x86-64: Optimize strcat/strncat, strcpy/strncpy and stpcpy/stpncpy with AVX2
Optimize x86-64 strcat/strncat, strcpy/strncpy and stpcpy/stpncpy with AVX2.
It uses vector comparison as much as possible. In general, the larger the
source string, the greater performance gain observed, reaching speedups of
1.6x compared to SSE2 unaligned routines. Select AVX2 strcat/strncat,
strcpy/strncpy and stpcpy/stpncpy on AVX2 machines where vzeroupper is
preferred and AVX unaligned load is fast.
* sysdeps/x86_64/multiarch/Makefile (sysdep_routines): Add
strcat-avx2, strncat-avx2, strcpy-avx2, strncpy-avx2,
stpcpy-avx2 and stpncpy-avx2.
* sysdeps/x86_64/multiarch/ifunc-impl-list.c:
(__libc_ifunc_impl_list): Add tests for __strcat_avx2,
__strncat_avx2, __strcpy_avx2, __strncpy_avx2, __stpcpy_avx2
and __stpncpy_avx2.
* sysdeps/x86_64/multiarch/{ifunc-unaligned-ssse3.h =>
ifunc-strcpy.h}: rename header for a more generic name.
* sysdeps/x86_64/multiarch/ifunc-strcpy.h:
(IFUNC_SELECTOR): Return OPTIMIZE (avx2) on AVX 2 machines if
AVX unaligned load is fast and vzeroupper is preferred.
* sysdeps/x86_64/multiarch/stpcpy-avx2.S: New file
* sysdeps/x86_64/multiarch/stpncpy-avx2.S: Likewise
* sysdeps/x86_64/multiarch/strcat-avx2.S: Likewise
* sysdeps/x86_64/multiarch/strcpy-avx2.S: Likewise
* sysdeps/x86_64/multiarch/strncat-avx2.S: Likewise
* sysdeps/x86_64/multiarch/strncpy-avx2.S: Likewise
This commit is contained in:
Leonardo Sandoval
parent
a1b02ae763
commit
1a153e47fc
21
ChangeLog
21
ChangeLog
|
|
@ -1,3 +1,24 @@
|
|||
2019-01-14 Leonardo Sandoval <leonardo.sandoval.gonzalez@intel.com>
|
||||
|
||||
* sysdeps/x86_64/multiarch/Makefile (sysdep_routines): Add
|
||||
strcat-avx2, strncat-avx2, strcpy-avx2, strncpy-avx2,
|
||||
stpcpy-avx2 and stpncpy-avx2.
|
||||
* sysdeps/x86_64/multiarch/ifunc-impl-list.c:
|
||||
(__libc_ifunc_impl_list): Add tests for __strcat_avx2,
|
||||
__strncat_avx2, __strcpy_avx2, __strncpy_avx2, __stpcpy_avx2
|
||||
and __stpncpy_avx2.
|
||||
* sysdeps/x86_64/multiarch/{ifunc-unaligned-ssse3.h =>
|
||||
ifunc-strcpy.h}: rename header for a more generic name.
|
||||
* sysdeps/x86_64/multiarch/ifunc-strcpy.h:
|
||||
(IFUNC_SELECTOR): Return OPTIMIZE (avx2) on AVX 2 machines if
|
||||
AVX unaligned load is fast and vzeroupper is preferred.
|
||||
* sysdeps/x86_64/multiarch/stpcpy-avx2.S: New file.
|
||||
* sysdeps/x86_64/multiarch/stpncpy-avx2.S: Likewise.
|
||||
* sysdeps/x86_64/multiarch/strcat-avx2.S: Likewise.
|
||||
* sysdeps/x86_64/multiarch/strcpy-avx2.S: Likewise.
|
||||
* sysdeps/x86_64/multiarch/strncat-avx2.S: Likewise.
|
||||
* sysdeps/x86_64/multiarch/strncpy-avx2.S: Likewise.
|
||||
|
||||
2019-01-12 Dmitry V. Levin <ldv@altlinux.org>
|
||||
|
||||
* argp/argp-help.c: Fix typo in comment.
|
||||
|
|
|
|||
|
|
@ -24,11 +24,14 @@ sysdep_routines += strncat-c stpncpy-c strncpy-c \
|
|||
strchr-sse2 strchrnul-sse2 strchr-avx2 strchrnul-avx2 \
|
||||
strrchr-sse2 strrchr-avx2 \
|
||||
strlen-sse2 strnlen-sse2 strlen-avx2 strnlen-avx2 \
|
||||
strcat-avx2 strncat-avx2 \
|
||||
strcat-ssse3 strncat-ssse3\
|
||||
strcpy-avx2 strncpy-avx2 \
|
||||
strcpy-sse2 stpcpy-sse2 \
|
||||
strcpy-ssse3 strncpy-ssse3 stpcpy-ssse3 stpncpy-ssse3 \
|
||||
strcpy-sse2-unaligned strncpy-sse2-unaligned \
|
||||
stpcpy-sse2-unaligned stpncpy-sse2-unaligned \
|
||||
stpcpy-avx2 stpncpy-avx2 \
|
||||
strcat-sse2 \
|
||||
strcat-sse2-unaligned strncat-sse2-unaligned \
|
||||
strchr-sse2-no-bsf memcmp-ssse3 strstr-sse2-unaligned \
|
||||
|
|
|
|||
|
|
@ -199,6 +199,8 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array,
|
|||
IFUNC_IMPL (i, name, stpncpy,
|
||||
IFUNC_IMPL_ADD (array, i, stpncpy, HAS_CPU_FEATURE (SSSE3),
|
||||
__stpncpy_ssse3)
|
||||
IFUNC_IMPL_ADD (array, i, stpncpy, HAS_ARCH_FEATURE (AVX2_Usable),
|
||||
__stpncpy_avx2)
|
||||
IFUNC_IMPL_ADD (array, i, stpncpy, 1,
|
||||
__stpncpy_sse2_unaligned)
|
||||
IFUNC_IMPL_ADD (array, i, stpncpy, 1, __stpncpy_sse2))
|
||||
|
|
@ -207,6 +209,8 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array,
|
|||
IFUNC_IMPL (i, name, stpcpy,
|
||||
IFUNC_IMPL_ADD (array, i, stpcpy, HAS_CPU_FEATURE (SSSE3),
|
||||
__stpcpy_ssse3)
|
||||
IFUNC_IMPL_ADD (array, i, stpcpy, HAS_ARCH_FEATURE (AVX2_Usable),
|
||||
__stpcpy_avx2)
|
||||
IFUNC_IMPL_ADD (array, i, stpcpy, 1, __stpcpy_sse2_unaligned)
|
||||
IFUNC_IMPL_ADD (array, i, stpcpy, 1, __stpcpy_sse2))
|
||||
|
||||
|
|
@ -239,6 +243,8 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array,
|
|||
|
||||
/* Support sysdeps/x86_64/multiarch/strcat.c. */
|
||||
IFUNC_IMPL (i, name, strcat,
|
||||
IFUNC_IMPL_ADD (array, i, strcat, HAS_ARCH_FEATURE (AVX2_Usable),
|
||||
__strcat_avx2)
|
||||
IFUNC_IMPL_ADD (array, i, strcat, HAS_CPU_FEATURE (SSSE3),
|
||||
__strcat_ssse3)
|
||||
IFUNC_IMPL_ADD (array, i, strcat, 1, __strcat_sse2_unaligned)
|
||||
|
|
@ -280,6 +286,8 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array,
|
|||
|
||||
/* Support sysdeps/x86_64/multiarch/strcpy.c. */
|
||||
IFUNC_IMPL (i, name, strcpy,
|
||||
IFUNC_IMPL_ADD (array, i, strcpy, HAS_ARCH_FEATURE (AVX2_Usable),
|
||||
__strcpy_avx2)
|
||||
IFUNC_IMPL_ADD (array, i, strcpy, HAS_CPU_FEATURE (SSSE3),
|
||||
__strcpy_ssse3)
|
||||
IFUNC_IMPL_ADD (array, i, strcpy, 1, __strcpy_sse2_unaligned)
|
||||
|
|
@ -321,6 +329,8 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array,
|
|||
|
||||
/* Support sysdeps/x86_64/multiarch/strncat.c. */
|
||||
IFUNC_IMPL (i, name, strncat,
|
||||
IFUNC_IMPL_ADD (array, i, strncat, HAS_ARCH_FEATURE (AVX2_Usable),
|
||||
__strncat_avx2)
|
||||
IFUNC_IMPL_ADD (array, i, strncat, HAS_CPU_FEATURE (SSSE3),
|
||||
__strncat_ssse3)
|
||||
IFUNC_IMPL_ADD (array, i, strncat, 1,
|
||||
|
|
@ -329,6 +339,8 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array,
|
|||
|
||||
/* Support sysdeps/x86_64/multiarch/strncpy.c. */
|
||||
IFUNC_IMPL (i, name, strncpy,
|
||||
IFUNC_IMPL_ADD (array, i, strncpy, HAS_ARCH_FEATURE (AVX2_Usable),
|
||||
__strncpy_avx2)
|
||||
IFUNC_IMPL_ADD (array, i, strncpy, HAS_CPU_FEATURE (SSSE3),
|
||||
__strncpy_ssse3)
|
||||
IFUNC_IMPL_ADD (array, i, strncpy, 1,
|
||||
|
|
|
|||
|
|
@ -24,12 +24,18 @@ extern __typeof (REDIRECT_NAME) OPTIMIZE (sse2) attribute_hidden;
|
|||
extern __typeof (REDIRECT_NAME) OPTIMIZE (sse2_unaligned)
|
||||
attribute_hidden;
|
||||
extern __typeof (REDIRECT_NAME) OPTIMIZE (ssse3) attribute_hidden;
|
||||
extern __typeof (REDIRECT_NAME) OPTIMIZE (avx2) attribute_hidden;
|
||||
|
||||
static inline void *
|
||||
IFUNC_SELECTOR (void)
|
||||
{
|
||||
const struct cpu_features* cpu_features = __get_cpu_features ();
|
||||
|
||||
if (!CPU_FEATURES_ARCH_P (cpu_features, Prefer_No_VZEROUPPER)
|
||||
&& CPU_FEATURES_ARCH_P (cpu_features, AVX2_Usable)
|
||||
&& CPU_FEATURES_ARCH_P (cpu_features, AVX_Fast_Unaligned_Load))
|
||||
return OPTIMIZE (avx2);
|
||||
|
||||
if (CPU_FEATURES_ARCH_P (cpu_features, Fast_Unaligned_Load))
|
||||
return OPTIMIZE (sse2_unaligned);
|
||||
|
||||
|
|
@ -0,0 +1,3 @@
|
|||
#define USE_AS_STPCPY
|
||||
#define STRCPY __stpcpy_avx2
|
||||
#include "strcpy-avx2.S"
|
||||
|
|
@ -28,7 +28,7 @@
|
|||
# undef __stpcpy
|
||||
|
||||
# define SYMBOL_NAME stpcpy
|
||||
# include "ifunc-unaligned-ssse3.h"
|
||||
# include "ifunc-strcpy.h"
|
||||
|
||||
libc_ifunc_redirected (__redirect_stpcpy, __stpcpy, IFUNC_SELECTOR ());
|
||||
|
||||
|
|
|
|||
|
|
@ -0,0 +1,4 @@
|
|||
#define USE_AS_STPCPY
|
||||
#define USE_AS_STRNCPY
|
||||
#define STRCPY __stpncpy_avx2
|
||||
#include "strcpy-avx2.S"
|
||||
|
|
@ -26,7 +26,7 @@
|
|||
# undef __stpncpy
|
||||
|
||||
# define SYMBOL_NAME stpncpy
|
||||
# include "ifunc-unaligned-ssse3.h"
|
||||
# include "ifunc-strcpy.h"
|
||||
|
||||
libc_ifunc_redirected (__redirect_stpncpy, __stpncpy, IFUNC_SELECTOR ());
|
||||
|
||||
|
|
|
|||
|
|
@ -0,0 +1,275 @@
|
|||
/* strcat with AVX2
|
||||
Copyright (C) 2011-2018 Free Software Foundation, Inc.
|
||||
Contributed by Intel Corporation.
|
||||
This file is part of the GNU C Library.
|
||||
|
||||
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/>. */
|
||||
|
||||
#if IS_IN (libc)
|
||||
|
||||
# include <sysdep.h>
|
||||
|
||||
# ifndef STRCAT
|
||||
# define STRCAT __strcat_avx2
|
||||
# endif
|
||||
|
||||
# define USE_AS_STRCAT
|
||||
|
||||
/* Number of bytes in a vector register */
|
||||
# define VEC_SIZE 32
|
||||
|
||||
.section .text.avx,"ax",@progbits
|
||||
ENTRY (STRCAT)
|
||||
mov %rdi, %r9
|
||||
# ifdef USE_AS_STRNCAT
|
||||
mov %rdx, %r8
|
||||
# endif
|
||||
|
||||
xor %eax, %eax
|
||||
mov %edi, %ecx
|
||||
and $((VEC_SIZE * 4) - 1), %ecx
|
||||
vpxor %xmm6, %xmm6, %xmm6
|
||||
cmp $(VEC_SIZE * 3), %ecx
|
||||
ja L(fourth_vector_boundary)
|
||||
vpcmpeqb (%rdi), %ymm6, %ymm0
|
||||
vpmovmskb %ymm0, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_first_vector)
|
||||
mov %rdi, %rax
|
||||
and $-VEC_SIZE, %rax
|
||||
jmp L(align_vec_size_start)
|
||||
L(fourth_vector_boundary):
|
||||
mov %rdi, %rax
|
||||
and $-VEC_SIZE, %rax
|
||||
vpcmpeqb (%rax), %ymm6, %ymm0
|
||||
mov $-1, %r10d
|
||||
sub %rax, %rcx
|
||||
shl %cl, %r10d
|
||||
vpmovmskb %ymm0, %edx
|
||||
and %r10d, %edx
|
||||
jnz L(exit)
|
||||
|
||||
L(align_vec_size_start):
|
||||
vpcmpeqb VEC_SIZE(%rax), %ymm6, %ymm0
|
||||
vpmovmskb %ymm0, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_second_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 2)(%rax), %ymm6, %ymm1
|
||||
vpmovmskb %ymm1, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_third_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 3)(%rax), %ymm6, %ymm2
|
||||
vpmovmskb %ymm2, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_fourth_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 4)(%rax), %ymm6, %ymm3
|
||||
vpmovmskb %ymm3, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_fifth_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 5)(%rax), %ymm6, %ymm0
|
||||
add $(VEC_SIZE * 4), %rax
|
||||
vpmovmskb %ymm0, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_second_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 2)(%rax), %ymm6, %ymm1
|
||||
vpmovmskb %ymm1, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_third_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 3)(%rax), %ymm6, %ymm2
|
||||
vpmovmskb %ymm2, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_fourth_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 4)(%rax), %ymm6, %ymm3
|
||||
vpmovmskb %ymm3, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_fifth_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 5)(%rax), %ymm6, %ymm0
|
||||
add $(VEC_SIZE * 4), %rax
|
||||
vpmovmskb %ymm0, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_second_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 2)(%rax), %ymm6, %ymm1
|
||||
vpmovmskb %ymm1, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_third_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 3)(%rax), %ymm6, %ymm2
|
||||
vpmovmskb %ymm2, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_fourth_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 4)(%rax), %ymm6, %ymm3
|
||||
vpmovmskb %ymm3, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_fifth_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 5)(%rax), %ymm6, %ymm0
|
||||
add $(VEC_SIZE * 4), %rax
|
||||
vpmovmskb %ymm0, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_second_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 2)(%rax), %ymm6, %ymm1
|
||||
vpmovmskb %ymm1, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_third_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 3)(%rax), %ymm6, %ymm2
|
||||
vpmovmskb %ymm2, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_fourth_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 4)(%rax), %ymm6, %ymm3
|
||||
vpmovmskb %ymm3, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_fifth_vector)
|
||||
|
||||
test $((VEC_SIZE * 4) - 1), %rax
|
||||
jz L(align_four_vec_loop)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 5)(%rax), %ymm6, %ymm0
|
||||
add $(VEC_SIZE * 5), %rax
|
||||
vpmovmskb %ymm0, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit)
|
||||
|
||||
test $((VEC_SIZE * 4) - 1), %rax
|
||||
jz L(align_four_vec_loop)
|
||||
|
||||
vpcmpeqb VEC_SIZE(%rax), %ymm6, %ymm1
|
||||
add $VEC_SIZE, %rax
|
||||
vpmovmskb %ymm1, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit)
|
||||
|
||||
test $((VEC_SIZE * 4) - 1), %rax
|
||||
jz L(align_four_vec_loop)
|
||||
|
||||
vpcmpeqb VEC_SIZE(%rax), %ymm6, %ymm2
|
||||
add $VEC_SIZE, %rax
|
||||
vpmovmskb %ymm2, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit)
|
||||
|
||||
test $((VEC_SIZE * 4) - 1), %rax
|
||||
jz L(align_four_vec_loop)
|
||||
|
||||
vpcmpeqb VEC_SIZE(%rax), %ymm6, %ymm3
|
||||
add $VEC_SIZE, %rax
|
||||
vpmovmskb %ymm3, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit)
|
||||
|
||||
add $VEC_SIZE, %rax
|
||||
|
||||
.p2align 4
|
||||
L(align_four_vec_loop):
|
||||
vmovaps (%rax), %ymm4
|
||||
vpminub VEC_SIZE(%rax), %ymm4, %ymm4
|
||||
vmovaps (VEC_SIZE * 2)(%rax), %ymm5
|
||||
vpminub (VEC_SIZE * 3)(%rax), %ymm5, %ymm5
|
||||
add $(VEC_SIZE * 4), %rax
|
||||
vpminub %ymm4, %ymm5, %ymm5
|
||||
vpcmpeqb %ymm5, %ymm6, %ymm5
|
||||
vpmovmskb %ymm5, %edx
|
||||
test %edx, %edx
|
||||
jz L(align_four_vec_loop)
|
||||
|
||||
vpcmpeqb -(VEC_SIZE * 4)(%rax), %ymm6, %ymm0
|
||||
sub $(VEC_SIZE * 5), %rax
|
||||
vpmovmskb %ymm0, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_second_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 2)(%rax), %ymm6, %ymm1
|
||||
vpmovmskb %ymm1, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_third_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 3)(%rax), %ymm6, %ymm2
|
||||
vpmovmskb %ymm2, %edx
|
||||
test %edx, %edx
|
||||
jnz L(exit_null_on_fourth_vector)
|
||||
|
||||
vpcmpeqb (VEC_SIZE * 4)(%rax), %ymm6, %ymm3
|
||||
vpmovmskb %ymm3, %edx
|
||||
sub %rdi, %rax
|
||||
bsf %rdx, %rdx
|
||||
add %rdx, %rax
|
||||
add $(VEC_SIZE * 4), %rax
|
||||
jmp L(StartStrcpyPart)
|
||||
|
||||
.p2align 4
|
||||
L(exit):
|
||||
sub %rdi, %rax
|
||||
L(exit_null_on_first_vector):
|
||||
bsf %rdx, %rdx
|
||||
add %rdx, %rax
|
||||
jmp L(StartStrcpyPart)
|
||||
|
||||
.p2align 4
|
||||
L(exit_null_on_second_vector):
|
||||
sub %rdi, %rax
|
||||
bsf %rdx, %rdx
|
||||
add %rdx, %rax
|
||||
add $VEC_SIZE, %rax
|
||||
jmp L(StartStrcpyPart)
|
||||
|
||||
.p2align 4
|
||||
L(exit_null_on_third_vector):
|
||||
sub %rdi, %rax
|
||||
bsf %rdx, %rdx
|
||||
add %rdx, %rax
|
||||
add $(VEC_SIZE * 2), %rax
|
||||
jmp L(StartStrcpyPart)
|
||||
|
||||
.p2align 4
|
||||
L(exit_null_on_fourth_vector):
|
||||
sub %rdi, %rax
|
||||
bsf %rdx, %rdx
|
||||
add %rdx, %rax
|
||||
add $(VEC_SIZE * 3), %rax
|
||||
jmp L(StartStrcpyPart)
|
||||
|
||||
.p2align 4
|
||||
L(exit_null_on_fifth_vector):
|
||||
sub %rdi, %rax
|
||||
bsf %rdx, %rdx
|
||||
add %rdx, %rax
|
||||
add $(VEC_SIZE * 4), %rax
|
||||
|
||||
.p2align 4
|
||||
L(StartStrcpyPart):
|
||||
lea (%r9, %rax), %rdi
|
||||
mov %rsi, %rcx
|
||||
mov %r9, %rax /* save result */
|
||||
|
||||
# ifdef USE_AS_STRNCAT
|
||||
test %r8, %r8
|
||||
jz L(ExitZero)
|
||||
# define USE_AS_STRNCPY
|
||||
# endif
|
||||
|
||||
# include "strcpy-avx2.S"
|
||||
#endif
|
||||
|
|
@ -24,7 +24,7 @@
|
|||
# undef strcat
|
||||
|
||||
# define SYMBOL_NAME strcat
|
||||
# include "ifunc-unaligned-ssse3.h"
|
||||
# include "ifunc-strcpy.h"
|
||||
|
||||
libc_ifunc_redirected (__redirect_strcat, strcat, IFUNC_SELECTOR ());
|
||||
|
||||
|
|
|
|||
File diff suppressed because it is too large
Load Diff
|
|
@ -24,7 +24,7 @@
|
|||
# undef strcpy
|
||||
|
||||
# define SYMBOL_NAME strcpy
|
||||
# include "ifunc-unaligned-ssse3.h"
|
||||
# include "ifunc-strcpy.h"
|
||||
|
||||
libc_ifunc_redirected (__redirect_strcpy, strcpy, IFUNC_SELECTOR ());
|
||||
|
||||
|
|
|
|||
|
|
@ -0,0 +1,3 @@
|
|||
#define USE_AS_STRNCAT
|
||||
#define STRCAT __strncat_avx2
|
||||
#include "strcat-avx2.S"
|
||||
|
|
@ -24,7 +24,7 @@
|
|||
# undef strncat
|
||||
|
||||
# define SYMBOL_NAME strncat
|
||||
# include "ifunc-unaligned-ssse3.h"
|
||||
# include "ifunc-strcpy.h"
|
||||
|
||||
libc_ifunc_redirected (__redirect_strncat, strncat, IFUNC_SELECTOR ());
|
||||
strong_alias (strncat, __strncat);
|
||||
|
|
|
|||
|
|
@ -0,0 +1,3 @@
|
|||
#define USE_AS_STRNCPY
|
||||
#define STRCPY __strncpy_avx2
|
||||
#include "strcpy-avx2.S"
|
||||
|
|
@ -24,7 +24,7 @@
|
|||
# undef strncpy
|
||||
|
||||
# define SYMBOL_NAME strncpy
|
||||
# include "ifunc-unaligned-ssse3.h"
|
||||
# include "ifunc-strcpy.h"
|
||||
|
||||
libc_ifunc_redirected (__redirect_strncpy, strncpy, IFUNC_SELECTOR ());
|
||||
|
||||
|
|
|
|||
Loading…
Reference in New Issue