集成电路与嵌入式系统2026，Vol.26Issue(1)：47-53,7.DOI:10.20193/j.ices2097-4191.2025.0082
基于开源处理器的间接访问数据预取器设计
Design of differential-matching prefetcher based on open-source CPU
摘要
Abstract
Indirect memory accesses,prevalent in data-intensive applications such as graph processing and sparse linear algebra,exhibit irregular patterns that severely degrade cache performance due to their low spatial/temporal locality.Traditional stride-based prefetchers fail to capture such patterns where target addresses are dynamically computed through index arrays.This paper proposes a dynamic multi-pattern-aware prefetcher(DMP)to address these challenges.DMP introduces a lightweight shifted differential matching mecha-nism to autonomously identify indirect access patterns by comparing index data sequences with target address sequence.Implemented on the open-source XuanTie C910 RISC-V processor,DMP reduces L1 data cache miss rates by 27.3%and achieves speedups of 1.07～1.22 timed for Sparse Matrix-Vector Multiplication(SpMV)algorithm.This work provides a hardware-efficient solution for non-contiguous memory access patterns in modern processors.关键词
数据预取/间接内存访问/缓存优化/非连续访问/硬件高效性/非规则访存Key words
data prefetching/indirect memory access/cache optimization/non-contiguous access/hardware efficiency/irregular workloads分类
信息技术与安全科学引用本文复制引用
ZONG Pengchen,QU Shaoru,ZHAO Wenzhe,REN Pengju,XIA Tian..基于开源处理器的间接访问数据预取器设计[J].集成电路与嵌入式系统,2026,26(1):47-53,7.基金项目
国家自然科学基金(62088102) （62088102）
国家自然科学基金(62302381) （62302381）
国家重点研发计划项目(2022YFB4500500) （2022YFB4500500）
