[mirror_ubuntu-bionic-kernel.git] / arch / arm / lib / xor-neon.c

/*
 * linux/arch/arm/lib/xor-neon.c
 *
 * Copyright (C) 2013 Linaro Ltd <ard.biesheuvel@linaro.org>
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 */

#include <linux/raid/xor.h>

#ifndef __ARM_NEON__
#error You should compile this file with '-mfloat-abi=softfp -mfpu=neon'
#endif

/*
 * Pull in the reference implementations while instructing GCC (through
 * -ftree-vectorize) to attempt to exploit implicit parallelism and emit
 * NEON instructions.
 */
#if __GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 6)
#pragma GCC optimize "tree-vectorize"
#else
/*
 * While older versions of GCC do not generate incorrect code, they fail to
 * recognize the parallel nature of these functions, and emit plain ARM code,
 * which is known to be slower than the optimized ARM code in asm-arm/xor.h.
 */
#warning This code requires at least version 4.6 of GCC
#endif

#pragma GCC diagnostic ignored "-Wunused-variable"
#include <asm-generic/xor.h>

struct xor_block_template const xor_block_neon_inner = {
	.name	= "__inner_neon__",
	.do_2	= xor_8regs_2,
	.do_3	= xor_8regs_3,
	.do_4	= xor_8regs_4,
	.do_5	= xor_8regs_5,
};
Commit	Line	Data
01956597 AB	1	/*
	2	* linux/arch/arm/lib/xor-neon.c
	3	*
	4	* Copyright (C) 2013 Linaro Ltd <ard.biesheuvel@linaro.org>
	5	*
	6	* This program is free software; you can redistribute it and/or modify
	7	* it under the terms of the GNU General Public License version 2 as
	8	* published by the Free Software Foundation.
	9	*/
	10
	11	#include <linux/raid/xor.h>
	12
	13	#ifndef __ARM_NEON__
	14	#error You should compile this file with '-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	#endif
	32
	33	#pragma GCC diagnostic ignored "-Wunused-variable"
	34	#include <asm-generic/xor.h>
	35
	36	struct xor_block_template const xor_block_neon_inner = {
	37	.name = "__inner_neon__",
	38	.do_2 = xor_8regs_2,
	39	.do_3 = xor_8regs_3,
	40	.do_4 = xor_8regs_4,
	41	.do_5 = xor_8regs_5,
	42	};