基于指令距离的存储相关性预测方法

doi:10.11772/j.issn.1001-9081.2013.07.1903

计算机应用 ›› 2013, Vol. 33 ›› Issue (07): 1903-1907.DOI: 10.11772/j.issn.1001-9081.2013.07.1903

基于指令距离的存储相关性预测方法

路冬冬,何军,杨剑新,王飙

上海高性能集成电路设计中心前端设计部，上海 201204

收稿日期:2013-01-07 修回日期:2013-03-01 发布日期:2013-07-06 出版日期:2013-07-01
通讯作者: 路冬冬
作者简介:路冬冬(1988-)，男，河南安阳人，助理工程师，硕士研究生，主要研究方向：微处理器设计；何军(1980-)，男，湖北汉川人，工程师，博士研究生，主要研究方向：微处理器设计；杨剑新(1976-)，男，四川成都人，高级工程师，硕士，主要研究方向：微处理器设计、微处理器验证。

Memory dependence prediction method based on instruction distance

LU Dongdong,HE Jun,YANG Jianxin,WANG Biao

The Front-end Design Department, Shanghai High Performance IC Design Centre, Shanghai 201204, China

Received:2013-01-07 Revised:2013-03-01 Online:2013-07-06 Published:2013-07-01
Contact: LU Dongdong

摘要/Abstract

摘要： 存储相关性预测对于减少存储相关性冲突、提高微处理器性能具有十分重要的作用。针对传统相关性预测器硬件开销大、可实现性较差的缺点，通过对存储相关性的局部性分析，提出了一种基于指令距离的存储相关性预测方法。该方法充分利用了发生存储相关性冲突的指令在指令距离上的局部性，预测冲突指令的指令距离，进而控制部分访存指令的发射时机，大大减少了存储相关性冲突的次数。实验结果表明，在硬件开销约为1KB的情况下，使用基于指令距离的相关性预测器后，每个时钟周期平均执行的指令数可以提高1.70%，最高可以提高5.11%。在硬件开销较小的情况下，较大程度提高了微处理器的性能。

关键词: 指令级并行, 访存指令, 存储相关性预测, 指令距离

Abstract: Memory dependence prediction plays a very important role to reduce memory order violation and improve microprocessor performance. However, the traditional methods usually have large hardware overhead and poor realizability. Through the analysis of memory dependence's locality, this paper proposed a new memory predictor based on instruction distance. Compared to other memory dependence predictors, this predictor made full use of memory dependence's locality on instruction distance, predicted memory instruction' violation distance, controlled the speculation of a few instructions, finally deduced the number of memory order violation and improved the performance. The simulation results show that with only 1KB hardware budget, average Instruction Per Cycle (IPC) get a 1.70% speedup, and the most improvement is 5.11%. In the case of a small hardware overhead, the performance is greatly improved.

Key words: Instruction Level Parallelism (ILP), memory instruction, memory dependence prediction, instruction distance

中图分类号:

TP302.1

路冬冬何军杨剑新王飙. 基于指令距离的存储相关性预测方法[J]. 计算机应用, 2013, 33(07): 1903-1907.

LU Dongdong HE Jun YANG Jianxin WANG Biao. Memory dependence prediction method based on instruction distance[J]. Journal of Computer Applications, 2013, 33(07): 1903-1907.

参考文献

［1］CRISTAL A, SANTANA O J, VALERO M, et al. Toward kilo-instruction processors ［J］. ACM Transactions on Architecture and Code Optimization, 2004, 1(4)： 389-417.

［2］SUBRAMANIAM S. Improving processor efficiency by exploiting common-case behaviors of memory instructions ［D］. Atlanta： Georgia Institute of Technology, 2009.

［3］SHA T. Exploiting aggressive memory dependence speculation to simplify the store-load datapath ［D］. Philadelphia：University of Pennsylvania, 2011.

［4］LIPASTI M H, SHEN J P. The performance potential of value and dependence prediction ［C］// Proceedings of the 3rd International Euro-Par Conference on Parallel Processing. Berlin： Springer, 1997： 1043-1052.

［5］MOSHOVOS A, BREACH S E, VIJAYKUMAR T N, et al. Dynamic speculation and synchronization of data dependencies ［J］. ACM SIGARCH Computer Architecture News, 1997, 25(2)：181-193.

［6］MOSHOVOS A. Memory dependence prediction ［D］. Madison： University of Wisconsin, 1998.

［7］CHRYSOS G Z, EMER J S. Memory dependence prediction using store sets ［C］// Proceedings of the 25th Annual International Symposium on Computer Architecture. Washington, DC： IEEE Computer Society, 1998： 142-153.

［8］SUBRAMANIAM S, LOH G H. Store vectors for scalable memory dependence prediction and scheduling ［C］// Proceedings of the 12th International Symposium on High Performance Computer Architecture. Atlanta, GA： Conference Publications, 2006： 64-75.

［9］STONE S S, WOLEY K M, FRANK M I. Address-indexed memory disambiguation and store-to-load forwarding ［C］// Proceedings of the 38th Annual IEEE/ACM International Symposium on Microarchitecture. Washington, DC： IEEE Computer Society, 2005： 171-182.

［10］KESSLER R E. The Alpha 21264 microprocessor ［J］. IEEE Micro Magazine, 1999, 19(2)： 24-36.

［11］DOWECK J. Inside Intel core microarchitecture and smart memory access ［EB/OL］. ［2012-12-05］. http：//www.intel.com/pressroom/kits/core2duo/pdf/ICM_whitepaper.pdf.

［12］ROESNER F, BURGER D, KECKLER S W. Counting dependence predictors ［C］// Proceedings of the 35th Annual International Symposium on Computer Architecture. Washington, DC： IEEE Computer Society, 2008： 215-226.

［13］FANG C, CARR S, ONDER S, et al. Feedback-directed memory disambiguation through store distance analysis ［C］// Proceedings of the 20th Annual International Conference on Supercomputing. New York： ACM Press, 2006： 278-287.

[1]	王寿成, 徐进辉, 严迎建, 李功丽, 贾永旺. 面向密码流处理器的AES算法软件流水实现方法[J]. 计算机应用, 2017, 37(6): 1620-1624.
[2]	黄胜兵, 郑启龙, 郭连伟. 分簇VLIW DSP上支持单双字模式选择的SIMD编译优化[J]. 计算机应用, 2015, 35(8): 2371-2374.
[3]	邱鹏飞洪一耿锐徐云. 基于数据流图的异构VLIW DSP分簇方法[J]. 计算机应用, 2011, 31(04): 935-937.

基于指令距离的存储相关性预测方法

Memory dependence prediction method based on instruction distance

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics