1/* Memset for falkor.
2   Copyright (C) 2017-2022 Free Software Foundation, Inc.
3
4   This file is part of the GNU C Library.
5
6   The GNU C Library is free software; you can redistribute it and/or
7   modify it under the terms of the GNU Lesser General Public
8   License as published by the Free Software Foundation; either
9   version 2.1 of the License, or (at your option) any later version.
10
11   The GNU C Library is distributed in the hope that it will be useful,
12   but WITHOUT ANY WARRANTY; without even the implied warranty of
13   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
14   Lesser General Public License for more details.
15
16   You should have received a copy of the GNU Lesser General Public
17   License along with the GNU C Library.  If not, see
18   <https://www.gnu.org/licenses/>.  */
19
20#include <sysdep.h>
21#include <memset-reg.h>
22
23/* Reading dczid_el0 is expensive on falkor so move it into the ifunc
24   resolver and assume ZVA size of 64 bytes.  The IFUNC resolver takes care to
25   use this function only when ZVA is enabled.  */
26
27#if IS_IN (libc)
28.macro zva_macro
29	.p2align 4
30	/* Write the first and last 64 byte aligned block using stp rather
31	   than using DC ZVA.  This is faster on some cores.  */
32	str	q0, [dst, 16]
33	stp	q0, q0, [dst, 32]
34	bic	dst, dst, 63
35	stp	q0, q0, [dst, 64]
36	stp	q0, q0, [dst, 96]
37	sub	count, dstend, dst	/* Count is now 128 too large.	*/
38	sub	count, count, 128+64+64	/* Adjust count and bias for loop.  */
39	add	dst, dst, 128
401:	dc	zva, dst
41	add	dst, dst, 64
42	subs	count, count, 64
43	b.hi	1b
44	stp	q0, q0, [dst, 0]
45	stp	q0, q0, [dst, 32]
46	stp	q0, q0, [dstend, -64]
47	stp	q0, q0, [dstend, -32]
48	ret
49.endm
50
51# define ZVA_MACRO zva_macro
52# define MEMSET __memset_falkor
53# include <sysdeps/aarch64/memset.S>
54#endif
55