/* Highly optimized version for i586.

   Copyright (C) 1997-2018 Free Software Foundation, Inc.

   This file is part of the GNU C Library.

   Contributed by Ulrich Drepper <drepper@cygnus.com>, 1997.

   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>

#include "asm-syntax.h"

#define PARMS	4+8	/* space for 2 saved regs */

#define RTN	PARMS

#define DEST	RTN

#define SRC	DEST+4

#define LEN	SRC+4

        .text

#if defined PIC && IS_IN (libc)

ENTRY (__memcpy_chk)

	movl	12(%esp), %eax

	cmpl	%eax, 16(%esp)

	jb	HIDDEN_JUMPTARGET (__chk_fail)

END (__memcpy_chk)

#endif

ENTRY (memcpy)

	pushl	%edi

	cfi_adjust_cfa_offset (4)

	pushl	%esi

	cfi_adjust_cfa_offset (4)

	movl	DEST(%esp), %edi

	cfi_rel_offset (edi, 4)

	movl	SRC(%esp), %esi

	cfi_rel_offset (esi, 0)

	movl	LEN(%esp), %ecx

	movl	%edi, %eax

	/* We need this in any case.  */

	cld

	/* Cutoff for the big loop is a size of 32 bytes since otherwise

	   the loop will never be entered.  */

	cmpl	$32, %ecx

	jbe	L(1)

	negl	%eax

	andl	$3, %eax

	subl	%eax, %ecx

	xchgl	%eax, %ecx

	rep; movsb

	movl	%eax, %ecx

	subl	$32, %ecx

	js	L(2)

	/* Read ahead to make sure we write in the cache since the stupid

	   i586 designers haven't implemented read-on-write-miss.  */

	movl	(%edi), %eax

L(3):	movl	28(%edi), %edx

	/* Now correct the loop counter.  Please note that in the following

	   code the flags are not changed anymore.  */

	subl	$32, %ecx

	movl	(%esi), %eax

	movl	4(%esi), %edx

	movl	%eax, (%edi)

	movl	%edx, 4(%edi)

	movl	8(%esi), %eax

	movl	12(%esi), %edx

	movl	%eax, 8(%edi)

	movl	%edx, 12(%edi)

	movl	16(%esi), %eax

	movl	20(%esi), %edx

	movl	%eax, 16(%edi)

	movl	%edx, 20(%edi)

	movl	24(%esi), %eax

	movl	28(%esi), %edx

	movl	%eax, 24(%edi)

	movl	%edx, 28(%edi)

	leal	32(%esi), %esi

	leal	32(%edi), %edi

	jns	L(3)

	/* Correct extra loop counter modification.  */

L(2):	addl	$32, %ecx

#ifndef USE_AS_MEMPCPY

	movl	DEST(%esp), %eax

#endif

L(1):	rep; movsb

#ifdef USE_AS_MEMPCPY

	movl	%edi, %eax

#endif

	popl	%esi

	cfi_adjust_cfa_offset (-4)

	cfi_restore (esi)

	popl	%edi

	cfi_adjust_cfa_offset (-4)

	cfi_restore (edi)

	ret

END (memcpy)

#ifndef USE_AS_MEMPCPY

libc_hidden_builtin_def (memcpy)

#endif