Merge branch 'core/urgent' into core/futexes

Merge reason: this branch was on an pre -rc1 base, merge it up to -rc6+
              to get the latest upstream fixes.

Conflicts:
	kernel/futex.c

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

23 files changed, 133 insertions, 71 deletions

diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c
index cd1b687544f..3e33fb933d9 100644
--- a/arch/powerpc/kernel/cputable.c
+++ b/arch/powerpc/kernel/cputable.c
@@ -382,7 +382,8 @@ static struct cpu_spec __initdata cpu_specs[] = {
 		.icache_bsize		= 128,
 		.dcache_bsize		= 128,
 		.machine_check		= machine_check_generic,
-		.oprofile_cpu_type	= "ppc64/compat-power5+",
+		.oprofile_cpu_type	= "ppc64/ibm-compat-v1",
+		.oprofile_type		= PPC_OPROFILE_POWER4,
 		.platform		= "power5+",
 	},
 	{	/* Power6 */
@@ -416,7 +417,8 @@ static struct cpu_spec __initdata cpu_specs[] = {
 		.icache_bsize		= 128,
 		.dcache_bsize		= 128,
 		.machine_check		= machine_check_generic,
-		.oprofile_cpu_type	= "ppc64/compat-power6",
+		.oprofile_cpu_type	= "ppc64/ibm-compat-v1",
+		.oprofile_type		= PPC_OPROFILE_POWER4,
 		.platform		= "power6",
 	},
 	{	/* 2.06-compliant processor, i.e. Power7 "architected" mode */
@@ -429,7 +431,8 @@ static struct cpu_spec __initdata cpu_specs[] = {
 		.icache_bsize		= 128,
 		.dcache_bsize		= 128,
 		.machine_check		= machine_check_generic,
-		.oprofile_cpu_type	= "ppc64/compat-power7",
+		.oprofile_type		= PPC_OPROFILE_POWER4,
+		.oprofile_cpu_type	= "ppc64/ibm-compat-v1",
 		.platform		= "power7",
 	},
 	{	/* Power7 */
@@ -1833,8 +1836,10 @@ static void __init setup_cpu_spec(unsigned long offset, struct cpu_spec *s)
 		 * and, in that case, keep the current value for
 		 * oprofile_cpu_type.
 		 */
-		if (old.oprofile_cpu_type == NULL)
-			t->oprofile_cpu_type = s->oprofile_cpu_type;
+		if (old.oprofile_cpu_type != NULL) {
+			t->oprofile_cpu_type = old.oprofile_cpu_type;
+			t->oprofile_type = old.oprofile_type;
+		}
 	}
 
 	*PTRRELOC(&cur_cpu_spec) = &the_cpu_spec;
diff --git a/arch/powerpc/kernel/dma.c b/arch/powerpc/kernel/dma.c
index 1c5c8a6fc12..53c7788cba7 100644
--- a/arch/powerpc/kernel/dma.c
+++ b/arch/powerpc/kernel/dma.c
@@ -94,7 +94,7 @@ static int dma_direct_dma_supported(struct device *dev, u64 mask)
 	 * done via some global so platforms can set the limit in case
 	 * they have limited DMA windows
 	 */
-	return mask >= DMA_32BIT_MASK;
+	return mask >= DMA_BIT_MASK(32);
 #else
 	return 1;
 #endif
diff --git a/arch/powerpc/kernel/fpu.S b/arch/powerpc/kernel/fpu.S
index a088c064ae4..2436df33c6f 100644
--- a/arch/powerpc/kernel/fpu.S
+++ b/arch/powerpc/kernel/fpu.S
@@ -145,6 +145,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
 	beq	1f
 	PPC_LL	r4,_MSR-STACK_FRAME_OVERHEAD(r5)
 	li	r3,MSR_FP|MSR_FE0|MSR_FE1
+#ifdef CONFIG_VSX
+BEGIN_FTR_SECTION
+	oris	r3,r3,MSR_VSX@h
+END_FTR_SECTION_IFSET(CPU_FTR_VSX)
+#endif
 	andc	r4,r4,r3		/* disable FP for previous task */
 	PPC_STL	r4,_MSR-STACK_FRAME_OVERHEAD(r5)
 1:
diff --git a/arch/powerpc/kernel/ftrace.c b/arch/powerpc/kernel/ftrace.c
index 5b5d16b2cac..2d182f119d1 100644
--- a/arch/powerpc/kernel/ftrace.c
+++ b/arch/powerpc/kernel/ftrace.c
@@ -157,7 +157,7 @@ __ftrace_make_nop(struct module *mod,
 	 * 0xe8, 0x4c, 0x00, 0x28,    ld      r2,40(r12)
 	 */
 
-	pr_debug("ip:%lx jumps to %lx r2: %lx", ip, tramp, mod->arch.toc);
+	pr_devel("ip:%lx jumps to %lx r2: %lx", ip, tramp, mod->arch.toc);
 
 	/* Find where the trampoline jumps to */
 	if (probe_kernel_read(jmp, (void *)tramp, sizeof(jmp))) {
@@ -165,7 +165,7 @@ __ftrace_make_nop(struct module *mod,
 		return -EFAULT;
 	}
 
-	pr_debug(" %08x %08x", jmp[0], jmp[1]);
+	pr_devel(" %08x %08x", jmp[0], jmp[1]);
 
 	/* verify that this is what we expect it to be */
 	if (((jmp[0] & 0xffff0000) != 0x3d820000) ||
@@ -181,18 +181,18 @@ __ftrace_make_nop(struct module *mod,
 	offset = ((unsigned)((unsigned short)jmp[0]) << 16) +
 		(int)((short)jmp[1]);
 
-	pr_debug(" %x ", offset);
+	pr_devel(" %x ", offset);
 
 	/* get the address this jumps too */
 	tramp = mod->arch.toc + offset + 32;
-	pr_debug("toc: %lx", tramp);
+	pr_devel("toc: %lx", tramp);
 
 	if (probe_kernel_read(jmp, (void *)tramp, 8)) {
 		printk(KERN_ERR "Failed to read %lx\n", tramp);
 		return -EFAULT;
 	}
 
-	pr_debug(" %08x %08x\n", jmp[0], jmp[1]);
+	pr_devel(" %08x %08x\n", jmp[0], jmp[1]);
 
 	ptr = ((unsigned long)jmp[0] << 32) + jmp[1];
 
@@ -269,7 +269,7 @@ __ftrace_make_nop(struct module *mod,
 	 *  0x4e, 0x80, 0x04, 0x20  bctr
 	 */
 
-	pr_debug("ip:%lx jumps to %lx", ip, tramp);
+	pr_devel("ip:%lx jumps to %lx", ip, tramp);
 
 	/* Find where the trampoline jumps to */
 	if (probe_kernel_read(jmp, (void *)tramp, sizeof(jmp))) {
@@ -277,7 +277,7 @@ __ftrace_make_nop(struct module *mod,
 		return -EFAULT;
 	}
 
-	pr_debug(" %08x %08x ", jmp[0], jmp[1]);
+	pr_devel(" %08x %08x ", jmp[0], jmp[1]);
 
 	/* verify that this is what we expect it to be */
 	if (((jmp[0] & 0xffff0000) != 0x3d600000) ||
@@ -293,7 +293,7 @@ __ftrace_make_nop(struct module *mod,
 	if (tramp & 0x8000)
 		tramp -= 0x10000;
 
-	pr_debug(" %x ", tramp);
+	pr_devel(" %lx ", tramp);
 
 	if (tramp != addr) {
 		printk(KERN_ERR
@@ -402,7 +402,7 @@ __ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr)
 	/* ld r2,40(r1) */
 	op[1] = 0xe8410028;
 
-	pr_debug("write to %lx\n", rec->ip);
+	pr_devel("write to %lx\n", rec->ip);
 
 	if (probe_kernel_write((void *)ip, op, MCOUNT_INSN_SIZE * 2))
 		return -EPERM;
@@ -442,7 +442,7 @@ __ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr)
 		return -EINVAL;
 	}
 
-	pr_debug("write to %lx\n", rec->ip);
+	pr_devel("write to %lx\n", rec->ip);
 
 	if (probe_kernel_write((void *)ip, &op, MCOUNT_INSN_SIZE))
 		return -EPERM;
@@ -557,7 +557,6 @@ extern void mod_return_to_handler(void);
 void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr)
 {
 	unsigned long old;
-	unsigned long long calltime;
 	int faulted;
 	struct ftrace_graph_ent trace;
 	unsigned long return_hooker = (unsigned long)&return_to_handler;
@@ -565,7 +564,7 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr)
 	if (unlikely(atomic_read(&current->tracing_graph_pause)))
 		return;
 
-#if CONFIG_PPC64
+#ifdef CONFIG_PPC64
 	/* non core kernel code needs to save and restore the TOC */
 	if (REGION_ID(self_addr) != KERNEL_REGION_ID)
 		return_hooker = (unsigned long)&mod_return_to_handler;
@@ -595,7 +594,7 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr)
 			PPC_LONG "2b,4b\n"
 		".previous"
 
-		: [old] "=r" (old), [faulted] "=r" (faulted)
+		: [old] "=&r" (old), [faulted] "=r" (faulted)
 		: [parent] "r" (parent), [return_hooker] "r" (return_hooker)
 		: "memory"
 	);
@@ -606,10 +605,7 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr)
 		return;
 	}
 
-	calltime = cpu_clock(raw_smp_processor_id());
-
-	if (ftrace_push_return_trace(old, calltime,
-				self_addr, &trace.depth) == -EBUSY) {
+	if (ftrace_push_return_trace(old, self_addr, &trace.depth) == -EBUSY) {
 		*parent = old;
 		return;
 	}
diff --git a/arch/powerpc/kernel/head_32.S b/arch/powerpc/kernel/head_32.S
index 54e68c11ae1..c01467f952d 100644
--- a/arch/powerpc/kernel/head_32.S
+++ b/arch/powerpc/kernel/head_32.S
@@ -21,6 +21,7 @@
  *
  */
 
+#include <linux/init.h>
 #include <asm/reg.h>
 #include <asm/page.h>
 #include <asm/mmu.h>
@@ -50,7 +51,7 @@
 	mtspr	SPRN_DBAT##n##L,RB;	\
 1:
 
-	.section	.text.head, "ax"
+	__HEAD
 	.stabs	"arch/powerpc/kernel/",N_SO,0,0,0f
 	.stabs	"head_32.S",N_SO,0,0,0f
 0:
diff --git a/arch/powerpc/kernel/head_40x.S b/arch/powerpc/kernel/head_40x.S
index 56d8e5d90c5..0c96911d429 100644
--- a/arch/powerpc/kernel/head_40x.S
+++ b/arch/powerpc/kernel/head_40x.S
@@ -31,6 +31,7 @@
  *
  */
 
+#include <linux/init.h>
 #include <asm/processor.h>
 #include <asm/page.h>
 #include <asm/mmu.h>
@@ -52,7 +53,7 @@
  *
  * This is all going to change RSN when we add bi_recs.......  -- Dan
  */
-	.section	.text.head, "ax"
+	__HEAD
 _ENTRY(_stext);
 _ENTRY(_start);
 
diff --git a/arch/powerpc/kernel/head_44x.S b/arch/powerpc/kernel/head_44x.S
index b56fecc93a1..18d8a1677c4 100644
--- a/arch/powerpc/kernel/head_44x.S
+++ b/arch/powerpc/kernel/head_44x.S
@@ -28,6 +28,7 @@
  * option) any later version.
  */
 
+#include <linux/init.h>
 #include <asm/processor.h>
 #include <asm/page.h>
 #include <asm/mmu.h>
@@ -50,7 +51,7 @@
  *   r7 - End of kernel command line string
  *
  */
-	.section	.text.head, "ax"
+	__HEAD
 _ENTRY(_stext);
 _ENTRY(_start);
 	/*
diff --git a/arch/powerpc/kernel/head_8xx.S b/arch/powerpc/kernel/head_8xx.S
index 3c9452d4308..52ff8c53b93 100644
--- a/arch/powerpc/kernel/head_8xx.S
+++ b/arch/powerpc/kernel/head_8xx.S
@@ -19,6 +19,7 @@
  *
  */
 
+#include <linux/init.h>
 #include <asm/processor.h>
 #include <asm/page.h>
 #include <asm/mmu.h>
@@ -38,7 +39,7 @@
 #else
 #define DO_8xx_CPU6(val, reg)
 #endif
-	.section	.text.head, "ax"
+	__HEAD
 _ENTRY(_stext);
 _ENTRY(_start);
 
diff --git a/arch/powerpc/kernel/head_fsl_booke.S b/arch/powerpc/kernel/head_fsl_booke.S
index 4c22620d009..5bdcc06d294 100644
--- a/arch/powerpc/kernel/head_fsl_booke.S
+++ b/arch/powerpc/kernel/head_fsl_booke.S
@@ -30,6 +30,7 @@
  * option) any later version.
  */
 
+#include <linux/init.h>
 #include <linux/threads.h>
 #include <asm/processor.h>
 #include <asm/page.h>
@@ -53,7 +54,7 @@
  *   r7 - End of kernel command line string
  *
  */
-	.section	.text.head, "ax"
+	__HEAD
 _ENTRY(_stext);
 _ENTRY(_start);
 	/*
diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c
index 5576147e57b..8c1a4966867 100644
--- a/arch/powerpc/kernel/irq.c
+++ b/arch/powerpc/kernel/irq.c
@@ -672,11 +672,13 @@ unsigned int irq_create_mapping(struct irq_host *host,
 			return NO_IRQ;
 		}
 	}
-	pr_debug("irq: -> obtained virq %d\n", virq);
 
 	if (irq_setup_virq(host, virq, hwirq))
 		return NO_IRQ;
 
+	printk(KERN_DEBUG "irq: irq %lu on host %s mapped to virtual irq %u\n",
+		hwirq, host->of_node ? host->of_node->full_name : "null", virq);
+
 	return virq;
 }
 EXPORT_SYMBOL_GPL(irq_create_mapping);
diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c
index 182e0f642f3..78b3f7840ad 100644
--- a/arch/powerpc/kernel/lparcfg.c
+++ b/arch/powerpc/kernel/lparcfg.c
@@ -434,6 +434,21 @@ static void pseries_cmo_data(struct seq_file *m)
 	seq_printf(m, "cmo_page_size=%lu\n", cmo_get_page_size());
 }
 
+static void splpar_dispatch_data(struct seq_file *m)
+{
+	int cpu;
+	unsigned long dispatches = 0;
+	unsigned long dispatch_dispersions = 0;
+
+	for_each_possible_cpu(cpu) {
+		dispatches += lppaca[cpu].yield_count;
+		dispatch_dispersions += lppaca[cpu].dispersion_count;
+	}
+
+	seq_printf(m, "dispatches=%lu\n", dispatches);
+	seq_printf(m, "dispatch_dispersions=%lu\n", dispatch_dispersions);
+}
+
 static int pseries_lparcfg_data(struct seq_file *m, void *v)
 {
 	int partition_potential_processors;
@@ -460,6 +475,7 @@ static int pseries_lparcfg_data(struct seq_file *m, void *v)
 		parse_ppp_data(m);
 		parse_mpp_data(m);
 		pseries_cmo_data(m);
+		splpar_dispatch_data(m);
 
 		seq_printf(m, "purr=%ld\n", get_purr());
 	} else {		/* non SPLPAR case */
diff --git a/arch/powerpc/kernel/machine_kexec.c b/arch/powerpc/kernel/machine_kexec.c
index d59e2b1bdcb..bb3d893a835 100644
--- a/arch/powerpc/kernel/machine_kexec.c
+++ b/arch/powerpc/kernel/machine_kexec.c
@@ -125,8 +125,8 @@ void __init reserve_crashkernel(void)
 	/* Crash kernel trumps memory limit */
 	if (memory_limit && memory_limit <= crashk_res.end) {
 		memory_limit = crashk_res.end + 1;
-		printk("Adjusted memory limit for crashkernel, now 0x%lx\n",
-				memory_limit);
+		printk("Adjusted memory limit for crashkernel, now 0x%llx\n",
+		       (unsigned long long)memory_limit);
 	}
 
 	printk(KERN_INFO "Reserving %ldMB of memory at %ldMB "
diff --git a/arch/powerpc/kernel/misc_64.S b/arch/powerpc/kernel/misc_64.S
index 3053fe5c62f..b9530b2395a 100644
--- a/arch/powerpc/kernel/misc_64.S
+++ b/arch/powerpc/kernel/misc_64.S
@@ -495,7 +495,15 @@ _GLOBAL(giveup_altivec)
 	stvx	vr0,r4,r3
 	beq	1f
 	ld	r4,_MSR-STACK_FRAME_OVERHEAD(r5)
+#ifdef CONFIG_VSX
+BEGIN_FTR_SECTION
+	lis	r3,(MSR_VEC|MSR_VSX)@h
+FTR_SECTION_ELSE
+	lis	r3,MSR_VEC@h
+ALT_FTR_SECTION_END_IFSET(CPU_FTR_VSX)
+#else
 	lis	r3,MSR_VEC@h
+#endif
 	andc	r4,r4,r3		/* disable FP for previous task */
 	std	r4,_MSR-STACK_FRAME_OVERHEAD(r5)
 1:
diff --git a/arch/powerpc/kernel/of_platform.c b/arch/powerpc/kernel/of_platform.c
index 3f37a6e6277..87df428e358 100644
--- a/arch/powerpc/kernel/of_platform.c
+++ b/arch/powerpc/kernel/of_platform.c
@@ -76,7 +76,7 @@ struct of_device* of_platform_device_create(struct device_node *np,
 		return NULL;
 
 	dev->dma_mask = 0xffffffffUL;
-	dev->dev.coherent_dma_mask = DMA_32BIT_MASK;
+	dev->dev.coherent_dma_mask = DMA_BIT_MASK(32);
 
 	dev->dev.bus = &of_platform_bus_type;
 
diff --git a/arch/powerpc/kernel/pci-common.c b/arch/powerpc/kernel/pci-common.c
index 9c69e7e145c..4fee63cb53f 100644
--- a/arch/powerpc/kernel/pci-common.c
+++ b/arch/powerpc/kernel/pci-common.c
@@ -1366,12 +1366,17 @@ static void __init pcibios_allocate_resources(int pass)
 
 	for_each_pci_dev(dev) {
 		pci_read_config_word(dev, PCI_COMMAND, &command);
-		for (idx = 0; idx < 6; idx++) {
+		for (idx = 0; idx <= PCI_ROM_RESOURCE; idx++) {
 			r = &dev->resource[idx];
 			if (r->parent)		/* Already allocated */
 				continue;
 			if (!r->flags || (r->flags & IORESOURCE_UNSET))
 				continue;	/* Not assigned at all */
+			/* We only allocate ROMs on pass 1 just in case they
+			 * have been screwed up by firmware
+			 */
+			if (idx == PCI_ROM_RESOURCE )
+				disabled = 1;
 			if (r->flags & IORESOURCE_IO)
 				disabled = !(command & PCI_COMMAND_IO);
 			else
@@ -1382,17 +1387,19 @@ static void __init pcibios_allocate_resources(int pass)
 		if (pass)
 			continue;
 		r = &dev->resource[PCI_ROM_RESOURCE];
-		if (r->flags & IORESOURCE_ROM_ENABLE) {
+		if (r->flags) {
 			/* Turn the ROM off, leave the resource region,
 			 * but keep it unregistered.
 			 */
 			u32 reg;
-			pr_debug("PCI: Switching off ROM of %s\n",
-				 pci_name(dev));
-			r->flags &= ~IORESOURCE_ROM_ENABLE;
 			pci_read_config_dword(dev, dev->rom_base_reg, &reg);
-			pci_write_config_dword(dev, dev->rom_base_reg,
-					       reg & ~PCI_ROM_ADDRESS_ENABLE);
+			if (reg & PCI_ROM_ADDRESS_ENABLE) {
+				pr_debug("PCI: Switching off ROM of %s\n",
+					 pci_name(dev));
+				r->flags &= ~IORESOURCE_ROM_ENABLE;
+				pci_write_config_dword(dev, dev->rom_base_reg,
+						       reg & ~PCI_ROM_ADDRESS_ENABLE);
+			}
 		}
 	}
 }
diff --git a/arch/powerpc/kernel/pci_64.c b/arch/powerpc/kernel/pci_64.c
index be574fc0d92..96edb6f8bab 100644
--- a/arch/powerpc/kernel/pci_64.c
+++ b/arch/powerpc/kernel/pci_64.c
@@ -64,7 +64,7 @@ static u32 get_int_prop(struct device_node *np, const char *name, u32 def)
 	return def;
 }
 
-static unsigned int pci_parse_of_flags(u32 addr0)
+static unsigned int pci_parse_of_flags(u32 addr0, int bridge)
 {
 	unsigned int flags = 0;
 
@@ -75,8 +75,17 @@ static unsigned int pci_parse_of_flags(u32 addr0)
 		if (addr0 & 0x40000000)
 			flags |= IORESOURCE_PREFETCH
 				 | PCI_BASE_ADDRESS_MEM_PREFETCH;
+		/* Note: We don't know whether the ROM has been left enabled
+		 * by the firmware or not. We mark it as disabled (ie, we do
+		 * not set the IORESOURCE_ROM_ENABLE flag) for now rather than
+		 * do a config space read, it will be force-enabled if needed
+		 */
+		if (!bridge && (addr0 & 0xff) == 0x30)
+			flags |= IORESOURCE_READONLY;
 	} else if (addr0 & 0x01000000)
 		flags = IORESOURCE_IO | PCI_BASE_ADDRESS_SPACE_IO;
+	if (flags)
+		flags |= IORESOURCE_SIZEALIGN;
 	return flags;
 }
 
@@ -95,7 +104,7 @@ static void pci_parse_of_addrs(struct device_node *node, struct pci_dev *dev)
 		return;
 	pr_debug("    parse addresses (%d bytes) @ %p\n", proplen, addrs);
 	for (; proplen >= 20; proplen -= 20, addrs += 5) {
-		flags = pci_parse_of_flags(addrs[0]);
+		flags = pci_parse_of_flags(addrs[0], 0);
 		if (!flags)
 			continue;
 		base = of_read_number(&addrs[1], 2);
@@ -293,7 +302,7 @@ void __devinit of_scan_pci_bridge(struct device_node *node,
 	}
 	i = 1;
 	for (; len >= 32; len -= 32, ranges += 8) {
-		flags = pci_parse_of_flags(ranges[0]);
+		flags = pci_parse_of_flags(ranges[0], 1);
 		size = of_read_number(&ranges[6], 2);
 		if (flags == 0 || size == 0)
 			continue;
diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
index 5ec6a9e2393..ce01ff2474d 100644
--- a/arch/powerpc/kernel/prom.c
+++ b/arch/powerpc/kernel/prom.c
@@ -426,7 +426,7 @@ static int __init early_parse_mem(char *p)
 		return 1;
 
 	memory_limit = PAGE_ALIGN(memparse(p, &p));
-	DBG("memory limit = 0x%lx\n", memory_limit);
+	DBG("memory limit = 0x%llx\n", (unsigned long long)memory_limit);
 
 	return 0;
 }
@@ -1160,7 +1160,7 @@ static inline void __init phyp_dump_reserve_mem(void) {}
 
 void __init early_init_devtree(void *params)
 {
-	unsigned long limit;
+	phys_addr_t limit;
 
 	DBG(" -> early_init_devtree(%p)\n", params);
 
@@ -1204,7 +1204,7 @@ void __init early_init_devtree(void *params)
 
 	limit = memory_limit;
 	if (! limit) {
-		unsigned long memsize;
+		phys_addr_t memsize;
 
 		/* Ensure that total memory size is page-aligned, because
 		 * otherwise mark_bootmem() gets upset. */
@@ -1218,7 +1218,7 @@ void __init early_init_devtree(void *params)
 	lmb_analyze();
 	lmb_dump_all();
 
-	DBG("Phys. mem: %lx\n", lmb_phys_mem_size());
+	DBG("Phys. mem: %llx\n", lmb_phys_mem_size());
 
 	/* We may need to relocate the flat tree, do it now.
 	 * FIXME .. and the initrd too? */
diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c
index 2e026c0407d..2f0e64b5364 100644
--- a/arch/powerpc/kernel/prom_init.c
+++ b/arch/powerpc/kernel/prom_init.c
@@ -614,7 +614,7 @@ static unsigned char ibm_architecture_vec[] = {
 	W(0xffffffff),			/* virt_base */
 	W(0xffffffff),			/* virt_size */
 	W(0xffffffff),			/* load_base */
-	W(64),				/* 128MB min RMA */
+	W(64),				/* 64MB min RMA */
 	W(0xffffffff),			/* full client load */
 	0,				/* min RMA percentage of total RAM */
 	48,				/* max log_2(hash table size) */
@@ -732,15 +732,17 @@ static void __init prom_send_capabilities(void)
 	root = call_prom("open", 1, 1, ADDR("/"));
 	if (root != 0) {
 		/* try calling the ibm,client-architecture-support method */
+		prom_printf("Calling ibm,client-architecture...");
 		if (call_prom_ret("call-method", 3, 2, &ret,
 				  ADDR("ibm,client-architecture-support"),
 				  root,
 				  ADDR(ibm_architecture_vec)) == 0) {
 			/* the call exists... */
 			if (ret)
-				prom_printf("WARNING: ibm,client-architecture"
+				prom_printf("\nWARNING: ibm,client-architecture"
 					    "-support call FAILED!\n");
 			call_prom("close", 1, 0, root);
+			prom_printf(" done\n");
 			return;
 		}
 		call_prom("close", 1, 0, root);
@@ -1083,7 +1085,7 @@ static void __init prom_instantiate_rtas(void)
 		return;
 	}
 
-	prom_printf("instantiating rtas at 0x%x ...", base);
+	prom_printf("instantiating rtas at 0x%x...", base);
 
 	if (call_prom_ret("call-method", 3, 2, &entry,
 			  ADDR("instantiate-rtas"),
@@ -1495,7 +1497,7 @@ static int __init prom_find_machine_type(void)
 		return PLATFORM_GENERIC;
 	x = prom_getproplen(rtas, "ibm,hypertas-functions");
 	if (x != PROM_ERROR) {
-		prom_printf("Hypertas detected, assuming LPAR !\n");
+		prom_debug("Hypertas detected, assuming LPAR !\n");
 		return PLATFORM_PSERIES_LPAR;
 	}
 	return PLATFORM_PSERIES;
@@ -1544,7 +1546,7 @@ static void __init prom_check_displays(void)
 	};
 	const unsigned char *clut;
 
-	prom_printf("Looking for displays\n");
+	prom_debug("Looking for displays\n");
 	for (node = 0; prom_next_node(&node); ) {
 		memset(type, 0, sizeof(type));
 		prom_getprop(node, "device_type", type, sizeof(type));
@@ -1562,7 +1564,7 @@ static void __init prom_check_displays(void)
 		if (call_prom("package-to-path", 3, 1, node, path,
 			      PROM_SCRATCH_SIZE-10) == PROM_ERROR)
 			continue;
-		prom_printf("found display   : %s, opening ... ", path);
+		prom_printf("found display   : %s, opening... ", path);
 		
 		ih = call_prom("open", 1, 1, path);
 		if (ih == 0) {
@@ -2387,7 +2389,7 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,
 	/*
 	 * Now finally create the flattened device-tree
 	 */
-	prom_printf("copying OF device tree ...\n");
+	prom_printf("copying OF device tree...\n");
 	flatten_device_tree();
 
 	/*
@@ -2402,7 +2404,7 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,
 	 * Call OF "quiesce" method to shut down pending DMA's from
 	 * devices etc...
 	 */
-	prom_printf("Calling quiesce ...\n");
+	prom_printf("Calling quiesce...\n");
 	call_prom("quiesce", 0, 0);
 
 	/*
diff --git a/arch/powerpc/kernel/prom_parse.c b/arch/powerpc/kernel/prom_parse.c
index 8f0856f312d..8362620c9e6 100644
--- a/arch/powerpc/kernel/prom_parse.c
+++ b/arch/powerpc/kernel/prom_parse.c
@@ -971,7 +971,7 @@ int of_irq_map_one(struct device_node *device, int index, struct of_irq *out_irq
 	struct device_node *p;
 	const u32 *intspec, *tmp, *addr;
 	u32 intsize, intlen;
-	int res;
+	int res = -EINVAL;
 
 	DBG("of_irq_map_one: dev=%s, index=%d\n", device->full_name, index);
 
@@ -995,21 +995,20 @@ int of_irq_map_one(struct device_node *device, int index, struct of_irq *out_irq
 
 	/* Get size of interrupt specifier */
 	tmp = of_get_property(p, "#interrupt-cells", NULL);
-	if (tmp == NULL) {
-		of_node_put(p);
-		return -EINVAL;
-	}
+	if (tmp == NULL)
+		goto out;
 	intsize = *tmp;
 
 	DBG(" intsize=%d intlen=%d\n", intsize, intlen);
 
 	/* Check index */
 	if ((index + 1) * intsize > intlen)
-		return -EINVAL;
+		goto out;
 
 	/* Get new specifier and map it */
 	res = of_irq_map_raw(p, intspec + index * intsize, intsize,
 			     addr, out_irq);
+out:
 	of_node_put(p);
 	return res;
 }
diff --git a/arch/powerpc/kernel/ptrace32.c b/arch/powerpc/kernel/ptrace32.c
index 197d49c790a..297632cba04 100644
--- a/arch/powerpc/kernel/ptrace32.c
+++ b/arch/powerpc/kernel/ptrace32.c
@@ -67,7 +67,8 @@ static long compat_ptrace_old(struct task_struct *child, long request,
 /* Macros to workout the correct index for the FPR in the thread struct */
 #define FPRNUMBER(i) (((i) - PT_FPR0) >> 1)
 #define FPRHALF(i) (((i) - PT_FPR0) & 1)
-#define FPRINDEX(i) TS_FPRWIDTH * FPRNUMBER(i) + FPRHALF(i)
+#define FPRINDEX(i) TS_FPRWIDTH * FPRNUMBER(i) * 2 + FPRHALF(i)
+#define FPRINDEX_3264(i) (TS_FPRWIDTH * ((i) - PT_FPR0))
 
 long compat_arch_ptrace(struct task_struct *child, compat_long_t request,
 			compat_ulong_t caddr, compat_ulong_t cdata)
@@ -168,8 +169,9 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request,
 		CHECK_FULL_REGS(child->thread.regs);
 		if (numReg >= PT_FPR0) {
 			flush_fp_to_thread(child);
-			tmp = ((unsigned long int *)child->thread.fpr)
-				[FPRINDEX(numReg)];
+			/* get 64 bit FPR */
+			tmp = ((u64 *)child->thread.fpr)
+				[FPRINDEX_3264(numReg)];
 		} else { /* register within PT_REGS struct */
 			tmp = ptrace_get_reg(child, numReg);
 		} 
@@ -262,8 +264,13 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request,
 				freg = (freg & 0xfffffffful) | (data << 32);
 			ret = ptrace_put_reg(child, numReg, freg);
 		} else {
+			u64 *tmp;
 			flush_fp_to_thread(child);
-			((unsigned int *)child->thread.regs)[index] = data;
+			/* get 64 bit FPR ... */
+			tmp = &(((u64 *)child->thread.fpr)
+				[FPRINDEX_3264(numReg)]);
+			/* ... write the 32 bit part we want */
+			((u32 *)tmp)[index % 2] = data;
 			ret = 0;
 		}
 		break;
diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c
index 9774f9fed96..02fed27af7f 100644
--- a/arch/powerpc/kernel/setup-common.c
+++ b/arch/powerpc/kernel/setup-common.c
@@ -36,7 +36,6 @@
 #include <linux/percpu.h>
 #include <linux/lmb.h>
 #include <linux/of_platform.h>
-#include <linux/platform_device.h>
 #include <asm/io.h>
 #include <asm/prom.h>
 #include <asm/processor.h>
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index 926ea864e34..48571ac56fb 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -77,7 +77,7 @@
 #include <linux/clockchips.h>
 #include <linux/clocksource.h>
 
-static cycle_t rtc_read(void);
+static cycle_t rtc_read(struct clocksource *);
 static struct clocksource clocksource_rtc = {
 	.name         = "rtc",
 	.rating       = 400,
@@ -88,7 +88,7 @@ static struct clocksource clocksource_rtc = {
 	.read         = rtc_read,
 };
 
-static cycle_t timebase_read(void);
+static cycle_t timebase_read(struct clocksource *);
 static struct clocksource clocksource_timebase = {
 	.name         = "timebase",
 	.rating       = 400,
@@ -766,12 +766,12 @@ unsigned long read_persistent_clock(void)
 }
 
 /* clocksource code */
-static cycle_t rtc_read(void)
+static cycle_t rtc_read(struct clocksource *cs)
 {
 	return (cycle_t)get_rtc();
 }
 
-static cycle_t timebase_read(void)
+static cycle_t timebase_read(struct clocksource *cs)
 {
 	return (cycle_t)get_tb();
 }
diff --git a/arch/powerpc/kernel/vmlinux.lds.S b/arch/powerpc/kernel/vmlinux.lds.S
index b9ef1644a72..8ef8a14abc9 100644
--- a/arch/powerpc/kernel/vmlinux.lds.S
+++ b/arch/powerpc/kernel/vmlinux.lds.S
@@ -52,9 +52,10 @@ SECTIONS
 	/* Text and gots */
 	.text : AT(ADDR(.text) - LOAD_OFFSET) {
 		ALIGN_FUNCTION();
-		*(.text.head)
+		HEAD_TEXT
 		_text = .;
-		*(.text .fixup .text.init.refok .exit.text.refok __ftr_alt_*)
+		/* careful! __ftr_alt_* sections need to be close to .text */
+		*(.text .fixup __ftr_alt_* .ref.text)
 		SCHED_TEXT
 		LOCK_TEXT
 		KPROBES_TEXT
@@ -263,6 +264,7 @@ SECTIONS
 		*(.data.page_aligned)
 	}
 
+	. = ALIGN(L1_CACHE_BYTES);
 	.data.cacheline_aligned : AT(ADDR(.data.cacheline_aligned) - LOAD_OFFSET) {
 		*(.data.cacheline_aligned)
 	}