diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index 0e49d39..1be5f61 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -26,6 +26,7 @@
 	select ODD_RT_SIGACTION
 	select OLD_SIGSUSPEND
 	select CPU_NO_EFFICIENT_FFS if !ALPHA_EV67
+	select ARCH_HAS_RAW_COPY_USER
 	help
 	  The Alpha is a 64-bit general-purpose processor designed and
 	  marketed by the Digital Equipment Corporation of blessed memory,
diff --git a/arch/alpha/include/asm/extable.h b/arch/alpha/include/asm/extable.h
new file mode 100644
index 0000000..048e209
--- /dev/null
+++ b/arch/alpha/include/asm/extable.h
@@ -0,0 +1,55 @@
+#ifndef _ASM_EXTABLE_H
+#define _ASM_EXTABLE_H
+
+/*
+ * About the exception table:
+ *
+ * - insn is a 32-bit pc-relative offset from the faulting insn.
+ * - nextinsn is a 16-bit offset off of the faulting instruction
+ *   (not off of the *next* instruction as branches are).
+ * - errreg is the register in which to place -EFAULT.
+ * - valreg is the final target register for the load sequence
+ *   and will be zeroed.
+ *
+ * Either errreg or valreg may be $31, in which case nothing happens.
+ *
+ * The exception fixup information "just so happens" to be arranged
+ * as in a MEM format instruction.  This lets us emit our three
+ * values like so:
+ *
+ *      lda valreg, nextinsn(errreg)
+ *
+ */
+
+struct exception_table_entry
+{
+	signed int insn;
+	union exception_fixup {
+		unsigned unit;
+		struct {
+			signed int nextinsn : 16;
+			unsigned int errreg : 5;
+			unsigned int valreg : 5;
+		} bits;
+	} fixup;
+};
+
+/* Returns the new pc */
+#define fixup_exception(map_reg, _fixup, pc)			\
+({								\
+	if ((_fixup)->fixup.bits.valreg != 31)			\
+		map_reg((_fixup)->fixup.bits.valreg) = 0;	\
+	if ((_fixup)->fixup.bits.errreg != 31)			\
+		map_reg((_fixup)->fixup.bits.errreg) = -EFAULT;	\
+	(pc) + (_fixup)->fixup.bits.nextinsn;			\
+})
+
+#define ARCH_HAS_RELATIVE_EXTABLE
+
+#define swap_ex_entry_fixup(a, b, tmp, delta)			\
+	do {							\
+		(a)->fixup.unit = (b)->fixup.unit;		\
+		(b)->fixup.unit = (tmp).fixup.unit;		\
+	} while (0)
+
+#endif
diff --git a/arch/alpha/include/asm/futex.h b/arch/alpha/include/asm/futex.h
index f939794..fb01dfb 100644
--- a/arch/alpha/include/asm/futex.h
+++ b/arch/alpha/include/asm/futex.h
@@ -19,12 +19,8 @@
 	"3:	.subsection 2\n"				\
 	"4:	br	1b\n"					\
 	"	.previous\n"					\
-	"	.section __ex_table,\"a\"\n"			\
-	"	.long	1b-.\n"					\
-	"	lda	$31,3b-1b(%1)\n"			\
-	"	.long	2b-.\n"					\
-	"	lda	$31,3b-2b(%1)\n"			\
-	"	.previous\n"					\
+	EXC(1b,3b,%1,$31)					\
+	EXC(2b,3b,%1,$31)					\
 	:	"=&r" (oldval), "=&r"(ret)			\
 	:	"r" (uaddr), "r"(oparg)				\
 	:	"memory")
@@ -101,12 +97,8 @@ futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
 	"3:	.subsection 2\n"
 	"4:	br	1b\n"
 	"	.previous\n"
-	"	.section __ex_table,\"a\"\n"
-	"	.long	1b-.\n"
-	"	lda	$31,3b-1b(%0)\n"
-	"	.long	2b-.\n"
-	"	lda	$31,3b-2b(%0)\n"
-	"	.previous\n"
+	EXC(1b,3b,%0,$31)
+	EXC(2b,3b,%0,$31)
 	:	"+r"(ret), "=&r"(prev), "=&r"(cmp)
 	:	"r"(uaddr), "r"((long)(int)oldval), "r"(newval)
 	:	"memory");
diff --git a/arch/alpha/include/asm/uaccess.h b/arch/alpha/include/asm/uaccess.h
index b2f62b9..7b82dc9 100644
--- a/arch/alpha/include/asm/uaccess.h
+++ b/arch/alpha/include/asm/uaccess.h
@@ -32,13 +32,13 @@
  *  - AND "addr+size" doesn't have any high-bits set
  *  - OR we are in kernel mode.
  */
-#define __access_ok(addr, size, segment) \
-	(((segment).seg & (addr | size | (addr+size))) == 0)
+#define __access_ok(addr, size) \
+	((get_fs().seg & (addr | size | (addr+size))) == 0)
 
-#define access_ok(type, addr, size)				\
-({								\
-	__chk_user_ptr(addr);					\
-	__access_ok(((unsigned long)(addr)), (size), get_fs());	\
+#define access_ok(type, addr, size)			\
+({							\
+	__chk_user_ptr(addr);				\
+	__access_ok(((unsigned long)(addr)), (size));	\
 })
 
 /*
@@ -54,9 +54,9 @@
  * (b) require any knowledge of processes at this stage
  */
 #define put_user(x, ptr) \
-  __put_user_check((__typeof__(*(ptr)))(x), (ptr), sizeof(*(ptr)), get_fs())
+  __put_user_check((__typeof__(*(ptr)))(x), (ptr), sizeof(*(ptr)))
 #define get_user(x, ptr) \
-  __get_user_check((x), (ptr), sizeof(*(ptr)), get_fs())
+  __get_user_check((x), (ptr), sizeof(*(ptr)))
 
 /*
  * The "__xxx" versions do not do address space checking, useful when
@@ -74,6 +74,11 @@
  * more extensive comments with fixup_inline_exception below for
  * more information.
  */
+#define EXC(label,cont,res,err)				\
+	".section __ex_table,\"a\"\n"			\
+	"	.long "#label"-.\n"			\
+	"	lda "#res","#cont"-"#label"("#err")\n"	\
+	".previous\n"
 
 extern void __get_user_unknown(void);
 
@@ -93,23 +98,23 @@ extern void __get_user_unknown(void);
 	__gu_err;						\
 })
 
-#define __get_user_check(x, ptr, size, segment)				\
-({									\
-	long __gu_err = -EFAULT;					\
-	unsigned long __gu_val = 0;					\
-	const __typeof__(*(ptr)) __user *__gu_addr = (ptr);		\
-	if (__access_ok((unsigned long)__gu_addr, size, segment)) {	\
-		__gu_err = 0;						\
-		switch (size) {						\
-		  case 1: __get_user_8(__gu_addr); break;		\
-		  case 2: __get_user_16(__gu_addr); break;		\
-		  case 4: __get_user_32(__gu_addr); break;		\
-		  case 8: __get_user_64(__gu_addr); break;		\
-		  default: __get_user_unknown(); break;			\
-		}							\
-	}								\
-	(x) = (__force __typeof__(*(ptr))) __gu_val;			\
-	__gu_err;							\
+#define __get_user_check(x, ptr, size)				\
+({								\
+	long __gu_err = -EFAULT;				\
+	unsigned long __gu_val = 0;				\
+	const __typeof__(*(ptr)) __user *__gu_addr = (ptr);	\
+	if (__access_ok((unsigned long)__gu_addr, size)) {	\
+		__gu_err = 0;					\
+		switch (size) {					\
+		  case 1: __get_user_8(__gu_addr); break;	\
+		  case 2: __get_user_16(__gu_addr); break;	\
+		  case 4: __get_user_32(__gu_addr); break;	\
+		  case 8: __get_user_64(__gu_addr); break;	\
+		  default: __get_user_unknown(); break;		\
+		}						\
+	}							\
+	(x) = (__force __typeof__(*(ptr))) __gu_val;		\
+	__gu_err;						\
 })
 
 struct __large_struct { unsigned long buf[100]; };
@@ -118,20 +123,14 @@ struct __large_struct { unsigned long buf[100]; };
 #define __get_user_64(addr)				\
 	__asm__("1: ldq %0,%2\n"			\
 	"2:\n"						\
-	".section __ex_table,\"a\"\n"			\
-	"	.long 1b - .\n"				\
-	"	lda %0, 2b-1b(%1)\n"			\
-	".previous"					\
+	EXC(1b,2b,%0,%1)				\
 		: "=r"(__gu_val), "=r"(__gu_err)	\
 		: "m"(__m(addr)), "1"(__gu_err))
 
 #define __get_user_32(addr)				\
 	__asm__("1: ldl %0,%2\n"			\
 	"2:\n"						\
-	".section __ex_table,\"a\"\n"			\
-	"	.long 1b - .\n"				\
-	"	lda %0, 2b-1b(%1)\n"			\
-	".previous"					\
+	EXC(1b,2b,%0,%1)				\
 		: "=r"(__gu_val), "=r"(__gu_err)	\
 		: "m"(__m(addr)), "1"(__gu_err))
 
@@ -141,20 +140,14 @@ struct __large_struct { unsigned long buf[100]; };
 #define __get_user_16(addr)				\
 	__asm__("1: ldwu %0,%2\n"			\
 	"2:\n"						\
-	".section __ex_table,\"a\"\n"			\
-	"	.long 1b - .\n"				\
-	"	lda %0, 2b-1b(%1)\n"			\
-	".previous"					\
+	EXC(1b,2b,%0,%1)				\
 		: "=r"(__gu_val), "=r"(__gu_err)	\
 		: "m"(__m(addr)), "1"(__gu_err))
 
 #define __get_user_8(addr)				\
 	__asm__("1: ldbu %0,%2\n"			\
 	"2:\n"						\
-	".section __ex_table,\"a\"\n"			\
-	"	.long 1b - .\n"				\
-	"	lda %0, 2b-1b(%1)\n"			\
-	".previous"					\
+	EXC(1b,2b,%0,%1)				\
 		: "=r"(__gu_val), "=r"(__gu_err)	\
 		: "m"(__m(addr)), "1"(__gu_err))
 #else
@@ -170,12 +163,8 @@ struct __large_struct { unsigned long buf[100]; };
 	"	extwh %1,%3,%1\n"					\
 	"	or %0,%1,%0\n"						\
 	"3:\n"								\
-	".section __ex_table,\"a\"\n"					\
-	"	.long 1b - .\n"						\
-	"	lda %0, 3b-1b(%2)\n"					\
-	"	.long 2b - .\n"						\
-	"	lda %0, 3b-2b(%2)\n"					\
-	".previous"							\
+	EXC(1b,3b,%0,%2)						\
+	EXC(2b,3b,%0,%2)						\
 		: "=&r"(__gu_val), "=&r"(__gu_tmp), "=r"(__gu_err)	\
 		: "r"(addr), "2"(__gu_err));				\
 }
@@ -184,10 +173,7 @@ struct __large_struct { unsigned long buf[100]; };
 	__asm__("1: ldq_u %0,0(%2)\n"					\
 	"	extbl %0,%2,%0\n"					\
 	"2:\n"								\
-	".section __ex_table,\"a\"\n"					\
-	"	.long 1b - .\n"						\
-	"	lda %0, 2b-1b(%1)\n"					\
-	".previous"							\
+	EXC(1b,2b,%0,%1)						\
 		: "=&r"(__gu_val), "=r"(__gu_err)			\
 		: "r"(addr), "1"(__gu_err))
 #endif
@@ -208,21 +194,21 @@ extern void __put_user_unknown(void);
 	__pu_err;						\
 })
 
-#define __put_user_check(x, ptr, size, segment)				\
-({									\
-	long __pu_err = -EFAULT;					\
-	__typeof__(*(ptr)) __user *__pu_addr = (ptr);			\
-	if (__access_ok((unsigned long)__pu_addr, size, segment)) {	\
-		__pu_err = 0;						\
-		switch (size) {						\
-		  case 1: __put_user_8(x, __pu_addr); break;		\
-		  case 2: __put_user_16(x, __pu_addr); break;		\
-		  case 4: __put_user_32(x, __pu_addr); break;		\
-		  case 8: __put_user_64(x, __pu_addr); break;		\
-		  default: __put_user_unknown(); break;			\
-		}							\
-	}								\
-	__pu_err;							\
+#define __put_user_check(x, ptr, size)				\
+({								\
+	long __pu_err = -EFAULT;				\
+	__typeof__(*(ptr)) __user *__pu_addr = (ptr);		\
+	if (__access_ok((unsigned long)__pu_addr, size)) {	\
+		__pu_err = 0;					\
+		switch (size) {					\
+		  case 1: __put_user_8(x, __pu_addr); break;	\
+		  case 2: __put_user_16(x, __pu_addr); break;	\
+		  case 4: __put_user_32(x, __pu_addr); break;	\
+		  case 8: __put_user_64(x, __pu_addr); break;	\
+		  default: __put_user_unknown(); break;		\
+		}						\
+	}							\
+	__pu_err;						\
 })
 
 /*
@@ -233,20 +219,14 @@ extern void __put_user_unknown(void);
 #define __put_user_64(x, addr)					\
 __asm__ __volatile__("1: stq %r2,%1\n"				\
 	"2:\n"							\
-	".section __ex_table,\"a\"\n"				\
-	"	.long 1b - .\n"					\
-	"	lda $31,2b-1b(%0)\n"				\
-	".previous"						\
+	EXC(1b,2b,$31,%0)					\
 		: "=r"(__pu_err)				\
 		: "m" (__m(addr)), "rJ" (x), "0"(__pu_err))
 
 #define __put_user_32(x, addr)					\
 __asm__ __volatile__("1: stl %r2,%1\n"				\
 	"2:\n"							\
-	".section __ex_table,\"a\"\n"				\
-	"	.long 1b - .\n"					\
-	"	lda $31,2b-1b(%0)\n"				\
-	".previous"						\
+	EXC(1b,2b,$31,%0)					\
 		: "=r"(__pu_err)				\
 		: "m"(__m(addr)), "rJ"(x), "0"(__pu_err))
 
@@ -256,20 +236,14 @@ __asm__ __volatile__("1: stl %r2,%1\n"				\
 #define __put_user_16(x, addr)					\
 __asm__ __volatile__("1: stw %r2,%1\n"				\
 	"2:\n"							\
-	".section __ex_table,\"a\"\n"				\
-	"	.long 1b - .\n"					\
-	"	lda $31,2b-1b(%0)\n"				\
-	".previous"						\
+	EXC(1b,2b,$31,%0)					\
 		: "=r"(__pu_err)				\
 		: "m"(__m(addr)), "rJ"(x), "0"(__pu_err))
 
 #define __put_user_8(x, addr)					\
 __asm__ __volatile__("1: stb %r2,%1\n"				\
 	"2:\n"							\
-	".section __ex_table,\"a\"\n"				\
-	"	.long 1b - .\n"					\
-	"	lda $31,2b-1b(%0)\n"				\
-	".previous"						\
+	EXC(1b,2b,$31,%0)					\
 		: "=r"(__pu_err)				\
 		: "m"(__m(addr)), "rJ"(x), "0"(__pu_err))
 #else
@@ -291,16 +265,10 @@ __asm__ __volatile__("1: stb %r2,%1\n"				\
 	"3:	stq_u %2,1(%5)\n"				\
 	"4:	stq_u %1,0(%5)\n"				\
 	"5:\n"							\
-	".section __ex_table,\"a\"\n"				\
-	"	.long 1b - .\n"					\
-	"	lda $31, 5b-1b(%0)\n"				\
-	"	.long 2b - .\n"					\
-	"	lda $31, 5b-2b(%0)\n"				\
-	"	.long 3b - .\n"					\
-	"	lda $31, 5b-3b(%0)\n"				\
-	"	.long 4b - .\n"					\
-	"	lda $31, 5b-4b(%0)\n"				\
-	".previous"						\
+	EXC(1b,5b,$31,%0)					\
+	EXC(2b,5b,$31,%0)					\
+	EXC(3b,5b,$31,%0)					\
+	EXC(4b,5b,$31,%0)					\
 		: "=r"(__pu_err), "=&r"(__pu_tmp1), 		\
 		  "=&r"(__pu_tmp2), "=&r"(__pu_tmp3), 		\
 		  "=&r"(__pu_tmp4)				\
@@ -317,12 +285,8 @@ __asm__ __volatile__("1: stb %r2,%1\n"				\
 	"	or %1,%2,%1\n"					\
 	"2:	stq_u %1,0(%4)\n"				\
 	"3:\n"							\
-	".section __ex_table,\"a\"\n"				\
-	"	.long 1b - .\n"					\
-	"	lda $31, 3b-1b(%0)\n"				\
-	"	.long 2b - .\n"					\
-	"	lda $31, 3b-2b(%0)\n"				\
-	".previous"						\
+	EXC(1b,3b,$31,%0)					\
+	EXC(2b,3b,$31,%0)					\
 		: "=r"(__pu_err), 				\
 	  	  "=&r"(__pu_tmp1), "=&r"(__pu_tmp2)		\
 		: "r"((unsigned long)(x)), "r"(addr), "0"(__pu_err)); \
@@ -334,96 +298,30 @@ __asm__ __volatile__("1: stb %r2,%1\n"				\
  * Complex access routines
  */
 
-/* This little bit of silliness is to get the GP loaded for a function
-   that ordinarily wouldn't.  Otherwise we could have it done by the macro
-   directly, which can be optimized the linker.  */
-#ifdef MODULE
-#define __module_address(sym)		"r"(sym),
-#define __module_call(ra, arg, sym)	"jsr $" #ra ",(%" #arg ")," #sym
-#else
-#define __module_address(sym)
-#define __module_call(ra, arg, sym)	"bsr $" #ra "," #sym " !samegp"
-#endif
+extern long __copy_user(void *to, const void *from, long len);
 
-extern void __copy_user(void);
-
-extern inline long
-__copy_tofrom_user_nocheck(void *to, const void *from, long len)
+static inline unsigned long
+raw_copy_from_user(void *to, const void __user *from, unsigned long len)
 {
-	register void * __cu_to __asm__("$6") = to;
-	register const void * __cu_from __asm__("$7") = from;
-	register long __cu_len __asm__("$0") = len;
-
-	__asm__ __volatile__(
-		__module_call(28, 3, __copy_user)
-		: "=r" (__cu_len), "=r" (__cu_from), "=r" (__cu_to)
-		: __module_address(__copy_user)
-		  "0" (__cu_len), "1" (__cu_from), "2" (__cu_to)
-		: "$1", "$2", "$3", "$4", "$5", "$28", "memory");
-
-	return __cu_len;
+	return __copy_user(to, (__force const void *)from, len);
 }
 
-#define __copy_to_user(to, from, n)					\
-({									\
-	__chk_user_ptr(to);						\
-	__copy_tofrom_user_nocheck((__force void *)(to), (from), (n));	\
-})
-#define __copy_from_user(to, from, n)					\
-({									\
-	__chk_user_ptr(from);						\
-	__copy_tofrom_user_nocheck((to), (__force void *)(from), (n));	\
-})
-
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
-
-extern inline long
-copy_to_user(void __user *to, const void *from, long n)
+static inline unsigned long
+raw_copy_to_user(void __user *to, const void *from, unsigned long len)
 {
-	if (likely(__access_ok((unsigned long)to, n, get_fs())))
-		n = __copy_tofrom_user_nocheck((__force void *)to, from, n);
-	return n;
+	return __copy_user((__force void *)to, from, len);
 }
 
-extern inline long
-copy_from_user(void *to, const void __user *from, long n)
-{
-	long res = n;
-	if (likely(__access_ok((unsigned long)from, n, get_fs())))
-		res = __copy_from_user_inatomic(to, from, n);
-	if (unlikely(res))
-		memset(to + (n - res), 0, res);
-	return res;
-}
-
-extern void __do_clear_user(void);
-
-extern inline long
-__clear_user(void __user *to, long len)
-{
-	register void __user * __cl_to __asm__("$6") = to;
-	register long __cl_len __asm__("$0") = len;
-	__asm__ __volatile__(
-		__module_call(28, 2, __do_clear_user)
-		: "=r"(__cl_len), "=r"(__cl_to)
-		: __module_address(__do_clear_user)
-		  "0"(__cl_len), "1"(__cl_to)
-		: "$1", "$2", "$3", "$4", "$5", "$28", "memory");
-	return __cl_len;
-}
+extern long __clear_user(void __user *to, long len);
 
 extern inline long
 clear_user(void __user *to, long len)
 {
-	if (__access_ok((unsigned long)to, len, get_fs()))
+	if (__access_ok((unsigned long)to, len))
 		len = __clear_user(to, len);
 	return len;
 }
 
-#undef __module_address
-#undef __module_call
-
 #define user_addr_max() \
         (uaccess_kernel() ? ~0UL : TASK_SIZE)
 
@@ -431,56 +329,6 @@ extern long strncpy_from_user(char *dest, const char __user *src, long count);
 extern __must_check long strlen_user(const char __user *str);
 extern __must_check long strnlen_user(const char __user *str, long n);
 
-/*
- * About the exception table:
- *
- * - insn is a 32-bit pc-relative offset from the faulting insn.
- * - nextinsn is a 16-bit offset off of the faulting instruction
- *   (not off of the *next* instruction as branches are).
- * - errreg is the register in which to place -EFAULT.
- * - valreg is the final target register for the load sequence
- *   and will be zeroed.
- *
- * Either errreg or valreg may be $31, in which case nothing happens.
- *
- * The exception fixup information "just so happens" to be arranged
- * as in a MEM format instruction.  This lets us emit our three
- * values like so:
- *
- *      lda valreg, nextinsn(errreg)
- *
- */
-
-struct exception_table_entry
-{
-	signed int insn;
-	union exception_fixup {
-		unsigned unit;
-		struct {
-			signed int nextinsn : 16;
-			unsigned int errreg : 5;
-			unsigned int valreg : 5;
-		} bits;
-	} fixup;
-};
-
-/* Returns the new pc */
-#define fixup_exception(map_reg, _fixup, pc)			\
-({								\
-	if ((_fixup)->fixup.bits.valreg != 31)			\
-		map_reg((_fixup)->fixup.bits.valreg) = 0;	\
-	if ((_fixup)->fixup.bits.errreg != 31)			\
-		map_reg((_fixup)->fixup.bits.errreg) = -EFAULT;	\
-	(pc) + (_fixup)->fixup.bits.nextinsn;			\
-})
-
-#define ARCH_HAS_RELATIVE_EXTABLE
-
-#define swap_ex_entry_fixup(a, b, tmp, delta)			\
-	do {							\
-		(a)->fixup.unit = (b)->fixup.unit;		\
-		(b)->fixup.unit = (tmp).fixup.unit;		\
-	} while (0)
-
+#include <asm/extable.h>
 
 #endif /* __ALPHA_UACCESS_H */
diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c
index 0b96109..6d76e52 100644
--- a/arch/alpha/kernel/osf_sys.c
+++ b/arch/alpha/kernel/osf_sys.c
@@ -1290,7 +1290,7 @@ SYSCALL_DEFINE1(old_adjtimex, struct timex32 __user *, txc_p)
 	/* copy relevant bits of struct timex. */
 	if (copy_from_user(&txc, txc_p, offsetof(struct timex32, time)) ||
 	    copy_from_user(&txc.tick, &txc_p->tick, sizeof(struct timex32) - 
-			   offsetof(struct timex32, time)))
+			   offsetof(struct timex32, tick)))
 	  return -EFAULT;
 
 	ret = do_adjtimex(&txc);	
diff --git a/arch/alpha/kernel/traps.c b/arch/alpha/kernel/traps.c
index b137390..65bb102 100644
--- a/arch/alpha/kernel/traps.c
+++ b/arch/alpha/kernel/traps.c
@@ -482,12 +482,8 @@ do_entUna(void * va, unsigned long opcode, unsigned long reg,
 		"	extwl %1,%3,%1\n"
 		"	extwh %2,%3,%2\n"
 		"3:\n"
-		".section __ex_table,\"a\"\n"
-		"	.long 1b - .\n"
-		"	lda %1,3b-1b(%0)\n"
-		"	.long 2b - .\n"
-		"	lda %2,3b-2b(%0)\n"
-		".previous"
+		EXC(1b,3b,%1,%0)
+		EXC(2b,3b,%2,%0)
 			: "=r"(error), "=&r"(tmp1), "=&r"(tmp2)
 			: "r"(va), "0"(0));
 		if (error)
@@ -502,12 +498,8 @@ do_entUna(void * va, unsigned long opcode, unsigned long reg,
 		"	extll %1,%3,%1\n"
 		"	extlh %2,%3,%2\n"
 		"3:\n"
-		".section __ex_table,\"a\"\n"
-		"	.long 1b - .\n"
-		"	lda %1,3b-1b(%0)\n"
-		"	.long 2b - .\n"
-		"	lda %2,3b-2b(%0)\n"
-		".previous"
+		EXC(1b,3b,%1,%0)
+		EXC(2b,3b,%2,%0)
 			: "=r"(error), "=&r"(tmp1), "=&r"(tmp2)
 			: "r"(va), "0"(0));
 		if (error)
@@ -522,12 +514,8 @@ do_entUna(void * va, unsigned long opcode, unsigned long reg,
 		"	extql %1,%3,%1\n"
 		"	extqh %2,%3,%2\n"
 		"3:\n"
-		".section __ex_table,\"a\"\n"
-		"	.long 1b - .\n"
-		"	lda %1,3b-1b(%0)\n"
-		"	.long 2b - .\n"
-		"	lda %2,3b-2b(%0)\n"
-		".previous"
+		EXC(1b,3b,%1,%0)
+		EXC(2b,3b,%2,%0)
 			: "=r"(error), "=&r"(tmp1), "=&r"(tmp2)
 			: "r"(va), "0"(0));
 		if (error)
@@ -551,16 +539,10 @@ do_entUna(void * va, unsigned long opcode, unsigned long reg,
 		"3:	stq_u %2,1(%5)\n"
 		"4:	stq_u %1,0(%5)\n"
 		"5:\n"
-		".section __ex_table,\"a\"\n"
-		"	.long 1b - .\n"
-		"	lda %2,5b-1b(%0)\n"
-		"	.long 2b - .\n"
-		"	lda %1,5b-2b(%0)\n"
-		"	.long 3b - .\n"
-		"	lda $31,5b-3b(%0)\n"
-		"	.long 4b - .\n"
-		"	lda $31,5b-4b(%0)\n"
-		".previous"
+		EXC(1b,5b,%2,%0)
+		EXC(2b,5b,%1,%0)
+		EXC(3b,5b,$31,%0)
+		EXC(4b,5b,$31,%0)
 			: "=r"(error), "=&r"(tmp1), "=&r"(tmp2),
 			  "=&r"(tmp3), "=&r"(tmp4)
 			: "r"(va), "r"(una_reg(reg)), "0"(0));
@@ -581,16 +563,10 @@ do_entUna(void * va, unsigned long opcode, unsigned long reg,
 		"3:	stq_u %2,3(%5)\n"
 		"4:	stq_u %1,0(%5)\n"
 		"5:\n"
-		".section __ex_table,\"a\"\n"
-		"	.long 1b - .\n"
-		"	lda %2,5b-1b(%0)\n"
-		"	.long 2b - .\n"
-		"	lda %1,5b-2b(%0)\n"
-		"	.long 3b - .\n"
-		"	lda $31,5b-3b(%0)\n"
-		"	.long 4b - .\n"
-		"	lda $31,5b-4b(%0)\n"
-		".previous"
+		EXC(1b,5b,%2,%0)
+		EXC(2b,5b,%1,%0)
+		EXC(3b,5b,$31,%0)
+		EXC(4b,5b,$31,%0)
 			: "=r"(error), "=&r"(tmp1), "=&r"(tmp2),
 			  "=&r"(tmp3), "=&r"(tmp4)
 			: "r"(va), "r"(una_reg(reg)), "0"(0));
@@ -611,16 +587,10 @@ do_entUna(void * va, unsigned long opcode, unsigned long reg,
 		"3:	stq_u %2,7(%5)\n"
 		"4:	stq_u %1,0(%5)\n"
 		"5:\n"
-		".section __ex_table,\"a\"\n\t"
-		"	.long 1b - .\n"
-		"	lda %2,5b-1b(%0)\n"
-		"	.long 2b - .\n"
-		"	lda %1,5b-2b(%0)\n"
-		"	.long 3b - .\n"
-		"	lda $31,5b-3b(%0)\n"
-		"	.long 4b - .\n"
-		"	lda $31,5b-4b(%0)\n"
-		".previous"
+		EXC(1b,5b,%2,%0)
+		EXC(2b,5b,%1,%0)
+		EXC(3b,5b,$31,%0)
+		EXC(4b,5b,$31,%0)
 			: "=r"(error), "=&r"(tmp1), "=&r"(tmp2),
 			  "=&r"(tmp3), "=&r"(tmp4)
 			: "r"(va), "r"(una_reg(reg)), "0"(0));
@@ -802,7 +772,7 @@ do_entUnaUser(void __user * va, unsigned long opcode,
 	/* Don't bother reading ds in the access check since we already
 	   know that this came from the user.  Also rely on the fact that
 	   the page at TASK_SIZE is unmapped and so can't be touched anyway. */
-	if (!__access_ok((unsigned long)va, 0, USER_DS))
+	if ((unsigned long)va >= TASK_SIZE)
 		goto give_sigsegv;
 
 	++unaligned[1].count;
@@ -835,12 +805,8 @@ do_entUnaUser(void __user * va, unsigned long opcode,
 		"	extwl %1,%3,%1\n"
 		"	extwh %2,%3,%2\n"
 		"3:\n"
-		".section __ex_table,\"a\"\n"
-		"	.long 1b - .\n"
-		"	lda %1,3b-1b(%0)\n"
-		"	.long 2b - .\n"
-		"	lda %2,3b-2b(%0)\n"
-		".previous"
+		EXC(1b,3b,%1,%0)
+		EXC(2b,3b,%2,%0)
 			: "=r"(error), "=&r"(tmp1), "=&r"(tmp2)
 			: "r"(va), "0"(0));
 		if (error)
@@ -855,12 +821,8 @@ do_entUnaUser(void __user * va, unsigned long opcode,
 		"	extll %1,%3,%1\n"
 		"	extlh %2,%3,%2\n"
 		"3:\n"
-		".section __ex_table,\"a\"\n"
-		"	.long 1b - .\n"
-		"	lda %1,3b-1b(%0)\n"
-		"	.long 2b - .\n"
-		"	lda %2,3b-2b(%0)\n"
-		".previous"
+		EXC(1b,3b,%1,%0)
+		EXC(2b,3b,%2,%0)
 			: "=r"(error), "=&r"(tmp1), "=&r"(tmp2)
 			: "r"(va), "0"(0));
 		if (error)
@@ -875,12 +837,8 @@ do_entUnaUser(void __user * va, unsigned long opcode,
 		"	extql %1,%3,%1\n"
 		"	extqh %2,%3,%2\n"
 		"3:\n"
-		".section __ex_table,\"a\"\n"
-		"	.long 1b - .\n"
-		"	lda %1,3b-1b(%0)\n"
-		"	.long 2b - .\n"
-		"	lda %2,3b-2b(%0)\n"
-		".previous"
+		EXC(1b,3b,%1,%0)
+		EXC(2b,3b,%2,%0)
 			: "=r"(error), "=&r"(tmp1), "=&r"(tmp2)
 			: "r"(va), "0"(0));
 		if (error)
@@ -895,12 +853,8 @@ do_entUnaUser(void __user * va, unsigned long opcode,
 		"	extll %1,%3,%1\n"
 		"	extlh %2,%3,%2\n"
 		"3:\n"
-		".section __ex_table,\"a\"\n"
-		"	.long 1b - .\n"
-		"	lda %1,3b-1b(%0)\n"
-		"	.long 2b - .\n"
-		"	lda %2,3b-2b(%0)\n"
-		".previous"
+		EXC(1b,3b,%1,%0)
+		EXC(2b,3b,%2,%0)
 			: "=r"(error), "=&r"(tmp1), "=&r"(tmp2)
 			: "r"(va), "0"(0));
 		if (error)
@@ -915,12 +869,8 @@ do_entUnaUser(void __user * va, unsigned long opcode,
 		"	extql %1,%3,%1\n"
 		"	extqh %2,%3,%2\n"
 		"3:\n"
-		".section __ex_table,\"a\"\n"
-		"	.long 1b - .\n"
-		"	lda %1,3b-1b(%0)\n"
-		"	.long 2b - .\n"
-		"	lda %2,3b-2b(%0)\n"
-		".previous"
+		EXC(1b,3b,%1,%0)
+		EXC(2b,3b,%2,%0)
 			: "=r"(error), "=&r"(tmp1), "=&r"(tmp2)
 			: "r"(va), "0"(0));
 		if (error)
@@ -944,16 +894,10 @@ do_entUnaUser(void __user * va, unsigned long opcode,
 		"3:	stq_u %2,1(%5)\n"
 		"4:	stq_u %1,0(%5)\n"
 		"5:\n"
-		".section __ex_table,\"a\"\n"
-		"	.long 1b - .\n"
-		"	lda %2,5b-1b(%0)\n"
-		"	.long 2b - .\n"
-		"	lda %1,5b-2b(%0)\n"
-		"	.long 3b - .\n"
-		"	lda $31,5b-3b(%0)\n"
-		"	.long 4b - .\n"
-		"	lda $31,5b-4b(%0)\n"
-		".previous"
+		EXC(1b,5b,%2,%0)
+		EXC(2b,5b,%1,%0)
+		EXC(3b,5b,$31,%0)
+		EXC(4b,5b,$31,%0)
 			: "=r"(error), "=&r"(tmp1), "=&r"(tmp2),
 			  "=&r"(tmp3), "=&r"(tmp4)
 			: "r"(va), "r"(*reg_addr), "0"(0));
@@ -978,16 +922,10 @@ do_entUnaUser(void __user * va, unsigned long opcode,
 		"3:	stq_u %2,3(%5)\n"
 		"4:	stq_u %1,0(%5)\n"
 		"5:\n"
-		".section __ex_table,\"a\"\n"
-		"	.long 1b - .\n"
-		"	lda %2,5b-1b(%0)\n"
-		"	.long 2b - .\n"
-		"	lda %1,5b-2b(%0)\n"
-		"	.long 3b - .\n"
-		"	lda $31,5b-3b(%0)\n"
-		"	.long 4b - .\n"
-		"	lda $31,5b-4b(%0)\n"
-		".previous"
+		EXC(1b,5b,%2,%0)
+		EXC(2b,5b,%1,%0)
+		EXC(3b,5b,$31,%0)
+		EXC(4b,5b,$31,%0)
 			: "=r"(error), "=&r"(tmp1), "=&r"(tmp2),
 			  "=&r"(tmp3), "=&r"(tmp4)
 			: "r"(va), "r"(*reg_addr), "0"(0));
@@ -1012,16 +950,10 @@ do_entUnaUser(void __user * va, unsigned long opcode,
 		"3:	stq_u %2,7(%5)\n"
 		"4:	stq_u %1,0(%5)\n"
 		"5:\n"
-		".section __ex_table,\"a\"\n\t"
-		"	.long 1b - .\n"
-		"	lda %2,5b-1b(%0)\n"
-		"	.long 2b - .\n"
-		"	lda %1,5b-2b(%0)\n"
-		"	.long 3b - .\n"
-		"	lda $31,5b-3b(%0)\n"
-		"	.long 4b - .\n"
-		"	lda $31,5b-4b(%0)\n"
-		".previous"
+		EXC(1b,5b,%2,%0)
+		EXC(2b,5b,%1,%0)
+		EXC(3b,5b,$31,%0)
+		EXC(4b,5b,$31,%0)
 			: "=r"(error), "=&r"(tmp1), "=&r"(tmp2),
 			  "=&r"(tmp3), "=&r"(tmp4)
 			: "r"(va), "r"(*reg_addr), "0"(0));
@@ -1047,7 +979,7 @@ do_entUnaUser(void __user * va, unsigned long opcode,
 	/* We need to replicate some of the logic in mm/fault.c,
 	   since we don't have access to the fault code in the
 	   exception handling return path.  */
-	if (!__access_ok((unsigned long)va, 0, USER_DS))
+	if ((unsigned long)va >= TASK_SIZE)
 		info.si_code = SEGV_ACCERR;
 	else {
 		struct mm_struct *mm = current->mm;
diff --git a/arch/alpha/lib/clear_user.S b/arch/alpha/lib/clear_user.S
index bf5b931..006f469f 100644
--- a/arch/alpha/lib/clear_user.S
+++ b/arch/alpha/lib/clear_user.S
@@ -8,21 +8,6 @@
  * right "bytes left to zero" value (and that it is updated only _after_
  * a successful copy).  There is also some rather minor exception setup
  * stuff.
- *
- * NOTE! This is not directly C-callable, because the calling semantics
- * are different:
- *
- * Inputs:
- *	length in $0
- *	destination address in $6
- *	exception pointer in $7
- *	return address in $28 (exceptions expect it there)
- *
- * Outputs:
- *	bytes left to copy in $0
- *
- * Clobbers:
- *	$1,$2,$3,$4,$5,$6
  */
 #include <asm/export.h>
 
@@ -38,62 +23,63 @@
 	.set noreorder
 	.align 4
 
-	.globl __do_clear_user
-	.ent __do_clear_user
-	.frame	$30, 0, $28
+	.globl __clear_user
+	.ent __clear_user
+	.frame	$30, 0, $26
 	.prologue 0
 
 $loop:
 	and	$1, 3, $4	# e0    :
 	beq	$4, 1f		# .. e1 :
 
-0:	EX( stq_u $31, 0($6) )	# e0    : zero one word
+0:	EX( stq_u $31, 0($16) )	# e0    : zero one word
 	subq	$0, 8, $0	# .. e1 :
 	subq	$4, 1, $4	# e0    :
-	addq	$6, 8, $6	# .. e1 :
+	addq	$16, 8, $16	# .. e1 :
 	bne	$4, 0b		# e1    :
 	unop			#       :
 
 1:	bic	$1, 3, $1	# e0    :
 	beq	$1, $tail	# .. e1 :
 
-2:	EX( stq_u $31, 0($6) )	# e0    : zero four words
+2:	EX( stq_u $31, 0($16) )	# e0    : zero four words
 	subq	$0, 8, $0	# .. e1 :
-	EX( stq_u $31, 8($6) )	# e0    :
+	EX( stq_u $31, 8($16) )	# e0    :
 	subq	$0, 8, $0	# .. e1 :
-	EX( stq_u $31, 16($6) )	# e0    :
+	EX( stq_u $31, 16($16) )	# e0    :
 	subq	$0, 8, $0	# .. e1 :
-	EX( stq_u $31, 24($6) )	# e0    :
+	EX( stq_u $31, 24($16) )	# e0    :
 	subq	$0, 8, $0	# .. e1 :
 	subq	$1, 4, $1	# e0    :
-	addq	$6, 32, $6	# .. e1 :
+	addq	$16, 32, $16	# .. e1 :
 	bne	$1, 2b		# e1    :
 
 $tail:
 	bne	$2, 1f		# e1    : is there a tail to do?
-	ret	$31, ($28), 1	# .. e1 :
+	ret	$31, ($26), 1	# .. e1 :
 
-1:	EX( ldq_u $5, 0($6) )	# e0    :
+1:	EX( ldq_u $5, 0($16) )	# e0    :
 	clr	$0		# .. e1 :
 	nop			# e1    :
 	mskqh	$5, $0, $5	# e0    :
-	EX( stq_u $5, 0($6) )	# e0    :
-	ret	$31, ($28), 1	# .. e1 :
+	EX( stq_u $5, 0($16) )	# e0    :
+	ret	$31, ($26), 1	# .. e1 :
 
-__do_clear_user:
-	and	$6, 7, $4	# e0    : find dest misalignment
+__clear_user:
+	and	$17, $17, $0
+	and	$16, 7, $4	# e0    : find dest misalignment
 	beq	$0, $zerolength # .. e1 :
 	addq	$0, $4, $1	# e0    : bias counter
 	and	$1, 7, $2	# e1    : number of bytes in tail
 	srl	$1, 3, $1	# e0    :
 	beq	$4, $loop	# .. e1 :
 
-	EX( ldq_u $5, 0($6) )	# e0    : load dst word to mask back in
+	EX( ldq_u $5, 0($16) )	# e0    : load dst word to mask back in
 	beq	$1, $oneword	# .. e1 : sub-word store?
 
-	mskql	$5, $6, $5	# e0    : take care of misaligned head
-	addq	$6, 8, $6	# .. e1 :
-	EX( stq_u $5, -8($6) )	# e0    :
+	mskql	$5, $16, $5	# e0    : take care of misaligned head
+	addq	$16, 8, $16	# .. e1 :
+	EX( stq_u $5, -8($16) )	# e0    :
 	addq	$0, $4, $0	# .. e1 : bytes left -= 8 - misalignment
 	subq	$1, 1, $1	# e0    :
 	subq	$0, 8, $0	# .. e1 :
@@ -101,15 +87,15 @@
 	unop			#       :
 
 $oneword:
-	mskql	$5, $6, $4	# e0    :
+	mskql	$5, $16, $4	# e0    :
 	mskqh	$5, $2, $5	# e0    :
 	or	$5, $4, $5	# e1    :
-	EX( stq_u $5, 0($6) )	# e0    :
+	EX( stq_u $5, 0($16) )	# e0    :
 	clr	$0		# .. e1 :
 
 $zerolength:
 $exception:
-	ret	$31, ($28), 1	# .. e1 :
+	ret	$31, ($26), 1	# .. e1 :
 
-	.end __do_clear_user
-	EXPORT_SYMBOL(__do_clear_user)
+	.end __clear_user
+	EXPORT_SYMBOL(__clear_user)
diff --git a/arch/alpha/lib/copy_user.S b/arch/alpha/lib/copy_user.S
index 509f62b..159f1b7 100644
--- a/arch/alpha/lib/copy_user.S
+++ b/arch/alpha/lib/copy_user.S
@@ -9,21 +9,6 @@
  * contains the right "bytes left to copy" value (and that it is updated
  * only _after_ a successful copy). There is also some rather minor
  * exception setup stuff..
- *
- * NOTE! This is not directly C-callable, because the calling semantics are
- * different:
- *
- * Inputs:
- *	length in $0
- *	destination address in $6
- *	source address in $7
- *	return address in $28
- *
- * Outputs:
- *	bytes left to copy in $0
- *
- * Clobbers:
- *	$1,$2,$3,$4,$5,$6,$7
  */
 
 #include <asm/export.h>
@@ -49,58 +34,59 @@
 	.ent __copy_user
 __copy_user:
 	.prologue 0
-	and $6,7,$3
+	and $18,$18,$0
+	and $16,7,$3
 	beq $0,$35
 	beq $3,$36
 	subq $3,8,$3
 	.align 4
 $37:
-	EXI( ldq_u $1,0($7) )
-	EXO( ldq_u $2,0($6) )
-	extbl $1,$7,$1
-	mskbl $2,$6,$2
-	insbl $1,$6,$1
+	EXI( ldq_u $1,0($17) )
+	EXO( ldq_u $2,0($16) )
+	extbl $1,$17,$1
+	mskbl $2,$16,$2
+	insbl $1,$16,$1
 	addq $3,1,$3
 	bis $1,$2,$1
-	EXO( stq_u $1,0($6) )
+	EXO( stq_u $1,0($16) )
 	subq $0,1,$0
-	addq $6,1,$6
-	addq $7,1,$7
+	addq $16,1,$16
+	addq $17,1,$17
 	beq $0,$41
 	bne $3,$37
 $36:
-	and $7,7,$1
+	and $17,7,$1
 	bic $0,7,$4
 	beq $1,$43
 	beq $4,$48
-	EXI( ldq_u $3,0($7) )
+	EXI( ldq_u $3,0($17) )
 	.align 4
 $50:
-	EXI( ldq_u $2,8($7) )
+	EXI( ldq_u $2,8($17) )
 	subq $4,8,$4
-	extql $3,$7,$3
-	extqh $2,$7,$1
+	extql $3,$17,$3
+	extqh $2,$17,$1
 	bis $3,$1,$1
-	EXO( stq $1,0($6) )
-	addq $7,8,$7
+	EXO( stq $1,0($16) )
+	addq $17,8,$17
 	subq $0,8,$0
-	addq $6,8,$6
+	addq $16,8,$16
 	bis $2,$2,$3
 	bne $4,$50
 $48:
 	beq $0,$41
 	.align 4
 $57:
-	EXI( ldq_u $1,0($7) )
-	EXO( ldq_u $2,0($6) )
-	extbl $1,$7,$1
-	mskbl $2,$6,$2
-	insbl $1,$6,$1
+	EXI( ldq_u $1,0($17) )
+	EXO( ldq_u $2,0($16) )
+	extbl $1,$17,$1
+	mskbl $2,$16,$2
+	insbl $1,$16,$1
 	bis $1,$2,$1
-	EXO( stq_u $1,0($6) )
+	EXO( stq_u $1,0($16) )
 	subq $0,1,$0
-	addq $6,1,$6
-	addq $7,1,$7
+	addq $16,1,$16
+	addq $17,1,$17
 	bne $0,$57
 	br $31,$41
 	.align 4
@@ -108,27 +94,27 @@
 	beq $4,$65
 	.align 4
 $66:
-	EXI( ldq $1,0($7) )
+	EXI( ldq $1,0($17) )
 	subq $4,8,$4
-	EXO( stq $1,0($6) )
-	addq $7,8,$7
+	EXO( stq $1,0($16) )
+	addq $17,8,$17
 	subq $0,8,$0
-	addq $6,8,$6
+	addq $16,8,$16
 	bne $4,$66
 $65:
 	beq $0,$41
-	EXI( ldq $2,0($7) )
-	EXO( ldq $1,0($6) )
+	EXI( ldq $2,0($17) )
+	EXO( ldq $1,0($16) )
 	mskql $2,$0,$2
 	mskqh $1,$0,$1
 	bis $2,$1,$2
-	EXO( stq $2,0($6) )
+	EXO( stq $2,0($16) )
 	bis $31,$31,$0
 $41:
 $35:
 $exitin:
 $exitout:
-	ret $31,($28),1
+	ret $31,($26),1
 
 	.end __copy_user
 EXPORT_SYMBOL(__copy_user)
diff --git a/arch/alpha/lib/csum_partial_copy.c b/arch/alpha/lib/csum_partial_copy.c
index 5dfb797..ab42afb 100644
--- a/arch/alpha/lib/csum_partial_copy.c
+++ b/arch/alpha/lib/csum_partial_copy.c
@@ -45,10 +45,7 @@ __asm__ __volatile__("insqh %1,%2,%0":"=r" (z):"r" (x),"r" (y))
 	__asm__ __volatile__(				\
 	"1:	ldq_u %0,%2\n"				\
 	"2:\n"						\
-	".section __ex_table,\"a\"\n"			\
-	"	.long 1b - .\n"				\
-	"	lda %0,2b-1b(%1)\n"			\
-	".previous"					\
+	EXC(1b,2b,%0,%1)				\
 		: "=r"(x), "=r"(__guu_err)		\
 		: "m"(__m(ptr)), "1"(0));		\
 	__guu_err;					\
@@ -60,10 +57,7 @@ __asm__ __volatile__("insqh %1,%2,%0":"=r" (z):"r" (x),"r" (y))
 	__asm__ __volatile__(				\
 	"1:	stq_u %2,%1\n"				\
 	"2:\n"						\
-	".section __ex_table,\"a\"\n"			\
-	"	.long 1b - ."				\
-	"	lda $31,2b-1b(%0)\n"			\
-	".previous"					\
+	EXC(1b,2b,$31,%0)				\
 		: "=r"(__puu_err)			\
 		: "m"(__m(addr)), "rJ"(x), "0"(0));	\
 	__puu_err;					\
diff --git a/arch/alpha/lib/ev6-clear_user.S b/arch/alpha/lib/ev6-clear_user.S
index 05bef6b..e179e47 100644
--- a/arch/alpha/lib/ev6-clear_user.S
+++ b/arch/alpha/lib/ev6-clear_user.S
@@ -9,21 +9,6 @@
  * a successful copy).  There is also some rather minor exception setup
  * stuff.
  *
- * NOTE! This is not directly C-callable, because the calling semantics
- * are different:
- *
- * Inputs:
- *	length in $0
- *	destination address in $6
- *	exception pointer in $7
- *	return address in $28 (exceptions expect it there)
- *
- * Outputs:
- *	bytes left to copy in $0
- *
- * Clobbers:
- *	$1,$2,$3,$4,$5,$6
- *
  * Much of the information about 21264 scheduling/coding comes from:
  *	Compiler Writer's Guide for the Alpha 21264
  *	abbreviated as 'CWG' in other comments here
@@ -56,14 +41,15 @@
 	.set noreorder
 	.align 4
 
-	.globl __do_clear_user
-	.ent __do_clear_user
-	.frame	$30, 0, $28
+	.globl __clear_user
+	.ent __clear_user
+	.frame	$30, 0, $26
 	.prologue 0
 
 				# Pipeline info : Slotting & Comments
-__do_clear_user:
-	and	$6, 7, $4	# .. E  .. ..	: find dest head misalignment
+__clear_user:
+	and	$17, $17, $0
+	and	$16, 7, $4	# .. E  .. ..	: find dest head misalignment
 	beq	$0, $zerolength # U  .. .. ..	:  U L U L
 
 	addq	$0, $4, $1	# .. .. .. E	: bias counter
@@ -75,14 +61,14 @@
 
 /*
  * Head is not aligned.  Write (8 - $4) bytes to head of destination
- * This means $6 is known to be misaligned
+ * This means $16 is known to be misaligned
  */
-	EX( ldq_u $5, 0($6) )	# .. .. .. L	: load dst word to mask back in
+	EX( ldq_u $5, 0($16) )	# .. .. .. L	: load dst word to mask back in
 	beq	$1, $onebyte	# .. .. U  ..	: sub-word store?
-	mskql	$5, $6, $5	# .. U  .. ..	: take care of misaligned head
-	addq	$6, 8, $6	# E  .. .. .. 	: L U U L
+	mskql	$5, $16, $5	# .. U  .. ..	: take care of misaligned head
+	addq	$16, 8, $16	# E  .. .. .. 	: L U U L
 
-	EX( stq_u $5, -8($6) )	# .. .. .. L	:
+	EX( stq_u $5, -8($16) )	# .. .. .. L	:
 	subq	$1, 1, $1	# .. .. E  ..	:
 	addq	$0, $4, $0	# .. E  .. ..	: bytes left -= 8 - misalignment
 	subq	$0, 8, $0	# E  .. .. ..	: U L U L
@@ -93,11 +79,11 @@
  * values upon initial entry to the loop
  * $1 is number of quadwords to clear (zero is a valid value)
  * $2 is number of trailing bytes (0..7) ($2 never used...)
- * $6 is known to be aligned 0mod8
+ * $16 is known to be aligned 0mod8
  */
 $headalign:
 	subq	$1, 16, $4	# .. .. .. E	: If < 16, we can not use the huge loop
-	and	$6, 0x3f, $2	# .. .. E  ..	: Forward work for huge loop
+	and	$16, 0x3f, $2	# .. .. E  ..	: Forward work for huge loop
 	subq	$2, 0x40, $3	# .. E  .. ..	: bias counter (huge loop)
 	blt	$4, $trailquad	# U  .. .. ..	: U L U L
 
@@ -114,21 +100,21 @@
 	beq	$3, $bigalign	# U  .. .. ..	: U L U L : Aligned 0mod64
 
 $alignmod64:
-	EX( stq_u $31, 0($6) )	# .. .. .. L
+	EX( stq_u $31, 0($16) )	# .. .. .. L
 	addq	$3, 8, $3	# .. .. E  ..
 	subq	$0, 8, $0	# .. E  .. ..
 	nop			# E  .. .. ..	: U L U L
 
 	nop			# .. .. .. E
 	subq	$1, 1, $1	# .. .. E  ..
-	addq	$6, 8, $6	# .. E  .. ..
+	addq	$16, 8, $16	# .. E  .. ..
 	blt	$3, $alignmod64	# U  .. .. ..	: U L U L
 
 $bigalign:
 /*
  * $0 is the number of bytes left
  * $1 is the number of quads left
- * $6 is aligned 0mod64
+ * $16 is aligned 0mod64
  * we know that we'll be taking a minimum of one trip through
  * CWG Section 3.7.6: do not expect a sustained store rate of > 1/cycle
  * We are _not_ going to update $0 after every single store.  That
@@ -145,39 +131,39 @@
 	nop			# E :
 	nop			# E :
 	nop			# E :
-	bis	$6,$6,$3	# E : U L U L : Initial wh64 address is dest
+	bis	$16,$16,$3	# E : U L U L : Initial wh64 address is dest
 	/* This might actually help for the current trip... */
 
 $do_wh64:
 	wh64	($3)		# .. .. .. L1	: memory subsystem hint
 	subq	$1, 16, $4	# .. .. E  ..	: Forward calculation - repeat the loop?
-	EX( stq_u $31, 0($6) )	# .. L  .. ..
+	EX( stq_u $31, 0($16) )	# .. L  .. ..
 	subq	$0, 8, $0	# E  .. .. ..	: U L U L
 
-	addq	$6, 128, $3	# E : Target address of wh64
-	EX( stq_u $31, 8($6) )	# L :
-	EX( stq_u $31, 16($6) )	# L :
+	addq	$16, 128, $3	# E : Target address of wh64
+	EX( stq_u $31, 8($16) )	# L :
+	EX( stq_u $31, 16($16) )	# L :
 	subq	$0, 16, $0	# E : U L L U
 
 	nop			# E :
-	EX( stq_u $31, 24($6) )	# L :
-	EX( stq_u $31, 32($6) )	# L :
+	EX( stq_u $31, 24($16) )	# L :
+	EX( stq_u $31, 32($16) )	# L :
 	subq	$0, 168, $5	# E : U L L U : two trips through the loop left?
 	/* 168 = 192 - 24, since we've already completed some stores */
 
 	subq	$0, 16, $0	# E :
-	EX( stq_u $31, 40($6) )	# L :
-	EX( stq_u $31, 48($6) )	# L :
-	cmovlt	$5, $6, $3	# E : U L L U : Latency 2, extra mapping cycle
+	EX( stq_u $31, 40($16) )	# L :
+	EX( stq_u $31, 48($16) )	# L :
+	cmovlt	$5, $16, $3	# E : U L L U : Latency 2, extra mapping cycle
 
 	subq	$1, 8, $1	# E :
 	subq	$0, 16, $0	# E :
-	EX( stq_u $31, 56($6) )	# L :
+	EX( stq_u $31, 56($16) )	# L :
 	nop			# E : U L U L
 
 	nop			# E :
 	subq	$0, 8, $0	# E :
-	addq	$6, 64, $6	# E :
+	addq	$16, 64, $16	# E :
 	bge	$4, $do_wh64	# U : U L U L
 
 $trailquad:
@@ -190,14 +176,14 @@
 	beq	$1, $trailbytes	# U  .. .. ..	: U L U L : Only 0..7 bytes to go
 
 $onequad:
-	EX( stq_u $31, 0($6) )	# .. .. .. L
+	EX( stq_u $31, 0($16) )	# .. .. .. L
 	subq	$1, 1, $1	# .. .. E  ..
 	subq	$0, 8, $0	# .. E  .. ..
 	nop			# E  .. .. ..	: U L U L
 
 	nop			# .. .. .. E
 	nop			# .. .. E  ..
-	addq	$6, 8, $6	# .. E  .. ..
+	addq	$16, 8, $16	# .. E  .. ..
 	bgt	$1, $onequad	# U  .. .. ..	: U L U L
 
 	# We have an unknown number of bytes left to go.
@@ -211,9 +197,9 @@
 	# so we will use $0 as the loop counter
 	# We know for a fact that $0 > 0 zero due to previous context
 $onebyte:
-	EX( stb $31, 0($6) )	# .. .. .. L
+	EX( stb $31, 0($16) )	# .. .. .. L
 	subq	$0, 1, $0	# .. .. E  ..	:
-	addq	$6, 1, $6	# .. E  .. ..	:
+	addq	$16, 1, $16	# .. E  .. ..	:
 	bgt	$0, $onebyte	# U  .. .. ..	: U L U L
 
 $zerolength:
@@ -221,6 +207,6 @@
 	nop			# .. .. .. E	:
 	nop			# .. .. E  ..	:
 	nop			# .. E  .. ..	:
-	ret	$31, ($28), 1	# L0 .. .. ..	: L U L U
-	.end __do_clear_user
-	EXPORT_SYMBOL(__do_clear_user)
+	ret	$31, ($26), 1	# L0 .. .. ..	: L U L U
+	.end __clear_user
+	EXPORT_SYMBOL(__clear_user)
diff --git a/arch/alpha/lib/ev6-copy_user.S b/arch/alpha/lib/ev6-copy_user.S
index be720b5..35e6710 100644
--- a/arch/alpha/lib/ev6-copy_user.S
+++ b/arch/alpha/lib/ev6-copy_user.S
@@ -12,21 +12,6 @@
  * only _after_ a successful copy). There is also some rather minor
  * exception setup stuff..
  *
- * NOTE! This is not directly C-callable, because the calling semantics are
- * different:
- *
- * Inputs:
- *	length in $0
- *	destination address in $6
- *	source address in $7
- *	return address in $28
- *
- * Outputs:
- *	bytes left to copy in $0
- *
- * Clobbers:
- *	$1,$2,$3,$4,$5,$6,$7
- *
  * Much of the information about 21264 scheduling/coding comes from:
  *	Compiler Writer's Guide for the Alpha 21264
  *	abbreviated as 'CWG' in other comments here
@@ -60,10 +45,11 @@
 				# Pipeline info: Slotting & Comments
 __copy_user:
 	.prologue 0
-	subq $0, 32, $1		# .. E  .. ..	: Is this going to be a small copy?
+	andq $18, $18, $0
+	subq $18, 32, $1	# .. E  .. ..	: Is this going to be a small copy?
 	beq $0, $zerolength	# U  .. .. ..	: U L U L
 
-	and $6,7,$3		# .. .. .. E	: is leading dest misalignment
+	and $16,7,$3		# .. .. .. E	: is leading dest misalignment
 	ble $1, $onebyteloop	# .. .. U  ..	: 1st branch : small amount of data
 	beq $3, $destaligned	# .. U  .. ..	: 2nd (one cycle fetcher stall)
 	subq $3, 8, $3		# E  .. .. ..	: L U U L : trip counter
@@ -73,17 +59,17 @@
  * We know we have at least one trip through this loop
  */
 $aligndest:
-	EXI( ldbu $1,0($7) )	# .. .. .. L	: Keep loads separate from stores
-	addq $6,1,$6		# .. .. E  ..	: Section 3.8 in the CWG
+	EXI( ldbu $1,0($17) )	# .. .. .. L	: Keep loads separate from stores
+	addq $16,1,$16		# .. .. E  ..	: Section 3.8 in the CWG
 	addq $3,1,$3		# .. E  .. ..	:
 	nop			# E  .. .. ..	: U L U L
 
 /*
- * the -1 is to compensate for the inc($6) done in a previous quadpack
+ * the -1 is to compensate for the inc($16) done in a previous quadpack
  * which allows us zero dependencies within either quadpack in the loop
  */
-	EXO( stb $1,-1($6) )	# .. .. .. L	:
-	addq $7,1,$7		# .. .. E  ..	: Section 3.8 in the CWG
+	EXO( stb $1,-1($16) )	# .. .. .. L	:
+	addq $17,1,$17		# .. .. E  ..	: Section 3.8 in the CWG
 	subq $0,1,$0		# .. E  .. ..	:
 	bne $3, $aligndest	# U  .. .. ..	: U L U L
 
@@ -92,29 +78,29 @@
  * If we arrived via branch, we have a minimum of 32 bytes
  */
 $destaligned:
-	and $7,7,$1		# .. .. .. E	: Check _current_ source alignment
+	and $17,7,$1		# .. .. .. E	: Check _current_ source alignment
 	bic $0,7,$4		# .. .. E  ..	: number bytes as a quadword loop
-	EXI( ldq_u $3,0($7) )	# .. L  .. ..	: Forward fetch for fallthrough code
+	EXI( ldq_u $3,0($17) )	# .. L  .. ..	: Forward fetch for fallthrough code
 	beq $1,$quadaligned	# U  .. .. ..	: U L U L
 
 /*
- * In the worst case, we've just executed an ldq_u here from 0($7)
+ * In the worst case, we've just executed an ldq_u here from 0($17)
  * and we'll repeat it once if we take the branch
  */
 
 /* Misaligned quadword loop - not unrolled.  Leave it that way. */
 $misquad:
-	EXI( ldq_u $2,8($7) )	# .. .. .. L	:
+	EXI( ldq_u $2,8($17) )	# .. .. .. L	:
 	subq $4,8,$4		# .. .. E  ..	:
-	extql $3,$7,$3		# .. U  .. ..	:
-	extqh $2,$7,$1		# U  .. .. ..	: U U L L
+	extql $3,$17,$3		# .. U  .. ..	:
+	extqh $2,$17,$1		# U  .. .. ..	: U U L L
 
 	bis $3,$1,$1		# .. .. .. E	:
-	EXO( stq $1,0($6) )	# .. .. L  ..	:
-	addq $7,8,$7		# .. E  .. ..	:
+	EXO( stq $1,0($16) )	# .. .. L  ..	:
+	addq $17,8,$17		# .. E  .. ..	:
 	subq $0,8,$0		# E  .. .. ..	: U L L U
 
-	addq $6,8,$6		# .. .. .. E	:
+	addq $16,8,$16		# .. .. .. E	:
 	bis $2,$2,$3		# .. .. E  ..	:
 	nop			# .. E  .. ..	:
 	bne $4,$misquad		# U  .. .. ..	: U L U L
@@ -125,8 +111,8 @@
 	beq $0,$zerolength	# U  .. .. ..	: U L U L
 
 /* We know we have at least one trip through the byte loop */
-	EXI ( ldbu $2,0($7) )	# .. .. .. L	: No loads in the same quad
-	addq $6,1,$6		# .. .. E  ..	: as the store (Section 3.8 in CWG)
+	EXI ( ldbu $2,0($17) )	# .. .. .. L	: No loads in the same quad
+	addq $16,1,$16		# .. .. E  ..	: as the store (Section 3.8 in CWG)
 	nop			# .. E  .. ..	:
 	br $31, $dirtyentry	# L0 .. .. ..	: L U U L
 /* Do the trailing byte loop load, then hop into the store part of the loop */
@@ -136,8 +122,8 @@
  * Based upon the usage context, it's worth the effort to unroll this loop
  * $0 - number of bytes to be moved
  * $4 - number of bytes to move as quadwords
- * $6 is current destination address
- * $7 is current source address
+ * $16 is current destination address
+ * $17 is current source address
  */
 $quadaligned:
 	subq	$4, 32, $2	# .. .. .. E	: do not unroll for small stuff
@@ -155,29 +141,29 @@
  * instruction memory hint instruction).
  */
 $unroll4:
-	EXI( ldq $1,0($7) )	# .. .. .. L
-	EXI( ldq $2,8($7) )	# .. .. L  ..
+	EXI( ldq $1,0($17) )	# .. .. .. L
+	EXI( ldq $2,8($17) )	# .. .. L  ..
 	subq	$4,32,$4	# .. E  .. ..
 	nop			# E  .. .. ..	: U U L L
 
-	addq	$7,16,$7	# .. .. .. E
-	EXO( stq $1,0($6) )	# .. .. L  ..
-	EXO( stq $2,8($6) )	# .. L  .. ..
+	addq	$17,16,$17	# .. .. .. E
+	EXO( stq $1,0($16) )	# .. .. L  ..
+	EXO( stq $2,8($16) )	# .. L  .. ..
 	subq	$0,16,$0	# E  .. .. ..	: U L L U
 
-	addq	$6,16,$6	# .. .. .. E
-	EXI( ldq $1,0($7) )	# .. .. L  ..
-	EXI( ldq $2,8($7) )	# .. L  .. ..
+	addq	$16,16,$16	# .. .. .. E
+	EXI( ldq $1,0($17) )	# .. .. L  ..
+	EXI( ldq $2,8($17) )	# .. L  .. ..
 	subq	$4, 32, $3	# E  .. .. ..	: U U L L : is there enough for another trip?
 
-	EXO( stq $1,0($6) )	# .. .. .. L
-	EXO( stq $2,8($6) )	# .. .. L  ..
+	EXO( stq $1,0($16) )	# .. .. .. L
+	EXO( stq $2,8($16) )	# .. .. L  ..
 	subq	$0,16,$0	# .. E  .. ..
-	addq	$7,16,$7	# E  .. .. ..	: U L L U
+	addq	$17,16,$17	# E  .. .. ..	: U L L U
 
 	nop			# .. .. .. E
 	nop			# .. .. E  ..
-	addq	$6,16,$6	# .. E  .. ..
+	addq	$16,16,$16	# .. E  .. ..
 	bgt	$3,$unroll4	# U  .. .. ..	: U L U L
 
 	nop
@@ -186,14 +172,14 @@
 	beq	$4, $noquads
 
 $onequad:
-	EXI( ldq $1,0($7) )
+	EXI( ldq $1,0($17) )
 	subq	$4,8,$4
-	addq	$7,8,$7
+	addq	$17,8,$17
 	nop
 
-	EXO( stq $1,0($6) )
+	EXO( stq $1,0($16) )
 	subq	$0,8,$0
-	addq	$6,8,$6
+	addq	$16,8,$16
 	bne	$4,$onequad
 
 $noquads:
@@ -207,23 +193,23 @@
  * There's no point in doing a lot of complex alignment calculations to try to
  * to quadword stuff for a small amount of data.
  *	$0 - remaining number of bytes left to copy
- *	$6 - current dest addr
- *	$7 - current source addr
+ *	$16 - current dest addr
+ *	$17 - current source addr
  */
 
 $onebyteloop:
-	EXI ( ldbu $2,0($7) )	# .. .. .. L	: No loads in the same quad
-	addq $6,1,$6		# .. .. E  ..	: as the store (Section 3.8 in CWG)
+	EXI ( ldbu $2,0($17) )	# .. .. .. L	: No loads in the same quad
+	addq $16,1,$16		# .. .. E  ..	: as the store (Section 3.8 in CWG)
 	nop			# .. E  .. ..	:
 	nop			# E  .. .. ..	: U L U L
 
 $dirtyentry:
 /*
- * the -1 is to compensate for the inc($6) done in a previous quadpack
+ * the -1 is to compensate for the inc($16) done in a previous quadpack
  * which allows us zero dependencies within either quadpack in the loop
  */
-	EXO ( stb $2,-1($6) )	# .. .. .. L	:
-	addq $7,1,$7		# .. .. E  ..	: quadpack as the load
+	EXO ( stb $2,-1($16) )	# .. .. .. L	:
+	addq $17,1,$17		# .. .. E  ..	: quadpack as the load
 	subq $0,1,$0		# .. E  .. ..	: change count _after_ copy
 	bgt $0,$onebyteloop	# U  .. .. ..	: U L U L
 
@@ -233,7 +219,7 @@
 	nop			# .. .. .. E
 	nop			# .. .. E  ..
 	nop			# .. E  .. ..
-	ret $31,($28),1		# L0 .. .. ..	: L U L U
+	ret $31,($26),1		# L0 .. .. ..	: L U L U
 
 	.end __copy_user
 	EXPORT_SYMBOL(__copy_user)
diff --git a/arch/arc/Kconfig b/arch/arc/Kconfig
index c9f30f4..7e213ff 100644
--- a/arch/arc/Kconfig
+++ b/arch/arc/Kconfig
@@ -44,6 +44,7 @@
 	select HAVE_GENERIC_DMA_COHERENT
 	select HAVE_KERNEL_GZIP
 	select HAVE_KERNEL_LZMA
+	select ARCH_HAS_RAW_COPY_USER
 
 config MIGHT_HAVE_PCI
 	bool
diff --git a/arch/arc/include/asm/uaccess.h b/arch/arc/include/asm/uaccess.h
index ffd14e6..6736b7d 100644
--- a/arch/arc/include/asm/uaccess.h
+++ b/arch/arc/include/asm/uaccess.h
@@ -69,6 +69,8 @@
 	__ret;							\
 })
 
+#define __copy_from_user __copy_from_user	/* have asm-generic leave it alone */
+
 /*
  * Returns 0 on success, -EFAULT if not.
  * @ret already contains 0 - given that errors will be less likely
@@ -394,9 +396,6 @@ __arc_copy_from_user(void *to, const void __user *from, unsigned long n)
 	return res;
 }
 
-extern unsigned long slowpath_copy_to_user(void __user *to, const void *from,
-					   unsigned long n);
-
 static inline unsigned long
 __arc_copy_to_user(void __user *to, const void *from, unsigned long n)
 {
@@ -724,8 +723,8 @@ static inline long __arc_strnlen_user(const char __user *s, long n)
 }
 
 #ifndef CONFIG_CC_OPTIMIZE_FOR_SIZE
-#define __copy_from_user(t, f, n)	__arc_copy_from_user(t, f, n)
-#define __copy_to_user(t, f, n)		__arc_copy_to_user(t, f, n)
+#define raw_copy_from_user		__arc_copy_from_user
+#define raw_copy_to_user		__arc_copy_to_user
 #define __clear_user(d, n)		__arc_clear_user(d, n)
 #define __strncpy_from_user(d, s, n)	__arc_strncpy_from_user(d, s, n)
 #define __strnlen_user(s, n)		__arc_strnlen_user(s, n)
@@ -740,8 +739,8 @@ extern long arc_strncpy_from_user_noinline (char *dst, const char __user *src,
 		long count);
 extern long arc_strnlen_user_noinline(const char __user *src, long n);
 
-#define __copy_from_user(t, f, n)	arc_copy_from_user_noinline(t, f, n)
-#define __copy_to_user(t, f, n)		arc_copy_to_user_noinline(t, f, n)
+#define raw_copy_from_user		arc_copy_from_user_noinline
+#define raw_copy_to_user		arc_copy_to_user_noinline
 #define __clear_user(d, n)		arc_clear_user_noinline(d, n)
 #define __strncpy_from_user(d, s, n)	arc_strncpy_from_user_noinline(d, s, n)
 #define __strnlen_user(s, n)		arc_strnlen_user_noinline(s, n)
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 0d4e71b..6fab7f3 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -96,6 +96,7 @@
 	select PERF_USE_VMALLOC
 	select RTC_LIB
 	select SYS_SUPPORTS_APM_EMULATION
+	select ARCH_HAS_RAW_COPY_USER
 	# Above selects are sorted alphabetically; please add new ones
 	# according to that.  Thanks.
 	help
diff --git a/arch/arm/include/asm/Kbuild b/arch/arm/include/asm/Kbuild
index b14e8c7..3a36d99 100644
--- a/arch/arm/include/asm/Kbuild
+++ b/arch/arm/include/asm/Kbuild
@@ -7,6 +7,7 @@
 generic-y += emergency-restart.h
 generic-y += errno.h
 generic-y += exec.h
+generic-y += extable.h
 generic-y += ioctl.h
 generic-y += ipcbuf.h
 generic-y += irq_regs.h
diff --git a/arch/arm/include/asm/uaccess.h b/arch/arm/include/asm/uaccess.h
index b635273..2577405 100644
--- a/arch/arm/include/asm/uaccess.h
+++ b/arch/arm/include/asm/uaccess.h
@@ -24,25 +24,7 @@
 #define __put_user_unaligned __put_user
 #endif
 
-/*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
-
-struct exception_table_entry
-{
-	unsigned long insn, fixup;
-};
-
-extern int fixup_exception(struct pt_regs *regs);
+#include <asm/extable.h>
 
 /*
  * These two functions allow hooking accesses to userspace to increase
@@ -473,7 +455,7 @@ extern unsigned long __must_check
 arm_copy_from_user(void *to, const void __user *from, unsigned long n);
 
 static inline unsigned long __must_check
-__arch_copy_from_user(void *to, const void __user *from, unsigned long n)
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
 {
 	unsigned int __ua_flags;
 
@@ -489,7 +471,7 @@ extern unsigned long __must_check
 __copy_to_user_std(void __user *to, const void *from, unsigned long n);
 
 static inline unsigned long __must_check
-__arch_copy_to_user(void __user *to, const void *from, unsigned long n)
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 {
 #ifndef CONFIG_UACCESS_WITH_MEMCPY
 	unsigned int __ua_flags;
@@ -517,54 +499,22 @@ __clear_user(void __user *addr, unsigned long n)
 }
 
 #else
-#define __arch_copy_from_user(to, from, n)	\
-					(memcpy(to, (void __force *)from, n), 0)
-#define __arch_copy_to_user(to, from, n)	\
-					(memcpy((void __force *)to, from, n), 0)
+static inline unsigned long
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
+{
+	memcpy(to, (const void __force *)from, n);
+	return 0;
+}
+static inline unsigned long
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
+{
+	memcpy((void __force *)to, from, n);
+	return 0;
+}
 #define __clear_user(addr, n)		(memset((void __force *)addr, 0, n), 0)
 #endif
-
-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_from_user(void *to, const void __user *from, unsigned long n)
-{
-	unsigned long res = n;
-
-	check_object_size(to, n, false);
-
-	if (likely(access_ok(VERIFY_READ, from, n)))
-		res = __arch_copy_from_user(to, from, n);
-	if (unlikely(res))
-		memset(to + (n - res), 0, res);
-	return res;
-}
-
-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_to_user(void __user *to, const void *from, unsigned long n)
-{
-	check_object_size(from, n, true);
-
-	if (access_ok(VERIFY_WRITE, to, n))
-		n = __arch_copy_to_user(to, from, n);
-	return n;
-}
-
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
+#define INLINE_COPY_TO_USER
+#define INLINE_COPY_FROM_USER
 
 static inline unsigned long __must_check clear_user(void __user *to, unsigned long n)
 {
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index a39029b..b72160e 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -115,6 +115,7 @@
 	select SPARSE_IRQ
 	select SYSCTL_EXCEPTION_TRACE
 	select THREAD_INFO_IN_TASK
+	select ARCH_HAS_RAW_COPY_USER
 	help
 	  ARM 64-bit (AArch64) Linux support.
 
diff --git a/arch/arm64/include/asm/extable.h b/arch/arm64/include/asm/extable.h
new file mode 100644
index 0000000..42f50f1
--- /dev/null
+++ b/arch/arm64/include/asm/extable.h
@@ -0,0 +1,25 @@
+#ifndef __ASM_EXTABLE_H
+#define __ASM_EXTABLE_H
+
+/*
+ * The exception table consists of pairs of relative offsets: the first
+ * is the relative offset to an instruction that is allowed to fault,
+ * and the second is the relative offset at which the program should
+ * continue. No registers are modified, so it is entirely up to the
+ * continuation code to figure out what to do.
+ *
+ * All the routines below use bits of fixup code that are out of line
+ * with the main instruction path.  This means when everything is well,
+ * we don't even have to jump over them.  Further, they do not intrude
+ * on our cache or tlb entries.
+ */
+
+struct exception_table_entry
+{
+	int insn, fixup;
+};
+
+#define ARCH_HAS_RELATIVE_EXTABLE
+
+extern int fixup_exception(struct pt_regs *regs);
+#endif
diff --git a/arch/arm64/include/asm/uaccess.h b/arch/arm64/include/asm/uaccess.h
index 7c514e1..ba49717 100644
--- a/arch/arm64/include/asm/uaccess.h
+++ b/arch/arm64/include/asm/uaccess.h
@@ -33,28 +33,7 @@
 #include <asm/ptrace.h>
 #include <asm/memory.h>
 #include <asm/compiler.h>
-
-/*
- * The exception table consists of pairs of relative offsets: the first
- * is the relative offset to an instruction that is allowed to fault,
- * and the second is the relative offset at which the program should
- * continue. No registers are modified, so it is entirely up to the
- * continuation code to figure out what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
-
-struct exception_table_entry
-{
-	int insn, fixup;
-};
-
-#define ARCH_HAS_RELATIVE_EXTABLE
-
-extern int fixup_exception(struct pt_regs *regs);
+#include <asm/extable.h>
 
 #define KERNEL_DS	(-1UL)
 #define get_ds()	(KERNEL_DS)
@@ -352,58 +331,13 @@ do {									\
 })
 
 extern unsigned long __must_check __arch_copy_from_user(void *to, const void __user *from, unsigned long n);
+#define raw_copy_from_user __arch_copy_from_user
 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);
+#define raw_copy_to_user __arch_copy_to_user
+extern unsigned long __must_check raw_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)
-{
-	kasan_check_write(to, 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)
-{
-	kasan_check_read(from, 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)
-{
-	unsigned long res = n;
-	kasan_check_write(to, n);
-	check_object_size(to, n, false);
-
-	if (access_ok(VERIFY_READ, from, n)) {
-		res = __arch_copy_from_user(to, from, n);
-	}
-	if (unlikely(res))
-		memset(to + (n - res), 0, res);
-	return res;
-}
-
-static inline unsigned long __must_check copy_to_user(void __user *to, const void *from, unsigned long n)
-{
-	kasan_check_read(from, n);
-	check_object_size(from, n, true);
-
-	if (access_ok(VERIFY_WRITE, to, n)) {
-		n = __arch_copy_to_user(to, from, n);
-	}
-	return n;
-}
-
-static inline unsigned long __must_check copy_in_user(void __user *to, const void __user *from, unsigned long n)
-{
-	if (access_ok(VERIFY_READ, from, n) && access_ok(VERIFY_WRITE, to, n))
-		n = __copy_in_user(to, from, n);
-	return n;
-}
-
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
+#define INLINE_COPY_TO_USER
+#define INLINE_COPY_FROM_USER
 
 static inline unsigned long __must_check clear_user(void __user *to, unsigned long n)
 {
diff --git a/arch/arm64/kernel/arm64ksyms.c b/arch/arm64/kernel/arm64ksyms.c
index e9c4dc9..67368c7 100644
--- a/arch/arm64/kernel/arm64ksyms.c
+++ b/arch/arm64/kernel/arm64ksyms.c
@@ -38,7 +38,7 @@ EXPORT_SYMBOL(clear_page);
 EXPORT_SYMBOL(__arch_copy_from_user);
 EXPORT_SYMBOL(__arch_copy_to_user);
 EXPORT_SYMBOL(__clear_user);
-EXPORT_SYMBOL(__copy_in_user);
+EXPORT_SYMBOL(raw_copy_in_user);
 
 	/* physical memory */
 EXPORT_SYMBOL(memstart_addr);
diff --git a/arch/arm64/lib/copy_in_user.S b/arch/arm64/lib/copy_in_user.S
index 47184c3..b24a830 100644
--- a/arch/arm64/lib/copy_in_user.S
+++ b/arch/arm64/lib/copy_in_user.S
@@ -64,14 +64,14 @@
 	.endm
 
 end	.req	x5
-ENTRY(__copy_in_user)
+ENTRY(raw_copy_in_user)
 	uaccess_enable_not_uao x3, x4
 	add	end, x0, x2
 #include "copy_template.S"
 	uaccess_disable_not_uao x3
 	mov	x0, #0
 	ret
-ENDPROC(__copy_in_user)
+ENDPROC(raw_copy_in_user)
 
 	.section .fixup,"ax"
 	.align	2
diff --git a/arch/avr32/Kconfig b/arch/avr32/Kconfig
index 7e75d45..8c349f2 100644
--- a/arch/avr32/Kconfig
+++ b/arch/avr32/Kconfig
@@ -19,6 +19,7 @@
 	select HAVE_MOD_ARCH_SPECIFIC
 	select MODULES_USE_ELF_RELA
 	select HAVE_NMI
+	select ARCH_HAS_RAW_COPY_USER
 	help
 	  AVR32 is a high-performance 32-bit RISC microprocessor core,
 	  designed for cost-sensitive embedded applications, with particular
diff --git a/arch/avr32/include/asm/Kbuild b/arch/avr32/include/asm/Kbuild
index 3d7ef2c..ff7d739 100644
--- a/arch/avr32/include/asm/Kbuild
+++ b/arch/avr32/include/asm/Kbuild
@@ -5,6 +5,7 @@
 generic-y += div64.h
 generic-y += emergency-restart.h
 generic-y += exec.h
+generic-y += extable.h
 generic-y += futex.h
 generic-y += irq_regs.h
 generic-y += irq_work.h
diff --git a/arch/avr32/include/asm/uaccess.h b/arch/avr32/include/asm/uaccess.h
index 7ca5cb3..619d50e 100644
--- a/arch/avr32/include/asm/uaccess.h
+++ b/arch/avr32/include/asm/uaccess.h
@@ -66,34 +66,18 @@ static inline void set_fs(mm_segment_t s)
 extern __kernel_size_t __copy_user(void *to, const void *from,
 				   __kernel_size_t n);
 
-extern __kernel_size_t copy_to_user(void __user *to, const void *from,
-				    __kernel_size_t n);
-extern __kernel_size_t ___copy_from_user(void *to, const void __user *from,
-				      __kernel_size_t n);
-
-static inline __kernel_size_t __copy_to_user(void __user *to, const void *from,
-					     __kernel_size_t n)
+static inline unsigned long
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 {
 	return __copy_user((void __force *)to, from, n);
 }
-static inline __kernel_size_t __copy_from_user(void *to,
-					       const void __user *from,
-					       __kernel_size_t n)
+static inline unsigned long
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
 {
 	return __copy_user(to, (const void __force *)from, n);
 }
-static inline __kernel_size_t copy_from_user(void *to,
-					       const void __user *from,
-					       __kernel_size_t n)
-{
-	size_t res = ___copy_from_user(to, from, n);
-	if (unlikely(res))
-		memset(to + (n - res), 0, res);
-	return res;
-}
-
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 /*
  * put_user: - Write a simple value into user space.
@@ -323,9 +307,6 @@ extern long __strnlen_user(const char __user *__s, long __n);
 
 #define strlen_user(s) strnlen_user(s, ~0UL >> 1)
 
-struct exception_table_entry
-{
-	unsigned long insn, fixup;
-};
+#include <asm/extable.h>
 
 #endif /* __ASM_AVR32_UACCESS_H */
diff --git a/arch/avr32/kernel/avr32_ksyms.c b/arch/avr32/kernel/avr32_ksyms.c
index 0d05fd0..0cdae8e 100644
--- a/arch/avr32/kernel/avr32_ksyms.c
+++ b/arch/avr32/kernel/avr32_ksyms.c
@@ -36,8 +36,6 @@ EXPORT_SYMBOL(copy_page);
 /*
  * Userspace access stuff.
  */
-EXPORT_SYMBOL(___copy_from_user);
-EXPORT_SYMBOL(copy_to_user);
 EXPORT_SYMBOL(__copy_user);
 EXPORT_SYMBOL(strncpy_from_user);
 EXPORT_SYMBOL(__strncpy_from_user);
diff --git a/arch/avr32/lib/copy_user.S b/arch/avr32/lib/copy_user.S
index 0753734..f432757 100644
--- a/arch/avr32/lib/copy_user.S
+++ b/arch/avr32/lib/copy_user.S
@@ -23,21 +23,6 @@
 	 */
 	.text
 	.align	1
-	.global	___copy_from_user
-	.type	___copy_from_user, @function
-___copy_from_user:
-	branch_if_kernel r8, __copy_user
-	ret_if_privileged r8, r11, r10, r10
-	rjmp	__copy_user
-	.size	___copy_from_user, . - ___copy_from_user
-
-	.global	copy_to_user
-	.type	copy_to_user, @function
-copy_to_user:
-	branch_if_kernel r8, __copy_user
-	ret_if_privileged r8, r12, r10, r10
-	.size	copy_to_user, . - copy_to_user
-
 	.global	__copy_user
 	.type	__copy_user, @function
 __copy_user:
diff --git a/arch/c6x/Kconfig b/arch/c6x/Kconfig
index 5aa8ea8..3c7bd9a 100644
--- a/arch/c6x/Kconfig
+++ b/arch/c6x/Kconfig
@@ -18,6 +18,7 @@
 	select GENERIC_CLOCKEVENTS
 	select MODULES_USE_ELF_RELA
 	select ARCH_NO_COHERENT_DMA_MMAP
+	select ARCH_HAS_RAW_COPY_USER
 
 config MMU
 	def_bool n
diff --git a/arch/c6x/include/asm/uaccess.h b/arch/c6x/include/asm/uaccess.h
index 79fb6f4..ba67568 100644
--- a/arch/c6x/include/asm/uaccess.h
+++ b/arch/c6x/include/asm/uaccess.h
@@ -13,17 +13,11 @@
 #include <linux/compiler.h>
 #include <linux/string.h>
 
-#ifdef CONFIG_ACCESS_CHECK
-#define __access_ok _access_ok
-#endif
-
 /*
- * __copy_from_user/copy_to_user are based on ones in asm-generic/uaccess.h
- *
  * C6X supports unaligned 32 and 64 bit loads and stores.
  */
-static inline __must_check long __copy_from_user(void *to,
-		const void __user *from, unsigned long n)
+static inline __must_check unsigned long
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
 {
 	u32 tmp32;
 	u64 tmp64;
@@ -58,8 +52,8 @@ static inline __must_check long __copy_from_user(void *to,
 	return 0;
 }
 
-static inline __must_check long __copy_to_user(void __user *to,
-		const void *from, unsigned long n)
+static inline __must_check unsigned long
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 {
 	u32 tmp32;
 	u64 tmp64;
@@ -93,6 +87,8 @@ static inline __must_check long __copy_to_user(void __user *to,
 	memcpy((void __force *)to, from, n);
 	return 0;
 }
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 extern int _access_ok(unsigned long addr, unsigned long size);
 #ifdef CONFIG_ACCESS_CHECK
diff --git a/arch/cris/Kconfig b/arch/cris/Kconfig
index 71b758d..36f94c4 100644
--- a/arch/cris/Kconfig
+++ b/arch/cris/Kconfig
@@ -71,6 +71,7 @@
 	select GENERIC_SCHED_CLOCK if ETRAX_ARCH_V32
 	select HAVE_DEBUG_BUGVERBOSE if ETRAX_ARCH_V32
 	select HAVE_NMI
+	select ARCH_HAS_RAW_COPY_USER
 
 config HZ
 	int
diff --git a/arch/cris/arch-v10/lib/usercopy.c b/arch/cris/arch-v10/lib/usercopy.c
index 1ba7cc0..48fa37f 100644
--- a/arch/cris/arch-v10/lib/usercopy.c
+++ b/arch/cris/arch-v10/lib/usercopy.c
@@ -188,11 +188,10 @@ unsigned long __copy_user(void __user *pdst, const void *psrc, unsigned long pn)
 }
 EXPORT_SYMBOL(__copy_user);
 
-/* Copy from user to kernel, zeroing the bytes that were inaccessible in
-   userland.  The return-value is the number of bytes that were
+/* Copy from user to kernel.  The return-value is the number of bytes that were
    inaccessible.  */
 
-unsigned long __copy_user_zeroing(void *pdst, const void __user *psrc,
+unsigned long __copy_user_in(void *pdst, const void __user *psrc,
 				  unsigned long pn)
 {
   /* We want the parameters put in special registers.
@@ -217,19 +216,17 @@ unsigned long __copy_user_zeroing(void *pdst, const void __user *psrc,
     {
       __asm_copy_from_user_1 (dst, src, retn);
       n--;
+      if (retn)
+         goto exception;
     }
 
     if (((unsigned long) src & 2) && n >= 2)
     {
       __asm_copy_from_user_2 (dst, src, retn);
       n -= 2;
+      if (retn)
+         goto exception;
     }
-
-    /* We only need one check after the unalignment-adjustments, because
-       if both adjustments were done, either both or neither reference
-       had an exception.  */
-    if (retn != 0)
-      goto copy_exception_bytes;
   }
 
   /* Decide which copying method to use. */
@@ -328,7 +325,7 @@ unsigned long __copy_user_zeroing(void *pdst, const void __user *psrc,
     n -= 4;
 
     if (retn)
-      goto copy_exception_bytes;
+      goto exception;
   }
 
   /* If we get here, there were no memory read faults.  */
@@ -356,20 +353,10 @@ unsigned long __copy_user_zeroing(void *pdst, const void __user *psrc,
      bytes.  */
   return retn;
 
-copy_exception_bytes:
-  /* We already have "retn" bytes cleared, and need to clear the
-     remaining "n" bytes.  A non-optimized simple byte-for-byte in-line
-     memset is preferred here, since this isn't speed-critical code and
-     we'd rather have this a leaf-function than calling memset.  */
-  {
-    char *endp;
-    for (endp = dst + n; dst < endp; dst++)
-      *dst = 0;
-  }
-
+exception:
   return retn + n;
 }
-EXPORT_SYMBOL(__copy_user_zeroing);
+EXPORT_SYMBOL(__copy_user_in);
 
 /* Zero userspace.  */
 unsigned long __do_clear_user(void __user *pto, unsigned long pn)
diff --git a/arch/cris/arch-v32/lib/usercopy.c b/arch/cris/arch-v32/lib/usercopy.c
index 05e58da..20b6080 100644
--- a/arch/cris/arch-v32/lib/usercopy.c
+++ b/arch/cris/arch-v32/lib/usercopy.c
@@ -156,10 +156,9 @@ unsigned long __copy_user(void __user *pdst, const void *psrc, unsigned long pn)
 }
 EXPORT_SYMBOL(__copy_user);
 
-/* Copy from user to kernel, zeroing the bytes that were inaccessible in
-   userland.  The return-value is the number of bytes that were
+/* Copy from user to kernel.  The return-value is the number of bytes that were
    inaccessible.  */
-unsigned long __copy_user_zeroing(void *pdst, const void __user *psrc,
+unsigned long __copy_user_in(void *pdst, const void __user *psrc,
 				  unsigned long pn)
 {
   /* We want the parameters put in special registers.
@@ -184,19 +183,18 @@ unsigned long __copy_user_zeroing(void *pdst, const void __user *psrc,
     {
       __asm_copy_from_user_1 (dst, src, retn);
       n--;
+      if (retn != 0)
+        goto exception;
     }
 
     if (((unsigned long) src & 2) && n >= 2)
     {
       __asm_copy_from_user_2 (dst, src, retn);
       n -= 2;
+      if (retn != 0)
+        goto exception;
     }
 
-    /* We only need one check after the unalignment-adjustments, because
-       if both adjustments were done, either both or neither reference
-       had an exception.  */
-    if (retn != 0)
-      goto copy_exception_bytes;
   }
 
   /* Movem is dirt cheap.  The overheap is low enough to always use the
@@ -279,7 +277,7 @@ unsigned long __copy_user_zeroing(void *pdst, const void __user *psrc,
     n -= 4;
 
     if (retn)
-      goto copy_exception_bytes;
+      goto exception;
   }
 
   /* If we get here, there were no memory read faults.  */
@@ -307,20 +305,10 @@ unsigned long __copy_user_zeroing(void *pdst, const void __user *psrc,
      bytes.  */
   return retn;
 
-copy_exception_bytes:
-  /* We already have "retn" bytes cleared, and need to clear the
-     remaining "n" bytes.  A non-optimized simple byte-for-byte in-line
-     memset is preferred here, since this isn't speed-critical code and
-     we'd rather have this a leaf-function than calling memset.  */
-  {
-    char *endp;
-    for (endp = dst + n; dst < endp; dst++)
-      *dst = 0;
-  }
-
+exception:
   return retn + n;
 }
-EXPORT_SYMBOL(__copy_user_zeroing);
+EXPORT_SYMBOL(__copy_user_in);
 
 /* Zero userspace.  */
 unsigned long __do_clear_user(void __user *pto, unsigned long pn)
diff --git a/arch/cris/include/arch-v10/arch/uaccess.h b/arch/cris/include/arch-v10/arch/uaccess.h
index 65b02d9..5477c98 100644
--- a/arch/cris/include/arch-v10/arch/uaccess.h
+++ b/arch/cris/include/arch-v10/arch/uaccess.h
@@ -172,16 +172,14 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 	__asm_copy_user_cont(to, from, ret,	\
 		"	move.b [%1+],$r9\n"	\
 		"2:	move.b $r9,[%0+]\n",	\
-		"3:	addq 1,%2\n"		\
-		"	clear.b [%0+]\n",	\
+		"3:	addq 1,%2\n",		\
 		"	.dword 2b,3b\n")
 
 #define __asm_copy_from_user_2x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
 	__asm_copy_user_cont(to, from, ret,		\
 		"	move.w [%1+],$r9\n"		\
 		"2:	move.w $r9,[%0+]\n" COPY,	\
-		"3:	addq 2,%2\n"			\
-		"	clear.w [%0+]\n" FIXUP,		\
+		"3:	addq 2,%2\n" FIXUP,		\
 		"	.dword 2b,3b\n" TENTRY)
 
 #define __asm_copy_from_user_2(to, from, ret) \
@@ -191,16 +189,14 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 	__asm_copy_from_user_2x_cont(to, from, ret,	\
 		"	move.b [%1+],$r9\n"		\
 		"4:	move.b $r9,[%0+]\n",		\
-		"5:	addq 1,%2\n"			\
-		"	clear.b [%0+]\n",		\
+		"5:	addq 1,%2\n",			\
 		"	.dword 4b,5b\n")
 
 #define __asm_copy_from_user_4x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
 	__asm_copy_user_cont(to, from, ret,		\
 		"	move.d [%1+],$r9\n"		\
 		"2:	move.d $r9,[%0+]\n" COPY,	\
-		"3:	addq 4,%2\n"			\
-		"	clear.d [%0+]\n" FIXUP,		\
+		"3:	addq 4,%2\n" FIXUP,		\
 		"	.dword 2b,3b\n" TENTRY)
 
 #define __asm_copy_from_user_4(to, from, ret) \
@@ -210,8 +206,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 	__asm_copy_from_user_4x_cont(to, from, ret,	\
 		"	move.b [%1+],$r9\n"		\
 		"4:	move.b $r9,[%0+]\n",		\
-		"5:	addq 1,%2\n"			\
-		"	clear.b [%0+]\n",		\
+		"5:	addq 1,%2\n",			\
 		"	.dword 4b,5b\n")
 
 #define __asm_copy_from_user_6x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
@@ -219,7 +214,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"	move.w [%1+],$r9\n"		\
 		"4:	move.w $r9,[%0+]\n" COPY,	\
 		"5:	addq 2,%2\n"			\
-		"	clear.w [%0+]\n" FIXUP,		\
+			FIXUP,				\
 		"	.dword 4b,5b\n" TENTRY)
 
 #define __asm_copy_from_user_6(to, from, ret) \
@@ -229,8 +224,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 	__asm_copy_from_user_6x_cont(to, from, ret,	\
 		"	move.b [%1+],$r9\n"		\
 		"6:	move.b $r9,[%0+]\n",		\
-		"7:	addq 1,%2\n"			\
-		"	clear.b [%0+]\n",		\
+		"7:	addq 1,%2\n",			\
 		"	.dword 6b,7b\n")
 
 #define __asm_copy_from_user_8x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
@@ -238,7 +232,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"	move.d [%1+],$r9\n"		\
 		"4:	move.d $r9,[%0+]\n" COPY,	\
 		"5:	addq 4,%2\n"			\
-		"	clear.d [%0+]\n" FIXUP,		\
+			FIXUP,				\
 		"	.dword 4b,5b\n" TENTRY)
 
 #define __asm_copy_from_user_8(to, from, ret) \
@@ -248,8 +242,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 	__asm_copy_from_user_8x_cont(to, from, ret,	\
 		"	move.b [%1+],$r9\n"		\
 		"6:	move.b $r9,[%0+]\n",		\
-		"7:	addq 1,%2\n"			\
-		"	clear.b [%0+]\n",		\
+		"7:	addq 1,%2\n",			\
 		"	.dword 6b,7b\n")
 
 #define __asm_copy_from_user_10x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
@@ -257,7 +250,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"	move.w [%1+],$r9\n"		\
 		"6:	move.w $r9,[%0+]\n" COPY,	\
 		"7:	addq 2,%2\n"			\
-		"	clear.w [%0+]\n" FIXUP,		\
+			FIXUP,				\
 		"	.dword 6b,7b\n" TENTRY)
 
 #define __asm_copy_from_user_10(to, from, ret) \
@@ -267,8 +260,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 	__asm_copy_from_user_10x_cont(to, from, ret,	\
 		"	move.b [%1+],$r9\n"		\
 		"8:	move.b $r9,[%0+]\n",		\
-		"9:	addq 1,%2\n"			\
-		"	clear.b [%0+]\n",		\
+		"9:	addq 1,%2\n",			\
 		"	.dword 8b,9b\n")
 
 #define __asm_copy_from_user_12x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
@@ -276,7 +268,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"	move.d [%1+],$r9\n"		\
 		"6:	move.d $r9,[%0+]\n" COPY,	\
 		"7:	addq 4,%2\n"			\
-		"	clear.d [%0+]\n" FIXUP,		\
+			FIXUP,				\
 		"	.dword 6b,7b\n" TENTRY)
 
 #define __asm_copy_from_user_12(to, from, ret) \
@@ -286,8 +278,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 	__asm_copy_from_user_12x_cont(to, from, ret,	\
 		"	move.b [%1+],$r9\n"		\
 		"8:	move.b $r9,[%0+]\n",		\
-		"9:	addq 1,%2\n"			\
-		"	clear.b [%0+]\n",		\
+		"9:	addq 1,%2\n",			\
 		"	.dword 8b,9b\n")
 
 #define __asm_copy_from_user_14x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
@@ -295,7 +286,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"	move.w [%1+],$r9\n"		\
 		"8:	move.w $r9,[%0+]\n" COPY,	\
 		"9:	addq 2,%2\n"			\
-		"	clear.w [%0+]\n" FIXUP,		\
+			FIXUP,				\
 		"	.dword 8b,9b\n" TENTRY)
 
 #define __asm_copy_from_user_14(to, from, ret) \
@@ -305,8 +296,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 	__asm_copy_from_user_14x_cont(to, from, ret,	\
 		"	move.b [%1+],$r9\n"		\
 		"10:	move.b $r9,[%0+]\n",		\
-		"11:	addq 1,%2\n"			\
-		"	clear.b [%0+]\n",		\
+		"11:	addq 1,%2\n",			\
 		"	.dword 10b,11b\n")
 
 #define __asm_copy_from_user_16x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
@@ -314,7 +304,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"	move.d [%1+],$r9\n"		\
 		"8:	move.d $r9,[%0+]\n" COPY,	\
 		"9:	addq 4,%2\n"			\
-		"	clear.d [%0+]\n" FIXUP,		\
+			FIXUP,				\
 		"	.dword 8b,9b\n" TENTRY)
 
 #define __asm_copy_from_user_16(to, from, ret) \
@@ -325,7 +315,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"	move.d [%1+],$r9\n"		\
 		"10:	move.d $r9,[%0+]\n" COPY,	\
 		"11:	addq 4,%2\n"			\
-		"	clear.d [%0+]\n" FIXUP,		\
+			FIXUP,				\
 		"	.dword 10b,11b\n" TENTRY)
 
 #define __asm_copy_from_user_20(to, from, ret) \
@@ -336,7 +326,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"	move.d [%1+],$r9\n"		\
 		"12:	move.d $r9,[%0+]\n" COPY,	\
 		"13:	addq 4,%2\n"			\
-		"	clear.d [%0+]\n" FIXUP,		\
+			FIXUP,				\
 		"	.dword 12b,13b\n" TENTRY)
 
 #define __asm_copy_from_user_24(to, from, ret) \
diff --git a/arch/cris/include/arch-v32/arch/uaccess.h b/arch/cris/include/arch-v32/arch/uaccess.h
index 3196019..dc2ce09 100644
--- a/arch/cris/include/arch-v32/arch/uaccess.h
+++ b/arch/cris/include/arch-v32/arch/uaccess.h
@@ -178,8 +178,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"2:	move.b [%1+],$acr\n"	\
 		"	move.b $acr,[%0+]\n",	\
 		"3:	addq 1,%2\n"		\
-		"	jump 1b\n"		\
-		"	clear.b [%0+]\n",	\
+		"	jump 1b\n",		\
 		"	.dword 2b,3b\n")
 
 #define __asm_copy_from_user_2x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
@@ -189,8 +188,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"	move.w $acr,[%0+]\n",		\
 			FIXUP				\
 		"3:	addq 2,%2\n"			\
-		"	jump 1b\n"			\
-		"	clear.w [%0+]\n",		\
+		"	jump 1b\n",			\
 			TENTRY				\
 		"	.dword 2b,3b\n")
 
@@ -201,8 +199,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 	__asm_copy_from_user_2x_cont(to, from, ret,	\
 		"4:	move.b [%1+],$acr\n"		\
 		"	move.b $acr,[%0+]\n",		\
-		"5:	addq 1,%2\n"			\
-		"	clear.b [%0+]\n",		\
+		"5:	addq 1,%2\n",			\
 		"	.dword 4b,5b\n")
 
 #define __asm_copy_from_user_4x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
@@ -212,8 +209,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"	move.d $acr,[%0+]\n",		\
 			FIXUP				\
 		"3:	addq 4,%2\n"			\
-		"	jump 1b\n"			\
-		"	clear.d [%0+]\n",		\
+		"	jump 1b\n",			\
 			TENTRY				\
 		"	.dword 2b,3b\n")
 
@@ -224,8 +220,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 	__asm_copy_from_user_4x_cont(to, from, ret,	\
 		"4:	move.b [%1+],$acr\n"		\
 		"	move.b $acr,[%0+]\n",		\
-		"5:	addq 1,%2\n"			\
-		"	clear.b [%0+]\n",		\
+		"5:	addq 1,%2\n",			\
 		"	.dword 4b,5b\n")
 
 #define __asm_copy_from_user_6x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
@@ -234,8 +229,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"4:	move.w [%1+],$acr\n"		\
 		"	move.w $acr,[%0+]\n",		\
 			FIXUP				\
-		"5:	addq 2,%2\n"			\
-		"	clear.w [%0+]\n",		\
+		"5:	addq 2,%2\n",			\
 			TENTRY				\
 		"	.dword 4b,5b\n")
 
@@ -246,8 +240,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 	__asm_copy_from_user_6x_cont(to, from, ret,	\
 		"6:	move.b [%1+],$acr\n"		\
 		"	move.b $acr,[%0+]\n",		\
-		"7:	addq 1,%2\n"			\
-		"	clear.b [%0+]\n",		\
+		"7:	addq 1,%2\n",			\
 		"	.dword 6b,7b\n")
 
 #define __asm_copy_from_user_8x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
@@ -256,8 +249,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"4:	move.d [%1+],$acr\n"		\
 		"	move.d $acr,[%0+]\n",		\
 			FIXUP				\
-		"5:	addq 4,%2\n"			\
-		"	clear.d [%0+]\n",		\
+		"5:	addq 4,%2\n",			\
 			TENTRY				\
 		"	.dword 4b,5b\n")
 
@@ -268,8 +260,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 	__asm_copy_from_user_8x_cont(to, from, ret,	\
 		"6:	move.b [%1+],$acr\n"		\
 		"	move.b $acr,[%0+]\n",		\
-		"7:	addq 1,%2\n"			\
-		"	clear.b [%0+]\n",		\
+		"7:	addq 1,%2\n",			\
 		"	.dword 6b,7b\n")
 
 #define __asm_copy_from_user_10x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
@@ -278,8 +269,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"6:	move.w [%1+],$acr\n"		\
 		"	move.w $acr,[%0+]\n",		\
 			FIXUP				\
-		"7:	addq 2,%2\n"			\
-		"	clear.w [%0+]\n",		\
+		"7:	addq 2,%2\n",			\
 			TENTRY				\
 		"	.dword 6b,7b\n")
 
@@ -290,8 +280,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 	__asm_copy_from_user_10x_cont(to, from, ret,	\
 		"8:	move.b [%1+],$acr\n"		\
 		"	move.b $acr,[%0+]\n",		\
-		"9:	addq 1,%2\n"			\
-		"	clear.b [%0+]\n",		\
+		"9:	addq 1,%2\n",			\
 		"	.dword 8b,9b\n")
 
 #define __asm_copy_from_user_12x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
@@ -300,8 +289,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"6:	move.d [%1+],$acr\n"		\
 		"	move.d $acr,[%0+]\n",		\
 			FIXUP				\
-		"7:	addq 4,%2\n"			\
-		"	clear.d [%0+]\n",		\
+		"7:	addq 4,%2\n",			\
 			TENTRY				\
 		"	.dword 6b,7b\n")
 
@@ -312,8 +300,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 	__asm_copy_from_user_12x_cont(to, from, ret,	\
 		"8:	move.b [%1+],$acr\n"		\
 		"	move.b $acr,[%0+]\n",		\
-		"9:	addq 1,%2\n"			\
-		"	clear.b [%0+]\n",		\
+		"9:	addq 1,%2\n",			\
 		"	.dword 8b,9b\n")
 
 #define __asm_copy_from_user_14x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
@@ -322,8 +309,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"8:	move.w [%1+],$acr\n"		\
 		"	move.w $acr,[%0+]\n",		\
 			FIXUP				\
-		"9:	addq 2,%2\n"			\
-		"	clear.w [%0+]\n",		\
+		"9:	addq 2,%2\n",			\
 			TENTRY				\
 		"	.dword 8b,9b\n")
 
@@ -334,8 +320,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 	__asm_copy_from_user_14x_cont(to, from, ret,	\
 		"10:	move.b [%1+],$acr\n"		\
 		"	move.b $acr,[%0+]\n",		\
-		"11:	addq 1,%2\n"			\
-		"	clear.b [%0+]\n",		\
+		"11:	addq 1,%2\n",			\
 		"	.dword 10b,11b\n")
 
 #define __asm_copy_from_user_16x_cont(to, from, ret, COPY, FIXUP, TENTRY) \
@@ -344,8 +329,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"8:	move.d [%1+],$acr\n"		\
 		"	move.d $acr,[%0+]\n",		\
 			FIXUP				\
-		"9:	addq 4,%2\n"			\
-		"	clear.d [%0+]\n",		\
+		"9:	addq 4,%2\n",			\
 			TENTRY				\
 		"	.dword 8b,9b\n")
 
@@ -358,8 +342,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"10:	move.d [%1+],$acr\n"		\
 		"	move.d $acr,[%0+]\n",		\
 			FIXUP				\
-		"11:	addq 4,%2\n"			\
-		"	clear.d [%0+]\n",		\
+		"11:	addq 4,%2\n",			\
 			TENTRY				\
 		"	.dword 10b,11b\n")
 
@@ -372,8 +355,7 @@ __do_strncpy_from_user(char *dst, const char *src, long count)
 		"12:	move.d [%1+],$acr\n"		\
 		"	move.d $acr,[%0+]\n",		\
 			FIXUP				\
-		"13:	addq 4,%2\n"			\
-		"	clear.d [%0+]\n",		\
+		"13:	addq 4,%2\n",			\
 			TENTRY				\
 		"	.dword 12b,13b\n")
 
diff --git a/arch/cris/include/asm/Kbuild b/arch/cris/include/asm/Kbuild
index 0f5132b..2890099 100644
--- a/arch/cris/include/asm/Kbuild
+++ b/arch/cris/include/asm/Kbuild
@@ -9,6 +9,7 @@
 generic-y += div64.h
 generic-y += errno.h
 generic-y += exec.h
+generic-y += extable.h
 generic-y += emergency-restart.h
 generic-y += fcntl.h
 generic-y += futex.h
diff --git a/arch/cris/include/asm/uaccess.h b/arch/cris/include/asm/uaccess.h
index 5f5b8f5..0d473ae 100644
--- a/arch/cris/include/asm/uaccess.h
+++ b/arch/cris/include/asm/uaccess.h
@@ -50,23 +50,7 @@
 #define access_ok(type, addr, size) __access_ok((unsigned long)(addr), (size))
 
 #include <arch/uaccess.h>
-
-/*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
-
-struct exception_table_entry {
-	unsigned long insn, fixup;
-};
+#include <asm/extable.h>
 
 /*
  * These are the main single-value transfer routines.  They automatically
@@ -185,7 +169,7 @@ extern long __get_user_bad(void);
    live in lib/usercopy.c  */
 
 extern unsigned long __copy_user(void __user *to, const void *from, unsigned long n);
-extern unsigned long __copy_user_zeroing(void *to, const void __user *from, unsigned long n);
+extern unsigned long __copy_user_in(void *to, const void __user *from, unsigned long n);
 extern unsigned long __do_clear_user(void __user *to, unsigned long n);
 
 static inline long
@@ -252,7 +236,7 @@ __constant_copy_from_user(void *to, const void __user *from, unsigned long n)
 	else if (n == 24)
 		__asm_copy_from_user_24(to, from, ret);
 	else
-		ret = __copy_user_zeroing(to, from, n);
+		ret = __copy_user_in(to, from, n);
 
 	return ret;
 }
@@ -352,62 +336,33 @@ static inline size_t clear_user(void __user *to, size_t n)
 		return __do_clear_user(to, n);
 }
 
-static inline size_t copy_from_user(void *to, const void __user *from, size_t n)
+static inline unsigned long
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
 {
-	if (unlikely(!access_ok(VERIFY_READ, from, n))) {
-		memset(to, 0, n);
-		return n;
-	}
 	if (__builtin_constant_p(n))
 		return __constant_copy_from_user(to, from, n);
 	else
-		return __copy_user_zeroing(to, from, n);
+		return __copy_user_in(to, from, n);
 }
 
-static inline size_t copy_to_user(void __user *to, const void *from, size_t n)
+static inline unsigned long
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 {
-	if (unlikely(!access_ok(VERIFY_WRITE, to, n)))
-		return n;
 	if (__builtin_constant_p(n))
 		return __constant_copy_to_user(to, from, n);
 	else
 		return __copy_user(to, from, n);
 }
 
-/* We let the __ versions of copy_from/to_user inline, because they're often
- * used in fast paths and have only a small space overhead.
- */
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 static inline unsigned long
-__generic_copy_from_user_nocheck(void *to, const void __user *from,
-				 unsigned long n)
-{
-	return __copy_user_zeroing(to, from, n);
-}
-
-static inline unsigned long
-__generic_copy_to_user_nocheck(void __user *to, const void *from,
-			       unsigned long n)
-{
-	return __copy_user(to, from, n);
-}
-
-static inline unsigned long
-__generic_clear_user_nocheck(void __user *to, unsigned long n)
+__clear_user(void __user *to, unsigned long n)
 {
 	return __do_clear_user(to, n);
 }
 
-/* without checking */
-
-#define __copy_to_user(to, from, n) \
-	__generic_copy_to_user_nocheck((to), (from), (n))
-#define __copy_from_user(to, from, n) \
-	__generic_copy_from_user_nocheck((to), (from), (n))
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
-#define __clear_user(to, n) __generic_clear_user_nocheck((to), (n))
-
 #define strlen_user(str)	strnlen_user((str), 0x7ffffffe)
 
 #endif	/* _CRIS_UACCESS_H */
diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig
index eefd9a4..e489fef1 100644
--- a/arch/frv/Kconfig
+++ b/arch/frv/Kconfig
@@ -16,6 +16,7 @@
 	select OLD_SIGACTION
 	select HAVE_DEBUG_STACKOVERFLOW
 	select ARCH_NO_COHERENT_DMA_MMAP
+	select ARCH_HAS_RAW_COPY_USER
 
 config ZONE_DMA
 	bool
diff --git a/arch/frv/include/asm/Kbuild b/arch/frv/include/asm/Kbuild
index c33b467..cce3bc36 100644
--- a/arch/frv/include/asm/Kbuild
+++ b/arch/frv/include/asm/Kbuild
@@ -1,6 +1,7 @@
 
 generic-y += clkdev.h
 generic-y += exec.h
+generic-y += extable.h
 generic-y += irq_work.h
 generic-y += mcs_spinlock.h
 generic-y += mm-arch-hooks.h
diff --git a/arch/frv/include/asm/uaccess.h b/arch/frv/include/asm/uaccess.h
index 55b3a69..e4e33b4 100644
--- a/arch/frv/include/asm/uaccess.h
+++ b/arch/frv/include/asm/uaccess.h
@@ -18,6 +18,7 @@
 #include <linux/mm.h>
 #include <asm/segment.h>
 #include <asm/sections.h>
+#include <asm/extable.h>
 
 #define __ptr(x) ((unsigned long __force *)(x))
 
@@ -59,26 +60,6 @@ static inline int ___range_ok(unsigned long addr, unsigned long size)
 #define access_ok(type,addr,size) (__range_ok((void __user *)(addr), (size)) == 0)
 #define __access_ok(addr,size) (__range_ok((addr), (size)) == 0)
 
-/*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
-struct exception_table_entry
-{
-	unsigned long insn, fixup;
-};
-
-/* Returns 0 if exception not found and fixup otherwise.  */
-extern unsigned long search_exception_table(unsigned long);
-
 
 /*
  * These are the main single-value transfer routines.  They automatically
@@ -252,23 +233,44 @@ do {							\
 /*
  *
  */
+
 #define ____force(x) (__force void *)(void __user *)(x)
 #ifdef CONFIG_MMU
 extern long __memset_user(void *dst, unsigned long count);
 extern long __memcpy_user(void *dst, const void *src, unsigned long count);
 
 #define __clear_user(dst,count)			__memset_user(____force(dst), (count))
-#define __copy_from_user_inatomic(to, from, n)	__memcpy_user((to), ____force(from), (n))
-#define __copy_to_user_inatomic(to, from, n)	__memcpy_user(____force(to), (from), (n))
 
 #else
 
 #define __clear_user(dst,count)			(memset(____force(dst), 0, (count)), 0)
-#define __copy_from_user_inatomic(to, from, n)	(memcpy((to), ____force(from), (n)), 0)
-#define __copy_to_user_inatomic(to, from, n)	(memcpy(____force(to), (from), (n)), 0)
 
 #endif
 
+static inline unsigned long
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
+{
+#ifdef CONFIG_MMU
+	return __memcpy_user(to, (__force const void *)from, n);
+#else
+	memcpy(to, (__force const void *)from, n);
+	return 0;
+#endif
+}
+
+static inline unsigned long
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
+{
+#ifdef CONFIG_MMU
+	return __memcpy_user((__force void *)to, from, n);
+#else
+	memcpy((__force void *)to, from, n);
+	return 0;
+#endif
+}
+#define INLINE_COPY_TO_USER
+#define INLINE_COPY_FROM_USER
+
 static inline unsigned long __must_check
 clear_user(void __user *to, unsigned long n)
 {
@@ -277,43 +279,9 @@ clear_user(void __user *to, unsigned long n)
 	return n;
 }
 
-static inline unsigned long __must_check
-__copy_to_user(void __user *to, const void *from, unsigned long n)
-{
-       might_fault();
-       return __copy_to_user_inatomic(to, from, n);
-}
-
-static inline unsigned long
-__copy_from_user(void *to, const void __user *from, unsigned long n)
-{
-       might_fault();
-       return __copy_from_user_inatomic(to, from, n);
-}
-
-static inline long copy_from_user(void *to, const void __user *from, unsigned long n)
-{
-	unsigned long ret = n;
-
-	if (likely(__access_ok(from, n)))
-		ret = __copy_from_user(to, from, n);
-
-	if (unlikely(ret != 0))
-		memset(to + (n - ret), 0, ret);
-
-	return ret;
-}
-
-static inline long copy_to_user(void __user *to, const void *from, unsigned long n)
-{
-	return likely(__access_ok(to, n)) ? __copy_to_user(to, from, n) : n;
-}
-
 extern long strncpy_from_user(char *dst, const char __user *src, long count);
 extern long strnlen_user(const char __user *src, long count);
 
 #define strlen_user(str) strnlen_user(str, 32767)
 
-extern unsigned long search_exception_table(unsigned long addr);
-
 #endif /* _ASM_UACCESS_H */
diff --git a/arch/frv/kernel/traps.c b/arch/frv/kernel/traps.c
index ce29991..fb08ebe 100644
--- a/arch/frv/kernel/traps.c
+++ b/arch/frv/kernel/traps.c
@@ -360,13 +360,8 @@ asmlinkage void memory_access_exception(unsigned long esr0,
 	siginfo_t info;
 
 #ifdef CONFIG_MMU
-	unsigned long fixup;
-
-	fixup = search_exception_table(__frame->pc);
-	if (fixup) {
-		__frame->pc = fixup;
+	if (fixup_exception(__frame))
 		return;
-	}
 #endif
 
 	die_if_kernel("-- Memory Access Exception --\n"
diff --git a/arch/frv/mm/extable.c b/arch/frv/mm/extable.c
index a0e8b3e..9198ddd 100644
--- a/arch/frv/mm/extable.c
+++ b/arch/frv/mm/extable.c
@@ -10,40 +10,39 @@ extern const void __memset_end, __memset_user_error_lr, __memset_user_error_hand
 extern const void __memcpy_end, __memcpy_user_error_lr, __memcpy_user_error_handler;
 extern spinlock_t modlist_lock;
 
-
-/*****************************************************************************/
-/*
- * see if there's a fixup handler available to deal with a kernel fault
- */
-unsigned long search_exception_table(unsigned long pc)
+int fixup_exception(struct pt_regs *regs)
 {
 	const struct exception_table_entry *extab;
+	unsigned long pc = regs->pc;
 
 	/* determine if the fault lay during a memcpy_user or a memset_user */
-	if (__frame->lr == (unsigned long) &__memset_user_error_lr &&
+	if (regs->lr == (unsigned long) &__memset_user_error_lr &&
 	    (unsigned long) &memset <= pc && pc < (unsigned long) &__memset_end
 	    ) {
 		/* the fault occurred in a protected memset
 		 * - we search for the return address (in LR) instead of the program counter
 		 * - it was probably during a clear_user()
 		 */
-		return (unsigned long) &__memset_user_error_handler;
+		regs->pc = (unsigned long) &__memset_user_error_handler;
+		return 1;
 	}
 
-	if (__frame->lr == (unsigned long) &__memcpy_user_error_lr &&
+	if (regs->lr == (unsigned long) &__memcpy_user_error_lr &&
 	    (unsigned long) &memcpy <= pc && pc < (unsigned long) &__memcpy_end
 	    ) {
 		/* the fault occurred in a protected memset
 		 * - we search for the return address (in LR) instead of the program counter
 		 * - it was probably during a copy_to/from_user()
 		 */
-		return (unsigned long) &__memcpy_user_error_handler;
+		regs->pc = (unsigned long) &__memcpy_user_error_handler;
+		return 1;
 	}
 
 	extab = search_exception_tables(pc);
-	if (extab)
-		return extab->fixup;
+	if (extab) {
+		regs->pc = extab->fixup;
+		return 1;
+	}
 
 	return 0;
-
-} /* end search_exception_table() */
+}
diff --git a/arch/frv/mm/fault.c b/arch/frv/mm/fault.c
index 614a46c..179e79e 100644
--- a/arch/frv/mm/fault.c
+++ b/arch/frv/mm/fault.c
@@ -33,7 +33,7 @@ asmlinkage void do_page_fault(int datammu, unsigned long esr0, unsigned long ear
 {
 	struct vm_area_struct *vma;
 	struct mm_struct *mm;
-	unsigned long _pme, lrai, lrad, fixup;
+	unsigned long _pme, lrai, lrad;
 	unsigned long flags = 0;
 	siginfo_t info;
 	pgd_t *pge;
@@ -201,10 +201,8 @@ asmlinkage void do_page_fault(int datammu, unsigned long esr0, unsigned long ear
 
  no_context:
 	/* are we prepared to handle this kernel fault? */
-	if ((fixup = search_exception_table(__frame->pc)) != 0) {
-		__frame->pc = fixup;
+	if (fixup_exception(__frame))
 		return;
-	}
 
 /*
  * Oops. The kernel tried to access some bad page. We'll have to
diff --git a/arch/h8300/Kconfig b/arch/h8300/Kconfig
index 3ae8525..4738834 100644
--- a/arch/h8300/Kconfig
+++ b/arch/h8300/Kconfig
@@ -22,6 +22,7 @@
 	select HAVE_ARCH_KGDB
 	select HAVE_ARCH_HASH
 	select CPU_NO_EFFICIENT_FFS
+	select ARCH_HAS_RAW_COPY_USER
 
 config RWSEM_GENERIC_SPINLOCK
 	def_bool y
diff --git a/arch/h8300/include/asm/uaccess.h b/arch/h8300/include/asm/uaccess.h
index c156a41..6f6144a 100644
--- a/arch/h8300/include/asm/uaccess.h
+++ b/arch/h8300/include/asm/uaccess.h
@@ -3,8 +3,8 @@
 
 #include <linux/string.h>
 
-static inline __must_check long __copy_from_user(void *to,
-		const void __user * from, unsigned long n)
+static inline __must_check unsigned long
+raw_copy_from_user(void *to, const void __user * from, unsigned long n)
 {
 	if (__builtin_constant_p(n)) {
 		switch(n) {
@@ -24,8 +24,8 @@ static inline __must_check long __copy_from_user(void *to,
 	return 0;
 }
 
-static inline __must_check long __copy_to_user(void __user *to,
-		const void *from, unsigned long n)
+static inline __must_check unsigned long
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 {
 	if (__builtin_constant_p(n)) {
 		switch(n) {
@@ -46,6 +46,8 @@ static inline __must_check long __copy_to_user(void __user *to,
 	memcpy((void __force *)to, from, n);
 	return 0;
 }
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 #include <asm-generic/uaccess.h>
 
diff --git a/arch/hexagon/Kconfig b/arch/hexagon/Kconfig
index 1941e4b..0c536a8 100644
--- a/arch/hexagon/Kconfig
+++ b/arch/hexagon/Kconfig
@@ -26,6 +26,7 @@
 	select GENERIC_CLOCKEVENTS_BROADCAST
 	select MODULES_USE_ELF_RELA
 	select GENERIC_CPU_DEVICES
+	select ARCH_HAS_RAW_COPY_USER
 	---help---
 	  Qualcomm Hexagon is a processor architecture designed for high
 	  performance and low power across a wide variety of applications.
diff --git a/arch/hexagon/include/asm/uaccess.h b/arch/hexagon/include/asm/uaccess.h
index 3a7f818..9ba0586 100644
--- a/arch/hexagon/include/asm/uaccess.h
+++ b/arch/hexagon/include/asm/uaccess.h
@@ -65,19 +65,12 @@
  */
 
 /*  Assembly somewhat optimized copy routines  */
-unsigned long __copy_from_user_hexagon(void *to, const void __user *from,
+unsigned long raw_copy_from_user(void *to, const void __user *from,
 				     unsigned long n);
-unsigned long __copy_to_user_hexagon(void __user *to, const void *from,
+unsigned long raw_copy_to_user(void __user *to, const void *from,
 				   unsigned long n);
-
-#define __copy_from_user(to, from, n) __copy_from_user_hexagon(to, from, n)
-#define __copy_to_user(to, from, n) __copy_to_user_hexagon(to, from, n)
-
-/*
- * XXX todo: some additonal performance gain is possible by
- * implementing __copy_to/from_user_inatomic, which is much
- * like __copy_to/from_user, but performs slightly less checking.
- */
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 __kernel_size_t __clear_user_hexagon(void __user *dest, unsigned long count);
 #define __clear_user(a, s) __clear_user_hexagon((a), (s))
diff --git a/arch/hexagon/kernel/hexagon_ksyms.c b/arch/hexagon/kernel/hexagon_ksyms.c
index af9dec4..00bcad9 100644
--- a/arch/hexagon/kernel/hexagon_ksyms.c
+++ b/arch/hexagon/kernel/hexagon_ksyms.c
@@ -25,8 +25,8 @@
 
 /* Additional functions */
 EXPORT_SYMBOL(__clear_user_hexagon);
-EXPORT_SYMBOL(__copy_from_user_hexagon);
-EXPORT_SYMBOL(__copy_to_user_hexagon);
+EXPORT_SYMBOL(raw_copy_from_user);
+EXPORT_SYMBOL(raw_copy_to_user);
 EXPORT_SYMBOL(__iounmap);
 EXPORT_SYMBOL(__strnlen_user);
 EXPORT_SYMBOL(__vmgetie);
diff --git a/arch/hexagon/mm/copy_from_user.S b/arch/hexagon/mm/copy_from_user.S
index 7fc94f3..7da066f 100644
--- a/arch/hexagon/mm/copy_from_user.S
+++ b/arch/hexagon/mm/copy_from_user.S
@@ -44,7 +44,7 @@
 #define bytes r2
 #define loopcount r5
 
-#define FUNCNAME __copy_from_user_hexagon
+#define FUNCNAME raw_copy_from_user
 #include "copy_user_template.S"
 
 	/* LOAD FAULTS from COPY_FROM_USER */
diff --git a/arch/hexagon/mm/copy_to_user.S b/arch/hexagon/mm/copy_to_user.S
index 0cfbcc0..a7b7f8d 100644
--- a/arch/hexagon/mm/copy_to_user.S
+++ b/arch/hexagon/mm/copy_to_user.S
@@ -43,7 +43,7 @@
 #define bytes r2
 #define loopcount r5
 
-#define FUNCNAME __copy_to_user_hexagon
+#define FUNCNAME raw_copy_to_user
 #include "copy_user_template.S"
 
 	/* STORE FAULTS from COPY_TO_USER */
diff --git a/arch/ia64/include/asm/extable.h b/arch/ia64/include/asm/extable.h
new file mode 100644
index 0000000..20376e7
--- /dev/null
+++ b/arch/ia64/include/asm/extable.h
@@ -0,0 +1,11 @@
+#ifndef _ASM_IA64_EXTABLE_H
+#define _ASM_IA64_EXTABLE_H
+
+#define ARCH_HAS_RELATIVE_EXTABLE
+
+struct exception_table_entry {
+	int insn;	/* location-relative address of insn this fixup is for */
+	int fixup;	/* location-relative continuation addr.; if bit 2 is set, r9 is set to 0 */
+};
+
+#endif
diff --git a/arch/ia64/include/asm/uaccess.h b/arch/ia64/include/asm/uaccess.h
index d471d1a..bd15972 100644
--- a/arch/ia64/include/asm/uaccess.h
+++ b/arch/ia64/include/asm/uaccess.h
@@ -39,6 +39,7 @@
 #include <asm/intrinsics.h>
 #include <asm/pgtable.h>
 #include <asm/io.h>
+#include <asm/extable.h>
 
 /*
  * For historical reasons, the following macros are grossly misnamed:
@@ -58,14 +59,14 @@
  * address TASK_SIZE is never valid.  We also need to make sure that the address doesn't
  * point inside the virtually mapped linear page table.
  */
-#define __access_ok(addr, size, segment)						\
-({											\
-	__chk_user_ptr(addr);								\
-	(likely((unsigned long) (addr) <= (segment).seg)				\
-	 && ((segment).seg == KERNEL_DS.seg						\
-	     || likely(REGION_OFFSET((unsigned long) (addr)) < RGN_MAP_LIMIT)));	\
-})
-#define access_ok(type, addr, size)	__access_ok((addr), (size), get_fs())
+static inline int __access_ok(const void __user *p, unsigned long size)
+{
+	unsigned long addr = (unsigned long)p;
+	unsigned long seg = get_fs().seg;
+	return likely(addr <= seg) &&
+	 (seg == KERNEL_DS.seg || likely(REGION_OFFSET(addr) < RGN_MAP_LIMIT));
+}
+#define access_ok(type, addr, size)	__access_ok((addr), (size))
 
 /*
  * These are the main single-value transfer routines.  They automatically
@@ -75,8 +76,8 @@
  * (a) re-use the arguments for side effects (sizeof/typeof is ok)
  * (b) require any knowledge of processes at this stage
  */
-#define put_user(x, ptr)	__put_user_check((__typeof__(*(ptr))) (x), (ptr), sizeof(*(ptr)), get_fs())
-#define get_user(x, ptr)	__get_user_check((x), (ptr), sizeof(*(ptr)), get_fs())
+#define put_user(x, ptr)	__put_user_check((__typeof__(*(ptr))) (x), (ptr), sizeof(*(ptr)))
+#define get_user(x, ptr)	__get_user_check((x), (ptr), sizeof(*(ptr)))
 
 /*
  * The "__xxx" versions do not do address space checking, useful when
@@ -179,13 +180,13 @@ extern void __get_user_unknown (void);
  * could clobber r8 and r9 (among others).  Thus, be careful not to evaluate it while
  * using r8/r9.
  */
-#define __do_get_user(check, x, ptr, size, segment)					\
+#define __do_get_user(check, x, ptr, size)						\
 ({											\
 	const __typeof__(*(ptr)) __user *__gu_ptr = (ptr);				\
 	__typeof__ (size) __gu_size = (size);						\
 	long __gu_err = -EFAULT;							\
 	unsigned long __gu_val = 0;							\
-	if (!check || __access_ok(__gu_ptr, size, segment))				\
+	if (!check || __access_ok(__gu_ptr, size))					\
 		switch (__gu_size) {							\
 		      case 1: __get_user_size(__gu_val, __gu_ptr, 1, __gu_err); break;	\
 		      case 2: __get_user_size(__gu_val, __gu_ptr, 2, __gu_err); break;	\
@@ -197,8 +198,8 @@ extern void __get_user_unknown (void);
 	__gu_err;									\
 })
 
-#define __get_user_nocheck(x, ptr, size)	__do_get_user(0, x, ptr, size, KERNEL_DS)
-#define __get_user_check(x, ptr, size, segment)	__do_get_user(1, x, ptr, size, segment)
+#define __get_user_nocheck(x, ptr, size)	__do_get_user(0, x, ptr, size)
+#define __get_user_check(x, ptr, size)	__do_get_user(1, x, ptr, size)
 
 extern void __put_user_unknown (void);
 
@@ -206,14 +207,14 @@ extern void __put_user_unknown (void);
  * Evaluating arguments X, PTR, SIZE, and SEGMENT may involve subroutine-calls, which
  * could clobber r8 (among others).  Thus, be careful not to evaluate them while using r8.
  */
-#define __do_put_user(check, x, ptr, size, segment)					\
+#define __do_put_user(check, x, ptr, size)						\
 ({											\
 	__typeof__ (x) __pu_x = (x);							\
 	__typeof__ (*(ptr)) __user *__pu_ptr = (ptr);					\
 	__typeof__ (size) __pu_size = (size);						\
 	long __pu_err = -EFAULT;							\
 											\
-	if (!check || __access_ok(__pu_ptr, __pu_size, segment))			\
+	if (!check || __access_ok(__pu_ptr, __pu_size))					\
 		switch (__pu_size) {							\
 		      case 1: __put_user_size(__pu_x, __pu_ptr, 1, __pu_err); break;	\
 		      case 2: __put_user_size(__pu_x, __pu_ptr, 2, __pu_err); break;	\
@@ -224,8 +225,8 @@ extern void __put_user_unknown (void);
 	__pu_err;									\
 })
 
-#define __put_user_nocheck(x, ptr, size)	__do_put_user(0, x, ptr, size, KERNEL_DS)
-#define __put_user_check(x, ptr, size, segment)	__do_put_user(1, x, ptr, size, segment)
+#define __put_user_nocheck(x, ptr, size)	__do_put_user(0, x, ptr, size)
+#define __put_user_check(x, ptr, size)	__do_put_user(1, x, ptr, size)
 
 /*
  * Complex access routines
@@ -257,7 +258,7 @@ __copy_from_user (void *to, const void __user *from, unsigned long count)
 	const void *__cu_from = (from);							\
 	long __cu_len = (n);								\
 											\
-	if (__access_ok(__cu_to, __cu_len, get_fs())) {					\
+	if (__access_ok(__cu_to, __cu_len)) {						\
 		check_object_size(__cu_from, __cu_len, true);			\
 		__cu_len = __copy_user(__cu_to, (__force void __user *)  __cu_from, __cu_len);	\
 	}										\
@@ -268,23 +269,13 @@ static inline unsigned long
 copy_from_user(void *to, const void __user *from, unsigned long n)
 {
 	check_object_size(to, n, false);
-	if (likely(__access_ok(from, n, get_fs())))
+	if (likely(__access_ok(from, n)))
 		n = __copy_user((__force void __user *) to, from, n);
 	else
 		memset(to, 0, n);
 	return n;
 }
 
-#define __copy_in_user(to, from, size)	__copy_user((to), (from), (size))
-
-static inline unsigned long
-copy_in_user (void __user *to, const void __user *from, unsigned long n)
-{
-	if (likely(access_ok(VERIFY_READ, from, n) && access_ok(VERIFY_WRITE, to, n)))
-		n = __copy_user(to, from, n);
-	return n;
-}
-
 extern unsigned long __do_clear_user (void __user *, unsigned long);
 
 #define __clear_user(to, n)		__do_clear_user(to, n)
@@ -292,7 +283,7 @@ extern unsigned long __do_clear_user (void __user *, unsigned long);
 #define clear_user(to, n)					\
 ({								\
 	unsigned long __cu_len = (n);				\
-	if (__access_ok(to, __cu_len, get_fs()))		\
+	if (__access_ok(to, __cu_len))				\
 		__cu_len = __do_clear_user(to, __cu_len);	\
 	__cu_len;						\
 })
@@ -308,7 +299,7 @@ extern long __must_check __strncpy_from_user (char *to, const char __user *from,
 ({									\
 	const char __user * __sfu_from = (from);			\
 	long __sfu_ret = -EFAULT;					\
-	if (__access_ok(__sfu_from, 0, get_fs()))			\
+	if (__access_ok(__sfu_from, 0))					\
 		__sfu_ret = __strncpy_from_user((to), __sfu_from, (n));	\
 	__sfu_ret;							\
 })
@@ -320,7 +311,7 @@ extern unsigned long __strlen_user (const char __user *);
 ({							\
 	const char __user *__su_str = (str);		\
 	unsigned long __su_ret = 0;			\
-	if (__access_ok(__su_str, 0, get_fs()))		\
+	if (__access_ok(__su_str, 0))			\
 		__su_ret = __strlen_user(__su_str);	\
 	__su_ret;					\
 })
@@ -336,18 +327,11 @@ extern unsigned long __strnlen_user (const char __user *, long);
 ({								\
 	const char __user *__su_str = (str);			\
 	unsigned long __su_ret = 0;				\
-	if (__access_ok(__su_str, 0, get_fs()))			\
+	if (__access_ok(__su_str, 0))				\
 		__su_ret = __strnlen_user(__su_str, len);	\
 	__su_ret;						\
 })
 
-#define ARCH_HAS_RELATIVE_EXTABLE
-
-struct exception_table_entry {
-	int insn;	/* location-relative address of insn this fixup is for */
-	int fixup;	/* location-relative continuation addr.; if bit 2 is set, r9 is set to 0 */
-};
-
 #define ARCH_HAS_TRANSLATE_MEM_PTR	1
 static __inline__ void *
 xlate_dev_mem_ptr(phys_addr_t p)
diff --git a/arch/ia64/mm/extable.c b/arch/ia64/mm/extable.c
index 4edb816..8d1c45d 100644
--- a/arch/ia64/mm/extable.c
+++ b/arch/ia64/mm/extable.c
@@ -5,7 +5,9 @@
  *	David Mosberger-Tang <davidm@hpl.hp.com>
  */
 
-#include <linux/uaccess.h>
+#include <asm/ptrace.h>
+#include <asm/extable.h>
+#include <asm/errno.h>
 
 void
 ia64_handle_exception (struct pt_regs *regs, const struct exception_table_entry *e)
diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig
index 9547446..b3e82bd 100644
--- a/arch/m32r/Kconfig
+++ b/arch/m32r/Kconfig
@@ -19,6 +19,7 @@
 	select HAVE_DEBUG_STACKOVERFLOW
 	select CPU_NO_EFFICIENT_FFS
 	select DMA_NOOP_OPS
+	select ARCH_HAS_RAW_COPY_USER
 
 config SBUS
 	bool
diff --git a/arch/m32r/include/asm/Kbuild b/arch/m32r/include/asm/Kbuild
index deb2987..c000ffa 100644
--- a/arch/m32r/include/asm/Kbuild
+++ b/arch/m32r/include/asm/Kbuild
@@ -2,6 +2,7 @@
 generic-y += clkdev.h
 generic-y += current.h
 generic-y += exec.h
+generic-y += extable.h
 generic-y += irq_work.h
 generic-y += kvm_para.h
 generic-y += mcs_spinlock.h
diff --git a/arch/m32r/include/asm/uaccess.h b/arch/m32r/include/asm/uaccess.h
index 96b0efd..16c155d 100644
--- a/arch/m32r/include/asm/uaccess.h
+++ b/arch/m32r/include/asm/uaccess.h
@@ -13,6 +13,7 @@
  */
 #include <asm/page.h>
 #include <asm/setup.h>
+#include <linux/prefetch.h>
 
 /*
  * The fs value determines whether argument validity checking should be
@@ -109,25 +110,7 @@ static inline int access_ok(int type, const void *addr, unsigned long size)
 }
 #endif /* CONFIG_MMU */
 
-/*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
-
-struct exception_table_entry
-{
-	unsigned long insn, fixup;
-};
-
-extern int fixup_exception(struct pt_regs *regs);
+#include <asm/extable.h>
 
 /*
  * These are the main single-value transfer routines.  They automatically
@@ -478,173 +461,24 @@ do {									\
 		: "r14", "memory");					\
 } while (0)
 
-#define __copy_user_zeroing(to, from, size)				\
-do {									\
-	unsigned long __dst, __src, __c;				\
-	__asm__ __volatile__ (						\
-		"	mv	r14, %0\n"				\
-		"	or	r14, %1\n"				\
-		"	beq	%0, %1, 9f\n"				\
-		"	beqz	%2, 9f\n"				\
-		"	and3	r14, r14, #3\n"				\
-		"	bnez	r14, 2f\n"				\
-		"	and3	%2, %2, #3\n"				\
-		"	beqz	%3, 2f\n"				\
-		"	addi	%0, #-4		; word_copy \n"		\
-		"	.fillinsn\n"					\
-		"0:	ld	r14, @%1+\n"				\
-		"	addi	%3, #-1\n"				\
-		"	.fillinsn\n"					\
-		"1:	st	r14, @+%0\n"				\
-		"	bnez	%3, 0b\n"				\
-		"	beqz	%2, 9f\n"				\
-		"	addi	%0, #4\n"				\
-		"	.fillinsn\n"					\
-		"2:	ldb	r14, @%1	; byte_copy \n"		\
-		"	.fillinsn\n"					\
-		"3:	stb	r14, @%0\n"				\
-		"	addi	%1, #1\n"				\
-		"	addi	%2, #-1\n"				\
-		"	addi	%0, #1\n"				\
-		"	bnez	%2, 2b\n"				\
-		"	.fillinsn\n"					\
-		"9:\n"							\
-		".section .fixup,\"ax\"\n"				\
-		"	.balign 4\n"					\
-		"5:	addi	%3, #1\n"				\
-		"	addi	%1, #-4\n"				\
-		"	.fillinsn\n"					\
-		"6:	slli	%3, #2\n"				\
-		"	add	%2, %3\n"				\
-		"	addi	%0, #4\n"				\
-		"	.fillinsn\n"					\
-		"7:	ldi	r14, #0		; store zero \n"	\
-		"	.fillinsn\n"					\
-		"8:	addi	%2, #-1\n"				\
-		"	stb	r14, @%0	; ACE? \n"		\
-		"	addi	%0, #1\n"				\
-		"	bnez	%2, 8b\n"				\
-		"	seth	r14, #high(9b)\n"			\
-		"	or3	r14, r14, #low(9b)\n"			\
-		"	jmp	r14\n"					\
-		".previous\n"						\
-		".section __ex_table,\"a\"\n"				\
-		"	.balign 4\n"					\
-		"	.long 0b,6b\n"					\
-		"	.long 1b,5b\n"					\
-		"	.long 2b,7b\n"					\
-		"	.long 3b,7b\n"					\
-		".previous\n"						\
-		: "=&r" (__dst), "=&r" (__src), "=&r" (size),		\
-		  "=&r" (__c)						\
-		: "0" (to), "1" (from), "2" (size), "3" (size / 4)	\
-		: "r14", "memory");					\
-} while (0)
-
-
 /* We let the __ versions of copy_from/to_user inline, because they're often
  * used in fast paths and have only a small space overhead.
  */
-static inline unsigned long __generic_copy_from_user_nocheck(void *to,
-	const void __user *from, unsigned long n)
+static inline unsigned long
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
 {
-	__copy_user_zeroing(to, from, n);
-	return n;
-}
-
-static inline unsigned long __generic_copy_to_user_nocheck(void __user *to,
-	const void *from, unsigned long n)
-{
+	prefetchw(to);
 	__copy_user(to, from, n);
 	return n;
 }
 
-unsigned long __generic_copy_to_user(void __user *, const void *, unsigned long);
-unsigned long __generic_copy_from_user(void *, const void __user *, unsigned long);
-
-/**
- * __copy_to_user: - Copy a block of data into user space, with less checking.
- * @to:   Destination address, in user space.
- * @from: Source address, in kernel space.
- * @n:    Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from kernel space to user space.  Caller must check
- * the specified block with access_ok() before calling this function.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- */
-#define __copy_to_user(to, from, n)			\
-	__generic_copy_to_user_nocheck((to), (from), (n))
-
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
-
-/**
- * copy_to_user: - Copy a block of data into user space.
- * @to:   Destination address, in user space.
- * @from: Source address, in kernel space.
- * @n:    Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from kernel space to user space.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- */
-#define copy_to_user(to, from, n)			\
-({							\
-	might_fault();					\
-	__generic_copy_to_user((to), (from), (n));	\
-})
-
-/**
- * __copy_from_user: - Copy a block of data from user space, with less checking. * @to:   Destination address, in kernel space.
- * @from: Source address, in user space.
- * @n:    Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from user space to kernel space.  Caller must check
- * the specified block with access_ok() before calling this function.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- *
- * If some data could not be copied, this function will pad the copied
- * data to the requested size using zero bytes.
- */
-#define __copy_from_user(to, from, n)			\
-	__generic_copy_from_user_nocheck((to), (from), (n))
-
-/**
- * copy_from_user: - Copy a block of data from user space.
- * @to:   Destination address, in kernel space.
- * @from: Source address, in user space.
- * @n:    Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from user space to kernel space.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- *
- * If some data could not be copied, this function will pad the copied
- * data to the requested size using zero bytes.
- */
-#define copy_from_user(to, from, n)			\
-({							\
-	might_fault();					\
-	__generic_copy_from_user((to), (from), (n));	\
-})
+static inline unsigned long
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
+{
+	prefetchw(from);
+	__copy_user(to, from, n);
+	return n;
+}
 
 long __must_check strncpy_from_user(char *dst, const char __user *src,
 				long count);
diff --git a/arch/m32r/lib/usercopy.c b/arch/m32r/lib/usercopy.c
index fd03f27..b3ef2c8 100644
--- a/arch/m32r/lib/usercopy.c
+++ b/arch/m32r/lib/usercopy.c
@@ -11,27 +11,6 @@
 #include <linux/thread_info.h>
 #include <linux/uaccess.h>
 
-unsigned long
-__generic_copy_to_user(void __user *to, const void *from, unsigned long n)
-{
-	prefetch(from);
-	if (access_ok(VERIFY_WRITE, to, n))
-		__copy_user(to,from,n);
-	return n;
-}
-
-unsigned long
-__generic_copy_from_user(void *to, const void __user *from, unsigned long n)
-{
-	prefetchw(to);
-	if (access_ok(VERIFY_READ, from, n))
-		__copy_user_zeroing(to,from,n);
-	else
-		memset(to, 0, n);
-	return n;
-}
-
-
 /*
  * Copy a null terminated string from userspace.
  */
diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig
index d140206..7d34575 100644
--- a/arch/m68k/Kconfig
+++ b/arch/m68k/Kconfig
@@ -22,6 +22,7 @@
 	select MODULES_USE_ELF_RELA
 	select OLD_SIGSUSPEND3
 	select OLD_SIGACTION
+	select ARCH_HAS_RAW_COPY_USER
 
 config RWSEM_GENERIC_SPINLOCK
 	bool
diff --git a/arch/m68k/include/asm/Kbuild b/arch/m68k/include/asm/Kbuild
index d4f9ccb..82005d2 100644
--- a/arch/m68k/include/asm/Kbuild
+++ b/arch/m68k/include/asm/Kbuild
@@ -5,6 +5,7 @@
 generic-y += emergency-restart.h
 generic-y += errno.h
 generic-y += exec.h
+generic-y += extable.h
 generic-y += futex.h
 generic-y += hw_irq.h
 generic-y += ioctl.h
diff --git a/arch/m68k/include/asm/processor.h b/arch/m68k/include/asm/processor.h
index f5f790c3..77239e8 100644
--- a/arch/m68k/include/asm/processor.h
+++ b/arch/m68k/include/asm/processor.h
@@ -122,16 +122,6 @@ static inline void start_thread(struct pt_regs * regs, unsigned long pc,
 	wrusp(usp);
 }
 
-#ifdef CONFIG_MMU
-extern int handle_kernel_fault(struct pt_regs *regs);
-#else
-static inline  int handle_kernel_fault(struct pt_regs *regs)
-{
-	/* Any fault in kernel is fatal on non-mmu */
-	return 0;
-}
-#endif
-
 /* Forward declaration, a strange C thing */
 struct task_struct;
 
diff --git a/arch/m68k/include/asm/uaccess.h b/arch/m68k/include/asm/uaccess.h
index 3fadc4a..67b3481 100644
--- a/arch/m68k/include/asm/uaccess.h
+++ b/arch/m68k/include/asm/uaccess.h
@@ -4,6 +4,7 @@
 #include <asm/uaccess_mm.h>
 #endif
 
+#include <asm/extable.h>
 #ifdef CONFIG_CPU_HAS_NO_UNALIGNED
 #include <asm-generic/uaccess-unaligned.h>
 #else
diff --git a/arch/m68k/include/asm/uaccess_mm.h b/arch/m68k/include/asm/uaccess_mm.h
index 14054a4..ef856ff 100644
--- a/arch/m68k/include/asm/uaccess_mm.h
+++ b/arch/m68k/include/asm/uaccess_mm.h
@@ -31,24 +31,6 @@ static inline int access_ok(int type, const void __user *addr,
 #define	MOVES	"move"
 #endif
 
-/*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
-
-struct exception_table_entry
-{
-	unsigned long insn, fixup;
-};
-
 extern int __put_user_bad(void);
 extern int __get_user_bad(void);
 
@@ -197,39 +179,55 @@ asm volatile ("\n"					\
 unsigned long __generic_copy_from_user(void *to, const void __user *from, unsigned long n);
 unsigned long __generic_copy_to_user(void __user *to, const void *from, unsigned long n);
 
-#define __constant_copy_from_user_asm(res, to, from, tmp, n, s1, s2, s3)\
+#define __suffix0
+#define __suffix1 b
+#define __suffix2 w
+#define __suffix4 l
+
+#define ____constant_copy_from_user_asm(res, to, from, tmp, n1, n2, n3, s1, s2, s3)\
 	asm volatile ("\n"						\
 		"1:	"MOVES"."#s1"	(%2)+,%3\n"			\
 		"	move."#s1"	%3,(%1)+\n"			\
+		"	.ifnc	\""#s2"\",\"\"\n"			\
 		"2:	"MOVES"."#s2"	(%2)+,%3\n"			\
 		"	move."#s2"	%3,(%1)+\n"			\
 		"	.ifnc	\""#s3"\",\"\"\n"			\
 		"3:	"MOVES"."#s3"	(%2)+,%3\n"			\
 		"	move."#s3"	%3,(%1)+\n"			\
 		"	.endif\n"					\
+		"	.endif\n"					\
 		"4:\n"							\
 		"	.section __ex_table,\"a\"\n"			\
 		"	.align	4\n"					\
 		"	.long	1b,10f\n"				\
+		"	.ifnc	\""#s2"\",\"\"\n"			\
 		"	.long	2b,20f\n"				\
 		"	.ifnc	\""#s3"\",\"\"\n"			\
 		"	.long	3b,30f\n"				\
 		"	.endif\n"					\
+		"	.endif\n"					\
 		"	.previous\n"					\
 		"\n"							\
 		"	.section .fixup,\"ax\"\n"			\
 		"	.even\n"					\
-		"10:	clr."#s1"	(%1)+\n"			\
-		"20:	clr."#s2"	(%1)+\n"			\
+		"10:	addq.l #"#n1",%0\n"				\
+		"	.ifnc	\""#s2"\",\"\"\n"			\
+		"20:	addq.l #"#n2",%0\n"				\
 		"	.ifnc	\""#s3"\",\"\"\n"			\
-		"30:	clr."#s3"	(%1)+\n"			\
+		"30:	addq.l #"#n3",%0\n"				\
 		"	.endif\n"					\
-		"	moveq.l	#"#n",%0\n"				\
+		"	.endif\n"					\
 		"	jra	4b\n"					\
 		"	.previous\n"					\
 		: "+d" (res), "+&a" (to), "+a" (from), "=&d" (tmp)	\
 		: : "memory")
 
+#define ___constant_copy_from_user_asm(res, to, from, tmp, n1, n2, n3, s1, s2, s3)\
+	____constant_copy_from_user_asm(res, to, from, tmp, n1, n2, n3, s1, s2, s3)
+#define __constant_copy_from_user_asm(res, to, from, tmp, n1, n2, n3)	\
+	___constant_copy_from_user_asm(res, to, from, tmp, n1, n2, n3,  \
+					__suffix##n1, __suffix##n2, __suffix##n3)
+
 static __always_inline unsigned long
 __constant_copy_from_user(void *to, const void __user *from, unsigned long n)
 {
@@ -237,37 +235,37 @@ __constant_copy_from_user(void *to, const void __user *from, unsigned long n)
 
 	switch (n) {
 	case 1:
-		__get_user_asm(res, *(u8 *)to, (u8 __user *)from, u8, b, d, 1);
+		__constant_copy_from_user_asm(res, to, from, tmp, 1, 0, 0);
 		break;
 	case 2:
-		__get_user_asm(res, *(u16 *)to, (u16 __user *)from, u16, w, r, 2);
+		__constant_copy_from_user_asm(res, to, from, tmp, 2, 0, 0);
 		break;
 	case 3:
-		__constant_copy_from_user_asm(res, to, from, tmp, 3, w, b,);
+		__constant_copy_from_user_asm(res, to, from, tmp, 2, 1, 0);
 		break;
 	case 4:
-		__get_user_asm(res, *(u32 *)to, (u32 __user *)from, u32, l, r, 4);
+		__constant_copy_from_user_asm(res, to, from, tmp, 4, 0, 0);
 		break;
 	case 5:
-		__constant_copy_from_user_asm(res, to, from, tmp, 5, l, b,);
+		__constant_copy_from_user_asm(res, to, from, tmp, 4, 1, 0);
 		break;
 	case 6:
-		__constant_copy_from_user_asm(res, to, from, tmp, 6, l, w,);
+		__constant_copy_from_user_asm(res, to, from, tmp, 4, 2, 0);
 		break;
 	case 7:
-		__constant_copy_from_user_asm(res, to, from, tmp, 7, l, w, b);
+		__constant_copy_from_user_asm(res, to, from, tmp, 4, 2, 1);
 		break;
 	case 8:
-		__constant_copy_from_user_asm(res, to, from, tmp, 8, l, l,);
+		__constant_copy_from_user_asm(res, to, from, tmp, 4, 4, 0);
 		break;
 	case 9:
-		__constant_copy_from_user_asm(res, to, from, tmp, 9, l, l, b);
+		__constant_copy_from_user_asm(res, to, from, tmp, 4, 4, 1);
 		break;
 	case 10:
-		__constant_copy_from_user_asm(res, to, from, tmp, 10, l, l, w);
+		__constant_copy_from_user_asm(res, to, from, tmp, 4, 4, 2);
 		break;
 	case 12:
-		__constant_copy_from_user_asm(res, to, from, tmp, 12, l, l, l);
+		__constant_copy_from_user_asm(res, to, from, tmp, 4, 4, 4);
 		break;
 	default:
 		/* we limit the inlined version to 3 moves */
@@ -358,21 +356,23 @@ __constant_copy_to_user(void __user *to, const void *from, unsigned long n)
 	return res;
 }
 
-#define __copy_from_user(to, from, n)		\
-(__builtin_constant_p(n) ?			\
- __constant_copy_from_user(to, from, n) :	\
- __generic_copy_from_user(to, from, n))
+static inline unsigned long
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
+{
+	if (__builtin_constant_p(n))
+		return __constant_copy_from_user(to, from, n);
+	return __generic_copy_from_user(to, from, n);
+}
 
-#define __copy_to_user(to, from, n)		\
-(__builtin_constant_p(n) ?			\
- __constant_copy_to_user(to, from, n) :		\
- __generic_copy_to_user(to, from, n))
-
-#define __copy_to_user_inatomic		__copy_to_user
-#define __copy_from_user_inatomic	__copy_from_user
-
-#define copy_from_user(to, from, n)	__copy_from_user(to, from, n)
-#define copy_to_user(to, from, n)	__copy_to_user(to, from, n)
+static inline unsigned long
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
+{
+	if (__builtin_constant_p(n))
+		return __constant_copy_to_user(to, from, n);
+	return __generic_copy_to_user(to, from, n);
+}
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 #define user_addr_max() \
 	(uaccess_kernel() ? ~0UL : TASK_SIZE)
diff --git a/arch/m68k/include/asm/uaccess_no.h b/arch/m68k/include/asm/uaccess_no.h
index e77ce66..e482c38 100644
--- a/arch/m68k/include/asm/uaccess_no.h
+++ b/arch/m68k/include/asm/uaccess_no.h
@@ -23,25 +23,6 @@ static inline int _access_ok(unsigned long addr, unsigned long size)
 }
 
 /*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
-
-struct exception_table_entry
-{
-	unsigned long insn, fixup;
-};
-
-
-/*
  * These are the main single-value transfer routines.  They automatically
  * use the right size if we just have the right pointer type.
  */
@@ -120,13 +101,21 @@ extern int __get_user_bad(void);
 		 : "=d" (x)					\
 		 : "m" (*__ptr(ptr)))
 
-#define copy_from_user(to, from, n)		(memcpy(to, from, n), 0)
-#define copy_to_user(to, from, n)		(memcpy(to, from, n), 0)
+static inline unsigned long
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
+{
+	memcpy(to, (__force const void *)from, n);
+	return 0;
+}
 
-#define __copy_from_user(to, from, n) copy_from_user(to, from, n)
-#define __copy_to_user(to, from, n) copy_to_user(to, from, n)
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
+static inline unsigned long
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
+{
+	memcpy((__force void *)to, from, n);
+	return 0;
+}
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 /*
  * Copy a null terminated string from userspace.
diff --git a/arch/m68k/kernel/signal.c b/arch/m68k/kernel/signal.c
index 093b7c4..6f945bb 100644
--- a/arch/m68k/kernel/signal.c
+++ b/arch/m68k/kernel/signal.c
@@ -88,7 +88,7 @@ static inline int frame_extra_sizes(int f)
 	return frame_size_change[f];
 }
 
-int handle_kernel_fault(struct pt_regs *regs)
+int fixup_exception(struct pt_regs *regs)
 {
 	const struct exception_table_entry *fixup;
 	struct pt_regs *tregs;
diff --git a/arch/m68k/kernel/traps.c b/arch/m68k/kernel/traps.c
index a926d2c..c1cc4e9 100644
--- a/arch/m68k/kernel/traps.c
+++ b/arch/m68k/kernel/traps.c
@@ -1016,8 +1016,13 @@ asmlinkage void trap_c(struct frame *fp)
 			/* traced a trapping instruction on a 68020/30,
 			 * real exception will be executed afterwards.
 			 */
-		} else if (!handle_kernel_fault(&fp->ptregs))
-			bad_super_trap(fp);
+			return;
+		}
+#ifdef CONFIG_MMU
+		if (fixup_exception(&fp->ptregs))
+			return;
+#endif
+		bad_super_trap(fp);
 		return;
 	}
 
diff --git a/arch/m68k/lib/uaccess.c b/arch/m68k/lib/uaccess.c
index a76b73a..7646e46 100644
--- a/arch/m68k/lib/uaccess.c
+++ b/arch/m68k/lib/uaccess.c
@@ -30,19 +30,13 @@ unsigned long __generic_copy_from_user(void *to, const void __user *from,
 		"6:\n"
 		"	.section .fixup,\"ax\"\n"
 		"	.even\n"
-		"10:	move.l	%0,%3\n"
-		"7:	clr.l	(%2)+\n"
-		"	subq.l	#1,%3\n"
-		"	jne	7b\n"
-		"	lsl.l	#2,%0\n"
+		"10:	lsl.l	#2,%0\n"
 		"	btst	#1,%5\n"
 		"	jeq	8f\n"
-		"30:	clr.w	(%2)+\n"
-		"	addq.l	#2,%0\n"
+		"30:	addq.l	#2,%0\n"
 		"8:	btst	#0,%5\n"
 		"	jeq	6b\n"
-		"50:	clr.b	(%2)+\n"
-		"	addq.l	#1,%0\n"
+		"50:	addq.l	#1,%0\n"
 		"	jra	6b\n"
 		"	.previous\n"
 		"\n"
diff --git a/arch/m68k/mm/fault.c b/arch/m68k/mm/fault.c
index bd66a0b..2795e4c 100644
--- a/arch/m68k/mm/fault.c
+++ b/arch/m68k/mm/fault.c
@@ -32,7 +32,7 @@ int send_fault_sig(struct pt_regs *regs)
 		force_sig_info(siginfo.si_signo,
 			       &siginfo, current);
 	} else {
-		if (handle_kernel_fault(regs))
+		if (fixup_exception(regs))
 			return -1;
 
 		//if (siginfo.si_signo == SIGBUS)
diff --git a/arch/metag/include/asm/Kbuild b/arch/metag/include/asm/Kbuild
index f9b9df5..8f940553 100644
--- a/arch/metag/include/asm/Kbuild
+++ b/arch/metag/include/asm/Kbuild
@@ -8,6 +8,7 @@
 generic-y += emergency-restart.h
 generic-y += errno.h
 generic-y += exec.h
+generic-y += extable.h
 generic-y += fb.h
 generic-y += fcntl.h
 generic-y += futex.h
diff --git a/arch/metag/include/asm/uaccess.h b/arch/metag/include/asm/uaccess.h
index 83de755..2613d91 100644
--- a/arch/metag/include/asm/uaccess.h
+++ b/arch/metag/include/asm/uaccess.h
@@ -47,28 +47,7 @@ static inline int __access_ok(unsigned long addr, unsigned long size)
 #define access_ok(type, addr, size) __access_ok((unsigned long)(addr),	\
 						(unsigned long)(size))
 
-static inline int verify_area(int type, const void *addr, unsigned long size)
-{
-	return access_ok(type, addr, size) ? 0 : -EFAULT;
-}
-
-/*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
-struct exception_table_entry {
-	unsigned long insn, fixup;
-};
-
-extern int fixup_exception(struct pt_regs *regs);
+#include <asm/extable.h>
 
 /*
  * These are the main single-value transfer routines.  They automatically
diff --git a/arch/microblaze/Kconfig b/arch/microblaze/Kconfig
index 85885a50..1aff365 100644
--- a/arch/microblaze/Kconfig
+++ b/arch/microblaze/Kconfig
@@ -34,6 +34,7 @@
 	select TRACING_SUPPORT
 	select VIRT_TO_BUS
 	select CPU_NO_EFFICIENT_FFS
+	select ARCH_HAS_RAW_COPY_USER
 
 config SWAP
 	def_bool n
diff --git a/arch/microblaze/include/asm/Kbuild b/arch/microblaze/include/asm/Kbuild
index 1732ec1..56830ff 100644
--- a/arch/microblaze/include/asm/Kbuild
+++ b/arch/microblaze/include/asm/Kbuild
@@ -3,6 +3,7 @@
 generic-y += clkdev.h
 generic-y += device.h
 generic-y += exec.h
+generic-y += extable.h
 generic-y += irq_work.h
 generic-y += mcs_spinlock.h
 generic-y += mm-arch-hooks.h
diff --git a/arch/microblaze/include/asm/uaccess.h b/arch/microblaze/include/asm/uaccess.h
index b132cd3..fe1712a 100644
--- a/arch/microblaze/include/asm/uaccess.h
+++ b/arch/microblaze/include/asm/uaccess.h
@@ -17,6 +17,7 @@
 #include <asm/mmu.h>
 #include <asm/page.h>
 #include <asm/pgtable.h>
+#include <asm/extable.h>
 #include <linux/string.h>
 
 /*
@@ -47,22 +48,6 @@
 
 # define segment_eq(a, b)	((a).seg == (b).seg)
 
-/*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue. No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path. This means when everything is well,
- * we don't even have to jump over them. Further, they do not intrude
- * on our cache or tlb entries.
- */
-struct exception_table_entry {
-	unsigned long insn, fixup;
-};
-
 #ifndef CONFIG_MMU
 
 /* Check against bounds of physical memory */
@@ -353,37 +338,19 @@ extern long __user_bad(void);
 
 
 /* copy_to_from_user */
-#define __copy_from_user(to, from, n)	\
-	__copy_tofrom_user((__force void __user *)(to), \
-				(void __user *)(from), (n))
-#define __copy_from_user_inatomic(to, from, n) \
-		__copy_from_user((to), (from), (n))
-
-static inline long copy_from_user(void *to,
-		const void __user *from, unsigned long n)
+static inline unsigned long
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
 {
-	unsigned long res = n;
-	might_fault();
-	if (likely(access_ok(VERIFY_READ, from, n)))
-		res = __copy_from_user(to, from, n);
-	if (unlikely(res))
-		memset(to + (n - res), 0, res);
-	return res;
+	return __copy_tofrom_user((__force void __user *)to, from, n);
 }
 
-#define __copy_to_user(to, from, n)	\
-		__copy_tofrom_user((void __user *)(to), \
-			(__force const void __user *)(from), (n))
-#define __copy_to_user_inatomic(to, from, n) __copy_to_user((to), (from), (n))
-
-static inline long copy_to_user(void __user *to,
-		const void *from, unsigned long n)
+static inline unsigned long
+raw_copy_from_user(void __user *to, const void *from, unsigned long n)
 {
-	might_fault();
-	if (access_ok(VERIFY_WRITE, to, n))
-		return __copy_to_user(to, from, n);
-	return n;
+	return __copy_tofrom_user(to, (__force const void __user *)from, n);
 }
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 /*
  * Copy a null terminated string from userspace.
diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index a008a9f..aff5633 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -69,6 +69,7 @@
 	select HAVE_EXIT_THREAD
 	select HAVE_REGS_AND_STACK_ACCESS_API
 	select HAVE_ARCH_HARDENED_USERCOPY
+	select ARCH_HAS_RAW_COPY_USER
 
 menu "Machine selection"
 
diff --git a/arch/mips/cavium-octeon/cpu.c b/arch/mips/cavium-octeon/cpu.c
index a5b4279..036d56c 100644
--- a/arch/mips/cavium-octeon/cpu.c
+++ b/arch/mips/cavium-octeon/cpu.c
@@ -10,7 +10,9 @@
 #include <linux/irqflags.h>
 #include <linux/notifier.h>
 #include <linux/prefetch.h>
+#include <linux/ptrace.h>
 #include <linux/sched.h>
+#include <linux/sched/task_stack.h>
 
 #include <asm/cop2.h>
 #include <asm/current.h>
diff --git a/arch/mips/cavium-octeon/crypto/octeon-crypto.c b/arch/mips/cavium-octeon/crypto/octeon-crypto.c
index 4d22365..cfb4a14 100644
--- a/arch/mips/cavium-octeon/crypto/octeon-crypto.c
+++ b/arch/mips/cavium-octeon/crypto/octeon-crypto.c
@@ -9,6 +9,7 @@
 #include <asm/cop2.h>
 #include <linux/export.h>
 #include <linux/interrupt.h>
+#include <linux/sched/task_stack.h>
 
 #include "octeon-crypto.h"
 
diff --git a/arch/mips/cavium-octeon/octeon-memcpy.S b/arch/mips/cavium-octeon/octeon-memcpy.S
index cfd97f6..0a7c983 100644
--- a/arch/mips/cavium-octeon/octeon-memcpy.S
+++ b/arch/mips/cavium-octeon/octeon-memcpy.S
@@ -140,15 +140,6 @@
 	.set	noat
 
 /*
- * t7 is used as a flag to note inatomic mode.
- */
-LEAF(__copy_user_inatomic)
-EXPORT_SYMBOL(__copy_user_inatomic)
-	b	__copy_user_common
-	 li	t7, 1
-	END(__copy_user_inatomic)
-
-/*
  * A combined memcpy/__copy_user
  * __copy_user sets len to 0 for success; else to an upper bound of
  * the number of uncopied bytes.
@@ -161,8 +152,6 @@
 __memcpy:
 FEXPORT(__copy_user)
 EXPORT_SYMBOL(__copy_user)
-	li	t7, 0				/* not inatomic */
-__copy_user_common:
 	/*
 	 * Note: dst & src may be unaligned, len may be 0
 	 * Temps
@@ -414,25 +403,7 @@
 	LOAD	t0, TI_TASK($28)
 	LOAD	t0, THREAD_BUADDR(t0)	# t0 is just past last good address
 	SUB	len, AT, t0		# len number of uncopied bytes
-	bnez	t7, 2f		/* Skip the zeroing out part if inatomic */
-	/*
-	 * Here's where we rely on src and dst being incremented in tandem,
-	 *   See (3) above.
-	 * dst += (fault addr - src) to put dst at first byte to clear
-	 */
-	ADD	dst, t0			# compute start address in a1
-	SUB	dst, src
-	/*
-	 * Clear len bytes starting at dst.  Can't call __bzero because it
-	 * might modify len.  An inefficient loop for these rare times...
-	 */
-	beqz	len, done
-	 SUB	src, len, 1
-1:	sb	zero, 0(dst)
-	ADD	dst, dst, 1
-	bnez	src, 1b
-	 SUB	src, src, 1
-2:	jr	ra
+	jr	ra
 	 nop
 
 
diff --git a/arch/mips/cavium-octeon/smp.c b/arch/mips/cavium-octeon/smp.c
index 4b94b7f..3de7865 100644
--- a/arch/mips/cavium-octeon/smp.c
+++ b/arch/mips/cavium-octeon/smp.c
@@ -12,6 +12,7 @@
 #include <linux/kernel_stat.h>
 #include <linux/sched.h>
 #include <linux/sched/hotplug.h>
+#include <linux/sched/task_stack.h>
 #include <linux/init.h>
 #include <linux/export.h>
 
diff --git a/arch/mips/include/asm/fpu.h b/arch/mips/include/asm/fpu.h
index 321752b..f94455f 100644
--- a/arch/mips/include/asm/fpu.h
+++ b/arch/mips/include/asm/fpu.h
@@ -12,6 +12,7 @@
 
 #include <linux/sched.h>
 #include <linux/sched/task_stack.h>
+#include <linux/ptrace.h>
 #include <linux/thread_info.h>
 #include <linux/bitops.h>
 
diff --git a/arch/mips/include/asm/uaccess.h b/arch/mips/include/asm/uaccess.h
index 70ca8ee..99e629a 100644
--- a/arch/mips/include/asm/uaccess.h
+++ b/arch/mips/include/asm/uaccess.h
@@ -128,23 +128,14 @@ static inline bool eva_kernel_access(void)
  * this function, memory access functions may still return -EFAULT.
  */
 
-#define __access_mask get_fs().seg
-
-#define __access_ok(addr, size, mask)					\
-({									\
-	unsigned long __addr = (unsigned long) (addr);			\
-	unsigned long __size = size;					\
-	unsigned long __mask = mask;					\
-	unsigned long __ok;						\
-									\
-	__chk_user_ptr(addr);						\
-	__ok = (signed long)(__mask & (__addr | (__addr + __size) |	\
-		__ua_size(__size)));					\
-	__ok == 0;							\
-})
+static inline int __access_ok(const void __user *p, unsigned long size)
+{
+	unsigned long addr = (unsigned long)p;
+	return (get_fs().seg & (addr | (addr + size) | __ua_size(size))) == 0;
+}
 
 #define access_ok(type, addr, size)					\
-	likely(__access_ok((addr), (size), __access_mask))
+	likely(__access_ok((addr), (size)))
 
 /*
  * put_user: - Write a simple value into user space.
@@ -806,8 +797,30 @@ extern void __put_user_unaligned_unknown(void);
 
 extern size_t __copy_user(void *__to, const void *__from, size_t __n);
 
-#ifndef CONFIG_EVA
-#define __invoke_copy_to_user(to, from, n)				\
+#define __invoke_copy_from(func, to, from, n)				\
+({									\
+	register void *__cu_to_r __asm__("$4");				\
+	register const void __user *__cu_from_r __asm__("$5");		\
+	register long __cu_len_r __asm__("$6");				\
+									\
+	__cu_to_r = (to);						\
+	__cu_from_r = (from);						\
+	__cu_len_r = (n);						\
+	__asm__ __volatile__(						\
+	".set\tnoreorder\n\t"						\
+	__MODULE_JAL(func)						\
+	".set\tnoat\n\t"						\
+	__UA_ADDU "\t$1, %1, %2\n\t"					\
+	".set\tat\n\t"							\
+	".set\treorder"							\
+	: "+r" (__cu_to_r), "+r" (__cu_from_r), "+r" (__cu_len_r)	\
+	:								\
+	: "$8", "$9", "$10", "$11", "$12", "$14", "$15", "$24", "$31",	\
+	  DADDI_SCRATCH, "memory");					\
+	__cu_len_r;							\
+})
+
+#define __invoke_copy_to(func, to, from, n)				\
 ({									\
 	register void __user *__cu_to_r __asm__("$4");			\
 	register const void *__cu_from_r __asm__("$5");			\
@@ -817,161 +830,7 @@ extern size_t __copy_user(void *__to, const void *__from, size_t __n);
 	__cu_from_r = (from);						\
 	__cu_len_r = (n);						\
 	__asm__ __volatile__(						\
-	__MODULE_JAL(__copy_user)					\
-	: "+r" (__cu_to_r), "+r" (__cu_from_r), "+r" (__cu_len_r)	\
-	:								\
-	: "$8", "$9", "$10", "$11", "$12", "$14", "$15", "$24", "$31",	\
-	  DADDI_SCRATCH, "memory");					\
-	__cu_len_r;							\
-})
-
-#define __invoke_copy_to_kernel(to, from, n)				\
-	__invoke_copy_to_user(to, from, n)
-
-#endif
-
-/*
- * __copy_to_user: - Copy a block of data into user space, with less checking.
- * @to:	  Destination address, in user space.
- * @from: Source address, in kernel space.
- * @n:	  Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from kernel space to user space.  Caller must check
- * the specified block with access_ok() before calling this function.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- */
-#define __copy_to_user(to, from, n)					\
-({									\
-	void __user *__cu_to;						\
-	const void *__cu_from;						\
-	long __cu_len;							\
-									\
-	__cu_to = (to);							\
-	__cu_from = (from);						\
-	__cu_len = (n);							\
-									\
-	check_object_size(__cu_from, __cu_len, true);			\
-	might_fault();							\
-									\
-	if (eva_kernel_access())					\
-		__cu_len = __invoke_copy_to_kernel(__cu_to, __cu_from,	\
-						   __cu_len);		\
-	else								\
-		__cu_len = __invoke_copy_to_user(__cu_to, __cu_from,	\
-						 __cu_len);		\
-	__cu_len;							\
-})
-
-extern size_t __copy_user_inatomic(void *__to, const void *__from, size_t __n);
-
-#define __copy_to_user_inatomic(to, from, n)				\
-({									\
-	void __user *__cu_to;						\
-	const void *__cu_from;						\
-	long __cu_len;							\
-									\
-	__cu_to = (to);							\
-	__cu_from = (from);						\
-	__cu_len = (n);							\
-									\
-	check_object_size(__cu_from, __cu_len, true);			\
-									\
-	if (eva_kernel_access())					\
-		__cu_len = __invoke_copy_to_kernel(__cu_to, __cu_from,	\
-						   __cu_len);		\
-	else								\
-		__cu_len = __invoke_copy_to_user(__cu_to, __cu_from,	\
-						 __cu_len);		\
-	__cu_len;							\
-})
-
-#define __copy_from_user_inatomic(to, from, n)				\
-({									\
-	void *__cu_to;							\
-	const void __user *__cu_from;					\
-	long __cu_len;							\
-									\
-	__cu_to = (to);							\
-	__cu_from = (from);						\
-	__cu_len = (n);							\
-									\
-	check_object_size(__cu_to, __cu_len, false);			\
-									\
-	if (eva_kernel_access())					\
-		__cu_len = __invoke_copy_from_kernel_inatomic(__cu_to,	\
-							      __cu_from,\
-							      __cu_len);\
-	else								\
-		__cu_len = __invoke_copy_from_user_inatomic(__cu_to,	\
-							    __cu_from,	\
-							    __cu_len);	\
-	__cu_len;							\
-})
-
-/*
- * copy_to_user: - Copy a block of data into user space.
- * @to:	  Destination address, in user space.
- * @from: Source address, in kernel space.
- * @n:	  Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from kernel space to user space.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- */
-#define copy_to_user(to, from, n)					\
-({									\
-	void __user *__cu_to;						\
-	const void *__cu_from;						\
-	long __cu_len;							\
-									\
-	__cu_to = (to);							\
-	__cu_from = (from);						\
-	__cu_len = (n);							\
-									\
-	check_object_size(__cu_from, __cu_len, true);			\
-									\
-	if (eva_kernel_access()) {					\
-		__cu_len = __invoke_copy_to_kernel(__cu_to,		\
-						   __cu_from,		\
-						   __cu_len);		\
-	} else {							\
-		if (access_ok(VERIFY_WRITE, __cu_to, __cu_len)) {       \
-			might_fault();                                  \
-			__cu_len = __invoke_copy_to_user(__cu_to,	\
-							 __cu_from,	\
-							 __cu_len);     \
-		}							\
-	}								\
-	__cu_len;							\
-})
-
-#ifndef CONFIG_EVA
-
-#define __invoke_copy_from_user(to, from, n)				\
-({									\
-	register void *__cu_to_r __asm__("$4");				\
-	register const void __user *__cu_from_r __asm__("$5");		\
-	register long __cu_len_r __asm__("$6");				\
-									\
-	__cu_to_r = (to);						\
-	__cu_from_r = (from);						\
-	__cu_len_r = (n);						\
-	__asm__ __volatile__(						\
-	".set\tnoreorder\n\t"						\
-	__MODULE_JAL(__copy_user)					\
-	".set\tnoat\n\t"						\
-	__UA_ADDU "\t$1, %1, %2\n\t"					\
-	".set\tat\n\t"							\
-	".set\treorder"							\
+	__MODULE_JAL(func)						\
 	: "+r" (__cu_to_r), "+r" (__cu_from_r), "+r" (__cu_len_r)	\
 	:								\
 	: "$8", "$9", "$10", "$11", "$12", "$14", "$15", "$24", "$31",	\
@@ -980,261 +839,78 @@ extern size_t __copy_user_inatomic(void *__to, const void *__from, size_t __n);
 })
 
 #define __invoke_copy_from_kernel(to, from, n)				\
-	__invoke_copy_from_user(to, from, n)
+	__invoke_copy_from(__copy_user, to, from, n)
 
-/* For userland <-> userland operations */
-#define ___invoke_copy_in_user(to, from, n)				\
-	__invoke_copy_from_user(to, from, n)
+#define __invoke_copy_to_kernel(to, from, n)				\
+	__invoke_copy_to(__copy_user, to, from, n)
 
-/* For kernel <-> kernel operations */
 #define ___invoke_copy_in_kernel(to, from, n)				\
-	__invoke_copy_from_user(to, from, n)
+	__invoke_copy_from(__copy_user, to, from, n)
 
-#define __invoke_copy_from_user_inatomic(to, from, n)			\
-({									\
-	register void *__cu_to_r __asm__("$4");				\
-	register const void __user *__cu_from_r __asm__("$5");		\
-	register long __cu_len_r __asm__("$6");				\
-									\
-	__cu_to_r = (to);						\
-	__cu_from_r = (from);						\
-	__cu_len_r = (n);						\
-	__asm__ __volatile__(						\
-	".set\tnoreorder\n\t"						\
-	__MODULE_JAL(__copy_user_inatomic)				\
-	".set\tnoat\n\t"						\
-	__UA_ADDU "\t$1, %1, %2\n\t"					\
-	".set\tat\n\t"							\
-	".set\treorder"							\
-	: "+r" (__cu_to_r), "+r" (__cu_from_r), "+r" (__cu_len_r)	\
-	:								\
-	: "$8", "$9", "$10", "$11", "$12", "$14", "$15", "$24", "$31",	\
-	  DADDI_SCRATCH, "memory");					\
-	__cu_len_r;							\
-})
+#ifndef CONFIG_EVA
+#define __invoke_copy_from_user(to, from, n)				\
+	__invoke_copy_from(__copy_user, to, from, n)
 
-#define __invoke_copy_from_kernel_inatomic(to, from, n)			\
-	__invoke_copy_from_user_inatomic(to, from, n)			\
+#define __invoke_copy_to_user(to, from, n)				\
+	__invoke_copy_to(__copy_user, to, from, n)
+
+#define ___invoke_copy_in_user(to, from, n)				\
+	__invoke_copy_from(__copy_user, to, from, n)
 
 #else
 
 /* EVA specific functions */
 
-extern size_t __copy_user_inatomic_eva(void *__to, const void *__from,
-				       size_t __n);
 extern size_t __copy_from_user_eva(void *__to, const void *__from,
 				   size_t __n);
 extern size_t __copy_to_user_eva(void *__to, const void *__from,
 				 size_t __n);
 extern size_t __copy_in_user_eva(void *__to, const void *__from, size_t __n);
 
-#define __invoke_copy_from_user_eva_generic(to, from, n, func_ptr)	\
-({									\
-	register void *__cu_to_r __asm__("$4");				\
-	register const void __user *__cu_from_r __asm__("$5");		\
-	register long __cu_len_r __asm__("$6");				\
-									\
-	__cu_to_r = (to);						\
-	__cu_from_r = (from);						\
-	__cu_len_r = (n);						\
-	__asm__ __volatile__(						\
-	".set\tnoreorder\n\t"						\
-	__MODULE_JAL(func_ptr)						\
-	".set\tnoat\n\t"						\
-	__UA_ADDU "\t$1, %1, %2\n\t"					\
-	".set\tat\n\t"							\
-	".set\treorder"							\
-	: "+r" (__cu_to_r), "+r" (__cu_from_r), "+r" (__cu_len_r)	\
-	:								\
-	: "$8", "$9", "$10", "$11", "$12", "$14", "$15", "$24", "$31",	\
-	  DADDI_SCRATCH, "memory");					\
-	__cu_len_r;							\
-})
-
-#define __invoke_copy_to_user_eva_generic(to, from, n, func_ptr)	\
-({									\
-	register void *__cu_to_r __asm__("$4");				\
-	register const void __user *__cu_from_r __asm__("$5");		\
-	register long __cu_len_r __asm__("$6");				\
-									\
-	__cu_to_r = (to);						\
-	__cu_from_r = (from);						\
-	__cu_len_r = (n);						\
-	__asm__ __volatile__(						\
-	__MODULE_JAL(func_ptr)						\
-	: "+r" (__cu_to_r), "+r" (__cu_from_r), "+r" (__cu_len_r)	\
-	:								\
-	: "$8", "$9", "$10", "$11", "$12", "$14", "$15", "$24", "$31",	\
-	  DADDI_SCRATCH, "memory");					\
-	__cu_len_r;							\
-})
-
 /*
  * Source or destination address is in userland. We need to go through
  * the TLB
  */
 #define __invoke_copy_from_user(to, from, n)				\
-	__invoke_copy_from_user_eva_generic(to, from, n, __copy_from_user_eva)
-
-#define __invoke_copy_from_user_inatomic(to, from, n)			\
-	__invoke_copy_from_user_eva_generic(to, from, n,		\
-					    __copy_user_inatomic_eva)
+	__invoke_copy_from(__copy_from_user_eva, to, from, n)
 
 #define __invoke_copy_to_user(to, from, n)				\
-	__invoke_copy_to_user_eva_generic(to, from, n, __copy_to_user_eva)
+	__invoke_copy_to(__copy_to_user_eva, to, from, n)
 
 #define ___invoke_copy_in_user(to, from, n)				\
-	__invoke_copy_from_user_eva_generic(to, from, n, __copy_in_user_eva)
-
-/*
- * Source or destination address in the kernel. We are not going through
- * the TLB
- */
-#define __invoke_copy_from_kernel(to, from, n)				\
-	__invoke_copy_from_user_eva_generic(to, from, n, __copy_user)
-
-#define __invoke_copy_from_kernel_inatomic(to, from, n)			\
-	__invoke_copy_from_user_eva_generic(to, from, n, __copy_user_inatomic)
-
-#define __invoke_copy_to_kernel(to, from, n)				\
-	__invoke_copy_to_user_eva_generic(to, from, n, __copy_user)
-
-#define ___invoke_copy_in_kernel(to, from, n)				\
-	__invoke_copy_from_user_eva_generic(to, from, n, __copy_user)
+	__invoke_copy_from(__copy_in_user_eva, to, from, n)
 
 #endif /* CONFIG_EVA */
 
-/*
- * __copy_from_user: - Copy a block of data from user space, with less checking.
- * @to:	  Destination address, in kernel space.
- * @from: Source address, in user space.
- * @n:	  Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from user space to kernel space.  Caller must check
- * the specified block with access_ok() before calling this function.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- *
- * If some data could not be copied, this function will pad the copied
- * data to the requested size using zero bytes.
- */
-#define __copy_from_user(to, from, n)					\
-({									\
-	void *__cu_to;							\
-	const void __user *__cu_from;					\
-	long __cu_len;							\
-									\
-	__cu_to = (to);							\
-	__cu_from = (from);						\
-	__cu_len = (n);							\
-									\
-	check_object_size(__cu_to, __cu_len, false);			\
-									\
-	if (eva_kernel_access()) {					\
-		__cu_len = __invoke_copy_from_kernel(__cu_to,		\
-						     __cu_from,		\
-						     __cu_len);		\
-	} else {							\
-		might_fault();						\
-		__cu_len = __invoke_copy_from_user(__cu_to, __cu_from,	\
-						   __cu_len);		\
-	}								\
-	__cu_len;							\
-})
+static inline unsigned long
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
+{
+	if (eva_kernel_access())
+		return __invoke_copy_to_kernel(to, from, n);
+	else
+		return __invoke_copy_to_user(to, from, n);
+}
 
-/*
- * copy_from_user: - Copy a block of data from user space.
- * @to:	  Destination address, in kernel space.
- * @from: Source address, in user space.
- * @n:	  Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from user space to kernel space.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- *
- * If some data could not be copied, this function will pad the copied
- * data to the requested size using zero bytes.
- */
-#define copy_from_user(to, from, n)					\
-({									\
-	void *__cu_to;							\
-	const void __user *__cu_from;					\
-	long __cu_len;							\
-									\
-	__cu_to = (to);							\
-	__cu_from = (from);						\
-	__cu_len = (n);							\
-									\
-	check_object_size(__cu_to, __cu_len, false);			\
-									\
-	if (eva_kernel_access()) {					\
-		__cu_len = __invoke_copy_from_kernel(__cu_to,		\
-						     __cu_from,		\
-						     __cu_len);		\
-	} else {							\
-		if (access_ok(VERIFY_READ, __cu_from, __cu_len)) {	\
-			might_fault();                                  \
-			__cu_len = __invoke_copy_from_user(__cu_to,	\
-							   __cu_from,	\
-							   __cu_len);   \
-		} else {						\
-			memset(__cu_to, 0, __cu_len);			\
-		}							\
-	}								\
-	__cu_len;							\
-})
+static inline unsigned long
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
+{
+	if (eva_kernel_access())
+		return __invoke_copy_from_kernel(to, from, n);
+	else
+		return __invoke_copy_from_user(to, from, n);
+}
 
-#define __copy_in_user(to, from, n)					\
-({									\
-	void __user *__cu_to;						\
-	const void __user *__cu_from;					\
-	long __cu_len;							\
-									\
-	__cu_to = (to);							\
-	__cu_from = (from);						\
-	__cu_len = (n);							\
-	if (eva_kernel_access()) {					\
-		__cu_len = ___invoke_copy_in_kernel(__cu_to, __cu_from,	\
-						    __cu_len);		\
-	} else {							\
-		might_fault();						\
-		__cu_len = ___invoke_copy_in_user(__cu_to, __cu_from,	\
-						  __cu_len);		\
-	}								\
-	__cu_len;							\
-})
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
-#define copy_in_user(to, from, n)					\
-({									\
-	void __user *__cu_to;						\
-	const void __user *__cu_from;					\
-	long __cu_len;							\
-									\
-	__cu_to = (to);							\
-	__cu_from = (from);						\
-	__cu_len = (n);							\
-	if (eva_kernel_access()) {					\
-		__cu_len = ___invoke_copy_in_kernel(__cu_to,__cu_from,	\
-						    __cu_len);		\
-	} else {							\
-		if (likely(access_ok(VERIFY_READ, __cu_from, __cu_len) &&\
-			   access_ok(VERIFY_WRITE, __cu_to, __cu_len))) {\
-			might_fault();					\
-			__cu_len = ___invoke_copy_in_user(__cu_to,	\
-							  __cu_from,	\
-							  __cu_len);	\
-		}							\
-	}								\
-	__cu_len;							\
-})
+static inline unsigned long
+raw_copy_in_user(void __user*to, const void __user *from, unsigned long n)
+{
+	if (eva_kernel_access())
+		return ___invoke_copy_in_kernel(to, from, n);
+	else
+		return ___invoke_copy_in_user(to, from,	n);
+}
 
 extern __kernel_size_t __bzero_kernel(void __user *addr, __kernel_size_t size);
 extern __kernel_size_t __bzero(void __user *addr, __kernel_size_t size);
diff --git a/arch/mips/kernel/smp-bmips.c b/arch/mips/kernel/smp-bmips.c
index 3daa2ca..1b070a7 100644
--- a/arch/mips/kernel/smp-bmips.c
+++ b/arch/mips/kernel/smp-bmips.c
@@ -11,6 +11,7 @@
 #include <linux/init.h>
 #include <linux/sched.h>
 #include <linux/sched/hotplug.h>
+#include <linux/sched/task_stack.h>
 #include <linux/mm.h>
 #include <linux/delay.h>
 #include <linux/smp.h>
diff --git a/arch/mips/kernel/smp-mt.c b/arch/mips/kernel/smp-mt.c
index e077ea3..e398cbc 100644
--- a/arch/mips/kernel/smp-mt.c
+++ b/arch/mips/kernel/smp-mt.c
@@ -23,6 +23,7 @@
 #include <linux/interrupt.h>
 #include <linux/irqchip/mips-gic.h>
 #include <linux/compiler.h>
+#include <linux/sched/task_stack.h>
 #include <linux/smp.h>
 
 #include <linux/atomic.h>
diff --git a/arch/mips/kernel/syscall.c b/arch/mips/kernel/syscall.c
index f1d17ec..1dfa7f5 100644
--- a/arch/mips/kernel/syscall.c
+++ b/arch/mips/kernel/syscall.c
@@ -98,7 +98,7 @@ static inline int mips_atomic_set(unsigned long addr, unsigned long new)
 	if (unlikely(addr & 3))
 		return -EINVAL;
 
-	if (unlikely(!access_ok(VERIFY_WRITE, addr, 4)))
+	if (unlikely(!access_ok(VERIFY_WRITE, (const void __user *)addr, 4)))
 		return -EINVAL;
 
 	if (cpu_has_llsc && R10000_LLSC_WAR) {
diff --git a/arch/mips/lib/memcpy.S b/arch/mips/lib/memcpy.S
index c3031f1..3114a2e 100644
--- a/arch/mips/lib/memcpy.S
+++ b/arch/mips/lib/memcpy.S
@@ -562,39 +562,9 @@
 	LOADK	t0, THREAD_BUADDR(t0)	# t0 is just past last good address
 	 nop
 	SUB	len, AT, t0		# len number of uncopied bytes
-	bnez	t6, .Ldone\@	/* Skip the zeroing part if inatomic */
-	/*
-	 * Here's where we rely on src and dst being incremented in tandem,
-	 *   See (3) above.
-	 * dst += (fault addr - src) to put dst at first byte to clear
-	 */
-	ADD	dst, t0			# compute start address in a1
-	SUB	dst, src
-	/*
-	 * Clear len bytes starting at dst.  Can't call __bzero because it
-	 * might modify len.  An inefficient loop for these rare times...
-	 */
-	.set	reorder				/* DADDI_WAR */
-	SUB	src, len, 1
-	beqz	len, .Ldone\@
-	.set	noreorder
-1:	sb	zero, 0(dst)
-	ADD	dst, dst, 1
-#ifndef CONFIG_CPU_DADDI_WORKAROUNDS
-	bnez	src, 1b
-	 SUB	src, src, 1
-#else
-	.set	push
-	.set	noat
-	li	v1, 1
-	bnez	src, 1b
-	 SUB	src, src, v1
-	.set	pop
-#endif
 	jr	ra
 	 nop
 
-
 #define SEXC(n)							\
 	.set	reorder;			/* DADDI_WAR */ \
 .Ls_exc_p ## n ## u\@:						\
@@ -673,15 +643,6 @@
 	END(__rmemcpy)
 
 /*
- * t6 is used as a flag to note inatomic mode.
- */
-LEAF(__copy_user_inatomic)
-EXPORT_SYMBOL(__copy_user_inatomic)
-	b	__copy_user_common
-	li	t6, 1
-	END(__copy_user_inatomic)
-
-/*
  * A combined memcpy/__copy_user
  * __copy_user sets len to 0 for success; else to an upper bound of
  * the number of uncopied bytes.
@@ -694,8 +655,6 @@
 .L__memcpy:
 FEXPORT(__copy_user)
 EXPORT_SYMBOL(__copy_user)
-	li	t6, 0	/* not inatomic */
-__copy_user_common:
 	/* Legacy Mode, user <-> user */
 	__BUILD_COPY_USER LEGACY_MODE USEROP USEROP
 
@@ -708,20 +667,12 @@
  * space
  */
 
-LEAF(__copy_user_inatomic_eva)
-EXPORT_SYMBOL(__copy_user_inatomic_eva)
-	b       __copy_from_user_common
-	li	t6, 1
-	END(__copy_user_inatomic_eva)
-
 /*
  * __copy_from_user (EVA)
  */
 
 LEAF(__copy_from_user_eva)
 EXPORT_SYMBOL(__copy_from_user_eva)
-	li	t6, 0	/* not inatomic */
-__copy_from_user_common:
 	__BUILD_COPY_USER EVA_MODE USEROP KERNELOP
 END(__copy_from_user_eva)
 
diff --git a/arch/mips/loongson64/loongson-3/cop2-ex.c b/arch/mips/loongson64/loongson-3/cop2-ex.c
index ea13764..621d6af 100644
--- a/arch/mips/loongson64/loongson-3/cop2-ex.c
+++ b/arch/mips/loongson64/loongson-3/cop2-ex.c
@@ -13,6 +13,7 @@
 #include <linux/init.h>
 #include <linux/sched.h>
 #include <linux/notifier.h>
+#include <linux/ptrace.h>
 
 #include <asm/fpu.h>
 #include <asm/cop2.h>
diff --git a/arch/mips/netlogic/common/smp.c b/arch/mips/netlogic/common/smp.c
index 10d86d5..bddf1ef 100644
--- a/arch/mips/netlogic/common/smp.c
+++ b/arch/mips/netlogic/common/smp.c
@@ -35,6 +35,7 @@
 #include <linux/kernel.h>
 #include <linux/delay.h>
 #include <linux/init.h>
+#include <linux/sched/task_stack.h>
 #include <linux/smp.h>
 #include <linux/irq.h>
 
diff --git a/arch/mips/netlogic/xlp/cop2-ex.c b/arch/mips/netlogic/xlp/cop2-ex.c
index 52bc5de..21e439b 100644
--- a/arch/mips/netlogic/xlp/cop2-ex.c
+++ b/arch/mips/netlogic/xlp/cop2-ex.c
@@ -9,11 +9,14 @@
  * Copyright (C) 2009 Wind River Systems,
  *   written by Ralf Baechle <ralf@linux-mips.org>
  */
+#include <linux/capability.h>
 #include <linux/init.h>
 #include <linux/irqflags.h>
 #include <linux/notifier.h>
 #include <linux/prefetch.h>
+#include <linux/ptrace.h>
 #include <linux/sched.h>
+#include <linux/sched/task_stack.h>
 
 #include <asm/cop2.h>
 #include <asm/current.h>
diff --git a/arch/mips/oprofile/backtrace.c b/arch/mips/oprofile/backtrace.c
index 5e645c9..16ace55 100644
--- a/arch/mips/oprofile/backtrace.c
+++ b/arch/mips/oprofile/backtrace.c
@@ -18,7 +18,7 @@ struct stackframe {
 static inline int get_mem(unsigned long addr, unsigned long *result)
 {
 	unsigned long *address = (unsigned long *) addr;
-	if (!access_ok(VERIFY_READ, addr, sizeof(unsigned long)))
+	if (!access_ok(VERIFY_READ, address, sizeof(unsigned long)))
 		return -1;
 	if (__copy_from_user_inatomic(result, address, sizeof(unsigned long)))
 		return -3;
diff --git a/arch/mips/sgi-ip22/ip28-berr.c b/arch/mips/sgi-ip22/ip28-berr.c
index 1f2a5bc..75460e1 100644
--- a/arch/mips/sgi-ip22/ip28-berr.c
+++ b/arch/mips/sgi-ip22/ip28-berr.c
@@ -9,6 +9,7 @@
 #include <linux/kernel.h>
 #include <linux/sched.h>
 #include <linux/sched/debug.h>
+#include <linux/sched/signal.h>
 #include <linux/seq_file.h>
 
 #include <asm/addrspace.h>
diff --git a/arch/mips/sgi-ip27/ip27-berr.c b/arch/mips/sgi-ip27/ip27-berr.c
index d12879e..83efe03 100644
--- a/arch/mips/sgi-ip27/ip27-berr.c
+++ b/arch/mips/sgi-ip27/ip27-berr.c
@@ -12,7 +12,9 @@
 #include <linux/signal.h>	/* for SIGBUS */
 #include <linux/sched.h>	/* schow_regs(), force_sig() */
 #include <linux/sched/debug.h>
+#include <linux/sched/signal.h>
 
+#include <asm/ptrace.h>
 #include <asm/sn/addrs.h>
 #include <asm/sn/arch.h>
 #include <asm/sn/sn0/hub.h>
diff --git a/arch/mips/sgi-ip27/ip27-smp.c b/arch/mips/sgi-ip27/ip27-smp.c
index f5ed45e..4cd47d2 100644
--- a/arch/mips/sgi-ip27/ip27-smp.c
+++ b/arch/mips/sgi-ip27/ip27-smp.c
@@ -8,10 +8,13 @@
  */
 #include <linux/init.h>
 #include <linux/sched.h>
+#include <linux/sched/task_stack.h>
 #include <linux/topology.h>
 #include <linux/nodemask.h>
+
 #include <asm/page.h>
 #include <asm/processor.h>
+#include <asm/ptrace.h>
 #include <asm/sn/arch.h>
 #include <asm/sn/gda.h>
 #include <asm/sn/intr.h>
diff --git a/arch/mips/sgi-ip32/ip32-berr.c b/arch/mips/sgi-ip32/ip32-berr.c
index 57d8c74..c1f12a9 100644
--- a/arch/mips/sgi-ip32/ip32-berr.c
+++ b/arch/mips/sgi-ip32/ip32-berr.c
@@ -11,6 +11,7 @@
 #include <linux/kernel.h>
 #include <linux/sched.h>
 #include <linux/sched/debug.h>
+#include <linux/sched/signal.h>
 #include <asm/traps.h>
 #include <linux/uaccess.h>
 #include <asm/addrspace.h>
diff --git a/arch/mips/sgi-ip32/ip32-reset.c b/arch/mips/sgi-ip32/ip32-reset.c
index 8bd415c..b3b442d 100644
--- a/arch/mips/sgi-ip32/ip32-reset.c
+++ b/arch/mips/sgi-ip32/ip32-reset.c
@@ -13,6 +13,7 @@
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/sched.h>
+#include <linux/sched/signal.h>
 #include <linux/notifier.h>
 #include <linux/delay.h>
 #include <linux/rtc/ds1685.h>
diff --git a/arch/mn10300/Kconfig b/arch/mn10300/Kconfig
index 38e3494..a96f3dc 100644
--- a/arch/mn10300/Kconfig
+++ b/arch/mn10300/Kconfig
@@ -16,6 +16,7 @@
 	select OLD_SIGACTION
 	select HAVE_DEBUG_STACKOVERFLOW
 	select ARCH_NO_COHERENT_DMA_MMAP
+	select ARCH_HAS_RAW_COPY_USER
 
 config AM33_2
 	def_bool n
diff --git a/arch/mn10300/include/asm/Kbuild b/arch/mn10300/include/asm/Kbuild
index 97f64c7..ed810e7 100644
--- a/arch/mn10300/include/asm/Kbuild
+++ b/arch/mn10300/include/asm/Kbuild
@@ -2,6 +2,7 @@
 generic-y += barrier.h
 generic-y += clkdev.h
 generic-y += exec.h
+generic-y += extable.h
 generic-y += irq_work.h
 generic-y += mcs_spinlock.h
 generic-y += mm-arch-hooks.h
diff --git a/arch/mn10300/include/asm/uaccess.h b/arch/mn10300/include/asm/uaccess.h
index 2da7b0f..c696647 100644
--- a/arch/mn10300/include/asm/uaccess.h
+++ b/arch/mn10300/include/asm/uaccess.h
@@ -66,26 +66,7 @@ static inline int ___range_ok(unsigned long addr, unsigned int size)
 #define access_ok(type, addr, size) (__range_ok((addr), (size)) == 0)
 #define __access_ok(addr, size)     (__range_ok((addr), (size)) == 0)
 
-/*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
-
-struct exception_table_entry
-{
-	unsigned long insn, fixup;
-};
-
-/* Returns 0 if exception not found and fixup otherwise.  */
-extern int fixup_exception(struct pt_regs *regs);
+#include <asm/extable.h>
 
 #define put_user(x, ptr) __put_user_check((x), (ptr), sizeof(*(ptr)))
 #define get_user(x, ptr) __get_user_check((x), (ptr), sizeof(*(ptr)))
@@ -294,171 +275,20 @@ do {									\
 	}								\
 } while (0)
 
-#define __copy_user_zeroing(to, from, size)				\
-do {									\
-	if (size) {							\
-		void *__to = to;					\
-		const void *__from = from;				\
-		int w;							\
-		asm volatile(						\
-			"0:     movbu	(%0),%3;\n"			\
-			"1:     movbu	%3,(%1);\n"			\
-			"	inc	%0;\n"				\
-			"	inc	%1;\n"				\
-			"       add	-1,%2;\n"			\
-			"       bne	0b;\n"				\
-			"2:\n"						\
-			"	.section .fixup,\"ax\"\n"		\
-			"3:\n"						\
-			"	mov	%2,%0\n"			\
-			"	clr	%3\n"				\
-			"4:     movbu	%3,(%1);\n"			\
-			"	inc	%1;\n"				\
-			"       add	-1,%2;\n"			\
-			"       bne	4b;\n"				\
-			"	mov	%0,%2\n"			\
-			"	jmp	2b\n"				\
-			"	.previous\n"				\
-			"	.section __ex_table,\"a\"\n"		\
-			"       .balign	4\n"				\
-			"       .long	0b,3b\n"			\
-			"       .long	1b,3b\n"			\
-			"	.previous\n"				\
-			: "=a"(__from), "=a"(__to), "=r"(size), "=&r"(w)\
-			: "0"(__from), "1"(__to), "2"(size)		\
-			: "cc", "memory");				\
-	}								\
-} while (0)
-
-/* We let the __ versions of copy_from/to_user inline, because they're often
- * used in fast paths and have only a small space overhead.
- */
-static inline
-unsigned long __generic_copy_from_user_nocheck(void *to, const void *from,
-					       unsigned long n)
-{
-	__copy_user_zeroing(to, from, n);
-	return n;
-}
-
-static inline
-unsigned long __generic_copy_to_user_nocheck(void *to, const void *from,
-					     unsigned long n)
+static inline unsigned long
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
 {
 	__copy_user(to, from, n);
 	return n;
 }
 
-
-#if 0
-#error "don't use - these macros don't increment to & from pointers"
-/* Optimize just a little bit when we know the size of the move. */
-#define __constant_copy_user(to, from, size)	\
-do {						\
-	asm volatile(				\
-		"       mov %0,a0;\n"		\
-		"0:     movbu (%1),d3;\n"	\
-		"1:     movbu d3,(%2);\n"	\
-		"       add -1,a0;\n"		\
-		"       bne 0b;\n"		\
-		"2:;"				\
-		".section .fixup,\"ax\"\n"	\
-		"3:	jmp 2b\n"		\
-		".previous\n"			\
-		".section __ex_table,\"a\"\n"	\
-		"       .balign 4\n"		\
-		"       .long 0b,3b\n"		\
-		"       .long 1b,3b\n"		\
-		".previous"			\
-		:				\
-		: "d"(size), "d"(to), "d"(from)	\
-		: "d3", "a0");			\
-} while (0)
-
-/* Optimize just a little bit when we know the size of the move. */
-#define __constant_copy_user_zeroing(to, from, size)	\
-do {							\
-	asm volatile(					\
-		"       mov %0,a0;\n"			\
-		"0:     movbu (%1),d3;\n"		\
-		"1:     movbu d3,(%2);\n"		\
-		"       add -1,a0;\n"			\
-		"       bne 0b;\n"			\
-		"2:;"					\
-		".section .fixup,\"ax\"\n"		\
-		"3:	jmp 2b\n"			\
-		".previous\n"				\
-		".section __ex_table,\"a\"\n"		\
-		"       .balign 4\n"			\
-		"       .long 0b,3b\n"			\
-		"       .long 1b,3b\n"			\
-		".previous"				\
-		:					\
-		: "d"(size), "d"(to), "d"(from)		\
-		: "d3", "a0");				\
-} while (0)
-
-static inline
-unsigned long __constant_copy_to_user(void *to, const void *from,
-				      unsigned long n)
+static inline unsigned long
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 {
-	if (access_ok(VERIFY_WRITE, to, n))
-		__constant_copy_user(to, from, n);
+	__copy_user(to, from, n);
 	return n;
 }
 
-static inline
-unsigned long __constant_copy_from_user(void *to, const void *from,
-					unsigned long n)
-{
-	if (access_ok(VERIFY_READ, from, n))
-		__constant_copy_user_zeroing(to, from, n);
-	return n;
-}
-
-static inline
-unsigned long __constant_copy_to_user_nocheck(void *to, const void *from,
-					      unsigned long n)
-{
-	__constant_copy_user(to, from, n);
-	return n;
-}
-
-static inline
-unsigned long __constant_copy_from_user_nocheck(void *to, const void *from,
-						unsigned long n)
-{
-	__constant_copy_user_zeroing(to, from, n);
-	return n;
-}
-#endif
-
-extern unsigned long __generic_copy_to_user(void __user *, const void *,
-					    unsigned long);
-extern unsigned long __generic_copy_from_user(void *, const void __user *,
-					      unsigned long);
-
-#define __copy_to_user_inatomic(to, from, n) \
-	__generic_copy_to_user_nocheck((to), (from), (n))
-#define __copy_from_user_inatomic(to, from, n) \
-	__generic_copy_from_user_nocheck((to), (from), (n))
-
-#define __copy_to_user(to, from, n)			\
-({							\
-	might_fault();					\
-	__copy_to_user_inatomic((to), (from), (n));	\
-})
-
-#define __copy_from_user(to, from, n)			\
-({							\
-	might_fault();					\
-	__copy_from_user_inatomic((to), (from), (n));	\
-})
-
-
-#define copy_to_user(to, from, n)   __generic_copy_to_user((to), (from), (n))
-#define copy_from_user(to, from, n) __generic_copy_from_user((to), (from), (n))
-
 extern long strncpy_from_user(char *dst, const char __user *src, long count);
 extern long __strncpy_from_user(char *dst, const char __user *src, long count);
 extern long strnlen_user(const char __user *str, long n);
diff --git a/arch/mn10300/kernel/mn10300_ksyms.c b/arch/mn10300/kernel/mn10300_ksyms.c
index ec6c4f8..5e9f919 100644
--- a/arch/mn10300/kernel/mn10300_ksyms.c
+++ b/arch/mn10300/kernel/mn10300_ksyms.c
@@ -26,8 +26,6 @@ EXPORT_SYMBOL(strncpy_from_user);
 EXPORT_SYMBOL(__strncpy_from_user);
 EXPORT_SYMBOL(clear_user);
 EXPORT_SYMBOL(__clear_user);
-EXPORT_SYMBOL(__generic_copy_from_user);
-EXPORT_SYMBOL(__generic_copy_to_user);
 EXPORT_SYMBOL(strnlen_user);
 
 extern u64 __ashrdi3(u64, unsigned);
diff --git a/arch/mn10300/lib/usercopy.c b/arch/mn10300/lib/usercopy.c
index ce8899e..cece179 100644
--- a/arch/mn10300/lib/usercopy.c
+++ b/arch/mn10300/lib/usercopy.c
@@ -11,24 +11,6 @@
  */
 #include <linux/uaccess.h>
 
-unsigned long
-__generic_copy_to_user(void *to, const void *from, unsigned long n)
-{
-	if (access_ok(VERIFY_WRITE, to, n))
-		__copy_user(to, from, n);
-	return n;
-}
-
-unsigned long
-__generic_copy_from_user(void *to, const void *from, unsigned long n)
-{
-	if (access_ok(VERIFY_READ, from, n))
-		__copy_user_zeroing(to, from, n);
-	else
-		memset(to, 0, n);
-	return n;
-}
-
 /*
  * Copy a null terminated string from userspace.
  */
diff --git a/arch/nios2/Kconfig b/arch/nios2/Kconfig
index 51a56c8..45b4727 100644
--- a/arch/nios2/Kconfig
+++ b/arch/nios2/Kconfig
@@ -16,6 +16,7 @@
 	select SPARSE_IRQ
 	select USB_ARCH_HAS_HCD if USB_SUPPORT
 	select CPU_NO_EFFICIENT_FFS
+	select ARCH_HAS_RAW_COPY_USER
 
 config GENERIC_CSUM
 	def_bool y
diff --git a/arch/nios2/include/asm/Kbuild b/arch/nios2/include/asm/Kbuild
index aaa3c21..87e70f2 100644
--- a/arch/nios2/include/asm/Kbuild
+++ b/arch/nios2/include/asm/Kbuild
@@ -13,6 +13,7 @@
 generic-y += emergency-restart.h
 generic-y += errno.h
 generic-y += exec.h
+generic-y += extable.h
 generic-y += fb.h
 generic-y += fcntl.h
 generic-y += ftrace.h
diff --git a/arch/nios2/include/asm/uaccess.h b/arch/nios2/include/asm/uaccess.h
index 198bbf1..727bd95 100644
--- a/arch/nios2/include/asm/uaccess.h
+++ b/arch/nios2/include/asm/uaccess.h
@@ -17,24 +17,7 @@
 
 #include <asm/page.h>
 
-/*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
-struct exception_table_entry {
-	unsigned long insn;
-	unsigned long fixup;
-};
-
-extern int fixup_exception(struct pt_regs *regs);
+#include <asm/extable.h>
 
 /*
  * Segment stuff
@@ -90,36 +73,17 @@ static inline unsigned long __must_check clear_user(void __user *to,
 	return __clear_user(to, n);
 }
 
-extern long __copy_from_user(void *to, const void __user *from,
-				unsigned long n);
-extern long __copy_to_user(void __user *to, const void *from, unsigned long n);
-
-static inline long copy_from_user(void *to, const void __user *from,
-				unsigned long n)
-{
-	unsigned long res = n;
-	if (access_ok(VERIFY_READ, from, n))
-		res = __copy_from_user(to, from, n);
-	if (unlikely(res))
-		memset(to + (n - res), 0, res);
-	return res;
-}
-
-static inline long copy_to_user(void __user *to, const void *from,
-				unsigned long n)
-{
-	if (!access_ok(VERIFY_WRITE, to, n))
-		return n;
-	return __copy_to_user(to, from, n);
-}
+extern unsigned long
+raw_copy_from_user(void *to, const void __user *from, unsigned long n);
+extern unsigned long
+raw_copy_to_user(void __user *to, const void *from, unsigned long n);
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 extern long strncpy_from_user(char *__to, const char __user *__from,
 				long __len);
 extern long strnlen_user(const char __user *s, long n);
 
-#define __copy_from_user_inatomic	__copy_from_user
-#define __copy_to_user_inatomic		__copy_to_user
-
 /* Optimized macros */
 #define __get_user_asm(val, insn, addr, err)				\
 {									\
diff --git a/arch/nios2/mm/uaccess.c b/arch/nios2/mm/uaccess.c
index 7663e15..8049833 100644
--- a/arch/nios2/mm/uaccess.c
+++ b/arch/nios2/mm/uaccess.c
@@ -10,9 +10,9 @@
 #include <linux/export.h>
 #include <linux/uaccess.h>
 
-asm(".global	__copy_from_user\n"
-	"   .type __copy_from_user, @function\n"
-	"__copy_from_user:\n"
+asm(".global	raw_copy_from_user\n"
+	"   .type raw_copy_from_user, @function\n"
+	"raw_copy_from_user:\n"
 	"   movi  r2,7\n"
 	"   mov   r3,r4\n"
 	"   bge   r2,r6,1f\n"
@@ -65,12 +65,12 @@ asm(".global	__copy_from_user\n"
 	".word 7b,13b\n"
 	".previous\n"
 	);
-EXPORT_SYMBOL(__copy_from_user);
+EXPORT_SYMBOL(raw_copy_from_user);
 
 asm(
-	"   .global __copy_to_user\n"
-	"   .type __copy_to_user, @function\n"
-	"__copy_to_user:\n"
+	"   .global raw_copy_to_user\n"
+	"   .type raw_copy_to_user, @function\n"
+	"raw_copy_to_user:\n"
 	"   movi  r2,7\n"
 	"   mov   r3,r4\n"
 	"   bge   r2,r6,1f\n"
@@ -127,7 +127,7 @@ asm(
 	".word 11b,13b\n"
 	".word 12b,13b\n"
 	".previous\n");
-EXPORT_SYMBOL(__copy_to_user);
+EXPORT_SYMBOL(raw_copy_to_user);
 
 long strncpy_from_user(char *__to, const char __user *__from, long __len)
 {
diff --git a/arch/openrisc/Kconfig b/arch/openrisc/Kconfig
index 1e95920..3895418 100644
--- a/arch/openrisc/Kconfig
+++ b/arch/openrisc/Kconfig
@@ -28,6 +28,7 @@
 	select OR1K_PIC
 	select CPU_NO_EFFICIENT_FFS if !OPENRISC_HAVE_INST_FF1
 	select NO_BOOTMEM
+	select ARCH_HAS_RAW_COPY_USER
 
 config MMU
 	def_bool y
diff --git a/arch/openrisc/include/asm/Kbuild b/arch/openrisc/include/asm/Kbuild
index fb01873..df8e2f7 100644
--- a/arch/openrisc/include/asm/Kbuild
+++ b/arch/openrisc/include/asm/Kbuild
@@ -16,6 +16,7 @@
 generic-y += emergency-restart.h
 generic-y += errno.h
 generic-y += exec.h
+generic-y += extable.h
 generic-y += fb.h
 generic-y += fcntl.h
 generic-y += ftrace.h
diff --git a/arch/openrisc/include/asm/uaccess.h b/arch/openrisc/include/asm/uaccess.h
index 227af1a..96fc425 100644
--- a/arch/openrisc/include/asm/uaccess.h
+++ b/arch/openrisc/include/asm/uaccess.h
@@ -25,6 +25,7 @@
 #include <linux/prefetch.h>
 #include <linux/string.h>
 #include <asm/page.h>
+#include <asm/extable.h>
 
 /*
  * The fs value determines whether argument validity checking should be
@@ -61,23 +62,6 @@
 	__range_ok((unsigned long)addr, (unsigned long)size)
 
 /*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
-
-struct exception_table_entry {
-	unsigned long insn, fixup;
-};
-
-/*
  * These are the main single-value transfer routines.  They automatically
  * use the right size if we just have the right pointer type.
  *
@@ -252,34 +236,18 @@ do {									\
 
 extern unsigned long __must_check
 __copy_tofrom_user(void *to, const void *from, unsigned long size);
-
-#define __copy_from_user(to, from, size) \
-	__copy_tofrom_user(to, from, size)
-#define __copy_to_user(to, from, size) \
-	__copy_tofrom_user(to, from, size)
-
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
-
 static inline unsigned long
-copy_from_user(void *to, const void *from, unsigned long n)
+raw_copy_from_user(void *to, const void __user *from, unsigned long size)
 {
-	unsigned long res = n;
-
-	if (likely(access_ok(VERIFY_READ, from, n)))
-		res = __copy_tofrom_user(to, from, n);
-	if (unlikely(res))
-		memset(to + (n - res), 0, res);
-	return res;
+	return __copy_tofrom_user(to, (__force const void *)from, size);
 }
-
 static inline unsigned long
-copy_to_user(void *to, const void *from, unsigned long n)
+raw_copy_to_user(void *to, const void __user *from, unsigned long size)
 {
-	if (likely(access_ok(VERIFY_WRITE, to, n)))
-		n = __copy_tofrom_user(to, from, n);
-	return n;
+	return __copy_tofrom_user((__force void *)to, from, size);
 }
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 extern unsigned long __clear_user(void *addr, unsigned long size);
 
diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index ad294b3..15b7b27 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -41,6 +41,7 @@
 	select GENERIC_CLOCKEVENTS
 	select ARCH_NO_COHERENT_DMA_MMAP
 	select CPU_NO_EFFICIENT_FFS
+	select ARCH_HAS_RAW_COPY_USER
 
 	help
 	  The PA-RISC microprocessor is designed by Hewlett-Packard and used
diff --git a/arch/parisc/include/asm/uaccess.h b/arch/parisc/include/asm/uaccess.h
index a0b4613..f09968b 100644
--- a/arch/parisc/include/asm/uaccess.h
+++ b/arch/parisc/include/asm/uaccess.h
@@ -177,9 +177,6 @@ struct exception_data {
  * Complex access routines -- external declarations
  */
 
-extern unsigned long lcopy_to_user(void __user *, const void *, unsigned long);
-extern unsigned long lcopy_from_user(void *, const void __user *, unsigned long);
-extern unsigned long lcopy_in_user(void __user *, const void __user *, unsigned long);
 extern long strncpy_from_user(char *, const char __user *, long);
 extern unsigned lclear_user(void __user *, unsigned long);
 extern long lstrnlen_user(const char __user *, long);
@@ -193,59 +190,14 @@ extern long lstrnlen_user(const char __user *, long);
 #define clear_user lclear_user
 #define __clear_user lclear_user
 
-unsigned long __must_check __copy_to_user(void __user *dst, const void *src,
-					  unsigned long len);
-unsigned long __must_check __copy_from_user(void *dst, const void __user *src,
-					  unsigned long len);
-unsigned long copy_in_user(void __user *dst, const void __user *src,
-			   unsigned long len);
-#define __copy_in_user copy_in_user
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
-
-extern void __compiletime_error("usercopy buffer size is too small")
-__bad_copy_user(void);
-
-static inline void copy_user_overflow(int size, unsigned long count)
-{
-	WARN(1, "Buffer overflow detected (%d < %lu)!\n", size, count);
-}
-
-static __always_inline unsigned long __must_check
-copy_from_user(void *to, const void __user *from, unsigned long n)
-{
-	int sz = __compiletime_object_size(to);
-	unsigned long ret = n;
-
-	if (likely(sz < 0 || sz >= n)) {
-		check_object_size(to, n, false);
-		ret = __copy_from_user(to, from, n);
-	} else if (!__builtin_constant_p(n))
-		copy_user_overflow(sz, n);
-	else
-		__bad_copy_user();
-
-	if (unlikely(ret))
-		memset(to + (n - ret), 0, ret);
-
-	return ret;
-}
-
-static __always_inline unsigned long __must_check
-copy_to_user(void __user *to, const void *from, unsigned long n)
-{
-	int sz = __compiletime_object_size(from);
-
-	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))
-		copy_user_overflow(sz, n);
-	else
-		__bad_copy_user();
-
-	return n;
-}
+unsigned long __must_check raw_copy_to_user(void __user *dst, const void *src,
+					    unsigned long len);
+unsigned long __must_check raw_copy_from_user(void *dst, const void __user *src,
+					    unsigned long len);
+unsigned long __must_check raw_copy_in_user(void __user *dst, const void __user *src,
+					    unsigned long len);
+#define INLINE_COPY_TO_USER
+#define INLINE_COPY_FROM_USER
 
 struct pt_regs;
 int fixup_exception(struct pt_regs *regs);
diff --git a/arch/parisc/lib/memcpy.c b/arch/parisc/lib/memcpy.c
index 66f5160..7ce9282 100644
--- a/arch/parisc/lib/memcpy.c
+++ b/arch/parisc/lib/memcpy.c
@@ -113,31 +113,10 @@ DECLARE_PER_CPU(struct exception_data, exception_data);
 	: _tt(_t)					\
 	: "r8")
 
-#define ldbma(_s, _a, _t, _e) def_load_ai_insn(ldbs,1,"=r",_s,_a,_t,_e)
-#define stbma(_s, _t, _a, _e) def_store_ai_insn(stbs,1,"r",_s,_a,_t,_e)
-#define ldwma(_s, _a, _t, _e) def_load_ai_insn(ldw,4,"=r",_s,_a,_t,_e)
-#define stwma(_s, _t, _a, _e) def_store_ai_insn(stw,4,"r",_s,_a,_t,_e)
-#define flddma(_s, _a, _t, _e) def_load_ai_insn(fldd,8,"=f",_s,_a,_t,_e)
-#define fstdma(_s, _t, _a, _e) def_store_ai_insn(fstd,8,"f",_s,_a,_t,_e)
-
-#define def_load_insn(_insn,_tt,_s,_o,_a,_t,_e) 	\
-	__asm__ __volatile__ (				\
-	"1:\t" #_insn " " #_o "(" _s ",%1), %0\n\t"	\
-	ASM_EXCEPTIONTABLE_ENTRY(1b,_e)			\
-	: _tt(_t) 					\
-	: "r"(_a)					\
-	: "r8")
-
-#define def_store_insn(_insn,_tt,_s,_t,_o,_a,_e) 	\
-	__asm__ __volatile__ (				\
-	"1:\t" #_insn " %0, " #_o "(" _s ",%1)\n\t" 	\
-	ASM_EXCEPTIONTABLE_ENTRY(1b,_e)			\
-	: 						\
-	: _tt(_t), "r"(_a)				\
-	: "r8")
-
-#define ldw(_s,_o,_a,_t,_e)	def_load_insn(ldw,"=r",_s,_o,_a,_t,_e)
-#define stw(_s,_t,_o,_a,_e) 	def_store_insn(stw,"r",_s,_t,_o,_a,_e)
+#define load1(_t, _e) def_load_ai_insn(ldbs,1,"=r",s_space,src,_t,_e)
+#define store1(_t, _e) def_store_ai_insn(stbs,1,"r",d_space,dst,_t,_e)
+#define load4(_t, _e) def_load_ai_insn(ldw,4,"=r",s_space,src,_t,_e)
+#define store4(_t, _e) def_store_ai_insn(stw,4,"r",d_space,dst,_t,_e)
 
 #ifdef  CONFIG_PREFETCH
 static inline void prefetch_src(const void *addr)
@@ -154,12 +133,13 @@ static inline void prefetch_dst(const void *addr)
 #define prefetch_dst(addr) do { } while(0)
 #endif
 
-#define PA_MEMCPY_OK		0
-#define PA_MEMCPY_LOAD_ERROR	1
-#define PA_MEMCPY_STORE_ERROR	2
-
 /* Copy from a not-aligned src to an aligned dst, using shifts. Handles 4 words
  * per loop.  This code is derived from glibc. 
+ *
+ * Return 0 on success,
+ *	  last address stored on load fault
+ *	  last address stored + 1 on store fault
+ *
  */
 static noinline unsigned long copy_dstaligned(unsigned long dst,
 					unsigned long src, unsigned long len)
@@ -183,41 +163,25 @@ static noinline unsigned long copy_dstaligned(unsigned long dst,
 	switch (len % 4)
 	{
 		case 2:
-			/* a1 = ((unsigned int *) src)[0];
-			   a2 = ((unsigned int *) src)[1]; */
-			ldw(s_space, 0, src, a1, cda_ldw_exc);
-			ldw(s_space, 4, src, a2, cda_ldw_exc);
-			src -= 1 * sizeof(unsigned int);
-			dst -= 3 * sizeof(unsigned int);
+			load4(a1, cda_ldw_exc);
+			load4(a2, cda_ldw_exc);
 			len += 2;
 			goto do1;
 		case 3:
-			/* a0 = ((unsigned int *) src)[0];
-			   a1 = ((unsigned int *) src)[1]; */
-			ldw(s_space, 0, src, a0, cda_ldw_exc);
-			ldw(s_space, 4, src, a1, cda_ldw_exc);
-			src -= 0 * sizeof(unsigned int);
-			dst -= 2 * sizeof(unsigned int);
+			load4(a0, cda_ldw_exc);
+			load4(a1, cda_ldw_exc);
 			len += 1;
 			goto do2;
 		case 0:
 			if (len == 0)
-				return PA_MEMCPY_OK;
-			/* a3 = ((unsigned int *) src)[0];
-			   a0 = ((unsigned int *) src)[1]; */
-			ldw(s_space, 0, src, a3, cda_ldw_exc);
-			ldw(s_space, 4, src, a0, cda_ldw_exc);
-			src -=-1 * sizeof(unsigned int);
-			dst -= 1 * sizeof(unsigned int);
+				return 0;
+			load4(a3, cda_ldw_exc);
+			load4(a0, cda_ldw_exc);
 			len += 0;
 			goto do3;
 		case 1:
-			/* a2 = ((unsigned int *) src)[0];
-			   a3 = ((unsigned int *) src)[1]; */
-			ldw(s_space, 0, src, a2, cda_ldw_exc);
-			ldw(s_space, 4, src, a3, cda_ldw_exc);
-			src -=-2 * sizeof(unsigned int);
-			dst -= 0 * sizeof(unsigned int);
+			load4(a2, cda_ldw_exc);
+			load4(a3, cda_ldw_exc);
 			len -= 1;
 			if (len == 0)
 				goto do0;
@@ -228,67 +192,52 @@ static noinline unsigned long copy_dstaligned(unsigned long dst,
 	{
 		/* prefetch_src((const void *)(src + 4 * sizeof(unsigned int))); */
 do4:
-		/* a0 = ((unsigned int *) src)[0]; */
-		ldw(s_space, 0, src, a0, cda_ldw_exc);
-		/* ((unsigned int *) dst)[0] = MERGE (a2, sh_1, a3, sh_2); */
-		stw(d_space, MERGE (a2, sh_1, a3, sh_2), 0, dst, cda_stw_exc);
+		load4(a0, cda_ldw_exc);
+		store4(MERGE (a2, sh_1, a3, sh_2), cda_stw_exc);
 do3:
-		/* a1 = ((unsigned int *) src)[1]; */
-		ldw(s_space, 4, src, a1, cda_ldw_exc);
-		/* ((unsigned int *) dst)[1] = MERGE (a3, sh_1, a0, sh_2); */
-		stw(d_space, MERGE (a3, sh_1, a0, sh_2), 4, dst, cda_stw_exc);
+		load4(a1, cda_ldw_exc);
+		store4(MERGE (a3, sh_1, a0, sh_2), cda_stw_exc);
 do2:
-		/* a2 = ((unsigned int *) src)[2]; */
-		ldw(s_space, 8, src, a2, cda_ldw_exc);
-		/* ((unsigned int *) dst)[2] = MERGE (a0, sh_1, a1, sh_2); */
-		stw(d_space, MERGE (a0, sh_1, a1, sh_2), 8, dst, cda_stw_exc);
+		load4(a2, cda_ldw_exc);
+		store4(MERGE (a0, sh_1, a1, sh_2), cda_stw_exc);
 do1:
-		/* a3 = ((unsigned int *) src)[3]; */
-		ldw(s_space, 12, src, a3, cda_ldw_exc);
-		/* ((unsigned int *) dst)[3] = MERGE (a1, sh_1, a2, sh_2); */
-		stw(d_space, MERGE (a1, sh_1, a2, sh_2), 12, dst, cda_stw_exc);
+		load4(a3, cda_ldw_exc);
+		store4(MERGE (a1, sh_1, a2, sh_2), cda_stw_exc);
 
-		src += 4 * sizeof(unsigned int);
-		dst += 4 * sizeof(unsigned int);
 		len -= 4;
 	}
 	while (len != 0);
 
 do0:
 	/* ((unsigned int *) dst)[0] = MERGE (a2, sh_1, a3, sh_2); */
-	stw(d_space, MERGE (a2, sh_1, a3, sh_2), 0, dst, cda_stw_exc);
+	store4(MERGE (a2, sh_1, a3, sh_2), cda_stw_exc);
 
 	preserve_branch(handle_load_error);
 	preserve_branch(handle_store_error);
 
-	return PA_MEMCPY_OK;
+	return 0;
 
 handle_load_error:
 	__asm__ __volatile__ ("cda_ldw_exc:\n");
-	return PA_MEMCPY_LOAD_ERROR;
+	return dst;
 
 handle_store_error:
 	__asm__ __volatile__ ("cda_stw_exc:\n");
-	return PA_MEMCPY_STORE_ERROR;
+	return dst + 1;
 }
 
 
-/* Returns PA_MEMCPY_OK, PA_MEMCPY_LOAD_ERROR or PA_MEMCPY_STORE_ERROR.
- * In case of an access fault the faulty address can be read from the per_cpu
- * exception data struct. */
-static noinline unsigned long pa_memcpy_internal(void *dstp, const void *srcp,
+/* Returns 0 for success, otherwise, returns number of bytes not transferred. */
+static noinline unsigned long pa_memcpy(void *dstp, const void *srcp,
 					unsigned long len)
 {
 	register unsigned long src, dst, t1, t2, t3;
-	register unsigned char *pcs, *pcd;
-	register unsigned int *pws, *pwd;
-	register double *pds, *pdd;
-	unsigned long ret;
+	unsigned long ret, end;
 
 	src = (unsigned long)srcp;
 	dst = (unsigned long)dstp;
-	pcs = (unsigned char *)srcp;
-	pcd = (unsigned char *)dstp;
+	end = dst + len;
+	asm volatile("": : "r"(end));
 
 	/* prefetch_src((const void *)srcp); */
 
@@ -297,217 +246,136 @@ static noinline unsigned long pa_memcpy_internal(void *dstp, const void *srcp,
 
 	/* Check alignment */
 	t1 = (src ^ dst);
-	if (unlikely(t1 & (sizeof(double)-1)))
+	if (unlikely(t1 & (sizeof(unsigned int)-1)))
 		goto unaligned_copy;
 
 	/* src and dst have same alignment. */
 
 	/* Copy bytes till we are double-aligned. */
-	t2 = src & (sizeof(double) - 1);
+	t2 = src & (sizeof(unsigned int) - 1);
 	if (unlikely(t2 != 0)) {
-		t2 = sizeof(double) - t2;
+		t2 = sizeof(unsigned int) - t2;
 		while (t2 && len) {
-			/* *pcd++ = *pcs++; */
-			ldbma(s_space, pcs, t3, pmc_load_exc);
+			load1(t3, pmc_done);
 			len--;
-			stbma(d_space, t3, pcd, pmc_store_exc);
+			store1(t3, pmc_done);
 			t2--;
 		}
 	}
 
-	pds = (double *)pcs;
-	pdd = (double *)pcd;
-
-#if 0
-	/* Copy 8 doubles at a time */
-	while (len >= 8*sizeof(double)) {
-		register double r1, r2, r3, r4, r5, r6, r7, r8;
-		/* prefetch_src((char *)pds + L1_CACHE_BYTES); */
-		flddma(s_space, pds, r1, pmc_load_exc);
-		flddma(s_space, pds, r2, pmc_load_exc);
-		flddma(s_space, pds, r3, pmc_load_exc);
-		flddma(s_space, pds, r4, pmc_load_exc);
-		fstdma(d_space, r1, pdd, pmc_store_exc);
-		fstdma(d_space, r2, pdd, pmc_store_exc);
-		fstdma(d_space, r3, pdd, pmc_store_exc);
-		fstdma(d_space, r4, pdd, pmc_store_exc);
-
-#if 0
-		if (L1_CACHE_BYTES <= 32)
-			prefetch_src((char *)pds + L1_CACHE_BYTES);
-#endif
-		flddma(s_space, pds, r5, pmc_load_exc);
-		flddma(s_space, pds, r6, pmc_load_exc);
-		flddma(s_space, pds, r7, pmc_load_exc);
-		flddma(s_space, pds, r8, pmc_load_exc);
-		fstdma(d_space, r5, pdd, pmc_store_exc);
-		fstdma(d_space, r6, pdd, pmc_store_exc);
-		fstdma(d_space, r7, pdd, pmc_store_exc);
-		fstdma(d_space, r8, pdd, pmc_store_exc);
-		len -= 8*sizeof(double);
-	}
-#endif
-
-	pws = (unsigned int *)pds;
-	pwd = (unsigned int *)pdd;
-
-word_copy:
 	while (len >= 8*sizeof(unsigned int)) {
 		register unsigned int r1,r2,r3,r4,r5,r6,r7,r8;
-		/* prefetch_src((char *)pws + L1_CACHE_BYTES); */
-		ldwma(s_space, pws, r1, pmc_load_exc);
-		ldwma(s_space, pws, r2, pmc_load_exc);
-		ldwma(s_space, pws, r3, pmc_load_exc);
-		ldwma(s_space, pws, r4, pmc_load_exc);
-		stwma(d_space, r1, pwd, pmc_store_exc);
-		stwma(d_space, r2, pwd, pmc_store_exc);
-		stwma(d_space, r3, pwd, pmc_store_exc);
-		stwma(d_space, r4, pwd, pmc_store_exc);
+		/* prefetch_src((char *)src + L1_CACHE_BYTES); */
+		load4(r1, pmc_done);
+		load4(r2, pmc_byte4);		// bytecopy src-4
+		load4(r3, pmc_byte8);		// bytecopy src-8
+		load4(r4, pmc_byte12);		// bytecopy src-12
+		store4(r1, pmc_done);
+		store4(r2, pmc_done);
+		store4(r3, pmc_done);
+		store4(r4, pmc_done);
 
-		ldwma(s_space, pws, r5, pmc_load_exc);
-		ldwma(s_space, pws, r6, pmc_load_exc);
-		ldwma(s_space, pws, r7, pmc_load_exc);
-		ldwma(s_space, pws, r8, pmc_load_exc);
-		stwma(d_space, r5, pwd, pmc_store_exc);
-		stwma(d_space, r6, pwd, pmc_store_exc);
-		stwma(d_space, r7, pwd, pmc_store_exc);
-		stwma(d_space, r8, pwd, pmc_store_exc);
+		load4(r5, pmc_done);
+		load4(r6, pmc_byte4);		// bytecopy src-4
+		load4(r7, pmc_byte8);		// bytecopy src-8
+		load4(r8, pmc_byte12);		// bytecopy src-12
+		store4(r5, pmc_done);
+		store4(r6, pmc_done);
+		store4(r7, pmc_done);
+		store4(r8, pmc_done);
 		len -= 8*sizeof(unsigned int);
 	}
 
 	while (len >= 4*sizeof(unsigned int)) {
 		register unsigned int r1,r2,r3,r4;
-		ldwma(s_space, pws, r1, pmc_load_exc);
-		ldwma(s_space, pws, r2, pmc_load_exc);
-		ldwma(s_space, pws, r3, pmc_load_exc);
-		ldwma(s_space, pws, r4, pmc_load_exc);
-		stwma(d_space, r1, pwd, pmc_store_exc);
-		stwma(d_space, r2, pwd, pmc_store_exc);
-		stwma(d_space, r3, pwd, pmc_store_exc);
-		stwma(d_space, r4, pwd, pmc_store_exc);
+		load4(r1, pmc_done);
+		load4(r2, pmc_byte4);		// bytecopy src-4
+		load4(r3, pmc_byte8);		// bytecopy src-8
+		load4(r4, pmc_byte12);		// bytecopy src-12
+		store4(r1, pmc_done);
+		store4(r2, pmc_done);
+		store4(r3, pmc_done);
+		store4(r4, pmc_done);
 		len -= 4*sizeof(unsigned int);
 	}
 
-	pcs = (unsigned char *)pws;
-	pcd = (unsigned char *)pwd;
-
 byte_copy:
 	while (len) {
-		/* *pcd++ = *pcs++; */
-		ldbma(s_space, pcs, t3, pmc_load_exc);
-		stbma(d_space, t3, pcd, pmc_store_exc);
+		load1(t3, pmc_done);
+		store1(t3, pmc_done);
 		len--;
 	}
 
-	return PA_MEMCPY_OK;
+	return 0;
 
 unaligned_copy:
-	/* possibly we are aligned on a word, but not on a double... */
-	if (likely((t1 & (sizeof(unsigned int)-1)) == 0)) {
-		t2 = src & (sizeof(unsigned int) - 1);
-
-		if (unlikely(t2 != 0)) {
-			t2 = sizeof(unsigned int) - t2;
-			while (t2) {
-				/* *pcd++ = *pcs++; */
-				ldbma(s_space, pcs, t3, pmc_load_exc);
-				stbma(d_space, t3, pcd, pmc_store_exc);
-				len--;
-				t2--;
-			}
-		}
-
-		pws = (unsigned int *)pcs;
-		pwd = (unsigned int *)pcd;
-		goto word_copy;
-	}
-
-	/* Align the destination.  */
 	if (unlikely((dst & (sizeof(unsigned int) - 1)) != 0)) {
 		t2 = sizeof(unsigned int) - (dst & (sizeof(unsigned int) - 1));
 		while (t2) {
-			/* *pcd++ = *pcs++; */
-			ldbma(s_space, pcs, t3, pmc_load_exc);
-			stbma(d_space, t3, pcd, pmc_store_exc);
+			load1(t3, pmc_done);
+			store1(t3, pmc_done);
 			len--;
 			t2--;
 		}
-		dst = (unsigned long)pcd;
-		src = (unsigned long)pcs;
 	}
 
 	ret = copy_dstaligned(dst, src, len / sizeof(unsigned int));
-	if (ret)
-		return ret;
+	if (unlikely(ret)) {
+		if (ret & 1)
+			return len - (ret - 1 - dst);
+		src += ret - dst;
+		len -= ret - dst;
+		dst = ret;
+		goto byte_copy;
+	}
 
-	pcs += (len & -sizeof(unsigned int));
-	pcd += (len & -sizeof(unsigned int));
+	src += (len & -sizeof(unsigned int));
+	dst += (len & -sizeof(unsigned int));
 	len %= sizeof(unsigned int);
 
-	preserve_branch(handle_load_error);
-	preserve_branch(handle_store_error);
-
+	preserve_branch(handle_error1);
+	preserve_branch(handle_error2);
+	preserve_branch(handle_error3);
+	preserve_branch(handle_error4);
 	goto byte_copy;
 
-handle_load_error:
-	__asm__ __volatile__ ("pmc_load_exc:\n");
-	return PA_MEMCPY_LOAD_ERROR;
-
-handle_store_error:
-	__asm__ __volatile__ ("pmc_store_exc:\n");
-	return PA_MEMCPY_STORE_ERROR;
-}
-
-
-/* Returns 0 for success, otherwise, returns number of bytes not transferred. */
-static unsigned long pa_memcpy(void *dstp, const void *srcp, unsigned long len)
-{
-	unsigned long ret, fault_addr, reference;
-	struct exception_data *d;
-
-	ret = pa_memcpy_internal(dstp, srcp, len);
-	if (likely(ret == PA_MEMCPY_OK))
-		return 0;
-
-	/* if a load or store fault occured we can get the faulty addr */
-	d = this_cpu_ptr(&exception_data);
-	fault_addr = d->fault_addr;
-
-	/* error in load or store? */
-	if (ret == PA_MEMCPY_LOAD_ERROR)
-		reference = (unsigned long) srcp;
-	else
-		reference = (unsigned long) dstp;
-
-	DPRINTF("pa_memcpy: fault type = %lu, len=%lu fault_addr=%lu ref=%lu\n",
-		ret, len, fault_addr, reference);
-
-	if (fault_addr >= reference)
-		return len - (fault_addr - reference);
-	else
-		return len;
+handle_error1:
+	__asm__ __volatile__ ("pmc_done:\n");
+	return end - dst;
+handle_error2:
+	__asm__ __volatile__ ("pmc_byte4:\n");
+	src -= 4;
+	goto byte_copy;
+handle_error3:
+	__asm__ __volatile__ ("pmc_byte8:\n");
+	src -= 8;
+	goto byte_copy;
+handle_error4:
+	__asm__ __volatile__ ("pmc_byte12:\n");
+	src -= 12;
+	goto byte_copy;
 }
 
 #ifdef __KERNEL__
-unsigned long __copy_to_user(void __user *dst, const void *src,
-			     unsigned long len)
+unsigned long raw_copy_to_user(void __user *dst, const void *src,
+			       unsigned long len)
 {
 	mtsp(get_kernel_space(), 1);
 	mtsp(get_user_space(), 2);
 	return pa_memcpy((void __force *)dst, src, len);
 }
-EXPORT_SYMBOL(__copy_to_user);
+EXPORT_SYMBOL(raw_copy_to_user);
 
-unsigned long __copy_from_user(void *dst, const void __user *src,
+unsigned long raw_copy_from_user(void *dst, const void __user *src,
 			       unsigned long len)
 {
 	mtsp(get_user_space(), 1);
 	mtsp(get_kernel_space(), 2);
 	return pa_memcpy(dst, (void __force *)src, len);
 }
-EXPORT_SYMBOL(__copy_from_user);
+EXPORT_SYMBOL(raw_copy_from_user);
 
-unsigned long copy_in_user(void __user *dst, const void __user *src, unsigned long len)
+unsigned long raw_copy_in_user(void __user *dst, const void __user *src, unsigned long len)
 {
 	mtsp(get_user_space(), 1);
 	mtsp(get_user_space(), 2);
@@ -523,7 +391,7 @@ void * memcpy(void * dst,const void *src, size_t count)
 	return dst;
 }
 
-EXPORT_SYMBOL(copy_in_user);
+EXPORT_SYMBOL(raw_copy_in_user);
 EXPORT_SYMBOL(memcpy);
 
 long probe_kernel_read(void *dst, const void *src, size_t size)
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 4940917..191451f 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -167,6 +167,7 @@
 	select HAVE_ARCH_HARDENED_USERCOPY
 	select HAVE_KERNEL_GZIP
 	select HAVE_CONTEXT_TRACKING if PPC64
+	select ARCH_HAS_RAW_COPY_USER
 
 config GENERIC_CSUM
 	def_bool n
diff --git a/arch/powerpc/include/asm/extable.h b/arch/powerpc/include/asm/extable.h
new file mode 100644
index 0000000..07cc45c
--- /dev/null
+++ b/arch/powerpc/include/asm/extable.h
@@ -0,0 +1,29 @@
+#ifndef _ARCH_POWERPC_EXTABLE_H
+#define _ARCH_POWERPC_EXTABLE_H
+
+/*
+ * The exception table consists of pairs of relative addresses: the first is
+ * the address of an instruction that is allowed to fault, and the second is
+ * the address at which the program should continue.  No registers are
+ * modified, so it is entirely up to the continuation code to figure out what
+ * to do.
+ *
+ * All the routines below use bits of fixup code that are out of line with the
+ * main instruction path.  This means when everything is well, we don't even
+ * have to jump over them.  Further, they do not intrude on our cache or tlb
+ * entries.
+ */
+
+#define ARCH_HAS_RELATIVE_EXTABLE
+
+struct exception_table_entry {
+	int insn;
+	int fixup;
+};
+
+static inline unsigned long extable_fixup(const struct exception_table_entry *x)
+{
+	return (unsigned long)&x->fixup + x->fixup;
+}
+
+#endif
diff --git a/arch/powerpc/include/asm/uaccess.h b/arch/powerpc/include/asm/uaccess.h
index 3904040..5c0d8a8 100644
--- a/arch/powerpc/include/asm/uaccess.h
+++ b/arch/powerpc/include/asm/uaccess.h
@@ -5,6 +5,7 @@
 #include <asm/ppc_asm.h>
 #include <asm/processor.h>
 #include <asm/page.h>
+#include <asm/extable.h>
 
 /*
  * The fs value determines whether argument validity checking should be
@@ -56,31 +57,6 @@
 	 __access_ok((__force unsigned long)(addr), (size), get_fs()))
 
 /*
- * The exception table consists of pairs of relative addresses: the first is
- * the address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out what
- * to do.
- *
- * All the routines below use bits of fixup code that are out of line with the
- * main instruction path.  This means when everything is well, we don't even
- * have to jump over them.  Further, they do not intrude on our cache or tlb
- * entries.
- */
-
-#define ARCH_HAS_RELATIVE_EXTABLE
-
-struct exception_table_entry {
-	int insn;
-	int fixup;
-};
-
-static inline unsigned long extable_fixup(const struct exception_table_entry *x)
-{
-	return (unsigned long)&x->fixup + x->fixup;
-}
-
-/*
  * These are the main single-value transfer routines.  They automatically
  * use the right size if we just have the right pointer type.
  *
@@ -293,42 +269,19 @@ extern unsigned long __copy_tofrom_user(void __user *to,
 
 #ifndef __powerpc64__
 
-static inline unsigned long copy_from_user(void *to,
-		const void __user *from, unsigned long n)
-{
-	if (likely(access_ok(VERIFY_READ, from, n))) {
-		check_object_size(to, n, false);
-		return __copy_tofrom_user((__force void __user *)to, from, n);
-	}
-	memset(to, 0, n);
-	return n;
-}
-
-static inline unsigned long copy_to_user(void __user *to,
-		const void *from, unsigned long n)
-{
-	if (access_ok(VERIFY_WRITE, to, n)) {
-		check_object_size(from, n, true);
-		return __copy_tofrom_user(to, (__force void __user *)from, n);
-	}
-	return n;
-}
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 #else /* __powerpc64__ */
 
-#define __copy_in_user(to, from, size) \
-	__copy_tofrom_user((to), (from), (size))
-
-extern unsigned long copy_from_user(void *to, const void __user *from,
-				    unsigned long n);
-extern unsigned long copy_to_user(void __user *to, const void *from,
-				  unsigned long n);
-extern unsigned long copy_in_user(void __user *to, const void __user *from,
-				  unsigned long n);
-
+static inline unsigned long
+raw_copy_in_user(void __user *to, const void __user *from, unsigned long n)
+{
+	return __copy_tofrom_user(to, from, n);
+}
 #endif /* __powerpc64__ */
 
-static inline unsigned long __copy_from_user_inatomic(void *to,
+static inline unsigned long raw_copy_from_user(void *to,
 		const void __user *from, unsigned long n)
 {
 	if (__builtin_constant_p(n) && (n <= 8)) {
@@ -352,12 +305,10 @@ static inline unsigned long __copy_from_user_inatomic(void *to,
 			return 0;
 	}
 
-	check_object_size(to, n, false);
-
 	return __copy_tofrom_user((__force void __user *)to, from, n);
 }
 
-static inline unsigned long __copy_to_user_inatomic(void __user *to,
+static inline unsigned long raw_copy_to_user(void __user *to,
 		const void *from, unsigned long n)
 {
 	if (__builtin_constant_p(n) && (n <= 8)) {
@@ -381,25 +332,9 @@ static inline unsigned long __copy_to_user_inatomic(void __user *to,
 			return 0;
 	}
 
-	check_object_size(from, n, true);
-
 	return __copy_tofrom_user(to, (__force const void __user *)from, n);
 }
 
-static inline unsigned long __copy_from_user(void *to,
-		const void __user *from, unsigned long size)
-{
-	might_fault();
-	return __copy_from_user_inatomic(to, from, size);
-}
-
-static inline unsigned long __copy_to_user(void __user *to,
-		const void *from, unsigned long size)
-{
-	might_fault();
-	return __copy_to_user_inatomic(to, from, size);
-}
-
 extern unsigned long __clear_user(void __user *addr, unsigned long size);
 
 static inline unsigned long clear_user(void __user *addr, unsigned long size)
diff --git a/arch/powerpc/lib/Makefile b/arch/powerpc/lib/Makefile
index 0e649d7..99275e6 100644
--- a/arch/powerpc/lib/Makefile
+++ b/arch/powerpc/lib/Makefile
@@ -14,7 +14,7 @@
 
 obj-$(CONFIG_PPC32)	+= div64.o copy_32.o
 
-obj64-y	+= copypage_64.o copyuser_64.o usercopy_64.o mem_64.o hweight_64.o \
+obj64-y	+= copypage_64.o copyuser_64.o mem_64.o hweight_64.o \
 	   copyuser_power7.o string_64.o copypage_power7.o memcpy_power7.o \
 	   memcpy_64.o memcmp_64.o
 
diff --git a/arch/powerpc/lib/copy_32.S b/arch/powerpc/lib/copy_32.S
index ff0d894..f015b06 100644
--- a/arch/powerpc/lib/copy_32.S
+++ b/arch/powerpc/lib/copy_32.S
@@ -477,18 +477,6 @@
 	bdnz	130b
 /* then clear out the destination: r3 bytes starting at 4(r6) */
 132:	mfctr	r3
-	srwi.	r0,r3,2
-	li	r9,0
-	mtctr	r0
-	beq	113f
-112:	stwu	r9,4(r6)
-	bdnz	112b
-113:	andi.	r0,r3,3
-	mtctr	r0
-	beq	120f
-114:	stb	r9,4(r6)
-	addi	r6,r6,1
-	bdnz	114b
 120:	blr
 
 	EX_TABLE(30b,108b)
diff --git a/arch/powerpc/lib/copyuser_64.S b/arch/powerpc/lib/copyuser_64.S
index aee6e24..08da06e 100644
--- a/arch/powerpc/lib/copyuser_64.S
+++ b/arch/powerpc/lib/copyuser_64.S
@@ -319,32 +319,9 @@
 	blr
 
 /*
- * here we have trapped again, need to clear ctr bytes starting at r3
+ * here we have trapped again, amount remaining is in ctr.
  */
-143:	mfctr	r5
-	li	r0,0
-	mr	r4,r3
-	mr	r3,r5		/* return the number of bytes not copied */
-1:	andi.	r9,r4,7
-	beq	3f
-90:	stb	r0,0(r4)
-	addic.	r5,r5,-1
-	addi	r4,r4,1
-	bne	1b
-	blr
-3:	cmpldi	cr1,r5,8
-	srdi	r9,r5,3
-	andi.	r5,r5,7
-	blt	cr1,93f
-	mtctr	r9
-91:	std	r0,0(r4)
-	addi	r4,r4,8
-	bdnz	91b
-93:	beqlr
-	mtctr	r5	
-92:	stb	r0,0(r4)
-	addi	r4,r4,1
-	bdnz	92b
+143:	mfctr	r3
 	blr
 
 /*
@@ -389,10 +366,7 @@
 	ld	r5,-8(r1)
 	add	r6,r6,r5
 	subf	r3,r3,r6	/* #bytes not copied */
-190:
-191:
-192:
-	blr			/* #bytes not copied in r3 */
+	blr
 
 	EX_TABLE(20b,120b)
 	EX_TABLE(220b,320b)
@@ -451,9 +425,6 @@
 	EX_TABLE(88b,188b)
 	EX_TABLE(43b,143b)
 	EX_TABLE(89b,189b)
-	EX_TABLE(90b,190b)
-	EX_TABLE(91b,191b)
-	EX_TABLE(92b,192b)
 
 /*
  * Routine to copy a whole page of data, optimized for POWER4.
diff --git a/arch/powerpc/lib/usercopy_64.c b/arch/powerpc/lib/usercopy_64.c
deleted file mode 100644
index 9bd3a3d..0000000
--- a/arch/powerpc/lib/usercopy_64.c
+++ /dev/null
@@ -1,41 +0,0 @@
-/*
- * Functions which are too large to be inlined.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
- */
-#include <linux/module.h>
-#include <linux/uaccess.h>
-
-unsigned long copy_from_user(void *to, const void __user *from, unsigned long n)
-{
-	if (likely(access_ok(VERIFY_READ, from, n)))
-		n = __copy_from_user(to, from, n);
-	else
-		memset(to, 0, n);
-	return n;
-}
-
-unsigned long copy_to_user(void __user *to, const void *from, unsigned long n)
-{
-	if (likely(access_ok(VERIFY_WRITE, to, n)))
-		n = __copy_to_user(to, from, n);
-	return n;
-}
-
-unsigned long copy_in_user(void __user *to, const void __user *from,
-			   unsigned long n)
-{
-	might_sleep();
-	if (likely(access_ok(VERIFY_READ, from, n) &&
-	    access_ok(VERIFY_WRITE, to, n)))
-		n =__copy_tofrom_user(to, from, n);
-	return n;
-}
-
-EXPORT_SYMBOL(copy_from_user);
-EXPORT_SYMBOL(copy_to_user);
-EXPORT_SYMBOL(copy_in_user);
-
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index a2dcef0..0724dbf 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -178,6 +178,7 @@
 	select ARCH_HAS_SCALED_CPUTIME
 	select VIRT_TO_BUS
 	select HAVE_NMI
+	select ARCH_HAS_RAW_COPY_USER
 
 
 config SCHED_OMIT_FRAME_POINTER
diff --git a/arch/s390/include/asm/extable.h b/arch/s390/include/asm/extable.h
new file mode 100644
index 0000000..16cfe2d
--- /dev/null
+++ b/arch/s390/include/asm/extable.h
@@ -0,0 +1,28 @@
+#ifndef __S390_EXTABLE_H
+#define __S390_EXTABLE_H
+/*
+ * The exception table consists of pairs of addresses: the first is the
+ * address of an instruction that is allowed to fault, and the second is
+ * the address at which the program should continue.  No registers are
+ * modified, so it is entirely up to the continuation code to figure out
+ * what to do.
+ *
+ * All the routines below use bits of fixup code that are out of line
+ * with the main instruction path.  This means when everything is well,
+ * we don't even have to jump over them.  Further, they do not intrude
+ * on our cache or tlb entries.
+ */
+
+struct exception_table_entry
+{
+	int insn, fixup;
+};
+
+static inline unsigned long extable_fixup(const struct exception_table_entry *x)
+{
+	return (unsigned long)&x->fixup + x->fixup;
+}
+
+#define ARCH_HAS_RELATIVE_EXTABLE
+
+#endif
diff --git a/arch/s390/include/asm/uaccess.h b/arch/s390/include/asm/uaccess.h
index 7228ed8..72d69b9 100644
--- a/arch/s390/include/asm/uaccess.h
+++ b/arch/s390/include/asm/uaccess.h
@@ -14,6 +14,7 @@
  */
 #include <asm/processor.h>
 #include <asm/ctl_reg.h>
+#include <asm/extable.h>
 
 
 /*
@@ -59,72 +60,14 @@ static inline int __range_ok(unsigned long addr, unsigned long size)
 
 #define access_ok(type, addr, size) __access_ok(addr, size)
 
-/*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
+unsigned long __must_check
+raw_copy_from_user(void *to, const void __user *from, unsigned long n);
 
-struct exception_table_entry
-{
-	int insn, fixup;
-};
+unsigned long __must_check
+raw_copy_to_user(void __user *to, const void *from, unsigned long n);
 
-static inline unsigned long extable_fixup(const struct exception_table_entry *x)
-{
-	return (unsigned long)&x->fixup + x->fixup;
-}
-
-#define ARCH_HAS_RELATIVE_EXTABLE
-
-/**
- * __copy_from_user: - Copy a block of data from user space, with less checking.
- * @to:   Destination address, in kernel space.
- * @from: Source address, in user space.
- * @n:	  Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from user space to kernel space.  Caller must check
- * the specified block with access_ok() before calling this function.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- *
- * If some data could not be copied, this function will pad the copied
- * data to the requested size using zero bytes.
- */
-unsigned long __must_check __copy_from_user(void *to, const void __user *from,
-					    unsigned long n);
-
-/**
- * __copy_to_user: - Copy a block of data into user space, with less checking.
- * @to:   Destination address, in user space.
- * @from: Source address, in kernel space.
- * @n:	  Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from kernel space to user space.  Caller must check
- * the specified block with access_ok() before calling this function.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- */
-unsigned long __must_check __copy_to_user(void __user *to, const void *from,
-					  unsigned long n);
-
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 #ifdef CONFIG_HAVE_MARCH_Z10_FEATURES
 
@@ -213,13 +156,13 @@ static inline int __get_user_fn(void *x, const void __user *ptr, unsigned long s
 
 static inline int __put_user_fn(void *x, void __user *ptr, unsigned long size)
 {
-	size = __copy_to_user(ptr, x, size);
+	size = raw_copy_to_user(ptr, x, size);
 	return size ? -EFAULT : 0;
 }
 
 static inline int __get_user_fn(void *x, const void __user *ptr, unsigned long size)
 {
-	size = __copy_from_user(x, ptr, size);
+	size = raw_copy_from_user(x, ptr, size);
 	return size ? -EFAULT : 0;
 }
 
@@ -309,77 +252,8 @@ int __get_user_bad(void) __attribute__((noreturn));
 #define __put_user_unaligned __put_user
 #define __get_user_unaligned __get_user
 
-extern void __compiletime_error("usercopy buffer size is too small")
-__bad_copy_user(void);
-
-static inline void copy_user_overflow(int size, unsigned long count)
-{
-	WARN(1, "Buffer overflow detected (%d < %lu)!\n", size, count);
-}
-
-/**
- * copy_to_user: - Copy a block of data into user space.
- * @to:   Destination address, in user space.
- * @from: Source address, in kernel space.
- * @n:    Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from kernel space to user space.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- */
-static inline unsigned long __must_check
-copy_to_user(void __user *to, const void *from, unsigned long n)
-{
-	might_fault();
-	return __copy_to_user(to, from, n);
-}
-
-/**
- * copy_from_user: - Copy a block of data from user space.
- * @to:   Destination address, in kernel space.
- * @from: Source address, in user space.
- * @n:    Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from user space to kernel space.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- *
- * If some data could not be copied, this function will pad the copied
- * data to the requested size using zero bytes.
- */
-static inline unsigned long __must_check
-copy_from_user(void *to, const void __user *from, unsigned long n)
-{
-	unsigned int sz = __compiletime_object_size(to);
-
-	might_fault();
-	if (unlikely(sz != -1 && sz < n)) {
-		if (!__builtin_constant_p(n))
-			copy_user_overflow(sz, n);
-		else
-			__bad_copy_user();
-		return n;
-	}
-	return __copy_from_user(to, from, n);
-}
-
 unsigned long __must_check
-__copy_in_user(void __user *to, const void __user *from, unsigned long n);
-
-static inline unsigned long __must_check
-copy_in_user(void __user *to, const void __user *from, unsigned long n)
-{
-	might_fault();
-	return __copy_in_user(to, from, n);
-}
+raw_copy_in_user(void __user *to, const void __user *from, unsigned long n);
 
 /*
  * Copy a null terminated string from userspace.
diff --git a/arch/s390/lib/uaccess.c b/arch/s390/lib/uaccess.c
index f481fcd..b55172c 100644
--- a/arch/s390/lib/uaccess.c
+++ b/arch/s390/lib/uaccess.c
@@ -26,7 +26,7 @@ static inline unsigned long copy_from_user_mvcos(void *x, const void __user *ptr
 	tmp1 = -4096UL;
 	asm volatile(
 		"0: .insn ss,0xc80000000000,0(%0,%2),0(%1),0\n"
-		"9: jz    7f\n"
+		"6: jz    4f\n"
 		"1: algr  %0,%3\n"
 		"   slgr  %1,%3\n"
 		"   slgr  %2,%3\n"
@@ -37,21 +37,11 @@ static inline unsigned long copy_from_user_mvcos(void *x, const void __user *ptr
 		"   clgr  %0,%4\n"	/* copy crosses next page boundary? */
 		"   jnh   4f\n"
 		"3: .insn ss,0xc80000000000,0(%4,%2),0(%1),0\n"
-		"10:slgr  %0,%4\n"
-		"   algr  %2,%4\n"
-		"4: lghi  %4,-1\n"
-		"   algr  %4,%0\n"	/* copy remaining size, subtract 1 */
-		"   bras  %3,6f\n"	/* memset loop */
-		"   xc    0(1,%2),0(%2)\n"
-		"5: xc    0(256,%2),0(%2)\n"
-		"   la    %2,256(%2)\n"
-		"6: aghi  %4,-256\n"
-		"   jnm   5b\n"
-		"   ex    %4,0(%3)\n"
-		"   j     8f\n"
-		"7: slgr  %0,%0\n"
-		"8:\n"
-		EX_TABLE(0b,2b) EX_TABLE(3b,4b) EX_TABLE(9b,2b) EX_TABLE(10b,4b)
+		"7: slgr  %0,%4\n"
+		"   j     5f\n"
+		"4: slgr  %0,%0\n"
+		"5:\n"
+		EX_TABLE(0b,2b) EX_TABLE(3b,5b) EX_TABLE(6b,2b) EX_TABLE(7b,5b)
 		: "+a" (size), "+a" (ptr), "+a" (x), "+a" (tmp1), "=a" (tmp2)
 		: "d" (reg0) : "cc", "memory");
 	return size;
@@ -67,49 +57,38 @@ static inline unsigned long copy_from_user_mvcp(void *x, const void __user *ptr,
 	asm volatile(
 		"   sacf  0\n"
 		"0: mvcp  0(%0,%2),0(%1),%3\n"
-		"10:jz    8f\n"
+		"7: jz    5f\n"
 		"1: algr  %0,%3\n"
 		"   la    %1,256(%1)\n"
 		"   la    %2,256(%2)\n"
 		"2: mvcp  0(%0,%2),0(%1),%3\n"
-		"11:jnz   1b\n"
-		"   j     8f\n"
+		"8: jnz   1b\n"
+		"   j     5f\n"
 		"3: la    %4,255(%1)\n"	/* %4 = ptr + 255 */
 		"   lghi  %3,-4096\n"
 		"   nr    %4,%3\n"	/* %4 = (ptr + 255) & -4096 */
 		"   slgr  %4,%1\n"
 		"   clgr  %0,%4\n"	/* copy crosses next page boundary? */
-		"   jnh   5f\n"
+		"   jnh   6f\n"
 		"4: mvcp  0(%4,%2),0(%1),%3\n"
-		"12:slgr  %0,%4\n"
-		"   algr  %2,%4\n"
-		"5: lghi  %4,-1\n"
-		"   algr  %4,%0\n"	/* copy remaining size, subtract 1 */
-		"   bras  %3,7f\n"	/* memset loop */
-		"   xc    0(1,%2),0(%2)\n"
-		"6: xc    0(256,%2),0(%2)\n"
-		"   la    %2,256(%2)\n"
-		"7: aghi  %4,-256\n"
-		"   jnm   6b\n"
-		"   ex    %4,0(%3)\n"
-		"   j     9f\n"
-		"8: slgr  %0,%0\n"
-		"9: sacf  768\n"
-		EX_TABLE(0b,3b) EX_TABLE(2b,3b) EX_TABLE(4b,5b)
-		EX_TABLE(10b,3b) EX_TABLE(11b,3b) EX_TABLE(12b,5b)
+		"9: slgr  %0,%4\n"
+		"   j     6f\n"
+		"5: slgr  %0,%0\n"
+		"6: sacf  768\n"
+		EX_TABLE(0b,3b) EX_TABLE(2b,3b) EX_TABLE(4b,6b)
+		EX_TABLE(7b,3b) EX_TABLE(8b,3b) EX_TABLE(9b,6b)
 		: "+a" (size), "+a" (ptr), "+a" (x), "+a" (tmp1), "=a" (tmp2)
 		: : "cc", "memory");
 	return size;
 }
 
-unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n)
+unsigned long raw_copy_from_user(void *to, const void __user *from, unsigned long n)
 {
-	check_object_size(to, n, false);
 	if (static_branch_likely(&have_mvcos))
 		return copy_from_user_mvcos(to, from, n);
 	return copy_from_user_mvcp(to, from, n);
 }
-EXPORT_SYMBOL(__copy_from_user);
+EXPORT_SYMBOL(raw_copy_from_user);
 
 static inline unsigned long copy_to_user_mvcos(void __user *ptr, const void *x,
 					       unsigned long size)
@@ -176,14 +155,13 @@ static inline unsigned long copy_to_user_mvcs(void __user *ptr, const void *x,
 	return size;
 }
 
-unsigned long __copy_to_user(void __user *to, const void *from, unsigned long n)
+unsigned long raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 {
-	check_object_size(from, n, true);
 	if (static_branch_likely(&have_mvcos))
 		return copy_to_user_mvcos(to, from, n);
 	return copy_to_user_mvcs(to, from, n);
 }
-EXPORT_SYMBOL(__copy_to_user);
+EXPORT_SYMBOL(raw_copy_to_user);
 
 static inline unsigned long copy_in_user_mvcos(void __user *to, const void __user *from,
 					       unsigned long size)
@@ -240,13 +218,13 @@ static inline unsigned long copy_in_user_mvc(void __user *to, const void __user
 	return size;
 }
 
-unsigned long __copy_in_user(void __user *to, const void __user *from, unsigned long n)
+unsigned long raw_copy_in_user(void __user *to, const void __user *from, unsigned long n)
 {
 	if (static_branch_likely(&have_mvcos))
 		return copy_in_user_mvcos(to, from, n);
 	return copy_in_user_mvc(to, from, n);
 }
-EXPORT_SYMBOL(__copy_in_user);
+EXPORT_SYMBOL(raw_copy_in_user);
 
 static inline unsigned long clear_user_mvcos(void __user *to, unsigned long size)
 {
diff --git a/arch/score/Kconfig b/arch/score/Kconfig
index 507d631..4d241cf 100644
--- a/arch/score/Kconfig
+++ b/arch/score/Kconfig
@@ -15,6 +15,7 @@
 	select MODULES_USE_ELF_REL
 	select CLONE_BACKWARDS
 	select CPU_NO_EFFICIENT_FFS
+	select ARCH_HAS_RAW_COPY_USER
 
 choice
 	prompt "System type"
diff --git a/arch/score/include/asm/Kbuild b/arch/score/include/asm/Kbuild
index 926943a..e3a8d0f 100644
--- a/arch/score/include/asm/Kbuild
+++ b/arch/score/include/asm/Kbuild
@@ -4,6 +4,7 @@
 generic-y += barrier.h
 generic-y += clkdev.h
 generic-y += current.h
+generic-y += extable.h
 generic-y += irq_work.h
 generic-y += mcs_spinlock.h
 generic-y += mm-arch-hooks.h
diff --git a/arch/score/include/asm/extable.h b/arch/score/include/asm/extable.h
deleted file mode 100644
index c4423ccf..0000000
--- a/arch/score/include/asm/extable.h
+++ /dev/null
@@ -1,11 +0,0 @@
-#ifndef _ASM_SCORE_EXTABLE_H
-#define _ASM_SCORE_EXTABLE_H
-
-struct exception_table_entry {
-	unsigned long insn;
-	unsigned long fixup;
-};
-
-struct pt_regs;
-extern int fixup_exception(struct pt_regs *regs);
-#endif
diff --git a/arch/score/include/asm/uaccess.h b/arch/score/include/asm/uaccess.h
index 7a6c698..916e5db 100644
--- a/arch/score/include/asm/uaccess.h
+++ b/arch/score/include/asm/uaccess.h
@@ -295,61 +295,19 @@ extern void __put_user_unknown(void);
 extern int __copy_tofrom_user(void *to, const void *from, unsigned long len);
 
 static inline unsigned long
-copy_from_user(void *to, const void *from, unsigned long len)
+raw_copy_from_user(void *to, const void __user *from, unsigned long len)
 {
-	unsigned long res = len;
-
-	if (likely(access_ok(VERIFY_READ, from, len)))
-		res = __copy_tofrom_user(to, from, len);
-
-	if (unlikely(res))
-		memset(to + (len - res), 0, res);
-
-	return res;
+	return __copy_tofrom_user(to, (__force const void *)from, len);
 }
 
 static inline unsigned long
-copy_to_user(void *to, const void *from, unsigned long len)
+raw_copy_to_user(void __user *to, const void *from, unsigned long len)
 {
-	if (likely(access_ok(VERIFY_WRITE, to, len)))
-		len = __copy_tofrom_user(to, from, len);
-
-	return len;
+	return __copy_tofrom_user((__force void *)to, from, len);
 }
 
-static inline unsigned long
-__copy_from_user(void *to, const void *from, unsigned long len)
-{
-	unsigned long left = __copy_tofrom_user(to, from, len);
-	if (unlikely(left))
-		memset(to + (len - left), 0, left);
-	return left;
-}
-
-#define __copy_to_user(to, from, len)		\
-		__copy_tofrom_user((to), (from), (len))
-
-static inline unsigned long
-__copy_to_user_inatomic(void *to, const void *from, unsigned long len)
-{
-	return __copy_to_user(to, from, len);
-}
-
-static inline unsigned long
-__copy_from_user_inatomic(void *to, const void *from, unsigned long len)
-{
-	return __copy_tofrom_user(to, from, len);
-}
-
-#define __copy_in_user(to, from, len)	__copy_tofrom_user(to, from, len)
-
-static inline unsigned long
-copy_in_user(void *to, const void *from, unsigned long len)
-{
-	if (access_ok(VERIFY_READ, from, len) &&
-		      access_ok(VERFITY_WRITE, to, len))
-		return __copy_tofrom_user(to, from, len);
-}
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 /*
  * __clear_user: - Zero a block of memory in user space, with less checking.
diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index ee08695..c689645 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -48,6 +48,7 @@
 	select HAVE_ARCH_AUDITSYSCALL
 	select HAVE_FUTEX_CMPXCHG if FUTEX
 	select HAVE_NMI
+	select ARCH_HAS_RAW_COPY_USER
 	help
 	  The SuperH is a RISC processor targeted for use in embedded systems
 	  and consumer electronics; it was also used in the Sega Dreamcast
diff --git a/arch/sh/include/asm/extable.h b/arch/sh/include/asm/extable.h
new file mode 100644
index 0000000..df2ee2f
--- /dev/null
+++ b/arch/sh/include/asm/extable.h
@@ -0,0 +1,10 @@
+#ifndef __ASM_SH_EXTABLE_H
+#define __ASM_SH_EXTABLE_H
+
+#include <asm-generic/extable.h>
+
+#if defined(CONFIG_SUPERH64) && defined(CONFIG_MMU)
+#define ARCH_HAS_SEARCH_EXTABLE
+#endif
+
+#endif
diff --git a/arch/sh/include/asm/uaccess.h b/arch/sh/include/asm/uaccess.h
index 89a28df..2722b61 100644
--- a/arch/sh/include/asm/uaccess.h
+++ b/arch/sh/include/asm/uaccess.h
@@ -2,6 +2,7 @@
 #define __ASM_SH_UACCESS_H
 
 #include <asm/segment.h>
+#include <asm/extable.h>
 
 #define __addr_ok(addr) \
 	((unsigned long __force)(addr) < current_thread_info()->addr_limit.seg)
@@ -107,19 +108,18 @@ extern __must_check long strnlen_user(const char __user *str, long n);
 __kernel_size_t __copy_user(void *to, const void *from, __kernel_size_t n);
 
 static __always_inline unsigned long
-__copy_from_user(void *to, const void __user *from, unsigned long n)
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
 {
 	return __copy_user(to, (__force void *)from, n);
 }
 
 static __always_inline unsigned long __must_check
-__copy_to_user(void __user *to, const void *from, unsigned long n)
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 {
 	return __copy_user((__force void *)to, from, n);
 }
-
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 /*
  * Clear the area and return remaining number of bytes
@@ -139,55 +139,6 @@ __kernel_size_t __clear_user(void *addr, __kernel_size_t size);
 	__cl_size;							\
 })
 
-static inline unsigned long
-copy_from_user(void *to, const void __user *from, unsigned long n)
-{
-	unsigned long __copy_from = (unsigned long) from;
-	__kernel_size_t __copy_size = (__kernel_size_t) n;
-
-	if (__copy_size && __access_ok(__copy_from, __copy_size))
-		__copy_size = __copy_user(to, from, __copy_size);
-
-	if (unlikely(__copy_size))
-		memset(to + (n - __copy_size), 0, __copy_size);
-
-	return __copy_size;
-}
-
-static inline unsigned long
-copy_to_user(void __user *to, const void *from, unsigned long n)
-{
-	unsigned long __copy_to = (unsigned long) to;
-	__kernel_size_t __copy_size = (__kernel_size_t) n;
-
-	if (__copy_size && __access_ok(__copy_to, __copy_size))
-		return __copy_user(to, from, __copy_size);
-
-	return __copy_size;
-}
-
-/*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
-struct exception_table_entry {
-	unsigned long insn, fixup;
-};
-
-#if defined(CONFIG_SUPERH64) && defined(CONFIG_MMU)
-#define ARCH_HAS_SEARCH_EXTABLE
-#endif
-
-int fixup_exception(struct pt_regs *regs);
-
 extern void *set_exception_table_vec(unsigned int vec, void *handler);
 
 static inline void *set_exception_table_evt(unsigned int evt, void *handler)
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 68ac5c7..6835304 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -45,6 +45,7 @@
 	select HAVE_ARCH_HARDENED_USERCOPY
 	select PROVE_LOCKING_SMALL if PROVE_LOCKING
 	select ARCH_WANT_RELAX_ORDER
+	select ARCH_HAS_RAW_COPY_USER
 
 config SPARC32
 	def_bool !64BIT
diff --git a/arch/sparc/include/asm/uaccess_32.h b/arch/sparc/include/asm/uaccess_32.h
index a59a1e8..12ebee2 100644
--- a/arch/sparc/include/asm/uaccess_32.h
+++ b/arch/sparc/include/asm/uaccess_32.h
@@ -71,8 +71,6 @@ struct exception_table_entry
 /* Returns 0 if exception not found and fixup otherwise.  */
 unsigned long search_extables_range(unsigned long addr, unsigned long *g2);
 
-void __ret_efault(void);
-
 /* Uh, these should become the main single-value transfer routines..
  * They automatically use the right size if we just have the right
  * pointer type..
@@ -237,39 +235,18 @@ int __get_user_bad(void);
 
 unsigned long __copy_user(void __user *to, const void __user *from, unsigned long size);
 
-static inline unsigned long copy_to_user(void __user *to, const void *from, unsigned long n)
+static inline unsigned long raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 {
-	if (n && __access_ok((unsigned long) to, n)) {
-		check_object_size(from, n, true);
-		return __copy_user(to, (__force void __user *) from, n);
-	} else
-		return n;
-}
-
-static inline unsigned long __copy_to_user(void __user *to, const void *from, unsigned long n)
-{
-	check_object_size(from, n, true);
 	return __copy_user(to, (__force void __user *) from, n);
 }
 
-static inline unsigned long copy_from_user(void *to, const void __user *from, unsigned long n)
-{
-	if (n && __access_ok((unsigned long) from, n)) {
-		check_object_size(to, n, false);
-		return __copy_user((__force void __user *) to, from, n);
-	} else {
-		memset(to, 0, n);
-		return n;
-	}
-}
-
-static inline unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n)
+static inline unsigned long raw_copy_from_user(void *to, const void __user *from, unsigned long n)
 {
 	return __copy_user((__force void __user *) to, from, n);
 }
 
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 static inline unsigned long __clear_user(void __user *addr, unsigned long size)
 {
diff --git a/arch/sparc/include/asm/uaccess_64.h b/arch/sparc/include/asm/uaccess_64.h
index 7afb4f6..6096d67 100644
--- a/arch/sparc/include/asm/uaccess_64.h
+++ b/arch/sparc/include/asm/uaccess_64.h
@@ -176,39 +176,19 @@ __asm__ __volatile__(							\
 
 int __get_user_bad(void);
 
-unsigned long __must_check ___copy_from_user(void *to,
+unsigned long __must_check raw_copy_from_user(void *to,
 					     const void __user *from,
 					     unsigned long size);
-static inline unsigned long __must_check
-copy_from_user(void *to, const void __user *from, unsigned long size)
-{
-	check_object_size(to, size, false);
 
-	return ___copy_from_user(to, from, size);
-}
-#define __copy_from_user copy_from_user
-
-unsigned long __must_check ___copy_to_user(void __user *to,
+unsigned long __must_check raw_copy_to_user(void __user *to,
 					   const void *from,
 					   unsigned long size);
-static inline unsigned long __must_check
-copy_to_user(void __user *to, const void *from, unsigned long size)
-{
-	check_object_size(from, size, true);
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
-	return ___copy_to_user(to, from, size);
-}
-#define __copy_to_user copy_to_user
-
-unsigned long __must_check ___copy_in_user(void __user *to,
+unsigned long __must_check raw_copy_in_user(void __user *to,
 					   const void __user *from,
 					   unsigned long size);
-static inline unsigned long __must_check
-copy_in_user(void __user *to, void __user *from, unsigned long size)
-{
-	return ___copy_in_user(to, from, size);
-}
-#define __copy_in_user copy_in_user
 
 unsigned long __must_check __clear_user(void __user *, unsigned long);
 
@@ -217,9 +197,6 @@ unsigned long __must_check __clear_user(void __user *, unsigned long);
 __must_check long strlen_user(const char __user *str);
 __must_check long strnlen_user(const char __user *str, long n);
 
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
-
 struct pt_regs;
 unsigned long compute_effective_address(struct pt_regs *,
 					unsigned int insn,
diff --git a/arch/sparc/kernel/head_32.S b/arch/sparc/kernel/head_32.S
index 7bb317b..7274e43 100644
--- a/arch/sparc/kernel/head_32.S
+++ b/arch/sparc/kernel/head_32.S
@@ -809,10 +809,3 @@
 	.word	0
 	.word	0
 	.word	t_irq14
-
-        .section        ".fixup",#alloc,#execinstr
-        .globl  __ret_efault
-__ret_efault:
-        ret
-         restore %g0, -EFAULT, %o0
-EXPORT_SYMBOL(__ret_efault)
diff --git a/arch/sparc/lib/GENpatch.S b/arch/sparc/lib/GENpatch.S
index fab9e89..95e2f1f 100644
--- a/arch/sparc/lib/GENpatch.S
+++ b/arch/sparc/lib/GENpatch.S
@@ -26,8 +26,8 @@
 	.type	generic_patch_copyops,#function
 generic_patch_copyops:
 	GEN_DO_PATCH(memcpy, GENmemcpy)
-	GEN_DO_PATCH(___copy_from_user, GENcopy_from_user)
-	GEN_DO_PATCH(___copy_to_user, GENcopy_to_user)
+	GEN_DO_PATCH(raw_copy_from_user, GENcopy_from_user)
+	GEN_DO_PATCH(raw_copy_to_user, GENcopy_to_user)
 	retl
 	 nop
 	.size	generic_patch_copyops,.-generic_patch_copyops
diff --git a/arch/sparc/lib/NG2patch.S b/arch/sparc/lib/NG2patch.S
index 28c36f0..56ccc19 100644
--- a/arch/sparc/lib/NG2patch.S
+++ b/arch/sparc/lib/NG2patch.S
@@ -26,8 +26,8 @@
 	.type	niagara2_patch_copyops,#function
 niagara2_patch_copyops:
 	NG_DO_PATCH(memcpy, NG2memcpy)
-	NG_DO_PATCH(___copy_from_user, NG2copy_from_user)
-	NG_DO_PATCH(___copy_to_user, NG2copy_to_user)
+	NG_DO_PATCH(raw_copy_from_user, NG2copy_from_user)
+	NG_DO_PATCH(raw_copy_to_user, NG2copy_to_user)
 	retl
 	 nop
 	.size	niagara2_patch_copyops,.-niagara2_patch_copyops
diff --git a/arch/sparc/lib/NG4patch.S b/arch/sparc/lib/NG4patch.S
index a114cbc..3cc0f8c 100644
--- a/arch/sparc/lib/NG4patch.S
+++ b/arch/sparc/lib/NG4patch.S
@@ -26,8 +26,8 @@
 	.type	niagara4_patch_copyops,#function
 niagara4_patch_copyops:
 	NG_DO_PATCH(memcpy, NG4memcpy)
-	NG_DO_PATCH(___copy_from_user, NG4copy_from_user)
-	NG_DO_PATCH(___copy_to_user, NG4copy_to_user)
+	NG_DO_PATCH(raw_copy_from_user, NG4copy_from_user)
+	NG_DO_PATCH(raw_copy_to_user, NG4copy_to_user)
 	retl
 	 nop
 	.size	niagara4_patch_copyops,.-niagara4_patch_copyops
diff --git a/arch/sparc/lib/NGpatch.S b/arch/sparc/lib/NGpatch.S
index 3b0674f..62ccda7 100644
--- a/arch/sparc/lib/NGpatch.S
+++ b/arch/sparc/lib/NGpatch.S
@@ -26,8 +26,8 @@
 	.type	niagara_patch_copyops,#function
 niagara_patch_copyops:
 	NG_DO_PATCH(memcpy, NGmemcpy)
-	NG_DO_PATCH(___copy_from_user, NGcopy_from_user)
-	NG_DO_PATCH(___copy_to_user, NGcopy_to_user)
+	NG_DO_PATCH(raw_copy_from_user, NGcopy_from_user)
+	NG_DO_PATCH(raw_copy_to_user, NGcopy_to_user)
 	retl
 	 nop
 	.size	niagara_patch_copyops,.-niagara_patch_copyops
diff --git a/arch/sparc/lib/U1copy_from_user.S b/arch/sparc/lib/U1copy_from_user.S
index bb6ff73..69f95c3 100644
--- a/arch/sparc/lib/U1copy_from_user.S
+++ b/arch/sparc/lib/U1copy_from_user.S
@@ -19,7 +19,7 @@
 	.text;			\
 	.align 4;
 
-#define FUNC_NAME		___copy_from_user
+#define FUNC_NAME		raw_copy_from_user
 #define LOAD(type,addr,dest)	type##a [addr] %asi, dest
 #define LOAD_BLK(addr,dest)	ldda [addr] ASI_BLK_AIUS, dest
 #define EX_RETVAL(x)		0
diff --git a/arch/sparc/lib/U1copy_to_user.S b/arch/sparc/lib/U1copy_to_user.S
index ed92ce73..26b9136 100644
--- a/arch/sparc/lib/U1copy_to_user.S
+++ b/arch/sparc/lib/U1copy_to_user.S
@@ -19,7 +19,7 @@
 	.text;			\
 	.align 4;
 
-#define FUNC_NAME		___copy_to_user
+#define FUNC_NAME		raw_copy_to_user
 #define STORE(type,src,addr)	type##a src, [addr] ASI_AIUS
 #define STORE_BLK(src,addr)	stda src, [addr] ASI_BLK_AIUS
 #define EX_RETVAL(x)		0
diff --git a/arch/sparc/lib/U3patch.S b/arch/sparc/lib/U3patch.S
index ecc3026..91cd653 100644
--- a/arch/sparc/lib/U3patch.S
+++ b/arch/sparc/lib/U3patch.S
@@ -26,8 +26,8 @@
 	.type	cheetah_patch_copyops,#function
 cheetah_patch_copyops:
 	ULTRA3_DO_PATCH(memcpy, U3memcpy)
-	ULTRA3_DO_PATCH(___copy_from_user, U3copy_from_user)
-	ULTRA3_DO_PATCH(___copy_to_user, U3copy_to_user)
+	ULTRA3_DO_PATCH(raw_copy_from_user, U3copy_from_user)
+	ULTRA3_DO_PATCH(raw_copy_to_user, U3copy_to_user)
 	retl
 	 nop
 	.size	cheetah_patch_copyops,.-cheetah_patch_copyops
diff --git a/arch/sparc/lib/copy_in_user.S b/arch/sparc/lib/copy_in_user.S
index 0252b21..1b73bb8 100644
--- a/arch/sparc/lib/copy_in_user.S
+++ b/arch/sparc/lib/copy_in_user.S
@@ -44,7 +44,7 @@
 	 * to copy register windows around during thread cloning.
 	 */
 
-ENTRY(___copy_in_user)	/* %o0=dst, %o1=src, %o2=len */
+ENTRY(raw_copy_in_user)	/* %o0=dst, %o1=src, %o2=len */
 	cmp		%o2, 0
 	be,pn		%XCC, 85f
 	 or		%o0, %o1, %o3
@@ -105,5 +105,5 @@
 	 add		%o0, 1, %o0
 	retl
 	 clr		%o0
-ENDPROC(___copy_in_user)
-EXPORT_SYMBOL(___copy_in_user)
+ENDPROC(raw_copy_in_user)
+EXPORT_SYMBOL(raw_copy_in_user)
diff --git a/arch/sparc/lib/copy_user.S b/arch/sparc/lib/copy_user.S
index cea644d..bc243ee 100644
--- a/arch/sparc/lib/copy_user.S
+++ b/arch/sparc/lib/copy_user.S
@@ -364,21 +364,7 @@
 97:
 	mov	%o2, %g3
 fixupretl:
-	sethi   %hi(PAGE_OFFSET), %g1
-	cmp	%o0, %g1
-	blu	1f
-	 cmp	%o1, %g1
-	bgeu	1f
-	 ld	[%g6 + TI_PREEMPT], %g1
-	cmp	%g1, 0
-	bne	1f
-	 nop
-	save	%sp, -64, %sp
-	mov	%i0, %o0
-	call	__bzero
-	 mov	%g3, %o1
-	restore
-1:	retl
+	retl
 	 mov	%g3, %o0
 
 /* exception routine sets %g2 to (broken_insn - first_insn)>>2 */
diff --git a/arch/tile/Kconfig b/arch/tile/Kconfig
index 4583c03..27808d8 100644
--- a/arch/tile/Kconfig
+++ b/arch/tile/Kconfig
@@ -33,6 +33,7 @@
 	select USER_STACKTRACE_SUPPORT
 	select USE_PMC if PERF_EVENTS
 	select VIRT_TO_BUS
+	select ARCH_HAS_RAW_COPY_USER
 
 config MMU
 	def_bool y
diff --git a/arch/tile/include/asm/Kbuild b/arch/tile/include/asm/Kbuild
index aa48b6e..24c44e9 100644
--- a/arch/tile/include/asm/Kbuild
+++ b/arch/tile/include/asm/Kbuild
@@ -7,6 +7,7 @@
 generic-y += emergency-restart.h
 generic-y += errno.h
 generic-y += exec.h
+generic-y += extable.h
 generic-y += fb.h
 generic-y += fcntl.h
 generic-y += hw_irq.h
diff --git a/arch/tile/include/asm/uaccess.h b/arch/tile/include/asm/uaccess.h
index 14ea3d1..a803f6b 100644
--- a/arch/tile/include/asm/uaccess.h
+++ b/arch/tile/include/asm/uaccess.h
@@ -98,24 +98,7 @@ int __range_ok(unsigned long addr, unsigned long size);
 	likely(__range_ok((unsigned long)(addr), (size)) == 0);	\
 })
 
-/*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
-
-struct exception_table_entry {
-	unsigned long insn, fixup;
-};
-
-extern int fixup_exception(struct pt_regs *regs);
+#include <asm/extable.h>
 
 /*
  * This is a type: either unsigned long, if the argument fits into
@@ -330,145 +313,16 @@ extern int __put_user_bad(void)
 		((x) = 0, -EFAULT);					\
 })
 
-/**
- * __copy_to_user() - copy data into user space, with less checking.
- * @to:   Destination address, in user space.
- * @from: Source address, in kernel space.
- * @n:    Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from kernel space to user space.  Caller must check
- * the specified block with access_ok() before calling this function.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- *
- * An alternate version - __copy_to_user_inatomic() - is designed
- * to be called from atomic context, typically bracketed by calls
- * to pagefault_disable() and pagefault_enable().
- */
-extern unsigned long __must_check __copy_to_user_inatomic(
-	void __user *to, const void *from, unsigned long n);
-
-static inline unsigned long __must_check
-__copy_to_user(void __user *to, const void *from, unsigned long n)
-{
-	might_fault();
-	return __copy_to_user_inatomic(to, from, 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))
-		n = __copy_to_user(to, from, n);
-	return n;
-}
-
-/**
- * __copy_from_user() - copy data from user space, with less checking.
- * @to:   Destination address, in kernel space.
- * @from: Source address, in user space.
- * @n:    Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from user space to kernel space.  Caller must check
- * the specified block with access_ok() before calling this function.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- *
- * If some data could not be copied, this function will pad the copied
- * data to the requested size using zero bytes.
- *
- * An alternate version - __copy_from_user_inatomic() - is designed
- * to be called from atomic context, typically bracketed by calls
- * to pagefault_disable() and pagefault_enable().  This version
- * does *NOT* pad with zeros.
- */
-extern unsigned long __must_check __copy_from_user_inatomic(
-	void *to, const void __user *from, unsigned long n);
-extern unsigned long __must_check __copy_from_user_zeroing(
-	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)
-{
-       might_fault();
-       return __copy_from_user_zeroing(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))
-		n = __copy_from_user(to, from, n);
-	else
-		memset(to, 0, n);
-	return n;
-}
-
-extern void __compiletime_error("usercopy buffer size is too small")
-__bad_copy_user(void);
-
-static inline void copy_user_overflow(int size, unsigned long count)
-{
-	WARN(1, "Buffer overflow detected (%d < %lu)!\n", size, count);
-}
-
-static inline unsigned long __must_check copy_from_user(void *to,
-					  const void __user *from,
-					  unsigned long n)
-{
-	int sz = __compiletime_object_size(to);
-
-	if (likely(sz == -1 || sz >= n))
-		n = _copy_from_user(to, from, n);
-	else if (!__builtin_constant_p(n))
-		copy_user_overflow(sz, n);
-	else
-		__bad_copy_user();
-
-	return n;
-}
+extern unsigned long __must_check
+raw_copy_to_user(void __user *to, const void *from, unsigned long n);
+extern unsigned long __must_check
+raw_copy_from_user(void *to, const void __user *from, unsigned long n);
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 #ifdef __tilegx__
-/**
- * __copy_in_user() - copy data within user space, with less checking.
- * @to:   Destination address, in user space.
- * @from: Source address, in user space.
- * @n:    Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from user space to user space.  Caller must check
- * the specified blocks with access_ok() before calling this function.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- */
-extern unsigned long __copy_in_user_inatomic(
+extern unsigned long raw_copy_in_user(
 	void __user *to, const void __user *from, unsigned long n);
-
-static inline unsigned long __must_check
-__copy_in_user(void __user *to, const void __user *from, unsigned long n)
-{
-	might_fault();
-	return __copy_in_user_inatomic(to, from, n);
-}
-
-static inline unsigned long __must_check
-copy_in_user(void __user *to, const void __user *from, unsigned long n)
-{
-	if (access_ok(VERIFY_WRITE, to, n) && access_ok(VERIFY_READ, from, n))
-		n = __copy_in_user(to, from, n);
-	return n;
-}
 #endif
 
 
diff --git a/arch/tile/lib/exports.c b/arch/tile/lib/exports.c
index c5369fe..ecce8e1 100644
--- a/arch/tile/lib/exports.c
+++ b/arch/tile/lib/exports.c
@@ -38,11 +38,10 @@ EXPORT_SYMBOL(__mcount);
 
 /* arch/tile/lib/, various memcpy files */
 EXPORT_SYMBOL(memcpy);
-EXPORT_SYMBOL(__copy_to_user_inatomic);
-EXPORT_SYMBOL(__copy_from_user_inatomic);
-EXPORT_SYMBOL(__copy_from_user_zeroing);
+EXPORT_SYMBOL(raw_copy_to_user);
+EXPORT_SYMBOL(raw_copy_from_user);
 #ifdef __tilegx__
-EXPORT_SYMBOL(__copy_in_user_inatomic);
+EXPORT_SYMBOL(raw_copy_in_user);
 #endif
 
 /* hypervisor glue */
diff --git a/arch/tile/lib/memcpy_32.S b/arch/tile/lib/memcpy_32.S
index a2771ae..270f126 100644
--- a/arch/tile/lib/memcpy_32.S
+++ b/arch/tile/lib/memcpy_32.S
@@ -24,7 +24,6 @@
 
 #define IS_MEMCPY	  0
 #define IS_COPY_FROM_USER  1
-#define IS_COPY_FROM_USER_ZEROING  2
 #define IS_COPY_TO_USER   -1
 
 	.section .text.memcpy_common, "ax"
@@ -42,40 +41,31 @@
 	9
 
 
-/* __copy_from_user_inatomic takes the kernel target address in r0,
+/* raw_copy_from_user takes the kernel target address in r0,
  * the user source in r1, and the bytes to copy in r2.
  * It returns the number of uncopiable bytes (hopefully zero) in r0.
  */
-ENTRY(__copy_from_user_inatomic)
-.type __copy_from_user_inatomic, @function
-	FEEDBACK_ENTER_EXPLICIT(__copy_from_user_inatomic, \
+ENTRY(raw_copy_from_user)
+.type raw_copy_from_user, @function
+	FEEDBACK_ENTER_EXPLICIT(raw_copy_from_user, \
 	  .text.memcpy_common, \
-	  .Lend_memcpy_common - __copy_from_user_inatomic)
+	  .Lend_memcpy_common - raw_copy_from_user)
 	{ movei r29, IS_COPY_FROM_USER; j memcpy_common }
-	.size __copy_from_user_inatomic, . - __copy_from_user_inatomic
+	.size raw_copy_from_user, . - raw_copy_from_user
 
-/* __copy_from_user_zeroing is like __copy_from_user_inatomic, but
- * any uncopiable bytes are zeroed in the target.
- */
-ENTRY(__copy_from_user_zeroing)
-.type __copy_from_user_zeroing, @function
-	FEEDBACK_REENTER(__copy_from_user_inatomic)
-	{ movei r29, IS_COPY_FROM_USER_ZEROING; j memcpy_common }
-	.size __copy_from_user_zeroing, . - __copy_from_user_zeroing
-
-/* __copy_to_user_inatomic takes the user target address in r0,
+/* raw_copy_to_user takes the user target address in r0,
  * the kernel source in r1, and the bytes to copy in r2.
  * It returns the number of uncopiable bytes (hopefully zero) in r0.
  */
-ENTRY(__copy_to_user_inatomic)
-.type __copy_to_user_inatomic, @function
-	FEEDBACK_REENTER(__copy_from_user_inatomic)
+ENTRY(raw_copy_to_user)
+.type raw_copy_to_user, @function
+	FEEDBACK_REENTER(raw_copy_from_user)
 	{ movei r29, IS_COPY_TO_USER; j memcpy_common }
-	.size __copy_to_user_inatomic, . - __copy_to_user_inatomic
+	.size raw_copy_to_user, . - raw_copy_to_user
 
 ENTRY(memcpy)
 .type memcpy, @function
-	FEEDBACK_REENTER(__copy_from_user_inatomic)
+	FEEDBACK_REENTER(raw_copy_from_user)
 	{ movei r29, IS_MEMCPY }
 	.size memcpy, . - memcpy
 	/* Fall through */
@@ -520,12 +510,7 @@
 	{ bnzt r2, copy_from_user_fixup_loop }
 
 .Lcopy_from_user_fixup_zero_remainder:
-	{ bbs r29, 2f }  /* low bit set means IS_COPY_FROM_USER */
-	/* byte-at-a-time loop faulted, so zero the rest. */
-	{ move r3, r2; bz r2, 2f /* should be impossible, but handle it. */ }
-1:      { sb r0, zero; addi r0, r0, 1; addi r3, r3, -1 }
-	{ bnzt r3, 1b }
-2:	move lr, r27
+	move lr, r27
 	{ move r0, r2; jrp lr }
 
 copy_to_user_fixup_loop:
diff --git a/arch/tile/lib/memcpy_user_64.c b/arch/tile/lib/memcpy_user_64.c
index 97bbb60..a3fea9f 100644
--- a/arch/tile/lib/memcpy_user_64.c
+++ b/arch/tile/lib/memcpy_user_64.c
@@ -51,7 +51,7 @@
 		__v;						\
 	})
 
-#define USERCOPY_FUNC __copy_to_user_inatomic
+#define USERCOPY_FUNC raw_copy_to_user
 #define ST1(p, v) _ST((p), st1, (v))
 #define ST2(p, v) _ST((p), st2, (v))
 #define ST4(p, v) _ST((p), st4, (v))
@@ -62,7 +62,7 @@
 #define LD8 LD
 #include "memcpy_64.c"
 
-#define USERCOPY_FUNC __copy_from_user_inatomic
+#define USERCOPY_FUNC raw_copy_from_user
 #define ST1 ST
 #define ST2 ST
 #define ST4 ST
@@ -73,7 +73,7 @@
 #define LD8(p) _LD((p), ld)
 #include "memcpy_64.c"
 
-#define USERCOPY_FUNC __copy_in_user_inatomic
+#define USERCOPY_FUNC raw_copy_in_user
 #define ST1(p, v) _ST((p), st1, (v))
 #define ST2(p, v) _ST((p), st2, (v))
 #define ST4(p, v) _ST((p), st4, (v))
@@ -83,12 +83,3 @@
 #define LD4(p) _LD((p), ld4u)
 #define LD8(p) _LD((p), ld)
 #include "memcpy_64.c"
-
-unsigned long __copy_from_user_zeroing(void *to, const void __user *from,
-				       unsigned long n)
-{
-	unsigned long rc = __copy_from_user_inatomic(to, from, n);
-	if (unlikely(rc))
-		memset(to + n - rc, 0, rc);
-	return rc;
-}
diff --git a/arch/um/Kconfig.common b/arch/um/Kconfig.common
index fd44385..e13f763 100644
--- a/arch/um/Kconfig.common
+++ b/arch/um/Kconfig.common
@@ -13,6 +13,7 @@
 	select GENERIC_CLOCKEVENTS
 	select HAVE_GCC_PLUGINS
 	select TTY # Needed for line.c
+	select ARCH_HAS_RAW_COPY_USER
 
 config MMU
 	bool
diff --git a/arch/um/include/asm/uaccess.h b/arch/um/include/asm/uaccess.h
index 9396c8a..cc00fc5 100644
--- a/arch/um/include/asm/uaccess.h
+++ b/arch/um/include/asm/uaccess.h
@@ -21,8 +21,8 @@
 #define __addr_range_nowrap(addr, size) \
 	((unsigned long) (addr) <= ((unsigned long) (addr) + (size)))
 
-extern long __copy_from_user(void *to, const void __user *from, unsigned long n);
-extern long __copy_to_user(void __user *to, const void *from, unsigned long n);
+extern unsigned long raw_copy_from_user(void *to, const void __user *from, unsigned long n);
+extern unsigned long raw_copy_to_user(void __user *to, const void *from, unsigned long n);
 extern long __strncpy_from_user(char *dst, const char __user *src, long count);
 extern long __strnlen_user(const void __user *str, long len);
 extern unsigned long __clear_user(void __user *mem, unsigned long len);
@@ -33,8 +33,8 @@ static inline int __access_ok(unsigned long addr, unsigned long size);
 #define __clear_user __clear_user
 #define __strnlen_user __strnlen_user
 #define __strncpy_from_user __strncpy_from_user
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 #include <asm-generic/uaccess.h>
 
diff --git a/arch/um/kernel/skas/uaccess.c b/arch/um/kernel/skas/uaccess.c
index 22c9f79..d450797 100644
--- a/arch/um/kernel/skas/uaccess.c
+++ b/arch/um/kernel/skas/uaccess.c
@@ -139,7 +139,7 @@ static int copy_chunk_from_user(unsigned long from, int len, void *arg)
 	return 0;
 }
 
-long __copy_from_user(void *to, const void __user *from, unsigned long n)
+unsigned long raw_copy_from_user(void *to, const void __user *from, unsigned long n)
 {
 	if (uaccess_kernel()) {
 		memcpy(to, (__force void*)from, n);
@@ -148,7 +148,7 @@ long __copy_from_user(void *to, const void __user *from, unsigned long n)
 
 	return buffer_op((unsigned long) from, n, 0, copy_chunk_from_user, &to);
 }
-EXPORT_SYMBOL(__copy_from_user);
+EXPORT_SYMBOL(raw_copy_from_user);
 
 static int copy_chunk_to_user(unsigned long to, int len, void *arg)
 {
@@ -159,7 +159,7 @@ static int copy_chunk_to_user(unsigned long to, int len, void *arg)
 	return 0;
 }
 
-long __copy_to_user(void __user *to, const void *from, unsigned long n)
+unsigned long raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 {
 	if (uaccess_kernel()) {
 		memcpy((__force void *) to, from, n);
@@ -168,7 +168,7 @@ long __copy_to_user(void __user *to, const void *from, unsigned long n)
 
 	return buffer_op((unsigned long) to, n, 1, copy_chunk_to_user, &from);
 }
-EXPORT_SYMBOL(__copy_to_user);
+EXPORT_SYMBOL(raw_copy_to_user);
 
 static int strncpy_chunk_from_user(unsigned long from, int len, void *arg)
 {
diff --git a/arch/unicore32/Kconfig b/arch/unicore32/Kconfig
index 0769066..319519e 100644
--- a/arch/unicore32/Kconfig
+++ b/arch/unicore32/Kconfig
@@ -18,6 +18,7 @@
 	select ARCH_WANT_FRAME_POINTERS
 	select GENERIC_IOMAP
 	select MODULES_USE_ELF_REL
+	select ARCH_HAS_RAW_COPY_USER
 	help
 	  UniCore-32 is 32-bit Instruction Set Architecture,
 	  including a series of low-power-consumption RISC chip
diff --git a/arch/unicore32/include/asm/uaccess.h b/arch/unicore32/include/asm/uaccess.h
index b2bcab6..1d55f2f 100644
--- a/arch/unicore32/include/asm/uaccess.h
+++ b/arch/unicore32/include/asm/uaccess.h
@@ -24,15 +24,17 @@
 #define __access_ok(addr, size)	(__kernel_ok || __user_ok((addr), (size)))
 
 extern unsigned long __must_check
-__copy_from_user(void *to, const void __user *from, unsigned long n);
+raw_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);
+raw_copy_to_user(void __user *to, const void *from, unsigned long n);
 extern unsigned long __must_check
 __clear_user(void __user *addr, unsigned long n);
 extern unsigned long __must_check
 __strncpy_from_user(char *to, const char __user *from, unsigned long count);
 extern unsigned long
 __strnlen_user(const char __user *s, long n);
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 #include <asm-generic/uaccess.h>
 
diff --git a/arch/unicore32/kernel/ksyms.c b/arch/unicore32/kernel/ksyms.c
index 0323528..dcc72ee 100644
--- a/arch/unicore32/kernel/ksyms.c
+++ b/arch/unicore32/kernel/ksyms.c
@@ -46,8 +46,8 @@ EXPORT_SYMBOL(__strncpy_from_user);
 
 EXPORT_SYMBOL(copy_page);
 
-EXPORT_SYMBOL(__copy_from_user);
-EXPORT_SYMBOL(__copy_to_user);
+EXPORT_SYMBOL(raw_copy_from_user);
+EXPORT_SYMBOL(raw_copy_to_user);
 EXPORT_SYMBOL(__clear_user);
 
 EXPORT_SYMBOL(__ashldi3);
diff --git a/arch/unicore32/lib/copy_from_user.S b/arch/unicore32/lib/copy_from_user.S
index ab0767ea..5f80fcb 100644
--- a/arch/unicore32/lib/copy_from_user.S
+++ b/arch/unicore32/lib/copy_from_user.S
@@ -16,7 +16,7 @@
 /*
  * Prototype:
  *
- *	size_t __copy_from_user(void *to, const void *from, size_t n)
+ *	size_t raw_copy_from_user(void *to, const void *from, size_t n)
  *
  * Purpose:
  *
@@ -87,22 +87,18 @@
 
 	.text
 
-ENTRY(__copy_from_user)
+ENTRY(raw_copy_from_user)
 
 #include "copy_template.S"
 
-ENDPROC(__copy_from_user)
+ENDPROC(raw_copy_from_user)
 
 	.pushsection .fixup,"ax"
 	.align 0
 	copy_abort_preamble
-	ldm.w	(r1, r2), [sp]+
-	sub	r3, r0, r1
-	rsub	r2, r3, r2
-	stw	r2, [sp]
-	mov	r1, #0
-	b.l	memset
-	ldw.w	r0, [sp]+, #4
+	ldm.w	(r1, r2, r3), [sp]+
+	sub	r0, r0, r1
+	rsub	r0, r0, r2
 	copy_abort_end
 	.popsection
 
diff --git a/arch/unicore32/lib/copy_to_user.S b/arch/unicore32/lib/copy_to_user.S
index 6e22151..857c681 100644
--- a/arch/unicore32/lib/copy_to_user.S
+++ b/arch/unicore32/lib/copy_to_user.S
@@ -16,7 +16,7 @@
 /*
  * Prototype:
  *
- *	size_t __copy_to_user(void *to, const void *from, size_t n)
+ *	size_t raw_copy_to_user(void *to, const void *from, size_t n)
  *
  * Purpose:
  *
@@ -79,11 +79,11 @@
 
 	.text
 
-WEAK(__copy_to_user)
+WEAK(raw_copy_to_user)
 
 #include "copy_template.S"
 
-ENDPROC(__copy_to_user)
+ENDPROC(raw_copy_to_user)
 
 	.pushsection .fixup,"ax"
 	.align 0
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index cc98d5a..5f59fc3 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -175,6 +175,7 @@
 	select USER_STACKTRACE_SUPPORT
 	select VIRT_TO_BUS
 	select X86_FEATURE_NAMES		if PROC_FS
+	select ARCH_HAS_RAW_COPY_USER
 
 config INSTRUCTION_DECODER
 	def_bool y
diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h
index 0522d88..68766b2 100644
--- a/arch/x86/include/asm/uaccess.h
+++ b/arch/x86/include/asm/uaccess.h
@@ -379,6 +379,18 @@ do {									\
 		     : "=r" (err), ltype(x)				\
 		     : "m" (__m(addr)), "i" (errret), "0" (err))
 
+#define __get_user_asm_nozero(x, addr, err, itype, rtype, ltype, errret)	\
+	asm volatile("\n"						\
+		     "1:	mov"itype" %2,%"rtype"1\n"		\
+		     "2:\n"						\
+		     ".section .fixup,\"ax\"\n"				\
+		     "3:	mov %3,%0\n"				\
+		     "	jmp 2b\n"					\
+		     ".previous\n"					\
+		     _ASM_EXTABLE(1b, 3b)				\
+		     : "=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.
@@ -670,59 +682,6 @@ extern struct movsl_mask {
 # include <asm/uaccess_64.h>
 #endif
 
-unsigned long __must_check _copy_from_user(void *to, const void __user *from,
-					   unsigned n);
-unsigned long __must_check _copy_to_user(void __user *to, const void *from,
-					 unsigned n);
-
-extern void __compiletime_error("usercopy buffer size is too small")
-__bad_copy_user(void);
-
-static inline void copy_user_overflow(int size, unsigned long count)
-{
-	WARN(1, "Buffer overflow detected (%d < %lu)!\n", size, count);
-}
-
-static __always_inline unsigned long __must_check
-copy_from_user(void *to, const void __user *from, unsigned long n)
-{
-	int sz = __compiletime_object_size(to);
-
-	might_fault();
-
-	kasan_check_write(to, 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))
-		copy_user_overflow(sz, n);
-	else
-		__bad_copy_user();
-
-	return n;
-}
-
-static __always_inline unsigned long __must_check
-copy_to_user(void __user *to, const void *from, unsigned long n)
-{
-	int sz = __compiletime_object_size(from);
-
-	kasan_check_read(from, n);
-
-	might_fault();
-
-	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))
-		copy_user_overflow(sz, n);
-	else
-		__bad_copy_user();
-
-	return n;
-}
-
 /*
  * We rely on the nested NMI work to allow atomic faults from the NMI path; the
  * nested NMI paths are careful to preserve CR2.
diff --git a/arch/x86/include/asm/uaccess_32.h b/arch/x86/include/asm/uaccess_32.h
index 5268ecc..aeda9bb 100644
--- a/arch/x86/include/asm/uaccess_32.h
+++ b/arch/x86/include/asm/uaccess_32.h
@@ -8,143 +8,48 @@
 #include <asm/asm.h>
 #include <asm/page.h>
 
-unsigned long __must_check __copy_to_user_ll
-		(void __user *to, const void *from, unsigned long n);
-unsigned long __must_check __copy_from_user_ll
-		(void *to, const void __user *from, unsigned long n);
-unsigned long __must_check __copy_from_user_ll_nozero
-		(void *to, const void __user *from, unsigned long n);
-unsigned long __must_check __copy_from_user_ll_nocache
-		(void *to, const void __user *from, unsigned long n);
+unsigned long __must_check __copy_user_ll
+		(void *to, const void *from, unsigned long n);
 unsigned long __must_check __copy_from_user_ll_nocache_nozero
 		(void *to, const void __user *from, unsigned long n);
 
-/**
- * __copy_to_user_inatomic: - Copy a block of data into user space, with less checking.
- * @to:   Destination address, in user space.
- * @from: Source address, in kernel space.
- * @n:    Number of bytes to copy.
- *
- * Context: User context only.
- *
- * Copy data from kernel space to user space.  Caller must check
- * the specified block with access_ok() before calling this function.
- * The caller should also make sure he pins the user space address
- * so that we don't result in page fault and sleep.
- */
 static __always_inline unsigned long __must_check
-__copy_to_user_inatomic(void __user *to, const void *from, unsigned long n)
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 {
-	check_object_size(from, n, true);
-	return __copy_to_user_ll(to, from, n);
-}
-
-/**
- * __copy_to_user: - Copy a block of data into user space, with less checking.
- * @to:   Destination address, in user space.
- * @from: Source address, in kernel space.
- * @n:    Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from kernel space to user space.  Caller must check
- * the specified block with access_ok() before calling this function.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- */
-static __always_inline unsigned long __must_check
-__copy_to_user(void __user *to, const void *from, unsigned long n)
-{
-	might_fault();
-	return __copy_to_user_inatomic(to, from, n);
+	return __copy_user_ll((__force void *)to, from, n);
 }
 
 static __always_inline unsigned long
-__copy_from_user_inatomic(void *to, const void __user *from, unsigned long n)
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
 {
-	return __copy_from_user_ll_nozero(to, from, n);
-}
-
-/**
- * __copy_from_user: - Copy a block of data from user space, with less checking.
- * @to:   Destination address, in kernel space.
- * @from: Source address, in user space.
- * @n:    Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from user space to kernel space.  Caller must check
- * the specified block with access_ok() before calling this function.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- *
- * If some data could not be copied, this function will pad the copied
- * data to the requested size using zero bytes.
- *
- * An alternate version - __copy_from_user_inatomic() - may be called from
- * atomic context and will fail rather than sleep.  In this case the
- * uncopied bytes will *NOT* be padded with zeros.  See fs/filemap.h
- * for explanation of why this is needed.
- */
-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;
 
 		switch (n) {
 		case 1:
+			ret = 0;
 			__uaccess_begin();
-			__get_user_size(*(u8 *)to, from, 1, ret, 1);
+			__get_user_asm_nozero(*(u8 *)to, from, ret,
+					      "b", "b", "=q", 1);
 			__uaccess_end();
 			return ret;
 		case 2:
+			ret = 0;
 			__uaccess_begin();
-			__get_user_size(*(u16 *)to, from, 2, ret, 2);
+			__get_user_asm_nozero(*(u16 *)to, from, ret,
+					      "w", "w", "=r", 2);
 			__uaccess_end();
 			return ret;
 		case 4:
+			ret = 0;
 			__uaccess_begin();
-			__get_user_size(*(u32 *)to, from, 4, ret, 4);
+			__get_user_asm_nozero(*(u32 *)to, from, ret,
+					      "l", "k", "=r", 4);
 			__uaccess_end();
 			return ret;
 		}
 	}
-	return __copy_from_user_ll(to, from, n);
-}
-
-static __always_inline unsigned long __copy_from_user_nocache(void *to,
-				const void __user *from, unsigned long n)
-{
-	might_fault();
-	if (__builtin_constant_p(n)) {
-		unsigned long ret;
-
-		switch (n) {
-		case 1:
-			__uaccess_begin();
-			__get_user_size(*(u8 *)to, from, 1, ret, 1);
-			__uaccess_end();
-			return ret;
-		case 2:
-			__uaccess_begin();
-			__get_user_size(*(u16 *)to, from, 2, ret, 2);
-			__uaccess_end();
-			return ret;
-		case 4:
-			__uaccess_begin();
-			__get_user_size(*(u32 *)to, from, 4, ret, 4);
-			__uaccess_end();
-			return ret;
-		}
-	}
-	return __copy_from_user_ll_nocache(to, from, n);
+	return __copy_user_ll(to, (__force const void *)from, n);
 }
 
 static __always_inline unsigned long
diff --git a/arch/x86/include/asm/uaccess_64.h b/arch/x86/include/asm/uaccess_64.h
index 8ddadd93..f83b42a 100644
--- a/arch/x86/include/asm/uaccess_64.h
+++ b/arch/x86/include/asm/uaccess_64.h
@@ -45,58 +45,54 @@ copy_user_generic(void *to, const void *from, unsigned len)
 	return ret;
 }
 
-__must_check unsigned long
-copy_in_user(void __user *to, const void __user *from, unsigned len);
-
-static __always_inline __must_check
-int __copy_from_user_nocheck(void *dst, const void __user *src, unsigned size)
+static __always_inline __must_check unsigned long
+raw_copy_from_user(void *dst, const void __user *src, unsigned long 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:
 		__uaccess_begin();
-		__get_user_asm(*(u8 *)dst, (u8 __user *)src,
+		__get_user_asm_nozero(*(u8 *)dst, (u8 __user *)src,
 			      ret, "b", "b", "=q", 1);
 		__uaccess_end();
 		return ret;
 	case 2:
 		__uaccess_begin();
-		__get_user_asm(*(u16 *)dst, (u16 __user *)src,
+		__get_user_asm_nozero(*(u16 *)dst, (u16 __user *)src,
 			      ret, "w", "w", "=r", 2);
 		__uaccess_end();
 		return ret;
 	case 4:
 		__uaccess_begin();
-		__get_user_asm(*(u32 *)dst, (u32 __user *)src,
+		__get_user_asm_nozero(*(u32 *)dst, (u32 __user *)src,
 			      ret, "l", "k", "=r", 4);
 		__uaccess_end();
 		return ret;
 	case 8:
 		__uaccess_begin();
-		__get_user_asm(*(u64 *)dst, (u64 __user *)src,
+		__get_user_asm_nozero(*(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,
+		__get_user_asm_nozero(*(u64 *)dst, (u64 __user *)src,
 			       ret, "q", "", "=r", 10);
 		if (likely(!ret))
-			__get_user_asm(*(u16 *)(8 + (char *)dst),
+			__get_user_asm_nozero(*(u16 *)(8 + (char *)dst),
 				       (u16 __user *)(8 + (char __user *)src),
 				       ret, "w", "w", "=r", 2);
 		__uaccess_end();
 		return ret;
 	case 16:
 		__uaccess_begin();
-		__get_user_asm(*(u64 *)dst, (u64 __user *)src,
+		__get_user_asm_nozero(*(u64 *)dst, (u64 __user *)src,
 			       ret, "q", "", "=r", 16);
 		if (likely(!ret))
-			__get_user_asm(*(u64 *)(8 + (char *)dst),
+			__get_user_asm_nozero(*(u64 *)(8 + (char *)dst),
 				       (u64 __user *)(8 + (char __user *)src),
 				       ret, "q", "", "=r", 8);
 		__uaccess_end();
@@ -106,20 +102,11 @@ int __copy_from_user_nocheck(void *dst, const void __user *src, unsigned size)
 	}
 }
 
-static __always_inline __must_check
-int __copy_from_user(void *dst, const void __user *src, unsigned size)
-{
-	might_fault();
-	kasan_check_write(dst, size);
-	return __copy_from_user_nocheck(dst, src, size);
-}
-
-static __always_inline __must_check
-int __copy_to_user_nocheck(void __user *dst, const void *src, unsigned size)
+static __always_inline __must_check unsigned long
+raw_copy_to_user(void __user *dst, const void *src, unsigned long 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) {
@@ -175,100 +162,16 @@ int __copy_to_user_nocheck(void __user *dst, const void *src, unsigned size)
 }
 
 static __always_inline __must_check
-int __copy_to_user(void __user *dst, const void *src, unsigned size)
+int raw_copy_in_user(void __user *dst, const void __user *src, unsigned size)
 {
-	might_fault();
-	kasan_check_read(src, size);
-	return __copy_to_user_nocheck(dst, src, size);
-}
-
-static __always_inline __must_check
-int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
-{
-	int ret = 0;
-
-	might_fault();
-	if (!__builtin_constant_p(size))
-		return copy_user_generic((__force void *)dst,
-					 (__force void *)src, 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:
-		return copy_user_generic((__force void *)dst,
-					 (__force void *)src, size);
-	}
-}
-
-static __must_check __always_inline int
-__copy_from_user_inatomic(void *dst, const void __user *src, unsigned size)
-{
-	kasan_check_write(dst, size);
-	return __copy_from_user_nocheck(dst, src, size);
-}
-
-static __must_check __always_inline int
-__copy_to_user_inatomic(void __user *dst, const void *src, unsigned size)
-{
-	kasan_check_read(src, size);
-	return __copy_to_user_nocheck(dst, src, size);
+	return copy_user_generic((__force void *)dst,
+				 (__force void *)src, size);
 }
 
 extern long __copy_user_nocache(void *dst, const void __user *src,
 				unsigned size, int zerorest);
 
 static inline int
-__copy_from_user_nocache(void *dst, const void __user *src, unsigned size)
-{
-	might_fault();
-	kasan_check_write(dst, size);
-	return __copy_user_nocache(dst, src, size, 1);
-}
-
-static inline int
 __copy_from_user_inatomic_nocache(void *dst, const void __user *src,
 				  unsigned size)
 {
diff --git a/arch/x86/lib/usercopy.c b/arch/x86/lib/usercopy.c
index c074799..c8c6ad0 100644
--- a/arch/x86/lib/usercopy.c
+++ b/arch/x86/lib/usercopy.c
@@ -4,12 +4,9 @@
  *  For licencing details see kernel-base/COPYING
  */
 
-#include <linux/highmem.h>
+#include <linux/uaccess.h>
 #include <linux/export.h>
 
-#include <asm/word-at-a-time.h>
-#include <linux/sched.h>
-
 /*
  * We rely on the nested NMI work to allow atomic faults from the NMI path; the
  * nested NMI paths are careful to preserve CR2.
@@ -34,52 +31,3 @@ copy_from_user_nmi(void *to, const void __user *from, unsigned long n)
 	return ret;
 }
 EXPORT_SYMBOL_GPL(copy_from_user_nmi);
-
-/**
- * copy_to_user: - Copy a block of data into user space.
- * @to:   Destination address, in user space.
- * @from: Source address, in kernel space.
- * @n:    Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from kernel space to user space.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- */
-unsigned long _copy_to_user(void __user *to, const void *from, unsigned n)
-{
-	if (access_ok(VERIFY_WRITE, to, n))
-		n = __copy_to_user(to, from, n);
-	return n;
-}
-EXPORT_SYMBOL(_copy_to_user);
-
-/**
- * copy_from_user: - Copy a block of data from user space.
- * @to:   Destination address, in kernel space.
- * @from: Source address, in user space.
- * @n:    Number of bytes to copy.
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Copy data from user space to kernel space.
- *
- * Returns number of bytes that could not be copied.
- * On success, this will be zero.
- *
- * If some data could not be copied, this function will pad the copied
- * data to the requested size using zero bytes.
- */
-unsigned long _copy_from_user(void *to, const void __user *from, unsigned n)
-{
-	if (access_ok(VERIFY_READ, from, n))
-		n = __copy_from_user(to, from, n);
-	else
-		memset(to, 0, n);
-	return n;
-}
-EXPORT_SYMBOL(_copy_from_user);
diff --git a/arch/x86/lib/usercopy_32.c b/arch/x86/lib/usercopy_32.c
index 1f65ff65..bd057a4 100644
--- a/arch/x86/lib/usercopy_32.c
+++ b/arch/x86/lib/usercopy_32.c
@@ -5,12 +5,7 @@
  * Copyright 1997 Andi Kleen <ak@muc.de>
  * Copyright 1997 Linus Torvalds
  */
-#include <linux/mm.h>
-#include <linux/highmem.h>
-#include <linux/blkdev.h>
 #include <linux/export.h>
-#include <linux/backing-dev.h>
-#include <linux/interrupt.h>
 #include <linux/uaccess.h>
 #include <asm/mmx.h>
 #include <asm/asm.h>
@@ -201,197 +196,6 @@ __copy_user_intel(void __user *to, const void *from, unsigned long size)
 	return size;
 }
 
-static unsigned long
-__copy_user_zeroing_intel(void *to, const void __user *from, unsigned long size)
-{
-	int d0, d1;
-	__asm__ __volatile__(
-		       "        .align 2,0x90\n"
-		       "0:      movl 32(%4), %%eax\n"
-		       "        cmpl $67, %0\n"
-		       "        jbe 2f\n"
-		       "1:      movl 64(%4), %%eax\n"
-		       "        .align 2,0x90\n"
-		       "2:      movl 0(%4), %%eax\n"
-		       "21:     movl 4(%4), %%edx\n"
-		       "        movl %%eax, 0(%3)\n"
-		       "        movl %%edx, 4(%3)\n"
-		       "3:      movl 8(%4), %%eax\n"
-		       "31:     movl 12(%4),%%edx\n"
-		       "        movl %%eax, 8(%3)\n"
-		       "        movl %%edx, 12(%3)\n"
-		       "4:      movl 16(%4), %%eax\n"
-		       "41:     movl 20(%4), %%edx\n"
-		       "        movl %%eax, 16(%3)\n"
-		       "        movl %%edx, 20(%3)\n"
-		       "10:     movl 24(%4), %%eax\n"
-		       "51:     movl 28(%4), %%edx\n"
-		       "        movl %%eax, 24(%3)\n"
-		       "        movl %%edx, 28(%3)\n"
-		       "11:     movl 32(%4), %%eax\n"
-		       "61:     movl 36(%4), %%edx\n"
-		       "        movl %%eax, 32(%3)\n"
-		       "        movl %%edx, 36(%3)\n"
-		       "12:     movl 40(%4), %%eax\n"
-		       "71:     movl 44(%4), %%edx\n"
-		       "        movl %%eax, 40(%3)\n"
-		       "        movl %%edx, 44(%3)\n"
-		       "13:     movl 48(%4), %%eax\n"
-		       "81:     movl 52(%4), %%edx\n"
-		       "        movl %%eax, 48(%3)\n"
-		       "        movl %%edx, 52(%3)\n"
-		       "14:     movl 56(%4), %%eax\n"
-		       "91:     movl 60(%4), %%edx\n"
-		       "        movl %%eax, 56(%3)\n"
-		       "        movl %%edx, 60(%3)\n"
-		       "        addl $-64, %0\n"
-		       "        addl $64, %4\n"
-		       "        addl $64, %3\n"
-		       "        cmpl $63, %0\n"
-		       "        ja  0b\n"
-		       "5:      movl  %0, %%eax\n"
-		       "        shrl  $2, %0\n"
-		       "        andl $3, %%eax\n"
-		       "        cld\n"
-		       "6:      rep; movsl\n"
-		       "        movl %%eax,%0\n"
-		       "7:      rep; movsb\n"
-		       "8:\n"
-		       ".section .fixup,\"ax\"\n"
-		       "9:      lea 0(%%eax,%0,4),%0\n"
-		       "16:     pushl %0\n"
-		       "        pushl %%eax\n"
-		       "        xorl %%eax,%%eax\n"
-		       "        rep; stosb\n"
-		       "        popl %%eax\n"
-		       "        popl %0\n"
-		       "        jmp 8b\n"
-		       ".previous\n"
-		       _ASM_EXTABLE(0b,16b)
-		       _ASM_EXTABLE(1b,16b)
-		       _ASM_EXTABLE(2b,16b)
-		       _ASM_EXTABLE(21b,16b)
-		       _ASM_EXTABLE(3b,16b)
-		       _ASM_EXTABLE(31b,16b)
-		       _ASM_EXTABLE(4b,16b)
-		       _ASM_EXTABLE(41b,16b)
-		       _ASM_EXTABLE(10b,16b)
-		       _ASM_EXTABLE(51b,16b)
-		       _ASM_EXTABLE(11b,16b)
-		       _ASM_EXTABLE(61b,16b)
-		       _ASM_EXTABLE(12b,16b)
-		       _ASM_EXTABLE(71b,16b)
-		       _ASM_EXTABLE(13b,16b)
-		       _ASM_EXTABLE(81b,16b)
-		       _ASM_EXTABLE(14b,16b)
-		       _ASM_EXTABLE(91b,16b)
-		       _ASM_EXTABLE(6b,9b)
-		       _ASM_EXTABLE(7b,16b)
-		       : "=&c"(size), "=&D" (d0), "=&S" (d1)
-		       :  "1"(to), "2"(from), "0"(size)
-		       : "eax", "edx", "memory");
-	return size;
-}
-
-/*
- * Non Temporal Hint version of __copy_user_zeroing_intel.  It is cache aware.
- * hyoshiok@miraclelinux.com
- */
-
-static unsigned long __copy_user_zeroing_intel_nocache(void *to,
-				const void __user *from, unsigned long size)
-{
-	int d0, d1;
-
-	__asm__ __volatile__(
-	       "        .align 2,0x90\n"
-	       "0:      movl 32(%4), %%eax\n"
-	       "        cmpl $67, %0\n"
-	       "        jbe 2f\n"
-	       "1:      movl 64(%4), %%eax\n"
-	       "        .align 2,0x90\n"
-	       "2:      movl 0(%4), %%eax\n"
-	       "21:     movl 4(%4), %%edx\n"
-	       "        movnti %%eax, 0(%3)\n"
-	       "        movnti %%edx, 4(%3)\n"
-	       "3:      movl 8(%4), %%eax\n"
-	       "31:     movl 12(%4),%%edx\n"
-	       "        movnti %%eax, 8(%3)\n"
-	       "        movnti %%edx, 12(%3)\n"
-	       "4:      movl 16(%4), %%eax\n"
-	       "41:     movl 20(%4), %%edx\n"
-	       "        movnti %%eax, 16(%3)\n"
-	       "        movnti %%edx, 20(%3)\n"
-	       "10:     movl 24(%4), %%eax\n"
-	       "51:     movl 28(%4), %%edx\n"
-	       "        movnti %%eax, 24(%3)\n"
-	       "        movnti %%edx, 28(%3)\n"
-	       "11:     movl 32(%4), %%eax\n"
-	       "61:     movl 36(%4), %%edx\n"
-	       "        movnti %%eax, 32(%3)\n"
-	       "        movnti %%edx, 36(%3)\n"
-	       "12:     movl 40(%4), %%eax\n"
-	       "71:     movl 44(%4), %%edx\n"
-	       "        movnti %%eax, 40(%3)\n"
-	       "        movnti %%edx, 44(%3)\n"
-	       "13:     movl 48(%4), %%eax\n"
-	       "81:     movl 52(%4), %%edx\n"
-	       "        movnti %%eax, 48(%3)\n"
-	       "        movnti %%edx, 52(%3)\n"
-	       "14:     movl 56(%4), %%eax\n"
-	       "91:     movl 60(%4), %%edx\n"
-	       "        movnti %%eax, 56(%3)\n"
-	       "        movnti %%edx, 60(%3)\n"
-	       "        addl $-64, %0\n"
-	       "        addl $64, %4\n"
-	       "        addl $64, %3\n"
-	       "        cmpl $63, %0\n"
-	       "        ja  0b\n"
-	       "        sfence \n"
-	       "5:      movl  %0, %%eax\n"
-	       "        shrl  $2, %0\n"
-	       "        andl $3, %%eax\n"
-	       "        cld\n"
-	       "6:      rep; movsl\n"
-	       "        movl %%eax,%0\n"
-	       "7:      rep; movsb\n"
-	       "8:\n"
-	       ".section .fixup,\"ax\"\n"
-	       "9:      lea 0(%%eax,%0,4),%0\n"
-	       "16:     pushl %0\n"
-	       "        pushl %%eax\n"
-	       "        xorl %%eax,%%eax\n"
-	       "        rep; stosb\n"
-	       "        popl %%eax\n"
-	       "        popl %0\n"
-	       "        jmp 8b\n"
-	       ".previous\n"
-	       _ASM_EXTABLE(0b,16b)
-	       _ASM_EXTABLE(1b,16b)
-	       _ASM_EXTABLE(2b,16b)
-	       _ASM_EXTABLE(21b,16b)
-	       _ASM_EXTABLE(3b,16b)
-	       _ASM_EXTABLE(31b,16b)
-	       _ASM_EXTABLE(4b,16b)
-	       _ASM_EXTABLE(41b,16b)
-	       _ASM_EXTABLE(10b,16b)
-	       _ASM_EXTABLE(51b,16b)
-	       _ASM_EXTABLE(11b,16b)
-	       _ASM_EXTABLE(61b,16b)
-	       _ASM_EXTABLE(12b,16b)
-	       _ASM_EXTABLE(71b,16b)
-	       _ASM_EXTABLE(13b,16b)
-	       _ASM_EXTABLE(81b,16b)
-	       _ASM_EXTABLE(14b,16b)
-	       _ASM_EXTABLE(91b,16b)
-	       _ASM_EXTABLE(6b,9b)
-	       _ASM_EXTABLE(7b,16b)
-	       : "=&c"(size), "=&D" (d0), "=&S" (d1)
-	       :  "1"(to), "2"(from), "0"(size)
-	       : "eax", "edx", "memory");
-	return size;
-}
-
 static unsigned long __copy_user_intel_nocache(void *to,
 				const void __user *from, unsigned long size)
 {
@@ -486,12 +290,8 @@ static unsigned long __copy_user_intel_nocache(void *to,
  * Leave these declared but undefined.  They should not be any references to
  * them
  */
-unsigned long __copy_user_zeroing_intel(void *to, const void __user *from,
-					unsigned long size);
 unsigned long __copy_user_intel(void __user *to, const void *from,
 					unsigned long size);
-unsigned long __copy_user_zeroing_intel_nocache(void *to,
-				const void __user *from, unsigned long size);
 #endif /* CONFIG_X86_INTEL_USERCOPY */
 
 /* Generic arbitrary sized copy.  */
@@ -528,47 +328,7 @@ do {									\
 		: "memory");						\
 } while (0)
 
-#define __copy_user_zeroing(to, from, size)				\
-do {									\
-	int __d0, __d1, __d2;						\
-	__asm__ __volatile__(						\
-		"	cmp  $7,%0\n"					\
-		"	jbe  1f\n"					\
-		"	movl %1,%0\n"					\
-		"	negl %0\n"					\
-		"	andl $7,%0\n"					\
-		"	subl %0,%3\n"					\
-		"4:	rep; movsb\n"					\
-		"	movl %3,%0\n"					\
-		"	shrl $2,%0\n"					\
-		"	andl $3,%3\n"					\
-		"	.align 2,0x90\n"				\
-		"0:	rep; movsl\n"					\
-		"	movl %3,%0\n"					\
-		"1:	rep; movsb\n"					\
-		"2:\n"							\
-		".section .fixup,\"ax\"\n"				\
-		"5:	addl %3,%0\n"					\
-		"	jmp 6f\n"					\
-		"3:	lea 0(%3,%0,4),%0\n"				\
-		"6:	pushl %0\n"					\
-		"	pushl %%eax\n"					\
-		"	xorl %%eax,%%eax\n"				\
-		"	rep; stosb\n"					\
-		"	popl %%eax\n"					\
-		"	popl %0\n"					\
-		"	jmp 2b\n"					\
-		".previous\n"						\
-		_ASM_EXTABLE(4b,5b)					\
-		_ASM_EXTABLE(0b,3b)					\
-		_ASM_EXTABLE(1b,6b)					\
-		: "=&c"(size), "=&D" (__d0), "=&S" (__d1), "=r"(__d2)	\
-		: "3"(size), "0"(size), "1"(to), "2"(from)		\
-		: "memory");						\
-} while (0)
-
-unsigned long __copy_to_user_ll(void __user *to, const void *from,
-				unsigned long n)
+unsigned long __copy_user_ll(void *to, const void *from, unsigned long n)
 {
 	stac();
 	if (movsl_is_ok(to, from, n))
@@ -578,51 +338,7 @@ unsigned long __copy_to_user_ll(void __user *to, const void *from,
 	clac();
 	return n;
 }
-EXPORT_SYMBOL(__copy_to_user_ll);
-
-unsigned long __copy_from_user_ll(void *to, const void __user *from,
-					unsigned long n)
-{
-	stac();
-	if (movsl_is_ok(to, from, n))
-		__copy_user_zeroing(to, from, n);
-	else
-		n = __copy_user_zeroing_intel(to, from, n);
-	clac();
-	return n;
-}
-EXPORT_SYMBOL(__copy_from_user_ll);
-
-unsigned long __copy_from_user_ll_nozero(void *to, const void __user *from,
-					 unsigned long n)
-{
-	stac();
-	if (movsl_is_ok(to, from, n))
-		__copy_user(to, from, n);
-	else
-		n = __copy_user_intel((void __user *)to,
-				      (const void *)from, n);
-	clac();
-	return n;
-}
-EXPORT_SYMBOL(__copy_from_user_ll_nozero);
-
-unsigned long __copy_from_user_ll_nocache(void *to, const void __user *from,
-					unsigned long n)
-{
-	stac();
-#ifdef CONFIG_X86_INTEL_USERCOPY
-	if (n > 64 && static_cpu_has(X86_FEATURE_XMM2))
-		n = __copy_user_zeroing_intel_nocache(to, from, n);
-	else
-		__copy_user_zeroing(to, from, n);
-#else
-	__copy_user_zeroing(to, from, n);
-#endif
-	clac();
-	return n;
-}
-EXPORT_SYMBOL(__copy_from_user_ll_nocache);
+EXPORT_SYMBOL(__copy_user_ll);
 
 unsigned long __copy_from_user_ll_nocache_nozero(void *to, const void __user *from,
 					unsigned long n)
diff --git a/arch/x86/lib/usercopy_64.c b/arch/x86/lib/usercopy_64.c
index 6987358..3b7c40a 100644
--- a/arch/x86/lib/usercopy_64.c
+++ b/arch/x86/lib/usercopy_64.c
@@ -54,15 +54,6 @@ unsigned long clear_user(void __user *to, unsigned long n)
 }
 EXPORT_SYMBOL(clear_user);
 
-unsigned long copy_in_user(void __user *to, const void __user *from, unsigned len)
-{
-	if (access_ok(VERIFY_WRITE, to, len) && access_ok(VERIFY_READ, from, len)) { 
-		return copy_user_generic((__force void *)to, (__force void *)from, len);
-	} 
-	return len;		
-}
-EXPORT_SYMBOL(copy_in_user);
-
 /*
  * Try to copy last bytes and clear the rest if needed.
  * Since protection fault in copy_from/to_user is not a normal situation,
@@ -80,9 +71,5 @@ copy_user_handle_tail(char *to, char *from, unsigned len)
 			break;
 	}
 	clac();
-
-	/* If the destination is a kernel buffer, we always clear the end */
-	if (!__addr_ok(to))
-		memset(to, 0, len);
 	return len;
 }
diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig
index f4126cf..043d37d 100644
--- a/arch/xtensa/Kconfig
+++ b/arch/xtensa/Kconfig
@@ -29,6 +29,7 @@
 	select NO_BOOTMEM
 	select PERF_USE_VMALLOC
 	select VIRT_TO_BUS
+	select ARCH_HAS_RAW_COPY_USER
 	help
 	  Xtensa processors are 32-bit RISC machines designed by Tensilica
 	  primarily for embedded systems.  These processors are both
diff --git a/arch/xtensa/include/asm/Kbuild b/arch/xtensa/include/asm/Kbuild
index f41408c..cc23e9e 100644
--- a/arch/xtensa/include/asm/Kbuild
+++ b/arch/xtensa/include/asm/Kbuild
@@ -6,6 +6,7 @@
 generic-y += emergency-restart.h
 generic-y += errno.h
 generic-y += exec.h
+generic-y += extable.h
 generic-y += fcntl.h
 generic-y += hardirq.h
 generic-y += ioctl.h
diff --git a/arch/xtensa/include/asm/uaccess.h b/arch/xtensa/include/asm/uaccess.h
index 2651269..8e93ed8 100644
--- a/arch/xtensa/include/asm/uaccess.h
+++ b/arch/xtensa/include/asm/uaccess.h
@@ -18,6 +18,7 @@
 
 #include <linux/prefetch.h>
 #include <asm/types.h>
+#include <asm/extable.h>
 
 /*
  * The fs value determines whether argument validity checking should
@@ -233,60 +234,22 @@ __asm__ __volatile__(			\
  * Copy to/from user space
  */
 
-/*
- * We use a generic, arbitrary-sized copy subroutine.  The Xtensa
- * architecture would cause heavy code bloat if we tried to inline
- * these functions and provide __constant_copy_* equivalents like the
- * i386 versions.  __xtensa_copy_user is quite efficient.  See the
- * .fixup section of __xtensa_copy_user for a discussion on the
- * X_zeroing equivalents for Xtensa.
- */
-
 extern unsigned __xtensa_copy_user(void *to, const void *from, unsigned n);
-#define __copy_user(to, from, size) __xtensa_copy_user(to, from, size)
-
 
 static inline unsigned long
-__generic_copy_from_user_nocheck(void *to, const void *from, unsigned long n)
-{
-	return __copy_user(to, from, n);
-}
-
-static inline unsigned long
-__generic_copy_to_user_nocheck(void *to, const void *from, unsigned long n)
-{
-	return __copy_user(to, from, n);
-}
-
-static inline unsigned long
-__generic_copy_to_user(void *to, const void *from, unsigned long n)
-{
-	prefetch(from);
-	if (access_ok(VERIFY_WRITE, to, n))
-		return __copy_user(to, from, n);
-	return n;
-}
-
-static inline unsigned long
-__generic_copy_from_user(void *to, const void *from, unsigned long n)
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
 {
 	prefetchw(to);
-	if (access_ok(VERIFY_READ, from, n))
-		return __copy_user(to, from, n);
-	else
-		memset(to, 0, n);
-	return n;
+	return __xtensa_copy_user(to, (__force const void *)from, n);
 }
-
-#define copy_to_user(to, from, n) __generic_copy_to_user((to), (from), (n))
-#define copy_from_user(to, from, n) __generic_copy_from_user((to), (from), (n))
-#define __copy_to_user(to, from, n) \
-	__generic_copy_to_user_nocheck((to), (from), (n))
-#define __copy_from_user(to, from, n) \
-	__generic_copy_from_user_nocheck((to), (from), (n))
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
-
+static inline unsigned long
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
+{
+	prefetchw(from);
+	return __xtensa_copy_user((__force void *)to, from, n);
+}
+#define INLINE_COPY_FROM_USER
+#define INLINE_COPY_TO_USER
 
 /*
  * We need to return the number of bytes not cleared.  Our memset()
@@ -342,10 +305,4 @@ static inline long strnlen_user(const char *str, long len)
 	return __strnlen_user(str, len);
 }
 
-
-struct exception_table_entry
-{
-	unsigned long insn, fixup;
-};
-
 #endif	/* _XTENSA_UACCESS_H */
diff --git a/arch/xtensa/lib/usercopy.S b/arch/xtensa/lib/usercopy.S
index 7ea4dd6..d9cd766 100644
--- a/arch/xtensa/lib/usercopy.S
+++ b/arch/xtensa/lib/usercopy.S
@@ -102,9 +102,9 @@
 	bltui	a4, 7, .Lbytecopy	# do short copies byte by byte
 
 	# copy 1 byte
-	EX(l8ui, a6, a3, 0, l_fixup)
+	EX(l8ui, a6, a3, 0, fixup)
 	addi	a3, a3,  1
-	EX(s8i, a6, a5,  0, s_fixup)
+	EX(s8i, a6, a5,  0, fixup)
 	addi	a5, a5,  1
 	addi	a4, a4, -1
 	bbci.l	a5, 1, .Ldstaligned	# if dst is now aligned, then
@@ -112,11 +112,11 @@
 .Ldst2mod4:	# dst 16-bit aligned
 	# copy 2 bytes
 	bltui	a4, 6, .Lbytecopy	# do short copies byte by byte
-	EX(l8ui, a6, a3, 0, l_fixup)
-	EX(l8ui, a7, a3, 1, l_fixup)
+	EX(l8ui, a6, a3, 0, fixup)
+	EX(l8ui, a7, a3, 1, fixup)
 	addi	a3, a3,  2
-	EX(s8i, a6, a5,  0, s_fixup)
-	EX(s8i, a7, a5,  1, s_fixup)
+	EX(s8i, a6, a5,  0, fixup)
+	EX(s8i, a7, a5,  1, fixup)
 	addi	a5, a5,  2
 	addi	a4, a4, -2
 	j	.Ldstaligned	# dst is now aligned, return to main algorithm
@@ -135,9 +135,9 @@
 	add	a7, a3, a4	# a7 = end address for source
 #endif /* !XCHAL_HAVE_LOOPS */
 .Lnextbyte:
-	EX(l8ui, a6, a3, 0, l_fixup)
+	EX(l8ui, a6, a3, 0, fixup)
 	addi	a3, a3, 1
-	EX(s8i, a6, a5, 0, s_fixup)
+	EX(s8i, a6, a5, 0, fixup)
 	addi	a5, a5, 1
 #if !XCHAL_HAVE_LOOPS
 	blt	a3, a7, .Lnextbyte
@@ -161,15 +161,15 @@
 	add	a8, a8, a3	# a8 = end of last 16B source chunk
 #endif /* !XCHAL_HAVE_LOOPS */
 .Loop1:
-	EX(l32i, a6, a3,  0, l_fixup)
-	EX(l32i, a7, a3,  4, l_fixup)
-	EX(s32i, a6, a5,  0, s_fixup)
-	EX(l32i, a6, a3,  8, l_fixup)
-	EX(s32i, a7, a5,  4, s_fixup)
-	EX(l32i, a7, a3, 12, l_fixup)
-	EX(s32i, a6, a5,  8, s_fixup)
+	EX(l32i, a6, a3,  0, fixup)
+	EX(l32i, a7, a3,  4, fixup)
+	EX(s32i, a6, a5,  0, fixup)
+	EX(l32i, a6, a3,  8, fixup)
+	EX(s32i, a7, a5,  4, fixup)
+	EX(l32i, a7, a3, 12, fixup)
+	EX(s32i, a6, a5,  8, fixup)
 	addi	a3, a3, 16
-	EX(s32i, a7, a5, 12, s_fixup)
+	EX(s32i, a7, a5, 12, fixup)
 	addi	a5, a5, 16
 #if !XCHAL_HAVE_LOOPS
 	blt	a3, a8, .Loop1
@@ -177,31 +177,31 @@
 .Loop1done:
 	bbci.l	a4, 3, .L2
 	# copy 8 bytes
-	EX(l32i, a6, a3,  0, l_fixup)
-	EX(l32i, a7, a3,  4, l_fixup)
+	EX(l32i, a6, a3,  0, fixup)
+	EX(l32i, a7, a3,  4, fixup)
 	addi	a3, a3,  8
-	EX(s32i, a6, a5,  0, s_fixup)
-	EX(s32i, a7, a5,  4, s_fixup)
+	EX(s32i, a6, a5,  0, fixup)
+	EX(s32i, a7, a5,  4, fixup)
 	addi	a5, a5,  8
 .L2:
 	bbci.l	a4, 2, .L3
 	# copy 4 bytes
-	EX(l32i, a6, a3,  0, l_fixup)
+	EX(l32i, a6, a3,  0, fixup)
 	addi	a3, a3,  4
-	EX(s32i, a6, a5,  0, s_fixup)
+	EX(s32i, a6, a5,  0, fixup)
 	addi	a5, a5,  4
 .L3:
 	bbci.l	a4, 1, .L4
 	# copy 2 bytes
-	EX(l16ui, a6, a3,  0, l_fixup)
+	EX(l16ui, a6, a3,  0, fixup)
 	addi	a3, a3,  2
-	EX(s16i,  a6, a5,  0, s_fixup)
+	EX(s16i,  a6, a5,  0, fixup)
 	addi	a5, a5,  2
 .L4:
 	bbci.l	a4, 0, .L5
 	# copy 1 byte
-	EX(l8ui, a6, a3,  0, l_fixup)
-	EX(s8i,  a6, a5,  0, s_fixup)
+	EX(l8ui, a6, a3,  0, fixup)
+	EX(s8i,  a6, a5,  0, fixup)
 .L5:
 	movi	a2, 0		# return success for len bytes copied
 	retw
@@ -217,7 +217,7 @@
 	# copy 16 bytes per iteration for word-aligned dst and unaligned src
 	and	a10, a3, a8	# save unalignment offset for below
 	sub	a3, a3, a10	# align a3 (to avoid sim warnings only; not needed for hardware)
-	EX(l32i, a6, a3, 0, l_fixup)	# load first word
+	EX(l32i, a6, a3, 0, fixup)	# load first word
 #if XCHAL_HAVE_LOOPS
 	loopnez	a7, .Loop2done
 #else /* !XCHAL_HAVE_LOOPS */
@@ -226,19 +226,19 @@
 	add	a12, a12, a3	# a12 = end of last 16B source chunk
 #endif /* !XCHAL_HAVE_LOOPS */
 .Loop2:
-	EX(l32i, a7, a3,  4, l_fixup)
-	EX(l32i, a8, a3,  8, l_fixup)
+	EX(l32i, a7, a3,  4, fixup)
+	EX(l32i, a8, a3,  8, fixup)
 	ALIGN(	a6, a6, a7)
-	EX(s32i, a6, a5,  0, s_fixup)
-	EX(l32i, a9, a3, 12, l_fixup)
+	EX(s32i, a6, a5,  0, fixup)
+	EX(l32i, a9, a3, 12, fixup)
 	ALIGN(	a7, a7, a8)
-	EX(s32i, a7, a5,  4, s_fixup)
-	EX(l32i, a6, a3, 16, l_fixup)
+	EX(s32i, a7, a5,  4, fixup)
+	EX(l32i, a6, a3, 16, fixup)
 	ALIGN(	a8, a8, a9)
-	EX(s32i, a8, a5,  8, s_fixup)
+	EX(s32i, a8, a5,  8, fixup)
 	addi	a3, a3, 16
 	ALIGN(	a9, a9, a6)
-	EX(s32i, a9, a5, 12, s_fixup)
+	EX(s32i, a9, a5, 12, fixup)
 	addi	a5, a5, 16
 #if !XCHAL_HAVE_LOOPS
 	blt	a3, a12, .Loop2
@@ -246,39 +246,39 @@
 .Loop2done:
 	bbci.l	a4, 3, .L12
 	# copy 8 bytes
-	EX(l32i, a7, a3,  4, l_fixup)
-	EX(l32i, a8, a3,  8, l_fixup)
+	EX(l32i, a7, a3,  4, fixup)
+	EX(l32i, a8, a3,  8, fixup)
 	ALIGN(	a6, a6, a7)
-	EX(s32i, a6, a5,  0, s_fixup)
+	EX(s32i, a6, a5,  0, fixup)
 	addi	a3, a3,  8
 	ALIGN(	a7, a7, a8)
-	EX(s32i, a7, a5,  4, s_fixup)
+	EX(s32i, a7, a5,  4, fixup)
 	addi	a5, a5,  8
 	mov	a6, a8
 .L12:
 	bbci.l	a4, 2, .L13
 	# copy 4 bytes
-	EX(l32i, a7, a3,  4, l_fixup)
+	EX(l32i, a7, a3,  4, fixup)
 	addi	a3, a3,  4
 	ALIGN(	a6, a6, a7)
-	EX(s32i, a6, a5,  0, s_fixup)
+	EX(s32i, a6, a5,  0, fixup)
 	addi	a5, a5,  4
 	mov	a6, a7
 .L13:
 	add	a3, a3, a10	# readjust a3 with correct misalignment
 	bbci.l	a4, 1, .L14
 	# copy 2 bytes
-	EX(l8ui, a6, a3,  0, l_fixup)
-	EX(l8ui, a7, a3,  1, l_fixup)
+	EX(l8ui, a6, a3,  0, fixup)
+	EX(l8ui, a7, a3,  1, fixup)
 	addi	a3, a3,  2
-	EX(s8i, a6, a5,  0, s_fixup)
-	EX(s8i, a7, a5,  1, s_fixup)
+	EX(s8i, a6, a5,  0, fixup)
+	EX(s8i, a7, a5,  1, fixup)
 	addi	a5, a5,  2
 .L14:
 	bbci.l	a4, 0, .L15
 	# copy 1 byte
-	EX(l8ui, a6, a3,  0, l_fixup)
-	EX(s8i,  a6, a5,  0, s_fixup)
+	EX(l8ui, a6, a3,  0, fixup)
+	EX(s8i,  a6, a5,  0, fixup)
 .L15:
 	movi	a2, 0		# return success for len bytes copied
 	retw
@@ -291,30 +291,10 @@
  * bytes_copied = a5 - a2
  * retval = bytes_not_copied = original len - bytes_copied
  * retval = a11 - (a5 - a2)
- *
- * Clearing the remaining pieces of kernel memory plugs security
- * holes.  This functionality is the equivalent of the *_zeroing
- * functions that some architectures provide.
  */
 
-.Lmemset:
-	.word	memset
 
-s_fixup:
+fixup:
 	sub	a2, a5, a2	/* a2 <-- bytes copied */
 	sub	a2, a11, a2	/* a2 <-- bytes not copied */
 	retw
-
-l_fixup:
-	sub	a2, a5, a2	/* a2 <-- bytes copied */
-	sub	a2, a11, a2	/* a2 <-- bytes not copied == return value */
-
-	/* void *memset(void *s, int c, size_t n); */
-	mov	a6, a5		/* s */
-	movi	a7, 0		/* c */
-	mov	a8, a2		/* n */
-	l32r	a4, .Lmemset
-	callx4	a4
-	/* Ignore memset return value in a6. */
-	/* a2 still contains bytes not copied. */
-	retw
diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c
index 4348027..58fb1bc 100644
--- a/fs/ocfs2/cluster/tcp.c
+++ b/fs/ocfs2/cluster/tcp.c
@@ -1460,27 +1460,10 @@ static void o2net_rx_until_empty(struct work_struct *work)
 
 static int o2net_set_nodelay(struct socket *sock)
 {
-	int ret, val = 1;
-	mm_segment_t oldfs;
+	int val = 1;
 
-	oldfs = get_fs();
-	set_fs(KERNEL_DS);
-
-	/*
-	 * Dear unsuspecting programmer,
-	 *
-	 * Don't use sock_setsockopt() for SOL_TCP.  It doesn't check its level
-	 * argument and assumes SOL_SOCKET so, say, your TCP_NODELAY will
-	 * silently turn into SO_DEBUG.
-	 *
-	 * Yours,
-	 * Keeper of hilariously fragile interfaces.
-	 */
-	ret = sock->ops->setsockopt(sock, SOL_TCP, TCP_NODELAY,
-				    (char __user *)&val, sizeof(val));
-
-	set_fs(oldfs);
-	return ret;
+	return kernel_setsockopt(sock, SOL_TCP, TCP_NODELAY,
+				    (void *)&val, sizeof(val));
 }
 
 static int o2net_set_usertimeout(struct socket *sock)
@@ -1488,7 +1471,7 @@ static int o2net_set_usertimeout(struct socket *sock)
 	int user_timeout = O2NET_TCP_USER_TIMEOUT;
 
 	return kernel_setsockopt(sock, SOL_TCP, TCP_USER_TIMEOUT,
-				(char *)&user_timeout, sizeof(user_timeout));
+				(void *)&user_timeout, sizeof(user_timeout));
 }
 
 static void o2net_initialize_handshake(void)
diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h
index 5f76bc9..7fc2104 100644
--- a/include/linux/uaccess.h
+++ b/include/linux/uaccess.h
@@ -261,12 +261,6 @@ static inline unsigned long __copy_from_user_inatomic_nocache(void *to,
 	return __copy_from_user_inatomic(to, from, n);
 }
 
-static inline unsigned long __copy_from_user_nocache(void *to,
-				const void __user *from, unsigned long n)
-{
-	return __copy_from_user(to, from, n);
-}
-
 #endif		/* ARCH_HAS_NOCACHE_UACCESS */
 
 /*
diff --git a/lib/iov_iter.c b/lib/iov_iter.c
index 97db876..672c32f 100644
--- a/lib/iov_iter.c
+++ b/lib/iov_iter.c
@@ -604,7 +604,7 @@ size_t copy_from_iter_nocache(void *addr, size_t bytes, struct iov_iter *i)
 		return 0;
 	}
 	iterate_and_advance(i, bytes, v,
-		__copy_from_user_nocache((to += v.iov_len) - v.iov_len,
+		__copy_from_user_inatomic_nocache((to += v.iov_len) - v.iov_len,
 					 v.iov_base, v.iov_len),
 		memcpy_from_page((to += v.bv_len) - v.bv_len, v.bv_page,
 				 v.bv_offset, v.bv_len),
@@ -625,7 +625,7 @@ bool copy_from_iter_full_nocache(void *addr, size_t bytes, struct iov_iter *i)
 	if (unlikely(i->count < bytes))
 		return false;
 	iterate_all_kinds(i, bytes, v, ({
-		if (__copy_from_user_nocache((to += v.iov_len) - v.iov_len,
+		if (__copy_from_user_inatomic_nocache((to += v.iov_len) - v.iov_len,
 					     v.iov_base, v.iov_len))
 			return false;
 		0;}),
diff --git a/net/rds/tcp.c b/net/rds/tcp.c
index a973d3b..731282f 100644
--- a/net/rds/tcp.c
+++ b/net/rds/tcp.c
@@ -84,13 +84,10 @@ static struct ctl_table rds_tcp_sysctl_table[] = {
 /* doing it this way avoids calling tcp_sk() */
 void rds_tcp_nonagle(struct socket *sock)
 {
-	mm_segment_t oldfs = get_fs();
 	int val = 1;
 
-	set_fs(KERNEL_DS);
-	sock->ops->setsockopt(sock, SOL_TCP, TCP_NODELAY, (char __user *)&val,
+	kernel_setsockopt(sock, SOL_TCP, TCP_NODELAY, (void *)&val,
 			      sizeof(val));
-	set_fs(oldfs);
 }
 
 u32 rds_tcp_snd_nxt(struct rds_tcp_connection *tc)
diff --git a/net/rds/tcp_send.c b/net/rds/tcp_send.c
index dcf4742..98755b7 100644
--- a/net/rds/tcp_send.c
+++ b/net/rds/tcp_send.c
@@ -40,13 +40,8 @@
 
 static void rds_tcp_cork(struct socket *sock, int val)
 {
-	mm_segment_t oldfs;
-
-	oldfs = get_fs();
-	set_fs(KERNEL_DS);
-	sock->ops->setsockopt(sock, SOL_TCP, TCP_CORK, (char __user *)&val,
+	kernel_setsockopt(sock, SOL_TCP, TCP_CORK, (char __user *)&val,
 			      sizeof(val));
-	set_fs(oldfs);
 }
 
 void rds_tcp_xmit_path_prepare(struct rds_conn_path *cp)