ll = u, vv.ll = v;
 
-  w.ll = __umulsidi3 (uu.s.low, vv.s.low);
-  w.s.high += ((u32) uu.s.low * (u32) vv.s.high
-               + (u32) uu.s.high * (u32) vv.s.low);
+	w.ll = __umulsidi3(uu.s.low, vv.s.low);
+	w.s.high += ((u32) uu.s.low * (u32) vv.s.high
+		     + (u32) uu.s.high * (u32) vv.s.low);
 
-  return w.ll;
+	return w.ll;
 }
-
diff --git a/arch/arm/lib/ucmpdi2.c b/arch/arm/lib/ucmpdi2.c
index 3eb438c065b..57f3f2df385 100644
--- a/arch/arm/lib/ucmpdi2.c
+++ b/arch/arm/lib/ucmpdi2.c
@@ -31,21 +31,19 @@ Boston, MA 02111-1307, USA.  */
 
 #include "gcclib.h"
 
-int
-__ucmpdi2 (s64 a, s64 b)
+int __ucmpdi2(s64 a, s64 b)
 {
-  DIunion au, bu;
-
-  au.ll = a, bu.ll = b;
-
-  if ((u32) au.s.high < (u32) bu.s.high)
-    return 0;
-  else if ((u32) au.s.high > (u32) bu.s.high)
-    return 2;
-  if ((u32) au.s.low < (u32) bu.s.low)
-    return 0;
-  else if ((u32) au.s.low > (u32) bu.s.low)
-    return 2;
-  return 1;
+	DIunion au, bu;
+
+	au.ll = a, bu.ll = b;
+
+	if ((u32) au.s.high < (u32) bu.s.high)
+		return 0;
+	else if ((u32) au.s.high > (u32) bu.s.high)
+		return 2;
+	if ((u32) au.s.low < (u32) bu.s.low)
+		return 0;
+	else if ((u32) au.s.low > (u32) bu.s.low)
+		return 2;
+	return 1;
 }
-
diff --git a/arch/arm/lib/udivdi3.c b/arch/arm/lib/udivdi3.c
index df1d5ef62bc..e343be4c664 100644
--- a/arch/arm/lib/udivdi3.c
+++ b/arch/arm/lib/udivdi3.c
@@ -32,211 +32,191 @@ Boston, MA 02111-1307, USA.  */
 #include "gcclib.h"
 #include "longlong.h"
 
-static const u8 __clz_tab[] =
-{
-  0,1,2,2,3,3,3,3,4,4,4,4,4,4,4,4,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,
-  6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,
-  7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,
-  7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,
-  8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
-  8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
-  8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
-  8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
+static const u8 __clz_tab[] = {
+	0, 1, 2, 2, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 5, 5, 5, 5,
+	    5, 5, 5, 5, 5, 5, 5, 5,
+	6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
+	    6, 6, 6, 6, 6, 6, 6, 6,
+	7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+	    7, 7, 7, 7, 7, 7, 7, 7,
+	7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+	    7, 7, 7, 7, 7, 7, 7, 7,
+	8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
+	    8, 8, 8, 8, 8, 8, 8, 8,
+	8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
+	    8, 8, 8, 8, 8, 8, 8, 8,
+	8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
+	    8, 8, 8, 8, 8, 8, 8, 8,
+	8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
+	    8, 8, 8, 8, 8, 8, 8, 8,
 };
 
-u64
-__udivmoddi4 (u64 n, u64 d, u64 *rp)
+u64 __udivmoddi4(u64 n, u64 d, u64 * rp)
 {
-  DIunion ww;
-  DIunion nn, dd;
-  DIunion rr;
-  u32 d0, d1, n0, n1, n2;
-  u32 q0, q1;
-  u32 b, bm;
-
-  nn.ll = n;
-  dd.ll = d;
-
-  d0 = dd.s.low;
-  d1 = dd.s.high;
-  n0 = nn.s.low;
-  n1 = nn.s.high;
-
-  if (d1 == 0)
-    {
-      if (d0 > n1)
-        {
-          /* 0q = nn / 0D */
-
-          count_leading_zeros (bm, d0);
-
-          if (bm != 0)
-            {
-              /* Normalize, i.e. make the most significant bit of the
-                 denominator set.  */
-
-              d0 = d0 << bm;
-              n1 = (n1 << bm) | (n0 >> (SI_TYPE_SIZE - bm));
-              n0 = n0 << bm;
-            }
-
-          udiv_qrnnd (q0, n0, n1, n0, d0);
-          q1 = 0;
-
-          /* Remainder in n0 >> bm.  */
-        }
-      else
-        {
-          /* qq = NN / 0d */
-
-          if (d0 == 0)
-            d0 = 1 / d0;        /* Divide intentionally by zero.  */
-
-          count_leading_zeros (bm, d0);
-
-          if (bm == 0)
-            {
-              /* From (n1 >= d0) /\ (the most significant bit of d0 is set),
-                 conclude (the most significant bit of n1 is set) /\ (the
-                 leading quotient digit q1 = 1).
-
-                 This special case is necessary, not an optimization.
-                 (Shifts counts of SI_TYPE_SIZE are undefined.)  */
-
-              n1 -= d0;
-              q1 = 1;
-            }
-          else
-            {
-              /* Normalize.  */
-
-              b = SI_TYPE_SIZE - bm;
-
-              d0 = d0 << bm;
-              n2 = n1 >> b;
-              n1 = (n1 << bm) | (n0 >> b);
-              n0 = n0 << bm;
-
-              udiv_qrnnd (q1, n1, n2, n1, d0);
-            }
-
-          /* n1 != d0...  */
-
-          udiv_qrnnd (q0, n0, n1, n0, d0);
-
-          /* Remainder in n0 >> bm.  */
-        }
-
-      if (rp != 0)
-        {
-          rr.s.low = n0 >> bm;
-          rr.s.high = 0;
-          *rp = rr.ll;
-        }
-    }
-  else
-    {
-      if (d1 > n1)
-        {
-          /* 00 = nn / DD */
-
-          q0 = 0;
-          q1 = 0;
-
-          /* Remainder in n1n0.  */
-          if (rp != 0)
-            {
-              rr.s.low = n0;
-              rr.s.high = n1;
-              *rp = rr.ll;
-            }
-        }
-      else
-        {
-          /* 0q = NN / dd */
-
-          count_leading_zeros (bm, d1);
-          if (bm == 0)
-            {
-              /* From (n1 >= d1) /\ (the most significant bit of d1 is set),
-                 conclude (the most significant bit of n1 is set) /\ (the
-                 quotient digit q0 = 0 or 1).
-
-                 This special case is necessary, not an optimization.  */
-
-              /* The condition on the next line takes advantage of that
-                 n1 >= d1 (true due to program flow).  */
-              if (n1 > d1 || n0 >= d0)
-                {
-                  q0 = 1;
-                  sub_ddmmss (n1, n0, n1, n0, d1, d0);
-                }
-              else
-                q0 = 0;
-
-              q1 = 0;
-
-              if (rp != 0)
-                {
-                  rr.s.low = n0;
-                  rr.s.high = n1;
-                  *rp = rr.ll;
-                }
-            }
-          else
-            {
-              u32 m1, m0;
-              /* Normalize.  */
-
-              b = SI_TYPE_SIZE - bm;
-
-              d1 = (d1 << bm) | (d0 >> b);
-              d0 = d0 << bm;
-              n2 = n1 >> b;
-              n1 = (n1 << bm) | (n0 >> b);
-              n0 = n0 << bm;
-
-              udiv_qrnnd (q0, n1, n2, n1, d1);
-              umul_ppmm (m1, m0, q0, d0);
-
-              if (m1 > n1 || (m1 == n1 && m0 > n0))
-                {
-                  q0--;
-                  sub_ddmmss (m1, m0, m1, m0, d1, d0);
-                }
-
-              q1 = 0;
-
-              /* Remainder in (n1n0 - m1m0) >> bm.  */
-              if (rp != 0)
-                {
-                  sub_ddmmss (n1, n0, n1, n0, m1, m0);
-                  rr.s.low = (n1 << b) | (n0 >> bm);
-                  rr.s.high = n1 >> bm;
-                  *rp = rr.ll;
-                }
-            }
-        }
-    }
-
-  ww.s.low = q0;
-  ww.s.high = q1;
-  return ww.ll;
+	DIunion ww;
+	DIunion nn, dd;
+	DIunion rr;
+	u32 d0, d1, n0, n1, n2;
+	u32 q0, q1;
+	u32 b, bm;
+
+	nn.ll = n;
+	dd.ll = d;
+
+	d0 = dd.s.low;
+	d1 = dd.s.high;
+	n0 = nn.s.low;
+	n1 = nn.s.high;
+
+	if (d1 == 0) {
+		if (d0 > n1) {
+			/* 0q = nn / 0D */
+
+			count_leading_zeros(bm, d0);
+
+			if (bm != 0) {
+				/* Normalize, i.e. make the most significant bit of the
+				   denominator set.  */
+
+				d0 = d0 << bm;
+				n1 = (n1 << bm) | (n0 >> (SI_TYPE_SIZE - bm));
+				n0 = n0 << bm;
+			}
+
+			udiv_qrnnd(q0, n0, n1, n0, d0);
+			q1 = 0;
+
+			/* Remainder in n0 >> bm.  */
+		} else {
+			/* qq = NN / 0d */
+
+			if (d0 == 0)
+				d0 = 1 / d0;	/* Divide intentionally by zero.  */
+
+			count_leading_zeros(bm, d0);
+
+			if (bm == 0) {
+				/* From (n1 >= d0) /\ (the most significant bit of d0 is set),
+				   conclude (the most significant bit of n1 is set) /\ (the
+				   leading quotient digit q1 = 1).
+
+				   This special case is necessary, not an optimization.
+				   (Shifts counts of SI_TYPE_SIZE are undefined.)  */
+
+				n1 -= d0;
+				q1 = 1;
+			} else {
+				/* Normalize.  */
+
+				b = SI_TYPE_SIZE - bm;
+
+				d0 = d0 << bm;
+				n2 = n1 >> b;
+				n1 = (n1 << bm) | (n0 >> b);
+				n0 = n0 << bm;
+
+				udiv_qrnnd(q1, n1, n2, n1, d0);
+			}
+
+			/* n1 != d0...  */
+
+			udiv_qrnnd(q0, n0, n1, n0, d0);
+
+			/* Remainder in n0 >> bm.  */
+		}
+
+		if (rp != 0) {
+			rr.s.low = n0 >> bm;
+			rr.s.high = 0;
+			*rp = rr.ll;
+		}
+	} else {
+		if (d1 > n1) {
+			/* 00 = nn / DD */
+
+			q0 = 0;
+			q1 = 0;
+
+			/* Remainder in n1n0.  */
+			if (rp != 0) {
+				rr.s.low = n0;
+				rr.s.high = n1;
+				*rp = rr.ll;
+			}
+		} else {
+			/* 0q = NN / dd */
+
+			count_leading_zeros(bm, d1);
+			if (bm == 0) {
+				/* From (n1 >= d1) /\ (the most significant bit of d1 is set),
+				   conclude (the most significant bit of n1 is set) /\ (the
+				   quotient digit q0 = 0 or 1).
+
+				   This special case is necessary, not an optimization.  */
+
+				/* The condition on the next line takes advantage of that
+				   n1 >= d1 (true due to program flow).  */
+				if (n1 > d1 || n0 >= d0) {
+					q0 = 1;
+					sub_ddmmss(n1, n0, n1, n0, d1, d0);
+				} else
+					q0 = 0;
+
+				q1 = 0;
+
+				if (rp != 0) {
+					rr.s.low = n0;
+					rr.s.high = n1;
+					*rp = rr.ll;
+				}
+			} else {
+				u32 m1, m0;
+				/* Normalize.  */
+
+				b = SI_TYPE_SIZE - bm;
+
+				d1 = (d1 << bm) | (d0 >> b);
+				d0 = d0 << bm;
+				n2 = n1 >> b;
+				n1 = (n1 << bm) | (n0 >> b);
+				n0 = n0 << bm;
+
+				udiv_qrnnd(q0, n1, n2, n1, d1);
+				umul_ppmm(m1, m0, q0, d0);
+
+				if (m1 > n1 || (m1 == n1 && m0 > n0)) {
+					q0--;
+					sub_ddmmss(m1, m0, m1, m0, d1, d0);
+				}
+
+				q1 = 0;
+
+				/* Remainder in (n1n0 - m1m0) >> bm.  */
+				if (rp != 0) {
+					sub_ddmmss(n1, n0, n1, n0, m1, m0);
+					rr.s.low = (n1 << b) | (n0 >> bm);
+					rr.s.high = n1 >> bm;
+					*rp = rr.ll;
+				}
+			}
+		}
+	}
+
+	ww.s.low = q0;
+	ww.s.high = q1;
+	return ww.ll;
 }
 
-u64
-__udivdi3 (u64 n, u64 d)
+u64 __udivdi3(u64 n, u64 d)
 {
-  return __udivmoddi4 (n, d, (u64 *) 0);
+	return __udivmoddi4(n, d, (u64 *) 0);
 }
 
-u64
-__umoddi3 (u64 u, u64 v)
+u64 __umoddi3(u64 u, u64 v)
 {
-  u64 w;
+	u64 w;
 
-  (void) __udivmoddi4 (u ,v, &w);
+	(void)__udivmoddi4(u, v, &w);
 
-  return w;
+	return w;
 }
-
-- 
cgit v1.2.3-70-g09d2


From 14eb75b6f83e3213d752dfe821df2fa57efbe519 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 20 Jun 2005 16:56:08 +0100
Subject: [PATCH] ARM: Add missed include for dmabounce.c

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/common/dmabounce.c | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'arch')

diff --git a/arch/arm/common/dmabounce.c b/arch/arm/common/dmabounce.c
index 39dd33ff6f3..9d63a01214e 100644
--- a/arch/arm/common/dmabounce.c
+++ b/arch/arm/common/dmabounce.c
@@ -30,6 +30,8 @@
 #include <linux/dmapool.h>
 #include <linux/list.h>
 
+#include <asm/cacheflush.h>
+
 #undef DEBUG
 
 #undef STATS
-- 
cgit v1.2.3-70-g09d2


From a507ef3ac68f0671fdd3f824885cd7f1ea1040b1 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 20 Jun 2005 16:57:17 +0100
Subject: [PATCH] ARM: Remove nmi_tick() from Integrator.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-integrator/core.c | 1 -
 1 file changed, 1 deletion(-)

(limited to 'arch')

diff --git a/arch/arm/mach-integrator/core.c b/arch/arm/mach-integrator/core.c
index d302f0405fd..bd1e5e3c9d3 100644
--- a/arch/arm/mach-integrator/core.c
+++ b/arch/arm/mach-integrator/core.c
@@ -227,7 +227,6 @@ integrator_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 	 * primary CPU
 	 */
 	if (hard_smp_processor_id() == 0) {
-		nmi_tick();
 		timer_tick(regs);
 #ifdef CONFIG_SMP
 		smp_send_timer();
-- 
cgit v1.2.3-70-g09d2


From 34b727c135ff651f153be5757056d25678b6d018 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Thu, 9 Jun 2005 22:36:00 -0700
Subject: [IA64] Drop spurious paren in entry.h

The latest assembler catches this typo.  (reported by Jim Wilson).

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/entry.h b/arch/ia64/kernel/entry.h
index 6d4ecec989b..78eeb079341 100644
--- a/arch/ia64/kernel/entry.h
+++ b/arch/ia64/kernel/entry.h
@@ -60,7 +60,7 @@
 	.spillsp @priunat,SW(AR_UNAT)+16+(off);					\
 	.spillsp ar.rnat,SW(AR_RNAT)+16+(off);					\
 	.spillsp ar.bspstore,SW(AR_BSPSTORE)+16+(off);				\
-	.spillsp pr,SW(PR)+16+(off))
+	.spillsp pr,SW(PR)+16+(off)
 
 #define DO_SAVE_SWITCH_STACK			\
 	movl r28=1f;				\
-- 
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 'arch')

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 'arch')

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 'arch')

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 'arch')

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 d67947a1bd6530791ad3663b93d91b44de89c2ca Mon Sep 17 00:00:00 2001
From: Richard Purdie <rpurdie@net.rmk.(none)>
Date: Mon, 20 Jun 2005 18:51:07 +0100
Subject: [PATCH] ARM: 2716/1: SharpSL Param: Fix typo

Patch from Richard Purdie

Fix typo in sharpsl_param.c so it works correctly on collie.

Signed-off-by: Richard Purdie
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/common/sharpsl_param.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/arm/common/sharpsl_param.c b/arch/arm/common/sharpsl_param.c
index c2c557a224c..c94864c5b1a 100644
--- a/arch/arm/common/sharpsl_param.c
+++ b/arch/arm/common/sharpsl_param.c
@@ -22,7 +22,7 @@
  * them early in the boot process, then pass them to the appropriate drivers.
  * Not all devices use all paramaters but the format is common to all.
  */
-#ifdef ARCH_SA1100
+#ifdef CONFIG_ARCH_SA1100
 #define PARAM_BASE	0xe8ffc000
 #else
 #define PARAM_BASE	0xa0000a00
-- 
cgit v1.2.3-70-g09d2


From f0ffeddc897a5cc5d5c6cc1b99a8799a8b34b28f Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@org.rmk.(none)>
Date: Mon, 20 Jun 2005 18:51:08 +0100
Subject: [PATCH] ARM: 2719/1: enable module support in ixp2000 defconfigs by
 default

Patch from Lennert Buytenhek

The ixp2000 defconfigs are among the few that do not enable module
support by default.  I keep enabling module support by hand for every
new kernel version, so let's just make this change upstream.

Signed-off-by: Lennert Buytenhek
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/configs/enp2611_defconfig  | 8 +++++++-
 arch/arm/configs/ixdp2400_defconfig | 8 +++++++-
 arch/arm/configs/ixdp2401_defconfig | 8 +++++++-
 arch/arm/configs/ixdp2800_defconfig | 8 +++++++-
 arch/arm/configs/ixdp2801_defconfig | 8 +++++++-
 5 files changed, 35 insertions(+), 5 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/configs/enp2611_defconfig b/arch/arm/configs/enp2611_defconfig
index e8f9fccffe8..06fae4b6277 100644
--- a/arch/arm/configs/enp2611_defconfig
+++ b/arch/arm/configs/enp2611_defconfig
@@ -50,7 +50,13 @@ CONFIG_BASE_SMALL=0
 #
 # Loadable module support
 #
-# CONFIG_MODULES is not set
+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
 
 #
 # System Type
diff --git a/arch/arm/configs/ixdp2400_defconfig b/arch/arm/configs/ixdp2400_defconfig
index 4fd663ecbe3..810a450a55d 100644
--- a/arch/arm/configs/ixdp2400_defconfig
+++ b/arch/arm/configs/ixdp2400_defconfig
@@ -50,7 +50,13 @@ CONFIG_BASE_SMALL=0
 #
 # Loadable module support
 #
-# CONFIG_MODULES is not set
+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
 
 #
 # System Type
diff --git a/arch/arm/configs/ixdp2401_defconfig b/arch/arm/configs/ixdp2401_defconfig
index 6f51c98084a..72e1b940e97 100644
--- a/arch/arm/configs/ixdp2401_defconfig
+++ b/arch/arm/configs/ixdp2401_defconfig
@@ -50,7 +50,13 @@ CONFIG_BASE_SMALL=0
 #
 # Loadable module support
 #
-# CONFIG_MODULES is not set
+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
 
 #
 # System Type
diff --git a/arch/arm/configs/ixdp2800_defconfig b/arch/arm/configs/ixdp2800_defconfig
index 7be3521f91f..1592e45f027 100644
--- a/arch/arm/configs/ixdp2800_defconfig
+++ b/arch/arm/configs/ixdp2800_defconfig
@@ -50,7 +50,13 @@ CONFIG_BASE_SMALL=0
 #
 # Loadable module support
 #
-# CONFIG_MODULES is not set
+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
 
 #
 # System Type
diff --git a/arch/arm/configs/ixdp2801_defconfig b/arch/arm/configs/ixdp2801_defconfig
index cd84a20f30f..f1afe3d09ec 100644
--- a/arch/arm/configs/ixdp2801_defconfig
+++ b/arch/arm/configs/ixdp2801_defconfig
@@ -50,7 +50,13 @@ CONFIG_BASE_SMALL=0
 #
 # Loadable module support
 #
-# CONFIG_MODULES is not set
+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
 
 #
 # System Type
-- 
cgit v1.2.3-70-g09d2


From 8874b414ffe037c39e73bb262ddf69653a13c0a4 Mon Sep 17 00:00:00 2001
From: "gregkh@suse.de" <gregkh@suse.de>
Date: Wed, 23 Mar 2005 09:56:34 -0800
Subject: [PATCH] class: convert arch/* to use the new class api instead of
 class_simple

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/i386/kernel/cpuid.c | 22 +++++++++++-----------
 arch/i386/kernel/msr.c   | 22 +++++++++++-----------
 2 files changed, 22 insertions(+), 22 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/cpuid.c b/arch/i386/kernel/cpuid.c
index 2e2756345bb..4647db4ad6d 100644
--- a/arch/i386/kernel/cpuid.c
+++ b/arch/i386/kernel/cpuid.c
@@ -45,7 +45,7 @@
 #include <asm/uaccess.h>
 #include <asm/system.h>
 
-static struct class_simple *cpuid_class;
+static struct class *cpuid_class;
 
 #ifdef CONFIG_SMP
 
@@ -158,12 +158,12 @@ static struct file_operations cpuid_fops = {
 	.open = cpuid_open,
 };
 
-static int cpuid_class_simple_device_add(int i) 
+static int cpuid_class_device_create(int i)
 {
 	int err = 0;
 	struct class_device *class_err;
 
-	class_err = class_simple_device_add(cpuid_class, MKDEV(CPUID_MAJOR, i), NULL, "cpu%d",i);
+	class_err = class_device_create(cpuid_class, MKDEV(CPUID_MAJOR, i), NULL, "cpu%d",i);
 	if (IS_ERR(class_err))
 		err = PTR_ERR(class_err);
 	return err;
@@ -175,10 +175,10 @@ static int __devinit cpuid_class_cpu_callback(struct notifier_block *nfb, unsign
 
 	switch (action) {
 	case CPU_ONLINE:
-		cpuid_class_simple_device_add(cpu);
+		cpuid_class_device_create(cpu);
 		break;
 	case CPU_DEAD:
-		class_simple_device_remove(MKDEV(CPUID_MAJOR, cpu));
+		class_device_destroy(cpuid_class, MKDEV(CPUID_MAJOR, cpu));
 		break;
 	}
 	return NOTIFY_OK;
@@ -200,13 +200,13 @@ static int __init cpuid_init(void)
 		err = -EBUSY;
 		goto out;
 	}
-	cpuid_class = class_simple_create(THIS_MODULE, "cpuid");
+	cpuid_class = class_create(THIS_MODULE, "cpuid");
 	if (IS_ERR(cpuid_class)) {
 		err = PTR_ERR(cpuid_class);
 		goto out_chrdev;
 	}
 	for_each_online_cpu(i) {
-		err = cpuid_class_simple_device_add(i);
+		err = cpuid_class_device_create(i);
 		if (err != 0) 
 			goto out_class;
 	}
@@ -218,9 +218,9 @@ static int __init cpuid_init(void)
 out_class:
 	i = 0;
 	for_each_online_cpu(i) {
-		class_simple_device_remove(MKDEV(CPUID_MAJOR, i));
+		class_device_destroy(cpuid_class, MKDEV(CPUID_MAJOR, i));
 	}
-	class_simple_destroy(cpuid_class);
+	class_destroy(cpuid_class);
 out_chrdev:
 	unregister_chrdev(CPUID_MAJOR, "cpu/cpuid");	
 out:
@@ -232,8 +232,8 @@ static void __exit cpuid_exit(void)
 	int cpu = 0;
 
 	for_each_online_cpu(cpu)
-		class_simple_device_remove(MKDEV(CPUID_MAJOR, cpu));
-	class_simple_destroy(cpuid_class);
+		class_device_destroy(cpuid_class, MKDEV(CPUID_MAJOR, cpu));
+	class_destroy(cpuid_class);
 	unregister_chrdev(CPUID_MAJOR, "cpu/cpuid");
 	unregister_cpu_notifier(&cpuid_class_cpu_notifier);
 }
diff --git a/arch/i386/kernel/msr.c b/arch/i386/kernel/msr.c
index 05d9f8f363a..b2f03c39a6f 100644
--- a/arch/i386/kernel/msr.c
+++ b/arch/i386/kernel/msr.c
@@ -44,7 +44,7 @@
 #include <asm/uaccess.h>
 #include <asm/system.h>
 
-static struct class_simple *msr_class;
+static struct class *msr_class;
 
 /* Note: "err" is handled in a funny way below.  Otherwise one version
    of gcc or another breaks. */
@@ -260,12 +260,12 @@ static struct file_operations msr_fops = {
 	.open = msr_open,
 };
 
-static int msr_class_simple_device_add(int i)
+static int msr_class_device_create(int i)
 {
 	int err = 0;
 	struct class_device *class_err;
 
-	class_err = class_simple_device_add(msr_class, MKDEV(MSR_MAJOR, i), NULL, "msr%d",i);
+	class_err = class_device_create(msr_class, MKDEV(MSR_MAJOR, i), NULL, "msr%d",i);
 	if (IS_ERR(class_err)) 
 		err = PTR_ERR(class_err);
 	return err;
@@ -277,10 +277,10 @@ static int __devinit msr_class_cpu_callback(struct notifier_block *nfb, unsigned
 
 	switch (action) {
 	case CPU_ONLINE:
-		msr_class_simple_device_add(cpu);
+		msr_class_device_create(cpu);
 		break;
 	case CPU_DEAD:
-		class_simple_device_remove(MKDEV(MSR_MAJOR, cpu));	
+		class_device_destroy(msr_class, MKDEV(MSR_MAJOR, cpu));
 		break;
 	}
 	return NOTIFY_OK;
@@ -302,13 +302,13 @@ static int __init msr_init(void)
 		err = -EBUSY;
 		goto out;
 	}
-	msr_class = class_simple_create(THIS_MODULE, "msr");
+	msr_class = class_create(THIS_MODULE, "msr");
 	if (IS_ERR(msr_class)) {
 		err = PTR_ERR(msr_class);
 		goto out_chrdev;
 	}
 	for_each_online_cpu(i) {
-		err = msr_class_simple_device_add(i);
+		err = msr_class_device_create(i);
 		if (err != 0)
 			goto out_class;
 	}
@@ -320,8 +320,8 @@ static int __init msr_init(void)
 out_class:
 	i = 0;
 	for_each_online_cpu(i)
-		class_simple_device_remove(MKDEV(MSR_MAJOR, i));
-	class_simple_destroy(msr_class);
+		class_device_destroy(msr_class, MKDEV(MSR_MAJOR, i));
+	class_destroy(msr_class);
 out_chrdev:
 	unregister_chrdev(MSR_MAJOR, "cpu/msr");
 out:
@@ -332,8 +332,8 @@ static void __exit msr_exit(void)
 {
 	int cpu = 0;
 	for_each_online_cpu(cpu)
-		class_simple_device_remove(MKDEV(MSR_MAJOR, cpu));
-	class_simple_destroy(msr_class);
+		class_device_destroy(msr_class, MKDEV(MSR_MAJOR, cpu));
+	class_destroy(msr_class);
 	unregister_chrdev(MSR_MAJOR, "cpu/msr");
 	unregister_cpu_notifier(&msr_class_cpu_notifier);
 }
-- 
cgit v1.2.3-70-g09d2


From 6623415687eaffef49429292ab062bb046ee3311 Mon Sep 17 00:00:00 2001
From: Patrick Mochel <mochel@digitalimplant.org>
Date: Thu, 28 Apr 2005 17:11:52 -0700
Subject: [PATCH] sn: fixes due to driver core changes

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/ia64/sn/kernel/tiocx.c | 21 +++++++++------------
 1 file changed, 9 insertions(+), 12 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/sn/kernel/tiocx.c b/arch/ia64/sn/kernel/tiocx.c
index ab9b5f35c2a..b6bcfe0b0f0 100644
--- a/arch/ia64/sn/kernel/tiocx.c
+++ b/arch/ia64/sn/kernel/tiocx.c
@@ -518,25 +518,22 @@ static int __init tiocx_init(void)
 	return 0;
 }
 
-static void __exit tiocx_exit(void)
+static int cx_remove_device(struct device * dev, void * data)
 {
-	struct device *dev;
-	struct device *tdev;
+	struct cx_dev *cx_dev = to_cx_dev(dev);
+	device_remove_file(dev, &dev_attr_cxdev_control);
+	cx_device_unregister(cx_dev);
+	return 0;
+}
 
+static void __exit tiocx_exit(void)
+{
 	DBG("tiocx_exit\n");
 
 	/*
 	 * Unregister devices.
 	 */
-	list_for_each_entry_safe(dev, tdev, &tiocx_bus_type.devices.list,
-				 bus_list) {
-		if (dev) {
-			struct cx_dev *cx_dev = to_cx_dev(dev);
-			device_remove_file(dev, &dev_attr_cxdev_control);
-			cx_device_unregister(cx_dev);
-		}
-	}
-
+	bus_for_each_dev(&tiocx_bus_type, NULL, NULL, cx_remove_device);
 	bus_unregister(&tiocx_bus_type);
 }
 
-- 
cgit v1.2.3-70-g09d2


From ff381d2223a30ee70752791fd9c3588d8f1cab77 Mon Sep 17 00:00:00 2001
From: Yani Ioannou <yani.ioannou@gmail.com>
Date: Tue, 17 May 2005 06:40:51 -0400
Subject: [PATCH] Driver Core: arch: update device attribute callbacks

Signed-off-by: Yani Ioannou <yani.ioannou@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/arm/common/amba.c        |  2 +-
 arch/arm/kernel/ecard.c       | 12 ++++++------
 arch/arm26/kernel/ecard.c     | 10 +++++-----
 arch/ia64/sn/kernel/tiocx.c   |  4 ++--
 arch/parisc/kernel/drivers.c  |  2 +-
 arch/ppc/kernel/pci.c         |  2 +-
 arch/ppc/syslib/ocp.c         |  2 +-
 arch/ppc/syslib/of_device.c   |  2 +-
 arch/ppc64/kernel/of_device.c |  2 +-
 arch/ppc64/kernel/pci.c       |  2 +-
 arch/ppc64/kernel/vio.c       |  4 ++--
 11 files changed, 22 insertions(+), 22 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/common/amba.c b/arch/arm/common/amba.c
index a0507f8c33f..c6beb751f2a 100644
--- a/arch/arm/common/amba.c
+++ b/arch/arm/common/amba.c
@@ -169,7 +169,7 @@ static void amba_device_release(struct device *dev)
 }
 
 #define amba_attr(name,fmt,arg...)				\
-static ssize_t show_##name(struct device *_dev, char *buf)	\
+static ssize_t show_##name(struct device *_dev, struct device_attribute *attr, char *buf)	\
 {								\
 	struct amba_device *dev = to_amba_device(_dev);		\
 	return sprintf(buf, fmt, arg);				\
diff --git a/arch/arm/kernel/ecard.c b/arch/arm/kernel/ecard.c
index 3dc15b131f5..6540db69133 100644
--- a/arch/arm/kernel/ecard.c
+++ b/arch/arm/kernel/ecard.c
@@ -866,19 +866,19 @@ static struct expansion_card *__init ecard_alloc_card(int type, int slot)
 	return ec;
 }
 
-static ssize_t ecard_show_irq(struct device *dev, char *buf)
+static ssize_t ecard_show_irq(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct expansion_card *ec = ECARD_DEV(dev);
 	return sprintf(buf, "%u\n", ec->irq);
 }
 
-static ssize_t ecard_show_dma(struct device *dev, char *buf)
+static ssize_t ecard_show_dma(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct expansion_card *ec = ECARD_DEV(dev);
 	return sprintf(buf, "%u\n", ec->dma);
 }
 
-static ssize_t ecard_show_resources(struct device *dev, char *buf)
+static ssize_t ecard_show_resources(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct expansion_card *ec = ECARD_DEV(dev);
 	char *str = buf;
@@ -893,19 +893,19 @@ static ssize_t ecard_show_resources(struct device *dev, char *buf)
 	return str - buf;
 }
 
-static ssize_t ecard_show_vendor(struct device *dev, char *buf)
+static ssize_t ecard_show_vendor(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct expansion_card *ec = ECARD_DEV(dev);
 	return sprintf(buf, "%u\n", ec->cid.manufacturer);
 }
 
-static ssize_t ecard_show_device(struct device *dev, char *buf)
+static ssize_t ecard_show_device(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct expansion_card *ec = ECARD_DEV(dev);
 	return sprintf(buf, "%u\n", ec->cid.product);
 }
 
-static ssize_t ecard_show_type(struct device *dev, char *buf)
+static ssize_t ecard_show_type(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct expansion_card *ec = ECARD_DEV(dev);
 	return sprintf(buf, "%s\n", ec->type == ECARD_EASI ? "EASI" : "IOC");
diff --git a/arch/arm26/kernel/ecard.c b/arch/arm26/kernel/ecard.c
index 824c6b571ad..f2278aadac8 100644
--- a/arch/arm26/kernel/ecard.c
+++ b/arch/arm26/kernel/ecard.c
@@ -562,31 +562,31 @@ static void __init ecard_init_resources(struct expansion_card *ec)
 	}
 }
 
-static ssize_t ecard_show_irq(struct device *dev, char *buf)
+static ssize_t ecard_show_irq(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct expansion_card *ec = ECARD_DEV(dev);
 	return sprintf(buf, "%u\n", ec->irq);
 }
 
-static ssize_t ecard_show_vendor(struct device *dev, char *buf)
+static ssize_t ecard_show_vendor(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct expansion_card *ec = ECARD_DEV(dev);
 	return sprintf(buf, "%u\n", ec->cid.manufacturer);
 }
 
-static ssize_t ecard_show_device(struct device *dev, char *buf)
+static ssize_t ecard_show_device(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct expansion_card *ec = ECARD_DEV(dev);
 	return sprintf(buf, "%u\n", ec->cid.product);
 }
 
-static ssize_t ecard_show_dma(struct device *dev, char *buf)
+static ssize_t ecard_show_dma(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct expansion_card *ec = ECARD_DEV(dev);
 	return sprintf(buf, "%u\n", ec->dma);
 }
 
-static ssize_t ecard_show_resources(struct device *dev, char *buf)
+static ssize_t ecard_show_resources(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct expansion_card *ec = ECARD_DEV(dev);
 	char *str = buf;
diff --git a/arch/ia64/sn/kernel/tiocx.c b/arch/ia64/sn/kernel/tiocx.c
index b6bcfe0b0f0..a087b274847 100644
--- a/arch/ia64/sn/kernel/tiocx.c
+++ b/arch/ia64/sn/kernel/tiocx.c
@@ -432,7 +432,7 @@ static int tiocx_reload(struct cx_dev *cx_dev)
 	return cx_device_reload(cx_dev);
 }
 
-static ssize_t show_cxdev_control(struct device *dev, char *buf)
+static ssize_t show_cxdev_control(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct cx_dev *cx_dev = to_cx_dev(dev);
 
@@ -442,7 +442,7 @@ static ssize_t show_cxdev_control(struct device *dev, char *buf)
 		       tiocx_btchar_get(cx_dev->cx_id.nasid));
 }
 
-static ssize_t store_cxdev_control(struct device *dev, const char *buf,
+static ssize_t store_cxdev_control(struct device *dev, struct device_attribute *attr, const char *buf,
 				   size_t count)
 {
 	int n;
diff --git a/arch/parisc/kernel/drivers.c b/arch/parisc/kernel/drivers.c
index ebf186656af..d34bbe7ae0e 100644
--- a/arch/parisc/kernel/drivers.c
+++ b/arch/parisc/kernel/drivers.c
@@ -466,7 +466,7 @@ static int parisc_generic_match(struct device *dev, struct device_driver *drv)
 }
 
 #define pa_dev_attr(name, field, format_string)				\
-static ssize_t name##_show(struct device *dev, char *buf)		\
+static ssize_t name##_show(struct device *dev, struct device_attribute *attr, char *buf)		\
 {									\
 	struct parisc_device *padev = to_parisc_device(dev);		\
 	return sprintf(buf, format_string, padev->field);		\
diff --git a/arch/ppc/kernel/pci.c b/arch/ppc/kernel/pci.c
index 47a15306823..6d7b92d7245 100644
--- a/arch/ppc/kernel/pci.c
+++ b/arch/ppc/kernel/pci.c
@@ -1003,7 +1003,7 @@ pci_create_OF_bus_map(void)
 	}
 }
 
-static ssize_t pci_show_devspec(struct device *dev, char *buf)
+static ssize_t pci_show_devspec(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct pci_dev *pdev;
 	struct device_node *np;
diff --git a/arch/ppc/syslib/ocp.c b/arch/ppc/syslib/ocp.c
index a5156c5179a..e5fd2ae503e 100644
--- a/arch/ppc/syslib/ocp.c
+++ b/arch/ppc/syslib/ocp.c
@@ -68,7 +68,7 @@ static int ocp_inited;
 /* Sysfs support */
 #define OCP_DEF_ATTR(field, format_string)				\
 static ssize_t								\
-show_##field(struct device *dev, char *buf)				\
+show_##field(struct device *dev, struct device_attribute *attr, char *buf)				\
 {									\
 	struct ocp_device *odev = to_ocp_dev(dev);			\
 									\
diff --git a/arch/ppc/syslib/of_device.c b/arch/ppc/syslib/of_device.c
index 46269ed21ae..49c0e34e2d6 100644
--- a/arch/ppc/syslib/of_device.c
+++ b/arch/ppc/syslib/of_device.c
@@ -161,7 +161,7 @@ void of_unregister_driver(struct of_platform_driver *drv)
 }
 
 
-static ssize_t dev_show_devspec(struct device *dev, char *buf)
+static ssize_t dev_show_devspec(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct of_device *ofdev;
 
diff --git a/arch/ppc64/kernel/of_device.c b/arch/ppc64/kernel/of_device.c
index f4c825a69fa..66bd5ab7c25 100644
--- a/arch/ppc64/kernel/of_device.c
+++ b/arch/ppc64/kernel/of_device.c
@@ -161,7 +161,7 @@ void of_unregister_driver(struct of_platform_driver *drv)
 }
 
 
-static ssize_t dev_show_devspec(struct device *dev, char *buf)
+static ssize_t dev_show_devspec(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct of_device *ofdev;
 
diff --git a/arch/ppc64/kernel/pci.c b/arch/ppc64/kernel/pci.c
index d786d4b6af0..2bf0513f3ec 100644
--- a/arch/ppc64/kernel/pci.c
+++ b/arch/ppc64/kernel/pci.c
@@ -507,7 +507,7 @@ int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
 }
 
 #ifdef CONFIG_PPC_MULTIPLATFORM
-static ssize_t pci_show_devspec(struct device *dev, char *buf)
+static ssize_t pci_show_devspec(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct pci_dev *pdev;
 	struct device_node *np;
diff --git a/arch/ppc64/kernel/vio.c b/arch/ppc64/kernel/vio.c
index cdd830cb276..79f2dc7a983 100644
--- a/arch/ppc64/kernel/vio.c
+++ b/arch/ppc64/kernel/vio.c
@@ -300,7 +300,7 @@ static void __devinit vio_dev_release(struct device *dev)
 }
 
 #ifdef CONFIG_PPC_PSERIES
-static ssize_t viodev_show_devspec(struct device *dev, char *buf)
+static ssize_t viodev_show_devspec(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct device_node *of_node = dev->platform_data;
 
@@ -309,7 +309,7 @@ static ssize_t viodev_show_devspec(struct device *dev, char *buf)
 DEVICE_ATTR(devspec, S_IRUSR | S_IRGRP | S_IROTH, viodev_show_devspec, NULL);
 #endif
 
-static ssize_t viodev_show_name(struct device *dev, char *buf)
+static ssize_t viodev_show_name(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	return sprintf(buf, "%s\n", to_vio_dev(dev)->name);
 }
-- 
cgit v1.2.3-70-g09d2


From 95dccdfe29a76c513e29f88f366612d7570adab3 Mon Sep 17 00:00:00 2001
From: Greg Edwards <edwardsg@sgi.com>
Date: Tue, 21 Jun 2005 10:43:00 -0700
Subject: [IA64] refresh arch/ia64/defconfig

Refresh arch/ia64/defconfig, as it was getting a bit stale.  The only
manual changes I made were:

CONFIG_SCSI_SATA=y		needed for some Altix base I/O cards
CONFIG_SCSI_SATA_VITESSE=y

CONFIG_DM_MULTIPATH=m		the rest are already modules

CONFIG_FUSION_SPI=y		new driver breakout
CONFIG_FUSION_FC=m

CONFIG_SGI_TIOCX=y		enable some other SGI drivers
CONFIG_SGI_MBCS=m
CONFIG_AGP_SGI_TIOCA=m

Boot tested on Altix, HP rx2600 and Intel Tiger

Signed-off-by: Greg Edwards <edwardsg@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/defconfig | 138 ++++++++++++++++++++++++++++++++++------------------
 1 file changed, 91 insertions(+), 47 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/defconfig b/arch/ia64/defconfig
index 7539e83bf05..2aea0f9e6e1 100644
--- a/arch/ia64/defconfig
+++ b/arch/ia64/defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.10
-# Thu Jan  6 11:13:13 2005
+# Linux kernel version: 2.6.12
+# Tue Jun 21 11:30:42 2005
 #
 
 #
@@ -10,6 +10,7 @@
 CONFIG_EXPERIMENTAL=y
 CONFIG_CLEAN_COMPILE=y
 CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -21,24 +22,27 @@ CONFIG_POSIX_MQUEUE=y
 # CONFIG_BSD_PROCESS_ACCT is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
-CONFIG_LOG_BUF_SHIFT=20
 CONFIG_HOTPLUG=y
 CONFIG_KOBJECT_UEVENT=y
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
+# CONFIG_CPUSETS is not set
 # CONFIG_EMBEDDED is not set
 CONFIG_KALLSYMS=y
 CONFIG_KALLSYMS_ALL=y
 # CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_BASE_FULL=y
 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
+CONFIG_BASE_SMALL=0
 
 #
 # Loadable module support
@@ -63,9 +67,11 @@ CONFIG_GENERIC_CALIBRATE_DELAY=y
 CONFIG_TIME_INTERPOLATION=y
 CONFIG_EFI=y
 CONFIG_GENERIC_IOMAP=y
+CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y
 CONFIG_IA64_GENERIC=y
 # CONFIG_IA64_DIG is not set
 # CONFIG_IA64_HP_ZX1 is not set
+# CONFIG_IA64_HP_ZX1_SWIOTLB is not set
 # CONFIG_IA64_SGI_SN2 is not set
 # CONFIG_IA64_HP_SIM is not set
 # CONFIG_ITANIUM is not set
@@ -77,6 +83,7 @@ CONFIG_IA64_PAGE_SIZE_16KB=y
 CONFIG_IA64_L1_CACHE_SHIFT=7
 CONFIG_NUMA=y
 CONFIG_VIRTUAL_MEM_MAP=y
+CONFIG_HOLES_IN_ZONE=y
 CONFIG_DISCONTIGMEM=y
 CONFIG_IA64_CYCLONE=y
 CONFIG_IOSAPIC=y
@@ -84,6 +91,7 @@ CONFIG_FORCE_MAX_ZONEORDER=18
 CONFIG_SMP=y
 CONFIG_NR_CPUS=512
 CONFIG_HOTPLUG_CPU=y
+# CONFIG_SCHED_SMT is not set
 # CONFIG_PREEMPT is not set
 CONFIG_HAVE_DEC_LOCK=y
 CONFIG_IA32_SUPPORT=y
@@ -135,6 +143,7 @@ CONFIG_PCI_DOMAINS=y
 # CONFIG_PCI_MSI is not set
 CONFIG_PCI_LEGACY_PROC=y
 CONFIG_PCI_NAMES=y
+# CONFIG_PCI_DEBUG is not set
 
 #
 # PCI Hotplug Support
@@ -144,7 +153,6 @@ CONFIG_HOTPLUG_PCI=m
 CONFIG_HOTPLUG_PCI_ACPI=m
 # CONFIG_HOTPLUG_PCI_ACPI_IBM is not set
 # CONFIG_HOTPLUG_PCI_CPCI is not set
-# CONFIG_HOTPLUG_PCI_PCIE is not set
 # CONFIG_HOTPLUG_PCI_SHPC is not set
 
 #
@@ -152,10 +160,6 @@ CONFIG_HOTPLUG_PCI_ACPI=m
 #
 # CONFIG_PCCARD is not set
 
-#
-# PC-card bridges
-#
-
 #
 # Device Drivers
 #
@@ -190,6 +194,7 @@ CONFIG_PREVENT_FIRMWARE_BUILD=y
 # CONFIG_BLK_CPQ_CISS_DA is not set
 # CONFIG_BLK_DEV_DAC960 is not set
 # CONFIG_BLK_DEV_UMEM is not set
+# CONFIG_BLK_DEV_COW_COMMON is not set
 CONFIG_BLK_DEV_LOOP=m
 CONFIG_BLK_DEV_CRYPTOLOOP=m
 CONFIG_BLK_DEV_NBD=m
@@ -209,6 +214,7 @@ 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
@@ -283,6 +289,7 @@ CONFIG_CHR_DEV_ST=m
 CONFIG_BLK_DEV_SR=m
 # CONFIG_BLK_DEV_SR_VENDOR is not set
 CONFIG_CHR_DEV_SG=m
+# CONFIG_CHR_DEV_SCH is not set
 
 #
 # Some SCSI devices (e.g. CD jukebox) support multiple LUNs
@@ -310,13 +317,21 @@ CONFIG_SCSI_FC_ATTRS=y
 # CONFIG_SCSI_AIC79XX is not set
 # CONFIG_MEGARAID_NEWGEN is not set
 # CONFIG_MEGARAID_LEGACY is not set
-# CONFIG_SCSI_SATA is not set
-# CONFIG_SCSI_BUSLOGIC is not set
+CONFIG_SCSI_SATA=y
+# CONFIG_SCSI_SATA_AHCI is not set
+# CONFIG_SCSI_SATA_SVW is not set
+# CONFIG_SCSI_ATA_PIIX is not set
+# CONFIG_SCSI_SATA_NV is not set
+# CONFIG_SCSI_SATA_PROMISE is not set
+# CONFIG_SCSI_SATA_QSTOR is not set
+# CONFIG_SCSI_SATA_SX4 is not set
+# CONFIG_SCSI_SATA_SIL is not set
+# CONFIG_SCSI_SATA_SIS is not set
+# CONFIG_SCSI_SATA_ULI is not set
+# CONFIG_SCSI_SATA_VIA is not set
+CONFIG_SCSI_SATA_VITESSE=y
 # CONFIG_SCSI_DMX3191D is not set
-# CONFIG_SCSI_EATA is not set
-# CONFIG_SCSI_EATA_PIO is not set
 # CONFIG_SCSI_FUTURE_DOMAIN is not set
-# CONFIG_SCSI_GDTH is not set
 # CONFIG_SCSI_IPS is not set
 # CONFIG_SCSI_INITIO is not set
 # CONFIG_SCSI_INIA100 is not set
@@ -326,7 +341,6 @@ CONFIG_SCSI_SYM53C8XX_DEFAULT_TAGS=16
 CONFIG_SCSI_SYM53C8XX_MAX_TAGS=64
 # CONFIG_SCSI_SYM53C8XX_IOMAPPED is not set
 # CONFIG_SCSI_IPR is not set
-# CONFIG_SCSI_QLOGIC_ISP is not set
 CONFIG_SCSI_QLOGIC_FC=y
 # CONFIG_SCSI_QLOGIC_FC_FIRMWARE is not set
 CONFIG_SCSI_QLOGIC_1280=y
@@ -337,6 +351,7 @@ CONFIG_SCSI_QLA22XX=m
 CONFIG_SCSI_QLA2300=m
 CONFIG_SCSI_QLA2322=m
 # CONFIG_SCSI_QLA6312 is not set
+# CONFIG_SCSI_LPFC is not set
 # CONFIG_SCSI_DC395x is not set
 # CONFIG_SCSI_DC390T is not set
 # CONFIG_SCSI_DEBUG is not set
@@ -359,12 +374,16 @@ CONFIG_DM_CRYPT=m
 CONFIG_DM_SNAPSHOT=m
 CONFIG_DM_MIRROR=m
 CONFIG_DM_ZERO=m
+CONFIG_DM_MULTIPATH=m
+# CONFIG_DM_MULTIPATH_EMC is not set
 
 #
 # Fusion MPT device support
 #
 CONFIG_FUSION=y
-CONFIG_FUSION_MAX_SGE=40
+CONFIG_FUSION_SPI=y
+CONFIG_FUSION_FC=m
+CONFIG_FUSION_MAX_SGE=128
 # CONFIG_FUSION_CTL is not set
 
 #
@@ -387,7 +406,6 @@ CONFIG_NET=y
 #
 CONFIG_PACKET=y
 # CONFIG_PACKET_MMAP is not set
-CONFIG_NETLINK_DEV=y
 CONFIG_UNIX=y
 # CONFIG_NET_KEY is not set
 CONFIG_INET=y
@@ -447,7 +465,6 @@ CONFIG_DUMMY=m
 # CONFIG_BONDING is not set
 # CONFIG_EQUALIZER is not set
 # CONFIG_TUN is not set
-# CONFIG_ETHERTAP is not set
 
 #
 # ARCnet devices
@@ -484,9 +501,7 @@ CONFIG_NET_PCI=y
 # CONFIG_FORCEDETH is not set
 # CONFIG_DGRS is not set
 CONFIG_EEPRO100=m
-# CONFIG_EEPRO100_PIO is not set
 CONFIG_E100=m
-# CONFIG_E100_NAPI is not set
 # CONFIG_FEALNX is not set
 # CONFIG_NATSEMI is not set
 # CONFIG_NE2K_PCI is not set
@@ -508,9 +523,11 @@ CONFIG_E1000=y
 # CONFIG_HAMACHI is not set
 # CONFIG_YELLOWFIN is not set
 # CONFIG_R8169 is not set
+# CONFIG_SKGE is not set
 # CONFIG_SK98LIN is not set
 # CONFIG_VIA_VELOCITY is not set
 CONFIG_TIGON3=y
+# CONFIG_BNX2 is not set
 
 #
 # Ethernet (10000 Mbit)
@@ -567,25 +584,6 @@ CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
 # CONFIG_INPUT_EVDEV is not set
 # CONFIG_INPUT_EVBUG is not set
 
-#
-# Input I/O drivers
-#
-CONFIG_GAMEPORT=m
-CONFIG_SOUND_GAMEPORT=m
-# CONFIG_GAMEPORT_NS558 is not set
-# CONFIG_GAMEPORT_L4 is not set
-# CONFIG_GAMEPORT_EMU10K1 is not set
-# CONFIG_GAMEPORT_VORTEX is not set
-# CONFIG_GAMEPORT_FM801 is not set
-# CONFIG_GAMEPORT_CS461x is not set
-CONFIG_SERIO=y
-CONFIG_SERIO_I8042=y
-# CONFIG_SERIO_SERPORT is not set
-# CONFIG_SERIO_CT82C710 is not set
-# CONFIG_SERIO_PCIPS2 is not set
-CONFIG_SERIO_LIBPS2=y
-# CONFIG_SERIO_RAW is not set
-
 #
 # Input Device Drivers
 #
@@ -603,6 +601,23 @@ CONFIG_MOUSE_PS2=y
 # CONFIG_INPUT_TOUCHSCREEN is not set
 # CONFIG_INPUT_MISC is not set
 
+#
+# Hardware I/O ports
+#
+CONFIG_SERIO=y
+CONFIG_SERIO_I8042=y
+# CONFIG_SERIO_SERPORT is not set
+# CONFIG_SERIO_PCIPS2 is not set
+CONFIG_SERIO_LIBPS2=y
+# CONFIG_SERIO_RAW is not set
+CONFIG_GAMEPORT=m
+# CONFIG_GAMEPORT_NS558 is not set
+# CONFIG_GAMEPORT_L4 is not set
+# CONFIG_GAMEPORT_EMU10K1 is not set
+# CONFIG_GAMEPORT_VORTEX is not set
+# CONFIG_GAMEPORT_FM801 is not set
+# CONFIG_GAMEPORT_CS461X is not set
+
 #
 # Character devices
 #
@@ -613,11 +628,15 @@ CONFIG_SERIAL_NONSTANDARD=y
 # CONFIG_ROCKETPORT is not set
 # CONFIG_CYCLADES is not set
 # CONFIG_MOXA_SMARTIO is not set
-# CONFIG_SYNCLINK is not set
+# CONFIG_ISI is not set
 # CONFIG_SYNCLINKMP is not set
 # CONFIG_N_HDLC is not set
+# CONFIG_SPECIALIX is not set
+# CONFIG_SX is not set
 # CONFIG_STALDRV is not set
 CONFIG_SGI_SNSC=y
+CONFIG_SGI_TIOCX=y
+CONFIG_SGI_MBCS=m
 
 #
 # Serial drivers
@@ -638,6 +657,7 @@ CONFIG_SERIAL_8250_SHARE_IRQ=y
 CONFIG_SERIAL_CORE=y
 CONFIG_SERIAL_CORE_CONSOLE=y
 CONFIG_SERIAL_SGI_L1_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
 CONFIG_UNIX98_PTYS=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
@@ -663,6 +683,7 @@ CONFIG_EFI_RTC=y
 CONFIG_AGP=m
 CONFIG_AGP_I460=m
 CONFIG_AGP_HP_ZX1=m
+CONFIG_AGP_SGI_TIOCA=m
 CONFIG_DRM=m
 CONFIG_DRM_TDFX=m
 CONFIG_DRM_R128=m
@@ -674,8 +695,14 @@ CONFIG_HPET=y
 # CONFIG_HPET_RTC_IRQ is not set
 CONFIG_HPET_MMAP=y
 CONFIG_MAX_RAW_DEVS=256
+# CONFIG_HANGCHECK_TIMER is not set
 CONFIG_MMTIMER=y
 
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
+
 #
 # I2C support
 #
@@ -760,6 +787,8 @@ CONFIG_SND_CS46XX=m
 CONFIG_SND_CS46XX_NEW_DSP=y
 CONFIG_SND_CS4281=m
 CONFIG_SND_EMU10K1=m
+# CONFIG_SND_EMU10K1X is not set
+# CONFIG_SND_CA0106 is not set
 # CONFIG_SND_KORG1212 is not set
 # CONFIG_SND_MIXART is not set
 # CONFIG_SND_NM256 is not set
@@ -784,13 +813,14 @@ CONFIG_SND_FM801=m
 # CONFIG_SND_INTEL8X0M is not set
 # CONFIG_SND_SONICVIBES is not set
 # CONFIG_SND_VIA82XX is not set
+# CONFIG_SND_VIA82XX_MODEM is not set
 # CONFIG_SND_VX222 is not set
+# CONFIG_SND_HDA_INTEL is not set
 
 #
 # USB devices
 #
 # CONFIG_SND_USB_AUDIO is not set
-# CONFIG_SND_USB_USX2Y is not set
 
 #
 # Open Sound System
@@ -800,6 +830,8 @@ CONFIG_SND_FM801=m
 #
 # USB support
 #
+CONFIG_USB_ARCH_HAS_HCD=y
+CONFIG_USB_ARCH_HAS_OHCI=y
 CONFIG_USB=m
 # CONFIG_USB_DEBUG is not set
 
@@ -811,8 +843,6 @@ CONFIG_USB_DEVICEFS=y
 # CONFIG_USB_DYNAMIC_MINORS is not set
 # CONFIG_USB_SUSPEND is not set
 # CONFIG_USB_OTG is not set
-CONFIG_USB_ARCH_HAS_HCD=y
-CONFIG_USB_ARCH_HAS_OHCI=y
 
 #
 # USB Host Controller Drivers
@@ -821,6 +851,8 @@ CONFIG_USB_EHCI_HCD=m
 # CONFIG_USB_EHCI_SPLIT_ISO is not set
 # CONFIG_USB_EHCI_ROOT_HUB_TT is not set
 CONFIG_USB_OHCI_HCD=m
+# CONFIG_USB_OHCI_BIG_ENDIAN is not set
+CONFIG_USB_OHCI_LITTLE_ENDIAN=y
 CONFIG_USB_UHCI_HCD=m
 # CONFIG_USB_SL811_HCD is not set
 
@@ -838,12 +870,11 @@ CONFIG_USB_UHCI_HCD=m
 #
 CONFIG_USB_STORAGE=m
 # CONFIG_USB_STORAGE_DEBUG is not set
-# CONFIG_USB_STORAGE_RW_DETECT is not set
 # CONFIG_USB_STORAGE_DATAFAB is not set
 # CONFIG_USB_STORAGE_FREECOM is not set
 # CONFIG_USB_STORAGE_ISD200 is not set
 # CONFIG_USB_STORAGE_DPCM is not set
-# CONFIG_USB_STORAGE_HP8200e is not set
+# CONFIG_USB_STORAGE_USBAT is not set
 # CONFIG_USB_STORAGE_SDDR09 is not set
 # CONFIG_USB_STORAGE_SDDR55 is not set
 # CONFIG_USB_STORAGE_JUMPSHOT is not set
@@ -893,6 +924,7 @@ CONFIG_USB_HIDINPUT=y
 # CONFIG_USB_PEGASUS is not set
 # CONFIG_USB_RTL8150 is not set
 # CONFIG_USB_USBNET is not set
+CONFIG_USB_MON=m
 
 #
 # USB port drivers
@@ -908,7 +940,6 @@ CONFIG_USB_HIDINPUT=y
 #
 # CONFIG_USB_EMI62 is not set
 # CONFIG_USB_EMI26 is not set
-# CONFIG_USB_TIGL is not set
 # CONFIG_USB_AUERSWALD is not set
 # CONFIG_USB_RIO500 is not set
 # CONFIG_USB_LEGOTOWER is not set
@@ -917,6 +948,8 @@ CONFIG_USB_HIDINPUT=y
 # CONFIG_USB_CYTHERM is not set
 # CONFIG_USB_PHIDGETKIT is not set
 # CONFIG_USB_PHIDGETSERVO is not set
+# CONFIG_USB_IDMOUSE is not set
+# CONFIG_USB_SISUSBVGA is not set
 # CONFIG_USB_TEST is not set
 
 #
@@ -964,7 +997,12 @@ CONFIG_REISERFS_FS_POSIX_ACL=y
 CONFIG_REISERFS_FS_SECURITY=y
 # CONFIG_JFS_FS is not set
 CONFIG_FS_POSIX_ACL=y
+
+#
+# XFS support
+#
 CONFIG_XFS_FS=y
+CONFIG_XFS_EXPORT=y
 # CONFIG_XFS_RT is not set
 # CONFIG_XFS_QUOTA is not set
 # CONFIG_XFS_SECURITY is not set
@@ -1042,7 +1080,7 @@ CONFIG_NFSD_V4=y
 CONFIG_NFSD_TCP=y
 CONFIG_LOCKD=m
 CONFIG_LOCKD_V4=y
-CONFIG_EXPORTFS=m
+CONFIG_EXPORTFS=y
 CONFIG_SUNRPC=m
 CONFIG_SUNRPC_GSS=m
 CONFIG_RPCSEC_GSS_KRB5=m
@@ -1128,6 +1166,8 @@ CONFIG_NLS_UTF8=m
 # CONFIG_CRC_CCITT is not set
 CONFIG_CRC32=y
 # CONFIG_LIBCRC32C is not set
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
 
 #
 # HP Simulator drivers
@@ -1144,14 +1184,17 @@ CONFIG_CRC32=y
 #
 # Kernel hacking
 #
+# CONFIG_PRINTK_TIME is not set
 CONFIG_DEBUG_KERNEL=y
 CONFIG_MAGIC_SYSRQ=y
+CONFIG_LOG_BUF_SHIFT=20
 # CONFIG_SCHEDSTATS is not set
 # CONFIG_DEBUG_SLAB is not set
 # CONFIG_DEBUG_SPINLOCK is not set
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_KOBJECT is not set
 # CONFIG_DEBUG_INFO is not set
+# CONFIG_DEBUG_FS is not set
 CONFIG_IA64_GRANULE_16MB=y
 # CONFIG_IA64_GRANULE_64MB is not set
 # CONFIG_IA64_PRINT_HAZARDS is not set
@@ -1178,6 +1221,7 @@ CONFIG_CRYPTO_MD5=m
 # CONFIG_CRYPTO_SHA256 is not set
 # CONFIG_CRYPTO_SHA512 is not set
 # CONFIG_CRYPTO_WP512 is not set
+# CONFIG_CRYPTO_TGR192 is not set
 CONFIG_CRYPTO_DES=m
 # CONFIG_CRYPTO_BLOWFISH is not set
 # CONFIG_CRYPTO_TWOFISH is not set
-- 
cgit v1.2.3-70-g09d2


From bd91c4bb134a3c5b04969f5ab960a3fd48778bcc Mon Sep 17 00:00:00 2001
From: Tony Luck <tony.luck@intel.com>
Date: Tue, 21 Jun 2005 14:16:00 -0700
Subject: [IA64] Refresh tiger_defconfig

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/configs/tiger_defconfig | 19 ++++++++++---------
 1 file changed, 10 insertions(+), 9 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/configs/tiger_defconfig b/arch/ia64/configs/tiger_defconfig
index 9086b789f6a..47f45341ac6 100644
--- a/arch/ia64/configs/tiger_defconfig
+++ b/arch/ia64/configs/tiger_defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.12-rc3
-# Tue May  3 15:55:04 2005
+# Linux kernel version: 2.6.12-20050621
+# Tue Jun 21 14:03:24 2005
 #
 
 #
@@ -67,6 +67,7 @@ CONFIG_GENERIC_CALIBRATE_DELAY=y
 CONFIG_TIME_INTERPOLATION=y
 CONFIG_EFI=y
 CONFIG_GENERIC_IOMAP=y
+CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y
 # CONFIG_IA64_GENERIC is not set
 CONFIG_IA64_DIG=y
 # CONFIG_IA64_HP_ZX1 is not set
@@ -285,6 +286,7 @@ CONFIG_CHR_DEV_ST=m
 CONFIG_BLK_DEV_SR=m
 # CONFIG_BLK_DEV_SR_VENDOR is not set
 CONFIG_CHR_DEV_SG=m
+# CONFIG_CHR_DEV_SCH is not set
 
 #
 # Some SCSI devices (e.g. CD jukebox) support multiple LUNs
@@ -313,11 +315,8 @@ CONFIG_SCSI_FC_ATTRS=y
 # CONFIG_MEGARAID_NEWGEN is not set
 # CONFIG_MEGARAID_LEGACY is not set
 # CONFIG_SCSI_SATA is not set
-# CONFIG_SCSI_BUSLOGIC is not set
 # CONFIG_SCSI_DMX3191D is not set
-# CONFIG_SCSI_EATA is not set
 # CONFIG_SCSI_FUTURE_DOMAIN is not set
-# CONFIG_SCSI_GDTH is not set
 # CONFIG_SCSI_IPS is not set
 # CONFIG_SCSI_INITIO is not set
 # CONFIG_SCSI_INIA100 is not set
@@ -366,8 +365,10 @@ CONFIG_DM_ZERO=m
 # Fusion MPT device support
 #
 CONFIG_FUSION=y
-CONFIG_FUSION_MAX_SGE=40
-# CONFIG_FUSION_CTL is not set
+CONFIG_FUSION_SPI=y
+CONFIG_FUSION_FC=y
+CONFIG_FUSION_MAX_SGE=128
+CONFIG_FUSION_CTL=y
 
 #
 # IEEE 1394 (FireWire) support
@@ -506,9 +507,11 @@ CONFIG_E1000=y
 # CONFIG_HAMACHI is not set
 # CONFIG_YELLOWFIN is not set
 # CONFIG_R8169 is not set
+# CONFIG_SKGE is not set
 # CONFIG_SK98LIN is not set
 # CONFIG_VIA_VELOCITY is not set
 CONFIG_TIGON3=y
+# CONFIG_BNX2 is not set
 
 #
 # Ethernet (10000 Mbit)
@@ -598,7 +601,6 @@ CONFIG_GAMEPORT=m
 # CONFIG_GAMEPORT_VORTEX is not set
 # CONFIG_GAMEPORT_FM801 is not set
 # CONFIG_GAMEPORT_CS461X is not set
-CONFIG_SOUND_GAMEPORT=m
 
 #
 # Character devices
@@ -611,7 +613,6 @@ CONFIG_SERIAL_NONSTANDARD=y
 # CONFIG_CYCLADES is not set
 # CONFIG_MOXA_SMARTIO is not set
 # CONFIG_ISI is not set
-# CONFIG_SYNCLINK is not set
 # CONFIG_SYNCLINKMP is not set
 # CONFIG_N_HDLC is not set
 # CONFIG_SPECIALIX is not set
-- 
cgit v1.2.3-70-g09d2


From a35f1e03b868a3179fa12344ec0d46a44fbf7574 Mon Sep 17 00:00:00 2001
From: Greg Edwards <edwardsg@sgi.com>
Date: Tue, 21 Jun 2005 11:30:00 -0700
Subject: [IA64] enable SGI simulator for generic kernels

Allow the SGI simulator (medusa) to work on generic kernels.  There is
no inherent dependency on an sn2-specific kernel.

Boot tested on Altix, medusa and HP rx2600.

Signed-off-by: Greg Edwards <edwardsg@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 3ad2c4af099..992bcfff791 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -216,7 +216,7 @@ config IOSAPIC
 
 config IA64_SGI_SN_SIM
 	bool "SGI Medusa Simulator Support"
-	depends on IA64_SGI_SN2
+	depends on IA64_SGI_SN2 || IA64_GENERIC
 	help
 	  If you are compiling a kernel that will run under SGI's IA-64
 	  simulator (Medusa) then say Y, otherwise say N.
-- 
cgit v1.2.3-70-g09d2


From 52a0de2cd2d3da8f90d88e2eccb63d0cadd6ae26 Mon Sep 17 00:00:00 2001
From: Christophe Lucas <clucas@rotomalug.org>
Date: Mon, 20 Jun 2005 14:50:00 -0700
Subject: [IA64] printk needs KERN_INFO arch/ia64/kernel/smp.c

printk() calls should include appropriate KERN_* constant.

Signed-off-by: Christophe Lucas <clucas@rotomalug.org>
Signed-off-by: Domen Puncer <domen@coderock.org>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/smp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/smp.c b/arch/ia64/kernel/smp.c
index 953095e2ce1..b49d4ddaab9 100644
--- a/arch/ia64/kernel/smp.c
+++ b/arch/ia64/kernel/smp.c
@@ -269,7 +269,7 @@ smp_call_function_single (int cpuid, void (*func) (void *info), void *info, int
 	int me = get_cpu(); /* prevent preemption and reschedule on another processor */
 
 	if (cpuid == me) {
-		printk("%s: trying to call self\n", __FUNCTION__);
+		printk(KERN_INFO "%s: trying to call self\n", __FUNCTION__);
 		put_cpu();
 		return -EBUSY;
 	}
-- 
cgit v1.2.3-70-g09d2


From 0393eed5c3220c9c3823a09a2d02329b8ff08b45 Mon Sep 17 00:00:00 2001
From: Ken Chen <kenneth.w.chen@intel.com>
Date: Tue, 21 Jun 2005 14:40:31 -0700
Subject: [IA64] fix nested_dtlb_miss handler for hugetlb address

The nested_dtlb_miss handler currently does not handle fault from
hugetlb address correctly.  It walks the page table assuming PAGE_SIZE.
Thus when taking a fault triggered from hugetlb address, it would not
calculate the pgd/pmd/pte address correctly and thus result an incorrect
invocation of ia64_do_page_fault().  In there, kernel will signal SIGBUS
and application dies (The faulting address is perfectly legal and we
have a valid pte for the corresponding user hugetlb address as well).
This patch fix the described kernel bug.  Since nested_dtlb_miss is a
rare event and a slow path anyway, I'm making the change without #ifdef
CONFIG_HUGETLB_PAGE for code readability.  Tony, please apply.

Signed-off-by: Ken Chen <kenneth.w.chen@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/ivt.S | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/ivt.S b/arch/ia64/kernel/ivt.S
index d9c05d53435..2bc085a73e3 100644
--- a/arch/ia64/kernel/ivt.S
+++ b/arch/ia64/kernel/ivt.S
@@ -405,17 +405,22 @@ ENTRY(nested_dtlb_miss)
 	 *		r30:	continuation address
 	 *		r31:	saved pr
 	 *
-	 * Clobbered:	b0, r18, r19, r21, psr.dt (cleared)
+	 * Clobbered:	b0, r18, r19, r21, r22, psr.dt (cleared)
 	 */
 	rsm psr.dt				// switch to using physical data addressing
 	mov r19=IA64_KR(PT_BASE)		// get the page table base address
 	shl r21=r16,3				// shift bit 60 into sign bit
+	mov r18=cr.itir
 	;;
 	shr.u r17=r16,61			// get the region number into r17
+	extr.u r18=r18,2,6			// get the faulting page size
 	;;
 	cmp.eq p6,p7=5,r17			// is faulting address in region 5?
-	shr.u r18=r16,PGDIR_SHIFT		// get bits 33-63 of faulting address
+	add r22=-PAGE_SHIFT,r18			// adjustment for hugetlb address
+	add r18=PGDIR_SHIFT-PAGE_SHIFT,r18
 	;;
+	shr.u r22=r16,r22
+	shr.u r18=r16,r18
 (p7)	dep r17=r17,r19,(PAGE_SHIFT-3),3	// put region number bits in place
 
 	srlz.d
@@ -428,7 +433,7 @@ ENTRY(nested_dtlb_miss)
 (p6)	dep r17=r18,r19,3,(PAGE_SHIFT-3)	// r17=PTA + IFA(33,42)*8
 (p7)	dep r17=r18,r17,3,(PAGE_SHIFT-6)	// r17=PTA + (((IFA(61,63) << 7) | IFA(33,39))*8)
 	cmp.eq p7,p6=0,r21			// unused address bits all zeroes?
-	shr.u r18=r16,PMD_SHIFT			// shift L2 index into position
+	shr.u r18=r22,PMD_SHIFT			// shift L2 index into position
 	;;
 	ld8 r17=[r17]				// fetch the L1 entry (may be 0)
 	;;
@@ -436,7 +441,7 @@ ENTRY(nested_dtlb_miss)
 	dep r17=r18,r17,3,(PAGE_SHIFT-3)	// compute address of L2 page table entry
 	;;
 (p7)	ld8 r17=[r17]				// fetch the L2 entry (may be 0)
-	shr.u r19=r16,PAGE_SHIFT		// shift L3 index into position
+	shr.u r19=r22,PAGE_SHIFT		// shift L3 index into position
 	;;
 (p7)	cmp.eq.or.andcm p6,p7=r17,r0		// was L2 entry NULL?
 	dep r17=r19,r17,3,(PAGE_SHIFT-3)	// compute address of L3 page table entry
-- 
cgit v1.2.3-70-g09d2


From 7b404b3459db1326bbdc7085edb1f58b98da5826 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Mon, 13 Jun 2005 11:27:00 -0700
Subject: [IA64] remove "pci=routeirq" option

Remove "pci=routeirq" option for ia64.  This was a workaround
after ACPI IRQ routing was changed from "all at boot for everything
in _PRT" to "do it when the device is enabled" in case there were
drivers that didn't use pci_enable_device().

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/pci/pci.c | 20 --------------------
 1 file changed, 20 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c
index 88641e5095b..e3fc4edea11 100644
--- a/arch/ia64/pci/pci.c
+++ b/arch/ia64/pci/pci.c
@@ -33,8 +33,6 @@
 #include <asm/hw_irq.h>
 
 
-static int pci_routeirq;
-
 /*
  * Low-level SAL-based PCI configuration access functions. Note that SAL
  * calls are already serialized (via sal_lock), so we don't need another
@@ -139,24 +137,8 @@ static void acpi_map_iosapics(void)
 static int __init
 pci_acpi_init (void)
 {
-	struct pci_dev *dev = NULL;
-
-	printk(KERN_INFO "PCI: Using ACPI for IRQ routing\n");
-
 	acpi_map_iosapics();
 
-	if (pci_routeirq) {
-		/*
-		 * PCI IRQ routing is set up by pci_enable_device(), but we
-		 * also do it here in case there are still broken drivers that
-		 * don't use pci_enable_device().
-		 */
-		printk(KERN_INFO "PCI: Routing interrupts for all devices because \"pci=routeirq\" specified\n");
-		for_each_pci_dev(dev)
-			acpi_pci_irq_enable(dev);
-	} else
-		printk(KERN_INFO "PCI: If a device doesn't work, try \"pci=routeirq\".  If it helps, post a report\n");
-
 	return 0;
 }
 
@@ -500,8 +482,6 @@ pcibios_align_resource (void *data, struct resource *res,
 char * __init
 pcibios_setup (char *str)
 {
-	if (!strcmp(str, "routeirq"))
-		pci_routeirq = 1;
 	return NULL;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 8005aba69a6440a535a4cc2aed99ffca580847e0 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 21 Jun 2005 15:39:22 -0700
Subject: [SPARC64]: Fix cmsg length checks in Solaris emulation layer.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc64/solaris/socket.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/sparc64/solaris/socket.c b/arch/sparc64/solaris/socket.c
index ec8e074c4ea..06740582717 100644
--- a/arch/sparc64/solaris/socket.c
+++ b/arch/sparc64/solaris/socket.c
@@ -317,8 +317,10 @@ asmlinkage int solaris_sendmsg(int fd, struct sol_nmsghdr __user *user_msg, unsi
 		unsigned long *kcmsg;
 		compat_size_t cmlen;
 
-		if(kern_msg.msg_controllen > sizeof(ctl) &&
-		   kern_msg.msg_controllen <= 256) {
+		if (kern_msg.msg_controllen <= sizeof(compat_size_t))
+			return -EINVAL;
+
+		if(kern_msg.msg_controllen > sizeof(ctl)) {
 			err = -ENOBUFS;
 			ctl_buf = kmalloc(kern_msg.msg_controllen, GFP_KERNEL);
 			if(!ctl_buf)
-- 
cgit v1.2.3-70-g09d2


From 4ea78729b8dbfc400fe165a57b90a394a7275a54 Mon Sep 17 00:00:00 2001
From: Matthew Chapman <matthewc@hp.com>
Date: Tue, 21 Jun 2005 16:19:20 -0700
Subject: [IA64] ptrace and restore_sigcontext() allow ar.rsc.pl==0

This patch fixes handling of accesses to ar.rsc via ptrace & restore_sigcontext
[With Thanks to Chris Wright for noticing the restore_sigcontext path]

Signed-off-by: Matthew Chapman <matthewc@hp.com>
Acked-by: David Mosberger <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/ptrace.c | 15 ++++++++++-----
 arch/ia64/kernel/signal.c |  5 +++--
 2 files changed, 13 insertions(+), 7 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/ptrace.c b/arch/ia64/kernel/ptrace.c
index 575a8f657b3..6d57aebad48 100644
--- a/arch/ia64/kernel/ptrace.c
+++ b/arch/ia64/kernel/ptrace.c
@@ -945,6 +945,13 @@ access_uarea (struct task_struct *child, unsigned long addr,
 				*data = (pt->cr_ipsr & IPSR_MASK);
 			return 0;
 
+		      case PT_AR_RSC:
+			if (write_access)
+				pt->ar_rsc = *data | (3 << 2); /* force PL3 */
+			else
+				*data = pt->ar_rsc;
+			return 0;
+
 		      case PT_AR_RNAT:
 			urbs_end = ia64_get_user_rbs_end(child, pt, NULL);
 			rnat_addr = (long) ia64_rse_rnat_addr((long *)
@@ -996,9 +1003,6 @@ access_uarea (struct task_struct *child, unsigned long addr,
 		      case PT_AR_BSPSTORE:
 			ptr = pt_reg_addr(pt, ar_bspstore);
 			break;
-		      case PT_AR_RSC:
-			ptr = pt_reg_addr(pt, ar_rsc);
-			break;
 		      case PT_AR_UNAT:
 			ptr = pt_reg_addr(pt, ar_unat);
 			break;
@@ -1234,7 +1238,7 @@ ptrace_getregs (struct task_struct *child, struct pt_all_user_regs __user *ppr)
 static long
 ptrace_setregs (struct task_struct *child, struct pt_all_user_regs __user *ppr)
 {
-	unsigned long psr, ec, lc, rnat, bsp, cfm, nat_bits, val = 0;
+	unsigned long psr, rsc, ec, lc, rnat, bsp, cfm, nat_bits, val = 0;
 	struct unw_frame_info info;
 	struct switch_stack *sw;
 	struct ia64_fpreg fpval;
@@ -1267,7 +1271,7 @@ ptrace_setregs (struct task_struct *child, struct pt_all_user_regs __user *ppr)
 	/* app regs */
 
 	retval |= __get_user(pt->ar_pfs, &ppr->ar[PT_AUR_PFS]);
-	retval |= __get_user(pt->ar_rsc, &ppr->ar[PT_AUR_RSC]);
+	retval |= __get_user(rsc, &ppr->ar[PT_AUR_RSC]);
 	retval |= __get_user(pt->ar_bspstore, &ppr->ar[PT_AUR_BSPSTORE]);
 	retval |= __get_user(pt->ar_unat, &ppr->ar[PT_AUR_UNAT]);
 	retval |= __get_user(pt->ar_ccv, &ppr->ar[PT_AUR_CCV]);
@@ -1365,6 +1369,7 @@ ptrace_setregs (struct task_struct *child, struct pt_all_user_regs __user *ppr)
 	retval |= __get_user(nat_bits, &ppr->nat);
 
 	retval |= access_uarea(child, PT_CR_IPSR, &psr, 1);
+	retval |= access_uarea(child, PT_AR_RSC, &rsc, 1);
 	retval |= access_uarea(child, PT_AR_EC, &ec, 1);
 	retval |= access_uarea(child, PT_AR_LC, &lc, 1);
 	retval |= access_uarea(child, PT_AR_RNAT, &rnat, 1);
diff --git a/arch/ia64/kernel/signal.c b/arch/ia64/kernel/signal.c
index 499b7e5317c..edd9f07860b 100644
--- a/arch/ia64/kernel/signal.c
+++ b/arch/ia64/kernel/signal.c
@@ -94,7 +94,7 @@ sys_sigaltstack (const stack_t __user *uss, stack_t __user *uoss, long arg2,
 static long
 restore_sigcontext (struct sigcontext __user *sc, struct sigscratch *scr)
 {
-	unsigned long ip, flags, nat, um, cfm;
+	unsigned long ip, flags, nat, um, cfm, rsc;
 	long err;
 
 	/* Always make any pending restarted system calls return -EINTR */
@@ -106,7 +106,7 @@ restore_sigcontext (struct sigcontext __user *sc, struct sigscratch *scr)
 	err |= __get_user(ip, &sc->sc_ip);			/* instruction pointer */
 	err |= __get_user(cfm, &sc->sc_cfm);
 	err |= __get_user(um, &sc->sc_um);			/* user mask */
-	err |= __get_user(scr->pt.ar_rsc, &sc->sc_ar_rsc);
+	err |= __get_user(rsc, &sc->sc_ar_rsc);
 	err |= __get_user(scr->pt.ar_unat, &sc->sc_ar_unat);
 	err |= __get_user(scr->pt.ar_fpsr, &sc->sc_ar_fpsr);
 	err |= __get_user(scr->pt.ar_pfs, &sc->sc_ar_pfs);
@@ -119,6 +119,7 @@ restore_sigcontext (struct sigcontext __user *sc, struct sigscratch *scr)
 	err |= __copy_from_user(&scr->pt.r15, &sc->sc_gr[15], 8);	/* r15 */
 
 	scr->pt.cr_ifs = cfm | (1UL << 63);
+	scr->pt.ar_rsc = rsc | (3 << 2); /* force PL3 */
 
 	/* establish new instruction pointer: */
 	scr->pt.cr_iip = ip & ~0x3UL;
-- 
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 'arch')

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 'arch')

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 'arch')

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 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 'arch')

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 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 'arch')

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 'arch')

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 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 'arch')

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 'arch')

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 2894801db1cea527885f7a8679e5a88fb7377d9c Mon Sep 17 00:00:00 2001
From: Coywolf Qi Hunt <coywolf@lovecn.org>
Date: Tue, 21 Jun 2005 17:15:15 -0700
Subject: [PATCH] kbuild: display compile version

I am always trying to make sure I've booted the right kernel after a new
install.  Too paranoid maybe.  But I guess there're other people like me.
So let's make kbuild display the compile version number at the end to give
us a hint.  I know we may be booting vmlinux someday, but don't care about
it for now.

Signed-off-by: Coywolf Qi Hunt <coywolf@lovecn.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/boot/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/i386/boot/Makefile b/arch/i386/boot/Makefile
index aa7064a75ee..43cd6220ee4 100644
--- a/arch/i386/boot/Makefile
+++ b/arch/i386/boot/Makefile
@@ -48,7 +48,7 @@ cmd_image = $(obj)/tools/build $(BUILDFLAGS) $(obj)/bootsect $(obj)/setup \
 $(obj)/zImage $(obj)/bzImage: $(obj)/bootsect $(obj)/setup \
 			      $(obj)/vmlinux.bin $(obj)/tools/build FORCE
 	$(call if_changed,image)
-	@echo 'Kernel: $@ is ready'
+	@echo 'Kernel: $@ is ready' ' (#'`cat .version`')'
 
 $(obj)/vmlinux.bin: $(obj)/compressed/vmlinux FORCE
 	$(call if_changed,objcopy)
-- 
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 'arch')

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 'arch')

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 c93fcff6952068d484f2f7dae187037f46235647 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Tue, 21 Jun 2005 17:15:20 -0700
Subject: [PATCH] ppc32: Removed dependency on CONFIG_CPM2 for building
 mpc85xx_device.c

Previously we needed CONFIG_CPM2 enabled to get the proper IRQ ifdef's for
CPM interrupts.  Recent changes have caused that to be no longer necessary.

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/syslib/mpc85xx_devices.c | 2 --
 1 file changed, 2 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc/syslib/mpc85xx_devices.c b/arch/ppc/syslib/mpc85xx_devices.c
index b30c57d4518..8af322dd476 100644
--- a/arch/ppc/syslib/mpc85xx_devices.c
+++ b/arch/ppc/syslib/mpc85xx_devices.c
@@ -321,7 +321,6 @@ struct platform_device ppc_sys_platform_devices[] = {
 			},
 		},
 	},
-#ifdef CONFIG_CPM2
 	[MPC85xx_CPM_FCC1] = {
 		.name = "fsl-cpm-fcc",
 		.id	= 1,
@@ -575,7 +574,6 @@ struct platform_device ppc_sys_platform_devices[] = {
 			},
 		},
 	},
-#endif /* CONFIG_CPM2 */
 	[MPC85xx_eTSEC1] = {
 		.name = "fsl-gianfar",
 		.id	= 1,
-- 
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 'arch')

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 1e5aa8c86571deea5d8e8462fedcf610e21af717 Mon Sep 17 00:00:00 2001
From: Matt Porter <mporter@kernel.crashing.org>
Date: Tue, 21 Jun 2005 17:15:22 -0700
Subject: [PATCH] ppc32: fix CONFIG_TASK_SIZE handling on 40x

This patch is virtually identical to my previous 44x one.  It removes
0x8000'0000 TASK_SIZE hardcoded assumption from head_4xx.S.

Signed-off-by: Eugene Surovegin <ebs@ebshome.net>
Signed-off-by: Matt Porter <mporter@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/kernel/head_4xx.S | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc/kernel/head_4xx.S b/arch/ppc/kernel/head_4xx.S
index 6f5d380e234..23fb51819ba 100644
--- a/arch/ppc/kernel/head_4xx.S
+++ b/arch/ppc/kernel/head_4xx.S
@@ -291,8 +291,9 @@ label:
 	/* If we are faulting a kernel address, we have to use the
 	 * kernel page tables.
 	 */
-	andis.	r11, r10, 0x8000
-	beq	3f
+	lis	r11, TASK_SIZE@h
+	cmplw	r10, r11
+	blt+	3f
 	lis	r11, swapper_pg_dir@h
 	ori	r11, r11, swapper_pg_dir@l
 	li	r9, 0
@@ -479,8 +480,9 @@ label:
 	/* If we are faulting a kernel address, we have to use the
 	 * kernel page tables.
 	 */
-	andis.	r11, r10, 0x8000
-	beq	3f
+	lis	r11, TASK_SIZE@h
+	cmplw	r10, r11
+	blt+	3f
 	lis	r11, swapper_pg_dir@h
 	ori	r11, r11, swapper_pg_dir@l
 	li	r9, 0
@@ -578,8 +580,9 @@ label:
 	/* If we are faulting a kernel address, we have to use the
 	 * kernel page tables.
 	 */
-	andis.	r11, r10, 0x8000
-	beq	3f
+	lis	r11, TASK_SIZE@h
+	cmplw	r10, r11
+	blt+	3f
 	lis	r11, swapper_pg_dir@h
 	ori	r11, r11, swapper_pg_dir@l
 	li	r9, 0
-- 
cgit v1.2.3-70-g09d2


From 682afbbd1456e9e7b64f482eb05676b933173cdf Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Tue, 21 Jun 2005 17:15:23 -0700
Subject: [PATCH] ppc32: Add support for MPC8245 8250 serial ports on Sandpoint

Added platform device initialization for the two 8250 style UARTs that
exist on the MPC8245.  Additionally, updated the Sandpoint code to enable
one of these UARTs if an MPC8245 is connected to it.

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/platforms/sandpoint.c  | 20 ++++++++++++++++++++
 arch/ppc/syslib/mpc10x_common.c | 20 ++++++++++++++++++++
 2 files changed, 40 insertions(+)

(limited to 'arch')

diff --git a/arch/ppc/platforms/sandpoint.c b/arch/ppc/platforms/sandpoint.c
index 531bfa0e451..70e58f43f2b 100644
--- a/arch/ppc/platforms/sandpoint.c
+++ b/arch/ppc/platforms/sandpoint.c
@@ -81,6 +81,7 @@
 #include <linux/serial.h>
 #include <linux/tty.h>	/* for linux/serial_core.h */
 #include <linux/serial_core.h>
+#include <linux/serial_8250.h>
 
 #include <asm/system.h>
 #include <asm/pgtable.h>
@@ -99,6 +100,7 @@
 #include <asm/mpc10x.h>
 #include <asm/pci-bridge.h>
 #include <asm/kgdb.h>
+#include <asm/ppc_sys.h>
 
 #include "sandpoint.h"
 
@@ -305,6 +307,24 @@ sandpoint_setup_arch(void)
 	/* Lookup PCI host bridges */
 	sandpoint_find_bridges();
 
+	if (strncmp (cur_ppc_sys_spec->ppc_sys_name, "8245", 4) == 0)
+	{
+		bd_t *bp = (bd_t *)__res;
+		struct plat_serial8250_port *pdata;
+		pdata = (struct plat_serial8250_port *) ppc_sys_get_pdata(MPC10X_DUART);
+
+		if (pdata)
+		{
+			pdata[0].uartclk = bp->bi_busfreq;
+			pdata[0].membase = ioremap(pdata[0].mapbase, 0x100);
+
+			/* this disables the 2nd serial port on the DUART
+			 * since the sandpoint does not have it connected */
+			pdata[1].uartclk = 0;
+			pdata[1].irq = 0;
+			pdata[1].mapbase = 0;
+		}
+
 	printk(KERN_INFO "Motorola SPS Sandpoint Test Platform\n");
 	printk(KERN_INFO "Port by MontaVista Software, Inc. (source@mvista.com)\n");
 
diff --git a/arch/ppc/syslib/mpc10x_common.c b/arch/ppc/syslib/mpc10x_common.c
index 0130aeb4a2d..8fc5f415452 100644
--- a/arch/ppc/syslib/mpc10x_common.c
+++ b/arch/ppc/syslib/mpc10x_common.c
@@ -44,10 +44,12 @@
 #define MPC10X_I2C_IRQ (EPIC_IRQ_BASE + NUM_8259_INTERRUPTS)
 #define MPC10X_DMA0_IRQ (EPIC_IRQ_BASE + 1 + NUM_8259_INTERRUPTS)
 #define MPC10X_DMA1_IRQ (EPIC_IRQ_BASE + 2 + NUM_8259_INTERRUPTS)
+#define MPC10X_UART0_IRQ (EPIC_IRQ_BASE + 4 + NUM_8259_INTERRUPTS)
 #else
 #define MPC10X_I2C_IRQ -1
 #define MPC10X_DMA0_IRQ -1
 #define MPC10X_DMA1_IRQ -1
+#define MPC10X_UART0_IRQ -1
 #endif
 
 static struct fsl_i2c_platform_data mpc10x_i2c_pdata = {
@@ -55,6 +57,16 @@ static struct fsl_i2c_platform_data mpc10x_i2c_pdata = {
 };
 
 static struct plat_serial8250_port serial_platform_data[] = {
+	[0] = {
+		.mapbase	= 0x4500,
+		.iotype		= UPIO_MEM,
+		.flags		= UPF_BOOT_AUTOCONF | UPF_SKIP_TEST,
+	},
+	[1] = {
+		.mapbase	= 0x4600,
+		.iotype		= UPIO_MEM,
+		.flags		= UPF_BOOT_AUTOCONF | UPF_SKIP_TEST,
+	},
 	{ },
 };
 
@@ -399,6 +411,12 @@ mpc10x_bridge_init(struct pci_controller *hose,
 	ppc_sys_platform_devices[MPC10X_DMA1].resource[1].start = MPC10X_DMA1_IRQ;
 	ppc_sys_platform_devices[MPC10X_DMA1].resource[1].end = MPC10X_DMA1_IRQ;
 
+	serial_platform_data[0].mapbase += phys_eumb_base;
+	serial_platform_data[0].irq = MPC10X_UART0_IRQ;
+
+	serial_platform_data[1].mapbase += phys_eumb_base;
+	serial_platform_data[1].irq = MPC10X_UART0_IRQ + 1;
+
 	/*
 	 * 8240 erratum 26, 8241/8245 erratum 29, 107 erratum 23: speculative
 	 * PCI reads may return stale data so turn off.
@@ -597,6 +615,8 @@ void __init mpc10x_set_openpic(void)
 	openpic_set_sources(EPIC_IRQ_BASE, 3, OpenPIC_Addr + 0x11020);
 	/* Skip reserved space and map Message Unit Interrupt (I2O) */
 	openpic_set_sources(EPIC_IRQ_BASE + 3, 1, OpenPIC_Addr + 0x110C0);
+	/* Skip reserved space and map Serial Interupts */
+	openpic_set_sources(EPIC_IRQ_BASE + 4, 2, OpenPIC_Addr + 0x11120);
 
 	openpic_init(NUM_8259_INTERRUPTS);
 }
-- 
cgit v1.2.3-70-g09d2


From eee4146ab908188e556641f1cd7a10e894d10edb Mon Sep 17 00:00:00 2001
From: Matt Porter <mporter@kernel.crashing.org>
Date: Tue, 21 Jun 2005 17:15:24 -0700
Subject: [PATCH] ppc32: remove orphaned ppc4xx_kgdb.c

Removes ppc4xx_kgdb.c which is no longer being used.  Pointed out by Andrei
Konovalov.

Signed-off-by: Matt Porter <mporter@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/syslib/ppc4xx_kgdb.c | 124 ------------------------------------------
 1 file changed, 124 deletions(-)
 delete mode 100644 arch/ppc/syslib/ppc4xx_kgdb.c

(limited to 'arch')

diff --git a/arch/ppc/syslib/ppc4xx_kgdb.c b/arch/ppc/syslib/ppc4xx_kgdb.c
deleted file mode 100644
index fe8668bf813..00000000000
--- a/arch/ppc/syslib/ppc4xx_kgdb.c
+++ /dev/null
@@ -1,124 +0,0 @@
-#include <linux/config.h>
-#include <linux/types.h>
-#include <asm/ibm4xx.h>
-#include <linux/kernel.h>
-
-
-
-#define LSR_DR		0x01 /* Data ready */
-#define LSR_OE		0x02 /* Overrun */
-#define LSR_PE		0x04 /* Parity error */
-#define LSR_FE		0x08 /* Framing error */
-#define LSR_BI		0x10 /* Break */
-#define LSR_THRE	0x20 /* Xmit holding register empty */
-#define LSR_TEMT	0x40 /* Xmitter empty */
-#define LSR_ERR		0x80 /* Error */
-
-#include <platforms/4xx/ibm_ocp.h>
-
-extern struct NS16550* COM_PORTS[];
-#ifndef NULL
-#define NULL 0x00
-#endif
-
-static volatile struct NS16550 *kgdb_debugport = NULL;
-
-volatile struct NS16550 *
-NS16550_init(int chan)
-{
-	volatile struct NS16550 *com_port;
-	int quot;
-#ifdef BASE_BAUD
-	quot = BASE_BAUD / 9600;
-#else
-	quot = 0x000c; /* 0xc = 9600 baud (on a pc) */
-#endif
-
-	com_port = (struct NS16550 *) COM_PORTS[chan];
-
-	com_port->lcr = 0x00;
-	com_port->ier = 0xFF;
-	com_port->ier = 0x00;
-	com_port->lcr = com_port->lcr | 0x80; /* Access baud rate */
-	com_port->dll = ( quot & 0x00ff ); /* 0xc = 9600 baud */
-	com_port->dlm = ( quot & 0xff00 ) >> 8;
-	com_port->lcr = 0x03; /* 8 data, 1 stop, no parity */
-	com_port->mcr = 0x00; /* RTS/DTR */
-	com_port->fcr = 0x07; /* Clear & enable FIFOs */
-
-	return( com_port );
-}
-
-
-void
-NS16550_putc(volatile struct NS16550 *com_port, unsigned char c)
-{
-	while ((com_port->lsr & LSR_THRE) == 0)
-		;
-	com_port->thr = c;
-	return;
-}
-
-unsigned char
-NS16550_getc(volatile struct NS16550 *com_port)
-{
-	while ((com_port->lsr & LSR_DR) == 0)
-		;
-	return (com_port->rbr);
-}
-
-unsigned char
-NS16550_tstc(volatile struct NS16550 *com_port)
-{
-	return ((com_port->lsr & LSR_DR) != 0);
-}
-
-
-#if defined(CONFIG_KGDB_TTYS0)
-#define KGDB_PORT 0
-#elif defined(CONFIG_KGDB_TTYS1)
-#define KGDB_PORT 1
-#elif defined(CONFIG_KGDB_TTYS2)
-#define KGDB_PORT 2
-#elif defined(CONFIG_KGDB_TTYS3)
-#define KGDB_PORT 3
-#else
-#error "invalid kgdb_tty port"
-#endif
-
-void putDebugChar( unsigned char c )
-{
-	if ( kgdb_debugport == NULL )
-		kgdb_debugport = NS16550_init(KGDB_PORT);
-	NS16550_putc( kgdb_debugport, c );
-}
-
-int getDebugChar( void )
-{
-	if (kgdb_debugport == NULL)
-		kgdb_debugport = NS16550_init(KGDB_PORT);
-
-	return(NS16550_getc(kgdb_debugport));
-}
-
-void kgdb_interruptible(int enable)
-{
-	return;
-}
-
-void putDebugString(char* str)
-{
-	while (*str != '\0') {
-		putDebugChar(*str);
-		str++;
-	}
-	putDebugChar('\r');
-	return;
-}
-
-void
-kgdb_map_scc(void)
-{
-	printk("kgdb init \n");
-	kgdb_debugport = NS16550_init(KGDB_PORT);
-}
-- 
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 'arch')

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 'arch')

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 1492ec8069ea6f82bc32df27cabbec72e554e2b8 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Tue, 21 Jun 2005 17:15:27 -0700
Subject: [PATCH] ppc32: Factor out common exception code into macro's for
 4xx/Book-E

4xx and Book-E PPC's have several exception levels.  The code to handle
each level is fairly regular.  Turning the code into macro's will ease the
handling of future exception levels (debug) in forth coming chips.

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/entry.S      | 164 +++++++++++++++++--------------------------
 arch/ppc/kernel/head_booke.h |  94 +++++++------------------
 2 files changed, 87 insertions(+), 171 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc/kernel/entry.S b/arch/ppc/kernel/entry.S
index 661523707e8..8377b6ca26d 100644
--- a/arch/ppc/kernel/entry.S
+++ b/arch/ppc/kernel/entry.S
@@ -46,26 +46,23 @@
 
 #ifdef CONFIG_BOOKE
 #include "head_booke.h"
+#define TRANSFER_TO_HANDLER_EXC_LEVEL(exc_level)	\
+	mtspr	exc_level##_SPRG,r8;			\
+	BOOKE_LOAD_EXC_LEVEL_STACK(exc_level);		\
+	lwz	r0,GPR10-INT_FRAME_SIZE(r8);		\
+	stw	r0,GPR10(r11);				\
+	lwz	r0,GPR11-INT_FRAME_SIZE(r8);		\
+	stw	r0,GPR11(r11);				\
+	mfspr	r8,exc_level##_SPRG
+
 	.globl	mcheck_transfer_to_handler
 mcheck_transfer_to_handler:
-	mtspr	MCHECK_SPRG,r8
-	BOOKE_LOAD_MCHECK_STACK
-	lwz	r0,GPR10-INT_FRAME_SIZE(r8)
-	stw	r0,GPR10(r11)
-	lwz	r0,GPR11-INT_FRAME_SIZE(r8)
-	stw	r0,GPR11(r11)
-	mfspr	r8,MCHECK_SPRG
+	TRANSFER_TO_HANDLER_EXC_LEVEL(MCHECK)
 	b	transfer_to_handler_full
 
 	.globl	crit_transfer_to_handler
 crit_transfer_to_handler:
-	mtspr	CRIT_SPRG,r8
-	BOOKE_LOAD_CRIT_STACK
-	lwz	r0,GPR10-INT_FRAME_SIZE(r8)
-	stw	r0,GPR10(r11)
-	lwz	r0,GPR11-INT_FRAME_SIZE(r8)
-	stw	r0,GPR11(r11)
-	mfspr	r8,CRIT_SPRG
+	TRANSFER_TO_HANDLER_EXC_LEVEL(CRIT)
 	/* fall through */
 #endif
 
@@ -783,99 +780,64 @@ exc_exit_restart_end:
  * time of the critical interrupt.
  *
  */
-	.globl	ret_from_crit_exc
-ret_from_crit_exc:
-	REST_NVGPRS(r1)
-	lwz	r3,_MSR(r1)
-	andi.	r3,r3,MSR_PR
-	LOAD_MSR_KERNEL(r10,MSR_KERNEL)
-	bne	user_exc_return
-
-	lwz	r0,GPR0(r1)
-	lwz	r2,GPR2(r1)
-	REST_4GPRS(3, r1)
-	REST_2GPRS(7, r1)
-
-	lwz	r10,_XER(r1)
-	lwz	r11,_CTR(r1)
-	mtspr	SPRN_XER,r10
-	mtctr	r11
-
-	PPC405_ERR77(0,r1)
-	stwcx.	r0,0,r1			/* to clear the reservation */
-
-	lwz	r11,_LINK(r1)
-	mtlr	r11
-	lwz	r10,_CCR(r1)
-	mtcrf	0xff,r10
 #ifdef CONFIG_40x
-	/* avoid any possible TLB misses here by turning off MSR.DR, we
-	 * assume the instructions here are mapped by a pinned TLB entry */
-	li	r10,MSR_IR
-	mtmsr	r10
-	isync
-	tophys(r1, r1)
+#define PPC_40x_TURN_OFF_MSR_DR						    \
+	/* avoid any possible TLB misses here by turning off MSR.DR, we	    \
+	 * assume the instructions here are mapped by a pinned TLB entry */ \
+	li	r10,MSR_IR;						    \
+	mtmsr	r10;							    \
+	isync;								    \
+	tophys(r1, r1);
+#else
+#define PPC_40x_TURN_OFF_MSR_DR
 #endif
-	lwz	r9,_DEAR(r1)
-	lwz	r10,_ESR(r1)
-	mtspr	SPRN_DEAR,r9
-	mtspr	SPRN_ESR,r10
-	lwz	r11,_NIP(r1)
-	lwz	r12,_MSR(r1)
-	mtspr	SPRN_CSRR0,r11
-	mtspr	SPRN_CSRR1,r12
-	lwz	r9,GPR9(r1)
-	lwz	r12,GPR12(r1)
-	lwz	r10,GPR10(r1)
-	lwz	r11,GPR11(r1)
-	lwz	r1,GPR1(r1)
-	PPC405_ERR77_SYNC
-	rfci
-	b	.		/* prevent prefetch past rfci */
+
+#define RET_FROM_EXC_LEVEL(exc_lvl_srr0, exc_lvl_srr1, exc_lvl_rfi)	\
+	REST_NVGPRS(r1);						\
+	lwz	r3,_MSR(r1);						\
+	andi.	r3,r3,MSR_PR;						\
+	LOAD_MSR_KERNEL(r10,MSR_KERNEL);				\
+	bne	user_exc_return;					\
+	lwz	r0,GPR0(r1);						\
+	lwz	r2,GPR2(r1);						\
+	REST_4GPRS(3, r1);						\
+	REST_2GPRS(7, r1);						\
+	lwz	r10,_XER(r1);						\
+	lwz	r11,_CTR(r1);						\
+	mtspr	SPRN_XER,r10;						\
+	mtctr	r11;							\
+	PPC405_ERR77(0,r1);						\
+	stwcx.	r0,0,r1;		/* to clear the reservation */	\
+	lwz	r11,_LINK(r1);						\
+	mtlr	r11;							\
+	lwz	r10,_CCR(r1);						\
+	mtcrf	0xff,r10;						\
+	PPC_40x_TURN_OFF_MSR_DR;					\
+	lwz	r9,_DEAR(r1);						\
+	lwz	r10,_ESR(r1);						\
+	mtspr	SPRN_DEAR,r9;						\
+	mtspr	SPRN_ESR,r10;						\
+	lwz	r11,_NIP(r1);						\
+	lwz	r12,_MSR(r1);						\
+	mtspr	exc_lvl_srr0,r11;					\
+	mtspr	exc_lvl_srr1,r12;					\
+	lwz	r9,GPR9(r1);						\
+	lwz	r12,GPR12(r1);						\
+	lwz	r10,GPR10(r1);						\
+	lwz	r11,GPR11(r1);						\
+	lwz	r1,GPR1(r1);						\
+	PPC405_ERR77_SYNC;						\
+	exc_lvl_rfi;							\
+	b	.;		/* prevent prefetch past exc_lvl_rfi */
+
+	.globl	ret_from_crit_exc
+ret_from_crit_exc:
+	RET_FROM_EXC_LEVEL(SPRN_CSRR0, SPRN_CSRR1, RFCI)
 
 #ifdef CONFIG_BOOKE
-/*
- * Return from a machine check interrupt, similar to a critical
- * interrupt.
- */
 	.globl	ret_from_mcheck_exc
 ret_from_mcheck_exc:
-	REST_NVGPRS(r1)
-	lwz	r3,_MSR(r1)
-	andi.	r3,r3,MSR_PR
-	LOAD_MSR_KERNEL(r10,MSR_KERNEL)
-	bne	user_exc_return
-
-	lwz	r0,GPR0(r1)
-	lwz	r2,GPR2(r1)
-	REST_4GPRS(3, r1)
-	REST_2GPRS(7, r1)
-
-	lwz	r10,_XER(r1)
-	lwz	r11,_CTR(r1)
-	mtspr	SPRN_XER,r10
-	mtctr	r11
-
-	stwcx.	r0,0,r1			/* to clear the reservation */
-
-	lwz	r11,_LINK(r1)
-	mtlr	r11
-	lwz	r10,_CCR(r1)
-	mtcrf	0xff,r10
-	lwz	r9,_DEAR(r1)
-	lwz	r10,_ESR(r1)
-	mtspr	SPRN_DEAR,r9
-	mtspr	SPRN_ESR,r10
-	lwz	r11,_NIP(r1)
-	lwz	r12,_MSR(r1)
-	mtspr	SPRN_MCSRR0,r11
-	mtspr	SPRN_MCSRR1,r12
-	lwz	r9,GPR9(r1)
-	lwz	r12,GPR12(r1)
-	lwz	r10,GPR10(r1)
-	lwz	r11,GPR11(r1)
-	lwz	r1,GPR1(r1)
-	RFMCI
+	RET_FROM_EXC_LEVEL(SPRN_MCSRR0, SPRN_MCSRR1, RFMCI)
 #endif /* CONFIG_BOOKE */
 
 /*
diff --git a/arch/ppc/kernel/head_booke.h b/arch/ppc/kernel/head_booke.h
index f213d12eec0..9c50f9d2657 100644
--- a/arch/ppc/kernel/head_booke.h
+++ b/arch/ppc/kernel/head_booke.h
@@ -67,46 +67,36 @@
 #define CRIT_STACK_TOP		(exception_stack_top)
 
 #ifdef CONFIG_SMP
-#define BOOKE_LOAD_CRIT_STACK				\
+#define BOOKE_LOAD_EXC_LEVEL_STACK(level)		\
 	mfspr	r8,SPRN_PIR;				\
 	mulli	r8,r8,BOOKE_EXCEPTION_STACK_SIZE;	\
 	neg	r8,r8;					\
-	addis	r8,r8,CRIT_STACK_TOP@ha;		\
-	addi	r8,r8,CRIT_STACK_TOP@l
-#define BOOKE_LOAD_MCHECK_STACK				\
-	mfspr	r8,SPRN_PIR;				\
-	mulli	r8,r8,BOOKE_EXCEPTION_STACK_SIZE;	\
-	neg	r8,r8;					\
-	addis	r8,r8,MCHECK_STACK_TOP@ha;		\
-	addi	r8,r8,MCHECK_STACK_TOP@l
+	addis	r8,r8,level##_STACK_TOP@ha;		\
+	addi	r8,r8,level##_STACK_TOP@l
 #else
-#define BOOKE_LOAD_CRIT_STACK				\
-	lis	r8,CRIT_STACK_TOP@h;			\
-	ori	r8,r8,CRIT_STACK_TOP@l
-#define BOOKE_LOAD_MCHECK_STACK				\
-	lis	r8,MCHECK_STACK_TOP@h;			\
-	ori	r8,r8,MCHECK_STACK_TOP@l
+#define BOOKE_LOAD_EXC_LEVEL_STACK(level)		\
+	lis	r8,level##_STACK_TOP@h;			\
+	ori	r8,r8,level##_STACK_TOP@l
 #endif
 
 /*
- * Exception prolog for critical exceptions.  This is a little different
- * from the normal exception prolog above since a critical exception
- * can potentially occur at any point during normal exception processing.
- * Thus we cannot use the same SPRG registers as the normal prolog above.
- * Instead we use a portion of the critical exception stack at low physical
- * addresses.
+ * Exception prolog for critical/machine check exceptions.  This is a
+ * little different from the normal exception prolog above since a
+ * critical/machine check exception can potentially occur at any point
+ * during normal exception processing. Thus we cannot use the same SPRG
+ * registers as the normal prolog above. Instead we use a portion of the
+ * critical/machine check exception stack at low physical addresses.
  */
-
-#define CRITICAL_EXCEPTION_PROLOG					     \
-	mtspr	CRIT_SPRG,r8;						     \
-	BOOKE_LOAD_CRIT_STACK;		/* r8 points to the crit stack */    \
+#define EXC_LEVEL_EXCEPTION_PROLOG(exc_level, exc_level_srr0, exc_level_srr1) \
+	mtspr	exc_level##_SPRG,r8;					     \
+	BOOKE_LOAD_EXC_LEVEL_STACK(exc_level);/* r8 points to the exc_level stack*/ \
 	stw	r10,GPR10-INT_FRAME_SIZE(r8);				     \
 	stw	r11,GPR11-INT_FRAME_SIZE(r8);				     \
 	mfcr	r10;			/* save CR in r10 for now	   */\
-	mfspr	r11,SPRN_CSRR1;		/* check whether user or kernel    */\
+	mfspr	r11,exc_level_srr1;	/* check whether user or kernel    */\
 	andi.	r11,r11,MSR_PR;						     \
 	mr	r11,r8;							     \
-	mfspr	r8,CRIT_SPRG;						     \
+	mfspr	r8,exc_level##_SPRG;					     \
 	beq	1f;							     \
 	/* COMING FROM USER MODE */					     \
 	mfspr	r11,SPRN_SPRG3;		/* if from user, start at top of   */\
@@ -122,9 +112,9 @@
 	stw	r12,_DEAR(r11);		/* since they may have had stuff   */\
 	mfspr	r9,SPRN_ESR;		/* in them at the point where the  */\
 	stw	r9,_ESR(r11);		/* exception was taken		   */\
-	mfspr	r12,SPRN_CSRR0;						     \
+	mfspr	r12,exc_level_srr0;					     \
 	stw	r1,GPR1(r11);						     \
-	mfspr	r9,SPRN_CSRR1;						     \
+	mfspr	r9,exc_level_srr1;					     \
 	stw	r1,0(r11);						     \
 	mr	r1,r11;							     \
 	rlwinm	r9,r9,0,14,12;		/* clear MSR_WE (necessary?)	   */\
@@ -132,45 +122,10 @@
 	SAVE_4GPRS(3, r11);						     \
 	SAVE_2GPRS(7, r11)
 
-/*
- * Exception prolog for machine check exceptions.  This is similar to
- * the critical exception prolog, except that machine check exceptions
- * have their stack.
- */
-#define MCHECK_EXCEPTION_PROLOG					     \
-	mtspr	MCHECK_SPRG,r8;						     \
-	BOOKE_LOAD_MCHECK_STACK;	/* r8 points to the mcheck stack   */\
-	stw	r10,GPR10-INT_FRAME_SIZE(r8);				     \
-	stw	r11,GPR11-INT_FRAME_SIZE(r8);				     \
-	mfcr	r10;			/* save CR in r10 for now	   */\
-	mfspr	r11,SPRN_MCSRR1;	/* check whether user or kernel    */\
-	andi.	r11,r11,MSR_PR;						     \
-	mr	r11,r8;							     \
-	mfspr	r8,MCHECK_SPRG;						     \
-	beq	1f;							     \
-	/* COMING FROM USER MODE */					     \
-	mfspr	r11,SPRN_SPRG3;		/* if from user, start at top of   */\
-	lwz	r11,THREAD_INFO-THREAD(r11); /* this thread's kernel stack */\
-	addi	r11,r11,THREAD_SIZE;					     \
-1:	subi	r11,r11,INT_FRAME_SIZE;	/* Allocate an exception frame     */\
-	stw	r10,_CCR(r11);          /* save various registers	   */\
-	stw	r12,GPR12(r11);						     \
-	stw	r9,GPR9(r11);						     \
-	mflr	r10;							     \
-	stw	r10,_LINK(r11);						     \
-	mfspr	r12,SPRN_DEAR;		/* save DEAR and ESR in the frame  */\
-	stw	r12,_DEAR(r11);		/* since they may have had stuff   */\
-	mfspr	r9,SPRN_ESR;		/* in them at the point where the  */\
-	stw	r9,_ESR(r11);		/* exception was taken		   */\
-	mfspr	r12,SPRN_MCSRR0;					     \
-	stw	r1,GPR1(r11);						     \
-	mfspr	r9,SPRN_MCSRR1;						     \
-	stw	r1,0(r11);						     \
-	mr	r1,r11;							     \
-	rlwinm	r9,r9,0,14,12;		/* clear MSR_WE (necessary?)	   */\
-	stw	r0,GPR0(r11);						     \
-	SAVE_4GPRS(3, r11);						     \
-	SAVE_2GPRS(7, r11)
+#define CRITICAL_EXCEPTION_PROLOG \
+		EXC_LEVEL_EXCEPTION_PROLOG(CRIT, SPRN_CSRR0, SPRN_CSRR1)
+#define MCHECK_EXCEPTION_PROLOG \
+		EXC_LEVEL_EXCEPTION_PROLOG(MCHECK, SPRN_MCSRR0, SPRN_MCSRR1)
 
 /*
  * Exception vectors.
@@ -237,7 +192,6 @@ label:
 	EXC_XFER_TEMPLATE(hdlr, n+1, MSR_KERNEL, COPY_EE, transfer_to_handler, \
 			  ret_from_except)
 
-
 /* Check for a single step debug exception while in an exception
  * handler before state has been saved.  This is to catch the case
  * where an instruction that we are trying to single step causes
@@ -291,7 +245,7 @@ label:
 	lwz	r9,GPR9(r11);						      \
 	lwz	r12,GPR12(r11);						      \
 	mtspr	CRIT_SPRG,r8;						      \
-	BOOKE_LOAD_CRIT_STACK;		/* r8 points to the crit stack */     \
+	BOOKE_LOAD_EXC_LEVEL_STACK(CRIT); /* r8 points to the debug stack */  \
 	lwz	r10,GPR10-INT_FRAME_SIZE(r8);				      \
 	lwz	r11,GPR11-INT_FRAME_SIZE(r8);				      \
 	mfspr	r8,CRIT_SPRG;						      \
-- 
cgit v1.2.3-70-g09d2


From ba8c6d534a0144375f910e764ec965d5e8059f7a Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Tue, 21 Jun 2005 17:15:28 -0700
Subject: [PATCH] ppc32: remove some unnecessary includes of prom.h

Fight the Good Fight: Limit prom.h header creep.

Signed-off-by: Jon Loeliger <jdl@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/platforms/83xx/mpc834x_sys.c        | 1 -
 arch/ppc/platforms/85xx/mpc8540_ads.c        | 1 -
 arch/ppc/platforms/85xx/mpc8560_ads.c        | 1 -
 arch/ppc/platforms/85xx/mpc85xx_ads_common.c | 1 -
 arch/ppc/platforms/85xx/mpc85xx_cds_common.c | 1 -
 arch/ppc/platforms/85xx/sbc8560.c            | 1 -
 arch/ppc/platforms/85xx/sbc85xx.c            | 1 -
 arch/ppc/platforms/85xx/stx_gp3.c            | 1 -
 arch/ppc/syslib/open_pic.c                   | 1 -
 arch/ppc/syslib/open_pic2.c                  | 1 -
 arch/ppc/syslib/ppc83xx_setup.c              | 1 -
 arch/ppc/syslib/ppc85xx_setup.c              | 3 ++-
 12 files changed, 2 insertions(+), 12 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc/platforms/83xx/mpc834x_sys.c b/arch/ppc/platforms/83xx/mpc834x_sys.c
index e6348b5a1dd..86ca5cf8126 100644
--- a/arch/ppc/platforms/83xx/mpc834x_sys.c
+++ b/arch/ppc/platforms/83xx/mpc834x_sys.c
@@ -41,7 +41,6 @@
 #include <asm/time.h>
 #include <asm/io.h>
 #include <asm/machdep.h>
-#include <asm/prom.h>
 #include <asm/ipic.h>
 #include <asm/bootinfo.h>
 #include <asm/pci-bridge.h>
diff --git a/arch/ppc/platforms/85xx/mpc8540_ads.c b/arch/ppc/platforms/85xx/mpc8540_ads.c
index d1c2c6cb300..a2ed611cd93 100644
--- a/arch/ppc/platforms/85xx/mpc8540_ads.c
+++ b/arch/ppc/platforms/85xx/mpc8540_ads.c
@@ -41,7 +41,6 @@
 #include <asm/time.h>
 #include <asm/io.h>
 #include <asm/machdep.h>
-#include <asm/prom.h>
 #include <asm/open_pic.h>
 #include <asm/bootinfo.h>
 #include <asm/pci-bridge.h>
diff --git a/arch/ppc/platforms/85xx/mpc8560_ads.c b/arch/ppc/platforms/85xx/mpc8560_ads.c
index 761b8c7b25d..d87dfd5ce0a 100644
--- a/arch/ppc/platforms/85xx/mpc8560_ads.c
+++ b/arch/ppc/platforms/85xx/mpc8560_ads.c
@@ -41,7 +41,6 @@
 #include <asm/time.h>
 #include <asm/io.h>
 #include <asm/machdep.h>
-#include <asm/prom.h>
 #include <asm/open_pic.h>
 #include <asm/bootinfo.h>
 #include <asm/pci-bridge.h>
diff --git a/arch/ppc/platforms/85xx/mpc85xx_ads_common.c b/arch/ppc/platforms/85xx/mpc85xx_ads_common.c
index 30397623fa6..18e952d1767 100644
--- a/arch/ppc/platforms/85xx/mpc85xx_ads_common.c
+++ b/arch/ppc/platforms/85xx/mpc85xx_ads_common.c
@@ -36,7 +36,6 @@
 #include <asm/time.h>
 #include <asm/io.h>
 #include <asm/machdep.h>
-#include <asm/prom.h>
 #include <asm/open_pic.h>
 #include <asm/bootinfo.h>
 #include <asm/pci-bridge.h>
diff --git a/arch/ppc/platforms/85xx/mpc85xx_cds_common.c b/arch/ppc/platforms/85xx/mpc85xx_cds_common.c
index 51964b3081e..203b2ca61df 100644
--- a/arch/ppc/platforms/85xx/mpc85xx_cds_common.c
+++ b/arch/ppc/platforms/85xx/mpc85xx_cds_common.c
@@ -42,7 +42,6 @@
 #include <asm/todc.h>
 #include <asm/io.h>
 #include <asm/machdep.h>
-#include <asm/prom.h>
 #include <asm/open_pic.h>
 #include <asm/i8259.h>
 #include <asm/bootinfo.h>
diff --git a/arch/ppc/platforms/85xx/sbc8560.c b/arch/ppc/platforms/85xx/sbc8560.c
index 42e960a0bf3..3dbdd73618e 100644
--- a/arch/ppc/platforms/85xx/sbc8560.c
+++ b/arch/ppc/platforms/85xx/sbc8560.c
@@ -41,7 +41,6 @@
 #include <asm/time.h>
 #include <asm/io.h>
 #include <asm/machdep.h>
-#include <asm/prom.h>
 #include <asm/open_pic.h>
 #include <asm/bootinfo.h>
 #include <asm/pci-bridge.h>
diff --git a/arch/ppc/platforms/85xx/sbc85xx.c b/arch/ppc/platforms/85xx/sbc85xx.c
index a7cb625b2d0..4f6d1ddd6fb 100644
--- a/arch/ppc/platforms/85xx/sbc85xx.c
+++ b/arch/ppc/platforms/85xx/sbc85xx.c
@@ -35,7 +35,6 @@
 #include <asm/time.h>
 #include <asm/io.h>
 #include <asm/machdep.h>
-#include <asm/prom.h>
 #include <asm/open_pic.h>
 #include <asm/bootinfo.h>
 #include <asm/pci-bridge.h>
diff --git a/arch/ppc/platforms/85xx/stx_gp3.c b/arch/ppc/platforms/85xx/stx_gp3.c
index bc0a99638ff..9455bb6b45e 100644
--- a/arch/ppc/platforms/85xx/stx_gp3.c
+++ b/arch/ppc/platforms/85xx/stx_gp3.c
@@ -46,7 +46,6 @@
 #include <asm/time.h>
 #include <asm/io.h>
 #include <asm/machdep.h>
-#include <asm/prom.h>
 #include <asm/open_pic.h>
 #include <asm/bootinfo.h>
 #include <asm/pci-bridge.h>
diff --git a/arch/ppc/syslib/open_pic.c b/arch/ppc/syslib/open_pic.c
index 000ba47c67c..b45d8268bf9 100644
--- a/arch/ppc/syslib/open_pic.c
+++ b/arch/ppc/syslib/open_pic.c
@@ -21,7 +21,6 @@
 #include <asm/signal.h>
 #include <asm/io.h>
 #include <asm/irq.h>
-#include <asm/prom.h>
 #include <asm/sections.h>
 #include <asm/open_pic.h>
 #include <asm/i8259.h>
diff --git a/arch/ppc/syslib/open_pic2.c b/arch/ppc/syslib/open_pic2.c
index ea26da0d8b6..7e272c51a49 100644
--- a/arch/ppc/syslib/open_pic2.c
+++ b/arch/ppc/syslib/open_pic2.c
@@ -25,7 +25,6 @@
 #include <asm/signal.h>
 #include <asm/io.h>
 #include <asm/irq.h>
-#include <asm/prom.h>
 #include <asm/sections.h>
 #include <asm/open_pic.h>
 #include <asm/i8259.h>
diff --git a/arch/ppc/syslib/ppc83xx_setup.c b/arch/ppc/syslib/ppc83xx_setup.c
index 843cf8873e6..602a86891f7 100644
--- a/arch/ppc/syslib/ppc83xx_setup.c
+++ b/arch/ppc/syslib/ppc83xx_setup.c
@@ -23,7 +23,6 @@
 #include <linux/serial_core.h>
 #include <linux/serial_8250.h>
 
-#include <asm/prom.h>
 #include <asm/time.h>
 #include <asm/mpc83xx.h>
 #include <asm/mmu.h>
diff --git a/arch/ppc/syslib/ppc85xx_setup.c b/arch/ppc/syslib/ppc85xx_setup.c
index 455473f589d..ca95d79a704 100644
--- a/arch/ppc/syslib/ppc85xx_setup.c
+++ b/arch/ppc/syslib/ppc85xx_setup.c
@@ -23,7 +23,6 @@
 #include <linux/serial_core.h>
 #include <linux/serial_8250.h>
 
-#include <asm/prom.h>
 #include <asm/time.h>
 #include <asm/mpc85xx.h>
 #include <asm/immap_85xx.h>
@@ -33,6 +32,8 @@
 
 #include <syslib/ppc85xx_setup.h>
 
+extern void abort(void);
+
 /* Return the amount of memory */
 unsigned long __init
 mpc85xx_find_end_of_memory(void)
-- 
cgit v1.2.3-70-g09d2


From a70d439345875d476ede258094356e2acd09b1a1 Mon Sep 17 00:00:00 2001
From: Jakub Bogusz <qboosh@pld-linux.org>
Date: Tue, 21 Jun 2005 17:15:29 -0700
Subject: [PATCH] ppc32: don't recursively crash in die() on CHRP/PReP machines

This patch avoids recursive crash (leading to kernel stack overflow) in
die() on CHRP/PReP machines when CONFIG_PMAC_BACKLIGHT=y.  set_backlight_*
functions are placed in pmac section, which is discarded when _machine !=
_MACH_Pmac.

Signed-off-by: Jakub Bogusz <qboosh@pld-linux.org>
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/kernel/traps.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc/kernel/traps.c b/arch/ppc/kernel/traps.c
index c65731e8bc6..2ca8ecfeefd 100644
--- a/arch/ppc/kernel/traps.c
+++ b/arch/ppc/kernel/traps.c
@@ -81,8 +81,10 @@ void die(const char * str, struct pt_regs * fp, long err)
 	console_verbose();
 	spin_lock_irq(&die_lock);
 #ifdef CONFIG_PMAC_BACKLIGHT
-	set_backlight_enable(1);
-	set_backlight_level(BACKLIGHT_MAX);
+	if (_machine == _MACH_Pmac) {
+		set_backlight_enable(1);
+		set_backlight_level(BACKLIGHT_MAX);
+	}
 #endif
 	printk("Oops: %s, sig: %ld [#%d]\n", str, err, ++die_counter);
 #ifdef CONFIG_PREEMPT
-- 
cgit v1.2.3-70-g09d2


From 6879dc137ea4efad65cab8bf8a7c0b742bcf92cc Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 21 Jun 2005 17:15:30 -0700
Subject: [PATCH] ppc32: Kill embedded system.map, use kallsyms

This patch kills the whole embedded System.map mecanism and the
bootloader-passed System.map that was used to provide symbol resolution in
xmon.  Instead, xmon now uses kallsyms like ppc64 does.

No hurry getting that in Linus tree, let it be tested in -mm for a while
first and make sure it doesn't break various embedded configs.

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/boot/ld.script             |   3 -
 arch/ppc/boot/openfirmware/Makefile |   5 +-
 arch/ppc/boot/openfirmware/common.c |  16 ---
 arch/ppc/boot/simple/Makefile       |   4 +-
 arch/ppc/boot/utils/addSystemMap.c  | 186 -------------------------
 arch/ppc/kernel/setup.c             |   7 -
 arch/ppc/mm/init.c                  |  16 +--
 arch/ppc/xmon/xmon.c                | 270 ++++++++++--------------------------
 arch/ppc64/xmon/xmon.c              |   9 +-
 9 files changed, 84 insertions(+), 432 deletions(-)
 delete mode 100644 arch/ppc/boot/utils/addSystemMap.c

(limited to 'arch')

diff --git a/arch/ppc/boot/ld.script b/arch/ppc/boot/ld.script
index 6ee602d8b6a..9362193742a 100644
--- a/arch/ppc/boot/ld.script
+++ b/arch/ppc/boot/ld.script
@@ -58,9 +58,6 @@ SECTIONS
     *(.ramdisk)
     __ramdisk_end = .;
     . = ALIGN(4096);
-    __sysmap_begin = .;
-    *(.sysmap)
-    __sysmap_end = .;
     CONSTRUCTORS
   }
   _edata  =  .;
diff --git a/arch/ppc/boot/openfirmware/Makefile b/arch/ppc/boot/openfirmware/Makefile
index 4eacbd8c772..03415238fab 100644
--- a/arch/ppc/boot/openfirmware/Makefile
+++ b/arch/ppc/boot/openfirmware/Makefile
@@ -54,13 +54,10 @@ $(images)/ramdisk.image.gz:
 	@echo '          RAM disk image must be provided separately'
 	@/bin/false
 
-objcpxmon-$(CONFIG_XMON) := --add-section=.sysmap=System.map \
-	--set-section-flags=.sysmap=contents,alloc,load,readonly,data
 quiet_cmd_genimage = GEN     $@
       cmd_genimage = $(OBJCOPY) -R .comment       \
 	--add-section=.image=$(images)/vmlinux.gz \
-	--set-section-flags=.image=contents,alloc,load,readonly,data \
-	$(objcpxmon-y) $< $@
+	--set-section-flags=.image=contents,alloc,load,readonly,data $< $@
 
 targets += image.o
 $(obj)/image.o: $(obj)/dummy.o $(images)/vmlinux.gz FORCE
diff --git a/arch/ppc/boot/openfirmware/common.c b/arch/ppc/boot/openfirmware/common.c
index 9e6952781f1..0f46756a903 100644
--- a/arch/ppc/boot/openfirmware/common.c
+++ b/arch/ppc/boot/openfirmware/common.c
@@ -15,7 +15,6 @@
 #include <asm/page.h>
 
 /* Information from the linker */
-extern char __sysmap_begin, __sysmap_end;
 
 extern int strcmp(const char *s1, const char *s2);
 extern char *avail_ram, *avail_high;
@@ -116,14 +115,8 @@ void gunzip(void *dst, int dstlen, unsigned char *src, int *lenp)
 void make_bi_recs(unsigned long addr, char *name, unsigned int mach,
 		unsigned long progend)
 {
-	unsigned long sysmap_size;
 	struct bi_record *rec;
 
-	/* Figure out the size of a possible System.map we're going to
-	 * pass along.
-	 * */
-	sysmap_size = (unsigned long)(&__sysmap_end) -
-		(unsigned long)(&__sysmap_begin);
 
 	/* leave a 1MB gap then align to the next 1MB boundary */
 	addr = _ALIGN(addr+ (1<<20) - 1, (1<<20));
@@ -147,15 +140,6 @@ void make_bi_recs(unsigned long addr, char *name, unsigned int mach,
 	rec->size = sizeof(struct bi_record) + 2 * sizeof(unsigned long);
 	rec = (struct bi_record *)((unsigned long)rec + rec->size);
 
-	if (sysmap_size) {
-		rec->tag = BI_SYSMAP;
-		rec->data[0] = (unsigned long)(&__sysmap_begin);
-		rec->data[1] = sysmap_size;
-		rec->size = sizeof(struct bi_record) + 2 *
-			sizeof(unsigned long);
-		rec = (struct bi_record *)((unsigned long)rec + rec->size);
-	}
-
 	rec->tag = BI_LAST;
 	rec->size = sizeof(struct bi_record);
 	rec = (struct bi_record *)((unsigned long)rec + rec->size);
diff --git a/arch/ppc/boot/simple/Makefile b/arch/ppc/boot/simple/Makefile
index c28061ad598..991b4cbb83c 100644
--- a/arch/ppc/boot/simple/Makefile
+++ b/arch/ppc/boot/simple/Makefile
@@ -203,7 +203,7 @@ $(obj)/zvmlinux: $(OBJS) $(LIBS) $(srctree)/$(boot)/ld.script \
 		$(obj)/dummy.o $(obj)/image.o
 	$(LD) $(LD_ARGS) -o $@ $(OBJS) $(obj)/image.o $(LIBS)
 	$(OBJCOPY) $(OBJCOPY_ARGS) $@ $@ -R .comment -R .stab \
-		-R .stabstr -R .ramdisk -R .sysmap
+		-R .stabstr -R .ramdisk
 
 $(obj)/zvmlinux.initrd: $(OBJS) $(LIBS) $(srctree)/$(boot)/ld.script \
 		$(images)/vmlinux.gz $(obj)/dummy.o
@@ -215,7 +215,7 @@ $(obj)/zvmlinux.initrd: $(OBJS) $(LIBS) $(srctree)/$(boot)/ld.script \
 		$(obj)/dummy.o $(obj)/image.o
 	$(LD) $(LD_ARGS) -o $@ $(OBJS) $(obj)/image.o $(LIBS)
 	$(OBJCOPY) $(OBJCOPY_ARGS) $@ $@ -R .comment -R .stab \
-		-R .stabstr -R .sysmap
+		-R .stabstr
 
 # Sort-of dummy rules, that let us format the image we want.
 zImage: $(images)/$(zimage-y) $(obj)/zvmlinux
diff --git a/arch/ppc/boot/utils/addSystemMap.c b/arch/ppc/boot/utils/addSystemMap.c
deleted file mode 100644
index 4654f891b27..00000000000
--- a/arch/ppc/boot/utils/addSystemMap.c
+++ /dev/null
@@ -1,186 +0,0 @@
-#include <stdio.h>
-#include <stdlib.h>
-#include <byteswap.h>
-#include <sys/types.h>
-#include <sys/stat.h>
-
-void xlate( char * inb, char * trb, unsigned len )
-{
-    unsigned i;
-    for (  i=0; i<len; ++i ) {
-	char c = *inb++;
-	char c1 = c >> 4;
-	char c2 = c & 0xf;
-	if ( c1 > 9 )
-	    c1 = c1 + 'A' - 10;
-	else
-	    c1 = c1 + '0';
-	if ( c2 > 9 )
-	    c2 = c2 + 'A' - 10;
-	else
-	    c2 = c2 + '0';
-	*trb++ = c1;
-	*trb++ = c2;
-    }
-    *trb = 0;
-}
-
-#define ElfHeaderSize  (64 * 1024)
-#define ElfPages  (ElfHeaderSize / 4096)
-#define KERNELBASE (0xc0000000)
-
-void get4k( /*istream *inf*/FILE *file, char *buf )
-{
-    unsigned j;
-    unsigned num = fread(buf, 1, 4096, file);
-    for (  j=num; j<4096; ++j )
-	buf[j] = 0;
-}
-
-void put4k( /*ostream *outf*/FILE *file, char *buf )
-{
-    fwrite(buf, 1, 4096, file);
-}
-
-int main(int argc, char **argv)
-{
-    char inbuf[4096];
-    FILE *ramDisk = NULL;
-    FILE *inputVmlinux = NULL;
-    FILE *outputVmlinux = NULL;
-    unsigned i = 0;
-    unsigned long ramFileLen = 0;
-    unsigned long ramLen = 0;
-    unsigned long roundR = 0;
-    unsigned long kernelLen = 0;
-    unsigned long actualKernelLen = 0;
-    unsigned long round = 0;
-    unsigned long roundedKernelLen = 0;
-    unsigned long ramStartOffs = 0;
-    unsigned long ramPages = 0;
-    unsigned long roundedKernelPages = 0;
-    if ( argc < 2 ) {
-	printf("Name of System Map file missing.\n");
-	exit(1);
-    }
-
-    if ( argc < 3 ) {
-	printf("Name of vmlinux file missing.\n");
-	exit(1);
-    }
-
-    if ( argc < 4 ) {
-	printf("Name of vmlinux output file missing.\n");
-	exit(1);
-    }
-
-    ramDisk = fopen(argv[1], "r");
-    if ( ! ramDisk ) {
-	printf("System Map file \"%s\" failed to open.\n", argv[1]);
-	exit(1);
-    }
-    inputVmlinux = fopen(argv[2], "r");
-    if ( ! inputVmlinux ) {
-	printf("vmlinux file \"%s\" failed to open.\n", argv[2]);
-	exit(1);
-    }
-    outputVmlinux = fopen(argv[3], "w");
-    if ( ! outputVmlinux ) {
-	printf("output vmlinux file \"%s\" failed to open.\n", argv[3]);
-	exit(1);
-    }
-    fseek(ramDisk, 0, SEEK_END);
-    ramFileLen = ftell(ramDisk);
-    fseek(ramDisk, 0, SEEK_SET);
-    printf("%s file size = %ld\n", argv[1], ramFileLen);
-
-    ramLen = ramFileLen;
-
-    roundR = 4096 - (ramLen % 4096);
-    if ( roundR ) {
-	printf("Rounding System Map file up to a multiple of 4096, adding %ld\n", roundR);
-	ramLen += roundR;
-    }
-
-    printf("Rounded System Map size is %ld\n", ramLen);
-    fseek(inputVmlinux, 0, SEEK_END);
-    kernelLen = ftell(inputVmlinux);
-    fseek(inputVmlinux, 0, SEEK_SET);
-    printf("kernel file size = %ld\n", kernelLen);
-    if ( kernelLen == 0 ) {
-	printf("You must have a linux kernel specified as argv[2]\n");
-	exit(1);
-    }
-
-    actualKernelLen = kernelLen - ElfHeaderSize;
-
-    printf("actual kernel length (minus ELF header) = %ld\n", actualKernelLen);
-
-    round = actualKernelLen % 4096;
-    roundedKernelLen = actualKernelLen;
-    if ( round )
-	roundedKernelLen += (4096 - round);
-
-    printf("actual kernel length rounded up to a 4k multiple = %ld\n", roundedKernelLen);
-
-    ramStartOffs = roundedKernelLen;
-    ramPages = ramLen / 4096;
-
-    printf("System map pages to copy = %ld\n", ramPages);
-
-    // Copy 64K ELF header
-      for (i=0; i<(ElfPages); ++i) {
-	  get4k( inputVmlinux, inbuf );
-	  put4k( outputVmlinux, inbuf );
-      }
-
-
-
-    roundedKernelPages = roundedKernelLen / 4096;
-
-    fseek(inputVmlinux, ElfHeaderSize, SEEK_SET);
-
-    {
-	for ( i=0; i<roundedKernelPages; ++i ) {
-	    get4k( inputVmlinux, inbuf );
-	    if ( i == 0 ) {
-		unsigned long * p;
-		printf("Storing embedded_sysmap_start at 0x3c\n");
-		p = (unsigned long *)(inbuf + 0x3c);
-
-#if (BYTE_ORDER == __BIG_ENDIAN)
-		*p = ramStartOffs;
-#else
-		*p = bswap_32(ramStartOffs);
-#endif
-
-		printf("Storing embedded_sysmap_end at 0x44\n");
-		p = (unsigned long *)(inbuf + 0x44);
-#if (BYTE_ORDER == __BIG_ENDIAN)
-		*p = ramStartOffs + ramFileLen;
-#else
-		*p = bswap_32(ramStartOffs + ramFileLen);
-#endif
-	    }
-	    put4k( outputVmlinux, inbuf );
-	}
-    }
-
-    {
-	for ( i=0; i<ramPages; ++i ) {
-	    get4k( ramDisk, inbuf );
-	    put4k( outputVmlinux, inbuf );
-	}
-    }
-
-
-    fclose(ramDisk);
-    fclose(inputVmlinux);
-    fclose(outputVmlinux);
-    /* Set permission to executable */
-    chmod(argv[3], S_IRUSR|S_IWUSR|S_IXUSR|S_IRGRP|S_IXGRP|S_IROTH|S_IXOTH);
-
-    return 0;
-
-}
-
diff --git a/arch/ppc/kernel/setup.c b/arch/ppc/kernel/setup.c
index c344c66e68f..c42f7532693 100644
--- a/arch/ppc/kernel/setup.c
+++ b/arch/ppc/kernel/setup.c
@@ -61,8 +61,6 @@ extern void power4_idle(void);
 
 extern boot_infos_t *boot_infos;
 struct ide_machdep_calls ppc_ide_md;
-char *sysmap;
-unsigned long sysmap_size;
 
 /* Used with the BI_MEMSIZE bootinfo parameter to store the memory
    size value reported by the boot loader. */
@@ -578,11 +576,6 @@ void parse_bootinfo(struct bi_record *rec)
 		case BI_CMD_LINE:
 			strlcpy(cmd_line, (void *)data, sizeof(cmd_line));
 			break;
-		case BI_SYSMAP:
-			sysmap = (char *)((data[0] >= (KERNELBASE)) ? data[0] :
-					  (data[0]+KERNELBASE));
-			sysmap_size = data[1];
-			break;
 #ifdef CONFIG_BLK_DEV_INITRD
 		case BI_INITRD:
 			initrd_start = data[0] + KERNELBASE;
diff --git a/arch/ppc/mm/init.c b/arch/ppc/mm/init.c
index c8529d004cf..334ef4150d9 100644
--- a/arch/ppc/mm/init.c
+++ b/arch/ppc/mm/init.c
@@ -96,9 +96,6 @@ extern struct task_struct *current_set[NR_CPUS];
 char *klimit = _end;
 struct mem_pieces phys_avail;
 
-extern char *sysmap;
-extern unsigned long sysmap_size;
-
 /*
  * this tells the system to map all of ram with the segregs
  * (i.e. page tables) instead of the bats.
@@ -442,12 +439,6 @@ void __init mem_init(void)
 	if (agp_special_page)
 		SetPageReserved(virt_to_page(agp_special_page));
 #endif
-	if ( sysmap )
-		for (addr = (unsigned long)sysmap;
-		     addr < PAGE_ALIGN((unsigned long)sysmap+sysmap_size) ;
-		     addr += PAGE_SIZE)
-			SetPageReserved(virt_to_page(addr));
-
 	for (addr = PAGE_OFFSET; addr < (unsigned long)high_memory;
 	     addr += PAGE_SIZE) {
 		if (!PageReserved(virt_to_page(addr)))
@@ -482,9 +473,7 @@ void __init mem_init(void)
 	       codepages<< (PAGE_SHIFT-10), datapages<< (PAGE_SHIFT-10),
 	       initpages<< (PAGE_SHIFT-10),
 	       (unsigned long) (totalhigh_pages << (PAGE_SHIFT-10)));
-	if (sysmap)
-		printk("System.map loaded at 0x%08x for debugger, size: %ld bytes\n",
-			(unsigned int)sysmap, sysmap_size);
+
 #ifdef CONFIG_PPC_PMAC
 	if (agp_special_page)
 		printk(KERN_INFO "AGP special page: 0x%08lx\n", agp_special_page);
@@ -534,9 +523,6 @@ set_phys_avail(unsigned long total_memory)
 	if (rtas_data)
 		mem_pieces_remove(&phys_avail, rtas_data, rtas_size, 1);
 #endif
-	/* remove the sysmap pages from the available memory */
-	if (sysmap)
-		mem_pieces_remove(&phys_avail, __pa(sysmap), sysmap_size, 1);
 #ifdef CONFIG_PPC_PMAC
 	/* Because of some uninorth weirdness, we need a page of
 	 * memory as high as possible (it must be outside of the
diff --git a/arch/ppc/xmon/xmon.c b/arch/ppc/xmon/xmon.c
index 8565f49b8b0..be7869e3946 100644
--- a/arch/ppc/xmon/xmon.c
+++ b/arch/ppc/xmon/xmon.c
@@ -9,6 +9,7 @@
 #include <linux/smp.h>
 #include <linux/interrupt.h>
 #include <linux/bitops.h>
+#include <linux/kallsyms.h>
 #include <asm/ptrace.h>
 #include <asm/string.h>
 #include <asm/prom.h>
@@ -93,8 +94,7 @@ static void take_input(char *);
 static unsigned read_spr(int);
 static void write_spr(int, unsigned);
 static void super_regs(void);
-static void print_sysmap(void);
-static void sysmap_lookup(void);
+static void symbol_lookup(void);
 static void remove_bpts(void);
 static void insert_bpts(void);
 static struct bpt *at_breakpoint(unsigned pc);
@@ -103,7 +103,6 @@ static void cacheflush(void);
 #ifdef CONFIG_SMP
 static void cpu_cmd(void);
 #endif /* CONFIG_SMP */
-static int pretty_print_addr(unsigned long addr);
 static void csum(void);
 #ifdef CONFIG_BOOTX_TEXT
 static void vidcmds(void);
@@ -120,8 +119,6 @@ extern void longjmp(u_int *, int);
 
 extern void xmon_enter(void);
 extern void xmon_leave(void);
-extern char* xmon_find_symbol(unsigned long addr, unsigned long* saddr);
-extern unsigned long xmon_symbol_to_addr(char* symbol);
 
 static unsigned start_tb[NR_CPUS][2];
 static unsigned stop_tb[NR_CPUS][2];
@@ -148,7 +145,6 @@ Commands:\n\
   mm	move a block of memory\n\
   ms	set a block of memory\n\
   md	compare two blocks of memory\n\
-  M	print System.map\n\
   r	print registers\n\
   S	print special registers\n\
   t	print backtrace\n\
@@ -175,6 +171,35 @@ extern inline void __delay(unsigned int loops)
 				     "r" (loops) : "ctr");
 }
 
+/* Print an address in numeric and symbolic form (if possible) */
+static void xmon_print_symbol(unsigned long address, const char *mid,
+			      const char *after)
+{
+	char *modname;
+	const char *name = NULL;
+	unsigned long offset, size;
+	static char tmpstr[128];
+
+	printf("%.8lx", address);
+	if (setjmp(bus_error_jmp) == 0) {
+		debugger_fault_handler = handle_fault;
+		sync();
+		name = kallsyms_lookup(address, &size, &offset, &modname,
+				       tmpstr);
+		sync();
+		/* wait a little while to see if we get a machine check */
+		__delay(200);
+	}
+	debugger_fault_handler = NULL;
+
+	if (name) {
+		printf("%s%s+%#lx/%#lx", mid, name, offset, size);
+		if (modname)
+			printf(" [%s]", modname);
+	}
+	printf("%s", after);
+}
+
 static void get_tb(unsigned *p)
 {
 	unsigned hi, lo, hiagain;
@@ -454,7 +479,7 @@ cmds(struct pt_regs *excp)
 			dump();
 			break;
 		case 'l':
-			sysmap_lookup();
+			symbol_lookup();
 			break;
 		case 'r':
 			if (excp != NULL)
@@ -466,9 +491,6 @@ cmds(struct pt_regs *excp)
 			else
 				excprint(excp);
 			break;
-		case 'M':
-			print_sysmap();
-			break;
 		case 'S':
 			super_regs();
 			break;
@@ -825,20 +847,19 @@ backtrace(struct pt_regs *excp)
 	for (; sp != 0; sp = stack[0]) {
 		if (mread(sp, stack, sizeof(stack)) != sizeof(stack))
 			break;
-		pretty_print_addr(stack[1]);
-		printf(" ");
+		printf("[%.8lx] ", stack);
+		xmon_print_symbol(stack[1], " ", "\n");
 		if (stack[1] == (unsigned) &ret_from_except
 		    || stack[1] == (unsigned) &ret_from_except_full
 		    || stack[1] == (unsigned) &ret_from_syscall) {
 			if (mread(sp+16, &regs, sizeof(regs)) != sizeof(regs))
 				break;
-			printf("\nexception:%x [%x] %x ", regs.trap, sp+16,
+			printf("exception:%x [%x] %x\n", regs.trap, sp+16,
 			       regs.nip);
 			sp = regs.gpr[1];
 			if (mread(sp, stack, sizeof(stack)) != sizeof(stack))
 				break;
 		}
-		printf("\n");
 	}
 }
 
@@ -859,11 +880,10 @@ excprint(struct pt_regs *fp)
 #ifdef CONFIG_SMP
 	printf("cpu %d: ", smp_processor_id());
 #endif /* CONFIG_SMP */
-	printf("vector: %x at pc = ", fp->trap);
-	pretty_print_addr(fp->nip);
-	printf(", lr = ");
-	pretty_print_addr(fp->link);
-	printf("\nmsr = %x, sp = %x [%x]\n", fp->msr, fp->gpr[1], fp);
+	printf("vector: %x at pc=", fp->trap);
+	xmon_print_symbol(fp->nip, ": ", ", lr=");
+	xmon_print_symbol(fp->link, ": ", "\n");
+	printf("msr = %x, sp = %x [%x]\n", fp->msr, fp->gpr[1], fp);
 	trap = TRAP(fp);
 	if (trap == 0x300 || trap == 0x600)
 		printf("dar = %x, dsisr = %x\n", fp->dar, fp->dsisr);
@@ -950,24 +970,6 @@ static unsigned int regno;
 extern char exc_prolog;
 extern char dec_exc;
 
-void
-print_sysmap(void)
-{
-	extern char *sysmap;
-	if ( sysmap ) {
-		printf("System.map: \n");
-		if( setjmp(bus_error_jmp) == 0 ) {
-			debugger_fault_handler = handle_fault;
-			sync();
-			xmon_puts(sysmap);
-			sync();
-		}
-		debugger_fault_handler = NULL;
-	}
-	else
-		printf("No System.map\n");
-}
-
 void
 super_regs(void)
 {
@@ -1738,7 +1740,7 @@ scanhex(unsigned *vp)
 		printf("invalid register name '%%%s'\n", regname);
 		return 0;
 	} else if (c == '$') {
-		static char symname[64];
+		static char symname[128];
 		int i;
 		for (i=0; i<63; i++) {
 			c = inchar();
@@ -1749,7 +1751,14 @@ scanhex(unsigned *vp)
 			symname[i] = c;
 		}
 		symname[i++] = 0;
-		*vp = xmon_symbol_to_addr(symname);
+		*vp = 0;
+		if (setjmp(bus_error_jmp) == 0) {
+			debugger_fault_handler = handle_fault;
+			sync();
+			*vp = kallsyms_lookup_name(symname);
+			sync();
+		}
+		debugger_fault_handler = NULL;
 		if (!(*vp)) {
 			printf("unknown symbol\n");
 			return 0;
@@ -1840,169 +1849,34 @@ take_input(char *str)
 	lineptr = str;
 }
 
-void
-sysmap_lookup(void)
+static void
+symbol_lookup(void)
 {
 	int type = inchar();
 	unsigned addr;
-	static char tmp[64];
-	char* cur;
-
-	extern char *sysmap;
-	extern unsigned long sysmap_size;
-	if ( !sysmap || !sysmap_size )
-		return;
-
-	switch(type) {
-		case 'a':
-			if (scanhex(&addr)) {
-				pretty_print_addr(addr);
-				printf("\n");
-			}
-			termch = 0;
-			break;
-		case 's':
-			getstring(tmp, 64);
-			if( setjmp(bus_error_jmp) == 0 ) {
-				debugger_fault_handler = handle_fault;
-				sync();
-				cur = sysmap;
-				do {
-					cur = strstr(cur, tmp);
-					if (cur) {
-						static char res[64];
-						char *p, *d;
-						p = cur;
-						while(p > sysmap && *p != 10)
-							p--;
-						if (*p == 10) p++;
-						d = res;
-						while(*p && p < (sysmap + sysmap_size) && *p != 10)
-							*(d++) = *(p++);
-						*(d++) = 0;
-						printf("%s\n", res);
-						cur++;
-					}
-				} while (cur);
-				sync();
-			}
-			debugger_fault_handler = NULL;
-			termch = 0;
-			break;
-	}
-}
+	static char tmp[128];
 
-static int
-pretty_print_addr(unsigned long addr)
-{
-	char *sym;
-	unsigned long saddr;
-	
-	printf("%08x", addr);
-	sym = xmon_find_symbol(addr, &saddr);
-	if (sym)
-		printf(" (%s+0x%x)", sym, addr-saddr);
-	return (sym != 0);
-}
-
-char*
-xmon_find_symbol(unsigned long addr, unsigned long* saddr)
-{
-	static char rbuffer[64];
-	char *p, *ep, *limit;
-	unsigned long prev, next;
-	char* psym;
-
-	extern char *sysmap;
-	extern unsigned long sysmap_size;
-	if ( !sysmap || !sysmap_size )
-		return NULL;
-	
-	prev = 0;
-	psym = NULL;
-	p = sysmap;
-	limit = p + sysmap_size;
-	if( setjmp(bus_error_jmp) == 0 ) {
-		debugger_fault_handler = handle_fault;
-		sync();
-		do {
-			next = simple_strtoul(p, &p, 16);
-			if (next > addr && prev <= addr) {
-				if (!psym)
-					goto bail;
-				ep = rbuffer;
-				p = psym;
-				while(*p && p < limit && *p == 32)
-					p++;
-				while(*p && p < limit && *p != 10 && (ep - rbuffer) < 63)
-					*(ep++) = *(p++);
-				*(ep++) = 0;
-				if (saddr)
-					*saddr = prev;
-				debugger_fault_handler = NULL;
-				return rbuffer;
-			}
-			prev = next;
-			psym = p;
-			while(*p && p < limit && *p != 10)
-				p++;
-			if (*p) p++;
-		} while(*p && p < limit && next);
-bail:
-		sync();
+	switch (type) {
+	case 'a':
+		if (scanhex(&addr))
+			xmon_print_symbol(addr, ": ", "\n");
+		termch = 0;
+		break;
+	case 's':
+		getstring(tmp, 64);
+		if (setjmp(bus_error_jmp) == 0) {
+			debugger_fault_handler = handle_fault;
+			sync();
+			addr = kallsyms_lookup_name(tmp);
+			if (addr)
+				printf("%s: %lx\n", tmp, addr);
+			else
+				printf("Symbol '%s' not found.\n", tmp);
+			sync();
+		}
+		debugger_fault_handler = NULL;
+		termch = 0;
+		break;
 	}
-	debugger_fault_handler = NULL;
-	return NULL;
 }
 
-unsigned long
-xmon_symbol_to_addr(char* symbol)
-{
-	char *p, *cur;
-	char *match = NULL;
-	int goodness = 0;
-	int result = 0;
-	
-	extern char *sysmap;
-	extern unsigned long sysmap_size;
-	if ( !sysmap || !sysmap_size )
-		return 0;
-
-	if( setjmp(bus_error_jmp) == 0 ) {
-		debugger_fault_handler = handle_fault;
-		sync();
-		cur = sysmap;
-		while(cur) {
-			cur = strstr(cur, symbol);
-			if (cur) {
-				int gd = 1;
-
-				/* best match if equal, better match if
-				 * begins with
-				 */
-				if (cur == sysmap || *(cur-1) == ' ') {
-					gd++;
-					if (cur[strlen(symbol)] == 10)
-						gd++;
-				}
-				if (gd > goodness) {
-					match = cur;
-					goodness = gd;
-					if (gd == 3)
-						break;
-				}
-				cur++;
-			}
-		}	
-		if (goodness) {
-			p = match;
-			while(p > sysmap && *p != 10)
-				p--;
-			if (*p == 10) p++;
-			result = simple_strtoul(p, &p, 16);
-		}
-		sync();
-	}
-	debugger_fault_handler = NULL;
-	return result;
-}		
diff --git a/arch/ppc64/xmon/xmon.c b/arch/ppc64/xmon/xmon.c
index 3c0ccb2623a..7f6e13a4b71 100644
--- a/arch/ppc64/xmon/xmon.c
+++ b/arch/ppc64/xmon/xmon.c
@@ -2247,7 +2247,14 @@ scanhex(unsigned long *vp)
 			tmpstr[i] = c;
 		}
 		tmpstr[i++] = 0;
-		*vp = kallsyms_lookup_name(tmpstr);
+		*vp = 0;
+		if (setjmp(bus_error_jmp) == 0) {
+			catch_memory_errors = 1;
+			sync();
+			*vp = kallsyms_lookup_name(tmpstr);
+			sync();
+		}
+		catch_memory_errors = 0;
 		if (!(*vp)) {
 			printf("unknown symbol '%s'\n", tmpstr);
 			return 0;
-- 
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 'arch')

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 723e2b35e43dcbcfd737c40453caa7d198092d23 Mon Sep 17 00:00:00 2001
From: Sven Luther <sven.luther@wanadoo.fr>
Date: Tue, 21 Jun 2005 17:15:32 -0700
Subject: [PATCH] ppc64: override command line AS/LD/CC variables when adding
 -m64 and co for biarch compilers

The following kind of calls currently fails :

  make ARCH=ppc64 CC="gcc-3.4"

Since the code for detecting a biarch compiler and adding the needed 64bit
magic argument fails if the AS/LD/CC commands are overriden in the command
line.

The attached patch fixes this by using the make override and += directive,
but i am not 100% sure this will work without gmake, as i am no Makefile
expert.

Cc: Paul Mackerras <paulus@samba.org>
Cc: Anton Blanchard <anton@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>
---
 arch/ppc64/Makefile | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/Makefile b/arch/ppc64/Makefile
index 691f3008e69..33c752ceca4 100644
--- a/arch/ppc64/Makefile
+++ b/arch/ppc64/Makefile
@@ -35,9 +35,9 @@ CROSS32AS	:= $(AS) -a32
 CROSS32LD	:= $(LD) -m elf32ppc
 CROSS32OBJCOPY	:= $(OBJCOPY)
 endif
-AS              := $(AS) -a64
-LD              := $(LD) -m elf64ppc
-CC		:= $(CC) -m64
+override AS	+= -a64
+override LD	+= -m elf64ppc
+override CC	+= -m64
 endif
 
 export CROSS32CC CROSS32AS CROSS32LD CROSS32OBJCOPY
-- 
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 'arch')

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 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 'arch')

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 'arch')

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 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 'arch')

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 'arch')

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 'arch')

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 'arch')

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 57ca86d4f0e44f81be10bc77a5d0e26760e4844f Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:45 -0700
Subject: [PATCH] ppc64 iSeries: iSeries_VpdInfo.c cleanups

Clean up iSeries_VpdInfo.c:
	- white space and comment fixes
	- make a function static
	- the functions here are only called from iSeries_pci.c, so
	  CONFIG_PCI will be set (so remove check)
	- only build when CONFIG_PCI is set
	- remove unneeded includes and cast

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          |  5 +--
 arch/ppc64/kernel/iSeries_VpdInfo.c | 84 ++++++++++++++++++-------------------
 2 files changed, 43 insertions(+), 46 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/Makefile b/arch/ppc64/kernel/Makefile
index 9132d927171..ada9be84dde 100644
--- a/arch/ppc64/kernel/Makefile
+++ b/arch/ppc64/kernel/Makefile
@@ -16,13 +16,12 @@ obj-y += vdso32/ vdso64/
 
 obj-$(CONFIG_PPC_OF) +=	of_device.o
 
-pci-obj-$(CONFIG_PPC_ISERIES)	+= iSeries_pci.o
+pci-obj-$(CONFIG_PPC_ISERIES)	+= iSeries_pci.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-$(CONFIG_PPC_ISERIES) += iSeries_irq.o \
-			     iSeries_VpdInfo.o XmPciLpEvent.o \
+obj-$(CONFIG_PPC_ISERIES) += iSeries_irq.o XmPciLpEvent.o \
 			     HvCall.o HvLpConfig.o LparData.o \
 			     iSeries_setup.o ItLpQueue.o hvCall.o \
 			     mf.o HvLpEvent.o iSeries_proc.o iSeries_htab.o \
diff --git a/arch/ppc64/kernel/iSeries_VpdInfo.c b/arch/ppc64/kernel/iSeries_VpdInfo.c
index a82caf98d2e..9c89ce9a3d0 100644
--- a/arch/ppc64/kernel/iSeries_VpdInfo.c
+++ b/arch/ppc64/kernel/iSeries_VpdInfo.c
@@ -1,31 +1,31 @@
-/************************************************************************/
-/* File iSeries_vpdInfo.c created by Allan Trautman on Fri Feb  2 2001. */
-/************************************************************************/
-/* This code gets the card location of the hardware                     */
-/* 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, Feb 2, 2001                                               */
-/*   Ported to ppc64, August 20, 2001                                   */
-/* End Change Activity                                                  */
-/************************************************************************/
-#include <linux/config.h>
+/*
+ * File iSeries_vpdInfo.c created by Allan Trautman on Fri Feb  2 2001.
+ *
+ * This code gets the card location of the hardware
+ * Copyright (C) 2001  <Allan H Trautman> <IBM Corp>
+ * Copyright (C) 2005  Stephen Rothwel, 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, Feb 2, 2001
+ *   Ported to ppc64, August 20, 2001
+ * End Change Activity
+ */
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/pci.h>
@@ -34,14 +34,13 @@
 
 #include <asm/iSeries/HvCallPci.h>
 #include <asm/iSeries/HvTypes.h>
-#include <asm/iSeries/mf.h>
 #include <asm/iSeries/iSeries_pci.h>
-#include "pci.h"
 
 /*
  * Size of Bus VPD data
  */
 #define BUS_VPDSIZE      1024
+
 /*
  * Bus Vpd Tags
  */
@@ -49,6 +48,7 @@
 #define  VpdEndOfAreaTag   0x79
 #define  VpdIdStringTag    0x82
 #define  VpdVendorAreaTag  0x84
+
 /*
  * Mfg Area Tags
  */
@@ -78,7 +78,7 @@ struct SlotMapStruct {
 	char CardLocation[3];
 	char Parms[8];
 	char Reserved[2];
-}; 
+};
 typedef struct SlotMapStruct SlotMap;
 #define SLOT_ENTRY_SIZE   16
 
@@ -111,28 +111,26 @@ int iSeries_Device_Information(struct pci_dev *PciDev, char *buffer,
 			PciDev->vendor);
 	len += sprintf(buffer + len, "Frame%3d, Card %4s  ",
 			DevNode->FrameId, DevNode->CardLocation);
-#ifdef CONFIG_PCI
 	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));
-#endif
 	return len;
 }
 
 /*
  * Parse the Slot Area
  */
-void iSeries_Parse_SlotArea(SlotMap *MapPtr, int MapLen,
+static void iSeries_Parse_SlotArea(SlotMap *MapPtr, int MapLen,
 		struct iSeries_Device_Node *DevNode)
 {
 	int SlotMapLen = MapLen;
 	SlotMap *SlotMapPtr = MapPtr;
 
 	/*
-	 * Parse Slot label until we find the one requrested
+	 * Parse Slot label until we find the one requested
 	 */
 	while (SlotMapLen > 0) {
 		if (SlotMapPtr->AgentId == DevNode->AgentId ) {
@@ -182,7 +180,7 @@ static void iSeries_Parse_MfgArea(u8 *AreaData, int AreaLen,
 			if (SlotMapFmt == 0x1004)
 				SlotMapPtr = (SlotMap *)((char *)MfgAreaPtr
 						+ MFG_ENTRY_SIZE + 1);
- 	    		else
+			else
 				SlotMapPtr = (SlotMap *)((char *)MfgAreaPtr
 						+ MFG_ENTRY_SIZE);
 	    		iSeries_Parse_SlotArea(SlotMapPtr, MfgTagLen, DevNode);
@@ -193,8 +191,8 @@ static void iSeries_Parse_MfgArea(u8 *AreaData, int AreaLen,
 		 */
 		MfgAreaPtr = (MfgArea *)((char *)MfgAreaPtr + MfgTagLen
 				+ MFG_ENTRY_SIZE);
-		MfgAreaLen -= (MfgTagLen + MFG_ENTRY_SIZE); 
-	}	
+		MfgAreaLen -= (MfgTagLen + MFG_ENTRY_SIZE);
+	}
 }
 
 /*
@@ -205,7 +203,7 @@ static int iSeries_Parse_PhbId(u8 *AreaPtr, int AreaLength)
 {
 	u8 *PhbPtr = AreaPtr;
 	int DataLen = AreaLength;
-	char PhbId = 0xFF;                   
+	char PhbId = 0xFF;
 
 	while (DataLen > 0) {
 		if ((*PhbPtr == 'B') && (*(PhbPtr + 1) == 'U')
@@ -215,7 +213,7 @@ static int iSeries_Parse_PhbId(u8 *AreaPtr, int AreaLength)
 				++PhbPtr;
 			PhbId = (*PhbPtr & 0x0F);
 			break;
-        	}
+		}
 		++PhbPtr;
 		--DataLen;
 	}
@@ -232,7 +230,7 @@ static void iSeries_Parse_Vpd(u8 *VpdData, int VpdDataLen,
 	int DataLen = VpdDataLen - 3;
 
 	while ((*TagPtr != VpdEndOfAreaTag) && (DataLen > 0)) {
-		int AreaLen = *(TagPtr + 1) + (*(TagPtr + 2) * 256);	
+		int AreaLen = *(TagPtr + 1) + (*(TagPtr + 2) * 256);
 		u8 *AreaData  = TagPtr + 3;
 
 		if (*TagPtr == VpdIdStringTag)
@@ -243,12 +241,12 @@ static void iSeries_Parse_Vpd(u8 *VpdData, int VpdDataLen,
 		TagPtr  = AreaData + AreaLen;
 		DataLen -= AreaLen;
 	}
-}    
+}
 
 void iSeries_Get_Location_Code(struct iSeries_Device_Node *DevNode)
 {
 	int BusVpdLen = 0;
-	u8 *BusVpdPtr = (u8 *)kmalloc(BUS_VPDSIZE, GFP_KERNEL);
+	u8 *BusVpdPtr = kmalloc(BUS_VPDSIZE, GFP_KERNEL);
 
 	if (BusVpdPtr == NULL) {
 		printk("PCI: Bus VPD Buffer allocation failure.\n");
-- 
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 'arch')

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 'arch')

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 'arch')

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 'arch')

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 89ef68f0be53df6c4f1552fc329fe5abb1d5ed33 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:50 -0700
Subject: [PATCH] ppc64 iSeries: remove XmPciLpEvent.c

This patch just merges XmPciLpEvent.c into iSeries_irq.c (the only caller of
its only external function).  XmPciLpEvent.c just contained the lowlevel
iSeries irq code.

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/XmPciLpEvent.c | 183 ---------------------------------------
 arch/ppc64/kernel/iSeries_irq.c  | 170 +++++++++++++++++++++++++++++++++++-
 3 files changed, 169 insertions(+), 186 deletions(-)
 delete mode 100644 arch/ppc64/kernel/XmPciLpEvent.c

(limited to 'arch')

diff --git a/arch/ppc64/kernel/Makefile b/arch/ppc64/kernel/Makefile
index ada9be84dde..ebf9b3cbf33 100644
--- a/arch/ppc64/kernel/Makefile
+++ b/arch/ppc64/kernel/Makefile
@@ -21,7 +21,7 @@ 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-$(CONFIG_PPC_ISERIES) += iSeries_irq.o XmPciLpEvent.o \
+obj-$(CONFIG_PPC_ISERIES) += iSeries_irq.o \
 			     HvCall.o HvLpConfig.o LparData.o \
 			     iSeries_setup.o ItLpQueue.o hvCall.o \
 			     mf.o HvLpEvent.o iSeries_proc.o iSeries_htab.o \
diff --git a/arch/ppc64/kernel/XmPciLpEvent.c b/arch/ppc64/kernel/XmPciLpEvent.c
deleted file mode 100644
index 3db53074483..00000000000
--- a/arch/ppc64/kernel/XmPciLpEvent.c
+++ /dev/null
@@ -1,183 +0,0 @@
-/*
- * 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>
-#include <linux/threads.h>
-#include <linux/smp.h>
-#include <linux/param.h>
-#include <linux/string.h>
-#include <linux/bootmem.h>
-#include <linux/ide.h>
-
-#include <asm/iSeries/HvTypes.h>
-#include <asm/iSeries/HvLpEvent.h>
-#include <asm/iSeries/HvCallPci.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
-};
-
-struct XmPciLpEvent_BusInterrupt {
-	HvBusNumber	busNumber;
-	HvSubBusNumber	subBusNumber;
-};
-
-struct XmPciLpEvent_NodeInterrupt {
-	HvBusNumber	busNumber;
-	HvSubBusNumber	subBusNumber;
-	HvAgentId	deviceId;
-};
-
-struct XmPciLpEvent {
-	struct HvLpEvent hvLpEvent;
-
-	union {
-		u64 alignData;			// Align on an 8-byte boundary
-
-		struct {
-			u32		fisr;
-			HvBusNumber	busNumber;
-			HvSubBusNumber	subBusNumber;
-			HvAgentId	deviceId;
-		} slotInterrupt;
-
-		struct XmPciLpEvent_BusInterrupt busFailed;
-		struct XmPciLpEvent_BusInterrupt busRecovered;
-		struct XmPciLpEvent_BusInterrupt busCreated;
-
-		struct XmPciLpEvent_NodeInterrupt nodeFailed;
-		struct XmPciLpEvent_NodeInterrupt nodeRecovered;
-
-	} eventData;
-
-};
-
-static void intReceived(struct XmPciLpEvent *eventParm,
-		struct pt_regs *regsParm)
-{
-	int irq;
-
-	++Pci_Interrupt_Count;
-#if 0
-	PPCDBG(PPCDBG_BUSWALK, "PCI: XmPciLpEvent.c: intReceived\n");
-#endif
-
-	switch (eventParm->hvLpEvent.xSubtype) {
-	case XmPciLpEvent_SlotInterrupt:
-		irq = eventParm->hvLpEvent.xCorrelationToken;
-		/* Dispatch the interrupt handlers for this irq */
-		ppc_irq_dispatch_handler(regsParm, irq);
-		HvCallPci_eoi(eventParm->eventData.slotInterrupt.busNumber,
-			eventParm->eventData.slotInterrupt.subBusNumber,
-			eventParm->eventData.slotInterrupt.deviceId);
-		break;
-		/* Ignore error recovery events for now */
-	case XmPciLpEvent_BusCreated:
-		printk(KERN_INFO "XmPciLpEvent.c: 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",
-			eventParm->eventData.busFailed.busNumber);
-		break;
-	case XmPciLpEvent_BusRecovered:
-	case XmPciLpEvent_UnQuiesceBus:
-		printk(KERN_INFO "XmPciLpEvent.c: 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",
-			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",
-			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",
-			eventParm->hvLpEvent.xSubtype);
-		break;
-	}
-}
-
-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()
-{
-	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 "
-					"failed with rc 0x%x\n", xRc);
-	} else
-		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 41902e358e7..96a1f8cd67e 100644
--- a/arch/ppc64/kernel/iSeries_irq.c
+++ b/arch/ppc64/kernel/iSeries_irq.c
@@ -1,6 +1,7 @@
 /*
  * This module supports the iSeries PCI bus interrupt handling
  * Copyright (C) 20yy  <Robert L Holtorf> <IBM Corp>
+ * Copyright (C) 2004-2005 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
@@ -22,6 +23,7 @@
  *   Created, December 13, 2000 by Wayne Holm
  * End Change Activity
  */
+#include <linux/config.h>
 #include <linux/pci.h>
 #include <linux/init.h>
 #include <linux/threads.h>
@@ -30,11 +32,12 @@
 #include <linux/string.h>
 #include <linux/bootmem.h>
 #include <linux/ide.h>
-
 #include <linux/irq.h>
 #include <linux/spinlock.h>
-#include <asm/ppcdebug.h>
 
+#include <asm/ppcdebug.h>
+#include <asm/iSeries/HvTypes.h>
+#include <asm/iSeries/HvLpEvent.h>
 #include <asm/iSeries/HvCallPci.h>
 #include <asm/iSeries/HvCallXm.h>
 #include <asm/iSeries/iSeries_irq.h>
@@ -46,6 +49,169 @@ unsigned int virt_irq_to_real_map[NR_IRQS];
 /* Note: the pcnet32 driver assumes irq numbers < 2 aren't valid. :( */
 static int next_virtual_irq = 2;
 
+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
+};
+
+struct XmPciLpEvent_BusInterrupt {
+	HvBusNumber	busNumber;
+	HvSubBusNumber	subBusNumber;
+};
+
+struct XmPciLpEvent_NodeInterrupt {
+	HvBusNumber	busNumber;
+	HvSubBusNumber	subBusNumber;
+	HvAgentId	deviceId;
+};
+
+struct XmPciLpEvent {
+	struct HvLpEvent hvLpEvent;
+
+	union {
+		u64 alignData;			// Align on an 8-byte boundary
+
+		struct {
+			u32		fisr;
+			HvBusNumber	busNumber;
+			HvSubBusNumber	subBusNumber;
+			HvAgentId	deviceId;
+		} slotInterrupt;
+
+		struct XmPciLpEvent_BusInterrupt busFailed;
+		struct XmPciLpEvent_BusInterrupt busRecovered;
+		struct XmPciLpEvent_BusInterrupt busCreated;
+
+		struct XmPciLpEvent_NodeInterrupt nodeFailed;
+		struct XmPciLpEvent_NodeInterrupt nodeRecovered;
+
+	} eventData;
+
+};
+
+static void intReceived(struct XmPciLpEvent *eventParm,
+		struct pt_regs *regsParm)
+{
+	int irq;
+
+	++Pci_Interrupt_Count;
+#if 0
+	PPCDBG(PPCDBG_BUSWALK, "PCI: XmPciLpEvent.c: intReceived\n");
+#endif
+
+	switch (eventParm->hvLpEvent.xSubtype) {
+	case XmPciLpEvent_SlotInterrupt:
+		irq = eventParm->hvLpEvent.xCorrelationToken;
+		/* Dispatch the interrupt handlers for this irq */
+		ppc_irq_dispatch_handler(regsParm, irq);
+		HvCallPci_eoi(eventParm->eventData.slotInterrupt.busNumber,
+			eventParm->eventData.slotInterrupt.subBusNumber,
+			eventParm->eventData.slotInterrupt.deviceId);
+		break;
+		/* Ignore error recovery events for now */
+	case XmPciLpEvent_BusCreated:
+		printk(KERN_INFO "XmPciLpEvent.c: 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",
+			eventParm->eventData.busFailed.busNumber);
+		break;
+	case XmPciLpEvent_BusRecovered:
+	case XmPciLpEvent_UnQuiesceBus:
+		printk(KERN_INFO "XmPciLpEvent.c: 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",
+			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",
+			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",
+			eventParm->hvLpEvent.xSubtype);
+		break;
+	}
+}
+
+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()
+{
+	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 "
+					"failed with rc 0x%x\n", xRc);
+	} else
+		printk(KERN_ERR "XmPciLpEvent.c: 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)
 {
-- 
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 'arch')

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 'arch')

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 ac5b33c9bcf1c05ae949b026a51d7f8f9573591c Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 21 Jun 2005 17:15:54 -0700
Subject: [PATCH] ppc64: tidy up vio devices fake parent

Currently we dynamically allocate the fake parent device for all devices on
the vio bus.  This patch statically allocates it.  This also allows us to
reuse it for the iSeries "generic" vio device (that is used for passing to
dma routines when communicating with the hypervisor without a device
involved).  Also unexport vio_bus_type as it is never used in modules.

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/vio.c | 37 ++++++++++++++++---------------------
 1 file changed, 16 insertions(+), 21 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/vio.c b/arch/ppc64/kernel/vio.c
index 79f2dc7a983..0c0ba71ac0e 100644
--- a/arch/ppc64/kernel/vio.c
+++ b/arch/ppc64/kernel/vio.c
@@ -41,20 +41,25 @@ static const struct vio_device_id *vio_match_device(
 static struct iommu_table *vio_build_iommu_table(struct vio_dev *);
 static int vio_num_address_cells;
 #endif
-static struct vio_dev *vio_bus_device; /* fake "parent" device */
-
 #ifdef CONFIG_PPC_ISERIES
-static struct vio_dev *__init vio_register_device_iseries(char *type,
-		uint32_t unit_num);
-
 static struct iommu_table veth_iommu_table;
 static struct iommu_table vio_iommu_table;
-
-static struct vio_dev _vio_dev  = {
+#endif
+static struct vio_dev vio_bus_device  = { /* fake "parent" device */
+	.name = vio_bus_device.dev.bus_id,
+	.type = "",
+#ifdef CONFIG_PPC_ISERIES
 	.iommu_table = &vio_iommu_table,
-	.dev.bus = &vio_bus_type
+#endif
+	.dev.bus_id = "vio",
+	.dev.bus = &vio_bus_type,
 };
-struct device *iSeries_vio_dev = &_vio_dev.dev;
+
+#ifdef CONFIG_PPC_ISERIES
+static struct vio_dev *__init vio_register_device_iseries(char *type,
+		uint32_t unit_num);
+
+struct device *iSeries_vio_dev = &vio_bus_device.dev;
 EXPORT_SYMBOL(iSeries_vio_dev);
 
 #define device_is_compatible(a, b)	1
@@ -260,18 +265,10 @@ static int __init vio_bus_init(void)
 	}
 
 	/* the fake parent of all vio devices, just to give us a nice directory */
-	vio_bus_device = kmalloc(sizeof(struct vio_dev), GFP_KERNEL);
-	if (!vio_bus_device) {
-		return 1;
-	}
-	memset(vio_bus_device, 0, sizeof(struct vio_dev));
-	strcpy(vio_bus_device->dev.bus_id, "vio");
-
-	err = device_register(&vio_bus_device->dev);
+	err = device_register(&vio_bus_device.dev);
 	if (err) {
 		printk(KERN_WARNING "%s: device_register returned %i\n", __FUNCTION__,
 			err);
-		kfree(vio_bus_device);
 		return err;
 	}
 
@@ -326,7 +323,7 @@ static struct vio_dev * __devinit vio_register_device_common(
 	viodev->unit_address = unit_address;
 	viodev->iommu_table = iommu_table;
 	/* init generic 'struct device' fields: */
-	viodev->dev.parent = &vio_bus_device->dev;
+	viodev->dev.parent = &vio_bus_device.dev;
 	viodev->dev.bus = &vio_bus_type;
 	viodev->dev.release = vio_dev_release;
 
@@ -636,5 +633,3 @@ struct bus_type vio_bus_type = {
 	.name = "vio",
 	.match = vio_bus_match,
 };
-
-EXPORT_SYMBOL(vio_bus_type);
-- 
cgit v1.2.3-70-g09d2


From 515bae9cdc6a78eda0879e1f158056d73ec808b7 Mon Sep 17 00:00:00 2001
From: Anton Blanchard <anton@samba.org>
Date: Tue, 21 Jun 2005 17:15:55 -0700
Subject: [PATCH] ppc64: Mark kernel hptes dirty

We dont use the hardware referenced and changed bits and setting them early
avoids a store to memory.  We already do this for userspace hptes but not
kernel ones.  Do it.

Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/mm/hash_utils.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ppc64/mm/hash_utils.c b/arch/ppc64/mm/hash_utils.c
index 87d0525f56f..1647b1c6f28 100644
--- a/arch/ppc64/mm/hash_utils.c
+++ b/arch/ppc64/mm/hash_utils.c
@@ -195,7 +195,7 @@ void __init htab_initialize(void)
 		memset((void *)table, 0, htab_size_bytes);
 	}
 
-	mode_rw = _PAGE_ACCESSED | _PAGE_COHERENT | PP_RWXX;
+	mode_rw = _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_COHERENT | PP_RWXX;
 
 	/* On U3 based machines, we need to reserve the DART area and
 	 * _NOT_ map it to avoid cache paradoxes as it's remapped non
-- 
cgit v1.2.3-70-g09d2


From 9b843cda193c56f5e12fedeaf95e0126b706d57b Mon Sep 17 00:00:00 2001
From: Anton Blanchard <anton@samba.org>
Date: Tue, 21 Jun 2005 17:15:55 -0700
Subject: [PATCH] ppc64: set/clear SMT capable bit at boot

Allow the SMT bit to be set/reset at boot, like the ALTIVEC bit.  This
means we will enable SMT on unknown cpus that support it.

Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/prom.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/prom.c b/arch/ppc64/kernel/prom.c
index eb6538b5800..47727a6f734 100644
--- a/arch/ppc64/kernel/prom.c
+++ b/arch/ppc64/kernel/prom.c
@@ -884,6 +884,7 @@ static int __init early_init_dt_scan_cpus(unsigned long node,
 {
 	char *type = get_flat_dt_prop(node, "device_type", NULL);
 	u32 *prop;
+	unsigned long size;
 
 	/* We are scanning "cpu" nodes only */
 	if (type == NULL || strcmp(type, "cpu") != 0)
@@ -929,6 +930,17 @@ static int __init early_init_dt_scan_cpus(unsigned long node,
 		cur_cpu_spec->cpu_user_features |= PPC_FEATURE_HAS_ALTIVEC;
 	}
 
+	/*
+	 * Check for an SMT capable CPU and set the CPU feature. We do
+	 * this by looking at the size of the ibm,ppc-interrupt-server#s
+	 * property
+	 */
+	prop = (u32 *)get_flat_dt_prop(node, "ibm,ppc-interrupt-server#s",
+				       &size);
+	cur_cpu_spec->cpu_features &= ~CPU_FTR_SMT;
+	if (prop && ((size / sizeof(u32)) > 1))
+		cur_cpu_spec->cpu_features |= CPU_FTR_SMT;
+
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From e5d310b349b2cbcc0dab31139c92201f332695bb 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: CONFIG split

The SGI IOC4 I/O controller chip drivers are currently all configured by
CONFIG_BLK_DEV_SGIIOC4.  This is undesirable as not all IOC4 hardware features
are needed by all systems.

This patch adds two configuration variables, CONFIG_SGI_IOC4 for core IOC4
driver support (see patch 1/3 in this series for further explanation) and
CONFIG_SERIAL_SGI_IOC4 to independently enable serial port support.

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>
---
 arch/ia64/configs/sn2_defconfig |  6 ++++++
 arch/ia64/defconfig             |  6 ++++++
 drivers/Kconfig                 |  2 ++
 drivers/Makefile                |  2 +-
 drivers/ide/Kconfig             |  4 ++--
 drivers/serial/Kconfig          |  9 +++++++++
 drivers/serial/Makefile         |  2 +-
 drivers/sn/Kconfig              | 20 ++++++++++++++++++++
 drivers/sn/Makefile             |  2 +-
 9 files changed, 48 insertions(+), 5 deletions(-)
 create mode 100644 drivers/sn/Kconfig

(limited to 'arch')

diff --git a/arch/ia64/configs/sn2_defconfig b/arch/ia64/configs/sn2_defconfig
index 6ff7107fee4..a01bb02d074 100644
--- a/arch/ia64/configs/sn2_defconfig
+++ b/arch/ia64/configs/sn2_defconfig
@@ -588,6 +588,7 @@ CONFIG_SGI_MBCS=m
 CONFIG_SERIAL_CORE=y
 CONFIG_SERIAL_CORE_CONSOLE=y
 CONFIG_SERIAL_SGI_L1_CONSOLE=y
+CONFIG_SERIAL_SGI_IOC4=y
 CONFIG_UNIX98_PTYS=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
@@ -787,6 +788,11 @@ CONFIG_INFINIBAND_MTHCA=m
 CONFIG_INFINIBAND_IPOIB=m
 # CONFIG_INFINIBAND_IPOIB_DEBUG is not set
 
+#
+# SN Devices
+#
+CONFIG_SGI_IOC4=y
+
 #
 # File systems
 #
diff --git a/arch/ia64/defconfig b/arch/ia64/defconfig
index 7539e83bf05..9997ef45ab2 100644
--- a/arch/ia64/defconfig
+++ b/arch/ia64/defconfig
@@ -638,6 +638,7 @@ CONFIG_SERIAL_8250_SHARE_IRQ=y
 CONFIG_SERIAL_CORE=y
 CONFIG_SERIAL_CORE_CONSOLE=y
 CONFIG_SERIAL_SGI_L1_CONSOLE=y
+CONFIG_SERIAL_SGI_IOC4=y
 CONFIG_UNIX98_PTYS=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
@@ -942,6 +943,11 @@ CONFIG_INFINIBAND_MTHCA=m
 CONFIG_INFINIBAND_IPOIB=m
 # CONFIG_INFINIBAND_IPOIB_DEBUG is not set
 
+#
+# SN Devices
+#
+CONFIG_SGI_IOC4=y
+
 #
 # File systems
 #
diff --git a/drivers/Kconfig b/drivers/Kconfig
index ed41d9036bf..aed4a9b97c1 100644
--- a/drivers/Kconfig
+++ b/drivers/Kconfig
@@ -58,4 +58,6 @@ source "drivers/mmc/Kconfig"
 
 source "drivers/infiniband/Kconfig"
 
+source "drivers/sn/Kconfig"
+
 endmenu
diff --git a/drivers/Makefile b/drivers/Makefile
index 15681de688f..3167be54fed 100644
--- a/drivers/Makefile
+++ b/drivers/Makefile
@@ -61,6 +61,6 @@ obj-$(CONFIG_EISA)		+= eisa/
 obj-$(CONFIG_CPU_FREQ)		+= cpufreq/
 obj-$(CONFIG_MMC)		+= mmc/
 obj-$(CONFIG_INFINIBAND)	+= infiniband/
-obj-$(CONFIG_BLK_DEV_SGIIOC4)	+= sn/
+obj-$(CONFIG_SGI_IOC4)		+= sn/
 obj-y				+= firmware/
 obj-$(CONFIG_CRYPTO)		+= crypto/
diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig
index 3ac0a535b4a..0273f124a4f 100644
--- a/drivers/ide/Kconfig
+++ b/drivers/ide/Kconfig
@@ -672,8 +672,8 @@ config BLK_DEV_SVWKS
 	  chipsets.
 
 config BLK_DEV_SGIIOC4
-	tristate "Silicon Graphics IOC4 chipset support"
-	depends on IA64_SGI_SN2 || IA64_GENERIC
+	tristate "Silicon Graphics IOC4 chipset ATA/ATAPI support"
+	depends on (IA64_SGI_SN2 || IA64_GENERIC) && SGI_IOC4
 	help
 	  This driver adds PIO & MultiMode DMA-2 support for the SGI IOC4
 	  chipset, which has one channel and can support two devices.
diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig
index 6a15703f1cb..25fcef2c42d 100644
--- a/drivers/serial/Kconfig
+++ b/drivers/serial/Kconfig
@@ -843,4 +843,13 @@ config SERIAL_JSM
           To compile this driver as a module, choose M here: the
           module will be called jsm.
 
+config SERIAL_SGI_IOC4
+	tristate "SGI IOC4 controller serial support"
+	depends on (IA64_GENERIC || IA64_SGI_SN2) && SGI_IOC4
+	select SERIAL_CORE
+	help
+		If you have an SGI Altix with an IOC4 based Base IO card
+		and wish to use the serial ports on this card, say Y.
+		Otherwise, say N.
+
 endmenu
diff --git a/drivers/serial/Makefile b/drivers/serial/Makefile
index 81b77d769b8..8f1cdde7dbe 100644
--- a/drivers/serial/Makefile
+++ b/drivers/serial/Makefile
@@ -51,4 +51,4 @@ obj-$(CONFIG_ETRAX_SERIAL) += crisv10.o
 obj-$(CONFIG_SERIAL_JSM) += jsm/
 obj-$(CONFIG_SERIAL_TXX9) += serial_txx9.o
 obj-$(CONFIG_SERIAL_VR41XX) += vr41xx_siu.o
-obj-$(CONFIG_BLK_DEV_SGIIOC4) += ioc4_serial.o
+obj-$(CONFIG_SERIAL_SGI_IOC4) += ioc4_serial.o
diff --git a/drivers/sn/Kconfig b/drivers/sn/Kconfig
new file mode 100644
index 00000000000..20f7515ab83
--- /dev/null
+++ b/drivers/sn/Kconfig
@@ -0,0 +1,20 @@
+#
+# Miscellaneous SN-specific devices
+#
+
+menu "SN Devices"
+
+config SGI_IOC4
+	tristate "SGI IOC4 Base IO support"
+	depends on IA64_GENERIC || IA64_SGI_SN2
+	default m
+	---help---
+	This option enables basic support for the SGI IOC4-based Base IO
+	controller card.  This option does not enable any specific
+	functions on such a card, but provides necessary infrastructure
+	for other drivers to utilize.
+
+	If you have an SGI Altix with an IOC4-based
+	I/O controller say Y.  Otherwise say N.
+
+endmenu
diff --git a/drivers/sn/Makefile b/drivers/sn/Makefile
index 631e5495844..c2a28418537 100644
--- a/drivers/sn/Makefile
+++ b/drivers/sn/Makefile
@@ -3,4 +3,4 @@
 #
 #
 
-obj-$(CONFIG_BLK_DEV_SGIIOC4) += ioc4.o
+obj-$(CONFIG_SGI_IOC4) += ioc4.o
-- 
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 'arch')

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 6f973b001a9b511900fb73d816adf77d8755838c Mon Sep 17 00:00:00 2001
From: Hirokazu Takata <takata@linux-m32r.org>
Date: Tue, 21 Jun 2005 17:16:13 -0700
Subject: [PATCH] m32r: Update setup_xxxxx.c

Change coding styles of hw_interrupt_type struct's initialization portions.

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/kernel/setup_m32700ut.c | 56 +++++++++++++++++++--------------------
 arch/m32r/kernel/setup_mappi.c    | 14 +++++-----
 arch/m32r/kernel/setup_mappi2.c   | 14 +++++-----
 arch/m32r/kernel/setup_mappi3.c   | 14 +++++-----
 arch/m32r/kernel/setup_oaks32r.c  | 14 +++++-----
 arch/m32r/kernel/setup_opsput.c   | 28 ++++++++++----------
 arch/m32r/kernel/setup_usrv.c     | 28 ++++++++++----------
 7 files changed, 84 insertions(+), 84 deletions(-)

(limited to 'arch')

diff --git a/arch/m32r/kernel/setup_m32700ut.c b/arch/m32r/kernel/setup_m32700ut.c
index 488aa87bab7..b014e2c1e52 100644
--- a/arch/m32r/kernel/setup_m32700ut.c
+++ b/arch/m32r/kernel/setup_m32700ut.c
@@ -78,13 +78,13 @@ static void shutdown_m32700ut_irq(unsigned int irq)
 
 static struct hw_interrupt_type m32700ut_irq_type =
 {
-	"M32700UT-IRQ",
-	startup_m32700ut_irq,
-	shutdown_m32700ut_irq,
-	enable_m32700ut_irq,
-	disable_m32700ut_irq,
-	mask_and_ack_m32700ut,
-	end_m32700ut_irq
+	.typename = "M32700UT-IRQ",
+	.startup = startup_m32700ut_irq,
+	.shutdown = shutdown_m32700ut_irq,
+	.enable = enable_m32700ut_irq,
+	.disable = disable_m32700ut_irq,
+	.ack = mask_and_ack_m32700ut,
+	.end = end_m32700ut_irq
 };
 
 /*
@@ -155,13 +155,13 @@ static void shutdown_m32700ut_pld_irq(unsigned int irq)
 
 static struct hw_interrupt_type m32700ut_pld_irq_type =
 {
-	"M32700UT-PLD-IRQ",
-	startup_m32700ut_pld_irq,
-	shutdown_m32700ut_pld_irq,
-	enable_m32700ut_pld_irq,
-	disable_m32700ut_pld_irq,
-	mask_and_ack_m32700ut_pld,
-	end_m32700ut_pld_irq
+	.typename = "M32700UT-PLD-IRQ",
+	.startup = startup_m32700ut_pld_irq,
+	.shutdown = shutdown_m32700ut_pld_irq,
+	.enable = enable_m32700ut_pld_irq,
+	.disable = disable_m32700ut_pld_irq,
+	.ack = mask_and_ack_m32700ut_pld,
+	.end = end_m32700ut_pld_irq
 };
 
 /*
@@ -224,13 +224,13 @@ static void shutdown_m32700ut_lanpld_irq(unsigned int irq)
 
 static struct hw_interrupt_type m32700ut_lanpld_irq_type =
 {
-	"M32700UT-PLD-LAN-IRQ",
-	startup_m32700ut_lanpld_irq,
-	shutdown_m32700ut_lanpld_irq,
-	enable_m32700ut_lanpld_irq,
-	disable_m32700ut_lanpld_irq,
-	mask_and_ack_m32700ut_lanpld,
-	end_m32700ut_lanpld_irq
+	.typename = "M32700UT-PLD-LAN-IRQ",
+	.startup = startup_m32700ut_lanpld_irq,
+	.shutdown = shutdown_m32700ut_lanpld_irq,
+	.enable = enable_m32700ut_lanpld_irq,
+	.disable = disable_m32700ut_lanpld_irq,
+	.ack = mask_and_ack_m32700ut_lanpld,
+	.end = end_m32700ut_lanpld_irq
 };
 
 /*
@@ -293,13 +293,13 @@ static void shutdown_m32700ut_lcdpld_irq(unsigned int irq)
 
 static struct hw_interrupt_type m32700ut_lcdpld_irq_type =
 {
-	"M32700UT-PLD-LCD-IRQ",
-	startup_m32700ut_lcdpld_irq,
-	shutdown_m32700ut_lcdpld_irq,
-	enable_m32700ut_lcdpld_irq,
-	disable_m32700ut_lcdpld_irq,
-	mask_and_ack_m32700ut_lcdpld,
-	end_m32700ut_lcdpld_irq
+	.typename = "M32700UT-PLD-LCD-IRQ",
+	.startup = startup_m32700ut_lcdpld_irq,
+	.shutdown = shutdown_m32700ut_lcdpld_irq,
+	.enable = enable_m32700ut_lcdpld_irq,
+	.disable = disable_m32700ut_lcdpld_irq,
+	.ack = mask_and_ack_m32700ut_lcdpld,
+	.end = end_m32700ut_lcdpld_irq
 };
 
 void __init init_IRQ(void)
diff --git a/arch/m32r/kernel/setup_mappi.c b/arch/m32r/kernel/setup_mappi.c
index 1e74110f067..aaf8e569b93 100644
--- a/arch/m32r/kernel/setup_mappi.c
+++ b/arch/m32r/kernel/setup_mappi.c
@@ -70,13 +70,13 @@ static void shutdown_mappi_irq(unsigned int irq)
 
 static struct hw_interrupt_type mappi_irq_type =
 {
-	"MAPPI-IRQ",
-	startup_mappi_irq,
-	shutdown_mappi_irq,
-	enable_mappi_irq,
-	disable_mappi_irq,
-	mask_and_ack_mappi,
-	end_mappi_irq
+	.typename = "MAPPI-IRQ",
+	.startup = startup_mappi_irq,
+	.shutdown = shutdown_mappi_irq,
+	.enable = enable_mappi_irq,
+	.disable = disable_mappi_irq,
+	.ack = mask_and_ack_mappi,
+	.end = end_mappi_irq
 };
 
 void __init init_IRQ(void)
diff --git a/arch/m32r/kernel/setup_mappi2.c b/arch/m32r/kernel/setup_mappi2.c
index 248fc658dbd..38d5e9a4142 100644
--- a/arch/m32r/kernel/setup_mappi2.c
+++ b/arch/m32r/kernel/setup_mappi2.c
@@ -79,13 +79,13 @@ static void shutdown_mappi2_irq(unsigned int irq)
 
 static struct hw_interrupt_type mappi2_irq_type =
 {
-	"MAPPI2-IRQ",
-	startup_mappi2_irq,
-	shutdown_mappi2_irq,
-	enable_mappi2_irq,
-	disable_mappi2_irq,
-	mask_and_ack_mappi2,
-	end_mappi2_irq
+	.typename = "MAPPI2-IRQ",
+	.startup = startup_mappi2_irq,
+	.shutdown = shutdown_mappi2_irq,
+	.enable = enable_mappi2_irq,
+	.disable = disable_mappi2_irq,
+	.ack = mask_and_ack_mappi2,
+	.end = end_mappi2_irq
 };
 
 void __init init_IRQ(void)
diff --git a/arch/m32r/kernel/setup_mappi3.c b/arch/m32r/kernel/setup_mappi3.c
index 72b5b7609ce..3d60a85aaec 100644
--- a/arch/m32r/kernel/setup_mappi3.c
+++ b/arch/m32r/kernel/setup_mappi3.c
@@ -79,13 +79,13 @@ static void shutdown_mappi3_irq(unsigned int irq)
 
 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
+	.typename = "MAPPI3-IRQ",
+	.startup = startup_mappi3_irq,
+	.shutdown = shutdown_mappi3_irq,
+	.enable = enable_mappi3_irq,
+	.disable = disable_mappi3_irq,
+	.ack = mask_and_ack_mappi3,
+	.end = end_mappi3_irq
 };
 
 void __init init_IRQ(void)
diff --git a/arch/m32r/kernel/setup_oaks32r.c b/arch/m32r/kernel/setup_oaks32r.c
index b04834526c9..d656640badc 100644
--- a/arch/m32r/kernel/setup_oaks32r.c
+++ b/arch/m32r/kernel/setup_oaks32r.c
@@ -70,13 +70,13 @@ static void shutdown_oaks32r_irq(unsigned int irq)
 
 static struct hw_interrupt_type oaks32r_irq_type =
 {
-	"OAKS32R-IRQ",
-	startup_oaks32r_irq,
-	shutdown_oaks32r_irq,
-	enable_oaks32r_irq,
-	disable_oaks32r_irq,
-	mask_and_ack_mappi,
-	end_oaks32r_irq
+	.typename = "OAKS32R-IRQ",
+	.startup = startup_oaks32r_irq,
+	.shutdown = shutdown_oaks32r_irq,
+	.enable = enable_oaks32r_irq,
+	.disable = disable_oaks32r_irq,
+	.ack = mask_and_ack_mappi,
+	.end = end_oaks32r_irq
 };
 
 void __init init_IRQ(void)
diff --git a/arch/m32r/kernel/setup_opsput.c b/arch/m32r/kernel/setup_opsput.c
index 84315e344c5..86f4cf2a86c 100644
--- a/arch/m32r/kernel/setup_opsput.c
+++ b/arch/m32r/kernel/setup_opsput.c
@@ -79,13 +79,13 @@ static void shutdown_opsput_irq(unsigned int irq)
 
 static struct hw_interrupt_type opsput_irq_type =
 {
-	"OPSPUT-IRQ",
-	startup_opsput_irq,
-	shutdown_opsput_irq,
-	enable_opsput_irq,
-	disable_opsput_irq,
-	mask_and_ack_opsput,
-	end_opsput_irq
+	.typename = "OPSPUT-IRQ",
+	.startup = startup_opsput_irq,
+	.shutdown = shutdown_opsput_irq,
+	.enable = enable_opsput_irq,
+	.disable = disable_opsput_irq,
+	.ack = mask_and_ack_opsput,
+	.end = end_opsput_irq
 };
 
 /*
@@ -156,13 +156,13 @@ static void shutdown_opsput_pld_irq(unsigned int irq)
 
 static struct hw_interrupt_type opsput_pld_irq_type =
 {
-	"OPSPUT-PLD-IRQ",
-	startup_opsput_pld_irq,
-	shutdown_opsput_pld_irq,
-	enable_opsput_pld_irq,
-	disable_opsput_pld_irq,
-	mask_and_ack_opsput_pld,
-	end_opsput_pld_irq
+	.typename = "OPSPUT-PLD-IRQ",
+	.startup = startup_opsput_pld_irq,
+	.shutdown = shutdown_opsput_pld_irq,
+	.enable = enable_opsput_pld_irq,
+	.disable = disable_opsput_pld_irq,
+	.ack = mask_and_ack_opsput_pld,
+	.end = end_opsput_pld_irq
 };
 
 /*
diff --git a/arch/m32r/kernel/setup_usrv.c b/arch/m32r/kernel/setup_usrv.c
index fe417be5e3e..634741bf9d3 100644
--- a/arch/m32r/kernel/setup_usrv.c
+++ b/arch/m32r/kernel/setup_usrv.c
@@ -70,13 +70,13 @@ static void shutdown_mappi_irq(unsigned int irq)
 
 static struct hw_interrupt_type mappi_irq_type =
 {
-	"M32700-IRQ",
-	startup_mappi_irq,
-	shutdown_mappi_irq,
-	enable_mappi_irq,
-	disable_mappi_irq,
-	mask_and_ack_mappi,
-	end_mappi_irq
+	.typename = "M32700-IRQ",
+	.startup = startup_mappi_irq,
+	.shutdown = shutdown_mappi_irq,
+	.enable = enable_mappi_irq,
+	.disable = disable_mappi_irq,
+	.ack = mask_and_ack_mappi,
+	.end = end_mappi_irq
 };
 
 /*
@@ -143,13 +143,13 @@ static void shutdown_m32700ut_pld_irq(unsigned int irq)
 
 static struct hw_interrupt_type m32700ut_pld_irq_type =
 {
-	"USRV-PLD-IRQ",
-	startup_m32700ut_pld_irq,
-	shutdown_m32700ut_pld_irq,
-	enable_m32700ut_pld_irq,
-	disable_m32700ut_pld_irq,
-	mask_and_ack_m32700ut_pld,
-	end_m32700ut_pld_irq
+	.typename = "USRV-PLD-IRQ",
+	.startup = startup_m32700ut_pld_irq,
+	.shutdown = shutdown_m32700ut_pld_irq,
+	.enable = enable_m32700ut_pld_irq,
+	.disable = disable_m32700ut_pld_irq,
+	.ack = mask_and_ack_m32700ut_pld,
+	.end = end_m32700ut_pld_irq
 };
 
 void __init init_IRQ(void)
-- 
cgit v1.2.3-70-g09d2


From 1cc1265e9a8579be0029cd675ec22a6151893e04 Mon Sep 17 00:00:00 2001
From: Hirokazu Takata <takata@linux-m32r.org>
Date: Tue, 21 Jun 2005 17:16:15 -0700
Subject: [PATCH] m32r: Cleanup arch/m32r/mm/extable.c

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/mm/extable.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

(limited to 'arch')

diff --git a/arch/m32r/mm/extable.c b/arch/m32r/mm/extable.c
index 9a97363b652..1743f23d49a 100644
--- a/arch/m32r/mm/extable.c
+++ b/arch/m32r/mm/extable.c
@@ -1,10 +1,8 @@
 /*
- * linux/arch/i386/mm/extable.c
+ * linux/arch/m32r/mm/extable.c
  */
 
-#include <linux/config.h>
 #include <linux/module.h>
-#include <linux/spinlock.h>
 #include <asm/uaccess.h>
 
 int fixup_exception(struct pt_regs *regs)
@@ -19,4 +17,3 @@ int fixup_exception(struct pt_regs *regs)
 
 	return 0;
 }
-
-- 
cgit v1.2.3-70-g09d2


From 960c2a89a051333d2e6793a416d9a33c4b116b41 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: Update defconfig files

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/defconfig                       | 52 ++++++++++++++++++------------
 arch/m32r/m32700ut/defconfig.m32700ut.smp | 53 +++++++++++++++++++------------
 arch/m32r/m32700ut/defconfig.m32700ut.up  | 52 ++++++++++++++++++------------
 arch/m32r/mappi/defconfig.nommu           | 46 +++++++++++++++------------
 arch/m32r/mappi/defconfig.smp             | 49 +++++++++++++++-------------
 arch/m32r/mappi/defconfig.up              | 48 +++++++++++++++-------------
 arch/m32r/mappi2/defconfig.vdec2          | 46 +++++++++++++++------------
 arch/m32r/oaks32r/defconfig.nommu         | 50 ++++++++++++++---------------
 arch/m32r/opsput/defconfig.opsput         | 45 ++++++++++++++------------
 9 files changed, 247 insertions(+), 194 deletions(-)

(limited to 'arch')

diff --git a/arch/m32r/defconfig b/arch/m32r/defconfig
index 417c95ba87e..85309309316 100644
--- a/arch/m32r/defconfig
+++ b/arch/m32r/defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.11-rc4
-# Wed Feb 16 21:10:44 2005
+# Linux kernel version: 2.6.12-rc5
+# Fri Jun  3 16:20:11 2005
 #
 CONFIG_M32R=y
 # CONFIG_UID16 is not set
@@ -16,6 +16,7 @@ CONFIG_EXPERIMENTAL=y
 CONFIG_CLEAN_COMPILE=y
 CONFIG_BROKEN_ON_SMP=y
 CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -28,13 +29,15 @@ CONFIG_BSD_PROCESS_ACCT=y
 # CONFIG_BSD_PROCESS_ACCT_V3 is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
-CONFIG_LOG_BUF_SHIFT=14
 CONFIG_HOTPLUG=y
 CONFIG_KOBJECT_UEVENT=y
 CONFIG_IKCONFIG=y
 # CONFIG_IKCONFIG_PROC 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
@@ -44,6 +47,7 @@ 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
@@ -65,6 +69,7 @@ CONFIG_PLAT_M32700UT=y
 # CONFIG_PLAT_OPSPUT is not set
 # CONFIG_PLAT_OAKS32R is not set
 # CONFIG_PLAT_MAPPI2 is not set
+# CONFIG_PLAT_MAPPI3 is not set
 CONFIG_CHIP_M32700=y
 # CONFIG_CHIP_M32102 is not set
 # CONFIG_CHIP_VDEC2 is not set
@@ -268,7 +273,6 @@ CONFIG_NET=y
 #
 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
@@ -392,18 +396,6 @@ CONFIG_INPUT=y
 # 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=y
-# CONFIG_SERIO_I8042 is not set
-CONFIG_SERIO_SERPORT=y
-# CONFIG_SERIO_CT82C710 is not set
-# CONFIG_SERIO_LIBPS2 is not set
-# CONFIG_SERIO_RAW is not set
-
 #
 # Input Device Drivers
 #
@@ -413,6 +405,17 @@ CONFIG_SERIO_SERPORT=y
 # 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=y
+# CONFIG_SERIO_LIBPS2 is not set
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+CONFIG_SOUND_GAMEPORT=y
+
 #
 # Character devices
 #
@@ -464,6 +467,10 @@ CONFIG_DS1302=y
 # CONFIG_SYNCLINK_CS is not set
 # CONFIG_RAW_DRIVER is not set
 
+#
+# TPM devices
+#
+
 #
 # I2C support
 #
@@ -508,8 +515,14 @@ CONFIG_VIDEO_M32R_AR_M64278=y
 # Graphics support
 #
 CONFIG_FB=y
+# CONFIG_FB_CFB_FILLRECT is not set
+# CONFIG_FB_CFB_COPYAREA is not set
+# CONFIG_FB_CFB_IMAGEBLIT is not set
+# CONFIG_FB_SOFT_CURSOR is not set
+# CONFIG_FB_MACMODES is not set
 # CONFIG_FB_MODE_HELPERS is not set
 # CONFIG_FB_TILEBLITTING is not set
+# CONFIG_FB_S1D13XXX is not set
 # CONFIG_FB_VIRTUAL is not set
 
 #
@@ -542,10 +555,6 @@ CONFIG_LOGO_LINUX_CLUT224=y
 # 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
 #
@@ -719,8 +728,9 @@ CONFIG_NLS_DEFAULT="iso8859-1"
 #
 # Kernel hacking
 #
+# CONFIG_PRINTK_TIME is not set
 # CONFIG_DEBUG_KERNEL is not set
-CONFIG_DEBUG_PREEMPT=y
+CONFIG_LOG_BUF_SHIFT=14
 # CONFIG_DEBUG_BUGVERBOSE is not set
 # CONFIG_FRAME_POINTER is not set
 
diff --git a/arch/m32r/m32700ut/defconfig.m32700ut.smp b/arch/m32r/m32700ut/defconfig.m32700ut.smp
index 1c002181df7..3e607d90b5f 100644
--- a/arch/m32r/m32700ut/defconfig.m32700ut.smp
+++ b/arch/m32r/m32700ut/defconfig.m32700ut.smp
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.11-rc4
-# Wed Feb 16 21:10:50 2005
+# Linux kernel version: 2.6.12-rc5
+# Fri Jun  3 16:20:58 2005
 #
 CONFIG_M32R=y
 # CONFIG_UID16 is not set
@@ -15,6 +15,7 @@ CONFIG_GENERIC_IRQ_PROBE=y
 CONFIG_EXPERIMENTAL=y
 CONFIG_CLEAN_COMPILE=y
 CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -27,13 +28,16 @@ CONFIG_BSD_PROCESS_ACCT=y
 # CONFIG_BSD_PROCESS_ACCT_V3 is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
-CONFIG_LOG_BUF_SHIFT=15
 CONFIG_HOTPLUG=y
 CONFIG_KOBJECT_UEVENT=y
 CONFIG_IKCONFIG=y
 # CONFIG_IKCONFIG_PROC is not set
+# 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
@@ -43,6 +47,7 @@ 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
@@ -65,6 +70,7 @@ CONFIG_PLAT_M32700UT=y
 # CONFIG_PLAT_OPSPUT is not set
 # CONFIG_PLAT_OAKS32R is not set
 # CONFIG_PLAT_MAPPI2 is not set
+# CONFIG_PLAT_MAPPI3 is not set
 CONFIG_CHIP_M32700=y
 # CONFIG_CHIP_M32102 is not set
 # CONFIG_CHIP_VDEC2 is not set
@@ -271,7 +277,6 @@ CONFIG_NET=y
 #
 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
@@ -395,18 +400,6 @@ CONFIG_INPUT=y
 # 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=y
-# CONFIG_SERIO_I8042 is not set
-CONFIG_SERIO_SERPORT=y
-# CONFIG_SERIO_CT82C710 is not set
-# CONFIG_SERIO_LIBPS2 is not set
-# CONFIG_SERIO_RAW is not set
-
 #
 # Input Device Drivers
 #
@@ -416,6 +409,17 @@ CONFIG_SERIO_SERPORT=y
 # 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=y
+# CONFIG_SERIO_LIBPS2 is not set
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+CONFIG_SOUND_GAMEPORT=y
+
 #
 # Character devices
 #
@@ -467,6 +471,10 @@ CONFIG_DS1302=y
 # CONFIG_SYNCLINK_CS is not set
 # CONFIG_RAW_DRIVER is not set
 
+#
+# TPM devices
+#
+
 #
 # I2C support
 #
@@ -511,8 +519,14 @@ CONFIG_VIDEO_M32R_AR_M64278=y
 # Graphics support
 #
 CONFIG_FB=y
+# CONFIG_FB_CFB_FILLRECT is not set
+# CONFIG_FB_CFB_COPYAREA is not set
+# CONFIG_FB_CFB_IMAGEBLIT is not set
+# CONFIG_FB_SOFT_CURSOR is not set
+# CONFIG_FB_MACMODES is not set
 # CONFIG_FB_MODE_HELPERS is not set
 # CONFIG_FB_TILEBLITTING is not set
+# CONFIG_FB_S1D13XXX is not set
 # CONFIG_FB_VIRTUAL is not set
 
 #
@@ -545,10 +559,6 @@ CONFIG_LOGO_LINUX_CLUT224=y
 # 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
 #
@@ -722,8 +732,9 @@ CONFIG_NLS_DEFAULT="iso8859-1"
 #
 # Kernel hacking
 #
+# CONFIG_PRINTK_TIME is not set
 # CONFIG_DEBUG_KERNEL is not set
-CONFIG_DEBUG_PREEMPT=y
+CONFIG_LOG_BUF_SHIFT=15
 # CONFIG_DEBUG_BUGVERBOSE is not set
 # CONFIG_FRAME_POINTER is not set
 
diff --git a/arch/m32r/m32700ut/defconfig.m32700ut.up b/arch/m32r/m32700ut/defconfig.m32700ut.up
index 805357f7bda..2d3e7cda8f4 100644
--- a/arch/m32r/m32700ut/defconfig.m32700ut.up
+++ b/arch/m32r/m32700ut/defconfig.m32700ut.up
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.11-rc4
-# Wed Feb 16 21:10:54 2005
+# Linux kernel version: 2.6.12-rc5
+# Fri Jun  3 16:21:34 2005
 #
 CONFIG_M32R=y
 # CONFIG_UID16 is not set
@@ -16,6 +16,7 @@ CONFIG_EXPERIMENTAL=y
 CONFIG_CLEAN_COMPILE=y
 CONFIG_BROKEN_ON_SMP=y
 CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -28,13 +29,15 @@ CONFIG_BSD_PROCESS_ACCT=y
 # CONFIG_BSD_PROCESS_ACCT_V3 is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
-CONFIG_LOG_BUF_SHIFT=14
 CONFIG_HOTPLUG=y
 CONFIG_KOBJECT_UEVENT=y
 CONFIG_IKCONFIG=y
 # CONFIG_IKCONFIG_PROC 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
@@ -44,6 +47,7 @@ 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
@@ -65,6 +69,7 @@ CONFIG_PLAT_M32700UT=y
 # CONFIG_PLAT_OPSPUT is not set
 # CONFIG_PLAT_OAKS32R is not set
 # CONFIG_PLAT_MAPPI2 is not set
+# CONFIG_PLAT_MAPPI3 is not set
 CONFIG_CHIP_M32700=y
 # CONFIG_CHIP_M32102 is not set
 # CONFIG_CHIP_VDEC2 is not set
@@ -268,7 +273,6 @@ CONFIG_NET=y
 #
 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
@@ -392,18 +396,6 @@ CONFIG_INPUT=y
 # 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=y
-# CONFIG_SERIO_I8042 is not set
-CONFIG_SERIO_SERPORT=y
-# CONFIG_SERIO_CT82C710 is not set
-# CONFIG_SERIO_LIBPS2 is not set
-# CONFIG_SERIO_RAW is not set
-
 #
 # Input Device Drivers
 #
@@ -413,6 +405,17 @@ CONFIG_SERIO_SERPORT=y
 # 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=y
+# CONFIG_SERIO_LIBPS2 is not set
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+CONFIG_SOUND_GAMEPORT=y
+
 #
 # Character devices
 #
@@ -464,6 +467,10 @@ CONFIG_DS1302=y
 # CONFIG_SYNCLINK_CS is not set
 # CONFIG_RAW_DRIVER is not set
 
+#
+# TPM devices
+#
+
 #
 # I2C support
 #
@@ -508,8 +515,14 @@ CONFIG_VIDEO_M32R_AR_M64278=y
 # Graphics support
 #
 CONFIG_FB=y
+# CONFIG_FB_CFB_FILLRECT is not set
+# CONFIG_FB_CFB_COPYAREA is not set
+# CONFIG_FB_CFB_IMAGEBLIT is not set
+# CONFIG_FB_SOFT_CURSOR is not set
+# CONFIG_FB_MACMODES is not set
 # CONFIG_FB_MODE_HELPERS is not set
 # CONFIG_FB_TILEBLITTING is not set
+# CONFIG_FB_S1D13XXX is not set
 # CONFIG_FB_VIRTUAL is not set
 
 #
@@ -542,10 +555,6 @@ CONFIG_LOGO_LINUX_CLUT224=y
 # 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
 #
@@ -719,8 +728,9 @@ CONFIG_NLS_DEFAULT="iso8859-1"
 #
 # Kernel hacking
 #
+# CONFIG_PRINTK_TIME is not set
 # CONFIG_DEBUG_KERNEL is not set
-CONFIG_DEBUG_PREEMPT=y
+CONFIG_LOG_BUF_SHIFT=14
 # CONFIG_DEBUG_BUGVERBOSE is not set
 # CONFIG_FRAME_POINTER is not set
 
diff --git a/arch/m32r/mappi/defconfig.nommu b/arch/m32r/mappi/defconfig.nommu
index 714aa6e7034..a8425fba340 100644
--- a/arch/m32r/mappi/defconfig.nommu
+++ b/arch/m32r/mappi/defconfig.nommu
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.11-rc4
-# Wed Feb 16 21:10:57 2005
+# Linux kernel version: 2.6.12-rc5
+# Fri Jun  3 16:21:46 2005
 #
 CONFIG_M32R=y
 # CONFIG_UID16 is not set
@@ -16,6 +16,7 @@ CONFIG_EXPERIMENTAL=y
 CONFIG_CLEAN_COMPILE=y
 CONFIG_BROKEN_ON_SMP=y
 CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -26,13 +27,15 @@ CONFIG_BSD_PROCESS_ACCT=y
 # CONFIG_BSD_PROCESS_ACCT_V3 is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
-CONFIG_LOG_BUF_SHIFT=14
 CONFIG_HOTPLUG=y
 CONFIG_KOBJECT_UEVENT=y
 CONFIG_IKCONFIG=y
 # CONFIG_IKCONFIG_PROC 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
@@ -41,6 +44,7 @@ CONFIG_CC_ALIGN_LABELS=0
 CONFIG_CC_ALIGN_LOOPS=0
 CONFIG_CC_ALIGN_JUMPS=0
 CONFIG_TINY_SHMEM=y
+CONFIG_BASE_SMALL=0
 
 #
 # Loadable module support
@@ -62,6 +66,7 @@ CONFIG_PLAT_MAPPI=y
 # CONFIG_PLAT_OPSPUT is not set
 # CONFIG_PLAT_OAKS32R is not set
 # CONFIG_PLAT_MAPPI2 is not set
+# CONFIG_PLAT_MAPPI3 is not set
 CONFIG_CHIP_M32700=y
 # CONFIG_CHIP_M32102 is not set
 # CONFIG_CHIP_VDEC2 is not set
@@ -202,7 +207,6 @@ CONFIG_NET=y
 #
 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
@@ -324,18 +328,6 @@ CONFIG_INPUT=y
 # 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=y
-# CONFIG_SERIO_I8042 is not set
-CONFIG_SERIO_SERPORT=y
-# CONFIG_SERIO_CT82C710 is not set
-# CONFIG_SERIO_LIBPS2 is not set
-# CONFIG_SERIO_RAW is not set
-
 #
 # Input Device Drivers
 #
@@ -345,6 +337,17 @@ CONFIG_SERIO_SERPORT=y
 # 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=y
+# CONFIG_SERIO_LIBPS2 is not set
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+CONFIG_SOUND_GAMEPORT=y
+
 #
 # Character devices
 #
@@ -393,6 +396,10 @@ CONFIG_LEGACY_PTY_COUNT=256
 # CONFIG_SYNCLINK_CS is not set
 # CONFIG_RAW_DRIVER is not set
 
+#
+# TPM devices
+#
+
 #
 # I2C support
 #
@@ -433,10 +440,6 @@ CONFIG_LEGACY_PTY_COUNT=256
 # 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
 #
@@ -595,8 +598,9 @@ CONFIG_NLS_DEFAULT="iso8859-1"
 #
 # Kernel hacking
 #
+# CONFIG_PRINTK_TIME is not set
 # CONFIG_DEBUG_KERNEL is not set
-CONFIG_DEBUG_PREEMPT=y
+CONFIG_LOG_BUF_SHIFT=14
 # CONFIG_DEBUG_BUGVERBOSE is not set
 # CONFIG_FRAME_POINTER is not set
 
diff --git a/arch/m32r/mappi/defconfig.smp b/arch/m32r/mappi/defconfig.smp
index 956a8e2f98d..1a7f3cd4224 100644
--- a/arch/m32r/mappi/defconfig.smp
+++ b/arch/m32r/mappi/defconfig.smp
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.11-rc4
-# Wed Feb 16 21:11:02 2005
+# Linux kernel version: 2.6.12-rc5
+# Fri Jun  3 16:21:52 2005
 #
 CONFIG_M32R=y
 # CONFIG_UID16 is not set
@@ -17,6 +17,7 @@ CONFIG_EXPERIMENTAL=y
 CONFIG_BROKEN=y
 CONFIG_BROKEN_ON_SMP=y
 CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -28,13 +29,16 @@ CONFIG_SYSVIPC=y
 # CONFIG_BSD_PROCESS_ACCT is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
-CONFIG_LOG_BUF_SHIFT=15
 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
@@ -44,6 +48,7 @@ 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
@@ -66,6 +71,7 @@ CONFIG_PLAT_MAPPI=y
 # CONFIG_PLAT_OPSPUT is not set
 # CONFIG_PLAT_OAKS32R is not set
 # CONFIG_PLAT_MAPPI2 is not set
+# CONFIG_PLAT_MAPPI3 is not set
 CONFIG_CHIP_M32700=y
 # CONFIG_CHIP_M32102 is not set
 # CONFIG_CHIP_VDEC2 is not set
@@ -139,8 +145,8 @@ CONFIG_PREVENT_FIRMWARE_BUILD=y
 #
 CONFIG_MTD=y
 # CONFIG_MTD_DEBUG is not set
-CONFIG_MTD_PARTITIONS=y
 # 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
@@ -294,7 +300,6 @@ CONFIG_NET=y
 # Networking options
 #
 # CONFIG_PACKET is not set
-# CONFIG_NETLINK_DEV is not set
 CONFIG_UNIX=y
 # CONFIG_NET_KEY is not set
 CONFIG_INET=y
@@ -419,18 +424,6 @@ CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
 # 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=y
-# CONFIG_SERIO_I8042 is not set
-# CONFIG_SERIO_SERPORT is not set
-# CONFIG_SERIO_CT82C710 is not set
-# CONFIG_SERIO_LIBPS2 is not set
-# CONFIG_SERIO_RAW is not set
-
 #
 # Input Device Drivers
 #
@@ -440,6 +433,17 @@ CONFIG_SERIO=y
 # 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
 #
@@ -488,6 +492,10 @@ CONFIG_LEGACY_PTY_COUNT=256
 # CONFIG_SYNCLINK_CS is not set
 # CONFIG_RAW_DRIVER is not set
 
+#
+# TPM devices
+#
+
 #
 # I2C support
 #
@@ -528,10 +536,6 @@ CONFIG_LEGACY_PTY_COUNT=256
 # 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
 #
@@ -708,8 +712,9 @@ CONFIG_NLS_DEFAULT="iso8859-1"
 #
 # Kernel hacking
 #
+# CONFIG_PRINTK_TIME is not set
 # CONFIG_DEBUG_KERNEL is not set
-CONFIG_DEBUG_PREEMPT=y
+CONFIG_LOG_BUF_SHIFT=15
 # CONFIG_DEBUG_BUGVERBOSE is not set
 # CONFIG_FRAME_POINTER is not set
 
diff --git a/arch/m32r/mappi/defconfig.up b/arch/m32r/mappi/defconfig.up
index c9253e94efb..38910fb7669 100644
--- a/arch/m32r/mappi/defconfig.up
+++ b/arch/m32r/mappi/defconfig.up
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.11-rc4
-# Wed Feb 16 21:11:07 2005
+# Linux kernel version: 2.6.12-rc5
+# Fri Jun  3 16:21:59 2005
 #
 CONFIG_M32R=y
 # CONFIG_UID16 is not set
@@ -17,6 +17,7 @@ CONFIG_EXPERIMENTAL=y
 CONFIG_BROKEN=y
 CONFIG_BROKEN_ON_SMP=y
 CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -28,13 +29,15 @@ CONFIG_SYSVIPC=y
 # CONFIG_BSD_PROCESS_ACCT is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
-CONFIG_LOG_BUF_SHIFT=14
 CONFIG_HOTPLUG=y
 CONFIG_KOBJECT_UEVENT=y
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
 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
@@ -44,6 +47,7 @@ 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
@@ -65,6 +69,7 @@ CONFIG_PLAT_MAPPI=y
 # CONFIG_PLAT_OPSPUT is not set
 # CONFIG_PLAT_OAKS32R is not set
 # CONFIG_PLAT_MAPPI2 is not set
+# CONFIG_PLAT_MAPPI3 is not set
 CONFIG_CHIP_M32700=y
 # CONFIG_CHIP_M32102 is not set
 # CONFIG_CHIP_VDEC2 is not set
@@ -135,8 +140,8 @@ CONFIG_PREVENT_FIRMWARE_BUILD=y
 #
 CONFIG_MTD=y
 # CONFIG_MTD_DEBUG is not set
-CONFIG_MTD_PARTITIONS=y
 # 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
@@ -290,7 +295,6 @@ CONFIG_NET=y
 # Networking options
 #
 # CONFIG_PACKET is not set
-# CONFIG_NETLINK_DEV is not set
 CONFIG_UNIX=y
 # CONFIG_NET_KEY is not set
 CONFIG_INET=y
@@ -415,18 +419,6 @@ CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
 # 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=y
-# CONFIG_SERIO_I8042 is not set
-# CONFIG_SERIO_SERPORT is not set
-# CONFIG_SERIO_CT82C710 is not set
-# CONFIG_SERIO_LIBPS2 is not set
-# CONFIG_SERIO_RAW is not set
-
 #
 # Input Device Drivers
 #
@@ -436,6 +428,17 @@ CONFIG_SERIO=y
 # 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
 #
@@ -484,6 +487,10 @@ CONFIG_LEGACY_PTY_COUNT=256
 # CONFIG_SYNCLINK_CS is not set
 # CONFIG_RAW_DRIVER is not set
 
+#
+# TPM devices
+#
+
 #
 # I2C support
 #
@@ -524,10 +531,6 @@ CONFIG_LEGACY_PTY_COUNT=256
 # 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
 #
@@ -704,8 +707,9 @@ CONFIG_NLS_DEFAULT="iso8859-1"
 #
 # Kernel hacking
 #
+# CONFIG_PRINTK_TIME is not set
 # CONFIG_DEBUG_KERNEL is not set
-CONFIG_DEBUG_PREEMPT=y
+CONFIG_LOG_BUF_SHIFT=14
 # CONFIG_DEBUG_BUGVERBOSE is not set
 # CONFIG_FRAME_POINTER is not set
 
diff --git a/arch/m32r/mappi2/defconfig.vdec2 b/arch/m32r/mappi2/defconfig.vdec2
index c14791d2730..56f287b6589 100644
--- a/arch/m32r/mappi2/defconfig.vdec2
+++ b/arch/m32r/mappi2/defconfig.vdec2
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.11-rc4
-# Wed Feb 16 21:11:10 2005
+# Linux kernel version: 2.6.12-rc5
+# Fri Jun  3 16:22:02 2005
 #
 CONFIG_M32R=y
 # CONFIG_UID16 is not set
@@ -16,6 +16,7 @@ CONFIG_EXPERIMENTAL=y
 CONFIG_CLEAN_COMPILE=y
 CONFIG_BROKEN_ON_SMP=y
 CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -28,13 +29,15 @@ CONFIG_BSD_PROCESS_ACCT=y
 # CONFIG_BSD_PROCESS_ACCT_V3 is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
-CONFIG_LOG_BUF_SHIFT=14
 CONFIG_HOTPLUG=y
 CONFIG_KOBJECT_UEVENT=y
 CONFIG_IKCONFIG=y
 # CONFIG_IKCONFIG_PROC 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
@@ -44,6 +47,7 @@ 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
@@ -65,6 +69,7 @@ CONFIG_KMOD=y
 # CONFIG_PLAT_OPSPUT is not set
 # CONFIG_PLAT_OAKS32R is not set
 CONFIG_PLAT_MAPPI2=y
+# CONFIG_PLAT_MAPPI3 is not set
 # CONFIG_CHIP_M32700 is not set
 # CONFIG_CHIP_M32102 is not set
 CONFIG_CHIP_VDEC2=y
@@ -264,7 +269,6 @@ CONFIG_NET=y
 #
 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
@@ -388,18 +392,6 @@ CONFIG_INPUT=y
 # 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=y
-# CONFIG_SERIO_I8042 is not set
-CONFIG_SERIO_SERPORT=y
-# CONFIG_SERIO_CT82C710 is not set
-# CONFIG_SERIO_LIBPS2 is not set
-# CONFIG_SERIO_RAW is not set
-
 #
 # Input Device Drivers
 #
@@ -409,6 +401,17 @@ CONFIG_SERIO_SERPORT=y
 # 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=y
+# CONFIG_SERIO_LIBPS2 is not set
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+CONFIG_SOUND_GAMEPORT=y
+
 #
 # Character devices
 #
@@ -459,6 +462,10 @@ CONFIG_LEGACY_PTY_COUNT=256
 # CONFIG_SYNCLINK_CS is not set
 # CONFIG_RAW_DRIVER is not set
 
+#
+# TPM devices
+#
+
 #
 # I2C support
 #
@@ -520,10 +527,6 @@ CONFIG_DUMMY_CONSOLE=y
 # 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
 #
@@ -697,8 +700,9 @@ CONFIG_NLS_DEFAULT="iso8859-1"
 #
 # Kernel hacking
 #
+# CONFIG_PRINTK_TIME is not set
 # CONFIG_DEBUG_KERNEL is not set
-CONFIG_DEBUG_PREEMPT=y
+CONFIG_LOG_BUF_SHIFT=14
 # CONFIG_DEBUG_BUGVERBOSE is not set
 # CONFIG_FRAME_POINTER is not set
 
diff --git a/arch/m32r/oaks32r/defconfig.nommu b/arch/m32r/oaks32r/defconfig.nommu
index f2da9be726e..3f9fe519acb 100644
--- a/arch/m32r/oaks32r/defconfig.nommu
+++ b/arch/m32r/oaks32r/defconfig.nommu
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.11-rc4
-# Wed Feb 16 21:11:13 2005
+# Linux kernel version: 2.6.12-rc5
+# Fri Jun  3 16:22:04 2005
 #
 CONFIG_M32R=y
 # CONFIG_UID16 is not set
@@ -16,6 +16,7 @@ CONFIG_EXPERIMENTAL=y
 CONFIG_CLEAN_COMPILE=y
 CONFIG_BROKEN_ON_SMP=y
 CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -26,12 +27,14 @@ CONFIG_BSD_PROCESS_ACCT=y
 # CONFIG_BSD_PROCESS_ACCT_V3 is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
-CONFIG_LOG_BUF_SHIFT=14
 CONFIG_HOTPLUG=y
 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 is not set
 # CONFIG_EPOLL is not set
 # CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
@@ -40,6 +43,7 @@ CONFIG_CC_ALIGN_LABELS=0
 CONFIG_CC_ALIGN_LOOPS=0
 CONFIG_CC_ALIGN_JUMPS=0
 CONFIG_TINY_SHMEM=y
+CONFIG_BASE_SMALL=0
 
 #
 # Loadable module support
@@ -61,6 +65,7 @@ CONFIG_KMOD=y
 # CONFIG_PLAT_OPSPUT is not set
 CONFIG_PLAT_OAKS32R=y
 # CONFIG_PLAT_MAPPI2 is not set
+# CONFIG_PLAT_MAPPI3 is not set
 # CONFIG_CHIP_M32700 is not set
 CONFIG_CHIP_M32102=y
 # CONFIG_CHIP_VDEC2 is not set
@@ -91,10 +96,6 @@ CONFIG_PREEMPT=y
 #
 # CONFIG_PCCARD is not set
 
-#
-# PC-card bridges
-#
-
 #
 # PCI Hotplug Support
 #
@@ -193,7 +194,6 @@ CONFIG_NET=y
 #
 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
@@ -310,18 +310,6 @@ CONFIG_INPUT=y
 # 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=y
-# CONFIG_SERIO_I8042 is not set
-CONFIG_SERIO_SERPORT=y
-# CONFIG_SERIO_CT82C710 is not set
-# CONFIG_SERIO_LIBPS2 is not set
-# CONFIG_SERIO_RAW is not set
-
 #
 # Input Device Drivers
 #
@@ -331,6 +319,17 @@ CONFIG_SERIO_SERPORT=y
 # 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=y
+# CONFIG_SERIO_LIBPS2 is not set
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+CONFIG_SOUND_GAMEPORT=y
+
 #
 # Character devices
 #
@@ -374,6 +373,10 @@ CONFIG_LEGACY_PTY_COUNT=256
 # CONFIG_DRM is not set
 # CONFIG_RAW_DRIVER is not set
 
+#
+# TPM devices
+#
+
 #
 # I2C support
 #
@@ -414,10 +417,6 @@ CONFIG_LEGACY_PTY_COUNT=256
 # 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
 #
@@ -574,8 +573,9 @@ CONFIG_NLS_DEFAULT="iso8859-1"
 #
 # Kernel hacking
 #
+# CONFIG_PRINTK_TIME is not set
 # CONFIG_DEBUG_KERNEL is not set
-CONFIG_DEBUG_PREEMPT=y
+CONFIG_LOG_BUF_SHIFT=14
 # CONFIG_DEBUG_BUGVERBOSE is not set
 # CONFIG_FRAME_POINTER is not set
 
diff --git a/arch/m32r/opsput/defconfig.opsput b/arch/m32r/opsput/defconfig.opsput
index a87e1ea4c30..66adec6a701 100644
--- a/arch/m32r/opsput/defconfig.opsput
+++ b/arch/m32r/opsput/defconfig.opsput
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.11-rc4
-# Wed Feb 16 21:11:41 2005
+# Linux kernel version: 2.6.12-rc5
+# Fri Jun  3 16:22:06 2005
 #
 CONFIG_M32R=y
 # CONFIG_UID16 is not set
@@ -15,6 +15,7 @@ CONFIG_GENERIC_IRQ_PROBE=y
 CONFIG_EXPERIMENTAL=y
 CONFIG_CLEAN_COMPILE=y
 CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -27,13 +28,15 @@ CONFIG_BSD_PROCESS_ACCT=y
 # CONFIG_BSD_PROCESS_ACCT_V3 is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
-CONFIG_LOG_BUF_SHIFT=14
 CONFIG_HOTPLUG=y
 CONFIG_KOBJECT_UEVENT=y
 CONFIG_IKCONFIG=y
 # CONFIG_IKCONFIG_PROC 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
@@ -43,6 +46,7 @@ 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
@@ -64,6 +68,7 @@ CONFIG_KMOD=y
 CONFIG_PLAT_OPSPUT=y
 # CONFIG_PLAT_OAKS32R is not set
 # CONFIG_PLAT_MAPPI2 is not set
+# CONFIG_PLAT_MAPPI3 is not set
 # CONFIG_CHIP_M32700 is not set
 # CONFIG_CHIP_M32102 is not set
 # CONFIG_CHIP_VDEC2 is not set
@@ -243,7 +248,6 @@ CONFIG_NET=y
 #
 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
@@ -367,18 +371,6 @@ CONFIG_INPUT=y
 # 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=y
-# CONFIG_SERIO_I8042 is not set
-CONFIG_SERIO_SERPORT=y
-# CONFIG_SERIO_CT82C710 is not set
-# CONFIG_SERIO_LIBPS2 is not set
-# CONFIG_SERIO_RAW is not set
-
 #
 # Input Device Drivers
 #
@@ -388,6 +380,17 @@ CONFIG_SERIO_SERPORT=y
 # 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=y
+# CONFIG_SERIO_LIBPS2 is not set
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+CONFIG_SOUND_GAMEPORT=y
+
 #
 # Character devices
 #
@@ -437,6 +440,10 @@ CONFIG_DS1302=y
 # CONFIG_SYNCLINK_CS is not set
 # CONFIG_RAW_DRIVER is not set
 
+#
+# TPM devices
+#
+
 #
 # I2C support
 #
@@ -477,10 +484,6 @@ CONFIG_DS1302=y
 # 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
 #
@@ -654,8 +657,10 @@ CONFIG_NLS_DEFAULT="iso8859-1"
 #
 # Kernel hacking
 #
+# CONFIG_PRINTK_TIME is not set
 CONFIG_DEBUG_KERNEL=y
 # CONFIG_MAGIC_SYSRQ is not set
+CONFIG_LOG_BUF_SHIFT=14
 # CONFIG_SCHEDSTATS is not set
 # CONFIG_DEBUG_SLAB is not set
 # CONFIG_DEBUG_SPINLOCK is not set
-- 
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 'arch')

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 faec1e99ba9ca7371d9aee1656938373133c4b21 Mon Sep 17 00:00:00 2001
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Date: Tue, 21 Jun 2005 17:16:21 -0700
Subject: [PATCH] uml: complete hw_controller_type->release conversion

This occurrence of free_irq_by_irq_and_dev() was missed when converting UML
to the use of hw_controller_type->release.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/drivers/line.c | 1 -
 1 file changed, 1 deletion(-)

(limited to 'arch')

diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c
index 562f864254b..0f59736db32 100644
--- a/arch/um/drivers/line.c
+++ b/arch/um/drivers/line.c
@@ -756,7 +756,6 @@ static void unregister_winch(struct tty_struct *tty)
         if(winch->pid != -1)
                 os_kill_process(winch->pid, 1);
 
-        free_irq_by_irq_and_dev(WINCH_IRQ, winch);
         free_irq(WINCH_IRQ, winch);
         list_del(&winch->list);
         kfree(winch);
-- 
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 'arch')

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 60b2737de1b1ddfdb90f3ba622634eb49d6f3603 Mon Sep 17 00:00:00 2001
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Date: Tue, 21 Jun 2005 17:16:25 -0700
Subject: [PATCH] uml: fix linkage of tt mode against NPTL

With Al Viro <viro@parcelfarce.linux.theplanet.co.uk>

To make sure switcheroo() can execute when we remap all the executable
image, we used a trick to make it use a local copy of errno...  this trick
does not work with NPTL glibc, only with LinuxThreads, so use another
(simpler) one to make it work anyway.

Hopefully, a lot improved thanks to merging with the version of Al Viro
(which had his part of problems, though, i.e.  removing a fix to another
bug and not fixing the problem on i386).

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/Makefile               |  2 +-
 arch/um/kernel/tt/Makefile     | 15 ---------------
 arch/um/kernel/tt/unmap.c      | 31 -------------------------------
 arch/um/kernel/uml.lds.S       | 15 +++------------
 arch/um/scripts/Makefile.rules |  6 ++++++
 arch/um/scripts/Makefile.unmap | 22 ++++++++++++++++++++++
 arch/um/sys-i386/Makefile      |  2 ++
 arch/um/sys-i386/unmap.c       | 25 +++++++++++++++++++++++++
 arch/um/sys-x86_64/Makefile    |  2 ++
 arch/um/sys-x86_64/unmap.c     | 25 +++++++++++++++++++++++++
 10 files changed, 86 insertions(+), 59 deletions(-)
 delete mode 100644 arch/um/kernel/tt/unmap.c
 create mode 100644 arch/um/scripts/Makefile.unmap
 create mode 100644 arch/um/sys-i386/unmap.c
 create mode 100644 arch/um/sys-x86_64/unmap.c

(limited to 'arch')

diff --git a/arch/um/Makefile b/arch/um/Makefile
index f2a0c40a920..3f073902351 100644
--- a/arch/um/Makefile
+++ b/arch/um/Makefile
@@ -122,7 +122,7 @@ endif
 CPPFLAGS_vmlinux.lds = $(shell echo -U$(SUBARCH) \
 	-DSTART=$(START) -DELF_ARCH=$(ELF_ARCH) \
 	-DELF_FORMAT=\"$(ELF_FORMAT)\" $(CPP_MODE-y) \
-	-DKERNEL_STACK_SIZE=$(STACK_SIZE))
+	-DKERNEL_STACK_SIZE=$(STACK_SIZE) -DSUBARCH=$(SUBARCH))
 
 #The wrappers will select whether using "malloc" or the kernel allocator.
 LINK_WRAPS = -Wl,--wrap,malloc -Wl,--wrap,free -Wl,--wrap,calloc
diff --git a/arch/um/kernel/tt/Makefile b/arch/um/kernel/tt/Makefile
index c3faea21a99..3fd2554e60b 100644
--- a/arch/um/kernel/tt/Makefile
+++ b/arch/um/kernel/tt/Makefile
@@ -3,10 +3,6 @@
 # Licensed under the GPL
 #
 
-extra-y := unmap_fin.o
-targets := unmap.o
-clean-files := unmap_tmp.o
-
 obj-y = exec_kern.o exec_user.o gdb.o ksyms.o mem.o mem_user.o process_kern.o \
 	syscall_kern.o syscall_user.o time.o tlb.o tracer.o trap_user.o \
 	uaccess.o uaccess_user.o
@@ -16,14 +12,3 @@ obj-$(CONFIG_PT_PROXY) += gdb_kern.o ptproxy/
 USER_OBJS := gdb.o time.o tracer.o
 
 include arch/um/scripts/Makefile.rules
-
-UNMAP_CFLAGS := $(patsubst -pg -DPROFILING,,$(USER_CFLAGS))
-UNMAP_CFLAGS := $(patsubst -fprofile-arcs -ftest-coverage,,$(UNMAP_CFLAGS))
-
-#XXX: partially copied from arch/um/scripts/Makefile.rules
-$(obj)/unmap.o: c_flags = -Wp,-MD,$(depfile) $(UNMAP_CFLAGS)
-
-$(obj)/unmap_fin.o : $(obj)/unmap.o
-	$(LD) -r -o $(obj)/unmap_tmp.o $< $(shell $(CC) -print-file-name=libc.a)
-	$(OBJCOPY) $(obj)/unmap_tmp.o $@ -G switcheroo
-
diff --git a/arch/um/kernel/tt/unmap.c b/arch/um/kernel/tt/unmap.c
deleted file mode 100644
index 3f7aecdbe53..00000000000
--- a/arch/um/kernel/tt/unmap.c
+++ /dev/null
@@ -1,31 +0,0 @@
-/* 
- * Copyright (C) 2000 Jeff Dike (jdike@karaya.com)
- * Licensed under the GPL
- */
-
-#include <sys/mman.h>
-
-int switcheroo(int fd, int prot, void *from, void *to, int size)
-{
-	if(munmap(to, size) < 0){
-		return(-1);
-	}
-	if(mmap(to, size, prot,	MAP_SHARED | MAP_FIXED, fd, 0) != to){
-		return(-1);
-	}
-	if(munmap(from, size) < 0){
-		return(-1);
-	}
-	return(0);
-}
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
diff --git a/arch/um/kernel/uml.lds.S b/arch/um/kernel/uml.lds.S
index dd5355500bd..61dfd4fef75 100644
--- a/arch/um/kernel/uml.lds.S
+++ b/arch/um/kernel/uml.lds.S
@@ -14,19 +14,10 @@ SECTIONS
   /* Used in arch/um/kernel/mem.c. Any memory between START and __binary_start
    * is remapped.*/
   __binary_start = .;
-#ifdef MODE_TT
-  .thread_private : {
-    __start_thread_private = .;
-    errno = .;
-    . += 4;
-    arch/um/kernel/tt/unmap_fin.o (.data)
-    __end_thread_private = .;
-  }
-  . = ALIGN(4096);
-  .remap : { arch/um/kernel/tt/unmap_fin.o (.text) }
 
-  /* We want it only if we are in MODE_TT. In both cases, however, when MODE_TT
-   * is off the resulting binary segfaults.*/
+#ifdef MODE_TT
+  .remap_data : { arch/um/sys-SUBARCH/unmap_fin.o (.data .bss) }
+  .remap : { arch/um/sys-SUBARCH/unmap_fin.o (.text) }
 
   . = ALIGN(4096);		/* Init code and data */
 #endif
diff --git a/arch/um/scripts/Makefile.rules b/arch/um/scripts/Makefile.rules
index 98346c71149..7459d09c233 100644
--- a/arch/um/scripts/Makefile.rules
+++ b/arch/um/scripts/Makefile.rules
@@ -10,6 +10,12 @@ USER_OBJS := $(foreach file,$(USER_OBJS),$(obj)/$(file))
 $(USER_OBJS) : c_flags = -Wp,-MD,$(depfile) $(USER_CFLAGS) \
 	$(CFLAGS_$(notdir $@))
 
+# The stubs and unmap.o can't try to call mcount or update basic block data
+define unprofile
+	$(patsubst -pg,,$(patsubst -fprofile-arcs -ftest-coverage,,$(1)))
+endef
+
+
 quiet_cmd_make_link = SYMLINK $@
 cmd_make_link       = ln -sf $(srctree)/arch/$(SUBARCH)/$($(notdir $@)-dir)/$(notdir $@) $@
 
diff --git a/arch/um/scripts/Makefile.unmap b/arch/um/scripts/Makefile.unmap
new file mode 100644
index 00000000000..37a8f976529
--- /dev/null
+++ b/arch/um/scripts/Makefile.unmap
@@ -0,0 +1,22 @@
+clean-files += unmap_tmp.o unmap_fin.o unmap.o
+
+ifdef CONFIG_MODE_TT
+
+#Always build unmap_fin.o
+extra-y += unmap_fin.o
+#Do dependency tracking for unmap.o (it will be always built, but won't get the tracking unless we use this).
+targets += unmap.o
+
+#XXX: partially copied from arch/um/scripts/Makefile.rules
+$(obj)/unmap.o: _c_flags = $(call unprofile,$(CFLAGS))
+
+quiet_cmd_wrapld = LD      $@
+define cmd_wrapld
+	$(LD) -r -o $(obj)/unmap_tmp.o $< $(shell $(CC) -print-file-name=libc.a); \
+	$(OBJCOPY) $(obj)/unmap_tmp.o $@ -G switcheroo
+endef
+
+$(obj)/unmap_fin.o : $(obj)/unmap.o FORCE
+	$(call if_changed,wrapld)
+
+endif
diff --git a/arch/um/sys-i386/Makefile b/arch/um/sys-i386/Makefile
index 4351e560550..095bcdb0b9c 100644
--- a/arch/um/sys-i386/Makefile
+++ b/arch/um/sys-i386/Makefile
@@ -17,3 +17,5 @@ highmem.c-dir = mm
 module.c-dir = kernel
 
 subdir- := util
+
+include arch/um/scripts/Makefile.unmap
diff --git a/arch/um/sys-i386/unmap.c b/arch/um/sys-i386/unmap.c
new file mode 100644
index 00000000000..136875263d2
--- /dev/null
+++ b/arch/um/sys-i386/unmap.c
@@ -0,0 +1,25 @@
+/*
+ * Copyright (C) 2000 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include <linux/mman.h>
+#include <asm/unistd.h>
+
+static int errno;
+
+static inline _syscall2(int,munmap,void *,start,size_t,len)
+static inline _syscall6(void *,mmap2,void *,addr,size_t,len,int,prot,int,flags,int,fd,off_t,offset)
+int switcheroo(int fd, int prot, void *from, void *to, int size)
+{
+	if(munmap(to, size) < 0){
+		return(-1);
+	}
+	if(mmap2(to, size, prot, MAP_SHARED | MAP_FIXED, fd, 0) != to){
+		return(-1);
+	}
+	if(munmap(from, size) < 0){
+		return(-1);
+	}
+	return(0);
+}
diff --git a/arch/um/sys-x86_64/Makefile b/arch/um/sys-x86_64/Makefile
index 608466ad6b2..2bc6f684901 100644
--- a/arch/um/sys-x86_64/Makefile
+++ b/arch/um/sys-x86_64/Makefile
@@ -29,3 +29,5 @@ thunk.S-dir = lib
 module.c-dir = kernel
 
 subdir- := util
+
+include arch/um/scripts/Makefile.unmap
diff --git a/arch/um/sys-x86_64/unmap.c b/arch/um/sys-x86_64/unmap.c
new file mode 100644
index 00000000000..bc7094cce47
--- /dev/null
+++ b/arch/um/sys-x86_64/unmap.c
@@ -0,0 +1,25 @@
+/*
+ * Copyright (C) 2000 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include <linux/mman.h>
+#include <asm/unistd.h>
+
+static int errno;
+
+static inline _syscall2(int,munmap,void *,start,size_t,len)
+static inline _syscall6(void *,mmap,void *,addr,size_t,len,int,prot,int,flags,int,fd,off_t,offset)
+int switcheroo(int fd, int prot, void *from, void *to, int size)
+{
+	if(munmap(to, size) < 0){
+		return(-1);
+	}
+	if(mmap(to, size, prot, MAP_SHARED | MAP_FIXED, fd, 0) != to){
+		return(-1);
+	}
+	if(munmap(from, size) < 0){
+		return(-1);
+	}
+	return(0);
+}
-- 
cgit v1.2.3-70-g09d2


From 14651c798a2a4d15ccc53d064df28bfd0283bad6 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Tue, 21 Jun 2005 17:16:28 -0700
Subject: [PATCH] s390: #ifdefs in compat_ioctls

Remove superflous #if .. #endif pairs from compat_ioctl.c.

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/compat_ioctl.c | 6 ------
 1 file changed, 6 deletions(-)

(limited to 'arch')

diff --git a/arch/s390/kernel/compat_ioctl.c b/arch/s390/kernel/compat_ioctl.c
index 03d03c6d3cb..24a1e9f069a 100644
--- a/arch/s390/kernel/compat_ioctl.c
+++ b/arch/s390/kernel/compat_ioctl.c
@@ -42,7 +42,6 @@ struct ioctl_trans ioctl_start[] = {
 #include "../../../fs/compat_ioctl.c"
 
 /* s390 only ioctls */
-#if defined(CONFIG_DASD) || defined(CONFIG_DASD_MODULE)
 COMPATIBLE_IOCTL(DASDAPIVER)
 COMPATIBLE_IOCTL(BIODASDDISABLE)
 COMPATIBLE_IOCTL(BIODASDENABLE)
@@ -59,16 +58,11 @@ COMPATIBLE_IOCTL(BIODASDPRRD)
 COMPATIBLE_IOCTL(BIODASDPSRD)
 COMPATIBLE_IOCTL(BIODASDGATTR)
 COMPATIBLE_IOCTL(BIODASDSATTR)
-#if defined(CONFIG_DASD_CMB) || defined(CONFIG_DASD_CMB_MODULE)
 COMPATIBLE_IOCTL(BIODASDCMFENABLE)
 COMPATIBLE_IOCTL(BIODASDCMFDISABLE)
 COMPATIBLE_IOCTL(BIODASDREADALLCMB)
-#endif
-#endif
 
-#if defined(CONFIG_S390_TAPE) || defined(CONFIG_S390_TAPE_MODULE)
 COMPATIBLE_IOCTL(TAPE390_DISPLAY)
-#endif
 
 /* s390 doesn't need handlers here */
 COMPATIBLE_IOCTL(TIOCGSERIAL)
-- 
cgit v1.2.3-70-g09d2


From 77eb65cbc18d86a9c334964ed4c3d178f5560918 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 21 Jun 2005 17:16:28 -0700
Subject: [PATCH] s390: kernel stack overflow panic

die() doesn't return, therefore print registers and then panic instead.

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/traps.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
index 8b90e9528b9..ca34b6f34b3 100644
--- a/arch/s390/kernel/traps.c
+++ b/arch/s390/kernel/traps.c
@@ -668,7 +668,10 @@ asmlinkage void space_switch_exception(struct pt_regs * regs, long int_code)
 
 asmlinkage void kernel_stack_overflow(struct pt_regs * regs)
 {
-	die("Kernel stack overflow", regs, 0);
+	bust_spinlocks(1);
+	printk("Kernel stack overflow.\n");
+	show_regs(regs);
+	bust_spinlocks(0);
 	panic("Corrupt kernel stack, can't continue.");
 }
 
-- 
cgit v1.2.3-70-g09d2


From 447570cfde680d5bf09c47b9c8d1dcf5bcb18f10 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 21 Jun 2005 17:16:29 -0700
Subject: [PATCH] s390: cmm sender parameter visibility

Make cmm module parameter "sender" visible in sysfs.

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/mm/cmm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/s390/mm/cmm.c b/arch/s390/mm/cmm.c
index f5a5bc09b8f..2d5cb138575 100644
--- a/arch/s390/mm/cmm.c
+++ b/arch/s390/mm/cmm.c
@@ -21,7 +21,7 @@
 #include <asm/uaccess.h>
 
 static char *sender = "VMRMSVM";
-module_param(sender, charp, 0);
+module_param(sender, charp, 0400);
 MODULE_PARM_DESC(sender,
 		 "Guest name that may send SMSG messages (default VMRMSVM)");
 
-- 
cgit v1.2.3-70-g09d2


From e9b9a04796eade5241452a913ec6f3433437b4f5 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 21 Jun 2005 17:16:30 -0700
Subject: [PATCH] s390: memory detection > 32GB

The kernel takes a very long time to boot if the memory size is bigger then
32767 MB.  The memory size is contained in a structure created by an sclp
call.  The kernel accesses the field with a LH instrution which performs a
sign extension of a 16 bit word.  In the case of a memory size with bit 2^15
set this results in a very large value and the memory detection just loops for
a long time.  In addition if more then 64 GB are used on a 64 bit system the
memory size is read from an incorrect storage location.

Use zero-extention to read the 16 bit memory size and the correct offset to
read the 4 byte memory size on 64 bit.

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/head.S   | 8 ++++----
 arch/s390/kernel/head64.S | 6 +++---
 2 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'arch')

diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S
index b804c55bd91..658e8de74f8 100644
--- a/arch/s390/kernel/head.S
+++ b/arch/s390/kernel/head.S
@@ -517,10 +517,10 @@ startup:basr  %r13,0                     # get base
 	l     %r2, .Lrcp2-.LPG1(%r13)	# try with Read SCP
 	b     .Lservicecall-.LPG1(%r13)
 .Lprocsccb:
-	lh    %r1,.Lscpincr1-PARMAREA(%r4) # use this one if != 0
-	chi   %r1,0x00
-	jne   .Lscnd
-	l     %r1,.Lscpincr2-PARMAREA(%r4) # otherwise use this one
+	lhi   %r1,0
+	icm   %r1,3,.Lscpincr1-PARMAREA(%r4) # use this one if != 0
+	jnz   .Lscnd
+	l     %r1,.Lscpincr2-PARMAREA+4(%r4) # otherwise use this one
 .Lscnd:
 	xr    %r3,%r3			# same logic
 	ic    %r3,.Lscpa1-PARMAREA(%r4)
diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S
index 8366793bc37..8e2caf9dc18 100644
--- a/arch/s390/kernel/head64.S
+++ b/arch/s390/kernel/head64.S
@@ -518,9 +518,9 @@ startup:basr  %r13,0                     # get base
 	l     %r2,.Lrcp2-.LPG1(%r13)	# try with Read SCP
 	b     .Lservicecall-.LPG1(%r13)
 .Lprocsccb:
-	lh    %r1,.Lscpincr1-PARMAREA(%r4) # use this one if != 0
-	chi   %r1,0x00
-	jne   .Lscnd
+	lghi  %r1,0
+	icm   %r1,3,.Lscpincr1-PARMAREA(%r4) # use this one if != 0
+	jnz   .Lscnd
 	lg    %r1,.Lscpincr2-PARMAREA(%r4) # otherwise use this one
 .Lscnd:
 	xr    %r3,%r3			# same logic
-- 
cgit v1.2.3-70-g09d2


From 2b071886170df456e230c38a3e504da4a11fff79 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 21 Jun 2005 17:16:31 -0700
Subject: [PATCH] s390: pending interrupt after ipl from reader

Wait for interrupt and clear status pending after resetting the reader.

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/head.S   | 19 +++++++++++++++++--
 arch/s390/kernel/head64.S | 19 +++++++++++++++++--
 2 files changed, 34 insertions(+), 4 deletions(-)

(limited to 'arch')

diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S
index 658e8de74f8..fc8bf5e285f 100644
--- a/arch/s390/kernel/head.S
+++ b/arch/s390/kernel/head.S
@@ -345,10 +345,25 @@ iplstart:
 	bno   .Lnoreset
         la    %r2,.Lreset              
         lhi   %r3,26
-        .long 0x83230008
+	diag  %r2,%r3,8
+	mvc   0x78(8),.Lrdrnewpsw              # set up IO interrupt psw
+.Lwaitrdrirq:
+	lpsw  .Lrdrwaitpsw
+.Lrdrint:
+	c     %r1,0xb8                         # compare subchannel number
+	bne   .Lwaitrdrirq
+	la    %r5,.Lirb
+	tsch  0(%r5)
 .Lnoreset:
+        b     .Lnoload
+
+	.align 8
+.Lrdrnewpsw:
+	.long  0x00080000,0x80000000+.Lrdrint
+.Lrdrwaitpsw:
+	.long  0x020a0000,0x80000000+.Lrdrint
 #endif
-	
+
 #
 # everything loaded, go for it
 #
diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S
index 8e2caf9dc18..f525c0c2125 100644
--- a/arch/s390/kernel/head64.S
+++ b/arch/s390/kernel/head64.S
@@ -344,10 +344,25 @@ iplstart:
 	bno   .Lnoreset
         la    %r2,.Lreset              
         lhi   %r3,26
-        .long 0x83230008
+	diag  %r2,%r3,8
+	mvc   0x78(8),.Lrdrnewpsw	       # set up IO interrupt psw
+.Lwaitrdrirq:
+	lpsw  .Lrdrwaitpsw
+.Lrdrint:
+	c     %r1,0xb8			       # compare subchannel number
+	bne   .Lwaitrdrirq
+	la    %r5,.Lirb
+	tsch  0(%r5)
 .Lnoreset:
+	b     .Lnoload
+
+	.align 8
+.Lrdrnewpsw:
+	.long  0x00080000,0x80000000+.Lrdrint
+.Lrdrwaitpsw:
+	.long  0x020a0000,0x80000000+.Lrdrint
 #endif
-	
+
 #
 # everything loaded, go for it
 #
-- 
cgit v1.2.3-70-g09d2


From bdca3f0aedde85552099aa95ab1449bf81e4f6f5 Mon Sep 17 00:00:00 2001
From: Randy Vinson <rvinson@mvista.com>
Date: Fri, 3 Jun 2005 14:43:56 -0700
Subject: [PATCH] I2C: Add support for Maxim/Dallas DS1374 Real-Time Clock Chip
 (2/2)

This change provides support for the DS1374 Real-Time Clock chip present
on the MPC8349ADS board. It depends on a previous patch which adds I2C
support for the DS1374.

Signed-off-by: Randy Vinson <rvinson@mvista.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/ppc/platforms/83xx/mpc834x_sys.c | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

(limited to 'arch')

diff --git a/arch/ppc/platforms/83xx/mpc834x_sys.c b/arch/ppc/platforms/83xx/mpc834x_sys.c
index 86ca5cf8126..37ece154279 100644
--- a/arch/ppc/platforms/83xx/mpc834x_sys.c
+++ b/arch/ppc/platforms/83xx/mpc834x_sys.c
@@ -185,6 +185,26 @@ mpc834x_sys_init_IRQ(void)
 	ipic_set_default_priority();
 }
 
+#if defined(CONFIG_I2C_MPC) && defined(CONFIG_SENSORS_DS1374)
+extern ulong	ds1374_get_rtc_time(void);
+extern int	ds1374_set_rtc_time(ulong);
+
+static int __init
+mpc834x_rtc_hookup(void)
+{
+	struct timespec	tv;
+
+	ppc_md.get_rtc_time = ds1374_get_rtc_time;
+	ppc_md.set_rtc_time = ds1374_set_rtc_time;
+
+	tv.tv_nsec = 0;
+	tv.tv_sec = (ppc_md.get_rtc_time)();
+	do_settimeofday(&tv);
+
+	return 0;
+}
+late_initcall(mpc834x_rtc_hookup);
+#endif
 static __inline__ void
 mpc834x_sys_set_bat(void)
 {
-- 
cgit v1.2.3-70-g09d2


From 522c37b9d3bc2554264c2d7cbba439571a2043fb Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 22 Jun 2005 09:52:26 +0100
Subject: [PATCH] ARM: Fix sa1111.c build error caused by klist changes

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/common/sa1111.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/common/sa1111.c b/arch/arm/common/sa1111.c
index 21fce3414ed..38c2eb667eb 100644
--- a/arch/arm/common/sa1111.c
+++ b/arch/arm/common/sa1111.c
@@ -721,16 +721,17 @@ __sa1111_probe(struct device *me, struct resource *mem, int irq)
 	return ret;
 }
 
+static int sa1111_remove_one(struct device *dev, void *data)
+{
+	device_unregister(dev);
+	return 0;
+}
+
 static void __sa1111_remove(struct sa1111 *sachip)
 {
-	struct list_head *l, *n;
 	void __iomem *irqbase = sachip->base + SA1111_INTC;
 
-	list_for_each_safe(l, n, &sachip->dev->children) {
-		struct device *d = list_to_dev(l);
-
-		device_unregister(d);
-	}
+	device_for_each_child(sachip->dev, NULL, sa1111_remove_one);
 
 	/* disable all IRQs */
 	sa1111_writel(0, irqbase + SA1111_INTEN0);
-- 
cgit v1.2.3-70-g09d2


From ebe2a9ffa148746bae62d0f7188590a85f29f9ed Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 22 Jun 2005 09:55:04 +0100
Subject: [PATCH] ARM: Allow riscpc to parse "acorn" boot info tag

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-rpc/riscpc.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/mach-rpc/riscpc.c b/arch/arm/mach-rpc/riscpc.c
index 815c53225cd..43710688143 100644
--- a/arch/arm/mach-rpc/riscpc.c
+++ b/arch/arm/mach-rpc/riscpc.c
@@ -32,10 +32,7 @@
 
 extern void rpc_init_irq(void);
 
-extern unsigned int vram_size;
-
-#if 0
-
+unsigned int vram_size;
 unsigned int memc_ctrl_reg;
 unsigned int number_mfm_drives;
 
@@ -63,8 +60,6 @@ static int __init parse_tag_acorn(const struct tag *tag)
 
 __tagtable(ATAG_ACORN, parse_tag_acorn);
 
-#endif
-
 static struct map_desc rpc_io_desc[] __initdata = {
  { SCREEN_BASE,	SCREEN_START,	2*1048576, MT_DEVICE }, /* VRAM		*/
  { (u32)IO_BASE, IO_START,	IO_SIZE	 , MT_DEVICE }, /* IO space	*/
-- 
cgit v1.2.3-70-g09d2


From 052162198b89e64d37c20238412674152d614997 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 22 Jun 2005 09:56:57 +0100
Subject: [PATCH] ARM: Allow clps7500 to build without parsing "acorn" tag

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-clps7500/core.c | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'arch')

diff --git a/arch/arm/mach-clps7500/core.c b/arch/arm/mach-clps7500/core.c
index fdfededfd96..0bc7da48861 100644
--- a/arch/arm/mach-clps7500/core.c
+++ b/arch/arm/mach-clps7500/core.c
@@ -26,6 +26,8 @@
 #include <asm/irq.h>
 #include <asm/mach-types.h>
 
+unsigned int vram_size;
+
 static void cl7500_ack_irq_a(unsigned int irq)
 {
 	unsigned int val, mask;
-- 
cgit v1.2.3-70-g09d2


From e00d349e7781a92cf35b242259c9e5341a9661bb Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 22 Jun 2005 20:26:05 +0100
Subject: [PATCH] ARM: Move signal return code into vector page

Move the signal return code into the vector page instead of placing
it on the user mode stack, which will allow us to avoid flushing
the instruction cache on signals, as well as eventually allowing
non-exec stack.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/signal.c | 29 +++++++++++++++++++----------
 arch/arm/kernel/signal.h | 12 ++++++++++++
 arch/arm/kernel/traps.c  |  9 +++++++++
 3 files changed, 40 insertions(+), 10 deletions(-)
 create mode 100644 arch/arm/kernel/signal.h

(limited to 'arch')

diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c
index 931919fd512..07ddeed6176 100644
--- a/arch/arm/kernel/signal.c
+++ b/arch/arm/kernel/signal.c
@@ -19,6 +19,7 @@
 #include <asm/unistd.h>
 
 #include "ptrace.h"
+#include "signal.h"
 
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 
@@ -35,7 +36,7 @@
 #define SWI_THUMB_SIGRETURN	(0xdf00 << 16 | 0x2700 | (__NR_sigreturn - __NR_SYSCALL_BASE))
 #define SWI_THUMB_RT_SIGRETURN	(0xdf00 << 16 | 0x2700 | (__NR_rt_sigreturn - __NR_SYSCALL_BASE))
 
-static const unsigned long retcodes[4] = {
+const unsigned long sigreturn_codes[4] = {
 	SWI_SYS_SIGRETURN,	SWI_THUMB_SIGRETURN,
 	SWI_SYS_RT_SIGRETURN,	SWI_THUMB_RT_SIGRETURN
 };
@@ -500,17 +501,25 @@ setup_return(struct pt_regs *regs, struct k_sigaction *ka,
 		if (ka->sa.sa_flags & SA_SIGINFO)
 			idx += 2;
 
-		if (__put_user(retcodes[idx], rc))
+		if (__put_user(sigreturn_codes[idx], rc))
 			return 1;
 
-		/*
-		 * Ensure that the instruction cache sees
-		 * the return code written onto the stack.
-		 */
-		flush_icache_range((unsigned long)rc,
-				   (unsigned long)(rc + 1));
-
-		retcode = ((unsigned long)rc) + thumb;
+		if (cpsr & MODE32_BIT) {
+			/*
+			 * 32-bit code can use the new high-page
+			 * signal return code support.
+			 */
+			retcode = KERN_SIGRETURN_CODE + (idx << 2) + thumb;
+		} else {
+			/*
+			 * Ensure that the instruction cache sees
+			 * the return code written onto the stack.
+			 */
+			flush_icache_range((unsigned long)rc,
+					   (unsigned long)(rc + 1));
+
+			retcode = ((unsigned long)rc) + thumb;
+		}
 	}
 
 	regs->ARM_r0 = usig;
diff --git a/arch/arm/kernel/signal.h b/arch/arm/kernel/signal.h
new file mode 100644
index 00000000000..91d26faca62
--- /dev/null
+++ b/arch/arm/kernel/signal.h
@@ -0,0 +1,12 @@
+/*
+ *  linux/arch/arm/kernel/signal.h
+ *
+ *  Copyright (C) 2005 Russell King.
+ *
+ * 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.
+ */
+#define KERN_SIGRETURN_CODE	0xffff0500
+
+extern const unsigned long sigreturn_codes[4];
diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c
index 45d2a032d89..2fb0a4cfb37 100644
--- a/arch/arm/kernel/traps.c
+++ b/arch/arm/kernel/traps.c
@@ -30,6 +30,7 @@
 #include <asm/traps.h>
 
 #include "ptrace.h"
+#include "signal.h"
 
 const char *processor_modes[]=
 { "USER_26", "FIQ_26" , "IRQ_26" , "SVC_26" , "UK4_26" , "UK5_26" , "UK6_26" , "UK7_26" ,
@@ -683,6 +684,14 @@ void __init trap_init(void)
 	memcpy((void *)0xffff0000, __vectors_start, __vectors_end - __vectors_start);
 	memcpy((void *)0xffff0200, __stubs_start, __stubs_end - __stubs_start);
 	memcpy((void *)0xffff1000 - kuser_sz, __kuser_helper_start, kuser_sz);
+
+	/*
+	 * Copy signal return handlers into the vector page, and
+	 * set sigreturn to be a pointer to these.
+	 */
+	memcpy((void *)KERN_SIGRETURN_CODE, sigreturn_codes,
+	       sizeof(sigreturn_codes));
+
 	flush_icache_range(0xffff0000, 0xffff0000 + PAGE_SIZE);
 	modify_domain(DOMAIN_USER, DOMAIN_CLIENT);
 }
-- 
cgit v1.2.3-70-g09d2


From f1b04770b0d073a9d70e5b3b873d274c1c19e1f6 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Wed, 22 Jun 2005 15:10:02 -0500
Subject: [PATCH] ppc32: Fix building MPC8555 CDS

Adding support for MPC8548 w/o PCI support, broke building MPC8555 CDS
by trying to remove a loop variable that was used when PCI is enabled.

Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org)
---
 arch/ppc/platforms/85xx/mpc85xx_cds_common.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'arch')

diff --git a/arch/ppc/platforms/85xx/mpc85xx_cds_common.c b/arch/ppc/platforms/85xx/mpc85xx_cds_common.c
index 203b2ca61df..b52c4317fef 100644
--- a/arch/ppc/platforms/85xx/mpc85xx_cds_common.c
+++ b/arch/ppc/platforms/85xx/mpc85xx_cds_common.c
@@ -149,6 +149,7 @@ 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;
-- 
cgit v1.2.3-70-g09d2


From b46a58fd4e3cdd107ffccedaa5e233deeb94c5c0 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 22 Jun 2005 21:25:58 +0100
Subject: [PATCH] ARM: Use list_for_each_entry() for dmabounce

Convert dmabounce.c to use list_for_each_entry() instead of
list_for_each() + list_entry().

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/common/dmabounce.c | 18 +++++-------------
 1 file changed, 5 insertions(+), 13 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/common/dmabounce.c b/arch/arm/common/dmabounce.c
index 9d63a01214e..cbf2165476b 100644
--- a/arch/arm/common/dmabounce.c
+++ b/arch/arm/common/dmabounce.c
@@ -93,15 +93,12 @@ static void print_alloc_stats(struct dmabounce_device_info *device_info)
 static inline struct dmabounce_device_info *
 find_dmabounce_dev(struct device *dev)
 {
-	struct list_head *entry;
-
-	list_for_each(entry, &dmabounce_devs) {
-		struct dmabounce_device_info *d =
-			list_entry(entry, struct dmabounce_device_info, node);
+	struct dmabounce_device_info *d;
 
+	list_for_each_entry(d, &dmabounce_devs, node)
 		if (d->dev == dev)
 			return d;
-	}
+
 	return NULL;
 }
 
@@ -172,15 +169,11 @@ alloc_safe_buffer(struct dmabounce_device_info *device_info, void *ptr,
 static inline struct safe_buffer *
 find_safe_buffer(struct dmabounce_device_info *device_info, dma_addr_t safe_dma_addr)
 {
-	struct list_head *entry;
-
-	list_for_each(entry, &device_info->safe_buffers) {
-		struct safe_buffer *b =
-			list_entry(entry, struct safe_buffer, node);
+	struct safe_buffer *b;
 
+	list_for_each_entry(b, &device_info->safe_buffers, node)
 		if (b->safe_dma_addr == safe_dma_addr)
 			return b;
-	}
 
 	return NULL;
 }
@@ -301,7 +294,6 @@ unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
 			__func__, buf->ptr, (void *) virt_to_dma(dev, buf->ptr),
 			buf->safe, (void *) buf->safe_dma_addr);
 
-
 		DO_STATS ( device_info->bounce_count++ );
 
 		if (dir == DMA_FROM_DEVICE || dir == DMA_BIDIRECTIONAL) {
-- 
cgit v1.2.3-70-g09d2


From 3a66941106855215127f8bf1afd06099b72dc75b Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 22 Jun 2005 21:43:10 +0100
Subject: [PATCH] ARM: Ensure memory information is page aligned

Ensure that meminfo.bank[] array contains page-aligned start/size
information.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/setup.c | 25 ++++++++++++++++---------
 1 file changed, 16 insertions(+), 9 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index 9fed5fa194d..8cf733daa80 100644
--- a/arch/arm/kernel/setup.c
+++ b/arch/arm/kernel/setup.c
@@ -395,6 +395,20 @@ static void __init early_initrd(char **p)
 }
 __early_param("initrd=", early_initrd);
 
+static void __init add_memory(unsigned long start, unsigned long size)
+{
+	/*
+	 * Ensure that start/size are aligned to a page boundary.
+	 * Size is appropriately rounded down, start is rounded up.
+	 */
+	size -= start & ~PAGE_MASK;
+
+	meminfo.bank[meminfo.nr_banks].start = PAGE_ALIGN(start);
+	meminfo.bank[meminfo.nr_banks].size  = size & PAGE_MASK;
+	meminfo.bank[meminfo.nr_banks].node  = PHYS_TO_NID(start);
+	meminfo.nr_banks += 1;
+}
+
 /*
  * Pick out the memory size.  We look for mem=size@start,
  * where start and size are "size[KkMm]"
@@ -419,10 +433,7 @@ static void __init early_mem(char **p)
 	if (**p == '@')
 		start = memparse(*p + 1, p);
 
-	meminfo.bank[meminfo.nr_banks].start = start;
-	meminfo.bank[meminfo.nr_banks].size  = size;
-	meminfo.bank[meminfo.nr_banks].node  = PHYS_TO_NID(start);
-	meminfo.nr_banks += 1;
+	add_memory(start, size);
 }
 __early_param("mem=", early_mem);
 
@@ -564,11 +575,7 @@ static int __init parse_tag_mem32(const struct tag *tag)
 			tag->u.mem.start, tag->u.mem.size / 1024);
 		return -EINVAL;
 	}
-	meminfo.bank[meminfo.nr_banks].start = tag->u.mem.start;
-	meminfo.bank[meminfo.nr_banks].size  = tag->u.mem.size;
-	meminfo.bank[meminfo.nr_banks].node  = PHYS_TO_NID(tag->u.mem.start);
-	meminfo.nr_banks += 1;
-
+	add_memory(tag->u.mem.start, tag->u.mem.size);
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 92a8cbed29eb9bf6e8eec16ca29d54015bc0e8a2 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 22 Jun 2005 21:47:25 +0100
Subject: [PATCH] ARM: Remove explicit page-alignments in memory init

Since meminfo.bank[] array contains page-aligned start/size, we
no longer need to explicitly round up/down the addresses when
converting to PFNs.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/init.c | 19 ++++++-------------
 1 file changed, 6 insertions(+), 13 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index 72a2b8cee31..c08710b1ff0 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -93,14 +93,7 @@ struct node_info {
 };
 
 #define O_PFN_DOWN(x)	((x) >> PAGE_SHIFT)
-#define V_PFN_DOWN(x)	O_PFN_DOWN(__pa(x))
-
 #define O_PFN_UP(x)	(PAGE_ALIGN(x) >> PAGE_SHIFT)
-#define V_PFN_UP(x)	O_PFN_UP(__pa(x))
-
-#define PFN_SIZE(x)	((x) >> PAGE_SHIFT)
-#define PFN_RANGE(s,e)	PFN_SIZE(PAGE_ALIGN((unsigned long)(e)) - \
-				(((unsigned long)(s)) & PAGE_MASK))
 
 /*
  * FIXME: We really want to avoid allocating the bootmap bitmap
@@ -113,7 +106,7 @@ find_bootmap_pfn(int node, struct meminfo *mi, unsigned int bootmap_pages)
 {
 	unsigned int start_pfn, bank, bootmap_pfn;
 
-	start_pfn   = V_PFN_UP(&_end);
+	start_pfn   = O_PFN_UP(__pa(&_end));
 	bootmap_pfn = 0;
 
 	for (bank = 0; bank < mi->nr_banks; bank ++) {
@@ -122,9 +115,9 @@ find_bootmap_pfn(int node, struct meminfo *mi, unsigned int bootmap_pages)
 		if (mi->bank[bank].node != node)
 			continue;
 
-		start = O_PFN_UP(mi->bank[bank].start);
-		end   = O_PFN_DOWN(mi->bank[bank].size +
-				   mi->bank[bank].start);
+		start = mi->bank[bank].start >> PAGE_SHIFT;
+		end   = (mi->bank[bank].size +
+			 mi->bank[bank].start) >> PAGE_SHIFT;
 
 		if (end < start_pfn)
 			continue;
@@ -191,8 +184,8 @@ find_memend_and_nodes(struct meminfo *mi, struct node_info *np)
 		/*
 		 * Get the start and end pfns for this bank
 		 */
-		start = O_PFN_UP(mi->bank[i].start);
-		end   = O_PFN_DOWN(mi->bank[i].start + mi->bank[i].size);
+		start = mi->bank[i].start >> PAGE_SHIFT;
+		end   = (mi->bank[i].start + mi->bank[i].size) >> PAGE_SHIFT;
 
 		if (np[node].start > start)
 			np[node].start = start;
-- 
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 'arch')

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 'arch')

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 c5a3c2e52af1bcb118022ffac9a0fd1d42d43bd3 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 23 Jun 2005 09:43:23 +1000
Subject: [PATCH] ppc64: Split out generic rtas code from pSeries_pci.c.

BPA is using rtas for PCI but should not be confused by
pSeries code. This also avoids some #ifdefs. Other
platforms that want to use rtas_pci.c could create
their own platform_pci.c with platform specific fixups.

Signed-off-by: Arnd Bergmann <arndb@de.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/Makefile      |   3 +-
 arch/ppc64/kernel/mpic.h        |   3 +
 arch/ppc64/kernel/pSeries_pci.c | 476 +-------------------------------------
 arch/ppc64/kernel/rtas_pci.c    | 495 ++++++++++++++++++++++++++++++++++++++++
 4 files changed, 506 insertions(+), 471 deletions(-)
 create mode 100644 arch/ppc64/kernel/rtas_pci.c

(limited to 'arch')

diff --git a/arch/ppc64/kernel/Makefile b/arch/ppc64/kernel/Makefile
index b5e167cf1a0..f389f2453da 100644
--- a/arch/ppc64/kernel/Makefile
+++ b/arch/ppc64/kernel/Makefile
@@ -31,13 +31,14 @@ obj-$(CONFIG_PPC_MULTIPLATFORM) += nvram.o i8259.o prom_init.o prom.o mpic.o
 
 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 rtas.o pSeries_setup.o pSeries_iommu.o
+			     xics.o pSeries_setup.o pSeries_iommu.o
 
 obj-$(CONFIG_EEH)		+= eeh.o
 obj-$(CONFIG_PROC_FS)		+= proc_ppc64.o
 obj-$(CONFIG_RTAS_FLASH)	+= rtas_flash.o
 obj-$(CONFIG_SMP)		+= smp.o
 obj-$(CONFIG_MODULES)		+= module.o ppc_ksyms.o
+obj-$(CONFIG_PPC_RTAS)		+= rtas.o rtas_pci.o
 obj-$(CONFIG_RTAS_PROC)		+= rtas-proc.o
 obj-$(CONFIG_SCANLOG)		+= scanlog.o
 obj-$(CONFIG_VIOPATH)		+= viopath.o
diff --git a/arch/ppc64/kernel/mpic.h b/arch/ppc64/kernel/mpic.h
index 571b3c99e06..63e177143ea 100644
--- a/arch/ppc64/kernel/mpic.h
+++ b/arch/ppc64/kernel/mpic.h
@@ -265,3 +265,6 @@ extern void mpic_send_ipi(unsigned int ipi_no, unsigned int cpu_mask);
 extern int mpic_get_one_irq(struct mpic *mpic, struct pt_regs *regs);
 /* This one gets to the primary mpic */
 extern int mpic_get_irq(struct pt_regs *regs);
+
+/* global mpic for pSeries */
+extern struct mpic *pSeries_mpic;
diff --git a/arch/ppc64/kernel/pSeries_pci.c b/arch/ppc64/kernel/pSeries_pci.c
index 0b1cca28140..dfa6d3d3e9f 100644
--- a/arch/ppc64/kernel/pSeries_pci.c
+++ b/arch/ppc64/kernel/pSeries_pci.c
@@ -1,13 +1,11 @@
 /*
- * pSeries_pci.c
+ * arch/ppc64/kernel/pSeries_pci.c
  *
  * Copyright (C) 2001 Dave Engebretsen, IBM Corporation
  * Copyright (C) 2003 Anton Blanchard <anton@au.ibm.com>, IBM
  *
  * pSeries specific routines for PCI.
  * 
- * Based on code from pci.c and chrp_pci.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
@@ -23,430 +21,18 @@
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
  */
 
+#include <linux/init.h>
+#include <linux/ioport.h>
 #include <linux/kernel.h>
-#include <linux/threads.h>
 #include <linux/pci.h>
 #include <linux/string.h>
-#include <linux/init.h>
-#include <linux/bootmem.h>
 
-#include <asm/io.h>
-#include <asm/pgtable.h>
-#include <asm/irq.h>
-#include <asm/prom.h>
-#include <asm/machdep.h>
 #include <asm/pci-bridge.h>
-#include <asm/iommu.h>
-#include <asm/rtas.h>
+#include <asm/prom.h>
 
-#include "mpic.h"
 #include "pci.h"
 
-/* RTAS tokens */
-static int read_pci_config;
-static int write_pci_config;
-static int ibm_read_pci_config;
-static int ibm_write_pci_config;
-
-static int s7a_workaround;
-
-extern struct mpic *pSeries_mpic;
-
-static int config_access_valid(struct device_node *dn, int where)
-{
-	if (where < 256)
-		return 1;
-	if (where < 4096 && dn->pci_ext_config_space)
-		return 1;
-
-	return 0;
-}
-
-static int rtas_read_config(struct device_node *dn, int where, int size, u32 *val)
-{
-	int returnval = -1;
-	unsigned long buid, addr;
-	int ret;
-
-	if (!dn)
-		return PCIBIOS_DEVICE_NOT_FOUND;
-	if (!config_access_valid(dn, where))
-		return PCIBIOS_BAD_REGISTER_NUMBER;
-
-	addr = ((where & 0xf00) << 20) | (dn->busno << 16) |
-		(dn->devfn << 8) | (where & 0xff);
-	buid = dn->phb->buid;
-	if (buid) {
-		ret = rtas_call(ibm_read_pci_config, 4, 2, &returnval,
-				addr, buid >> 32, buid & 0xffffffff, size);
-	} else {
-		ret = rtas_call(read_pci_config, 2, 2, &returnval, addr, size);
-	}
-	*val = returnval;
-
-	if (ret)
-		return PCIBIOS_DEVICE_NOT_FOUND;
-
-	if (returnval == EEH_IO_ERROR_VALUE(size)
-	    && eeh_dn_check_failure (dn, NULL))
-		return PCIBIOS_DEVICE_NOT_FOUND;
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int rtas_pci_read_config(struct pci_bus *bus,
-				unsigned int devfn,
-				int where, int size, u32 *val)
-{
-	struct device_node *busdn, *dn;
-
-	if (bus->self)
-		busdn = pci_device_to_OF_node(bus->self);
-	else
-		busdn = bus->sysdata;	/* must be a phb */
-
-	/* Search only direct children of the bus */
-	for (dn = busdn->child; dn; dn = dn->sibling)
-		if (dn->devfn == devfn)
-			return rtas_read_config(dn, where, size, val);
-	return PCIBIOS_DEVICE_NOT_FOUND;
-}
-
-static int rtas_write_config(struct device_node *dn, int where, int size, u32 val)
-{
-	unsigned long buid, addr;
-	int ret;
-
-	if (!dn)
-		return PCIBIOS_DEVICE_NOT_FOUND;
-	if (!config_access_valid(dn, where))
-		return PCIBIOS_BAD_REGISTER_NUMBER;
-
-	addr = ((where & 0xf00) << 20) | (dn->busno << 16) |
-		(dn->devfn << 8) | (where & 0xff);
-	buid = dn->phb->buid;
-	if (buid) {
-		ret = rtas_call(ibm_write_pci_config, 5, 1, NULL, addr, buid >> 32, buid & 0xffffffff, size, (ulong) val);
-	} else {
-		ret = rtas_call(write_pci_config, 3, 1, NULL, addr, size, (ulong)val);
-	}
-
-	if (ret)
-		return PCIBIOS_DEVICE_NOT_FOUND;
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int rtas_pci_write_config(struct pci_bus *bus,
-				 unsigned int devfn,
-				 int where, int size, u32 val)
-{
-	struct device_node *busdn, *dn;
-
-	if (bus->self)
-		busdn = pci_device_to_OF_node(bus->self);
-	else
-		busdn = bus->sysdata;	/* must be a phb */
-
-	/* Search only direct children of the bus */
-	for (dn = busdn->child; dn; dn = dn->sibling)
-		if (dn->devfn == devfn)
-			return rtas_write_config(dn, where, size, val);
-	return PCIBIOS_DEVICE_NOT_FOUND;
-}
-
-struct pci_ops rtas_pci_ops = {
-	rtas_pci_read_config,
-	rtas_pci_write_config
-};
-
-int is_python(struct device_node *dev)
-{
-	char *model = (char *)get_property(dev, "model", NULL);
-
-	if (model && strstr(model, "Python"))
-		return 1;
-
-	return 0;
-}
-
-static int get_phb_reg_prop(struct device_node *dev,
-			    unsigned int addr_size_words,
-			    struct reg_property64 *reg)
-{
-	unsigned int *ui_ptr = NULL, len;
-
-	/* Found a PHB, now figure out where his registers are mapped. */
-	ui_ptr = (unsigned int *)get_property(dev, "reg", &len);
-	if (ui_ptr == NULL)
-		return 1;
-
-	if (addr_size_words == 1) {
-		reg->address = ((struct reg_property32 *)ui_ptr)->address;
-		reg->size    = ((struct reg_property32 *)ui_ptr)->size;
-	} else {
-		*reg = *((struct reg_property64 *)ui_ptr);
-	}
-
-	return 0;
-}
-
-static void python_countermeasures(struct device_node *dev,
-				   unsigned int addr_size_words)
-{
-	struct reg_property64 reg_struct;
-	void __iomem *chip_regs;
-	volatile u32 val;
-
-	if (get_phb_reg_prop(dev, addr_size_words, &reg_struct))
-		return;
-
-	/* Python's register file is 1 MB in size. */
-	chip_regs = ioremap(reg_struct.address & ~(0xfffffUL), 0x100000);
-
-	/* 
-	 * Firmware doesn't always clear this bit which is critical
-	 * for good performance - Anton
-	 */
-
-#define PRG_CL_RESET_VALID 0x00010000
-
-	val = in_be32(chip_regs + 0xf6030);
-	if (val & PRG_CL_RESET_VALID) {
-		printk(KERN_INFO "Python workaround: ");
-		val &= ~PRG_CL_RESET_VALID;
-		out_be32(chip_regs + 0xf6030, val);
-		/*
-		 * We must read it back for changes to
-		 * take effect
-		 */
-		val = in_be32(chip_regs + 0xf6030);
-		printk("reg0: %x\n", val);
-	}
-
-	iounmap(chip_regs);
-}
-
-void __init init_pci_config_tokens (void)
-{
-	read_pci_config = rtas_token("read-pci-config");
-	write_pci_config = rtas_token("write-pci-config");
-	ibm_read_pci_config = rtas_token("ibm,read-pci-config");
-	ibm_write_pci_config = rtas_token("ibm,write-pci-config");
-}
-
-unsigned long __devinit get_phb_buid (struct device_node *phb)
-{
-	int addr_cells;
-	unsigned int *buid_vals;
-	unsigned int len;
-	unsigned long buid;
-
-	if (ibm_read_pci_config == -1) return 0;
-
-	/* PHB's will always be children of the root node,
-	 * or so it is promised by the current firmware. */
-	if (phb->parent == NULL)
-		return 0;
-	if (phb->parent->parent)
-		return 0;
-
-	buid_vals = (unsigned int *) get_property(phb, "reg", &len);
-	if (buid_vals == NULL)
-		return 0;
-
-	addr_cells = prom_n_addr_cells(phb);
-	if (addr_cells == 1) {
-		buid = (unsigned long) buid_vals[0];
-	} else {
-		buid = (((unsigned long)buid_vals[0]) << 32UL) |
-			(((unsigned long)buid_vals[1]) & 0xffffffff);
-	}
-	return buid;
-}
-
-static int phb_set_bus_ranges(struct device_node *dev,
-			      struct pci_controller *phb)
-{
-	int *bus_range;
-	unsigned int len;
-
-	bus_range = (int *) get_property(dev, "bus-range", &len);
-	if (bus_range == NULL || len < 2 * sizeof(int)) {
-		return 1;
- 	}
- 
-	phb->first_busno =  bus_range[0];
-	phb->last_busno  =  bus_range[1];
-
-	return 0;
-}
-
-static int __devinit setup_phb(struct device_node *dev,
-			       struct pci_controller *phb,
-			       unsigned int addr_size_words)
-{
-	pci_setup_pci_controller(phb);
-
-	if (is_python(dev))
-		python_countermeasures(dev, addr_size_words);
-
-	if (phb_set_bus_ranges(dev, phb))
-		return 1;
-
-	phb->arch_data = dev;
-	phb->ops = &rtas_pci_ops;
-	phb->buid = get_phb_buid(dev);
-
-	return 0;
-}
-
-static void __devinit add_linux_pci_domain(struct device_node *dev,
-					   struct pci_controller *phb,
-					   struct property *of_prop)
-{
-	memset(of_prop, 0, sizeof(struct property));
-	of_prop->name = "linux,pci-domain";
-	of_prop->length = sizeof(phb->global_number);
-	of_prop->value = (unsigned char *)&of_prop[1];
-	memcpy(of_prop->value, &phb->global_number, sizeof(phb->global_number));
-	prom_add_property(dev, of_prop);
-}
-
-static struct pci_controller * __init alloc_phb(struct device_node *dev,
-						unsigned int addr_size_words)
-{
-	struct pci_controller *phb;
-	struct property *of_prop;
-
-	phb = alloc_bootmem(sizeof(struct pci_controller));
-	if (phb == NULL)
-		return NULL;
-
-	of_prop = alloc_bootmem(sizeof(struct property) +
-				sizeof(phb->global_number));
-	if (!of_prop)
-		return NULL;
-
-	if (setup_phb(dev, phb, addr_size_words))
-		return NULL;
-
-	add_linux_pci_domain(dev, phb, of_prop);
-
-	return phb;
-}
-
-static struct pci_controller * __devinit alloc_phb_dynamic(struct device_node *dev, unsigned int addr_size_words)
-{
-	struct pci_controller *phb;
-
-	phb = (struct pci_controller *)kmalloc(sizeof(struct pci_controller),
-					       GFP_KERNEL);
-	if (phb == NULL)
-		return NULL;
-
-	if (setup_phb(dev, phb, addr_size_words))
-		return NULL;
-
-	phb->is_dynamic = 1;
-
-	/* TODO: linux,pci-domain? */
-
- 	return phb;
-}
-
-unsigned long __init find_and_init_phbs(void)
-{
-	struct device_node *node;
-	struct pci_controller *phb;
-	unsigned int root_size_cells = 0;
-	unsigned int index;
-	unsigned int *opprop = NULL;
-	struct device_node *root = of_find_node_by_path("/");
-
-	if (ppc64_interrupt_controller == IC_OPEN_PIC) {
-		opprop = (unsigned int *)get_property(root,
-				"platform-open-pic", NULL);
-	}
-
-	root_size_cells = prom_n_size_cells(root);
-
-	index = 0;
-
-	for (node = of_get_next_child(root, NULL);
-	     node != NULL;
-	     node = of_get_next_child(root, node)) {
-		if (node->type == NULL || strcmp(node->type, "pci") != 0)
-			continue;
-
-		phb = alloc_phb(node, root_size_cells);
-		if (!phb)
-			continue;
-
-		pci_process_bridge_OF_ranges(phb, node);
-		pci_setup_phb_io(phb, index == 0);
-
-		if (ppc64_interrupt_controller == IC_OPEN_PIC && pSeries_mpic) {
-			int addr = root_size_cells * (index + 2) - 1;
-			mpic_assign_isu(pSeries_mpic, index, opprop[addr]);
-		}
-
-		index++;
-	}
-
-	of_node_put(root);
-	pci_devs_phb_init();
-
-	/*
-	 * pci_probe_only and pci_assign_all_buses can be set via properties
-	 * in chosen.
-	 */
-	if (of_chosen) {
-		int *prop;
-
-		prop = (int *)get_property(of_chosen, "linux,pci-probe-only",
-					   NULL);
-		if (prop)
-			pci_probe_only = *prop;
-
-		prop = (int *)get_property(of_chosen,
-					   "linux,pci-assign-all-buses", NULL);
-		if (prop)
-			pci_assign_all_buses = *prop;
-	}
-
-	return 0;
-}
-
-struct pci_controller * __devinit init_phb_dynamic(struct device_node *dn)
-{
-	struct device_node *root = of_find_node_by_path("/");
-	unsigned int root_size_cells = 0;
-	struct pci_controller *phb;
-	struct pci_bus *bus;
-	int primary;
-
-	root_size_cells = prom_n_size_cells(root);
-
-	primary = list_empty(&hose_list);
-	phb = alloc_phb_dynamic(dn, root_size_cells);
-	if (!phb)
-		return NULL;
-
-	pci_process_bridge_OF_ranges(phb, dn);
-
-	pci_setup_phb_io_dynamic(phb, primary);
-	of_node_put(root);
-
-	pci_devs_phb_init_dynamic(phb);
-	phb->last_busno = 0xff;
-	bus = pci_scan_bus(phb->first_busno, phb->ops, phb->arch_data);
-	phb->bus = bus;
-	phb->last_busno = bus->subordinate;
-
-	return phb;
-}
-EXPORT_SYMBOL(init_phb_dynamic);
+static int __initdata s7a_workaround;
 
 #if 0
 void pcibios_name_device(struct pci_dev *dev)
@@ -474,7 +60,7 @@ void pcibios_name_device(struct pci_dev *dev)
 DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pcibios_name_device);
 #endif
 
-static void check_s7a(void)
+static void __init check_s7a(void)
 {
 	struct device_node *root;
 	char *model;
@@ -488,56 +74,6 @@ static void check_s7a(void)
 	}
 }
 
-/* RPA-specific bits for removing PHBs */
-int pcibios_remove_root_bus(struct pci_controller *phb)
-{
-	struct pci_bus *b = phb->bus;
-	struct resource *res;
-	int rc, i;
-
-	res = b->resource[0];
-	if (!res->flags) {
-		printk(KERN_ERR "%s: no IO resource for PHB %s\n", __FUNCTION__,
-				b->name);
-		return 1;
-	}
-
-	rc = unmap_bus_range(b);
-	if (rc) {
-		printk(KERN_ERR "%s: failed to unmap IO on bus %s\n",
-			__FUNCTION__, b->name);
-		return 1;
-	}
-
-	if (release_resource(res)) {
-		printk(KERN_ERR "%s: failed to release IO on bus %s\n",
-				__FUNCTION__, b->name);
-		return 1;
-	}
-
-	for (i = 1; i < 3; ++i) {
-		res = b->resource[i];
-		if (!res->flags && i == 0) {
-			printk(KERN_ERR "%s: no MEM resource for PHB %s\n",
-				__FUNCTION__, b->name);
-			return 1;
-		}
-		if (res->flags && release_resource(res)) {
-			printk(KERN_ERR
-			       "%s: failed to release IO %d on bus %s\n",
-				__FUNCTION__, i, b->name);
-			return 1;
-		}
-	}
-
-	list_del(&phb->list_node);
-	if (phb->is_dynamic)
-		kfree(phb);
-
-	return 0;
-}
-EXPORT_SYMBOL(pcibios_remove_root_bus);
-
 static void __init pSeries_request_regions(void)
 {
 	if (!isa_io_base)
diff --git a/arch/ppc64/kernel/rtas_pci.c b/arch/ppc64/kernel/rtas_pci.c
new file mode 100644
index 00000000000..1048817befb
--- /dev/null
+++ b/arch/ppc64/kernel/rtas_pci.c
@@ -0,0 +1,495 @@
+/*
+ * arch/ppc64/kernel/rtas_pci.c
+ *
+ * Copyright (C) 2001 Dave Engebretsen, IBM Corporation
+ * Copyright (C) 2003 Anton Blanchard <anton@au.ibm.com>, IBM
+ *
+ * RTAS specific routines for PCI.
+ * 
+ * Based on code from pci.c, chrp_pci.c and pSeries_pci.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
+ */
+
+#include <linux/kernel.h>
+#include <linux/threads.h>
+#include <linux/pci.h>
+#include <linux/string.h>
+#include <linux/init.h>
+#include <linux/bootmem.h>
+
+#include <asm/io.h>
+#include <asm/pgtable.h>
+#include <asm/irq.h>
+#include <asm/prom.h>
+#include <asm/machdep.h>
+#include <asm/pci-bridge.h>
+#include <asm/iommu.h>
+#include <asm/rtas.h>
+
+#include "mpic.h"
+#include "pci.h"
+
+/* RTAS tokens */
+static int read_pci_config;
+static int write_pci_config;
+static int ibm_read_pci_config;
+static int ibm_write_pci_config;
+
+static int config_access_valid(struct device_node *dn, int where)
+{
+	if (where < 256)
+		return 1;
+	if (where < 4096 && dn->pci_ext_config_space)
+		return 1;
+
+	return 0;
+}
+
+static int rtas_read_config(struct device_node *dn, int where, int size, u32 *val)
+{
+	int returnval = -1;
+	unsigned long buid, addr;
+	int ret;
+
+	if (!dn)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+	if (!config_access_valid(dn, where))
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+
+	addr = ((where & 0xf00) << 20) | (dn->busno << 16) |
+		(dn->devfn << 8) | (where & 0xff);
+	buid = dn->phb->buid;
+	if (buid) {
+		ret = rtas_call(ibm_read_pci_config, 4, 2, &returnval,
+				addr, buid >> 32, buid & 0xffffffff, size);
+	} else {
+		ret = rtas_call(read_pci_config, 2, 2, &returnval, addr, size);
+	}
+	*val = returnval;
+
+	if (ret)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	if (returnval == EEH_IO_ERROR_VALUE(size)
+	    && eeh_dn_check_failure (dn, NULL))
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int rtas_pci_read_config(struct pci_bus *bus,
+				unsigned int devfn,
+				int where, int size, u32 *val)
+{
+	struct device_node *busdn, *dn;
+
+	if (bus->self)
+		busdn = pci_device_to_OF_node(bus->self);
+	else
+		busdn = bus->sysdata;	/* must be a phb */
+
+	/* Search only direct children of the bus */
+	for (dn = busdn->child; dn; dn = dn->sibling)
+		if (dn->devfn == devfn)
+			return rtas_read_config(dn, where, size, val);
+	return PCIBIOS_DEVICE_NOT_FOUND;
+}
+
+static int rtas_write_config(struct device_node *dn, int where, int size, u32 val)
+{
+	unsigned long buid, addr;
+	int ret;
+
+	if (!dn)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+	if (!config_access_valid(dn, where))
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+
+	addr = ((where & 0xf00) << 20) | (dn->busno << 16) |
+		(dn->devfn << 8) | (where & 0xff);
+	buid = dn->phb->buid;
+	if (buid) {
+		ret = rtas_call(ibm_write_pci_config, 5, 1, NULL, addr, buid >> 32, buid & 0xffffffff, size, (ulong) val);
+	} else {
+		ret = rtas_call(write_pci_config, 3, 1, NULL, addr, size, (ulong)val);
+	}
+
+	if (ret)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int rtas_pci_write_config(struct pci_bus *bus,
+				 unsigned int devfn,
+				 int where, int size, u32 val)
+{
+	struct device_node *busdn, *dn;
+
+	if (bus->self)
+		busdn = pci_device_to_OF_node(bus->self);
+	else
+		busdn = bus->sysdata;	/* must be a phb */
+
+	/* Search only direct children of the bus */
+	for (dn = busdn->child; dn; dn = dn->sibling)
+		if (dn->devfn == devfn)
+			return rtas_write_config(dn, where, size, val);
+	return PCIBIOS_DEVICE_NOT_FOUND;
+}
+
+struct pci_ops rtas_pci_ops = {
+	rtas_pci_read_config,
+	rtas_pci_write_config
+};
+
+int is_python(struct device_node *dev)
+{
+	char *model = (char *)get_property(dev, "model", NULL);
+
+	if (model && strstr(model, "Python"))
+		return 1;
+
+	return 0;
+}
+
+static int get_phb_reg_prop(struct device_node *dev,
+			    unsigned int addr_size_words,
+			    struct reg_property64 *reg)
+{
+	unsigned int *ui_ptr = NULL, len;
+
+	/* Found a PHB, now figure out where his registers are mapped. */
+	ui_ptr = (unsigned int *)get_property(dev, "reg", &len);
+	if (ui_ptr == NULL)
+		return 1;
+
+	if (addr_size_words == 1) {
+		reg->address = ((struct reg_property32 *)ui_ptr)->address;
+		reg->size    = ((struct reg_property32 *)ui_ptr)->size;
+	} else {
+		*reg = *((struct reg_property64 *)ui_ptr);
+	}
+
+	return 0;
+}
+
+static void python_countermeasures(struct device_node *dev,
+				   unsigned int addr_size_words)
+{
+	struct reg_property64 reg_struct;
+	void __iomem *chip_regs;
+	volatile u32 val;
+
+	if (get_phb_reg_prop(dev, addr_size_words, &reg_struct))
+		return;
+
+	/* Python's register file is 1 MB in size. */
+	chip_regs = ioremap(reg_struct.address & ~(0xfffffUL), 0x100000);
+
+	/* 
+	 * Firmware doesn't always clear this bit which is critical
+	 * for good performance - Anton
+	 */
+
+#define PRG_CL_RESET_VALID 0x00010000
+
+	val = in_be32(chip_regs + 0xf6030);
+	if (val & PRG_CL_RESET_VALID) {
+		printk(KERN_INFO "Python workaround: ");
+		val &= ~PRG_CL_RESET_VALID;
+		out_be32(chip_regs + 0xf6030, val);
+		/*
+		 * We must read it back for changes to
+		 * take effect
+		 */
+		val = in_be32(chip_regs + 0xf6030);
+		printk("reg0: %x\n", val);
+	}
+
+	iounmap(chip_regs);
+}
+
+void __init init_pci_config_tokens (void)
+{
+	read_pci_config = rtas_token("read-pci-config");
+	write_pci_config = rtas_token("write-pci-config");
+	ibm_read_pci_config = rtas_token("ibm,read-pci-config");
+	ibm_write_pci_config = rtas_token("ibm,write-pci-config");
+}
+
+unsigned long __devinit get_phb_buid (struct device_node *phb)
+{
+	int addr_cells;
+	unsigned int *buid_vals;
+	unsigned int len;
+	unsigned long buid;
+
+	if (ibm_read_pci_config == -1) return 0;
+
+	/* PHB's will always be children of the root node,
+	 * or so it is promised by the current firmware. */
+	if (phb->parent == NULL)
+		return 0;
+	if (phb->parent->parent)
+		return 0;
+
+	buid_vals = (unsigned int *) get_property(phb, "reg", &len);
+	if (buid_vals == NULL)
+		return 0;
+
+	addr_cells = prom_n_addr_cells(phb);
+	if (addr_cells == 1) {
+		buid = (unsigned long) buid_vals[0];
+	} else {
+		buid = (((unsigned long)buid_vals[0]) << 32UL) |
+			(((unsigned long)buid_vals[1]) & 0xffffffff);
+	}
+	return buid;
+}
+
+static int phb_set_bus_ranges(struct device_node *dev,
+			      struct pci_controller *phb)
+{
+	int *bus_range;
+	unsigned int len;
+
+	bus_range = (int *) get_property(dev, "bus-range", &len);
+	if (bus_range == NULL || len < 2 * sizeof(int)) {
+		return 1;
+ 	}
+ 
+	phb->first_busno =  bus_range[0];
+	phb->last_busno  =  bus_range[1];
+
+	return 0;
+}
+
+static int __devinit setup_phb(struct device_node *dev,
+			       struct pci_controller *phb,
+			       unsigned int addr_size_words)
+{
+	pci_setup_pci_controller(phb);
+
+	if (is_python(dev))
+		python_countermeasures(dev, addr_size_words);
+
+	if (phb_set_bus_ranges(dev, phb))
+		return 1;
+
+	phb->arch_data = dev;
+	phb->ops = &rtas_pci_ops;
+	phb->buid = get_phb_buid(dev);
+
+	return 0;
+}
+
+static void __devinit add_linux_pci_domain(struct device_node *dev,
+					   struct pci_controller *phb,
+					   struct property *of_prop)
+{
+	memset(of_prop, 0, sizeof(struct property));
+	of_prop->name = "linux,pci-domain";
+	of_prop->length = sizeof(phb->global_number);
+	of_prop->value = (unsigned char *)&of_prop[1];
+	memcpy(of_prop->value, &phb->global_number, sizeof(phb->global_number));
+	prom_add_property(dev, of_prop);
+}
+
+static struct pci_controller * __init alloc_phb(struct device_node *dev,
+						unsigned int addr_size_words)
+{
+	struct pci_controller *phb;
+	struct property *of_prop;
+
+	phb = alloc_bootmem(sizeof(struct pci_controller));
+	if (phb == NULL)
+		return NULL;
+
+	of_prop = alloc_bootmem(sizeof(struct property) +
+				sizeof(phb->global_number));
+	if (!of_prop)
+		return NULL;
+
+	if (setup_phb(dev, phb, addr_size_words))
+		return NULL;
+
+	add_linux_pci_domain(dev, phb, of_prop);
+
+	return phb;
+}
+
+static struct pci_controller * __devinit alloc_phb_dynamic(struct device_node *dev, unsigned int addr_size_words)
+{
+	struct pci_controller *phb;
+
+	phb = (struct pci_controller *)kmalloc(sizeof(struct pci_controller),
+					       GFP_KERNEL);
+	if (phb == NULL)
+		return NULL;
+
+	if (setup_phb(dev, phb, addr_size_words))
+		return NULL;
+
+	phb->is_dynamic = 1;
+
+	/* TODO: linux,pci-domain? */
+
+ 	return phb;
+}
+
+unsigned long __init find_and_init_phbs(void)
+{
+	struct device_node *node;
+	struct pci_controller *phb;
+	unsigned int root_size_cells = 0;
+	unsigned int index;
+	unsigned int *opprop = NULL;
+	struct device_node *root = of_find_node_by_path("/");
+
+	if (ppc64_interrupt_controller == IC_OPEN_PIC) {
+		opprop = (unsigned int *)get_property(root,
+				"platform-open-pic", NULL);
+	}
+
+	root_size_cells = prom_n_size_cells(root);
+
+	index = 0;
+
+	for (node = of_get_next_child(root, NULL);
+	     node != NULL;
+	     node = of_get_next_child(root, node)) {
+		if (node->type == NULL || strcmp(node->type, "pci") != 0)
+			continue;
+
+		phb = alloc_phb(node, root_size_cells);
+		if (!phb)
+			continue;
+
+		pci_process_bridge_OF_ranges(phb, node);
+		pci_setup_phb_io(phb, index == 0);
+#ifdef CONFIG_PPC_PSERIES
+		if (ppc64_interrupt_controller == IC_OPEN_PIC && pSeries_mpic) {
+			int addr = root_size_cells * (index + 2) - 1;
+			mpic_assign_isu(pSeries_mpic, index, opprop[addr]);
+		}
+#endif
+		index++;
+	}
+
+	of_node_put(root);
+	pci_devs_phb_init();
+
+	/*
+	 * pci_probe_only and pci_assign_all_buses can be set via properties
+	 * in chosen.
+	 */
+	if (of_chosen) {
+		int *prop;
+
+		prop = (int *)get_property(of_chosen, "linux,pci-probe-only",
+					   NULL);
+		if (prop)
+			pci_probe_only = *prop;
+
+		prop = (int *)get_property(of_chosen,
+					   "linux,pci-assign-all-buses", NULL);
+		if (prop)
+			pci_assign_all_buses = *prop;
+	}
+
+	return 0;
+}
+
+struct pci_controller * __devinit init_phb_dynamic(struct device_node *dn)
+{
+	struct device_node *root = of_find_node_by_path("/");
+	unsigned int root_size_cells = 0;
+	struct pci_controller *phb;
+	struct pci_bus *bus;
+	int primary;
+
+	root_size_cells = prom_n_size_cells(root);
+
+	primary = list_empty(&hose_list);
+	phb = alloc_phb_dynamic(dn, root_size_cells);
+	if (!phb)
+		return NULL;
+
+	pci_process_bridge_OF_ranges(phb, dn);
+
+	pci_setup_phb_io_dynamic(phb, primary);
+	of_node_put(root);
+
+	pci_devs_phb_init_dynamic(phb);
+	phb->last_busno = 0xff;
+	bus = pci_scan_bus(phb->first_busno, phb->ops, phb->arch_data);
+	phb->bus = bus;
+	phb->last_busno = bus->subordinate;
+
+	return phb;
+}
+EXPORT_SYMBOL(init_phb_dynamic);
+
+/* RPA-specific bits for removing PHBs */
+int pcibios_remove_root_bus(struct pci_controller *phb)
+{
+	struct pci_bus *b = phb->bus;
+	struct resource *res;
+	int rc, i;
+
+	res = b->resource[0];
+	if (!res->flags) {
+		printk(KERN_ERR "%s: no IO resource for PHB %s\n", __FUNCTION__,
+				b->name);
+		return 1;
+	}
+
+	rc = unmap_bus_range(b);
+	if (rc) {
+		printk(KERN_ERR "%s: failed to unmap IO on bus %s\n",
+			__FUNCTION__, b->name);
+		return 1;
+	}
+
+	if (release_resource(res)) {
+		printk(KERN_ERR "%s: failed to release IO on bus %s\n",
+				__FUNCTION__, b->name);
+		return 1;
+	}
+
+	for (i = 1; i < 3; ++i) {
+		res = b->resource[i];
+		if (!res->flags && i == 0) {
+			printk(KERN_ERR "%s: no MEM resource for PHB %s\n",
+				__FUNCTION__, b->name);
+			return 1;
+		}
+		if (res->flags && release_resource(res)) {
+			printk(KERN_ERR
+			       "%s: failed to release IO %d on bus %s\n",
+				__FUNCTION__, i, b->name);
+			return 1;
+		}
+	}
+
+	list_del(&phb->list_node);
+	if (phb->is_dynamic)
+		kfree(phb);
+
+	return 0;
+}
+EXPORT_SYMBOL(pcibios_remove_root_bus);
-- 
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 'arch')

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 'arch')

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 'arch')

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 cebf589c822b5de87098b57644024d16f8dbc1bb Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 23 Jun 2005 09:43:43 +1000
Subject: [PATCH] ppc64: Add driver for BPA interrupt controllers

Add support for the integrated interrupt controller on BPA
CPUs. There is one of those for each SMT thread.

The mapping of interrupt numbers to HW interrupt sources
is described in arch/ppc64/kernel/bpa_iic.h.

This version hardcodes the 'Spider' chip as the secondary
interrupt controller. That is not really generic for the
architecture, but at the moment it is the only secondary
PIC that exists.

A little more work will be needed on this as soon as
we have boards with multiple external interrupt controllers.

Signed-off-by: Arnd Bergmann <arndb@de.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/Kconfig              |  15 +++
 arch/ppc64/kernel/Makefile      |   8 +-
 arch/ppc64/kernel/bpa_iic.c     | 270 ++++++++++++++++++++++++++++++++++++++++
 arch/ppc64/kernel/bpa_iic.h     |  62 +++++++++
 arch/ppc64/kernel/bpa_setup.c   |   6 +-
 arch/ppc64/kernel/pSeries_smp.c |  69 +++++++++-
 arch/ppc64/kernel/smp.c         |   4 +-
 arch/ppc64/kernel/spider-pic.c  | 191 ++++++++++++++++++++++++++++
 8 files changed, 614 insertions(+), 11 deletions(-)
 create mode 100644 arch/ppc64/kernel/bpa_iic.c
 create mode 100644 arch/ppc64/kernel/bpa_iic.h
 create mode 100644 arch/ppc64/kernel/spider-pic.c

(limited to 'arch')

diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig
index c7f2f0a4d85..bae56ec76ea 100644
--- a/arch/ppc64/Kconfig
+++ b/arch/ppc64/Kconfig
@@ -110,6 +110,21 @@ config PPC_OF
 	bool
 	default y
 
+config XICS
+	depends on PPC_PSERIES
+	bool
+	default y
+
+config MPIC
+	depends on PPC_PSERIES || PPC_PMAC || PPC_MAPLE
+	bool
+	default y
+
+config BPA_IIC
+	depends on PPC_BPA
+	bool
+	default y
+
 # VMX is pSeries only for now until somebody writes the iSeries
 # exception vectors for it
 config ALTIVEC
diff --git a/arch/ppc64/kernel/Makefile b/arch/ppc64/kernel/Makefile
index c89983ab909..bc58d80ee3d 100644
--- a/arch/ppc64/kernel/Makefile
+++ b/arch/ppc64/kernel/Makefile
@@ -27,13 +27,13 @@ obj-$(CONFIG_PPC_ISERIES) += HvCall.o HvLpConfig.o LparData.o \
 			     mf.o HvLpEvent.o iSeries_proc.o iSeries_htab.o \
 			     iSeries_iommu.o
 
-obj-$(CONFIG_PPC_MULTIPLATFORM) += nvram.o i8259.o prom_init.o prom.o mpic.o
+obj-$(CONFIG_PPC_MULTIPLATFORM) += nvram.o i8259.o prom_init.o prom.o
 
 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
+			     pSeries_setup.o pSeries_iommu.o
 
-obj-$(CONFIG_PPC_BPA) += bpa_setup.o bpa_nvram.o
+obj-$(CONFIG_PPC_BPA) += bpa_setup.o bpa_nvram.o bpa_iic.o spider-pic.o
 
 obj-$(CONFIG_EEH)		+= eeh.o
 obj-$(CONFIG_PROC_FS)		+= proc_ppc64.o
@@ -49,6 +49,8 @@ obj-$(CONFIG_HVC_CONSOLE)	+= hvconsole.o
 obj-$(CONFIG_BOOTX_TEXT)	+= btext.o
 obj-$(CONFIG_HVCS)		+= hvcserver.o
 obj-$(CONFIG_IBMVIO)		+= vio.o
+obj-$(CONFIG_XICS)		+= xics.o
+obj-$(CONFIG_MPIC)		+= mpic.o
 
 obj-$(CONFIG_PPC_PMAC)		+= pmac_setup.o pmac_feature.o pmac_pci.o \
 				   pmac_time.o pmac_nvram.o pmac_low_i2c.o
diff --git a/arch/ppc64/kernel/bpa_iic.c b/arch/ppc64/kernel/bpa_iic.c
new file mode 100644
index 00000000000..c8f3dc3fad7
--- /dev/null
+++ b/arch/ppc64/kernel/bpa_iic.c
@@ -0,0 +1,270 @@
+/*
+ * BPA Internal Interrupt Controller
+ *
+ * (C) Copyright IBM Deutschland Entwicklung GmbH 2005
+ *
+ * Author: Arnd Bergmann <arndb@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/config.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/percpu.h>
+#include <linux/types.h>
+
+#include <asm/io.h>
+#include <asm/pgtable.h>
+#include <asm/prom.h>
+#include <asm/ptrace.h>
+
+#include "bpa_iic.h"
+
+struct iic_pending_bits {
+	u32 data;
+	u8 flags;
+	u8 class;
+	u8 source;
+	u8 prio;
+};
+
+enum iic_pending_flags {
+	IIC_VALID = 0x80,
+	IIC_IPI   = 0x40,
+};
+
+struct iic_regs {
+	struct iic_pending_bits pending;
+	struct iic_pending_bits pending_destr;
+	u64 generate;
+	u64 prio;
+};
+
+struct iic {
+	struct iic_regs __iomem *regs;
+};
+
+static DEFINE_PER_CPU(struct iic, iic);
+
+void iic_local_enable(void)
+{
+	out_be64(&__get_cpu_var(iic).regs->prio, 0xff);
+}
+
+void iic_local_disable(void)
+{
+	out_be64(&__get_cpu_var(iic).regs->prio, 0x0);
+}
+
+static unsigned int iic_startup(unsigned int irq)
+{
+	return 0;
+}
+
+static void iic_enable(unsigned int irq)
+{
+	iic_local_enable();
+}
+
+static void iic_disable(unsigned int irq)
+{
+}
+
+static void iic_end(unsigned int irq)
+{
+	iic_local_enable();
+}
+
+static struct hw_interrupt_type iic_pic = {
+	.typename = " BPA-IIC  ",
+	.startup = iic_startup,
+	.enable = iic_enable,
+	.disable = iic_disable,
+	.end = iic_end,
+};
+
+static int iic_external_get_irq(struct iic_pending_bits pending)
+{
+	int irq;
+	unsigned char node, unit;
+
+	node = pending.source >> 4;
+	unit = pending.source & 0xf;
+	irq = -1;
+
+	/*
+	 * This mapping is specific to the Broadband
+	 * Engine. We might need to get the numbers
+	 * from the device tree to support future CPUs.
+	 */
+	switch (unit) {
+	case 0x00:
+	case 0x0b:
+		/*
+		 * One of these units can be connected
+		 * to an external interrupt controller.
+		 */
+		if (pending.prio > 0x3f ||
+		    pending.class != 2)
+			break;
+		irq = IIC_EXT_OFFSET
+			+ spider_get_irq(pending.prio + node * IIC_NODE_STRIDE)
+			+ node * IIC_NODE_STRIDE;
+		break;
+	case 0x01 ... 0x04:
+	case 0x07 ... 0x0a:
+		/*
+		 * These units are connected to the SPEs
+		 */
+		if (pending.class > 2)
+			break;
+		irq = IIC_SPE_OFFSET
+			+ pending.class * IIC_CLASS_STRIDE
+			+ node * IIC_NODE_STRIDE
+			+ unit;
+		break;
+	}
+	if (irq == -1)
+		printk(KERN_WARNING "Unexpected interrupt class %02x, "
+			"source %02x, prio %02x, cpu %02x\n", pending.class,
+			pending.source, pending.prio, smp_processor_id());
+	return irq;
+}
+
+/* Get an IRQ number from the pending state register of the IIC */
+int iic_get_irq(struct pt_regs *regs)
+{
+	struct iic *iic;
+	int irq;
+	struct iic_pending_bits pending;
+
+	iic = &__get_cpu_var(iic);
+	*(unsigned long *) &pending = 
+		in_be64((unsigned long __iomem *) &iic->regs->pending_destr);
+
+	irq = -1;
+	if (pending.flags & IIC_VALID) {
+		if (pending.flags & IIC_IPI) {
+			irq = IIC_IPI_OFFSET + (pending.prio >> 4);
+/*
+			if (irq > 0x80)
+				printk(KERN_WARNING "Unexpected IPI prio %02x"
+					"on CPU %02x\n", pending.prio,
+							smp_processor_id());
+*/
+		} else {
+			irq = iic_external_get_irq(pending);
+		}
+	}
+	return irq;
+}
+
+static struct iic_regs __iomem *find_iic(int cpu)
+{
+	struct device_node *np;
+	int nodeid = cpu / 2;
+	unsigned long regs;
+	struct iic_regs __iomem *iic_regs;
+
+	for (np = of_find_node_by_type(NULL, "cpu");
+	     np;
+	     np = of_find_node_by_type(np, "cpu")) {
+		if (nodeid == *(int *)get_property(np, "node-id", NULL))
+			break;
+	}
+
+	if (!np) {
+		printk(KERN_WARNING "IIC: CPU %d not found\n", cpu);
+		iic_regs = NULL;
+	} else {
+		regs = *(long *)get_property(np, "iic", NULL);
+
+		/* hack until we have decided on the devtree info */
+		regs += 0x400;
+		if (cpu & 1)
+			regs += 0x20;
+
+		printk(KERN_DEBUG "IIC for CPU %d at %lx\n", cpu, regs);
+		iic_regs = __ioremap(regs, sizeof(struct iic_regs),
+						 _PAGE_NO_CACHE);
+	}
+	return iic_regs;
+}
+
+#ifdef CONFIG_SMP
+void iic_setup_cpu(void)
+{
+	out_be64(&__get_cpu_var(iic).regs->prio, 0xff);
+}
+
+void iic_cause_IPI(int cpu, int mesg)
+{
+	out_be64(&per_cpu(iic, cpu).regs->generate, mesg);
+}
+
+static irqreturn_t iic_ipi_action(int irq, void *dev_id, struct pt_regs *regs)
+{
+
+	smp_message_recv(irq - IIC_IPI_OFFSET, regs);
+	return IRQ_HANDLED;
+}
+
+static void iic_request_ipi(int irq, const char *name)
+{
+	/* IPIs are marked SA_INTERRUPT as they must run with irqs
+	 * disabled */
+	get_irq_desc(irq)->handler = &iic_pic;
+	get_irq_desc(irq)->status |= IRQ_PER_CPU;
+	request_irq(irq, iic_ipi_action, SA_INTERRUPT, name, NULL);
+}
+
+void iic_request_IPIs(void)
+{
+	iic_request_ipi(IIC_IPI_OFFSET + PPC_MSG_CALL_FUNCTION, "IPI-call");
+	iic_request_ipi(IIC_IPI_OFFSET + PPC_MSG_RESCHEDULE, "IPI-resched");
+#ifdef CONFIG_DEBUGGER
+	iic_request_ipi(IIC_IPI_OFFSET + PPC_MSG_DEBUGGER_BREAK, "IPI-debug");
+#endif /* CONFIG_DEBUGGER */
+}
+#endif /* CONFIG_SMP */
+
+static void iic_setup_spe_handlers(void)
+{
+	int be, isrc;
+
+	/* Assume two threads per BE are present */
+	for (be=0; be < num_present_cpus() / 2; be++) {
+		for (isrc = 0; isrc < IIC_CLASS_STRIDE * 3; isrc++) {
+			int irq = IIC_NODE_STRIDE * be + IIC_SPE_OFFSET + isrc;
+			get_irq_desc(irq)->handler = &iic_pic;
+		}
+	}
+}
+
+void iic_init_IRQ(void)
+{
+	int cpu, irq_offset;
+	struct iic *iic;
+
+	irq_offset = 0;
+	for_each_cpu(cpu) {
+		iic = &per_cpu(iic, cpu);
+		iic->regs = find_iic(cpu);
+		if (iic->regs)
+			out_be64(&iic->regs->prio, 0xff);
+	}
+	iic_setup_spe_handlers();
+}
diff --git a/arch/ppc64/kernel/bpa_iic.h b/arch/ppc64/kernel/bpa_iic.h
new file mode 100644
index 00000000000..6833c302216
--- /dev/null
+++ b/arch/ppc64/kernel/bpa_iic.h
@@ -0,0 +1,62 @@
+#ifndef ASM_BPA_IIC_H
+#define ASM_BPA_IIC_H
+#ifdef __KERNEL__
+/*
+ * Mapping of IIC pending bits into per-node
+ * interrupt numbers.
+ *
+ * IRQ     FF CC SS PP   FF CC SS PP	Description
+ *
+ * 00-3f   80 02 +0 00 - 80 02 +0 3f	South Bridge
+ * 00-3f   80 02 +b 00 - 80 02 +b 3f	South Bridge
+ * 41-4a   80 00 +1 ** - 80 00 +a **	SPU Class 0
+ * 51-5a   80 01 +1 ** - 80 01 +a **	SPU Class 1
+ * 61-6a   80 02 +1 ** - 80 02 +a **	SPU Class 2
+ * 70-7f   C0 ** ** 00 - C0 ** ** 0f	IPI
+ *
+ *    F flags
+ *    C class
+ *    S source
+ *    P Priority
+ *    + node number
+ *    * don't care
+ *
+ * A node consists of a Broadband Engine and an optional
+ * south bridge device providing a maximum of 64 IRQs.
+ * The south bridge may be connected to either IOIF0
+ * or IOIF1.
+ * Each SPE is represented as three IRQ lines, one per
+ * interrupt class.
+ * 16 IRQ numbers are reserved for inter processor
+ * interruptions, although these are only used in the
+ * range of the first node.
+ *
+ * This scheme needs 128 IRQ numbers per BIF node ID,
+ * which means that with the total of 512 lines
+ * available, we can have a maximum of four nodes.
+ */
+
+enum {
+	IIC_EXT_OFFSET   = 0x00, /* Start of south bridge IRQs */
+	IIC_NUM_EXT      = 0x40, /* Number of south bridge IRQs */
+	IIC_SPE_OFFSET   = 0x40, /* Start of SPE interrupts */
+	IIC_CLASS_STRIDE = 0x10, /* SPE IRQs per class    */
+	IIC_IPI_OFFSET   = 0x70, /* Start of IPI IRQs */
+	IIC_NUM_IPIS     = 0x10, /* IRQs reserved for IPI */
+	IIC_NODE_STRIDE  = 0x80, /* Total IRQs per node   */
+};
+
+extern void iic_init_IRQ(void);
+extern int  iic_get_irq(struct pt_regs *regs);
+extern void iic_cause_IPI(int cpu, int mesg);
+extern void iic_request_IPIs(void);
+extern void iic_setup_cpu(void);
+extern void iic_local_enable(void);
+extern void iic_local_disable(void);
+
+
+extern void spider_init_IRQ(void);
+extern int spider_get_irq(unsigned long int_pending);
+
+#endif
+#endif /* ASM_BPA_IIC_H */
diff --git a/arch/ppc64/kernel/bpa_setup.c b/arch/ppc64/kernel/bpa_setup.c
index d1992dd2d61..0a43d8a93d7 100644
--- a/arch/ppc64/kernel/bpa_setup.c
+++ b/arch/ppc64/kernel/bpa_setup.c
@@ -45,6 +45,7 @@
 #include <asm/cputable.h>
 
 #include "pci.h"
+#include "bpa_iic.h"
 
 #ifdef DEBUG
 #define DBG(fmt...) udbg_printf(fmt)
@@ -71,6 +72,9 @@ static void bpa_progress(char *s, unsigned short hex)
 
 static void __init bpa_setup_arch(void)
 {
+	ppc_md.init_IRQ       = iic_init_IRQ;
+	ppc_md.get_irq        = iic_get_irq;
+
 #ifdef CONFIG_SMP
 	smp_init_pSeries();
 #endif
@@ -86,7 +90,7 @@ static void __init bpa_setup_arch(void)
 	/* Find and initialize PCI host bridges */
 	init_pci_config_tokens();
 	find_and_init_phbs();
-
+	spider_init_IRQ();
 #ifdef CONFIG_DUMMY_CONSOLE
 	conswitchp = &dummy_con;
 #endif
diff --git a/arch/ppc64/kernel/pSeries_smp.c b/arch/ppc64/kernel/pSeries_smp.c
index 4203bd020c8..30154140f7e 100644
--- a/arch/ppc64/kernel/pSeries_smp.c
+++ b/arch/ppc64/kernel/pSeries_smp.c
@@ -1,5 +1,5 @@
 /*
- * SMP support for pSeries machines.
+ * SMP support for pSeries and BPA machines.
  *
  * Dave Engebretsen, Peter Bergner, and
  * Mike Corrigan {engebret|bergner|mikec}@us.ibm.com
@@ -47,6 +47,7 @@
 #include <asm/pSeries_reconfig.h>
 
 #include "mpic.h"
+#include "bpa_iic.h"
 
 #ifdef DEBUG
 #define DBG(fmt...) udbg_printf(fmt)
@@ -286,6 +287,7 @@ static inline int __devinit smp_startup_cpu(unsigned int lcpu)
 	return 1;
 }
 
+#ifdef CONFIG_XICS
 static inline void smp_xics_do_message(int cpu, int msg)
 {
 	set_bit(msg, &xics_ipi_message[cpu].value);
@@ -327,6 +329,37 @@ static void __devinit smp_xics_setup_cpu(int cpu)
 	cpu_clear(cpu, of_spin_map);
 
 }
+#endif /* CONFIG_XICS */
+#ifdef CONFIG_BPA_IIC
+static void smp_iic_message_pass(int target, int msg)
+{
+	unsigned int i;
+
+	if (target < NR_CPUS) {
+		iic_cause_IPI(target, msg);
+	} else {
+		for_each_online_cpu(i) {
+			if (target == MSG_ALL_BUT_SELF
+			    && i == smp_processor_id())
+				continue;
+			iic_cause_IPI(i, msg);
+		}
+	}
+}
+
+static int __init smp_iic_probe(void)
+{
+	iic_request_IPIs();
+
+	return cpus_weight(cpu_possible_map);
+}
+
+static void __devinit smp_iic_setup_cpu(int cpu)
+{
+	if (cpu != boot_cpuid)
+		iic_setup_cpu();
+}
+#endif /* CONFIG_BPA_IIC */
 
 static DEFINE_SPINLOCK(timebase_lock);
 static unsigned long timebase = 0;
@@ -381,14 +414,15 @@ static int smp_pSeries_cpu_bootable(unsigned int nr)
 
 	return 1;
 }
-
+#ifdef CONFIG_MPIC
 static struct smp_ops_t pSeries_mpic_smp_ops = {
 	.message_pass	= smp_mpic_message_pass,
 	.probe		= smp_mpic_probe,
 	.kick_cpu	= smp_pSeries_kick_cpu,
 	.setup_cpu	= smp_mpic_setup_cpu,
 };
-
+#endif
+#ifdef CONFIG_XICS
 static struct smp_ops_t pSeries_xics_smp_ops = {
 	.message_pass	= smp_xics_message_pass,
 	.probe		= smp_xics_probe,
@@ -396,6 +430,16 @@ static struct smp_ops_t pSeries_xics_smp_ops = {
 	.setup_cpu	= smp_xics_setup_cpu,
 	.cpu_bootable	= smp_pSeries_cpu_bootable,
 };
+#endif
+#ifdef CONFIG_BPA_IIC
+static struct smp_ops_t bpa_iic_smp_ops = {
+	.message_pass	= smp_iic_message_pass,
+	.probe		= smp_iic_probe,
+	.kick_cpu	= smp_pSeries_kick_cpu,
+	.setup_cpu	= smp_iic_setup_cpu,
+	.cpu_bootable	= smp_pSeries_cpu_bootable,
+};
+#endif
 
 /* This is called very early */
 void __init smp_init_pSeries(void)
@@ -404,10 +448,25 @@ void __init smp_init_pSeries(void)
 
 	DBG(" -> smp_init_pSeries()\n");
 
-	if (ppc64_interrupt_controller == IC_OPEN_PIC)
+	switch (ppc64_interrupt_controller) {
+#ifdef CONFIG_MPIC
+	case IC_OPEN_PIC:
 		smp_ops = &pSeries_mpic_smp_ops;
-	else
+		break;
+#endif
+#ifdef CONFIG_XICS
+	case IC_PPC_XIC:
 		smp_ops = &pSeries_xics_smp_ops;
+		break;
+#endif
+#ifdef CONFIG_BPA_IIC
+	case IC_BPA_IIC:
+		smp_ops = &bpa_iic_smp_ops;
+		break;
+#endif
+	default:
+		panic("Invalid interrupt controller");
+	}
 
 #ifdef CONFIG_HOTPLUG_CPU
 	smp_ops->cpu_disable = pSeries_cpu_disable;
diff --git a/arch/ppc64/kernel/smp.c b/arch/ppc64/kernel/smp.c
index 9ef5d36d6b2..2fcddfcb594 100644
--- a/arch/ppc64/kernel/smp.c
+++ b/arch/ppc64/kernel/smp.c
@@ -71,7 +71,7 @@ void smp_call_function_interrupt(void);
 
 int smt_enabled_at_boot = 1;
 
-#ifdef CONFIG_PPC_MULTIPLATFORM
+#ifdef CONFIG_MPIC
 void smp_mpic_message_pass(int target, int msg)
 {
 	/* make sure we're sending something that translates to an IPI */
@@ -128,7 +128,7 @@ void __devinit smp_generic_kick_cpu(int nr)
 	smp_mb();
 }
 
-#endif /* CONFIG_PPC_MULTIPLATFORM */
+#endif /* CONFIG_MPIC */
 
 static void __init smp_space_timers(unsigned int max_cpus)
 {
diff --git a/arch/ppc64/kernel/spider-pic.c b/arch/ppc64/kernel/spider-pic.c
new file mode 100644
index 00000000000..d5c9a02fb11
--- /dev/null
+++ b/arch/ppc64/kernel/spider-pic.c
@@ -0,0 +1,191 @@
+/*
+ * External Interrupt Controller on Spider South Bridge
+ *
+ * (C) Copyright IBM Deutschland Entwicklung GmbH 2005
+ *
+ * Author: Arnd Bergmann <arndb@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/interrupt.h>
+#include <linux/irq.h>
+
+#include <asm/pgtable.h>
+#include <asm/prom.h>
+#include <asm/io.h>
+
+#include "bpa_iic.h"
+
+/* register layout taken from Spider spec, table 7.4-4 */
+enum {
+	TIR_DEN		= 0x004, /* Detection Enable Register */
+	TIR_MSK		= 0x084, /* Mask Level Register */
+	TIR_EDC		= 0x0c0, /* Edge Detection Clear Register */
+	TIR_PNDA	= 0x100, /* Pending Register A */
+	TIR_PNDB	= 0x104, /* Pending Register B */
+	TIR_CS		= 0x144, /* Current Status Register */
+	TIR_LCSA	= 0x150, /* Level Current Status Register A */
+	TIR_LCSB	= 0x154, /* Level Current Status Register B */
+	TIR_LCSC	= 0x158, /* Level Current Status Register C */
+	TIR_LCSD	= 0x15c, /* Level Current Status Register D */
+	TIR_CFGA	= 0x200, /* Setting Register A0 */
+	TIR_CFGB	= 0x204, /* Setting Register B0 */
+			/* 0x208 ... 0x3ff Setting Register An/Bn */
+	TIR_PPNDA	= 0x400, /* Packet Pending Register A */
+	TIR_PPNDB	= 0x404, /* Packet Pending Register B */
+	TIR_PIERA	= 0x408, /* Packet Output Error Register A */
+	TIR_PIERB	= 0x40c, /* Packet Output Error Register B */
+	TIR_PIEN	= 0x444, /* Packet Output Enable Register */
+	TIR_PIPND	= 0x454, /* Packet Output Pending Register */
+	TIRDID		= 0x484, /* Spider Device ID Register */
+	REISTIM		= 0x500, /* Reissue Command Timeout Time Setting */
+	REISTIMEN	= 0x504, /* Reissue Command Timeout Setting */
+	REISWAITEN	= 0x508, /* Reissue Wait Control*/
+};
+
+static void __iomem *spider_pics[4];
+
+static void __iomem *spider_get_pic(int irq)
+{
+	int node = irq / IIC_NODE_STRIDE;
+	irq %= IIC_NODE_STRIDE;
+
+	if (irq >= IIC_EXT_OFFSET &&
+	    irq < IIC_EXT_OFFSET + IIC_NUM_EXT &&
+	    spider_pics)
+		return spider_pics[node];
+	return NULL;
+}
+
+static int spider_get_nr(unsigned int irq)
+{
+	return (irq % IIC_NODE_STRIDE) - IIC_EXT_OFFSET;
+}
+
+static void __iomem *spider_get_irq_config(int irq)
+{
+	void __iomem *pic;
+	pic = spider_get_pic(irq);
+	return pic + TIR_CFGA + 8 * spider_get_nr(irq);
+}
+
+static void spider_enable_irq(unsigned int irq)
+{
+	void __iomem *cfg = spider_get_irq_config(irq);
+	irq = spider_get_nr(irq);
+
+	out_be32(cfg, in_be32(cfg) | 0x3107000eu);
+	out_be32(cfg + 4, in_be32(cfg + 4) | 0x00020000u | irq);
+}
+
+static void spider_disable_irq(unsigned int irq)
+{
+	void __iomem *cfg = spider_get_irq_config(irq);
+	irq = spider_get_nr(irq);
+
+	out_be32(cfg, in_be32(cfg) & ~0x30000000u);
+}
+
+static unsigned int spider_startup_irq(unsigned int irq)
+{
+	spider_enable_irq(irq);
+	return 0;
+}
+
+static void spider_shutdown_irq(unsigned int irq)
+{
+	spider_disable_irq(irq);
+}
+
+static void spider_end_irq(unsigned int irq)
+{
+	spider_enable_irq(irq);
+}
+
+static void spider_ack_irq(unsigned int irq)
+{
+	spider_disable_irq(irq);
+	iic_local_enable();
+}
+
+static struct hw_interrupt_type spider_pic = {
+	.typename = " SPIDER   ",
+	.startup = spider_startup_irq,
+	.shutdown = spider_shutdown_irq,
+	.enable = spider_enable_irq,
+	.disable = spider_disable_irq,
+	.ack = spider_ack_irq,
+	.end = spider_end_irq,
+};
+
+
+int spider_get_irq(unsigned long int_pending)
+{
+	void __iomem *regs = spider_get_pic(int_pending);
+	unsigned long cs;
+	int irq;
+
+	cs = in_be32(regs + TIR_CS);
+
+	irq = cs >> 24;
+	if (irq != 63)
+		return irq;
+
+	return -1;
+}
+ 
+void spider_init_IRQ(void)
+{
+	int node;
+	struct device_node *dn;
+	unsigned int *property;
+	long spiderpic;
+	int n;
+
+/* FIXME: detect multiple PICs as soon as the device tree has them */
+	for (node = 0; node < 1; node++) {
+		dn = of_find_node_by_path("/");
+		n = prom_n_addr_cells(dn);
+		property = (unsigned int *) get_property(dn,
+				"platform-spider-pic", NULL);
+
+		if (!property)
+			continue;
+		for (spiderpic = 0; n > 0; --n)
+			spiderpic = (spiderpic << 32) + *property++;
+		printk(KERN_DEBUG "SPIDER addr: %lx\n", spiderpic);
+		spider_pics[node] = __ioremap(spiderpic, 0x800, _PAGE_NO_CACHE);
+		for (n = 0; n < IIC_NUM_EXT; n++) {
+			int irq = n + IIC_EXT_OFFSET + node * IIC_NODE_STRIDE;
+			get_irq_desc(irq)->handler = &spider_pic;
+
+ 		/* do not mask any interrupts because of level */
+ 		out_be32(spider_pics[node] + TIR_MSK, 0x0);
+ 		
+ 		/* disable edge detection clear */
+ 		/* out_be32(spider_pics[node] + TIR_EDC, 0x0); */
+ 		
+ 		/* enable interrupt packets to be output */
+ 		out_be32(spider_pics[node] + TIR_PIEN,
+			in_be32(spider_pics[node] + TIR_PIEN) | 0x1);
+ 		
+ 		/* Enable the interrupt detection enable bit. Do this last! */
+ 		out_be32(spider_pics[node] + TIR_DEN,
+			in_be32(spider_pics[node] +TIR_DEN) | 0x1);
+
+		}
+	}
+}
-- 
cgit v1.2.3-70-g09d2


From ae209cf10086b97e92e39af7cec0f84b21b6fca3 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 23 Jun 2005 09:43:54 +1000
Subject: [PATCH] ppc64: Add driver for BPA iommu

Implementation of software load support for the BE iommu. This is very
different from other iommu code on ppc64, since we only do a static mapping.
The mapping is currently hardcoded but should really be read from the
firmware, but they don't set up the device nodes yet. There is a single
512MB DMA window for PCI, USB and ethernet at 0x20000000 for our RAM.

The Cell processor can put the I/O page table either in memory like
the hashed page table (hardware load) or have the operating system
write the entries into memory mapped CPU registers (software load).

I use the software load mechanism because I know that all I/O page
table entries for the amount of installed physical memory fit into
the IO TLB cache. At the point when we get machines with more than
4GB of installed memory, we can either use hardware I/O page table
access like the other platforms do or dynamically update the I/O
TLB entries when a page fault occurs in the I/O subsystem.

The software load can then use the macros that I have implemented
for the static mapping in order to do the TLB cache updates.

Signed-off-by: Arnd Bergmann <arndb@de.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/Makefile    |   3 +-
 arch/ppc64/kernel/bpa_iommu.c | 377 ++++++++++++++++++++++++++++++++++++++++++
 arch/ppc64/kernel/bpa_iommu.h |  65 ++++++++
 arch/ppc64/kernel/bpa_setup.c |   3 +-
 4 files changed, 446 insertions(+), 2 deletions(-)
 create mode 100644 arch/ppc64/kernel/bpa_iommu.c
 create mode 100644 arch/ppc64/kernel/bpa_iommu.h

(limited to 'arch')

diff --git a/arch/ppc64/kernel/Makefile b/arch/ppc64/kernel/Makefile
index bc58d80ee3d..dffbfb7ac8d 100644
--- a/arch/ppc64/kernel/Makefile
+++ b/arch/ppc64/kernel/Makefile
@@ -33,7 +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 \
 			     pSeries_setup.o pSeries_iommu.o
 
-obj-$(CONFIG_PPC_BPA) += bpa_setup.o bpa_nvram.o bpa_iic.o spider-pic.o
+obj-$(CONFIG_PPC_BPA) += bpa_setup.o bpa_iommu.o bpa_nvram.o \
+			 bpa_iic.o spider-pic.o
 
 obj-$(CONFIG_EEH)		+= eeh.o
 obj-$(CONFIG_PROC_FS)		+= proc_ppc64.o
diff --git a/arch/ppc64/kernel/bpa_iommu.c b/arch/ppc64/kernel/bpa_iommu.c
new file mode 100644
index 00000000000..f33a7bccb0d
--- /dev/null
+++ b/arch/ppc64/kernel/bpa_iommu.c
@@ -0,0 +1,377 @@
+/*
+ * IOMMU implementation for Broadband Processor Architecture
+ * We just establish a linear mapping at boot by setting all the
+ * IOPT cache entries in the CPU.
+ * The mapping functions should be identical to pci_direct_iommu, 
+ * except for the handling of the high order bit that is required
+ * by the Spider bridge. These should be split into a separate
+ * file at the point where we get a different bridge chip.
+ *
+ * Copyright (C) 2005 IBM Deutschland Entwicklung GmbH,
+ *			 Arnd Bergmann <arndb@de.ibm.com>
+ *
+ * Based on linear mapping
+ * Copyright (C) 2003 Benjamin Herrenschmidt (benh@kernel.crashing.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.
+ */
+
+#undef DEBUG
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/string.h>
+#include <linux/init.h>
+#include <linux/bootmem.h>
+#include <linux/mm.h>
+#include <linux/dma-mapping.h>
+
+#include <asm/sections.h>
+#include <asm/iommu.h>
+#include <asm/io.h>
+#include <asm/prom.h>
+#include <asm/pci-bridge.h>
+#include <asm/machdep.h>
+#include <asm/pmac_feature.h>
+#include <asm/abs_addr.h>
+#include <asm/system.h>
+
+#include "pci.h"
+#include "bpa_iommu.h"
+
+static inline unsigned long 
+get_iopt_entry(unsigned long real_address, unsigned long ioid,
+			 unsigned long prot)
+{
+	return (prot & IOPT_PROT_MASK)
+	     | (IOPT_COHERENT)
+	     | (IOPT_ORDER_VC)
+	     | (real_address & IOPT_RPN_MASK)
+	     | (ioid & IOPT_IOID_MASK);
+}
+
+typedef struct {
+	unsigned long val;
+} ioste;
+
+static inline ioste
+mk_ioste(unsigned long val)
+{
+	ioste ioste = { .val = val, };
+	return ioste;
+}
+
+static inline ioste
+get_iost_entry(unsigned long iopt_base, unsigned long io_address, unsigned page_size)
+{
+	unsigned long ps;
+	unsigned long iostep;
+	unsigned long nnpt;
+	unsigned long shift;
+
+	switch (page_size) {
+	case 0x1000000:
+		ps = IOST_PS_16M;
+		nnpt = 0;  /* one page per segment */
+		shift = 5; /* segment has 16 iopt entries */
+		break;
+
+	case 0x100000:
+		ps = IOST_PS_1M;
+		nnpt = 0;  /* one page per segment */
+		shift = 1; /* segment has 256 iopt entries */
+		break;
+
+	case 0x10000:
+		ps = IOST_PS_64K;
+		nnpt = 0x07; /* 8 pages per io page table */
+		shift = 0;   /* all entries are used */
+		break;
+
+	case 0x1000:
+		ps = IOST_PS_4K;
+		nnpt = 0x7f; /* 128 pages per io page table */
+		shift = 0;   /* all entries are used */
+		break;
+
+	default: /* not a known compile time constant */
+		BUILD_BUG_ON(1);
+		break;
+	}
+
+	iostep = iopt_base +
+			 /* need 8 bytes per iopte */
+			(((io_address / page_size * 8)
+			 /* align io page tables on 4k page boundaries */
+				 << shift) 
+			 /* nnpt+1 pages go into each iopt */
+				 & ~(nnpt << 12));
+
+	nnpt++; /* this seems to work, but the documentation is not clear
+		   about wether we put nnpt or nnpt-1 into the ioste bits.
+		   In theory, this can't work for 4k pages. */
+	return mk_ioste(IOST_VALID_MASK
+			| (iostep & IOST_PT_BASE_MASK)
+			| ((nnpt << 5) & IOST_NNPT_MASK)
+			| (ps & IOST_PS_MASK));
+}
+
+/* compute the address of an io pte */
+static inline unsigned long
+get_ioptep(ioste iost_entry, unsigned long io_address)
+{
+	unsigned long iopt_base;
+	unsigned long page_size;
+	unsigned long page_number;
+	unsigned long iopt_offset;
+
+	iopt_base = iost_entry.val & IOST_PT_BASE_MASK;
+	page_size = iost_entry.val & IOST_PS_MASK;
+
+	/* decode page size to compute page number */
+	page_number = (io_address & 0x0fffffff) >> (10 + 2 * page_size);
+	/* page number is an offset into the io page table */
+	iopt_offset = (page_number << 3) & 0x7fff8ul;
+	return iopt_base + iopt_offset;
+}
+
+/* compute the tag field of the iopt cache entry */
+static inline unsigned long
+get_ioc_tag(ioste iost_entry, unsigned long io_address)
+{
+	unsigned long iopte = get_ioptep(iost_entry, io_address);
+
+	return IOPT_VALID_MASK
+	     | ((iopte & 0x00000000000000ff8ul) >> 3)
+	     | ((iopte & 0x0000003fffffc0000ul) >> 9);
+}
+
+/* compute the hashed 6 bit index for the 4-way associative pte cache */
+static inline unsigned long
+get_ioc_hash(ioste iost_entry, unsigned long io_address)
+{
+	unsigned long iopte = get_ioptep(iost_entry, io_address);
+
+	return ((iopte & 0x000000000000001f8ul) >> 3)
+	     ^ ((iopte & 0x00000000000020000ul) >> 17)
+	     ^ ((iopte & 0x00000000000010000ul) >> 15)
+	     ^ ((iopte & 0x00000000000008000ul) >> 13)
+	     ^ ((iopte & 0x00000000000004000ul) >> 11)
+	     ^ ((iopte & 0x00000000000002000ul) >> 9)
+	     ^ ((iopte & 0x00000000000001000ul) >> 7);
+}
+
+/* same as above, but pretend that we have a simpler 1-way associative
+   pte cache with an 8 bit index */
+static inline unsigned long
+get_ioc_hash_1way(ioste iost_entry, unsigned long io_address)
+{
+	unsigned long iopte = get_ioptep(iost_entry, io_address);
+
+	return ((iopte & 0x000000000000001f8ul) >> 3)
+	     ^ ((iopte & 0x00000000000020000ul) >> 17)
+	     ^ ((iopte & 0x00000000000010000ul) >> 15)
+	     ^ ((iopte & 0x00000000000008000ul) >> 13)
+	     ^ ((iopte & 0x00000000000004000ul) >> 11)
+	     ^ ((iopte & 0x00000000000002000ul) >> 9)
+	     ^ ((iopte & 0x00000000000001000ul) >> 7)
+	     ^ ((iopte & 0x0000000000000c000ul) >> 8);
+}
+
+static inline ioste
+get_iost_cache(void __iomem *base, unsigned long index)
+{
+	unsigned long __iomem *p = (base + IOC_ST_CACHE_DIR);
+	return mk_ioste(in_be64(&p[index]));
+}
+
+static inline void
+set_iost_cache(void __iomem *base, unsigned long index, ioste ste)
+{
+	unsigned long __iomem *p = (base + IOC_ST_CACHE_DIR);
+	pr_debug("ioste %02lx was %016lx, store %016lx", index,
+			get_iost_cache(base, index).val, ste.val);
+	out_be64(&p[index], ste.val);
+	pr_debug(" now %016lx\n", get_iost_cache(base, index).val);
+}
+
+static inline unsigned long
+get_iopt_cache(void __iomem *base, unsigned long index, unsigned long *tag)
+{
+	unsigned long __iomem *tags = (void *)(base + IOC_PT_CACHE_DIR);
+	unsigned long __iomem *p = (void *)(base + IOC_PT_CACHE_REG);	
+
+	*tag = tags[index];
+	rmb();
+	return *p;
+}
+
+static inline void
+set_iopt_cache(void __iomem *base, unsigned long index,
+		 unsigned long tag, unsigned long val)
+{
+	unsigned long __iomem *tags = base + IOC_PT_CACHE_DIR;
+	unsigned long __iomem *p = base + IOC_PT_CACHE_REG;
+	pr_debug("iopt %02lx was v%016lx/t%016lx, store v%016lx/t%016lx\n",
+		index, get_iopt_cache(base, index, &oldtag), oldtag, val, tag);
+
+	out_be64(p, val);
+	out_be64(&tags[index], tag);
+}
+
+static inline void
+set_iost_origin(void __iomem *base)
+{
+	unsigned long __iomem *p = base + IOC_ST_ORIGIN;
+	unsigned long origin = IOSTO_ENABLE | IOSTO_SW;
+
+	pr_debug("iost_origin %016lx, now %016lx\n", in_be64(p), origin);
+	out_be64(p, origin);
+}
+
+static inline void
+set_iocmd_config(void __iomem *base)
+{
+	unsigned long __iomem *p = base + 0xc00;
+	unsigned long conf;
+
+	conf = in_be64(p);
+	pr_debug("iost_conf %016lx, now %016lx\n", conf, conf | IOCMD_CONF_TE);
+	out_be64(p, conf | IOCMD_CONF_TE);
+}
+
+/* FIXME: get these from the device tree */
+#define ioc_base	0x20000511000ull
+#define ioc_mmio_base	0x20000510000ull
+#define ioid		0x48a
+#define iopt_phys_offset (- 0x20000000) /* We have a 512MB offset from the SB */
+#define io_page_size	0x1000000
+
+static unsigned long map_iopt_entry(unsigned long address)
+{
+	switch (address >> 20) {
+	case 0x600:
+		address = 0x24020000000ull; /* spider i/o */
+		break;
+	default:
+		address += iopt_phys_offset;
+		break;
+	}
+
+	return get_iopt_entry(address, ioid, IOPT_PROT_RW);
+}
+
+static void iommu_bus_setup_null(struct pci_bus *b) { }
+static void iommu_dev_setup_null(struct pci_dev *d) { }
+
+/* initialize the iommu to support a simple linear mapping
+ * for each DMA window used by any device. For now, we
+ * happen to know that there is only one DMA window in use,
+ * starting at iopt_phys_offset. */
+static void bpa_map_iommu(void)
+{
+	unsigned long address;
+	void __iomem *base;
+	ioste ioste;
+	unsigned long index;
+
+	base = __ioremap(ioc_base, 0x1000, _PAGE_NO_CACHE);
+	pr_debug("%lx mapped to %p\n", ioc_base, base);
+	set_iocmd_config(base);
+	iounmap(base);
+
+	base = __ioremap(ioc_mmio_base, 0x1000, _PAGE_NO_CACHE);
+	pr_debug("%lx mapped to %p\n", ioc_mmio_base, base);
+
+	set_iost_origin(base);
+
+	for (address = 0; address < 0x100000000ul; address += io_page_size) {
+		ioste = get_iost_entry(0x10000000000ul, address, io_page_size);
+		if ((address & 0xfffffff) == 0) /* segment start */
+			set_iost_cache(base, address >> 28, ioste);
+		index = get_ioc_hash_1way(ioste, address);
+		pr_debug("addr %08lx, index %02lx, ioste %016lx\n",
+					 address, index, ioste.val);
+		set_iopt_cache(base,
+			get_ioc_hash_1way(ioste, address),
+			get_ioc_tag(ioste, address),
+			map_iopt_entry(address));
+	}
+	iounmap(base);
+}
+
+
+static void *bpa_alloc_coherent(struct device *hwdev, size_t size,
+			   dma_addr_t *dma_handle, unsigned int __nocast flag)
+{
+	void *ret;
+
+	ret = (void *)__get_free_pages(flag, get_order(size));
+	if (ret != NULL) {
+		memset(ret, 0, size);
+		*dma_handle = virt_to_abs(ret) | BPA_DMA_VALID;
+	}
+	return ret;
+}
+
+static void bpa_free_coherent(struct device *hwdev, size_t size,
+				 void *vaddr, dma_addr_t dma_handle)
+{
+	free_pages((unsigned long)vaddr, get_order(size));
+}
+
+static dma_addr_t bpa_map_single(struct device *hwdev, void *ptr,
+		size_t size, enum dma_data_direction direction)
+{
+	return virt_to_abs(ptr) | BPA_DMA_VALID;
+}
+
+static void bpa_unmap_single(struct device *hwdev, dma_addr_t dma_addr,
+		size_t size, enum dma_data_direction direction)
+{
+}
+
+static int bpa_map_sg(struct device *hwdev, struct scatterlist *sg,
+		int nents, enum dma_data_direction direction)
+{
+	int i;
+
+	for (i = 0; i < nents; i++, sg++) {
+		sg->dma_address = (page_to_phys(sg->page) + sg->offset)
+					| BPA_DMA_VALID;
+		sg->dma_length = sg->length;
+	}
+
+	return nents;
+}
+
+static void bpa_unmap_sg(struct device *hwdev, struct scatterlist *sg,
+		int nents, enum dma_data_direction direction)
+{
+}
+
+static int bpa_dma_supported(struct device *dev, u64 mask)
+{
+	return mask < 0x100000000ull;
+}
+
+void bpa_init_iommu(void)
+{
+	bpa_map_iommu();
+
+	/* Direct I/O, IOMMU off */
+	ppc_md.iommu_dev_setup = iommu_dev_setup_null;
+	ppc_md.iommu_bus_setup = iommu_bus_setup_null;
+
+	pci_dma_ops.alloc_coherent = bpa_alloc_coherent;
+	pci_dma_ops.free_coherent = bpa_free_coherent;
+	pci_dma_ops.map_single = bpa_map_single;
+	pci_dma_ops.unmap_single = bpa_unmap_single;
+	pci_dma_ops.map_sg = bpa_map_sg;
+	pci_dma_ops.unmap_sg = bpa_unmap_sg;
+	pci_dma_ops.dma_supported = bpa_dma_supported;
+}
diff --git a/arch/ppc64/kernel/bpa_iommu.h b/arch/ppc64/kernel/bpa_iommu.h
new file mode 100644
index 00000000000..e547d77dfa0
--- /dev/null
+++ b/arch/ppc64/kernel/bpa_iommu.h
@@ -0,0 +1,65 @@
+#ifndef BPA_IOMMU_H
+#define BPA_IOMMU_H
+
+/* some constants */
+enum {
+	/* segment table entries */
+	IOST_VALID_MASK	  = 0x8000000000000000ul,
+	IOST_TAG_MASK     = 0x3000000000000000ul,
+	IOST_PT_BASE_MASK = 0x000003fffffff000ul,
+	IOST_NNPT_MASK	  = 0x0000000000000fe0ul,
+	IOST_PS_MASK	  = 0x000000000000000ful,
+
+	IOST_PS_4K	  = 0x1,
+	IOST_PS_64K	  = 0x3,
+	IOST_PS_1M	  = 0x5,
+	IOST_PS_16M	  = 0x7,
+
+	/* iopt tag register */
+	IOPT_VALID_MASK   = 0x0000000200000000ul,
+	IOPT_TAG_MASK	  = 0x00000001fffffffful,
+
+	/* iopt cache register */
+	IOPT_PROT_MASK	  = 0xc000000000000000ul,
+	IOPT_PROT_NONE	  = 0x0000000000000000ul,
+	IOPT_PROT_READ	  = 0x4000000000000000ul,
+	IOPT_PROT_WRITE	  = 0x8000000000000000ul,
+	IOPT_PROT_RW	  = 0xc000000000000000ul,
+	IOPT_COHERENT	  = 0x2000000000000000ul,
+	
+	IOPT_ORDER_MASK	  = 0x1800000000000000ul,
+	/* order access to same IOID/VC on same address */
+	IOPT_ORDER_ADDR	  = 0x0800000000000000ul,
+	/* similar, but only after a write access */
+	IOPT_ORDER_WRITES = 0x1000000000000000ul,
+	/* Order all accesses to same IOID/VC */
+	IOPT_ORDER_VC	  = 0x1800000000000000ul,
+	
+	IOPT_RPN_MASK	  = 0x000003fffffff000ul,
+	IOPT_HINT_MASK	  = 0x0000000000000800ul,
+	IOPT_IOID_MASK	  = 0x00000000000007fful,
+
+	IOSTO_ENABLE	  = 0x8000000000000000ul,
+	IOSTO_ORIGIN	  = 0x000003fffffff000ul,
+	IOSTO_HW	  = 0x0000000000000800ul,
+	IOSTO_SW	  = 0x0000000000000400ul,
+
+	IOCMD_CONF_TE	  = 0x0000800000000000ul,
+
+	/* memory mapped registers */
+	IOC_PT_CACHE_DIR  = 0x000,
+	IOC_ST_CACHE_DIR  = 0x800,
+	IOC_PT_CACHE_REG  = 0x910,
+	IOC_ST_ORIGIN     = 0x918,
+	IOC_CONF	  = 0x930,
+
+	/* The high bit needs to be set on every DMA address,
+	   only 2GB are addressable */
+	BPA_DMA_VALID	  = 0x80000000,
+	BPA_DMA_MASK	  = 0x7fffffff,
+};
+
+
+void bpa_init_iommu(void);
+
+#endif
diff --git a/arch/ppc64/kernel/bpa_setup.c b/arch/ppc64/kernel/bpa_setup.c
index 0a43d8a93d7..57b3db66f45 100644
--- a/arch/ppc64/kernel/bpa_setup.c
+++ b/arch/ppc64/kernel/bpa_setup.c
@@ -46,6 +46,7 @@
 
 #include "pci.h"
 #include "bpa_iic.h"
+#include "bpa_iommu.h"
 
 #ifdef DEBUG
 #define DBG(fmt...) udbg_printf(fmt)
@@ -107,7 +108,7 @@ static void __init bpa_init_early(void)
 
 	hpte_init_native();
 
-	pci_direct_iommu_init();
+	bpa_init_iommu();
 
 	ppc64_interrupt_controller = IC_BPA_IIC;
 
-- 
cgit v1.2.3-70-g09d2


From 8f586b2243198194240626fd9695da5564ffa7ee Mon Sep 17 00:00:00 2001
From: Mike Strosaker <strosake@austin.ibm.com>
Date: Thu, 23 Jun 2005 16:09:41 +1000
Subject: [PATCH] correct printing to operator panel

This patch corrects the printing of progress indicators to the op
panel on p/iSeries ppc64 systems.  Each discrete reference code should
begin with a form feed char to clear the op panel, and the first and
second lines should be separated with a CR/LF sequence.  Padding with
spaces is not necessary.

Also, capitalize the hex value printed on the first line, to be
consistent with the values printed by firmware, service processor,
etc.

It turns out that there's an ibm,form-feed property; this patch uses
it in the pSeries-specific progress routine.  This patch also checks
the number of rows and the specific width of each row (the second row
on power5 systems can actually hold 80 characters).  If the displayed
text is too wide for the physical display, it can be viewed in the ASM
menus, or by selecting option 14 on the op panel.

Signed-off-by: Mike Strosaker <strosake@austin.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/rtas.c  | 59 +++++++++++++++++++++++++++++------------------
 arch/ppc64/kernel/setup.c |  6 ++---
 2 files changed, 40 insertions(+), 25 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/rtas.c b/arch/ppc64/kernel/rtas.c
index 43e1518653d..5e8eb33b8e5 100644
--- a/arch/ppc64/kernel/rtas.c
+++ b/arch/ppc64/kernel/rtas.c
@@ -98,21 +98,29 @@ rtas_progress(char *s, unsigned short hex)
 	int width, *p;
 	char *os;
 	static int display_character, set_indicator;
-	static int max_width;
+	static int display_width, display_lines, *row_width, form_feed;
 	static DEFINE_SPINLOCK(progress_lock);
+	static int current_line;
 	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;
+	if (display_width == 0) {
+		display_width = 0x10;
+		if ((root = find_path_device("/rtas"))) {
+			if ((p = (unsigned int *)get_property(root,
+					"ibm,display-line-length", NULL)))
+				display_width = *p;
+			if ((p = (unsigned int *)get_property(root,
+					"ibm,form-feed", NULL)))
+				form_feed = *p;
+			if ((p = (unsigned int *)get_property(root,
+					"ibm,display-number-of-lines", NULL)))
+				display_lines = *p;
+			row_width = (unsigned int *)get_property(root,
+					"ibm,display-truncation-length", NULL);
+		}
 		display_character = rtas_token("display-character");
 		set_indicator = rtas_token("set-indicator");
 	}
@@ -131,31 +139,39 @@ rtas_progress(char *s, unsigned short hex)
 	 * 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 no newline is pending and form feed is supported, clear the
+	 * display with a form feed; otherwise, 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');
+		current_line = 0;
+		if (form_feed)
+			rtas_call(display_character, 1, 1, NULL,
+				  (char)form_feed);
+		else
+			rtas_call(display_character, 1, 1, NULL, '\r');
 	}
  
-	width = max_width;
+	if (row_width)
+		width = row_width[current_line];
+	else
+		width = display_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;
+				current_line++;
+				if (current_line > display_lines-1)
+					current_line = display_lines-1;
 				spin_unlock(&progress_lock);
 				return;
 			}
@@ -172,7 +188,10 @@ rtas_progress(char *s, unsigned short hex)
 				rtas_call(display_character, 1, 1, NULL, *os);
 			}
  
-			width = max_width;
+			if (row_width)
+				width = row_width[current_line];
+			else
+				width = display_width;
 		} else {
 			width--;
 			rtas_call(display_character, 1, 1, NULL, *os);
@@ -186,10 +205,6 @@ rtas_progress(char *s, unsigned short hex)
 				os++;
 	}
  
-	/* Blank to end of line. */
-	while (width-- > 0)
-		rtas_call(display_character, 1, 1, NULL, ' ');
-
 	spin_unlock(&progress_lock);
 }
 
diff --git a/arch/ppc64/kernel/setup.c b/arch/ppc64/kernel/setup.c
index 10222008fe2..8a1ca695f8a 100644
--- a/arch/ppc64/kernel/setup.c
+++ b/arch/ppc64/kernel/setup.c
@@ -1080,11 +1080,11 @@ void __init setup_arch(char **cmdline_p)
 static void ppc64_do_msg(unsigned int src, const char *msg)
 {
 	if (ppc_md.progress) {
-		char buf[32];
+		char buf[128];
 
-		sprintf(buf, "%08x        \n", src);
+		sprintf(buf, "%08X\n", src);
 		ppc_md.progress(buf, 0);
-		sprintf(buf, "%-16s", msg);
+		snprintf(buf, 128, "%s", msg);
 		ppc_md.progress(buf, 0);
 	}
 }
-- 
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 'arch')

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 d7152fe14cad075d6dd4ee4194acd131aed0244e Mon Sep 17 00:00:00 2001
From: David Gibson <david@gibson.dropbear.id.au>
Date: Thu, 23 Jun 2005 17:14:39 +1000
Subject: [PATCH] Maple powerdown patch

Currently reset and powerdown are not implemented on the Maple board,
and attempting to do so will (incorrectly return).  This implements
the proper communication with the service processor, allowing correct
reset and powerdown on the Maple board, by communicating with the
service processor.  If somehow it's unable to communicate with the
service processor it will loop forever instead.

Note that powerdown on the Maple will power down the CPUs, but not the
fans or other board components due to hardware and firmware
limitations.

Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Frank Rowand <frowand@mvista.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/maple_setup.c | 62 ++++++++++++++++++++++++++++++++++++++++-
 arch/ppc64/kernel/setup.c       | 18 ++++++++++++
 2 files changed, 79 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/maple_setup.c b/arch/ppc64/kernel/maple_setup.c
index 3dabedb8553..da8900b51f4 100644
--- a/arch/ppc64/kernel/maple_setup.c
+++ b/arch/ppc64/kernel/maple_setup.c
@@ -78,17 +78,77 @@ extern int maple_pci_get_legacy_ide_irq(struct pci_dev *dev, int channel);
 extern void generic_find_legacy_serial_ports(u64 *physport,
 		unsigned int *default_speed);
 
-
 static void maple_restart(char *cmd)
 {
+	unsigned int maple_nvram_base;
+	unsigned int maple_nvram_offset;
+	unsigned int maple_nvram_command;
+	struct device_node *rtcs;
+
+	/* find NVRAM device */
+	rtcs = find_compatible_devices("nvram", "AMD8111");
+	if (rtcs && rtcs->addrs) {
+		maple_nvram_base = rtcs->addrs[0].address;
+	} else {
+		printk(KERN_EMERG "Maple: Unable to find NVRAM\n");
+		printk(KERN_EMERG "Maple: Manual Restart Required\n");
+		return;
+	}
+
+	/* find service processor device */
+	rtcs = find_devices("service-processor");
+	if (!rtcs) {
+		printk(KERN_EMERG "Maple: Unable to find Service Processor\n");
+		printk(KERN_EMERG "Maple: Manual Restart Required\n");
+		return;
+	}
+	maple_nvram_offset = *(unsigned int*) get_property(rtcs,
+			"restart-addr", NULL);
+	maple_nvram_command = *(unsigned int*) get_property(rtcs,
+			"restart-value", NULL);
+
+	/* send command */
+	outb_p(maple_nvram_command, maple_nvram_base + maple_nvram_offset);
+	for (;;) ;
 }
 
 static void maple_power_off(void)
 {
+	unsigned int maple_nvram_base;
+	unsigned int maple_nvram_offset;
+	unsigned int maple_nvram_command;
+	struct device_node *rtcs;
+
+	/* find NVRAM device */
+	rtcs = find_compatible_devices("nvram", "AMD8111");
+	if (rtcs && rtcs->addrs) {
+		maple_nvram_base = rtcs->addrs[0].address;
+	} else {
+		printk(KERN_EMERG "Maple: Unable to find NVRAM\n");
+		printk(KERN_EMERG "Maple: Manual Power-Down Required\n");
+		return;
+	}
+
+	/* find service processor device */
+	rtcs = find_devices("service-processor");
+	if (!rtcs) {
+		printk(KERN_EMERG "Maple: Unable to find Service Processor\n");
+		printk(KERN_EMERG "Maple: Manual Power-Down Required\n");
+		return;
+	}
+	maple_nvram_offset = *(unsigned int*) get_property(rtcs,
+			"power-off-addr", NULL);
+	maple_nvram_command = *(unsigned int*) get_property(rtcs,
+			"power-off-value", NULL);
+
+	/* send command */
+	outb_p(maple_nvram_command, maple_nvram_base + maple_nvram_offset);
+	for (;;) ;
 }
 
 static void maple_halt(void)
 {
+	maple_power_off();
 }
 
 #ifdef CONFIG_SMP
diff --git a/arch/ppc64/kernel/setup.c b/arch/ppc64/kernel/setup.c
index 8a1ca695f8a..7d060ddb5e9 100644
--- a/arch/ppc64/kernel/setup.c
+++ b/arch/ppc64/kernel/setup.c
@@ -683,6 +683,12 @@ void machine_restart(char *cmd)
 	if (ppc_md.nvram_sync)
 		ppc_md.nvram_sync();
 	ppc_md.restart(cmd);
+#ifdef CONFIG_SMP
+	smp_send_stop();
+#endif
+	printk(KERN_EMERG "System Halted, OK to turn off power\n");
+	local_irq_disable();
+	while (1) ;
 }
 
 EXPORT_SYMBOL(machine_restart);
@@ -692,6 +698,12 @@ void machine_power_off(void)
 	if (ppc_md.nvram_sync)
 		ppc_md.nvram_sync();
 	ppc_md.power_off();
+#ifdef CONFIG_SMP
+	smp_send_stop();
+#endif
+	printk(KERN_EMERG "System Halted, OK to turn off power\n");
+	local_irq_disable();
+	while (1) ;
 }
 
 EXPORT_SYMBOL(machine_power_off);
@@ -701,6 +713,12 @@ void machine_halt(void)
 	if (ppc_md.nvram_sync)
 		ppc_md.nvram_sync();
 	ppc_md.halt();
+#ifdef CONFIG_SMP
+	smp_send_stop();
+#endif
+	printk(KERN_EMERG "System Halted, OK to turn off power\n");
+	local_irq_disable();
+	while (1) ;
 }
 
 EXPORT_SYMBOL(machine_halt);
-- 
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 'arch')

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 c2ebaa425e6630adcbf757b004d257dd4204925b Mon Sep 17 00:00:00 2001
From: Dave Hansen <haveblue@us.ibm.com>
Date: Thu, 23 Jun 2005 00:07:38 -0700
Subject: [PATCH] sparsemem base: early_pfn_to_nid() (works before sparse is
 initialized)

The following four patches provide the last needed changes before the
introduction of sparsemem.  For a more complete description of what this
will do, please see this patch:

http://www.sr71.net/patches/2.6.11/2.6.11-bk7-mhp1/broken-out/B-sparse-150-sparsemem.patch

or previous posts on the subject:
http://marc.theaimsgroup.com/?t=110868540700001&r=1&w=2
http://marc.theaimsgroup.com/?l=linux-mm&m=109897373315016&w=2

Three of these are i386-only, but one of them reorganizes the macros
used to manage the space in page->flags, and will affect all platforms.
There are analogous patches to the i386 ones for ppc64, ia64, and
x86_64, but those will be submitted by the normal arch maintainers.

The combination of the four patches has been test-booted on a variety of
i386 hardware, and compiled for ppc64, i386, and x86-64 with about 17
different .configs.  It's also been runtime-tested on ia64 configs (with
more patches on top).

This patch:

We _know_ which node pages in general belong to, at least at a very gross
level in node_{start,end}_pfn[].  Use those to target the allocations of
pages.

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/mm/discontig.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

(limited to 'arch')

diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c
index 1726b4096b1..85d2fcbe107 100644
--- a/arch/i386/mm/discontig.c
+++ b/arch/i386/mm/discontig.c
@@ -146,6 +146,21 @@ static void __init find_max_pfn_node(int nid)
 		BUG();
 }
 
+/* Find the owning node for a pfn. */
+int early_pfn_to_nid(unsigned long pfn)
+{
+	int nid;
+
+	for_each_node(nid) {
+		if (node_end_pfn[nid] == 0)
+			break;
+		if (node_start_pfn[nid] <= pfn && node_end_pfn[nid] >= pfn)
+			return nid;
+	}
+
+	return 0;
+}
+
 /* 
  * Allocate memory for the pg_data_t for this node via a crude pre-bootmem
  * method.  For node zero take this from the bottom of memory, for
-- 
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 'arch')

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 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 'arch')

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 3f22ab276b931b72ea04b184c155b34d0362bfc3 Mon Sep 17 00:00:00 2001
From: Dave Hansen <haveblue@us.ibm.com>
Date: Thu, 23 Jun 2005 00:07:43 -0700
Subject: [PATCH] make each arch use mm/Kconfig

For all architectures, this just means that you'll see a "Memory Model"
choice in your architecture menu.  For those that implement DISCONTIGMEM,
you may eventually want to make your ARCH_DISCONTIGMEM_ENABLE a "def_bool
y" and make your users select DISCONTIGMEM right out of the new choice
menu.  The only disadvantage might be if you have some specific things that
you need in your help option to explain something about DISCONTIGMEM.

Signed-off-by: Dave Hansen <haveblue@us.ibm.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/alpha/Kconfig     | 4 +++-
 arch/arm/Kconfig       | 4 +++-
 arch/arm26/Kconfig     | 2 ++
 arch/cris/Kconfig      | 2 ++
 arch/frv/Kconfig       | 2 ++
 arch/h8300/Kconfig.cpu | 3 +++
 arch/i386/Kconfig      | 4 +++-
 arch/ia64/Kconfig      | 4 +++-
 arch/m32r/Kconfig      | 4 +++-
 arch/m68k/Kconfig      | 2 ++
 arch/m68knommu/Kconfig | 2 ++
 arch/mips/Kconfig      | 6 +++++-
 arch/parisc/Kconfig    | 8 +++++++-
 arch/ppc/Kconfig       | 2 ++
 arch/ppc64/Kconfig     | 4 +++-
 arch/s390/Kconfig      | 2 ++
 arch/sh/Kconfig        | 8 +++++++-
 arch/sh64/Kconfig      | 2 ++
 arch/sparc/Kconfig     | 2 ++
 arch/sparc64/Kconfig   | 2 ++
 arch/um/Kconfig        | 1 +
 arch/v850/Kconfig      | 2 ++
 arch/x86_64/Kconfig    | 4 +++-
 23 files changed, 66 insertions(+), 10 deletions(-)

(limited to 'arch')

diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index f7c96635d3b..c5739d6309d 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -509,7 +509,7 @@ config NR_CPUS
 	depends on SMP
 	default "64"
 
-config DISCONTIGMEM
+config ARCH_DISCONTIGMEM_ENABLE
 	bool "Discontiguous Memory Support (EXPERIMENTAL)"
 	depends on EXPERIMENTAL
 	help
@@ -518,6 +518,8 @@ config DISCONTIGMEM
 	  or have huge holes in the physical address space for other reasons.
 	  See <file:Documentation/vm/numa> for more.
 
+source "mm/Kconfig"
+
 config NUMA
 	bool "NUMA Support (EXPERIMENTAL)"
 	depends on DISCONTIGMEM
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index ee8a9ad7bbd..07ba77c19f6 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -346,7 +346,7 @@ 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 DISCONTIGMEM
+config ARCH_DISCONTIGMEM_ENABLE
 	bool
 	default (ARCH_LH7A40X && !LH7A40X_CONTIGMEM)
 	help
@@ -355,6 +355,8 @@ config DISCONTIGMEM
 	  or have huge holes in the physical address space for other reasons.
 	  See <file:Documentation/vm/numa> for more.
 
+source "mm/Kconfig"
+
 config LEDS
 	bool "Timer and CPU usage LEDs"
 	depends on ARCH_CDB89712 || ARCH_CO285 || ARCH_EBSA110 || \
diff --git a/arch/arm26/Kconfig b/arch/arm26/Kconfig
index 6caed90661f..dc0c1936969 100644
--- a/arch/arm26/Kconfig
+++ b/arch/arm26/Kconfig
@@ -179,6 +179,8 @@ config CMDLINE
 	  time by entering them here. As a minimum, you should specify the
 	  memory size and the root device (e.g., mem=64M root=/dev/nfs).
 
+source "mm/Kconfig"
+
 endmenu
 
 source "drivers/base/Kconfig"
diff --git a/arch/cris/Kconfig b/arch/cris/Kconfig
index 4332ca348d5..f848e376149 100644
--- a/arch/cris/Kconfig
+++ b/arch/cris/Kconfig
@@ -74,6 +74,8 @@ 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.
 
+source mm/Kconfig
+
 endmenu
 
 menu "Hardware setup"
diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig
index 2b19372767e..c93f95146cc 100644
--- a/arch/frv/Kconfig
+++ b/arch/frv/Kconfig
@@ -74,6 +74,8 @@ config HIGHPTE
 	  with a lot of RAM, this can be wasteful of precious low memory.
 	  Setting this option will put user-space page tables in high memory.
 
+source "mm/Kconfig"
+
 choice
 	prompt "uClinux kernel load address"
 	depends on !MMU
diff --git a/arch/h8300/Kconfig.cpu b/arch/h8300/Kconfig.cpu
index d9dd62a565a..a380167a13c 100644
--- a/arch/h8300/Kconfig.cpu
+++ b/arch/h8300/Kconfig.cpu
@@ -180,4 +180,7 @@ config CPU_H8S
 config PREEMPT
 	bool "Preemptible Kernel"
 	default n
+
+source "mm/Kconfig"
+
 endmenu
diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index 35ca3a17ed2..8e5242c8e09 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -783,7 +783,7 @@ 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 DISCONTIGMEM
+config ARCH_DISCONTIGMEM_ENABLE
 	bool
 	depends on NUMA
 	default y
@@ -808,6 +808,8 @@ config HAVE_ARCH_ALLOC_REMAP
 	depends on NUMA
 	default y
 
+source "mm/Kconfig"
+
 config HIGHPTE
 	bool "Allocate 3rd-level pagetables from highmem"
 	depends on HIGHMEM4G || HIGHMEM64G
diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index ce4dfa8b834..9f6a46caba9 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -197,7 +197,7 @@ config HOLES_IN_ZONE
 	bool
 	default y if VIRTUAL_MEM_MAP
 
-config DISCONTIGMEM
+config ARCH_DISCONTIGMEM_ENABLE
 	bool "Discontiguous memory support"
 	depends on (IA64_DIG || IA64_SGI_SN2 || IA64_GENERIC || IA64_HP_ZX1 || IA64_HP_ZX1_SWIOTLB) && NUMA && VIRTUAL_MEM_MAP
 	default y if (IA64_SGI_SN2 || IA64_GENERIC) && NUMA
@@ -300,6 +300,8 @@ 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.
 
+source "mm/Kconfig"
+
 config HAVE_DEC_LOCK
 	bool
 	depends on (SMP || PREEMPT)
diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig
index 64c133344af..42ca8a39798 100644
--- a/arch/m32r/Kconfig
+++ b/arch/m32r/Kconfig
@@ -172,11 +172,13 @@ config NOHIGHMEM
 	bool
 	default y
 
-config DISCONTIGMEM
+config ARCH_DISCONTIGMEM_ENABLE
 	bool "Internal RAM Support"
 	depends on CHIP_M32700 || CHIP_M32102 || CHIP_VDEC2 || CHIP_OPSP
 	default y
 
+source "mm/Kconfig"
+
 config IRAM_START
 	hex "Internal memory start address (hex)"
 	default "00f00000"
diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig
index d0713c7d9f0..691a2469ff3 100644
--- a/arch/m68k/Kconfig
+++ b/arch/m68k/Kconfig
@@ -357,6 +357,8 @@ config 060_WRITETHROUGH
 	  is hardwired on.  The 53c710 SCSI driver is known to suffer from
 	  this problem.
 
+source "mm/Kconfig"
+
 endmenu
 
 menu "General setup"
diff --git a/arch/m68knommu/Kconfig b/arch/m68knommu/Kconfig
index e729bd28062..dbfcdc8e608 100644
--- a/arch/m68knommu/Kconfig
+++ b/arch/m68knommu/Kconfig
@@ -532,6 +532,8 @@ config ROMKERNEL
 
 endchoice
 
+source "mm/Kconfig"
+
 endmenu
 
 config ISA_DMA_API
diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index ab9944693f1..2710018ac80 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -492,7 +492,7 @@ config SGI_SN0_N_MODE
 	  which allows for more memory.  Your system is most probably
 	  running in M-Mode, so you should say N here.
 
-config DISCONTIGMEM
+config ARCH_DISCONTIGMEM_ENABLE
 	bool
 	default y if SGI_IP27
 	help
@@ -501,6 +501,10 @@ config DISCONTIGMEM
 	  or have huge holes in the physical address space for other reasons.
 	  See <file:Documentation/vm/numa> for more.
 
+config ARCH_FLATMEM_DISABLE
+	def_bool y
+	depends on ARCH_DISCONTIGMEM_ENABLE
+
 config NUMA
 	bool "NUMA Support"
 	depends on SGI_IP27
diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index e7e7c56fc21..ab7cc4eb7fc 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -148,7 +148,7 @@ config HOTPLUG_CPU
 	default y if SMP
 	select HOTPLUG
 
-config DISCONTIGMEM
+config ARCH_DISCONTIGMEM_ENABLE
 	bool "Discontiguous memory support (EXPERIMENTAL)"
 	depends on EXPERIMENTAL
 	help
@@ -157,6 +157,12 @@ config DISCONTIGMEM
 	  or have huge holes in the physical address space for other reasons.
 	  See <file:Documentation/vm/numa> for more.
 
+config ARCH_FLATMEM_DISABLE
+	def_bool y
+	depends on ARCH_DISCONTIGMEM_ENABLE
+
+source "mm/Kconfig"
+
 config PREEMPT
 	bool
 #	bool "Preemptible Kernel"
diff --git a/arch/ppc/Kconfig b/arch/ppc/Kconfig
index 10162b187bc..848f43970a4 100644
--- a/arch/ppc/Kconfig
+++ b/arch/ppc/Kconfig
@@ -905,6 +905,8 @@ config PREEMPT
 config HIGHMEM
 	bool "High memory support"
 
+source "mm/Kconfig"
+
 source "fs/Kconfig.binfmt"
 
 config PROC_DEVICETREE
diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig
index 0f1fa289744..87344f32423 100644
--- a/arch/ppc64/Kconfig
+++ b/arch/ppc64/Kconfig
@@ -198,10 +198,12 @@ config HMT
 	  This option enables hardware multithreading on RS64 cpus.
 	  pSeries systems p620 and p660 have such a cpu type.
 
-config DISCONTIGMEM
+config ARCH_DISCONTIGMEM_ENABLE
 	bool "Discontiguous Memory Support"
 	depends on SMP && PPC_PSERIES
 
+source "mm/Kconfig"
+
 config NUMA
 	bool "NUMA support"
 	depends on DISCONTIGMEM
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index ab79af84699..32696c1d928 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -226,6 +226,8 @@ config WARN_STACK_SIZE
 	  This allows you to specify the maximum frame size a function may
 	  have without the compiler complaining about it.
 
+source "mm/Kconfig"
+
 comment "I/O subsystem configuration"
 
 config MACHCHK_WARNING
diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 3468d512722..ac2b865359b 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -486,7 +486,7 @@ config CPU_SUBTYPE_ST40
        depends on CPU_SUBTYPE_ST40STB1 || CPU_SUBTYPE_ST40GX1
        default y
 
-config DISCONTIGMEM
+config ARCH_DISCONTIGMEM_ENABLE
 	bool
 	depends on SH_HP690
 	default y
@@ -496,6 +496,12 @@ config DISCONTIGMEM
 	  or have huge holes in the physical address space for other reasons.
 	  See <file:Documentation/vm/numa> for more.
 
+config ARCH_FLATMEM_DISABLE
+	def_bool y
+	depends on ARCH_DISCONTIGMEM_ENABLE
+
+source "mm/Kconfig"
+
 config ZERO_PAGE_OFFSET
 	hex "Zero page offset"
 	default "0x00001000" if !(SH_MPC1211 || SH_SH03)
diff --git a/arch/sh64/Kconfig b/arch/sh64/Kconfig
index 76eb81fba45..708e59736a4 100644
--- a/arch/sh64/Kconfig
+++ b/arch/sh64/Kconfig
@@ -217,6 +217,8 @@ config PREEMPT
 	bool "Preemptible Kernel (EXPERIMENTAL)"
 	depends on EXPERIMENTAL
 
+source "mm/Kconfig"
+
 endmenu
 
 menu "Bus options (PCI, PCMCIA, EISA, MCA, ISA)"
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 237f922520f..310d3f0c445 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -291,6 +291,8 @@ config PRINTER
 	  If you have more than 8 printers, you need to increase the LP_NO
 	  macro in lp.c and the PARPORT_MAX macro in parport.h.
 
+source "mm/Kconfig"
+
 endmenu
 
 source "drivers/base/Kconfig"
diff --git a/arch/sparc64/Kconfig b/arch/sparc64/Kconfig
index a72fd15d5ea..e2b050eb3b9 100644
--- a/arch/sparc64/Kconfig
+++ b/arch/sparc64/Kconfig
@@ -484,6 +484,8 @@ config CMDLINE
 
 	  NOTE: This option WILL override the PROM bootargs setting!
 
+source "mm/Kconfig"
+
 endmenu
 
 source "drivers/base/Kconfig"
diff --git a/arch/um/Kconfig b/arch/um/Kconfig
index b8e952c88fd..9469e77303e 100644
--- a/arch/um/Kconfig
+++ b/arch/um/Kconfig
@@ -74,6 +74,7 @@ config MODE_SKAS
 	option will shrink the UML binary slightly.
 
 source "arch/um/Kconfig_arch"
+source "mm/Kconfig"
 
 config LD_SCRIPT_STATIC
 	bool
diff --git a/arch/v850/Kconfig b/arch/v850/Kconfig
index 90cd4baa75e..27febd6ffa8 100644
--- a/arch/v850/Kconfig
+++ b/arch/v850/Kconfig
@@ -218,6 +218,8 @@ menu "Processor type and features"
 	     a lot of RAM, and you need to able to allocate very large
 	     contiguous chunks. If unsure, say N.
 
+source "mm/Kconfig"
+
 endmenu
 
 
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 289f448ac89..9f5b99e7b1e 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -265,7 +265,7 @@ config NUMA_EMU
 	  into virtual nodes when booted with "numa=fake=N", where N is the
 	  number of nodes. This is only useful for debugging.
 
-config DISCONTIGMEM
+config ARCH_DISCONTIGMEM_ENABLE
        bool
        depends on NUMA
        default y
@@ -274,6 +274,8 @@ config NUMA
        bool
        default n
 
+source "mm/Kconfig"
+
 config HAVE_DEC_LOCK
 	bool
 	depends on SMP
-- 
cgit v1.2.3-70-g09d2


From 0e19243e9a19ef8e5994852671bd06bb51630811 Mon Sep 17 00:00:00 2001
From: Dave Hansen <haveblue@us.ibm.com>
Date: Thu, 23 Jun 2005 00:07:45 -0700
Subject: [PATCH] update all defconfigs for ARCH_DISCONTIGMEM_ENABLE

This will at least suppress one prompt that users would have received the
first time they compile with the new DISCONTIG arch option.  They'll still
get the "Memory Model" prompt, but 99% of them will have the default work
there.

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/alpha/defconfig                 | 2 +-
 arch/ia64/configs/sn2_defconfig      | 2 +-
 arch/ia64/defconfig                  | 2 +-
 arch/mips/configs/ip27_defconfig     | 2 +-
 arch/ppc64/configs/pSeries_defconfig | 2 +-
 arch/ppc64/defconfig                 | 2 +-
 6 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'arch')

diff --git a/arch/alpha/defconfig b/arch/alpha/defconfig
index 5e39b7a7c8f..6da9c3dbde4 100644
--- a/arch/alpha/defconfig
+++ b/arch/alpha/defconfig
@@ -96,7 +96,7 @@ CONFIG_ALPHA_CORE_AGP=y
 CONFIG_ALPHA_BROKEN_IRQ_MASK=y
 CONFIG_EISA=y
 # CONFIG_SMP is not set
-# CONFIG_DISCONTIGMEM is not set
+# CONFIG_ARCH_DISCONTIGMEM_ENABLE is not set
 CONFIG_VERBOSE_MCHECK=y
 CONFIG_VERBOSE_MCHECK_ON=1
 CONFIG_PCI_LEGACY_PROC=y
diff --git a/arch/ia64/configs/sn2_defconfig b/arch/ia64/configs/sn2_defconfig
index a01bb02d074..487d2e36b0a 100644
--- a/arch/ia64/configs/sn2_defconfig
+++ b/arch/ia64/configs/sn2_defconfig
@@ -78,7 +78,7 @@ CONFIG_IA64_L1_CACHE_SHIFT=7
 CONFIG_NUMA=y
 CONFIG_VIRTUAL_MEM_MAP=y
 CONFIG_HOLES_IN_ZONE=y
-CONFIG_DISCONTIGMEM=y
+CONFIG_ARCH_DISCONTIGMEM_ENABLE=y
 # CONFIG_IA64_CYCLONE is not set
 CONFIG_IOSAPIC=y
 CONFIG_IA64_SGI_SN_SIM=y
diff --git a/arch/ia64/defconfig b/arch/ia64/defconfig
index 7be8096e056..8444add7638 100644
--- a/arch/ia64/defconfig
+++ b/arch/ia64/defconfig
@@ -84,7 +84,7 @@ CONFIG_IA64_L1_CACHE_SHIFT=7
 CONFIG_NUMA=y
 CONFIG_VIRTUAL_MEM_MAP=y
 CONFIG_HOLES_IN_ZONE=y
-CONFIG_DISCONTIGMEM=y
+CONFIG_ARCH_DISCONTIGMEM_ENABLE=y
 CONFIG_IA64_CYCLONE=y
 CONFIG_IOSAPIC=y
 CONFIG_FORCE_MAX_ZONEORDER=18
diff --git a/arch/mips/configs/ip27_defconfig b/arch/mips/configs/ip27_defconfig
index 13472292d0e..b5bab3a42fc 100644
--- a/arch/mips/configs/ip27_defconfig
+++ b/arch/mips/configs/ip27_defconfig
@@ -82,7 +82,7 @@ CONFIG_STOP_MACHINE=y
 # CONFIG_SGI_IP22 is not set
 CONFIG_SGI_IP27=y
 # CONFIG_SGI_SN0_N_MODE is not set
-CONFIG_DISCONTIGMEM=y
+CONFIG_ARCH_DISCONTIGMEM_ENABLE=y
 CONFIG_NUMA=y
 # CONFIG_MAPPED_KERNEL is not set
 # CONFIG_REPLICATE_KTEXT is not set
diff --git a/arch/ppc64/configs/pSeries_defconfig b/arch/ppc64/configs/pSeries_defconfig
index 3eb5ef25d3a..d0db8b5966c 100644
--- a/arch/ppc64/configs/pSeries_defconfig
+++ b/arch/ppc64/configs/pSeries_defconfig
@@ -88,7 +88,7 @@ CONFIG_IBMVIO=y
 CONFIG_IOMMU_VMERGE=y
 CONFIG_SMP=y
 CONFIG_NR_CPUS=128
-CONFIG_DISCONTIGMEM=y
+CONFIG_ARCH_DISCONTIGMEM_ENABLE=y
 CONFIG_NUMA=y
 CONFIG_SCHED_SMT=y
 # CONFIG_PREEMPT is not set
diff --git a/arch/ppc64/defconfig b/arch/ppc64/defconfig
index 2f31bf3046f..b8e2066dde7 100644
--- a/arch/ppc64/defconfig
+++ b/arch/ppc64/defconfig
@@ -89,7 +89,7 @@ CONFIG_BOOTX_TEXT=y
 CONFIG_IOMMU_VMERGE=y
 CONFIG_SMP=y
 CONFIG_NR_CPUS=32
-CONFIG_DISCONTIGMEM=y
+CONFIG_ARCH_DISCONTIGMEM_ENABLE=y
 # CONFIG_NUMA is not set
 # CONFIG_SCHED_SMT is not set
 # CONFIG_PREEMPT is not set
-- 
cgit v1.2.3-70-g09d2


From 074ccf8016b61f4b40066f8d737ab31e17a6afd1 Mon Sep 17 00:00:00 2001
From: Dave Hansen <haveblue@us.ibm.com>
Date: Thu, 23 Jun 2005 00:07:48 -0700
Subject: [PATCH] mm/Kconfig: kill unused ARCH_FLATMEM_DISABLE

This used to be used to disable FLATMEM selection, but I decided to change it
to be done generically when DISCONTIG is enabled.  The option is unused, so
this kills it.

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/mips/Kconfig   | 4 ----
 arch/parisc/Kconfig | 4 ----
 arch/sh/Kconfig     | 4 ----
 3 files changed, 12 deletions(-)

(limited to 'arch')

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 2710018ac80..94f5a8eb2c2 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -501,10 +501,6 @@ config ARCH_DISCONTIGMEM_ENABLE
 	  or have huge holes in the physical address space for other reasons.
 	  See <file:Documentation/vm/numa> for more.
 
-config ARCH_FLATMEM_DISABLE
-	def_bool y
-	depends on ARCH_DISCONTIGMEM_ENABLE
-
 config NUMA
 	bool "NUMA Support"
 	depends on SGI_IP27
diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index ab7cc4eb7fc..ce327c799b4 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -157,10 +157,6 @@ config ARCH_DISCONTIGMEM_ENABLE
 	  or have huge holes in the physical address space for other reasons.
 	  See <file:Documentation/vm/numa> for more.
 
-config ARCH_FLATMEM_DISABLE
-	def_bool y
-	depends on ARCH_DISCONTIGMEM_ENABLE
-
 source "mm/Kconfig"
 
 config PREEMPT
diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index ac2b865359b..a7c8bfc1160 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -496,10 +496,6 @@ config ARCH_DISCONTIGMEM_ENABLE
 	  or have huge holes in the physical address space for other reasons.
 	  See <file:Documentation/vm/numa> for more.
 
-config ARCH_FLATMEM_DISABLE
-	def_bool y
-	depends on ARCH_DISCONTIGMEM_ENABLE
-
 source "mm/Kconfig"
 
 config ZERO_PAGE_OFFSET
-- 
cgit v1.2.3-70-g09d2


From 368a0a3afad08069b2679ecaff80fc18c10a6e2a Mon Sep 17 00:00:00 2001
From: Mike Kravetz <kravetz@us.ibm.com>
Date: Thu, 23 Jun 2005 00:07:51 -0700
Subject: [PATCH] ppc64: Kconfig memory models

This patch changes some of the default behavior in the ppc64 Kconfig file
that was recently changed/added to 2.6.12-rc2-mm1 by Dave Hansen in
preparation for SPARSEMEM.  Patch allows the display of both FLAT and
DISCONTIG models on pseries.  As before, default is DISCONTIG for SMP and
PSERIES and FLAT for others.

Signed-off-by: Mike Kravetz <kravetz@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/Kconfig | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig
index 87344f32423..011b5c0bf1d 100644
--- a/arch/ppc64/Kconfig
+++ b/arch/ppc64/Kconfig
@@ -199,9 +199,16 @@ config HMT
 	  pSeries systems p620 and p660 have such a cpu type.
 
 config ARCH_DISCONTIGMEM_ENABLE
-	bool "Discontiguous Memory Support"
+	def_bool y
 	depends on SMP && PPC_PSERIES
 
+config ARCH_DISCONTIGMEM_DEFAULT
+	def_bool y
+	depends on ARCH_DISCONTIGMEM_ENABLE
+
+config ARCH_FLATMEM_ENABLE
+	def_bool y
+
 source "mm/Kconfig"
 
 config NUMA
-- 
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 'arch')

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 af705362ab6018071310c5fcd436a6b457517d5f Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Thu, 23 Jun 2005 00:07:53 -0700
Subject: [PATCH] generify memory present

Allow architectures to indicate that they will be providing hooks to indice
installed memory areas, memory_present().  Provide prototypes for the i386
implementation.

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 | 2 +-
 mm/Kconfig        | 4 ++++
 2 files changed, 5 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index a8128f99733..3b7248126d2 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -793,7 +793,7 @@ config HAVE_ARCH_BOOTMEM_NODE
 	depends on NUMA
 	default y
 
-config HAVE_MEMORY_PRESENT
+config ARCH_HAVE_MEMORY_PRESENT
 	bool
 	depends on DISCONTIGMEM
 	default y
diff --git a/mm/Kconfig b/mm/Kconfig
index 8283685f3bb..5127441561b 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -53,3 +53,7 @@ config FLATMEM
 config NEED_MULTIPLE_NODES
 	def_bool y
 	depends on DISCONTIGMEM || NUMA
+
+config HAVE_MEMORY_PRESENT
+	def_bool y
+	depends on ARCH_HAVE_MEMORY_PRESENT
-- 
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 'arch')

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 'arch')

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 'arch')

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 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 'arch')

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 74b30be2e183bd9a12d0350698996e3d1969f290 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Thu, 23 Jun 2005 00:08:02 -0700
Subject: [PATCH] ppc64: add memory present

Provide hooks for PPC64 to allow memory models to be informed of installed
memory areas.  This allows SPARSEMEM to instantiate mem_map for the populated
areas.

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 ++--
 arch/ppc64/mm/numa.c | 3 +++
 2 files changed, 5 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig
index 3c8bfd0345c..c92d48fe06e 100644
--- a/arch/ppc64/Kconfig
+++ b/arch/ppc64/Kconfig
@@ -212,8 +212,8 @@ config ARCH_FLATMEM_ENABLE
 source "mm/Kconfig"
 
 config HAVE_ARCH_EARLY_PFN_TO_NID
-	bool
-	default y
+	def_bool y
+	depends on NEED_MULTIPLE_NODES
 
 # Some NUMA nodes have memory ranges that span
 # other nodes.  Even though a pfn is valid and
diff --git a/arch/ppc64/mm/numa.c b/arch/ppc64/mm/numa.c
index ea862ec643d..cafd91aef28 100644
--- a/arch/ppc64/mm/numa.c
+++ b/arch/ppc64/mm/numa.c
@@ -440,6 +440,8 @@ new_range:
 		for (i = start ; i < (start+size); i += MEMORY_INCREMENT)
 			numa_memory_lookup_table[i >> MEMORY_INCREMENT_SHIFT] =
 				numa_domain;
+		memory_present(numa_domain, start >> PAGE_SHIFT,
+						(start + size) >> PAGE_SHIFT);
 
 		if (--ranges)
 			goto new_range;
@@ -481,6 +483,7 @@ static void __init setup_nonnuma(void)
 
 	for (i = 0 ; i < top_of_ram; i += MEMORY_INCREMENT)
 		numa_memory_lookup_table[i >> MEMORY_INCREMENT_SHIFT] = 0;
+	memory_present(0, 0, init_node_data[0].node_end_pfn);
 }
 
 static void __init dump_numa_topology(void)
-- 
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 'arch')

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 073326634bec2f36e165583d21fb0a9fad47ac0a Mon Sep 17 00:00:00 2001
From: Matt Tolentino <metolent@snoqualmie.dp.intel.com>
Date: Thu, 23 Jun 2005 00:08:03 -0700
Subject: [PATCH] remove direct ref to contig_page_data for x86-64

This patch pulls out all remaining direct references to contig_page_data
from arch/x86-64, thus saving an ifdef in one case.

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/aperture.c | 4 ----
 arch/x86_64/kernel/setup.c    | 2 +-
 2 files changed, 1 insertion(+), 5 deletions(-)

(limited to 'arch')

diff --git a/arch/x86_64/kernel/aperture.c b/arch/x86_64/kernel/aperture.c
index 504e6347499..c9a6b812e92 100644
--- a/arch/x86_64/kernel/aperture.c
+++ b/arch/x86_64/kernel/aperture.c
@@ -40,11 +40,7 @@ int fix_aperture __initdata = 1;
 
 static u32 __init allocate_aperture(void) 
 {
-#ifdef CONFIG_DISCONTIGMEM
 	pg_data_t *nd0 = NODE_DATA(0);
-#else
-	pg_data_t *nd0 = &contig_page_data;
-#endif	
 	u32 aper_size;
 	void *p; 
 
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 99f038ede23..125a0bb1f5d 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -386,7 +386,7 @@ static void __init contig_initmem_init(void)
         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(&contig_page_data, 0, end_pfn << PAGE_SHIFT); 
+        e820_bootmem_free(NODE_DATA(0), 0, end_pfn << PAGE_SHIFT);
         reserve_bootmem(bootmap, bootmap_size);
 } 
 #endif
-- 
cgit v1.2.3-70-g09d2


From 1035faf1b19efb83d5626985240f52cd149dd39b Mon Sep 17 00:00:00 2001
From: Matt Tolentino <metolent@snoqualmie.dp.intel.com>
Date: Thu, 23 Jun 2005 00:08:05 -0700
Subject: [PATCH] add x86-64 Kconfig options for sparsemem

Add the requisite arch specific Kconfig options to enable the use of the
sparsemem implementation for NUMA kernels on x86-64.

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/Kconfig | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

(limited to 'arch')

diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 9f5b99e7b1e..61ed1665234 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -274,8 +274,27 @@ config NUMA
        bool
        default n
 
+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_FLATMEM_ENABLE
+	def_bool y
+	depends on !NUMA
+
 source "mm/Kconfig"
 
+config HAVE_ARCH_EARLY_PFN_TO_NID
+	def_bool y
+
 config HAVE_DEC_LOCK
 	bool
 	depends on SMP
-- 
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 'arch')

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 'arch')

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 6f4e1e5061c44a93337338af4bf9bed10ee9f32e Mon Sep 17 00:00:00 2001
From: "Martin J. Bligh" <mbligh@mbligh.org>
Date: Thu, 23 Jun 2005 00:08:08 -0700
Subject: [PATCH] add page_state info to show_mem

This helps a lot when debugging out of memory stuff - useful especially to
see if all the memory is sucked into slab, etc.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mm/pgtable.c | 8 ++++++++
 1 file changed, 8 insertions(+)

(limited to 'arch')

diff --git a/arch/i386/mm/pgtable.c b/arch/i386/mm/pgtable.c
index 80c84cdf22e..270c59f099a 100644
--- a/arch/i386/mm/pgtable.c
+++ b/arch/i386/mm/pgtable.c
@@ -30,6 +30,7 @@ void show_mem(void)
 	struct page *page;
 	pg_data_t *pgdat;
 	unsigned long i;
+	struct page_state ps;
 
 	printk("Mem-info:\n");
 	show_free_areas();
@@ -53,6 +54,13 @@ void show_mem(void)
 	printk("%d reserved pages\n",reserved);
 	printk("%d pages shared\n",shared);
 	printk("%d pages swap cached\n",cached);
+
+	get_page_state(&ps);
+	printk("%lu pages dirty\n", ps.nr_dirty);
+	printk("%lu pages writeback\n", ps.nr_writeback);
+	printk("%lu pages mapped\n", ps.nr_mapped);
+	printk("%lu pages slab\n", ps.nr_slab);
+	printk("%lu pages pagetables\n", ps.nr_page_table_pages);
 }
 
 /*
-- 
cgit v1.2.3-70-g09d2


From d0e7feb03d5ac48069c2fd57bbba61522e0ca493 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Thu, 23 Jun 2005 00:08:09 -0700
Subject: [PATCH] biarch compiler support for i386

This allows the i386 architecture to be built on a system with a biarch
compiler that defaults to x86-64, merely by specifying ARCH=i386.

As previously discussed, this uses the equivalent logic to the ppc port.

Signed-Off-By: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Makefile | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'arch')

diff --git a/arch/i386/Makefile b/arch/i386/Makefile
index 1c36ca332a9..bf7c9ba709f 100644
--- a/arch/i386/Makefile
+++ b/arch/i386/Makefile
@@ -17,6 +17,13 @@
 # 20050320  Kianusch Sayah Karadji <kianusch@sk-tech.net>
 #           Added support for GEODE CPU
 
+HAS_BIARCH      := $(call cc-option-yn, -m32)
+ifeq ($(HAS_BIARCH),y)
+AS              := $(AS) --32
+LD              := $(LD) -m elf_i386
+CC              := $(CC) -m32
+endif
+
 LDFLAGS		:= -m elf_i386
 OBJCOPYFLAGS	:= -O binary -R .note -R .comment -S
 LDFLAGS_vmlinux :=
-- 
cgit v1.2.3-70-g09d2


From 0f8e2d62fa04441cd12c08ce521e84e5bd3f8a46 Mon Sep 17 00:00:00 2001
From: Ian Campbell <icampbell@arcom.com>
Date: Thu, 23 Jun 2005 00:08:10 -0700
Subject: [PATCH] use ${CROSS_COMPILE}installkernel in arch/*/boot/install.sh

The attached patch causes the various arch specific install.sh scripts to
look for ${CROSS_COMPILE}installkernel rather than just installkernel (in
both /sbin/ and ~/bin/ where the script already did this).  This allows you
to have e.g.  arm-linux-installkernel as a handy way to install on your
cross target.  It also prevents the script picking up on the host
/sbin/installkernel which causes the script to fall through and do the
install itself (which is what I actually use myself, with $INSTALL_PATH
set).

I don't believe it causes back-compatibility problems since calling the
host installkernel was never likely to work or be what you wanted when
cross compiling anyway.  If $CROSS_COMPILE isn't set then nothing changes.

I only use ARM and i386 myself but I figured it couldn't hurt to do the
whole lot.  I've cc'd those who I hope are the arch maintainers for files
that I've touched.

Signed-off-by: Ian Campbell <icampbell@arcom.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/arm/boot/install.sh    | 4 ++--
 arch/arm26/boot/install.sh  | 4 ++--
 arch/i386/boot/install.sh   | 4 ++--
 arch/ppc64/boot/install.sh  | 4 ++--
 arch/s390/boot/install.sh   | 4 ++--
 arch/x86_64/boot/install.sh | 4 ++--
 6 files changed, 12 insertions(+), 12 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/boot/install.sh b/arch/arm/boot/install.sh
index 935bb27369e..9f9bed20734 100644
--- a/arch/arm/boot/install.sh
+++ b/arch/arm/boot/install.sh
@@ -21,8 +21,8 @@
 #
 
 # User may have a custom install script
-if [ -x ~/bin/installkernel ]; then exec ~/bin/installkernel "$@"; fi
-if [ -x /sbin/installkernel ]; then exec /sbin/installkernel "$@"; fi
+if [ -x ~/bin/${CROSS_COMPILE}installkernel ]; then exec ~/bin/${CROSS_COMPILE}installkernel "$@"; fi
+if [ -x /sbin/${CROSS_COMPILE}installkernel ]; then exec /sbin/${CROSS_COMPILE}installkernel "$@"; fi
 
 if [ "$(basename $2)" = "zImage" ]; then
 # Compressed install
diff --git a/arch/arm26/boot/install.sh b/arch/arm26/boot/install.sh
index c628328dd9e..8a8399b26cf 100644
--- a/arch/arm26/boot/install.sh
+++ b/arch/arm26/boot/install.sh
@@ -23,8 +23,8 @@
 
 # User may have a custom install script
 
-if [ -x /sbin/installkernel ]; then
-  exec /sbin/installkernel "$@"
+if [ -x /sbin/${CROSS_COMPILE}installkernel ]; then
+  exec /sbin/${CROSS_COMPILE}installkernel "$@"
 fi
 
 if [ "$2" = "zImage" ]; then
diff --git a/arch/i386/boot/install.sh b/arch/i386/boot/install.sh
index 90f2452b3b9..f17b40dfc0f 100644
--- a/arch/i386/boot/install.sh
+++ b/arch/i386/boot/install.sh
@@ -21,8 +21,8 @@
 
 # User may have a custom install script
 
-if [ -x ~/bin/installkernel ]; then exec ~/bin/installkernel "$@"; fi
-if [ -x /sbin/installkernel ]; then exec /sbin/installkernel "$@"; fi
+if [ -x ~/bin/${CROSS_COMPILE}installkernel ]; then exec ~/bin/${CROSS_COMPILE}installkernel "$@"; fi
+if [ -x /sbin/${CROSS_COMPILE}installkernel ]; then exec /sbin/${CROSS_COMPILE}installkernel "$@"; fi
 
 # Default install - same as make zlilo
 
diff --git a/arch/ppc64/boot/install.sh b/arch/ppc64/boot/install.sh
index 955c5681db6..cb2d6626b55 100644
--- a/arch/ppc64/boot/install.sh
+++ b/arch/ppc64/boot/install.sh
@@ -22,8 +22,8 @@
 
 # User may have a custom install script
 
-if [ -x ~/bin/installkernel ]; then exec ~/bin/installkernel "$@"; fi
-if [ -x /sbin/installkernel ]; then exec /sbin/installkernel "$@"; fi
+if [ -x ~/bin/${CROSS_COMPILE}installkernel ]; then exec ~/bin/${CROSS_COMPILE}installkernel "$@"; fi
+if [ -x /sbin/${CROSS_COMPILE}installkernel ]; then exec /sbin/${CROSS_COMPILE}installkernel "$@"; fi
 
 # Default install
 
diff --git a/arch/s390/boot/install.sh b/arch/s390/boot/install.sh
index 278a8139cb1..d4026f62cb0 100644
--- a/arch/s390/boot/install.sh
+++ b/arch/s390/boot/install.sh
@@ -21,8 +21,8 @@
 
 # User may have a custom install script
 
-if [ -x ~/bin/installkernel ]; then exec ~/bin/installkernel "$@"; fi
-if [ -x /sbin/installkernel ]; then exec /sbin/installkernel "$@"; fi
+if [ -x ~/bin/${CROSS_COMPILE}installkernel ]; then exec ~/bin/${CROSS_COMPILE}installkernel "$@"; fi
+if [ -x /sbin/${CROSS_COMPILE}installkernel ]; then exec /sbin/${CROSS_COMPILE}installkernel "$@"; fi
 
 # Default install - same as make zlilo
 
diff --git a/arch/x86_64/boot/install.sh b/arch/x86_64/boot/install.sh
index 90f2452b3b9..f17b40dfc0f 100644
--- a/arch/x86_64/boot/install.sh
+++ b/arch/x86_64/boot/install.sh
@@ -21,8 +21,8 @@
 
 # User may have a custom install script
 
-if [ -x ~/bin/installkernel ]; then exec ~/bin/installkernel "$@"; fi
-if [ -x /sbin/installkernel ]; then exec /sbin/installkernel "$@"; fi
+if [ -x ~/bin/${CROSS_COMPILE}installkernel ]; then exec ~/bin/${CROSS_COMPILE}installkernel "$@"; fi
+if [ -x /sbin/${CROSS_COMPILE}installkernel ]; then exec /sbin/${CROSS_COMPILE}installkernel "$@"; fi
 
 # Default install - same as make zlilo
 
-- 
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 'arch')

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 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 'arch')

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 7c1def1652c6c1a95eafca2991baace34afaed0f Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Thu, 23 Jun 2005 00:08:21 -0700
Subject: [PATCH] i386: never block forced SIGSEGV

This problem was first noticed on PPC and has already been fixed there.
But the exact same issue applies to other platforms in the same way.  The
signal blocking for sa_mask and the handled signal takes place after the
handler setup.  When the stack is bogus, the handler setup forces a
SIGSEGV.  But then this will be blocked, and returning to user mode will
fault again and iterate.  This patch fixes the problem by checking whether
signal handler setup failed, and not doing the signal-blocking if so.  This
copies what was done in the ppc code.  I think all architectures' signal
handler setup code follows this pattern and needs the change.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/signal.c | 27 ++++++++++++++++-----------
 1 file changed, 16 insertions(+), 11 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/signal.c b/arch/i386/kernel/signal.c
index ea46d028af0..344400787c3 100644
--- a/arch/i386/kernel/signal.c
+++ b/arch/i386/kernel/signal.c
@@ -346,8 +346,8 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs * regs, size_t frame_size)
 extern void __user __kernel_sigreturn;
 extern void __user __kernel_rt_sigreturn;
 
-static void setup_frame(int sig, struct k_sigaction *ka,
-			sigset_t *set, struct pt_regs * regs)
+static int setup_frame(int sig, struct k_sigaction *ka,
+		       sigset_t *set, struct pt_regs * regs)
 {
 	void __user *restorer;
 	struct sigframe __user *frame;
@@ -429,13 +429,14 @@ static void setup_frame(int sig, struct k_sigaction *ka,
 		current->comm, current->pid, frame, regs->eip, frame->pretcode);
 #endif
 
-	return;
+	return 1;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
+	return 0;
 }
 
-static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 			   sigset_t *set, struct pt_regs * regs)
 {
 	void __user *restorer;
@@ -522,20 +523,23 @@ static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 		current->comm, current->pid, frame, regs->eip, frame->pretcode);
 #endif
 
-	return;
+	return 1;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
+	return 0;
 }
 
 /*
  * OK, we're invoking a handler
  */	
 
-static void
+static int
 handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	      sigset_t *oldset,	struct pt_regs * regs)
 {
+	int ret;
+
 	/* Are we from a system call? */
 	if (regs->orig_eax >= 0) {
 		/* If so, check system call restarting.. */
@@ -569,17 +573,19 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 
 	/* Set up the stack frame */
 	if (ka->sa.sa_flags & SA_SIGINFO)
-		setup_rt_frame(sig, ka, info, oldset, regs);
+		ret = setup_rt_frame(sig, ka, info, oldset, regs);
 	else
-		setup_frame(sig, ka, oldset, regs);
+		ret = setup_frame(sig, ka, oldset, regs);
 
-	if (!(ka->sa.sa_flags & SA_NODEFER)) {
+	if (ret && !(ka->sa.sa_flags & SA_NODEFER)) {
 		spin_lock_irq(&current->sighand->siglock);
 		sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
 		sigaddset(&current->blocked,sig);
 		recalc_sigpending();
 		spin_unlock_irq(&current->sighand->siglock);
 	}
+
+	return ret;
 }
 
 /*
@@ -622,8 +628,7 @@ int fastcall do_signal(struct pt_regs *regs, sigset_t *oldset)
 		}
 
 		/* Whee!  Actually deliver the signal.  */
-		handle_signal(signr, &info, &ka, oldset, regs);
-		return 1;
+		return handle_signal(signr, &info, &ka, oldset, regs);
 	}
 
  no_signal:
-- 
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 'arch')

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 7fbb4f6e6873593a2defb8f66512f55d08d88106 Mon Sep 17 00:00:00 2001
From: Jan Beulich <JBeulich@novell.com>
Date: Thu, 23 Jun 2005 00:08:23 -0700
Subject: [PATCH] adjust i386 watchdog tick calculation

Get the i386 watchdog tick calculation into a state where it can also be used
on CPUs with frequencies beyond 4GHz, and it consolidates the calculation into
a single place (for potential furture adjustments).

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/nmi.c | 24 +++++++++++++++---------
 1 file changed, 15 insertions(+), 9 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index 2c0ee9c2d02..da6c46d667c 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -28,8 +28,7 @@
 #include <linux/sysctl.h>
 
 #include <asm/smp.h>
-#include <asm/mtrr.h>
-#include <asm/mpspec.h>
+#include <asm/div64.h>
 #include <asm/nmi.h>
 
 #include "mach_traps.h"
@@ -324,6 +323,16 @@ static void clear_msr_range(unsigned int base, unsigned int n)
 		wrmsr(base+i, 0, 0);
 }
 
+static inline void write_watchdog_counter(const char *descr)
+{
+	u64 count = (u64)cpu_khz * 1000;
+
+	do_div(count, nmi_hz);
+	if(descr)
+		Dprintk("setting %s to -0x%08Lx\n", descr, count);
+	wrmsrl(nmi_perfctr_msr, 0 - count);
+}
+
 static void setup_k7_watchdog(void)
 {
 	unsigned int evntsel;
@@ -339,8 +348,7 @@ static void setup_k7_watchdog(void)
 		| K7_NMI_EVENT;
 
 	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
-	Dprintk("setting K7_PERFCTR0 to %08lx\n", -(cpu_khz/nmi_hz*1000));
-	wrmsr(MSR_K7_PERFCTR0, -(cpu_khz/nmi_hz*1000), -1);
+	write_watchdog_counter("K7_PERFCTR0");
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	evntsel |= K7_EVNTSEL_ENABLE;
 	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
@@ -361,8 +369,7 @@ static void setup_p6_watchdog(void)
 		| P6_NMI_EVENT;
 
 	wrmsr(MSR_P6_EVNTSEL0, evntsel, 0);
-	Dprintk("setting P6_PERFCTR0 to %08lx\n", -(cpu_khz/nmi_hz*1000));
-	wrmsr(MSR_P6_PERFCTR0, -(cpu_khz/nmi_hz*1000), 0);
+	write_watchdog_counter("P6_PERFCTR0");
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	evntsel |= P6_EVNTSEL0_ENABLE;
 	wrmsr(MSR_P6_EVNTSEL0, evntsel, 0);
@@ -402,8 +409,7 @@ static int setup_p4_watchdog(void)
 
 	wrmsr(MSR_P4_CRU_ESCR0, P4_NMI_CRU_ESCR0, 0);
 	wrmsr(MSR_P4_IQ_CCCR0, P4_NMI_IQ_CCCR0 & ~P4_CCCR_ENABLE, 0);
-	Dprintk("setting P4_IQ_COUNTER0 to 0x%08lx\n", -(cpu_khz/nmi_hz*1000));
-	wrmsr(MSR_P4_IQ_COUNTER0, -(cpu_khz/nmi_hz*1000), -1);
+	write_watchdog_counter("P4_IQ_COUNTER0");
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
 	return 1;
@@ -518,7 +524,7 @@ void nmi_watchdog_tick (struct pt_regs * regs)
 			 * other P6 variant */
 			apic_write(APIC_LVTPC, APIC_DM_NMI);
 		}
-		wrmsr(nmi_perfctr_msr, -(cpu_khz/nmi_hz*1000), -1);
+		write_watchdog_counter(NULL);
 	}
 }
 
-- 
cgit v1.2.3-70-g09d2


From 799d19f6ec5ca2102c61122f5219a17f1c4e961a Mon Sep 17 00:00:00 2001
From: Jan Beulich <JBeulich@novell.com>
Date: Thu, 23 Jun 2005 00:08:24 -0700
Subject: [PATCH] allow early printk to use more than 25 lines

Allow early printk code to take advantage of the full size of the screen, not
just the first 25 lines.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Acked-by: 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/early_printk.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

(limited to 'arch')

diff --git a/arch/x86_64/kernel/early_printk.c b/arch/x86_64/kernel/early_printk.c
index e3a19e8ebbf..9631c747c5e 100644
--- a/arch/x86_64/kernel/early_printk.c
+++ b/arch/x86_64/kernel/early_printk.c
@@ -2,20 +2,24 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/string.h>
+#include <linux/tty.h>
 #include <asm/io.h>
 #include <asm/processor.h>
 
 /* Simple VGA output */
 
 #ifdef __i386__
+#include <asm/setup.h>
 #define VGABASE		(__ISA_IO_base + 0xb8000)
 #else
+#include <asm/bootsetup.h>
 #define VGABASE		((void __iomem *)0xffffffff800b8000UL)
 #endif
 
-#define MAX_YPOS	25
-#define MAX_XPOS	80
+#define MAX_YPOS	max_ypos
+#define MAX_XPOS	max_xpos
 
+static int max_ypos = 25, max_xpos = 80;
 static int current_ypos = 1, current_xpos = 0; 
 
 static void early_vga_write(struct console *con, const char *str, unsigned n)
@@ -196,7 +200,10 @@ int __init setup_early_printk(char *opt)
 	} else if (!strncmp(buf, "ttyS", 4)) { 
 		early_serial_init(buf);
 		early_console = &early_serial_console;		
-	} else if (!strncmp(buf, "vga", 3)) {
+	} else if (!strncmp(buf, "vga", 3)
+	           && SCREEN_INFO.orig_video_isVGA == 1) {
+		max_xpos = SCREEN_INFO.orig_video_cols;
+		max_ypos = SCREEN_INFO.orig_video_lines;
 		early_console = &early_vga_console; 
 	}
 	early_console_initialized = 1;
-- 
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 'arch')

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 'arch')

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 c434b7a6aedfe428ad17cd61b21b125a7b7a29ce Mon Sep 17 00:00:00 2001
From: Natalie Protasevich <Natalie.Protasevich@unisys.com>
Date: Thu, 23 Jun 2005 00:08:29 -0700
Subject: [PATCH] x86: avoid wasting IRQs for PCI devices

I have submitted the patch for x86_64, this is submission for i386.

The patch changes the way IRQs are handed out to PCI devices.  Currently,
each I/O APIC pin gets associated with an IRQ, no matter if the pin is used
or not.  This imposes severe limitation on systems that have designs that
employ many I/O APICs, only utilizing couple lines of each, such as P64H2
chipset.  It is used in ES7000, and currently, there is no way to boot the
system with more that 9 I/O APICs.

The simple change below allows to boot a system with say 64 (or more) I/O
APICs, each providing 1 slot, which otherwise impossible because of the IRQ
gaps created for unused lines on each I/O APIC.  It does not resolve the
problem with number of devices that exceeds number of possible IRQs, but
eases up a tension for IRQs on any large system with potentually large
number of devices.

Signed-off-by: Natalie Protasevich <Natalie.Protasevich@unisys.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/mpparse.c | 26 +++++++++++++++++++++++++-
 1 file changed, 25 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index 0a061056b82..383a11600d2 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -1058,11 +1058,20 @@ void __init mp_config_acpi_legacy_irqs (void)
 	}
 }
 
+#define MAX_GSI_NUM	4096
+
 int mp_register_gsi (u32 gsi, int edge_level, int active_high_low)
 {
 	int			ioapic = -1;
 	int			ioapic_pin = 0;
 	int			idx, bit = 0;
+	static int		pci_irq = 16;
+	/*
+	 * Mapping between Global System Interrups, which
+	 * represent all possible interrupts, and IRQs
+	 * assigned to actual devices.
+	 */
+	static int		gsi_to_irq[MAX_GSI_NUM];
 
 #ifdef CONFIG_ACPI_BUS
 	/* Don't set up the ACPI SCI because it's already set up */
@@ -1097,11 +1106,26 @@ int mp_register_gsi (u32 gsi, int edge_level, int active_high_low)
 	if ((1<<bit) & mp_ioapic_routing[ioapic].pin_programmed[idx]) {
 		Dprintk(KERN_DEBUG "Pin %d-%d already programmed\n",
 			mp_ioapic_routing[ioapic].apic_id, ioapic_pin);
-		return gsi;
+		return gsi_to_irq[gsi];
 	}
 
 	mp_ioapic_routing[ioapic].pin_programmed[idx] |= (1<<bit);
 
+	if (edge_level) {
+		/*
+		 * For PCI devices assign IRQs in order, avoiding gaps
+		 * due to unused I/O APIC pins.
+		 */
+		int irq = gsi;
+		if (gsi < MAX_GSI_NUM) {
+			gsi = pci_irq++;
+			gsi_to_irq[irq] = gsi;
+		} else {
+			printk(KERN_ERR "GSI %u is too high\n", gsi);
+			return gsi;
+		}
+	}
+
 	io_apic_set_pci_routing(ioapic, ioapic_pin, gsi,
 		    edge_level == ACPI_EDGE_SENSITIVE ? 0 : 1,
 		    active_high_low == ACPI_ACTIVE_HIGH ? 0 : 1);
-- 
cgit v1.2.3-70-g09d2


From 80bb82afea9eeb94ae3bfed418d079930da5d3b3 Mon Sep 17 00:00:00 2001
From: Aleksey Gorelov <Aleksey_Gorelov@Phoenix.com>
Date: Thu, 23 Jun 2005 00:08:29 -0700
Subject: [PATCH] VIA 82C586B IRQ routing fix

According to the VIA 82C586B datasheet (still available from
http://gkernel.sourceforge.net/specs/via/586b.pdf.bz2) this chip need a
special PIRQ mapping.

Signed-off-by: Karsten Keil <kkeil@suse.de>
Signed-off-by: Aleksey Gorelov <aleksey_gorelov@phoenix.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/pci/irq.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

(limited to 'arch')

diff --git a/arch/i386/pci/irq.c b/arch/i386/pci/irq.c
index da21b1d07c1..83458f81e66 100644
--- a/arch/i386/pci/irq.c
+++ b/arch/i386/pci/irq.c
@@ -226,6 +226,24 @@ static int pirq_via_set(struct pci_dev *router, struct pci_dev *dev, int pirq, i
 	return 1;
 }
 
+/*
+ * The VIA pirq rules are nibble-based, like ALI,
+ * but without the ugly irq number munging.
+ * However, for 82C586, nibble map is different .
+ */
+static int pirq_via586_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
+{
+	static unsigned int pirqmap[4] = { 3, 2, 5, 1 };
+	return read_config_nybble(router, 0x55, pirqmap[pirq-1]);
+}
+
+static int pirq_via586_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
+{
+	static unsigned int pirqmap[4] = { 3, 2, 5, 1 };
+	write_config_nybble(router, 0x55, pirqmap[pirq-1], irq);
+	return 1;
+}
+
 /*
  * ITE 8330G pirq rules are nibble-based
  * FIXME: pirqmap may be { 1, 0, 3, 2 },
@@ -512,6 +530,10 @@ static __init int via_router_probe(struct irq_router *r, struct pci_dev *router,
 	switch(device)
 	{
 		case PCI_DEVICE_ID_VIA_82C586_0:
+			r->name = "VIA";
+			r->get = pirq_via586_get;
+			r->set = pirq_via586_set;
+			return 1;
 		case PCI_DEVICE_ID_VIA_82C596:
 		case PCI_DEVICE_ID_VIA_82C686:
 		case PCI_DEVICE_ID_VIA_8231:
-- 
cgit v1.2.3-70-g09d2


From 129f69465b411592247c408f93d7106939223be1 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Thu, 23 Jun 2005 00:08:33 -0700
Subject: [PATCH] Remove i386_ksyms.c, almost.

* EXPORT_SYMBOL's moved to other files
* #include <linux/config.h>, <linux/module.h> where needed
* #include's in i386_ksyms.c cleaned up
* After copy-paste, redundant due to Makefiles rules preprocessor directives
  removed:

	#ifdef CONFIG_FOO
	EXPORT_SYMBOL(foo);
	#endif

	obj-$(CONFIG_FOO) += foo.o

* Tiny reformat to fit in 80 columns

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/i386_ksyms.c | 160 +-----------------------------------------
 arch/i386/kernel/i387.c       |   3 +
 arch/i386/kernel/io_apic.c    |   3 +-
 arch/i386/kernel/pci-dma.c    |   3 +
 arch/i386/kernel/process.c    |   7 ++
 arch/i386/kernel/reboot.c     |   5 ++
 arch/i386/kernel/setup.c      |  20 ++++++
 arch/i386/kernel/smp.c        |   3 +
 arch/i386/kernel/smpboot.c    |  12 ++++
 arch/i386/kernel/time.c       |   4 ++
 arch/i386/kernel/traps.c      |   3 +
 arch/i386/lib/dec_and_lock.c  |   2 +
 arch/i386/lib/delay.c         |   6 ++
 arch/i386/lib/mmx.c           |   5 ++
 arch/i386/lib/usercopy.c      |   8 ++-
 arch/i386/mm/discontig.c      |   3 +
 arch/i386/mm/highmem.c        |   6 ++
 arch/i386/mm/init.c           |   1 +
 arch/i386/mm/ioremap.c        |   5 +-
 arch/i386/pci/pcbios.c        |   4 +-
 20 files changed, 101 insertions(+), 162 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/i386_ksyms.c b/arch/i386/kernel/i386_ksyms.c
index 903190a4b3f..180f070d03c 100644
--- a/arch/i386/kernel/i386_ksyms.c
+++ b/arch/i386/kernel/i386_ksyms.c
@@ -1,97 +1,17 @@
 #include <linux/config.h>
 #include <linux/module.h>
-#include <linux/smp.h>
-#include <linux/user.h>
-#include <linux/elfcore.h>
-#include <linux/mca.h>
-#include <linux/sched.h>
-#include <linux/in6.h>
-#include <linux/interrupt.h>
-#include <linux/smp_lock.h>
-#include <linux/pm.h>
-#include <linux/pci.h>
-#include <linux/apm_bios.h>
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <linux/tty.h>
-#include <linux/highmem.h>
-#include <linux/time.h>
-
-#include <asm/semaphore.h>
-#include <asm/processor.h>
-#include <asm/i387.h>
-#include <asm/uaccess.h>
 #include <asm/checksum.h>
-#include <asm/io.h>
-#include <asm/delay.h>
-#include <asm/irq.h>
-#include <asm/mmx.h>
 #include <asm/desc.h>
-#include <asm/pgtable.h>
-#include <asm/tlbflush.h>
-#include <asm/nmi.h>
-#include <asm/ist.h>
-#include <asm/kdebug.h>
-
-extern void dump_thread(struct pt_regs *, struct user *);
-extern spinlock_t rtc_lock;
 
 /* This is definitely a GPL-only symbol */
 EXPORT_SYMBOL_GPL(cpu_gdt_table);
 
-#if defined(CONFIG_APM_MODULE)
-extern void machine_real_restart(unsigned char *, int);
-EXPORT_SYMBOL(machine_real_restart);
-extern void default_idle(void);
-EXPORT_SYMBOL(default_idle);
-#endif
-
-#ifdef CONFIG_SMP
-extern void FASTCALL( __write_lock_failed(rwlock_t *rw));
-extern void FASTCALL( __read_lock_failed(rwlock_t *rw));
-#endif
-
-#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_HD) || defined(CONFIG_BLK_DEV_IDE_MODULE) || defined(CONFIG_BLK_DEV_HD_MODULE)
-extern struct drive_info_struct drive_info;
-EXPORT_SYMBOL(drive_info);
-#endif
-
-extern unsigned long cpu_khz;
-extern unsigned long get_cmos_time(void);
-
-/* platform dependent support */
-EXPORT_SYMBOL(boot_cpu_data);
-#ifdef CONFIG_DISCONTIGMEM
-EXPORT_SYMBOL(node_data);
-EXPORT_SYMBOL(physnode_map);
-#endif
-#ifdef CONFIG_X86_NUMAQ
-EXPORT_SYMBOL(xquad_portio);
-#endif
-EXPORT_SYMBOL(dump_thread);
-EXPORT_SYMBOL(dump_fpu);
-EXPORT_SYMBOL_GPL(kernel_fpu_begin);
-EXPORT_SYMBOL(__ioremap);
-EXPORT_SYMBOL(ioremap_nocache);
-EXPORT_SYMBOL(iounmap);
-EXPORT_SYMBOL(kernel_thread);
-EXPORT_SYMBOL(pm_idle);
-EXPORT_SYMBOL(pm_power_off);
-EXPORT_SYMBOL(get_cmos_time);
-EXPORT_SYMBOL(cpu_khz);
-EXPORT_SYMBOL(apm_info);
-
 EXPORT_SYMBOL(__down_failed);
 EXPORT_SYMBOL(__down_failed_interruptible);
 EXPORT_SYMBOL(__down_failed_trylock);
 EXPORT_SYMBOL(__up_wakeup);
 /* Networking helper routines. */
 EXPORT_SYMBOL(csum_partial_copy_generic);
-/* Delay loops */
-EXPORT_SYMBOL(__ndelay);
-EXPORT_SYMBOL(__udelay);
-EXPORT_SYMBOL(__delay);
-EXPORT_SYMBOL(__const_udelay);
 
 EXPORT_SYMBOL(__get_user_1);
 EXPORT_SYMBOL(__get_user_2);
@@ -105,87 +25,11 @@ EXPORT_SYMBOL(__put_user_8);
 EXPORT_SYMBOL(strpbrk);
 EXPORT_SYMBOL(strstr);
 
-EXPORT_SYMBOL(strncpy_from_user);
-EXPORT_SYMBOL(__strncpy_from_user);
-EXPORT_SYMBOL(clear_user);
-EXPORT_SYMBOL(__clear_user);
-EXPORT_SYMBOL(__copy_from_user_ll);
-EXPORT_SYMBOL(__copy_to_user_ll);
-EXPORT_SYMBOL(strnlen_user);
-
-EXPORT_SYMBOL(dma_alloc_coherent);
-EXPORT_SYMBOL(dma_free_coherent);
-
-#ifdef CONFIG_PCI
-EXPORT_SYMBOL(pci_mem_start);
-#endif
-
-#ifdef CONFIG_PCI_BIOS
-EXPORT_SYMBOL(pcibios_set_irq_routing);
-EXPORT_SYMBOL(pcibios_get_irq_routing_table);
-#endif
-
-#ifdef CONFIG_X86_USE_3DNOW
-EXPORT_SYMBOL(_mmx_memcpy);
-EXPORT_SYMBOL(mmx_clear_page);
-EXPORT_SYMBOL(mmx_copy_page);
-#endif
-
-#ifdef CONFIG_X86_HT
-EXPORT_SYMBOL(smp_num_siblings);
-EXPORT_SYMBOL(cpu_sibling_map);
-#endif
-
 #ifdef CONFIG_SMP
-EXPORT_SYMBOL(cpu_data);
-EXPORT_SYMBOL(cpu_online_map);
-EXPORT_SYMBOL(cpu_callout_map);
+extern void FASTCALL( __write_lock_failed(rwlock_t *rw));
+extern void FASTCALL( __read_lock_failed(rwlock_t *rw));
 EXPORT_SYMBOL(__write_lock_failed);
 EXPORT_SYMBOL(__read_lock_failed);
-
-/* Global SMP stuff */
-EXPORT_SYMBOL(smp_call_function);
-
-/* TLB flushing */
-EXPORT_SYMBOL(flush_tlb_page);
-#endif
-
-#ifdef CONFIG_X86_IO_APIC
-EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector);
-#endif
-
-#ifdef CONFIG_MCA
-EXPORT_SYMBOL(machine_id);
-#endif
-
-#ifdef CONFIG_VT
-EXPORT_SYMBOL(screen_info);
-#endif
-
-EXPORT_SYMBOL(get_wchan);
-
-EXPORT_SYMBOL(rtc_lock);
-
-EXPORT_SYMBOL_GPL(set_nmi_callback);
-EXPORT_SYMBOL_GPL(unset_nmi_callback);
-
-EXPORT_SYMBOL(register_die_notifier);
-#ifdef CONFIG_HAVE_DEC_LOCK
-EXPORT_SYMBOL(_atomic_dec_and_lock);
-#endif
-
-EXPORT_SYMBOL(__PAGE_KERNEL);
-
-#ifdef CONFIG_HIGHMEM
-EXPORT_SYMBOL(kmap);
-EXPORT_SYMBOL(kunmap);
-EXPORT_SYMBOL(kmap_atomic);
-EXPORT_SYMBOL(kunmap_atomic);
-EXPORT_SYMBOL(kmap_atomic_to_page);
-#endif
-
-#if defined(CONFIG_X86_SPEEDSTEP_SMI) || defined(CONFIG_X86_SPEEDSTEP_SMI_MODULE)
-EXPORT_SYMBOL(ist_info);
 #endif
 
 EXPORT_SYMBOL(csum_partial);
diff --git a/arch/i386/kernel/i387.c b/arch/i386/kernel/i387.c
index c55e037f08f..b817168d9c6 100644
--- a/arch/i386/kernel/i387.c
+++ b/arch/i386/kernel/i387.c
@@ -10,6 +10,7 @@
 
 #include <linux/config.h>
 #include <linux/sched.h>
+#include <linux/module.h>
 #include <asm/processor.h>
 #include <asm/i387.h>
 #include <asm/math_emu.h>
@@ -79,6 +80,7 @@ void kernel_fpu_begin(void)
 	}
 	clts();
 }
+EXPORT_SYMBOL_GPL(kernel_fpu_begin);
 
 void restore_fpu( struct task_struct *tsk )
 {
@@ -526,6 +528,7 @@ int dump_fpu( struct pt_regs *regs, struct user_i387_struct *fpu )
 
 	return fpvalid;
 }
+EXPORT_SYMBOL(dump_fpu);
 
 int dump_task_fpu(struct task_struct *tsk, struct user_i387_struct *fpu)
 {
diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index 51f7a5d8721..08540bc4ba3 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -31,7 +31,7 @@
 #include <linux/mc146818rtc.h>
 #include <linux/compiler.h>
 #include <linux/acpi.h>
-
+#include <linux/module.h>
 #include <linux/sysdev.h>
 #include <asm/io.h>
 #include <asm/smp.h>
@@ -812,6 +812,7 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin)
 	}
 	return best_guess;
 }
+EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector);
 
 /*
  * This function currently is only a helper for the i386 smp boot process where 
diff --git a/arch/i386/kernel/pci-dma.c b/arch/i386/kernel/pci-dma.c
index 4de2e03c7b4..1e51427cc9e 100644
--- a/arch/i386/kernel/pci-dma.c
+++ b/arch/i386/kernel/pci-dma.c
@@ -11,6 +11,7 @@
 #include <linux/mm.h>
 #include <linux/string.h>
 #include <linux/pci.h>
+#include <linux/module.h>
 #include <asm/io.h>
 
 struct dma_coherent_mem {
@@ -54,6 +55,7 @@ void *dma_alloc_coherent(struct device *dev, size_t size,
 	}
 	return ret;
 }
+EXPORT_SYMBOL(dma_alloc_coherent);
 
 void dma_free_coherent(struct device *dev, size_t size,
 			 void *vaddr, dma_addr_t dma_handle)
@@ -68,6 +70,7 @@ void dma_free_coherent(struct device *dev, size_t size,
 	} else
 		free_pages((unsigned long)vaddr, order);
 }
+EXPORT_SYMBOL(dma_free_coherent);
 
 int dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
 				dma_addr_t device_addr, size_t size, int flags)
diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index 96e3ea6b17c..3c3f245cca5 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -73,6 +73,7 @@ unsigned long thread_saved_pc(struct task_struct *tsk)
  * Powermanagement idle function, if any..
  */
 void (*pm_idle)(void);
+EXPORT_SYMBOL(pm_idle);
 static DEFINE_PER_CPU(unsigned int, cpu_idle_state);
 
 void disable_hlt(void)
@@ -105,6 +106,9 @@ void default_idle(void)
 		cpu_relax();
 	}
 }
+#ifdef CONFIG_APM_MODULE
+EXPORT_SYMBOL(default_idle);
+#endif
 
 /*
  * On SMP it's slightly faster (but much more power-consuming!)
@@ -325,6 +329,7 @@ int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
 	/* Ok, create the new process.. */
 	return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, &regs, 0, NULL, NULL);
 }
+EXPORT_SYMBOL(kernel_thread);
 
 /*
  * Free current thread data structures etc..
@@ -508,6 +513,7 @@ void dump_thread(struct pt_regs * regs, struct user * dump)
 
 	dump->u_fpvalid = dump_fpu (regs, &dump->i387);
 }
+EXPORT_SYMBOL(dump_thread);
 
 /* 
  * Capture the user space registers if the task is not running (in user space)
@@ -731,6 +737,7 @@ unsigned long get_wchan(struct task_struct *p)
 	} while (count++ < 16);
 	return 0;
 }
+EXPORT_SYMBOL(get_wchan);
 
 /*
  * sys_alloc_thread_area: get a yet unused TLS descriptor index.
diff --git a/arch/i386/kernel/reboot.c b/arch/i386/kernel/reboot.c
index 6dc27eb70ee..db912209a8d 100644
--- a/arch/i386/kernel/reboot.c
+++ b/arch/i386/kernel/reboot.c
@@ -2,6 +2,7 @@
  *  linux/arch/i386/kernel/reboot.c
  */
 
+#include <linux/config.h>
 #include <linux/mm.h>
 #include <linux/module.h>
 #include <linux/delay.h>
@@ -19,6 +20,7 @@
  * Power off function, if any
  */
 void (*pm_power_off)(void);
+EXPORT_SYMBOL(pm_power_off);
 
 static int reboot_mode;
 static int reboot_thru_bios;
@@ -295,6 +297,9 @@ void machine_real_restart(unsigned char *code, int length)
 				:
 				: "i" ((void *) (0x1000 - sizeof (real_mode_switch) - 100)));
 }
+#ifdef CONFIG_APM_MODULE
+EXPORT_SYMBOL(machine_real_restart);
+#endif
 
 void machine_restart(char * __unused)
 {
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index 0d689335c8d..30406fd0b64 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -23,6 +23,7 @@
  * This file handles the architecture-dependent parts of initialization
  */
 
+#include <linux/config.h>
 #include <linux/sched.h>
 #include <linux/mm.h>
 #include <linux/mmzone.h>
@@ -74,6 +75,7 @@ EXPORT_SYMBOL(efi_enabled);
 struct cpuinfo_x86 new_cpu_data __initdata = { 0, 0, 0, 0, -1, 1, 0, 0, -1 };
 /* common cpu data for all cpus */
 struct cpuinfo_x86 boot_cpu_data = { 0, 0, 0, 0, -1, 1, 0, 0, -1 };
+EXPORT_SYMBOL(boot_cpu_data);
 
 unsigned long mmu_cr4_features;
 
@@ -91,12 +93,18 @@ extern acpi_interrupt_flags	acpi_sci_flags;
 
 /* for MCA, but anyone else can use it if they want */
 unsigned int machine_id;
+#ifdef CONFIG_MCA
+EXPORT_SYMBOL(machine_id);
+#endif
 unsigned int machine_submodel_id;
 unsigned int BIOS_revision;
 unsigned int mca_pentium_flag;
 
 /* For PCI or other memory-mapped resources */
 unsigned long pci_mem_start = 0x10000000;
+#ifdef CONFIG_PCI
+EXPORT_SYMBOL(pci_mem_start);
+#endif
 
 /* Boot loader ID as an integer, for the benefit of proc_dointvec */
 int bootloader_type;
@@ -108,14 +116,26 @@ static unsigned int highmem_pages = -1;
  * Setup options
  */
 struct drive_info_struct { char dummy[32]; } drive_info;
+#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_HD) || \
+    defined(CONFIG_BLK_DEV_IDE_MODULE) || defined(CONFIG_BLK_DEV_HD_MODULE)
+EXPORT_SYMBOL(drive_info);
+#endif
 struct screen_info screen_info;
+#ifdef CONFIG_VT
+EXPORT_SYMBOL(screen_info);
+#endif
 struct apm_info apm_info;
+EXPORT_SYMBOL(apm_info);
 struct sys_desc_table_struct {
 	unsigned short length;
 	unsigned char table[0];
 };
 struct edid_info edid_info;
 struct ist_info ist_info;
+#if defined(CONFIG_X86_SPEEDSTEP_SMI) || \
+	defined(CONFIG_X86_SPEEDSTEP_SMI_MODULE)
+EXPORT_SYMBOL(ist_info);
+#endif
 struct e820map e820;
 
 extern void early_cpu_init(void);
diff --git a/arch/i386/kernel/smp.c b/arch/i386/kernel/smp.c
index 6223c33ac91..68be7d0c723 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/module.h>
 
 #include <asm/mtrr.h>
 #include <asm/tlbflush.h>
@@ -452,6 +453,7 @@ void flush_tlb_page(struct vm_area_struct * vma, unsigned long va)
 
 	preempt_enable();
 }
+EXPORT_SYMBOL(flush_tlb_page);
 
 static void do_flush_tlb_all(void* info)
 {
@@ -547,6 +549,7 @@ int smp_call_function (void (*func) (void *info), void *info, int nonatomic,
 
 	return 0;
 }
+EXPORT_SYMBOL(smp_call_function);
 
 static void stop_this_cpu (void * dummy)
 {
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index bc1bb6919e6..aaad95e7f87 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -60,6 +60,9 @@ static int __initdata smp_b_stepping;
 
 /* Number of siblings per CPU package */
 int smp_num_siblings = 1;
+#ifdef CONFIG_X86_HT
+EXPORT_SYMBOL(smp_num_siblings);
+#endif
 int phys_proc_id[NR_CPUS]; /* Package ID of each logical CPU */
 EXPORT_SYMBOL(phys_proc_id);
 int cpu_core_id[NR_CPUS]; /* Core ID of each logical CPU */
@@ -67,13 +70,16 @@ EXPORT_SYMBOL(cpu_core_id);
 
 /* bitmap of online cpus */
 cpumask_t cpu_online_map;
+EXPORT_SYMBOL(cpu_online_map);
 
 cpumask_t cpu_callin_map;
 cpumask_t cpu_callout_map;
+EXPORT_SYMBOL(cpu_callout_map);
 static cpumask_t smp_commenced_mask;
 
 /* Per CPU bogomips and other parameters */
 struct cpuinfo_x86 cpu_data[NR_CPUS] __cacheline_aligned;
+EXPORT_SYMBOL(cpu_data);
 
 u8 x86_cpu_to_apicid[NR_CPUS] =
 			{ [0 ... NR_CPUS-1] = 0xff };
@@ -885,8 +891,14 @@ static void smp_tune_scheduling (void)
 static int boot_cpu_logical_apicid;
 /* Where the IO area was mapped on multiquad, always 0 otherwise */
 void *xquad_portio;
+#ifdef CONFIG_X86_NUMAQ
+EXPORT_SYMBOL(xquad_portio);
+#endif
 
 cpumask_t cpu_sibling_map[NR_CPUS] __cacheline_aligned;
+#ifdef CONFIG_X86_HT
+EXPORT_SYMBOL(cpu_sibling_map);
+#endif
 cpumask_t cpu_core_map[NR_CPUS] __cacheline_aligned;
 EXPORT_SYMBOL(cpu_core_map);
 
diff --git a/arch/i386/kernel/time.c b/arch/i386/kernel/time.c
index a0dcb7c87c3..8bc8363fbb4 100644
--- a/arch/i386/kernel/time.c
+++ b/arch/i386/kernel/time.c
@@ -78,10 +78,12 @@ u64 jiffies_64 = INITIAL_JIFFIES;
 EXPORT_SYMBOL(jiffies_64);
 
 unsigned long cpu_khz;	/* Detected as we calibrate the TSC */
+EXPORT_SYMBOL(cpu_khz);
 
 extern unsigned long wall_jiffies;
 
 DEFINE_SPINLOCK(rtc_lock);
+EXPORT_SYMBOL(rtc_lock);
 
 DEFINE_SPINLOCK(i8253_lock);
 EXPORT_SYMBOL(i8253_lock);
@@ -324,6 +326,8 @@ unsigned long get_cmos_time(void)
 
 	return retval;
 }
+EXPORT_SYMBOL(get_cmos_time);
+
 static void sync_cmos_clock(unsigned long dummy);
 
 static struct timer_list sync_cmos_timer =
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 83c579e82a8..7f729665d29 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -104,6 +104,7 @@ int register_die_notifier(struct notifier_block *nb)
 	spin_unlock_irqrestore(&die_notifier_lock, flags);
 	return err;
 }
+EXPORT_SYMBOL(register_die_notifier);
 
 static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
 {
@@ -636,11 +637,13 @@ void set_nmi_callback(nmi_callback_t callback)
 {
 	nmi_callback = callback;
 }
+EXPORT_SYMBOL_GPL(set_nmi_callback);
 
 void unset_nmi_callback(void)
 {
 	nmi_callback = dummy_nmi_callback;
 }
+EXPORT_SYMBOL_GPL(unset_nmi_callback);
 
 #ifdef CONFIG_KPROBES
 fastcall void do_int3(struct pt_regs *regs, long error_code)
diff --git a/arch/i386/lib/dec_and_lock.c b/arch/i386/lib/dec_and_lock.c
index ab43394dc77..8b81b2524fa 100644
--- a/arch/i386/lib/dec_and_lock.c
+++ b/arch/i386/lib/dec_and_lock.c
@@ -8,6 +8,7 @@
  */
 
 #include <linux/spinlock.h>
+#include <linux/module.h>
 #include <asm/atomic.h>
 
 int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock)
@@ -38,3 +39,4 @@ slow_path:
 	spin_unlock(lock);
 	return 0;
 }
+EXPORT_SYMBOL(_atomic_dec_and_lock);
diff --git a/arch/i386/lib/delay.c b/arch/i386/lib/delay.c
index eb0cdfe9280..c49a6acbee5 100644
--- a/arch/i386/lib/delay.c
+++ b/arch/i386/lib/delay.c
@@ -13,6 +13,7 @@
 #include <linux/config.h>
 #include <linux/sched.h>
 #include <linux/delay.h>
+#include <linux/module.h>
 #include <asm/processor.h>
 #include <asm/delay.h>
 #include <asm/timer.h>
@@ -47,3 +48,8 @@ void __ndelay(unsigned long nsecs)
 {
 	__const_udelay(nsecs * 0x00005);  /* 2**32 / 1000000000 (rounded up) */
 }
+
+EXPORT_SYMBOL(__delay);
+EXPORT_SYMBOL(__const_udelay);
+EXPORT_SYMBOL(__udelay);
+EXPORT_SYMBOL(__ndelay);
diff --git a/arch/i386/lib/mmx.c b/arch/i386/lib/mmx.c
index 01f8b1a2cc8..2afda94dffd 100644
--- a/arch/i386/lib/mmx.c
+++ b/arch/i386/lib/mmx.c
@@ -3,6 +3,7 @@
 #include <linux/string.h>
 #include <linux/sched.h>
 #include <linux/hardirq.h>
+#include <linux/module.h>
 
 #include <asm/i387.h>
 
@@ -397,3 +398,7 @@ void mmx_copy_page(void *to, void *from)
 	else
 		fast_copy_page(to, from);
 }
+
+EXPORT_SYMBOL(_mmx_memcpy);
+EXPORT_SYMBOL(mmx_clear_page);
+EXPORT_SYMBOL(mmx_copy_page);
diff --git a/arch/i386/lib/usercopy.c b/arch/i386/lib/usercopy.c
index 51aa2bbb026..4cf981d70f4 100644
--- a/arch/i386/lib/usercopy.c
+++ b/arch/i386/lib/usercopy.c
@@ -84,6 +84,7 @@ __strncpy_from_user(char *dst, const char __user *src, long count)
 	__do_strncpy_from_user(dst, src, count, res);
 	return res;
 }
+EXPORT_SYMBOL(__strncpy_from_user);
 
 /**
  * strncpy_from_user: - Copy a NUL terminated string from userspace.
@@ -111,7 +112,7 @@ strncpy_from_user(char *dst, const char __user *src, long count)
 		__do_strncpy_from_user(dst, src, count, res);
 	return res;
 }
-
+EXPORT_SYMBOL(strncpy_from_user);
 
 /*
  * Zero Userspace
@@ -157,6 +158,7 @@ clear_user(void __user *to, unsigned long n)
 		__do_clear_user(to, n);
 	return n;
 }
+EXPORT_SYMBOL(clear_user);
 
 /**
  * __clear_user: - Zero a block of memory in user space, with less checking.
@@ -175,6 +177,7 @@ __clear_user(void __user *to, unsigned long n)
 	__do_clear_user(to, n);
 	return n;
 }
+EXPORT_SYMBOL(__clear_user);
 
 /**
  * strlen_user: - Get the size of a string in user space.
@@ -218,6 +221,7 @@ long strnlen_user(const char __user *s, long n)
 		:"cc");
 	return res & mask;
 }
+EXPORT_SYMBOL(strnlen_user);
 
 #ifdef CONFIG_X86_INTEL_USERCOPY
 static unsigned long
@@ -570,6 +574,7 @@ survive:
 		n = __copy_user_intel(to, from, n);
 	return n;
 }
+EXPORT_SYMBOL(__copy_to_user_ll);
 
 unsigned long
 __copy_from_user_ll(void *to, const void __user *from, unsigned long n)
@@ -581,6 +586,7 @@ __copy_from_user_ll(void *to, const void __user *from, unsigned long n)
 		n = __copy_user_zeroing_intel(to, from, n);
 	return n;
 }
+EXPORT_SYMBOL(__copy_from_user_ll);
 
 /**
  * copy_to_user: - Copy a block of data into user space.
diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c
index 0efeb96ba5d..f429c871e84 100644
--- a/arch/i386/mm/discontig.c
+++ b/arch/i386/mm/discontig.c
@@ -29,12 +29,14 @@
 #include <linux/highmem.h>
 #include <linux/initrd.h>
 #include <linux/nodemask.h>
+#include <linux/module.h>
 #include <asm/e820.h>
 #include <asm/setup.h>
 #include <asm/mmzone.h>
 #include <bios_ebda.h>
 
 struct pglist_data *node_data[MAX_NUMNODES];
+EXPORT_SYMBOL(node_data);
 bootmem_data_t node0_bdata;
 
 /*
@@ -63,6 +65,7 @@ unsigned long node_end_pfn[MAX_NUMNODES];
  *     physnode_map[8- ] = -1;
  */
 s8 physnode_map[MAX_ELEMENTS] = { [0 ... (MAX_ELEMENTS - 1)] = -1};
+EXPORT_SYMBOL(physnode_map);
 
 void memory_present(int nid, unsigned long start, unsigned long end)
 {
diff --git a/arch/i386/mm/highmem.c b/arch/i386/mm/highmem.c
index fc4c4cad4e9..4b7aaf99d7e 100644
--- a/arch/i386/mm/highmem.c
+++ b/arch/i386/mm/highmem.c
@@ -1,4 +1,5 @@
 #include <linux/highmem.h>
+#include <linux/module.h>
 
 void *kmap(struct page *page)
 {
@@ -87,3 +88,8 @@ struct page *kmap_atomic_to_page(void *ptr)
 	return pte_page(*pte);
 }
 
+EXPORT_SYMBOL(kmap);
+EXPORT_SYMBOL(kunmap);
+EXPORT_SYMBOL(kmap_atomic);
+EXPORT_SYMBOL(kunmap_atomic);
+EXPORT_SYMBOL(kmap_atomic_to_page);
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index 48ebfab77a3..3672e2ef51a 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -295,6 +295,7 @@ static void __init set_highmem_pages_init(int bad_ppro)
 #endif /* CONFIG_HIGHMEM */
 
 unsigned long long __PAGE_KERNEL = _PAGE_KERNEL;
+EXPORT_SYMBOL(__PAGE_KERNEL);
 unsigned long long __PAGE_KERNEL_EXEC = _PAGE_KERNEL_EXEC;
 
 #ifdef CONFIG_NUMA
diff --git a/arch/i386/mm/ioremap.c b/arch/i386/mm/ioremap.c
index ab542792b27..d393eefc705 100644
--- a/arch/i386/mm/ioremap.c
+++ b/arch/i386/mm/ioremap.c
@@ -11,6 +11,7 @@
 #include <linux/vmalloc.h>
 #include <linux/init.h>
 #include <linux/slab.h>
+#include <linux/module.h>
 #include <asm/io.h>
 #include <asm/fixmap.h>
 #include <asm/cacheflush.h>
@@ -165,7 +166,7 @@ void __iomem * __ioremap(unsigned long phys_addr, unsigned long size, unsigned l
 	}
 	return (void __iomem *) (offset + (char __iomem *)addr);
 }
-
+EXPORT_SYMBOL(__ioremap);
 
 /**
  * ioremap_nocache     -   map bus memory into CPU space
@@ -222,6 +223,7 @@ void __iomem *ioremap_nocache (unsigned long phys_addr, unsigned long size)
 
 	return p;					
 }
+EXPORT_SYMBOL(ioremap_nocache);
 
 void iounmap(volatile void __iomem *addr)
 {
@@ -255,6 +257,7 @@ out_unlock:
 	write_unlock(&vmlist_lock);
 	kfree(p); 
 }
+EXPORT_SYMBOL(iounmap);
 
 void __init *bt_ioremap(unsigned long phys_addr, unsigned long size)
 {
diff --git a/arch/i386/pci/pcbios.c b/arch/i386/pci/pcbios.c
index 141421b673b..b9d65f0bc2d 100644
--- a/arch/i386/pci/pcbios.c
+++ b/arch/i386/pci/pcbios.c
@@ -4,6 +4,7 @@
 
 #include <linux/pci.h>
 #include <linux/init.h>
+#include <linux/module.h>
 #include "pci.h"
 #include "pci-functions.h"
 
@@ -456,7 +457,7 @@ struct irq_routing_table * __devinit pcibios_get_irq_routing_table(void)
 	free_page(page);
 	return rt;
 }
-
+EXPORT_SYMBOL(pcibios_get_irq_routing_table);
 
 int pcibios_set_irq_routing(struct pci_dev *dev, int pin, int irq)
 {
@@ -473,6 +474,7 @@ int pcibios_set_irq_routing(struct pci_dev *dev, int pin, int irq)
 		  "S" (&pci_indirect));
 	return !(ret & 0xff00);
 }
+EXPORT_SYMBOL(pcibios_set_irq_routing);
 
 static int __init pci_pcbios_init(void)
 {
-- 
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 'arch')

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 c92c6ffdb16990872acf4ce8b24f82f98fcbbb68 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Thu, 23 Jun 2005 00:08:35 -0700
Subject: [PATCH] mtrr size-and-base debugging

Consolidate the mtrr sanity checking, add a dump_stack().

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/cpu/mtrr/main.c | 23 +++++++++++++++--------
 1 file changed, 15 insertions(+), 8 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/cpu/mtrr/main.c b/arch/i386/kernel/cpu/mtrr/main.c
index e1c2042b9b7..d66b09e0c82 100644
--- a/arch/i386/kernel/cpu/mtrr/main.c
+++ b/arch/i386/kernel/cpu/mtrr/main.c
@@ -375,6 +375,19 @@ int mtrr_add_page(unsigned long base, unsigned long size,
 	return error;
 }
 
+static int mtrr_check(unsigned long base, unsigned long size)
+{
+	if ((base & (PAGE_SIZE - 1)) || (size & (PAGE_SIZE - 1))) {
+		printk(KERN_WARNING
+			"mtrr: size and base must be multiples of 4 kiB\n");
+		printk(KERN_DEBUG
+			"mtrr: size: 0x%lx  base: 0x%lx\n", size, base);
+		dump_stack();
+		return -1;
+	}
+	return 0;
+}
+
 /**
  *	mtrr_add - Add a memory type region
  *	@base: Physical base address of region
@@ -415,11 +428,8 @@ int
 mtrr_add(unsigned long base, unsigned long size, unsigned int type,
 	 char increment)
 {
-	if ((base & (PAGE_SIZE - 1)) || (size & (PAGE_SIZE - 1))) {
-		printk(KERN_WARNING "mtrr: size and base must be multiples of 4 kiB\n");
-		printk(KERN_DEBUG "mtrr: size: 0x%lx  base: 0x%lx\n", size, base);
+	if (mtrr_check(base, size))
 		return -EINVAL;
-	}
 	return mtrr_add_page(base >> PAGE_SHIFT, size >> PAGE_SHIFT, type,
 			     increment);
 }
@@ -511,11 +521,8 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size)
 int
 mtrr_del(int reg, unsigned long base, unsigned long size)
 {
-	if ((base & (PAGE_SIZE - 1)) || (size & (PAGE_SIZE - 1))) {
-		printk(KERN_INFO "mtrr: size and base must be multiples of 4 kiB\n");
-		printk(KERN_DEBUG "mtrr: size: 0x%lx  base: 0x%lx\n", size, base);
+	if (mtrr_check(base, size))
 		return -EINVAL;
-	}
 	return mtrr_del_page(reg, base >> PAGE_SHIFT, size >> PAGE_SHIFT);
 }
 
-- 
cgit v1.2.3-70-g09d2


From c0a88c987878e533fc21fbf684198021a3b2c279 Mon Sep 17 00:00:00 2001
From: Alexander Nyberg <alexn@telia.com>
Date: Thu, 23 Jun 2005 00:08:35 -0700
Subject: [PATCH] x86_64: i8259.c iso99 structure initialization

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/i8259.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

(limited to 'arch')

diff --git a/arch/x86_64/kernel/i8259.c b/arch/x86_64/kernel/i8259.c
index 7873d9ba881..19eafa0aa95 100644
--- a/arch/x86_64/kernel/i8259.c
+++ b/arch/x86_64/kernel/i8259.c
@@ -157,14 +157,13 @@ static unsigned int startup_8259A_irq(unsigned int irq)
 }
 
 static struct hw_interrupt_type i8259A_irq_type = {
-	"XT-PIC",
-	startup_8259A_irq,
-	shutdown_8259A_irq,
-	enable_8259A_irq,
-	disable_8259A_irq,
-	mask_and_ack_8259A,
-	end_8259A_irq,
-	NULL
+	.typename = "XT-PIC",
+	.startup = startup_8259A_irq,
+	.shutdown = shutdown_8259A_irq,
+	.enable = enable_8259A_irq,
+	.disable = disable_8259A_irq,
+	.ack = mask_and_ack_8259A,
+	.end = end_8259A_irq,
 };
 
 /*
-- 
cgit v1.2.3-70-g09d2


From a3a00751ad8970c13d0563c2e92ee68c655a8e6b Mon Sep 17 00:00:00 2001
From: john stultz <johnstul@us.ibm.com>
Date: Thu, 23 Jun 2005 00:08:36 -0700
Subject: [PATCH] x86_64: fix hpet for systems that don't support legacy
 replacement

Currently the x86-64 HPET code assumes the entire HPET implementation from
the spec is present.  This breaks on boxes that do not implement the
optional legacy timer replacement functionality portion of the spec.

This patch fixes this issue, allowing x86-64 systems that cannot use the
HPET for the timer interrupt and RTC to still use the HPET as a time
source.  I've tested this patch on a system systems without HPET, with HPET
but without legacy timer replacement, as well as HPET with legacy timer
replacement.

This version adds a minor check to cap the HPET counter value in
gettimeoffset_hpet to avoid possible time inconsistencies.  Please ignore
the A2 version I sent to you earlier.

Acked-by: 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/time.c | 42 ++++++++++++++++++++++++++++--------------
 1 file changed, 28 insertions(+), 14 deletions(-)

(limited to 'arch')

diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index fb8c809b4cd..66bf6ddeb0c 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -64,6 +64,7 @@ static int notsc __initdata = 0;
 unsigned int cpu_khz;					/* TSC clocks / usec, not used here */
 static unsigned long hpet_period;			/* fsecs / HPET clock */
 unsigned long hpet_tick;				/* HPET clocks / interrupt */
+static int hpet_use_timer;
 unsigned long vxtime_hz = PIT_TICK_RATE;
 int report_lost_ticks;				/* command line option */
 unsigned long long monotonic_base;
@@ -105,7 +106,9 @@ static inline unsigned int do_gettimeoffset_tsc(void)
 
 static inline unsigned int do_gettimeoffset_hpet(void)
 {
-	return ((hpet_readl(HPET_COUNTER) - vxtime.last) * vxtime.quot) >> 32;
+	/* cap counter read to one tick to avoid inconsistencies */
+	unsigned long counter = hpet_readl(HPET_COUNTER) - vxtime.last;
+	return (min(counter,hpet_tick) * vxtime.quot) >> 32;
 }
 
 unsigned int (*do_gettimeoffset)(void) = do_gettimeoffset_tsc;
@@ -301,7 +304,7 @@ unsigned long long monotonic_clock(void)
 
 			last_offset = vxtime.last;
 			base = monotonic_base;
-			this_offset = hpet_readl(HPET_T0_CMP) - hpet_tick;
+			this_offset = hpet_readl(HPET_COUNTER);
 
 		} while (read_seqretry(&xtime_lock, seq));
 		offset = (this_offset - last_offset);
@@ -377,7 +380,14 @@ static irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 	write_seqlock(&xtime_lock);
 
-	if (vxtime.hpet_address) {
+	if (vxtime.hpet_address)
+		offset = hpet_readl(HPET_COUNTER);
+
+	if (hpet_use_timer) {
+		/* if we're using the hpet timer functionality,
+		 * we can more accurately know the counter value
+		 * when the timer interrupt occured.
+		 */
 		offset = hpet_readl(HPET_T0_CMP) - hpet_tick;
 		delay = hpet_readl(HPET_COUNTER) - offset;
 	} else {
@@ -803,17 +813,18 @@ static int hpet_timer_stop_set_go(unsigned long tick)
  * Set up timer 0, as periodic with first interrupt to happen at hpet_tick,
  * and period also hpet_tick.
  */
-
-	hpet_writel(HPET_TN_ENABLE | HPET_TN_PERIODIC | HPET_TN_SETVAL |
+	if (hpet_use_timer) {
+		hpet_writel(HPET_TN_ENABLE | HPET_TN_PERIODIC | HPET_TN_SETVAL |
 		    HPET_TN_32BIT, HPET_T0_CFG);
-	hpet_writel(hpet_tick, HPET_T0_CMP);
-	hpet_writel(hpet_tick, HPET_T0_CMP); /* AK: why twice? */
-
+		hpet_writel(hpet_tick, HPET_T0_CMP);
+		hpet_writel(hpet_tick, HPET_T0_CMP); /* AK: why twice? */
+		cfg |= HPET_CFG_LEGACY;
+	}
 /*
  * Go!
  */
 
-	cfg |= HPET_CFG_ENABLE | HPET_CFG_LEGACY;
+	cfg |= HPET_CFG_ENABLE;
 	hpet_writel(cfg, HPET_CFG);
 
 	return 0;
@@ -834,8 +845,7 @@ static int hpet_init(void)
 
 	id = hpet_readl(HPET_ID);
 
-	if (!(id & HPET_ID_VENDOR) || !(id & HPET_ID_NUMBER) ||
-	    !(id & HPET_ID_LEGSUP))
+	if (!(id & HPET_ID_VENDOR) || !(id & HPET_ID_NUMBER))
 		return -1;
 
 	hpet_period = hpet_readl(HPET_PERIOD);
@@ -845,6 +855,8 @@ static int hpet_init(void)
 	hpet_tick = (1000000000L * (USEC_PER_SEC / HZ) + hpet_period / 2) /
 		hpet_period;
 
+	hpet_use_timer = (id & HPET_ID_LEGSUP);
+
 	return hpet_timer_stop_set_go(hpet_tick);
 }
 
@@ -901,9 +913,11 @@ void __init time_init(void)
 	set_normalized_timespec(&wall_to_monotonic,
 	                        -xtime.tv_sec, -xtime.tv_nsec);
 
-	if (!hpet_init()) {
+	if (!hpet_init())
                 vxtime_hz = (1000000000000000L + hpet_period / 2) /
 			hpet_period;
+
+	if (hpet_use_timer) {
 		cpu_khz = hpet_calibrate_tsc();
 		timename = "HPET";
 #ifdef CONFIG_X86_PM_TIMER
@@ -968,7 +982,7 @@ void __init time_init_gtod(void)
 	if (unsynchronized_tsc())
 		notsc = 1;
 	if (vxtime.hpet_address && notsc) {
-		timetype = "HPET";
+		timetype = hpet_use_timer ? "HPET" : "PIT/HPET";
 		vxtime.last = hpet_readl(HPET_T0_CMP) - hpet_tick;
 		vxtime.mode = VXTIME_HPET;
 		do_gettimeoffset = do_gettimeoffset_hpet;
@@ -983,7 +997,7 @@ void __init time_init_gtod(void)
 		printk(KERN_INFO "Disabling vsyscall due to use of PM timer\n");
 #endif
 	} else {
-		timetype = vxtime.hpet_address ? "HPET/TSC" : "PIT/TSC";
+		timetype = hpet_use_timer ? "HPET/TSC" : "PIT/TSC";
 		vxtime.mode = VXTIME_TSC;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 0928d6ef7f204979749fb241a90a04a35dae133a Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Thu, 23 Jun 2005 00:08:37 -0700
Subject: [PATCH] x86_64: never block forced SIGSEGV

This is the x86_64 version of the signal fix I just posted for i386.

This problem was first noticed on PPC and has already been fixed there.
But the exact same issue applies to other platforms in the same way.  The
signal blocking for sa_mask and the handled signal takes place after the
handler setup.  When the stack is bogus, the handler setup forces a
SIGSEGV.  But then this will be blocked, and returning to user mode will
fault again and iterate.  This patch fixes the problem by checking whether
signal handler setup failed, and not doing the signal-blocking if so.  This
copies what was done in the ppc code.  I think all architectures' signal
handler setup code follows this pattern and needs the change.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/ia32/ia32_signal.c | 15 ++++++++-------
 arch/x86_64/kernel/signal.c    | 26 +++++++++++++++-----------
 2 files changed, 23 insertions(+), 18 deletions(-)

(limited to 'arch')

diff --git a/arch/x86_64/ia32/ia32_signal.c b/arch/x86_64/ia32/ia32_signal.c
index fbd09b5126c..66e2821533d 100644
--- a/arch/x86_64/ia32/ia32_signal.c
+++ b/arch/x86_64/ia32/ia32_signal.c
@@ -428,8 +428,8 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs * regs, size_t frame_size)
 	return (void __user *)((rsp - frame_size) & -8UL);
 }
 
-void ia32_setup_frame(int sig, struct k_sigaction *ka,
-			compat_sigset_t *set, struct pt_regs * regs)
+int ia32_setup_frame(int sig, struct k_sigaction *ka,
+		     compat_sigset_t *set, struct pt_regs * regs)
 {
 	struct sigframe __user *frame;
 	int err = 0;
@@ -514,14 +514,15 @@ void ia32_setup_frame(int sig, struct k_sigaction *ka,
 		current->comm, current->pid, frame, regs->rip, frame->pretcode);
 #endif
 
-	return;
+	return 1;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
+	return 0;
 }
 
-void ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
-			   compat_sigset_t *set, struct pt_regs * regs)
+int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+			compat_sigset_t *set, struct pt_regs * regs)
 {
 	struct rt_sigframe __user *frame;
 	int err = 0;
@@ -613,9 +614,9 @@ void ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 		current->comm, current->pid, frame, regs->rip, frame->pretcode);
 #endif
 
-	return;
+	return 1;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
+	return 0;
 }
-
diff --git a/arch/x86_64/kernel/signal.c b/arch/x86_64/kernel/signal.c
index 429c0269dc4..3c5f3089371 100644
--- a/arch/x86_64/kernel/signal.c
+++ b/arch/x86_64/kernel/signal.c
@@ -34,9 +34,9 @@
 
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 
-void ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
                sigset_t *set, struct pt_regs * regs); 
-void ia32_setup_frame(int sig, struct k_sigaction *ka,
+int ia32_setup_frame(int sig, struct k_sigaction *ka,
             sigset_t *set, struct pt_regs * regs); 
 
 asmlinkage long
@@ -238,7 +238,7 @@ get_stack(struct k_sigaction *ka, struct pt_regs *regs, unsigned long size)
 	return (void __user *)round_down(rsp - size, 16); 
 }
 
-static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 			   sigset_t *set, struct pt_regs * regs)
 {
 	struct rt_sigframe __user *frame;
@@ -327,20 +327,23 @@ static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 		current->comm, current->pid, frame, regs->rip, frame->pretcode);
 #endif
 
-	return;
+	return 1;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
+	return 0;
 }
 
 /*
  * OK, we're invoking a handler
  */	
 
-static void
+static int
 handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 		sigset_t *oldset, struct pt_regs *regs)
 {
+	int ret;
+
 #ifdef DEBUG_SIG
 	printk("handle_signal pid:%d sig:%lu rip:%lx rsp:%lx regs=%p\n",
 		current->pid, sig,
@@ -384,20 +387,22 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 #ifdef CONFIG_IA32_EMULATION
 	if (test_thread_flag(TIF_IA32)) {
 		if (ka->sa.sa_flags & SA_SIGINFO)
-			ia32_setup_rt_frame(sig, ka, info, oldset, regs);
+			ret = ia32_setup_rt_frame(sig, ka, info, oldset, regs);
 		else
-			ia32_setup_frame(sig, ka, oldset, regs);
+			ret = ia32_setup_frame(sig, ka, oldset, regs);
 	} else 
 #endif
-	setup_rt_frame(sig, ka, info, oldset, regs);
+	ret = setup_rt_frame(sig, ka, info, oldset, regs);
 
-	if (!(ka->sa.sa_flags & SA_NODEFER)) {
+	if (ret && !(ka->sa.sa_flags & SA_NODEFER)) {
 		spin_lock_irq(&current->sighand->siglock);
 		sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
 		sigaddset(&current->blocked,sig);
 		recalc_sigpending();
 		spin_unlock_irq(&current->sighand->siglock);
 	}
+
+	return ret;
 }
 
 /*
@@ -437,8 +442,7 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
 			asm volatile("movq %0,%%db7"	: : "r" (current->thread.debugreg7));
 
 		/* Whee!  Actually deliver the signal.  */
-		handle_signal(signr, &info, &ka, oldset, regs);
-		return 1;
+		return handle_signal(signr, &info, &ka, oldset, regs);
 	}
 
  no_signal:
-- 
cgit v1.2.3-70-g09d2


From 701067c4661ebcdc155cc8f696acb24c016c058b Mon Sep 17 00:00:00 2001
From: Natalie Protasevich <Natalie.Protasevich@unisys.com>
Date: Thu, 23 Jun 2005 00:08:41 -0700
Subject: [PATCH] x86_64: avoid wasting IRQs

I suggest to change the way IRQs are handed out to PCI devices.

Currently, each I/O APIC pin gets associated with an IRQ, no matter if the
pin is used or not.  It is expected that each pin can potentually be
engaged by a device inserted into the corresponding PCI slot.  However,
this imposes severe limitation on systems that have designs that employ
many I/O APICs, only utilizing couple lines of each, such as P64H2 chipset.

It is used in ES7000, and currently, there is no way to boot the system
with more that 9 I/O APICs.

The simple change below allows to boot a system with say 64 (or more) I/O
APICs, each providing 1 slot, which otherwise impossible because of the IRQ
gaps created for unused lines on each I/O APIC.  It does not resolve the
problem with number of devices that exceeds number of possible IRQs, but
eases up a tension for IRQs on any large system with potentually large
number of devices.

I only implemented this for the ACPI boot, since if the system is this big
and using newer chipsets it is probably (better be!) an ACPI based system
:).  The change is completely "mechanical" and does not alter any internal
structures or interrupt model/implementation.  The patch works for both
i386 and x86_64 archs.  It works with MSIs just fine, and should not
intervene with implementations like shared vectors, when they get worked
out and incorporated.

To illustrate, below is the interrupt distribution for 2-cell ES7000 with
20 I/O APICs, and an Ethernet card in the last slot, which should be eth1
and which was not configured because its IRQ exceeded allowable number (it
actially turned out huge - 480!):

zorro-tb2:~ # cat /proc/interrupts
           CPU0       CPU1       CPU2       CPU3       CPU4       CPU5       CPU6       CPU7
  0:      65716      30012      30007      30002      30009      30010      30010      30010    IO-APIC-edge  timer
  4:        373          0        725        280          0          0          0          0    IO-APIC-edge  serial
  8:          0          0          0          0          0          0          0          0    IO-APIC-edge  rtc
  9:          0          0          0          0          0          0          0          0   IO-APIC-level  acpi
 14:         39          3          0          0          0          0          0          0    IO-APIC-edge  ide0
 16:        108         13          0          0          0          0          0          0   IO-APIC-level  uhci_hcd:usb1
 18:          0          0          0          0          0          0          0          0   IO-APIC-level  uhci_hcd:usb3
 19:         15          0          0          0          0          0          0          0   IO-APIC-level  uhci_hcd:usb2
 23:          3          0          0          0          0          0          0          0   IO-APIC-level  ehci_hcd:usb4
 96:       4240        397         18          0          0          0          0          0   IO-APIC-level  aic7xxx
 97:         15          0          0          0          0          0          0          0   IO-APIC-level  aic7xxx
192:        847          0          0          0          0          0          0          0   IO-APIC-level  eth0
NMI:          0          0          0          0          0          0          0          0
LOC:     273423     274528     272829     274228     274092     273761     273827     273694
ERR:          7
MIS:          0

Even though the system doesn't have that many devices, some don't get
enabled only because of IRQ numbering model.

This is the IRQ picture after the patch was applied:

zorro-tb2:~ # cat /proc/interrupts
           CPU0       CPU1       CPU2       CPU3       CPU4       CPU5       CPU6       CPU7
  0:      44169      10004      10004      10001      10004      10003      10004       6135    IO-APIC-edge  timer
  4:        345          0          0          0          0        244          0          0    IO-APIC-edge  serial
  8:          0          0          0          0          0          0          0          0    IO-APIC-edge  rtc
  9:          0          0          0          0          0          0          0          0   IO-APIC-level  acpi
 14:         39          0          3          0          0          0          0          0    IO-APIC-edge  ide0
 17:       4425          0          9          0          0          0          0          0   IO-APIC-level  aic7xxx
 18:         15          0          0          0          0          0          0          0   IO-APIC-level  aic7xxx, uhci_hcd:usb3
 21:        231          0          0          0          0          0          0          0   IO-APIC-level  uhci_hcd:usb1
 22:         26          0          0          0          0          0          0          0   IO-APIC-level  uhci_hcd:usb2
 23:          3          0          0          0          0          0          0          0   IO-APIC-level  ehci_hcd:usb4
 24:        348          0          0          0          0          0          0          0   IO-APIC-level  eth0
 25:          6        192          0          0          0          0          0          0   IO-APIC-level  eth1
NMI:          0          0          0          0          0          0          0          0
LOC:     107981     107636     108899     108698     108489     108326     108331     108254
ERR:          7
MIS:          0

Not only we see the card in the last I/O APIC, but we are not even close to
using up available IRQs, since we didn't waste any.

Signed-off-by: Natalie Protasevich <Natalie.Protasevich@unisys.com>
Acked-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/mpparse.c | 21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index ed6a5588146..9c5aa2a790c 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -906,11 +906,20 @@ void __init mp_config_acpi_legacy_irqs (void)
 	return;
 }
 
+#define MAX_GSI_NUM	4096
+
 int mp_register_gsi(u32 gsi, int edge_level, int active_high_low)
 {
 	int			ioapic = -1;
 	int			ioapic_pin = 0;
 	int			idx, bit = 0;
+	static int		pci_irq = 16;
+	/*
+	 * Mapping between Global System Interrupts, which
+	 * represent all possible interrupts, to the IRQs
+	 * assigned to actual devices.
+	 */
+	static int		gsi_to_irq[MAX_GSI_NUM];
 
 	if (acpi_irq_model != ACPI_IRQ_MODEL_IOAPIC)
 		return gsi;
@@ -945,11 +954,21 @@ int mp_register_gsi(u32 gsi, int edge_level, int active_high_low)
 	if ((1<<bit) & mp_ioapic_routing[ioapic].pin_programmed[idx]) {
 		Dprintk(KERN_DEBUG "Pin %d-%d already programmed\n",
 			mp_ioapic_routing[ioapic].apic_id, ioapic_pin);
-		return gsi;
+		return gsi_to_irq[gsi];
 	}
 
 	mp_ioapic_routing[ioapic].pin_programmed[idx] |= (1<<bit);
 
+	if (edge_level) {
+		/*
+		 * For PCI devices assign IRQs in order, avoiding gaps
+		 * due to unused I/O APIC pins.
+		 */
+		int irq = gsi;
+		gsi = pci_irq++;
+		gsi_to_irq[irq] = gsi;
+	}
+
 	io_apic_set_pci_routing(ioapic, ioapic_pin, gsi,
 		edge_level == ACPI_EDGE_SENSITIVE ? 0 : 1,
 		active_high_low == ACPI_ACTIVE_HIGH ? 0 : 1);
-- 
cgit v1.2.3-70-g09d2


From 1cc6f12e03ebc064b74161c684f987284ce9d0cc Mon Sep 17 00:00:00 2001
From: Vincent Hanquez <vincent.hanquez@cl.cam.ac.uk>
Date: Thu, 23 Jun 2005 00:08:43 -0700
Subject: [PATCH] xen: x86: Use new macro for debugreg

Make use of the 2 new macro set_debugreg and get_debugreg.

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/cpu/common.c |  2 +-
 arch/i386/kernel/process.c    | 12 ++++++------
 arch/i386/kernel/signal.c     |  2 +-
 arch/i386/kernel/traps.c      |  6 ++----
 arch/i386/power/cpu.c         | 14 +++++++-------
 5 files changed, 17 insertions(+), 19 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index d199e525680..b9954248d0a 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -635,7 +635,7 @@ void __init cpu_init (void)
 
 	/* Clear all 6 debug registers: */
 
-#define CD(register) __asm__("movl %0,%%db" #register ::"r"(0) );
+#define CD(register) set_debugreg(0, register)
 
 	CD(0); CD(1); CD(2); CD(3); /* no db4 and db5 */; CD(6); CD(7);
 
diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index 3c3f245cca5..2468ab70c38 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -633,13 +633,13 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas
 	 * Now maybe reload the debug registers
 	 */
 	if (unlikely(next->debugreg[7])) {
-		loaddebug(next, 0);
-		loaddebug(next, 1);
-		loaddebug(next, 2);
-		loaddebug(next, 3);
+		set_debugreg(current->thread.debugreg[0], 0);
+		set_debugreg(current->thread.debugreg[1], 1);
+		set_debugreg(current->thread.debugreg[2], 2);
+		set_debugreg(current->thread.debugreg[3], 3);
 		/* no 4 and 5 */
-		loaddebug(next, 6);
-		loaddebug(next, 7);
+		set_debugreg(current->thread.debugreg[6], 6);
+		set_debugreg(current->thread.debugreg[7], 7);
 	}
 
 	if (unlikely(prev->io_bitmap_ptr || next->io_bitmap_ptr))
diff --git a/arch/i386/kernel/signal.c b/arch/i386/kernel/signal.c
index 344400787c3..839d4dc88cd 100644
--- a/arch/i386/kernel/signal.c
+++ b/arch/i386/kernel/signal.c
@@ -624,7 +624,7 @@ int fastcall do_signal(struct pt_regs *regs, sigset_t *oldset)
 		 * inside the kernel.
 		 */
 		if (unlikely(current->thread.debugreg[7])) {
-			loaddebug(&current->thread, 7);
+			set_debugreg(current->thread.debugreg[7], 7);
 		}
 
 		/* Whee!  Actually deliver the signal.  */
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 7f729665d29..c01d7ba6d7e 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -685,7 +685,7 @@ fastcall void do_debug(struct pt_regs * regs, long error_code)
 	unsigned int condition;
 	struct task_struct *tsk = current;
 
-	__asm__ __volatile__("movl %%db6,%0" : "=r" (condition));
+	get_debugreg(condition, 6);
 
 	if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code,
 					SIGTRAP) == NOTIFY_STOP)
@@ -727,9 +727,7 @@ fastcall void do_debug(struct pt_regs * regs, long error_code)
 	 * the signal is delivered.
 	 */
 clear_dr7:
-	__asm__("movl %0,%%db7"
-		: /* no output */
-		: "r" (0));
+	set_debugreg(0, 7);
 	return;
 
 debug_vm86:
diff --git a/arch/i386/power/cpu.c b/arch/i386/power/cpu.c
index cf337c673d9..6f521cf19a1 100644
--- a/arch/i386/power/cpu.c
+++ b/arch/i386/power/cpu.c
@@ -94,13 +94,13 @@ static void fix_processor_context(void)
 	 * Now maybe reload the debug registers
 	 */
 	if (current->thread.debugreg[7]){
-                loaddebug(&current->thread, 0);
-                loaddebug(&current->thread, 1);
-                loaddebug(&current->thread, 2);
-                loaddebug(&current->thread, 3);
-                /* no 4 and 5 */
-                loaddebug(&current->thread, 6);
-                loaddebug(&current->thread, 7);
+		set_debugreg(current->thread.debugreg[0], 0);
+		set_debugreg(current->thread.debugreg[1], 1);
+		set_debugreg(current->thread.debugreg[2], 2);
+		set_debugreg(current->thread.debugreg[3], 3);
+		/* no 4 and 5 */
+		set_debugreg(current->thread.debugreg[6], 6);
+		set_debugreg(current->thread.debugreg[7], 7);
 	}
 
 }
-- 
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 'arch')

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 717b594a415bfaf2dbd5e8266636488f2564c689 Mon Sep 17 00:00:00 2001
From: Vincent Hanquez <vincent.hanquez@cl.cam.ac.uk>
Date: Thu, 23 Jun 2005 00:08:45 -0700
Subject: [PATCH] xen: x86: Use more usermode macro

Use the user_mode macro where it's possible.

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/process.c |  2 +-
 arch/i386/kernel/signal.c  |  2 +-
 arch/i386/kernel/traps.c   | 12 ++++++------
 3 files changed, 8 insertions(+), 8 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index 2468ab70c38..be3efba7caf 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -266,7 +266,7 @@ void show_regs(struct pt_regs * regs)
 	printk("EIP: %04x:[<%08lx>] CPU: %d\n",0xffff & regs->xcs,regs->eip, smp_processor_id());
 	print_symbol("EIP is at %s\n", regs->eip);
 
-	if (regs->xcs & 3)
+	if (user_mode(regs))
 		printk(" ESP: %04x:%08lx",0xffff & regs->xss,regs->esp);
 	printk(" EFLAGS: %08lx    %s  (%s)\n",
 	       regs->eflags, print_tainted(), system_utsname.release);
diff --git a/arch/i386/kernel/signal.c b/arch/i386/kernel/signal.c
index 839d4dc88cd..b9b8f4e20fa 100644
--- a/arch/i386/kernel/signal.c
+++ b/arch/i386/kernel/signal.c
@@ -605,7 +605,7 @@ int fastcall do_signal(struct pt_regs *regs, sigset_t *oldset)
 	 * kernel mode. Just return without doing anything
 	 * if so.
 	 */
-	if ((regs->xcs & 3) != 3)
+	if (!user_mode(regs))
 		return 1;
 
 	if (current->flags & PF_FREEZE) {
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index c01d7ba6d7e..e4d4e2162c7 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -210,7 +210,7 @@ void show_registers(struct pt_regs *regs)
 
 	esp = (unsigned long) (&regs->esp);
 	ss = __KERNEL_DS;
-	if (regs->xcs & 3) {
+	if (user_mode(regs)) {
 		in_kernel = 0;
 		esp = regs->esp;
 		ss = regs->xss & 0xffff;
@@ -266,7 +266,7 @@ static void handle_BUG(struct pt_regs *regs)
 	char c;
 	unsigned long eip;
 
-	if (regs->xcs & 3)
+	if (user_mode(regs))
 		goto no_bug;		/* Not in kernel */
 
 	eip = regs->eip;
@@ -354,7 +354,7 @@ void die(const char * str, struct pt_regs * regs, long err)
 
 static inline void die_if_kernel(const char * str, struct pt_regs * regs, long err)
 {
-	if (!(regs->eflags & VM_MASK) && !(3 & regs->xcs))
+	if (!user_mode_vm(regs))
 		die(str, regs, err);
 }
 
@@ -367,7 +367,7 @@ static void do_trap(int trapnr, int signr, char *str, int vm86,
 		goto trap_signal;
 	}
 
-	if (!(regs->xcs & 3))
+	if (!user_mode(regs))
 		goto kernel_trap;
 
 	trap_signal: {
@@ -489,7 +489,7 @@ fastcall void do_general_protection(struct pt_regs * regs, long error_code)
 	if (regs->eflags & VM_MASK)
 		goto gp_in_vm86;
 
-	if (!(regs->xcs & 3))
+	if (!user_mode(regs))
 		goto gp_in_kernel;
 
 	current->thread.error_code = error_code;
@@ -716,7 +716,7 @@ fastcall void do_debug(struct pt_regs * regs, long error_code)
 		 * check for kernel mode by just checking the CPL
 		 * of CS.
 		 */
-		if ((regs->xcs & 3) == 0)
+		if (!user_mode(regs))
 			goto clear_TF_reenable;
 	}
 
-- 
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 'arch')

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 76381fee7e8feb4c22be636aa5d4765dbe4fbf9e 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: use more usermode macro

Make use of the user_mode macro where it's possible.  This is useful for Xen
because it will need only to redefine only the 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  | 18 +++++++++---------
 arch/x86_64/mm/fault.c      |  4 ++--
 3 files changed, 12 insertions(+), 12 deletions(-)

(limited to 'arch')

diff --git a/arch/x86_64/kernel/signal.c b/arch/x86_64/kernel/signal.c
index 0446e8b96a2..98b7ba95d58 100644
--- a/arch/x86_64/kernel/signal.c
+++ b/arch/x86_64/kernel/signal.c
@@ -422,7 +422,7 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
 	 * kernel mode. Just return without doing anything
 	 * if so.
 	 */
-	if ((regs->cs & 3) != 3)
+	if (!user_mode(regs))
 		return 1;
 
 	if (try_to_freeze(0))
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 21706c1a0f9..121646fc43f 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -274,7 +274,7 @@ EXPORT_SYMBOL(dump_stack);
 void show_registers(struct pt_regs *regs)
 {
 	int i;
-	int in_kernel = (regs->cs & 3) == 0;
+	int in_kernel = !user_mode(regs);
 	unsigned long rsp;
 	const int cpu = safe_smp_processor_id(); 
 	struct task_struct *cur = cpu_pda[cpu].pcurrent; 
@@ -318,7 +318,7 @@ void handle_BUG(struct pt_regs *regs)
 	struct bug_frame f;
 	char tmp;
 
-	if (regs->cs & 3)
+	if (user_mode(regs))
 		return; 
 	if (__copy_from_user(&f, (struct bug_frame *) regs->rip, 
 			     sizeof(struct bug_frame)))
@@ -437,7 +437,7 @@ static void do_trap(int trapnr, int signr, char *str,
        }
 #endif
 
-	if ((regs->cs & 3)  != 0) { 
+	if (user_mode(regs)) {
 		struct task_struct *tsk = current;
 
 		if (exception_trace && unhandled_signal(tsk, signr))
@@ -522,7 +522,7 @@ asmlinkage void do_general_protection(struct pt_regs * regs, long error_code)
        }
 #endif
 
-	if ((regs->cs & 3)!=0) { 
+	if (user_mode(regs)) {
 		struct task_struct *tsk = current;
 
 		if (exception_trace && unhandled_signal(tsk, SIGSEGV))
@@ -638,7 +638,7 @@ asmlinkage struct pt_regs *sync_regs(struct pt_regs *eregs)
 	if (eregs == (struct pt_regs *)eregs->rsp)
 		;
 	/* Exception from user space */
-	else if (eregs->cs & 3)
+	else if (user_mode(eregs))
 		regs = ((struct pt_regs *)current->thread.rsp0) - 1;
 	/* Exception from kernel and interrupts are enabled. Move to
  	   kernel process stack. */
@@ -697,7 +697,7 @@ asmlinkage void do_debug(struct pt_regs * regs, unsigned long error_code)
 		 * allowing programs to debug themselves without the ptrace()
 		 * interface.
 		 */
-                if ((regs->cs & 3) == 0)
+                if (!user_mode(regs))
                        goto clear_TF_reenable;
 		/*
 		 * Was the TF flag set by a debugger? If so, clear it now,
@@ -715,7 +715,7 @@ asmlinkage void do_debug(struct pt_regs * regs, unsigned long error_code)
 	info.si_signo = SIGTRAP;
 	info.si_errno = 0;
 	info.si_code = TRAP_BRKPT;
-	if ((regs->cs & 3) == 0) 
+	if (!user_mode(regs))
 		goto clear_dr7; 
 
 	info.si_addr = (void __user *)regs->rip;
@@ -756,7 +756,7 @@ asmlinkage void do_coprocessor_error(struct pt_regs *regs)
 	unsigned short cwd, swd;
 
 	conditional_sti(regs);
-	if ((regs->cs & 3) == 0 &&
+	if (!user_mode(regs) &&
 	    kernel_math_error(regs, "kernel x87 math error"))
 		return;
 
@@ -822,7 +822,7 @@ asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs)
 	unsigned short mxcsr;
 
 	conditional_sti(regs);
-	if ((regs->cs & 3) == 0 &&
+	if (!user_mode(regs) &&
         	kernel_math_error(regs, "kernel simd math error"))
 		return;
 
diff --git a/arch/x86_64/mm/fault.c b/arch/x86_64/mm/fault.c
index 57d3ab15a5c..2f187986f94 100644
--- a/arch/x86_64/mm/fault.c
+++ b/arch/x86_64/mm/fault.c
@@ -74,7 +74,7 @@ static noinline int is_prefetch(struct pt_regs *regs, unsigned long addr,
 	instr = (unsigned char *)convert_rip_to_linear(current, regs);
 	max_instr = instr + 15;
 
-	if ((regs->cs & 3) != 0 && instr >= (unsigned char *)TASK_SIZE)
+	if (user_mode(regs) && instr >= (unsigned char *)TASK_SIZE)
 		return 0;
 
 	while (scan_more && instr < max_instr) { 
@@ -106,7 +106,7 @@ static noinline int is_prefetch(struct pt_regs *regs, unsigned long addr,
 			/* Could check the LDT for lm, but for now it's good
 			   enough to assume that long mode only uses well known
 			   segments or kernel. */
-			scan_more = ((regs->cs & 3) == 0) || (regs->cs == __USER_CS);
+			scan_more = (!user_mode(regs)) || (regs->cs == __USER_CS);
 			break;
 			
 		case 0x60:
-- 
cgit v1.2.3-70-g09d2


From 15d20bfd606c4b4454aeaa05fc86f77994e48c92 Mon Sep 17 00:00:00 2001
From: Domen Puncer <domen@coderock.org>
Date: Thu, 23 Jun 2005 00:08:47 -0700
Subject: [PATCH] ptrace_h8300: condition bugfix

Assignment doesn't make much sense here as condition would always be true.

Signed-off-by: Domen Puncer <domen@coderock.org>
Signed-off-by: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/h8300/platform/h8300h/ptrace_h8300h.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/h8300/platform/h8300h/ptrace_h8300h.c b/arch/h8300/platform/h8300h/ptrace_h8300h.c
index 18e51a7167d..6ac93c05a1a 100644
--- a/arch/h8300/platform/h8300h/ptrace_h8300h.c
+++ b/arch/h8300/platform/h8300h/ptrace_h8300h.c
@@ -245,12 +245,12 @@ static unsigned short *getnextpc(struct task_struct *child, unsigned short *pc)
 						addr = h8300_get_reg(child, regno-1+PT_ER1);
 					return (unsigned short *)addr;
 				case relb:
-					if ((inst = 0x55) || isbranch(child,inst & 0x0f))
+					if (inst == 0x55 || isbranch(child,inst & 0x0f))
 						pc = (unsigned short *)((unsigned long)pc +
 								       ((signed char)(*fetch_p)));
 					return pc+1; /* skip myself */
 				case relw:
-					if ((inst = 0x5c) || isbranch(child,(*fetch_p & 0xf0) >> 4))
+					if (inst == 0x5c || isbranch(child,(*fetch_p & 0xf0) >> 4))
 						pc = (unsigned short *)((unsigned long)pc +
 								       ((signed short)(*(pc+1))));
 					return pc+2; /* skip myself */
-- 
cgit v1.2.3-70-g09d2


From dfe52244e004f5103478966cd88351feb5c50d79 Mon Sep 17 00:00:00 2001
From: Robert Love <rml@novell.com>
Date: Thu, 23 Jun 2005 00:09:04 -0700
Subject: [PATCH] kstrdup: convert a few existing implementations

Convert a bunch of strdup() implementations and their callers to the new
kstrdup().  A few remain, for example see sound/core, and there are tons of
open coded strdup()'s around.  Sigh.  But this is a start.

Signed-off-by: Robert Love <rml@novell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/process_kern.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

(limited to 'arch')

diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c
index 804c6bbdf67..157584ae479 100644
--- a/arch/um/kernel/process_kern.c
+++ b/arch/um/kernel/process_kern.c
@@ -8,6 +8,7 @@
 #include "linux/kernel.h"
 #include "linux/sched.h"
 #include "linux/interrupt.h"
+#include "linux/string.h"
 #include "linux/mm.h"
 #include "linux/slab.h"
 #include "linux/utsname.h"
@@ -322,12 +323,7 @@ void do_uml_exitcalls(void)
 
 char *uml_strdup(char *string)
 {
-	char *new;
-
-	new = kmalloc(strlen(string) + 1, GFP_KERNEL);
-	if(new == NULL) return(NULL);
-	strcpy(new, string);
-	return(new);
+	return kstrdup(string, GFP_KERNEL);
 }
 
 int copy_to_user_proc(void __user *to, void *from, int size)
-- 
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 'arch')

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 'arch')

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 'arch')

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 'arch')

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 'arch')

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 'arch')

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 'arch')

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 'arch')

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 'arch')

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 13608d6433eb34840224ef632cc444f3eb59bc13 Mon Sep 17 00:00:00 2001
From: Rusty Lynch <rusty.lynch@intel.com>
Date: Thu, 23 Jun 2005 00:09:31 -0700
Subject: [PATCH] Kprobes ia64 qp fix

Fix a bug where a kprobe still fires when the instruction is predicated
off.  So given the p6=0, and we have an instruction like:

(p6) move loc1=0

we should not be triggering the kprobe.  This is handled by carrying over
the qp section of the original instruction into the break instruction.

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/kprobes.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index b7a204137fb..98bef04d948 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -115,19 +115,19 @@ int arch_prepare_kprobe(struct kprobe *p)
 	case 0:
  		major_opcode = (bundle->quad0.slot0 >> SLOT0_OPCODE_SHIFT);
  		kprobe_inst = bundle->quad0.slot0;
-		bundle->quad0.slot0 = BREAK_INST;
+		bundle->quad0.slot0 = BREAK_INST | (0x3f & kprobe_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->quad0.slot1_p0 = BREAK_INST | (0x3f & kprobe_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;
+		bundle->quad1.slot2 = BREAK_INST | (0x3f & kprobe_inst);
 		break;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From a5403183d84d419651672f1aee5ff2273d185efa Mon Sep 17 00:00:00 2001
From: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Date: Thu, 23 Jun 2005 00:09:32 -0700
Subject: [PATCH] Kprobes IA64: arch_prepare_kprobes() cleanup

arch_prepare_kprobes() was doing lots of functionality
in just one single function. This patch
attempts to clean up arch_prepare_kprobes() by moving
specific sub task to the following (new)functions
1)valid_kprobe_addr() -->> validate the given kprobe address
2)get_kprobe_inst(slot..)->> Retrives the instruction for a given slot from the bundle
3)prepare_break_inst() -->> Prepares break instruction within the bundle
	3a)update_kprobe_inst_flag()-->>Updates the internal flags, required
			for proper emulation of the instruction at later
			point in time.

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 | 191 ++++++++++++++++++++++++++++-----------------
 1 file changed, 121 insertions(+), 70 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index 98bef04d948..5d534468155 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -81,90 +81,141 @@ static enum instruction_type bundle_encoding[32][3] = {
   { u, u, u },  			/* 1F */
 };
 
-int arch_prepare_kprobe(struct kprobe *p)
+/*
+ * In this function we check to see if the instruction
+ * is IP relative instruction and update the kprobe
+ * inst flag accordingly
+ */
+static void update_kprobe_inst_flag(uint template, uint  slot, uint major_opcode,
+	unsigned long kprobe_inst, struct kprobe *p)
 {
-	unsigned long addr = (unsigned long) p->addr;
-	unsigned long *bundle_addr = (unsigned long *)(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.insn.bundle;
-
-	memcpy(&p->opcode.bundle, bundle_addr, sizeof(bundle_t));
-	memcpy(&p->ainsn.insn.bundle, bundle_addr, sizeof(bundle_t));
-
 	p->ainsn.inst_flag = 0;
 	p->ainsn.target_br_reg = 0;
 
- 	template = bundle->quad0.template;
+	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;
+		}
+ 	} else if (bundle_encoding[template][slot] == X) {
+		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;
+		}
+	}
+	return;
+}
 
-	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;
+/*
+ * In this function we override the bundle with
+ * the break instruction at the given slot.
+ */
+static void prepare_break_inst(uint template, uint  slot, uint major_opcode,
+	unsigned long kprobe_inst, struct kprobe *p)
+{
+	unsigned long break_inst = BREAK_INST;
+	bundle_t *bundle = &p->ainsn.insn.bundle;
+
+	/*
+	 * Copy the original kprobe_inst qualifying predicate(qp)
+	 * to the break instruction
+	 */
+	break_inst |= (0x3f & kprobe_inst);
+
+	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;
 	}
 
- 	if (slot == 1 && bundle_encoding[template][1] == L) {
- 		lx_type_inst = 1;
-  		slot = 2;
- 	}
+	/*
+	 * Update the instruction flag, so that we can
+	 * emulate the instruction properly after we
+	 * single step on original instruction
+	 */
+	update_kprobe_inst_flag(template, slot, major_opcode, kprobe_inst, p);
+}
+
+static inline void get_kprobe_inst(bundle_t *bundle, uint slot,
+	       	unsigned long *kprobe_inst, uint *major_opcode)
+{
+	unsigned long kprobe_inst_p0, kprobe_inst_p1;
+	unsigned int template;
+
+	template = bundle->quad0.template;
 
 	switch (slot) {
-	case 0:
- 		major_opcode = (bundle->quad0.slot0 >> SLOT0_OPCODE_SHIFT);
- 		kprobe_inst = bundle->quad0.slot0;
-		bundle->quad0.slot0 = BREAK_INST | (0x3f & kprobe_inst);
+	  case 0:
+ 		*major_opcode = (bundle->quad0.slot0 >> SLOT0_OPCODE_SHIFT);
+ 		*kprobe_inst = bundle->quad0.slot0;
 		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 | (0x3f & kprobe_inst);
-		bundle->quad1.slot1_p1 = (BREAK_INST >> (64-46));
+	  case 1:
+ 		*major_opcode = (bundle->quad1.slot1_p1 >> SLOT1_p1_OPCODE_SHIFT);
+  		kprobe_inst_p0 = bundle->quad0.slot1_p0;
+  		kprobe_inst_p1 = bundle->quad1.slot1_p1;
+  		*kprobe_inst = kprobe_inst_p0 | (kprobe_inst_p1 << (64-46));
 		break;
-	case 2:
- 		major_opcode = (bundle->quad1.slot2 >> SLOT2_OPCODE_SHIFT);
- 		kprobe_inst = bundle->quad1.slot2;
-		bundle->quad1.slot2 = BREAK_INST | (0x3f & kprobe_inst);
+	  case 2:
+ 		*major_opcode = (bundle->quad1.slot2 >> SLOT2_OPCODE_SHIFT);
+ 		*kprobe_inst = bundle->quad1.slot2;
 		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;
- 		}
+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);
+		return -EINVAL;
 	}
+	return 0;
+}
+
+int arch_prepare_kprobe(struct kprobe *p)
+{
+	unsigned long addr = (unsigned long) p->addr;
+	unsigned long *kprobe_addr = (unsigned long *)(addr & ~0xFULL);
+	unsigned long kprobe_inst=0;
+	unsigned int slot = addr & 0xf, template, major_opcode = 0;
+	bundle_t *bundle = &p->ainsn.insn.bundle;
+
+	memcpy(&p->opcode.bundle, kprobe_addr, sizeof(bundle_t));
+	memcpy(&p->ainsn.insn.bundle, kprobe_addr, sizeof(bundle_t));
+
+ 	template = bundle->quad0.template;
+
+	if(valid_kprobe_addr(template, slot, addr))
+		return -EINVAL;
+
+	/* Move to slot 2, if bundle is MLX type and kprobe slot is 1 */
+ 	if (slot == 1 && bundle_encoding[template][1] == L)
+  		slot++;
+
+	/* Get kprobe_inst and major_opcode from the bundle */
+	get_kprobe_inst(bundle, slot, &kprobe_inst, &major_opcode);
+
+	prepare_break_inst(template, slot, major_opcode, kprobe_inst, p);
 
 	return 0;
 }
@@ -260,7 +311,7 @@ static void resume_execution(struct kprobe *p, struct pt_regs *regs)
  		if (regs->cr_iip == bundle_addr) {
  			regs->cr_iip = resume_addr;
  		}
- 	}
+	}
 
 turn_ss_off:
   	/* Turn off Single Step bit */
-- 
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 'arch')

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 708de8f11c2901cc49fd7725baf4a0fbd7264e73 Mon Sep 17 00:00:00 2001
From: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Date: Thu, 23 Jun 2005 00:09:34 -0700
Subject: [PATCH] Kprobes IA64: safe register kprobe

The current kprobes does not yet handle register kprobes on some of the
following kind of instruction which needs to be emulated in a special way.

1) mov r1=ip
2) chk -- Speculation check instruction

This patch attempts to fail register_kprobes() when user tries to insert
kprobes on the above kind of instruction.

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 | 45 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 45 insertions(+)

(limited to 'arch')

diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index 51bc08bd046..027d656664d 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -119,6 +119,48 @@ static void update_kprobe_inst_flag(uint template, uint  slot, uint major_opcode
 	return;
 }
 
+/*
+ * In this function we check to see if the instruction
+ * on which we are inserting kprobe is supported.
+ * Returns 0 if supported
+ * Returns -EINVAL if unsupported
+ */
+static int unsupported_inst(uint template, uint  slot, uint major_opcode,
+	unsigned long kprobe_inst, struct kprobe *p)
+{
+	unsigned long addr = (unsigned long)p->addr;
+
+	if (bundle_encoding[template][slot] == I) {
+		switch (major_opcode) {
+			case 0x0: //I_UNIT_MISC_OPCODE:
+			/*
+			 * Check for Integer speculation instruction
+			 * - Bit 33-35 to be equal to 0x1
+			 */
+			if (((kprobe_inst >> 33) & 0x7) == 1) {
+				printk(KERN_WARNING
+					"Kprobes on speculation inst at <0x%lx> not supported\n",
+					addr);
+				return -EINVAL;
+			}
+
+			/*
+			 * IP relative mov instruction
+			 *  - Bit 27-35 to be equal to 0x30
+			 */
+			if (((kprobe_inst >> 27) & 0x1FF) == 0x30) {
+				printk(KERN_WARNING
+					"Kprobes on \"mov r1=ip\" at <0x%lx> not supported\n",
+					addr);
+				return -EINVAL;
+
+			}
+		}
+	}
+	return 0;
+}
+
+
 /*
  * In this function we check to see if the instruction
  * (qp) cmpx.crel.ctype p1,p2=r2,r3
@@ -254,6 +296,9 @@ int arch_prepare_kprobe(struct kprobe *p)
 	/* Get kprobe_inst and major_opcode from the bundle */
 	get_kprobe_inst(bundle, slot, &kprobe_inst, &major_opcode);
 
+	if (unsupported_inst(template, slot, major_opcode, kprobe_inst, p))
+			return -EINVAL;
+
 	prepare_break_inst(template, slot, major_opcode, kprobe_inst, p);
 
 	return 0;
-- 
cgit v1.2.3-70-g09d2


From 89cb14c0dd0e4a7d0315d19f449389c4d49237ee Mon Sep 17 00:00:00 2001
From: Keshavamurthy Anil S <anil.s.keshavamurthy@intel.com>
Date: Thu, 23 Jun 2005 00:09:35 -0700
Subject: [PATCH] Kprobes/IA64: check jprobe break before handling

Once the jprobe instrumented function returns, it executes a jprobe_break
which is a break instruction with __IA64_JPROBE_BREAK value.  The current
patch checks for this break value, before assuming that jprobe instrumented
function just completed.

The previous code was not checking for this value and that was a bug.

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 | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index 027d656664d..41e80b42d3f 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -419,10 +419,11 @@ static void prepare_ss(struct kprobe *p, struct pt_regs *regs)
 	ia64_psr(regs)->ss = 1;
 }
 
-static int pre_kprobes_handler(struct pt_regs *regs)
+static int pre_kprobes_handler(struct die_args *args)
 {
 	struct kprobe *p;
 	int ret = 0;
+	struct pt_regs *regs = args->regs;
 	kprobe_opcode_t *addr = (kprobe_opcode_t *)instruction_pointer(regs);
 
 	preempt_disable();
@@ -437,7 +438,7 @@ static int pre_kprobes_handler(struct pt_regs *regs)
 			}
 			arch_disarm_kprobe(p);
 			ret = 1;
-		} else {
+		} else if (args->err == __IA64_BREAK_JPROBE) {
 			/*
 			 * jprobe instrumented function just completed
 			 */
@@ -445,6 +446,9 @@ static int pre_kprobes_handler(struct pt_regs *regs)
 			if (p->break_handler && p->break_handler(p, regs)) {
 				goto ss_probe;
 			}
+		} else {
+			/* Not our break */
+			goto no_kprobe;
 		}
 	}
 
@@ -515,7 +519,7 @@ int kprobe_exceptions_notify(struct notifier_block *self, unsigned long val,
 	struct die_args *args = (struct die_args *)data;
 	switch(val) {
 	case DIE_BREAK:
-		if (pre_kprobes_handler(args->regs))
+		if (pre_kprobes_handler(args))
 			return NOTIFY_STOP;
 		break;
 	case DIE_SS:
-- 
cgit v1.2.3-70-g09d2


From 417c8da6511b54843e6b557d94d8112d80e32156 Mon Sep 17 00:00:00 2001
From: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Date: Thu, 23 Jun 2005 00:09:37 -0700
Subject: [PATCH] kprobes: Temporary disarming of reentrant probe for i386

This patch includes i386 architecture specific 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>
---
 arch/i386/kernel/kprobes.c | 62 ++++++++++++++++++++++++++++++++++++----------
 1 file changed, 49 insertions(+), 13 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c
index b8e2bae0ab4..3762f6b35ab 100644
--- a/arch/i386/kernel/kprobes.c
+++ b/arch/i386/kernel/kprobes.c
@@ -37,12 +37,10 @@
 #include <asm/kdebug.h>
 #include <asm/desc.h>
 
-/* kprobe_status settings */
-#define KPROBE_HIT_ACTIVE	0x00000001
-#define KPROBE_HIT_SS		0x00000002
-
 static struct kprobe *current_kprobe;
 static unsigned long kprobe_status, kprobe_old_eflags, kprobe_saved_eflags;
+static struct kprobe *kprobe_prev;
+static unsigned long kprobe_status_prev, kprobe_old_eflags_prev, kprobe_saved_eflags_prev;
 static struct pt_regs jprobe_saved_regs;
 static long *jprobe_saved_esp;
 /* copy of the kernel stack at the probe fire time */
@@ -93,6 +91,31 @@ void arch_remove_kprobe(struct kprobe *p)
 {
 }
 
+static inline void save_previous_kprobe(void)
+{
+	kprobe_prev = current_kprobe;
+	kprobe_status_prev = kprobe_status;
+	kprobe_old_eflags_prev = kprobe_old_eflags;
+	kprobe_saved_eflags_prev = kprobe_saved_eflags;
+}
+
+static inline void restore_previous_kprobe(void)
+{
+	current_kprobe = kprobe_prev;
+	kprobe_status = kprobe_status_prev;
+	kprobe_old_eflags = kprobe_old_eflags_prev;
+	kprobe_saved_eflags = kprobe_saved_eflags_prev;
+}
+
+static inline void set_current_kprobe(struct kprobe *p, struct pt_regs *regs)
+{
+	current_kprobe = p;
+	kprobe_saved_eflags = kprobe_old_eflags
+		= (regs->eflags & (TF_MASK | IF_MASK));
+	if (is_IF_modifier(p->opcode))
+		kprobe_saved_eflags &= ~IF_MASK;
+}
+
 static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 {
 	regs->eflags |= TF_MASK;
@@ -184,9 +207,18 @@ static int kprobe_handler(struct pt_regs *regs)
 				unlock_kprobes();
 				goto no_kprobe;
 			}
-			arch_disarm_kprobe(p);
-			regs->eip = (unsigned long)p->addr;
-			ret = 1;
+			/* We have reentered the kprobe_handler(), since
+			 * another probe was hit while within the handler.
+			 * We here save the original kprobes variables and
+			 * just single step on the instruction of the new probe
+			 * without calling any user handlers.
+			 */
+			save_previous_kprobe();
+			set_current_kprobe(p, regs);
+			p->nmissed++;
+			prepare_singlestep(p, regs);
+			kprobe_status = KPROBE_REENTER;
+			return 1;
 		} else {
 			p = current_kprobe;
 			if (p->break_handler && p->break_handler(p, regs)) {
@@ -221,11 +253,7 @@ static int kprobe_handler(struct pt_regs *regs)
 	}
 
 	kprobe_status = KPROBE_HIT_ACTIVE;
-	current_kprobe = p;
-	kprobe_saved_eflags = kprobe_old_eflags
-	    = (regs->eflags & (TF_MASK | IF_MASK));
-	if (is_IF_modifier(p->opcode))
-		kprobe_saved_eflags &= ~IF_MASK;
+	set_current_kprobe(p, regs);
 
 	if (p->pre_handler && p->pre_handler(p, regs))
 		/* handler has already set things up, so skip ss setup */
@@ -370,14 +398,22 @@ static inline int post_kprobe_handler(struct pt_regs *regs)
 	if (!kprobe_running())
 		return 0;
 
-	if (current_kprobe->post_handler)
+	if ((kprobe_status != KPROBE_REENTER) && current_kprobe->post_handler) {
+		kprobe_status = KPROBE_HIT_SSDONE;
 		current_kprobe->post_handler(current_kprobe, regs, 0);
+	}
 
 	if (current_kprobe->post_handler != trampoline_post_handler)
 		resume_execution(current_kprobe, regs);
 	regs->eflags |= kprobe_saved_eflags;
 
+	/*Restore back the original saved kprobes variables and continue. */
+	if (kprobe_status == KPROBE_REENTER) {
+		restore_previous_kprobe();
+		goto out;
+	}
 	unlock_kprobes();
+out:
 	preempt_enable_no_resched();
 
 	/*
-- 
cgit v1.2.3-70-g09d2


From aa3d7e3d782bbcf567b1da6101d8fbff7cf7e7a6 Mon Sep 17 00:00:00 2001
From: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Date: Thu, 23 Jun 2005 00:09:37 -0700
Subject: [PATCH] kprobes: Temporary disarming of reentrant probe for x86_64

This patch includes x86_64 architecture specific 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>
---
 arch/x86_64/kernel/kprobes.c | 76 ++++++++++++++++++++++++++++++++++++--------
 1 file changed, 62 insertions(+), 14 deletions(-)

(limited to 'arch')

diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c
index 324bf57925a..4e680f87a75 100644
--- a/arch/x86_64/kernel/kprobes.c
+++ b/arch/x86_64/kernel/kprobes.c
@@ -45,12 +45,10 @@
 
 static DECLARE_MUTEX(kprobe_mutex);
 
-/* kprobe_status settings */
-#define KPROBE_HIT_ACTIVE	0x00000001
-#define KPROBE_HIT_SS		0x00000002
-
 static struct kprobe *current_kprobe;
 static unsigned long kprobe_status, kprobe_old_rflags, kprobe_saved_rflags;
+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);
@@ -240,6 +238,31 @@ void arch_remove_kprobe(struct kprobe *p)
 	down(&kprobe_mutex);
 }
 
+static inline void save_previous_kprobe(void)
+{
+	kprobe_prev = current_kprobe;
+	kprobe_status_prev = kprobe_status;
+	kprobe_old_rflags_prev = kprobe_old_rflags;
+	kprobe_saved_rflags_prev = kprobe_saved_rflags;
+}
+
+static inline void restore_previous_kprobe(void)
+{
+	current_kprobe = kprobe_prev;
+	kprobe_status = kprobe_status_prev;
+	kprobe_old_rflags = kprobe_old_rflags_prev;
+	kprobe_saved_rflags = kprobe_saved_rflags_prev;
+}
+
+static inline void set_current_kprobe(struct kprobe *p, struct pt_regs *regs)
+{
+	current_kprobe = p;
+	kprobe_saved_rflags = kprobe_old_rflags
+		= (regs->eflags & (TF_MASK | IF_MASK));
+	if (is_IF_modifier(p->ainsn.insn))
+		kprobe_saved_rflags &= ~IF_MASK;
+}
+
 static void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 {
 	regs->eflags |= TF_MASK;
@@ -319,10 +342,30 @@ int kprobe_handler(struct pt_regs *regs)
 				regs->eflags |= kprobe_saved_rflags;
 				unlock_kprobes();
 				goto no_kprobe;
+			} else if (kprobe_status == KPROBE_HIT_SSDONE) {
+				/* TODO: Provide re-entrancy from
+				 * post_kprobes_handler() and avoid exception
+				 * stack corruption while single-stepping on
+				 * the instruction of the new probe.
+				 */
+				arch_disarm_kprobe(p);
+				regs->rip = (unsigned long)p->addr;
+				ret = 1;
+			} else {
+				/* We have reentered the kprobe_handler(), since
+				 * another probe was hit while within the
+				 * handler. We here save the original kprobe
+				 * variables and just single step on instruction
+				 * of the new probe without calling any user
+				 * handlers.
+				 */
+				save_previous_kprobe();
+				set_current_kprobe(p, regs);
+				p->nmissed++;
+				prepare_singlestep(p, regs);
+				kprobe_status = KPROBE_REENTER;
+				return 1;
 			}
-			arch_disarm_kprobe(p);
-			regs->rip = (unsigned long)p->addr;
-			ret = 1;
 		} else {
 			p = current_kprobe;
 			if (p->break_handler && p->break_handler(p, regs)) {
@@ -352,11 +395,7 @@ int kprobe_handler(struct pt_regs *regs)
 	}
 
 	kprobe_status = KPROBE_HIT_ACTIVE;
-	current_kprobe = p;
-	kprobe_saved_rflags = kprobe_old_rflags
-	    = (regs->eflags & (TF_MASK | IF_MASK));
-	if (is_IF_modifier(p->ainsn.insn))
-		kprobe_saved_rflags &= ~IF_MASK;
+	set_current_kprobe(p, regs);
 
 	if (p->pre_handler && p->pre_handler(p, regs))
 		/* handler has already set things up, so skip ss setup */
@@ -506,14 +545,23 @@ int post_kprobe_handler(struct pt_regs *regs)
 	if (!kprobe_running())
 		return 0;
 
-	if (current_kprobe->post_handler)
+	if ((kprobe_status != KPROBE_REENTER) && current_kprobe->post_handler) {
+		kprobe_status = KPROBE_HIT_SSDONE;
 		current_kprobe->post_handler(current_kprobe, regs, 0);
+	}
 
 	if (current_kprobe->post_handler != trampoline_post_handler)
 		resume_execution(current_kprobe, regs);
 	regs->eflags |= kprobe_saved_rflags;
 
-	unlock_kprobes();
+	/* Restore the original saved kprobes variables and continue. */
+	if (kprobe_status == KPROBE_REENTER) {
+		restore_previous_kprobe();
+		goto out;
+	} else {
+		unlock_kprobes();
+	}
+out:
 	preempt_enable_no_resched();
 
 	/*
-- 
cgit v1.2.3-70-g09d2


From 42cc20600a3450efcf1f724fa3891ffbff4fcb2b Mon Sep 17 00:00:00 2001
From: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Date: Thu, 23 Jun 2005 00:09:38 -0700
Subject: [PATCH] kprobes: Temporary disarming of reentrant probe for ppc64

This patch includes ppc64 architecture specific 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>
---
 arch/ppc64/kernel/kprobes.c | 46 +++++++++++++++++++++++++++++++++++++--------
 1 file changed, 38 insertions(+), 8 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/kprobes.c b/arch/ppc64/kernel/kprobes.c
index 8c0920a6d03..782ce3efa2c 100644
--- a/arch/ppc64/kernel/kprobes.c
+++ b/arch/ppc64/kernel/kprobes.c
@@ -36,12 +36,10 @@
 #include <asm/kdebug.h>
 #include <asm/sstep.h>
 
-/* kprobe_status settings */
-#define KPROBE_HIT_ACTIVE	0x00000001
-#define KPROBE_HIT_SS		0x00000002
-
 static struct kprobe *current_kprobe;
 static unsigned long kprobe_status, kprobe_saved_msr;
+static struct kprobe *kprobe_prev;
+static unsigned long kprobe_status_prev, kprobe_saved_msr_prev;
 static struct pt_regs jprobe_saved_regs;
 
 int arch_prepare_kprobe(struct kprobe *p)
@@ -93,6 +91,20 @@ static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 		regs->nip = (unsigned long)&p->ainsn.insn;
 }
 
+static inline void save_previous_kprobe(void)
+{
+	kprobe_prev = current_kprobe;
+	kprobe_status_prev = kprobe_status;
+	kprobe_saved_msr_prev = kprobe_saved_msr;
+}
+
+static inline void restore_previous_kprobe(void)
+{
+	current_kprobe = kprobe_prev;
+	kprobe_status = kprobe_status_prev;
+	kprobe_saved_msr = kprobe_saved_msr_prev;
+}
+
 static inline int kprobe_handler(struct pt_regs *regs)
 {
 	struct kprobe *p;
@@ -111,9 +123,19 @@ static inline int kprobe_handler(struct pt_regs *regs)
 				unlock_kprobes();
 				goto no_kprobe;
 			}
-			arch_disarm_kprobe(p);
-			regs->nip = (unsigned long)p->addr;
-			ret = 1;
+			/* We have reentered the kprobe_handler(), since
+			 * another probe was hit while within the handler.
+			 * We here save the original kprobes variables and
+			 * just single step on the instruction of the new probe
+			 * without calling any user handlers.
+			 */
+			save_previous_kprobe();
+			current_kprobe = p;
+			kprobe_saved_msr = regs->msr;
+			p->nmissed++;
+			prepare_singlestep(p, regs);
+			kprobe_status = KPROBE_REENTER;
+			return 1;
 		} else {
 			p = current_kprobe;
 			if (p->break_handler && p->break_handler(p, regs)) {
@@ -195,13 +217,21 @@ static inline int post_kprobe_handler(struct pt_regs *regs)
 	if (!kprobe_running())
 		return 0;
 
-	if (current_kprobe->post_handler)
+	if ((kprobe_status != KPROBE_REENTER) && current_kprobe->post_handler) {
+		kprobe_status = KPROBE_HIT_SSDONE;
 		current_kprobe->post_handler(current_kprobe, regs, 0);
+	}
 
 	resume_execution(current_kprobe, regs);
 	regs->msr |= kprobe_saved_msr;
 
+	/*Restore back the original saved kprobes variables and continue. */
+	if (kprobe_status == KPROBE_REENTER) {
+		restore_previous_kprobe();
+		goto out;
+	}
 	unlock_kprobes();
+out:
 	preempt_enable_no_resched();
 
 	/*
-- 
cgit v1.2.3-70-g09d2


From e539c2331414e73a5a1b79fb57369d79447c1cf8 Mon Sep 17 00:00:00 2001
From: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Date: Thu, 23 Jun 2005 00:09:39 -0700
Subject: [PATCH] kprobes: Temporary disarming of reentrant probe for sparc64

This patch includes sparc64 architecture specific changes to support temporary
disarming on reentrancy of probes.

Signed-of-by: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/sparc64/kernel/kprobes.c | 62 ++++++++++++++++++++++++++++++++++---------
 1 file changed, 49 insertions(+), 13 deletions(-)

(limited to 'arch')

diff --git a/arch/sparc64/kernel/kprobes.c b/arch/sparc64/kernel/kprobes.c
index d67195ba3fa..bdac631cf01 100644
--- a/arch/sparc64/kernel/kprobes.c
+++ b/arch/sparc64/kernel/kprobes.c
@@ -65,19 +65,40 @@ void arch_remove_kprobe(struct kprobe *p)
 {
 }
 
-/* kprobe_status settings */
-#define KPROBE_HIT_ACTIVE	0x00000001
-#define KPROBE_HIT_SS		0x00000002
-
 static struct kprobe *current_kprobe;
 static unsigned long current_kprobe_orig_tnpc;
 static unsigned long current_kprobe_orig_tstate_pil;
 static unsigned int kprobe_status;
+static struct kprobe *kprobe_prev;
+static unsigned long kprobe_orig_tnpc_prev;
+static unsigned long kprobe_orig_tstate_pil_prev;
+static unsigned int kprobe_status_prev;
 
-static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
+static inline void save_previous_kprobe(void)
+{
+	kprobe_status_prev = kprobe_status;
+	kprobe_orig_tnpc_prev = current_kprobe_orig_tnpc;
+	kprobe_orig_tstate_pil_prev = current_kprobe_orig_tstate_pil;
+	kprobe_prev = current_kprobe;
+}
+
+static inline void restore_previous_kprobe(void)
+{
+	kprobe_status = kprobe_status_prev;
+	current_kprobe_orig_tnpc = kprobe_orig_tnpc_prev;
+	current_kprobe_orig_tstate_pil = kprobe_orig_tstate_pil_prev;
+	current_kprobe = kprobe_prev;
+}
+
+static inline void set_current_kprobe(struct kprobe *p, struct pt_regs *regs)
 {
 	current_kprobe_orig_tnpc = regs->tnpc;
 	current_kprobe_orig_tstate_pil = (regs->tstate & TSTATE_PIL);
+	current_kprobe = p;
+}
+
+static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
+{
 	regs->tstate |= TSTATE_PIL;
 
 	/*single step inline, if it a breakpoint instruction*/
@@ -110,12 +131,18 @@ static int kprobe_handler(struct pt_regs *regs)
 				unlock_kprobes();
 				goto no_kprobe;
 			}
-			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;
+			/* We have reentered the kprobe_handler(), since
+			 * another probe was hit while within the handler.
+			 * We here save the original kprobes variables and
+			 * just single step on the instruction of the new probe
+			 * without calling any user handlers.
+			 */
+			save_previous_kprobe();
+			set_current_kprobe(p, regs);
+			p->nmissed++;
+			kprobe_status = KPROBE_REENTER;
+			prepare_singlestep(p, regs);
+			return 1;
 		} else {
 			p = current_kprobe;
 			if (p->break_handler && p->break_handler(p, regs))
@@ -143,8 +170,8 @@ static int kprobe_handler(struct pt_regs *regs)
 		goto no_kprobe;
 	}
 
+	set_current_kprobe(p, regs);
 	kprobe_status = KPROBE_HIT_ACTIVE;
-	current_kprobe = p;
 	if (p->pre_handler && p->pre_handler(p, regs))
 		return 1;
 
@@ -250,12 +277,20 @@ static inline int post_kprobe_handler(struct pt_regs *regs)
 	if (!kprobe_running())
 		return 0;
 
-	if (current_kprobe->post_handler)
+	if ((kprobe_status != KPROBE_REENTER) && current_kprobe->post_handler) {
+		kprobe_status = KPROBE_HIT_SSDONE;
 		current_kprobe->post_handler(current_kprobe, regs, 0);
+	}
 
 	resume_execution(current_kprobe, regs);
 
+	/*Restore back the original saved kprobes variables and continue. */
+	if (kprobe_status == KPROBE_REENTER) {
+		restore_previous_kprobe();
+		goto out;
+	}
 	unlock_kprobes();
+out:
 	preempt_enable_no_resched();
 
 	return 1;
@@ -397,3 +432,4 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 	}
 	return 0;
 }
+
-- 
cgit v1.2.3-70-g09d2


From 852caccc89d3883522e87a91bfa89fd9c9cfe15a Mon Sep 17 00:00:00 2001
From: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Date: Thu, 23 Jun 2005 00:09:40 -0700
Subject: [PATCH] Kprobes/ia64: temporary disarming of reentrant probe

This patch includes IA64 architecture specific changes(ported form i386) to
support temporary disarming on reentrancy of probes.

In case of reentrancy we single step without calling user handler.

Signed-of-by: Anil S Keshavamurth <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 | 49 ++++++++++++++++++++++++++++++++++++++++------
 1 file changed, 43 insertions(+), 6 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index 41e80b42d3f..5978823d5c6 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -41,8 +41,8 @@ extern void jprobe_inst_return(void);
 #define KPROBE_HIT_ACTIVE	0x00000001
 #define KPROBE_HIT_SS		0x00000002
 
-static struct kprobe *current_kprobe;
-static unsigned long kprobe_status;
+static struct kprobe *current_kprobe, *kprobe_prev;
+static unsigned long kprobe_status, kprobe_status_prev;
 static struct pt_regs jprobe_saved_regs;
 
 enum instruction_type {A, I, M, F, B, L, X, u};
@@ -273,6 +273,23 @@ static int valid_kprobe_addr(int template, int slot, unsigned long addr)
 	return 0;
 }
 
+static inline void save_previous_kprobe(void)
+{
+	kprobe_prev = current_kprobe;
+	kprobe_status_prev = kprobe_status;
+}
+
+static inline void restore_previous_kprobe(void)
+{
+	current_kprobe = kprobe_prev;
+	kprobe_status = kprobe_status_prev;
+}
+
+static inline void set_current_kprobe(struct kprobe *p)
+{
+	current_kprobe = p;
+}
+
 int arch_prepare_kprobe(struct kprobe *p)
 {
 	unsigned long addr = (unsigned long) p->addr;
@@ -436,8 +453,18 @@ static int pre_kprobes_handler(struct die_args *args)
 				unlock_kprobes();
 				goto no_kprobe;
 			}
-			arch_disarm_kprobe(p);
-			ret = 1;
+			/* We have reentered the pre_kprobe_handler(), since
+			 * another probe was hit while within the handler.
+			 * We here save the original kprobes variables and
+			 * just single step on the instruction of the new probe
+			 * without calling any user handlers.
+			 */
+			save_previous_kprobe();
+			set_current_kprobe(p);
+			p->nmissed++;
+			prepare_ss(p, regs);
+			kprobe_status = KPROBE_REENTER;
+			return 1;
 		} else if (args->err == __IA64_BREAK_JPROBE) {
 			/*
 			 * jprobe instrumented function just completed
@@ -460,7 +487,7 @@ static int pre_kprobes_handler(struct die_args *args)
 	}
 
 	kprobe_status = KPROBE_HIT_ACTIVE;
-	current_kprobe = p;
+	set_current_kprobe(p);
 
 	if (p->pre_handler && p->pre_handler(p, regs))
 		/*
@@ -485,12 +512,22 @@ static int post_kprobes_handler(struct pt_regs *regs)
 	if (!kprobe_running())
 		return 0;
 
-	if (current_kprobe->post_handler)
+	if ((kprobe_status != KPROBE_REENTER) && current_kprobe->post_handler) {
+		kprobe_status = KPROBE_HIT_SSDONE;
 		current_kprobe->post_handler(current_kprobe, regs, 0);
+	}
 
 	resume_execution(current_kprobe, regs);
 
+	/*Restore back the original saved kprobes variables and continue. */
+	if (kprobe_status == KPROBE_REENTER) {
+		restore_previous_kprobe();
+		goto out;
+	}
+
 	unlock_kprobes();
+
+out:
 	preempt_enable_no_resched();
 	return 1;
 }
-- 
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 'arch')

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 30aaa80885df5d1518d498a450e73282cea8d8b1 Mon Sep 17 00:00:00 2001
From: William Lee Irwin III <wli@holomorphy.com>
Date: Thu, 23 Jun 2005 00:10:18 -0700
Subject: [PATCH] use drivers/Kconfig for sparc32

Kconfig is spitting out massive numbers of errors and so on.  This patch
switches arch/sparc/Kconfig to use drivers/Kconfig so those stop.

Signed-off-by: William Irwin <wli@holomorphy.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/sparc/Kconfig | 16 +++++-----------
 1 file changed, 5 insertions(+), 11 deletions(-)

(limited to 'arch')

diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 310d3f0c445..262e13d086f 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -264,7 +264,11 @@ config SUNOS_EMUL
 	  want to run SunOS binaries on an Ultra you must also say Y to
 	  "Kernel support for 32-bit a.out binaries" above.
 
-source "drivers/parport/Kconfig"
+source "mm/Kconfig"
+
+endmenu
+
+source "drivers/Kconfig"
 
 config PRINTER
 	tristate "Parallel printer support"
@@ -374,18 +378,8 @@ config UNIX98_PTY_COUNT
 
 endmenu
 
-source "drivers/input/Kconfig"
-
 source "fs/Kconfig"
 
-source "sound/Kconfig"
-
-source "drivers/usb/Kconfig"
-
-source "drivers/infiniband/Kconfig"
-
-source "drivers/char/watchdog/Kconfig"
-
 source "arch/sparc/Kconfig.debug"
 
 source "security/Kconfig"
-- 
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 'arch')

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 c1241c4c3a1507d76c7b987130f2f02f53ecc09f Mon Sep 17 00:00:00 2001
From: Nicolas Pitre <nico@cam.org>
Date: Thu, 23 Jun 2005 21:56:46 +0100
Subject: [PATCH] ARM: 2722/1: remove reliance on udivdi3 for nwfpe

Patch from Nicolas Pitre

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/nwfpe/softfloat-macros | 22 ++++++++++++++++++----
 arch/arm/nwfpe/softfloat.c      | 12 ++++++++++--
 2 files changed, 28 insertions(+), 6 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/nwfpe/softfloat-macros b/arch/arm/nwfpe/softfloat-macros
index 5469989f2c5..5a060f95a58 100644
--- a/arch/arm/nwfpe/softfloat-macros
+++ b/arch/arm/nwfpe/softfloat-macros
@@ -563,8 +563,14 @@ static bits64 estimateDiv128To64( bits64 a0, bits64 a1, bits64 b )
     bits64 rem0, rem1, term0, term1;
     bits64 z;
     if ( b <= a0 ) return LIT64( 0xFFFFFFFFFFFFFFFF );
-    b0 = b>>32;
-    z = ( b0<<32 <= a0 ) ? LIT64( 0xFFFFFFFF00000000 ) : ( a0 / b0 )<<32;
+    b0 = b>>32;  /* hence b0 is 32 bits wide now */
+    if ( b0<<32 <= a0 ) {
+        z = LIT64( 0xFFFFFFFF00000000 );
+    }  else {
+        z = a0;
+        do_div( z, b0 );
+        z <<= 32;
+    }
     mul64To128( b, z, &term0, &term1 );
     sub128( a0, a1, term0, term1, &rem0, &rem1 );
     while ( ( (sbits64) rem0 ) < 0 ) {
@@ -573,7 +579,12 @@ static bits64 estimateDiv128To64( bits64 a0, bits64 a1, bits64 b )
         add128( rem0, rem1, b0, b1, &rem0, &rem1 );
     }
     rem0 = ( rem0<<32 ) | ( rem1>>32 );
-    z |= ( b0<<32 <= rem0 ) ? 0xFFFFFFFF : rem0 / b0;
+    if ( b0<<32 <= rem0 ) {
+        z |= 0xFFFFFFFF;
+    } else {
+        do_div( rem0, b0 );
+        z |= rem0;
+    }
     return z;
 
 }
@@ -601,6 +612,7 @@ static bits32 estimateSqrt32( int16 aExp, bits32 a )
     };
     int8 index;
     bits32 z;
+    bits64 A;
 
     index = ( a>>27 ) & 15;
     if ( aExp & 1 ) {
@@ -614,7 +626,9 @@ static bits32 estimateSqrt32( int16 aExp, bits32 a )
         z = ( 0x20000 <= z ) ? 0xFFFF8000 : ( z<<15 );
         if ( z <= a ) return (bits32) ( ( (sbits32) a )>>1 );
     }
-    return ( (bits32) ( ( ( (bits64) a )<<31 ) / z ) ) + ( z>>1 );
+    A = ( (bits64) a )<<31;
+    do_div( A, z );
+    return ( (bits32) A ) + ( z>>1 );
 
 }
 
diff --git a/arch/arm/nwfpe/softfloat.c b/arch/arm/nwfpe/softfloat.c
index 9d743ae2906..e038dd3be9b 100644
--- a/arch/arm/nwfpe/softfloat.c
+++ b/arch/arm/nwfpe/softfloat.c
@@ -28,6 +28,8 @@ this code that are retained.
 ===============================================================================
 */
 
+#include <asm/div64.h>
+
 #include "fpa11.h"
 //#include "milieu.h"
 //#include "softfloat.h"
@@ -1331,7 +1333,11 @@ float32 float32_div( float32 a, float32 b )
         aSig >>= 1;
         ++zExp;
     }
-    zSig = ( ( (bits64) aSig )<<32 ) / bSig;
+    {
+        bits64 tmp = ( (bits64) aSig )<<32;
+        do_div( tmp, bSig );
+        zSig = tmp;
+    }
     if ( ( zSig & 0x3F ) == 0 ) {
         zSig |= ( ( (bits64) bSig ) * zSig != ( (bits64) aSig )<<32 );
     }
@@ -1397,7 +1403,9 @@ float32 float32_rem( float32 a, float32 b )
         q = ( bSig <= aSig );
         if ( q ) aSig -= bSig;
         if ( 0 < expDiff ) {
-            q = ( ( (bits64) aSig )<<32 ) / bSig;
+            bits64 tmp = ( (bits64) aSig )<<32;
+            do_div( tmp, bSig );
+            q = tmp;
             q >>= 32 - expDiff;
             bSig >>= 2;
             aSig = ( ( aSig>>1 )<<( expDiff - 1 ) ) - bSig * q;
-- 
cgit v1.2.3-70-g09d2


From d97a666f36cf051e1b1c60505be3d6e9b51f785f Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Thu, 23 Jun 2005 21:56:47 +0100
Subject: [PATCH] ARM: 2729/1: DM9000 platform support for S3C2410 machines
 (BAST, VR1000)

Patch from Ben Dooks

Add platform_device information for DM9000 chip(s) on the
Simtec BAST and the VR1000 board.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-s3c2410/mach-bast.c   | 49 ++++++++++++++++++++---
 arch/arm/mach-s3c2410/mach-vr1000.c | 77 +++++++++++++++++++++++++++++++++++--
 2 files changed, 117 insertions(+), 9 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/mach-s3c2410/mach-bast.c b/arch/arm/mach-s3c2410/mach-bast.c
index 3bb97eb6e69..f3e970039b6 100644
--- a/arch/arm/mach-s3c2410/mach-bast.c
+++ b/arch/arm/mach-s3c2410/mach-bast.c
@@ -26,6 +26,7 @@
  *     03-Mar-2005 BJD  Ensured that bast-cpld.h is included
  *     10-Mar-2005 LCVR Changed S3C2410_VA to S3C24XX_VA
  *     14-Mar-2006 BJD  Updated for __iomem changes
+ *     22-Jun-2006 BJD  Added DM9000 platform information
 */
 
 #include <linux/kernel.h>
@@ -35,6 +36,7 @@
 #include <linux/timer.h>
 #include <linux/init.h>
 #include <linux/device.h>
+#include <linux/dm9000.h>
 
 #include <asm/mach/arch.h>
 #include <asm/mach/map.h>
@@ -53,6 +55,7 @@
 #include <asm/arch/regs-serial.h>
 #include <asm/arch/regs-gpio.h>
 #include <asm/arch/regs-mem.h>
+#include <asm/arch/regs-lcd.h>
 #include <asm/arch/nand.h>
 
 #include <linux/mtd/mtd.h>
@@ -112,7 +115,6 @@ static struct map_desc bast_iodesc[] __initdata = {
   { VA_C2(BAST_VA_ISAMEM),  PA_CS2(BAST_PA_ISAMEM),   SZ_16M, MT_DEVICE },
   { VA_C2(BAST_VA_ASIXNET), PA_CS3(BAST_PA_ASIXNET),  SZ_1M,  MT_DEVICE },
   { VA_C2(BAST_VA_SUPERIO), PA_CS2(BAST_PA_SUPERIO),  SZ_1M,  MT_DEVICE },
-  { VA_C2(BAST_VA_DM9000),  PA_CS2(BAST_PA_DM9000),   SZ_1M,  MT_DEVICE },
   { VA_C2(BAST_VA_IDEPRI),  PA_CS3(BAST_PA_IDEPRI),   SZ_1M,  MT_DEVICE },
   { VA_C2(BAST_VA_IDESEC),  PA_CS3(BAST_PA_IDESEC),   SZ_1M,  MT_DEVICE },
   { VA_C2(BAST_VA_IDEPRIAUX), PA_CS3(BAST_PA_IDEPRIAUX), SZ_1M, MT_DEVICE },
@@ -123,7 +125,6 @@ static struct map_desc bast_iodesc[] __initdata = {
   { VA_C3(BAST_VA_ISAMEM),  PA_CS3(BAST_PA_ISAMEM),   SZ_16M, MT_DEVICE },
   { VA_C3(BAST_VA_ASIXNET), PA_CS3(BAST_PA_ASIXNET),  SZ_1M,  MT_DEVICE },
   { VA_C3(BAST_VA_SUPERIO), PA_CS3(BAST_PA_SUPERIO),  SZ_1M,  MT_DEVICE },
-  { VA_C3(BAST_VA_DM9000),  PA_CS3(BAST_PA_DM9000),   SZ_1M,  MT_DEVICE },
   { VA_C3(BAST_VA_IDEPRI),  PA_CS3(BAST_PA_IDEPRI),   SZ_1M,  MT_DEVICE },
   { VA_C3(BAST_VA_IDESEC),  PA_CS3(BAST_PA_IDESEC),   SZ_1M,  MT_DEVICE },
   { VA_C3(BAST_VA_IDEPRIAUX), PA_CS3(BAST_PA_IDEPRIAUX), SZ_1M, MT_DEVICE },
@@ -134,7 +135,6 @@ static struct map_desc bast_iodesc[] __initdata = {
   { VA_C4(BAST_VA_ISAMEM),  PA_CS4(BAST_PA_ISAMEM),   SZ_16M, MT_DEVICE },
   { VA_C4(BAST_VA_ASIXNET), PA_CS5(BAST_PA_ASIXNET),  SZ_1M,  MT_DEVICE },
   { VA_C4(BAST_VA_SUPERIO), PA_CS4(BAST_PA_SUPERIO),  SZ_1M,  MT_DEVICE },
-  { VA_C4(BAST_VA_DM9000),  PA_CS4(BAST_PA_DM9000),   SZ_1M,  MT_DEVICE },
   { VA_C4(BAST_VA_IDEPRI),  PA_CS5(BAST_PA_IDEPRI),   SZ_1M,  MT_DEVICE },
   { VA_C4(BAST_VA_IDESEC),  PA_CS5(BAST_PA_IDESEC),   SZ_1M,  MT_DEVICE },
   { VA_C4(BAST_VA_IDEPRIAUX), PA_CS5(BAST_PA_IDEPRIAUX), SZ_1M, MT_DEVICE },
@@ -145,7 +145,6 @@ static struct map_desc bast_iodesc[] __initdata = {
   { VA_C5(BAST_VA_ISAMEM),  PA_CS5(BAST_PA_ISAMEM),   SZ_16M, MT_DEVICE },
   { VA_C5(BAST_VA_ASIXNET), PA_CS5(BAST_PA_ASIXNET),  SZ_1M,  MT_DEVICE },
   { VA_C5(BAST_VA_SUPERIO), PA_CS5(BAST_PA_SUPERIO),  SZ_1M,  MT_DEVICE },
-  { VA_C5(BAST_VA_DM9000),  PA_CS5(BAST_PA_DM9000),   SZ_1M,  MT_DEVICE },
   { VA_C5(BAST_VA_IDEPRI),  PA_CS5(BAST_PA_IDEPRI),   SZ_1M,  MT_DEVICE },
   { VA_C5(BAST_VA_IDESEC),  PA_CS5(BAST_PA_IDESEC),   SZ_1M,  MT_DEVICE },
   { VA_C5(BAST_VA_IDEPRIAUX), PA_CS5(BAST_PA_IDEPRIAUX), SZ_1M, MT_DEVICE },
@@ -313,6 +312,45 @@ static struct s3c2410_platform_nand bast_nand_info = {
 	.select_chip	= bast_nand_select,
 };
 
+/* DM9000 */
+
+static struct resource bast_dm9k_resource[] = {
+	[0] = {
+		.start = S3C2410_CS5 + BAST_PA_DM9000,
+		.end   = S3C2410_CS5 + BAST_PA_DM9000 + 3,
+		.flags = IORESOURCE_MEM
+	},
+	[1] = {
+		.start = S3C2410_CS5 + BAST_PA_DM9000 + 0x40,
+		.end   = S3C2410_CS5 + BAST_PA_DM9000 + 0x40 + 0x3f,
+		.flags = IORESOURCE_MEM
+	},
+	[2] = {
+		.start = IRQ_DM9000,
+		.end   = IRQ_DM9000,
+		.flags = IORESOURCE_IRQ
+	}
+
+};
+
+/* for the moment we limit ourselves to 16bit IO until some
+ * better IO routines can be written and tested
+*/
+
+struct dm9000_plat_data bast_dm9k_platdata = {
+	.flags		= DM9000_PLATF_16BITONLY
+};
+
+static struct platform_device bast_device_dm9k = {
+	.name		= "dm9000",
+	.id		= 0,
+	.num_resources	= ARRAY_SIZE(bast_dm9k_resource),
+	.resource	= bast_dm9k_resource,
+	.dev		= {
+		.platform_data = &bast_dm9k_platdata,
+	}
+};
+
 
 /* Standard BAST devices */
 
@@ -324,7 +362,8 @@ static struct platform_device *bast_devices[] __initdata = {
 	&s3c_device_iis,
  	&s3c_device_rtc,
 	&s3c_device_nand,
-	&bast_device_nor
+	&bast_device_nor,
+	&bast_device_dm9k,
 };
 
 static struct clk *bast_clocks[] = {
diff --git a/arch/arm/mach-s3c2410/mach-vr1000.c b/arch/arm/mach-s3c2410/mach-vr1000.c
index 5512146b1ce..76be074944a 100644
--- a/arch/arm/mach-s3c2410/mach-vr1000.c
+++ b/arch/arm/mach-s3c2410/mach-vr1000.c
@@ -27,6 +27,7 @@
  *     10-Feb-2005 BJD  Added power-off capability
  *     10-Mar-2005 LCVR Changed S3C2410_VA to S3C24XX_VA
  *     14-Mar-2006 BJD  void __iomem fixes
+ *     22-Jun-2006 BJD  Added DM9000 platform information
 */
 
 #include <linux/kernel.h>
@@ -35,6 +36,7 @@
 #include <linux/list.h>
 #include <linux/timer.h>
 #include <linux/init.h>
+#include <linux/dm9000.h>
 
 #include <linux/serial.h>
 #include <linux/tty.h>
@@ -98,28 +100,24 @@ static struct map_desc vr1000_iodesc[] __initdata = {
    * are only 8bit */
 
   /* slow, byte */
-  { VA_C2(VR1000_VA_DM9000),  PA_CS2(VR1000_PA_DM9000),	  SZ_1M,  MT_DEVICE },
   { VA_C2(VR1000_VA_IDEPRI),  PA_CS3(VR1000_PA_IDEPRI),	  SZ_1M,  MT_DEVICE },
   { VA_C2(VR1000_VA_IDESEC),  PA_CS3(VR1000_PA_IDESEC),	  SZ_1M,  MT_DEVICE },
   { VA_C2(VR1000_VA_IDEPRIAUX), PA_CS3(VR1000_PA_IDEPRIAUX), SZ_1M, MT_DEVICE },
   { VA_C2(VR1000_VA_IDESECAUX), PA_CS3(VR1000_PA_IDESECAUX), SZ_1M, MT_DEVICE },
 
   /* slow, word */
-  { VA_C3(VR1000_VA_DM9000),  PA_CS3(VR1000_PA_DM9000),	  SZ_1M,  MT_DEVICE },
   { VA_C3(VR1000_VA_IDEPRI),  PA_CS3(VR1000_PA_IDEPRI),	  SZ_1M,  MT_DEVICE },
   { VA_C3(VR1000_VA_IDESEC),  PA_CS3(VR1000_PA_IDESEC),	  SZ_1M,  MT_DEVICE },
   { VA_C3(VR1000_VA_IDEPRIAUX), PA_CS3(VR1000_PA_IDEPRIAUX), SZ_1M, MT_DEVICE },
   { VA_C3(VR1000_VA_IDESECAUX), PA_CS3(VR1000_PA_IDESECAUX), SZ_1M, MT_DEVICE },
 
   /* fast, byte */
-  { VA_C4(VR1000_VA_DM9000),  PA_CS4(VR1000_PA_DM9000),	  SZ_1M,  MT_DEVICE },
   { VA_C4(VR1000_VA_IDEPRI),  PA_CS5(VR1000_PA_IDEPRI),	  SZ_1M,  MT_DEVICE },
   { VA_C4(VR1000_VA_IDESEC),  PA_CS5(VR1000_PA_IDESEC),	  SZ_1M,  MT_DEVICE },
   { VA_C4(VR1000_VA_IDEPRIAUX), PA_CS5(VR1000_PA_IDEPRIAUX), SZ_1M, MT_DEVICE },
   { VA_C4(VR1000_VA_IDESECAUX), PA_CS5(VR1000_PA_IDESECAUX), SZ_1M, MT_DEVICE },
 
   /* fast, word */
-  { VA_C5(VR1000_VA_DM9000),  PA_CS5(VR1000_PA_DM9000),	  SZ_1M,  MT_DEVICE },
   { VA_C5(VR1000_VA_IDEPRI),  PA_CS5(VR1000_PA_IDEPRI),	  SZ_1M,  MT_DEVICE },
   { VA_C5(VR1000_VA_IDESEC),  PA_CS5(VR1000_PA_IDESEC),	  SZ_1M,  MT_DEVICE },
   { VA_C5(VR1000_VA_IDEPRIAUX), PA_CS5(VR1000_PA_IDEPRIAUX), SZ_1M, MT_DEVICE },
@@ -246,6 +244,74 @@ static struct platform_device vr1000_nor = {
 	.resource	= vr1000_nor_resource,
 };
 
+/* DM9000 ethernet devices */
+
+static struct resource vr1000_dm9k0_resource[] = {
+	[0] = {
+		.start = S3C2410_CS5 + VR1000_PA_DM9000,
+		.end   = S3C2410_CS5 + VR1000_PA_DM9000 + 3,
+		.flags = IORESOURCE_MEM
+	},
+	[1] = {
+		.start = S3C2410_CS5 + VR1000_PA_DM9000 + 0x40,
+		.end   = S3C2410_CS5 + VR1000_PA_DM9000 + 0x7f,
+		.flags = IORESOURCE_MEM
+	},
+	[2] = {
+		.start = IRQ_VR1000_DM9000A,
+		.end   = IRQ_VR1000_DM9000A,
+		.flags = IORESOURCE_IRQ
+	}
+
+};
+
+static struct resource vr1000_dm9k1_resource[] = {
+	[0] = {
+		.start = S3C2410_CS5 + VR1000_PA_DM9000 + 0x80,
+		.end   = S3C2410_CS5 + VR1000_PA_DM9000 + 0x83,
+		.flags = IORESOURCE_MEM
+	},
+	[1] = {
+		.start = S3C2410_CS5 + VR1000_PA_DM9000 + 0xC0,
+		.end   = S3C2410_CS5 + VR1000_PA_DM9000 + 0xFF,
+		.flags = IORESOURCE_MEM
+	},
+	[2] = {
+		.start = IRQ_VR1000_DM9000N,
+		.end   = IRQ_VR1000_DM9000N,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+/* for the moment we limit ourselves to 16bit IO until some
+ * better IO routines can be written and tested
+*/
+
+struct dm9000_plat_data vr1000_dm9k_platdata = {
+	.flags		= DM9000_PLATF_16BITONLY,
+};
+
+static struct platform_device vr1000_dm9k0 = {
+	.name		= "dm9000",
+	.id		= 0,
+	.num_resources	= ARRAY_SIZE(vr1000_dm9k0_resource),
+	.resource	= vr1000_dm9k0_resource,
+	.dev		= {
+		.platform_data = &vr1000_dm9k_platdata,
+	}
+};
+
+static struct platform_device vr1000_dm9k1 = {
+	.name		= "dm9000",
+	.id		= 1,
+	.num_resources	= ARRAY_SIZE(vr1000_dm9k1_resource),
+	.resource	= vr1000_dm9k1_resource,
+	.dev		= {
+		.platform_data = &vr1000_dm9k_platdata,
+	}
+};
+
+/* devices for this board */
 
 static struct platform_device *vr1000_devices[] __initdata = {
 	&s3c_device_usb,
@@ -253,8 +319,11 @@ static struct platform_device *vr1000_devices[] __initdata = {
 	&s3c_device_wdt,
 	&s3c_device_i2c,
 	&s3c_device_iis,
+	&s3c_device_adc,
 	&serial_device,
 	&vr1000_nor,
+	&vr1000_dm9k0,
+	&vr1000_dm9k1
 };
 
 static struct clk *vr1000_clocks[] = {
-- 
cgit v1.2.3-70-g09d2


From 691027b91be99413dc60fab0902b366434555015 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Thu, 23 Jun 2005 21:56:48 +0100
Subject: [PATCH] ARM: 2730/1: S3C2410 default configuration update

Patch from Ben Dooks

Add support for the DM9000 and bring default configuration
up-to-date with the latest 2.6.12 kernel release

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/configs/s3c2410_defconfig | 27 ++++++++++++++++++---------
 1 file changed, 18 insertions(+), 9 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/configs/s3c2410_defconfig b/arch/arm/configs/s3c2410_defconfig
index 2a63fb27719..98b72ff3883 100644
--- a/arch/arm/configs/s3c2410_defconfig
+++ b/arch/arm/configs/s3c2410_defconfig
@@ -1,14 +1,13 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.12-rc1-bk2
-# Sun Mar 27 17:47:45 2005
+# Linux kernel version: 2.6.12-git4
+# Wed Jun 22 15:56:42 2005
 #
 CONFIG_ARM=y
 CONFIG_MMU=y
 CONFIG_UID16=y
 CONFIG_RWSEM_GENERIC_SPINLOCK=y
 CONFIG_GENERIC_CALIBRATE_DELAY=y
-CONFIG_GENERIC_IOMAP=y
 
 #
 # Code maturity level options
@@ -17,6 +16,7 @@ CONFIG_EXPERIMENTAL=y
 # CONFIG_CLEAN_COMPILE is not set
 CONFIG_BROKEN=y
 CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -35,6 +35,8 @@ CONFIG_KOBJECT_UEVENT=y
 CONFIG_KALLSYMS=y
 # CONFIG_KALLSYMS_ALL is not set
 # CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
 CONFIG_BASE_FULL=y
 CONFIG_FUTEX=y
 CONFIG_EPOLL=y
@@ -81,6 +83,7 @@ CONFIG_ARCH_S3C2410=y
 # CONFIG_ARCH_VERSATILE is not set
 # CONFIG_ARCH_IMX is not set
 # CONFIG_ARCH_H720X is not set
+# CONFIG_ARCH_AAEC2000 is not set
 
 #
 # S3C24XX Implementations
@@ -134,6 +137,7 @@ CONFIG_CPU_TLB_V4WBI=y
 #
 # Bus support
 #
+CONFIG_ISA_DMA_API=y
 
 #
 # PCCARD (PCMCIA/CardBus) support
@@ -143,7 +147,9 @@ CONFIG_CPU_TLB_V4WBI=y
 #
 # Kernel Features
 #
+# CONFIG_SMP is not set
 # CONFIG_PREEMPT is not set
+# CONFIG_DISCONTIGMEM is not set
 CONFIG_ALIGNMENT_TRAP=y
 
 #
@@ -297,7 +303,6 @@ CONFIG_PARPORT_1284=y
 #
 # Block devices
 #
-# CONFIG_BLK_DEV_FD is not set
 # CONFIG_PARIDE is not set
 # CONFIG_BLK_DEV_COW_COMMON is not set
 CONFIG_BLK_DEV_LOOP=y
@@ -359,6 +364,7 @@ CONFIG_BLK_DEV_IDE_BAST=y
 #
 # Fusion MPT device support
 #
+# CONFIG_FUSION is not set
 
 #
 # IEEE 1394 (FireWire) support
@@ -378,10 +384,11 @@ CONFIG_NET=y
 # Networking options
 #
 # CONFIG_PACKET is not set
-# CONFIG_NETLINK_DEV is not set
 CONFIG_UNIX=y
 # CONFIG_NET_KEY is not set
 CONFIG_INET=y
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_FIB_TRIE is not set
 # CONFIG_IP_MULTICAST is not set
 # CONFIG_IP_ADVANCED_ROUTER is not set
 CONFIG_IP_PNP=y
@@ -443,8 +450,9 @@ CONFIG_NETDEVICES=y
 # Ethernet (10 or 100Mbit)
 #
 CONFIG_NET_ETHERNET=y
-# CONFIG_MII is not set
+CONFIG_MII=m
 # CONFIG_SMC91X is not set
+CONFIG_DM9000=m
 
 #
 # Ethernet (1000 Mbit)
@@ -521,7 +529,6 @@ CONFIG_SERIO_SERPORT=y
 CONFIG_SERIO_LIBPS2=y
 # CONFIG_SERIO_RAW is not set
 # CONFIG_GAMEPORT is not set
-CONFIG_SOUND_GAMEPORT=y
 
 #
 # Character devices
@@ -605,7 +612,6 @@ CONFIG_S3C2410_RTC=y
 #
 # TPM devices
 #
-# CONFIG_TCG_TPM is not set
 
 #
 # I2C support
@@ -654,6 +660,7 @@ CONFIG_SENSORS_LM78=m
 CONFIG_SENSORS_LM85=m
 # 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
@@ -665,6 +672,7 @@ CONFIG_SENSORS_LM85=m
 #
 # Other I2C Chip support
 #
+# CONFIG_SENSORS_DS1337 is not set
 CONFIG_SENSORS_EEPROM=m
 # CONFIG_SENSORS_PCF8574 is not set
 # CONFIG_SENSORS_PCF8591 is not set
@@ -696,8 +704,10 @@ CONFIG_FB=y
 # CONFIG_FB_CFB_COPYAREA is not set
 # CONFIG_FB_CFB_IMAGEBLIT is not set
 # CONFIG_FB_SOFT_CURSOR is not set
+# CONFIG_FB_MACMODES is not set
 CONFIG_FB_MODE_HELPERS=y
 # CONFIG_FB_TILEBLITTING is not set
+# CONFIG_FB_S1D13XXX is not set
 # CONFIG_FB_VIRTUAL is not set
 
 #
@@ -782,7 +792,6 @@ CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
 #
 CONFIG_PROC_FS=y
 CONFIG_SYSFS=y
-# CONFIG_DEVFS_FS is not set
 # CONFIG_DEVPTS_FS_XATTR is not set
 # CONFIG_TMPFS is not set
 # CONFIG_HUGETLBFS is not set
-- 
cgit v1.2.3-70-g09d2


From 7286aa9b9ab35f20b1ff16d867f4535701df99b5 Mon Sep 17 00:00:00 2001
From: Andrea Arcangeli <andrea@cpushare.com>
Date: Thu, 23 Jun 2005 22:00:45 -0700
Subject: [PATCH] ppc64: fix seccomp with 32-bit userland

The seccomp check has to happen when entering the syscall and not when
exiting it or regs->gpr[0] contains garabge during signal handling in
ppc64_rt_sigreturn (this actually might be a bug too, but an orthogonal
one, since we really have to run the check before invoking the syscall and
not after it).

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/ppc64/kernel/ptrace.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/ptrace.c b/arch/ppc64/kernel/ptrace.c
index 9f8c6087ae5..2993f108d96 100644
--- a/arch/ppc64/kernel/ptrace.c
+++ b/arch/ppc64/kernel/ptrace.c
@@ -305,6 +305,8 @@ static void do_syscall_trace(void)
 
 void do_syscall_trace_enter(struct pt_regs *regs)
 {
+	secure_computing(regs->gpr[0]);
+
 	if (test_thread_flag(TIF_SYSCALL_TRACE)
 	    && (current->ptrace & PT_PTRACED))
 		do_syscall_trace();
@@ -320,8 +322,6 @@ void do_syscall_trace_enter(struct pt_regs *regs)
 
 void do_syscall_trace_leave(struct pt_regs *regs)
 {
-	secure_computing(regs->gpr[0]);
-
 	if (unlikely(current->audit_context))
 		audit_syscall_exit(current, 
 				   (regs->ccr&0x1000)?AUDITSC_FAILURE:AUDITSC_SUCCESS,
-- 
cgit v1.2.3-70-g09d2


From 8e1a6dd2fddcc73c9e933758361e3d9c076c688a Mon Sep 17 00:00:00 2001
From: Chris Zankel <czankel@tensilica.com>
Date: Thu, 23 Jun 2005 22:01:10 -0700
Subject: [PATCH] xtensa: Architecture support for Tensilica Xtensa Part 1

The attached patches provides part 1 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/Kconfig                  | 258 ++++++++++++++
 arch/xtensa/Kconfig.debug            |   7 +
 arch/xtensa/Makefile                 | 102 ++++++
 arch/xtensa/configs/common_defconfig | 662 +++++++++++++++++++++++++++++++++++
 4 files changed, 1029 insertions(+)
 create mode 100644 arch/xtensa/Kconfig
 create mode 100644 arch/xtensa/Kconfig.debug
 create mode 100644 arch/xtensa/Makefile
 create mode 100644 arch/xtensa/configs/common_defconfig

(limited to 'arch')

diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig
new file mode 100644
index 00000000000..3e89767cea7
--- /dev/null
+++ b/arch/xtensa/Kconfig
@@ -0,0 +1,258 @@
+# For a description of the syntax of this configuration file,
+# see Documentation/kbuild/config-language.txt.
+
+mainmenu "Linux/Xtensa Kernel Configuration"
+
+config FRAME_POINTER
+	bool
+	default n
+
+config XTENSA
+	bool
+	default y
+	help
+	  Xtensa processors are 32-bit RISC machines designed by Tensilica
+	  primarily for embedded systems.  These processors are both
+	  configurable and extensible.  The Linux port to the Xtensa
+	  architecture supports all processor configurations and extensions,
+	  with reasonable minimum requirements.  The Xtensa Linux project has
+	  a home page at <http://xtensa.sourceforge.net/>.
+
+config UID16
+	bool
+	default n
+
+config RWSEM_XCHGADD_ALGORITHM
+	bool
+	default y
+
+config HAVE_DEC_LOCK
+	bool
+	default y
+
+config GENERIC_HARDIRQS
+	bool
+	default y
+
+source "init/Kconfig"
+
+menu "Processor type and features"
+
+choice
+	prompt "Xtensa Processor Configuration"
+	default XTENSA_CPU_LINUX_BE
+
+config XTENSA_CPU_LINUX_BE
+	bool "linux_be"
+	---help---
+	  The linux_be processor configuration is the baseline Xtensa
+	  configurations included in this kernel and also used by
+	  binutils, gcc, and gdb. It contains no TIE, no coprocessors,
+	  and the following configuration options:
+
+	  Code Density Option                2 Misc Special Registers
+	  NSA/NSAU Instructions              128-bit Data Bus Width
+	  Processor ID                       8K, 2-way I and D Caches
+	  Zero-Overhead Loops                2 Inst Address Break Registers
+	  Big Endian                         2 Data Address Break Registers
+	  64 General-Purpose Registers       JTAG Interface and Trace Port
+	  17 Interrupts                      MMU w/ TLBs and Autorefill
+	  3 Interrupt Levels                 8 Autorefill Ways (I/D TLBs)
+	  3 Timers                           Unaligned Exceptions
+endchoice
+
+config MMU
+	bool
+	default y
+
+config XTENSA_UNALIGNED_USER
+	bool "Unaligned memory access in use space"
+	---help---
+	   The Xtensa architecture currently does not handle unaligned
+	   memory accesses in hardware but through an exception handler.
+	   Per default, unaligned memory accesses are disabled in user space.
+
+	   Say Y here to enable unaligned memory access in user space.
+
+config PREEMPT
+	bool "Preemptible Kernel"
+	---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.
+           Unfortunately the kernel code has some race conditions if both
+           CONFIG_SMP and CONFIG_PREEMPT are enabled, so this option is
+           currently disabled if you are building an SMP kernel.
+
+           Say Y here if you are building a kernel for a desktop, embedded
+           or real-time system.  Say N if you are unsure.
+
+config MATH_EMULATION
+	bool "Math emulation"
+	help
+	Can we use information of configuration file?
+
+config HIGHMEM
+	bool "High memory support"
+
+endmenu
+
+menu "Platform options"
+
+choice
+	prompt "Xtensa System Type"
+	default XTENSA_PLATFORM_ISS
+
+config XTENSA_PLATFORM_ISS
+	bool "ISS"
+	help
+	  ISS is an acronym for Tensilica's Instruction Set Simulator.
+
+config XTENSA_PLATFORM_XT2000
+	bool "XT2000"
+	help
+	  XT2000 is the name of Tensilica's feature-rich emulation platform.
+	  This hardware is capable of running a full Linux distribution.
+
+endchoice
+
+
+config XTENSA_CALIBRATE_CCOUNT
+	bool "Auto calibration of the CPU clock rate"
+	---help---
+	  On some platforms (XT2000, for example), the CPU clock rate can
+	  vary.  The frequency can be determined, however, by measuring
+	  against a well known, fixed frequency, such as an UART oscillator.
+
+config XTENSA_CPU_CLOCK
+	int "CPU clock rate [MHz]"
+	depends on !XTENSA_CALIBRATE_CCOUNT
+	default "16"
+
+config GENERIC_CALIBRATE_DELAY
+	bool "Auto calibration of the BogoMIPS value"
+	---help---
+	  The BogoMIPS value can easily derived from the CPU frequency.
+
+config CMDLINE_BOOL
+	bool "Default bootloader kernel arguments"
+
+config CMDLINE
+	string "Initial kernel command string"
+	depends on CMDLINE_BOOL
+	default "console=ttyS0,38400 root=/dev/ram"
+	help
+	  On some architectures (EBSA110 and CATS), there is currently no way
+	  for the boot loader to pass arguments to the kernel. For these
+	  architectures, you should supply some command-line options at build
+	  time by entering them here. As a minimum, you should specify the
+	  memory size and the root device (e.g., mem=64M root=/dev/nfs).
+
+config SERIAL_CONSOLE
+	bool
+	depends on XTENSA_PLATFORM_ISS
+	default y
+
+config XTENSA_ISS_NETWORK
+	bool
+	depends on XTENSA_PLATFORM_ISS
+	default y
+
+endmenu
+
+menu "Bus options"
+
+config PCI
+	bool "PCI support" if !XTENSA_PLATFORM_ISS
+	depends on !XTENSA_PLATFORM_ISS
+	default y
+	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
+	  your box. Other bus systems are ISA, EISA, MicroChannel (MCA) or
+	  VESA. If you have PCI, say Y, otherwise N.
+
+	  The PCI-HOWTO, available from
+	  <http://www.linuxdoc.org/docs.html#howto>, contains valuable
+	  information about which PCI hardware does work under Linux and which
+	  doesn't
+
+source "drivers/pci/Kconfig"
+
+config HOTPLUG
+
+	bool "Support for hot-pluggable devices"
+	---help---
+	Say Y here if you want to plug devices into your computer while
+	the system is running, and be able to use them quickly.  In many
+	cases, the devices can likewise be unplugged at any time too.
+
+	One well known example of this is PCMCIA- or PC-cards, credit-card
+	size devices such as network cards, modems or hard drives which are
+	plugged into slots found on all modern laptop computers.  Another
+	example, used on modern desktops as well as laptops, is USB.
+
+	Enable HOTPLUG and KMOD, and build a modular kernel.  Get agent
+	software (at <http://linux-hotplug.sourceforge.net/>) and install it.
+	Then your kernel will automatically call out to a user mode "policy
+	agent" (/sbin/hotplug) to load modules and set up software needed
+	to use devices as you hotplug them.
+
+source "drivers/pcmcia/Kconfig"
+
+source "drivers/pci/hotplug/Kconfig"
+
+endmenu
+
+menu "Exectuable file formats"
+
+# only elf supported
+config KCORE_ELF
+        bool
+        depends on PROC_FS
+        default y
+        help
+          If you enabled support for /proc file system then the file
+          /proc/kcore will contain the kernel core image in ELF format. This
+          can be used in gdb:
+
+          $ cd /usr/src/linux ; gdb vmlinux /proc/kcore
+
+          This is especially useful if you have compiled the kernel with the
+          "-g" option to preserve debugging information. It is mainly used
+	  for examining kernel data structures on the live kernel.
+
+source "fs/Kconfig.binfmt"
+
+endmenu
+
+source "drivers/Kconfig"
+
+source "fs/Kconfig"
+
+menu "Xtensa initrd options"
+	depends on BLK_DEV_INITRD
+
+	config EMBEDDED_RAMDISK
+	bool "Embed root filesystem ramdisk into the kernel"
+
+config EMBEDDED_RAMDISK_IMAGE
+	string "Filename of gziped ramdisk image"
+	depends on EMBEDDED_RAMDISK
+	default "ramdisk.gz"
+	help
+	  This is the filename of the ramdisk image to be built into the
+	  kernel.  Relative pathnames are relative to arch/xtensa/boot/ramdisk/.
+	  The ramdisk image is not part of the kernel distribution; you must
+	  provide one yourself.
+endmenu
+
+source "arch/xtensa/Kconfig.debug"
+
+source "security/Kconfig"
+
+source "crypto/Kconfig"
+
+source "lib/Kconfig"
+
+
diff --git a/arch/xtensa/Kconfig.debug b/arch/xtensa/Kconfig.debug
new file mode 100644
index 00000000000..11c585295dd
--- /dev/null
+++ b/arch/xtensa/Kconfig.debug
@@ -0,0 +1,7 @@
+menu "Kernel hacking"
+
+source "lib/Kconfig.debug"
+
+endmenu
+
+
diff --git a/arch/xtensa/Makefile b/arch/xtensa/Makefile
new file mode 100644
index 00000000000..4fa27453b1f
--- /dev/null
+++ b/arch/xtensa/Makefile
@@ -0,0 +1,102 @@
+#
+# 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.
+#
+# This file is included by the global makefile so that you can add your own
+# architecture-specific flags and dependencies. Remember to do have actions
+# for "archclean" and "archdep" for cleaning up and making dependencies for
+# this architecture
+
+# Core configuration.
+# (Use CPU=<xtensa_config> to use another default compiler.)
+
+cpu-$(CONFIG_XTENSA_CPU_LINUX_BE)	:= linux_be
+cpu-$(CONFIG_XTENSA_CPU_LINUX_CUSTOM)	:= linux_custom
+
+CPU = $(cpu-y)
+export CPU
+
+# Platform configuration
+
+platform-y					:= common
+platform-$(CONFIG_XTENSA_PLATFORM_XT2000)	:= xt2000
+platform-$(CONFIG_XTENSA_PLATFORM_ISS)		:= iss
+
+PLATFORM = $(platform-y)
+export PLATFORM
+
+#LDFLAGS_vmlinux		:= -T$(word 1,$(LINKSCRIPT))
+AFLAGS_vmlinux.lds.o	:= -Uxtensa
+CPPFLAGS		+= -Iarch/xtensa -Iinclude/asm  -mlongcalls  -g
+AFLAGS			+= -Iarch/xtensa -Iinclude/asm
+CPP			= $(CC) -E $(CFLAGS)
+
+cflags-y	+= -Iarch/xtensa -pipe -mlongcalls
+
+
+KBUILD_DEFCONFIG := common_defconfig
+
+# ramdisk/initrd support
+# You need a compressed ramdisk image, named ramdisk.gz in
+# arch/xtensa/boot/ramdisk
+
+core-$(CONFIG_EMBEDDED_RAMDISK)	+= arch/xtensa/boot/ramdisk/
+
+# Test for cross compiling
+
+ifneq ($(CPU),)
+  COMPILE_ARCH = $(shell uname -m)
+
+  ifneq ($(COMPILE_ARCH), xtensa)
+    ifndef CROSS_COMPILE
+      CROSS_COMPILE = xtensa_$(CPU)-
+    endif
+  endif
+endif
+
+#
+
+LIBGCC := $(shell $(CC) $(CFLAGS) -print-libgcc-file-name)
+
+head-y		:= arch/xtensa/kernel/head.o
+core-y		+= arch/xtensa/kernel/ \
+		   arch/xtensa/mm/ arch/xtensa/platform-$(PLATFORM)/
+libs-y		+= arch/xtensa/lib/ $(LIBGCC)
+
+boot	:= arch/xtensa/boot
+
+arch/xtensa/kernel/asm-offsets.s: \
+	arch/xtensa/kernel/asm-offsets.c \
+	include/asm-xtensa/.platform
+
+include/asm-xtensa/offsets.h: arch/xtensa/kernel/asm-offsets.s
+	$(call filechk,gen-asm-offsets)
+
+prepare: include/asm-xtensa/.platform include/asm-xtensa/offsets.h
+
+# Update machine cpu and platform symlinks if something which affects
+# them changed.
+
+include/asm-xtensa/.platform: $(wildcard include/config/arch/*.h)
+	@echo '  Setting up cpu ($(CPU)) and platform ($(PLATFORM)) symlinks'
+	$(Q)rm -f include/asm-xtensa/platform
+	$(Q)rm -f include/asm-xtensa/xtensa/config
+	$(Q)(cd include/asm-xtensa/; ln -sf platform-$(PLATFORM) platform)
+	$(Q)(cd include/asm-xtensa/xtensa; ln -sf config-$(CPU) config)
+
+all: zImage
+
+bzImage : zImage
+
+zImage zImage.initrd: vmlinux
+	$(Q)$(MAKE) $(build)=$(boot) $@
+
+CLEAN_FILES	+= arch/xtensa/vmlinux.lds include/asm-xtensa/offset.h
+
+define archhelp
+  @echo '* zImage      - Compressed kernel image (arch/xtensa/boot/images/zImage.*)'
+endef
+
diff --git a/arch/xtensa/configs/common_defconfig b/arch/xtensa/configs/common_defconfig
new file mode 100644
index 00000000000..1d230ee081b
--- /dev/null
+++ b/arch/xtensa/configs/common_defconfig
@@ -0,0 +1,662 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.11-rc2
+# Tue Mar  1 16:36:53 2005
+#
+# CONFIG_FRAME_POINTER is not set
+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=y
+# CONFIG_BSD_PROCESS_ACCT_V3 is not set
+CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_HOTPLUG is not set
+CONFIG_KOBJECT_UEVENT=y
+# CONFIG_IKCONFIG is not set
+# CONFIG_EMBEDDED is not set
+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=y
+# CONFIG_MODULE_UNLOAD is not set
+CONFIG_OBSOLETE_MODPARM=y
+CONFIG_MODVERSIONS=y
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+CONFIG_KMOD=y
+
+#
+# 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 is not set
+CONFIG_XTENSA_PLATFORM_XT2000=y
+CONFIG_XTENSA_CALIBRATE_CCOUNT=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttyS0,38400 ip=bootp root=nfs nfsroot=/opt/montavista/pro/devkit/xtensa/linux_be/target"
+
+#
+# Bus options
+#
+CONFIG_PCI=y
+# CONFIG_PCI_LEGACY_PROC is not set
+# CONFIG_PCI_NAMES is not set
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+# CONFIG_PCCARD is not set
+
+#
+# PC-card bridges
+#
+
+#
+# PCI Hotplug Support
+#
+# CONFIG_HOTPLUG_PCI is not set
+
+#
+# Exectuable file formats
+#
+CONFIG_KCORE_ELF=y
+CONFIG_BINFMT_ELF=y
+CONFIG_BINFMT_MISC=y
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+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_CPQ_DA is not set
+# CONFIG_BLK_CPQ_CISS_DA is not set
+# CONFIG_BLK_DEV_DAC960 is not set
+# CONFIG_BLK_DEV_UMEM 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_SX8 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=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
+#
+# CONFIG_IEEE1394 is not set
+
+#
+# I2O device support
+#
+# CONFIG_I2O is not set
+
+#
+# Networking support
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+# CONFIG_PACKET is not set
+# CONFIG_NETLINK_DEV is not set
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+CONFIG_IP_MULTICAST=y
+CONFIG_IP_ADVANCED_ROUTER=y
+CONFIG_IP_MULTIPLE_TABLES=y
+CONFIG_IP_ROUTE_MULTIPATH=y
+CONFIG_IP_ROUTE_VERBOSE=y
+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_IP_MROUTE 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_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=y
+CONFIG_NET_SCH_CLK_JIFFIES=y
+# CONFIG_NET_SCH_CLK_GETTIMEOFDAY is not set
+# CONFIG_NET_SCH_CLK_CPU is not set
+CONFIG_NET_SCH_CBQ=m
+CONFIG_NET_SCH_HTB=m
+# CONFIG_NET_SCH_HFSC is not set
+CONFIG_NET_SCH_PRIO=m
+CONFIG_NET_SCH_RED=m
+CONFIG_NET_SCH_SFQ=m
+CONFIG_NET_SCH_TEQL=m
+CONFIG_NET_SCH_TBF=m
+CONFIG_NET_SCH_GRED=m
+CONFIG_NET_SCH_DSMARK=m
+# CONFIG_NET_SCH_NETEM is not set
+CONFIG_NET_SCH_INGRESS=m
+CONFIG_NET_QOS=y
+CONFIG_NET_ESTIMATOR=y
+CONFIG_NET_CLS=y
+CONFIG_NET_CLS_TCINDEX=m
+CONFIG_NET_CLS_ROUTE4=m
+CONFIG_NET_CLS_ROUTE=y
+CONFIG_NET_CLS_FW=m
+CONFIG_NET_CLS_U32=m
+# CONFIG_CLS_U32_PERF is not set
+# CONFIG_NET_CLS_IND is not set
+CONFIG_NET_CLS_RSVP=m
+CONFIG_NET_CLS_RSVP6=m
+# CONFIG_NET_CLS_ACT is not set
+CONFIG_NET_CLS_POLICE=y
+
+#
+# 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=y
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+
+#
+# ARCnet devices
+#
+# CONFIG_ARCNET is not set
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+# CONFIG_MII is not set
+CONFIG_XT2000_SONIC=y
+# CONFIG_HAPPYMEAL is not set
+# CONFIG_SUNGEM is not set
+# CONFIG_NET_VENDOR_3COM is not set
+
+#
+# Tulip family network device support
+#
+# CONFIG_NET_TULIP is not set
+# CONFIG_HP100 is not set
+# CONFIG_NET_PCI is not set
+
+#
+# Ethernet (1000 Mbit)
+#
+# CONFIG_ACENIC is not set
+# CONFIG_DL2K is not set
+# CONFIG_E1000 is not set
+# CONFIG_NS83820 is not set
+# CONFIG_HAMACHI is not set
+# CONFIG_YELLOWFIN is not set
+# CONFIG_R8169 is not set
+# CONFIG_SK98LIN is not set
+# CONFIG_TIGON3 is not set
+
+#
+# Ethernet (10000 Mbit)
+#
+# CONFIG_IXGB is not set
+# CONFIG_S2IO is not set
+
+#
+# Token Ring devices
+#
+# CONFIG_TR is not set
+
+#
+# Wireless LAN (non-hamradio)
+#
+CONFIG_NET_RADIO=y
+
+#
+# Obsolete Wireless cards support (pre-802.11)
+#
+CONFIG_STRIP=m
+
+#
+# Wireless 802.11b ISA/PCI cards support
+#
+CONFIG_HERMES=m
+# CONFIG_PLX_HERMES is not set
+# CONFIG_TMD_HERMES is not set
+# CONFIG_PCI_HERMES is not set
+# CONFIG_ATMEL is not set
+
+#
+# Prism GT/Duette 802.11(a/b/g) PCI/Cardbus support
+#
+# CONFIG_PRISM54 is not set
+CONFIG_NET_WIRELESS=y
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+# CONFIG_FDDI is not set
+# CONFIG_HIPPI 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 is not set
+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 I/O drivers
+#
+# CONFIG_GAMEPORT is not set
+CONFIG_SOUND_GAMEPORT=y
+CONFIG_SERIO=y
+# CONFIG_SERIO_I8042 is not set
+# CONFIG_SERIO_SERPORT is not set
+# CONFIG_SERIO_CT82C710 is not set
+# CONFIG_SERIO_PCIPS2 is not set
+# CONFIG_SERIO_RAW 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=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_RTC is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+# CONFIG_APPLICOM 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 is not set
+CONFIG_USB_ARCH_HAS_HCD=y
+CONFIG_USB_ARCH_HAS_OHCI=y
+
+#
+# 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=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 is not set
+CONFIG_SYSFS=y
+CONFIG_DEVFS_FS=y
+# CONFIG_DEVFS_MOUNT is not set
+# CONFIG_DEVFS_DEBUG is not set
+# CONFIG_DEVPTS_FS_XATTR is not set
+# CONFIG_TMPFS 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 is not set
+# 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=y
+# 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
-- 
cgit v1.2.3-70-g09d2


From 4bedea94545165364618d403d03b61d797acba0b Mon Sep 17 00:00:00 2001
From: Chris Zankel <czankel@tensilica.com>
Date: Thu, 23 Jun 2005 22:01:12 -0700
Subject: [PATCH] xtensa: Architecture support for Tensilica Xtensa Part 2

The attached patches provides part 2 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/boot/Makefile                 |   37 +
 arch/xtensa/boot/boot-elf/Makefile        |   52 +
 arch/xtensa/boot/boot-elf/boot.ld         |   71 +
 arch/xtensa/boot/boot-elf/bootstrap.S     |   37 +
 arch/xtensa/boot/boot-redboot/Makefile    |   35 +
 arch/xtensa/boot/boot-redboot/boot.ld     |   66 +
 arch/xtensa/boot/boot-redboot/bootstrap.S |  246 ++++
 arch/xtensa/boot/include/zlib.h           |  433 ++++++
 arch/xtensa/boot/lib/Makefile             |    6 +
 arch/xtensa/boot/lib/memcpy.S             |   36 +
 arch/xtensa/boot/lib/zlib.c               | 2150 +++++++++++++++++++++++++++++
 arch/xtensa/boot/lib/zmem.c               |   87 ++
 arch/xtensa/boot/ramdisk/Makefile         |   23 +
 13 files changed, 3279 insertions(+)
 create mode 100644 arch/xtensa/boot/Makefile
 create mode 100644 arch/xtensa/boot/boot-elf/Makefile
 create mode 100644 arch/xtensa/boot/boot-elf/boot.ld
 create mode 100644 arch/xtensa/boot/boot-elf/bootstrap.S
 create mode 100644 arch/xtensa/boot/boot-redboot/Makefile
 create mode 100644 arch/xtensa/boot/boot-redboot/boot.ld
 create mode 100644 arch/xtensa/boot/boot-redboot/bootstrap.S
 create mode 100644 arch/xtensa/boot/include/zlib.h
 create mode 100644 arch/xtensa/boot/lib/Makefile
 create mode 100644 arch/xtensa/boot/lib/memcpy.S
 create mode 100644 arch/xtensa/boot/lib/zlib.c
 create mode 100644 arch/xtensa/boot/lib/zmem.c
 create mode 100644 arch/xtensa/boot/ramdisk/Makefile

(limited to 'arch')

diff --git a/arch/xtensa/boot/Makefile b/arch/xtensa/boot/Makefile
new file mode 100644
index 00000000000..260f456ccf0
--- /dev/null
+++ b/arch/xtensa/boot/Makefile
@@ -0,0 +1,37 @@
+#
+# arch/xtensa/boot/Makefile
+#
+# 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.
+#
+#
+
+
+CFLAGS 		+= -fno-builtin -Iarch/$(ARCH)/boot/include
+HOSTFLAGS	+= -Iarch/$(ARCH)/boot/include
+
+BIG_ENDIAN	:= $(shell echo -e "\#ifdef __XTENSA_EL__\nint little;\n\#else\nint big;\n\#endif" | $(CC) -E -|grep -c big)
+
+
+export CFLAGS
+export AFLAGS
+export BIG_ENDIAN
+
+# Subdirs for the boot loader(s)
+
+bootdir-$(CONFIG_XTENSA_PLATFORM_ISS)	 += boot-elf
+bootdir-$(CONFIG_XTENSA_PLATFORM_XT2000) += boot-redboot boot-elf
+
+subdir-y	:= lib/
+
+subdir-y	+= boot-elf/ boot-redboot/
+
+zImage zImage.initrd Image Image.initrd: $(bootdir-y)
+
+$(bootdir-y): $(addprefix $(obj)/,$(subdir-y)) \
+	      $(addprefix $(obj)/,$(host-progs))
+	$(Q)$(MAKE) $(build)=$(obj)/$@ $(MAKECMDGOALS)
+
+
+
diff --git a/arch/xtensa/boot/boot-elf/Makefile b/arch/xtensa/boot/boot-elf/Makefile
new file mode 100644
index 00000000000..f6ef6a36966
--- /dev/null
+++ b/arch/xtensa/boot/boot-elf/Makefile
@@ -0,0 +1,52 @@
+#
+# This file is subject to the terms and conditions of the GNU General Public
+# License.  See the file "COPYING" in the main directory of this archive
+# for more details.
+#
+
+GZIP = gzip
+GZIP_FLAGS = -v9fc
+
+ifeq ($(BIG_ENDIAN),1)
+OBJCOPY_ARGS    := -O elf32-xtensa-be
+else
+OBJCOPY_ARGS    := -O elf32-xtensa-le
+endif
+
+export OBJCOPY_ARGS
+
+boot-y		:= bootstrap.o
+
+OBJS		:= $(addprefix $(obj)/,$(boot-y))
+
+Image:	vmlinux $(OBJS)
+	$(OBJCOPY) --strip-all -R .comment -R .xt.insn -O binary \
+		vmlinux vmlinux.tmp
+	$(OBJCOPY) $(OBJCOPY_ARGS) -R .comment \
+		--add-section image=vmlinux.tmp \
+		--set-section-flags image=contents,alloc,load,load,data \
+		$(OBJS) $@.tmp
+	$(LD) $(LDFLAGS) $(LDFLAGS_vmlinux) \
+		-T arch/$(ARCH)/boot/boot-elf/boot.ld \
+		-o arch/$(ARCH)/boot/$@.elf $@.tmp
+	rm -f $@.tmp vmlinux.tmp
+
+Image.initrd:	vmlinux $(OBJS)
+	$(OBJCOPY) --strip-all -R .comment -R .xt.insn -O binary \
+		--add-section .initrd=arch/$(ARCH)/boot/ramdisk \
+		--set-section-flags .initrd=contents,alloc,load,load,data \
+		vmlinux vmlinux.tmp
+	$(OBJCOPY) $(OBJCOPY_ARGS) -R .comment \
+		--add-section image=vmlinux.tmp \
+		--set-section-flags image=contents,alloc,load,load,data \
+		$(OBJS) $@.tmp
+	$(LD) $(LDFLAGS) $(LDFLAGS_vmlinux) \
+		-T arch/$(ARCH)/boot/boot-elf/boot.ld \
+		-o arch/$(ARCH)/boot/$@.elf $@.tmp
+	rm -f $@.tmp vmlinux.tmp
+
+
+zImage:	Image
+
+zImage.initrd:	Image.initrd
+
diff --git a/arch/xtensa/boot/boot-elf/boot.ld b/arch/xtensa/boot/boot-elf/boot.ld
new file mode 100644
index 00000000000..4ab06a0a7a6
--- /dev/null
+++ b/arch/xtensa/boot/boot-elf/boot.ld
@@ -0,0 +1,71 @@
+OUTPUT_ARCH(xtensa)
+
+SECTIONS
+{
+	.start 0xD0000000 : { *(.start) }
+
+	.text 0xD0000000:
+	{
+		__reloc_start = . ;
+		_text_start = . ;
+		*(.literal .text.literal .text)
+		_text_end = . ;
+	}
+
+	.rodata ALIGN(0x04):
+	{
+		*(.rodata)
+		*(.rodata1)
+	}
+
+	.data ALIGN(0x04):
+	{
+		*(.data)
+		*(.data1)
+		*(.sdata)
+		*(.sdata2)
+		*(.got.plt)
+		*(.got)
+		*(.dynamic)
+	}
+
+	__reloc_end = . ;
+
+	.initrd ALIGN(0x10) :
+	{
+		boot_initrd_start = . ;
+		*(.initrd)
+		boot_initrd_end = .;
+	}
+
+	. = ALIGN(0x10);
+	__image_load = . ;
+	.image 0xd0001000:
+	{
+		_image_start = .;
+		*(image)
+		. = (. + 3) & ~ 3;
+		_image_end = .	;
+	}
+
+
+	.bss ((LOADADDR(.image) + SIZEOF(.image) + 3) & ~ 3):
+	{
+		__bss_start = .;
+		*(.sbss)
+		*(.scommon)
+		*(.dynbss)
+		*(.bss)
+		__bss_end = .;
+	}
+	_end = .;
+	_param_start = .;
+
+	.ResetVector.text 0xfe000020 :
+	{
+		*(.ResetVector.text)
+	}
+
+
+	PROVIDE (end = .);
+}
diff --git a/arch/xtensa/boot/boot-elf/bootstrap.S b/arch/xtensa/boot/boot-elf/bootstrap.S
new file mode 100644
index 00000000000..7cba94abdab
--- /dev/null
+++ b/arch/xtensa/boot/boot-elf/bootstrap.S
@@ -0,0 +1,37 @@
+
+#include <xtensa/config/specreg.h>
+#include <xtensa/config/core.h>
+
+#include <linux/config.h>
+#include <asm/bootparam.h>
+
+
+/* ResetVector
+ */
+	.section        .ResetVector.text, "ax"
+	.global         _ResetVector
+_ResetVector:
+	_j reset
+	.align 4
+RomInitAddr:
+	.word 0xd0001000
+RomBootParam:
+	.word _bootparam
+reset:
+	l32r    a0, RomInitAddr
+	l32r	a2, RomBootParam
+	movi	a3, 0
+	movi	a4, 0
+	jx      a0
+
+	.align 4
+	.section .bootstrap.data, "aw"
+
+	.globl _bootparam
+_bootparam:
+	.short	BP_TAG_FIRST
+	.short	4
+	.long	BP_VERSION
+	.short	BP_TAG_LAST
+	.short	0
+	.long	0
diff --git a/arch/xtensa/boot/boot-redboot/Makefile b/arch/xtensa/boot/boot-redboot/Makefile
new file mode 100644
index 00000000000..ca8a68bc847
--- /dev/null
+++ b/arch/xtensa/boot/boot-redboot/Makefile
@@ -0,0 +1,35 @@
+#
+# 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.
+#
+
+GZIP = gzip
+GZIP_FLAGS = -v9fc
+ifeq ($(BIG_ENDIAN),1)
+OBJCOPY_ARGS 	:= -O elf32-xtensa-be
+else
+OBJCOPY_ARGS 	:= -O elf32-xtensa-le
+endif
+
+LD_ARGS	= -T $(obj)/boot.ld
+
+boot-y	:= bootstrap.o
+
+OBJS	:= $(addprefix $(obj)/,$(boot-y))
+LIBS	:= arch/$(ARCH)/boot/lib/lib.a arch/$(ARCH)/lib/lib.a
+
+LIBGCC := $(shell $(CC) $(CFLAGS) -print-libgcc-file-name)
+
+zImage: vmlinux $(OBJS) $(LIBS)
+	$(OBJCOPY) --strip-all -R .comment -R .xt.insn -O binary \
+		$(TOPDIR)/vmlinux vmlinux.tmp
+	gzip -vf9 vmlinux.tmp
+	$(OBJCOPY) $(OBJCOPY_ARGS) -R .comment \
+		--add-section image=vmlinux.tmp.gz \
+		--set-section-flags image=contents,alloc,load,load,data \
+		$(OBJS) $@.tmp
+	$(LD) $(LD_ARGS) -o $@.elf $@.tmp $(LIBS) -L/xtensa-elf/lib $(LIBGCC)
+	$(OBJCOPY) -S -O binary $@.elf arch/$(ARCH)/boot/images/$@.redboot
+#	rm -f $@.tmp $@.elf vmlinux.tmp.gz
+
diff --git a/arch/xtensa/boot/boot-redboot/boot.ld b/arch/xtensa/boot/boot-redboot/boot.ld
new file mode 100644
index 00000000000..65b726410e8
--- /dev/null
+++ b/arch/xtensa/boot/boot-redboot/boot.ld
@@ -0,0 +1,66 @@
+OUTPUT_ARCH(xtensa)
+
+SECTIONS
+{
+	.start 0xD0200000 : { *(.start) }
+
+	.text :
+	{
+		__reloc_start = . ;
+		_text_start = . ;
+		*(.literal .text.literal .text)
+		_text_end = . ;
+	}
+
+	.rodata ALIGN(0x04):
+	{
+		*(.rodata)
+		*(.rodata1)
+	}
+
+	.data ALIGN(0x04):
+	{
+		*(.data)
+		*(.data1)
+		*(.sdata)
+		*(.sdata2)
+		*(.got.plt)
+		*(.got)
+		*(.dynamic)
+	}
+
+	__reloc_end = . ;
+
+	.initrd ALIGN(0x10) :
+	{
+		boot_initrd_start = . ;
+		*(.initrd)
+		boot_initrd_end = .;
+	}
+
+	. = ALIGN(0x10);
+	__image_load = . ;
+	.image 0xd0001000: AT(__image_load)
+	{
+		_image_start = .;
+		*(image)
+		. = (. + 3) & ~ 3;
+		_image_end = .	;
+	}
+
+
+	.bss ((LOADADDR(.image) + SIZEOF(.image) + 3) & ~ 3):
+	{
+		__bss_start = .;
+		*(.sbss)
+		*(.scommon)
+		*(.dynbss)
+		*(.bss)
+		__bss_end = .;
+	}
+	_end = .;
+	_param_start = .;
+
+
+	PROVIDE (end = .);
+}
diff --git a/arch/xtensa/boot/boot-redboot/bootstrap.S b/arch/xtensa/boot/boot-redboot/bootstrap.S
new file mode 100644
index 00000000000..ee636b0da81
--- /dev/null
+++ b/arch/xtensa/boot/boot-redboot/bootstrap.S
@@ -0,0 +1,246 @@
+
+#define _ASMLANGUAGE
+#include <xtensa/config/specreg.h>
+#include <xtensa/config/core.h>
+#include <xtensa/cacheasm.h>
+
+	/*
+	 * RB-Data: RedBoot data/bss
+	 * P:	    Boot-Parameters
+	 * L:	    Kernel-Loader
+	 *
+	 * The Linux-Kernel image including the loader must be loaded
+	 * to a position so that the kernel and the boot parameters
+	 * can fit in the space before the load address.
+	 *  ______________________________________________________
+	 * |_RB-Data_|_P_|__________|_L_|___Linux-Kernel___|______|
+	 *                          ^
+	 *                          ^ Load address
+	 *  ______________________________________________________
+	 * |___Linux-Kernel___|_P_|_L_|___________________________|
+	 *
+	 * The loader copies the parameter to the position that will
+	 * be the end of the kernel and itself to the end of the
+	 * parameter list.
+	 */
+
+/* Make sure we have enough space for the 'uncompressor' */
+
+#define STACK_SIZE 32768
+#define HEAP_SIZE (131072*4)
+
+	# a2: Parameter list
+	# a3: Size of parameter list
+
+	.section .start, "ax"
+
+	.globl __start
+	/* this must be the first byte of the loader! */
+__start:
+	entry	sp, 32		# we do not intend to return
+	_call0	_start
+__start_a0:
+	.align 4
+
+	.section .text, "ax"
+	.begin literal_prefix .text
+
+	/* put literals in here! */
+
+	.globl _start
+_start:
+
+	/* 'reset' window registers */
+
+	movi	a4, 1
+	wsr	a4, PS
+	rsync
+
+	rsr	a5, WINDOWBASE
+	ssl	a5
+	sll	a4, a4
+	wsr	a4, WINDOWSTART
+	rsync
+
+	movi	a4, 0x00040000
+	wsr	a4, PS
+	rsync
+
+	/* copy the loader to its address
+	 * Note: The loader itself is a very small piece, so we assume we
+	 *       don't partially overlap. We also assume (even more important)
+	 *	 that the kernel image is out of the way. Usually, when the
+	 *	 load address of this image is not at an arbitrary address,
+	 *	 but aligned to some 10K's we shouldn't overlap.
+	 */
+
+	/* Note: The assembler cannot relax "addi a0, a0, ..." to an
+	   l32r, so we load to a4 first. */
+
+	addi	a4, a0, __start - __start_a0
+	mov	a0, a4
+	movi	a4, __start
+	movi	a5, __reloc_end
+
+	# a0: address where this code has been loaded
+	# a4: compiled address of __start
+	# a5: compiled end address
+
+	mov.n	a7, a0
+	mov.n	a8, a4
+
+1:
+	l32i	a10, a7, 0
+	l32i	a11, a7, 4
+	s32i	a10, a8, 0
+	s32i	a11, a8, 4
+	l32i	a10, a7, 8
+	l32i	a11, a7, 12
+	s32i	a10, a8, 8
+	s32i	a11, a8, 12
+	addi	a8, a8, 16
+	addi	a7, a7, 16
+	blt	a8, a5, 1b
+
+
+	/* We have to flush and invalidate the caches here before we jump. */
+
+#if XCHAL_DCACHE_IS_WRITEBACK
+	dcache_writeback_all  a5, a6
+#endif
+	icache_invalidate_all a5, a6
+
+	movi	a11, _reloc
+	jx	a11
+
+	.globl _reloc
+_reloc:
+
+	/* RedBoot is now at the end of the memory, so we don't have
+	 * to copy the parameter list. Keep the code around; in case
+	 * we need it again. */
+#if 0
+	# a0: load address
+	# a2: start address of parameter list
+	# a3: length of parameter list
+	# a4: __start
+
+	/* copy the parameter list out of the way */
+
+	movi	a6, _param_start
+	add	a3, a2, a3
+2:
+	l32i	a8, a2, 0
+	s32i	a8, a6, 0
+	addi	a2, a2, 4
+	addi	a6, a6, 4
+	blt	a2, a3, 2b
+#endif
+
+	/* clear BSS section */
+	movi	a6, __bss_start
+	movi	a7, __bss_end
+	movi.n	a5, 0
+3:
+	s32i	a5, a6, 0
+	addi	a6, a6, 4
+	blt	a6, a7, 3b
+
+	movi	a5, -16
+	movi	a1, _stack + STACK_SIZE
+	and	a1, a1, a5
+
+	/* Uncompress the kernel */
+
+	# a0: load address
+	# a2: boot parameter
+	# a4: __start
+
+	movi	a3, __image_load
+	sub	a4, a3, a4
+	add	a8, a0, a4
+
+	# a1  Stack
+	# a8(a4)  Load address of the image
+
+	movi	a6, _image_start
+	movi	a10, _image_end
+	movi	a7, 0x1000000
+	sub	a11, a10, a6
+	movi	a9, complen
+	s32i	a11, a9, 0
+
+	movi	a0, 0
+
+	# a6 destination
+	# a7 maximum size of destination
+	# a8 source
+	# a9 ptr to length
+
+	.extern gunzip
+	movi	a4, gunzip
+	beqz	a4, 1f
+
+	callx4	a4
+
+	j	2f
+
+
+	# a6 destination start
+	# a7 maximum size of destination
+	# a8 source start
+	# a9 ptr to length
+	# a10 destination end
+
+1:
+        l32i    a9, a8, 0
+        l32i    a11, a8, 4
+        s32i    a9, a6, 0
+        s32i    a11, a6, 4
+        l32i    a9, a8, 8
+        l32i    a11, a8, 12
+        s32i    a9, a6, 8
+        s32i    a11, a6, 12
+        addi    a6, a6, 16
+        addi    a8, a8, 16
+        blt     a6, a10, 1b
+
+
+	/* jump to the kernel */
+2:
+#if XCHAL_DCACHE_IS_WRITEBACK
+	dcache_writeback_all a5, a6
+#endif
+	icache_invalidate_all a5, a6
+
+	movi	a5, __start
+	movi	a3, boot_initrd_start
+	movi	a4, boot_initrd_end
+	sub	a3, a3, a5
+	sub	a4, a4, a5
+	add	a3, a0, a3
+	add	a4, a0, a4
+
+	# a2  Boot parameter list
+	# a3  initrd_start (virtual load address)
+	# a4  initrd_end   (virtual load address)
+
+	movi	a0, _image_start
+	jx	a0
+
+	.align 16
+	.data
+	.globl avail_ram
+avail_ram:
+	.long	_heap
+	.globl end_avail
+end_avail:
+	.long	_heap + HEAP_SIZE
+
+	.comm _stack, STACK_SIZE
+	.comm _heap, HEAP_SIZE
+
+	.globl end_avail
+	.comm complen, 4
+
+	.end	literal_prefix
diff --git a/arch/xtensa/boot/include/zlib.h b/arch/xtensa/boot/include/zlib.h
new file mode 100644
index 00000000000..ea29b623785
--- /dev/null
+++ b/arch/xtensa/boot/include/zlib.h
@@ -0,0 +1,433 @@
+/*
+ * BK Id: SCCS/s.zlib.h 1.8 05/18/01 15:17:23 cort
+ */
+/*
+ * This file is derived from zlib.h and zconf.h from the zlib-0.95
+ * distribution by Jean-loup Gailly and Mark Adler, with some additions
+ * by Paul Mackerras to aid in implementing Deflate compression and
+ * decompression for PPP packets.
+ */
+
+/*
+ *  ==FILEVERSION 960122==
+ *
+ * This marker is used by the Linux installation script to determine
+ * whether an up-to-date version of this file is already installed.
+ */
+
+/* zlib.h -- interface of the 'zlib' general purpose compression library
+  version 0.95, Aug 16th, 1995.
+
+  Copyright (C) 1995 Jean-loup Gailly and Mark Adler
+
+  This software is provided 'as-is', without any express or implied
+  warranty.  In no event will the authors be held liable for any damages
+  arising from the use of this software.
+
+  Permission is granted to anyone to use this software for any purpose,
+  including commercial applications, and to alter it and redistribute it
+  freely, subject to the following restrictions:
+
+  1. The origin of this software must not be misrepresented; you must not
+     claim that you wrote the original software. If you use this software
+     in a product, an acknowledgment in the product documentation would be
+     appreciated but is not required.
+  2. Altered source versions must be plainly marked as such, and must not be
+     misrepresented as being the original software.
+  3. This notice may not be removed or altered from any source distribution.
+
+  Jean-loup Gailly        Mark Adler
+  gzip@prep.ai.mit.edu    madler@alumni.caltech.edu
+ */
+
+#ifndef _ZLIB_H
+#define _ZLIB_H
+
+/* #include "zconf.h" */	/* included directly here */
+
+/* zconf.h -- configuration of the zlib compression library
+ * Copyright (C) 1995 Jean-loup Gailly.
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* From: zconf.h,v 1.12 1995/05/03 17:27:12 jloup Exp */
+
+/*
+     The library does not install any signal handler. It is recommended to
+  add at least a handler for SIGSEGV when decompressing; the library checks
+  the consistency of the input data whenever possible but may go nuts
+  for some forms of corrupted input.
+ */
+
+/*
+ * Compile with -DMAXSEG_64K if the alloc function cannot allocate more
+ * than 64k bytes at a time (needed on systems with 16-bit int).
+ * Compile with -DUNALIGNED_OK if it is OK to access shorts or ints
+ * at addresses which are not a multiple of their size.
+ * Under DOS, -DFAR=far or -DFAR=__far may be needed.
+ */
+
+#ifndef STDC
+#  if defined(MSDOS) || defined(__STDC__) || defined(__cplusplus)
+#    define STDC
+#  endif
+#endif
+
+#ifdef	__MWERKS__ /* Metrowerks CodeWarrior declares fileno() in unix.h */
+#  include <unix.h>
+#endif
+
+/* Maximum value for memLevel in deflateInit2 */
+#ifndef MAX_MEM_LEVEL
+#  ifdef MAXSEG_64K
+#    define MAX_MEM_LEVEL 8
+#  else
+#    define MAX_MEM_LEVEL 9
+#  endif
+#endif
+
+#ifndef FAR
+#  define FAR
+#endif
+
+/* Maximum value for windowBits in deflateInit2 and inflateInit2 */
+#ifndef MAX_WBITS
+#  define MAX_WBITS   15 /* 32K LZ77 window */
+#endif
+
+/* The memory requirements for deflate are (in bytes):
+            1 << (windowBits+2)   +  1 << (memLevel+9)
+ that is: 128K for windowBits=15  +  128K for memLevel = 8  (default values)
+ plus a few kilobytes for small objects. For example, if you want to reduce
+ the default memory requirements from 256K to 128K, compile with
+     make CFLAGS="-O -DMAX_WBITS=14 -DMAX_MEM_LEVEL=7"
+ Of course this will generally degrade compression (there's no free lunch).
+
+   The memory requirements for inflate are (in bytes) 1 << windowBits
+ that is, 32K for windowBits=15 (default value) plus a few kilobytes
+ for small objects.
+*/
+
+                        /* Type declarations */
+
+#ifndef OF /* function prototypes */
+#  ifdef STDC
+#    define OF(args)  args
+#  else
+#    define OF(args)  ()
+#  endif
+#endif
+
+typedef unsigned char  Byte;  /* 8 bits */
+typedef unsigned int   uInt;  /* 16 bits or more */
+typedef unsigned long  uLong; /* 32 bits or more */
+
+typedef Byte FAR Bytef;
+typedef char FAR charf;
+typedef int FAR intf;
+typedef uInt FAR uIntf;
+typedef uLong FAR uLongf;
+
+#ifdef STDC
+   typedef void FAR *voidpf;
+   typedef void     *voidp;
+#else
+   typedef Byte FAR *voidpf;
+   typedef Byte     *voidp;
+#endif
+
+/* end of original zconf.h */
+
+#define ZLIB_VERSION "0.95P"
+
+/*
+     The 'zlib' compression library provides in-memory compression and
+  decompression functions, including integrity checks of the uncompressed
+  data.  This version of the library supports only one compression method
+  (deflation) but other algorithms may be added later and will have the same
+  stream interface.
+
+     For compression the application must provide the output buffer and
+  may optionally provide the input buffer for optimization. For decompression,
+  the application must provide the input buffer and may optionally provide
+  the output buffer for optimization.
+
+     Compression can be done in a single step if the buffers are large
+  enough (for example if an input file is mmap'ed), or can be done by
+  repeated calls of the compression function.  In the latter case, the
+  application must provide more input and/or consume the output
+  (providing more output space) before each call.
+*/
+
+typedef voidpf (*alloc_func) OF((voidpf opaque, uInt items, uInt size));
+typedef void   (*free_func)  OF((voidpf opaque, voidpf address, uInt nbytes));
+
+struct internal_state;
+
+typedef struct z_stream_s {
+    Bytef    *next_in;  /* next input byte */
+    uInt     avail_in;  /* number of bytes available at next_in */
+    uLong    total_in;  /* total nb of input bytes read so far */
+
+    Bytef    *next_out; /* next output byte should be put there */
+    uInt     avail_out; /* remaining free space at next_out */
+    uLong    total_out; /* total nb of bytes output so far */
+
+    char     *msg;      /* last error message, NULL if no error */
+    struct internal_state FAR *state; /* not visible by applications */
+
+    alloc_func zalloc;  /* used to allocate the internal state */
+    free_func  zfree;   /* used to free the internal state */
+    voidp      opaque;  /* private data object passed to zalloc and zfree */
+
+    Byte     data_type; /* best guess about the data type: ascii or binary */
+
+} z_stream;
+
+/*
+   The application must update next_in and avail_in when avail_in has
+   dropped to zero. It must update next_out and avail_out when avail_out
+   has dropped to zero. The application must initialize zalloc, zfree and
+   opaque before calling the init function. All other fields are set by the
+   compression library and must not be updated by the application.
+
+   The opaque value provided by the application will be passed as the first
+   parameter for calls of zalloc and zfree. This can be useful for custom
+   memory management. The compression library attaches no meaning to the
+   opaque value.
+
+   zalloc must return Z_NULL if there is not enough memory for the object.
+   On 16-bit systems, the functions zalloc and zfree must be able to allocate
+   exactly 65536 bytes, but will not be required to allocate more than this
+   if the symbol MAXSEG_64K is defined (see zconf.h). WARNING: On MSDOS,
+   pointers returned by zalloc for objects of exactly 65536 bytes *must*
+   have their offset normalized to zero. The default allocation function
+   provided by this library ensures this (see zutil.c). To reduce memory
+   requirements and avoid any allocation of 64K objects, at the expense of
+   compression ratio, compile the library with -DMAX_WBITS=14 (see zconf.h).
+
+   The fields total_in and total_out can be used for statistics or
+   progress reports. After compression, total_in holds the total size of
+   the uncompressed data and may be saved for use in the decompressor
+   (particularly if the decompressor wants to decompress everything in
+   a single step).
+*/
+
+                        /* constants */
+
+#define Z_NO_FLUSH      0
+#define Z_PARTIAL_FLUSH 1
+#define Z_FULL_FLUSH    2
+#define Z_SYNC_FLUSH    3 /* experimental: partial_flush + byte align */
+#define Z_FINISH        4
+#define Z_PACKET_FLUSH	5
+/* See deflate() below for the usage of these constants */
+
+#define Z_OK            0
+#define Z_STREAM_END    1
+#define Z_ERRNO        (-1)
+#define Z_STREAM_ERROR (-2)
+#define Z_DATA_ERROR   (-3)
+#define Z_MEM_ERROR    (-4)
+#define Z_BUF_ERROR    (-5)
+/* error codes for the compression/decompression functions */
+
+#define Z_BEST_SPEED             1
+#define Z_BEST_COMPRESSION       9
+#define Z_DEFAULT_COMPRESSION  (-1)
+/* compression levels */
+
+#define Z_FILTERED            1
+#define Z_HUFFMAN_ONLY        2
+#define Z_DEFAULT_STRATEGY    0
+
+#define Z_BINARY   0
+#define Z_ASCII    1
+#define Z_UNKNOWN  2
+/* Used to set the data_type field */
+
+#define Z_NULL  0  /* for initializing zalloc, zfree, opaque */
+
+extern char *zlib_version;
+/* The application can compare zlib_version and ZLIB_VERSION for consistency.
+   If the first character differs, the library code actually used is
+   not compatible with the zlib.h header file used by the application.
+ */
+
+                        /* basic functions */
+
+extern int inflateInit OF((z_stream *strm));
+/*
+     Initializes the internal stream state for decompression. The fields
+   zalloc and zfree must be initialized before by the caller.  If zalloc and
+   zfree are set to Z_NULL, inflateInit updates them to use default allocation
+   functions.
+
+     inflateInit returns Z_OK if success, Z_MEM_ERROR if there was not
+   enough memory.  msg is set to null if there is no error message.
+   inflateInit does not perform any decompression: this will be done by
+   inflate().
+*/
+
+
+extern int inflate OF((z_stream *strm, int flush));
+/*
+  Performs one or both of the following actions:
+
+  - Decompress more input starting at next_in and update next_in and avail_in
+    accordingly. If not all input can be processed (because there is not
+    enough room in the output buffer), next_in is updated and processing
+    will resume at this point for the next call of inflate().
+
+  - Provide more output starting at next_out and update next_out and avail_out
+    accordingly.  inflate() always provides as much output as possible
+    (until there is no more input data or no more space in the output buffer).
+
+  Before the call of inflate(), the application should ensure that at least
+  one of the actions is possible, by providing more input and/or consuming
+  more output, and updating the next_* and avail_* values accordingly.
+  The application can consume the uncompressed output when it wants, for
+  example when the output buffer is full (avail_out == 0), or after each
+  call of inflate().
+
+    If the parameter flush is set to Z_PARTIAL_FLUSH or Z_PACKET_FLUSH,
+  inflate flushes as much output as possible to the output buffer. The
+  flushing behavior of inflate is not specified for values of the flush
+  parameter other than Z_PARTIAL_FLUSH, Z_PACKET_FLUSH or Z_FINISH, but the
+  current implementation actually flushes as much output as possible
+  anyway.  For Z_PACKET_FLUSH, inflate checks that once all the input data
+  has been consumed, it is expecting to see the length field of a stored
+  block; if not, it returns Z_DATA_ERROR.
+
+    inflate() should normally be called until it returns Z_STREAM_END or an
+  error. However if all decompression is to be performed in a single step
+  (a single call of inflate), the parameter flush should be set to
+  Z_FINISH. In this case all pending input is processed and all pending
+  output is flushed; avail_out must be large enough to hold all the
+  uncompressed data. (The size of the uncompressed data may have been saved
+  by the compressor for this purpose.) The next operation on this stream must
+  be inflateEnd to deallocate the decompression state. The use of Z_FINISH
+  is never required, but can be used to inform inflate that a faster routine
+  may be used for the single inflate() call.
+
+    inflate() returns Z_OK if some progress has been made (more input
+  processed or more output produced), Z_STREAM_END if the end of the
+  compressed data has been reached and all uncompressed output has been
+  produced, Z_DATA_ERROR if the input data was corrupted, Z_STREAM_ERROR if
+  the stream structure was inconsistent (for example if next_in or next_out
+  was NULL), Z_MEM_ERROR if there was not enough memory, Z_BUF_ERROR if no
+  progress is possible or if there was not enough room in the output buffer
+  when Z_FINISH is used. In the Z_DATA_ERROR case, the application may then
+  call inflateSync to look for a good compression block.  */
+
+
+extern int inflateEnd OF((z_stream *strm));
+/*
+     All dynamically allocated data structures for this stream are freed.
+   This function discards any unprocessed input and does not flush any
+   pending output.
+
+     inflateEnd returns Z_OK if success, Z_STREAM_ERROR if the stream state
+   was inconsistent. In the error case, msg may be set but then points to a
+   static string (which must not be deallocated).
+*/
+
+                        /* advanced functions */
+
+extern int inflateInit2 OF((z_stream *strm,
+                            int  windowBits));
+/*
+     This is another version of inflateInit with more compression options. The
+   fields next_out, zalloc and zfree must be initialized before by the caller.
+
+     The windowBits parameter is the base two logarithm of the maximum window
+   size (the size of the history buffer).  It should be in the range 8..15 for
+   this version of the library (the value 16 will be allowed soon). The
+   default value is 15 if inflateInit is used instead. If a compressed stream
+   with a larger window size is given as input, inflate() will return with
+   the error code Z_DATA_ERROR instead of trying to allocate a larger window.
+
+     If next_out is not null, the library will use this buffer for the history
+   buffer; the buffer must either be large enough to hold the entire output
+   data, or have at least 1<<windowBits bytes.  If next_out is null, the
+   library will allocate its own buffer (and leave next_out null). next_in
+   need not be provided here but must be provided by the application for the
+   next call of inflate().
+
+     If the history buffer is provided by the application, next_out must
+   never be changed by the application since the decompressor maintains
+   history information inside this buffer from call to call; the application
+   can only reset next_out to the beginning of the history buffer when
+   avail_out is zero and all output has been consumed.
+
+      inflateInit2 returns Z_OK if success, Z_MEM_ERROR if there was
+   not enough memory, Z_STREAM_ERROR if a parameter is invalid (such as
+   windowBits < 8). msg is set to null if there is no error message.
+   inflateInit2 does not perform any decompression: this will be done by
+   inflate().
+*/
+
+extern int inflateSync OF((z_stream *strm));
+/*
+    Skips invalid compressed data until the special marker (see deflate()
+  above) can be found, or until all available input is skipped. No output
+  is provided.
+
+    inflateSync returns Z_OK if the special marker has been found, Z_BUF_ERROR
+  if no more input was provided, Z_DATA_ERROR if no marker has been found,
+  or Z_STREAM_ERROR if the stream structure was inconsistent. In the success
+  case, the application may save the current current value of total_in which
+  indicates where valid compressed data was found. In the error case, the
+  application may repeatedly call inflateSync, providing more input each time,
+  until success or end of the input data.
+*/
+
+extern int inflateReset OF((z_stream *strm));
+/*
+     This function is equivalent to inflateEnd followed by inflateInit,
+   but does not free and reallocate all the internal decompression state.
+   The stream will keep attributes that may have been set by inflateInit2.
+
+      inflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source
+   stream state was inconsistent (such as zalloc or state being NULL).
+*/
+
+extern int inflateIncomp OF((z_stream *strm));
+/*
+     This function adds the data at next_in (avail_in bytes) to the output
+   history without performing any output.  There must be no pending output,
+   and the decompressor must be expecting to see the start of a block.
+   Calling this function is equivalent to decompressing a stored block
+   containing the data at next_in (except that the data is not output).
+*/
+
+                        /* checksum functions */
+
+/*
+     This function is not related to compression but is exported
+   anyway because it might be useful in applications using the
+   compression library.
+*/
+
+extern uLong adler32 OF((uLong adler, Bytef *buf, uInt len));
+
+/*
+     Update a running Adler-32 checksum with the bytes buf[0..len-1] and
+   return the updated checksum. If buf is NULL, this function returns
+   the required initial value for the checksum.
+   An Adler-32 checksum is almost as reliable as a CRC32 but can be computed
+   much faster. Usage example:
+
+     uLong adler = adler32(0L, Z_NULL, 0);
+
+     while (read_buffer(buffer, length) != EOF) {
+       adler = adler32(adler, buffer, length);
+     }
+     if (adler != original_adler) error();
+*/
+
+#ifndef _Z_UTIL_H
+    struct internal_state {int dummy;}; /* hack for buggy compilers */
+#endif
+
+#endif /* _ZLIB_H */
diff --git a/arch/xtensa/boot/lib/Makefile b/arch/xtensa/boot/lib/Makefile
new file mode 100644
index 00000000000..c0a74dc3a0d
--- /dev/null
+++ b/arch/xtensa/boot/lib/Makefile
@@ -0,0 +1,6 @@
+#
+# Makefile for some libs needed by zImage.
+#
+
+
+lib-y := zlib.o zmem.o
diff --git a/arch/xtensa/boot/lib/memcpy.S b/arch/xtensa/boot/lib/memcpy.S
new file mode 100644
index 00000000000..a029f5df2d5
--- /dev/null
+++ b/arch/xtensa/boot/lib/memcpy.S
@@ -0,0 +1,36 @@
+/*
+ *  arch/xtensa/lib/memcpy.S
+ *
+ *  ANSI C standard library function memcpy
+ *
+ *  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.
+ */
+
+#define _ASMLANGUAGE
+#include <xtensa/config/core.h>
+
+.text
+.align	4
+.global	bcopy
+.type   bcopy,@function
+bcopy:
+        movi    a14, xthal_bcopy        // a14 safe to use regardless of whether caller
+                                        //   used call4 or call8 (can't have used call12)
+        jx      a14                     // let the Core HAL do the work
+
+.text
+.align	4
+.global	memcpy
+.type   memcpy,@function
+memcpy:
+.global	memmove
+.type   memmove,@function
+memmove:
+        movi    a14, xthal_memcpy      // a14 safe to use regardless of whether caller
+                                        //   used call4 or call8 (can't have used call12)
+        jx      a14                     // let the Core HAL do the work
+
diff --git a/arch/xtensa/boot/lib/zlib.c b/arch/xtensa/boot/lib/zlib.c
new file mode 100644
index 00000000000..e3859f63107
--- /dev/null
+++ b/arch/xtensa/boot/lib/zlib.c
@@ -0,0 +1,2150 @@
+/*
+ * BK Id: SCCS/s.zlib.c 1.8 05/18/01 15:17:24 cort
+ */
+/*
+ * This file is derived from various .h and .c files from the zlib-0.95
+ * distribution by Jean-loup Gailly and Mark Adler, with some additions
+ * by Paul Mackerras to aid in implementing Deflate compression and
+ * decompression for PPP packets.  See zlib.h for conditions of
+ * distribution and use.
+ *
+ * Changes that have been made include:
+ * - changed functions not used outside this file to "local"
+ * - added minCompression parameter to deflateInit2
+ * - added Z_PACKET_FLUSH (see zlib.h for details)
+ * - added inflateIncomp
+ *
+ */
+
+/*+++++*/
+/* zutil.h -- internal interface and configuration of the compression library
+ * Copyright (C) 1995 Jean-loup Gailly.
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* WARNING: this file should *not* be used by applications. It is
+   part of the implementation of the compression library and is
+   subject to change. Applications should only use zlib.h.
+ */
+
+/* From: zutil.h,v 1.9 1995/05/03 17:27:12 jloup Exp */
+
+#define _Z_UTIL_H
+
+#include "zlib.h"
+
+#ifndef local
+#  define local static
+#endif
+/* compile with -Dlocal if your debugger can't find static symbols */
+
+#define FAR
+
+typedef unsigned char  uch;
+typedef uch FAR uchf;
+typedef unsigned short ush;
+typedef ush FAR ushf;
+typedef unsigned long  ulg;
+
+extern char *z_errmsg[]; /* indexed by 1-zlib_error */
+
+#define ERR_RETURN(strm,err) return (strm->msg=z_errmsg[1-err], err)
+/* To be used only when the state is known to be valid */
+
+#ifndef NULL
+#define NULL	((void *) 0)
+#endif
+
+        /* common constants */
+
+#define DEFLATED   8
+
+#ifndef DEF_WBITS
+#  define DEF_WBITS MAX_WBITS
+#endif
+/* default windowBits for decompression. MAX_WBITS is for compression only */
+
+#if MAX_MEM_LEVEL >= 8
+#  define DEF_MEM_LEVEL 8
+#else
+#  define DEF_MEM_LEVEL  MAX_MEM_LEVEL
+#endif
+/* default memLevel */
+
+#define STORED_BLOCK 0
+#define STATIC_TREES 1
+#define DYN_TREES    2
+/* The three kinds of block type */
+
+#define MIN_MATCH  3
+#define MAX_MATCH  258
+/* The minimum and maximum match lengths */
+
+         /* functions */
+
+#include <linux/string.h>
+#define zmemcpy memcpy
+#define zmemzero(dest, len)	memset(dest, 0, len)
+
+/* Diagnostic functions */
+#ifdef DEBUG_ZLIB
+#  include <stdio.h>
+#  ifndef verbose
+#    define verbose 0
+#  endif
+#  define Assert(cond,msg) {if(!(cond)) z_error(msg);}
+#  define Trace(x) fprintf x
+#  define Tracev(x) {if (verbose) fprintf x ;}
+#  define Tracevv(x) {if (verbose>1) fprintf x ;}
+#  define Tracec(c,x) {if (verbose && (c)) fprintf x ;}
+#  define Tracecv(c,x) {if (verbose>1 && (c)) fprintf x ;}
+#else
+#  define Assert(cond,msg)
+#  define Trace(x)
+#  define Tracev(x)
+#  define Tracevv(x)
+#  define Tracec(c,x)
+#  define Tracecv(c,x)
+#endif
+
+
+typedef uLong (*check_func) OF((uLong check, Bytef *buf, uInt len));
+
+/* voidpf zcalloc OF((voidpf opaque, unsigned items, unsigned size)); */
+/* void   zcfree  OF((voidpf opaque, voidpf ptr)); */
+
+#define ZALLOC(strm, items, size) \
+           (*((strm)->zalloc))((strm)->opaque, (items), (size))
+#define ZFREE(strm, addr, size)	\
+	   (*((strm)->zfree))((strm)->opaque, (voidpf)(addr), (size))
+#define TRY_FREE(s, p, n) {if (p) ZFREE(s, p, n);}
+
+/* deflate.h -- internal compression state
+ * Copyright (C) 1995 Jean-loup Gailly
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* WARNING: this file should *not* be used by applications. It is
+   part of the implementation of the compression library and is
+   subject to change. Applications should only use zlib.h.
+ */
+
+/*+++++*/
+/* infblock.h -- header to use infblock.c
+ * Copyright (C) 1995 Mark Adler
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* WARNING: this file should *not* be used by applications. It is
+   part of the implementation of the compression library and is
+   subject to change. Applications should only use zlib.h.
+ */
+
+struct inflate_blocks_state;
+typedef struct inflate_blocks_state FAR inflate_blocks_statef;
+
+local inflate_blocks_statef * inflate_blocks_new OF((
+    z_stream *z,
+    check_func c,               /* check function */
+    uInt w));                   /* window size */
+
+local int inflate_blocks OF((
+    inflate_blocks_statef *,
+    z_stream *,
+    int));                      /* initial return code */
+
+local void inflate_blocks_reset OF((
+    inflate_blocks_statef *,
+    z_stream *,
+    uLongf *));                  /* check value on output */
+
+local int inflate_blocks_free OF((
+    inflate_blocks_statef *,
+    z_stream *,
+    uLongf *));                  /* check value on output */
+
+local int inflate_addhistory OF((
+    inflate_blocks_statef *,
+    z_stream *));
+
+local int inflate_packet_flush OF((
+    inflate_blocks_statef *));
+
+/*+++++*/
+/* inftrees.h -- header to use inftrees.c
+ * Copyright (C) 1995 Mark Adler
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* WARNING: this file should *not* be used by applications. It is
+   part of the implementation of the compression library and is
+   subject to change. Applications should only use zlib.h.
+ */
+
+/* Huffman code lookup table entry--this entry is four bytes for machines
+   that have 16-bit pointers (e.g. PC's in the small or medium model). */
+
+typedef struct inflate_huft_s FAR inflate_huft;
+
+struct inflate_huft_s {
+  union {
+    struct {
+      Byte Exop;        /* number of extra bits or operation */
+      Byte Bits;        /* number of bits in this code or subcode */
+    } what;
+    uInt Nalloc;	/* number of these allocated here */
+    Bytef *pad;         /* pad structure to a power of 2 (4 bytes for */
+  } word;               /*  16-bit, 8 bytes for 32-bit machines) */
+  union {
+    uInt Base;          /* literal, length base, or distance base */
+    inflate_huft *Next; /* pointer to next level of table */
+  } more;
+};
+
+#ifdef DEBUG_ZLIB
+  local uInt inflate_hufts;
+#endif
+
+local int inflate_trees_bits OF((
+    uIntf *,                    /* 19 code lengths */
+    uIntf *,                    /* bits tree desired/actual depth */
+    inflate_huft * FAR *,       /* bits tree result */
+    z_stream *));               /* for zalloc, zfree functions */
+
+local int inflate_trees_dynamic OF((
+    uInt,                       /* number of literal/length codes */
+    uInt,                       /* number of distance codes */
+    uIntf *,                    /* that many (total) code lengths */
+    uIntf *,                    /* literal desired/actual bit depth */
+    uIntf *,                    /* distance desired/actual bit depth */
+    inflate_huft * FAR *,       /* literal/length tree result */
+    inflate_huft * FAR *,       /* distance tree result */
+    z_stream *));               /* for zalloc, zfree functions */
+
+local int inflate_trees_fixed OF((
+    uIntf *,                    /* literal desired/actual bit depth */
+    uIntf *,                    /* distance desired/actual bit depth */
+    inflate_huft * FAR *,       /* literal/length tree result */
+    inflate_huft * FAR *));     /* distance tree result */
+
+local int inflate_trees_free OF((
+    inflate_huft *,             /* tables to free */
+    z_stream *));               /* for zfree function */
+
+
+/*+++++*/
+/* infcodes.h -- header to use infcodes.c
+ * Copyright (C) 1995 Mark Adler
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* WARNING: this file should *not* be used by applications. It is
+   part of the implementation of the compression library and is
+   subject to change. Applications should only use zlib.h.
+ */
+
+struct inflate_codes_state;
+typedef struct inflate_codes_state FAR inflate_codes_statef;
+
+local inflate_codes_statef *inflate_codes_new OF((
+    uInt, uInt,
+    inflate_huft *, inflate_huft *,
+    z_stream *));
+
+local int inflate_codes OF((
+    inflate_blocks_statef *,
+    z_stream *,
+    int));
+
+local void inflate_codes_free OF((
+    inflate_codes_statef *,
+    z_stream *));
+
+
+/*+++++*/
+/* inflate.c -- zlib interface to inflate modules
+ * Copyright (C) 1995 Mark Adler
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* inflate private state */
+struct internal_state {
+
+  /* mode */
+  enum {
+      METHOD,   /* waiting for method byte */
+      FLAG,     /* waiting for flag byte */
+      BLOCKS,   /* decompressing blocks */
+      CHECK4,   /* four check bytes to go */
+      CHECK3,   /* three check bytes to go */
+      CHECK2,   /* two check bytes to go */
+      CHECK1,   /* one check byte to go */
+      DONE,     /* finished check, done */
+      BAD}      /* got an error--stay here */
+    mode;               /* current inflate mode */
+
+  /* mode dependent information */
+  union {
+    uInt method;        /* if FLAGS, method byte */
+    struct {
+      uLong was;                /* computed check value */
+      uLong need;               /* stream check value */
+    } check;            /* if CHECK, check values to compare */
+    uInt marker;        /* if BAD, inflateSync's marker bytes count */
+  } sub;        /* submode */
+
+  /* mode independent information */
+  int  nowrap;          /* flag for no wrapper */
+  uInt wbits;           /* log2(window size)  (8..15, defaults to 15) */
+  inflate_blocks_statef
+    *blocks;            /* current inflate_blocks state */
+
+};
+
+
+int inflateReset(z)
+z_stream *z;
+{
+  uLong c;
+
+  if (z == Z_NULL || z->state == Z_NULL)
+    return Z_STREAM_ERROR;
+  z->total_in = z->total_out = 0;
+  z->msg = Z_NULL;
+  z->state->mode = z->state->nowrap ? BLOCKS : METHOD;
+  inflate_blocks_reset(z->state->blocks, z, &c);
+  Trace((stderr, "inflate: reset\n"));
+  return Z_OK;
+}
+
+
+int inflateEnd(z)
+z_stream *z;
+{
+  uLong c;
+
+  if (z == Z_NULL || z->state == Z_NULL || z->zfree == Z_NULL)
+    return Z_STREAM_ERROR;
+  if (z->state->blocks != Z_NULL)
+    inflate_blocks_free(z->state->blocks, z, &c);
+  ZFREE(z, z->state, sizeof(struct internal_state));
+  z->state = Z_NULL;
+  Trace((stderr, "inflate: end\n"));
+  return Z_OK;
+}
+
+
+int inflateInit2(z, w)
+z_stream *z;
+int w;
+{
+  /* initialize state */
+  if (z == Z_NULL)
+    return Z_STREAM_ERROR;
+/*  if (z->zalloc == Z_NULL) z->zalloc = zcalloc; */
+/*  if (z->zfree == Z_NULL) z->zfree = zcfree; */
+  if ((z->state = (struct internal_state FAR *)
+       ZALLOC(z,1,sizeof(struct internal_state))) == Z_NULL)
+    return Z_MEM_ERROR;
+  z->state->blocks = Z_NULL;
+
+  /* handle undocumented nowrap option (no zlib header or check) */
+  z->state->nowrap = 0;
+  if (w < 0)
+  {
+    w = - w;
+    z->state->nowrap = 1;
+  }
+
+  /* set window size */
+  if (w < 8 || w > 15)
+  {
+    inflateEnd(z);
+    return Z_STREAM_ERROR;
+  }
+  z->state->wbits = (uInt)w;
+
+  /* create inflate_blocks state */
+  if ((z->state->blocks =
+       inflate_blocks_new(z, z->state->nowrap ? Z_NULL : adler32, 1 << w))
+      == Z_NULL)
+  {
+    inflateEnd(z);
+    return Z_MEM_ERROR;
+  }
+  Trace((stderr, "inflate: allocated\n"));
+
+  /* reset state */
+  inflateReset(z);
+  return Z_OK;
+}
+
+
+int inflateInit(z)
+z_stream *z;
+{
+  return inflateInit2(z, DEF_WBITS);
+}
+
+
+#define NEEDBYTE {if(z->avail_in==0)goto empty;r=Z_OK;}
+#define NEXTBYTE (z->avail_in--,z->total_in++,*z->next_in++)
+
+int inflate(z, f)
+z_stream *z;
+int f;
+{
+  int r;
+  uInt b;
+
+  if (z == Z_NULL || z->next_in == Z_NULL)
+    return Z_STREAM_ERROR;
+  r = Z_BUF_ERROR;
+  while (1) switch (z->state->mode)
+  {
+    case METHOD:
+      NEEDBYTE
+      if (((z->state->sub.method = NEXTBYTE) & 0xf) != DEFLATED)
+      {
+        z->state->mode = BAD;
+        z->msg = "unknown compression method";
+        z->state->sub.marker = 5;       /* can't try inflateSync */
+        break;
+      }
+      if ((z->state->sub.method >> 4) + 8 > z->state->wbits)
+      {
+        z->state->mode = BAD;
+        z->msg = "invalid window size";
+        z->state->sub.marker = 5;       /* can't try inflateSync */
+        break;
+      }
+      z->state->mode = FLAG;
+    case FLAG:
+      NEEDBYTE
+      if ((b = NEXTBYTE) & 0x20)
+      {
+        z->state->mode = BAD;
+        z->msg = "invalid reserved bit";
+        z->state->sub.marker = 5;       /* can't try inflateSync */
+        break;
+      }
+      if (((z->state->sub.method << 8) + b) % 31)
+      {
+        z->state->mode = BAD;
+        z->msg = "incorrect header check";
+        z->state->sub.marker = 5;       /* can't try inflateSync */
+        break;
+      }
+      Trace((stderr, "inflate: zlib header ok\n"));
+      z->state->mode = BLOCKS;
+    case BLOCKS:
+      r = inflate_blocks(z->state->blocks, z, r);
+      if (f == Z_PACKET_FLUSH && z->avail_in == 0 && z->avail_out != 0)
+	  r = inflate_packet_flush(z->state->blocks);
+      if (r == Z_DATA_ERROR)
+      {
+        z->state->mode = BAD;
+        z->state->sub.marker = 0;       /* can try inflateSync */
+        break;
+      }
+      if (r != Z_STREAM_END)
+        return r;
+      r = Z_OK;
+      inflate_blocks_reset(z->state->blocks, z, &z->state->sub.check.was);
+      if (z->state->nowrap)
+      {
+        z->state->mode = DONE;
+        break;
+      }
+      z->state->mode = CHECK4;
+    case CHECK4:
+      NEEDBYTE
+      z->state->sub.check.need = (uLong)NEXTBYTE << 24;
+      z->state->mode = CHECK3;
+    case CHECK3:
+      NEEDBYTE
+      z->state->sub.check.need += (uLong)NEXTBYTE << 16;
+      z->state->mode = CHECK2;
+    case CHECK2:
+      NEEDBYTE
+      z->state->sub.check.need += (uLong)NEXTBYTE << 8;
+      z->state->mode = CHECK1;
+    case CHECK1:
+      NEEDBYTE
+      z->state->sub.check.need += (uLong)NEXTBYTE;
+
+      if (z->state->sub.check.was != z->state->sub.check.need)
+      {
+        z->state->mode = BAD;
+        z->msg = "incorrect data check";
+        z->state->sub.marker = 5;       /* can't try inflateSync */
+        break;
+      }
+      Trace((stderr, "inflate: zlib check ok\n"));
+      z->state->mode = DONE;
+    case DONE:
+      return Z_STREAM_END;
+    case BAD:
+      return Z_DATA_ERROR;
+    default:
+      return Z_STREAM_ERROR;
+  }
+
+ empty:
+  if (f != Z_PACKET_FLUSH)
+    return r;
+  z->state->mode = BAD;
+  z->state->sub.marker = 0;       /* can try inflateSync */
+  return Z_DATA_ERROR;
+}
+
+/*
+ * This subroutine adds the data at next_in/avail_in to the output history
+ * without performing any output.  The output buffer must be "caught up";
+ * i.e. no pending output (hence s->read equals s->write), and the state must
+ * be BLOCKS (i.e. we should be willing to see the start of a series of
+ * BLOCKS).  On exit, the output will also be caught up, and the checksum
+ * will have been updated if need be.
+ */
+
+int inflateIncomp(z)
+z_stream *z;
+{
+    if (z->state->mode != BLOCKS)
+	return Z_DATA_ERROR;
+    return inflate_addhistory(z->state->blocks, z);
+}
+
+
+int inflateSync(z)
+z_stream *z;
+{
+  uInt n;       /* number of bytes to look at */
+  Bytef *p;     /* pointer to bytes */
+  uInt m;       /* number of marker bytes found in a row */
+  uLong r, w;   /* temporaries to save total_in and total_out */
+
+  /* set up */
+  if (z == Z_NULL || z->state == Z_NULL)
+    return Z_STREAM_ERROR;
+  if (z->state->mode != BAD)
+  {
+    z->state->mode = BAD;
+    z->state->sub.marker = 0;
+  }
+  if ((n = z->avail_in) == 0)
+    return Z_BUF_ERROR;
+  p = z->next_in;
+  m = z->state->sub.marker;
+
+  /* search */
+  while (n && m < 4)
+  {
+    if (*p == (Byte)(m < 2 ? 0 : 0xff))
+      m++;
+    else if (*p)
+      m = 0;
+    else
+      m = 4 - m;
+    p++, n--;
+  }
+
+  /* restore */
+  z->total_in += p - z->next_in;
+  z->next_in = p;
+  z->avail_in = n;
+  z->state->sub.marker = m;
+
+  /* return no joy or set up to restart on a new block */
+  if (m != 4)
+    return Z_DATA_ERROR;
+  r = z->total_in;  w = z->total_out;
+  inflateReset(z);
+  z->total_in = r;  z->total_out = w;
+  z->state->mode = BLOCKS;
+  return Z_OK;
+}
+
+#undef NEEDBYTE
+#undef NEXTBYTE
+
+/*+++++*/
+/* infutil.h -- types and macros common to blocks and codes
+ * Copyright (C) 1995 Mark Adler
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* WARNING: this file should *not* be used by applications. It is
+   part of the implementation of the compression library and is
+   subject to change. Applications should only use zlib.h.
+ */
+
+/* inflate blocks semi-private state */
+struct inflate_blocks_state {
+
+  /* mode */
+  enum {
+      TYPE,     /* get type bits (3, including end bit) */
+      LENS,     /* get lengths for stored */
+      STORED,   /* processing stored block */
+      TABLE,    /* get table lengths */
+      BTREE,    /* get bit lengths tree for a dynamic block */
+      DTREE,    /* get length, distance trees for a dynamic block */
+      CODES,    /* processing fixed or dynamic block */
+      DRY,      /* output remaining window bytes */
+      DONEB,     /* finished last block, done */
+      BADB}      /* got a data error--stuck here */
+    mode;               /* current inflate_block mode */
+
+  /* mode dependent information */
+  union {
+    uInt left;          /* if STORED, bytes left to copy */
+    struct {
+      uInt table;               /* table lengths (14 bits) */
+      uInt index;               /* index into blens (or border) */
+      uIntf *blens;             /* bit lengths of codes */
+      uInt bb;                  /* bit length tree depth */
+      inflate_huft *tb;         /* bit length decoding tree */
+      int nblens;		/* # elements allocated at blens */
+    } trees;            /* if DTREE, decoding info for trees */
+    struct {
+      inflate_huft *tl, *td;    /* trees to free */
+      inflate_codes_statef
+         *codes;
+    } decode;           /* if CODES, current state */
+  } sub;                /* submode */
+  uInt last;            /* true if this block is the last block */
+
+  /* mode independent information */
+  uInt bitk;            /* bits in bit buffer */
+  uLong bitb;           /* bit buffer */
+  Bytef *window;        /* sliding window */
+  Bytef *end;           /* one byte after sliding window */
+  Bytef *read;          /* window read pointer */
+  Bytef *write;         /* window write pointer */
+  check_func checkfn;   /* check function */
+  uLong check;          /* check on output */
+
+};
+
+
+/* defines for inflate input/output */
+/*   update pointers and return */
+#define UPDBITS {s->bitb=b;s->bitk=k;}
+#define UPDIN {z->avail_in=n;z->total_in+=p-z->next_in;z->next_in=p;}
+#define UPDOUT {s->write=q;}
+#define UPDATE {UPDBITS UPDIN UPDOUT}
+#define LEAVE {UPDATE return inflate_flush(s,z,r);}
+/*   get bytes and bits */
+#define LOADIN {p=z->next_in;n=z->avail_in;b=s->bitb;k=s->bitk;}
+#define NEEDBYTE {if(n)r=Z_OK;else LEAVE}
+#define NEXTBYTE (n--,*p++)
+#define NEEDBITS(j) {while(k<(j)){NEEDBYTE;b|=((uLong)NEXTBYTE)<<k;k+=8;}}
+#define DUMPBITS(j) {b>>=(j);k-=(j);}
+/*   output bytes */
+#define WAVAIL (q<s->read?s->read-q-1:s->end-q)
+#define LOADOUT {q=s->write;m=WAVAIL;}
+#define WRAP {if(q==s->end&&s->read!=s->window){q=s->window;m=WAVAIL;}}
+#define FLUSH {UPDOUT r=inflate_flush(s,z,r); LOADOUT}
+#define NEEDOUT {if(m==0){WRAP if(m==0){FLUSH WRAP if(m==0) LEAVE}}r=Z_OK;}
+#define OUTBYTE(a) {*q++=(Byte)(a);m--;}
+/*   load local pointers */
+#define LOAD {LOADIN LOADOUT}
+
+/*
+ * The IBM 150 firmware munges the data right after _etext[].  This
+ * protects it. -- Cort
+ */
+local uInt protect_mask[] = {0, 0, 0, 0, 0, 0, 0, 0, 0 ,0 ,0 ,0};
+/* And'ing with mask[n] masks the lower n bits */
+local uInt inflate_mask[] = {
+    0x0000,
+    0x0001, 0x0003, 0x0007, 0x000f, 0x001f, 0x003f, 0x007f, 0x00ff,
+    0x01ff, 0x03ff, 0x07ff, 0x0fff, 0x1fff, 0x3fff, 0x7fff, 0xffff
+};
+
+/* copy as much as possible from the sliding window to the output area */
+local int inflate_flush OF((
+    inflate_blocks_statef *,
+    z_stream *,
+    int));
+
+/*+++++*/
+/* inffast.h -- header to use inffast.c
+ * Copyright (C) 1995 Mark Adler
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* WARNING: this file should *not* be used by applications. It is
+   part of the implementation of the compression library and is
+   subject to change. Applications should only use zlib.h.
+ */
+
+local int inflate_fast OF((
+    uInt,
+    uInt,
+    inflate_huft *,
+    inflate_huft *,
+    inflate_blocks_statef *,
+    z_stream *));
+
+
+/*+++++*/
+/* infblock.c -- interpret and process block types to last block
+ * Copyright (C) 1995 Mark Adler
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* Table for deflate from PKZIP's appnote.txt. */
+local uInt border[] = { /* Order of the bit length code lengths */
+        16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15};
+
+/*
+   Notes beyond the 1.93a appnote.txt:
+
+   1. Distance pointers never point before the beginning of the output
+      stream.
+   2. Distance pointers can point back across blocks, up to 32k away.
+   3. There is an implied maximum of 7 bits for the bit length table and
+      15 bits for the actual data.
+   4. If only one code exists, then it is encoded using one bit.  (Zero
+      would be more efficient, but perhaps a little confusing.)  If two
+      codes exist, they are coded using one bit each (0 and 1).
+   5. There is no way of sending zero distance codes--a dummy must be
+      sent if there are none.  (History: a pre 2.0 version of PKZIP would
+      store blocks with no distance codes, but this was discovered to be
+      too harsh a criterion.)  Valid only for 1.93a.  2.04c does allow
+      zero distance codes, which is sent as one code of zero bits in
+      length.
+   6. There are up to 286 literal/length codes.  Code 256 represents the
+      end-of-block.  Note however that the static length tree defines
+      288 codes just to fill out the Huffman codes.  Codes 286 and 287
+      cannot be used though, since there is no length base or extra bits
+      defined for them.  Similarily, there are up to 30 distance codes.
+      However, static trees define 32 codes (all 5 bits) to fill out the
+      Huffman codes, but the last two had better not show up in the data.
+   7. Unzip can check dynamic Huffman blocks for complete code sets.
+      The exception is that a single code would not be complete (see #4).
+   8. The five bits following the block type is really the number of
+      literal codes sent minus 257.
+   9. Length codes 8,16,16 are interpreted as 13 length codes of 8 bits
+      (1+6+6).  Therefore, to output three times the length, you output
+      three codes (1+1+1), whereas to output four times the same length,
+      you only need two codes (1+3).  Hmm.
+  10. In the tree reconstruction algorithm, Code = Code + Increment
+      only if BitLength(i) is not zero.  (Pretty obvious.)
+  11. Correction: 4 Bits: # of Bit Length codes - 4     (4 - 19)
+  12. Note: length code 284 can represent 227-258, but length code 285
+      really is 258.  The last length deserves its own, short code
+      since it gets used a lot in very redundant files.  The length
+      258 is special since 258 - 3 (the min match length) is 255.
+  13. The literal/length and distance code bit lengths are read as a
+      single stream of lengths.  It is possible (and advantageous) for
+      a repeat code (16, 17, or 18) to go across the boundary between
+      the two sets of lengths.
+ */
+
+
+local void inflate_blocks_reset(s, z, c)
+inflate_blocks_statef *s;
+z_stream *z;
+uLongf *c;
+{
+  if (s->checkfn != Z_NULL)
+    *c = s->check;
+  if (s->mode == BTREE || s->mode == DTREE)
+    ZFREE(z, s->sub.trees.blens, s->sub.trees.nblens * sizeof(uInt));
+  if (s->mode == CODES)
+  {
+    inflate_codes_free(s->sub.decode.codes, z);
+    inflate_trees_free(s->sub.decode.td, z);
+    inflate_trees_free(s->sub.decode.tl, z);
+  }
+  s->mode = TYPE;
+  s->bitk = 0;
+  s->bitb = 0;
+  s->read = s->write = s->window;
+  if (s->checkfn != Z_NULL)
+    s->check = (*s->checkfn)(0L, Z_NULL, 0);
+  Trace((stderr, "inflate:   blocks reset\n"));
+}
+
+
+local inflate_blocks_statef *inflate_blocks_new(z, c, w)
+z_stream *z;
+check_func c;
+uInt w;
+{
+  inflate_blocks_statef *s;
+
+  if ((s = (inflate_blocks_statef *)ZALLOC
+       (z,1,sizeof(struct inflate_blocks_state))) == Z_NULL)
+    return s;
+  if ((s->window = (Bytef *)ZALLOC(z, 1, w)) == Z_NULL)
+  {
+    ZFREE(z, s, sizeof(struct inflate_blocks_state));
+    return Z_NULL;
+  }
+  s->end = s->window + w;
+  s->checkfn = c;
+  s->mode = TYPE;
+  Trace((stderr, "inflate:   blocks allocated\n"));
+  inflate_blocks_reset(s, z, &s->check);
+  return s;
+}
+
+
+local int inflate_blocks(s, z, r)
+inflate_blocks_statef *s;
+z_stream *z;
+int r;
+{
+  uInt t;               /* temporary storage */
+  uLong b;              /* bit buffer */
+  uInt k;               /* bits in bit buffer */
+  Bytef *p;             /* input data pointer */
+  uInt n;               /* bytes available there */
+  Bytef *q;             /* output window write pointer */
+  uInt m;               /* bytes to end of window or read pointer */
+
+  /* copy input/output information to locals (UPDATE macro restores) */
+  LOAD
+
+  /* process input based on current state */
+  while (1) switch (s->mode)
+  {
+    case TYPE:
+      NEEDBITS(3)
+      t = (uInt)b & 7;
+      s->last = t & 1;
+      switch (t >> 1)
+      {
+        case 0:                         /* stored */
+          Trace((stderr, "inflate:     stored block%s\n",
+                 s->last ? " (last)" : ""));
+          DUMPBITS(3)
+          t = k & 7;                    /* go to byte boundary */
+          DUMPBITS(t)
+          s->mode = LENS;               /* get length of stored block */
+          break;
+        case 1:                         /* fixed */
+          Trace((stderr, "inflate:     fixed codes block%s\n",
+                 s->last ? " (last)" : ""));
+          {
+            uInt bl, bd;
+            inflate_huft *tl, *td;
+
+            inflate_trees_fixed(&bl, &bd, &tl, &td);
+            s->sub.decode.codes = inflate_codes_new(bl, bd, tl, td, z);
+            if (s->sub.decode.codes == Z_NULL)
+            {
+              r = Z_MEM_ERROR;
+              LEAVE
+            }
+            s->sub.decode.tl = Z_NULL;  /* don't try to free these */
+            s->sub.decode.td = Z_NULL;
+          }
+          DUMPBITS(3)
+          s->mode = CODES;
+          break;
+        case 2:                         /* dynamic */
+          Trace((stderr, "inflate:     dynamic codes block%s\n",
+                 s->last ? " (last)" : ""));
+          DUMPBITS(3)
+          s->mode = TABLE;
+          break;
+        case 3:                         /* illegal */
+          DUMPBITS(3)
+          s->mode = BADB;
+          z->msg = "invalid block type";
+          r = Z_DATA_ERROR;
+          LEAVE
+      }
+      break;
+    case LENS:
+      NEEDBITS(32)
+      if (((~b) >> 16) != (b & 0xffff))
+      {
+        s->mode = BADB;
+        z->msg = "invalid stored block lengths";
+        r = Z_DATA_ERROR;
+        LEAVE
+      }
+      s->sub.left = (uInt)b & 0xffff;
+      b = k = 0;                      /* dump bits */
+      Tracev((stderr, "inflate:       stored length %u\n", s->sub.left));
+      s->mode = s->sub.left ? STORED : TYPE;
+      break;
+    case STORED:
+      if (n == 0)
+        LEAVE
+      NEEDOUT
+      t = s->sub.left;
+      if (t > n) t = n;
+      if (t > m) t = m;
+      zmemcpy(q, p, t);
+      p += t;  n -= t;
+      q += t;  m -= t;
+      if ((s->sub.left -= t) != 0)
+        break;
+      Tracev((stderr, "inflate:       stored end, %lu total out\n",
+              z->total_out + (q >= s->read ? q - s->read :
+              (s->end - s->read) + (q - s->window))));
+      s->mode = s->last ? DRY : TYPE;
+      break;
+    case TABLE:
+      NEEDBITS(14)
+      s->sub.trees.table = t = (uInt)b & 0x3fff;
+#ifndef PKZIP_BUG_WORKAROUND
+      if ((t & 0x1f) > 29 || ((t >> 5) & 0x1f) > 29)
+      {
+        s->mode = BADB;
+        z->msg = "too many length or distance symbols";
+        r = Z_DATA_ERROR;
+        LEAVE
+      }
+#endif
+      t = 258 + (t & 0x1f) + ((t >> 5) & 0x1f);
+      if (t < 19)
+        t = 19;
+      if ((s->sub.trees.blens = (uIntf*)ZALLOC(z, t, sizeof(uInt))) == Z_NULL)
+      {
+        r = Z_MEM_ERROR;
+        LEAVE
+      }
+      s->sub.trees.nblens = t;
+      DUMPBITS(14)
+      s->sub.trees.index = 0;
+      Tracev((stderr, "inflate:       table sizes ok\n"));
+      s->mode = BTREE;
+    case BTREE:
+      while (s->sub.trees.index < 4 + (s->sub.trees.table >> 10))
+      {
+        NEEDBITS(3)
+        s->sub.trees.blens[border[s->sub.trees.index++]] = (uInt)b & 7;
+        DUMPBITS(3)
+      }
+      while (s->sub.trees.index < 19)
+        s->sub.trees.blens[border[s->sub.trees.index++]] = 0;
+      s->sub.trees.bb = 7;
+      t = inflate_trees_bits(s->sub.trees.blens, &s->sub.trees.bb,
+                             &s->sub.trees.tb, z);
+      if (t != Z_OK)
+      {
+        r = t;
+        if (r == Z_DATA_ERROR)
+          s->mode = BADB;
+        LEAVE
+      }
+      s->sub.trees.index = 0;
+      Tracev((stderr, "inflate:       bits tree ok\n"));
+      s->mode = DTREE;
+    case DTREE:
+      while (t = s->sub.trees.table,
+             s->sub.trees.index < 258 + (t & 0x1f) + ((t >> 5) & 0x1f))
+      {
+        inflate_huft *h;
+        uInt i, j, c;
+
+        t = s->sub.trees.bb;
+        NEEDBITS(t)
+        h = s->sub.trees.tb + ((uInt)b & inflate_mask[t]);
+        t = h->word.what.Bits;
+        c = h->more.Base;
+        if (c < 16)
+        {
+          DUMPBITS(t)
+          s->sub.trees.blens[s->sub.trees.index++] = c;
+        }
+        else /* c == 16..18 */
+        {
+          i = c == 18 ? 7 : c - 14;
+          j = c == 18 ? 11 : 3;
+          NEEDBITS(t + i)
+          DUMPBITS(t)
+          j += (uInt)b & inflate_mask[i];
+          DUMPBITS(i)
+          i = s->sub.trees.index;
+          t = s->sub.trees.table;
+          if (i + j > 258 + (t & 0x1f) + ((t >> 5) & 0x1f) ||
+              (c == 16 && i < 1))
+          {
+            s->mode = BADB;
+            z->msg = "invalid bit length repeat";
+            r = Z_DATA_ERROR;
+            LEAVE
+          }
+          c = c == 16 ? s->sub.trees.blens[i - 1] : 0;
+          do {
+            s->sub.trees.blens[i++] = c;
+          } while (--j);
+          s->sub.trees.index = i;
+        }
+      }
+      inflate_trees_free(s->sub.trees.tb, z);
+      s->sub.trees.tb = Z_NULL;
+      {
+        uInt bl, bd;
+        inflate_huft *tl, *td;
+        inflate_codes_statef *c;
+
+        bl = 9;         /* must be <= 9 for lookahead assumptions */
+        bd = 6;         /* must be <= 9 for lookahead assumptions */
+        t = s->sub.trees.table;
+        t = inflate_trees_dynamic(257 + (t & 0x1f), 1 + ((t >> 5) & 0x1f),
+                                  s->sub.trees.blens, &bl, &bd, &tl, &td, z);
+        if (t != Z_OK)
+        {
+          if (t == (uInt)Z_DATA_ERROR)
+            s->mode = BADB;
+          r = t;
+          LEAVE
+        }
+        Tracev((stderr, "inflate:       trees ok\n"));
+        if ((c = inflate_codes_new(bl, bd, tl, td, z)) == Z_NULL)
+        {
+          inflate_trees_free(td, z);
+          inflate_trees_free(tl, z);
+          r = Z_MEM_ERROR;
+          LEAVE
+        }
+        ZFREE(z, s->sub.trees.blens, s->sub.trees.nblens * sizeof(uInt));
+        s->sub.decode.codes = c;
+        s->sub.decode.tl = tl;
+        s->sub.decode.td = td;
+      }
+      s->mode = CODES;
+    case CODES:
+      UPDATE
+      if ((r = inflate_codes(s, z, r)) != Z_STREAM_END)
+        return inflate_flush(s, z, r);
+      r = Z_OK;
+      inflate_codes_free(s->sub.decode.codes, z);
+      inflate_trees_free(s->sub.decode.td, z);
+      inflate_trees_free(s->sub.decode.tl, z);
+      LOAD
+      Tracev((stderr, "inflate:       codes end, %lu total out\n",
+              z->total_out + (q >= s->read ? q - s->read :
+              (s->end - s->read) + (q - s->window))));
+      if (!s->last)
+      {
+        s->mode = TYPE;
+        break;
+      }
+      if (k > 7)              /* return unused byte, if any */
+      {
+        Assert(k < 16, "inflate_codes grabbed too many bytes")
+        k -= 8;
+        n++;
+        p--;                    /* can always return one */
+      }
+      s->mode = DRY;
+    case DRY:
+      FLUSH
+      if (s->read != s->write)
+        LEAVE
+      s->mode = DONEB;
+    case DONEB:
+      r = Z_STREAM_END;
+      LEAVE
+    case BADB:
+      r = Z_DATA_ERROR;
+      LEAVE
+    default:
+      r = Z_STREAM_ERROR;
+      LEAVE
+  }
+}
+
+
+local int inflate_blocks_free(s, z, c)
+inflate_blocks_statef *s;
+z_stream *z;
+uLongf *c;
+{
+  inflate_blocks_reset(s, z, c);
+  ZFREE(z, s->window, s->end - s->window);
+  ZFREE(z, s, sizeof(struct inflate_blocks_state));
+  Trace((stderr, "inflate:   blocks freed\n"));
+  return Z_OK;
+}
+
+/*
+ * This subroutine adds the data at next_in/avail_in to the output history
+ * without performing any output.  The output buffer must be "caught up";
+ * i.e. no pending output (hence s->read equals s->write), and the state must
+ * be BLOCKS (i.e. we should be willing to see the start of a series of
+ * BLOCKS).  On exit, the output will also be caught up, and the checksum
+ * will have been updated if need be.
+ */
+local int inflate_addhistory(s, z)
+inflate_blocks_statef *s;
+z_stream *z;
+{
+    uLong b;              /* bit buffer */  /* NOT USED HERE */
+    uInt k;               /* bits in bit buffer */ /* NOT USED HERE */
+    uInt t;               /* temporary storage */
+    Bytef *p;             /* input data pointer */
+    uInt n;               /* bytes available there */
+    Bytef *q;             /* output window write pointer */
+    uInt m;               /* bytes to end of window or read pointer */
+
+    if (s->read != s->write)
+	return Z_STREAM_ERROR;
+    if (s->mode != TYPE)
+	return Z_DATA_ERROR;
+
+    /* we're ready to rock */
+    LOAD
+    /* while there is input ready, copy to output buffer, moving
+     * pointers as needed.
+     */
+    while (n) {
+	t = n;  /* how many to do */
+	/* is there room until end of buffer? */
+	if (t > m) t = m;
+	/* update check information */
+	if (s->checkfn != Z_NULL)
+	    s->check = (*s->checkfn)(s->check, q, t);
+	zmemcpy(q, p, t);
+	q += t;
+	p += t;
+	n -= t;
+	z->total_out += t;
+	s->read = q;    /* drag read pointer forward */
+/*      WRAP  */ 	/* expand WRAP macro by hand to handle s->read */
+	if (q == s->end) {
+	    s->read = q = s->window;
+	    m = WAVAIL;
+	}
+    }
+    UPDATE
+    return Z_OK;
+}
+
+
+/*
+ * At the end of a Deflate-compressed PPP packet, we expect to have seen
+ * a `stored' block type value but not the (zero) length bytes.
+ */
+local int inflate_packet_flush(s)
+    inflate_blocks_statef *s;
+{
+    if (s->mode != LENS)
+	return Z_DATA_ERROR;
+    s->mode = TYPE;
+    return Z_OK;
+}
+
+
+/*+++++*/
+/* inftrees.c -- generate Huffman trees for efficient decoding
+ * Copyright (C) 1995 Mark Adler
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* simplify the use of the inflate_huft type with some defines */
+#define base more.Base
+#define next more.Next
+#define exop word.what.Exop
+#define bits word.what.Bits
+
+
+local int huft_build OF((
+    uIntf *,            /* code lengths in bits */
+    uInt,               /* number of codes */
+    uInt,               /* number of "simple" codes */
+    uIntf *,            /* list of base values for non-simple codes */
+    uIntf *,            /* list of extra bits for non-simple codes */
+    inflate_huft * FAR*,/* result: starting table */
+    uIntf *,            /* maximum lookup bits (returns actual) */
+    z_stream *));       /* for zalloc function */
+
+local voidpf falloc OF((
+    voidpf,             /* opaque pointer (not used) */
+    uInt,               /* number of items */
+    uInt));             /* size of item */
+
+local void ffree OF((
+    voidpf q,           /* opaque pointer (not used) */
+    voidpf p,           /* what to free (not used) */
+    uInt n));		/* number of bytes (not used) */
+
+/* Tables for deflate from PKZIP's appnote.txt. */
+local uInt cplens[] = { /* Copy lengths for literal codes 257..285 */
+        3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 15, 17, 19, 23, 27, 31,
+        35, 43, 51, 59, 67, 83, 99, 115, 131, 163, 195, 227, 258, 0, 0};
+        /* actually lengths - 2; also see note #13 above about 258 */
+local uInt cplext[] = { /* Extra bits for literal codes 257..285 */
+        0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2,
+        3, 3, 3, 3, 4, 4, 4, 4, 5, 5, 5, 5, 0, 192, 192}; /* 192==invalid */
+local uInt cpdist[] = { /* Copy offsets for distance codes 0..29 */
+        1, 2, 3, 4, 5, 7, 9, 13, 17, 25, 33, 49, 65, 97, 129, 193,
+        257, 385, 513, 769, 1025, 1537, 2049, 3073, 4097, 6145,
+        8193, 12289, 16385, 24577};
+local uInt cpdext[] = { /* Extra bits for distance codes */
+        0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6,
+        7, 7, 8, 8, 9, 9, 10, 10, 11, 11,
+        12, 12, 13, 13};
+
+/*
+   Huffman code decoding is performed using a multi-level table lookup.
+   The fastest way to decode is to simply build a lookup table whose
+   size is determined by the longest code.  However, the time it takes
+   to build this table can also be a factor if the data being decoded
+   is not very long.  The most common codes are necessarily the
+   shortest codes, so those codes dominate the decoding time, and hence
+   the speed.  The idea is you can have a shorter table that decodes the
+   shorter, more probable codes, and then point to subsidiary tables for
+   the longer codes.  The time it costs to decode the longer codes is
+   then traded against the time it takes to make longer tables.
+
+   This results of this trade are in the variables lbits and dbits
+   below.  lbits is the number of bits the first level table for literal/
+   length codes can decode in one step, and dbits is the same thing for
+   the distance codes.  Subsequent tables are also less than or equal to
+   those sizes.  These values may be adjusted either when all of the
+   codes are shorter than that, in which case the longest code length in
+   bits is used, or when the shortest code is *longer* than the requested
+   table size, in which case the length of the shortest code in bits is
+   used.
+
+   There are two different values for the two tables, since they code a
+   different number of possibilities each.  The literal/length table
+   codes 286 possible values, or in a flat code, a little over eight
+   bits.  The distance table codes 30 possible values, or a little less
+   than five bits, flat.  The optimum values for speed end up being
+   about one bit more than those, so lbits is 8+1 and dbits is 5+1.
+   The optimum values may differ though from machine to machine, and
+   possibly even between compilers.  Your mileage may vary.
+ */
+
+
+/* If BMAX needs to be larger than 16, then h and x[] should be uLong. */
+#define BMAX 15         /* maximum bit length of any code */
+#define N_MAX 288       /* maximum number of codes in any set */
+
+#ifdef DEBUG_ZLIB
+  uInt inflate_hufts;
+#endif
+
+local int huft_build(b, n, s, d, e, t, m, zs)
+uIntf *b;               /* code lengths in bits (all assumed <= BMAX) */
+uInt n;                 /* number of codes (assumed <= N_MAX) */
+uInt s;                 /* number of simple-valued codes (0..s-1) */
+uIntf *d;               /* list of base values for non-simple codes */
+uIntf *e;               /* list of extra bits for non-simple codes */
+inflate_huft * FAR *t;  /* result: starting table */
+uIntf *m;               /* maximum lookup bits, returns actual */
+z_stream *zs;           /* for zalloc function */
+/* Given a list of code lengths and a maximum table size, make a set of
+   tables to decode that set of codes.  Return Z_OK on success, Z_BUF_ERROR
+   if the given code set is incomplete (the tables are still built in this
+   case), Z_DATA_ERROR if the input is invalid (all zero length codes or an
+   over-subscribed set of lengths), or Z_MEM_ERROR if not enough memory. */
+{
+
+  uInt a;                       /* counter for codes of length k */
+  uInt c[BMAX+1];               /* bit length count table */
+  uInt f;                       /* i repeats in table every f entries */
+  int g;                        /* maximum code length */
+  int h;                        /* table level */
+  register uInt i;              /* counter, current code */
+  register uInt j;              /* counter */
+  register int k;               /* number of bits in current code */
+  int l;                        /* bits per table (returned in m) */
+  register uIntf *p;            /* pointer into c[], b[], or v[] */
+  inflate_huft *q;              /* points to current table */
+  struct inflate_huft_s r;      /* table entry for structure assignment */
+  inflate_huft *u[BMAX];        /* table stack */
+  uInt v[N_MAX];                /* values in order of bit length */
+  register int w;               /* bits before this table == (l * h) */
+  uInt x[BMAX+1];               /* bit offsets, then code stack */
+  uIntf *xp;                    /* pointer into x */
+  int y;                        /* number of dummy codes added */
+  uInt z;                       /* number of entries in current table */
+
+
+  /* Generate counts for each bit length */
+  p = c;
+#define C0 *p++ = 0;
+#define C2 C0 C0 C0 C0
+#define C4 C2 C2 C2 C2
+  C4                            /* clear c[]--assume BMAX+1 is 16 */
+  p = b;  i = n;
+  do {
+    c[*p++]++;                  /* assume all entries <= BMAX */
+  } while (--i);
+  if (c[0] == n)                /* null input--all zero length codes */
+  {
+    *t = (inflate_huft *)Z_NULL;
+    *m = 0;
+    return Z_OK;
+  }
+
+
+  /* Find minimum and maximum length, bound *m by those */
+  l = *m;
+  for (j = 1; j <= BMAX; j++)
+    if (c[j])
+      break;
+  k = j;                        /* minimum code length */
+  if ((uInt)l < j)
+    l = j;
+  for (i = BMAX; i; i--)
+    if (c[i])
+      break;
+  g = i;                        /* maximum code length */
+  if ((uInt)l > i)
+    l = i;
+  *m = l;
+
+
+  /* Adjust last length count to fill out codes, if needed */
+  for (y = 1 << j; j < i; j++, y <<= 1)
+    if ((y -= c[j]) < 0)
+      return Z_DATA_ERROR;
+  if ((y -= c[i]) < 0)
+    return Z_DATA_ERROR;
+  c[i] += y;
+
+
+  /* Generate starting offsets into the value table for each length */
+  x[1] = j = 0;
+  p = c + 1;  xp = x + 2;
+  while (--i) {                 /* note that i == g from above */
+    *xp++ = (j += *p++);
+  }
+
+
+  /* Make a table of values in order of bit lengths */
+  p = b;  i = 0;
+  do {
+    if ((j = *p++) != 0)
+      v[x[j]++] = i;
+  } while (++i < n);
+
+
+  /* Generate the Huffman codes and for each, make the table entries */
+  x[0] = i = 0;                 /* first Huffman code is zero */
+  p = v;                        /* grab values in bit order */
+  h = -1;                       /* no tables yet--level -1 */
+  w = -l;                       /* bits decoded == (l * h) */
+  u[0] = (inflate_huft *)Z_NULL;        /* just to keep compilers happy */
+  q = (inflate_huft *)Z_NULL;   /* ditto */
+  z = 0;                        /* ditto */
+
+  /* go through the bit lengths (k already is bits in shortest code) */
+  for (; k <= g; k++)
+  {
+    a = c[k];
+    while (a--)
+    {
+      /* here i is the Huffman code of length k bits for value *p */
+      /* make tables up to required level */
+      while (k > w + l)
+      {
+        h++;
+        w += l;                 /* previous table always l bits */
+
+        /* compute minimum size table less than or equal to l bits */
+        z = (z = g - w) > (uInt)l ? l : z;      /* table size upper limit */
+        if ((f = 1 << (j = k - w)) > a + 1)     /* try a k-w bit table */
+        {                       /* too few codes for k-w bit table */
+          f -= a + 1;           /* deduct codes from patterns left */
+          xp = c + k;
+          if (j < z)
+            while (++j < z)     /* try smaller tables up to z bits */
+            {
+              if ((f <<= 1) <= *++xp)
+                break;          /* enough codes to use up j bits */
+              f -= *xp;         /* else deduct codes from patterns */
+            }
+        }
+        z = 1 << j;             /* table entries for j-bit table */
+
+        /* allocate and link in new table */
+        if ((q = (inflate_huft *)ZALLOC
+             (zs,z + 1,sizeof(inflate_huft))) == Z_NULL)
+        {
+          if (h)
+            inflate_trees_free(u[0], zs);
+          return Z_MEM_ERROR;   /* not enough memory */
+        }
+	q->word.Nalloc = z + 1;
+#ifdef DEBUG_ZLIB
+        inflate_hufts += z + 1;
+#endif
+        *t = q + 1;             /* link to list for huft_free() */
+        *(t = &(q->next)) = Z_NULL;
+        u[h] = ++q;             /* table starts after link */
+
+        /* connect to last table, if there is one */
+        if (h)
+        {
+          x[h] = i;             /* save pattern for backing up */
+          r.bits = (Byte)l;     /* bits to dump before this table */
+          r.exop = (Byte)j;     /* bits in this table */
+          r.next = q;           /* pointer to this table */
+          j = i >> (w - l);     /* (get around Turbo C bug) */
+          u[h-1][j] = r;        /* connect to last table */
+        }
+      }
+
+      /* set up table entry in r */
+      r.bits = (Byte)(k - w);
+      if (p >= v + n)
+        r.exop = 128 + 64;      /* out of values--invalid code */
+      else if (*p < s)
+      {
+        r.exop = (Byte)(*p < 256 ? 0 : 32 + 64);     /* 256 is end-of-block */
+        r.base = *p++;          /* simple code is just the value */
+      }
+      else
+      {
+        r.exop = (Byte)e[*p - s] + 16 + 64; /* non-simple--look up in lists */
+        r.base = d[*p++ - s];
+      }
+
+      /* fill code-like entries with r */
+      f = 1 << (k - w);
+      for (j = i >> w; j < z; j += f)
+        q[j] = r;
+
+      /* backwards increment the k-bit code i */
+      for (j = 1 << (k - 1); i & j; j >>= 1)
+        i ^= j;
+      i ^= j;
+
+      /* backup over finished tables */
+      while ((i & ((1 << w) - 1)) != x[h])
+      {
+        h--;                    /* don't need to update q */
+        w -= l;
+      }
+    }
+  }
+
+
+  /* Return Z_BUF_ERROR if we were given an incomplete table */
+  return y != 0 && g != 1 ? Z_BUF_ERROR : Z_OK;
+}
+
+
+local int inflate_trees_bits(c, bb, tb, z)
+uIntf *c;               /* 19 code lengths */
+uIntf *bb;              /* bits tree desired/actual depth */
+inflate_huft * FAR *tb; /* bits tree result */
+z_stream *z;            /* for zfree function */
+{
+  int r;
+
+  r = huft_build(c, 19, 19, (uIntf*)Z_NULL, (uIntf*)Z_NULL, tb, bb, z);
+  if (r == Z_DATA_ERROR)
+    z->msg = "oversubscribed dynamic bit lengths tree";
+  else if (r == Z_BUF_ERROR)
+  {
+    inflate_trees_free(*tb, z);
+    z->msg = "incomplete dynamic bit lengths tree";
+    r = Z_DATA_ERROR;
+  }
+  return r;
+}
+
+
+local int inflate_trees_dynamic(nl, nd, c, bl, bd, tl, td, z)
+uInt nl;                /* number of literal/length codes */
+uInt nd;                /* number of distance codes */
+uIntf *c;               /* that many (total) code lengths */
+uIntf *bl;              /* literal desired/actual bit depth */
+uIntf *bd;              /* distance desired/actual bit depth */
+inflate_huft * FAR *tl; /* literal/length tree result */
+inflate_huft * FAR *td; /* distance tree result */
+z_stream *z;            /* for zfree function */
+{
+  int r;
+
+  /* build literal/length tree */
+  if ((r = huft_build(c, nl, 257, cplens, cplext, tl, bl, z)) != Z_OK)
+  {
+    if (r == Z_DATA_ERROR)
+      z->msg = "oversubscribed literal/length tree";
+    else if (r == Z_BUF_ERROR)
+    {
+      inflate_trees_free(*tl, z);
+      z->msg = "incomplete literal/length tree";
+      r = Z_DATA_ERROR;
+    }
+    return r;
+  }
+
+  /* build distance tree */
+  if ((r = huft_build(c + nl, nd, 0, cpdist, cpdext, td, bd, z)) != Z_OK)
+  {
+    if (r == Z_DATA_ERROR)
+      z->msg = "oversubscribed literal/length tree";
+    else if (r == Z_BUF_ERROR) {
+#ifdef PKZIP_BUG_WORKAROUND
+      r = Z_OK;
+    }
+#else
+      inflate_trees_free(*td, z);
+      z->msg = "incomplete literal/length tree";
+      r = Z_DATA_ERROR;
+    }
+    inflate_trees_free(*tl, z);
+    return r;
+#endif
+  }
+
+  /* done */
+  return Z_OK;
+}
+
+
+/* build fixed tables only once--keep them here */
+local int fixed_lock = 0;
+local int fixed_built = 0;
+#define FIXEDH 530      /* number of hufts used by fixed tables */
+local uInt fixed_left = FIXEDH;
+local inflate_huft fixed_mem[FIXEDH];
+local uInt fixed_bl;
+local uInt fixed_bd;
+local inflate_huft *fixed_tl;
+local inflate_huft *fixed_td;
+
+
+local voidpf falloc(q, n, s)
+voidpf q;        /* opaque pointer (not used) */
+uInt n;         /* number of items */
+uInt s;         /* size of item */
+{
+  Assert(s == sizeof(inflate_huft) && n <= fixed_left,
+         "inflate_trees falloc overflow");
+  if (q) s++; /* to make some compilers happy */
+  fixed_left -= n;
+  return (voidpf)(fixed_mem + fixed_left);
+}
+
+
+local void ffree(q, p, n)
+voidpf q;
+voidpf p;
+uInt n;
+{
+  Assert(0, "inflate_trees ffree called!");
+  if (q) q = p; /* to make some compilers happy */
+}
+
+
+local int inflate_trees_fixed(bl, bd, tl, td)
+uIntf *bl;               /* literal desired/actual bit depth */
+uIntf *bd;               /* distance desired/actual bit depth */
+inflate_huft * FAR *tl;  /* literal/length tree result */
+inflate_huft * FAR *td;  /* distance tree result */
+{
+  /* build fixed tables if not built already--lock out other instances */
+  while (++fixed_lock > 1)
+    fixed_lock--;
+  if (!fixed_built)
+  {
+    int k;              /* temporary variable */
+    unsigned c[288];    /* length list for huft_build */
+    z_stream z;         /* for falloc function */
+
+    /* set up fake z_stream for memory routines */
+    z.zalloc = falloc;
+    z.zfree = ffree;
+    z.opaque = Z_NULL;
+
+    /* literal table */
+    for (k = 0; k < 144; k++)
+      c[k] = 8;
+    for (; k < 256; k++)
+      c[k] = 9;
+    for (; k < 280; k++)
+      c[k] = 7;
+    for (; k < 288; k++)
+      c[k] = 8;
+    fixed_bl = 7;
+    huft_build(c, 288, 257, cplens, cplext, &fixed_tl, &fixed_bl, &z);
+
+    /* distance table */
+    for (k = 0; k < 30; k++)
+      c[k] = 5;
+    fixed_bd = 5;
+    huft_build(c, 30, 0, cpdist, cpdext, &fixed_td, &fixed_bd, &z);
+
+    /* done */
+    fixed_built = 1;
+  }
+  fixed_lock--;
+  *bl = fixed_bl;
+  *bd = fixed_bd;
+  *tl = fixed_tl;
+  *td = fixed_td;
+  return Z_OK;
+}
+
+
+local int inflate_trees_free(t, z)
+inflate_huft *t;        /* table to free */
+z_stream *z;            /* for zfree function */
+/* Free the malloc'ed tables built by huft_build(), which makes a linked
+   list of the tables it made, with the links in a dummy first entry of
+   each table. */
+{
+  register inflate_huft *p, *q;
+
+  /* Go through linked list, freeing from the malloced (t[-1]) address. */
+  p = t;
+  while (p != Z_NULL)
+  {
+    q = (--p)->next;
+    ZFREE(z, p, p->word.Nalloc * sizeof(inflate_huft));
+    p = q;
+  }
+  return Z_OK;
+}
+
+/*+++++*/
+/* infcodes.c -- process literals and length/distance pairs
+ * Copyright (C) 1995 Mark Adler
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* simplify the use of the inflate_huft type with some defines */
+#define base more.Base
+#define next more.Next
+#define exop word.what.Exop
+#define bits word.what.Bits
+
+/* inflate codes private state */
+struct inflate_codes_state {
+
+  /* mode */
+  enum {        /* waiting for "i:"=input, "o:"=output, "x:"=nothing */
+      START,    /* x: set up for LEN */
+      LEN,      /* i: get length/literal/eob next */
+      LENEXT,   /* i: getting length extra (have base) */
+      DIST,     /* i: get distance next */
+      DISTEXT,  /* i: getting distance extra */
+      COPY,     /* o: copying bytes in window, waiting for space */
+      LIT,      /* o: got literal, waiting for output space */
+      WASH,     /* o: got eob, possibly still output waiting */
+      END,      /* x: got eob and all data flushed */
+      BADCODE}  /* x: got error */
+    mode;               /* current inflate_codes mode */
+
+  /* mode dependent information */
+  uInt len;
+  union {
+    struct {
+      inflate_huft *tree;       /* pointer into tree */
+      uInt need;                /* bits needed */
+    } code;             /* if LEN or DIST, where in tree */
+    uInt lit;           /* if LIT, literal */
+    struct {
+      uInt get;                 /* bits to get for extra */
+      uInt dist;                /* distance back to copy from */
+    } copy;             /* if EXT or COPY, where and how much */
+  } sub;                /* submode */
+
+  /* mode independent information */
+  Byte lbits;           /* ltree bits decoded per branch */
+  Byte dbits;           /* dtree bits decoder per branch */
+  inflate_huft *ltree;          /* literal/length/eob tree */
+  inflate_huft *dtree;          /* distance tree */
+
+};
+
+
+local inflate_codes_statef *inflate_codes_new(bl, bd, tl, td, z)
+uInt bl, bd;
+inflate_huft *tl, *td;
+z_stream *z;
+{
+  inflate_codes_statef *c;
+
+  if ((c = (inflate_codes_statef *)
+       ZALLOC(z,1,sizeof(struct inflate_codes_state))) != Z_NULL)
+  {
+    c->mode = START;
+    c->lbits = (Byte)bl;
+    c->dbits = (Byte)bd;
+    c->ltree = tl;
+    c->dtree = td;
+    Tracev((stderr, "inflate:       codes new\n"));
+  }
+  return c;
+}
+
+
+local int inflate_codes(s, z, r)
+inflate_blocks_statef *s;
+z_stream *z;
+int r;
+{
+  uInt j;               /* temporary storage */
+  inflate_huft *t;      /* temporary pointer */
+  uInt e;               /* extra bits or operation */
+  uLong b;              /* bit buffer */
+  uInt k;               /* bits in bit buffer */
+  Bytef *p;             /* input data pointer */
+  uInt n;               /* bytes available there */
+  Bytef *q;             /* output window write pointer */
+  uInt m;               /* bytes to end of window or read pointer */
+  Bytef *f;             /* pointer to copy strings from */
+  inflate_codes_statef *c = s->sub.decode.codes;  /* codes state */
+
+  /* copy input/output information to locals (UPDATE macro restores) */
+  LOAD
+
+  /* process input and output based on current state */
+  while (1) switch (c->mode)
+  {             /* waiting for "i:"=input, "o:"=output, "x:"=nothing */
+    case START:         /* x: set up for LEN */
+#ifndef SLOW
+      if (m >= 258 && n >= 10)
+      {
+        UPDATE
+        r = inflate_fast(c->lbits, c->dbits, c->ltree, c->dtree, s, z);
+        LOAD
+        if (r != Z_OK)
+        {
+          c->mode = r == Z_STREAM_END ? WASH : BADCODE;
+          break;
+        }
+      }
+#endif /* !SLOW */
+      c->sub.code.need = c->lbits;
+      c->sub.code.tree = c->ltree;
+      c->mode = LEN;
+    case LEN:           /* i: get length/literal/eob next */
+      j = c->sub.code.need;
+      NEEDBITS(j)
+      t = c->sub.code.tree + ((uInt)b & inflate_mask[j]);
+      DUMPBITS(t->bits)
+      e = (uInt)(t->exop);
+      if (e == 0)               /* literal */
+      {
+        c->sub.lit = t->base;
+        Tracevv((stderr, t->base >= 0x20 && t->base < 0x7f ?
+                 "inflate:         literal '%c'\n" :
+                 "inflate:         literal 0x%02x\n", t->base));
+        c->mode = LIT;
+        break;
+      }
+      if (e & 16)               /* length */
+      {
+        c->sub.copy.get = e & 15;
+        c->len = t->base;
+        c->mode = LENEXT;
+        break;
+      }
+      if ((e & 64) == 0)        /* next table */
+      {
+        c->sub.code.need = e;
+        c->sub.code.tree = t->next;
+        break;
+      }
+      if (e & 32)               /* end of block */
+      {
+        Tracevv((stderr, "inflate:         end of block\n"));
+        c->mode = WASH;
+        break;
+      }
+      c->mode = BADCODE;        /* invalid code */
+      z->msg = "invalid literal/length code";
+      r = Z_DATA_ERROR;
+      LEAVE
+    case LENEXT:        /* i: getting length extra (have base) */
+      j = c->sub.copy.get;
+      NEEDBITS(j)
+      c->len += (uInt)b & inflate_mask[j];
+      DUMPBITS(j)
+      c->sub.code.need = c->dbits;
+      c->sub.code.tree = c->dtree;
+      Tracevv((stderr, "inflate:         length %u\n", c->len));
+      c->mode = DIST;
+    case DIST:          /* i: get distance next */
+      j = c->sub.code.need;
+      NEEDBITS(j)
+      t = c->sub.code.tree + ((uInt)b & inflate_mask[j]);
+      DUMPBITS(t->bits)
+      e = (uInt)(t->exop);
+      if (e & 16)               /* distance */
+      {
+        c->sub.copy.get = e & 15;
+        c->sub.copy.dist = t->base;
+        c->mode = DISTEXT;
+        break;
+      }
+      if ((e & 64) == 0)        /* next table */
+      {
+        c->sub.code.need = e;
+        c->sub.code.tree = t->next;
+        break;
+      }
+      c->mode = BADCODE;        /* invalid code */
+      z->msg = "invalid distance code";
+      r = Z_DATA_ERROR;
+      LEAVE
+    case DISTEXT:       /* i: getting distance extra */
+      j = c->sub.copy.get;
+      NEEDBITS(j)
+      c->sub.copy.dist += (uInt)b & inflate_mask[j];
+      DUMPBITS(j)
+      Tracevv((stderr, "inflate:         distance %u\n", c->sub.copy.dist));
+      c->mode = COPY;
+    case COPY:          /* o: copying bytes in window, waiting for space */
+#ifndef __TURBOC__ /* Turbo C bug for following expression */
+      f = (uInt)(q - s->window) < c->sub.copy.dist ?
+          s->end - (c->sub.copy.dist - (q - s->window)) :
+          q - c->sub.copy.dist;
+#else
+      f = q - c->sub.copy.dist;
+      if ((uInt)(q - s->window) < c->sub.copy.dist)
+        f = s->end - (c->sub.copy.dist - (q - s->window));
+#endif
+      while (c->len)
+      {
+        NEEDOUT
+        OUTBYTE(*f++)
+        if (f == s->end)
+          f = s->window;
+        c->len--;
+      }
+      c->mode = START;
+      break;
+    case LIT:           /* o: got literal, waiting for output space */
+      NEEDOUT
+      OUTBYTE(c->sub.lit)
+      c->mode = START;
+      break;
+    case WASH:          /* o: got eob, possibly more output */
+      FLUSH
+      if (s->read != s->write)
+        LEAVE
+      c->mode = END;
+    case END:
+      r = Z_STREAM_END;
+      LEAVE
+    case BADCODE:       /* x: got error */
+      r = Z_DATA_ERROR;
+      LEAVE
+    default:
+      r = Z_STREAM_ERROR;
+      LEAVE
+  }
+}
+
+
+local void inflate_codes_free(c, z)
+inflate_codes_statef *c;
+z_stream *z;
+{
+  ZFREE(z, c, sizeof(struct inflate_codes_state));
+  Tracev((stderr, "inflate:       codes free\n"));
+}
+
+/*+++++*/
+/* inflate_util.c -- data and routines common to blocks and codes
+ * Copyright (C) 1995 Mark Adler
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* copy as much as possible from the sliding window to the output area */
+local int inflate_flush(s, z, r)
+inflate_blocks_statef *s;
+z_stream *z;
+int r;
+{
+  uInt n;
+  Bytef *p, *q;
+
+  /* local copies of source and destination pointers */
+  p = z->next_out;
+  q = s->read;
+
+  /* compute number of bytes to copy as far as end of window */
+  n = (uInt)((q <= s->write ? s->write : s->end) - q);
+  if (n > z->avail_out) n = z->avail_out;
+  if (n && r == Z_BUF_ERROR) r = Z_OK;
+
+  /* update counters */
+  z->avail_out -= n;
+  z->total_out += n;
+
+  /* update check information */
+  if (s->checkfn != Z_NULL)
+    s->check = (*s->checkfn)(s->check, q, n);
+
+  /* copy as far as end of window */
+  zmemcpy(p, q, n);
+  p += n;
+  q += n;
+
+  /* see if more to copy at beginning of window */
+  if (q == s->end)
+  {
+    /* wrap pointers */
+    q = s->window;
+    if (s->write == s->end)
+      s->write = s->window;
+
+    /* compute bytes to copy */
+    n = (uInt)(s->write - q);
+    if (n > z->avail_out) n = z->avail_out;
+    if (n && r == Z_BUF_ERROR) r = Z_OK;
+
+    /* update counters */
+    z->avail_out -= n;
+    z->total_out += n;
+
+    /* update check information */
+    if (s->checkfn != Z_NULL)
+      s->check = (*s->checkfn)(s->check, q, n);
+
+    /* copy */
+    zmemcpy(p, q, n);
+    p += n;
+    q += n;
+  }
+
+  /* update pointers */
+  z->next_out = p;
+  s->read = q;
+
+  /* done */
+  return r;
+}
+
+
+/*+++++*/
+/* inffast.c -- process literals and length/distance pairs fast
+ * Copyright (C) 1995 Mark Adler
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* simplify the use of the inflate_huft type with some defines */
+#define base more.Base
+#define next more.Next
+#define exop word.what.Exop
+#define bits word.what.Bits
+
+/* macros for bit input with no checking and for returning unused bytes */
+#define GRABBITS(j) {while(k<(j)){b|=((uLong)NEXTBYTE)<<k;k+=8;}}
+#define UNGRAB {n+=(c=k>>3);p-=c;k&=7;}
+
+/* Called with number of bytes left to write in window at least 258
+   (the maximum string length) and number of input bytes available
+   at least ten.  The ten bytes are six bytes for the longest length/
+   distance pair plus four bytes for overloading the bit buffer. */
+
+local int inflate_fast(bl, bd, tl, td, s, z)
+uInt bl, bd;
+inflate_huft *tl, *td;
+inflate_blocks_statef *s;
+z_stream *z;
+{
+  inflate_huft *t;      /* temporary pointer */
+  uInt e;               /* extra bits or operation */
+  uLong b;              /* bit buffer */
+  uInt k;               /* bits in bit buffer */
+  Bytef *p;             /* input data pointer */
+  uInt n;               /* bytes available there */
+  Bytef *q;             /* output window write pointer */
+  uInt m;               /* bytes to end of window or read pointer */
+  uInt ml;              /* mask for literal/length tree */
+  uInt md;              /* mask for distance tree */
+  uInt c;               /* bytes to copy */
+  uInt d;               /* distance back to copy from */
+  Bytef *r;             /* copy source pointer */
+
+  /* load input, output, bit values */
+  LOAD
+
+  /* initialize masks */
+  ml = inflate_mask[bl];
+  md = inflate_mask[bd];
+
+  /* do until not enough input or output space for fast loop */
+  do {                          /* assume called with m >= 258 && n >= 10 */
+    /* get literal/length code */
+    GRABBITS(20)                /* max bits for literal/length code */
+    if ((e = (t = tl + ((uInt)b & ml))->exop) == 0)
+    {
+      DUMPBITS(t->bits)
+      Tracevv((stderr, t->base >= 0x20 && t->base < 0x7f ?
+                "inflate:         * literal '%c'\n" :
+                "inflate:         * literal 0x%02x\n", t->base));
+      *q++ = (Byte)t->base;
+      m--;
+      continue;
+    }
+    do {
+      DUMPBITS(t->bits)
+      if (e & 16)
+      {
+        /* get extra bits for length */
+        e &= 15;
+        c = t->base + ((uInt)b & inflate_mask[e]);
+        DUMPBITS(e)
+        Tracevv((stderr, "inflate:         * length %u\n", c));
+
+        /* decode distance base of block to copy */
+        GRABBITS(15);           /* max bits for distance code */
+        e = (t = td + ((uInt)b & md))->exop;
+        do {
+          DUMPBITS(t->bits)
+          if (e & 16)
+          {
+            /* get extra bits to add to distance base */
+            e &= 15;
+            GRABBITS(e)         /* get extra bits (up to 13) */
+            d = t->base + ((uInt)b & inflate_mask[e]);
+            DUMPBITS(e)
+            Tracevv((stderr, "inflate:         * distance %u\n", d));
+
+            /* do the copy */
+            m -= c;
+            if ((uInt)(q - s->window) >= d)     /* offset before dest */
+            {                                   /*  just copy */
+              r = q - d;
+              *q++ = *r++;  c--;        /* minimum count is three, */
+              *q++ = *r++;  c--;        /*  so unroll loop a little */
+            }
+            else                        /* else offset after destination */
+            {
+              e = d - (q - s->window);  /* bytes from offset to end */
+              r = s->end - e;           /* pointer to offset */
+              if (c > e)                /* if source crosses, */
+              {
+                c -= e;                 /* copy to end of window */
+                do {
+                  *q++ = *r++;
+                } while (--e);
+                r = s->window;          /* copy rest from start of window */
+              }
+            }
+            do {                        /* copy all or what's left */
+              *q++ = *r++;
+            } while (--c);
+            break;
+          }
+          else if ((e & 64) == 0)
+            e = (t = t->next + ((uInt)b & inflate_mask[e]))->exop;
+          else
+          {
+            z->msg = "invalid distance code";
+            UNGRAB
+            UPDATE
+            return Z_DATA_ERROR;
+          }
+        } while (1);
+        break;
+      }
+      if ((e & 64) == 0)
+      {
+        if ((e = (t = t->next + ((uInt)b & inflate_mask[e]))->exop) == 0)
+        {
+          DUMPBITS(t->bits)
+          Tracevv((stderr, t->base >= 0x20 && t->base < 0x7f ?
+                    "inflate:         * literal '%c'\n" :
+                    "inflate:         * literal 0x%02x\n", t->base));
+          *q++ = (Byte)t->base;
+          m--;
+          break;
+        }
+      }
+      else if (e & 32)
+      {
+        Tracevv((stderr, "inflate:         * end of block\n"));
+        UNGRAB
+        UPDATE
+        return Z_STREAM_END;
+      }
+      else
+      {
+        z->msg = "invalid literal/length code";
+        UNGRAB
+        UPDATE
+        return Z_DATA_ERROR;
+      }
+    } while (1);
+  } while (m >= 258 && n >= 10);
+
+  /* not enough input or output--restore pointers and return */
+  UNGRAB
+  UPDATE
+  return Z_OK;
+}
+
+
+/*+++++*/
+/* zutil.c -- target dependent utility functions for the compression library
+ * Copyright (C) 1995 Jean-loup Gailly.
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* From: zutil.c,v 1.8 1995/05/03 17:27:12 jloup Exp */
+
+char *zlib_version = ZLIB_VERSION;
+
+char *z_errmsg[] = {
+"stream end",          /* Z_STREAM_END    1 */
+"",                    /* Z_OK            0 */
+"file error",          /* Z_ERRNO        (-1) */
+"stream error",        /* Z_STREAM_ERROR (-2) */
+"data error",          /* Z_DATA_ERROR   (-3) */
+"insufficient memory", /* Z_MEM_ERROR    (-4) */
+"buffer error",        /* Z_BUF_ERROR    (-5) */
+""};
+
+
+/*+++++*/
+/* adler32.c -- compute the Adler-32 checksum of a data stream
+ * Copyright (C) 1995 Mark Adler
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* From: adler32.c,v 1.6 1995/05/03 17:27:08 jloup Exp */
+
+#define BASE 65521L /* largest prime smaller than 65536 */
+#define NMAX 5552
+/* NMAX is the largest n such that 255n(n+1)/2 + (n+1)(BASE-1) <= 2^32-1 */
+
+#define DO1(buf)  {s1 += *buf++; s2 += s1;}
+#define DO2(buf)  DO1(buf); DO1(buf);
+#define DO4(buf)  DO2(buf); DO2(buf);
+#define DO8(buf)  DO4(buf); DO4(buf);
+#define DO16(buf) DO8(buf); DO8(buf);
+
+/* ========================================================================= */
+uLong adler32(adler, buf, len)
+    uLong adler;
+    Bytef *buf;
+    uInt len;
+{
+    unsigned long s1 = adler & 0xffff;
+    unsigned long s2 = (adler >> 16) & 0xffff;
+    int k;
+
+    if (buf == Z_NULL) return 1L;
+
+    while (len > 0) {
+        k = len < NMAX ? len : NMAX;
+        len -= k;
+        while (k >= 16) {
+            DO16(buf);
+            k -= 16;
+        }
+        if (k != 0) do {
+            DO1(buf);
+        } while (--k);
+        s1 %= BASE;
+        s2 %= BASE;
+    }
+    return (s2 << 16) | s1;
+}
diff --git a/arch/xtensa/boot/lib/zmem.c b/arch/xtensa/boot/lib/zmem.c
new file mode 100644
index 00000000000..7848f126d67
--- /dev/null
+++ b/arch/xtensa/boot/lib/zmem.c
@@ -0,0 +1,87 @@
+#include "zlib.h"
+
+/* bits taken from ppc */
+
+extern void *avail_ram, *end_avail;
+
+void exit (void)
+{
+  for (;;);
+}
+
+void *zalloc(void *x, unsigned items, unsigned size)
+{
+        void *p = avail_ram;
+
+        size *= items;
+        size = (size + 7) & -8;
+        avail_ram += size;
+        if (avail_ram > end_avail) {
+                //puts("oops... out of memory\n");
+                //pause();
+                exit ();
+        }
+        return p;
+}
+
+void zfree(void *x, void *addr, unsigned nb)
+{
+}
+
+
+#define HEAD_CRC        2
+#define EXTRA_FIELD     4
+#define ORIG_NAME       8
+#define COMMENT         0x10
+#define RESERVED        0xe0
+
+#define DEFLATED        8
+
+void gunzip (void *dst, int dstlen, unsigned char *src, int *lenp)
+{
+	z_stream s;
+	int r, i, flags;
+
+        /* skip header */
+
+        i = 10;
+        flags = src[3];
+        if (src[2] != DEFLATED || (flags & RESERVED) != 0) {
+                //puts("bad gzipped data\n");
+                exit();
+        }
+        if ((flags & EXTRA_FIELD) != 0)
+                i = 12 + src[10] + (src[11] << 8);
+        if ((flags & ORIG_NAME) != 0)
+                while (src[i++] != 0)
+                        ;
+        if ((flags & COMMENT) != 0)
+                while (src[i++] != 0)
+                        ;
+        if ((flags & HEAD_CRC) != 0)
+                i += 2;
+        if (i >= *lenp) {
+                //puts("gunzip: ran out of data in header\n");
+                exit();
+        }
+
+        s.zalloc = zalloc;
+        s.zfree = zfree;
+        r = inflateInit2(&s, -MAX_WBITS);
+        if (r != Z_OK) {
+                //puts("inflateInit2 returned "); puthex(r); puts("\n");
+                exit();
+        }
+        s.next_in = src + i;
+        s.avail_in = *lenp - i;
+        s.next_out = dst;
+        s.avail_out = dstlen;
+        r = inflate(&s, Z_FINISH);
+        if (r != Z_OK && r != Z_STREAM_END) {
+                //puts("inflate returned "); puthex(r); puts("\n");
+                exit();
+        }
+        *lenp = s.next_out - (unsigned char *) dst;
+        inflateEnd(&s);
+}
+
diff --git a/arch/xtensa/boot/ramdisk/Makefile b/arch/xtensa/boot/ramdisk/Makefile
new file mode 100644
index 00000000000..b12f7635243
--- /dev/null
+++ b/arch/xtensa/boot/ramdisk/Makefile
@@ -0,0 +1,23 @@
+#
+# Makefile for a ramdisk image
+#
+
+BIG_ENDIAN	:= $(shell echo -e "\#ifdef __XTENSA_EL__\nint little;\n\#else\nint big;\n\#endif" | $(CC) -E -|grep -c big)
+
+ifeq ($(BIG_ENDIAN),1)
+OBJCOPY_ARGS    := -O elf32-xtensa-be
+else
+OBJCOPY_ARGS    := -O elf32-xtensa-le
+endif
+
+obj-y = ramdisk.o
+
+RAMDISK_IMAGE = arch/$(ARCH)/boot/ramdisk/$(CONFIG_EMBEDDED_RAMDISK_IMAGE)
+
+arch/$(ARCH)/boot/ramdisk/ramdisk.o:
+	$(Q)echo -e "dummy:" | $(AS) -o $@;
+	$(Q)$(OBJCOPY) $(OBJCOPY_ARGS)					    \
+		--add-section .initrd=$(RAMDISK_IMAGE)			    \
+		--set-section-flags .initrd=contents,alloc,load,load,data   \
+		arch/$(ARCH)/boot/ramdisk/ramdisk.o $@
+
-- 
cgit v1.2.3-70-g09d2


From 5a0015d62668e64c8b6e02e360fbbea121bfd5e6 Mon Sep 17 00:00:00 2001
From: Chris Zankel <czankel@tensilica.com>
Date: Thu, 23 Jun 2005 22:01:16 -0700
Subject: [PATCH] xtensa: Architecture support for Tensilica Xtensa Part 3

The attached patches provides part 3 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/kernel/Makefile       |   18 +
 arch/xtensa/kernel/align.S        |  459 +++++++++
 arch/xtensa/kernel/asm-offsets.c  |   94 ++
 arch/xtensa/kernel/coprocessor.S  |  201 ++++
 arch/xtensa/kernel/entry.S        | 1996 +++++++++++++++++++++++++++++++++++++
 arch/xtensa/kernel/head.S         |  237 +++++
 arch/xtensa/kernel/irq.c          |  192 ++++
 arch/xtensa/kernel/module.c       |   78 ++
 arch/xtensa/kernel/pci-dma.c      |   73 ++
 arch/xtensa/kernel/pci.c          |  563 +++++++++++
 arch/xtensa/kernel/platform.c     |   49 +
 arch/xtensa/kernel/process.c      |  482 +++++++++
 arch/xtensa/kernel/ptrace.c       |  407 ++++++++
 arch/xtensa/kernel/semaphore.c    |  226 +++++
 arch/xtensa/kernel/setup.c        |  520 ++++++++++
 arch/xtensa/kernel/signal.c       |  713 +++++++++++++
 arch/xtensa/kernel/syscalls.c     |  418 ++++++++
 arch/xtensa/kernel/syscalls.h     |  248 +++++
 arch/xtensa/kernel/time.c         |  227 +++++
 arch/xtensa/kernel/traps.c        |  498 +++++++++
 arch/xtensa/kernel/vectors.S      |  464 +++++++++
 arch/xtensa/kernel/vmlinux.lds.S  |  341 +++++++
 arch/xtensa/kernel/xtensa_ksyms.c |  123 +++
 23 files changed, 8627 insertions(+)
 create mode 100644 arch/xtensa/kernel/Makefile
 create mode 100644 arch/xtensa/kernel/align.S
 create mode 100644 arch/xtensa/kernel/asm-offsets.c
 create mode 100644 arch/xtensa/kernel/coprocessor.S
 create mode 100644 arch/xtensa/kernel/entry.S
 create mode 100644 arch/xtensa/kernel/head.S
 create mode 100644 arch/xtensa/kernel/irq.c
 create mode 100644 arch/xtensa/kernel/module.c
 create mode 100644 arch/xtensa/kernel/pci-dma.c
 create mode 100644 arch/xtensa/kernel/pci.c
 create mode 100644 arch/xtensa/kernel/platform.c
 create mode 100644 arch/xtensa/kernel/process.c
 create mode 100644 arch/xtensa/kernel/ptrace.c
 create mode 100644 arch/xtensa/kernel/semaphore.c
 create mode 100644 arch/xtensa/kernel/setup.c
 create mode 100644 arch/xtensa/kernel/signal.c
 create mode 100644 arch/xtensa/kernel/syscalls.c
 create mode 100644 arch/xtensa/kernel/syscalls.h
 create mode 100644 arch/xtensa/kernel/time.c
 create mode 100644 arch/xtensa/kernel/traps.c
 create mode 100644 arch/xtensa/kernel/vectors.S
 create mode 100644 arch/xtensa/kernel/vmlinux.lds.S
 create mode 100644 arch/xtensa/kernel/xtensa_ksyms.c

(limited to 'arch')

diff --git a/arch/xtensa/kernel/Makefile b/arch/xtensa/kernel/Makefile
new file mode 100644
index 00000000000..d573017a5dd
--- /dev/null
+++ b/arch/xtensa/kernel/Makefile
@@ -0,0 +1,18 @@
+#
+# Makefile for the Linux/Xtensa kernel.
+#
+
+extra-y := head.o vmlinux.lds
+
+
+obj-y := align.o entry.o irq.o coprocessor.o process.o ptrace.o semaphore.o  \
+	 setup.o signal.o syscalls.o time.o traps.o vectors.o platform.o  \
+	 pci-dma.o
+
+## windowspill.o
+
+obj-$(CONFIG_KGDB) += xtensa-stub.o
+obj-$(CONFIG_PCI) += pci.o
+obj-$(CONFIG_MODULES) += xtensa_ksyms.o module.o
+
+
diff --git a/arch/xtensa/kernel/align.S b/arch/xtensa/kernel/align.S
new file mode 100644
index 00000000000..74b1e90ef08
--- /dev/null
+++ b/arch/xtensa/kernel/align.S
@@ -0,0 +1,459 @@
+/*
+ * arch/xtensa/kernel/align.S
+ *
+ * Handle unalignment exceptions in kernel space.
+ *
+ * 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.
+ *
+ * Rewritten by Chris Zankel <chris@zankel.net>
+ *
+ * Based on work from Joe Taylor <joe@tensilica.com, joetylr@yahoo.com>
+ * and Marc Gauthier <marc@tensilica.com, marc@alimni.uwaterloo.ca>
+ */
+
+#include <linux/linkage.h>
+#include <asm/ptrace.h>
+#include <asm/ptrace.h>
+#include <asm/current.h>
+#include <asm/offsets.h>
+#include <asm/pgtable.h>
+#include <asm/processor.h>
+#include <asm/page.h>
+#include <asm/thread_info.h>
+
+#if XCHAL_UNALIGNED_LOAD_EXCEPTION || XCHAL_UNALIGNED_STORE_EXCEPTION
+
+/*  First-level exception handler for unaligned exceptions.
+ *
+ *  Note: This handler works only for kernel exceptions.  Unaligned user
+ *        access should get a seg fault.
+ */
+
+/* Big and little endian 16-bit values are located in
+ * different halves of a register.  HWORD_START helps to
+ * abstract the notion of extracting a 16-bit value from a
+ * register.
+ * We also have to define new shifting instructions because
+ * lsb and msb are on 'opposite' ends in a register for
+ * different endian machines.
+ *
+ * Assume a memory region in ascending address:
+ *   	0 1 2 3|4 5 6 7
+ *
+ * When loading one word into a register, the content of that register is:
+ *  LE	3 2 1 0, 7 6 5 4
+ *  BE  0 1 2 3, 4 5 6 7
+ *
+ * Masking the bits of the higher/lower address means:
+ *  LE  X X 0 0, 0 0 X X
+ *  BE	0 0 X X, X X 0 0
+ *
+ * Shifting to higher/lower addresses, means:
+ *  LE  shift left / shift right
+ *  BE  shift right / shift left
+ *
+ * Extracting 16 bits from a 32 bit reg. value to higher/lower address means:
+ *  LE  mask 0 0 X X / shift left
+ *  BE  shift left / mask 0 0 X X
+ */
+
+#define UNALIGNED_USER_EXCEPTION
+
+#if XCHAL_HAVE_BE
+
+#define HWORD_START	16
+#define	INSN_OP0	28
+#define	INSN_T		24
+#define	INSN_OP1	16
+
+.macro __src_b	r, w0, w1;	src	\r, \w0, \w1;	.endm
+.macro __ssa8	r;		ssa8b	\r;		.endm
+.macro __ssa8r	r;		ssa8l	\r;		.endm
+.macro __sh	r, s;		srl	\r, \s;		.endm
+.macro __sl	r, s;		sll	\r, \s;		.endm
+.macro __exth	r, s;		extui	\r, \s, 0, 16;	.endm
+.macro __extl	r, s;		slli	\r, \s, 16;	.endm
+
+#else
+
+#define HWORD_START	0
+#define	INSN_OP0	0
+#define	INSN_T		4
+#define	INSN_OP1	12
+
+.macro __src_b	r, w0, w1;	src	\r, \w1, \w0;	.endm
+.macro __ssa8	r;		ssa8l	\r;		.endm
+.macro __ssa8r	r;		ssa8b	\r;		.endm
+.macro __sh	r, s;		sll	\r, \s;		.endm
+.macro __sl	r, s;		srl	\r, \s;		.endm
+.macro __exth	r, s;		slli	\r, \s, 16;	.endm
+.macro __extl	r, s;		extui	\r, \s, 0, 16;	.endm
+
+#endif
+
+/*
+ *	xxxx xxxx = imm8 field
+ *	     yyyy = imm4 field
+ *	     ssss = s field
+ *	     tttt = t field
+ *
+ *	       		 16		    0
+ *		          -------------------
+ *	L32I.N		  yyyy ssss tttt 1000
+ *	S32I.N	          yyyy ssss tttt 1001
+ *
+ *	       23			    0
+ *		-----------------------------
+ *	res	          0000           0010
+ *	L16UI	xxxx xxxx 0001 ssss tttt 0010
+ *	L32I	xxxx xxxx 0010 ssss tttt 0010
+ *	XXX	          0011 ssss tttt 0010
+ *	XXX	          0100 ssss tttt 0010
+ *	S16I	xxxx xxxx 0101 ssss tttt 0010
+ *	S32I	xxxx xxxx 0110 ssss tttt 0010
+ *	XXX	          0111 ssss tttt 0010
+ *	XXX	          1000 ssss tttt 0010
+ *	L16SI	xxxx xxxx 1001 ssss tttt 0010
+ *	XXX	          1010           0010
+ *      **L32AI	xxxx xxxx 1011 ssss tttt 0010 unsupported
+ *	XXX	          1100           0010
+ *	XXX	          1101           0010
+ *	XXX	          1110           0010
+ *	**S32RI	xxxx xxxx 1111 ssss tttt 0010 unsupported
+ *		-----------------------------
+ *                           ^         ^    ^
+ *    sub-opcode (NIBBLE_R) -+         |    |
+ *       t field (NIBBLE_T) -----------+    |
+ *  major opcode (NIBBLE_OP0) --------------+
+ */
+
+#define OP0_L32I_N	0x8		/* load immediate narrow */
+#define OP0_S32I_N	0x9		/* store immediate narrow */
+#define OP1_SI_MASK	0x4		/* OP1 bit set for stores */
+#define OP1_SI_BIT	2		/* OP1 bit number for stores */
+
+#define OP1_L32I	0x2
+#define OP1_L16UI	0x1
+#define OP1_L16SI	0x9
+#define OP1_L32AI	0xb
+
+#define OP1_S32I	0x6
+#define OP1_S16I	0x5
+#define OP1_S32RI	0xf
+
+/*
+ * Entry condition:
+ *
+ *   a0:	trashed, original value saved on stack (PT_AREG0)
+ *   a1:	a1
+ *   a2:	new stack pointer, original in DEPC
+ *   a3:	dispatch table
+ *   depc:	a2, original value saved on stack (PT_DEPC)
+ *   excsave_1:	a3
+ *
+ *   PT_DEPC >= VALID_DOUBLE_EXCEPTION_ADDRESS: double exception, DEPC
+ *	     <  VALID_DOUBLE_EXCEPTION_ADDRESS: regular exception
+ */
+
+
+ENTRY(fast_unaligned)
+
+	/* Note: We don't expect the address to be aligned on a word
+	 *       boundary. After all, the processor generated that exception
+	 *       and it would be a hardware fault.
+	 */
+
+	/* Save some working register */
+
+	s32i	a4, a2, PT_AREG4
+	s32i	a5, a2, PT_AREG5
+	s32i	a6, a2, PT_AREG6
+	s32i	a7, a2, PT_AREG7
+	s32i	a8, a2, PT_AREG8
+
+	rsr	a0, DEPC
+	xsr	a3, EXCSAVE_1
+	s32i	a0, a2, PT_AREG2
+	s32i	a3, a2, PT_AREG3
+
+	/* Keep value of SAR in a0 */
+
+	rsr	a0, SAR
+	rsr	a8, EXCVADDR		# load unaligned memory address
+
+	/* Now, identify one of the following load/store instructions.
+	 *
+	 * The only possible danger of a double exception on the
+	 * following l32i instructions is kernel code in vmalloc
+	 * memory. The processor was just executing at the EPC_1
+	 * address, and indeed, already fetched the instruction.  That
+	 * guarantees a TLB mapping, which hasn't been replaced by
+	 * this unaligned exception handler that uses only static TLB
+	 * mappings. However, high-level interrupt handlers might
+	 * modify TLB entries, so for the generic case, we register a
+	 * TABLE_FIXUP handler here, too.
+	 */
+
+	/* a3...a6 saved on stack, a2 = SP */
+
+	/* Extract the instruction that caused the unaligned access. */
+
+	rsr	a7, EPC_1	# load exception address
+	movi	a3, ~3
+	and	a3, a3, a7	# mask lower bits
+
+	l32i	a4, a3, 0	# load 2 words
+	l32i	a5, a3, 4
+
+	__ssa8	a7
+	__src_b	a4, a4, a5	# a4 has the instruction
+
+	/* Analyze the instruction (load or store?). */
+
+	extui	a5, a4, INSN_OP0, 4	# get insn.op0 nibble
+
+#if XCHAL_HAVE_NARROW
+	_beqi	a5, OP0_L32I_N, .Lload	# L32I.N, jump
+	addi	a6, a5, -OP0_S32I_N
+	_beqz	a6, .Lstore		# S32I.N, do a store
+#endif
+	/* 'store indicator bit' not set, jump */
+	_bbci.l	a4, OP1_SI_BIT + INSN_OP1, .Lload
+
+	/* Store: Jump to table entry to get the value in the source register.*/
+
+.Lstore:movi	a5, .Lstore_table	# table
+	extui	a6, a4, INSN_T, 4	# get source register
+	addx8	a5, a6, a5
+	jx	a5			# jump into table
+
+	/* Invalid instruction, CRITICAL! */
+.Linvalid_instruction_load:
+	j	.Linvalid_instruction
+
+	/* Load: Load memory address. */
+
+.Lload: movi	a3, ~3
+	and	a3, a3, a8		# align memory address
+
+	__ssa8	a8
+#ifdef UNALIGNED_USER_EXCEPTION
+	addi	a3, a3, 8
+	l32e	a5, a3, -8
+	l32e	a6, a3, -4
+#else
+	l32i	a5, a3, 0
+	l32i	a6, a3, 4
+#endif
+	__src_b	a3, a5, a6		# a3 has the data word
+
+#if XCHAL_HAVE_NARROW
+	addi	a7, a7, 2		# increment PC (assume 16-bit insn)
+
+	extui	a5, a4, INSN_OP0, 4
+	_beqi	a5, OP0_L32I_N, 1f	# l32i.n: jump
+
+	addi	a7, a7, 1
+#else
+	addi	a7, a7, 3
+#endif
+
+	extui	a5, a4, INSN_OP1, 4
+	_beqi	a5, OP1_L32I, 1f	# l32i: jump
+
+	extui	a3, a3, 0, 16		# extract lower 16 bits
+	_beqi	a5, OP1_L16UI, 1f
+	addi	a5, a5, -OP1_L16SI
+	_bnez	a5, .Linvalid_instruction_load
+
+	/* sign extend value */
+
+	slli	a3, a3, 16
+	srai	a3, a3, 16
+
+	/* Set target register. */
+
+1:
+
+#if XCHAL_HAVE_LOOP
+	rsr	a3, LEND		# check if we reached LEND
+	bne	a7, a3, 1f
+	rsr	a3, LCOUNT		# and LCOUNT != 0
+	beqz	a3, 1f
+	addi	a3, a3, -1		# decrement LCOUNT and set
+	rsr	a7, LBEG		# set PC to LBEGIN
+	wsr	a3, LCOUNT
+#endif
+
+1:	wsr	a7, EPC_1		# skip load instruction
+	extui	a4, a4, INSN_T, 4	# extract target register
+	movi	a5, .Lload_table
+	addx8	a4, a4, a5
+	jx	a4			# jump to entry for target register
+
+	.align	8
+.Lload_table:
+	s32i	a3, a2, PT_AREG0;	_j .Lexit;	.align 8
+	mov	a1, a3;			_j .Lexit;	.align 8 # fishy??
+	s32i	a3, a2, PT_AREG2;	_j .Lexit;	.align 8
+	s32i	a3, a2, PT_AREG3;	_j .Lexit;	.align 8
+	s32i	a3, a2, PT_AREG4;	_j .Lexit;	.align 8
+	s32i	a3, a2, PT_AREG5;	_j .Lexit;	.align 8
+	s32i	a3, a2, PT_AREG6;	_j .Lexit;	.align 8
+	s32i	a3, a2, PT_AREG7;	_j .Lexit;	.align 8
+	s32i	a3, a2, PT_AREG8;	_j .Lexit;	.align 8
+	mov	a9, a3		;	_j .Lexit;	.align 8
+	mov	a10, a3		;	_j .Lexit;	.align 8
+	mov	a11, a3		;	_j .Lexit;	.align 8
+	mov	a12, a3		;	_j .Lexit;	.align 8
+	mov	a13, a3		;	_j .Lexit;	.align 8
+	mov	a14, a3		;	_j .Lexit;	.align 8
+	mov	a15, a3		;	_j .Lexit;	.align 8
+
+.Lstore_table:
+	l32i	a3, a2, PT_AREG0;	_j 1f;	.align 8
+	mov	a3, a1;			_j 1f;	.align 8	# fishy??
+	l32i	a3, a2, PT_AREG2;	_j 1f;	.align 8
+	l32i	a3, a2, PT_AREG3;	_j 1f;	.align 8
+	l32i	a3, a2, PT_AREG4;	_j 1f;	.align 8
+	l32i	a3, a2, PT_AREG5;	_j 1f;	.align 8
+	l32i	a3, a2, PT_AREG6;	_j 1f;	.align 8
+	l32i	a3, a2, PT_AREG7;	_j 1f;	.align 8
+	l32i	a3, a2, PT_AREG8;	_j 1f;	.align 8
+	mov	a3, a9		;	_j 1f;	.align 8
+	mov	a3, a10		;	_j 1f;	.align 8
+	mov	a3, a11		;	_j 1f;	.align 8
+	mov	a3, a12		;	_j 1f;	.align 8
+	mov	a3, a13		;	_j 1f;	.align 8
+	mov	a3, a14		;	_j 1f;	.align 8
+	mov	a3, a15		;	_j 1f;	.align 8
+
+1: 	# a7: instruction pointer, a4: instruction, a3: value
+
+	movi	a6, 0			# mask: ffffffff:00000000
+
+#if XCHAL_HAVE_NARROW
+	addi	a7, a7, 2		# incr. PC,assume 16-bit instruction
+
+	extui	a5, a4, INSN_OP0, 4	# extract OP0
+	addi	a5, a5, -OP0_S32I_N
+	_beqz	a5, 1f			# s32i.n: jump
+
+	addi	a7, a7, 1		# increment PC, 32-bit instruction
+#else
+	addi	a7, a7, 3		# increment PC, 32-bit instruction
+#endif
+
+	extui	a5, a4, INSN_OP1, 4	# extract OP1
+	_beqi	a5, OP1_S32I, 1f	# jump if 32 bit store
+	_bnei	a5, OP1_S16I, .Linvalid_instruction_store
+
+	movi	a5, -1
+	__extl	a3, a3			# get 16-bit value
+	__exth	a6, a5			# get 16-bit mask ffffffff:ffff0000
+
+	/* Get memory address */
+
+1:
+#if XCHAL_HAVE_LOOP
+	rsr	a3, LEND		# check if we reached LEND
+	bne	a7, a3, 1f
+	rsr	a3, LCOUNT		# and LCOUNT != 0
+	beqz	a3, 1f
+	addi	a3, a3, -1		# decrement LCOUNT and set
+	rsr	a7, LBEG		# set PC to LBEGIN
+	wsr	a3, LCOUNT
+#endif
+
+1:	wsr	a7, EPC_1		# skip store instruction
+	movi	a4, ~3
+	and	a4, a4, a8		# align memory address
+
+	/* Insert value into memory */
+
+	movi	a5, -1			# mask: ffffffff:XXXX0000
+#ifdef UNALIGNED_USER_EXCEPTION
+	addi	a4, a4, 8
+#endif
+
+	__ssa8r a8
+	__src_b	a7, a5, a6		# lo-mask  F..F0..0 (BE) 0..0F..F (LE)
+	__src_b	a6, a6, a5		# hi-mask  0..0F..F (BE) F..F0..0 (LE)
+#ifdef UNALIGNED_USER_EXCEPTION
+	l32e	a5, a4, -8
+#else
+	l32i	a5, a4, 0		# load lower address word
+#endif
+	and	a5, a5, a7		# mask
+	__sh	a7, a3 			# shift value
+	or	a5, a5, a7		# or with original value
+#ifdef UNALIGNED_USER_EXCEPTION
+	s32e	a5, a4, -8
+	l32e	a7, a4, -4
+#else
+	s32i	a5, a4, 0		# store
+	l32i	a7, a4, 4		# same for upper address word
+#endif
+	__sl	a5, a3
+	and	a6, a7, a6
+	or	a6, a6, a5
+#ifdef UNALIGNED_USER_EXCEPTION
+	s32e	a6, a4, -4
+#else
+	s32i	a6, a4, 4
+#endif
+
+	/* Done. restore stack and return */
+
+.Lexit:
+	movi	a4, 0
+	rsr	a3, EXCSAVE_1
+	s32i	a4, a3, EXC_TABLE_FIXUP
+
+	/* Restore working register */
+
+	l32i	a7, a2, PT_AREG7
+	l32i	a6, a2, PT_AREG6
+	l32i	a5, a2, PT_AREG5
+	l32i	a4, a2, PT_AREG4
+	l32i	a3, a2, PT_AREG3
+
+	/* restore SAR and return */
+
+	wsr	a0, SAR
+	l32i	a0, a2, PT_AREG0
+	l32i	a2, a2, PT_AREG2
+	rfe
+
+	/* We cannot handle this exception. */
+
+	.extern _kernel_exception
+.Linvalid_instruction_store:
+.Linvalid_instruction:
+
+	/* Restore a4...a8 and SAR, set SP, and jump to default exception. */
+
+	l32i	a8, a2, PT_AREG8
+	l32i	a7, a2, PT_AREG7
+	l32i	a6, a2, PT_AREG6
+	l32i	a5, a2, PT_AREG5
+	l32i	a4, a2, PT_AREG4
+	wsr	a0, SAR
+	mov	a1, a2
+
+	rsr	a0, PS
+        bbsi.l  a2, PS_UM_SHIFT, 1f     # jump if user mode
+
+	movi	a0, _kernel_exception
+	jx	a0
+
+1:	movi	a0, _user_exception
+	jx	a0
+
+
+#endif /* XCHAL_UNALIGNED_LOAD_EXCEPTION || XCHAL_UNALIGNED_STORE_EXCEPTION */
+
diff --git a/arch/xtensa/kernel/asm-offsets.c b/arch/xtensa/kernel/asm-offsets.c
new file mode 100644
index 00000000000..840cd9a1d3d
--- /dev/null
+++ b/arch/xtensa/kernel/asm-offsets.c
@@ -0,0 +1,94 @@
+/*
+ * arch/xtensa/kernel/asm-offsets.c
+ *
+ * Generates definitions from c-type structures used by assembly sources.
+ *
+ * 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.
+ *
+ * Chris Zankel <chris@zankel.net>
+ */
+
+#include <asm/processor.h>
+
+#include <linux/types.h>
+#include <linux/sched.h>
+#include <linux/stddef.h>
+#include <linux/thread_info.h>
+#include <linux/ptrace.h>
+#include <asm/ptrace.h>
+#include <asm/processor.h>
+#include <asm/uaccess.h>
+
+#define DEFINE(sym, val) asm volatile("\n->" #sym " %0 " #val : : "i" (val))
+#define BLANK() asm volatile("\n->" : : )
+
+int main(void)
+{
+	/* struct pt_regs */
+	DEFINE(PT_PC, offsetof (struct pt_regs, pc));
+	DEFINE(PT_PS, offsetof (struct pt_regs, ps));
+	DEFINE(PT_DEPC, offsetof (struct pt_regs, depc));
+	DEFINE(PT_EXCCAUSE, offsetof (struct pt_regs, exccause));
+	DEFINE(PT_EXCVADDR, offsetof (struct pt_regs, excvaddr));
+	DEFINE(PT_DEBUGCAUSE, offsetof (struct pt_regs, debugcause));
+	DEFINE(PT_WMASK, offsetof (struct pt_regs, wmask));
+	DEFINE(PT_LBEG, offsetof (struct pt_regs, lbeg));
+	DEFINE(PT_LEND, offsetof (struct pt_regs, lend));
+	DEFINE(PT_LCOUNT, offsetof (struct pt_regs, lcount));
+	DEFINE(PT_SAR, offsetof (struct pt_regs, sar));
+	DEFINE(PT_SYSCALL, offsetof (struct pt_regs, syscall));
+	DEFINE(PT_AREG, offsetof (struct pt_regs, areg[0]));
+	DEFINE(PT_AREG0, offsetof (struct pt_regs, areg[0]));
+	DEFINE(PT_AREG1, offsetof (struct pt_regs, areg[1]));
+	DEFINE(PT_AREG2, offsetof (struct pt_regs, areg[2]));
+	DEFINE(PT_AREG3, offsetof (struct pt_regs, areg[3]));
+	DEFINE(PT_AREG4, offsetof (struct pt_regs, areg[4]));
+	DEFINE(PT_AREG5, offsetof (struct pt_regs, areg[5]));
+	DEFINE(PT_AREG6, offsetof (struct pt_regs, areg[6]));
+	DEFINE(PT_AREG7, offsetof (struct pt_regs, areg[7]));
+	DEFINE(PT_AREG8, offsetof (struct pt_regs, areg[8]));
+	DEFINE(PT_AREG9, offsetof (struct pt_regs, areg[9]));
+	DEFINE(PT_AREG10, offsetof (struct pt_regs, areg[10]));
+	DEFINE(PT_AREG11, offsetof (struct pt_regs, areg[11]));
+	DEFINE(PT_AREG12, offsetof (struct pt_regs, areg[12]));
+	DEFINE(PT_AREG13, offsetof (struct pt_regs, areg[13]));
+	DEFINE(PT_AREG14, offsetof (struct pt_regs, areg[14]));
+	DEFINE(PT_AREG15, offsetof (struct pt_regs, areg[15]));
+	DEFINE(PT_WINDOWBASE, offsetof (struct pt_regs, windowbase));
+	DEFINE(PT_WINDOWSTART, offsetof(struct pt_regs, windowstart));
+	DEFINE(PT_SIZE, sizeof(struct pt_regs));
+	DEFINE(PT_AREG_END, offsetof (struct pt_regs, areg[XCHAL_NUM_AREGS]));
+	DEFINE(PT_USER_SIZE, offsetof(struct pt_regs, areg[XCHAL_NUM_AREGS]));
+	BLANK();
+
+	/* struct task_struct */
+	DEFINE(TASK_PTRACE, offsetof (struct task_struct, ptrace));
+	DEFINE(TASK_MM, offsetof (struct task_struct, mm));
+	DEFINE(TASK_ACTIVE_MM, offsetof (struct task_struct, active_mm));
+	DEFINE(TASK_PID, offsetof (struct task_struct, pid));
+	DEFINE(TASK_THREAD, offsetof (struct task_struct, thread));
+	DEFINE(TASK_THREAD_INFO, offsetof (struct task_struct, thread_info));
+	DEFINE(TASK_STRUCT_SIZE, sizeof (struct task_struct));
+	BLANK();
+
+	/* struct thread_info (offset from start_struct) */
+	DEFINE(THREAD_RA, offsetof (struct task_struct, thread.ra));
+	DEFINE(THREAD_SP, offsetof (struct task_struct, thread.sp));
+	DEFINE(THREAD_CP_SAVE, offsetof (struct task_struct, thread.cp_save));
+	DEFINE(THREAD_CURRENT_DS, offsetof (struct task_struct, thread.current_ds));
+	BLANK();
+
+	/* struct mm_struct */
+	DEFINE(MM_USERS, offsetof(struct mm_struct, mm_users));
+	DEFINE(MM_PGD, offsetof (struct mm_struct, pgd));
+	DEFINE(MM_CONTEXT, offsetof (struct mm_struct, context));
+	BLANK();
+	DEFINE(PT_SINGLESTEP_BIT, PT_SINGLESTEP_BIT);
+	return 0;
+}
+
+
diff --git a/arch/xtensa/kernel/coprocessor.S b/arch/xtensa/kernel/coprocessor.S
new file mode 100644
index 00000000000..356192a4d39
--- /dev/null
+++ b/arch/xtensa/kernel/coprocessor.S
@@ -0,0 +1,201 @@
+/*
+ * arch/xtensa/kernel/coprocessor.S
+ *
+ * Xtensa processor configuration-specific table of coprocessor and
+ * other custom register layout information.
+ *
+ * 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.
+ *
+ * Marc Gauthier <marc@tensilica.com> <marc@alumni.uwaterloo.ca>
+ */
+
+/*
+ * This module contains a table that describes the layout of the various
+ * custom registers and states associated with each coprocessor, as well
+ * as those not associated with any coprocessor ("extra state").
+ * This table is included with core dumps and is available via the ptrace
+ * interface, allowing the layout of such register/state information to
+ * be modified in the kernel without affecting the debugger.  Each
+ * register or state is identified using a 32-bit "libdb target number"
+ * assigned when the Xtensa processor is generated.
+ */
+
+#include <linux/config.h>
+#include <linux/linkage.h>
+#include <asm/processor.h>
+
+#if XCHAL_HAVE_CP
+
+#define CP_LAST ((XCHAL_CP_MAX - 1) * COPROCESSOR_INFO_SIZE)
+
+ENTRY(release_coprocessors)
+
+	entry	a1, 16
+						# a2: task
+	movi	a3, 1 << XCHAL_CP_MAX 		# a3: coprocessor-bit
+	movi	a4, coprocessor_info+CP_LAST	# a4: owner-table
+						# a5: tmp
+	movi	a6, 0				# a6: 0
+	rsil	a7, LOCKLEVEL			# a7: PS
+
+1:	/* Check if task is coprocessor owner of coprocessor[i]. */
+
+	l32i	a5, a4, COPROCESSOR_INFO_OWNER
+	srli	a3, a3, 1
+	beqz	a3, 1f
+	addi	a4, a4, -8
+	beq	a2, a5, 1b
+
+	/* Found an entry: Clear entry CPENABLE bit to disable CP. */
+
+	rsr	a5, CPENABLE
+	s32i	a6, a4, COPROCESSOR_INFO_OWNER
+	xor	a5, a3, a5
+	wsr	a5, CPENABLE
+
+	bnez	a3, 1b
+
+1:	wsr	a7, PS
+	rsync
+	retw
+
+
+ENTRY(disable_coprocessor)
+	entry	sp, 16
+	rsil	a7, LOCKLEVEL
+	rsr	a3, CPENABLE
+	movi	a4, 1
+	ssl	a2
+	sll	a4, a4
+	and	a4, a3, a4
+	xor	a3, a3, a4
+	wsr	a3, CPENABLE
+	wsr	a7, PS
+	rsync
+	retw
+
+ENTRY(enable_coprocessor)
+	entry	sp, 16
+	rsil	a7, LOCKLEVEL
+	rsr	a3, CPENABLE
+	movi	a4, 1
+	ssl	a2
+	sll	a4, a4
+	or	a3, a3, a4
+	wsr	a3, CPENABLE
+	wsr	a7, PS
+	rsync
+	retw
+
+#endif
+
+ENTRY(save_coprocessor_extra)
+	entry	sp, 16
+	xchal_extra_store_funcbody
+	retw
+
+ENTRY(restore_coprocessor_extra)
+	entry	sp, 16
+	xchal_extra_load_funcbody
+	retw
+
+ENTRY(save_coprocessor_registers)
+	entry	sp, 16
+	xchal_cpi_store_funcbody
+	retw
+
+ENTRY(restore_coprocessor_registers)
+	entry	sp, 16
+	xchal_cpi_load_funcbody
+	retw
+
+
+/*
+ *  The Xtensa compile-time HAL (core.h) XCHAL_*_SA_CONTENTS_LIBDB macros
+ *  describe the contents of coprocessor & extra save areas in terms of
+ *  undefined CONTENTS_LIBDB_{SREG,UREG,REGF} macros.  We define these
+ *  latter macros here; they expand into a table of the format we want.
+ *  The general format is:
+ *
+ *	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)
+ *
+ *  For this table, we only care about the <libdbnum>, <offset> and <size>
+ *  fields.
+ */
+
+/*  Map all XCHAL CONTENTS macros to the reg_entry asm macro defined below:  */
+
+#define CONTENTS_LIBDB_SREG(libdbnum,offset,size,align,rsv1,name,sregnum,     \
+			    bitmask, rsv2, rsv3)			      \
+		reg_entry libdbnum, offset, size ;
+#define CONTENTS_LIBDB_UREG(libdbnum,offset,size,align,rsv1,name,uregnum,     \
+			    bitmask, rsv2, rsv3)			      \
+		reg_entry libdbnum, offset, size ;
+#define CONTENTS_LIBDB_REGF(libdbnum, offset, size, align, rsv1, name, index, \
+			    numentries, contentsize, regname_base,	      \
+			    regfile_name, rsv2, rsv3)			      \
+		reg_entry libdbnum, offset, size ;
+
+/* A single table entry: */
+	.macro	reg_entry	libdbnum, offset, size
+	 .ifne	(__last_offset-(__last_group_offset+\offset))
+	  /* padding entry */
+	  .word	(0xFC000000+__last_offset-(__last_group_offset+\offset))
+	 .endif
+	 .word	\libdbnum				/* actual entry */
+	 .set	__last_offset, __last_group_offset+\offset+\size
+	.endm	/* reg_entry */
+
+
+/* Table entry that marks the beginning of a group (coprocessor or "extra"): */
+	.macro	reg_group	cpnum, num_entries, align
+	 .set	__last_group_offset, (__last_offset + \align- 1) & -\align
+	 .ifne	\num_entries
+	  .word	0xFD000000+(\cpnum<<16)+\num_entries
+	 .endif
+	.endm	/* reg_group */
+
+/*
+ * Register info tables.
+ */
+
+	.section .rodata, "a"
+	.globl	_xtensa_reginfo_tables
+	.globl	_xtensa_reginfo_table_size
+	.align	4
+_xtensa_reginfo_table_size:
+	.word	_xtensa_reginfo_table_end - _xtensa_reginfo_tables
+
+_xtensa_reginfo_tables:
+	.set	__last_offset, 0
+	reg_group 0xFF, XCHAL_EXTRA_SA_CONTENTS_LIBDB_NUM, XCHAL_EXTRA_SA_ALIGN
+	XCHAL_EXTRA_SA_CONTENTS_LIBDB
+	reg_group 0, XCHAL_CP0_SA_CONTENTS_LIBDB_NUM, XCHAL_CP0_SA_ALIGN
+	XCHAL_CP0_SA_CONTENTS_LIBDB
+	reg_group 1, XCHAL_CP1_SA_CONTENTS_LIBDB_NUM, XCHAL_CP1_SA_ALIGN
+	XCHAL_CP1_SA_CONTENTS_LIBDB
+	reg_group 2, XCHAL_CP2_SA_CONTENTS_LIBDB_NUM, XCHAL_CP2_SA_ALIGN
+	XCHAL_CP2_SA_CONTENTS_LIBDB
+	reg_group 3, XCHAL_CP3_SA_CONTENTS_LIBDB_NUM, XCHAL_CP3_SA_ALIGN
+	XCHAL_CP3_SA_CONTENTS_LIBDB
+	reg_group 4, XCHAL_CP4_SA_CONTENTS_LIBDB_NUM, XCHAL_CP4_SA_ALIGN
+	XCHAL_CP4_SA_CONTENTS_LIBDB
+	reg_group 5, XCHAL_CP5_SA_CONTENTS_LIBDB_NUM, XCHAL_CP5_SA_ALIGN
+	XCHAL_CP5_SA_CONTENTS_LIBDB
+	reg_group 6, XCHAL_CP6_SA_CONTENTS_LIBDB_NUM, XCHAL_CP6_SA_ALIGN
+	XCHAL_CP6_SA_CONTENTS_LIBDB
+	reg_group 7, XCHAL_CP7_SA_CONTENTS_LIBDB_NUM, XCHAL_CP7_SA_ALIGN
+	XCHAL_CP7_SA_CONTENTS_LIBDB
+	.word	0xFC000000	/* invalid register number,marks end of table*/
+_xtensa_reginfo_table_end:
+
diff --git a/arch/xtensa/kernel/entry.S b/arch/xtensa/kernel/entry.S
new file mode 100644
index 00000000000..c64a01f71de
--- /dev/null
+++ b/arch/xtensa/kernel/entry.S
@@ -0,0 +1,1996 @@
+/*
+ * arch/xtensa/kernel/entry.S
+ *
+ * Low-level exception handling
+ *
+ * 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) 2004-2005 by Tensilica Inc.
+ *
+ * Chris Zankel <chris@zankel.net>
+ *
+ */
+
+#include <linux/linkage.h>
+#include <asm/offsets.h>
+#include <asm/processor.h>
+#include <asm/thread_info.h>
+#include <asm/uaccess.h>
+#include <asm/unistd.h>
+#include <asm/ptrace.h>
+#include <asm/current.h>
+#include <asm/pgtable.h>
+#include <asm/page.h>
+#include <asm/signal.h>
+#include <xtensa/coreasm.h>
+
+/* Unimplemented features. */
+
+#undef SIGNAL_HANDLING_IN_DOUBLE_EXCEPTION
+#undef KERNEL_STACK_OVERFLOW_CHECK
+#undef PREEMPTIBLE_KERNEL
+#undef ALLOCA_EXCEPTION_IN_IRAM
+
+/* Not well tested.
+ *
+ * - fast_coprocessor
+ */
+
+/*
+ * Macro to find first bit set in WINDOWBASE from the left + 1
+ *
+ * 100....0 -> 1
+ * 010....0 -> 2
+ * 000....1 -> WSBITS
+ */
+
+	.macro ffs_ws bit mask
+
+#if XCHAL_HAVE_NSA
+	nsau    \bit, \mask			# 32-WSBITS ... 31 (32 iff 0)
+	addi    \bit, \bit, WSBITS - 32 + 1   	# uppest bit set -> return 1
+#else
+	movi    \bit, WSBITS
+#if WSBITS > 16
+	_bltui  \mask, 0x10000, 99f
+	addi    \bit, \bit, -16
+	extui   \mask, \mask, 16, 16
+#endif
+#if WSBITS > 8
+99:	_bltui  \mask, 0x100, 99f
+	addi    \bit, \bit, -8
+	srli    \mask, \mask, 8
+#endif
+99:	_bltui  \mask, 0x10, 99f
+	addi    \bit, \bit, -4
+	srli    \mask, \mask, 4
+99:	_bltui  \mask, 0x4, 99f
+	addi    \bit, \bit, -2
+	srli    \mask, \mask, 2
+99:	_bltui  \mask, 0x2, 99f
+	addi    \bit, \bit, -1
+99:
+
+#endif
+	.endm
+
+/* ----------------- DEFAULT FIRST LEVEL EXCEPTION HANDLERS ----------------- */
+
+/*
+ * First-level exception handler for user exceptions.
+ * Save some special registers, extra states and all registers in the AR
+ * register file that were in use in the user task, and jump to the common
+ * exception code.
+ * We save SAR (used to calculate WMASK), and WB and WS (we don't have to
+ * save them for kernel exceptions).
+ *
+ * Entry condition for user_exception:
+ *
+ *   a0:	trashed, original value saved on stack (PT_AREG0)
+ *   a1:	a1
+ *   a2:	new stack pointer, original value in depc
+ *   a3:	dispatch table
+ *   depc:	a2, original value saved on stack (PT_DEPC)
+ *   excsave1:	a3
+ *
+ *   PT_DEPC >= VALID_DOUBLE_EXCEPTION_ADDRESS: double exception, DEPC
+ *	     <  VALID_DOUBLE_EXCEPTION_ADDRESS: regular exception
+ *
+ * Entry condition for _user_exception:
+ *
+ *   a0-a3 and depc have been saved to PT_AREG0...PT_AREG3 and PT_DEPC
+ *   excsave has been restored, and
+ *   stack pointer (a1) has been set.
+ *
+ * Note: _user_exception might be at an odd adress. Don't use call0..call12
+ */
+
+ENTRY(user_exception)
+
+	/* Save a2, a3, and depc, restore excsave_1 and set SP. */
+
+	xsr	a3, EXCSAVE_1
+	rsr	a0, DEPC
+	s32i	a1, a2, PT_AREG1
+	s32i	a0, a2, PT_AREG2
+	s32i	a3, a2, PT_AREG3
+	mov	a1, a2
+
+	.globl _user_exception
+_user_exception:
+
+	/* Save SAR and turn off single stepping */
+
+	movi	a2, 0
+	rsr	a3, SAR
+	wsr	a2, ICOUNTLEVEL
+	s32i	a3, a1, PT_SAR
+
+	/* Rotate ws so that the current windowbase is at bit0. */
+	/* Assume ws = xxwww1yyyy. Rotate ws right, so that a2 = yyyyxxwww1 */
+
+	rsr	a2, WINDOWBASE
+	rsr	a3, WINDOWSTART
+	ssr	a2
+	s32i	a2, a1, PT_WINDOWBASE
+	s32i	a3, a1, PT_WINDOWSTART
+	slli	a2, a3, 32-WSBITS
+	src	a2, a3, a2
+	srli	a2, a2, 32-WSBITS
+	s32i	a2, a1, PT_WMASK	# needed for restoring registers
+
+	/* Save only live registers. */
+
+	_bbsi.l	a2, 1, 1f
+	s32i	a4, a1, PT_AREG4
+	s32i	a5, a1, PT_AREG5
+	s32i	a6, a1, PT_AREG6
+	s32i	a7, a1, PT_AREG7
+	_bbsi.l	a2, 2, 1f
+	s32i	a8, a1, PT_AREG8
+	s32i	a9, a1, PT_AREG9
+	s32i	a10, a1, PT_AREG10
+	s32i	a11, a1, PT_AREG11
+	_bbsi.l	a2, 3, 1f
+	s32i	a12, a1, PT_AREG12
+	s32i	a13, a1, PT_AREG13
+	s32i	a14, a1, PT_AREG14
+	s32i	a15, a1, PT_AREG15
+	_bnei	a2, 1, 1f		# only one valid frame?
+
+	/* Only one valid frame, skip saving regs. */
+
+	j	2f
+
+	/* Save the remaining registers.
+	 * We have to save all registers up to the first '1' from
+	 * the right, except the current frame (bit 0).
+	 * Assume a2 is:  001001000110001
+	 * All regiser frames starting from the top fiel to the marked '1'
+	 * must be saved.
+	 */
+
+1:	addi	a3, a2, -1		# eliminate '1' in bit 0: yyyyxxww0
+	neg	a3, a3			# yyyyxxww0 -> YYYYXXWW1+1
+	and	a3, a3, a2		# max. only one bit is set
+
+	/* Find number of frames to save */
+
+	ffs_ws	a0, a3			# number of frames to the '1' from left
+
+	/* Store information into WMASK:
+	 * bits 0..3: xxx1 masked lower 4 bits of the rotated windowstart,
+	 * bits 4...: number of valid 4-register frames
+	 */
+
+	slli	a3, a0, 4		# number of frames to save in bits 8..4
+	extui	a2, a2, 0, 4		# mask for the first 16 registers
+	or	a2, a3, a2
+	s32i	a2, a1, PT_WMASK	# needed when we restore the reg-file
+
+	/* Save 4 registers at a time */
+
+1:	rotw	-1
+	s32i	a0, a5, PT_AREG_END - 16
+	s32i	a1, a5, PT_AREG_END - 12
+	s32i	a2, a5, PT_AREG_END - 8
+	s32i	a3, a5, PT_AREG_END - 4
+	addi	a0, a4, -1
+	addi	a1, a5, -16
+	_bnez	a0, 1b
+
+	/* WINDOWBASE still in SAR! */
+
+	rsr	a2, SAR			# original WINDOWBASE
+	movi	a3, 1
+	ssl	a2
+	sll	a3, a3
+	wsr	a3, WINDOWSTART		# set corresponding WINDOWSTART bit
+	wsr	a2, WINDOWBASE		# and WINDOWSTART
+	rsync
+
+	/* We are back to the original stack pointer (a1) */
+
+2:
+#if XCHAL_EXTRA_SA_SIZE
+
+	/* For user exceptions, save the extra state into the user's TCB.
+	 * Note: We must assume that xchal_extra_store_funcbody destroys a2..a15
+	 */
+
+	GET_CURRENT(a2,a1)
+	addi	a2, a2, THREAD_CP_SAVE
+	xchal_extra_store_funcbody
+#endif
+
+	/* Now, jump to the common exception handler. */
+
+	j	common_exception
+
+
+/*
+ * First-level exit handler for kernel exceptions
+ * Save special registers and the live window frame.
+ * Note: Even though we changes the stack pointer, we don't have to do a
+ *	 MOVSP here, as we do that when we return from the exception.
+ *	 (See comment in the kernel exception exit code)
+ *
+ * Entry condition for kernel_exception:
+ *
+ *   a0:	trashed, original value saved on stack (PT_AREG0)
+ *   a1:	a1
+ *   a2:	new stack pointer, original in DEPC
+ *   a3:	dispatch table
+ *   depc:	a2, original value saved on stack (PT_DEPC)
+ *   excsave_1:	a3
+ *
+ *   PT_DEPC >= VALID_DOUBLE_EXCEPTION_ADDRESS: double exception, DEPC
+ *	     <  VALID_DOUBLE_EXCEPTION_ADDRESS: regular exception
+ *
+ * Entry condition for _kernel_exception:
+ *
+ *   a0-a3 and depc have been saved to PT_AREG0...PT_AREG3 and PT_DEPC
+ *   excsave has been restored, and
+ *   stack pointer (a1) has been set.
+ *
+ * Note: _kernel_exception might be at an odd adress. Don't use call0..call12
+ */
+
+ENTRY(kernel_exception)
+
+	/* Save a0, a2, a3, DEPC and set SP. */
+
+	xsr	a3, EXCSAVE_1		# restore a3, excsave_1
+	rsr	a0, DEPC		# get a2
+	s32i	a1, a2, PT_AREG1
+	s32i	a0, a2, PT_AREG2
+	s32i	a3, a2, PT_AREG3
+	mov	a1, a2
+
+	.globl _kernel_exception
+_kernel_exception:
+
+	/* Save SAR and turn off single stepping */
+
+	movi	a2, 0
+	rsr	a3, SAR
+	wsr	a2, ICOUNTLEVEL
+	s32i	a3, a1, PT_SAR
+
+	/* Rotate ws so that the current windowbase is at bit0. */
+	/* Assume ws = xxwww1yyyy. Rotate ws right, so that a2 = yyyyxxwww1 */
+
+	rsr	a2, WINDOWBASE		# don't need to save these, we only
+	rsr	a3, WINDOWSTART		# need shifted windowstart: windowmask
+	ssr	a2
+	slli	a2, a3, 32-WSBITS
+	src	a2, a3, a2
+	srli	a2, a2, 32-WSBITS
+	s32i	a2, a1, PT_WMASK	# needed for kernel_exception_exit
+
+	/* Save only the live window-frame */
+
+	_bbsi.l	a2, 1, 1f
+	s32i	a4, a1, PT_AREG4
+	s32i	a5, a1, PT_AREG5
+	s32i	a6, a1, PT_AREG6
+	s32i	a7, a1, PT_AREG7
+	_bbsi.l	a2, 2, 1f
+	s32i	a8, a1, PT_AREG8
+	s32i	a9, a1, PT_AREG9
+	s32i	a10, a1, PT_AREG10
+	s32i	a11, a1, PT_AREG11
+	_bbsi.l	a2, 3, 1f
+	s32i	a12, a1, PT_AREG12
+	s32i	a13, a1, PT_AREG13
+	s32i	a14, a1, PT_AREG14
+	s32i	a15, a1, PT_AREG15
+
+1:
+
+#ifdef KERNEL_STACK_OVERFLOW_CHECK
+
+	/*  Stack overflow check, for debugging  */
+	extui	a2, a1, TASK_SIZE_BITS,XX
+	movi	a3, SIZE??
+	_bge	a2, a3, out_of_stack_panic
+
+#endif
+
+/*
+ * This is the common exception handler.
+ * We get here from the user exception handler or simply by falling through
+ * from the kernel exception handler.
+ * Save the remaining special registers, switch to kernel mode, and jump
+ * to the second-level exception handler.
+ *
+ */
+
+common_exception:
+
+	/* Save EXCVADDR, DEBUGCAUSE, and PC, and clear LCOUNT */
+
+	rsr	a2, DEBUGCAUSE
+	rsr	a3, EPC_1
+	s32i	a2, a1, PT_DEBUGCAUSE
+	s32i	a3, a1, PT_PC
+
+	rsr	a3, EXCVADDR
+	movi	a2, 0
+	s32i	a3, a1, PT_EXCVADDR
+	xsr	a2, LCOUNT
+	s32i	a2, a1, PT_LCOUNT
+
+	/* It is now save to restore the EXC_TABLE_FIXUP variable. */
+
+	rsr	a0, EXCCAUSE
+	movi	a3, 0
+	rsr	a2, EXCSAVE_1
+	s32i	a0, a1, PT_EXCCAUSE
+	s32i	a3, a2, EXC_TABLE_FIXUP
+
+	/* All unrecoverable states are saved on stack, now, and a1 is valid,
+	 * so we can allow exceptions and interrupts (*) again.
+	 * Set PS(EXCM = 0, UM = 0, RING = 0, OWB = 0, WOE = 1, INTLEVEL = X)
+	 *
+	 * (*) We only allow interrupts if PS.INTLEVEL was not set to 1 before
+	 *     (interrupts disabled) and if this exception is not an interrupt.
+	 */
+
+	rsr	a3, PS
+	addi	a0, a0, -4
+	movi	a2, 1
+	extui	a3, a3, 0, 1		# a3 = PS.INTLEVEL[0]
+	moveqz	a3, a2, a0		# a3 = 1 iff interrupt exception
+	movi	a2, PS_WOE_MASK
+	or	a3, a3, a2
+	rsr	a0, EXCCAUSE
+	xsr	a3, PS
+
+	s32i	a3, a1, PT_PS		# save ps
+
+	/* Save LBEG, LEND */
+
+	rsr	a2, LBEG
+	rsr	a3, LEND
+	s32i	a2, a1, PT_LBEG
+	s32i	a3, a1, PT_LEND
+
+	/* Go to second-level dispatcher. Set up parameters to pass to the
+	 * exception handler and call the exception handler.
+	 */
+
+	movi	a4, exc_table
+	mov	a6, a1			# pass stack frame
+	mov	a7, a0			# pass EXCCAUSE
+	addx4	a4, a0, a4
+	l32i	a4, a4, EXC_TABLE_DEFAULT		# load handler
+
+	/* Call the second-level handler */
+
+	callx4	a4
+
+	/* Jump here for exception exit */
+
+common_exception_return:
+
+	/* Jump if we are returning from kernel exceptions. */
+
+1:	l32i	a3, a1, PT_PS
+	_bbsi.l	a3, PS_UM_SHIFT, 2f
+	j	kernel_exception_exit
+
+	/* Specific to a user exception exit:
+	 * We need to check some flags for signal handling and rescheduling,
+	 * and have to restore WB and WS, extra states, and all registers
+	 * in the register file that were in use in the user task.
+	 */
+
+2:	wsr	a3, PS		/* disable interrupts */
+
+	/* Check for signals (keep interrupts disabled while we read TI_FLAGS)
+	 * Note: PS.INTLEVEL = 0, PS.EXCM = 1
+	 */
+
+	GET_THREAD_INFO(a2,a1)
+	l32i	a4, a2, TI_FLAGS
+
+	/* Enable interrupts again.
+	 * Note: When we get here, we certainly have handled any interrupts.
+	 *       (Hint: There is only one user exception frame on stack)
+	 */
+
+	movi	a3, PS_WOE_MASK
+
+	_bbsi.l	a4, TIF_NEED_RESCHED, 3f
+	_bbci.l	a4, TIF_SIGPENDING, 4f
+
+#ifndef SIGNAL_HANDLING_IN_DOUBLE_EXCEPTION
+	l32i	a4, a1, PT_DEPC
+	bgeui	a4, VALID_DOUBLE_EXCEPTION_ADDRESS, 4f
+#endif
+
+	/* Reenable interrupts and call do_signal() */
+
+	wsr	a3, PS
+	movi	a4, do_signal	# int do_signal(struct pt_regs*, sigset_t*)
+	mov	a6, a1
+	movi	a7, 0
+	callx4	a4
+	j	1b
+
+3:	/* Reenable interrupts and reschedule */
+
+	wsr	a3, PS
+	movi	a4, schedule	# void schedule (void)
+	callx4	a4
+	j	1b
+
+	/* Restore the state of the task and return from the exception. */
+
+
+	/* If we are returning from a user exception, and the process
+	 * to run next has PT_SINGLESTEP set, we want to setup
+	 * ICOUNT and ICOUNTLEVEL to step one instruction.
+	 * PT_SINGLESTEP is set by sys_ptrace (ptrace.c)
+	 */
+
+4:	/* a2 holds GET_CURRENT(a2,a1)  */
+
+	l32i	a3, a2, TI_TASK
+	l32i	a3, a3, TASK_PTRACE
+	bbci.l	a3, PT_SINGLESTEP_BIT, 1f # jump if single-step flag is not set
+
+	movi	a3, -2			# PT_SINGLESTEP flag is set,
+	movi	a4, 1			# icountlevel of 1 means it won't
+	wsr	a3, ICOUNT		# start counting until after rfe
+	wsr	a4, ICOUNTLEVEL		# so setup icount & icountlevel.
+	isync
+
+1:
+
+#if XCHAL_EXTRA_SA_SIZE
+
+	/* For user exceptions, restore the extra state from the user's TCB. */
+
+	/* Note: a2 still contains GET_CURRENT(a2,a1) */
+	addi	a2, a2, THREAD_CP_SAVE
+	xchal_extra_load_funcbody
+
+	/* We must assume that xchal_extra_store_funcbody destroys
+	 * registers a2..a15.  FIXME, this list can eventually be
+	 * reduced once real register requirements of the macro are
+	 * finalized. */
+
+#endif /* XCHAL_EXTRA_SA_SIZE */
+
+
+	/* Switch to the user thread WINDOWBASE. Save SP temporarily in DEPC */
+
+	l32i	a2, a1, PT_WINDOWBASE
+	l32i	a3, a1, PT_WINDOWSTART
+	wsr	a1, DEPC		# use DEPC as temp storage
+	wsr	a3, WINDOWSTART		# restore WINDOWSTART
+	ssr	a2			# preserve user's WB in the SAR
+	wsr	a2, WINDOWBASE		# switch to user's saved WB
+	rsync
+	rsr	a1, DEPC		# restore stack pointer
+	l32i	a2, a1, PT_WMASK	# register frames saved (in bits 4...9)
+	rotw	-1			# we restore a4..a7
+	_bltui	a6, 16, 1f		# only have to restore current window?
+
+	/* The working registers are a0 and a3.  We are restoring to
+	 * a4..a7.  Be careful not to destroy what we have just restored.
+	 * Note: wmask has the format YYYYM:
+	 *       Y: number of registers saved in groups of 4
+	 *       M: 4 bit mask of first 16 registers
+	 */
+
+	mov	a2, a6
+	mov	a3, a5
+
+2:	rotw	-1			# a0..a3 become a4..a7
+	addi	a3, a7, -4*4		# next iteration
+	addi	a2, a6, -16		# decrementing Y in WMASK
+	l32i	a4, a3, PT_AREG_END + 0
+	l32i	a5, a3, PT_AREG_END + 4
+	l32i	a6, a3, PT_AREG_END + 8
+	l32i	a7, a3, PT_AREG_END + 12
+	_bgeui	a2, 16, 2b
+
+	/* Clear unrestored registers (don't leak anything to user-land */
+
+1:	rsr	a0, WINDOWBASE
+	rsr	a3, SAR
+	sub	a3, a0, a3
+	beqz	a3, 2f
+	extui	a3, a3, 0, WBBITS
+
+1:	rotw	-1
+	addi	a3, a7, -1
+	movi	a4, 0
+	movi	a5, 0
+	movi	a6, 0
+	movi	a7, 0
+	bgei	a3, 1, 1b
+
+	/* We are back were we were when we started.
+	 * Note: a2 still contains WMASK (if we've returned to the original
+	 *	 frame where we had loaded a2), or at least the lower 4 bits
+	 *	 (if we have restored WSBITS-1 frames).
+	 */
+
+2:	j	common_exception_exit
+
+	/* This is the kernel exception exit.
+	 * We avoided to do a MOVSP when we entered the exception, but we
+	 * have to do it here.
+	 */
+
+kernel_exception_exit:
+
+	/* Disable interrupts (a3 holds PT_PS) */
+
+	wsr	a3, PS
+
+#ifdef PREEMPTIBLE_KERNEL
+
+#ifdef CONFIG_PREEMPT
+
+	/*
+	 * Note: We've just returned from a call4, so we have
+	 * at least 4 addt'l regs.
+	 */
+
+	/* Check current_thread_info->preempt_count */
+
+	GET_THREAD_INFO(a2)
+	l32i	a3, a2, TI_PREEMPT
+	bnez	a3, 1f
+
+	l32i	a2, a2, TI_FLAGS
+
+1:
+
+#endif
+
+#endif
+
+	/* Check if we have to do a movsp.
+	 *
+	 * We only have to do a movsp if the previous window-frame has
+	 * been spilled to the *temporary* exception stack instead of the
+	 * task's stack. This is the case if the corresponding bit in
+	 * WINDOWSTART for the previous window-frame was set before
+	 * (not spilled) but is zero now (spilled).
+	 * If this bit is zero, all other bits except the one for the
+	 * current window frame are also zero. So, we can use a simple test:
+	 * 'and' WINDOWSTART and WINDOWSTART-1:
+	 *
+	 *  (XXXXXX1[0]* - 1) AND XXXXXX1[0]* = XXXXXX0[0]*
+	 *
+	 * The result is zero only if one bit was set.
+	 *
+	 * (Note: We might have gone through several task switches before
+	 *        we come back to the current task, so WINDOWBASE might be
+	 *        different from the time the exception occurred.)
+	 */
+
+	/* Test WINDOWSTART before and after the exception.
+	 * We actually have WMASK, so we only have to test if it is 1 or not.
+	 */
+
+	l32i	a2, a1, PT_WMASK
+	_beqi	a2, 1, common_exception_exit	# Spilled before exception,jump
+
+	/* Test WINDOWSTART now. If spilled, do the movsp */
+
+	rsr     a3, WINDOWSTART
+	addi	a0, a3, -1
+	and     a3, a3, a0
+	_bnez	a3, common_exception_exit
+
+	/* Do a movsp (we returned from a call4, so we have at least a0..a7) */
+
+	addi    a0, a1, -16
+	l32i    a3, a0, 0
+	l32i    a4, a0, 4
+	s32i    a3, a1, PT_SIZE+0
+	s32i    a4, a1, PT_SIZE+4
+	l32i    a3, a0, 8
+	l32i    a4, a0, 12
+	s32i    a3, a1, PT_SIZE+8
+	s32i    a4, a1, PT_SIZE+12
+
+	/* Common exception exit.
+	 * We restore the special register and the current window frame, and
+	 * return from the exception.
+	 *
+	 * Note: We expect a2 to hold PT_WMASK
+	 */
+
+common_exception_exit:
+
+	_bbsi.l	a2, 1, 1f
+	l32i	a4,  a1, PT_AREG4
+	l32i	a5,  a1, PT_AREG5
+	l32i	a6,  a1, PT_AREG6
+	l32i	a7,  a1, PT_AREG7
+	_bbsi.l	a2, 2, 1f
+	l32i	a8,  a1, PT_AREG8
+	l32i	a9,  a1, PT_AREG9
+	l32i	a10, a1, PT_AREG10
+	l32i	a11, a1, PT_AREG11
+	_bbsi.l	a2, 3, 1f
+	l32i	a12, a1, PT_AREG12
+	l32i	a13, a1, PT_AREG13
+	l32i	a14, a1, PT_AREG14
+	l32i	a15, a1, PT_AREG15
+
+	/* Restore PC, SAR */
+
+1:	l32i	a2, a1, PT_PC
+	l32i	a3, a1, PT_SAR
+	wsr	a2, EPC_1
+	wsr	a3, SAR
+
+	/* Restore LBEG, LEND, LCOUNT */
+
+	l32i	a2, a1, PT_LBEG
+	l32i	a3, a1, PT_LEND
+	wsr	a2, LBEG
+	l32i	a2, a1, PT_LCOUNT
+	wsr	a3, LEND
+	wsr	a2, LCOUNT
+
+	/* Check if it was double exception. */
+
+	l32i	a0, a1, PT_DEPC
+	l32i	a3, a1, PT_AREG3
+	l32i	a2, a1, PT_AREG2
+	_bgeui	a0, VALID_DOUBLE_EXCEPTION_ADDRESS, 1f
+
+	/* Restore a0...a3 and return */
+
+	l32i	a0, a1, PT_AREG0
+	l32i	a1, a1, PT_AREG1
+	rfe
+
+1:	wsr	a0, DEPC
+	l32i	a0, a1, PT_AREG0
+	l32i	a1, a1, PT_AREG1
+	rfde
+
+/*
+ * Debug exception handler.
+ *
+ * Currently, we don't support KGDB, so only user application can be debugged.
+ *
+ * When we get here,  a0 is trashed and saved to excsave[debuglevel]
+ */
+
+ENTRY(debug_exception)
+
+	rsr	a0, EPS + XCHAL_DEBUGLEVEL
+	bbsi.l	a0, PS_EXCM_SHIFT, 1f	# exception mode
+
+	/* Set EPC_1 and EXCCAUSE */
+
+	wsr	a2, DEPC		# save a2 temporarily
+	rsr	a2, EPC + XCHAL_DEBUGLEVEL
+	wsr	a2, EPC_1
+
+	movi	a2, EXCCAUSE_MAPPED_DEBUG
+	wsr	a2, EXCCAUSE
+
+	/* Restore PS to the value before the debug exc but with PS.EXCM set.*/
+
+	movi	a2, 1 << PS_EXCM_SHIFT
+	or	a2, a0, a2
+	movi	a0, debug_exception	# restore a3, debug jump vector
+	wsr	a2, PS
+	xsr	a0, EXCSAVE + XCHAL_DEBUGLEVEL
+
+	/* Switch to kernel/user stack, restore jump vector, and save a0 */
+
+	bbsi.l	a2, PS_UM_SHIFT, 2f	# jump if user mode
+
+	addi	a2, a1, -16-PT_SIZE	# assume kernel stack
+	s32i	a0, a2, PT_AREG0
+	movi	a0, 0
+	s32i	a1, a2, PT_AREG1
+	s32i	a0, a2, PT_DEPC		# mark it as a regular exception
+	xsr	a0, DEPC
+	s32i	a3, a2, PT_AREG3
+	s32i	a0, a2, PT_AREG2
+	mov	a1, a2
+	j	_kernel_exception
+
+2:	rsr	a2, EXCSAVE_1
+	l32i	a2, a2, EXC_TABLE_KSTK	# load kernel stack pointer
+	s32i	a0, a2, PT_AREG0
+	movi	a0, 0
+	s32i	a1, a2, PT_AREG1
+	s32i	a0, a2, PT_DEPC
+	xsr	a0, DEPC
+	s32i	a3, a2, PT_AREG3
+	s32i	a0, a2, PT_AREG2
+	mov	a1, a2
+	j	_user_exception
+
+	/* Debug exception while in exception mode. */
+1:	j	1b	// FIXME!!
+
+
+/*
+ * We get here in case of an unrecoverable exception.
+ * The only thing we can do is to be nice and print a panic message.
+ * We only produce a single stack frame for panic, so ???
+ *
+ *
+ * Entry conditions:
+ *
+ *   - a0 contains the caller address; original value saved in excsave1.
+ *   - the original a0 contains a valid return address (backtrace) or 0.
+ *   - a2 contains a valid stackpointer
+ *
+ * Notes:
+ *
+ *   - If the stack pointer could be invalid, the caller has to setup a
+ *     dummy stack pointer (e.g. the stack of the init_task)
+ *
+ *   - If the return address could be invalid, the caller has to set it
+ *     to 0, so the backtrace would stop.
+ *
+ */
+	.align 4
+unrecoverable_text:
+	.ascii "Unrecoverable error in exception handler\0"
+
+ENTRY(unrecoverable_exception)
+
+	movi	a0, 1
+	movi	a1, 0
+
+	wsr	a0, WINDOWSTART
+	wsr	a1, WINDOWBASE
+	rsync
+
+	movi	a1, PS_WOE_MASK | 1
+	wsr	a1, PS
+	rsync
+
+	movi	a1, init_task
+	movi	a0, 0
+	addi	a1, a1, PT_REGS_OFFSET
+
+	movi	a4, panic
+	movi	a6, unrecoverable_text
+
+	callx4	a4
+
+1:	j	1b
+
+
+/* -------------------------- FAST EXCEPTION HANDLERS ----------------------- */
+
+/*
+ * Fast-handler for alloca exceptions
+ *
+ *  The ALLOCA handler is entered when user code executes the MOVSP
+ *  instruction and the caller's frame is not in the register file.
+ *  In this case, the caller frame's a0..a3 are on the stack just
+ *  below sp (a1), and this handler moves them.
+ *
+ *  For "MOVSP <ar>,<as>" without destination register a1, this routine
+ *  simply moves the value from <as> to <ar> without moving the save area.
+ *
+ * Entry condition:
+ *
+ *   a0:	trashed, original value saved on stack (PT_AREG0)
+ *   a1:	a1
+ *   a2:	new stack pointer, original in DEPC
+ *   a3:	dispatch table
+ *   depc:	a2, original value saved on stack (PT_DEPC)
+ *   excsave_1:	a3
+ *
+ *   PT_DEPC >= VALID_DOUBLE_EXCEPTION_ADDRESS: double exception, DEPC
+ *	     <  VALID_DOUBLE_EXCEPTION_ADDRESS: regular exception
+ */
+
+#if XCHAL_HAVE_BE
+#define _EXTUI_MOVSP_SRC(ar)	extui ar, ar, 4, 4
+#define _EXTUI_MOVSP_DST(ar)	extui ar, ar, 0, 4
+#else
+#define _EXTUI_MOVSP_SRC(ar)	extui ar, ar, 0, 4
+#define _EXTUI_MOVSP_DST(ar)	extui ar, ar, 4, 4
+#endif
+
+ENTRY(fast_alloca)
+
+	/* We shouldn't be in a double exception. */
+
+	l32i	a0, a2, PT_DEPC
+	_bgeui	a0, VALID_DOUBLE_EXCEPTION_ADDRESS, .Lunhandled_double
+
+	rsr	a0, DEPC		# get a2
+	s32i	a4, a2, PT_AREG4	# save a4 and
+	s32i	a0, a2, PT_AREG2	# a2 to stack
+
+	/* Exit critical section. */
+
+	movi	a0, 0
+	s32i	a0, a3, EXC_TABLE_FIXUP
+
+	/* Restore a3, excsave_1 */
+
+	xsr	a3, EXCSAVE_1		# make sure excsave_1 is valid for dbl.
+	rsr	a4, EPC_1		# get exception address
+	s32i	a3, a2, PT_AREG3	# save a3 to stack
+
+#ifdef ALLOCA_EXCEPTION_IN_IRAM
+#error	iram not supported
+#else
+	/* Note: l8ui not allowed in IRAM/IROM!! */
+	l8ui	a0, a4, 1		# read as(src) from MOVSP instruction
+#endif
+	movi	a3, .Lmovsp_src
+	_EXTUI_MOVSP_SRC(a0)		# extract source register number
+	addx8	a3, a0, a3
+	jx	a3
+
+.Lunhandled_double:
+	wsr	a0, EXCSAVE_1
+	movi	a0, unrecoverable_exception
+	callx0	a0
+
+	.align 8
+.Lmovsp_src:
+	l32i	a3, a2, PT_AREG0;	_j 1f;	.align 8
+	mov	a3, a1;			_j 1f;	.align 8
+	l32i	a3, a2, PT_AREG2;	_j 1f;	.align 8
+	l32i	a3, a2, PT_AREG3;	_j 1f;	.align 8
+	l32i	a3, a2, PT_AREG4;	_j 1f;	.align 8
+	mov	a3, a5;			_j 1f;	.align 8
+	mov	a3, a6;			_j 1f;	.align 8
+	mov	a3, a7;			_j 1f;	.align 8
+	mov	a3, a8;			_j 1f;	.align 8
+	mov	a3, a9;			_j 1f;	.align 8
+	mov	a3, a10;		_j 1f;	.align 8
+	mov	a3, a11;		_j 1f;	.align 8
+	mov	a3, a12;		_j 1f;	.align 8
+	mov	a3, a13;		_j 1f;	.align 8
+	mov	a3, a14;		_j 1f;	.align 8
+	mov	a3, a15;		_j 1f;	.align 8
+
+1:
+
+#ifdef ALLOCA_EXCEPTION_IN_IRAM
+#error	iram not supported
+#else
+	l8ui	a0, a4, 0		# read ar(dst) from MOVSP instruction
+#endif
+	addi	a4, a4, 3		# step over movsp
+	_EXTUI_MOVSP_DST(a0)		# extract destination register
+	wsr	a4, EPC_1		# save new epc_1
+
+	_bnei	a0, 1, 1f		# no 'movsp a1, ax': jump
+
+        /* Move the save area. This implies the use of the L32E
+	 * and S32E instructions, because this move must be done with
+	 * the user's PS.RING privilege levels, not with ring 0
+	 * (kernel's) privileges currently active with PS.EXCM
+	 * set. Note that we have stil registered a fixup routine with the
+	 * double exception vector in case a double exception occurs.
+	 */
+
+	/* a0,a4:avail a1:old user stack a2:exc. stack a3:new user stack. */
+
+	l32e	a0, a1, -16
+	l32e	a4, a1, -12
+	s32e	a0, a3, -16
+	s32e	a4, a3, -12
+	l32e	a0, a1, -8
+	l32e	a4, a1, -4
+	s32e	a0, a3, -8
+	s32e	a4, a3, -4
+
+	/* Restore stack-pointer and all the other saved registers. */
+
+	mov	a1, a3
+
+	l32i	a4, a2, PT_AREG4
+	l32i	a3, a2, PT_AREG3
+	l32i	a0, a2, PT_AREG0
+	l32i	a2, a2, PT_AREG2
+	rfe
+
+	/*  MOVSP <at>,<as>  was invoked with <at> != a1.
+	 *  Because the stack pointer is not being modified,
+	 *  we should be able to just modify the pointer
+	 *  without moving any save area.
+	 *  The processor only traps these occurrences if the
+	 *  caller window isn't live, so unfortunately we can't
+	 *  use this as an alternate trap mechanism.
+	 *  So we just do the move.  This requires that we
+	 *  resolve the destination register, not just the source,
+	 *  so there's some extra work.
+	 *  (PERHAPS NOT REALLY NEEDED, BUT CLEANER...)
+	 */
+
+	/* a0 dst-reg, a1 user-stack, a2 stack, a3 value of src reg. */
+
+1:	movi	a4, .Lmovsp_dst
+	addx8	a4, a0, a4
+	jx	a4
+
+	.align 8
+.Lmovsp_dst:
+	s32i	a3, a2, PT_AREG0;	_j 1f;	.align 8
+	mov	a1, a3;			_j 1f;	.align 8
+	s32i	a3, a2, PT_AREG2;	_j 1f;	.align 8
+	s32i	a3, a2, PT_AREG3;	_j 1f;	.align 8
+	s32i	a3, a2, PT_AREG4;	_j 1f;	.align 8
+	mov	a5, a3;			_j 1f;	.align 8
+	mov	a6, a3;			_j 1f;	.align 8
+	mov	a7, a3;			_j 1f;	.align 8
+	mov	a8, a3;			_j 1f;	.align 8
+	mov	a9, a3;			_j 1f;	.align 8
+	mov	a10, a3;		_j 1f;	.align 8
+	mov	a11, a3;		_j 1f;	.align 8
+	mov	a12, a3;		_j 1f;	.align 8
+	mov	a13, a3;		_j 1f;	.align 8
+	mov	a14, a3;		_j 1f;	.align 8
+	mov	a15, a3;		_j 1f;	.align 8
+
+1:	l32i	a4, a2, PT_AREG4
+	l32i	a3, a2, PT_AREG3
+	l32i	a0, a2, PT_AREG0
+	l32i	a2, a2, PT_AREG2
+	rfe
+
+
+/*
+ * fast system calls.
+ *
+ * WARNING:  The kernel doesn't save the entire user context before
+ * handling a fast system call.  These functions are small and short,
+ * usually offering some functionality not available to user tasks.
+ *
+ * BE CAREFUL TO PRESERVE THE USER'S CONTEXT.
+ *
+ * Entry condition:
+ *
+ *   a0:	trashed, original value saved on stack (PT_AREG0)
+ *   a1:	a1
+ *   a2:	new stack pointer, original in DEPC
+ *   a3:	dispatch table
+ *   depc:	a2, original value saved on stack (PT_DEPC)
+ *   excsave_1:	a3
+ */
+
+ENTRY(fast_syscall_kernel)
+
+	/* Skip syscall. */
+
+	rsr	a0, EPC_1
+	addi	a0, a0, 3
+	wsr	a0, EPC_1
+
+	l32i	a0, a2, PT_DEPC
+	bgeui	a0, VALID_DOUBLE_EXCEPTION_ADDRESS, fast_syscall_unrecoverable
+
+	rsr	a0, DEPC			# get syscall-nr
+	_beqz	a0, fast_syscall_spill_registers
+
+	addi	a0, a0, -__NR_sysxtensa
+	_beqz	a0, fast_syscall_sysxtensa
+
+	j	kernel_exception
+
+
+ENTRY(fast_syscall_user)
+
+	/* Skip syscall. */
+
+	rsr	a0, EPC_1
+	addi	a0, a0, 3
+	wsr	a0, EPC_1
+
+	l32i	a0, a2, PT_DEPC
+	bgeui	a0, VALID_DOUBLE_EXCEPTION_ADDRESS, fast_syscall_unrecoverable
+
+	rsr	a0, DEPC			# get syscall-nr
+	_beqz	a0, fast_syscall_spill_registers
+
+	addi	a0, a0, -__NR_sysxtensa
+	_beqz	a0, fast_syscall_sysxtensa
+
+	j	user_exception
+
+ENTRY(fast_syscall_unrecoverable)
+
+        /* Restore all states. */
+
+        l32i    a0, a2, PT_AREG0        # restore a0
+        xsr     a2, DEPC                # restore a2, depc
+        rsr     a3, EXCSAVE_1
+
+        wsr     a0, EXCSAVE_1
+        movi    a0, unrecoverable_exception
+        callx0  a0
+
+
+
+/*
+ * sysxtensa syscall handler
+ *
+ * int sysxtensa (XTENSA_ATOMIC_SET, ptr, val, unused);
+ * int sysxtensa (XTENSA_ATOMIC_ADD, ptr, val, unused);
+ * int sysxtensa (XTENSA_ATOMIC_EXG_ADD, ptr, val, unused);
+ * int sysxtensa (XTENSA_ATOMIC_CMP_SWP, ptr, oldval, newval);
+ * a2                    a6              a3    a4      a5
+ *
+ * Entry condition:
+ *
+ *   a0:	trashed, original value saved on stack (PT_AREG0)
+ *   a1:	a1
+ *   a2:	new stack pointer, original in DEPC
+ *   a3:	dispatch table
+ *   depc:	a2, original value saved on stack (PT_DEPC)
+ *   excsave_1:	a3
+ *
+ *   PT_DEPC >= VALID_DOUBLE_EXCEPTION_ADDRESS: double exception, DEPC
+ *	     <  VALID_DOUBLE_EXCEPTION_ADDRESS: regular exception
+ *
+ * Note: we don't have to save a2; a2 holds the return value
+ *
+ * We use the two macros TRY and CATCH:
+ *
+ * TRY	 adds an entry to the __ex_table fixup table for the immediately
+ *	 following instruction.
+ *
+ * CATCH catches any exception that occurred at one of the preceeding TRY
+ *       statements and continues from there
+ *
+ * Usage TRY	l32i	a0, a1, 0
+ *		<other code>
+ *	 done:	rfe
+ *	 CATCH	<set return code>
+ *		j done
+ */
+
+#define TRY								\
+	.section __ex_table, "a";					\
+	.word	66f, 67f;						\
+	.text;								\
+66:
+
+#define CATCH								\
+67:
+
+ENTRY(fast_syscall_sysxtensa)
+
+	_beqz	a6, 1f
+	_blti	a6, SYSXTENSA_COUNT, 2f
+
+1:	j	user_exception
+
+2:	xsr	a3, EXCSAVE_1		# restore a3, excsave1
+	s32i	a7, a2, PT_AREG7
+
+	movi	a7, 4			# sizeof(unsigned int)
+	verify_area a3, a7, a0, a2, .Leac
+
+	_beqi	a6, SYSXTENSA_ATOMIC_SET, .Lset
+	_beqi	a6, SYSXTENSA_ATOMIC_EXG_ADD, .Lexg
+	_beqi	a6, SYSXTENSA_ATOMIC_ADD, .Ladd
+
+	/* Fall through for SYSXTENSA_ATOMIC_CMP_SWP */
+
+.Lswp:	/* Atomic compare and swap */
+
+TRY	l32i	a7, a3, 0		# read old value
+	bne	a7, a4, 1f		# same as old value? jump
+	s32i	a5, a3, 0		# different, modify value
+	movi	a7, 1			# and return 1
+	j	.Lret
+
+1:	movi	a7, 0			# same values: return 0
+	j	.Lret
+
+.Ladd:	/* Atomic add */
+.Lexg:	/* Atomic (exchange) add */
+
+TRY	l32i	a7, a3, 0
+	add	a4, a4, a7
+	s32i	a4, a3, 0
+	j	.Lret
+
+.Lset:	/* Atomic set */
+
+TRY	l32i	a7, a3, 0		# read old value as return value
+	s32i	a4, a3, 0		# write new value
+
+.Lret:	mov	a0, a2
+	mov	a2, a7
+	l32i	a7, a0, PT_AREG7
+	l32i	a3, a0, PT_AREG3
+	l32i	a0, a0, PT_AREG0
+	rfe
+
+CATCH
+.Leac:	movi	a7, -EFAULT
+	j	.Lret
+
+
+
+/* fast_syscall_spill_registers.
+ *
+ * Entry condition:
+ *
+ *   a0:	trashed, original value saved on stack (PT_AREG0)
+ *   a1:	a1
+ *   a2:	new stack pointer, original in DEPC
+ *   a3:	dispatch table
+ *   depc:	a2, original value saved on stack (PT_DEPC)
+ *   excsave_1:	a3
+ *
+ * Note: We assume the stack pointer is EXC_TABLE_KSTK in the fixup handler.
+ * Note: We don't need to save a2 in depc (return value)
+ */
+
+ENTRY(fast_syscall_spill_registers)
+
+	/* Register a FIXUP handler (pass current wb as a parameter) */
+
+	movi	a0, fast_syscall_spill_registers_fixup
+	s32i	a0, a3, EXC_TABLE_FIXUP
+	rsr	a0, WINDOWBASE
+	s32i	a0, a3, EXC_TABLE_PARAM
+
+	/* Save a3 and SAR on stack. */
+
+	rsr	a0, SAR
+	xsr	a3, EXCSAVE_1		# restore a3 and excsave_1
+	s32i	a0, a2, PT_AREG4	# store SAR to PT_AREG4
+	s32i	a3, a2, PT_AREG3
+
+	/* The spill routine might clobber a7, a11, and a15. */
+
+	s32i	a7, a2, PT_AREG5
+	s32i	a11, a2, PT_AREG6
+	s32i	a15, a2, PT_AREG7
+
+	call0	_spill_registers	# destroys a3, DEPC, and SAR
+
+	/* Advance PC, restore registers and SAR, and return from exception. */
+
+	l32i	a3, a2, PT_AREG4
+	l32i	a0, a2, PT_AREG0
+	wsr	a3, SAR
+	l32i	a3, a2, PT_AREG3
+
+	/* Restore clobbered registers. */
+
+	l32i	a7, a2, PT_AREG5
+	l32i	a11, a2, PT_AREG6
+	l32i	a15, a2, PT_AREG7
+
+	movi	a2, 0
+	rfe
+
+/* Fixup handler.
+ *
+ * We get here if the spill routine causes an exception, e.g. tlb miss.
+ * We basically restore WINDOWBASE and WINDOWSTART to the condition when
+ * we entered the spill routine and jump to the user exception handler.
+ *
+ * a0: value of depc, original value in depc
+ * a2: trashed, original value in EXC_TABLE_DOUBLE_SAVE
+ * a3: exctable, original value in excsave1
+ */
+
+fast_syscall_spill_registers_fixup:
+
+	rsr	a2, WINDOWBASE	# get current windowbase (a2 is saved)
+	xsr	a0, DEPC	# restore depc and a0
+	ssl	a2		# set shift (32 - WB)
+
+	/* We need to make sure the current registers (a0-a3) are preserved.
+	 * To do this, we simply set the bit for the current window frame
+	 * in WS, so that the exception handlers save them to the task stack.
+	 */
+
+	rsr	a3, EXCSAVE_1	# get spill-mask
+	slli	a2, a3, 1	# shift left by one
+
+	slli	a3, a2, 32-WSBITS
+	src	a2, a2, a3	# a1 = xxwww1yyxxxwww1yy......
+	wsr	a2, WINDOWSTART	# set corrected windowstart
+
+	movi	a3, exc_table
+	l32i	a2, a3, EXC_TABLE_DOUBLE_SAVE	# restore a2
+	l32i	a3, a3, EXC_TABLE_PARAM	# original WB (in user task)
+
+	/* Return to the original (user task) WINDOWBASE.
+	 * We leave the following frame behind:
+	 * a0, a1, a2	same
+	 * a3:		trashed (saved in excsave_1)
+	 * depc:	depc (we have to return to that address)
+	 * excsave_1:	a3
+	 */
+
+	wsr	a3, WINDOWBASE
+	rsync
+
+	/* We are now in the original frame when we entered _spill_registers:
+	 *  a0: return address
+	 *  a1: used, stack pointer
+	 *  a2: kernel stack pointer
+	 *  a3: available, saved in EXCSAVE_1
+	 *  depc: exception address
+	 *  excsave: a3
+	 * Note: This frame might be the same as above.
+	 */
+
+#ifdef SIGNAL_HANDLING_IN_DOUBLE_EXCEPTION
+	/* Restore registers we precautiously saved.
+	 * We have the value of the 'right' a3
+	 */
+
+	l32i	a7, a2, PT_AREG5
+	l32i	a11, a2, PT_AREG6
+	l32i	a15, a2, PT_AREG7
+#endif
+
+	/* Setup stack pointer. */
+
+	addi	a2, a2, -PT_USER_SIZE
+	s32i	a0, a2, PT_AREG0
+
+	/* Make sure we return to this fixup handler. */
+
+	movi	a3, fast_syscall_spill_registers_fixup_return
+	s32i	a3, a2, PT_DEPC		# setup depc
+
+	/* Jump to the exception handler. */
+
+	movi	a3, exc_table
+	rsr	a0, EXCCAUSE
+        addx4   a0, a0, a3              	# find entry in table
+        l32i    a0, a0, EXC_TABLE_FAST_USER     # load handler
+        jx      a0
+
+fast_syscall_spill_registers_fixup_return:
+
+	/* When we return here, all registers have been restored (a2: DEPC) */
+
+	wsr	a2, DEPC		# exception address
+
+	/* Restore fixup handler. */
+
+	xsr	a3, EXCSAVE_1
+	movi	a2, fast_syscall_spill_registers_fixup
+	s32i	a2, a3, EXC_TABLE_FIXUP
+	rsr	a2, WINDOWBASE
+	s32i	a2, a3, EXC_TABLE_PARAM
+	l32i	a2, a3, EXC_TABLE_KSTK
+
+#ifdef SIGNAL_HANDLING_IN_DOUBLE_EXCEPTION
+	/* Save registers again that might be clobbered. */
+
+	s32i	a7, a2, PT_AREG5
+	s32i	a11, a2, PT_AREG6
+	s32i	a15, a2, PT_AREG7
+#endif
+
+	/* Load WB at the time the exception occurred. */
+
+	rsr	a3, SAR			# WB is still in SAR
+	neg	a3, a3
+	wsr	a3, WINDOWBASE
+	rsync
+
+	/* Restore a3 and return. */
+
+	movi	a3, exc_table
+	xsr	a3, EXCSAVE_1
+
+	rfde
+
+
+/*
+ * spill all registers.
+ *
+ * This is not a real function. The following conditions must be met:
+ *
+ *  - must be called with call0.
+ *  - uses DEPC, a3 and SAR.
+ *  - the last 'valid' register of each frame are clobbered.
+ *  - the caller must have registered a fixup handler
+ *    (or be inside a critical section)
+ *  - PS_EXCM must be set (PS_WOE cleared?)
+ */
+
+ENTRY(_spill_registers)
+
+	/*
+	 * Rotate ws so that the current windowbase is at bit 0.
+	 * Assume ws = xxxwww1yy (www1 current window frame).
+	 * Rotate ws right so that a2 = yyxxxwww1.
+	 */
+
+	wsr	a2, DEPC		# preserve a2
+	rsr	a2, WINDOWBASE
+	rsr	a3, WINDOWSTART
+	ssr	a2			# holds WB
+	slli	a2, a3, WSBITS
+	or	a3, a3, a2		# a2 = xxxwww1yyxxxwww1yy
+	srl	a3, a3
+
+	/* We are done if there are no more than the current register frame. */
+
+	extui	a3, a3, 1, WSBITS-2	# a3 = 0yyxxxwww
+	movi	a2, (1 << (WSBITS-1))
+	_beqz	a3, .Lnospill		# only one active frame? jump
+
+	/* We want 1 at the top, so that we return to the current windowbase */
+
+	or	a3, a3, a2		# 1yyxxxwww
+
+	/* Skip empty frames - get 'oldest' WINDOWSTART-bit. */
+
+	wsr	a3, WINDOWSTART		# save shifted windowstart
+	neg	a2, a3
+	and	a3, a2, a3		# first bit set from right: 000010000
+
+	ffs_ws	a2, a3			# a2: shifts to skip empty frames
+	movi	a3, WSBITS
+	sub	a2, a3, a2		# WSBITS-a2:number of 0-bits from right
+	ssr	a2			# save in SAR for later.
+
+	rsr	a3, WINDOWBASE
+	add	a3, a3, a2
+	rsr	a2, DEPC		# restore a2
+	wsr	a3, WINDOWBASE
+	rsync
+
+	rsr	a3, WINDOWSTART
+	srl	a3, a3			# shift windowstart
+
+	/* WB is now just one frame below the oldest frame in the register
+	   window. WS is shifted so the oldest frame is in bit 0, thus, WB
+	   and WS differ by one 4-register frame. */
+
+	/* Save frames. Depending what call was used (call4, call8, call12),
+	 * we have to save 4,8. or 12 registers.
+	 */
+
+	_bbsi.l	a3, 1, .Lc4
+	_bbsi.l	a3, 2, .Lc8
+
+	/* Special case: we have a call12-frame starting at a4. */
+
+	_bbci.l	a3, 3, .Lc12	# bit 3 shouldn't be zero! (Jump to Lc12 first)
+
+	s32e	a4, a1, -16	# a1 is valid with an empty spill area
+	l32e	a4, a5, -12
+	s32e	a8, a4, -48
+	mov	a8, a4
+	l32e	a4, a1, -16
+	j	.Lc12c
+
+.Lloop: _bbsi.l	a3, 1, .Lc4
+	_bbci.l	a3, 2, .Lc12
+
+.Lc8:	s32e	a4, a13, -16
+	l32e	a4, a5, -12
+	s32e	a8, a4, -32
+	s32e	a5, a13, -12
+	s32e	a6, a13, -8
+	s32e	a7, a13, -4
+	s32e	a9, a4, -28
+	s32e	a10, a4, -24
+	s32e	a11, a4, -20
+
+	srli	a11, a3, 2		# shift windowbase by 2
+	rotw	2
+	_bnei	a3, 1, .Lloop
+
+.Lexit: /* Done. Do the final rotation, set WS, and return. */
+
+	rotw	1
+	rsr	a3, WINDOWBASE
+	ssl	a3
+	movi	a3, 1
+	sll	a3, a3
+	wsr	a3, WINDOWSTART
+
+.Lnospill:
+	jx	a0
+
+.Lc4:	s32e	a4, a9, -16
+	s32e	a5, a9, -12
+	s32e	a6, a9, -8
+	s32e	a7, a9, -4
+
+	srli	a7, a3, 1
+	rotw	1
+	_bnei	a3, 1, .Lloop
+	j	.Lexit
+
+.Lc12:	_bbci.l	a3, 3, .Linvalid_mask	# bit 2 shouldn't be zero!
+
+	/* 12-register frame (call12) */
+
+	l32e	a2, a5, -12
+	s32e	a8, a2, -48
+	mov	a8, a2
+
+.Lc12c: s32e	a9, a8, -44
+	s32e	a10, a8, -40
+	s32e	a11, a8, -36
+	s32e	a12, a8, -32
+	s32e	a13, a8, -28
+	s32e	a14, a8, -24
+	s32e	a15, a8, -20
+	srli	a15, a3, 3
+
+	/* The stack pointer for a4..a7 is out of reach, so we rotate the
+	 * window, grab the stackpointer, and rotate back.
+	 * Alternatively, we could also use the following approach, but that
+	 * makes the fixup routine much more complicated:
+	 * rotw	1
+	 * s32e	a0, a13, -16
+	 * ...
+	 * rotw 2
+	 */
+
+	rotw	1
+	mov	a5, a13
+	rotw	-1
+
+	s32e	a4, a9, -16
+	s32e	a5, a9, -12
+	s32e	a6, a9, -8
+	s32e	a7, a9, -4
+
+	rotw	3
+
+	_beqi	a3, 1, .Lexit
+	j	.Lloop
+
+.Linvalid_mask:
+
+	/* We get here because of an unrecoverable error in the window
+	 * registers. If we are in user space, we kill the application,
+	 * however, this condition is unrecoverable in kernel space.
+	 */
+
+	rsr	a0, PS
+	_bbci.l	a0, PS_UM_SHIFT, 1f
+
+ 	/* User space: Setup a dummy frame and kill application.
+	 * Note: We assume EXC_TABLE_KSTK contains a valid stack pointer.
+	 */
+
+	movi	a0, 1
+	movi	a1, 0
+
+	wsr	a0, WINDOWSTART
+	wsr	a1, WINDOWBASE
+	rsync
+
+	movi	a0, 0
+
+	movi	a3, exc_table
+	l32i	a1, a3, EXC_TABLE_KSTK
+	wsr	a3, EXCSAVE_1
+
+	movi	a4, PS_WOE_MASK | 1
+	wsr	a4, PS
+	rsync
+
+	movi	a6, SIGSEGV
+	movi	a4, do_exit
+	callx4	a4
+
+1:	/* Kernel space: PANIC! */
+
+	wsr	a0, EXCSAVE_1
+	movi	a0, unrecoverable_exception
+	callx0	a0		# should not return
+1:	j	1b
+
+/*
+ * We should never get here. Bail out!
+ */
+
+ENTRY(fast_second_level_miss_double_kernel)
+
+1:	movi	a0, unrecoverable_exception
+	callx0	a0		# should not return
+1:	j	1b
+
+/* First-level entry handler for user, kernel, and double 2nd-level
+ * TLB miss exceptions.  Note that for now, user and kernel miss
+ * exceptions share the same entry point and are handled identically.
+ *
+ * An old, less-efficient C version of this function used to exist.
+ * We include it below, interleaved as comments, for reference.
+ *
+ * Entry condition:
+ *
+ *   a0:	trashed, original value saved on stack (PT_AREG0)
+ *   a1:	a1
+ *   a2:	new stack pointer, original in DEPC
+ *   a3:	dispatch table
+ *   depc:	a2, original value saved on stack (PT_DEPC)
+ *   excsave_1:	a3
+ *
+ *   PT_DEPC >= VALID_DOUBLE_EXCEPTION_ADDRESS: double exception, DEPC
+ *	     <  VALID_DOUBLE_EXCEPTION_ADDRESS: regular exception
+ */
+
+ENTRY(fast_second_level_miss)
+
+	/* Save a1. Note: we don't expect a double exception. */
+
+	s32i	a1, a2, PT_AREG1
+
+	/* We need to map the page of PTEs for the user task.  Find
+	 * the pointer to that page.  Also, it's possible for tsk->mm
+	 * to be NULL while tsk->active_mm is nonzero if we faulted on
+	 * a vmalloc address.  In that rare case, we must use
+	 * active_mm instead to avoid a fault in this handler.  See
+	 *
+	 * http://mail.nl.linux.org/linux-mm/2002-08/msg00258.html
+	 *   (or search Internet on "mm vs. active_mm")
+	 *
+	 *	if (!mm)
+	 *		mm = tsk->active_mm;
+	 *	pgd = pgd_offset (mm, regs->excvaddr);
+	 *	pmd = pmd_offset (pgd, regs->excvaddr);
+	 *	pmdval = *pmd;
+	 */
+
+	GET_CURRENT(a1,a2)
+	l32i	a0, a1, TASK_MM		# tsk->mm
+	beqz	a0, 9f
+
+8:	rsr	a1, EXCVADDR		# fault address
+	_PGD_OFFSET(a0, a1, a1)
+	l32i	a0, a0, 0		# read pmdval
+	//beqi	a0, _PAGE_USER, 2f
+	beqz	a0, 2f
+
+	/* Read ptevaddr and convert to top of page-table page.
+	 *
+	 * 	vpnval = read_ptevaddr_register() & PAGE_MASK;
+	 * 	vpnval += DTLB_WAY_PGTABLE;
+	 *	pteval = mk_pte (virt_to_page(pmd_val(pmdval)), PAGE_KERNEL);
+	 *	write_dtlb_entry (pteval, vpnval);
+	 *
+	 * The messy computation for 'pteval' above really simplifies
+	 * into the following:
+	 *
+	 * pteval = ((pmdval - PAGE_OFFSET) & PAGE_MASK) | PAGE_KERNEL
+	 */
+
+	movi	a1, -PAGE_OFFSET
+	add	a0, a0, a1		# pmdval - PAGE_OFFSET
+	extui	a1, a0, 0, PAGE_SHIFT	# ... & PAGE_MASK
+	xor	a0, a0, a1
+
+
+	movi	a1, PAGE_DIRECTORY
+	or	a0, a0, a1		# ... | PAGE_DIRECTORY
+
+	rsr	a1, PTEVADDR
+	srli	a1, a1, PAGE_SHIFT
+	slli	a1, a1, PAGE_SHIFT	# ptevaddr & PAGE_MASK
+	addi	a1, a1, DTLB_WAY_PGTABLE	# ... + way_number
+
+	wdtlb	a0, a1
+	dsync
+
+	/* Exit critical section. */
+
+	movi	a0, 0
+	s32i	a0, a3, EXC_TABLE_FIXUP
+
+	/* Restore the working registers, and return. */
+
+	l32i	a0, a2, PT_AREG0
+	l32i	a1, a2, PT_AREG1
+	l32i	a2, a2, PT_DEPC
+	xsr	a3, EXCSAVE_1
+
+	bgeui	a2, VALID_DOUBLE_EXCEPTION_ADDRESS, 1f
+
+	/* Restore excsave1 and return. */
+
+	rsr	a2, DEPC
+	rfe
+
+	/* Return from double exception. */
+
+1:	xsr	a2, DEPC
+	esync
+	rfde
+
+9:	l32i	a0, a1, TASK_ACTIVE_MM	# unlikely case mm == 0
+	j	8b
+
+2:	/* Invalid PGD, default exception handling */
+
+	rsr	a1, DEPC
+	xsr	a3, EXCSAVE_1
+	s32i	a1, a2, PT_AREG2
+	s32i	a3, a2, PT_AREG3
+	mov	a1, a2
+
+	rsr	a2, PS
+	bbsi.l	a2, PS_UM_SHIFT, 1f
+	j	_kernel_exception
+1:	j	_user_exception
+
+
+/*
+ * StoreProhibitedException
+ *
+ * Update the pte and invalidate the itlb mapping for this pte.
+ *
+ * Entry condition:
+ *
+ *   a0:	trashed, original value saved on stack (PT_AREG0)
+ *   a1:	a1
+ *   a2:	new stack pointer, original in DEPC
+ *   a3:	dispatch table
+ *   depc:	a2, original value saved on stack (PT_DEPC)
+ *   excsave_1:	a3
+ *
+ *   PT_DEPC >= VALID_DOUBLE_EXCEPTION_ADDRESS: double exception, DEPC
+ *	     <  VALID_DOUBLE_EXCEPTION_ADDRESS: regular exception
+ */
+
+ENTRY(fast_store_prohibited)
+
+	/* Save a1 and a4. */
+
+	s32i	a1, a2, PT_AREG1
+	s32i	a4, a2, PT_AREG4
+
+	GET_CURRENT(a1,a2)
+	l32i	a0, a1, TASK_MM		# tsk->mm
+	beqz	a0, 9f
+
+8:	rsr	a1, EXCVADDR		# fault address
+	_PGD_OFFSET(a0, a1, a4)
+	l32i	a0, a0, 0
+	//beqi	a0, _PAGE_USER, 2f	# FIXME use _PAGE_INVALID
+	beqz	a0, 2f
+
+	_PTE_OFFSET(a0, a1, a4)
+	l32i	a4, a0, 0		# read pteval
+	movi	a1, _PAGE_VALID | _PAGE_RW
+	bnall	a4, a1, 2f
+
+	movi	a1, _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_WRENABLE
+	or	a4, a4, a1
+	rsr	a1, EXCVADDR
+	s32i	a4, a0, 0
+
+	/* We need to flush the cache if we have page coloring. */
+#if (DCACHE_WAY_SIZE > PAGE_SIZE) && XCHAL_DCACHE_IS_WRITEBACK
+	dhwb	a0, 0
+#endif
+	pdtlb	a0, a1
+	beqz	a0, 1f
+	idtlb	a0		// FIXME do we need this?
+	wdtlb	a4, a0
+1:
+
+	/* Exit critical section. */
+
+	movi	a0, 0
+	s32i	a0, a3, EXC_TABLE_FIXUP
+
+	/* Restore the working registers, and return. */
+
+	l32i	a4, a2, PT_AREG4
+	l32i	a1, a2, PT_AREG1
+	l32i	a0, a2, PT_AREG0
+	l32i	a2, a2, PT_DEPC
+
+	/* Restore excsave1 and a3. */
+
+	xsr	a3, EXCSAVE_1
+	bgeui	a2, VALID_DOUBLE_EXCEPTION_ADDRESS, 1f
+
+	rsr	a2, DEPC
+	rfe
+
+	/* Double exception. Restore FIXUP handler and return. */
+
+1:	xsr	a2, DEPC
+	esync
+	rfde
+
+9:	l32i	a0, a1, TASK_ACTIVE_MM	# unlikely case mm == 0
+	j	8b
+
+2:	/* If there was a problem, handle fault in C */
+
+	rsr	a4, DEPC	# still holds a2
+	xsr	a3, EXCSAVE_1
+	s32i	a4, a2, PT_AREG2
+	s32i	a3, a2, PT_AREG3
+	l32i	a4, a2, PT_AREG4
+	mov	a1, a2
+
+	rsr	a2, PS
+	bbsi.l	a2, PS_UM_SHIFT, 1f
+	j	_kernel_exception
+1:	j	_user_exception
+
+
+#if XCHAL_EXTRA_SA_SIZE
+
+#warning fast_coprocessor untested
+
+/*
+ * Entry condition:
+ *
+ *   a0:	trashed, original value saved on stack (PT_AREG0)
+ *   a1:	a1
+ *   a2:	new stack pointer, original in DEPC
+ *   a3:	dispatch table
+ *   depc:	a2, original value saved on stack (PT_DEPC)
+ *   excsave_1:	a3
+ *
+ *   PT_DEPC >= VALID_DOUBLE_EXCEPTION_ADDRESS: double exception, DEPC
+ *	     <  VALID_DOUBLE_EXCEPTION_ADDRESS: regular exception
+ */
+
+ENTRY(fast_coprocessor_double)
+	wsr	a0, EXCSAVE_1
+	movi	a0, unrecoverable_exception
+	callx0	a0
+
+ENTRY(fast_coprocessor)
+
+	/* Fatal if we are in a double exception. */
+
+	l32i	a0, a2, PT_DEPC
+	_bgeui	a0, VALID_DOUBLE_EXCEPTION_ADDRESS, fast_coprocessor_double
+
+	/* Save some registers a1, a3, a4, SAR */
+
+	xsr	a3, EXCSAVE_1
+	s32i	a3, a2, PT_AREG3
+	rsr	a3, SAR
+	s32i	a4, a2, PT_AREG4
+	s32i	a1, a2, PT_AREG1
+	s32i	a5, a1, PT_AREG5
+	s32i	a3, a2, PT_SAR
+	mov	a1, a2
+
+	/* Currently, the HAL macros only guarantee saving a0 and a1.
+	 * These can and will be refined in the future, but for now,
+	 * just save the remaining registers of a2...a15.
+	 */
+	s32i	a6, a1, PT_AREG6
+	s32i	a7, a1, PT_AREG7
+	s32i	a8, a1, PT_AREG8
+	s32i	a9, a1, PT_AREG9
+	s32i	a10, a1, PT_AREG10
+	s32i	a11, a1, PT_AREG11
+	s32i	a12, a1, PT_AREG12
+	s32i	a13, a1, PT_AREG13
+	s32i	a14, a1, PT_AREG14
+	s32i	a15, a1, PT_AREG15
+
+	/* Find coprocessor number. Subtract first CP EXCCAUSE from EXCCAUSE */
+
+	rsr	a0, EXCCAUSE
+	addi	a3, a0, -XCHAL_EXCCAUSE_COPROCESSOR0_DISABLED
+
+	/* Set corresponding CPENABLE bit */
+
+	movi	a4, 1
+	ssl	a3			# SAR: 32 - coprocessor_number
+	rsr	a5, CPENABLE
+	sll	a4, a4
+	or	a4, a5, a4
+	wsr	a4, CPENABLE
+	rsync
+	movi	a5, coprocessor_info	# list of owner and offset into cp_save
+	addx8	a0, a4, a5		# entry for CP
+
+	bne	a4, a5, .Lload		# bit wasn't set before, cp not in use
+
+	/* Now compare the current task with the owner of the coprocessor.
+	 * If they are the same, there is no reason to save or restore any
+	 * coprocessor state. Having already enabled the coprocessor,
+	 * branch ahead to return.
+	 */
+	GET_CURRENT(a5,a1)
+	l32i	a4, a0, COPROCESSOR_INFO_OWNER	# a4: current owner for this CP
+	beq	a4, a5, .Ldone
+
+	/* Find location to dump current coprocessor state:
+	 *  task_struct->task_cp_save_offset + coprocessor_offset[coprocessor]
+	 *
+	 * Note: a0 pointer to the entry in the coprocessor owner table,
+	 *	 a3 coprocessor number,
+         *	 a4 current owner of coprocessor.
+	 */
+	l32i	a5, a0, COPROCESSOR_INFO_OFFSET
+	addi	a2, a4, THREAD_CP_SAVE
+	add	a2, a2, a5
+
+	/* Store current coprocessor states. (a5 still has CP number) */
+
+	xchal_cpi_store_funcbody
+
+	/* The macro might have destroyed a3 (coprocessor number), but
+	 * SAR still has 32 - coprocessor_number!
+	 */
+	movi	a3, 32
+	rsr	a4, SAR
+	sub	a3, a3, a4
+
+.Lload:	/* A new task now owns the corpocessors. Save its TCB pointer into
+	 * the coprocessor owner table.
+	 *
+	 * Note: a0 pointer to the entry in the coprocessor owner table,
+	 *	 a3 coprocessor number.
+	 */
+	GET_CURRENT(a4,a1)
+	s32i	a4, a0, 0
+
+	/* Find location from where to restore the current coprocessor state.*/
+
+	l32i	a5, a0, COPROCESSOR_INFO_OFFSET
+	addi	a2, a4, THREAD_CP_SAVE
+	add	a2, a2, a4
+
+	xchal_cpi_load_funcbody
+
+	/* We must assume that the xchal_cpi_store_funcbody macro destroyed
+	 * registers a2..a15.
+	 */
+
+.Ldone:	l32i	a15, a1, PT_AREG15
+	l32i	a14, a1, PT_AREG14
+	l32i	a13, a1, PT_AREG13
+	l32i	a12, a1, PT_AREG12
+	l32i	a11, a1, PT_AREG11
+	l32i	a10, a1, PT_AREG10
+	l32i	a9, a1, PT_AREG9
+	l32i	a8, a1, PT_AREG8
+	l32i	a7, a1, PT_AREG7
+	l32i	a6, a1, PT_AREG6
+	l32i	a5, a1, PT_AREG5
+	l32i	a4, a1, PT_AREG4
+	l32i	a3, a1, PT_AREG3
+	l32i	a2, a1, PT_AREG2
+	l32i	a0, a1, PT_AREG0
+	l32i	a1, a1, PT_AREG1
+
+	rfe
+
+#endif /* XCHAL_EXTRA_SA_SIZE */
+
+/*
+ * Task switch.
+ *
+ * struct task*  _switch_to (struct task* prev, struct task* next)
+ *         a2                              a2                 a3
+ */
+
+ENTRY(_switch_to)
+
+	entry	a1, 16
+
+	mov	a4, a3			# preserve a3
+
+	s32i	a0, a2, THREAD_RA	# save return address
+	s32i	a1, a2, THREAD_SP	# save stack pointer
+
+	/* Disable ints while we manipulate the stack pointer; spill regs. */
+
+	movi	a5, PS_EXCM_MASK | LOCKLEVEL
+	xsr	a5, PS
+	rsr	a3, EXCSAVE_1
+	rsync
+	s32i	a3, a3, EXC_TABLE_FIXUP	/* enter critical section */
+
+	call0	_spill_registers
+
+	/* Set kernel stack (and leave critical section)
+	 * Note: It's save to set it here. The stack will not be overwritten
+	 *       because the kernel stack will only be loaded again after
+	 *       we return from kernel space.
+	 */
+
+	l32i	a0, a4, TASK_THREAD_INFO
+	rsr	a3, EXCSAVE_1		# exc_table
+	movi	a1, 0
+	addi	a0, a0, PT_REGS_OFFSET
+	s32i	a1, a3, EXC_TABLE_FIXUP
+	s32i	a0, a3, EXC_TABLE_KSTK
+
+	/* restore context of the task that 'next' addresses */
+
+	l32i	a0, a4, THREAD_RA	/* restore return address */
+	l32i	a1, a4, THREAD_SP	/* restore stack pointer */
+
+	wsr	a5, PS
+	rsync
+
+	retw
+
+
+ENTRY(ret_from_fork)
+
+	/* void schedule_tail (struct task_struct *prev)
+	 * Note: prev is still in a6 (return value from fake call4 frame)
+	 */
+	movi	a4, schedule_tail
+	callx4	a4
+
+	movi	a4, do_syscall_trace
+	callx4	a4
+
+	j	common_exception_return
+
+
+
+/*
+ * Table of syscalls
+ */
+
+.data
+.align  4
+.global sys_call_table
+sys_call_table:
+
+#define SYSCALL(call, narg) .word call
+#include "syscalls.h"
+
+/*
+ * Number of arguments of each syscall
+ */
+
+.global sys_narg_table
+sys_narg_table:
+
+#undef SYSCALL
+#define SYSCALL(call, narg) .byte narg
+#include "syscalls.h"
+
diff --git a/arch/xtensa/kernel/head.S b/arch/xtensa/kernel/head.S
new file mode 100644
index 00000000000..6e9b5225b8f
--- /dev/null
+++ b/arch/xtensa/kernel/head.S
@@ -0,0 +1,237 @@
+/*
+ * arch/xtensa/kernel/head.S
+ *
+ * Xtensa Processor startup code.
+ *
+ * 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.
+ *
+ * Chris Zankel <chris@zankel.net>
+ * Marc Gauthier <marc@tensilica.com, marc@alumni.uwaterloo.ca>
+ * Joe Taylor <joe@tensilica.com, joetylr@yahoo.com>
+ * Kevin Chea
+ */
+
+#include <xtensa/cacheasm.h>
+#include <linux/config.h>
+#include <asm/processor.h>
+#include <asm/page.h>
+
+/*
+ * This module contains the entry code for kernel images. It performs the
+ * minimal setup needed to call the generic C routines.
+ *
+ * Prerequisites:
+ *
+ * - The kernel image has been loaded to the actual address where it was
+ *   compiled to.
+ * - a2 contains either 0 or a pointer to a list of boot parameters.
+ *   (see setup.c for more details)
+ *
+ */
+
+	.macro	iterate	from, to , cmd
+		.ifeq	((\to - \from) & ~0xfff)
+			\cmd	\from
+			iterate "(\from+1)", \to, \cmd
+		.endif
+	.endm
+
+/*
+ *  _start
+ *
+ *  The bootloader passes a pointer to a list of boot parameters in a2.
+ */
+
+	/* The first bytes of the kernel image must be an instruction, so we
+	 * manually allocate and define the literal constant we need for a jx
+	 * instruction.
+	 */
+
+	.section .head.text, "ax"
+	.globl _start
+_start:	_j	2f
+	.align	4
+1:	.word	_startup
+2:	l32r	a0, 1b
+	jx	a0
+
+	.text
+	.align 4
+_startup:
+
+	/* Disable interrupts and exceptions. */
+
+	movi	a0, XCHAL_PS_EXCM_MASK
+	wsr	a0, PS
+
+	/* Preserve the pointer to the boot parameter list in EXCSAVE_1 */
+
+	wsr	a2, EXCSAVE_1
+
+	/* Start with a fresh windowbase and windowstart.  */
+
+	movi	a1, 1
+	movi	a0, 0
+	wsr	a1, WINDOWSTART
+	wsr	a0, WINDOWBASE
+	rsync
+
+	/* Set a0 to 0 for the remaining initialization. */
+
+	movi	a0, 0
+
+	/* Clear debugging registers. */
+
+#if XCHAL_HAVE_DEBUG
+	wsr	a0, IBREAKENABLE
+	wsr	a0, ICOUNT
+	movi	a1, 15
+	wsr	a0, ICOUNTLEVEL
+
+	.macro reset_dbreak num
+	wsr	a0, DBREAKC + \num
+	.endm
+
+        iterate 0, XCHAL_NUM_IBREAK-1, reset_dbreak
+#endif
+
+	/* Clear CCOUNT (not really necessary, but nice) */
+
+	wsr	a0, CCOUNT	# not really necessary, but nice
+
+	/* Disable zero-loops. */
+
+#if XCHAL_HAVE_LOOPS
+	wsr	a0, LCOUNT
+#endif
+
+	/* Disable all timers. */
+
+	.macro	reset_timer	num
+	wsr	a0, CCOMPARE_0 + \num
+	.endm
+	iterate	0, XCHAL_NUM_TIMERS-1, reset_timer
+
+	/* Interrupt initialization. */
+
+	movi	a2, XCHAL_INTTYPE_MASK_SOFTWARE | XCHAL_INTTYPE_MASK_EXTERN_EDGE
+	wsr	a0, INTENABLE
+	wsr	a2, INTCLEAR
+
+	/* Disable coprocessors. */
+
+#if XCHAL_CP_NUM > 0
+	wsr	a0, CPENABLE
+#endif
+
+	/* Set PS.INTLEVEL=1, PS.WOE=0, kernel stack, PS.EXCM=0
+	 *
+	 * Note: PS.EXCM must be cleared before using any loop
+	 *	 instructions; otherwise, they are silently disabled, and
+	 * 	 at most one iteration of the loop is executed.
+	 */
+
+	movi	a1, 1
+	wsr	a1, PS
+	rsync
+
+	/*  Initialize the caches.
+	 *  Does not include flushing writeback d-cache.
+	 *  a6, a7 are just working registers (clobbered).
+	 */
+
+	icache_reset  a2, a3
+	dcache_reset  a2, a3
+
+	/* Unpack data sections
+	 *
+	 * The linker script used to build the Linux kernel image
+	 * creates a table located at __boot_reloc_table_start
+	 * that contans the information what data needs to be unpacked.
+	 *
+	 * Uses a2-a7.
+	 */
+
+	movi	a2, __boot_reloc_table_start
+	movi	a3, __boot_reloc_table_end
+
+1:	beq	a2, a3, 3f	# no more entries?
+	l32i	a4, a2, 0	# start destination (in RAM)
+	l32i	a5, a2, 4	# end desination (in RAM)
+	l32i	a6, a2, 8	# start source (in ROM)
+	addi	a2, a2, 12	# next entry
+	beq	a4, a5, 1b	# skip, empty entry
+	beq	a4, a6, 1b	# skip, source and dest. are the same
+
+2:	l32i	a7, a6, 0	# load word
+	addi	a6, a6, 4
+	s32i	a7, a4, 0	# store word
+	addi	a4, a4, 4
+	bltu	a4, a5, 2b
+	j	1b
+
+3:
+	/* All code and initialized data segments have been copied.
+	 * Now clear the BSS segment.
+	 */
+
+	movi	a2, _bss_start	# start of BSS
+	movi	a3, _bss_end	# end of BSS
+
+1:	addi	a2, a2, 4
+	s32i	a0, a2, 0
+	blt	a2, a3, 1b
+
+#if XCHAL_DCACHE_IS_WRITEBACK
+
+	/* After unpacking, flush the writeback cache to memory so the
+	 * instructions/data are available.
+	 */
+
+	dcache_writeback_all	a2, a3
+#endif
+
+	/* Setup stack and enable window exceptions (keep irqs disabled) */
+
+	movi	a1, init_thread_union
+	addi	a1, a1, KERNEL_STACK_SIZE
+
+	movi	a2, 0x00040001		# WOE=1, INTLEVEL=1, UM=0
+	wsr	a2, PS			# (enable reg-windows; progmode stack)
+	rsync
+
+	/* Set up EXCSAVE[DEBUGLEVEL] to point to the Debug Exception Handler.*/
+
+	movi	a2, debug_exception
+	wsr	a2, EXCSAVE + XCHAL_DEBUGLEVEL
+
+	/* Set up EXCSAVE[1] to point to the exc_table. */
+
+	movi	a6, exc_table
+	xsr	a6, EXCSAVE_1
+
+	/* init_arch kick-starts the linux kernel */
+
+	movi	a4, init_arch
+	callx4	a4
+
+	movi	a4, start_kernel
+	callx4	a4
+
+should_never_return:
+	j	should_never_return
+
+	/* Define some common data structures here.  We define them
+	 * here in this assembly file due to their unusual alignment
+	 * requirements.
+	 */
+
+	.comm   swapper_pg_dir,PAGE_SIZE,PAGE_SIZE
+	.comm	empty_bad_page_table,PAGE_SIZE,PAGE_SIZE
+	.comm	empty_bad_page,PAGE_SIZE,PAGE_SIZE
+	.comm	empty_zero_page,PAGE_SIZE,PAGE_SIZE
+
diff --git a/arch/xtensa/kernel/irq.c b/arch/xtensa/kernel/irq.c
new file mode 100644
index 00000000000..4cbf6d91571
--- /dev/null
+++ b/arch/xtensa/kernel/irq.c
@@ -0,0 +1,192 @@
+/*
+ * linux/arch/xtensa/kernel/irq.c
+ *
+ * Xtensa built-in interrupt controller and some generic functions copied
+ * from i386.
+ *
+ * Copyright (C) 2002 - 2005 Tensilica, Inc.
+ * Copyright (C) 1992, 1998 Linus Torvalds, Ingo Molnar
+ *
+ *
+ * Chris Zankel <chris@zankel.net>
+ * Kevin Chea
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/seq_file.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/kernel_stat.h>
+
+#include <asm/uaccess.h>
+#include <asm/platform.h>
+
+static void enable_xtensa_irq(unsigned int irq);
+static void disable_xtensa_irq(unsigned int irq);
+static void mask_and_ack_xtensa(unsigned int irq);
+static void end_xtensa_irq(unsigned int irq);
+
+static unsigned int cached_irq_mask;
+
+atomic_t irq_err_count;
+
+/*
+ * 'what should we do if we get a hw irq event on an illegal vector'.
+ * each architecture has to answer this themselves.
+ */
+void ack_bad_irq(unsigned int irq)
+{
+          printk("unexpected IRQ trap at vector %02x\n", irq);
+}
+
+/*
+ * do_IRQ handles all normal device IRQ's (the special
+ * SMP cross-CPU interrupts have their own specific
+ * handlers).
+ */
+
+unsigned int  do_IRQ(int irq, struct pt_regs *regs)
+{
+	irq_enter();
+
+#ifdef CONFIG_DEBUG_STACKOVERFLOW
+	/* Debugging check for stack overflow: is there less than 1KB free? */
+	{
+		unsigned long sp;
+
+		__asm__ __volatile__ ("mov %0, a1\n" : "=a" (sp));
+		sp &= THREAD_SIZE - 1;
+
+		if (unlikely(sp < (sizeof(thread_info) + 1024)))
+			printk("Stack overflow in do_IRQ: %ld\n",
+			       sp - sizeof(struct thread_info));
+	}
+#endif
+
+	__do_IRQ(irq, regs);
+
+	irq_exit();
+
+	return 1;
+}
+
+/*
+ * Generic, controller-independent functions:
+ */
+
+int show_interrupts(struct seq_file *p, void *v)
+{
+	int i = *(loff_t *) v, j;
+	struct irqaction * action;
+	unsigned long flags;
+
+	if (i == 0) {
+		seq_printf(p, "           ");
+		for (j=0; j<NR_CPUS; j++)
+			if (cpu_online(j))
+				seq_printf(p, "CPU%d       ",j);
+		seq_putc(p, '\n');
+	}
+
+	if (i < NR_IRQS) {
+		spin_lock_irqsave(&irq_desc[i].lock, flags);
+		action = irq_desc[i].action;
+		if (!action)
+			goto skip;
+		seq_printf(p, "%3d: ",i);
+#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]);
+#endif
+		seq_printf(p, " %14s", irq_desc[i].handler->typename);
+		seq_printf(p, "  %s", action->name);
+
+		for (action=action->next; action; action = action->next)
+			seq_printf(p, ", %s", action->name);
+
+		seq_putc(p, '\n');
+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));
+		seq_putc(p, '\n');
+		seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count));
+	}
+	return 0;
+}
+/* shutdown is same as "disable" */
+#define shutdown_xtensa_irq disable_xtensa_irq
+
+static unsigned int startup_xtensa_irq(unsigned int irq)
+{
+	enable_xtensa_irq(irq);
+	return 0;               /* never anything pending */
+}
+
+static struct hw_interrupt_type xtensa_irq_type = {
+	"Xtensa-IRQ",
+	startup_xtensa_irq,
+	shutdown_xtensa_irq,
+	enable_xtensa_irq,
+	disable_xtensa_irq,
+	mask_and_ack_xtensa,
+	end_xtensa_irq
+};
+
+static inline void mask_irq(unsigned int irq)
+{
+	cached_irq_mask &= ~(1 << irq);
+	set_sr (cached_irq_mask, INTENABLE);
+}
+
+static inline void unmask_irq(unsigned int irq)
+{
+	cached_irq_mask |= 1 << irq;
+	set_sr (cached_irq_mask, INTENABLE);
+}
+
+static void disable_xtensa_irq(unsigned int irq)
+{
+	unsigned long flags;
+	local_save_flags(flags);
+	mask_irq(irq);
+	local_irq_restore(flags);
+}
+
+static void enable_xtensa_irq(unsigned int irq)
+{
+	unsigned long flags;
+	local_save_flags(flags);
+	unmask_irq(irq);
+	local_irq_restore(flags);
+}
+
+static void mask_and_ack_xtensa(unsigned int irq)
+{
+        disable_xtensa_irq(irq);
+}
+
+static void end_xtensa_irq(unsigned int irq)
+{
+        enable_xtensa_irq(irq);
+}
+
+
+void __init init_IRQ(void)
+{
+	int i;
+
+	for (i=0; i < XTENSA_NR_IRQS; i++)
+		irq_desc[i].handler = &xtensa_irq_type;
+
+	cached_irq_mask = 0;
+
+	platform_init_irq();
+}
diff --git a/arch/xtensa/kernel/module.c b/arch/xtensa/kernel/module.c
new file mode 100644
index 00000000000..d1683cfa19a
--- /dev/null
+++ b/arch/xtensa/kernel/module.c
@@ -0,0 +1,78 @@
+/*
+ * arch/xtensa/kernel/platform.c
+ *
+ * Module support.
+ *
+ * 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.
+ *
+ * Chris Zankel <chris@zankel.net>
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/moduleloader.h>
+#include <linux/elf.h>
+#include <linux/vmalloc.h>
+#include <linux/fs.h>
+#include <linux/string.h>
+#include <linux/kernel.h>
+#include <linux/cache.h>
+
+LIST_HEAD(module_buf_list);
+
+void *module_alloc(unsigned long size)
+{
+  panic("module_alloc not implemented");
+}
+
+void module_free(struct module *mod, void *module_region)
+{
+  panic("module_free not implemented");
+}
+
+int module_frob_arch_sections(Elf32_Ehdr *hdr,
+    			      Elf32_Shdr *sechdrs,
+			      char *secstrings,
+			      struct module *me)
+{
+  panic("module_frob_arch_sections not implemented");
+}
+
+int apply_relocate(Elf32_Shdr *sechdrs,
+    		   const char *strtab,
+		   unsigned int symindex,
+		   unsigned int relsec,
+		   struct module *module)
+{
+  panic ("apply_relocate not implemented");
+}
+
+int apply_relocate_add(Elf32_Shdr *sechdrs,
+		       const char *strtab,
+		       unsigned int symindex,
+		       unsigned int relsec,
+		       struct module *module)
+{
+  panic("apply_relocate_add not implemented");
+}
+
+int module_finalize(const Elf_Ehdr *hdr,
+    		    const Elf_Shdr *sechdrs,
+		    struct module *me)
+{
+  panic ("module_finalize not implemented");
+}
+
+void module_arch_cleanup(struct module *mod)
+{
+  panic("module_arch_cleanup not implemented");
+}
+
+struct bug_entry *module_find_bug(unsigned long bugaddr)
+{
+  panic("module_find_bug not implemented");
+}
diff --git a/arch/xtensa/kernel/pci-dma.c b/arch/xtensa/kernel/pci-dma.c
new file mode 100644
index 00000000000..84fde258cf8
--- /dev/null
+++ b/arch/xtensa/kernel/pci-dma.c
@@ -0,0 +1,73 @@
+/*
+ * arch/xtensa/pci-dma.c
+ *
+ * DMA coherent memory allocation.
+ *
+ * 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.
+ *
+ * Copyright (C) 2002 - 2005 Tensilica Inc.
+ *
+ * Based on version for i386.
+ *
+ * Chris Zankel <chris@zankel.net>
+ * Joe Taylor <joe@tensilica.com, joetylr@yahoo.com>
+ */
+
+#include <linux/types.h>
+#include <linux/mm.h>
+#include <linux/string.h>
+#include <linux/pci.h>
+#include <asm/io.h>
+#include <asm/cacheflush.h>
+
+/*
+ * Note: We assume that the full memory space is always mapped to 'kseg'
+ *	 Otherwise we have to use page attributes (not implemented).
+ */
+
+void *
+dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *handle, int gfp)
+{
+	void *ret;
+
+	/* ignore region speicifiers */
+	gfp &= ~(__GFP_DMA | __GFP_HIGHMEM);
+
+	if (dev == NULL || (*dev->dma_mask < 0xffffffff))
+		gfp |= GFP_DMA;
+	ret = (void *)__get_free_pages(gfp, get_order(size));
+
+	if (ret != NULL) {
+		memset(ret, 0, size);
+		*handle = virt_to_bus(ret);
+	}
+	return (void*) BYPASS_ADDR((unsigned long)ret);
+}
+
+void dma_free_coherent(struct device *hwdev, size_t size,
+			 void *vaddr, dma_addr_t dma_handle)
+{
+	free_pages(CACHED_ADDR((unsigned long)vaddr), get_order(size));
+}
+
+
+void consistent_sync(void *vaddr, size_t size, int direction)
+{
+	switch (direction) {
+	case PCI_DMA_NONE:
+		BUG();
+	case PCI_DMA_FROMDEVICE:        /* invalidate only */
+		__invalidate_dcache_range((unsigned long)vaddr,
+				          (unsigned long)size);
+		break;
+
+	case PCI_DMA_TODEVICE:          /* writeback only */
+	case PCI_DMA_BIDIRECTIONAL:     /* writeback and invalidate */
+		__flush_invalidate_dcache_range((unsigned long)vaddr,
+				    		(unsigned long)size);
+		break;
+	}
+}
diff --git a/arch/xtensa/kernel/pci.c b/arch/xtensa/kernel/pci.c
new file mode 100644
index 00000000000..d29a8164863
--- /dev/null
+++ b/arch/xtensa/kernel/pci.c
@@ -0,0 +1,563 @@
+/*
+ * arch/xtensa/pcibios.c
+ *
+ * PCI bios-type initialisation for PCI machines
+ *
+ * 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.
+ *
+ * Copyright (C) 2001-2005 Tensilica Inc.
+ *
+ * Based largely on work from Cort (ppc/kernel/pci.c)
+ * IO functions copied from sparc.
+ *
+ * Chris Zankel <chris@zankel.net>
+ *
+ */
+
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/string.h>
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/errno.h>
+#include <linux/bootmem.h>
+
+#include <asm/pci-bridge.h>
+#include <asm/platform.h>
+
+#undef DEBUG
+
+#ifdef DEBUG
+#define DBG(x...) printk(x)
+#else
+#define DBG(x...)
+#endif
+
+/* PCI Controller */
+
+
+/*
+ * pcibios_alloc_controller
+ * pcibios_enable_device
+ * pcibios_fixups
+ * pcibios_align_resource
+ * pcibios_fixup_bus
+ * pcibios_setup
+ * pci_bus_add_device
+ * pci_mmap_page_range
+ */
+
+struct pci_controller* pci_ctrl_head;
+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
+ * addresses to be allocated in the 0x000-0x0ff region
+ * modulo 0x400.
+ *
+ * Why? Because some silly external IO cards only decode
+ * the low 10 bits of the IO address. The 0x00-0xff region
+ * is reserved for motherboard devices that decode all 16
+ * bits, so it's ok to allocate at, say, 0x2800-0x28ff,
+ * but we want to try to avoid allocating at 0x2900-0x2bff
+ * which might have be mirrored at 0x0100-0x03ff..
+ */
+void
+pcibios_align_resource(void *data, struct resource *res, unsigned long size,
+    		       unsigned long align)
+{
+	struct pci_dev *dev = data;
+
+	if (res->flags & IORESOURCE_IO) {
+		unsigned long start = res->start;
+
+		if (size > 0x100) {
+			printk(KERN_ERR "PCI: I/O Region %s/%d too large"
+			       " (%ld bytes)\n", dev->slot_name,
+			       dev->resource - res, size);
+		}
+
+		if (start & 0x300) {
+			start = (start + 0x3ff) & ~0x3ff;
+			res->start = start;
+		}
+	}
+}
+
+int
+pcibios_enable_resources(struct pci_dev *dev, int mask)
+{
+	u16 cmd, old_cmd;
+	int idx;
+	struct resource *r;
+
+	pci_read_config_word(dev, PCI_COMMAND, &cmd);
+	old_cmd = cmd;
+	for(idx=0; idx<6; idx++) {
+		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);
+			return -EINVAL;
+		}
+		if (r->flags & IORESOURCE_IO)
+			cmd |= PCI_COMMAND_IO;
+		if (r->flags & IORESOURCE_MEM)
+			cmd |= PCI_COMMAND_MEMORY;
+	}
+	if (dev->resource[PCI_ROM_RESOURCE].start)
+		cmd |= PCI_COMMAND_MEMORY;
+	if (cmd != old_cmd) {
+		printk("PCI: Enabling device %s (%04x -> %04x)\n",
+			dev->slot_name, old_cmd, cmd);
+		pci_write_config_word(dev, PCI_COMMAND, cmd);
+	}
+	return 0;
+}
+
+struct pci_controller * __init pcibios_alloc_controller(void)
+{
+	struct pci_controller *pci_ctrl;
+
+	pci_ctrl = (struct pci_controller *)alloc_bootmem(sizeof(*pci_ctrl));
+	memset(pci_ctrl, 0, sizeof(struct pci_controller));
+
+	*pci_ctrl_tail = pci_ctrl;
+	pci_ctrl_tail = &pci_ctrl->next;
+
+	return pci_ctrl;
+}
+
+static int __init pcibios_init(void)
+{
+	struct pci_controller *pci_ctrl;
+	struct pci_bus *bus;
+	int next_busno = 0, i;
+
+	printk("PCI: Probing PCI hardware\n");
+
+	/* Scan all of the recorded PCI controllers.  */
+	for (pci_ctrl = pci_ctrl_head; pci_ctrl; pci_ctrl = pci_ctrl->next) {
+		pci_ctrl->last_busno = 0xff;
+		bus = pci_scan_bus(pci_ctrl->first_busno, pci_ctrl->ops,
+				   pci_ctrl);
+		if (pci_ctrl->io_resource.flags) {
+			unsigned long offs;
+
+			offs = (unsigned long)pci_ctrl->io_space.base;
+			pci_ctrl->io_resource.start += offs;
+			pci_ctrl->io_resource.end += offs;
+			bus->resource[0] = &pci_ctrl->io_resource;
+		}
+		for (i = 0; i < 3; ++i)
+			if (pci_ctrl->mem_resources[i].flags)
+				bus->resource[i+1] =&pci_ctrl->mem_resources[i];
+		pci_ctrl->bus = bus;
+		pci_ctrl->last_busno = bus->subordinate;
+		if (next_busno <= pci_ctrl->last_busno)
+			next_busno = pci_ctrl->last_busno+1;
+	}
+	pci_bus_count = next_busno;
+
+	return platform_pcibios_fixup();
+}
+
+subsys_initcall(pcibios_init);
+
+void __init pcibios_fixup_bus(struct pci_bus *bus)
+{
+	struct pci_controller *pci_ctrl = bus->sysdata;
+	struct resource *res;
+	unsigned long io_offset;
+	int i;
+
+	io_offset = (unsigned long)pci_ctrl->io_space.base;
+	if (bus->parent == NULL) {
+		/* this is a host bridge - fill in its resources */
+		pci_ctrl->bus = bus;
+
+		bus->resource[0] = res = &pci_ctrl->io_resource;
+		if (!res->flags) {
+			if (io_offset)
+				printk (KERN_ERR "I/O resource not set for host"
+					" bridge %d\n", pci_ctrl->index);
+			res->start = 0;
+			res->end = IO_SPACE_LIMIT;
+			res->flags = IORESOURCE_IO;
+		}
+		res->start += io_offset;
+		res->end += io_offset;
+
+		for (i = 0; i < 3; i++) {
+			res = &pci_ctrl->mem_resources[i];
+			if (!res->flags) {
+				if (i > 0)
+					continue;
+				printk(KERN_ERR "Memory resource not set for "
+				       "host bridge %d\n", pci_ctrl->index);
+				res->start = 0;
+				res->end = ~0U;
+				res->flags = IORESOURCE_MEM;
+			}
+			bus->resource[i+1] = res;
+		}
+	} else {
+		/* This is a subordinate bridge */
+		pci_read_bridge_bases(bus);
+
+		for (i = 0; i < 4; i++) {
+			if ((res = bus->resource[i]) == NULL || !res->flags)
+				continue;
+			if (io_offset && (res->flags & IORESOURCE_IO)) {
+				res->start += io_offset;
+				res->end += io_offset;
+			}
+		}
+	}
+}
+
+char __init *pcibios_setup(char *str)
+{
+	return str;
+}
+
+/* the next one is stolen from the alpha port... */
+
+void __init
+pcibios_update_irq(struct pci_dev *dev, int irq)
+{
+	pci_write_config_byte(dev, PCI_INTERRUPT_LINE, irq);
+}
+
+int pcibios_enable_device(struct pci_dev *dev, int mask)
+{
+	u16 cmd, old_cmd;
+	int idx;
+	struct resource *r;
+
+	pci_read_config_word(dev, PCI_COMMAND, &cmd);
+	old_cmd = cmd;
+	for (idx=0; idx<6; idx++) {
+		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);
+			return -EINVAL;
+		}
+		if (r->flags & IORESOURCE_IO)
+			cmd |= PCI_COMMAND_IO;
+		if (r->flags & IORESOURCE_MEM)
+			cmd |= PCI_COMMAND_MEMORY;
+	}
+	if (cmd != old_cmd) {
+		printk("PCI: Enabling device %s (%04x -> %04x)\n",
+		       dev->slot_name, old_cmd, cmd);
+		pci_write_config_word(dev, PCI_COMMAND, cmd);
+	}
+
+	return 0;
+}
+
+#ifdef CONFIG_PROC_FS
+
+/*
+ * Return the index of the PCI controller for device pdev.
+ */
+
+int
+pci_controller_num(struct pci_dev *dev)
+{
+	struct pci_controller *pci_ctrl = (struct pci_controller*) dev->sysdata;
+	return pci_ctrl->index;
+}
+
+#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.
+ *  -- paulus.
+ */
+
+/*
+ * Adjust vm_pgoff of VMA such that it is the physical page offset
+ * corresponding to the 32-bit pci bus offset for DEV requested by the user.
+ *
+ * Basically, the user finds the base address for his device which he wishes
+ * to mmap.  They read the 32-bit value from the config space base register,
+ * add whatever PAGE_SIZE multiple offset they wish, and feed this into the
+ * offset parameter of mmap on /proc/bus/pci/XXX for that device.
+ *
+ * Returns negative error code on failure, zero on success.
+ */
+static __inline__ int
+__pci_mmap_make_offset(struct pci_dev *dev, struct vm_area_struct *vma,
+		       enum pci_mmap_state mmap_state)
+{
+	struct pci_controller *pci_ctrl = (struct pci_controller*) dev->sysdata;
+	unsigned long offset = vma->vm_pgoff << PAGE_SHIFT;
+	unsigned long io_offset = 0;
+	int i, res_bit;
+
+	if (pci_ctrl == 0)
+		return -EINVAL;		/* should never happen */
+
+	/* If memory, add on the PCI bridge address offset */
+	if (mmap_state == pci_mmap_mem) {
+		res_bit = IORESOURCE_MEM;
+	} else {
+		io_offset = (unsigned long)pci_ctrl->io_space.base;
+		offset += io_offset;
+		res_bit = IORESOURCE_IO;
+	}
+
+	/*
+	 * Check that the offset requested corresponds to one of the
+	 * resources of the device.
+	 */
+	for (i = 0; i <= PCI_ROM_RESOURCE; i++) {
+		struct resource *rp = &dev->resource[i];
+		int flags = rp->flags;
+
+		/* treat ROM as memory (should be already) */
+		if (i == PCI_ROM_RESOURCE)
+			flags |= IORESOURCE_MEM;
+
+		/* Active and same type? */
+		if ((flags & res_bit) == 0)
+			continue;
+
+		/* In the range of this resource? */
+		if (offset < (rp->start & PAGE_MASK) || offset > rp->end)
+			continue;
+
+		/* found it! construct the final physical address */
+		if (mmap_state == pci_mmap_io)
+			offset += pci_ctrl->io_space.start - io_offset;
+		vma->vm_pgoff = offset >> PAGE_SHIFT;
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
+/*
+ * Set vm_flags of VMA, as appropriate for this architecture, for a pci device
+ * mapping.
+ */
+static __inline__ void
+__pci_mmap_set_flags(struct pci_dev *dev, struct vm_area_struct *vma,
+		     enum pci_mmap_state mmap_state)
+{
+	vma->vm_flags |= VM_SHM | VM_LOCKED | VM_IO;
+}
+
+/*
+ * Set vm_page_prot of VMA, as appropriate for this architecture, for a pci
+ * device mapping.
+ */
+static __inline__ void
+__pci_mmap_set_pgprot(struct pci_dev *dev, struct vm_area_struct *vma,
+		      enum pci_mmap_state mmap_state, int write_combine)
+{
+	int prot = pgprot_val(vma->vm_page_prot);
+
+	/* Set to write-through */
+	prot &= ~_PAGE_NO_CACHE;
+#if 0
+	if (!write_combine)
+		prot |= _PAGE_WRITETHRU;
+#endif
+	vma->vm_page_prot = __pgprot(prot);
+}
+
+/*
+ * Perform the actual remap of the pages for a PCI device mapping, as
+ * appropriate for this architecture.  The region in the process to map
+ * is described by vm_start and vm_end members of VMA, the base physical
+ * address is found in vm_pgoff.
+ * The pci device structure is provided so that architectures may make mapping
+ * decisions on a per-device or per-bus basis.
+ *
+ * Returns a negative error code on failure, zero on success.
+ */
+int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
+			enum pci_mmap_state mmap_state,
+			int write_combine)
+{
+	int ret;
+
+	ret = __pci_mmap_make_offset(dev, vma, mmap_state);
+	if (ret < 0)
+		return ret;
+
+	__pci_mmap_set_flags(dev, vma, mmap_state);
+	__pci_mmap_set_pgprot(dev, vma, mmap_state, write_combine);
+
+	ret = io_remap_page_range(vma, vma->vm_start, vma->vm_pgoff<<PAGE_SHIFT,
+			       vma->vm_end - vma->vm_start, vma->vm_page_prot);
+
+	return ret;
+}
+
+/*
+ * This probably belongs here rather than ioport.c because
+ * we do not want this crud linked into SBus kernels.
+ * Also, think for a moment about likes of floppy.c that
+ * include architecture specific parts. They may want to redefine ins/outs.
+ *
+ * We do not use horroble macroses here because we want to
+ * advance pointer by sizeof(size).
+ */
+void outsb(unsigned long addr, const void *src, unsigned long count) {
+        while (count) {
+                count -= 1;
+                writeb(*(const char *)src, addr);
+                src += 1;
+                addr += 1;
+        }
+}
+
+void outsw(unsigned long addr, const void *src, unsigned long count) {
+        while (count) {
+                count -= 2;
+                writew(*(const short *)src, addr);
+                src += 2;
+                addr += 2;
+        }
+}
+
+void outsl(unsigned long addr, const void *src, unsigned long count) {
+        while (count) {
+                count -= 4;
+                writel(*(const long *)src, addr);
+                src += 4;
+                addr += 4;
+        }
+}
+
+void insb(unsigned long addr, void *dst, unsigned long count) {
+        while (count) {
+                count -= 1;
+                *(unsigned char *)dst = readb(addr);
+                dst += 1;
+                addr += 1;
+        }
+}
+
+void insw(unsigned long addr, void *dst, unsigned long count) {
+        while (count) {
+                count -= 2;
+                *(unsigned short *)dst = readw(addr);
+                dst += 2;
+                addr += 2;
+        }
+}
+
+void insl(unsigned long addr, void *dst, unsigned long count) {
+        while (count) {
+                count -= 4;
+                /*
+                 * XXX I am sure we are in for an unaligned trap here.
+                 */
+                *(unsigned long *)dst = readl(addr);
+                dst += 4;
+                addr += 4;
+        }
+}
+
+
+
diff --git a/arch/xtensa/kernel/platform.c b/arch/xtensa/kernel/platform.c
new file mode 100644
index 00000000000..cf136278444
--- /dev/null
+++ b/arch/xtensa/kernel/platform.c
@@ -0,0 +1,49 @@
+/*
+ * arch/xtensa/kernel/platform.c
+ *
+ * Default platform 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) 2005 Tensilica Inc.
+ *
+ * Chris Zankel <chris@zankel.net>
+ */
+
+#include <linux/config.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include <linux/time.h>
+#include <asm/platform.h>
+#include <asm/timex.h>
+
+#define _F(r,f,a,b)							\
+	r __platform_##f a b;                                   	\
+	r platform_##f a __attribute__((weak, alias("__platform_"#f)))
+
+/*
+ * Default functions that are used if no platform specific function is defined.
+ * (Please, refer to include/asm-xtensa/platform.h for more information)
+ */
+
+_F(void, setup, (char** cmd), { });
+_F(void, init_irq, (void), { });
+_F(void, restart, (void), { while(1); });
+_F(void, halt, (void), { while(1); });
+_F(void, power_off, (void), { while(1); });
+_F(void, idle, (void), { __asm__ __volatile__ ("waiti 0" ::: "memory"); });
+_F(void, heartbeat, (void), { });
+_F(int,  pcibios_fixup, (void), { return 0; });
+_F(int, get_rtc_time, (time_t* t), { return 0; });
+_F(int, set_rtc_time, (time_t t), { return 0; });
+
+#if CONFIG_XTENSA_CALIBRATE_CCOUNT
+_F(void, calibrate_ccount, (void),
+{
+  printk ("ERROR: Cannot calibrate cpu frequency! Assuming 100MHz.\n");
+  ccount_per_jiffy = 100 * (1000000UL/HZ);
+});
+#endif
+
diff --git a/arch/xtensa/kernel/process.c b/arch/xtensa/kernel/process.c
new file mode 100644
index 00000000000..4099703b14b
--- /dev/null
+++ b/arch/xtensa/kernel/process.c
@@ -0,0 +1,482 @@
+// TODO	verify coprocessor handling
+/*
+ * arch/xtensa/kernel/process.c
+ *
+ * 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.
+ *
+ * Joe Taylor <joe@tensilica.com, joetylr@yahoo.com>
+ * Chris Zankel <chris@zankel.net>
+ * Marc Gauthier <marc@tensilica.com, marc@alumni.uwaterloo.ca>
+ * Kevin Chea
+ */
+
+#include <linux/config.h>
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/smp_lock.h>
+#include <linux/stddef.h>
+#include <linux/unistd.h>
+#include <linux/ptrace.h>
+#include <linux/slab.h>
+#include <linux/elf.h>
+#include <linux/init.h>
+#include <linux/prctl.h>
+#include <linux/init_task.h>
+#include <linux/module.h>
+#include <linux/mqueue.h>
+
+#include <asm/pgtable.h>
+#include <asm/uaccess.h>
+#include <asm/system.h>
+#include <asm/io.h>
+#include <asm/processor.h>
+#include <asm/platform.h>
+#include <asm/mmu.h>
+#include <asm/irq.h>
+#include <asm/atomic.h>
+#include <asm/offsets.h>
+#include <asm/coprocessor.h>
+
+extern void ret_from_fork(void);
+
+static struct fs_struct init_fs = INIT_FS;
+static struct files_struct init_files = INIT_FILES;
+static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
+static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
+struct mm_struct init_mm = INIT_MM(init_mm);
+EXPORT_SYMBOL(init_mm);
+
+union thread_union init_thread_union
+	__attribute__((__section__(".data.init_task"))) =
+{ INIT_THREAD_INFO(init_task) };
+
+struct task_struct init_task = INIT_TASK(init_task);
+EXPORT_SYMBOL(init_task);
+
+struct task_struct *current_set[NR_CPUS] = {&init_task, };
+
+
+#if XCHAL_CP_NUM > 0
+
+/*
+ * Coprocessor ownership.
+ */
+
+coprocessor_info_t coprocessor_info[] = {
+	{ 0, XTENSA_CPE_CP0_OFFSET },
+	{ 0, XTENSA_CPE_CP1_OFFSET },
+	{ 0, XTENSA_CPE_CP2_OFFSET },
+	{ 0, XTENSA_CPE_CP3_OFFSET },
+	{ 0, XTENSA_CPE_CP4_OFFSET },
+	{ 0, XTENSA_CPE_CP5_OFFSET },
+	{ 0, XTENSA_CPE_CP6_OFFSET },
+	{ 0, XTENSA_CPE_CP7_OFFSET },
+};
+
+#endif
+
+/*
+ * Powermanagement idle function, if any is provided by the platform.
+ */
+
+void cpu_idle(void)
+{
+  	local_irq_enable();
+
+	/* endless idle loop with no priority at all */
+	while (1) {
+		while (!need_resched())
+			platform_idle();
+		preempt_enable();
+		schedule();
+	}
+}
+
+/*
+ * Free current thread data structures etc..
+ */
+
+void exit_thread(void)
+{
+	release_coprocessors(current);	/* Empty macro if no CPs are defined */
+}
+
+void flush_thread(void)
+{
+	release_coprocessors(current);	/* Empty macro if no CPs are defined */
+}
+
+/*
+ * Copy thread.
+ *
+ * The stack layout for the new thread looks like this:
+ *
+ *	+------------------------+ <- sp in childregs (= tos)
+ *	|       childregs        |
+ *	+------------------------+ <- thread.sp = sp in dummy-frame
+ *	|      dummy-frame       |    (saved in dummy-frame spill-area)
+ *	+------------------------+
+ *
+ * We create a dummy frame to return to ret_from_fork:
+ *   a0 points to ret_from_fork (simulating a call4)
+ *   sp points to itself (thread.sp)
+ *   a2, a3 are unused.
+ *
+ * Note: This is a pristine frame, so we don't need any spill region on top of
+ *       childregs.
+ */
+
+int copy_thread(int nr, unsigned long clone_flags, unsigned long usp,
+		unsigned long unused,
+                struct task_struct * p, struct pt_regs * regs)
+{
+	struct pt_regs *childregs;
+	unsigned long tos;
+	int user_mode = user_mode(regs);
+
+	/* Set up new TSS. */
+	tos = (unsigned long)p->thread_info + THREAD_SIZE;
+	if (user_mode)
+		childregs = (struct pt_regs*)(tos - PT_USER_SIZE);
+	else
+		childregs = (struct pt_regs*)tos - 1;
+
+	*childregs = *regs;
+
+	/* Create a call4 dummy-frame: a0 = 0, a1 = childregs. */
+	*((int*)childregs - 3) = (unsigned long)childregs;
+	*((int*)childregs - 4) = 0;
+
+	childregs->areg[1] = tos;
+	childregs->areg[2] = 0;
+	p->set_child_tid = p->clear_child_tid = NULL;
+	p->thread.ra = MAKE_RA_FOR_CALL((unsigned long)ret_from_fork, 0x1);
+	p->thread.sp = (unsigned long)childregs;
+	if (user_mode(regs)) {
+
+		int len = childregs->wmask & ~0xf;
+		childregs->areg[1] = usp;
+		memcpy(&childregs->areg[XCHAL_NUM_AREGS - len/4],
+		       &regs->areg[XCHAL_NUM_AREGS - len/4], len);
+
+		if (clone_flags & CLONE_SETTLS)
+			childregs->areg[2] = childregs->areg[6];
+
+	} else {
+		/* In kernel space, we start a new thread with a new stack. */
+		childregs->wmask = 1;
+	}
+	return 0;
+}
+
+
+/*
+ * Create a kernel thread
+ */
+
+int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
+{
+	long retval;
+	__asm__ __volatile__
+		("mov           a5, %4\n\t" /* preserve fn in a5 */
+		 "mov           a6, %3\n\t" /* preserve and setup arg in a6 */
+		 "movi		a2, %1\n\t" /* load __NR_clone for syscall*/
+		 "mov		a3, sp\n\t" /* sp check and sys_clone */
+		 "mov		a4, %5\n\t" /* load flags for syscall */
+		 "syscall\n\t"
+		 "beq		a3, sp, 1f\n\t" /* branch if parent */
+		 "callx4	a5\n\t"     /* call fn */
+		 "movi		a2, %2\n\t" /* load __NR_exit for syscall */
+		 "mov		a3, a6\n\t" /* load fn return value */
+		 "syscall\n"
+		 "1:\n\t"
+		 "mov		%0, a2\n\t" /* parent returns zero */
+		 :"=r" (retval)
+		 :"i" (__NR_clone), "i" (__NR_exit),
+		 "r" (arg), "r" (fn),
+		 "r" (flags | CLONE_VM)
+		 : "a2", "a3", "a4", "a5", "a6" );
+	return retval;
+}
+
+
+/*
+ * These bracket the sleeping functions..
+ */
+
+unsigned long get_wchan(struct task_struct *p)
+{
+	unsigned long sp, pc;
+	unsigned long stack_page = (unsigned long) p->thread_info;
+	int count = 0;
+
+	if (!p || p == current || p->state == TASK_RUNNING)
+		return 0;
+
+	sp = p->thread.sp;
+	pc = MAKE_PC_FROM_RA(p->thread.ra, p->thread.sp);
+
+	do {
+		if (sp < stack_page + sizeof(struct task_struct) ||
+		    sp >= (stack_page + THREAD_SIZE) ||
+		    pc == 0)
+			return 0;
+		if (!in_sched_functions(pc))
+			return pc;
+
+		/* Stack layout: sp-4: ra, sp-3: sp' */
+
+		pc = MAKE_PC_FROM_RA(*(unsigned long*)sp - 4, sp);
+		sp = *(unsigned long *)sp - 3;
+	} while (count++ < 16);
+	return 0;
+}
+
+/*
+ * do_copy_regs() gathers information from 'struct pt_regs' and
+ * 'current->thread.areg[]' to fill in the xtensa_gregset_t
+ * structure.
+ *
+ * xtensa_gregset_t and 'struct pt_regs' are vastly different formats
+ * of processor registers.  Besides different ordering,
+ * xtensa_gregset_t contains non-live register information that
+ * 'struct pt_regs' does not.  Exception handling (primarily) uses
+ * 'struct pt_regs'.  Core files and ptrace use xtensa_gregset_t.
+ *
+ */
+
+void do_copy_regs (xtensa_gregset_t *elfregs, struct pt_regs *regs,
+		   struct task_struct *tsk)
+{
+	int i, n, wb_offset;
+
+	elfregs->xchal_config_id0 = XCHAL_HW_CONFIGID0;
+	elfregs->xchal_config_id1 = XCHAL_HW_CONFIGID1;
+
+	__asm__ __volatile__ ("rsr  %0, 176\n" : "=a" (i));
+ 	elfregs->cpux = i;
+	__asm__ __volatile__ ("rsr  %0, 208\n" : "=a" (i));
+ 	elfregs->cpuy = i;
+
+	/* Note:  PS.EXCM is not set while user task is running; its
+	 * being set in regs->ps is for exception handling convenience.
+	 */
+
+	elfregs->pc		= regs->pc;
+	elfregs->ps		= (regs->ps & ~XCHAL_PS_EXCM_MASK);
+	elfregs->exccause	= regs->exccause;
+	elfregs->excvaddr	= regs->excvaddr;
+	elfregs->windowbase	= regs->windowbase;
+	elfregs->windowstart	= regs->windowstart;
+	elfregs->lbeg		= regs->lbeg;
+	elfregs->lend		= regs->lend;
+	elfregs->lcount		= regs->lcount;
+	elfregs->sar		= regs->sar;
+	elfregs->syscall	= regs->syscall;
+
+	/* Copy register file.
+	 * The layout looks like this:
+	 *
+	 * |  a0 ... a15  | Z ... Z |  arX ... arY  |
+	 *  current window  unused    saved frames
+	 */
+
+	memset (elfregs->ar, 0, sizeof(elfregs->ar));
+
+	wb_offset = regs->windowbase * 4;
+	n = (regs->wmask&1)? 4 : (regs->wmask&2)? 8 : (regs->wmask&4)? 12 : 16;
+
+	for (i = 0; i < n; i++)
+		elfregs->ar[(wb_offset + i) % XCHAL_NUM_AREGS] = regs->areg[i];
+
+	n = (regs->wmask >> 4) * 4;
+
+	for (i = XCHAL_NUM_AREGS - n; n > 0; i++, n--)
+		elfregs->ar[(wb_offset + i) % XCHAL_NUM_AREGS] = regs->areg[i];
+}
+
+void xtensa_elf_core_copy_regs (xtensa_gregset_t *elfregs, struct pt_regs *regs)
+{
+	do_copy_regs ((xtensa_gregset_t *)elfregs, regs, current);
+}
+
+
+/* The inverse of do_copy_regs().  No error or sanity checking. */
+
+void do_restore_regs (xtensa_gregset_t *elfregs, struct pt_regs *regs,
+		      struct task_struct *tsk)
+{
+	int i, n, wb_offset;
+
+	/* Note:  PS.EXCM is not set while user task is running; it
+	 * needs to be set in regs->ps is for exception handling convenience.
+	 */
+
+	regs->pc		= elfregs->pc;
+	regs->ps		= (elfregs->ps | XCHAL_PS_EXCM_MASK);
+	regs->exccause		= elfregs->exccause;
+	regs->excvaddr		= elfregs->excvaddr;
+	regs->windowbase	= elfregs->windowbase;
+	regs->windowstart	= elfregs->windowstart;
+	regs->lbeg		= elfregs->lbeg;
+	regs->lend		= elfregs->lend;
+	regs->lcount		= elfregs->lcount;
+	regs->sar		= elfregs->sar;
+	regs->syscall	= elfregs->syscall;
+
+	/* Clear everything. */
+
+	memset (regs->areg, 0, sizeof(regs->areg));
+
+	/* Copy regs from live window frame. */
+
+	wb_offset = regs->windowbase * 4;
+	n = (regs->wmask&1)? 4 : (regs->wmask&2)? 8 : (regs->wmask&4)? 12 : 16;
+
+	for (i = 0; i < n; i++)
+		regs->areg[(wb_offset+i) % XCHAL_NUM_AREGS] = elfregs->ar[i];
+
+	n = (regs->wmask >> 4) * 4;
+
+	for (i = XCHAL_NUM_AREGS - n; n > 0; i++, n--)
+		regs->areg[(wb_offset+i) % XCHAL_NUM_AREGS] = elfregs->ar[i];
+}
+
+/*
+ * do_save_fpregs() gathers information from 'struct pt_regs' and
+ * 'current->thread' to fill in the elf_fpregset_t structure.
+ *
+ * Core files and ptrace use elf_fpregset_t.
+ */
+
+void do_save_fpregs (elf_fpregset_t *fpregs, struct pt_regs *regs,
+		     struct task_struct *tsk)
+{
+#if XCHAL_HAVE_CP
+
+	extern unsigned char	_xtensa_reginfo_tables[];
+	extern unsigned		_xtensa_reginfo_table_size;
+	int i;
+	unsigned long flags;
+
+	/* Before dumping coprocessor state from memory,
+	 * ensure any live coprocessor contents for this
+	 * task are first saved to memory:
+	 */
+	local_irq_save(flags);
+
+	for (i = 0; i < XCHAL_CP_MAX; i++) {
+		if (tsk == coprocessor_info[i].owner) {
+			enable_coprocessor(i);
+			save_coprocessor_registers(
+			    tsk->thread.cp_save+coprocessor_info[i].offset,i);
+			disable_coprocessor(i);
+		}
+	}
+
+	local_irq_restore(flags);
+
+	/* Now dump coprocessor & extra state: */
+	memcpy((unsigned char*)fpregs,
+		_xtensa_reginfo_tables, _xtensa_reginfo_table_size);
+	memcpy((unsigned char*)fpregs + _xtensa_reginfo_table_size,
+		tsk->thread.cp_save, XTENSA_CP_EXTRA_SIZE);
+#endif
+}
+
+/*
+ * The inverse of do_save_fpregs().
+ * Copies coprocessor and extra state from fpregs into regs and tsk->thread.
+ * Returns 0 on success, non-zero if layout doesn't match.
+ */
+
+int  do_restore_fpregs (elf_fpregset_t *fpregs, struct pt_regs *regs,
+		        struct task_struct *tsk)
+{
+#if XCHAL_HAVE_CP
+
+	extern unsigned char	_xtensa_reginfo_tables[];
+	extern unsigned		_xtensa_reginfo_table_size;
+	int i;
+	unsigned long flags;
+
+	/* Make sure save area layouts match.
+	 * FIXME:  in the future we could allow restoring from
+	 * a different layout of the same registers, by comparing
+	 * fpregs' table with _xtensa_reginfo_tables and matching
+	 * entries and copying registers one at a time.
+	 * Not too sure yet whether that's very useful.
+	 */
+
+	if( memcmp((unsigned char*)fpregs,
+		_xtensa_reginfo_tables, _xtensa_reginfo_table_size) ) {
+	    return -1;
+	}
+
+	/* Before restoring coprocessor state from memory,
+	 * ensure any live coprocessor contents for this
+	 * task are first invalidated.
+	 */
+
+	local_irq_save(flags);
+
+	for (i = 0; i < XCHAL_CP_MAX; i++) {
+		if (tsk == coprocessor_info[i].owner) {
+			enable_coprocessor(i);
+			save_coprocessor_registers(
+			    tsk->thread.cp_save+coprocessor_info[i].offset,i);
+			coprocessor_info[i].owner = 0;
+			disable_coprocessor(i);
+		}
+	}
+
+	local_irq_restore(flags);
+
+	/*  Now restore coprocessor & extra state:  */
+
+	memcpy(tsk->thread.cp_save,
+		(unsigned char*)fpregs + _xtensa_reginfo_table_size,
+		XTENSA_CP_EXTRA_SIZE);
+#endif
+	return 0;
+}
+/*
+ * Fill in the CP structure for a core dump for a particular task.
+ */
+
+int
+dump_task_fpu(struct pt_regs *regs, struct task_struct *task, elf_fpregset_t *r)
+{
+/* see asm/coprocessor.h for this magic number 16 */
+#if TOTAL_CPEXTRA_SIZE > 16
+	do_save_fpregs (r, regs, task);
+
+	/*  For now, bit 16 means some extra state may be present:  */
+// FIXME!! need to track to return more accurate mask
+	return 0x10000 | XCHAL_CP_MASK;
+#else
+	return 0;	/* no coprocessors active on this processor */
+#endif
+}
+
+/*
+ * Fill in the CP structure for a core dump.
+ * This includes any FPU coprocessor.
+ * Here, we dump all coprocessors, and other ("extra") custom state.
+ *
+ * This function is called by elf_core_dump() in fs/binfmt_elf.c
+ * (in which case 'regs' comes from calls to do_coredump, see signals.c).
+ */
+int  dump_fpu(struct pt_regs *regs, elf_fpregset_t *r)
+{
+	return dump_task_fpu(regs, current, r);
+}
diff --git a/arch/xtensa/kernel/ptrace.c b/arch/xtensa/kernel/ptrace.c
new file mode 100644
index 00000000000..9ef07a4dd2a
--- /dev/null
+++ b/arch/xtensa/kernel/ptrace.c
@@ -0,0 +1,407 @@
+// TODO some minor issues
+/*
+ * 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.
+ *
+ * Joe Taylor	<joe@tensilica.com, joetylr@yahoo.com>
+ * Chris Zankel <chris@zankel.net>
+ * Scott Foehner<sfoehner@yahoo.com>,
+ * Kevin Chea
+ * Marc Gauthier<marc@tensilica.com> <marc@alumni.uwaterloo.ca>
+ */
+
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/errno.h>
+#include <linux/ptrace.h>
+#include <linux/smp.h>
+#include <linux/smp_lock.h>
+#include <linux/security.h>
+
+#include <asm/pgtable.h>
+#include <asm/page.h>
+#include <asm/system.h>
+#include <asm/uaccess.h>
+#include <asm/ptrace.h>
+#include <asm/elf.h>
+
+#define TEST_KERNEL	// verify kernel operations FIXME: remove
+
+
+/*
+ * Called by kernel/ptrace.c when detaching..
+ *
+ * Make sure single step bits etc are not set.
+ */
+
+void ptrace_disable(struct task_struct *child)
+{
+	/* Nothing to do.. */
+}
+
+int sys_ptrace(long request, long pid, long addr, long data)
+{
+	struct task_struct *child;
+	int ret = -EPERM;
+
+	lock_kernel();
+
+#if 0
+	if ((int)request != 1)
+	printk("ptrace(r=%d,pid=%d,addr=%08lx,data=%08lx)\n",
+	       (int) request, (int) pid, (unsigned long) addr,
+	       (unsigned long) data);
+#endif
+
+	if (request == PTRACE_TRACEME) {
+
+		/* Are we already being traced? */
+
+		if (current->ptrace & PT_PTRACED)
+			goto out;
+
+		if ((ret = security_ptrace(current->parent, current)))
+			goto out;
+
+		/* Set the ptrace bit in the process flags. */
+
+		current->ptrace |= PT_PTRACED;
+		ret = 0;
+		goto out;
+	}
+
+	ret = -ESRCH;
+	read_lock(&tasklist_lock);
+	child = find_task_by_pid(pid);
+	if (child)
+		get_task_struct(child);
+	read_unlock(&tasklist_lock);
+	if (!child)
+		goto out;
+
+	ret = -EPERM;
+	if (pid == 1)		/* you may not mess with init */
+		goto out;
+
+	if (request == PTRACE_ATTACH) {
+		ret = ptrace_attach(child);
+		goto out_tsk;
+	}
+
+	if ((ret = ptrace_check_attach(child, request == PTRACE_KILL)) < 0)
+		goto out_tsk;
+
+	switch (request) {
+	case PTRACE_PEEKTEXT: /* read word at location addr. */
+	case PTRACE_PEEKDATA:
+	{
+		unsigned long tmp;
+		int copied;
+
+		copied = access_process_vm(child, addr, &tmp, sizeof(tmp), 0);
+		ret = -EIO;
+		if (copied != sizeof(tmp))
+			break;
+		ret = put_user(tmp,(unsigned long *) data);
+
+		goto out;
+	}
+
+	/* Read the word at location addr in the USER area.  */
+
+	case PTRACE_PEEKUSR:
+		{
+		struct pt_regs *regs;
+		unsigned long tmp;
+
+		regs = xtensa_pt_regs(child);
+		tmp = 0;  /* Default return value. */
+
+		switch(addr) {
+
+		case REG_AR_BASE ... REG_AR_BASE + XCHAL_NUM_AREGS - 1:
+			{
+			int ar = addr - REG_AR_BASE - regs->windowbase * 4;
+			ar &= (XCHAL_NUM_AREGS - 1);
+			if (ar < 16 && ar + (regs->wmask >> 4) * 4 >= 0)
+				tmp = regs->areg[ar];
+			else
+				ret = -EIO;
+			break;
+			}
+		case REG_A_BASE ... REG_A_BASE + 15:
+			tmp = regs->areg[addr - REG_A_BASE];
+			break;
+		case REG_PC:
+			tmp = regs->pc;
+			break;
+		case REG_PS:
+			/* Note:  PS.EXCM is not set while user task is running;
+			 * its being set in regs is for exception handling
+			 * convenience.  */
+			tmp = (regs->ps & ~XCHAL_PS_EXCM_MASK);
+			break;
+		case REG_WB:
+			tmp = regs->windowbase;
+			break;
+		case REG_WS:
+			tmp = regs->windowstart;
+			break;
+		case REG_LBEG:
+			tmp = regs->lbeg;
+			break;
+		case REG_LEND:
+			tmp = regs->lend;
+			break;
+		case REG_LCOUNT:
+			tmp = regs->lcount;
+			break;
+		case REG_SAR:
+			tmp = regs->sar;
+			break;
+		case REG_DEPC:
+			tmp = regs->depc;
+			break;
+		case REG_EXCCAUSE:
+			tmp = regs->exccause;
+			break;
+		case REG_EXCVADDR:
+			tmp = regs->excvaddr;
+			break;
+		case SYSCALL_NR:
+			tmp = regs->syscall;
+			break;
+		default:
+			tmp = 0;
+			ret = -EIO;
+			goto out;
+		}
+		ret = put_user(tmp, (unsigned long *) data);
+		goto out;
+		}
+
+	case PTRACE_POKETEXT: /* write the word at location addr. */
+	case PTRACE_POKEDATA:
+		if (access_process_vm(child, addr, &data, sizeof(data), 1)
+		    == sizeof(data))
+			break;
+		ret = -EIO;
+		goto out;
+
+	case PTRACE_POKEUSR:
+		{
+		struct pt_regs *regs;
+		regs = xtensa_pt_regs(child);
+
+		switch (addr) {
+		case REG_AR_BASE ... REG_AR_BASE + XCHAL_NUM_AREGS - 1:
+			{
+			int ar = addr - REG_AR_BASE - regs->windowbase * 4;
+			if (ar < 16 && ar + (regs->wmask >> 4) * 4 >= 0)
+				regs->areg[ar & (XCHAL_NUM_AREGS - 1)] = data;
+			else
+				ret = -EIO;
+			break;
+			}
+		case REG_A_BASE ... REG_A_BASE + 15:
+			regs->areg[addr - REG_A_BASE] = data;
+			break;
+		case REG_PC:
+			regs->pc = data;
+			break;
+		case SYSCALL_NR:
+			regs->syscall = data;
+			break;
+#ifdef TEST_KERNEL
+		case REG_WB:
+			regs->windowbase = data;
+			break;
+		case REG_WS:
+			regs->windowstart = data;
+			break;
+#endif
+
+		default:
+			/* The rest are not allowed. */
+			ret = -EIO;
+			break;
+		}
+		break;
+		}
+
+	/* continue and stop at next (return from) syscall */
+	case PTRACE_SYSCALL:
+	case PTRACE_CONT: /* restart after signal. */
+	{
+		ret = -EIO;
+		if ((unsigned long) data > _NSIG)
+			break;
+		if (request == PTRACE_SYSCALL)
+			set_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+		else
+			clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+		child->exit_code = data;
+		/* Make sure the single step bit is not set. */
+		child->ptrace &= ~PT_SINGLESTEP;
+		wake_up_process(child);
+		ret = 0;
+		break;
+	}
+
+	/*
+	 * make the child exit.  Best I can do is send it a sigkill.
+	 * perhaps it should be put in the status that it wants to
+	 * exit.
+	 */
+	case PTRACE_KILL:
+		ret = 0;
+		if (child->state == EXIT_ZOMBIE)	/* already dead */
+			break;
+		child->exit_code = SIGKILL;
+		child->ptrace &= ~PT_SINGLESTEP;
+		wake_up_process(child);
+		break;
+
+	case PTRACE_SINGLESTEP:
+		ret = -EIO;
+		if ((unsigned long) data > _NSIG)
+			break;
+		clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+		child->ptrace |= PT_SINGLESTEP;
+		child->exit_code = data;
+		wake_up_process(child);
+		ret = 0;
+		break;
+
+	case PTRACE_GETREGS:
+	{
+		/* 'data' points to user memory in which to write.
+		 * Mainly due to the non-live register values, we
+		 * reformat the register values into something more
+		 * standard.  For convenience, we use the handy
+		 * elf_gregset_t format. */
+
+		xtensa_gregset_t format;
+		struct pt_regs *regs = xtensa_pt_regs(child);
+
+		do_copy_regs (&format, regs, child);
+
+		/* Now, copy to user space nice and easy... */
+		ret = 0;
+		if (copy_to_user((void *)data, &format, sizeof(elf_gregset_t)))
+			ret = -EFAULT;
+		break;
+	}
+
+	case PTRACE_SETREGS:
+	{
+		/* 'data' points to user memory that contains the new
+		 * values in the elf_gregset_t format. */
+
+		xtensa_gregset_t format;
+		struct pt_regs *regs = xtensa_pt_regs(child);
+
+		if (copy_from_user(&format,(void *)data,sizeof(elf_gregset_t))){
+			ret = -EFAULT;
+			break;
+		}
+
+		/* FIXME: Perhaps we want some sanity checks on
+		 * these user-space values?  See ARM version.  Are
+		 * debuggers a security concern? */
+
+		do_restore_regs (&format, regs, child);
+
+		ret = 0;
+		break;
+	}
+
+	case PTRACE_GETFPREGS:
+	{
+		/* 'data' points to user memory in which to write.
+		 * For convenience, we use the handy
+		 * elf_fpregset_t format. */
+
+		elf_fpregset_t fpregs;
+		struct pt_regs *regs = xtensa_pt_regs(child);
+
+		do_save_fpregs (&fpregs, regs, child);
+
+		/* Now, copy to user space nice and easy... */
+		ret = 0;
+		if (copy_to_user((void *)data, &fpregs, sizeof(elf_fpregset_t)))
+			ret = -EFAULT;
+
+		break;
+	}
+
+	case PTRACE_SETFPREGS:
+	{
+		/* 'data' points to user memory that contains the new
+		 * values in the elf_fpregset_t format.
+		 */
+		elf_fpregset_t fpregs;
+		struct pt_regs *regs = xtensa_pt_regs(child);
+
+		ret = 0;
+		if (copy_from_user(&fpregs, (void *)data, sizeof(elf_fpregset_t))) {
+			ret = -EFAULT;
+			break;
+		}
+
+		if (do_restore_fpregs (&fpregs, regs, child))
+			ret = -EIO;
+		break;
+	}
+
+	case PTRACE_GETFPREGSIZE:
+		/* 'data' points to 'unsigned long' set to the size
+		 * of elf_fpregset_t
+		 */
+		ret = put_user(sizeof(elf_fpregset_t), (unsigned long *) data);
+		break;
+
+	case PTRACE_DETACH: /* detach a process that was attached. */
+		ret = ptrace_detach(child, data);
+		break;
+
+	default:
+		ret = ptrace_request(child, request, addr, data);
+		goto out;
+	}
+out_tsk:
+	put_task_struct(child);
+out:
+	unlock_kernel();
+	return ret;
+}
+
+void do_syscall_trace(void)
+{
+	if (!test_thread_flag(TIF_SYSCALL_TRACE))
+		return;
+
+	if (!(current->ptrace & PT_PTRACED))
+		return;
+
+	/*
+	 * The 0x80 provides a way for the tracing parent to distinguish
+	 * between a syscall stop and SIGTRAP delivery
+	 */
+	ptrace_notify(SIGTRAP|((current->ptrace & PT_TRACESYSGOOD) ? 0x80 : 0));
+
+	/*
+	 * this isn't the same as continuing with a signal, but it will do
+	 * for normal use.  strace only continues with a signal if the
+	 * stopping signal is not SIGTRAP.  -brl
+	 */
+	if (current->exit_code) {
+		send_sig(current->exit_code, current, 1);
+		current->exit_code = 0;
+	}
+}
diff --git a/arch/xtensa/kernel/semaphore.c b/arch/xtensa/kernel/semaphore.c
new file mode 100644
index 00000000000..d40f4b1b75a
--- /dev/null
+++ b/arch/xtensa/kernel/semaphore.c
@@ -0,0 +1,226 @@
+/*
+ * arch/xtensa/kernel/semaphore.c
+ *
+ * Generic semaphore code. Buyer beware. Do your own specific changes
+ * in <asm/semaphore-helper.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.
+ *
+ * Joe Taylor	<joe@tensilica.com, joetylr@yahoo.com>
+ * Chris Zankel	<chris@zankel.net>
+ * Marc Gauthier<marc@tensilica.com, marc@alumni.uwaterloo.ca>
+ * Kevin Chea
+ */
+
+#include <linux/sched.h>
+#include <linux/wait.h>
+#include <linux/init.h>
+#include <asm/semaphore.h>
+#include <asm/errno.h>
+
+/*
+ * These two _must_ execute atomically wrt each other.
+ */
+
+static __inline__ void wake_one_more(struct semaphore * sem)
+{
+	atomic_inc((atomic_t *)&sem->sleepers);
+}
+
+static __inline__ int waking_non_zero(struct semaphore *sem)
+{
+	unsigned long flags;
+	int ret = 0;
+
+	spin_lock_irqsave(&semaphore_wake_lock, flags);
+	if (sem->sleepers > 0) {
+		sem->sleepers--;
+		ret = 1;
+	}
+	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
+	return ret;
+}
+
+/*
+ * waking_non_zero_interruptible:
+ *	1	got the lock
+ *	0	go to sleep
+ *	-EINTR	interrupted
+ *
+ * We must undo the sem->count down_interruptible() increment while we are
+ * protected by the spinlock in order to make atomic this atomic_inc() with the
+ * atomic_read() in wake_one_more(), otherwise we can race. -arca
+ */
+
+static __inline__ int waking_non_zero_interruptible(struct semaphore *sem,
+						struct task_struct *tsk)
+{
+	unsigned long flags;
+	int ret = 0;
+
+	spin_lock_irqsave(&semaphore_wake_lock, flags);
+	if (sem->sleepers > 0) {
+		sem->sleepers--;
+		ret = 1;
+	} else if (signal_pending(tsk)) {
+		atomic_inc(&sem->count);
+		ret = -EINTR;
+	}
+	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
+	return ret;
+}
+
+/*
+ * waking_non_zero_trylock:
+ *	1	failed to lock
+ *	0	got the lock
+ *
+ * We must undo the sem->count down_trylock() increment while we are
+ * protected by the spinlock in order to make atomic this atomic_inc() with the
+ * atomic_read() in wake_one_more(), otherwise we can race. -arca
+ */
+
+static __inline__ int waking_non_zero_trylock(struct semaphore *sem)
+{
+	unsigned long flags;
+	int ret = 1;
+
+	spin_lock_irqsave(&semaphore_wake_lock, flags);
+	if (sem->sleepers <= 0)
+		atomic_inc(&sem->count);
+	else {
+		sem->sleepers--;
+		ret = 0;
+	}
+	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
+	return ret;
+}
+
+spinlock_t semaphore_wake_lock;
+
+/*
+ * Semaphores are implemented using a two-way counter:
+ * The "count" variable is decremented for each process
+ * that tries to sleep, while the "waking" variable is
+ * incremented when the "up()" code goes to wake up waiting
+ * processes.
+ *
+ * Notably, the inline "up()" and "down()" functions can
+ * efficiently test if they need to do any extra work (up
+ * needs to do something only if count was negative before
+ * the increment operation.
+ *
+ * waking_non_zero() (from asm/semaphore.h) must execute
+ * atomically.
+ *
+ * When __up() is called, the count was negative before
+ * incrementing it, and we need to wake up somebody.
+ *
+ * This routine adds one to the count of processes that need to
+ * wake up and exit.  ALL waiting processes actually wake up but
+ * only the one that gets to the "waking" field first will gate
+ * through and acquire the semaphore.  The others will go back
+ * to sleep.
+ *
+ * Note that these functions are only called when there is
+ * contention on the lock, and as such all this is the
+ * "non-critical" part of the whole semaphore business. The
+ * critical part is the inline stuff in <asm/semaphore.h>
+ * where we want to avoid any extra jumps and calls.
+ */
+
+void __up(struct semaphore *sem)
+{
+	wake_one_more(sem);
+	wake_up(&sem->wait);
+}
+
+/*
+ * Perform the "down" function.  Return zero for semaphore acquired,
+ * return negative for signalled out of the function.
+ *
+ * If called from __down, the return is ignored and the wait loop is
+ * not interruptible.  This means that a task waiting on a semaphore
+ * using "down()" cannot be killed until someone does an "up()" on
+ * the semaphore.
+ *
+ * If called from __down_interruptible, the return value gets checked
+ * upon return.  If the return value is negative then the task continues
+ * with the negative value in the return register (it can be tested by
+ * the caller).
+ *
+ * Either form may be used in conjunction with "up()".
+ *
+ */
+
+#define DOWN_VAR				\
+	struct task_struct *tsk = current;	\
+	wait_queue_t wait;			\
+	init_waitqueue_entry(&wait, tsk);
+
+#define DOWN_HEAD(task_state)						\
+									\
+									\
+	tsk->state = (task_state);					\
+	add_wait_queue(&sem->wait, &wait);				\
+									\
+	/*								\
+	 * Ok, we're set up.  sem->count is known to be less than zero	\
+	 * so we must wait.						\
+	 *								\
+	 * We can let go the lock for purposes of waiting.		\
+	 * We re-acquire it after awaking so as to protect		\
+	 * all semaphore operations.					\
+	 *								\
+	 * If "up()" is called before we call waking_non_zero() then	\
+	 * we will catch it right away.  If it is called later then	\
+	 * we will have to go through a wakeup cycle to catch it.	\
+	 *								\
+	 * Multiple waiters contend for the semaphore lock to see	\
+	 * who gets to gate through and who has to wait some more.	\
+	 */								\
+	for (;;) {
+
+#define DOWN_TAIL(task_state)			\
+		tsk->state = (task_state);	\
+	}					\
+	tsk->state = TASK_RUNNING;		\
+	remove_wait_queue(&sem->wait, &wait);
+
+void __sched __down(struct semaphore * sem)
+{
+	DOWN_VAR
+	DOWN_HEAD(TASK_UNINTERRUPTIBLE)
+	if (waking_non_zero(sem))
+		break;
+	schedule();
+	DOWN_TAIL(TASK_UNINTERRUPTIBLE)
+}
+
+int __sched __down_interruptible(struct semaphore * sem)
+{
+	int ret = 0;
+	DOWN_VAR
+	DOWN_HEAD(TASK_INTERRUPTIBLE)
+
+	ret = waking_non_zero_interruptible(sem, tsk);
+	if (ret)
+	{
+		if (ret == 1)
+			/* ret != 0 only if we get interrupted -arca */
+			ret = 0;
+		break;
+	}
+	schedule();
+	DOWN_TAIL(TASK_INTERRUPTIBLE)
+	return ret;
+}
+
+int __down_trylock(struct semaphore * sem)
+{
+	return waking_non_zero_trylock(sem);
+}
diff --git a/arch/xtensa/kernel/setup.c b/arch/xtensa/kernel/setup.c
new file mode 100644
index 00000000000..1f5bf5d624e
--- /dev/null
+++ b/arch/xtensa/kernel/setup.c
@@ -0,0 +1,520 @@
+/*
+ * arch/xtensa/setup.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) 1995  Linus Torvalds
+ * Copyright (C) 2001 - 2005  Tensilica Inc.
+ *
+ * Chris Zankel	<chris@zankel.net>
+ * Joe Taylor	<joe@tensilica.com, joetylr@yahoo.com>
+ * Kevin Chea
+ * Marc Gauthier<marc@tensilica.com> <marc@alumni.uwaterloo.ca>
+ */
+
+#include <linux/config.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/proc_fs.h>
+#include <linux/tty.h>
+#include <linux/bootmem.h>
+#include <linux/kernel.h>
+
+#if defined(CONFIG_VGA_CONSOLE) || defined(CONFIG_DUMMY_CONSOLE)
+# include <linux/console.h>
+#endif
+
+#ifdef CONFIG_RTC
+# include <linux/timex.h>
+#endif
+
+#ifdef CONFIG_PROC_FS
+# include <linux/seq_file.h>
+#endif
+
+#include <asm/system.h>
+#include <asm/bootparam.h>
+#include <asm/pgtable.h>
+#include <asm/processor.h>
+#include <asm/timex.h>
+#include <asm/platform.h>
+#include <asm/page.h>
+#include <asm/setup.h>
+
+#include <xtensa/config/system.h>
+
+#if defined(CONFIG_VGA_CONSOLE) || defined(CONFIG_DUMMY_CONSOLE)
+struct screen_info screen_info = { 0, 24, 0, 0, 0, 80, 0, 0, 0, 24, 1, 16};
+#endif
+
+#ifdef CONFIG_BLK_DEV_FD
+extern struct fd_ops no_fd_ops;
+struct fd_ops *fd_ops;
+#endif
+
+#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_IDE_MODULE)
+extern struct ide_ops no_ide_ops;
+struct ide_ops *ide_ops;
+#endif
+
+extern struct rtc_ops no_rtc_ops;
+struct rtc_ops *rtc_ops;
+
+#ifdef CONFIG_PC_KEYB
+extern struct kbd_ops no_kbd_ops;
+struct kbd_ops *kbd_ops;
+#endif
+
+#ifdef CONFIG_BLK_DEV_INITRD
+extern void *initrd_start;
+extern void *initrd_end;
+extern void *__initrd_start;
+extern void *__initrd_end;
+int initrd_is_mapped = 0;
+extern int initrd_below_start_ok;
+#endif
+
+unsigned char aux_device_present;
+extern unsigned long loops_per_jiffy;
+
+/* Command line specified as configuration option. */
+
+static char command_line[COMMAND_LINE_SIZE];
+
+#ifdef CONFIG_CMDLINE_BOOL
+static char default_command_line[COMMAND_LINE_SIZE] __initdata = CONFIG_CMDLINE;
+#endif
+
+sysmem_info_t __initdata sysmem;
+
+#ifdef CONFIG_BLK_DEV_INITRD
+int initrd_is_mapped;
+#endif
+
+extern void init_mmu(void);
+
+/*
+ * Boot parameter parsing.
+ *
+ * The Xtensa port uses a list of variable-sized tags to pass data to
+ * the kernel. The first tag must be a BP_TAG_FIRST tag for the list
+ * to be recognised. The list is terminated with a zero-sized
+ * BP_TAG_LAST tag.
+ */
+
+typedef struct tagtable {
+	u32 tag;
+	int (*parse)(const bp_tag_t*);
+} tagtable_t;
+
+#define __tagtable(tag, fn) static tagtable_t __tagtable_##fn 		\
+	__attribute__((unused, __section__(".taglist"))) = { tag, fn }
+
+/* parse current tag */
+
+static int __init parse_tag_mem(const bp_tag_t *tag)
+{
+	meminfo_t *mi = (meminfo_t*)(tag->data);
+
+	if (mi->type != MEMORY_TYPE_CONVENTIONAL)
+		return -1;
+
+	if (sysmem.nr_banks >= SYSMEM_BANKS_MAX) {
+		printk(KERN_WARNING
+		       "Ignoring memory bank 0x%08lx size %ldKB\n",
+		       (unsigned long)mi->start,
+		       (unsigned long)mi->end - (unsigned long)mi->start);
+		return -EINVAL;
+	}
+	sysmem.bank[sysmem.nr_banks].type  = mi->type;
+	sysmem.bank[sysmem.nr_banks].start = PAGE_ALIGN(mi->start);
+	sysmem.bank[sysmem.nr_banks].end   = mi->end & PAGE_SIZE;
+	sysmem.nr_banks++;
+
+	return 0;
+}
+
+__tagtable(BP_TAG_MEMORY, parse_tag_mem);
+
+#ifdef CONFIG_BLK_DEV_INITRD
+
+static int __init parse_tag_initrd(const bp_tag_t* tag)
+{
+	meminfo_t* mi;
+	mi = (meminfo_t*)(tag->data);
+	initrd_start = (void*)(mi->start);
+	initrd_end = (void*)(mi->end);
+
+	return 0;
+}
+
+__tagtable(BP_TAG_INITRD, parse_tag_initrd);
+
+#endif /* CONFIG_BLK_DEV_INITRD */
+
+static int __init parse_tag_cmdline(const bp_tag_t* tag)
+{
+	strncpy(command_line, (char*)(tag->data), COMMAND_LINE_SIZE);
+	command_line[COMMAND_LINE_SIZE - 1] = '\0';
+	return 0;
+}
+
+__tagtable(BP_TAG_COMMAND_LINE, parse_tag_cmdline);
+
+static int __init parse_bootparam(const bp_tag_t* tag)
+{
+	extern tagtable_t __tagtable_begin, __tagtable_end;
+	tagtable_t *t;
+
+	/* Boot parameters must start with a BP_TAG_FIRST tag. */
+
+	if (tag->id != BP_TAG_FIRST) {
+		printk(KERN_WARNING "Invalid boot parameters!\n");
+		return 0;
+	}
+
+	tag = (bp_tag_t*)((unsigned long)tag + sizeof(bp_tag_t) + tag->size);
+
+	/* Parse all tags. */
+
+	while (tag != NULL && tag->id != BP_TAG_LAST) {
+	 	for (t = &__tagtable_begin; t < &__tagtable_end; t++) {
+			if (tag->id == t->tag) {
+				t->parse(tag);
+				break;
+			}
+		}
+		if (t == &__tagtable_end)
+			printk(KERN_WARNING "Ignoring tag "
+			       "0x%08x\n", tag->id);
+		tag = (bp_tag_t*)((unsigned long)(tag + 1) + tag->size);
+	}
+
+	return 0;
+}
+
+/*
+ * Initialize architecture. (Early stage)
+ */
+
+void __init init_arch(bp_tag_t *bp_start)
+{
+
+#ifdef CONFIG_BLK_DEV_INITRD
+	initrd_start = &__initrd_start;
+	initrd_end = &__initrd_end;
+#endif
+
+	sysmem.nr_banks = 0;
+
+#ifdef CONFIG_CMDLINE_BOOL
+	strcpy(command_line, default_command_line);
+#endif
+
+	/* Parse boot parameters */
+
+        if (bp_start)
+	  parse_bootparam(bp_start);
+
+	if (sysmem.nr_banks == 0) {
+		sysmem.nr_banks = 1;
+		sysmem.bank[0].start = PLATFORM_DEFAULT_MEM_START;
+		sysmem.bank[0].end = PLATFORM_DEFAULT_MEM_START
+				     + PLATFORM_DEFAULT_MEM_SIZE;
+	}
+
+	/* Early hook for platforms */
+
+	platform_init(bp_start);
+
+	/* Initialize MMU. */
+
+	init_mmu();
+}
+
+/*
+ * Initialize system. Setup memory and reserve regions.
+ */
+
+extern char _end;
+extern char _stext;
+extern char _WindowVectors_text_start;
+extern char _WindowVectors_text_end;
+extern char _DebugInterruptVector_literal_start;
+extern char _DebugInterruptVector_text_end;
+extern char _KernelExceptionVector_literal_start;
+extern char _KernelExceptionVector_text_end;
+extern char _UserExceptionVector_literal_start;
+extern char _UserExceptionVector_text_end;
+extern char _DoubleExceptionVector_literal_start;
+extern char _DoubleExceptionVector_text_end;
+
+void __init setup_arch(char **cmdline_p)
+{
+	extern int mem_reserve(unsigned long, unsigned long, int);
+	extern void bootmem_init(void);
+
+	memcpy(saved_command_line, command_line, COMMAND_LINE_SIZE);
+	saved_command_line[COMMAND_LINE_SIZE-1] = '\0';
+	*cmdline_p = command_line;
+
+	/* Reserve some memory regions */
+
+#ifdef CONFIG_BLK_DEV_INITRD
+	if (initrd_start < initrd_end) {
+		initrd_is_mapped = mem_reserve(__pa(initrd_start),
+					       __pa(initrd_end), 0);
+		initrd_below_start_ok = 1;
+ 	} else {
+		initrd_start = 0;
+	}
+#endif
+
+	mem_reserve(__pa(&_stext),__pa(&_end), 1);
+
+	mem_reserve(__pa(&_WindowVectors_text_start),
+		    __pa(&_WindowVectors_text_end), 0);
+
+	mem_reserve(__pa(&_DebugInterruptVector_literal_start),
+		    __pa(&_DebugInterruptVector_text_end), 0);
+
+	mem_reserve(__pa(&_KernelExceptionVector_literal_start),
+		    __pa(&_KernelExceptionVector_text_end), 0);
+
+	mem_reserve(__pa(&_UserExceptionVector_literal_start),
+		    __pa(&_UserExceptionVector_text_end), 0);
+
+	mem_reserve(__pa(&_DoubleExceptionVector_literal_start),
+		    __pa(&_DoubleExceptionVector_text_end), 0);
+
+	bootmem_init();
+
+	platform_setup(cmdline_p);
+
+
+	paging_init();
+
+#ifdef CONFIG_VT
+# if defined(CONFIG_VGA_CONSOLE)
+	conswitchp = &vga_con;
+# elif defined(CONFIG_DUMMY_CONSOLE)
+	conswitchp = &dummy_con;
+# endif
+#endif
+
+#if CONFIG_PCI
+	platform_pcibios_init();
+#endif
+}
+
+void machine_restart(char * cmd)
+{
+	platform_restart();
+}
+
+void machine_halt(void)
+{
+	platform_halt();
+	while (1);
+}
+
+void machine_power_off(void)
+{
+	platform_power_off();
+	while (1);
+}
+#ifdef CONFIG_PROC_FS
+
+/*
+ * Display some core information through /proc/cpuinfo.
+ */
+
+static int
+c_show(struct seq_file *f, void *slot)
+{
+	/* high-level stuff */
+	seq_printf(f,"processor\t: 0\n"
+		     "vendor_id\t: Tensilica\n"
+		     "model\t\t: Xtensa " XCHAL_HW_RELEASE_NAME "\n"
+		     "core ID\t\t: " XCHAL_CORE_ID "\n"
+		     "build ID\t: 0x%x\n"
+		     "byte order\t: %s\n"
+ 		     "cpu MHz\t\t: %lu.%02lu\n"
+		     "bogomips\t: %lu.%02lu\n",
+		     XCHAL_BUILD_UNIQUE_ID,
+		     XCHAL_HAVE_BE ?  "big" : "little",
+		     CCOUNT_PER_JIFFY/(1000000/HZ),
+		     (CCOUNT_PER_JIFFY/(10000/HZ)) % 100,
+		     loops_per_jiffy/(500000/HZ),
+		     (loops_per_jiffy/(5000/HZ)) % 100);
+
+	seq_printf(f,"flags\t\t: "
+#if XCHAL_HAVE_NMI
+		     "nmi "
+#endif
+#if XCHAL_HAVE_DEBUG
+		     "debug "
+# if XCHAL_HAVE_OCD
+		     "ocd "
+# endif
+#endif
+#if XCHAL_HAVE_DENSITY
+	    	     "density "
+#endif
+#if XCHAL_HAVE_BOOLEANS
+		     "boolean "
+#endif
+#if XCHAL_HAVE_LOOPS
+		     "loop "
+#endif
+#if XCHAL_HAVE_NSA
+		     "nsa "
+#endif
+#if XCHAL_HAVE_MINMAX
+		     "minmax "
+#endif
+#if XCHAL_HAVE_SEXT
+		     "sext "
+#endif
+#if XCHAL_HAVE_CLAMPS
+		     "clamps "
+#endif
+#if XCHAL_HAVE_MAC16
+		     "mac16 "
+#endif
+#if XCHAL_HAVE_MUL16
+		     "mul16 "
+#endif
+#if XCHAL_HAVE_MUL32
+		     "mul32 "
+#endif
+#if XCHAL_HAVE_MUL32_HIGH
+		     "mul32h "
+#endif
+#if XCHAL_HAVE_FP
+		     "fpu "
+#endif
+		     "\n");
+
+	/* Registers. */
+	seq_printf(f,"physical aregs\t: %d\n"
+		     "misc regs\t: %d\n"
+		     "ibreak\t\t: %d\n"
+		     "dbreak\t\t: %d\n",
+		     XCHAL_NUM_AREGS,
+		     XCHAL_NUM_MISC_REGS,
+		     XCHAL_NUM_IBREAK,
+		     XCHAL_NUM_DBREAK);
+
+
+	/* Interrupt. */
+	seq_printf(f,"num ints\t: %d\n"
+		     "ext ints\t: %d\n"
+		     "int levels\t: %d\n"
+		     "timers\t\t: %d\n"
+		     "debug level\t: %d\n",
+		     XCHAL_NUM_INTERRUPTS,
+		     XCHAL_NUM_EXTINTERRUPTS,
+		     XCHAL_NUM_INTLEVELS,
+		     XCHAL_NUM_TIMERS,
+		     XCHAL_DEBUGLEVEL);
+
+	/* Coprocessors */
+#if XCHAL_HAVE_CP
+	seq_printf(f, "coprocessors\t: %d\n", XCHAL_CP_NUM);
+#else
+	seq_printf(f, "coprocessors\t: none\n");
+#endif
+
+	/* {I,D}{RAM,ROM} and XLMI */
+	seq_printf(f,"inst ROMs\t: %d\n"
+		     "inst RAMs\t: %d\n"
+		     "data ROMs\t: %d\n"
+		     "data RAMs\t: %d\n"
+		     "XLMI ports\t: %d\n",
+		     XCHAL_NUM_IROM,
+		     XCHAL_NUM_IRAM,
+		     XCHAL_NUM_DROM,
+		     XCHAL_NUM_DRAM,
+		     XCHAL_NUM_XLMI);
+
+	/* Cache */
+	seq_printf(f,"icache line size: %d\n"
+		     "icache ways\t: %d\n"
+		     "icache size\t: %d\n"
+		     "icache flags\t: "
+#if XCHAL_ICACHE_LINE_LOCKABLE
+		     "lock"
+#endif
+		     "\n"
+		     "dcache line size: %d\n"
+		     "dcache ways\t: %d\n"
+		     "dcache size\t: %d\n"
+		     "dcache flags\t: "
+#if XCHAL_DCACHE_IS_WRITEBACK
+		     "writeback"
+#endif
+#if XCHAL_DCACHE_LINE_LOCKABLE
+		     "lock"
+#endif
+		     "\n",
+		     XCHAL_ICACHE_LINESIZE,
+		     XCHAL_ICACHE_WAYS,
+		     XCHAL_ICACHE_SIZE,
+		     XCHAL_DCACHE_LINESIZE,
+		     XCHAL_DCACHE_WAYS,
+		     XCHAL_DCACHE_SIZE);
+
+	/* MMU */
+	seq_printf(f,"ASID bits\t: %d\n"
+		     "ASID invalid\t: %d\n"
+		     "ASID kernel\t: %d\n"
+		     "rings\t\t: %d\n"
+		     "itlb ways\t: %d\n"
+		     "itlb AR ways\t: %d\n"
+		     "dtlb ways\t: %d\n"
+		     "dtlb AR ways\t: %d\n",
+		     XCHAL_MMU_ASID_BITS,
+		     XCHAL_MMU_ASID_INVALID,
+		     XCHAL_MMU_ASID_KERNEL,
+		     XCHAL_MMU_RINGS,
+		     XCHAL_ITLB_WAYS,
+		     XCHAL_ITLB_ARF_WAYS,
+		     XCHAL_DTLB_WAYS,
+		     XCHAL_DTLB_ARF_WAYS);
+
+	return 0;
+}
+
+/*
+ * We show only CPU #0 info.
+ */
+static void *
+c_start(struct seq_file *f, loff_t *pos)
+{
+	return (void *) ((*pos == 0) ? (void *)1 : NULL);
+}
+
+static void *
+c_next(struct seq_file *f, void *v, loff_t *pos)
+{
+	return NULL;
+}
+
+static void
+c_stop(struct seq_file *f, void *v)
+{
+}
+
+struct seq_operations cpuinfo_op =
+{
+	start:  c_start,
+	next:   c_next,
+	stop:   c_stop,
+	show:   c_show
+};
+
+#endif /* CONFIG_PROC_FS */
+
diff --git a/arch/xtensa/kernel/signal.c b/arch/xtensa/kernel/signal.c
new file mode 100644
index 00000000000..df6e1e17b09
--- /dev/null
+++ b/arch/xtensa/kernel/signal.c
@@ -0,0 +1,713 @@
+// TODO coprocessor stuff
+/*
+ *  linux/arch/xtensa/kernel/signal.c
+ *
+ *  Copyright (C) 1991, 1992  Linus Torvalds
+ *  1997-11-28  Modified for POSIX.1b signals by Richard Henderson
+ *
+ *  Joe Taylor <joe@tensilica.com>
+ *  Chris Zankel <chris@zankel.net>
+ *
+ *
+ *
+ */
+
+#include <xtensa/config/core.h>
+#include <xtensa/hal.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/smp_lock.h>
+#include <linux/kernel.h>
+#include <linux/signal.h>
+#include <linux/errno.h>
+#include <linux/wait.h>
+#include <linux/ptrace.h>
+#include <linux/unistd.h>
+#include <linux/stddef.h>
+#include <linux/personality.h>
+#include <asm/ucontext.h>
+#include <asm/uaccess.h>
+#include <asm/pgtable.h>
+#include <asm/cacheflush.h>
+
+#define DEBUG_SIG  0
+
+#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
+
+asmlinkage long sys_wait4(pid_t pid,unsigned int * stat_addr, int options,
+			  struct rusage * ru);
+asmlinkage int do_signal(struct pt_regs *regs, sigset_t *oldset);
+
+extern struct task_struct *coproc_owners[];
+
+
+/*
+ * Atomically swap in the new signal mask, and wait for a signal.
+ */
+
+int sys_sigsuspend(struct pt_regs *regs)
+{
+	old_sigset_t mask = (old_sigset_t) regs->areg[3];
+	sigset_t saveset;
+
+	mask &= _BLOCKABLE;
+	spin_lock_irq(&current->sighand->siglock);
+	saveset = current->blocked;
+	siginitset(&current->blocked, mask);
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	regs->areg[2] = -EINTR;
+	while (1) {
+		current->state = TASK_INTERRUPTIBLE;
+		schedule();
+		if (do_signal(regs, &saveset))
+			return -EINTR;
+	}
+}
+
+asmlinkage int
+sys_rt_sigsuspend(struct pt_regs *regs)
+{
+	sigset_t *unewset = (sigset_t *) regs->areg[4];
+	size_t sigsetsize = (size_t) regs->areg[3];
+	sigset_t saveset, newset;
+	/* XXX: Don't preclude handling different sized sigset_t's.  */
+	if (sigsetsize != sizeof(sigset_t))
+		return -EINVAL;
+
+	if (copy_from_user(&newset, unewset, sizeof(newset)))
+		return -EFAULT;
+	sigdelsetmask(&newset, ~_BLOCKABLE);
+	spin_lock_irq(&current->sighand->siglock);
+	saveset = current->blocked;
+	current->blocked = newset;
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	regs->areg[2] = -EINTR;
+	while (1) {
+		current->state = TASK_INTERRUPTIBLE;
+		schedule();
+		if (do_signal(regs, &saveset))
+			return -EINTR;
+	}
+}
+
+asmlinkage int
+sys_sigaction(int sig, const struct old_sigaction *act,
+	      struct old_sigaction *oact)
+{
+	struct k_sigaction new_ka, old_ka;
+	int ret;
+
+	if (act) {
+		old_sigset_t mask;
+		if (verify_area(VERIFY_READ, act, sizeof(*act)) ||
+		    __get_user(new_ka.sa.sa_handler, &act->sa_handler) ||
+		    __get_user(new_ka.sa.sa_restorer, &act->sa_restorer))
+			return -EFAULT;
+		__get_user(new_ka.sa.sa_flags, &act->sa_flags);
+		__get_user(mask, &act->sa_mask);
+		siginitset(&new_ka.sa.sa_mask, mask);
+	}
+
+	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
+
+	if (!ret && oact) {
+		if (verify_area(VERIFY_WRITE, oact, sizeof(*oact)) ||
+		    __put_user(old_ka.sa.sa_handler, &oact->sa_handler) ||
+		    __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer))
+			return -EFAULT;
+		__put_user(old_ka.sa.sa_flags, &oact->sa_flags);
+		__put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask);
+	}
+
+	return ret;
+}
+
+asmlinkage int
+sys_sigaltstack(struct pt_regs *regs)
+{
+	const stack_t *uss = (stack_t *) regs->areg[4];
+	stack_t *uoss = (stack_t *) regs->areg[3];
+
+	if (regs->depc > 64)
+		panic ("Double exception sys_sigreturn\n");
+
+
+	return do_sigaltstack(uss, uoss, regs->areg[1]);
+}
+
+
+/*
+ * Do a signal return; undo the signal stack.
+ */
+
+struct sigframe
+{
+	struct sigcontext sc;
+	struct _cpstate cpstate;
+	unsigned long extramask[_NSIG_WORDS-1];
+	unsigned char retcode[6];
+	unsigned int reserved[4]; /* Reserved area for chaining */
+	unsigned int window[4]; /* Window of 4 registers for initial context */
+};
+
+struct rt_sigframe
+{
+	struct siginfo info;
+	struct ucontext uc;
+	struct _cpstate cpstate;
+	unsigned char retcode[6];
+	unsigned int reserved[4]; /* Reserved area for chaining */
+	unsigned int window[4]; /* Window of 4 registers for initial context */
+};
+
+extern void release_all_cp (struct task_struct *);
+
+
+// FIXME restore_cpextra
+static inline int
+restore_cpextra (struct _cpstate *buf)
+{
+#if 0
+	/* The signal handler may have used coprocessors in which
+	 * case they are still enabled.  We disable them to force a
+	 * reloading of the original task's CP state by the lazy
+	 * context-switching mechanisms of CP exception handling.
+	 * Also, we essentially discard any coprocessor state that the
+	 * signal handler created. */
+
+	struct task_struct *tsk = current;
+	release_all_cp(tsk);
+	return __copy_from_user(tsk->thread.cpextra, buf, TOTAL_CPEXTRA_SIZE);
+#endif
+	return 0;
+}
+
+/* Note: We don't copy double exception 'tregs', we have to finish double exc. first before we return to signal handler! This dbl.exc.handler might cause another double exception, but I think we are fine as the situation is the same as if we had returned to the signal handerl and got an interrupt immediately...
+ */
+
+
+static int
+restore_sigcontext(struct pt_regs *regs, struct sigcontext *sc)
+{
+	struct thread_struct *thread;
+	unsigned int err = 0;
+	unsigned long ps;
+	struct _cpstate *buf;
+
+#define COPY(x)	err |= __get_user(regs->x, &sc->sc_##x)
+	COPY(pc);
+	COPY(depc);
+	COPY(wmask);
+	COPY(lbeg);
+	COPY(lend);
+	COPY(lcount);
+	COPY(sar);
+	COPY(windowbase);
+	COPY(windowstart);
+#undef COPY
+
+	/* For PS, restore only PS.CALLINC.
+	 * Assume that all other bits are either the same as for the signal
+	 * handler, or the user mode value doesn't matter (e.g. PS.OWB).
+	 */
+	err |= __get_user(ps, &sc->sc_ps);
+	regs->ps = (regs->ps & ~XCHAL_PS_CALLINC_MASK)
+		| (ps & XCHAL_PS_CALLINC_MASK);
+
+	/* Additional corruption checks */
+
+	if ((regs->windowbase >= (XCHAL_NUM_AREGS/4))
+	|| ((regs->windowstart & ~((1<<(XCHAL_NUM_AREGS/4)) - 1)) != 0) )
+		err = 1;
+	if ((regs->lcount > 0)
+	&& ((regs->lbeg > TASK_SIZE) || (regs->lend > TASK_SIZE)) )
+		err = 1;
+
+	/* Restore extended register state.
+	 * See struct thread_struct in processor.h.
+	 */
+	thread = &current->thread;
+
+	err |= __copy_from_user (regs->areg, sc->sc_areg, XCHAL_NUM_AREGS*4);
+	err |= __get_user(buf, &sc->sc_cpstate);
+	if (buf) {
+		if (verify_area(VERIFY_READ, buf, sizeof(*buf)))
+			goto badframe;
+		err |= restore_cpextra(buf);
+	}
+
+	regs->syscall = -1;		/* disable syscall checks */
+	return err;
+
+badframe:
+	return 1;
+}
+
+static inline void
+flush_my_cpstate(struct task_struct *tsk)
+{
+	unsigned long flags;
+	local_irq_save(flags);
+
+#if 0	// FIXME
+	for (i = 0; i < XCHAL_CP_NUM; i++) {
+		if (tsk == coproc_owners[i]) {
+			xthal_validate_cp(i);
+			xthal_save_cpregs(tsk->thread.cpregs_ptr[i], i);
+
+			/* Invalidate and "disown" the cp to allow
+			 * callers the chance to reset cp state in the
+			 * task_struct. */
+
+			xthal_invalidate_cp(i);
+			coproc_owners[i] = 0;
+		}
+	}
+#endif
+	local_irq_restore(flags);
+}
+
+/* Return codes:
+	0:  nothing saved
+	1:  stuff to save, successful
+       -1:  stuff to save, error happened
+*/
+static int
+save_cpextra (struct _cpstate *buf)
+{
+#if (XCHAL_EXTRA_SA_SIZE == 0) && (XCHAL_CP_NUM == 0)
+	return 0;
+#else
+
+	/* FIXME: If a task has never used a coprocessor, there is
+	 * no need to save and restore anything.  Tracking this
+	 * information would allow us to optimize this section.
+	 * Perhaps we can use current->used_math or (current->flags &
+	 * PF_USEDFPU) or define a new field in the thread
+	 * structure. */
+
+	/* We flush any live, task-owned cp state to the task_struct,
+	 * then copy it all to the sigframe.  Then we clear all
+	 * cp/extra state in the task_struct, effectively
+	 * clearing/resetting all cp/extra state for the signal
+	 * handler (cp-exception handling will load these new values
+	 * into the cp/extra registers.)  This step is important for
+	 * things like a floating-point cp, where the OS must reset
+	 * the FCR to the default rounding mode. */
+
+	int err = 0;
+	struct task_struct *tsk = current;
+
+	flush_my_cpstate(tsk);
+	/* Note that we just copy everything: 'extra' and 'cp' state together.*/
+	err |= __copy_to_user(buf, tsk->thread.cp_save, XTENSA_CP_EXTRA_SIZE);
+	memset(tsk->thread.cp_save, 0, XTENSA_CP_EXTRA_SIZE);
+
+#if (XTENSA_CP_EXTRA_SIZE == 0)
+#error Sanity check on memset above, cpextra_size should not be zero.
+#endif
+
+	return err ? -1 : 1;
+#endif
+}
+
+static int
+setup_sigcontext(struct sigcontext *sc, struct _cpstate *cpstate,
+		 struct pt_regs *regs, unsigned long mask)
+{
+	struct thread_struct *thread;
+	int err = 0;
+
+//printk("setup_sigcontext\n");
+#define COPY(x)	err |= __put_user(regs->x, &sc->sc_##x)
+	COPY(pc);
+	COPY(ps);
+	COPY(depc);
+	COPY(wmask);
+	COPY(lbeg);
+	COPY(lend);
+	COPY(lcount);
+	COPY(sar);
+	COPY(windowbase);
+	COPY(windowstart);
+#undef COPY
+
+	/* Save extended register state.
+	 * See struct thread_struct in processor.h.
+	 */
+	thread = &current->thread;
+	err |= __copy_to_user (sc->sc_areg, regs->areg, XCHAL_NUM_AREGS * 4);
+	err |= save_cpextra(cpstate);
+	err |= __put_user(err ? NULL : cpstate, &sc->sc_cpstate);
+	/* non-iBCS2 extensions.. */
+	err |= __put_user(mask, &sc->oldmask);
+
+	return err;
+}
+
+asmlinkage int sys_sigreturn(struct pt_regs *regs)
+{
+	struct sigframe *frame = (struct sigframe *)regs->areg[1];
+	sigset_t set;
+	if (regs->depc > 64)
+		panic ("Double exception sys_sigreturn\n");
+
+	if (verify_area(VERIFY_READ, frame, sizeof(*frame)))
+		goto badframe;
+
+	if (__get_user(set.sig[0], &frame->sc.oldmask)
+	    || (_NSIG_WORDS > 1
+		&& __copy_from_user(&set.sig[1], &frame->extramask,
+				    sizeof(frame->extramask))))
+		goto badframe;
+
+	sigdelsetmask(&set, ~_BLOCKABLE);
+
+	spin_lock_irq(&current->sighand->siglock);
+	current->blocked = set;
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	if (restore_sigcontext(regs, &frame->sc))
+		goto badframe;
+	return regs->areg[2];
+
+badframe:
+	force_sig(SIGSEGV, current);
+	return 0;
+}
+
+asmlinkage int sys_rt_sigreturn(struct pt_regs *regs)
+{
+	struct rt_sigframe *frame = (struct rt_sigframe *)regs->areg[1];
+	sigset_t set;
+	stack_t st;
+	int ret;
+	if (regs->depc > 64)
+	{
+		printk("!!!!!!! DEPC !!!!!!!\n");
+		return 0;
+	}
+
+	if (verify_area(VERIFY_READ, frame, sizeof(*frame)))
+		goto badframe;
+
+	if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set)))
+		goto badframe;
+
+	sigdelsetmask(&set, ~_BLOCKABLE);
+	spin_lock_irq(&current->sighand->siglock);
+	current->blocked = set;
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	if (restore_sigcontext(regs, &frame->uc.uc_mcontext))
+		goto badframe;
+	ret = regs->areg[2];
+
+	if (__copy_from_user(&st, &frame->uc.uc_stack, sizeof(st)))
+		goto badframe;
+	/* It is more difficult to avoid calling this function than to
+	   call it and ignore errors.  */
+	do_sigaltstack(&st, NULL, regs->areg[1]);
+
+	return ret;
+
+badframe:
+	force_sig(SIGSEGV, current);
+	return 0;
+}
+
+/*
+ * Set up a signal frame.
+ */
+
+/*
+ * Determine which stack to use..
+ */
+static inline void *
+get_sigframe(struct k_sigaction *ka, unsigned long sp, size_t frame_size)
+{
+	if ((ka->sa.sa_flags & SA_ONSTACK) != 0 && ! on_sig_stack(sp))
+		sp = current->sas_ss_sp + current->sas_ss_size;
+
+	return (void *)((sp - frame_size) & -16ul);
+}
+
+#define USE_SIGRETURN		0
+#define USE_RT_SIGRETURN	1
+
+static int
+gen_return_code(unsigned char *codemem, unsigned int use_rt_sigreturn)
+{
+	unsigned int retcall;
+	int err = 0;
+
+#if 0
+	/* Ignoring SA_RESTORER for now; it's supposed to be obsolete,
+	 * and the xtensa glibc doesn't use it.
+	 */
+	if (ka->sa.sa_flags & SA_RESTORER) {
+		regs->pr = (unsigned long) ka->sa.sa_restorer;
+	} else
+#endif /* 0 */
+	{
+
+#if (__NR_sigreturn > 255) || (__NR_rt_sigreturn > 255)
+
+/* The 12-bit immediate is really split up within the 24-bit MOVI
+ * instruction.  As long as the above system call numbers fit within
+ * 8-bits, the following code works fine. See the Xtensa ISA for
+ * details.
+ */
+
+#error Generating the MOVI instruction below breaks!
+#endif
+
+		retcall = use_rt_sigreturn ? __NR_rt_sigreturn : __NR_sigreturn;
+
+#ifdef __XTENSA_EB__   /* Big Endian version */
+		/* Generate instruction:  MOVI a2, retcall */
+		err |= __put_user(0x22, &codemem[0]);
+		err |= __put_user(0x0a, &codemem[1]);
+		err |= __put_user(retcall, &codemem[2]);
+		/* Generate instruction:  SYSCALL */
+		err |= __put_user(0x00, &codemem[3]);
+		err |= __put_user(0x05, &codemem[4]);
+		err |= __put_user(0x00, &codemem[5]);
+
+#elif defined __XTENSA_EL__   /* Little Endian version */
+		/* Generate instruction:  MOVI a2, retcall */
+		err |= __put_user(0x22, &codemem[0]);
+		err |= __put_user(0xa0, &codemem[1]);
+		err |= __put_user(retcall, &codemem[2]);
+		/* Generate instruction:  SYSCALL */
+		err |= __put_user(0x00, &codemem[3]);
+		err |= __put_user(0x50, &codemem[4]);
+		err |= __put_user(0x00, &codemem[5]);
+#else
+#error Must use compiler for Xtensa processors.
+#endif
+	}
+
+	/* Flush generated code out of the data cache */
+
+	if (err == 0)
+		__flush_invalidate_cache_range((unsigned long)codemem, 6UL);
+
+	return err;
+}
+
+static void
+set_thread_state(struct pt_regs *regs, void *stack, unsigned char *retaddr,
+	void *handler, unsigned long arg1, void *arg2, void *arg3)
+{
+	/* Set up registers for signal handler */
+	start_thread(regs, (unsigned long) handler, (unsigned long) stack);
+
+	/* Set up a stack frame for a call4
+	 * Note: PS.CALLINC is set to one by start_thread
+	 */
+	regs->areg[4] = (((unsigned long) retaddr) & 0x3fffffff) | 0x40000000;
+	regs->areg[6] = arg1;
+	regs->areg[7] = (unsigned long) arg2;
+	regs->areg[8] = (unsigned long) arg3;
+}
+
+static void setup_frame(int sig, struct k_sigaction *ka,
+			sigset_t *set, struct pt_regs *regs)
+{
+	struct sigframe *frame;
+	int err = 0;
+	int signal;
+
+	frame = get_sigframe(ka, regs->areg[1], sizeof(*frame));
+	if (regs->depc > 64)
+	{
+		printk("!!!!!!! DEPC !!!!!!!\n");
+		return;
+	}
+
+
+	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
+		goto give_sigsegv;
+
+	signal = current_thread_info()->exec_domain
+		&& current_thread_info()->exec_domain->signal_invmap
+		&& sig < 32
+		? current_thread_info()->exec_domain->signal_invmap[sig]
+		: sig;
+
+	err |= setup_sigcontext(&frame->sc, &frame->cpstate, regs, set->sig[0]);
+
+	if (_NSIG_WORDS > 1) {
+		err |= __copy_to_user(frame->extramask, &set->sig[1],
+				      sizeof(frame->extramask));
+	}
+
+	/* Create sys_sigreturn syscall in stack frame */
+	err |= gen_return_code(frame->retcode, USE_SIGRETURN);
+
+	if (err)
+		goto give_sigsegv;
+
+	/* Create signal handler execution context.
+	 * Return context not modified until this point.
+	 */
+	set_thread_state(regs, frame, frame->retcode,
+		ka->sa.sa_handler, signal, &frame->sc, NULL);
+
+	/* Set access mode to USER_DS.  Nomenclature is outdated, but
+	 * functionality is used in uaccess.h
+	 */
+	set_fs(USER_DS);
+
+
+#if DEBUG_SIG
+	printk("SIG deliver (%s:%d): signal=%d sp=%p pc=%08x\n",
+		current->comm, current->pid, signal, frame, regs->pc);
+#endif
+
+	return;
+
+give_sigsegv:
+	if (sig == SIGSEGV)
+		ka->sa.sa_handler = SIG_DFL;
+	force_sig(SIGSEGV, current);
+}
+
+static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+			   sigset_t *set, struct pt_regs *regs)
+{
+	struct rt_sigframe *frame;
+	int err = 0;
+	int signal;
+
+	frame = get_sigframe(ka, regs->areg[1], sizeof(*frame));
+	if (regs->depc > 64)
+		panic ("Double exception sys_sigreturn\n");
+
+	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
+		goto give_sigsegv;
+
+	signal = current_thread_info()->exec_domain
+		&& current_thread_info()->exec_domain->signal_invmap
+		&& sig < 32
+		? current_thread_info()->exec_domain->signal_invmap[sig]
+		: sig;
+
+	err |= copy_siginfo_to_user(&frame->info, info);
+
+	/* Create the ucontext.  */
+	err |= __put_user(0, &frame->uc.uc_flags);
+	err |= __put_user(0, &frame->uc.uc_link);
+	err |= __put_user((void *)current->sas_ss_sp,
+			  &frame->uc.uc_stack.ss_sp);
+	err |= __put_user(sas_ss_flags(regs->areg[1]),
+			  &frame->uc.uc_stack.ss_flags);
+	err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size);
+	err |= setup_sigcontext(&frame->uc.uc_mcontext, &frame->cpstate,
+			        regs, set->sig[0]);
+	err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
+
+	/* Create sys_rt_sigreturn syscall in stack frame */
+	err |= gen_return_code(frame->retcode, USE_RT_SIGRETURN);
+
+	if (err)
+		goto give_sigsegv;
+
+	/* Create signal handler execution context.
+	 * Return context not modified until this point.
+	 */
+	set_thread_state(regs, frame, frame->retcode,
+		ka->sa.sa_handler, signal, &frame->info, &frame->uc);
+
+	/* Set access mode to USER_DS.  Nomenclature is outdated, but
+	 * functionality is used in uaccess.h
+	 */
+	set_fs(USER_DS);
+
+#if DEBUG_SIG
+	printk("SIG rt deliver (%s:%d): signal=%d sp=%p pc=%08x\n",
+		current->comm, current->pid, signal, frame, regs->pc);
+#endif
+
+	return;
+
+give_sigsegv:
+	if (sig == SIGSEGV)
+		ka->sa.sa_handler = SIG_DFL;
+	force_sig(SIGSEGV, current);
+}
+
+
+
+/*
+ * Note that 'init' is a special process: it doesn't get signals it doesn't
+ * want to handle. Thus you cannot kill init even with a SIGKILL even by
+ * mistake.
+ *
+ * Note that we go through the signals twice: once to check the signals that
+ * the kernel can handle, and then we build all the user-level signal handling
+ * stack-frames in one go after that.
+ */
+int do_signal(struct pt_regs *regs, sigset_t *oldset)
+{
+	siginfo_t info;
+	int signr;
+	struct k_sigaction ka;
+
+	if (!oldset)
+		oldset = &current->blocked;
+
+	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
+
+	/* Are we from a system call? */
+	if (regs->syscall >= 0) {
+		/* If so, check system call restarting.. */
+		switch (regs->areg[2]) {
+			case ERESTARTNOHAND:
+			case ERESTART_RESTARTBLOCK:
+				regs->areg[2] = -EINTR;
+				break;
+
+			case ERESTARTSYS:
+				if (!(ka.sa.sa_flags & SA_RESTART)) {
+					regs->areg[2] = -EINTR;
+					break;
+				}
+			/* fallthrough */
+			case ERESTARTNOINTR:
+				regs->areg[2] = regs->syscall;
+				regs->pc -= 3;
+		}
+	}
+
+	if (signr == 0)
+		return 0;		/* no signals delivered */
+
+	/* Whee!  Actually deliver the signal.  */
+
+	/* Set up the stack frame */
+	if (ka.sa.sa_flags & SA_SIGINFO)
+		setup_rt_frame(signr, &ka, &info, oldset, regs);
+	else
+		setup_frame(signr, &ka, oldset, regs);
+
+	if (ka.sa.sa_flags & SA_ONESHOT)
+		ka.sa.sa_handler = SIG_DFL;
+
+	if (!(ka.sa.sa_flags & SA_NODEFER)) {
+		spin_lock_irq(&current->sighand->siglock);
+		sigorsets(&current->blocked, &current->blocked, &ka.sa.sa_mask);
+		sigaddset(&current->blocked, signr);
+		recalc_sigpending();
+		spin_unlock_irq(&current->sighand->siglock);
+	}
+	return 1;
+}
diff --git a/arch/xtensa/kernel/syscalls.c b/arch/xtensa/kernel/syscalls.c
new file mode 100644
index 00000000000..abc8ed6c702
--- /dev/null
+++ b/arch/xtensa/kernel/syscalls.c
@@ -0,0 +1,418 @@
+/*
+ * arch/xtensa/kernel/syscall.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.
+ * Copyright (C) 2000 Silicon Graphics, Inc.
+ * Copyright (C) 1995 - 2000 by Ralf Baechle
+ *
+ * Joe Taylor <joe@tensilica.com, joetylr@yahoo.com>
+ * Marc Gauthier <marc@tensilica.com, marc@alumni.uwaterloo.ca>
+ * Chris Zankel <chris@zankel.net>
+ * Kevin Chea
+ *
+ */
+
+#define DEBUG	0
+
+#include <linux/config.h>
+#include <linux/linkage.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/smp_lock.h>
+#include <linux/mman.h>
+#include <linux/sched.h>
+#include <linux/file.h>
+#include <linux/slab.h>
+#include <linux/utsname.h>
+#include <linux/unistd.h>
+#include <linux/stringify.h>
+#include <linux/syscalls.h>
+#include <linux/sem.h>
+#include <linux/msg.h>
+#include <linux/shm.h>
+#include <linux/errno.h>
+#include <asm/ptrace.h>
+#include <asm/signal.h>
+#include <asm/uaccess.h>
+#include <asm/hardirq.h>
+#include <asm/mman.h>
+#include <asm/shmparam.h>
+#include <asm/page.h>
+#include <asm/ipc.h>
+
+extern void do_syscall_trace(void);
+typedef int (*syscall_t)(void *a0,...);
+extern int (*do_syscalls)(struct pt_regs *regs, syscall_t fun,
+				     int narg);
+extern syscall_t sys_call_table[];
+extern unsigned char sys_narg_table[];
+
+/*
+ * sys_pipe() is the normal C calling standard for creating a pipe. It's not
+ * the way unix traditional does this, though.
+ */
+
+int sys_pipe(int __user *userfds)
+{
+	int fd[2];
+	int error;
+
+	error = do_pipe(fd);
+	if (!error) {
+		if (copy_to_user(userfds, fd, 2 * sizeof(int)))
+			error = -EFAULT;
+	}
+	return error;
+}
+
+/*
+ * Common code for old and new mmaps.
+ */
+
+static inline long do_mmap2(unsigned long addr, unsigned long len,
+    			    unsigned long prot, unsigned long flags,
+			    unsigned long fd, unsigned long pgoff)
+{
+	int error = -EBADF;
+	struct file * file = NULL;
+
+	flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE);
+	if (!(flags & MAP_ANONYMOUS)) {
+		file = fget(fd);
+		if (!file)
+			goto out;
+	}
+
+	down_write(&current->mm->mmap_sem);
+	error = do_mmap_pgoff(file, addr, len, prot, flags, pgoff);
+	up_write(&current->mm->mmap_sem);
+
+	if (file)
+		fput(file);
+out:
+	return error;
+}
+
+unsigned long old_mmap(unsigned long addr, size_t len, int prot,
+		       int flags, int fd, off_t offset)
+{
+	return do_mmap2(addr, len, prot, flags, fd, offset >> PAGE_SHIFT);
+}
+
+long sys_mmap2(unsigned long addr, unsigned long len, unsigned long prot,
+	       unsigned long flags, unsigned long fd, unsigned long pgoff)
+{
+	return do_mmap2(addr, len, prot, flags, fd, pgoff);
+}
+
+int sys_fork(struct pt_regs *regs)
+{
+	return do_fork(SIGCHLD, regs->areg[1], regs, 0, NULL, NULL);
+}
+
+int sys_vfork(struct pt_regs *regs)
+{
+	return do_fork(CLONE_VFORK|CLONE_VM|SIGCHLD, regs->areg[1],
+		       regs, 0, NULL, NULL);
+}
+
+int sys_clone(struct pt_regs *regs)
+{
+	unsigned long clone_flags;
+	unsigned long newsp;
+	int __user *parent_tidptr, *child_tidptr;
+	clone_flags = regs->areg[4];
+	newsp = regs->areg[3];
+	parent_tidptr = (int __user *)regs->areg[5];
+	child_tidptr = (int __user *)regs->areg[6];
+	if (!newsp)
+		newsp = regs->areg[1];
+	return do_fork(clone_flags,newsp,regs,0,parent_tidptr,child_tidptr);
+}
+
+/*
+ * sys_execve() executes a new program.
+ */
+
+int sys_execve(struct pt_regs *regs)
+{
+	int error;
+	char * filename;
+
+	filename = getname((char *) (long)regs->areg[5]);
+	error = PTR_ERR(filename);
+	if (IS_ERR(filename))
+		goto out;
+	error = do_execve(filename, (char **) (long)regs->areg[3],
+	                  (char **) (long)regs->areg[4], regs);
+	putname(filename);
+
+out:
+	return error;
+}
+
+int sys_uname(struct old_utsname * name)
+{
+	if (name && !copy_to_user(name, &system_utsname, sizeof (*name)))
+		return 0;
+	return -EFAULT;
+}
+
+int sys_olduname(struct oldold_utsname * name)
+{
+	int error;
+
+	if (!name)
+		return -EFAULT;
+	if (!access_ok(VERIFY_WRITE,name,sizeof(struct oldold_utsname)))
+		return -EFAULT;
+
+	error = __copy_to_user(&name->sysname,&system_utsname.sysname,__OLD_UTS_LEN);
+	error -= __put_user(0,name->sysname+__OLD_UTS_LEN);
+	error -= __copy_to_user(&name->nodename,&system_utsname.nodename,__OLD_UTS_LEN);
+	error -= __put_user(0,name->nodename+__OLD_UTS_LEN);
+	error -= __copy_to_user(&name->release,&system_utsname.release,__OLD_UTS_LEN);
+	error -= __put_user(0,name->release+__OLD_UTS_LEN);
+	error -= __copy_to_user(&name->version,&system_utsname.version,__OLD_UTS_LEN);
+	error -= __put_user(0,name->version+__OLD_UTS_LEN);
+	error -= __copy_to_user(&name->machine,&system_utsname.machine,__OLD_UTS_LEN);
+	error -= __put_user(0,name->machine+__OLD_UTS_LEN);
+
+	return error ? -EFAULT : 0;
+}
+
+
+/*
+ * Build the string table for the builtin "poor man's strace".
+ */
+
+#if DEBUG
+#define SYSCALL(fun, narg) #fun,
+static char *sfnames[] = {
+#include "syscalls.h"
+};
+#undef SYS
+#endif
+
+void system_call (struct pt_regs *regs)
+{
+	syscall_t syscall;
+	unsigned long parm0, parm1, parm2, parm3, parm4, parm5;
+	int nargs, res;
+	unsigned int syscallnr;
+	int ps;
+
+#if DEBUG
+	int i;
+	unsigned long parms[6];
+	char *sysname;
+#endif
+
+	regs->syscall = regs->areg[2];
+
+	do_syscall_trace();
+
+	/* Have to load after syscall_trace because strace
+	 * sometimes changes regs->syscall.
+	 */
+	syscallnr = regs->syscall;
+
+	parm0 = parm1 = parm2 = parm3 = parm4 = parm5 = 0;
+
+	/* Restore interrupt level to syscall invoker's.
+	 * If this were in assembly, we wouldn't disable
+	 * interrupts in the first place:
+	 */
+	local_save_flags (ps);
+	local_irq_restore((ps & ~XCHAL_PS_INTLEVEL_MASK) |
+			  (regs->ps & XCHAL_PS_INTLEVEL_MASK) );
+
+	if (syscallnr > __NR_Linux_syscalls) {
+		regs->areg[2] = -ENOSYS;
+		return;
+	}
+
+	syscall = sys_call_table[syscallnr];
+	nargs = sys_narg_table[syscallnr];
+
+	if (syscall == NULL) {
+		regs->areg[2] = -ENOSYS;
+		return;
+	}
+
+	/* There shouldn't be more than six arguments in the table! */
+
+	if (nargs > 6)
+		panic("Internal error - too many syscall arguments (%d)!\n",
+		      nargs);
+
+	/* Linux takes system-call arguments in registers.  The ABI
+         * and Xtensa 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
+	 */
+	if (nargs) {
+		parm0 = regs->areg[nargs+2];
+		parm1 = regs->areg[3];
+		parm2 = regs->areg[4];
+		parm3 = regs->areg[5];
+		parm4 = regs->areg[6];
+		parm5 = regs->areg[7];
+	} else /* nargs == 0 */
+		parm0 = (unsigned long) regs;
+
+#if DEBUG
+	parms[0] = parm0;
+	parms[1] = parm1;
+	parms[2] = parm2;
+	parms[3] = parm3;
+	parms[4] = parm4;
+	parms[5] = parm5;
+
+	sysname = sfnames[syscallnr];
+	if (strncmp(sysname, "sys_", 4) == 0)
+		sysname = sysname + 4;
+
+	printk("\017SYSCALL:I:%x:%d:%s  %s(", regs->pc, current->pid,
+	       current->comm, sysname);
+	for (i = 0; i < nargs; i++)
+		printk((i>0) ? ", %#lx" : "%#lx", parms[i]);
+	printk(")\n");
+#endif
+
+	res = syscall((void *)parm0, parm1, parm2, parm3, parm4, parm5);
+
+#if DEBUG
+	printk("\017SYSCALL:O:%d:%s  %s(",current->pid, current->comm, sysname);
+	for (i = 0; i < nargs; i++)
+		printk((i>0) ? ", %#lx" : "%#lx", parms[i]);
+	if (res < 4096)
+		printk(") = %d\n", res);
+	else
+		printk(") = %#x\n", res);
+#endif /* DEBUG */
+
+	regs->areg[2] = res;
+	do_syscall_trace();
+}
+
+/*
+ * sys_ipc() is the de-multiplexer for the SysV IPC calls..
+ *
+ * This is really horribly ugly.
+ */
+
+int sys_ipc (uint call, int first, int second,
+			int third, void __user *ptr, long fifth)
+{
+	int version, ret;
+
+	version = call >> 16; /* hack for backward compatibility */
+	call &= 0xffff;
+	ret = -ENOSYS;
+
+	switch (call) {
+	case SEMOP:
+		ret = sys_semtimedop (first, (struct sembuf __user *)ptr,
+				     second, NULL);
+		break;
+
+	case SEMTIMEDOP:
+		ret = sys_semtimedop (first, (struct sembuf __user *)ptr,
+				      second, (const struct timespec *) fifth);
+		break;
+
+	case SEMGET:
+		ret = sys_semget (first, second, third);
+		break;
+
+	case SEMCTL: {
+		union semun fourth;
+
+		if (ptr && !get_user(fourth.__pad, (void *__user *) ptr))
+			ret = sys_semctl (first, second, third, fourth);
+		break;
+		}
+
+	case MSGSND:
+		ret = sys_msgsnd (first, (struct msgbuf __user*) ptr,
+				  second, third);
+		break;
+
+	case MSGRCV:
+		switch (version) {
+		case 0: {
+			struct ipc_kludge tmp;
+
+			if (ptr && !copy_from_user(&tmp,
+					   (struct ipc_kludge *) ptr,
+					   sizeof (tmp)))
+				ret = sys_msgrcv (first, tmp.msgp, second,
+						  tmp.msgtyp, third);
+			break;
+			}
+
+		default:
+			ret = sys_msgrcv (first, (struct msgbuf __user *) ptr,
+					  second, 0, third);
+			break;
+		}
+		break;
+
+	case MSGGET:
+		ret = sys_msgget ((key_t) first, second);
+		break;
+
+	case MSGCTL:
+		ret = sys_msgctl (first, second, (struct msqid_ds __user*) ptr);
+		break;
+
+	case SHMAT: {
+		ulong raddr;
+		ret = do_shmat (first, (char __user *) ptr, second, &raddr);
+
+		if (!ret)
+			ret = put_user (raddr, (ulong __user *) third);
+
+		break;
+		}
+
+	case SHMDT:
+		ret = sys_shmdt ((char __user *)ptr);
+		break;
+
+	case SHMGET:
+		ret = sys_shmget (first, second, third);
+		break;
+
+	case SHMCTL:
+		ret = sys_shmctl (first, second, (struct shmid_ds __user*) ptr);
+		break;
+	}
+	return ret;
+}
+
diff --git a/arch/xtensa/kernel/syscalls.h b/arch/xtensa/kernel/syscalls.h
new file mode 100644
index 00000000000..5b3f75f50fe
--- /dev/null
+++ b/arch/xtensa/kernel/syscalls.h
@@ -0,0 +1,248 @@
+/*
+ * arch/xtensa/kernel/syscalls.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.
+ *
+ * Changes by Joe Taylor <joe@tensilica.com>
+ */
+
+/*
+ * This file is being included twice - once to build a list of all
+ * syscalls and once to build a table of how many arguments each syscall
+ * accepts.  Syscalls that receive a pointer to the saved registers are
+ * marked as having zero arguments.
+ *
+ * The binary compatibility calls are in a separate list.
+ *
+ * Entry '0' used to be system_call.  It's removed to disable indirect
+ * system calls for now so user tasks can't recurse.  See mips'
+ * sys_syscall for a comparable example.
+ */
+
+SYSCALL(0, 0)		                /* 00 */
+
+SYSCALL(sys_exit, 1)
+SYSCALL(sys_fork, 0)
+SYSCALL(sys_read, 3)
+SYSCALL(sys_write, 3)
+SYSCALL(sys_open, 3)			/* 05 */
+SYSCALL(sys_close, 1)
+SYSCALL(sys_waitpid, 3)
+SYSCALL(sys_creat, 2)
+SYSCALL(sys_link, 2)
+SYSCALL(sys_unlink, 1)			/* 10 */
+SYSCALL(sys_execve, 0)
+SYSCALL(sys_chdir, 1)
+SYSCALL(sys_time, 1)
+SYSCALL(sys_mknod, 3)
+SYSCALL(sys_chmod, 2)			/* 15 */
+SYSCALL(sys_lchown, 3)
+SYSCALL(sys_ni_syscall, 0)
+SYSCALL(sys_stat, 2)
+SYSCALL(sys_lseek, 3)
+SYSCALL(sys_getpid, 0)			/* 20 */
+SYSCALL(sys_mount, 5)
+SYSCALL(sys_oldumount, 1)
+SYSCALL(sys_setuid, 1)
+SYSCALL(sys_getuid, 0)
+SYSCALL(sys_stime, 1)			/* 25 */
+SYSCALL(sys_ptrace, 4)
+SYSCALL(sys_alarm, 1)
+SYSCALL(sys_fstat, 2)
+SYSCALL(sys_pause, 0)
+SYSCALL(sys_utime, 2)			/* 30 */
+SYSCALL(sys_ni_syscall, 0)
+SYSCALL(sys_ni_syscall, 0)
+SYSCALL(sys_access, 2)
+SYSCALL(sys_nice, 1)
+SYSCALL(sys_ni_syscall, 0)		/* 35 */
+SYSCALL(sys_sync, 0)
+SYSCALL(sys_kill, 2)
+SYSCALL(sys_rename, 2)
+SYSCALL(sys_mkdir, 2)
+SYSCALL(sys_rmdir, 1)			/* 40 */
+SYSCALL(sys_dup, 1)
+SYSCALL(sys_pipe, 1)
+SYSCALL(sys_times, 1)
+SYSCALL(sys_ni_syscall, 0)
+SYSCALL(sys_brk, 1)			/* 45 */
+SYSCALL(sys_setgid, 1)
+SYSCALL(sys_getgid, 0)
+SYSCALL(sys_ni_syscall, 0)		/* was signal(2) */
+SYSCALL(sys_geteuid, 0)
+SYSCALL(sys_getegid, 0)			/* 50 */
+SYSCALL(sys_acct, 1)
+SYSCALL(sys_umount, 2)
+SYSCALL(sys_ni_syscall, 0)
+SYSCALL(sys_ioctl, 3)
+SYSCALL(sys_fcntl, 3)			/* 55 */
+SYSCALL(sys_ni_syscall, 2)
+SYSCALL(sys_setpgid, 2)
+SYSCALL(sys_ni_syscall, 0)
+SYSCALL(sys_olduname, 1)
+SYSCALL(sys_umask, 1)			/* 60 */
+SYSCALL(sys_chroot, 1)
+SYSCALL(sys_ustat, 2)
+SYSCALL(sys_dup2, 2)
+SYSCALL(sys_getppid, 0)
+SYSCALL(sys_getpgrp, 0)			/* 65 */
+SYSCALL(sys_setsid, 0)
+SYSCALL(sys_sigaction, 3)
+SYSCALL(sys_sgetmask, 0)
+SYSCALL(sys_ssetmask, 1)
+SYSCALL(sys_setreuid, 2)		/* 70 */
+SYSCALL(sys_setregid, 2)
+SYSCALL(sys_sigsuspend, 0)
+SYSCALL(sys_sigpending, 1)
+SYSCALL(sys_sethostname, 2)
+SYSCALL(sys_setrlimit, 2)		/* 75 */
+SYSCALL(sys_getrlimit, 2)
+SYSCALL(sys_getrusage, 2)
+SYSCALL(sys_gettimeofday, 2)
+SYSCALL(sys_settimeofday, 2)
+SYSCALL(sys_getgroups, 2)		/* 80 */
+SYSCALL(sys_setgroups, 2)
+SYSCALL(sys_ni_syscall, 0)		 /* old_select */
+SYSCALL(sys_symlink, 2)
+SYSCALL(sys_lstat, 2)
+SYSCALL(sys_readlink, 3)		/* 85 */
+SYSCALL(sys_uselib, 1)
+SYSCALL(sys_swapon, 2)
+SYSCALL(sys_reboot, 3)
+SYSCALL(old_readdir, 3)
+SYSCALL(old_mmap, 6)			/* 90 */
+SYSCALL(sys_munmap, 2)
+SYSCALL(sys_truncate, 2)
+SYSCALL(sys_ftruncate, 2)
+SYSCALL(sys_fchmod, 2)
+SYSCALL(sys_fchown, 3)			/* 95 */
+SYSCALL(sys_getpriority, 2)
+SYSCALL(sys_setpriority, 3)
+SYSCALL(sys_ni_syscall, 0)
+SYSCALL(sys_statfs, 2)
+SYSCALL(sys_fstatfs, 2)			/* 100 */
+SYSCALL(sys_ni_syscall, 3)
+SYSCALL(sys_socketcall, 2)
+SYSCALL(sys_syslog, 3)
+SYSCALL(sys_setitimer, 3)
+SYSCALL(sys_getitimer, 2)		/* 105 */
+SYSCALL(sys_newstat, 2)
+SYSCALL(sys_newlstat, 2)
+SYSCALL(sys_newfstat, 2)
+SYSCALL(sys_uname, 1)
+SYSCALL(sys_ni_syscall, 0)		/* 110 */
+SYSCALL(sys_vhangup, 0)
+SYSCALL(sys_ni_syscall, 0)		/* was sys_idle() */
+SYSCALL(sys_ni_syscall, 0)
+SYSCALL(sys_wait4, 4)
+SYSCALL(sys_swapoff, 1)			/* 115 */
+SYSCALL(sys_sysinfo, 1)
+SYSCALL(sys_ipc, 5)   			/* 6 really, but glibc uses only 5) */
+SYSCALL(sys_fsync, 1)
+SYSCALL(sys_sigreturn, 0)
+SYSCALL(sys_clone, 0)			/* 120 */
+SYSCALL(sys_setdomainname, 2)
+SYSCALL(sys_newuname, 1)
+SYSCALL(sys_ni_syscall, 0) 		/* sys_modify_ldt */
+SYSCALL(sys_adjtimex, 1)
+SYSCALL(sys_mprotect, 3)		/* 125 */
+SYSCALL(sys_sigprocmask, 3)
+SYSCALL(sys_ni_syscall, 2)		/* old sys_create_module */
+SYSCALL(sys_init_module, 2)
+SYSCALL(sys_delete_module, 1)
+SYSCALL(sys_ni_syscall, 1)		/* old sys_get_kernel_sysm */	/* 130 */
+SYSCALL(sys_quotactl, 0)
+SYSCALL(sys_getpgid, 1)
+SYSCALL(sys_fchdir, 1)
+SYSCALL(sys_bdflush, 2)
+SYSCALL(sys_sysfs, 3)			/* 135 */
+SYSCALL(sys_personality, 1)
+SYSCALL(sys_ni_syscall, 0)		/* for afs_syscall */
+SYSCALL(sys_setfsuid, 1)
+SYSCALL(sys_setfsgid, 1)
+SYSCALL(sys_llseek, 5)			/* 140 */
+SYSCALL(sys_getdents, 3)
+SYSCALL(sys_select, 5)
+SYSCALL(sys_flock, 2)
+SYSCALL(sys_msync, 3)
+SYSCALL(sys_readv, 3)			/* 145 */
+SYSCALL(sys_writev, 3)
+SYSCALL(sys_ni_syscall, 3)
+SYSCALL(sys_ni_syscall, 3)
+SYSCALL(sys_ni_syscall, 4)		/* handled in fast syscall handler. */
+SYSCALL(sys_ni_syscall, 0)		/* 150 */
+SYSCALL(sys_getsid, 1)
+SYSCALL(sys_fdatasync, 1)
+SYSCALL(sys_sysctl, 1)
+SYSCALL(sys_mlock, 2)
+SYSCALL(sys_munlock, 2)			/* 155 */
+SYSCALL(sys_mlockall, 1)
+SYSCALL(sys_munlockall, 0)
+SYSCALL(sys_sched_setparam,2)
+SYSCALL(sys_sched_getparam,2)
+SYSCALL(sys_sched_setscheduler,3)	/* 160 */
+SYSCALL(sys_sched_getscheduler,1)
+SYSCALL(sys_sched_yield,0)
+SYSCALL(sys_sched_get_priority_max,1)
+SYSCALL(sys_sched_get_priority_min,1)
+SYSCALL(sys_sched_rr_get_interval,2)	/* 165 */
+SYSCALL(sys_nanosleep,2)
+SYSCALL(sys_mremap,4)
+SYSCALL(sys_accept, 3)
+SYSCALL(sys_bind, 3)
+SYSCALL(sys_connect, 3)			/* 170 */
+SYSCALL(sys_getpeername, 3)
+SYSCALL(sys_getsockname, 3)
+SYSCALL(sys_getsockopt, 5)
+SYSCALL(sys_listen, 2)
+SYSCALL(sys_recv, 4)			/* 175 */
+SYSCALL(sys_recvfrom, 6)
+SYSCALL(sys_recvmsg, 3)
+SYSCALL(sys_send, 4)
+SYSCALL(sys_sendmsg, 3)
+SYSCALL(sys_sendto, 6)			/* 180 */
+SYSCALL(sys_setsockopt, 5)
+SYSCALL(sys_shutdown, 2)
+SYSCALL(sys_socket, 3)
+SYSCALL(sys_socketpair, 4)
+SYSCALL(sys_setresuid, 3)		/* 185 */
+SYSCALL(sys_getresuid, 3)
+SYSCALL(sys_ni_syscall, 5)		/* old sys_query_module */
+SYSCALL(sys_poll, 3)
+SYSCALL(sys_nfsservctl, 3)
+SYSCALL(sys_setresgid, 3)		/* 190 */
+SYSCALL(sys_getresgid, 3)
+SYSCALL(sys_prctl, 5)
+SYSCALL(sys_rt_sigreturn, 0)
+SYSCALL(sys_rt_sigaction, 4)
+SYSCALL(sys_rt_sigprocmask, 4)		/* 195 */
+SYSCALL(sys_rt_sigpending, 2)
+SYSCALL(sys_rt_sigtimedwait, 4)
+SYSCALL(sys_rt_sigqueueinfo, 3)
+SYSCALL(sys_rt_sigsuspend, 0)
+SYSCALL(sys_pread64, 5)			/* 200 */
+SYSCALL(sys_pwrite64, 5)
+SYSCALL(sys_chown, 3)
+SYSCALL(sys_getcwd, 2)
+SYSCALL(sys_capget, 2)
+SYSCALL(sys_capset, 2)			/* 205 */
+SYSCALL(sys_sigaltstack, 0)
+SYSCALL(sys_sendfile, 4)
+SYSCALL(sys_ni_syscall, 0)
+SYSCALL(sys_ni_syscall, 0)
+SYSCALL(sys_mmap2, 6)			/* 210 */
+SYSCALL(sys_truncate64, 2)
+SYSCALL(sys_ftruncate64, 2)
+SYSCALL(sys_stat64, 2)
+SYSCALL(sys_lstat64, 2)
+SYSCALL(sys_fstat64, 2)			/* 215 */
+SYSCALL(sys_pivot_root, 2)
+SYSCALL(sys_mincore, 3)
+SYSCALL(sys_madvise, 3)
+SYSCALL(sys_getdents64, 3)
+SYSCALL(sys_vfork, 0)			/* 220 */
diff --git a/arch/xtensa/kernel/time.c b/arch/xtensa/kernel/time.c
new file mode 100644
index 00000000000..e07287db5a4
--- /dev/null
+++ b/arch/xtensa/kernel/time.c
@@ -0,0 +1,227 @@
+/*
+ * arch/xtensa/kernel/time.c
+ *
+ * Timer and clock support.
+ *
+ * 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.
+ *
+ * Chris Zankel <chris@zankel.net>
+ */
+
+#include <linux/config.h>
+#include <linux/errno.h>
+#include <linux/time.h>
+#include <linux/timex.h>
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <linux/profile.h>
+#include <linux/delay.h>
+
+#include <asm/timex.h>
+#include <asm/platform.h>
+
+
+extern volatile unsigned long wall_jiffies;
+
+u64 jiffies_64 = INITIAL_JIFFIES;
+EXPORT_SYMBOL(jiffies_64);
+
+spinlock_t rtc_lock = SPIN_LOCK_UNLOCKED;
+EXPORT_SYMBOL(rtc_lock);
+
+
+#ifdef CONFIG_XTENSA_CALIBRATE_CCOUNT
+unsigned long ccount_per_jiffy;		/* per 1/HZ */
+unsigned long ccount_nsec;		/* nsec per ccount increment */
+#endif
+
+unsigned int last_ccount_stamp;
+static long last_rtc_update = 0;
+
+/*
+ * Scheduler clock - returns current tim in nanosec units.
+ */
+
+unsigned long long sched_clock(void)
+{
+	return (unsigned long long)jiffies * (1000000000 / HZ);
+}
+
+static irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs);
+static struct irqaction timer_irqaction = {
+	.handler =	timer_interrupt,
+	.flags =	SA_INTERRUPT,
+	.name =		"timer",
+};
+
+void __init time_init(void)
+{
+	time_t sec_o, sec_n = 0;
+
+	/* The platform must provide a function to calibrate the processor
+	 * speed for the CALIBRATE.
+	 */
+
+#if CONFIG_XTENSA_CALIBRATE_CCOUNT
+	printk("Calibrating CPU frequency ");
+	platform_calibrate_ccount();
+	printk("%d.%02d MHz\n", (int)ccount_per_jiffy/(1000000/HZ),
+			(int)(ccount_per_jiffy/(10000/HZ))%100);
+#endif
+
+	/* Set time from RTC (if provided) */
+
+	if (platform_get_rtc_time(&sec_o) == 0)
+		while (platform_get_rtc_time(&sec_n))
+			if (sec_o != sec_n)
+				break;
+
+	xtime.tv_nsec = 0;
+	last_rtc_update = xtime.tv_sec = sec_n;
+	last_ccount_stamp = get_ccount();
+
+	set_normalized_timespec(&wall_to_monotonic,
+		-xtime.tv_sec, -xtime.tv_nsec);
+
+	/* Initialize the linux timer interrupt. */
+
+	setup_irq(LINUX_TIMER_INT, &timer_irqaction);
+	set_linux_timer(get_ccount() + CCOUNT_PER_JIFFY);
+}
+
+
+int do_settimeofday(struct timespec *tv)
+{
+	time_t wtm_sec, sec = tv->tv_sec;
+	long wtm_nsec, nsec = tv->tv_nsec;
+	unsigned long ccount;
+
+	if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC)
+		return -EINVAL;
+
+	write_seqlock_irq(&xtime_lock);
+
+	/* This is revolting. We need to set "xtime" correctly. However, the
+	 * value in this location is the value at the most recent update of
+	 * wall time.  Discover what correction gettimeofday() would have
+	 * made, and then undo it!
+	 */
+	ccount = get_ccount();
+	nsec -= (ccount - last_ccount_stamp) * CCOUNT_NSEC;
+	nsec -= (jiffies - wall_jiffies) * CCOUNT_PER_JIFFY * CCOUNT_NSEC;
+
+	wtm_sec  = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec);
+	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec);
+
+	set_normalized_timespec(&xtime, sec, nsec);
+	set_normalized_timespec(&wall_to_monotonic, wtm_sec, wtm_nsec);
+
+	time_adjust = 0;                /* stop active adjtime() */
+	time_status |= STA_UNSYNC;
+	time_maxerror = NTP_PHASE_LIMIT;
+	time_esterror = NTP_PHASE_LIMIT;
+	write_sequnlock_irq(&xtime_lock);
+	return 0;
+}
+
+EXPORT_SYMBOL(do_settimeofday);
+
+
+void do_gettimeofday(struct timeval *tv)
+{
+	unsigned long flags;
+	unsigned long sec, usec, delta, lost, seq;
+
+	do {
+		seq = read_seqbegin_irqsave(&xtime_lock, flags);
+
+		delta = get_ccount() - last_ccount_stamp;
+		sec = xtime.tv_sec;
+		usec = (xtime.tv_nsec / NSEC_PER_USEC);
+
+		lost = jiffies - wall_jiffies;
+
+	} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));
+
+	usec += lost * (1000000UL/HZ) + (delta * CCOUNT_NSEC) / NSEC_PER_USEC;
+	for (; usec >= 1000000; sec++, usec -= 1000000)
+		;
+
+	tv->tv_sec = sec;
+	tv->tv_usec = usec;
+}
+
+EXPORT_SYMBOL(do_gettimeofday);
+
+/*
+ * The timer interrupt is called HZ times per second.
+ */
+
+irqreturn_t timer_interrupt (int irq, void *dev_id, struct pt_regs *regs)
+{
+
+	unsigned long next;
+
+	next = get_linux_timer();
+
+again:
+	while ((signed long)(get_ccount() - next) > 0) {
+
+		profile_tick(CPU_PROFILING, regs);
+#ifndef CONFIG_SMP
+		update_process_times(user_mode(regs));
+#endif
+
+		write_seqlock(&xtime_lock);
+
+		last_ccount_stamp = next;
+		next += CCOUNT_PER_JIFFY;
+		do_timer (regs); /* Linux handler in kernel/timer.c */
+
+		if ((time_status & STA_UNSYNC) == 0 &&
+		    xtime.tv_sec - last_rtc_update >= 659 &&
+		    abs((xtime.tv_nsec/1000)-(1000000-1000000/HZ))<5000000/HZ &&
+		    jiffies - wall_jiffies == 1) {
+
+			if (platform_set_rtc_time(xtime.tv_sec+1) == 0)
+				last_rtc_update = xtime.tv_sec+1;
+			else
+				/* Do it again in 60 s */
+				last_rtc_update += 60;
+		}
+		write_sequnlock(&xtime_lock);
+	}
+
+	/* NOTE: writing CCOMPAREn clears the interrupt.  */
+
+	set_linux_timer (next);
+
+	/* Make sure we didn't miss any tick... */
+
+	if ((signed long)(get_ccount() - next) > 0)
+		goto again;
+
+	/* Allow platform to do something usefull (Wdog). */
+
+	platform_heartbeat();
+
+	return IRQ_HANDLED;
+}
+
+#ifndef CONFIG_GENERIC_CALIBRATE_DELAY
+void __devinit calibrate_delay(void)
+{
+	loops_per_jiffy = CCOUNT_PER_JIFFY;
+	printk("Calibrating delay loop (skipped)... "
+	       "%lu.%02lu BogoMIPS preset\n",
+	       loops_per_jiffy/(1000000/HZ),
+	       (loops_per_jiffy/(10000/HZ)) % 100);
+}
+#endif
+
diff --git a/arch/xtensa/kernel/traps.c b/arch/xtensa/kernel/traps.c
new file mode 100644
index 00000000000..804246e743b
--- /dev/null
+++ b/arch/xtensa/kernel/traps.c
@@ -0,0 +1,498 @@
+/*
+ * arch/xtensa/kernel/traps.c
+ *
+ * Exception handling.
+ *
+ * Derived from code with the following copyrights:
+ * Copyright (C) 1994 - 1999 by Ralf Baechle
+ * Modified for R3000 by Paul M. Antoine, 1995, 1996
+ * Complete output from die() by Ulf Carlsson, 1998
+ * Copyright (C) 1999 Silicon Graphics, Inc.
+ *
+ * Essentially rewritten for the Xtensa architecture port.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ *
+ * Joe Taylor	<joe@tensilica.com, joetylr@yahoo.com>
+ * Chris Zankel	<chris@zankel.net>
+ * Marc Gauthier<marc@tensilica.com, marc@alumni.uwaterloo.ca>
+ * Kevin Chea
+ *
+ * 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.
+ */
+
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/stringify.h>
+#include <linux/kallsyms.h>
+
+#include <asm/ptrace.h>
+#include <asm/timex.h>
+#include <asm/uaccess.h>
+#include <asm/pgtable.h>
+#include <asm/processor.h>
+
+#ifdef CONFIG_KGDB
+extern int gdb_enter;
+extern int return_from_debug_flag;
+#endif
+
+/*
+ * Machine specific interrupt handlers
+ */
+
+extern void kernel_exception(void);
+extern void user_exception(void);
+
+extern void fast_syscall_kernel(void);
+extern void fast_syscall_user(void);
+extern void fast_alloca(void);
+extern void fast_unaligned(void);
+extern void fast_second_level_miss(void);
+extern void fast_store_prohibited(void);
+extern void fast_coprocessor(void);
+
+extern void do_illegal_instruction (struct pt_regs*);
+extern void do_interrupt (struct pt_regs*);
+extern void do_unaligned_user (struct pt_regs*);
+extern void do_multihit (struct pt_regs*, unsigned long);
+extern void do_page_fault (struct pt_regs*, unsigned long);
+extern void do_debug (struct pt_regs*);
+extern void system_call (struct pt_regs*);
+
+/*
+ * The vector table must be preceded by a save area (which
+ * implies it must be in RAM, unless one places RAM immediately
+ * before a ROM and puts the vector at the start of the ROM (!))
+ */
+
+#define KRNL		0x01
+#define USER		0x02
+
+#define COPROCESSOR(x)							\
+{ XCHAL_EXCCAUSE_COPROCESSOR ## x ## _DISABLED, USER, fast_coprocessor }
+
+typedef struct {
+	int cause;
+	int fast;
+	void* handler;
+} dispatch_init_table_t;
+
+dispatch_init_table_t __init dispatch_init_table[] = {
+
+{ XCHAL_EXCCAUSE_ILLEGAL_INSTRUCTION,	0,	   do_illegal_instruction},
+{ XCHAL_EXCCAUSE_SYSTEM_CALL,		KRNL,	   fast_syscall_kernel },
+{ XCHAL_EXCCAUSE_SYSTEM_CALL,		USER,	   fast_syscall_user },
+{ XCHAL_EXCCAUSE_SYSTEM_CALL,		0,	   system_call },
+/* XCHAL_EXCCAUSE_INSTRUCTION_FETCH unhandled */
+/* XCHAL_EXCCAUSE_LOAD_STORE_ERROR unhandled*/
+{ XCHAL_EXCCAUSE_LEVEL1_INTERRUPT,	0,	   do_interrupt },
+{ XCHAL_EXCCAUSE_ALLOCA,		USER|KRNL, fast_alloca },
+/* XCHAL_EXCCAUSE_INTEGER_DIVIDE_BY_ZERO unhandled */
+/* XCHAL_EXCCAUSE_PRIVILEGED unhandled */
+#if XCHAL_UNALIGNED_LOAD_EXCEPTION || XCHAL_UNALIGNED_STORE_EXCEPTION
+#ifdef CONFIG_UNALIGNED_USER
+{ XCHAL_EXCCAUSE_UNALIGNED,		USER,	   fast_unaligned },
+#else
+{ XCHAL_EXCCAUSE_UNALIGNED,		0,	   do_unaligned_user },
+#endif
+{ XCHAL_EXCCAUSE_UNALIGNED,		KRNL,	   fast_unaligned },
+#endif
+{ XCHAL_EXCCAUSE_ITLB_MISS,		0,	   do_page_fault },
+{ XCHAL_EXCCAUSE_ITLB_MISS,		USER|KRNL, fast_second_level_miss},
+{ XCHAL_EXCCAUSE_ITLB_MULTIHIT,		0,	   do_multihit },
+{ XCHAL_EXCCAUSE_ITLB_PRIVILEGE,	0,	   do_page_fault },
+/* XCHAL_EXCCAUSE_SIZE_RESTRICTION unhandled */
+{ XCHAL_EXCCAUSE_FETCH_CACHE_ATTRIBUTE,	0,	   do_page_fault },
+{ XCHAL_EXCCAUSE_DTLB_MISS,		USER|KRNL, fast_second_level_miss},
+{ XCHAL_EXCCAUSE_DTLB_MISS,		0,	   do_page_fault },
+{ XCHAL_EXCCAUSE_DTLB_MULTIHIT,		0,	   do_multihit },
+{ XCHAL_EXCCAUSE_DTLB_PRIVILEGE,	0,	   do_page_fault },
+/* XCHAL_EXCCAUSE_DTLB_SIZE_RESTRICTION unhandled */
+{ XCHAL_EXCCAUSE_STORE_CACHE_ATTRIBUTE,	USER|KRNL, fast_store_prohibited },
+{ XCHAL_EXCCAUSE_STORE_CACHE_ATTRIBUTE,	0,	   do_page_fault },
+{ XCHAL_EXCCAUSE_LOAD_CACHE_ATTRIBUTE,	0,	   do_page_fault },
+/* XCCHAL_EXCCAUSE_FLOATING_POINT unhandled */
+#if (XCHAL_CP_MASK & 1)
+COPROCESSOR(0),
+#endif
+#if (XCHAL_CP_MASK & 2)
+COPROCESSOR(1),
+#endif
+#if (XCHAL_CP_MASK & 4)
+COPROCESSOR(2),
+#endif
+#if (XCHAL_CP_MASK & 8)
+COPROCESSOR(3),
+#endif
+#if (XCHAL_CP_MASK & 16)
+COPROCESSOR(4),
+#endif
+#if (XCHAL_CP_MASK & 32)
+COPROCESSOR(5),
+#endif
+#if (XCHAL_CP_MASK & 64)
+COPROCESSOR(6),
+#endif
+#if (XCHAL_CP_MASK & 128)
+COPROCESSOR(7),
+#endif
+{ EXCCAUSE_MAPPED_DEBUG,		0,		do_debug },
+{ -1, -1, 0 }
+
+};
+
+/* The exception table <exc_table> serves two functions:
+ * 1. it contains three dispatch tables (fast_user, fast_kernel, default-c)
+ * 2. it is a temporary memory buffer for the exception handlers.
+ */
+
+unsigned long exc_table[EXC_TABLE_SIZE/4];
+
+void die(const char*, struct pt_regs*, long);
+
+static inline void
+__die_if_kernel(const char *str, struct pt_regs *regs, long err)
+{
+	if (!user_mode(regs))
+		die(str, regs, err);
+}
+
+/*
+ * Unhandled Exceptions. Kill user task or panic if in kernel space.
+ */
+
+void do_unhandled(struct pt_regs *regs, unsigned long exccause)
+{
+	__die_if_kernel("Caught unhandled exception - should not happen",
+	    		regs, SIGKILL);
+
+	/* If in user mode, send SIGILL signal to current process */
+	printk("Caught unhandled exception in '%s' "
+	       "(pid = %d, pc = %#010lx) - should not happen\n"
+	       "\tEXCCAUSE is %ld\n",
+	       current->comm, current->pid, regs->pc, exccause);
+	force_sig(SIGILL, current);
+}
+
+/*
+ * Multi-hit exception. This if fatal!
+ */
+
+void do_multihit(struct pt_regs *regs, unsigned long exccause)
+{
+	die("Caught multihit exception", regs, SIGKILL);
+}
+
+/*
+ * Level-1 interrupt.
+ * We currently have no priority encoding.
+ */
+
+unsigned long ignored_level1_interrupts;
+extern void do_IRQ(int, struct pt_regs *);
+
+void do_interrupt (struct pt_regs *regs)
+{
+	unsigned long intread = get_sr (INTREAD);
+	unsigned long intenable = get_sr (INTENABLE);
+	int i, mask;
+
+	/* Handle all interrupts (no priorities).
+	 * (Clear the interrupt before processing, in case it's
+	 *  edge-triggered or software-generated)
+	 */
+
+	for (i=0, mask = 1; i < XCHAL_NUM_INTERRUPTS; i++, mask <<= 1) {
+		if (mask & (intread & intenable)) {
+			set_sr (mask, INTCLEAR);
+			do_IRQ (i,regs);
+		}
+	}
+}
+
+/*
+ * Illegal instruction. Fatal if in kernel space.
+ */
+
+void
+do_illegal_instruction(struct pt_regs *regs)
+{
+	__die_if_kernel("Illegal instruction in kernel", regs, SIGKILL);
+
+	/* If in user mode, send SIGILL signal to current process. */
+
+	printk("Illegal Instruction in '%s' (pid = %d, pc = %#010lx)\n",
+	    current->comm, current->pid, regs->pc);
+	force_sig(SIGILL, current);
+}
+
+
+/*
+ * Handle unaligned memory accesses from user space. Kill task.
+ *
+ * If CONFIG_UNALIGNED_USER is not set, we don't allow unaligned memory
+ * accesses causes from user space.
+ */
+
+#if XCHAL_UNALIGNED_LOAD_EXCEPTION || XCHAL_UNALIGNED_STORE_EXCEPTION
+#ifndef CONFIG_UNALIGNED_USER
+void
+do_unaligned_user (struct pt_regs *regs)
+{
+	siginfo_t info;
+
+	__die_if_kernel("Unhandled unaligned exception in kernel",
+	    		regs, SIGKILL);
+
+	current->thread.bad_vaddr = regs->excvaddr;
+	current->thread.error_code = -3;
+	printk("Unaligned memory access to %08lx in '%s' "
+	       "(pid = %d, pc = %#010lx)\n",
+	       regs->excvaddr, current->comm, current->pid, regs->pc);
+	info.si_signo = SIGBUS;
+	info.si_errno = 0;
+	info.si_code = BUS_ADRALN;
+	info.si_addr = (void *) regs->excvaddr;
+	force_sig_info(SIGSEGV, &info, current);
+
+}
+#endif
+#endif
+
+void
+do_debug(struct pt_regs *regs)
+{
+#ifdef CONFIG_KGDB
+	/* If remote debugging is configured AND enabled, we give control to
+	 * kgdb.  Otherwise, we fall through, perhaps giving control to the
+	 * native debugger.
+	 */
+
+	if (gdb_enter) {
+		extern void gdb_handle_exception(struct pt_regs *);
+		gdb_handle_exception(regs);
+		return_from_debug_flag = 1;
+		return;
+	}
+#endif
+
+	__die_if_kernel("Breakpoint in kernel", regs, SIGKILL);
+
+	/* If in user mode, send SIGTRAP signal to current process */
+
+	force_sig(SIGTRAP, current);
+}
+
+
+/*
+ * Initialize dispatch tables.
+ *
+ * The exception vectors are stored compressed the __init section in the
+ * dispatch_init_table. This function initializes the following three tables
+ * from that compressed table:
+ * - fast user		first dispatch table for user exceptions
+ * - fast kernel	first dispatch table for kernel exceptions
+ * - default C-handler	C-handler called by the default fast handler.
+ *
+ * See vectors.S for more details.
+ */
+
+#define set_handler(idx,handler) (exc_table[idx] = (unsigned long) (handler))
+
+void trap_init(void)
+{
+	int i;
+
+	/* Setup default vectors. */
+
+	for(i = 0; i < 64; i++) {
+		set_handler(EXC_TABLE_FAST_USER/4   + i, user_exception);
+		set_handler(EXC_TABLE_FAST_KERNEL/4 + i, kernel_exception);
+		set_handler(EXC_TABLE_DEFAULT/4 + i, do_unhandled);
+	}
+
+	/* Setup specific handlers. */
+
+	for(i = 0; dispatch_init_table[i].cause >= 0; i++) {
+
+		int fast = dispatch_init_table[i].fast;
+		int cause = dispatch_init_table[i].cause;
+		void *handler = dispatch_init_table[i].handler;
+
+		if (fast == 0)
+			set_handler (EXC_TABLE_DEFAULT/4 + cause, handler);
+		if (fast && fast & USER)
+			set_handler (EXC_TABLE_FAST_USER/4 + cause, handler);
+		if (fast && fast & KRNL)
+			set_handler (EXC_TABLE_FAST_KERNEL/4 + cause, handler);
+	}
+
+	/* Initialize EXCSAVE_1 to hold the address of the exception table. */
+
+	i = (unsigned long)exc_table;
+	__asm__ __volatile__("wsr  %0, "__stringify(EXCSAVE_1)"\n" : : "a" (i));
+}
+
+/*
+ * This function dumps the current valid window frame and other base registers.
+ */
+
+void show_regs(struct pt_regs * regs)
+{
+	int i, wmask;
+
+	wmask = regs->wmask & ~1;
+
+	for (i = 0; i < 32; i++) {
+		if (wmask & (1 << (i / 4)))
+			break;
+		if ((i % 8) == 0)
+			printk ("\n" KERN_INFO "a%02d: ", i);
+		printk("%08lx ", regs->areg[i]);
+	}
+	printk("\n");
+
+	printk("pc: %08lx, ps: %08lx, depc: %08lx, excvaddr: %08lx\n",
+	       regs->pc, regs->ps, regs->depc, regs->excvaddr);
+	printk("lbeg: %08lx, lend: %08lx lcount: %08lx, sar: %08lx\n",
+	       regs->lbeg, regs->lend, regs->lcount, regs->sar);
+	if (user_mode(regs))
+		printk("wb: %08lx, ws: %08lx, wmask: %08lx, syscall: %ld\n",
+		       regs->windowbase, regs->windowstart, regs->wmask,
+		       regs->syscall);
+}
+
+void show_trace(struct task_struct *task, unsigned long *sp)
+{
+	unsigned long a0, a1, pc;
+	unsigned long sp_start, sp_end;
+
+	a1 = (unsigned long)sp;
+
+	if (a1 == 0)
+		__asm__ __volatile__ ("mov %0, a1\n" : "=a"(a1));
+
+
+	sp_start = a1 & ~(THREAD_SIZE-1);
+	sp_end = sp_start + THREAD_SIZE;
+
+	printk("Call Trace:");
+#ifdef CONFIG_KALLSYMS
+	printk("\n");
+#endif
+	spill_registers();
+
+	while (a1 > sp_start && a1 < sp_end) {
+		sp = (unsigned long*)a1;
+
+		a0 = *(sp - 4);
+		a1 = *(sp - 3);
+
+		if (a1 <= (unsigned long) sp)
+			break;
+
+		pc = MAKE_PC_FROM_RA(a0, a1);
+
+		if (kernel_text_address(pc)) {
+			printk(" [<%08lx>] ", pc);
+			print_symbol("%s\n", pc);
+		}
+	}
+	printk("\n");
+}
+
+/*
+ * This routine abuses get_user()/put_user() to reference pointers
+ * with at least a bit of error checking ...
+ */
+
+static int kstack_depth_to_print = 24;
+
+void show_stack(struct task_struct *task, unsigned long *sp)
+{
+	int i = 0;
+	unsigned long *stack;
+
+	if (sp == 0)
+		__asm__ __volatile__ ("mov %0, a1\n" : "=a"(sp));
+
+ 	stack = sp;
+
+	printk("\nStack: ");
+
+	for (i = 0; i < kstack_depth_to_print; i++) {
+		if (kstack_end(sp))
+			break;
+		if (i && ((i % 8) == 0))
+			printk("\n       ");
+		printk("%08lx ", *sp++);
+	}
+	printk("\n");
+	show_trace(task, stack);
+}
+
+void dump_stack(void)
+{
+	show_stack(current, NULL);
+}
+
+EXPORT_SYMBOL(dump_stack);
+
+
+void show_code(unsigned int *pc)
+{
+	long i;
+
+	printk("\nCode:");
+
+	for(i = -3 ; i < 6 ; i++) {
+		unsigned long insn;
+		if (__get_user(insn, pc + i)) {
+			printk(" (Bad address in pc)\n");
+			break;
+		}
+		printk("%c%08lx%c",(i?' ':'<'),insn,(i?' ':'>'));
+	}
+}
+
+spinlock_t die_lock = SPIN_LOCK_UNLOCKED;
+
+void die(const char * str, struct pt_regs * regs, long err)
+{
+	static int die_counter;
+	int nl = 0;
+
+	console_verbose();
+	spin_lock_irq(&die_lock);
+
+	printk("%s: sig: %ld [#%d]\n", str, err, ++die_counter);
+#ifdef CONFIG_PREEMPT
+	printk("PREEMPT ");
+	nl = 1;
+#endif
+	if (nl)
+		printk("\n");
+	show_regs(regs);
+	if (!user_mode(regs))
+		show_stack(NULL, (unsigned long*)regs->areg[1]);
+
+	spin_unlock_irq(&die_lock);
+
+	if (in_interrupt())
+		panic("Fatal exception in interrupt");
+
+	if (panic_on_oops) {
+		printk(KERN_EMERG "Fatal exception: panic in 5 seconds\n");
+		set_current_state(TASK_UNINTERRUPTIBLE);
+		schedule_timeout(5 * HZ);
+		panic("Fatal exception");
+	}
+	do_exit(err);
+}
+
+
diff --git a/arch/xtensa/kernel/vectors.S b/arch/xtensa/kernel/vectors.S
new file mode 100644
index 00000000000..81808f0c674
--- /dev/null
+++ b/arch/xtensa/kernel/vectors.S
@@ -0,0 +1,464 @@
+/*
+ * arch/xtensa/kernel/vectors.S
+ *
+ * This file contains all exception vectors (user, kernel, and double),
+ * as well as the window vectors (overflow and underflow), and the debug
+ * vector. These are the primary vectors executed by the processor if an
+ * exception occurs.
+ *
+ * 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.
+ *
+ * Chris Zankel <chris@zankel.net>
+ *
+ */
+
+/*
+ * We use a two-level table approach. The user and kernel exception vectors
+ * use a first-level dispatch table to dispatch the exception to a registered
+ * fast handler or the default handler, if no fast handler was registered.
+ * The default handler sets up a C-stack and dispatches the exception to a
+ * registerd C handler in the second-level dispatch table.
+ *
+ * Fast handler entry condition:
+ *
+ *   a0:	trashed, original value saved on stack (PT_AREG0)
+ *   a1:	a1
+ *   a2:	new stack pointer, original value in depc
+ *   a3:	dispatch table
+ *   depc:	a2, original value saved on stack (PT_DEPC)
+ *   excsave_1:	a3
+ *
+ * The value for PT_DEPC saved to stack also functions as a boolean to
+ * indicate that the exception is either a double or a regular exception:
+ *
+ *   PT_DEPC	>= VALID_DOUBLE_EXCEPTION_ADDRESS: double exception
+ *		<  VALID_DOUBLE_EXCEPTION_ADDRESS: regular exception
+ *
+ * Note:  Neither the kernel nor the user exception handler generate literals.
+ *
+ */
+
+#include <linux/linkage.h>
+#include <asm/ptrace.h>
+#include <asm/ptrace.h>
+#include <asm/current.h>
+#include <asm/offsets.h>
+#include <asm/pgtable.h>
+#include <asm/processor.h>
+#include <asm/page.h>
+#include <asm/thread_info.h>
+#include <asm/processor.h>
+
+
+/*
+ * User exception vector. (Exceptions with PS.UM == 1, PS.EXCM == 0)
+ *
+ * We get here when an exception occurred while we were in userland.
+ * We switch to the kernel stack and jump to the first level handler
+ * associated to the exception cause.
+ *
+ * Note: the saved kernel stack pointer (EXC_TABLE_KSTK) is already
+ *       decremented by PT_USER_SIZE.
+ */
+
+	.section .UserExceptionVector.text, "ax"
+
+ENTRY(_UserExceptionVector)
+
+	xsr	a3, EXCSAVE_1		# save a3 and get dispatch table
+	wsr	a2, DEPC		# save a2
+	l32i	a2, a3, EXC_TABLE_KSTK	# load kernel stack to a2
+	s32i	a0, a2, PT_AREG0	# save a0 to ESF
+	rsr	a0, EXCCAUSE		# retrieve exception cause
+	s32i	a0, a2, PT_DEPC		# mark it as a regular exception
+	addx4	a0, a0, a3		# find entry in table
+	l32i	a0, a0, EXC_TABLE_FAST_USER	# load handler
+	jx	a0
+
+/*
+ * Kernel exception vector. (Exceptions with PS.UM == 0, PS.EXCM == 0)
+ *
+ * We get this exception when we were already in kernel space.
+ * We decrement the current stack pointer (kernel) by PT_SIZE and
+ * jump to the first-level handler associated with the exception cause.
+ *
+ * Note: we need to preserve space for the spill region.
+ */
+
+	.section .KernelExceptionVector.text, "ax"
+
+ENTRY(_KernelExceptionVector)
+
+	xsr	a3, EXCSAVE_1		# save a3, and get dispatch table
+	wsr	a2, DEPC		# save a2
+	addi	a2, a1, -16-PT_SIZE	# adjust stack pointer
+	s32i	a0, a2, PT_AREG0	# save a0 to ESF
+	rsr	a0, EXCCAUSE		# retrieve exception cause
+	s32i	a0, a2, PT_DEPC		# mark it as a regular exception
+	addx4	a0, a0, a3		# find entry in table
+	l32i	a0, a0, EXC_TABLE_FAST_KERNEL	# load handler address
+	jx	a0
+
+
+/*
+ * Double exception vector (Exceptions with PS.EXCM == 1)
+ * We get this exception when another exception occurs while were are
+ * already in an exception, such as window overflow/underflow exception,
+ * or 'expected' exceptions, for example memory exception when we were trying
+ * to read data from an invalid address in user space.
+ *
+ * Note that this vector is never invoked for level-1 interrupts, because such
+ * interrupts are disabled (masked) when PS.EXCM is set.
+ *
+ * We decode the exception and take the appropriate action.  However, the
+ * double exception vector is much more careful, because a lot more error
+ * cases go through the double exception vector than through the user and
+ * kernel exception vectors.
+ *
+ * Occasionally, the kernel expects a double exception to occur.  This usually
+ * happens when accessing user-space memory with the user's permissions
+ * (l32e/s32e instructions).  The kernel state, though, is not always suitable
+ * for immediate transfer of control to handle_double, where "normal" exception
+ * processing occurs. Also in kernel mode, TLB misses can occur if accessing
+ * vmalloc memory, possibly requiring repair in a double exception handler.
+ *
+ * The variable at TABLE_FIXUP offset from the pointer in EXCSAVE_1 doubles as
+ * a boolean variable and a pointer to a fixup routine. If the variable
+ * EXC_TABLE_FIXUP is non-zero, this handler jumps to that address. A value of
+ * zero indicates to use the default kernel/user exception handler.
+ * There is only one exception, when the value is identical to the exc_table
+ * label, the kernel is in trouble. This mechanism is used to protect critical
+ * sections, mainly when the handler writes to the stack to assert the stack
+ * pointer is valid. Once the fixup/default handler leaves that area, the
+ * EXC_TABLE_FIXUP variable is reset to the fixup handler or zero.
+ *
+ * Procedures wishing to use this mechanism should set EXC_TABLE_FIXUP to the
+ * nonzero address of a fixup routine before it could cause a double exception
+ * and reset it before it returns.
+ *
+ * Some other things to take care of when a fast exception handler doesn't
+ * specify a particular fixup handler but wants to use the default handlers:
+ *
+ *  - The original stack pointer (in a1) must not be modified. The fast
+ *    exception handler should only use a2 as the stack pointer.
+ *
+ *  - If the fast handler manipulates the stack pointer (in a2), it has to
+ *    register a valid fixup handler and cannot use the default handlers.
+ *
+ *  - The handler can use any other generic register from a3 to a15, but it
+ *    must save the content of these registers to stack (PT_AREG3...PT_AREGx)
+ *
+ *  - These registers must be saved before a double exception can occur.
+ *
+ *  - If we ever implement handling signals while in double exceptions, the
+ *    number of registers a fast handler has saved (excluding a0 and a1) must
+ *    be written to  PT_AREG1. (1 if only a3 is used, 2 for a3 and a4, etc. )
+ *
+ * The fixup handlers are special handlers:
+ *
+ *  - Fixup entry conditions differ from regular exceptions:
+ *
+ *	a0:	   DEPC
+ *	a1: 	   a1
+ *	a2:	   trashed, original value in EXC_TABLE_DOUBLE_A2
+ *	a3:	   exctable
+ *	depc:	   a0
+ *	excsave_1: a3
+ *
+ *  - When the kernel enters the fixup handler, it still assumes it is in a
+ *    critical section, so EXC_TABLE_FIXUP variable is set to exc_table.
+ *    The fixup handler, therefore, has to re-register itself as the fixup
+ *    handler before it returns from the double exception.
+ *
+ *  - Fixup handler can share the same exception frame with the fast handler.
+ *    The kernel stack pointer is not changed when entering the fixup handler.
+ *
+ *  - Fixup handlers can jump to the default kernel and user exception
+ *    handlers. Before it jumps, though, it has to setup a exception frame
+ *    on stack. Because the default handler resets the register fixup handler
+ *    the fixup handler must make sure that the default handler returns to
+ *    it instead of the exception address, so it can re-register itself as
+ *    the fixup handler.
+ *
+ * In case of a critical condition where the kernel cannot recover, we jump
+ * to unrecoverable_exception with the following entry conditions.
+ * All registers a0...a15 are unchanged from the last exception, except:
+ *
+ *	a0:	   last address before we jumped to the unrecoverable_exception.
+ *	excsave_1: a0
+ *
+ *
+ * See the handle_alloca_user and spill_registers routines for example clients.
+ *
+ * FIXME: Note: we currently don't allow signal handling coming from a double
+ *        exception, so the item markt with (*) is not required.
+ */
+
+	.section .DoubleExceptionVector.text, "ax"
+	.begin literal_prefix .DoubleExceptionVector
+
+ENTRY(_DoubleExceptionVector)
+
+	/* Deliberately destroy excsave (don't assume it's value was valid). */
+
+	wsr	a3, EXCSAVE_1		# save a3
+
+	/* Check for kernel double exception (usually fatal). */
+
+	rsr	a3, PS
+	_bbci.l	a3, PS_UM_SHIFT, .Lksp
+
+	/* Check if we are currently handling a window exception. */
+	/* Note: We don't need to indicate that we enter a critical section. */
+
+	xsr	a0, DEPC		# get DEPC, save a0
+
+	movi	a3, XCHAL_WINDOW_VECTORS_VADDR
+	_bltu	a0, a3, .Lfixup
+	addi	a3, a3, XSHAL_WINDOW_VECTORS_SIZE
+	_bgeu	a0, a3, .Lfixup
+
+	/* Window overflow/underflow exception. Get stack pointer. */
+
+	mov	a3, a2
+	movi	a2, exc_table
+	l32i	a2, a2, EXC_TABLE_KSTK
+
+	/* Check for overflow/underflow exception, jump if overflow. */
+
+	_bbci.l	a0, 6, .Lovfl
+
+	/* a0: depc, a1: a1, a2: kstk, a3: a2, depc: a0, excsave: a3  */
+
+	/* Restart window underflow exception.
+	 * We return to the instruction in user space that caused the window
+	 * underflow exception. Therefore, we change window base to the value
+	 * before we entered the window underflow exception and prepare the
+	 * registers to return as if we were coming from a regular exception
+	 * by changing depc (in a0).
+	 * Note: We can trash the current window frame (a0...a3) and depc!
+	 */
+
+	wsr	a2, DEPC		# save stack pointer temporarily
+	rsr	a0, PS
+	extui	a0, a0, XCHAL_PS_OWB_SHIFT, XCHAL_PS_OWB_BITS
+	wsr	a0, WINDOWBASE
+	rsync
+
+	/* We are now in the previous window frame. Save registers again. */
+
+	xsr	a2, DEPC		# save a2 and get stack pointer
+	s32i	a0, a2, PT_AREG0
+
+	wsr	a3, EXCSAVE_1		# save a3
+	movi	a3, exc_table
+
+	rsr	a0, EXCCAUSE
+	s32i	a0, a2, PT_DEPC		# mark it as a regular exception
+	addx4	a0, a0, a3
+	l32i	a0, a0, EXC_TABLE_FAST_USER
+	jx	a0
+
+.Lfixup:/* Check for a fixup handler or if we were in a critical section. */
+
+	/* a0: depc, a1: a1, a2: a2, a3: trashed, depc: a0, excsave1: a3 */
+
+	movi	a3, exc_table
+	s32i	a2, a3, EXC_TABLE_DOUBLE_SAVE	# temporary variable
+
+	/* Enter critical section. */
+
+	l32i	a2, a3, EXC_TABLE_FIXUP
+	s32i	a3, a3, EXC_TABLE_FIXUP
+	beq	a2, a3, .Lunrecoverable_fixup	# critical!
+	beqz	a2, .Ldflt			# no handler was registered
+
+	/* a0: depc, a1: a1, a2: trash, a3: exctable, depc: a0, excsave: a3 */
+
+	jx	a2
+
+.Ldflt:	/* Get stack pointer. */
+
+	l32i	a3, a3, EXC_TABLE_DOUBLE_SAVE
+	addi	a2, a3, -PT_USER_SIZE
+
+.Lovfl:	/* Jump to default handlers. */
+
+	/* a0: depc, a1: a1, a2: kstk, a3: a2, depc: a0, excsave: a3 */
+
+	xsr	a3, DEPC
+	s32i	a0, a2, PT_DEPC
+	s32i	a3, a2, PT_AREG0
+
+	/* a0: avail, a1: a1, a2: kstk, a3: avail, depc: a2, excsave: a3 */
+
+	movi	a3, exc_table
+	rsr	a0, EXCCAUSE
+	addx4	a0, a0, a3
+	l32i	a0, a0, EXC_TABLE_FAST_USER
+	jx	a0
+
+	/*
+	 * We only allow the ITLB miss exception if we are in kernel space.
+	 * All other exceptions are unexpected and thus unrecoverable!
+	 */
+
+	.extern fast_second_level_miss_double_kernel
+
+.Lksp:	/* a0: a0, a1: a1, a2: a2, a3: trashed, depc: depc, excsave: a3 */
+
+	rsr	a3, EXCCAUSE
+	beqi	a3, XCHAL_EXCCAUSE_ITLB_MISS, 1f
+	addi	a3, a3, -XCHAL_EXCCAUSE_DTLB_MISS
+	bnez	a3, .Lunrecoverable
+1:	movi	a3, fast_second_level_miss_double_kernel
+	jx	a3
+
+	/* Critical! We can't handle this situation. PANIC! */
+
+	.extern unrecoverable_exception
+
+.Lunrecoverable_fixup:
+	l32i	a2, a3, EXC_TABLE_DOUBLE_SAVE
+	xsr	a0, DEPC
+
+.Lunrecoverable:
+	rsr	a3, EXCSAVE_1
+	wsr	a0, EXCSAVE_1
+	movi	a0, unrecoverable_exception
+	callx0	a0
+
+	.end literal_prefix
+
+
+/*
+ * Debug interrupt vector
+ *
+ * There is not much space here, so simply jump to another handler.
+ * EXCSAVE[DEBUGLEVEL] has been set to that handler.
+ */
+
+	.section .DebugInterruptVector.text, "ax"
+
+ENTRY(_DebugInterruptVector)
+	xsr	a0, EXCSAVE + XCHAL_DEBUGLEVEL
+	jx	a0
+
+
+
+/* Window overflow and underflow handlers.
+ * The handlers must be 64 bytes apart, first starting with the underflow
+ * handlers underflow-4 to underflow-12, then the overflow handlers
+ * overflow-4 to overflow-12.
+ *
+ * Note: We rerun the underflow handlers if we hit an exception, so
+ *	 we try to access any page that would cause a page fault early.
+ */
+
+	.section		.WindowVectors.text, "ax"
+
+
+/* 4-Register Window Overflow Vector (Handler) */
+
+	.align 64
+.global _WindowOverflow4
+_WindowOverflow4:
+	s32e	a0, a5, -16
+	s32e	a1, a5, -12
+	s32e	a2, a5,  -8
+	s32e	a3, a5,  -4
+	rfwo
+
+
+/* 4-Register Window Underflow Vector (Handler) */
+
+	.align 64
+.global _WindowUnderflow4
+_WindowUnderflow4:
+	l32e	a0, a5, -16
+	l32e	a1, a5, -12
+	l32e	a2, a5,  -8
+	l32e	a3, a5,  -4
+	rfwu
+
+
+/* 8-Register Window Overflow Vector (Handler) */
+
+	.align 64
+.global _WindowOverflow8
+_WindowOverflow8:
+	s32e	a0, a9, -16
+	l32e	a0, a1, -12
+	s32e	a2, a9,  -8
+	s32e	a1, a9, -12
+	s32e	a3, a9,  -4
+	s32e	a4, a0, -32
+	s32e	a5, a0, -28
+	s32e	a6, a0, -24
+	s32e	a7, a0, -20
+	rfwo
+
+/* 8-Register Window Underflow Vector (Handler) */
+
+	.align 64
+.global _WindowUnderflow8
+_WindowUnderflow8:
+	l32e	a1, a9, -12
+	l32e	a0, a9, -16
+	l32e	a7, a1, -12
+	l32e	a2, a9,  -8
+	l32e	a4, a7, -32
+	l32e	a3, a9,  -4
+	l32e	a5, a7, -28
+	l32e	a6, a7, -24
+	l32e	a7, a7, -20
+	rfwu
+
+
+/* 12-Register Window Overflow Vector (Handler) */
+
+	.align 64
+.global _WindowOverflow12
+_WindowOverflow12:
+	s32e	a0,  a13, -16
+	l32e	a0,  a1,  -12
+	s32e	a1,  a13, -12
+	s32e	a2,  a13,  -8
+	s32e	a3,  a13,  -4
+	s32e	a4,  a0,  -48
+	s32e	a5,  a0,  -44
+	s32e	a6,  a0,  -40
+	s32e	a7,  a0,  -36
+	s32e	a8,  a0,  -32
+	s32e	a9,  a0,  -28
+	s32e	a10, a0,  -24
+	s32e	a11, a0,  -20
+	rfwo
+
+/* 12-Register Window Underflow Vector (Handler) */
+
+	.align 64
+.global _WindowUnderflow12
+_WindowUnderflow12:
+	l32e	a1,  a13, -12
+	l32e	a0,  a13, -16
+	l32e	a11, a1,  -12
+	l32e	a2,  a13,  -8
+	l32e	a4,  a11, -48
+	l32e	a8,  a11, -32
+	l32e	a3,  a13,  -4
+	l32e	a5,  a11, -44
+	l32e	a6,  a11, -40
+	l32e	a7,  a11, -36
+	l32e	a9,  a11, -28
+	l32e	a10, a11, -24
+	l32e	a11, a11, -20
+	rfwu
+
+	.text
+
+
diff --git a/arch/xtensa/kernel/vmlinux.lds.S b/arch/xtensa/kernel/vmlinux.lds.S
new file mode 100644
index 00000000000..476b2b53cd0
--- /dev/null
+++ b/arch/xtensa/kernel/vmlinux.lds.S
@@ -0,0 +1,341 @@
+/*
+ * arch/xtensa/kernel/vmlinux.lds.S
+ *
+ * Xtensa linker script
+ *
+ * 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.
+ *
+ * Chris Zankel <chris@zankel.net>
+ * Marc Gauthier <marc@tensilica.com, marc@alumni.uwaterloo.ca>
+ * Joe Taylor <joe@tensilica.com, joetylr@yahoo.com>
+ */
+
+#include <asm-generic/vmlinux.lds.h>
+
+#include <linux/config.h>
+#define _NOCLANGUAGE
+#include <xtensa/config/core.h>
+#include <xtensa/config/system.h>
+OUTPUT_ARCH(xtensa)
+ENTRY(_start)
+
+#if XCHAL_MEMORY_ORDER == XTHAL_BIGENDIAN
+jiffies = jiffies_64 + 4;
+#else
+jiffies = jiffies_64;
+#endif
+
+#define KERNELOFFSET 0x1000
+
+/* Note: In the following macros, it would be nice to specify only the
+   vector name and section kind and construct "sym" and "section" using
+   CPP concatenation, but that does not work reliably.  Concatenating a
+   string with "." produces an invalid token.  CPP will not print a
+   warning because it thinks this is an assembly file, but it leaves
+   them as multiple tokens and there may or may not be whitespace
+   between them.  */
+
+/* Macro for a relocation entry */
+
+#define RELOCATE_ENTRY(sym, section)		\
+	LONG(sym ## _start);			\
+	LONG(sym ## _end);			\
+	LONG(LOADADDR(section))
+
+/* Macro to define a section for a vector.
+ *
+ * Use of the MIN function catches the types of errors illustrated in
+ * the following example:
+ *
+ * Assume the section .DoubleExceptionVector.literal is completely
+ * full.  Then a programmer adds code to .DoubleExceptionVector.text
+ * that produces another literal.  The final literal position will
+ * overlay onto the first word of the adjacent code section
+ * .DoubleExceptionVector.text.  (In practice, the literals will
+ * overwrite the code, and the first few instructions will be
+ * garbage.)
+ */
+
+#define SECTION_VECTOR(sym, section, addr, max_prevsec_size, prevsec)       \
+  section addr : AT((MIN(LOADADDR(prevsec) + max_prevsec_size,		    \
+		         LOADADDR(prevsec) + SIZEOF(prevsec)) + 3) & ~ 3)   \
+  {									    \
+    . = ALIGN(4);							    \
+    sym ## _start = ABSOLUTE(.);		 			    \
+    *(section)								    \
+    sym ## _end = ABSOLUTE(.);						    \
+  }
+
+/*
+ *  Mapping of input sections to output sections when linking.
+ */
+
+SECTIONS
+{
+  . = XCHAL_KSEG_CACHED_VADDR + KERNELOFFSET;
+  /* .text section */
+
+  _text = .;
+  _stext = .;
+  _ftext = .;
+
+  .text :
+  {
+    /* The .head.text section must be the first section! */
+    *(.head.text)
+    *(.literal .text)
+    *(.srom.text)
+    VMLINUX_SYMBOL(__sched_text_start) = .;
+    *(.sched.text.literal .sched.text)
+    VMLINUX_SYMBOL(__sched_text_end) = .;
+    VMLINUX_SYMBOL(__lock_text_start) = .;
+    *(.spinlock.text.literal .spinlock.text)
+    VMLINUX_SYMBOL(__lock_text_end) = .;
+
+  }
+  _etext = .;
+
+  . = ALIGN(16);
+
+  RODATA
+
+  /*  Relocation table */
+
+  . = ALIGN(16);
+  __boot_reloc_table_start = ABSOLUTE(.);
+
+  __relocate : {
+
+    RELOCATE_ENTRY(_WindowVectors_text,
+		   .WindowVectors.text);
+#if 0
+    RELOCATE_ENTRY(_KernelExceptionVector_literal,
+		   .KernelExceptionVector.literal);
+#endif
+    RELOCATE_ENTRY(_KernelExceptionVector_text,
+		   .KernelExceptionVector.text);
+#if 0
+    RELOCATE_ENTRY(_UserExceptionVector_literal,
+		   .UserExceptionVector.literal);
+#endif
+    RELOCATE_ENTRY(_UserExceptionVector_text,
+		   .UserExceptionVector.text);
+    RELOCATE_ENTRY(_DoubleExceptionVector_literal,
+		   .DoubleExceptionVector.literal);
+    RELOCATE_ENTRY(_DoubleExceptionVector_text,
+		   .DoubleExceptionVector.text);
+  }
+  __boot_reloc_table_end = ABSOLUTE(.) ;
+
+  .fixup   : { *(.fixup) }
+
+  . = ALIGN(16);
+
+  __ex_table : {
+    __start___ex_table = .;
+    *(__ex_table)
+    __stop___ex_table = .;
+  }
+
+  /* Data section */
+
+  . = ALIGN(XCHAL_ICACHE_LINESIZE);
+  _fdata = .;
+  .data :
+  {
+    *(.data) CONSTRUCTORS
+    . = ALIGN(XCHAL_ICACHE_LINESIZE);
+    *(.data.cacheline_aligned)
+  }
+
+  _edata = .;
+
+  /* The initial task */
+  . = ALIGN(8192);
+  .data.init_task : { *(.data.init_task) }
+
+  /* Initialization code and data: */
+
+  . = ALIGN(1<<XCHAL_MMU_MIN_PTE_PAGE_SIZE);
+  __init_begin = .;
+  .init.text : {
+  	_sinittext = .;
+	*(.init.text.literal) *(.init.text)
+	_einittext = .;
+  }
+
+  .init.data :
+  {
+    *(.init.data)
+    . = ALIGN(0x4);
+    __tagtable_begin = .;
+    *(.taglist)
+    __tagtable_end = .;
+  }
+
+  . = ALIGN(XCHAL_ICACHE_LINESIZE);
+
+  __setup_start = .;
+  .init.setup : { *(.init.setup) }
+  __setup_end = .;
+
+  __initcall_start = .;
+  .initcall.init : {
+  	*(.initcall1.init)
+  	*(.initcall2.init)
+  	*(.initcall3.init)
+  	*(.initcall4.init)
+  	*(.initcall5.init)
+  	*(.initcall6.init)
+  	*(.initcall7.init)
+  }
+  __initcall_end = .;
+
+  __con_initcall_start = .;
+  .con_initcall.init : { *(.con_initcall.init) }
+  __con_initcall_end = .;
+
+  SECURITY_INIT
+
+  . = ALIGN(4);
+
+  __start___ftr_fixup = .;
+  __ftr_fixup : { *(__ftr_fixup) }
+  __stop___ftr_fixup = .;
+
+  . = ALIGN(32);
+  __per_cpu_start = .;
+  .data.percpu  : { *(.data.percpu) }
+  __per_cpu_end = .;
+
+  . = ALIGN(4096);
+  __initramfs_start =.;
+  .init.ramfs : { *(.init.ramfs) }
+  __initramfs_end = .;
+
+  /* We need this dummy segment here */
+
+  . = ALIGN(4);
+  .dummy : { LONG(0) }
+
+  /* The vectors are relocated to the real position at startup time */
+
+  SECTION_VECTOR (_WindowVectors_text,
+		  .WindowVectors.text,
+		  XCHAL_WINDOW_VECTORS_VADDR, 4,
+		  .dummy)
+  SECTION_VECTOR (_DebugInterruptVector_literal,
+		  .DebugInterruptVector.literal,
+		  XCHAL_INTLEVEL_VECTOR_VADDR(XCHAL_DEBUGLEVEL) - 4,
+		  SIZEOF(.WindowVectors.text),
+		  .WindowVectors.text)
+  SECTION_VECTOR (_DebugInterruptVector_text,
+		  .DebugInterruptVector.text,
+		  XCHAL_INTLEVEL_VECTOR_VADDR(XCHAL_DEBUGLEVEL),
+		  4,
+		  .DebugInterruptVector.literal)
+  SECTION_VECTOR (_KernelExceptionVector_literal,
+		  .KernelExceptionVector.literal,
+		  XCHAL_KERNELEXC_VECTOR_VADDR - 4,
+		  SIZEOF(.DebugInterruptVector.text),
+		  .DebugInterruptVector.text)
+  SECTION_VECTOR (_KernelExceptionVector_text,
+		  .KernelExceptionVector.text,
+		  XCHAL_KERNELEXC_VECTOR_VADDR,
+		  4,
+		  .KernelExceptionVector.literal)
+  SECTION_VECTOR (_UserExceptionVector_literal,
+		  .UserExceptionVector.literal,
+		  XCHAL_USEREXC_VECTOR_VADDR - 4,
+		  SIZEOF(.KernelExceptionVector.text),
+		  .KernelExceptionVector.text)
+  SECTION_VECTOR (_UserExceptionVector_text,
+		  .UserExceptionVector.text,
+		  XCHAL_USEREXC_VECTOR_VADDR,
+		  4,
+		  .UserExceptionVector.literal)
+  SECTION_VECTOR (_DoubleExceptionVector_literal,
+		  .DoubleExceptionVector.literal,
+		  XCHAL_DOUBLEEXC_VECTOR_VADDR - 16,
+		  SIZEOF(.UserExceptionVector.text),
+		  .UserExceptionVector.text)
+  SECTION_VECTOR (_DoubleExceptionVector_text,
+		  .DoubleExceptionVector.text,
+		  XCHAL_DOUBLEEXC_VECTOR_VADDR,
+		  32,
+		  .DoubleExceptionVector.literal)
+
+  . = (LOADADDR( .DoubleExceptionVector.text ) + SIZEOF( .DoubleExceptionVector.text ) + 3) & ~ 3;
+  . = ALIGN(1<<XCHAL_MMU_MIN_PTE_PAGE_SIZE);
+
+  __init_end = .;
+
+  . = ALIGN(8192);
+
+  /* BSS section */
+  _bss_start = .;
+  .sbss : { *(.sbss) *(.scommon) }
+  .bss : { *(COMMON) *(.bss) }
+  _bss_end = .;
+  _end = .;
+
+  /* only used by the boot loader  */
+
+  . = ALIGN(0x10);
+  .bootstrap : { *(.bootstrap.literal .bootstrap.text .bootstrap.data) }
+
+  . = ALIGN(0x1000);
+  __initrd_start = .;
+  .initrd : { *(.initrd) }
+  __initrd_end = .;
+
+  .ResetVector.text XCHAL_RESET_VECTOR_VADDR :
+  {
+    *(.ResetVector.text)
+  }
+
+
+  /* Sections to be discarded */
+  /DISCARD/ :
+  {
+        *(.text.exit)
+	*(.text.exit.literal)
+        *(.data.exit)
+        *(.exitcall.exit)
+  }
+
+
+  .debug  0 :  { *(.debug) }
+  .line  0 :  { *(.line) }
+  .debug_srcinfo  0 :  { *(.debug_srcinfo) }
+  .debug_sfnames  0 :  { *(.debug_sfnames) }
+  .debug_aranges  0 :  { *(.debug_aranges) }
+  .debug_pubnames  0 :  { *(.debug_pubnames) }
+  .debug_info  0 :  { *(.debug_info) }
+  .debug_abbrev  0 :  { *(.debug_abbrev) }
+  .debug_line  0 :  { *(.debug_line) }
+  .debug_frame  0 :  { *(.debug_frame) }
+  .debug_str  0 :  { *(.debug_str) }
+  .debug_loc  0 :  { *(.debug_loc) }
+  .debug_macinfo  0 :  { *(.debug_macinfo) }
+  .debug_weaknames  0 :  { *(.debug_weaknames) }
+  .debug_funcnames  0 :  { *(.debug_funcnames) }
+  .debug_typenames  0 :  { *(.debug_typenames) }
+  .debug_varnames  0 :  { *(.debug_varnames) }
+
+  .xt.insn 0 :
+  {
+    *(.xt.insn)
+    *(.gnu.linkonce.x*)
+  }
+
+  .xt.lit 0 :
+  {
+    *(.xt.lit)
+    *(.gnu.linkonce.p*)
+  }
+}
diff --git a/arch/xtensa/kernel/xtensa_ksyms.c b/arch/xtensa/kernel/xtensa_ksyms.c
new file mode 100644
index 00000000000..efae56a5147
--- /dev/null
+++ b/arch/xtensa/kernel/xtensa_ksyms.c
@@ -0,0 +1,123 @@
+/*
+ * arch/xtensa/kernel/xtensa_ksyms.c
+ *
+ * Export Xtensa-specific functions for loadable modules.
+ *
+ * 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.
+ *
+ * Joe Taylor <joe@tensilica.com>
+ */
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/string.h>
+#include <linux/mm.h>
+#include <linux/interrupt.h>
+#include <asm/irq.h>
+#include <linux/in6.h>
+#include <linux/pci.h>
+#include <linux/ide.h>
+
+#include <asm/uaccess.h>
+#include <asm/checksum.h>
+#include <asm/dma.h>
+#include <asm/io.h>
+#include <asm/page.h>
+#include <asm/pgalloc.h>
+#include <asm/semaphore.h>
+#ifdef CONFIG_BLK_DEV_FD
+#include <asm/floppy.h>
+#endif
+#ifdef CONFIG_NET
+#include <net/checksum.h>
+#endif /* CONFIG_NET */
+
+
+/*
+ * String functions
+ */
+EXPORT_SYMBOL(memcmp);
+EXPORT_SYMBOL(memset);
+EXPORT_SYMBOL(memcpy);
+EXPORT_SYMBOL(memmove);
+EXPORT_SYMBOL(memchr);
+EXPORT_SYMBOL(strcat);
+EXPORT_SYMBOL(strchr);
+EXPORT_SYMBOL(strlen);
+EXPORT_SYMBOL(strpbrk);
+EXPORT_SYMBOL(strncat);
+EXPORT_SYMBOL(strnlen);
+EXPORT_SYMBOL(strrchr);
+EXPORT_SYMBOL(strstr);
+
+EXPORT_SYMBOL(enable_irq);
+EXPORT_SYMBOL(disable_irq);
+EXPORT_SYMBOL(kernel_thread);
+
+/*
+ * gcc internal math functions
+ */
+extern long long __ashrdi3(long long, int);
+extern long long __ashldi3(long long, int);
+extern long long __lshrdi3(long long, int);
+extern int __divsi3(int, int);
+extern int __modsi3(int, int);
+extern long long __muldi3(long long, long long);
+extern int __mulsi3(int, int);
+extern unsigned int __udivsi3(unsigned int, unsigned int);
+extern unsigned int __umodsi3(unsigned int, unsigned int);
+extern unsigned long long __umoddi3(unsigned long long, unsigned long long);
+extern unsigned long long __udivdi3(unsigned long long, unsigned long long);
+
+EXPORT_SYMBOL(__ashldi3);
+EXPORT_SYMBOL(__ashrdi3);
+EXPORT_SYMBOL(__lshrdi3);
+EXPORT_SYMBOL(__divsi3);
+EXPORT_SYMBOL(__modsi3);
+EXPORT_SYMBOL(__muldi3);
+EXPORT_SYMBOL(__mulsi3);
+EXPORT_SYMBOL(__udivsi3);
+EXPORT_SYMBOL(__umodsi3);
+EXPORT_SYMBOL(__udivdi3);
+EXPORT_SYMBOL(__umoddi3);
+
+/*
+ * Semaphore operations
+ */
+EXPORT_SYMBOL(__down);
+EXPORT_SYMBOL(__down_interruptible);
+EXPORT_SYMBOL(__down_trylock);
+EXPORT_SYMBOL(__up);
+
+#ifdef CONFIG_NET
+/*
+ * Networking support
+ */
+EXPORT_SYMBOL(csum_partial_copy_generic);
+#endif /* CONFIG_NET */
+
+/*
+ * Architecture-specific symbols
+ */
+EXPORT_SYMBOL(__xtensa_copy_user);
+
+/*
+ * Kernel hacking ...
+ */
+
+#if defined(CONFIG_VGA_CONSOLE) || defined(CONFIG_DUMMY_CONSOLE)
+// FIXME EXPORT_SYMBOL(screen_info);
+#endif
+
+EXPORT_SYMBOL(get_wchan);
+
+EXPORT_SYMBOL(outsb);
+EXPORT_SYMBOL(outsw);
+EXPORT_SYMBOL(outsl);
+EXPORT_SYMBOL(insb);
+EXPORT_SYMBOL(insw);
+EXPORT_SYMBOL(insl);
-- 
cgit v1.2.3-70-g09d2


From 249ac17e96811acc3c6402317dd5d5c89d2cbf68 Mon Sep 17 00:00:00 2001
From: Chris Zankel <czankel@tensilica.com>
Date: Thu, 23 Jun 2005 22:01:20 -0700
Subject: [PATCH] xtensa: Architecture support for Tensilica Xtensa Part 4

The attached patches provides part 4 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/lib/Makefile       |   7 +
 arch/xtensa/lib/checksum.S     | 410 +++++++++++++++++++++++++++++++++++++++++
 arch/xtensa/lib/memcopy.S      | 315 +++++++++++++++++++++++++++++++
 arch/xtensa/lib/memset.S       | 160 ++++++++++++++++
 arch/xtensa/lib/pci-auto.c     | 352 +++++++++++++++++++++++++++++++++++
 arch/xtensa/lib/strcasecmp.c   |  32 ++++
 arch/xtensa/lib/strncpy_user.S | 224 ++++++++++++++++++++++
 arch/xtensa/lib/strnlen_user.S | 147 +++++++++++++++
 arch/xtensa/lib/usercopy.S     | 321 ++++++++++++++++++++++++++++++++
 9 files changed, 1968 insertions(+)
 create mode 100644 arch/xtensa/lib/Makefile
 create mode 100644 arch/xtensa/lib/checksum.S
 create mode 100644 arch/xtensa/lib/memcopy.S
 create mode 100644 arch/xtensa/lib/memset.S
 create mode 100644 arch/xtensa/lib/pci-auto.c
 create mode 100644 arch/xtensa/lib/strcasecmp.c
 create mode 100644 arch/xtensa/lib/strncpy_user.S
 create mode 100644 arch/xtensa/lib/strnlen_user.S
 create mode 100644 arch/xtensa/lib/usercopy.S

(limited to 'arch')

diff --git a/arch/xtensa/lib/Makefile b/arch/xtensa/lib/Makefile
new file mode 100644
index 00000000000..ed935b58e8a
--- /dev/null
+++ b/arch/xtensa/lib/Makefile
@@ -0,0 +1,7 @@
+#
+# Makefile for Xtensa-specific library files.
+#
+
+lib-y	+= memcopy.o memset.o checksum.o strcasecmp.o \
+	   usercopy.o strncpy_user.o strnlen_user.o
+lib-$(CONFIG_PCI) += pci-auto.o
diff --git a/arch/xtensa/lib/checksum.S b/arch/xtensa/lib/checksum.S
new file mode 100644
index 00000000000..e2d64dfd530
--- /dev/null
+++ b/arch/xtensa/lib/checksum.S
@@ -0,0 +1,410 @@
+/*
+ * INET		An implementation of the TCP/IP protocol suite for the LINUX
+ *		operating system.  INET is implemented using the  BSD Socket
+ *		interface as the means of communication with the user level.
+ *
+ *		IP/TCP/UDP checksumming routines
+ *
+ * Xtensa version:  Copyright (C) 2001 Tensilica, Inc. by Kevin Chea
+ *                  Optimized by Joe Taylor
+ *
+ *		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 <asm/errno.h>
+#include <linux/linkage.h>
+#define _ASMLANGUAGE
+#include <xtensa/config/core.h>
+
+/*
+ * computes a partial checksum, e.g. for TCP/UDP fragments
+ */
+
+/*
+ * unsigned int csum_partial(const unsigned char *buf, int len,
+ *                           unsigned int sum);
+ *    a2 = buf
+ *    a3 = len
+ *    a4 = sum
+ *
+ * This function assumes 2- or 4-byte alignment.  Other alignments will fail!
+ */
+
+/* ONES_ADD converts twos-complement math to ones-complement. */
+#define ONES_ADD(sum, val)	  \
+	add	sum, sum, val	; \
+	bgeu	sum, val, 99f	; \
+	addi	sum, sum, 1	; \
+99:				;
+
+.text
+ENTRY(csum_partial)
+	  /*
+	   * Experiments with Ethernet and SLIP connections show that buf
+	   * is aligned on either a 2-byte or 4-byte boundary.
+	   */
+	entry	sp, 32
+	extui	a5, a2, 0, 2
+	bnez	a5, 8f		/* branch if 2-byte aligned */
+	/* Fall-through on common case, 4-byte alignment */
+1:
+	srli	a5, a3, 5	/* 32-byte chunks */
+#if XCHAL_HAVE_LOOPS
+	loopgtz	a5, 2f
+#else
+	beqz	a5, 2f
+	slli	a5, a5, 5
+	add	a5, a5, a2	/* a5 = end of last 32-byte chunk */
+.Loop1:
+#endif
+	l32i	a6, a2, 0
+	l32i	a7, a2, 4
+	ONES_ADD(a4, a6)
+	ONES_ADD(a4, a7)
+	l32i	a6, a2, 8
+	l32i	a7, a2, 12
+	ONES_ADD(a4, a6)
+	ONES_ADD(a4, a7)
+	l32i	a6, a2, 16
+	l32i	a7, a2, 20
+	ONES_ADD(a4, a6)
+	ONES_ADD(a4, a7)
+	l32i	a6, a2, 24
+	l32i	a7, a2, 28
+	ONES_ADD(a4, a6)
+	ONES_ADD(a4, a7)
+	addi	a2, a2, 4*8
+#if !XCHAL_HAVE_LOOPS
+	blt	a2, a5, .Loop1
+#endif
+2:
+	extui	a5, a3, 2, 3	/* remaining 4-byte chunks */
+#if XCHAL_HAVE_LOOPS
+	loopgtz	a5, 3f
+#else
+	beqz	a5, 3f
+	slli	a5, a5, 2
+	add	a5, a5, a2	/* a5 = end of last 4-byte chunk */
+.Loop2:
+#endif
+	l32i	a6, a2, 0
+	ONES_ADD(a4, a6)
+	addi	a2, a2, 4
+#if !XCHAL_HAVE_LOOPS
+	blt	a2, a5, .Loop2
+#endif
+3:
+	_bbci.l	a3, 1, 5f	/* remaining 2-byte chunk */
+	l16ui	a6, a2, 0
+	ONES_ADD(a4, a6)
+	addi	a2, a2, 2
+5:
+	_bbci.l	a3, 0, 7f	/* remaining 1-byte chunk */
+6:	l8ui	a6, a2, 0
+#ifdef __XTENSA_EB__
+	slli	a6, a6, 8	/* load byte into bits 8..15 */
+#endif
+	ONES_ADD(a4, a6)
+7:
+	mov	a2, a4
+	retw
+
+	/* uncommon case, buf is 2-byte aligned */
+8:
+	beqz	a3, 7b		/* branch if len == 0 */
+	beqi	a3, 1, 6b	/* branch if len == 1 */
+
+	extui	a5, a2, 0, 1
+	bnez	a5, 8f		/* branch if 1-byte aligned */
+
+	l16ui	a6, a2, 0	/* common case, len >= 2 */
+	ONES_ADD(a4, a6)
+	addi	a2, a2, 2	/* adjust buf */
+	addi	a3, a3, -2	/* adjust len */
+	j	1b		/* now buf is 4-byte aligned */
+
+	/* case: odd-byte aligned, len > 1
+	 * This case is dog slow, so don't give us an odd address.
+	 * (I don't think this ever happens, but just in case.)
+	 */
+8:
+	srli	a5, a3, 2	/* 4-byte chunks */
+#if XCHAL_HAVE_LOOPS
+	loopgtz	a5, 2f
+#else
+	beqz	a5, 2f
+	slli	a5, a5, 2
+	add	a5, a5, a2	/* a5 = end of last 4-byte chunk */
+.Loop3:
+#endif
+	l8ui	a6, a2, 0	/* bits 24..31 */
+	l16ui	a7, a2, 1	/* bits  8..23 */
+	l8ui	a8, a2, 3	/* bits  0.. 8 */
+#ifdef	__XTENSA_EB__
+	slli	a6, a6, 24
+#else
+	slli	a8, a8, 24
+#endif
+	slli	a7, a7, 8
+	or	a7, a7, a6
+	or	a7, a7, a8
+	ONES_ADD(a4, a7)
+	addi	a2, a2, 4
+#if !XCHAL_HAVE_LOOPS
+	blt	a2, a5, .Loop3
+#endif
+2:
+	_bbci.l	a3, 1, 3f	/* remaining 2-byte chunk, still odd addr */
+	l8ui	a6, a2, 0
+	l8ui	a7, a2, 1
+#ifdef	__XTENSA_EB__
+	slli	a6, a6, 8
+#else
+	slli	a7, a7, 8
+#endif
+	or	a7, a7, a6
+	ONES_ADD(a4, a7)
+	addi	a2, a2, 2
+3:
+	j	5b		/* branch to handle the remaining byte */
+
+
+
+/*
+ * Copy from ds while checksumming, otherwise like csum_partial
+ *
+ * The macros SRC and DST specify the type of access for the instruction.
+ * thus we can call a custom exception handler for each access type.
+ */
+
+#define SRC(y...)			\
+	9999: y;			\
+	.section __ex_table, "a";	\
+	.long 9999b, 6001f	;	\
+	.previous
+
+#define DST(y...)			\
+	9999: y;			\
+	.section __ex_table, "a";	\
+	.long 9999b, 6002f	;	\
+	.previous
+
+/*
+unsigned int csum_partial_copy_generic (const char *src, char *dst, int len,
+					int sum, int *src_err_ptr, int *dst_err_ptr)
+	a2  = src
+	a3  = dst
+	a4  = len
+	a5  = sum
+	a6  = src_err_ptr
+	a7  = dst_err_ptr
+	a8  = temp
+	a9  = temp
+	a10 = temp
+	a11 = original len for exception handling
+	a12 = original dst for exception handling
+
+    This function is optimized for 4-byte aligned addresses.  Other
+    alignments work, but not nearly as efficiently.
+ */
+
+ENTRY(csum_partial_copy_generic)
+	entry	sp, 32
+	mov	a12, a3
+	mov	a11, a4
+	or	a10, a2, a3
+
+	/* We optimize the following alignment tests for the 4-byte
+	aligned case.  Two bbsi.l instructions might seem more optimal
+	(commented out below).  However, both labels 5: and 3: are out
+	of the imm8 range, so the assembler relaxes them into
+	equivalent bbci.l, j combinations, which is actually
+	slower. */
+
+	extui	a9, a10, 0, 2
+	beqz	a9, 1f		/* branch if both are 4-byte aligned */
+	bbsi.l	a10, 0, 5f	/* branch if one address is odd */
+	j	3f		/* one address is 2-byte aligned */
+
+/*	_bbsi.l	a10, 0, 5f */	/* branch if odd address */
+/*	_bbsi.l	a10, 1, 3f */	/* branch if 2-byte-aligned address */
+
+1:
+	/* src and dst are both 4-byte aligned */
+	srli	a10, a4, 5	/* 32-byte chunks */
+#if XCHAL_HAVE_LOOPS
+	loopgtz	a10, 2f
+#else
+	beqz	a10, 2f
+	slli	a10, a10, 5
+	add	a10, a10, a2	/* a10 = end of last 32-byte src chunk */
+.Loop5:
+#endif
+SRC(	l32i	a9, a2, 0	)
+SRC(	l32i	a8, a2, 4	)
+DST(	s32i	a9, a3, 0	)
+DST(	s32i	a8, a3, 4	)
+	ONES_ADD(a5, a9)
+	ONES_ADD(a5, a8)
+SRC(	l32i	a9, a2, 8	)
+SRC(	l32i	a8, a2, 12	)
+DST(	s32i	a9, a3, 8	)
+DST(	s32i	a8, a3, 12	)
+	ONES_ADD(a5, a9)
+	ONES_ADD(a5, a8)
+SRC(	l32i	a9, a2, 16	)
+SRC(	l32i	a8, a2, 20	)
+DST(	s32i	a9, a3, 16	)
+DST(	s32i	a8, a3, 20	)
+	ONES_ADD(a5, a9)
+	ONES_ADD(a5, a8)
+SRC(	l32i	a9, a2, 24	)
+SRC(	l32i	a8, a2, 28	)
+DST(	s32i	a9, a3, 24	)
+DST(	s32i	a8, a3, 28	)
+	ONES_ADD(a5, a9)
+	ONES_ADD(a5, a8)
+	addi	a2, a2, 32
+	addi	a3, a3, 32
+#if !XCHAL_HAVE_LOOPS
+	blt	a2, a10, .Loop5
+#endif
+2:
+	extui	a10, a4, 2, 3	/* remaining 4-byte chunks */
+	extui	a4, a4, 0, 2	/* reset len for general-case, 2-byte chunks */
+#if XCHAL_HAVE_LOOPS
+	loopgtz	a10, 3f
+#else
+	beqz	a10, 3f
+	slli	a10, a10, 2
+	add	a10, a10, a2	/* a10 = end of last 4-byte src chunk */
+.Loop6:
+#endif
+SRC(	l32i	a9, a2, 0	)
+DST(	s32i	a9, a3, 0	)
+	ONES_ADD(a5, a9)
+	addi	a2, a2, 4
+	addi	a3, a3, 4
+#if !XCHAL_HAVE_LOOPS
+	blt	a2, a10, .Loop6
+#endif
+3:
+	/*
+	Control comes to here in two cases: (1) It may fall through
+	to here from the 4-byte alignment case to process, at most,
+	one 2-byte chunk.  (2) It branches to here from above if
+	either src or dst is 2-byte aligned, and we process all bytes
+	here, except for perhaps a trailing odd byte.  It's
+	inefficient, so align your addresses to 4-byte boundaries.
+
+	a2 = src
+	a3 = dst
+	a4 = len
+	a5 = sum
+	*/
+	srli	a10, a4, 1	/* 2-byte chunks */
+#if XCHAL_HAVE_LOOPS
+	loopgtz	a10, 4f
+#else
+	beqz	a10, 4f
+	slli	a10, a10, 1
+	add	a10, a10, a2	/* a10 = end of last 2-byte src chunk */
+.Loop7:
+#endif
+SRC(	l16ui	a9, a2, 0	)
+DST(	s16i	a9, a3, 0	)
+	ONES_ADD(a5, a9)
+	addi	a2, a2, 2
+	addi	a3, a3, 2
+#if !XCHAL_HAVE_LOOPS
+	blt	a2, a10, .Loop7
+#endif
+4:
+	/* This section processes a possible trailing odd byte. */
+	_bbci.l	a4, 0, 8f	/* 1-byte chunk */
+SRC(	l8ui	a9, a2, 0	)
+DST(	s8i	a9, a3, 0	)
+#ifdef __XTENSA_EB__
+	slli	a9, a9, 8	/* shift byte to bits 8..15 */
+#endif
+	ONES_ADD(a5, a9)
+8:
+	mov	a2, a5
+	retw
+
+5:
+	/* Control branch to here when either src or dst is odd.  We
+	process all bytes using 8-bit accesses.  Grossly inefficient,
+	so don't feed us an odd address. */
+
+	srli	a10, a4, 1	/* handle in pairs for 16-bit csum */
+#if XCHAL_HAVE_LOOPS
+	loopgtz	a10, 6f
+#else
+	beqz	a10, 6f
+	slli	a10, a10, 1
+	add	a10, a10, a2	/* a10 = end of last odd-aligned, 2-byte src chunk */
+.Loop8:
+#endif
+SRC(	l8ui	a9, a2, 0	)
+SRC(	l8ui	a8, a2, 1	)
+DST(	s8i	a9, a3, 0	)
+DST(	s8i	a8, a3, 1	)
+#ifdef __XTENSA_EB__
+	slli	a9, a9, 8	/* combine into a single 16-bit value */
+#else				/* for checksum computation */
+	slli	a8, a8, 8
+#endif
+	or	a9, a9, a8
+	ONES_ADD(a5, a9)
+	addi	a2, a2, 2
+	addi	a3, a3, 2
+#if !XCHAL_HAVE_LOOPS
+	blt	a2, a10, .Loop8
+#endif
+6:
+	j	4b		/* process the possible trailing odd byte */
+
+
+# Exception handler:
+.section .fixup, "ax"
+/*
+	a6  = src_err_ptr
+	a7  = dst_err_ptr
+	a11 = original len for exception handling
+	a12 = original dst for exception handling
+*/
+
+6001:
+	_movi	a2, -EFAULT
+	s32i	a2, a6, 0	/* src_err_ptr */
+
+	# clear the complete destination - computing the rest
+	# is too much work
+	movi	a2, 0
+#if XCHAL_HAVE_LOOPS
+	loopgtz	a11, 2f
+#else
+	beqz	a11, 2f
+	add	a11, a11, a12	/* a11 = ending address */
+.Leloop:
+#endif
+	s8i	a2, a12, 0
+	addi	a12, a12, 1
+#if !XCHAL_HAVE_LOOPS
+	blt	a12, a11, .Leloop
+#endif
+2:
+	retw
+
+6002:
+	movi	a2, -EFAULT
+	s32i	a2, a7, 0	/* dst_err_ptr */
+	movi	a2, 0
+	retw
+
+.previous
+
diff --git a/arch/xtensa/lib/memcopy.S b/arch/xtensa/lib/memcopy.S
new file mode 100644
index 00000000000..e8f6d7eb722
--- /dev/null
+++ b/arch/xtensa/lib/memcopy.S
@@ -0,0 +1,315 @@
+/*
+ * arch/xtensa/lib/hal/memcopy.S -- Core HAL library functions
+ * xthal_memcpy and xthal_bcopy
+ *
+ * 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.
+ */
+
+#include <xtensa/coreasm.h>
+
+	.macro	src_b	r, w0, w1
+#ifdef __XTENSA_EB__
+	src	\r, \w0, \w1
+#else
+	src	\r, \w1, \w0
+#endif
+	.endm
+
+	.macro	ssa8	r
+#ifdef __XTENSA_EB__
+	ssa8b	\r
+#else
+	ssa8l	\r
+#endif
+	.endm
+
+
+/*
+ * void *memcpy(void *dst, const void *src, size_t len);
+ * void *memmove(void *dst, const void *src, size_t len);
+ * void *bcopy(const void *src, void *dst, size_t len);
+ *
+ * This function is intended to do the same thing as the standard
+ * library function memcpy() (or bcopy()) for most cases.
+ * However, where the source and/or destination references
+ * an instruction RAM or ROM or a data RAM or ROM, that
+ * source and/or destination will always be accessed with
+ * 32-bit load and store instructions (as required for these
+ * types of devices).
+ *
+ * !!!!!!!  XTFIXME:
+ * !!!!!!!  Handling of IRAM/IROM has not yet
+ * !!!!!!!  been implemented.
+ *
+ * The bcopy version is provided here to avoid the overhead
+ * of an extra call, for callers that require this convention.
+ *
+ * The (general case) algorithm is as follows:
+ *   If destination is unaligned, align it by conditionally
+ *     copying 1 and 2 bytes.
+ *   If source is aligned,
+ *     do 16 bytes with a loop, and then finish up with
+ *     8, 4, 2, and 1 byte copies conditional on the length;
+ *   else (if source is unaligned),
+ *     do the same, but use SRC to align the source data.
+ *   This code tries to use fall-through branches for the common
+ *     case of aligned source and destination and multiple
+ *     of 4 (or 8) length.
+ *
+ * Register use:
+ *	a0/ return address
+ *	a1/ stack pointer
+ *	a2/ return value
+ *	a3/ src
+ *	a4/ length
+ *	a5/ dst
+ *	a6/ tmp
+ *	a7/ tmp
+ *	a8/ tmp
+ *	a9/ tmp
+ *	a10/ tmp
+ *	a11/ tmp
+ */
+
+	.text
+	.align	4
+	.global	bcopy
+	.type   bcopy,@function
+bcopy:
+	entry	sp, 16		# minimal stack frame
+	# a2=src, a3=dst, a4=len
+	mov	a5, a3		# copy dst so that a2 is return value
+	mov	a3, a2
+	mov	a2, a5
+	j	.Lcommon	# go to common code for memcpy+bcopy
+
+
+/*
+ * Byte by byte copy
+ */
+	.align	4
+	.byte	0		# 1 mod 4 alignment for LOOPNEZ
+				# (0 mod 4 alignment for LBEG)
+.Lbytecopy:
+#if XCHAL_HAVE_LOOPS
+	loopnez	a4, .Lbytecopydone
+#else /* !XCHAL_HAVE_LOOPS */
+	beqz	a4, .Lbytecopydone
+	add	a7, a3, a4	# a7 = end address for source
+#endif /* !XCHAL_HAVE_LOOPS */
+.Lnextbyte:
+	l8ui	a6, a3, 0
+	addi	a3, a3, 1
+	s8i	a6, a5, 0
+	addi	a5, a5, 1
+#if !XCHAL_HAVE_LOOPS
+	blt	a3, a7, .Lnextbyte
+#endif /* !XCHAL_HAVE_LOOPS */
+.Lbytecopydone:
+	retw
+
+/*
+ * Destination is unaligned
+ */
+
+	.align	4
+.Ldst1mod2:	# dst is only byte aligned
+	_bltui	a4, 7, .Lbytecopy	# do short copies byte by byte
+
+	# copy 1 byte
+	l8ui	a6, a3,  0
+	addi	a3, a3,  1
+	addi	a4, a4, -1
+	s8i	a6, a5,  0
+	addi	a5, a5,  1
+	_bbci.l	a5, 1, .Ldstaligned	# if dst is now aligned, then
+					# return to main algorithm
+.Ldst2mod4:	# dst 16-bit aligned
+	# copy 2 bytes
+	_bltui	a4, 6, .Lbytecopy	# do short copies byte by byte
+	l8ui	a6, a3,  0
+	l8ui	a7, a3,  1
+	addi	a3, a3,  2
+	addi	a4, a4, -2
+	s8i	a6, a5,  0
+	s8i	a7, a5,  1
+	addi	a5, a5,  2
+	j	.Ldstaligned	# dst is now aligned, return to main algorithm
+
+	.align	4
+	.global	memcpy
+	.type   memcpy,@function
+memcpy:
+	.global	memmove
+	.type   memmove,@function
+memmove:
+
+	entry	sp, 16		# minimal stack frame
+	# a2/ dst, a3/ src, a4/ len
+	mov	a5, a2		# copy dst so that a2 is return value
+.Lcommon:
+	_bbsi.l	a2, 0, .Ldst1mod2	# if dst is 1 mod 2
+	_bbsi.l	a2, 1, .Ldst2mod4	# if dst is 2 mod 4
+.Ldstaligned:	# return here from .Ldst?mod? once dst is aligned
+	srli	a7, a4, 4	# number of loop iterations with 16B
+				# per iteration
+	movi	a8, 3		# if source is not aligned,
+	_bany	a3, a8, .Lsrcunaligned	# then use shifting copy
+	/*
+	 * Destination and source are word-aligned, use word copy.
+	 */
+	# copy 16 bytes per iteration for word-aligned dst and word-aligned src
+#if XCHAL_HAVE_LOOPS
+	loopnez	a7, .Loop1done
+#else /* !XCHAL_HAVE_LOOPS */
+	beqz	a7, .Loop1done
+	slli	a8, a7, 4
+	add	a8, a8, a3	# a8 = end of last 16B source chunk
+#endif /* !XCHAL_HAVE_LOOPS */
+.Loop1:
+	l32i	a6, a3,  0
+	l32i	a7, a3,  4
+	s32i	a6, a5,  0
+	l32i	a6, a3,  8
+	s32i	a7, a5,  4
+	l32i	a7, a3, 12
+	s32i	a6, a5,  8
+	addi	a3, a3, 16
+	s32i	a7, a5, 12
+	addi	a5, a5, 16
+#if !XCHAL_HAVE_LOOPS
+	blt	a3, a8, .Loop1
+#endif /* !XCHAL_HAVE_LOOPS */
+.Loop1done:
+	bbci.l	a4, 3, .L2
+	# copy 8 bytes
+	l32i	a6, a3,  0
+	l32i	a7, a3,  4
+	addi	a3, a3,  8
+	s32i	a6, a5,  0
+	s32i	a7, a5,  4
+	addi	a5, a5,  8
+.L2:
+	bbsi.l	a4, 2, .L3
+	bbsi.l	a4, 1, .L4
+	bbsi.l	a4, 0, .L5
+	retw
+.L3:
+	# copy 4 bytes
+	l32i	a6, a3,  0
+	addi	a3, a3,  4
+	s32i	a6, a5,  0
+	addi	a5, a5,  4
+	bbsi.l	a4, 1, .L4
+	bbsi.l	a4, 0, .L5
+	retw
+.L4:
+	# copy 2 bytes
+	l16ui	a6, a3,  0
+	addi	a3, a3,  2
+	s16i	a6, a5,  0
+	addi	a5, a5,  2
+	bbsi.l	a4, 0, .L5
+	retw
+.L5:
+	# copy 1 byte
+	l8ui	a6, a3,  0
+	s8i	a6, a5,  0
+	retw
+
+/*
+ * Destination is aligned, Source is unaligned
+ */
+
+	.align	4
+.Lsrcunaligned:
+	_beqz	a4, .Ldone	# avoid loading anything for zero-length copies
+	# copy 16 bytes per iteration for word-aligned dst and unaligned src
+	ssa8	a3		# set shift amount from byte offset
+#define SIM_CHECKS_ALIGNMENT	1	/* set to 1 when running on ISS (simulator) with the
+					   lint or ferret client, or 0 to save a few cycles */
+#if XCHAL_UNALIGNED_LOAD_EXCEPTION || SIM_CHECKS_ALIGNMENT
+	and	a11, a3, a8	# save unalignment offset for below
+	sub	a3, a3, a11	# align a3
+#endif
+	l32i	a6, a3, 0	# load first word
+#if XCHAL_HAVE_LOOPS
+	loopnez	a7, .Loop2done
+#else /* !XCHAL_HAVE_LOOPS */
+	beqz	a7, .Loop2done
+	slli	a10, a7, 4
+	add	a10, a10, a3	# a10 = end of last 16B source chunk
+#endif /* !XCHAL_HAVE_LOOPS */
+.Loop2:
+	l32i	a7, a3,  4
+	l32i	a8, a3,  8
+	src_b	a6, a6, a7
+	s32i	a6, a5,  0
+	l32i	a9, a3, 12
+	src_b	a7, a7, a8
+	s32i	a7, a5,  4
+	l32i	a6, a3, 16
+	src_b	a8, a8, a9
+	s32i	a8, a5,  8
+	addi	a3, a3, 16
+	src_b	a9, a9, a6
+	s32i	a9, a5, 12
+	addi	a5, a5, 16
+#if !XCHAL_HAVE_LOOPS
+	blt	a3, a10, .Loop2
+#endif /* !XCHAL_HAVE_LOOPS */
+.Loop2done:
+	bbci.l	a4, 3, .L12
+	# copy 8 bytes
+	l32i	a7, a3,  4
+	l32i	a8, a3,  8
+	src_b	a6, a6, a7
+	s32i	a6, a5,  0
+	addi	a3, a3,  8
+	src_b	a7, a7, a8
+	s32i	a7, a5,  4
+	addi	a5, a5,  8
+	mov	a6, a8
+.L12:
+	bbci.l	a4, 2, .L13
+	# copy 4 bytes
+	l32i	a7, a3,  4
+	addi	a3, a3,  4
+	src_b	a6, a6, a7
+	s32i	a6, a5,  0
+	addi	a5, a5,  4
+	mov	a6, a7
+.L13:
+#if XCHAL_UNALIGNED_LOAD_EXCEPTION || SIM_CHECKS_ALIGNMENT
+	add	a3, a3, a11	# readjust a3 with correct misalignment
+#endif
+	bbsi.l	a4, 1, .L14
+	bbsi.l	a4, 0, .L15
+.Ldone:	retw
+.L14:
+	# copy 2 bytes
+	l8ui	a6, a3,  0
+	l8ui	a7, a3,  1
+	addi	a3, a3,  2
+	s8i	a6, a5,  0
+	s8i	a7, a5,  1
+	addi	a5, a5,  2
+	bbsi.l	a4, 0, .L15
+	retw
+.L15:
+	# copy 1 byte
+	l8ui	a6, a3,  0
+	s8i	a6, a5,  0
+	retw
+
+/*
+ * Local Variables:
+ * mode:fundamental
+ * comment-start: "# "
+ * comment-start-skip: "# *"
+ * End:
+ */
diff --git a/arch/xtensa/lib/memset.S b/arch/xtensa/lib/memset.S
new file mode 100644
index 00000000000..4de25134bc6
--- /dev/null
+++ b/arch/xtensa/lib/memset.S
@@ -0,0 +1,160 @@
+/*
+ *  arch/xtensa/lib/memset.S
+ *
+ *  ANSI C standard library function memset
+ *  (Well, almost.  .fixup code might return zero.)
+ *
+ *  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>
+
+/*
+ * void *memset(void *dst, int c, size_t length)
+ *
+ * The algorithm is as follows:
+ *   Create a word with c in all byte positions
+ *   If the destination is aligned,
+ *     do 16B chucks with a loop, and then finish up with
+ *     8B, 4B, 2B, and 1B stores conditional on the length.
+ *   If destination is unaligned, align it by conditionally
+ *     setting 1B and 2B and then go to aligned case.
+ *   This code tries to use fall-through branches for the common
+ *     case of an aligned destination (except for the branches to
+ *     the alignment labels).
+ */
+
+/* Load or store instructions that may cause exceptions use the EX macro. */
+
+#define EX(insn,reg1,reg2,offset,handler)	\
+9:	insn	reg1, reg2, offset;		\
+	.section __ex_table, "a";		\
+	.word	9b, handler;			\
+	.previous
+
+
+.text
+.align	4
+.global	memset
+.type	memset,@function
+memset:
+	entry	sp, 16		# minimal stack frame
+	# a2/ dst, a3/ c, a4/ length
+	extui	a3, a3, 0, 8	# mask to just 8 bits
+	slli	a7, a3, 8	# duplicate character in all bytes of word
+	or	a3, a3, a7	# ...
+	slli	a7, a3, 16	# ...
+	or	a3, a3, a7	# ...
+	mov	a5, a2		# copy dst so that a2 is return value
+	movi	a6, 3		# for alignment tests
+	bany	a2, a6, .Ldstunaligned # if dst is unaligned
+.L0:	# return here from .Ldstunaligned when dst is aligned
+	srli	a7, a4, 4	# number of loop iterations with 16B
+				# per iteration
+	bnez	a4, .Laligned
+	retw
+
+/*
+ * Destination is word-aligned.
+ */
+	# set 16 bytes per iteration for word-aligned dst
+	.align	4		# 1 mod 4 alignment for LOOPNEZ
+	.byte	0		# (0 mod 4 alignment for LBEG)
+.Laligned:
+#if XCHAL_HAVE_LOOPS
+	loopnez	a7, .Loop1done
+#else /* !XCHAL_HAVE_LOOPS */
+	beqz	a7, .Loop1done
+	slli	a6, a7, 4
+	add	a6, a6, a5	# a6 = end of last 16B chunk
+#endif /* !XCHAL_HAVE_LOOPS */
+.Loop1:
+	EX(s32i, a3, a5,  0, memset_fixup)
+	EX(s32i, a3, a5,  4, memset_fixup)
+	EX(s32i, a3, a5,  8, memset_fixup)
+	EX(s32i, a3, a5, 12, memset_fixup)
+	addi	a5, a5, 16
+#if !XCHAL_HAVE_LOOPS
+	blt	a5, a6, .Loop1
+#endif /* !XCHAL_HAVE_LOOPS */
+.Loop1done:
+	bbci.l	a4, 3, .L2
+	# set 8 bytes
+	EX(s32i, a3, a5,  0, memset_fixup)
+	EX(s32i, a3, a5,  4, memset_fixup)
+	addi	a5, a5,  8
+.L2:
+	bbci.l	a4, 2, .L3
+	# set 4 bytes
+	EX(s32i, a3, a5,  0, memset_fixup)
+	addi	a5, a5,  4
+.L3:
+	bbci.l	a4, 1, .L4
+	# set 2 bytes
+	EX(s16i, a3, a5,  0, memset_fixup)
+	addi	a5, a5,  2
+.L4:
+	bbci.l	a4, 0, .L5
+	# set 1 byte
+	EX(s8i, a3, a5,  0, memset_fixup)
+.L5:
+.Lret1:
+	retw
+
+/*
+ * Destination is unaligned
+ */
+
+.Ldstunaligned:
+	bltui	a4, 8, .Lbyteset	# do short copies byte by byte
+	bbci.l	a5, 0, .L20		# branch if dst alignment half-aligned
+	# dst is only byte aligned
+	# set 1 byte
+	EX(s8i, a3, a5,  0, memset_fixup)
+	addi	a5, a5,  1
+	addi	a4, a4, -1
+	# now retest if dst aligned
+	bbci.l	a5, 1, .L0	# if now aligned, return to main algorithm
+.L20:
+	# dst half-aligned
+	# set 2 bytes
+	EX(s16i, a3, a5,  0, memset_fixup)
+	addi	a5, a5,  2
+	addi	a4, a4, -2
+	j	.L0		# dst is now aligned, return to main algorithm
+
+/*
+ * Byte by byte set
+ */
+	.align	4
+	.byte	0		# 1 mod 4 alignment for LOOPNEZ
+				# (0 mod 4 alignment for LBEG)
+.Lbyteset:
+#if XCHAL_HAVE_LOOPS
+	loopnez	a4, .Lbytesetdone
+#else /* !XCHAL_HAVE_LOOPS */
+	beqz	a4, .Lbytesetdone
+	add	a6, a5, a4	# a6 = ending address
+#endif /* !XCHAL_HAVE_LOOPS */
+.Lbyteloop:
+	EX(s8i, a3, a5, 0, memset_fixup)
+	addi	a5, a5, 1
+#if !XCHAL_HAVE_LOOPS
+	blt	a5, a6, .Lbyteloop
+#endif /* !XCHAL_HAVE_LOOPS */
+.Lbytesetdone:
+	retw
+
+
+	.section .fixup, "ax"
+	.align	4
+
+/* We return zero if a failure occurred. */
+
+memset_fixup:
+	movi	a2, 0
+	retw
diff --git a/arch/xtensa/lib/pci-auto.c b/arch/xtensa/lib/pci-auto.c
new file mode 100644
index 00000000000..90c790f6123
--- /dev/null
+++ b/arch/xtensa/lib/pci-auto.c
@@ -0,0 +1,352 @@
+/*
+ * arch/xtensa/kernel/pci-auto.c
+ *
+ * PCI autoconfiguration library
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ *
+ * Chris Zankel <zankel@tensilica.com, cez@zankel.net>
+ *
+ * Based on work from Matt Porter <mporter@mvista.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/kernel.h>
+#include <linux/init.h>
+#include <linux/pci.h>
+
+#include <asm/pci-bridge.h>
+
+
+/*
+ *
+ * Setting up a PCI
+ *
+ * pci_ctrl->first_busno = <first bus number (0)>
+ * pci_ctrl->last_busno = <last bus number (0xff)>
+ * pci_ctrl->ops = <PCI config operations>
+ * pci_ctrl->map_irq = <function to return the interrupt number for a device>
+ *
+ * pci_ctrl->io_space.start = <IO space start address (PCI view)>
+ * pci_ctrl->io_space.end = <IO space end address (PCI view)>
+ * pci_ctrl->io_space.base = <IO space offset: address 0 from CPU space>
+ * pci_ctrl->mem_space.start = <MEM space start address (PCI view)>
+ * pci_ctrl->mem_space.end = <MEM space end address (PCI view)>
+ * pci_ctrl->mem_space.base = <MEM space offset: address 0 from CPU space>
+ *
+ * pcibios_init_resource(&pci_ctrl->io_resource, <IO space start>,
+ * 			 <IO space end>, IORESOURCE_IO, "PCI host bridge");
+ * pcibios_init_resource(&pci_ctrl->mem_resources[0], <MEM space start>,
+ * 			 <MEM space end>, IORESOURCE_MEM, "PCI host bridge");
+ *
+ * pci_ctrl->last_busno = pciauto_bus_scan(pci_ctrl,pci_ctrl->first_busno);
+ *
+ * int __init pciauto_bus_scan(struct pci_controller *pci_ctrl, int current_bus)
+ *
+ */
+
+
+/* define DEBUG to print some debugging messages. */
+
+#undef DEBUG
+
+#ifdef DEBUG
+# define DBG(x...) printk(x)
+#else
+# define DBG(x...)
+#endif
+
+static int pciauto_upper_iospc;
+static int pciauto_upper_memspc;
+
+static struct pci_dev pciauto_dev;
+static struct pci_bus pciauto_bus;
+
+/*
+ * Helper functions
+ */
+
+/* Initialize the bars of a PCI device.  */
+
+static void __init
+pciauto_setup_bars(struct pci_dev *dev, int bar_limit)
+{
+	int bar_size;
+	int bar, bar_nr;
+	int *upper_limit;
+	int found_mem64 = 0;
+
+	for (bar = PCI_BASE_ADDRESS_0, bar_nr = 0;
+	     bar <= bar_limit;
+	     bar+=4, bar_nr++)
+	{
+		/* Tickle the BAR and get the size */
+		pci_write_config_dword(dev, bar, 0xffffffff);
+		pci_read_config_dword(dev, bar, &bar_size);
+
+		/* If BAR is not implemented go to the next BAR */
+		if (!bar_size)
+			continue;
+
+		/* Check the BAR type and set our address mask */
+		if (bar_size & PCI_BASE_ADDRESS_SPACE_IO)
+		{
+			bar_size &= PCI_BASE_ADDRESS_IO_MASK;
+			upper_limit = &pciauto_upper_iospc;
+			DBG("PCI Autoconfig: BAR %d, I/O, ", bar_nr);
+		}
+		else
+		{
+			if ((bar_size & PCI_BASE_ADDRESS_MEM_TYPE_MASK) ==
+			    PCI_BASE_ADDRESS_MEM_TYPE_64)
+				found_mem64 = 1;
+
+			bar_size &= PCI_BASE_ADDRESS_MEM_MASK;
+			upper_limit = &pciauto_upper_memspc;
+			DBG("PCI Autoconfig: BAR %d, Mem, ", bar_nr);
+		}
+
+		/* Allocate a base address (bar_size is negative!) */
+		*upper_limit = (*upper_limit + bar_size) & bar_size;
+
+		/* Write it out and update our limit */
+		pci_write_config_dword(dev, bar, *upper_limit);
+
+		/*
+		 * If we are a 64-bit decoder then increment to the
+		 * upper 32 bits of the bar and force it to locate
+		 * in the lower 4GB of memory.
+		 */
+
+		if (found_mem64)
+			pci_write_config_dword(dev, (bar+=4), 0x00000000);
+
+		DBG("size=0x%x, address=0x%x\n", ~bar_size + 1, *upper_limit);
+	}
+}
+
+/* Initialize the interrupt number. */
+
+static void __init
+pciauto_setup_irq(struct pci_controller* pci_ctrl,struct pci_dev *dev,int devfn)
+{
+	u8 pin;
+	int irq = 0;
+
+	pci_read_config_byte(dev, PCI_INTERRUPT_PIN, &pin);
+
+	/* Fix illegal pin numbers. */
+
+	if (pin == 0 || pin > 4)
+		pin = 1;
+
+	if (pci_ctrl->map_irq)
+		irq = pci_ctrl->map_irq(dev, PCI_SLOT(devfn), pin);
+
+	if (irq == -1)
+		irq = 0;
+
+	DBG("PCI Autoconfig: Interrupt %d, pin %d\n", irq, pin);
+
+	pci_write_config_byte(dev, PCI_INTERRUPT_LINE, irq);
+}
+
+
+static void __init
+pciauto_prescan_setup_bridge(struct pci_dev *dev, int current_bus,
+			     int sub_bus, int *iosave, int *memsave)
+{
+	/* Configure bus number registers */
+	pci_write_config_byte(dev, PCI_PRIMARY_BUS, current_bus);
+	pci_write_config_byte(dev, PCI_SECONDARY_BUS, sub_bus + 1);
+	pci_write_config_byte(dev, PCI_SUBORDINATE_BUS,	0xff);
+
+	/* Round memory allocator to 1MB boundary */
+	pciauto_upper_memspc &= ~(0x100000 - 1);
+	*memsave = pciauto_upper_memspc;
+
+	/* Round I/O allocator to 4KB boundary */
+	pciauto_upper_iospc &= ~(0x1000 - 1);
+	*iosave = pciauto_upper_iospc;
+
+	/* Set up memory and I/O filter limits, assume 32-bit I/O space */
+	pci_write_config_word(dev, PCI_MEMORY_LIMIT,
+			      ((pciauto_upper_memspc - 1) & 0xfff00000) >> 16);
+	pci_write_config_byte(dev, PCI_IO_LIMIT,
+			      ((pciauto_upper_iospc - 1) & 0x0000f000) >> 8);
+	pci_write_config_word(dev, PCI_IO_LIMIT_UPPER16,
+			      ((pciauto_upper_iospc - 1) & 0xffff0000) >> 16);
+}
+
+static void __init
+pciauto_postscan_setup_bridge(struct pci_dev *dev, int current_bus, int sub_bus,
+			      int *iosave, int *memsave)
+{
+	int cmdstat;
+
+	/* Configure bus number registers */
+	pci_write_config_byte(dev, PCI_SUBORDINATE_BUS,	sub_bus);
+
+	/*
+	 * Round memory allocator to 1MB boundary.
+	 * If no space used, allocate minimum.
+	 */
+	pciauto_upper_memspc &= ~(0x100000 - 1);
+	if (*memsave == pciauto_upper_memspc)
+		pciauto_upper_memspc -= 0x00100000;
+
+	pci_write_config_word(dev, PCI_MEMORY_BASE, pciauto_upper_memspc >> 16);
+
+	/* Allocate 1MB for pre-fretch */
+	pci_write_config_word(dev, PCI_PREF_MEMORY_LIMIT,
+			      ((pciauto_upper_memspc - 1) & 0xfff00000) >> 16);
+
+	pciauto_upper_memspc -= 0x100000;
+
+	pci_write_config_word(dev, PCI_PREF_MEMORY_BASE,
+			      pciauto_upper_memspc >> 16);
+
+	/* Round I/O allocator to 4KB boundary */
+	pciauto_upper_iospc &= ~(0x1000 - 1);
+	if (*iosave == pciauto_upper_iospc)
+		pciauto_upper_iospc -= 0x1000;
+
+	pci_write_config_byte(dev, PCI_IO_BASE,
+			      (pciauto_upper_iospc & 0x0000f000) >> 8);
+	pci_write_config_word(dev, PCI_IO_BASE_UPPER16,
+			      pciauto_upper_iospc >> 16);
+
+	/* Enable memory and I/O accesses, enable bus master */
+	pci_read_config_dword(dev, PCI_COMMAND, &cmdstat);
+	pci_write_config_dword(dev, PCI_COMMAND,
+			       cmdstat |
+			       PCI_COMMAND_IO |
+			       PCI_COMMAND_MEMORY |
+			       PCI_COMMAND_MASTER);
+}
+
+/*
+ * Scan the current PCI bus.
+ */
+
+
+int __init pciauto_bus_scan(struct pci_controller *pci_ctrl, int current_bus)
+{
+	int sub_bus, pci_devfn, pci_class, cmdstat, found_multi=0;
+	unsigned short vid;
+	unsigned char header_type;
+	struct pci_dev *dev = &pciauto_dev;
+
+        pciauto_dev.bus = &pciauto_bus;
+        pciauto_dev.sysdata = pci_ctrl;
+	pciauto_bus.ops = pci_ctrl->ops;
+
+	/*
+	 * Fetch our I/O and memory space upper boundaries used
+	 * to allocated base addresses on this pci_controller.
+	 */
+
+	if (current_bus == pci_ctrl->first_busno)
+	{
+		pciauto_upper_iospc = pci_ctrl->io_resource.end + 1;
+		pciauto_upper_memspc = pci_ctrl->mem_resources[0].end + 1;
+	}
+
+	sub_bus = current_bus;
+
+	for (pci_devfn = 0; pci_devfn < 0xff; pci_devfn++)
+	{
+		/* Skip our host bridge */
+		if ((current_bus == pci_ctrl->first_busno) && (pci_devfn == 0))
+			continue;
+
+		if (PCI_FUNC(pci_devfn) && !found_multi)
+			continue;
+
+		pciauto_bus.number = current_bus;
+		pciauto_dev.devfn = pci_devfn;
+
+		/* If config space read fails from this device, move on */
+		if (pci_read_config_byte(dev, PCI_HEADER_TYPE, &header_type))
+			continue;
+
+		if (!PCI_FUNC(pci_devfn))
+			found_multi = header_type & 0x80;
+		pci_read_config_word(dev, PCI_VENDOR_ID, &vid);
+
+		if (vid == 0xffff || vid == 0x0000) {
+			found_multi = 0;
+			continue;
+		}
+
+		pci_read_config_dword(dev, PCI_CLASS_REVISION, &pci_class);
+
+		if ((pci_class >> 16) == PCI_CLASS_BRIDGE_PCI) {
+
+			int iosave, memsave;
+
+			DBG("PCI Autoconfig: Found P2P bridge, device %d\n",
+			    PCI_SLOT(pci_devfn));
+
+			/* Allocate PCI I/O and/or memory space */
+			pciauto_setup_bars(dev, PCI_BASE_ADDRESS_1);
+
+			pciauto_prescan_setup_bridge(dev, current_bus, sub_bus,
+					&iosave, &memsave);
+			sub_bus = pciauto_bus_scan(pci_ctrl, sub_bus+1);
+			pciauto_postscan_setup_bridge(dev, current_bus, sub_bus,
+					&iosave, &memsave);
+			pciauto_bus.number = current_bus;
+
+			continue;
+
+		}
+
+
+#if 0
+		/* Skip legacy mode IDE controller */
+
+		if ((pci_class >> 16) == PCI_CLASS_STORAGE_IDE) {
+
+			unsigned char prg_iface;
+			pci_read_config_byte(dev, PCI_CLASS_PROG, &prg_iface);
+
+			if (!(prg_iface & PCIAUTO_IDE_MODE_MASK)) {
+				DBG("PCI Autoconfig: Skipping legacy mode "
+				    "IDE controller\n");
+				continue;
+			}
+		}
+#endif
+
+		/*
+		 * Found a peripheral, enable some standard
+		 * settings
+		 */
+
+		pci_read_config_dword(dev, PCI_COMMAND,	&cmdstat);
+		pci_write_config_dword(dev, PCI_COMMAND,
+				cmdstat |
+					PCI_COMMAND_IO |
+					PCI_COMMAND_MEMORY |
+					PCI_COMMAND_MASTER);
+		pci_write_config_byte(dev, PCI_LATENCY_TIMER, 0x80);
+
+		/* Allocate PCI I/O and/or memory space */
+		DBG("PCI Autoconfig: Found Bus %d, Device %d, Function %d\n",
+		    current_bus, PCI_SLOT(pci_devfn), PCI_FUNC(pci_devfn) );
+
+		pciauto_setup_bars(dev, PCI_BASE_ADDRESS_5);
+		pciauto_setup_irq(pci_ctrl, dev, pci_devfn);
+	}
+	return sub_bus;
+}
+
+
+
+
+
diff --git a/arch/xtensa/lib/strcasecmp.c b/arch/xtensa/lib/strcasecmp.c
new file mode 100644
index 00000000000..165b2d6effa
--- /dev/null
+++ b/arch/xtensa/lib/strcasecmp.c
@@ -0,0 +1,32 @@
+/*
+ *  linux/arch/xtensa/lib/strcasecmp.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) 2002 Tensilica Inc.
+ */
+
+#include <linux/string.h>
+
+
+/* We handle nothing here except the C locale.  Since this is used in
+   only one place, on strings known to contain only 7 bit ASCII, this
+   is ok.  */
+
+int strcasecmp(const char *a, const char *b)
+{
+	int ca, cb;
+
+	do {
+		ca = *a++ & 0xff;
+		cb = *b++ & 0xff;
+		if (ca >= 'A' && ca <= 'Z')
+			ca += 'a' - 'A';
+		if (cb >= 'A' && cb <= 'Z')
+			cb += 'a' - 'A';
+	} while (ca == cb && ca != '\0');
+
+	return ca - cb;
+}
diff --git a/arch/xtensa/lib/strncpy_user.S b/arch/xtensa/lib/strncpy_user.S
new file mode 100644
index 00000000000..71d55df4389
--- /dev/null
+++ b/arch/xtensa/lib/strncpy_user.S
@@ -0,0 +1,224 @@
+/*
+ *  arch/xtensa/lib/strncpy_user.S
+ *
+ *  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.
+ *
+ *  Returns: -EFAULT if exception before terminator, N if the entire
+ *  buffer filled, else strlen.
+ *
+ *  Copyright (C) 2002 Tensilica Inc.
+ */
+
+#include <xtensa/coreasm.h>
+#include <linux/errno.h>
+
+/* Load or store instructions that may cause exceptions use the EX macro. */
+
+#define EX(insn,reg1,reg2,offset,handler)	\
+9:	insn	reg1, reg2, offset;		\
+	.section __ex_table, "a";		\
+	.word	9b, handler;			\
+	.previous
+
+/*
+ * char *__strncpy_user(char *dst, const char *src, size_t len)
+ */
+.text
+.begin literal
+.align	4
+.Lmask0:
+	.byte	0xff, 0x00, 0x00, 0x00
+.Lmask1:
+	.byte	0x00, 0xff, 0x00, 0x00
+.Lmask2:
+	.byte	0x00, 0x00, 0xff, 0x00
+.Lmask3:
+	.byte	0x00, 0x00, 0x00, 0xff
+.end literal
+
+# Register use
+#   a0/ return address
+#   a1/ stack pointer
+#   a2/ return value
+#   a3/ src
+#   a4/ len
+#   a5/ mask0
+#   a6/ mask1
+#   a7/ mask2
+#   a8/ mask3
+#   a9/ tmp
+#   a10/ tmp
+#   a11/ dst
+#   a12/ tmp
+
+.align	4
+.global	__strncpy_user
+.type	__strncpy_user,@function
+__strncpy_user:
+	entry	sp, 16		# minimal stack frame
+	# a2/ dst, a3/ src, a4/ len
+	mov	a11, a2		# leave dst in return value register
+	beqz	a4, .Lret	# if len is zero
+	l32r	a5, .Lmask0	# mask for byte 0
+	l32r	a6, .Lmask1	# mask for byte 1
+	l32r	a7, .Lmask2	# mask for byte 2
+	l32r	a8, .Lmask3	# mask for byte 3
+	bbsi.l	a3, 0, .Lsrc1mod2 # if only  8-bit aligned
+	bbsi.l	a3, 1, .Lsrc2mod4 # if only 16-bit aligned
+.Lsrcaligned:	# return here when src is word-aligned
+	srli	a12, a4, 2	# number of loop iterations with 4B per loop
+	movi	a9, 3
+	bnone	a11, a9, .Laligned
+	j	.Ldstunaligned
+
+.Lsrc1mod2:	# src address is odd
+	EX(l8ui, a9, a3, 0, fixup_l)	# get byte 0
+	addi	a3, a3, 1		# advance src pointer
+	EX(s8i, a9, a11, 0, fixup_s)	# store byte 0
+	beqz	a9, .Lret		# if byte 0 is zero
+	addi	a11, a11, 1		# advance dst pointer
+	addi	a4, a4, -1		# decrement len
+	beqz	a4, .Lret		# if len is zero
+	bbci.l	a3, 1, .Lsrcaligned	# if src is now word-aligned
+
+.Lsrc2mod4:	# src address is 2 mod 4
+	EX(l8ui, a9, a3, 0, fixup_l)	# get byte 0
+	/* 1-cycle interlock */
+	EX(s8i, a9, a11, 0, fixup_s)	# store byte 0
+	beqz	a9, .Lret		# if byte 0 is zero
+	addi	a11, a11, 1		# advance dst pointer
+	addi	a4, a4, -1		# decrement len
+	beqz	a4, .Lret		# if len is zero
+	EX(l8ui, a9, a3, 1, fixup_l)	# get byte 0
+	addi	a3, a3, 2		# advance src pointer
+	EX(s8i, a9, a11, 0, fixup_s)	# store byte 0
+	beqz	a9, .Lret		# if byte 0 is zero
+	addi	a11, a11, 1		# advance dst pointer
+	addi	a4, a4, -1		# decrement len
+	bnez	a4, .Lsrcaligned	# if len is nonzero
+.Lret:
+	sub	a2, a11, a2		# compute strlen
+	retw
+
+/*
+ * dst is word-aligned, src is word-aligned
+ */
+	.align	4		# 1 mod 4 alignment for LOOPNEZ
+	.byte	0		# (0 mod 4 alignment for LBEG)
+.Laligned:
+#if XCHAL_HAVE_LOOPS
+	loopnez	a12, .Loop1done
+#else
+	beqz	a12, .Loop1done
+	slli	a12, a12, 2
+	add	a12, a12, a11	# a12 = end of last 4B chunck
+#endif
+.Loop1:
+	EX(l32i, a9, a3, 0, fixup_l)	# get word from src
+	addi	a3, a3, 4		# advance src pointer
+	bnone	a9, a5, .Lz0		# if byte 0 is zero
+	bnone	a9, a6, .Lz1		# if byte 1 is zero
+	bnone	a9, a7, .Lz2		# if byte 2 is zero
+	EX(s32i, a9, a11, 0, fixup_s)	# store word to dst
+	bnone	a9, a8, .Lz3		# if byte 3 is zero
+	addi	a11, a11, 4		# advance dst pointer
+#if !XCHAL_HAVE_LOOPS
+	blt	a11, a12, .Loop1
+#endif
+
+.Loop1done:
+	bbci.l	a4, 1, .L100
+	# copy 2 bytes
+	EX(l16ui, a9, a3, 0, fixup_l)
+	addi	a3, a3, 2		# advance src pointer
+#ifdef __XTENSA_EB__
+	bnone	a9, a7, .Lz0		# if byte 2 is zero
+	bnone	a9, a8, .Lz1		# if byte 3 is zero
+#else
+	bnone	a9, a5, .Lz0		# if byte 0 is zero
+	bnone	a9, a6, .Lz1		# if byte 1 is zero
+#endif
+	EX(s16i, a9, a11, 0, fixup_s)
+	addi	a11, a11, 2		# advance dst pointer
+.L100:
+	bbci.l	a4, 0, .Lret
+	EX(l8ui, a9, a3, 0, fixup_l)
+	/* slot */
+	EX(s8i, a9, a11, 0, fixup_s)
+	beqz	a9, .Lret		# if byte is zero
+	addi	a11, a11, 1-3		# advance dst ptr 1, but also cancel
+					# the effect of adding 3 in .Lz3 code
+	/* fall thru to .Lz3 and "retw" */
+
+.Lz3:	# byte 3 is zero
+	addi	a11, a11, 3		# advance dst pointer
+	sub	a2, a11, a2		# compute strlen
+	retw
+.Lz0:	# byte 0 is zero
+#ifdef __XTENSA_EB__
+	movi	a9, 0
+#endif /* __XTENSA_EB__ */
+	EX(s8i, a9, a11, 0, fixup_s)
+	sub	a2, a11, a2		# compute strlen
+	retw
+.Lz1:	# byte 1 is zero
+#ifdef __XTENSA_EB__
+        extui   a9, a9, 16, 16
+#endif /* __XTENSA_EB__ */
+	EX(s16i, a9, a11, 0, fixup_s)
+	addi	a11, a11, 1		# advance dst pointer
+	sub	a2, a11, a2		# compute strlen
+	retw
+.Lz2:	# byte 2 is zero
+#ifdef __XTENSA_EB__
+        extui   a9, a9, 16, 16
+#endif /* __XTENSA_EB__ */
+	EX(s16i, a9, a11, 0, fixup_s)
+	movi	a9, 0
+	EX(s8i, a9, a11, 2, fixup_s)
+	addi	a11, a11, 2		# advance dst pointer
+	sub	a2, a11, a2		# compute strlen
+	retw
+
+	.align	4		# 1 mod 4 alignment for LOOPNEZ
+	.byte	0		# (0 mod 4 alignment for LBEG)
+.Ldstunaligned:
+/*
+ * for now just use byte copy loop
+ */
+#if XCHAL_HAVE_LOOPS
+	loopnez	a4, .Lunalignedend
+#else
+	beqz	a4, .Lunalignedend
+	add	a12, a11, a4		# a12 = ending address
+#endif /* XCHAL_HAVE_LOOPS */
+.Lnextbyte:
+	EX(l8ui, a9, a3, 0, fixup_l)
+	addi	a3, a3, 1
+	EX(s8i, a9, a11, 0, fixup_s)
+	beqz	a9, .Lunalignedend
+	addi	a11, a11, 1
+#if !XCHAL_HAVE_LOOPS
+	blt	a11, a12, .Lnextbyte
+#endif
+
+.Lunalignedend:
+	sub	a2, a11, a2		# compute strlen
+	retw
+
+
+	.section .fixup, "ax"
+	.align	4
+
+	/* For now, just return -EFAULT.  Future implementations might
+	 * like to clear remaining kernel space, like the fixup
+	 * implementation in memset().  Thus, we differentiate between
+	 * load/store fixups. */
+
+fixup_s:
+fixup_l:
+	movi	a2, -EFAULT
+	retw
+
diff --git a/arch/xtensa/lib/strnlen_user.S b/arch/xtensa/lib/strnlen_user.S
new file mode 100644
index 00000000000..cdff4d670f3
--- /dev/null
+++ b/arch/xtensa/lib/strnlen_user.S
@@ -0,0 +1,147 @@
+/*
+ *  arch/xtensa/lib/strnlen_user.S
+ *
+ *  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.
+ *
+ *  Returns strnlen, including trailing zero terminator.
+ *  Zero indicates error.
+ *
+ *  Copyright (C) 2002 Tensilica Inc.
+ */
+
+#include <xtensa/coreasm.h>
+
+/* Load or store instructions that may cause exceptions use the EX macro. */
+
+#define EX(insn,reg1,reg2,offset,handler)	\
+9:	insn	reg1, reg2, offset;		\
+	.section __ex_table, "a";		\
+	.word	9b, handler;			\
+	.previous
+
+/*
+ * size_t __strnlen_user(const char *s, size_t len)
+ */
+.text
+.begin literal
+.align	4
+.Lmask0:
+	.byte	0xff, 0x00, 0x00, 0x00
+.Lmask1:
+	.byte	0x00, 0xff, 0x00, 0x00
+.Lmask2:
+	.byte	0x00, 0x00, 0xff, 0x00
+.Lmask3:
+	.byte	0x00, 0x00, 0x00, 0xff
+.end literal
+
+# Register use:
+#   a2/ src
+#   a3/ len
+#   a4/ tmp
+#   a5/ mask0
+#   a6/ mask1
+#   a7/ mask2
+#   a8/ mask3
+#   a9/ tmp
+#   a10/ tmp
+
+.align	4
+.global	__strnlen_user
+.type	__strnlen_user,@function
+__strnlen_user:
+	entry	sp, 16		# minimal stack frame
+	# a2/ s, a3/ len
+	addi	a4, a2, -4	# because we overincrement at the end;
+				# we compensate with load offsets of 4
+	l32r	a5, .Lmask0	# mask for byte 0
+	l32r	a6, .Lmask1	# mask for byte 1
+	l32r	a7, .Lmask2	# mask for byte 2
+	l32r	a8, .Lmask3	# mask for byte 3
+	bbsi.l	a2, 0, .L1mod2	# if only  8-bit aligned
+	bbsi.l	a2, 1, .L2mod4	# if only 16-bit aligned
+
+/*
+ * String is word-aligned.
+ */
+.Laligned:
+	srli	a10, a3, 2	# number of loop iterations with 4B per loop
+#if XCHAL_HAVE_LOOPS
+	loopnez	a10, .Ldone
+#else
+	beqz	a10, .Ldone
+	slli	a10, a10, 2
+	add	a10, a10, a4	# a10 = end of last 4B chunk
+#endif /* XCHAL_HAVE_LOOPS */
+.Loop:
+	EX(l32i, a9, a4, 4, lenfixup)	# get next word of string
+	addi	a4, a4, 4		# advance string pointer
+	bnone	a9, a5, .Lz0		# if byte 0 is zero
+	bnone	a9, a6, .Lz1		# if byte 1 is zero
+	bnone	a9, a7, .Lz2		# if byte 2 is zero
+	bnone	a9, a8, .Lz3		# if byte 3 is zero
+#if !XCHAL_HAVE_LOOPS
+	blt	a4, a10, .Loop
+#endif
+
+.Ldone:
+	EX(l32i, a9, a4, 4, lenfixup)	# load 4 bytes for remaining checks
+
+	bbci.l	a3, 1, .L100
+	# check two more bytes (bytes 0, 1 of word)
+	addi	a4, a4, 2	# advance string pointer
+	bnone	a9, a5, .Lz0	# if byte 0 is zero
+	bnone	a9, a6, .Lz1	# if byte 1 is zero
+.L100:
+	bbci.l	a3, 0, .L101
+	# check one more byte (byte 2 of word)
+	# Actually, we don't need to check.  Zero or nonzero, we'll add one.
+	# Do not add an extra one for the NULL terminator since we have
+	#  exhausted the original len parameter.
+	addi	a4, a4, 1	# advance string pointer
+.L101:
+	sub	a2, a4, a2	# compute length
+	retw
+
+# NOTE that in several places below, we point to the byte just after
+# the zero byte in order to include the NULL terminator in the count.
+
+.Lz3:	# byte 3 is zero
+	addi	a4, a4, 3	# point to zero byte
+.Lz0:	# byte 0 is zero
+	addi	a4, a4, 1	# point just beyond zero byte
+	sub	a2, a4, a2	# subtract to get length
+	retw
+.Lz1:	# byte 1 is zero
+	addi	a4, a4, 1+1	# point just beyond zero byte
+	sub	a2, a4, a2	# subtract to get length
+	retw
+.Lz2:	# byte 2 is zero
+	addi	a4, a4, 2+1	# point just beyond zero byte
+	sub	a2, a4, a2	# subtract to get length
+	retw
+
+.L1mod2:	# address is odd
+	EX(l8ui, a9, a4, 4, lenfixup)	# get byte 0
+	addi	a4, a4, 1		# advance string pointer
+	beqz	a9, .Lz3		# if byte 0 is zero
+	bbci.l	a4, 1, .Laligned	# if string pointer is now word-aligned
+
+.L2mod4:	# address is 2 mod 4
+	addi	a4, a4, 2	# advance ptr for aligned access
+	EX(l32i, a9, a4, 0, lenfixup)	# get word with first two bytes of string
+	bnone	a9, a7, .Lz2	# if byte 2 (of word, not string) is zero
+	bany	a9, a8, .Laligned # if byte 3 (of word, not string) is nonzero
+	# byte 3 is zero
+	addi	a4, a4, 3+1	# point just beyond zero byte
+	sub	a2, a4, a2	# subtract to get length
+	retw
+
+	.section .fixup, "ax"
+	.align	4
+lenfixup:
+	movi	a2, 0
+	retw
+
diff --git a/arch/xtensa/lib/usercopy.S b/arch/xtensa/lib/usercopy.S
new file mode 100644
index 00000000000..265db2693cb
--- /dev/null
+++ b/arch/xtensa/lib/usercopy.S
@@ -0,0 +1,321 @@
+/*
+ *  arch/xtensa/lib/usercopy.S
+ *
+ *  Copy to/from user space (derived from arch/xtensa/lib/hal/memcopy.S)
+ *
+ *  DO NOT COMBINE this function with <arch/xtensa/lib/hal/memcopy.S>.
+ *  It needs to remain separate and distinct.  The hal files are part
+ *  of the the Xtensa link-time HAL, and those files may differ per
+ *  processor configuration.  Patching the kernel for another
+ *  processor configuration includes replacing the hal files, and we
+ *  could loose the special functionality for accessing user-space
+ *  memory during such a patch.  We sacrifice a little code space here
+ *  in favor to simplify code maintenance.
+ *
+ *  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.
+ */
+
+
+/*
+ * size_t __xtensa_copy_user (void *dst, const void *src, size_t len);
+ *
+ * The returned value is the number of bytes not copied.  Implies zero
+ * is success.
+ *
+ * The general case algorithm is as follows:
+ *   If the destination and source are both aligned,
+ *     do 16B chunks with a loop, and then finish up with
+ *     8B, 4B, 2B, and 1B copies conditional on the length.
+ *   If destination is aligned and source unaligned,
+ *     do the same, but use SRC to align the source data.
+ *   If destination is unaligned, align it by conditionally
+ *     copying 1B and 2B and then retest.
+ *   This code tries to use fall-through braches for the common
+ *     case of aligned destinations (except for the branches to
+ *     the alignment label).
+ *
+ * Register use:
+ *	a0/ return address
+ *	a1/ stack pointer
+ *	a2/ return value
+ *	a3/ src
+ *	a4/ length
+ *	a5/ dst
+ *	a6/ tmp
+ *	a7/ tmp
+ *	a8/ tmp
+ *	a9/ tmp
+ *	a10/ tmp
+ *	a11/ original length
+ */
+
+#include <xtensa/coreasm.h>
+
+#ifdef __XTENSA_EB__
+#define ALIGN(R, W0, W1) src	R, W0, W1
+#define SSA8(R)	ssa8b R
+#else
+#define ALIGN(R, W0, W1) src	R, W1, W0
+#define SSA8(R)	ssa8l R
+#endif
+
+/* Load or store instructions that may cause exceptions use the EX macro. */
+
+#define EX(insn,reg1,reg2,offset,handler)	\
+9:	insn	reg1, reg2, offset;		\
+	.section __ex_table, "a";		\
+	.word	9b, handler;			\
+	.previous
+
+
+	.text
+	.align	4
+	.global	__xtensa_copy_user
+	.type	__xtensa_copy_user,@function
+__xtensa_copy_user:
+	entry	sp, 16		# minimal stack frame
+	# a2/ dst, a3/ src, a4/ len
+	mov	a5, a2		# copy dst so that a2 is return value
+	mov	a11, a4		# preserve original len for error case
+.Lcommon:
+	bbsi.l	a2, 0, .Ldst1mod2 # if dst is 1 mod 2
+	bbsi.l	a2, 1, .Ldst2mod4 # if dst is 2 mod 4
+.Ldstaligned:	# return here from .Ldstunaligned when dst is aligned
+	srli	a7, a4, 4	# number of loop iterations with 16B
+				# per iteration
+	movi	a8, 3		  # if source is also aligned,
+	bnone	a3, a8, .Laligned # then use word copy
+	SSA8(	a3)		# set shift amount from byte offset
+	bnez	a4, .Lsrcunaligned
+	movi	a2, 0		# return success for len==0
+	retw
+
+/*
+ * Destination is unaligned
+ */
+
+.Ldst1mod2:	# dst is only byte aligned
+	bltui	a4, 7, .Lbytecopy	# do short copies byte by byte
+
+	# copy 1 byte
+	EX(l8ui, a6, a3, 0, l_fixup)
+	addi	a3, a3,  1
+	EX(s8i, a6, a5,  0, s_fixup)
+	addi	a5, a5,  1
+	addi	a4, a4, -1
+	bbci.l	a5, 1, .Ldstaligned	# if dst is now aligned, then
+					# return to main algorithm
+.Ldst2mod4:	# dst 16-bit aligned
+	# copy 2 bytes
+	bltui	a4, 6, .Lbytecopy	# do short copies byte by byte
+	EX(l8ui, a6, a3, 0, l_fixup)
+	EX(l8ui, a7, a3, 1, l_fixup)
+	addi	a3, a3,  2
+	EX(s8i, a6, a5,  0, s_fixup)
+	EX(s8i, a7, a5,  1, s_fixup)
+	addi	a5, a5,  2
+	addi	a4, a4, -2
+	j	.Ldstaligned	# dst is now aligned, return to main algorithm
+
+/*
+ * Byte by byte copy
+ */
+	.align	4
+	.byte	0		# 1 mod 4 alignment for LOOPNEZ
+				# (0 mod 4 alignment for LBEG)
+.Lbytecopy:
+#if XCHAL_HAVE_LOOPS
+	loopnez	a4, .Lbytecopydone
+#else /* !XCHAL_HAVE_LOOPS */
+	beqz	a4, .Lbytecopydone
+	add	a7, a3, a4	# a7 = end address for source
+#endif /* !XCHAL_HAVE_LOOPS */
+.Lnextbyte:
+	EX(l8ui, a6, a3, 0, l_fixup)
+	addi	a3, a3, 1
+	EX(s8i, a6, a5, 0, s_fixup)
+	addi	a5, a5, 1
+#if !XCHAL_HAVE_LOOPS
+	blt	a3, a7, .Lnextbyte
+#endif /* !XCHAL_HAVE_LOOPS */
+.Lbytecopydone:
+	movi	a2, 0		# return success for len bytes copied
+	retw
+
+/*
+ * Destination and source are word-aligned.
+ */
+	# copy 16 bytes per iteration for word-aligned dst and word-aligned src
+	.align	4		# 1 mod 4 alignment for LOOPNEZ
+	.byte	0		# (0 mod 4 alignment for LBEG)
+.Laligned:
+#if XCHAL_HAVE_LOOPS
+	loopnez	a7, .Loop1done
+#else /* !XCHAL_HAVE_LOOPS */
+	beqz	a7, .Loop1done
+	slli	a8, a7, 4
+	add	a8, a8, a3	# a8 = end of last 16B source chunk
+#endif /* !XCHAL_HAVE_LOOPS */
+.Loop1:
+	EX(l32i, a6, a3,  0, l_fixup)
+	EX(l32i, a7, a3,  4, l_fixup)
+	EX(s32i, a6, a5,  0, s_fixup)
+	EX(l32i, a6, a3,  8, l_fixup)
+	EX(s32i, a7, a5,  4, s_fixup)
+	EX(l32i, a7, a3, 12, l_fixup)
+	EX(s32i, a6, a5,  8, s_fixup)
+	addi	a3, a3, 16
+	EX(s32i, a7, a5, 12, s_fixup)
+	addi	a5, a5, 16
+#if !XCHAL_HAVE_LOOPS
+	blt	a3, a8, .Loop1
+#endif /* !XCHAL_HAVE_LOOPS */
+.Loop1done:
+	bbci.l	a4, 3, .L2
+	# copy 8 bytes
+	EX(l32i, a6, a3,  0, l_fixup)
+	EX(l32i, a7, a3,  4, l_fixup)
+	addi	a3, a3,  8
+	EX(s32i, a6, a5,  0, s_fixup)
+	EX(s32i, a7, a5,  4, s_fixup)
+	addi	a5, a5,  8
+.L2:
+	bbci.l	a4, 2, .L3
+	# copy 4 bytes
+	EX(l32i, a6, a3,  0, l_fixup)
+	addi	a3, a3,  4
+	EX(s32i, a6, a5,  0, s_fixup)
+	addi	a5, a5,  4
+.L3:
+	bbci.l	a4, 1, .L4
+	# copy 2 bytes
+	EX(l16ui, a6, a3,  0, l_fixup)
+	addi	a3, a3,  2
+	EX(s16i,  a6, a5,  0, s_fixup)
+	addi	a5, a5,  2
+.L4:
+	bbci.l	a4, 0, .L5
+	# copy 1 byte
+	EX(l8ui, a6, a3,  0, l_fixup)
+	EX(s8i,  a6, a5,  0, s_fixup)
+.L5:
+	movi	a2, 0		# return success for len bytes copied
+	retw
+
+/*
+ * Destination is aligned, Source is unaligned
+ */
+
+	.align	4
+	.byte	0		# 1 mod 4 alignement for LOOPNEZ
+				# (0 mod 4 alignment for LBEG)
+.Lsrcunaligned:
+	# copy 16 bytes per iteration for word-aligned dst and unaligned src
+	and	a10, a3, a8	# save unalignment offset for below
+	sub	a3, a3, a10	# align a3 (to avoid sim warnings only; not needed for hardware)
+	EX(l32i, a6, a3, 0, l_fixup)	# load first word
+#if XCHAL_HAVE_LOOPS
+	loopnez	a7, .Loop2done
+#else /* !XCHAL_HAVE_LOOPS */
+	beqz	a7, .Loop2done
+	slli	a10, a7, 4
+	add	a10, a10, a3	# a10 = end of last 16B source chunk
+#endif /* !XCHAL_HAVE_LOOPS */
+.Loop2:
+	EX(l32i, a7, a3,  4, l_fixup)
+	EX(l32i, a8, a3,  8, l_fixup)
+	ALIGN(	a6, a6, a7)
+	EX(s32i, a6, a5,  0, s_fixup)
+	EX(l32i, a9, a3, 12, l_fixup)
+	ALIGN(	a7, a7, a8)
+	EX(s32i, a7, a5,  4, s_fixup)
+	EX(l32i, a6, a3, 16, l_fixup)
+	ALIGN(	a8, a8, a9)
+	EX(s32i, a8, a5,  8, s_fixup)
+	addi	a3, a3, 16
+	ALIGN(	a9, a9, a6)
+	EX(s32i, a9, a5, 12, s_fixup)
+	addi	a5, a5, 16
+#if !XCHAL_HAVE_LOOPS
+	blt	a3, a10, .Loop2
+#endif /* !XCHAL_HAVE_LOOPS */
+.Loop2done:
+	bbci.l	a4, 3, .L12
+	# copy 8 bytes
+	EX(l32i, a7, a3,  4, l_fixup)
+	EX(l32i, a8, a3,  8, l_fixup)
+	ALIGN(	a6, a6, a7)
+	EX(s32i, a6, a5,  0, s_fixup)
+	addi	a3, a3,  8
+	ALIGN(	a7, a7, a8)
+	EX(s32i, a7, a5,  4, s_fixup)
+	addi	a5, a5,  8
+	mov	a6, a8
+.L12:
+	bbci.l	a4, 2, .L13
+	# copy 4 bytes
+	EX(l32i, a7, a3,  4, l_fixup)
+	addi	a3, a3,  4
+	ALIGN(	a6, a6, a7)
+	EX(s32i, a6, a5,  0, s_fixup)
+	addi	a5, a5,  4
+	mov	a6, a7
+.L13:
+	add	a3, a3, a10	# readjust a3 with correct misalignment
+	bbci.l	a4, 1, .L14
+	# copy 2 bytes
+	EX(l8ui, a6, a3,  0, l_fixup)
+	EX(l8ui, a7, a3,  1, l_fixup)
+	addi	a3, a3,  2
+	EX(s8i, a6, a5,  0, s_fixup)
+	EX(s8i, a7, a5,  1, s_fixup)
+	addi	a5, a5,  2
+.L14:
+	bbci.l	a4, 0, .L15
+	# copy 1 byte
+	EX(l8ui, a6, a3,  0, l_fixup)
+	EX(s8i,  a6, a5,  0, s_fixup)
+.L15:
+	movi	a2, 0		# return success for len bytes copied
+	retw
+
+
+	.section .fixup, "ax"
+	.align	4
+
+/* a2 = original dst; a5 = current dst; a11= original len
+ * bytes_copied = a5 - a2
+ * retval = bytes_not_copied = original len - bytes_copied
+ * retval = a11 - (a5 - a2)
+ *
+ * Clearing the remaining pieces of kernel memory plugs security
+ * holes.  This functionality is the equivalent of the *_zeroing
+ * functions that some architectures provide.
+ */
+
+.Lmemset:
+	.word	memset
+
+s_fixup:
+	sub	a2, a5, a2	/* a2 <-- bytes copied */
+	sub	a2, a11, a2	/* a2 <-- bytes not copied */
+	retw
+
+l_fixup:
+	sub	a2, a5, a2	/* a2 <-- bytes copied */
+	sub	a2, a11, a2	/* a2 <-- bytes not copied == return value */
+
+	/* void *memset(void *s, int c, size_t n); */
+	mov	a6, a5		/* s */
+	movi	a7, 0		/* c */
+	mov	a8, a2		/* n */
+	l32r	a4, .Lmemset
+	callx4	a4
+	/* Ignore memset return value in a6. */
+	/* a2 still contains bytes not copied. */
+	retw
+
-- 
cgit v1.2.3-70-g09d2


From 3f65ce4d141e435e54c20ed2379d983d362a2cb5 Mon Sep 17 00:00:00 2001
From: Chris Zankel <czankel@tensilica.com>
Date: Thu, 23 Jun 2005 22:01:24 -0700
Subject: [PATCH] xtensa: Architecture support for Tensilica Xtensa Part 5

The attached patches provides part 5 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/mm/Makefile  |  13 ++
 arch/xtensa/mm/fault.c   | 241 +++++++++++++++++++++
 arch/xtensa/mm/init.c    | 551 +++++++++++++++++++++++++++++++++++++++++++++++
 arch/xtensa/mm/misc.S    | 374 ++++++++++++++++++++++++++++++++
 arch/xtensa/mm/pgtable.c |  76 +++++++
 arch/xtensa/mm/tlb.c     | 545 ++++++++++++++++++++++++++++++++++++++++++++++
 6 files changed, 1800 insertions(+)
 create mode 100644 arch/xtensa/mm/Makefile
 create mode 100644 arch/xtensa/mm/fault.c
 create mode 100644 arch/xtensa/mm/init.c
 create mode 100644 arch/xtensa/mm/misc.S
 create mode 100644 arch/xtensa/mm/pgtable.c
 create mode 100644 arch/xtensa/mm/tlb.c

(limited to 'arch')

diff --git a/arch/xtensa/mm/Makefile b/arch/xtensa/mm/Makefile
new file mode 100644
index 00000000000..a5aed5932d7
--- /dev/null
+++ b/arch/xtensa/mm/Makefile
@@ -0,0 +1,13 @@
+#
+# Makefile for the Linux/Xtensa-specific parts of the memory manager.
+#
+# 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 definition is now in the main makefile...
+
+obj-y	 := init.o fault.o tlb.o misc.o
+obj-m	 :=
+obj-n	 :=
+obj-	 :=
diff --git a/arch/xtensa/mm/fault.c b/arch/xtensa/mm/fault.c
new file mode 100644
index 00000000000..a945a33e85a
--- /dev/null
+++ b/arch/xtensa/mm/fault.c
@@ -0,0 +1,241 @@
+// TODO VM_EXEC flag work-around, cache aliasing
+/*
+ * arch/xtensa/mm/fault.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.
+ *
+ * Chris Zankel <chris@zankel.net>
+ * Joe Taylor	<joe@tensilica.com, joetylr@yahoo.com>
+ */
+
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <asm/mmu_context.h>
+#include <asm/cacheflush.h>
+#include <asm/hardirq.h>
+#include <asm/uaccess.h>
+#include <asm/system.h>
+#include <asm/pgalloc.h>
+
+unsigned long asid_cache = ASID_FIRST_VERSION;
+void bad_page_fault(struct pt_regs*, unsigned long, int);
+
+/*
+ * This routine handles page faults.  It determines the address,
+ * and the problem, and then passes it off to one of the appropriate
+ * routines.
+ *
+ * Note: does not handle Miss and MultiHit.
+ */
+
+void do_page_fault(struct pt_regs *regs)
+{
+	struct vm_area_struct * vma;
+	struct mm_struct *mm = current->mm;
+	unsigned int exccause = regs->exccause;
+	unsigned int address = regs->excvaddr;
+	siginfo_t info;
+
+	int is_write, is_exec;
+
+	info.si_code = SEGV_MAPERR;
+
+	/* We fault-in kernel-space virtual memory on-demand. The
+	 * 'reference' page table is init_mm.pgd.
+	 */
+	if (address >= TASK_SIZE && !user_mode(regs))
+		goto vmalloc_fault;
+
+	/* If we're in an interrupt or have no user
+	 * context, we must not take the fault..
+	 */
+	if (in_atomic() || !mm) {
+		bad_page_fault(regs, address, SIGSEGV);
+		return;
+	}
+
+	is_write = (exccause == XCHAL_EXCCAUSE_STORE_CACHE_ATTRIBUTE) ? 1 : 0;
+	is_exec =  (exccause == XCHAL_EXCCAUSE_ITLB_PRIVILEGE ||
+		    exccause == XCHAL_EXCCAUSE_ITLB_MISS ||
+		    exccause == XCHAL_EXCCAUSE_FETCH_CACHE_ATTRIBUTE) ? 1 : 0;
+
+#if 0
+	printk("[%s:%d:%08x:%d:%08x:%s%s]\n", current->comm, current->pid,
+	       address, exccause, regs->pc, is_write? "w":"", is_exec? "x":"");
+#endif
+
+	down_read(&mm->mmap_sem);
+	vma = find_vma(mm, address);
+
+	if (!vma)
+		goto bad_area;
+	if (vma->vm_start <= address)
+		goto good_area;
+	if (!(vma->vm_flags & VM_GROWSDOWN))
+		goto bad_area;
+	if (expand_stack(vma, address))
+		goto bad_area;
+
+	/* Ok, we have a good vm_area for this memory access, so
+	 * we can handle it..
+	 */
+
+good_area:
+	info.si_code = SEGV_ACCERR;
+
+	if (is_write) {
+		if (!(vma->vm_flags & VM_WRITE))
+			goto bad_area;
+	} else if (is_exec) {
+		if (!(vma->vm_flags & VM_EXEC))
+			goto bad_area;
+	} else	/* Allow read even from write-only pages. */
+		if (!(vma->vm_flags & (VM_READ | VM_WRITE)))
+			goto bad_area;
+
+	/* If for any reason at all we couldn't handle the fault,
+	 * make sure we exit gracefully rather than endlessly redo
+	 * the fault.
+	 */
+survive:
+	switch (handle_mm_fault(mm, vma, address, is_write)) {
+	case VM_FAULT_MINOR:
+		current->min_flt++;
+		break;
+	case VM_FAULT_MAJOR:
+		current->maj_flt++;
+		break;
+	case VM_FAULT_SIGBUS:
+		goto do_sigbus;
+	case VM_FAULT_OOM:
+		goto out_of_memory;
+	default:
+		BUG();
+	}
+
+	up_read(&mm->mmap_sem);
+	return;
+
+	/* Something tried to access memory that isn't in our memory map..
+	 * Fix it, but check if it's kernel or user first..
+	 */
+bad_area:
+	up_read(&mm->mmap_sem);
+	if (user_mode(regs)) {
+		current->thread.bad_vaddr = address;
+		current->thread.error_code = is_write;
+		info.si_signo = SIGSEGV;
+		info.si_errno = 0;
+		/* info.si_code has been set above */
+		info.si_addr = (void *) address;
+		force_sig_info(SIGSEGV, &info, current);
+		return;
+	}
+	bad_page_fault(regs, address, SIGSEGV);
+	return;
+
+
+	/* We ran out of memory, or some other thing happened to us that made
+	 * us unable to handle the page fault gracefully.
+	 */
+out_of_memory:
+	up_read(&mm->mmap_sem);
+	if (current->pid == 1) {
+		yield();
+		down_read(&mm->mmap_sem);
+		goto survive;
+	}
+	printk("VM: killing process %s\n", current->comm);
+	if (user_mode(regs))
+		do_exit(SIGKILL);
+	bad_page_fault(regs, address, SIGKILL);
+	return;
+
+do_sigbus:
+	up_read(&mm->mmap_sem);
+
+	/* Send a sigbus, regardless of whether we were in kernel
+	 * or user mode.
+	 */
+	current->thread.bad_vaddr = address;
+	info.si_code = SIGBUS;
+	info.si_errno = 0;
+	info.si_code = BUS_ADRERR;
+	info.si_addr = (void *) address;
+	force_sig_info(SIGBUS, &info, current);
+
+	/* Kernel mode? Handle exceptions or die */
+	if (!user_mode(regs))
+		bad_page_fault(regs, address, SIGBUS);
+
+vmalloc_fault:
+	{
+		/* Synchronize this task's top level page-table
+		 * with the 'reference' page table.
+		 */
+		struct mm_struct *act_mm = current->active_mm;
+		int index = pgd_index(address);
+		pgd_t *pgd, *pgd_k;
+		pmd_t *pmd, *pmd_k;
+		pte_t *pte_k;
+
+		if (act_mm == NULL)
+			goto bad_page_fault;
+
+		pgd = act_mm->pgd + index;
+		pgd_k = init_mm.pgd + index;
+
+		if (!pgd_present(*pgd_k))
+			goto bad_page_fault;
+
+		pgd_val(*pgd) = pgd_val(*pgd_k);
+
+		pmd = pmd_offset(pgd, address);
+		pmd_k = pmd_offset(pgd_k, address);
+		if (!pmd_present(*pmd) || !pmd_present(*pmd_k))
+			goto bad_page_fault;
+
+		pmd_val(*pmd) = pmd_val(*pmd_k);
+		pte_k = pte_offset_kernel(pmd_k, address);
+
+		if (!pte_present(*pte_k))
+			goto bad_page_fault;
+		return;
+	}
+bad_page_fault:
+	bad_page_fault(regs, address, SIGKILL);
+	return;
+}
+
+
+void
+bad_page_fault(struct pt_regs *regs, unsigned long address, int sig)
+{
+	extern void die(const char*, struct pt_regs*, long);
+	const struct exception_table_entry *entry;
+
+	/* Are we prepared to handle this kernel fault?  */
+	if ((entry = search_exception_tables(regs->pc)) != NULL) {
+#if 1
+		printk(KERN_DEBUG "%s: Exception at pc=%#010lx (%lx)\n",
+				current->comm, regs->pc, entry->fixup);
+#endif
+		current->thread.bad_uaddr = address;
+		regs->pc = entry->fixup;
+		return;
+	}
+
+	/* Oops. The kernel tried to access some bad page. We'll have to
+	 * terminate things with extreme prejudice.
+	 */
+	printk(KERN_ALERT "Unable to handle kernel paging request at virtual "
+	       "address %08lx\n pc = %08lx, ra = %08lx\n",
+	       address, regs->pc, regs->areg[0]);
+	die("Oops", regs, sig);
+	do_exit(sig);
+}
+
diff --git a/arch/xtensa/mm/init.c b/arch/xtensa/mm/init.c
new file mode 100644
index 00000000000..56aace84aae
--- /dev/null
+++ b/arch/xtensa/mm/init.c
@@ -0,0 +1,551 @@
+/*
+ * arch/xtensa/mm/init.c
+ *
+ * Derived from MIPS, PPC.
+ *
+ * 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.
+ *
+ * Chris Zankel	<chris@zankel.net>
+ * Joe Taylor	<joe@tensilica.com, joetylr@yahoo.com>
+ * Marc Gauthier
+ * Kevin Chea
+ */
+
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/bootmem.h>
+#include <linux/swap.h>
+
+#include <asm/pgtable.h>
+#include <asm/bootparam.h>
+#include <asm/mmu_context.h>
+#include <asm/tlb.h>
+#include <asm/tlbflush.h>
+#include <asm/page.h>
+#include <asm/pgalloc.h>
+#include <asm/pgtable.h>
+
+
+#define DEBUG 0
+
+DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
+//static DEFINE_SPINLOCK(tlb_lock);
+
+/*
+ * This flag is used to indicate that the page was mapped and modified in
+ * kernel space, so the cache is probably dirty at that address.
+ * If cache aliasing is enabled and the page color mismatches, update_mmu_cache
+ * synchronizes the caches if this bit is set.
+ */
+
+#define PG_cache_clean PG_arch_1
+
+/* References to section boundaries */
+
+extern char _ftext, _etext, _fdata, _edata, _rodata_end;
+extern char __init_begin, __init_end;
+
+/*
+ * mem_reserve(start, end, must_exist)
+ *
+ * Reserve some memory from the memory pool.
+ *
+ * Parameters:
+ *  start	Start of region,
+ *  end		End of region,
+ *  must_exist	Must exist in memory pool.
+ *
+ * Returns:
+ *  0 (memory area couldn't be mapped)
+ * -1 (success)
+ */
+
+int __init mem_reserve(unsigned long start, unsigned long end, int must_exist)
+{
+	int i;
+
+	if (start == end)
+		return 0;
+
+	start = start & PAGE_MASK;
+	end = PAGE_ALIGN(end);
+
+	for (i = 0; i < sysmem.nr_banks; i++)
+		if (start < sysmem.bank[i].end
+		    && end >= sysmem.bank[i].start)
+			break;
+
+	if (i == sysmem.nr_banks) {
+		if (must_exist)
+			printk (KERN_WARNING "mem_reserve: [0x%0lx, 0x%0lx) "
+				"not in any region!\n", start, end);
+		return 0;
+	}
+
+	if (start > sysmem.bank[i].start) {
+		if (end < sysmem.bank[i].end) {
+			/* split entry */
+			if (sysmem.nr_banks >= SYSMEM_BANKS_MAX)
+				panic("meminfo overflow\n");
+			sysmem.bank[sysmem.nr_banks].start = end;
+			sysmem.bank[sysmem.nr_banks].end = sysmem.bank[i].end;
+			sysmem.nr_banks++;
+		}
+		sysmem.bank[i].end = start;
+	} else {
+		if (end < sysmem.bank[i].end)
+			sysmem.bank[i].start = end;
+		else {
+			/* remove entry */
+			sysmem.nr_banks--;
+			sysmem.bank[i].start = sysmem.bank[sysmem.nr_banks].start;
+			sysmem.bank[i].end   = sysmem.bank[sysmem.nr_banks].end;
+		}
+	}
+	return -1;
+}
+
+
+/*
+ * Initialize the bootmem system and give it all the memory we have available.
+ */
+
+void __init bootmem_init(void)
+{
+	unsigned long pfn;
+	unsigned long bootmap_start, bootmap_size;
+	int i;
+
+	max_low_pfn = max_pfn = 0;
+	min_low_pfn = ~0;
+
+	for (i=0; i < sysmem.nr_banks; i++) {
+		pfn = PAGE_ALIGN(sysmem.bank[i].start) >> PAGE_SHIFT;
+		if (pfn < min_low_pfn)
+			min_low_pfn = pfn;
+		pfn = PAGE_ALIGN(sysmem.bank[i].end - 1) >> PAGE_SHIFT;
+		if (pfn > max_pfn)
+			max_pfn = pfn;
+	}
+
+	if (min_low_pfn > max_pfn)
+		panic("No memory found!\n");
+
+	max_low_pfn = max_pfn < MAX_LOW_MEMORY >> PAGE_SHIFT ?
+		max_pfn : MAX_LOW_MEMORY >> PAGE_SHIFT;
+
+	/* Find an area to use for the bootmem bitmap. */
+
+	bootmap_size = bootmem_bootmap_pages(max_low_pfn) << PAGE_SHIFT;
+	bootmap_start = ~0;
+
+	for (i=0; i<sysmem.nr_banks; i++)
+		if (sysmem.bank[i].end - sysmem.bank[i].start >= bootmap_size) {
+			bootmap_start = sysmem.bank[i].start;
+			break;
+		}
+
+	if (bootmap_start == ~0UL)
+		panic("Cannot find %ld bytes for bootmap\n", bootmap_size);
+
+	/* Reserve the bootmem bitmap area */
+
+	mem_reserve(bootmap_start, bootmap_start + bootmap_size, 1);
+	bootmap_size = init_bootmem_node(NODE_DATA(0), min_low_pfn,
+					 bootmap_start >> PAGE_SHIFT,
+					 max_low_pfn);
+
+	/* Add all remaining memory pieces into the bootmem map */
+
+	for (i=0; i<sysmem.nr_banks; i++)
+		free_bootmem(sysmem.bank[i].start,
+			     sysmem.bank[i].end - sysmem.bank[i].start);
+
+}
+
+
+void __init paging_init(void)
+{
+	unsigned long zones_size[MAX_NR_ZONES];
+	int i;
+
+	/* All pages are DMA-able, so we put them all in the DMA zone. */
+
+	zones_size[ZONE_DMA] = max_low_pfn;
+	for (i = 1; i < MAX_NR_ZONES; i++)
+		zones_size[i] = 0;
+
+#ifdef CONFIG_HIGHMEM
+	zones_size[ZONE_HIGHMEM] = max_pfn - max_low_pfn;
+#endif
+
+	/* Initialize the kernel's page tables. */
+
+	memset(swapper_pg_dir, 0, PAGE_SIZE);
+
+	free_area_init(zones_size);
+}
+
+/*
+ * Flush the mmu and reset associated register to default values.
+ */
+
+void __init init_mmu (void)
+{
+	/* Writing zeros to the <t>TLBCFG special registers ensure
+	 * that valid values exist in the register.  For existing
+	 * PGSZID<w> fields, zero selects the first element of the
+	 * page-size array.  For nonexistant PGSZID<w> fields, zero is
+	 * the best value to write.  Also, when changing PGSZID<w>
+	 * fields, the corresponding TLB must be flushed.
+	 */
+	set_itlbcfg_register (0);
+	set_dtlbcfg_register (0);
+	flush_tlb_all ();
+
+	/* Set rasid register to a known value. */
+
+	set_rasid_register (ASID_ALL_RESERVED);
+
+	/* Set PTEVADDR special register to the start of the page
+	 * table, which is in kernel mappable space (ie. not
+	 * statically mapped).  This register's value is undefined on
+	 * reset.
+	 */
+	set_ptevaddr_register (PGTABLE_START);
+}
+
+/*
+ * Initialize memory pages.
+ */
+
+void __init mem_init(void)
+{
+	unsigned long codesize, reservedpages, datasize, initsize;
+	unsigned long highmemsize, tmp, ram;
+
+	max_mapnr = num_physpages = max_low_pfn;
+	high_memory = (void *) __va(max_mapnr << PAGE_SHIFT);
+	highmemsize = 0;
+
+#if CONFIG_HIGHMEM
+#error HIGHGMEM not implemented in init.c
+#endif
+
+	totalram_pages += free_all_bootmem();
+
+	reservedpages = ram = 0;
+	for (tmp = 0; tmp < max_low_pfn; tmp++) {
+		ram++;
+		if (PageReserved(mem_map+tmp))
+			reservedpages++;
+	}
+
+	codesize =  (unsigned long) &_etext - (unsigned long) &_ftext;
+	datasize =  (unsigned long) &_edata - (unsigned long) &_fdata;
+	initsize =  (unsigned long) &__init_end - (unsigned long) &__init_begin;
+
+	printk("Memory: %luk/%luk available (%ldk kernel code, %ldk reserved, "
+	       "%ldk data, %ldk init %ldk highmem)\n",
+	       (unsigned long) nr_free_pages() << (PAGE_SHIFT-10),
+	       ram << (PAGE_SHIFT-10),
+	       codesize >> 10,
+	       reservedpages << (PAGE_SHIFT-10),
+	       datasize >> 10,
+	       initsize >> 10,
+	       highmemsize >> 10);
+}
+
+void
+free_reserved_mem(void *start, void *end)
+{
+	for (; start < end; start += PAGE_SIZE) {
+		ClearPageReserved(virt_to_page(start));
+		set_page_count(virt_to_page(start), 1);
+		free_page((unsigned long)start);
+		totalram_pages++;
+	}
+}
+
+#ifdef CONFIG_BLK_DEV_INITRD
+extern int initrd_is_mapped;
+
+void free_initrd_mem(unsigned long start, unsigned long end)
+{
+	if (initrd_is_mapped) {
+		free_reserved_mem((void*)start, (void*)end);
+		printk ("Freeing initrd memory: %ldk freed\n",(end-start)>>10);
+	}
+}
+#endif
+
+void free_initmem(void)
+{
+	free_reserved_mem(&__init_begin, &__init_end);
+	printk("Freeing unused kernel memory: %dk freed\n",
+	       (&__init_end - &__init_begin) >> 10);
+}
+
+void show_mem(void)
+{
+	int i, free = 0, total = 0, reserved = 0;
+	int shared = 0, cached = 0;
+
+	printk("Mem-info:\n");
+	show_free_areas();
+	printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
+	i = max_mapnr;
+	while (i-- > 0) {
+		total++;
+		if (PageReserved(mem_map+i))
+			reserved++;
+		else if (PageSwapCache(mem_map+i))
+			cached++;
+		else if (!page_count(mem_map + i))
+			free++;
+		else
+			shared += page_count(mem_map + i) - 1;
+	}
+	printk("%d pages of RAM\n", total);
+	printk("%d reserved pages\n", reserved);
+	printk("%d pages shared\n", shared);
+	printk("%d pages swap cached\n",cached);
+	printk("%d free pages\n", free);
+}
+
+/* ------------------------------------------------------------------------- */
+
+#if (DCACHE_WAY_SIZE > PAGE_SIZE)
+
+/*
+ * With cache aliasing, the page color of the page in kernel space and user
+ * space might mismatch. We temporarily map the page to a different virtual
+ * address with the same color and clear the page there.
+ */
+
+void clear_user_page(void *kaddr, unsigned long vaddr, struct page* page)
+{
+
+  	/*  There shouldn't be any entries for this page. */
+
+	__flush_invalidate_dcache_page_phys(__pa(page_address(page)));
+
+	if (!PAGE_COLOR_EQ(vaddr, kaddr)) {
+		unsigned long v, p;
+
+		/* Temporarily map page to DTLB_WAY_DCACHE_ALIAS0. */
+
+		spin_lock(&tlb_lock);
+
+		p = (unsigned long)pte_val((mk_pte(page,PAGE_KERNEL)));
+		kaddr = (void*)PAGE_COLOR_MAP0(vaddr);
+		v = (unsigned long)kaddr | DTLB_WAY_DCACHE_ALIAS0;
+		__asm__ __volatile__("wdtlb %0,%1; dsync" : :"a" (p), "a" (v));
+
+		clear_page(kaddr);
+
+		spin_unlock(&tlb_lock);
+	} else {
+		clear_page(kaddr);
+	}
+
+	/* We need to make sure that i$ and d$ are coherent. */
+
+	clear_bit(PG_cache_clean, &page->flags);
+}
+
+/*
+ * With cache aliasing, we have to make sure that the page color of the page
+ * in kernel space matches that of the virtual user address before we read
+ * the page. If the page color differ, we create a temporary DTLB entry with
+ * the corrent page color and use this 'temporary' address as the source.
+ * We then use the same approach as in clear_user_page and copy the data
+ * to the kernel space and clear the PG_cache_clean bit to synchronize caches
+ * later.
+ *
+ * Note:
+ * Instead of using another 'way' for the temporary DTLB entry, we could
+ * probably use the same entry that points to the kernel address (after
+ * saving the original value and restoring it when we are done).
+ */
+
+void copy_user_page(void* to, void* from, unsigned long vaddr,
+    		    struct page* to_page)
+{
+	/* There shouldn't be any entries for the new page. */
+
+	__flush_invalidate_dcache_page_phys(__pa(page_address(to_page)));
+
+	spin_lock(&tlb_lock);
+
+	if (!PAGE_COLOR_EQ(vaddr, from)) {
+		unsigned long v, p, t;
+
+		__asm__ __volatile__ ("pdtlb %1,%2; rdtlb1 %0,%1"
+				      : "=a"(p), "=a"(t) : "a"(from));
+		from = (void*)PAGE_COLOR_MAP0(vaddr);
+		v = (unsigned long)from | DTLB_WAY_DCACHE_ALIAS0;
+		__asm__ __volatile__ ("wdtlb %0,%1; dsync" ::"a" (p), "a" (v));
+	}
+
+	if (!PAGE_COLOR_EQ(vaddr, to)) {
+		unsigned long v, p;
+
+		p = (unsigned long)pte_val((mk_pte(to_page,PAGE_KERNEL)));
+		to = (void*)PAGE_COLOR_MAP1(vaddr);
+		v = (unsigned long)to | DTLB_WAY_DCACHE_ALIAS1;
+		__asm__ __volatile__ ("wdtlb %0,%1; dsync" ::"a" (p), "a" (v));
+	}
+	copy_page(to, from);
+
+	spin_unlock(&tlb_lock);
+
+	/* We need to make sure that i$ and d$ are coherent. */
+
+	clear_bit(PG_cache_clean, &to_page->flags);
+}
+
+
+
+/*
+ * Any time the kernel writes to a user page cache page, or it is about to
+ * read from a page cache page this routine is called.
+ *
+ * Note:
+ * The kernel currently only provides one architecture bit in the page
+ * flags that we use for I$/D$ coherency. Maybe, in future, we can
+ * use a sepearte bit for deferred dcache aliasing:
+ * If the page is not mapped yet, we only need to set a flag,
+ * if mapped, we need to invalidate the page.
+ */
+// FIXME: we probably need this for WB caches not only for Page Coloring..
+
+void flush_dcache_page(struct page *page)
+{
+	unsigned long addr = __pa(page_address(page));
+	struct address_space *mapping = page_mapping(page);
+
+	__flush_invalidate_dcache_page_phys(addr);
+
+	if (!test_bit(PG_cache_clean, &page->flags))
+		return;
+
+	/* If this page hasn't been mapped, yet, handle I$/D$ coherency later.*/
+#if 0
+	if (mapping && !mapping_mapped(mapping))
+		clear_bit(PG_cache_clean, &page->flags);
+	else
+#endif
+		__invalidate_icache_page_phys(addr);
+}
+
+void flush_cache_range(struct vm_area_struct* vma, unsigned long s,
+		       unsigned long e)
+{
+	__flush_invalidate_cache_all();
+}
+
+void flush_cache_page(struct vm_area_struct* vma, unsigned long address,
+    		      unsigned long pfn)
+{
+	struct page *page = pfn_to_page(pfn);
+
+	/* Remove any entry for the old mapping. */
+
+	if (current->active_mm == vma->vm_mm) {
+		unsigned long addr = __pa(page_address(page));
+		__flush_invalidate_dcache_page_phys(addr);
+		if ((vma->vm_flags & VM_EXEC) != 0)
+			__invalidate_icache_page_phys(addr);
+	} else {
+		BUG();
+	}
+}
+
+#endif	/* (DCACHE_WAY_SIZE > PAGE_SIZE) */
+
+
+pte_t* pte_alloc_one_kernel (struct mm_struct* mm, unsigned long addr)
+{
+	pte_t* pte = (pte_t*)__get_free_pages(GFP_KERNEL|__GFP_REPEAT, 0);
+	if (likely(pte)) {
+	       	pte_t* ptep = (pte_t*)(pte_val(*pte) + PAGE_OFFSET);
+		int i;
+		for (i = 0; i < 1024; i++, ptep++)
+			pte_clear(mm, addr, ptep);
+	}
+	return pte;
+}
+
+struct page* pte_alloc_one(struct mm_struct *mm, unsigned long addr)
+{
+	struct page *page;
+
+	page = alloc_pages(GFP_KERNEL | __GFP_REPEAT, 0);
+
+	if (likely(page)) {
+		pte_t* ptep = kmap_atomic(page, KM_USER0);
+		int i;
+
+		for (i = 0; i < 1024; i++, ptep++)
+			pte_clear(mm, addr, ptep);
+
+		kunmap_atomic(ptep, KM_USER0);
+	}
+	return page;
+}
+
+
+/*
+ * Handle D$/I$ coherency.
+ *
+ * Note:
+ * We only have one architecture bit for the page flags, so we cannot handle
+ * cache aliasing, yet.
+ */
+
+void
+update_mmu_cache(struct vm_area_struct * vma, unsigned long addr, pte_t pte)
+{
+	unsigned long pfn = pte_pfn(pte);
+	struct page *page;
+	unsigned long vaddr = addr & PAGE_MASK;
+
+	if (!pfn_valid(pfn))
+		return;
+
+	page = pfn_to_page(pfn);
+
+	invalidate_itlb_mapping(addr);
+	invalidate_dtlb_mapping(addr);
+
+	/* We have a new mapping. Use it. */
+
+	write_dtlb_entry(pte, dtlb_probe(addr));
+
+	/* If the processor can execute from this page, synchronize D$/I$. */
+
+	if ((vma->vm_flags & VM_EXEC) != 0) {
+
+		write_itlb_entry(pte, itlb_probe(addr));
+
+		/* Synchronize caches, if not clean. */
+
+		if (!test_and_set_bit(PG_cache_clean, &page->flags)) {
+			__flush_dcache_page(vaddr);
+			__invalidate_icache_page(vaddr);
+		}
+	}
+}
+
diff --git a/arch/xtensa/mm/misc.S b/arch/xtensa/mm/misc.S
new file mode 100644
index 00000000000..327c0f17187
--- /dev/null
+++ b/arch/xtensa/mm/misc.S
@@ -0,0 +1,374 @@
+/*
+ * arch/xtensa/mm/misc.S
+ *
+ * Miscellaneous assembly 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.
+ *
+ * Chris Zankel	<chris@zankel.net>
+ */
+
+/* Note: we might want to implement some of the loops as zero-overhead-loops,
+ *	 where applicable and if supported by the processor.
+ */
+
+#include <linux/linkage.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+
+#include <xtensa/cacheasm.h>
+#include <xtensa/cacheattrasm.h>
+
+/* clear_page (page) */
+
+ENTRY(clear_page)
+	entry	a1, 16
+	addi	a4, a2, PAGE_SIZE
+	movi	a3, 0
+
+1:	s32i	a3, a2, 0
+	s32i	a3, a2, 4
+	s32i	a3, a2, 8
+	s32i	a3, a2, 12
+	s32i	a3, a2, 16
+	s32i	a3, a2, 20
+	s32i	a3, a2, 24
+	s32i	a3, a2, 28
+	addi	a2, a2, 32
+	blt	a2, a4, 1b
+
+	retw
+
+/*
+ * copy_page (void *to, void *from)
+ *                  a2        a3
+ */
+
+ENTRY(copy_page)
+	entry	a1, 16
+	addi	a4, a2, PAGE_SIZE
+
+1:	l32i	a5, a3, 0
+	l32i	a6, a3, 4
+	l32i	a7, a3, 8
+	s32i	a5, a2, 0
+	s32i	a6, a2, 4
+	s32i	a7, a2, 8
+	l32i	a5, a3, 12
+	l32i	a6, a3, 16
+	l32i	a7, a3, 20
+	s32i	a5, a2, 12
+	s32i	a6, a2, 16
+	s32i	a7, a2, 20
+	l32i	a5, a3, 24
+	l32i	a6, a3, 28
+	s32i	a5, a2, 24
+	s32i	a6, a2, 28
+	addi	a2, a2, 32
+	addi	a3, a3, 32
+	blt	a2, a4, 1b
+
+	retw
+
+
+/*
+ * void __flush_invalidate_cache_all(void)
+ */
+
+ENTRY(__flush_invalidate_cache_all)
+	entry	sp, 16
+	dcache_writeback_inv_all a2, a3
+	icache_invalidate_all a2, a3
+	retw
+
+/*
+ * void __invalidate_icache_all(void)
+ */
+
+ENTRY(__invalidate_icache_all)
+	entry	sp, 16
+	icache_invalidate_all a2, a3
+	retw
+
+/*
+ * void __flush_invalidate_dcache_all(void)
+ */
+
+ENTRY(__flush_invalidate_dcache_all)
+	entry	sp, 16
+	dcache_writeback_inv_all a2, a3
+	retw
+
+
+/*
+ * void __flush_invalidate_cache_range(ulong start, ulong size)
+ */
+
+ENTRY(__flush_invalidate_cache_range)
+	entry	sp, 16
+	mov	a4, a2
+	mov	a5, a3
+	dcache_writeback_inv_region a4, a5, a6
+	icache_invalidate_region a2, a3, a4
+	retw
+
+/*
+ * void __invalidate_icache_page(ulong start)
+ */
+
+ENTRY(__invalidate_icache_page)
+	entry	sp, 16
+	movi	a3, PAGE_SIZE
+	icache_invalidate_region a2, a3, a4
+	retw
+
+/*
+ * void __invalidate_dcache_page(ulong start)
+ */
+
+ENTRY(__invalidate_dcache_page)
+	entry	sp, 16
+	movi	a3, PAGE_SIZE
+	dcache_invalidate_region a2, a3, a4
+	retw
+
+/*
+ * void __invalidate_icache_range(ulong start, ulong size)
+ */
+
+ENTRY(__invalidate_icache_range)
+	entry	sp, 16
+	icache_invalidate_region a2, a3, a4
+	retw
+
+/*
+ * void __invalidate_dcache_range(ulong start, ulong size)
+ */
+
+ENTRY(__invalidate_dcache_range)
+	entry	sp, 16
+	dcache_invalidate_region a2, a3, a4
+	retw
+
+/*
+ * void __flush_dcache_page(ulong start)
+ */
+
+ENTRY(__flush_dcache_page)
+	entry	sp, 16
+	movi	a3, PAGE_SIZE
+	dcache_writeback_region a2, a3, a4
+	retw
+
+/*
+ * void __flush_invalidate_dcache_page(ulong start)
+ */
+
+ENTRY(__flush_invalidate_dcache_page)
+	entry	sp, 16
+	movi	a3, PAGE_SIZE
+	dcache_writeback_inv_region a2, a3, a4
+	retw
+
+/*
+ * void __flush_invalidate_dcache_range(ulong start, ulong size)
+ */
+
+ENTRY(__flush_invalidate_dcache_range)
+	entry	sp, 16
+	dcache_writeback_inv_region a2, a3, a4
+	retw
+
+/*
+ * void __invalidate_dcache_all(void)
+ */
+
+ENTRY(__invalidate_dcache_all)
+	entry	sp, 16
+	dcache_invalidate_all a2, a3
+	retw
+
+/*
+ * void __flush_invalidate_dcache_page_phys(ulong start)
+ */
+
+ENTRY(__flush_invalidate_dcache_page_phys)
+	entry	sp, 16
+
+	movi	a3, XCHAL_DCACHE_SIZE
+	movi	a4, PAGE_MASK | 1
+	addi	a2, a2, 1
+
+1:	addi	a3, a3, -XCHAL_DCACHE_LINESIZE
+
+	ldct	a6, a3
+	dsync
+	and	a6, a6, a4
+	beq	a6, a2, 2f
+	bgeui	a3, 2, 1b
+	retw
+
+2:	diwbi	a3, 0
+	bgeui	a3, 2, 1b
+	retw
+
+ENTRY(check_dcache_low0)
+	entry	sp, 16
+
+	movi	a3, XCHAL_DCACHE_SIZE / 4
+	movi	a4, PAGE_MASK | 1
+	addi	a2, a2, 1
+
+1:	addi	a3, a3, -XCHAL_DCACHE_LINESIZE
+
+	ldct	a6, a3
+	dsync
+	and	a6, a6, a4
+	beq	a6, a2, 2f
+	bgeui	a3, 2, 1b
+	retw
+
+2:	j 2b
+
+ENTRY(check_dcache_high0)
+	entry	sp, 16
+
+	movi	a5, XCHAL_DCACHE_SIZE / 4
+	movi	a3, XCHAL_DCACHE_SIZE / 2
+	movi	a4, PAGE_MASK | 1
+	addi	a2, a2, 1
+
+1:	addi	a3, a3, -XCHAL_DCACHE_LINESIZE
+	addi	a5, a5, -XCHAL_DCACHE_LINESIZE
+
+	ldct	a6, a3
+	dsync
+	and	a6, a6, a4
+	beq	a6, a2, 2f
+	bgeui	a5, 2, 1b
+	retw
+
+2:	j 2b
+
+ENTRY(check_dcache_low1)
+	entry	sp, 16
+
+	movi	a5, XCHAL_DCACHE_SIZE / 4
+	movi	a3, XCHAL_DCACHE_SIZE * 3 / 4
+	movi	a4, PAGE_MASK | 1
+	addi	a2, a2, 1
+
+1:	addi	a3, a3, -XCHAL_DCACHE_LINESIZE
+	addi	a5, a5, -XCHAL_DCACHE_LINESIZE
+
+	ldct	a6, a3
+	dsync
+	and	a6, a6, a4
+	beq	a6, a2, 2f
+	bgeui	a5, 2, 1b
+	retw
+
+2:	j 2b
+
+ENTRY(check_dcache_high1)
+	entry	sp, 16
+
+	movi	a5, XCHAL_DCACHE_SIZE / 4
+	movi	a3, XCHAL_DCACHE_SIZE
+	movi	a4, PAGE_MASK | 1
+	addi	a2, a2, 1
+
+1:	addi	a3, a3, -XCHAL_DCACHE_LINESIZE
+	addi	a5, a5, -XCHAL_DCACHE_LINESIZE
+
+	ldct	a6, a3
+	dsync
+	and	a6, a6, a4
+	beq	a6, a2, 2f
+	bgeui	a5, 2, 1b
+	retw
+
+2:	j 2b
+
+
+/*
+ * void __invalidate_icache_page_phys(ulong start)
+ */
+
+ENTRY(__invalidate_icache_page_phys)
+	entry	sp, 16
+
+	movi	a3, XCHAL_ICACHE_SIZE
+	movi	a4, PAGE_MASK | 1
+	addi	a2, a2, 1
+
+1:	addi	a3, a3, -XCHAL_ICACHE_LINESIZE
+
+	lict	a6, a3
+	isync
+	and	a6, a6, a4
+	beq	a6, a2, 2f
+	bgeui	a3, 2, 1b
+	retw
+
+2:	iii	a3, 0
+	bgeui	a3, 2, 1b
+	retw
+
+
+#if 0
+
+	movi	a3, XCHAL_DCACHE_WAYS - 1
+	movi	a4, PAGE_SIZE
+
+1:	mov	a5, a2
+	add	a6, a2, a4
+
+2:	diwbi	a5, 0
+	diwbi	a5, XCHAL_DCACHE_LINESIZE
+	diwbi	a5, XCHAL_DCACHE_LINESIZE * 2
+	diwbi	a5, XCHAL_DCACHE_LINESIZE * 3
+
+	addi	a5, a5, XCHAL_DCACHE_LINESIZE * 4
+	blt	a5, a6, 2b
+
+	addi	a3, a3, -1
+	addi	a2, a2, XCHAL_DCACHE_SIZE / XCHAL_DCACHE_WAYS
+	bgez	a3, 1b
+
+	retw
+
+ENTRY(__invalidate_icache_page_index)
+	entry	sp, 16
+
+	movi	a3, XCHAL_ICACHE_WAYS - 1
+	movi	a4, PAGE_SIZE
+
+1:	mov	a5, a2
+	add	a6, a2, a4
+
+2:	iii	a5, 0
+	iii	a5, XCHAL_ICACHE_LINESIZE
+	iii	a5, XCHAL_ICACHE_LINESIZE * 2
+	iii	a5, XCHAL_ICACHE_LINESIZE * 3
+
+	addi	a5, a5, XCHAL_ICACHE_LINESIZE * 4
+	blt	a5, a6, 2b
+
+	addi	a3, a3, -1
+	addi	a2, a2, XCHAL_ICACHE_SIZE / XCHAL_ICACHE_WAYS
+	bgez	a3, 2b
+
+	retw
+
+#endif
+
+
+
+
+
+
diff --git a/arch/xtensa/mm/pgtable.c b/arch/xtensa/mm/pgtable.c
new file mode 100644
index 00000000000..e5e119c820e
--- /dev/null
+++ b/arch/xtensa/mm/pgtable.c
@@ -0,0 +1,76 @@
+/*
+ * arch/xtensa/mm/fault.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.
+ *
+ * Chris Zankel <chris@zankel.net>
+ */
+
+#if (DCACHE_SIZE > PAGE_SIZE)
+
+pte_t* pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address)
+{
+	pte_t *pte, p;
+	int color = ADDR_COLOR(address);
+	int i;
+
+	p = (pte_t*) __get_free_pages(GFP_KERNEL|__GFP_REPEAT, COLOR_ORDER);
+
+	if (likely(p)) {
+		struct page *page;
+
+		for (i = 0; i < COLOR_SIZE; i++, p++) {
+			page = virt_to_page(pte);
+
+			set_page_count(page, 1);
+			ClearPageCompound(page);
+
+			if (ADDR_COLOR(p) == color)
+				pte = p;
+			else
+				free_page(p);
+		}
+		clear_page(pte);
+	}
+	return pte;
+}
+
+#ifdef PROFILING
+
+int mask;
+int hit;
+int flush;
+
+#endif
+
+struct page* pte_alloc_one(struct mm_struct *mm, unsigned long address)
+{
+	struct page *page, p;
+	int color = ADDR_COLOR(address);
+
+	p = alloc_pages(GFP_KERNEL | __GFP_REPEAT, PTE_ORDER);
+
+	if (likely(p)) {
+		for (i = 0; i < PAGE_ORDER; i++) {
+			set_page_count(p, 1);
+			ClearPageCompound(p);
+
+			if (PADDR_COLOR(page_address(pg)) == color)
+				page = p;
+			else
+				free_page(p);
+		}
+		clear_highpage(page);
+	}
+
+	return page;
+}
+
+#endif
+
+
+
diff --git a/arch/xtensa/mm/tlb.c b/arch/xtensa/mm/tlb.c
new file mode 100644
index 00000000000..d3bd3bfc3b3
--- /dev/null
+++ b/arch/xtensa/mm/tlb.c
@@ -0,0 +1,545 @@
+/*
+ * arch/xtensa/mm/mmu.c
+ *
+ * Logic that manipulates the Xtensa MMU.  Derived 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) 2001 - 2003 Tensilica Inc.
+ *
+ * Joe Taylor
+ * Chris Zankel	<chris@zankel.net>
+ * Marc Gauthier
+ */
+
+#include <linux/mm.h>
+#include <asm/processor.h>
+#include <asm/mmu_context.h>
+#include <asm/tlbflush.h>
+#include <asm/system.h>
+#include <asm/cacheflush.h>
+
+
+static inline void __flush_itlb_all (void)
+{
+	int way, index;
+
+	for (way = 0; way < XCHAL_ITLB_ARF_WAYS; way++) {
+		for (index = 0; index < ITLB_ENTRIES_PER_ARF_WAY; index++) {
+			int entry = way + (index << PAGE_SHIFT);
+			invalidate_itlb_entry_no_isync (entry);
+		}
+	}
+	asm volatile ("isync\n");
+}
+
+static inline void __flush_dtlb_all (void)
+{
+	int way, index;
+
+	for (way = 0; way < XCHAL_DTLB_ARF_WAYS; way++) {
+		for (index = 0; index < DTLB_ENTRIES_PER_ARF_WAY; index++) {
+			int entry = way + (index << PAGE_SHIFT);
+			invalidate_dtlb_entry_no_isync (entry);
+		}
+	}
+	asm volatile ("isync\n");
+}
+
+
+void flush_tlb_all (void)
+{
+	__flush_itlb_all();
+	__flush_dtlb_all();
+}
+
+/* If mm is current, we simply assign the current task a new ASID, thus,
+ * invalidating all previous tlb entries. If mm is someone else's user mapping,
+ * wie invalidate the context, thus, when that user mapping is swapped in,
+ * a new context will be assigned to it.
+ */
+
+void flush_tlb_mm(struct mm_struct *mm)
+{
+#if 0
+	printk("[tlbmm<%lx>]\n", (unsigned long)mm->context);
+#endif
+
+	if (mm == current->active_mm) {
+		int flags;
+		local_save_flags(flags);
+		get_new_mmu_context(mm, asid_cache);
+		set_rasid_register(ASID_INSERT(mm->context));
+		local_irq_restore(flags);
+	}
+	else
+		mm->context = 0;
+}
+
+void flush_tlb_range (struct vm_area_struct *vma,
+    		      unsigned long start, unsigned long end)
+{
+	struct mm_struct *mm = vma->vm_mm;
+	unsigned long flags;
+
+	if (mm->context == NO_CONTEXT)
+		return;
+
+#if 0
+	printk("[tlbrange<%02lx,%08lx,%08lx>]\n",
+			(unsigned long)mm->context, start, end);
+#endif
+	local_save_flags(flags);
+
+	if (end-start + (PAGE_SIZE-1) <= SMALLEST_NTLB_ENTRIES << PAGE_SHIFT) {
+		int oldpid = get_rasid_register();
+		set_rasid_register (ASID_INSERT(mm->context));
+		start &= PAGE_MASK;
+ 		if (vma->vm_flags & VM_EXEC)
+			while(start < end) {
+				invalidate_itlb_mapping(start);
+				invalidate_dtlb_mapping(start);
+				start += PAGE_SIZE;
+			}
+		else
+			while(start < end) {
+				invalidate_dtlb_mapping(start);
+				start += PAGE_SIZE;
+			}
+
+		set_rasid_register(oldpid);
+	} else {
+		get_new_mmu_context(mm, asid_cache);
+		if (mm == current->active_mm)
+			set_rasid_register(ASID_INSERT(mm->context));
+	}
+	local_irq_restore(flags);
+}
+
+void flush_tlb_page (struct vm_area_struct *vma, unsigned long page)
+{
+	struct mm_struct* mm = vma->vm_mm;
+	unsigned long flags;
+	int oldpid;
+#if 0
+	printk("[tlbpage<%02lx,%08lx>]\n",
+			(unsigned long)mm->context, page);
+#endif
+
+	if(mm->context == NO_CONTEXT)
+		return;
+
+	local_save_flags(flags);
+
+       	oldpid = get_rasid_register();
+
+	if (vma->vm_flags & VM_EXEC)
+		invalidate_itlb_mapping(page);
+	invalidate_dtlb_mapping(page);
+
+	set_rasid_register(oldpid);
+
+	local_irq_restore(flags);
+
+#if 0
+	flush_tlb_all();
+	return;
+#endif
+}
+
+
+#ifdef DEBUG_TLB
+
+#define USE_ITLB  0
+#define USE_DTLB  1
+
+struct way_config_t {
+	int indicies;
+	int indicies_log2;
+	int pgsz_log2;
+	int arf;
+};
+
+static struct way_config_t itlb[XCHAL_ITLB_WAYS] =
+{
+	{ XCHAL_ITLB_SET(XCHAL_ITLB_WAY0_SET, ENTRIES),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY0_SET, ENTRIES_LOG2),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY0_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY0_SET, ARF)
+	},
+	{ XCHAL_ITLB_SET(XCHAL_ITLB_WAY1_SET, ENTRIES),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY1_SET, ENTRIES_LOG2),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY1_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY1_SET, ARF)
+	},
+	{ XCHAL_ITLB_SET(XCHAL_ITLB_WAY2_SET, ENTRIES),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY2_SET, ENTRIES_LOG2),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY2_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY2_SET, ARF)
+	},
+	{ XCHAL_ITLB_SET(XCHAL_ITLB_WAY3_SET, ENTRIES),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY3_SET, ENTRIES_LOG2),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY3_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY3_SET, ARF)
+	},
+	{ XCHAL_ITLB_SET(XCHAL_ITLB_WAY4_SET, ENTRIES),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY4_SET, ENTRIES_LOG2),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY4_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY4_SET, ARF)
+	},
+	{ XCHAL_ITLB_SET(XCHAL_ITLB_WAY5_SET, ENTRIES),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY5_SET, ENTRIES_LOG2),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY5_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY5_SET, ARF)
+	},
+	{ XCHAL_ITLB_SET(XCHAL_ITLB_WAY6_SET, ENTRIES),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY6_SET, ENTRIES_LOG2),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY6_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_ITLB_SET(XCHAL_ITLB_WAY6_SET, ARF)
+	}
+};
+
+static struct way_config_t dtlb[XCHAL_DTLB_WAYS] =
+{
+	{ XCHAL_DTLB_SET(XCHAL_DTLB_WAY0_SET, ENTRIES),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY0_SET, ENTRIES_LOG2),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY0_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY0_SET, ARF)
+	},
+	{ XCHAL_DTLB_SET(XCHAL_DTLB_WAY1_SET, ENTRIES),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY1_SET, ENTRIES_LOG2),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY1_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY1_SET, ARF)
+	},
+	{ XCHAL_DTLB_SET(XCHAL_DTLB_WAY2_SET, ENTRIES),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY2_SET, ENTRIES_LOG2),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY2_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY2_SET, ARF)
+	},
+	{ XCHAL_DTLB_SET(XCHAL_DTLB_WAY3_SET, ENTRIES),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY3_SET, ENTRIES_LOG2),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY3_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY3_SET, ARF)
+	},
+	{ XCHAL_DTLB_SET(XCHAL_DTLB_WAY4_SET, ENTRIES),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY4_SET, ENTRIES_LOG2),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY4_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY4_SET, ARF)
+	},
+	{ XCHAL_DTLB_SET(XCHAL_DTLB_WAY5_SET, ENTRIES),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY5_SET, ENTRIES_LOG2),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY5_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY5_SET, ARF)
+	},
+	{ XCHAL_DTLB_SET(XCHAL_DTLB_WAY6_SET, ENTRIES),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY6_SET, ENTRIES_LOG2),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY6_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY6_SET, ARF)
+	},
+	{ XCHAL_DTLB_SET(XCHAL_DTLB_WAY7_SET, ENTRIES),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY7_SET, ENTRIES_LOG2),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY7_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY7_SET, ARF)
+	},
+	{ XCHAL_DTLB_SET(XCHAL_DTLB_WAY8_SET, ENTRIES),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY8_SET, ENTRIES_LOG2),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY8_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY8_SET, ARF)
+	},
+	{ XCHAL_DTLB_SET(XCHAL_DTLB_WAY9_SET, ENTRIES),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY9_SET, ENTRIES_LOG2),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY9_SET, PAGESZ_LOG2_MIN),
+	  XCHAL_DTLB_SET(XCHAL_DTLB_WAY9_SET, ARF)
+	}
+};
+
+/*  Total number of entries:  */
+#define ITLB_TOTAL_ENTRIES	\
+		XCHAL_ITLB_SET(XCHAL_ITLB_WAY0_SET, ENTRIES) + \
+		XCHAL_ITLB_SET(XCHAL_ITLB_WAY1_SET, ENTRIES) + \
+		XCHAL_ITLB_SET(XCHAL_ITLB_WAY2_SET, ENTRIES) + \
+		XCHAL_ITLB_SET(XCHAL_ITLB_WAY3_SET, ENTRIES) + \
+		XCHAL_ITLB_SET(XCHAL_ITLB_WAY4_SET, ENTRIES) + \
+		XCHAL_ITLB_SET(XCHAL_ITLB_WAY5_SET, ENTRIES) + \
+		XCHAL_ITLB_SET(XCHAL_ITLB_WAY6_SET, ENTRIES)
+#define DTLB_TOTAL_ENTRIES	\
+		XCHAL_DTLB_SET(XCHAL_DTLB_WAY0_SET, ENTRIES) + \
+		XCHAL_DTLB_SET(XCHAL_DTLB_WAY1_SET, ENTRIES) + \
+		XCHAL_DTLB_SET(XCHAL_DTLB_WAY2_SET, ENTRIES) + \
+		XCHAL_DTLB_SET(XCHAL_DTLB_WAY3_SET, ENTRIES) + \
+		XCHAL_DTLB_SET(XCHAL_DTLB_WAY4_SET, ENTRIES) + \
+		XCHAL_DTLB_SET(XCHAL_DTLB_WAY5_SET, ENTRIES) + \
+		XCHAL_DTLB_SET(XCHAL_DTLB_WAY6_SET, ENTRIES) + \
+		XCHAL_DTLB_SET(XCHAL_DTLB_WAY7_SET, ENTRIES) + \
+		XCHAL_DTLB_SET(XCHAL_DTLB_WAY8_SET, ENTRIES) + \
+		XCHAL_DTLB_SET(XCHAL_DTLB_WAY9_SET, ENTRIES)
+
+
+typedef struct {
+    unsigned		va;
+    unsigned		pa;
+    unsigned char	asid;
+    unsigned char	ca;
+    unsigned char	way;
+    unsigned char	index;
+    unsigned char	pgsz_log2;	/* 0 .. 32 */
+    unsigned char	type;		/* 0=ITLB 1=DTLB */
+} tlb_dump_entry_t;
+
+/*  Return -1 if a precedes b, +1 if a follows b, 0 if same:  */
+int cmp_tlb_dump_info( tlb_dump_entry_t *a, tlb_dump_entry_t *b )
+{
+    if (a->asid < b->asid) return -1;
+    if (a->asid > b->asid) return  1;
+    if (a->va < b->va) return -1;
+    if (a->va > b->va) return  1;
+    if (a->pa < b->pa) return -1;
+    if (a->pa > b->pa) return  1;
+    if (a->ca < b->ca) return -1;
+    if (a->ca > b->ca) return  1;
+    if (a->way < b->way) return -1;
+    if (a->way > b->way) return  1;
+    if (a->index < b->index) return -1;
+    if (a->index > b->index) return  1;
+    return 0;
+}
+
+void sort_tlb_dump_info( tlb_dump_entry_t *t, int n )
+{
+    int i, j;
+    /*  Simple O(n*n) sort:  */
+    for (i = 0; i < n-1; i++)
+	for (j = i+1; j < n; j++)
+	    if (cmp_tlb_dump_info(t+i, t+j) > 0) {
+		tlb_dump_entry_t tmp = t[i];
+		t[i] = t[j];
+		t[j] = tmp;
+	    }
+}
+
+
+static tlb_dump_entry_t itlb_dump_info[ITLB_TOTAL_ENTRIES];
+static tlb_dump_entry_t dtlb_dump_info[DTLB_TOTAL_ENTRIES];
+
+
+static inline char *way_type (int type)
+{
+	return type ? "autorefill" : "non-autorefill";
+}
+
+void print_entry (struct way_config_t *way_info,
+		  unsigned int way,
+		  unsigned int index,
+		  unsigned int virtual,
+		  unsigned int translation)
+{
+	char valid_chr;
+	unsigned int va, pa, asid, ca;
+
+	va = virtual &
+	  	~((1 << (way_info->pgsz_log2 + way_info->indicies_log2)) - 1);
+	asid = virtual & ((1 << XCHAL_MMU_ASID_BITS) - 1);
+	pa = translation & ~((1 << way_info->pgsz_log2) - 1);
+	ca = translation & ((1 << XCHAL_MMU_CA_BITS) - 1);
+	valid_chr = asid ? 'V' : 'I';
+
+	/* Compute and incorporate the effect of the index bits on the
+	 * va.  It's more useful for kernel debugging, since we always
+	 * want to know the effective va anyway. */
+
+	va += index << way_info->pgsz_log2;
+
+	printk ("\t[%d,%d] (%c) vpn 0x%.8x  ppn 0x%.8x  asid 0x%.2x  am 0x%x\n",
+		way, index, valid_chr, va, pa, asid, ca);
+}
+
+void print_itlb_entry (struct way_config_t *way_info, int way, int index)
+{
+	print_entry (way_info, way, index,
+		     read_itlb_virtual (way + (index << way_info->pgsz_log2)),
+		     read_itlb_translation (way + (index << way_info->pgsz_log2)));
+}
+
+void print_dtlb_entry (struct way_config_t *way_info, int way, int index)
+{
+	print_entry (way_info, way, index,
+		     read_dtlb_virtual (way + (index << way_info->pgsz_log2)),
+		     read_dtlb_translation (way + (index << way_info->pgsz_log2)));
+}
+
+void dump_itlb (void)
+{
+	int way, index;
+
+	printk ("\nITLB: ways = %d\n", XCHAL_ITLB_WAYS);
+
+	for (way = 0; way < XCHAL_ITLB_WAYS; way++) {
+		printk ("\nWay: %d, Entries: %d, MinPageSize: %d, Type: %s\n",
+			way, itlb[way].indicies,
+			itlb[way].pgsz_log2, way_type(itlb[way].arf));
+		for (index = 0; index < itlb[way].indicies; index++) {
+			print_itlb_entry(&itlb[way], way, index);
+		}
+	}
+}
+
+void dump_dtlb (void)
+{
+	int way, index;
+
+	printk ("\nDTLB: ways = %d\n", XCHAL_DTLB_WAYS);
+
+	for (way = 0; way < XCHAL_DTLB_WAYS; way++) {
+		printk ("\nWay: %d, Entries: %d, MinPageSize: %d, Type: %s\n",
+			way, dtlb[way].indicies,
+			dtlb[way].pgsz_log2, way_type(dtlb[way].arf));
+		for (index = 0; index < dtlb[way].indicies; index++) {
+			print_dtlb_entry(&dtlb[way], way, index);
+		}
+	}
+}
+
+void dump_tlb (tlb_dump_entry_t *tinfo, struct way_config_t *config,
+		int entries, int ways, int type, int show_invalid)
+{
+    tlb_dump_entry_t *e = tinfo;
+    int way, i;
+
+    /*  Gather all info:  */
+    for (way = 0; way < ways; way++) {
+	struct way_config_t *cfg = config + way;
+	for (i = 0; i < cfg->indicies; i++) {
+	    unsigned wayindex = way + (i << cfg->pgsz_log2);
+	    unsigned vv = (type ? read_dtlb_virtual (wayindex)
+		    		: read_itlb_virtual (wayindex));
+	    unsigned pp = (type ? read_dtlb_translation (wayindex)
+		    		: read_itlb_translation (wayindex));
+
+	    /* Compute and incorporate the effect of the index bits on the
+	     * va.  It's more useful for kernel debugging, since we always
+	     * want to know the effective va anyway. */
+
+	    e->va = (vv & ~((1 << (cfg->pgsz_log2 + cfg->indicies_log2)) - 1));
+	    e->va += (i << cfg->pgsz_log2);
+	    e->pa = (pp & ~((1 << cfg->pgsz_log2) - 1));
+	    e->asid = (vv & ((1 << XCHAL_MMU_ASID_BITS) - 1));
+	    e->ca = (pp & ((1 << XCHAL_MMU_CA_BITS) - 1));
+	    e->way = way;
+	    e->index = i;
+	    e->pgsz_log2 = cfg->pgsz_log2;
+	    e->type = type;
+	    e++;
+	}
+    }
+#if 1
+    /*  Sort by ASID and VADDR:  */
+    sort_tlb_dump_info (tinfo, entries);
+#endif
+
+    /*  Display all sorted info:  */
+    printk ("\n%cTLB dump:\n", (type ? 'D' : 'I'));
+    for (e = tinfo, i = 0; i < entries; i++, e++) {
+#if 0
+	if (e->asid == 0 && !show_invalid)
+	    continue;
+#endif
+	printk ("%c way=%d i=%d  ASID=%02X V=%08X -> P=%08X CA=%X (%d %cB)\n",
+		(e->type ? 'D' : 'I'), e->way, e->index,
+		e->asid, e->va, e->pa, e->ca,
+		(1 << (e->pgsz_log2 % 10)),
+		" kMG"[e->pgsz_log2 / 10]
+		);
+    }
+}
+
+void dump_tlbs2 (int showinv)
+{
+    dump_tlb (itlb_dump_info, itlb, ITLB_TOTAL_ENTRIES, XCHAL_ITLB_WAYS, 0, showinv);
+    dump_tlb (dtlb_dump_info, dtlb, DTLB_TOTAL_ENTRIES, XCHAL_DTLB_WAYS, 1, showinv);
+}
+
+void dump_all_tlbs (void)
+{
+    dump_tlbs2 (1);
+}
+
+void dump_valid_tlbs (void)
+{
+    dump_tlbs2 (0);
+}
+
+
+void dump_tlbs (void)
+{
+	dump_itlb();
+	dump_dtlb();
+}
+
+void dump_cache_tag(int dcache, int idx)
+{
+	int w, i, s, e;
+	unsigned long tag, index;
+	unsigned long num_lines, num_ways, cache_size, line_size;
+
+	num_ways = dcache ? XCHAL_DCACHE_WAYS : XCHAL_ICACHE_WAYS;
+	cache_size = dcache ? XCHAL_DCACHE_SIZE : XCHAL_ICACHE_SIZE;
+	line_size = dcache ? XCHAL_DCACHE_LINESIZE : XCHAL_ICACHE_LINESIZE;
+
+	num_lines = cache_size / num_ways;
+
+	s = 0; e = num_lines;
+
+	if (idx >= 0)
+		e = (s = idx * line_size) + 1;
+
+	for (i = s; i < e; i+= line_size) {
+		printk("\nline %#08x:", i);
+		for (w = 0; w < num_ways; w++) {
+			index = w * num_lines + i;
+			if (dcache)
+				__asm__ __volatile__("ldct %0, %1\n\t"
+						: "=a"(tag) : "a"(index));
+			else
+				__asm__ __volatile__("lict %0, %1\n\t"
+						: "=a"(tag) : "a"(index));
+
+			printk(" %#010lx", tag);
+		}
+	}
+	printk ("\n");
+}
+
+void dump_icache(int index)
+{
+	unsigned long data, addr;
+	int w, i;
+
+	const unsigned long num_ways = XCHAL_ICACHE_WAYS;
+	const unsigned long cache_size = XCHAL_ICACHE_SIZE;
+	const unsigned long line_size = XCHAL_ICACHE_LINESIZE;
+	const unsigned long num_lines = cache_size / num_ways / line_size;
+
+	for (w = 0; w < num_ways; w++) {
+		printk ("\nWay %d", w);
+
+		for (i = 0; i < line_size; i+= 4) {
+			addr = w * num_lines + index * line_size + i;
+			__asm__ __volatile__("licw %0, %1\n\t"
+					: "=a"(data) : "a"(addr));
+			printk(" %#010lx", data);
+		}
+	}
+	printk ("\n");
+}
+
+void dump_cache_tags(void)
+{
+	printk("Instruction cache\n");
+	dump_cache_tag(0, -1);
+	printk("Data cache\n");
+	dump_cache_tag(1, -1);
+}
+
+#endif
-- 
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 'arch')

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 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 'arch')

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 'arch')

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 79042f087b5ac7bba819de03dc3e7462bab8aad9 Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Fri, 24 Jun 2005 21:27:39 +0100
Subject: [PATCH] ARM: 2698/1: Enable kernel r/w access to user pages on ARMv6

Patch from Catalin Marinas

cpu_v6_set_pte() sets the kernel access rights to r/o for user
pages (L_PTE_USER) when neither L_PTE_WRITE nor L_PTE_DIRTY are
set. This causes a kernel data abort when writing the TLS value
in the 0xffff0000 page. This patch enables the kernel r/w access.

Signed-off-by: Catalin Marinas
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/proc-v6.S | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/mm/proc-v6.S b/arch/arm/mm/proc-v6.S
index 0aa73d41478..e3d8510f434 100644
--- a/arch/arm/mm/proc-v6.S
+++ b/arch/arm/mm/proc-v6.S
@@ -132,8 +132,8 @@ ENTRY(cpu_v6_switch_mm)
  *	  100x   1   0   1	r/o	no acc
  *	  10x0   1   0   1	r/o	no acc
  *	  1011   0   0   1	r/w	no acc
- *	  110x   1   1   0	r/o	r/o
- *	  11x0   1   1   0	r/o	r/o
+ *	  110x   0   1   0	r/w	r/o
+ *	  11x0   0   1   0	r/w	r/o
  *	  1111   0   1   1	r/w	r/w
  */
 ENTRY(cpu_v6_set_pte)
@@ -150,7 +150,7 @@ ENTRY(cpu_v6_set_pte)
 	tst	r1, #L_PTE_USER
 	orrne	r2, r2, #AP1 | nG
 	tstne	r2, #APX
-	eorne	r2, r2, #AP0
+	bicne	r2, r2, #APX | AP0
 
 	tst	r1, #L_PTE_YOUNG
 	biceq	r2, r2, #APX | AP1 | AP0
-- 
cgit v1.2.3-70-g09d2


From 456b6b863a6a1a80e369e42568817747a427b072 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Sat, 25 Jun 2005 16:58:21 +0100
Subject: [PATCH] ARM: 2749/1: update ixp2000 defconfigs to 2.6.12-git6

Patch from Lennert Buytenhek

Update the defconfigs for the ixp2000 platforms to 2.6.12-git6.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/configs/enp2611_defconfig  | 58 +++++++++++++++++++++++++++++++------
 arch/arm/configs/ixdp2400_defconfig | 58 +++++++++++++++++++++++++++++++------
 arch/arm/configs/ixdp2401_defconfig | 58 +++++++++++++++++++++++++++++++------
 arch/arm/configs/ixdp2800_defconfig | 58 +++++++++++++++++++++++++++++++------
 arch/arm/configs/ixdp2801_defconfig | 58 +++++++++++++++++++++++++++++++------
 5 files changed, 245 insertions(+), 45 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/configs/enp2611_defconfig b/arch/arm/configs/enp2611_defconfig
index 06fae4b6277..af215342456 100644
--- a/arch/arm/configs/enp2611_defconfig
+++ b/arch/arm/configs/enp2611_defconfig
@@ -1,14 +1,13 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.12-rc1-bk2
-# Sun Mar 27 22:08:24 2005
+# Linux kernel version: 2.6.12-git6
+# Sat Jun 25 00:57:29 2005
 #
 CONFIG_ARM=y
 CONFIG_MMU=y
 CONFIG_UID16=y
 CONFIG_RWSEM_GENERIC_SPINLOCK=y
 CONFIG_GENERIC_CALIBRATE_DELAY=y
-CONFIG_GENERIC_IOMAP=y
 
 #
 # Code maturity level options
@@ -16,6 +15,7 @@ CONFIG_GENERIC_IOMAP=y
 CONFIG_EXPERIMENTAL=y
 CONFIG_CLEAN_COMPILE=y
 CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -35,6 +35,8 @@ CONFIG_EMBEDDED=y
 CONFIG_KALLSYMS=y
 # CONFIG_KALLSYMS_ALL is not set
 # CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
 CONFIG_BASE_FULL=y
 CONFIG_FUTEX=y
 CONFIG_EPOLL=y
@@ -82,6 +84,7 @@ CONFIG_ARCH_IXP2000=y
 # CONFIG_ARCH_VERSATILE is not set
 # CONFIG_ARCH_IMX is not set
 # CONFIG_ARCH_H720X is not set
+# CONFIG_ARCH_AAEC2000 is not set
 CONFIG_ARCH_SUPPORTS_BIG_ENDIAN=y
 
 #
@@ -106,7 +109,6 @@ CONFIG_CPU_32v5=y
 CONFIG_CPU_ABRT_EV5T=y
 CONFIG_CPU_CACHE_VIVT=y
 CONFIG_CPU_TLB_V4WBI=y
-CONFIG_CPU_MINICACHE=y
 
 #
 # Processor Features
@@ -118,9 +120,11 @@ CONFIG_XSCALE_PMU=y
 #
 # Bus support
 #
+CONFIG_ISA_DMA_API=y
 CONFIG_PCI=y
 CONFIG_PCI_LEGACY_PROC=y
 CONFIG_PCI_NAMES=y
+# CONFIG_PCI_DEBUG is not set
 
 #
 # PCCARD (PCMCIA/CardBus) support
@@ -130,7 +134,15 @@ CONFIG_PCI_NAMES=y
 #
 # Kernel Features
 #
+# CONFIG_SMP is not set
 # CONFIG_PREEMPT is not set
+# CONFIG_ARCH_DISCONTIGMEM_ENABLE is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
 CONFIG_ALIGNMENT_TRAP=y
 
 #
@@ -269,7 +281,6 @@ CONFIG_MTD_IXP2000=y
 #
 # Block devices
 #
-# CONFIG_BLK_DEV_FD is not set
 # CONFIG_BLK_CPQ_DA is not set
 # CONFIG_BLK_CPQ_CISS_DA is not set
 # CONFIG_BLK_DEV_DAC960 is not set
@@ -308,6 +319,7 @@ CONFIG_IOSCHED_CFQ=y
 #
 # Fusion MPT device support
 #
+# CONFIG_FUSION is not set
 
 #
 # IEEE 1394 (FireWire) support
@@ -329,10 +341,11 @@ CONFIG_NET=y
 #
 CONFIG_PACKET=y
 CONFIG_PACKET_MMAP=y
-# CONFIG_NETLINK_DEV is not set
 CONFIG_UNIX=y
 # CONFIG_NET_KEY is not set
 CONFIG_INET=y
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_FIB_TRIE is not set
 # CONFIG_IP_MULTICAST is not set
 # CONFIG_IP_ADVANCED_ROUTER is not set
 CONFIG_IP_PNP=y
@@ -349,6 +362,17 @@ CONFIG_SYN_COOKIES=y
 # CONFIG_INET_TUNNEL is not set
 # CONFIG_IP_TCPDIAG is not set
 # CONFIG_IP_TCPDIAG_IPV6 is not set
+
+#
+# TCP congestion control
+#
+CONFIG_TCP_CONG_BIC=y
+CONFIG_TCP_CONG_WESTWOOD=m
+CONFIG_TCP_CONG_HTCP=m
+# CONFIG_TCP_CONG_HSTCP is not set
+# CONFIG_TCP_CONG_HYBLA is not set
+# CONFIG_TCP_CONG_VEGAS is not set
+# CONFIG_TCP_CONG_SCALABLE is not set
 # CONFIG_IPV6 is not set
 # CONFIG_NETFILTER is not set
 
@@ -404,6 +428,7 @@ CONFIG_MII=y
 # CONFIG_SUNGEM is not set
 # CONFIG_NET_VENDOR_3COM is not set
 # CONFIG_SMC91X is not set
+# CONFIG_DM9000 is not set
 
 #
 # Tulip family network device support
@@ -440,9 +465,11 @@ CONFIG_EEPRO100=y
 # CONFIG_HAMACHI is not set
 # CONFIG_YELLOWFIN is not set
 # CONFIG_R8169 is not set
+# CONFIG_SKGE is not set
 # CONFIG_SK98LIN is not set
 # CONFIG_VIA_VELOCITY is not set
 # CONFIG_TIGON3 is not set
+# CONFIG_BNX2 is not set
 
 #
 # Ethernet (10000 Mbit)
@@ -464,6 +491,7 @@ CONFIG_EEPRO100=y
 # Wan interfaces
 #
 CONFIG_WAN=y
+# CONFIG_DSCC4 is not set
 # CONFIG_LANMEDIA is not set
 # CONFIG_SYNCLINK_SYNCPPP is not set
 CONFIG_HDLC=y
@@ -526,7 +554,6 @@ CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
 #
 # CONFIG_SERIO is not set
 # CONFIG_GAMEPORT is not set
-CONFIG_SOUND_GAMEPORT=y
 
 #
 # Character devices
@@ -547,6 +574,7 @@ CONFIG_SERIAL_8250_NR_UARTS=2
 #
 CONFIG_SERIAL_CORE=y
 CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
 CONFIG_UNIX98_PTYS=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
@@ -613,17 +641,18 @@ CONFIG_I2C_ALGOBIT=y
 # CONFIG_I2C_AMD8111 is not set
 # CONFIG_I2C_I801 is not set
 # CONFIG_I2C_I810 is not set
+# CONFIG_I2C_PIIX4 is not set
 # CONFIG_I2C_ISA is not set
 # CONFIG_I2C_IXP2000 is not set
 # CONFIG_I2C_NFORCE2 is not set
 # CONFIG_I2C_PARPORT_LIGHT is not set
-# CONFIG_I2C_PIIX4 is not set
 # CONFIG_I2C_PROSAVAGE is not set
 # CONFIG_I2C_SAVAGE4 is not set
 # CONFIG_SCx200_ACB is not set
 # CONFIG_I2C_SIS5595 is not set
 # CONFIG_I2C_SIS630 is not set
 # CONFIG_I2C_SIS96X is not set
+# CONFIG_I2C_STUB is not set
 # CONFIG_I2C_VIA is not set
 # CONFIG_I2C_VIAPRO is not set
 # CONFIG_I2C_VOODOO3 is not set
@@ -637,7 +666,9 @@ CONFIG_I2C_SENSOR=y
 # CONFIG_SENSORS_ADM1025 is not set
 # CONFIG_SENSORS_ADM1026 is not set
 # CONFIG_SENSORS_ADM1031 is not set
+# CONFIG_SENSORS_ADM9240 is not set
 # CONFIG_SENSORS_ASB100 is not set
+# CONFIG_SENSORS_ATXP1 is not set
 # CONFIG_SENSORS_DS1621 is not set
 # CONFIG_SENSORS_FSCHER is not set
 # CONFIG_SENSORS_FSCPOS is not set
@@ -653,6 +684,7 @@ CONFIG_I2C_SENSOR=y
 # 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
@@ -662,14 +694,19 @@ CONFIG_I2C_SENSOR=y
 # CONFIG_SENSORS_W83781D is not set
 # CONFIG_SENSORS_W83L785TS is not set
 # CONFIG_SENSORS_W83627HF is not set
+# CONFIG_SENSORS_W83627EHF is not set
 
 #
 # Other I2C Chip support
 #
+# CONFIG_SENSORS_DS1337 is not set
+# CONFIG_SENSORS_DS1374 is not set
 CONFIG_SENSORS_EEPROM=y
 # CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_SENSORS_PCA9539 is not set
 # CONFIG_SENSORS_PCF8591 is not set
 # CONFIG_SENSORS_RTC8564 is not set
+# CONFIG_SENSORS_MAX6875 is not set
 # CONFIG_I2C_DEBUG_CORE is not set
 # CONFIG_I2C_DEBUG_ALGO is not set
 # CONFIG_I2C_DEBUG_BUS is not set
@@ -723,6 +760,7 @@ CONFIG_EXT2_FS=y
 CONFIG_EXT2_FS_XATTR=y
 CONFIG_EXT2_FS_POSIX_ACL=y
 # CONFIG_EXT2_FS_SECURITY is not set
+# CONFIG_EXT2_FS_XIP is not set
 CONFIG_EXT3_FS=y
 CONFIG_EXT3_FS_XATTR=y
 CONFIG_EXT3_FS_POSIX_ACL=y
@@ -763,7 +801,6 @@ CONFIG_DNOTIFY=y
 #
 CONFIG_PROC_FS=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
@@ -801,12 +838,14 @@ CONFIG_JFFS2_RTIME=y
 #
 CONFIG_NFS_FS=y
 CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL 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_LOCKD_V4=y
+CONFIG_NFS_COMMON=y
 CONFIG_SUNRPC=y
 # CONFIG_RPCSEC_GSS_KRB5 is not set
 # CONFIG_RPCSEC_GSS_SPKM3 is not set
@@ -891,3 +930,4 @@ CONFIG_CRC32=y
 # CONFIG_LIBCRC32C is not set
 CONFIG_ZLIB_INFLATE=y
 CONFIG_ZLIB_DEFLATE=y
+# CONFIG_TEXTSEARCH is not set
diff --git a/arch/arm/configs/ixdp2400_defconfig b/arch/arm/configs/ixdp2400_defconfig
index 810a450a55d..a7ee1a442f3 100644
--- a/arch/arm/configs/ixdp2400_defconfig
+++ b/arch/arm/configs/ixdp2400_defconfig
@@ -1,14 +1,13 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.12-rc1-bk2
-# Sun Mar 27 21:13:38 2005
+# Linux kernel version: 2.6.12-git6
+# Sat Jun 25 00:58:38 2005
 #
 CONFIG_ARM=y
 CONFIG_MMU=y
 CONFIG_UID16=y
 CONFIG_RWSEM_GENERIC_SPINLOCK=y
 CONFIG_GENERIC_CALIBRATE_DELAY=y
-CONFIG_GENERIC_IOMAP=y
 
 #
 # Code maturity level options
@@ -16,6 +15,7 @@ CONFIG_GENERIC_IOMAP=y
 CONFIG_EXPERIMENTAL=y
 CONFIG_CLEAN_COMPILE=y
 CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -35,6 +35,8 @@ CONFIG_EMBEDDED=y
 CONFIG_KALLSYMS=y
 # CONFIG_KALLSYMS_ALL is not set
 # CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
 CONFIG_BASE_FULL=y
 CONFIG_FUTEX=y
 CONFIG_EPOLL=y
@@ -82,6 +84,7 @@ CONFIG_ARCH_IXP2000=y
 # CONFIG_ARCH_VERSATILE is not set
 # CONFIG_ARCH_IMX is not set
 # CONFIG_ARCH_H720X is not set
+# CONFIG_ARCH_AAEC2000 is not set
 CONFIG_ARCH_SUPPORTS_BIG_ENDIAN=y
 
 #
@@ -107,7 +110,6 @@ CONFIG_CPU_32v5=y
 CONFIG_CPU_ABRT_EV5T=y
 CONFIG_CPU_CACHE_VIVT=y
 CONFIG_CPU_TLB_V4WBI=y
-CONFIG_CPU_MINICACHE=y
 
 #
 # Processor Features
@@ -119,9 +121,11 @@ CONFIG_XSCALE_PMU=y
 #
 # Bus support
 #
+CONFIG_ISA_DMA_API=y
 CONFIG_PCI=y
 CONFIG_PCI_LEGACY_PROC=y
 CONFIG_PCI_NAMES=y
+# CONFIG_PCI_DEBUG is not set
 
 #
 # PCCARD (PCMCIA/CardBus) support
@@ -131,7 +135,15 @@ CONFIG_PCI_NAMES=y
 #
 # Kernel Features
 #
+# CONFIG_SMP is not set
 # CONFIG_PREEMPT is not set
+# CONFIG_ARCH_DISCONTIGMEM_ENABLE is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
 CONFIG_ALIGNMENT_TRAP=y
 
 #
@@ -270,7 +282,6 @@ CONFIG_MTD_IXP2000=y
 #
 # Block devices
 #
-# CONFIG_BLK_DEV_FD is not set
 # CONFIG_BLK_CPQ_DA is not set
 # CONFIG_BLK_CPQ_CISS_DA is not set
 # CONFIG_BLK_DEV_DAC960 is not set
@@ -309,6 +320,7 @@ CONFIG_IOSCHED_CFQ=y
 #
 # Fusion MPT device support
 #
+# CONFIG_FUSION is not set
 
 #
 # IEEE 1394 (FireWire) support
@@ -330,10 +342,11 @@ CONFIG_NET=y
 #
 CONFIG_PACKET=y
 CONFIG_PACKET_MMAP=y
-# CONFIG_NETLINK_DEV is not set
 CONFIG_UNIX=y
 # CONFIG_NET_KEY is not set
 CONFIG_INET=y
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_FIB_TRIE is not set
 # CONFIG_IP_MULTICAST is not set
 # CONFIG_IP_ADVANCED_ROUTER is not set
 CONFIG_IP_PNP=y
@@ -350,6 +363,17 @@ CONFIG_SYN_COOKIES=y
 # CONFIG_INET_TUNNEL is not set
 # CONFIG_IP_TCPDIAG is not set
 # CONFIG_IP_TCPDIAG_IPV6 is not set
+
+#
+# TCP congestion control
+#
+CONFIG_TCP_CONG_BIC=y
+CONFIG_TCP_CONG_WESTWOOD=m
+CONFIG_TCP_CONG_HTCP=m
+# CONFIG_TCP_CONG_HSTCP is not set
+# CONFIG_TCP_CONG_HYBLA is not set
+# CONFIG_TCP_CONG_VEGAS is not set
+# CONFIG_TCP_CONG_SCALABLE is not set
 # CONFIG_IPV6 is not set
 # CONFIG_NETFILTER is not set
 
@@ -405,6 +429,7 @@ CONFIG_MII=y
 # CONFIG_SUNGEM is not set
 # CONFIG_NET_VENDOR_3COM is not set
 # CONFIG_SMC91X is not set
+# CONFIG_DM9000 is not set
 
 #
 # Tulip family network device support
@@ -441,9 +466,11 @@ CONFIG_EEPRO100=y
 # CONFIG_HAMACHI is not set
 # CONFIG_YELLOWFIN is not set
 # CONFIG_R8169 is not set
+# CONFIG_SKGE is not set
 # CONFIG_SK98LIN is not set
 # CONFIG_VIA_VELOCITY is not set
 # CONFIG_TIGON3 is not set
+# CONFIG_BNX2 is not set
 
 #
 # Ethernet (10000 Mbit)
@@ -465,6 +492,7 @@ CONFIG_EEPRO100=y
 # Wan interfaces
 #
 CONFIG_WAN=y
+# CONFIG_DSCC4 is not set
 # CONFIG_LANMEDIA is not set
 # CONFIG_SYNCLINK_SYNCPPP is not set
 CONFIG_HDLC=y
@@ -527,7 +555,6 @@ CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
 #
 # CONFIG_SERIO is not set
 # CONFIG_GAMEPORT is not set
-CONFIG_SOUND_GAMEPORT=y
 
 #
 # Character devices
@@ -548,6 +575,7 @@ CONFIG_SERIAL_8250_NR_UARTS=2
 #
 CONFIG_SERIAL_CORE=y
 CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
 CONFIG_UNIX98_PTYS=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
@@ -614,17 +642,18 @@ CONFIG_I2C_ALGOBIT=y
 # CONFIG_I2C_AMD8111 is not set
 # CONFIG_I2C_I801 is not set
 # CONFIG_I2C_I810 is not set
+# CONFIG_I2C_PIIX4 is not set
 # CONFIG_I2C_ISA is not set
 # CONFIG_I2C_IXP2000 is not set
 # CONFIG_I2C_NFORCE2 is not set
 # CONFIG_I2C_PARPORT_LIGHT is not set
-# CONFIG_I2C_PIIX4 is not set
 # CONFIG_I2C_PROSAVAGE is not set
 # CONFIG_I2C_SAVAGE4 is not set
 # CONFIG_SCx200_ACB is not set
 # CONFIG_I2C_SIS5595 is not set
 # CONFIG_I2C_SIS630 is not set
 # CONFIG_I2C_SIS96X is not set
+# CONFIG_I2C_STUB is not set
 # CONFIG_I2C_VIA is not set
 # CONFIG_I2C_VIAPRO is not set
 # CONFIG_I2C_VOODOO3 is not set
@@ -638,7 +667,9 @@ CONFIG_I2C_SENSOR=y
 # CONFIG_SENSORS_ADM1025 is not set
 # CONFIG_SENSORS_ADM1026 is not set
 # CONFIG_SENSORS_ADM1031 is not set
+# CONFIG_SENSORS_ADM9240 is not set
 # CONFIG_SENSORS_ASB100 is not set
+# CONFIG_SENSORS_ATXP1 is not set
 # CONFIG_SENSORS_DS1621 is not set
 # CONFIG_SENSORS_FSCHER is not set
 # CONFIG_SENSORS_FSCPOS is not set
@@ -654,6 +685,7 @@ CONFIG_I2C_SENSOR=y
 # 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
@@ -663,14 +695,19 @@ CONFIG_I2C_SENSOR=y
 # CONFIG_SENSORS_W83781D is not set
 # CONFIG_SENSORS_W83L785TS is not set
 # CONFIG_SENSORS_W83627HF is not set
+# CONFIG_SENSORS_W83627EHF is not set
 
 #
 # Other I2C Chip support
 #
+# CONFIG_SENSORS_DS1337 is not set
+# CONFIG_SENSORS_DS1374 is not set
 CONFIG_SENSORS_EEPROM=y
 # CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_SENSORS_PCA9539 is not set
 # CONFIG_SENSORS_PCF8591 is not set
 # CONFIG_SENSORS_RTC8564 is not set
+# CONFIG_SENSORS_MAX6875 is not set
 # CONFIG_I2C_DEBUG_CORE is not set
 # CONFIG_I2C_DEBUG_ALGO is not set
 # CONFIG_I2C_DEBUG_BUS is not set
@@ -724,6 +761,7 @@ CONFIG_EXT2_FS=y
 CONFIG_EXT2_FS_XATTR=y
 CONFIG_EXT2_FS_POSIX_ACL=y
 # CONFIG_EXT2_FS_SECURITY is not set
+# CONFIG_EXT2_FS_XIP is not set
 CONFIG_EXT3_FS=y
 CONFIG_EXT3_FS_XATTR=y
 CONFIG_EXT3_FS_POSIX_ACL=y
@@ -764,7 +802,6 @@ CONFIG_DNOTIFY=y
 #
 CONFIG_PROC_FS=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
@@ -802,12 +839,14 @@ CONFIG_JFFS2_RTIME=y
 #
 CONFIG_NFS_FS=y
 CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL 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_LOCKD_V4=y
+CONFIG_NFS_COMMON=y
 CONFIG_SUNRPC=y
 # CONFIG_RPCSEC_GSS_KRB5 is not set
 # CONFIG_RPCSEC_GSS_SPKM3 is not set
@@ -892,3 +931,4 @@ CONFIG_CRC32=y
 # CONFIG_LIBCRC32C is not set
 CONFIG_ZLIB_INFLATE=y
 CONFIG_ZLIB_DEFLATE=y
+# CONFIG_TEXTSEARCH is not set
diff --git a/arch/arm/configs/ixdp2401_defconfig b/arch/arm/configs/ixdp2401_defconfig
index 72e1b940e97..2da48fca5c1 100644
--- a/arch/arm/configs/ixdp2401_defconfig
+++ b/arch/arm/configs/ixdp2401_defconfig
@@ -1,14 +1,13 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.12-rc1-bk2
-# Sun Mar 27 21:53:55 2005
+# Linux kernel version: 2.6.12-git6
+# Sat Jun 25 00:59:35 2005
 #
 CONFIG_ARM=y
 CONFIG_MMU=y
 CONFIG_UID16=y
 CONFIG_RWSEM_GENERIC_SPINLOCK=y
 CONFIG_GENERIC_CALIBRATE_DELAY=y
-CONFIG_GENERIC_IOMAP=y
 
 #
 # Code maturity level options
@@ -16,6 +15,7 @@ CONFIG_GENERIC_IOMAP=y
 CONFIG_EXPERIMENTAL=y
 CONFIG_CLEAN_COMPILE=y
 CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -35,6 +35,8 @@ CONFIG_EMBEDDED=y
 CONFIG_KALLSYMS=y
 # CONFIG_KALLSYMS_ALL is not set
 # CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
 CONFIG_BASE_FULL=y
 CONFIG_FUTEX=y
 CONFIG_EPOLL=y
@@ -82,6 +84,7 @@ CONFIG_ARCH_IXP2000=y
 # CONFIG_ARCH_VERSATILE is not set
 # CONFIG_ARCH_IMX is not set
 # CONFIG_ARCH_H720X is not set
+# CONFIG_ARCH_AAEC2000 is not set
 CONFIG_ARCH_SUPPORTS_BIG_ENDIAN=y
 
 #
@@ -107,7 +110,6 @@ CONFIG_CPU_32v5=y
 CONFIG_CPU_ABRT_EV5T=y
 CONFIG_CPU_CACHE_VIVT=y
 CONFIG_CPU_TLB_V4WBI=y
-CONFIG_CPU_MINICACHE=y
 
 #
 # Processor Features
@@ -119,9 +121,11 @@ CONFIG_XSCALE_PMU=y
 #
 # Bus support
 #
+CONFIG_ISA_DMA_API=y
 CONFIG_PCI=y
 CONFIG_PCI_LEGACY_PROC=y
 CONFIG_PCI_NAMES=y
+# CONFIG_PCI_DEBUG is not set
 
 #
 # PCCARD (PCMCIA/CardBus) support
@@ -131,7 +135,15 @@ CONFIG_PCI_NAMES=y
 #
 # Kernel Features
 #
+# CONFIG_SMP is not set
 # CONFIG_PREEMPT is not set
+# CONFIG_ARCH_DISCONTIGMEM_ENABLE is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
 CONFIG_ALIGNMENT_TRAP=y
 
 #
@@ -270,7 +282,6 @@ CONFIG_MTD_IXP2000=y
 #
 # Block devices
 #
-# CONFIG_BLK_DEV_FD is not set
 # CONFIG_BLK_CPQ_DA is not set
 # CONFIG_BLK_CPQ_CISS_DA is not set
 # CONFIG_BLK_DEV_DAC960 is not set
@@ -309,6 +320,7 @@ CONFIG_IOSCHED_CFQ=y
 #
 # Fusion MPT device support
 #
+# CONFIG_FUSION is not set
 
 #
 # IEEE 1394 (FireWire) support
@@ -330,10 +342,11 @@ CONFIG_NET=y
 #
 CONFIG_PACKET=y
 CONFIG_PACKET_MMAP=y
-# CONFIG_NETLINK_DEV is not set
 CONFIG_UNIX=y
 # CONFIG_NET_KEY is not set
 CONFIG_INET=y
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_FIB_TRIE is not set
 # CONFIG_IP_MULTICAST is not set
 # CONFIG_IP_ADVANCED_ROUTER is not set
 CONFIG_IP_PNP=y
@@ -350,6 +363,17 @@ CONFIG_SYN_COOKIES=y
 # CONFIG_INET_TUNNEL is not set
 CONFIG_IP_TCPDIAG=y
 # CONFIG_IP_TCPDIAG_IPV6 is not set
+
+#
+# TCP congestion control
+#
+CONFIG_TCP_CONG_BIC=y
+CONFIG_TCP_CONG_WESTWOOD=m
+CONFIG_TCP_CONG_HTCP=m
+# CONFIG_TCP_CONG_HSTCP is not set
+# CONFIG_TCP_CONG_HYBLA is not set
+# CONFIG_TCP_CONG_VEGAS is not set
+# CONFIG_TCP_CONG_SCALABLE is not set
 # CONFIG_IPV6 is not set
 # CONFIG_NETFILTER is not set
 
@@ -405,6 +429,7 @@ CONFIG_MII=y
 # CONFIG_SUNGEM is not set
 # CONFIG_NET_VENDOR_3COM is not set
 # CONFIG_SMC91X is not set
+# CONFIG_DM9000 is not set
 
 #
 # Tulip family network device support
@@ -442,9 +467,11 @@ CONFIG_EEPRO100=y
 # CONFIG_HAMACHI is not set
 # CONFIG_YELLOWFIN is not set
 # CONFIG_R8169 is not set
+# CONFIG_SKGE is not set
 # CONFIG_SK98LIN is not set
 # CONFIG_VIA_VELOCITY is not set
 # CONFIG_TIGON3 is not set
+# CONFIG_BNX2 is not set
 
 #
 # Ethernet (10000 Mbit)
@@ -466,6 +493,7 @@ CONFIG_EEPRO100=y
 # Wan interfaces
 #
 CONFIG_WAN=y
+# CONFIG_DSCC4 is not set
 # CONFIG_LANMEDIA is not set
 # CONFIG_SYNCLINK_SYNCPPP is not set
 CONFIG_HDLC=y
@@ -528,7 +556,6 @@ CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
 #
 # CONFIG_SERIO is not set
 # CONFIG_GAMEPORT is not set
-CONFIG_SOUND_GAMEPORT=y
 
 #
 # Character devices
@@ -549,6 +576,7 @@ CONFIG_SERIAL_8250_NR_UARTS=2
 #
 CONFIG_SERIAL_CORE=y
 CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
 CONFIG_UNIX98_PTYS=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
@@ -615,17 +643,18 @@ CONFIG_I2C_ALGOBIT=y
 # CONFIG_I2C_AMD8111 is not set
 # CONFIG_I2C_I801 is not set
 # CONFIG_I2C_I810 is not set
+# CONFIG_I2C_PIIX4 is not set
 # CONFIG_I2C_ISA is not set
 # CONFIG_I2C_IXP2000 is not set
 # CONFIG_I2C_NFORCE2 is not set
 # CONFIG_I2C_PARPORT_LIGHT is not set
-# CONFIG_I2C_PIIX4 is not set
 # CONFIG_I2C_PROSAVAGE is not set
 # CONFIG_I2C_SAVAGE4 is not set
 # CONFIG_SCx200_ACB is not set
 # CONFIG_I2C_SIS5595 is not set
 # CONFIG_I2C_SIS630 is not set
 # CONFIG_I2C_SIS96X is not set
+# CONFIG_I2C_STUB is not set
 # CONFIG_I2C_VIA is not set
 # CONFIG_I2C_VIAPRO is not set
 # CONFIG_I2C_VOODOO3 is not set
@@ -639,7 +668,9 @@ CONFIG_I2C_SENSOR=y
 # CONFIG_SENSORS_ADM1025 is not set
 # CONFIG_SENSORS_ADM1026 is not set
 # CONFIG_SENSORS_ADM1031 is not set
+# CONFIG_SENSORS_ADM9240 is not set
 # CONFIG_SENSORS_ASB100 is not set
+# CONFIG_SENSORS_ATXP1 is not set
 # CONFIG_SENSORS_DS1621 is not set
 # CONFIG_SENSORS_FSCHER is not set
 # CONFIG_SENSORS_FSCPOS is not set
@@ -655,6 +686,7 @@ CONFIG_I2C_SENSOR=y
 # 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
@@ -664,14 +696,19 @@ CONFIG_I2C_SENSOR=y
 # CONFIG_SENSORS_W83781D is not set
 # CONFIG_SENSORS_W83L785TS is not set
 # CONFIG_SENSORS_W83627HF is not set
+# CONFIG_SENSORS_W83627EHF is not set
 
 #
 # Other I2C Chip support
 #
+# CONFIG_SENSORS_DS1337 is not set
+# CONFIG_SENSORS_DS1374 is not set
 CONFIG_SENSORS_EEPROM=y
 # CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_SENSORS_PCA9539 is not set
 # CONFIG_SENSORS_PCF8591 is not set
 # CONFIG_SENSORS_RTC8564 is not set
+# CONFIG_SENSORS_MAX6875 is not set
 # CONFIG_I2C_DEBUG_CORE is not set
 # CONFIG_I2C_DEBUG_ALGO is not set
 # CONFIG_I2C_DEBUG_BUS is not set
@@ -725,6 +762,7 @@ CONFIG_EXT2_FS=y
 CONFIG_EXT2_FS_XATTR=y
 CONFIG_EXT2_FS_POSIX_ACL=y
 # CONFIG_EXT2_FS_SECURITY is not set
+# CONFIG_EXT2_FS_XIP is not set
 CONFIG_EXT3_FS=y
 CONFIG_EXT3_FS_XATTR=y
 CONFIG_EXT3_FS_POSIX_ACL=y
@@ -765,7 +803,6 @@ CONFIG_DNOTIFY=y
 #
 CONFIG_PROC_FS=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
@@ -803,12 +840,14 @@ CONFIG_JFFS2_RTIME=y
 #
 CONFIG_NFS_FS=y
 CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL 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_LOCKD_V4=y
+CONFIG_NFS_COMMON=y
 CONFIG_SUNRPC=y
 # CONFIG_RPCSEC_GSS_KRB5 is not set
 # CONFIG_RPCSEC_GSS_SPKM3 is not set
@@ -893,3 +932,4 @@ CONFIG_CRC32=y
 # CONFIG_LIBCRC32C is not set
 CONFIG_ZLIB_INFLATE=y
 CONFIG_ZLIB_DEFLATE=y
+# CONFIG_TEXTSEARCH is not set
diff --git a/arch/arm/configs/ixdp2800_defconfig b/arch/arm/configs/ixdp2800_defconfig
index 1592e45f027..7ba867f751f 100644
--- a/arch/arm/configs/ixdp2800_defconfig
+++ b/arch/arm/configs/ixdp2800_defconfig
@@ -1,14 +1,13 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.12-rc1-bk2
-# Sun Mar 27 22:15:23 2005
+# Linux kernel version: 2.6.12-git6
+# Sat Jun 25 01:00:27 2005
 #
 CONFIG_ARM=y
 CONFIG_MMU=y
 CONFIG_UID16=y
 CONFIG_RWSEM_GENERIC_SPINLOCK=y
 CONFIG_GENERIC_CALIBRATE_DELAY=y
-CONFIG_GENERIC_IOMAP=y
 
 #
 # Code maturity level options
@@ -16,6 +15,7 @@ CONFIG_GENERIC_IOMAP=y
 CONFIG_EXPERIMENTAL=y
 CONFIG_CLEAN_COMPILE=y
 CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -35,6 +35,8 @@ CONFIG_EMBEDDED=y
 CONFIG_KALLSYMS=y
 # CONFIG_KALLSYMS_ALL is not set
 # CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
 CONFIG_BASE_FULL=y
 CONFIG_FUTEX=y
 CONFIG_EPOLL=y
@@ -82,6 +84,7 @@ CONFIG_ARCH_IXP2000=y
 # CONFIG_ARCH_VERSATILE is not set
 # CONFIG_ARCH_IMX is not set
 # CONFIG_ARCH_H720X is not set
+# CONFIG_ARCH_AAEC2000 is not set
 CONFIG_ARCH_SUPPORTS_BIG_ENDIAN=y
 
 #
@@ -107,7 +110,6 @@ CONFIG_CPU_32v5=y
 CONFIG_CPU_ABRT_EV5T=y
 CONFIG_CPU_CACHE_VIVT=y
 CONFIG_CPU_TLB_V4WBI=y
-CONFIG_CPU_MINICACHE=y
 
 #
 # Processor Features
@@ -119,9 +121,11 @@ CONFIG_XSCALE_PMU=y
 #
 # Bus support
 #
+CONFIG_ISA_DMA_API=y
 CONFIG_PCI=y
 CONFIG_PCI_LEGACY_PROC=y
 CONFIG_PCI_NAMES=y
+# CONFIG_PCI_DEBUG is not set
 
 #
 # PCCARD (PCMCIA/CardBus) support
@@ -131,7 +135,15 @@ CONFIG_PCI_NAMES=y
 #
 # Kernel Features
 #
+# CONFIG_SMP is not set
 # CONFIG_PREEMPT is not set
+# CONFIG_ARCH_DISCONTIGMEM_ENABLE is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
 CONFIG_ALIGNMENT_TRAP=y
 
 #
@@ -270,7 +282,6 @@ CONFIG_MTD_IXP2000=y
 #
 # Block devices
 #
-# CONFIG_BLK_DEV_FD is not set
 # CONFIG_BLK_CPQ_DA is not set
 # CONFIG_BLK_CPQ_CISS_DA is not set
 # CONFIG_BLK_DEV_DAC960 is not set
@@ -309,6 +320,7 @@ CONFIG_IOSCHED_CFQ=y
 #
 # Fusion MPT device support
 #
+# CONFIG_FUSION is not set
 
 #
 # IEEE 1394 (FireWire) support
@@ -330,10 +342,11 @@ CONFIG_NET=y
 #
 CONFIG_PACKET=y
 CONFIG_PACKET_MMAP=y
-# CONFIG_NETLINK_DEV is not set
 CONFIG_UNIX=y
 # CONFIG_NET_KEY is not set
 CONFIG_INET=y
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_FIB_TRIE is not set
 # CONFIG_IP_MULTICAST is not set
 # CONFIG_IP_ADVANCED_ROUTER is not set
 CONFIG_IP_PNP=y
@@ -350,6 +363,17 @@ CONFIG_SYN_COOKIES=y
 # CONFIG_INET_TUNNEL is not set
 # CONFIG_IP_TCPDIAG is not set
 # CONFIG_IP_TCPDIAG_IPV6 is not set
+
+#
+# TCP congestion control
+#
+CONFIG_TCP_CONG_BIC=y
+CONFIG_TCP_CONG_WESTWOOD=m
+CONFIG_TCP_CONG_HTCP=m
+# CONFIG_TCP_CONG_HSTCP is not set
+# CONFIG_TCP_CONG_HYBLA is not set
+# CONFIG_TCP_CONG_VEGAS is not set
+# CONFIG_TCP_CONG_SCALABLE is not set
 # CONFIG_IPV6 is not set
 # CONFIG_NETFILTER is not set
 
@@ -405,6 +429,7 @@ CONFIG_MII=y
 # CONFIG_SUNGEM is not set
 # CONFIG_NET_VENDOR_3COM is not set
 # CONFIG_SMC91X is not set
+# CONFIG_DM9000 is not set
 
 #
 # Tulip family network device support
@@ -441,9 +466,11 @@ CONFIG_EEPRO100=y
 # CONFIG_HAMACHI is not set
 # CONFIG_YELLOWFIN is not set
 # CONFIG_R8169 is not set
+# CONFIG_SKGE is not set
 # CONFIG_SK98LIN is not set
 # CONFIG_VIA_VELOCITY is not set
 # CONFIG_TIGON3 is not set
+# CONFIG_BNX2 is not set
 
 #
 # Ethernet (10000 Mbit)
@@ -465,6 +492,7 @@ CONFIG_EEPRO100=y
 # Wan interfaces
 #
 CONFIG_WAN=y
+# CONFIG_DSCC4 is not set
 # CONFIG_LANMEDIA is not set
 # CONFIG_SYNCLINK_SYNCPPP is not set
 CONFIG_HDLC=y
@@ -527,7 +555,6 @@ CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
 #
 # CONFIG_SERIO is not set
 # CONFIG_GAMEPORT is not set
-CONFIG_SOUND_GAMEPORT=y
 
 #
 # Character devices
@@ -548,6 +575,7 @@ CONFIG_SERIAL_8250_NR_UARTS=2
 #
 CONFIG_SERIAL_CORE=y
 CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
 CONFIG_UNIX98_PTYS=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
@@ -614,17 +642,18 @@ CONFIG_I2C_ALGOBIT=y
 # CONFIG_I2C_AMD8111 is not set
 # CONFIG_I2C_I801 is not set
 # CONFIG_I2C_I810 is not set
+# CONFIG_I2C_PIIX4 is not set
 # CONFIG_I2C_ISA is not set
 # CONFIG_I2C_IXP2000 is not set
 # CONFIG_I2C_NFORCE2 is not set
 # CONFIG_I2C_PARPORT_LIGHT is not set
-# CONFIG_I2C_PIIX4 is not set
 # CONFIG_I2C_PROSAVAGE is not set
 # CONFIG_I2C_SAVAGE4 is not set
 # CONFIG_SCx200_ACB is not set
 # CONFIG_I2C_SIS5595 is not set
 # CONFIG_I2C_SIS630 is not set
 # CONFIG_I2C_SIS96X is not set
+# CONFIG_I2C_STUB is not set
 # CONFIG_I2C_VIA is not set
 # CONFIG_I2C_VIAPRO is not set
 # CONFIG_I2C_VOODOO3 is not set
@@ -638,7 +667,9 @@ CONFIG_I2C_SENSOR=y
 # CONFIG_SENSORS_ADM1025 is not set
 # CONFIG_SENSORS_ADM1026 is not set
 # CONFIG_SENSORS_ADM1031 is not set
+# CONFIG_SENSORS_ADM9240 is not set
 # CONFIG_SENSORS_ASB100 is not set
+# CONFIG_SENSORS_ATXP1 is not set
 # CONFIG_SENSORS_DS1621 is not set
 # CONFIG_SENSORS_FSCHER is not set
 # CONFIG_SENSORS_FSCPOS is not set
@@ -654,6 +685,7 @@ CONFIG_I2C_SENSOR=y
 # 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
@@ -663,14 +695,19 @@ CONFIG_I2C_SENSOR=y
 # CONFIG_SENSORS_W83781D is not set
 # CONFIG_SENSORS_W83L785TS is not set
 # CONFIG_SENSORS_W83627HF is not set
+# CONFIG_SENSORS_W83627EHF is not set
 
 #
 # Other I2C Chip support
 #
+# CONFIG_SENSORS_DS1337 is not set
+# CONFIG_SENSORS_DS1374 is not set
 CONFIG_SENSORS_EEPROM=y
 # CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_SENSORS_PCA9539 is not set
 # CONFIG_SENSORS_PCF8591 is not set
 # CONFIG_SENSORS_RTC8564 is not set
+# CONFIG_SENSORS_MAX6875 is not set
 # CONFIG_I2C_DEBUG_CORE is not set
 # CONFIG_I2C_DEBUG_ALGO is not set
 # CONFIG_I2C_DEBUG_BUS is not set
@@ -724,6 +761,7 @@ CONFIG_EXT2_FS=y
 CONFIG_EXT2_FS_XATTR=y
 CONFIG_EXT2_FS_POSIX_ACL=y
 # CONFIG_EXT2_FS_SECURITY is not set
+# CONFIG_EXT2_FS_XIP is not set
 CONFIG_EXT3_FS=y
 CONFIG_EXT3_FS_XATTR=y
 CONFIG_EXT3_FS_POSIX_ACL=y
@@ -764,7 +802,6 @@ CONFIG_DNOTIFY=y
 #
 CONFIG_PROC_FS=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
@@ -802,12 +839,14 @@ CONFIG_JFFS2_RTIME=y
 #
 CONFIG_NFS_FS=y
 CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL 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_LOCKD_V4=y
+CONFIG_NFS_COMMON=y
 CONFIG_SUNRPC=y
 # CONFIG_RPCSEC_GSS_KRB5 is not set
 # CONFIG_RPCSEC_GSS_SPKM3 is not set
@@ -892,3 +931,4 @@ CONFIG_CRC32=y
 # CONFIG_LIBCRC32C is not set
 CONFIG_ZLIB_INFLATE=y
 CONFIG_ZLIB_DEFLATE=y
+# CONFIG_TEXTSEARCH is not set
diff --git a/arch/arm/configs/ixdp2801_defconfig b/arch/arm/configs/ixdp2801_defconfig
index f1afe3d09ec..c4df0ec34b0 100644
--- a/arch/arm/configs/ixdp2801_defconfig
+++ b/arch/arm/configs/ixdp2801_defconfig
@@ -1,14 +1,13 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.12-rc1-bk2
-# Sun Mar 27 22:39:19 2005
+# Linux kernel version: 2.6.12-git6
+# Sat Jun 25 01:01:18 2005
 #
 CONFIG_ARM=y
 CONFIG_MMU=y
 CONFIG_UID16=y
 CONFIG_RWSEM_GENERIC_SPINLOCK=y
 CONFIG_GENERIC_CALIBRATE_DELAY=y
-CONFIG_GENERIC_IOMAP=y
 
 #
 # Code maturity level options
@@ -16,6 +15,7 @@ CONFIG_GENERIC_IOMAP=y
 CONFIG_EXPERIMENTAL=y
 CONFIG_CLEAN_COMPILE=y
 CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -35,6 +35,8 @@ CONFIG_EMBEDDED=y
 CONFIG_KALLSYMS=y
 # CONFIG_KALLSYMS_ALL is not set
 # CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
 CONFIG_BASE_FULL=y
 CONFIG_FUTEX=y
 CONFIG_EPOLL=y
@@ -82,6 +84,7 @@ CONFIG_ARCH_IXP2000=y
 # CONFIG_ARCH_VERSATILE is not set
 # CONFIG_ARCH_IMX is not set
 # CONFIG_ARCH_H720X is not set
+# CONFIG_ARCH_AAEC2000 is not set
 CONFIG_ARCH_SUPPORTS_BIG_ENDIAN=y
 
 #
@@ -107,7 +110,6 @@ CONFIG_CPU_32v5=y
 CONFIG_CPU_ABRT_EV5T=y
 CONFIG_CPU_CACHE_VIVT=y
 CONFIG_CPU_TLB_V4WBI=y
-CONFIG_CPU_MINICACHE=y
 
 #
 # Processor Features
@@ -119,9 +121,11 @@ CONFIG_XSCALE_PMU=y
 #
 # Bus support
 #
+CONFIG_ISA_DMA_API=y
 CONFIG_PCI=y
 CONFIG_PCI_LEGACY_PROC=y
 CONFIG_PCI_NAMES=y
+# CONFIG_PCI_DEBUG is not set
 
 #
 # PCCARD (PCMCIA/CardBus) support
@@ -131,7 +135,15 @@ CONFIG_PCI_NAMES=y
 #
 # Kernel Features
 #
+# CONFIG_SMP is not set
 # CONFIG_PREEMPT is not set
+# CONFIG_ARCH_DISCONTIGMEM_ENABLE is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
 CONFIG_ALIGNMENT_TRAP=y
 
 #
@@ -270,7 +282,6 @@ CONFIG_MTD_IXP2000=y
 #
 # Block devices
 #
-# CONFIG_BLK_DEV_FD is not set
 # CONFIG_BLK_CPQ_DA is not set
 # CONFIG_BLK_CPQ_CISS_DA is not set
 # CONFIG_BLK_DEV_DAC960 is not set
@@ -309,6 +320,7 @@ CONFIG_IOSCHED_CFQ=y
 #
 # Fusion MPT device support
 #
+# CONFIG_FUSION is not set
 
 #
 # IEEE 1394 (FireWire) support
@@ -330,10 +342,11 @@ CONFIG_NET=y
 #
 CONFIG_PACKET=y
 CONFIG_PACKET_MMAP=y
-# CONFIG_NETLINK_DEV is not set
 CONFIG_UNIX=y
 # CONFIG_NET_KEY is not set
 CONFIG_INET=y
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_FIB_TRIE is not set
 # CONFIG_IP_MULTICAST is not set
 # CONFIG_IP_ADVANCED_ROUTER is not set
 CONFIG_IP_PNP=y
@@ -350,6 +363,17 @@ CONFIG_SYN_COOKIES=y
 # CONFIG_INET_TUNNEL is not set
 # CONFIG_IP_TCPDIAG is not set
 # CONFIG_IP_TCPDIAG_IPV6 is not set
+
+#
+# TCP congestion control
+#
+CONFIG_TCP_CONG_BIC=y
+CONFIG_TCP_CONG_WESTWOOD=m
+CONFIG_TCP_CONG_HTCP=m
+# CONFIG_TCP_CONG_HSTCP is not set
+# CONFIG_TCP_CONG_HYBLA is not set
+# CONFIG_TCP_CONG_VEGAS is not set
+# CONFIG_TCP_CONG_SCALABLE is not set
 # CONFIG_IPV6 is not set
 # CONFIG_NETFILTER is not set
 
@@ -405,6 +429,7 @@ CONFIG_MII=y
 # CONFIG_SUNGEM is not set
 # CONFIG_NET_VENDOR_3COM is not set
 # CONFIG_SMC91X is not set
+# CONFIG_DM9000 is not set
 
 #
 # Tulip family network device support
@@ -442,9 +467,11 @@ CONFIG_EEPRO100=y
 # CONFIG_HAMACHI is not set
 # CONFIG_YELLOWFIN is not set
 # CONFIG_R8169 is not set
+# CONFIG_SKGE is not set
 # CONFIG_SK98LIN is not set
 # CONFIG_VIA_VELOCITY is not set
 # CONFIG_TIGON3 is not set
+# CONFIG_BNX2 is not set
 
 #
 # Ethernet (10000 Mbit)
@@ -466,6 +493,7 @@ CONFIG_EEPRO100=y
 # Wan interfaces
 #
 CONFIG_WAN=y
+# CONFIG_DSCC4 is not set
 # CONFIG_LANMEDIA is not set
 # CONFIG_SYNCLINK_SYNCPPP is not set
 CONFIG_HDLC=y
@@ -528,7 +556,6 @@ CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
 #
 # CONFIG_SERIO is not set
 # CONFIG_GAMEPORT is not set
-CONFIG_SOUND_GAMEPORT=y
 
 #
 # Character devices
@@ -549,6 +576,7 @@ CONFIG_SERIAL_8250_NR_UARTS=2
 #
 CONFIG_SERIAL_CORE=y
 CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
 CONFIG_UNIX98_PTYS=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
@@ -615,17 +643,18 @@ CONFIG_I2C_ALGOBIT=y
 # CONFIG_I2C_AMD8111 is not set
 # CONFIG_I2C_I801 is not set
 # CONFIG_I2C_I810 is not set
+# CONFIG_I2C_PIIX4 is not set
 # CONFIG_I2C_ISA is not set
 # CONFIG_I2C_IXP2000 is not set
 # CONFIG_I2C_NFORCE2 is not set
 # CONFIG_I2C_PARPORT_LIGHT is not set
-# CONFIG_I2C_PIIX4 is not set
 # CONFIG_I2C_PROSAVAGE is not set
 # CONFIG_I2C_SAVAGE4 is not set
 # CONFIG_SCx200_ACB is not set
 # CONFIG_I2C_SIS5595 is not set
 # CONFIG_I2C_SIS630 is not set
 # CONFIG_I2C_SIS96X is not set
+# CONFIG_I2C_STUB is not set
 # CONFIG_I2C_VIA is not set
 # CONFIG_I2C_VIAPRO is not set
 # CONFIG_I2C_VOODOO3 is not set
@@ -639,7 +668,9 @@ CONFIG_I2C_SENSOR=y
 # CONFIG_SENSORS_ADM1025 is not set
 # CONFIG_SENSORS_ADM1026 is not set
 # CONFIG_SENSORS_ADM1031 is not set
+# CONFIG_SENSORS_ADM9240 is not set
 # CONFIG_SENSORS_ASB100 is not set
+# CONFIG_SENSORS_ATXP1 is not set
 # CONFIG_SENSORS_DS1621 is not set
 # CONFIG_SENSORS_FSCHER is not set
 # CONFIG_SENSORS_FSCPOS is not set
@@ -655,6 +686,7 @@ CONFIG_I2C_SENSOR=y
 # 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
@@ -664,14 +696,19 @@ CONFIG_I2C_SENSOR=y
 # CONFIG_SENSORS_W83781D is not set
 # CONFIG_SENSORS_W83L785TS is not set
 # CONFIG_SENSORS_W83627HF is not set
+# CONFIG_SENSORS_W83627EHF is not set
 
 #
 # Other I2C Chip support
 #
+# CONFIG_SENSORS_DS1337 is not set
+# CONFIG_SENSORS_DS1374 is not set
 CONFIG_SENSORS_EEPROM=y
 # CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_SENSORS_PCA9539 is not set
 # CONFIG_SENSORS_PCF8591 is not set
 # CONFIG_SENSORS_RTC8564 is not set
+# CONFIG_SENSORS_MAX6875 is not set
 # CONFIG_I2C_DEBUG_CORE is not set
 # CONFIG_I2C_DEBUG_ALGO is not set
 # CONFIG_I2C_DEBUG_BUS is not set
@@ -725,6 +762,7 @@ CONFIG_EXT2_FS=y
 CONFIG_EXT2_FS_XATTR=y
 CONFIG_EXT2_FS_POSIX_ACL=y
 # CONFIG_EXT2_FS_SECURITY is not set
+# CONFIG_EXT2_FS_XIP is not set
 CONFIG_EXT3_FS=y
 CONFIG_EXT3_FS_XATTR=y
 CONFIG_EXT3_FS_POSIX_ACL=y
@@ -765,7 +803,6 @@ CONFIG_DNOTIFY=y
 #
 CONFIG_PROC_FS=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
@@ -803,12 +840,14 @@ CONFIG_JFFS2_RTIME=y
 #
 CONFIG_NFS_FS=y
 CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL 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_LOCKD_V4=y
+CONFIG_NFS_COMMON=y
 CONFIG_SUNRPC=y
 # CONFIG_RPCSEC_GSS_KRB5 is not set
 # CONFIG_RPCSEC_GSS_SPKM3 is not set
@@ -893,3 +932,4 @@ CONFIG_CRC32=y
 # CONFIG_LIBCRC32C is not set
 CONFIG_ZLIB_INFLATE=y
 CONFIG_ZLIB_DEFLATE=y
+# CONFIG_TEXTSEARCH is not set
-- 
cgit v1.2.3-70-g09d2


From ea23d1ac7e380aefc00d691c7c9a3e747bc2531f Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Sat, 25 Jun 2005 16:58:21 +0100
Subject: [PATCH] ARM: 2750/1: add i2c platform device for enp2611 on-board i2c
 bus

Patch from Lennert Buytenhek

On the enp2611, GPIO 7 and 6 are connected to an on-board i2c bus that
attaches to the SODIMM module slot (for SPD) and an LM84 temperature
sensor.  Add a platform device for this i2c bus.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-ixp2000/enp2611.c | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/arm/mach-ixp2000/enp2611.c b/arch/arm/mach-ixp2000/enp2611.c
index 04b38bcf9aa..f3a291b6a9f 100644
--- a/arch/arm/mach-ixp2000/enp2611.c
+++ b/arch/arm/mach-ixp2000/enp2611.c
@@ -197,8 +197,23 @@ static struct platform_device enp2611_flash = {
 	.resource	= &enp2611_flash_resource,
 };
 
+static struct ixp2000_i2c_pins enp2611_i2c_gpio_pins = {
+	.sda_pin	= ENP2611_GPIO_SDA,
+	.scl_pin	= ENP2611_GPIO_SCL,
+};
+
+static struct platform_device enp2611_i2c_controller = {
+	.name		= "IXP2000-I2C",
+	.id		= 0,
+	.dev		= {
+		.platform_data = &enp2611_i2c_gpio_pins
+	},
+	.num_resources	= 0
+};
+
 static struct platform_device *enp2611_devices[] __initdata = {
-	&enp2611_flash
+	&enp2611_flash,
+	&enp2611_i2c_controller
 };
 
 static void __init enp2611_init_machine(void)
-- 
cgit v1.2.3-70-g09d2


From 8144f56bd1e49015f94b8da99b24c4d04643b66d Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Sat, 25 Jun 2005 16:58:22 +0100
Subject: [PATCH] ARM: 2751/1: ixp2000 gpio cleanup broke ixdp2800 build

Patch from Lennert Buytenhek

The ixp2000 gpio cleanup broke the ixdp2800 build as it moved some
gpio-related functions from arch/platform.h to arch/gpio.h and the
ixdp2x00 support code used those functions but didn't include the
latter header file.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-ixp2000/ixdp2x00.c | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'arch')

diff --git a/arch/arm/mach-ixp2000/ixdp2x00.c b/arch/arm/mach-ixp2000/ixdp2x00.c
index 21c41fe15b9..5e4380747b5 100644
--- a/arch/arm/mach-ixp2000/ixdp2x00.c
+++ b/arch/arm/mach-ixp2000/ixdp2x00.c
@@ -42,6 +42,9 @@
 #include <asm/mach/flash.h>
 #include <asm/mach/arch.h>
 
+#include <asm/arch/gpio.h>
+
+
 /*************************************************************************
  * IXDP2x00 IRQ Initialization
  *************************************************************************/
-- 
cgit v1.2.3-70-g09d2


From 3cd9e19ebc91593c9f076410d6f979be188f01a0 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Sat, 25 Jun 2005 19:29:34 +0100
Subject: [PATCH] ARM: Fix discontigmem

The merge of sparsemem broke ARM discontigmem.  Fix it.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig               | 2 +-
 arch/arm/mach-clps711x/Kconfig | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 07ba77c19f6..bfcf4228036 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -157,7 +157,7 @@ config ARCH_RPC
 config ARCH_SA1100
 	bool "SA1100-based"
 	select ISA
-	select DISCONTIGMEM
+	select ARCH_DISCONTIGMEM_ENABLE
 
 config ARCH_S3C2410
 	bool "Samsung S3C2410"
diff --git a/arch/arm/mach-clps711x/Kconfig b/arch/arm/mach-clps711x/Kconfig
index 45c930ccd06..0793dcf54f2 100644
--- a/arch/arm/mach-clps711x/Kconfig
+++ b/arch/arm/mach-clps711x/Kconfig
@@ -28,7 +28,7 @@ config ARCH_CLEP7312
 config ARCH_EDB7211
 	bool "EDB7211"
 	select ISA
-	select DISCONTIGMEM
+	select ARCH_DISCONTIGMEM_ENABLE
 	help
 	  Say Y here if you intend to run this kernel on a Cirrus Logic EDB-7211
 	  evaluation board.
-- 
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 'arch')

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 'arch')

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 'arch')

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 821fe94727e72d63e1abd1bc0b044c72dfad9fb6 Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Sat, 25 Jun 2005 14:54:29 -0700
Subject: [PATCH] gcc4 compile fix for recent ia64 xpc changes

Gcc4 doesn't like volatile casts as lvalues.  Make the structure members
volatile instead.

Signed-off-by: Dave Jones <davej@redhat.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/sn/kernel/xpc.h           | 12 ++++++------
 arch/ia64/sn/kernel/xpc_channel.c   | 12 ++++++------
 arch/ia64/sn/kernel/xpc_partition.c |  2 +-
 3 files changed, 13 insertions(+), 13 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/sn/kernel/xpc.h b/arch/ia64/sn/kernel/xpc.h
index 1a0aed8490d..d0ee635daf2 100644
--- a/arch/ia64/sn/kernel/xpc.h
+++ b/arch/ia64/sn/kernel/xpc.h
@@ -87,7 +87,7 @@ struct xpc_rsvd_page {
 	u8 partid;		/* partition ID from SAL */
 	u8 version;
 	u8 pad[6];		/* pad to u64 align */
-	u64 vars_pa;
+	volatile u64 vars_pa;
 	u64 part_nasids[XP_NASID_MASK_WORDS] ____cacheline_aligned;
 	u64 mach_nasids[XP_NASID_MASK_WORDS] ____cacheline_aligned;
 };
@@ -138,7 +138,7 @@ struct xpc_vars {
  * occupies half a cacheline.
  */
 struct xpc_vars_part {
-	u64 magic;
+	volatile u64 magic;
 
 	u64 openclose_args_pa;	/* physical address of open and close args */
 	u64 GPs_pa;		/* physical address of Get/Put values */
@@ -185,8 +185,8 @@ struct xpc_vars_part {
  * Define a Get/Put value pair (pointers) used with a message queue.
  */
 struct xpc_gp {
-	s64 get;	/* Get value */
-	s64 put;	/* Put value */
+	volatile s64 get;	/* Get value */
+	volatile s64 put;	/* Put value */
 };
 
 #define XPC_GP_SIZE \
@@ -231,7 +231,7 @@ struct xpc_openclose_args {
  */
 struct xpc_notify {
 	struct semaphore sema;		/* notify semaphore */
-	u8 type;			/* type of notification */
+	volatile u8 type;			/* type of notification */
 
 	/* the following two fields are only used if type == XPC_N_CALL */
 	xpc_notify_func func;		/* user's notify function */
@@ -439,7 +439,7 @@ struct xpc_partition {
 
 	/* XPC infrastructure referencing and teardown control */
 
-	u8 setup_state;			/* infrastructure setup state */
+	volatile u8 setup_state;			/* infrastructure setup state */
 	wait_queue_head_t teardown_wq;	/* kthread waiting to teardown infra */
 	atomic_t references;		/* #of references to infrastructure */
 
diff --git a/arch/ia64/sn/kernel/xpc_channel.c b/arch/ia64/sn/kernel/xpc_channel.c
index 0bf6fbcc46d..6d02dac8056 100644
--- a/arch/ia64/sn/kernel/xpc_channel.c
+++ b/arch/ia64/sn/kernel/xpc_channel.c
@@ -209,7 +209,7 @@ xpc_setup_infrastructure(struct xpc_partition *part)
 	 * With the setting of the partition setup_state to XPC_P_SETUP, we're
 	 * declaring that this partition is ready to go.
 	 */
-	(volatile u8) part->setup_state = XPC_P_SETUP;
+	part->setup_state = XPC_P_SETUP;
 
 
 	/*
@@ -227,7 +227,7 @@ xpc_setup_infrastructure(struct xpc_partition *part)
 	xpc_vars_part[partid].IPI_phys_cpuid =
 					cpu_physical_id(smp_processor_id());
 	xpc_vars_part[partid].nchannels = part->nchannels;
-	(volatile u64) xpc_vars_part[partid].magic = XPC_VP_MAGIC1;
+	xpc_vars_part[partid].magic = XPC_VP_MAGIC1;
 
 	return xpcSuccess;
 }
@@ -355,7 +355,7 @@ xpc_pull_remote_vars_part(struct xpc_partition *part)
 
 		/* let the other side know that we've pulled their variables */
 
-		(volatile u64) xpc_vars_part[partid].magic = XPC_VP_MAGIC2;
+		xpc_vars_part[partid].magic = XPC_VP_MAGIC2;
 	}
 
 	if (pulled_entry->magic == XPC_VP_MAGIC1) {
@@ -1183,7 +1183,7 @@ xpc_process_msg_IPI(struct xpc_partition *part, int ch_number)
 		 */
 		xpc_clear_local_msgqueue_flags(ch);
 
-		(volatile s64) ch->w_remote_GP.get = ch->remote_GP.get;
+		ch->w_remote_GP.get = ch->remote_GP.get;
 
 		dev_dbg(xpc_chan, "w_remote_GP.get changed to %ld, partid=%d, "
 			"channel=%d\n", ch->w_remote_GP.get, ch->partid,
@@ -1211,7 +1211,7 @@ xpc_process_msg_IPI(struct xpc_partition *part, int ch_number)
 		 */
 		xpc_clear_remote_msgqueue_flags(ch);
 
-		(volatile s64) ch->w_remote_GP.put = ch->remote_GP.put;
+		ch->w_remote_GP.put = ch->remote_GP.put;
 
 		dev_dbg(xpc_chan, "w_remote_GP.put changed to %ld, partid=%d, "
 			"channel=%d\n", ch->w_remote_GP.put, ch->partid,
@@ -1875,7 +1875,7 @@ xpc_send_msg(struct xpc_channel *ch, struct xpc_msg *msg, u8 notify_type,
 		notify = &ch->notify_queue[msg_number % ch->local_nentries];
 		notify->func = func;
 		notify->key = key;
-		(volatile u8) notify->type = notify_type;
+		notify->type = notify_type;
 
 		// >>> is a mb() needed here?
 
diff --git a/arch/ia64/sn/kernel/xpc_partition.c b/arch/ia64/sn/kernel/xpc_partition.c
index cd7ed73f0e7..578265ea9e6 100644
--- a/arch/ia64/sn/kernel/xpc_partition.c
+++ b/arch/ia64/sn/kernel/xpc_partition.c
@@ -253,7 +253,7 @@ xpc_rsvd_page_init(void)
 	 * This signifies to the remote partition that our reserved
 	 * page is initialized.
 	 */
-	(volatile u64) rp->vars_pa = __pa(xpc_vars);
+	rp->vars_pa = __pa(xpc_vars);
 
 	return rp;
 }
-- 
cgit v1.2.3-70-g09d2


From 7919a693bd735ed0aa93fc359ae09a588cfeb3bc Mon Sep 17 00:00:00 2001
From: Russell King <rmk+lkml@arm.linux.org.uk>
Date: Sat, 25 Jun 2005 14:54:30 -0700
Subject: [PATCH] Serial: remove unnecessary register_serial/unregister_serial

A couple of drivers declare register_serial/unregister_serial prototypes
but don't use them.  FRV contains a commented out call to register_serial.
Since these are deprecated, remove these unnecessary references.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/frv/kernel/setup.c            | 2 --
 arch/xtensa/platform-iss/console.c | 3 ---
 drivers/char/amiserial.c           | 4 ----
 3 files changed, 9 deletions(-)

(limited to 'arch')

diff --git a/arch/frv/kernel/setup.c b/arch/frv/kernel/setup.c
index ef6865f0b97..767ebb55bd8 100644
--- a/arch/frv/kernel/setup.c
+++ b/arch/frv/kernel/setup.c
@@ -790,12 +790,10 @@ void __init setup_arch(char **cmdline_p)
 #ifndef CONFIG_GDBSTUB_UART0
 	__reg(UART0_BASE + UART_IER * 8) = 0;
 	early_serial_setup(&__frv_uart0);
-//	register_serial(&__frv_uart0);
 #endif
 #ifndef CONFIG_GDBSTUB_UART1
 	__reg(UART1_BASE + UART_IER * 8) = 0;
 	early_serial_setup(&__frv_uart1);
-//	register_serial(&__frv_uart1);
 #endif
 
 #if defined(CONFIG_CHR_DEV_FLASH) || defined(CONFIG_BLK_DEV_FLASH)
diff --git a/arch/xtensa/platform-iss/console.c b/arch/xtensa/platform-iss/console.c
index 9e2b53f6a90..4fbddf92a92 100644
--- a/arch/xtensa/platform-iss/console.c
+++ b/arch/xtensa/platform-iss/console.c
@@ -198,9 +198,6 @@ static int rs_read_proc(char *page, char **start, off_t off, int count,
 }
 
 
-int register_serial(struct serial_struct*);
-void unregister_serial(int);
-
 static struct tty_operations serial_ops = {
 	.open = rs_open,
 	.close = rs_close,
diff --git a/drivers/char/amiserial.c b/drivers/char/amiserial.c
index 777bc499bbb..2a36561eec6 100644
--- a/drivers/char/amiserial.c
+++ b/drivers/char/amiserial.c
@@ -1973,10 +1973,6 @@ static _INLINE_ void show_serial_version(void)
 }
 
 
-int register_serial(struct serial_struct *req);
-void unregister_serial(int line);
-
-
 static struct tty_operations serial_ops = {
 	.open = rs_open,
 	.close = rs_close,
-- 
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 'arch')

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 62aa751d16399637325852bc0a1fcf13c2476dd7 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Sat, 25 Jun 2005 14:54:36 -0700
Subject: [PATCH] ppc32: Check return of ppc_sys_get_pdata before accessing
 pointer

Ensure that the returned pointer from ppc_sys_get_pdata is not NULL before we
start using it.  This handles any cases where we have variants of processors
on the same board with different functionality.

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/83xx/mpc834x_sys.c | 28 ++++++++++++---------
 arch/ppc/platforms/85xx/mpc8540_ads.c | 46 ++++++++++++++++++++---------------
 arch/ppc/platforms/85xx/mpc8560_ads.c | 28 ++++++++++++---------
 arch/ppc/platforms/85xx/sbc8560.c     | 28 ++++++++++++---------
 arch/ppc/platforms/85xx/stx_gp3.c     | 26 +++++++++++---------
 5 files changed, 89 insertions(+), 67 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc/platforms/83xx/mpc834x_sys.c b/arch/ppc/platforms/83xx/mpc834x_sys.c
index 37ece154279..ddd04d4c1ea 100644
--- a/arch/ppc/platforms/83xx/mpc834x_sys.c
+++ b/arch/ppc/platforms/83xx/mpc834x_sys.c
@@ -94,20 +94,24 @@ mpc834x_sys_setup_arch(void)
 
 	/* setup the board related information for the enet controllers */
 	pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC83xx_TSEC1);
-	pdata->board_flags = FSL_GIANFAR_BRD_HAS_PHY_INTR;
-	pdata->interruptPHY = MPC83xx_IRQ_EXT1;
-	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 = MPC83xx_IRQ_EXT1;
+		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(MPC83xx_TSEC2);
-	pdata->board_flags = FSL_GIANFAR_BRD_HAS_PHY_INTR;
-	pdata->interruptPHY = MPC83xx_IRQ_EXT2;
-	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 = MPC83xx_IRQ_EXT2;
+		pdata->phyid = 1;
+		/* fixup phy address */
+		pdata->phy_reg_addr += binfo->bi_immr_base;
+		memcpy(pdata->mac_addr, binfo->bi_enet1addr, 6);
+	}
 
 #ifdef CONFIG_BLK_DEV_INITRD
 	if (initrd_start)
diff --git a/arch/ppc/platforms/85xx/mpc8540_ads.c b/arch/ppc/platforms/85xx/mpc8540_ads.c
index a2ed611cd93..ddd2e9a5bb1 100644
--- a/arch/ppc/platforms/85xx/mpc8540_ads.c
+++ b/arch/ppc/platforms/85xx/mpc8540_ads.c
@@ -92,28 +92,34 @@ mpc8540ads_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);
-
-	pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC85xx_FEC);
-	pdata->board_flags = 0;
-	pdata->interruptPHY = MPC85xx_IRQ_EXT5;
-	pdata->phyid = 3;
-	/* fixup phy address */
-	pdata->phy_reg_addr += binfo->bi_immr_base;
-	memcpy(pdata->mac_addr, binfo->bi_enet2addr, 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);
+	}
+
+	if (pdata) {
+		pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC85xx_FEC);
+		pdata->board_flags = 0;
+		pdata->interruptPHY = MPC85xx_IRQ_EXT5;
+		pdata->phyid = 3;
+		/* fixup phy address */
+		pdata->phy_reg_addr += binfo->bi_immr_base;
+		memcpy(pdata->mac_addr, binfo->bi_enet2addr, 6);
+	}
 
 #ifdef CONFIG_BLK_DEV_INITRD
 	if (initrd_start)
diff --git a/arch/ppc/platforms/85xx/mpc8560_ads.c b/arch/ppc/platforms/85xx/mpc8560_ads.c
index d87dfd5ce0a..e18380258b6 100644
--- a/arch/ppc/platforms/85xx/mpc8560_ads.c
+++ b/arch/ppc/platforms/85xx/mpc8560_ads.c
@@ -90,20 +90,24 @@ mpc8560ads_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);
+	}
 
 #ifdef CONFIG_BLK_DEV_INITRD
 	if (initrd_start)
diff --git a/arch/ppc/platforms/85xx/sbc8560.c b/arch/ppc/platforms/85xx/sbc8560.c
index 3dbdd73618e..165df94d4aa 100644
--- a/arch/ppc/platforms/85xx/sbc8560.c
+++ b/arch/ppc/platforms/85xx/sbc8560.c
@@ -129,20 +129,24 @@ sbc8560_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_EXT6;
-	pdata->phyid = 25;
-	/* 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_EXT6;
+		pdata->phyid = 25;
+		/* 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_EXT7;
-	pdata->phyid = 26;
-	/* 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_EXT7;
+		pdata->phyid = 26;
+		/* fixup phy address */
+		pdata->phy_reg_addr += binfo->bi_immr_base;
+		memcpy(pdata->mac_addr, binfo->bi_enet1addr, 6);
+	}
 
 #ifdef CONFIG_BLK_DEV_INITRD
 	if (initrd_start)
diff --git a/arch/ppc/platforms/85xx/stx_gp3.c b/arch/ppc/platforms/85xx/stx_gp3.c
index 9455bb6b45e..bb41265cfc8 100644
--- a/arch/ppc/platforms/85xx/stx_gp3.c
+++ b/arch/ppc/platforms/85xx/stx_gp3.c
@@ -122,19 +122,23 @@ gp3_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 = 2;
-	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 = 2;
+		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 = 4;
-	/* 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 = 4;
+		/* fixup phy address */
+		pdata->phy_reg_addr += binfo->bi_immr_base;
+		memcpy(pdata->mac_addr, binfo->bi_enet1addr, 6);
+	}
 
 #ifdef CONFIG_BLK_DEV_INITRD
 	if (initrd_start)
-- 
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 'arch')

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 'arch')

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 2ec19faf617e61d56fb8da78ccb3ef895c00fae7 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Sat, 25 Jun 2005 14:54:41 -0700
Subject: [PATCH] ppc32: remove some unnecessary includes of bootmem.h

Continue the Good Fight:  Limit bootmem.h include creep.

Signed-off-by: Jon Loeliger <jdl@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/mm/44x_mmu.c          | 1 -
 arch/ppc/mm/4xx_mmu.c          | 1 -
 arch/ppc/mm/fsl_booke_mmu.c    | 1 -
 arch/ppc/platforms/chrp_pci.c  | 1 -
 arch/ppc/platforms/katana.c    | 2 +-
 arch/ppc/platforms/pmac_pci.c  | 1 -
 arch/ppc/syslib/cpm2_common.c  | 2 +-
 arch/ppc/syslib/indirect_pci.c | 1 -
 arch/ppc/syslib/mv64x60.c      | 1 -
 arch/ppc/syslib/mv64x60_win.c  | 1 -
 10 files changed, 2 insertions(+), 10 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc/mm/44x_mmu.c b/arch/ppc/mm/44x_mmu.c
index 72f7c0d1c0e..3d79ce281b6 100644
--- a/arch/ppc/mm/44x_mmu.c
+++ b/arch/ppc/mm/44x_mmu.c
@@ -39,7 +39,6 @@
 #include <linux/vmalloc.h>
 #include <linux/init.h>
 #include <linux/delay.h>
-#include <linux/bootmem.h>
 #include <linux/highmem.h>
 
 #include <asm/pgalloc.h>
diff --git a/arch/ppc/mm/4xx_mmu.c b/arch/ppc/mm/4xx_mmu.c
index a7f61614038..b7bcbc232f3 100644
--- a/arch/ppc/mm/4xx_mmu.c
+++ b/arch/ppc/mm/4xx_mmu.c
@@ -36,7 +36,6 @@
 #include <linux/vmalloc.h>
 #include <linux/init.h>
 #include <linux/delay.h>
-#include <linux/bootmem.h>
 #include <linux/highmem.h>
 
 #include <asm/pgalloc.h>
diff --git a/arch/ppc/mm/fsl_booke_mmu.c b/arch/ppc/mm/fsl_booke_mmu.c
index 9fa884de5c7..af9ca0eb6d5 100644
--- a/arch/ppc/mm/fsl_booke_mmu.c
+++ b/arch/ppc/mm/fsl_booke_mmu.c
@@ -41,7 +41,6 @@
 #include <linux/vmalloc.h>
 #include <linux/init.h>
 #include <linux/delay.h>
-#include <linux/bootmem.h>
 #include <linux/highmem.h>
 
 #include <asm/pgalloc.h>
diff --git a/arch/ppc/platforms/chrp_pci.c b/arch/ppc/platforms/chrp_pci.c
index 7d0ee308f66..7d3fbb5c5db 100644
--- a/arch/ppc/platforms/chrp_pci.c
+++ b/arch/ppc/platforms/chrp_pci.c
@@ -9,7 +9,6 @@
 #include <linux/string.h>
 #include <linux/init.h>
 #include <linux/ide.h>
-#include <linux/bootmem.h>
 
 #include <asm/io.h>
 #include <asm/pgtable.h>
diff --git a/arch/ppc/platforms/katana.c b/arch/ppc/platforms/katana.c
index eda922ac316..169dbf6534b 100644
--- a/arch/ppc/platforms/katana.c
+++ b/arch/ppc/platforms/katana.c
@@ -27,12 +27,12 @@
 #include <linux/root_dev.h>
 #include <linux/delay.h>
 #include <linux/seq_file.h>
-#include <linux/bootmem.h>
 #include <linux/mtd/physmap.h>
 #include <linux/mv643xx.h>
 #ifdef CONFIG_BOOTIMG
 #include <linux/bootimg.h>
 #endif
+#include <asm/io.h>
 #include <asm/page.h>
 #include <asm/time.h>
 #include <asm/smp.h>
diff --git a/arch/ppc/platforms/pmac_pci.c b/arch/ppc/platforms/pmac_pci.c
index f6ff5192406..719fb49fe2b 100644
--- a/arch/ppc/platforms/pmac_pci.c
+++ b/arch/ppc/platforms/pmac_pci.c
@@ -17,7 +17,6 @@
 #include <linux/delay.h>
 #include <linux/string.h>
 #include <linux/init.h>
-#include <linux/bootmem.h>
 
 #include <asm/sections.h>
 #include <asm/io.h>
diff --git a/arch/ppc/syslib/cpm2_common.c b/arch/ppc/syslib/cpm2_common.c
index ea5e77080e8..4c19a4ac716 100644
--- a/arch/ppc/syslib/cpm2_common.c
+++ b/arch/ppc/syslib/cpm2_common.c
@@ -21,8 +21,8 @@
 #include <linux/string.h>
 #include <linux/mm.h>
 #include <linux/interrupt.h>
-#include <linux/bootmem.h>
 #include <linux/module.h>
+#include <asm/io.h>
 #include <asm/irq.h>
 #include <asm/mpc8260.h>
 #include <asm/page.h>
diff --git a/arch/ppc/syslib/indirect_pci.c b/arch/ppc/syslib/indirect_pci.c
index a5a752609e2..e7148846970 100644
--- a/arch/ppc/syslib/indirect_pci.c
+++ b/arch/ppc/syslib/indirect_pci.c
@@ -14,7 +14,6 @@
 #include <linux/delay.h>
 #include <linux/string.h>
 #include <linux/init.h>
-#include <linux/bootmem.h>
 
 #include <asm/io.h>
 #include <asm/prom.h>
diff --git a/arch/ppc/syslib/mv64x60.c b/arch/ppc/syslib/mv64x60.c
index 7b241e7876b..cc77177fa1c 100644
--- a/arch/ppc/syslib/mv64x60.c
+++ b/arch/ppc/syslib/mv64x60.c
@@ -17,7 +17,6 @@
 #include <linux/slab.h>
 #include <linux/module.h>
 #include <linux/string.h>
-#include <linux/bootmem.h>
 #include <linux/spinlock.h>
 #include <linux/mv643xx.h>
 
diff --git a/arch/ppc/syslib/mv64x60_win.c b/arch/ppc/syslib/mv64x60_win.c
index b6f0f5dcf6e..5b827e2bbe2 100644
--- a/arch/ppc/syslib/mv64x60_win.c
+++ b/arch/ppc/syslib/mv64x60_win.c
@@ -17,7 +17,6 @@
 #include <linux/slab.h>
 #include <linux/module.h>
 #include <linux/string.h>
-#include <linux/bootmem.h>
 #include <linux/mv643xx.h>
 
 #include <asm/byteorder.h>
-- 
cgit v1.2.3-70-g09d2


From 856509d5da74aefeabe92a8df039cc472f4f7c5f Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Sat, 25 Jun 2005 14:54:42 -0700
Subject: [PATCH] ppc64: Fix compile warnings in arch/ppc64/kernel/lparcfg.c

Stephen's patch to remove LparData.h missed an include in lparcfg.c This
fixes a few compile warnings.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc64/kernel/lparcfg.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/lparcfg.c b/arch/ppc64/kernel/lparcfg.c
index 387923fcf9b..02e96627fa6 100644
--- a/arch/ppc64/kernel/lparcfg.c
+++ b/arch/ppc64/kernel/lparcfg.c
@@ -34,6 +34,7 @@
 #include <asm/system.h>
 #include <asm/time.h>
 #include <asm/iSeries/ItExtVpdPanel.h>
+#include <asm/prom.h>
 
 #define MODULE_VERS "1.6"
 #define MODULE_NAME "lparcfg"
-- 
cgit v1.2.3-70-g09d2


From aea00143a8db8c0b31dca85bff3c325444d93f0f Mon Sep 17 00:00:00 2001
From: Andrey Panin <pazke@donpac.ru>
Date: Sat, 25 Jun 2005 14:54:42 -0700
Subject: [PATCH] dmi: move ACPI boot quirk

This patch moves ACPI boot quirks out of dmi_scan.c

Signed-off-by: Andrey Panin <pazke@donpac.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/acpi/boot.c | 219 ++++++++++++++++++++++++++++++++++++++++++-
 arch/i386/kernel/dmi_scan.c  | 163 --------------------------------
 2 files changed, 217 insertions(+), 165 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index 848bb97af7c..0771596c549 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -29,6 +29,7 @@
 #include <linux/efi.h>
 #include <linux/irq.h>
 #include <linux/module.h>
+#include <linux/dmi.h>
 
 #include <asm/pgtable.h>
 #include <asm/io_apic.h>
@@ -815,6 +816,218 @@ acpi_process_madt(void)
 	return;
 }
 
+extern int acpi_force;
+
+#ifdef __i386__
+
+#ifdef	CONFIG_ACPI_PCI
+static int __init disable_acpi_irq(struct dmi_system_id *d)
+{
+	if (!acpi_force) {
+		printk(KERN_NOTICE "%s detected: force use of acpi=noirq\n",
+		       d->ident);
+		acpi_noirq_set();
+	}
+	return 0;
+}
+
+static int __init disable_acpi_pci(struct dmi_system_id *d)
+{
+	if (!acpi_force) {
+		printk(KERN_NOTICE "%s detected: force use of pci=noacpi\n",
+		       d->ident);
+		acpi_disable_pci();
+	}
+	return 0;
+}
+#endif
+
+static int __init dmi_disable_acpi(struct dmi_system_id *d)
+{
+	if (!acpi_force) {
+		printk(KERN_NOTICE "%s detected: acpi off\n",d->ident);
+		disable_acpi();
+	} else {
+		printk(KERN_NOTICE
+		       "Warning: DMI blacklist says broken, but acpi forced\n");
+	}
+	return 0;
+}
+
+/*
+ * Limit ACPI to CPU enumeration for HT
+ */
+static int __init force_acpi_ht(struct dmi_system_id *d)
+{
+	if (!acpi_force) {
+		printk(KERN_NOTICE "%s detected: force use of acpi=ht\n", d->ident);
+		disable_acpi();
+		acpi_ht = 1;
+	} else {
+		printk(KERN_NOTICE
+		       "Warning: acpi=force overrules DMI blacklist: acpi=ht\n");
+	}
+	return 0;
+}
+
+/*
+ * If your system is blacklisted here, but you find that acpi=force
+ * works for you, please contact acpi-devel@sourceforge.net
+ */
+static struct dmi_system_id __initdata acpi_dmi_table[] = {
+	/*
+	 * Boxes that need ACPI disabled
+	 */
+	{
+		.callback = dmi_disable_acpi,
+		.ident = "IBM Thinkpad",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "IBM"),
+			DMI_MATCH(DMI_BOARD_NAME, "2629H1G"),
+		},
+	},
+
+	/*
+	 * Boxes that need acpi=ht
+	 */
+	{
+		.callback = force_acpi_ht,
+		.ident = "FSC Primergy T850",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "FUJITSU SIEMENS"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "PRIMERGY T850"),
+		},
+	},
+	{
+		.callback = force_acpi_ht,
+		.ident = "DELL GX240",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "Dell Computer Corporation"),
+			DMI_MATCH(DMI_BOARD_NAME, "OptiPlex GX240"),
+		},
+	},
+	{
+		.callback = force_acpi_ht,
+		.ident = "HP VISUALIZE NT Workstation",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "Hewlett-Packard"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "HP VISUALIZE NT Workstation"),
+		},
+	},
+	{
+		.callback = force_acpi_ht,
+		.ident = "Compaq Workstation W8000",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Compaq"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Workstation W8000"),
+		},
+	},
+	{
+		.callback = force_acpi_ht,
+		.ident = "ASUS P4B266",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
+			DMI_MATCH(DMI_BOARD_NAME, "P4B266"),
+		},
+	},
+	{
+		.callback = force_acpi_ht,
+		.ident = "ASUS P2B-DS",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
+			DMI_MATCH(DMI_BOARD_NAME, "P2B-DS"),
+		},
+	},
+	{
+		.callback = force_acpi_ht,
+		.ident = "ASUS CUR-DLS",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
+			DMI_MATCH(DMI_BOARD_NAME, "CUR-DLS"),
+		},
+	},
+	{
+		.callback = force_acpi_ht,
+		.ident = "ABIT i440BX-W83977",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "ABIT <http://www.abit.com>"),
+			DMI_MATCH(DMI_BOARD_NAME, "i440BX-W83977 (BP6)"),
+		},
+	},
+	{
+		.callback = force_acpi_ht,
+		.ident = "IBM Bladecenter",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "IBM"),
+			DMI_MATCH(DMI_BOARD_NAME, "IBM eServer BladeCenter HS20"),
+		},
+	},
+	{
+		.callback = force_acpi_ht,
+		.ident = "IBM eServer xSeries 360",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "IBM"),
+			DMI_MATCH(DMI_BOARD_NAME, "eServer xSeries 360"),
+		},
+	},
+	{
+		.callback = force_acpi_ht,
+		.ident = "IBM eserver xSeries 330",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "IBM"),
+			DMI_MATCH(DMI_BOARD_NAME, "eserver xSeries 330"),
+		},
+	},
+	{
+		.callback = force_acpi_ht,
+		.ident = "IBM eserver xSeries 440",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "IBM"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "eserver xSeries 440"),
+		},
+	},
+
+#ifdef	CONFIG_ACPI_PCI
+	/*
+	 * Boxes that need ACPI PCI IRQ routing disabled
+	 */
+	{
+		.callback = disable_acpi_irq,
+		.ident = "ASUS A7V",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC"),
+			DMI_MATCH(DMI_BOARD_NAME, "<A7V>"),
+			/* newer BIOS, Revision 1011, does work */
+			DMI_MATCH(DMI_BIOS_VERSION, "ASUS A7V ACPI BIOS Revision 1007"),
+		},
+	},
+
+	/*
+	 * Boxes that need ACPI PCI IRQ routing and PCI scan disabled
+	 */
+	{	/* _BBN 0 bug */
+		.callback = disable_acpi_pci,
+		.ident = "ASUS PR-DLS",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
+			DMI_MATCH(DMI_BOARD_NAME, "PR-DLS"),
+			DMI_MATCH(DMI_BIOS_VERSION, "ASUS PR-DLS ACPI BIOS Revision 1010"),
+			DMI_MATCH(DMI_BIOS_DATE, "03/21/2003")
+		},
+	},
+	{
+		.callback = disable_acpi_pci,
+		.ident = "Acer TravelMate 36x Laptop",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Acer"),
+ 			DMI_MATCH(DMI_PRODUCT_NAME, "TravelMate 360"),
+		},
+	},
+#endif
+};
+
+#endif	/* __i386__ */
+
 /*
  * acpi_boot_table_init() and acpi_boot_init()
  *  called from setup_arch(), always.
@@ -843,6 +1056,10 @@ acpi_boot_table_init(void)
 {
 	int error;
 
+#ifdef __i386__
+	dmi_check_system(acpi_dmi_table);
+#endif
+
 	/*
 	 * If acpi_disabled, bail out
 	 * One exception: acpi=ht continues far enough to enumerate LAPICs
@@ -870,8 +1087,6 @@ acpi_boot_table_init(void)
 	 */
 	error = acpi_blacklisted();
 	if (error) {
-		extern int acpi_force;
-
 		if (acpi_force) {
 			printk(KERN_WARNING PREFIX "acpi=force override\n");
 		} else {
diff --git a/arch/i386/kernel/dmi_scan.c b/arch/i386/kernel/dmi_scan.c
index 3facd20212b..b4bae60f506 100644
--- a/arch/i386/kernel/dmi_scan.c
+++ b/arch/i386/kernel/dmi_scan.c
@@ -188,59 +188,6 @@ static __init int reset_videomode_after_s3(struct dmi_blacklist *d)
 #endif
 
 
-#ifdef	CONFIG_ACPI_BOOT
-extern int acpi_force;
-
-static __init __attribute__((unused)) int dmi_disable_acpi(struct dmi_blacklist *d) 
-{ 
-	if (!acpi_force) { 
-		printk(KERN_NOTICE "%s detected: acpi off\n",d->ident); 
-		disable_acpi();
-	} else { 
-		printk(KERN_NOTICE 
-		       "Warning: DMI blacklist says broken, but acpi forced\n"); 
-	}
-	return 0;
-} 
-
-/*
- * Limit ACPI to CPU enumeration for HT
- */
-static __init __attribute__((unused)) int force_acpi_ht(struct dmi_blacklist *d) 
-{ 
-	if (!acpi_force) { 
-		printk(KERN_NOTICE "%s detected: force use of acpi=ht\n", d->ident); 
-		disable_acpi();
-		acpi_ht = 1; 
-	} else { 
-		printk(KERN_NOTICE 
-		       "Warning: acpi=force overrules DMI blacklist: acpi=ht\n"); 
-	}
-	return 0;
-} 
-#endif
-
-#ifdef	CONFIG_ACPI_PCI
-static __init int disable_acpi_irq(struct dmi_blacklist *d) 
-{
-	if (!acpi_force) {
-		printk(KERN_NOTICE "%s detected: force use of acpi=noirq\n",
-		       d->ident); 	
-		acpi_noirq_set();
-	}
-	return 0;
-}
-static __init int disable_acpi_pci(struct dmi_blacklist *d) 
-{
-	if (!acpi_force) {
-		printk(KERN_NOTICE "%s detected: force use of pci=noacpi\n",
-		       d->ident); 	
-		acpi_disable_pci();
-	}
-	return 0;
-}  
-#endif
-
 /*
  *	Process the DMI blacklists
  */
@@ -264,116 +211,6 @@ static __initdata struct dmi_blacklist dmi_blacklist[]={
 			} },
 #endif
 
-#ifdef	CONFIG_ACPI_BOOT
-	/*
-	 * If your system is blacklisted here, but you find that acpi=force
-	 * works for you, please contact acpi-devel@sourceforge.net
-	 */
-
-	/*
-	 *	Boxes that need ACPI disabled
-	 */
-
-	{ dmi_disable_acpi, "IBM Thinkpad", {
-			MATCH(DMI_BOARD_VENDOR, "IBM"),
-			MATCH(DMI_BOARD_NAME, "2629H1G"),
-			NO_MATCH, NO_MATCH }},
-
-	/*
-	 *	Boxes that need acpi=ht 
-	 */
-
-	{ force_acpi_ht, "FSC Primergy T850", {
-			MATCH(DMI_SYS_VENDOR, "FUJITSU SIEMENS"),
-			MATCH(DMI_PRODUCT_NAME, "PRIMERGY T850"),
-			NO_MATCH, NO_MATCH }},
-
-	{ force_acpi_ht, "DELL GX240", {
-			MATCH(DMI_BOARD_VENDOR, "Dell Computer Corporation"),
-			MATCH(DMI_BOARD_NAME, "OptiPlex GX240"),
-			NO_MATCH, NO_MATCH }},
-
-	{ force_acpi_ht, "HP VISUALIZE NT Workstation", {
-			MATCH(DMI_BOARD_VENDOR, "Hewlett-Packard"),
-			MATCH(DMI_PRODUCT_NAME, "HP VISUALIZE NT Workstation"),
-			NO_MATCH, NO_MATCH }},
-
-	{ force_acpi_ht, "Compaq Workstation W8000", {
-			MATCH(DMI_SYS_VENDOR, "Compaq"),
-			MATCH(DMI_PRODUCT_NAME, "Workstation W8000"),
-			NO_MATCH, NO_MATCH }},
-
-	{ force_acpi_ht, "ASUS P4B266", {
-			MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
-			MATCH(DMI_BOARD_NAME, "P4B266"),
-			NO_MATCH, NO_MATCH }},
-
-	{ force_acpi_ht, "ASUS P2B-DS", {
-			MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
-			MATCH(DMI_BOARD_NAME, "P2B-DS"),
-			NO_MATCH, NO_MATCH }},
-
-	{ force_acpi_ht, "ASUS CUR-DLS", {
-			MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
-			MATCH(DMI_BOARD_NAME, "CUR-DLS"),
-			NO_MATCH, NO_MATCH }},
-
-	{ force_acpi_ht, "ABIT i440BX-W83977", {
-			MATCH(DMI_BOARD_VENDOR, "ABIT <http://www.abit.com>"),
-			MATCH(DMI_BOARD_NAME, "i440BX-W83977 (BP6)"),
-			NO_MATCH, NO_MATCH }},
-
-	{ force_acpi_ht, "IBM Bladecenter", {
-			MATCH(DMI_BOARD_VENDOR, "IBM"),
-			MATCH(DMI_BOARD_NAME, "IBM eServer BladeCenter HS20"),
-			NO_MATCH, NO_MATCH }},
-
-	{ force_acpi_ht, "IBM eServer xSeries 360", {
-			MATCH(DMI_BOARD_VENDOR, "IBM"),
-			MATCH(DMI_BOARD_NAME, "eServer xSeries 360"),
-			NO_MATCH, NO_MATCH }},
-
-	{ force_acpi_ht, "IBM eserver xSeries 330", {
-			MATCH(DMI_BOARD_VENDOR, "IBM"),
-			MATCH(DMI_BOARD_NAME, "eserver xSeries 330"),
-			NO_MATCH, NO_MATCH }},
-
-	{ force_acpi_ht, "IBM eserver xSeries 440", {
-			MATCH(DMI_BOARD_VENDOR, "IBM"),
-			MATCH(DMI_PRODUCT_NAME, "eserver xSeries 440"),
-			NO_MATCH, NO_MATCH }},
-
-#endif	// CONFIG_ACPI_BOOT
-
-#ifdef	CONFIG_ACPI_PCI
-	/*
-	 *	Boxes that need ACPI PCI IRQ routing disabled
-	 */
-
-	{ disable_acpi_irq, "ASUS A7V", {
-			MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC"),
-			MATCH(DMI_BOARD_NAME, "<A7V>"),
-			/* newer BIOS, Revision 1011, does work */
-			MATCH(DMI_BIOS_VERSION, "ASUS A7V ACPI BIOS Revision 1007"),
-			NO_MATCH }},
-
-	/*
-	 *	Boxes that need ACPI PCI IRQ routing and PCI scan disabled
-	 */
-	{ disable_acpi_pci, "ASUS PR-DLS", {	/* _BBN 0 bug */
-			MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
-			MATCH(DMI_BOARD_NAME, "PR-DLS"),
-			MATCH(DMI_BIOS_VERSION, "ASUS PR-DLS ACPI BIOS Revision 1010"),
-			MATCH(DMI_BIOS_DATE, "03/21/2003") }},
-
- 	{ disable_acpi_pci, "Acer TravelMate 36x Laptop", {
- 			MATCH(DMI_SYS_VENDOR, "Acer"),
- 			MATCH(DMI_PRODUCT_NAME, "TravelMate 360"),
- 			NO_MATCH, NO_MATCH
- 			} },
-
-#endif
-
 	{ NULL, }
 };
 
-- 
cgit v1.2.3-70-g09d2


From 0f8133a8db81ff824a4abbe5bb0f15bf034d31c3 Mon Sep 17 00:00:00 2001
From: Andrey Panin <pazke@donpac.ru>
Date: Sat, 25 Jun 2005 14:54:45 -0700
Subject: [PATCH] dmi: move ACPI sleep quirk

This patch moves ACPI sleep quirk out of dmi_scan.c

Signed-off-by: Andrey Panin <pazke@donpac.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/acpi/boot.c  |  1 +
 arch/i386/kernel/acpi/sleep.c | 27 +++++++++++++++++++++++++++
 arch/i386/kernel/dmi_scan.c   | 16 ----------------
 3 files changed, 28 insertions(+), 16 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index 0771596c549..9f63ae0f404 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -1024,6 +1024,7 @@ static struct dmi_system_id __initdata acpi_dmi_table[] = {
 		},
 	},
 #endif
+	{ }
 };
 
 #endif	/* __i386__ */
diff --git a/arch/i386/kernel/acpi/sleep.c b/arch/i386/kernel/acpi/sleep.c
index 28bb0514bb6..c1af93032ff 100644
--- a/arch/i386/kernel/acpi/sleep.c
+++ b/arch/i386/kernel/acpi/sleep.c
@@ -7,6 +7,7 @@
 
 #include <linux/acpi.h>
 #include <linux/bootmem.h>
+#include <linux/dmi.h>
 #include <asm/smp.h>
 #include <asm/tlbflush.h>
 
@@ -91,3 +92,29 @@ static int __init acpi_sleep_setup(char *str)
 
 
 __setup("acpi_sleep=", acpi_sleep_setup);
+
+
+static __init int reset_videomode_after_s3(struct dmi_system_id *d)
+{
+	acpi_video_flags |= 2;
+	return 0;
+}
+
+static __initdata struct dmi_system_id acpisleep_dmi_table[] = {
+	{	/* Reset video mode after returning from ACPI S3 sleep */
+		.callback = reset_videomode_after_s3,
+		.ident = "Toshiba Satellite 4030cdt",
+		.matches = {
+			DMI_MATCH(DMI_PRODUCT_NAME, "S4030CDT/4.3"),
+		},
+	},
+	{ }
+};
+
+static int __init acpisleep_dmi_init(void)
+{
+	dmi_check_system(acpisleep_dmi_table);
+	return 0;
+}
+
+core_initcall(acpisleep_dmi_init);
diff --git a/arch/i386/kernel/dmi_scan.c b/arch/i386/kernel/dmi_scan.c
index b4bae60f506..065c30a73c1 100644
--- a/arch/i386/kernel/dmi_scan.c
+++ b/arch/i386/kernel/dmi_scan.c
@@ -177,16 +177,6 @@ static __init int broken_toshiba_keyboard(struct dmi_blacklist *d)
 }
 
 
-#ifdef CONFIG_ACPI_SLEEP
-static __init int reset_videomode_after_s3(struct dmi_blacklist *d)
-{
-	/* See acpi_wakeup.S */
-	extern long acpi_video_flags;
-	acpi_video_flags |= 2;
-	return 0;
-}
-#endif
-
 
 /*
  *	Process the DMI blacklists
@@ -204,12 +194,6 @@ static __initdata struct dmi_blacklist dmi_blacklist[]={
 			MATCH(DMI_PRODUCT_NAME, "S4030CDT/4.3"),
 			NO_MATCH, NO_MATCH, NO_MATCH
 			} },
-#ifdef CONFIG_ACPI_SLEEP
-	{ reset_videomode_after_s3, "Toshiba Satellite 4030cdt", { /* Reset video mode after returning from ACPI S3 sleep */
-			MATCH(DMI_PRODUCT_NAME, "S4030CDT/4.3"),
-			NO_MATCH, NO_MATCH, NO_MATCH
-			} },
-#endif
 
 	{ NULL, }
 };
-- 
cgit v1.2.3-70-g09d2


From b625883f24d018c989173aeb727f6de954fb154d Mon Sep 17 00:00:00 2001
From: Andrey Panin <pazke@donpac.ru>
Date: Sat, 25 Jun 2005 14:54:46 -0700
Subject: [PATCH] dmi: remove central blacklist

Since last dmi quirk looks useless (it just prints 404 compliant url) we can
finally remove central dmi blacklist.

Signed-off-by: Andrey Panin <pazke@donpac.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/dmi_scan.c | 45 +--------------------------------------------
 1 file changed, 1 insertion(+), 44 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/dmi_scan.c b/arch/i386/kernel/dmi_scan.c
index 065c30a73c1..ad4b369a583 100644
--- a/arch/i386/kernel/dmi_scan.c
+++ b/arch/i386/kernel/dmi_scan.c
@@ -159,51 +159,11 @@ static void __init dmi_save_ident(struct dmi_header *dm, int slot, int string)
 		printk(KERN_ERR "dmi_save_ident: out of memory.\n");
 }
 
-/*
- * Ugly compatibility crap.
- */
-#define dmi_blacklist	dmi_system_id
-#define NO_MATCH	{ DMI_NONE, NULL}
-#define MATCH		DMI_MATCH
-
-/*
- * Toshiba keyboard likes to repeat keys when they are not repeated.
- */
-
-static __init int broken_toshiba_keyboard(struct dmi_blacklist *d)
-{
-	printk(KERN_WARNING "Toshiba with broken keyboard detected. If your keyboard sometimes generates 3 keypresses instead of one, see http://davyd.ucc.asn.au/projects/toshiba/README\n");
-	return 0;
-}
-
-
-
-/*
- *	Process the DMI blacklists
- */
- 
-
-/*
- *	This will be expanded over time to force things like the APM 
- *	interrupt mask settings according to the laptop
- */
- 
-static __initdata struct dmi_blacklist dmi_blacklist[]={
-
-	{ broken_toshiba_keyboard, "Toshiba Satellite 4030cdt", { /* Keyboard generates spurious repeats */
-			MATCH(DMI_PRODUCT_NAME, "S4030CDT/4.3"),
-			NO_MATCH, NO_MATCH, NO_MATCH
-			} },
-
-	{ NULL, }
-};
-
 /*
  *	Process a DMI table entry. Right now all we care about are the BIOS
  *	and machine entries. For 2.5 we should pull the smbus controller info
  *	out of here.
  */
-
 static void __init dmi_decode(struct dmi_header *dm)
 {
 #ifdef DMI_DEBUG
@@ -253,10 +213,7 @@ static void __init dmi_decode(struct dmi_header *dm)
 
 void __init dmi_scan_machine(void)
 {
-	int err = dmi_iterate(dmi_decode);
-	if(err == 0)
- 		dmi_check_system(dmi_blacklist);
-	else
+	if (dmi_iterate(dmi_decode))
 		printk(KERN_INFO "DMI not present.\n");
 }
 
-- 
cgit v1.2.3-70-g09d2


From 1249c5138f573890eae0c01f13d627094edcd55c Mon Sep 17 00:00:00 2001
From: Andrey Panin <pazke@donpac.ru>
Date: Sat, 25 Jun 2005 14:54:47 -0700
Subject: [PATCH] dmi: spring cleanup

Whitespace and CodingStyle cleanup. No functionality changes.

Signed-off-by: Andrey Panin <pazke@donpac.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/dmi_scan.c | 163 +++++++++++++++++++-------------------------
 1 file changed, 70 insertions(+), 93 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/dmi_scan.c b/arch/i386/kernel/dmi_scan.c
index ad4b369a583..a3cdf894302 100644
--- a/arch/i386/kernel/dmi_scan.c
+++ b/arch/i386/kernel/dmi_scan.c
@@ -1,22 +1,15 @@
 #include <linux/types.h>
-#include <linux/kernel.h>
 #include <linux/string.h>
 #include <linux/init.h>
 #include <linux/module.h>
-#include <linux/slab.h>
-#include <linux/acpi.h>
-#include <asm/io.h>
-#include <linux/pm.h>
-#include <asm/system.h>
 #include <linux/dmi.h>
 #include <linux/bootmem.h>
 
 
-struct dmi_header
-{
-	u8	type;
-	u8	length;
-	u16	handle;
+struct dmi_header {
+	u8 type;
+	u8 length;
+	u16 handle;
 };
 
 #undef DMI_DEBUG
@@ -29,15 +22,13 @@ struct dmi_header
 
 static char * __init dmi_string(struct dmi_header *dm, u8 s)
 {
-	u8 *bp=(u8 *)dm;
-	bp+=dm->length;
-	if(!s)
+	u8 *bp = ((u8 *) dm) + dm->length;
+
+	if (!s)
 		return "";
 	s--;
-	while(s>0 && *bp)
-	{
-		bp+=strlen(bp);
-		bp++;
+	while (s > 0 && *bp) {
+		bp += strlen(bp) + 1;
 		s--;
 	}
 	return bp;
@@ -47,16 +38,14 @@ static char * __init dmi_string(struct dmi_header *dm, u8 s)
  *	We have to be cautious here. We have seen BIOSes with DMI pointers
  *	pointing to completely the wrong place for example
  */
- 
-static int __init dmi_table(u32 base, int len, int num, void (*decode)(struct dmi_header *))
+static int __init dmi_table(u32 base, int len, int num,
+			    void (*decode)(struct dmi_header *))
 {
-	u8 *buf;
-	struct dmi_header *dm;
-	u8 *data;
-	int i=0;
+	u8 *buf, *data;
+	int i = 0;
 		
 	buf = bt_ioremap(base, len);
-	if(buf==NULL)
+	if (buf == NULL)
 		return -1;
 
 	data = buf;
@@ -65,36 +54,34 @@ static int __init dmi_table(u32 base, int len, int num, void (*decode)(struct dm
  	 *	Stop when we see all the items the table claimed to have
  	 *	OR we run off the end of the table (also happens)
  	 */
- 
-	while(i<num && data-buf+sizeof(struct dmi_header)<=len)
-	{
-		dm=(struct dmi_header *)data;
+	while ((i < num) && (data - buf + sizeof(struct dmi_header)) <= len) {
+		struct dmi_header *dm = (struct dmi_header *)data;
 		/*
 		 *  We want to know the total length (formated area and strings)
 		 *  before decoding to make sure we won't run off the table in
 		 *  dmi_decode or dmi_string
 		 */
-		data+=dm->length;
-		while(data-buf<len-1 && (data[0] || data[1]))
+		data += dm->length;
+		while ((data - buf < len - 1) && (data[0] || data[1]))
 			data++;
-		if(data-buf<len-1)
+		if (data - buf < len - 1)
 			decode(dm);
-		data+=2;
+		data += 2;
 		i++;
 	}
 	bt_iounmap(buf, len);
 	return 0;
 }
 
-
-inline static int __init dmi_checksum(u8 *buf)
+static int __init dmi_checksum(u8 *buf)
 {
-	u8 sum=0;
+	u8 sum = 0;
 	int a;
 	
-	for(a=0; a<15; a++)
-		sum+=buf[a];
-	return (sum==0);
+	for (a = 0; a < 15; a++)
+		sum += buf[a];
+
+	return sum == 0;
 }
 
 static int __init dmi_iterate(void (*decode)(struct dmi_header *))
@@ -110,28 +97,30 @@ static int __init dmi_iterate(void (*decode)(struct dmi_header *))
 	p = ioremap(0xF0000, 0x10000);
 	if (p == NULL)
 		return -1;
+
 	for (q = p; q < p + 0x10000; q += 16) {
 		memcpy_fromio(buf, q, 15);
-		if(memcmp(buf, "_DMI_", 5)==0 && dmi_checksum(buf))
-		{
-			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];
+		if ((memcmp(buf, "_DMI_", 5) == 0) && dmi_checksum(buf)) {
+			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];
 
 			/*
 			 * DMI version 0.0 means that the real version is taken from
 			 * the SMBIOS version, which we don't know at this point.
 			 */
-			if(buf[14]!=0)
+			if (buf[14] != 0)
 				printk(KERN_INFO "DMI %d.%d present.\n",
-					buf[14]>>4, buf[14]&0x0F);
+					buf[14] >> 4, buf[14] & 0xF);
 			else
 				printk(KERN_INFO "DMI present.\n");
+
 			dmi_printk((KERN_INFO "%d structures occupying %d bytes.\n",
 				num, len));
-			dmi_printk((KERN_INFO "DMI table at 0x%08X.\n",
-				base));
-			if(dmi_table(base,len, num, decode)==0)
+			dmi_printk((KERN_INFO "DMI table at 0x%08X.\n", base));
+
+			if (dmi_table(base,len, num, decode) == 0)
 				return 0;
 		}
 	}
@@ -143,16 +132,17 @@ static char *dmi_ident[DMI_STRING_MAX];
 /*
  *	Save a DMI string
  */
- 
 static void __init dmi_save_ident(struct dmi_header *dm, int slot, int string)
 {
 	char *d = (char*)dm;
 	char *p = dmi_string(dm, d[string]);
-	if(p==NULL || *p == 0)
+
+	if (p == NULL || *p == 0)
 		return;
 	if (dmi_ident[slot])
 		return;
-	dmi_ident[slot] = alloc_bootmem(strlen(p)+1);
+
+	dmi_ident[slot] = alloc_bootmem(strlen(p) + 1);
 	if(dmi_ident[slot])
 		strcpy(dmi_ident[slot], p);
 	else
@@ -166,48 +156,35 @@ static void __init dmi_save_ident(struct dmi_header *dm, int slot, int string)
  */
 static void __init dmi_decode(struct dmi_header *dm)
 {
-#ifdef DMI_DEBUG
-	u8 *data = (u8 *)dm;
-#endif
+	u8 *data __attribute__((__unused__)) = (u8 *)dm;
 	
-	switch(dm->type)
-	{
-		case  0:
-			dmi_printk(("BIOS Vendor: %s\n",
-				dmi_string(dm, data[4])));
-			dmi_save_ident(dm, DMI_BIOS_VENDOR, 4);
-			dmi_printk(("BIOS Version: %s\n", 
-				dmi_string(dm, data[5])));
-			dmi_save_ident(dm, DMI_BIOS_VERSION, 5);
-			dmi_printk(("BIOS Release: %s\n",
-				dmi_string(dm, data[8])));
-			dmi_save_ident(dm, DMI_BIOS_DATE, 8);
-			break;
-		case 1:
-			dmi_printk(("System Vendor: %s\n",
-				dmi_string(dm, data[4])));
-			dmi_save_ident(dm, DMI_SYS_VENDOR, 4);
-			dmi_printk(("Product Name: %s\n",
-				dmi_string(dm, data[5])));
-			dmi_save_ident(dm, DMI_PRODUCT_NAME, 5);
-			dmi_printk(("Version: %s\n",
-				dmi_string(dm, data[6])));
-			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",
-				dmi_string(dm, data[4])));
-			dmi_save_ident(dm, DMI_BOARD_VENDOR, 4);
-			dmi_printk(("Board Name: %s\n",
-				dmi_string(dm, data[5])));
-			dmi_save_ident(dm, DMI_BOARD_NAME, 5);
-			dmi_printk(("Board Version: %s\n",
-				dmi_string(dm, data[6])));
-			dmi_save_ident(dm, DMI_BOARD_VERSION, 6);
-			break;
+	switch(dm->type) {
+	case  0:
+		dmi_printk(("BIOS Vendor: %s\n", dmi_string(dm, data[4])));
+		dmi_save_ident(dm, DMI_BIOS_VENDOR, 4);
+		dmi_printk(("BIOS Version: %s\n", dmi_string(dm, data[5])));
+		dmi_save_ident(dm, DMI_BIOS_VERSION, 5);
+		dmi_printk(("BIOS Release: %s\n", dmi_string(dm, data[8])));
+		dmi_save_ident(dm, DMI_BIOS_DATE, 8);
+		break;
+	case 1:
+		dmi_printk(("System Vendor: %s\n", dmi_string(dm, data[4])));
+		dmi_save_ident(dm, DMI_SYS_VENDOR, 4);
+		dmi_printk(("Product Name: %s\n", dmi_string(dm, data[5])));
+		dmi_save_ident(dm, DMI_PRODUCT_NAME, 5);
+		dmi_printk(("Version: %s\n", dmi_string(dm, data[6])));
+		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", dmi_string(dm, data[4])));
+		dmi_save_ident(dm, DMI_BOARD_VENDOR, 4);
+		dmi_printk(("Board Name: %s\n", dmi_string(dm, data[5])));
+		dmi_save_ident(dm, DMI_BOARD_NAME, 5);
+		dmi_printk(("Board Version: %s\n", dmi_string(dm, data[6])));
+		dmi_save_ident(dm, DMI_BOARD_VERSION, 6);
+		break;
 	}
 }
 
-- 
cgit v1.2.3-70-g09d2


From a13db56624c2a9d6c0dae0a693b25b0e58de9ea3 Mon Sep 17 00:00:00 2001
From: Shaohua Li <shaohua.li@intel.com>
Date: Sat, 25 Jun 2005 14:54:48 -0700
Subject: [PATCH] CPU hotplug: fix hpet sectioning

With hpet enabled, cpu hotplug uses some routines marked with __init.

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/kernel/time_hpet.c     | 2 +-
 arch/i386/kernel/timers/common.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/time_hpet.c b/arch/i386/kernel/time_hpet.c
index 10a0cbb88e7..658c0629ba6 100644
--- a/arch/i386/kernel/time_hpet.c
+++ b/arch/i386/kernel/time_hpet.c
@@ -50,7 +50,7 @@ static void hpet_writel(unsigned long d, unsigned long a)
  * comparator value and continue. Next tick can be caught by checking
  * for a change in the comparator value. Used in apic.c.
  */
-static void __init wait_hpet_tick(void)
+static void __devinit wait_hpet_tick(void)
 {
 	unsigned int start_cmp_val, end_cmp_val;
 
diff --git a/arch/i386/kernel/timers/common.c b/arch/i386/kernel/timers/common.c
index 37353bd3180..8163fe0cf1f 100644
--- a/arch/i386/kernel/timers/common.c
+++ b/arch/i386/kernel/timers/common.c
@@ -86,7 +86,7 @@ bad_ctc:
 #define CALIBRATE_CNT_HPET 	(5 * hpet_tick)
 #define CALIBRATE_TIME_HPET 	(5 * KERNEL_TICK_USEC)
 
-unsigned long __init calibrate_tsc_hpet(unsigned long *tsc_hpet_quotient_ptr)
+unsigned long __devinit calibrate_tsc_hpet(unsigned long *tsc_hpet_quotient_ptr)
 {
 	unsigned long tsc_startlow, tsc_starthigh;
 	unsigned long tsc_endlow, tsc_endhigh;
-- 
cgit v1.2.3-70-g09d2


From d92de65cab5980c16d4a1c326c1ef9a591892883 Mon Sep 17 00:00:00 2001
From: Shaohua Li <shaohua.li@intel.com>
Date: Sat, 25 Jun 2005 14:54:49 -0700
Subject: [PATCH] variable overflow after hundreds round of hotplug CPU

I'm doing the cpu hotplug stress test and found a variable ('ready') is
overflow after several hundreds rounds of cpu hotplug.  Here is a fix.

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/kernel/head.S | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S
index e966fc8c44c..4477bb10709 100644
--- a/arch/i386/kernel/head.S
+++ b/arch/i386/kernel/head.S
@@ -299,7 +299,6 @@ is386:	movl $2,%ecx		# set MP
 	movl %eax,%cr0
 
 	call check_x87
-	incb ready
 	lgdt cpu_gdt_descr
 	lidt idt_descr
 	ljmp $(__KERNEL_CS),$1f
@@ -316,8 +315,9 @@ is386:	movl $2,%ecx		# set MP
 	lldt %ax
 	cld			# gcc2 wants the direction flag cleared at all times
 #ifdef CONFIG_SMP
-	movb ready, %cl	
-	cmpb $1,%cl
+	movb ready, %cl
+	movb $1, ready
+	cmpb $0,%cl
 	je 1f			# the first CPU calls start_kernel
 				# all other CPUs call initialize_secondary
 	call initialize_secondary
-- 
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 'arch')

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 'arch')

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 'arch')

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 d720803a9365d360b3e5ea02033f0c11b5b1226a Mon Sep 17 00:00:00 2001
From: Li Shaohua <shaohua.li@intel.com>
Date: Sat, 25 Jun 2005 14:54:54 -0700
Subject: [PATCH] sibling map initializing rework

Make sibling map init per-cpu.  Hotplug CPU may change the map at runtime.

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/smpboot.c | 97 ++++++++++++++++++++++++----------------------
 1 file changed, 50 insertions(+), 47 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index c5517f33230..09b4ceb832b 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -66,11 +66,21 @@ int smp_num_siblings = 1;
 #ifdef CONFIG_X86_HT
 EXPORT_SYMBOL(smp_num_siblings);
 #endif
-int phys_proc_id[NR_CPUS]; /* Package ID of each logical CPU */
+
+/* Package ID of each logical CPU */
+int phys_proc_id[NR_CPUS] = {[0 ... NR_CPUS-1] = BAD_APICID};
 EXPORT_SYMBOL(phys_proc_id);
-int cpu_core_id[NR_CPUS]; /* Core ID of each logical CPU */
+
+/* Core ID of each logical CPU */
+int cpu_core_id[NR_CPUS] = {[0 ... NR_CPUS-1] = BAD_APICID};
 EXPORT_SYMBOL(cpu_core_id);
 
+cpumask_t cpu_sibling_map[NR_CPUS];
+EXPORT_SYMBOL(cpu_sibling_map);
+
+cpumask_t cpu_core_map[NR_CPUS];
+EXPORT_SYMBOL(cpu_core_map);
+
 /* bitmap of online cpus */
 cpumask_t cpu_online_map;
 EXPORT_SYMBOL(cpu_online_map);
@@ -423,6 +433,38 @@ static void __init smp_callin(void)
 
 static int cpucount;
 
+static inline void
+set_cpu_sibling_map(int cpu)
+{
+	int i;
+
+	if (smp_num_siblings > 1) {
+		for (i = 0; i < NR_CPUS; i++) {
+			if (!cpu_isset(i, cpu_callout_map))
+				continue;
+			if (cpu_core_id[cpu] == cpu_core_id[i]) {
+				cpu_set(i, cpu_sibling_map[cpu]);
+				cpu_set(cpu, cpu_sibling_map[i]);
+			}
+		}
+	} else {
+		cpu_set(cpu, cpu_sibling_map[cpu]);
+	}
+
+	if (current_cpu_data.x86_num_cores > 1) {
+		for (i = 0; i < NR_CPUS; i++) {
+			if (!cpu_isset(i, cpu_callout_map))
+				continue;
+			if (phys_proc_id[cpu] == phys_proc_id[i]) {
+				cpu_set(i, cpu_core_map[cpu]);
+				cpu_set(cpu, cpu_core_map[i]);
+			}
+		}
+	} else {
+		cpu_core_map[cpu] = cpu_sibling_map[cpu];
+	}
+}
+
 /*
  * Activate a secondary processor.
  */
@@ -450,6 +492,10 @@ static void __init start_secondary(void *unused)
 	 */
 	local_flush_tlb();
 
+	/* This must be done before setting cpu_online_map */
+	set_cpu_sibling_map(raw_smp_processor_id());
+	wmb();
+
 	/*
 	 * We need to hold call_lock, so there is no inconsistency
 	 * between the time smp_call_function() determines number of
@@ -912,13 +958,6 @@ void *xquad_portio;
 EXPORT_SYMBOL(xquad_portio);
 #endif
 
-cpumask_t cpu_sibling_map[NR_CPUS] __cacheline_aligned;
-#ifdef CONFIG_X86_HT
-EXPORT_SYMBOL(cpu_sibling_map);
-#endif
-cpumask_t cpu_core_map[NR_CPUS] __cacheline_aligned;
-EXPORT_SYMBOL(cpu_core_map);
-
 static void __init smp_boot_cpus(unsigned int max_cpus)
 {
 	int apicid, cpu, bit, kicked;
@@ -1082,44 +1121,8 @@ static void __init smp_boot_cpus(unsigned int max_cpus)
 		cpus_clear(cpu_core_map[cpu]);
 	}
 
-	for (cpu = 0; cpu < NR_CPUS; cpu++) {
-		struct cpuinfo_x86 *c = cpu_data + cpu;
-		int siblings = 0;
-		int i;
-		if (!cpu_isset(cpu, cpu_callout_map))
-			continue;
-
-		if (smp_num_siblings > 1) {
-			for (i = 0; i < NR_CPUS; i++) {
-				if (!cpu_isset(i, cpu_callout_map))
-					continue;
-				if (cpu_core_id[cpu] == cpu_core_id[i]) {
-					siblings++;
-					cpu_set(i, cpu_sibling_map[cpu]);
-				}
-			}
-		} else {
-			siblings++;
-			cpu_set(cpu, cpu_sibling_map[cpu]);
-		}
-
-		if (siblings != smp_num_siblings) {
-			printk(KERN_WARNING "WARNING: %d siblings found for CPU%d, should be %d\n", siblings, cpu, smp_num_siblings);
-			smp_num_siblings = siblings;
-		}
-
-		if (c->x86_num_cores > 1) {
-			for (i = 0; i < NR_CPUS; i++) {
-				if (!cpu_isset(i, cpu_callout_map))
-					continue;
-				if (phys_proc_id[cpu] == phys_proc_id[i]) {
-					cpu_set(i, cpu_core_map[cpu]);
-				}
-			}
-		} else {
-			cpu_core_map[cpu] = cpu_sibling_map[cpu];
-		}
-	}
+	cpu_set(0, cpu_sibling_map[0]);
+	cpu_set(0, cpu_core_map[0]);
 
 	smpboot_setup_io_apic();
 
-- 
cgit v1.2.3-70-g09d2


From 0bb3184df537002a742bafddf3f4fb482b7fe610 Mon Sep 17 00:00:00 2001
From: Li Shaohua <shaohua.li@intel.com>
Date: Sat, 25 Jun 2005 14:54:55 -0700
Subject: [PATCH] init call cleanup

Trival patch for CPU hotplug.  In CPU identify part, only did cleaup for intel
CPUs.  Need do for other CPUs if they support S3 SMP.

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/apic.c                | 14 +++++++-------
 arch/i386/kernel/cpu/common.c          | 30 +++++++++++++++---------------
 arch/i386/kernel/cpu/intel.c           | 12 ++++++------
 arch/i386/kernel/cpu/intel_cacheinfo.c |  4 ++--
 arch/i386/kernel/cpu/mcheck/mce.c      |  2 +-
 arch/i386/kernel/cpu/mcheck/p5.c       |  2 +-
 arch/i386/kernel/process.c             |  2 +-
 arch/i386/kernel/setup.c               |  2 +-
 arch/i386/kernel/smpboot.c             | 18 +++++++++---------
 arch/i386/kernel/timers/timer_tsc.c    |  2 +-
 10 files changed, 44 insertions(+), 44 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index a28a088f3e7..b905d7bb9a0 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -364,7 +364,7 @@ void __init init_bsp_APIC(void)
 	apic_write_around(APIC_LVT1, value);
 }
 
-void __init setup_local_APIC (void)
+void __devinit setup_local_APIC(void)
 {
 	unsigned long oldvalue, value, ver, maxlvt;
 
@@ -635,7 +635,7 @@ static struct sys_device device_lapic = {
 	.cls	= &lapic_sysclass,
 };
 
-static void __init apic_pm_activate(void)
+static void __devinit apic_pm_activate(void)
 {
 	apic_pm_state.active = 1;
 }
@@ -856,7 +856,7 @@ fake_ioapic_page:
  * but we do not accept timer interrupts yet. We only allow the BP
  * to calibrate.
  */
-static unsigned int __init get_8254_timer_count(void)
+static unsigned int __devinit get_8254_timer_count(void)
 {
 	extern spinlock_t i8253_lock;
 	unsigned long flags;
@@ -875,7 +875,7 @@ static unsigned int __init get_8254_timer_count(void)
 }
 
 /* next tick in 8254 can be caught by catching timer wraparound */
-static void __init wait_8254_wraparound(void)
+static void __devinit wait_8254_wraparound(void)
 {
 	unsigned int curr_count, prev_count;
 
@@ -895,7 +895,7 @@ static void __init wait_8254_wraparound(void)
  * Default initialization for 8254 timers. If we use other timers like HPET,
  * we override this later
  */
-void (*wait_timer_tick)(void) __initdata = wait_8254_wraparound;
+void (*wait_timer_tick)(void) __devinitdata = wait_8254_wraparound;
 
 /*
  * This function sets up the local APIC timer, with a timeout of
@@ -931,7 +931,7 @@ static void __setup_APIC_LVTT(unsigned int clocks)
 	apic_write_around(APIC_TMICT, clocks/APIC_DIVISOR);
 }
 
-static void __init setup_APIC_timer(unsigned int clocks)
+static void __devinit setup_APIC_timer(unsigned int clocks)
 {
 	unsigned long flags;
 
@@ -1044,7 +1044,7 @@ void __init setup_boot_APIC_clock(void)
 	local_irq_enable();
 }
 
-void __init setup_secondary_APIC_clock(void)
+void __devinit setup_secondary_APIC_clock(void)
 {
 	setup_APIC_timer(calibration_result);
 }
diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index d58e169fbdb..aac74758caf 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -24,9 +24,9 @@ EXPORT_PER_CPU_SYMBOL(cpu_gdt_table);
 DEFINE_PER_CPU(unsigned char, cpu_16bit_stack[CPU_16BIT_STACK_SIZE]);
 EXPORT_PER_CPU_SYMBOL(cpu_16bit_stack);
 
-static int cachesize_override __initdata = -1;
-static int disable_x86_fxsr __initdata = 0;
-static int disable_x86_serial_nr __initdata = 1;
+static int cachesize_override __devinitdata = -1;
+static int disable_x86_fxsr __devinitdata = 0;
+static int disable_x86_serial_nr __devinitdata = 1;
 
 struct cpu_dev * cpu_devs[X86_VENDOR_NUM] = {};
 
@@ -59,7 +59,7 @@ static int __init cachesize_setup(char *str)
 }
 __setup("cachesize=", cachesize_setup);
 
-int __init get_model_name(struct cpuinfo_x86 *c)
+int __devinit get_model_name(struct cpuinfo_x86 *c)
 {
 	unsigned int *v;
 	char *p, *q;
@@ -89,7 +89,7 @@ int __init get_model_name(struct cpuinfo_x86 *c)
 }
 
 
-void __init display_cacheinfo(struct cpuinfo_x86 *c)
+void __devinit display_cacheinfo(struct cpuinfo_x86 *c)
 {
 	unsigned int n, dummy, ecx, edx, l2size;
 
@@ -130,7 +130,7 @@ void __init display_cacheinfo(struct cpuinfo_x86 *c)
 /* in particular, if CPUID levels 0x80000002..4 are supported, this isn't used */
 
 /* Look up CPU names by table lookup. */
-static char __init *table_lookup_model(struct cpuinfo_x86 *c)
+static char __devinit *table_lookup_model(struct cpuinfo_x86 *c)
 {
 	struct cpu_model_info *info;
 
@@ -151,7 +151,7 @@ static char __init *table_lookup_model(struct cpuinfo_x86 *c)
 }
 
 
-void __init get_cpu_vendor(struct cpuinfo_x86 *c, int early)
+void __devinit get_cpu_vendor(struct cpuinfo_x86 *c, int early)
 {
 	char *v = c->x86_vendor_id;
 	int i;
@@ -202,7 +202,7 @@ static inline int flag_is_changeable_p(u32 flag)
 
 
 /* Probe for the CPUID instruction */
-static int __init have_cpuid_p(void)
+static int __devinit have_cpuid_p(void)
 {
 	return flag_is_changeable_p(X86_EFLAGS_ID);
 }
@@ -249,7 +249,7 @@ static void __init early_cpu_detect(void)
 #endif
 }
 
-void __init generic_identify(struct cpuinfo_x86 * c)
+void __devinit generic_identify(struct cpuinfo_x86 * c)
 {
 	u32 tfms, xlvl;
 	int junk;
@@ -296,7 +296,7 @@ void __init generic_identify(struct cpuinfo_x86 * c)
 	}
 }
 
-static void __init squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
+static void __devinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
 {
 	if (cpu_has(c, X86_FEATURE_PN) && disable_x86_serial_nr ) {
 		/* Disable processor serial number */
@@ -324,7 +324,7 @@ __setup("serialnumber", x86_serial_nr_setup);
 /*
  * This does the hard work of actually picking apart the CPU stuff...
  */
-void __init identify_cpu(struct cpuinfo_x86 *c)
+void __devinit identify_cpu(struct cpuinfo_x86 *c)
 {
 	int i;
 
@@ -438,7 +438,7 @@ void __init identify_cpu(struct cpuinfo_x86 *c)
 }
 
 #ifdef CONFIG_X86_HT
-void __init detect_ht(struct cpuinfo_x86 *c)
+void __devinit detect_ht(struct cpuinfo_x86 *c)
 {
 	u32 	eax, ebx, ecx, edx;
 	int 	index_msb, tmp;
@@ -493,7 +493,7 @@ void __init detect_ht(struct cpuinfo_x86 *c)
 }
 #endif
 
-void __init print_cpu_info(struct cpuinfo_x86 *c)
+void __devinit print_cpu_info(struct cpuinfo_x86 *c)
 {
 	char *vendor = NULL;
 
@@ -516,7 +516,7 @@ void __init print_cpu_info(struct cpuinfo_x86 *c)
 		printk("\n");
 }
 
-cpumask_t cpu_initialized __initdata = CPU_MASK_NONE;
+cpumask_t cpu_initialized __devinitdata = CPU_MASK_NONE;
 
 /* This is hacky. :)
  * We're emulating future behavior.
@@ -563,7 +563,7 @@ void __init early_cpu_init(void)
  * and IDT. We reload them nevertheless, this function acts as a
  * 'CPU state barrier', nothing should get across.
  */
-void __init cpu_init (void)
+void __devinit cpu_init(void)
 {
 	int cpu = smp_processor_id();
 	struct tss_struct * t = &per_cpu(init_tss, cpu);
diff --git a/arch/i386/kernel/cpu/intel.c b/arch/i386/kernel/cpu/intel.c
index 121aa2176e6..96a75d04583 100644
--- a/arch/i386/kernel/cpu/intel.c
+++ b/arch/i386/kernel/cpu/intel.c
@@ -28,7 +28,7 @@ extern int trap_init_f00f_bug(void);
 struct movsl_mask movsl_mask;
 #endif
 
-void __init early_intel_workaround(struct cpuinfo_x86 *c)
+void __devinit early_intel_workaround(struct cpuinfo_x86 *c)
 {
 	if (c->x86_vendor != X86_VENDOR_INTEL)
 		return;
@@ -43,7 +43,7 @@ void __init early_intel_workaround(struct cpuinfo_x86 *c)
  *	This is called before we do cpu ident work
  */
  
-int __init ppro_with_ram_bug(void)
+int __devinit ppro_with_ram_bug(void)
 {
 	/* Uses data from early_cpu_detect now */
 	if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL &&
@@ -61,7 +61,7 @@ int __init ppro_with_ram_bug(void)
  * P4 Xeon errata 037 workaround.
  * Hardware prefetcher may cause stale data to be loaded into the cache.
  */
-static void __init Intel_errata_workarounds(struct cpuinfo_x86 *c)
+static void __devinit Intel_errata_workarounds(struct cpuinfo_x86 *c)
 {
 	unsigned long lo, hi;
 
@@ -80,7 +80,7 @@ static void __init Intel_errata_workarounds(struct cpuinfo_x86 *c)
 /*
  * find out the number of processor cores on the die
  */
-static int __init num_cpu_cores(struct cpuinfo_x86 *c)
+static int __devinit num_cpu_cores(struct cpuinfo_x86 *c)
 {
 	unsigned int eax;
 
@@ -98,7 +98,7 @@ static int __init num_cpu_cores(struct cpuinfo_x86 *c)
 		return 1;
 }
 
-static void __init init_intel(struct cpuinfo_x86 *c)
+static void __devinit init_intel(struct cpuinfo_x86 *c)
 {
 	unsigned int l2 = 0;
 	char *p = NULL;
@@ -204,7 +204,7 @@ static unsigned int intel_size_cache(struct cpuinfo_x86 * c, unsigned int size)
 	return size;
 }
 
-static struct cpu_dev intel_cpu_dev __initdata = {
+static struct cpu_dev intel_cpu_dev __devinitdata = {
 	.c_vendor	= "Intel",
 	.c_ident 	= { "GenuineIntel" },
 	.c_models = {
diff --git a/arch/i386/kernel/cpu/intel_cacheinfo.c b/arch/i386/kernel/cpu/intel_cacheinfo.c
index a710dc4eb20..1d768b26326 100644
--- a/arch/i386/kernel/cpu/intel_cacheinfo.c
+++ b/arch/i386/kernel/cpu/intel_cacheinfo.c
@@ -28,7 +28,7 @@ struct _cache_table
 };
 
 /* all the cache descriptor types we care about (no TLB or trace cache entries) */
-static struct _cache_table cache_table[] __initdata =
+static struct _cache_table cache_table[] __devinitdata =
 {
 	{ 0x06, LVL_1_INST, 8 },	/* 4-way set assoc, 32 byte line size */
 	{ 0x08, LVL_1_INST, 16 },	/* 4-way set assoc, 32 byte line size */
@@ -160,7 +160,7 @@ static int __init find_num_cache_leaves(void)
 	return retval;
 }
 
-unsigned int __init init_intel_cacheinfo(struct cpuinfo_x86 *c)
+unsigned int __devinit init_intel_cacheinfo(struct cpuinfo_x86 *c)
 {
 	unsigned int trace = 0, l1i = 0, l1d = 0, l2 = 0, l3 = 0; /* Cache sizes */
 	unsigned int new_l1d = 0, new_l1i = 0; /* Cache sizes from cpuid(4) */
diff --git a/arch/i386/kernel/cpu/mcheck/mce.c b/arch/i386/kernel/cpu/mcheck/mce.c
index bf6d1aefafc..7218a7341fb 100644
--- a/arch/i386/kernel/cpu/mcheck/mce.c
+++ b/arch/i386/kernel/cpu/mcheck/mce.c
@@ -31,7 +31,7 @@ static fastcall void unexpected_machine_check(struct pt_regs * regs, long error_
 void fastcall (*machine_check_vector)(struct pt_regs *, long error_code) = unexpected_machine_check;
 
 /* This has to be run for each processor */
-void __init mcheck_init(struct cpuinfo_x86 *c)
+void __devinit mcheck_init(struct cpuinfo_x86 *c)
 {
 	if (mce_disabled==1)
 		return;
diff --git a/arch/i386/kernel/cpu/mcheck/p5.c b/arch/i386/kernel/cpu/mcheck/p5.c
index c45a1b485c8..ec0614cd292 100644
--- a/arch/i386/kernel/cpu/mcheck/p5.c
+++ b/arch/i386/kernel/cpu/mcheck/p5.c
@@ -29,7 +29,7 @@ static fastcall void pentium_machine_check(struct pt_regs * regs, long error_cod
 }
 
 /* Set up machine check reporting for processors with Intel style MCE */
-void __init intel_p5_mcheck_init(struct cpuinfo_x86 *c)
+void __devinit intel_p5_mcheck_init(struct cpuinfo_x86 *c)
 {
 	u32 l, h;
 	
diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index c1b11e8df60..e06f2dc7123 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -260,7 +260,7 @@ static void mwait_idle(void)
 	}
 }
 
-void __init select_idle_routine(const struct cpuinfo_x86 *c)
+void __devinit select_idle_routine(const struct cpuinfo_x86 *c)
 {
 	if (cpu_has(c, X86_FEATURE_MWAIT)) {
 		printk("monitor/mwait feature present.\n");
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index 30406fd0b64..cba67e4ba0a 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -60,7 +60,7 @@
    address, and must not be in the .bss segment! */
 unsigned long init_pg_tables_end __initdata = ~0UL;
 
-int disable_pse __initdata = 0;
+int disable_pse __devinitdata = 0;
 
 /*
  * Machine setup..
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index 09b4ceb832b..fb0b200d1d8 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -59,7 +59,7 @@
 #include <smpboot_hooks.h>
 
 /* Set if we find a B stepping CPU */
-static int __initdata smp_b_stepping;
+static int __devinitdata smp_b_stepping;
 
 /* Number of siblings per CPU package */
 int smp_num_siblings = 1;
@@ -118,7 +118,7 @@ DEFINE_PER_CPU(int, cpu_state) = { 0 };
  * has made sure it's suitably aligned.
  */
 
-static unsigned long __init setup_trampoline(void)
+static unsigned long __devinit setup_trampoline(void)
 {
 	memcpy(trampoline_base, trampoline_data, trampoline_end - trampoline_data);
 	return virt_to_phys(trampoline_base);
@@ -148,7 +148,7 @@ void __init smp_alloc_memory(void)
  * a given CPU
  */
 
-static void __init smp_store_cpu_info(int id)
+static void __devinit smp_store_cpu_info(int id)
 {
 	struct cpuinfo_x86 *c = cpu_data + id;
 
@@ -342,7 +342,7 @@ extern void calibrate_delay(void);
 
 static atomic_t init_deasserted;
 
-static void __init smp_callin(void)
+static void __devinit smp_callin(void)
 {
 	int cpuid, phys_id;
 	unsigned long timeout;
@@ -468,7 +468,7 @@ set_cpu_sibling_map(int cpu)
 /*
  * Activate a secondary processor.
  */
-static void __init start_secondary(void *unused)
+static void __devinit start_secondary(void *unused)
 {
 	/*
 	 * Dont put anything before smp_callin(), SMP
@@ -521,7 +521,7 @@ static void __init start_secondary(void *unused)
  * from the task structure
  * This function must not return.
  */
-void __init initialize_secondary(void)
+void __devinit initialize_secondary(void)
 {
 	/*
 	 * We don't actually need to load the full TSS,
@@ -635,7 +635,7 @@ static inline void __inquire_remote_apic(int apicid)
  * INIT, INIT, STARTUP sequence will reset the chip hard for us, and this
  * won't ... remember to clear down the APIC, etc later.
  */
-static int __init
+static int __devinit
 wakeup_secondary_cpu(int logical_apicid, unsigned long start_eip)
 {
 	unsigned long send_status = 0, accept_status = 0;
@@ -681,7 +681,7 @@ wakeup_secondary_cpu(int logical_apicid, unsigned long start_eip)
 #endif	/* WAKE_SECONDARY_VIA_NMI */
 
 #ifdef WAKE_SECONDARY_VIA_INIT
-static int __init
+static int __devinit
 wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip)
 {
 	unsigned long send_status = 0, accept_status = 0;
@@ -817,7 +817,7 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip)
 
 extern cpumask_t cpu_initialized;
 
-static int __init do_boot_cpu(int apicid)
+static int __devinit do_boot_cpu(int apicid)
 /*
  * NOTE - on most systems this is a PHYSICAL apic ID, but on multiquad
  * (ie clustered apic addressing mode), this is a LOGICAL apic ID.
diff --git a/arch/i386/kernel/timers/timer_tsc.c b/arch/i386/kernel/timers/timer_tsc.c
index 54c36b18202..f46e625bab6 100644
--- a/arch/i386/kernel/timers/timer_tsc.c
+++ b/arch/i386/kernel/timers/timer_tsc.c
@@ -33,7 +33,7 @@ static struct timer_opts timer_tsc;
 
 static inline void cpufreq_delayed_get(void);
 
-int tsc_disable __initdata = 0;
+int tsc_disable __devinitdata = 0;
 
 extern spinlock_t i8253_lock;
 
-- 
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 'arch')

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 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 'arch')

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 'arch')

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 cb0cd8d49a9b81aff7a02e2ed826b5cfdfe9a172 Mon Sep 17 00:00:00 2001
From: Ashok Raj <ashok.raj@intel.com>
Date: Sat, 25 Jun 2005 14:55:01 -0700
Subject: [PATCH] x86_64: CPU hotplug sibling map cleanup

This patch is a minor cleanup to the cpu sibling/core map.  It is required
that this setup happens on a per-cpu bringup time.

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/smpboot.c | 84 +++++++++++++++++++-------------------------
 1 file changed, 36 insertions(+), 48 deletions(-)

(limited to 'arch')

diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index 5a3f955b657..571a55462fa 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -445,6 +445,33 @@ void __cpuinit smp_callin(void)
 	cpu_set(cpuid, cpu_callin_map);
 }
 
+static inline void set_cpu_sibling_map(int cpu)
+{
+	int i;
+
+	if (smp_num_siblings > 1) {
+		for_each_cpu(i) {
+			if (cpu_core_id[cpu] == cpu_core_id[i]) {
+				cpu_set(i, cpu_sibling_map[cpu]);
+				cpu_set(cpu, cpu_sibling_map[i]);
+			}
+		}
+	} else {
+		cpu_set(cpu, cpu_sibling_map[cpu]);
+	}
+
+	if (current_cpu_data.x86_num_cores > 1) {
+		for_each_cpu(i) {
+			if (phys_proc_id[cpu] == phys_proc_id[i]) {
+				cpu_set(i, cpu_core_map[cpu]);
+				cpu_set(cpu, cpu_core_map[i]);
+			}
+		}
+	} else {
+		cpu_core_map[cpu] = cpu_sibling_map[cpu];
+	}
+}
+
 /*
  * Setup code on secondary processor (after comming out of the trampoline)
  */
@@ -474,6 +501,12 @@ void __cpuinit start_secondary(void)
 
 	enable_APIC_timer();
 
+	/*
+	 * The sibling maps must be set before turing the online map on for
+	 * this cpu
+	 */
+	set_cpu_sibling_map(smp_processor_id());
+
 	/*
 	 * Allow the master to continue.
 	 */
@@ -816,51 +849,6 @@ do_rest:
 cycles_t cacheflush_time;
 unsigned long cache_decay_ticks;
 
-/*
- * Construct cpu_sibling_map[], so that we can tell the sibling CPU
- * on SMT systems efficiently.
- */
-static __cpuinit void detect_siblings(void)
-{
-	int cpu;
-
-	for (cpu = 0; cpu < NR_CPUS; cpu++) {
-		cpus_clear(cpu_sibling_map[cpu]);
-		cpus_clear(cpu_core_map[cpu]);
-	}
-
-	for_each_online_cpu (cpu) {
-		struct cpuinfo_x86 *c = cpu_data + cpu;
-		int siblings = 0;
-		int i;
-		if (smp_num_siblings > 1) {
-			for_each_online_cpu (i) {
-				if (cpu_core_id[cpu] == cpu_core_id[i]) {
-					siblings++;
-					cpu_set(i, cpu_sibling_map[cpu]);
-				}
-			}
-		} else {
-			siblings++;
-			cpu_set(cpu, cpu_sibling_map[cpu]);
-		}
-
-		if (siblings != smp_num_siblings) {
-			printk(KERN_WARNING
-	       "WARNING: %d siblings found for CPU%d, should be %d\n",
-			       siblings, cpu, smp_num_siblings);
-			smp_num_siblings = siblings;
-		}
-		if (c->x86_num_cores > 1) {
-			for_each_online_cpu(i) {
-				if (phys_proc_id[cpu] == phys_proc_id[i])
-					cpu_set(i, cpu_core_map[cpu]);
-			}
-		} else
-			cpu_core_map[cpu] = cpu_sibling_map[cpu];
-	}
-}
-
 /*
  * Cleanup possible dangling ends...
  */
@@ -1048,6 +1036,8 @@ void __init smp_prepare_boot_cpu(void)
 	int me = smp_processor_id();
 	cpu_set(me, cpu_online_map);
 	cpu_set(me, cpu_callout_map);
+	cpu_set(0, cpu_sibling_map[0]);
+	cpu_set(0, cpu_core_map[0]);
 }
 
 /*
@@ -1107,7 +1097,6 @@ void __init smp_cpus_done(unsigned int max_cpus)
 	setup_ioapic_dest();
 #endif
 
-	detect_siblings();
 	time_init_gtod();
 
 	check_nmi_watchdog();
@@ -1115,8 +1104,7 @@ void __init smp_cpus_done(unsigned int max_cpus)
 
 #ifdef CONFIG_HOTPLUG_CPU
 
-static void
-remove_siblinginfo(int cpu)
+static void remove_siblinginfo(int cpu)
 {
 	int sibling;
 
-- 
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 'arch')

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 a02c4cb67e4ccd5ce7a13c7f04c2fedb06c35431 Mon Sep 17 00:00:00 2001
From: Ashok Raj <ashok.raj@intel.com>
Date: Sat, 25 Jun 2005 14:55:03 -0700
Subject: [PATCH] x86_64: Provide ability to choose using shortcuts for IPI in
 flat mode.

This patch provides an option to switch broadcast or use mask version for
sending IPI's.  If CONFIG_HOTPLUG_CPU is defined, we choose not to use
broadcast shortcuts by default, otherwise we choose broadcast mode as default.

both cases, one can change this via startup cmd line option, to choose
no-broadcast mode.

	no_ipi_broadcast=1

This is provided on request from Andi Kleen, since he doesnt agree with
replacing IPI shortcuts as a solution for CPU hotplug.  Without removing
broadcast IPI's, it would mean lots of new code for __cpu_up() path, which
would acheive the same results.

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/genapic_flat.c | 96 +++++++++++++++++++++++++++++++++------
 1 file changed, 82 insertions(+), 14 deletions(-)

(limited to 'arch')

diff --git a/arch/x86_64/kernel/genapic_flat.c b/arch/x86_64/kernel/genapic_flat.c
index 00f3fa6df71..28284696508 100644
--- a/arch/x86_64/kernel/genapic_flat.c
+++ b/arch/x86_64/kernel/genapic_flat.c
@@ -20,6 +20,46 @@
 #include <asm/smp.h>
 #include <asm/ipi.h>
 
+/*
+ * The following permit choosing broadcast IPI shortcut v.s sending IPI only
+ * to online cpus via the send_IPI_mask varient.
+ * The mask version is my preferred option, since it eliminates a lot of
+ * other extra code that would need to be written to cleanup intrs sent
+ * to a CPU while offline.
+ *
+ * Sending broadcast introduces lots of trouble in CPU hotplug situations.
+ * These IPI's are delivered to cpu's irrespective of their offline status
+ * and could pickup stale intr data when these CPUS are turned online.
+ *
+ * Not using broadcast is a cleaner approach IMO, but Andi Kleen disagrees with
+ * the idea of not using broadcast IPI's anymore. Hence the run time check
+ * is introduced, on his request so we can choose an alternate mechanism.
+ *
+ * Initial wacky performance tests that collect cycle counts show
+ * no increase in using mask v.s broadcast version. In fact they seem
+ * identical in terms of cycle counts.
+ *
+ * if we need to use broadcast, we need to do the following.
+ *
+ * cli;
+ * hold call_lock;
+ * clear any pending IPI, just ack and clear all pending intr
+ * set cpu_online_map;
+ * release call_lock;
+ * sti;
+ *
+ * The complicated dummy irq processing shown above is not required if
+ * we didnt sent IPI's to wrong CPU's in the first place.
+ *
+ * - Ashok Raj <ashok.raj@intel.com>
+ */
+#ifdef CONFIG_HOTPLUG_CPU
+#define DEFAULT_SEND_IPI	(1)
+#else
+#define DEFAULT_SEND_IPI	(0)
+#endif
+
+static int no_broadcast=DEFAULT_SEND_IPI;
 
 static cpumask_t flat_target_cpus(void)
 {
@@ -79,28 +119,37 @@ static void flat_send_IPI_mask(cpumask_t cpumask, int vector)
 	local_irq_restore(flags);
 }
 
-static void flat_send_IPI_allbutself(int vector)
+static inline void __local_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 (no_broadcast) {
+		cpumask_t mask = cpu_online_map;
+		int this_cpu = get_cpu();
 
-	if (cpus_weight(mask) >= 1)
+		cpu_clear(this_cpu, mask);
 		flat_send_IPI_mask(mask, vector);
+		put_cpu();
+	}
+	else
+		__send_IPI_shortcut(APIC_DEST_ALLBUT, vector, APIC_DEST_LOGICAL);
+}
 
-	put_cpu();
+static inline void __local_flat_send_IPI_all(int vector)
+{
+	if (no_broadcast)
+		flat_send_IPI_mask(cpu_online_map, vector);
+	else
+		__send_IPI_shortcut(APIC_DEST_ALLINC, vector, APIC_DEST_LOGICAL);
+}
+
+static void flat_send_IPI_allbutself(int vector)
+{
+	if (((num_online_cpus()) - 1) >= 1)
+		__local_flat_send_IPI_allbutself(vector);
 }
 
 static void flat_send_IPI_all(int vector)
 {
-	flat_send_IPI_mask(cpu_online_map, vector);
+	__local_flat_send_IPI_all(vector);
 }
 
 static int flat_apic_id_registered(void)
@@ -121,6 +170,16 @@ static unsigned int phys_pkg_id(int index_msb)
 	return ((ebx >> 24) & 0xFF) >> index_msb;
 }
 
+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);
+
 struct genapic apic_flat =  {
 	.name = "flat",
 	.int_delivery_mode = dest_LowestPrio,
@@ -135,3 +194,12 @@ struct genapic apic_flat =  {
 	.cpu_mask_to_apicid = flat_cpu_mask_to_apicid,
 	.phys_pkg_id = phys_pkg_id,
 };
+
+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);
-- 
cgit v1.2.3-70-g09d2


From a9fa06c26f7b7914c8cdf4d309b74df3151cc227 Mon Sep 17 00:00:00 2001
From: Shaohua Li <shaohua.li@intel.com>
Date: Sat, 25 Jun 2005 14:55:05 -0700
Subject: [PATCH] set cpu_state for CPU hotplug (ia64)

Dead CPU notifies online CPU that it's dead using cpu_state variable.
After switching to physical cpu hotplug, we forgot setting the variable.
This patch fixes it.  Currently only __cpu_die uses it.  We changed other
locations for consistency in case others use it.

Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Acked-by: Ashok Raj <ashok.raj@intel.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/smpboot.c | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'arch')

diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c
index a888ddc10f7..623b0a54670 100644
--- a/arch/ia64/kernel/smpboot.c
+++ b/arch/ia64/kernel/smpboot.c
@@ -346,6 +346,7 @@ smp_callin (void)
 	lock_ipi_calllock();
 	cpu_set(cpuid, cpu_online_map);
 	unlock_ipi_calllock();
+	per_cpu(cpu_state, cpuid) = CPU_ONLINE;
 
 	smp_setup_percpu_timer();
 
@@ -611,6 +612,7 @@ void __devinit smp_prepare_boot_cpu(void)
 {
 	cpu_set(smp_processor_id(), cpu_online_map);
 	cpu_set(smp_processor_id(), cpu_callin_map);
+	per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE;
 }
 
 /*
@@ -775,6 +777,7 @@ __cpu_up (unsigned int cpu)
 	if (cpu_isset(cpu, cpu_callin_map))
 		return -EINVAL;
 
+	per_cpu(cpu_state, cpu) = CPU_UP_PREPARE;
 	/* Processor goes to start_secondary(), sets online flag */
 	ret = do_boot_cpu(sapicid, cpu);
 	if (ret < 0)
-- 
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 'arch')

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 648be3188135add682349e86d46d07cc11c8eb57 Mon Sep 17 00:00:00 2001
From: Pavel Machek <pavel@ucw.cz>
Date: Sat, 25 Jun 2005 14:55:09 -0700
Subject: [PATCH] swsusp: kill config_pm_disk

CONFIG_PM_DISK is long gone, but it still managed to survived at few
places.

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/defconfig | 1 -
 arch/i386/mm/init.c | 2 +-
 mm/page_io.c        | 2 +-
 3 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/defconfig b/arch/i386/defconfig
index 28e62038379..ca07b95c06b 100644
--- a/arch/i386/defconfig
+++ b/arch/i386/defconfig
@@ -126,7 +126,6 @@ CONFIG_HAVE_DEC_LOCK=y
 #
 CONFIG_PM=y
 CONFIG_SOFTWARE_SUSPEND=y
-# CONFIG_PM_DISK is not set
 
 #
 # ACPI (Advanced Configuration and Power Interface) Support
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index 3672e2ef51a..12216b52e28 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -352,7 +352,7 @@ static void __init pagetable_init (void)
 #endif
 }
 
-#if defined(CONFIG_PM_DISK) || defined(CONFIG_SOFTWARE_SUSPEND)
+#ifdef CONFIG_SOFTWARE_SUSPEND
 /*
  * Swap suspend & friends need this for resume because things like the intel-agp
  * driver might have split up a kernel 4MB mapping.
diff --git a/mm/page_io.c b/mm/page_io.c
index 667c76df1ec..2e605a19ce5 100644
--- a/mm/page_io.c
+++ b/mm/page_io.c
@@ -127,7 +127,7 @@ out:
 	return ret;
 }
 
-#if defined(CONFIG_SOFTWARE_SUSPEND) || defined(CONFIG_PM_DISK)
+#ifdef CONFIG_SOFTWARE_SUSPEND
 /*
  * A scruffy utility function to read or write an arbitrary swap page
  * and wait on the I/O.  The caller must have a ref on the page.
-- 
cgit v1.2.3-70-g09d2


From 343c3f642898cb2cb5c3e4f948e3e0a1bbc0351b Mon Sep 17 00:00:00 2001
From: Pavel Machek <pavel@ucw.cz>
Date: Sat, 25 Jun 2005 14:55:09 -0700
Subject: [PATCH] s-t-RAM: load gdt the right way

Sleep code uses wrong version of lgdt, that does the wrong thing when
gdt is beyond 16MB or so.

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/x86_64/kernel/acpi/wakeup.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/x86_64/kernel/acpi/wakeup.S b/arch/x86_64/kernel/acpi/wakeup.S
index a4c630034cd..185faa911db 100644
--- a/arch/x86_64/kernel/acpi/wakeup.S
+++ b/arch/x86_64/kernel/acpi/wakeup.S
@@ -67,7 +67,7 @@ wakeup_code:
 	shll	$4, %eax
 	addl	$(gdta - wakeup_code), %eax
 	movl	%eax, gdt_48a +2 - wakeup_code
-	lgdt	%ds:gdt_48a - wakeup_code		# load gdt with whatever is
+	lgdtl	%ds:gdt_48a - wakeup_code	# load gdt with whatever is
 						# appropriate
 
 	movl	$1, %eax			# protected mode (PE) bit
-- 
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 'arch')

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 350d5bd84ecda038bb482ab5f2596bdca68109d9 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Sat, 25 Jun 2005 14:55:19 -0700
Subject: [PATCH] uml: fix sizeof usage

Size of pointer doesn't seem right, but maybe my solution isn't either
(sig_size maybe?).

Signed-off-by: Domen Puncer <domen@coderock.org>
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/sys-i386/signal.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/um/sys-i386/signal.c b/arch/um/sys-i386/signal.c
index 03913ca5d25..4efc69a039d 100644
--- a/arch/um/sys-i386/signal.c
+++ b/arch/um/sys-i386/signal.c
@@ -312,7 +312,7 @@ long sys_sigreturn(struct pt_regs regs)
 	unsigned long __user *extramask = frame->extramask;
 	int sig_size = (_NSIG_WORDS - 1) * sizeof(unsigned long);
 
-	if(copy_from_user(&set.sig[0], oldmask, sizeof(&set.sig[0])) ||
+	if(copy_from_user(&set.sig[0], oldmask, sizeof(set.sig[0])) ||
 	   copy_from_user(&set.sig[1], extramask, sig_size))
 		goto segfault;
 
-- 
cgit v1.2.3-70-g09d2


From 41f2148a67f28803d64bf5ff538591af90a5ab57 Mon Sep 17 00:00:00 2001
From: Jesper Juhl <juhl-lkml@dif.dk>
Date: Sat, 25 Jun 2005 14:55:20 -0700
Subject: [PATCH] uml: kfree cleanup

Here's a small patch to remove a few unnessesary NULL pointer checks before
kfree() in arch/um/drivers/daemon_user.c

Signed-off-by: Jesper Juhl <juhl-lkml@dif.dk>
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/drivers/daemon_user.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'arch')

diff --git a/arch/um/drivers/daemon_user.c b/arch/um/drivers/daemon_user.c
index cf15b4a8b51..c1b03f7c1da 100644
--- a/arch/um/drivers/daemon_user.c
+++ b/arch/um/drivers/daemon_user.c
@@ -157,9 +157,9 @@ static void daemon_remove(void *data)
 
 	os_close_file(pri->fd);
 	os_close_file(pri->control);
-	if(pri->data_addr != NULL) kfree(pri->data_addr);
-	if(pri->ctl_addr != NULL) kfree(pri->ctl_addr);
-	if(pri->local_addr != NULL) kfree(pri->local_addr);
+	kfree(pri->data_addr);
+	kfree(pri->ctl_addr);
+	kfree(pri->local_addr);
 }
 
 int daemon_user_write(int fd, void *buf, int len, struct daemon_data *pri)
-- 
cgit v1.2.3-70-g09d2


From e0877f07e85a46e4fde32bd84f08551d360839fe Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Sat, 25 Jun 2005 14:55:21 -0700
Subject: [PATCH] uml: fork cleanup

Fix the do_fork calling convention: normal arch pass the regs and the new sp
value to do_fork instead of NULL.

Currently the arch-independent code ignores these values, while the UML code
(actually it's copy_thread) gets the right values by itself.

With this patch, things are fixed up.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/process_kern.c      |  4 ++--
 arch/um/kernel/skas/process_kern.c | 14 +-------------
 arch/um/kernel/syscall_kern.c      | 19 +++++--------------
 arch/um/kernel/tt/process_kern.c   | 17 +++--------------
 arch/um/sys-i386/syscalls.c        | 23 ++++-------------------
 arch/um/sys-x86_64/syscalls.c      | 23 ++++-------------------
 6 files changed, 19 insertions(+), 81 deletions(-)

(limited to 'arch')

diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c
index 157584ae479..0c57bc6e374 100644
--- a/arch/um/kernel/process_kern.c
+++ b/arch/um/kernel/process_kern.c
@@ -96,8 +96,8 @@ int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
 
 	current->thread.request.u.thread.proc = fn;
 	current->thread.request.u.thread.arg = arg;
-	pid = do_fork(CLONE_VM | CLONE_UNTRACED | flags, 0, NULL, 0, NULL,
-		      NULL);
+	pid = do_fork(CLONE_VM | CLONE_UNTRACED | flags, 0,
+		      &current->thread.regs, 0, NULL, NULL);
 	if(pid < 0)
 		panic("do_fork failed in kernel_thread, errno = %d", pid);
 	return(pid);
diff --git a/arch/um/kernel/skas/process_kern.c b/arch/um/kernel/skas/process_kern.c
index fc71ef29578..8f0607f54b7 100644
--- a/arch/um/kernel/skas/process_kern.c
+++ b/arch/um/kernel/skas/process_kern.c
@@ -111,8 +111,7 @@ int copy_thread_skas(int nr, unsigned long clone_flags, unsigned long sp,
   	void (*handler)(int);
 
 	if(current->thread.forking){
-	  	memcpy(&p->thread.regs.regs.skas, 
-		       &current->thread.regs.regs.skas, 
+	  	memcpy(&p->thread.regs.regs.skas, &regs->regs.skas,
 		       sizeof(p->thread.regs.regs.skas));
 		REGS_SET_SYSCALL_RETURN(p->thread.regs.regs.skas.regs, 0);
 		if(sp != 0) REGS_SP(p->thread.regs.regs.skas.regs) = sp;
@@ -201,14 +200,3 @@ int thread_pid_skas(struct task_struct *task)
 #warning Need to look up userspace_pid by cpu
 	return(userspace_pid[0]);
 }
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
diff --git a/arch/um/kernel/syscall_kern.c b/arch/um/kernel/syscall_kern.c
index b7a55251e89..8e1a3501ff4 100644
--- a/arch/um/kernel/syscall_kern.c
+++ b/arch/um/kernel/syscall_kern.c
@@ -31,7 +31,8 @@ long sys_fork(void)
 	long ret;
 
 	current->thread.forking = 1;
-        ret = do_fork(SIGCHLD, 0, NULL, 0, NULL, NULL);
+	ret = do_fork(SIGCHLD, UPT_SP(&current->thread.regs.regs),
+		      &current->thread.regs, 0, NULL, NULL);
 	current->thread.forking = 0;
 	return(ret);
 }
@@ -41,8 +42,9 @@ long sys_vfork(void)
 	long ret;
 
 	current->thread.forking = 1;
-	ret = do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, 0, NULL, 0, NULL,
-		      NULL);
+	ret = do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD,
+		      UPT_SP(&current->thread.regs.regs),
+		      &current->thread.regs, 0, NULL, NULL);
 	current->thread.forking = 0;
 	return(ret);
 }
@@ -162,14 +164,3 @@ int next_syscall_index(int limit)
 	spin_unlock(&syscall_lock);
 	return(ret);
 }
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
diff --git a/arch/um/kernel/tt/process_kern.c b/arch/um/kernel/tt/process_kern.c
index 776310fd5b8..a189a2b9293 100644
--- a/arch/um/kernel/tt/process_kern.c
+++ b/arch/um/kernel/tt/process_kern.c
@@ -266,10 +266,10 @@ int copy_thread_tt(int nr, unsigned long clone_flags, unsigned long sp,
 	}
 
 	if(current->thread.forking){
-		sc_to_sc(UPT_SC(&p->thread.regs.regs), 
-			 UPT_SC(&current->thread.regs.regs));
+		sc_to_sc(UPT_SC(&p->thread.regs.regs), UPT_SC(&regs->regs));
 		SC_SET_SYSCALL_RETURN(UPT_SC(&p->thread.regs.regs), 0);
-		if(sp != 0) SC_SP(UPT_SC(&p->thread.regs.regs)) = sp;
+		if(sp != 0)
+			SC_SP(UPT_SC(&p->thread.regs.regs)) = sp;
 	}
 	p->thread.mode.tt.extern_pid = new_pid;
 
@@ -459,14 +459,3 @@ int is_valid_pid(int pid)
 	read_unlock(&tasklist_lock);
 	return(0);
 }
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
diff --git a/arch/um/sys-i386/syscalls.c b/arch/um/sys-i386/syscalls.c
index 335e2d89504..83e9be820a8 100644
--- a/arch/um/sys-i386/syscalls.c
+++ b/arch/um/sys-i386/syscalls.c
@@ -69,15 +69,11 @@ long sys_clone(unsigned long clone_flags, unsigned long newsp,
 {
 	long ret;
 
-	/* XXX: normal arch do here this pass, and also pass the regs to
-	 * do_fork, instead of NULL. Currently the arch-independent code
-	 * ignores these values, while the UML code (actually it's
-	 * copy_thread) does the right thing. But this should change,
-	 probably. */
-	/*if (!newsp)
-		newsp = UPT_SP(current->thread.regs);*/
+	if (!newsp)
+		newsp = UPT_SP(&current->thread.regs.regs);
 	current->thread.forking = 1;
-	ret = do_fork(clone_flags, newsp, NULL, 0, parent_tid, child_tid);
+	ret = do_fork(clone_flags, newsp, &current->thread.regs, 0, parent_tid,
+		      child_tid);
 	current->thread.forking = 0;
 	return(ret);
 }
@@ -197,14 +193,3 @@ long sys_sigaction(int sig, const struct old_sigaction __user *act,
 
 	return ret;
 }
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
diff --git a/arch/um/sys-x86_64/syscalls.c b/arch/um/sys-x86_64/syscalls.c
index 6f44f40204e..3259a4db453 100644
--- a/arch/um/sys-x86_64/syscalls.c
+++ b/arch/um/sys-x86_64/syscalls.c
@@ -174,26 +174,11 @@ long sys_clone(unsigned long clone_flags, unsigned long newsp,
 {
 	long ret;
 
-	/* XXX: normal arch do here this pass, and also pass the regs to
-	 * do_fork, instead of NULL. Currently the arch-independent code
-	 * ignores these values, while the UML code (actually it's
-	 * copy_thread) does the right thing. But this should change,
-	 probably. */
-	/*if (!newsp)
-		newsp = UPT_SP(current->thread.regs);*/
+	if (!newsp)
+		newsp = UPT_SP(&current->thread.regs.regs);
 	current->thread.forking = 1;
-	ret = do_fork(clone_flags, newsp, NULL, 0, parent_tid, child_tid);
+	ret = do_fork(clone_flags, newsp, &current->thread.regs, 0, parent_tid,
+		      child_tid);
 	current->thread.forking = 0;
 	return(ret);
 }
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
-- 
cgit v1.2.3-70-g09d2


From a6f4e3cf75538a75ee4fab620e3c8466f2154458 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Sat, 25 Jun 2005 14:55:22 -0700
Subject: [PATCH] uml: fix timer initialization

In skas mode, the call to uml_idle_timer permanently shut off the virtual
timer, resulting in no timer ticks to anything but the idle thread.  This is
likely the cause of the soft lockups that are seen sporadically in recent
UMLs.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/process_kern.c      | 2 +-
 arch/um/kernel/skas/process_kern.c | 1 -
 2 files changed, 1 insertion(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c
index 0c57bc6e374..d4036ed680b 100644
--- a/arch/um/kernel/process_kern.c
+++ b/arch/um/kernel/process_kern.c
@@ -169,7 +169,7 @@ int current_pid(void)
 
 void default_idle(void)
 {
-	uml_idle_timer();
+	CHOOSE_MODE(uml_idle_timer(), (void) 0);
 
 	atomic_inc(&init_mm.mm_count);
 	current->mm = &init_mm;
diff --git a/arch/um/kernel/skas/process_kern.c b/arch/um/kernel/skas/process_kern.c
index 8f0607f54b7..0a7b8aa55db 100644
--- a/arch/um/kernel/skas/process_kern.c
+++ b/arch/um/kernel/skas/process_kern.c
@@ -180,7 +180,6 @@ int start_uml_skas(void)
 	start_userspace(0);
 
 	init_new_thread_signals(1);
-	uml_idle_timer();
 
 	init_task.thread.request.u.thread.proc = start_kernel_proc;
 	init_task.thread.request.u.thread.arg = NULL;
-- 
cgit v1.2.3-70-g09d2


From 026549d28469f7d4ca7e5a4707f0d2dc4f2c164c Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Sat, 25 Jun 2005 14:55:23 -0700
Subject: [PATCH] uml: always disable kmalloc during shutdown

kmalloc wasn't being disabled during panic.  This patch ensures that, no
matter how UML is exiting, it is disabled.  This matters because part of the
cleanup is to remove the umid file, which involves readdir, which calls
malloc.  This must map to libc malloc, rather than kmalloc or vmalloc.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/main.c   | 1 -
 arch/um/kernel/reboot.c | 9 ++++-----
 2 files changed, 4 insertions(+), 6 deletions(-)

(limited to 'arch')

diff --git a/arch/um/kernel/main.c b/arch/um/kernel/main.c
index e59f5815267..1e1a87f1c51 100644
--- a/arch/um/kernel/main.c
+++ b/arch/um/kernel/main.c
@@ -69,7 +69,6 @@ static __init void do_uml_initcalls(void)
 
 static void last_ditch_exit(int sig)
 {
-        kmalloc_ok = 0;
 	signal(SIGINT, SIG_DFL);
 	signal(SIGTERM, SIG_DFL);
 	signal(SIGHUP, SIG_DFL);
diff --git a/arch/um/kernel/reboot.c b/arch/um/kernel/reboot.c
index 207f89d7490..fcec51da1d3 100644
--- a/arch/um/kernel/reboot.c
+++ b/arch/um/kernel/reboot.c
@@ -38,14 +38,14 @@ static void kill_off_processes(void)
 
 void uml_cleanup(void)
 {
-	kill_off_processes();
+        kmalloc_ok = 0;
 	do_uml_exitcalls();
+	kill_off_processes();
 }
 
 void machine_restart(char * __unused)
 {
-	do_uml_exitcalls();
-	kill_off_processes();
+        uml_cleanup();
 	CHOOSE_MODE(reboot_tt(), reboot_skas());
 }
 
@@ -53,8 +53,7 @@ EXPORT_SYMBOL(machine_restart);
 
 void machine_power_off(void)
 {
-	do_uml_exitcalls();
-	kill_off_processes();
+        uml_cleanup();
 	CHOOSE_MODE(halt_tt(), halt_skas());
 }
 
-- 
cgit v1.2.3-70-g09d2


From fc47a0d18a1994b4a18d2235fcde1b75dfa72552 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Sat, 25 Jun 2005 14:55:24 -0700
Subject: [PATCH] uml: time initialization tidying

user_time_init_skas and user_time_init_tt were essentially the same.  So, this
merges them, deleting the mode-specific functions and declarations.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/include/time_user.h             |  2 +-
 arch/um/kernel/skas/Makefile            |  4 ++--
 arch/um/kernel/skas/include/mode-skas.h |  1 -
 arch/um/kernel/skas/time.c              | 30 ------------------------------
 arch/um/kernel/time.c                   | 31 ++++++++++++++-----------------
 arch/um/kernel/time_kern.c              |  2 +-
 arch/um/kernel/tt/Makefile              |  4 ++--
 arch/um/kernel/tt/include/mode-tt.h     |  1 -
 arch/um/kernel/tt/time.c                | 28 ----------------------------
 9 files changed, 20 insertions(+), 83 deletions(-)
 delete mode 100644 arch/um/kernel/skas/time.c
 delete mode 100644 arch/um/kernel/tt/time.c

(limited to 'arch')

diff --git a/arch/um/include/time_user.h b/arch/um/include/time_user.h
index 6793a2fcd0a..f64ef77019a 100644
--- a/arch/um/include/time_user.h
+++ b/arch/um/include/time_user.h
@@ -8,11 +8,11 @@
 
 extern void timer(void);
 extern void switch_timers(int to_real);
-extern void set_interval(int timer_type);
 extern void idle_sleep(int secs);
 extern void enable_timer(void);
 extern void disable_timer(void);
 extern unsigned long time_lock(void);
 extern void time_unlock(unsigned long);
+extern void user_time_init(void);
 
 #endif
diff --git a/arch/um/kernel/skas/Makefile b/arch/um/kernel/skas/Makefile
index d37d1bfcd6f..ff69c4b312c 100644
--- a/arch/um/kernel/skas/Makefile
+++ b/arch/um/kernel/skas/Makefile
@@ -4,10 +4,10 @@
 #
 
 obj-y := exec_kern.o mem.o mem_user.o mmu.o process.o process_kern.o \
-	syscall_kern.o syscall_user.o time.o tlb.o trap_user.o uaccess.o \
+	syscall_kern.o syscall_user.o tlb.o trap_user.o uaccess.o \
 
 subdir- := util
 
-USER_OBJS := process.o time.o
+USER_OBJS := process.o
 
 include arch/um/scripts/Makefile.rules
diff --git a/arch/um/kernel/skas/include/mode-skas.h b/arch/um/kernel/skas/include/mode-skas.h
index c1e33bd788d..bcd26a6a388 100644
--- a/arch/um/kernel/skas/include/mode-skas.h
+++ b/arch/um/kernel/skas/include/mode-skas.h
@@ -13,7 +13,6 @@ extern unsigned long exec_fp_regs[];
 extern unsigned long exec_fpx_regs[];
 extern int have_fpx_regs;
 
-extern void user_time_init_skas(void);
 extern void sig_handler_common_skas(int sig, void *sc_ptr);
 extern void halt_skas(void);
 extern void reboot_skas(void);
diff --git a/arch/um/kernel/skas/time.c b/arch/um/kernel/skas/time.c
deleted file mode 100644
index 98091494b89..00000000000
--- a/arch/um/kernel/skas/time.c
+++ /dev/null
@@ -1,30 +0,0 @@
-/* 
- * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
- * Licensed under the GPL
- */
-
-#include <sys/signal.h>
-#include <sys/time.h>
-#include "time_user.h"
-#include "process.h"
-#include "user.h"
-
-void user_time_init_skas(void)
-{
-        if(signal(SIGALRM, (__sighandler_t) alarm_handler) == SIG_ERR)
-                panic("Couldn't set SIGALRM handler");
- 	if(signal(SIGVTALRM, (__sighandler_t) alarm_handler) == SIG_ERR)
- 		panic("Couldn't set SIGVTALRM handler");
-	set_interval(ITIMER_VIRTUAL);
-}
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c
index c40c86a3f91..f829b309b63 100644
--- a/arch/um/kernel/time.c
+++ b/arch/um/kernel/time.c
@@ -33,7 +33,7 @@ void timer(void)
 	timeradd(&xtime, &local_offset, &xtime);
 }
 
-void set_interval(int timer_type)
+static void set_interval(int timer_type)
 {
 	int usec = 1000000/hz();
 	struct itimerval interval = ((struct itimerval) { { 0, usec },
@@ -45,12 +45,7 @@ void set_interval(int timer_type)
 
 void enable_timer(void)
 {
-	int usec = 1000000/hz();
-	struct itimerval enable = ((struct itimerval) { { 0, usec },
-							{ 0, usec }});
-	if(setitimer(ITIMER_VIRTUAL, &enable, NULL))
-		printk("enable_timer - setitimer failed, errno = %d\n",
-		       errno);
+	set_interval(ITIMER_VIRTUAL);
 }
 
 void disable_timer(void)
@@ -155,13 +150,15 @@ void idle_sleep(int secs)
 	nanosleep(&ts, NULL);
 }
 
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
+/* XXX This partly duplicates init_irq_signals */
+
+void user_time_init(void)
+{
+	set_handler(SIGVTALRM, (__sighandler_t) alarm_handler,
+		    SA_ONSTACK | SA_RESTART, SIGUSR1, SIGIO, SIGWINCH,
+		    SIGALRM, SIGUSR2, -1);
+	set_handler(SIGALRM, (__sighandler_t) alarm_handler,
+		    SA_ONSTACK | SA_RESTART, SIGUSR1, SIGIO, SIGWINCH,
+		    SIGVTALRM, SIGUSR2, -1);
+	set_interval(ITIMER_VIRTUAL);
+}
diff --git a/arch/um/kernel/time_kern.c b/arch/um/kernel/time_kern.c
index 6516fc52afe..a8b4ef601f5 100644
--- a/arch/um/kernel/time_kern.c
+++ b/arch/um/kernel/time_kern.c
@@ -162,7 +162,7 @@ int __init timer_init(void)
 {
 	int err;
 
-	CHOOSE_MODE(user_time_init_tt(), user_time_init_skas());
+	user_time_init();
 	err = request_irq(TIMER_IRQ, um_timer, SA_INTERRUPT, "timer", NULL);
 	if(err != 0)
 		printk(KERN_ERR "timer_init : request_irq failed - "
diff --git a/arch/um/kernel/tt/Makefile b/arch/um/kernel/tt/Makefile
index 3fd2554e60b..6939e5af847 100644
--- a/arch/um/kernel/tt/Makefile
+++ b/arch/um/kernel/tt/Makefile
@@ -4,11 +4,11 @@
 #
 
 obj-y = exec_kern.o exec_user.o gdb.o ksyms.o mem.o mem_user.o process_kern.o \
-	syscall_kern.o syscall_user.o time.o tlb.o tracer.o trap_user.o \
+	syscall_kern.o syscall_user.o tlb.o tracer.o trap_user.o \
 	uaccess.o uaccess_user.o
 
 obj-$(CONFIG_PT_PROXY) += gdb_kern.o ptproxy/
 
-USER_OBJS := gdb.o time.o tracer.o
+USER_OBJS := gdb.o tracer.o
 
 include arch/um/scripts/Makefile.rules
diff --git a/arch/um/kernel/tt/include/mode-tt.h b/arch/um/kernel/tt/include/mode-tt.h
index efe46201906..e171e15fead 100644
--- a/arch/um/kernel/tt/include/mode-tt.h
+++ b/arch/um/kernel/tt/include/mode-tt.h
@@ -13,7 +13,6 @@ enum { OP_NONE, OP_EXEC, OP_FORK, OP_TRACE_ON, OP_REBOOT, OP_HALT, OP_CB };
 extern int tracing_pid;
 
 extern int tracer(int (*init_proc)(void *), void *sp);
-extern void user_time_init_tt(void);
 extern void sig_handler_common_tt(int sig, void *sc);
 extern void syscall_handler_tt(int sig, union uml_pt_regs *regs);
 extern void reboot_tt(void);
diff --git a/arch/um/kernel/tt/time.c b/arch/um/kernel/tt/time.c
deleted file mode 100644
index 8565b71b07c..00000000000
--- a/arch/um/kernel/tt/time.c
+++ /dev/null
@@ -1,28 +0,0 @@
-/* 
- * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com)
- * Licensed under the GPL
- */
-
-#include <signal.h>
-#include <sys/time.h>
-#include <time_user.h>
-#include "process.h"
-#include "user.h"
-
-void user_time_init_tt(void)
-{
-	if(signal(SIGVTALRM, (__sighandler_t) alarm_handler) == SIG_ERR)
-		panic("Couldn't set SIGVTALRM handler");
-	set_interval(ITIMER_VIRTUAL);
-}
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
-- 
cgit v1.2.3-70-g09d2


From 29d56cfe3ca599ddc3ae9156e7e469c044d97b96 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Sat, 25 Jun 2005 14:55:25 -0700
Subject: [PATCH] uml: hot-unplug code cleanup

Clean up the hot-unplugging code.  There is now an id procedure which is
called to figure out what device we're talking to.  The error messages from
that are now done from mconsole_remove instead of the driver.  remove is now
called with the device number, after it has been checked, so doesn't need to
do sanity checking on it.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/drivers/line.c            | 19 +++++++++++++++++--
 arch/um/drivers/mconsole_kern.c   | 36 ++++++++++++++++++++++++++++++++----
 arch/um/drivers/net_kern.c        | 31 +++++++++++++++++++++----------
 arch/um/drivers/ssl.c             |  9 +++++----
 arch/um/drivers/stdio_console.c   |  7 ++++---
 arch/um/drivers/ubd_kern.c        | 37 ++++++++++++++++++++++++-------------
 arch/um/include/line.h            |  3 ++-
 arch/um/include/mconsole_kern.h   |  3 ++-
 arch/um/kernel/tt/gdb.c           |  4 ++--
 arch/um/kernel/tt/gdb_kern.c      |  2 +-
 arch/um/kernel/tt/include/debug.h | 17 +++--------------
 11 files changed, 113 insertions(+), 55 deletions(-)

(limited to 'arch')

diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c
index 0f59736db32..2bb4c4f5dec 100644
--- a/arch/um/drivers/line.c
+++ b/arch/um/drivers/line.c
@@ -602,11 +602,26 @@ int line_get_config(char *name, struct line *lines, unsigned int num, char *str,
 	return n;
 }
 
-int line_remove(struct line *lines, unsigned int num, char *str)
+int line_id(char **str, int *start_out, int *end_out)
+{
+	char *end;
+        int n;
+
+	n = simple_strtoul(*str, &end, 0);
+	if((*end != '\0') || (end == *str))
+                return -1;
+
+        *str = end;
+        *start_out = n;
+        *end_out = n;
+        return n;
+}
+
+int line_remove(struct line *lines, unsigned int num, int n)
 {
 	char config[sizeof("conxxxx=none\0")];
 
-	sprintf(config, "%s=none", str);
+	sprintf(config, "%d=none", n);
 	return !line_setup(lines, num, config, 0);
 }
 
diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c
index d7c7adcc0a6..404de41a4f6 100644
--- a/arch/um/drivers/mconsole_kern.c
+++ b/arch/um/drivers/mconsole_kern.c
@@ -419,8 +419,9 @@ void mconsole_config(struct mc_request *req)
 void mconsole_remove(struct mc_request *req)
 {
 	struct mc_device *dev;	
-	char *ptr = req->request.data;
-	int err;
+	char *ptr = req->request.data, *err_msg = "";
+        char error[256];
+	int err, start, end, n;
 
 	ptr += strlen("remove");
 	while(isspace(*ptr)) ptr++;
@@ -429,8 +430,35 @@ void mconsole_remove(struct mc_request *req)
 		mconsole_reply(req, "Bad remove option", 1, 0);
 		return;
 	}
-	err = (*dev->remove)(&ptr[strlen(dev->name)]);
-	mconsole_reply(req, "", err, 0);
+
+        ptr = &ptr[strlen(dev->name)];
+
+        err = 1;
+        n = (*dev->id)(&ptr, &start, &end);
+        if(n < 0){
+                err_msg = "Couldn't parse device number";
+                goto out;
+        }
+        else if((n < start) || (n > end)){
+                sprintf(error, "Invalid device number - must be between "
+                        "%d and %d", start, end);
+                err_msg = error;
+                goto out;
+        }
+
+	err = (*dev->remove)(n);
+        switch(err){
+        case -ENODEV:
+                err_msg = "Device doesn't exist";
+                break;
+        case -EBUSY:
+                err_msg = "Device is currently open";
+                break;
+        default:
+                break;
+        }
+ out:
+	mconsole_reply(req, err_msg, err, 0);
 }
 
 #ifdef CONFIG_MAGIC_SYSRQ
diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c
index 5388a742869..1495007bf6c 100644
--- a/arch/um/drivers/net_kern.c
+++ b/arch/um/drivers/net_kern.c
@@ -612,25 +612,35 @@ static int net_config(char *str)
 	return(err);
 }
 
-static int net_remove(char *str)
+static int net_id(char **str, int *start_out, int *end_out)
+{
+        char *end;
+        int n;
+
+	n = simple_strtoul(*str, &end, 0);
+	if((*end != '\0') || (end == *str))
+		return -1;
+
+        *start_out = n;
+        *end_out = n;
+        *str = end;
+        return n;
+}
+
+static int net_remove(int n)
 {
 	struct uml_net *device;
 	struct net_device *dev;
 	struct uml_net_private *lp;
-	char *end;
-	int n;
-
-	n = simple_strtoul(str, &end, 0);
-	if((*end != '\0') || (end == str))
-		return(-1);
 
 	device = find_device(n);
 	if(device == NULL)
-		return(0);
+		return -ENODEV;
 
 	dev = device->dev;
 	lp = dev->priv;
-	if(lp->fd > 0) return(-1);
+	if(lp->fd > 0)
+                return -EBUSY;
 	if(lp->remove != NULL) (*lp->remove)(&lp->user);
 	unregister_netdev(dev);
 	platform_device_unregister(&device->pdev);
@@ -638,13 +648,14 @@ static int net_remove(char *str)
 	list_del(&device->list);
 	kfree(device);
 	free_netdev(dev);
-	return(0);
+	return 0;
 }
 
 static struct mc_device net_mc = {
 	.name		= "eth",
 	.config		= net_config,
 	.get_config	= NULL,
+        .id		= net_id,
 	.remove		= net_remove,
 };
 
diff --git a/arch/um/drivers/ssl.c b/arch/um/drivers/ssl.c
index b32a77010fb..62e04ecfada 100644
--- a/arch/um/drivers/ssl.c
+++ b/arch/um/drivers/ssl.c
@@ -49,7 +49,7 @@ static struct chan_opts opts = {
 
 static int ssl_config(char *str);
 static int ssl_get_config(char *dev, char *str, int size, char **error_out);
-static int ssl_remove(char *str);
+static int ssl_remove(int n);
 
 static struct line_driver driver = {
 	.name 			= "UML serial line",
@@ -69,6 +69,7 @@ static struct line_driver driver = {
 		.name  		= "ssl",
 		.config 	= ssl_config,
 		.get_config 	= ssl_get_config,
+                .id		= line_id,
 		.remove 	= ssl_remove,
 	},
 };
@@ -94,10 +95,10 @@ static int ssl_get_config(char *dev, char *str, int size, char **error_out)
 			       str, size, error_out));
 }
 
-static int ssl_remove(char *str)
+static int ssl_remove(int n)
 {
-	return(line_remove(serial_lines, 
-			   sizeof(serial_lines)/sizeof(serial_lines[0]), str));
+        return line_remove(serial_lines,
+                           sizeof(serial_lines)/sizeof(serial_lines[0]), n);
 }
 
 int ssl_open(struct tty_struct *tty, struct file *filp)
diff --git a/arch/um/drivers/stdio_console.c b/arch/um/drivers/stdio_console.c
index afbe1e71ed8..005aa6333b6 100644
--- a/arch/um/drivers/stdio_console.c
+++ b/arch/um/drivers/stdio_console.c
@@ -55,7 +55,7 @@ static struct chan_opts opts = {
 
 static int con_config(char *str);
 static int con_get_config(char *dev, char *str, int size, char **error_out);
-static int con_remove(char *str);
+static int con_remove(int n);
 
 static struct line_driver driver = {
 	.name 			= "UML console",
@@ -75,6 +75,7 @@ static struct line_driver driver = {
 		.name  		= "con",
 		.config 	= con_config,
 		.get_config 	= con_get_config,
+                .id		= line_id,
 		.remove 	= con_remove,
 	},
 };
@@ -99,9 +100,9 @@ static int con_get_config(char *dev, char *str, int size, char **error_out)
 			       size, error_out));
 }
 
-static int con_remove(char *str)
+static int con_remove(int n)
 {
-	return(line_remove(vts, sizeof(vts)/sizeof(vts[0]), str));
+        return line_remove(vts, sizeof(vts)/sizeof(vts[0]), n);
 }
 
 static int con_open(struct tty_struct *tty, struct file *filp)
diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c
index 2a7f6892c55..344b24d09a7 100644
--- a/arch/um/drivers/ubd_kern.c
+++ b/arch/um/drivers/ubd_kern.c
@@ -754,24 +754,34 @@ static int ubd_get_config(char *name, char *str, int size, char **error_out)
 	return(len);
 }
 
-static int ubd_remove(char *str)
+static int ubd_id(char **str, int *start_out, int *end_out)
+{
+        int n;
+
+	n = parse_unit(str);
+        *start_out = 0;
+        *end_out = MAX_DEV - 1;
+        return n;
+}
+
+static int ubd_remove(int n)
 {
 	struct ubd *dev;
-	int n, err = -ENODEV;
+	int err = -ENODEV;
 
-	n = parse_unit(&str);
+	spin_lock(&ubd_lock);
 
-	if((n < 0) || (n >= MAX_DEV))
-		return(err);
+	if(ubd_gendisk[n] == NULL)
+		goto out;
 
 	dev = &ubd_dev[n];
-	if(dev->count > 0)
-		return(-EBUSY);	/* you cannot remove a open disk */
 
-	err = 0;
- 	spin_lock(&ubd_lock);
+	if(dev->file == NULL)
+		goto out;
 
-	if(ubd_gendisk[n] == NULL)
+	/* you cannot remove a open disk */
+	err = -EBUSY;
+	if(dev->count > 0)
 		goto out;
 
 	del_gendisk(ubd_gendisk[n]);
@@ -787,15 +797,16 @@ static int ubd_remove(char *str)
 	platform_device_unregister(&dev->pdev);
 	*dev = ((struct ubd) DEFAULT_UBD);
 	err = 0;
- out:
- 	spin_unlock(&ubd_lock);
-	return(err);
+out:
+	spin_unlock(&ubd_lock);
+	return err;
 }
 
 static struct mc_device ubd_mc = {
 	.name		= "ubd",
 	.config		= ubd_config,
  	.get_config	= ubd_get_config,
+	.id		= ubd_id,
 	.remove		= ubd_remove,
 };
 
diff --git a/arch/um/include/line.h b/arch/um/include/line.h
index 4c5e92c04cc..5323d22a6ca 100644
--- a/arch/um/include/line.h
+++ b/arch/um/include/line.h
@@ -101,7 +101,8 @@ extern void lines_init(struct line *lines, int nlines);
 extern void close_lines(struct line *lines, int nlines);
 
 extern int line_config(struct line *lines, unsigned int sizeof_lines, char *str);
-extern int line_remove(struct line *lines, unsigned int sizeof_lines, char *str);
+extern int line_id(char **str, int *start_out, int *end_out);
+extern int line_remove(struct line *lines, unsigned int sizeof_lines, int n);
 extern int line_get_config(char *dev, struct line *lines, unsigned int sizeof_lines, char *str,
 			   int size, char **error_out);
 
diff --git a/arch/um/include/mconsole_kern.h b/arch/um/include/mconsole_kern.h
index 61c274fcee5..d86ee14260c 100644
--- a/arch/um/include/mconsole_kern.h
+++ b/arch/um/include/mconsole_kern.h
@@ -20,7 +20,8 @@ struct mc_device {
 	char *name;
 	int (*config)(char *);
 	int (*get_config)(char *, char *, int, char **);
-	int (*remove)(char *);
+        int (*id)(char **, int *, int *);
+	int (*remove)(int);
 };
 
 #define CONFIG_CHUNK(str, size, current, chunk, end) \
diff --git a/arch/um/kernel/tt/gdb.c b/arch/um/kernel/tt/gdb.c
index 19a0ad7b35b..37e22d71a0d 100644
--- a/arch/um/kernel/tt/gdb.c
+++ b/arch/um/kernel/tt/gdb.c
@@ -153,10 +153,10 @@ void remove_gdb_cb(void *unused)
 	exit_debugger_cb(NULL);
 }
 
-int gdb_remove(char *unused)
+int gdb_remove(int unused)
 {
 	initial_thread_cb(remove_gdb_cb, NULL);
-	return(0);
+        return 0;
 }
 
 void signal_usr1(int sig)
diff --git a/arch/um/kernel/tt/gdb_kern.c b/arch/um/kernel/tt/gdb_kern.c
index 93fb121f86a..26506388a6a 100644
--- a/arch/um/kernel/tt/gdb_kern.c
+++ b/arch/um/kernel/tt/gdb_kern.c
@@ -10,7 +10,7 @@
 #ifdef CONFIG_MCONSOLE
 
 extern int gdb_config(char *str);
-extern int gdb_remove(char *unused);
+extern int gdb_remove(int n);
 
 static struct mc_device gdb_mc = {
 	.name		= "gdb",
diff --git a/arch/um/kernel/tt/include/debug.h b/arch/um/kernel/tt/include/debug.h
index 8eff674107c..738435461e1 100644
--- a/arch/um/kernel/tt/include/debug.h
+++ b/arch/um/kernel/tt/include/debug.h
@@ -4,8 +4,8 @@
  * Licensed under the GPL
  */
 
-#ifndef __DEBUG_H
-#define __DEBUG_H
+#ifndef __UML_TT_DEBUG_H
+#define __UML_TT_DEBUG_H
 
 extern int debugger_proxy(int status, pid_t pid);
 extern void child_proxy(pid_t pid, int status);
@@ -13,17 +13,6 @@ extern void init_proxy (pid_t pid, int waiting, int status);
 extern int start_debugger(char *prog, int startup, int stop, int *debugger_fd);
 extern void fake_child_exit(void);
 extern int gdb_config(char *str);
-extern int gdb_remove(char *unused);
+extern int gdb_remove(int unused);
 
 #endif
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
-- 
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 'arch')

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 'arch')

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 'arch')

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 'arch')

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 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 'arch')

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 7f1867a5b3dc3034cbea403b229d65eed4a7f62e Mon Sep 17 00:00:00 2001
From: Dinakar Guniguntala <dino@in.ibm.com>
Date: Sat, 25 Jun 2005 14:57:36 -0700
Subject: [PATCH] Dynamic sched domains: ia64 changes

ia64 changes similar to kernel/sched.c.

Signed-off-by: Dinakar Guniguntala <dino@in.ibm.com>
Acked-by: Paul Jackson <pj@sgi.com>
Acked-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 | 76 ++++++++++++++++++++++++++++-------------------
 1 file changed, 45 insertions(+), 31 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/domain.c b/arch/ia64/kernel/domain.c
index afbde79c3b3..d65e87b6394 100644
--- a/arch/ia64/kernel/domain.c
+++ b/arch/ia64/kernel/domain.c
@@ -27,7 +27,7 @@
  *
  * Should use nodemask_t.
  */
-static int __devinit find_next_best_node(int node, unsigned long *used_nodes)
+static int find_next_best_node(int node, unsigned long *used_nodes)
 {
 	int i, n, val, min_val, best_node = 0;
 
@@ -66,7 +66,7 @@ static int __devinit find_next_best_node(int node, unsigned long *used_nodes)
  * should be one that prevents unnecessary balancing, but also spreads tasks
  * out optimally.
  */
-static cpumask_t __devinit sched_domain_node_span(int node)
+static cpumask_t sched_domain_node_span(int node)
 {
 	int i;
 	cpumask_t span, nodemask;
@@ -96,7 +96,7 @@ static cpumask_t __devinit sched_domain_node_span(int node)
 #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;
 }
@@ -104,7 +104,7 @@ 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]);
@@ -125,44 +125,36 @@ static struct sched_group *sched_group_nodes[MAX_NUMNODES];
 static DEFINE_PER_CPU(struct sched_domain, allnodes_domains);
 static struct sched_group sched_group_allnodes[MAX_NUMNODES];
 
-static int __devinit cpu_to_allnodes_group(int cpu)
+static int cpu_to_allnodes_group(int cpu)
 {
 	return cpu_to_node(cpu);
 }
 #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
  */
-void __devinit arch_init_sched_domains(void)
+void build_sched_domains(const cpumask_t *cpu_map)
 {
 	int i;
-	cpumask_t cpu_default_map;
 
 	/*
-	 * 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.
+	 * Set up domains for cpus specified by the cpu_map.
 	 */
-	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 dummy domain.
-	 */
-	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
 		if (num_online_cpus()
 				> SD_NODES_PER_DOMAIN*cpus_weight(nodemask)) {
 			sd = &per_cpu(allnodes_domains, i);
 			*sd = SD_ALLNODES_INIT;
-			sd->span = cpu_default_map;
+			sd->span = *cpu_map;
 			group = cpu_to_allnodes_group(i);
 			sd->groups = &sched_group_allnodes[group];
 			p = sd;
@@ -173,7 +165,7 @@ void __devinit arch_init_sched_domains(void)
 		*sd = SD_NODE_INIT;
 		sd->span = sched_domain_node_span(cpu_to_node(i));
 		sd->parent = p;
-		cpus_and(sd->span, sd->span, cpu_default_map);
+		cpus_and(sd->span, sd->span, *cpu_map);
 #endif
 
 		p = sd;
@@ -190,7 +182,7 @@ 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
@@ -198,9 +190,9 @@ void __devinit arch_init_sched_domains(void)
 
 #ifdef CONFIG_SCHED_SMT
 	/* Set up CPU (sibling) groups */
-	for_each_cpu_mask(i, cpu_default_map) {
+	for_each_cpu_mask(i, *cpu_map) {
 		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;
 
@@ -213,7 +205,7 @@ 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;
 
@@ -222,7 +214,7 @@ void __devinit arch_init_sched_domains(void)
 	}
 
 #ifdef CONFIG_NUMA
-	init_sched_build_groups(sched_group_allnodes, cpu_default_map,
+	init_sched_build_groups(sched_group_allnodes, *cpu_map,
 				&cpu_to_allnodes_group);
 
 	for (i = 0; i < MAX_NUMNODES; i++) {
@@ -233,12 +225,12 @@ void __devinit arch_init_sched_domains(void)
 		cpumask_t covered = CPU_MASK_NONE;
 		int j;
 
-		cpus_and(nodemask, nodemask, cpu_default_map);
+		cpus_and(nodemask, nodemask, *cpu_map);
 		if (cpus_empty(nodemask))
 			continue;
 
 		domainspan = sched_domain_node_span(i);
-		cpus_and(domainspan, domainspan, cpu_default_map);
+		cpus_and(domainspan, domainspan, *cpu_map);
 
 		sg = kmalloc(sizeof(struct sched_group), GFP_KERNEL);
 		sched_group_nodes[i] = sg;
@@ -266,7 +258,7 @@ void __devinit arch_init_sched_domains(void)
 			int n = (i + j) % MAX_NUMNODES;
 
 			cpus_complement(notcovered, covered);
-			cpus_and(tmp, notcovered, cpu_default_map);
+			cpus_and(tmp, notcovered, *cpu_map);
 			cpus_and(tmp, tmp, domainspan);
 			if (cpus_empty(tmp))
 				break;
@@ -293,7 +285,7 @@ void __devinit arch_init_sched_domains(void)
 #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
@@ -359,13 +351,35 @@ next_sg:
 		cpu_attach_domain(sd, i);
 	}
 }
+/*
+ * Set up scheduler domains and groups.  Callers must hold the hotplug lock.
+ */
+void arch_init_sched_domains(const cpumask_t *cpu_map)
+{
+	cpumask_t cpu_default_map;
+
+	/*
+	 * 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);
+}
 
-void __devinit arch_destroy_sched_domains(void)
+void arch_destroy_sched_domains(const cpumask_t *cpu_map)
 {
 #ifdef CONFIG_NUMA
 	int i;
 	for (i = 0; i < MAX_NUMNODES; i++) {
+		cpumask_t nodemask = node_to_cpumask(i);
 		struct sched_group *oldsg, *sg = sched_group_nodes[i];
+
+		cpus_and(nodemask, nodemask, *cpu_map);
+		if (cpus_empty(nodemask))
+			continue;
+
 		if (sg == NULL)
 			continue;
 		sg = sg->next;
-- 
cgit v1.2.3-70-g09d2


From cc19ca86a023fcd552c78e77a7be6ce271f92a28 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sat, 25 Jun 2005 14:57:36 -0700
Subject: [PATCH] consolidate PREEMPT options into kernel/Kconfig.preempt

This patch consolidates the CONFIG_PREEMPT and CONFIG_PREEMPT_BKL
preemption options into kernel/Kconfig.preempt.  This, besides reducing
source-code, also enables more centralized tweaking of preemption related
options.

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/Kconfig      | 23 +----------------------
 arch/ppc64/Kconfig     | 21 +--------------------
 arch/x86_64/Kconfig    | 29 ++---------------------------
 kernel/Kconfig.preempt | 24 ++++++++++++++++++++++++
 4 files changed, 28 insertions(+), 69 deletions(-)
 create mode 100644 kernel/Kconfig.preempt

(limited to 'arch')

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index b4cd11e5845..961ab20032f 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -510,28 +510,7 @@ config SCHED_SMT
 	  cost of slightly increased overhead in some places. If unsure say
 	  N here.
 
-config PREEMPT
-	bool "Preemptible Kernel"
-	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
-	  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.
-
-config PREEMPT_BKL
-	bool "Preempt The Big Kernel Lock"
-	depends on PREEMPT
-	default y
-	help
-	  This option reduces the latency of the kernel by making the
-	  big kernel lock preemptible.
-
-	  Say Y here if you are building a kernel for a desktop system.
-	  Say N if you are unsure.
+source "kernel/Kconfig.preempt"
 
 config X86_UP_APIC
 	bool "Local APIC support on uniprocessors"
diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig
index cb27068bfcd..5f40b438b58 100644
--- a/arch/ppc64/Kconfig
+++ b/arch/ppc64/Kconfig
@@ -270,26 +270,7 @@ config SCHED_SMT
 	  when dealing with POWER5 cpus at a cost of slightly increased
 	  overhead in some places. If unsure say N here.
 
-config PREEMPT
-	bool "Preemptible Kernel"
-	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.
-
-	  Say Y here if you are building a kernel for a desktop, embedded
-	  or real-time system.  Say N if you are unsure.
-
-config PREEMPT_BKL
-	bool "Preempt The Big Kernel Lock"
-	depends on PREEMPT
-	default y
-	help
-	  This option reduces the latency of the kernel by making the
-	  big kernel lock preemptible.
-
-	  Say Y here if you are building a kernel for a desktop system.
-	  Say N if you are unsure.
+source "kernel/Kconfig.preempt"
 
 config EEH
 	bool "PCI Extended Error Handling (EEH)" if EMBEDDED
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 92f5a526602..a853d87ff7e 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -207,33 +207,6 @@ config SMP
 
 	  If you don't know what to do here, say N.
 
-config PREEMPT
-	bool "Preemptible Kernel"
-	---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
-	  under load. On contrary it may also break your drivers and add
-	  priority inheritance problems to your system. Don't select it if
-	  you rely on a stable system or have slightly obscure hardware.
-	  It's also not very well tested on x86-64 currently.
-	  You have been warned.
-
-	  Say Y here if you are feeling brave and building a kernel for a
-	  desktop, embedded or real-time system.  Say N if you are unsure.
-
-config PREEMPT_BKL
-	bool "Preempt The Big Kernel Lock"
-	depends on PREEMPT
-	default y
-	help
-	  This option reduces the latency of the kernel by making the
-	  big kernel lock preemptible.
-
-	  Say Y here if you are building a kernel for a desktop system.
-	  Say N if you are unsure.
-
 config SCHED_SMT
 	bool "SMT (Hyperthreading) scheduler support"
 	depends on SMP
@@ -244,6 +217,8 @@ config SCHED_SMT
 	  cost of slightly increased overhead in some places. If unsure say
 	  N here.
 
+source "kernel/Kconfig.preempt"
+
 config K8_NUMA
        bool "K8 NUMA support"
        select NUMA
diff --git a/kernel/Kconfig.preempt b/kernel/Kconfig.preempt
new file mode 100644
index 00000000000..587328be821
--- /dev/null
+++ b/kernel/Kconfig.preempt
@@ -0,0 +1,24 @@
+
+config PREEMPT
+	bool "Preemptible Kernel"
+	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
+	  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.
+
+config PREEMPT_BKL
+	bool "Preempt The Big Kernel Lock"
+	depends on PREEMPT
+	default y
+	help
+	  This option reduces the latency of the kernel by making the
+	  big kernel lock preemptible.
+
+	  Say Y here if you are building a kernel for a desktop system.
+	  Say N if you are unsure.
+
-- 
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 'arch')

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 70adada4282b52130ef31c7722d3fd8d94dacbc2 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:42 -0700
Subject: [PATCH] kexec: x86_64: e820 64bit fix

From: Eric W. Biederman <ebiederm@xmission.com>

It is ok to reserve resources > 4G on x86_64 struct resource is 64bit now :)

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/e820.c | 2 --
 1 file changed, 2 deletions(-)

(limited to 'arch')

diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index 7c154dfff64..cf4878309b0 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -191,8 +191,6 @@ void __init e820_reserve_resources(void)
 	int i;
 	for (i = 0; i < e820.nr_map; i++) {
 		struct resource *res;
-		if (e820.map[i].addr + e820.map[i].size > 0x100000000ULL)
-			continue;
 		res = alloc_bootmem_low(sizeof(struct resource));
 		switch (e820.map[i].type) {
 		case E820_RAM:	res->name = "System RAM"; break;
-- 
cgit v1.2.3-70-g09d2


From cee5dab4856f51c5cad3aecc630ad0a4d2217a85 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:43 -0700
Subject: [PATCH] kexec: x86: i8259 shutdown: disable interrupts

From: Eric W. Biederman <ebiederm@xmission.com>

This patch disables interrupt generation from the legacy pic on reboot.  Now
that there is a sys_device class it should not be called while drivers are
still using interrupts.

There is a report about this breaking ACPI power off on some systems.
http://bugme.osdl.org/show_bug.cgi?id=4041
However the final comment seems to exonerate this code.  So until
I get more information I believe that was a false positive.

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/i8259.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'arch')

diff --git a/arch/i386/kernel/i8259.c b/arch/i386/kernel/i8259.c
index 2c4813b47e5..178f4e9bac9 100644
--- a/arch/i386/kernel/i8259.c
+++ b/arch/i386/kernel/i8259.c
@@ -268,10 +268,22 @@ static int i8259A_suspend(struct sys_device *dev, pm_message_t state)
 	return 0;
 }
 
+static int i8259A_shutdown(struct sys_device *dev)
+{
+	/* Put the i8259A into a quiescent state that
+	 * the kernel initialization code can get it
+	 * out of.
+	 */
+	outb(0xff, 0x21);	/* mask all of 8259A-1 */
+	outb(0xff, 0xA1);	/* mask all of 8259A-1 */
+	return 0;
+}
+
 static struct sysdev_class i8259_sysdev_class = {
 	set_kset_name("i8259"),
 	.suspend = i8259A_suspend,
 	.resume = i8259A_resume,
+	.shutdown = i8259A_shutdown,
 };
 
 static struct sys_device device_i8259A = {
-- 
cgit v1.2.3-70-g09d2


From 719e711050482be667dafd39ec787859d353931c Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:43 -0700
Subject: [PATCH] kexec: x86_64: add i8259 shutdown method

From: Eric W. Biederman <ebiederm@xmission.com

The following patch simply adds a shutdown method to the x86_64 i8259 code.

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/i8259.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'arch')

diff --git a/arch/x86_64/kernel/i8259.c b/arch/x86_64/kernel/i8259.c
index 19eafa0aa95..a8916909512 100644
--- a/arch/x86_64/kernel/i8259.c
+++ b/arch/x86_64/kernel/i8259.c
@@ -414,10 +414,22 @@ static int i8259A_suspend(struct sys_device *dev, pm_message_t state)
 	return 0;
 }
 
+static int i8259A_shutdown(struct sys_device *dev)
+{
+	/* Put the i8259A into a quiescent state that
+	 * the kernel initialization code can get it
+	 * out of.
+	 */
+	outb(0xff, 0x21);	/* mask all of 8259A-1 */
+	outb(0xff, 0xA1);	/* mask all of 8259A-1 */
+	return 0;
+}
+
 static struct sysdev_class i8259_sysdev_class = {
 	set_kset_name("i8259"),
 	.suspend = i8259A_suspend,
 	.resume = i8259A_resume,
+	.shutdown = i8259A_shutdown,
 };
 
 static struct sys_device device_i8259A = {
-- 
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 'arch')

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 'arch')

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 ad0d75ebacdbf1004d004803df0ba371c6bdbe2a Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:47 -0700
Subject: [PATCH] kexec: x86: vmlinux: fix physical addresses

The vmlinux on i386 does not report the correct physical address of
the kernel.  Instead in the physical address field it currently
reports the virtual address of the kernel.

This is patch is a bug fix that corrects vmlinux to report the
proper physical addresses.

This is potentially a help for crash dump analysis tools.

This definitiely allows bootloaders that load vmlinux as a standard
ELF executable.  Bootloaders directly loading vmlinux become of
practical importance when we consider the kexec on panic case.

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/vmlinux.lds.S | 59 +++++++++++++++++++++++++++---------------
 1 file changed, 38 insertions(+), 21 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S
index e0512cc8bea..e17ab69c1f0 100644
--- a/arch/i386/kernel/vmlinux.lds.S
+++ b/arch/i386/kernel/vmlinux.lds.S
@@ -2,20 +2,23 @@
  * Written by Martin Mares <mj@atrey.karlin.mff.cuni.cz>;
  */
 
+#define LOAD_OFFSET __PAGE_OFFSET
+
 #include <asm-generic/vmlinux.lds.h>
 #include <asm/thread_info.h>
 #include <asm/page.h>
 
 OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386")
 OUTPUT_ARCH(i386)
-ENTRY(startup_32)
+ENTRY(phys_startup_32)
 jiffies = jiffies_64;
 SECTIONS
 {
-  . = __PAGE_OFFSET + 0x100000;
+  . = LOAD_OFFSET + 0x100000;
+  phys_startup_32 = startup_32 - LOAD_OFFSET;
   /* read-only */
   _text = .;			/* Text and read-only data */
-  .text : {
+  .text : AT(ADDR(.text) - LOAD_OFFSET) {
 	*(.text)
 	SCHED_TEXT
 	LOCK_TEXT
@@ -27,49 +30,55 @@ SECTIONS
 
   . = ALIGN(16);		/* Exception table */
   __start___ex_table = .;
-  __ex_table : { *(__ex_table) }
+  __ex_table : AT(ADDR(__ex_table) - LOAD_OFFSET) { *(__ex_table) }
   __stop___ex_table = .;
 
   RODATA
 
   /* writeable */
-  .data : {			/* Data */
+  .data : AT(ADDR(.data) - LOAD_OFFSET) {	/* Data */
 	*(.data)
 	CONSTRUCTORS
 	}
 
   . = ALIGN(4096);
   __nosave_begin = .;
-  .data_nosave : { *(.data.nosave) }
+  .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { *(.data.nosave) }
   . = ALIGN(4096);
   __nosave_end = .;
 
   . = ALIGN(4096);
-  .data.page_aligned : { *(.data.idt) }
+  .data.page_aligned : AT(ADDR(.data.page_aligned) - LOAD_OFFSET) {
+	*(.data.idt)
+  }
 
   . = ALIGN(32);
-  .data.cacheline_aligned : { *(.data.cacheline_aligned) }
+  .data.cacheline_aligned : AT(ADDR(.data.cacheline_aligned) - LOAD_OFFSET) {
+	*(.data.cacheline_aligned)
+  }
 
   _edata = .;			/* End of data section */
 
   . = ALIGN(THREAD_SIZE);	/* init_task */
-  .data.init_task : { *(.data.init_task) }
+  .data.init_task : AT(ADDR(.data.init_task) - LOAD_OFFSET) {
+	*(.data.init_task)
+  }
 
   /* will be freed after init */
   . = ALIGN(4096);		/* Init code and data */
   __init_begin = .;
-  .init.text : { 
+  .init.text : AT(ADDR(.init.text) - LOAD_OFFSET) {
 	_sinittext = .;
 	*(.init.text)
 	_einittext = .;
   }
-  .init.data : { *(.init.data) }
+  .init.data : AT(ADDR(.init.data) - LOAD_OFFSET) { *(.init.data) }
   . = ALIGN(16);
   __setup_start = .;
-  .init.setup : { *(.init.setup) }
+  .init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET) { *(.init.setup) }
   __setup_end = .;
   __initcall_start = .;
-  .initcall.init : {
+  .initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET) {
 	*(.initcall1.init) 
 	*(.initcall2.init) 
 	*(.initcall3.init) 
@@ -80,33 +89,41 @@ SECTIONS
   }
   __initcall_end = .;
   __con_initcall_start = .;
-  .con_initcall.init : { *(.con_initcall.init) }
+  .con_initcall.init : AT(ADDR(.con_initcall.init) - LOAD_OFFSET) {
+	*(.con_initcall.init)
+  }
   __con_initcall_end = .;
   SECURITY_INIT
   . = ALIGN(4);
   __alt_instructions = .;
-  .altinstructions : { *(.altinstructions) } 
+  .altinstructions : AT(ADDR(.altinstructions) - LOAD_OFFSET) {
+	*(.altinstructions)
+  }
   __alt_instructions_end = .; 
- .altinstr_replacement : { *(.altinstr_replacement) } 
+  .altinstr_replacement : AT(ADDR(.altinstr_replacement) - LOAD_OFFSET) {
+	*(.altinstr_replacement)
+  }
   /* .exit.text is discard at runtime, not link time, to deal with references
      from .altinstructions and .eh_frame */
-  .exit.text : { *(.exit.text) }
-  .exit.data : { *(.exit.data) }
+  .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET) { *(.exit.text) }
+  .exit.data : AT(ADDR(.exit.data) - LOAD_OFFSET) { *(.exit.data) }
   . = ALIGN(4096);
   __initramfs_start = .;
-  .init.ramfs : { *(.init.ramfs) }
+  .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) { *(.init.ramfs) }
   __initramfs_end = .;
   . = ALIGN(32);
   __per_cpu_start = .;
-  .data.percpu  : { *(.data.percpu) }
+  .data.percpu  : AT(ADDR(.data.percpu) - LOAD_OFFSET) { *(.data.percpu) }
   __per_cpu_end = .;
   . = ALIGN(4096);
   __init_end = .;
   /* freed after init ends here */
 	
   __bss_start = .;		/* BSS */
-  .bss : {
+  .bss.page_aligned : AT(ADDR(.bss.page_aligned) - LOAD_OFFSET) {
 	*(.bss.page_aligned)
+  }
+  .bss : AT(ADDR(.bss) - LOAD_OFFSET) {
 	*(.bss)
   }
   . = ALIGN(4);
-- 
cgit v1.2.3-70-g09d2


From 5ded01e83ec3b60191b03b9f88f53acd4e6112f5 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:48 -0700
Subject: [PATCH] kexec: x86_64: vmlinux: fix physical addresses

The vmlinux on x86_64 does not report the correct physical address of
the kernel.  Instead in the physical address field it currently
reports the virtual address of the kernel.

This is patch is a bug fix that corrects vmlinux to report the
proper physical addresses.

This is potentially a help for crash dump analysis tools.

This definitiely allows bootloaders that load vmlinux as a standard
ELF executable.  Bootloaders directly loading vmlinux become of
practical importance when we consider the kexec on panic case.

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/Makefile             |   2 +-
 arch/x86_64/kernel/vmlinux.lds.S | 128 ++++++++++++++++++++++++++-------------
 2 files changed, 87 insertions(+), 43 deletions(-)

(limited to 'arch')

diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile
index 6f90c246c41..8a73794f9b9 100644
--- a/arch/x86_64/Makefile
+++ b/arch/x86_64/Makefile
@@ -35,7 +35,7 @@ export IA32_CC IA32_LD IA32_AS IA32_OBJCOPY IA32_CPP
 
 LDFLAGS		:= -m elf_x86_64
 OBJCOPYFLAGS	:= -O binary -R .note -R .comment -S
-LDFLAGS_vmlinux := -e stext
+LDFLAGS_vmlinux :=
 
 CHECKFLAGS      += -D__x86_64__ -m64
 
diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S
index 59ebd5beda8..73389f51c4e 100644
--- a/arch/x86_64/kernel/vmlinux.lds.S
+++ b/arch/x86_64/kernel/vmlinux.lds.S
@@ -2,7 +2,10 @@
  * Written by Martin Mares <mj@atrey.karlin.mff.cuni.cz>;
  */
 
+#define LOAD_OFFSET __START_KERNEL_map
+
 #include <asm-generic/vmlinux.lds.h>
+#include <asm/page.h>
 #include <linux/config.h>
 
 OUTPUT_FORMAT("elf64-x86-64", "elf64-x86-64", "elf64-x86-64")
@@ -11,28 +14,30 @@ ENTRY(phys_startup_64)
 jiffies_64 = jiffies;
 SECTIONS
 {
-  . = 0xffffffff80100000;
+  . = __START_KERNEL;
   phys_startup_64 = startup_64 - LOAD_OFFSET;
   _text = .;			/* Text and read-only data */
-  .text : {
+  .text :  AT(ADDR(.text) - LOAD_OFFSET) {
 	*(.text)
 	SCHED_TEXT
 	LOCK_TEXT
 	*(.fixup)
 	*(.gnu.warning)
 	} = 0x9090
-  .text.lock : { *(.text.lock) }	/* out-of-line lock text */
+  				/* out-of-line lock text */
+  .text.lock : AT(ADDR(.text.lock) - LOAD_OFFSET) { *(.text.lock) }
 
   _etext = .;			/* End of text section */
 
   . = ALIGN(16);		/* Exception table */
   __start___ex_table = .;
-  __ex_table : { *(__ex_table) }
+  __ex_table : AT(ADDR(__ex_table) - LOAD_OFFSET) { *(__ex_table) }
   __stop___ex_table = .;
 
   RODATA
 
-  .data : {			/* Data */
+				/* Data */
+  .data : AT(ADDR(.data) - LOAD_OFFSET) {
 	*(.data)
 	CONSTRUCTORS
 	}
@@ -40,62 +45,95 @@ SECTIONS
   _edata = .;			/* End of data section */
 
   __bss_start = .;		/* BSS */
-  .bss : {
+  .bss : AT(ADDR(.bss) - LOAD_OFFSET) {
 	*(.bss.page_aligned)	
 	*(.bss)
 	}
   __bss_end = .;
 
+  . = ALIGN(PAGE_SIZE);
   . = ALIGN(CONFIG_X86_L1_CACHE_BYTES);
-  .data.cacheline_aligned : { *(.data.cacheline_aligned) }
+  .data.cacheline_aligned : AT(ADDR(.data.cacheline_aligned) - LOAD_OFFSET) {
+	*(.data.cacheline_aligned)
+  }
+
+#define VSYSCALL_ADDR (-10*1024*1024)
+#define VSYSCALL_PHYS_ADDR ((LOADADDR(.data.cacheline_aligned) + SIZEOF(.data.cacheline_aligned) + 4095) & ~(4095))
+#define VSYSCALL_VIRT_ADDR ((ADDR(.data.cacheline_aligned) + SIZEOF(.data.cacheline_aligned) + 4095) & ~(4095))
+
+#define VLOAD_OFFSET (VSYSCALL_ADDR - VSYSCALL_PHYS_ADDR)
+#define VLOAD(x) (ADDR(x) - VLOAD_OFFSET)
+
+#define VVIRT_OFFSET (VSYSCALL_ADDR - VSYSCALL_VIRT_ADDR)
+#define VVIRT(x) (ADDR(x) - VVIRT_OFFSET)
 
-#define AFTER(x)      BINALIGN(LOADADDR(x) + SIZEOF(x), 16)
-#define BINALIGN(x,y) (((x) + (y) - 1)  & ~((y) - 1))
-#define CACHE_ALIGN(x) BINALIGN(x, CONFIG_X86_L1_CACHE_BYTES)
+  . = VSYSCALL_ADDR;
+  .vsyscall_0 :	 AT(VSYSCALL_PHYS_ADDR) { *(.vsyscall_0) }
+  __vsyscall_0 = VSYSCALL_VIRT_ADDR;
 
-  .vsyscall_0 -10*1024*1024: AT ((LOADADDR(.data.cacheline_aligned) + SIZEOF(.data.cacheline_aligned) + 4095) & ~(4095)) { *(.vsyscall_0) }
-  __vsyscall_0 = LOADADDR(.vsyscall_0);
   . = ALIGN(CONFIG_X86_L1_CACHE_BYTES);
-  .xtime_lock : AT CACHE_ALIGN(AFTER(.vsyscall_0)) { *(.xtime_lock) }
-  xtime_lock = LOADADDR(.xtime_lock);
-  .vxtime : AT AFTER(.xtime_lock) { *(.vxtime) }
-  vxtime = LOADADDR(.vxtime);
-  .wall_jiffies : AT AFTER(.vxtime) { *(.wall_jiffies) }
-  wall_jiffies = LOADADDR(.wall_jiffies);
-  .sys_tz : AT AFTER(.wall_jiffies) { *(.sys_tz) }
-  sys_tz = LOADADDR(.sys_tz);
-  .sysctl_vsyscall : AT AFTER(.sys_tz) { *(.sysctl_vsyscall) }
-  sysctl_vsyscall = LOADADDR(.sysctl_vsyscall); 
-  .xtime : AT AFTER(.sysctl_vsyscall) { *(.xtime) }
-  xtime = LOADADDR(.xtime);
+  .xtime_lock : AT(VLOAD(.xtime_lock)) { *(.xtime_lock) }
+  xtime_lock = VVIRT(.xtime_lock);
+
+  .vxtime : AT(VLOAD(.vxtime)) { *(.vxtime) }
+  vxtime = VVIRT(.vxtime);
+
+  .wall_jiffies : AT(VLOAD(.wall_jiffies)) { *(.wall_jiffies) }
+  wall_jiffies = VVIRT(.wall_jiffies);
+
+  .sys_tz : AT(VLOAD(.sys_tz)) { *(.sys_tz) }
+  sys_tz = VVIRT(.sys_tz);
+
+  .sysctl_vsyscall : AT(VLOAD(.sysctl_vsyscall)) { *(.sysctl_vsyscall) }
+  sysctl_vsyscall = VVIRT(.sysctl_vsyscall);
+
+  .xtime : AT(VLOAD(.xtime)) { *(.xtime) }
+  xtime = VVIRT(.xtime);
+
   . = ALIGN(CONFIG_X86_L1_CACHE_BYTES);
-  .jiffies : AT CACHE_ALIGN(AFTER(.xtime)) { *(.jiffies) }
-  jiffies = LOADADDR(.jiffies);
-  .vsyscall_1 ADDR(.vsyscall_0) + 1024: AT (LOADADDR(.vsyscall_0) + 1024) { *(.vsyscall_1) }
-  . = LOADADDR(.vsyscall_0) + 4096;
+  .jiffies : AT(VLOAD(.jiffies)) { *(.jiffies) }
+  jiffies = VVIRT(.jiffies);
+
+  .vsyscall_1 ADDR(.vsyscall_0) + 1024: AT(VLOAD(.vsyscall_1)) { *(.vsyscall_1) }
+  .vsyscall_2 ADDR(.vsyscall_0) + 2048: AT(VLOAD(.vsyscall_2)) { *(.vsyscall_2) }
+  .vsyscall_3 ADDR(.vsyscall_0) + 3072: AT(VLOAD(.vsyscall_3)) { *(.vsyscall_3) }
+
+  . = VSYSCALL_VIRT_ADDR + 4096;
+
+#undef VSYSCALL_ADDR
+#undef VSYSCALL_PHYS_ADDR
+#undef VSYSCALL_VIRT_ADDR
+#undef VLOAD_OFFSET
+#undef VLOAD
+#undef VVIRT_OFFSET
+#undef VVIRT
 
   . = ALIGN(8192);		/* init_task */
-  .data.init_task : { *(.data.init_task) }
+  .data.init_task : AT(ADDR(.data.init_task) - LOAD_OFFSET) {
+	*(.data.init_task)
+  }
 
   . = ALIGN(4096);
-  .data.page_aligned : { *(.data.page_aligned) }
+  .data.page_aligned : AT(ADDR(.data.page_aligned) - LOAD_OFFSET) {
+	*(.data.page_aligned)
+  }
 
   . = ALIGN(4096);		/* Init code and data */
   __init_begin = .;
-  .init.text : { 
+  .init.text : AT(ADDR(.init.text) - LOAD_OFFSET) {
 	_sinittext = .;
 	*(.init.text)
 	_einittext = .;
   }
   __initdata_begin = .;
-  .init.data : { *(.init.data) }
+  .init.data : AT(ADDR(.init.data) - LOAD_OFFSET) { *(.init.data) }
   __initdata_end = .;
   . = ALIGN(16);
   __setup_start = .;
-  .init.setup : { *(.init.setup) }
+  .init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET) { *(.init.setup) }
   __setup_end = .;
   __initcall_start = .;
-  .initcall.init : {
+  .initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET) {
 	*(.initcall1.init) 
 	*(.initcall2.init) 
 	*(.initcall3.init) 
@@ -106,32 +144,38 @@ SECTIONS
   }
   __initcall_end = .;
   __con_initcall_start = .;
-  .con_initcall.init : { *(.con_initcall.init) }
+  .con_initcall.init : AT(ADDR(.con_initcall.init) - LOAD_OFFSET) {
+	*(.con_initcall.init)
+  }
   __con_initcall_end = .;
   SECURITY_INIT
   . = ALIGN(8);
   __alt_instructions = .;
-  .altinstructions : { *(.altinstructions) } 
+  .altinstructions : AT(ADDR(.altinstructions) - LOAD_OFFSET) {
+	*(.altinstructions)
+  }
   __alt_instructions_end = .; 
- .altinstr_replacement : { *(.altinstr_replacement) }
+  .altinstr_replacement : AT(ADDR(.altinstr_replacement) - LOAD_OFFSET) {
+	*(.altinstr_replacement)
+  }
   /* .exit.text is discard at runtime, not link time, to deal with references
      from .altinstructions and .eh_frame */
-  .exit.text : { *(.exit.text) }
-  .exit.data : { *(.exit.data) }	
+  .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET) { *(.exit.text) }
+  .exit.data : AT(ADDR(.exit.data) - LOAD_OFFSET) { *(.exit.data) }
   . = ALIGN(4096);
   __initramfs_start = .;
-  .init.ramfs : { *(.init.ramfs) }
+  .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) { *(.init.ramfs) }
   __initramfs_end = .;	
   . = ALIGN(32);
   __per_cpu_start = .;
-  .data.percpu  : { *(.data.percpu) }
+  .data.percpu  : AT(ADDR(.data.percpu) - LOAD_OFFSET) { *(.data.percpu) }
   __per_cpu_end = .;
   . = ALIGN(4096);
   __init_end = .;
 
   . = ALIGN(4096);
   __nosave_begin = .;
-  .data_nosave : { *(.data.nosave) }
+  .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { *(.data.nosave) }
   . = ALIGN(4096);
   __nosave_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 'arch')

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 8a9190853c34289d9181acd9c620c76143bf88ca Mon Sep 17 00:00:00 2001
From: Vivek Goyal <vgoyal@in.ibm.com>
Date: Sat, 25 Jun 2005 14:57:51 -0700
Subject: [PATCH] kexec: reserve Bootmem fix for booting nondefault location
 kernel

This patch fixes a problem with reserving memory during boot up of a kernel
built for non-default location.  Currently boot memory allocator reserves
the memory required by kernel image, boot allocaotor bitmap etc.  It
assumes that kernel is loaded at 1MB (HIGH_MEMORY hard coded to 1024*1024).
 But kernel can be built for non-default locatoin, hence existing
hardcoding will lead to reserving unnecessary memory.  This patch fixes it.

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 | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index f1ad9fdeaad..6337157f5df 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -1124,8 +1124,8 @@ void __init setup_bootmem_allocator(void)
 	 * the (very unlikely) case of us accidentally initializing the
 	 * bootmem allocator with an invalid RAM area.
 	 */
-	reserve_bootmem(HIGH_MEMORY, (PFN_PHYS(min_low_pfn) +
-			 bootmap_size + PAGE_SIZE-1) - (HIGH_MEMORY));
+	reserve_bootmem(__PHYSICAL_START, (PFN_PHYS(min_low_pfn) +
+			 bootmap_size + PAGE_SIZE-1) - (__PHYSICAL_START));
 
 	/*
 	 * reserve physical page 0 - it's a special BIOS page on many boxes,
-- 
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 'arch')

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 dd2a13054ffc25783a74afb5e4a0f2115e45f9cd Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:55 -0700
Subject: [PATCH] kexec: x86: factor out apic shutdown code

Factor out the apic and smp shutdown code from machine_restart so it can be
called by in the kexec reboot path as well.

By switching to the bootstrap cpu by default on reboot I can delete/simplify
some motherboard fixups well.

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/reboot.c | 82 +++++++++++++++--------------------------------
 1 file changed, 26 insertions(+), 56 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/reboot.c b/arch/i386/kernel/reboot.c
index db912209a8d..b3e58484996 100644
--- a/arch/i386/kernel/reboot.c
+++ b/arch/i386/kernel/reboot.c
@@ -26,7 +26,6 @@ static int reboot_mode;
 static int reboot_thru_bios;
 
 #ifdef CONFIG_SMP
-int reboot_smp = 0;
 static int reboot_cpu = -1;
 /* shamelessly grabbed from lib/vsprintf.c for readability */
 #define is_digit(c)	((c) >= '0' && (c) <= '9')
@@ -49,7 +48,6 @@ static int __init reboot_setup(char *str)
 			break;
 #ifdef CONFIG_SMP
 		case 's': /* "smp" reboot by executing reset on BSP or other CPU*/
-			reboot_smp = 1;
 			if (is_digit(*(str+1))) {
 				reboot_cpu = (int) (*(str+1) - '0');
 				if (is_digit(*(str+2))) 
@@ -88,33 +86,9 @@ static int __init set_bios_reboot(struct dmi_system_id *d)
 	return 0;
 }
 
-/*
- * Some machines require the "reboot=s"  commandline option, this quirk makes that automatic.
- */
-static int __init set_smp_reboot(struct dmi_system_id *d)
-{
-#ifdef CONFIG_SMP
-	if (!reboot_smp) {
-		reboot_smp = 1;
-		printk(KERN_INFO "%s series board detected. Selecting SMP-method for reboots.\n", d->ident);
-	}
-#endif
-	return 0;
-}
-
-/*
- * Some machines require the "reboot=b,s"  commandline option, this quirk makes that automatic.
- */
-static int __init set_smp_bios_reboot(struct dmi_system_id *d)
-{
-	set_smp_reboot(d);
-	set_bios_reboot(d);
-	return 0;
-}
-
 static struct dmi_system_id __initdata reboot_dmi_table[] = {
 	{	/* Handle problems with rebooting on Dell 1300's */
-		.callback = set_smp_bios_reboot,
+		.callback = set_bios_reboot,
 		.ident = "Dell PowerEdge 1300",
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "Dell Computer Corporation"),
@@ -301,41 +275,32 @@ void machine_real_restart(unsigned char *code, int length)
 EXPORT_SYMBOL(machine_real_restart);
 #endif
 
-void machine_restart(char * __unused)
+void machine_shutdown(void)
 {
 #ifdef CONFIG_SMP
-	int cpuid;
-	
-	cpuid = GET_APIC_ID(apic_read(APIC_ID));
-
-	if (reboot_smp) {
-
-		/* check to see if reboot_cpu is valid 
-		   if its not, default to the BSP */
-		if ((reboot_cpu == -1) ||  
-		      (reboot_cpu > (NR_CPUS -1))  || 
-		      !physid_isset(cpuid, phys_cpu_present_map))
-			reboot_cpu = boot_cpu_physical_apicid;
-
-		reboot_smp = 0;  /* use this as a flag to only go through this once*/
-		/* re-run this function on the other CPUs
-		   it will fall though this section since we have 
-		   cleared reboot_smp, and do the reboot if it is the
-		   correct CPU, otherwise it halts. */
-		if (reboot_cpu != cpuid)
-			smp_call_function((void *)machine_restart , NULL, 1, 0);
+	int reboot_cpu_id;
+
+	/* The boot cpu is always logical cpu 0 */
+	reboot_cpu_id = 0;
+
+	/* See if there has been given a command line override */
+	if ((reboot_cpu_id != -1) && (reboot_cpu < NR_CPUS) &&
+		cpu_isset(reboot_cpu, cpu_online_map)) {
+		reboot_cpu_id = reboot_cpu;
 	}
 
-	/* if reboot_cpu is still -1, then we want a tradional reboot, 
-	   and if we are not running on the reboot_cpu,, halt */
-	if ((reboot_cpu != -1) && (cpuid != reboot_cpu)) {
-		for (;;)
-		__asm__ __volatile__ ("hlt");
+	/* Make certain the cpu I'm rebooting on is online */
+	if (!cpu_isset(reboot_cpu_id, cpu_online_map)) {
+		reboot_cpu_id = smp_processor_id();
 	}
-	/*
-	 * Stop all CPUs and turn off local APICs and the IO-APIC, so
-	 * other OSs see a clean IRQ state.
+
+	/* Make certain I only run on the appropriate processor */
+	set_cpus_allowed(current, cpumask_of_cpu(reboot_cpu_id));
+
+	/* O.K. Now that I'm on the appropriate processor, stop
+	 * all of the others, and disable their local APICs.
 	 */
+
 	smp_send_stop();
 #endif /* CONFIG_SMP */
 
@@ -344,6 +309,11 @@ void machine_restart(char * __unused)
 #ifdef CONFIG_X86_IO_APIC
 	disable_IO_APIC();
 #endif
+}
+
+void machine_restart(char * __unused)
+{
+	machine_shutdown();
 
 	if (!reboot_thru_bios) {
 		if (efi_enabled) {
-- 
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 'arch')

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 c4ac4263a019c791e906f284bb03891d3c25a845 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:58 -0700
Subject: [PATCH] crashdump: x86: add NMI handler to capture other CPUs

One of the dangers when switching from one kernel to another is what happens
to all of the other cpus that were running in the crashed kernel.  In an
attempt to avoid that problem this patch adds a nmi handler and attempts to
shoot down the other cpus by sending them non maskable interrupts.

The code then waits for 1 second or until all known cpus have stopped running
and then jumps from the running kernel that has crashed to the kernel in
reserved memory.

The kernel spin loop is used for the delay as that should behave continue to
be safe even in after a crash.

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/crash.c | 56 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 56 insertions(+)

(limited to 'arch')

diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c
index fa27a6c2abb..882779c0787 100644
--- a/arch/i386/kernel/crash.c
+++ b/arch/i386/kernel/crash.c
@@ -23,12 +23,65 @@
 #include <asm/hardirq.h>
 #include <asm/nmi.h>
 #include <asm/hw_irq.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];
 
+#ifdef CONFIG_SMP
+static atomic_t waiting_for_crash_ipi;
+
+static int crash_nmi_callback(struct pt_regs *regs, int cpu)
+{
+	local_irq_disable();
+	atomic_dec(&waiting_for_crash_ipi);
+	/* Assume hlt works */
+	__asm__("hlt");
+	for(;;);
+	return 1;
+}
+
+/*
+ * By using the NMI code instead of a vector we just sneak thru the
+ * word generator coming out with just what we want.  AND it does
+ * not matter if clustered_apic_mode is set or not.
+ */
+static void smp_send_nmi_allbutself(void)
+{
+	send_IPI_allbutself(APIC_DM_NMI);
+}
+
+static void nmi_shootdown_cpus(void)
+{
+	unsigned long msecs;
+	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
+	 * out the NMI
+	 */
+	wmb();
+
+	smp_send_nmi_allbutself();
+
+	msecs = 1000; /* Wait at most a second for the other cpus to stop */
+	while ((atomic_read(&waiting_for_crash_ipi) > 0) && msecs) {
+		mdelay(1);
+		msecs--;
+	}
+
+	/* Leave the nmi callback set */
+}
+#else
+static void nmi_shootdown_cpus(void)
+{
+	/* There are no cpus to shootdown */
+}
+#endif
+
 void machine_crash_shutdown(void)
 {
 	/* This function is only called after the system
@@ -39,4 +92,7 @@ void machine_crash_shutdown(void)
 	 * In practice this means shooting down the other cpus in
 	 * an SMP system.
 	 */
+	/* The kernel is broken so disable interrupts */
+	local_irq_disable();
+	nmi_shootdown_cpus();
 }
-- 
cgit v1.2.3-70-g09d2


From 2c818b45a202b8f632b4031edcba8599efda42b7 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:57:59 -0700
Subject: [PATCH] kexec: x86: snapshot registers during crash shutdown

After the kernel panics if we wish to generate an entire machine core file it
is very nice to know the register state at the time the machine crashed.

After long discussion it was realized that if you are going to be saving the
information anyway it is reasonable to store the information in a format that
it will be used and recognized in so the register state is stored in the
standard ELF note format.

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/crash.c | 80 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 80 insertions(+)

(limited to 'arch')

diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c
index 882779c0787..1bb5dd98d1d 100644
--- a/arch/i386/kernel/crash.c
+++ b/arch/i386/kernel/crash.c
@@ -30,12 +30,91 @@ typedef u32 note_buf_t[MAX_NOTE_BYTES/4];
 
 note_buf_t crash_notes[NR_CPUS];
 
+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;
+	memcpy(buf, &note, sizeof(note));
+	buf += (sizeof(note) +3)/4;
+	memcpy(buf, name, note.n_namesz);
+	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)) {
+		return;
+	}
+	/* Using ELF notes here is opportunistic.
+	 * I need a well defined structure format
+	 * for the data I pass, and I need tags
+	 * on the data to indicate what information I have
+	 * squirrelled away.  ELF notes happen to provide
+	 * all of that that no need to invent something new.
+	 */
+	buf = &crash_notes[cpu][0];
+	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));
+
+	final_note(buf);
+}
+
+static void crash_get_current_regs(struct pt_regs *regs)
+{
+	__asm__ __volatile__("movl %%ebx,%0" : "=m"(regs->ebx));
+	__asm__ __volatile__("movl %%ecx,%0" : "=m"(regs->ecx));
+	__asm__ __volatile__("movl %%edx,%0" : "=m"(regs->edx));
+	__asm__ __volatile__("movl %%esi,%0" : "=m"(regs->esi));
+	__asm__ __volatile__("movl %%edi,%0" : "=m"(regs->edi));
+	__asm__ __volatile__("movl %%ebp,%0" : "=m"(regs->ebp));
+	__asm__ __volatile__("movl %%eax,%0" : "=m"(regs->eax));
+	__asm__ __volatile__("movl %%esp,%0" : "=m"(regs->esp));
+	__asm__ __volatile__("movw %%ss, %%ax;" :"=a"(regs->xss));
+	__asm__ __volatile__("movw %%cs, %%ax;" :"=a"(regs->xcs));
+	__asm__ __volatile__("movw %%ds, %%ax;" :"=a"(regs->xds));
+	__asm__ __volatile__("movw %%es, %%ax;" :"=a"(regs->xes));
+	__asm__ __volatile__("pushfl; popl %0" :"=m"(regs->eflags));
+
+	regs->eip = (unsigned long)current_text_addr();
+}
+
+static void crash_save_self(void)
+{
+	struct pt_regs regs;
+	int cpu;
+	cpu = smp_processor_id();
+	crash_get_current_regs(&regs);
+	crash_save_this_cpu(&regs, cpu);
+}
+
 #ifdef CONFIG_SMP
 static atomic_t waiting_for_crash_ipi;
 
 static int crash_nmi_callback(struct pt_regs *regs, int cpu)
 {
 	local_irq_disable();
+	crash_save_this_cpu(regs, cpu);
 	atomic_dec(&waiting_for_crash_ipi);
 	/* Assume hlt works */
 	__asm__("hlt");
@@ -95,4 +174,5 @@ void machine_crash_shutdown(void)
 	/* The kernel is broken so disable interrupts */
 	local_irq_disable();
 	nmi_shootdown_cpus();
+	crash_save_self();
 }
-- 
cgit v1.2.3-70-g09d2


From 63d30298efc387c72557d11e2a7b467554c05a64 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:58:00 -0700
Subject: [PATCH] kexec: x86 shutdown APICs during crash_shutdown

In the case of a crash/panic an architecture specific function
machine_crash_shutdown is called.  This patch adds to the x86 machine_crash
function the standard kernel code for shutting down apics.

Every line of code added to that function increases the risk that we will call
code after a kernel panic that is not safe.

This patch should not make it to the stable kernel without a being reviewed a
lot more.  It is unclear how much a hardned kernel can take when it comes to
misconfigured apics.  So since a normal kernel has problems this patch does a
clean shutdown.

It is my expectation this patch will be dropped from future generations of the
kexec work.  But for the moment it is a crutch to keep from breaking
everything.

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/crash.c | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'arch')

diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c
index 1bb5dd98d1d..59b92d21746 100644
--- a/arch/i386/kernel/crash.c
+++ b/arch/i386/kernel/crash.c
@@ -23,6 +23,7 @@
 #include <asm/hardirq.h>
 #include <asm/nmi.h>
 #include <asm/hw_irq.h>
+#include <asm/apic.h>
 #include <mach_ipi.h>
 
 #define MAX_NOTE_BYTES 1024
@@ -115,6 +116,7 @@ static int crash_nmi_callback(struct pt_regs *regs, int cpu)
 {
 	local_irq_disable();
 	crash_save_this_cpu(regs, cpu);
+	disable_local_APIC();
 	atomic_dec(&waiting_for_crash_ipi);
 	/* Assume hlt works */
 	__asm__("hlt");
@@ -153,6 +155,7 @@ static void nmi_shootdown_cpus(void)
 	}
 
 	/* Leave the nmi callback set */
+	disable_local_APIC();
 }
 #else
 static void nmi_shootdown_cpus(void)
@@ -174,5 +177,9 @@ void machine_crash_shutdown(void)
 	/* The kernel is broken so disable interrupts */
 	local_irq_disable();
 	nmi_shootdown_cpus();
+	lapic_shutdown();
+#if defined(CONFIG_X86_IO_APIC)
+	disable_IO_APIC();
+#endif
 	crash_save_self();
 }
-- 
cgit v1.2.3-70-g09d2


From 1bc3b91aeed71a904e431d12ca90e9b6bcb42c91 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:58:01 -0700
Subject: [PATCH] crashdump: x86 crashkernel option

This is the x86 implementation of the crashkernel option.  It reserves a
window of memory very early in the bootup process, so we never use it for
anything but the kernel to switch to when the running kernel panics.

In addition to reserving this memory a resource structure is registered so
looking at /proc/iomem it is clear what happened to that memory.

ISSUES:
Is it possible to implement this in a architecture generic way?
What should be done with architectures that always use an iommu and
thus don't report their RAM memory resources in /proc/iomem?

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/kernel/efi.c   |  4 ++++
 arch/i386/kernel/setup.c | 32 ++++++++++++++++++++++++++++++++
 arch/i386/mm/discontig.c |  2 ++
 3 files changed, 38 insertions(+)

(limited to 'arch')

diff --git a/arch/i386/kernel/efi.c b/arch/i386/kernel/efi.c
index f732f427b41..385883ea8c1 100644
--- a/arch/i386/kernel/efi.c
+++ b/arch/i386/kernel/efi.c
@@ -30,6 +30,7 @@
 #include <linux/ioport.h>
 #include <linux/module.h>
 #include <linux/efi.h>
+#include <linux/kexec.h>
 
 #include <asm/setup.h>
 #include <asm/io.h>
@@ -598,6 +599,9 @@ efi_initialize_iomem_resources(struct resource *code_resource,
 		if (md->type == EFI_CONVENTIONAL_MEMORY) {
 			request_resource(res, code_resource);
 			request_resource(res, data_resource);
+#ifdef CONFIG_KEXEC
+			request_resource(res, &crashk_res);
+#endif
 		}
 	}
 }
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index 6337157f5df..d88ebdfa6cc 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -43,7 +43,10 @@
 #include <linux/init.h>
 #include <linux/edd.h>
 #include <linux/nodemask.h>
+#include <linux/kexec.h>
+
 #include <video/edid.h>
+
 #include <asm/apic.h>
 #include <asm/e820.h>
 #include <asm/mpspec.h>
@@ -846,6 +849,27 @@ static void __init parse_cmdline_early (char ** cmdline_p)
 			lapic_disable();
 #endif /* CONFIG_X86_LOCAL_APIC */
 
+#ifdef CONFIG_KEXEC
+		/* crashkernel=size@addr specifies the location to reserve for
+		 * a crash kernel.  By reserving this memory we guarantee
+		 * that linux never set's it up as a DMA target.
+		 * Useful for holding code to do something appropriate
+		 * after a kernel panic.
+		 */
+		else if (!memcmp(from, "crashkernel=", 12)) {
+			unsigned long size, base;
+			size = memparse(from+12, &from);
+			if (*from == '@') {
+				base = memparse(from+1, &from);
+				/* FIXME: Do I want a sanity check
+				 * to validate the memory range?
+				 */
+				crashk_res.start = base;
+				crashk_res.end   = base + size - 1;
+			}
+		}
+#endif
+
 		/*
 		 * highmem=size forces highmem to be exactly 'size' bytes.
 		 * This works even on boxes that have no highmem otherwise.
@@ -1181,6 +1205,11 @@ void __init setup_bootmem_allocator(void)
 		}
 	}
 #endif
+#ifdef CONFIG_KEXEC
+	if (crashk_res.start != crashk_res.end)
+		reserve_bootmem(crashk_res.start,
+			crashk_res.end - crashk_res.start + 1);
+#endif
 }
 
 /*
@@ -1234,6 +1263,9 @@ legacy_init_iomem_resources(struct resource *code_resource, struct resource *dat
 			 */
 			request_resource(res, code_resource);
 			request_resource(res, data_resource);
+#ifdef CONFIG_KEXEC
+			request_resource(res, &crashk_res);
+#endif
 		}
 	}
 }
diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c
index f429c871e84..b358f0702a4 100644
--- a/arch/i386/mm/discontig.c
+++ b/arch/i386/mm/discontig.c
@@ -30,6 +30,8 @@
 #include <linux/initrd.h>
 #include <linux/nodemask.h>
 #include <linux/module.h>
+#include <linux/kexec.h>
+
 #include <asm/e820.h>
 #include <asm/setup.h>
 #include <asm/mmzone.h>
-- 
cgit v1.2.3-70-g09d2


From d89559589a588d1a654329d8cd9a3ad33aaad9be 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: factor out apic shutdown code

Factor out the apic and smp shutdown code from machine_restart so it can be
called by in the kexec reboot path as well.

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/reboot.c | 62 ++++++++++++++++++++++++---------------------
 1 file changed, 33 insertions(+), 29 deletions(-)

(limited to 'arch')

diff --git a/arch/x86_64/kernel/reboot.c b/arch/x86_64/kernel/reboot.c
index be4b36f762c..57e71dbdfd6 100644
--- a/arch/x86_64/kernel/reboot.c
+++ b/arch/x86_64/kernel/reboot.c
@@ -66,41 +66,47 @@ static int __init reboot_setup(char *str)
 
 __setup("reboot=", reboot_setup);
 
-#ifdef CONFIG_SMP
-static void smp_halt(void)
+static inline void kb_wait(void)
 {
-	int cpuid = safe_smp_processor_id(); 
-	static int first_entry = 1;
+	int i;
 
-	if (reboot_force)
-		return;
+	for (i=0; i<0x10000; i++)
+		if ((inb_p(0x64) & 0x02) == 0)
+			break;
+}
 
-	if (first_entry) {
-		first_entry = 0;
-		smp_call_function((void *)machine_restart, NULL, 1, 0);
-	}
-			
-	smp_stop_cpu(); 
+void machine_shutdown(void)
+{
+	/* Stop the cpus and apics */
+#ifdef CONFIG_SMP
+	int reboot_cpu_id;
 
-	/* AP calling this. Just halt */
-	if (cpuid != boot_cpu_id) { 
-		for (;;) 
-			asm("hlt");
+	/* The boot cpu is always logical cpu 0 */
+	reboot_cpu_id = 0;
+
+	/* Make certain the cpu I'm about to reboot on is online */
+	if (!cpu_isset(reboot_cpu_id, cpu_online_map)) {
+		reboot_cpu_id = smp_processor_id();
 	}
 
-	/* Wait for all other CPUs to have run smp_stop_cpu */
-	while (!cpus_empty(cpu_online_map))
-		rep_nop(); 
-}
+	/* Make certain I only run on the appropriate processor */
+	set_cpus_allowed(current, cpumask_of_cpu(reboot_cpu_id));
+
+	/* O.K Now that I'm on the appropriate processor,
+	 * stop all of the others.
+	 */
+	smp_send_stop();
 #endif
 
-static inline void kb_wait(void)
-{
-	int i;
+	local_irq_disable();
 
-	for (i=0; i<0x10000; i++)
-		if ((inb_p(0x64) & 0x02) == 0)
-			break;
+#ifndef CONFIG_SMP
+	disable_local_APIC();
+#endif
+
+	disable_IO_APIC();
+
+	local_irq_enable();
 }
 
 void machine_restart(char * __unused)
@@ -109,9 +115,7 @@ void machine_restart(char * __unused)
 
 	printk("machine restart\n");
 
-#ifdef CONFIG_SMP
-	smp_halt(); 
-#endif
+	machine_shutdown();
 
 	if (!reboot_force) {
 		local_irq_disable();
-- 
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 'arch')

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 5f5609df0c943b005847d3b2765d6dd47b624011 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 25 Jun 2005 14:58:04 -0700
Subject: [PATCH] crashdump: x86_64: crashkernel option

This is the x86_64 implementation of the crashkernel option.  It reserves
a window of memory very early in the bootup process, so we never use
it for anything but the kernel to switch to when the running
kernel panics.

In addition to reserving this memory a resource structure is registered
so looking at /proc/iomem it is clear what happened to that memory.

ISSUES:
Is it possible to implement this in a architecture generic way?
What should be done with architectures that always use an iommu and
thus don't report their RAM memory resources in /proc/iomem?

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/e820.c  |  4 ++++
 arch/x86_64/kernel/setup.c | 29 +++++++++++++++++++++++++++++
 2 files changed, 33 insertions(+)

(limited to 'arch')

diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index cf4878309b0..6ded3a50dfe 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -16,6 +16,7 @@
 #include <linux/bootmem.h>
 #include <linux/ioport.h>
 #include <linux/string.h>
+#include <linux/kexec.h>
 #include <asm/page.h>
 #include <asm/e820.h>
 #include <asm/proto.h>
@@ -210,6 +211,9 @@ void __init e820_reserve_resources(void)
 			 */
 			request_resource(res, &code_resource);
 			request_resource(res, &data_resource);
+#ifdef CONFIG_KEXEC
+			request_resource(res, &crashk_res);
+#endif
 		}
 	}
 }
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index ca9a915a6d4..b02d921da4f 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -41,6 +41,7 @@
 #include <linux/kallsyms.h>
 #include <linux/edd.h>
 #include <linux/mmzone.h>
+#include <linux/kexec.h>
 
 #include <asm/mtrr.h>
 #include <asm/uaccess.h>
@@ -367,6 +368,27 @@ static __init void parse_cmdline_early (char ** cmdline_p)
 		if (!memcmp(from, "noexec=", 7))
 			nonx_setup(from + 7);
 
+#ifdef CONFIG_KEXEC
+		/* crashkernel=size@addr specifies the location to reserve for
+		 * a crash kernel.  By reserving this memory we guarantee
+		 * that linux never set's it up as a DMA target.
+		 * Useful for holding code to do something appropriate
+		 * after a kernel panic.
+		 */
+		else if (!memcmp(from, "crashkernel=", 12)) {
+			unsigned long size, base;
+			size = memparse(from+12, &from);
+			if (*from == '@') {
+				base = memparse(from+1, &from);
+				/* FIXME: Do I want a sanity check
+				 * to validate the memory range?
+				 */
+				crashk_res.start = base;
+				crashk_res.end   = base + size - 1;
+			}
+		}
+#endif
+
 	next_char:
 		c = *(from++);
 		if (!c)
@@ -625,6 +647,13 @@ void __init setup_arch(char **cmdline_p)
 #endif
 
 	sparse_init();
+
+#ifdef CONFIG_KEXEC
+	if (crashk_res.start != crashk_res.end) {
+		reserve_bootmem(crashk_res.start,
+			crashk_res.end - crashk_res.start + 1);
+	}
+#endif
 	paging_init();
 
 	check_ioapic();
-- 
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 'arch')

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 'arch')

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 'arch')

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 'arch')

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 'arch')

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 4d55476c3f889e0f30d88e22da4682b5f10394ff Mon Sep 17 00:00:00 2001
From: Vivek Goyal <vgoyal@in.ibm.com>
Date: Sat, 25 Jun 2005 14:58:13 -0700
Subject: [PATCH] kdump: NMI handler segment selector, stack pointer fix

CPU does not save ss and esp on stack if execution was already in kernel mode
at the time of NMI occurrence.  This leads to saving of erractic values for ss
and esp.  This patch fixes the issue.

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/crash.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

(limited to 'arch')

diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c
index 3645ad7ac20..31e077bb0ca 100644
--- a/arch/i386/kernel/crash.c
+++ b/arch/i386/kernel/crash.c
@@ -112,7 +112,20 @@ static atomic_t waiting_for_crash_ipi;
 
 static int crash_nmi_callback(struct pt_regs *regs, int cpu)
 {
+	struct pt_regs fixed_regs;
 	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));
+		regs = &fixed_regs;
+	}
 	crash_save_this_cpu(regs, cpu);
 	disable_local_APIC();
 	atomic_dec(&waiting_for_crash_ipi);
-- 
cgit v1.2.3-70-g09d2


From a3ea8ac8468f5c7fc65684331dfba3260d5b2d93 Mon Sep 17 00:00:00 2001
From: Vivek Goyal <vgoyal@in.ibm.com>
Date: Sat, 25 Jun 2005 14:58:14 -0700
Subject: [PATCH] Kexec: Kexec on panic fix with nmi watchdog enabled

o Problem: Kexec on panic hangs if first kernel is booted with nmi_watchdog
  command line parameter. This problem occurs because kexec crash shutdown
  code replaces the NMI callback handler. This handler saves the cpu register
  states and halts the cpu. If system is booted with nmi_watchdog parameter,
  then crashing cpu also runs this nmi handler and halts itself.

o This patch fixes the problem by keeping a track of crashing cpu and not
  executing the new nmi handler on crashing cpu.

o There is a dependence on smp_processor_id() function which might return
  insane value for cpu, if cpu field of thread_info is corrupted.

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/crash.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'arch')

diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c
index 31e077bb0ca..a021681d21f 100644
--- a/arch/i386/kernel/crash.c
+++ b/arch/i386/kernel/crash.c
@@ -28,6 +28,8 @@
 
 
 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)
@@ -113,6 +115,13 @@ static atomic_t waiting_for_crash_ipi;
 static int crash_nmi_callback(struct pt_regs *regs, int cpu)
 {
 	struct pt_regs fixed_regs;
+
+	/* Don't do anything if this handler is invoked on crashing cpu.
+	 * Otherwise, system will completely hang. Crashing cpu can get
+	 * an NMI if system was initially booted with nmi_watchdog parameter.
+	 */
+	if (cpu == crashing_cpu)
+		return 1;
 	local_irq_disable();
 
 	/* CPU does not save ss and esp on stack if execution is already
@@ -187,6 +196,9 @@ void machine_crash_shutdown(void)
 	 */
 	/* The kernel is broken so disable interrupts */
 	local_irq_disable();
+
+	/* Make a note of crashing cpu. Will be used in NMI callback.*/
+	crashing_cpu = smp_processor_id();
 	nmi_shootdown_cpus();
 	lapic_shutdown();
 #if defined(CONFIG_X86_IO_APIC)
-- 
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 'arch')

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 5f016456c96868c27df248a54d1cc919e7b70a23 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: Kconfig

- config option CONFIG_CRASH_DUMP

- Made it dependent on HIGHMEM.  This is required as capture kernel treats
  the previous kernel's memory as high memmory and stitches a PTE for
  accessing it.

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/Kconfig | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'arch')

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index 0f391cbf116..6c02336fe2e 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -970,6 +970,13 @@ config KEXEC
 	  support.  As of this writing the exact hardware interface is
 	  strongly in flux, so no good recommendation can be made.
 
+config CRASH_DUMP
+	bool "kernel crash dumps (EXPERIMENTAL)"
+	depends on EMBEDDED
+	depends on EXPERIMENTAL
+	depends on HIGHMEM
+	help
+	  Generate crash dump after being started by kexec.
 endmenu
 
 
-- 
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 'arch')

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 'arch')

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 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 'arch')

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 4f339ecb30c759f94a29992d4635d9194132b6cf Mon Sep 17 00:00:00 2001
From: Alexander Nyberg <alexn@telia.com>
Date: Sat, 25 Jun 2005 14:58:27 -0700
Subject: [PATCH] kdump: Save trap information for later analysis

If we are faulting in kernel it is quite possible this will lead to a
panic.  Save trap number, cr2 (in case of page fault) and error_code in the
current thread (these fields already exist for signal delivery but are not
used here).

This helps later kdump crash analyzing from user-space (a script has been
submitted to dig this info out in gdb).

Signed-off-by: Alexander Nyberg <alexn@telia.com>
Cc: <fastboot@lists.osdl.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/traps.c | 12 ++++++++----
 arch/i386/mm/fault.c     |  3 +++
 2 files changed, 11 insertions(+), 4 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index e458463ebc0..c6077a5ef7c 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -369,6 +369,10 @@ static inline void die_if_kernel(const char * str, struct pt_regs * regs, long e
 static void do_trap(int trapnr, int signr, char *str, int vm86,
 			   struct pt_regs * regs, long error_code, siginfo_t *info)
 {
+	struct task_struct *tsk = current;
+	tsk->thread.error_code = error_code;
+	tsk->thread.trap_no = trapnr;
+
 	if (regs->eflags & VM_MASK) {
 		if (vm86)
 			goto vm86_trap;
@@ -379,9 +383,6 @@ static void do_trap(int trapnr, int signr, char *str, int vm86,
 		goto kernel_trap;
 
 	trap_signal: {
-		struct task_struct *tsk = current;
-		tsk->thread.error_code = error_code;
-		tsk->thread.trap_no = trapnr;
 		if (info)
 			force_sig_info(signr, info, tsk);
 		else
@@ -494,6 +495,9 @@ fastcall void do_general_protection(struct pt_regs * regs, long error_code)
 	}
 	put_cpu();
 
+	current->thread.error_code = error_code;
+	current->thread.trap_no = 13;
+
 	if (regs->eflags & VM_MASK)
 		goto gp_in_vm86;
 
@@ -897,9 +901,9 @@ fastcall void do_simd_coprocessor_error(struct pt_regs * regs,
 					  error_code);
 			return;
 		}
-		die_if_kernel("cache flush denied", regs, error_code);
 		current->thread.trap_no = 19;
 		current->thread.error_code = error_code;
+		die_if_kernel("cache flush denied", regs, error_code);
 		force_sig(SIGSEGV, current);
 	}
 }
diff --git a/arch/i386/mm/fault.c b/arch/i386/mm/fault.c
index a509237c481..92ed6c0a55d 100644
--- a/arch/i386/mm/fault.c
+++ b/arch/i386/mm/fault.c
@@ -463,6 +463,9 @@ no_context:
 		printk(KERN_ALERT "*pte = %08lx\n", page);
 	}
 #endif
+	tsk->thread.cr2 = address;
+	tsk->thread.trap_no = 14;
+	tsk->thread.error_code = error_code;
 	die("Oops", regs, error_code);
 	bust_spinlocks(0);
 	do_exit(SIGKILL);
-- 
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 'arch')

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 09dbb4768c962ac00f9ff416328a7e6bfbd92b21 Mon Sep 17 00:00:00 2001
From: randy_dunlap <rdunlap@xenotime.net>
Date: Sat, 25 Jun 2005 14:58:39 -0700
Subject: [PATCH] x86-64: add memcpy/memset prototypes

Put function prototypes for memset() and memcpy() ahead of where
there are used, to kill sparse warnings:

arch/x86_64/boot/compressed/../../../../lib/inflate.c:317:3: warning: undefined identifier 'memset'
arch/x86_64/boot/compressed/../../../../lib/inflate.c:601:11: warning: undefined identifier 'memcpy'
arch/x86_64/boot/compressed/misc.c:151:2: warning: undefined identifier 'memcpy'
arch/x86_64/boot/compressed/../../../../lib/inflate.c:317:3: warning: call with no type!
arch/x86_64/boot/compressed/../../../../lib/inflate.c:601:17: warning: call with no type!
arch/x86_64/boot/compressed/misc.c:151:9: warning: call with no type!

Signed-off-by: randy_dunlap <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/boot/compressed/misc.c | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'arch')

diff --git a/arch/x86_64/boot/compressed/misc.c b/arch/x86_64/boot/compressed/misc.c
index c43cedb5c1d..f992aef141b 100644
--- a/arch/x86_64/boot/compressed/misc.c
+++ b/arch/x86_64/boot/compressed/misc.c
@@ -93,6 +93,9 @@ static unsigned long output_ptr = 0;
 static void *malloc(int size);
 static void free(void *where);
  
+void* memset(void* s, int c, unsigned n);
+void* memcpy(void* dest, const void* src, unsigned n);
+
 static void putstr(const char *);
   
 extern int end;
-- 
cgit v1.2.3-70-g09d2


From 3f3ae3471f4c5921274c3869eef21b5611e85b50 Mon Sep 17 00:00:00 2001
From: Domen Puncer <domen@coderock.org>
Date: Sat, 25 Jun 2005 14:58:44 -0700
Subject: [PATCH] arch/i386/kernel/traps.c: fix sparse warnings

Signed-off-by: Alexey Dobriyan <adobriyan@mail.ru>
Signed-off-by: Domen Puncer <domen@coderock.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/traps.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index c6077a5ef7c..a61f33d06ea 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -235,22 +235,22 @@ void show_registers(struct pt_regs *regs)
 	 * time of the fault..
 	 */
 	if (in_kernel) {
-		u8 *eip;
+		u8 __user *eip;
 
 		printk("\nStack: ");
 		show_stack(NULL, (unsigned long*)esp);
 
 		printk("Code: ");
 
-		eip = (u8 *)regs->eip - 43;
+		eip = (u8 __user *)regs->eip - 43;
 		for (i = 0; i < 64; i++, eip++) {
 			unsigned char c;
 
-			if (eip < (u8 *)PAGE_OFFSET || __get_user(c, eip)) {
+			if (eip < (u8 __user *)PAGE_OFFSET || __get_user(c, eip)) {
 				printk(" Bad EIP value.");
 				break;
 			}
-			if (eip == (u8 *)regs->eip)
+			if (eip == (u8 __user *)regs->eip)
 				printk("<%02x> ", c);
 			else
 				printk("%02x ", c);
@@ -274,13 +274,13 @@ static void handle_BUG(struct pt_regs *regs)
 
 	if (eip < PAGE_OFFSET)
 		goto no_bug;
-	if (__get_user(ud2, (unsigned short *)eip))
+	if (__get_user(ud2, (unsigned short __user *)eip))
 		goto no_bug;
 	if (ud2 != 0x0b0f)
 		goto no_bug;
-	if (__get_user(line, (unsigned short *)(eip + 2)))
+	if (__get_user(line, (unsigned short __user *)(eip + 2)))
 		goto bug;
-	if (__get_user(file, (char **)(eip + 4)) ||
+	if (__get_user(file, (char * __user *)(eip + 4)) ||
 		(unsigned long)file < PAGE_OFFSET || __get_user(c, file))
 		file = "<bad filename>";
 
-- 
cgit v1.2.3-70-g09d2


From 77617bd8068470556049ebcdd0625acc7ab71a2b Mon Sep 17 00:00:00 2001
From: Domen Puncer <domen@coderock.org>
Date: Sat, 25 Jun 2005 14:58:45 -0700
Subject: [PATCH] arch/i386/kernel/apm.c: fix sparse warnings

Signed-off-by: Alexey Dobriyan <adobriyan@mail.ru>
Signed-off-by: Domen Puncer <domen@coderock.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/apm.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/apm.c b/arch/i386/kernel/apm.c
index 0ff65abcd56..d48ce929096 100644
--- a/arch/i386/kernel/apm.c
+++ b/arch/i386/kernel/apm.c
@@ -346,10 +346,10 @@ extern int (*console_blank_hook)(int);
 struct apm_user {
 	int		magic;
 	struct apm_user *	next;
-	int		suser: 1;
-	int		writer: 1;
-	int		reader: 1;
-	int		suspend_wait: 1;
+	unsigned int	suser: 1;
+	unsigned int	writer: 1;
+	unsigned int	reader: 1;
+	unsigned int	suspend_wait: 1;
 	int		suspend_result;
 	int		suspends_pending;
 	int		standbys_pending;
-- 
cgit v1.2.3-70-g09d2


From c7c5844526f55b93fc9a44170cb7568c9bc95c4a Mon Sep 17 00:00:00 2001
From: Domen Puncer <domen@coderock.org>
Date: Sat, 25 Jun 2005 14:58:46 -0700
Subject: [PATCH] arch/i386/mm/fault.c: fix sparse warnings

Signed-off-by: Alexey Dobriyan <adobriyan@mail.ru>
Signed-off-by: Domen Puncer <domen@coderock.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mm/fault.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/mm/fault.c b/arch/i386/mm/fault.c
index 92ed6c0a55d..8e90339d6ea 100644
--- a/arch/i386/mm/fault.c
+++ b/arch/i386/mm/fault.c
@@ -146,7 +146,7 @@ static int __is_prefetch(struct pt_regs *regs, unsigned long addr)
 
 		if (instr > limit)
 			break;
-		if (__get_user(opcode, (unsigned char *) instr))
+		if (__get_user(opcode, (unsigned char __user *) instr))
 			break; 
 
 		instr_hi = opcode & 0xf0; 
@@ -173,7 +173,7 @@ static int __is_prefetch(struct pt_regs *regs, unsigned long addr)
 			scan_more = 0;
 			if (instr > limit)
 				break;
-			if (__get_user(opcode, (unsigned char *) instr)) 
+			if (__get_user(opcode, (unsigned char __user *) instr))
 				break;
 			prefetch = (instr_lo == 0xF) &&
 				(opcode == 0x0D || opcode == 0x18);
-- 
cgit v1.2.3-70-g09d2


From 40086ea17e8cf12112438b3a855f44b26f7cb385 Mon Sep 17 00:00:00 2001
From: Domen Puncer <domen@coderock.org>
Date: Sat, 25 Jun 2005 14:58:46 -0700
Subject: [PATCH] arch/i386/crypto/aes.c: fix sparse warnings

Signed-off-by: Alexey Dobriyan <adobriyan@mail.ru>
Signed-off-by: Domen Puncer <domen@coderock.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/crypto/aes.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/i386/crypto/aes.c b/arch/i386/crypto/aes.c
index 1019430fc1f..88ee85c3b43 100644
--- a/arch/i386/crypto/aes.c
+++ b/arch/i386/crypto/aes.c
@@ -59,7 +59,7 @@ struct aes_ctx {
 };
 
 #define WPOLY 0x011b
-#define u32_in(x) le32_to_cpu(*(const u32 *)(x))
+#define u32_in(x) le32_to_cpup((const __le32 *)(x))
 #define bytes2word(b0, b1, b2, b3)  \
 	(((u32)(b3) << 24) | ((u32)(b2) << 16) | ((u32)(b1) << 8) | (b0))
 
-- 
cgit v1.2.3-70-g09d2


From 4ae6673e029d609da7ef4311440d6de501d6967a Mon Sep 17 00:00:00 2001
From: Jesper Juhl <juhl-lkml@dif.dk>
Date: Sat, 25 Jun 2005 14:58:48 -0700
Subject: [PATCH] get rid of redundant NULL checks before kfree() in arch/i386/

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>
---
 arch/i386/kernel/cpu/cpufreq/powernow-k7.c | 4 +---
 arch/i386/kernel/cpu/mtrr/generic.c        | 3 +--
 arch/i386/kernel/io_apic.c                 | 6 ++----
 3 files changed, 4 insertions(+), 9 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/cpu/cpufreq/powernow-k7.c b/arch/i386/kernel/cpu/cpufreq/powernow-k7.c
index 5c530064eb7..73a5dc5b26b 100644
--- a/arch/i386/kernel/cpu/cpufreq/powernow-k7.c
+++ b/arch/i386/kernel/cpu/cpufreq/powernow-k7.c
@@ -648,9 +648,7 @@ static int powernow_cpu_exit (struct cpufreq_policy *policy) {
 	}
 #endif
 
-	if (powernow_table)
-		kfree(powernow_table);
-
+	kfree(powernow_table);
 	return 0;
 }
 
diff --git a/arch/i386/kernel/cpu/mtrr/generic.c b/arch/i386/kernel/cpu/mtrr/generic.c
index f468a979e9a..64d91f73a0a 100644
--- a/arch/i386/kernel/cpu/mtrr/generic.c
+++ b/arch/i386/kernel/cpu/mtrr/generic.c
@@ -70,8 +70,7 @@ void __init get_mtrr_state(void)
 /*  Free resources associated with a struct mtrr_state  */
 void __init finalize_mtrr_state(void)
 {
-	if (mtrr_state.var_ranges)
-		kfree(mtrr_state.var_ranges);
+	kfree(mtrr_state.var_ranges);
 	mtrr_state.var_ranges = NULL;
 }
 
diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index fc81c8cddd2..060c4ca4d69 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -632,10 +632,8 @@ static int __init balanced_irq_init(void)
 		printk(KERN_ERR "balanced_irq_init: failed to spawn balanced_irq");
 failed:
 	for (i = 0; i < NR_CPUS; i++) {
-		if(irq_cpu_data[i].irq_delta)
-			kfree(irq_cpu_data[i].irq_delta);
-		if(irq_cpu_data[i].last_irq)
-			kfree(irq_cpu_data[i].last_irq);
+		kfree(irq_cpu_data[i].irq_delta);
+		kfree(irq_cpu_data[i].last_irq);
 	}
 	return 0;
 }
-- 
cgit v1.2.3-70-g09d2


From f45494480f31342125870c1a184999d7c5a59471 Mon Sep 17 00:00:00 2001
From: Domen Puncer <domen@coderock.org>
Date: Sat, 25 Jun 2005 14:58:59 -0700
Subject: [PATCH] x86_64: coding style and whitespace fixups

Remove some of the unnecessary differences between arch/i386 and
arch/x86_64.  This patch fixes more whitespace issues, some miscellaneous
typos, a wrong URL and a factually incorrect statement about the current
boot sector code.

Signed-off-by: Domen Puncer <domen@coderock.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/boot/Makefile            | 4 ++--
 arch/i386/boot/compressed/misc.c   | 1 -
 arch/i386/boot/edd.S               | 2 +-
 arch/i386/boot/setup.S             | 6 +++---
 arch/i386/boot/tools/build.c       | 5 ++---
 arch/x86_64/boot/compressed/head.S | 4 +---
 arch/x86_64/boot/compressed/misc.c | 2 +-
 arch/x86_64/boot/install.sh        | 2 +-
 arch/x86_64/boot/setup.S           | 8 ++++----
 arch/x86_64/boot/tools/build.c     | 5 ++---
 10 files changed, 17 insertions(+), 22 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/boot/Makefile b/arch/i386/boot/Makefile
index 43cd6220ee4..1e71382d413 100644
--- a/arch/i386/boot/Makefile
+++ b/arch/i386/boot/Makefile
@@ -25,8 +25,8 @@ SVGA_MODE := -DSVGA_MODE=NORMAL_VGA
 
 #RAMDISK := -DRAMDISK=512
 
-targets		:= vmlinux.bin bootsect bootsect.o setup setup.o \
-		   zImage bzImage
+targets		:= vmlinux.bin bootsect bootsect.o \
+		   setup setup.o zImage bzImage
 subdir- 	:= compressed
 
 hostprogs-y	:= tools/build
diff --git a/arch/i386/boot/compressed/misc.c b/arch/i386/boot/compressed/misc.c
index e1330766965..82a807f9f5e 100644
--- a/arch/i386/boot/compressed/misc.c
+++ b/arch/i386/boot/compressed/misc.c
@@ -354,7 +354,6 @@ static void close_output_buffer_if_we_run_high(struct moveparams *mv)
 	}
 }
 
-
 asmlinkage int decompress_kernel(struct moveparams *mv, void *rmode)
 {
 	real_mode = rmode;
diff --git a/arch/i386/boot/edd.S b/arch/i386/boot/edd.S
index 027d6b354ff..d8d69f2b911 100644
--- a/arch/i386/boot/edd.S
+++ b/arch/i386/boot/edd.S
@@ -6,7 +6,7 @@
  *   projects 1572D, 1484D, 1386D, 1226DT
  * disk signature read by Matt Domsch <Matt_Domsch@dell.com>
  *	and Andrew Wilks <Andrew_Wilks@dell.com> September 2003, June 2004
- * legacy CHS retreival by Patrick J. LoPresti <patl@users.sourceforge.net>
+ * legacy CHS retrieval by Patrick J. LoPresti <patl@users.sourceforge.net>
  *      March 2004
  * Command line option parsing, Matt Domsch, November 2004
  */
diff --git a/arch/i386/boot/setup.S b/arch/i386/boot/setup.S
index caa1fde6904..8cb420f40c5 100644
--- a/arch/i386/boot/setup.S
+++ b/arch/i386/boot/setup.S
@@ -33,7 +33,7 @@
  * Transcribed from Intel (as86) -> AT&T (gas) by Chris Noe, May 1999.
  * <stiker@northlink.com>
  *
- * Fix to work around buggy BIOSes which dont use carry bit correctly
+ * Fix to work around buggy BIOSes which don't use carry bit correctly
  * and/or report extended memory in CX/DX for e801h memory size detection 
  * call.  As a result the kernel got wrong figures.  The int15/e801h docs
  * from Ralf Brown interrupt list seem to indicate AX/BX should be used
@@ -357,7 +357,7 @@ bail820:
 
 meme801:
 	stc					# fix to work around buggy
-	xorw	%cx,%cx				# BIOSes which dont clear/set
+	xorw	%cx,%cx				# BIOSes which don't clear/set
 	xorw	%dx,%dx				# carry on pass/error of
 						# e801h memory size call
 						# or merely pass cx,dx though
@@ -847,7 +847,7 @@ flush_instr:
 #
 #	but we yet haven't reloaded the CS register, so the default size 
 #	of the target offset still is 16 bit.
-#       However, using an operand prefix (0x66), the CPU will properly
+#	However, using an operand prefix (0x66), the CPU will properly
 #	take our 48 bit far pointer. (INTeL 80386 Programmer's Reference
 #	Manual, Mixing 16-bit and 32-bit code, page 16-6)
 
diff --git a/arch/i386/boot/tools/build.c b/arch/i386/boot/tools/build.c
index 26509b826ae..4a17956512e 100644
--- a/arch/i386/boot/tools/build.c
+++ b/arch/i386/boot/tools/build.c
@@ -1,6 +1,4 @@
 /*
- *  $Id: build.c,v 1.5 1997/05/19 12:29:58 mj Exp $
- *
  *  Copyright (C) 1991, 1992  Linus Torvalds
  *  Copyright (C) 1997 Martin Mares
  */
@@ -8,7 +6,8 @@
 /*
  * This file builds a disk-image from three different files:
  *
- * - bootsect: exactly 512 bytes of 8086 machine code, loads the rest
+ * - bootsect: compatibility mbr which prints an error message if
+ *             someone tries to boot the kernel directly.
  * - setup: 8086 machine code, sets up system parm
  * - system: 80386 code for actual system
  *
diff --git a/arch/x86_64/boot/compressed/head.S b/arch/x86_64/boot/compressed/head.S
index b0df2aeff5d..6f55565e4d4 100644
--- a/arch/x86_64/boot/compressed/head.S
+++ b/arch/x86_64/boot/compressed/head.S
@@ -2,8 +2,6 @@
  *  linux/boot/head.S
  *
  *  Copyright (C) 1991, 1992, 1993  Linus Torvalds
- *
- *  $Id: head.S,v 1.3 2001/04/20 00:59:28 ak Exp $	 		
  */
 
 /*
@@ -21,7 +19,7 @@
  */
 
 /*
- * High loaded stuff by Hans Lermen & Werner Almesberger, Feb. 1996	
+ * High loaded stuff by Hans Lermen & Werner Almesberger, Feb. 1996
  */
 .code32
 .text
diff --git a/arch/x86_64/boot/compressed/misc.c b/arch/x86_64/boot/compressed/misc.c
index f992aef141b..b38d5b8b5fb 100644
--- a/arch/x86_64/boot/compressed/misc.c
+++ b/arch/x86_64/boot/compressed/misc.c
@@ -97,7 +97,7 @@ void* memset(void* s, int c, unsigned n);
 void* memcpy(void* dest, const void* src, unsigned n);
 
 static void putstr(const char *);
-  
+
 extern int end;
 static long free_mem_ptr = (long)&end;
 static long free_mem_end_ptr;
diff --git a/arch/x86_64/boot/install.sh b/arch/x86_64/boot/install.sh
index f17b40dfc0f..198af15a775 100644
--- a/arch/x86_64/boot/install.sh
+++ b/arch/x86_64/boot/install.sh
@@ -1,6 +1,6 @@
 #!/bin/sh
 #
-# arch/i386/boot/install.sh
+# arch/x86_64/boot/install.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
diff --git a/arch/x86_64/boot/setup.S b/arch/x86_64/boot/setup.S
index 75d4d2ad93b..ff58b2832b7 100644
--- a/arch/x86_64/boot/setup.S
+++ b/arch/x86_64/boot/setup.S
@@ -33,7 +33,7 @@
  * Transcribed from Intel (as86) -> AT&T (gas) by Chris Noe, May 1999.
  * <stiker@northlink.com>
  *
- * Fix to work around buggy BIOSes which dont use carry bit correctly
+ * Fix to work around buggy BIOSes which don't use carry bit correctly
  * and/or report extended memory in CX/DX for e801h memory size detection 
  * call.  As a result the kernel got wrong figures.  The int15/e801h docs
  * from Ralf Brown interrupt list seem to indicate AX/BX should be used
@@ -383,7 +383,7 @@ sse_ok:
 # a whole bunch of different types, and allows memory holes and
 # everything.  We scan through this memory map and build a list
 # of the first 32 memory areas, which we return at [E820MAP].
-# This is documented at http://www.teleport.com/~acpi/acpihtml/topic245.htm
+# This is documented at http://www.acpi.info/, in the ACPI 2.0 specification.
 
 #define SMAP  0x534d4150
 
@@ -436,7 +436,7 @@ bail820:
 
 meme801:
 	stc					# fix to work around buggy
-	xorw	%cx,%cx				# BIOSes which dont clear/set
+	xorw	%cx,%cx				# BIOSes which don't clear/set
 	xorw	%dx,%dx				# carry on pass/error of
 						# e801h memory size call
 						# or merely pass cx,dx though
@@ -733,7 +733,7 @@ flush_instr:
 #
 #	but we yet haven't reloaded the CS register, so the default size 
 #	of the target offset still is 16 bit.
-#       However, using an operant prefix (0x66), the CPU will properly
+#	However, using an operand prefix (0x66), the CPU will properly
 #	take our 48 bit far pointer. (INTeL 80386 Programmer's Reference
 #	Manual, Mixing 16-bit and 32-bit code, page 16-6)
 
diff --git a/arch/x86_64/boot/tools/build.c b/arch/x86_64/boot/tools/build.c
index c2fa6631317..18b5bac1c42 100644
--- a/arch/x86_64/boot/tools/build.c
+++ b/arch/x86_64/boot/tools/build.c
@@ -1,6 +1,4 @@
 /*
- *  $Id: build.c,v 1.3 2001/06/26 15:14:50 pavel Exp $
- *
  *  Copyright (C) 1991, 1992  Linus Torvalds
  *  Copyright (C) 1997 Martin Mares
  */
@@ -8,7 +6,8 @@
 /*
  * This file builds a disk-image from three different files:
  *
- * - bootsect: exactly 512 bytes of 8086 machine code, loads the rest
+ * - bootsect: compatibility mbr which prints an error message if
+ *             someone tries to boot the kernel directly.
  * - setup: 8086 machine code, sets up system parm
  * - system: 80386 code for actual system
  *
-- 
cgit v1.2.3-70-g09d2


From 9e84d1c36a505765c12e90289c9dbd575332840b Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Sat, 25 Jun 2005 14:59:12 -0700
Subject: [PATCH] i386: cleanup boot_cpu_logical_apicid variables

There are currently two different boot_cpu_logical_apicid variables:
- a global one in mpparse.c
- a static one in smpboot.c

Of these two, only the one in smpboot.c might be used (through
boot_cpu_apicid).

This patch therefore removes the one in mpparse.c .

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrey Panin <pazke@donpac.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/mpparse.c     | 2 --
 arch/i386/mach-visws/mpparse.c | 5 +----
 2 files changed, 1 insertion(+), 6 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index 383a11600d2..af917f609c7 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -67,7 +67,6 @@ unsigned long mp_lapic_addr;
 
 /* Processor that is doing the boot up */
 unsigned int boot_cpu_physical_apicid = -1U;
-unsigned int boot_cpu_logical_apicid = -1U;
 /* Internal processor count */
 static unsigned int __initdata num_processors;
 
@@ -180,7 +179,6 @@ static void __init MP_processor_info (struct mpc_config_processor *m)
 	if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
 		Dprintk("    Bootup CPU\n");
 		boot_cpu_physical_apicid = m->mpc_apicid;
-		boot_cpu_logical_apicid = apicid;
 	}
 
 	if (num_processors >= NR_CPUS) {
diff --git a/arch/i386/mach-visws/mpparse.c b/arch/i386/mach-visws/mpparse.c
index 5a22082147f..5f3d7e6de37 100644
--- a/arch/i386/mach-visws/mpparse.c
+++ b/arch/i386/mach-visws/mpparse.c
@@ -23,7 +23,6 @@ unsigned long mp_lapic_addr;
 
 /* Processor that is doing the boot up */
 unsigned int boot_cpu_physical_apicid = -1U;
-unsigned int boot_cpu_logical_apicid = -1U;
 
 /* Bitmask of physically existing CPUs */
 physid_mask_t phys_cpu_present_map;
@@ -52,10 +51,8 @@ static void __init MP_processor_info (struct mpc_config_processor *m)
 		(m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
 		m->mpc_apicver);
 
-	if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
+	if (m->mpc_cpuflag & CPU_BOOTPROCESSOR)
 		boot_cpu_physical_apicid = m->mpc_apicid;
-		boot_cpu_logical_apicid = logical_apicid;
-	}
 
 	ver = m->mpc_apicver;
 	if ((ver >= 0x14 && m->mpc_apicid >= 0xff) || m->mpc_apicid >= 0xf) {
-- 
cgit v1.2.3-70-g09d2


From f14c6fd0fc9fbaf242254c84ba2632decb25e91a Mon Sep 17 00:00:00 2001
From: Olaf Hering <olh@suse.de>
Date: Sat, 25 Jun 2005 14:59:21 -0700
Subject: [PATCH] update comment about gzip scratch size

fix a comment about the array size.

Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/boot/openfirmware/chrpmain.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ppc/boot/openfirmware/chrpmain.c b/arch/ppc/boot/openfirmware/chrpmain.c
index 6fb4f738728..effe4a0624b 100644
--- a/arch/ppc/boot/openfirmware/chrpmain.c
+++ b/arch/ppc/boot/openfirmware/chrpmain.c
@@ -39,7 +39,7 @@ char *avail_high;
 
 #define SCRATCH_SIZE	(128 << 10)
 
-static char scratch[SCRATCH_SIZE];	/* 1MB of scratch space for gunzip */
+static char scratch[SCRATCH_SIZE];	/* 128k of scratch space for gunzip */
 
 typedef void (*kernel_start_t)(int, int, void *, unsigned int, unsigned int);
 
-- 
cgit v1.2.3-70-g09d2


From ee48dd579977c35f35760bc893115e649b026306 Mon Sep 17 00:00:00 2001
From: Christophe Lucas <clucas@rotomalug.org>
Date: Sat, 25 Jun 2005 14:59:24 -0700
Subject: [PATCH] printk: arch/i386/mm/ioremap.c

printk() calls should include appropriate KERN_* constant.

Signed-off-by: Christophe Lucas <clucas@rotomalug.org>
Signed-off-by: Domen Puncer <domen@coderock.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mm/ioremap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/i386/mm/ioremap.c b/arch/i386/mm/ioremap.c
index d393eefc705..6b25afc933b 100644
--- a/arch/i386/mm/ioremap.c
+++ b/arch/i386/mm/ioremap.c
@@ -243,7 +243,7 @@ void iounmap(volatile void __iomem *addr)
 	write_lock(&vmlist_lock);
 	p = __remove_vm_area((void *) (PAGE_MASK & (unsigned long __force) addr));
 	if (!p) { 
-		printk("iounmap: bad address %p\n", addr);
+		printk(KERN_WARNING "iounmap: bad address %p\n", addr);
 		goto out_unlock;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From f90e7185ee00c1faa9d07b2f669d124772fdba17 Mon Sep 17 00:00:00 2001
From: Christophe Lucas <clucas@rotomalug.org>
Date: Sat, 25 Jun 2005 14:59:24 -0700
Subject: [PATCH] printk: arch/i386/mm/pgtable.c

printk() calls should include appropriate KERN_* constant.

Signed-off-by: Christophe Lucas <clucas@rotomalug.org>
Signed-off-by: Domen Puncer <domen@coderock.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mm/pgtable.c | 30 +++++++++++++++---------------
 1 file changed, 15 insertions(+), 15 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/mm/pgtable.c b/arch/i386/mm/pgtable.c
index 270c59f099a..bd2f7afc7a2 100644
--- a/arch/i386/mm/pgtable.c
+++ b/arch/i386/mm/pgtable.c
@@ -32,9 +32,9 @@ void show_mem(void)
 	unsigned long i;
 	struct page_state ps;
 
-	printk("Mem-info:\n");
+	printk(KERN_INFO "Mem-info:\n");
 	show_free_areas();
-	printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
+	printk(KERN_INFO "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_page_nr(pgdat, i);
@@ -49,18 +49,18 @@ void show_mem(void)
 				shared += page_count(page) - 1;
 		}
 	}
-	printk("%d pages of RAM\n", total);
-	printk("%d pages of HIGHMEM\n",highmem);
-	printk("%d reserved pages\n",reserved);
-	printk("%d pages shared\n",shared);
-	printk("%d pages swap cached\n",cached);
+	printk(KERN_INFO "%d pages of RAM\n", total);
+	printk(KERN_INFO "%d pages of HIGHMEM\n", highmem);
+	printk(KERN_INFO "%d reserved pages\n", reserved);
+	printk(KERN_INFO "%d pages shared\n", shared);
+	printk(KERN_INFO "%d pages swap cached\n", cached);
 
 	get_page_state(&ps);
-	printk("%lu pages dirty\n", ps.nr_dirty);
-	printk("%lu pages writeback\n", ps.nr_writeback);
-	printk("%lu pages mapped\n", ps.nr_mapped);
-	printk("%lu pages slab\n", ps.nr_slab);
-	printk("%lu pages pagetables\n", ps.nr_page_table_pages);
+	printk(KERN_INFO "%lu pages dirty\n", ps.nr_dirty);
+	printk(KERN_INFO "%lu pages writeback\n", ps.nr_writeback);
+	printk(KERN_INFO "%lu pages mapped\n", ps.nr_mapped);
+	printk(KERN_INFO "%lu pages slab\n", ps.nr_slab);
+	printk(KERN_INFO "%lu pages pagetables\n", ps.nr_page_table_pages);
 }
 
 /*
@@ -113,16 +113,16 @@ void set_pmd_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags)
 	pmd_t *pmd;
 
 	if (vaddr & (PMD_SIZE-1)) {		/* vaddr is misaligned */
-		printk ("set_pmd_pfn: vaddr misaligned\n");
+		printk(KERN_WARNING "set_pmd_pfn: vaddr misaligned\n");
 		return; /* BUG(); */
 	}
 	if (pfn & (PTRS_PER_PTE-1)) {		/* pfn is misaligned */
-		printk ("set_pmd_pfn: pfn misaligned\n");
+		printk(KERN_WARNING "set_pmd_pfn: pfn misaligned\n");
 		return; /* BUG(); */
 	}
 	pgd = swapper_pg_dir + pgd_index(vaddr);
 	if (pgd_none(*pgd)) {
-		printk ("set_pmd_pfn: pgd_none\n");
+		printk(KERN_WARNING "set_pmd_pfn: pgd_none\n");
 		return; /* BUG(); */
 	}
 	pud = pud_offset(pgd, vaddr);
-- 
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 'arch')

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 6f0dcb72d6a053fff9288b742d3bcc9b23f0f6db Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@ppc970.osdl.org>
Date: Sat, 25 Jun 2005 20:09:12 -0700
Subject: Fix up try_to_freeze() usage in arch/i386/kernel/signal.c

The parentheses were missing. Noted by Pavel Machek.
---
 arch/i386/kernel/signal.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/signal.c b/arch/i386/kernel/signal.c
index ac5b1e975c5..89ef7adc63a 100644
--- a/arch/i386/kernel/signal.c
+++ b/arch/i386/kernel/signal.c
@@ -608,7 +608,7 @@ int fastcall do_signal(struct pt_regs *regs, sigset_t *oldset)
 	if (!user_mode(regs))
 		return 1;
 
-	if (try_to_freeze)
+	if (try_to_freeze())
 		goto no_signal;
 
 	if (!oldset)
-- 
cgit v1.2.3-70-g09d2


From bdb94f3a78366d46bc73c8c8d8fe0dfb9522ff36 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Sun, 26 Jun 2005 03:27:21 -0700
Subject: [PATCH] arm: swsusp build fix

Another swsusp fixup.

Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/arm/kernel/signal.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c
index 07ddeed6176..5e435e42dac 100644
--- a/arch/arm/kernel/signal.c
+++ b/arch/arm/kernel/signal.c
@@ -697,7 +697,7 @@ static int do_signal(sigset_t *oldset, struct pt_regs *regs, int syscall)
 	if (!user_mode(regs))
 		return 0;
 
-	if (try_to_freeze(0))
+	if (try_to_freeze())
 		goto no_signal;
 
 	if (current->ptrace & PT_SINGLESTEP)
-- 
cgit v1.2.3-70-g09d2


From 7d681b23d6cc14a8c026ea6756242cb522cbbcae Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Sun, 26 Jun 2005 09:14:01 -0500
Subject: [PATCH] ppc32: Fix MPC83xx IPIC external interrupt pending register
 offset

The pending registers for IRQ1-IRQ7 were pointing to the interrupt pending
register instead of the external one.

Signed-off-by: Tony Li <Tony.Li@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/syslib/ipic.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc/syslib/ipic.c b/arch/ppc/syslib/ipic.c
index 580ed658e87..8f01e0f1d84 100644
--- a/arch/ppc/syslib/ipic.c
+++ b/arch/ppc/syslib/ipic.c
@@ -79,7 +79,7 @@ static struct ipic_info ipic_info[] = {
 		.prio_mask = 7,
 	},
 	[17] = {
-		.pend	= IPIC_SIPNR_H,
+		.pend	= IPIC_SEPNR,
 		.mask	= IPIC_SEMSR,
 		.prio	= IPIC_SMPRR_A,
 		.force	= IPIC_SEFCR,
@@ -87,7 +87,7 @@ static struct ipic_info ipic_info[] = {
 		.prio_mask = 5,
 	},
 	[18] = {
-		.pend	= IPIC_SIPNR_H,
+		.pend	= IPIC_SEPNR,
 		.mask	= IPIC_SEMSR,
 		.prio	= IPIC_SMPRR_A,
 		.force	= IPIC_SEFCR,
@@ -95,7 +95,7 @@ static struct ipic_info ipic_info[] = {
 		.prio_mask = 6,
 	},
 	[19] = {
-		.pend	= IPIC_SIPNR_H,
+		.pend	= IPIC_SEPNR,
 		.mask	= IPIC_SEMSR,
 		.prio	= IPIC_SMPRR_A,
 		.force	= IPIC_SEFCR,
@@ -103,7 +103,7 @@ static struct ipic_info ipic_info[] = {
 		.prio_mask = 7,
 	},
 	[20] = {
-		.pend	= IPIC_SIPNR_H,
+		.pend	= IPIC_SEPNR,
 		.mask	= IPIC_SEMSR,
 		.prio	= IPIC_SMPRR_B,
 		.force	= IPIC_SEFCR,
@@ -111,7 +111,7 @@ static struct ipic_info ipic_info[] = {
 		.prio_mask = 4,
 	},
 	[21] = {
-		.pend	= IPIC_SIPNR_H,
+		.pend	= IPIC_SEPNR,
 		.mask	= IPIC_SEMSR,
 		.prio	= IPIC_SMPRR_B,
 		.force	= IPIC_SEFCR,
@@ -119,7 +119,7 @@ static struct ipic_info ipic_info[] = {
 		.prio_mask = 5,
 	},
 	[22] = {
-		.pend	= IPIC_SIPNR_H,
+		.pend	= IPIC_SEPNR,
 		.mask	= IPIC_SEMSR,
 		.prio	= IPIC_SMPRR_B,
 		.force	= IPIC_SEFCR,
@@ -127,7 +127,7 @@ static struct ipic_info ipic_info[] = {
 		.prio_mask = 6,
 	},
 	[23] = {
-		.pend	= IPIC_SIPNR_H,
+		.pend	= IPIC_SEPNR,
 		.mask	= IPIC_SEMSR,
 		.prio	= IPIC_SMPRR_B,
 		.force	= IPIC_SEFCR,
-- 
cgit v1.2.3-70-g09d2


From 09b8b5f843afc21daf710cc610e5ca890ee94696 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Sun, 26 Jun 2005 17:06:36 +0100
Subject: [PATCH] ARM: Add SA_TIMER flag to timer interrupts

VST needs to know which timer handler is for the timer interrupt.
Mark all timer interrupts with the SA_TIMER flag.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-aaec2000/core.c            |  4 ++--
 arch/arm/mach-clps711x/time.c            |  4 ++--
 arch/arm/mach-clps7500/core.c            |  4 ++--
 arch/arm/mach-ebsa110/core.c             |  4 ++--
 arch/arm/mach-epxa10db/time.c            |  4 ++--
 arch/arm/mach-footbridge/dc21285-timer.c |  2 +-
 arch/arm/mach-footbridge/isa-timer.c     |  2 +-
 arch/arm/mach-h720x/cpu-h7201.c          |  4 ++--
 arch/arm/mach-h720x/cpu-h7202.c          |  4 ++--
 arch/arm/mach-imx/time.c                 |  4 ++--
 arch/arm/mach-integrator/core.c          |  4 ++--
 arch/arm/mach-iop3xx/iop321-time.c       |  2 +-
 arch/arm/mach-iop3xx/iop331-time.c       |  2 +-
 arch/arm/mach-ixp2000/core.c             |  4 ++--
 arch/arm/mach-ixp4xx/common.c            |  4 ++--
 arch/arm/mach-lh7a40x/time.c             |  4 ++--
 arch/arm/mach-omap/time.c                | 10 +++++-----
 arch/arm/mach-pxa/time.c                 |  4 ++--
 arch/arm/mach-s3c2410/time.c             |  4 ++--
 arch/arm/mach-sa1100/h3600.c             |  2 +-
 arch/arm/mach-sa1100/time.c              |  4 ++--
 arch/arm/mach-shark/core.c               |  4 ++--
 arch/arm/mach-versatile/core.c           |  4 ++--
 23 files changed, 44 insertions(+), 44 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/mach-aaec2000/core.c b/arch/arm/mach-aaec2000/core.c
index fc145b3768f..aece0cd4f0a 100644
--- a/arch/arm/mach-aaec2000/core.c
+++ b/arch/arm/mach-aaec2000/core.c
@@ -128,8 +128,8 @@ aaec2000_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 static struct irqaction aaec2000_timer_irq = {
 	.name		= "AAEC-2000 Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= aaec2000_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= aaec2000_timer_interrupt,
 };
 
 static void __init aaec2000_timer_init(void)
diff --git a/arch/arm/mach-clps711x/time.c b/arch/arm/mach-clps711x/time.c
index 383d4e0c6e3..1a23f0dcd4b 100644
--- a/arch/arm/mach-clps711x/time.c
+++ b/arch/arm/mach-clps711x/time.c
@@ -57,8 +57,8 @@ p720t_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 static struct irqaction clps711x_timer_irq = {
 	.name		= "CLPS711x Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= p720t_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= p720t_timer_interrupt,
 };
 
 static void __init clps711x_timer_init(void)
diff --git a/arch/arm/mach-clps7500/core.c b/arch/arm/mach-clps7500/core.c
index 0bc7da48861..90e85f434f6 100644
--- a/arch/arm/mach-clps7500/core.c
+++ b/arch/arm/mach-clps7500/core.c
@@ -298,8 +298,8 @@ clps7500_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 static struct irqaction clps7500_timer_irq = {
 	.name		= "CLPS7500 Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= clps7500_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= clps7500_timer_interrupt,
 };
 
 /*
diff --git a/arch/arm/mach-ebsa110/core.c b/arch/arm/mach-ebsa110/core.c
index ef362d44949..86ffdbb5626 100644
--- a/arch/arm/mach-ebsa110/core.c
+++ b/arch/arm/mach-ebsa110/core.c
@@ -173,8 +173,8 @@ ebsa110_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 static struct irqaction ebsa110_timer_irq = {
 	.name		= "EBSA110 Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= ebsa110_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= ebsa110_timer_interrupt,
 };
 
 /*
diff --git a/arch/arm/mach-epxa10db/time.c b/arch/arm/mach-epxa10db/time.c
index 1b991f3cc3c..4b1084dde8d 100644
--- a/arch/arm/mach-epxa10db/time.c
+++ b/arch/arm/mach-epxa10db/time.c
@@ -56,8 +56,8 @@ epxa10db_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 static struct irqaction epxa10db_timer_irq = {
 	.name		= "Excalibur Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= epxa10db_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= epxa10db_timer_interrupt,
 };
 
 /*
diff --git a/arch/arm/mach-footbridge/dc21285-timer.c b/arch/arm/mach-footbridge/dc21285-timer.c
index da5b9b7623c..14a62d6008f 100644
--- a/arch/arm/mach-footbridge/dc21285-timer.c
+++ b/arch/arm/mach-footbridge/dc21285-timer.c
@@ -43,7 +43,7 @@ timer1_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 static struct irqaction footbridge_timer_irq = {
 	.name		= "Timer1 timer tick",
 	.handler	= timer1_interrupt,
-	.flags		= SA_INTERRUPT,
+	.flags		= SA_INTERRUPT | SA_TIMER,
 };
 
 /*
diff --git a/arch/arm/mach-footbridge/isa-timer.c b/arch/arm/mach-footbridge/isa-timer.c
index a4fefa0bb5a..c1d74f7ab66 100644
--- a/arch/arm/mach-footbridge/isa-timer.c
+++ b/arch/arm/mach-footbridge/isa-timer.c
@@ -72,7 +72,7 @@ isa_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 static struct irqaction isa_timer_irq = {
 	.name		= "ISA timer tick",
 	.handler	= isa_timer_interrupt,
-	.flags		= SA_INTERRUPT,
+	.flags		= SA_INTERRUPT | SA_TIMER,
 };
 
 static void __init isa_timer_init(void)
diff --git a/arch/arm/mach-h720x/cpu-h7201.c b/arch/arm/mach-h720x/cpu-h7201.c
index 743656881ed..af9e4a5d5ea 100644
--- a/arch/arm/mach-h720x/cpu-h7201.c
+++ b/arch/arm/mach-h720x/cpu-h7201.c
@@ -41,8 +41,8 @@ h7201_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 static struct irqaction h7201_timer_irq = {
 	.name		= "h7201 Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= h7201_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= h7201_timer_interrupt,
 };
 
 /*
diff --git a/arch/arm/mach-h720x/cpu-h7202.c b/arch/arm/mach-h720x/cpu-h7202.c
index 21b8fb6122c..593b6a2a30e 100644
--- a/arch/arm/mach-h720x/cpu-h7202.c
+++ b/arch/arm/mach-h720x/cpu-h7202.c
@@ -171,8 +171,8 @@ static struct irqchip h7202_timerx_chip = {
 
 static struct irqaction h7202_timer_irq = {
 	.name		= "h7202 Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= h7202_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= h7202_timer_interrupt,
 };
 
 /*
diff --git a/arch/arm/mach-imx/time.c b/arch/arm/mach-imx/time.c
index 11f1e56c36b..ea805bfa5e5 100644
--- a/arch/arm/mach-imx/time.c
+++ b/arch/arm/mach-imx/time.c
@@ -72,8 +72,8 @@ imx_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 static struct irqaction imx_timer_irq = {
 	.name		= "i.MX Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= imx_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= imx_timer_interrupt,
 };
 
 /*
diff --git a/arch/arm/mach-integrator/core.c b/arch/arm/mach-integrator/core.c
index bd1e5e3c9d3..9222e57bd87 100644
--- a/arch/arm/mach-integrator/core.c
+++ b/arch/arm/mach-integrator/core.c
@@ -247,8 +247,8 @@ integrator_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 static struct irqaction integrator_timer_irq = {
 	.name		= "Integrator Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= integrator_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= integrator_timer_interrupt,
 };
 
 /*
diff --git a/arch/arm/mach-iop3xx/iop321-time.c b/arch/arm/mach-iop3xx/iop321-time.c
index 9b7dd64d1b8..d53af166950 100644
--- a/arch/arm/mach-iop3xx/iop321-time.c
+++ b/arch/arm/mach-iop3xx/iop321-time.c
@@ -86,7 +86,7 @@ iop321_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 static struct irqaction iop321_timer_irq = {
 	.name		= "IOP321 Timer Tick",
 	.handler	= iop321_timer_interrupt,
-	.flags		= SA_INTERRUPT
+	.flags		= SA_INTERRUPT | SA_TIMER,
 };
 
 static void __init iop321_timer_init(void)
diff --git a/arch/arm/mach-iop3xx/iop331-time.c b/arch/arm/mach-iop3xx/iop331-time.c
index e0169676926..1a6d9d661e4 100644
--- a/arch/arm/mach-iop3xx/iop331-time.c
+++ b/arch/arm/mach-iop3xx/iop331-time.c
@@ -83,7 +83,7 @@ iop331_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 static struct irqaction iop331_timer_irq = {
 	.name		= "IOP331 Timer Tick",
 	.handler	= iop331_timer_interrupt,
-	.flags		= SA_INTERRUPT
+	.flags		= SA_INTERRUPT | SA_TIMER,
 };
 
 static void __init iop331_timer_init(void)
diff --git a/arch/arm/mach-ixp2000/core.c b/arch/arm/mach-ixp2000/core.c
index 0ee34acb8d7..efd77eee065 100644
--- a/arch/arm/mach-ixp2000/core.c
+++ b/arch/arm/mach-ixp2000/core.c
@@ -194,8 +194,8 @@ static int ixp2000_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 static struct irqaction ixp2000_timer_irq = {
 	.name		= "IXP2000 Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= ixp2000_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= ixp2000_timer_interrupt,
 };
 
 void __init ixp2000_init_time(unsigned long tick_rate)
diff --git a/arch/arm/mach-ixp4xx/common.c b/arch/arm/mach-ixp4xx/common.c
index f39e8408488..04490a9f8f6 100644
--- a/arch/arm/mach-ixp4xx/common.c
+++ b/arch/arm/mach-ixp4xx/common.c
@@ -298,8 +298,8 @@ static irqreturn_t ixp4xx_timer_interrupt(int irq, void *dev_id, struct pt_regs
 
 static struct irqaction ixp4xx_timer_irq = {
 	.name		= "IXP4xx Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= ixp4xx_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= ixp4xx_timer_interrupt,
 };
 
 static void __init ixp4xx_timer_init(void)
diff --git a/arch/arm/mach-lh7a40x/time.c b/arch/arm/mach-lh7a40x/time.c
index 51e1c814b40..be377e331f2 100644
--- a/arch/arm/mach-lh7a40x/time.c
+++ b/arch/arm/mach-lh7a40x/time.c
@@ -53,8 +53,8 @@ lh7a40x_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 static struct irqaction lh7a40x_timer_irq = {
 	.name		= "LHA740x Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= lh7a40x_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= lh7a40x_timer_interrupt,
 };
 
 static void __init lh7a40x_timer_init(void)
diff --git a/arch/arm/mach-omap/time.c b/arch/arm/mach-omap/time.c
index 4205fdcb632..589e8b2740d 100644
--- a/arch/arm/mach-omap/time.c
+++ b/arch/arm/mach-omap/time.c
@@ -188,8 +188,8 @@ static irqreturn_t omap_mpu_timer_interrupt(int irq, void *dev_id,
 
 static struct irqaction omap_mpu_timer_irq = {
 	.name		= "mpu timer",
-	.flags		= SA_INTERRUPT,
-	.handler	= omap_mpu_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= omap_mpu_timer_interrupt,
 };
 
 static unsigned long omap_mpu_timer1_overflows;
@@ -203,7 +203,7 @@ static irqreturn_t omap_mpu_timer1_interrupt(int irq, void *dev_id,
 static struct irqaction omap_mpu_timer1_irq = {
 	.name		= "mpu timer1 overflow",
 	.flags		= SA_INTERRUPT,
-	.handler	= omap_mpu_timer1_interrupt
+	.handler	= omap_mpu_timer1_interrupt,
 };
 
 static __init void omap_init_mpu_timer(void)
@@ -349,8 +349,8 @@ static irqreturn_t omap_32k_timer_interrupt(int irq, void *dev_id,
 
 static struct irqaction omap_32k_timer_irq = {
 	.name		= "32KHz timer",
-	.flags		= SA_INTERRUPT,
-	.handler	= omap_32k_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= omap_32k_timer_interrupt,
 };
 
 static __init void omap_init_32k_timer(void)
diff --git a/arch/arm/mach-pxa/time.c b/arch/arm/mach-pxa/time.c
index 473fb6173f7..6e5202154f9 100644
--- a/arch/arm/mach-pxa/time.c
+++ b/arch/arm/mach-pxa/time.c
@@ -105,8 +105,8 @@ pxa_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 static struct irqaction pxa_timer_irq = {
 	.name		= "PXA Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= pxa_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= pxa_timer_interrupt,
 };
 
 static void __init pxa_timer_init(void)
diff --git a/arch/arm/mach-s3c2410/time.c b/arch/arm/mach-s3c2410/time.c
index 179f0e031af..765a3a9ae03 100644
--- a/arch/arm/mach-s3c2410/time.c
+++ b/arch/arm/mach-s3c2410/time.c
@@ -137,8 +137,8 @@ s3c2410_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 static struct irqaction s3c2410_timer_irq = {
 	.name		= "S3C2410 Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= s3c2410_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= s3c2410_timer_interrupt,
 };
 
 /*
diff --git a/arch/arm/mach-sa1100/h3600.c b/arch/arm/mach-sa1100/h3600.c
index 84c86543501..65dbe991426 100644
--- a/arch/arm/mach-sa1100/h3600.c
+++ b/arch/arm/mach-sa1100/h3600.c
@@ -727,7 +727,7 @@ static void h3800_IRQ_demux(unsigned int irq, struct irqdesc *desc, struct pt_re
 static struct irqaction h3800_irq = {
 	.name		= "h3800_asic",
 	.handler	= h3800_IRQ_demux,
-	.flags		= SA_INTERRUPT,
+	.flags		= SA_INTERRUPT | SA_TIMER,
 };
 
 u32 kpio_int_shadow = 0;
diff --git a/arch/arm/mach-sa1100/time.c b/arch/arm/mach-sa1100/time.c
index 19b0c0fd637..0eeb3616ffe 100644
--- a/arch/arm/mach-sa1100/time.c
+++ b/arch/arm/mach-sa1100/time.c
@@ -99,8 +99,8 @@ sa1100_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 static struct irqaction sa1100_timer_irq = {
 	.name		= "SA11xx Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= sa1100_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= sa1100_timer_interrupt,
 };
 
 static void __init sa1100_timer_init(void)
diff --git a/arch/arm/mach-shark/core.c b/arch/arm/mach-shark/core.c
index a9bc5d0dbd8..aa0e2f6e02f 100644
--- a/arch/arm/mach-shark/core.c
+++ b/arch/arm/mach-shark/core.c
@@ -84,8 +84,8 @@ shark_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 
 static struct irqaction shark_timer_irq = {
 	.name		= "Shark Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= shark_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= shark_timer_interrupt,
 };
 
 /*
diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c
index 6a7cbea5e09..9d1f2253e98 100644
--- a/arch/arm/mach-versatile/core.c
+++ b/arch/arm/mach-versatile/core.c
@@ -875,8 +875,8 @@ static irqreturn_t versatile_timer_interrupt(int irq, void *dev_id, struct pt_re
 
 static struct irqaction versatile_timer_irq = {
 	.name		= "Versatile Timer Tick",
-	.flags		= SA_INTERRUPT,
-	.handler	= versatile_timer_interrupt
+	.flags		= SA_INTERRUPT | SA_TIMER,
+	.handler	= versatile_timer_interrupt,
 };
 
 /*
-- 
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 'arch')

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 26799e675e47c5aa3104628e2081a7299ea46557 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Sun, 26 Jun 2005 22:24:19 +0100
Subject: [PATCH] ARM: 2757/1: remove ixdp2400_init_irq from ixdp2800 code

Patch from Lennert Buytenhek

Compiling one kernel that supports both ixdp2400 and ixdp2800 gives
an error, as a copy of the ixdp2400 irq init routing accidentally
ended up in ixdp2800.c somehow.

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/ixdp2800.c | 6 ------
 1 file changed, 6 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/mach-ixp2000/ixdp2800.c b/arch/arm/mach-ixp2000/ixdp2800.c
index aec13c7108a..468a4bbfb72 100644
--- a/arch/arm/mach-ixp2000/ixdp2800.c
+++ b/arch/arm/mach-ixp2000/ixdp2800.c
@@ -42,12 +42,6 @@
 #include <asm/mach/flash.h>
 #include <asm/mach/arch.h>
 
-
-void ixdp2400_init_irq(void)
-{
-	ixdp2x00_init_irq(IXDP2800_CPLD_INT_STAT, IXDP2800_CPLD_INT_MASK, IXDP2400_NR_IRQS);
-}
-
 /*************************************************************************
  * IXDP2800 timer tick
  *************************************************************************/
-- 
cgit v1.2.3-70-g09d2


From 99bcc0590806c4f7a4ecf1a11add335b56cde963 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 27 Jun 2005 13:59:43 +0100
Subject: [PATCH] ARM: Add missed AAEC2000 file

My scripts missed committing this file.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-aaec2000/Makefile.boot | 1 +
 1 file changed, 1 insertion(+)
 create mode 100644 arch/arm/mach-aaec2000/Makefile.boot

(limited to 'arch')

diff --git a/arch/arm/mach-aaec2000/Makefile.boot b/arch/arm/mach-aaec2000/Makefile.boot
new file mode 100644
index 00000000000..8f5a8b7c53c
--- /dev/null
+++ b/arch/arm/mach-aaec2000/Makefile.boot
@@ -0,0 +1 @@
+	zreladdr-y := 0xf0008000
-- 
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 'arch')

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 a343e6075a396e07eeff52c0da5629c8fd396be2 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 27 Jun 2005 14:08:56 +0100
Subject: [PATCH] ARM: Move PGD kernel page table initialisation

It doesn't make sense to have the PGD kernel pointers initialisation
separate from the PGD user pointers, especially when we clean the
data cache over the whole range.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/mm-armv.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/mm/mm-armv.c b/arch/arm/mm/mm-armv.c
index 2c2b93d77d4..d79864a0dfa 100644
--- a/arch/arm/mm/mm-armv.c
+++ b/arch/arm/mm/mm-armv.c
@@ -169,7 +169,14 @@ pgd_t *get_pgd_slow(struct mm_struct *mm)
 
 	memzero(new_pgd, FIRST_KERNEL_PGD_NR * sizeof(pgd_t));
 
+	/*
+	 * Copy over the kernel and IO PGD entries
+	 */
 	init_pgd = pgd_offset_k(0);
+	memcpy(new_pgd + FIRST_KERNEL_PGD_NR, init_pgd + FIRST_KERNEL_PGD_NR,
+		       (PTRS_PER_PGD - FIRST_KERNEL_PGD_NR) * sizeof(pgd_t));
+
+	clean_dcache_area(new_pgd, PTRS_PER_PGD * sizeof(pgd_t));
 
 	if (!vectors_high()) {
 		/*
@@ -198,14 +205,6 @@ pgd_t *get_pgd_slow(struct mm_struct *mm)
 		spin_unlock(&mm->page_table_lock);
 	}
 
-	/*
-	 * Copy over the kernel and IO PGD entries
-	 */
-	memcpy(new_pgd + FIRST_KERNEL_PGD_NR, init_pgd + FIRST_KERNEL_PGD_NR,
-		       (PTRS_PER_PGD - FIRST_KERNEL_PGD_NR) * sizeof(pgd_t));
-
-	clean_dcache_area(new_pgd, PTRS_PER_PGD * sizeof(pgd_t));
-
 	return new_pgd;
 
 no_pte:
-- 
cgit v1.2.3-70-g09d2


From a013053d4965d9a45300938e713a4b512e0257d8 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 27 Jun 2005 14:16:47 +0100
Subject: [PATCH] ARM: Move memmap freeing into init.c

It doesn't make sense for this to be in mm-armv.c now that 26-bit
ARM support is no longer integrated into arch/arm.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/init.c    | 71 +++++++++++++++++++++++++++++++++++++++++++++-----
 arch/arm/mm/mm-armv.c | 72 ---------------------------------------------------
 2 files changed, 65 insertions(+), 78 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index c08710b1ff0..6dcb23d64bf 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -522,6 +522,69 @@ static inline void free_area(unsigned long addr, unsigned long end, char *s)
 		printk(KERN_INFO "Freeing %s memory: %dK\n", s, size);
 }
 
+static inline void
+free_memmap(int node, unsigned long start_pfn, unsigned long end_pfn)
+{
+	struct page *start_pg, *end_pg;
+	unsigned long pg, pgend;
+
+	/*
+	 * Convert start_pfn/end_pfn to a struct page pointer.
+	 */
+	start_pg = pfn_to_page(start_pfn);
+	end_pg = pfn_to_page(end_pfn);
+
+	/*
+	 * Convert to physical addresses, and
+	 * round start upwards and end downwards.
+	 */
+	pg = PAGE_ALIGN(__pa(start_pg));
+	pgend = __pa(end_pg) & PAGE_MASK;
+
+	/*
+	 * If there are free pages between these,
+	 * free the section of the memmap array.
+	 */
+	if (pg < pgend)
+		free_bootmem_node(NODE_DATA(node), pg, pgend - pg);
+}
+
+/*
+ * The mem_map array can get very big.  Free the unused area of the memory map.
+ */
+static void __init free_unused_memmap_node(int node, struct meminfo *mi)
+{
+	unsigned long bank_start, prev_bank_end = 0;
+	unsigned int i;
+
+	/*
+	 * [FIXME] This relies on each bank being in address order.  This
+	 * may not be the case, especially if the user has provided the
+	 * information on the command line.
+	 */
+	for (i = 0; i < mi->nr_banks; i++) {
+		if (mi->bank[i].size == 0 || mi->bank[i].node != node)
+			continue;
+
+		bank_start = mi->bank[i].start >> PAGE_SHIFT;
+		if (bank_start < prev_bank_end) {
+			printk(KERN_ERR "MEM: unordered memory banks.  "
+				"Not freeing memmap.\n");
+			break;
+		}
+
+		/*
+		 * If we had a previous bank, and there is a space
+		 * between the current bank and the previous, free it.
+		 */
+		if (prev_bank_end && prev_bank_end != bank_start)
+			free_memmap(node, prev_bank_end, bank_start);
+
+		prev_bank_end = (mi->bank[i].start +
+				 mi->bank[i].size) >> PAGE_SHIFT;
+	}
+}
+
 /*
  * mem_init() marks the free areas in the mem_map and tells us how much
  * memory is free.  This is done after various parts of the system have
@@ -540,16 +603,12 @@ void __init mem_init(void)
 	max_mapnr   = virt_to_page(high_memory) - mem_map;
 #endif
 
-	/*
-	 * We may have non-contiguous memory.
-	 */
-	if (meminfo.nr_banks != 1)
-		create_memmap_holes(&meminfo);
-
 	/* this will put all unused low memory onto the freelists */
 	for_each_online_node(node) {
 		pg_data_t *pgdat = NODE_DATA(node);
 
+		free_unused_memmap_node(node, &meminfo);
+
 		if (pgdat->node_spanned_pages != 0)
 			totalram_pages += free_all_bootmem_node(pgdat);
 	}
diff --git a/arch/arm/mm/mm-armv.c b/arch/arm/mm/mm-armv.c
index d79864a0dfa..052ab443ec4 100644
--- a/arch/arm/mm/mm-armv.c
+++ b/arch/arm/mm/mm-armv.c
@@ -697,75 +697,3 @@ void __init iotable_init(struct map_desc *io_desc, int nr)
 	for (i = 0; i < nr; i++)
 		create_mapping(io_desc + i);
 }
-
-static inline void
-free_memmap(int node, unsigned long start_pfn, unsigned long end_pfn)
-{
-	struct page *start_pg, *end_pg;
-	unsigned long pg, pgend;
-
-	/*
-	 * Convert start_pfn/end_pfn to a struct page pointer.
-	 */
-	start_pg = pfn_to_page(start_pfn);
-	end_pg = pfn_to_page(end_pfn);
-
-	/*
-	 * Convert to physical addresses, and
-	 * round start upwards and end downwards.
-	 */
-	pg = PAGE_ALIGN(__pa(start_pg));
-	pgend = __pa(end_pg) & PAGE_MASK;
-
-	/*
-	 * If there are free pages between these,
-	 * free the section of the memmap array.
-	 */
-	if (pg < pgend)
-		free_bootmem_node(NODE_DATA(node), pg, pgend - pg);
-}
-
-static inline void free_unused_memmap_node(int node, struct meminfo *mi)
-{
-	unsigned long bank_start, prev_bank_end = 0;
-	unsigned int i;
-
-	/*
-	 * [FIXME] This relies on each bank being in address order.  This
-	 * may not be the case, especially if the user has provided the
-	 * information on the command line.
-	 */
-	for (i = 0; i < mi->nr_banks; i++) {
-		if (mi->bank[i].size == 0 || mi->bank[i].node != node)
-			continue;
-
-		bank_start = mi->bank[i].start >> PAGE_SHIFT;
-		if (bank_start < prev_bank_end) {
-			printk(KERN_ERR "MEM: unordered memory banks.  "
-				"Not freeing memmap.\n");
-			break;
-		}
-
-		/*
-		 * If we had a previous bank, and there is a space
-		 * between the current bank and the previous, free it.
-		 */
-		if (prev_bank_end && prev_bank_end != bank_start)
-			free_memmap(node, prev_bank_end, bank_start);
-
-		prev_bank_end = PAGE_ALIGN(mi->bank[i].start +
-					   mi->bank[i].size) >> PAGE_SHIFT;
-	}
-}
-
-/*
- * The mem_map array can get very big.  Free
- * the unused area of the memory map.
- */
-void __init create_memmap_holes(struct meminfo *mi)
-{
-	int node;
-
-	for_each_online_node(node)
-		free_unused_memmap_node(node, mi);
-}
-- 
cgit v1.2.3-70-g09d2


From f3bb742640338eca0d8c3fa8071df89168efbf0a Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 27 Jun 2005 14:49:10 +0100
Subject: [PATCH] ARM: Update mach-types

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/tools/mach-types | 69 ++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 66 insertions(+), 3 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/tools/mach-types b/arch/arm/tools/mach-types
index 30c1dfbb052..6d3a79e5fef 100644
--- a/arch/arm/tools/mach-types
+++ b/arch/arm/tools/mach-types
@@ -6,7 +6,7 @@
 # To add an entry into this database, please see Documentation/arm/README,
 # or contact rmk@arm.linux.org.uk
 #
-# Last update: Thu Mar 24 14:34:50 2005
+# Last update: Thu Jun 23 20:19:33 2005
 #
 # machine_is_xxx	CONFIG_xxxx		MACH_TYPE_xxx		number
 #
@@ -243,7 +243,7 @@ yoho			ARCH_YOHO		YOHO			231
 jasper			ARCH_JASPER		JASPER			232
 dsc25			ARCH_DSC25		DSC25			233
 omap_innovator		MACH_OMAP_INNOVATOR	OMAP_INNOVATOR		234
-ramses			ARCH_RAMSES		RAMSES			235
+mnci			ARCH_RAMSES		RAMSES			235
 s28x			ARCH_S28X		S28X			236
 mport3			ARCH_MPORT3		MPORT3			237
 pxa_eagle250		ARCH_PXA_EAGLE250	PXA_EAGLE250		238
@@ -323,7 +323,7 @@ nimbra29x		ARCH_NIMBRA29X		NIMBRA29X		311
 nimbra210		ARCH_NIMBRA210		NIMBRA210		312
 hhp_d95xx		ARCH_HHP_D95XX		HHP_D95XX		313
 labarm			ARCH_LABARM		LABARM			314
-m825xx			ARCH_M825XX		M825XX			315
+comcerto		ARCH_M825XX		M825XX			315
 m7100			SA1100_M7100		M7100			316
 nipc2			ARCH_NIPC2		NIPC2			317
 fu7202			ARCH_FU7202		FU7202			318
@@ -724,3 +724,66 @@ lpc22xx			MACH_LPC22XX		LPC22XX			715
 omap_comet3		MACH_COMET3		COMET3			716
 omap_comet4		MACH_COMET4		COMET4			717
 csb625			MACH_CSB625		CSB625			718
+fortunet2		MACH_FORTUNET2		FORTUNET2		719
+s5h2200			MACH_S5H2200		S5H2200			720
+optorm920		MACH_OPTORM920		OPTORM920		721
+adsbitsyxb		MACH_ADSBITSYXB		ADSBITSYXB		722
+adssphere		MACH_ADSSPHERE		ADSSPHERE		723
+adsportal		MACH_ADSPORTAL		ADSPORTAL		724
+ln2410sbc		MACH_LN2410SBC		LN2410SBC		725
+cb3rufc			MACH_CB3RUFC		CB3RUFC			726
+mp2usb			MACH_MP2USB		MP2USB			727
+ntnp425c		MACH_NTNP425C		NTNP425C		728
+colibri			MACH_COLIBRI		COLIBRI			729
+pcm7220			MACH_PCM7220		PCM7220			730
+gateway7001		MACH_GATEWAY7001	GATEWAY7001		731
+pcm027			MACH_PCM027		PCM027			732
+cmpxa			MACH_CMPXA		CMPXA			733
+anubis			MACH_ANUBIS		ANUBIS			734
+ite8152			MACH_ITE8152		ITE8152			735
+lpc3xxx			MACH_LPC3XXX		LPC3XXX			736
+puppeteer		MACH_PUPPETEER		PUPPETEER		737
+vt001			MACH_MACH_VADATECH	MACH_VADATECH		738
+e570			MACH_E570		E570			739
+x50			MACH_X50		X50			740
+recon			MACH_RECON		RECON			741
+xboardgp8		MACH_XBOARDGP8		XBOARDGP8		742
+fpic2			MACH_FPIC2		FPIC2			743
+akita			MACH_AKITA		AKITA			744
+a81			MACH_A81		A81			745
+svm_sc25x		MACH_SVM_SC25X		SVM_SC25X		746
+vt020			MACH_VADATECH020	VADATECH020		747
+tli			MACH_TLI		TLI			748
+edb9315lc		MACH_EDB9315LC		EDB9315LC		749
+passec			MACH_PASSEC		PASSEC			750
+ds_tiger		MACH_DS_TIGER		DS_TIGER		751
+e310			MACH_E310		E310			752
+e330			MACH_E330		E330			753
+rt3000			MACH_RT3000		RT3000			754
+nokia770		MACH_NOKIA770		NOKIA770		755
+pnx0106			MACH_PNX0106		PNX0106			756
+hx21xx			MACH_HX21XX		HX21XX			757
+faraday			MACH_FARADAY		FARADAY			758
+sbc9312			MACH_SBC9312		SBC9312			759
+batman			MACH_BATMAN		BATMAN			760
+jpd201			MACH_JPD201		JPD201			761
+mipsa			MACH_MIPSA		MIPSA			762
+kacom			MACH_KACOM		KACOM			763
+swarcocpu		MACH_SWARCOCPU		SWARCOCPU		764
+swarcodsl		MACH_SWARCODSL		SWARCODSL		765
+blueangel		MACH_BLUEANGEL		BLUEANGEL		766
+hairygrama		MACH_HAIRYGRAMA		HAIRYGRAMA		767
+banff			MACH_BANFF		BANFF			768
+carmeva			MACH_CARMEVA		CARMEVA			769
+sam255			MACH_SAM255		SAM255			770
+ppm10			MACH_PPM10		PPM10			771
+edb9315a		MACH_EDB9315A		EDB9315A		772
+sunset			MACH_SUNSET		SUNSET			773
+stargate2		MACH_STARGATE2		STARGATE2		774
+intelmote2		MACH_INTELMOTE2		INTELMOTE2		775
+trizeps4		MACH_TRIZEPS4		TRIZEPS4		776
+mainstone2		MACH_MAINSTONE2		MAINSTONE2		777
+ez_ixp42x		MACH_EZ_IXP42X		EZ_IXP42X		778
+tapwave_zodiac		MACH_TAPWAVE_ZODIAC	TAPWAVE_ZODIAC		779
+universalmeter		MACH_UNIVERSALMETER	UNIVERSALMETER		780
+hicoarm9		MACH_HICOARM9		HICOARM9		781
-- 
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 'arch')

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 313980c92724cf42877a7bdafdef439ee9d68ccb Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Mon, 11 Apr 2005 15:38:25 -0700
Subject: [PATCH] USB: omap_udc updates (mostly cleanups)

Various USB patches, mostly for portability:

  - Fifo mode 1 didn't work previously (oopsed), so now it's fixed and
    (why not) defines even more endpoints for composite devices.

  - OMAP 1710 doesn't have an internal transceiver.

  - Small PM update:  if the USB link is suspended, don't disconnect on
    entry to deep sleep.

  - Be more correct about handling zero length control reads.  OMAP
    seems to mis-handle that protocol peculiarity though; best avoided.

  - Platform device resources (for UDC and OTG controllers) now use
    physical addresses, so /proc/iomem is more consistent.

  - Minor cleanups, notably (by volume) for "sparse" NULL warnings.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
---
 arch/arm/mach-omap/usb.c      |   8 +--
 drivers/usb/gadget/omap_udc.c | 150 ++++++++++++++++++++++++++----------------
 2 files changed, 97 insertions(+), 61 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/mach-omap/usb.c b/arch/arm/mach-omap/usb.c
index 6e805d451d0..7f37857b1a2 100644
--- a/arch/arm/mach-omap/usb.c
+++ b/arch/arm/mach-omap/usb.c
@@ -288,8 +288,8 @@ static void usb_release(struct device *dev)
 static struct resource udc_resources[] = {
 	/* order is significant! */
 	{		/* registers */
-		.start		= IO_ADDRESS(UDC_BASE),
-		.end		= IO_ADDRESS(UDC_BASE + 0xff),
+		.start		= UDC_BASE,
+		.end		= UDC_BASE + 0xff,
 		.flags		= IORESOURCE_MEM,
 	}, {		/* general IRQ */
 		.start		= IH2_BASE + 20,
@@ -355,8 +355,8 @@ static struct platform_device ohci_device = {
 static struct resource otg_resources[] = {
 	/* order is significant! */
 	{
-		.start		= IO_ADDRESS(OTG_BASE),
-		.end		= IO_ADDRESS(OTG_BASE + 0xff),
+		.start		= OTG_BASE,
+		.end		= OTG_BASE + 0xff,
 		.flags		= IORESOURCE_MEM,
 	}, {
 		.start		= IH2_BASE + 8,
diff --git a/drivers/usb/gadget/omap_udc.c b/drivers/usb/gadget/omap_udc.c
index 98cbcbc16cc..4ec91a68f96 100644
--- a/drivers/usb/gadget/omap_udc.c
+++ b/drivers/usb/gadget/omap_udc.c
@@ -52,7 +52,6 @@
 #include <asm/mach-types.h>
 
 #include <asm/arch/dma.h>
-#include <asm/arch/mux.h>
 #include <asm/arch/usb.h>
 
 #include "omap_udc.h"
@@ -253,7 +252,7 @@ static int omap_ep_disable(struct usb_ep *_ep)
 	}
 
 	spin_lock_irqsave(&ep->udc->lock, flags);
-	ep->desc = 0;
+	ep->desc = NULL;
 	nuke (ep, -ESHUTDOWN);
 	ep->ep.maxpacket = ep->maxpacket;
 	ep->has_dma = 0;
@@ -388,8 +387,8 @@ done(struct omap_ep *ep, struct omap_req *req, int status)
 
 /*-------------------------------------------------------------------------*/
 
-#define	FIFO_FULL	(UDC_NON_ISO_FIFO_FULL | UDC_ISO_FIFO_FULL)
-#define	FIFO_UNWRITABLE	(UDC_EP_HALTED | FIFO_FULL)
+#define UDC_FIFO_FULL		(UDC_NON_ISO_FIFO_FULL | UDC_ISO_FIFO_FULL)
+#define UDC_FIFO_UNWRITABLE	(UDC_EP_HALTED | UDC_FIFO_FULL)
 
 #define FIFO_EMPTY	(UDC_NON_ISO_FIFO_EMPTY | UDC_ISO_FIFO_EMPTY)
 #define FIFO_UNREADABLE (UDC_EP_HALTED | FIFO_EMPTY)
@@ -433,7 +432,7 @@ static int write_fifo(struct omap_ep *ep, struct omap_req *req)
 
 	/* PIO-IN isn't double buffered except for iso */
 	ep_stat = UDC_STAT_FLG_REG;
-	if (ep_stat & FIFO_UNWRITABLE)
+	if (ep_stat & UDC_FIFO_UNWRITABLE)
 		return 0;
 
 	count = ep->ep.maxpacket;
@@ -504,7 +503,7 @@ static int read_fifo(struct omap_ep *ep, struct omap_req *req)
 		if (ep_stat & UDC_EP_HALTED)
 			break;
 
-		if (ep_stat & FIFO_FULL)
+		if (ep_stat & UDC_FIFO_FULL)
 			avail = ep->ep.maxpacket;
 		else  {
 			avail = UDC_RXFSTAT_REG;
@@ -856,7 +855,7 @@ static void dma_channel_release(struct omap_ep *ep)
 	if (!list_empty(&ep->queue))
 		req = container_of(ep->queue.next, struct omap_req, queue);
 	else
-		req = 0;
+		req = NULL;
 
 	active = ((1 << 7) & omap_readl(OMAP_DMA_CCR(ep->lch))) != 0;
 
@@ -997,18 +996,19 @@ omap_ep_queue(struct usb_ep *_ep, struct usb_request *_req, int gfp_flags)
 					UDC_IRQ_EN_REG = irq_en;
 				}
 
-				/* STATUS is reverse direction */
-				UDC_EP_NUM_REG = is_in
-						? UDC_EP_SEL
-						: (UDC_EP_SEL|UDC_EP_DIR);
+				/* STATUS for zero length DATA stages is
+				 * always an IN ... even for IN transfers,
+				 * a wierd case which seem to stall OMAP.
+				 */
+				UDC_EP_NUM_REG = (UDC_EP_SEL|UDC_EP_DIR);
 				UDC_CTRL_REG = UDC_CLR_EP;
 				UDC_CTRL_REG = UDC_SET_FIFO_EN;
-				UDC_EP_NUM_REG = udc->ep0_in ? 0 : UDC_EP_DIR;
+				UDC_EP_NUM_REG = UDC_EP_DIR;
 
 				/* cleanup */
 				udc->ep0_pending = 0;
 				done(ep, req, 0);
-				req = 0;
+				req = NULL;
 
 			/* non-empty DATA stage */
 			} else if (is_in) {
@@ -1029,7 +1029,7 @@ omap_ep_queue(struct usb_ep *_ep, struct usb_request *_req, int gfp_flags)
 			(is_in ? next_in_dma : next_out_dma)(ep, req);
 		else if (req) {
 			if ((is_in ? write_fifo : read_fifo)(ep, req) == 1)
-				req = 0;
+				req = NULL;
 			deselect_ep();
 			if (!is_in) {
 				UDC_CTRL_REG = UDC_SET_FIFO_EN;
@@ -1041,7 +1041,7 @@ omap_ep_queue(struct usb_ep *_ep, struct usb_request *_req, int gfp_flags)
 
 irq_wait:
 	/* irq handler advances the queue */
-	if (req != 0)
+	if (req != NULL)
 		list_add_tail(&req->queue, &ep->queue);
 	spin_unlock_irqrestore(&udc->lock, flags);
 
@@ -1238,6 +1238,8 @@ static int can_pullup(struct omap_udc *udc)
 
 static void pullup_enable(struct omap_udc *udc)
 {
+	udc->gadget.dev.parent->power.power_state = PMSG_ON;
+	udc->gadget.dev.power.power_state = PMSG_ON;
 	UDC_SYSCON1_REG |= UDC_PULLUP_EN;
 #ifndef CONFIG_USB_OTG
 	if (!cpu_is_omap15xx())
@@ -1382,7 +1384,7 @@ static void update_otg(struct omap_udc *udc)
 static void ep0_irq(struct omap_udc *udc, u16 irq_src)
 {
 	struct omap_ep	*ep0 = &udc->ep[0];
-	struct omap_req	*req = 0;
+	struct omap_req	*req = NULL;
 
 	ep0->irqs++;
 
@@ -1438,7 +1440,7 @@ static void ep0_irq(struct omap_udc *udc, u16 irq_src)
 				if (req)
 					done(ep0, req, 0);
 			}
-			req = 0;
+			req = NULL;
 		} else if (stat & UDC_STALL) {
 			UDC_CTRL_REG = UDC_CLR_HALT;
 			UDC_EP_NUM_REG = UDC_EP_DIR;
@@ -1511,9 +1513,6 @@ static void ep0_irq(struct omap_udc *udc, u16 irq_src)
 			u.word[3] = UDC_DATA_REG;
 			UDC_EP_NUM_REG = 0;
 		} while (UDC_IRQ_SRC_REG & UDC_SETUP);
-		le16_to_cpus (&u.r.wValue);
-		le16_to_cpus (&u.r.wIndex);
-		le16_to_cpus (&u.r.wLength);
 
 		/* Delegate almost all control requests to the gadget driver,
 		 * except for a handful of ch9 status/feature requests that
@@ -1566,6 +1565,11 @@ static void ep0_irq(struct omap_udc *udc, u16 irq_src)
 					UDC_CTRL_REG = UDC_SET_FIFO_EN;
 					ep->ackwait = 1 + ep->double_buf;
 				}
+				/* NOTE:  assumes the host behaves sanely,
+				 * only clearing real halts.  Else we may
+				 * need to kill pending transfers and then
+				 * restart the queue... very messy for DMA!
+				 */
 			}
 			VDBG("%s halt cleared by host\n", ep->name);
 			goto ep0out_status_stage;
@@ -2013,7 +2017,7 @@ int usb_gadget_register_driver (struct usb_gadget_driver *driver)
 	udc->softconnect = 1;
 
 	/* hook up the driver */
-	driver->driver.bus = 0;
+	driver->driver.bus = NULL;
 	udc->driver = driver;
 	udc->gadget.dev.driver = &driver->driver;
 	spin_unlock_irqrestore(&udc->lock, flags);
@@ -2021,8 +2025,8 @@ int usb_gadget_register_driver (struct usb_gadget_driver *driver)
 	status = driver->bind (&udc->gadget);
 	if (status) {
 		DBG("bind to %s --> %d\n", driver->driver.name, status);
-		udc->gadget.dev.driver = 0;
-		udc->driver = 0;
+		udc->gadget.dev.driver = NULL;
+		udc->driver = NULL;
 		goto done;
 	}
 	DBG("bound to driver %s\n", driver->driver.name);
@@ -2035,8 +2039,8 @@ int usb_gadget_register_driver (struct usb_gadget_driver *driver)
 		if (status < 0) {
 			ERR("can't bind to transceiver\n");
 			driver->unbind (&udc->gadget);
-			udc->gadget.dev.driver = 0;
-			udc->driver = 0;
+			udc->gadget.dev.driver = NULL;
+			udc->driver = NULL;
 			goto done;
 		}
 	} else {
@@ -2071,7 +2075,7 @@ int usb_gadget_unregister_driver (struct usb_gadget_driver *driver)
 		omap_vbus_session(&udc->gadget, 0);
 
 	if (udc->transceiver)
-		(void) otg_set_peripheral(udc->transceiver, 0);
+		(void) otg_set_peripheral(udc->transceiver, NULL);
 	else
 		pullup_disable(udc);
 
@@ -2080,9 +2084,8 @@ int usb_gadget_unregister_driver (struct usb_gadget_driver *driver)
 	spin_unlock_irqrestore(&udc->lock, flags);
 
 	driver->unbind(&udc->gadget);
-	udc->gadget.dev.driver = 0;
-	udc->driver = 0;
-
+	udc->gadget.dev.driver = NULL;
+	udc->driver = NULL;
 
 	DBG("unregistered driver '%s'\n", driver->driver.name);
 	return status;
@@ -2178,7 +2181,7 @@ static int proc_otg_show(struct seq_file *s)
 
 	tmp = OTG_REV_REG;
 	trans = USB_TRANSCEIVER_CTRL_REG;
-	seq_printf(s, "OTG rev %d.%d, transceiver_ctrl %03x\n",
+	seq_printf(s, "\nOTG rev %d.%d, transceiver_ctrl %03x\n",
 		tmp >> 4, tmp & 0xf, trans);
 	tmp = OTG_SYSCON_1_REG;
 	seq_printf(s, "otg_syscon1 %08x usb2 %s, usb1 %s, usb0 %s,"
@@ -2235,6 +2238,7 @@ static int proc_otg_show(struct seq_file *s)
 	seq_printf(s, "otg_outctrl %04x" "\n", tmp);
 	tmp = OTG_TEST_REG;
 	seq_printf(s, "otg_test    %04x" "\n", tmp);
+	return 0;
 }
 
 static int proc_udc_show(struct seq_file *s, void *_)
@@ -2378,7 +2382,7 @@ static int proc_udc_show(struct seq_file *s, void *_)
 
 static int proc_udc_open(struct inode *inode, struct file *file)
 {
-	return single_open(file, proc_udc_show, 0);
+	return single_open(file, proc_udc_show, NULL);
 }
 
 static struct file_operations proc_ops = {
@@ -2399,7 +2403,7 @@ static void create_proc_file(void)
 
 static void remove_proc_file(void)
 {
-	remove_proc_entry(proc_filename, 0);
+	remove_proc_entry(proc_filename, NULL);
 }
 
 #else
@@ -2505,7 +2509,7 @@ static void omap_udc_release(struct device *dev)
 {
 	complete(udc->done);
 	kfree (udc);
-	udc = 0;
+	udc = NULL;
 }
 
 static int __init
@@ -2577,23 +2581,33 @@ omap_udc_setup(struct platform_device *odev, struct otg_transceiver *xceiv)
 	case 1:
 		OMAP_BULK_EP("ep1in",  USB_DIR_IN  | 1);
 		OMAP_BULK_EP("ep2out", USB_DIR_OUT | 2);
+		OMAP_INT_EP("ep9in",   USB_DIR_IN  | 9, 16);
+
 		OMAP_BULK_EP("ep3in",  USB_DIR_IN  | 3);
 		OMAP_BULK_EP("ep4out", USB_DIR_OUT | 4);
+		OMAP_INT_EP("ep10in",  USB_DIR_IN  | 10, 16);
 
 		OMAP_BULK_EP("ep5in",  USB_DIR_IN  | 5);
 		OMAP_BULK_EP("ep5out", USB_DIR_OUT | 5);
+		OMAP_INT_EP("ep11in",  USB_DIR_IN  | 11, 16);
+
 		OMAP_BULK_EP("ep6in",  USB_DIR_IN  | 6);
 		OMAP_BULK_EP("ep6out", USB_DIR_OUT | 6);
+		OMAP_INT_EP("ep12in",  USB_DIR_IN  | 12, 16);
 
 		OMAP_BULK_EP("ep7in",  USB_DIR_IN  | 7);
 		OMAP_BULK_EP("ep7out", USB_DIR_OUT | 7);
+		OMAP_INT_EP("ep13in",  USB_DIR_IN  | 13, 16);
+		OMAP_INT_EP("ep13out", USB_DIR_OUT | 13, 16);
+
 		OMAP_BULK_EP("ep8in",  USB_DIR_IN  | 8);
 		OMAP_BULK_EP("ep8out", USB_DIR_OUT | 8);
+		OMAP_INT_EP("ep14in",  USB_DIR_IN  | 14, 16);
+		OMAP_INT_EP("ep14out", USB_DIR_OUT | 14, 16);
+
+		OMAP_BULK_EP("ep15in",  USB_DIR_IN  | 15);
+		OMAP_BULK_EP("ep15out", USB_DIR_OUT | 15);
 
-		OMAP_INT_EP("ep9in",   USB_DIR_IN  | 9, 16);
-		OMAP_INT_EP("ep10out", USB_DIR_IN  | 10, 16);
-		OMAP_INT_EP("ep11in",  USB_DIR_IN  | 9, 16);
-		OMAP_INT_EP("ep12out", USB_DIR_IN  | 10, 16);
 		break;
 
 #ifdef	USE_ISO
@@ -2640,8 +2654,8 @@ static int __init omap_udc_probe(struct device *dev)
 	struct platform_device	*odev = to_platform_device(dev);
 	int			status = -ENODEV;
 	int			hmc;
-	struct otg_transceiver	*xceiv = 0;
-	const char		*type = 0;
+	struct otg_transceiver	*xceiv = NULL;
+	const char		*type = NULL;
 	struct omap_usb_config	*config = dev->platform_data;
 
 	/* NOTE:  "knows" the order of the resources! */
@@ -2678,6 +2692,15 @@ static int __init omap_udc_probe(struct device *dev)
 	} else {
 		hmc = HMC_1610;
 		switch (hmc) {
+		case 0:			/* POWERUP DEFAULT == 0 */
+		case 4:
+		case 12:
+		case 20:
+			if (!cpu_is_omap1710()) {
+				type = "integrated";
+				break;
+			}
+			/* FALL THROUGH */
 		case 3:
 		case 11:
 		case 16:
@@ -2688,21 +2711,15 @@ static int __init omap_udc_probe(struct device *dev)
 				DBG("external transceiver not registered!\n");
 				if (config->otg)
 					goto cleanup0;
-				type = "(unknown external)";
+				type = "unknown";
 			} else
 				type = xceiv->label;
 			break;
-		case 0:			/* POWERUP DEFAULT == 0 */
-		case 4:
-		case 12:
-		case 20:
-			type = "INTEGRATED";
-			break;
 		case 21:			/* internal loopback */
-			type = "(loopback)";
+			type = "loopback";
 			break;
 		case 14:			/* transceiverless */
-			type = "(none)";
+			type = "no";
 			break;
 
 		default:
@@ -2710,14 +2727,14 @@ static int __init omap_udc_probe(struct device *dev)
 			return -ENODEV;
 		}
 	}
-	INFO("hmc mode %d, transceiver %s\n", hmc, type);
+	INFO("hmc mode %d, %s transceiver\n", hmc, type);
 
 	/* a "gadget" abstracts/virtualizes the controller */
 	status = omap_udc_setup(odev, xceiv);
 	if (status) {
 		goto cleanup0;
 	}
-	xceiv = 0;
+	xceiv = NULL;
 	// "udc" is now valid
 	pullup_disable(udc);
 #if	defined(CONFIG_USB_OHCI_HCD) || defined(CONFIG_USB_OHCI_HCD_MODULE)
@@ -2765,7 +2782,7 @@ cleanup2:
 
 cleanup1:
 	kfree (udc);
-	udc = 0;
+	udc = NULL;
 
 cleanup0:
 	if (xceiv)
@@ -2788,7 +2805,7 @@ static int __exit omap_udc_remove(struct device *dev)
 	pullup_disable(udc);
 	if (udc->transceiver) {
 		put_device(udc->transceiver->dev);
-		udc->transceiver = 0;
+		udc->transceiver = NULL;
 	}
 	UDC_SYSCON1_REG = 0;
 
@@ -2809,13 +2826,32 @@ static int __exit omap_udc_remove(struct device *dev)
 	return 0;
 }
 
-static int omap_udc_suspend(struct device *dev, pm_message_t state, u32 level)
+/* suspend/resume/wakeup from sysfs (echo > power/state) or when the
+ * system is forced into deep sleep
+ *
+ * REVISIT we should probably reject suspend requests when there's a host
+ * session active, rather than disconnecting, at least on boards that can
+ * report VBUS irqs (UDC_DEVSTAT_REG.UDC_ATT).  And in any case, we need to
+ * make host resumes and VBUS detection trigger OMAP wakeup events; that
+ * may involve talking to an external transceiver (e.g. isp1301).
+ */
+static int omap_udc_suspend(struct device *dev, pm_message_t message, u32 level)
 {
-	if (level != 0)
+	u32	devstat;
+
+	if (level != SUSPEND_POWER_DOWN)
 		return 0;
+	devstat = UDC_DEVSTAT_REG;
+
+	/* we're requesting 48 MHz clock if the pullup is enabled
+	 * (== we're attached to the host) and we're not suspended,
+	 * which would prevent entry to deep sleep...
+	 */
+	if ((devstat & UDC_ATT) != 0 && (devstat & UDC_SUS) == 0) {
+		WARN("session active; suspend requires disconnect\n");
+		omap_pullup(&udc->gadget, 0);
+	}
 
-	DBG("suspend, state %d\n", state);
-	omap_pullup(&udc->gadget, 0);
 	udc->gadget.dev.power.power_state = PMSG_SUSPEND;
 	udc->gadget.dev.parent->power.power_state = PMSG_SUSPEND;
 	return 0;
@@ -2823,7 +2859,7 @@ static int omap_udc_suspend(struct device *dev, pm_message_t state, u32 level)
 
 static int omap_udc_resume(struct device *dev, u32 level)
 {
-	if (level != 0)
+	if (level != RESUME_POWER_ON)
 		return 0;
 
 	DBG("resume + wakeup/SRP\n");
-- 
cgit v1.2.3-70-g09d2


From 9c4142a133f7efee08238722b157656c3da7ca97 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Mon, 27 Jun 2005 14:36:16 -0700
Subject: [PATCH] ppc32: Fix compiling of sandpoint platform

Lost a curly brace in translation.  Everything is better now.

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/platforms/sandpoint.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'arch')

diff --git a/arch/ppc/platforms/sandpoint.c b/arch/ppc/platforms/sandpoint.c
index 70e58f43f2b..8b149c2fc54 100644
--- a/arch/ppc/platforms/sandpoint.c
+++ b/arch/ppc/platforms/sandpoint.c
@@ -324,6 +324,7 @@ sandpoint_setup_arch(void)
 			pdata[1].irq = 0;
 			pdata[1].mapbase = 0;
 		}
+	}
 
 	printk(KERN_INFO "Motorola SPS Sandpoint Test Platform\n");
 	printk(KERN_INFO "Port by MontaVista Software, Inc. (source@mvista.com)\n");
-- 
cgit v1.2.3-70-g09d2


From d4b3a80e399c989028acd5185c792fab82eda035 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
Date: Mon, 27 Jun 2005 14:36:30 -0700
Subject: [PATCH] mips: fixed try_to_freeze build error

arch/mips/kernel/signal.c: In function 'do_signal':
arch/mips/kernel/signal.c:460: error: too many arguments to function 'try_to_freeze'

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/kernel/signal.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/mips/kernel/signal.c b/arch/mips/kernel/signal.c
index 508026ae584..65ee15396ff 100644
--- a/arch/mips/kernel/signal.c
+++ b/arch/mips/kernel/signal.c
@@ -457,7 +457,7 @@ static int do_signal(sigset_t *oldset, struct pt_regs *regs)
 	if (!user_mode(regs))
 		return 1;
 
-	if (try_to_freeze(0))
+	if (try_to_freeze())
 		goto no_signal;
 
 	if (!oldset)
-- 
cgit v1.2.3-70-g09d2


From bb1657468152c5e5232c7bf35cf0e9c41b5d9910 Mon Sep 17 00:00:00 2001
From: Marcelo Tosatti <marcelo.tosatti@cyclades.com>
Date: Mon, 27 Jun 2005 13:09:00 -0300
Subject: [PATCH] 8xx: avoid "dcbst" misbehaviour with unpopulated TLB

The proposed _tlbie call at update_mmu_cache() is safe because:

Addresses for which update_mmu_cache() gets invocated are never inside the
static kernel virtual mapping, meaning that there is no risk for the
_tlbie() here to be thrashing the pinned entry, as Dan suspected.

The intermediate TLB state in which this bug can be triggered is not
visible by userspace or any other contexts, except the page fault handling
path.  So there is no need to worry about userspace dcbxxx users.

The other solution to this is to avoid dcbst misbehaviour in the first
place, which involves changing in-kernel "dcbst" callers to use 8xx
specific SPR's.

Summary:

On 8xx, cache control instructions (particularly "dcbst" from
flush_dcache_icache) fault as write operation if there is an unpopulated
TLB entry for the address in question.  To workaround that, we invalidate
the TLB here, thus avoiding dcbst misbehaviour.

Signed-off-by: Marcelo Tosatti <marcelo.tosatti@cyclades.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/mm/init.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc/mm/init.c b/arch/ppc/mm/init.c
index 334ef4150d9..6164a2b3473 100644
--- a/arch/ppc/mm/init.c
+++ b/arch/ppc/mm/init.c
@@ -606,9 +606,19 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long address,
 		struct page *page = pfn_to_page(pfn);
 		if (!PageReserved(page)
 		    && !test_bit(PG_arch_1, &page->flags)) {
-			if (vma->vm_mm == current->active_mm)
+			if (vma->vm_mm == current->active_mm) {
+#ifdef CONFIG_8xx
+			/* On 8xx, cache control instructions (particularly 
+		 	 * "dcbst" from flush_dcache_icache) fault as write 
+			 * operation if there is an unpopulated TLB entry 
+			 * for the address in question. To workaround that, 
+			 * we invalidate the TLB here, thus avoiding dcbst 
+			 * misbehaviour.
+			 */
+				_tlbie(address);
+#endif
 				__flush_dcache_icache((void *) address);
-			else
+			} else
 				flush_dcache_icache_page(page);
 			set_bit(PG_arch_1, &page->flags);
 		}
-- 
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 'arch')

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 'arch')

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 6ae3db110e62b0846aae1b5c6e661484ee3a5ed1 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Mon, 27 Jun 2005 14:36:35 -0700
Subject: [PATCH] ppc64: Add missing exports

This patch adds a couple of missing symbol exports.  flush_dcache_page is
used by the AGP driver and rtc_lock by the RTC driver.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
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/ppc64/kernel/ppc_ksyms.c | 1 +
 arch/ppc64/kernel/time.c      | 1 +
 2 files changed, 2 insertions(+)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/ppc_ksyms.c b/arch/ppc64/kernel/ppc_ksyms.c
index b230a63fe4c..705742f4eec 100644
--- a/arch/ppc64/kernel/ppc_ksyms.c
+++ b/arch/ppc64/kernel/ppc_ksyms.c
@@ -75,6 +75,7 @@ EXPORT_SYMBOL(giveup_fpu);
 EXPORT_SYMBOL(giveup_altivec);
 #endif
 EXPORT_SYMBOL(__flush_icache_range);
+EXPORT_SYMBOL(flush_dcache_range);
 
 #ifdef CONFIG_SMP
 #ifdef CONFIG_PPC_ISERIES
diff --git a/arch/ppc64/kernel/time.c b/arch/ppc64/kernel/time.c
index 2348a75e050..2a532db9138 100644
--- a/arch/ppc64/kernel/time.c
+++ b/arch/ppc64/kernel/time.c
@@ -91,6 +91,7 @@ unsigned long tb_to_xs;
 unsigned      tb_to_us;
 unsigned long processor_freq;
 DEFINE_SPINLOCK(rtc_lock);
+EXPORT_SYMBOL_GPL(rtc_lock);
 
 unsigned long tb_to_ns_scale;
 unsigned long tb_to_ns_shift;
-- 
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 'arch')

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 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 'arch')

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 4bdbd37f6d01abc4c002bb8de90ea2c3bc7abe7e Mon Sep 17 00:00:00 2001
From: Rusty Lynch <rusty.lynch@intel.com>
Date: Mon, 27 Jun 2005 15:17:09 -0700
Subject: [PATCH] Return probe redesign: i386 specific changes

The following patch contains the i386 specific changes for the new
return probe design.  Changes include:

 * Removing the architecture specific functions for querying a return probe
   instance off a stack address
 * Complete rework onf arch_prepare_kretprobe() and trampoline_probe_handler()
 * Removing trampoline_post_handler()
 * Adding arch_init() so that now we handle registering the return probe
   trampoline instead of kernel/kprobes.c doing it

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/i386/kernel/kprobes.c | 133 ++++++++++++++++++++++++---------------------
 1 file changed, 70 insertions(+), 63 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c
index 3762f6b35ab..fc8b1752176 100644
--- a/arch/i386/kernel/kprobes.c
+++ b/arch/i386/kernel/kprobes.c
@@ -127,48 +127,23 @@ 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;
+        struct kretprobe_instance *ri;
+
+        if ((ri = get_free_rp_inst(rp)) != NULL) {
+                ri->rp = rp;
+                ri->task = current;
+		ri->ret_addr = (kprobe_opcode_t *) *sara;
 
-	/*
-	 * 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);
-	}
+                add_rp_inst(ri);
+        } else {
+                rp->nmissed++;
+        }
 }
 
 /*
@@ -286,36 +261,59 @@ no_kprobe:
  */
 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;
-}
+        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;
 
-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;
+        head = kretprobe_inst_table_head(current);
 
-	while ((ri = get_rp_inst(sara))) {
-		regs->eip = (unsigned long)ri->ret_addr;
+	/*
+	 * 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;
 	}
-	regs->eflags &= ~TF_MASK;
+
+	BUG_ON(!orig_ret_address || (orig_ret_address == trampoline_address));
+	regs->eip = 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;
 }
 
 /*
@@ -403,8 +401,7 @@ static inline int post_kprobe_handler(struct pt_regs *regs)
 		current_kprobe->post_handler(current_kprobe, regs, 0);
 	}
 
-	if (current_kprobe->post_handler != trampoline_post_handler)
-		resume_execution(current_kprobe, regs);
+	resume_execution(current_kprobe, regs);
 	regs->eflags |= kprobe_saved_eflags;
 
 	/*Restore back the original saved kprobes variables and continue. */
@@ -534,3 +531,13 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 	}
 	return 0;
 }
+
+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);
+}
-- 
cgit v1.2.3-70-g09d2


From ba8af12f432c4f00ddb0bc1068b57b20aac93ecf Mon Sep 17 00:00:00 2001
From: Rusty Lynch <rusty.lynch@intel.com>
Date: Mon, 27 Jun 2005 15:17:10 -0700
Subject: [PATCH] Return probe redesign: x86_64 specific changes

The following patch contains the x86_64 specific changes for the new
return probe design.  Changes include:
 * Removing the architecture specific functions for querying a return probe
   instance off a stack address
 * Complete rework onf arch_prepare_kretprobe() and trampoline_probe_handler()
 * Removing trampoline_post_handler()
 * Adding arch_init() so that now we handle registering the return probe
   trampoline instead of kernel/kprobes.c doing it

NOTE:
Note that with this new design, the dependency on calculating a pointer to
the task off the stack pointer no longer exist (resolving the problem of
interruption stacks as pointed out in the original feedback to this port.)

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/x86_64/kernel/kprobes.c | 133 +++++++++++++++++++++++--------------------
 1 file changed, 70 insertions(+), 63 deletions(-)

(limited to 'arch')

diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c
index 6a1c88376be..acd2a778ebe 100644
--- a/arch/x86_64/kernel/kprobes.c
+++ b/arch/x86_64/kernel/kprobes.c
@@ -272,48 +272,23 @@ 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;
+        struct kretprobe_instance *ri;
+
+        if ((ri = get_free_rp_inst(rp)) != NULL) {
+                ri->rp = rp;
+                ri->task = current;
+		ri->ret_addr = (kprobe_opcode_t *) *sara;
 
-	/*
-	 * 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);
-	}
+                add_rp_inst(ri);
+        } else {
+                rp->nmissed++;
+        }
 }
 
 /*
@@ -426,36 +401,59 @@ no_kprobe:
  */
 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;
-}
+        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;
 
-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;
+        head = kretprobe_inst_table_head(current);
 
-	while ((ri = get_rp_inst(sara))) {
-		regs->rip = (unsigned long)ri->ret_addr;
+	/*
+	 * 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;
 	}
-	regs->eflags &= ~TF_MASK;
+
+	BUG_ON(!orig_ret_address || (orig_ret_address == trampoline_address));
+	regs->rip = 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;
 }
 
 /*
@@ -548,8 +546,7 @@ int post_kprobe_handler(struct pt_regs *regs)
 		current_kprobe->post_handler(current_kprobe, regs, 0);
 	}
 
-	if (current_kprobe->post_handler != trampoline_post_handler)
-		resume_execution(current_kprobe, regs);
+	resume_execution(current_kprobe, regs);
 	regs->eflags |= kprobe_saved_rflags;
 
 	/* Restore the original saved kprobes variables and continue. */
@@ -679,3 +676,13 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 	}
 	return 0;
 }
+
+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);
+}
-- 
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 'arch')

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 'arch')

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 a528e21c235862cc1ae50e7809eb9116dc40ea0c 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] kprobes/ia64: refuse inserting kprobe on slot 1

Without the ability to atomically write 16 bytes, we can not update the
middle slot of a bundle, slot 1, unless we stop the machine first.  This
patch will ensure the ability to robustly insert and remove a kprobe by
refusing to insert a kprobe on slot 1 until a mechanism is in place to
safely handle this case.

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 | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'arch')

diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index c97e18e634c..ec2ceade12b 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -270,6 +270,13 @@ static int valid_kprobe_addr(int template, int slot, unsigned long addr)
 				addr);
 		return -EINVAL;
 	}
+
+	if (slot == 1 && bundle_encoding[template][1] != L) {
+		printk(KERN_WARNING "Inserting kprobes on slot #1 "
+		       "is not supported\n");
+		return -EINVAL;
+	}
+
 	return 0;
 }
 
-- 
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 'arch')

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 'arch')

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 'arch')

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 120bb4246a99cc6e9cc976573fcbcd0ee9d544ef Mon Sep 17 00:00:00 2001
From: "jayalk@intworks.biz" <jayalk@intworks.biz>
Date: Mon, 21 Mar 2005 20:20:42 -0800
Subject: [PATCH] PCI Allow OutOfRange PIRQ table address

I updated this to remove unnecessary variable initialization, make
check_routing be inline only and not __init, switch to strtoul, and
formatting fixes as per Randy Dunlap's recommendations.

I updated this to change pirq_table_addr to a long, and to add a warning
msg if the PIRQ table wasn't found at the specified address, as per thread
with Matthew Wilcox.

In our hardware situation, the BIOS is unable to store or generate it's PIRQ
table in the F0000h-100000h standard range. This patch adds a pci kernel
parameter, pirqaddr to allow the bootloader (or BIOS based loader) to inform
the kernel where the PIRQ table got stored. A beneficial side-effect is that,
if one's BIOS uses a static address each time for it's PIRQ table, then
pirqaddr can be used to avoid the $pirq search through that address block each
time at boot for normal PIRQ BIOSes.

Signed-off-by: Jaya Kumar <jayalk@intworks.biz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/kernel-parameters.txt |  4 +++
 arch/i386/pci/common.c              |  6 ++++-
 arch/i386/pci/irq.c                 | 51 +++++++++++++++++++++++++++----------
 arch/i386/pci/pci.h                 |  1 +
 4 files changed, 47 insertions(+), 15 deletions(-)

(limited to 'arch')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index f44bb5567c5..96d924eb7b6 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1030,6 +1030,10 @@ running once the system is up.
 		irqmask=0xMMMM		[IA-32] Set a bit mask of IRQs allowed to be assigned
 					automatically to PCI devices. You can make the kernel
 					exclude IRQs of your ISA cards this way.
+		pirqaddr=0xAAAAA	[IA-32] Specify the physical address
+					of the PIRQ table (normally generated
+					by the BIOS) if it is outside the
+					F0000h-100000h range.
 		lastbus=N		[IA-32] Scan all buses till bus #N. Can be useful
 					if the kernel is unable to find your secondary buses
 					and you want to tell it explicitly which ones they are.
diff --git a/arch/i386/pci/common.c b/arch/i386/pci/common.c
index 720975e1af5..2a2e79fbfef 100644
--- a/arch/i386/pci/common.c
+++ b/arch/i386/pci/common.c
@@ -25,7 +25,8 @@ unsigned int pci_probe = PCI_PROBE_BIOS | PCI_PROBE_CONF1 | PCI_PROBE_CONF2 |
 
 int pci_routeirq;
 int pcibios_last_bus = -1;
-struct pci_bus *pci_root_bus = NULL;
+unsigned long pirq_table_addr;
+struct pci_bus *pci_root_bus;
 struct pci_raw_ops *raw_pci_ops;
 
 static int pci_read(struct pci_bus *bus, unsigned int devfn, int where, int size, u32 *value)
@@ -188,6 +189,9 @@ char * __devinit  pcibios_setup(char *str)
 	} else if (!strcmp(str, "biosirq")) {
 		pci_probe |= PCI_BIOS_IRQ_SCAN;
 		return NULL;
+	} else if (!strncmp(str, "pirqaddr=", 9)) {
+		pirq_table_addr = simple_strtoul(str+9, NULL, 0);
+		return NULL;
 	}
 #endif
 #ifdef CONFIG_PCI_DIRECT
diff --git a/arch/i386/pci/irq.c b/arch/i386/pci/irq.c
index 83458f81e66..78ca1ecbb90 100644
--- a/arch/i386/pci/irq.c
+++ b/arch/i386/pci/irq.c
@@ -57,6 +57,35 @@ struct irq_router_handler {
 
 int (*pcibios_enable_irq)(struct pci_dev *dev) = NULL;
 
+/*
+ *  Check passed address for the PCI IRQ Routing Table signature
+ *  and perform checksum verification.
+ */
+
+static inline struct irq_routing_table * pirq_check_routing_table(u8 *addr)
+{
+	struct irq_routing_table *rt;
+	int i;
+	u8 sum;
+
+	rt = (struct irq_routing_table *) addr;
+	if (rt->signature != PIRQ_SIGNATURE ||
+	    rt->version != PIRQ_VERSION ||
+	    rt->size % 16 ||
+	    rt->size < sizeof(struct irq_routing_table))
+		return NULL;
+	sum = 0;
+	for (i=0; i < rt->size; i++)
+		sum += addr[i];
+	if (!sum) {
+		DBG("PCI: Interrupt Routing Table found at 0x%p\n", rt);
+		return rt;
+	}
+	return NULL;
+}
+
+
+
 /*
  *  Search 0xf0000 -- 0xfffff for the PCI IRQ Routing Table.
  */
@@ -65,23 +94,17 @@ static struct irq_routing_table * __init pirq_find_routing_table(void)
 {
 	u8 *addr;
 	struct irq_routing_table *rt;
-	int i;
-	u8 sum;
 
+	if (pirq_table_addr) {
+		rt = pirq_check_routing_table((u8 *) __va(pirq_table_addr));
+		if (rt)
+			return rt;
+		printk(KERN_WARNING "PCI: PIRQ table NOT found at pirqaddr\n");
+	}
 	for(addr = (u8 *) __va(0xf0000); addr < (u8 *) __va(0x100000); addr += 16) {
-		rt = (struct irq_routing_table *) addr;
-		if (rt->signature != PIRQ_SIGNATURE ||
-		    rt->version != PIRQ_VERSION ||
-		    rt->size % 16 ||
-		    rt->size < sizeof(struct irq_routing_table))
-			continue;
-		sum = 0;
-		for(i=0; i<rt->size; i++)
-			sum += addr[i];
-		if (!sum) {
-			DBG("PCI: Interrupt Routing Table found at 0x%p\n", rt);
+		rt = pirq_check_routing_table(addr);
+		if (rt)
 			return rt;
-		}
 	}
 	return NULL;
 }
diff --git a/arch/i386/pci/pci.h b/arch/i386/pci/pci.h
index a8fc80ca69f..a80f0f55ff5 100644
--- a/arch/i386/pci/pci.h
+++ b/arch/i386/pci/pci.h
@@ -27,6 +27,7 @@
 #define PCI_ASSIGN_ALL_BUSSES	0x4000
 
 extern unsigned int pci_probe;
+extern unsigned long pirq_table_addr;
 
 /* pci-i386.c */
 
-- 
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 'arch')

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 fab3fb0ac8c83072465b28ca859c420da6c6511c 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: Fix pci_enable_device() for p2p bridges

When checking if a PCI to PCI bridge should be enabled to decode memory and/or
IO resources, we need to look at all device resources not just the first 6.
This is needed to allow PCI bridges to pass down memory and IO accesses to
child devices even when the bridge itself does not consume resources in its
PCI BARs.

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/ia64/pci/pci.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c
index c0661d3382e..2a148775849 100644
--- a/arch/ia64/pci/pci.c
+++ b/arch/ia64/pci/pci.c
@@ -418,18 +418,24 @@ pcibios_enable_resources (struct pci_dev *dev, int mask)
 	u16 cmd, old_cmd;
 	int idx;
 	struct resource *r;
+	unsigned long type_mask = IORESOURCE_IO | IORESOURCE_MEM;
 
 	if (!dev)
 		return -EINVAL;
 
 	pci_read_config_word(dev, PCI_COMMAND, &cmd);
 	old_cmd = cmd;
-	for (idx=0; idx<6; idx++) {
+	for (idx=0; idx<PCI_NUM_RESOURCES; idx++) {
 		/* Only set up the desired resources.  */
 		if (!(mask & (1 << idx)))
 			continue;
 
 		r = &dev->resource[idx];
+		if (!(r->flags & type_mask))
+			continue;
+		if ((idx == PCI_ROM_RESOURCE) &&
+				(!(r->flags & IORESOURCE_ROM_ENABLE)))
+			continue;
 		if (!r->start && r->end) {
 			printk(KERN_ERR
 			       "PCI: Device %s not available because of resource collisions\n",
@@ -441,8 +447,6 @@ pcibios_enable_resources (struct pci_dev *dev, int mask)
 		if (r->flags & IORESOURCE_MEM)
 			cmd |= PCI_COMMAND_MEMORY;
 	}
-	if (dev->resource[PCI_ROM_RESOURCE].start)
-		cmd |= PCI_COMMAND_MEMORY;
 	if (cmd != old_cmd) {
 		printk("PCI: Enabling device %s (%04x -> %04x)\n", pci_name(dev), old_cmd, cmd);
 		pci_write_config_word(dev, PCI_COMMAND, cmd);
-- 
cgit v1.2.3-70-g09d2


From 71c3511c22e8e0648094672abec898b3bf84c18b Mon Sep 17 00:00:00 2001
From: Rajesh Shah <rajesh.shah@intel.com>
Date: Thu, 28 Apr 2005 00:25:46 -0700
Subject: [PATCH] acpi bridge hotadd: Make pcibios_fixup_bus() hot-plug safe

PCI scan code calls the arch specific pcibios_fixup_bus() each time it scans a
new bridge.  For root bridge hot-plug, the bridge and it's attached devices
may not have been configured properly yet, so it's not safe to claim those
resources at this time.

This code goes away when we clean up the way pci resources are claimed (in
pci_enable_device()), so this is only a stopgap fix.

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/ia64/pci/pci.c | 22 +++++++++++++++++++++-
 1 file changed, 21 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c
index 2a148775849..d929858cfb3 100644
--- a/arch/ia64/pci/pci.c
+++ b/arch/ia64/pci/pci.c
@@ -373,6 +373,25 @@ void pcibios_bus_to_resource(struct pci_dev *dev,
 	res->end = region->end + offset;
 }
 
+static int __devinit is_valid_resource(struct pci_dev *dev, int idx)
+{
+	unsigned int i, type_mask = IORESOURCE_IO | IORESOURCE_MEM;
+	struct resource *devr = &dev->resource[idx];
+
+	if (!dev->bus)
+		return 0;
+	for (i=0; i<PCI_BUS_NUM_RESOURCES; i++) {
+		struct resource *busr = dev->bus->resource[i];
+
+		if (!busr || ((busr->flags ^ devr->flags) & type_mask))
+			continue;
+		if ((devr->start) && (devr->start >= busr->start) &&
+				(devr->end <= busr->end))
+			return 1;
+	}
+	return 0;
+}
+
 static void __devinit pcibios_fixup_device_resources(struct pci_dev *dev)
 {
 	struct pci_bus_region region;
@@ -386,7 +405,8 @@ static void __devinit pcibios_fixup_device_resources(struct pci_dev *dev)
 		region.start = dev->resource[i].start;
 		region.end = dev->resource[i].end;
 		pcibios_bus_to_resource(dev, &dev->resource[i], &region);
-		pci_claim_resource(dev, i);
+		if ((is_valid_resource(dev, i)))
+			pci_claim_resource(dev, i);
 	}
 }
 
-- 
cgit v1.2.3-70-g09d2


From f7d473d919627262816459f8dba70d72812be074 Mon Sep 17 00:00:00 2001
From: Rajesh Shah <rajesh.shah@intel.com>
Date: Thu, 28 Apr 2005 00:25:51 -0700
Subject: [PATCH] acpi bridge hotadd: Read bridge resources when fixing up the
 bus

Read bridge io/mem/pfmem ranges when fixing up the bus so that bus resources
are tracked.  This is required to properly support pci end device and bridge
hotplug.

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/ia64/pci/pci.c | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'arch')

diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c
index d929858cfb3..720a861f88b 100644
--- a/arch/ia64/pci/pci.c
+++ b/arch/ia64/pci/pci.c
@@ -418,6 +418,10 @@ pcibios_fixup_bus (struct pci_bus *b)
 {
 	struct pci_dev *dev;
 
+	if (b->self) {
+		pci_read_bridge_bases(b);
+		pcibios_fixup_device_resources(b->self);
+	}
 	list_for_each_entry(dev, &b->devices, bus_list)
 		pcibios_fixup_device_resources(dev);
 
-- 
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 'arch')

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 'arch')

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 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 'arch')

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 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 'arch')

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 d57e26ceb7dbf44cd08128cb6146116d4281b58b 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: use the MCFG table to properly access pci devices (i386)

Now that we have access to the whole MCFG table, let's properly use it
for all pci device accesses (as that's what it is there for, some boxes
don't put all the busses into one entry.)

If, for some reason, the table is incorrect, we fallback to the "old
style" of mmconfig accesses, namely, we just assume the first entry in
the table is the one for us, and blindly use it.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/i386/pci/mmconfig.c | 29 +++++++++++++++++++++++++----
 1 file changed, 25 insertions(+), 4 deletions(-)

(limited to 'arch')

diff --git a/arch/i386/pci/mmconfig.c b/arch/i386/pci/mmconfig.c
index 5fbaa913225..60f0e7a1162 100644
--- a/arch/i386/pci/mmconfig.c
+++ b/arch/i386/pci/mmconfig.c
@@ -22,10 +22,31 @@ static u32 mmcfg_last_accessed_device;
 /*
  * Functions for accessing PCI configuration space with MMCONFIG accesses
  */
+static u32 get_base_addr(unsigned int seg, int bus)
+{
+	int cfg_num = -1;
+	struct acpi_table_mcfg_config *cfg;
+
+	while (1) {
+		++cfg_num;
+		if (cfg_num >= pci_mmcfg_config_num) {
+			/* something bad is going on, no cfg table is found. */
+			/* so we fall back to the old way we used to do this */
+			/* and just rely on the first entry to be correct. */
+			return pci_mmcfg_config[0].base_address;
+		}
+		cfg = &pci_mmcfg_config[cfg_num];
+		if (cfg->pci_segment_group_number != seg)
+			continue;
+		if ((cfg->start_bus_number <= bus) &&
+		    (cfg->end_bus_number >= bus))
+			return cfg->base_address;
+	}
+}
 
-static inline void pci_exp_set_dev_base(int bus, int devfn)
+static inline void pci_exp_set_dev_base(unsigned int seg, int bus, int devfn)
 {
-	u32 dev_base = pci_mmcfg_config[0].base_address | (bus << 20) | (devfn << 12);
+	u32 dev_base = get_base_addr(seg, bus) | (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);
@@ -42,7 +63,7 @@ static int pci_mmcfg_read(unsigned int seg, unsigned int bus,
 
 	spin_lock_irqsave(&pci_config_lock, flags);
 
-	pci_exp_set_dev_base(bus, devfn);
+	pci_exp_set_dev_base(seg, bus, devfn);
 
 	switch (len) {
 	case 1:
@@ -71,7 +92,7 @@ static int pci_mmcfg_write(unsigned int seg, unsigned int bus,
 
 	spin_lock_irqsave(&pci_config_lock, flags);
 
-	pci_exp_set_dev_base(bus, devfn);
+	pci_exp_set_dev_base(seg, bus, devfn);
 
 	switch (len) {
 	case 1:
-- 
cgit v1.2.3-70-g09d2


From 1cde8a16815bd85c8137d1ea556398983c597c11 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: use the MCFG table to properly access pci devices
 (x86-64)

Now that we have access to the whole MCFG table, let's properly use it
for all pci device accesses (as that's what it is there for, some boxes
don't put all the busses into one entry.)

If, for some reason, the table is incorrect, we fallback to the "old
style" of mmconfig accesses, namely, we just assume the first entry in
the table is the one for us, and blindly use it.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/x86_64/pci/mmconfig.c | 58 ++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 48 insertions(+), 10 deletions(-)

(limited to 'arch')

diff --git a/arch/x86_64/pci/mmconfig.c b/arch/x86_64/pci/mmconfig.c
index 09cfcc1234b..657e88aa090 100644
--- a/arch/x86_64/pci/mmconfig.c
+++ b/arch/x86_64/pci/mmconfig.c
@@ -13,17 +13,44 @@
 #define MMCONFIG_APER_SIZE (256*1024*1024)
 
 /* Static virtual mapping of the MMCONFIG aperture */
-static char *pci_mmcfg_virt;
+struct mmcfg_virt {
+	struct acpi_table_mcfg_config *cfg;
+	char *virt;
+};
+static struct mmcfg_virt *pci_mmcfg_virt;
 
-static inline char *pci_dev_base(unsigned int bus, unsigned int devfn)
+static char *get_virt(unsigned int seg, int bus)
 {
-	return pci_mmcfg_virt + ((bus << 20) | (devfn << 12));
+	int cfg_num = -1;
+	struct acpi_table_mcfg_config *cfg;
+
+	while (1) {
+		++cfg_num;
+		if (cfg_num >= pci_mmcfg_config_num) {
+			/* something bad is going on, no cfg table is found. */
+			/* so we fall back to the old way we used to do this */
+			/* and just rely on the first entry to be correct. */
+			return pci_mmcfg_virt[0].virt;
+		}
+		cfg = pci_mmcfg_virt[cfg_num].cfg;
+		if (cfg->pci_segment_group_number != seg)
+			continue;
+		if ((cfg->start_bus_number <= bus) &&
+		    (cfg->end_bus_number >= bus))
+			return pci_mmcfg_virt[cfg_num].virt;
+	}
+}
+
+static inline char *pci_dev_base(unsigned int seg, unsigned int bus, unsigned int devfn)
+{
+
+	return get_virt(seg, bus) + ((bus << 20) | (devfn << 12));
 }
 
 static int pci_mmcfg_read(unsigned int seg, unsigned int bus,
 			  unsigned int devfn, int reg, int len, u32 *value)
 {
-	char *addr = pci_dev_base(bus, devfn); 
+	char *addr = pci_dev_base(seg, bus, devfn);
 
 	if (unlikely(!value || (bus > 255) || (devfn > 255) || (reg > 4095)))
 		return -EINVAL;
@@ -46,7 +73,7 @@ static int pci_mmcfg_read(unsigned int seg, unsigned int bus,
 static int pci_mmcfg_write(unsigned int seg, unsigned int bus,
 			   unsigned int devfn, int reg, int len, u32 value)
 {
-	char *addr = pci_dev_base(bus,devfn);
+	char *addr = pci_dev_base(seg, bus, devfn);
 
 	if (unlikely((bus > 255) || (devfn > 255) || (reg > 4095)))
 		return -EINVAL;
@@ -73,6 +100,8 @@ static struct pci_raw_ops pci_mmcfg = {
 
 static int __init pci_mmcfg_init(void)
 {
+	int i;
+
 	if ((pci_probe & PCI_PROBE_MMCONF) == 0)
 		return 0;
 
@@ -90,13 +119,22 @@ 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_config[0].base_address, MMCONFIG_APER_SIZE);
-	if (!pci_mmcfg_virt) { 
-		printk("PCI: Cannot map mmconfig aperture\n");
+	pci_mmcfg_virt = kmalloc(sizeof(*pci_mmcfg_virt) * pci_mmcfg_config_num, GFP_KERNEL);
+	if (pci_mmcfg_virt == NULL) {
+		printk("PCI: Can not allocate memory for mmconfig structures\n");
 		return 0;
-	}	
+	}
+	for (i = 0; i < pci_mmcfg_config_num; ++i) {
+		pci_mmcfg_virt[i].cfg = &pci_mmcfg_config[i];
+		pci_mmcfg_virt[i].virt = ioremap_nocache(pci_mmcfg_config[i].base_address, MMCONFIG_APER_SIZE);
+		if (!pci_mmcfg_virt[i].virt) {
+			printk("PCI: Cannot map mmconfig aperture for segment %d\n",
+			       pci_mmcfg_config[i].pci_segment_group_number);
+			return 0;
+		}
+		printk(KERN_INFO "PCI: Using MMCONFIG at %x\n", pci_mmcfg_config[i].base_address);
+	}
 
-	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;
 
-- 
cgit v1.2.3-70-g09d2


From a341ad97245d01c923995cfe7deacd0c8aee6e16 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Tue, 28 Jun 2005 20:33:49 +1000
Subject: [PATCH] ppc64: simplify nvram partition scanning code

Convert nvram_create_os_partition to use list_for_each_entry
instead of list_for_each, as this reduces the code size by
two lines.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/nvram.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/nvram.c b/arch/ppc64/kernel/nvram.c
index 4e71781a441..4fb1a9f5060 100644
--- a/arch/ppc64/kernel/nvram.c
+++ b/arch/ppc64/kernel/nvram.c
@@ -338,9 +338,8 @@ static int nvram_remove_os_partition(void)
  */
 static int nvram_create_os_partition(void)
 {
-	struct list_head * p;
-	struct nvram_partition *part = NULL;
-	struct nvram_partition *new_part = NULL;
+	struct nvram_partition *part;
+	struct nvram_partition *new_part;
 	struct nvram_partition *free_part = NULL;
 	int seq_init[2] = { 0, 0 };
 	loff_t tmp_index;
@@ -349,8 +348,7 @@ static int nvram_create_os_partition(void)
 	
 	/* Find a free partition that will give us the maximum needed size 
 	   If can't find one that will give us the minimum size needed */
-	list_for_each(p, &nvram_part->partition) {
-		part = list_entry(p, struct nvram_partition, partition);
+	list_for_each_entry(part, &nvram_part->partition, partition) {
 		if (part->header.signature != NVRAM_SIG_FREE)
 			continue;
 
-- 
cgit v1.2.3-70-g09d2


From f5f1cc5437961a4bd93d615099f26780819e72d3 Mon Sep 17 00:00:00 2001
From: Nathan Lynch <ntl@pobox.com>
Date: Tue, 28 Jun 2005 20:55:25 +1000
Subject: [PATCH] ppc64: don't create spurious symlinks under node0 sysdev

On partitioned systems we can wind up creating spurious symlinks in
/sys/devices/system/node/node0 to non-present cpus.  The symlinks are
not broken; the problem is that we're potentially misinforming
userspace that there is a relationship between node0 and cpus which
are to be added later.  There's no guarantee at all that a cpu which
is added later will belong to node 0.

Signed-off-by: Nathan Lynch <ntl@pobox.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/kernel/sysfs.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/sysfs.c b/arch/ppc64/kernel/sysfs.c
index c8fa6569b2f..2f704a2cafb 100644
--- a/arch/ppc64/kernel/sysfs.c
+++ b/arch/ppc64/kernel/sysfs.c
@@ -400,7 +400,12 @@ static int __init topology_init(void)
 		struct cpu *c = &per_cpu(cpu_devices, cpu);
 
 #ifdef CONFIG_NUMA
-		parent = &node_devices[cpu_to_node(cpu)];
+		/* The node to which a cpu belongs can't be known
+		 * until the cpu is made present.
+		 */
+		parent = NULL;
+		if (cpu_present(cpu))
+			parent = &node_devices[cpu_to_node(cpu)];
 #endif
 		/*
 		 * For now, we just see if the system supports making
-- 
cgit v1.2.3-70-g09d2


From 45891f7660c02fdd7b044a04d71b965c795e9df5 Mon Sep 17 00:00:00 2001
From: Olaf Hering <olh@suse.de>
Date: Tue, 28 Jun 2005 21:01:21 +1000
Subject: [PATCH] remove unused arch/ppc64/boot/piggyback.c

piggyback is not called in arch/ppc64/boot/Makefile

Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/boot/Makefile    |  5 +--
 arch/ppc64/boot/piggyback.c | 83 ---------------------------------------------
 2 files changed, 1 insertion(+), 87 deletions(-)
 delete mode 100644 arch/ppc64/boot/piggyback.c

(limited to 'arch')

diff --git a/arch/ppc64/boot/Makefile b/arch/ppc64/boot/Makefile
index d3e1d6af920..683b2d43c15 100644
--- a/arch/ppc64/boot/Makefile
+++ b/arch/ppc64/boot/Makefile
@@ -52,7 +52,7 @@ obj-sec = $(foreach section, $(1), $(patsubst %,$(obj)/kernel-%.o, $(section)))
 src-sec = $(foreach section, $(1), $(patsubst %,$(obj)/kernel-%.c, $(section)))
 gz-sec  = $(foreach section, $(1), $(patsubst %,$(obj)/kernel-%.gz, $(section)))
 
-hostprogs-y		:= piggy addnote addRamDisk
+hostprogs-y		:= addnote addRamDisk
 targets 		+= zImage zImage.initrd imagesize.c \
 			   $(patsubst $(obj)/%,%, $(call obj-sec, $(required) $(initrd))) \
 			   $(patsubst $(obj)/%,%, $(call src-sec, $(required) $(initrd))) \
@@ -78,9 +78,6 @@ addsection = $(CROSS32OBJCOPY) $(1) \
 quiet_cmd_addnote = ADDNOTE $@ 
       cmd_addnote = $(CROSS32LD) $(BOOTLFLAGS) -o $@ $(obj-boot) && $(obj)/addnote $@
 
-quiet_cmd_piggy = PIGGY   $@
-      cmd_piggy = $(obj)/piggyback $(@:.o=) < $< | $(CROSS32AS) -o $@
-
 $(call gz-sec, $(required)): $(obj)/kernel-%.gz: % FORCE
 	$(call if_changed,gzip)
 
diff --git a/arch/ppc64/boot/piggyback.c b/arch/ppc64/boot/piggyback.c
deleted file mode 100644
index 235c7a87269..00000000000
--- a/arch/ppc64/boot/piggyback.c
+++ /dev/null
@@ -1,83 +0,0 @@
-/*
- * Copyright 2001 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.
- */
-#include <stdio.h>
-#include <unistd.h>
-#include <string.h>
-
-extern long ce_exec_config[];
-
-int main(int argc, char *argv[])
-{
-	int i, cnt, pos, len;
-	unsigned int cksum, val;
-	unsigned char *lp;
-	unsigned char buf[8192];
-	char *varname;
-	if (argc != 2)
-	{
-		fprintf(stderr, "usage: %s name <in-file >out-file\n",
-			argv[0]);
-		exit(1);
-	}
-
-	varname = strrchr(argv[1], '/');
-	if (varname)
-		varname++;
-	else
-		varname = argv[1];
-
-	fprintf(stdout, "#\n");
-	fprintf(stdout, "# Miscellaneous data structures:\n");
-	fprintf(stdout, "# WARNING - this file is automatically generated!\n");
-	fprintf(stdout, "#\n");
-	fprintf(stdout, "\n");
-	fprintf(stdout, "\t.data\n");
-	fprintf(stdout, "\t.globl %s_data\n", varname);
-	fprintf(stdout, "%s_data:\n", varname);
-	pos = 0;
-	cksum = 0;
-	while ((len = read(0, buf, sizeof(buf))) > 0)
-	{
-		cnt = 0;
-		lp = (unsigned char *)buf;
-		len = (len + 3) & ~3;  /* Round up to longwords */
-		for (i = 0;  i < len;  i += 4)
-		{
-			if (cnt == 0)
-			{
-				fprintf(stdout, "\t.long\t");
-			}
-			fprintf(stdout, "0x%02X%02X%02X%02X", lp[0], lp[1], lp[2], lp[3]);
-			val = *(unsigned long *)lp;
-			cksum ^= val;
-			lp += 4;
-			if (++cnt == 4)
-			{
-				cnt = 0;
-				fprintf(stdout, " # %x \n", pos+i-12);
-				fflush(stdout);
-			} else
-			{
-				fprintf(stdout, ",");
-			}
-		}
-		if (cnt)
-		{
-			fprintf(stdout, "0\n");
-		}
-		pos += len;
-	}
-	fprintf(stdout, "\t.globl %s_len\n", varname);
-	fprintf(stdout, "%s_len:\t.long\t0x%x\n", varname, pos);
-	fflush(stdout);
-	fclose(stdout);
-	fprintf(stderr, "cksum = %x\n", cksum);
-	exit(0);
-}
-
-- 
cgit v1.2.3-70-g09d2


From a58dfbbb2a350808253a8b5037f5ec5b9a68516d Mon Sep 17 00:00:00 2001
From: Olaf Hering <olh@suse.de>
Date: Tue, 28 Jun 2005 21:01:28 +1000
Subject: [PATCH] remove unused arch/ppc64/boot/mknote.c

mknote is not called in arch/ppc64/boot/Makefile

Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/boot/mknote.c | 43 -------------------------------------------
 1 file changed, 43 deletions(-)
 delete mode 100644 arch/ppc64/boot/mknote.c

(limited to 'arch')

diff --git a/arch/ppc64/boot/mknote.c b/arch/ppc64/boot/mknote.c
deleted file mode 100644
index 120cc1d8973..00000000000
--- a/arch/ppc64/boot/mknote.c
+++ /dev/null
@@ -1,43 +0,0 @@
-/*
- * Copyright (C) Cort Dougan 1999.
- *
- * 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.
- *
- * Generate a note section as per the CHRP specification.
- *
- */
-
-#include <stdio.h>
-
-#define PL(x) printf("%c%c%c%c", ((x)>>24)&0xff, ((x)>>16)&0xff, ((x)>>8)&0xff, (x)&0xff );
-
-int main(void)
-{
-/* header */
-	/* namesz */
-	PL(strlen("PowerPC")+1);
-	/* descrsz */
-	PL(6*4);
-	/* type */
-	PL(0x1275);
-	/* name */
-	printf("PowerPC"); printf("%c", 0);
-	
-/* descriptor */
-	/* real-mode */
-	PL(0xffffffff);
-	/* real-base */
-	PL(0x00c00000);
-	/* real-size */
-	PL(0xffffffff);
-	/* virt-base */
-	PL(0xffffffff);
-	/* virt-size */
-	PL(0xffffffff);
-	/* load-base */
-	PL(0x4000);
-	return 0;
-}
-- 
cgit v1.2.3-70-g09d2


From e6019db5a7f110839c62cefc073b6dc1143d6403 Mon Sep 17 00:00:00 2001
From: Olaf Hering <olh@suse.de>
Date: Tue, 28 Jun 2005 21:01:35 +1000
Subject: [PATCH] remove printk usage in arch/ppc64/boot/prom.c

remove the printk usage in the zImage. we are not there, yet.

Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/boot/main.c |  1 -
 arch/ppc64/boot/prom.c | 16 ++--------------
 2 files changed, 2 insertions(+), 15 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/boot/main.c b/arch/ppc64/boot/main.c
index da12ea2ca46..677e4187497 100644
--- a/arch/ppc64/boot/main.c
+++ b/arch/ppc64/boot/main.c
@@ -17,7 +17,6 @@
 
 extern void *finddevice(const char *);
 extern int getprop(void *, const char *, void *, int);
-extern void printk(char *fmt, ...);
 extern void printf(const char *fmt, ...);
 extern int sprintf(char *buf, const char *fmt, ...);
 void gunzip(void *, int, unsigned char *, int *);
diff --git a/arch/ppc64/boot/prom.c b/arch/ppc64/boot/prom.c
index d5218b15824..5e48b80ff5a 100644
--- a/arch/ppc64/boot/prom.c
+++ b/arch/ppc64/boot/prom.c
@@ -40,7 +40,7 @@ void *finddevice(const char *name);
 int getprop(void *phandle, const char *name, void *buf, int buflen);
 void chrpboot(int a1, int a2, void *prom);	/* in main.c */
 
-void printk(char *fmt, ...);
+int printf(char *fmt, ...);
 
 /* there is no convenient header to get this from...  -- paulus */
 extern unsigned long strlen(const char *);
@@ -220,7 +220,7 @@ readchar(void)
 		case 1:
 			return ch;
 		case -1:
-			printk("read(stdin) returned -1\r\n");
+			printf("read(stdin) returned -1\r\n");
 			return -1;
 		}
 	}
@@ -627,18 +627,6 @@ int sprintf(char * buf, const char *fmt, ...)
 
 static char sprint_buf[1024];
 
-void
-printk(char *fmt, ...)
-{
-	va_list args;
-	int n;
-
-	va_start(args, fmt);
-	n = vsprintf(sprint_buf, fmt, args);
-	va_end(args);
-	write(stdout, sprint_buf, n);
-}
-
 int
 printf(char *fmt, ...)
 {
-- 
cgit v1.2.3-70-g09d2


From b1bdfbd0a29d6da4dbe42736faac02c43a9afe76 Mon Sep 17 00:00:00 2001
From: Olaf Hering <olh@suse.de>
Date: Tue, 28 Jun 2005 21:01:46 +1000
Subject: [PATCH] remove duplicate printf in arch/ppc64/boot/main.c

initrd size is printed as hex, add a missing 0x
remove a duplicate printf when initrd is used.
remove use of kernel type to access the first bytes of the initrd memarea.

Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc64/boot/main.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/boot/main.c b/arch/ppc64/boot/main.c
index 677e4187497..199d9804f61 100644
--- a/arch/ppc64/boot/main.c
+++ b/arch/ppc64/boot/main.c
@@ -146,10 +146,10 @@ void start(unsigned long a1, unsigned long a2, void *promptr)
 		}
 		a1 = initrd.addr;
 		a2 = initrd.size;
-		printf("initial ramdisk moving 0x%lx <- 0x%lx (%lx bytes)\n\r",
+		printf("initial ramdisk moving 0x%lx <- 0x%lx (0x%lx bytes)\n\r",
 		       initrd.addr, (unsigned long)_initrd_start, initrd.size);
 		memmove((void *)initrd.addr, (void *)_initrd_start, initrd.size);
-		printf("initrd head: 0x%lx\n\r", *((u32 *)initrd.addr));
+		printf("initrd head: 0x%lx\n\r", *((unsigned long *)initrd.addr));
 	}
 
 	/* Eventually gunzip the kernel */
@@ -200,9 +200,6 @@ void start(unsigned long a1, unsigned long a2, void *promptr)
 
 	flush_cache((void *)vmlinux.addr, vmlinux.size);
 
-	if (a1)
-		printf("initrd head: 0x%lx\n\r", *((u32 *)initrd.addr));
-
 	kernel_entry = (kernel_entry_t)vmlinux.addr;
 #ifdef DEBUG
 	printf( "kernel:\n\r"
-- 
cgit v1.2.3-70-g09d2


From 564c90aa07cd43dc434d46cef8a15773a23d49a2 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Tue, 28 Jun 2005 13:46:09 +0100
Subject: [PATCH] ARM SMP: Use local_flush_tlb* where we really want to be
 local

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/init.c    | 2 +-
 arch/arm/mm/mm-armv.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index 6dcb23d64bf..edffa47a4b2 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -437,7 +437,7 @@ void __init paging_init(struct meminfo *mi, struct machine_desc *mdesc)
 	memtable_init(mi);
 	if (mdesc->map_io)
 		mdesc->map_io();
-	flush_tlb_all();
+	local_flush_tlb_all();
 
 	/*
 	 * initialise the zones within each node
diff --git a/arch/arm/mm/mm-armv.c b/arch/arm/mm/mm-armv.c
index 052ab443ec4..c3bd503b43a 100644
--- a/arch/arm/mm/mm-armv.c
+++ b/arch/arm/mm/mm-armv.c
@@ -682,7 +682,7 @@ void __init memtable_init(struct meminfo *mi)
 	}
 
 	flush_cache_all();
-	flush_tlb_all();
+	local_flush_tlb_all();
 
 	top_pmd = pmd_off_k(0xffff0000);
 }
-- 
cgit v1.2.3-70-g09d2


From 4b0ef3b1127776d4a2787d7530ac0c4da82c2331 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Tue, 28 Jun 2005 13:49:16 +0100
Subject: [PATCH] ARM SMP: Add IPI support code for SMP TLB flushing

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/smp.c | 123 ++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 123 insertions(+)

(limited to 'arch')

diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 34892758f09..a931409c8fe 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -502,3 +502,126 @@ int __init setup_profiling_timer(unsigned int multiplier)
 {
 	return -EINVAL;
 }
+
+static int
+on_each_cpu_mask(void (*func)(void *), void *info, int retry, int wait,
+		 cpumask_t mask)
+{
+	int ret = 0;
+
+	preempt_disable();
+
+	ret = smp_call_function_on_cpu(func, info, retry, wait, mask);
+	if (cpu_isset(smp_processor_id(), mask))
+		func(info);
+
+	preempt_enable();
+
+	return ret;
+}
+
+/**********************************************************************/
+
+/*
+ * TLB operations
+ */
+struct tlb_args {
+	struct vm_area_struct *ta_vma;
+	unsigned long ta_start;
+	unsigned long ta_end;
+};
+
+static inline void ipi_flush_tlb_all(void *ignored)
+{
+	local_flush_tlb_all();
+}
+
+static inline void ipi_flush_tlb_mm(void *arg)
+{
+	struct mm_struct *mm = (struct mm_struct *)arg;
+
+	local_flush_tlb_mm(mm);
+}
+
+static inline void ipi_flush_tlb_page(void *arg)
+{
+	struct tlb_args *ta = (struct tlb_args *)arg;
+
+	local_flush_tlb_page(ta->ta_vma, ta->ta_start);
+}
+
+static inline void ipi_flush_tlb_kernel_page(void *arg)
+{
+	struct tlb_args *ta = (struct tlb_args *)arg;
+
+	local_flush_tlb_kernel_page(ta->ta_start);
+}
+
+static inline void ipi_flush_tlb_range(void *arg)
+{
+	struct tlb_args *ta = (struct tlb_args *)arg;
+
+	local_flush_tlb_range(ta->ta_vma, ta->ta_start, ta->ta_end);
+}
+
+static inline void ipi_flush_tlb_kernel_range(void *arg)
+{
+	struct tlb_args *ta = (struct tlb_args *)arg;
+
+	local_flush_tlb_kernel_range(ta->ta_start, ta->ta_end);
+}
+
+void flush_tlb_all(void)
+{
+	on_each_cpu(ipi_flush_tlb_all, NULL, 1, 1);
+}
+
+void flush_tlb_mm(struct mm_struct *mm)
+{
+	cpumask_t mask = mm->cpu_vm_mask;
+
+	on_each_cpu_mask(ipi_flush_tlb_mm, mm, 1, 1, mask);
+}
+
+void flush_tlb_page(struct vm_area_struct *vma, unsigned long uaddr)
+{
+	cpumask_t mask = vma->vm_mm->cpu_vm_mask;
+	struct tlb_args ta;
+
+	ta.ta_vma = vma;
+	ta.ta_start = uaddr;
+
+	on_each_cpu_mask(ipi_flush_tlb_page, &ta, 1, 1, mask);
+}
+
+void flush_tlb_kernel_page(unsigned long kaddr)
+{
+	struct tlb_args ta;
+
+	ta.ta_start = kaddr;
+
+	on_each_cpu(ipi_flush_tlb_kernel_page, &ta, 1, 1);
+}
+
+void flush_tlb_range(struct vm_area_struct *vma,
+                     unsigned long start, unsigned long end)
+{
+	cpumask_t mask = vma->vm_mm->cpu_vm_mask;
+	struct tlb_args ta;
+
+	ta.ta_vma = vma;
+	ta.ta_start = start;
+	ta.ta_end = end;
+
+	on_each_cpu_mask(ipi_flush_tlb_range, &ta, 1, 1, mask);
+}
+
+void flush_tlb_kernel_range(unsigned long start, unsigned long end)
+{
+	struct tlb_args ta;
+
+	ta.ta_start = start;
+	ta.ta_end = end;
+
+	on_each_cpu(ipi_flush_tlb_kernel_range, &ta, 1, 1);
+}
-- 
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 'arch')

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 a9f9de7378e3a81f7266f542f2290399a298ef52 Mon Sep 17 00:00:00 2001
From: Mark Maule <maule@sgi.com>
Date: Tue, 26 Apr 2005 08:01:00 -0700
Subject: [IA64-SGI] altix: enable vgacon support

Altix patch to enable use of vgacon driver on that platform.  Depends on the
PCDP generalization patch discussed at:

	http://marc.theaimsgroup.com/?l=linux-ia64&m=111446235101939&w=2

Signed-off-by: Mark Maule <maule@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/configs/sn2_defconfig |  4 ++--
 arch/ia64/sn/kernel/iomv.c      |  6 ++++++
 arch/ia64/sn/kernel/setup.c     | 16 ++++++++++------
 3 files changed, 18 insertions(+), 8 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/configs/sn2_defconfig b/arch/ia64/configs/sn2_defconfig
index 487d2e36b0a..c0561398030 100644
--- a/arch/ia64/configs/sn2_defconfig
+++ b/arch/ia64/configs/sn2_defconfig
@@ -99,7 +99,7 @@ CONFIG_ACPI_DEALLOCATE_IRQ=y
 # Firmware Drivers
 #
 CONFIG_EFI_VARS=y
-# CONFIG_EFI_PCDP is not set
+CONFIG_EFI_PCDP=y
 CONFIG_BINFMT_ELF=y
 # CONFIG_BINFMT_MISC is not set
 
@@ -650,7 +650,7 @@ CONFIG_MMTIMER=y
 #
 # Console display driver support
 #
-# CONFIG_VGA_CONSOLE is not set
+CONFIG_VGA_CONSOLE=y
 CONFIG_DUMMY_CONSOLE=y
 
 #
diff --git a/arch/ia64/sn/kernel/iomv.c b/arch/ia64/sn/kernel/iomv.c
index fec6d8b8237..7ce3cdad627 100644
--- a/arch/ia64/sn/kernel/iomv.c
+++ b/arch/ia64/sn/kernel/iomv.c
@@ -9,12 +9,16 @@
 #include <linux/module.h>
 #include <asm/io.h>
 #include <asm/delay.h>
+#include <asm/vga.h>
 #include <asm/sn/nodepda.h>
 #include <asm/sn/simulator.h>
 #include <asm/sn/pda.h>
 #include <asm/sn/sn_cpuid.h>
 #include <asm/sn/shub_mmr.h>
 
+#define IS_LEGACY_VGA_IOPORT(p) \
+	(((p) >= 0x3b0 && (p) <= 0x3bb) || ((p) >= 0x3c0 && (p) <= 0x3df))
+
 /**
  * sn_io_addr - convert an in/out port to an i/o address
  * @port: port to convert
@@ -26,6 +30,8 @@
 void *sn_io_addr(unsigned long port)
 {
 	if (!IS_RUNNING_ON_SIMULATOR()) {
+		if (IS_LEGACY_VGA_IOPORT(port))
+			port += vga_console_iobase;
 		/* On sn2, legacy I/O ports don't point at anything */
 		if (port < (64 * 1024))
 			return NULL;
diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c
index 44bfc7f318c..3e5ba34ac2e 100644
--- a/arch/ia64/sn/kernel/setup.c
+++ b/arch/ia64/sn/kernel/setup.c
@@ -36,6 +36,7 @@
 #include <asm/machvec.h>
 #include <asm/system.h>
 #include <asm/processor.h>
+#include <asm/vga.h>
 #include <asm/sn/arch.h>
 #include <asm/sn/addrs.h>
 #include <asm/sn/pda.h>
@@ -273,14 +274,17 @@ void __init sn_setup(char **cmdline_p)
 
 	ia64_sn_plat_set_error_handling_features();
 
+#if defined(CONFIG_VT) && defined(CONFIG_VGA_CONSOLE)
 	/*
-	 * If the generic code has enabled vga console support - lets
-	 * get rid of it again. This is a kludge for the fact that ACPI
-	 * currtently has no way of informing us if legacy VGA is available
-	 * or not.
+	 * If there was a primary vga adapter identified through the
+	 * EFI PCDP table, make it the preferred console.  Otherwise
+	 * zero out conswitchp.
 	 */
-#if defined(CONFIG_VT) && defined(CONFIG_VGA_CONSOLE)
-	if (conswitchp == &vga_con) {
+
+	if (vga_console_membase) {
+		/* usable vga ... make tty0 the preferred default console */
+		add_preferred_console("tty", 0, NULL);
+	} else {
 		printk(KERN_DEBUG "SGI: Disabling VGA console\n");
 #ifdef CONFIG_DUMMY_CONSOLE
 		conswitchp = &dummy_con;
-- 
cgit v1.2.3-70-g09d2


From e15da40176f16050c8d92b36387a49370317e276 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Tue, 3 May 2005 12:07:00 -0700
Subject: [IA64] Recognize HP sx2000 chipset

No functional change, just identify the device nicely:

  -IOC: Unknown (103c:12ec) 0.1 HPA 0xf8020002000 IOVA space 1024Mb at 0x40000000
  +IOC: sx2000 0.1 HPA 0xf8020002000 IOVA space 1024Mb at 0x40000000

We used to create fake PCI devices for these chips, but we no longer do that.
So I don't think there's any reason to touch pci_ids.h now.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/hp/common/sba_iommu.c | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'arch')

diff --git a/arch/ia64/hp/common/sba_iommu.c b/arch/ia64/hp/common/sba_iommu.c
index b8db6e3e5e8..11957598a8b 100644
--- a/arch/ia64/hp/common/sba_iommu.c
+++ b/arch/ia64/hp/common/sba_iommu.c
@@ -156,10 +156,13 @@
 */
 #define DELAYED_RESOURCE_CNT	64
 
+#define PCI_DEVICE_ID_HP_SX2000_IOC	0x12ec
+
 #define ZX1_IOC_ID	((PCI_DEVICE_ID_HP_ZX1_IOC << 16) | PCI_VENDOR_ID_HP)
 #define ZX2_IOC_ID	((PCI_DEVICE_ID_HP_ZX2_IOC << 16) | PCI_VENDOR_ID_HP)
 #define REO_IOC_ID	((PCI_DEVICE_ID_HP_REO_IOC << 16) | PCI_VENDOR_ID_HP)
 #define SX1000_IOC_ID	((PCI_DEVICE_ID_HP_SX1000_IOC << 16) | PCI_VENDOR_ID_HP)
+#define SX2000_IOC_ID	((PCI_DEVICE_ID_HP_SX2000_IOC << 16) | PCI_VENDOR_ID_HP)
 
 #define ZX1_IOC_OFFSET	0x1000	/* ACPI reports SBA, we want IOC */
 
@@ -1726,6 +1729,7 @@ static struct ioc_iommu ioc_iommu_info[] __initdata = {
 	{ ZX1_IOC_ID, "zx1", ioc_zx1_init },
 	{ ZX2_IOC_ID, "zx2", NULL },
 	{ SX1000_IOC_ID, "sx1000", NULL },
+	{ SX2000_IOC_ID, "sx2000", NULL },
 };
 
 static struct ioc * __init
-- 
cgit v1.2.3-70-g09d2


From 458f935527372499b714bf4f8e646a68bb0f52e3 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 4 May 2005 13:25:00 -0700
Subject: [IA64] Speed up lfetch.fault [NULL]

This patch greatly speeds up the handling of lfetch.fault instructions
which result in NaT consumption. Due to the NaT-page mapped at address
0, this is guaranteed to happen when lfetch.fault'ing a NULL pointer.
With this patch in place, we can even define prefetch()/prefetchw() as
lfetch.fault without significant performance degradation.  More
importantly, it allows compilers to be more aggressive with using
lfetch.fault on pointers that might be NULL.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/ivt.S | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

(limited to 'arch')

diff --git a/arch/ia64/kernel/ivt.S b/arch/ia64/kernel/ivt.S
index b28d2212a77..3bb3a13c404 100644
--- a/arch/ia64/kernel/ivt.S
+++ b/arch/ia64/kernel/ivt.S
@@ -1243,6 +1243,25 @@ END(disabled_fp_reg)
 // 0x5600 Entry 26 (size 16 bundles) Nat Consumption (11,23,37,50)
 ENTRY(nat_consumption)
 	DBG_FAULT(26)
+
+	mov r16=cr.ipsr
+	mov r17=cr.isr
+	mov r31=pr				// save PR
+	;;
+	and r18=0xf,r17				// r18 = cr.ipsr.code{3:0}
+	tbit.z p6,p0=r17,IA64_ISR_NA_BIT
+	;;
+	cmp.ne.or p6,p0=IA64_ISR_CODE_LFETCH,r18
+	dep r16=-1,r16,IA64_PSR_ED_BIT,1
+(p6)	br.cond.spnt 1f		// branch if (cr.ispr.na == 0 || cr.ipsr.code{3:0} != LFETCH)
+	;;
+	mov cr.ipsr=r16		// set cr.ipsr.na
+	mov pr=r31,-1
+	;;
+	rfi
+
+1:	mov pr=r31,-1
+	;;
 	FAULT(26)
 END(nat_consumption)
 
-- 
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 'arch')

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 'arch')

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 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 'arch')

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 'arch')

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 819c67e69c4e0062787e27dd989f5f9d00d4834e Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Thu, 9 Jun 2005 22:40:00 -0700
Subject: [IA64] Replace stale KDB-code with useful MAGIC_SYSRQ code in
 simserial.c

Patch makes it possible to use the "F4" function key to do
magic sysrq in the HP Ski simulator.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/hp/sim/simserial.c | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/hp/sim/simserial.c b/arch/ia64/hp/sim/simserial.c
index 786e70718ce..7a8ae0f4b38 100644
--- a/arch/ia64/hp/sim/simserial.c
+++ b/arch/ia64/hp/sim/simserial.c
@@ -30,6 +30,7 @@
 #include <linux/module.h>
 #include <linux/serial.h>
 #include <linux/serialP.h>
+#include <linux/sysrq.h>
 
 #include <asm/irq.h>
 #include <asm/hw_irq.h>
@@ -149,12 +150,17 @@ static  void receive_chars(struct tty_struct *tty, struct pt_regs *regs)
 				seen_esc = 2;
 				continue;
 			} else if ( seen_esc == 2 ) {
-				if ( ch == 'P' ) show_state();		/* F1 key */
-#ifdef CONFIG_KDB
-				if ( ch == 'S' )
-					kdb(KDB_REASON_KEYBOARD, 0, (kdb_eframe_t) regs);
+				if ( ch == 'P' ) /* F1 */
+					show_state();
+#ifdef CONFIG_MAGIC_SYSRQ
+				if ( ch == 'S' ) { /* F4 */
+					do
+						ch = ia64_ssc(0, 0, 0, 0,
+							      SSC_GETCHAR);
+					while (!ch);
+					handle_sysrq(ch, regs, NULL);
+				}
 #endif
-
 				seen_esc = 0;
 				continue;
 			}
-- 
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 'arch')

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 fa0ebff62a8a25952a082641db0ed6b1cfc6c38c Mon Sep 17 00:00:00 2001
From: Richard Purdie <rpurdie@rpsys.net>
Date: Tue, 28 Jun 2005 21:01:03 +0100
Subject: [PATCH] ARM: 2761/1: OProfile: Add call graphing support for arm

Patch from Richard Purdie

Add functions to generate backtraces of both kernel and user processes
which allows oprofile's call graphing functionality to be used on arm.
This requires unstripped binaries/libs which use a frame pointer.

Signed-off-by: Richard Purdie
Signed-off-by: Zwane Mwaikambo <zwane@arm.linux.org.uk>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/oprofile/Makefile       |   2 +-
 arch/arm/oprofile/backtrace.c    | 144 +++++++++++++++++++++++++++++++++++++++
 arch/arm/oprofile/init.c         |   2 +
 arch/arm/oprofile/op_arm_model.h |   2 +
 4 files changed, 149 insertions(+), 1 deletion(-)
 create mode 100644 arch/arm/oprofile/backtrace.c

(limited to 'arch')

diff --git a/arch/arm/oprofile/Makefile b/arch/arm/oprofile/Makefile
index ba1a6e9f2b2..8ffb523e6c7 100644
--- a/arch/arm/oprofile/Makefile
+++ b/arch/arm/oprofile/Makefile
@@ -6,6 +6,6 @@ DRIVER_OBJS = $(addprefix ../../../drivers/oprofile/, \
 		oprofilefs.o oprofile_stats.o \
 		timer_int.o )
 
-oprofile-y				:= $(DRIVER_OBJS) init.o
+oprofile-y				:= $(DRIVER_OBJS) init.o backtrace.o
 oprofile-$(CONFIG_CPU_XSCALE)		+= common.o op_model_xscale.o
 
diff --git a/arch/arm/oprofile/backtrace.c b/arch/arm/oprofile/backtrace.c
new file mode 100644
index 00000000000..ec58d3e2eb8
--- /dev/null
+++ b/arch/arm/oprofile/backtrace.c
@@ -0,0 +1,144 @@
+/*
+ * Arm specific backtracing code for oprofile
+ *
+ * Copyright 2005 Openedhand Ltd.
+ *
+ * Author: Richard Purdie <rpurdie@openedhand.com>
+ *
+ * Based on i386 oprofile backtrace code by John Levon, David Smith
+ *
+ * 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/oprofile.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <asm/ptrace.h>
+#include <asm/uaccess.h>
+
+
+/*
+ * The registers we're interested in are at the end of the variable
+ * length saved register structure. The fp points at the end of this
+ * structure so the address of this struct is:
+ * (struct frame_tail *)(xxx->fp)-1
+ */
+struct frame_tail {
+	struct frame_tail *fp;
+	unsigned long sp;
+	unsigned long lr;
+} __attribute__((packed));
+
+
+#ifdef CONFIG_FRAME_POINTER
+static struct frame_tail* kernel_backtrace(struct frame_tail *tail)
+{
+	oprofile_add_trace(tail->lr);
+
+	/* frame pointers should strictly progress back up the stack
+	 * (towards higher addresses) */
+	if (tail >= tail->fp)
+		return NULL;
+
+	return tail->fp-1;
+}
+#endif
+
+static struct frame_tail* user_backtrace(struct frame_tail *tail)
+{
+	struct frame_tail buftail;
+
+	/* hardware pte might not be valid due to dirty/accessed bit emulation
+	 * so we use copy_from_user and benefit from exception fixups */
+	if (copy_from_user(&buftail, tail, sizeof(struct frame_tail)))
+		return NULL;
+
+	oprofile_add_trace(buftail.lr);
+
+	/* frame pointers should strictly progress back up the stack
+	 * (towards higher addresses) */
+	if (tail >= buftail.fp)
+		return NULL;
+
+	return buftail.fp-1;
+}
+
+/* Compare two addresses and see if they're on the same page */
+#define CMP_ADDR_EQUAL(x,y,offset) ((((unsigned long) x) >> PAGE_SHIFT) \
+	== ((((unsigned long) y) + offset) >> PAGE_SHIFT))
+
+/* check that the page(s) containing the frame tail are present */
+static int pages_present(struct frame_tail *tail)
+{
+	struct mm_struct * mm = current->mm;
+
+	if (!check_user_page_readable(mm, (unsigned long)tail))
+		return 0;
+
+	if (CMP_ADDR_EQUAL(tail, tail, 8))
+		return 1;
+
+	if (!check_user_page_readable(mm, ((unsigned long)tail) + 8))
+		return 0;
+
+	return 1;
+}
+
+/*
+ * |             | /\ Higher addresses
+ * |             |
+ * --------------- stack base (address of current_thread_info)
+ * | thread info |
+ * .             .
+ * |    stack    |
+ * --------------- saved regs->ARM_fp value if valid (frame_tail address)
+ * .             .
+ * --------------- struct pt_regs stored on stack (struct pt_regs *)
+ * |             |
+ * .             .
+ * |             |
+ * --------------- %esp
+ * |             |
+ * |             | \/ Lower addresses
+ *
+ * Thus, &pt_regs <-> stack base restricts the valid(ish) fp values
+ */
+static int valid_kernel_stack(struct frame_tail *tail, struct pt_regs *regs)
+{
+	unsigned long tailaddr = (unsigned long)tail;
+	unsigned long stack = (unsigned long)regs;
+	unsigned long stack_base = (stack & ~(THREAD_SIZE - 1)) + THREAD_SIZE;
+
+	return (tailaddr > stack) && (tailaddr < stack_base);
+}
+
+void arm_backtrace(struct pt_regs const *regs, unsigned int depth)
+{
+	struct frame_tail *tail;
+	unsigned long last_address = 0;
+
+	tail = ((struct frame_tail *) regs->ARM_fp) - 1;
+
+	if (!user_mode(regs)) {
+
+#ifdef CONFIG_FRAME_POINTER
+		while (depth-- && tail && valid_kernel_stack(tail, regs)) {
+			tail = kernel_backtrace(tail);
+		}
+#endif
+		return;
+	}
+
+	while (depth-- && tail && !((unsigned long) tail & 3)) {
+		if ((!CMP_ADDR_EQUAL(last_address, tail, 0)
+			|| !CMP_ADDR_EQUAL(last_address, tail, 8))
+				&& !pages_present(tail))
+			return;
+		last_address = (unsigned long) tail;
+		tail = user_backtrace(tail);
+	}
+}
+
diff --git a/arch/arm/oprofile/init.c b/arch/arm/oprofile/init.c
index cce3d3015eb..d315a3a86c8 100644
--- a/arch/arm/oprofile/init.c
+++ b/arch/arm/oprofile/init.c
@@ -20,6 +20,8 @@ int __init oprofile_arch_init(struct oprofile_operations *ops)
 	ret = pmu_init(ops, &op_xscale_spec);
 #endif
 
+	ops->backtrace = arm_backtrace;
+
 	return ret;
 }
 
diff --git a/arch/arm/oprofile/op_arm_model.h b/arch/arm/oprofile/op_arm_model.h
index 2d4caf4781a..2148d07484b 100644
--- a/arch/arm/oprofile/op_arm_model.h
+++ b/arch/arm/oprofile/op_arm_model.h
@@ -24,6 +24,8 @@ struct op_arm_model_spec {
 extern struct op_arm_model_spec op_xscale_spec;
 #endif
 
+extern void arm_backtrace(struct pt_regs * const regs, unsigned int depth);
+
 extern int __init pmu_init(struct oprofile_operations *ops, struct op_arm_model_spec *spec);
 extern void pmu_exit(void);
 #endif /* OP_ARM_MODEL_H */
-- 
cgit v1.2.3-70-g09d2


From f2be64b3b15f52ceb83ff26abc10705a31ae5352 Mon Sep 17 00:00:00 2001
From: Tony Lindgren <tony@atomide.com>
Date: Tue, 28 Jun 2005 21:01:16 +0100
Subject: [PATCH] ARM: 2760/1: Add a warning on dynamic tick timekeeping on
 some platforms

Patch from Tony Lindgren

This patch was suggested by RMK, and adds a warning on the accuracy
of timekeeping when using dynamic tick on some platforms. Depending
on the timer implementation, dynamic tick may affect the accuracy of
timekeeping.
Currently at least OMAP is known to have accurate timekeeping with
dynamic tick.

Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'arch')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index c8d94dcd8ef..620f2ca94ed 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -361,6 +361,11 @@ config NO_IDLE_HZ
 	  Alternatively, if you want dynamic tick automatically enabled
 	  during boot, pass "dyntick=enable" via the kernel command string.
 
+	  Please note that dynamic tick may affect the accuracy of
+	  timekeeping on some platforms depending on the implementation.
+	  Currently at least OMAP platform is known to have accurate
+	  timekeeping with dynamic tick.
+
 config ARCH_DISCONTIGMEM_ENABLE
 	bool
 	default (ARCH_LH7A40X && !LH7A40X_CONTIGMEM)
-- 
cgit v1.2.3-70-g09d2


From 50273978975ee62938b7ca612e281927cd789c13 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Tue, 28 Jun 2005 22:42:06 +0100
Subject: [PATCH] ARM: 2763/1: S3C24XX - mark IRQ_LCD as valid

Patch from Ben Dooks

Fix the IRQ_LCD so that it is marked as valid
since we no longer de-mux this in the main IRQ
handler.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-s3c2410/irq.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/mach-s3c2410/irq.c b/arch/arm/mach-s3c2410/irq.c
index b668c48f439..cf9f46d8806 100644
--- a/arch/arm/mach-s3c2410/irq.c
+++ b/arch/arm/mach-s3c2410/irq.c
@@ -40,8 +40,11 @@
  *   04-Nov-2004  Ben Dooks
  *		  Fix standard IRQ wake for EINT0..4 and RTC
  *
- *   22-Feb-2004  Ben Dooks
+ *   22-Feb-2005  Ben Dooks
  *		  Fixed edge-triggering on ADC IRQ
+ *
+ *   28-Jun-2005  Ben Dooks
+ *		  Mark IRQ_LCD valid
 */
 
 #include <linux/init.h>
@@ -366,7 +369,6 @@ static struct irqchip s3c_irq_eint0t4 = {
 #define INTMSK_UART1	 (1UL << (IRQ_UART1 - IRQ_EINT0))
 #define INTMSK_UART2	 (1UL << (IRQ_UART2 - IRQ_EINT0))
 #define INTMSK_ADCPARENT (1UL << (IRQ_ADCPARENT - IRQ_EINT0))
-#define INTMSK_LCD	 (1UL << (IRQ_LCD - IRQ_EINT0))
 
 static inline void
 s3c_irqsub_mask(unsigned int irqno, unsigned int parentbit,
@@ -716,7 +718,6 @@ void __init s3c24xx_init_irq(void)
 		case IRQ_UART0:
 		case IRQ_UART1:
 		case IRQ_UART2:
-		case IRQ_LCD:
 		case IRQ_ADCPARENT:
 			set_irq_chip(irqno, &s3c_irq_level_chip);
 			set_irq_handler(irqno, do_level_IRQ);
-- 
cgit v1.2.3-70-g09d2


From a8f5034540195307362d071a8b387226b410469f Mon Sep 17 00:00:00 2001
From: "KAMBAROV, ZAUR" <kambarov@berkeley.edu>
Date: Tue, 28 Jun 2005 20:45:06 -0700
Subject: [PATCH] coverity: i386: build.c: negative return to unsigned fix

Variable "c" was declared as an unsigned int, but used in:

125  		for (i=0 ; (c=read(fd, buf, sizeof(buf)))>0 ; i+=c )
126  			if (write(1, buf, c) != c)
127  				die("Write call failed");

(akpm: read() can return -1.  If it does, we fill the disk up with garbage).

Signed-off-by: Zaur Kambarov <zkambarov@coverity.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/boot/tools/build.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/i386/boot/tools/build.c b/arch/i386/boot/tools/build.c
index 4a17956512e..6835f6d47c3 100644
--- a/arch/i386/boot/tools/build.c
+++ b/arch/i386/boot/tools/build.c
@@ -70,7 +70,8 @@ void usage(void)
 
 int main(int argc, char ** argv)
 {
-	unsigned int i, c, sz, setup_sectors;
+	unsigned int i, sz, setup_sectors;
+	int c;
 	u32 sys_size;
 	byte major_root, minor_root;
 	struct stat sb;
-- 
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 'arch')

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 2668fb91cbfbd5e6799017a3579f20309aac758b Mon Sep 17 00:00:00 2001
From: William Lee Irwin III <wli@holomorphy.com>
Date: Tue, 28 Jun 2005 20:45:31 -0700
Subject: [PATCH] sparc32: Kconfig fixups

Something reverted most of the arch/sparc/Kconfig changes, leaving
arch/sparc/ unconfigurable.  This patch re-removes the parts made redundant
by drivers/Kconfig in addition to a mysterious, spurious second instance of
source "mm/Kconfig".  cvs strikes again?

Signed-off-by: William Irwin <wli@holomorphy.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/sparc/Kconfig | 56 ------------------------------------------------------
 1 file changed, 56 deletions(-)

(limited to 'arch')

diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 262e13d086f..7a117ef473c 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -270,66 +270,10 @@ endmenu
 
 source "drivers/Kconfig"
 
-config PRINTER
-	tristate "Parallel printer support"
-	depends on PARPORT
-	---help---
-	  If you intend to attach a printer to the parallel port of your Linux
-	  box (as opposed to using a serial printer; if the connector at the
-	  printer has 9 or 25 holes ["female"], then it's serial), say Y.
-	  Also read the Printing-HOWTO, available from
-	  <http://www.tldp.org/docs.html#howto>.
-
-	  It is possible to share one parallel port among several devices
-	  (e.g. printer and ZIP drive) and it is safe to compile the
-	  corresponding drivers into the kernel.  If you want to compile this
-	  driver as a module however, choose M here and read
-	  <file:Documentation/parport.txt>.  The module will be called lp.
-
-	  If you have several parallel ports, you can specify which ports to
-	  use with the "lp" kernel command line option.  (Try "man bootparam"
-	  or see the documentation of your boot loader (silo) about how to pass
-	  options to the kernel at boot time.)  The syntax of the "lp" command
-	  line option can be found in <file:drivers/char/lp.c>.
-
-	  If you have more than 8 printers, you need to increase the LP_NO
-	  macro in lp.c and the PARPORT_MAX macro in parport.h.
-
-source "mm/Kconfig"
-
-endmenu
-
-source "drivers/base/Kconfig"
-
-source "drivers/video/Kconfig"
-
-source "drivers/mtd/Kconfig"
-
-source "drivers/serial/Kconfig"
-
 if !SUN4
 source "drivers/sbus/char/Kconfig"
 endif
 
-source "drivers/block/Kconfig"
-
-# Don't frighten a common SBus user
-if PCI
-
-source "drivers/ide/Kconfig"
-
-endif
-
-source "drivers/isdn/Kconfig"
-
-source "drivers/scsi/Kconfig"
-
-source "drivers/fc4/Kconfig"
-
-source "drivers/md/Kconfig"
-
-source "net/Kconfig"
-
 # This one must be before the filesystem configs. -DaveM
 
 menu "Unix98 PTY support"
-- 
cgit v1.2.3-70-g09d2


From f705b1aed7002495a504086705ff9725ed9412dc Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Wed, 29 Jun 2005 11:09:15 +0100
Subject: [PATCH] ARM: 2764/1: S3C24XX - Common PM functions for Simtec boards

Patch from Ben Dooks

All current S3C24XX implementations from Simtec share the same
requirements for suspend/resume information.
This patch moves the save code out of the mach-bast.c file,
and into it's own so it can be shared by all the current
Simtec S3C24XX implementations.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-s3c2410/Kconfig     |  5 +++
 arch/arm/mach-s3c2410/Makefile    |  1 +
 arch/arm/mach-s3c2410/mach-bast.c | 39 +++--------------------
 arch/arm/mach-s3c2410/pm-simtec.c | 65 +++++++++++++++++++++++++++++++++++++++
 4 files changed, 75 insertions(+), 35 deletions(-)
 create mode 100644 arch/arm/mach-s3c2410/pm-simtec.c

(limited to 'arch')

diff --git a/arch/arm/mach-s3c2410/Kconfig b/arch/arm/mach-s3c2410/Kconfig
index 534df0c6c77..d4d03d0daae 100644
--- a/arch/arm/mach-s3c2410/Kconfig
+++ b/arch/arm/mach-s3c2410/Kconfig
@@ -154,6 +154,11 @@ config S3C2410_PM_CHECK_CHUNKSIZE
 	  the CRC data block will take more memory, but wil identify any
 	  faults with better precision.
 
+config PM_SIMTEC
+	bool
+	depends on PM && (ARCH_BAST || MACH_VR1000)
+	default y
+
 config S3C2410_LOWLEVEL_UART_PORT
 	int "S3C2410 UART to use for low-level messages"
 	default 0
diff --git a/arch/arm/mach-s3c2410/Makefile b/arch/arm/mach-s3c2410/Makefile
index 7c379aad5d6..f99b689e439 100644
--- a/arch/arm/mach-s3c2410/Makefile
+++ b/arch/arm/mach-s3c2410/Makefile
@@ -18,6 +18,7 @@ obj-$(CONFIG_S3C2410_DMA)  += dma.o
 # Power Management support
 
 obj-$(CONFIG_PM)	   += pm.o sleep.o
+obj-$(CONFIG_PM_SIMTEC)	   += pm-simtec.o
 
 # S3C2440 support
 
diff --git a/arch/arm/mach-s3c2410/mach-bast.c b/arch/arm/mach-s3c2410/mach-bast.c
index f3e970039b6..549bcb1f32c 100644
--- a/arch/arm/mach-s3c2410/mach-bast.c
+++ b/arch/arm/mach-s3c2410/mach-bast.c
@@ -27,6 +27,7 @@
  *     10-Mar-2005 LCVR Changed S3C2410_VA to S3C24XX_VA
  *     14-Mar-2006 BJD  Updated for __iomem changes
  *     22-Jun-2006 BJD  Added DM9000 platform information
+ *     28-Jun-2006 BJD  Moved pm functionality out to common code
 */
 
 #include <linux/kernel.h>
@@ -67,7 +68,6 @@
 #include "devs.h"
 #include "cpu.h"
 #include "usb-simtec.h"
-#include "pm.h"
 
 #define COPYRIGHT ", (c) 2004-2005 Simtec Electronics"
 
@@ -405,44 +405,13 @@ void __init bast_map_io(void)
 	usb_simtec_init();
 }
 
-void __init bast_init_irq(void)
-{
-	s3c24xx_init_irq();
-}
-
-#ifdef CONFIG_PM
-
-/* bast_init_machine
- *
- * enable the power management functions for the EB2410ITX
-*/
-
-static __init void bast_init_machine(void)
-{
-	unsigned long gstatus4;
-
-	printk(KERN_INFO "BAST Power Manangement" COPYRIGHT "\n");
-
-	gstatus4  = (__raw_readl(S3C2410_BANKCON7) & 0x3) << 30;
-	gstatus4 |= (__raw_readl(S3C2410_BANKCON6) & 0x3) << 28;
-	gstatus4 |= (__raw_readl(S3C2410_BANKSIZE) & S3C2410_BANKSIZE_MASK);
-
-	__raw_writel(gstatus4, S3C2410_GSTATUS4);
-
-	s3c2410_pm_init();
-}
-
-#else
-#define bast_init_machine NULL
-#endif
-
 
 MACHINE_START(BAST, "Simtec-BAST")
      MAINTAINER("Ben Dooks <ben@simtec.co.uk>")
      BOOT_MEM(S3C2410_SDRAM_PA, S3C2410_PA_UART, (u32)S3C24XX_VA_UART)
      BOOT_PARAMS(S3C2410_SDRAM_PA + 0x100)
-     MAPIO(bast_map_io)
-     INITIRQ(bast_init_irq)
-	.init_machine	= bast_init_machine,
+
+	.map_io		= bast_map_io,
+	.init_irq	= s3c24xx_init_irq,
 	.timer		= &s3c24xx_timer,
 MACHINE_END
diff --git a/arch/arm/mach-s3c2410/pm-simtec.c b/arch/arm/mach-s3c2410/pm-simtec.c
new file mode 100644
index 00000000000..2cb79883222
--- /dev/null
+++ b/arch/arm/mach-s3c2410/pm-simtec.c
@@ -0,0 +1,65 @@
+/* linux/arch/arm/mach-s3c2410/pm-simtec.c
+ *
+ * Copyright (c) 2004 Simtec Electronics
+ *	Ben Dooks <ben@simtec.co.uk>
+ *
+ * http://armlinux.simtec.co.uk/
+ *
+ * Power Management helpers for Simtec S3C24XX implementations
+ *
+ * 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/kernel.h>
+#include <linux/types.h>
+#include <linux/interrupt.h>
+#include <linux/list.h>
+#include <linux/timer.h>
+#include <linux/init.h>
+#include <linux/device.h>
+
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+
+#include <asm/hardware.h>
+#include <asm/io.h>
+
+#include <asm/arch/map.h>
+#include <asm/arch/regs-serial.h>
+#include <asm/arch/regs-gpio.h>
+#include <asm/arch/regs-mem.h>
+
+#include <asm/mach-types.h>
+
+#include "pm.h"
+
+#define COPYRIGHT ", (c) 2005 Simtec Electronics"
+
+/* pm_simtec_init
+ *
+ * enable the power management functions
+*/
+
+static __init int pm_simtec_init(void)
+{
+	unsigned long gstatus4;
+
+	/* check which machine we are running on */
+
+	if (!machine_is_bast() && !machine_is_vr1000())
+		return 0;
+
+	printk(KERN_INFO "Simtec Board Power Manangement" COPYRIGHT "\n");
+
+	gstatus4  = (__raw_readl(S3C2410_BANKCON7) & 0x3) << 30;
+	gstatus4 |= (__raw_readl(S3C2410_BANKCON6) & 0x3) << 28;
+	gstatus4 |= (__raw_readl(S3C2410_BANKSIZE) & S3C2410_BANKSIZE_MASK);
+
+	__raw_writel(gstatus4, S3C2410_GSTATUS4);
+
+	return s3c2410_pm_init();
+}
+
+arch_initcall(pm_simtec_init);
-- 
cgit v1.2.3-70-g09d2


From 6904b2465cf30265a67711025815784aa345bfad Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Wed, 29 Jun 2005 11:09:15 +0100
Subject: [PATCH] ARM: 2765/1: S3C24XX - small cleanups in
 arch/arm/mach-s3c2410

Patch from Ben Dooks

Re-tab the devs.c file, and change the initialiser for the
mach-vr1000.c to use `.xxx = yyy` form.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-s3c2410/devs.c        | 4 ++--
 arch/arm/mach-s3c2410/mach-vr1000.c | 8 ++------
 2 files changed, 4 insertions(+), 8 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/mach-s3c2410/devs.c b/arch/arm/mach-s3c2410/devs.c
index 64792f67866..4664bd11adc 100644
--- a/arch/arm/mach-s3c2410/devs.c
+++ b/arch/arm/mach-s3c2410/devs.c
@@ -96,8 +96,8 @@ struct platform_device s3c_device_lcd = {
 	.num_resources	  = ARRAY_SIZE(s3c_lcd_resource),
 	.resource	  = s3c_lcd_resource,
 	.dev              = {
-		.dma_mask = &s3c_device_lcd_dmamask,
-		.coherent_dma_mask = 0xffffffffUL
+		.dma_mask		= &s3c_device_lcd_dmamask,
+		.coherent_dma_mask	= 0xffffffffUL
 	}
 };
 
diff --git a/arch/arm/mach-s3c2410/mach-vr1000.c b/arch/arm/mach-s3c2410/mach-vr1000.c
index 76be074944a..1db2855e3e5 100644
--- a/arch/arm/mach-s3c2410/mach-vr1000.c
+++ b/arch/arm/mach-s3c2410/mach-vr1000.c
@@ -371,16 +371,12 @@ void __init vr1000_map_io(void)
 	usb_simtec_init();
 }
 
-void __init vr1000_init_irq(void)
-{
-	s3c24xx_init_irq();
-}
 
 MACHINE_START(VR1000, "Thorcom-VR1000")
      MAINTAINER("Ben Dooks <ben@simtec.co.uk>")
      BOOT_MEM(S3C2410_SDRAM_PA, S3C2410_PA_UART, (u32)S3C24XX_VA_UART)
      BOOT_PARAMS(S3C2410_SDRAM_PA + 0x100)
-     MAPIO(vr1000_map_io)
-     INITIRQ(vr1000_init_irq)
+	.map_io		= vr1000_map_io,
+	.init_irq	= s3c24xx_init_irq,
 	.timer		= &s3c24xx_timer,
 MACHINE_END
-- 
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 'arch')

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 aaaa3f9e51245be3215ff67bb3c7aaf5abb82f00 Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Wed, 29 Jun 2005 15:34:39 +0100
Subject: [PATCH] ARM: 2769/1: cpu_init() stack setup fix

Patch from Catalin Marinas

The compiler allocates r14 for the stk variable in the __asm__ directive.
This is a shadowed register and gets changed when the mode is changed,
causing random values in the SP register. The patch adds a clobber for
the r14 register.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/setup.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index 8cf733daa80..35b7273cfdb 100644
--- a/arch/arm/kernel/setup.c
+++ b/arch/arm/kernel/setup.c
@@ -359,7 +359,8 @@ void cpu_init(void)
 	      "I" (offsetof(struct stack, abt[0])),
 	      "I" (PSR_F_BIT | PSR_I_BIT | UND_MODE),
 	      "I" (offsetof(struct stack, und[0])),
-	      "I" (PSR_F_BIT | PSR_I_BIT | SVC_MODE));
+	      "I" (PSR_F_BIT | PSR_I_BIT | SVC_MODE)
+	    : "r14");
 }
 
 static struct machine_desc * __init setup_machine(unsigned int nr)
-- 
cgit v1.2.3-70-g09d2


From c7e788766610bdc764d7150e69ace2e0c4196cf0 Mon Sep 17 00:00:00 2001
From: Nicolas Pitre <nico@cam.org>
Date: Wed, 29 Jun 2005 18:10:54 +0100
Subject: [PATCH] ARM: 2723/2: remove __udivdi3 and __umoddi3 from the kernel

Patch from Nicolas Pitre

Those are big, slow and generally not recommended for kernel code.
They are even not present on i386.  So it should be concluded that
one could as well get away with do_div() alone.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/armksyms.c |   6 --
 arch/arm/lib/Makefile      |   2 +-
 arch/arm/lib/longlong.h    | 183 -------------------------------------
 arch/arm/lib/udivdi3.c     | 222 ---------------------------------------------
 drivers/net/arm/etherh.c   |  16 ++--
 5 files changed, 8 insertions(+), 421 deletions(-)
 delete mode 100644 arch/arm/lib/longlong.h
 delete mode 100644 arch/arm/lib/udivdi3.c

(limited to 'arch')

diff --git a/arch/arm/kernel/armksyms.c b/arch/arm/kernel/armksyms.c
index 4c38bd8bc29..b713c44c6fb 100644
--- a/arch/arm/kernel/armksyms.c
+++ b/arch/arm/kernel/armksyms.c
@@ -30,9 +30,6 @@ extern void __lshrdi3(void);
 extern void __modsi3(void);
 extern void __muldi3(void);
 extern void __ucmpdi2(void);
-extern void __udivdi3(void);
-extern void __umoddi3(void);
-extern void __udivmoddi4(void);
 extern void __udivsi3(void);
 extern void __umodsi3(void);
 extern void __do_div64(void);
@@ -134,9 +131,6 @@ EXPORT_SYMBOL(__lshrdi3);
 EXPORT_SYMBOL(__modsi3);
 EXPORT_SYMBOL(__muldi3);
 EXPORT_SYMBOL(__ucmpdi2);
-EXPORT_SYMBOL(__udivdi3);
-EXPORT_SYMBOL(__umoddi3);
-EXPORT_SYMBOL(__udivmoddi4);
 EXPORT_SYMBOL(__udivsi3);
 EXPORT_SYMBOL(__umodsi3);
 EXPORT_SYMBOL(__do_div64);
diff --git a/arch/arm/lib/Makefile b/arch/arm/lib/Makefile
index c0e65833ffc..8725d63e421 100644
--- a/arch/arm/lib/Makefile
+++ b/arch/arm/lib/Makefile
@@ -11,7 +11,7 @@ lib-y		:= backtrace.o changebit.o csumipv6.o csumpartial.o   \
 		   strnlen_user.o strchr.o strrchr.o testchangebit.o  \
 		   testclearbit.o testsetbit.o uaccess.o getuser.o    \
 		   putuser.o ashldi3.o ashrdi3.o lshrdi3.o muldi3.o   \
-		   ucmpdi2.o udivdi3.o lib1funcs.o div64.o	      \
+		   ucmpdi2.o lib1funcs.o div64.o	              \
 		   io-readsb.o io-writesb.o io-readsl.o io-writesl.o
 
 ifeq ($(CONFIG_CPU_32v3),y)
diff --git a/arch/arm/lib/longlong.h b/arch/arm/lib/longlong.h
deleted file mode 100644
index 90ae647e4d7..00000000000
--- a/arch/arm/lib/longlong.h
+++ /dev/null
@@ -1,183 +0,0 @@
-/* longlong.h -- based on code from gcc-2.95.3
-
-   definitions for mixed size 32/64 bit arithmetic.
-   Copyright (C) 1991, 92, 94, 95, 96, 1997, 1998 Free Software Foundation, Inc.
-
-   This definition file 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 definition file 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.  */
-
-/* Borrowed from GCC 2.95.3, I Molton 29/07/01 */
-
-#ifndef SI_TYPE_SIZE
-#define SI_TYPE_SIZE 32
-#endif
-
-#define __BITS4 (SI_TYPE_SIZE / 4)
-#define __ll_B (1L << (SI_TYPE_SIZE / 2))
-#define __ll_lowpart(t) ((u32) (t) % __ll_B)
-#define __ll_highpart(t) ((u32) (t) / __ll_B)
-
-/* Define auxiliary asm macros.
-
-   1) umul_ppmm(high_prod, low_prod, multipler, multiplicand)
-   multiplies two u32 integers MULTIPLER and MULTIPLICAND,
-   and generates a two-part u32 product in HIGH_PROD and
-   LOW_PROD.
-
-   2) __umulsidi3(a,b) multiplies two u32 integers A and B,
-   and returns a u64 product.  This is just a variant of umul_ppmm.
-
-   3) udiv_qrnnd(quotient, remainder, high_numerator, low_numerator,
-   denominator) divides a two-word unsigned integer, composed by the
-   integers HIGH_NUMERATOR and LOW_NUMERATOR, by DENOMINATOR and
-   places the quotient in QUOTIENT and the remainder in REMAINDER.
-   HIGH_NUMERATOR must be less than DENOMINATOR for correct operation.
-   If, in addition, the most significant bit of DENOMINATOR must be 1,
-   then the pre-processor symbol UDIV_NEEDS_NORMALIZATION is defined to 1.
-
-   4) sdiv_qrnnd(quotient, remainder, high_numerator, low_numerator,
-   denominator).  Like udiv_qrnnd but the numbers are signed.  The
-   quotient is rounded towards 0.
-
-   5) count_leading_zeros(count, x) counts the number of zero-bits from
-   the msb to the first non-zero bit.  This is the number of steps X
-   needs to be shifted left to set the msb.  Undefined for X == 0.
-
-   6) add_ssaaaa(high_sum, low_sum, high_addend_1, low_addend_1,
-   high_addend_2, low_addend_2) adds two two-word unsigned integers,
-   composed by HIGH_ADDEND_1 and LOW_ADDEND_1, and HIGH_ADDEND_2 and
-   LOW_ADDEND_2 respectively.  The result is placed in HIGH_SUM and
-   LOW_SUM.  Overflow (i.e. carry out) is not stored anywhere, and is
-   lost.
-
-   7) sub_ddmmss(high_difference, low_difference, high_minuend,
-   low_minuend, high_subtrahend, low_subtrahend) subtracts two
-   two-word unsigned integers, composed by HIGH_MINUEND_1 and
-   LOW_MINUEND_1, and HIGH_SUBTRAHEND_2 and LOW_SUBTRAHEND_2
-   respectively.  The result is placed in HIGH_DIFFERENCE and
-   LOW_DIFFERENCE.  Overflow (i.e. carry out) is not stored anywhere,
-   and is lost.
-
-   If any of these macros are left undefined for a particular CPU,
-   C macros are used.  */
-
-#if defined (__arm__)
-#define add_ssaaaa(sh, sl, ah, al, bh, bl) \
-  __asm__ ("adds	%1, %4, %5					\n\
-	adc	%0, %2, %3"						\
-	   : "=r" ((u32) (sh)),					\
-	     "=&r" ((u32) (sl))					\
-	   : "%r" ((u32) (ah)),					\
-	     "rI" ((u32) (bh)),					\
-	     "%r" ((u32) (al)),					\
-	     "rI" ((u32) (bl)))
-#define sub_ddmmss(sh, sl, ah, al, bh, bl) \
-  __asm__ ("subs	%1, %4, %5					\n\
-	sbc	%0, %2, %3"						\
-	   : "=r" ((u32) (sh)),					\
-	     "=&r" ((u32) (sl))					\
-	   : "r" ((u32) (ah)),					\
-	     "rI" ((u32) (bh)),					\
-	     "r" ((u32) (al)),					\
-	     "rI" ((u32) (bl)))
-#define umul_ppmm(xh, xl, a, b) \
-{register u32 __t0, __t1, __t2;					\
-  __asm__ ("%@ Inlined umul_ppmm					\n\
-	mov	%2, %5, lsr #16						\n\
-	mov	%0, %6, lsr #16						\n\
-	bic	%3, %5, %2, lsl #16					\n\
-	bic	%4, %6, %0, lsl #16					\n\
-	mul	%1, %3, %4						\n\
-	mul	%4, %2, %4						\n\
-	mul	%3, %0, %3						\n\
-	mul	%0, %2, %0						\n\
-	adds	%3, %4, %3						\n\
-	addcs	%0, %0, #65536						\n\
-	adds	%1, %1, %3, lsl #16					\n\
-	adc	%0, %0, %3, lsr #16"					\
-	   : "=&r" ((u32) (xh)),					\
-	     "=r" ((u32) (xl)),					\
-	     "=&r" (__t0), "=&r" (__t1), "=r" (__t2)			\
-	   : "r" ((u32) (a)),					\
-	     "r" ((u32) (b)));}
-#define UMUL_TIME 20
-#define UDIV_TIME 100
-#endif				/* __arm__ */
-
-#define __umulsidi3(u, v) \
-  ({DIunion __w;							\
-    umul_ppmm (__w.s.high, __w.s.low, u, v);				\
-    __w.ll; })
-
-#define __udiv_qrnnd_c(q, r, n1, n0, d) \
-  do {									\
-    u32 __d1, __d0, __q1, __q0;					\
-    u32 __r1, __r0, __m;						\
-    __d1 = __ll_highpart (d);						\
-    __d0 = __ll_lowpart (d);						\
-									\
-    __r1 = (n1) % __d1;							\
-    __q1 = (n1) / __d1;							\
-    __m = (u32) __q1 * __d0;					\
-    __r1 = __r1 * __ll_B | __ll_highpart (n0);				\
-    if (__r1 < __m)							\
-      {									\
-	__q1--, __r1 += (d);						\
-	if (__r1 >= (d)) /* i.e. we didn't get carry when adding to __r1 */\
-	  if (__r1 < __m)						\
-	    __q1--, __r1 += (d);					\
-      }									\
-    __r1 -= __m;							\
-									\
-    __r0 = __r1 % __d1;							\
-    __q0 = __r1 / __d1;							\
-    __m = (u32) __q0 * __d0;					\
-    __r0 = __r0 * __ll_B | __ll_lowpart (n0);				\
-    if (__r0 < __m)							\
-      {									\
-	__q0--, __r0 += (d);						\
-	if (__r0 >= (d))						\
-	  if (__r0 < __m)						\
-	    __q0--, __r0 += (d);					\
-      }									\
-    __r0 -= __m;							\
-									\
-    (q) = (u32) __q1 * __ll_B | __q0;				\
-    (r) = __r0;								\
-  } while (0)
-
-#define UDIV_NEEDS_NORMALIZATION 1
-#define udiv_qrnnd __udiv_qrnnd_c
-
-#define count_leading_zeros(count, x) \
-  do {									\
-    u32 __xr = (x);							\
-    u32 __a;							\
-									\
-    if (SI_TYPE_SIZE <= 32)						\
-      {									\
-	__a = __xr < ((u32)1<<2*__BITS4)				\
-	  ? (__xr < ((u32)1<<__BITS4) ? 0 : __BITS4)		\
-	  : (__xr < ((u32)1<<3*__BITS4) ?  2*__BITS4 : 3*__BITS4);	\
-      }									\
-    else								\
-      {									\
-	for (__a = SI_TYPE_SIZE - 8; __a > 0; __a -= 8)			\
-	  if (((__xr >> __a) & 0xff) != 0)				\
-	    break;							\
-      }									\
-									\
-    (count) = SI_TYPE_SIZE - (__clz_tab[__xr >> __a] + __a);		\
-  } while (0)
diff --git a/arch/arm/lib/udivdi3.c b/arch/arm/lib/udivdi3.c
deleted file mode 100644
index e343be4c664..00000000000
--- a/arch/arm/lib/udivdi3.c
+++ /dev/null
@@ -1,222 +0,0 @@
-/* More subroutines needed by GCC output code on some machines.  */
-/* Compile this one with gcc.  */
-/* Copyright (C) 1989, 92-98, 1999 Free Software Foundation, Inc.
-
-This file is part of GNU CC.
-
-GNU CC 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.
-
-GNU CC 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 GNU CC; see the file COPYING.  If not, write to
-the Free Software Foundation, 59 Temple Place - Suite 330,
-Boston, MA 02111-1307, USA.  */
-
-/* As a special exception, if you link this library with other files,
-   some of which are compiled with GCC, to produce an executable,
-   this library does not by itself cause the resulting executable
-   to be covered by the GNU General Public License.
-   This exception does not however invalidate any other reasons why
-   the executable file might be covered by the GNU General Public License.
- */
-/* support functions required by the kernel. based on code from gcc-2.95.3 */
-/* I Molton     29/07/01 */
-
-#include "gcclib.h"
-#include "longlong.h"
-
-static const u8 __clz_tab[] = {
-	0, 1, 2, 2, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 5, 5, 5, 5,
-	    5, 5, 5, 5, 5, 5, 5, 5,
-	6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
-	    6, 6, 6, 6, 6, 6, 6, 6,
-	7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
-	    7, 7, 7, 7, 7, 7, 7, 7,
-	7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
-	    7, 7, 7, 7, 7, 7, 7, 7,
-	8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
-	    8, 8, 8, 8, 8, 8, 8, 8,
-	8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
-	    8, 8, 8, 8, 8, 8, 8, 8,
-	8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
-	    8, 8, 8, 8, 8, 8, 8, 8,
-	8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
-	    8, 8, 8, 8, 8, 8, 8, 8,
-};
-
-u64 __udivmoddi4(u64 n, u64 d, u64 * rp)
-{
-	DIunion ww;
-	DIunion nn, dd;
-	DIunion rr;
-	u32 d0, d1, n0, n1, n2;
-	u32 q0, q1;
-	u32 b, bm;
-
-	nn.ll = n;
-	dd.ll = d;
-
-	d0 = dd.s.low;
-	d1 = dd.s.high;
-	n0 = nn.s.low;
-	n1 = nn.s.high;
-
-	if (d1 == 0) {
-		if (d0 > n1) {
-			/* 0q = nn / 0D */
-
-			count_leading_zeros(bm, d0);
-
-			if (bm != 0) {
-				/* Normalize, i.e. make the most significant bit of the
-				   denominator set.  */
-
-				d0 = d0 << bm;
-				n1 = (n1 << bm) | (n0 >> (SI_TYPE_SIZE - bm));
-				n0 = n0 << bm;
-			}
-
-			udiv_qrnnd(q0, n0, n1, n0, d0);
-			q1 = 0;
-
-			/* Remainder in n0 >> bm.  */
-		} else {
-			/* qq = NN / 0d */
-
-			if (d0 == 0)
-				d0 = 1 / d0;	/* Divide intentionally by zero.  */
-
-			count_leading_zeros(bm, d0);
-
-			if (bm == 0) {
-				/* From (n1 >= d0) /\ (the most significant bit of d0 is set),
-				   conclude (the most significant bit of n1 is set) /\ (the
-				   leading quotient digit q1 = 1).
-
-				   This special case is necessary, not an optimization.
-				   (Shifts counts of SI_TYPE_SIZE are undefined.)  */
-
-				n1 -= d0;
-				q1 = 1;
-			} else {
-				/* Normalize.  */
-
-				b = SI_TYPE_SIZE - bm;
-
-				d0 = d0 << bm;
-				n2 = n1 >> b;
-				n1 = (n1 << bm) | (n0 >> b);
-				n0 = n0 << bm;
-
-				udiv_qrnnd(q1, n1, n2, n1, d0);
-			}
-
-			/* n1 != d0...  */
-
-			udiv_qrnnd(q0, n0, n1, n0, d0);
-
-			/* Remainder in n0 >> bm.  */
-		}
-
-		if (rp != 0) {
-			rr.s.low = n0 >> bm;
-			rr.s.high = 0;
-			*rp = rr.ll;
-		}
-	} else {
-		if (d1 > n1) {
-			/* 00 = nn / DD */
-
-			q0 = 0;
-			q1 = 0;
-
-			/* Remainder in n1n0.  */
-			if (rp != 0) {
-				rr.s.low = n0;
-				rr.s.high = n1;
-				*rp = rr.ll;
-			}
-		} else {
-			/* 0q = NN / dd */
-
-			count_leading_zeros(bm, d1);
-			if (bm == 0) {
-				/* From (n1 >= d1) /\ (the most significant bit of d1 is set),
-				   conclude (the most significant bit of n1 is set) /\ (the
-				   quotient digit q0 = 0 or 1).
-
-				   This special case is necessary, not an optimization.  */
-
-				/* The condition on the next line takes advantage of that
-				   n1 >= d1 (true due to program flow).  */
-				if (n1 > d1 || n0 >= d0) {
-					q0 = 1;
-					sub_ddmmss(n1, n0, n1, n0, d1, d0);
-				} else
-					q0 = 0;
-
-				q1 = 0;
-
-				if (rp != 0) {
-					rr.s.low = n0;
-					rr.s.high = n1;
-					*rp = rr.ll;
-				}
-			} else {
-				u32 m1, m0;
-				/* Normalize.  */
-
-				b = SI_TYPE_SIZE - bm;
-
-				d1 = (d1 << bm) | (d0 >> b);
-				d0 = d0 << bm;
-				n2 = n1 >> b;
-				n1 = (n1 << bm) | (n0 >> b);
-				n0 = n0 << bm;
-
-				udiv_qrnnd(q0, n1, n2, n1, d1);
-				umul_ppmm(m1, m0, q0, d0);
-
-				if (m1 > n1 || (m1 == n1 && m0 > n0)) {
-					q0--;
-					sub_ddmmss(m1, m0, m1, m0, d1, d0);
-				}
-
-				q1 = 0;
-
-				/* Remainder in (n1n0 - m1m0) >> bm.  */
-				if (rp != 0) {
-					sub_ddmmss(n1, n0, n1, n0, m1, m0);
-					rr.s.low = (n1 << b) | (n0 >> bm);
-					rr.s.high = n1 >> bm;
-					*rp = rr.ll;
-				}
-			}
-		}
-	}
-
-	ww.s.low = q0;
-	ww.s.high = q1;
-	return ww.ll;
-}
-
-u64 __udivdi3(u64 n, u64 d)
-{
-	return __udivmoddi4(n, d, (u64 *) 0);
-}
-
-u64 __umoddi3(u64 u, u64 v)
-{
-	u64 w;
-
-	(void)__udivmoddi4(u, v, &w);
-
-	return w;
-}
diff --git a/drivers/net/arm/etherh.c b/drivers/net/arm/etherh.c
index 2e28c201dcc..942a2819576 100644
--- a/drivers/net/arm/etherh.c
+++ b/drivers/net/arm/etherh.c
@@ -68,7 +68,6 @@ struct etherh_priv {
 	void __iomem	*dma_base;
 	unsigned int	id;
 	void __iomem	*ctrl_port;
-	void __iomem	*base;
 	unsigned char	ctrl;
 	u32		supported;
 };
@@ -178,7 +177,7 @@ etherh_setif(struct net_device *dev)
 	switch (etherh_priv(dev)->id) {
 	case PROD_I3_ETHERLAN600:
 	case PROD_I3_ETHERLAN600A:
-		addr = etherh_priv(dev)->base + EN0_RCNTHI;
+		addr = (void *)dev->base_addr + EN0_RCNTHI;
 
 		switch (dev->if_port) {
 		case IF_PORT_10BASE2:
@@ -219,7 +218,7 @@ etherh_getifstat(struct net_device *dev)
 	switch (etherh_priv(dev)->id) {
 	case PROD_I3_ETHERLAN600:
 	case PROD_I3_ETHERLAN600A:
-		addr = etherh_priv(dev)->base + EN0_RCNTHI;
+		addr = (void *)dev->base_addr + EN0_RCNTHI;
 		switch (dev->if_port) {
 		case IF_PORT_10BASE2:
 			stat = 1;
@@ -282,7 +281,7 @@ static void
 etherh_reset(struct net_device *dev)
 {
 	struct ei_device *ei_local = netdev_priv(dev);
-	void __iomem *addr = etherh_priv(dev)->base;
+	void __iomem *addr = (void *)dev->base_addr;
 
 	writeb(E8390_NODMA+E8390_PAGE0+E8390_STOP, addr);
 
@@ -328,7 +327,7 @@ etherh_block_output (struct net_device *dev, int count, const unsigned char *buf
 
 	ei_local->dmaing = 1;
 
-	addr = etherh_priv(dev)->base;
+	addr = (void *)dev->base_addr;
 	dma_base = etherh_priv(dev)->dma_base;
 
 	count = (count + 1) & ~1;
@@ -388,7 +387,7 @@ etherh_block_input (struct net_device *dev, int count, struct sk_buff *skb, int
 
 	ei_local->dmaing = 1;
 
-	addr = etherh_priv(dev)->base;
+	addr = (void *)dev->base_addr;
 	dma_base = etherh_priv(dev)->dma_base;
 
 	buf = skb->data;
@@ -428,7 +427,7 @@ etherh_get_header (struct net_device *dev, struct e8390_pkt_hdr *hdr, int ring_p
 
 	ei_local->dmaing = 1;
 
-	addr = etherh_priv(dev)->base;
+	addr = (void *)dev->base_addr;
 	dma_base = etherh_priv(dev)->dma_base;
 
 	writeb (E8390_NODMA | E8390_PAGE0 | E8390_START, addr + E8390_CMD);
@@ -697,8 +696,7 @@ etherh_probe(struct expansion_card *ec, const struct ecard_id *id)
 		eh->ctrl_port = eh->ioc_fast;
 	}
 
-	eh->base = eh->memc + data->ns8390_offset;
-	dev->base_addr = (unsigned long)eh->base;
+	dev->base_addr = (unsigned long)eh->memc + data->ns8390_offset;
 	eh->dma_base = eh->memc + data->dataport_offset;
 	eh->ctrl_port += data->ctrlport_offset;
 
-- 
cgit v1.2.3-70-g09d2


From 20f720ccce8864583365d844d48d3fe8d0d171ad Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 29 Jun 2005 18:26:19 +0100
Subject: [PATCH] ARM: Remove nmi_tick from integrator platforms

The nmi watchdog code hasn't been merged for this platform yet.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-integrator/core.c | 1 -
 1 file changed, 1 deletion(-)

(limited to 'arch')

diff --git a/arch/arm/mach-integrator/core.c b/arch/arm/mach-integrator/core.c
index 797c7fddaa0..dacbf504dae 100644
--- a/arch/arm/mach-integrator/core.c
+++ b/arch/arm/mach-integrator/core.c
@@ -215,7 +215,6 @@ integrator_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 	 * primary CPU
 	 */
 	if (hard_smp_processor_id() == 0) {
-		nmi_tick();
 		timer_tick(regs);
 #ifdef CONFIG_SMP
 		smp_send_timer();
-- 
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 'arch')

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 b3402cf50efead37dd9553b90fbf1486e09fb78e Mon Sep 17 00:00:00 2001
From: Tony Lindgren <tony@atomide.com>
Date: Wed, 29 Jun 2005 19:59:48 +0100
Subject: [PATCH] ARM: 2771/1: Dynamic Tick support for OMAP, take 4

Patch from Tony Lindgren

This patch adds support for Dynamic Tick Timer for OMAP.
This patch is an updated version of ARM patch 2642/1 to
make it work with the already integrated generic ARM
dyntick support.

Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-omap/pm.c   | 16 ++++++++++------
 arch/arm/mach-omap/time.c | 44 ++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 52 insertions(+), 8 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/mach-omap/pm.c b/arch/arm/mach-omap/pm.c
index 00fac155df2..6b03ccdc1e9 100644
--- a/arch/arm/mach-omap/pm.c
+++ b/arch/arm/mach-omap/pm.c
@@ -41,7 +41,9 @@
 #include <linux/pm.h>
 
 #include <asm/io.h>
+#include <asm/mach/time.h>
 #include <asm/mach-types.h>
+
 #include <asm/arch/omap16xx.h>
 #include <asm/arch/pm.h>
 #include <asm/arch/mux.h>
@@ -80,13 +82,13 @@ void omap_pm_idle(void)
 		return;
 	}
 	mask32 = omap_readl(ARM_SYSST);
-	local_fiq_enable();
-	local_irq_enable();
 
-#if defined(CONFIG_OMAP_32K_TIMER) && defined(CONFIG_NO_IDLE_HZ)
-	/* Override timer to use VST for the next cycle */
-	omap_32k_timer_next_vst_interrupt();
-#endif
+	/*
+	 * Since an interrupt may set up a timer, we don't want to
+	 * reprogram the hardware timer with interrupts enabled.
+	 * Re-enable interrupts only after returning from idle.
+	 */
+	timer_dyn_reprogram();
 
 	if ((mask32 & DSP_IDLE) == 0) {
 		__asm__ volatile ("mcr	p15, 0, r0, c7, c0, 4");
@@ -102,6 +104,8 @@ void omap_pm_idle(void)
 
 		func_ptr();
 	}
+	local_fiq_enable();
+	local_irq_enable();
 }
 
 /*
diff --git a/arch/arm/mach-omap/time.c b/arch/arm/mach-omap/time.c
index 589e8b2740d..dd34e9f4c41 100644
--- a/arch/arm/mach-omap/time.c
+++ b/arch/arm/mach-omap/time.c
@@ -4,7 +4,7 @@
  * OMAP Timers
  *
  * Copyright (C) 2004 Nokia Corporation
- * Partial timer rewrite and additional VST timer support by
+ * Partial timer rewrite and additional dynamic tick timer support by
  * Tony Lindgen <tony@atomide.com> and
  * Tuukka Tikkanen <tuukka.tikkanen@elektrobit.com>
  *
@@ -261,7 +261,6 @@ unsigned long long sched_clock(void)
  * so with HZ = 100, TVR = 327.68.
  */
 #define OMAP_32K_TIMER_TICK_PERIOD	((32768 / HZ) - 1)
-#define MAX_SKIP_JIFFIES		25
 #define TIMER_32K_SYNCHRONIZED		0xfffbc410
 
 #define JIFFIES_TO_HW_TICKS(nr_jiffies, clock_rate)			\
@@ -347,6 +346,42 @@ static irqreturn_t omap_32k_timer_interrupt(int irq, void *dev_id,
 	return IRQ_HANDLED;
 }
 
+#ifdef CONFIG_NO_IDLE_HZ
+/*
+ * Programs the next timer interrupt needed. Called when dynamic tick is
+ * enabled, and to reprogram the ticks to skip from pm_idle. Note that
+ * we can keep the timer continuous, and don't need to set it to run in
+ * one-shot mode. This is because the timer will get reprogrammed again
+ * after next interrupt.
+ */
+void omap_32k_timer_reprogram(unsigned long next_tick)
+{
+	omap_32k_timer_start(JIFFIES_TO_HW_TICKS(next_tick, 32768) + 1);
+}
+
+static struct irqaction omap_32k_timer_irq;
+extern struct timer_update_handler timer_update;
+
+static int omap_32k_timer_enable_dyn_tick(void)
+{
+	/* No need to reprogram timer, just use the next interrupt */
+	return 0;
+}
+
+static int omap_32k_timer_disable_dyn_tick(void)
+{
+	omap_32k_timer_start(OMAP_32K_TIMER_TICK_PERIOD);
+	return 0;
+}
+
+static struct dyn_tick_timer omap_dyn_tick_timer = {
+	.enable		= omap_32k_timer_enable_dyn_tick,
+	.disable	= omap_32k_timer_disable_dyn_tick,
+	.reprogram	= omap_32k_timer_reprogram,
+	.handler	= omap_32k_timer_interrupt,
+};
+#endif	/* CONFIG_NO_IDLE_HZ */
+
 static struct irqaction omap_32k_timer_irq = {
 	.name		= "32KHz timer",
 	.flags		= SA_INTERRUPT | SA_TIMER,
@@ -355,6 +390,11 @@ static struct irqaction omap_32k_timer_irq = {
 
 static __init void omap_init_32k_timer(void)
 {
+
+#ifdef CONFIG_NO_IDLE_HZ
+	omap_timer.dyn_tick = &omap_dyn_tick_timer;
+#endif
+
 	setup_irq(INT_OS_TIMER, &omap_32k_timer_irq);
 	omap_timer.offset  = omap_32k_timer_gettimeoffset;
 	omap_32k_last_tick = omap_32k_sync_timer_read();
-- 
cgit v1.2.3-70-g09d2


From 438a76167959061e371025f727fabec2ad9e70a7 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 29 Jun 2005 23:01:02 +0100
Subject: [PATCH] ARM: Fix VFP to use do_div()

VFP used __divdi3 64-bit division needlessly.  Convert it to use
our 64-bit by 32-bit division instead.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/vfp/vfp.h       | 15 +++++++++++++--
 arch/arm/vfp/vfpdouble.c |  2 ++
 arch/arm/vfp/vfpsingle.c | 14 ++++++++++++--
 3 files changed, 27 insertions(+), 4 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/vfp/vfp.h b/arch/arm/vfp/vfp.h
index 55a02bc994a..4b97950984e 100644
--- a/arch/arm/vfp/vfp.h
+++ b/arch/arm/vfp/vfp.h
@@ -117,7 +117,13 @@ static inline u64 vfp_estimate_div128to64(u64 nh, u64 nl, u64 m)
 	if (nh >= m)
 		return ~0ULL;
 	mh = m >> 32;
-	z = (mh << 32 <= nh) ? 0xffffffff00000000ULL : (nh / mh) << 32;
+	if (mh << 32 <= nh) {
+		z = 0xffffffff00000000ULL;
+	} else {
+		z = nh;
+		do_div(z, mh);
+		z <<= 32;
+	}
 	mul64to128(&termh, &terml, m, z);
 	sub128(&remh, &reml, nh, nl, termh, terml);
 	ml = m << 32;
@@ -126,7 +132,12 @@ static inline u64 vfp_estimate_div128to64(u64 nh, u64 nl, u64 m)
 		add128(&remh, &reml, remh, reml, mh, ml);
 	}
 	remh = (remh << 32) | (reml >> 32);
-	z |= (mh << 32 <= remh) ? 0xffffffff : remh / mh;
+	if (mh << 32 <= remh) {
+		z |= 0xffffffff;
+	} else {
+		do_div(remh, mh);
+		z |= remh;
+	}
 	return z;
 }
 
diff --git a/arch/arm/vfp/vfpdouble.c b/arch/arm/vfp/vfpdouble.c
index fa3053e84db..b801cd66b6e 100644
--- a/arch/arm/vfp/vfpdouble.c
+++ b/arch/arm/vfp/vfpdouble.c
@@ -32,6 +32,8 @@
  */
 #include <linux/kernel.h>
 #include <linux/bitops.h>
+
+#include <asm/div64.h>
 #include <asm/ptrace.h>
 #include <asm/vfp.h>
 
diff --git a/arch/arm/vfp/vfpsingle.c b/arch/arm/vfp/vfpsingle.c
index 6849fe35cb2..14dd696ddeb 100644
--- a/arch/arm/vfp/vfpsingle.c
+++ b/arch/arm/vfp/vfpsingle.c
@@ -32,6 +32,8 @@
  */
 #include <linux/kernel.h>
 #include <linux/bitops.h>
+
+#include <asm/div64.h>
 #include <asm/ptrace.h>
 #include <asm/vfp.h>
 
@@ -303,7 +305,11 @@ u32 vfp_estimate_sqrt_significand(u32 exponent, u32 significand)
 		if (z <= a)
 			return (s32)a >> 1;
 	}
-	return (u32)(((u64)a << 31) / z) + (z >> 1);
+	{
+		u64 v = (u64)a << 31;
+		do_div(v, z);
+		return v + (z >> 1);
+	}
 }
 
 static u32 vfp_single_fsqrt(int sd, int unused, s32 m, u32 fpscr)
@@ -1107,7 +1113,11 @@ static u32 vfp_single_fdiv(int sd, int sn, s32 m, u32 fpscr)
 		vsn.significand >>= 1;
 		vsd.exponent++;
 	}
-	vsd.significand = ((u64)vsn.significand << 32) / vsm.significand;
+	{
+		u64 significand = (u64)vsn.significand << 32;
+		do_div(significand, vsm.significand);
+		vsd.significand = significand;
+	}
 	if ((vsd.significand & 0x3f) == 0)
 		vsd.significand |= ((u64)vsm.significand * vsd.significand != (u64)vsn.significand << 32);
 
-- 
cgit v1.2.3-70-g09d2


From da41119af78864d27ccbf505949df788d5e8aaf5 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 29 Jun 2005 23:02:02 +0100
Subject: [PATCH] ARM: Don't force SIGFPE

We were forcing SIGFPE on to a user program for no good reason.
Use send_sig_info() instead.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/vfp/vfpmodule.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/arm/vfp/vfpmodule.c b/arch/arm/vfp/vfpmodule.c
index 3aeedd2afc7..22f3da4e082 100644
--- a/arch/arm/vfp/vfpmodule.c
+++ b/arch/arm/vfp/vfpmodule.c
@@ -89,7 +89,7 @@ void vfp_raise_sigfpe(unsigned int sicode, struct pt_regs *regs)
 	current->thread.error_code = 0;
 	current->thread.trap_no = 6;
 
-	force_sig_info(SIGFPE, &info, current);
+	send_sig_info(SIGFPE, &info, current);
 }
 
 static void vfp_panic(char *reason)
-- 
cgit v1.2.3-70-g09d2


From 06697eea4ce18870fbce82dffbb375e216145f47 Mon Sep 17 00:00:00 2001
From: Tony Luck <tony.luck@intel.com>
Date: Wed, 29 Jun 2005 15:45:41 -0700
Subject: [IA64] Update tiger_defconfig

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/configs/tiger_defconfig | 39 +++++++++++++++++++++++++++++++--------
 1 file changed, 31 insertions(+), 8 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/configs/tiger_defconfig b/arch/ia64/configs/tiger_defconfig
index 47f45341ac6..73454eee26f 100644
--- a/arch/ia64/configs/tiger_defconfig
+++ b/arch/ia64/configs/tiger_defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.12-20050621
-# Tue Jun 21 14:03:24 2005
+# Linux kernel version: 2.6.13-rc1-20050629
+# Wed Jun 29 15:28:12 2005
 #
 
 #
@@ -80,18 +80,29 @@ CONFIG_MCKINLEY=y
 # CONFIG_IA64_PAGE_SIZE_8KB is not set
 CONFIG_IA64_PAGE_SIZE_16KB=y
 # CONFIG_IA64_PAGE_SIZE_64KB is not set
+# CONFIG_HZ_100 is not set
+CONFIG_HZ_250=y
+# CONFIG_HZ_1000 is not set
+CONFIG_HZ=250
 CONFIG_IA64_L1_CACHE_SHIFT=7
 # CONFIG_NUMA is not set
 CONFIG_VIRTUAL_MEM_MAP=y
 CONFIG_HOLES_IN_ZONE=y
 CONFIG_IA64_CYCLONE=y
 CONFIG_IOSAPIC=y
+# CONFIG_IA64_SGI_SN_XP is not set
 CONFIG_FORCE_MAX_ZONEORDER=18
 CONFIG_SMP=y
 CONFIG_NR_CPUS=4
 CONFIG_HOTPLUG_CPU=y
 # CONFIG_SCHED_SMT is not set
 # CONFIG_PREEMPT is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
 CONFIG_HAVE_DEC_LOCK=y
 CONFIG_IA32_SUPPORT=y
 CONFIG_COMPAT=y
@@ -257,6 +268,7 @@ CONFIG_BLK_DEV_CMD64X=y
 # CONFIG_BLK_DEV_HPT366 is not set
 # CONFIG_BLK_DEV_SC1200 is not set
 CONFIG_BLK_DEV_PIIX=y
+# CONFIG_BLK_DEV_IT821X is not set
 # CONFIG_BLK_DEV_NS87415 is not set
 # CONFIG_BLK_DEV_PDC202XX_OLD is not set
 # CONFIG_BLK_DEV_PDC202XX_NEW is not set
@@ -395,6 +407,7 @@ CONFIG_UNIX=y
 CONFIG_INET=y
 CONFIG_IP_MULTICAST=y
 # CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
 # CONFIG_IP_PNP is not set
 # CONFIG_NET_IPIP is not set
 # CONFIG_NET_IPGRE is not set
@@ -407,6 +420,8 @@ CONFIG_SYN_COOKIES=y
 # CONFIG_INET_TUNNEL is not set
 CONFIG_IP_TCPDIAG=y
 # CONFIG_IP_TCPDIAG_IPV6 is not set
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
 # CONFIG_IPV6 is not set
 # CONFIG_NETFILTER is not set
 
@@ -598,9 +613,7 @@ CONFIG_GAMEPORT=m
 # CONFIG_GAMEPORT_NS558 is not set
 # CONFIG_GAMEPORT_L4 is not set
 # CONFIG_GAMEPORT_EMU10K1 is not set
-# CONFIG_GAMEPORT_VORTEX is not set
 # CONFIG_GAMEPORT_FM801 is not set
-# CONFIG_GAMEPORT_CS461X is not set
 
 #
 # Character devices
@@ -629,7 +642,6 @@ CONFIG_SERIAL_8250_NR_UARTS=6
 CONFIG_SERIAL_8250_EXTENDED=y
 CONFIG_SERIAL_8250_SHARE_IRQ=y
 # CONFIG_SERIAL_8250_DETECT_IRQ is not set
-# CONFIG_SERIAL_8250_MULTIPORT is not set
 # CONFIG_SERIAL_8250_RSA is not set
 
 #
@@ -743,6 +755,7 @@ CONFIG_USB_DEVICEFS=y
 CONFIG_USB_EHCI_HCD=m
 # CONFIG_USB_EHCI_SPLIT_ISO is not set
 # CONFIG_USB_EHCI_ROOT_HUB_TT is not set
+# CONFIG_USB_ISP116X_HCD is not set
 CONFIG_USB_OHCI_HCD=m
 # CONFIG_USB_OHCI_BIG_ENDIAN is not set
 CONFIG_USB_OHCI_LITTLE_ENDIAN=y
@@ -779,9 +792,11 @@ CONFIG_USB_HIDINPUT=y
 # CONFIG_USB_HIDDEV is not set
 # CONFIG_USB_AIPTEK is not set
 # CONFIG_USB_WACOM is not set
+# CONFIG_USB_ACECAD is not set
 # CONFIG_USB_KBTAB is not set
 # CONFIG_USB_POWERMATE is not set
 # CONFIG_USB_MTOUCH is not set
+# CONFIG_USB_ITMTOUCH is not set
 # CONFIG_USB_EGALAX is not set
 # CONFIG_USB_XPAD is not set
 # CONFIG_USB_ATI_REMOTE is not set
@@ -838,7 +853,7 @@ CONFIG_USB_HIDINPUT=y
 # CONFIG_USB_TEST is not set
 
 #
-# USB ATM/DSL drivers
+# USB DSL modem support
 #
 
 #
@@ -856,6 +871,10 @@ CONFIG_USB_HIDINPUT=y
 #
 # CONFIG_INFINIBAND is not set
 
+#
+# SN Devices
+#
+
 #
 # File systems
 #
@@ -863,6 +882,7 @@ CONFIG_EXT2_FS=y
 CONFIG_EXT2_FS_XATTR=y
 CONFIG_EXT2_FS_POSIX_ACL=y
 CONFIG_EXT2_FS_SECURITY=y
+# CONFIG_EXT2_FS_XIP is not set
 CONFIG_EXT3_FS=y
 CONFIG_EXT3_FS_XATTR=y
 CONFIG_EXT3_FS_POSIX_ACL=y
@@ -922,7 +942,6 @@ CONFIG_NTFS_FS=m
 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=y
@@ -953,15 +972,18 @@ CONFIG_RAMFS=y
 #
 CONFIG_NFS_FS=m
 CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
 CONFIG_NFS_V4=y
 CONFIG_NFS_DIRECTIO=y
 CONFIG_NFSD=m
 CONFIG_NFSD_V3=y
+# CONFIG_NFSD_V3_ACL is not set
 CONFIG_NFSD_V4=y
 CONFIG_NFSD_TCP=y
 CONFIG_LOCKD=m
 CONFIG_LOCKD_V4=y
 CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
 CONFIG_SUNRPC=m
 CONFIG_SUNRPC_GSS=m
 CONFIG_RPCSEC_GSS_KRB5=m
@@ -1069,6 +1091,7 @@ CONFIG_LOG_BUF_SHIFT=20
 # CONFIG_DEBUG_KOBJECT is not set
 # CONFIG_DEBUG_INFO is not set
 # CONFIG_DEBUG_FS is not set
+# CONFIG_KPROBES is not set
 CONFIG_IA64_GRANULE_16MB=y
 # CONFIG_IA64_GRANULE_64MB is not set
 # CONFIG_IA64_PRINT_HAZARDS is not set
@@ -1090,7 +1113,7 @@ CONFIG_CRYPTO=y
 # CONFIG_CRYPTO_HMAC is not set
 # CONFIG_CRYPTO_NULL is not set
 # CONFIG_CRYPTO_MD4 is not set
-CONFIG_CRYPTO_MD5=m
+CONFIG_CRYPTO_MD5=y
 # CONFIG_CRYPTO_SHA1 is not set
 # CONFIG_CRYPTO_SHA256 is not set
 # CONFIG_CRYPTO_SHA512 is not set
-- 
cgit v1.2.3-70-g09d2


From 0647d8cfd40f7974335884d73deef83feeffc2a0 Mon Sep 17 00:00:00 2001
From: Tony Luck <tony.luck@intel.com>
Date: Wed, 29 Jun 2005 16:01:40 -0700
Subject: [IA64] Update zx1_defconfig

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/configs/zx1_defconfig | 166 +++++++++++++++++++++++++++++-----------
 1 file changed, 122 insertions(+), 44 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/configs/zx1_defconfig b/arch/ia64/configs/zx1_defconfig
index 21d6f9bab5e..b7755e4436d 100644
--- a/arch/ia64/configs/zx1_defconfig
+++ b/arch/ia64/configs/zx1_defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.10
-# Wed Dec 29 09:05:48 2004
+# Linux kernel version: 2.6.13-rc1-20050629
+# Wed Jun 29 15:31:11 2005
 #
 
 #
@@ -12,6 +12,7 @@ CONFIG_EXPERIMENTAL=y
 CONFIG_BROKEN=y
 CONFIG_BROKEN_ON_SMP=y
 CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -24,23 +25,26 @@ CONFIG_BSD_PROCESS_ACCT=y
 # CONFIG_BSD_PROCESS_ACCT_V3 is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
-CONFIG_LOG_BUF_SHIFT=17
 CONFIG_HOTPLUG=y
 CONFIG_KOBJECT_UEVENT=y
 # CONFIG_IKCONFIG is not set
+# CONFIG_CPUSETS is not set
 # CONFIG_EMBEDDED is not set
 CONFIG_KALLSYMS=y
 # CONFIG_KALLSYMS_ALL is not set
 # CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_BASE_FULL=y
 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
+CONFIG_BASE_SMALL=0
 
 #
 # Loadable module support
@@ -59,12 +63,15 @@ CONFIG_IA64=y
 CONFIG_64BIT=y
 CONFIG_MMU=y
 CONFIG_RWSEM_XCHGADD_ALGORITHM=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
 CONFIG_TIME_INTERPOLATION=y
 CONFIG_EFI=y
 CONFIG_GENERIC_IOMAP=y
+CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y
 # CONFIG_IA64_GENERIC is not set
 # CONFIG_IA64_DIG is not set
 CONFIG_IA64_HP_ZX1=y
+# CONFIG_IA64_HP_ZX1_SWIOTLB is not set
 # CONFIG_IA64_SGI_SN2 is not set
 # CONFIG_IA64_HP_SIM is not set
 # CONFIG_ITANIUM is not set
@@ -73,22 +80,36 @@ CONFIG_MCKINLEY=y
 # CONFIG_IA64_PAGE_SIZE_8KB is not set
 CONFIG_IA64_PAGE_SIZE_16KB=y
 # CONFIG_IA64_PAGE_SIZE_64KB is not set
+# CONFIG_HZ_100 is not set
+CONFIG_HZ_250=y
+# CONFIG_HZ_1000 is not set
+CONFIG_HZ=250
 CONFIG_IA64_L1_CACHE_SHIFT=7
 # CONFIG_NUMA is not set
 CONFIG_VIRTUAL_MEM_MAP=y
+CONFIG_HOLES_IN_ZONE=y
 # CONFIG_IA64_CYCLONE is not set
 CONFIG_IOSAPIC=y
+# CONFIG_IA64_SGI_SN_XP is not set
 CONFIG_FORCE_MAX_ZONEORDER=18
 CONFIG_SMP=y
 CONFIG_NR_CPUS=16
 # CONFIG_HOTPLUG_CPU is not set
+# CONFIG_SCHED_SMT is not set
 # CONFIG_PREEMPT is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
 CONFIG_HAVE_DEC_LOCK=y
 CONFIG_IA32_SUPPORT=y
 CONFIG_COMPAT=y
 CONFIG_IA64_MCA_RECOVERY=y
 CONFIG_PERFMON=y
 CONFIG_IA64_PALINFO=y
+CONFIG_ACPI_DEALLOCATE_IRQ=y
 
 #
 # Firmware Drivers
@@ -120,6 +141,7 @@ CONFIG_ACPI_BUS=y
 CONFIG_ACPI_POWER=y
 CONFIG_ACPI_PCI=y
 CONFIG_ACPI_SYSTEM=y
+# CONFIG_ACPI_CONTAINER is not set
 
 #
 # Bus options (PCI, PCMCIA)
@@ -129,6 +151,7 @@ CONFIG_PCI_DOMAINS=y
 # CONFIG_PCI_MSI is not set
 CONFIG_PCI_LEGACY_PROC=y
 CONFIG_PCI_NAMES=y
+# CONFIG_PCI_DEBUG is not set
 
 #
 # PCI Hotplug Support
@@ -138,7 +161,6 @@ CONFIG_HOTPLUG_PCI=y
 CONFIG_HOTPLUG_PCI_ACPI=y
 # CONFIG_HOTPLUG_PCI_ACPI_IBM is not set
 # CONFIG_HOTPLUG_PCI_CPCI is not set
-# CONFIG_HOTPLUG_PCI_PCIE is not set
 # CONFIG_HOTPLUG_PCI_SHPC is not set
 
 #
@@ -146,10 +168,6 @@ CONFIG_HOTPLUG_PCI_ACPI=y
 #
 # CONFIG_PCCARD is not set
 
-#
-# PC-card bridges
-#
-
 #
 # Device Drivers
 #
@@ -184,6 +202,7 @@ CONFIG_PREVENT_FIRMWARE_BUILD=y
 # CONFIG_BLK_CPQ_CISS_DA is not set
 # CONFIG_BLK_DEV_DAC960 is not set
 # CONFIG_BLK_DEV_UMEM 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
@@ -203,6 +222,7 @@ 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
@@ -246,6 +266,7 @@ CONFIG_BLK_DEV_CMD64X=y
 # CONFIG_BLK_DEV_HPT366 is not set
 # CONFIG_BLK_DEV_SC1200 is not set
 # CONFIG_BLK_DEV_PIIX is not set
+# CONFIG_BLK_DEV_IT821X is not set
 # CONFIG_BLK_DEV_NS87415 is not set
 # CONFIG_BLK_DEV_PDC202XX_OLD is not set
 # CONFIG_BLK_DEV_PDC202XX_NEW is not set
@@ -275,6 +296,7 @@ CONFIG_CHR_DEV_OSST=y
 CONFIG_BLK_DEV_SR=y
 CONFIG_BLK_DEV_SR_VENDOR=y
 CONFIG_CHR_DEV_SG=y
+# CONFIG_CHR_DEV_SCH is not set
 
 #
 # Some SCSI devices (e.g. CD jukebox) support multiple LUNs
@@ -288,6 +310,7 @@ CONFIG_SCSI_LOGGING=y
 #
 CONFIG_SCSI_SPI_ATTRS=y
 # CONFIG_SCSI_FC_ATTRS is not set
+# CONFIG_SCSI_ISCSI_ATTRS is not set
 
 #
 # SCSI low-level drivers
@@ -303,13 +326,10 @@ CONFIG_SCSI_SPI_ATTRS=y
 # CONFIG_MEGARAID_NEWGEN is not set
 # CONFIG_MEGARAID_LEGACY is not set
 # CONFIG_SCSI_SATA is not set
-# CONFIG_SCSI_BUSLOGIC is not set
 # CONFIG_SCSI_CPQFCTS is not set
 # CONFIG_SCSI_DMX3191D is not set
-# CONFIG_SCSI_EATA is not set
 # CONFIG_SCSI_EATA_PIO is not set
 # CONFIG_SCSI_FUTURE_DOMAIN is not set
-# CONFIG_SCSI_GDTH is not set
 # CONFIG_SCSI_IPS is not set
 # CONFIG_SCSI_INITIO is not set
 # CONFIG_SCSI_INIA100 is not set
@@ -319,8 +339,6 @@ CONFIG_SCSI_SYM53C8XX_DEFAULT_TAGS=16
 CONFIG_SCSI_SYM53C8XX_MAX_TAGS=64
 # CONFIG_SCSI_SYM53C8XX_IOMAPPED is not set
 # CONFIG_SCSI_IPR is not set
-# CONFIG_SCSI_PCI2000 is not set
-# CONFIG_SCSI_PCI2220I is not set
 # CONFIG_SCSI_QLOGIC_ISP is not set
 # CONFIG_SCSI_QLOGIC_FC is not set
 CONFIG_SCSI_QLOGIC_1280=y
@@ -331,7 +349,7 @@ CONFIG_SCSI_QLA2XXX=y
 # CONFIG_SCSI_QLA2300 is not set
 # CONFIG_SCSI_QLA2322 is not set
 # CONFIG_SCSI_QLA6312 is not set
-# CONFIG_SCSI_QLA6322 is not set
+# CONFIG_SCSI_LPFC is not set
 # CONFIG_SCSI_DC395x is not set
 # CONFIG_SCSI_DC390T is not set
 # CONFIG_SCSI_DEBUG is not set
@@ -344,9 +362,9 @@ CONFIG_SCSI_QLA2XXX=y
 #
 # Fusion MPT device support
 #
-CONFIG_FUSION=y
-CONFIG_FUSION_MAX_SGE=40
-# CONFIG_FUSION_CTL is not set
+# CONFIG_FUSION is not set
+# CONFIG_FUSION_SPI is not set
+# CONFIG_FUSION_FC is not set
 
 #
 # IEEE 1394 (FireWire) support
@@ -368,12 +386,12 @@ CONFIG_NET=y
 #
 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=y
 # CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
 # CONFIG_IP_PNP is not set
 # CONFIG_NET_IPIP is not set
 # CONFIG_NET_IPGRE is not set
@@ -386,6 +404,8 @@ CONFIG_IP_MULTICAST=y
 # CONFIG_INET_TUNNEL is not set
 # CONFIG_IP_TCPDIAG is not set
 # CONFIG_IP_TCPDIAG_IPV6 is not set
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
 
 #
 # IP: Virtual Server Configuration
@@ -405,8 +425,6 @@ CONFIG_NETFILTER=y
 CONFIG_IP_NF_ARPTABLES=y
 # CONFIG_IP_NF_ARPFILTER is not set
 # CONFIG_IP_NF_ARP_MANGLE is not set
-# CONFIG_IP_NF_COMPAT_IPCHAINS is not set
-# CONFIG_IP_NF_COMPAT_IPFWADM is not set
 
 #
 # SCTP Configuration (EXPERIMENTAL)
@@ -483,7 +501,6 @@ CONFIG_NET_PCI=y
 # CONFIG_DGRS is not set
 # CONFIG_EEPRO100 is not set
 CONFIG_E100=y
-# CONFIG_E100_NAPI is not set
 # CONFIG_FEALNX is not set
 # CONFIG_NATSEMI is not set
 # CONFIG_NE2K_PCI is not set
@@ -505,9 +522,11 @@ CONFIG_E1000=y
 # CONFIG_HAMACHI is not set
 # CONFIG_YELLOWFIN is not set
 # CONFIG_R8169 is not set
+# CONFIG_SKGE is not set
 # CONFIG_SK98LIN is not set
 # CONFIG_VIA_VELOCITY is not set
 CONFIG_TIGON3=y
+# CONFIG_BNX2 is not set
 
 #
 # Ethernet (10000 Mbit)
@@ -564,18 +583,6 @@ CONFIG_INPUT_JOYDEV=y
 CONFIG_INPUT_EVDEV=y
 # CONFIG_INPUT_EVBUG is not set
 
-#
-# Input I/O drivers
-#
-# CONFIG_GAMEPORT is not set
-CONFIG_SOUND_GAMEPORT=y
-CONFIG_SERIO=y
-# CONFIG_SERIO_I8042 is not set
-# CONFIG_SERIO_SERPORT is not set
-# CONFIG_SERIO_CT82C710 is not set
-# CONFIG_SERIO_PCIPS2 is not set
-# CONFIG_SERIO_RAW is not set
-
 #
 # Input Device Drivers
 #
@@ -585,6 +592,16 @@ CONFIG_SERIO=y
 # 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_PCIPS2 is not set
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+
 #
 # Character devices
 #
@@ -603,7 +620,6 @@ CONFIG_SERIAL_8250_NR_UARTS=8
 CONFIG_SERIAL_8250_EXTENDED=y
 CONFIG_SERIAL_8250_SHARE_IRQ=y
 # CONFIG_SERIAL_8250_DETECT_IRQ is not set
-# CONFIG_SERIAL_8250_MULTIPORT is not set
 # CONFIG_SERIAL_8250_RSA is not set
 
 #
@@ -611,6 +627,7 @@ CONFIG_SERIAL_8250_SHARE_IRQ=y
 #
 CONFIG_SERIAL_CORE=y
 CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
 CONFIG_UNIX98_PTYS=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
@@ -644,6 +661,12 @@ CONFIG_DRM_RADEON=y
 # CONFIG_DRM_SIS is not set
 # CONFIG_RAW_DRIVER is not set
 # CONFIG_HPET is not set
+# CONFIG_HANGCHECK_TIMER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
 
 #
 # I2C support
@@ -668,6 +691,7 @@ CONFIG_I2C_ALGOPCF=y
 # CONFIG_I2C_AMD8111 is not set
 # CONFIG_I2C_I801 is not set
 # CONFIG_I2C_I810 is not set
+# CONFIG_I2C_PIIX4 is not set
 # CONFIG_I2C_ISA is not set
 # CONFIG_I2C_NFORCE2 is not set
 # CONFIG_I2C_PARPORT_LIGHT is not set
@@ -691,10 +715,14 @@ CONFIG_I2C_ALGOPCF=y
 # CONFIG_SENSORS_ADM1025 is not set
 # CONFIG_SENSORS_ADM1026 is not set
 # CONFIG_SENSORS_ADM1031 is not set
+# CONFIG_SENSORS_ADM9240 is not set
 # CONFIG_SENSORS_ASB100 is not set
+# CONFIG_SENSORS_ATXP1 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
@@ -705,21 +733,29 @@ CONFIG_I2C_ALGOPCF=y
 # 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_SIS5595 is not set
 # CONFIG_SENSORS_SMSC47M1 is not set
 # CONFIG_SENSORS_VIA686A is not set
 # CONFIG_SENSORS_W83781D is not set
 # CONFIG_SENSORS_W83L785TS is not set
 # CONFIG_SENSORS_W83627HF is not set
+# CONFIG_SENSORS_W83627EHF is not set
 
 #
 # Other I2C Chip support
 #
+# CONFIG_SENSORS_DS1337 is not set
+# CONFIG_SENSORS_DS1374 is not set
 # CONFIG_SENSORS_EEPROM is not set
 # CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_SENSORS_PCA9539 is not set
 # CONFIG_SENSORS_PCF8591 is not set
 # CONFIG_SENSORS_RTC8564 is not set
+# CONFIG_SENSORS_MAX6875 is not set
 # CONFIG_I2C_DEBUG_CORE is not set
 # CONFIG_I2C_DEBUG_ALGO is not set
 # CONFIG_I2C_DEBUG_BUS is not set
@@ -746,6 +782,7 @@ CONFIG_VIDEO_DEV=y
 #
 # Video Adapters
 #
+# CONFIG_TUNER_MULTI_I2C is not set
 # CONFIG_VIDEO_BT848 is not set
 # CONFIG_VIDEO_CPIA is not set
 # CONFIG_VIDEO_SAA5246A is not set
@@ -778,6 +815,11 @@ CONFIG_VIDEO_DEV=y
 # Graphics support
 #
 CONFIG_FB=y
+CONFIG_FB_CFB_FILLRECT=y
+CONFIG_FB_CFB_COPYAREA=y
+CONFIG_FB_CFB_IMAGEBLIT=y
+CONFIG_FB_SOFT_CURSOR=y
+# CONFIG_FB_MACMODES is not set
 CONFIG_FB_MODE_HELPERS=y
 # CONFIG_FB_TILEBLITTING is not set
 # CONFIG_FB_CIRRUS is not set
@@ -785,6 +827,7 @@ CONFIG_FB_MODE_HELPERS=y
 # CONFIG_FB_CYBER2000 is not set
 # CONFIG_FB_ASILIANT is not set
 # CONFIG_FB_IMSTT is not set
+# CONFIG_FB_NVIDIA is not set
 # CONFIG_FB_RIVA is not set
 # CONFIG_FB_MATROX is not set
 # CONFIG_FB_RADEON_OLD is not set
@@ -801,6 +844,7 @@ CONFIG_FB_RADEON_DEBUG=y
 # CONFIG_FB_VOODOO1 is not set
 # CONFIG_FB_TRIDENT is not set
 # CONFIG_FB_PM3 is not set
+# CONFIG_FB_S1D13XXX is not set
 # CONFIG_FB_VIRTUAL is not set
 
 #
@@ -820,6 +864,7 @@ CONFIG_LOGO=y
 # CONFIG_LOGO_LINUX_MONO is not set
 # CONFIG_LOGO_LINUX_VGA16 is not set
 CONFIG_LOGO_LINUX_CLUT224=y
+# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
 
 #
 # Sound
@@ -869,6 +914,8 @@ CONFIG_SND_AC97_CODEC=y
 # CONFIG_SND_CS46XX is not set
 # CONFIG_SND_CS4281 is not set
 # CONFIG_SND_EMU10K1 is not set
+# CONFIG_SND_EMU10K1X is not set
+# CONFIG_SND_CA0106 is not set
 # CONFIG_SND_KORG1212 is not set
 # CONFIG_SND_MIXART is not set
 # CONFIG_SND_NM256 is not set
@@ -876,6 +923,7 @@ CONFIG_SND_AC97_CODEC=y
 # CONFIG_SND_RME96 is not set
 # CONFIG_SND_RME9652 is not set
 # CONFIG_SND_HDSP is not set
+# CONFIG_SND_HDSPM is not set
 # CONFIG_SND_TRIDENT is not set
 # CONFIG_SND_YMFPCI is not set
 # CONFIG_SND_ALS4000 is not set
@@ -893,13 +941,14 @@ CONFIG_SND_FM801_TEA575X=y
 # CONFIG_SND_INTEL8X0M is not set
 # CONFIG_SND_SONICVIBES is not set
 # CONFIG_SND_VIA82XX is not set
+# CONFIG_SND_VIA82XX_MODEM is not set
 # CONFIG_SND_VX222 is not set
+# CONFIG_SND_HDA_INTEL is not set
 
 #
 # USB devices
 #
 # CONFIG_SND_USB_AUDIO is not set
-# CONFIG_SND_USB_USX2Y is not set
 
 #
 # Open Sound System
@@ -909,6 +958,8 @@ CONFIG_SND_FM801_TEA575X=y
 #
 # USB support
 #
+CONFIG_USB_ARCH_HAS_HCD=y
+CONFIG_USB_ARCH_HAS_OHCI=y
 CONFIG_USB=y
 # CONFIG_USB_DEBUG is not set
 
@@ -920,8 +971,6 @@ CONFIG_USB_BANDWIDTH=y
 # CONFIG_USB_DYNAMIC_MINORS is not set
 # CONFIG_USB_SUSPEND is not set
 # CONFIG_USB_OTG is not set
-CONFIG_USB_ARCH_HAS_HCD=y
-CONFIG_USB_ARCH_HAS_OHCI=y
 
 #
 # USB Host Controller Drivers
@@ -929,7 +978,10 @@ CONFIG_USB_ARCH_HAS_OHCI=y
 CONFIG_USB_EHCI_HCD=y
 # CONFIG_USB_EHCI_SPLIT_ISO is not set
 # CONFIG_USB_EHCI_ROOT_HUB_TT is not set
+# CONFIG_USB_ISP116X_HCD is not set
 CONFIG_USB_OHCI_HCD=y
+# CONFIG_USB_OHCI_BIG_ENDIAN is not set
+CONFIG_USB_OHCI_LITTLE_ENDIAN=y
 CONFIG_USB_UHCI_HCD=y
 # CONFIG_USB_SL811_HCD is not set
 
@@ -947,12 +999,11 @@ CONFIG_USB_UHCI_HCD=y
 #
 CONFIG_USB_STORAGE=y
 # CONFIG_USB_STORAGE_DEBUG is not set
-# CONFIG_USB_STORAGE_RW_DETECT is not set
 # CONFIG_USB_STORAGE_DATAFAB is not set
 # CONFIG_USB_STORAGE_FREECOM is not set
 # CONFIG_USB_STORAGE_ISD200 is not set
 # CONFIG_USB_STORAGE_DPCM is not set
-# CONFIG_USB_STORAGE_HP8200e is not set
+# CONFIG_USB_STORAGE_USBAT is not set
 # CONFIG_USB_STORAGE_SDDR09 is not set
 # CONFIG_USB_STORAGE_SDDR55 is not set
 # CONFIG_USB_STORAGE_JUMPSHOT is not set
@@ -966,9 +1017,11 @@ CONFIG_USB_HIDINPUT=y
 CONFIG_USB_HIDDEV=y
 # CONFIG_USB_AIPTEK is not set
 # CONFIG_USB_WACOM is not set
+# CONFIG_USB_ACECAD is not set
 # CONFIG_USB_KBTAB is not set
 # CONFIG_USB_POWERMATE is not set
 # CONFIG_USB_MTOUCH is not set
+# CONFIG_USB_ITMTOUCH is not set
 # CONFIG_USB_EGALAX is not set
 # CONFIG_USB_XPAD is not set
 # CONFIG_USB_ATI_REMOTE is not set
@@ -978,7 +1031,6 @@ CONFIG_USB_HIDDEV=y
 #
 # CONFIG_USB_MDC800 is not set
 # CONFIG_USB_MICROTEK is not set
-# CONFIG_USB_HPUSBSCSI is not set
 
 #
 # USB Multimedia devices
@@ -992,6 +1044,7 @@ CONFIG_USB_HIDDEV=y
 # CONFIG_USB_SE401 is not set
 # CONFIG_USB_SN9C102 is not set
 # CONFIG_USB_STV680 is not set
+# CONFIG_USB_PWC is not set
 
 #
 # USB Network Adapters
@@ -1001,6 +1054,7 @@ CONFIG_USB_HIDDEV=y
 # CONFIG_USB_PEGASUS is not set
 # CONFIG_USB_RTL8150 is not set
 # CONFIG_USB_USBNET is not set
+CONFIG_USB_MON=y
 
 #
 # USB port drivers
@@ -1016,7 +1070,6 @@ CONFIG_USB_HIDDEV=y
 #
 # CONFIG_USB_EMI62 is not set
 # CONFIG_USB_EMI26 is not set
-# CONFIG_USB_TIGL is not set
 # CONFIG_USB_AUERSWALD is not set
 # CONFIG_USB_RIO500 is not set
 # CONFIG_USB_LEGOTOWER is not set
@@ -1025,9 +1078,11 @@ CONFIG_USB_HIDDEV=y
 # CONFIG_USB_CYTHERM is not set
 # CONFIG_USB_PHIDGETKIT is not set
 # CONFIG_USB_PHIDGETSERVO is not set
+# CONFIG_USB_IDMOUSE is not set
+# CONFIG_USB_SISUSBVGA is not set
 
 #
-# USB ATM/DSL drivers
+# USB DSL modem support
 #
 
 #
@@ -1040,6 +1095,15 @@ CONFIG_USB_HIDDEV=y
 #
 # CONFIG_MMC is not set
 
+#
+# InfiniBand support
+#
+# CONFIG_INFINIBAND is not set
+
+#
+# SN Devices
+#
+
 #
 # File systems
 #
@@ -1047,6 +1111,7 @@ CONFIG_EXT2_FS=y
 CONFIG_EXT2_FS_XATTR=y
 # CONFIG_EXT2_FS_POSIX_ACL is not set
 # CONFIG_EXT2_FS_SECURITY is not set
+# CONFIG_EXT2_FS_XIP is not set
 CONFIG_EXT3_FS=y
 CONFIG_EXT3_FS_XATTR=y
 # CONFIG_EXT3_FS_POSIX_ACL is not set
@@ -1056,6 +1121,10 @@ CONFIG_JBD=y
 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
@@ -1089,7 +1158,6 @@ CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
 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=y
@@ -1120,15 +1188,18 @@ CONFIG_RAMFS=y
 #
 CONFIG_NFS_FS=y
 CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
 CONFIG_NFS_V4=y
 # CONFIG_NFS_DIRECTIO is not set
 CONFIG_NFSD=y
 CONFIG_NFSD_V3=y
+# CONFIG_NFSD_V3_ACL is not set
 # CONFIG_NFSD_V4 is not set
 # CONFIG_NFSD_TCP is not set
 CONFIG_LOCKD=y
 CONFIG_LOCKD_V4=y
 CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
 CONFIG_SUNRPC=y
 CONFIG_SUNRPC_GSS=y
 CONFIG_RPCSEC_GSS_KRB5=y
@@ -1209,6 +1280,8 @@ CONFIG_NLS_UTF8=y
 # CONFIG_CRC_CCITT is not set
 CONFIG_CRC32=y
 # CONFIG_LIBCRC32C is not set
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
 
 #
 # Profiling support
@@ -1218,14 +1291,18 @@ CONFIG_CRC32=y
 #
 # Kernel hacking
 #
+# CONFIG_PRINTK_TIME is not set
 CONFIG_DEBUG_KERNEL=y
 CONFIG_MAGIC_SYSRQ=y
+CONFIG_LOG_BUF_SHIFT=17
 # CONFIG_SCHEDSTATS is not set
 # CONFIG_DEBUG_SLAB is not set
 # CONFIG_DEBUG_SPINLOCK is not set
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_KOBJECT is not set
 # CONFIG_DEBUG_INFO is not set
+# CONFIG_DEBUG_FS is not set
+CONFIG_KPROBES=y
 CONFIG_IA64_GRANULE_16MB=y
 # CONFIG_IA64_GRANULE_64MB is not set
 CONFIG_IA64_PRINT_HAZARDS=y
@@ -1252,6 +1329,7 @@ CONFIG_CRYPTO_MD5=y
 # CONFIG_CRYPTO_SHA256 is not set
 # CONFIG_CRYPTO_SHA512 is not set
 # CONFIG_CRYPTO_WP512 is not set
+# CONFIG_CRYPTO_TGR192 is not set
 CONFIG_CRYPTO_DES=y
 # CONFIG_CRYPTO_BLOWFISH is not set
 # CONFIG_CRYPTO_TWOFISH is not set
-- 
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 'arch')

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 fc07695386067a4abbd5ac8c7974059eae0b0a79 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:07:21 +1000
Subject: [PATCH] ppc64: Spread lpevents by default on iSeries

With the previous patch in place, spreading lpevents by default becomes
a one liner.

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 | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index e90dca8bd13..d3779346780 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -69,7 +69,7 @@ struct HvLpEvent * ItLpQueue_getNextLpEvent( struct ItLpQueue * lpQueue )
 	return nextLpEvent;
 }
 
-unsigned long spread_lpevents = 1;
+unsigned long spread_lpevents = NR_CPUS;
 
 int ItLpQueue_isLpIntPending( struct ItLpQueue * lpQueue )
 {
-- 
cgit v1.2.3-70-g09d2


From 0c885c175c06bdfe13e88d974d56a5e93ad2f544 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:07:33 +1000
Subject: [PATCH] ppc64: Move set_spread_lpevents() into ItLpQueue.c

The only code outside ItLpQueue.c that refers to spread_lpevents is in
set_apread_lpevents(), so move it inside ItLpQueue.c and make spread_lpevents
static.

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 | 20 --------------------
 2 files changed, 21 insertions(+), 21 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index d3779346780..61be23ed500 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -69,7 +69,7 @@ struct HvLpEvent * ItLpQueue_getNextLpEvent( struct ItLpQueue * lpQueue )
 	return nextLpEvent;
 }
 
-unsigned long spread_lpevents = NR_CPUS;
+static unsigned long spread_lpevents = NR_CPUS;
 
 int ItLpQueue_isLpIntPending( struct ItLpQueue * lpQueue )
 {
@@ -166,3 +166,23 @@ unsigned ItLpQueue_process( struct ItLpQueue * lpQueue, struct pt_regs *regs )
 
 	return numIntsProcessed;
 }
+
+static int set_spread_lpevents(char *str)
+{
+	unsigned long val = simple_strtoul(str, NULL, 0);
+
+	/*
+	 * The parameter is the number of processors to share in processing
+	 * lp events.
+	 */
+	if (( val > 0) && (val <= NR_CPUS)) {
+		spread_lpevents = val;
+		printk("lpevent processing spread over %ld processors\n", val);
+	} else {
+		printk("invalid spread_lpevents %ld\n", val);
+	}
+
+	return 1;
+}
+__setup("spread_lpevents=", set_spread_lpevents);
+
diff --git a/arch/ppc64/kernel/iSeries_setup.c b/arch/ppc64/kernel/iSeries_setup.c
index 2049b6dbafc..78f28356025 100644
--- a/arch/ppc64/kernel/iSeries_setup.c
+++ b/arch/ppc64/kernel/iSeries_setup.c
@@ -853,26 +853,6 @@ static int __init iSeries_src_init(void)
 
 late_initcall(iSeries_src_init);
 
-static int set_spread_lpevents(char *str)
-{
-	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)) {
-		spread_lpevents = val;
-		printk("lpevent processing spread over %ld processors\n", val);
-	} else {
-		printk("invalid spread_lpevents %ld\n", val);
-	}
-
-	return 1;
-}
-__setup("spread_lpevents=", set_spread_lpevents);
-
 #ifndef CONFIG_PCI
 void __init iSeries_init_IRQ(void) { }
 #endif
-- 
cgit v1.2.3-70-g09d2


From ee48444b85f498d99592835f61125385d8e9c975 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:07:48 +1000
Subject: [PATCH] ppc64: Reorganise the paca initialisation macros

This patch updates the macros that initialise the paca to remove the lpq
parameter. It also rearranges them a bit with the hope of making them a
bit clearer.

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/pacaData.c | 211 +++++++++++++------------------------------
 1 file changed, 65 insertions(+), 146 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/pacaData.c b/arch/ppc64/kernel/pacaData.c
index ebfb517019e..6316188737b 100644
--- a/arch/ppc64/kernel/pacaData.c
+++ b/arch/ppc64/kernel/pacaData.c
@@ -42,20 +42,7 @@ extern unsigned long __toc_start;
  * processors.  The processor VPD array needs one entry per physical
  * processor (not thread).
  */
-#ifdef CONFIG_PPC_ISERIES
-#define EXTRA_INITS(number, lpq)					    \
-	.lppaca_ptr = &paca[number].lppaca,				    \
-	.reg_save_ptr = &paca[number].reg_save,				    \
-	.reg_save = {							    \
-		.xDesc = 0xd397d9e2,	/* "LpRS" */			    \
-		.xSize = sizeof(struct ItLpRegSave)			    \
-	},
-#else
-#define EXTRA_INITS(number, lpq)
-#endif
-
-#define PACAINITDATA(number,start,lpq,asrr,asrv)			    \
-{									    \
+#define PACA_INIT_COMMON(number, start, asrr, asrv)			    \
 	.lock_token = 0x8000,						    \
 	.paca_index = (number),		/* Paca Index */		    \
 	.default_decr = 0x00ff0000,	/* Initial Decr */		    \
@@ -73,147 +60,79 @@ extern unsigned long __toc_start;
 		.end_of_quantum = 0xfffffffffffffffful,			    \
 		.slb_count = 64,					    \
 	},								    \
-	EXTRA_INITS((number), (lpq))					    \
-}
 
-struct paca_struct paca[] = {
 #ifdef CONFIG_PPC_ISERIES
-	PACAINITDATA( 0, 1, &xItLpQueue, 0, STAB0_VIRT_ADDR),
+#define PACA_INIT_ISERIES(number)					    \
+	.lppaca_ptr = &paca[number].lppaca,				    \
+	.reg_save_ptr = &paca[number].reg_save,				    \
+	.reg_save = {							    \
+		.xDesc = 0xd397d9e2,	/* "LpRS" */			    \
+		.xSize = sizeof(struct ItLpRegSave)			    \
+	}
+
+#define PACA_INIT(number)						    \
+{									    \
+	PACA_INIT_COMMON(number, 0, 0, 0)				    \
+	PACA_INIT_ISERIES(number)					    \
+}
+
+#define BOOTCPU_PACA_INIT(number)					    \
+{									    \
+	PACA_INIT_COMMON(number, 1, 0, STAB0_VIRT_ADDR)			    \
+	PACA_INIT_ISERIES(number)					    \
+}
+
 #else
-	PACAINITDATA( 0, 1, NULL, STAB0_PHYS_ADDR, STAB0_VIRT_ADDR),
+#define PACA_INIT(number)						    \
+{									    \
+	PACA_INIT_COMMON(number, 0, 0, 0)				    \
+}
+
+#define BOOTCPU_PACA_INIT(number)					    \
+{									    \
+	PACA_INIT_COMMON(number, 1, STAB0_PHYS_ADDR, STAB0_VIRT_ADDR)	    \
+}
 #endif
+
+struct paca_struct paca[] = {
+	BOOTCPU_PACA_INIT(0),
 #if NR_CPUS > 1
-	PACAINITDATA( 1, 0, NULL, 0, 0),
-	PACAINITDATA( 2, 0, NULL, 0, 0),
-	PACAINITDATA( 3, 0, NULL, 0, 0),
+	PACA_INIT(  1), PACA_INIT(  2), PACA_INIT(  3),
 #if NR_CPUS > 4
-	PACAINITDATA( 4, 0, NULL, 0, 0),
-	PACAINITDATA( 5, 0, NULL, 0, 0),
-	PACAINITDATA( 6, 0, NULL, 0, 0),
-	PACAINITDATA( 7, 0, NULL, 0, 0),
+	PACA_INIT(  4), PACA_INIT(  5), PACA_INIT(  6), PACA_INIT(  7),
 #if NR_CPUS > 8
-	PACAINITDATA( 8, 0, NULL, 0, 0),
-	PACAINITDATA( 9, 0, NULL, 0, 0),
-	PACAINITDATA(10, 0, NULL, 0, 0),
-	PACAINITDATA(11, 0, NULL, 0, 0),
-	PACAINITDATA(12, 0, NULL, 0, 0),
-	PACAINITDATA(13, 0, NULL, 0, 0),
-	PACAINITDATA(14, 0, NULL, 0, 0),
-	PACAINITDATA(15, 0, NULL, 0, 0),
-	PACAINITDATA(16, 0, NULL, 0, 0),
-	PACAINITDATA(17, 0, NULL, 0, 0),
-	PACAINITDATA(18, 0, NULL, 0, 0),
-	PACAINITDATA(19, 0, NULL, 0, 0),
-	PACAINITDATA(20, 0, NULL, 0, 0),
-	PACAINITDATA(21, 0, NULL, 0, 0),
-	PACAINITDATA(22, 0, NULL, 0, 0),
-	PACAINITDATA(23, 0, NULL, 0, 0),
-	PACAINITDATA(24, 0, NULL, 0, 0),
-	PACAINITDATA(25, 0, NULL, 0, 0),
-	PACAINITDATA(26, 0, NULL, 0, 0),
-	PACAINITDATA(27, 0, NULL, 0, 0),
-	PACAINITDATA(28, 0, NULL, 0, 0),
-	PACAINITDATA(29, 0, NULL, 0, 0),
-	PACAINITDATA(30, 0, NULL, 0, 0),
-	PACAINITDATA(31, 0, NULL, 0, 0),
+	PACA_INIT(  8), PACA_INIT(  9), PACA_INIT( 10), PACA_INIT( 11),
+	PACA_INIT( 12), PACA_INIT( 13), PACA_INIT( 14), PACA_INIT( 15),
+	PACA_INIT( 16), PACA_INIT( 17), PACA_INIT( 18), PACA_INIT( 19),
+	PACA_INIT( 20), PACA_INIT( 21), PACA_INIT( 22), PACA_INIT( 23),
+	PACA_INIT( 24), PACA_INIT( 25), PACA_INIT( 26), PACA_INIT( 27),
+	PACA_INIT( 28), PACA_INIT( 29), PACA_INIT( 30), PACA_INIT( 31),
 #if NR_CPUS > 32
-	PACAINITDATA(32, 0, NULL, 0, 0),
-	PACAINITDATA(33, 0, NULL, 0, 0),
-	PACAINITDATA(34, 0, NULL, 0, 0),
-	PACAINITDATA(35, 0, NULL, 0, 0),
-	PACAINITDATA(36, 0, NULL, 0, 0),
-	PACAINITDATA(37, 0, NULL, 0, 0),
-	PACAINITDATA(38, 0, NULL, 0, 0),
-	PACAINITDATA(39, 0, NULL, 0, 0),
-	PACAINITDATA(40, 0, NULL, 0, 0),
-	PACAINITDATA(41, 0, NULL, 0, 0),
-	PACAINITDATA(42, 0, NULL, 0, 0),
-	PACAINITDATA(43, 0, NULL, 0, 0),
-	PACAINITDATA(44, 0, NULL, 0, 0),
-	PACAINITDATA(45, 0, NULL, 0, 0),
-	PACAINITDATA(46, 0, NULL, 0, 0),
-	PACAINITDATA(47, 0, NULL, 0, 0),
-	PACAINITDATA(48, 0, NULL, 0, 0),
-	PACAINITDATA(49, 0, NULL, 0, 0),
-	PACAINITDATA(50, 0, NULL, 0, 0),
-	PACAINITDATA(51, 0, NULL, 0, 0),
-	PACAINITDATA(52, 0, NULL, 0, 0),
-	PACAINITDATA(53, 0, NULL, 0, 0),
-	PACAINITDATA(54, 0, NULL, 0, 0),
-	PACAINITDATA(55, 0, NULL, 0, 0),
-	PACAINITDATA(56, 0, NULL, 0, 0),
-	PACAINITDATA(57, 0, NULL, 0, 0),
-	PACAINITDATA(58, 0, NULL, 0, 0),
-	PACAINITDATA(59, 0, NULL, 0, 0),
-	PACAINITDATA(60, 0, NULL, 0, 0),
-	PACAINITDATA(61, 0, NULL, 0, 0),
-	PACAINITDATA(62, 0, NULL, 0, 0),
-	PACAINITDATA(63, 0, NULL, 0, 0),
+	PACA_INIT( 32), PACA_INIT( 33), PACA_INIT( 34), PACA_INIT( 35),
+	PACA_INIT( 36), PACA_INIT( 37), PACA_INIT( 38), PACA_INIT( 39),
+	PACA_INIT( 40), PACA_INIT( 41), PACA_INIT( 42), PACA_INIT( 43),
+	PACA_INIT( 44), PACA_INIT( 45), PACA_INIT( 46), PACA_INIT( 47),
+	PACA_INIT( 48), PACA_INIT( 49), PACA_INIT( 50), PACA_INIT( 51),
+	PACA_INIT( 52), PACA_INIT( 53), PACA_INIT( 54), PACA_INIT( 55),
+	PACA_INIT( 56), PACA_INIT( 57), PACA_INIT( 58), PACA_INIT( 59),
+	PACA_INIT( 60), PACA_INIT( 61), PACA_INIT( 62), PACA_INIT( 63),
 #if NR_CPUS > 64
-	PACAINITDATA(64, 0, NULL, 0, 0),
-	PACAINITDATA(65, 0, NULL, 0, 0),
-	PACAINITDATA(66, 0, NULL, 0, 0),
-	PACAINITDATA(67, 0, NULL, 0, 0),
-	PACAINITDATA(68, 0, NULL, 0, 0),
-	PACAINITDATA(69, 0, NULL, 0, 0),
-	PACAINITDATA(70, 0, NULL, 0, 0),
-	PACAINITDATA(71, 0, NULL, 0, 0),
-	PACAINITDATA(72, 0, NULL, 0, 0),
-	PACAINITDATA(73, 0, NULL, 0, 0),
-	PACAINITDATA(74, 0, NULL, 0, 0),
-	PACAINITDATA(75, 0, NULL, 0, 0),
-	PACAINITDATA(76, 0, NULL, 0, 0),
-	PACAINITDATA(77, 0, NULL, 0, 0),
-	PACAINITDATA(78, 0, NULL, 0, 0),
-	PACAINITDATA(79, 0, NULL, 0, 0),
-	PACAINITDATA(80, 0, NULL, 0, 0),
-	PACAINITDATA(81, 0, NULL, 0, 0),
-	PACAINITDATA(82, 0, NULL, 0, 0),
-	PACAINITDATA(83, 0, NULL, 0, 0),
-	PACAINITDATA(84, 0, NULL, 0, 0),
-	PACAINITDATA(85, 0, NULL, 0, 0),
-	PACAINITDATA(86, 0, NULL, 0, 0),
-	PACAINITDATA(87, 0, NULL, 0, 0),
-	PACAINITDATA(88, 0, NULL, 0, 0),
-	PACAINITDATA(89, 0, NULL, 0, 0),
-	PACAINITDATA(90, 0, NULL, 0, 0),
-	PACAINITDATA(91, 0, NULL, 0, 0),
-	PACAINITDATA(92, 0, NULL, 0, 0),
-	PACAINITDATA(93, 0, NULL, 0, 0),
-	PACAINITDATA(94, 0, NULL, 0, 0),
-	PACAINITDATA(95, 0, NULL, 0, 0),
-	PACAINITDATA(96, 0, NULL, 0, 0),
-	PACAINITDATA(97, 0, NULL, 0, 0),
-	PACAINITDATA(98, 0, NULL, 0, 0),
-	PACAINITDATA(99, 0, NULL, 0, 0),
-	PACAINITDATA(100, 0, NULL, 0, 0),
-	PACAINITDATA(101, 0, NULL, 0, 0),
-	PACAINITDATA(102, 0, NULL, 0, 0),
-	PACAINITDATA(103, 0, NULL, 0, 0),
-	PACAINITDATA(104, 0, NULL, 0, 0),
-	PACAINITDATA(105, 0, NULL, 0, 0),
-	PACAINITDATA(106, 0, NULL, 0, 0),
-	PACAINITDATA(107, 0, NULL, 0, 0),
-	PACAINITDATA(108, 0, NULL, 0, 0),
-	PACAINITDATA(109, 0, NULL, 0, 0),
-	PACAINITDATA(110, 0, NULL, 0, 0),
-	PACAINITDATA(111, 0, NULL, 0, 0),
-	PACAINITDATA(112, 0, NULL, 0, 0),
-	PACAINITDATA(113, 0, NULL, 0, 0),
-	PACAINITDATA(114, 0, NULL, 0, 0),
-	PACAINITDATA(115, 0, NULL, 0, 0),
-	PACAINITDATA(116, 0, NULL, 0, 0),
-	PACAINITDATA(117, 0, NULL, 0, 0),
-	PACAINITDATA(118, 0, NULL, 0, 0),
-	PACAINITDATA(119, 0, NULL, 0, 0),
-	PACAINITDATA(120, 0, NULL, 0, 0),
-	PACAINITDATA(121, 0, NULL, 0, 0),
-	PACAINITDATA(122, 0, NULL, 0, 0),
-	PACAINITDATA(123, 0, NULL, 0, 0),
-	PACAINITDATA(124, 0, NULL, 0, 0),
-	PACAINITDATA(125, 0, NULL, 0, 0),
-	PACAINITDATA(126, 0, NULL, 0, 0),
-	PACAINITDATA(127, 0, NULL, 0, 0),
+	PACA_INIT( 64), PACA_INIT( 65), PACA_INIT( 66), PACA_INIT( 67),
+	PACA_INIT( 68), PACA_INIT( 69), PACA_INIT( 70), PACA_INIT( 71),
+	PACA_INIT( 72), PACA_INIT( 73), PACA_INIT( 74), PACA_INIT( 75),
+	PACA_INIT( 76), PACA_INIT( 77), PACA_INIT( 78), PACA_INIT( 79),
+	PACA_INIT( 80), PACA_INIT( 81), PACA_INIT( 82), PACA_INIT( 83),
+	PACA_INIT( 84), PACA_INIT( 85), PACA_INIT( 86), PACA_INIT( 87),
+	PACA_INIT( 88), PACA_INIT( 89), PACA_INIT( 90), PACA_INIT( 91),
+	PACA_INIT( 92), PACA_INIT( 93), PACA_INIT( 94), PACA_INIT( 95),
+	PACA_INIT( 96), PACA_INIT( 97), PACA_INIT( 98), PACA_INIT( 99),
+	PACA_INIT(100), PACA_INIT(101), PACA_INIT(102), PACA_INIT(103),
+	PACA_INIT(104), PACA_INIT(105), PACA_INIT(106), PACA_INIT(107),
+	PACA_INIT(108), PACA_INIT(109), PACA_INIT(110), PACA_INIT(111),
+	PACA_INIT(112), PACA_INIT(113), PACA_INIT(114), PACA_INIT(115),
+	PACA_INIT(116), PACA_INIT(117), PACA_INIT(118), PACA_INIT(119),
+	PACA_INIT(120), PACA_INIT(121), PACA_INIT(122), PACA_INIT(123),
+	PACA_INIT(124), PACA_INIT(125), PACA_INIT(126), PACA_INIT(127),
 #endif
 #endif
 #endif
-- 
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 'arch')

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 'arch')

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 7b01328d455b50ff040d3a06b342ca370b1d8b0a Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:08:44 +1000
Subject: [PATCH] ppc64: Move xItLpQueue proc code into ItLpQueue.c

Move the code that displays xItLpQueue values in /proc into
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    | 59 ++++++++++++++++++++++++++++++++++++++++
 arch/ppc64/kernel/iSeries_proc.c | 48 --------------------------------
 2 files changed, 59 insertions(+), 48 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index 091aaed4842..11cd31dfcef 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -12,12 +12,26 @@
 #include <linux/kernel.h>
 #include <linux/sched.h>
 #include <linux/bootmem.h>
+#include <linux/seq_file.h>
+#include <linux/proc_fs.h>
 #include <asm/system.h>
 #include <asm/paca.h>
 #include <asm/iSeries/ItLpQueue.h>
 #include <asm/iSeries/HvLpEvent.h>
 #include <asm/iSeries/HvCallEvent.h>
 
+static char *event_types[9] = {
+	"Hypervisor\t\t",
+	"Machine Facilities\t",
+	"Session Manager\t",
+	"SPD I/O\t\t",
+	"Virtual Bus\t\t",
+	"PCI I/O\t\t",
+	"RIO I/O\t\t",
+	"Virtual Lan\t\t",
+	"Virtual I/O\t\t"
+};
+
 static __inline__ int set_inUse(void)
 {
 	int t;
@@ -208,3 +222,48 @@ void setup_hvlpevent_queue(void)
 					(LpEventStackSize - LpEventMaxSize);
 	xItLpQueue.xIndex = 0;
 }
+
+static int proc_lpevents_show(struct seq_file *m, void *v)
+{
+	unsigned int i;
+
+	seq_printf(m, "LpEventQueue 0\n");
+	seq_printf(m, "  events processed:\t%lu\n",
+		   (unsigned long)xItLpQueue.xLpIntCount);
+
+	for (i = 0; i < 9; ++i)
+		seq_printf(m, "    %s %10lu\n", event_types[i],
+			   (unsigned long)xItLpQueue.xLpIntCountByType[i]);
+
+	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);
+
+	return 0;
+}
+
+static int proc_lpevents_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, proc_lpevents_show, NULL);
+}
+
+static struct file_operations proc_lpevents_operations = {
+	.open		= proc_lpevents_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+static int __init proc_lpevents_init(void)
+{
+	struct proc_dir_entry *e;
+
+	e = create_proc_entry("iSeries/lpevents", S_IFREG|S_IRUGO, NULL);
+	if (e)
+		e->proc_fops = &proc_lpevents_operations;
+
+	return 0;
+}
+__initcall(proc_lpevents_init);
+
diff --git a/arch/ppc64/kernel/iSeries_proc.c b/arch/ppc64/kernel/iSeries_proc.c
index 356bd9931fc..0fe3116eba2 100644
--- a/arch/ppc64/kernel/iSeries_proc.c
+++ b/arch/ppc64/kernel/iSeries_proc.c
@@ -40,50 +40,6 @@ static int __init iseries_proc_create(void)
 }
 core_initcall(iseries_proc_create);
 
-static char *event_types[9] = {
-	"Hypervisor\t\t",
-	"Machine Facilities\t",
-	"Session Manager\t",
-	"SPD I/O\t\t",
-	"Virtual Bus\t\t",
-	"PCI I/O\t\t",
-	"RIO I/O\t\t",
-	"Virtual Lan\t\t",
-	"Virtual I/O\t\t"
-};
-
-static int proc_lpevents_show(struct seq_file *m, void *v)
-{
-	unsigned int i;
-
-	seq_printf(m, "LpEventQueue 0\n");
-	seq_printf(m, "  events processed:\t%lu\n",
-		   (unsigned long)xItLpQueue.xLpIntCount);
-
-	for (i = 0; i < 9; ++i)
-		seq_printf(m, "    %s %10lu\n", event_types[i],
-			   (unsigned long)xItLpQueue.xLpIntCountByType[i]);
-
-	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);
-
-	return 0;
-}
-
-static int proc_lpevents_open(struct inode *inode, struct file *file)
-{
-	return single_open(file, proc_lpevents_show, NULL);
-}
-
-static struct file_operations proc_lpevents_operations = {
-	.open		= proc_lpevents_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
-};
-
 static unsigned long startTitan = 0;
 static unsigned long startTb = 0;
 
@@ -148,10 +104,6 @@ static int __init iseries_proc_init(void)
 {
 	struct proc_dir_entry *e;
 
-	e = create_proc_entry("iSeries/lpevents", S_IFREG|S_IRUGO, NULL);
-	if (e)
-		e->proc_fops = &proc_lpevents_operations;
-
 	e = create_proc_entry("iSeries/titanTod", S_IFREG|S_IRUGO, NULL);
 	if (e)
 		e->proc_fops = &proc_titantod_operations;
-- 
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 'arch')

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 ab354b637924beb33dcc23eedc9482f2c692188f Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:12:21 +1000
Subject: [PATCH] ppc64: Move definition of xItLpQueue

The xItLpQueue is declared in LparData.c, move it into 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 | 8 ++++++++
 arch/ppc64/kernel/LparData.c  | 7 -------
 2 files changed, 8 insertions(+), 7 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index ac0b05a6e06..7ddbfb9750d 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -20,6 +20,14 @@
 #include <asm/iSeries/HvLpEvent.h>
 #include <asm/iSeries/HvCallEvent.h>
 
+/*
+ * The LpQueue is used to pass event data from the hypervisor to
+ * the partition.  This is where I/O interrupt events are communicated.
+ *
+ * It is written to by the hypervisor so cannot end up in the BSS.
+ */
+struct ItLpQueue xItLpQueue __attribute__((__section__(".data")));
+
 static char *event_types[9] = {
 	"Hypervisor\t\t",
 	"Machine Facilities\t",
diff --git a/arch/ppc64/kernel/LparData.c b/arch/ppc64/kernel/LparData.c
index badc5a44361..f42ee35f927 100644
--- a/arch/ppc64/kernel/LparData.c
+++ b/arch/ppc64/kernel/LparData.c
@@ -28,13 +28,6 @@
 #include <asm/iSeries/IoHriProcessorVpd.h>
 #include <asm/iSeries/ItSpCommArea.h>
 
-/* The LpQueue is used to pass event data from the hypervisor to
- * the partition.  This is where I/O interrupt events are communicated.
- */
-
-/* May be filled in by the hypervisor so cannot end up in the BSS */
-struct ItLpQueue xItLpQueue __attribute__((__section__(".data")));
-
 
 /* The HvReleaseData is the root of the information shared between 
  * the hypervisor and Linux.  
-- 
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 'arch')

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 'arch')

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 'arch')

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 'arch')

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 9b0470200a2441766599ad84f92ab9daca8ed86d Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:16:18 +1000
Subject: [PATCH] ppc64: Cleanup proc printing of event types

The code that prints event counts by type uses a hand-coded number of tabs
to get the alignment right. Instead use a printf alignment which will allow
allow us to use the event_type strings elsewhere in the future.

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 | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index 4a6ab9de629..83fb36a9bc0 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -31,15 +31,15 @@ struct hvlpevent_queue hvlpevent_queue __attribute__((__section__(".data")));
 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",
-	"SPD I/O\t\t",
-	"Virtual Bus\t\t",
-	"PCI I/O\t\t",
-	"RIO I/O\t\t",
-	"Virtual Lan\t\t",
-	"Virtual I/O\t\t"
+	"Hypervisor",
+	"Machine Facilities",
+	"Session Manager",
+	"SPD I/O",
+	"Virtual Bus",
+	"PCI I/O",
+	"RIO I/O",
+	"Virtual Lan",
+	"Virtual I/O"
 };
 
 static __inline__ int set_inUse(void)
@@ -248,7 +248,7 @@ static int proc_lpevents_show(struct seq_file *m, void *v)
 			sum += per_cpu(hvlpevent_counts, cpu)[i];
 		}
 
-		seq_printf(m, "    %s %10lu\n", event_types[i], sum);
+		seq_printf(m, "    %-20s %10lu\n", event_types[i], sum);
 	}
 
 	seq_printf(m, "\n  events processed by processor:\n");
-- 
cgit v1.2.3-70-g09d2


From 38fcdcfe38fc3f8972c906db64cd7d540b7760e8 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:16:28 +1000
Subject: [PATCH] ppc64: Cleanup whitespace in arch/ppc64/kernel/ItLpQueue.c

Just cleanup white space.

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 | 58 +++++++++++++++++++++----------------------
 1 file changed, 29 insertions(+), 29 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index 83fb36a9bc0..61a9dbdd295 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -1,7 +1,7 @@
 /*
  * ItLpQueue.c
  * 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
@@ -74,21 +74,21 @@ unsigned long ItLpQueueInProcess = 0;
 
 static struct HvLpEvent * get_next_hvlpevent(void)
 {
-	struct HvLpEvent * nextLpEvent = 
+	struct HvLpEvent * nextLpEvent =
 		(struct HvLpEvent *)hvlpevent_queue.xSlicCurEventPtr;
-	if ( nextLpEvent->xFlags.xValid ) {
+	if (nextLpEvent->xFlags.xValid) {
 		/* rmb() needed only for weakly consistent machines (regatta) */
 		rmb();
 		/* Set pointer to next potential event */
 		hvlpevent_queue.xSlicCurEventPtr += ((nextLpEvent->xSizeMinus1 +
-				      LpEventAlign ) /
-				      LpEventAlign ) *
+				      LpEventAlign) /
+				      LpEventAlign) *
 				      LpEventAlign;
 		/* Wrap to beginning if no room at end */
 		if (hvlpevent_queue.xSlicCurEventPtr > hvlpevent_queue.xSlicLastValidEventPtr)
 			hvlpevent_queue.xSlicCurEventPtr = hvlpevent_queue.xSlicEventStackPtr;
 	}
-	else 
+	else
 		nextLpEvent = NULL;
 
 	return nextLpEvent;
@@ -107,23 +107,23 @@ int hvlpevent_is_pending(void)
 	return next_event->xFlags.xValid | hvlpevent_queue.xPlicOverflowIntPending;
 }
 
-static void hvlpevent_clear_valid( 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
 	 * look like valid bits (on 64-byte boundaries)
-   	 */
-	unsigned extra = (( event->xSizeMinus1 + LpEventAlign ) /
-						 LpEventAlign ) - 1;
-	switch ( extra ) {
-	  case 3:
+	 */
+	unsigned extra = ((event->xSizeMinus1 + LpEventAlign) /
+						 LpEventAlign) - 1;
+	switch (extra) {
+	case 3:
 	   ((struct HvLpEvent*)((char*)event+3*LpEventAlign))->xFlags.xValid=0;
-	  case 2:
+	case 2:
 	   ((struct HvLpEvent*)((char*)event+2*LpEventAlign))->xFlags.xValid=0;
-	  case 1:
+	case 1:
 	   ((struct HvLpEvent*)((char*)event+1*LpEventAlign))->xFlags.xValid=0;
-	  case 0:
-	   ;	
+	case 0:
+		;
 	}
 	mb();
 	event->xFlags.xValid = 0;
@@ -136,7 +136,7 @@ void process_hvlpevents(struct pt_regs *regs)
 	/* If we have recursed, just return */
 	if ( !set_inUse() )
 		return;
-	
+
 	if (ItLpQueueInProcess == 0)
 		ItLpQueueInProcess = 1;
 	else
@@ -144,35 +144,35 @@ void process_hvlpevents(struct pt_regs *regs)
 
 	for (;;) {
 		nextLpEvent = get_next_hvlpevent();
-		if ( nextLpEvent ) {
-			/* Call appropriate handler here, passing 
+		if (nextLpEvent) {
+			/* Call appropriate handler here, passing
 			 * a pointer to the LpEvent.  The handler
 			 * must make a copy of the LpEvent if it
 			 * needs it in a bottom half. (perhaps for
 			 * an ACK)
-			 *	
-			 *  Handlers are responsible for ACK processing 
+			 *
+			 *  Handlers are responsible for ACK processing
 			 *
 			 * The Hypervisor guarantees that LpEvents will
 			 * only be delivered with types that we have
 			 * registered for, so no type check is necessary
 			 * here!
-  			 */
-			if ( nextLpEvent->xType < HvLpEvent_Type_NumTypes )
+			 */
+			if (nextLpEvent->xType < HvLpEvent_Type_NumTypes)
 				__get_cpu_var(hvlpevent_counts)[nextLpEvent->xType]++;
-			if ( nextLpEvent->xType < HvLpEvent_Type_NumTypes &&
-			     lpEventHandler[nextLpEvent->xType] ) 
+			if (nextLpEvent->xType < HvLpEvent_Type_NumTypes &&
+					lpEventHandler[nextLpEvent->xType])
 				lpEventHandler[nextLpEvent->xType](nextLpEvent, regs);
 			else
 				printk(KERN_INFO "Unexpected Lp Event type=%d\n", nextLpEvent->xType );
-			
-			hvlpevent_clear_valid( nextLpEvent );
-		} else if ( hvlpevent_queue.xPlicOverflowIntPending )
+
+			hvlpevent_clear_valid(nextLpEvent);
+		} else if (hvlpevent_queue.xPlicOverflowIntPending)
 			/*
 			 * No more valid events. If overflow events are
 			 * pending process them
 			 */
-			HvCallEvent_getOverflowLpEvents( hvlpevent_queue.xIndex);
+			HvCallEvent_getOverflowLpEvents(hvlpevent_queue.xIndex);
 		else
 			break;
 	}
-- 
cgit v1.2.3-70-g09d2


From ffe1b7e14e6b606bd84cab564aa2f481dbd4e418 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 30 Jun 2005 15:16:48 +1000
Subject: [PATCH] ppc64: Formatting cleanups in arch/ppc64/kernel/ItLpQueue.c

Just formatting cleanups:
 * rename some "nextLpEvent" variables to just "event"
 * make code fit in 80 columns
 * use brackets around if/else
 * use a temporary to make hvlpevent_clear_valid clearer

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 | 72 ++++++++++++++++++++++++-------------------
 1 file changed, 41 insertions(+), 31 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c
index 61a9dbdd295..a4f32cbf529 100644
--- a/arch/ppc64/kernel/ItLpQueue.c
+++ b/arch/ppc64/kernel/ItLpQueue.c
@@ -74,24 +74,27 @@ unsigned long ItLpQueueInProcess = 0;
 
 static struct HvLpEvent * get_next_hvlpevent(void)
 {
-	struct HvLpEvent * nextLpEvent =
-		(struct HvLpEvent *)hvlpevent_queue.xSlicCurEventPtr;
-	if (nextLpEvent->xFlags.xValid) {
+	struct HvLpEvent * event;
+	event = (struct HvLpEvent *)hvlpevent_queue.xSlicCurEventPtr;
+
+	if (event->xFlags.xValid) {
 		/* rmb() needed only for weakly consistent machines (regatta) */
 		rmb();
 		/* Set pointer to next potential event */
-		hvlpevent_queue.xSlicCurEventPtr += ((nextLpEvent->xSizeMinus1 +
-				      LpEventAlign) /
-				      LpEventAlign) *
-				      LpEventAlign;
+		hvlpevent_queue.xSlicCurEventPtr += ((event->xSizeMinus1 +
+				LpEventAlign) / LpEventAlign) * LpEventAlign;
+
 		/* Wrap to beginning if no room at end */
-		if (hvlpevent_queue.xSlicCurEventPtr > hvlpevent_queue.xSlicLastValidEventPtr)
-			hvlpevent_queue.xSlicCurEventPtr = hvlpevent_queue.xSlicEventStackPtr;
+		if (hvlpevent_queue.xSlicCurEventPtr >
+				hvlpevent_queue.xSlicLastValidEventPtr) {
+			hvlpevent_queue.xSlicCurEventPtr =
+				hvlpevent_queue.xSlicEventStackPtr;
+		}
+	} else {
+		event = NULL;
 	}
-	else
-		nextLpEvent = NULL;
 
-	return nextLpEvent;
+	return event;
 }
 
 static unsigned long spread_lpevents = NR_CPUS;
@@ -104,34 +107,41 @@ int hvlpevent_is_pending(void)
 		return 0;
 
 	next_event = (struct HvLpEvent *)hvlpevent_queue.xSlicCurEventPtr;
-	return next_event->xFlags.xValid | hvlpevent_queue.xPlicOverflowIntPending;
+
+	return next_event->xFlags.xValid |
+		hvlpevent_queue.xPlicOverflowIntPending;
 }
 
 static void hvlpevent_clear_valid(struct HvLpEvent * event)
 {
-	/* Clear the valid bit of the event
-	 * Also clear bits within this event that might
-	 * look like valid bits (on 64-byte boundaries)
+	/* Tell the Hypervisor that we're done with this event.
+	 * Also clear bits within this event that might look like valid bits.
+	 * ie. on 64-byte boundaries.
 	 */
+	struct HvLpEvent *tmp;
 	unsigned extra = ((event->xSizeMinus1 + LpEventAlign) /
 						 LpEventAlign) - 1;
+
 	switch (extra) {
 	case 3:
-	   ((struct HvLpEvent*)((char*)event+3*LpEventAlign))->xFlags.xValid=0;
+		tmp = (struct HvLpEvent*)((char*)event + 3 * LpEventAlign);
+		tmp->xFlags.xValid = 0;
 	case 2:
-	   ((struct HvLpEvent*)((char*)event+2*LpEventAlign))->xFlags.xValid=0;
+		tmp = (struct HvLpEvent*)((char*)event + 2 * LpEventAlign);
+		tmp->xFlags.xValid = 0;
 	case 1:
-	   ((struct HvLpEvent*)((char*)event+1*LpEventAlign))->xFlags.xValid=0;
-	case 0:
-		;
+		tmp = (struct HvLpEvent*)((char*)event + 1 * LpEventAlign);
+		tmp->xFlags.xValid = 0;
 	}
+
 	mb();
+
 	event->xFlags.xValid = 0;
 }
 
 void process_hvlpevents(struct pt_regs *regs)
 {
-	struct HvLpEvent * nextLpEvent;
+	struct HvLpEvent * event;
 
 	/* If we have recursed, just return */
 	if ( !set_inUse() )
@@ -143,8 +153,8 @@ void process_hvlpevents(struct pt_regs *regs)
 		BUG();
 
 	for (;;) {
-		nextLpEvent = get_next_hvlpevent();
-		if (nextLpEvent) {
+		event = get_next_hvlpevent();
+		if (event) {
 			/* Call appropriate handler here, passing
 			 * a pointer to the LpEvent.  The handler
 			 * must make a copy of the LpEvent if it
@@ -158,15 +168,15 @@ void process_hvlpevents(struct pt_regs *regs)
 			 * registered for, so no type check is necessary
 			 * here!
 			 */
-			if (nextLpEvent->xType < HvLpEvent_Type_NumTypes)
-				__get_cpu_var(hvlpevent_counts)[nextLpEvent->xType]++;
-			if (nextLpEvent->xType < HvLpEvent_Type_NumTypes &&
-					lpEventHandler[nextLpEvent->xType])
-				lpEventHandler[nextLpEvent->xType](nextLpEvent, regs);
+			if (event->xType < HvLpEvent_Type_NumTypes)
+				__get_cpu_var(hvlpevent_counts)[event->xType]++;
+			if (event->xType < HvLpEvent_Type_NumTypes &&
+					lpEventHandler[event->xType])
+				lpEventHandler[event->xType](event, regs);
 			else
-				printk(KERN_INFO "Unexpected Lp Event type=%d\n", nextLpEvent->xType );
+				printk(KERN_INFO "Unexpected Lp Event type=%d\n", event->xType );
 
-			hvlpevent_clear_valid(nextLpEvent);
+			hvlpevent_clear_valid(event);
 		} else if (hvlpevent_queue.xPlicOverflowIntPending)
 			/*
 			 * No more valid events. If overflow events are
-- 
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 'arch')

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 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 'arch')

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 ee93b43a05506667d771c4fb0c384301edd7d036 Mon Sep 17 00:00:00 2001
From: Olaf Hering <olh@suse.de>
Date: Thu, 30 Jun 2005 02:58:49 -0700
Subject: [PATCH] ppc32: use correct register names in
 arch/ppc/kernel/relocate_kernel.S

CONFIG_KEXEC=y doesnt work:

arch/ppc/kernel/relocate_kernel.S:37: Error: unsupported relocation against SRR1
arch/ppc/kernel/relocate_kernel.S:39: Error: unsupported relocation against SRR0

Signed-off-by: Olaf Hering <olh@suse.de>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ppc/kernel/relocate_kernel.S | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/ppc/kernel/relocate_kernel.S b/arch/ppc/kernel/relocate_kernel.S
index 7ff69c4af92..9b2ad48e988 100644
--- a/arch/ppc/kernel/relocate_kernel.S
+++ b/arch/ppc/kernel/relocate_kernel.S
@@ -34,9 +34,9 @@ relocate_new_kernel:
 
 	mr	r8, r0
 	ori     r8, r8, MSR_RI|MSR_ME
-	mtspr	SRR1, r8
+	mtspr	SPRN_SRR1, r8
 	addi	r8, r4, 1f - relocate_new_kernel
-	mtspr	SRR0, r8
+	mtspr	SPRN_SRR0, r8
 	sync
 	rfi
 
-- 
cgit v1.2.3-70-g09d2


From bcbda35ca7470bf0123a7ae685899776f67814b2 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@freescale.com>
Date: Thu, 30 Jun 2005 02:58:55 -0700
Subject: [PATCH] ppc32: Fix pointer check for MPC8540 ADS device

Editor snafu in which the call to ppc_sys_get_pdata got inside the if check
instead of before it.  Oops.

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/mpc8540_ads.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ppc/platforms/85xx/mpc8540_ads.c b/arch/ppc/platforms/85xx/mpc8540_ads.c
index ddd2e9a5bb1..f761fdf160d 100644
--- a/arch/ppc/platforms/85xx/mpc8540_ads.c
+++ b/arch/ppc/platforms/85xx/mpc8540_ads.c
@@ -111,8 +111,8 @@ mpc8540ads_setup_arch(void)
 		memcpy(pdata->mac_addr, binfo->bi_enet1addr, 6);
 	}
 
+	pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC85xx_FEC);
 	if (pdata) {
-		pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC85xx_FEC);
 		pdata->board_flags = 0;
 		pdata->interruptPHY = MPC85xx_IRQ_EXT5;
 		pdata->phyid = 3;
-- 
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 'arch')

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 0ee23b50f1541aacc3b975edae170a1b995b84f5 Mon Sep 17 00:00:00 2001
From: Jesper Juhl <juhl-lkml@dif.dk>
Date: Thu, 30 Jun 2005 02:58:56 -0700
Subject: [PATCH] xtensa: use valid_signal()

xtensa should use valid_signal() instead of testing _NSIG directly like
everyone else.

Signed-off-by: Jesper Juhl <juhl-lkml@dif.dk>
Cc: Chris Zankel <chris@zankel.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/xtensa/kernel/ptrace.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/xtensa/kernel/ptrace.c b/arch/xtensa/kernel/ptrace.c
index 9ef07a4dd2a..2659efdd4e9 100644
--- a/arch/xtensa/kernel/ptrace.c
+++ b/arch/xtensa/kernel/ptrace.c
@@ -22,6 +22,7 @@
 #include <linux/smp.h>
 #include <linux/smp_lock.h>
 #include <linux/security.h>
+#include <linux/signal.h>
 
 #include <asm/pgtable.h>
 #include <asm/page.h>
@@ -239,7 +240,7 @@ int sys_ptrace(long request, long pid, long addr, long data)
 	case PTRACE_CONT: /* restart after signal. */
 	{
 		ret = -EIO;
-		if ((unsigned long) data > _NSIG)
+		if (!valid_signal(data))
 			break;
 		if (request == PTRACE_SYSCALL)
 			set_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
@@ -269,7 +270,7 @@ int sys_ptrace(long request, long pid, long addr, long data)
 
 	case PTRACE_SINGLESTEP:
 		ret = -EIO;
-		if ((unsigned long) data > _NSIG)
+		if (!valid_signal(data))
 			break;
 		clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
 		child->ptrace |= PT_SINGLESTEP;
-- 
cgit v1.2.3-70-g09d2


From 82300bf479d7cdf87214b81ca5dc003bbc4f7e8f Mon Sep 17 00:00:00 2001
From: Chris Zankel <czankel@tensilica.com>
Date: Thu, 30 Jun 2005 02:58:58 -0700
Subject: [PATCH] xtensa: Added mm/Kconfig to get a flat memory layout

Added 'mm/Kconfig' to the xtensa Kconfig file to get a flat memory layout.
Fixed a typo in one of the help texts (thanks Geert for pointing it out)

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/Kconfig | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig
index 3e89767cea7..c9b5d298e3c 100644
--- a/arch/xtensa/Kconfig
+++ b/arch/xtensa/Kconfig
@@ -132,7 +132,7 @@ config XTENSA_CPU_CLOCK
 config GENERIC_CALIBRATE_DELAY
 	bool "Auto calibration of the BogoMIPS value"
 	---help---
-	  The BogoMIPS value can easily derived from the CPU frequency.
+	  The BogoMIPS value can easily be derived from the CPU frequency.
 
 config CMDLINE_BOOL
 	bool "Default bootloader kernel arguments"
@@ -158,6 +158,8 @@ config XTENSA_ISS_NETWORK
 	depends on XTENSA_PLATFORM_ISS
 	default y
 
+source "mm/Kconfig"
+
 endmenu
 
 menu "Bus options"
-- 
cgit v1.2.3-70-g09d2


From e7d163f7666560c90b163907b9d96ec6207e0f6f Mon Sep 17 00:00:00 2001
From: Chris Zankel <czankel@tensilica.com>
Date: Thu, 30 Jun 2005 02:58:59 -0700
Subject: [PATCH] xtensa: Removed local copy of zlib and fixed O= support

Removed an unnecessary local copy of zlib (sorry for the add'l traffic).
Fixed 'O=' support (thanks to Jan Dittmer for pointing it out).  Some minor
clean-ups in the make files.

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/Makefile                   |   48 +-
 arch/xtensa/boot/Makefile              |   10 +-
 arch/xtensa/boot/boot-elf/Makefile     |    4 +-
 arch/xtensa/boot/boot-redboot/Makefile |   10 +-
 arch/xtensa/boot/include/zlib.h        |  433 -------
 arch/xtensa/boot/lib/Makefile          |   13 +-
 arch/xtensa/boot/lib/memcpy.S          |   36 -
 arch/xtensa/boot/lib/zlib.c            | 2150 --------------------------------
 arch/xtensa/boot/lib/zmem.c            |   20 +-
 9 files changed, 53 insertions(+), 2671 deletions(-)
 delete mode 100644 arch/xtensa/boot/include/zlib.h
 delete mode 100644 arch/xtensa/boot/lib/memcpy.S
 delete mode 100644 arch/xtensa/boot/lib/zlib.c

(limited to 'arch')

diff --git a/arch/xtensa/Makefile b/arch/xtensa/Makefile
index 4fa27453b1f..27847e4ffcb 100644
--- a/arch/xtensa/Makefile
+++ b/arch/xtensa/Makefile
@@ -21,23 +21,17 @@ export CPU
 
 # Platform configuration
 
-platform-y					:= common
 platform-$(CONFIG_XTENSA_PLATFORM_XT2000)	:= xt2000
 platform-$(CONFIG_XTENSA_PLATFORM_ISS)		:= iss
 
 PLATFORM = $(platform-y)
 export PLATFORM
 
-#LDFLAGS_vmlinux		:= -T$(word 1,$(LINKSCRIPT))
-AFLAGS_vmlinux.lds.o	:= -Uxtensa
-CPPFLAGS		+= -Iarch/xtensa -Iinclude/asm  -mlongcalls  -g
-AFLAGS			+= -Iarch/xtensa -Iinclude/asm
-CPP			= $(CC) -E $(CFLAGS)
+CPPFLAGS	+= $(if $(KBUILD_SRC),-I$(srctree)/include/asm-xtensa/)
+CPPFLAGS	+= -Iinclude/asm
+CFLAGS		+= -pipe -mlongcalls
 
-cflags-y	+= -Iarch/xtensa -pipe -mlongcalls
-
-
-KBUILD_DEFCONFIG := common_defconfig
+KBUILD_DEFCONFIG := iss_defconfig
 
 # ramdisk/initrd support
 # You need a compressed ramdisk image, named ramdisk.gz in
@@ -62,30 +56,36 @@ endif
 LIBGCC := $(shell $(CC) $(CFLAGS) -print-libgcc-file-name)
 
 head-y		:= arch/xtensa/kernel/head.o
-core-y		+= arch/xtensa/kernel/ \
-		   arch/xtensa/mm/ arch/xtensa/platform-$(PLATFORM)/
+core-y		+= arch/xtensa/kernel/ arch/xtensa/mm/
+ifneq ($(PLATFORM),)
+core-y		+= arch/xtensa/platform-$(PLATFORM)/
+endif
 libs-y		+= arch/xtensa/lib/ $(LIBGCC)
 
-boot	:= arch/xtensa/boot
+boot		:= arch/xtensa/boot
+
+archinc		:= include/asm-xtensa
 
 arch/xtensa/kernel/asm-offsets.s: \
-	arch/xtensa/kernel/asm-offsets.c \
-	include/asm-xtensa/.platform
+	arch/xtensa/kernel/asm-offsets.c $(archinc)/.platform
 
 include/asm-xtensa/offsets.h: arch/xtensa/kernel/asm-offsets.s
 	$(call filechk,gen-asm-offsets)
 
-prepare: include/asm-xtensa/.platform include/asm-xtensa/offsets.h
+prepare: $(archinc)/.platform $(archinc)/offsets.h
 
 # Update machine cpu and platform symlinks if something which affects
 # them changed.
 
-include/asm-xtensa/.platform: $(wildcard include/config/arch/*.h)
-	@echo '  Setting up cpu ($(CPU)) and platform ($(PLATFORM)) symlinks'
-	$(Q)rm -f include/asm-xtensa/platform
-	$(Q)rm -f include/asm-xtensa/xtensa/config
-	$(Q)(cd include/asm-xtensa/; ln -sf platform-$(PLATFORM) platform)
-	$(Q)(cd include/asm-xtensa/xtensa; ln -sf config-$(CPU) config)
+$(archinc)/.platform: $(wildcard include/config/arch/*.h) include/config/MARKER
+	@echo '  SYMLINK $(archinc)/xtensa/config -> $(archinc)/xtensa/config-$(CPU)'
+	$(Q)mkdir -p $(archinc)
+	$(Q)mkdir -p $(archinc)/xtensa
+	$(Q)ln -fsn $(srctree)/$(archinc)/xtensa/config-$(CPU) $(archinc)/xtensa/config
+	@echo '  SYMLINK $(archinc)/platform -> $(archinc)/platform-$(PLATFORM)'
+	$(Q)ln -fsn $(srctree)/$(archinc)/platform-$(PLATFORM) $(archinc)/platform
+	@touch $@
+
 
 all: zImage
 
@@ -94,7 +94,9 @@ bzImage : zImage
 zImage zImage.initrd: vmlinux
 	$(Q)$(MAKE) $(build)=$(boot) $@
 
-CLEAN_FILES	+= arch/xtensa/vmlinux.lds include/asm-xtensa/offset.h
+CLEAN_FILES	+= arch/xtensa/vmlinux.lds $(archinc)/offset.h \
+		   $(archinc)/platform $(archinc)/xtensa/config \
+		   $(archinc)/.platform
 
 define archhelp
   @echo '* zImage      - Compressed kernel image (arch/xtensa/boot/images/zImage.*)'
diff --git a/arch/xtensa/boot/Makefile b/arch/xtensa/boot/Makefile
index 260f456ccf0..820b31d10ae 100644
--- a/arch/xtensa/boot/Makefile
+++ b/arch/xtensa/boot/Makefile
@@ -11,21 +11,19 @@
 CFLAGS 		+= -fno-builtin -Iarch/$(ARCH)/boot/include
 HOSTFLAGS	+= -Iarch/$(ARCH)/boot/include
 
-BIG_ENDIAN	:= $(shell echo -e "\#ifdef __XTENSA_EL__\nint little;\n\#else\nint big;\n\#endif" | $(CC) -E -|grep -c big)
-
+BIG_ENDIAN	:= $(shell echo -e __XTENSA_EB__ | $(CC) -E - | grep -v "\#")
 
 export CFLAGS
 export AFLAGS
 export BIG_ENDIAN
 
+subdir-y	:= lib
+
 # Subdirs for the boot loader(s)
 
 bootdir-$(CONFIG_XTENSA_PLATFORM_ISS)	 += boot-elf
 bootdir-$(CONFIG_XTENSA_PLATFORM_XT2000) += boot-redboot boot-elf
 
-subdir-y	:= lib/
-
-subdir-y	+= boot-elf/ boot-redboot/
 
 zImage zImage.initrd Image Image.initrd: $(bootdir-y)
 
@@ -33,5 +31,3 @@ $(bootdir-y): $(addprefix $(obj)/,$(subdir-y)) \
 	      $(addprefix $(obj)/,$(host-progs))
 	$(Q)$(MAKE) $(build)=$(obj)/$@ $(MAKECMDGOALS)
 
-
-
diff --git a/arch/xtensa/boot/boot-elf/Makefile b/arch/xtensa/boot/boot-elf/Makefile
index f6ef6a36966..734db7f7658 100644
--- a/arch/xtensa/boot/boot-elf/Makefile
+++ b/arch/xtensa/boot/boot-elf/Makefile
@@ -27,7 +27,7 @@ Image:	vmlinux $(OBJS)
 		--set-section-flags image=contents,alloc,load,load,data \
 		$(OBJS) $@.tmp
 	$(LD) $(LDFLAGS) $(LDFLAGS_vmlinux) \
-		-T arch/$(ARCH)/boot/boot-elf/boot.ld \
+		-T $(srctree)/arch/$(ARCH)/boot/boot-elf/boot.ld \
 		-o arch/$(ARCH)/boot/$@.elf $@.tmp
 	rm -f $@.tmp vmlinux.tmp
 
@@ -41,7 +41,7 @@ Image.initrd:	vmlinux $(OBJS)
 		--set-section-flags image=contents,alloc,load,load,data \
 		$(OBJS) $@.tmp
 	$(LD) $(LDFLAGS) $(LDFLAGS_vmlinux) \
-		-T arch/$(ARCH)/boot/boot-elf/boot.ld \
+		-T $(srctree)/arch/$(ARCH)/boot/boot-elf/boot.ld \
 		-o arch/$(ARCH)/boot/$@.elf $@.tmp
 	rm -f $@.tmp vmlinux.tmp
 
diff --git a/arch/xtensa/boot/boot-redboot/Makefile b/arch/xtensa/boot/boot-redboot/Makefile
index ca8a68bc847..f53262c2e1f 100644
--- a/arch/xtensa/boot/boot-redboot/Makefile
+++ b/arch/xtensa/boot/boot-redboot/Makefile
@@ -12,24 +12,24 @@ else
 OBJCOPY_ARGS 	:= -O elf32-xtensa-le
 endif
 
-LD_ARGS	= -T $(obj)/boot.ld
+LD_ARGS	= -T $(srctree)/$(obj)/boot.ld
 
 boot-y	:= bootstrap.o
 
 OBJS	:= $(addprefix $(obj)/,$(boot-y))
-LIBS	:= arch/$(ARCH)/boot/lib/lib.a arch/$(ARCH)/lib/lib.a
+LIBS	:= arch/xtensa/boot/lib/lib.a arch/xtensa/lib/lib.a
 
 LIBGCC := $(shell $(CC) $(CFLAGS) -print-libgcc-file-name)
 
 zImage: vmlinux $(OBJS) $(LIBS)
 	$(OBJCOPY) --strip-all -R .comment -R .xt.insn -O binary \
-		$(TOPDIR)/vmlinux vmlinux.tmp
+		vmlinux vmlinux.tmp
 	gzip -vf9 vmlinux.tmp
 	$(OBJCOPY) $(OBJCOPY_ARGS) -R .comment \
 		--add-section image=vmlinux.tmp.gz \
 		--set-section-flags image=contents,alloc,load,load,data \
 		$(OBJS) $@.tmp
 	$(LD) $(LD_ARGS) -o $@.elf $@.tmp $(LIBS) -L/xtensa-elf/lib $(LIBGCC)
-	$(OBJCOPY) -S -O binary $@.elf arch/$(ARCH)/boot/images/$@.redboot
-#	rm -f $@.tmp $@.elf vmlinux.tmp.gz
+	$(OBJCOPY) -S -O binary $@.elf arch/$(ARCH)/boot/$@.redboot
+	rm -f $@.tmp $@.elf vmlinux.tmp.gz
 
diff --git a/arch/xtensa/boot/include/zlib.h b/arch/xtensa/boot/include/zlib.h
deleted file mode 100644
index ea29b623785..00000000000
--- a/arch/xtensa/boot/include/zlib.h
+++ /dev/null
@@ -1,433 +0,0 @@
-/*
- * BK Id: SCCS/s.zlib.h 1.8 05/18/01 15:17:23 cort
- */
-/*
- * This file is derived from zlib.h and zconf.h from the zlib-0.95
- * distribution by Jean-loup Gailly and Mark Adler, with some additions
- * by Paul Mackerras to aid in implementing Deflate compression and
- * decompression for PPP packets.
- */
-
-/*
- *  ==FILEVERSION 960122==
- *
- * This marker is used by the Linux installation script to determine
- * whether an up-to-date version of this file is already installed.
- */
-
-/* zlib.h -- interface of the 'zlib' general purpose compression library
-  version 0.95, Aug 16th, 1995.
-
-  Copyright (C) 1995 Jean-loup Gailly and Mark Adler
-
-  This software is provided 'as-is', without any express or implied
-  warranty.  In no event will the authors be held liable for any damages
-  arising from the use of this software.
-
-  Permission is granted to anyone to use this software for any purpose,
-  including commercial applications, and to alter it and redistribute it
-  freely, subject to the following restrictions:
-
-  1. The origin of this software must not be misrepresented; you must not
-     claim that you wrote the original software. If you use this software
-     in a product, an acknowledgment in the product documentation would be
-     appreciated but is not required.
-  2. Altered source versions must be plainly marked as such, and must not be
-     misrepresented as being the original software.
-  3. This notice may not be removed or altered from any source distribution.
-
-  Jean-loup Gailly        Mark Adler
-  gzip@prep.ai.mit.edu    madler@alumni.caltech.edu
- */
-
-#ifndef _ZLIB_H
-#define _ZLIB_H
-
-/* #include "zconf.h" */	/* included directly here */
-
-/* zconf.h -- configuration of the zlib compression library
- * Copyright (C) 1995 Jean-loup Gailly.
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* From: zconf.h,v 1.12 1995/05/03 17:27:12 jloup Exp */
-
-/*
-     The library does not install any signal handler. It is recommended to
-  add at least a handler for SIGSEGV when decompressing; the library checks
-  the consistency of the input data whenever possible but may go nuts
-  for some forms of corrupted input.
- */
-
-/*
- * Compile with -DMAXSEG_64K if the alloc function cannot allocate more
- * than 64k bytes at a time (needed on systems with 16-bit int).
- * Compile with -DUNALIGNED_OK if it is OK to access shorts or ints
- * at addresses which are not a multiple of their size.
- * Under DOS, -DFAR=far or -DFAR=__far may be needed.
- */
-
-#ifndef STDC
-#  if defined(MSDOS) || defined(__STDC__) || defined(__cplusplus)
-#    define STDC
-#  endif
-#endif
-
-#ifdef	__MWERKS__ /* Metrowerks CodeWarrior declares fileno() in unix.h */
-#  include <unix.h>
-#endif
-
-/* Maximum value for memLevel in deflateInit2 */
-#ifndef MAX_MEM_LEVEL
-#  ifdef MAXSEG_64K
-#    define MAX_MEM_LEVEL 8
-#  else
-#    define MAX_MEM_LEVEL 9
-#  endif
-#endif
-
-#ifndef FAR
-#  define FAR
-#endif
-
-/* Maximum value for windowBits in deflateInit2 and inflateInit2 */
-#ifndef MAX_WBITS
-#  define MAX_WBITS   15 /* 32K LZ77 window */
-#endif
-
-/* The memory requirements for deflate are (in bytes):
-            1 << (windowBits+2)   +  1 << (memLevel+9)
- that is: 128K for windowBits=15  +  128K for memLevel = 8  (default values)
- plus a few kilobytes for small objects. For example, if you want to reduce
- the default memory requirements from 256K to 128K, compile with
-     make CFLAGS="-O -DMAX_WBITS=14 -DMAX_MEM_LEVEL=7"
- Of course this will generally degrade compression (there's no free lunch).
-
-   The memory requirements for inflate are (in bytes) 1 << windowBits
- that is, 32K for windowBits=15 (default value) plus a few kilobytes
- for small objects.
-*/
-
-                        /* Type declarations */
-
-#ifndef OF /* function prototypes */
-#  ifdef STDC
-#    define OF(args)  args
-#  else
-#    define OF(args)  ()
-#  endif
-#endif
-
-typedef unsigned char  Byte;  /* 8 bits */
-typedef unsigned int   uInt;  /* 16 bits or more */
-typedef unsigned long  uLong; /* 32 bits or more */
-
-typedef Byte FAR Bytef;
-typedef char FAR charf;
-typedef int FAR intf;
-typedef uInt FAR uIntf;
-typedef uLong FAR uLongf;
-
-#ifdef STDC
-   typedef void FAR *voidpf;
-   typedef void     *voidp;
-#else
-   typedef Byte FAR *voidpf;
-   typedef Byte     *voidp;
-#endif
-
-/* end of original zconf.h */
-
-#define ZLIB_VERSION "0.95P"
-
-/*
-     The 'zlib' compression library provides in-memory compression and
-  decompression functions, including integrity checks of the uncompressed
-  data.  This version of the library supports only one compression method
-  (deflation) but other algorithms may be added later and will have the same
-  stream interface.
-
-     For compression the application must provide the output buffer and
-  may optionally provide the input buffer for optimization. For decompression,
-  the application must provide the input buffer and may optionally provide
-  the output buffer for optimization.
-
-     Compression can be done in a single step if the buffers are large
-  enough (for example if an input file is mmap'ed), or can be done by
-  repeated calls of the compression function.  In the latter case, the
-  application must provide more input and/or consume the output
-  (providing more output space) before each call.
-*/
-
-typedef voidpf (*alloc_func) OF((voidpf opaque, uInt items, uInt size));
-typedef void   (*free_func)  OF((voidpf opaque, voidpf address, uInt nbytes));
-
-struct internal_state;
-
-typedef struct z_stream_s {
-    Bytef    *next_in;  /* next input byte */
-    uInt     avail_in;  /* number of bytes available at next_in */
-    uLong    total_in;  /* total nb of input bytes read so far */
-
-    Bytef    *next_out; /* next output byte should be put there */
-    uInt     avail_out; /* remaining free space at next_out */
-    uLong    total_out; /* total nb of bytes output so far */
-
-    char     *msg;      /* last error message, NULL if no error */
-    struct internal_state FAR *state; /* not visible by applications */
-
-    alloc_func zalloc;  /* used to allocate the internal state */
-    free_func  zfree;   /* used to free the internal state */
-    voidp      opaque;  /* private data object passed to zalloc and zfree */
-
-    Byte     data_type; /* best guess about the data type: ascii or binary */
-
-} z_stream;
-
-/*
-   The application must update next_in and avail_in when avail_in has
-   dropped to zero. It must update next_out and avail_out when avail_out
-   has dropped to zero. The application must initialize zalloc, zfree and
-   opaque before calling the init function. All other fields are set by the
-   compression library and must not be updated by the application.
-
-   The opaque value provided by the application will be passed as the first
-   parameter for calls of zalloc and zfree. This can be useful for custom
-   memory management. The compression library attaches no meaning to the
-   opaque value.
-
-   zalloc must return Z_NULL if there is not enough memory for the object.
-   On 16-bit systems, the functions zalloc and zfree must be able to allocate
-   exactly 65536 bytes, but will not be required to allocate more than this
-   if the symbol MAXSEG_64K is defined (see zconf.h). WARNING: On MSDOS,
-   pointers returned by zalloc for objects of exactly 65536 bytes *must*
-   have their offset normalized to zero. The default allocation function
-   provided by this library ensures this (see zutil.c). To reduce memory
-   requirements and avoid any allocation of 64K objects, at the expense of
-   compression ratio, compile the library with -DMAX_WBITS=14 (see zconf.h).
-
-   The fields total_in and total_out can be used for statistics or
-   progress reports. After compression, total_in holds the total size of
-   the uncompressed data and may be saved for use in the decompressor
-   (particularly if the decompressor wants to decompress everything in
-   a single step).
-*/
-
-                        /* constants */
-
-#define Z_NO_FLUSH      0
-#define Z_PARTIAL_FLUSH 1
-#define Z_FULL_FLUSH    2
-#define Z_SYNC_FLUSH    3 /* experimental: partial_flush + byte align */
-#define Z_FINISH        4
-#define Z_PACKET_FLUSH	5
-/* See deflate() below for the usage of these constants */
-
-#define Z_OK            0
-#define Z_STREAM_END    1
-#define Z_ERRNO        (-1)
-#define Z_STREAM_ERROR (-2)
-#define Z_DATA_ERROR   (-3)
-#define Z_MEM_ERROR    (-4)
-#define Z_BUF_ERROR    (-5)
-/* error codes for the compression/decompression functions */
-
-#define Z_BEST_SPEED             1
-#define Z_BEST_COMPRESSION       9
-#define Z_DEFAULT_COMPRESSION  (-1)
-/* compression levels */
-
-#define Z_FILTERED            1
-#define Z_HUFFMAN_ONLY        2
-#define Z_DEFAULT_STRATEGY    0
-
-#define Z_BINARY   0
-#define Z_ASCII    1
-#define Z_UNKNOWN  2
-/* Used to set the data_type field */
-
-#define Z_NULL  0  /* for initializing zalloc, zfree, opaque */
-
-extern char *zlib_version;
-/* The application can compare zlib_version and ZLIB_VERSION for consistency.
-   If the first character differs, the library code actually used is
-   not compatible with the zlib.h header file used by the application.
- */
-
-                        /* basic functions */
-
-extern int inflateInit OF((z_stream *strm));
-/*
-     Initializes the internal stream state for decompression. The fields
-   zalloc and zfree must be initialized before by the caller.  If zalloc and
-   zfree are set to Z_NULL, inflateInit updates them to use default allocation
-   functions.
-
-     inflateInit returns Z_OK if success, Z_MEM_ERROR if there was not
-   enough memory.  msg is set to null if there is no error message.
-   inflateInit does not perform any decompression: this will be done by
-   inflate().
-*/
-
-
-extern int inflate OF((z_stream *strm, int flush));
-/*
-  Performs one or both of the following actions:
-
-  - Decompress more input starting at next_in and update next_in and avail_in
-    accordingly. If not all input can be processed (because there is not
-    enough room in the output buffer), next_in is updated and processing
-    will resume at this point for the next call of inflate().
-
-  - Provide more output starting at next_out and update next_out and avail_out
-    accordingly.  inflate() always provides as much output as possible
-    (until there is no more input data or no more space in the output buffer).
-
-  Before the call of inflate(), the application should ensure that at least
-  one of the actions is possible, by providing more input and/or consuming
-  more output, and updating the next_* and avail_* values accordingly.
-  The application can consume the uncompressed output when it wants, for
-  example when the output buffer is full (avail_out == 0), or after each
-  call of inflate().
-
-    If the parameter flush is set to Z_PARTIAL_FLUSH or Z_PACKET_FLUSH,
-  inflate flushes as much output as possible to the output buffer. The
-  flushing behavior of inflate is not specified for values of the flush
-  parameter other than Z_PARTIAL_FLUSH, Z_PACKET_FLUSH or Z_FINISH, but the
-  current implementation actually flushes as much output as possible
-  anyway.  For Z_PACKET_FLUSH, inflate checks that once all the input data
-  has been consumed, it is expecting to see the length field of a stored
-  block; if not, it returns Z_DATA_ERROR.
-
-    inflate() should normally be called until it returns Z_STREAM_END or an
-  error. However if all decompression is to be performed in a single step
-  (a single call of inflate), the parameter flush should be set to
-  Z_FINISH. In this case all pending input is processed and all pending
-  output is flushed; avail_out must be large enough to hold all the
-  uncompressed data. (The size of the uncompressed data may have been saved
-  by the compressor for this purpose.) The next operation on this stream must
-  be inflateEnd to deallocate the decompression state. The use of Z_FINISH
-  is never required, but can be used to inform inflate that a faster routine
-  may be used for the single inflate() call.
-
-    inflate() returns Z_OK if some progress has been made (more input
-  processed or more output produced), Z_STREAM_END if the end of the
-  compressed data has been reached and all uncompressed output has been
-  produced, Z_DATA_ERROR if the input data was corrupted, Z_STREAM_ERROR if
-  the stream structure was inconsistent (for example if next_in or next_out
-  was NULL), Z_MEM_ERROR if there was not enough memory, Z_BUF_ERROR if no
-  progress is possible or if there was not enough room in the output buffer
-  when Z_FINISH is used. In the Z_DATA_ERROR case, the application may then
-  call inflateSync to look for a good compression block.  */
-
-
-extern int inflateEnd OF((z_stream *strm));
-/*
-     All dynamically allocated data structures for this stream are freed.
-   This function discards any unprocessed input and does not flush any
-   pending output.
-
-     inflateEnd returns Z_OK if success, Z_STREAM_ERROR if the stream state
-   was inconsistent. In the error case, msg may be set but then points to a
-   static string (which must not be deallocated).
-*/
-
-                        /* advanced functions */
-
-extern int inflateInit2 OF((z_stream *strm,
-                            int  windowBits));
-/*
-     This is another version of inflateInit with more compression options. The
-   fields next_out, zalloc and zfree must be initialized before by the caller.
-
-     The windowBits parameter is the base two logarithm of the maximum window
-   size (the size of the history buffer).  It should be in the range 8..15 for
-   this version of the library (the value 16 will be allowed soon). The
-   default value is 15 if inflateInit is used instead. If a compressed stream
-   with a larger window size is given as input, inflate() will return with
-   the error code Z_DATA_ERROR instead of trying to allocate a larger window.
-
-     If next_out is not null, the library will use this buffer for the history
-   buffer; the buffer must either be large enough to hold the entire output
-   data, or have at least 1<<windowBits bytes.  If next_out is null, the
-   library will allocate its own buffer (and leave next_out null). next_in
-   need not be provided here but must be provided by the application for the
-   next call of inflate().
-
-     If the history buffer is provided by the application, next_out must
-   never be changed by the application since the decompressor maintains
-   history information inside this buffer from call to call; the application
-   can only reset next_out to the beginning of the history buffer when
-   avail_out is zero and all output has been consumed.
-
-      inflateInit2 returns Z_OK if success, Z_MEM_ERROR if there was
-   not enough memory, Z_STREAM_ERROR if a parameter is invalid (such as
-   windowBits < 8). msg is set to null if there is no error message.
-   inflateInit2 does not perform any decompression: this will be done by
-   inflate().
-*/
-
-extern int inflateSync OF((z_stream *strm));
-/*
-    Skips invalid compressed data until the special marker (see deflate()
-  above) can be found, or until all available input is skipped. No output
-  is provided.
-
-    inflateSync returns Z_OK if the special marker has been found, Z_BUF_ERROR
-  if no more input was provided, Z_DATA_ERROR if no marker has been found,
-  or Z_STREAM_ERROR if the stream structure was inconsistent. In the success
-  case, the application may save the current current value of total_in which
-  indicates where valid compressed data was found. In the error case, the
-  application may repeatedly call inflateSync, providing more input each time,
-  until success or end of the input data.
-*/
-
-extern int inflateReset OF((z_stream *strm));
-/*
-     This function is equivalent to inflateEnd followed by inflateInit,
-   but does not free and reallocate all the internal decompression state.
-   The stream will keep attributes that may have been set by inflateInit2.
-
-      inflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source
-   stream state was inconsistent (such as zalloc or state being NULL).
-*/
-
-extern int inflateIncomp OF((z_stream *strm));
-/*
-     This function adds the data at next_in (avail_in bytes) to the output
-   history without performing any output.  There must be no pending output,
-   and the decompressor must be expecting to see the start of a block.
-   Calling this function is equivalent to decompressing a stored block
-   containing the data at next_in (except that the data is not output).
-*/
-
-                        /* checksum functions */
-
-/*
-     This function is not related to compression but is exported
-   anyway because it might be useful in applications using the
-   compression library.
-*/
-
-extern uLong adler32 OF((uLong adler, Bytef *buf, uInt len));
-
-/*
-     Update a running Adler-32 checksum with the bytes buf[0..len-1] and
-   return the updated checksum. If buf is NULL, this function returns
-   the required initial value for the checksum.
-   An Adler-32 checksum is almost as reliable as a CRC32 but can be computed
-   much faster. Usage example:
-
-     uLong adler = adler32(0L, Z_NULL, 0);
-
-     while (read_buffer(buffer, length) != EOF) {
-       adler = adler32(adler, buffer, length);
-     }
-     if (adler != original_adler) error();
-*/
-
-#ifndef _Z_UTIL_H
-    struct internal_state {int dummy;}; /* hack for buggy compilers */
-#endif
-
-#endif /* _ZLIB_H */
diff --git a/arch/xtensa/boot/lib/Makefile b/arch/xtensa/boot/lib/Makefile
index c0a74dc3a0d..9e73bb8aeb7 100644
--- a/arch/xtensa/boot/lib/Makefile
+++ b/arch/xtensa/boot/lib/Makefile
@@ -2,5 +2,16 @@
 # Makefile for some libs needed by zImage.
 #
 
+zlib	:= infblock.c infcodes.c inffast.c inflate.c inftrees.c infutil.c
 
-lib-y := zlib.o zmem.o
+lib-y	+= $(zlib:.c=.o) zmem.o
+
+EXTRA_CFLAGS	+= -Ilib/zlib_inflate
+
+quiet_cmd_copy_zlib = COPY    $@
+      cmd_copy_zlib = cat $< > $@
+
+$(addprefix $(obj)/,$(zlib)): $(obj)/%: $(srctree)/lib/zlib_inflate/%
+	$(call cmd,copy_zlib)
+
+clean-files	:= $(zlib)
diff --git a/arch/xtensa/boot/lib/memcpy.S b/arch/xtensa/boot/lib/memcpy.S
deleted file mode 100644
index a029f5df2d5..00000000000
--- a/arch/xtensa/boot/lib/memcpy.S
+++ /dev/null
@@ -1,36 +0,0 @@
-/*
- *  arch/xtensa/lib/memcpy.S
- *
- *  ANSI C standard library function memcpy
- *
- *  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.
- */
-
-#define _ASMLANGUAGE
-#include <xtensa/config/core.h>
-
-.text
-.align	4
-.global	bcopy
-.type   bcopy,@function
-bcopy:
-        movi    a14, xthal_bcopy        // a14 safe to use regardless of whether caller
-                                        //   used call4 or call8 (can't have used call12)
-        jx      a14                     // let the Core HAL do the work
-
-.text
-.align	4
-.global	memcpy
-.type   memcpy,@function
-memcpy:
-.global	memmove
-.type   memmove,@function
-memmove:
-        movi    a14, xthal_memcpy      // a14 safe to use regardless of whether caller
-                                        //   used call4 or call8 (can't have used call12)
-        jx      a14                     // let the Core HAL do the work
-
diff --git a/arch/xtensa/boot/lib/zlib.c b/arch/xtensa/boot/lib/zlib.c
deleted file mode 100644
index e3859f63107..00000000000
--- a/arch/xtensa/boot/lib/zlib.c
+++ /dev/null
@@ -1,2150 +0,0 @@
-/*
- * BK Id: SCCS/s.zlib.c 1.8 05/18/01 15:17:24 cort
- */
-/*
- * This file is derived from various .h and .c files from the zlib-0.95
- * distribution by Jean-loup Gailly and Mark Adler, with some additions
- * by Paul Mackerras to aid in implementing Deflate compression and
- * decompression for PPP packets.  See zlib.h for conditions of
- * distribution and use.
- *
- * Changes that have been made include:
- * - changed functions not used outside this file to "local"
- * - added minCompression parameter to deflateInit2
- * - added Z_PACKET_FLUSH (see zlib.h for details)
- * - added inflateIncomp
- *
- */
-
-/*+++++*/
-/* zutil.h -- internal interface and configuration of the compression library
- * Copyright (C) 1995 Jean-loup Gailly.
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* WARNING: this file should *not* be used by applications. It is
-   part of the implementation of the compression library and is
-   subject to change. Applications should only use zlib.h.
- */
-
-/* From: zutil.h,v 1.9 1995/05/03 17:27:12 jloup Exp */
-
-#define _Z_UTIL_H
-
-#include "zlib.h"
-
-#ifndef local
-#  define local static
-#endif
-/* compile with -Dlocal if your debugger can't find static symbols */
-
-#define FAR
-
-typedef unsigned char  uch;
-typedef uch FAR uchf;
-typedef unsigned short ush;
-typedef ush FAR ushf;
-typedef unsigned long  ulg;
-
-extern char *z_errmsg[]; /* indexed by 1-zlib_error */
-
-#define ERR_RETURN(strm,err) return (strm->msg=z_errmsg[1-err], err)
-/* To be used only when the state is known to be valid */
-
-#ifndef NULL
-#define NULL	((void *) 0)
-#endif
-
-        /* common constants */
-
-#define DEFLATED   8
-
-#ifndef DEF_WBITS
-#  define DEF_WBITS MAX_WBITS
-#endif
-/* default windowBits for decompression. MAX_WBITS is for compression only */
-
-#if MAX_MEM_LEVEL >= 8
-#  define DEF_MEM_LEVEL 8
-#else
-#  define DEF_MEM_LEVEL  MAX_MEM_LEVEL
-#endif
-/* default memLevel */
-
-#define STORED_BLOCK 0
-#define STATIC_TREES 1
-#define DYN_TREES    2
-/* The three kinds of block type */
-
-#define MIN_MATCH  3
-#define MAX_MATCH  258
-/* The minimum and maximum match lengths */
-
-         /* functions */
-
-#include <linux/string.h>
-#define zmemcpy memcpy
-#define zmemzero(dest, len)	memset(dest, 0, len)
-
-/* Diagnostic functions */
-#ifdef DEBUG_ZLIB
-#  include <stdio.h>
-#  ifndef verbose
-#    define verbose 0
-#  endif
-#  define Assert(cond,msg) {if(!(cond)) z_error(msg);}
-#  define Trace(x) fprintf x
-#  define Tracev(x) {if (verbose) fprintf x ;}
-#  define Tracevv(x) {if (verbose>1) fprintf x ;}
-#  define Tracec(c,x) {if (verbose && (c)) fprintf x ;}
-#  define Tracecv(c,x) {if (verbose>1 && (c)) fprintf x ;}
-#else
-#  define Assert(cond,msg)
-#  define Trace(x)
-#  define Tracev(x)
-#  define Tracevv(x)
-#  define Tracec(c,x)
-#  define Tracecv(c,x)
-#endif
-
-
-typedef uLong (*check_func) OF((uLong check, Bytef *buf, uInt len));
-
-/* voidpf zcalloc OF((voidpf opaque, unsigned items, unsigned size)); */
-/* void   zcfree  OF((voidpf opaque, voidpf ptr)); */
-
-#define ZALLOC(strm, items, size) \
-           (*((strm)->zalloc))((strm)->opaque, (items), (size))
-#define ZFREE(strm, addr, size)	\
-	   (*((strm)->zfree))((strm)->opaque, (voidpf)(addr), (size))
-#define TRY_FREE(s, p, n) {if (p) ZFREE(s, p, n);}
-
-/* deflate.h -- internal compression state
- * Copyright (C) 1995 Jean-loup Gailly
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* WARNING: this file should *not* be used by applications. It is
-   part of the implementation of the compression library and is
-   subject to change. Applications should only use zlib.h.
- */
-
-/*+++++*/
-/* infblock.h -- header to use infblock.c
- * Copyright (C) 1995 Mark Adler
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* WARNING: this file should *not* be used by applications. It is
-   part of the implementation of the compression library and is
-   subject to change. Applications should only use zlib.h.
- */
-
-struct inflate_blocks_state;
-typedef struct inflate_blocks_state FAR inflate_blocks_statef;
-
-local inflate_blocks_statef * inflate_blocks_new OF((
-    z_stream *z,
-    check_func c,               /* check function */
-    uInt w));                   /* window size */
-
-local int inflate_blocks OF((
-    inflate_blocks_statef *,
-    z_stream *,
-    int));                      /* initial return code */
-
-local void inflate_blocks_reset OF((
-    inflate_blocks_statef *,
-    z_stream *,
-    uLongf *));                  /* check value on output */
-
-local int inflate_blocks_free OF((
-    inflate_blocks_statef *,
-    z_stream *,
-    uLongf *));                  /* check value on output */
-
-local int inflate_addhistory OF((
-    inflate_blocks_statef *,
-    z_stream *));
-
-local int inflate_packet_flush OF((
-    inflate_blocks_statef *));
-
-/*+++++*/
-/* inftrees.h -- header to use inftrees.c
- * Copyright (C) 1995 Mark Adler
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* WARNING: this file should *not* be used by applications. It is
-   part of the implementation of the compression library and is
-   subject to change. Applications should only use zlib.h.
- */
-
-/* Huffman code lookup table entry--this entry is four bytes for machines
-   that have 16-bit pointers (e.g. PC's in the small or medium model). */
-
-typedef struct inflate_huft_s FAR inflate_huft;
-
-struct inflate_huft_s {
-  union {
-    struct {
-      Byte Exop;        /* number of extra bits or operation */
-      Byte Bits;        /* number of bits in this code or subcode */
-    } what;
-    uInt Nalloc;	/* number of these allocated here */
-    Bytef *pad;         /* pad structure to a power of 2 (4 bytes for */
-  } word;               /*  16-bit, 8 bytes for 32-bit machines) */
-  union {
-    uInt Base;          /* literal, length base, or distance base */
-    inflate_huft *Next; /* pointer to next level of table */
-  } more;
-};
-
-#ifdef DEBUG_ZLIB
-  local uInt inflate_hufts;
-#endif
-
-local int inflate_trees_bits OF((
-    uIntf *,                    /* 19 code lengths */
-    uIntf *,                    /* bits tree desired/actual depth */
-    inflate_huft * FAR *,       /* bits tree result */
-    z_stream *));               /* for zalloc, zfree functions */
-
-local int inflate_trees_dynamic OF((
-    uInt,                       /* number of literal/length codes */
-    uInt,                       /* number of distance codes */
-    uIntf *,                    /* that many (total) code lengths */
-    uIntf *,                    /* literal desired/actual bit depth */
-    uIntf *,                    /* distance desired/actual bit depth */
-    inflate_huft * FAR *,       /* literal/length tree result */
-    inflate_huft * FAR *,       /* distance tree result */
-    z_stream *));               /* for zalloc, zfree functions */
-
-local int inflate_trees_fixed OF((
-    uIntf *,                    /* literal desired/actual bit depth */
-    uIntf *,                    /* distance desired/actual bit depth */
-    inflate_huft * FAR *,       /* literal/length tree result */
-    inflate_huft * FAR *));     /* distance tree result */
-
-local int inflate_trees_free OF((
-    inflate_huft *,             /* tables to free */
-    z_stream *));               /* for zfree function */
-
-
-/*+++++*/
-/* infcodes.h -- header to use infcodes.c
- * Copyright (C) 1995 Mark Adler
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* WARNING: this file should *not* be used by applications. It is
-   part of the implementation of the compression library and is
-   subject to change. Applications should only use zlib.h.
- */
-
-struct inflate_codes_state;
-typedef struct inflate_codes_state FAR inflate_codes_statef;
-
-local inflate_codes_statef *inflate_codes_new OF((
-    uInt, uInt,
-    inflate_huft *, inflate_huft *,
-    z_stream *));
-
-local int inflate_codes OF((
-    inflate_blocks_statef *,
-    z_stream *,
-    int));
-
-local void inflate_codes_free OF((
-    inflate_codes_statef *,
-    z_stream *));
-
-
-/*+++++*/
-/* inflate.c -- zlib interface to inflate modules
- * Copyright (C) 1995 Mark Adler
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* inflate private state */
-struct internal_state {
-
-  /* mode */
-  enum {
-      METHOD,   /* waiting for method byte */
-      FLAG,     /* waiting for flag byte */
-      BLOCKS,   /* decompressing blocks */
-      CHECK4,   /* four check bytes to go */
-      CHECK3,   /* three check bytes to go */
-      CHECK2,   /* two check bytes to go */
-      CHECK1,   /* one check byte to go */
-      DONE,     /* finished check, done */
-      BAD}      /* got an error--stay here */
-    mode;               /* current inflate mode */
-
-  /* mode dependent information */
-  union {
-    uInt method;        /* if FLAGS, method byte */
-    struct {
-      uLong was;                /* computed check value */
-      uLong need;               /* stream check value */
-    } check;            /* if CHECK, check values to compare */
-    uInt marker;        /* if BAD, inflateSync's marker bytes count */
-  } sub;        /* submode */
-
-  /* mode independent information */
-  int  nowrap;          /* flag for no wrapper */
-  uInt wbits;           /* log2(window size)  (8..15, defaults to 15) */
-  inflate_blocks_statef
-    *blocks;            /* current inflate_blocks state */
-
-};
-
-
-int inflateReset(z)
-z_stream *z;
-{
-  uLong c;
-
-  if (z == Z_NULL || z->state == Z_NULL)
-    return Z_STREAM_ERROR;
-  z->total_in = z->total_out = 0;
-  z->msg = Z_NULL;
-  z->state->mode = z->state->nowrap ? BLOCKS : METHOD;
-  inflate_blocks_reset(z->state->blocks, z, &c);
-  Trace((stderr, "inflate: reset\n"));
-  return Z_OK;
-}
-
-
-int inflateEnd(z)
-z_stream *z;
-{
-  uLong c;
-
-  if (z == Z_NULL || z->state == Z_NULL || z->zfree == Z_NULL)
-    return Z_STREAM_ERROR;
-  if (z->state->blocks != Z_NULL)
-    inflate_blocks_free(z->state->blocks, z, &c);
-  ZFREE(z, z->state, sizeof(struct internal_state));
-  z->state = Z_NULL;
-  Trace((stderr, "inflate: end\n"));
-  return Z_OK;
-}
-
-
-int inflateInit2(z, w)
-z_stream *z;
-int w;
-{
-  /* initialize state */
-  if (z == Z_NULL)
-    return Z_STREAM_ERROR;
-/*  if (z->zalloc == Z_NULL) z->zalloc = zcalloc; */
-/*  if (z->zfree == Z_NULL) z->zfree = zcfree; */
-  if ((z->state = (struct internal_state FAR *)
-       ZALLOC(z,1,sizeof(struct internal_state))) == Z_NULL)
-    return Z_MEM_ERROR;
-  z->state->blocks = Z_NULL;
-
-  /* handle undocumented nowrap option (no zlib header or check) */
-  z->state->nowrap = 0;
-  if (w < 0)
-  {
-    w = - w;
-    z->state->nowrap = 1;
-  }
-
-  /* set window size */
-  if (w < 8 || w > 15)
-  {
-    inflateEnd(z);
-    return Z_STREAM_ERROR;
-  }
-  z->state->wbits = (uInt)w;
-
-  /* create inflate_blocks state */
-  if ((z->state->blocks =
-       inflate_blocks_new(z, z->state->nowrap ? Z_NULL : adler32, 1 << w))
-      == Z_NULL)
-  {
-    inflateEnd(z);
-    return Z_MEM_ERROR;
-  }
-  Trace((stderr, "inflate: allocated\n"));
-
-  /* reset state */
-  inflateReset(z);
-  return Z_OK;
-}
-
-
-int inflateInit(z)
-z_stream *z;
-{
-  return inflateInit2(z, DEF_WBITS);
-}
-
-
-#define NEEDBYTE {if(z->avail_in==0)goto empty;r=Z_OK;}
-#define NEXTBYTE (z->avail_in--,z->total_in++,*z->next_in++)
-
-int inflate(z, f)
-z_stream *z;
-int f;
-{
-  int r;
-  uInt b;
-
-  if (z == Z_NULL || z->next_in == Z_NULL)
-    return Z_STREAM_ERROR;
-  r = Z_BUF_ERROR;
-  while (1) switch (z->state->mode)
-  {
-    case METHOD:
-      NEEDBYTE
-      if (((z->state->sub.method = NEXTBYTE) & 0xf) != DEFLATED)
-      {
-        z->state->mode = BAD;
-        z->msg = "unknown compression method";
-        z->state->sub.marker = 5;       /* can't try inflateSync */
-        break;
-      }
-      if ((z->state->sub.method >> 4) + 8 > z->state->wbits)
-      {
-        z->state->mode = BAD;
-        z->msg = "invalid window size";
-        z->state->sub.marker = 5;       /* can't try inflateSync */
-        break;
-      }
-      z->state->mode = FLAG;
-    case FLAG:
-      NEEDBYTE
-      if ((b = NEXTBYTE) & 0x20)
-      {
-        z->state->mode = BAD;
-        z->msg = "invalid reserved bit";
-        z->state->sub.marker = 5;       /* can't try inflateSync */
-        break;
-      }
-      if (((z->state->sub.method << 8) + b) % 31)
-      {
-        z->state->mode = BAD;
-        z->msg = "incorrect header check";
-        z->state->sub.marker = 5;       /* can't try inflateSync */
-        break;
-      }
-      Trace((stderr, "inflate: zlib header ok\n"));
-      z->state->mode = BLOCKS;
-    case BLOCKS:
-      r = inflate_blocks(z->state->blocks, z, r);
-      if (f == Z_PACKET_FLUSH && z->avail_in == 0 && z->avail_out != 0)
-	  r = inflate_packet_flush(z->state->blocks);
-      if (r == Z_DATA_ERROR)
-      {
-        z->state->mode = BAD;
-        z->state->sub.marker = 0;       /* can try inflateSync */
-        break;
-      }
-      if (r != Z_STREAM_END)
-        return r;
-      r = Z_OK;
-      inflate_blocks_reset(z->state->blocks, z, &z->state->sub.check.was);
-      if (z->state->nowrap)
-      {
-        z->state->mode = DONE;
-        break;
-      }
-      z->state->mode = CHECK4;
-    case CHECK4:
-      NEEDBYTE
-      z->state->sub.check.need = (uLong)NEXTBYTE << 24;
-      z->state->mode = CHECK3;
-    case CHECK3:
-      NEEDBYTE
-      z->state->sub.check.need += (uLong)NEXTBYTE << 16;
-      z->state->mode = CHECK2;
-    case CHECK2:
-      NEEDBYTE
-      z->state->sub.check.need += (uLong)NEXTBYTE << 8;
-      z->state->mode = CHECK1;
-    case CHECK1:
-      NEEDBYTE
-      z->state->sub.check.need += (uLong)NEXTBYTE;
-
-      if (z->state->sub.check.was != z->state->sub.check.need)
-      {
-        z->state->mode = BAD;
-        z->msg = "incorrect data check";
-        z->state->sub.marker = 5;       /* can't try inflateSync */
-        break;
-      }
-      Trace((stderr, "inflate: zlib check ok\n"));
-      z->state->mode = DONE;
-    case DONE:
-      return Z_STREAM_END;
-    case BAD:
-      return Z_DATA_ERROR;
-    default:
-      return Z_STREAM_ERROR;
-  }
-
- empty:
-  if (f != Z_PACKET_FLUSH)
-    return r;
-  z->state->mode = BAD;
-  z->state->sub.marker = 0;       /* can try inflateSync */
-  return Z_DATA_ERROR;
-}
-
-/*
- * This subroutine adds the data at next_in/avail_in to the output history
- * without performing any output.  The output buffer must be "caught up";
- * i.e. no pending output (hence s->read equals s->write), and the state must
- * be BLOCKS (i.e. we should be willing to see the start of a series of
- * BLOCKS).  On exit, the output will also be caught up, and the checksum
- * will have been updated if need be.
- */
-
-int inflateIncomp(z)
-z_stream *z;
-{
-    if (z->state->mode != BLOCKS)
-	return Z_DATA_ERROR;
-    return inflate_addhistory(z->state->blocks, z);
-}
-
-
-int inflateSync(z)
-z_stream *z;
-{
-  uInt n;       /* number of bytes to look at */
-  Bytef *p;     /* pointer to bytes */
-  uInt m;       /* number of marker bytes found in a row */
-  uLong r, w;   /* temporaries to save total_in and total_out */
-
-  /* set up */
-  if (z == Z_NULL || z->state == Z_NULL)
-    return Z_STREAM_ERROR;
-  if (z->state->mode != BAD)
-  {
-    z->state->mode = BAD;
-    z->state->sub.marker = 0;
-  }
-  if ((n = z->avail_in) == 0)
-    return Z_BUF_ERROR;
-  p = z->next_in;
-  m = z->state->sub.marker;
-
-  /* search */
-  while (n && m < 4)
-  {
-    if (*p == (Byte)(m < 2 ? 0 : 0xff))
-      m++;
-    else if (*p)
-      m = 0;
-    else
-      m = 4 - m;
-    p++, n--;
-  }
-
-  /* restore */
-  z->total_in += p - z->next_in;
-  z->next_in = p;
-  z->avail_in = n;
-  z->state->sub.marker = m;
-
-  /* return no joy or set up to restart on a new block */
-  if (m != 4)
-    return Z_DATA_ERROR;
-  r = z->total_in;  w = z->total_out;
-  inflateReset(z);
-  z->total_in = r;  z->total_out = w;
-  z->state->mode = BLOCKS;
-  return Z_OK;
-}
-
-#undef NEEDBYTE
-#undef NEXTBYTE
-
-/*+++++*/
-/* infutil.h -- types and macros common to blocks and codes
- * Copyright (C) 1995 Mark Adler
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* WARNING: this file should *not* be used by applications. It is
-   part of the implementation of the compression library and is
-   subject to change. Applications should only use zlib.h.
- */
-
-/* inflate blocks semi-private state */
-struct inflate_blocks_state {
-
-  /* mode */
-  enum {
-      TYPE,     /* get type bits (3, including end bit) */
-      LENS,     /* get lengths for stored */
-      STORED,   /* processing stored block */
-      TABLE,    /* get table lengths */
-      BTREE,    /* get bit lengths tree for a dynamic block */
-      DTREE,    /* get length, distance trees for a dynamic block */
-      CODES,    /* processing fixed or dynamic block */
-      DRY,      /* output remaining window bytes */
-      DONEB,     /* finished last block, done */
-      BADB}      /* got a data error--stuck here */
-    mode;               /* current inflate_block mode */
-
-  /* mode dependent information */
-  union {
-    uInt left;          /* if STORED, bytes left to copy */
-    struct {
-      uInt table;               /* table lengths (14 bits) */
-      uInt index;               /* index into blens (or border) */
-      uIntf *blens;             /* bit lengths of codes */
-      uInt bb;                  /* bit length tree depth */
-      inflate_huft *tb;         /* bit length decoding tree */
-      int nblens;		/* # elements allocated at blens */
-    } trees;            /* if DTREE, decoding info for trees */
-    struct {
-      inflate_huft *tl, *td;    /* trees to free */
-      inflate_codes_statef
-         *codes;
-    } decode;           /* if CODES, current state */
-  } sub;                /* submode */
-  uInt last;            /* true if this block is the last block */
-
-  /* mode independent information */
-  uInt bitk;            /* bits in bit buffer */
-  uLong bitb;           /* bit buffer */
-  Bytef *window;        /* sliding window */
-  Bytef *end;           /* one byte after sliding window */
-  Bytef *read;          /* window read pointer */
-  Bytef *write;         /* window write pointer */
-  check_func checkfn;   /* check function */
-  uLong check;          /* check on output */
-
-};
-
-
-/* defines for inflate input/output */
-/*   update pointers and return */
-#define UPDBITS {s->bitb=b;s->bitk=k;}
-#define UPDIN {z->avail_in=n;z->total_in+=p-z->next_in;z->next_in=p;}
-#define UPDOUT {s->write=q;}
-#define UPDATE {UPDBITS UPDIN UPDOUT}
-#define LEAVE {UPDATE return inflate_flush(s,z,r);}
-/*   get bytes and bits */
-#define LOADIN {p=z->next_in;n=z->avail_in;b=s->bitb;k=s->bitk;}
-#define NEEDBYTE {if(n)r=Z_OK;else LEAVE}
-#define NEXTBYTE (n--,*p++)
-#define NEEDBITS(j) {while(k<(j)){NEEDBYTE;b|=((uLong)NEXTBYTE)<<k;k+=8;}}
-#define DUMPBITS(j) {b>>=(j);k-=(j);}
-/*   output bytes */
-#define WAVAIL (q<s->read?s->read-q-1:s->end-q)
-#define LOADOUT {q=s->write;m=WAVAIL;}
-#define WRAP {if(q==s->end&&s->read!=s->window){q=s->window;m=WAVAIL;}}
-#define FLUSH {UPDOUT r=inflate_flush(s,z,r); LOADOUT}
-#define NEEDOUT {if(m==0){WRAP if(m==0){FLUSH WRAP if(m==0) LEAVE}}r=Z_OK;}
-#define OUTBYTE(a) {*q++=(Byte)(a);m--;}
-/*   load local pointers */
-#define LOAD {LOADIN LOADOUT}
-
-/*
- * The IBM 150 firmware munges the data right after _etext[].  This
- * protects it. -- Cort
- */
-local uInt protect_mask[] = {0, 0, 0, 0, 0, 0, 0, 0, 0 ,0 ,0 ,0};
-/* And'ing with mask[n] masks the lower n bits */
-local uInt inflate_mask[] = {
-    0x0000,
-    0x0001, 0x0003, 0x0007, 0x000f, 0x001f, 0x003f, 0x007f, 0x00ff,
-    0x01ff, 0x03ff, 0x07ff, 0x0fff, 0x1fff, 0x3fff, 0x7fff, 0xffff
-};
-
-/* copy as much as possible from the sliding window to the output area */
-local int inflate_flush OF((
-    inflate_blocks_statef *,
-    z_stream *,
-    int));
-
-/*+++++*/
-/* inffast.h -- header to use inffast.c
- * Copyright (C) 1995 Mark Adler
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* WARNING: this file should *not* be used by applications. It is
-   part of the implementation of the compression library and is
-   subject to change. Applications should only use zlib.h.
- */
-
-local int inflate_fast OF((
-    uInt,
-    uInt,
-    inflate_huft *,
-    inflate_huft *,
-    inflate_blocks_statef *,
-    z_stream *));
-
-
-/*+++++*/
-/* infblock.c -- interpret and process block types to last block
- * Copyright (C) 1995 Mark Adler
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* Table for deflate from PKZIP's appnote.txt. */
-local uInt border[] = { /* Order of the bit length code lengths */
-        16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15};
-
-/*
-   Notes beyond the 1.93a appnote.txt:
-
-   1. Distance pointers never point before the beginning of the output
-      stream.
-   2. Distance pointers can point back across blocks, up to 32k away.
-   3. There is an implied maximum of 7 bits for the bit length table and
-      15 bits for the actual data.
-   4. If only one code exists, then it is encoded using one bit.  (Zero
-      would be more efficient, but perhaps a little confusing.)  If two
-      codes exist, they are coded using one bit each (0 and 1).
-   5. There is no way of sending zero distance codes--a dummy must be
-      sent if there are none.  (History: a pre 2.0 version of PKZIP would
-      store blocks with no distance codes, but this was discovered to be
-      too harsh a criterion.)  Valid only for 1.93a.  2.04c does allow
-      zero distance codes, which is sent as one code of zero bits in
-      length.
-   6. There are up to 286 literal/length codes.  Code 256 represents the
-      end-of-block.  Note however that the static length tree defines
-      288 codes just to fill out the Huffman codes.  Codes 286 and 287
-      cannot be used though, since there is no length base or extra bits
-      defined for them.  Similarily, there are up to 30 distance codes.
-      However, static trees define 32 codes (all 5 bits) to fill out the
-      Huffman codes, but the last two had better not show up in the data.
-   7. Unzip can check dynamic Huffman blocks for complete code sets.
-      The exception is that a single code would not be complete (see #4).
-   8. The five bits following the block type is really the number of
-      literal codes sent minus 257.
-   9. Length codes 8,16,16 are interpreted as 13 length codes of 8 bits
-      (1+6+6).  Therefore, to output three times the length, you output
-      three codes (1+1+1), whereas to output four times the same length,
-      you only need two codes (1+3).  Hmm.
-  10. In the tree reconstruction algorithm, Code = Code + Increment
-      only if BitLength(i) is not zero.  (Pretty obvious.)
-  11. Correction: 4 Bits: # of Bit Length codes - 4     (4 - 19)
-  12. Note: length code 284 can represent 227-258, but length code 285
-      really is 258.  The last length deserves its own, short code
-      since it gets used a lot in very redundant files.  The length
-      258 is special since 258 - 3 (the min match length) is 255.
-  13. The literal/length and distance code bit lengths are read as a
-      single stream of lengths.  It is possible (and advantageous) for
-      a repeat code (16, 17, or 18) to go across the boundary between
-      the two sets of lengths.
- */
-
-
-local void inflate_blocks_reset(s, z, c)
-inflate_blocks_statef *s;
-z_stream *z;
-uLongf *c;
-{
-  if (s->checkfn != Z_NULL)
-    *c = s->check;
-  if (s->mode == BTREE || s->mode == DTREE)
-    ZFREE(z, s->sub.trees.blens, s->sub.trees.nblens * sizeof(uInt));
-  if (s->mode == CODES)
-  {
-    inflate_codes_free(s->sub.decode.codes, z);
-    inflate_trees_free(s->sub.decode.td, z);
-    inflate_trees_free(s->sub.decode.tl, z);
-  }
-  s->mode = TYPE;
-  s->bitk = 0;
-  s->bitb = 0;
-  s->read = s->write = s->window;
-  if (s->checkfn != Z_NULL)
-    s->check = (*s->checkfn)(0L, Z_NULL, 0);
-  Trace((stderr, "inflate:   blocks reset\n"));
-}
-
-
-local inflate_blocks_statef *inflate_blocks_new(z, c, w)
-z_stream *z;
-check_func c;
-uInt w;
-{
-  inflate_blocks_statef *s;
-
-  if ((s = (inflate_blocks_statef *)ZALLOC
-       (z,1,sizeof(struct inflate_blocks_state))) == Z_NULL)
-    return s;
-  if ((s->window = (Bytef *)ZALLOC(z, 1, w)) == Z_NULL)
-  {
-    ZFREE(z, s, sizeof(struct inflate_blocks_state));
-    return Z_NULL;
-  }
-  s->end = s->window + w;
-  s->checkfn = c;
-  s->mode = TYPE;
-  Trace((stderr, "inflate:   blocks allocated\n"));
-  inflate_blocks_reset(s, z, &s->check);
-  return s;
-}
-
-
-local int inflate_blocks(s, z, r)
-inflate_blocks_statef *s;
-z_stream *z;
-int r;
-{
-  uInt t;               /* temporary storage */
-  uLong b;              /* bit buffer */
-  uInt k;               /* bits in bit buffer */
-  Bytef *p;             /* input data pointer */
-  uInt n;               /* bytes available there */
-  Bytef *q;             /* output window write pointer */
-  uInt m;               /* bytes to end of window or read pointer */
-
-  /* copy input/output information to locals (UPDATE macro restores) */
-  LOAD
-
-  /* process input based on current state */
-  while (1) switch (s->mode)
-  {
-    case TYPE:
-      NEEDBITS(3)
-      t = (uInt)b & 7;
-      s->last = t & 1;
-      switch (t >> 1)
-      {
-        case 0:                         /* stored */
-          Trace((stderr, "inflate:     stored block%s\n",
-                 s->last ? " (last)" : ""));
-          DUMPBITS(3)
-          t = k & 7;                    /* go to byte boundary */
-          DUMPBITS(t)
-          s->mode = LENS;               /* get length of stored block */
-          break;
-        case 1:                         /* fixed */
-          Trace((stderr, "inflate:     fixed codes block%s\n",
-                 s->last ? " (last)" : ""));
-          {
-            uInt bl, bd;
-            inflate_huft *tl, *td;
-
-            inflate_trees_fixed(&bl, &bd, &tl, &td);
-            s->sub.decode.codes = inflate_codes_new(bl, bd, tl, td, z);
-            if (s->sub.decode.codes == Z_NULL)
-            {
-              r = Z_MEM_ERROR;
-              LEAVE
-            }
-            s->sub.decode.tl = Z_NULL;  /* don't try to free these */
-            s->sub.decode.td = Z_NULL;
-          }
-          DUMPBITS(3)
-          s->mode = CODES;
-          break;
-        case 2:                         /* dynamic */
-          Trace((stderr, "inflate:     dynamic codes block%s\n",
-                 s->last ? " (last)" : ""));
-          DUMPBITS(3)
-          s->mode = TABLE;
-          break;
-        case 3:                         /* illegal */
-          DUMPBITS(3)
-          s->mode = BADB;
-          z->msg = "invalid block type";
-          r = Z_DATA_ERROR;
-          LEAVE
-      }
-      break;
-    case LENS:
-      NEEDBITS(32)
-      if (((~b) >> 16) != (b & 0xffff))
-      {
-        s->mode = BADB;
-        z->msg = "invalid stored block lengths";
-        r = Z_DATA_ERROR;
-        LEAVE
-      }
-      s->sub.left = (uInt)b & 0xffff;
-      b = k = 0;                      /* dump bits */
-      Tracev((stderr, "inflate:       stored length %u\n", s->sub.left));
-      s->mode = s->sub.left ? STORED : TYPE;
-      break;
-    case STORED:
-      if (n == 0)
-        LEAVE
-      NEEDOUT
-      t = s->sub.left;
-      if (t > n) t = n;
-      if (t > m) t = m;
-      zmemcpy(q, p, t);
-      p += t;  n -= t;
-      q += t;  m -= t;
-      if ((s->sub.left -= t) != 0)
-        break;
-      Tracev((stderr, "inflate:       stored end, %lu total out\n",
-              z->total_out + (q >= s->read ? q - s->read :
-              (s->end - s->read) + (q - s->window))));
-      s->mode = s->last ? DRY : TYPE;
-      break;
-    case TABLE:
-      NEEDBITS(14)
-      s->sub.trees.table = t = (uInt)b & 0x3fff;
-#ifndef PKZIP_BUG_WORKAROUND
-      if ((t & 0x1f) > 29 || ((t >> 5) & 0x1f) > 29)
-      {
-        s->mode = BADB;
-        z->msg = "too many length or distance symbols";
-        r = Z_DATA_ERROR;
-        LEAVE
-      }
-#endif
-      t = 258 + (t & 0x1f) + ((t >> 5) & 0x1f);
-      if (t < 19)
-        t = 19;
-      if ((s->sub.trees.blens = (uIntf*)ZALLOC(z, t, sizeof(uInt))) == Z_NULL)
-      {
-        r = Z_MEM_ERROR;
-        LEAVE
-      }
-      s->sub.trees.nblens = t;
-      DUMPBITS(14)
-      s->sub.trees.index = 0;
-      Tracev((stderr, "inflate:       table sizes ok\n"));
-      s->mode = BTREE;
-    case BTREE:
-      while (s->sub.trees.index < 4 + (s->sub.trees.table >> 10))
-      {
-        NEEDBITS(3)
-        s->sub.trees.blens[border[s->sub.trees.index++]] = (uInt)b & 7;
-        DUMPBITS(3)
-      }
-      while (s->sub.trees.index < 19)
-        s->sub.trees.blens[border[s->sub.trees.index++]] = 0;
-      s->sub.trees.bb = 7;
-      t = inflate_trees_bits(s->sub.trees.blens, &s->sub.trees.bb,
-                             &s->sub.trees.tb, z);
-      if (t != Z_OK)
-      {
-        r = t;
-        if (r == Z_DATA_ERROR)
-          s->mode = BADB;
-        LEAVE
-      }
-      s->sub.trees.index = 0;
-      Tracev((stderr, "inflate:       bits tree ok\n"));
-      s->mode = DTREE;
-    case DTREE:
-      while (t = s->sub.trees.table,
-             s->sub.trees.index < 258 + (t & 0x1f) + ((t >> 5) & 0x1f))
-      {
-        inflate_huft *h;
-        uInt i, j, c;
-
-        t = s->sub.trees.bb;
-        NEEDBITS(t)
-        h = s->sub.trees.tb + ((uInt)b & inflate_mask[t]);
-        t = h->word.what.Bits;
-        c = h->more.Base;
-        if (c < 16)
-        {
-          DUMPBITS(t)
-          s->sub.trees.blens[s->sub.trees.index++] = c;
-        }
-        else /* c == 16..18 */
-        {
-          i = c == 18 ? 7 : c - 14;
-          j = c == 18 ? 11 : 3;
-          NEEDBITS(t + i)
-          DUMPBITS(t)
-          j += (uInt)b & inflate_mask[i];
-          DUMPBITS(i)
-          i = s->sub.trees.index;
-          t = s->sub.trees.table;
-          if (i + j > 258 + (t & 0x1f) + ((t >> 5) & 0x1f) ||
-              (c == 16 && i < 1))
-          {
-            s->mode = BADB;
-            z->msg = "invalid bit length repeat";
-            r = Z_DATA_ERROR;
-            LEAVE
-          }
-          c = c == 16 ? s->sub.trees.blens[i - 1] : 0;
-          do {
-            s->sub.trees.blens[i++] = c;
-          } while (--j);
-          s->sub.trees.index = i;
-        }
-      }
-      inflate_trees_free(s->sub.trees.tb, z);
-      s->sub.trees.tb = Z_NULL;
-      {
-        uInt bl, bd;
-        inflate_huft *tl, *td;
-        inflate_codes_statef *c;
-
-        bl = 9;         /* must be <= 9 for lookahead assumptions */
-        bd = 6;         /* must be <= 9 for lookahead assumptions */
-        t = s->sub.trees.table;
-        t = inflate_trees_dynamic(257 + (t & 0x1f), 1 + ((t >> 5) & 0x1f),
-                                  s->sub.trees.blens, &bl, &bd, &tl, &td, z);
-        if (t != Z_OK)
-        {
-          if (t == (uInt)Z_DATA_ERROR)
-            s->mode = BADB;
-          r = t;
-          LEAVE
-        }
-        Tracev((stderr, "inflate:       trees ok\n"));
-        if ((c = inflate_codes_new(bl, bd, tl, td, z)) == Z_NULL)
-        {
-          inflate_trees_free(td, z);
-          inflate_trees_free(tl, z);
-          r = Z_MEM_ERROR;
-          LEAVE
-        }
-        ZFREE(z, s->sub.trees.blens, s->sub.trees.nblens * sizeof(uInt));
-        s->sub.decode.codes = c;
-        s->sub.decode.tl = tl;
-        s->sub.decode.td = td;
-      }
-      s->mode = CODES;
-    case CODES:
-      UPDATE
-      if ((r = inflate_codes(s, z, r)) != Z_STREAM_END)
-        return inflate_flush(s, z, r);
-      r = Z_OK;
-      inflate_codes_free(s->sub.decode.codes, z);
-      inflate_trees_free(s->sub.decode.td, z);
-      inflate_trees_free(s->sub.decode.tl, z);
-      LOAD
-      Tracev((stderr, "inflate:       codes end, %lu total out\n",
-              z->total_out + (q >= s->read ? q - s->read :
-              (s->end - s->read) + (q - s->window))));
-      if (!s->last)
-      {
-        s->mode = TYPE;
-        break;
-      }
-      if (k > 7)              /* return unused byte, if any */
-      {
-        Assert(k < 16, "inflate_codes grabbed too many bytes")
-        k -= 8;
-        n++;
-        p--;                    /* can always return one */
-      }
-      s->mode = DRY;
-    case DRY:
-      FLUSH
-      if (s->read != s->write)
-        LEAVE
-      s->mode = DONEB;
-    case DONEB:
-      r = Z_STREAM_END;
-      LEAVE
-    case BADB:
-      r = Z_DATA_ERROR;
-      LEAVE
-    default:
-      r = Z_STREAM_ERROR;
-      LEAVE
-  }
-}
-
-
-local int inflate_blocks_free(s, z, c)
-inflate_blocks_statef *s;
-z_stream *z;
-uLongf *c;
-{
-  inflate_blocks_reset(s, z, c);
-  ZFREE(z, s->window, s->end - s->window);
-  ZFREE(z, s, sizeof(struct inflate_blocks_state));
-  Trace((stderr, "inflate:   blocks freed\n"));
-  return Z_OK;
-}
-
-/*
- * This subroutine adds the data at next_in/avail_in to the output history
- * without performing any output.  The output buffer must be "caught up";
- * i.e. no pending output (hence s->read equals s->write), and the state must
- * be BLOCKS (i.e. we should be willing to see the start of a series of
- * BLOCKS).  On exit, the output will also be caught up, and the checksum
- * will have been updated if need be.
- */
-local int inflate_addhistory(s, z)
-inflate_blocks_statef *s;
-z_stream *z;
-{
-    uLong b;              /* bit buffer */  /* NOT USED HERE */
-    uInt k;               /* bits in bit buffer */ /* NOT USED HERE */
-    uInt t;               /* temporary storage */
-    Bytef *p;             /* input data pointer */
-    uInt n;               /* bytes available there */
-    Bytef *q;             /* output window write pointer */
-    uInt m;               /* bytes to end of window or read pointer */
-
-    if (s->read != s->write)
-	return Z_STREAM_ERROR;
-    if (s->mode != TYPE)
-	return Z_DATA_ERROR;
-
-    /* we're ready to rock */
-    LOAD
-    /* while there is input ready, copy to output buffer, moving
-     * pointers as needed.
-     */
-    while (n) {
-	t = n;  /* how many to do */
-	/* is there room until end of buffer? */
-	if (t > m) t = m;
-	/* update check information */
-	if (s->checkfn != Z_NULL)
-	    s->check = (*s->checkfn)(s->check, q, t);
-	zmemcpy(q, p, t);
-	q += t;
-	p += t;
-	n -= t;
-	z->total_out += t;
-	s->read = q;    /* drag read pointer forward */
-/*      WRAP  */ 	/* expand WRAP macro by hand to handle s->read */
-	if (q == s->end) {
-	    s->read = q = s->window;
-	    m = WAVAIL;
-	}
-    }
-    UPDATE
-    return Z_OK;
-}
-
-
-/*
- * At the end of a Deflate-compressed PPP packet, we expect to have seen
- * a `stored' block type value but not the (zero) length bytes.
- */
-local int inflate_packet_flush(s)
-    inflate_blocks_statef *s;
-{
-    if (s->mode != LENS)
-	return Z_DATA_ERROR;
-    s->mode = TYPE;
-    return Z_OK;
-}
-
-
-/*+++++*/
-/* inftrees.c -- generate Huffman trees for efficient decoding
- * Copyright (C) 1995 Mark Adler
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* simplify the use of the inflate_huft type with some defines */
-#define base more.Base
-#define next more.Next
-#define exop word.what.Exop
-#define bits word.what.Bits
-
-
-local int huft_build OF((
-    uIntf *,            /* code lengths in bits */
-    uInt,               /* number of codes */
-    uInt,               /* number of "simple" codes */
-    uIntf *,            /* list of base values for non-simple codes */
-    uIntf *,            /* list of extra bits for non-simple codes */
-    inflate_huft * FAR*,/* result: starting table */
-    uIntf *,            /* maximum lookup bits (returns actual) */
-    z_stream *));       /* for zalloc function */
-
-local voidpf falloc OF((
-    voidpf,             /* opaque pointer (not used) */
-    uInt,               /* number of items */
-    uInt));             /* size of item */
-
-local void ffree OF((
-    voidpf q,           /* opaque pointer (not used) */
-    voidpf p,           /* what to free (not used) */
-    uInt n));		/* number of bytes (not used) */
-
-/* Tables for deflate from PKZIP's appnote.txt. */
-local uInt cplens[] = { /* Copy lengths for literal codes 257..285 */
-        3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 15, 17, 19, 23, 27, 31,
-        35, 43, 51, 59, 67, 83, 99, 115, 131, 163, 195, 227, 258, 0, 0};
-        /* actually lengths - 2; also see note #13 above about 258 */
-local uInt cplext[] = { /* Extra bits for literal codes 257..285 */
-        0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2,
-        3, 3, 3, 3, 4, 4, 4, 4, 5, 5, 5, 5, 0, 192, 192}; /* 192==invalid */
-local uInt cpdist[] = { /* Copy offsets for distance codes 0..29 */
-        1, 2, 3, 4, 5, 7, 9, 13, 17, 25, 33, 49, 65, 97, 129, 193,
-        257, 385, 513, 769, 1025, 1537, 2049, 3073, 4097, 6145,
-        8193, 12289, 16385, 24577};
-local uInt cpdext[] = { /* Extra bits for distance codes */
-        0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6,
-        7, 7, 8, 8, 9, 9, 10, 10, 11, 11,
-        12, 12, 13, 13};
-
-/*
-   Huffman code decoding is performed using a multi-level table lookup.
-   The fastest way to decode is to simply build a lookup table whose
-   size is determined by the longest code.  However, the time it takes
-   to build this table can also be a factor if the data being decoded
-   is not very long.  The most common codes are necessarily the
-   shortest codes, so those codes dominate the decoding time, and hence
-   the speed.  The idea is you can have a shorter table that decodes the
-   shorter, more probable codes, and then point to subsidiary tables for
-   the longer codes.  The time it costs to decode the longer codes is
-   then traded against the time it takes to make longer tables.
-
-   This results of this trade are in the variables lbits and dbits
-   below.  lbits is the number of bits the first level table for literal/
-   length codes can decode in one step, and dbits is the same thing for
-   the distance codes.  Subsequent tables are also less than or equal to
-   those sizes.  These values may be adjusted either when all of the
-   codes are shorter than that, in which case the longest code length in
-   bits is used, or when the shortest code is *longer* than the requested
-   table size, in which case the length of the shortest code in bits is
-   used.
-
-   There are two different values for the two tables, since they code a
-   different number of possibilities each.  The literal/length table
-   codes 286 possible values, or in a flat code, a little over eight
-   bits.  The distance table codes 30 possible values, or a little less
-   than five bits, flat.  The optimum values for speed end up being
-   about one bit more than those, so lbits is 8+1 and dbits is 5+1.
-   The optimum values may differ though from machine to machine, and
-   possibly even between compilers.  Your mileage may vary.
- */
-
-
-/* If BMAX needs to be larger than 16, then h and x[] should be uLong. */
-#define BMAX 15         /* maximum bit length of any code */
-#define N_MAX 288       /* maximum number of codes in any set */
-
-#ifdef DEBUG_ZLIB
-  uInt inflate_hufts;
-#endif
-
-local int huft_build(b, n, s, d, e, t, m, zs)
-uIntf *b;               /* code lengths in bits (all assumed <= BMAX) */
-uInt n;                 /* number of codes (assumed <= N_MAX) */
-uInt s;                 /* number of simple-valued codes (0..s-1) */
-uIntf *d;               /* list of base values for non-simple codes */
-uIntf *e;               /* list of extra bits for non-simple codes */
-inflate_huft * FAR *t;  /* result: starting table */
-uIntf *m;               /* maximum lookup bits, returns actual */
-z_stream *zs;           /* for zalloc function */
-/* Given a list of code lengths and a maximum table size, make a set of
-   tables to decode that set of codes.  Return Z_OK on success, Z_BUF_ERROR
-   if the given code set is incomplete (the tables are still built in this
-   case), Z_DATA_ERROR if the input is invalid (all zero length codes or an
-   over-subscribed set of lengths), or Z_MEM_ERROR if not enough memory. */
-{
-
-  uInt a;                       /* counter for codes of length k */
-  uInt c[BMAX+1];               /* bit length count table */
-  uInt f;                       /* i repeats in table every f entries */
-  int g;                        /* maximum code length */
-  int h;                        /* table level */
-  register uInt i;              /* counter, current code */
-  register uInt j;              /* counter */
-  register int k;               /* number of bits in current code */
-  int l;                        /* bits per table (returned in m) */
-  register uIntf *p;            /* pointer into c[], b[], or v[] */
-  inflate_huft *q;              /* points to current table */
-  struct inflate_huft_s r;      /* table entry for structure assignment */
-  inflate_huft *u[BMAX];        /* table stack */
-  uInt v[N_MAX];                /* values in order of bit length */
-  register int w;               /* bits before this table == (l * h) */
-  uInt x[BMAX+1];               /* bit offsets, then code stack */
-  uIntf *xp;                    /* pointer into x */
-  int y;                        /* number of dummy codes added */
-  uInt z;                       /* number of entries in current table */
-
-
-  /* Generate counts for each bit length */
-  p = c;
-#define C0 *p++ = 0;
-#define C2 C0 C0 C0 C0
-#define C4 C2 C2 C2 C2
-  C4                            /* clear c[]--assume BMAX+1 is 16 */
-  p = b;  i = n;
-  do {
-    c[*p++]++;                  /* assume all entries <= BMAX */
-  } while (--i);
-  if (c[0] == n)                /* null input--all zero length codes */
-  {
-    *t = (inflate_huft *)Z_NULL;
-    *m = 0;
-    return Z_OK;
-  }
-
-
-  /* Find minimum and maximum length, bound *m by those */
-  l = *m;
-  for (j = 1; j <= BMAX; j++)
-    if (c[j])
-      break;
-  k = j;                        /* minimum code length */
-  if ((uInt)l < j)
-    l = j;
-  for (i = BMAX; i; i--)
-    if (c[i])
-      break;
-  g = i;                        /* maximum code length */
-  if ((uInt)l > i)
-    l = i;
-  *m = l;
-
-
-  /* Adjust last length count to fill out codes, if needed */
-  for (y = 1 << j; j < i; j++, y <<= 1)
-    if ((y -= c[j]) < 0)
-      return Z_DATA_ERROR;
-  if ((y -= c[i]) < 0)
-    return Z_DATA_ERROR;
-  c[i] += y;
-
-
-  /* Generate starting offsets into the value table for each length */
-  x[1] = j = 0;
-  p = c + 1;  xp = x + 2;
-  while (--i) {                 /* note that i == g from above */
-    *xp++ = (j += *p++);
-  }
-
-
-  /* Make a table of values in order of bit lengths */
-  p = b;  i = 0;
-  do {
-    if ((j = *p++) != 0)
-      v[x[j]++] = i;
-  } while (++i < n);
-
-
-  /* Generate the Huffman codes and for each, make the table entries */
-  x[0] = i = 0;                 /* first Huffman code is zero */
-  p = v;                        /* grab values in bit order */
-  h = -1;                       /* no tables yet--level -1 */
-  w = -l;                       /* bits decoded == (l * h) */
-  u[0] = (inflate_huft *)Z_NULL;        /* just to keep compilers happy */
-  q = (inflate_huft *)Z_NULL;   /* ditto */
-  z = 0;                        /* ditto */
-
-  /* go through the bit lengths (k already is bits in shortest code) */
-  for (; k <= g; k++)
-  {
-    a = c[k];
-    while (a--)
-    {
-      /* here i is the Huffman code of length k bits for value *p */
-      /* make tables up to required level */
-      while (k > w + l)
-      {
-        h++;
-        w += l;                 /* previous table always l bits */
-
-        /* compute minimum size table less than or equal to l bits */
-        z = (z = g - w) > (uInt)l ? l : z;      /* table size upper limit */
-        if ((f = 1 << (j = k - w)) > a + 1)     /* try a k-w bit table */
-        {                       /* too few codes for k-w bit table */
-          f -= a + 1;           /* deduct codes from patterns left */
-          xp = c + k;
-          if (j < z)
-            while (++j < z)     /* try smaller tables up to z bits */
-            {
-              if ((f <<= 1) <= *++xp)
-                break;          /* enough codes to use up j bits */
-              f -= *xp;         /* else deduct codes from patterns */
-            }
-        }
-        z = 1 << j;             /* table entries for j-bit table */
-
-        /* allocate and link in new table */
-        if ((q = (inflate_huft *)ZALLOC
-             (zs,z + 1,sizeof(inflate_huft))) == Z_NULL)
-        {
-          if (h)
-            inflate_trees_free(u[0], zs);
-          return Z_MEM_ERROR;   /* not enough memory */
-        }
-	q->word.Nalloc = z + 1;
-#ifdef DEBUG_ZLIB
-        inflate_hufts += z + 1;
-#endif
-        *t = q + 1;             /* link to list for huft_free() */
-        *(t = &(q->next)) = Z_NULL;
-        u[h] = ++q;             /* table starts after link */
-
-        /* connect to last table, if there is one */
-        if (h)
-        {
-          x[h] = i;             /* save pattern for backing up */
-          r.bits = (Byte)l;     /* bits to dump before this table */
-          r.exop = (Byte)j;     /* bits in this table */
-          r.next = q;           /* pointer to this table */
-          j = i >> (w - l);     /* (get around Turbo C bug) */
-          u[h-1][j] = r;        /* connect to last table */
-        }
-      }
-
-      /* set up table entry in r */
-      r.bits = (Byte)(k - w);
-      if (p >= v + n)
-        r.exop = 128 + 64;      /* out of values--invalid code */
-      else if (*p < s)
-      {
-        r.exop = (Byte)(*p < 256 ? 0 : 32 + 64);     /* 256 is end-of-block */
-        r.base = *p++;          /* simple code is just the value */
-      }
-      else
-      {
-        r.exop = (Byte)e[*p - s] + 16 + 64; /* non-simple--look up in lists */
-        r.base = d[*p++ - s];
-      }
-
-      /* fill code-like entries with r */
-      f = 1 << (k - w);
-      for (j = i >> w; j < z; j += f)
-        q[j] = r;
-
-      /* backwards increment the k-bit code i */
-      for (j = 1 << (k - 1); i & j; j >>= 1)
-        i ^= j;
-      i ^= j;
-
-      /* backup over finished tables */
-      while ((i & ((1 << w) - 1)) != x[h])
-      {
-        h--;                    /* don't need to update q */
-        w -= l;
-      }
-    }
-  }
-
-
-  /* Return Z_BUF_ERROR if we were given an incomplete table */
-  return y != 0 && g != 1 ? Z_BUF_ERROR : Z_OK;
-}
-
-
-local int inflate_trees_bits(c, bb, tb, z)
-uIntf *c;               /* 19 code lengths */
-uIntf *bb;              /* bits tree desired/actual depth */
-inflate_huft * FAR *tb; /* bits tree result */
-z_stream *z;            /* for zfree function */
-{
-  int r;
-
-  r = huft_build(c, 19, 19, (uIntf*)Z_NULL, (uIntf*)Z_NULL, tb, bb, z);
-  if (r == Z_DATA_ERROR)
-    z->msg = "oversubscribed dynamic bit lengths tree";
-  else if (r == Z_BUF_ERROR)
-  {
-    inflate_trees_free(*tb, z);
-    z->msg = "incomplete dynamic bit lengths tree";
-    r = Z_DATA_ERROR;
-  }
-  return r;
-}
-
-
-local int inflate_trees_dynamic(nl, nd, c, bl, bd, tl, td, z)
-uInt nl;                /* number of literal/length codes */
-uInt nd;                /* number of distance codes */
-uIntf *c;               /* that many (total) code lengths */
-uIntf *bl;              /* literal desired/actual bit depth */
-uIntf *bd;              /* distance desired/actual bit depth */
-inflate_huft * FAR *tl; /* literal/length tree result */
-inflate_huft * FAR *td; /* distance tree result */
-z_stream *z;            /* for zfree function */
-{
-  int r;
-
-  /* build literal/length tree */
-  if ((r = huft_build(c, nl, 257, cplens, cplext, tl, bl, z)) != Z_OK)
-  {
-    if (r == Z_DATA_ERROR)
-      z->msg = "oversubscribed literal/length tree";
-    else if (r == Z_BUF_ERROR)
-    {
-      inflate_trees_free(*tl, z);
-      z->msg = "incomplete literal/length tree";
-      r = Z_DATA_ERROR;
-    }
-    return r;
-  }
-
-  /* build distance tree */
-  if ((r = huft_build(c + nl, nd, 0, cpdist, cpdext, td, bd, z)) != Z_OK)
-  {
-    if (r == Z_DATA_ERROR)
-      z->msg = "oversubscribed literal/length tree";
-    else if (r == Z_BUF_ERROR) {
-#ifdef PKZIP_BUG_WORKAROUND
-      r = Z_OK;
-    }
-#else
-      inflate_trees_free(*td, z);
-      z->msg = "incomplete literal/length tree";
-      r = Z_DATA_ERROR;
-    }
-    inflate_trees_free(*tl, z);
-    return r;
-#endif
-  }
-
-  /* done */
-  return Z_OK;
-}
-
-
-/* build fixed tables only once--keep them here */
-local int fixed_lock = 0;
-local int fixed_built = 0;
-#define FIXEDH 530      /* number of hufts used by fixed tables */
-local uInt fixed_left = FIXEDH;
-local inflate_huft fixed_mem[FIXEDH];
-local uInt fixed_bl;
-local uInt fixed_bd;
-local inflate_huft *fixed_tl;
-local inflate_huft *fixed_td;
-
-
-local voidpf falloc(q, n, s)
-voidpf q;        /* opaque pointer (not used) */
-uInt n;         /* number of items */
-uInt s;         /* size of item */
-{
-  Assert(s == sizeof(inflate_huft) && n <= fixed_left,
-         "inflate_trees falloc overflow");
-  if (q) s++; /* to make some compilers happy */
-  fixed_left -= n;
-  return (voidpf)(fixed_mem + fixed_left);
-}
-
-
-local void ffree(q, p, n)
-voidpf q;
-voidpf p;
-uInt n;
-{
-  Assert(0, "inflate_trees ffree called!");
-  if (q) q = p; /* to make some compilers happy */
-}
-
-
-local int inflate_trees_fixed(bl, bd, tl, td)
-uIntf *bl;               /* literal desired/actual bit depth */
-uIntf *bd;               /* distance desired/actual bit depth */
-inflate_huft * FAR *tl;  /* literal/length tree result */
-inflate_huft * FAR *td;  /* distance tree result */
-{
-  /* build fixed tables if not built already--lock out other instances */
-  while (++fixed_lock > 1)
-    fixed_lock--;
-  if (!fixed_built)
-  {
-    int k;              /* temporary variable */
-    unsigned c[288];    /* length list for huft_build */
-    z_stream z;         /* for falloc function */
-
-    /* set up fake z_stream for memory routines */
-    z.zalloc = falloc;
-    z.zfree = ffree;
-    z.opaque = Z_NULL;
-
-    /* literal table */
-    for (k = 0; k < 144; k++)
-      c[k] = 8;
-    for (; k < 256; k++)
-      c[k] = 9;
-    for (; k < 280; k++)
-      c[k] = 7;
-    for (; k < 288; k++)
-      c[k] = 8;
-    fixed_bl = 7;
-    huft_build(c, 288, 257, cplens, cplext, &fixed_tl, &fixed_bl, &z);
-
-    /* distance table */
-    for (k = 0; k < 30; k++)
-      c[k] = 5;
-    fixed_bd = 5;
-    huft_build(c, 30, 0, cpdist, cpdext, &fixed_td, &fixed_bd, &z);
-
-    /* done */
-    fixed_built = 1;
-  }
-  fixed_lock--;
-  *bl = fixed_bl;
-  *bd = fixed_bd;
-  *tl = fixed_tl;
-  *td = fixed_td;
-  return Z_OK;
-}
-
-
-local int inflate_trees_free(t, z)
-inflate_huft *t;        /* table to free */
-z_stream *z;            /* for zfree function */
-/* Free the malloc'ed tables built by huft_build(), which makes a linked
-   list of the tables it made, with the links in a dummy first entry of
-   each table. */
-{
-  register inflate_huft *p, *q;
-
-  /* Go through linked list, freeing from the malloced (t[-1]) address. */
-  p = t;
-  while (p != Z_NULL)
-  {
-    q = (--p)->next;
-    ZFREE(z, p, p->word.Nalloc * sizeof(inflate_huft));
-    p = q;
-  }
-  return Z_OK;
-}
-
-/*+++++*/
-/* infcodes.c -- process literals and length/distance pairs
- * Copyright (C) 1995 Mark Adler
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* simplify the use of the inflate_huft type with some defines */
-#define base more.Base
-#define next more.Next
-#define exop word.what.Exop
-#define bits word.what.Bits
-
-/* inflate codes private state */
-struct inflate_codes_state {
-
-  /* mode */
-  enum {        /* waiting for "i:"=input, "o:"=output, "x:"=nothing */
-      START,    /* x: set up for LEN */
-      LEN,      /* i: get length/literal/eob next */
-      LENEXT,   /* i: getting length extra (have base) */
-      DIST,     /* i: get distance next */
-      DISTEXT,  /* i: getting distance extra */
-      COPY,     /* o: copying bytes in window, waiting for space */
-      LIT,      /* o: got literal, waiting for output space */
-      WASH,     /* o: got eob, possibly still output waiting */
-      END,      /* x: got eob and all data flushed */
-      BADCODE}  /* x: got error */
-    mode;               /* current inflate_codes mode */
-
-  /* mode dependent information */
-  uInt len;
-  union {
-    struct {
-      inflate_huft *tree;       /* pointer into tree */
-      uInt need;                /* bits needed */
-    } code;             /* if LEN or DIST, where in tree */
-    uInt lit;           /* if LIT, literal */
-    struct {
-      uInt get;                 /* bits to get for extra */
-      uInt dist;                /* distance back to copy from */
-    } copy;             /* if EXT or COPY, where and how much */
-  } sub;                /* submode */
-
-  /* mode independent information */
-  Byte lbits;           /* ltree bits decoded per branch */
-  Byte dbits;           /* dtree bits decoder per branch */
-  inflate_huft *ltree;          /* literal/length/eob tree */
-  inflate_huft *dtree;          /* distance tree */
-
-};
-
-
-local inflate_codes_statef *inflate_codes_new(bl, bd, tl, td, z)
-uInt bl, bd;
-inflate_huft *tl, *td;
-z_stream *z;
-{
-  inflate_codes_statef *c;
-
-  if ((c = (inflate_codes_statef *)
-       ZALLOC(z,1,sizeof(struct inflate_codes_state))) != Z_NULL)
-  {
-    c->mode = START;
-    c->lbits = (Byte)bl;
-    c->dbits = (Byte)bd;
-    c->ltree = tl;
-    c->dtree = td;
-    Tracev((stderr, "inflate:       codes new\n"));
-  }
-  return c;
-}
-
-
-local int inflate_codes(s, z, r)
-inflate_blocks_statef *s;
-z_stream *z;
-int r;
-{
-  uInt j;               /* temporary storage */
-  inflate_huft *t;      /* temporary pointer */
-  uInt e;               /* extra bits or operation */
-  uLong b;              /* bit buffer */
-  uInt k;               /* bits in bit buffer */
-  Bytef *p;             /* input data pointer */
-  uInt n;               /* bytes available there */
-  Bytef *q;             /* output window write pointer */
-  uInt m;               /* bytes to end of window or read pointer */
-  Bytef *f;             /* pointer to copy strings from */
-  inflate_codes_statef *c = s->sub.decode.codes;  /* codes state */
-
-  /* copy input/output information to locals (UPDATE macro restores) */
-  LOAD
-
-  /* process input and output based on current state */
-  while (1) switch (c->mode)
-  {             /* waiting for "i:"=input, "o:"=output, "x:"=nothing */
-    case START:         /* x: set up for LEN */
-#ifndef SLOW
-      if (m >= 258 && n >= 10)
-      {
-        UPDATE
-        r = inflate_fast(c->lbits, c->dbits, c->ltree, c->dtree, s, z);
-        LOAD
-        if (r != Z_OK)
-        {
-          c->mode = r == Z_STREAM_END ? WASH : BADCODE;
-          break;
-        }
-      }
-#endif /* !SLOW */
-      c->sub.code.need = c->lbits;
-      c->sub.code.tree = c->ltree;
-      c->mode = LEN;
-    case LEN:           /* i: get length/literal/eob next */
-      j = c->sub.code.need;
-      NEEDBITS(j)
-      t = c->sub.code.tree + ((uInt)b & inflate_mask[j]);
-      DUMPBITS(t->bits)
-      e = (uInt)(t->exop);
-      if (e == 0)               /* literal */
-      {
-        c->sub.lit = t->base;
-        Tracevv((stderr, t->base >= 0x20 && t->base < 0x7f ?
-                 "inflate:         literal '%c'\n" :
-                 "inflate:         literal 0x%02x\n", t->base));
-        c->mode = LIT;
-        break;
-      }
-      if (e & 16)               /* length */
-      {
-        c->sub.copy.get = e & 15;
-        c->len = t->base;
-        c->mode = LENEXT;
-        break;
-      }
-      if ((e & 64) == 0)        /* next table */
-      {
-        c->sub.code.need = e;
-        c->sub.code.tree = t->next;
-        break;
-      }
-      if (e & 32)               /* end of block */
-      {
-        Tracevv((stderr, "inflate:         end of block\n"));
-        c->mode = WASH;
-        break;
-      }
-      c->mode = BADCODE;        /* invalid code */
-      z->msg = "invalid literal/length code";
-      r = Z_DATA_ERROR;
-      LEAVE
-    case LENEXT:        /* i: getting length extra (have base) */
-      j = c->sub.copy.get;
-      NEEDBITS(j)
-      c->len += (uInt)b & inflate_mask[j];
-      DUMPBITS(j)
-      c->sub.code.need = c->dbits;
-      c->sub.code.tree = c->dtree;
-      Tracevv((stderr, "inflate:         length %u\n", c->len));
-      c->mode = DIST;
-    case DIST:          /* i: get distance next */
-      j = c->sub.code.need;
-      NEEDBITS(j)
-      t = c->sub.code.tree + ((uInt)b & inflate_mask[j]);
-      DUMPBITS(t->bits)
-      e = (uInt)(t->exop);
-      if (e & 16)               /* distance */
-      {
-        c->sub.copy.get = e & 15;
-        c->sub.copy.dist = t->base;
-        c->mode = DISTEXT;
-        break;
-      }
-      if ((e & 64) == 0)        /* next table */
-      {
-        c->sub.code.need = e;
-        c->sub.code.tree = t->next;
-        break;
-      }
-      c->mode = BADCODE;        /* invalid code */
-      z->msg = "invalid distance code";
-      r = Z_DATA_ERROR;
-      LEAVE
-    case DISTEXT:       /* i: getting distance extra */
-      j = c->sub.copy.get;
-      NEEDBITS(j)
-      c->sub.copy.dist += (uInt)b & inflate_mask[j];
-      DUMPBITS(j)
-      Tracevv((stderr, "inflate:         distance %u\n", c->sub.copy.dist));
-      c->mode = COPY;
-    case COPY:          /* o: copying bytes in window, waiting for space */
-#ifndef __TURBOC__ /* Turbo C bug for following expression */
-      f = (uInt)(q - s->window) < c->sub.copy.dist ?
-          s->end - (c->sub.copy.dist - (q - s->window)) :
-          q - c->sub.copy.dist;
-#else
-      f = q - c->sub.copy.dist;
-      if ((uInt)(q - s->window) < c->sub.copy.dist)
-        f = s->end - (c->sub.copy.dist - (q - s->window));
-#endif
-      while (c->len)
-      {
-        NEEDOUT
-        OUTBYTE(*f++)
-        if (f == s->end)
-          f = s->window;
-        c->len--;
-      }
-      c->mode = START;
-      break;
-    case LIT:           /* o: got literal, waiting for output space */
-      NEEDOUT
-      OUTBYTE(c->sub.lit)
-      c->mode = START;
-      break;
-    case WASH:          /* o: got eob, possibly more output */
-      FLUSH
-      if (s->read != s->write)
-        LEAVE
-      c->mode = END;
-    case END:
-      r = Z_STREAM_END;
-      LEAVE
-    case BADCODE:       /* x: got error */
-      r = Z_DATA_ERROR;
-      LEAVE
-    default:
-      r = Z_STREAM_ERROR;
-      LEAVE
-  }
-}
-
-
-local void inflate_codes_free(c, z)
-inflate_codes_statef *c;
-z_stream *z;
-{
-  ZFREE(z, c, sizeof(struct inflate_codes_state));
-  Tracev((stderr, "inflate:       codes free\n"));
-}
-
-/*+++++*/
-/* inflate_util.c -- data and routines common to blocks and codes
- * Copyright (C) 1995 Mark Adler
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* copy as much as possible from the sliding window to the output area */
-local int inflate_flush(s, z, r)
-inflate_blocks_statef *s;
-z_stream *z;
-int r;
-{
-  uInt n;
-  Bytef *p, *q;
-
-  /* local copies of source and destination pointers */
-  p = z->next_out;
-  q = s->read;
-
-  /* compute number of bytes to copy as far as end of window */
-  n = (uInt)((q <= s->write ? s->write : s->end) - q);
-  if (n > z->avail_out) n = z->avail_out;
-  if (n && r == Z_BUF_ERROR) r = Z_OK;
-
-  /* update counters */
-  z->avail_out -= n;
-  z->total_out += n;
-
-  /* update check information */
-  if (s->checkfn != Z_NULL)
-    s->check = (*s->checkfn)(s->check, q, n);
-
-  /* copy as far as end of window */
-  zmemcpy(p, q, n);
-  p += n;
-  q += n;
-
-  /* see if more to copy at beginning of window */
-  if (q == s->end)
-  {
-    /* wrap pointers */
-    q = s->window;
-    if (s->write == s->end)
-      s->write = s->window;
-
-    /* compute bytes to copy */
-    n = (uInt)(s->write - q);
-    if (n > z->avail_out) n = z->avail_out;
-    if (n && r == Z_BUF_ERROR) r = Z_OK;
-
-    /* update counters */
-    z->avail_out -= n;
-    z->total_out += n;
-
-    /* update check information */
-    if (s->checkfn != Z_NULL)
-      s->check = (*s->checkfn)(s->check, q, n);
-
-    /* copy */
-    zmemcpy(p, q, n);
-    p += n;
-    q += n;
-  }
-
-  /* update pointers */
-  z->next_out = p;
-  s->read = q;
-
-  /* done */
-  return r;
-}
-
-
-/*+++++*/
-/* inffast.c -- process literals and length/distance pairs fast
- * Copyright (C) 1995 Mark Adler
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* simplify the use of the inflate_huft type with some defines */
-#define base more.Base
-#define next more.Next
-#define exop word.what.Exop
-#define bits word.what.Bits
-
-/* macros for bit input with no checking and for returning unused bytes */
-#define GRABBITS(j) {while(k<(j)){b|=((uLong)NEXTBYTE)<<k;k+=8;}}
-#define UNGRAB {n+=(c=k>>3);p-=c;k&=7;}
-
-/* Called with number of bytes left to write in window at least 258
-   (the maximum string length) and number of input bytes available
-   at least ten.  The ten bytes are six bytes for the longest length/
-   distance pair plus four bytes for overloading the bit buffer. */
-
-local int inflate_fast(bl, bd, tl, td, s, z)
-uInt bl, bd;
-inflate_huft *tl, *td;
-inflate_blocks_statef *s;
-z_stream *z;
-{
-  inflate_huft *t;      /* temporary pointer */
-  uInt e;               /* extra bits or operation */
-  uLong b;              /* bit buffer */
-  uInt k;               /* bits in bit buffer */
-  Bytef *p;             /* input data pointer */
-  uInt n;               /* bytes available there */
-  Bytef *q;             /* output window write pointer */
-  uInt m;               /* bytes to end of window or read pointer */
-  uInt ml;              /* mask for literal/length tree */
-  uInt md;              /* mask for distance tree */
-  uInt c;               /* bytes to copy */
-  uInt d;               /* distance back to copy from */
-  Bytef *r;             /* copy source pointer */
-
-  /* load input, output, bit values */
-  LOAD
-
-  /* initialize masks */
-  ml = inflate_mask[bl];
-  md = inflate_mask[bd];
-
-  /* do until not enough input or output space for fast loop */
-  do {                          /* assume called with m >= 258 && n >= 10 */
-    /* get literal/length code */
-    GRABBITS(20)                /* max bits for literal/length code */
-    if ((e = (t = tl + ((uInt)b & ml))->exop) == 0)
-    {
-      DUMPBITS(t->bits)
-      Tracevv((stderr, t->base >= 0x20 && t->base < 0x7f ?
-                "inflate:         * literal '%c'\n" :
-                "inflate:         * literal 0x%02x\n", t->base));
-      *q++ = (Byte)t->base;
-      m--;
-      continue;
-    }
-    do {
-      DUMPBITS(t->bits)
-      if (e & 16)
-      {
-        /* get extra bits for length */
-        e &= 15;
-        c = t->base + ((uInt)b & inflate_mask[e]);
-        DUMPBITS(e)
-        Tracevv((stderr, "inflate:         * length %u\n", c));
-
-        /* decode distance base of block to copy */
-        GRABBITS(15);           /* max bits for distance code */
-        e = (t = td + ((uInt)b & md))->exop;
-        do {
-          DUMPBITS(t->bits)
-          if (e & 16)
-          {
-            /* get extra bits to add to distance base */
-            e &= 15;
-            GRABBITS(e)         /* get extra bits (up to 13) */
-            d = t->base + ((uInt)b & inflate_mask[e]);
-            DUMPBITS(e)
-            Tracevv((stderr, "inflate:         * distance %u\n", d));
-
-            /* do the copy */
-            m -= c;
-            if ((uInt)(q - s->window) >= d)     /* offset before dest */
-            {                                   /*  just copy */
-              r = q - d;
-              *q++ = *r++;  c--;        /* minimum count is three, */
-              *q++ = *r++;  c--;        /*  so unroll loop a little */
-            }
-            else                        /* else offset after destination */
-            {
-              e = d - (q - s->window);  /* bytes from offset to end */
-              r = s->end - e;           /* pointer to offset */
-              if (c > e)                /* if source crosses, */
-              {
-                c -= e;                 /* copy to end of window */
-                do {
-                  *q++ = *r++;
-                } while (--e);
-                r = s->window;          /* copy rest from start of window */
-              }
-            }
-            do {                        /* copy all or what's left */
-              *q++ = *r++;
-            } while (--c);
-            break;
-          }
-          else if ((e & 64) == 0)
-            e = (t = t->next + ((uInt)b & inflate_mask[e]))->exop;
-          else
-          {
-            z->msg = "invalid distance code";
-            UNGRAB
-            UPDATE
-            return Z_DATA_ERROR;
-          }
-        } while (1);
-        break;
-      }
-      if ((e & 64) == 0)
-      {
-        if ((e = (t = t->next + ((uInt)b & inflate_mask[e]))->exop) == 0)
-        {
-          DUMPBITS(t->bits)
-          Tracevv((stderr, t->base >= 0x20 && t->base < 0x7f ?
-                    "inflate:         * literal '%c'\n" :
-                    "inflate:         * literal 0x%02x\n", t->base));
-          *q++ = (Byte)t->base;
-          m--;
-          break;
-        }
-      }
-      else if (e & 32)
-      {
-        Tracevv((stderr, "inflate:         * end of block\n"));
-        UNGRAB
-        UPDATE
-        return Z_STREAM_END;
-      }
-      else
-      {
-        z->msg = "invalid literal/length code";
-        UNGRAB
-        UPDATE
-        return Z_DATA_ERROR;
-      }
-    } while (1);
-  } while (m >= 258 && n >= 10);
-
-  /* not enough input or output--restore pointers and return */
-  UNGRAB
-  UPDATE
-  return Z_OK;
-}
-
-
-/*+++++*/
-/* zutil.c -- target dependent utility functions for the compression library
- * Copyright (C) 1995 Jean-loup Gailly.
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* From: zutil.c,v 1.8 1995/05/03 17:27:12 jloup Exp */
-
-char *zlib_version = ZLIB_VERSION;
-
-char *z_errmsg[] = {
-"stream end",          /* Z_STREAM_END    1 */
-"",                    /* Z_OK            0 */
-"file error",          /* Z_ERRNO        (-1) */
-"stream error",        /* Z_STREAM_ERROR (-2) */
-"data error",          /* Z_DATA_ERROR   (-3) */
-"insufficient memory", /* Z_MEM_ERROR    (-4) */
-"buffer error",        /* Z_BUF_ERROR    (-5) */
-""};
-
-
-/*+++++*/
-/* adler32.c -- compute the Adler-32 checksum of a data stream
- * Copyright (C) 1995 Mark Adler
- * For conditions of distribution and use, see copyright notice in zlib.h
- */
-
-/* From: adler32.c,v 1.6 1995/05/03 17:27:08 jloup Exp */
-
-#define BASE 65521L /* largest prime smaller than 65536 */
-#define NMAX 5552
-/* NMAX is the largest n such that 255n(n+1)/2 + (n+1)(BASE-1) <= 2^32-1 */
-
-#define DO1(buf)  {s1 += *buf++; s2 += s1;}
-#define DO2(buf)  DO1(buf); DO1(buf);
-#define DO4(buf)  DO2(buf); DO2(buf);
-#define DO8(buf)  DO4(buf); DO4(buf);
-#define DO16(buf) DO8(buf); DO8(buf);
-
-/* ========================================================================= */
-uLong adler32(adler, buf, len)
-    uLong adler;
-    Bytef *buf;
-    uInt len;
-{
-    unsigned long s1 = adler & 0xffff;
-    unsigned long s2 = (adler >> 16) & 0xffff;
-    int k;
-
-    if (buf == Z_NULL) return 1L;
-
-    while (len > 0) {
-        k = len < NMAX ? len : NMAX;
-        len -= k;
-        while (k >= 16) {
-            DO16(buf);
-            k -= 16;
-        }
-        if (k != 0) do {
-            DO1(buf);
-        } while (--k);
-        s1 %= BASE;
-        s2 %= BASE;
-    }
-    return (s2 << 16) | s1;
-}
diff --git a/arch/xtensa/boot/lib/zmem.c b/arch/xtensa/boot/lib/zmem.c
index 7848f126d67..d9862aa8ca2 100644
--- a/arch/xtensa/boot/lib/zmem.c
+++ b/arch/xtensa/boot/lib/zmem.c
@@ -1,4 +1,4 @@
-#include "zlib.h"
+#include <linux/zlib.h>
 
 /* bits taken from ppc */
 
@@ -9,11 +9,10 @@ void exit (void)
   for (;;);
 }
 
-void *zalloc(void *x, unsigned items, unsigned size)
+void *zalloc(unsigned size)
 {
         void *p = avail_ram;
 
-        size *= items;
         size = (size + 7) & -8;
         avail_ram += size;
         if (avail_ram > end_avail) {
@@ -24,11 +23,6 @@ void *zalloc(void *x, unsigned items, unsigned size)
         return p;
 }
 
-void zfree(void *x, void *addr, unsigned nb)
-{
-}
-
-
 #define HEAD_CRC        2
 #define EXTRA_FIELD     4
 #define ORIG_NAME       8
@@ -43,7 +37,6 @@ void gunzip (void *dst, int dstlen, unsigned char *src, int *lenp)
 	int r, i, flags;
 
         /* skip header */
-
         i = 10;
         flags = src[3];
         if (src[2] != DEFLATED || (flags & RESERVED) != 0) {
@@ -65,9 +58,8 @@ void gunzip (void *dst, int dstlen, unsigned char *src, int *lenp)
                 exit();
         }
 
-        s.zalloc = zalloc;
-        s.zfree = zfree;
-        r = inflateInit2(&s, -MAX_WBITS);
+	s.workspace = zalloc(zlib_inflate_workspacesize());
+        r = zlib_inflateInit2(&s, -MAX_WBITS);
         if (r != Z_OK) {
                 //puts("inflateInit2 returned "); puthex(r); puts("\n");
                 exit();
@@ -76,12 +68,12 @@ void gunzip (void *dst, int dstlen, unsigned char *src, int *lenp)
         s.avail_in = *lenp - i;
         s.next_out = dst;
         s.avail_out = dstlen;
-        r = inflate(&s, Z_FINISH);
+        r = zlib_inflate(&s, Z_FINISH);
         if (r != Z_OK && r != Z_STREAM_END) {
                 //puts("inflate returned "); puthex(r); puts("\n");
                 exit();
         }
         *lenp = s.next_out - (unsigned char *) dst;
-        inflateEnd(&s);
+        zlib_inflateEnd(&s);
 }
 
-- 
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 'arch')

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


From c19cb1df809dcf343dd1eb6fe60d53639dafcb8c Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Thu, 30 Jun 2005 17:04:13 +0100
Subject: [PATCH] ARM: 2777/1: Fix broken comment arch/arm/mm/proc-arm1020.S

Patch from Catalin Marinas

This patch fixes a broken comment in the proc-arm1020.S file which
prevents the file compilation

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/proc-arm1020.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/arm/mm/proc-arm1020.S b/arch/arm/mm/proc-arm1020.S
index 1f325231b9e..e69f1940ab3 100644
--- a/arch/arm/mm/proc-arm1020.S
+++ b/arch/arm/mm/proc-arm1020.S
@@ -445,7 +445,7 @@ __arm1020_setup:
 	/*
 	 *  R
 	 * .RVI ZFRS BLDP WCAM
-	 * .0.1 1001 ..11 0101	/* FIXME: why no V bit? */
+	 * .0.1 1001 ..11 0101	FIXME: why no V bit?
 	 */
 	.type	arm1020_cr1_clear, #object
 	.type	arm1020_cr1_set, #object
-- 
cgit v1.2.3-70-g09d2


From c28a814f25d48f193565003223df0ae617796892 Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Thu, 30 Jun 2005 17:04:14 +0100
Subject: [PATCH] ARM: 2778/1: Add -mno-thumb-interwork to CFLAGS_ABI

Patch from Catalin Marinas

The new EABI gcc adds -mthumb-interwork by default, even if
-mabi=apcs-gnu is passed. This causes a warning for every compiled C
file when -march=armv4 is used. The patch adds -mno-thumb-interwork
if the option is supported. This is also useful since we don't need
any ARM/Thumb interworking in the kernel

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index 8330495e244..eb933dcafba 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -56,7 +56,7 @@ tune-$(CONFIG_CPU_XSCALE)	:=$(call cc-option,-mtune=xscale,-mtune=strongarm110)
 tune-$(CONFIG_CPU_V6)		:=-mtune=strongarm
 
 # Need -Uarm for gcc < 3.x
-CFLAGS_ABI	:=$(call cc-option,-mapcs-32,-mabi=apcs-gnu)
+CFLAGS_ABI	:=$(call cc-option,-mapcs-32,-mabi=apcs-gnu) $(call cc-option,-mno-thumb-interwork,)
 CFLAGS		+=$(CFLAGS_ABI) $(arch-y) $(tune-y) $(call cc-option,-mshort-load-bytes,$(call cc-option,-malignment-traps,)) -msoft-float -Uarm
 AFLAGS		+=$(CFLAGS_ABI) $(arch-y) $(tune-y) -msoft-float
 
-- 
cgit v1.2.3-70-g09d2


From abaf48a05a8f097654e746af2a5afb2ab95861a1 Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Thu, 30 Jun 2005 17:04:14 +0100
Subject: [PATCH] ARM: 2779/1: Fix the V bit setting for the ARM1020x CPUs

Patch from Catalin Marinas

This patch fixes the V bit setting for the ARM1020x processors. At
reset, this bit is automatically set to the value of the HIVECSINIT
input signal which just happened to be 1 but it is not mandatory.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/proc-arm1020.S  | 4 ++--
 arch/arm/mm/proc-arm1020e.S | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'arch')

diff --git a/arch/arm/mm/proc-arm1020.S b/arch/arm/mm/proc-arm1020.S
index e69f1940ab3..5c0ae5260d1 100644
--- a/arch/arm/mm/proc-arm1020.S
+++ b/arch/arm/mm/proc-arm1020.S
@@ -445,14 +445,14 @@ __arm1020_setup:
 	/*
 	 *  R
 	 * .RVI ZFRS BLDP WCAM
-	 * .0.1 1001 ..11 0101	FIXME: why no V bit?
+	 * .011 1001 ..11 0101
 	 */
 	.type	arm1020_cr1_clear, #object
 	.type	arm1020_cr1_set, #object
 arm1020_cr1_clear:
 	.word	0x593f
 arm1020_cr1_set:
-	.word	0x1935
+	.word	0x3935
 
 	__INITDATA
 
diff --git a/arch/arm/mm/proc-arm1020e.S b/arch/arm/mm/proc-arm1020e.S
index 142a2c2d6f0..d69389c4d4b 100644
--- a/arch/arm/mm/proc-arm1020e.S
+++ b/arch/arm/mm/proc-arm1020e.S
@@ -427,14 +427,14 @@ __arm1020e_setup:
 	/*
 	 *  R
 	 * .RVI ZFRS BLDP WCAM
-	 * .0.1 1001 ..11 0101	/* FIXME: why no V bit? */
+	 * .011 1001 ..11 0101
 	 */
 	.type	arm1020e_cr1_clear, #object
 	.type	arm1020e_cr1_set, #object
 arm1020e_cr1_clear:
 	.word	0x5f3f
 arm1020e_cr1_set:
-	.word	0x1935
+	.word	0x3935
 
 	__INITDATA
 
-- 
cgit v1.2.3-70-g09d2


From eaf05be039cf5adfba5b1846452ce89646110fdb Mon Sep 17 00:00:00 2001
From: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Date: Thu, 30 Jun 2005 20:02:18 +0400
Subject: [PATCH] alpha smp fix

As usual, the reason of this breakage is quite silly: in do_entIF, we
are checking for PS == 0 to see whether it was a kernel BUG() or
userspace trap.

It works, unless BUG() happens in interrupt - PS is not 0 in kernel mode
due to non-zero IPL, and the things get messed up horribly then.  In
this particular case it was BUG_ON(!irqs_disabled()) triggered in
run_posix_cpu_timers(), so we ended up shooting "current" with the
bursts of one SIGTRAP and three SIGILLs on every timer tick.  ;-)
---
 arch/alpha/kernel/traps.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/alpha/kernel/traps.c b/arch/alpha/kernel/traps.c
index fd7bd17cc96..6f509a644bd 100644
--- a/arch/alpha/kernel/traps.c
+++ b/arch/alpha/kernel/traps.c
@@ -240,7 +240,7 @@ do_entIF(unsigned long type, struct pt_regs *regs)
 	siginfo_t info;
 	int signo, code;
 
-	if (regs->ps == 0) {
+	if ((regs->ps & ~IPL_MAX) == 0) {
 		if (type == 1) {
 			const unsigned int *data
 			  = (const unsigned int *) regs->pc;
-- 
cgit v1.2.3-70-g09d2


From 4a89a04f1ee21a7c1f4413f1ad7dcfac50ff9b63 Mon Sep 17 00:00:00 2001
From: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Date: Fri, 1 Jul 2005 16:46:26 +0400
Subject: [PATCH] alpha smp fix (part #2)

This fixes the bug that caused BUG_ON(!irqs_disabled()) to trigger in
run_posix_cpu_timers() on alpha/smp.  We didn't disable interrupts
properly before calling smp_percpu_timer_interrupt().

We *do* disable interrupts everywhere except this unfortunate
smp_percpu_timer_interrupt().  Fixed thus.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/alpha/kernel/irq_alpha.c | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'arch')

diff --git a/arch/alpha/kernel/irq_alpha.c b/arch/alpha/kernel/irq_alpha.c
index e6ded33c6e2..9d34ce26e5e 100644
--- a/arch/alpha/kernel/irq_alpha.c
+++ b/arch/alpha/kernel/irq_alpha.c
@@ -55,6 +55,8 @@ do_entInt(unsigned long type, unsigned long vector,
 #ifdef CONFIG_SMP
 	  {
 		long cpu;
+
+		local_irq_disable();
 		smp_percpu_timer_interrupt(regs);
 		cpu = smp_processor_id();
 		if (cpu != boot_cpuid) {
-- 
cgit v1.2.3-70-g09d2