Merge commit 'v2.6.37-rc8' into x86/apic

Conflicts:
	arch/x86/include/asm/io_apic.h

Merge reason: move to a fresh -rc, resolve the conflict.

Signed-off-by: Ingo Molnar <mingo@elte.hu>

7 files changed, 116 insertions, 126 deletions

diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index 79b0b372d2d0..7d90ceb882a4 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -11,6 +11,7 @@
 #include <linux/kprobes.h>              /* __kprobes, ...               */
 #include <linux/mmiotrace.h>            /* kmmio_handler, ...           */
 #include <linux/perf_event.h>           /* perf_sw_event                */
+#include <linux/hugetlb.h>              /* hstate_index_to_shift        */
 
 #include <asm/traps.h>                  /* dotraplinkage, ...           */
 #include <asm/pgalloc.h>                /* pgd_*(), ...                 */
@@ -160,15 +161,20 @@ is_prefetch(struct pt_regs *regs, unsigned long error_code, unsigned long addr)
 
 static void
 force_sig_info_fault(int si_signo, int si_code, unsigned long address,
-                     struct task_struct *tsk)
+                     struct task_struct *tsk, int fault)
 {
+        unsigned lsb = 0;
         siginfo_t info;
 
         info.si_signo   = si_signo;
         info.si_errno   = 0;
         info.si_code    = si_code;
         info.si_addr    = (void __user *)address;
-        info.si_addr_lsb = si_code == BUS_MCEERR_AR ? PAGE_SHIFT : 0;
+        if (fault & VM_FAULT_HWPOISON_LARGE)
+                lsb = hstate_index_to_shift(VM_FAULT_GET_HINDEX(fault)); 
+        if (fault & VM_FAULT_HWPOISON)
+                lsb = PAGE_SHIFT;
+        info.si_addr_lsb = lsb;
 
         force_sig_info(si_signo, &info, tsk);
 }
@@ -722,7 +728,7 @@ __bad_area_nosemaphore(struct pt_regs *regs, unsigned long error_code,
                 tsk->thread.error_code  = error_code | (address >= TASK_SIZE);
                 tsk->thread.trap_no     = 14;
 
-                force_sig_info_fault(SIGSEGV, si_code, address, tsk);
+                force_sig_info_fault(SIGSEGV, si_code, address, tsk, 0);
 
                 return;
         }
@@ -807,14 +813,14 @@ do_sigbus(struct pt_regs *regs, unsigned long error_code, unsigned long address,
         tsk->thread.trap_no     = 14;
 
 #ifdef CONFIG_MEMORY_FAILURE
-        if (fault & VM_FAULT_HWPOISON) {
+        if (fault & (VM_FAULT_HWPOISON|VM_FAULT_HWPOISON_LARGE)) {
                 printk(KERN_ERR
         "MCE: Killing %s:%d due to hardware memory corruption fault at %lx\n",
                         tsk->comm, tsk->pid, address);
                 code = BUS_MCEERR_AR;
         }
 #endif
-        force_sig_info_fault(SIGBUS, code, address, tsk);
+        force_sig_info_fault(SIGBUS, code, address, tsk, fault);
 }
 
 static noinline void
@@ -824,7 +830,8 @@ mm_fault_error(struct pt_regs *regs, unsigned long error_code,
         if (fault & VM_FAULT_OOM) {
                 out_of_memory(regs, error_code, address);
         } else {
-                if (fault & (VM_FAULT_SIGBUS|VM_FAULT_HWPOISON))
+                if (fault & (VM_FAULT_SIGBUS|VM_FAULT_HWPOISON|
+                             VM_FAULT_HWPOISON_LARGE))
                         do_sigbus(regs, error_code, address, fault);
                 else
                         BUG();
@@ -912,9 +919,9 @@ spurious_fault(unsigned long error_code, unsigned long address)
 int show_unhandled_signals = 1;
 
 static inline int
-access_error(unsigned long error_code, int write, struct vm_area_struct *vma)
+access_error(unsigned long error_code, struct vm_area_struct *vma)
 {
-        if (write) {
+        if (error_code & PF_WRITE) {
                 /* write, present and write, not present: */
                 if (unlikely(!(vma->vm_flags & VM_WRITE)))
                         return 1;
@@ -949,8 +956,10 @@ do_page_fault(struct pt_regs *regs, unsigned long error_code)
         struct task_struct *tsk;
         unsigned long address;
         struct mm_struct *mm;
-        int write;
         int fault;
+        int write = error_code & PF_WRITE;
+        unsigned int flags = FAULT_FLAG_ALLOW_RETRY |
+                                        (write ? FAULT_FLAG_WRITE : 0);
 
         tsk = current;
         mm = tsk->mm;
@@ -1061,6 +1070,7 @@ do_page_fault(struct pt_regs *regs, unsigned long error_code)
                         bad_area_nosemaphore(regs, error_code, address);
                         return;
                 }
+retry:
                 down_read(&mm->mmap_sem);
         } else {
                 /*
@@ -1104,9 +1114,7 @@ do_page_fault(struct pt_regs *regs, unsigned long error_code)
          * we can handle it..
          */
 good_area:
-        write = error_code & PF_WRITE;
-
-        if (unlikely(access_error(error_code, write, vma))) {
+        if (unlikely(access_error(error_code, vma))) {
                 bad_area_access_error(regs, error_code, address);
                 return;
         }
@@ -1116,21 +1124,34 @@ good_area:
          * make sure we exit gracefully rather than endlessly redo
          * the fault:
          */
-        fault = handle_mm_fault(mm, vma, address, write ? FAULT_FLAG_WRITE : 0);
+        fault = handle_mm_fault(mm, vma, address, flags);
 
         if (unlikely(fault & VM_FAULT_ERROR)) {
                 mm_fault_error(regs, error_code, address, fault);
                 return;
         }
 
-        if (fault & VM_FAULT_MAJOR) {
-                tsk->maj_flt++;
-                perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1, 0,
-                                     regs, address);
-        } else {
-                tsk->min_flt++;
-                perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1, 0,
-                                     regs, address);
+        /*
+         * Major/minor page fault accounting is only done on the
+         * initial attempt. If we go through a retry, it is extremely
+         * likely that the page will be found in page cache at that point.
+         */
+        if (flags & FAULT_FLAG_ALLOW_RETRY) {
+                if (fault & VM_FAULT_MAJOR) {
+                        tsk->maj_flt++;
+                        perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1, 0,
+                                      regs, address);
+                } else {
+                        tsk->min_flt++;
+                        perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1, 0,
+                                      regs, address);
+                }
+                if (fault & VM_FAULT_RETRY) {
+                        /* Clear FAULT_FLAG_ALLOW_RETRY to avoid any risk
+                         * of starvation. */
+                        flags &= ~FAULT_FLAG_ALLOW_RETRY;
+                        goto retry;
+                }
         }
 
         check_v8086_mode(regs, address, tsk);
diff --git a/arch/x86/mm/highmem_32.c b/arch/x86/mm/highmem_32.c
index 5e8fa12ef861..b49962662101 100644
--- a/arch/x86/mm/highmem_32.c
+++ b/arch/x86/mm/highmem_32.c
@@ -9,6 +9,7 @@ void *kmap(struct page *page)
                 return page_address(page);
         return kmap_high(page);
 }
+EXPORT_SYMBOL(kmap);
 
 void kunmap(struct page *page)
 {
@@ -18,6 +19,7 @@ void kunmap(struct page *page)
                 return;
         kunmap_high(page);
 }
+EXPORT_SYMBOL(kunmap);
 
 /*
  * kmap_atomic/kunmap_atomic is significantly faster than kmap/kunmap because
@@ -27,10 +29,10 @@ void kunmap(struct page *page)
  * However when holding an atomic kmap it is not legal to sleep, so atomic
  * kmaps are appropriate for short, tight code paths only.
  */
-void *kmap_atomic_prot(struct page *page, enum km_type type, pgprot_t prot)
+void *kmap_atomic_prot(struct page *page, pgprot_t prot)
 {
-        enum fixed_addresses idx;
         unsigned long vaddr;
+        int idx, type;
 
         /* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */
         pagefault_disable();
@@ -38,8 +40,7 @@ void *kmap_atomic_prot(struct page *page, enum km_type type, pgprot_t prot)
         if (!PageHighMem(page))
                 return page_address(page);
 
-        debug_kmap_atomic(type);
-
+        type = kmap_atomic_idx_push();
         idx = type + KM_TYPE_NR*smp_processor_id();
         vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
         BUG_ON(!pte_none(*(kmap_pte-idx)));
@@ -47,44 +48,57 @@ void *kmap_atomic_prot(struct page *page, enum km_type type, pgprot_t prot)
 
         return (void *)vaddr;
 }
+EXPORT_SYMBOL(kmap_atomic_prot);
+
+void *__kmap_atomic(struct page *page)
+{
+        return kmap_atomic_prot(page, kmap_prot);
+}
+EXPORT_SYMBOL(__kmap_atomic);
 
-void *kmap_atomic(struct page *page, enum km_type type)
+/*
+ * 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)
 {
-        return kmap_atomic_prot(page, type, kmap_prot);
+        return kmap_atomic_prot_pfn(pfn, kmap_prot);
 }
+EXPORT_SYMBOL_GPL(kmap_atomic_pfn);
 
-void kunmap_atomic_notypecheck(void *kvaddr, enum km_type type)
+void __kunmap_atomic(void *kvaddr)
 {
         unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK;
-        enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id();
-
-        /*
-         * Force other mappings to Oops if they'll try to access this pte
-         * without first remap it.  Keeping stale mappings around is a bad idea
-         * also, in case the page changes cacheability attributes or becomes
-         * a protected page in a hypervisor.
-         */
-        if (vaddr == __fix_to_virt(FIX_KMAP_BEGIN+idx))
+
+        if (vaddr >= __fix_to_virt(FIX_KMAP_END) &&
+            vaddr <= __fix_to_virt(FIX_KMAP_BEGIN)) {
+                int idx, type;
+
+                type = kmap_atomic_idx();
+                idx = type + KM_TYPE_NR * smp_processor_id();
+
+#ifdef CONFIG_DEBUG_HIGHMEM
+                WARN_ON_ONCE(vaddr != __fix_to_virt(FIX_KMAP_BEGIN + idx));
+#endif
+                /*
+                 * Force other mappings to Oops if they'll try to access this
+                 * pte without first remap it.  Keeping stale mappings around
+                 * is a bad idea also, in case the page changes cacheability
+                 * attributes or becomes a protected page in a hypervisor.
+                 */
                 kpte_clear_flush(kmap_pte-idx, vaddr);
-        else {
+                kmap_atomic_idx_pop();
+        }
 #ifdef CONFIG_DEBUG_HIGHMEM
+        else {
                 BUG_ON(vaddr < PAGE_OFFSET);
                 BUG_ON(vaddr >= (unsigned long)high_memory);
-#endif
         }
+#endif
 
         pagefault_enable();
 }
-
-/*
- * 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)
-{
-        return kmap_atomic_prot_pfn(pfn, type, kmap_prot);
-}
-EXPORT_SYMBOL_GPL(kmap_atomic_pfn); /* temporarily in use by i915 GEM until vmap */
+EXPORT_SYMBOL(__kunmap_atomic);
 
 struct page *kmap_atomic_to_page(void *ptr)
 {
@@ -98,12 +112,6 @@ struct page *kmap_atomic_to_page(void *ptr)
         pte = kmap_pte - (idx - FIX_KMAP_BEGIN);
         return pte_page(*pte);
 }
-
-EXPORT_SYMBOL(kmap);
-EXPORT_SYMBOL(kunmap);
-EXPORT_SYMBOL(kmap_atomic);
-EXPORT_SYMBOL(kunmap_atomic_notypecheck);
-EXPORT_SYMBOL(kmap_atomic_prot);
 EXPORT_SYMBOL(kmap_atomic_to_page);
 
 void __init set_highmem_pages_init(void)
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 5d0a6711c282..0e969f9f401b 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -528,48 +528,6 @@ static void __init pagetable_init(void)
         permanent_kmaps_init(pgd_base);
 }
 
-#ifdef CONFIG_ACPI_SLEEP
-/*
- * ACPI suspend needs this for resume, because things like the intel-agp
- * driver might have split up a kernel 4MB mapping.
- */
-char swsusp_pg_dir[PAGE_SIZE]
-        __attribute__ ((aligned(PAGE_SIZE)));
-
-static inline void save_pg_dir(void)
-{
-        copy_page(swsusp_pg_dir, swapper_pg_dir);
-}
-#else /* !CONFIG_ACPI_SLEEP */
-static inline void save_pg_dir(void)
-{
-}
-#endif /* !CONFIG_ACPI_SLEEP */
-
-void zap_low_mappings(bool early)
-{
-        int i;
-
-        /*
-         * Zap initial low-memory mappings.
-         *
-         * Note that "pgd_clear()" doesn't do it for
-         * us, because pgd_clear() is a no-op on i386.
-         */
-        for (i = 0; i < KERNEL_PGD_BOUNDARY; i++) {
-#ifdef CONFIG_X86_PAE
-                set_pgd(swapper_pg_dir+i, __pgd(1 + __pa(empty_zero_page)));
-#else
-                set_pgd(swapper_pg_dir+i, __pgd(0));
-#endif
-        }
-
-        if (early)
-                __flush_tlb();
-        else
-                flush_tlb_all();
-}
-
 pteval_t __supported_pte_mask __read_mostly = ~(_PAGE_NX | _PAGE_GLOBAL | _PAGE_IOMAP);
 EXPORT_SYMBOL_GPL(__supported_pte_mask);
 
@@ -882,9 +840,6 @@ void __init mem_init(void)
 
         if (boot_cpu_data.wp_works_ok < 0)
                 test_wp_bit();
-
-        save_pg_dir();
-        zap_low_mappings(true);
 }
 
 #ifdef CONFIG_MEMORY_HOTPLUG
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 84346200e783..71a59296af80 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -51,7 +51,6 @@
 #include <asm/numa.h>
 #include <asm/cacheflush.h>
 #include <asm/init.h>
-#include <linux/bootmem.h>
 
 static int __init parse_direct_gbpages_off(char *arg)
 {
diff --git a/arch/x86/mm/iomap_32.c b/arch/x86/mm/iomap_32.c
index 72fc70cf6184..7b179b499fa3 100644
--- a/arch/x86/mm/iomap_32.c
+++ b/arch/x86/mm/iomap_32.c
@@ -48,21 +48,20 @@ int iomap_create_wc(resource_size_t base, unsigned long size, pgprot_t *prot)
 }
 EXPORT_SYMBOL_GPL(iomap_create_wc);
 
-void
-iomap_free(resource_size_t base, unsigned long size)
+void iomap_free(resource_size_t base, unsigned long size)
 {
         io_free_memtype(base, base + size);
 }
 EXPORT_SYMBOL_GPL(iomap_free);
 
-void *kmap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot)
+void *kmap_atomic_prot_pfn(unsigned long pfn, pgprot_t prot)
 {
-        enum fixed_addresses idx;
         unsigned long vaddr;
+        int idx, type;
 
         pagefault_disable();
 
-        debug_kmap_atomic(type);
+        type = kmap_atomic_idx_push();
         idx = type + KM_TYPE_NR * smp_processor_id();
         vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
         set_pte(kmap_pte - idx, pfn_pte(pfn, prot));
@@ -72,10 +71,10 @@ void *kmap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot)
 }
 
 /*
- * Map 'pfn' using fixed map 'type' and protections 'prot'
+ * Map 'pfn' using protections 'prot'
  */
 void __iomem *
-iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot)
+iomap_atomic_prot_pfn(unsigned long pfn, pgprot_t prot)
 {
         /*
          * For non-PAT systems, promote PAGE_KERNEL_WC to PAGE_KERNEL_UC_MINUS.
@@ -86,24 +85,34 @@ iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot)
         if (!pat_enabled && pgprot_val(prot) == pgprot_val(PAGE_KERNEL_WC))
                 prot = PAGE_KERNEL_UC_MINUS;
 
-        return (void __force __iomem *) kmap_atomic_prot_pfn(pfn, type, prot);
+        return (void __force __iomem *) kmap_atomic_prot_pfn(pfn, prot);
 }
 EXPORT_SYMBOL_GPL(iomap_atomic_prot_pfn);
 
 void
-iounmap_atomic(void __iomem *kvaddr, enum km_type type)
+iounmap_atomic(void __iomem *kvaddr)
 {
         unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK;
-        enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id();
 
-        /*
-         * Force other mappings to Oops if they'll try to access this pte
-         * without first remap it.  Keeping stale mappings around is a bad idea
-         * also, in case the page changes cacheability attributes or becomes
-         * a protected page in a hypervisor.
-         */
-        if (vaddr == __fix_to_virt(FIX_KMAP_BEGIN+idx))
+        if (vaddr >= __fix_to_virt(FIX_KMAP_END) &&
+            vaddr <= __fix_to_virt(FIX_KMAP_BEGIN)) {
+                int idx, type;
+
+                type = kmap_atomic_idx();
+                idx = type + KM_TYPE_NR * smp_processor_id();
+
+#ifdef CONFIG_DEBUG_HIGHMEM
+                WARN_ON_ONCE(vaddr != __fix_to_virt(FIX_KMAP_BEGIN + idx));
+#endif
+                /*
+                 * Force other mappings to Oops if they'll try to access this
+                 * pte without first remap it.  Keeping stale mappings around
+                 * is a bad idea also, in case the page changes cacheability
+                 * attributes or becomes a protected page in a hypervisor.
+                 */
                 kpte_clear_flush(kmap_pte-idx, vaddr);
+                kmap_atomic_idx_pop();
+        }
 
         pagefault_enable();
 }
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 60f498511dd6..7ffc9b727efd 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -178,11 +178,8 @@ static void * __init early_node_mem(int nodeid, unsigned long start,
 
         /* extend the search scope */
         end = max_pfn_mapped << PAGE_SHIFT;
-        if (end > (MAX_DMA32_PFN<<PAGE_SHIFT))
-                start = MAX_DMA32_PFN<<PAGE_SHIFT;
-        else
-                start = MAX_DMA_PFN<<PAGE_SHIFT;
-        mem = memblock_x86_find_in_range_node(nodeid, start, end, size, align);
+        start = MAX_DMA_PFN << PAGE_SHIFT;
+        mem = memblock_find_in_range(start, end, size, align);
         if (mem != MEMBLOCK_ERROR)
                 return __va(mem);
 
diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index 49358481c733..6acc724d5d8f 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -223,7 +223,7 @@ void native_flush_tlb_others(const struct cpumask *cpumask,
 
 static void __cpuinit calculate_tlb_offset(void)
 {
-        int cpu, node, nr_node_vecs;
+        int cpu, node, nr_node_vecs, idx = 0;
         /*
          * we are changing tlb_vector_offset for each CPU in runtime, but this
          * will not cause inconsistency, as the write is atomic under X86. we
@@ -239,7 +239,7 @@ static void __cpuinit calculate_tlb_offset(void)
                 nr_node_vecs = NUM_INVALIDATE_TLB_VECTORS/nr_online_nodes;
 
         for_each_online_node(node) {
-                int node_offset = (node % NUM_INVALIDATE_TLB_VECTORS) *
+                int node_offset = (idx % NUM_INVALIDATE_TLB_VECTORS) *
                         nr_node_vecs;
                 int cpu_offset = 0;
                 for_each_cpu(cpu, cpumask_of_node(node)) {
@@ -248,10 +248,11 @@ static void __cpuinit calculate_tlb_offset(void)
                         cpu_offset++;
                         cpu_offset = cpu_offset % nr_node_vecs;
                 }
+                idx++;
         }
 }
 
-static int tlb_cpuhp_notify(struct notifier_block *n,
+static int __cpuinit tlb_cpuhp_notify(struct notifier_block *n,
                 unsigned long action, void *hcpu)
 {
         switch (action & 0xf) {