Re: Linux 4.8.10

[Greg KH]

diff --git a/Makefile b/Makefile
index c1519ab85258..7cf2b4985703 100644
--- a/Makefile
+++ b/Makefile
@@ -1,6 +1,6 @@
 VERSION = 4
 PATCHLEVEL = 8
-SUBLEVEL = 9
+SUBLEVEL = 10
 EXTRAVERSION =
 NAME = Psychotic Stoned Sheep
 
diff --git a/arch/sparc/include/asm/uaccess_64.h b/arch/sparc/include/asm/uaccess_64.h
index 37a315d0ddd4..a6847fc05a6d 100644
--- a/arch/sparc/include/asm/uaccess_64.h
+++ b/arch/sparc/include/asm/uaccess_64.h
@@ -98,7 +98,6 @@ struct exception_table_entry {
         unsigned int insn, fixup;
 };
 
-void __ret_efault(void);
 void __retl_efault(void);
 
 /* Uh, these should become the main single-value transfer routines..
@@ -205,55 +204,34 @@ int __get_user_bad(void);
 unsigned long __must_check ___copy_from_user(void *to,
 					     const void __user *from,
 					     unsigned long size);
-unsigned long copy_from_user_fixup(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)
 {
-	unsigned long ret;
-
 	check_object_size(to, size, false);
 
-	ret = ___copy_from_user(to, from, size);
-	if (unlikely(ret))
-		ret = copy_from_user_fixup(to, from, size);
-
-	return ret;
+	return ___copy_from_user(to, from, size);
 }
 #define __copy_from_user copy_from_user
 
 unsigned long __must_check ___copy_to_user(void __user *to,
 					   const void *from,
 					   unsigned long size);
-unsigned long copy_to_user_fixup(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)
 {
-	unsigned long ret;
-
 	check_object_size(from, size, true);
 
-	ret = ___copy_to_user(to, from, size);
-	if (unlikely(ret))
-		ret = copy_to_user_fixup(to, from, size);
-	return ret;
+	return ___copy_to_user(to, from, size);
 }
 #define __copy_to_user copy_to_user
 
 unsigned long __must_check ___copy_in_user(void __user *to,
 					   const void __user *from,
 					   unsigned long size);
-unsigned long copy_in_user_fixup(void __user *to, void __user *from,
-				 unsigned long size);
 static inline unsigned long __must_check
 copy_in_user(void __user *to, void __user *from, unsigned long size)
 {
-	unsigned long ret = ___copy_in_user(to, from, size);
-
-	if (unlikely(ret))
-		ret = copy_in_user_fixup(to, from, size);
-	return ret;
+	return ___copy_in_user(to, from, size);
 }
 #define __copy_in_user copy_in_user
 
diff --git a/arch/sparc/kernel/head_64.S b/arch/sparc/kernel/head_64.S
index a076b4249e62..5f1f3ae21657 100644
--- a/arch/sparc/kernel/head_64.S
+++ b/arch/sparc/kernel/head_64.S
@@ -922,47 +922,11 @@ prom_tba:	.xword	0
 tlb_type:	.word	0	/* Must NOT end up in BSS */
 	.section	".fixup",#alloc,#execinstr
 
-	.globl	__ret_efault, __retl_efault, __ret_one, __retl_one
-ENTRY(__ret_efault)
-	ret
-	 restore %g0, -EFAULT, %o0
-ENDPROC(__ret_efault)
-
 ENTRY(__retl_efault)
 	retl
 	 mov	-EFAULT, %o0
 ENDPROC(__retl_efault)
 
-ENTRY(__retl_one)
-	retl
-	 mov	1, %o0
-ENDPROC(__retl_one)
-
-ENTRY(__retl_one_fp)
-	VISExitHalf
-	retl
-	 mov	1, %o0
-ENDPROC(__retl_one_fp)
-
-ENTRY(__ret_one_asi)
-	wr	%g0, ASI_AIUS, %asi
-	ret
-	 restore %g0, 1, %o0
-ENDPROC(__ret_one_asi)
-
-ENTRY(__retl_one_asi)
-	wr	%g0, ASI_AIUS, %asi
-	retl
-	 mov	1, %o0
-ENDPROC(__retl_one_asi)
-
-ENTRY(__retl_one_asi_fp)
-	wr	%g0, ASI_AIUS, %asi
-	VISExitHalf
-	retl
-	 mov	1, %o0
-ENDPROC(__retl_one_asi_fp)
-
 ENTRY(__retl_o1)
 	retl
 	 mov	%o1, %o0
diff --git a/arch/sparc/kernel/jump_label.c b/arch/sparc/kernel/jump_label.c
index 59bbeff55024..07933b9e9ce0 100644
--- a/arch/sparc/kernel/jump_label.c
+++ b/arch/sparc/kernel/jump_label.c
@@ -13,19 +13,30 @@
 void arch_jump_label_transform(struct jump_entry *entry,
 			       enum jump_label_type type)
 {
-	u32 val;
 	u32 *insn = (u32 *) (unsigned long) entry->code;
+	u32 val;
 
 	if (type == JUMP_LABEL_JMP) {
 		s32 off = (s32)entry->target - (s32)entry->code;
+		bool use_v9_branch = false;
+
+		BUG_ON(off & 3);
 
 #ifdef CONFIG_SPARC64
-		/* ba,pt %xcc, . + (off << 2) */
-		val = 0x10680000 | ((u32) off >> 2);
-#else
-		/* ba . + (off << 2) */
-		val = 0x10800000 | ((u32) off >> 2);
+		if (off <= 0xfffff && off >= -0x100000)
+			use_v9_branch = true;
 #endif
+		if (use_v9_branch) {
+			/* WDISP19 - target is . + immed << 2 */
+			/* ba,pt %xcc, . + off */
+			val = 0x10680000 | (((u32) off >> 2) & 0x7ffff);
+		} else {
+			/* WDISP22 - target is . + immed << 2 */
+			BUG_ON(off > 0x7fffff);
+			BUG_ON(off < -0x800000);
+			/* ba . + off */
+			val = 0x10800000 | (((u32) off >> 2) & 0x3fffff);
+		}
 	} else {
 		val = 0x01000000;
 	}
diff --git a/arch/sparc/kernel/sparc_ksyms_64.c b/arch/sparc/kernel/sparc_ksyms_64.c
index 9e034f29dcc5..20ffb052fe38 100644
--- a/arch/sparc/kernel/sparc_ksyms_64.c
+++ b/arch/sparc/kernel/sparc_ksyms_64.c
@@ -27,7 +27,6 @@ EXPORT_SYMBOL(__flushw_user);
 EXPORT_SYMBOL_GPL(real_hard_smp_processor_id);
 
 /* from head_64.S */
-EXPORT_SYMBOL(__ret_efault);
 EXPORT_SYMBOL(tlb_type);
 EXPORT_SYMBOL(sun4v_chip_type);
 EXPORT_SYMBOL(prom_root_node);
diff --git a/arch/sparc/lib/GENcopy_from_user.S b/arch/sparc/lib/GENcopy_from_user.S
index b7d0bd6b1406..69a439fa2fc1 100644
--- a/arch/sparc/lib/GENcopy_from_user.S
+++ b/arch/sparc/lib/GENcopy_from_user.S
@@ -3,11 +3,11 @@
  * Copyright (C) 2007 David S. Miller (davem@xxxxxxxxxxxxx)
  */
 
-#define EX_LD(x)		\
+#define EX_LD(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one;	\
+	.word 98b, y;		\
 	.text;			\
 	.align 4;
 
diff --git a/arch/sparc/lib/GENcopy_to_user.S b/arch/sparc/lib/GENcopy_to_user.S
index 780550e1afc7..9947427ce354 100644
--- a/arch/sparc/lib/GENcopy_to_user.S
+++ b/arch/sparc/lib/GENcopy_to_user.S
@@ -3,11 +3,11 @@
  * Copyright (C) 2007 David S. Miller (davem@xxxxxxxxxxxxx)
  */
 
-#define EX_ST(x)		\
+#define EX_ST(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one;	\
+	.word 98b, y;		\
 	.text;			\
 	.align 4;
 
diff --git a/arch/sparc/lib/GENmemcpy.S b/arch/sparc/lib/GENmemcpy.S
index 89358ee94851..059ea24ad73d 100644
--- a/arch/sparc/lib/GENmemcpy.S
+++ b/arch/sparc/lib/GENmemcpy.S
@@ -4,21 +4,18 @@
  */
 
 #ifdef __KERNEL__
+#include <linux/linkage.h>
 #define GLOBAL_SPARE	%g7
 #else
 #define GLOBAL_SPARE	%g5
 #endif
 
 #ifndef EX_LD
-#define EX_LD(x)	x
+#define EX_LD(x,y)	x
 #endif
 
 #ifndef EX_ST
-#define EX_ST(x)	x
-#endif
-
-#ifndef EX_RETVAL
-#define EX_RETVAL(x)	x
+#define EX_ST(x,y)	x
 #endif
 
 #ifndef LOAD
@@ -45,6 +42,29 @@
 	.register	%g3,#scratch
 
 	.text
+
+#ifndef EX_RETVAL
+#define EX_RETVAL(x)	x
+ENTRY(GEN_retl_o4_1)
+	add	%o4, %o2, %o4
+	retl
+	 add	%o4, 1, %o0
+ENDPROC(GEN_retl_o4_1)
+ENTRY(GEN_retl_g1_8)
+	add	%g1, %o2, %g1
+	retl
+	 add	%g1, 8, %o0
+ENDPROC(GEN_retl_g1_8)
+ENTRY(GEN_retl_o2_4)
+	retl
+	 add	%o2, 4, %o0
+ENDPROC(GEN_retl_o2_4)
+ENTRY(GEN_retl_o2_1)
+	retl
+	 add	%o2, 1, %o0
+ENDPROC(GEN_retl_o2_1)
+#endif
+
 	.align		64
 
 	.globl	FUNC_NAME
@@ -73,8 +93,8 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	sub		%g0, %o4, %o4
 	sub		%o2, %o4, %o2
 1:	subcc		%o4, 1, %o4
-	EX_LD(LOAD(ldub, %o1, %g1))
-	EX_ST(STORE(stb, %g1, %o0))
+	EX_LD(LOAD(ldub, %o1, %g1),GEN_retl_o4_1)
+	EX_ST(STORE(stb, %g1, %o0),GEN_retl_o4_1)
 	add		%o1, 1, %o1
 	bne,pt		%XCC, 1b
 	add		%o0, 1, %o0
@@ -82,8 +102,8 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	andn		%o2, 0x7, %g1
 	sub		%o2, %g1, %o2
 1:	subcc		%g1, 0x8, %g1
-	EX_LD(LOAD(ldx, %o1, %g2))
-	EX_ST(STORE(stx, %g2, %o0))
+	EX_LD(LOAD(ldx, %o1, %g2),GEN_retl_g1_8)
+	EX_ST(STORE(stx, %g2, %o0),GEN_retl_g1_8)
 	add		%o1, 0x8, %o1
 	bne,pt		%XCC, 1b
 	 add		%o0, 0x8, %o0
@@ -100,8 +120,8 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 
 1:
 	subcc		%o2, 4, %o2
-	EX_LD(LOAD(lduw, %o1, %g1))
-	EX_ST(STORE(stw, %g1, %o1 + %o3))
+	EX_LD(LOAD(lduw, %o1, %g1),GEN_retl_o2_4)
+	EX_ST(STORE(stw, %g1, %o1 + %o3),GEN_retl_o2_4)
 	bgu,pt		%XCC, 1b
 	 add		%o1, 4, %o1
 
@@ -111,8 +131,8 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	.align		32
 90:
 	subcc		%o2, 1, %o2
-	EX_LD(LOAD(ldub, %o1, %g1))
-	EX_ST(STORE(stb, %g1, %o1 + %o3))
+	EX_LD(LOAD(ldub, %o1, %g1),GEN_retl_o2_1)
+	EX_ST(STORE(stb, %g1, %o1 + %o3),GEN_retl_o2_1)
 	bgu,pt		%XCC, 90b
 	 add		%o1, 1, %o1
 	retl
diff --git a/arch/sparc/lib/Makefile b/arch/sparc/lib/Makefile
index 3269b0234093..4f2384a4286a 100644
--- a/arch/sparc/lib/Makefile
+++ b/arch/sparc/lib/Makefile
@@ -38,7 +38,7 @@ lib-$(CONFIG_SPARC64) +=  NG4patch.o NG4copy_page.o NG4clear_page.o NG4memset.o
 lib-$(CONFIG_SPARC64) += GENmemcpy.o GENcopy_from_user.o GENcopy_to_user.o
 lib-$(CONFIG_SPARC64) += GENpatch.o GENpage.o GENbzero.o
 
-lib-$(CONFIG_SPARC64) += copy_in_user.o user_fixup.o memmove.o
+lib-$(CONFIG_SPARC64) += copy_in_user.o memmove.o
 lib-$(CONFIG_SPARC64) += mcount.o ipcsum.o xor.o hweight.o ffs.o
 
 obj-$(CONFIG_SPARC64) += iomap.o
diff --git a/arch/sparc/lib/NG2copy_from_user.S b/arch/sparc/lib/NG2copy_from_user.S
index d5242b8c4f94..b79a6998d87c 100644
--- a/arch/sparc/lib/NG2copy_from_user.S
+++ b/arch/sparc/lib/NG2copy_from_user.S
@@ -3,19 +3,19 @@
  * Copyright (C) 2007 David S. Miller (davem@xxxxxxxxxxxxx)
  */
 
-#define EX_LD(x)		\
+#define EX_LD(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one_asi;\
+	.word 98b, y;		\
 	.text;			\
 	.align 4;
 
-#define EX_LD_FP(x)		\
+#define EX_LD_FP(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one_asi_fp;\
+	.word 98b, y##_fp;	\
 	.text;			\
 	.align 4;
 
diff --git a/arch/sparc/lib/NG2copy_to_user.S b/arch/sparc/lib/NG2copy_to_user.S
index 4e962d993b10..dcec55f254ab 100644
--- a/arch/sparc/lib/NG2copy_to_user.S
+++ b/arch/sparc/lib/NG2copy_to_user.S
@@ -3,19 +3,19 @@
  * Copyright (C) 2007 David S. Miller (davem@xxxxxxxxxxxxx)
  */
 
-#define EX_ST(x)		\
+#define EX_ST(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one_asi;\
+	.word 98b, y;		\
 	.text;			\
 	.align 4;
 
-#define EX_ST_FP(x)		\
+#define EX_ST_FP(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one_asi_fp;\
+	.word 98b, y##_fp;	\
 	.text;			\
 	.align 4;
 
diff --git a/arch/sparc/lib/NG2memcpy.S b/arch/sparc/lib/NG2memcpy.S
index d5f585df2f3f..c629dbd121b6 100644
--- a/arch/sparc/lib/NG2memcpy.S
+++ b/arch/sparc/lib/NG2memcpy.S
@@ -4,6 +4,7 @@
  */
 
 #ifdef __KERNEL__
+#include <linux/linkage.h>
 #include <asm/visasm.h>
 #include <asm/asi.h>
 #define GLOBAL_SPARE	%g7
@@ -32,21 +33,17 @@
 #endif
 
 #ifndef EX_LD
-#define EX_LD(x)	x
+#define EX_LD(x,y)	x
 #endif
 #ifndef EX_LD_FP
-#define EX_LD_FP(x)	x
+#define EX_LD_FP(x,y)	x
 #endif
 
 #ifndef EX_ST
-#define EX_ST(x)	x
+#define EX_ST(x,y)	x
 #endif
 #ifndef EX_ST_FP
-#define EX_ST_FP(x)	x
-#endif
-
-#ifndef EX_RETVAL
-#define EX_RETVAL(x)	x
+#define EX_ST_FP(x,y)	x
 #endif
 
 #ifndef LOAD
@@ -140,45 +137,110 @@
 	fsrc2		%x6, %f12; \
 	fsrc2		%x7, %f14;
 #define FREG_LOAD_1(base, x0) \
-	EX_LD_FP(LOAD(ldd, base + 0x00, %x0))
+	EX_LD_FP(LOAD(ldd, base + 0x00, %x0), NG2_retl_o2_plus_g1)
 #define FREG_LOAD_2(base, x0, x1) \
-	EX_LD_FP(LOAD(ldd, base + 0x00, %x0)); \
-	EX_LD_FP(LOAD(ldd, base + 0x08, %x1));
+	EX_LD_FP(LOAD(ldd, base + 0x00, %x0), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x08, %x1), NG2_retl_o2_plus_g1);
 #define FREG_LOAD_3(base, x0, x1, x2) \
-	EX_LD_FP(LOAD(ldd, base + 0x00, %x0)); \
-	EX_LD_FP(LOAD(ldd, base + 0x08, %x1)); \
-	EX_LD_FP(LOAD(ldd, base + 0x10, %x2));
+	EX_LD_FP(LOAD(ldd, base + 0x00, %x0), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x08, %x1), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x10, %x2), NG2_retl_o2_plus_g1);
 #define FREG_LOAD_4(base, x0, x1, x2, x3) \
-	EX_LD_FP(LOAD(ldd, base + 0x00, %x0)); \
-	EX_LD_FP(LOAD(ldd, base + 0x08, %x1)); \
-	EX_LD_FP(LOAD(ldd, base + 0x10, %x2)); \
-	EX_LD_FP(LOAD(ldd, base + 0x18, %x3));
+	EX_LD_FP(LOAD(ldd, base + 0x00, %x0), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x08, %x1), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x10, %x2), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x18, %x3), NG2_retl_o2_plus_g1);
 #define FREG_LOAD_5(base, x0, x1, x2, x3, x4) \
-	EX_LD_FP(LOAD(ldd, base + 0x00, %x0)); \
-	EX_LD_FP(LOAD(ldd, base + 0x08, %x1)); \
-	EX_LD_FP(LOAD(ldd, base + 0x10, %x2)); \
-	EX_LD_FP(LOAD(ldd, base + 0x18, %x3)); \
-	EX_LD_FP(LOAD(ldd, base + 0x20, %x4));
+	EX_LD_FP(LOAD(ldd, base + 0x00, %x0), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x08, %x1), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x10, %x2), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x18, %x3), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x20, %x4), NG2_retl_o2_plus_g1);
 #define FREG_LOAD_6(base, x0, x1, x2, x3, x4, x5) \
-	EX_LD_FP(LOAD(ldd, base + 0x00, %x0)); \
-	EX_LD_FP(LOAD(ldd, base + 0x08, %x1)); \
-	EX_LD_FP(LOAD(ldd, base + 0x10, %x2)); \
-	EX_LD_FP(LOAD(ldd, base + 0x18, %x3)); \
-	EX_LD_FP(LOAD(ldd, base + 0x20, %x4)); \
-	EX_LD_FP(LOAD(ldd, base + 0x28, %x5));
+	EX_LD_FP(LOAD(ldd, base + 0x00, %x0), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x08, %x1), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x10, %x2), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x18, %x3), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x20, %x4), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x28, %x5), NG2_retl_o2_plus_g1);
 #define FREG_LOAD_7(base, x0, x1, x2, x3, x4, x5, x6) \
-	EX_LD_FP(LOAD(ldd, base + 0x00, %x0)); \
-	EX_LD_FP(LOAD(ldd, base + 0x08, %x1)); \
-	EX_LD_FP(LOAD(ldd, base + 0x10, %x2)); \
-	EX_LD_FP(LOAD(ldd, base + 0x18, %x3)); \
-	EX_LD_FP(LOAD(ldd, base + 0x20, %x4)); \
-	EX_LD_FP(LOAD(ldd, base + 0x28, %x5)); \
-	EX_LD_FP(LOAD(ldd, base + 0x30, %x6));
+	EX_LD_FP(LOAD(ldd, base + 0x00, %x0), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x08, %x1), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x10, %x2), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x18, %x3), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x20, %x4), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x28, %x5), NG2_retl_o2_plus_g1); \
+	EX_LD_FP(LOAD(ldd, base + 0x30, %x6), NG2_retl_o2_plus_g1);
 
 	.register	%g2,#scratch
 	.register	%g3,#scratch
 
 	.text
+#ifndef EX_RETVAL
+#define EX_RETVAL(x)	x
+__restore_fp:
+	VISExitHalf
+__restore_asi:
+	retl
+	 wr	%g0, ASI_AIUS, %asi
+ENTRY(NG2_retl_o2)
+	ba,pt	%xcc, __restore_asi
+	 mov	%o2, %o0
+ENDPROC(NG2_retl_o2)
+ENTRY(NG2_retl_o2_plus_1)
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, 1, %o0
+ENDPROC(NG2_retl_o2_plus_1)
+ENTRY(NG2_retl_o2_plus_4)
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, 4, %o0
+ENDPROC(NG2_retl_o2_plus_4)
+ENTRY(NG2_retl_o2_plus_8)
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, 8, %o0
+ENDPROC(NG2_retl_o2_plus_8)
+ENTRY(NG2_retl_o2_plus_o4_plus_1)
+	add	%o4, 1, %o4
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o4, %o0
+ENDPROC(NG2_retl_o2_plus_o4_plus_1)
+ENTRY(NG2_retl_o2_plus_o4_plus_8)
+	add	%o4, 8, %o4
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o4, %o0
+ENDPROC(NG2_retl_o2_plus_o4_plus_8)
+ENTRY(NG2_retl_o2_plus_o4_plus_16)
+	add	%o4, 16, %o4
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o4, %o0
+ENDPROC(NG2_retl_o2_plus_o4_plus_16)
+ENTRY(NG2_retl_o2_plus_g1_fp)
+	ba,pt	%xcc, __restore_fp
+	 add	%o2, %g1, %o0
+ENDPROC(NG2_retl_o2_plus_g1_fp)
+ENTRY(NG2_retl_o2_plus_g1_plus_64_fp)
+	add	%g1, 64, %g1
+	ba,pt	%xcc, __restore_fp
+	 add	%o2, %g1, %o0
+ENDPROC(NG2_retl_o2_plus_g1_plus_64_fp)
+ENTRY(NG2_retl_o2_plus_g1_plus_1)
+	add	%g1, 1, %g1
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %g1, %o0
+ENDPROC(NG2_retl_o2_plus_g1_plus_1)
+ENTRY(NG2_retl_o2_and_7_plus_o4)
+	and	%o2, 7, %o2
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o4, %o0
+ENDPROC(NG2_retl_o2_and_7_plus_o4)
+ENTRY(NG2_retl_o2_and_7_plus_o4_plus_8)
+	and	%o2, 7, %o2
+	add	%o4, 8, %o4
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o4, %o0
+ENDPROC(NG2_retl_o2_and_7_plus_o4_plus_8)
+#endif
+
 	.align		64
 
 	.globl	FUNC_NAME
@@ -230,8 +292,8 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	sub		%g0, %o4, %o4	! bytes to align dst
 	sub		%o2, %o4, %o2
 1:	subcc		%o4, 1, %o4
-	EX_LD(LOAD(ldub, %o1, %g1))
-	EX_ST(STORE(stb, %g1, %o0))
+	EX_LD(LOAD(ldub, %o1, %g1), NG2_retl_o2_plus_o4_plus_1)
+	EX_ST(STORE(stb, %g1, %o0), NG2_retl_o2_plus_o4_plus_1)
 	add		%o1, 1, %o1
 	bne,pt		%XCC, 1b
 	add		%o0, 1, %o0
@@ -281,11 +343,11 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	 nop
 	/* fall through for 0 < low bits < 8 */
 110:	sub		%o4, 64, %g2
-	EX_LD_FP(LOAD_BLK(%g2, %f0))
-1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
-	EX_LD_FP(LOAD_BLK(%o4, %f16))
+	EX_LD_FP(LOAD_BLK(%g2, %f0), NG2_retl_o2_plus_g1)
+1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+	EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
 	FREG_FROB(f0, f2, f4, f6, f8, f10, f12, f14, f16)
-	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
 	FREG_MOVE_8(f16, f18, f20, f22, f24, f26, f28, f30)
 	subcc		%g1, 64, %g1
 	add		%o4, 64, %o4
@@ -296,10 +358,10 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 
 120:	sub		%o4, 56, %g2
 	FREG_LOAD_7(%g2, f0, f2, f4, f6, f8, f10, f12)
-1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
-	EX_LD_FP(LOAD_BLK(%o4, %f16))
+1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+	EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
 	FREG_FROB(f0, f2, f4, f6, f8, f10, f12, f16, f18)
-	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
 	FREG_MOVE_7(f18, f20, f22, f24, f26, f28, f30)
 	subcc		%g1, 64, %g1
 	add		%o4, 64, %o4
@@ -310,10 +372,10 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 
 130:	sub		%o4, 48, %g2
 	FREG_LOAD_6(%g2, f0, f2, f4, f6, f8, f10)
-1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
-	EX_LD_FP(LOAD_BLK(%o4, %f16))
+1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+	EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
 	FREG_FROB(f0, f2, f4, f6, f8, f10, f16, f18, f20)
-	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
 	FREG_MOVE_6(f20, f22, f24, f26, f28, f30)
 	subcc		%g1, 64, %g1
 	add		%o4, 64, %o4
@@ -324,10 +386,10 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 
 140:	sub		%o4, 40, %g2
 	FREG_LOAD_5(%g2, f0, f2, f4, f6, f8)
-1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
-	EX_LD_FP(LOAD_BLK(%o4, %f16))
+1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+	EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
 	FREG_FROB(f0, f2, f4, f6, f8, f16, f18, f20, f22)
-	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
 	FREG_MOVE_5(f22, f24, f26, f28, f30)
 	subcc		%g1, 64, %g1
 	add		%o4, 64, %o4
@@ -338,10 +400,10 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 
 150:	sub		%o4, 32, %g2
 	FREG_LOAD_4(%g2, f0, f2, f4, f6)
-1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
-	EX_LD_FP(LOAD_BLK(%o4, %f16))
+1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+	EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
 	FREG_FROB(f0, f2, f4, f6, f16, f18, f20, f22, f24)
-	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
 	FREG_MOVE_4(f24, f26, f28, f30)
 	subcc		%g1, 64, %g1
 	add		%o4, 64, %o4
@@ -352,10 +414,10 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 
 160:	sub		%o4, 24, %g2
 	FREG_LOAD_3(%g2, f0, f2, f4)
-1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
-	EX_LD_FP(LOAD_BLK(%o4, %f16))
+1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+	EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
 	FREG_FROB(f0, f2, f4, f16, f18, f20, f22, f24, f26)
-	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
 	FREG_MOVE_3(f26, f28, f30)
 	subcc		%g1, 64, %g1
 	add		%o4, 64, %o4
@@ -366,10 +428,10 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 
 170:	sub		%o4, 16, %g2
 	FREG_LOAD_2(%g2, f0, f2)
-1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
-	EX_LD_FP(LOAD_BLK(%o4, %f16))
+1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+	EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
 	FREG_FROB(f0, f2, f16, f18, f20, f22, f24, f26, f28)
-	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
 	FREG_MOVE_2(f28, f30)
 	subcc		%g1, 64, %g1
 	add		%o4, 64, %o4
@@ -380,10 +442,10 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 
 180:	sub		%o4, 8, %g2
 	FREG_LOAD_1(%g2, f0)
-1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
-	EX_LD_FP(LOAD_BLK(%o4, %f16))
+1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+	EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
 	FREG_FROB(f0, f16, f18, f20, f22, f24, f26, f28, f30)
-	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
 	FREG_MOVE_1(f30)
 	subcc		%g1, 64, %g1
 	add		%o4, 64, %o4
@@ -393,10 +455,10 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	 nop
 
 190:
-1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
+1:	EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
 	subcc		%g1, 64, %g1
-	EX_LD_FP(LOAD_BLK(%o4, %f0))
-	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+	EX_LD_FP(LOAD_BLK(%o4, %f0), NG2_retl_o2_plus_g1_plus_64)
+	EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1_plus_64)
 	add		%o4, 64, %o4
 	bne,pt		%xcc, 1b
 	 LOAD(prefetch, %o4 + 64, #one_read)
@@ -423,28 +485,28 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	andn		%o2, 0xf, %o4
 	and		%o2, 0xf, %o2
 1:	subcc		%o4, 0x10, %o4
-	EX_LD(LOAD(ldx, %o1, %o5))
+	EX_LD(LOAD(ldx, %o1, %o5), NG2_retl_o2_plus_o4_plus_16)
 	add		%o1, 0x08, %o1
-	EX_LD(LOAD(ldx, %o1, %g1))
+	EX_LD(LOAD(ldx, %o1, %g1), NG2_retl_o2_plus_o4_plus_16)
 	sub		%o1, 0x08, %o1
-	EX_ST(STORE(stx, %o5, %o1 + GLOBAL_SPARE))
+	EX_ST(STORE(stx, %o5, %o1 + GLOBAL_SPARE), NG2_retl_o2_plus_o4_plus_16)
 	add		%o1, 0x8, %o1
-	EX_ST(STORE(stx, %g1, %o1 + GLOBAL_SPARE))
+	EX_ST(STORE(stx, %g1, %o1 + GLOBAL_SPARE), NG2_retl_o2_plus_o4_plus_8)
 	bgu,pt		%XCC, 1b
 	 add		%o1, 0x8, %o1
 73:	andcc		%o2, 0x8, %g0
 	be,pt		%XCC, 1f
 	 nop
 	sub		%o2, 0x8, %o2
-	EX_LD(LOAD(ldx, %o1, %o5))
-	EX_ST(STORE(stx, %o5, %o1 + GLOBAL_SPARE))
+	EX_LD(LOAD(ldx, %o1, %o5), NG2_retl_o2_plus_8)
+	EX_ST(STORE(stx, %o5, %o1 + GLOBAL_SPARE), NG2_retl_o2_plus_8)
 	add		%o1, 0x8, %o1
 1:	andcc		%o2, 0x4, %g0
 	be,pt		%XCC, 1f
 	 nop
 	sub		%o2, 0x4, %o2
-	EX_LD(LOAD(lduw, %o1, %o5))
-	EX_ST(STORE(stw, %o5, %o1 + GLOBAL_SPARE))
+	EX_LD(LOAD(lduw, %o1, %o5), NG2_retl_o2_plus_4)
+	EX_ST(STORE(stw, %o5, %o1 + GLOBAL_SPARE), NG2_retl_o2_plus_4)
 	add		%o1, 0x4, %o1
 1:	cmp		%o2, 0
 	be,pt		%XCC, 85f
@@ -460,8 +522,8 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	sub		%o2, %g1, %o2
 
 1:	subcc		%g1, 1, %g1
-	EX_LD(LOAD(ldub, %o1, %o5))
-	EX_ST(STORE(stb, %o5, %o1 + GLOBAL_SPARE))
+	EX_LD(LOAD(ldub, %o1, %o5), NG2_retl_o2_plus_g1_plus_1)
+	EX_ST(STORE(stb, %o5, %o1 + GLOBAL_SPARE), NG2_retl_o2_plus_g1_plus_1)
 	bgu,pt		%icc, 1b
 	 add		%o1, 1, %o1
 
@@ -477,16 +539,16 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 
 8:	mov		64, GLOBAL_SPARE
 	andn		%o1, 0x7, %o1
-	EX_LD(LOAD(ldx, %o1, %g2))
+	EX_LD(LOAD(ldx, %o1, %g2), NG2_retl_o2)
 	sub		GLOBAL_SPARE, %g1, GLOBAL_SPARE
 	andn		%o2, 0x7, %o4
 	sllx		%g2, %g1, %g2
 1:	add		%o1, 0x8, %o1
-	EX_LD(LOAD(ldx, %o1, %g3))
+	EX_LD(LOAD(ldx, %o1, %g3), NG2_retl_o2_and_7_plus_o4)
 	subcc		%o4, 0x8, %o4
 	srlx		%g3, GLOBAL_SPARE, %o5
 	or		%o5, %g2, %o5
-	EX_ST(STORE(stx, %o5, %o0))
+	EX_ST(STORE(stx, %o5, %o0), NG2_retl_o2_and_7_plus_o4_plus_8)
 	add		%o0, 0x8, %o0
 	bgu,pt		%icc, 1b
 	 sllx		%g3, %g1, %g2
@@ -506,8 +568,8 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 
 1:
 	subcc		%o2, 4, %o2
-	EX_LD(LOAD(lduw, %o1, %g1))
-	EX_ST(STORE(stw, %g1, %o1 + GLOBAL_SPARE))
+	EX_LD(LOAD(lduw, %o1, %g1), NG2_retl_o2_plus_4)
+	EX_ST(STORE(stw, %g1, %o1 + GLOBAL_SPARE), NG2_retl_o2_plus_4)
 	bgu,pt		%XCC, 1b
 	 add		%o1, 4, %o1
 
@@ -517,8 +579,8 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	.align		32
 90:
 	subcc		%o2, 1, %o2
-	EX_LD(LOAD(ldub, %o1, %g1))
-	EX_ST(STORE(stb, %g1, %o1 + GLOBAL_SPARE))
+	EX_LD(LOAD(ldub, %o1, %g1), NG2_retl_o2_plus_1)
+	EX_ST(STORE(stb, %g1, %o1 + GLOBAL_SPARE), NG2_retl_o2_plus_1)
 	bgu,pt		%XCC, 90b
 	 add		%o1, 1, %o1
 	retl
diff --git a/arch/sparc/lib/NG4copy_from_user.S b/arch/sparc/lib/NG4copy_from_user.S
index 2e8ee7ad07a9..16a286c1a528 100644
--- a/arch/sparc/lib/NG4copy_from_user.S
+++ b/arch/sparc/lib/NG4copy_from_user.S
@@ -3,19 +3,19 @@
  * Copyright (C) 2012 David S. Miller (davem@xxxxxxxxxxxxx)
  */
 
-#define EX_LD(x)		\
+#define EX_LD(x, y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one_asi;\
+	.word 98b, y;		\
 	.text;			\
 	.align 4;
 
-#define EX_LD_FP(x)		\
+#define EX_LD_FP(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one_asi_fp;\
+	.word 98b, y##_fp;	\
 	.text;			\
 	.align 4;
 
diff --git a/arch/sparc/lib/NG4copy_to_user.S b/arch/sparc/lib/NG4copy_to_user.S
index be0bf4590df8..6b0276ffc858 100644
--- a/arch/sparc/lib/NG4copy_to_user.S
+++ b/arch/sparc/lib/NG4copy_to_user.S
@@ -3,19 +3,19 @@
  * Copyright (C) 2012 David S. Miller (davem@xxxxxxxxxxxxx)
  */
 
-#define EX_ST(x)		\
+#define EX_ST(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one_asi;\
+	.word 98b, y;		\
 	.text;			\
 	.align 4;
 
-#define EX_ST_FP(x)		\
+#define EX_ST_FP(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one_asi_fp;\
+	.word 98b, y##_fp;	\
 	.text;			\
 	.align 4;
 
diff --git a/arch/sparc/lib/NG4memcpy.S b/arch/sparc/lib/NG4memcpy.S
index 8e13ee1f4454..75bb93b1437f 100644
--- a/arch/sparc/lib/NG4memcpy.S
+++ b/arch/sparc/lib/NG4memcpy.S
@@ -4,6 +4,7 @@
  */
 
 #ifdef __KERNEL__
+#include <linux/linkage.h>
 #include <asm/visasm.h>
 #include <asm/asi.h>
 #define GLOBAL_SPARE	%g7
@@ -46,22 +47,19 @@
 #endif
 
 #ifndef EX_LD
-#define EX_LD(x)	x
+#define EX_LD(x,y)	x
 #endif
 #ifndef EX_LD_FP
-#define EX_LD_FP(x)	x
+#define EX_LD_FP(x,y)	x
 #endif
 
 #ifndef EX_ST
-#define EX_ST(x)	x
+#define EX_ST(x,y)	x
 #endif
 #ifndef EX_ST_FP
-#define EX_ST_FP(x)	x
+#define EX_ST_FP(x,y)	x
 #endif
 
-#ifndef EX_RETVAL
-#define EX_RETVAL(x)	x
-#endif
 
 #ifndef LOAD
 #define LOAD(type,addr,dest)	type [addr], dest
@@ -94,6 +92,158 @@
 	.register	%g3,#scratch
 
 	.text
+#ifndef EX_RETVAL
+#define EX_RETVAL(x)	x
+__restore_asi_fp:
+	VISExitHalf
+__restore_asi:
+	retl
+	 wr	%g0, ASI_AIUS, %asi
+
+ENTRY(NG4_retl_o2)
+	ba,pt	%xcc, __restore_asi
+	 mov	%o2, %o0
+ENDPROC(NG4_retl_o2)
+ENTRY(NG4_retl_o2_plus_1)
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, 1, %o0
+ENDPROC(NG4_retl_o2_plus_1)
+ENTRY(NG4_retl_o2_plus_4)
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, 4, %o0
+ENDPROC(NG4_retl_o2_plus_4)
+ENTRY(NG4_retl_o2_plus_o5)
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o5, %o0
+ENDPROC(NG4_retl_o2_plus_o5)
+ENTRY(NG4_retl_o2_plus_o5_plus_4)
+	add	%o5, 4, %o5
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o5, %o0
+ENDPROC(NG4_retl_o2_plus_o5_plus_4)
+ENTRY(NG4_retl_o2_plus_o5_plus_8)
+	add	%o5, 8, %o5
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o5, %o0
+ENDPROC(NG4_retl_o2_plus_o5_plus_8)
+ENTRY(NG4_retl_o2_plus_o5_plus_16)
+	add	%o5, 16, %o5
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o5, %o0
+ENDPROC(NG4_retl_o2_plus_o5_plus_16)
+ENTRY(NG4_retl_o2_plus_o5_plus_24)
+	add	%o5, 24, %o5
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o5, %o0
+ENDPROC(NG4_retl_o2_plus_o5_plus_24)
+ENTRY(NG4_retl_o2_plus_o5_plus_32)
+	add	%o5, 32, %o5
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o5, %o0
+ENDPROC(NG4_retl_o2_plus_o5_plus_32)
+ENTRY(NG4_retl_o2_plus_g1)
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %g1, %o0
+ENDPROC(NG4_retl_o2_plus_g1)
+ENTRY(NG4_retl_o2_plus_g1_plus_1)
+	add	%g1, 1, %g1
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %g1, %o0
+ENDPROC(NG4_retl_o2_plus_g1_plus_1)
+ENTRY(NG4_retl_o2_plus_g1_plus_8)
+	add	%g1, 8, %g1
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %g1, %o0
+ENDPROC(NG4_retl_o2_plus_g1_plus_8)
+ENTRY(NG4_retl_o2_plus_o4)
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4)
+ENTRY(NG4_retl_o2_plus_o4_plus_8)
+	add	%o4, 8, %o4
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_8)
+ENTRY(NG4_retl_o2_plus_o4_plus_16)
+	add	%o4, 16, %o4
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_16)
+ENTRY(NG4_retl_o2_plus_o4_plus_24)
+	add	%o4, 24, %o4
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_24)
+ENTRY(NG4_retl_o2_plus_o4_plus_32)
+	add	%o4, 32, %o4
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_32)
+ENTRY(NG4_retl_o2_plus_o4_plus_40)
+	add	%o4, 40, %o4
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_40)
+ENTRY(NG4_retl_o2_plus_o4_plus_48)
+	add	%o4, 48, %o4
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_48)
+ENTRY(NG4_retl_o2_plus_o4_plus_56)
+	add	%o4, 56, %o4
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_56)
+ENTRY(NG4_retl_o2_plus_o4_plus_64)
+	add	%o4, 64, %o4
+	ba,pt	%xcc, __restore_asi
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_64)
+ENTRY(NG4_retl_o2_plus_o4_fp)
+	ba,pt	%xcc, __restore_asi_fp
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_8_fp)
+	add	%o4, 8, %o4
+	ba,pt	%xcc, __restore_asi_fp
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_8_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_16_fp)
+	add	%o4, 16, %o4
+	ba,pt	%xcc, __restore_asi_fp
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_16_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_24_fp)
+	add	%o4, 24, %o4
+	ba,pt	%xcc, __restore_asi_fp
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_24_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_32_fp)
+	add	%o4, 32, %o4
+	ba,pt	%xcc, __restore_asi_fp
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_32_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_40_fp)
+	add	%o4, 40, %o4
+	ba,pt	%xcc, __restore_asi_fp
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_40_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_48_fp)
+	add	%o4, 48, %o4
+	ba,pt	%xcc, __restore_asi_fp
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_48_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_56_fp)
+	add	%o4, 56, %o4
+	ba,pt	%xcc, __restore_asi_fp
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_56_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_64_fp)
+	add	%o4, 64, %o4
+	ba,pt	%xcc, __restore_asi_fp
+	 add	%o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_64_fp)
+#endif
 	.align		64
 
 	.globl	FUNC_NAME
@@ -124,12 +274,13 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	brz,pt		%g1, 51f
 	 sub		%o2, %g1, %o2
 
-1:	EX_LD(LOAD(ldub, %o1 + 0x00, %g2))
+
+1:	EX_LD(LOAD(ldub, %o1 + 0x00, %g2), NG4_retl_o2_plus_g1)
 	add		%o1, 1, %o1
 	subcc		%g1, 1, %g1
 	add		%o0, 1, %o0
 	bne,pt		%icc, 1b
-	 EX_ST(STORE(stb, %g2, %o0 - 0x01))
+	 EX_ST(STORE(stb, %g2, %o0 - 0x01), NG4_retl_o2_plus_g1_plus_1)
 
 51:	LOAD(prefetch, %o1 + 0x040, #n_reads_strong)
 	LOAD(prefetch, %o1 + 0x080, #n_reads_strong)
@@ -154,43 +305,43 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	brz,pt		%g1, .Llarge_aligned
 	 sub		%o2, %g1, %o2
 
-1:	EX_LD(LOAD(ldx, %o1 + 0x00, %g2))
+1:	EX_LD(LOAD(ldx, %o1 + 0x00, %g2), NG4_retl_o2_plus_g1)
 	add		%o1, 8, %o1
 	subcc		%g1, 8, %g1
 	add		%o0, 8, %o0
 	bne,pt		%icc, 1b
-	 EX_ST(STORE(stx, %g2, %o0 - 0x08))
+	 EX_ST(STORE(stx, %g2, %o0 - 0x08), NG4_retl_o2_plus_g1_plus_8)
 
 .Llarge_aligned:
 	/* len >= 0x80 && src 8-byte aligned && dest 8-byte aligned */
 	andn		%o2, 0x3f, %o4
 	sub		%o2, %o4, %o2
 
-1:	EX_LD(LOAD(ldx, %o1 + 0x00, %g1))
+1:	EX_LD(LOAD(ldx, %o1 + 0x00, %g1), NG4_retl_o2_plus_o4)
 	add		%o1, 0x40, %o1
-	EX_LD(LOAD(ldx, %o1 - 0x38, %g2))
+	EX_LD(LOAD(ldx, %o1 - 0x38, %g2), NG4_retl_o2_plus_o4)
 	subcc		%o4, 0x40, %o4
-	EX_LD(LOAD(ldx, %o1 - 0x30, %g3))
-	EX_LD(LOAD(ldx, %o1 - 0x28, GLOBAL_SPARE))
-	EX_LD(LOAD(ldx, %o1 - 0x20, %o5))
-	EX_ST(STORE_INIT(%g1, %o0))
+	EX_LD(LOAD(ldx, %o1 - 0x30, %g3), NG4_retl_o2_plus_o4_plus_64)
+	EX_LD(LOAD(ldx, %o1 - 0x28, GLOBAL_SPARE), NG4_retl_o2_plus_o4_plus_64)
+	EX_LD(LOAD(ldx, %o1 - 0x20, %o5), NG4_retl_o2_plus_o4_plus_64)
+	EX_ST(STORE_INIT(%g1, %o0), NG4_retl_o2_plus_o4_plus_64)
 	add		%o0, 0x08, %o0
-	EX_ST(STORE_INIT(%g2, %o0))
+	EX_ST(STORE_INIT(%g2, %o0), NG4_retl_o2_plus_o4_plus_56)
 	add		%o0, 0x08, %o0
-	EX_LD(LOAD(ldx, %o1 - 0x18, %g2))
-	EX_ST(STORE_INIT(%g3, %o0))
+	EX_LD(LOAD(ldx, %o1 - 0x18, %g2), NG4_retl_o2_plus_o4_plus_48)
+	EX_ST(STORE_INIT(%g3, %o0), NG4_retl_o2_plus_o4_plus_48)
 	add		%o0, 0x08, %o0
-	EX_LD(LOAD(ldx, %o1 - 0x10, %g3))
-	EX_ST(STORE_INIT(GLOBAL_SPARE, %o0))
+	EX_LD(LOAD(ldx, %o1 - 0x10, %g3), NG4_retl_o2_plus_o4_plus_40)
+	EX_ST(STORE_INIT(GLOBAL_SPARE, %o0), NG4_retl_o2_plus_o4_plus_40)
 	add		%o0, 0x08, %o0
-	EX_LD(LOAD(ldx, %o1 - 0x08, GLOBAL_SPARE))
-	EX_ST(STORE_INIT(%o5, %o0))
+	EX_LD(LOAD(ldx, %o1 - 0x08, GLOBAL_SPARE), NG4_retl_o2_plus_o4_plus_32)
+	EX_ST(STORE_INIT(%o5, %o0), NG4_retl_o2_plus_o4_plus_32)
 	add		%o0, 0x08, %o0
-	EX_ST(STORE_INIT(%g2, %o0))
+	EX_ST(STORE_INIT(%g2, %o0), NG4_retl_o2_plus_o4_plus_24)
 	add		%o0, 0x08, %o0
-	EX_ST(STORE_INIT(%g3, %o0))
+	EX_ST(STORE_INIT(%g3, %o0), NG4_retl_o2_plus_o4_plus_16)
 	add		%o0, 0x08, %o0
-	EX_ST(STORE_INIT(GLOBAL_SPARE, %o0))
+	EX_ST(STORE_INIT(GLOBAL_SPARE, %o0), NG4_retl_o2_plus_o4_plus_8)
 	add		%o0, 0x08, %o0
 	bne,pt		%icc, 1b
 	 LOAD(prefetch, %o1 + 0x200, #n_reads_strong)
@@ -216,17 +367,17 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	sub		%o2, %o4, %o2
 	alignaddr	%o1, %g0, %g1
 	add		%o1, %o4, %o1
-	EX_LD_FP(LOAD(ldd, %g1 + 0x00, %f0))
-1:	EX_LD_FP(LOAD(ldd, %g1 + 0x08, %f2))
+	EX_LD_FP(LOAD(ldd, %g1 + 0x00, %f0), NG4_retl_o2_plus_o4)
+1:	EX_LD_FP(LOAD(ldd, %g1 + 0x08, %f2), NG4_retl_o2_plus_o4)
 	subcc		%o4, 0x40, %o4
-	EX_LD_FP(LOAD(ldd, %g1 + 0x10, %f4))
-	EX_LD_FP(LOAD(ldd, %g1 + 0x18, %f6))
-	EX_LD_FP(LOAD(ldd, %g1 + 0x20, %f8))
-	EX_LD_FP(LOAD(ldd, %g1 + 0x28, %f10))
-	EX_LD_FP(LOAD(ldd, %g1 + 0x30, %f12))
-	EX_LD_FP(LOAD(ldd, %g1 + 0x38, %f14))
+	EX_LD_FP(LOAD(ldd, %g1 + 0x10, %f4), NG4_retl_o2_plus_o4_plus_64)
+	EX_LD_FP(LOAD(ldd, %g1 + 0x18, %f6), NG4_retl_o2_plus_o4_plus_64)
+	EX_LD_FP(LOAD(ldd, %g1 + 0x20, %f8), NG4_retl_o2_plus_o4_plus_64)
+	EX_LD_FP(LOAD(ldd, %g1 + 0x28, %f10), NG4_retl_o2_plus_o4_plus_64)
+	EX_LD_FP(LOAD(ldd, %g1 + 0x30, %f12), NG4_retl_o2_plus_o4_plus_64)
+	EX_LD_FP(LOAD(ldd, %g1 + 0x38, %f14), NG4_retl_o2_plus_o4_plus_64)
 	faligndata	%f0, %f2, %f16
-	EX_LD_FP(LOAD(ldd, %g1 + 0x40, %f0))
+	EX_LD_FP(LOAD(ldd, %g1 + 0x40, %f0), NG4_retl_o2_plus_o4_plus_64)
 	faligndata	%f2, %f4, %f18
 	add		%g1, 0x40, %g1
 	faligndata	%f4, %f6, %f20
@@ -235,14 +386,14 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	faligndata	%f10, %f12, %f26
 	faligndata	%f12, %f14, %f28
 	faligndata	%f14, %f0, %f30
-	EX_ST_FP(STORE(std, %f16, %o0 + 0x00))
-	EX_ST_FP(STORE(std, %f18, %o0 + 0x08))
-	EX_ST_FP(STORE(std, %f20, %o0 + 0x10))
-	EX_ST_FP(STORE(std, %f22, %o0 + 0x18))
-	EX_ST_FP(STORE(std, %f24, %o0 + 0x20))
-	EX_ST_FP(STORE(std, %f26, %o0 + 0x28))
-	EX_ST_FP(STORE(std, %f28, %o0 + 0x30))
-	EX_ST_FP(STORE(std, %f30, %o0 + 0x38))
+	EX_ST_FP(STORE(std, %f16, %o0 + 0x00), NG4_retl_o2_plus_o4_plus_64)
+	EX_ST_FP(STORE(std, %f18, %o0 + 0x08), NG4_retl_o2_plus_o4_plus_56)
+	EX_ST_FP(STORE(std, %f20, %o0 + 0x10), NG4_retl_o2_plus_o4_plus_48)
+	EX_ST_FP(STORE(std, %f22, %o0 + 0x18), NG4_retl_o2_plus_o4_plus_40)
+	EX_ST_FP(STORE(std, %f24, %o0 + 0x20), NG4_retl_o2_plus_o4_plus_32)
+	EX_ST_FP(STORE(std, %f26, %o0 + 0x28), NG4_retl_o2_plus_o4_plus_24)
+	EX_ST_FP(STORE(std, %f28, %o0 + 0x30), NG4_retl_o2_plus_o4_plus_16)
+	EX_ST_FP(STORE(std, %f30, %o0 + 0x38), NG4_retl_o2_plus_o4_plus_8)
 	add		%o0, 0x40, %o0
 	bne,pt		%icc, 1b
 	 LOAD(prefetch, %g1 + 0x200, #n_reads_strong)
@@ -270,37 +421,38 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	andncc		%o2, 0x20 - 1, %o5
 	be,pn		%icc, 2f
 	 sub		%o2, %o5, %o2
-1:	EX_LD(LOAD(ldx, %o1 + 0x00, %g1))
-	EX_LD(LOAD(ldx, %o1 + 0x08, %g2))
-	EX_LD(LOAD(ldx, %o1 + 0x10, GLOBAL_SPARE))
-	EX_LD(LOAD(ldx, %o1 + 0x18, %o4))
+1:	EX_LD(LOAD(ldx, %o1 + 0x00, %g1), NG4_retl_o2_plus_o5)
+	EX_LD(LOAD(ldx, %o1 + 0x08, %g2), NG4_retl_o2_plus_o5)
+	EX_LD(LOAD(ldx, %o1 + 0x10, GLOBAL_SPARE), NG4_retl_o2_plus_o5)
+	EX_LD(LOAD(ldx, %o1 + 0x18, %o4), NG4_retl_o2_plus_o5)
 	add		%o1, 0x20, %o1
 	subcc		%o5, 0x20, %o5
-	EX_ST(STORE(stx, %g1, %o0 + 0x00))
-	EX_ST(STORE(stx, %g2, %o0 + 0x08))
-	EX_ST(STORE(stx, GLOBAL_SPARE, %o0 + 0x10))
-	EX_ST(STORE(stx, %o4, %o0 + 0x18))
+	EX_ST(STORE(stx, %g1, %o0 + 0x00), NG4_retl_o2_plus_o5_plus_32)
+	EX_ST(STORE(stx, %g2, %o0 + 0x08), NG4_retl_o2_plus_o5_plus_24)
+	EX_ST(STORE(stx, GLOBAL_SPARE, %o0 + 0x10), NG4_retl_o2_plus_o5_plus_24)
+	EX_ST(STORE(stx, %o4, %o0 + 0x18), NG4_retl_o2_plus_o5_plus_8)
 	bne,pt		%icc, 1b
 	 add		%o0, 0x20, %o0
 2:	andcc		%o2, 0x18, %o5
 	be,pt		%icc, 3f
 	 sub		%o2, %o5, %o2
-1:	EX_LD(LOAD(ldx, %o1 + 0x00, %g1))
+
+1:	EX_LD(LOAD(ldx, %o1 + 0x00, %g1), NG4_retl_o2_plus_o5)
 	add		%o1, 0x08, %o1
 	add		%o0, 0x08, %o0
 	subcc		%o5, 0x08, %o5
 	bne,pt		%icc, 1b
-	 EX_ST(STORE(stx, %g1, %o0 - 0x08))
+	 EX_ST(STORE(stx, %g1, %o0 - 0x08), NG4_retl_o2_plus_o5_plus_8)
 3:	brz,pt		%o2, .Lexit
 	 cmp		%o2, 0x04
 	bl,pn		%icc, .Ltiny
 	 nop
-	EX_LD(LOAD(lduw, %o1 + 0x00, %g1))
+	EX_LD(LOAD(lduw, %o1 + 0x00, %g1), NG4_retl_o2)
 	add		%o1, 0x04, %o1
 	add		%o0, 0x04, %o0
 	subcc		%o2, 0x04, %o2
 	bne,pn		%icc, .Ltiny
-	 EX_ST(STORE(stw, %g1, %o0 - 0x04))
+	 EX_ST(STORE(stw, %g1, %o0 - 0x04), NG4_retl_o2_plus_4)
 	ba,a,pt		%icc, .Lexit
 .Lmedium_unaligned:
 	/* First get dest 8 byte aligned.  */
@@ -309,12 +461,12 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	brz,pt		%g1, 2f
 	 sub		%o2, %g1, %o2
 
-1:	EX_LD(LOAD(ldub, %o1 + 0x00, %g2))
+1:	EX_LD(LOAD(ldub, %o1 + 0x00, %g2), NG4_retl_o2_plus_g1)
 	add		%o1, 1, %o1
 	subcc		%g1, 1, %g1
 	add		%o0, 1, %o0
 	bne,pt		%icc, 1b
-	 EX_ST(STORE(stb, %g2, %o0 - 0x01))
+	 EX_ST(STORE(stb, %g2, %o0 - 0x01), NG4_retl_o2_plus_g1_plus_1)
 2:
 	and		%o1, 0x7, %g1
 	brz,pn		%g1, .Lmedium_noprefetch
@@ -322,16 +474,16 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	mov		64, %g2
 	sub		%g2, %g1, %g2
 	andn		%o1, 0x7, %o1
-	EX_LD(LOAD(ldx, %o1 + 0x00, %o4))
+	EX_LD(LOAD(ldx, %o1 + 0x00, %o4), NG4_retl_o2)
 	sllx		%o4, %g1, %o4
 	andn		%o2, 0x08 - 1, %o5
 	sub		%o2, %o5, %o2
-1:	EX_LD(LOAD(ldx, %o1 + 0x08, %g3))
+1:	EX_LD(LOAD(ldx, %o1 + 0x08, %g3), NG4_retl_o2_plus_o5)
 	add		%o1, 0x08, %o1
 	subcc		%o5, 0x08, %o5
 	srlx		%g3, %g2, GLOBAL_SPARE
 	or		GLOBAL_SPARE, %o4, GLOBAL_SPARE
-	EX_ST(STORE(stx, GLOBAL_SPARE, %o0 + 0x00))
+	EX_ST(STORE(stx, GLOBAL_SPARE, %o0 + 0x00), NG4_retl_o2_plus_o5_plus_8)
 	add		%o0, 0x08, %o0
 	bne,pt		%icc, 1b
 	 sllx		%g3, %g1, %o4
@@ -342,17 +494,17 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	ba,pt		%icc, .Lsmall_unaligned
 
 .Ltiny:
-	EX_LD(LOAD(ldub, %o1 + 0x00, %g1))
+	EX_LD(LOAD(ldub, %o1 + 0x00, %g1), NG4_retl_o2)
 	subcc		%o2, 1, %o2
 	be,pn		%icc, .Lexit
-	 EX_ST(STORE(stb, %g1, %o0 + 0x00))
-	EX_LD(LOAD(ldub, %o1 + 0x01, %g1))
+	 EX_ST(STORE(stb, %g1, %o0 + 0x00), NG4_retl_o2_plus_1)
+	EX_LD(LOAD(ldub, %o1 + 0x01, %g1), NG4_retl_o2)
 	subcc		%o2, 1, %o2
 	be,pn		%icc, .Lexit
-	 EX_ST(STORE(stb, %g1, %o0 + 0x01))
-	EX_LD(LOAD(ldub, %o1 + 0x02, %g1))
+	 EX_ST(STORE(stb, %g1, %o0 + 0x01), NG4_retl_o2_plus_1)
+	EX_LD(LOAD(ldub, %o1 + 0x02, %g1), NG4_retl_o2)
 	ba,pt		%icc, .Lexit
-	 EX_ST(STORE(stb, %g1, %o0 + 0x02))
+	 EX_ST(STORE(stb, %g1, %o0 + 0x02), NG4_retl_o2)
 
 .Lsmall:
 	andcc		%g2, 0x3, %g0
@@ -360,22 +512,22 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	 andn		%o2, 0x4 - 1, %o5
 	sub		%o2, %o5, %o2
 1:
-	EX_LD(LOAD(lduw, %o1 + 0x00, %g1))
+	EX_LD(LOAD(lduw, %o1 + 0x00, %g1), NG4_retl_o2_plus_o5)
 	add		%o1, 0x04, %o1
 	subcc		%o5, 0x04, %o5
 	add		%o0, 0x04, %o0
 	bne,pt		%icc, 1b
-	 EX_ST(STORE(stw, %g1, %o0 - 0x04))
+	 EX_ST(STORE(stw, %g1, %o0 - 0x04), NG4_retl_o2_plus_o5_plus_4)
 	brz,pt		%o2, .Lexit
 	 nop
 	ba,a,pt		%icc, .Ltiny
 
 .Lsmall_unaligned:
-1:	EX_LD(LOAD(ldub, %o1 + 0x00, %g1))
+1:	EX_LD(LOAD(ldub, %o1 + 0x00, %g1), NG4_retl_o2)
 	add		%o1, 1, %o1
 	add		%o0, 1, %o0
 	subcc		%o2, 1, %o2
 	bne,pt		%icc, 1b
-	 EX_ST(STORE(stb, %g1, %o0 - 0x01))
+	 EX_ST(STORE(stb, %g1, %o0 - 0x01), NG4_retl_o2_plus_1)
 	ba,a,pt		%icc, .Lexit
 	.size		FUNC_NAME, .-FUNC_NAME
diff --git a/arch/sparc/lib/NGcopy_from_user.S b/arch/sparc/lib/NGcopy_from_user.S
index 5d1e4d1ac21e..9cd42fcbc781 100644
--- a/arch/sparc/lib/NGcopy_from_user.S
+++ b/arch/sparc/lib/NGcopy_from_user.S
@@ -3,11 +3,11 @@
  * Copyright (C) 2006, 2007 David S. Miller (davem@xxxxxxxxxxxxx)
  */
 
-#define EX_LD(x)		\
+#define EX_LD(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __ret_one_asi;\
+	.word 98b, y;		\
 	.text;			\
 	.align 4;
 
diff --git a/arch/sparc/lib/NGcopy_to_user.S b/arch/sparc/lib/NGcopy_to_user.S
index ff630dcb273c..5c358afd464e 100644
--- a/arch/sparc/lib/NGcopy_to_user.S
+++ b/arch/sparc/lib/NGcopy_to_user.S
@@ -3,11 +3,11 @@
  * Copyright (C) 2006, 2007 David S. Miller (davem@xxxxxxxxxxxxx)
  */
 
-#define EX_ST(x)		\
+#define EX_ST(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __ret_one_asi;\
+	.word 98b, y;		\
 	.text;			\
 	.align 4;
 
diff --git a/arch/sparc/lib/NGmemcpy.S b/arch/sparc/lib/NGmemcpy.S
index 96a14caf6966..d88c4ed50a00 100644
--- a/arch/sparc/lib/NGmemcpy.S
+++ b/arch/sparc/lib/NGmemcpy.S
@@ -4,6 +4,7 @@
  */
 
 #ifdef __KERNEL__
+#include <linux/linkage.h>
 #include <asm/asi.h>
 #include <asm/thread_info.h>
 #define GLOBAL_SPARE	%g7
@@ -27,15 +28,11 @@
 #endif
 
 #ifndef EX_LD
-#define EX_LD(x)	x
+#define EX_LD(x,y)	x
 #endif
 
 #ifndef EX_ST
-#define EX_ST(x)	x
-#endif
-
-#ifndef EX_RETVAL
-#define EX_RETVAL(x)	x
+#define EX_ST(x,y)	x
 #endif
 
 #ifndef LOAD
@@ -79,6 +76,92 @@
 	.register	%g3,#scratch
 
 	.text
+#ifndef EX_RETVAL
+#define EX_RETVAL(x)	x
+__restore_asi:
+	ret
+	wr	%g0, ASI_AIUS, %asi
+	 restore
+ENTRY(NG_ret_i2_plus_i4_plus_1)
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, %i5, %i0
+ENDPROC(NG_ret_i2_plus_i4_plus_1)
+ENTRY(NG_ret_i2_plus_g1)
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1)
+ENTRY(NG_ret_i2_plus_g1_minus_8)
+	sub	%g1, 8, %g1
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_minus_8)
+ENTRY(NG_ret_i2_plus_g1_minus_16)
+	sub	%g1, 16, %g1
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_minus_16)
+ENTRY(NG_ret_i2_plus_g1_minus_24)
+	sub	%g1, 24, %g1
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_minus_24)
+ENTRY(NG_ret_i2_plus_g1_minus_32)
+	sub	%g1, 32, %g1
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_minus_32)
+ENTRY(NG_ret_i2_plus_g1_minus_40)
+	sub	%g1, 40, %g1
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_minus_40)
+ENTRY(NG_ret_i2_plus_g1_minus_48)
+	sub	%g1, 48, %g1
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_minus_48)
+ENTRY(NG_ret_i2_plus_g1_minus_56)
+	sub	%g1, 56, %g1
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_minus_56)
+ENTRY(NG_ret_i2_plus_i4)
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, %i4, %i0
+ENDPROC(NG_ret_i2_plus_i4)
+ENTRY(NG_ret_i2_plus_i4_minus_8)
+	sub	%i4, 8, %i4
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, %i4, %i0
+ENDPROC(NG_ret_i2_plus_i4_minus_8)
+ENTRY(NG_ret_i2_plus_8)
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, 8, %i0
+ENDPROC(NG_ret_i2_plus_8)
+ENTRY(NG_ret_i2_plus_4)
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, 4, %i0
+ENDPROC(NG_ret_i2_plus_4)
+ENTRY(NG_ret_i2_plus_1)
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, 1, %i0
+ENDPROC(NG_ret_i2_plus_1)
+ENTRY(NG_ret_i2_plus_g1_plus_1)
+	add	%g1, 1, %g1
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_plus_1)
+ENTRY(NG_ret_i2)
+	ba,pt	%xcc, __restore_asi
+	 mov	%i2, %i0
+ENDPROC(NG_ret_i2)
+ENTRY(NG_ret_i2_and_7_plus_i4)
+	and	%i2, 7, %i2
+	ba,pt	%xcc, __restore_asi
+	 add	%i2, %i4, %i0
+ENDPROC(NG_ret_i2_and_7_plus_i4)
+#endif
+
 	.align		64
 
 	.globl	FUNC_NAME
@@ -126,8 +209,8 @@ FUNC_NAME:	/* %i0=dst, %i1=src, %i2=len */
 	sub		%g0, %i4, %i4	! bytes to align dst
 	sub		%i2, %i4, %i2
 1:	subcc		%i4, 1, %i4
-	EX_LD(LOAD(ldub, %i1, %g1))
-	EX_ST(STORE(stb, %g1, %o0))
+	EX_LD(LOAD(ldub, %i1, %g1), NG_ret_i2_plus_i4_plus_1)
+	EX_ST(STORE(stb, %g1, %o0), NG_ret_i2_plus_i4_plus_1)
 	add		%i1, 1, %i1
 	bne,pt		%XCC, 1b
 	add		%o0, 1, %o0
@@ -160,7 +243,7 @@ FUNC_NAME:	/* %i0=dst, %i1=src, %i2=len */
 	and		%i4, 0x7, GLOBAL_SPARE
 	sll		GLOBAL_SPARE, 3, GLOBAL_SPARE
 	mov		64, %i5
-	EX_LD(LOAD_TWIN(%i1, %g2, %g3))
+	EX_LD(LOAD_TWIN(%i1, %g2, %g3), NG_ret_i2_plus_g1)
 	sub		%i5, GLOBAL_SPARE, %i5
 	mov		16, %o4
 	mov		32, %o5
@@ -178,31 +261,31 @@ FUNC_NAME:	/* %i0=dst, %i1=src, %i2=len */
 	srlx		WORD3, PRE_SHIFT, TMP; \
 	or		WORD2, TMP, WORD2;
 
-8:	EX_LD(LOAD_TWIN(%i1 + %o4, %o2, %o3))
+8:	EX_LD(LOAD_TWIN(%i1 + %o4, %o2, %o3), NG_ret_i2_plus_g1)
 	MIX_THREE_WORDS(%g2, %g3, %o2, %i5, GLOBAL_SPARE, %o1)
 	LOAD(prefetch, %i1 + %i3, #one_read)
 
-	EX_ST(STORE_INIT(%g2, %o0 + 0x00))
-	EX_ST(STORE_INIT(%g3, %o0 + 0x08))
+	EX_ST(STORE_INIT(%g2, %o0 + 0x00), NG_ret_i2_plus_g1)
+	EX_ST(STORE_INIT(%g3, %o0 + 0x08), NG_ret_i2_plus_g1_minus_8)
 
-	EX_LD(LOAD_TWIN(%i1 + %o5, %g2, %g3))
+	EX_LD(LOAD_TWIN(%i1 + %o5, %g2, %g3), NG_ret_i2_plus_g1_minus_16)
 	MIX_THREE_WORDS(%o2, %o3, %g2, %i5, GLOBAL_SPARE, %o1)
 
-	EX_ST(STORE_INIT(%o2, %o0 + 0x10))
-	EX_ST(STORE_INIT(%o3, %o0 + 0x18))
+	EX_ST(STORE_INIT(%o2, %o0 + 0x10), NG_ret_i2_plus_g1_minus_16)
+	EX_ST(STORE_INIT(%o3, %o0 + 0x18), NG_ret_i2_plus_g1_minus_24)
 
-	EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3))
+	EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3), NG_ret_i2_plus_g1_minus_32)
 	MIX_THREE_WORDS(%g2, %g3, %o2, %i5, GLOBAL_SPARE, %o1)
 
-	EX_ST(STORE_INIT(%g2, %o0 + 0x20))
-	EX_ST(STORE_INIT(%g3, %o0 + 0x28))
+	EX_ST(STORE_INIT(%g2, %o0 + 0x20), NG_ret_i2_plus_g1_minus_32)
+	EX_ST(STORE_INIT(%g3, %o0 + 0x28), NG_ret_i2_plus_g1_minus_40)
 
-	EX_LD(LOAD_TWIN(%i1 + %i3, %g2, %g3))
+	EX_LD(LOAD_TWIN(%i1 + %i3, %g2, %g3), NG_ret_i2_plus_g1_minus_48)
 	add		%i1, 64, %i1
 	MIX_THREE_WORDS(%o2, %o3, %g2, %i5, GLOBAL_SPARE, %o1)
 
-	EX_ST(STORE_INIT(%o2, %o0 + 0x30))
-	EX_ST(STORE_INIT(%o3, %o0 + 0x38))
+	EX_ST(STORE_INIT(%o2, %o0 + 0x30), NG_ret_i2_plus_g1_minus_48)
+	EX_ST(STORE_INIT(%o3, %o0 + 0x38), NG_ret_i2_plus_g1_minus_56)
 
 	subcc		%g1, 64, %g1
 	bne,pt		%XCC, 8b
@@ -211,31 +294,31 @@ FUNC_NAME:	/* %i0=dst, %i1=src, %i2=len */
 	ba,pt		%XCC, 60f
 	 add		%i1, %i4, %i1
 
-9:	EX_LD(LOAD_TWIN(%i1 + %o4, %o2, %o3))
+9:	EX_LD(LOAD_TWIN(%i1 + %o4, %o2, %o3), NG_ret_i2_plus_g1)
 	MIX_THREE_WORDS(%g3, %o2, %o3, %i5, GLOBAL_SPARE, %o1)
 	LOAD(prefetch, %i1 + %i3, #one_read)
 
-	EX_ST(STORE_INIT(%g3, %o0 + 0x00))
-	EX_ST(STORE_INIT(%o2, %o0 + 0x08))
+	EX_ST(STORE_INIT(%g3, %o0 + 0x00), NG_ret_i2_plus_g1)
+	EX_ST(STORE_INIT(%o2, %o0 + 0x08), NG_ret_i2_plus_g1_minus_8)
 
-	EX_LD(LOAD_TWIN(%i1 + %o5, %g2, %g3))
+	EX_LD(LOAD_TWIN(%i1 + %o5, %g2, %g3), NG_ret_i2_plus_g1_minus_16)
 	MIX_THREE_WORDS(%o3, %g2, %g3, %i5, GLOBAL_SPARE, %o1)
 
-	EX_ST(STORE_INIT(%o3, %o0 + 0x10))
-	EX_ST(STORE_INIT(%g2, %o0 + 0x18))
+	EX_ST(STORE_INIT(%o3, %o0 + 0x10), NG_ret_i2_plus_g1_minus_16)
+	EX_ST(STORE_INIT(%g2, %o0 + 0x18), NG_ret_i2_plus_g1_minus_24)
 
-	EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3))
+	EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3), NG_ret_i2_plus_g1_minus_32)
 	MIX_THREE_WORDS(%g3, %o2, %o3, %i5, GLOBAL_SPARE, %o1)
 
-	EX_ST(STORE_INIT(%g3, %o0 + 0x20))
-	EX_ST(STORE_INIT(%o2, %o0 + 0x28))
+	EX_ST(STORE_INIT(%g3, %o0 + 0x20), NG_ret_i2_plus_g1_minus_32)
+	EX_ST(STORE_INIT(%o2, %o0 + 0x28), NG_ret_i2_plus_g1_minus_40)
 
-	EX_LD(LOAD_TWIN(%i1 + %i3, %g2, %g3))
+	EX_LD(LOAD_TWIN(%i1 + %i3, %g2, %g3), NG_ret_i2_plus_g1_minus_48)
 	add		%i1, 64, %i1
 	MIX_THREE_WORDS(%o3, %g2, %g3, %i5, GLOBAL_SPARE, %o1)
 
-	EX_ST(STORE_INIT(%o3, %o0 + 0x30))
-	EX_ST(STORE_INIT(%g2, %o0 + 0x38))
+	EX_ST(STORE_INIT(%o3, %o0 + 0x30), NG_ret_i2_plus_g1_minus_48)
+	EX_ST(STORE_INIT(%g2, %o0 + 0x38), NG_ret_i2_plus_g1_minus_56)
 
 	subcc		%g1, 64, %g1
 	bne,pt		%XCC, 9b
@@ -249,25 +332,25 @@ FUNC_NAME:	/* %i0=dst, %i1=src, %i2=len */
 	 * one twin load ahead, then add 8 back into source when
 	 * we finish the loop.
 	 */
-	EX_LD(LOAD_TWIN(%i1, %o4, %o5))
+	EX_LD(LOAD_TWIN(%i1, %o4, %o5), NG_ret_i2_plus_g1)
 	mov	16, %o7
 	mov	32, %g2
 	mov	48, %g3
 	mov	64, %o1
-1:	EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3))
+1:	EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3), NG_ret_i2_plus_g1)
 	LOAD(prefetch, %i1 + %o1, #one_read)
-	EX_ST(STORE_INIT(%o5, %o0 + 0x00))	! initializes cache line
-	EX_ST(STORE_INIT(%o2, %o0 + 0x08))
-	EX_LD(LOAD_TWIN(%i1 + %g2, %o4, %o5))
-	EX_ST(STORE_INIT(%o3, %o0 + 0x10))
-	EX_ST(STORE_INIT(%o4, %o0 + 0x18))
-	EX_LD(LOAD_TWIN(%i1 + %g3, %o2, %o3))
-	EX_ST(STORE_INIT(%o5, %o0 + 0x20))
-	EX_ST(STORE_INIT(%o2, %o0 + 0x28))
-	EX_LD(LOAD_TWIN(%i1 + %o1, %o4, %o5))
+	EX_ST(STORE_INIT(%o5, %o0 + 0x00), NG_ret_i2_plus_g1)	! initializes cache line
+	EX_ST(STORE_INIT(%o2, %o0 + 0x08), NG_ret_i2_plus_g1_minus_8)
+	EX_LD(LOAD_TWIN(%i1 + %g2, %o4, %o5), NG_ret_i2_plus_g1_minus_16)
+	EX_ST(STORE_INIT(%o3, %o0 + 0x10), NG_ret_i2_plus_g1_minus_16)
+	EX_ST(STORE_INIT(%o4, %o0 + 0x18), NG_ret_i2_plus_g1_minus_24)
+	EX_LD(LOAD_TWIN(%i1 + %g3, %o2, %o3), NG_ret_i2_plus_g1_minus_32)
+	EX_ST(STORE_INIT(%o5, %o0 + 0x20), NG_ret_i2_plus_g1_minus_32)
+	EX_ST(STORE_INIT(%o2, %o0 + 0x28), NG_ret_i2_plus_g1_minus_40)
+	EX_LD(LOAD_TWIN(%i1 + %o1, %o4, %o5), NG_ret_i2_plus_g1_minus_48)
 	add		%i1, 64, %i1
-	EX_ST(STORE_INIT(%o3, %o0 + 0x30))
-	EX_ST(STORE_INIT(%o4, %o0 + 0x38))
+	EX_ST(STORE_INIT(%o3, %o0 + 0x30), NG_ret_i2_plus_g1_minus_48)
+	EX_ST(STORE_INIT(%o4, %o0 + 0x38), NG_ret_i2_plus_g1_minus_56)
 	subcc		%g1, 64, %g1
 	bne,pt		%XCC, 1b
 	 add		%o0, 64, %o0
@@ -282,20 +365,20 @@ FUNC_NAME:	/* %i0=dst, %i1=src, %i2=len */
 	mov	32, %g2
 	mov	48, %g3
 	mov	64, %o1
-1:	EX_LD(LOAD_TWIN(%i1 + %g0, %o4, %o5))
-	EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3))
+1:	EX_LD(LOAD_TWIN(%i1 + %g0, %o4, %o5), NG_ret_i2_plus_g1)
+	EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3), NG_ret_i2_plus_g1)
 	LOAD(prefetch, %i1 + %o1, #one_read)
-	EX_ST(STORE_INIT(%o4, %o0 + 0x00))	! initializes cache line
-	EX_ST(STORE_INIT(%o5, %o0 + 0x08))
-	EX_LD(LOAD_TWIN(%i1 + %g2, %o4, %o5))
-	EX_ST(STORE_INIT(%o2, %o0 + 0x10))
-	EX_ST(STORE_INIT(%o3, %o0 + 0x18))
-	EX_LD(LOAD_TWIN(%i1 + %g3, %o2, %o3))
+	EX_ST(STORE_INIT(%o4, %o0 + 0x00), NG_ret_i2_plus_g1)	! initializes cache line
+	EX_ST(STORE_INIT(%o5, %o0 + 0x08), NG_ret_i2_plus_g1_minus_8)
+	EX_LD(LOAD_TWIN(%i1 + %g2, %o4, %o5), NG_ret_i2_plus_g1_minus_16)
+	EX_ST(STORE_INIT(%o2, %o0 + 0x10), NG_ret_i2_plus_g1_minus_16)
+	EX_ST(STORE_INIT(%o3, %o0 + 0x18), NG_ret_i2_plus_g1_minus_24)
+	EX_LD(LOAD_TWIN(%i1 + %g3, %o2, %o3), NG_ret_i2_plus_g1_minus_32)
 	add	%i1, 64, %i1
-	EX_ST(STORE_INIT(%o4, %o0 + 0x20))
-	EX_ST(STORE_INIT(%o5, %o0 + 0x28))
-	EX_ST(STORE_INIT(%o2, %o0 + 0x30))
-	EX_ST(STORE_INIT(%o3, %o0 + 0x38))
+	EX_ST(STORE_INIT(%o4, %o0 + 0x20), NG_ret_i2_plus_g1_minus_32)
+	EX_ST(STORE_INIT(%o5, %o0 + 0x28), NG_ret_i2_plus_g1_minus_40)
+	EX_ST(STORE_INIT(%o2, %o0 + 0x30), NG_ret_i2_plus_g1_minus_48)
+	EX_ST(STORE_INIT(%o3, %o0 + 0x38), NG_ret_i2_plus_g1_minus_56)
 	subcc	%g1, 64, %g1
 	bne,pt	%XCC, 1b
 	 add	%o0, 64, %o0
@@ -321,28 +404,28 @@ FUNC_NAME:	/* %i0=dst, %i1=src, %i2=len */
 	andn		%i2, 0xf, %i4
 	and		%i2, 0xf, %i2
 1:	subcc		%i4, 0x10, %i4
-	EX_LD(LOAD(ldx, %i1, %o4))
+	EX_LD(LOAD(ldx, %i1, %o4), NG_ret_i2_plus_i4)
 	add		%i1, 0x08, %i1
-	EX_LD(LOAD(ldx, %i1, %g1))
+	EX_LD(LOAD(ldx, %i1, %g1), NG_ret_i2_plus_i4)
 	sub		%i1, 0x08, %i1
-	EX_ST(STORE(stx, %o4, %i1 + %i3))
+	EX_ST(STORE(stx, %o4, %i1 + %i3), NG_ret_i2_plus_i4)
 	add		%i1, 0x8, %i1
-	EX_ST(STORE(stx, %g1, %i1 + %i3))
+	EX_ST(STORE(stx, %g1, %i1 + %i3), NG_ret_i2_plus_i4_minus_8)
 	bgu,pt		%XCC, 1b
 	 add		%i1, 0x8, %i1
 73:	andcc		%i2, 0x8, %g0
 	be,pt		%XCC, 1f
 	 nop
 	sub		%i2, 0x8, %i2
-	EX_LD(LOAD(ldx, %i1, %o4))
-	EX_ST(STORE(stx, %o4, %i1 + %i3))
+	EX_LD(LOAD(ldx, %i1, %o4), NG_ret_i2_plus_8)
+	EX_ST(STORE(stx, %o4, %i1 + %i3), NG_ret_i2_plus_8)
 	add		%i1, 0x8, %i1
 1:	andcc		%i2, 0x4, %g0
 	be,pt		%XCC, 1f
 	 nop
 	sub		%i2, 0x4, %i2
-	EX_LD(LOAD(lduw, %i1, %i5))
-	EX_ST(STORE(stw, %i5, %i1 + %i3))
+	EX_LD(LOAD(lduw, %i1, %i5), NG_ret_i2_plus_4)
+	EX_ST(STORE(stw, %i5, %i1 + %i3), NG_ret_i2_plus_4)
 	add		%i1, 0x4, %i1
 1:	cmp		%i2, 0
 	be,pt		%XCC, 85f
@@ -358,8 +441,8 @@ FUNC_NAME:	/* %i0=dst, %i1=src, %i2=len */
 	sub		%i2, %g1, %i2
 
 1:	subcc		%g1, 1, %g1
-	EX_LD(LOAD(ldub, %i1, %i5))
-	EX_ST(STORE(stb, %i5, %i1 + %i3))
+	EX_LD(LOAD(ldub, %i1, %i5), NG_ret_i2_plus_g1_plus_1)
+	EX_ST(STORE(stb, %i5, %i1 + %i3), NG_ret_i2_plus_g1_plus_1)
 	bgu,pt		%icc, 1b
 	 add		%i1, 1, %i1
 
@@ -375,16 +458,16 @@ FUNC_NAME:	/* %i0=dst, %i1=src, %i2=len */
 
 8:	mov		64, %i3
 	andn		%i1, 0x7, %i1
-	EX_LD(LOAD(ldx, %i1, %g2))
+	EX_LD(LOAD(ldx, %i1, %g2), NG_ret_i2)
 	sub		%i3, %g1, %i3
 	andn		%i2, 0x7, %i4
 	sllx		%g2, %g1, %g2
 1:	add		%i1, 0x8, %i1
-	EX_LD(LOAD(ldx, %i1, %g3))
+	EX_LD(LOAD(ldx, %i1, %g3), NG_ret_i2_and_7_plus_i4)
 	subcc		%i4, 0x8, %i4
 	srlx		%g3, %i3, %i5
 	or		%i5, %g2, %i5
-	EX_ST(STORE(stx, %i5, %o0))
+	EX_ST(STORE(stx, %i5, %o0), NG_ret_i2_and_7_plus_i4)
 	add		%o0, 0x8, %o0
 	bgu,pt		%icc, 1b
 	 sllx		%g3, %g1, %g2
@@ -404,8 +487,8 @@ FUNC_NAME:	/* %i0=dst, %i1=src, %i2=len */
 
 1:
 	subcc		%i2, 4, %i2
-	EX_LD(LOAD(lduw, %i1, %g1))
-	EX_ST(STORE(stw, %g1, %i1 + %i3))
+	EX_LD(LOAD(lduw, %i1, %g1), NG_ret_i2_plus_4)
+	EX_ST(STORE(stw, %g1, %i1 + %i3), NG_ret_i2_plus_4)
 	bgu,pt		%XCC, 1b
 	 add		%i1, 4, %i1
 
@@ -415,8 +498,8 @@ FUNC_NAME:	/* %i0=dst, %i1=src, %i2=len */
 	.align		32
 90:
 	subcc		%i2, 1, %i2
-	EX_LD(LOAD(ldub, %i1, %g1))
-	EX_ST(STORE(stb, %g1, %i1 + %i3))
+	EX_LD(LOAD(ldub, %i1, %g1), NG_ret_i2_plus_1)
+	EX_ST(STORE(stb, %g1, %i1 + %i3), NG_ret_i2_plus_1)
 	bgu,pt		%XCC, 90b
 	 add		%i1, 1, %i1
 	ret
diff --git a/arch/sparc/lib/U1copy_from_user.S b/arch/sparc/lib/U1copy_from_user.S
index ecc5692fa2b4..bb6ff73229e3 100644
--- a/arch/sparc/lib/U1copy_from_user.S
+++ b/arch/sparc/lib/U1copy_from_user.S
@@ -3,19 +3,19 @@
  * Copyright (C) 1999, 2000, 2004 David S. Miller (davem@xxxxxxxxxx)
  */
 
-#define EX_LD(x)		\
+#define EX_LD(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one;	\
+	.word 98b, y;		\
 	.text;			\
 	.align 4;
 
-#define EX_LD_FP(x)		\
+#define EX_LD_FP(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one_fp;\
+	.word 98b, y;		\
 	.text;			\
 	.align 4;
 
diff --git a/arch/sparc/lib/U1copy_to_user.S b/arch/sparc/lib/U1copy_to_user.S
index 9eea392e44d4..ed92ce739558 100644
--- a/arch/sparc/lib/U1copy_to_user.S
+++ b/arch/sparc/lib/U1copy_to_user.S
@@ -3,19 +3,19 @@
  * Copyright (C) 1999, 2000, 2004 David S. Miller (davem@xxxxxxxxxx)
  */
 
-#define EX_ST(x)		\
+#define EX_ST(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one;	\
+	.word 98b, y;		\
 	.text;			\
 	.align 4;
 
-#define EX_ST_FP(x)		\
+#define EX_ST_FP(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one_fp;\
+	.word 98b, y;		\
 	.text;			\
 	.align 4;
 
diff --git a/arch/sparc/lib/U1memcpy.S b/arch/sparc/lib/U1memcpy.S
index 3e6209ebb7d7..f30d2ab2c371 100644
--- a/arch/sparc/lib/U1memcpy.S
+++ b/arch/sparc/lib/U1memcpy.S
@@ -5,6 +5,7 @@
  */
 
 #ifdef __KERNEL__
+#include <linux/linkage.h>
 #include <asm/visasm.h>
 #include <asm/asi.h>
 #define GLOBAL_SPARE	g7
@@ -23,21 +24,17 @@
 #endif
 
 #ifndef EX_LD
-#define EX_LD(x)	x
+#define EX_LD(x,y)	x
 #endif
 #ifndef EX_LD_FP
-#define EX_LD_FP(x)	x
+#define EX_LD_FP(x,y)	x
 #endif
 
 #ifndef EX_ST
-#define EX_ST(x)	x
+#define EX_ST(x,y)	x
 #endif
 #ifndef EX_ST_FP
-#define EX_ST_FP(x)	x
-#endif
-
-#ifndef EX_RETVAL
-#define EX_RETVAL(x)	x
+#define EX_ST_FP(x,y)	x
 #endif
 
 #ifndef LOAD
@@ -78,53 +75,169 @@
 	faligndata		%f7, %f8, %f60;			\
 	faligndata		%f8, %f9, %f62;
 
-#define MAIN_LOOP_CHUNK(src, dest, fdest, fsrc, len, jmptgt)	\
-	EX_LD_FP(LOAD_BLK(%src, %fdest));				\
-	EX_ST_FP(STORE_BLK(%fsrc, %dest));				\
-	add			%src, 0x40, %src;		\
-	subcc			%len, 0x40, %len;		\
-	be,pn			%xcc, jmptgt;			\
-	 add			%dest, 0x40, %dest;		\
-
-#define LOOP_CHUNK1(src, dest, len, branch_dest)		\
-	MAIN_LOOP_CHUNK(src, dest, f0,  f48, len, branch_dest)
-#define LOOP_CHUNK2(src, dest, len, branch_dest)		\
-	MAIN_LOOP_CHUNK(src, dest, f16, f48, len, branch_dest)
-#define LOOP_CHUNK3(src, dest, len, branch_dest)		\
-	MAIN_LOOP_CHUNK(src, dest, f32, f48, len, branch_dest)
+#define MAIN_LOOP_CHUNK(src, dest, fdest, fsrc, jmptgt)			\
+	EX_LD_FP(LOAD_BLK(%src, %fdest), U1_gs_80_fp);			\
+	EX_ST_FP(STORE_BLK(%fsrc, %dest), U1_gs_80_fp);			\
+	add			%src, 0x40, %src;			\
+	subcc			%GLOBAL_SPARE, 0x40, %GLOBAL_SPARE;	\
+	be,pn			%xcc, jmptgt;				\
+	 add			%dest, 0x40, %dest;			\
+
+#define LOOP_CHUNK1(src, dest, branch_dest)		\
+	MAIN_LOOP_CHUNK(src, dest, f0,  f48, branch_dest)
+#define LOOP_CHUNK2(src, dest, branch_dest)		\
+	MAIN_LOOP_CHUNK(src, dest, f16, f48, branch_dest)
+#define LOOP_CHUNK3(src, dest, branch_dest)		\
+	MAIN_LOOP_CHUNK(src, dest, f32, f48, branch_dest)
 
 #define DO_SYNC			membar	#Sync;
 #define STORE_SYNC(dest, fsrc)				\
-	EX_ST_FP(STORE_BLK(%fsrc, %dest));			\
+	EX_ST_FP(STORE_BLK(%fsrc, %dest), U1_gs_80_fp);	\
 	add			%dest, 0x40, %dest;	\
 	DO_SYNC
 
 #define STORE_JUMP(dest, fsrc, target)			\
-	EX_ST_FP(STORE_BLK(%fsrc, %dest));			\
+	EX_ST_FP(STORE_BLK(%fsrc, %dest), U1_gs_40_fp);	\
 	add			%dest, 0x40, %dest;	\
 	ba,pt			%xcc, target;		\
 	 nop;
 
-#define FINISH_VISCHUNK(dest, f0, f1, left)	\
-	subcc			%left, 8, %left;\
-	bl,pn			%xcc, 95f;	\
-	 faligndata		%f0, %f1, %f48;	\
-	EX_ST_FP(STORE(std, %f48, %dest));		\
+#define FINISH_VISCHUNK(dest, f0, f1)			\
+	subcc			%g3, 8, %g3;		\
+	bl,pn			%xcc, 95f;		\
+	 faligndata		%f0, %f1, %f48;		\
+	EX_ST_FP(STORE(std, %f48, %dest), U1_g3_8_fp);	\
 	add			%dest, 8, %dest;
 
-#define UNEVEN_VISCHUNK_LAST(dest, f0, f1, left)	\
-	subcc			%left, 8, %left;	\
-	bl,pn			%xcc, 95f;		\
+#define UNEVEN_VISCHUNK_LAST(dest, f0, f1)	\
+	subcc			%g3, 8, %g3;	\
+	bl,pn			%xcc, 95f;	\
 	 fsrc2			%f0, %f1;
 
-#define UNEVEN_VISCHUNK(dest, f0, f1, left)		\
-	UNEVEN_VISCHUNK_LAST(dest, f0, f1, left)	\
+#define UNEVEN_VISCHUNK(dest, f0, f1)		\
+	UNEVEN_VISCHUNK_LAST(dest, f0, f1)	\
 	ba,a,pt			%xcc, 93f;
 
 	.register	%g2,#scratch
 	.register	%g3,#scratch
 
 	.text
+#ifndef EX_RETVAL
+#define EX_RETVAL(x)	x
+ENTRY(U1_g1_1_fp)
+	VISExitHalf
+	add		%g1, 1, %g1
+	add		%g1, %g2, %g1
+	retl
+	 add		%g1, %o2, %o0
+ENDPROC(U1_g1_1_fp)
+ENTRY(U1_g2_0_fp)
+	VISExitHalf
+	retl
+	 add		%g2, %o2, %o0
+ENDPROC(U1_g2_0_fp)
+ENTRY(U1_g2_8_fp)
+	VISExitHalf
+	add		%g2, 8, %g2
+	retl
+	 add		%g2, %o2, %o0
+ENDPROC(U1_g2_8_fp)
+ENTRY(U1_gs_0_fp)
+	VISExitHalf
+	add		%GLOBAL_SPARE, %g3, %o0
+	retl
+	 add		%o0, %o2, %o0
+ENDPROC(U1_gs_0_fp)
+ENTRY(U1_gs_80_fp)
+	VISExitHalf
+	add		%GLOBAL_SPARE, 0x80, %GLOBAL_SPARE
+	add		%GLOBAL_SPARE, %g3, %o0
+	retl
+	 add		%o0, %o2, %o0
+ENDPROC(U1_gs_80_fp)
+ENTRY(U1_gs_40_fp)
+	VISExitHalf
+	add		%GLOBAL_SPARE, 0x40, %GLOBAL_SPARE
+	add		%GLOBAL_SPARE, %g3, %o0
+	retl
+	 add		%o0, %o2, %o0
+ENDPROC(U1_gs_40_fp)
+ENTRY(U1_g3_0_fp)
+	VISExitHalf
+	retl
+	 add		%g3, %o2, %o0
+ENDPROC(U1_g3_0_fp)
+ENTRY(U1_g3_8_fp)
+	VISExitHalf
+	add		%g3, 8, %g3
+	retl
+	 add		%g3, %o2, %o0
+ENDPROC(U1_g3_8_fp)
+ENTRY(U1_o2_0_fp)
+	VISExitHalf
+	retl
+	 mov		%o2, %o0
+ENDPROC(U1_o2_0_fp)
+ENTRY(U1_o2_1_fp)
+	VISExitHalf
+	retl
+	 add		%o2, 1, %o0
+ENDPROC(U1_o2_1_fp)
+ENTRY(U1_gs_0)
+	VISExitHalf
+	retl
+	 add		%GLOBAL_SPARE, %o2, %o0
+ENDPROC(U1_gs_0)
+ENTRY(U1_gs_8)
+	VISExitHalf
+	add		%GLOBAL_SPARE, %o2, %GLOBAL_SPARE
+	retl
+	 add		%GLOBAL_SPARE, 0x8, %o0
+ENDPROC(U1_gs_8)
+ENTRY(U1_gs_10)
+	VISExitHalf
+	add		%GLOBAL_SPARE, %o2, %GLOBAL_SPARE
+	retl
+	 add		%GLOBAL_SPARE, 0x10, %o0
+ENDPROC(U1_gs_10)
+ENTRY(U1_o2_0)
+	retl
+	 mov		%o2, %o0
+ENDPROC(U1_o2_0)
+ENTRY(U1_o2_8)
+	retl
+	 add		%o2, 8, %o0
+ENDPROC(U1_o2_8)
+ENTRY(U1_o2_4)
+	retl
+	 add		%o2, 4, %o0
+ENDPROC(U1_o2_4)
+ENTRY(U1_o2_1)
+	retl
+	 add		%o2, 1, %o0
+ENDPROC(U1_o2_1)
+ENTRY(U1_g1_0)
+	retl
+	 add		%g1, %o2, %o0
+ENDPROC(U1_g1_0)
+ENTRY(U1_g1_1)
+	add		%g1, 1, %g1
+	retl
+	 add		%g1, %o2, %o0
+ENDPROC(U1_g1_1)
+ENTRY(U1_gs_0_o2_adj)
+	and		%o2, 7, %o2
+	retl
+	 add		%GLOBAL_SPARE, %o2, %o0
+ENDPROC(U1_gs_0_o2_adj)
+ENTRY(U1_gs_8_o2_adj)
+	and		%o2, 7, %o2
+	add		%GLOBAL_SPARE, 8, %GLOBAL_SPARE
+	retl
+	 add		%GLOBAL_SPARE, %o2, %o0
+ENDPROC(U1_gs_8_o2_adj)
+#endif
+
 	.align		64
 
 	.globl		FUNC_NAME
@@ -166,8 +279,8 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	 and		%g2, 0x38, %g2
 
 1:	subcc		%g1, 0x1, %g1
-	EX_LD_FP(LOAD(ldub, %o1 + 0x00, %o3))
-	EX_ST_FP(STORE(stb, %o3, %o1 + %GLOBAL_SPARE))
+	EX_LD_FP(LOAD(ldub, %o1 + 0x00, %o3), U1_g1_1_fp)
+	EX_ST_FP(STORE(stb, %o3, %o1 + %GLOBAL_SPARE), U1_g1_1_fp)
 	bgu,pt		%XCC, 1b
 	 add		%o1, 0x1, %o1
 
@@ -178,20 +291,20 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	be,pt		%icc, 3f
 	 alignaddr	%o1, %g0, %o1
 
-	EX_LD_FP(LOAD(ldd, %o1, %f4))
-1:	EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f6))
+	EX_LD_FP(LOAD(ldd, %o1, %f4), U1_g2_0_fp)
+1:	EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f6), U1_g2_0_fp)
 	add		%o1, 0x8, %o1
 	subcc		%g2, 0x8, %g2
 	faligndata	%f4, %f6, %f0
-	EX_ST_FP(STORE(std, %f0, %o0))
+	EX_ST_FP(STORE(std, %f0, %o0), U1_g2_8_fp)
 	be,pn		%icc, 3f
 	 add		%o0, 0x8, %o0
 
-	EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f4))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f4), U1_g2_0_fp)
 	add		%o1, 0x8, %o1
 	subcc		%g2, 0x8, %g2
 	faligndata	%f6, %f4, %f0
-	EX_ST_FP(STORE(std, %f0, %o0))
+	EX_ST_FP(STORE(std, %f0, %o0), U1_g2_8_fp)
 	bne,pt		%icc, 1b
 	 add		%o0, 0x8, %o0
 
@@ -214,13 +327,13 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	add		%g1, %GLOBAL_SPARE, %g1
 	subcc		%o2, %g3, %o2
 
-	EX_LD_FP(LOAD_BLK(%o1, %f0))
+	EX_LD_FP(LOAD_BLK(%o1, %f0), U1_gs_0_fp)
 	add		%o1, 0x40, %o1
 	add		%g1, %g3, %g1
-	EX_LD_FP(LOAD_BLK(%o1, %f16))
+	EX_LD_FP(LOAD_BLK(%o1, %f16), U1_gs_0_fp)
 	add		%o1, 0x40, %o1
 	sub		%GLOBAL_SPARE, 0x80, %GLOBAL_SPARE
-	EX_LD_FP(LOAD_BLK(%o1, %f32))
+	EX_LD_FP(LOAD_BLK(%o1, %f32), U1_gs_80_fp)
 	add		%o1, 0x40, %o1
 
 	/* There are 8 instances of the unrolled loop,
@@ -240,11 +353,11 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 
 	.align		64
 1:	FREG_FROB(f0, f2, f4, f6, f8, f10,f12,f14,f16)
-	LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+	LOOP_CHUNK1(o1, o0, 1f)
 	FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32)
-	LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+	LOOP_CHUNK2(o1, o0, 2f)
 	FREG_FROB(f32,f34,f36,f38,f40,f42,f44,f46,f0)
-	LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+	LOOP_CHUNK3(o1, o0, 3f)
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f0, %f2, %f48
 1:	FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32)
@@ -261,11 +374,11 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	STORE_JUMP(o0, f48, 56f)
 
 1:	FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18)
-	LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+	LOOP_CHUNK1(o1, o0, 1f)
 	FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34)
-	LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+	LOOP_CHUNK2(o1, o0, 2f)
 	FREG_FROB(f34,f36,f38,f40,f42,f44,f46,f0, f2)
-	LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+	LOOP_CHUNK3(o1, o0, 3f)
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f2, %f4, %f48
 1:	FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34)
@@ -282,11 +395,11 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	STORE_JUMP(o0, f48, 57f)
 
 1:	FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20)
-	LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+	LOOP_CHUNK1(o1, o0, 1f)
 	FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36)
-	LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+	LOOP_CHUNK2(o1, o0, 2f)
 	FREG_FROB(f36,f38,f40,f42,f44,f46,f0, f2, f4)
-	LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+	LOOP_CHUNK3(o1, o0, 3f)
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f4, %f6, %f48
 1:	FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36)
@@ -303,11 +416,11 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	STORE_JUMP(o0, f48, 58f)
 
 1:	FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22)
-	LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+	LOOP_CHUNK1(o1, o0, 1f)
 	FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38)
-	LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+	LOOP_CHUNK2(o1, o0, 2f)
 	FREG_FROB(f38,f40,f42,f44,f46,f0, f2, f4, f6) 
-	LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+	LOOP_CHUNK3(o1, o0, 3f)
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f6, %f8, %f48
 1:	FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38)
@@ -324,11 +437,11 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	STORE_JUMP(o0, f48, 59f)
 
 1:	FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24)
-	LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+	LOOP_CHUNK1(o1, o0, 1f)
 	FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40)
-	LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+	LOOP_CHUNK2(o1, o0, 2f)
 	FREG_FROB(f40,f42,f44,f46,f0, f2, f4, f6, f8)
-	LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+	LOOP_CHUNK3(o1, o0, 3f)
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f8, %f10, %f48
 1:	FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40)
@@ -345,11 +458,11 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	STORE_JUMP(o0, f48, 60f)
 
 1:	FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26)
-	LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+	LOOP_CHUNK1(o1, o0, 1f)
 	FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42)
-	LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+	LOOP_CHUNK2(o1, o0, 2f)
 	FREG_FROB(f42,f44,f46,f0, f2, f4, f6, f8, f10)
-	LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+	LOOP_CHUNK3(o1, o0, 3f)
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f10, %f12, %f48
 1:	FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42)
@@ -366,11 +479,11 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	STORE_JUMP(o0, f48, 61f)
 
 1:	FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28)
-	LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+	LOOP_CHUNK1(o1, o0, 1f)
 	FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44)
-	LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+	LOOP_CHUNK2(o1, o0, 2f)
 	FREG_FROB(f44,f46,f0, f2, f4, f6, f8, f10,f12)
-	LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+	LOOP_CHUNK3(o1, o0, 3f)
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f12, %f14, %f48
 1:	FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44)
@@ -387,11 +500,11 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	STORE_JUMP(o0, f48, 62f)
 
 1:	FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30)
-	LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+	LOOP_CHUNK1(o1, o0, 1f)
 	FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46)
-	LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+	LOOP_CHUNK2(o1, o0, 2f)
 	FREG_FROB(f46,f0, f2, f4, f6, f8, f10,f12,f14)
-	LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+	LOOP_CHUNK3(o1, o0, 3f)
 	ba,pt		%xcc, 1b+4
 	 faligndata	%f14, %f16, %f48
 1:	FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46)
@@ -407,53 +520,53 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46)
 	STORE_JUMP(o0, f48, 63f)
 
-40:	FINISH_VISCHUNK(o0, f0,  f2,  g3)
-41:	FINISH_VISCHUNK(o0, f2,  f4,  g3)
-42:	FINISH_VISCHUNK(o0, f4,  f6,  g3)
-43:	FINISH_VISCHUNK(o0, f6,  f8,  g3)
-44:	FINISH_VISCHUNK(o0, f8,  f10, g3)
-45:	FINISH_VISCHUNK(o0, f10, f12, g3)
-46:	FINISH_VISCHUNK(o0, f12, f14, g3)
-47:	UNEVEN_VISCHUNK(o0, f14, f0,  g3)
-48:	FINISH_VISCHUNK(o0, f16, f18, g3)
-49:	FINISH_VISCHUNK(o0, f18, f20, g3)
-50:	FINISH_VISCHUNK(o0, f20, f22, g3)
-51:	FINISH_VISCHUNK(o0, f22, f24, g3)
-52:	FINISH_VISCHUNK(o0, f24, f26, g3)
-53:	FINISH_VISCHUNK(o0, f26, f28, g3)
-54:	FINISH_VISCHUNK(o0, f28, f30, g3)
-55:	UNEVEN_VISCHUNK(o0, f30, f0,  g3)
-56:	FINISH_VISCHUNK(o0, f32, f34, g3)
-57:	FINISH_VISCHUNK(o0, f34, f36, g3)
-58:	FINISH_VISCHUNK(o0, f36, f38, g3)
-59:	FINISH_VISCHUNK(o0, f38, f40, g3)
-60:	FINISH_VISCHUNK(o0, f40, f42, g3)
-61:	FINISH_VISCHUNK(o0, f42, f44, g3)
-62:	FINISH_VISCHUNK(o0, f44, f46, g3)
-63:	UNEVEN_VISCHUNK_LAST(o0, f46, f0,  g3)
-
-93:	EX_LD_FP(LOAD(ldd, %o1, %f2))
+40:	FINISH_VISCHUNK(o0, f0,  f2)
+41:	FINISH_VISCHUNK(o0, f2,  f4)
+42:	FINISH_VISCHUNK(o0, f4,  f6)
+43:	FINISH_VISCHUNK(o0, f6,  f8)
+44:	FINISH_VISCHUNK(o0, f8,  f10)
+45:	FINISH_VISCHUNK(o0, f10, f12)
+46:	FINISH_VISCHUNK(o0, f12, f14)
+47:	UNEVEN_VISCHUNK(o0, f14, f0)
+48:	FINISH_VISCHUNK(o0, f16, f18)
+49:	FINISH_VISCHUNK(o0, f18, f20)
+50:	FINISH_VISCHUNK(o0, f20, f22)
+51:	FINISH_VISCHUNK(o0, f22, f24)
+52:	FINISH_VISCHUNK(o0, f24, f26)
+53:	FINISH_VISCHUNK(o0, f26, f28)
+54:	FINISH_VISCHUNK(o0, f28, f30)
+55:	UNEVEN_VISCHUNK(o0, f30, f0)
+56:	FINISH_VISCHUNK(o0, f32, f34)
+57:	FINISH_VISCHUNK(o0, f34, f36)
+58:	FINISH_VISCHUNK(o0, f36, f38)
+59:	FINISH_VISCHUNK(o0, f38, f40)
+60:	FINISH_VISCHUNK(o0, f40, f42)
+61:	FINISH_VISCHUNK(o0, f42, f44)
+62:	FINISH_VISCHUNK(o0, f44, f46)
+63:	UNEVEN_VISCHUNK_LAST(o0, f46, f0)
+
+93:	EX_LD_FP(LOAD(ldd, %o1, %f2), U1_g3_0_fp)
 	add		%o1, 8, %o1
 	subcc		%g3, 8, %g3
 	faligndata	%f0, %f2, %f8
-	EX_ST_FP(STORE(std, %f8, %o0))
+	EX_ST_FP(STORE(std, %f8, %o0), U1_g3_8_fp)
 	bl,pn		%xcc, 95f
 	 add		%o0, 8, %o0
-	EX_LD_FP(LOAD(ldd, %o1, %f0))
+	EX_LD_FP(LOAD(ldd, %o1, %f0), U1_g3_0_fp)
 	add		%o1, 8, %o1
 	subcc		%g3, 8, %g3
 	faligndata	%f2, %f0, %f8
-	EX_ST_FP(STORE(std, %f8, %o0))
+	EX_ST_FP(STORE(std, %f8, %o0), U1_g3_8_fp)
 	bge,pt		%xcc, 93b
 	 add		%o0, 8, %o0
 
 95:	brz,pt		%o2, 2f
 	 mov		%g1, %o1
 
-1:	EX_LD_FP(LOAD(ldub, %o1, %o3))
+1:	EX_LD_FP(LOAD(ldub, %o1, %o3), U1_o2_0_fp)
 	add		%o1, 1, %o1
 	subcc		%o2, 1, %o2
-	EX_ST_FP(STORE(stb, %o3, %o0))
+	EX_ST_FP(STORE(stb, %o3, %o0), U1_o2_1_fp)
 	bne,pt		%xcc, 1b
 	 add		%o0, 1, %o0
 
@@ -469,27 +582,27 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 
 72:	andn		%o2, 0xf, %GLOBAL_SPARE
 	and		%o2, 0xf, %o2
-1:	EX_LD(LOAD(ldx, %o1 + 0x00, %o5))
-	EX_LD(LOAD(ldx, %o1 + 0x08, %g1))
+1:	EX_LD(LOAD(ldx, %o1 + 0x00, %o5), U1_gs_0)
+	EX_LD(LOAD(ldx, %o1 + 0x08, %g1), U1_gs_0)
 	subcc		%GLOBAL_SPARE, 0x10, %GLOBAL_SPARE
-	EX_ST(STORE(stx, %o5, %o1 + %o3))
+	EX_ST(STORE(stx, %o5, %o1 + %o3), U1_gs_10)
 	add		%o1, 0x8, %o1
-	EX_ST(STORE(stx, %g1, %o1 + %o3))
+	EX_ST(STORE(stx, %g1, %o1 + %o3), U1_gs_8)
 	bgu,pt		%XCC, 1b
 	 add		%o1, 0x8, %o1
 73:	andcc		%o2, 0x8, %g0
 	be,pt		%XCC, 1f
 	 nop
-	EX_LD(LOAD(ldx, %o1, %o5))
+	EX_LD(LOAD(ldx, %o1, %o5), U1_o2_0)
 	sub		%o2, 0x8, %o2
-	EX_ST(STORE(stx, %o5, %o1 + %o3))
+	EX_ST(STORE(stx, %o5, %o1 + %o3), U1_o2_8)
 	add		%o1, 0x8, %o1
 1:	andcc		%o2, 0x4, %g0
 	be,pt		%XCC, 1f
 	 nop
-	EX_LD(LOAD(lduw, %o1, %o5))
+	EX_LD(LOAD(lduw, %o1, %o5), U1_o2_0)
 	sub		%o2, 0x4, %o2
-	EX_ST(STORE(stw, %o5, %o1 + %o3))
+	EX_ST(STORE(stw, %o5, %o1 + %o3), U1_o2_4)
 	add		%o1, 0x4, %o1
 1:	cmp		%o2, 0
 	be,pt		%XCC, 85f
@@ -503,9 +616,9 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	 sub		%g0, %g1, %g1
 	sub		%o2, %g1, %o2
 
-1:	EX_LD(LOAD(ldub, %o1, %o5))
+1:	EX_LD(LOAD(ldub, %o1, %o5), U1_g1_0)
 	subcc		%g1, 1, %g1
-	EX_ST(STORE(stb, %o5, %o1 + %o3))
+	EX_ST(STORE(stb, %o5, %o1 + %o3), U1_g1_1)
 	bgu,pt		%icc, 1b
 	 add		%o1, 1, %o1
 
@@ -521,16 +634,16 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 
 8:	mov		64, %o3
 	andn		%o1, 0x7, %o1
-	EX_LD(LOAD(ldx, %o1, %g2))
+	EX_LD(LOAD(ldx, %o1, %g2), U1_o2_0)
 	sub		%o3, %g1, %o3
 	andn		%o2, 0x7, %GLOBAL_SPARE
 	sllx		%g2, %g1, %g2
-1:	EX_LD(LOAD(ldx, %o1 + 0x8, %g3))
+1:	EX_LD(LOAD(ldx, %o1 + 0x8, %g3), U1_gs_0_o2_adj)
 	subcc		%GLOBAL_SPARE, 0x8, %GLOBAL_SPARE
 	add		%o1, 0x8, %o1
 	srlx		%g3, %o3, %o5
 	or		%o5, %g2, %o5
-	EX_ST(STORE(stx, %o5, %o0))
+	EX_ST(STORE(stx, %o5, %o0), U1_gs_8_o2_adj)
 	add		%o0, 0x8, %o0
 	bgu,pt		%icc, 1b
 	 sllx		%g3, %g1, %g2
@@ -548,9 +661,9 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	bne,pn		%XCC, 90f
 	 sub		%o0, %o1, %o3
 
-1:	EX_LD(LOAD(lduw, %o1, %g1))
+1:	EX_LD(LOAD(lduw, %o1, %g1), U1_o2_0)
 	subcc		%o2, 4, %o2
-	EX_ST(STORE(stw, %g1, %o1 + %o3))
+	EX_ST(STORE(stw, %g1, %o1 + %o3), U1_o2_4)
 	bgu,pt		%XCC, 1b
 	 add		%o1, 4, %o1
 
@@ -558,9 +671,9 @@ FUNC_NAME:		/* %o0=dst, %o1=src, %o2=len */
 	 mov		EX_RETVAL(%o4), %o0
 
 	.align		32
-90:	EX_LD(LOAD(ldub, %o1, %g1))
+90:	EX_LD(LOAD(ldub, %o1, %g1), U1_o2_0)
 	subcc		%o2, 1, %o2
-	EX_ST(STORE(stb, %g1, %o1 + %o3))
+	EX_ST(STORE(stb, %g1, %o1 + %o3), U1_o2_1)
 	bgu,pt		%XCC, 90b
 	 add		%o1, 1, %o1
 	retl
diff --git a/arch/sparc/lib/U3copy_from_user.S b/arch/sparc/lib/U3copy_from_user.S
index 88ad73d86fe4..db73010a1af8 100644
--- a/arch/sparc/lib/U3copy_from_user.S
+++ b/arch/sparc/lib/U3copy_from_user.S
@@ -3,19 +3,19 @@
  * Copyright (C) 1999, 2000, 2004 David S. Miller (davem@xxxxxxxxxx)
  */
 
-#define EX_LD(x)		\
+#define EX_LD(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one;	\
+	.word 98b, y;		\
 	.text;			\
 	.align 4;
 
-#define EX_LD_FP(x)		\
+#define EX_LD_FP(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one_fp;\
+	.word 98b, y##_fp;	\
 	.text;			\
 	.align 4;
 
diff --git a/arch/sparc/lib/U3copy_to_user.S b/arch/sparc/lib/U3copy_to_user.S
index 845139d75537..c4ee858e352a 100644
--- a/arch/sparc/lib/U3copy_to_user.S
+++ b/arch/sparc/lib/U3copy_to_user.S
@@ -3,19 +3,19 @@
  * Copyright (C) 1999, 2000, 2004 David S. Miller (davem@xxxxxxxxxx)
  */
 
-#define EX_ST(x)		\
+#define EX_ST(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one;	\
+	.word 98b, y;		\
 	.text;			\
 	.align 4;
 
-#define EX_ST_FP(x)		\
+#define EX_ST_FP(x,y)		\
 98:	x;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one_fp;\
+	.word 98b, y##_fp;	\
 	.text;			\
 	.align 4;
 
diff --git a/arch/sparc/lib/U3memcpy.S b/arch/sparc/lib/U3memcpy.S
index 491ee69e4995..54f98706b03b 100644
--- a/arch/sparc/lib/U3memcpy.S
+++ b/arch/sparc/lib/U3memcpy.S
@@ -4,6 +4,7 @@
  */
 
 #ifdef __KERNEL__
+#include <linux/linkage.h>
 #include <asm/visasm.h>
 #include <asm/asi.h>
 #define GLOBAL_SPARE	%g7
@@ -22,21 +23,17 @@
 #endif
 
 #ifndef EX_LD
-#define EX_LD(x)	x
+#define EX_LD(x,y)	x
 #endif
 #ifndef EX_LD_FP
-#define EX_LD_FP(x)	x
+#define EX_LD_FP(x,y)	x
 #endif
 
 #ifndef EX_ST
-#define EX_ST(x)	x
+#define EX_ST(x,y)	x
 #endif
 #ifndef EX_ST_FP
-#define EX_ST_FP(x)	x
-#endif
-
-#ifndef EX_RETVAL
-#define EX_RETVAL(x)	x
+#define EX_ST_FP(x,y)	x
 #endif
 
 #ifndef LOAD
@@ -77,6 +74,87 @@
 	 */
 
 	.text
+#ifndef EX_RETVAL
+#define EX_RETVAL(x)	x
+__restore_fp:
+	VISExitHalf
+	retl
+	 nop
+ENTRY(U3_retl_o2_plus_g2_plus_g1_plus_1_fp)
+	add	%g1, 1, %g1
+	add	%g2, %g1, %g2
+	ba,pt	%xcc, __restore_fp
+	 add	%o2, %g2, %o0
+ENDPROC(U3_retl_o2_plus_g2_plus_g1_plus_1_fp)
+ENTRY(U3_retl_o2_plus_g2_fp)
+	ba,pt	%xcc, __restore_fp
+	 add	%o2, %g2, %o0
+ENDPROC(U3_retl_o2_plus_g2_fp)
+ENTRY(U3_retl_o2_plus_g2_plus_8_fp)
+	add	%g2, 8, %g2
+	ba,pt	%xcc, __restore_fp
+	 add	%o2, %g2, %o0
+ENDPROC(U3_retl_o2_plus_g2_plus_8_fp)
+ENTRY(U3_retl_o2)
+	retl
+	 mov	%o2, %o0
+ENDPROC(U3_retl_o2)
+ENTRY(U3_retl_o2_plus_1)
+	retl
+	 add	%o2, 1, %o0
+ENDPROC(U3_retl_o2_plus_1)
+ENTRY(U3_retl_o2_plus_4)
+	retl
+	 add	%o2, 4, %o0
+ENDPROC(U3_retl_o2_plus_4)
+ENTRY(U3_retl_o2_plus_8)
+	retl
+	 add	%o2, 8, %o0
+ENDPROC(U3_retl_o2_plus_8)
+ENTRY(U3_retl_o2_plus_g1_plus_1)
+	add	%g1, 1, %g1
+	retl
+	 add	%o2, %g1, %o0
+ENDPROC(U3_retl_o2_plus_g1_plus_1)
+ENTRY(U3_retl_o2_fp)
+	ba,pt	%xcc, __restore_fp
+	 mov	%o2, %o0
+ENDPROC(U3_retl_o2_fp)
+ENTRY(U3_retl_o2_plus_o3_sll_6_plus_0x80_fp)
+	sll	%o3, 6, %o3
+	add	%o3, 0x80, %o3
+	ba,pt	%xcc, __restore_fp
+	 add	%o2, %o3, %o0
+ENDPROC(U3_retl_o2_plus_o3_sll_6_plus_0x80_fp)
+ENTRY(U3_retl_o2_plus_o3_sll_6_plus_0x40_fp)
+	sll	%o3, 6, %o3
+	add	%o3, 0x40, %o3
+	ba,pt	%xcc, __restore_fp
+	 add	%o2, %o3, %o0
+ENDPROC(U3_retl_o2_plus_o3_sll_6_plus_0x40_fp)
+ENTRY(U3_retl_o2_plus_GS_plus_0x10)
+	add	GLOBAL_SPARE, 0x10, GLOBAL_SPARE
+	retl
+	 add	%o2, GLOBAL_SPARE, %o0
+ENDPROC(U3_retl_o2_plus_GS_plus_0x10)
+ENTRY(U3_retl_o2_plus_GS_plus_0x08)
+	add	GLOBAL_SPARE, 0x08, GLOBAL_SPARE
+	retl
+	 add	%o2, GLOBAL_SPARE, %o0
+ENDPROC(U3_retl_o2_plus_GS_plus_0x08)
+ENTRY(U3_retl_o2_and_7_plus_GS)
+	and	%o2, 7, %o2
+	retl
+	 add	%o2, GLOBAL_SPARE, %o2
+ENDPROC(U3_retl_o2_and_7_plus_GS)
+ENTRY(U3_retl_o2_and_7_plus_GS_plus_8)
+	add	GLOBAL_SPARE, 8, GLOBAL_SPARE
+	and	%o2, 7, %o2
+	retl
+	 add	%o2, GLOBAL_SPARE, %o2
+ENDPROC(U3_retl_o2_and_7_plus_GS_plus_8)
+#endif
+
 	.align		64
 
 	/* The cheetah's flexible spine, oversized liver, enlarged heart,
@@ -126,8 +204,8 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	 and		%g2, 0x38, %g2
 
 1:	subcc		%g1, 0x1, %g1
-	EX_LD_FP(LOAD(ldub, %o1 + 0x00, %o3))
-	EX_ST_FP(STORE(stb, %o3, %o1 + GLOBAL_SPARE))
+	EX_LD_FP(LOAD(ldub, %o1 + 0x00, %o3), U3_retl_o2_plus_g2_plus_g1_plus_1)
+	EX_ST_FP(STORE(stb, %o3, %o1 + GLOBAL_SPARE), U3_retl_o2_plus_g2_plus_g1_plus_1)
 	bgu,pt		%XCC, 1b
 	 add		%o1, 0x1, %o1
 
@@ -138,20 +216,20 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	be,pt		%icc, 3f
 	 alignaddr	%o1, %g0, %o1
 
-	EX_LD_FP(LOAD(ldd, %o1, %f4))
-1:	EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f6))
+	EX_LD_FP(LOAD(ldd, %o1, %f4), U3_retl_o2_plus_g2)
+1:	EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f6), U3_retl_o2_plus_g2)
 	add		%o1, 0x8, %o1
 	subcc		%g2, 0x8, %g2
 	faligndata	%f4, %f6, %f0
-	EX_ST_FP(STORE(std, %f0, %o0))
+	EX_ST_FP(STORE(std, %f0, %o0), U3_retl_o2_plus_g2_plus_8)
 	be,pn		%icc, 3f
 	 add		%o0, 0x8, %o0
 
-	EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f4))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f4), U3_retl_o2_plus_g2)
 	add		%o1, 0x8, %o1
 	subcc		%g2, 0x8, %g2
 	faligndata	%f6, %f4, %f2
-	EX_ST_FP(STORE(std, %f2, %o0))
+	EX_ST_FP(STORE(std, %f2, %o0), U3_retl_o2_plus_g2_plus_8)
 	bne,pt		%icc, 1b
 	 add		%o0, 0x8, %o0
 
@@ -161,25 +239,25 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	LOAD(prefetch, %o1 + 0x080, #one_read)
 	LOAD(prefetch, %o1 + 0x0c0, #one_read)
 	LOAD(prefetch, %o1 + 0x100, #one_read)
-	EX_LD_FP(LOAD(ldd, %o1 + 0x000, %f0))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x000, %f0), U3_retl_o2)
 	LOAD(prefetch, %o1 + 0x140, #one_read)
-	EX_LD_FP(LOAD(ldd, %o1 + 0x008, %f2))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x008, %f2), U3_retl_o2)
 	LOAD(prefetch, %o1 + 0x180, #one_read)
-	EX_LD_FP(LOAD(ldd, %o1 + 0x010, %f4))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x010, %f4), U3_retl_o2)
 	LOAD(prefetch, %o1 + 0x1c0, #one_read)
 	faligndata	%f0, %f2, %f16
-	EX_LD_FP(LOAD(ldd, %o1 + 0x018, %f6))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x018, %f6), U3_retl_o2)
 	faligndata	%f2, %f4, %f18
-	EX_LD_FP(LOAD(ldd, %o1 + 0x020, %f8))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x020, %f8), U3_retl_o2)
 	faligndata	%f4, %f6, %f20
-	EX_LD_FP(LOAD(ldd, %o1 + 0x028, %f10))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x028, %f10), U3_retl_o2)
 	faligndata	%f6, %f8, %f22
 
-	EX_LD_FP(LOAD(ldd, %o1 + 0x030, %f12))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x030, %f12), U3_retl_o2)
 	faligndata	%f8, %f10, %f24
-	EX_LD_FP(LOAD(ldd, %o1 + 0x038, %f14))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x038, %f14), U3_retl_o2)
 	faligndata	%f10, %f12, %f26
-	EX_LD_FP(LOAD(ldd, %o1 + 0x040, %f0))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x040, %f0), U3_retl_o2)
 
 	subcc		GLOBAL_SPARE, 0x80, GLOBAL_SPARE
 	add		%o1, 0x40, %o1
@@ -190,26 +268,26 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 
 	.align		64
 1:
-	EX_LD_FP(LOAD(ldd, %o1 + 0x008, %f2))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x008, %f2), U3_retl_o2_plus_o3_sll_6_plus_0x80)
 	faligndata	%f12, %f14, %f28
-	EX_LD_FP(LOAD(ldd, %o1 + 0x010, %f4))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x010, %f4), U3_retl_o2_plus_o3_sll_6_plus_0x80)
 	faligndata	%f14, %f0, %f30
-	EX_ST_FP(STORE_BLK(%f16, %o0))
-	EX_LD_FP(LOAD(ldd, %o1 + 0x018, %f6))
+	EX_ST_FP(STORE_BLK(%f16, %o0), U3_retl_o2_plus_o3_sll_6_plus_0x80)
+	EX_LD_FP(LOAD(ldd, %o1 + 0x018, %f6), U3_retl_o2_plus_o3_sll_6_plus_0x40)
 	faligndata	%f0, %f2, %f16
 	add		%o0, 0x40, %o0
 
-	EX_LD_FP(LOAD(ldd, %o1 + 0x020, %f8))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x020, %f8), U3_retl_o2_plus_o3_sll_6_plus_0x40)
 	faligndata	%f2, %f4, %f18
-	EX_LD_FP(LOAD(ldd, %o1 + 0x028, %f10))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x028, %f10), U3_retl_o2_plus_o3_sll_6_plus_0x40)
 	faligndata	%f4, %f6, %f20
-	EX_LD_FP(LOAD(ldd, %o1 + 0x030, %f12))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x030, %f12), U3_retl_o2_plus_o3_sll_6_plus_0x40)
 	subcc		%o3, 0x01, %o3
 	faligndata	%f6, %f8, %f22
-	EX_LD_FP(LOAD(ldd, %o1 + 0x038, %f14))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x038, %f14), U3_retl_o2_plus_o3_sll_6_plus_0x80)
 
 	faligndata	%f8, %f10, %f24
-	EX_LD_FP(LOAD(ldd, %o1 + 0x040, %f0))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x040, %f0), U3_retl_o2_plus_o3_sll_6_plus_0x80)
 	LOAD(prefetch, %o1 + 0x1c0, #one_read)
 	faligndata	%f10, %f12, %f26
 	bg,pt		%XCC, 1b
@@ -217,29 +295,29 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 
 	/* Finally we copy the last full 64-byte block. */
 2:
-	EX_LD_FP(LOAD(ldd, %o1 + 0x008, %f2))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x008, %f2), U3_retl_o2_plus_o3_sll_6_plus_0x80)
 	faligndata	%f12, %f14, %f28
-	EX_LD_FP(LOAD(ldd, %o1 + 0x010, %f4))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x010, %f4), U3_retl_o2_plus_o3_sll_6_plus_0x80)
 	faligndata	%f14, %f0, %f30
-	EX_ST_FP(STORE_BLK(%f16, %o0))
-	EX_LD_FP(LOAD(ldd, %o1 + 0x018, %f6))
+	EX_ST_FP(STORE_BLK(%f16, %o0), U3_retl_o2_plus_o3_sll_6_plus_0x80)
+	EX_LD_FP(LOAD(ldd, %o1 + 0x018, %f6), U3_retl_o2_plus_o3_sll_6_plus_0x40)
 	faligndata	%f0, %f2, %f16
-	EX_LD_FP(LOAD(ldd, %o1 + 0x020, %f8))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x020, %f8), U3_retl_o2_plus_o3_sll_6_plus_0x40)
 	faligndata	%f2, %f4, %f18
-	EX_LD_FP(LOAD(ldd, %o1 + 0x028, %f10))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x028, %f10), U3_retl_o2_plus_o3_sll_6_plus_0x40)
 	faligndata	%f4, %f6, %f20
-	EX_LD_FP(LOAD(ldd, %o1 + 0x030, %f12))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x030, %f12), U3_retl_o2_plus_o3_sll_6_plus_0x40)
 	faligndata	%f6, %f8, %f22
-	EX_LD_FP(LOAD(ldd, %o1 + 0x038, %f14))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x038, %f14), U3_retl_o2_plus_o3_sll_6_plus_0x40)
 	faligndata	%f8, %f10, %f24
 	cmp		%g1, 0
 	be,pt		%XCC, 1f
 	 add		%o0, 0x40, %o0
-	EX_LD_FP(LOAD(ldd, %o1 + 0x040, %f0))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x040, %f0), U3_retl_o2_plus_o3_sll_6_plus_0x40)
 1:	faligndata	%f10, %f12, %f26
 	faligndata	%f12, %f14, %f28
 	faligndata	%f14, %f0, %f30
-	EX_ST_FP(STORE_BLK(%f16, %o0))
+	EX_ST_FP(STORE_BLK(%f16, %o0), U3_retl_o2_plus_o3_sll_6_plus_0x40)
 	add		%o0, 0x40, %o0
 	add		%o1, 0x40, %o1
 	membar		#Sync
@@ -259,20 +337,20 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 
 	sub		%o2, %g2, %o2
 	be,a,pt		%XCC, 1f
-	 EX_LD_FP(LOAD(ldd, %o1 + 0x00, %f0))
+	 EX_LD_FP(LOAD(ldd, %o1 + 0x00, %f0), U3_retl_o2_plus_g2)
 
-1:	EX_LD_FP(LOAD(ldd, %o1 + 0x08, %f2))
+1:	EX_LD_FP(LOAD(ldd, %o1 + 0x08, %f2), U3_retl_o2_plus_g2)
 	add		%o1, 0x8, %o1
 	subcc		%g2, 0x8, %g2
 	faligndata	%f0, %f2, %f8
-	EX_ST_FP(STORE(std, %f8, %o0))
+	EX_ST_FP(STORE(std, %f8, %o0), U3_retl_o2_plus_g2_plus_8)
 	be,pn		%XCC, 2f
 	 add		%o0, 0x8, %o0
-	EX_LD_FP(LOAD(ldd, %o1 + 0x08, %f0))
+	EX_LD_FP(LOAD(ldd, %o1 + 0x08, %f0), U3_retl_o2_plus_g2)
 	add		%o1, 0x8, %o1
 	subcc		%g2, 0x8, %g2
 	faligndata	%f2, %f0, %f8
-	EX_ST_FP(STORE(std, %f8, %o0))
+	EX_ST_FP(STORE(std, %f8, %o0), U3_retl_o2_plus_g2_plus_8)
 	bne,pn		%XCC, 1b
 	 add		%o0, 0x8, %o0
 
@@ -292,30 +370,33 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	 andcc		%o2, 0x8, %g0
 	be,pt		%icc, 1f
 	 nop
-	EX_LD(LOAD(ldx, %o1, %o5))
-	EX_ST(STORE(stx, %o5, %o1 + %o3))
+	EX_LD(LOAD(ldx, %o1, %o5), U3_retl_o2)
+	EX_ST(STORE(stx, %o5, %o1 + %o3), U3_retl_o2)
 	add		%o1, 0x8, %o1
+	sub		%o2, 8, %o2
 
 1:	andcc		%o2, 0x4, %g0
 	be,pt		%icc, 1f
 	 nop
-	EX_LD(LOAD(lduw, %o1, %o5))
-	EX_ST(STORE(stw, %o5, %o1 + %o3))
+	EX_LD(LOAD(lduw, %o1, %o5), U3_retl_o2)
+	EX_ST(STORE(stw, %o5, %o1 + %o3), U3_retl_o2)
 	add		%o1, 0x4, %o1
+	sub		%o2, 4, %o2
 
 1:	andcc		%o2, 0x2, %g0
 	be,pt		%icc, 1f
 	 nop
-	EX_LD(LOAD(lduh, %o1, %o5))
-	EX_ST(STORE(sth, %o5, %o1 + %o3))
+	EX_LD(LOAD(lduh, %o1, %o5), U3_retl_o2)
+	EX_ST(STORE(sth, %o5, %o1 + %o3), U3_retl_o2)
 	add		%o1, 0x2, %o1
+	sub		%o2, 2, %o2
 
 1:	andcc		%o2, 0x1, %g0
 	be,pt		%icc, 85f
 	 nop
-	EX_LD(LOAD(ldub, %o1, %o5))
+	EX_LD(LOAD(ldub, %o1, %o5), U3_retl_o2)
 	ba,pt		%xcc, 85f
-	 EX_ST(STORE(stb, %o5, %o1 + %o3))
+	 EX_ST(STORE(stb, %o5, %o1 + %o3), U3_retl_o2)
 
 	.align		64
 70: /* 16 < len <= 64 */
@@ -326,26 +407,26 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	andn		%o2, 0xf, GLOBAL_SPARE
 	and		%o2, 0xf, %o2
 1:	subcc		GLOBAL_SPARE, 0x10, GLOBAL_SPARE
-	EX_LD(LOAD(ldx, %o1 + 0x00, %o5))
-	EX_LD(LOAD(ldx, %o1 + 0x08, %g1))
-	EX_ST(STORE(stx, %o5, %o1 + %o3))
+	EX_LD(LOAD(ldx, %o1 + 0x00, %o5), U3_retl_o2_plus_GS_plus_0x10)
+	EX_LD(LOAD(ldx, %o1 + 0x08, %g1), U3_retl_o2_plus_GS_plus_0x10)
+	EX_ST(STORE(stx, %o5, %o1 + %o3), U3_retl_o2_plus_GS_plus_0x10)
 	add		%o1, 0x8, %o1
-	EX_ST(STORE(stx, %g1, %o1 + %o3))
+	EX_ST(STORE(stx, %g1, %o1 + %o3), U3_retl_o2_plus_GS_plus_0x08)
 	bgu,pt		%XCC, 1b
 	 add		%o1, 0x8, %o1
 73:	andcc		%o2, 0x8, %g0
 	be,pt		%XCC, 1f
 	 nop
 	sub		%o2, 0x8, %o2
-	EX_LD(LOAD(ldx, %o1, %o5))
-	EX_ST(STORE(stx, %o5, %o1 + %o3))
+	EX_LD(LOAD(ldx, %o1, %o5), U3_retl_o2_plus_8)
+	EX_ST(STORE(stx, %o5, %o1 + %o3), U3_retl_o2_plus_8)
 	add		%o1, 0x8, %o1
 1:	andcc		%o2, 0x4, %g0
 	be,pt		%XCC, 1f
 	 nop
 	sub		%o2, 0x4, %o2
-	EX_LD(LOAD(lduw, %o1, %o5))
-	EX_ST(STORE(stw, %o5, %o1 + %o3))
+	EX_LD(LOAD(lduw, %o1, %o5), U3_retl_o2_plus_4)
+	EX_ST(STORE(stw, %o5, %o1 + %o3), U3_retl_o2_plus_4)
 	add		%o1, 0x4, %o1
 1:	cmp		%o2, 0
 	be,pt		%XCC, 85f
@@ -361,8 +442,8 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	sub		%o2, %g1, %o2
 
 1:	subcc		%g1, 1, %g1
-	EX_LD(LOAD(ldub, %o1, %o5))
-	EX_ST(STORE(stb, %o5, %o1 + %o3))
+	EX_LD(LOAD(ldub, %o1, %o5), U3_retl_o2_plus_g1_plus_1)
+	EX_ST(STORE(stb, %o5, %o1 + %o3), U3_retl_o2_plus_g1_plus_1)
 	bgu,pt		%icc, 1b
 	 add		%o1, 1, %o1
 
@@ -378,16 +459,16 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 
 8:	mov		64, %o3
 	andn		%o1, 0x7, %o1
-	EX_LD(LOAD(ldx, %o1, %g2))
+	EX_LD(LOAD(ldx, %o1, %g2), U3_retl_o2)
 	sub		%o3, %g1, %o3
 	andn		%o2, 0x7, GLOBAL_SPARE
 	sllx		%g2, %g1, %g2
-1:	EX_LD(LOAD(ldx, %o1 + 0x8, %g3))
+1:	EX_LD(LOAD(ldx, %o1 + 0x8, %g3), U3_retl_o2_and_7_plus_GS)
 	subcc		GLOBAL_SPARE, 0x8, GLOBAL_SPARE
 	add		%o1, 0x8, %o1
 	srlx		%g3, %o3, %o5
 	or		%o5, %g2, %o5
-	EX_ST(STORE(stx, %o5, %o0))
+	EX_ST(STORE(stx, %o5, %o0), U3_retl_o2_and_7_plus_GS_plus_8)
 	add		%o0, 0x8, %o0
 	bgu,pt		%icc, 1b
 	 sllx		%g3, %g1, %g2
@@ -407,8 +488,8 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 
 1:
 	subcc		%o2, 4, %o2
-	EX_LD(LOAD(lduw, %o1, %g1))
-	EX_ST(STORE(stw, %g1, %o1 + %o3))
+	EX_LD(LOAD(lduw, %o1, %g1), U3_retl_o2_plus_4)
+	EX_ST(STORE(stw, %g1, %o1 + %o3), U3_retl_o2_plus_4)
 	bgu,pt		%XCC, 1b
 	 add		%o1, 4, %o1
 
@@ -418,8 +499,8 @@ FUNC_NAME:	/* %o0=dst, %o1=src, %o2=len */
 	.align		32
 90:
 	subcc		%o2, 1, %o2
-	EX_LD(LOAD(ldub, %o1, %g1))
-	EX_ST(STORE(stb, %g1, %o1 + %o3))
+	EX_LD(LOAD(ldub, %o1, %g1), U3_retl_o2_plus_1)
+	EX_ST(STORE(stb, %g1, %o1 + %o3), U3_retl_o2_plus_1)
 	bgu,pt		%XCC, 90b
 	 add		%o1, 1, %o1
 	retl
diff --git a/arch/sparc/lib/copy_in_user.S b/arch/sparc/lib/copy_in_user.S
index 302c0e60dc2c..4c89b486fa0d 100644
--- a/arch/sparc/lib/copy_in_user.S
+++ b/arch/sparc/lib/copy_in_user.S
@@ -8,18 +8,33 @@
 
 #define XCC xcc
 
-#define EX(x,y)			\
+#define EX(x,y,z)		\
 98:	x,y;			\
 	.section __ex_table,"a";\
 	.align 4;		\
-	.word 98b, __retl_one;	\
+	.word 98b, z;		\
 	.text;			\
 	.align 4;
 
+#define EX_O4(x,y) EX(x,y,__retl_o4_plus_8)
+#define EX_O2_4(x,y) EX(x,y,__retl_o2_plus_4)
+#define EX_O2_1(x,y) EX(x,y,__retl_o2_plus_1)
+
 	.register	%g2,#scratch
 	.register	%g3,#scratch
 
 	.text
+__retl_o4_plus_8:
+	add	%o4, %o2, %o4
+	retl
+	 add	%o4, 8, %o0
+__retl_o2_plus_4:
+	retl
+	 add	%o2, 4, %o0
+__retl_o2_plus_1:
+	retl
+	 add	%o2, 1, %o0
+
 	.align	32
 
 	/* Don't try to get too fancy here, just nice and
@@ -44,8 +59,8 @@ ENTRY(___copy_in_user)	/* %o0=dst, %o1=src, %o2=len */
 	andn		%o2, 0x7, %o4
 	and		%o2, 0x7, %o2
 1:	subcc		%o4, 0x8, %o4
-	EX(ldxa [%o1] %asi, %o5)
-	EX(stxa %o5, [%o0] %asi)
+	EX_O4(ldxa [%o1] %asi, %o5)
+	EX_O4(stxa %o5, [%o0] %asi)
 	add		%o1, 0x8, %o1
 	bgu,pt		%XCC, 1b
 	 add		%o0, 0x8, %o0
@@ -53,8 +68,8 @@ ENTRY(___copy_in_user)	/* %o0=dst, %o1=src, %o2=len */
 	be,pt		%XCC, 1f
 	 nop
 	sub		%o2, 0x4, %o2
-	EX(lduwa [%o1] %asi, %o5)
-	EX(stwa %o5, [%o0] %asi)
+	EX_O2_4(lduwa [%o1] %asi, %o5)
+	EX_O2_4(stwa %o5, [%o0] %asi)
 	add		%o1, 0x4, %o1
 	add		%o0, 0x4, %o0
 1:	cmp		%o2, 0
@@ -70,8 +85,8 @@ ENTRY(___copy_in_user)	/* %o0=dst, %o1=src, %o2=len */
 
 82:
 	subcc		%o2, 4, %o2
-	EX(lduwa [%o1] %asi, %g1)
-	EX(stwa %g1, [%o0] %asi)
+	EX_O2_4(lduwa [%o1] %asi, %g1)
+	EX_O2_4(stwa %g1, [%o0] %asi)
 	add		%o1, 4, %o1
 	bgu,pt		%XCC, 82b
 	 add		%o0, 4, %o0
@@ -82,8 +97,8 @@ ENTRY(___copy_in_user)	/* %o0=dst, %o1=src, %o2=len */
 	.align	32
 90:
 	subcc		%o2, 1, %o2
-	EX(lduba [%o1] %asi, %g1)
-	EX(stba %g1, [%o0] %asi)
+	EX_O2_1(lduba [%o1] %asi, %g1)
+	EX_O2_1(stba %g1, [%o0] %asi)
 	add		%o1, 1, %o1
 	bgu,pt		%XCC, 90b
 	 add		%o0, 1, %o0
diff --git a/arch/sparc/lib/user_fixup.c b/arch/sparc/lib/user_fixup.c
deleted file mode 100644
index ac96ae236709..000000000000
--- a/arch/sparc/lib/user_fixup.c
+++ /dev/null
@@ -1,71 +0,0 @@
-/* user_fixup.c: Fix up user copy faults.
- *
- * Copyright (C) 2004 David S. Miller <davem@xxxxxxxxxx>
- */
-
-#include <linux/compiler.h>
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <linux/errno.h>
-#include <linux/module.h>
-
-#include <asm/uaccess.h>
-
-/* Calculating the exact fault address when using
- * block loads and stores can be very complicated.
- *
- * Instead of trying to be clever and handling all
- * of the cases, just fix things up simply here.
- */
-
-static unsigned long compute_size(unsigned long start, unsigned long size, unsigned long *offset)
-{
-	unsigned long fault_addr = current_thread_info()->fault_address;
-	unsigned long end = start + size;
-
-	if (fault_addr < start || fault_addr >= end) {
-		*offset = 0;
-	} else {
-		*offset = fault_addr - start;
-		size = end - fault_addr;
-	}
-	return size;
-}
-
-unsigned long copy_from_user_fixup(void *to, const void __user *from, unsigned long size)
-{
-	unsigned long offset;
-
-	size = compute_size((unsigned long) from, size, &offset);
-	if (likely(size))
-		memset(to + offset, 0, size);
-
-	return size;
-}
-EXPORT_SYMBOL(copy_from_user_fixup);
-
-unsigned long copy_to_user_fixup(void __user *to, const void *from, unsigned long size)
-{
-	unsigned long offset;
-
-	return compute_size((unsigned long) to, size, &offset);
-}
-EXPORT_SYMBOL(copy_to_user_fixup);
-
-unsigned long copy_in_user_fixup(void __user *to, void __user *from, unsigned long size)
-{
-	unsigned long fault_addr = current_thread_info()->fault_address;
-	unsigned long start = (unsigned long) to;
-	unsigned long end = start + size;
-
-	if (fault_addr >= start && fault_addr < end)
-		return end - fault_addr;
-
-	start = (unsigned long) from;
-	end = start + size;
-	if (fault_addr >= start && fault_addr < end)
-		return end - fault_addr;
-
-	return size;
-}
-EXPORT_SYMBOL(copy_in_user_fixup);
diff --git a/arch/sparc/mm/tsb.c b/arch/sparc/mm/tsb.c
index f2b77112e9d8..e20fbbafb0b0 100644
--- a/arch/sparc/mm/tsb.c
+++ b/arch/sparc/mm/tsb.c
@@ -27,6 +27,20 @@ static inline int tag_compare(unsigned long tag, unsigned long vaddr)
 	return (tag == (vaddr >> 22));
 }
 
+static void flush_tsb_kernel_range_scan(unsigned long start, unsigned long end)
+{
+	unsigned long idx;
+
+	for (idx = 0; idx < KERNEL_TSB_NENTRIES; idx++) {
+		struct tsb *ent = &swapper_tsb[idx];
+		unsigned long match = idx << 13;
+
+		match |= (ent->tag << 22);
+		if (match >= start && match < end)
+			ent->tag = (1UL << TSB_TAG_INVALID_BIT);
+	}
+}
+
 /* TSB flushes need only occur on the processor initiating the address
  * space modification, not on each cpu the address space has run on.
  * Only the TLB flush needs that treatment.
@@ -36,6 +50,9 @@ void flush_tsb_kernel_range(unsigned long start, unsigned long end)
 {
 	unsigned long v;
 
+	if ((end - start) >> PAGE_SHIFT >= 2 * KERNEL_TSB_NENTRIES)
+		return flush_tsb_kernel_range_scan(start, end);
+
 	for (v = start; v < end; v += PAGE_SIZE) {
 		unsigned long hash = tsb_hash(v, PAGE_SHIFT,
 					      KERNEL_TSB_NENTRIES);
diff --git a/arch/sparc/mm/ultra.S b/arch/sparc/mm/ultra.S
index b4f4733abc6e..5d2fd6cd3189 100644
--- a/arch/sparc/mm/ultra.S
+++ b/arch/sparc/mm/ultra.S
@@ -30,7 +30,7 @@
 	.text
 	.align		32
 	.globl		__flush_tlb_mm
-__flush_tlb_mm:		/* 18 insns */
+__flush_tlb_mm:		/* 19 insns */
 	/* %o0=(ctx & TAG_CONTEXT_BITS), %o1=SECONDARY_CONTEXT */
 	ldxa		[%o1] ASI_DMMU, %g2
 	cmp		%g2, %o0
@@ -81,7 +81,7 @@ __flush_tlb_page:	/* 22 insns */
 
 	.align		32
 	.globl		__flush_tlb_pending
-__flush_tlb_pending:	/* 26 insns */
+__flush_tlb_pending:	/* 27 insns */
 	/* %o0 = context, %o1 = nr, %o2 = vaddrs[] */
 	rdpr		%pstate, %g7
 	sllx		%o1, 3, %o1
@@ -113,12 +113,14 @@ __flush_tlb_pending:	/* 26 insns */
 
 	.align		32
 	.globl		__flush_tlb_kernel_range
-__flush_tlb_kernel_range:	/* 16 insns */
+__flush_tlb_kernel_range:	/* 31 insns */
 	/* %o0=start, %o1=end */
 	cmp		%o0, %o1
 	be,pn		%xcc, 2f
+	 sub		%o1, %o0, %o3
+	srlx		%o3, 18, %o4
+	brnz,pn		%o4, __spitfire_flush_tlb_kernel_range_slow
 	 sethi		%hi(PAGE_SIZE), %o4
-	sub		%o1, %o0, %o3
 	sub		%o3, %o4, %o3
 	or		%o0, 0x20, %o0		! Nucleus
 1:	stxa		%g0, [%o0 + %o3] ASI_DMMU_DEMAP
@@ -131,6 +133,41 @@ __flush_tlb_kernel_range:	/* 16 insns */
 	retl
 	 nop
 	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+
+__spitfire_flush_tlb_kernel_range_slow:
+	mov		63 * 8, %o4
+1:	ldxa		[%o4] ASI_ITLB_DATA_ACCESS, %o3
+	andcc		%o3, 0x40, %g0			/* _PAGE_L_4U */
+	bne,pn		%xcc, 2f
+	 mov		TLB_TAG_ACCESS, %o3
+	stxa		%g0, [%o3] ASI_IMMU
+	stxa		%g0, [%o4] ASI_ITLB_DATA_ACCESS
+	membar		#Sync
+2:	ldxa		[%o4] ASI_DTLB_DATA_ACCESS, %o3
+	andcc		%o3, 0x40, %g0
+	bne,pn		%xcc, 2f
+	 mov		TLB_TAG_ACCESS, %o3
+	stxa		%g0, [%o3] ASI_DMMU
+	stxa		%g0, [%o4] ASI_DTLB_DATA_ACCESS
+	membar		#Sync
+2:	sub		%o4, 8, %o4
+	brgez,pt	%o4, 1b
+	 nop
+	retl
+	 nop
 
 __spitfire_flush_tlb_mm_slow:
 	rdpr		%pstate, %g1
@@ -285,6 +322,40 @@ __cheetah_flush_tlb_pending:	/* 27 insns */
 	retl
 	 wrpr		%g7, 0x0, %pstate
 
+__cheetah_flush_tlb_kernel_range:	/* 31 insns */
+	/* %o0=start, %o1=end */
+	cmp		%o0, %o1
+	be,pn		%xcc, 2f
+	 sub		%o1, %o0, %o3
+	srlx		%o3, 18, %o4
+	brnz,pn		%o4, 3f
+	 sethi		%hi(PAGE_SIZE), %o4
+	sub		%o3, %o4, %o3
+	or		%o0, 0x20, %o0		! Nucleus
+1:	stxa		%g0, [%o0 + %o3] ASI_DMMU_DEMAP
+	stxa		%g0, [%o0 + %o3] ASI_IMMU_DEMAP
+	membar		#Sync
+	brnz,pt		%o3, 1b
+	 sub		%o3, %o4, %o3
+2:	sethi		%hi(KERNBASE), %o3
+	flush		%o3
+	retl
+	 nop
+3:	mov		0x80, %o4
+	stxa		%g0, [%o4] ASI_DMMU_DEMAP
+	membar		#Sync
+	stxa		%g0, [%o4] ASI_IMMU_DEMAP
+	membar		#Sync
+	retl
+	 nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+
 #ifdef DCACHE_ALIASING_POSSIBLE
 __cheetah_flush_dcache_page: /* 11 insns */
 	sethi		%hi(PAGE_OFFSET), %g1
@@ -309,19 +380,28 @@ __hypervisor_tlb_tl0_error:
 	ret
 	 restore
 
-__hypervisor_flush_tlb_mm: /* 10 insns */
+__hypervisor_flush_tlb_mm: /* 19 insns */
 	mov		%o0, %o2	/* ARG2: mmu context */
 	mov		0, %o0		/* ARG0: CPU lists unimplemented */
 	mov		0, %o1		/* ARG1: CPU lists unimplemented */
 	mov		HV_MMU_ALL, %o3	/* ARG3: flags */
 	mov		HV_FAST_MMU_DEMAP_CTX, %o5
 	ta		HV_FAST_TRAP
-	brnz,pn		%o0, __hypervisor_tlb_tl0_error
+	brnz,pn		%o0, 1f
 	 mov		HV_FAST_MMU_DEMAP_CTX, %o1
 	retl
 	 nop
+1:	sethi		%hi(__hypervisor_tlb_tl0_error), %o5
+	jmpl		%o5 + %lo(__hypervisor_tlb_tl0_error), %g0
+	 nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
 
-__hypervisor_flush_tlb_page: /* 11 insns */
+__hypervisor_flush_tlb_page: /* 22 insns */
 	/* %o0 = context, %o1 = vaddr */
 	mov		%o0, %g2
 	mov		%o1, %o0              /* ARG0: vaddr + IMMU-bit */
@@ -330,12 +410,23 @@ __hypervisor_flush_tlb_page: /* 11 insns */
 	srlx		%o0, PAGE_SHIFT, %o0
 	sllx		%o0, PAGE_SHIFT, %o0
 	ta		HV_MMU_UNMAP_ADDR_TRAP
-	brnz,pn		%o0, __hypervisor_tlb_tl0_error
+	brnz,pn		%o0, 1f
 	 mov		HV_MMU_UNMAP_ADDR_TRAP, %o1
 	retl
 	 nop
+1:	sethi		%hi(__hypervisor_tlb_tl0_error), %o2
+	jmpl		%o2 + %lo(__hypervisor_tlb_tl0_error), %g0
+	 nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
 
-__hypervisor_flush_tlb_pending: /* 16 insns */
+__hypervisor_flush_tlb_pending: /* 27 insns */
 	/* %o0 = context, %o1 = nr, %o2 = vaddrs[] */
 	sllx		%o1, 3, %g1
 	mov		%o2, %g2
@@ -347,31 +438,57 @@ __hypervisor_flush_tlb_pending: /* 16 insns */
 	srlx		%o0, PAGE_SHIFT, %o0
 	sllx		%o0, PAGE_SHIFT, %o0
 	ta		HV_MMU_UNMAP_ADDR_TRAP
-	brnz,pn		%o0, __hypervisor_tlb_tl0_error
+	brnz,pn		%o0, 1f
 	 mov		HV_MMU_UNMAP_ADDR_TRAP, %o1
 	brnz,pt		%g1, 1b
 	 nop
 	retl
 	 nop
+1:	sethi		%hi(__hypervisor_tlb_tl0_error), %o2
+	jmpl		%o2 + %lo(__hypervisor_tlb_tl0_error), %g0
+	 nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
 
-__hypervisor_flush_tlb_kernel_range: /* 16 insns */
+__hypervisor_flush_tlb_kernel_range: /* 31 insns */
 	/* %o0=start, %o1=end */
 	cmp		%o0, %o1
 	be,pn		%xcc, 2f
-	 sethi		%hi(PAGE_SIZE), %g3
-	mov		%o0, %g1
-	sub		%o1, %g1, %g2
+	 sub		%o1, %o0, %g2
+	srlx		%g2, 18, %g3
+	brnz,pn		%g3, 4f
+	 mov		%o0, %g1
+	sethi		%hi(PAGE_SIZE), %g3
 	sub		%g2, %g3, %g2
 1:	add		%g1, %g2, %o0	/* ARG0: virtual address */
 	mov		0, %o1		/* ARG1: mmu context */
 	mov		HV_MMU_ALL, %o2	/* ARG2: flags */
 	ta		HV_MMU_UNMAP_ADDR_TRAP
-	brnz,pn		%o0, __hypervisor_tlb_tl0_error
+	brnz,pn		%o0, 3f
 	 mov		HV_MMU_UNMAP_ADDR_TRAP, %o1
 	brnz,pt		%g2, 1b
 	 sub		%g2, %g3, %g2
 2:	retl
 	 nop
+3:	sethi		%hi(__hypervisor_tlb_tl0_error), %o2
+	jmpl		%o2 + %lo(__hypervisor_tlb_tl0_error), %g0
+	 nop
+4:	mov		0, %o0		/* ARG0: CPU lists unimplemented */
+	mov		0, %o1		/* ARG1: CPU lists unimplemented */
+	mov		0, %o2		/* ARG2: mmu context == nucleus */
+	mov		HV_MMU_ALL, %o3	/* ARG3: flags */
+	mov		HV_FAST_MMU_DEMAP_CTX, %o5
+	ta		HV_FAST_TRAP
+	brnz,pn		%o0, 3b
+	 mov		HV_FAST_MMU_DEMAP_CTX, %o1
+	retl
+	 nop
 
 #ifdef DCACHE_ALIASING_POSSIBLE
 	/* XXX Niagara and friends have an 8K cache, so no aliasing is
@@ -394,43 +511,6 @@ tlb_patch_one:
 	retl
 	 nop
 
-	.globl		cheetah_patch_cachetlbops
-cheetah_patch_cachetlbops:
-	save		%sp, -128, %sp
-
-	sethi		%hi(__flush_tlb_mm), %o0
-	or		%o0, %lo(__flush_tlb_mm), %o0
-	sethi		%hi(__cheetah_flush_tlb_mm), %o1
-	or		%o1, %lo(__cheetah_flush_tlb_mm), %o1
-	call		tlb_patch_one
-	 mov		19, %o2
-
-	sethi		%hi(__flush_tlb_page), %o0
-	or		%o0, %lo(__flush_tlb_page), %o0
-	sethi		%hi(__cheetah_flush_tlb_page), %o1
-	or		%o1, %lo(__cheetah_flush_tlb_page), %o1
-	call		tlb_patch_one
-	 mov		22, %o2
-
-	sethi		%hi(__flush_tlb_pending), %o0
-	or		%o0, %lo(__flush_tlb_pending), %o0
-	sethi		%hi(__cheetah_flush_tlb_pending), %o1
-	or		%o1, %lo(__cheetah_flush_tlb_pending), %o1
-	call		tlb_patch_one
-	 mov		27, %o2
-
-#ifdef DCACHE_ALIASING_POSSIBLE
-	sethi		%hi(__flush_dcache_page), %o0
-	or		%o0, %lo(__flush_dcache_page), %o0
-	sethi		%hi(__cheetah_flush_dcache_page), %o1
-	or		%o1, %lo(__cheetah_flush_dcache_page), %o1
-	call		tlb_patch_one
-	 mov		11, %o2
-#endif /* DCACHE_ALIASING_POSSIBLE */
-
-	ret
-	 restore
-
 #ifdef CONFIG_SMP
 	/* These are all called by the slaves of a cross call, at
 	 * trap level 1, with interrupts fully disabled.
@@ -447,7 +527,7 @@ cheetah_patch_cachetlbops:
 	 */
 	.align		32
 	.globl		xcall_flush_tlb_mm
-xcall_flush_tlb_mm:	/* 21 insns */
+xcall_flush_tlb_mm:	/* 24 insns */
 	mov		PRIMARY_CONTEXT, %g2
 	ldxa		[%g2] ASI_DMMU, %g3
 	srlx		%g3, CTX_PGSZ1_NUC_SHIFT, %g4
@@ -469,9 +549,12 @@ xcall_flush_tlb_mm:	/* 21 insns */
 	nop
 	nop
 	nop
+	nop
+	nop
+	nop
 
 	.globl		xcall_flush_tlb_page
-xcall_flush_tlb_page:	/* 17 insns */
+xcall_flush_tlb_page:	/* 20 insns */
 	/* %g5=context, %g1=vaddr */
 	mov		PRIMARY_CONTEXT, %g4
 	ldxa		[%g4] ASI_DMMU, %g2
@@ -490,15 +573,20 @@ xcall_flush_tlb_page:	/* 17 insns */
 	retry
 	nop
 	nop
+	nop
+	nop
+	nop
 
 	.globl		xcall_flush_tlb_kernel_range
-xcall_flush_tlb_kernel_range:	/* 25 insns */
+xcall_flush_tlb_kernel_range:	/* 44 insns */
 	sethi		%hi(PAGE_SIZE - 1), %g2
 	or		%g2, %lo(PAGE_SIZE - 1), %g2
 	andn		%g1, %g2, %g1
 	andn		%g7, %g2, %g7
 	sub		%g7, %g1, %g3
-	add		%g2, 1, %g2
+	srlx		%g3, 18, %g2
+	brnz,pn		%g2, 2f
+	 add		%g2, 1, %g2
 	sub		%g3, %g2, %g3
 	or		%g1, 0x20, %g1		! Nucleus
 1:	stxa		%g0, [%g1 + %g3] ASI_DMMU_DEMAP
@@ -507,8 +595,25 @@ xcall_flush_tlb_kernel_range:	/* 25 insns */
 	brnz,pt		%g3, 1b
 	 sub		%g3, %g2, %g3
 	retry
-	nop
-	nop
+2:	mov		63 * 8, %g1
+1:	ldxa		[%g1] ASI_ITLB_DATA_ACCESS, %g2
+	andcc		%g2, 0x40, %g0			/* _PAGE_L_4U */
+	bne,pn		%xcc, 2f
+	 mov		TLB_TAG_ACCESS, %g2
+	stxa		%g0, [%g2] ASI_IMMU
+	stxa		%g0, [%g1] ASI_ITLB_DATA_ACCESS
+	membar		#Sync
+2:	ldxa		[%g1] ASI_DTLB_DATA_ACCESS, %g2
+	andcc		%g2, 0x40, %g0
+	bne,pn		%xcc, 2f
+	 mov		TLB_TAG_ACCESS, %g2
+	stxa		%g0, [%g2] ASI_DMMU
+	stxa		%g0, [%g1] ASI_DTLB_DATA_ACCESS
+	membar		#Sync
+2:	sub		%g1, 8, %g1
+	brgez,pt	%g1, 1b
+	 nop
+	retry
 	nop
 	nop
 	nop
@@ -637,6 +742,52 @@ xcall_fetch_glob_pmu_n4:
 
 	retry
 
+__cheetah_xcall_flush_tlb_kernel_range:	/* 44 insns */
+	sethi		%hi(PAGE_SIZE - 1), %g2
+	or		%g2, %lo(PAGE_SIZE - 1), %g2
+	andn		%g1, %g2, %g1
+	andn		%g7, %g2, %g7
+	sub		%g7, %g1, %g3
+	srlx		%g3, 18, %g2
+	brnz,pn		%g2, 2f
+	 add		%g2, 1, %g2
+	sub		%g3, %g2, %g3
+	or		%g1, 0x20, %g1		! Nucleus
+1:	stxa		%g0, [%g1 + %g3] ASI_DMMU_DEMAP
+	stxa		%g0, [%g1 + %g3] ASI_IMMU_DEMAP
+	membar		#Sync
+	brnz,pt		%g3, 1b
+	 sub		%g3, %g2, %g3
+	retry
+2:	mov		0x80, %g2
+	stxa		%g0, [%g2] ASI_DMMU_DEMAP
+	membar		#Sync
+	stxa		%g0, [%g2] ASI_IMMU_DEMAP
+	membar		#Sync
+	retry
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+
 #ifdef DCACHE_ALIASING_POSSIBLE
 	.align		32
 	.globl		xcall_flush_dcache_page_cheetah
@@ -700,7 +851,7 @@ __hypervisor_tlb_xcall_error:
 	ba,a,pt	%xcc, rtrap
 
 	.globl		__hypervisor_xcall_flush_tlb_mm
-__hypervisor_xcall_flush_tlb_mm: /* 21 insns */
+__hypervisor_xcall_flush_tlb_mm: /* 24 insns */
 	/* %g5=ctx, g1,g2,g3,g4,g7=scratch, %g6=unusable */
 	mov		%o0, %g2
 	mov		%o1, %g3
@@ -714,7 +865,7 @@ __hypervisor_xcall_flush_tlb_mm: /* 21 insns */
 	mov		HV_FAST_MMU_DEMAP_CTX, %o5
 	ta		HV_FAST_TRAP
 	mov		HV_FAST_MMU_DEMAP_CTX, %g6
-	brnz,pn		%o0, __hypervisor_tlb_xcall_error
+	brnz,pn		%o0, 1f
 	 mov		%o0, %g5
 	mov		%g2, %o0
 	mov		%g3, %o1
@@ -723,9 +874,12 @@ __hypervisor_xcall_flush_tlb_mm: /* 21 insns */
 	mov		%g7, %o5
 	membar		#Sync
 	retry
+1:	sethi		%hi(__hypervisor_tlb_xcall_error), %g4
+	jmpl		%g4 + %lo(__hypervisor_tlb_xcall_error), %g0
+	 nop
 
 	.globl		__hypervisor_xcall_flush_tlb_page
-__hypervisor_xcall_flush_tlb_page: /* 17 insns */
+__hypervisor_xcall_flush_tlb_page: /* 20 insns */
 	/* %g5=ctx, %g1=vaddr */
 	mov		%o0, %g2
 	mov		%o1, %g3
@@ -737,42 +891,64 @@ __hypervisor_xcall_flush_tlb_page: /* 17 insns */
 	sllx		%o0, PAGE_SHIFT, %o0
 	ta		HV_MMU_UNMAP_ADDR_TRAP
 	mov		HV_MMU_UNMAP_ADDR_TRAP, %g6
-	brnz,a,pn	%o0, __hypervisor_tlb_xcall_error
+	brnz,a,pn	%o0, 1f
 	 mov		%o0, %g5
 	mov		%g2, %o0
 	mov		%g3, %o1
 	mov		%g4, %o2
 	membar		#Sync
 	retry
+1:	sethi		%hi(__hypervisor_tlb_xcall_error), %g4
+	jmpl		%g4 + %lo(__hypervisor_tlb_xcall_error), %g0
+	 nop
 
 	.globl		__hypervisor_xcall_flush_tlb_kernel_range
-__hypervisor_xcall_flush_tlb_kernel_range: /* 25 insns */
+__hypervisor_xcall_flush_tlb_kernel_range: /* 44 insns */
 	/* %g1=start, %g7=end, g2,g3,g4,g5,g6=scratch */
 	sethi		%hi(PAGE_SIZE - 1), %g2
 	or		%g2, %lo(PAGE_SIZE - 1), %g2
 	andn		%g1, %g2, %g1
 	andn		%g7, %g2, %g7
 	sub		%g7, %g1, %g3
+	srlx		%g3, 18, %g7
 	add		%g2, 1, %g2
 	sub		%g3, %g2, %g3
 	mov		%o0, %g2
 	mov		%o1, %g4
-	mov		%o2, %g7
+	brnz,pn		%g7, 2f
+	 mov		%o2, %g7
 1:	add		%g1, %g3, %o0	/* ARG0: virtual address */
 	mov		0, %o1		/* ARG1: mmu context */
 	mov		HV_MMU_ALL, %o2	/* ARG2: flags */
 	ta		HV_MMU_UNMAP_ADDR_TRAP
 	mov		HV_MMU_UNMAP_ADDR_TRAP, %g6
-	brnz,pn		%o0, __hypervisor_tlb_xcall_error
+	brnz,pn		%o0, 1f
 	 mov		%o0, %g5
 	sethi		%hi(PAGE_SIZE), %o2
 	brnz,pt		%g3, 1b
 	 sub		%g3, %o2, %g3
-	mov		%g2, %o0
+5:	mov		%g2, %o0
 	mov		%g4, %o1
 	mov		%g7, %o2
 	membar		#Sync
 	retry
+1:	sethi		%hi(__hypervisor_tlb_xcall_error), %g4
+	jmpl		%g4 + %lo(__hypervisor_tlb_xcall_error), %g0
+	 nop
+2:	mov		%o3, %g1
+	mov		%o5, %g3
+	mov		0, %o0		/* ARG0: CPU lists unimplemented */
+	mov		0, %o1		/* ARG1: CPU lists unimplemented */
+	mov		0, %o2		/* ARG2: mmu context == nucleus */
+	mov		HV_MMU_ALL, %o3	/* ARG3: flags */
+	mov		HV_FAST_MMU_DEMAP_CTX, %o5
+	ta		HV_FAST_TRAP
+	mov		%g1, %o3
+	brz,pt		%o0, 5b
+	 mov		%g3, %o5
+	mov		HV_FAST_MMU_DEMAP_CTX, %g6
+	ba,pt		%xcc, 1b
+	 clr		%g5
 
 	/* These just get rescheduled to PIL vectors. */
 	.globl		xcall_call_function
@@ -809,6 +985,58 @@ xcall_kgdb_capture:
 
 #endif /* CONFIG_SMP */
 
+	.globl		cheetah_patch_cachetlbops
+cheetah_patch_cachetlbops:
+	save		%sp, -128, %sp
+
+	sethi		%hi(__flush_tlb_mm), %o0
+	or		%o0, %lo(__flush_tlb_mm), %o0
+	sethi		%hi(__cheetah_flush_tlb_mm), %o1
+	or		%o1, %lo(__cheetah_flush_tlb_mm), %o1
+	call		tlb_patch_one
+	 mov		19, %o2
+
+	sethi		%hi(__flush_tlb_page), %o0
+	or		%o0, %lo(__flush_tlb_page), %o0
+	sethi		%hi(__cheetah_flush_tlb_page), %o1
+	or		%o1, %lo(__cheetah_flush_tlb_page), %o1
+	call		tlb_patch_one
+	 mov		22, %o2
+
+	sethi		%hi(__flush_tlb_pending), %o0
+	or		%o0, %lo(__flush_tlb_pending), %o0
+	sethi		%hi(__cheetah_flush_tlb_pending), %o1
+	or		%o1, %lo(__cheetah_flush_tlb_pending), %o1
+	call		tlb_patch_one
+	 mov		27, %o2
+
+	sethi		%hi(__flush_tlb_kernel_range), %o0
+	or		%o0, %lo(__flush_tlb_kernel_range), %o0
+	sethi		%hi(__cheetah_flush_tlb_kernel_range), %o1
+	or		%o1, %lo(__cheetah_flush_tlb_kernel_range), %o1
+	call		tlb_patch_one
+	 mov		31, %o2
+
+#ifdef DCACHE_ALIASING_POSSIBLE
+	sethi		%hi(__flush_dcache_page), %o0
+	or		%o0, %lo(__flush_dcache_page), %o0
+	sethi		%hi(__cheetah_flush_dcache_page), %o1
+	or		%o1, %lo(__cheetah_flush_dcache_page), %o1
+	call		tlb_patch_one
+	 mov		11, %o2
+#endif /* DCACHE_ALIASING_POSSIBLE */
+
+#ifdef CONFIG_SMP
+	sethi		%hi(xcall_flush_tlb_kernel_range), %o0
+	or		%o0, %lo(xcall_flush_tlb_kernel_range), %o0
+	sethi		%hi(__cheetah_xcall_flush_tlb_kernel_range), %o1
+	or		%o1, %lo(__cheetah_xcall_flush_tlb_kernel_range), %o1
+	call		tlb_patch_one
+	 mov		44, %o2
+#endif /* CONFIG_SMP */
+
+	ret
+	 restore
 
 	.globl		hypervisor_patch_cachetlbops
 hypervisor_patch_cachetlbops:
@@ -819,28 +1047,28 @@ hypervisor_patch_cachetlbops:
 	sethi		%hi(__hypervisor_flush_tlb_mm), %o1
 	or		%o1, %lo(__hypervisor_flush_tlb_mm), %o1
 	call		tlb_patch_one
-	 mov		10, %o2
+	 mov		19, %o2
 
 	sethi		%hi(__flush_tlb_page), %o0
 	or		%o0, %lo(__flush_tlb_page), %o0
 	sethi		%hi(__hypervisor_flush_tlb_page), %o1
 	or		%o1, %lo(__hypervisor_flush_tlb_page), %o1
 	call		tlb_patch_one
-	 mov		11, %o2
+	 mov		22, %o2
 
 	sethi		%hi(__flush_tlb_pending), %o0
 	or		%o0, %lo(__flush_tlb_pending), %o0
 	sethi		%hi(__hypervisor_flush_tlb_pending), %o1
 	or		%o1, %lo(__hypervisor_flush_tlb_pending), %o1
 	call		tlb_patch_one
-	 mov		16, %o2
+	 mov		27, %o2
 
 	sethi		%hi(__flush_tlb_kernel_range), %o0
 	or		%o0, %lo(__flush_tlb_kernel_range), %o0
 	sethi		%hi(__hypervisor_flush_tlb_kernel_range), %o1
 	or		%o1, %lo(__hypervisor_flush_tlb_kernel_range), %o1
 	call		tlb_patch_one
-	 mov		16, %o2
+	 mov		31, %o2
 
 #ifdef DCACHE_ALIASING_POSSIBLE
 	sethi		%hi(__flush_dcache_page), %o0
@@ -857,21 +1085,21 @@ hypervisor_patch_cachetlbops:
 	sethi		%hi(__hypervisor_xcall_flush_tlb_mm), %o1
 	or		%o1, %lo(__hypervisor_xcall_flush_tlb_mm), %o1
 	call		tlb_patch_one
-	 mov		21, %o2
+	 mov		24, %o2
 
 	sethi		%hi(xcall_flush_tlb_page), %o0
 	or		%o0, %lo(xcall_flush_tlb_page), %o0
 	sethi		%hi(__hypervisor_xcall_flush_tlb_page), %o1
 	or		%o1, %lo(__hypervisor_xcall_flush_tlb_page), %o1
 	call		tlb_patch_one
-	 mov		17, %o2
+	 mov		20, %o2
 
 	sethi		%hi(xcall_flush_tlb_kernel_range), %o0
 	or		%o0, %lo(xcall_flush_tlb_kernel_range), %o0
 	sethi		%hi(__hypervisor_xcall_flush_tlb_kernel_range), %o1
 	or		%o1, %lo(__hypervisor_xcall_flush_tlb_kernel_range), %o1
 	call		tlb_patch_one
-	 mov		25, %o2
+	 mov		44, %o2
 #endif /* CONFIG_SMP */
 
 	ret
diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c
index c4751ece76f6..45e87c9cc828 100644
--- a/drivers/net/ethernet/broadcom/bgmac.c
+++ b/drivers/net/ethernet/broadcom/bgmac.c
@@ -307,6 +307,10 @@ static void bgmac_dma_rx_enable(struct bgmac *bgmac,
 	u32 ctl;
 
 	ctl = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_RX_CTL);
+
+	/* preserve ONLY bits 16-17 from current hardware value */
+	ctl &= BGMAC_DMA_RX_ADDREXT_MASK;
+
 	if (bgmac->feature_flags & BGMAC_FEAT_RX_MASK_SETUP) {
 		ctl &= ~BGMAC_DMA_RX_BL_MASK;
 		ctl |= BGMAC_DMA_RX_BL_128 << BGMAC_DMA_RX_BL_SHIFT;
@@ -317,7 +321,6 @@ static void bgmac_dma_rx_enable(struct bgmac *bgmac,
 		ctl &= ~BGMAC_DMA_RX_PT_MASK;
 		ctl |= BGMAC_DMA_RX_PT_1 << BGMAC_DMA_RX_PT_SHIFT;
 	}
-	ctl &= BGMAC_DMA_RX_ADDREXT_MASK;
 	ctl |= BGMAC_DMA_RX_ENABLE;
 	ctl |= BGMAC_DMA_RX_PARITY_DISABLE;
 	ctl |= BGMAC_DMA_RX_OVERFLOW_CONT;
diff --git a/drivers/net/ethernet/broadcom/bnx2.c b/drivers/net/ethernet/broadcom/bnx2.c
index 505ceaf451e2..2c850a92ab15 100644
--- a/drivers/net/ethernet/broadcom/bnx2.c
+++ b/drivers/net/ethernet/broadcom/bnx2.c
@@ -49,6 +49,7 @@
 #include <linux/firmware.h>
 #include <linux/log2.h>
 #include <linux/aer.h>
+#include <linux/crash_dump.h>
 
 #if defined(CONFIG_CNIC) || defined(CONFIG_CNIC_MODULE)
 #define BCM_CNIC 1
@@ -4759,15 +4760,16 @@ bnx2_setup_msix_tbl(struct bnx2 *bp)
 	BNX2_WR(bp, BNX2_PCI_GRC_WINDOW3_ADDR, BNX2_MSIX_PBA_ADDR);
 }
 
-static int
-bnx2_reset_chip(struct bnx2 *bp, u32 reset_code)
+static void
+bnx2_wait_dma_complete(struct bnx2 *bp)
 {
 	u32 val;
-	int i, rc = 0;
-	u8 old_port;
+	int i;
 
-	/* Wait for the current PCI transaction to complete before
-	 * issuing a reset. */
+	/*
+	 * Wait for the current PCI transaction to complete before
+	 * issuing a reset.
+	 */
 	if ((BNX2_CHIP(bp) == BNX2_CHIP_5706) ||
 	    (BNX2_CHIP(bp) == BNX2_CHIP_5708)) {
 		BNX2_WR(bp, BNX2_MISC_ENABLE_CLR_BITS,
@@ -4791,6 +4793,21 @@ bnx2_reset_chip(struct bnx2 *bp, u32 reset_code)
 		}
 	}
 
+	return;
+}
+
+
+static int
+bnx2_reset_chip(struct bnx2 *bp, u32 reset_code)
+{
+	u32 val;
+	int i, rc = 0;
+	u8 old_port;
+
+	/* Wait for the current PCI transaction to complete before
+	 * issuing a reset. */
+	bnx2_wait_dma_complete(bp);
+
 	/* Wait for the firmware to tell us it is ok to issue a reset. */
 	bnx2_fw_sync(bp, BNX2_DRV_MSG_DATA_WAIT0 | reset_code, 1, 1);
 
@@ -6356,6 +6373,10 @@ bnx2_open(struct net_device *dev)
 	struct bnx2 *bp = netdev_priv(dev);
 	int rc;
 
+	rc = bnx2_request_firmware(bp);
+	if (rc < 0)
+		goto out;
+
 	netif_carrier_off(dev);
 
 	bnx2_disable_int(bp);
@@ -6424,6 +6445,7 @@ open_err:
 	bnx2_free_irq(bp);
 	bnx2_free_mem(bp);
 	bnx2_del_napi(bp);
+	bnx2_release_firmware(bp);
 	goto out;
 }
 
@@ -8570,12 +8592,15 @@ bnx2_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	pci_set_drvdata(pdev, dev);
 
-	rc = bnx2_request_firmware(bp);
-	if (rc < 0)
-		goto error;
-
+	/*
+	 * In-flight DMA from 1st kernel could continue going in kdump kernel.
+	 * New io-page table has been created before bnx2 does reset at open stage.
+	 * We have to wait for the in-flight DMA to complete to avoid it look up
+	 * into the newly created io-page table.
+	 */
+	if (is_kdump_kernel())
+		bnx2_wait_dma_complete(bp);
 
-	bnx2_reset_chip(bp, BNX2_DRV_MSG_CODE_RESET);
 	memcpy(dev->dev_addr, bp->mac_addr, ETH_ALEN);
 
 	dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_SG |
@@ -8608,7 +8633,6 @@ bnx2_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	return 0;
 
 error:
-	bnx2_release_firmware(bp);
 	pci_iounmap(pdev, bp->regview);
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index d48873bcbddf..5cdc96bdd444 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -231,7 +231,7 @@ mlxsw_sp_span_entry_create(struct mlxsw_sp_port *port)
 
 	span_entry->used = true;
 	span_entry->id = index;
-	span_entry->ref_count = 0;
+	span_entry->ref_count = 1;
 	span_entry->local_port = local_port;
 	return span_entry;
 }
@@ -268,6 +268,7 @@ struct mlxsw_sp_span_entry *mlxsw_sp_span_entry_get(struct mlxsw_sp_port *port)
 
 	span_entry = mlxsw_sp_span_entry_find(port);
 	if (span_entry) {
+		/* Already exists, just take a reference */
 		span_entry->ref_count++;
 		return span_entry;
 	}
@@ -278,6 +279,7 @@ struct mlxsw_sp_span_entry *mlxsw_sp_span_entry_get(struct mlxsw_sp_port *port)
 static int mlxsw_sp_span_entry_put(struct mlxsw_sp *mlxsw_sp,
 				   struct mlxsw_sp_span_entry *span_entry)
 {
+	WARN_ON(!span_entry->ref_count);
 	if (--span_entry->ref_count == 0)
 		mlxsw_sp_span_entry_destroy(mlxsw_sp, span_entry);
 	return 0;
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index 3f5c51da6d3e..62514b9bf988 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -777,6 +777,26 @@ static void mlxsw_sp_router_neigh_rec_process(struct mlxsw_sp *mlxsw_sp,
 	}
 }
 
+static bool mlxsw_sp_router_rauhtd_is_full(char *rauhtd_pl)
+{
+	u8 num_rec, last_rec_index, num_entries;
+
+	num_rec = mlxsw_reg_rauhtd_num_rec_get(rauhtd_pl);
+	last_rec_index = num_rec - 1;
+
+	if (num_rec < MLXSW_REG_RAUHTD_REC_MAX_NUM)
+		return false;
+	if (mlxsw_reg_rauhtd_rec_type_get(rauhtd_pl, last_rec_index) ==
+	    MLXSW_REG_RAUHTD_TYPE_IPV6)
+		return true;
+
+	num_entries = mlxsw_reg_rauhtd_ipv4_rec_num_entries_get(rauhtd_pl,
+								last_rec_index);
+	if (++num_entries == MLXSW_REG_RAUHTD_IPV4_ENT_PER_REC)
+		return true;
+	return false;
+}
+
 static int mlxsw_sp_router_neighs_update_rauhtd(struct mlxsw_sp *mlxsw_sp)
 {
 	char *rauhtd_pl;
@@ -803,7 +823,7 @@ static int mlxsw_sp_router_neighs_update_rauhtd(struct mlxsw_sp *mlxsw_sp)
 		for (i = 0; i < num_rec; i++)
 			mlxsw_sp_router_neigh_rec_process(mlxsw_sp, rauhtd_pl,
 							  i);
-	} while (num_rec);
+	} while (mlxsw_sp_router_rauhtd_is_full(rauhtd_pl));
 	rtnl_unlock();
 
 	kfree(rauhtd_pl);
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 4c8c60af7985..fe9e7b1979b8 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -871,6 +871,13 @@ static int stmmac_init_phy(struct net_device *dev)
 		return -ENODEV;
 	}
 
+	/* stmmac_adjust_link will change this to PHY_IGNORE_INTERRUPT to avoid
+	 * subsequent PHY polling, make sure we force a link transition if
+	 * we have a UP/DOWN/UP transition
+	 */
+	if (phydev->is_pseudo_fixed_link)
+		phydev->irq = PHY_POLL;
+
 	pr_debug("stmmac_init_phy:  %s: attached to PHY (UID 0x%x)"
 		 " Link = %d\n", dev->name, phydev->phy_id, phydev->link);
 
diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c
index 5c8429f23a89..3a5530d0511b 100644
--- a/drivers/usb/gadget/function/f_fs.c
+++ b/drivers/usb/gadget/function/f_fs.c
@@ -133,8 +133,60 @@ struct ffs_epfile {
 	/*
 	 * Buffer for holding data from partial reads which may happen since
 	 * weâre rounding user read requests to a multiple of a max packet size.
+	 *
+	 * The pointer is initialised with NULL value and may be set by
+	 * __ffs_epfile_read_data function to point to a temporary buffer.
+	 *
+	 * In normal operation, calls to __ffs_epfile_read_buffered will consume
+	 * data from said buffer and eventually free it.  Importantly, while the
+	 * function is using the buffer, it sets the pointer to NULL.  This is
+	 * all right since __ffs_epfile_read_data and __ffs_epfile_read_buffered
+	 * can never run concurrently (they are synchronised by epfile->mutex)
+	 * so the latter will not assign a new value to the pointer.
+	 *
+	 * Meanwhile ffs_func_eps_disable frees the buffer (if the pointer is
+	 * valid) and sets the pointer to READ_BUFFER_DROP value.  This special
+	 * value is crux of the synchronisation between ffs_func_eps_disable and
+	 * __ffs_epfile_read_data.
+	 *
+	 * Once __ffs_epfile_read_data is about to finish it will try to set the
+	 * pointer back to its old value (as described above), but seeing as the
+	 * pointer is not-NULL (namely READ_BUFFER_DROP) it will instead free
+	 * the buffer.
+	 *
+	 * == State transitions ==
+	 *
+	 * â ptr == NULL:  (initial state)
+	 *   â __ffs_epfile_read_buffer_free: go to ptr == DROP
+	 *   â __ffs_epfile_read_buffered:    nop
+	 *   â __ffs_epfile_read_data allocates temp buffer: go to ptr == buf
+	 *   â reading finishes:              n/a, not in âand readingâ state
+	 * â ptr == DROP:
+	 *   â __ffs_epfile_read_buffer_free: nop
+	 *   â __ffs_epfile_read_buffered:    go to ptr == NULL
+	 *   â __ffs_epfile_read_data allocates temp buffer: free buf, nop
+	 *   â reading finishes:              n/a, not in âand readingâ state
+	 * â ptr == buf:
+	 *   â __ffs_epfile_read_buffer_free: free buf, go to ptr == DROP
+	 *   â __ffs_epfile_read_buffered:    go to ptr == NULL and reading
+	 *   â __ffs_epfile_read_data:        n/a, __ffs_epfile_read_buffered
+	 *                                    is always called first
+	 *   â reading finishes:              n/a, not in âand readingâ state
+	 * â ptr == NULL and reading:
+	 *   â __ffs_epfile_read_buffer_free: go to ptr == DROP and reading
+	 *   â __ffs_epfile_read_buffered:    n/a, mutex is held
+	 *   â __ffs_epfile_read_data:        n/a, mutex is held
+	 *   â reading finishes and â
+	 *     â all data read:               free buf, go to ptr == NULL
+	 *     â otherwise:                   go to ptr == buf and reading
+	 * â ptr == DROP and reading:
+	 *   â __ffs_epfile_read_buffer_free: nop
+	 *   â __ffs_epfile_read_buffered:    n/a, mutex is held
+	 *   â __ffs_epfile_read_data:        n/a, mutex is held
+	 *   â reading finishes:              free buf, go to ptr == DROP
 	 */
-	struct ffs_buffer		*read_buffer;	/* P: epfile->mutex */
+	struct ffs_buffer		*read_buffer;
+#define READ_BUFFER_DROP ((struct ffs_buffer *)ERR_PTR(-ESHUTDOWN))
 
 	char				name[5];
 
@@ -733,25 +785,47 @@ static void ffs_epfile_async_io_complete(struct usb_ep *_ep,
 	schedule_work(&io_data->work);
 }
 
+static void __ffs_epfile_read_buffer_free(struct ffs_epfile *epfile)
+{
+	/*
+	 * See comment in struct ffs_epfile for full read_buffer pointer
+	 * synchronisation story.
+	 */
+	struct ffs_buffer *buf = xchg(&epfile->read_buffer, READ_BUFFER_DROP);
+	if (buf && buf != READ_BUFFER_DROP)
+		kfree(buf);
+}
+
 /* Assumes epfile->mutex is held. */
 static ssize_t __ffs_epfile_read_buffered(struct ffs_epfile *epfile,
 					  struct iov_iter *iter)
 {
-	struct ffs_buffer *buf = epfile->read_buffer;
+	/*
+	 * Null out epfile->read_buffer so ffs_func_eps_disable does not free
+	 * the buffer while we are using it.  See comment in struct ffs_epfile
+	 * for full read_buffer pointer synchronisation story.
+	 */
+	struct ffs_buffer *buf = xchg(&epfile->read_buffer, NULL);
 	ssize_t ret;
-	if (!buf)
+	if (!buf || buf == READ_BUFFER_DROP)
 		return 0;
 
 	ret = copy_to_iter(buf->data, buf->length, iter);
 	if (buf->length == ret) {
 		kfree(buf);
-		epfile->read_buffer = NULL;
-	} else if (unlikely(iov_iter_count(iter))) {
+		return ret;
+	}
+
+	if (unlikely(iov_iter_count(iter))) {
 		ret = -EFAULT;
 	} else {
 		buf->length -= ret;
 		buf->data += ret;
 	}
+
+	if (cmpxchg(&epfile->read_buffer, NULL, buf))
+		kfree(buf);
+
 	return ret;
 }
 
@@ -780,7 +854,15 @@ static ssize_t __ffs_epfile_read_data(struct ffs_epfile *epfile,
 	buf->length = data_len;
 	buf->data = buf->storage;
 	memcpy(buf->storage, data + ret, data_len);
-	epfile->read_buffer = buf;
+
+	/*
+	 * At this point read_buffer is NULL or READ_BUFFER_DROP (if
+	 * ffs_func_eps_disable has been called in the meanwhile).  See comment
+	 * in struct ffs_epfile for full read_buffer pointer synchronisation
+	 * story.
+	 */
+	if (unlikely(cmpxchg(&epfile->read_buffer, NULL, buf)))
+		kfree(buf);
 
 	return ret;
 }
@@ -1094,8 +1176,7 @@ ffs_epfile_release(struct inode *inode, struct file *file)
 
 	ENTER();
 
-	kfree(epfile->read_buffer);
-	epfile->read_buffer = NULL;
+	__ffs_epfile_read_buffer_free(epfile);
 	ffs_data_closed(epfile->ffs);
 
 	return 0;
@@ -1721,24 +1802,20 @@ static void ffs_func_eps_disable(struct ffs_function *func)
 	unsigned count            = func->ffs->eps_count;
 	unsigned long flags;
 
+	spin_lock_irqsave(&func->ffs->eps_lock, flags);
 	do {
-		if (epfile)
-			mutex_lock(&epfile->mutex);
-		spin_lock_irqsave(&func->ffs->eps_lock, flags);
 		/* pending requests get nuked */
 		if (likely(ep->ep))
 			usb_ep_disable(ep->ep);
 		++ep;
-		spin_unlock_irqrestore(&func->ffs->eps_lock, flags);
 
 		if (epfile) {
 			epfile->ep = NULL;
-			kfree(epfile->read_buffer);
-			epfile->read_buffer = NULL;
-			mutex_unlock(&epfile->mutex);
+			__ffs_epfile_read_buffer_free(epfile);
 			++epfile;
 		}
 	} while (--count);
+	spin_unlock_irqrestore(&func->ffs->eps_lock, flags);
 }
 
 static int ffs_func_eps_enable(struct ffs_function *func)
diff --git a/include/net/ip.h b/include/net/ip.h
index 156b0c11b524..0ccf6daf6f56 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -47,7 +47,6 @@ struct inet_skb_parm {
 #define IPSKB_REROUTED		BIT(4)
 #define IPSKB_DOREDIRECT	BIT(5)
 #define IPSKB_FRAG_PMTU		BIT(6)
-#define IPSKB_FRAG_SEGS		BIT(7)
 
 	u16			frag_max_size;
 };
diff --git a/include/net/ip6_tunnel.h b/include/net/ip6_tunnel.h
index 43a5a0e4524c..b01d5d1d7439 100644
--- a/include/net/ip6_tunnel.h
+++ b/include/net/ip6_tunnel.h
@@ -145,6 +145,7 @@ static inline void ip6tunnel_xmit(struct sock *sk, struct sk_buff *skb,
 {
 	int pkt_len, err;
 
+	memset(skb->cb, 0, sizeof(struct inet6_skb_parm));
 	pkt_len = skb->len - skb_inner_network_offset(skb);
 	err = ip6_local_out(dev_net(skb_dst(skb)->dev), sk, skb);
 	if (unlikely(net_xmit_eval(err)))
diff --git a/include/net/sock.h b/include/net/sock.h
index 8741988e6880..c26eab962ec7 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1587,11 +1587,11 @@ static inline void sock_put(struct sock *sk)
 void sock_gen_put(struct sock *sk);
 
 int __sk_receive_skb(struct sock *sk, struct sk_buff *skb, const int nested,
-		     unsigned int trim_cap);
+		     unsigned int trim_cap, bool refcounted);
 static inline int sk_receive_skb(struct sock *sk, struct sk_buff *skb,
 				 const int nested)
 {
-	return __sk_receive_skb(sk, skb, nested, 1);
+	return __sk_receive_skb(sk, skb, nested, 1, true);
 }
 
 static inline void sk_tx_queue_set(struct sock *sk, int tx_queue)
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 7717302cab91..0de698940793 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1164,6 +1164,7 @@ static inline void tcp_prequeue_init(struct tcp_sock *tp)
 }
 
 bool tcp_prequeue(struct sock *sk, struct sk_buff *skb);
+int tcp_filter(struct sock *sk, struct sk_buff *skb);
 
 #undef STATE_TRACE
 
diff --git a/include/uapi/linux/atm_zatm.h b/include/uapi/linux/atm_zatm.h
index 5cd4d4d2dd1d..9c9c6ad55f14 100644
--- a/include/uapi/linux/atm_zatm.h
+++ b/include/uapi/linux/atm_zatm.h
@@ -14,7 +14,6 @@
 
 #include <linux/atmapi.h>
 #include <linux/atmioc.h>
-#include <linux/time.h>
 
 #define ZATM_GETPOOL	_IOW('a',ATMIOC_SARPRV+1,struct atmif_sioc)
 						/* get pool statistics */
diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index 570eeca7bdfa..ad1bc67aff1b 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -687,7 +687,8 @@ static void delete_all_elements(struct bpf_htab *htab)
 
 		hlist_for_each_entry_safe(l, n, head, hash_node) {
 			hlist_del_rcu(&l->hash_node);
-			htab_elem_free(htab, l);
+			if (l->state != HTAB_EXTRA_ELEM_USED)
+				htab_elem_free(htab, l);
 		}
 	}
 }
diff --git a/net/core/dev.c b/net/core/dev.c
index 44b3ba462ba1..9ce9d7284ea7 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2484,7 +2484,7 @@ int skb_checksum_help(struct sk_buff *skb)
 			goto out;
 	}
 
-	*(__sum16 *)(skb->data + offset) = csum_fold(csum);
+	*(__sum16 *)(skb->data + offset) = csum_fold(csum) ?: CSUM_MANGLED_0;
 out_set_summed:
 	skb->ip_summed = CHECKSUM_NONE;
 out:
diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index 52742a02814f..5550a86f7264 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -118,7 +118,7 @@ bool __skb_flow_dissect(const struct sk_buff *skb,
 	struct flow_dissector_key_tags *key_tags;
 	struct flow_dissector_key_keyid *key_keyid;
 	u8 ip_proto = 0;
-	bool ret = false;
+	bool ret;
 
 	if (!data) {
 		data = skb->data;
@@ -481,12 +481,17 @@ ip_proto_again:
 out_good:
 	ret = true;
 
-out_bad:
+	key_control->thoff = (u16)nhoff;
+out:
 	key_basic->n_proto = proto;
 	key_basic->ip_proto = ip_proto;
-	key_control->thoff = (u16)nhoff;
 
 	return ret;
+
+out_bad:
+	ret = false;
+	key_control->thoff = min_t(u16, nhoff, skb ? skb->len : hlen);
+	goto out;
 }
 EXPORT_SYMBOL(__skb_flow_dissect);
 
diff --git a/net/core/sock.c b/net/core/sock.c
index fd7b41edf1ce..10acaccca5c8 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -453,7 +453,7 @@ int sock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
 EXPORT_SYMBOL(sock_queue_rcv_skb);
 
 int __sk_receive_skb(struct sock *sk, struct sk_buff *skb,
-		     const int nested, unsigned int trim_cap)
+		     const int nested, unsigned int trim_cap, bool refcounted)
 {
 	int rc = NET_RX_SUCCESS;
 
@@ -487,7 +487,8 @@ int __sk_receive_skb(struct sock *sk, struct sk_buff *skb,
 
 	bh_unlock_sock(sk);
 out:
-	sock_put(sk);
+	if (refcounted)
+		sock_put(sk);
 	return rc;
 discard_and_relse:
 	kfree_skb(skb);
@@ -1563,6 +1564,7 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority)
 		RCU_INIT_POINTER(newsk->sk_reuseport_cb, NULL);
 
 		newsk->sk_err	   = 0;
+		newsk->sk_err_soft = 0;
 		newsk->sk_priority = 0;
 		newsk->sk_incoming_cpu = raw_smp_processor_id();
 		atomic64_set(&newsk->sk_cookie, 0);
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 345a3aeb8c7e..b567c8725aea 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -235,7 +235,7 @@ static void dccp_v4_err(struct sk_buff *skb, u32 info)
 {
 	const struct iphdr *iph = (struct iphdr *)skb->data;
 	const u8 offset = iph->ihl << 2;
-	const struct dccp_hdr *dh = (struct dccp_hdr *)(skb->data + offset);
+	const struct dccp_hdr *dh;
 	struct dccp_sock *dp;
 	struct inet_sock *inet;
 	const int type = icmp_hdr(skb)->type;
@@ -245,11 +245,13 @@ static void dccp_v4_err(struct sk_buff *skb, u32 info)
 	int err;
 	struct net *net = dev_net(skb->dev);
 
-	if (skb->len < offset + sizeof(*dh) ||
-	    skb->len < offset + __dccp_basic_hdr_len(dh)) {
-		__ICMP_INC_STATS(net, ICMP_MIB_INERRORS);
-		return;
-	}
+	/* Only need dccph_dport & dccph_sport which are the first
+	 * 4 bytes in dccp header.
+	 * Our caller (icmp_socket_deliver()) already pulled 8 bytes for us.
+	 */
+	BUILD_BUG_ON(offsetofend(struct dccp_hdr, dccph_sport) > 8);
+	BUILD_BUG_ON(offsetofend(struct dccp_hdr, dccph_dport) > 8);
+	dh = (struct dccp_hdr *)(skb->data + offset);
 
 	sk = __inet_lookup_established(net, &dccp_hashinfo,
 				       iph->daddr, dh->dccph_dport,
@@ -868,7 +870,7 @@ lookup:
 		goto discard_and_relse;
 	nf_reset(skb);
 
-	return __sk_receive_skb(sk, skb, 1, dh->dccph_doff * 4);
+	return __sk_receive_skb(sk, skb, 1, dh->dccph_doff * 4, refcounted);
 
 no_dccp_socket:
 	if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 3828f94b234c..715e5d1dc107 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -70,7 +70,7 @@ static void dccp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 			u8 type, u8 code, int offset, __be32 info)
 {
 	const struct ipv6hdr *hdr = (const struct ipv6hdr *)skb->data;
-	const struct dccp_hdr *dh = (struct dccp_hdr *)(skb->data + offset);
+	const struct dccp_hdr *dh;
 	struct dccp_sock *dp;
 	struct ipv6_pinfo *np;
 	struct sock *sk;
@@ -78,12 +78,13 @@ static void dccp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	__u64 seq;
 	struct net *net = dev_net(skb->dev);
 
-	if (skb->len < offset + sizeof(*dh) ||
-	    skb->len < offset + __dccp_basic_hdr_len(dh)) {
-		__ICMP6_INC_STATS(net, __in6_dev_get(skb->dev),
-				  ICMP6_MIB_INERRORS);
-		return;
-	}
+	/* Only need dccph_dport & dccph_sport which are the first
+	 * 4 bytes in dccp header.
+	 * Our caller (icmpv6_notify()) already pulled 8 bytes for us.
+	 */
+	BUILD_BUG_ON(offsetofend(struct dccp_hdr, dccph_sport) > 8);
+	BUILD_BUG_ON(offsetofend(struct dccp_hdr, dccph_dport) > 8);
+	dh = (struct dccp_hdr *)(skb->data + offset);
 
 	sk = __inet6_lookup_established(net, &dccp_hashinfo,
 					&hdr->daddr, dh->dccph_dport,
@@ -738,7 +739,8 @@ lookup:
 	if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb))
 		goto discard_and_relse;
 
-	return __sk_receive_skb(sk, skb, 1, dh->dccph_doff * 4) ? -1 : 0;
+	return __sk_receive_skb(sk, skb, 1, dh->dccph_doff * 4,
+				refcounted) ? -1 : 0;
 
 no_dccp_socket:
 	if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb))
@@ -956,6 +958,7 @@ static const struct inet_connection_sock_af_ops dccp_ipv6_mapped = {
 	.getsockopt	   = ipv6_getsockopt,
 	.addr2sockaddr	   = inet6_csk_addr2sockaddr,
 	.sockaddr_len	   = sizeof(struct sockaddr_in6),
+	.bind_conflict	   = inet6_csk_bind_conflict,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt = compat_ipv6_setsockopt,
 	.compat_getsockopt = compat_ipv6_getsockopt,
diff --git a/net/dccp/proto.c b/net/dccp/proto.c
index 41e65804ddf5..9fe25bf63296 100644
--- a/net/dccp/proto.c
+++ b/net/dccp/proto.c
@@ -1009,6 +1009,10 @@ void dccp_close(struct sock *sk, long timeout)
 		__kfree_skb(skb);
 	}
 
+	/* If socket has been already reset kill it. */
+	if (sk->sk_state == DCCP_CLOSED)
+		goto adjudge_to_death;
+
 	if (data_was_unread) {
 		/* Unread data was tossed, send an appropriate Reset Code */
 		DCCP_WARN("ABORT with %u bytes unread\n", data_was_unread);
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index e2ffc2a5c7db..7ef703102dca 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -2455,22 +2455,19 @@ static struct key_vector *fib_route_get_idx(struct fib_route_iter *iter,
 	struct key_vector *l, **tp = &iter->tnode;
 	t_key key;
 
-	/* use cache location of next-to-find key */
+	/* use cached location of previously found key */
 	if (iter->pos > 0 && pos >= iter->pos) {
-		pos -= iter->pos;
 		key = iter->key;
 	} else {
-		iter->pos = 0;
+		iter->pos = 1;
 		key = 0;
 	}
 
-	while ((l = leaf_walk_rcu(tp, key)) != NULL) {
+	pos -= iter->pos;
+
+	while ((l = leaf_walk_rcu(tp, key)) && (pos-- > 0)) {
 		key = l->key + 1;
 		iter->pos++;
-
-		if (--pos <= 0)
-			break;
-
 		l = NULL;
 
 		/* handle unlikely case of a key wrap */
@@ -2479,7 +2476,7 @@ static struct key_vector *fib_route_get_idx(struct fib_route_iter *iter,
 	}
 
 	if (l)
-		iter->key = key;	/* remember it */
+		iter->key = l->key;	/* remember it */
 	else
 		iter->pos = 0;		/* forget it */
 
@@ -2507,7 +2504,7 @@ static void *fib_route_seq_start(struct seq_file *seq, loff_t *pos)
 		return fib_route_get_idx(iter, *pos);
 
 	iter->pos = 0;
-	iter->key = 0;
+	iter->key = KEY_MAX;
 
 	return SEQ_START_TOKEN;
 }
@@ -2516,7 +2513,7 @@ static void *fib_route_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 {
 	struct fib_route_iter *iter = seq->private;
 	struct key_vector *l = NULL;
-	t_key key = iter->key;
+	t_key key = iter->key + 1;
 
 	++*pos;
 
@@ -2525,7 +2522,7 @@ static void *fib_route_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 		l = leaf_walk_rcu(&iter->tnode, key);
 
 	if (l) {
-		iter->key = l->key + 1;
+		iter->key = l->key;
 		iter->pos++;
 	} else {
 		iter->pos = 0;
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 38abe70e595f..48734ee6293f 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -477,7 +477,7 @@ static struct rtable *icmp_route_lookup(struct net *net,
 	fl4->flowi4_proto = IPPROTO_ICMP;
 	fl4->fl4_icmp_type = type;
 	fl4->fl4_icmp_code = code;
-	fl4->flowi4_oif = l3mdev_master_ifindex(skb_in->dev);
+	fl4->flowi4_oif = l3mdev_master_ifindex(skb_dst(skb_in)->dev);
 
 	security_skb_classify_flow(skb_in, flowi4_to_flowi(fl4));
 	rt = __ip_route_output_key_hash(net, fl4,
@@ -502,7 +502,7 @@ static struct rtable *icmp_route_lookup(struct net *net,
 	if (err)
 		goto relookup_failed;
 
-	if (inet_addr_type_dev_table(net, skb_in->dev,
+	if (inet_addr_type_dev_table(net, skb_dst(skb_in)->dev,
 				     fl4_dec.saddr) == RTN_LOCAL) {
 		rt2 = __ip_route_output_key(net, &fl4_dec);
 		if (IS_ERR(rt2))
diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c
index 8b4ffd216839..9f0a7b96646f 100644
--- a/net/ipv4/ip_forward.c
+++ b/net/ipv4/ip_forward.c
@@ -117,7 +117,7 @@ int ip_forward(struct sk_buff *skb)
 	if (opt->is_strictroute && rt->rt_uses_gateway)
 		goto sr_failed;
 
-	IPCB(skb)->flags |= IPSKB_FORWARDED | IPSKB_FRAG_SEGS;
+	IPCB(skb)->flags |= IPSKB_FORWARDED;
 	mtu = ip_dst_mtu_maybe_forward(&rt->dst, true);
 	if (ip_exceeds_mtu(skb, mtu)) {
 		IP_INC_STATS(net, IPSTATS_MIB_FRAGFAILS);
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index dde37fb340bf..307daed9a4b9 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -223,11 +223,9 @@ static int ip_finish_output_gso(struct net *net, struct sock *sk,
 	struct sk_buff *segs;
 	int ret = 0;
 
-	/* common case: fragmentation of segments is not allowed,
-	 * or seglen is <= mtu
+	/* common case: seglen is <= mtu
 	 */
-	if (((IPCB(skb)->flags & IPSKB_FRAG_SEGS) == 0) ||
-	      skb_gso_validate_mtu(skb, mtu))
+	if (skb_gso_validate_mtu(skb, mtu))
 		return ip_finish_output2(net, sk, skb);
 
 	/* Slowpath -  GSO segment length is exceeding the dst MTU.
diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c
index 0f227db0e9ac..afd6b5968caf 100644
--- a/net/ipv4/ip_tunnel_core.c
+++ b/net/ipv4/ip_tunnel_core.c
@@ -63,7 +63,6 @@ void iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb,
 	int pkt_len = skb->len - skb_inner_network_offset(skb);
 	struct net *net = dev_net(rt->dst.dev);
 	struct net_device *dev = skb->dev;
-	int skb_iif = skb->skb_iif;
 	struct iphdr *iph;
 	int err;
 
@@ -73,16 +72,6 @@ void iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb,
 	skb_dst_set(skb, &rt->dst);
 	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
 
-	if (skb_iif && !(df & htons(IP_DF))) {
-		/* Arrived from an ingress interface, got encapsulated, with
-		 * fragmentation of encapulating frames allowed.
-		 * If skb is gso, the resulting encapsulated network segments
-		 * may exceed dst mtu.
-		 * Allow IP Fragmentation of segments.
-		 */
-		IPCB(skb)->flags |= IPSKB_FRAG_SEGS;
-	}
-
 	/* Push down and install the IP header. */
 	skb_push(skb, sizeof(struct iphdr));
 	skb_reset_network_header(skb);
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 5f006e13de56..27089f5ebbb1 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -1749,7 +1749,7 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt,
 		vif->dev->stats.tx_bytes += skb->len;
 	}
 
-	IPCB(skb)->flags |= IPSKB_FORWARDED | IPSKB_FRAG_SEGS;
+	IPCB(skb)->flags |= IPSKB_FORWARDED;
 
 	/* RFC1584 teaches, that DVMRP/PIM router must deliver packets locally
 	 * not only before forwarding, but after forwarding on all output
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 62c3ed0b7556..2f23ef1a8486 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -753,7 +753,9 @@ static void __ip_do_redirect(struct rtable *rt, struct sk_buff *skb, struct flow
 			goto reject_redirect;
 	}
 
-	n = ipv4_neigh_lookup(&rt->dst, NULL, &new_gw);
+	n = __ipv4_neigh_lookup(rt->dst.dev, new_gw);
+	if (!n)
+		n = neigh_create(&arp_tbl, &new_gw, rt->dst.dev);
 	if (!IS_ERR(n)) {
 		if (!(n->nud_state & NUD_VALID)) {
 			neigh_event_send(n, NULL);
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index ffbb218de520..c876f5ddc86c 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1145,7 +1145,7 @@ restart:
 
 	err = -EPIPE;
 	if (sk->sk_err || (sk->sk_shutdown & SEND_SHUTDOWN))
-		goto out_err;
+		goto do_error;
 
 	sg = !!(sk->sk_route_caps & NETIF_F_SG);
 
@@ -1219,7 +1219,7 @@ new_segment:
 
 			if (!skb_can_coalesce(skb, i, pfrag->page,
 					      pfrag->offset)) {
-				if (i == sysctl_max_skb_frags || !sg) {
+				if (i >= sysctl_max_skb_frags || !sg) {
 					tcp_mark_push(tp, skb);
 					goto new_segment;
 				}
diff --git a/net/ipv4/tcp_dctcp.c b/net/ipv4/tcp_dctcp.c
index 10d728b6804c..ab37c6775630 100644
--- a/net/ipv4/tcp_dctcp.c
+++ b/net/ipv4/tcp_dctcp.c
@@ -56,6 +56,7 @@ struct dctcp {
 	u32 next_seq;
 	u32 ce_state;
 	u32 delayed_ack_reserved;
+	u32 loss_cwnd;
 };
 
 static unsigned int dctcp_shift_g __read_mostly = 4; /* g = 1/2^4 */
@@ -96,6 +97,7 @@ static void dctcp_init(struct sock *sk)
 		ca->dctcp_alpha = min(dctcp_alpha_on_init, DCTCP_MAX_ALPHA);
 
 		ca->delayed_ack_reserved = 0;
+		ca->loss_cwnd = 0;
 		ca->ce_state = 0;
 
 		dctcp_reset(tp, ca);
@@ -111,9 +113,10 @@ static void dctcp_init(struct sock *sk)
 
 static u32 dctcp_ssthresh(struct sock *sk)
 {
-	const struct dctcp *ca = inet_csk_ca(sk);
+	struct dctcp *ca = inet_csk_ca(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
 
+	ca->loss_cwnd = tp->snd_cwnd;
 	return max(tp->snd_cwnd - ((tp->snd_cwnd * ca->dctcp_alpha) >> 11U), 2U);
 }
 
@@ -308,12 +311,20 @@ static size_t dctcp_get_info(struct sock *sk, u32 ext, int *attr,
 	return 0;
 }
 
+static u32 dctcp_cwnd_undo(struct sock *sk)
+{
+	const struct dctcp *ca = inet_csk_ca(sk);
+
+	return max(tcp_sk(sk)->snd_cwnd, ca->loss_cwnd);
+}
+
 static struct tcp_congestion_ops dctcp __read_mostly = {
 	.init		= dctcp_init,
 	.in_ack_event   = dctcp_update_alpha,
 	.cwnd_event	= dctcp_cwnd_event,
 	.ssthresh	= dctcp_ssthresh,
 	.cong_avoid	= tcp_reno_cong_avoid,
+	.undo_cwnd	= dctcp_cwnd_undo,
 	.set_state	= dctcp_state,
 	.get_info	= dctcp_get_info,
 	.flags		= TCP_CONG_NEEDS_ECN,
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 7158d4f8dae4..7b235fa12903 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1537,6 +1537,21 @@ bool tcp_prequeue(struct sock *sk, struct sk_buff *skb)
 }
 EXPORT_SYMBOL(tcp_prequeue);
 
+int tcp_filter(struct sock *sk, struct sk_buff *skb)
+{
+	struct tcphdr *th = (struct tcphdr *)skb->data;
+	unsigned int eaten = skb->len;
+	int err;
+
+	err = sk_filter_trim_cap(sk, skb, th->doff * 4);
+	if (!err) {
+		eaten -= skb->len;
+		TCP_SKB_CB(skb)->end_seq -= eaten;
+	}
+	return err;
+}
+EXPORT_SYMBOL(tcp_filter);
+
 /*
  *	From tcp_input.c
  */
@@ -1648,8 +1663,10 @@ process:
 
 	nf_reset(skb);
 
-	if (sk_filter(sk, skb))
+	if (tcp_filter(sk, skb))
 		goto discard_and_relse;
+	th = (const struct tcphdr *)skb->data;
+	iph = ip_hdr(skb);
 
 	skb->dev = NULL;
 
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index bd59c343d35f..7370ad2e693a 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -448,7 +448,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info,
 	if (__ipv6_addr_needs_scope_id(addr_type))
 		iif = skb->dev->ifindex;
 	else
-		iif = l3mdev_master_ifindex(skb->dev);
+		iif = l3mdev_master_ifindex(skb_dst(skb)->dev);
 
 	/*
 	 *	Must not send error if the source does not uniquely
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index fc67822c42e0..af6a09efad5b 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1228,7 +1228,7 @@ static int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb)
 	if (skb->protocol == htons(ETH_P_IP))
 		return tcp_v4_do_rcv(sk, skb);
 
-	if (sk_filter(sk, skb))
+	if (tcp_filter(sk, skb))
 		goto discard;
 
 	/*
@@ -1455,8 +1455,10 @@ process:
 	if (tcp_v6_inbound_md5_hash(sk, skb))
 		goto discard_and_relse;
 
-	if (sk_filter(sk, skb))
+	if (tcp_filter(sk, skb))
 		goto discard_and_relse;
+	th = (const struct tcphdr *)skb->data;
+	hdr = ipv6_hdr(skb);
 
 	skb->dev = NULL;
 
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index baccbf3c1c60..7b0e059bf13b 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -1214,9 +1214,12 @@ static int __sctp_connect(struct sock *sk,
 
 	timeo = sock_sndtimeo(sk, f_flags & O_NONBLOCK);
 
-	err = sctp_wait_for_connect(asoc, &timeo);
-	if ((err == 0 || err == -EINPROGRESS) && assoc_id)
+	if (assoc_id)
 		*assoc_id = asoc->assoc_id;
+	err = sctp_wait_for_connect(asoc, &timeo);
+	/* Note: the asoc may be freed after the return of
+	 * sctp_wait_for_connect.
+	 */
 
 	/* Don't free association on exit. */
 	asoc = NULL;
@@ -4278,19 +4281,18 @@ static void sctp_shutdown(struct sock *sk, int how)
 {
 	struct net *net = sock_net(sk);
 	struct sctp_endpoint *ep;
-	struct sctp_association *asoc;
 
 	if (!sctp_style(sk, TCP))
 		return;
 
-	if (how & SEND_SHUTDOWN) {
+	ep = sctp_sk(sk)->ep;
+	if (how & SEND_SHUTDOWN && !list_empty(&ep->asocs)) {
+		struct sctp_association *asoc;
+
 		sk->sk_state = SCTP_SS_CLOSING;
-		ep = sctp_sk(sk)->ep;
-		if (!list_empty(&ep->asocs)) {
-			asoc = list_entry(ep->asocs.next,
-					  struct sctp_association, asocs);
-			sctp_primitive_SHUTDOWN(net, asoc, NULL);
-		}
+		asoc = list_entry(ep->asocs.next,
+				  struct sctp_association, asocs);
+		sctp_primitive_SHUTDOWN(net, asoc, NULL);
 	}
 }
 
diff --git a/net/socket.c b/net/socket.c
index a1bd16106625..03bc2c289c94 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -2041,6 +2041,8 @@ int __sys_sendmmsg(int fd, struct mmsghdr __user *mmsg, unsigned int vlen,
 		if (err)
 			break;
 		++datagrams;
+		if (msg_data_left(&msg_sys))
+			break;
 		cond_resched();
 	}
 
diff --git a/tools/spi/spidev_test.c b/tools/spi/spidev_test.c
index f3825b676e38..f046b77cfefe 100644
--- a/tools/spi/spidev_test.c
+++ b/tools/spi/spidev_test.c
@@ -19,6 +19,7 @@
 #include <getopt.h>
 #include <fcntl.h>
 #include <sys/ioctl.h>
+#include <linux/ioctl.h>
 #include <sys/stat.h>
 #include <linux/types.h>
 #include <linux/spi/spidev.h>