From 5c102b99d332db54a2dddb321bf8562f275fab26 Mon Sep 17 00:00:00 2001 From: Sun Yimin Date: Fri, 29 Dec 2023 08:42:35 +0800 Subject: [PATCH] =?UTF-8?q?Updated=20SM2=E6=80=A7=E8=83=BD=E4=BC=98?= =?UTF-8?q?=E5=8C=96=EF=BC=88=E7=BB=AD=EF=BC=89=20(markdown)?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- SM2性能优化(续).md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/SM2性能优化(续).md b/SM2性能优化(续).md index 639cb31..32f08db 100644 --- a/SM2性能优化(续).md +++ b/SM2性能优化(续).md @@ -13,11 +13,11 @@ prime order elliptic curves, §A.2](https://eprint.iacr.org/2015/1060)。这里 * 基点标量乘法查找表生成,没有重新写生成方法,延用以前的,可以参考[gen_p256_table.go](https://github.com/emmansun/gmsm/blob/go_1.16/sm2/gen_p256_table.go)。 * 调整了点标量乘法的窗口大小,具体请参考[sm2 point scalar multiple ASM implementation issue: final p256PointAddAsm's input maybe equal](https://github.com/emmansun/gmsm/discussions/127) -## amd64架构优化 +### amd64架构优化 * 使用**MULX**优化, 可以参考这个提交 https://github.com/emmansun/gmsm/commit/a0c4a389b8ceea6cbad3ede4eef011b26482b2ad * AVX2优化,这里只是SELECT、MOVE等优化,不涉及计算。 -## amd64架构优化 +### amd64架构优化 * 使用SIMD实现SELECT、MOVE操作