nuttx/libs/libc/machine/xtensa/arch_strlen.S
zhuyanlin cfcff5f570 libc:machine:xtensa:add xtensa libc implement
N/A

Signed-off-by: zhuyanlin <zhuyanlin1@xiaomi.com>
2021-11-06 07:39:27 -05:00

124 lines
3.6 KiB
ArmAsm

/****************************************************************************
* libs/libc/machine/xtensa/arch_strlen.S
*
* 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.
*
****************************************************************************/
/****************************************************************************
* Included Files
****************************************************************************/
#include "xtensa_asm.h"
#include <arch/chip/core-isa.h>
#include <arch/xtensa/xtensa_abi.h>
/****************************************************************************
* Public Functions
****************************************************************************/
.section .text
.begin schedule
.align 4
.literal_position
.global strlen
.type strlen, @function
strlen:
ENTRY(16)
/* a2 = s */
addi a3, a2, -4 # because we overincrement at the end
movi a4, MASK0
movi a5, MASK1
movi a6, MASK2
movi a7, MASK3
bbsi.l a2, 0, .L1mod2
bbsi.l a2, 1, .L2mod4
j .Laligned
.L1mod2: # address is odd
l8ui a8, a3, 4 # get byte 0
addi a3, a3, 1 # advance string pointer
beqz a8, .Lz3 # if byte 0 is zero
bbci.l a3, 1, .Laligned # if string pointer is now word-aligned
.L2mod4: # address is 2 mod 4
addi a3, a3, 2 # advance ptr for aligned access
l32i a8, a3, 0 # get word with first two bytes of string
bnone a8, a6, .Lz2 # if byte 2 (of word, not string) is zero
bany a8, a7, .Laligned # if byte 3 (of word, not string) is nonzero
/* Byte 3 is zero. */
addi a3, a3, 3 # point to zero byte
sub a2, a3, a2 # subtract to get length
RET(16)
/* String is word-aligned. */
.align 4
#if XCHAL_HAVE_LOOPS
#if XCHAL_HAVE_DENSITY
/* (2 mod 4) alignment for loop instruction */
#else
/* (1 mod 4) alignment for loop instruction */
.byte 0
.byte 0
#endif
#endif
.Laligned:
#if XCHAL_HAVE_LOOPS
#if XCHAL_HAVE_DENSITY
_movi.n a8, 0 # set up for the maximum loop count
#else
_movi a8, 0 # set up for the maximum loop count
#endif
loop a8, .Lz3 # loop forever (almost anyway)
#endif
1: l32i a8, a3, 4 # get next word of string
addi a3, a3, 4 # advance string pointer
bnone a8, a4, .Lz0 # if byte 0 is zero
bnone a8, a5, .Lz1 # if byte 1 is zero
bnone a8, a6, .Lz2 # if byte 2 is zero
#if XCHAL_HAVE_LOOPS
bnone a8, a7, .Lz3 # if byte 3 is zero
#else
bany a8, a7, 1b # repeat if byte 3 is non-zero
#endif
.Lz3: /* Byte 3 is zero. */
addi a3, a3, 3 # point to zero byte
/* Fall through.... */
.Lz0: /* Byte 0 is zero. */
sub a2, a3, a2 # subtract to get length
RET(16)
.Lz1: /* Byte 1 is zero. */
addi a3, a3, 1 # point to zero byte
sub a2, a3, a2 # subtract to get length
RET(16)
.Lz2: /* Byte 2 is zero. */
addi a3, a3, 2 # point to zero byte
sub a2, a3, a2 # subtract to get length
RET(16)
.end schedule
.size strlen, . - strlen