2013-09-07 02:56:26 +00:00
|
|
|
/*
|
|
|
|
* Cryptographic API.
|
|
|
|
*
|
|
|
|
* T10 Data Integrity Field CRC16 Crypto Transform using PCLMULQDQ Instructions
|
|
|
|
*
|
|
|
|
* Copyright (C) 2013 Intel Corporation
|
|
|
|
* Author: Tim Chen <tim.c.chen@linux.intel.com>
|
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or modify it
|
|
|
|
* under the terms of the GNU General Public License as published by the Free
|
|
|
|
* Software Foundation; either version 2 of the License, or (at your option)
|
|
|
|
* any later version.
|
|
|
|
*
|
|
|
|
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
|
|
|
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
|
|
|
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
|
|
|
|
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
|
|
|
|
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
|
|
|
|
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
|
|
|
|
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
|
|
|
* SOFTWARE.
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/types.h>
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/crc-t10dif.h>
|
|
|
|
#include <crypto/internal/hash.h>
|
2019-03-13 05:12:48 +00:00
|
|
|
#include <crypto/internal/simd.h>
|
2013-09-07 02:56:26 +00:00
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/string.h>
|
|
|
|
#include <linux/kernel.h>
|
2016-01-26 21:12:04 +00:00
|
|
|
#include <asm/cpufeatures.h>
|
2013-09-07 02:56:26 +00:00
|
|
|
#include <asm/cpu_device_id.h>
|
2019-03-13 05:12:48 +00:00
|
|
|
#include <asm/simd.h>
|
2013-09-07 02:56:26 +00:00
|
|
|
|
crypto: x86/crct10dif-pcl - cleanup and optimizations
The x86, arm, and arm64 asm implementations of crct10dif are very
difficult to understand partly because many of the comments, labels, and
macros are named incorrectly: the lengths mentioned are usually off by a
factor of two from the actual code. Many other things are unnecessarily
convoluted as well, e.g. there are many more fold constants than
actually needed and some aren't fully reduced.
This series therefore cleans up all these implementations to be much
more maintainable. I also made some small optimizations where I saw
opportunities, resulting in slightly better performance.
This patch cleans up the x86 version.
As part of this, I removed support for len < 16 from the x86 assembly;
now the glue code falls back to the generic table-based implementation
in this case. Due to the overhead of kernel_fpu_begin(), this actually
significantly improves performance on these lengths. (And even if
kernel_fpu_begin() were free, the generic code is still faster for about
len < 11.) This removal also eliminates error-prone special cases and
makes the x86, arm32, and arm64 ports of the code match more closely.
Acked-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2019-01-31 04:42:40 +00:00
|
|
|
asmlinkage u16 crc_t10dif_pcl(u16 init_crc, const u8 *buf, size_t len);
|
2013-09-07 02:56:26 +00:00
|
|
|
|
|
|
|
struct chksum_desc_ctx {
|
|
|
|
__u16 crc;
|
|
|
|
};
|
|
|
|
|
|
|
|
static int chksum_init(struct shash_desc *desc)
|
|
|
|
{
|
|
|
|
struct chksum_desc_ctx *ctx = shash_desc_ctx(desc);
|
|
|
|
|
|
|
|
ctx->crc = 0;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int chksum_update(struct shash_desc *desc, const u8 *data,
|
|
|
|
unsigned int length)
|
|
|
|
{
|
|
|
|
struct chksum_desc_ctx *ctx = shash_desc_ctx(desc);
|
|
|
|
|
2019-03-13 05:12:48 +00:00
|
|
|
if (length >= 16 && crypto_simd_usable()) {
|
2013-09-07 02:56:26 +00:00
|
|
|
kernel_fpu_begin();
|
|
|
|
ctx->crc = crc_t10dif_pcl(ctx->crc, data, length);
|
|
|
|
kernel_fpu_end();
|
|
|
|
} else
|
|
|
|
ctx->crc = crc_t10dif_generic(ctx->crc, data, length);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int chksum_final(struct shash_desc *desc, u8 *out)
|
|
|
|
{
|
|
|
|
struct chksum_desc_ctx *ctx = shash_desc_ctx(desc);
|
|
|
|
|
|
|
|
*(__u16 *)out = ctx->crc;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2019-03-31 20:04:13 +00:00
|
|
|
static int __chksum_finup(__u16 crc, const u8 *data, unsigned int len, u8 *out)
|
2013-09-07 02:56:26 +00:00
|
|
|
{
|
2019-03-13 05:12:48 +00:00
|
|
|
if (len >= 16 && crypto_simd_usable()) {
|
2013-09-07 02:56:26 +00:00
|
|
|
kernel_fpu_begin();
|
2019-03-31 20:04:13 +00:00
|
|
|
*(__u16 *)out = crc_t10dif_pcl(crc, data, len);
|
2013-09-07 02:56:26 +00:00
|
|
|
kernel_fpu_end();
|
|
|
|
} else
|
2019-03-31 20:04:13 +00:00
|
|
|
*(__u16 *)out = crc_t10dif_generic(crc, data, len);
|
2013-09-07 02:56:26 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int chksum_finup(struct shash_desc *desc, const u8 *data,
|
|
|
|
unsigned int len, u8 *out)
|
|
|
|
{
|
|
|
|
struct chksum_desc_ctx *ctx = shash_desc_ctx(desc);
|
|
|
|
|
2019-03-31 20:04:13 +00:00
|
|
|
return __chksum_finup(ctx->crc, data, len, out);
|
2013-09-07 02:56:26 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static int chksum_digest(struct shash_desc *desc, const u8 *data,
|
|
|
|
unsigned int length, u8 *out)
|
|
|
|
{
|
2019-03-31 20:04:13 +00:00
|
|
|
return __chksum_finup(0, data, length, out);
|
2013-09-07 02:56:26 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static struct shash_alg alg = {
|
|
|
|
.digestsize = CRC_T10DIF_DIGEST_SIZE,
|
|
|
|
.init = chksum_init,
|
|
|
|
.update = chksum_update,
|
|
|
|
.final = chksum_final,
|
|
|
|
.finup = chksum_finup,
|
|
|
|
.digest = chksum_digest,
|
|
|
|
.descsize = sizeof(struct chksum_desc_ctx),
|
|
|
|
.base = {
|
|
|
|
.cra_name = "crct10dif",
|
|
|
|
.cra_driver_name = "crct10dif-pclmul",
|
|
|
|
.cra_priority = 200,
|
|
|
|
.cra_blocksize = CRC_T10DIF_BLOCK_SIZE,
|
|
|
|
.cra_module = THIS_MODULE,
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
static const struct x86_cpu_id crct10dif_cpu_id[] = {
|
2020-03-20 13:14:05 +00:00
|
|
|
X86_MATCH_FEATURE(X86_FEATURE_PCLMULQDQ, NULL),
|
2013-09-07 02:56:26 +00:00
|
|
|
{}
|
|
|
|
};
|
|
|
|
MODULE_DEVICE_TABLE(x86cpu, crct10dif_cpu_id);
|
|
|
|
|
|
|
|
static int __init crct10dif_intel_mod_init(void)
|
|
|
|
{
|
|
|
|
if (!x86_match_cpu(crct10dif_cpu_id))
|
|
|
|
return -ENODEV;
|
|
|
|
|
|
|
|
return crypto_register_shash(&alg);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void __exit crct10dif_intel_mod_fini(void)
|
|
|
|
{
|
|
|
|
crypto_unregister_shash(&alg);
|
|
|
|
}
|
|
|
|
|
|
|
|
module_init(crct10dif_intel_mod_init);
|
|
|
|
module_exit(crct10dif_intel_mod_fini);
|
|
|
|
|
|
|
|
MODULE_AUTHOR("Tim Chen <tim.c.chen@linux.intel.com>");
|
|
|
|
MODULE_DESCRIPTION("T10 DIF CRC calculation accelerated with PCLMULQDQ.");
|
|
|
|
MODULE_LICENSE("GPL");
|
|
|
|
|
2014-11-21 01:05:53 +00:00
|
|
|
MODULE_ALIAS_CRYPTO("crct10dif");
|
|
|
|
MODULE_ALIAS_CRYPTO("crct10dif-pclmul");
|