AuroraMiddleware/glibc
1
0
Fork 0
mirror of https://sourceware.org/git/glibc.git synced 2024-09-19 16:10:01 +00:00
Code Issues Packages Projects Releases Wiki Activity

x86: Unifies 'strnlen-evex' and 'strnlen-evex512' implementations.

Browse Source 
This commit uses a common implementation 'strnlen-evex-base.S' for both
'strnlen-evex' and 'strnlen-evex512'

This patch serves both to reduce the number of implementations, and it also does some small optimizations that benefit strnlen-evex and strnlen-evex512.

All tests pass on x86.

Benchmarks were taken on SKX.
https://www.intel.com/content/www/us/en/products/sku/123613/intel-core-i97900x-xseries-processor-13-75m-cache-up-to-4-30-ghz/specifications.html

Geometric mean for strnlen-evex over all benchmarks (N=10) was (new/old) 0.881
Geometric mean for strnlen-evex512 over all benchmarks (N=10) was (new/old) 0.953

Code Size Changes:
    strnlen-evex       :  +31 bytes
    strnlen-evex512    :  +156 bytes
Reviewed-by: Noah Goldstein <goldstein.w.n@gmail.com>
This commit is contained in:
Matthew Sterrett 2024-08-09 15:05:09 -07:00 committed by H.J. Lu
parent 25a5eb4010
commit 294a892769
3 changed files with 469 additions and 680 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

462
sysdeps/x86_64/multiarch/strnlen-evex-base.S Normal file
View File

@ -0,0 +1,462 @@
/* strnlen/wcsnlen optimized with 256/512-bit EVEX instructions.
Copyright (C) 2022-2024 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
<https://www.gnu.org/licenses/>. */
#include <isa-level.h>
#if ISA_SHOULD_BUILD (4)
# include <sysdep.h>
#ifdef USE_AS_WCSLEN
# define VPCMPEQ vpcmpeqd
# define VPTESTN vptestnmd
# define VPMINU vpminud
# define CHAR_SIZE 4
#else
# define VPCMPEQ vpcmpeqb
# define VPTESTN vptestnmb
# define VPMINU vpminub
# define CHAR_SIZE 1
#endif
#define XZERO VMM_128(0)
#define VZERO VMM(0)
#define PAGE_SIZE 4096
#define CHAR_PER_VEC (VEC_SIZE / CHAR_SIZE)
#if CHAR_PER_VEC == 32
# define SUB_SHORT(imm, reg) subb $(imm), %VGPR_SZ(reg, 8)
#else
# define SUB_SHORT(imm, reg) subl $(imm), %VGPR_SZ(reg, 32)
#endif
#ifdef USE_AS_WCSLEN
/* For wide-character, we care more about limitting code size
than optimally aligning targets, so just cap nop padding
reasonably low. */
# define P2ALIGN(...) .p2align 4,, 6
# define P2ALIGN_CLAMPED(...) P2ALIGN(__VA_ARGS__)
#else
# define P2ALIGN(x) .p2align x
# define P2ALIGN_CLAMPED(x, y) .p2align x,, y
#endif
.section SECTION(.text), "ax", @progbits
/* Aligning entry point to 64 byte, provides better performance for
one vector length string. */
ENTRY_P2ALIGN(STRNLEN, 6)
/* rdi is pointer to array, rsi is the upper limit. */
/* Check zero length. */
test %RSI_LP, %RSI_LP
jz L(zero)
#ifdef __ILP32__
/* Clear the upper 32 bits. */
movl %esi, %esi
#endif
vpxorq %XZERO, %XZERO, %XZERO
/* Check that we won't cross a page boundary with our first load. */
movl %edi, %eax
shll $20, %eax
cmpl $((PAGE_SIZE - VEC_SIZE) << 20), %eax
ja L(crosses_page_boundary)
/* Check the first VEC_SIZE bytes. Each bit in K0 represents a
null byte. */
VPCMPEQ (%rdi), %VZERO, %k0
KMOV %k0, %VRCX
/* If src (rcx) is zero, bsf does not change the result. NB:
Must use 64-bit bsf here so that upper bits of len are not
cleared. */
movq %rsi, %rax
bsfq %rcx, %rax
/* If rax > CHAR_PER_VEC then rcx must have been zero (no null
CHAR) and rsi must be > CHAR_PER_VEC. */
cmpq $CHAR_PER_VEC, %rax
ja L(more_1x_vec)
/* Check if first match in bounds. */
cmpq %rax, %rsi
cmovb %esi, %eax
ret
#if VEC_SIZE == 32
P2ALIGN_CLAMPED(4, 2)
L(zero):
L(max_0):
movl %esi, %eax
ret
#endif
P2ALIGN_CLAMPED(4, 10)
L(more_1x_vec):
L(cross_page_continue):
/* After this calculation, rax stores the number of elements
left to be processed The complexity comes from the fact some
elements get read twice due to alignment and we need to be
sure we don't count them twice (else, it would just be rsi -
CHAR_PER_VEC). */
#ifdef USE_AS_WCSLEN
/* Need to compute directly for wcslen as CHAR_SIZE * rsi can
overflow. */
movq %rdi, %rax
andq $(VEC_SIZE * -1), %rdi
subq %rdi, %rax
sarq $2, %rax
leaq -(CHAR_PER_VEC * 1)(%rax, %rsi), %rax
#else
/* Calculate ptr + N - VEC_SIZE, then mask off the low bits,
then subtract ptr to get the new aligned limit value. */
leaq (VEC_SIZE * -1)(%rsi, %rdi), %rax
andq $(VEC_SIZE * -1), %rdi
subq %rdi, %rax
#endif
VPCMPEQ VEC_SIZE(%rdi), %VZERO, %k0
/* Checking here is faster for 256-bit but not 512-bit */
#if VEC_SIZE == 0
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(last_vec_check)
#endif
cmpq $(CHAR_PER_VEC * 2), %rax
ja L(more_2x_vec)
L(last_2x_vec_or_less):
/* Checking here is faster for 512-bit but not 256-bit */
#if VEC_SIZE != 0
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(last_vec_check)
#endif
/* Check for the end of data. */
SUB_SHORT (CHAR_PER_VEC, rax)
jbe L(max_0)
/* Check the final remaining vector. */
VPCMPEQ (VEC_SIZE * 2)(%rdi), %VZERO, %k0
KMOV %k0, %VRDX
test %VRDX, %VRDX
#if VEC_SIZE == 32
jz L(max_0)
#else
jnz L(last_vec_check)
P2ALIGN_CLAMPED(4, 2)
L(zero):
L(max_0):
movl %esi, %eax
ret
#endif
P2ALIGN_CLAMPED(4, 4)
L(last_vec_check):
bsf %VRDX, %VRDX
sub %eax, %edx
lea (%rsi, %rdx), %eax
cmovae %esi, %eax
ret
#if VEC_SIZE == 32
P2ALIGN_CLAMPED(4, 8)
#endif
L(last_4x_vec_or_less):
addl $(CHAR_PER_VEC * -4), %eax
VPCMPEQ (VEC_SIZE * 5)(%rdi), %VZERO, %k0
#if VEC_SIZE == 64
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(last_vec_check)
#endif
subq $(VEC_SIZE * -4), %rdi
cmpl $(CHAR_PER_VEC * 2), %eax
jbe L(last_2x_vec_or_less)
P2ALIGN_CLAMPED(4, 6)
L(more_2x_vec):
/* Remaining length >= 2 * CHAR_PER_VEC so do VEC0/VEC1 without
rechecking bounds. */
/* Already checked in 256-bit case */
#if VEC_SIZE != 0
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(first_vec_x1)
#endif
VPCMPEQ (VEC_SIZE * 2)(%rdi), %VZERO, %k0
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(first_vec_x2)
cmpq $(CHAR_PER_VEC * 4), %rax
ja L(more_4x_vec)
VPCMPEQ (VEC_SIZE * 3)(%rdi), %VZERO, %k0
KMOV %k0, %VRDX
addl $(CHAR_PER_VEC * -2), %eax
test %VRDX, %VRDX
jnz L(last_vec_check)
subb $(CHAR_PER_VEC), %al
jbe L(max_1)
VPCMPEQ (VEC_SIZE * 4)(%rdi), %VZERO, %k0
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(last_vec_check)
L(max_1):
movl %esi, %eax
ret
P2ALIGN_CLAMPED(4, 14)
L(first_vec_x2):
#if VEC_SIZE == 64
/* If VEC_SIZE == 64 we can fit logic for full return label in
spare bytes before next cache line. */
bsf %VRDX, %VRDX
sub %eax, %esi
leal (CHAR_PER_VEC * 1)(%rsi, %rdx), %eax
ret
P2ALIGN_CLAMPED(4, 6)
#else
addl $CHAR_PER_VEC, %esi
#endif
L(first_vec_x1):
bsf %VRDX, %VRDX
sub %eax, %esi
leal (CHAR_PER_VEC * 0)(%rsi, %rdx), %eax
ret
#if VEC_SIZE == 64
P2ALIGN_CLAMPED(4, 6)
L(first_vec_x4):
# if VEC_SIZE == 64
/* If VEC_SIZE == 64 we can fit logic for full return label in
spare bytes before next cache line. */
bsf %VRDX, %VRDX
sub %eax, %esi
leal (CHAR_PER_VEC * 3)(%rsi, %rdx), %eax
ret
P2ALIGN_CLAMPED(4, 6)
# else
addl $CHAR_PER_VEC, %esi
# endif
L(first_vec_x3):
bsf %VRDX, %VRDX
sub %eax, %esi
leal (CHAR_PER_VEC * 2)(%rsi, %rdx), %eax
ret
#endif
P2ALIGN_CLAMPED(6, 20)
L(more_4x_vec):
VPCMPEQ (VEC_SIZE * 3)(%rdi), %VZERO, %k0
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(first_vec_x3)
VPCMPEQ (VEC_SIZE * 4)(%rdi), %VZERO, %k0
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(first_vec_x4)
/* Check if at last VEC_SIZE * 4 length before aligning for the
loop. */
cmpq $(CHAR_PER_VEC * 8), %rax
jbe L(last_4x_vec_or_less)
/* Compute number of words checked after aligning. */
#ifdef USE_AS_WCSLEN
/* Need to compute directly for wcslen as CHAR_SIZE * rsi can
overflow. */
leaq (VEC_SIZE * -3)(%rdi), %rdx
#else
leaq (VEC_SIZE * -3)(%rdi, %rax), %rax
#endif
subq $(VEC_SIZE * -1), %rdi
/* Align data to VEC_SIZE * 4. */
#if VEC_SIZE == 64
/* Saves code size. No evex512 processor has partial register
stalls. If that change this can be replaced with `andq
$-(VEC_SIZE * 4), %rdi`. */
xorb %dil, %dil
#else
andq $-(VEC_SIZE * 4), %rdi
#endif
#ifdef USE_AS_WCSLEN
subq %rdi, %rdx
sarq $2, %rdx
addq %rdx, %rax
#else
subq %rdi, %rax
#endif
// mov %rdi, %rdx
P2ALIGN(6)
L(loop):
/* VPMINU and VPCMP combination provide better performance as
compared to alternative combinations. */
VMOVA (VEC_SIZE * 4)(%rdi), %VMM(1)
VPMINU (VEC_SIZE * 5)(%rdi), %VMM(1), %VMM(2)
VMOVA (VEC_SIZE * 6)(%rdi), %VMM(3)
VPMINU (VEC_SIZE * 7)(%rdi), %VMM(3), %VMM(4)
VPTESTN %VMM(2), %VMM(2), %k0
VPTESTN %VMM(4), %VMM(4), %k1
subq $-(VEC_SIZE * 4), %rdi
KORTEST %k0, %k1
jnz L(loopend)
subq $(CHAR_PER_VEC * 4), %rax
ja L(loop)
mov %rsi, %rax
ret
#if VEC_SIZE == 32
P2ALIGN_CLAMPED(4, 6)
L(first_vec_x4):
# if VEC_SIZE == 64
/* If VEC_SIZE == 64 we can fit logic for full return label in
spare bytes before next cache line. */
bsf %VRDX, %VRDX
sub %eax, %esi
leal (CHAR_PER_VEC * 3)(%rsi, %rdx), %eax
ret
P2ALIGN_CLAMPED(4, 6)
# else
addl $CHAR_PER_VEC, %esi
# endif
L(first_vec_x3):
bsf %VRDX, %VRDX
sub %eax, %esi
leal (CHAR_PER_VEC * 2)(%rsi, %rdx), %eax
ret
#endif
P2ALIGN_CLAMPED(4, 11)
L(loopend):
/* We found a null terminator in one of the 4 vectors. */
/* Check the first vector. */
movq %rax, %r8
VPTESTN %VMM(1), %VMM(1), %k2
KMOV %k2, %VRCX
bsf %rcx, %r8
cmpq $(CHAR_PER_VEC), %r8
jbe L(end_vec)
/* Check the second vector. */
subq $(CHAR_PER_VEC), %rax
movq %rax, %r8
KMOV %k0, %VRCX
bsf %rcx, %r8
cmpq $(CHAR_PER_VEC), %r8
jbe L(end_vec)
/* Check the third vector. */
subq $(CHAR_PER_VEC), %rax
movq %rax, %r8
VPTESTN %VMM(3), %VMM(3), %k2
KMOV %k2, %VRCX
bsf %rcx, %r8
cmpq $(CHAR_PER_VEC), %r8
jbe L(end_vec)
/* It is in the fourth vector. */
subq $(CHAR_PER_VEC), %rax
movq %rax, %r8
KMOV %k1, %VRCX
bsf %rcx, %r8
P2ALIGN_CLAMPED(4, 3)
L(end_vec):
/* Get the number that has been processed. */
movq %rsi, %rcx
subq %rax, %rcx
/* Add that to the offset we found the null terminator at. */
leaq (%r8, %rcx), %rax
/* Take the min of that and the limit. */
cmpq %rsi, %rax
cmovnb %rsi, %rax
ret
P2ALIGN_CLAMPED(4, 11)
L(crosses_page_boundary):
/* Align data backwards to VEC_SIZE. */
shrl $20, %eax
movq %rdi, %rcx
andq $-VEC_SIZE, %rcx
VPCMPEQ (%rcx), %VZERO, %k0
KMOV %k0, %VRCX
#ifdef USE_AS_WCSLEN
shrl $2, %eax
andl $(CHAR_PER_VEC - 1), %eax
#endif
/* By this point rax contains number of bytes we need to skip. */
shrx %VRAX, %VRCX, %VRCX
/* Calculates CHAR_PER_VEC - eax and stores in eax. */
negl %eax
andl $(CHAR_PER_VEC - 1), %eax
movq %rsi, %rdx
bsf %VRCX, %VRDX
cmpq %rax, %rdx
ja L(cross_page_continue)
/* The vector had a null terminator or we are at the limit. */
movl %edx, %eax
cmpq %rdx, %rsi
cmovb %esi, %eax
ret
END(STRNLEN)
#endif

428
sysdeps/x86_64/multiarch/strnlen-evex.S
View File

@ -1,423 +1,7 @@
/* strnlen/wcsnlen optimized with 256-bit EVEX instructions.
Copyright (C) 2022-2024 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
<https://www.gnu.org/licenses/>. */
#include <isa-level.h>
#include <sysdep.h>
#if ISA_SHOULD_BUILD (4)
# ifndef VEC_SIZE
# include "x86-evex256-vecs.h"
# endif
# ifndef STRNLEN
# define STRNLEN __strnlen_evex
# endif
# ifdef USE_AS_WCSLEN
# define VPCMPEQ vpcmpeqd
# define VPCMPNEQ vpcmpneqd
# define VPTESTN vptestnmd
# define VPTEST vptestmd
# define VPMINU vpminud
# define CHAR_SIZE 4
# else
# define VPCMPEQ vpcmpeqb
# define VPCMPNEQ vpcmpneqb
# define VPTESTN vptestnmb
# define VPTEST vptestmb
# define VPMINU vpminub
# define CHAR_SIZE 1
# define REG_WIDTH VEC_SIZE
# endif
# define CHAR_PER_VEC (VEC_SIZE / CHAR_SIZE)
# include "reg-macros.h"
# if CHAR_PER_VEC == 32
# define SUB_SHORT(imm, reg) subb $(imm), %VGPR_SZ(reg, 8)
# else
# define SUB_SHORT(imm, reg) subl $(imm), %VGPR_SZ(reg, 32)
# endif
# if CHAR_PER_VEC == 64
# define FALLTHROUGH_RETURN_OFFSET (CHAR_PER_VEC * 3)
# else
# define FALLTHROUGH_RETURN_OFFSET (CHAR_PER_VEC * 2)
# endif
# define XZERO VMM_128(0)
# define VZERO VMM(0)
# define PAGE_SIZE 4096
.section SECTION(.text), "ax", @progbits
ENTRY_P2ALIGN (STRNLEN, 6)
/* Check zero length. */
test %RSI_LP, %RSI_LP
jz L(zero)
# ifdef __ILP32__
/* Clear the upper 32 bits. */
movl %esi, %esi
# endif
movl %edi, %eax
vpxorq %XZERO, %XZERO, %XZERO
andl $(PAGE_SIZE - 1), %eax
cmpl $(PAGE_SIZE - VEC_SIZE), %eax
ja L(cross_page_boundary)
/* Check the first VEC_SIZE bytes. Each bit in K0 represents a
null byte. */
VPCMPEQ (%rdi), %VZERO, %k0
KMOV %k0, %VRCX
movq %rsi, %rax
/* If src (rcx) is zero, bsf does not change the result. NB:
Must use 64-bit bsf here so that upper bits of len are not
cleared. */
bsfq %rcx, %rax
/* If rax > CHAR_PER_VEC then rcx must have been zero (no null
CHAR) and rsi must be > CHAR_PER_VEC. */
cmpq $CHAR_PER_VEC, %rax
ja L(more_1x_vec)
/* Check if first match in bounds. */
cmpq %rax, %rsi
cmovb %esi, %eax
ret
# if CHAR_PER_VEC != 32
.p2align 4,, 2
L(zero):
L(max_0):
movl %esi, %eax
ret
# endif
/* Aligned more for strnlen compares remaining length vs 2 *
CHAR_PER_VEC, 4 * CHAR_PER_VEC, and 8 * CHAR_PER_VEC before
going to the loop. */
.p2align 4,, 10
L(more_1x_vec):
L(cross_page_continue):
/* Compute number of words checked after aligning. */
# ifdef USE_AS_WCSLEN
/* Need to compute directly for wcslen as CHAR_SIZE * rsi can
overflow. */
movq %rdi, %rax
andq $(VEC_SIZE * -1), %rdi
subq %rdi, %rax
sarq $2, %rax
leaq -(CHAR_PER_VEC * 1)(%rax, %rsi), %rax
# else
leaq (VEC_SIZE * -1)(%rsi, %rdi), %rax
andq $(VEC_SIZE * -1), %rdi
subq %rdi, %rax
# endif
VPCMPEQ VEC_SIZE(%rdi), %VZERO, %k0
cmpq $(CHAR_PER_VEC * 2), %rax
ja L(more_2x_vec)
L(last_2x_vec_or_less):
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(last_vec_check)
/* Check the end of data. */
SUB_SHORT (CHAR_PER_VEC, rax)
jbe L(max_0)
VPCMPEQ (VEC_SIZE * 2)(%rdi), %VZERO, %k0
KMOV %k0, %VRDX
test %VRDX, %VRDX
jz L(max_0)
/* Best place for LAST_VEC_CHECK if ZMM. */
.p2align 4,, 8
L(last_vec_check):
bsf %VRDX, %VRDX
sub %eax, %edx
lea (%rsi, %rdx), %eax
cmovae %esi, %eax
ret
# if CHAR_PER_VEC == 32
.p2align 4,, 2
L(zero):
L(max_0):
movl %esi, %eax
ret
# endif
.p2align 4,, 8
L(last_4x_vec_or_less):
addl $(CHAR_PER_VEC * -4), %eax
VPCMPEQ (VEC_SIZE * 5)(%rdi), %VZERO, %k0
subq $(VEC_SIZE * -4), %rdi
cmpl $(CHAR_PER_VEC * 2), %eax
jbe L(last_2x_vec_or_less)
.p2align 4,, 6
L(more_2x_vec):
/* Remaining length >= 2 * CHAR_PER_VEC so do VEC0/VEC1 without
rechecking bounds. */
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(first_vec_x1)
VPCMPEQ (VEC_SIZE * 2)(%rdi), %VZERO, %k0
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(first_vec_x2)
cmpq $(CHAR_PER_VEC * 4), %rax
ja L(more_4x_vec)
VPCMPEQ (VEC_SIZE * 3)(%rdi), %VZERO, %k0
KMOV %k0, %VRDX
addl $(CHAR_PER_VEC * -2), %eax
test %VRDX, %VRDX
jnz L(last_vec_check)
subl $(CHAR_PER_VEC), %eax
jbe L(max_1)
VPCMPEQ (VEC_SIZE * 4)(%rdi), %VZERO, %k0
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(last_vec_check)
L(max_1):
movl %esi, %eax
ret
.p2align 4,, 3
L(first_vec_x2):
# if VEC_SIZE == 64
/* If VEC_SIZE == 64 we can fit logic for full return label in
spare bytes before next cache line. */
bsf %VRDX, %VRDX
sub %eax, %esi
leal (CHAR_PER_VEC * 1)(%rsi, %rdx), %eax
ret
.p2align 4,, 6
# else
addl $CHAR_PER_VEC, %esi
# endif
L(first_vec_x1):
bsf %VRDX, %VRDX
sub %eax, %esi
leal (CHAR_PER_VEC * 0)(%rsi, %rdx), %eax
ret
.p2align 4,, 6
L(first_vec_x4):
# if VEC_SIZE == 64
/* If VEC_SIZE == 64 we can fit logic for full return label in
spare bytes before next cache line. */
bsf %VRDX, %VRDX
sub %eax, %esi
leal (CHAR_PER_VEC * 3)(%rsi, %rdx), %eax
ret
.p2align 4,, 6
# else
addl $CHAR_PER_VEC, %esi
# endif
L(first_vec_x3):
bsf %VRDX, %VRDX
sub %eax, %esi
leal (CHAR_PER_VEC * 2)(%rsi, %rdx), %eax
ret
.p2align 4,, 5
L(more_4x_vec):
VPCMPEQ (VEC_SIZE * 3)(%rdi), %VZERO, %k0
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(first_vec_x3)
VPCMPEQ (VEC_SIZE * 4)(%rdi), %VZERO, %k0
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(first_vec_x4)
/* Check if at last VEC_SIZE * 4 length before aligning for the
loop. */
cmpq $(CHAR_PER_VEC * 8), %rax
jbe L(last_4x_vec_or_less)
/* Compute number of words checked after aligning. */
# ifdef USE_AS_WCSLEN
/* Need to compute directly for wcslen as CHAR_SIZE * rsi can
overflow. */
leaq (VEC_SIZE * -3)(%rdi), %rdx
# else
leaq (VEC_SIZE * -3)(%rdi, %rax), %rax
# endif
subq $(VEC_SIZE * -1), %rdi
/* Align data to VEC_SIZE * 4. */
# if VEC_SIZE == 64
/* Saves code size. No evex512 processor has partial register
stalls. If that change this can be replaced with `andq
$-(VEC_SIZE * 4), %rdi`. */
xorb %dil, %dil
# else
andq $-(VEC_SIZE * 4), %rdi
# endif
# ifdef USE_AS_WCSLEN
subq %rdi, %rdx
sarq $2, %rdx
addq %rdx, %rax
# else
subq %rdi, %rax
# endif
/* Compare 4 * VEC at a time forward. */
.p2align 4,, 11
L(loop_4x_vec):
VMOVA (VEC_SIZE * 4)(%rdi), %VMM(1)
VPMINU (VEC_SIZE * 5)(%rdi), %VMM(1), %VMM(2)
VMOVA (VEC_SIZE * 6)(%rdi), %VMM(3)
VPMINU (VEC_SIZE * 7)(%rdi), %VMM(3), %VMM(4)
VPTESTN %VMM(2), %VMM(2), %k0
VPTESTN %VMM(4), %VMM(4), %k2
subq $-(VEC_SIZE * 4), %rdi
/* Break if at end of length. */
subq $(CHAR_PER_VEC * 4), %rax
jbe L(loop_len_end)
KORTEST %k0, %k2
jz L(loop_4x_vec)
L(loop_last_4x_vec):
movq %rsi, %rcx
subq %rax, %rsi
VPTESTN %VMM(1), %VMM(1), %k1
KMOV %k1, %VRDX
test %VRDX, %VRDX
jnz L(last_vec_x0)
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(last_vec_x1)
VPTESTN %VMM(3), %VMM(3), %k0
/* Separate logic for VEC_SIZE == 64 and VEC_SIZE == 32 for
returning last 2x VEC. For VEC_SIZE == 64 we test each VEC
individually, for VEC_SIZE == 32 we combine them in a single
64-bit GPR. */
# if CHAR_PER_VEC == 64
KMOV %k0, %VRDX
test %VRDX, %VRDX
jnz L(last_vec_x2)
KMOV %k2, %VRDX
# else
/* We can only combine last 2x VEC masks if CHAR_PER_VEC <= 32.
*/
kmovd %k2, %edx
kmovd %k0, %eax
salq $CHAR_PER_VEC, %rdx
orq %rax, %rdx
# endif
/* first_vec_x3 for strlen-ZMM and first_vec_x2 for strlen-YMM.
*/
bsfq %rdx, %rdx
leaq (FALLTHROUGH_RETURN_OFFSET - CHAR_PER_VEC * 4)(%rsi, %rdx), %rax
cmpq %rax, %rcx
cmovb %rcx, %rax
ret
/* Handle last 4x VEC after loop. All VECs have been loaded. */
.p2align 4,, 4
L(loop_len_end):
KORTEST %k0, %k2
jnz L(loop_last_4x_vec)
movq %rsi, %rax
ret
# if CHAR_PER_VEC == 64
/* Since we can't combine the last 2x VEC for VEC_SIZE == 64
need return label for it. */
.p2align 4,, 8
L(last_vec_x2):
bsf %VRDX, %VRDX
leaq (CHAR_PER_VEC * -2)(%rsi, %rdx), %rax
cmpq %rax, %rcx
cmovb %rcx, %rax
ret
# endif
.p2align 4,, 10
L(last_vec_x1):
addq $CHAR_PER_VEC, %rsi
L(last_vec_x0):
bsf %VRDX, %VRDX
leaq (CHAR_PER_VEC * -4)(%rsi, %rdx), %rax
cmpq %rax, %rcx
cmovb %rcx, %rax
ret
.p2align 4,, 8
L(cross_page_boundary):
/* Align data to VEC_SIZE. */
movq %rdi, %rcx
andq $-VEC_SIZE, %rcx
VPCMPEQ (%rcx), %VZERO, %k0
KMOV %k0, %VRCX
# ifdef USE_AS_WCSLEN
shrl $2, %eax
andl $(CHAR_PER_VEC - 1), %eax
# endif
shrx %VRAX, %VRCX, %VRCX
negl %eax
andl $(CHAR_PER_VEC - 1), %eax
movq %rsi, %rdx
bsf %VRCX, %VRDX
cmpq %rax, %rdx
ja L(cross_page_continue)
movl %edx, %eax
cmpq %rdx, %rsi
cmovb %esi, %eax
ret
END (STRNLEN)
#ifndef STRNLEN
#define STRNLEN __strnlen_evex
#endif
#include "x86-evex256-vecs.h"
#include "reg-macros.h"
#include "strnlen-evex-base.S"

259
sysdeps/x86_64/multiarch/strnlen-evex512.S
View File

@ -1,264 +1,7 @@
/* Placeholder function, not used by any processor at the moment.
Copyright (C) 2022-2024 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
<https://www.gnu.org/licenses/>. */
#ifndef STRNLEN
#define STRNLEN __strnlen_evex512
#endif
#include "x86-evex512-vecs.h"
#include "reg-macros.h"
#include <isa-level.h>
#if ISA_SHOULD_BUILD (4)
# include <sysdep.h>
# ifdef USE_AS_WCSLEN
# define VPCMPEQ vpcmpeqd
# define VPTESTN vptestnmd
# define VPMINU vpminud
# define CHAR_SIZE 4
# else
# define VPCMPEQ vpcmpeqb
# define VPTESTN vptestnmb
# define VPMINU vpminub
# define CHAR_SIZE 1
# endif
# define PAGE_SIZE 4096
# define CHAR_PER_VEC (VEC_SIZE / CHAR_SIZE)
.section SECTION(.text),"ax",@progbits
/* Aligning entry point to 64 byte, provides better performance for
one vector length string. */
ENTRY_P2ALIGN (STRNLEN, 6)
/* Check zero length. */
test %RSI_LP, %RSI_LP
jz L(ret_max)
# ifdef __ILP32__
/* Clear the upper 32 bits. */
movl %esi, %esi
# endif
movl %edi, %eax
vpxorq %VMM_128(0), %VMM_128(0), %VMM_128(0)
sall $20, %eax
cmpl $((PAGE_SIZE - VEC_SIZE) << 20), %eax
ja L(page_cross)
/* Compare [w]char for null, mask bit will be set for match. */
VPCMPEQ (%rdi), %VMM(0), %k0
KMOV %k0, %VRCX
/* Store max length in rax. */
mov %rsi, %rax
/* If rcx is 0, rax will have max length. We can not use VRCX
and VRAX here for evex256 because, upper 32 bits may be
undefined for ecx and eax. */
bsfq %rcx, %rax
cmp $CHAR_PER_VEC, %rax
ja L(align_more)
cmpq %rax, %rsi
cmovb %esi, %eax
ret
/* At this point vector max length reached. */
.p2align 4,,3
L(ret_max):
movq %rsi, %rax
ret
L(align_more):
mov %rdi, %rax
/* Align rax to VEC_SIZE. */
andq $-VEC_SIZE, %rax
movq %rdi, %rdx
subq %rax, %rdx
# ifdef USE_AS_WCSLEN
shr $2, %VRDX
# endif
/* At this point rdx contains [w]chars already compared. */
leaq -CHAR_PER_VEC(%rsi, %rdx), %rdx
/* At this point rdx contains number of w[char] needs to go.
Now onwards rdx will keep decrementing with each compare. */
/* Loop unroll 4 times for 4 vector loop. */
VPCMPEQ VEC_SIZE(%rax), %VMM(0), %k0
subq $-VEC_SIZE, %rax
KMOV %k0, %VRCX
test %VRCX, %VRCX
jnz L(ret_vec_x1)
subq $CHAR_PER_VEC, %rdx
jbe L(ret_max)
VPCMPEQ VEC_SIZE(%rax), %VMM(0), %k0
KMOV %k0, %VRCX
test %VRCX, %VRCX
jnz L(ret_vec_x2)
subq $CHAR_PER_VEC, %rdx
jbe L(ret_max)
VPCMPEQ (VEC_SIZE * 2)(%rax), %VMM(0), %k0
KMOV %k0, %VRCX
test %VRCX, %VRCX
jnz L(ret_vec_x3)
subq $CHAR_PER_VEC, %rdx
jbe L(ret_max)
VPCMPEQ (VEC_SIZE * 3)(%rax), %VMM(0), %k0
KMOV %k0, %VRCX
test %VRCX, %VRCX
jnz L(ret_vec_x4)
subq $CHAR_PER_VEC, %rdx
jbe L(ret_max)
/* Save pointer before 4 x VEC_SIZE alignment. */
movq %rax, %rcx
/* Align address to VEC_SIZE * 4 for loop. */
andq $-(VEC_SIZE * 4), %rax
subq %rax, %rcx
# ifdef USE_AS_WCSLEN
shr $2, %VRCX
# endif
/* rcx contains number of [w]char will be recompared due to
alignment fixes. rdx must be incremented by rcx to offset
alignment adjustment. */
addq %rcx, %rdx
/* Need jump as we don't want to add/subtract rdx for first
iteration of 4 x VEC_SIZE aligned loop. */
.p2align 4,,11
L(loop):
/* VPMINU and VPCMP combination provide better performance as
compared to alternative combinations. */
VMOVA (VEC_SIZE * 4)(%rax), %VMM(1)
VPMINU (VEC_SIZE * 5)(%rax), %VMM(1), %VMM(2)
VMOVA (VEC_SIZE * 6)(%rax), %VMM(3)
VPMINU (VEC_SIZE * 7)(%rax), %VMM(3), %VMM(4)
VPTESTN %VMM(2), %VMM(2), %k0
VPTESTN %VMM(4), %VMM(4), %k1
subq $-(VEC_SIZE * 4), %rax
KORTEST %k0, %k1
jnz L(loopend)
subq $(CHAR_PER_VEC * 4), %rdx
ja L(loop)
mov %rsi, %rax
ret
L(loopend):
VPTESTN %VMM(1), %VMM(1), %k2
KMOV %k2, %VRCX
test %VRCX, %VRCX
jnz L(ret_vec_x1)
KMOV %k0, %VRCX
/* At this point, if k0 is non zero, null char must be in the
second vector. */
test %VRCX, %VRCX
jnz L(ret_vec_x2)
VPTESTN %VMM(3), %VMM(3), %k3
KMOV %k3, %VRCX
test %VRCX, %VRCX
jnz L(ret_vec_x3)
/* At this point null [w]char must be in the fourth vector so no
need to check. */
KMOV %k1, %VRCX
/* Fourth, third, second vector terminating are pretty much
same, implemented this way to avoid branching and reuse code
from pre loop exit condition. */
L(ret_vec_x4):
bsf %VRCX, %VRCX
subq %rdi, %rax
# ifdef USE_AS_WCSLEN
subq $-(VEC_SIZE * 3), %rax
shrq $2, %rax
addq %rcx, %rax
# else
leaq (VEC_SIZE * 3)(%rcx, %rax), %rax
# endif
cmpq %rsi, %rax
cmovnb %rsi, %rax
ret
L(ret_vec_x3):
bsf %VRCX, %VRCX
subq %rdi, %rax
# ifdef USE_AS_WCSLEN
subq $-(VEC_SIZE * 2), %rax
shrq $2, %rax
addq %rcx, %rax
# else
leaq (VEC_SIZE * 2)(%rcx, %rax), %rax
# endif
cmpq %rsi, %rax
cmovnb %rsi, %rax
ret
L(ret_vec_x2):
subq $-VEC_SIZE, %rax
L(ret_vec_x1):
bsf %VRCX, %VRCX
subq %rdi, %rax
# ifdef USE_AS_WCSLEN
shrq $2, %rax
# endif
addq %rcx, %rax
cmpq %rsi, %rax
cmovnb %rsi, %rax
ret
L(page_cross):
mov %rdi, %rax
movl %edi, %ecx
andl $(VEC_SIZE - 1), %ecx
# ifdef USE_AS_WCSLEN
sarl $2, %ecx
# endif
/* ecx contains number of w[char] to be skipped as a result
of address alignment. */
andq $-VEC_SIZE, %rax
VPCMPEQ (%rax), %VMM(0), %k0
KMOV %k0, %VRDX
/* Ignore number of character for alignment adjustment. */
shr %cl, %VRDX
jnz L(page_cross_end)
movl $CHAR_PER_VEC, %eax
sub %ecx, %eax
cmp %rax, %rsi
ja L(align_more)
L(page_cross_end):
bsf %VRDX, %VRAX
cmpq %rsi, %rax
cmovnb %esi, %eax
ret
END (STRNLEN)
#endif
#include "strnlen-evex-base.S"