linux/arch/x86/crypto/cast6_avx_glue.c

306 lines
8.1 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-or-later
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
/*
* Glue Code for the AVX assembler implementation of the Cast6 Cipher
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
*
* Copyright (C) 2012 Johannes Goetzfried
* <Johannes.Goetzfried@informatik.stud.uni-erlangen.de>
*
* Copyright © 2013 Jussi Kivilinna <jussi.kivilinna@iki.fi>
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
*/
#include <linux/module.h>
#include <linux/types.h>
#include <linux/crypto.h>
#include <linux/err.h>
#include <crypto/algapi.h>
#include <crypto/cast6.h>
#include <crypto/internal/simd.h>
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
#include <crypto/xts.h>
#include <asm/crypto/glue_helper.h>
#define CAST6_PARALLEL_BLOCKS 8
asmlinkage void cast6_ecb_enc_8way(struct cast6_ctx *ctx, u8 *dst,
const u8 *src);
asmlinkage void cast6_ecb_dec_8way(struct cast6_ctx *ctx, u8 *dst,
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
const u8 *src);
asmlinkage void cast6_cbc_dec_8way(struct cast6_ctx *ctx, u8 *dst,
const u8 *src);
asmlinkage void cast6_ctr_8way(struct cast6_ctx *ctx, u8 *dst, const u8 *src,
le128 *iv);
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
asmlinkage void cast6_xts_enc_8way(struct cast6_ctx *ctx, u8 *dst,
const u8 *src, le128 *iv);
asmlinkage void cast6_xts_dec_8way(struct cast6_ctx *ctx, u8 *dst,
const u8 *src, le128 *iv);
static int cast6_setkey_skcipher(struct crypto_skcipher *tfm,
const u8 *key, unsigned int keylen)
{
return cast6_setkey(&tfm->base, key, keylen);
}
static void cast6_xts_enc(void *ctx, u128 *dst, const u128 *src, le128 *iv)
{
glue_xts_crypt_128bit_one(ctx, dst, src, iv,
GLUE_FUNC_CAST(__cast6_encrypt));
}
static void cast6_xts_dec(void *ctx, u128 *dst, const u128 *src, le128 *iv)
{
glue_xts_crypt_128bit_one(ctx, dst, src, iv,
GLUE_FUNC_CAST(__cast6_decrypt));
}
static void cast6_crypt_ctr(void *ctx, u128 *dst, const u128 *src, le128 *iv)
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
{
be128 ctrblk;
le128_to_be128(&ctrblk, iv);
le128_inc(iv);
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
__cast6_encrypt(ctx, (u8 *)&ctrblk, (u8 *)&ctrblk);
u128_xor(dst, src, (u128 *)&ctrblk);
}
static const struct common_glue_ctx cast6_enc = {
.num_funcs = 2,
.fpu_blocks_limit = CAST6_PARALLEL_BLOCKS,
.funcs = { {
.num_blocks = CAST6_PARALLEL_BLOCKS,
.fn_u = { .ecb = GLUE_FUNC_CAST(cast6_ecb_enc_8way) }
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
}, {
.num_blocks = 1,
.fn_u = { .ecb = GLUE_FUNC_CAST(__cast6_encrypt) }
} }
};
static const struct common_glue_ctx cast6_ctr = {
.num_funcs = 2,
.fpu_blocks_limit = CAST6_PARALLEL_BLOCKS,
.funcs = { {
.num_blocks = CAST6_PARALLEL_BLOCKS,
.fn_u = { .ctr = GLUE_CTR_FUNC_CAST(cast6_ctr_8way) }
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
}, {
.num_blocks = 1,
.fn_u = { .ctr = GLUE_CTR_FUNC_CAST(cast6_crypt_ctr) }
} }
};
static const struct common_glue_ctx cast6_enc_xts = {
.num_funcs = 2,
.fpu_blocks_limit = CAST6_PARALLEL_BLOCKS,
.funcs = { {
.num_blocks = CAST6_PARALLEL_BLOCKS,
.fn_u = { .xts = GLUE_XTS_FUNC_CAST(cast6_xts_enc_8way) }
}, {
.num_blocks = 1,
.fn_u = { .xts = GLUE_XTS_FUNC_CAST(cast6_xts_enc) }
} }
};
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
static const struct common_glue_ctx cast6_dec = {
.num_funcs = 2,
.fpu_blocks_limit = CAST6_PARALLEL_BLOCKS,
.funcs = { {
.num_blocks = CAST6_PARALLEL_BLOCKS,
.fn_u = { .ecb = GLUE_FUNC_CAST(cast6_ecb_dec_8way) }
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
}, {
.num_blocks = 1,
.fn_u = { .ecb = GLUE_FUNC_CAST(__cast6_decrypt) }
} }
};
static const struct common_glue_ctx cast6_dec_cbc = {
.num_funcs = 2,
.fpu_blocks_limit = CAST6_PARALLEL_BLOCKS,
.funcs = { {
.num_blocks = CAST6_PARALLEL_BLOCKS,
.fn_u = { .cbc = GLUE_CBC_FUNC_CAST(cast6_cbc_dec_8way) }
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
}, {
.num_blocks = 1,
.fn_u = { .cbc = GLUE_CBC_FUNC_CAST(__cast6_decrypt) }
} }
};
static const struct common_glue_ctx cast6_dec_xts = {
.num_funcs = 2,
.fpu_blocks_limit = CAST6_PARALLEL_BLOCKS,
.funcs = { {
.num_blocks = CAST6_PARALLEL_BLOCKS,
.fn_u = { .xts = GLUE_XTS_FUNC_CAST(cast6_xts_dec_8way) }
}, {
.num_blocks = 1,
.fn_u = { .xts = GLUE_XTS_FUNC_CAST(cast6_xts_dec) }
} }
};
static int ecb_encrypt(struct skcipher_request *req)
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
{
return glue_ecb_req_128bit(&cast6_enc, req);
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
}
static int ecb_decrypt(struct skcipher_request *req)
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
{
return glue_ecb_req_128bit(&cast6_dec, req);
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
}
static int cbc_encrypt(struct skcipher_request *req)
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
{
return glue_cbc_encrypt_req_128bit(GLUE_FUNC_CAST(__cast6_encrypt),
req);
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
}
static int cbc_decrypt(struct skcipher_request *req)
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
{
return glue_cbc_decrypt_req_128bit(&cast6_dec_cbc, req);
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
}
static int ctr_crypt(struct skcipher_request *req)
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
{
return glue_ctr_req_128bit(&cast6_ctr, req);
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
}
struct cast6_xts_ctx {
struct cast6_ctx tweak_ctx;
struct cast6_ctx crypt_ctx;
};
static int xts_cast6_setkey(struct crypto_skcipher *tfm, const u8 *key,
unsigned int keylen)
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
{
struct cast6_xts_ctx *ctx = crypto_skcipher_ctx(tfm);
u32 *flags = &tfm->base.crt_flags;
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
int err;
err = xts_verify_key(tfm, key, keylen);
if (err)
return err;
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
/* first half of xts-key is for crypt */
err = __cast6_setkey(&ctx->crypt_ctx, key, keylen / 2, flags);
if (err)
return err;
/* second half of xts-key is for tweak */
return __cast6_setkey(&ctx->tweak_ctx, key + keylen / 2, keylen / 2,
flags);
}
static int xts_encrypt(struct skcipher_request *req)
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
{
struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
struct cast6_xts_ctx *ctx = crypto_skcipher_ctx(tfm);
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
return glue_xts_req_128bit(&cast6_enc_xts, req,
XTS_TWEAK_CAST(__cast6_encrypt),
&ctx->tweak_ctx, &ctx->crypt_ctx, false);
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
}
static int xts_decrypt(struct skcipher_request *req)
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
{
struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
struct cast6_xts_ctx *ctx = crypto_skcipher_ctx(tfm);
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
return glue_xts_req_128bit(&cast6_dec_xts, req,
XTS_TWEAK_CAST(__cast6_encrypt),
&ctx->tweak_ctx, &ctx->crypt_ctx, true);
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
}
static struct skcipher_alg cast6_algs[] = {
{
.base.cra_name = "__ecb(cast6)",
.base.cra_driver_name = "__ecb-cast6-avx",
.base.cra_priority = 200,
.base.cra_flags = CRYPTO_ALG_INTERNAL,
.base.cra_blocksize = CAST6_BLOCK_SIZE,
.base.cra_ctxsize = sizeof(struct cast6_ctx),
.base.cra_module = THIS_MODULE,
.min_keysize = CAST6_MIN_KEY_SIZE,
.max_keysize = CAST6_MAX_KEY_SIZE,
.setkey = cast6_setkey_skcipher,
.encrypt = ecb_encrypt,
.decrypt = ecb_decrypt,
}, {
.base.cra_name = "__cbc(cast6)",
.base.cra_driver_name = "__cbc-cast6-avx",
.base.cra_priority = 200,
.base.cra_flags = CRYPTO_ALG_INTERNAL,
.base.cra_blocksize = CAST6_BLOCK_SIZE,
.base.cra_ctxsize = sizeof(struct cast6_ctx),
.base.cra_module = THIS_MODULE,
.min_keysize = CAST6_MIN_KEY_SIZE,
.max_keysize = CAST6_MAX_KEY_SIZE,
.ivsize = CAST6_BLOCK_SIZE,
.setkey = cast6_setkey_skcipher,
.encrypt = cbc_encrypt,
.decrypt = cbc_decrypt,
}, {
.base.cra_name = "__ctr(cast6)",
.base.cra_driver_name = "__ctr-cast6-avx",
.base.cra_priority = 200,
.base.cra_flags = CRYPTO_ALG_INTERNAL,
.base.cra_blocksize = 1,
.base.cra_ctxsize = sizeof(struct cast6_ctx),
.base.cra_module = THIS_MODULE,
.min_keysize = CAST6_MIN_KEY_SIZE,
.max_keysize = CAST6_MAX_KEY_SIZE,
.ivsize = CAST6_BLOCK_SIZE,
.chunksize = CAST6_BLOCK_SIZE,
.setkey = cast6_setkey_skcipher,
.encrypt = ctr_crypt,
.decrypt = ctr_crypt,
}, {
.base.cra_name = "__xts(cast6)",
.base.cra_driver_name = "__xts-cast6-avx",
.base.cra_priority = 200,
.base.cra_flags = CRYPTO_ALG_INTERNAL,
.base.cra_blocksize = CAST6_BLOCK_SIZE,
.base.cra_ctxsize = sizeof(struct cast6_xts_ctx),
.base.cra_module = THIS_MODULE,
.min_keysize = 2 * CAST6_MIN_KEY_SIZE,
.max_keysize = 2 * CAST6_MAX_KEY_SIZE,
.ivsize = CAST6_BLOCK_SIZE,
.setkey = xts_cast6_setkey,
.encrypt = xts_encrypt,
.decrypt = xts_decrypt,
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
},
};
static struct simd_skcipher_alg *cast6_simd_algs[ARRAY_SIZE(cast6_algs)];
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
static int __init cast6_init(void)
{
const char *feature_name;
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
x86/fpu: Rename XSAVE macros There are two concepts that have some confusing naming: 1. Extended State Component numbers (currently called XFEATURE_BIT_*) 2. Extended State Component masks (currently called XSTATE_*) The numbers are (currently) from 0-9. State component 3 is the bounds registers for MPX, for instance. But when we want to enable "state component 3", we go set a bit in XCR0. The bit we set is 1<<3. We can check to see if a state component feature is enabled by looking at its bit. The current 'xfeature_bit's are at best xfeature bit _numbers_. Calling them bits is at best inconsistent with ending the enum list with 'XFEATURES_NR_MAX'. This patch renames the enum to be 'xfeature'. These also happen to be what the Intel documentation calls a "state component". We also want to differentiate these from the "XSTATE_*" macros. The "XSTATE_*" macros are a mask, and we rename them to match. These macros are reasonably widely used so this patch is a wee bit big, but this really is just a rename. The only non-mechanical part of this is the s/XSTATE_EXTEND_MASK/XFEATURE_MASK_EXTEND/ We need a better name for it, but that's another patch. Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Tim Chen <tim.c.chen@linux.intel.com> Cc: dave@sr71.net Cc: linux-kernel@vger.kernel.org Link: http://lkml.kernel.org/r/20150902233126.38653250@viggo.jf.intel.com [ Ported to v4.3-rc1. ] Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-09-03 07:31:26 +08:00
if (!cpu_has_xfeatures(XFEATURE_MASK_SSE | XFEATURE_MASK_YMM,
&feature_name)) {
pr_info("CPU feature '%s' is not supported.\n", feature_name);
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
return -ENODEV;
}
return simd_register_skciphers_compat(cast6_algs,
ARRAY_SIZE(cast6_algs),
cast6_simd_algs);
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
}
static void __exit cast6_exit(void)
{
simd_unregister_skciphers(cast6_algs, ARRAY_SIZE(cast6_algs),
cast6_simd_algs);
crypto: cast6 - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Cast6 block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the functions from the generic module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) cast6-avx-x86_64 vs. cast6-generic 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 1.00x 1.01x 1.01x 0.99x 0.97x 0.98x 1.01x 0.96x 0.98x 64B 0.98x 0.99x 1.02x 1.01x 0.99x 1.00x 1.01x 0.99x 1.00x 0.99x 256B 1.77x 1.84x 0.99x 1.85x 1.77x 1.77x 1.70x 1.74x 1.69x 1.72x 1024B 1.93x 1.95x 0.99x 1.96x 1.93x 1.93x 1.84x 1.85x 1.89x 1.87x 8192B 1.91x 1.95x 0.99x 1.97x 1.95x 1.91x 1.86x 1.87x 1.93x 1.90x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.97x 0.99x 1.02x 1.01x 0.98x 0.99x 1.00x 1.00x 0.98x 0.98x 64B 0.98x 0.99x 1.01x 1.00x 1.00x 1.00x 1.01x 1.01x 0.97x 1.00x 256B 1.77x 1.83x 1.00x 1.86x 1.79x 1.78x 1.70x 1.76x 1.71x 1.69x 1024B 1.92x 1.95x 0.99x 1.96x 1.93x 1.93x 1.83x 1.86x 1.89x 1.87x 8192B 1.94x 1.95x 0.99x 1.97x 1.95x 1.95x 1.87x 1.87x 1.93x 1.91x Signed-off-by: Johannes Goetzfried <Johannes.Goetzfried@informatik.stud.uni-erlangen.de> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-07-12 01:38:57 +08:00
}
module_init(cast6_init);
module_exit(cast6_exit);
MODULE_DESCRIPTION("Cast6 Cipher Algorithm, AVX optimized");
MODULE_LICENSE("GPL");
MODULE_ALIAS_CRYPTO("cast6");