期刊文献+

基于互补谓词的编译优化 被引量:2

Optimizing Compiler Based on Complementary Predicate
在线阅读 下载PDF
导出
摘要 许多通用和嵌入式高性能处理器都支持谓词执行,利用谓词执行可以简化程序的控制结构,而且指令调度、寄存器分配也可以利用谓词提高效率.基于二进制决策图(BDD)的谓词分析系统,提出了一种利用互补谓词机制对控制结构进行优化的方法;对传统着色图寄存器分配算法进行改进,给出了一种利用互补谓词机制建立统一简化干涉图的新算法,减少了代码溢出;利用互补谓词的特性指导指令调度,使之更加有效利用资源,并对目标体系结构进行了改进以支持指令调度.最后在YHFT-DSP/700芯片的编译器上对所提优化方法的有效性进行了验证. Predicated execution promises to reduce control flow overhead and to enhance optimization, provided that instruction scheduling and register allocation can utilize it efficiently. Based on binary decision diagram (BDD) predicate analysis system, this paper proposes the method to optimize control structures in programs utilizing complementary predicate. And based on traditional graph coloring register allocation, a new algorithm to construct unified and simplified interference graph utilizing complementary predicate is presented, which can reduce spill code. Moreover, complementary predicate-aware scheduling is proposed to reduce the superfluous commitment of resources to operations whose predicates evaluate to false at run-time, and the architecture is modified to support scheduling. Experiment results in YHFT- DSP/700 compiler show the effectiveness of proposal method.
出处 《电子学报》 EI CAS CSCD 北大核心 2006年第7期1280-1286,共7页 Acta Electronica Sinica
基金 国家863高技术研究发展计划基金(No.2004AA1Z1040) 国家自然科学基金(No.60473079)
关键词 编译器 互补谓词 谓词分析 指令调度 寄存器分配 compiler complementary predicate predicate analysis instruction scheduling register allocation
  • 相关文献

参考文献17

  • 1Triebel W.Itanium Architecture for Software Developers[M].Intel Press,2000.
  • 2TMS320C6000 CPU and Instruction Set Reference Guide (Rev.F)[M].Texas Instruments Inc.,2000.
  • 3SC140 DSP Core Reference Manual[M].Freescale Semiconductor,Inc.,2004.
  • 4Gillies D M,Ju D R,et al.Global predicate analysis and its application to register allocation[A].MICRO 29:Proceedings of the 29th Annual ACM/IEEE International Symposium on Microarchitecture[C].IEEE Computer Society,1996.114-125.
  • 5Mahlke S A,Lin D C,et al.Effective compiler support for predicated execution using the hyperblock[J].SIGMICRO Newsl.,1992,23 (1-2):45-54.
  • 6Schlansker M,Mahlke S,Johnson R.Control CPR:a branch height reduction optimization for EPIC architectures[A].PLDI'99:Proceedings of the ACM SIGPLAN 1999 Conference on Programming Language Design and Implementation[C].ACM Press,1999.155-168.
  • 7Johnson R,Schlansker M.Analysis techniques for predicated code[A].MICRO 29:Proceedings of the 29th Annual ACM/IEEE International Symposium on Microarchitecture[C].IEEE Computer Society,1996.100-113.
  • 8Leupers R.Exploiting conditional instructions in code generation for embedded VLIW processors[A].DATE'99:Proceedings of the Conference on Design,Automation and Test in Europe[C].ACM Press,1999.
  • 9Hwu W W,August D I,Sias J W.Program decision logic optimization using predication and control speculation[J].Proceedings of the IEEE,2001,89(11):1660 -1675.
  • 10Eichenberger A E,Davidson E S.Register allocation for predicated code[A].MICRO 28:Proceedings of the 28th Annual International Symposium on Microarchitecture[C].IEEE Computer Society Press,1995.180-191.

同被引文献50

  • 1胡定磊,陈书明.低功耗编译技术综述[J].电子学报,2005,33(4):676-682. 被引量:11
  • 2唐遇星,邓鹍,周兴铭.基于Trace-Cache的多级动态优化框架设计[J].电子学报,2005,33(11):1946-1951. 被引量:4
  • 3胡定磊,陈书明,刘春林.分簇结构超长指令字DSP编译器的设计与实现[J].小型微型计算机系统,2006,27(2):348-353. 被引量:7
  • 4陈书明,李振涛,万江华,胡定磊,郭阳,汪东,扈啸,孙书为.“银河飞腾”高性能数字信号处理器研究进展[J].计算机研究与发展,2006,43(6):993-1000. 被引量:29
  • 5TMS320C6000 CPU and Instruction Set Reference Guide(Rev.F)[G].Dallas,TX:texas Instruments Inc,2000.
  • 6J Fridman,Z Greenfield.The tiger SHARC DSP architecture[J].IEEE Micro,2000,20(1):66-76.
  • 7P Faraboschi,G Brown,et al.Lx:A technology platform for customizable VLIW embedded processing[C].In:Proc of the 27th Annual Int'l Symp on Computer Architecture.New York:ACM Press,2000.203-213.
  • 8B R Rau.Iterative modulo scheduling:An algorithm for software pipelining loops[C].In:Proc of the 27th Annual Int'l Symp on Microarchitecture.New York:ACM Press,1994.63-74.
  • 9P Faraboschi,J A Fisher,et al.Instruction scheduling for instruction level parallel processors[J].Proceedings of the IEEE,2001,89(11):1638-1659.
  • 10D M Lavery.Modulo scheduling for control-intensive generalpurpose programs:[Ph D dissertation][D].Urbana,IL:University of Illinois,1997.

引证文献2

二级引证文献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