iv/linux
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
0f961f9f67
linux/arch/x86/crypto/chacha20_glue.c

209 lines
5.8 KiB
C
Raw Normal View History

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>
2015-07-16 20:14:01 +03:00
/*
* ChaCha20 256-bit cipher algorithm, RFC7539, SIMD glue code
*
* Copyright (C) 2015 Martin Willi
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*/
#include <crypto/algapi.h>
crypto: chacha20-generic - refactor to allow varying number of rounds In preparation for adding XChaCha12 support, rename/refactor chacha20-generic to support different numbers of rounds. The justification for needing XChaCha12 support is explained in more detail in the patch "crypto: chacha - add XChaCha12 support". The only difference between ChaCha{8,12,20} are the number of rounds itself; all other parts of the algorithm are the same. Therefore, remove the "20" from all definitions, structures, functions, files, etc. that will be shared by all ChaCha versions. Also make ->setkey() store the round count in the chacha_ctx (previously chacha20_ctx). The generic code then passes the round count through to chacha_block(). There will be a ->setkey() function for each explicitly allowed round count; the encrypt/decrypt functions will be the same. I decided not to do it the opposite way (same ->setkey() function for all round counts, with different encrypt/decrypt functions) because that would have required more boilerplate code in architecture-specific implementations of ChaCha and XChaCha. Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by: Martin Willi <martin@strongswan.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-17 04:26:21 +03:00
#include <crypto/chacha.h>
crypto: chacha20 - convert generic and x86 versions to skcipher This converts the ChaCha20 code from a blkcipher to a skcipher, which is now the preferred way to implement symmetric block and stream ciphers. This ports the generic and x86 versions at the same time because the latter reuses routines of the former. Note that the skcipher_walk() API guarantees that all presented blocks except the final one are a multiple of the chunk size, so we can simplify the encrypt() routine somewhat. Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-12-09 17:33:51 +03:00
#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>
2015-07-16 20:14:01 +03:00
#include <linux/kernel.h>
#include <linux/module.h>
#include <asm/fpu/api.h>
#include <asm/simd.h>
#define CHACHA20_STATE_ALIGN 16
crypto: x86/chacha20 - Support partial lengths in 1-block SSSE3 variant Add a length argument to the single block function for SSSE3, so the block function may XOR only a partial length of the full block. Given that the setup code is rather cheap, the function does not process more than one block; this allows us to keep the block function selection in the C glue code. The required branching does not negatively affect performance for full block sizes. The partial XORing uses simple "rep movsb" to copy the data before and after doing XOR in SSE. This is rather efficient on modern processors; movsw can be slightly faster, but the additional complexity is probably not worth it. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:25 +03:00
asmlinkage void chacha20_block_xor_ssse3(u32 *state, u8 *dst, const u8 *src,
unsigned int len);
crypto: x86/chacha20 - Support partial lengths in 4-block SSSE3 variant Add a length argument to the quad block function for SSSE3, so the block function may XOR only a partial length of four blocks. As we already have the stack set up, the partial XORing does not need to. This gives a slightly different function trailer, so we keep that separate from the 1-block function. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:26 +03:00
asmlinkage void chacha20_4block_xor_ssse3(u32 *state, u8 *dst, const u8 *src,
unsigned int len);
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>
2015-07-16 20:14:03 +03:00
#ifdef CONFIG_AS_AVX2
crypto: x86/chacha20 - Add a 2-block AVX2 variant This variant uses the same principle as the single block SSSE3 variant by shuffling the state matrix after each round. With the wider AVX registers, we can do two blocks in parallel, though. This function can increase performance and efficiency significantly for lengths that would otherwise require a 4-block function. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:29 +03:00
asmlinkage void chacha20_2block_xor_avx2(u32 *state, u8 *dst, const u8 *src,
unsigned int len);
crypto: x86/chacha20 - Add a 4-block AVX2 variant This variant builds upon the idea of the 2-block AVX2 variant that shuffles words after each round. The shuffling has a rather high latency, so the arithmetic units are not optimally used. Given that we have plenty of registers in AVX, this version parallelizes the 2-block variant to do four blocks. While the first two blocks are shuffling, the CPU can do the XORing on the second two blocks and vice-versa, which makes this version much faster than the SSSE3 variant for four blocks. The latter is now mostly for systems that do not have AVX2, but there it is the work-horse, so we keep it in place. The partial XORing function trailer is very similar to the AVX2 2-block variant. While it could be shared, that code segment is rather short; profiling is also easier with the trailer integrated, so we keep it per function. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:30 +03:00
asmlinkage void chacha20_4block_xor_avx2(u32 *state, u8 *dst, const u8 *src,
unsigned int len);
crypto: x86/chacha20 - Support partial lengths in 8-block AVX2 variant Add a length argument to the eight block function for AVX2, so the block function may XOR only a partial length of eight blocks. To avoid unnecessary operations, we integrate XORing of the first four blocks in the final lane interleaving; this also avoids some work in the partial lengths path. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:27 +03:00
asmlinkage void chacha20_8block_xor_avx2(u32 *state, u8 *dst, const u8 *src,
unsigned int len);
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>
2015-07-16 20:14:03 +03:00
static bool chacha20_use_avx2;
crypto: x86/chacha20 - Add a 8-block AVX-512VL variant This variant is similar to the AVX2 version, but benefits from the AVX-512 rotate instructions and the additional registers, so it can operate without any data on the stack. It uses ymm registers only to avoid the massive core throttling on Skylake-X platforms. Nontheless does it bring a ~30% speed improvement compared to the AVX2 variant for random encryption lengths. The AVX2 version uses "rep movsb" for partial block XORing via the stack. With AVX-512, the new "vmovdqu8" can do this much more efficiently. The associated "kmov" instructions to work with dynamic masks is not part of the AVX-512VL instruction set, hence we depend on AVX-512BW as well. Given that the major AVX-512VL architectures provide AVX-512BW and this extension does not affect core clocking, this seems to be no problem at least for now. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-20 19:30:48 +03:00
#ifdef CONFIG_AS_AVX512
crypto: x86/chacha20 - Add a 2-block AVX-512VL variant This version uses the same principle as the AVX2 version. It benefits from the AVX-512VL rotate instructions and the more efficient partial block handling using "vmovdqu8", resulting in a speedup of ~20%. Unlike the AVX2 version, it is faster than the single block SSSE3 version to process a single block. Hence we engage that function for (partial) single block lengths as well. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-20 19:30:49 +03:00
asmlinkage void chacha20_2block_xor_avx512vl(u32 *state, u8 *dst, const u8 *src,
unsigned int len);
crypto: x86/chacha20 - Add a 4-block AVX-512VL variant This version uses the same principle as the AVX2 version by scheduling the operations for two block pairs in parallel. It benefits from the AVX-512VL rotate instructions and the more efficient partial block handling using "vmovdqu8", resulting in a speedup of the raw block function of ~20%. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-20 19:30:50 +03:00
asmlinkage void chacha20_4block_xor_avx512vl(u32 *state, u8 *dst, const u8 *src,
unsigned int len);
crypto: x86/chacha20 - Add a 8-block AVX-512VL variant This variant is similar to the AVX2 version, but benefits from the AVX-512 rotate instructions and the additional registers, so it can operate without any data on the stack. It uses ymm registers only to avoid the massive core throttling on Skylake-X platforms. Nontheless does it bring a ~30% speed improvement compared to the AVX2 variant for random encryption lengths. The AVX2 version uses "rep movsb" for partial block XORing via the stack. With AVX-512, the new "vmovdqu8" can do this much more efficiently. The associated "kmov" instructions to work with dynamic masks is not part of the AVX-512VL instruction set, hence we depend on AVX-512BW as well. Given that the major AVX-512VL architectures provide AVX-512BW and this extension does not affect core clocking, this seems to be no problem at least for now. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-20 19:30:48 +03:00
asmlinkage void chacha20_8block_xor_avx512vl(u32 *state, u8 *dst, const u8 *src,
unsigned int len);
static bool chacha20_use_avx512vl;
#endif
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>
2015-07-16 20:14:03 +03:00
#endif
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>
2015-07-16 20:14:01 +03:00
crypto: x86/chacha20 - Use larger block functions more aggressively Now that all block functions support partial lengths, engage the wider block sizes more aggressively. This prevents using smaller block functions multiple times, where the next larger block function would have been faster. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:28 +03:00
static unsigned int chacha20_advance(unsigned int len, unsigned int maxblocks)
{
crypto: chacha20-generic - refactor to allow varying number of rounds In preparation for adding XChaCha12 support, rename/refactor chacha20-generic to support different numbers of rounds. The justification for needing XChaCha12 support is explained in more detail in the patch "crypto: chacha - add XChaCha12 support". The only difference between ChaCha{8,12,20} are the number of rounds itself; all other parts of the algorithm are the same. Therefore, remove the "20" from all definitions, structures, functions, files, etc. that will be shared by all ChaCha versions. Also make ->setkey() store the round count in the chacha_ctx (previously chacha20_ctx). The generic code then passes the round count through to chacha_block(). There will be a ->setkey() function for each explicitly allowed round count; the encrypt/decrypt functions will be the same. I decided not to do it the opposite way (same ->setkey() function for all round counts, with different encrypt/decrypt functions) because that would have required more boilerplate code in architecture-specific implementations of ChaCha and XChaCha. Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by: Martin Willi <martin@strongswan.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-17 04:26:21 +03:00
len = min(len, maxblocks * CHACHA_BLOCK_SIZE);
return round_up(len, CHACHA_BLOCK_SIZE) / CHACHA_BLOCK_SIZE;
crypto: x86/chacha20 - Use larger block functions more aggressively Now that all block functions support partial lengths, engage the wider block sizes more aggressively. This prevents using smaller block functions multiple times, where the next larger block function would have been faster. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:28 +03:00
}
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>
2015-07-16 20:14:01 +03:00
static void chacha20_dosimd(u32 *state, u8 *dst, const u8 *src,
unsigned int bytes)
{
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>
2015-07-16 20:14:03 +03:00
#ifdef CONFIG_AS_AVX2
crypto: x86/chacha20 - Add a 8-block AVX-512VL variant This variant is similar to the AVX2 version, but benefits from the AVX-512 rotate instructions and the additional registers, so it can operate without any data on the stack. It uses ymm registers only to avoid the massive core throttling on Skylake-X platforms. Nontheless does it bring a ~30% speed improvement compared to the AVX2 variant for random encryption lengths. The AVX2 version uses "rep movsb" for partial block XORing via the stack. With AVX-512, the new "vmovdqu8" can do this much more efficiently. The associated "kmov" instructions to work with dynamic masks is not part of the AVX-512VL instruction set, hence we depend on AVX-512BW as well. Given that the major AVX-512VL architectures provide AVX-512BW and this extension does not affect core clocking, this seems to be no problem at least for now. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-20 19:30:48 +03:00
#ifdef CONFIG_AS_AVX512
if (chacha20_use_avx512vl) {
while (bytes >= CHACHA_BLOCK_SIZE * 8) {
chacha20_8block_xor_avx512vl(state, dst, src, bytes);
bytes -= CHACHA_BLOCK_SIZE * 8;
src += CHACHA_BLOCK_SIZE * 8;
dst += CHACHA_BLOCK_SIZE * 8;
state[12] += 8;
}
if (bytes > CHACHA_BLOCK_SIZE * 4) {
chacha20_8block_xor_avx512vl(state, dst, src, bytes);
state[12] += chacha20_advance(bytes, 8);
return;
}
crypto: x86/chacha20 - Add a 4-block AVX-512VL variant This version uses the same principle as the AVX2 version by scheduling the operations for two block pairs in parallel. It benefits from the AVX-512VL rotate instructions and the more efficient partial block handling using "vmovdqu8", resulting in a speedup of the raw block function of ~20%. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-20 19:30:50 +03:00
if (bytes > CHACHA_BLOCK_SIZE * 2) {
chacha20_4block_xor_avx512vl(state, dst, src, bytes);
state[12] += chacha20_advance(bytes, 4);
return;
}
crypto: x86/chacha20 - Add a 2-block AVX-512VL variant This version uses the same principle as the AVX2 version. It benefits from the AVX-512VL rotate instructions and the more efficient partial block handling using "vmovdqu8", resulting in a speedup of ~20%. Unlike the AVX2 version, it is faster than the single block SSSE3 version to process a single block. Hence we engage that function for (partial) single block lengths as well. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-20 19:30:49 +03:00
if (bytes) {
chacha20_2block_xor_avx512vl(state, dst, src, bytes);
state[12] += chacha20_advance(bytes, 2);
return;
}
crypto: x86/chacha20 - Add a 8-block AVX-512VL variant This variant is similar to the AVX2 version, but benefits from the AVX-512 rotate instructions and the additional registers, so it can operate without any data on the stack. It uses ymm registers only to avoid the massive core throttling on Skylake-X platforms. Nontheless does it bring a ~30% speed improvement compared to the AVX2 variant for random encryption lengths. The AVX2 version uses "rep movsb" for partial block XORing via the stack. With AVX-512, the new "vmovdqu8" can do this much more efficiently. The associated "kmov" instructions to work with dynamic masks is not part of the AVX-512VL instruction set, hence we depend on AVX-512BW as well. Given that the major AVX-512VL architectures provide AVX-512BW and this extension does not affect core clocking, this seems to be no problem at least for now. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-20 19:30:48 +03:00
}
#endif
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>
2015-07-16 20:14:03 +03:00
if (chacha20_use_avx2) {
crypto: chacha20-generic - refactor to allow varying number of rounds In preparation for adding XChaCha12 support, rename/refactor chacha20-generic to support different numbers of rounds. The justification for needing XChaCha12 support is explained in more detail in the patch "crypto: chacha - add XChaCha12 support". The only difference between ChaCha{8,12,20} are the number of rounds itself; all other parts of the algorithm are the same. Therefore, remove the "20" from all definitions, structures, functions, files, etc. that will be shared by all ChaCha versions. Also make ->setkey() store the round count in the chacha_ctx (previously chacha20_ctx). The generic code then passes the round count through to chacha_block(). There will be a ->setkey() function for each explicitly allowed round count; the encrypt/decrypt functions will be the same. I decided not to do it the opposite way (same ->setkey() function for all round counts, with different encrypt/decrypt functions) because that would have required more boilerplate code in architecture-specific implementations of ChaCha and XChaCha. Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by: Martin Willi <martin@strongswan.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-17 04:26:21 +03:00
while (bytes >= CHACHA_BLOCK_SIZE * 8) {
crypto: x86/chacha20 - Support partial lengths in 8-block AVX2 variant Add a length argument to the eight block function for AVX2, so the block function may XOR only a partial length of eight blocks. To avoid unnecessary operations, we integrate XORing of the first four blocks in the final lane interleaving; this also avoids some work in the partial lengths path. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:27 +03:00
chacha20_8block_xor_avx2(state, dst, src, bytes);
crypto: chacha20-generic - refactor to allow varying number of rounds In preparation for adding XChaCha12 support, rename/refactor chacha20-generic to support different numbers of rounds. The justification for needing XChaCha12 support is explained in more detail in the patch "crypto: chacha - add XChaCha12 support". The only difference between ChaCha{8,12,20} are the number of rounds itself; all other parts of the algorithm are the same. Therefore, remove the "20" from all definitions, structures, functions, files, etc. that will be shared by all ChaCha versions. Also make ->setkey() store the round count in the chacha_ctx (previously chacha20_ctx). The generic code then passes the round count through to chacha_block(). There will be a ->setkey() function for each explicitly allowed round count; the encrypt/decrypt functions will be the same. I decided not to do it the opposite way (same ->setkey() function for all round counts, with different encrypt/decrypt functions) because that would have required more boilerplate code in architecture-specific implementations of ChaCha and XChaCha. Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by: Martin Willi <martin@strongswan.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-17 04:26:21 +03:00
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>
2015-07-16 20:14:03 +03:00
state[12] += 8;
}
crypto: chacha20-generic - refactor to allow varying number of rounds In preparation for adding XChaCha12 support, rename/refactor chacha20-generic to support different numbers of rounds. The justification for needing XChaCha12 support is explained in more detail in the patch "crypto: chacha - add XChaCha12 support". The only difference between ChaCha{8,12,20} are the number of rounds itself; all other parts of the algorithm are the same. Therefore, remove the "20" from all definitions, structures, functions, files, etc. that will be shared by all ChaCha versions. Also make ->setkey() store the round count in the chacha_ctx (previously chacha20_ctx). The generic code then passes the round count through to chacha_block(). There will be a ->setkey() function for each explicitly allowed round count; the encrypt/decrypt functions will be the same. I decided not to do it the opposite way (same ->setkey() function for all round counts, with different encrypt/decrypt functions) because that would have required more boilerplate code in architecture-specific implementations of ChaCha and XChaCha. Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by: Martin Willi <martin@strongswan.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-17 04:26:21 +03:00
if (bytes > CHACHA_BLOCK_SIZE * 4) {
crypto: x86/chacha20 - Use larger block functions more aggressively Now that all block functions support partial lengths, engage the wider block sizes more aggressively. This prevents using smaller block functions multiple times, where the next larger block function would have been faster. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:28 +03:00
chacha20_8block_xor_avx2(state, dst, src, bytes);
state[12] += chacha20_advance(bytes, 8);
return;
}
crypto: chacha20-generic - refactor to allow varying number of rounds In preparation for adding XChaCha12 support, rename/refactor chacha20-generic to support different numbers of rounds. The justification for needing XChaCha12 support is explained in more detail in the patch "crypto: chacha - add XChaCha12 support". The only difference between ChaCha{8,12,20} are the number of rounds itself; all other parts of the algorithm are the same. Therefore, remove the "20" from all definitions, structures, functions, files, etc. that will be shared by all ChaCha versions. Also make ->setkey() store the round count in the chacha_ctx (previously chacha20_ctx). The generic code then passes the round count through to chacha_block(). There will be a ->setkey() function for each explicitly allowed round count; the encrypt/decrypt functions will be the same. I decided not to do it the opposite way (same ->setkey() function for all round counts, with different encrypt/decrypt functions) because that would have required more boilerplate code in architecture-specific implementations of ChaCha and XChaCha. Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by: Martin Willi <martin@strongswan.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-17 04:26:21 +03:00
if (bytes > CHACHA_BLOCK_SIZE * 2) {
crypto: x86/chacha20 - Add a 4-block AVX2 variant This variant builds upon the idea of the 2-block AVX2 variant that shuffles words after each round. The shuffling has a rather high latency, so the arithmetic units are not optimally used. Given that we have plenty of registers in AVX, this version parallelizes the 2-block variant to do four blocks. While the first two blocks are shuffling, the CPU can do the XORing on the second two blocks and vice-versa, which makes this version much faster than the SSSE3 variant for four blocks. The latter is now mostly for systems that do not have AVX2, but there it is the work-horse, so we keep it in place. The partial XORing function trailer is very similar to the AVX2 2-block variant. While it could be shared, that code segment is rather short; profiling is also easier with the trailer integrated, so we keep it per function. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:30 +03:00
chacha20_4block_xor_avx2(state, dst, src, bytes);
state[12] += chacha20_advance(bytes, 4);
return;
}
crypto: chacha20-generic - refactor to allow varying number of rounds In preparation for adding XChaCha12 support, rename/refactor chacha20-generic to support different numbers of rounds. The justification for needing XChaCha12 support is explained in more detail in the patch "crypto: chacha - add XChaCha12 support". The only difference between ChaCha{8,12,20} are the number of rounds itself; all other parts of the algorithm are the same. Therefore, remove the "20" from all definitions, structures, functions, files, etc. that will be shared by all ChaCha versions. Also make ->setkey() store the round count in the chacha_ctx (previously chacha20_ctx). The generic code then passes the round count through to chacha_block(). There will be a ->setkey() function for each explicitly allowed round count; the encrypt/decrypt functions will be the same. I decided not to do it the opposite way (same ->setkey() function for all round counts, with different encrypt/decrypt functions) because that would have required more boilerplate code in architecture-specific implementations of ChaCha and XChaCha. Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by: Martin Willi <martin@strongswan.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-17 04:26:21 +03:00
if (bytes > CHACHA_BLOCK_SIZE) {
crypto: x86/chacha20 - Add a 2-block AVX2 variant This variant uses the same principle as the single block SSSE3 variant by shuffling the state matrix after each round. With the wider AVX registers, we can do two blocks in parallel, though. This function can increase performance and efficiency significantly for lengths that would otherwise require a 4-block function. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:29 +03:00
chacha20_2block_xor_avx2(state, dst, src, bytes);
state[12] += chacha20_advance(bytes, 2);
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>
2015-07-16 20:14:03 +03:00
}
#endif
crypto: chacha20-generic - refactor to allow varying number of rounds In preparation for adding XChaCha12 support, rename/refactor chacha20-generic to support different numbers of rounds. The justification for needing XChaCha12 support is explained in more detail in the patch "crypto: chacha - add XChaCha12 support". The only difference between ChaCha{8,12,20} are the number of rounds itself; all other parts of the algorithm are the same. Therefore, remove the "20" from all definitions, structures, functions, files, etc. that will be shared by all ChaCha versions. Also make ->setkey() store the round count in the chacha_ctx (previously chacha20_ctx). The generic code then passes the round count through to chacha_block(). There will be a ->setkey() function for each explicitly allowed round count; the encrypt/decrypt functions will be the same. I decided not to do it the opposite way (same ->setkey() function for all round counts, with different encrypt/decrypt functions) because that would have required more boilerplate code in architecture-specific implementations of ChaCha and XChaCha. Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by: Martin Willi <martin@strongswan.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-17 04:26:21 +03:00
while (bytes >= CHACHA_BLOCK_SIZE * 4) {
crypto: x86/chacha20 - Support partial lengths in 4-block SSSE3 variant Add a length argument to the quad block function for SSSE3, so the block function may XOR only a partial length of four blocks. As we already have the stack set up, the partial XORing does not need to. This gives a slightly different function trailer, so we keep that separate from the 1-block function. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:26 +03:00
chacha20_4block_xor_ssse3(state, dst, src, bytes);
crypto: chacha20-generic - refactor to allow varying number of rounds In preparation for adding XChaCha12 support, rename/refactor chacha20-generic to support different numbers of rounds. The justification for needing XChaCha12 support is explained in more detail in the patch "crypto: chacha - add XChaCha12 support". The only difference between ChaCha{8,12,20} are the number of rounds itself; all other parts of the algorithm are the same. Therefore, remove the "20" from all definitions, structures, functions, files, etc. that will be shared by all ChaCha versions. Also make ->setkey() store the round count in the chacha_ctx (previously chacha20_ctx). The generic code then passes the round count through to chacha_block(). There will be a ->setkey() function for each explicitly allowed round count; the encrypt/decrypt functions will be the same. I decided not to do it the opposite way (same ->setkey() function for all round counts, with different encrypt/decrypt functions) because that would have required more boilerplate code in architecture-specific implementations of ChaCha and XChaCha. Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by: Martin Willi <martin@strongswan.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-17 04:26:21 +03:00
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>
2015-07-16 20:14:02 +03:00
state[12] += 4;
}
crypto: chacha20-generic - refactor to allow varying number of rounds In preparation for adding XChaCha12 support, rename/refactor chacha20-generic to support different numbers of rounds. The justification for needing XChaCha12 support is explained in more detail in the patch "crypto: chacha - add XChaCha12 support". The only difference between ChaCha{8,12,20} are the number of rounds itself; all other parts of the algorithm are the same. Therefore, remove the "20" from all definitions, structures, functions, files, etc. that will be shared by all ChaCha versions. Also make ->setkey() store the round count in the chacha_ctx (previously chacha20_ctx). The generic code then passes the round count through to chacha_block(). There will be a ->setkey() function for each explicitly allowed round count; the encrypt/decrypt functions will be the same. I decided not to do it the opposite way (same ->setkey() function for all round counts, with different encrypt/decrypt functions) because that would have required more boilerplate code in architecture-specific implementations of ChaCha and XChaCha. Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by: Martin Willi <martin@strongswan.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-17 04:26:21 +03:00
if (bytes > CHACHA_BLOCK_SIZE) {
crypto: x86/chacha20 - Use larger block functions more aggressively Now that all block functions support partial lengths, engage the wider block sizes more aggressively. This prevents using smaller block functions multiple times, where the next larger block function would have been faster. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:28 +03:00
chacha20_4block_xor_ssse3(state, dst, src, bytes);
state[12] += chacha20_advance(bytes, 4);
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>
2015-07-16 20:14:01 +03:00
}
if (bytes) {
crypto: x86/chacha20 - Support partial lengths in 1-block SSSE3 variant Add a length argument to the single block function for SSSE3, so the block function may XOR only a partial length of the full block. Given that the setup code is rather cheap, the function does not process more than one block; this allows us to keep the block function selection in the C glue code. The required branching does not negatively affect performance for full block sizes. The partial XORing uses simple "rep movsb" to copy the data before and after doing XOR in SSE. This is rather efficient on modern processors; movsw can be slightly faster, but the additional complexity is probably not worth it. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:25 +03:00
chacha20_block_xor_ssse3(state, dst, src, bytes);
crypto: x86/chacha20 - Use larger block functions more aggressively Now that all block functions support partial lengths, engage the wider block sizes more aggressively. This prevents using smaller block functions multiple times, where the next larger block function would have been faster. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:28 +03:00
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>
2015-07-16 20:14:01 +03:00
}
}
crypto: chacha20 - convert generic and x86 versions to skcipher This converts the ChaCha20 code from a blkcipher to a skcipher, which is now the preferred way to implement symmetric block and stream ciphers. This ports the generic and x86 versions at the same time because the latter reuses routines of the former. Note that the skcipher_walk() API guarantees that all presented blocks except the final one are a multiple of the chunk size, so we can simplify the encrypt() routine somewhat. Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-12-09 17:33:51 +03:00
static int chacha20_simd(struct skcipher_request *req)
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>
2015-07-16 20:14:01 +03:00
{
crypto: chacha20 - convert generic and x86 versions to skcipher This converts the ChaCha20 code from a blkcipher to a skcipher, which is now the preferred way to implement symmetric block and stream ciphers. This ports the generic and x86 versions at the same time because the latter reuses routines of the former. Note that the skcipher_walk() API guarantees that all presented blocks except the final one are a multiple of the chunk size, so we can simplify the encrypt() routine somewhat. Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-12-09 17:33:51 +03:00
struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
crypto: chacha20-generic - refactor to allow varying number of rounds In preparation for adding XChaCha12 support, rename/refactor chacha20-generic to support different numbers of rounds. The justification for needing XChaCha12 support is explained in more detail in the patch "crypto: chacha - add XChaCha12 support". The only difference between ChaCha{8,12,20} are the number of rounds itself; all other parts of the algorithm are the same. Therefore, remove the "20" from all definitions, structures, functions, files, etc. that will be shared by all ChaCha versions. Also make ->setkey() store the round count in the chacha_ctx (previously chacha20_ctx). The generic code then passes the round count through to chacha_block(). There will be a ->setkey() function for each explicitly allowed round count; the encrypt/decrypt functions will be the same. I decided not to do it the opposite way (same ->setkey() function for all round counts, with different encrypt/decrypt functions) because that would have required more boilerplate code in architecture-specific implementations of ChaCha and XChaCha. Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by: Martin Willi <martin@strongswan.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-17 04:26:21 +03:00
struct chacha_ctx *ctx = crypto_skcipher_ctx(tfm);
crypto: x86/chacha20 - Manually align stack buffer The kernel on x86-64 cannot use gcc attribute align to align to a 16-byte boundary. This patch reverts to the old way of aligning it by hand. Fixes: 9ae433bc79f9 ("crypto: chacha20 - convert generic and...") Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
2017-01-11 15:28:06 +03:00
u32 *state, state_buf[16 + 2] __aligned(8);
crypto: chacha20 - convert generic and x86 versions to skcipher This converts the ChaCha20 code from a blkcipher to a skcipher, which is now the preferred way to implement symmetric block and stream ciphers. This ports the generic and x86 versions at the same time because the latter reuses routines of the former. Note that the skcipher_walk() API guarantees that all presented blocks except the final one are a multiple of the chunk size, so we can simplify the encrypt() routine somewhat. Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-12-09 17:33:51 +03:00
struct skcipher_walk walk;
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>
2015-07-16 20:14:01 +03:00
int err;
crypto: x86/chacha20 - Manually align stack buffer The kernel on x86-64 cannot use gcc attribute align to align to a 16-byte boundary. This patch reverts to the old way of aligning it by hand. Fixes: 9ae433bc79f9 ("crypto: chacha20 - convert generic and...") Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
2017-01-11 15:28:06 +03:00
BUILD_BUG_ON(CHACHA20_STATE_ALIGN != 16);
state = PTR_ALIGN(state_buf + 0, CHACHA20_STATE_ALIGN);
crypto: chacha20-generic - refactor to allow varying number of rounds In preparation for adding XChaCha12 support, rename/refactor chacha20-generic to support different numbers of rounds. The justification for needing XChaCha12 support is explained in more detail in the patch "crypto: chacha - add XChaCha12 support". The only difference between ChaCha{8,12,20} are the number of rounds itself; all other parts of the algorithm are the same. Therefore, remove the "20" from all definitions, structures, functions, files, etc. that will be shared by all ChaCha versions. Also make ->setkey() store the round count in the chacha_ctx (previously chacha20_ctx). The generic code then passes the round count through to chacha_block(). There will be a ->setkey() function for each explicitly allowed round count; the encrypt/decrypt functions will be the same. I decided not to do it the opposite way (same ->setkey() function for all round counts, with different encrypt/decrypt functions) because that would have required more boilerplate code in architecture-specific implementations of ChaCha and XChaCha. Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by: Martin Willi <martin@strongswan.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-17 04:26:21 +03:00
if (req->cryptlen <= CHACHA_BLOCK_SIZE || !may_use_simd())
return crypto_chacha_crypt(req);
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>
2015-07-16 20:14:01 +03:00
crypto: chacha20 - convert generic and x86 versions to skcipher This converts the ChaCha20 code from a blkcipher to a skcipher, which is now the preferred way to implement symmetric block and stream ciphers. This ports the generic and x86 versions at the same time because the latter reuses routines of the former. Note that the skcipher_walk() API guarantees that all presented blocks except the final one are a multiple of the chunk size, so we can simplify the encrypt() routine somewhat. Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-12-09 17:33:51 +03:00
err = skcipher_walk_virt(&walk, req, true);
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>
2015-07-16 20:14:01 +03:00
crypto: chacha20-generic - refactor to allow varying number of rounds In preparation for adding XChaCha12 support, rename/refactor chacha20-generic to support different numbers of rounds. The justification for needing XChaCha12 support is explained in more detail in the patch "crypto: chacha - add XChaCha12 support". The only difference between ChaCha{8,12,20} are the number of rounds itself; all other parts of the algorithm are the same. Therefore, remove the "20" from all definitions, structures, functions, files, etc. that will be shared by all ChaCha versions. Also make ->setkey() store the round count in the chacha_ctx (previously chacha20_ctx). The generic code then passes the round count through to chacha_block(). There will be a ->setkey() function for each explicitly allowed round count; the encrypt/decrypt functions will be the same. I decided not to do it the opposite way (same ->setkey() function for all round counts, with different encrypt/decrypt functions) because that would have required more boilerplate code in architecture-specific implementations of ChaCha and XChaCha. Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by: Martin Willi <martin@strongswan.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-17 04:26:21 +03:00
crypto_chacha_init(state, ctx, walk.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>
2015-07-16 20:14:01 +03:00
kernel_fpu_begin();
crypto: x86/chacha20 - Use larger block functions more aggressively Now that all block functions support partial lengths, engage the wider block sizes more aggressively. This prevents using smaller block functions multiple times, where the next larger block function would have been faster. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:28 +03:00
while (walk.nbytes > 0) {
unsigned int nbytes = walk.nbytes;
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>
2015-07-16 20:14:01 +03:00
chacha20_dosimd(state, walk.dst.virt.addr, walk.src.virt.addr,
crypto: x86/chacha20 - Use larger block functions more aggressively Now that all block functions support partial lengths, engage the wider block sizes more aggressively. This prevents using smaller block functions multiple times, where the next larger block function would have been faster. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-11 12:36:28 +03:00
nbytes);
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>
2015-07-16 20:14:01 +03:00
}
kernel_fpu_end();
return err;
}
crypto: chacha20 - convert generic and x86 versions to skcipher This converts the ChaCha20 code from a blkcipher to a skcipher, which is now the preferred way to implement symmetric block and stream ciphers. This ports the generic and x86 versions at the same time because the latter reuses routines of the former. Note that the skcipher_walk() API guarantees that all presented blocks except the final one are a multiple of the chunk size, so we can simplify the encrypt() routine somewhat. Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-12-09 17:33:51 +03:00
static struct skcipher_alg alg = {
.base.cra_name = "chacha20",
.base.cra_driver_name = "chacha20-simd",
.base.cra_priority = 300,
.base.cra_blocksize = 1,
crypto: chacha20-generic - refactor to allow varying number of rounds In preparation for adding XChaCha12 support, rename/refactor chacha20-generic to support different numbers of rounds. The justification for needing XChaCha12 support is explained in more detail in the patch "crypto: chacha - add XChaCha12 support". The only difference between ChaCha{8,12,20} are the number of rounds itself; all other parts of the algorithm are the same. Therefore, remove the "20" from all definitions, structures, functions, files, etc. that will be shared by all ChaCha versions. Also make ->setkey() store the round count in the chacha_ctx (previously chacha20_ctx). The generic code then passes the round count through to chacha_block(). There will be a ->setkey() function for each explicitly allowed round count; the encrypt/decrypt functions will be the same. I decided not to do it the opposite way (same ->setkey() function for all round counts, with different encrypt/decrypt functions) because that would have required more boilerplate code in architecture-specific implementations of ChaCha and XChaCha. Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by: Martin Willi <martin@strongswan.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-17 04:26:21 +03:00
.base.cra_ctxsize = sizeof(struct chacha_ctx),
crypto: chacha20 - convert generic and x86 versions to skcipher This converts the ChaCha20 code from a blkcipher to a skcipher, which is now the preferred way to implement symmetric block and stream ciphers. This ports the generic and x86 versions at the same time because the latter reuses routines of the former. Note that the skcipher_walk() API guarantees that all presented blocks except the final one are a multiple of the chunk size, so we can simplify the encrypt() routine somewhat. Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-12-09 17:33:51 +03:00
.base.cra_module = THIS_MODULE,
crypto: chacha20-generic - refactor to allow varying number of rounds In preparation for adding XChaCha12 support, rename/refactor chacha20-generic to support different numbers of rounds. The justification for needing XChaCha12 support is explained in more detail in the patch "crypto: chacha - add XChaCha12 support". The only difference between ChaCha{8,12,20} are the number of rounds itself; all other parts of the algorithm are the same. Therefore, remove the "20" from all definitions, structures, functions, files, etc. that will be shared by all ChaCha versions. Also make ->setkey() store the round count in the chacha_ctx (previously chacha20_ctx). The generic code then passes the round count through to chacha_block(). There will be a ->setkey() function for each explicitly allowed round count; the encrypt/decrypt functions will be the same. I decided not to do it the opposite way (same ->setkey() function for all round counts, with different encrypt/decrypt functions) because that would have required more boilerplate code in architecture-specific implementations of ChaCha and XChaCha. Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by: Martin Willi <martin@strongswan.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-17 04:26:21 +03:00
.min_keysize = CHACHA_KEY_SIZE,
.max_keysize = CHACHA_KEY_SIZE,
.ivsize = CHACHA_IV_SIZE,
.chunksize = CHACHA_BLOCK_SIZE,
crypto: chacha20 - convert generic and x86 versions to skcipher This converts the ChaCha20 code from a blkcipher to a skcipher, which is now the preferred way to implement symmetric block and stream ciphers. This ports the generic and x86 versions at the same time because the latter reuses routines of the former. Note that the skcipher_walk() API guarantees that all presented blocks except the final one are a multiple of the chunk size, so we can simplify the encrypt() routine somewhat. Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-12-09 17:33:51 +03:00
.setkey = crypto_chacha20_setkey,
.encrypt = chacha20_simd,
.decrypt = chacha20_simd,
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>
2015-07-16 20:14:01 +03:00
};
static int __init chacha20_simd_mod_init(void)
{
x86/cpufeature: Remove unused and seldomly used cpu_has_xx macros Those are stupid and code should use static_cpu_has_safe() or boot_cpu_has() instead. Kill the least used and unused ones. The remaining ones need more careful inspection before a conversion can happen. On the TODO. Signed-off-by: Borislav Petkov <bp@suse.de> Link: http://lkml.kernel.org/r/1449481182-27541-4-git-send-email-bp@alien8.de Cc: David Sterba <dsterba@suse.com> Cc: Herbert Xu <herbert@gondor.apana.org.au> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Matt Mackall <mpm@selenic.com> Cc: Chris Mason <clm@fb.com> Cc: Josef Bacik <jbacik@fb.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2015-12-07 12:39:41 +03:00
if (!boot_cpu_has(X86_FEATURE_SSSE3))
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>
2015-07-16 20:14:01 +03:00
return -ENODEV;
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>
2015-07-16 20:14:03 +03:00
#ifdef CONFIG_AS_AVX2
x86/cpufeature: Replace cpu_has_avx with boot_cpu_has() usage Signed-off-by: Borislav Petkov <bp@suse.de> 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: 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: linux-crypto@vger.kernel.org Link: http://lkml.kernel.org/r/1459801503-15600-4-git-send-email-bp@alien8.de Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-04-04 23:24:56 +03:00
chacha20_use_avx2 = boot_cpu_has(X86_FEATURE_AVX) &&
boot_cpu_has(X86_FEATURE_AVX2) &&
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 02:31:26 +03:00
cpu_has_xfeatures(XFEATURE_MASK_SSE | XFEATURE_MASK_YMM, NULL);
crypto: x86/chacha20 - Add a 8-block AVX-512VL variant This variant is similar to the AVX2 version, but benefits from the AVX-512 rotate instructions and the additional registers, so it can operate without any data on the stack. It uses ymm registers only to avoid the massive core throttling on Skylake-X platforms. Nontheless does it bring a ~30% speed improvement compared to the AVX2 variant for random encryption lengths. The AVX2 version uses "rep movsb" for partial block XORing via the stack. With AVX-512, the new "vmovdqu8" can do this much more efficiently. The associated "kmov" instructions to work with dynamic masks is not part of the AVX-512VL instruction set, hence we depend on AVX-512BW as well. Given that the major AVX-512VL architectures provide AVX-512BW and this extension does not affect core clocking, this seems to be no problem at least for now. Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-11-20 19:30:48 +03:00
#ifdef CONFIG_AS_AVX512
chacha20_use_avx512vl = chacha20_use_avx2 &&
boot_cpu_has(X86_FEATURE_AVX512VL) &&
boot_cpu_has(X86_FEATURE_AVX512BW); /* kmovq */
#endif
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>
2015-07-16 20:14:03 +03:00
#endif
crypto: chacha20 - convert generic and x86 versions to skcipher This converts the ChaCha20 code from a blkcipher to a skcipher, which is now the preferred way to implement symmetric block and stream ciphers. This ports the generic and x86 versions at the same time because the latter reuses routines of the former. Note that the skcipher_walk() API guarantees that all presented blocks except the final one are a multiple of the chunk size, so we can simplify the encrypt() routine somewhat. Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-12-09 17:33:51 +03:00
return crypto_register_skcipher(&alg);
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>
2015-07-16 20:14:01 +03:00
}
static void __exit chacha20_simd_mod_fini(void)
{
crypto: chacha20 - convert generic and x86 versions to skcipher This converts the ChaCha20 code from a blkcipher to a skcipher, which is now the preferred way to implement symmetric block and stream ciphers. This ports the generic and x86 versions at the same time because the latter reuses routines of the former. Note that the skcipher_walk() API guarantees that all presented blocks except the final one are a multiple of the chunk size, so we can simplify the encrypt() routine somewhat. Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-12-09 17:33:51 +03:00
crypto_unregister_skcipher(&alg);
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>
2015-07-16 20:14:01 +03:00
}
module_init(chacha20_simd_mod_init);
module_exit(chacha20_simd_mod_fini);
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Martin Willi <martin@strongswan.org>");
MODULE_DESCRIPTION("chacha20 cipher algorithm, SIMD accelerated");
MODULE_ALIAS_CRYPTO("chacha20");
MODULE_ALIAS_CRYPTO("chacha20-simd");
Reference in New Issue Copy Permalink