/* SPDX-License-Identifier: GPL-2.0 */
/* Copyright 2002 Andi Kleen, SuSE Labs */

#include <linux/export.h>
#include <linux/linkage.h>
#include <linux/cfi_types.h>
#include <asm/cpufeatures.h>
#include <asm/alternative.h>

.section .noinstr.text, "ax"

/*
 * ISO C memset - set a memory block to a byte value. This function uses fast
 * string to get better performance than the original function. The code is
 * simpler and shorter than the original function as well.
 *
 * rdi   destination
 * rsi   value (char)
 * rdx   count (bytes)
 *
 * rax   original destination
 *
 * The FSRS alternative should be done inline (avoiding the call and
 * the disgusting return handling), but that would require some help
 * from the compiler for better calling conventions.
 *
 * The 'rep stosb' itself is small enough to replace the call, but all
 * the register moves blow up the code. And two of them are "needed"
 * only for the return value that is the same as the source input,
 * which the compiler could/should do much better anyway.
 */
SYM_TYPED_FUNC_START(__memset)
        ALTERNATIVE "jmp memset_orig", "", X86_FEATURE_FSRS

        movq %rdi,%r9
        movb %sil,%al
        movq %rdx,%rcx
        rep stosb
        movq %r9,%rax
        RET
SYM_FUNC_END(__memset)
EXPORT_SYMBOL(__memset)

SYM_FUNC_ALIAS_MEMFUNC(memset, __memset)
SYM_PIC_ALIAS(memset)
EXPORT_SYMBOL(memset)

SYM_FUNC_START_LOCAL(memset_orig)
        movq %rdi,%r10

        /* expand byte value  */
        movzbl %sil,%ecx
        movabs $0x0101010101010101,%rax
        imulq  %rcx,%rax

        /* align dst */
        movl  %edi,%r9d
        andl  $7,%r9d
        jnz  .Lbad_alignment
.Lafter_bad_alignment:

        movq  %rdx,%rcx
        shrq  $6,%rcx
        jz       .Lhandle_tail

        .p2align 4
.Lloop_64:
        decq  %rcx
        movq  %rax,(%rdi)
        movq  %rax,8(%rdi)
        movq  %rax,16(%rdi)
        movq  %rax,24(%rdi)
        movq  %rax,32(%rdi)
        movq  %rax,40(%rdi)
        movq  %rax,48(%rdi)
        movq  %rax,56(%rdi)
        leaq  64(%rdi),%rdi
        jnz    .Lloop_64

        /* Handle tail in loops. The loops should be faster than hard
           to predict jump tables. */
        .p2align 4
.Lhandle_tail:
        movl    %edx,%ecx
        andl    $63&(~7),%ecx
        jz              .Lhandle_7
        shrl    $3,%ecx
        .p2align 4
.Lloop_8:
        decl   %ecx
        movq  %rax,(%rdi)
        leaq  8(%rdi),%rdi
        jnz    .Lloop_8

.Lhandle_7:
        andl    $7,%edx
        jz      .Lende
        .p2align 4
.Lloop_1:
        decl    %edx
        movb    %al,(%rdi)
        leaq    1(%rdi),%rdi
        jnz     .Lloop_1

.Lende:
        movq    %r10,%rax
        RET

.Lbad_alignment:
        cmpq $7,%rdx
        jbe     .Lhandle_7
        movq %rax,(%rdi)        /* unaligned store */
        movq $8,%r8
        subq %r9,%r8
        addq %r8,%rdi
        subq %r8,%rdx
        jmp .Lafter_bad_alignment
.Lfinal:
SYM_FUNC_END(memset_orig)