期刊文献+

基于历史长度自适应的分支预测方法 被引量:2

Branch Prediction Based on Branch History Length Adaptation
下载PDF
导出
摘要 通过研究处理器动态分支预测器中预测效率与分支历史长度的关系,针对程序中各分支指令存在不同最优历史长度的规律,提出一种搜索各分支指令最佳历史长度的分支预测方法.该方法通过实时监测分支指令的预测准确率,在分支预测表硬件资源不变的情况下动态调整预测器的历史长度,以适应程序的动态运行特征.实验结果表明,在相同硬件资源下,文中方法相对于Gshare预测器错误率降低15.8%,相对于Bi-mode预测器预测错误率降低10.3%. In this paper, we study the relationship between predict accuracy and branch history length in dynamic branch predictor, and found that different branch instruction will achieve optimal prediction accuracy with differ-ent history length. Thus, we propose a branch prediction strategy, which adapts the branch history length for each branch instruction to achieve high prediction accuracy in total. This approach monitors branch misprediction rate, and dynamically changes the branch history length when the misprediction rate higher than a threshold, this ap-proach can avoid some critical branch alias during the program’s execution, leading the predictor to achieve the optimal prediction accuracy. The experimental results show that, in comparison Gshare, this method achieved 15.8% reduction in misprediction rate, while in comparison with Bi-mode, 10.3% reduction in mispredition rate was achieved.
出处 《计算机辅助设计与图形学学报》 EI CSCD 北大核心 2015年第4期764-770,共7页 Journal of Computer-Aided Design & Computer Graphics
关键词 分支预测 分支别名 错误率监测 历史长度自适应 branch prediction branch alias mis-prediction rate monitor history length adaptation
  • 相关文献

参考文献18

  • 1陈晨,陈志坚,孟建熠,严晓浪.基于预测极性动态变换的分支预测框架研究[J].电子与信息学报,2013,35(4):1001-1006. 被引量:2
  • 2苟鹏飞,王诗博,杨兵,喻明艳.改进的基于O-GEHL预测技术的EDGE块预测器[J].电子科技大学学报,2012,41(2):305-310. 被引量:1
  • 3Tse-Yu Yeh,and Yale N. Patt.Alternative Implementations of Two-Level Adaptive Branch Prediction. Proceedings of the 19th Annual InternationalSymposium on Computer Architecture . 1992
  • 4Seznec A.A new case for the TAGE branch predictor. Proceedings of the 44th Annual IEEE/ACM InternationalSymposium on Microarchitecture . 2011
  • 5Eden A N,Mudge T.The YAGS branch prediction scheme. International Symposium on Microarchitecture . 1998
  • 6Jiménez D.An optimized scaled neural branch predictor. 2011 IEEE 29th International Conference on ComputerDesign (ICCD) . 2011
  • 7Poovey, Jason A.,Conte, Thomas M.,Levy, Markus,Gal-On, Shay.A benchmark characterization of the EEMBC benchmark suite. IEEE Micro Magazine . 2009
  • 8Jimenez D A.Oh-snap:optimized hybrid scaled neural analog predictor[OL]. http://www.jilp.org/jwac-2/program/cbp3_02_jimenez.pdf . 2014
  • 9Mudge T,Lee C C,Sechrest S.Correlation and aliasing in dynamic branch predictors. Proceedings of the 23rd Annual International Symposium on Computer Architecture . 1996
  • 10Lee C C,Chen I C K,Mudge T N.The Bi-mode branch predictor. Proceedings of the 30th Annual IEEE/ACM International Symposium on Microarchitecture . 1997

二级参考文献32

  • 1Hennessy J and Patterson D. Computer Architecture: A Quantitative Approach[M]. Fifth Edition, Beijing: China Machine Press, 2012:162- 167.
  • 2Jim6nez D. An optimized scaled neural branch predictor[C]. 2011 IEEE 29th International Conference on Computer Design (ICCD), Amherst, 2011:113- 118.
  • 3Jim6nez D. Oh-snap: optimized hybrid scaled neural analog predictor[C]. Proceedings of the 3rd Championship on Branch Prediction, Detroit, 2011:9- 12.
  • 4Yell T and Patt Y. Two-level adaptive training branch prediction[C]. Proceedings of the 24th Annual International Symposium on Microarchitecture, New York, 1991: 51-61.
  • 5Zhang Long, Tao Fang, and Xiang Jin-feng. Researches on design and implementations of two 2-bit predictors[J]. Advanced Engineering Forum, 2011, 1(1): 241-246.
  • 6Young C, Gloy N, and Smith M. A comparative analysis of schemes for correlated branch prediction[C]. Proceedings of the 22nd Annual International Symposium on Computer Architecture. New York, 1995: 276-286.
  • 7TMcott A, Nemirovsky M, and Wood R. The influence of branch prediction table interference on branch prediction scheme performance[C]. Proceedings of tile 3rd Annual International Conference on Parallel Architectures and Compilation Techniques, Manchester, 1995: 89-98.
  • 8Sprangle E, Chappell R, Alsup M, et al.. The agree predictor: mechanism for reducing negative branch history interference[C]. Proceedings of the 24th Annual International Symposium on Computer Architecture, New York, 1997: 284-291.
  • 9Lee C, Chen I, and Mudge T. The bi-mode branch predictor [C]. 30th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO'97), Research Triangle Park, NC, 1997:4 -13.
  • 10Seznec A. A 64 kbytes ISL-TAGE branch predictor[C]. Proceedings of the 3rd Championship Branch Prediction, Detroit, 2011: 13-16.

共引文献1

同被引文献6

引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部