/*      $NetBSD: memset.S,v 1.5 2014/05/23 03:17:31 uebayasi Exp $      */

/*-
 * Copyright (c) 2003 The NetBSD Foundation, Inc.
 * All rights reserved.
 *
 * This code is derived from software contributed to The NetBSD Foundation
 * by David Laight.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 *
 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
 * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
 * POSSIBILITY OF SUCH DAMAGE.
 */

#ifdef __HAIKU__
#include <asm_defs.h>
#endif

#ifdef BZERO
ENTRY(bzero)
#else
.align 4
FUNCTION(memset):
#endif
#ifdef BZERO
        movl    8(%esp),%ecx
        xor     %eax,%eax
#else
        movl    12(%esp),%ecx
        movzbl  8(%esp),%eax            /* unsigned char, zero extend */
#endif
        cmpl    $0x0f,%ecx              /* avoid mispredicted branch... */

        pushl   %edi
        movl    8(%esp),%edi

        /*
         * if the string is too short, it's really not worth the overhead
         * of aligning to word boundries, etc.  So we jump to a plain
         * unaligned set.
         *
         * NB aligning the transfer is actually pointless on my athlon 700,
         * It does make a difference to a PII though.
         *
         * The PII, PIII and PIV all seem to have a massive performance
         * drop when the initial target address is an odd multiple of 4.
         */
        jbe     .Lby_bytes

#ifndef BZERO
        movb    %al,%ah                 /* copy char to all bytes in word */
        movl    %eax,%edx
        sall    $16,%eax
        orl     %edx,%eax
#endif

        movl    %edi,%edx               /* detect misalignment */
        neg     %edx
        andl    $7,%edx
        jnz     .Lalign
.Laligned:
        movl    %eax,-4(%edi,%ecx)      /* zap last 4 bytes */
        shrl    $2,%ecx                 /* zero by words */
        rep
        stosl
.Ldone:
#ifndef BZERO
        movl    8(%esp),%eax            /* return address of buffer */
#endif
        pop     %edi
        ret

.Lalign:
        movl    %eax,(%edi)             /* zap first 8 bytes */
        movl    %eax,4(%edi)
        subl    %edx,%ecx               /* remove from main count */
        add     %edx,%edi
        jmp     .Laligned

.Lby_bytes:
        rep
        stosb

#ifndef BZERO
        movl    8(%esp),%eax            /* return address of buffer */
#endif
        popl    %edi
        ret
#ifdef BZERO
END(bzero)
#else
FUNCTION_END(memset)
SYMBOL(memset_end):
#endif