Sparc string routines from Jakub.

1999-03-29  Jakub Jelinek  <jj@ultra.linux.cz>

	* sysdeps/sparc/sparc32/sparcv8/Makefile: New file.
	* sysdeps/sparc/sparc32/sparcv8/rem.S: Delay after write %y.
	* sysdeps/sparc/sparc32/sparcv8/sdiv.S: Likewise.
	* sysdeps/sparc/sparc32/sparcv8/udiv.S: Likewise.
	* sysdeps/sparc/sparc32/sparcv8/urem.S: Likewise.

	* sysdeps/sparc/sparc32/sparcv9/bcopy.c: New file.
	* sysdeps/sparc/sparc32/sparcv9/bzero.c: New file.
	* sysdeps/sparc/sparc32/sparcv9/memchr.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/memcmp.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/memcpy.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/memset.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/stpcpy.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/strcat.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/strchr.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/strcmp.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/strcpy.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/strcspn.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/strlen.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/strncpy.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/strpbrk.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/strrchr.c: New file.
	* sysdeps/sparc/sparc32/sparcv9/strspn.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/dotmul.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/rem.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/sdiv.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/udiv.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/umul.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/urem.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/Makefile: New file.
	* sysdeps/sparc/sparc32/sparcv9/stpncpy.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/strncmp.S: New file.

	* sysdeps/sparc/sparc32/bcopy.c: New file.
	* sysdeps/sparc/sparc32/bzero.c: New file.
	* sysdeps/sparc/sparc32/memchr.S: New file.
	* sysdeps/sparc/sparc32/memcpy.S: New file.
	* sysdeps/sparc/sparc32/memmove.c: New file.
	* sysdeps/sparc/sparc32/memset.S: New file.
	* sysdeps/sparc/sparc32/stpcpy.S: New file.
	* sysdeps/sparc/sparc32/strchr.S: New file.
	* sysdeps/sparc/sparc32/strrchr.c: New file.
	* sysdeps/sparc/sparc32/strcpy.S: New file.
	* sysdeps/sparc/sparc32/strlen.S: New file.
	* sysdeps/sparc/sparc32/strcat.S: New file.
	* sysdeps/sparc/sparc32/strcmp.S: New file.

	* sysdeps/sparc/sparc64/bcopy.c: New file.
	* sysdeps/sparc/sparc64/bzero.c: New file.
	* sysdeps/sparc/sparc64/memchr.S: New file.
	* sysdeps/sparc/sparc64/memcmp.S: New file.
	* sysdeps/sparc/sparc64/memcpy.S: New file.
	* sysdeps/sparc/sparc64/memset.S: New file.
	* sysdeps/sparc/sparc64/stpcpy.S: New file.
	* sysdeps/sparc/sparc64/strcat.S: New file.
	* sysdeps/sparc/sparc64/strchr.S: New file.
	* sysdeps/sparc/sparc64/strcmp.S: New file.
	* sysdeps/sparc/sparc64/strcpy.S: New file.
	* sysdeps/sparc/sparc64/strcspn.S: New file.
	* sysdeps/sparc/sparc64/strlen.S: New file.
	* sysdeps/sparc/sparc64/strncpy.S: New file.
	* sysdeps/sparc/sparc64/strpbrk.S: New file.
	* sysdeps/sparc/sparc64/strrchr.c: New file.
	* sysdeps/sparc/sparc64/strspn.S: New file.
	* sysdeps/sparc/sparc64/stpncpy.S: New file.
	* sysdeps/sparc/sparc64/strncmp.S: New file.

	* sysdeps/unix/sysv/linux/sparc/sparc32/Makefile (ASFLAGS-.os):
	Append -fPIC, don't replace.

	* configure.in: Recognize sparcv8plus / sparcv9.

1 files changed, 974 insertions, 0 deletions

diff --git a/sysdeps/sparc/sparc32/memcpy.S b/sysdeps/sparc/sparc32/memcpy.S
new file mode 100644
index 0000000000..24d9e04346
--- /dev/null
+++ b/sysdeps/sparc/sparc32/memcpy.S
@@ -0,0 +1,974 @@
+/* Copy SIZE bytes from SRC to DEST.
+   For SPARC v7.
+   Copyright (C) 1996, 1999 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 Library General Public License as
+   published by the Free Software Foundation; either version 2 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
+   Library General Public License for more details.
+
+   You should have received a copy of the GNU Library General Public
+   License along with the GNU C Library; see the file COPYING.LIB.  If not,
+   write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+   Boston, MA 02111-1307, USA.  */
+
+#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];
+
+/* Both these macros have to start with exactly the same insn */
+#define RMOVE_BIGCHUNK(src, dst, offset, t0, t1, t2, t3, t4, t5, t6, t7) 				\
+	ldd	[%src - offset - 0x20], %t0; 								\
+	ldd	[%src - offset - 0x18], %t2; 								\
+	ldd	[%src - offset - 0x10], %t4; 								\
+	ldd	[%src - offset - 0x08], %t6; 								\
+	st	%t0, [%dst - offset - 0x20]; 								\
+	st	%t1, [%dst - offset - 0x1c]; 								\
+	st	%t2, [%dst - offset - 0x18]; 								\
+	st	%t3, [%dst - offset - 0x14]; 								\
+	st	%t4, [%dst - offset - 0x10]; 								\
+	st	%t5, [%dst - offset - 0x0c]; 								\
+	st	%t6, [%dst - offset - 0x08]; 								\
+	st	%t7, [%dst - offset - 0x04];
+
+#define RMOVE_BIGALIGNCHUNK(src, dst, offset, t0, t1, t2, t3, t4, t5, t6, t7) 				\
+	ldd	[%src - offset - 0x20], %t0; 								\
+	ldd	[%src - offset - 0x18], %t2; 								\
+	ldd	[%src - offset - 0x10], %t4; 								\
+	ldd	[%src - offset - 0x08], %t6; 								\
+	std	%t0, [%dst - offset - 0x20]; 								\
+	std	%t2, [%dst - offset - 0x18]; 								\
+	std	%t4, [%dst - offset - 0x10]; 								\
+	std	%t6, [%dst - offset - 0x08];
+
+#define RMOVE_LASTCHUNK(src, dst, offset, t0, t1, t2, t3) 						\
+	ldd	[%src + offset + 0x00], %t0; 								\
+	ldd	[%src + offset + 0x08], %t2; 								\
+	st	%t0, [%dst + offset + 0x00]; 								\
+	st	%t1, [%dst + offset + 0x04]; 								\
+	st	%t2, [%dst + offset + 0x08]; 								\
+	st	%t3, [%dst + offset + 0x0c];
+
+#define RMOVE_SHORTCHUNK(src, dst, offset, t0, t1) 							\
+	ldub	[%src + offset + 0x00], %t0; 								\
+	ldub	[%src + offset + 0x01], %t1; 								\
+	stb	%t0, [%dst + offset + 0x00]; 								\
+	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
+	.align	4
+
+70:	andcc		%o1, 1, %g0
+	be		4f
+	 andcc		%o1, 2, %g0
+
+	ldub		[%o1 - 1], %g2
+	sub		%o1, 1, %o1
+	stb		%g2, [%o0 - 1]
+	sub		%o2, 1, %o2
+	be		3f
+	 sub		%o0, 1, %o0
+4:	lduh		[%o1 - 2], %g2
+	sub		%o1, 2, %o1
+	sth		%g2, [%o0 - 2]
+	sub		%o2, 2, %o2
+	b		3f
+	 sub		%o0, 2, %o0
+
+0:	retl
+	 nop		! Only bcopy returns here and it retuns void...
+
+ENTRY(bcopy)
+	mov		%o0, %o3
+	mov		%o1, %o0
+	mov		%o3, %o1
+	tst		%o2
+	bcs		0b
+	 /* Do the cmp in the delay slot */
+END(bcopy)
+ENTRY(memmove)
+	cmp		%o0, %o1
+	st		%o0, [%sp + 64]
+	bleu		9f
+	 sub		%o0, %o1, %o4
+
+	add		%o1, %o2, %o3
+	cmp		%o3, %o0
+	bleu		0f
+	 andcc		%o4, 3, %o5
+
+	add		%o1, %o2, %o1
+	add		%o0, %o2, %o0
+	bne		77f
+	 cmp		%o2, 15
+	bleu		91f
+	 andcc		%o1, 3, %g0
+	bne		70b
+3:	 andcc		%o1, 4, %g0
+
+	be		2f
+	 mov		%o2, %g1
+
+	ld		[%o1 - 4], %o4
+	sub		%g1, 4, %g1
+	st		%o4, [%o0 - 4]
+	sub		%o1, 4, %o1
+	sub		%o0, 4, %o0
+2:	andcc		%g1, 0xffffff80, %g7
+	be		3f
+	 andcc		%o0, 4, %g0
+
+	be		74f + 4
+5:	RMOVE_BIGCHUNK(o1, o0, 0x00, o2, o3, o4, o5, g2, g3, g4, g5)
+	RMOVE_BIGCHUNK(o1, o0, 0x20, o2, o3, o4, o5, g2, g3, g4, g5)
+	RMOVE_BIGCHUNK(o1, o0, 0x40, o2, o3, o4, o5, g2, g3, g4, g5)
+	RMOVE_BIGCHUNK(o1, o0, 0x60, o2, o3, o4, o5, g2, g3, g4, g5)
+	subcc		%g7, 128, %g7
+	sub		%o1, 128, %o1
+	bne		5b
+	 sub		%o0, 128, %o0
+
+3:	andcc		%g1, 0x70, %g7
+	be		72f
+	 andcc		%g1, 8, %g0
+
+	srl		%g7, 1, %o4
+	mov		%o7, %g2
+	add		%g7, %o4, %o4
+101:	call		100f
+	 sub		%o1, %g7, %o1
+	mov		%g2, %o7
+	jmpl		%o5 + (72f - 101b), %g0
+	 sub		%o0, %g7, %o0
+
+71:	RMOVE_LASTCHUNK(o1, o0, 0x60, g2, g3, g4, g5)
+	RMOVE_LASTCHUNK(o1, o0, 0x50, g2, g3, g4, g5)
+	RMOVE_LASTCHUNK(o1, o0, 0x40, g2, g3, g4, g5)
+	RMOVE_LASTCHUNK(o1, o0, 0x30, g2, g3, g4, g5)
+	RMOVE_LASTCHUNK(o1, o0, 0x20, g2, g3, g4, g5)
+	RMOVE_LASTCHUNK(o1, o0, 0x10, g2, g3, g4, g5)
+	RMOVE_LASTCHUNK(o1, o0, 0x00, g2, g3, g4, g5)
+72:	be		73f
+	 andcc		%g1, 4, %g0
+
+	ldd		[%o1 - 0x08], %g2
+	sub		%o0, 8, %o0
+	sub		%o1, 8, %o1
+	st		%g2, [%o0]
+	st		%g3, [%o0 + 0x04]
+73:	be		1f
+	 andcc		%g1, 2, %g0
+
+	ld		[%o1 - 4], %g2
+	sub		%o1, 4, %o1
+	st		%g2, [%o0 - 4]
+	sub		%o0, 4, %o0
+1:	be		1f
+	 andcc		%g1, 1, %g0
+
+	lduh		[%o1 - 2], %g2
+	sub		%o1, 2, %o1
+	sth		%g2, [%o0 - 2]
+	sub		%o0, 2, %o0
+1:	be		1f
+	 nop
+
+	ldub		[%o1 - 1], %g2
+	stb		%g2, [%o0 - 1]
+1:	retl
+	 ld		[%sp + 64], %o0
+
+74:	RMOVE_BIGALIGNCHUNK(o1, o0, 0x00, o2, o3, o4, o5, g2, g3, g4, g5)
+	RMOVE_BIGALIGNCHUNK(o1, o0, 0x20, o2, o3, o4, o5, g2, g3, g4, g5)
+	RMOVE_BIGALIGNCHUNK(o1, o0, 0x40, o2, o3, o4, o5, g2, g3, g4, g5)
+	RMOVE_BIGALIGNCHUNK(o1, o0, 0x60, o2, o3, o4, o5, g2, g3, g4, g5)
+	subcc		%g7, 128, %g7
+	sub		%o1, 128, %o1
+	bne		74b
+	 sub		%o0, 128, %o0
+
+	andcc		%g1, 0x70, %g7
+	be		72b
+	 andcc		%g1, 8, %g0
+
+	srl		%g7, 1, %o4
+	mov		%o7, %g2
+	add		%g7, %o4, %o4
+102:	call		100f
+	 sub		%o1, %g7, %o1
+	mov		%g2, %o7
+	jmpl		%o5 + (72b - 102b), %g0
+	 sub		%o0, %g7, %o0
+
+75:	and		%o2, 0xe, %o3
+	mov		%o7, %g2
+	sll		%o3, 3, %o4
+	sub		%o0, %o3, %o0
+103:	call		100f
+	 sub		%o1, %o3, %o1
+	mov		%g2, %o7
+	jmpl		%o5 + (76f - 103b), %g0
+	 andcc		%o2, 1, %g0
+
+	RMOVE_SHORTCHUNK(o1, o0, 0x0c, g2, g3)
+	RMOVE_SHORTCHUNK(o1, o0, 0x0a, g2, g3)
+	RMOVE_SHORTCHUNK(o1, o0, 0x08, g2, g3)
+	RMOVE_SHORTCHUNK(o1, o0, 0x06, g2, g3)
+	RMOVE_SHORTCHUNK(o1, o0, 0x04, g2, g3)
+	RMOVE_SHORTCHUNK(o1, o0, 0x02, g2, g3)
+	RMOVE_SHORTCHUNK(o1, o0, 0x00, g2, g3)
+
+76:	be		1f
+	 nop
+	ldub		[%o1 - 1], %g2
+	stb		%g2, [%o0 - 1]
+1:	retl
+	 ld		[%sp + 64], %o0
+
+91:	bne		75b
+	 andcc		%o2, 8, %g0
+
+	be		1f
+	 andcc		%o2, 4, %g0
+
+	ld		[%o1 - 0x08], %g2
+	ld		[%o1 - 0x04], %g3
+	sub		%o1, 8, %o1
+	st		%g2, [%o0 - 0x08]
+	st		%g3, [%o0 - 0x04]
+	sub		%o0, 8, %o0
+1:	b		73b
+	 mov		%o2, %g1
+
+77:	cmp		%o2, 15
+	bleu		75b
+	 andcc		%o0, 3, %g0
+	be		64f
+	 andcc		%o0, 1, %g0
+	be		63f
+	 andcc		%o0, 2, %g0
+	ldub		[%o1 - 1], %g5
+	sub		%o1, 1, %o1
+	stb		%g5, [%o0 - 1]
+	sub		%o0, 1, %o0
+	be		64f
+	 sub		%o2, 1, %o2
+
+63:	ldub		[%o1 - 1], %g5
+	sub		%o1, 2, %o1
+	stb		%g5, [%o0 - 1]
+	sub		%o0, 2, %o0
+	ldub		[%o1], %g5
+	sub		%o2, 2, %o2
+	stb		%g5, [%o0]
+64:	and		%o1, 3, %g2
+	and		%o1, -4, %o1
+	and		%o2, 0xc, %g3
+	add		%o1, 4, %o1
+	cmp		%g3, 4
+	sll		%g2, 3, %g4
+	mov		32, %g2
+	be		4f
+	 sub		%g2, %g4, %g7
+
+	blu		3f
+	 cmp		%g3, 8
+
+	be		2f
+	 srl		%o2, 2, %g3
+
+	ld		[%o1 - 4], %o3
+	add		%o0, -8, %o0
+	ld		[%o1 - 8], %o4
+	add		%o1, -16, %o1
+	b		7f
+	 add		%g3, 1, %g3
+2:	ld		[%o1 - 4], %o4
+	add		%o0, -4, %o0
+	ld		[%o1 - 8], %g1
+	add		%o1, -12, %o1
+	b		8f
+	 add		%g3, 2, %g3
+3:	ld		[%o1 - 4], %o5
+	add		%o0, -12, %o0
+	ld		[%o1 - 8], %o3
+	add		%o1, -20, %o1
+	b		6f
+	 srl		%o2, 2, %g3
+4:	ld		[%o1 - 4], %g1
+	srl		%o2, 2, %g3
+	ld		[%o1 - 8], %o5
+	add		%o1, -24, %o1
+	add		%o0, -16, %o0
+	add		%g3, -1, %g3
+
+	ld		[%o1 + 12], %o3
+5:	sll		%o5, %g4, %g2
+	srl		%g1, %g7, %g5
+	or		%g2, %g5, %g2
+	st		%g2, [%o0 + 12]
+6:	ld		[%o1 + 8], %o4
+	sll		%o3, %g4, %g2
+	srl		%o5, %g7, %g5
+	or		%g2, %g5, %g2
+	st		%g2, [%o0 + 8]
+7:	ld		[%o1 + 4], %g1
+	sll		%o4, %g4, %g2
+	srl		%o3, %g7, %g5
+	or		%g2, %g5, %g2
+	st		%g2, [%o0 + 4]
+8:	ld		[%o1], %o5
+	sll		%g1, %g4, %g2
+	srl		%o4, %g7, %g5
+	addcc		%g3, -4, %g3
+	or		%g2, %g5, %g2
+	add		%o1, -16, %o1
+	st		%g2, [%o0]
+	add		%o0, -16, %o0
+	bne,a		5b	
+	 ld		[%o1 + 12], %o3
+	sll		%o5, %g4, %g2
+	srl		%g1, %g7, %g5
+	srl		%g4, 3, %g3
+	or		%g2, %g5, %g2
+	add		%o1, %g3, %o1
+	andcc		%o2, 2, %g0
+	st		%g2, [%o0 + 12]
+	be		1f
+	 andcc		%o2, 1, %g0
+	
+	ldub		[%o1 + 15], %g5
+	add		%o1, -2, %o1
+	stb		%g5, [%o0 + 11]
+	add		%o0, -2, %o0
+	ldub		[%o1 + 16], %g5
+	stb		%g5, [%o0 + 12]
+1:	be		1f
+	 nop
+	ldub		[%o1 + 15], %g5
+	stb		%g5, [%o0 + 11]
+1:	retl
+	 ld		[%sp + 64], %o0
+
+78:	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
+	sth		%g2, [%o0]
+	sub		%o2, 2, %o2
+	b		3f
+	 add		%o0, 2, %o0
+END(memmove)
+
+ENTRY(memcpy)		/* %o0=dst %o1=src %o2=len */
+	sub		%o0, %o1, %o4
+	st		%o0, [%sp + 64]
+9:	andcc		%o4, 3, %o5
+0:	bne		86f
+	 cmp		%o2, 15
+
+	bleu		90f
+	 andcc		%o1, 3, %g0
+
+	bne		78b
+3:	 andcc		%o1, 4, %g0
+
+	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, %g7
+	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		%g7, 128, %g7
+	add		%o1, 128, %o1
+	bne		5b
+	 add		%o0, 128, %o0
+3:	andcc		%g1, 0x70, %g7
+	be		80f
+	 andcc		%g1, 8, %g0
+
+	srl		%g7, 1, %o4
+	mov		%o7, %g2
+	add		%g7, %o4, %o4
+	add		%o1, %g7, %o1
+104:	call		100f
+	 add		%o0, %g7, %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		%g7, 128, %g7
+	add		%o1, 128, %o1
+	bne		82b
+	 add		%o0, 128, %o0
+
+	andcc		%g1, 0x70, %g7
+	be		84f
+	 andcc		%g1, 8, %g0
+
+	mov		%o7, %g2
+111:	call		110f
+	 add		%o1, %g7, %o1
+	mov		%g2, %o7
+	jmpl		%o5 + (84f - 111b), %g0
+	 add		%o0, %g7, %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, %g7
+	
+	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, %g7, %g5
+	or		%g2, %g5, %g2
+	st		%g2, [%o0]
+7:	ld		[%o1 + 4], %o4
+	sll		%g1, %g4, %g2
+	srl		%o3, %g7, %g5
+	or		%g2, %g5, %g2
+	st		%g2, [%o0 + 4]
+8:	ld		[%o1 + 8], %o5
+	sll		%o3, %g4, %g2
+	srl		%o4, %g7, %g5
+	or		%g2, %g5, %g2
+	st		%g2, [%o0 + 8]
+9:	ld		[%o1 + 12], %g1
+	sll		%o4, %g4, %g2
+	srl		%o5, %g7, %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, %g7, %g5
+	srl		%g7, 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, g7, g1, 8, 24, -3)
+	SMOVE_CHUNK(o1, o0, 0x10, g2, g3, g4, g5, o4, o5, g7, g1, 8, 24, -3)
+	SMOVE_CHUNK(o1, o0, 0x20, g2, g3, g4, g5, o4, o5, g7, g1, 8, 24, -3)
+	SMOVE_CHUNK(o1, o0, 0x30, g2, g3, g4, g5, o4, o5, g7, 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, g7, 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, g7, g1, 16, 16, -2)
+	SMOVE_CHUNK(o1, o0, 0x10, g2, g3, g4, g5, o4, o5, g7, g1, 16, 16, -2)
+	SMOVE_CHUNK(o1, o0, 0x20, g2, g3, g4, g5, o4, o5, g7, g1, 16, 16, -2)
+	SMOVE_CHUNK(o1, o0, 0x30, g2, g3, g4, g5, o4, o5, g7, 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, g7, 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, g7, g1, 24, 8, -1)
+	SMOVE_CHUNK(o1, o0, 0x10, g2, g3, g4, g5, o4, o5, g7, g1, 24, 8, -1)
+	SMOVE_CHUNK(o1, o0, 0x20, g2, g3, g4, g5, o4, o5, g7, g1, 24, 8, -1)
+	SMOVE_CHUNK(o1, o0, 0x30, g2, g3, g4, g5, o4, o5, g7, 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, g7, 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, g7, g1, 8, 24, -3)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x10, g2, g3, g4, g5, o4, o5, g7, g1, 8, 24, -3)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x20, g2, g3, g4, g5, o4, o5, g7, g1, 8, 24, -3)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x30, g2, g3, g4, g5, o4, o5, g7, 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, g7, 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, g7, g1, 24, 8, 3)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x10, g2, g3, g4, g5, o4, o5, g7, g1, 24, 8, 3)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x20, g2, g3, g4, g5, o4, o5, g7, g1, 24, 8, 3)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x30, g2, g3, g4, g5, o4, o5, g7, 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, g7, 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, g7, g1, 16, 16, -2)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x10, g2, g3, g4, g5, o4, o5, g7, g1, 16, 16, -2)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x20, g2, g3, g4, g5, o4, o5, g7, g1, 16, 16, -2)
+	SMOVE_ALIGNCHUNK(o1, o0, 0x30, g2, g3, g4, g5, o4, o5, g7, 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, g7, 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, %g7, %o5
+END(memcpy)