x86-64: Optimize strlen/strnlen/wcslen/wcsnlen with AVX2

Optimize strlen/strnlen/wcslen/wcsnlen with AVX2 to check 32 bytes with
a single vector compare instruction.  It is as fast as SSE2 versions for
size <= 16 bytes and up to 1X faster for or size > 16 bytes on Haswell.
Select AVX2 version on AVX2 machines where vzeroupper is preferred and
AVX unaligned load is fast.

NB: It uses TZCNT instead of BSF since TZCNT produces the same result
as BSF for non-zero input.  TZCNT is faster than BSF and is executed
as BSF if machine doesn't support TZCNT.

	* sysdeps/x86_64/multiarch/Makefile (sysdep_routines): Add
	strlen-sse2, strnlen-sse2, strlen-avx2, strnlen-avx2,
	wcslen-sse2, wcslen-avx2 and wcsnlen-avx2.
	* sysdeps/x86_64/multiarch/ifunc-impl-list.c
	(__libc_ifunc_impl_list): Add tests for __strlen_avx2,
	__strlen_sse2, __strnlen_avx2, __strnlen_sse2, __wcslen_avx2,
	__wcslen_sse2 and __wcsnlen_avx2.
	* sysdeps/x86_64/multiarch/strlen-avx2.S: New file.
	* sysdeps/x86_64/multiarch/strlen-sse2.S: Likewise.
	* sysdeps/x86_64/multiarch/strlen.c: Likewise.
	* sysdeps/x86_64/multiarch/strnlen-avx2.S: Likewise.
	* sysdeps/x86_64/multiarch/strnlen-sse2.S: Likewise.
	* sysdeps/x86_64/multiarch/strnlen.c: Likewise.
	* sysdeps/x86_64/multiarch/wcslen-avx2.S: Likewise.
	* sysdeps/x86_64/multiarch/wcslen-sse2.S: Likewise.
	* sysdeps/x86_64/multiarch/wcslen.c: Likewise.
	* sysdeps/x86_64/multiarch/wcsnlen-avx2.S: Likewise.
	* sysdeps/x86_64/multiarch/wcsnlen.c (OPTIMIZE (avx2)): New.
	(IFUNC_SELECTOR): Return OPTIMIZE (avx2) on AVX2 machines where
	vzeroupper is preferred and AVX unaligned load is fast.
This commit is contained in:
H.J. Lu 2017-06-09 05:18:03 -07:00
parent 2f5d20ac99
commit dc485ceb2a
14 changed files with 644 additions and 1 deletions

View File

@ -1,3 +1,26 @@
2017-06-09 H.J. Lu <hongjiu.lu@intel.com>
* sysdeps/x86_64/multiarch/Makefile (sysdep_routines): Add
strlen-sse2, strnlen-sse2, strlen-avx2, strnlen-avx2,
wcslen-sse2, wcslen-avx2 and wcsnlen-avx2.
* sysdeps/x86_64/multiarch/ifunc-impl-list.c
(__libc_ifunc_impl_list): Add tests for __strlen_avx2,
__strlen_sse2, __strnlen_avx2, __strnlen_sse2, __wcslen_avx2,
__wcslen_sse2 and __wcsnlen_avx2.
* sysdeps/x86_64/multiarch/strlen-avx2.S: New file.
* sysdeps/x86_64/multiarch/strlen-sse2.S: Likewise.
* sysdeps/x86_64/multiarch/strlen.c: Likewise.
* sysdeps/x86_64/multiarch/strnlen-avx2.S: Likewise.
* sysdeps/x86_64/multiarch/strnlen-sse2.S: Likewise.
* sysdeps/x86_64/multiarch/strnlen.c: Likewise.
* sysdeps/x86_64/multiarch/wcslen-avx2.S: Likewise.
* sysdeps/x86_64/multiarch/wcslen-sse2.S: Likewise.
* sysdeps/x86_64/multiarch/wcslen.c: Likewise.
* sysdeps/x86_64/multiarch/wcsnlen-avx2.S: Likewise.
* sysdeps/x86_64/multiarch/wcsnlen.c (OPTIMIZE (avx2)): New.
(IFUNC_SELECTOR): Return OPTIMIZE (avx2) on AVX2 machines where
vzeroupper is preferred and AVX unaligned load is fast.
2017-06-09 H.J. Lu <hongjiu.lu@intel.com>
* sysdeps/x86_64/memchr.S (MEMCHR): New. Depending on if

View File

@ -13,6 +13,7 @@ sysdep_routines += strncat-c stpncpy-c strncpy-c strcmp-ssse3 \
memcpy-ssse3-back \
memmove-ssse3-back \
memmove-avx512-no-vzeroupper strcasecmp_l-ssse3 \
strlen-sse2 strnlen-sse2 strlen-avx2 strnlen-avx2 \
strncase_l-ssse3 strcat-ssse3 strncat-ssse3\
strcpy-ssse3 strncpy-ssse3 stpcpy-ssse3 stpncpy-ssse3 \
strcpy-sse2-unaligned strncpy-sse2-unaligned \
@ -36,7 +37,8 @@ sysdep_routines += wmemcmp-sse4 wmemcmp-ssse3 wmemcmp-c \
wmemcmp-avx2-movbe \
wmemchr-sse2 wmemchr-avx2 \
wcscpy-ssse3 wcscpy-c \
wcsnlen-sse4_1 wcsnlen-c
wcsnlen-sse4_1 wcsnlen-c \
wcslen-sse2 wcslen-avx2 wcsnlen-avx2
endif
ifeq ($(subdir),debug)

View File

@ -166,6 +166,20 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array,
__rawmemchr_avx2)
IFUNC_IMPL_ADD (array, i, rawmemchr, 1, __rawmemchr_sse2))
/* Support sysdeps/x86_64/multiarch/strlen.S. */
IFUNC_IMPL (i, name, strlen,
IFUNC_IMPL_ADD (array, i, strlen,
HAS_ARCH_FEATURE (AVX2_Usable),
__strlen_avx2)
IFUNC_IMPL_ADD (array, i, strlen, 1, __strlen_sse2))
/* Support sysdeps/x86_64/multiarch/strnlen.S. */
IFUNC_IMPL (i, name, strnlen,
IFUNC_IMPL_ADD (array, i, strnlen,
HAS_ARCH_FEATURE (AVX2_Usable),
__strnlen_avx2)
IFUNC_IMPL_ADD (array, i, strnlen, 1, __strnlen_sse2))
/* Support sysdeps/x86_64/multiarch/stpncpy.S. */
IFUNC_IMPL (i, name, stpncpy,
IFUNC_IMPL_ADD (array, i, stpncpy, HAS_CPU_FEATURE (SSSE3),
@ -310,8 +324,18 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array,
__wcscpy_ssse3)
IFUNC_IMPL_ADD (array, i, wcscpy, 1, __wcscpy_sse2))
/* Support sysdeps/x86_64/multiarch/wcslen.c. */
IFUNC_IMPL (i, name, wcslen,
IFUNC_IMPL_ADD (array, i, wcslen,
HAS_ARCH_FEATURE (AVX2_Usable),
__wcslen_avx2)
IFUNC_IMPL_ADD (array, i, wcslen, 1, __wcslen_sse2))
/* Support sysdeps/x86_64/multiarch/wcsnlen.c. */
IFUNC_IMPL (i, name, wcsnlen,
IFUNC_IMPL_ADD (array, i, wcsnlen,
HAS_ARCH_FEATURE (AVX2_Usable),
__wcsnlen_avx2)
IFUNC_IMPL_ADD (array, i, wcsnlen,
HAS_CPU_FEATURE (SSE4_1),
__wcsnlen_sse4_1)

View File

@ -0,0 +1,394 @@
/* strlen/strnlen/wcslen/wcsnlen optimized with AVX2.
Copyright (C) 2017 Free Software Foundation, Inc.
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 STRLEN
# define STRLEN __strlen_avx2
# endif
# ifdef USE_AS_WCSLEN
# define VPCMPEQ vpcmpeqd
# define VPMINU vpminud
# else
# define VPCMPEQ vpcmpeqb
# define VPMINU vpminub
# endif
# ifndef VZEROUPPER
# define VZEROUPPER vzeroupper
# endif
# define VEC_SIZE 32
.section .text.avx,"ax",@progbits
ENTRY (STRLEN)
# ifdef USE_AS_STRNLEN
/* Check for zero length. */
testq %rsi, %rsi
jz L(zero)
# ifdef USE_AS_WCSLEN
shl $2, %rsi
# endif
movq %rsi, %r8
# endif
movl %edi, %ecx
movq %rdi, %rdx
vpxor %xmm0, %xmm0, %xmm0
/* Check if we may cross page boundary with one vector load. */
andl $(2 * VEC_SIZE - 1), %ecx
cmpl $VEC_SIZE, %ecx
ja L(cros_page_boundary)
/* Check the first VEC_SIZE bytes. */
VPCMPEQ (%rdi), %ymm0, %ymm1
vpmovmskb %ymm1, %eax
testl %eax, %eax
# ifdef USE_AS_STRNLEN
jnz L(first_vec_x0_check)
/* Adjust length and check the end of data. */
subq $VEC_SIZE, %rsi
jbe L(max)
# else
jnz L(first_vec_x0)
# endif
/* Align data for aligned loads in the loop. */
addq $VEC_SIZE, %rdi
andl $(VEC_SIZE - 1), %ecx
andq $-VEC_SIZE, %rdi
# ifdef USE_AS_STRNLEN
/* Adjust length. */
addq %rcx, %rsi
subq $(VEC_SIZE * 4), %rsi
jbe L(last_4x_vec_or_less)
# endif
jmp L(more_4x_vec)
.p2align 4
L(cros_page_boundary):
andl $(VEC_SIZE - 1), %ecx
andq $-VEC_SIZE, %rdi
VPCMPEQ (%rdi), %ymm0, %ymm1
vpmovmskb %ymm1, %eax
/* Remove the leading bytes. */
sarl %cl, %eax
testl %eax, %eax
jz L(aligned_more)
tzcntl %eax, %eax
# ifdef USE_AS_STRNLEN
/* Check the end of data. */
cmpq %rax, %rsi
jbe L(max)
# endif
addq %rdi, %rax
addq %rcx, %rax
subq %rdx, %rax
# ifdef USE_AS_WCSLEN
shrq $2, %rax
# endif
VZEROUPPER
ret
.p2align 4
L(aligned_more):
# ifdef USE_AS_STRNLEN
/* "rcx" is less than VEC_SIZE. Calculate "rdx + rcx - VEC_SIZE"
with "rdx - (VEC_SIZE - rcx)" instead of "(rdx + rcx) - VEC_SIZE"
to void possible addition overflow. */
negq %rcx
addq $VEC_SIZE, %rcx
/* Check the end of data. */
subq %rcx, %rsi
jbe L(max)
# endif
addq $VEC_SIZE, %rdi
# ifdef USE_AS_STRNLEN
subq $(VEC_SIZE * 4), %rsi
jbe L(last_4x_vec_or_less)
# endif
L(more_4x_vec):
/* Check the first 4 * VEC_SIZE. Only one VEC_SIZE at a time
since data is only aligned to VEC_SIZE. */
VPCMPEQ (%rdi), %ymm0, %ymm1
vpmovmskb %ymm1, %eax
testl %eax, %eax
jnz L(first_vec_x0)
VPCMPEQ VEC_SIZE(%rdi), %ymm0, %ymm1
vpmovmskb %ymm1, %eax
testl %eax, %eax
jnz L(first_vec_x1)
VPCMPEQ (VEC_SIZE * 2)(%rdi), %ymm0, %ymm1
vpmovmskb %ymm1, %eax
testl %eax, %eax
jnz L(first_vec_x2)
VPCMPEQ (VEC_SIZE * 3)(%rdi), %ymm0, %ymm1
vpmovmskb %ymm1, %eax
testl %eax, %eax
jnz L(first_vec_x3)
addq $(VEC_SIZE * 4), %rdi
# ifdef USE_AS_STRNLEN
subq $(VEC_SIZE * 4), %rsi
jbe L(last_4x_vec_or_less)
# endif
/* Align data to 4 * VEC_SIZE. */
movq %rdi, %rcx
andl $(4 * VEC_SIZE - 1), %ecx
andq $-(4 * VEC_SIZE), %rdi
# ifdef USE_AS_STRNLEN
/* Adjust length. */
addq %rcx, %rsi
# endif
.p2align 4
L(loop_4x_vec):
/* Compare 4 * VEC at a time forward. */
vmovdqa (%rdi), %ymm1
vmovdqa VEC_SIZE(%rdi), %ymm2
vmovdqa (VEC_SIZE * 2)(%rdi), %ymm3
vmovdqa (VEC_SIZE * 3)(%rdi), %ymm4
VPMINU %ymm1, %ymm2, %ymm5
VPMINU %ymm3, %ymm4, %ymm6
VPMINU %ymm5, %ymm6, %ymm5
VPCMPEQ %ymm5, %ymm0, %ymm5
vpmovmskb %ymm5, %eax
testl %eax, %eax
jnz L(4x_vec_end)
addq $(VEC_SIZE * 4), %rdi
# ifndef USE_AS_STRNLEN
jmp L(loop_4x_vec)
# else
subq $(VEC_SIZE * 4), %rsi
ja L(loop_4x_vec)
L(last_4x_vec_or_less):
/* Less than 4 * VEC and aligned to VEC_SIZE. */
addl $(VEC_SIZE * 2), %esi
jle L(last_2x_vec)
VPCMPEQ (%rdi), %ymm0, %ymm1
vpmovmskb %ymm1, %eax
testl %eax, %eax
jnz L(first_vec_x0)
VPCMPEQ VEC_SIZE(%rdi), %ymm0, %ymm1
vpmovmskb %ymm1, %eax
testl %eax, %eax
jnz L(first_vec_x1)
VPCMPEQ (VEC_SIZE * 2)(%rdi), %ymm0, %ymm1
vpmovmskb %ymm1, %eax
testl %eax, %eax
jnz L(first_vec_x2_check)
subl $VEC_SIZE, %esi
jle L(max)
VPCMPEQ (VEC_SIZE * 3)(%rdi), %ymm0, %ymm1
vpmovmskb %ymm1, %eax
testl %eax, %eax
jnz L(first_vec_x3_check)
movq %r8, %rax
# ifdef USE_AS_WCSLEN
shrq $2, %rax
# endif
VZEROUPPER
ret
.p2align 4
L(last_2x_vec):
addl $(VEC_SIZE * 2), %esi
VPCMPEQ (%rdi), %ymm0, %ymm1
vpmovmskb %ymm1, %eax
testl %eax, %eax
jnz L(first_vec_x0_check)
subl $VEC_SIZE, %esi
jle L(max)
VPCMPEQ VEC_SIZE(%rdi), %ymm0, %ymm1
vpmovmskb %ymm1, %eax
testl %eax, %eax
jnz L(first_vec_x1_check)
movq %r8, %rax
# ifdef USE_AS_WCSLEN
shrq $2, %rax
# endif
VZEROUPPER
ret
.p2align 4
L(first_vec_x0_check):
tzcntl %eax, %eax
/* Check the end of data. */
cmpq %rax, %rsi
jbe L(max)
addq %rdi, %rax
subq %rdx, %rax
# ifdef USE_AS_WCSLEN
shrq $2, %rax
# endif
VZEROUPPER
ret
.p2align 4
L(first_vec_x1_check):
tzcntl %eax, %eax
/* Check the end of data. */
cmpq %rax, %rsi
jbe L(max)
addq $VEC_SIZE, %rax
addq %rdi, %rax
subq %rdx, %rax
# ifdef USE_AS_WCSLEN
shrq $2, %rax
# endif
VZEROUPPER
ret
.p2align 4
L(first_vec_x2_check):
tzcntl %eax, %eax
/* Check the end of data. */
cmpq %rax, %rsi
jbe L(max)
addq $(VEC_SIZE * 2), %rax
addq %rdi, %rax
subq %rdx, %rax
# ifdef USE_AS_WCSLEN
shrq $2, %rax
# endif
VZEROUPPER
ret
.p2align 4
L(first_vec_x3_check):
tzcntl %eax, %eax
/* Check the end of data. */
cmpq %rax, %rsi
jbe L(max)
addq $(VEC_SIZE * 3), %rax
addq %rdi, %rax
subq %rdx, %rax
# ifdef USE_AS_WCSLEN
shrq $2, %rax
# endif
VZEROUPPER
ret
.p2align 4
L(max):
movq %r8, %rax
# ifdef USE_AS_WCSLEN
shrq $2, %rax
# endif
VZEROUPPER
ret
.p2align 4
L(zero):
xorl %eax, %eax
ret
# endif
.p2align 4
L(first_vec_x0):
tzcntl %eax, %eax
addq %rdi, %rax
subq %rdx, %rax
# ifdef USE_AS_WCSLEN
shrq $2, %rax
# endif
VZEROUPPER
ret
.p2align 4
L(first_vec_x1):
tzcntl %eax, %eax
addq $VEC_SIZE, %rax
addq %rdi, %rax
subq %rdx, %rax
# ifdef USE_AS_WCSLEN
shrq $2, %rax
# endif
VZEROUPPER
ret
.p2align 4
L(first_vec_x2):
tzcntl %eax, %eax
addq $(VEC_SIZE * 2), %rax
addq %rdi, %rax
subq %rdx, %rax
# ifdef USE_AS_WCSLEN
shrq $2, %rax
# endif
VZEROUPPER
ret
.p2align 4
L(4x_vec_end):
VPCMPEQ %ymm1, %ymm0, %ymm1
vpmovmskb %ymm1, %eax
testl %eax, %eax
jnz L(first_vec_x0)
VPCMPEQ %ymm2, %ymm0, %ymm2
vpmovmskb %ymm2, %eax
testl %eax, %eax
jnz L(first_vec_x1)
VPCMPEQ %ymm3, %ymm0, %ymm3
vpmovmskb %ymm3, %eax
testl %eax, %eax
jnz L(first_vec_x2)
VPCMPEQ %ymm4, %ymm0, %ymm4
vpmovmskb %ymm4, %eax
testl %eax, %eax
L(first_vec_x3):
tzcntl %eax, %eax
addq $(VEC_SIZE * 3), %rax
addq %rdi, %rax
subq %rdx, %rax
# ifdef USE_AS_WCSLEN
shrq $2, %rax
# endif
VZEROUPPER
ret
END (STRLEN)
#endif

View File

@ -0,0 +1,23 @@
/* strlen optimized with SSE2.
Copyright (C) 2017 Free Software Foundation, Inc.
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)
# define strlen __strlen_sse2
#endif
#include "../strlen.S"

View File

@ -0,0 +1,34 @@
/* Multiple versions of strlen.
All versions must be listed in ifunc-impl-list.c.
Copyright (C) 2017 Free Software Foundation, Inc.
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/>. */
/* Define multiple versions only for the definition in libc. */
#if IS_IN (libc)
# define strlen __redirect_strlen
# include <string.h>
# undef strlen
# define SYMBOL_NAME strlen
# include "ifunc-avx2.h"
libc_ifunc_redirected (__redirect_strlen, strlen, IFUNC_SELECTOR ());
# ifdef SHARED
__hidden_ver1 (strlen, __GI_strlen, __redirect_strlen)
__attribute__((visibility ("hidden")));
# endif
#endif

View File

@ -0,0 +1,4 @@
#define STRLEN __strnlen_avx2
#define USE_AS_STRNLEN 1
#include "strlen-avx2.S"

View File

@ -0,0 +1,28 @@
/* strnlen optimized with SSE2.
Copyright (C) 2017 Free Software Foundation, Inc.
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)
# define __strnlen __strnlen_sse2
# undef weak_alias
# define weak_alias(__strnlen, strnlen)
# undef libc_hidden_builtin_def
# define libc_hidden_builtin_def(strnlen)
#endif
#include "../strnlen.S"

View File

@ -0,0 +1,39 @@
/* Multiple versions of strnlen.
All versions must be listed in ifunc-impl-list.c.
Copyright (C) 2017 Free Software Foundation, Inc.
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/>. */
/* Define multiple versions only for the definition in libc. */
#if IS_IN (libc)
# define strnlen __redirect_strnlen
# define __strnlen __redirect___strnlen
# include <string.h>
# undef __strnlen
# undef strnlen
# define SYMBOL_NAME strnlen
# include "ifunc-avx2.h"
libc_ifunc_redirected (__redirect_strnlen, __strnlen, IFUNC_SELECTOR ());
weak_alias (__strnlen, strnlen);
# ifdef SHARED
__hidden_ver1 (__strnlen, __GI___strnlen, __redirect___strnlen)
__attribute__((visibility ("hidden")));
__hidden_ver1 (strnlen, __GI_strnlen, __redirect_strnlen)
__attribute__((weak, visibility ("hidden")));
# endif
#endif

View File

@ -0,0 +1,4 @@
#define STRLEN __wcslen_avx2
#define USE_AS_WCSLEN 1
#include "strlen-avx2.S"

View File

@ -0,0 +1,26 @@
/* wcslen optimized with SSE2.
Copyright (C) 2017 Free Software Foundation, Inc.
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)
# define __wcslen __wcslen_sse2
# undef weak_alias
# define weak_alias(__wcslen, wcslen)
#endif
#include "../wcslen.S"

View File

@ -0,0 +1,31 @@
/* Multiple versions of wcslen.
All versions must be listed in ifunc-impl-list.c.
Copyright (C) 2017 Free Software Foundation, Inc.
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/>. */
/* Define multiple versions only for the definition in libc. */
#if IS_IN (libc)
# define __wcslen __redirect_wcslen
# include <wchar.h>
# undef __wcslen
# define SYMBOL_NAME wcslen
# include "ifunc-avx2.h"
libc_ifunc_redirected (__redirect_wcslen, __wcslen, IFUNC_SELECTOR ());
weak_alias (__wcslen, wcslen);
#endif

View File

@ -0,0 +1,5 @@
#define STRLEN __wcsnlen_avx2
#define USE_AS_WCSLEN 1
#define USE_AS_STRNLEN 1
#include "strlen-avx2.S"

View File

@ -28,12 +28,18 @@
extern __typeof (REDIRECT_NAME) OPTIMIZE (sse2) attribute_hidden;
extern __typeof (REDIRECT_NAME) OPTIMIZE (sse4_1) 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_CPU_P (cpu_features, SSE4_1))
return OPTIMIZE (sse4_1);