xref: /kernel/linux/linux-5.10/arch/arm/lib/xor-neon.c (revision 8c2ecf20)
1// SPDX-License-Identifier: GPL-2.0-only
2/*
3 * linux/arch/arm/lib/xor-neon.c
4 *
5 * Copyright (C) 2013 Linaro Ltd <ard.biesheuvel@linaro.org>
6 */
7
8#include <linux/raid/xor.h>
9#include <linux/module.h>
10
11MODULE_LICENSE("GPL");
12
13#ifndef __ARM_NEON__
14#error You should compile this file with '-march=armv7-a -mfloat-abi=softfp -mfpu=neon'
15#endif
16
17/*
18 * Pull in the reference implementations while instructing GCC (through
19 * -ftree-vectorize) to attempt to exploit implicit parallelism and emit
20 * NEON instructions.
21 */
22#if __GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 6)
23#pragma GCC optimize "tree-vectorize"
24#else
25/*
26 * While older versions of GCC do not generate incorrect code, they fail to
27 * recognize the parallel nature of these functions, and emit plain ARM code,
28 * which is known to be slower than the optimized ARM code in asm-arm/xor.h.
29 *
30 * #warning This code requires at least version 4.6 of GCC
31 */
32#endif
33
34#pragma GCC diagnostic ignored "-Wunused-variable"
35#include <asm-generic/xor.h>
36
37struct xor_block_template const xor_block_neon_inner = {
38	.name	= "__inner_neon__",
39	.do_2	= xor_8regs_2,
40	.do_3	= xor_8regs_3,
41	.do_4	= xor_8regs_4,
42	.do_5	= xor_8regs_5,
43};
44EXPORT_SYMBOL(xor_block_neon_inner);
45