基于KCPSM6指令扩展的SM4算法优化实现方法  

An Optimized Implementation of SM4 Algorithm Based on KCPSM6 Instruction Set Extension

作　　者：赵涔伶 周波[1] 王曦 杜薇 黄益盛[1] 张武 ZHAO Cenling;ZHOU Bo;WANG Xi;DU Wei;HUANG Yisheng;ZHANG Wu(No.30 Institute of CETC,Chengdu Sichuan 610041,China)

机构地区：[1]中国电子科技集团公司第三十研究所,四川成都610041

出　　处：《通信技术》2024年第6期640-646,共7页Communications Technology

摘　　要：密码算法的实现是低成本、资源受限的密码应用场景中的关键环节,而如何应用各类微处理器指令扩展技术对算法进行优化实现是当前研究重点之一。针对SM4算法,提出了一种基于KCPSM6微处理器指令扩展的算法优化实现方法。该方法实现了兼具密钥扩展函数和加解密函数功能的指令扩展单元,新增4个数异或计算的扩展指令,并采用双端口RAM交互接口设计了一款资源占用低的硬件电路结构。实验结果显示,LUT资源仅占用134个,相比不包含指令扩展的方法,所提方法消耗的时钟周期至少降低了71.38%,速率最大提升了4.79倍,有效实现了在较低硬件资源占用的情况下提升算法性能。The implementation of cryptographic algorithms is a key part in low-cost and resource limited cryptographic application scenarios,and how to apply various microprocessor instruction extension technologies to optimize the implementation of algorithms is currently one of the research focuses.For the SM4 algorithm,this paper proposes an optimized implementation method based on instruction set extension of KCPSM6 microprocessor.The method implements an instruction set extension unit that combines key extension function and encryption/decryption function,which adds extension instructions for XOR calculation of four numbers,and designs a low resource consumption hardware circuit structure by using an interactive interface of a dual-port RAM.The experimental results indicate that only 134 LUT resources have been occupied,and compared with the no-extension instruction,clock cycles are reduced by 71.38%at least,and the speed is maximally improved by 4.79 times,which effectively achieves the improvement of the algorithm performance with lower hardware resource consumption.

关 键 词：SM4 指令扩展 KCPSM6 低资源占用 

分 类 号：TP309.7[自动化与计算机技术—计算机系统结构]