期刊文献+

分支指令特性与分支预测器的性能研究 被引量:1

Research on Branch Characteristic and Branch Predictor Performance
下载PDF
导出
摘要 根据分支指令的特性,分析了分支行为与分支预测技术对单发射嵌入式处理器CPI栈(CPI stacks)组成的影响,并在RTL级设计了分支预测器的时序精确模型,通过硬件模拟方法对分支指令特性和分支预测器的性能进行了研究.实验考察了分支指令在分支预测器命中或缺失时的不同跳转统计特性,验证了分支预测器对CPI栈影响的理论推导,为单发射嵌入式处理器中分支预测器的设计与优化提供了精确的实验依据. Based on the characteristic of the branch instruction,the impact of branch behavior and branch prediction technique on the CPI stacks components of the embedded single-issue processor is analyzed in this paper,furthermore the accurate timing model of the branch predictor is designed in RTL-level,and the characteristic of branch instruction and the performance of branch predictor are also studied through hardware simulation.In our experiment the correspondent statistical properties of the branch instruction when the behavior of miss or hit in the branch predictor happens are recorded,proving the legitimate conclusion of the correlativity between the branch predictor and the CPI stacks components and providing the design and optimization of the branch predictor for the embedded single-issue processor with accurate experimental basis.
出处 《微电子学与计算机》 CSCD 北大核心 2010年第6期8-12,共5页 Microelectronics & Computer
关键词 CPI栈 分支预测器 单发射嵌入式处理器 硬件模型 CPI stacks branch predictor single-issue embedded processor hardware model
  • 相关文献

参考文献7

  • 1Evers M,Yeh T Y.Understanding branches and designing branch predictors for high-performance microprocessors[J].IEEE,2001,89(11):1610-1620.
  • 2陈恺冬,史江一,郝跃.一种基于双模结构的预测器[J].微电子学与计算机,2008,25(5):110-113. 被引量:2
  • 3Stijn Eyerman,Lieven Eeckhout,Tejas Karkhanis,et al.A performance counter architecture for computing accurate CPI components[C]//Proceedings of the 2006 ASPLOS Conference on ACM SIGARCH Computer Architecture News.San Jose,2006:175-184.
  • 4黄海林,范东睿,许彤,唐志敏.嵌入式处理器中访存部件的低功耗设计研究[J].计算机学报,2006,29(5):815-821. 被引量:11
  • 5Perleberg C,Smith A.Branch target buffer design and optimization[J].IEEE Transactions on Computers,1993,42(4):396-412.
  • 6武萌,沈海斌.一种gshare分支预测器的低功耗设计方法[J].微电子学与计算机,2007,24(3):200-202. 被引量:7
  • 7Lee J K L,Smith A J.Branch prediction strategies and branch target buffer design[J].Computer,1984,17(1):6-22.

二级参考文献25

  • 1刘岩,侯朝焕.一种静态可控功耗的数据Cache设计[J].微电子学与计算机,2004,21(11):135-137. 被引量:4
  • 2陈智勇.超标量处理器的转移预测方案研究[J].微电子学与计算机,2006,23(11):118-120. 被引量:1
  • 3武萌,沈海斌.一种gshare分支预测器的低功耗设计方法[J].微电子学与计算机,2007,24(3):200-202. 被引量:7
  • 4Hennessy J.L,Patterson D.A..Computer Architecture:A Quantitative Approach (2nd Edition).Beijing:China Machine Press,2002
  • 5Clark L.T,Choi B,Wilkerson M..Reducing translation lookaside buffer active power.In:Proceedings of the ISLPED,Seoul,Korea,2003,10~13
  • 6Juan T.et al.Reducing TLB power requirements.In:Proceedings of the ISLPED,Monterey,California,1997,196~201
  • 7Chen J.B.et al.A simulation based study of TLB performance.In:Proceedings of the ISCA,Queensland,Australia,1992,114~123
  • 8Inoue K..High-performance low-power cache memory architecture[Ph.D.dissertation].Kyushu University,Fukuoka,Japan,2001
  • 9Chang Yen-Jen,Ruan Shang-Jang,Lai Fei-Pei.Design and analysis of low-power cache using two-level filter scheme.IEEE Transactions on VLSI Systems,2003,11(4):568~580
  • 10Kin J,Gupta M,Mangione-Smith W.H..The filter cache:An energy efficient memory structure.In:Proceedings of the MICRO-97:ACM/IEEE International Symposium on Microarchitecture,Los Alamitos,USA,1997,184~193

共引文献17

同被引文献4

  • 1Evers M, Yeh TY. Understanding branches and designing branch predictors for high-performance micro- processors[J]. Proceedings of the IEEE, 2001, 89 (11) : 1610-1620.
  • 2Perleberg C, Smith A. Branch target buffer design and optimization [J]. IEEE Transactions on Computers, 1993, 42(4): 396-412.
  • 3Scott McFarling. Combining branch predictors[R]. California: Digital Western Research Laboratory, 1993.
  • 4鄢国丰,张盛兵,黄小平.“龙腾R”微处理器分支处理单元的研究与设计[J].微电子学与计算机,2010,27(6):122-127. 被引量:2

引证文献1

二级引证文献1

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