Merge branch 'for-linus2' of git://git.kernel.org/pub/scm/linux/kerne…

…l/git/vegard/kmemcheck

* 'for-linus2' of git://git.kernel.org/pub/scm/linux/kernel/git/vegard/kmemcheck: (39 commits)
  signal: fix __send_signal() false positive kmemcheck warning
  fs: fix do_mount_root() false positive kmemcheck warning
  fs: introduce __getname_gfp()
  trace: annotate bitfields in struct ring_buffer_event
  net: annotate struct sock bitfield
  c2port: annotate bitfield for kmemcheck
  net: annotate inet_timewait_sock bitfields
  ieee1394/csr1212: fix false positive kmemcheck report
  ieee1394: annotate bitfield
  net: annotate bitfields in struct inet_sock
  net: use kmemcheck bitfields API for skbuff
  kmemcheck: introduce bitfield API
  kmemcheck: add opcode self-testing at boot
  x86: unify pte_hidden
  x86: make _PAGE_HIDDEN conditional
  kmemcheck: make kconfig accessible for other architectures
  kmemcheck: enable in the x86 Kconfig
  kmemcheck: add hooks for the page allocator
  kmemcheck: add hooks for page- and sg-dma-mappings
  kmemcheck: don't track page tables
  ...

MAINTAINERS

@@ -3406,6 +3406,14 @@ F:	drivers/serial/kgdboc.c
 F:	include/linux/kgdb.h
 F:	kernel/kgdb.c
 
+KMEMCHECK
+P:	Vegard Nossum
+M:	[email protected]
+P	Pekka Enberg
+M:	[email protected]
+L:	[email protected]
+S:	Maintained
+
 KMEMLEAK
 P:	Catalin Marinas
 M:	[email protected]

arch/x86/Kconfig

@@ -46,6 +46,7 @@ config X86
 	select HAVE_KERNEL_GZIP
 	select HAVE_KERNEL_BZIP2
 	select HAVE_KERNEL_LZMA
+	select HAVE_ARCH_KMEMCHECK
 
 config OUTPUT_FORMAT
 	string

arch/x86/Makefile

@@ -81,6 +81,11 @@ ifdef CONFIG_CC_STACKPROTECTOR
         endif
 endif
 
+# Don't unroll struct assignments with kmemcheck enabled
+ifeq ($(CONFIG_KMEMCHECK),y)
+	KBUILD_CFLAGS += $(call cc-option,-fno-builtin-memcpy)
+endif
+
 # Stackpointer is addressed different for 32 bit and 64 bit x86
 sp-$(CONFIG_X86_32) := esp
 sp-$(CONFIG_X86_64) := rsp

arch/x86/include/asm/dma-mapping.h

@@ -6,6 +6,7 @@
  * Documentation/DMA-API.txt for documentation.
  */
 
+#include <linux/kmemcheck.h>
 #include <linux/scatterlist.h>
 #include <linux/dma-debug.h>
 #include <linux/dma-attrs.h>
@@ -60,6 +61,7 @@ dma_map_single(struct device *hwdev, void *ptr, size_t size,
 	dma_addr_t addr;
 
 	BUG_ON(!valid_dma_direction(dir));
+	kmemcheck_mark_initialized(ptr, size);
 	addr = ops->map_page(hwdev, virt_to_page(ptr),
 			     (unsigned long)ptr & ~PAGE_MASK, size,
 			     dir, NULL);
@@ -87,8 +89,12 @@ dma_map_sg(struct device *hwdev, struct scatterlist *sg,
 {
 	struct dma_map_ops *ops = get_dma_ops(hwdev);
 	int ents;
+	struct scatterlist *s;
+	int i;
 
 	BUG_ON(!valid_dma_direction(dir));
+	for_each_sg(sg, s, nents, i)
+		kmemcheck_mark_initialized(sg_virt(s), s->length);
 	ents = ops->map_sg(hwdev, sg, nents, dir, NULL);
 	debug_dma_map_sg(hwdev, sg, nents, ents, dir);
 
@@ -200,6 +206,7 @@ static inline dma_addr_t dma_map_page(struct device *dev, struct page *page,
 	dma_addr_t addr;
 
 	BUG_ON(!valid_dma_direction(dir));
+	kmemcheck_mark_initialized(page_address(page) + offset, size);
 	addr = ops->map_page(dev, page, offset, size, dir, NULL);
 	debug_dma_map_page(dev, page, offset, size, dir, addr, false);
 

arch/x86/include/asm/kmemcheck.h

@@ -0,0 +1,42 @@
+#ifndef ASM_X86_KMEMCHECK_H
+#define ASM_X86_KMEMCHECK_H
+
+#include <linux/types.h>
+#include <asm/ptrace.h>
+
+#ifdef CONFIG_KMEMCHECK
+bool kmemcheck_active(struct pt_regs *regs);
+
+void kmemcheck_show(struct pt_regs *regs);
+void kmemcheck_hide(struct pt_regs *regs);
+
+bool kmemcheck_fault(struct pt_regs *regs,
+	unsigned long address, unsigned long error_code);
+bool kmemcheck_trap(struct pt_regs *regs);
+#else
+static inline bool kmemcheck_active(struct pt_regs *regs)
+{
+	return false;
+}
+
+static inline void kmemcheck_show(struct pt_regs *regs)
+{
+}
+
+static inline void kmemcheck_hide(struct pt_regs *regs)
+{
+}
+
+static inline bool kmemcheck_fault(struct pt_regs *regs,
+	unsigned long address, unsigned long error_code)
+{
+	return false;
+}
+
+static inline bool kmemcheck_trap(struct pt_regs *regs)
+{
+	return false;
+}
+#endif /* CONFIG_KMEMCHECK */
+
+#endif

arch/x86/include/asm/pgtable.h

@@ -317,6 +317,11 @@ static inline int pte_present(pte_t a)
 	return pte_flags(a) & (_PAGE_PRESENT | _PAGE_PROTNONE);
 }
 
+static inline int pte_hidden(pte_t pte)
+{
+	return pte_flags(pte) & _PAGE_HIDDEN;
+}
+
 static inline int pmd_present(pmd_t pmd)
 {
 	return pmd_flags(pmd) & _PAGE_PRESENT;

arch/x86/include/asm/pgtable_types.h

@@ -18,7 +18,7 @@
 #define _PAGE_BIT_GLOBAL	8	/* Global TLB entry PPro+ */
 #define _PAGE_BIT_UNUSED1	9	/* available for programmer */
 #define _PAGE_BIT_IOMAP		10	/* flag used to indicate IO mapping */
-#define _PAGE_BIT_UNUSED3	11
+#define _PAGE_BIT_HIDDEN	11	/* hidden by kmemcheck */
 #define _PAGE_BIT_PAT_LARGE	12	/* On 2MB or 1GB pages */
 #define _PAGE_BIT_SPECIAL	_PAGE_BIT_UNUSED1
 #define _PAGE_BIT_CPA_TEST	_PAGE_BIT_UNUSED1
@@ -41,13 +41,18 @@
 #define _PAGE_GLOBAL	(_AT(pteval_t, 1) << _PAGE_BIT_GLOBAL)
 #define _PAGE_UNUSED1	(_AT(pteval_t, 1) << _PAGE_BIT_UNUSED1)
 #define _PAGE_IOMAP	(_AT(pteval_t, 1) << _PAGE_BIT_IOMAP)
-#define _PAGE_UNUSED3	(_AT(pteval_t, 1) << _PAGE_BIT_UNUSED3)
 #define _PAGE_PAT	(_AT(pteval_t, 1) << _PAGE_BIT_PAT)
 #define _PAGE_PAT_LARGE (_AT(pteval_t, 1) << _PAGE_BIT_PAT_LARGE)
 #define _PAGE_SPECIAL	(_AT(pteval_t, 1) << _PAGE_BIT_SPECIAL)
 #define _PAGE_CPA_TEST	(_AT(pteval_t, 1) << _PAGE_BIT_CPA_TEST)
 #define __HAVE_ARCH_PTE_SPECIAL
 
+#ifdef CONFIG_KMEMCHECK
+#define _PAGE_HIDDEN	(_AT(pteval_t, 1) << _PAGE_BIT_HIDDEN)
+#else
+#define _PAGE_HIDDEN	(_AT(pteval_t, 0))
+#endif
+
 #if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE)
 #define _PAGE_NX	(_AT(pteval_t, 1) << _PAGE_BIT_NX)
 #else

arch/x86/include/asm/string_32.h

@@ -177,10 +177,18 @@ static inline void *__memcpy3d(void *to, const void *from, size_t len)
  *	No 3D Now!
  */
 
+#ifndef CONFIG_KMEMCHECK
 #define memcpy(t, f, n)				\
 	(__builtin_constant_p((n))		\
 	 ? __constant_memcpy((t), (f), (n))	\
 	 : __memcpy((t), (f), (n)))
+#else
+/*
+ * kmemcheck becomes very happy if we use the REP instructions unconditionally,
+ * because it means that we know both memory operands in advance.
+ */
+#define memcpy(t, f, n) __memcpy((t), (f), (n))
+#endif
 
 #endif
 

arch/x86/include/asm/string_64.h

@@ -27,6 +27,7 @@ static __always_inline void *__inline_memcpy(void *to, const void *from, size_t
    function. */
 
 #define __HAVE_ARCH_MEMCPY 1
+#ifndef CONFIG_KMEMCHECK
 #if (__GNUC__ == 4 && __GNUC_MINOR__ >= 3) || __GNUC__ > 4
 extern void *memcpy(void *to, const void *from, size_t len);
 #else
@@ -42,6 +43,13 @@ extern void *__memcpy(void *to, const void *from, size_t len);
 	__ret;							\
 })
 #endif
+#else
+/*
+ * kmemcheck becomes very happy if we use the REP instructions unconditionally,
+ * because it means that we know both memory operands in advance.
+ */
+#define memcpy(dst, src, len) __inline_memcpy((dst), (src), (len))
+#endif
 
 #define __HAVE_ARCH_MEMSET
 void *memset(void *s, int c, size_t n);

arch/x86/include/asm/thread_info.h

@@ -154,9 +154,9 @@ struct thread_info {
 
 /* thread information allocation */
 #ifdef CONFIG_DEBUG_STACK_USAGE
-#define THREAD_FLAGS (GFP_KERNEL | __GFP_ZERO)
+#define THREAD_FLAGS (GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO)
 #else
-#define THREAD_FLAGS GFP_KERNEL
+#define THREAD_FLAGS (GFP_KERNEL | __GFP_NOTRACK)
 #endif
 
 #define __HAVE_ARCH_THREAD_INFO_ALLOCATOR

arch/x86/include/asm/xor.h

@@ -1,5 +1,10 @@
+#ifdef CONFIG_KMEMCHECK
+/* kmemcheck doesn't handle MMX/SSE/SSE2 instructions */
+# include <asm-generic/xor.h>
+#else
 #ifdef CONFIG_X86_32
 # include "xor_32.h"
 #else
 # include "xor_64.h"
 #endif
+#endif

arch/x86/kernel/cpu/intel.c

@@ -86,6 +86,29 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c)
 	 */
 	if (c->x86 == 6 && c->x86_model < 15)
 		clear_cpu_cap(c, X86_FEATURE_PAT);
+
+#ifdef CONFIG_KMEMCHECK
+	/*
+	 * P4s have a "fast strings" feature which causes single-
+	 * stepping REP instructions to only generate a #DB on
+	 * cache-line boundaries.
+	 *
+	 * Ingo Molnar reported a Pentium D (model 6) and a Xeon
+	 * (model 2) with the same problem.
+	 */
+	if (c->x86 == 15) {
+		u64 misc_enable;
+
+		rdmsrl(MSR_IA32_MISC_ENABLE, misc_enable);
+
+		if (misc_enable & MSR_IA32_MISC_ENABLE_FAST_STRING) {
+			printk(KERN_INFO "kmemcheck: Disabling fast string operations\n");
+
+			misc_enable &= ~MSR_IA32_MISC_ENABLE_FAST_STRING;
+			wrmsrl(MSR_IA32_MISC_ENABLE, misc_enable);
+		}
+	}
+#endif
 }
 
 #ifdef CONFIG_X86_32

arch/x86/kernel/process.c

@@ -63,7 +63,7 @@ void arch_task_cache_init(void)
         task_xstate_cachep =
         	kmem_cache_create("task_xstate", xstate_size,
 				  __alignof__(union thread_xstate),
-				  SLAB_PANIC, NULL);
+				  SLAB_PANIC | SLAB_NOTRACK, NULL);
 }
 
 /*

arch/x86/kernel/stacktrace.c

@@ -77,6 +77,13 @@ void save_stack_trace(struct stack_trace *trace)
 }
 EXPORT_SYMBOL_GPL(save_stack_trace);
 
+void save_stack_trace_bp(struct stack_trace *trace, unsigned long bp)
+{
+	dump_trace(current, NULL, NULL, bp, &save_stack_ops, trace);
+	if (trace->nr_entries < trace->max_entries)
+		trace->entries[trace->nr_entries++] = ULONG_MAX;
+}
+
 void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
 {
 	dump_trace(tsk, NULL, NULL, 0, &save_stack_ops_nosched, trace);

arch/x86/kernel/traps.c

@@ -45,6 +45,7 @@
 #include <linux/edac.h>
 #endif
 
+#include <asm/kmemcheck.h>
 #include <asm/stacktrace.h>
 #include <asm/processor.h>
 #include <asm/debugreg.h>
@@ -534,6 +535,10 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code)
 
 	get_debugreg(condition, 6);
 
+	/* Catch kmemcheck conditions first of all! */
+	if (condition & DR_STEP && kmemcheck_trap(regs))
+		return;
+
 	/*
 	 * The processor cleared BTF, so don't mark that we need it set.
 	 */

arch/x86/mm/Makefile

@@ -10,6 +10,8 @@ obj-$(CONFIG_X86_PTDUMP)	+= dump_pagetables.o
 
 obj-$(CONFIG_HIGHMEM)		+= highmem_32.o
 
+obj-$(CONFIG_KMEMCHECK)		+= kmemcheck/
+
 obj-$(CONFIG_MMIOTRACE)		+= mmiotrace.o
 mmiotrace-y			:= kmmio.o pf_in.o mmio-mod.o
 obj-$(CONFIG_MMIOTRACE_TEST)	+= testmmiotrace.o

arch/x86/mm/fault.c

@@ -14,6 +14,7 @@
 
 #include <asm/traps.h>			/* dotraplinkage, ...		*/
 #include <asm/pgalloc.h>		/* pgd_*(), ...			*/
+#include <asm/kmemcheck.h>		/* kmemcheck_*(), ...		*/
 
 /*
  * Page fault error code bits:
@@ -956,6 +957,13 @@ do_page_fault(struct pt_regs *regs, unsigned long error_code)
 	/* Get the faulting address: */
 	address = read_cr2();
 
+	/*
+	 * Detect and handle instructions that would cause a page fault for
+	 * both a tracked kernel page and a userspace page.
+	 */
+	if (kmemcheck_active(regs))
+		kmemcheck_hide(regs);
+
 	if (unlikely(kmmio_fault(regs, address)))
 		return;
 
@@ -973,9 +981,13 @@ do_page_fault(struct pt_regs *regs, unsigned long error_code)
 	 * protection error (error_code & 9) == 0.
 	 */
 	if (unlikely(fault_in_kernel_space(address))) {
-		if (!(error_code & (PF_RSVD|PF_USER|PF_PROT)) &&
-		    vmalloc_fault(address) >= 0)
-			return;
+		if (!(error_code & (PF_RSVD | PF_USER | PF_PROT))) {
+			if (vmalloc_fault(address) >= 0)
+				return;
+
+			if (kmemcheck_fault(regs, address, error_code))
+				return;
+		}
 
 		/* Can handle a stale RO->RW TLB: */
 		if (spurious_fault(error_code, address))

arch/x86/mm/init.c

@@ -213,7 +213,7 @@ unsigned long __init_refok init_memory_mapping(unsigned long start,
 	if (!after_bootmem)
 		init_gbpages();
 
-#ifdef CONFIG_DEBUG_PAGEALLOC
+#if defined(CONFIG_DEBUG_PAGEALLOC) || defined(CONFIG_KMEMCHECK)
 	/*
 	 * For CONFIG_DEBUG_PAGEALLOC, identity mapping will use small pages.
 	 * This will simplify cpa(), which otherwise needs to support splitting

arch/x86/mm/init_32.c

@@ -111,7 +111,7 @@ static pte_t * __init one_page_table_init(pmd_t *pmd)
 		pte_t *page_table = NULL;
 
 		if (after_bootmem) {
-#ifdef CONFIG_DEBUG_PAGEALLOC
+#if defined(CONFIG_DEBUG_PAGEALLOC) || defined(CONFIG_KMEMCHECK)
 			page_table = (pte_t *) alloc_bootmem_pages(PAGE_SIZE);
 #endif
 			if (!page_table)

arch/x86/mm/init_64.c

@@ -104,7 +104,7 @@ static __ref void *spp_getpage(void)
 	void *ptr;
 
 	if (after_bootmem)
-		ptr = (void *) get_zeroed_page(GFP_ATOMIC);
+		ptr = (void *) get_zeroed_page(GFP_ATOMIC | __GFP_NOTRACK);
 	else
 		ptr = alloc_bootmem_pages(PAGE_SIZE);
 
@@ -281,7 +281,7 @@ static __ref void *alloc_low_page(unsigned long *phys)
 	void *adr;
 
 	if (after_bootmem) {
-		adr = (void *)get_zeroed_page(GFP_ATOMIC);
+		adr = (void *)get_zeroed_page(GFP_ATOMIC | __GFP_NOTRACK);
 		*phys = __pa(adr);
 
 		return adr;

arch/x86/mm/kmemcheck/Makefile

@@ -0,0 +1 @@
+obj-y := error.o kmemcheck.o opcode.o pte.o selftest.o shadow.o