Merge commit 'linus/master' into tracing/kprobes

Conflicts:
	kernel/trace/Makefile
	kernel/trace/trace.h
	kernel/trace/trace_event_types.h
	kernel/trace/trace_export.c

Merge reason:
	Sync with latest significant tracing core changes.

15 files changed, 449 insertions, 263 deletions

diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile
index eefdeee8a871..9b5a9f59a478 100644
--- a/arch/x86/mm/Makefile
+++ b/arch/x86/mm/Makefile
@@ -1,5 +1,9 @@
 obj-y	:=  init.o init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \
-	    pat.o pgtable.o gup.o
+	    pat.o pgtable.o physaddr.o gup.o
+
+# Make sure __phys_addr has no stackprotector
+nostackp := $(call cc-option, -fno-stack-protector)
+CFLAGS_physaddr.o		:= $(nostackp)
 
 obj-$(CONFIG_SMP)		+= tlb.o
 
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index c322e59f2d10..923ea3fb7037 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -10,7 +10,7 @@
 #include <linux/bootmem.h>		/* max_low_pfn			*/
 #include <linux/kprobes.h>		/* __kprobes, ...		*/
 #include <linux/mmiotrace.h>		/* kmmio_handler, ...		*/
-#include <linux/perf_counter.h>		/* perf_swcounter_event		*/
+#include <linux/perf_event.h>		/* perf_sw_event		*/
 
 #include <asm/traps.h>			/* dotraplinkage, ...		*/
 #include <asm/pgalloc.h>		/* pgd_*(), ...			*/
@@ -286,26 +286,25 @@ check_v8086_mode(struct pt_regs *regs, unsigned long address,
 		tsk->thread.screen_bitmap |= 1 << bit;
 }
 
-static void dump_pagetable(unsigned long address)
+static bool low_pfn(unsigned long pfn)
 {
-	__typeof__(pte_val(__pte(0))) page;
+	return pfn < max_low_pfn;
+}
 
-	page = read_cr3();
-	page = ((__typeof__(page) *) __va(page))[address >> PGDIR_SHIFT];
+static void dump_pagetable(unsigned long address)
+{
+	pgd_t *base = __va(read_cr3());
+	pgd_t *pgd = &base[pgd_index(address)];
+	pmd_t *pmd;
+	pte_t *pte;
 
 #ifdef CONFIG_X86_PAE
-	printk("*pdpt = %016Lx ", page);
-	if ((page >> PAGE_SHIFT) < max_low_pfn
-	    && page & _PAGE_PRESENT) {
-		page &= PAGE_MASK;
-		page = ((__typeof__(page) *) __va(page))[(address >> PMD_SHIFT)
-							& (PTRS_PER_PMD - 1)];
-		printk(KERN_CONT "*pde = %016Lx ", page);
-		page &= ~_PAGE_NX;
-	}
-#else
-	printk("*pde = %08lx ", page);
+	printk("*pdpt = %016Lx ", pgd_val(*pgd));
+	if (!low_pfn(pgd_val(*pgd) >> PAGE_SHIFT) || !pgd_present(*pgd))
+		goto out;
 #endif
+	pmd = pmd_offset(pud_offset(pgd, address), address);
+	printk(KERN_CONT "*pde = %0*Lx ", sizeof(*pmd) * 2, (u64)pmd_val(*pmd));
 
 	/*
 	 * We must not directly access the pte in the highpte
@@ -313,16 +312,12 @@ static void dump_pagetable(unsigned long address)
 	 * And let's rather not kmap-atomic the pte, just in case
 	 * it's allocated already:
 	 */
-	if ((page >> PAGE_SHIFT) < max_low_pfn
-	    && (page & _PAGE_PRESENT)
-	    && !(page & _PAGE_PSE)) {
-
-		page &= PAGE_MASK;
-		page = ((__typeof__(page) *) __va(page))[(address >> PAGE_SHIFT)
-							& (PTRS_PER_PTE - 1)];
-		printk("*pte = %0*Lx ", sizeof(page)*2, (u64)page);
-	}
+	if (!low_pfn(pmd_pfn(*pmd)) || !pmd_present(*pmd) || pmd_large(*pmd))
+		goto out;
 
+	pte = pte_offset_kernel(pmd, address);
+	printk("*pte = %0*Lx ", sizeof(*pte) * 2, (u64)pte_val(*pte));
+out:
 	printk("\n");
 }
 
@@ -451,16 +446,12 @@ static int bad_address(void *p)
 
 static void dump_pagetable(unsigned long address)
 {
-	pgd_t *pgd;
+	pgd_t *base = __va(read_cr3() & PHYSICAL_PAGE_MASK);
+	pgd_t *pgd = base + pgd_index(address);
 	pud_t *pud;
 	pmd_t *pmd;
 	pte_t *pte;
 
-	pgd = (pgd_t *)read_cr3();
-
-	pgd = __va((unsigned long)pgd & PHYSICAL_PAGE_MASK);
-
-	pgd += pgd_index(address);
 	if (bad_address(pgd))
 		goto bad;
 
@@ -1027,7 +1018,7 @@ do_page_fault(struct pt_regs *regs, unsigned long error_code)
 	if (unlikely(error_code & PF_RSVD))
 		pgtable_bad(regs, error_code, address);
 
-	perf_swcounter_event(PERF_COUNT_SW_PAGE_FAULTS, 1, 0, regs, address);
+	perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, 0, regs, address);
 
 	/*
 	 * If we're in an interrupt, have no user context or are running
@@ -1124,11 +1115,11 @@ good_area:
 
 	if (fault & VM_FAULT_MAJOR) {
 		tsk->maj_flt++;
-		perf_swcounter_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1, 0,
+		perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1, 0,
 				     regs, address);
 	} else {
 		tsk->min_flt++;
-		perf_swcounter_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1, 0,
+		perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1, 0,
 				     regs, address);
 	}
 
diff --git a/arch/x86/mm/highmem_32.c b/arch/x86/mm/highmem_32.c
index 2112ed55e7ea..63a6ba66cbe0 100644
--- a/arch/x86/mm/highmem_32.c
+++ b/arch/x86/mm/highmem_32.c
@@ -24,7 +24,7 @@ void kunmap(struct page *page)
  * no global lock is needed and because the kmap code must perform a global TLB
  * invalidation when the kmap pool wraps.
  *
- * However when holding an atomic kmap is is not legal to sleep, so atomic
+ * 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)
@@ -104,6 +104,7 @@ EXPORT_SYMBOL(kunmap);
 EXPORT_SYMBOL(kmap_atomic);
 EXPORT_SYMBOL(kunmap_atomic);
 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 3cd7711bb949..30938c1d8d5d 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -84,7 +84,7 @@ static pmd_t * __init one_md_table_init(pgd_t *pgd)
 #ifdef CONFIG_X86_PAE
 	if (!(pgd_val(*pgd) & _PAGE_PRESENT)) {
 		if (after_bootmem)
-			pmd_table = (pmd_t *)alloc_bootmem_low_pages(PAGE_SIZE);
+			pmd_table = (pmd_t *)alloc_bootmem_pages(PAGE_SIZE);
 		else
 			pmd_table = (pmd_t *)alloc_low_page();
 		paravirt_alloc_pmd(&init_mm, __pa(pmd_table) >> PAGE_SHIFT);
@@ -116,7 +116,7 @@ static pte_t * __init one_page_table_init(pmd_t *pmd)
 #endif
 			if (!page_table)
 				page_table =
-				(pte_t *)alloc_bootmem_low_pages(PAGE_SIZE);
+				(pte_t *)alloc_bootmem_pages(PAGE_SIZE);
 		} else
 			page_table = (pte_t *)alloc_low_page();
 
@@ -857,8 +857,6 @@ static void __init test_wp_bit(void)
 	}
 }
 
-static struct kcore_list kcore_mem, kcore_vmalloc;
-
 void __init mem_init(void)
 {
 	int codesize, reservedpages, datasize, initsize;
@@ -886,13 +884,9 @@ void __init mem_init(void)
 	datasize =  (unsigned long) &_edata - (unsigned long) &_etext;
 	initsize =  (unsigned long) &__init_end - (unsigned long) &__init_begin;
 
-	kclist_add(&kcore_mem, __va(0), max_low_pfn << PAGE_SHIFT);
-	kclist_add(&kcore_vmalloc, (void *)VMALLOC_START,
-		   VMALLOC_END-VMALLOC_START);
-
 	printk(KERN_INFO "Memory: %luk/%luk available (%dk kernel code, "
 			"%dk reserved, %dk data, %dk init, %ldk highmem)\n",
-		(unsigned long) nr_free_pages() << (PAGE_SHIFT-10),
+		nr_free_pages() << (PAGE_SHIFT-10),
 		num_physpages << (PAGE_SHIFT-10),
 		codesize >> 10,
 		reservedpages << (PAGE_SHIFT-10),
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index ea56b8cbb6a6..5a4398a6006b 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -647,8 +647,7 @@ EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid);
 
 #endif /* CONFIG_MEMORY_HOTPLUG */
 
-static struct kcore_list kcore_mem, kcore_vmalloc, kcore_kernel,
-			 kcore_modules, kcore_vsyscall;
+static struct kcore_list kcore_vsyscall;
 
 void __init mem_init(void)
 {
@@ -677,17 +676,12 @@ void __init mem_init(void)
 	initsize =  (unsigned long) &__init_end - (unsigned long) &__init_begin;
 
 	/* Register memory areas for /proc/kcore */
-	kclist_add(&kcore_mem, __va(0), max_low_pfn << PAGE_SHIFT);
-	kclist_add(&kcore_vmalloc, (void *)VMALLOC_START,
-		   VMALLOC_END-VMALLOC_START);
-	kclist_add(&kcore_kernel, &_stext, _end - _stext);
-	kclist_add(&kcore_modules, (void *)MODULES_VADDR, MODULES_LEN);
 	kclist_add(&kcore_vsyscall, (void *)VSYSCALL_START,
-				 VSYSCALL_END - VSYSCALL_START);
+			 VSYSCALL_END - VSYSCALL_START, KCORE_OTHER);
 
 	printk(KERN_INFO "Memory: %luk/%luk available (%ldk kernel code, "
 			 "%ldk absent, %ldk reserved, %ldk data, %ldk init)\n",
-		(unsigned long) nr_free_pages() << (PAGE_SHIFT-10),
+		nr_free_pages() << (PAGE_SHIFT-10),
 		max_pfn << (PAGE_SHIFT-10),
 		codesize >> 10,
 		absent_pages << (PAGE_SHIFT-10),
diff --git a/arch/x86/mm/iomap_32.c b/arch/x86/mm/iomap_32.c
index fe6f84ca121e..84e236ce76ba 100644
--- a/arch/x86/mm/iomap_32.c
+++ b/arch/x86/mm/iomap_32.c
@@ -21,7 +21,7 @@
 #include <linux/module.h>
 #include <linux/highmem.h>
 
-int is_io_mapping_possible(resource_size_t base, unsigned long size)
+static int is_io_mapping_possible(resource_size_t base, unsigned long size)
 {
 #if !defined(CONFIG_X86_PAE) && defined(CONFIG_PHYS_ADDR_T_64BIT)
 	/* There is no way to map greater than 1 << 32 address without PAE */
@@ -30,7 +30,30 @@ int is_io_mapping_possible(resource_size_t base, unsigned long size)
 #endif
 	return 1;
 }
-EXPORT_SYMBOL_GPL(is_io_mapping_possible);
+
+int iomap_create_wc(resource_size_t base, unsigned long size, pgprot_t *prot)
+{
+	unsigned long flag = _PAGE_CACHE_WC;
+	int ret;
+
+	if (!is_io_mapping_possible(base, size))
+		return -EINVAL;
+
+	ret = io_reserve_memtype(base, base + size, &flag);
+	if (ret)
+		return ret;
+
+	*prot = __pgprot(__PAGE_KERNEL | flag);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(iomap_create_wc);
+
+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)
 {
diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index 8a450930834f..334e63ca7b2b 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -22,77 +22,7 @@
 #include <asm/pgalloc.h>
 #include <asm/pat.h>
 
-static inline int phys_addr_valid(resource_size_t addr)
-{
-#ifdef CONFIG_PHYS_ADDR_T_64BIT
-	return !(addr >> boot_cpu_data.x86_phys_bits);
-#else
-	return 1;
-#endif
-}
-
-#ifdef CONFIG_X86_64
-
-unsigned long __phys_addr(unsigned long x)
-{
-	if (x >= __START_KERNEL_map) {
-		x -= __START_KERNEL_map;
-		VIRTUAL_BUG_ON(x >= KERNEL_IMAGE_SIZE);
-		x += phys_base;
-	} else {
-		VIRTUAL_BUG_ON(x < PAGE_OFFSET);
-		x -= PAGE_OFFSET;
-		VIRTUAL_BUG_ON(!phys_addr_valid(x));
-	}
-	return x;
-}
-EXPORT_SYMBOL(__phys_addr);
-
-bool __virt_addr_valid(unsigned long x)
-{
-	if (x >= __START_KERNEL_map) {
-		x -= __START_KERNEL_map;
-		if (x >= KERNEL_IMAGE_SIZE)
-			return false;
-		x += phys_base;
-	} else {
-		if (x < PAGE_OFFSET)
-			return false;
-		x -= PAGE_OFFSET;
-		if (!phys_addr_valid(x))
-			return false;
-	}
-
-	return pfn_valid(x >> PAGE_SHIFT);
-}
-EXPORT_SYMBOL(__virt_addr_valid);
-
-#else
-
-#ifdef CONFIG_DEBUG_VIRTUAL
-unsigned long __phys_addr(unsigned long x)
-{
-	/* VMALLOC_* aren't constants  */
-	VIRTUAL_BUG_ON(x < PAGE_OFFSET);
-	VIRTUAL_BUG_ON(__vmalloc_start_set && is_vmalloc_addr((void *) x));
-	return x - PAGE_OFFSET;
-}
-EXPORT_SYMBOL(__phys_addr);
-#endif
-
-bool __virt_addr_valid(unsigned long x)
-{
-	if (x < PAGE_OFFSET)
-		return false;
-	if (__vmalloc_start_set && is_vmalloc_addr((void *) x))
-		return false;
-	if (x >= FIXADDR_START)
-		return false;
-	return pfn_valid((x - PAGE_OFFSET) >> PAGE_SHIFT);
-}
-EXPORT_SYMBOL(__virt_addr_valid);
-
-#endif
+#include "physaddr.h"
 
 int page_is_ram(unsigned long pagenr)
 {
@@ -228,24 +158,14 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr,
 	retval = reserve_memtype(phys_addr, (u64)phys_addr + size,
 						prot_val, &new_prot_val);
 	if (retval) {
-		pr_debug("Warning: reserve_memtype returned %d\n", retval);
+		printk(KERN_ERR "ioremap reserve_memtype failed %d\n", retval);
 		return NULL;
 	}
 
 	if (prot_val != new_prot_val) {
-		/*
-		 * Do not fallback to certain memory types with certain
-		 * requested type:
-		 * - request is uc-, return cannot be write-back
-		 * - request is uc-, return cannot be write-combine
-		 * - request is write-combine, return cannot be write-back
-		 */
-		if ((prot_val == _PAGE_CACHE_UC_MINUS &&
-		     (new_prot_val == _PAGE_CACHE_WB ||
-		      new_prot_val == _PAGE_CACHE_WC)) ||
-		    (prot_val == _PAGE_CACHE_WC &&
-		     new_prot_val == _PAGE_CACHE_WB)) {
-			pr_debug(
+		if (!is_new_memtype_allowed(phys_addr, size,
+					    prot_val, new_prot_val)) {
+			printk(KERN_ERR
 		"ioremap error for 0x%llx-0x%llx, requested 0x%lx, got 0x%lx\n",
 				(unsigned long long)phys_addr,
 				(unsigned long long)(phys_addr + size),
diff --git a/arch/x86/mm/kmemcheck/kmemcheck.c b/arch/x86/mm/kmemcheck/kmemcheck.c
index 2c55ed098654..8cc183344140 100644
--- a/arch/x86/mm/kmemcheck/kmemcheck.c
+++ b/arch/x86/mm/kmemcheck/kmemcheck.c
@@ -225,9 +225,6 @@ void kmemcheck_hide(struct pt_regs *regs)
 
 	BUG_ON(!irqs_disabled());
 
-	if (data->balance == 0)
-		return;
-
 	if (unlikely(data->balance != 1)) {
 		kmemcheck_show_all();
 		kmemcheck_error_save_bug(regs);
@@ -331,6 +328,20 @@ static void kmemcheck_read_strict(struct pt_regs *regs,
 	kmemcheck_shadow_set(shadow, size);
 }
 
+bool kmemcheck_is_obj_initialized(unsigned long addr, size_t size)
+{
+	enum kmemcheck_shadow status;
+	void *shadow;
+
+	shadow = kmemcheck_shadow_lookup(addr);
+	if (!shadow)
+		return true;
+
+	status = kmemcheck_shadow_test(shadow, size);
+
+	return status == KMEMCHECK_SHADOW_INITIALIZED;
+}
+
 /* Access may cross page boundary */
 static void kmemcheck_read(struct pt_regs *regs,
 	unsigned long addr, unsigned int size)
diff --git a/arch/x86/mm/kmemcheck/shadow.c b/arch/x86/mm/kmemcheck/shadow.c
index e773b6bd0079..3f66b82076a3 100644
--- a/arch/x86/mm/kmemcheck/shadow.c
+++ b/arch/x86/mm/kmemcheck/shadow.c
@@ -1,7 +1,6 @@
 #include <linux/kmemcheck.h>
 #include <linux/module.h>
 #include <linux/mm.h>
-#include <linux/module.h>
 
 #include <asm/page.h>
 #include <asm/pgtable.h>
diff --git a/arch/x86/mm/mmap.c b/arch/x86/mm/mmap.c
index 165829600566..c8191defc38a 100644
--- a/arch/x86/mm/mmap.c
+++ b/arch/x86/mm/mmap.c
@@ -29,13 +29,26 @@
 #include <linux/random.h>
 #include <linux/limits.h>
 #include <linux/sched.h>
+#include <asm/elf.h>
+
+static unsigned int stack_maxrandom_size(void)
+{
+	unsigned int max = 0;
+	if ((current->flags & PF_RANDOMIZE) &&
+		!(current->personality & ADDR_NO_RANDOMIZE)) {
+		max = ((-1U) & STACK_RND_MASK) << PAGE_SHIFT;
+	}
+
+	return max;
+}
+
 
 /*
  * Top of mmap area (just below the process stack).
  *
- * Leave an at least ~128 MB hole.
+ * Leave an at least ~128 MB hole with possible stack randomization.
  */
-#define MIN_GAP (128*1024*1024)
+#define MIN_GAP (128*1024*1024UL + stack_maxrandom_size())
 #define MAX_GAP (TASK_SIZE/6*5)
 
 /*
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 7e600c1962db..24952fdc7e40 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -12,6 +12,7 @@
 #include <linux/seq_file.h>
 #include <linux/debugfs.h>
 #include <linux/pfn.h>
+#include <linux/percpu.h>
 
 #include <asm/e820.h>
 #include <asm/processor.h>
@@ -686,7 +687,7 @@ static int cpa_process_alias(struct cpa_data *cpa)
 {
 	struct cpa_data alias_cpa;
 	unsigned long laddr = (unsigned long)__va(cpa->pfn << PAGE_SHIFT);
-	unsigned long vaddr, remapped;
+	unsigned long vaddr;
 	int ret;
 
 	if (cpa->pfn >= max_pfn_mapped)
@@ -744,24 +745,6 @@ static int cpa_process_alias(struct cpa_data *cpa)
 	}
 #endif
 
-	/*
-	 * If the PMD page was partially used for per-cpu remapping,
-	 * the recycled area needs to be split and modified.  Because
-	 * the area is always proper subset of a PMD page
-	 * cpa->numpages is guaranteed to be 1 for these areas, so
-	 * there's no need to loop over and check for further remaps.
-	 */
-	remapped = (unsigned long)pcpu_lpage_remapped((void *)laddr);
-	if (remapped) {
-		WARN_ON(cpa->numpages > 1);
-		alias_cpa = *cpa;
-		alias_cpa.vaddr = &remapped;
-		alias_cpa.flags &= ~(CPA_PAGES_ARRAY | CPA_ARRAY);
-		ret = __change_page_attr_set_clr(&alias_cpa, 0);
-		if (ret)
-			return ret;
-	}
-
 	return 0;
 }
 
@@ -822,6 +805,7 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages,
 {
 	struct cpa_data cpa;
 	int ret, cache, checkalias;
+	unsigned long baddr = 0;
 
 	/*
 	 * Check, if we are requested to change a not supported
@@ -853,6 +837,11 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages,
 			 */
 			WARN_ON_ONCE(1);
 		}
+		/*
+		 * Save address for cache flush. *addr is modified in the call
+		 * to __change_page_attr_set_clr() below.
+		 */
+		baddr = *addr;
 	}
 
 	/* Must avoid aliasing mappings in the highmem code */
@@ -900,7 +889,7 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages,
 			cpa_flush_array(addr, numpages, cache,
 					cpa.flags, pages);
 		} else
-			cpa_flush_range(*addr, numpages, cache);
+			cpa_flush_range(baddr, numpages, cache);
 	} else
 		cpa_flush_all(cache);
 
diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c
index 352aa9e927e2..7257cf3decf9 100644
--- a/arch/x86/mm/pat.c
+++ b/arch/x86/mm/pat.c
@@ -15,6 +15,7 @@
 #include <linux/gfp.h>
 #include <linux/mm.h>
 #include <linux/fs.h>
+#include <linux/rbtree.h>
 
 #include <asm/cacheflush.h>
 #include <asm/processor.h>
@@ -148,11 +149,10 @@ static char *cattr_name(unsigned long flags)
  * areas). All the aliases have the same cache attributes of course.
  * Zero attributes are represented as holes.
  *
- * Currently the data structure is a list because the number of mappings
- * are expected to be relatively small. If this should be a problem
- * it could be changed to a rbtree or similar.
+ * The data structure is a list that is also organized as an rbtree
+ * sorted on the start address of memtype range.
  *
- * memtype_lock protects the whole list.
+ * memtype_lock protects both the linear list and rbtree.
  */
 
 struct memtype {
@@ -160,11 +160,53 @@ struct memtype {
 	u64			end;
 	unsigned long		type;
 	struct list_head	nd;
+	struct rb_node		rb;
 };
 
+static struct rb_root memtype_rbroot = RB_ROOT;
 static LIST_HEAD(memtype_list);
 static DEFINE_SPINLOCK(memtype_lock);	/* protects memtype list */
 
+static struct memtype *memtype_rb_search(struct rb_root *root, u64 start)
+{
+	struct rb_node *node = root->rb_node;
+	struct memtype *last_lower = NULL;
+
+	while (node) {
+		struct memtype *data = container_of(node, struct memtype, rb);
+
+		if (data->start < start) {
+			last_lower = data;
+			node = node->rb_right;
+		} else if (data->start > start) {
+			node = node->rb_left;
+		} else
+			return data;
+	}
+
+	/* Will return NULL if there is no entry with its start <= start */
+	return last_lower;
+}
+
+static void memtype_rb_insert(struct rb_root *root, struct memtype *data)
+{
+	struct rb_node **new = &(root->rb_node);
+	struct rb_node *parent = NULL;
+
+	while (*new) {
+		struct memtype *this = container_of(*new, struct memtype, rb);
+
+		parent = *new;
+		if (data->start <= this->start)
+			new = &((*new)->rb_left);
+		else if (data->start > this->start)
+			new = &((*new)->rb_right);
+	}
+
+	rb_link_node(&data->rb, parent, new);
+	rb_insert_color(&data->rb, root);
+}
+
 /*
  * Does intersection of PAT memory type and MTRR memory type and returns
  * the resulting memory type as PAT understands it.
@@ -218,9 +260,6 @@ chk_conflict(struct memtype *new, struct memtype *entry, unsigned long *type)
 	return -EBUSY;
 }
 
-static struct memtype *cached_entry;
-static u64 cached_start;
-
 static int pat_pagerange_is_ram(unsigned long start, unsigned long end)
 {
 	int ram_page = 0, not_rampage = 0;
@@ -249,63 +288,61 @@ static int pat_pagerange_is_ram(unsigned long start, unsigned long end)
 }
 
 /*
- * For RAM pages, mark the pages as non WB memory type using
- * PageNonWB (PG_arch_1). We allow only one set_memory_uc() or
- * set_memory_wc() on a RAM page at a time before marking it as WB again.
- * This is ok, because only one driver will be owning the page and
- * doing set_memory_*() calls.
+ * For RAM pages, we use page flags to mark the pages with appropriate type.
+ * Here we do two pass:
+ * - Find the memtype of all the pages in the range, look for any conflicts
+ * - In case of no conflicts, set the new memtype for pages in the range
  *
- * For now, we use PageNonWB to track that the RAM page is being mapped
- * as non WB. In future, we will have to use one more flag
- * (or some other mechanism in page_struct) to distinguish between
- * UC and WC mapping.
+ * Caller must hold memtype_lock for atomicity.
  */
 static int reserve_ram_pages_type(u64 start, u64 end, unsigned long req_type,
 				  unsigned long *new_type)
 {
 	struct page *page;
-	u64 pfn, end_pfn;
+	u64 pfn;
+
+	if (req_type == _PAGE_CACHE_UC) {
+		/* We do not support strong UC */
+		WARN_ON_ONCE(1);
+		req_type = _PAGE_CACHE_UC_MINUS;
+	}
 
 	for (pfn = (start >> PAGE_SHIFT); pfn < (end >> PAGE_SHIFT); ++pfn) {
-		page = pfn_to_page(pfn);
-		if (page_mapped(page) || PageNonWB(page))
-			goto out;
+		unsigned long type;
 
-		SetPageNonWB(page);
+		page = pfn_to_page(pfn);
+		type = get_page_memtype(page);
+		if (type != -1) {
+			printk(KERN_INFO "reserve_ram_pages_type failed "
+				"0x%Lx-0x%Lx, track 0x%lx, req 0x%lx\n",
+				start, end, type, req_type);
+			if (new_type)
+				*new_type = type;
+
+			return -EBUSY;
+		}
 	}
-	return 0;
 
-out:
-	end_pfn = pfn;
-	for (pfn = (start >> PAGE_SHIFT); pfn < end_pfn; ++pfn) {
+	if (new_type)
+		*new_type = req_type;
+
+	for (pfn = (start >> PAGE_SHIFT); pfn < (end >> PAGE_SHIFT); ++pfn) {
 		page = pfn_to_page(pfn);
-		ClearPageNonWB(page);
+		set_page_memtype(page, req_type);
 	}
-
-	return -EINVAL;
+	return 0;
 }
 
 static int free_ram_pages_type(u64 start, u64 end)
 {
 	struct page *page;
-	u64 pfn, end_pfn;
+	u64 pfn;
 
 	for (pfn = (start >> PAGE_SHIFT); pfn < (end >> PAGE_SHIFT); ++pfn) {
 		page = pfn_to_page(pfn);
-		if (page_mapped(page) || !PageNonWB(page))
-			goto out;
-
-		ClearPageNonWB(page);
+		set_page_memtype(page, -1);
 	}
 	return 0;
-
-out:
-	end_pfn = pfn;
-	for (pfn = (start >> PAGE_SHIFT); pfn < end_pfn; ++pfn) {
-		page = pfn_to_page(pfn);
-		SetPageNonWB(page);
-	}
-	return -EINVAL;
 }
 
 /*
@@ -339,6 +376,8 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type,
 		if (new_type) {
 			if (req_type == -1)
 				*new_type = _PAGE_CACHE_WB;
+			else if (req_type == _PAGE_CACHE_WC)
+				*new_type = _PAGE_CACHE_UC_MINUS;
 			else
 				*new_type = req_type & _PAGE_CACHE_MASK;
 		}
@@ -364,11 +403,16 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type,
 		*new_type = actual_type;
 
 	is_range_ram = pat_pagerange_is_ram(start, end);
-	if (is_range_ram == 1)
-		return reserve_ram_pages_type(start, end, req_type,
-					      new_type);
-	else if (is_range_ram < 0)
+	if (is_range_ram == 1) {
+
+		spin_lock(&memtype_lock);
+		err = reserve_ram_pages_type(start, end, req_type, new_type);
+		spin_unlock(&memtype_lock);
+
+		return err;
+	} else if (is_range_ram < 0) {
 		return -EINVAL;
+	}
 
 	new  = kmalloc(sizeof(struct memtype), GFP_KERNEL);
 	if (!new)
@@ -380,17 +424,11 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type,
 
 	spin_lock(&memtype_lock);
 
-	if (cached_entry && start >= cached_start)
-		entry = cached_entry;
-	else
-		entry = list_entry(&memtype_list, struct memtype, nd);
-
 	/* Search for existing mapping that overlaps the current range */
 	where = NULL;
-	list_for_each_entry_continue(entry, &memtype_list, nd) {
+	list_for_each_entry(entry, &memtype_list, nd) {
 		if (end <= entry->start) {
 			where = entry->nd.prev;
-			cached_entry = list_entry(where, struct memtype, nd);
 			break;
 		} else if (start <= entry->start) { /* end > entry->start */
 			err = chk_conflict(new, entry, new_type);
@@ -398,8 +436,6 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type,
 				dprintk("Overlap at 0x%Lx-0x%Lx\n",
 					entry->start, entry->end);
 				where = entry->nd.prev;
-				cached_entry = list_entry(where,
-							struct memtype, nd);
 			}
 			break;
 		} else if (start < entry->end) { /* start > entry->start */
@@ -407,8 +443,6 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type,
 			if (!err) {
 				dprintk("Overlap at 0x%Lx-0x%Lx\n",
 					entry->start, entry->end);
-				cached_entry = list_entry(entry->nd.prev,
-							struct memtype, nd);
 
 				/*
 				 * Move to right position in the linked
@@ -436,13 +470,13 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type,
 		return err;
 	}
 
-	cached_start = start;
-
 	if (where)
 		list_add(&new->nd, where);
 	else
 		list_add_tail(&new->nd, &memtype_list);
 
+	memtype_rb_insert(&memtype_rbroot, new);
+
 	spin_unlock(&memtype_lock);
 
 	dprintk("reserve_memtype added 0x%Lx-0x%Lx, track %s, req %s, ret %s\n",
@@ -454,7 +488,7 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type,
 
 int free_memtype(u64 start, u64 end)
 {
-	struct memtype *entry;
+	struct memtype *entry, *saved_entry;
 	int err = -EINVAL;
 	int is_range_ram;
 
@@ -466,23 +500,58 @@ int free_memtype(u64 start, u64 end)
 		return 0;
 
 	is_range_ram = pat_pagerange_is_ram(start, end);
-	if (is_range_ram == 1)
-		return free_ram_pages_type(start, end);
-	else if (is_range_ram < 0)
+	if (is_range_ram == 1) {
+
+		spin_lock(&memtype_lock);
+		err = free_ram_pages_type(start, end);
+		spin_unlock(&memtype_lock);
+
+		return err;
+	} else if (is_range_ram < 0) {
 		return -EINVAL;
+	}
 
 	spin_lock(&memtype_lock);
-	list_for_each_entry(entry, &memtype_list, nd) {
+
+	entry = memtype_rb_search(&memtype_rbroot, start);
+	if (unlikely(entry == NULL))
+		goto unlock_ret;
+
+	/*
+	 * Saved entry points to an entry with start same or less than what
+	 * we searched for. Now go through the list in both directions to look
+	 * for the entry that matches with both start and end, with list stored
+	 * in sorted start address
+	 */
+	saved_entry = entry;
+	list_for_each_entry_from(entry, &memtype_list, nd) {
 		if (entry->start == start && entry->end == end) {
-			if (cached_entry == entry || cached_start == start)
-				cached_entry = NULL;
+			rb_erase(&entry->rb, &memtype_rbroot);
+			list_del(&entry->nd);
+			kfree(entry);
+			err = 0;
+			break;
+		} else if (entry->start > start) {
+			break;
+		}
+	}
+
+	if (!err)
+		goto unlock_ret;
 
+	entry = saved_entry;
+	list_for_each_entry_reverse(entry, &memtype_list, nd) {
+		if (entry->start == start && entry->end == end) {
+			rb_erase(&entry->rb, &memtype_rbroot);
 			list_del(&entry->nd);
 			kfree(entry);
 			err = 0;
 			break;
+		} else if (entry->start < start) {
+			break;
 		}
 	}
+unlock_ret:
 	spin_unlock(&memtype_lock);
 
 	if (err) {
@@ -496,6 +565,101 @@ int free_memtype(u64 start, u64 end)
 }
 
 
+/**
+ * lookup_memtype - Looksup the memory type for a physical address
+ * @paddr: physical address of which memory type needs to be looked up
+ *
+ * Only to be called when PAT is enabled
+ *
+ * Returns _PAGE_CACHE_WB, _PAGE_CACHE_WC, _PAGE_CACHE_UC_MINUS or
+ * _PAGE_CACHE_UC
+ */
+static unsigned long lookup_memtype(u64 paddr)
+{
+	int rettype = _PAGE_CACHE_WB;
+	struct memtype *entry;
+
+	if (is_ISA_range(paddr, paddr + PAGE_SIZE - 1))
+		return rettype;
+
+	if (pat_pagerange_is_ram(paddr, paddr + PAGE_SIZE)) {
+		struct page *page;
+		spin_lock(&memtype_lock);
+		page = pfn_to_page(paddr >> PAGE_SHIFT);
+		rettype = get_page_memtype(page);
+		spin_unlock(&memtype_lock);
+		/*
+		 * -1 from get_page_memtype() implies RAM page is in its
+		 * default state and not reserved, and hence of type WB
+		 */
+		if (rettype == -1)
+			rettype = _PAGE_CACHE_WB;
+
+		return rettype;
+	}
+
+	spin_lock(&memtype_lock);
+
+	entry = memtype_rb_search(&memtype_rbroot, paddr);
+	if (entry != NULL)
+		rettype = entry->type;
+	else
+		rettype = _PAGE_CACHE_UC_MINUS;
+
+	spin_unlock(&memtype_lock);
+	return rettype;
+}
+
+/**
+ * io_reserve_memtype - Request a memory type mapping for a region of memory
+ * @start: start (physical address) of the region
+ * @end: end (physical address) of the region
+ * @type: A pointer to memtype, with requested type. On success, requested
+ * or any other compatible type that was available for the region is returned
+ *
+ * On success, returns 0
+ * On failure, returns non-zero
+ */
+int io_reserve_memtype(resource_size_t start, resource_size_t end,
+			unsigned long *type)
+{
+	resource_size_t size = end - start;
+	unsigned long req_type = *type;
+	unsigned long new_type;
+	int ret;
+
+	WARN_ON_ONCE(iomem_map_sanity_check(start, size));
+
+	ret = reserve_memtype(start, end, req_type, &new_type);
+	if (ret)
+		goto out_err;
+
+	if (!is_new_memtype_allowed(start, size, req_type, new_type))
+		goto out_free;
+
+	if (kernel_map_sync_memtype(start, size, new_type) < 0)
+		goto out_free;
+
+	*type = new_type;
+	return 0;
+
+out_free:
+	free_memtype(start, end);
+	ret = -EBUSY;
+out_err:
+	return ret;
+}
+
+/**
+ * io_free_memtype - Release a memory type mapping for a region of memory
+ * @start: start (physical address) of the region
+ * @end: end (physical address) of the region
+ */
+void io_free_memtype(resource_size_t start, resource_size_t end)
+{
+	free_memtype(start, end);
+}
+
 pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
 				unsigned long size, pgprot_t vma_prot)
 {
@@ -577,7 +741,7 @@ int kernel_map_sync_memtype(u64 base, unsigned long size, unsigned long flags)
 {
 	unsigned long id_sz;
 
-	if (!pat_enabled || base >= __pa(high_memory))
+	if (base >= __pa(high_memory))
 		return 0;
 
 	id_sz = (__pa(high_memory) < base + size) ?
@@ -612,11 +776,29 @@ static int reserve_pfn_range(u64 paddr, unsigned long size, pgprot_t *vma_prot,
 	is_ram = pat_pagerange_is_ram(paddr, paddr + size);
 
 	/*
-	 * reserve_pfn_range() doesn't support RAM pages. Maintain the current
-	 * behavior with RAM pages by returning success.
+	 * reserve_pfn_range() for RAM pages. We do not refcount to keep
+	 * track of number of mappings of RAM pages. We can assert that
+	 * the type requested matches the type of first page in the range.
 	 */
-	if (is_ram != 0)
+	if (is_ram) {
+		if (!pat_enabled)
+			return 0;
+
+		flags = lookup_memtype(paddr);
+		if (want_flags != flags) {
+			printk(KERN_WARNING
+			"%s:%d map pfn RAM range req %s for %Lx-%Lx, got %s\n",
+				current->comm, current->pid,
+				cattr_name(want_flags),
+				(unsigned long long)paddr,
+				(unsigned long long)(paddr + size),
+				cattr_name(flags));
+			*vma_prot = __pgprot((pgprot_val(*vma_prot) &
+					      (~_PAGE_CACHE_MASK)) |
+					     flags);
+		}
 		return 0;
+	}
 
 	ret = reserve_memtype(paddr, paddr + size, want_flags, &flags);
 	if (ret)
@@ -678,14 +860,6 @@ int track_pfn_vma_copy(struct vm_area_struct *vma)
 	unsigned long vma_size = vma->vm_end - vma->vm_start;
 	pgprot_t pgprot;
 
-	if (!pat_enabled)
-		return 0;
-
-	/*
-	 * For now, only handle remap_pfn_range() vmas where
-	 * is_linear_pfn_mapping() == TRUE. Handling of
-	 * vm_insert_pfn() is TBD.
-	 */
 	if (is_linear_pfn_mapping(vma)) {
 		/*
 		 * reserve the whole chunk covered by vma. We need the
@@ -713,23 +887,24 @@ int track_pfn_vma_copy(struct vm_area_struct *vma)
 int track_pfn_vma_new(struct vm_area_struct *vma, pgprot_t *prot,
 			unsigned long pfn, unsigned long size)
 {
+	unsigned long flags;
 	resource_size_t paddr;
 	unsigned long vma_size = vma->vm_end - vma->vm_start;
 
-	if (!pat_enabled)
-		return 0;
-
-	/*
-	 * For now, only handle remap_pfn_range() vmas where
-	 * is_linear_pfn_mapping() == TRUE. Handling of
-	 * vm_insert_pfn() is TBD.
-	 */
 	if (is_linear_pfn_mapping(vma)) {
 		/* reserve the whole chunk starting from vm_pgoff */
 		paddr = (resource_size_t)vma->vm_pgoff << PAGE_SHIFT;
 		return reserve_pfn_range(paddr, vma_size, prot, 0);
 	}
 
+	if (!pat_enabled)
+		return 0;
+
+	/* for vm_insert_pfn and friends, we set prot based on lookup */
+	flags = lookup_memtype(pfn << PAGE_SHIFT);
+	*prot = __pgprot((pgprot_val(vma->vm_page_prot) & (~_PAGE_CACHE_MASK)) |
+			 flags);
+
 	return 0;
 }
 
@@ -744,14 +919,6 @@ void untrack_pfn_vma(struct vm_area_struct *vma, unsigned long pfn,
 	resource_size_t paddr;
 	unsigned long vma_size = vma->vm_end - vma->vm_start;
 
-	if (!pat_enabled)
-		return;
-
-	/*
-	 * For now, only handle remap_pfn_range() vmas where
-	 * is_linear_pfn_mapping() == TRUE. Handling of
-	 * vm_insert_pfn() is TBD.
-	 */
 	if (is_linear_pfn_mapping(vma)) {
 		/* free the whole chunk starting from vm_pgoff */
 		paddr = (resource_size_t)vma->vm_pgoff << PAGE_SHIFT;
@@ -827,7 +994,7 @@ static int memtype_seq_show(struct seq_file *seq, void *v)
 	return 0;
 }
 
-static struct seq_operations memtype_seq_ops = {
+static const struct seq_operations memtype_seq_ops = {
 	.start = memtype_seq_start,
 	.next  = memtype_seq_next,
 	.stop  = memtype_seq_stop,
diff --git a/arch/x86/mm/physaddr.c b/arch/x86/mm/physaddr.c
new file mode 100644
index 000000000000..d2e2735327b4
--- /dev/null
+++ b/arch/x86/mm/physaddr.c
@@ -0,0 +1,70 @@
+#include <linux/mmdebug.h>
+#include <linux/module.h>
+#include <linux/mm.h>
+
+#include <asm/page.h>
+
+#include "physaddr.h"
+
+#ifdef CONFIG_X86_64
+
+unsigned long __phys_addr(unsigned long x)
+{
+	if (x >= __START_KERNEL_map) {
+		x -= __START_KERNEL_map;
+		VIRTUAL_BUG_ON(x >= KERNEL_IMAGE_SIZE);
+		x += phys_base;
+	} else {
+		VIRTUAL_BUG_ON(x < PAGE_OFFSET);
+		x -= PAGE_OFFSET;
+		VIRTUAL_BUG_ON(!phys_addr_valid(x));
+	}
+	return x;
+}
+EXPORT_SYMBOL(__phys_addr);
+
+bool __virt_addr_valid(unsigned long x)
+{
+	if (x >= __START_KERNEL_map) {
+		x -= __START_KERNEL_map;
+		if (x >= KERNEL_IMAGE_SIZE)
+			return false;
+		x += phys_base;
+	} else {
+		if (x < PAGE_OFFSET)
+			return false;
+		x -= PAGE_OFFSET;
+		if (!phys_addr_valid(x))
+			return false;
+	}
+
+	return pfn_valid(x >> PAGE_SHIFT);
+}
+EXPORT_SYMBOL(__virt_addr_valid);
+
+#else
+
+#ifdef CONFIG_DEBUG_VIRTUAL
+unsigned long __phys_addr(unsigned long x)
+{
+	/* VMALLOC_* aren't constants  */
+	VIRTUAL_BUG_ON(x < PAGE_OFFSET);
+	VIRTUAL_BUG_ON(__vmalloc_start_set && is_vmalloc_addr((void *) x));
+	return x - PAGE_OFFSET;
+}
+EXPORT_SYMBOL(__phys_addr);
+#endif
+
+bool __virt_addr_valid(unsigned long x)
+{
+	if (x < PAGE_OFFSET)
+		return false;
+	if (__vmalloc_start_set && is_vmalloc_addr((void *) x))
+		return false;
+	if (x >= FIXADDR_START)
+		return false;
+	return pfn_valid((x - PAGE_OFFSET) >> PAGE_SHIFT);
+}
+EXPORT_SYMBOL(__virt_addr_valid);
+
+#endif	/* CONFIG_X86_64 */
diff --git a/arch/x86/mm/physaddr.h b/arch/x86/mm/physaddr.h
new file mode 100644
index 000000000000..a3cd5a0c97b3
--- /dev/null
+++ b/arch/x86/mm/physaddr.h
@@ -0,0 +1,10 @@
+#include <asm/processor.h>
+
+static inline int phys_addr_valid(resource_size_t addr)
+{
+#ifdef CONFIG_PHYS_ADDR_T_64BIT
+	return !(addr >> boot_cpu_data.x86_phys_bits);
+#else
+	return 1;
+#endif
+}
diff --git a/arch/x86/mm/srat_32.c b/arch/x86/mm/srat_32.c
index 29a0e37114f8..6f8aa33031c7 100644
--- a/arch/x86/mm/srat_32.c
+++ b/arch/x86/mm/srat_32.c
@@ -215,7 +215,7 @@ int __init get_memcfg_from_srat(void)
 		goto out_fail;
 
 	if (num_memory_chunks == 0) {
-		printk(KERN_WARNING
+		printk(KERN_DEBUG
 			 "could not find any ACPI SRAT memory areas.\n");
 		goto out_fail;
 	}
@@ -277,7 +277,7 @@ int __init get_memcfg_from_srat(void)
 	}
 	return 1;
 out_fail:
-	printk(KERN_ERR "failed to get NUMA memory information from SRAT"
+	printk(KERN_DEBUG "failed to get NUMA memory information from SRAT"
 			" table\n");
 	return 0;
 }