mirror of
https://github.com/apache/nuttx.git
synced 2026-10-07 14:25:25 +00:00
0790b52e64
memcmp, strncmp and strcmp reach their word loops only when both pointers
are already on a register boundary:
or t0, a0, a1
andi t0, t0, SZREG-1
That asks more than the loops need. They load from the two pointers at
the same boundary, so what matters is that the two agree about where a
boundary falls, not that either is already on one. A pair offset by the
same amount can be walked up to the boundary a byte at a time and
compared a register at a time from there.
The union also holds far less often than the difference. For arbitrary
pointers on RV64 it is true about one time in 64 against one in eight,
and the case it rejects, two strings carved out of the same buffer, is
the common one.
Test the difference of the pointers, and walk to the boundary first.
arch_strcpy.S and arch_memcpy.S already do this. Keeping every access
aligned is not only faster here: the base ISA does not require misaligned
loads and stores to be supported at all, so a routine in a machine
directory cannot assume one will work, whatever it costs.
Measured on a 1.4 GHz rv64, source and destination misaligned by one:
before after
memcmp 32K 34.4 458.0 MB/s
strncmp 32K 32.4 253.0 MB/s
strcmp 32K 41.0 280.0 MB/s
Each of those was the rate of the byte loop the word loop was meant to
replace. Pointers that genuinely disagree still take the byte loop, and
the aligned rates are unchanged.
The measurements come from the benchmark in apache/nuttx-apps#3706.
Assisted-by: Claude:claude-opus-5
Signed-off-by: Justin Hammond <justin@dynam.ac>
139 lines
3.1 KiB
ArmAsm
139 lines
3.1 KiB
ArmAsm
/****************************************************************************
|
|
* libs/libc/machine/risc-v/arch_strncmp.S
|
|
*
|
|
* SPDX-License-Identifier: Apache-2.0
|
|
*
|
|
* Licensed to the Apache Software Foundation (ASF) under one or more
|
|
* contributor license agreements. See the NOTICE file distributed with
|
|
* this work for additional information regarding copyright ownership. The
|
|
* ASF licenses this file to you under the Apache License, Version 2.0 (the
|
|
* "License"); you may not use this file except in compliance with the
|
|
* License. You may obtain a copy of the License at
|
|
*
|
|
* http://www.apache.org/licenses/LICENSE-2.0
|
|
*
|
|
* Unless required by applicable law or agreed to in writing, software
|
|
* distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
|
|
* WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
|
|
* License for the specific language governing permissions and limitations
|
|
* under the License.
|
|
*
|
|
****************************************************************************/
|
|
|
|
#include "libc.h"
|
|
|
|
#ifdef LIBC_BUILD_STRNCMP
|
|
|
|
#include "asm.h"
|
|
|
|
.text
|
|
.global ARCH_LIBCFUN(strncmp)
|
|
.type ARCH_LIBCFUN(strncmp), @function
|
|
.align 2
|
|
|
|
/************************************************************************************
|
|
* Name: strncmp
|
|
*
|
|
* int strncmp(const char *s1, const char *s2, size_t n)
|
|
*
|
|
* Word-at-a-time comparison with count limit.
|
|
************************************************************************************/
|
|
ARCH_LIBCFUN(strncmp):
|
|
.cfi_sections .debug_frame
|
|
.cfi_startproc
|
|
|
|
beqz a2, .Lequal
|
|
|
|
/* The word loop below loads from both pointers at a boundary, which
|
|
* asks that the two agree about where a boundary falls, not that
|
|
* either is already on one. Test the difference of the pointers,
|
|
* not their union.
|
|
*/
|
|
|
|
xor t0, a0, a1
|
|
andi t0, t0, SZREG-1
|
|
bnez t0, .Lbyte_loop
|
|
|
|
/* Same offset: walk both up to the boundary a byte at a time,
|
|
* stopping early on a difference or a terminator.
|
|
*/
|
|
|
|
.Lsnc_align_head:
|
|
andi t0, a0, SZREG-1
|
|
beqz t0, .Lsnc_aligned
|
|
lbu t0, 0(a0)
|
|
lbu t1, 0(a1)
|
|
bne t0, t1, .Ldiff
|
|
beqz t0, .Lequal
|
|
addi a0, a0, 1
|
|
addi a1, a1, 1
|
|
addi a2, a2, -1
|
|
beqz a2, .Lequal
|
|
j .Lsnc_align_head
|
|
|
|
.Lsnc_aligned:
|
|
|
|
/* Load masks */
|
|
|
|
#if SZREG == 8
|
|
lla t2, .Lsnc_mask01
|
|
ld t2, 0(t2)
|
|
lla t3, .Lsnc_mask80
|
|
ld t3, 0(t3)
|
|
#else
|
|
li t2, 0x01010101
|
|
li t3, 0x80808080
|
|
#endif
|
|
|
|
.Lword_loop:
|
|
li t0, SZREG
|
|
bltu a2, t0, .Lbyte_loop
|
|
REG_L t0, 0(a0)
|
|
REG_L t1, 0(a1)
|
|
|
|
bne t0, t1, .Lbyte_loop /* words differ */
|
|
|
|
/* DETECTNULL on t0 */
|
|
|
|
sub t4, t0, t2
|
|
not t5, t0
|
|
and t4, t4, t5
|
|
and t4, t4, t3
|
|
bnez t4, .Lequal /* null found, words equal → return 0 */
|
|
|
|
addi a0, a0, SZREG
|
|
addi a1, a1, SZREG
|
|
addi a2, a2, -SZREG
|
|
j .Lword_loop
|
|
|
|
.Lbyte_loop:
|
|
beqz a2, .Lequal
|
|
lbu t0, 0(a0)
|
|
lbu t1, 0(a1)
|
|
bne t0, t1, .Ldiff
|
|
beqz t0, .Lequal
|
|
addi a0, a0, 1
|
|
addi a1, a1, 1
|
|
addi a2, a2, -1
|
|
j .Lbyte_loop
|
|
|
|
.Lequal:
|
|
li a0, 0
|
|
ret
|
|
.Ldiff:
|
|
sub a0, t0, t1
|
|
ret
|
|
|
|
.cfi_endproc
|
|
.size ARCH_LIBCFUN(strncmp), .-ARCH_LIBCFUN(strncmp)
|
|
|
|
#if SZREG == 8
|
|
.section .srodata.cst8,"aM",@progbits,8
|
|
.align 3
|
|
.Lsnc_mask01:
|
|
.dword 0x0101010101010101
|
|
.Lsnc_mask80:
|
|
.dword 0x8080808080808080
|
|
#endif
|
|
|
|
#endif
|