期刊文献+

基于DICache的混合线索解释执行技术 被引量:1

A DICache-Based Hybrid Threaded Interpretation Technique
下载PDF
导出
摘要 由解释执行实现的指令集仿真是解决二进制兼容问题的有效手段。解释执行各步骤的组织方式对解释器性能有着重要影响。集中方式效率较低,而效率较高的线索方式由于译码过程过于复杂而无法用于CISC指令集的解释执行。本文提出了一种基于DICache的混合线索解释执行技术,DICache实现一种高效的硬件动态预译码,将源指令转换为一种中间表示,在解释例程中对DICache快速访问实现对CISC指令集的线索解释执行。本文在一个源为IA-32、目标为VLIW的解释器上,采用SPEC INT2000中的测试程序对基于DICache的混合线索解释执行技术进行评估。结果表明该方法可以显著提高解释器的性能。 Interpretation-based instruction set emulation provides a solution to the binary compatibility problem.The organization of the interpretation process has great impact on the performance of an interpreter.Centralized interpretation is inefficient while the traditional threaded interpretation is not suitable for interpreting the CISC Instruction Set Architecture (ISA) because of the complicated instruction decoding process.We propose a DICache-based hybrid interpretation technique.DICache can dynamically predecode the source instruction and convert them into an intermediate form.The DICache access code is appended at the end of each interpreter routine,which enables the threaded interpretation for CISC ISA.We implement an interpreter which interprets the IA-32 instructions on VLIW.We conduct some benchmarks from SPEC INT 2000 to evaluate the performance of the novel threaded interpretation method.It is demonstrated that DICache-based hybrid threaded interpretation can significantly improve the performance of an interpreter.
出处 《计算机工程与科学》 CSCD 北大核心 2012年第2期50-55,共6页 Computer Engineering & Science
基金 国家973计划资助项目(2007CB310901) 国家自然科学基金资助项目(60803041)
关键词 二进制兼容 指令集仿真 线索解释执行 DICache binary compatibility instruction set emulation threaded interpretation DICache
  • 相关文献

参考文献14

  • 1Mihocka D, Shwartsman S. Virtualization without Direct Execution or ] it ting , Designing a Portable Virtual Machine Infrastructure [C]/ /Proc of the 1st Workshop on Architectural and Microarchitectural Support for Binary Translation, 2008: 55-70.
  • 2Bochs 2.3. 6[EB/OL]. [2009-06-19]. http://sourceforge. net/projects/bochs/files/bochs/2. 3. 6/bochs-2. 3. 6/.
  • 3Ebcioglu K, Altman E R. DAISY: Dynamic Compilation for 100% Architectural Compatibility [C]/ /Proc of the 24th Int'l Syrnp on Computer Architecture, 1997:26-37.
  • 4Dehnertl C, GrantBK, Banning]P, eta!' The Transmeta Code Morphing P" Software: Using Speculation, Recovery, and Adaptive Retranslation to Address Real-Life Challenges [C]/ /Proc of the 1st Annual IEEE/ ACM Int' I Syrnp on Code Generation and Optimization, 2003: 15-24.
  • 5Smithl E, Nair R. Virtual Machines: Versatile Platforms for Sysetrns and Processes [M]. Beijing: Publishing House of Electronics Industry, 2006.
  • 6Klint P. Interpretation Techniques [J]. Software Practice and Experience, 1981, 11(9): 963-973.
  • 7Kogge P M. An Architecture Trail to Threaded-Code Systems [J]. IEEE Computer, 1982, 15(33) :22-34.
  • 8Intel Corporation. Intel 64 and IA-32 Architectures Software Developer's Manual, vol. 1: Basic Architecture[EB/OL]. [2009-09-19]. http://www. intel, com/ Assets/PDF/manual/253665. pdf.
  • 9Bell R. Threaded Code [J]. Communications of the ACM, 1973, 16(6): 370-372.
  • 10Chen Wei, Wang Zhiying, Lu Hongyi , et al. A Hardware Approach for Reducing Interpretation Overhead[C]//Proc of the 9th IEEE Int' I Conf on Computer and Information Technology, 2009:98-103.

同被引文献2

引证文献1

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