期刊文献+

龙芯链接后优化器设计与分析 被引量:2

GLTO Design and Analysis
下载PDF
导出
摘要 链接后优化技术是在编译链接后对整个程序再进行优化的一种技术·它克服了传统编译器优化局限于一个函数、一个模块的缺点,将优化范围扩展到整个程序,并且充分利用了链接后确定的信息·参照Arizona大学为Alpha处理器设计的链接后优化器ALTO,针对龙芯2号处理器的微体系结构和指令集的特征,设计了龙芯上的链接后优化器GLTO(Godsonlinktimeoptimizer)·GLTO使得龙芯处理器SPEC2000定点程序ref分值提高了9·4%,具有显著的优化效果·分析了主要优化策略的效果和产生的原因,提出了处理器的结构设计中的改进设想,并将GLTO与ALTO做了对比分析· Link time optimization is a technique which optimizes the whole program after compilation. It overcomes the limitations of traditional compilers by enlarging the optimizing scope from a single function or a module to the whole program, and fully utilizes the information only available at link time. Guided by the link time optimizer ALTO developed by Arizona University for the Compag Alpha, the GLTO (Godson link time optimizer) is designed with the consideration of the features of the microarchitecture and the instruction set of the Godson-2 processor. GLTO can achieve a 9.4% performance improvement of the SPEC2000 integer benchmark with ref input on average. In this paper, the effect and the cause of several major optimizing methods are explored, some improvements in the processor architecture design are proposed, and the differences between GLTO and ALTO are analyzed.
出处 《计算机研究与发展》 EI CSCD 北大核心 2006年第8期1450-1456,共7页 Journal of Computer Research and Development
基金 国家自然科学基金杰出青年基金项目(60325205) 国家"八六三"高技术研究发展计划重点基金项目(2002AA110010) 中国科学院知识创新工程重大项目(KGCX2-109)~~
关键词 链接后优化 GLTO 龙芯2号处理器 link time optimization GLTO Godson-2 processor
  • 相关文献

参考文献12

  • 1Saumya Debray, Robert Muth, Scott Watterson, et al.ALTO: A link-time optimizer for the Compaq Alpha [J].Software-Practice and Experience, 2001, 101 ( 1 ) : 31-67
  • 2胡伟武,张福新,等.龙芯项目资料[G].北京:中国科学院计算技术研究所,2004
  • 3V Santhanam, D Odnert. Register allocation across procedure and module boundaries [C]. In: Proc of the SIGPLAN'90 Conf on Programming Language Design and Implementation. New York: ACM Press, 1990. 28-39
  • 4D W Wall. Global register allocation at link time [ C]. In: Proc of the SIGPLAN' 86 Symposium on Compiler Construction.New York: ACM Press, 1986. 264-275
  • 5A Ayers, S de Jong, J Peyton, et al. Scalable cross-module optimization [C]. In: Proc of the SIGPLAN' 98 Conf on Programming Language Design and Implementation. New York: ACM Press, 1998. 301-312
  • 6R Cohn, D Goodwin, P G Lowney, et al. Optimizing Alpha executables on Windows NT with spike [J ]. Digital Technical Journal, 1997, 9(4): 3-20
  • 7T Romer, G Voelker, D Lee, et al. Instrumentation and optimization of Win32/Intel executables [C]. In: Proc of the USENIX Windows NT Workshop. Redmond, WA, USA:Microsoft Press, 1997. 1-7
  • 8Vasanth Bala, Evelyn Duesterwald, Sanjeev Banerjia. Dynamo:A transparent dynamic optimization system [ C]. In: Proc of the ACM SIGPLAN Conf on Programming Language Design and Implementation. New York: ACM Press, 2000. 1-12
  • 9陈瑜.龙芯链接后优化器实现[R].中国科学院计算技术研究所,技术报告:CS81—2,2004
  • 10Karl Pettis, Robert C Hansen. Profile guided code positioning[C]. In: Proc of the ACM SIGPLAN Conf on Programming Language Design and Implementation. New York : ACM Press,1990. 16-27

同被引文献16

  • 1杨灿群 李春江.异构多核处理器的编程模型和编译优化技术.中国计算机学会通讯,2007,3(2):32-39.
  • 2Bunda J,Fussell D,Jenevein R,et al.16-bit vs.32-bit Instructions for Pipelined Micro-processors [C]// Proceedings of the 20th Annual International Symposium on Computer Architecture.[S.l.]:IEEE Press,1993:237-246.
  • 3Gupta A R.Enhancing the Performance of 16-bit Code Using Augmenting Instructions [C]//Proceedings of ACM SIGPLAN Conference on Languages Compilers.[S.l.]:ACM Press,2003.
  • 4陈瑜.龙芯2号链接后优化器的实现与分析[D].北京:中国科学院计算技术研究所,2004.
  • 5Li D X,Ashok R,Hundt R.Lightweight Feedbackdirected Cross-module Optimization [C]//Proceedings of the 8th Annual IEEE/ACM International Symposium on Code Generation and Optimization.[S.l.]: ACM Press,2010:53-61.
  • 6Liu Xianhua,Zhang Jiyu,Cheng Xu.Efficient Code Size Reduction Without Performance Loss[C]//Proceedings of ACM Symposium on Applied Computing.[S.l.]: ACM Press,2007:666-672.
  • 7Bus B,Kaster D,Chanet D,et al.Post-pass Compaction Techniques 2003[J].Communications of the ACM,2003,46(8):41-46.
  • 8Phela R.Improving ARM Code Density and Performance [EB/OL].(2003-05-07).http:// www.cs.uiuc.edu/class/fa05/ cs433ug/PROCESSORS/Thumb2.pdf.
  • 9Sutter B.Link-time Binary Rewriting Techniques for Program Compaction [J].ACM Transactions on Programming Languages and Systems,2005,27(5):882-945.
  • 10Jones T M,Bartolini S,Maebe J,et al.Link-time Optimization for Power Efficiency in a Tagless Instruction Cache[C]// Proceedings of the 9th Annual IEEE/ ACM International Symposium on Code Generation and Optimization.[S.l.]:IEEE Press,2011:32-41.

引证文献2

二级引证文献5

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