2019-05-27 13:55:01 +07:00
|
|
|
// SPDX-License-Identifier: GPL-2.0-or-later
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
/*
|
2018-12-05 13:20:03 +07:00
|
|
|
* x64 SIMD accelerated ChaCha and XChaCha stream ciphers,
|
|
|
|
* including ChaCha20 (RFC7539)
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
*
|
|
|
|
* Copyright (C) 2015 Martin Willi
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <crypto/algapi.h>
|
2019-11-08 19:22:08 +07:00
|
|
|
#include <crypto/internal/chacha.h>
|
2019-03-13 12:12:48 +07:00
|
|
|
#include <crypto/internal/simd.h>
|
2016-12-09 21:33:51 +07:00
|
|
|
#include <crypto/internal/skcipher.h>
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <asm/simd.h>
|
|
|
|
|
2018-12-05 13:20:03 +07:00
|
|
|
#define CHACHA_STATE_ALIGN 16
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
|
2018-12-05 13:20:03 +07:00
|
|
|
asmlinkage void chacha_block_xor_ssse3(u32 *state, u8 *dst, const u8 *src,
|
|
|
|
unsigned int len, int nrounds);
|
|
|
|
asmlinkage void chacha_4block_xor_ssse3(u32 *state, u8 *dst, const u8 *src,
|
|
|
|
unsigned int len, int nrounds);
|
|
|
|
asmlinkage void hchacha_block_ssse3(const u32 *state, u32 *out, int nrounds);
|
2019-11-08 19:22:10 +07:00
|
|
|
|
2018-12-05 13:20:03 +07:00
|
|
|
asmlinkage void chacha_2block_xor_avx2(u32 *state, u8 *dst, const u8 *src,
|
|
|
|
unsigned int len, int nrounds);
|
|
|
|
asmlinkage void chacha_4block_xor_avx2(u32 *state, u8 *dst, const u8 *src,
|
|
|
|
unsigned int len, int nrounds);
|
|
|
|
asmlinkage void chacha_8block_xor_avx2(u32 *state, u8 *dst, const u8 *src,
|
|
|
|
unsigned int len, int nrounds);
|
2019-11-08 19:22:10 +07:00
|
|
|
|
2018-12-05 13:20:03 +07:00
|
|
|
asmlinkage void chacha_2block_xor_avx512vl(u32 *state, u8 *dst, const u8 *src,
|
|
|
|
unsigned int len, int nrounds);
|
|
|
|
asmlinkage void chacha_4block_xor_avx512vl(u32 *state, u8 *dst, const u8 *src,
|
|
|
|
unsigned int len, int nrounds);
|
|
|
|
asmlinkage void chacha_8block_xor_avx512vl(u32 *state, u8 *dst, const u8 *src,
|
|
|
|
unsigned int len, int nrounds);
|
2019-11-08 19:22:10 +07:00
|
|
|
|
|
|
|
static __ro_after_init DEFINE_STATIC_KEY_FALSE(chacha_use_simd);
|
|
|
|
static __ro_after_init DEFINE_STATIC_KEY_FALSE(chacha_use_avx2);
|
|
|
|
static __ro_after_init DEFINE_STATIC_KEY_FALSE(chacha_use_avx512vl);
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
|
2018-12-05 13:20:03 +07:00
|
|
|
static unsigned int chacha_advance(unsigned int len, unsigned int maxblocks)
|
2018-11-11 16:36:28 +07:00
|
|
|
{
|
2018-11-17 08:26:21 +07:00
|
|
|
len = min(len, maxblocks * CHACHA_BLOCK_SIZE);
|
|
|
|
return round_up(len, CHACHA_BLOCK_SIZE) / CHACHA_BLOCK_SIZE;
|
2018-11-11 16:36:28 +07:00
|
|
|
}
|
|
|
|
|
2018-12-05 13:20:03 +07:00
|
|
|
static void chacha_dosimd(u32 *state, u8 *dst, const u8 *src,
|
|
|
|
unsigned int bytes, int nrounds)
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
{
|
2019-11-08 19:22:10 +07:00
|
|
|
if (IS_ENABLED(CONFIG_AS_AVX512) &&
|
|
|
|
static_branch_likely(&chacha_use_avx512vl)) {
|
2018-11-20 23:30:48 +07:00
|
|
|
while (bytes >= CHACHA_BLOCK_SIZE * 8) {
|
2018-12-05 13:20:03 +07:00
|
|
|
chacha_8block_xor_avx512vl(state, dst, src, bytes,
|
|
|
|
nrounds);
|
2018-11-20 23:30:48 +07:00
|
|
|
bytes -= CHACHA_BLOCK_SIZE * 8;
|
|
|
|
src += CHACHA_BLOCK_SIZE * 8;
|
|
|
|
dst += CHACHA_BLOCK_SIZE * 8;
|
|
|
|
state[12] += 8;
|
|
|
|
}
|
|
|
|
if (bytes > CHACHA_BLOCK_SIZE * 4) {
|
2018-12-05 13:20:03 +07:00
|
|
|
chacha_8block_xor_avx512vl(state, dst, src, bytes,
|
|
|
|
nrounds);
|
|
|
|
state[12] += chacha_advance(bytes, 8);
|
2018-11-20 23:30:48 +07:00
|
|
|
return;
|
|
|
|
}
|
2018-11-20 23:30:50 +07:00
|
|
|
if (bytes > CHACHA_BLOCK_SIZE * 2) {
|
2018-12-05 13:20:03 +07:00
|
|
|
chacha_4block_xor_avx512vl(state, dst, src, bytes,
|
|
|
|
nrounds);
|
|
|
|
state[12] += chacha_advance(bytes, 4);
|
2018-11-20 23:30:50 +07:00
|
|
|
return;
|
|
|
|
}
|
2018-11-20 23:30:49 +07:00
|
|
|
if (bytes) {
|
2018-12-05 13:20:03 +07:00
|
|
|
chacha_2block_xor_avx512vl(state, dst, src, bytes,
|
|
|
|
nrounds);
|
|
|
|
state[12] += chacha_advance(bytes, 2);
|
2018-11-20 23:30:49 +07:00
|
|
|
return;
|
|
|
|
}
|
2018-11-20 23:30:48 +07:00
|
|
|
}
|
2019-11-08 19:22:10 +07:00
|
|
|
|
2020-03-27 03:26:00 +07:00
|
|
|
if (static_branch_likely(&chacha_use_avx2)) {
|
2018-11-17 08:26:21 +07:00
|
|
|
while (bytes >= CHACHA_BLOCK_SIZE * 8) {
|
2018-12-05 13:20:03 +07:00
|
|
|
chacha_8block_xor_avx2(state, dst, src, bytes, nrounds);
|
2018-11-17 08:26:21 +07:00
|
|
|
bytes -= CHACHA_BLOCK_SIZE * 8;
|
|
|
|
src += CHACHA_BLOCK_SIZE * 8;
|
|
|
|
dst += CHACHA_BLOCK_SIZE * 8;
|
crypto: chacha20 - Add an eight block AVX2 variant for x86_64
Extends the x86_64 ChaCha20 implementation by a function processing eight
ChaCha20 blocks in parallel using AVX2.
For large messages, throughput increases by ~55-70% compared to four block
SSSE3:
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 42249230 operations in 10 seconds (675987680 bytes)
test 1 (256 bit key, 64 byte blocks): 46441641 operations in 10 seconds (2972265024 bytes)
test 2 (256 bit key, 256 byte blocks): 33028112 operations in 10 seconds (8455196672 bytes)
test 3 (256 bit key, 1024 byte blocks): 11568759 operations in 10 seconds (11846409216 bytes)
test 4 (256 bit key, 8192 byte blocks): 1448761 operations in 10 seconds (11868250112 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 41999675 operations in 10 seconds (671994800 bytes)
test 1 (256 bit key, 64 byte blocks): 45805908 operations in 10 seconds (2931578112 bytes)
test 2 (256 bit key, 256 byte blocks): 32814947 operations in 10 seconds (8400626432 bytes)
test 3 (256 bit key, 1024 byte blocks): 19777167 operations in 10 seconds (20251819008 bytes)
test 4 (256 bit key, 8192 byte blocks): 2279321 operations in 10 seconds (18672197632 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:03 +07:00
|
|
|
state[12] += 8;
|
|
|
|
}
|
2018-11-17 08:26:21 +07:00
|
|
|
if (bytes > CHACHA_BLOCK_SIZE * 4) {
|
2018-12-05 13:20:03 +07:00
|
|
|
chacha_8block_xor_avx2(state, dst, src, bytes, nrounds);
|
|
|
|
state[12] += chacha_advance(bytes, 8);
|
2018-11-11 16:36:28 +07:00
|
|
|
return;
|
|
|
|
}
|
2018-11-17 08:26:21 +07:00
|
|
|
if (bytes > CHACHA_BLOCK_SIZE * 2) {
|
2018-12-05 13:20:03 +07:00
|
|
|
chacha_4block_xor_avx2(state, dst, src, bytes, nrounds);
|
|
|
|
state[12] += chacha_advance(bytes, 4);
|
2018-11-11 16:36:30 +07:00
|
|
|
return;
|
|
|
|
}
|
2018-11-17 08:26:21 +07:00
|
|
|
if (bytes > CHACHA_BLOCK_SIZE) {
|
2018-12-05 13:20:03 +07:00
|
|
|
chacha_2block_xor_avx2(state, dst, src, bytes, nrounds);
|
|
|
|
state[12] += chacha_advance(bytes, 2);
|
2018-11-11 16:36:29 +07:00
|
|
|
return;
|
|
|
|
}
|
crypto: chacha20 - Add an eight block AVX2 variant for x86_64
Extends the x86_64 ChaCha20 implementation by a function processing eight
ChaCha20 blocks in parallel using AVX2.
For large messages, throughput increases by ~55-70% compared to four block
SSSE3:
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 42249230 operations in 10 seconds (675987680 bytes)
test 1 (256 bit key, 64 byte blocks): 46441641 operations in 10 seconds (2972265024 bytes)
test 2 (256 bit key, 256 byte blocks): 33028112 operations in 10 seconds (8455196672 bytes)
test 3 (256 bit key, 1024 byte blocks): 11568759 operations in 10 seconds (11846409216 bytes)
test 4 (256 bit key, 8192 byte blocks): 1448761 operations in 10 seconds (11868250112 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 41999675 operations in 10 seconds (671994800 bytes)
test 1 (256 bit key, 64 byte blocks): 45805908 operations in 10 seconds (2931578112 bytes)
test 2 (256 bit key, 256 byte blocks): 32814947 operations in 10 seconds (8400626432 bytes)
test 3 (256 bit key, 1024 byte blocks): 19777167 operations in 10 seconds (20251819008 bytes)
test 4 (256 bit key, 8192 byte blocks): 2279321 operations in 10 seconds (18672197632 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:03 +07:00
|
|
|
}
|
2019-11-08 19:22:10 +07:00
|
|
|
|
2018-11-17 08:26:21 +07:00
|
|
|
while (bytes >= CHACHA_BLOCK_SIZE * 4) {
|
2018-12-05 13:20:03 +07:00
|
|
|
chacha_4block_xor_ssse3(state, dst, src, bytes, nrounds);
|
2018-11-17 08:26:21 +07:00
|
|
|
bytes -= CHACHA_BLOCK_SIZE * 4;
|
|
|
|
src += CHACHA_BLOCK_SIZE * 4;
|
|
|
|
dst += CHACHA_BLOCK_SIZE * 4;
|
crypto: chacha20 - Add a four block SSSE3 variant for x86_64
Extends the x86_64 SSSE3 ChaCha20 implementation by a function processing
four ChaCha20 blocks in parallel. This avoids the word shuffling needed
in the single block variant, further increasing throughput.
For large messages, throughput increases by ~110% compared to single block
SSSE3:
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 42249230 operations in 10 seconds (675987680 bytes)
test 1 (256 bit key, 64 byte blocks): 46441641 operations in 10 seconds (2972265024 bytes)
test 2 (256 bit key, 256 byte blocks): 33028112 operations in 10 seconds (8455196672 bytes)
test 3 (256 bit key, 1024 byte blocks): 11568759 operations in 10 seconds (11846409216 bytes)
test 4 (256 bit key, 8192 byte blocks): 1448761 operations in 10 seconds (11868250112 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:02 +07:00
|
|
|
state[12] += 4;
|
|
|
|
}
|
2018-11-17 08:26:21 +07:00
|
|
|
if (bytes > CHACHA_BLOCK_SIZE) {
|
2018-12-05 13:20:03 +07:00
|
|
|
chacha_4block_xor_ssse3(state, dst, src, bytes, nrounds);
|
|
|
|
state[12] += chacha_advance(bytes, 4);
|
2018-11-11 16:36:28 +07:00
|
|
|
return;
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
}
|
|
|
|
if (bytes) {
|
2018-12-05 13:20:03 +07:00
|
|
|
chacha_block_xor_ssse3(state, dst, src, bytes, nrounds);
|
2018-11-11 16:36:28 +07:00
|
|
|
state[12]++;
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2019-11-08 19:22:10 +07:00
|
|
|
void hchacha_block_arch(const u32 *state, u32 *stream, int nrounds)
|
|
|
|
{
|
|
|
|
state = PTR_ALIGN(state, CHACHA_STATE_ALIGN);
|
|
|
|
|
|
|
|
if (!static_branch_likely(&chacha_use_simd) || !crypto_simd_usable()) {
|
|
|
|
hchacha_block_generic(state, stream, nrounds);
|
|
|
|
} else {
|
|
|
|
kernel_fpu_begin();
|
|
|
|
hchacha_block_ssse3(state, stream, nrounds);
|
|
|
|
kernel_fpu_end();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(hchacha_block_arch);
|
|
|
|
|
|
|
|
void chacha_init_arch(u32 *state, const u32 *key, const u8 *iv)
|
|
|
|
{
|
|
|
|
state = PTR_ALIGN(state, CHACHA_STATE_ALIGN);
|
|
|
|
|
|
|
|
chacha_init_generic(state, key, iv);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(chacha_init_arch);
|
|
|
|
|
|
|
|
void chacha_crypt_arch(u32 *state, u8 *dst, const u8 *src, unsigned int bytes,
|
|
|
|
int nrounds)
|
|
|
|
{
|
|
|
|
state = PTR_ALIGN(state, CHACHA_STATE_ALIGN);
|
|
|
|
|
|
|
|
if (!static_branch_likely(&chacha_use_simd) || !crypto_simd_usable() ||
|
|
|
|
bytes <= CHACHA_BLOCK_SIZE)
|
|
|
|
return chacha_crypt_generic(state, dst, src, bytes, nrounds);
|
|
|
|
|
crypto: arch/lib - limit simd usage to 4k chunks
The initial Zinc patchset, after some mailing list discussion, contained
code to ensure that kernel_fpu_enable would not be kept on for more than
a 4k chunk, since it disables preemption. The choice of 4k isn't totally
scientific, but it's not a bad guess either, and it's what's used in
both the x86 poly1305, blake2s, and nhpoly1305 code already (in the form
of PAGE_SIZE, which this commit corrects to be explicitly 4k for the
former two).
Ard did some back of the envelope calculations and found that
at 5 cycles/byte (overestimate) on a 1ghz processor (pretty slow), 4k
means we have a maximum preemption disabling of 20us, which Sebastian
confirmed was probably a good limit.
Unfortunately the chunking appears to have been left out of the final
patchset that added the glue code. So, this commit adds it back in.
Fixes: 84e03fa39fbe ("crypto: x86/chacha - expose SIMD ChaCha routine as library function")
Fixes: b3aad5bad26a ("crypto: arm64/chacha - expose arm64 ChaCha routine as library function")
Fixes: a44a3430d71b ("crypto: arm/chacha - expose ARM ChaCha routine as library function")
Fixes: d7d7b8535662 ("crypto: x86/poly1305 - wire up faster implementations for kernel")
Fixes: f569ca164751 ("crypto: arm64/poly1305 - incorporate OpenSSL/CRYPTOGAMS NEON implementation")
Fixes: a6b803b3ddc7 ("crypto: arm/poly1305 - incorporate OpenSSL/CRYPTOGAMS NEON implementation")
Fixes: ed0356eda153 ("crypto: blake2s - x86_64 SIMD implementation")
Cc: Eric Biggers <ebiggers@google.com>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: stable@vger.kernel.org
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2020-04-23 06:18:53 +07:00
|
|
|
do {
|
|
|
|
unsigned int todo = min_t(unsigned int, bytes, SZ_4K);
|
|
|
|
|
|
|
|
kernel_fpu_begin();
|
|
|
|
chacha_dosimd(state, dst, src, todo, nrounds);
|
|
|
|
kernel_fpu_end();
|
|
|
|
|
|
|
|
bytes -= todo;
|
|
|
|
src += todo;
|
|
|
|
dst += todo;
|
|
|
|
} while (bytes);
|
2019-11-08 19:22:10 +07:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(chacha_crypt_arch);
|
|
|
|
|
2019-11-08 19:22:09 +07:00
|
|
|
static int chacha_simd_stream_xor(struct skcipher_request *req,
|
2019-06-03 12:47:14 +07:00
|
|
|
const struct chacha_ctx *ctx, const u8 *iv)
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
{
|
2017-01-11 19:28:06 +07:00
|
|
|
u32 *state, state_buf[16 + 2] __aligned(8);
|
2019-11-08 19:22:09 +07:00
|
|
|
struct skcipher_walk walk;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
err = skcipher_walk_virt(&walk, req, false);
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
|
2018-12-05 13:20:03 +07:00
|
|
|
BUILD_BUG_ON(CHACHA_STATE_ALIGN != 16);
|
|
|
|
state = PTR_ALIGN(state_buf + 0, CHACHA_STATE_ALIGN);
|
2017-01-11 19:28:06 +07:00
|
|
|
|
2019-11-08 19:22:09 +07:00
|
|
|
chacha_init_generic(state, ctx->key, iv);
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
|
2019-11-08 19:22:09 +07:00
|
|
|
while (walk.nbytes > 0) {
|
|
|
|
unsigned int nbytes = walk.nbytes;
|
2018-11-11 16:36:28 +07:00
|
|
|
|
2019-11-08 19:22:09 +07:00
|
|
|
if (nbytes < walk.total)
|
|
|
|
nbytes = round_down(nbytes, walk.stride);
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
|
2019-11-08 19:22:10 +07:00
|
|
|
if (!static_branch_likely(&chacha_use_simd) ||
|
|
|
|
!crypto_simd_usable()) {
|
2019-11-08 19:22:09 +07:00
|
|
|
chacha_crypt_generic(state, walk.dst.virt.addr,
|
|
|
|
walk.src.virt.addr, nbytes,
|
|
|
|
ctx->nrounds);
|
|
|
|
} else {
|
2018-12-05 13:20:05 +07:00
|
|
|
kernel_fpu_begin();
|
2019-11-08 19:22:09 +07:00
|
|
|
chacha_dosimd(state, walk.dst.virt.addr,
|
|
|
|
walk.src.virt.addr, nbytes,
|
|
|
|
ctx->nrounds);
|
|
|
|
kernel_fpu_end();
|
2018-12-05 13:20:05 +07:00
|
|
|
}
|
2019-11-08 19:22:09 +07:00
|
|
|
err = skcipher_walk_done(&walk, walk.nbytes - nbytes);
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
}
|
|
|
|
|
2018-12-05 13:20:02 +07:00
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2018-12-05 13:20:03 +07:00
|
|
|
static int chacha_simd(struct skcipher_request *req)
|
2018-12-05 13:20:02 +07:00
|
|
|
{
|
|
|
|
struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
|
|
|
|
struct chacha_ctx *ctx = crypto_skcipher_ctx(tfm);
|
|
|
|
|
2019-11-08 19:22:09 +07:00
|
|
|
return chacha_simd_stream_xor(req, ctx, req->iv);
|
2018-12-05 13:20:02 +07:00
|
|
|
}
|
|
|
|
|
2018-12-05 13:20:03 +07:00
|
|
|
static int xchacha_simd(struct skcipher_request *req)
|
2018-12-05 13:20:02 +07:00
|
|
|
{
|
|
|
|
struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
|
|
|
|
struct chacha_ctx *ctx = crypto_skcipher_ctx(tfm);
|
|
|
|
u32 *state, state_buf[16 + 2] __aligned(8);
|
2019-11-08 19:22:09 +07:00
|
|
|
struct chacha_ctx subctx;
|
2018-12-05 13:20:02 +07:00
|
|
|
u8 real_iv[16];
|
2018-12-16 03:40:17 +07:00
|
|
|
|
2018-12-05 13:20:03 +07:00
|
|
|
BUILD_BUG_ON(CHACHA_STATE_ALIGN != 16);
|
|
|
|
state = PTR_ALIGN(state_buf + 0, CHACHA_STATE_ALIGN);
|
2019-11-08 19:22:09 +07:00
|
|
|
chacha_init_generic(state, ctx->key, req->iv);
|
|
|
|
|
|
|
|
if (req->cryptlen > CHACHA_BLOCK_SIZE && crypto_simd_usable()) {
|
|
|
|
kernel_fpu_begin();
|
|
|
|
hchacha_block_ssse3(state, subctx.key, ctx->nrounds);
|
|
|
|
kernel_fpu_end();
|
|
|
|
} else {
|
|
|
|
hchacha_block_generic(state, subctx.key, ctx->nrounds);
|
|
|
|
}
|
2018-12-05 13:20:03 +07:00
|
|
|
subctx.nrounds = ctx->nrounds;
|
2018-12-05 13:20:02 +07:00
|
|
|
|
|
|
|
memcpy(&real_iv[0], req->iv + 24, 8);
|
|
|
|
memcpy(&real_iv[8], req->iv + 16, 8);
|
2019-11-08 19:22:09 +07:00
|
|
|
return chacha_simd_stream_xor(req, &subctx, real_iv);
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
}
|
|
|
|
|
2018-12-05 13:20:02 +07:00
|
|
|
static struct skcipher_alg algs[] = {
|
|
|
|
{
|
|
|
|
.base.cra_name = "chacha20",
|
|
|
|
.base.cra_driver_name = "chacha20-simd",
|
|
|
|
.base.cra_priority = 300,
|
|
|
|
.base.cra_blocksize = 1,
|
|
|
|
.base.cra_ctxsize = sizeof(struct chacha_ctx),
|
|
|
|
.base.cra_module = THIS_MODULE,
|
|
|
|
|
|
|
|
.min_keysize = CHACHA_KEY_SIZE,
|
|
|
|
.max_keysize = CHACHA_KEY_SIZE,
|
|
|
|
.ivsize = CHACHA_IV_SIZE,
|
|
|
|
.chunksize = CHACHA_BLOCK_SIZE,
|
2019-11-08 19:22:09 +07:00
|
|
|
.setkey = chacha20_setkey,
|
2018-12-05 13:20:03 +07:00
|
|
|
.encrypt = chacha_simd,
|
|
|
|
.decrypt = chacha_simd,
|
2018-12-05 13:20:02 +07:00
|
|
|
}, {
|
|
|
|
.base.cra_name = "xchacha20",
|
|
|
|
.base.cra_driver_name = "xchacha20-simd",
|
|
|
|
.base.cra_priority = 300,
|
|
|
|
.base.cra_blocksize = 1,
|
|
|
|
.base.cra_ctxsize = sizeof(struct chacha_ctx),
|
|
|
|
.base.cra_module = THIS_MODULE,
|
|
|
|
|
|
|
|
.min_keysize = CHACHA_KEY_SIZE,
|
|
|
|
.max_keysize = CHACHA_KEY_SIZE,
|
|
|
|
.ivsize = XCHACHA_IV_SIZE,
|
|
|
|
.chunksize = CHACHA_BLOCK_SIZE,
|
2019-11-08 19:22:09 +07:00
|
|
|
.setkey = chacha20_setkey,
|
2018-12-05 13:20:03 +07:00
|
|
|
.encrypt = xchacha_simd,
|
|
|
|
.decrypt = xchacha_simd,
|
2018-12-05 13:20:04 +07:00
|
|
|
}, {
|
|
|
|
.base.cra_name = "xchacha12",
|
|
|
|
.base.cra_driver_name = "xchacha12-simd",
|
|
|
|
.base.cra_priority = 300,
|
|
|
|
.base.cra_blocksize = 1,
|
|
|
|
.base.cra_ctxsize = sizeof(struct chacha_ctx),
|
|
|
|
.base.cra_module = THIS_MODULE,
|
|
|
|
|
|
|
|
.min_keysize = CHACHA_KEY_SIZE,
|
|
|
|
.max_keysize = CHACHA_KEY_SIZE,
|
|
|
|
.ivsize = XCHACHA_IV_SIZE,
|
|
|
|
.chunksize = CHACHA_BLOCK_SIZE,
|
2019-11-08 19:22:09 +07:00
|
|
|
.setkey = chacha12_setkey,
|
2018-12-05 13:20:04 +07:00
|
|
|
.encrypt = xchacha_simd,
|
|
|
|
.decrypt = xchacha_simd,
|
2018-12-05 13:20:02 +07:00
|
|
|
},
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
};
|
|
|
|
|
2018-12-05 13:20:03 +07:00
|
|
|
static int __init chacha_simd_mod_init(void)
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
{
|
2015-12-07 16:39:41 +07:00
|
|
|
if (!boot_cpu_has(X86_FEATURE_SSSE3))
|
2019-11-08 19:22:10 +07:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
static_branch_enable(&chacha_use_simd);
|
|
|
|
|
2020-03-27 03:26:00 +07:00
|
|
|
if (boot_cpu_has(X86_FEATURE_AVX) &&
|
2019-11-08 19:22:10 +07:00
|
|
|
boot_cpu_has(X86_FEATURE_AVX2) &&
|
|
|
|
cpu_has_xfeatures(XFEATURE_MASK_SSE | XFEATURE_MASK_YMM, NULL)) {
|
|
|
|
static_branch_enable(&chacha_use_avx2);
|
|
|
|
|
|
|
|
if (IS_ENABLED(CONFIG_AS_AVX512) &&
|
|
|
|
boot_cpu_has(X86_FEATURE_AVX512VL) &&
|
|
|
|
boot_cpu_has(X86_FEATURE_AVX512BW)) /* kmovq */
|
|
|
|
static_branch_enable(&chacha_use_avx512vl);
|
|
|
|
}
|
2019-11-25 17:31:12 +07:00
|
|
|
return IS_REACHABLE(CONFIG_CRYPTO_SKCIPHER) ?
|
|
|
|
crypto_register_skciphers(algs, ARRAY_SIZE(algs)) : 0;
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
}
|
|
|
|
|
2018-12-05 13:20:03 +07:00
|
|
|
static void __exit chacha_simd_mod_fini(void)
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
{
|
2019-11-25 17:31:12 +07:00
|
|
|
if (IS_REACHABLE(CONFIG_CRYPTO_SKCIPHER) && boot_cpu_has(X86_FEATURE_SSSE3))
|
2019-11-18 14:21:58 +07:00
|
|
|
crypto_unregister_skciphers(algs, ARRAY_SIZE(algs));
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
}
|
|
|
|
|
2018-12-05 13:20:03 +07:00
|
|
|
module_init(chacha_simd_mod_init);
|
|
|
|
module_exit(chacha_simd_mod_fini);
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
|
|
|
|
MODULE_LICENSE("GPL");
|
|
|
|
MODULE_AUTHOR("Martin Willi <martin@strongswan.org>");
|
2018-12-05 13:20:03 +07:00
|
|
|
MODULE_DESCRIPTION("ChaCha and XChaCha stream ciphers (x64 SIMD accelerated)");
|
crypto: chacha20 - Add a SSSE3 SIMD variant for x86_64
Implements an x86_64 assembler driver for the ChaCha20 stream cipher. This
single block variant works on a single state matrix using SSE instructions.
It requires SSSE3 due the use of pshufb for efficient 8/16-bit rotate
operations.
For large messages, throughput increases by ~65% compared to
chacha20-generic:
testing speed of chacha20 (chacha20-generic) encryption
test 0 (256 bit key, 16 byte blocks): 45089207 operations in 10 seconds (721427312 bytes)
test 1 (256 bit key, 64 byte blocks): 43839521 operations in 10 seconds (2805729344 bytes)
test 2 (256 bit key, 256 byte blocks): 12702056 operations in 10 seconds (3251726336 bytes)
test 3 (256 bit key, 1024 byte blocks): 3371173 operations in 10 seconds (3452081152 bytes)
test 4 (256 bit key, 8192 byte blocks): 422468 operations in 10 seconds (3460857856 bytes)
testing speed of chacha20 (chacha20-simd) encryption
test 0 (256 bit key, 16 byte blocks): 43141886 operations in 10 seconds (690270176 bytes)
test 1 (256 bit key, 64 byte blocks): 46845874 operations in 10 seconds (2998135936 bytes)
test 2 (256 bit key, 256 byte blocks): 18458512 operations in 10 seconds (4725379072 bytes)
test 3 (256 bit key, 1024 byte blocks): 5360533 operations in 10 seconds (5489185792 bytes)
test 4 (256 bit key, 8192 byte blocks): 692846 operations in 10 seconds (5675794432 bytes)
Benchmark results from a Core i5-4670T.
Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-07-17 00:14:01 +07:00
|
|
|
MODULE_ALIAS_CRYPTO("chacha20");
|
|
|
|
MODULE_ALIAS_CRYPTO("chacha20-simd");
|
2018-12-05 13:20:02 +07:00
|
|
|
MODULE_ALIAS_CRYPTO("xchacha20");
|
|
|
|
MODULE_ALIAS_CRYPTO("xchacha20-simd");
|
2018-12-05 13:20:04 +07:00
|
|
|
MODULE_ALIAS_CRYPTO("xchacha12");
|
|
|
|
MODULE_ALIAS_CRYPTO("xchacha12-simd");
|