about summary refs log tree commit diff
path: root/REORG.TODO/sysdeps/sparc/sparc32/memcpy.S
diff options
context:
space:
mode:
Diffstat (limited to 'REORG.TODO/sysdeps/sparc/sparc32/memcpy.S')
-rw-r--r--REORG.TODO/sysdeps/sparc/sparc32/memcpy.S653
1 files changed, 653 insertions, 0 deletions
diff --git a/REORG.TODO/sysdeps/sparc/sparc32/memcpy.S b/REORG.TODO/sysdeps/sparc/sparc32/memcpy.S
new file mode 100644
index 0000000000..da672bd120
--- /dev/null
+++ b/REORG.TODO/sysdeps/sparc/sparc32/memcpy.S
@@ -0,0 +1,653 @@
+/* Copy SIZE bytes from SRC to DEST.
+   For SPARC v7.
+   Copyright (C) 1996-2017 Free Software Foundation, Inc.
+   This file is part of the GNU C Library.
+   Contributed by David S. Miller <davem@caip.rutgers.edu>,
+		  Eddie C. Dost <ecd@skynet.be> and
+                  Jakub Jelinek <jj@ultra.linux.cz>.
+
+   The GNU C Library is free software; you can redistribute it and/or
+   modify it under the terms of the GNU Lesser General Public
+   License as published by the Free Software Foundation; either
+   version 2.1 of the License, or (at your option) any later version.
+
+   The GNU C Library is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+   Lesser General Public License for more details.
+
+   You should have received a copy of the GNU Lesser General Public
+   License along with the GNU C Library; if not, see
+   <http://www.gnu.org/licenses/>.  */
+
+#include <sysdep.h>
+
+/* Both these macros have to start with exactly the same insn */
+#define MOVE_BIGCHUNK(src, dst, offset, t0, t1, t2, t3, t4, t5, t6, t7) 				\
+	ldd	[%src + offset + 0x00], %t0; 								\
+	ldd	[%src + offset + 0x08], %t2; 								\
+	ldd	[%src + offset + 0x10], %t4; 								\
+	ldd	[%src + offset + 0x18], %t6; 								\
+	st	%t0, [%dst + offset + 0x00]; 								\
+	st	%t1, [%dst + offset + 0x04]; 								\
+	st	%t2, [%dst + offset + 0x08]; 								\
+	st	%t3, [%dst + offset + 0x0c]; 								\
+	st	%t4, [%dst + offset + 0x10]; 								\
+	st	%t5, [%dst + offset + 0x14]; 								\
+	st	%t6, [%dst + offset + 0x18]; 								\
+	st	%t7, [%dst + offset + 0x1c];
+
+#define MOVE_BIGALIGNCHUNK(src, dst, offset, t0, t1, t2, t3, t4, t5, t6, t7) 				\
+	ldd	[%src + offset + 0x00], %t0; 								\
+	ldd	[%src + offset + 0x08], %t2; 								\
+	ldd	[%src + offset + 0x10], %t4; 								\
+	ldd	[%src + offset + 0x18], %t6; 								\
+	std	%t0, [%dst + offset + 0x00]; 								\
+	std	%t2, [%dst + offset + 0x08]; 								\
+	std	%t4, [%dst + offset + 0x10]; 								\
+	std	%t6, [%dst + offset + 0x18];
+
+#define MOVE_LASTCHUNK(src, dst, offset, t0, t1, t2, t3) 						\
+	ldd	[%src - offset - 0x10], %t0; 								\
+	ldd	[%src - offset - 0x08], %t2; 								\
+	st	%t0, [%dst - offset - 0x10]; 								\
+	st	%t1, [%dst - offset - 0x0c]; 								\
+	st	%t2, [%dst - offset - 0x08]; 								\
+	st	%t3, [%dst - offset - 0x04];
+
+#define MOVE_LASTALIGNCHUNK(src, dst, offset, t0, t1, t2, t3) 						\
+	ldd	[%src - offset - 0x10], %t0; 								\
+	ldd	[%src - offset - 0x08], %t2; 								\
+	std	%t0, [%dst - offset - 0x10]; 								\
+	std	%t2, [%dst - offset - 0x08];
+
+#define MOVE_SHORTCHUNK(src, dst, offset, t0, t1) 							\
+	ldub	[%src - offset - 0x02], %t0; 								\
+	ldub	[%src - offset - 0x01], %t1; 								\
+	stb	%t0, [%dst - offset - 0x02]; 								\
+	stb	%t1, [%dst - offset - 0x01];
+
+#define SMOVE_CHUNK(src, dst, offset, t0, t1, t2, t3, t4, t5, t6, prev, shil, shir, offset2) 		\
+	ldd	[%src + offset + 0x00], %t0; 								\
+	ldd	[%src + offset + 0x08], %t2; 								\
+	srl	%t0, shir, %t5; 									\
+	srl	%t1, shir, %t6; 									\
+	sll	%t0, shil, %t0; 									\
+	or	%t5, %prev, %t5; 									\
+	sll	%t1, shil, %prev; 									\
+	or	%t6, %t0, %t0; 										\
+	srl	%t2, shir, %t1; 									\
+	srl	%t3, shir, %t6; 									\
+	sll	%t2, shil, %t2; 									\
+	or	%t1, %prev, %t1; 									\
+	std	%t4, [%dst + offset + offset2 - 0x04]; 							\
+	std	%t0, [%dst + offset + offset2 + 0x04];							\
+	sll	%t3, shil, %prev; 									\
+	or	%t6, %t2, %t4;
+
+#define SMOVE_ALIGNCHUNK(src, dst, offset, t0, t1, t2, t3, t4, t5, t6, prev, shil, shir, offset2) 	\
+	ldd	[%src + offset + 0x00], %t0; 								\
+	ldd	[%src + offset + 0x08], %t2; 								\
+	srl	%t0, shir, %t4; 									\
+	srl	%t1, shir, %t5; 									\
+	sll	%t0, shil, %t6; 									\
+	or	%t4, %prev, %t0; 									\
+	sll	%t1, shil, %prev; 									\
+	or	%t5, %t6, %t1; 										\
+	srl	%t2, shir, %t4; 									\
+	srl	%t3, shir, %t5; 									\
+	sll	%t2, shil, %t6; 									\
+	or	%t4, %prev, %t2; 									\
+	sll	%t3, shil, %prev; 									\
+	or	%t5, %t6, %t3;										\
+	std	%t0, [%dst + offset + offset2 + 0x00]; 							\
+	std	%t2, [%dst + offset + offset2 + 0x08];
+
+	.text
+ENTRY(__mempcpy)
+	add		%o0, %o2, %g1
+	ba		101f
+	 st		%g1, [%sp + 64]
+END(__mempcpy)
+
+	.align	4
+ENTRY(memcpy)		/* %o0=dst %o1=src %o2=len */
+	st		%o0, [%sp + 64]
+101:
+	sub		%o0, %o1, %o4
+9:	andcc		%o4, 3, %o5
+0:	bne		86f
+	 cmp		%o2, 15
+
+	bleu		90f
+	 andcc		%o1, 3, %g0
+
+	be		78f
+	 andcc		%o1, 4, %g0
+
+	andcc		%o1, 1, %g0
+	be		4f
+	 andcc		%o1, 2, %g0
+
+	ldub		[%o1], %g2
+	add		%o1, 1, %o1
+	stb		%g2, [%o0]
+	sub		%o2, 1, %o2
+	bne		77f
+	 add		%o0, 1, %o0
+4:	lduh		[%o1], %g2
+	add		%o1, 2, %o1
+	sth		%g2, [%o0]
+	sub		%o2, 2, %o2
+	add		%o0, 2, %o0
+
+77:	andcc		%o1, 4, %g0
+78:	be		2f
+	 mov		%o2, %g1
+
+	ld		[%o1], %o4
+	sub		%g1, 4, %g1
+	st		%o4, [%o0]
+	add		%o1, 4, %o1
+	add		%o0, 4, %o0
+2:	andcc		%g1, 0xffffff80, %g6
+	be		3f
+	 andcc		%o0, 4, %g0
+
+	be		82f + 4
+5:	MOVE_BIGCHUNK(o1, o0, 0x00, o2, o3, o4, o5, g2, g3, g4, g5)
+	MOVE_BIGCHUNK(o1, o0, 0x20, o2, o3, o4, o5, g2, g3, g4, g5)
+	MOVE_BIGCHUNK(o1, o0, 0x40, o2, o3, o4, o5, g2, g3, g4, g5)
+	MOVE_BIGCHUNK(o1, o0, 0x60, o2, o3, o4, o5, g2, g3, g4, g5)
+	subcc		%g6, 128, %g6
+	add		%o1, 128, %o1
+	bne		5b
+	 add		%o0, 128, %o0
+3:	andcc		%g1, 0x70, %g6
+	be		80f
+	 andcc		%g1, 8, %g0
+
+	srl		%g6, 1, %o4
+	mov		%o7, %g2
+	add		%g6, %o4, %o4
+	add		%o1, %g6, %o1
+104:	call		100f
+	 add		%o0, %g6, %o0
+	jmpl		%o5 + (80f - 104b), %g0
+	 mov		%g2, %o7
+
+79:	MOVE_LASTCHUNK(o1, o0, 0x60, g2, g3, g4, g5)
+	MOVE_LASTCHUNK(o1, o0, 0x50, g2, g3, g4, g5)
+	MOVE_LASTCHUNK(o1, o0, 0x40, g2, g3, g4, g5)
+	MOVE_LASTCHUNK(o1, o0, 0x30, g2, g3, g4, g5)
+	MOVE_LASTCHUNK(o1, o0, 0x20, g2, g3, g4, g5)
+	MOVE_LASTCHUNK(o1, o0, 0x10, g2, g3, g4, g5)
+	MOVE_LASTCHUNK(o1, o0, 0x00, g2, g3, g4, g5)
+
+80:	be		81f
+	 andcc		%g1, 4, %g0
+
+	ldd		[%o1], %g2
+	add		%o0, 8, %o0
+	st		%g2, [%o0 - 0x08]
+	add		%o1, 8, %o1
+	st		%g3, [%o0 - 0x04]
+
+81:	be		1f
+	 andcc		%g1, 2, %g0
+
+	ld		[%o1], %g2
+	add		%o1, 4, %o1
+	st		%g2, [%o0]
+	add		%o0, 4, %o0
+1:	be		1f
+	 andcc		%g1, 1, %g0
+
+	lduh		[%o1], %g2
+	add		%o1, 2, %o1
+	sth		%g2, [%o0]
+	add		%o0, 2, %o0
+1:	be		1f
+	 nop
+
+	ldub		[%o1], %g2
+	stb		%g2, [%o0]
+1:	retl
+	 ld		[%sp + 64], %o0
+
+82:	/* ldd_std */
+	MOVE_BIGALIGNCHUNK(o1, o0, 0x00, o2, o3, o4, o5, g2, g3, g4, g5)
+	MOVE_BIGALIGNCHUNK(o1, o0, 0x20, o2, o3, o4, o5, g2, g3, g4, g5)
+	MOVE_BIGALIGNCHUNK(o1, o0, 0x40, o2, o3, o4, o5, g2, g3, g4, g5)
+	MOVE_BIGALIGNCHUNK(o1, o0, 0x60, o2, o3, o4, o5, g2, g3, g4, g5)
+	subcc		%g6, 128, %g6
+	add		%o1, 128, %o1
+	bne		82b
+	 add		%o0, 128, %o0
+
+	andcc		%g1, 0x70, %g6
+	be		84f
+	 andcc		%g1, 8, %g0
+
+	mov		%o7, %g2
+111:	call		110f
+	 add		%o1, %g6, %o1
+	mov		%g2, %o7
+	jmpl		%o5 + (84f - 111b), %g0
+	 add		%o0, %g6, %o0
+
+83:	MOVE_LASTALIGNCHUNK(o1, o0, 0x60, g2, g3, g4, g5)
+	MOVE_LASTALIGNCHUNK(o1, o0, 0x50, g2, g3, g4, g5)
+	MOVE_LASTALIGNCHUNK(o1, o0, 0x40, g2, g3, g4, g5)
+	MOVE_LASTALIGNCHUNK(o1, o0, 0x30, g2, g3, g4, g5)
+	MOVE_LASTALIGNCHUNK(o1, o0, 0x20, g2, g3, g4, g5)
+	MOVE_LASTALIGNCHUNK(o1, o0, 0x10, g2, g3, g4, g5)
+	MOVE_LASTALIGNCHUNK(o1, o0, 0x00, g2, g3, g4, g5)
+
+84:	be		85f
+	 andcc		%g1, 4, %g0
+
+	ldd		[%o1], %g2
+	add		%o0, 8, %o0
+	std		%g2, [%o0 - 0x08]
+	add		%o1, 8, %o1
+85:	be		1f
+	 andcc		%g1, 2, %g0
+
+	ld		[%o1], %g2
+	add		%o1, 4, %o1
+	st		%g2, [%o0]
+	add		%o0, 4, %o0
+1:	be		1f
+	 andcc		%g1, 1, %g0
+
+	lduh		[%o1], %g2
+	add		%o1, 2, %o1
+	sth		%g2, [%o0]
+	add		%o0, 2, %o0
+1:	be		1f
+	 nop
+
+	ldub		[%o1], %g2
+	stb		%g2, [%o0]
+1:	retl
+	 ld		[%sp + 64], %o0
+
+86:	cmp		%o2, 6
+	bleu		88f
+
+	 cmp		%o2, 256
+	bcc		87f
+
+	 andcc		%o0, 3, %g0
+	be		61f
+	 andcc		%o0, 1, %g0
+	be		60f
+	 andcc		%o0, 2, %g0
+
+	ldub		[%o1], %g5
+	add		%o1, 1, %o1
+	stb		%g5, [%o0]
+	sub		%o2, 1, %o2
+	bne		61f
+	 add		%o0, 1, %o0
+60:	ldub		[%o1], %g3
+	add		%o1, 2, %o1
+	stb		%g3, [%o0]
+	sub		%o2, 2, %o2
+	ldub		[%o1 - 1], %g3
+	add		%o0, 2, %o0
+	stb		%g3, [%o0 - 1]
+61:	and		%o1, 3, %g2
+	and		%o2, 0xc, %g3
+	and		%o1, -4, %o1
+	cmp		%g3, 4
+	sll		%g2, 3, %g4
+	mov		32, %g2
+	be		4f
+	 sub		%g2, %g4, %g6
+
+	blu		3f
+	 cmp		%g3, 0x8
+
+	be		2f
+	 srl		%o2, 2, %g3
+
+	ld		[%o1], %o3
+	add		%o0, -8, %o0
+	ld		[%o1 + 4], %o4
+	b		8f
+	 add		%g3, 1, %g3
+2:	ld		[%o1], %o4
+	add		%o0, -12, %o0
+	ld		[%o1 + 4], %o5
+	add		%g3, 2, %g3
+	b		9f
+	 add		%o1, -4, %o1
+3:	ld		[%o1], %g1
+	add		%o0, -4, %o0
+	ld		[%o1 + 4], %o3
+	srl		%o2, 2, %g3
+	b		7f
+	 add		%o1, 4, %o1
+4:	ld		[%o1], %o5
+	cmp		%o2, 7
+	ld		[%o1 + 4], %g1
+	srl		%o2, 2, %g3
+	bleu		10f
+	 add		%o1, 8, %o1
+
+	ld		[%o1], %o3
+	add		%g3, -1, %g3
+5:	sll		%o5, %g4, %g2
+	srl		%g1, %g6, %g5
+	or		%g2, %g5, %g2
+	st		%g2, [%o0]
+7:	ld		[%o1 + 4], %o4
+	sll		%g1, %g4, %g2
+	srl		%o3, %g6, %g5
+	or		%g2, %g5, %g2
+	st		%g2, [%o0 + 4]
+8:	ld		[%o1 + 8], %o5
+	sll		%o3, %g4, %g2
+	srl		%o4, %g6, %g5
+	or		%g2, %g5, %g2
+	st		%g2, [%o0 + 8]
+9:	ld		[%o1 + 12], %g1
+	sll		%o4, %g4, %g2
+	srl		%o5, %g6, %g5
+	addcc		%g3, -4, %g3
+	or		%g2, %g5, %g2
+	add		%o1, 16, %o1
+	st		%g2, [%o0 + 12]
+	add		%o0, 16, %o0
+	bne,a		5b
+	 ld		[%o1], %o3
+10:	sll		%o5, %g4, %g2
+	srl		%g1, %g6, %g5
+	srl		%g6, 3, %g3
+	or		%g2, %g5, %g2
+	sub		%o1, %g3, %o1
+	andcc		%o2, 2, %g0
+	st		%g2, [%o0]
+	be		1f
+	 andcc		%o2, 1, %g0
+
+	ldub		[%o1], %g2
+	add		%o1, 2, %o1
+	stb		%g2, [%o0 + 4]
+	add		%o0, 2, %o0
+	ldub		[%o1 - 1], %g2
+	stb		%g2, [%o0 + 3]
+1:	be		1f
+	 nop
+	ldub		[%o1], %g2
+	stb		%g2, [%o0 + 4]
+1:	retl
+	 ld		[%sp + 64], %o0
+
+87:	andcc		%o1, 3, %g0
+	be		3f
+	 andcc		%o1, 1, %g0
+
+	be		4f
+	 andcc		%o1, 2, %g0
+
+	ldub		[%o1], %g2
+	add		%o1, 1, %o1
+	stb		%g2, [%o0]
+	sub		%o2, 1, %o2
+	bne		3f
+	 add		%o0, 1, %o0
+4:	lduh		[%o1], %g2
+	add		%o1, 2, %o1
+	srl		%g2, 8, %g3
+	sub		%o2, 2, %o2
+	stb		%g3, [%o0]
+	add		%o0, 2, %o0
+	stb		%g2, [%o0 - 1]
+3:	 andcc		%o1, 4, %g0
+
+	bne		2f
+	 cmp		%o5, 1
+
+	ld		[%o1], %o4
+	srl		%o4, 24, %g2
+	stb		%g2, [%o0]
+	srl		%o4, 16, %g3
+	stb		%g3, [%o0 + 1]
+	srl		%o4, 8, %g2
+	stb		%g2, [%o0 + 2]
+	sub		%o2, 4, %o2
+	stb		%o4, [%o0 + 3]
+	add		%o1, 4, %o1
+	add		%o0, 4, %o0
+2:	be		33f
+	 cmp		%o5, 2
+	be		32f
+	 sub		%o2, 4, %o2
+31:	ld		[%o1], %g2
+	add		%o1, 4, %o1
+	srl		%g2, 24, %g3
+	and		%o0, 7, %g5
+	stb		%g3, [%o0]
+	cmp		%g5, 7
+	sll		%g2, 8, %g1
+	add		%o0, 4, %o0
+	be		41f
+	 and		%o2, 0xffffffc0, %o3
+	ld		[%o0 - 7], %o4
+4:	SMOVE_CHUNK(o1, o0, 0x00, g2, g3, g4, g5, o4, o5, g6, g1, 8, 24, -3)
+	SMOVE_CHUNK(o1, o0, 0x10, g2, g3, g4, g5, o4, o5, g6, g1, 8, 24, -3)
+	SMOVE_CHUNK(o1, o0, 0x20, g2, g3, g4, g5, o4, o5, g6, g1, 8, 24, -3)
+	SMOVE_CHUNK(o1, o0, 0x30, g2, g3, g4, g5, o4, o5, g6, g1, 8, 24, -3)
+	subcc		%o3, 64, %o3
+	add		%o1, 64, %o1
+	bne		4b
+	 add		%o0, 64, %o0
+
+	andcc		%o2, 0x30, %o3
+	be,a		1f
+	 srl		%g1, 16, %g2
+4:	SMOVE_CHUNK(o1, o0, 0x00, g2, g3, g4, g5, o4, o5, g6, g1, 8, 24, -3)
+	subcc		%o3, 16, %o3
+	add		%o1, 16, %o1
+	bne		4b
+	 add		%o0, 16, %o0
+
+	srl		%g1, 16, %g2
+1:	st		%o4, [%o0 - 7]
+	sth		%g2, [%o0 - 3]
+	srl		%g1, 8, %g4
+	b		88f
+	 stb		%g4, [%o0 - 1]
+32:	ld		[%o1], %g2
+	add		%o1, 4, %o1
+	srl		%g2, 16, %g3
+	and		%o0, 7, %g5
+	sth		%g3, [%o0]
+	cmp		%g5, 6
+	sll		%g2, 16, %g1
+	add		%o0, 4, %o0
+	be		42f
+	 and		%o2, 0xffffffc0, %o3
+	ld		[%o0 - 6], %o4
+4:	SMOVE_CHUNK(o1, o0, 0x00, g2, g3, g4, g5, o4, o5, g6, g1, 16, 16, -2)
+	SMOVE_CHUNK(o1, o0, 0x10, g2, g3, g4, g5, o4, o5, g6, g1, 16, 16, -2)
+	SMOVE_CHUNK(o1, o0, 0x20, g2, g3, g4, g5, o4, o5, g6, g1, 16, 16, -2)
+	SMOVE_CHUNK(o1, o0, 0x30, g2, g3, g4, g5, o4, o5, g6, g1, 16, 16, -2)
+	subcc		%o3, 64, %o3
+	add		%o1, 64, %o1
+	bne		4b
+	 add		%o0, 64, %o0
+
+	andcc		%o2, 0x30, %o3
+	be,a		1f
+	 srl		%g1, 16, %g2
+4:	SMOVE_CHUNK(o1, o0, 0x00, g2, g3, g4, g5, o4, o5, g6, g1, 16, 16, -2)
+	subcc		%o3, 16, %o3
+	add		%o1, 16, %o1
+	bne		4b
+	 add		%o0, 16, %o0
+
+	srl		%g1, 16, %g2
+1:	st		%o4, [%o0 - 6]
+	b		88f
+	 sth		%g2, [%o0 - 2]
+33:	ld		[%o1], %g2
+	sub		%o2, 4, %o2
+	srl		%g2, 24, %g3
+	and		%o0, 7, %g5
+	stb		%g3, [%o0]
+	cmp		%g5, 5
+	srl		%g2, 8, %g4
+	sll		%g2, 24, %g1
+	sth		%g4, [%o0 + 1]
+	add		%o1, 4, %o1
+	be		43f
+	 and		%o2, 0xffffffc0, %o3
+
+	ld		[%o0 - 1], %o4
+	add		%o0, 4, %o0
+4:	SMOVE_CHUNK(o1, o0, 0x00, g2, g3, g4, g5, o4, o5, g6, g1, 24, 8, -1)
+	SMOVE_CHUNK(o1, o0, 0x10, g2, g3, g4, g5, o4, o5, g6, g1, 24, 8, -1)
+	SMOVE_CHUNK(o1, o0, 0x20, g2, g3, g4, g5, o4, o5, g6, g1, 24, 8, -1)
+	SMOVE_CHUNK(o1, o0, 0x30, g2, g3, g4, g5, o4, o5, g6, g1, 24, 8, -1)
+	subcc		%o3, 64, %o3
+	add		%o1, 64, %o1
+	bne		4b
+	 add		%o0, 64, %o0
+
+	andcc		%o2, 0x30, %o3
+	be,a		1f
+	 srl		%g1, 24, %g2
+4:	SMOVE_CHUNK(o1, o0, 0x00, g2, g3, g4, g5, o4, o5, g6, g1, 24, 8, -1)
+	subcc		%o3, 16, %o3
+	add		%o1, 16, %o1
+	bne		4b
+	 add		%o0, 16, %o0
+
+	srl		%g1, 24, %g2
+1:	st		%o4, [%o0 - 5]
+	b		88f
+	 stb		%g2, [%o0 - 1]
+41:	SMOVE_ALIGNCHUNK(o1, o0, 0x00, g2, g3, g4, g5, o4, o5, g6, g1, 8, 24, -3)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x10, g2, g3, g4, g5, o4, o5, g6, g1, 8, 24, -3)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x20, g2, g3, g4, g5, o4, o5, g6, g1, 8, 24, -3)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x30, g2, g3, g4, g5, o4, o5, g6, g1, 8, 24, -3)
+	subcc		%o3, 64, %o3
+	add		%o1, 64, %o1
+	bne		41b
+	 add		%o0, 64, %o0
+
+	andcc		%o2, 0x30, %o3
+	be,a		1f
+	 srl		%g1, 16, %g2
+4:	SMOVE_ALIGNCHUNK(o1, o0, 0x00, g2, g3, g4, g5, o4, o5, g6, g1, 8, 24, -3)
+	subcc		%o3, 16, %o3
+	add		%o1, 16, %o1
+	bne		4b
+	 add		%o0, 16, %o0
+
+	srl		%g1, 16, %g2
+1:	sth		%g2, [%o0 - 3]
+	srl		%g1, 8, %g4
+	b		88f
+	 stb		%g4, [%o0 - 1]
+43:	SMOVE_ALIGNCHUNK(o1, o0, 0x00, g2, g3, g4, g5, o4, o5, g6, g1, 24, 8, 3)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x10, g2, g3, g4, g5, o4, o5, g6, g1, 24, 8, 3)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x20, g2, g3, g4, g5, o4, o5, g6, g1, 24, 8, 3)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x30, g2, g3, g4, g5, o4, o5, g6, g1, 24, 8, 3)
+	subcc		%o3, 64, %o3
+	add		%o1, 64, %o1
+	bne		43b
+	 add		%o0, 64, %o0
+
+	andcc		%o2, 0x30, %o3
+	be,a		1f
+	 srl		%g1, 24, %g2
+4:	SMOVE_ALIGNCHUNK(o1, o0, 0x00, g2, g3, g4, g5, o4, o5, g6, g1, 24, 8, 3)
+	subcc		%o3, 16, %o3
+	add		%o1, 16, %o1
+	bne		4b
+	 add		%o0, 16, %o0
+
+	srl		%g1, 24, %g2
+1:	stb		%g2, [%o0 + 3]
+	b		88f
+	 add		%o0, 4, %o0
+42:	SMOVE_ALIGNCHUNK(o1, o0, 0x00, g2, g3, g4, g5, o4, o5, g6, g1, 16, 16, -2)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x10, g2, g3, g4, g5, o4, o5, g6, g1, 16, 16, -2)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x20, g2, g3, g4, g5, o4, o5, g6, g1, 16, 16, -2)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x30, g2, g3, g4, g5, o4, o5, g6, g1, 16, 16, -2)
+	subcc		%o3, 64, %o3
+	add		%o1, 64, %o1
+	bne		42b
+	 add		%o0, 64, %o0
+
+	andcc		%o2, 0x30, %o3
+	be,a		1f
+	 srl		%g1, 16, %g2
+4:	SMOVE_ALIGNCHUNK(o1, o0, 0x00, g2, g3, g4, g5, o4, o5, g6, g1, 16, 16, -2)
+	subcc		%o3, 16, %o3
+	add		%o1, 16, %o1
+	bne		4b
+	 add		%o0, 16, %o0
+
+	srl		%g1, 16, %g2
+1:	sth		%g2, [%o0 - 2]
+
+	/* Fall through */
+
+88:	and		%o2, 0xe, %o3
+	mov		%o7, %g2
+	sll		%o3, 3, %o4
+	add		%o0, %o3, %o0
+106:	call		100f
+	 add		%o1, %o3, %o1
+	mov		%g2, %o7
+	jmpl		%o5 + (89f - 106b), %g0
+	 andcc		%o2, 1, %g0
+
+	MOVE_SHORTCHUNK(o1, o0, 0x0c, g2, g3)
+	MOVE_SHORTCHUNK(o1, o0, 0x0a, g2, g3)
+	MOVE_SHORTCHUNK(o1, o0, 0x08, g2, g3)
+	MOVE_SHORTCHUNK(o1, o0, 0x06, g2, g3)
+	MOVE_SHORTCHUNK(o1, o0, 0x04, g2, g3)
+	MOVE_SHORTCHUNK(o1, o0, 0x02, g2, g3)
+	MOVE_SHORTCHUNK(o1, o0, 0x00, g2, g3)
+
+89:	be		1f
+	 nop
+
+	ldub		[%o1], %g2
+	stb		%g2, [%o0]
+1:	retl
+	 ld		[%sp + 64], %o0
+
+90:	bne		88b
+	 andcc		%o2, 8, %g0
+
+	be		1f
+	 andcc		%o2, 4, %g0
+
+	ld		[%o1 + 0x00], %g2
+	ld		[%o1 + 0x04], %g3
+	add		%o1, 8, %o1
+	st		%g2, [%o0 + 0x00]
+	st		%g3, [%o0 + 0x04]
+	add		%o0, 8, %o0
+1:	b		81b
+	 mov		%o2, %g1
+
+100:	retl
+	 sub		%o7, %o4, %o5
+110:	retl
+	 sub		%o7, %g6, %o5
+END(memcpy)
+
+libc_hidden_builtin_def (memcpy)
+
+libc_hidden_def (__mempcpy)
+weak_alias (__mempcpy, mempcpy)
+libc_hidden_builtin_def (mempcpy)