Merge branch 'linus' into stackprotector

Conflicts:
	arch/x86/kernel/Makefile
	include/asm-x86/pda.h

7 files changed, 159 insertions, 181 deletions

diff --git a/arch/sparc64/mm/fault.c b/arch/sparc64/mm/fault.c
index 236f4d228d2b..a9e474bf6385 100644
--- a/arch/sparc64/mm/fault.c
+++ b/arch/sparc64/mm/fault.c
@@ -1,7 +1,7 @@
 /*
  * arch/sparc64/mm/fault.c: Page fault handlers for the 64-bit Sparc.
  *
- * Copyright (C) 1996 David S. Miller (davem@caip.rutgers.edu)
+ * Copyright (C) 1996, 2008 David S. Miller (davem@davemloft.net)
  * Copyright (C) 1997, 1999 Jakub Jelinek (jj@ultra.linux.cz)
  */
 
@@ -18,7 +18,6 @@
 #include <linux/init.h>
 #include <linux/interrupt.h>
 #include <linux/kprobes.h>
-#include <linux/kallsyms.h>
 #include <linux/kdebug.h>
 
 #include <asm/page.h>
@@ -52,43 +51,6 @@ static inline int notify_page_fault(struct pt_regs *regs)
 }
 #endif
 
-/*
- * To debug kernel to catch accesses to certain virtual/physical addresses.
- * Mode = 0 selects physical watchpoints, mode = 1 selects virtual watchpoints.
- * flags = VM_READ watches memread accesses, flags = VM_WRITE watches memwrite accesses.
- * Caller passes in a 64bit aligned addr, with mask set to the bytes that need to be
- * watched. This is only useful on a single cpu machine for now. After the watchpoint
- * is detected, the process causing it will be killed, thus preventing an infinite loop.
- */
-void set_brkpt(unsigned long addr, unsigned char mask, int flags, int mode)
-{
-	unsigned long lsubits;
-
-	__asm__ __volatile__("ldxa [%%g0] %1, %0"
-			     : "=r" (lsubits)
-			     : "i" (ASI_LSU_CONTROL));
-	lsubits &= ~(LSU_CONTROL_PM | LSU_CONTROL_VM |
-		     LSU_CONTROL_PR | LSU_CONTROL_VR |
-		     LSU_CONTROL_PW | LSU_CONTROL_VW);
-
-	__asm__ __volatile__("stxa	%0, [%1] %2\n\t"
-			     "membar	#Sync"
-			     : /* no outputs */
-			     : "r" (addr), "r" (mode ? VIRT_WATCHPOINT : PHYS_WATCHPOINT),
-			       "i" (ASI_DMMU));
-
-	lsubits |= ((unsigned long)mask << (mode ? 25 : 33));
-	if (flags & VM_READ)
-		lsubits |= (mode ? LSU_CONTROL_VR : LSU_CONTROL_PR);
-	if (flags & VM_WRITE)
-		lsubits |= (mode ? LSU_CONTROL_VW : LSU_CONTROL_PW);
-	__asm__ __volatile__("stxa %0, [%%g0] %1\n\t"
-			     "membar #Sync"
-			     : /* no outputs */
-			     : "r" (lsubits), "i" (ASI_LSU_CONTROL)
-			     : "memory");
-}
-
 static void __kprobes unhandled_fault(unsigned long address,
 				      struct task_struct *tsk,
 				      struct pt_regs *regs)
@@ -115,7 +77,7 @@ static void bad_kernel_pc(struct pt_regs *regs, unsigned long vaddr)
 	printk(KERN_CRIT "OOPS: Bogus kernel PC [%016lx] in fault handler\n",
 	       regs->tpc);
 	printk(KERN_CRIT "OOPS: RPC [%016lx]\n", regs->u_regs[15]);
-	print_symbol("RPC: <%s>\n", regs->u_regs[15]);
+	printk("OOPS: RPC <%pS>\n", (void *) regs->u_regs[15]);
 	printk(KERN_CRIT "OOPS: Fault was to vaddr[%lx]\n", vaddr);
 	dump_stack();
 	unhandled_fault(regs->tpc, current, regs);
diff --git a/arch/sparc64/mm/hugetlbpage.c b/arch/sparc64/mm/hugetlbpage.c
index 6cfab2e4d340..f27d10369e0c 100644
--- a/arch/sparc64/mm/hugetlbpage.c
+++ b/arch/sparc64/mm/hugetlbpage.c
@@ -175,7 +175,7 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 		return -ENOMEM;
 
 	if (flags & MAP_FIXED) {
-		if (prepare_hugepage_range(addr, len))
+		if (prepare_hugepage_range(file, addr, len))
 			return -EINVAL;
 		return addr;
 	}
@@ -195,7 +195,8 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 				pgoff, flags);
 }
 
-pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_alloc(struct mm_struct *mm,
+			unsigned long addr, unsigned long sz)
 {
 	pgd_t *pgd;
 	pud_t *pud;
@@ -294,6 +295,11 @@ int pmd_huge(pmd_t pmd)
 	return 0;
 }
 
+int pud_huge(pud_t pud)
+{
+	return 0;
+}
+
 struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address,
 			     pmd_t *pmd, int write)
 {
@@ -344,7 +350,7 @@ void hugetlb_prefault_arch_hook(struct mm_struct *mm)
 			 * also executing in this address space.
 			 */
 			mm->context.sparc64_ctx_val = ctx;
-			on_each_cpu(context_reload, mm, 0, 0);
+			on_each_cpu(context_reload, mm, 0);
 		}
 		spin_unlock(&ctx_alloc_lock);
 	}
diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c
index 84898c44dd4d..3c10daf8fc01 100644
--- a/arch/sparc64/mm/init.c
+++ b/arch/sparc64/mm/init.c
@@ -46,14 +46,11 @@
 #include <asm/tsb.h>
 #include <asm/hypervisor.h>
 #include <asm/prom.h>
-#include <asm/sstate.h>
 #include <asm/mdesc.h>
 #include <asm/cpudata.h>
+#include <asm/irq.h>
 
-#define MAX_PHYS_ADDRESS	(1UL << 42UL)
-#define KPTE_BITMAP_CHUNK_SZ	(256UL * 1024UL * 1024UL)
-#define KPTE_BITMAP_BYTES	\
-	((MAX_PHYS_ADDRESS / KPTE_BITMAP_CHUNK_SZ) / 8)
+#include "init.h"
 
 unsigned long kern_linear_pte_xor[2] __read_mostly;
 
@@ -392,51 +389,6 @@ void __kprobes flush_icache_range(unsigned long start, unsigned long end)
 	}
 }
 
-void show_mem(void)
-{
-	unsigned long total = 0, reserved = 0;
-	unsigned long shared = 0, cached = 0;
-	pg_data_t *pgdat;
-
-	printk(KERN_INFO "Mem-info:\n");
-	show_free_areas();
-	printk(KERN_INFO "Free swap:       %6ldkB\n",
-	       nr_swap_pages << (PAGE_SHIFT-10));
-	for_each_online_pgdat(pgdat) {
-		unsigned long i, flags;
-
-		pgdat_resize_lock(pgdat, &flags);
-		for (i = 0; i < pgdat->node_spanned_pages; i++) {
-			struct page *page = pgdat_page_nr(pgdat, i);
-			total++;
-			if (PageReserved(page))
-				reserved++;
-			else if (PageSwapCache(page))
-				cached++;
-			else if (page_count(page))
-				shared += page_count(page) - 1;
-		}
-		pgdat_resize_unlock(pgdat, &flags);
-	}
-
-	printk(KERN_INFO "%lu pages of RAM\n", total);
-	printk(KERN_INFO "%lu reserved pages\n", reserved);
-	printk(KERN_INFO "%lu pages shared\n", shared);
-	printk(KERN_INFO "%lu pages swap cached\n", cached);
-
-	printk(KERN_INFO "%lu pages dirty\n",
-	       global_page_state(NR_FILE_DIRTY));
-	printk(KERN_INFO "%lu pages writeback\n",
-	       global_page_state(NR_WRITEBACK));
-	printk(KERN_INFO "%lu pages mapped\n",
-	       global_page_state(NR_FILE_MAPPED));
-	printk(KERN_INFO "%lu pages slab\n",
-		global_page_state(NR_SLAB_RECLAIMABLE) +
-		global_page_state(NR_SLAB_UNRECLAIMABLE));
-	printk(KERN_INFO "%lu pages pagetables\n",
-	       global_page_state(NR_PAGETABLE));
-}
-
 void mmu_info(struct seq_file *m)
 {
 	if (tlb_type == cheetah)
@@ -460,17 +412,9 @@ void mmu_info(struct seq_file *m)
 #endif /* CONFIG_DEBUG_DCFLUSH */
 }
 
-struct linux_prom_translation {
-	unsigned long virt;
-	unsigned long size;
-	unsigned long data;
-};
-
-/* Exported for kernel TLB miss handling in ktlb.S */
 struct linux_prom_translation prom_trans[512] __read_mostly;
 unsigned int prom_trans_ents __read_mostly;
 
-/* Exported for SMP bootup purposes. */
 unsigned long kern_locked_tte_data;
 
 /* The obp translations are saved based on 8k pagesize, since obp can
@@ -788,7 +732,6 @@ int numa_cpu_lookup_table[NR_CPUS];
 cpumask_t numa_cpumask_lookup_table[MAX_NUMNODES];
 
 #ifdef CONFIG_NEED_MULTIPLE_NODES
-static bootmem_data_t plat_node_bdata[MAX_NUMNODES];
 
 struct mdesc_mblock {
 	u64	base;
@@ -841,6 +784,9 @@ static unsigned long nid_range(unsigned long start, unsigned long end,
 		start += PAGE_SIZE;
 	}
 
+	if (start > end)
+		start = end;
+
 	return start;
 }
 #else
@@ -871,7 +817,7 @@ static void __init allocate_node_data(int nid)
 	NODE_DATA(nid) = __va(paddr);
 	memset(NODE_DATA(nid), 0, sizeof(struct pglist_data));
 
-	NODE_DATA(nid)->bdata = &plat_node_bdata[nid];
+	NODE_DATA(nid)->bdata = &bootmem_node_data[nid];
 #endif
 
 	p = NODE_DATA(nid);
@@ -980,6 +926,10 @@ int of_node_to_nid(struct device_node *dp)
 	int count, nid;
 	u64 grp;
 
+	/* This is the right thing to do on currently supported
+	 * SUN4U NUMA platforms as well, as the PCI controller does
+	 * not sit behind any particular memory controller.
+	 */
 	if (!mlgroups)
 		return -1;
 
@@ -1248,8 +1198,44 @@ out:
 	return err;
 }
 
+static int __init numa_parse_jbus(void)
+{
+	unsigned long cpu, index;
+
+	/* NUMA node id is encoded in bits 36 and higher, and there is
+	 * a 1-to-1 mapping from CPU ID to NUMA node ID.
+	 */
+	index = 0;
+	for_each_present_cpu(cpu) {
+		numa_cpu_lookup_table[cpu] = index;
+		numa_cpumask_lookup_table[index] = cpumask_of_cpu(cpu);
+		node_masks[index].mask = ~((1UL << 36UL) - 1UL);
+		node_masks[index].val = cpu << 36UL;
+
+		index++;
+	}
+	num_node_masks = index;
+
+	add_node_ranges();
+
+	for (index = 0; index < num_node_masks; index++) {
+		allocate_node_data(index);
+		node_set_online(index);
+	}
+
+	return 0;
+}
+
 static int __init numa_parse_sun4u(void)
 {
+	if (tlb_type == cheetah || tlb_type == cheetah_plus) {
+		unsigned long ver;
+
+		__asm__ ("rdpr %%ver, %0" : "=r" (ver));
+		if ((ver >> 32UL) == __JALAPENO_ID ||
+		    (ver >> 32UL) == __SERRANO_ID)
+			return numa_parse_jbus();
+	}
 	return -1;
 }
 
@@ -1675,8 +1661,6 @@ void __cpuinit sun4v_ktsb_register(void)
 
 /* paging_init() sets up the page tables */
 
-extern void central_probe(void);
-
 static unsigned long last_valid_pfn;
 pgd_t swapper_pg_dir[2048];
 
@@ -1721,8 +1705,6 @@ void __init paging_init(void)
 	kern_base = (prom_boot_mapping_phys_low >> 22UL) << 22UL;
 	kern_size = (unsigned long)&_end - (unsigned long)KERNBASE;
 
-	sstate_booting();
-
 	/* Invalidate both kernel TSBs.  */
 	memset(swapper_tsb, 0x40, sizeof(swapper_tsb));
 #ifndef CONFIG_DEBUG_PAGEALLOC
@@ -1768,8 +1750,7 @@ void __init paging_init(void)
 
 	find_ramdisk(phys_base);
 
-	if (cmdline_memory_size)
-		lmb_enforce_memory_limit(phys_base + cmdline_memory_size);
+	lmb_enforce_memory_limit(cmdline_memory_size);
 
 	lmb_analyze();
 	lmb_dump_all();
@@ -1817,6 +1798,16 @@ void __init paging_init(void)
 	if (tlb_type == hypervisor)
 		sun4v_mdesc_init();
 
+	/* Once the OF device tree and MDESC have been setup, we know
+	 * the list of possible cpus.  Therefore we can allocate the
+	 * IRQ stacks.
+	 */
+	for_each_possible_cpu(i) {
+		/* XXX Use node local allocations... XXX */
+		softirq_stack[i] = __va(lmb_alloc(THREAD_SIZE, THREAD_SIZE));
+		hardirq_stack[i] = __va(lmb_alloc(THREAD_SIZE, THREAD_SIZE));
+	}
+
 	/* Setup bootmem... */
 	last_valid_pfn = end_pfn = bootmem_init(phys_base);
 
@@ -1836,9 +1827,6 @@ void __init paging_init(void)
 	}
 
 	printk("Booting Linux...\n");
-
-	central_probe();
-	cpu_probe();
 }
 
 int __init page_in_phys_avail(unsigned long paddr)
@@ -1876,7 +1864,7 @@ static int pavail_rescan_ents __initdata;
  * memory list again, and make sure it provides at least as much
  * memory as 'pavail' does.
  */
-static void setup_valid_addr_bitmap_from_pavail(void)
+static void __init setup_valid_addr_bitmap_from_pavail(void)
 {
 	int i;
 
@@ -1996,6 +1984,15 @@ void __init mem_init(void)
 void free_initmem(void)
 {
 	unsigned long addr, initend;
+	int do_free = 1;
+
+	/* If the physical memory maps were trimmed by kernel command
+	 * line options, don't even try freeing this initmem stuff up.
+	 * The kernel image could have been in the trimmed out region
+	 * and if so the freeing below will free invalid page structs.
+	 */
+	if (cmdline_memory_size)
+		do_free = 0;
 
 	/*
 	 * The init section is aligned to 8k in vmlinux.lds. Page align for >8k pagesizes.
@@ -2010,13 +2007,16 @@ void free_initmem(void)
 			((unsigned long) __va(kern_base)) -
 			((unsigned long) KERNBASE));
 		memset((void *)addr, POISON_FREE_INITMEM, PAGE_SIZE);
-		p = virt_to_page(page);
 
-		ClearPageReserved(p);
-		init_page_count(p);
-		__free_page(p);
-		num_physpages++;
-		totalram_pages++;
+		if (do_free) {
+			p = virt_to_page(page);
+
+			ClearPageReserved(p);
+			init_page_count(p);
+			__free_page(p);
+			num_physpages++;
+			totalram_pages++;
+		}
 	}
 }
 
@@ -2053,7 +2053,6 @@ pgprot_t PAGE_COPY __read_mostly;
 pgprot_t PAGE_SHARED __read_mostly;
 EXPORT_SYMBOL(PAGE_SHARED);
 
-pgprot_t PAGE_EXEC __read_mostly;
 unsigned long pg_iobits __read_mostly;
 
 unsigned long _PAGE_IE __read_mostly;
@@ -2066,14 +2065,6 @@ unsigned long _PAGE_CACHE __read_mostly;
 EXPORT_SYMBOL(_PAGE_CACHE);
 
 #ifdef CONFIG_SPARSEMEM_VMEMMAP
-
-#define VMEMMAP_CHUNK_SHIFT	22
-#define VMEMMAP_CHUNK		(1UL << VMEMMAP_CHUNK_SHIFT)
-#define VMEMMAP_CHUNK_MASK	~(VMEMMAP_CHUNK - 1UL)
-#define VMEMMAP_ALIGN(x)	(((x)+VMEMMAP_CHUNK-1UL)&VMEMMAP_CHUNK_MASK)
-
-#define VMEMMAP_SIZE	((((1UL << MAX_PHYSADDR_BITS) >> PAGE_SHIFT) * \
-			  sizeof(struct page *)) >> VMEMMAP_CHUNK_SHIFT)
 unsigned long vmemmap_table[VMEMMAP_SIZE];
 
 int __meminit vmemmap_populate(struct page *start, unsigned long nr, int node)
@@ -2157,7 +2148,6 @@ static void __init sun4u_pgprot_init(void)
 				       _PAGE_CACHE_4U | _PAGE_P_4U |
 				       __ACCESS_BITS_4U | __DIRTY_BITS_4U |
 				       _PAGE_EXEC_4U | _PAGE_L_4U);
-	PAGE_EXEC = __pgprot(_PAGE_EXEC_4U);
 
 	_PAGE_IE = _PAGE_IE_4U;
 	_PAGE_E = _PAGE_E_4U;
@@ -2168,10 +2158,10 @@ static void __init sun4u_pgprot_init(void)
 
 #ifdef CONFIG_DEBUG_PAGEALLOC
 	kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZBITS_4U) ^
-		0xfffff80000000000;
+		0xfffff80000000000UL;
 #else
 	kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZ4MB_4U) ^
-		0xfffff80000000000;
+		0xfffff80000000000UL;
 #endif
 	kern_linear_pte_xor[0] |= (_PAGE_CP_4U | _PAGE_CV_4U |
 				   _PAGE_P_4U | _PAGE_W_4U);
@@ -2209,7 +2199,6 @@ static void __init sun4v_pgprot_init(void)
 				__ACCESS_BITS_4V | __DIRTY_BITS_4V |
 				_PAGE_EXEC_4V);
 	PAGE_KERNEL_LOCKED = PAGE_KERNEL;
-	PAGE_EXEC = __pgprot(_PAGE_EXEC_4V);
 
 	_PAGE_IE = _PAGE_IE_4V;
 	_PAGE_E = _PAGE_E_4V;
@@ -2217,20 +2206,20 @@ static void __init sun4v_pgprot_init(void)
 
 #ifdef CONFIG_DEBUG_PAGEALLOC
 	kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZBITS_4V) ^
-		0xfffff80000000000;
+		0xfffff80000000000UL;
 #else
 	kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZ4MB_4V) ^
-		0xfffff80000000000;
+		0xfffff80000000000UL;
 #endif
 	kern_linear_pte_xor[0] |= (_PAGE_CP_4V | _PAGE_CV_4V |
 				   _PAGE_P_4V | _PAGE_W_4V);
 
 #ifdef CONFIG_DEBUG_PAGEALLOC
 	kern_linear_pte_xor[1] = (_PAGE_VALID | _PAGE_SZBITS_4V) ^
-		0xfffff80000000000;
+		0xfffff80000000000UL;
 #else
 	kern_linear_pte_xor[1] = (_PAGE_VALID | _PAGE_SZ256MB_4V) ^
-		0xfffff80000000000;
+		0xfffff80000000000UL;
 #endif
 	kern_linear_pte_xor[1] |= (_PAGE_CP_4V | _PAGE_CV_4V |
 				   _PAGE_P_4V | _PAGE_W_4V);
diff --git a/arch/sparc64/mm/init.h b/arch/sparc64/mm/init.h
new file mode 100644
index 000000000000..16063870a489
--- /dev/null
+++ b/arch/sparc64/mm/init.h
@@ -0,0 +1,49 @@
+#ifndef _SPARC64_MM_INIT_H
+#define _SPARC64_MM_INIT_H
+
+/* Most of the symbols in this file are defined in init.c and
+ * marked non-static so that assembler code can get at them.
+ */
+
+#define MAX_PHYS_ADDRESS	(1UL << 42UL)
+#define KPTE_BITMAP_CHUNK_SZ	(256UL * 1024UL * 1024UL)
+#define KPTE_BITMAP_BYTES	\
+	((MAX_PHYS_ADDRESS / KPTE_BITMAP_CHUNK_SZ) / 8)
+
+extern unsigned long kern_linear_pte_xor[2];
+extern unsigned long kpte_linear_bitmap[KPTE_BITMAP_BYTES / sizeof(unsigned long)];
+extern unsigned int sparc64_highest_unlocked_tlb_ent;
+extern unsigned long sparc64_kern_pri_context;
+extern unsigned long sparc64_kern_pri_nuc_bits;
+extern unsigned long sparc64_kern_sec_context;
+extern void mmu_info(struct seq_file *m);
+
+struct linux_prom_translation {
+	unsigned long virt;
+	unsigned long size;
+	unsigned long data;
+};
+
+/* Exported for kernel TLB miss handling in ktlb.S */
+extern struct linux_prom_translation prom_trans[512];
+extern unsigned int prom_trans_ents;
+
+/* Exported for SMP bootup purposes. */
+extern unsigned long kern_locked_tte_data;
+
+extern void prom_world(int enter);
+
+extern void free_initmem(void);
+
+#ifdef CONFIG_SPARSEMEM_VMEMMAP
+#define VMEMMAP_CHUNK_SHIFT	22
+#define VMEMMAP_CHUNK		(1UL << VMEMMAP_CHUNK_SHIFT)
+#define VMEMMAP_CHUNK_MASK	~(VMEMMAP_CHUNK - 1UL)
+#define VMEMMAP_ALIGN(x)	(((x)+VMEMMAP_CHUNK-1UL)&VMEMMAP_CHUNK_MASK)
+
+#define VMEMMAP_SIZE	((((1UL << MAX_PHYSADDR_BITS) >> PAGE_SHIFT) * \
+			  sizeof(struct page *)) >> VMEMMAP_CHUNK_SHIFT)
+extern unsigned long vmemmap_table[VMEMMAP_SIZE];
+#endif
+
+#endif /* _SPARC64_MM_INIT_H */
diff --git a/arch/sparc64/mm/tlb.c b/arch/sparc64/mm/tlb.c
index ae24919cba7c..d8f21e24a82f 100644
--- a/arch/sparc64/mm/tlb.c
+++ b/arch/sparc64/mm/tlb.c
@@ -19,7 +19,7 @@
 
 /* Heavily inspired by the ppc64 code.  */
 
-DEFINE_PER_CPU(struct mmu_gather, mmu_gathers) = { 0, };
+DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 
 void flush_tlb_pending(void)
 {
diff --git a/arch/sparc64/mm/tsb.c b/arch/sparc64/mm/tsb.c
index fe70c8a557b5..587f8efb2e05 100644
--- a/arch/sparc64/mm/tsb.c
+++ b/arch/sparc64/mm/tsb.c
@@ -1,9 +1,10 @@
 /* arch/sparc64/mm/tsb.c
  *
- * Copyright (C) 2006 David S. Miller <davem@davemloft.net>
+ * Copyright (C) 2006, 2008 David S. Miller <davem@davemloft.net>
  */
 
 #include <linux/kernel.h>
+#include <linux/preempt.h>
 #include <asm/system.h>
 #include <asm/page.h>
 #include <asm/tlbflush.h>
@@ -96,12 +97,6 @@ void flush_tsb_user(struct mmu_gather *mp)
 #elif defined(CONFIG_SPARC64_PAGE_SIZE_64KB)
 #define HV_PGSZ_IDX_BASE	HV_PGSZ_IDX_64K
 #define HV_PGSZ_MASK_BASE	HV_PGSZ_MASK_64K
-#elif defined(CONFIG_SPARC64_PAGE_SIZE_512KB)
-#define HV_PGSZ_IDX_BASE	HV_PGSZ_IDX_512K
-#define HV_PGSZ_MASK_BASE	HV_PGSZ_MASK_512K
-#elif defined(CONFIG_SPARC64_PAGE_SIZE_4MB)
-#define HV_PGSZ_IDX_BASE	HV_PGSZ_IDX_4MB
-#define HV_PGSZ_MASK_BASE	HV_PGSZ_MASK_4MB
 #else
 #error Broken base page size setting...
 #endif
@@ -421,7 +416,9 @@ retry_tsb_alloc:
 		tsb_context_switch(mm);
 
 		/* Now force other processors to do the same.  */
+		preempt_disable();
 		smp_tsb_sync(mm);
+		preempt_enable();
 
 		/* Now it is safe to free the old tsb.  */
 		kmem_cache_free(tsb_caches[old_cache_index], old_tsb);
diff --git a/arch/sparc64/mm/ultra.S b/arch/sparc64/mm/ultra.S
index 9bb2d90a9df6..86773e89dc1b 100644
--- a/arch/sparc64/mm/ultra.S
+++ b/arch/sparc64/mm/ultra.S
@@ -480,42 +480,6 @@ xcall_sync_tick:
 	b		rtrap_xcall
 	 ldx		[%sp + PTREGS_OFF + PT_V9_TSTATE], %l1
 
-	/* NOTE: This is SPECIAL!!  We do etrap/rtrap however
-	 *       we choose to deal with the "BH's run with
-	 *       %pil==15" problem (described in asm/pil.h)
-	 *       by just invoking rtrap directly past where
-	 *       BH's are checked for.
-	 *
-	 *       We do it like this because we do not want %pil==15
-	 *       lockups to prevent regs being reported.
-	 */
-	.globl		xcall_report_regs
-xcall_report_regs:
-
-661:	rdpr		%pstate, %g2
-	wrpr		%g2, PSTATE_IG | PSTATE_AG, %pstate
-	.section	.sun4v_2insn_patch, "ax"
-	.word		661b
-	nop
-	nop
-	.previous
-
-	rdpr		%pil, %g2
-	wrpr		%g0, 15, %pil
-	sethi		%hi(109f), %g7
-	b,pt		%xcc, etrap_irq
-109:	 or		%g7, %lo(109b), %g7
-#ifdef CONFIG_TRACE_IRQFLAGS
-	call		trace_hardirqs_off
-	 nop
-#endif
-	call		__show_regs
-	 add		%sp, PTREGS_OFF, %o0
-	/* Has to be a non-v9 branch due to the large distance. */
-	b		rtrap_xcall
-	 ldx		[%sp + PTREGS_OFF + PT_V9_TSTATE], %l1
-
-#ifdef CONFIG_MAGIC_SYSRQ
 	.globl		xcall_fetch_glob_regs
 xcall_fetch_glob_regs:
 	sethi		%hi(global_reg_snapshot), %g1
@@ -531,6 +495,13 @@ xcall_fetch_glob_regs:
 	stx		%g7, [%g1 + GR_SNAP_TNPC]
 	stx		%o7, [%g1 + GR_SNAP_O7]
 	stx		%i7, [%g1 + GR_SNAP_I7]
+	/* Don't try this at home kids... */
+	rdpr		%cwp, %g2
+	sub		%g2, 1, %g7
+	wrpr		%g7, %cwp
+	mov		%i7, %g7
+	wrpr		%g2, %cwp
+	stx		%g7, [%g1 + GR_SNAP_RPC]
 	sethi		%hi(trap_block), %g7
 	or		%g7, %lo(trap_block), %g7
 	sllx		%g2, TRAP_BLOCK_SZ_SHIFT, %g2
@@ -539,7 +510,6 @@ xcall_fetch_glob_regs:
 	membar		#StoreStore
 	stx		%g3, [%g1 + GR_SNAP_THREAD]
 	retry
-#endif /* CONFIG_MAGIC_SYSRQ */
 
 #ifdef DCACHE_ALIASING_POSSIBLE
 	.align		32
@@ -688,6 +658,11 @@ xcall_call_function:
 	wr		%g0, (1 << PIL_SMP_CALL_FUNC), %set_softint
 	retry
 
+	.globl		xcall_call_function_single
+xcall_call_function_single:
+	wr		%g0, (1 << PIL_SMP_CALL_FUNC_SNGL), %set_softint
+	retry
+
 	.globl		xcall_receive_signal
 xcall_receive_signal:
 	wr		%g0, (1 << PIL_SMP_RECEIVE_SIGNAL), %set_softint