[thirdparty/glibc.git] / sysdeps / aarch64 / strlen.S

/* Copyright (C) 2012-2024 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 <sysdep.h>

/* Assumptions:
 *
 * ARMv8-a, AArch64, Advanced SIMD.
 * MTE compatible.
 */

#ifndef STRLEN
# define STRLEN __strlen
#endif

#define srcin		x0
#define result		x0

#define src		x1
#define	synd		x2
#define tmp		x3
#define shift		x4

#define data		q0
#define vdata		v0
#define vhas_nul	v1
#define vend		v2
#define dend		d2

/* Core algorithm:
   Process the string in 16-byte aligned chunks. Compute a 64-bit mask with
   four bits per byte using the shrn instruction. A count trailing zeros then
   identifies the first zero byte.  */

ENTRY (STRLEN)
	PTR_ARG (0)
	bic	src, srcin, 15
	ld1	{vdata.16b}, [src]
	cmeq	vhas_nul.16b, vdata.16b, 0
	lsl	shift, srcin, 2
	shrn	vend.8b, vhas_nul.8h, 4		/* 128->64 */
	fmov	synd, dend
	lsr	synd, synd, shift
	cbz	synd, L(loop)

	rbit	synd, synd
	clz	result, synd
	lsr	result, result, 2
	ret

	.p2align 5
L(loop):
	ldr	data, [src, 16]
	cmeq	vhas_nul.16b, vdata.16b, 0
	umaxp	vend.16b, vhas_nul.16b, vhas_nul.16b
	fmov	synd, dend
	cbnz	synd, L(loop_end)
	ldr	data, [src, 32]!
	cmeq	vhas_nul.16b, vdata.16b, 0
	umaxp	vend.16b, vhas_nul.16b, vhas_nul.16b
	fmov	synd, dend
	cbz	synd, L(loop)
	sub	src, src, 16
L(loop_end):
	shrn	vend.8b, vhas_nul.8h, 4		/* 128->64 */
	sub	result, src, srcin
	fmov	synd, dend
#ifndef __AARCH64EB__
	rbit	synd, synd
#endif
	add	result, result, 16
	clz	tmp, synd
	add	result, result, tmp, lsr 2
	ret

END (STRLEN)
weak_alias (STRLEN, strlen)
libc_hidden_builtin_def (strlen)
Commit	Line	Data
dff8da6b	1	/* Copyright (C) 2012-2024 Free Software Foundation, Inc.
d542f8ed MS	2
	3	This file is part of the GNU C Library.
	4
	5	The GNU C Library is free software; you can redistribute it and/or
	6	modify it under the terms of the GNU Lesser General Public
	7	License as published by the Free Software Foundation; either
	8	version 2.1 of the License, or (at your option) any later version.
	9
	10	The GNU C Library is distributed in the hope that it will be useful,
	11	but WITHOUT ANY WARRANTY; without even the implied warranty of
	12	MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
	13	Lesser General Public License for more details.
	14
	15	You should have received a copy of the GNU Lesser General Public
	16	License along with the GNU C Library. If not, see
5a82c748	17	<https://www.gnu.org/licenses/>. */
d542f8ed MS	18
	19	#include <sysdep.h>
	20
	21	/* Assumptions:
	22	*
a365ac45 AC	23	* ARMv8-a, AArch64, Advanced SIMD.
a365ac45 AC	24	* MTE compatible.
d542f8ed MS	25	*/
d542f8ed MS	26
436e4d5b SP	27	#ifndef STRLEN
	28	# define STRLEN __strlen
	29	#endif
	30
d542f8ed	31	#define srcin x0
a365ac45	32	#define result x0
d542f8ed	33
d542f8ed	34	#define src x1
a365ac45 AC	35	#define synd x2
a365ac45 AC	36	#define tmp x3
a365ac45 AC	37	#define shift x4
	38
	39	#define data q0
	40	#define vdata v0
	41	#define vhas_nul v1
3c998069 DK	42	#define vend v2
3c998069 DK	43	#define dend d2
a365ac45 AC	44
a365ac45 AC	45	/* Core algorithm:
03c8ce50 WD	46	Process the string in 16-byte aligned chunks. Compute a 64-bit mask with
	47	four bits per byte using the shrn instruction. A count trailing zeros then
	48	identifies the first zero byte. */
a365ac45 AC	49
a365ac45 AC	50	ENTRY (STRLEN)
45b1e17e	51	PTR_ARG (0)
a365ac45	52	bic src, srcin, 15
a365ac45	53	ld1 {vdata.16b}, [src]
a365ac45 AC	54	cmeq vhas_nul.16b, vdata.16b, 0
a365ac45 AC	55	lsl shift, srcin, 2
3c998069	56	shrn vend.8b, vhas_nul.8h, 4 /* 128->64 */
a365ac45 AC	57	fmov synd, dend
	58	lsr synd, synd, shift
	59	cbz synd, L(loop)
	60
	61	rbit synd, synd
	62	clz result, synd
	63	lsr result, result, 2
c435989f WD	64	ret
c435989f WD	65
a365ac45 AC	66	.p2align 5
a365ac45 AC	67	L(loop):
03c8ce50 WD	68	ldr data, [src, 16]
	69	cmeq vhas_nul.16b, vdata.16b, 0
	70	umaxp vend.16b, vhas_nul.16b, vhas_nul.16b
	71	fmov synd, dend
	72	cbnz synd, L(loop_end)
	73	ldr data, [src, 32]!
a365ac45 AC	74	cmeq vhas_nul.16b, vdata.16b, 0
	75	umaxp vend.16b, vhas_nul.16b, vhas_nul.16b
	76	fmov synd, dend
	77	cbz synd, L(loop)
03c8ce50 WD	78	sub src, src, 16
03c8ce50 WD	79	L(loop_end):
3c998069	80	shrn vend.8b, vhas_nul.8h, 4 /* 128->64 */
a365ac45 AC	81	sub result, src, srcin
	82	fmov synd, dend
	83	#ifndef __AARCH64EB__
	84	rbit synd, synd
d542f8ed	85	#endif
03c8ce50	86	add result, result, 16
a365ac45 AC	87	clz tmp, synd
a365ac45 AC	88	add result, result, tmp, lsr 2
c435989f WD	89	ret
c435989f WD	90
436e4d5b SP	91	END (STRLEN)
436e4d5b SP	92	weak_alias (STRLEN, strlen)
d542f8ed	93	libc_hidden_builtin_def (strlen)