lkml.org 
[lkml]   [2021]   [Dec]   [21]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [PATCH 5/6] crypto: x86/sm3 - add AVX assembly implementation
From


On 12/21/21 2:03 AM, Jussi Kivilinna wrote:
> On 20.12.2021 10.22, Tianjia Zhang wrote:
>> This patch adds AVX assembly accelerated implementation of SM3 secure
>> hash algorithm. From the benchmark data, compared to pure software
>> implementation sm3-generic, the performance increase is up to 38%.
>>
>> The main algorithm implementation based on SM3 AES/BMI2 accelerated
>> work by libgcrypt at:
>> https://gnupg.org/software/libgcrypt/index.html
>>
>> Benchmark on Intel i5-6200U 2.30GHz, performance data of two
>> implementations, pure software sm3-generic and sm3-avx acceleration.
>> The data comes from the 326 mode and 422 mode of tcrypt. The abscissas
>> are different lengths of per update. The data is tabulated and the
>> unit is Mb/s:
>>
>> update-size |     16      64     256    1024    2048    4096    8192
>> --------------------------------------------------------------------
>> sm3-generic | 105.97  129.60  182.12  189.62  188.06  193.66  194.88
>> sm3-avx     | 119.87  163.05  244.44  260.92  257.60  264.87  265.88
>>
>> Signed-off-by: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
>> ---
>>   arch/x86/crypto/Makefile         |   3 +
>>   arch/x86/crypto/sm3-avx-asm_64.S | 521 +++++++++++++++++++++++++++++++
>>   arch/x86/crypto/sm3_avx_glue.c   | 134 ++++++++
>>   crypto/Kconfig                   |  13 +
>>   4 files changed, 671 insertions(+)
>>   create mode 100644 arch/x86/crypto/sm3-avx-asm_64.S
>>   create mode 100644 arch/x86/crypto/sm3_avx_glue.c
>>
>> diff --git a/arch/x86/crypto/Makefile b/arch/x86/crypto/Makefile
>> index f307c93fc90a..7cbe860f6201 100644
>> --- a/arch/x86/crypto/Makefile
>> +++ b/arch/x86/crypto/Makefile
>> @@ -88,6 +88,9 @@ nhpoly1305-avx2-y := nh-avx2-x86_64.o
>> nhpoly1305-avx2-glue.o
>>   obj-$(CONFIG_CRYPTO_CURVE25519_X86) += curve25519-x86_64.o
>> +obj-$(CONFIG_CRYPTO_SM3_AVX_X86_64) += sm3-avx-x86_64.o
>> +sm3-avx-x86_64-y := sm3-avx-asm_64.o sm3_avx_glue.o
>> +
>>   obj-$(CONFIG_CRYPTO_SM4_AESNI_AVX_X86_64) += sm4-aesni-avx-x86_64.o
>>   sm4-aesni-avx-x86_64-y := sm4-aesni-avx-asm_64.o sm4_aesni_avx_glue.o
>> diff --git a/arch/x86/crypto/sm3-avx-asm_64.S
>> b/arch/x86/crypto/sm3-avx-asm_64.S
>> new file mode 100644
>> index 000000000000..e7a9a37f3609
>> --- /dev/null
>> +++ b/arch/x86/crypto/sm3-avx-asm_64.S
>> @@ -0,0 +1,521 @@
>> +/* SPDX-License-Identifier: GPL-2.0-or-later */
>> +/*
>> + * SM3 AVX accelerated transform.
>> + * specified in:
>> https://datatracker.ietf.org/doc/html/draft-sca-cfrg-sm3-02
>> + *
>> + * Copyright (C) 2021 Jussi Kivilinna <jussi.kivilinna@iki.fi>
>> + * Copyright (C) 2021 Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
>> + */
> <snip>
>> +
>> +#define R(i, a, b, c, d, e, f, g, h, round, widx,
>> wtype)                      \
>> +    /* rol(a, 12) => t0
>> */                                                \
>> +    roll3mov(12, a, t0); /* rorxl here would reduce perf by 6% on
>> zen3 */ \
>> +    /* rol (t0 + e + t), 7) => t1
>> */                                      \
>> +    addl3(t0, e,
>> t1);                                                     \
>> +    addl $K##round,
>> t1;                                                   \
>
> It's better to use "leal K##round(t0, e, 1), t1;" here and fix K0-K63
> macros
> instead as I noted at libgcrypt mailing-list:
>  https://lists.gnupg.org/pipermail/gcrypt-devel/2021-December/005209.html
>
> -Jussi

Thanks for pointing it out, I will fix it in the next patch.

Best regards,
Tianjia

\
 
 \ /
  Last update: 2021-12-21 08:40    [W:0.049 / U:1.888 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site