Merge remote-tracking branch 'common/android-4.4' into android-4.4.y

android/configs/android-base.cfg

@@ -141,6 +141,7 @@ CONFIG_PROFILING=y
 CONFIG_QUOTA=y
 CONFIG_RTC_CLASS=y
 CONFIG_RT_GROUP_SCHED=y
+CONFIG_SECCOMP=y
 CONFIG_SECURITY=y
 CONFIG_SECURITY_NETWORK=y
 CONFIG_SECURITY_PERF_EVENTS_RESTRICT=y

arch/Kconfig

@@ -423,6 +423,15 @@ config CC_STACKPROTECTOR_STRONG
 
 endchoice
 
+config HAVE_ARCH_WITHIN_STACK_FRAMES
+	bool
+	help
+	  An architecture should select this if it can walk the kernel stack
+	  frames to determine if an object is part of either the arguments
+	  or local variables (i.e. that it excludes saved return addresses,
+	  and similar) by implementing an inline arch_within_stack_frames(),
+	  which is used by CONFIG_HARDENED_USERCOPY.
+
 config HAVE_CONTEXT_TRACKING
 	bool
 	help

arch/arm/Kconfig

@@ -33,6 +33,7 @@ config ARM
 	select HARDIRQS_SW_RESEND
 	select HAVE_ARCH_AUDITSYSCALL if (AEABI && !OABI_COMPAT)
 	select HAVE_ARCH_BITREVERSE if (CPU_32v7M || CPU_32v7) && !CPU_32v6
+	select HAVE_ARCH_HARDENED_USERCOPY
 	select HAVE_ARCH_JUMP_LABEL if !XIP_KERNEL && !CPU_ENDIAN_BE32
 	select HAVE_ARCH_KGDB if !CPU_ENDIAN_BE32
 	select HAVE_ARCH_MMAP_RND_BITS if MMU

arch/arm/include/asm/uaccess.h

@@ -496,7 +496,10 @@ arm_copy_from_user(void *to, const void __user *from, unsigned long n);
 static inline unsigned long __must_check
 __copy_from_user(void *to, const void __user *from, unsigned long n)
 {
-	unsigned int __ua_flags = uaccess_save_and_enable();
+	unsigned int __ua_flags;
+
+	check_object_size(to, n, false);
+	__ua_flags = uaccess_save_and_enable();
 	n = arm_copy_from_user(to, from, n);
 	uaccess_restore(__ua_flags);
 	return n;
@@ -511,11 +514,15 @@ static inline unsigned long __must_check
 __copy_to_user(void __user *to, const void *from, unsigned long n)
 {
 #ifndef CONFIG_UACCESS_WITH_MEMCPY
-	unsigned int __ua_flags = uaccess_save_and_enable();
+	unsigned int __ua_flags;
+
+	check_object_size(from, n, true);
+	__ua_flags = uaccess_save_and_enable();
 	n = arm_copy_to_user(to, from, n);
 	uaccess_restore(__ua_flags);
 	return n;
 #else
+	check_object_size(from, n, true);
 	return arm_copy_to_user(to, from, n);
 #endif
 }

arch/arm/kernel/setup.c

@@ -772,7 +772,7 @@ static void __init request_standard_resources(const struct machine_desc *mdesc)
 	struct resource *res;
 
 	kernel_code.start   = virt_to_phys(_text);
-	kernel_code.end     = virt_to_phys(_etext - 1);
+	kernel_code.end     = virt_to_phys(__init_begin - 1);
 	kernel_data.start   = virt_to_phys(_sdata);
 	kernel_data.end     = virt_to_phys(_end - 1);
 

arch/arm/kernel/vmlinux.lds.S

@@ -120,6 +120,8 @@ SECTIONS
 #ifdef CONFIG_DEBUG_RODATA
 	. = ALIGN(1<<SECTION_SHIFT);
 #endif
+	_etext = .;			/* End of text section */
+
 	RO_DATA(PAGE_SIZE)
 
 	. = ALIGN(4);
@@ -150,8 +152,6 @@ SECTIONS
 
 	NOTES
 
-	_etext = .;			/* End of text and rodata section */
-
 #ifndef CONFIG_XIP_KERNEL
 # ifdef CONFIG_ARM_KERNMEM_PERMS
 	. = ALIGN(1<<SECTION_SHIFT);

arch/arm64/Kconfig

@@ -48,6 +48,7 @@ config ARM64
 	select HAVE_ALIGNED_STRUCT_PAGE if SLUB
 	select HAVE_ARCH_AUDITSYSCALL
 	select HAVE_ARCH_BITREVERSE
+	select HAVE_ARCH_HARDENED_USERCOPY
 	select HAVE_ARCH_JUMP_LABEL
 	select HAVE_ARCH_KASAN if SPARSEMEM_VMEMMAP && !(ARM64_16K_PAGES && ARM64_VA_BITS_48)
 	select HAVE_ARCH_KGDB

arch/arm64/include/asm/uaccess.h

@@ -247,24 +247,39 @@ do {									\
 		-EFAULT;						\
 })
 
-extern unsigned long __must_check __copy_from_user(void *to, const void __user *from, unsigned long n);
+extern unsigned long __must_check __arch_copy_from_user(void *to, const void __user *from, unsigned long n);
-extern unsigned long __must_check __copy_to_user(void __user *to, const void *from, unsigned long n);
+extern unsigned long __must_check __arch_copy_to_user(void __user *to, const void *from, unsigned long n);
 extern unsigned long __must_check __copy_in_user(void __user *to, const void __user *from, unsigned long n);
 extern unsigned long __must_check __clear_user(void __user *addr, unsigned long n);
 
+static inline unsigned long __must_check __copy_from_user(void *to, const void __user *from, unsigned long n)
+{
+	check_object_size(to, n, false);
+	return __arch_copy_from_user(to, from, n);
+}
+
+static inline unsigned long __must_check __copy_to_user(void __user *to, const void *from, unsigned long n)
+{
+	check_object_size(from, n, true);
+	return __arch_copy_to_user(to, from, n);
+}
+
 static inline unsigned long __must_check copy_from_user(void *to, const void __user *from, unsigned long n)
 {
-	if (access_ok(VERIFY_READ, from, n))
+	if (access_ok(VERIFY_READ, from, n)) {
-		n = __copy_from_user(to, from, n);
+		check_object_size(to, n, false);
-	else /* security hole - plug it */
+		n = __arch_copy_from_user(to, from, n);
+	} else /* security hole - plug it */
 		memset(to, 0, n);
 	return n;
 }
 
 static inline unsigned long __must_check copy_to_user(void __user *to, const void *from, unsigned long n)
 {
-	if (access_ok(VERIFY_WRITE, to, n))
+	if (access_ok(VERIFY_WRITE, to, n)) {
-		n = __copy_to_user(to, from, n);
+		check_object_size(from, n, true);
+		n = __arch_copy_to_user(to, from, n);
+	}
 	return n;
 }
 

arch/arm64/kernel/arm64ksyms.c

@@ -33,8 +33,8 @@ EXPORT_SYMBOL(copy_page);
 EXPORT_SYMBOL(clear_page);
 
 	/* user mem (segment) */
-EXPORT_SYMBOL(__copy_from_user);
+EXPORT_SYMBOL(__arch_copy_from_user);
-EXPORT_SYMBOL(__copy_to_user);
+EXPORT_SYMBOL(__arch_copy_to_user);
 EXPORT_SYMBOL(__clear_user);
 EXPORT_SYMBOL(__copy_in_user);
 

arch/arm64/kernel/setup.c

@@ -201,7 +201,7 @@ static void __init request_standard_resources(void)
 	struct resource *res;
 
 	kernel_code.start   = virt_to_phys(_text);
-	kernel_code.end     = virt_to_phys(_etext - 1);
+	kernel_code.end     = virt_to_phys(__init_begin - 1);
 	kernel_data.start   = virt_to_phys(_sdata);
 	kernel_data.end     = virt_to_phys(_end - 1);
 

arch/arm64/kernel/vmlinux.lds.S

@@ -114,11 +114,12 @@ SECTIONS
 	}
 
 	ALIGN_DEBUG_RO
+	_etext = .;			/* End of text section */
+
 	RO_DATA(PAGE_SIZE)
 	EXCEPTION_TABLE(8)
 	NOTES
 	ALIGN_DEBUG_RO
-	_etext = .;			/* End of text and rodata section */
 
 	ALIGN_DEBUG_RO_MIN(PAGE_SIZE)
 	__init_begin = .;

arch/arm64/lib/copy_from_user.S

@@ -66,7 +66,7 @@
 	.endm
 
 end	.req	x5
-ENTRY(__copy_from_user)
+ENTRY(__arch_copy_from_user)
 ALTERNATIVE("nop", __stringify(SET_PSTATE_PAN(0)), ARM64_HAS_PAN, \
 	    CONFIG_ARM64_PAN)
 	add	end, x0, x2
@@ -75,7 +75,7 @@ ALTERNATIVE("nop", __stringify(SET_PSTATE_PAN(1)), ARM64_HAS_PAN, \
 	    CONFIG_ARM64_PAN)
 	mov	x0, #0				// Nothing to copy
 	ret
-ENDPROC(__copy_from_user)
+ENDPROC(__arch_copy_from_user)
 
 	.section .fixup,"ax"
 	.align	2

arch/arm64/lib/copy_to_user.S

@@ -65,7 +65,7 @@
 	.endm
 
 end	.req	x5
-ENTRY(__copy_to_user)
+ENTRY(__arch_copy_to_user)
 ALTERNATIVE("nop", __stringify(SET_PSTATE_PAN(0)), ARM64_HAS_PAN, \
 	    CONFIG_ARM64_PAN)
 	add	end, x0, x2
@@ -74,7 +74,7 @@ ALTERNATIVE("nop", __stringify(SET_PSTATE_PAN(1)), ARM64_HAS_PAN, \
 	    CONFIG_ARM64_PAN)
 	mov	x0, #0
 	ret
-ENDPROC(__copy_to_user)
+ENDPROC(__arch_copy_to_user)
 
 	.section .fixup,"ax"
 	.align	2

arch/arm64/mm/mmu.c

@@ -337,7 +337,6 @@ static void __init __map_memblock(phys_addr_t start, phys_addr_t end)
 				end - kernel_x_end,
 				PAGE_KERNEL);
 	}
-
 }
 #else
 static void __init __map_memblock(phys_addr_t start, phys_addr_t end)
@@ -425,7 +424,7 @@ static void __init fixup_executable(void)
 void mark_rodata_ro(void)
 {
 	create_mapping_late(__pa(_stext), (unsigned long)_stext,
-				(unsigned long)_etext - (unsigned long)_stext,
+				(unsigned long)__init_begin - (unsigned long)_stext,
 				PAGE_KERNEL_ROX);
 
 }

arch/x86/Kconfig

@@ -79,6 +79,7 @@ config X86
 	select HAVE_ALIGNED_STRUCT_PAGE		if SLUB
 	select HAVE_AOUT			if X86_32
 	select HAVE_ARCH_AUDITSYSCALL
+	select HAVE_ARCH_HARDENED_USERCOPY
 	select HAVE_ARCH_HUGE_VMAP		if X86_64 || X86_PAE
 	select HAVE_ARCH_JUMP_LABEL
 	select HAVE_ARCH_KASAN			if X86_64 && SPARSEMEM_VMEMMAP
@@ -88,6 +89,7 @@ config X86
 	select HAVE_ARCH_SOFT_DIRTY		if X86_64
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_ARCH_TRANSPARENT_HUGEPAGE
+	select HAVE_ARCH_WITHIN_STACK_FRAMES
 	select HAVE_BPF_JIT			if X86_64
 	select HAVE_CC_STACKPROTECTOR
 	select HAVE_CMPXCHG_DOUBLE

arch/x86/include/asm/thread_info.h

@@ -177,6 +177,50 @@ static inline unsigned long current_stack_pointer(void)
 	return sp;
 }
 
+/*
+ * Walks up the stack frames to make sure that the specified object is
+ * entirely contained by a single stack frame.
+ *
+ * Returns:
+ *		 1 if within a frame
+ *		-1 if placed across a frame boundary (or outside stack)
+ *		 0 unable to determine (no frame pointers, etc)
+ */
+static inline int arch_within_stack_frames(const void * const stack,
+					   const void * const stackend,
+					   const void *obj, unsigned long len)
+{
+#if defined(CONFIG_FRAME_POINTER)
+	const void *frame = NULL;
+	const void *oldframe;
+
+	oldframe = __builtin_frame_address(1);
+	if (oldframe)
+		frame = __builtin_frame_address(2);
+	/*
+	 * low ----------------------------------------------> high
+	 * [saved bp][saved ip][args][local vars][saved bp][saved ip]
+	 *                     ^----------------^
+	 *               allow copies only within here
+	 */
+	while (stack <= frame && frame < stackend) {
+		/*
+		 * If obj + len extends past the last frame, this
+		 * check won't pass and the next frame will be 0,
+		 * causing us to bail out and correctly report
+		 * the copy as invalid.
+		 */
+		if (obj + len <= frame)
+			return obj >= oldframe + 2 * sizeof(void *) ? 1 : -1;
+		oldframe = frame;
+		frame = *(const void * const *)frame;
+	}
+	return -1;
+#else
+	return 0;
+#endif
+}
+
 #else /* !__ASSEMBLY__ */
 
 #ifdef CONFIG_X86_64

arch/x86/include/asm/uaccess.h

@@ -134,6 +134,9 @@ extern int __get_user_4(void);
 extern int __get_user_8(void);
 extern int __get_user_bad(void);
 
+#define __uaccess_begin() stac()
+#define __uaccess_end()   clac()
+
 /*
  * This is a type: either unsigned long, if the argument fits into
  * that type, or otherwise unsigned long long.
@@ -193,10 +196,10 @@ __typeof__(__builtin_choose_expr(sizeof(x) > sizeof(0UL), 0ULL, 0UL))
 
 #ifdef CONFIG_X86_32
 #define __put_user_asm_u64(x, addr, err, errret)			\
-	asm volatile(ASM_STAC "\n"					\
+	asm volatile("\n"						\
 		     "1:	movl %%eax,0(%2)\n"			\
 		     "2:	movl %%edx,4(%2)\n"			\
-		     "3: " ASM_CLAC "\n"				\
+		     "3:"						\
 		     ".section .fixup,\"ax\"\n"				\
 		     "4:	movl %3,%0\n"				\
 		     "	jmp 3b\n"					\
@@ -207,10 +210,10 @@ __typeof__(__builtin_choose_expr(sizeof(x) > sizeof(0UL), 0ULL, 0UL))
 		     : "A" (x), "r" (addr), "i" (errret), "0" (err))
 
 #define __put_user_asm_ex_u64(x, addr)					\
-	asm volatile(ASM_STAC "\n"					\
+	asm volatile("\n"						\
 		     "1:	movl %%eax,0(%1)\n"			\
 		     "2:	movl %%edx,4(%1)\n"			\
-		     "3: " ASM_CLAC "\n"				\
+		     "3:"						\
 		     _ASM_EXTABLE_EX(1b, 2b)				\
 		     _ASM_EXTABLE_EX(2b, 3b)				\
 		     : : "A" (x), "r" (addr))
@@ -304,6 +307,10 @@ do {									\
 	}								\
 } while (0)
 
+/*
+ * This doesn't do __uaccess_begin/end - the exception handling
+ * around it must do that.
+ */
 #define __put_user_size_ex(x, ptr, size)				\
 do {									\
 	__chk_user_ptr(ptr);						\
@@ -358,9 +365,9 @@ do {									\
 } while (0)
 
 #define __get_user_asm(x, addr, err, itype, rtype, ltype, errret)	\
-	asm volatile(ASM_STAC "\n"					\
+	asm volatile("\n"						\
 		     "1:	mov"itype" %2,%"rtype"1\n"		\
-		     "2: " ASM_CLAC "\n"				\
+		     "2:\n"						\
 		     ".section .fixup,\"ax\"\n"				\
 		     "3:	mov %3,%0\n"				\
 		     "	xor"itype" %"rtype"1,%"rtype"1\n"		\
@@ -370,6 +377,10 @@ do {									\
 		     : "=r" (err), ltype(x)				\
 		     : "m" (__m(addr)), "i" (errret), "0" (err))
 
+/*
+ * This doesn't do __uaccess_begin/end - the exception handling
+ * around it must do that.
+ */
 #define __get_user_size_ex(x, ptr, size)				\
 do {									\
 	__chk_user_ptr(ptr);						\
@@ -400,7 +411,9 @@ do {									\
 #define __put_user_nocheck(x, ptr, size)			\
 ({								\
 	int __pu_err;						\
+	__uaccess_begin();					\
 	__put_user_size((x), (ptr), (size), __pu_err, -EFAULT);	\
+	__uaccess_end();					\
 	__builtin_expect(__pu_err, 0);				\
 })
 
@@ -408,7 +421,9 @@ do {									\
 ({									\
 	int __gu_err;							\
 	unsigned long __gu_val;						\
+	__uaccess_begin();						\
 	__get_user_size(__gu_val, (ptr), (size), __gu_err, -EFAULT);	\
+	__uaccess_end();						\
 	(x) = (__force __typeof__(*(ptr)))__gu_val;			\
 	__builtin_expect(__gu_err, 0);					\
 })
@@ -423,9 +438,9 @@ struct __large_struct { unsigned long buf[100]; };
  * aliasing issues.
  */
 #define __put_user_asm(x, addr, err, itype, rtype, ltype, errret)	\
-	asm volatile(ASM_STAC "\n"					\
+	asm volatile("\n"						\
 		     "1:	mov"itype" %"rtype"1,%2\n"		\
-		     "2: " ASM_CLAC "\n"				\
+		     "2:\n"						\
 		     ".section .fixup,\"ax\"\n"				\
 		     "3:	mov %3,%0\n"				\
 		     "	jmp 2b\n"					\
@@ -445,11 +460,11 @@ struct __large_struct { unsigned long buf[100]; };
  */
 #define uaccess_try	do {						\
 	current_thread_info()->uaccess_err = 0;				\
-	stac();								\
+	__uaccess_begin();						\
 	barrier();
 
 #define uaccess_catch(err)						\
-	clac();								\
+	__uaccess_end();						\
 	(err) |= (current_thread_info()->uaccess_err ? -EFAULT : 0);	\
 } while (0)
 
@@ -547,12 +562,13 @@ extern void __cmpxchg_wrong_size(void)
 	__typeof__(ptr) __uval = (uval);				\
 	__typeof__(*(ptr)) __old = (old);				\
 	__typeof__(*(ptr)) __new = (new);				\
+	__uaccess_begin();						\
 	switch (size) {							\
 	case 1:								\
 	{								\
-		asm volatile("\t" ASM_STAC "\n"				\
+		asm volatile("\n"					\
 			"1:\t" LOCK_PREFIX "cmpxchgb %4, %2\n"		\
-			"2:\t" ASM_CLAC "\n"				\
+			"2:\n"						\
 			"\t.section .fixup, \"ax\"\n"			\
 			"3:\tmov     %3, %0\n"				\
 			"\tjmp     2b\n"				\
@@ -566,9 +582,9 @@ extern void __cmpxchg_wrong_size(void)
 	}								\
 	case 2:								\
 	{								\
-		asm volatile("\t" ASM_STAC "\n"				\
+		asm volatile("\n"					\
 			"1:\t" LOCK_PREFIX "cmpxchgw %4, %2\n"		\
-			"2:\t" ASM_CLAC "\n"				\
+			"2:\n"						\
 			"\t.section .fixup, \"ax\"\n"			\
 			"3:\tmov     %3, %0\n"				\
 			"\tjmp     2b\n"				\
@@ -582,9 +598,9 @@ extern void __cmpxchg_wrong_size(void)
 	}								\
 	case 4:								\
 	{								\
-		asm volatile("\t" ASM_STAC "\n"				\
+		asm volatile("\n"					\
 			"1:\t" LOCK_PREFIX "cmpxchgl %4, %2\n"		\
-			"2:\t" ASM_CLAC "\n"				\
+			"2:\n"						\
 			"\t.section .fixup, \"ax\"\n"			\
 			"3:\tmov     %3, %0\n"				\
 			"\tjmp     2b\n"				\
@@ -601,9 +617,9 @@ extern void __cmpxchg_wrong_size(void)
 		if (!IS_ENABLED(CONFIG_X86_64))				\
 			__cmpxchg_wrong_size();				\
 									\
-		asm volatile("\t" ASM_STAC "\n"				\
+		asm volatile("\n"					\
 			"1:\t" LOCK_PREFIX "cmpxchgq %4, %2\n"		\
-			"2:\t" ASM_CLAC "\n"				\
+			"2:\n"						\
 			"\t.section .fixup, \"ax\"\n"			\
 			"3:\tmov     %3, %0\n"				\
 			"\tjmp     2b\n"				\
@@ -618,6 +634,7 @@ extern void __cmpxchg_wrong_size(void)
 	default:							\
 		__cmpxchg_wrong_size();					\
 	}								\
+	__uaccess_end();						\
 	*__uval = __old;						\
 	__ret;								\
 })
@@ -714,9 +731,10 @@ copy_from_user(void *to, const void __user *from, unsigned long n)
 	 * case, and do only runtime checking for non-constant sizes.
 	 */
 
-	if (likely(sz < 0 || sz >= n))
+	if (likely(sz < 0 || sz >= n)) {
+		check_object_size(to, n, false);
 		n = _copy_from_user(to, from, n);
-	else if(__builtin_constant_p(n))
+	} else if (__builtin_constant_p(n))
 		copy_from_user_overflow();
 	else
 		__copy_from_user_overflow(sz, n);
@@ -732,9 +750,10 @@ copy_to_user(void __user *to, const void *from, unsigned long n)
 	might_fault();
 
 	/* See the comment in copy_from_user() above. */
-	if (likely(sz < 0 || sz >= n))
+	if (likely(sz < 0 || sz >= n)) {
+		check_object_size(from, n, true);
 		n = _copy_to_user(to, from, n);
-	else if(__builtin_constant_p(n))
+	} else if (__builtin_constant_p(n))
 		copy_to_user_overflow();
 	else
 		__copy_to_user_overflow(sz, n);
@@ -745,5 +764,30 @@ copy_to_user(void __user *to, const void *from, unsigned long n)
 #undef __copy_from_user_overflow
 #undef __copy_to_user_overflow
 
+/*
+ * The "unsafe" user accesses aren't really "unsafe", but the naming
+ * is a big fat warning: you have to not only do the access_ok()
+ * checking before using them, but you have to surround them with the
+ * user_access_begin/end() pair.
+ */
+#define user_access_begin()	__uaccess_begin()
+#define user_access_end()	__uaccess_end()
+
+#define unsafe_put_user(x, ptr, err_label)					\
+do {										\
+	int __pu_err;								\
+	__put_user_size((x), (ptr), sizeof(*(ptr)), __pu_err, -EFAULT);		\
+	if (unlikely(__pu_err)) goto err_label;					\
+} while (0)
+
+#define unsafe_get_user(x, ptr, err_label)					\
+do {										\
+	int __gu_err;								\
+	unsigned long __gu_val;							\
+	__get_user_size(__gu_val, (ptr), sizeof(*(ptr)), __gu_err, -EFAULT);	\
+	(x) = (__force __typeof__(*(ptr)))__gu_val;				\
+	if (unlikely(__gu_err)) goto err_label;					\
+} while (0)
+
 #endif /* _ASM_X86_UACCESS_H */
 

arch/x86/include/asm/uaccess_32.h

@@ -43,6 +43,7 @@ unsigned long __must_check __copy_from_user_ll_nocache_nozero
 static __always_inline unsigned long __must_check
 __copy_to_user_inatomic(void __user *to, const void *from, unsigned long n)
 {
+	check_object_size(from, n, true);
 	if (__builtin_constant_p(n)) {
 		unsigned long ret;
 
@@ -143,6 +144,7 @@ static __always_inline unsigned long
 __copy_from_user(void *to, const void __user *from, unsigned long n)
 {
 	might_fault();
+	check_object_size(to, n, false);
 	if (__builtin_constant_p(n)) {
 		unsigned long ret;
 

arch/x86/include/asm/uaccess_64.h

@@ -53,38 +53,53 @@ int __copy_from_user_nocheck(void *dst, const void __user *src, unsigned size)
 {
 	int ret = 0;
 
+	check_object_size(dst, size, false);
 	if (!__builtin_constant_p(size))
 		return copy_user_generic(dst, (__force void *)src, size);
 	switch (size) {
-	case 1:__get_user_asm(*(u8 *)dst, (u8 __user *)src,
+	case 1:
+		__uaccess_begin();
+		__get_user_asm(*(u8 *)dst, (u8 __user *)src,
 			      ret, "b", "b", "=q", 1);
+		__uaccess_end();
 		return ret;
-	case 2:__get_user_asm(*(u16 *)dst, (u16 __user *)src,
+	case 2:
+		__uaccess_begin();
+		__get_user_asm(*(u16 *)dst, (u16 __user *)src,
 			      ret, "w", "w", "=r", 2);
+		__uaccess_end();
 		return ret;
-	case 4:__get_user_asm(*(u32 *)dst, (u32 __user *)src,
+	case 4:
+		__uaccess_begin();
+		__get_user_asm(*(u32 *)dst, (u32 __user *)src,
 			      ret, "l", "k", "=r", 4);
+		__uaccess_end();
 		return ret;
-	case 8:__get_user_asm(*(u64 *)dst, (u64 __user *)src,
+	case 8:
+		__uaccess_begin();
+		__get_user_asm(*(u64 *)dst, (u64 __user *)src,
 			      ret, "q", "", "=r", 8);
+		__uaccess_end();
 		return ret;
 	case 10:
+		__uaccess_begin();
 		__get_user_asm(*(u64 *)dst, (u64 __user *)src,
 			       ret, "q", "", "=r", 10);
-		if (unlikely(ret))
+		if (likely(!ret))
-			return ret;
+			__get_user_asm(*(u16 *)(8 + (char *)dst),
-		__get_user_asm(*(u16 *)(8 + (char *)dst),
+				       (u16 __user *)(8 + (char __user *)src),
-			       (u16 __user *)(8 + (char __user *)src),
+				       ret, "w", "w", "=r", 2);
-			       ret, "w", "w", "=r", 2);
+		__uaccess_end();
 		return ret;
 	case 16:
+		__uaccess_begin();
 		__get_user_asm(*(u64 *)dst, (u64 __user *)src,
 			       ret, "q", "", "=r", 16);
-		if (unlikely(ret))
+		if (likely(!ret))
-			return ret;
+			__get_user_asm(*(u64 *)(8 + (char *)dst),
-		__get_user_asm(*(u64 *)(8 + (char *)dst),
+				       (u64 __user *)(8 + (char __user *)src),
-			       (u64 __user *)(8 + (char __user *)src),
+				       ret, "q", "", "=r", 8);
-			       ret, "q", "", "=r", 8);
+		__uaccess_end();
 		return ret;
 	default:
 		return copy_user_generic(dst, (__force void *)src, size);
@@ -103,38 +118,55 @@ int __copy_to_user_nocheck(void __user *dst, const void *src, unsigned size)
 {
 	int ret = 0;
 
+	check_object_size(src, size, true);
 	if (!__builtin_constant_p(size))
 		return copy_user_generic((__force void *)dst, src, size);
 	switch (size) {
-	case 1:__put_user_asm(*(u8 *)src, (u8 __user *)dst,
+	case 1:
+		__uaccess_begin();
+		__put_user_asm(*(u8 *)src, (u8 __user *)dst,
 			      ret, "b", "b", "iq", 1);
+		__uaccess_end();
 		return ret;
-	case 2:__put_user_asm(*(u16 *)src, (u16 __user *)dst,
+	case 2:
+		__uaccess_begin();
+		__put_user_asm(*(u16 *)src, (u16 __user *)dst,
 			      ret, "w", "w", "ir", 2);
+		__uaccess_end();
 		return ret;
-	case 4:__put_user_asm(*(u32 *)src, (u32 __user *)dst,
+	case 4:
+		__uaccess_begin();
+		__put_user_asm(*(u32 *)src, (u32 __user *)dst,
 			      ret, "l", "k", "ir", 4);
+		__uaccess_end();
 		return ret;
-	case 8:__put_user_asm(*(u64 *)src, (u64 __user *)dst,
+	case 8:
+		__uaccess_begin();
+		__put_user_asm(*(u64 *)src, (u64 __user *)dst,
 			      ret, "q", "", "er", 8);
+		__uaccess_end();
 		return ret;
 	case 10:
+		__uaccess_begin();
 		__put_user_asm(*(u64 *)src, (u64 __user *)dst,
 			       ret, "q", "", "er", 10);
-		if (unlikely(ret))
+		if (likely(!ret)) {
-			return ret;
+			asm("":::"memory");
-		asm("":::"memory");
+			__put_user_asm(4[(u16 *)src], 4 + (u16 __user *)dst,
-		__put_user_asm(4[(u16 *)src], 4 + (u16 __user *)dst,
+				       ret, "w", "w", "ir", 2);
-			       ret, "w", "w", "ir", 2);
+		}
+		__uaccess_end();
 		return ret;
 	case 16:
+		__uaccess_begin();
 		__put_user_asm(*(u64 *)src, (u64 __user *)dst,
 			       ret, "q", "", "er", 16);
-		if (unlikely(ret))
+		if (likely(!ret)) {
-			return ret;
+			asm("":::"memory");
-		asm("":::"memory");
+			__put_user_asm(1[(u64 *)src], 1 + (u64 __user *)dst,
-		__put_user_asm(1[(u64 *)src], 1 + (u64 __user *)dst,
+				       ret, "q", "", "er", 8);
-			       ret, "q", "", "er", 8);
+		}
+		__uaccess_end();
 		return ret;
 	default:
 		return copy_user_generic((__force void *)dst, src, size);
@@ -160,39 +192,47 @@ int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
 	switch (size) {
 	case 1: {
 		u8 tmp;
+		__uaccess_begin();
 		__get_user_asm(tmp, (u8 __user *)src,
 			       ret, "b", "b", "=q", 1);
 		if (likely(!ret))
 			__put_user_asm(tmp, (u8 __user *)dst,
 				       ret, "b", "b", "iq", 1);
+		__uaccess_end();
 		return ret;
 	}
 	case 2: {
 		u16 tmp;
+		__uaccess_begin();
 		__get_user_asm(tmp, (u16 __user *)src,
 			       ret, "w", "w", "=r", 2);
 		if (likely(!ret))
 			__put_user_asm(tmp, (u16 __user *)dst,
 				       ret, "w", "w", "ir", 2);
+		__uaccess_end();
 		return ret;
 	}
 
 	case 4: {
 		u32 tmp;
+		__uaccess_begin();
 		__get_user_asm(tmp, (u32 __user *)src,
 			       ret, "l", "k", "=r", 4);
 		if (likely(!ret))
 			__put_user_asm(tmp, (u32 __user *)dst,
 				       ret, "l", "k", "ir", 4);
+		__uaccess_end();
 		return ret;
 	}
 	case 8: {
 		u64 tmp;
+		__uaccess_begin();
 		__get_user_asm(tmp, (u64 __user *)src,
 			       ret, "q", "", "=r", 8);
 		if (likely(!ret))
 			__put_user_asm(tmp, (u64 __user *)dst,
 				       ret, "q", "", "er", 8);
+		__uaccess_end();
 		return ret;
 	}
 	default:

block/blk-core.c

@@ -40,6 +40,8 @@
 #include "blk.h"
 #include "blk-mq.h"
 
+#include <linux/math64.h>
+
 EXPORT_TRACEPOINT_SYMBOL_GPL(block_bio_remap);
 EXPORT_TRACEPOINT_SYMBOL_GPL(block_rq_remap);
 EXPORT_TRACEPOINT_SYMBOL_GPL(block_bio_complete);
@@ -3539,3 +3541,83 @@ int __init blk_dev_init(void)
 
 	return 0;
 }
+
+/*
+ * Blk IO latency support. We want this to be as cheap as possible, so doing
+ * this lockless (and avoiding atomics), a few off by a few errors in this
+ * code is not harmful, and we don't want to do anything that is
+ * perf-impactful.
+ * TODO : If necessary, we can make the histograms per-cpu and aggregate
+ * them when printing them out.
+ */
+void
+blk_zero_latency_hist(struct io_latency_state *s)
+{
+	memset(s->latency_y_axis_read, 0,
+	       sizeof(s->latency_y_axis_read));
+	memset(s->latency_y_axis_write, 0,
+	       sizeof(s->latency_y_axis_write));
+	s->latency_reads_elems = 0;
+	s->latency_writes_elems = 0;
+}
+
+ssize_t
+blk_latency_hist_show(struct io_latency_state *s, char *buf)
+{
+	int i;
+	int bytes_written = 0;
+	u_int64_t num_elem, elem;
+	int pct;
+
+	num_elem = s->latency_reads_elems;
+	if (num_elem > 0) {
+		bytes_written += scnprintf(buf + bytes_written,
+			   PAGE_SIZE - bytes_written,
+			   "IO svc_time Read Latency Histogram (n = %llu):\n",
+			   num_elem);
+		for (i = 0;
+		     i < ARRAY_SIZE(latency_x_axis_us);
+		     i++) {
+			elem = s->latency_y_axis_read[i];
+			pct = div64_u64(elem * 100, num_elem);
+			bytes_written += scnprintf(buf + bytes_written,
+						   PAGE_SIZE - bytes_written,
+						   "\t< %5lluus%15llu%15d%%\n",
+						   latency_x_axis_us[i],
+						   elem, pct);
+		}
+		/* Last element in y-axis table is overflow */
+		elem = s->latency_y_axis_read[i];
+		pct = div64_u64(elem * 100, num_elem);
+		bytes_written += scnprintf(buf + bytes_written,
+					   PAGE_SIZE - bytes_written,
+					   "\t> %5dms%15llu%15d%%\n", 10,
+					   elem, pct);
+	}
+	num_elem = s->latency_writes_elems;
+	if (num_elem > 0) {
+		bytes_written += scnprintf(buf + bytes_written,
+			   PAGE_SIZE - bytes_written,
+			   "IO svc_time Write Latency Histogram (n = %llu):\n",
+			   num_elem);
+		for (i = 0;
+		     i < ARRAY_SIZE(latency_x_axis_us);
+		     i++) {
+			elem = s->latency_y_axis_write[i];
+			pct = div64_u64(elem * 100, num_elem);
+			bytes_written += scnprintf(buf + bytes_written,
+						   PAGE_SIZE - bytes_written,
+						   "\t< %5lluus%15llu%15d%%\n",
+						   latency_x_axis_us[i],
+						   elem, pct);
+		}
+		/* Last element in y-axis table is overflow */
+		elem = s->latency_y_axis_write[i];
+		pct = div64_u64(elem * 100, num_elem);
+		bytes_written += scnprintf(buf + bytes_written,
+					   PAGE_SIZE - bytes_written,
+					   "\t> %5dms%15llu%15d%%\n", 10,
+					   elem, pct);
+	}
+	return bytes_written;
+}

drivers/media/tuners/tuner-xc2028.c

@@ -1403,11 +1403,14 @@ static int xc2028_set_config(struct dvb_frontend *fe, void *priv_cfg)
 	 * in order to avoid troubles during device release.
 	 */
 	kfree(priv->ctrl.fname);
+	priv->ctrl.fname = NULL;
 	memcpy(&priv->ctrl, p, sizeof(priv->ctrl));
 	if (p->fname) {
 		priv->ctrl.fname = kstrdup(p->fname, GFP_KERNEL);
-		if (priv->ctrl.fname == NULL)
+		if (priv->ctrl.fname == NULL) {
 			rc = -ENOMEM;
+			goto unlock;
+		}
 	}
 
 	/*
@@ -1439,6 +1442,7 @@ static int xc2028_set_config(struct dvb_frontend *fe, void *priv_cfg)
 		} else
 			priv->state = XC2028_WAITING_FIRMWARE;
 	}
+unlock:
 	mutex_unlock(&priv->lock);
 
 	return rc;

drivers/mmc/core/core.c

@@ -183,6 +183,17 @@ void mmc_request_done(struct mmc_host *host, struct mmc_request *mrq)
 			pr_debug("%s:     %d bytes transferred: %d\n",
 				mmc_hostname(host),
 				mrq->data->bytes_xfered, mrq->data->error);
+			if (mrq->lat_hist_enabled) {
+				ktime_t completion;
+				u_int64_t delta_us;
+
+				completion = ktime_get();
+				delta_us = ktime_us_delta(completion,
+							  mrq->io_start);
+				blk_update_latency_hist(&host->io_lat_s,
+					(mrq->data->flags & MMC_DATA_READ),
+					delta_us);
+			}
 			trace_mmc_blk_rw_end(cmd->opcode, cmd->arg, mrq->data);
 		}
 
@@ -627,6 +638,11 @@ struct mmc_async_req *mmc_start_req(struct mmc_host *host,
 	}
 
 	if (!err && areq) {
+		if (host->latency_hist_enabled) {
+			areq->mrq->io_start = ktime_get();
+			areq->mrq->lat_hist_enabled = 1;
+		} else
+			areq->mrq->lat_hist_enabled = 0;
 		trace_mmc_blk_rw_start(areq->mrq->cmd->opcode,
 				       areq->mrq->cmd->arg,
 				       areq->mrq->data);
@@ -1964,7 +1980,7 @@ void mmc_init_erase(struct mmc_card *card)
 }
 
 static unsigned int mmc_mmc_erase_timeout(struct mmc_card *card,
-				          unsigned int arg, unsigned int qty)
+					  unsigned int arg, unsigned int qty)
 {
 	unsigned int erase_timeout;
 
@@ -2907,6 +2923,54 @@ static void __exit mmc_exit(void)
 	destroy_workqueue(workqueue);
 }
 
+static ssize_t
+latency_hist_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct mmc_host *host = cls_dev_to_mmc_host(dev);
+
+	return blk_latency_hist_show(&host->io_lat_s, buf);
+}
+
+/*
+ * Values permitted 0, 1, 2.
+ * 0 -> Disable IO latency histograms (default)
+ * 1 -> Enable IO latency histograms
+ * 2 -> Zero out IO latency histograms
+ */
+static ssize_t
+latency_hist_store(struct device *dev, struct device_attribute *attr,
+		   const char *buf, size_t count)
+{
+	struct mmc_host *host = cls_dev_to_mmc_host(dev);
+	long value;
+
+	if (kstrtol(buf, 0, &value))
+		return -EINVAL;
+	if (value == BLK_IO_LAT_HIST_ZERO)
+		blk_zero_latency_hist(&host->io_lat_s);
+	else if (value == BLK_IO_LAT_HIST_ENABLE ||
+		 value == BLK_IO_LAT_HIST_DISABLE)
+		host->latency_hist_enabled = value;
+	return count;
+}
+
+static DEVICE_ATTR(latency_hist, S_IRUGO | S_IWUSR,
+		   latency_hist_show, latency_hist_store);
+
+void
+mmc_latency_hist_sysfs_init(struct mmc_host *host)
+{
+	if (device_create_file(&host->class_dev, &dev_attr_latency_hist))
+		dev_err(&host->class_dev,
+			"Failed to create latency_hist sysfs entry\n");
+}
+
+void
+mmc_latency_hist_sysfs_exit(struct mmc_host *host)
+{
+	device_remove_file(&host->class_dev, &dev_attr_latency_hist);
+}
+
 subsys_initcall(mmc_init);
 module_exit(mmc_exit);
 

drivers/mmc/core/host.c

@@ -32,8 +32,6 @@
 #include "slot-gpio.h"
 #include "pwrseq.h"
 
-#define cls_dev_to_mmc_host(d)	container_of(d, struct mmc_host, class_dev)
-
 static DEFINE_IDR(mmc_host_idr);
 static DEFINE_SPINLOCK(mmc_host_lock);
 
@@ -394,6 +392,8 @@ int mmc_add_host(struct mmc_host *host)
 	mmc_add_host_debugfs(host);
 #endif
 
+	mmc_latency_hist_sysfs_init(host);
+
 	mmc_start_host(host);
 	if (!(host->pm_flags & MMC_PM_IGNORE_PM_NOTIFY))
 		register_pm_notifier(&host->pm_notify);
@@ -422,6 +422,8 @@ void mmc_remove_host(struct mmc_host *host)
 	mmc_remove_host_debugfs(host);
 #endif
 
+	mmc_latency_hist_sysfs_exit(host);
+
 	device_del(&host->class_dev);
 
 	led_trigger_unregister_simple(host->led);

drivers/mmc/core/host.h

@@ -12,6 +12,8 @@
 #define _MMC_CORE_HOST_H
 #include <linux/mmc/host.h>
 
+#define cls_dev_to_mmc_host(d)	container_of(d, struct mmc_host, class_dev)
+
 int mmc_register_host_class(void);
 void mmc_unregister_host_class(void);
 
@@ -21,5 +23,8 @@ void mmc_retune_hold(struct mmc_host *host);
 void mmc_retune_release(struct mmc_host *host);
 int mmc_retune(struct mmc_host *host);
 
+void mmc_latency_hist_sysfs_init(struct mmc_host *host);
+void mmc_latency_hist_sysfs_exit(struct mmc_host *host);
+
 #endif
 

drivers/scsi/ufs/ufshcd.c

@@ -39,6 +39,7 @@
 
 #include <linux/async.h>
 #include <linux/devfreq.h>
+#include <linux/blkdev.h>
 
 #include "ufshcd.h"
 #include "unipro.h"
@@ -1332,6 +1333,17 @@ static int ufshcd_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *cmd)
 		clear_bit_unlock(tag, &hba->lrb_in_use);
 		goto out;
 	}
+
+	/* IO svc time latency histogram */
+	if (hba != NULL && cmd->request != NULL) {
+		if (hba->latency_hist_enabled &&
+		    (cmd->request->cmd_type == REQ_TYPE_FS)) {
+			cmd->request->lat_hist_io_start = ktime_get();
+			cmd->request->lat_hist_enabled = 1;
+		} else
+			cmd->request->lat_hist_enabled = 0;
+	}
+
 	WARN_ON(hba->clk_gating.state != CLKS_ON);
 
 	lrbp = &hba->lrb[tag];
@@ -3160,6 +3172,7 @@ static void ufshcd_transfer_req_compl(struct ufs_hba *hba)
 	u32 tr_doorbell;
 	int result;
 	int index;
+	struct request *req;
 
 	/* Resetting interrupt aggregation counters first and reading the
 	 * DOOR_BELL afterward allows us to handle all the completed requests.
@@ -3184,6 +3197,22 @@ static void ufshcd_transfer_req_compl(struct ufs_hba *hba)
 			/* Mark completed command as NULL in LRB */
 			lrbp->cmd = NULL;
 			clear_bit_unlock(index, &hba->lrb_in_use);
+			req = cmd->request;
+			if (req) {
+				/* Update IO svc time latency histogram */
+				if (req->lat_hist_enabled) {
+					ktime_t completion;
+					u_int64_t delta_us;
+
+					completion = ktime_get();
+					delta_us = ktime_us_delta(completion,
+						  req->lat_hist_io_start);
+					/* rq_data_dir() => true if WRITE */
+					blk_update_latency_hist(&hba->io_lat_s,
+						(rq_data_dir(req) == READ),
+						delta_us);
+				}
+			}
 			/* Do not touch lrbp after scsi done */
 			cmd->scsi_done(cmd);
 			__ufshcd_release(hba);
@@ -5327,6 +5356,54 @@ out:
 }
 EXPORT_SYMBOL(ufshcd_shutdown);
 
+/*
+ * Values permitted 0, 1, 2.
+ * 0 -> Disable IO latency histograms (default)
+ * 1 -> Enable IO latency histograms
+ * 2 -> Zero out IO latency histograms
+ */
+static ssize_t
+latency_hist_store(struct device *dev, struct device_attribute *attr,
+		   const char *buf, size_t count)
+{
+	struct ufs_hba *hba = dev_get_drvdata(dev);
+	long value;
+
+	if (kstrtol(buf, 0, &value))
+		return -EINVAL;
+	if (value == BLK_IO_LAT_HIST_ZERO)
+		blk_zero_latency_hist(&hba->io_lat_s);
+	else if (value == BLK_IO_LAT_HIST_ENABLE ||
+		 value == BLK_IO_LAT_HIST_DISABLE)
+		hba->latency_hist_enabled = value;
+	return count;
+}
+
+ssize_t
+latency_hist_show(struct device *dev, struct device_attribute *attr,
+		  char *buf)
+{
+	struct ufs_hba *hba = dev_get_drvdata(dev);
+
+	return blk_latency_hist_show(&hba->io_lat_s, buf);
+}
+
+static DEVICE_ATTR(latency_hist, S_IRUGO | S_IWUSR,
+		   latency_hist_show, latency_hist_store);
+
+static void
+ufshcd_init_latency_hist(struct ufs_hba *hba)
+{
+	if (device_create_file(hba->dev, &dev_attr_latency_hist))
+		dev_err(hba->dev, "Failed to create latency_hist sysfs entry\n");
+}
+
+static void
+ufshcd_exit_latency_hist(struct ufs_hba *hba)
+{
+	device_create_file(hba->dev, &dev_attr_latency_hist);
+}
+
 /**
  * ufshcd_remove - de-allocate SCSI host and host memory space
  *		data structure memory
@@ -5342,6 +5419,7 @@ void ufshcd_remove(struct ufs_hba *hba)
 	scsi_host_put(hba->host);
 
 	ufshcd_exit_clk_gating(hba);
+	ufshcd_exit_latency_hist(hba);
 	if (ufshcd_is_clkscaling_enabled(hba))
 		devfreq_remove_device(hba->devfreq);
 	ufshcd_hba_exit(hba);
@@ -5639,6 +5717,8 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq)
 	/* Hold auto suspend until async scan completes */
 	pm_runtime_get_sync(dev);
 
+	ufshcd_init_latency_hist(hba);
+
 	/*
 	 * The device-initialize-sequence hasn't been invoked yet.
 	 * Set the device to power-off state
@@ -5653,6 +5733,7 @@ out_remove_scsi_host:
 	scsi_remove_host(hba->host);
 exit_gating:
 	ufshcd_exit_clk_gating(hba);
+	ufshcd_exit_latency_hist(hba);
 out_disable:
 	hba->is_irq_enabled = false;
 	scsi_host_put(host);

drivers/scsi/ufs/ufshcd.h

@@ -532,6 +532,9 @@ struct ufs_hba {
 	struct devfreq *devfreq;
 	struct ufs_clk_scaling clk_scaling;
 	bool is_sys_suspended;
+
+	int			latency_hist_enabled;
+	struct io_latency_state io_lat_s;
 };
 
 /* Returns true if clocks can be gated. Otherwise false */

drivers/tty/tty_ldisc.c

@@ -417,6 +417,10 @@ EXPORT_SYMBOL_GPL(tty_ldisc_flush);
  *	they are not on hot paths so a little discipline won't do
  *	any harm.
  *
+ *	The line discipline-related tty_struct fields are reset to
+ *	prevent the ldisc driver from re-using stale information for
+ *	the new ldisc instance.
+ *
  *	Locking: takes termios_rwsem
  */
 
@@ -425,6 +429,9 @@ static void tty_set_termios_ldisc(struct tty_struct *tty, int num)
 	down_write(&tty->termios_rwsem);
 	tty->termios.c_line = num;
 	up_write(&tty->termios_rwsem);
+
+	tty->disc_data = NULL;
+	tty->receive_room = 0;
 }
 
 /**

include/linux/blkdev.h

@@ -197,6 +197,9 @@ struct request {
 
 	/* for bidi */
 	struct request *next_rq;
+
+	ktime_t			lat_hist_io_start;
+	int			lat_hist_enabled;
 };
 
 static inline unsigned short req_get_ioprio(struct request *req)
@@ -1656,6 +1659,79 @@ extern int bdev_write_page(struct block_device *, sector_t, struct page *,
 						struct writeback_control *);
 extern long bdev_direct_access(struct block_device *, sector_t,
 		void __pmem **addr, unsigned long *pfn, long size);
+
+/*
+ * X-axis for IO latency histogram support.
+ */
+static const u_int64_t latency_x_axis_us[] = {
+	100,
+	200,
+	300,
+	400,
+	500,
+	600,
+	700,
+	800,
+	900,
+	1000,
+	1200,
+	1400,
+	1600,
+	1800,
+	2000,
+	2500,
+	3000,
+	4000,
+	5000,
+	6000,
+	7000,
+	9000,
+	10000
+};
+
+#define BLK_IO_LAT_HIST_DISABLE         0
+#define BLK_IO_LAT_HIST_ENABLE          1
+#define BLK_IO_LAT_HIST_ZERO            2
+
+struct io_latency_state {
+	u_int64_t	latency_y_axis_read[ARRAY_SIZE(latency_x_axis_us) + 1];
+	u_int64_t	latency_reads_elems;
+	u_int64_t	latency_y_axis_write[ARRAY_SIZE(latency_x_axis_us) + 1];
+	u_int64_t	latency_writes_elems;
+};
+
+static inline void
+blk_update_latency_hist(struct io_latency_state *s,
+			int read,
+			u_int64_t delta_us)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(latency_x_axis_us); i++) {
+		if (delta_us < (u_int64_t)latency_x_axis_us[i]) {
+			if (read)
+				s->latency_y_axis_read[i]++;
+			else
+				s->latency_y_axis_write[i]++;
+			break;
+		}
+	}
+	if (i == ARRAY_SIZE(latency_x_axis_us)) {
+		/* Overflowed the histogram */
+		if (read)
+			s->latency_y_axis_read[i]++;
+		else
+			s->latency_y_axis_write[i]++;
+	}
+	if (read)
+		s->latency_reads_elems++;
+	else
+		s->latency_writes_elems++;
+}
+
+void blk_zero_latency_hist(struct io_latency_state *s);
+ssize_t blk_latency_hist_show(struct io_latency_state *s, char *buf);
+
 #else /* CONFIG_BLOCK */
 
 struct block_device;

include/linux/mmc/core.h

@@ -136,6 +136,8 @@ struct mmc_request {
 	struct completion	completion;
 	void			(*done)(struct mmc_request *);/* completion function */
 	struct mmc_host		*host;
+	ktime_t			io_start;
+	int			lat_hist_enabled;
 };
 
 struct mmc_card;

include/linux/mmc/host.h

@@ -16,6 +16,7 @@
 #include <linux/sched.h>
 #include <linux/device.h>
 #include <linux/fault-inject.h>
+#include <linux/blkdev.h>
 
 #include <linux/mmc/core.h>
 #include <linux/mmc/card.h>
@@ -379,6 +380,9 @@ struct mmc_host {
 	} embedded_sdio_data;
 #endif
 
+	int			latency_hist_enabled;
+	struct io_latency_state io_lat_s;
+
 	unsigned long		private[0] ____cacheline_aligned;
 };
 

include/linux/mmzone.h

@@ -65,8 +65,10 @@ enum {
 
 #ifdef CONFIG_CMA
 #  define is_migrate_cma(migratetype) unlikely((migratetype) == MIGRATE_CMA)
+#  define is_migrate_cma_page(_page) (get_pageblock_migratetype(_page) == MIGRATE_CMA)
 #else
 #  define is_migrate_cma(migratetype) false
+#  define is_migrate_cma_page(_page) false
 #endif
 
 #define for_each_migratetype_order(order, type) \

include/linux/slab.h

@@ -144,6 +144,18 @@ void kfree(const void *);
 void kzfree(const void *);
 size_t ksize(const void *);
 
+#ifdef CONFIG_HAVE_HARDENED_USERCOPY_ALLOCATOR
+const char *__check_heap_object(const void *ptr, unsigned long n,
+				struct page *page);
+#else
+static inline const char *__check_heap_object(const void *ptr,
+					      unsigned long n,
+					      struct page *page)
+{
+	return NULL;
+}
+#endif
+
 /*
  * Some archs want to perform DMA into kmalloc caches and need a guaranteed
  * alignment larger than the alignment of a 64-bit integer.

include/linux/slub_def.h

@@ -81,6 +81,7 @@ struct kmem_cache {
 	int reserved;		/* Reserved bytes at the end of slabs */
 	const char *name;	/* Name (only for display!) */
 	struct list_head list;	/* List of slab caches */
+	int red_left_pad;	/* Left redzone padding size */
 #ifdef CONFIG_SYSFS
 	struct kobject kobj;	/* For sysfs */
 #endif

include/linux/thread_info.h

@@ -145,6 +145,30 @@ static inline bool test_and_clear_restore_sigmask(void)
 #error "no set_restore_sigmask() provided and default one won't work"
 #endif
 
+#ifndef CONFIG_HAVE_ARCH_WITHIN_STACK_FRAMES
+static inline int arch_within_stack_frames(const void * const stack,
+					   const void * const stackend,
+					   const void *obj, unsigned long len)
+{
+	return 0;
+}
+#endif
+
+#ifdef CONFIG_HARDENED_USERCOPY
+extern void __check_object_size(const void *ptr, unsigned long n,
+					bool to_user);
+
+static inline void check_object_size(const void *ptr, unsigned long n,
+				     bool to_user)
+{
+	__check_object_size(ptr, n, to_user);
+}
+#else
+static inline void check_object_size(const void *ptr, unsigned long n,
+				     bool to_user)
+{ }
+#endif /* CONFIG_HARDENED_USERCOPY */
+
 #endif	/* __KERNEL__ */
 
 #endif /* _LINUX_THREAD_INFO_H */

include/linux/uaccess.h

@@ -111,4 +111,11 @@ extern long strncpy_from_unsafe(char *dst, const void *unsafe_addr, long count);
 #define probe_kernel_address(addr, retval)		\
 	probe_kernel_read(&retval, addr, sizeof(retval))
 
+#ifndef user_access_begin
+#define user_access_begin() do { } while (0)
+#define user_access_end() do { } while (0)
+#define unsafe_get_user(x, ptr, err) do { if (unlikely(__get_user(x, ptr))) goto err; } while (0)
+#define unsafe_put_user(x, ptr, err) do { if (unlikely(__put_user(x, ptr))) goto err; } while (0)
+#endif
+
 #endif		/* __LINUX_UACCESS_H__ */

include/net/tcp.h

@@ -1513,6 +1513,8 @@ static inline void tcp_check_send_head(struct sock *sk, struct sk_buff *skb_unli
 {
 	if (sk->sk_send_head == skb_unlinked)
 		sk->sk_send_head = NULL;
+	if (tcp_sk(sk)->highest_sack == skb_unlinked)
+		tcp_sk(sk)->highest_sack = NULL;
 }
 
 static inline void tcp_init_send_head(struct sock *sk)

init/Kconfig

@@ -1719,6 +1719,7 @@ choice
 
 config SLAB
 	bool "SLAB"
+	select HAVE_HARDENED_USERCOPY_ALLOCATOR
 	help
 	  The regular slab allocator that is established and known to work
 	  well in all environments. It organizes cache hot objects in
@@ -1726,6 +1727,7 @@ config SLAB
 
 config SLUB
 	bool "SLUB (Unqueued Allocator)"
+	select HAVE_HARDENED_USERCOPY_ALLOCATOR
 	help
 	   SLUB is a slab allocator that minimizes cache line usage
 	   instead of managing queues of cached objects (SLAB approach).

kernel/cpuset.c

@@ -98,6 +98,7 @@ struct cpuset {
 
 	/* user-configured CPUs and Memory Nodes allow to tasks */
 	cpumask_var_t cpus_allowed;
+	cpumask_var_t cpus_requested;
 	nodemask_t mems_allowed;
 
 	/* effective CPUs and Memory Nodes allow to tasks */
@@ -386,7 +387,7 @@ static void cpuset_update_task_spread_flag(struct cpuset *cs,
 
 static int is_cpuset_subset(const struct cpuset *p, const struct cpuset *q)
 {
-	return	cpumask_subset(p->cpus_allowed, q->cpus_allowed) &&
+	return	cpumask_subset(p->cpus_requested, q->cpus_requested) &&
 		nodes_subset(p->mems_allowed, q->mems_allowed) &&
 		is_cpu_exclusive(p) <= is_cpu_exclusive(q) &&
 		is_mem_exclusive(p) <= is_mem_exclusive(q);
@@ -486,7 +487,7 @@ static int validate_change(struct cpuset *cur, struct cpuset *trial)
 	cpuset_for_each_child(c, css, par) {
 		if ((is_cpu_exclusive(trial) || is_cpu_exclusive(c)) &&
 		    c != cur &&
-		    cpumask_intersects(trial->cpus_allowed, c->cpus_allowed))
+		    cpumask_intersects(trial->cpus_requested, c->cpus_requested))
 			goto out;
 		if ((is_mem_exclusive(trial) || is_mem_exclusive(c)) &&
 		    c != cur &&
@@ -945,17 +946,18 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs,
 	if (!*buf) {
 		cpumask_clear(trialcs->cpus_allowed);
 	} else {
-		retval = cpulist_parse(buf, trialcs->cpus_allowed);
+		retval = cpulist_parse(buf, trialcs->cpus_requested);
 		if (retval < 0)
 			return retval;
 
-		if (!cpumask_subset(trialcs->cpus_allowed,
+		if (!cpumask_subset(trialcs->cpus_requested, cpu_present_mask))
-				    top_cpuset.cpus_allowed))
 			return -EINVAL;
+
+		cpumask_and(trialcs->cpus_allowed, trialcs->cpus_requested, cpu_active_mask);
 	}
 
 	/* Nothing to do if the cpus didn't change */
-	if (cpumask_equal(cs->cpus_allowed, trialcs->cpus_allowed))
+	if (cpumask_equal(cs->cpus_requested, trialcs->cpus_requested))
 		return 0;
 
 	retval = validate_change(cs, trialcs);
@@ -964,6 +966,7 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs,
 
 	spin_lock_irq(&callback_lock);
 	cpumask_copy(cs->cpus_allowed, trialcs->cpus_allowed);
+	cpumask_copy(cs->cpus_requested, trialcs->cpus_requested);
 	spin_unlock_irq(&callback_lock);
 
 	/* use trialcs->cpus_allowed as a temp variable */
@@ -1754,7 +1757,7 @@ static int cpuset_common_seq_show(struct seq_file *sf, void *v)
 
 	switch (type) {
 	case FILE_CPULIST:
-		seq_printf(sf, "%*pbl\n", cpumask_pr_args(cs->cpus_allowed));
+		seq_printf(sf, "%*pbl\n", cpumask_pr_args(cs->cpus_requested));
 		break;
 	case FILE_MEMLIST:
 		seq_printf(sf, "%*pbl\n", nodemask_pr_args(&cs->mems_allowed));
@@ -1943,11 +1946,14 @@ cpuset_css_alloc(struct cgroup_subsys_state *parent_css)
 		return ERR_PTR(-ENOMEM);
 	if (!alloc_cpumask_var(&cs->cpus_allowed, GFP_KERNEL))
 		goto free_cs;
+	if (!alloc_cpumask_var(&cs->cpus_requested, GFP_KERNEL))
+		goto free_allowed;
 	if (!alloc_cpumask_var(&cs->effective_cpus, GFP_KERNEL))
-		goto free_cpus;
+		goto free_requested;
 
 	set_bit(CS_SCHED_LOAD_BALANCE, &cs->flags);
 	cpumask_clear(cs->cpus_allowed);
+	cpumask_clear(cs->cpus_requested);
 	nodes_clear(cs->mems_allowed);
 	cpumask_clear(cs->effective_cpus);
 	nodes_clear(cs->effective_mems);
@@ -1956,7 +1962,9 @@ cpuset_css_alloc(struct cgroup_subsys_state *parent_css)
 
 	return &cs->css;
 
-free_cpus:
+free_requested:
+	free_cpumask_var(cs->cpus_requested);
+free_allowed:
 	free_cpumask_var(cs->cpus_allowed);
 free_cs:
 	kfree(cs);
@@ -2019,6 +2027,7 @@ static int cpuset_css_online(struct cgroup_subsys_state *css)
 	cs->mems_allowed = parent->mems_allowed;
 	cs->effective_mems = parent->mems_allowed;
 	cpumask_copy(cs->cpus_allowed, parent->cpus_allowed);
+	cpumask_copy(cs->cpus_requested, parent->cpus_requested);
 	cpumask_copy(cs->effective_cpus, parent->cpus_allowed);
 	spin_unlock_irq(&callback_lock);
 out_unlock:
@@ -2053,6 +2062,7 @@ static void cpuset_css_free(struct cgroup_subsys_state *css)
 
 	free_cpumask_var(cs->effective_cpus);
 	free_cpumask_var(cs->cpus_allowed);
+	free_cpumask_var(cs->cpus_requested);
 	kfree(cs);
 }
 
@@ -2120,8 +2130,11 @@ int __init cpuset_init(void)
 		BUG();
 	if (!alloc_cpumask_var(&top_cpuset.effective_cpus, GFP_KERNEL))
 		BUG();
+	if (!alloc_cpumask_var(&top_cpuset.cpus_requested, GFP_KERNEL))
+		BUG();
 
 	cpumask_setall(top_cpuset.cpus_allowed);
+	cpumask_setall(top_cpuset.cpus_requested);
 	nodes_setall(top_cpuset.mems_allowed);
 	cpumask_setall(top_cpuset.effective_cpus);
 	nodes_setall(top_cpuset.effective_mems);
@@ -2255,7 +2268,7 @@ retry:
 		goto retry;
 	}
 
-	cpumask_and(&new_cpus, cs->cpus_allowed, parent_cs(cs)->effective_cpus);
+	cpumask_and(&new_cpus, cs->cpus_requested, parent_cs(cs)->effective_cpus);
 	nodes_and(new_mems, cs->mems_allowed, parent_cs(cs)->effective_mems);
 
 	cpus_updated = !cpumask_equal(&new_cpus, cs->effective_cpus);

kernel/rcu/sync.c

@@ -68,6 +68,7 @@ void rcu_sync_lockdep_assert(struct rcu_sync *rsp)
 	RCU_LOCKDEP_WARN(!gp_ops[rsp->gp_type].held(),
 			 "suspicious rcu_sync_is_idle() usage");
 }
+EXPORT_SYMBOL_GPL(rcu_sync_lockdep_assert);
 #endif
 
 /**

lib/strncpy_from_user.c

@@ -39,8 +39,8 @@ static inline long do_strncpy_from_user(char *dst, const char __user *src, long
 		unsigned long c, data;
 
 		/* Fall back to byte-at-a-time if we get a page fault */
-		if (unlikely(__get_user(c,(unsigned long __user *)(src+res))))
+		unsafe_get_user(c, (unsigned long __user *)(src+res), byte_at_a_time);
-			break;
+
 		*(unsigned long *)(dst+res) = c;
 		if (has_zero(c, &data, &constants)) {
 			data = prep_zero_mask(c, data, &constants);
@@ -55,8 +55,7 @@ byte_at_a_time:
 	while (max) {
 		char c;
 
-		if (unlikely(__get_user(c,src+res)))
+		unsafe_get_user(c,src+res, efault);
-			return -EFAULT;
 		dst[res] = c;
 		if (!c)
 			return res;
@@ -75,6 +74,7 @@ byte_at_a_time:
 	 * Nope: we hit the address space limit, and we still had more
 	 * characters the caller would have wanted. That's an EFAULT.
 	 */
+efault:
 	return -EFAULT;
 }
 
@@ -107,7 +107,12 @@ long strncpy_from_user(char *dst, const char __user *src, long count)
 	src_addr = (unsigned long)src;
 	if (likely(src_addr < max_addr)) {
 		unsigned long max = max_addr - src_addr;
-		return do_strncpy_from_user(dst, src, count, max);
+		long retval;
+
+		user_access_begin();
+		retval = do_strncpy_from_user(dst, src, count, max);
+		user_access_end();
+		return retval;
 	}
 	return -EFAULT;
 }

lib/strnlen_user.c

@@ -45,8 +45,7 @@ static inline long do_strnlen_user(const char __user *src, unsigned long count,
 	src -= align;
 	max += align;
 
-	if (unlikely(__get_user(c,(unsigned long __user *)src)))
+	unsafe_get_user(c, (unsigned long __user *)src, efault);
-		return 0;
 	c |= aligned_byte_mask(align);
 
 	for (;;) {
@@ -61,8 +60,7 @@ static inline long do_strnlen_user(const char __user *src, unsigned long count,
 		if (unlikely(max <= sizeof(unsigned long)))
 			break;
 		max -= sizeof(unsigned long);
-		if (unlikely(__get_user(c,(unsigned long __user *)(src+res))))
+		unsafe_get_user(c, (unsigned long __user *)(src+res), efault);
-			return 0;
 	}
 	res -= align;
 
@@ -77,6 +75,7 @@ static inline long do_strnlen_user(const char __user *src, unsigned long count,
 	 * Nope: we hit the address space limit, and we still had more
 	 * characters the caller would have wanted. That's 0.
 	 */
+efault:
 	return 0;
 }
 
@@ -112,7 +111,12 @@ long strnlen_user(const char __user *str, long count)
 	src_addr = (unsigned long)str;
 	if (likely(src_addr < max_addr)) {
 		unsigned long max = max_addr - src_addr;
-		return do_strnlen_user(str, count, max);
+		long retval;
+
+		user_access_begin();
+		retval = do_strnlen_user(str, count, max);
+		user_access_end();
+		return retval;
 	}
 	return 0;
 }
@@ -141,7 +145,12 @@ long strlen_user(const char __user *str)
 	src_addr = (unsigned long)str;
 	if (likely(src_addr < max_addr)) {
 		unsigned long max = max_addr - src_addr;
-		return do_strnlen_user(str, ~0ul, max);
+		long retval;
+
+		user_access_begin();
+		retval = do_strnlen_user(str, ~0ul, max);
+		user_access_end();
+		return retval;
 	}
 	return 0;
 }

mm/Makefile

@@ -5,6 +5,9 @@
 KASAN_SANITIZE_slab_common.o := n
 KASAN_SANITIZE_slub.o := n
 
+# Since __builtin_frame_address does work as used, disable the warning.
+CFLAGS_usercopy.o += $(call cc-disable-warning, frame-address)
+
 mmu-y			:= nommu.o
 mmu-$(CONFIG_MMU)	:= gup.o highmem.o memory.o mincore.o \
 			   mlock.o mmap.o mprotect.o mremap.o msync.o rmap.o \
@@ -81,3 +84,4 @@ obj-$(CONFIG_CMA_DEBUGFS) += cma_debug.o
 obj-$(CONFIG_USERFAULTFD) += userfaultfd.o
 obj-$(CONFIG_IDLE_PAGE_TRACKING) += page_idle.o
 obj-$(CONFIG_FRAME_VECTOR) += frame_vector.o
+obj-$(CONFIG_HARDENED_USERCOPY) += usercopy.o

mm/slab.c

@@ -4228,6 +4228,36 @@ static int __init slab_proc_init(void)
 module_init(slab_proc_init);
 #endif
 
+#ifdef CONFIG_HARDENED_USERCOPY
+/*
+ * Rejects objects that are incorrectly sized.
+ *
+ * Returns NULL if check passes, otherwise const char * to name of cache
+ * to indicate an error.
+ */
+const char *__check_heap_object(const void *ptr, unsigned long n,
+				struct page *page)
+{
+	struct kmem_cache *cachep;
+	unsigned int objnr;
+	unsigned long offset;
+
+	/* Find and validate object. */
+	cachep = page->slab_cache;
+	objnr = obj_to_index(cachep, page, (void *)ptr);
+	BUG_ON(objnr >= cachep->num);
+
+	/* Find offset within object. */
+	offset = ptr - index_to_obj(cachep, page, objnr) - obj_offset(cachep);
+
+	/* Allow address range falling entirely within object size. */
+	if (offset <= cachep->object_size && n <= cachep->object_size - offset)
+		return NULL;
+
+	return cachep->name;
+}
+#endif /* CONFIG_HARDENED_USERCOPY */
+
 /**
  * ksize - get the actual amount of memory allocated for a given object
  * @objp: Pointer to the object

mm/slub.c

@@ -124,6 +124,14 @@ static inline int kmem_cache_debug(struct kmem_cache *s)
 #endif
 }
 
+static inline void *fixup_red_left(struct kmem_cache *s, void *p)
+{
+	if (kmem_cache_debug(s) && s->flags & SLAB_RED_ZONE)
+		p += s->red_left_pad;
+
+	return p;
+}
+
 static inline bool kmem_cache_has_cpu_partial(struct kmem_cache *s)
 {
 #ifdef CONFIG_SLUB_CPU_PARTIAL
@@ -224,24 +232,6 @@ static inline void stat(const struct kmem_cache *s, enum stat_item si)
  * 			Core slab cache functions
  *******************************************************************/
 
-/* Verify that a pointer has an address that is valid within a slab page */
-static inline int check_valid_pointer(struct kmem_cache *s,
-				struct page *page, const void *object)
-{
-	void *base;
-
-	if (!object)
-		return 1;
-
-	base = page_address(page);
-	if (object < base || object >= base + page->objects * s->size ||
-		(object - base) % s->size) {
-		return 0;
-	}
-
-	return 1;
-}
-
 static inline void *get_freepointer(struct kmem_cache *s, void *object)
 {
 	return *(void **)(object + s->offset);
@@ -271,12 +261,14 @@ static inline void set_freepointer(struct kmem_cache *s, void *object, void *fp)
 
 /* Loop over all objects in a slab */
 #define for_each_object(__p, __s, __addr, __objects) \
-	for (__p = (__addr); __p < (__addr) + (__objects) * (__s)->size;\
+	for (__p = fixup_red_left(__s, __addr); \
-			__p += (__s)->size)
+		__p < (__addr) + (__objects) * (__s)->size; \
+		__p += (__s)->size)
 
 #define for_each_object_idx(__p, __idx, __s, __addr, __objects) \
-	for (__p = (__addr), __idx = 1; __idx <= __objects;\
+	for (__p = fixup_red_left(__s, __addr), __idx = 1; \
-			__p += (__s)->size, __idx++)
+		__idx <= __objects; \
+		__p += (__s)->size, __idx++)
 
 /* Determine object index from a given position */
 static inline int slab_index(void *p, struct kmem_cache *s, void *addr)
@@ -456,6 +448,22 @@ static void get_map(struct kmem_cache *s, struct page *page, unsigned long *map)
 		set_bit(slab_index(p, s, addr), map);
 }
 
+static inline int size_from_object(struct kmem_cache *s)
+{
+	if (s->flags & SLAB_RED_ZONE)
+		return s->size - s->red_left_pad;
+
+	return s->size;
+}
+
+static inline void *restore_red_left(struct kmem_cache *s, void *p)
+{
+	if (s->flags & SLAB_RED_ZONE)
+		p -= s->red_left_pad;
+
+	return p;
+}
+
 /*
  * Debug settings:
  */
@@ -489,6 +497,26 @@ static inline void metadata_access_disable(void)
 /*
  * Object debugging
  */
+
+/* Verify that a pointer has an address that is valid within a slab page */
+static inline int check_valid_pointer(struct kmem_cache *s,
+				struct page *page, void *object)
+{
+	void *base;
+
+	if (!object)
+		return 1;
+
+	base = page_address(page);
+	object = restore_red_left(s, object);
+	if (object < base || object >= base + page->objects * s->size ||
+		(object - base) % s->size) {
+		return 0;
+	}
+
+	return 1;
+}
+
 static void print_section(char *text, u8 *addr, unsigned int length)
 {
 	metadata_access_enable();
@@ -628,7 +656,9 @@ static void print_trailer(struct kmem_cache *s, struct page *page, u8 *p)
 	pr_err("INFO: Object 0x%p @offset=%tu fp=0x%p\n\n",
 	       p, p - addr, get_freepointer(s, p));
 
-	if (p > addr + 16)
+	if (s->flags & SLAB_RED_ZONE)
+		print_section("Redzone ", p - s->red_left_pad, s->red_left_pad);
+	else if (p > addr + 16)
 		print_section("Bytes b4 ", p - 16, 16);
 
 	print_section("Object ", p, min_t(unsigned long, s->object_size,
@@ -645,9 +675,9 @@ static void print_trailer(struct kmem_cache *s, struct page *page, u8 *p)
 	if (s->flags & SLAB_STORE_USER)
 		off += 2 * sizeof(struct track);
 
-	if (off != s->size)
+	if (off != size_from_object(s))
 		/* Beginning of the filler is the free pointer */
-		print_section("Padding ", p + off, s->size - off);
+		print_section("Padding ", p + off, size_from_object(s) - off);
 
 	dump_stack();
 }
@@ -677,6 +707,9 @@ static void init_object(struct kmem_cache *s, void *object, u8 val)
 {
 	u8 *p = object;
 
+	if (s->flags & SLAB_RED_ZONE)
+		memset(p - s->red_left_pad, val, s->red_left_pad);
+
 	if (s->flags & __OBJECT_POISON) {
 		memset(p, POISON_FREE, s->object_size - 1);
 		p[s->object_size - 1] = POISON_END;
@@ -769,11 +802,11 @@ static int check_pad_bytes(struct kmem_cache *s, struct page *page, u8 *p)
 		/* We also have user information there */
 		off += 2 * sizeof(struct track);
 
-	if (s->size == off)
+	if (size_from_object(s) == off)
 		return 1;
 
 	return check_bytes_and_report(s, page, p, "Object padding",
-				p + off, POISON_INUSE, s->size - off);
+			p + off, POISON_INUSE, size_from_object(s) - off);
 }
 
 /* Check the pad bytes at the end of a slab page */
@@ -817,6 +850,10 @@ static int check_object(struct kmem_cache *s, struct page *page,
 	u8 *endobject = object + s->object_size;
 
 	if (s->flags & SLAB_RED_ZONE) {
+		if (!check_bytes_and_report(s, page, object, "Redzone",
+			object - s->red_left_pad, val, s->red_left_pad))
+			return 0;
+
 		if (!check_bytes_and_report(s, page, object, "Redzone",
 			endobject, val, s->inuse - s->object_size))
 			return 0;
@@ -1468,7 +1505,7 @@ static struct page *allocate_slab(struct kmem_cache *s, gfp_t flags, int node)
 			set_freepointer(s, p, NULL);
 	}
 
-	page->freelist = start;
+	page->freelist = fixup_red_left(s, start);
 	page->inuse = page->objects;
 	page->frozen = 1;
 
@@ -3283,7 +3320,7 @@ static int calculate_sizes(struct kmem_cache *s, int forced_order)
 		 */
 		size += 2 * sizeof(struct track);
 
-	if (flags & SLAB_RED_ZONE)
+	if (flags & SLAB_RED_ZONE) {
 		/*
 		 * Add some empty padding so that we can catch
 		 * overwrites from earlier objects rather than let
@@ -3292,6 +3329,11 @@ static int calculate_sizes(struct kmem_cache *s, int forced_order)
 		 * of the object.
 		 */
 		size += sizeof(void *);
+
+		s->red_left_pad = sizeof(void *);
+		s->red_left_pad = ALIGN(s->red_left_pad, s->align);
+		size += s->red_left_pad;
+	}
 #endif
 
 	/*
@@ -3585,6 +3627,46 @@ void *__kmalloc_node(size_t size, gfp_t flags, int node)
 EXPORT_SYMBOL(__kmalloc_node);
 #endif
 
+#ifdef CONFIG_HARDENED_USERCOPY
+/*
+ * Rejects objects that are incorrectly sized.
+ *
+ * Returns NULL if check passes, otherwise const char * to name of cache
+ * to indicate an error.
+ */
+const char *__check_heap_object(const void *ptr, unsigned long n,
+				struct page *page)
+{
+	struct kmem_cache *s;
+	unsigned long offset;
+	size_t object_size;
+
+	/* Find object and usable object size. */
+	s = page->slab_cache;
+	object_size = slab_ksize(s);
+
+	/* Reject impossible pointers. */
+	if (ptr < page_address(page))
+		return s->name;
+
+	/* Find offset within object. */
+	offset = (ptr - page_address(page)) % s->size;
+
+	/* Adjust for redzone and reject if within the redzone. */
+	if (kmem_cache_debug(s) && s->flags & SLAB_RED_ZONE) {
+		if (offset < s->red_left_pad)
+			return s->name;
+		offset -= s->red_left_pad;
+	}
+
+	/* Allow address range falling entirely within object size. */
+	if (offset <= object_size && n <= object_size - offset)
+		return NULL;
+
+	return s->name;
+}
+#endif /* CONFIG_HARDENED_USERCOPY */
+
 static size_t __ksize(const void *object)
 {
 	struct page *page;

mm/usercopy.c

@@ -0,0 +1,269 @@
+/*
+ * This implements the various checks for CONFIG_HARDENED_USERCOPY*,
+ * which are designed to protect kernel memory from needless exposure
+ * and overwrite under many unintended conditions. This code is based
+ * on PAX_USERCOPY, which is:
+ *
+ * Copyright (C) 2001-2016 PaX Team, Bradley Spengler, Open Source
+ * Security Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/mm.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <asm/sections.h>
+
+enum {
+	BAD_STACK = -1,
+	NOT_STACK = 0,
+	GOOD_FRAME,
+	GOOD_STACK,
+};
+
+/*
+ * Checks if a given pointer and length is contained by the current
+ * stack frame (if possible).
+ *
+ * Returns:
+ *	NOT_STACK: not at all on the stack
+ *	GOOD_FRAME: fully within a valid stack frame
+ *	GOOD_STACK: fully on the stack (when can't do frame-checking)
+ *	BAD_STACK: error condition (invalid stack position or bad stack frame)
+ */
+static noinline int check_stack_object(const void *obj, unsigned long len)
+{
+	const void * const stack = task_stack_page(current);
+	const void * const stackend = stack + THREAD_SIZE;
+	int ret;
+
+	/* Object is not on the stack at all. */
+	if (obj + len <= stack || stackend <= obj)
+		return NOT_STACK;
+
+	/*
+	 * Reject: object partially overlaps the stack (passing the
+	 * the check above means at least one end is within the stack,
+	 * so if this check fails, the other end is outside the stack).
+	 */
+	if (obj < stack || stackend < obj + len)
+		return BAD_STACK;
+
+	/* Check if object is safely within a valid frame. */
+	ret = arch_within_stack_frames(stack, stackend, obj, len);
+	if (ret)
+		return ret;
+
+	return GOOD_STACK;
+}
+
+static void report_usercopy(const void *ptr, unsigned long len,
+			    bool to_user, const char *type)
+{
+	pr_emerg("kernel memory %s attempt detected %s %p (%s) (%lu bytes)\n",
+		to_user ? "exposure" : "overwrite",
+		to_user ? "from" : "to", ptr, type ? : "unknown", len);
+	/*
+	 * For greater effect, it would be nice to do do_group_exit(),
+	 * but BUG() actually hooks all the lock-breaking and per-arch
+	 * Oops code, so that is used here instead.
+	 */
+	BUG();
+}
+
+/* Returns true if any portion of [ptr,ptr+n) over laps with [low,high). */
+static bool overlaps(const void *ptr, unsigned long n, unsigned long low,
+		     unsigned long high)
+{
+	unsigned long check_low = (uintptr_t)ptr;
+	unsigned long check_high = check_low + n;
+
+	/* Does not overlap if entirely above or entirely below. */
+	if (check_low >= high || check_high <= low)
+		return false;
+
+	return true;
+}
+
+/* Is this address range in the kernel text area? */
+static inline const char *check_kernel_text_object(const void *ptr,
+						   unsigned long n)
+{
+	unsigned long textlow = (unsigned long)_stext;
+	unsigned long texthigh = (unsigned long)_etext;
+	unsigned long textlow_linear, texthigh_linear;
+
+	if (overlaps(ptr, n, textlow, texthigh))
+		return "<kernel text>";
+
+	/*
+	 * Some architectures have virtual memory mappings with a secondary
+	 * mapping of the kernel text, i.e. there is more than one virtual
+	 * kernel address that points to the kernel image. It is usually
+	 * when there is a separate linear physical memory mapping, in that
+	 * __pa() is not just the reverse of __va(). This can be detected
+	 * and checked:
+	 */
+	textlow_linear = (unsigned long)__va(__pa(textlow));
+	/* No different mapping: we're done. */
+	if (textlow_linear == textlow)
+		return NULL;
+
+	/* Check the secondary mapping... */
+	texthigh_linear = (unsigned long)__va(__pa(texthigh));
+	if (overlaps(ptr, n, textlow_linear, texthigh_linear))
+		return "<linear kernel text>";
+
+	return NULL;
+}
+
+static inline const char *check_bogus_address(const void *ptr, unsigned long n)
+{
+	/* Reject if object wraps past end of memory. */
+	if ((unsigned long)ptr + n < (unsigned long)ptr)
+		return "<wrapped address>";
+
+	/* Reject if NULL or ZERO-allocation. */
+	if (ZERO_OR_NULL_PTR(ptr))
+		return "<null>";
+
+	return NULL;
+}
+
+static inline const char *check_heap_object(const void *ptr, unsigned long n,
+					    bool to_user)
+{
+	struct page *page, *endpage;
+	const void *end = ptr + n - 1;
+	bool is_reserved, is_cma;
+
+	/*
+	 * Some architectures (arm64) return true for virt_addr_valid() on
+	 * vmalloced addresses. Work around this by checking for vmalloc
+	 * first.
+	 */
+	if (is_vmalloc_addr(ptr))
+		return NULL;
+
+	if (!virt_addr_valid(ptr))
+		return NULL;
+
+	page = virt_to_head_page(ptr);
+
+	/* Check slab allocator for flags and size. */
+	if (PageSlab(page))
+		return __check_heap_object(ptr, n, page);
+
+	/*
+	 * Sometimes the kernel data regions are not marked Reserved (see
+	 * check below). And sometimes [_sdata,_edata) does not cover
+	 * rodata and/or bss, so check each range explicitly.
+	 */
+
+	/* Allow reads of kernel rodata region (if not marked as Reserved). */
+	if (ptr >= (const void *)__start_rodata &&
+	    end <= (const void *)__end_rodata) {
+		if (!to_user)
+			return "<rodata>";
+		return NULL;
+	}
+
+	/* Allow kernel data region (if not marked as Reserved). */
+	if (ptr >= (const void *)_sdata && end <= (const void *)_edata)
+		return NULL;
+
+	/* Allow kernel bss region (if not marked as Reserved). */
+	if (ptr >= (const void *)__bss_start &&
+	    end <= (const void *)__bss_stop)
+		return NULL;
+
+	/* Is the object wholly within one base page? */
+	if (likely(((unsigned long)ptr & (unsigned long)PAGE_MASK) ==
+		   ((unsigned long)end & (unsigned long)PAGE_MASK)))
+		return NULL;
+
+	/* Allow if start and end are inside the same compound page. */
+	endpage = virt_to_head_page(end);
+	if (likely(endpage == page))
+		return NULL;
+
+	/*
+	 * Reject if range is entirely either Reserved (i.e. special or
+	 * device memory), or CMA. Otherwise, reject since the object spans
+	 * several independently allocated pages.
+	 */
+	is_reserved = PageReserved(page);
+	is_cma = is_migrate_cma_page(page);
+	if (!is_reserved && !is_cma)
+		goto reject;
+
+	for (ptr += PAGE_SIZE; ptr <= end; ptr += PAGE_SIZE) {
+		page = virt_to_head_page(ptr);
+		if (is_reserved && !PageReserved(page))
+			goto reject;
+		if (is_cma && !is_migrate_cma_page(page))
+			goto reject;
+	}
+
+	return NULL;
+
+reject:
+	return "<spans multiple pages>";
+}
+
+/*
+ * Validates that the given object is:
+ * - not bogus address
+ * - known-safe heap or stack object
+ * - not in kernel text
+ */
+void __check_object_size(const void *ptr, unsigned long n, bool to_user)
+{
+	const char *err;
+
+	/* Skip all tests if size is zero. */
+	if (!n)
+		return;
+
+	/* Check for invalid addresses. */
+	err = check_bogus_address(ptr, n);
+	if (err)
+		goto report;
+
+	/* Check for bad heap object. */
+	err = check_heap_object(ptr, n, to_user);
+	if (err)
+		goto report;
+
+	/* Check for bad stack object. */
+	switch (check_stack_object(ptr, n)) {
+	case NOT_STACK:
+		/* Object is not touching the current process stack. */
+		break;
+	case GOOD_FRAME:
+	case GOOD_STACK:
+		/*
+		 * Object is either in the correct frame (when it
+		 * is possible to check) or just generally on the
+		 * process stack (when frame checking not available).
+		 */
+		return;
+	default:
+		err = "<process stack>";
+		goto report;
+	}
+
+	/* Check for object in kernel to avoid text exposure. */
+	err = check_kernel_text_object(ptr, n);
+	if (!err)
+		return;
+
+report:
+	report_usercopy(ptr, n, to_user, err);
+}
+EXPORT_SYMBOL(__check_object_size);

net/netfilter/nfnetlink.c

@@ -326,10 +326,12 @@ replay:
 		nlh = nlmsg_hdr(skb);
 		err = 0;
 
-		if (nlmsg_len(nlh) < sizeof(struct nfgenmsg) ||
+		if (nlh->nlmsg_len < NLMSG_HDRLEN ||
-		    skb->len < nlh->nlmsg_len) {
+		    skb->len < nlh->nlmsg_len ||
-			err = -EINVAL;
+		    nlmsg_len(nlh) < sizeof(struct nfgenmsg)) {
-			goto ack;
+			nfnl_err_reset(&err_list);
+			status |= NFNL_BATCH_FAILURE;
+			goto done;
 		}
 
 		/* Only requests are handled by the kernel */

net/netfilter/xt_qtaguid.c

@@ -1946,7 +1946,7 @@ static int qtaguid_ctrl_proc_show(struct seq_file *m, void *v)
 			);
 		f_count = atomic_long_read(
 			&sock_tag_entry->socket->file->f_count);
-		seq_printf(m, "sock=%p tag=0x%llx (uid=%u) pid=%u "
+		seq_printf(m, "sock=%pK tag=0x%llx (uid=%u) pid=%u "
 			   "f_count=%lu\n",
 			   sock_tag_entry->sk,
 			   sock_tag_entry->tag, uid,
@@ -2548,8 +2548,7 @@ static int pp_stats_line(struct seq_file *m, struct tag_stat *ts_entry,
 	uid_t stat_uid = get_uid_from_tag(tag);
 	struct proc_print_info *ppi = m->private;
 	/* Detailed tags are not available to everybody */
-	if (get_atag_from_tag(tag) && !can_read_other_uid_stats(
+	if (!can_read_other_uid_stats(make_kuid(&init_user_ns,stat_uid))) {
-						make_kuid(&init_user_ns,stat_uid))) {
 		CT_DEBUG("qtaguid: stats line: "
 			 "%s 0x%llx %u: insufficient priv "
 			 "from pid=%u tgid=%u uid=%u stats.gid=%u\n",

security/Kconfig

@@ -127,6 +127,35 @@ config LSM_MMAP_MIN_ADDR
 	  this low address space will need the permission specific to the
 	  systems running LSM.
 
+config HAVE_HARDENED_USERCOPY_ALLOCATOR
+	bool
+	help
+	  The heap allocator implements __check_heap_object() for
+	  validating memory ranges against heap object sizes in
+	  support of CONFIG_HARDENED_USERCOPY.
+
+config HAVE_ARCH_HARDENED_USERCOPY
+	bool
+	help
+	  The architecture supports CONFIG_HARDENED_USERCOPY by
+	  calling check_object_size() just before performing the
+	  userspace copies in the low level implementation of
+	  copy_to_user() and copy_from_user().
+
+config HARDENED_USERCOPY
+	bool "Harden memory copies between kernel and userspace"
+	depends on HAVE_ARCH_HARDENED_USERCOPY
+	depends on HAVE_HARDENED_USERCOPY_ALLOCATOR
+	select BUG
+	help
+	  This option checks for obviously wrong memory regions when
+	  copying memory to/from the kernel (via copy_to_user() and
+	  copy_from_user() functions) by rejecting memory ranges that
+	  are larger than the specified heap object, span multiple
+	  separately allocates pages, are not on the process stack,
+	  or are part of the kernel text. This kills entire classes
+	  of heap overflow exploits and similar kernel memory exposures.
+
 source security/selinux/Kconfig
 source security/smack/Kconfig
 source security/tomoyo/Kconfig