期刊文献+

一种面向SIMD扩展部件的向量化统一架构

Unified Vectorization Framework for SIMD Extensions
下载PDF
导出
摘要 随着多媒体应用的普及和高性能计算的需求,越来越多的处理器集成了SIMD扩展。为了针对不同SIMD扩展部件自动生成高效的向量化代码,设计了一套虚拟向量指令集,在此基础上构建了一种面向SIMD扩展部件的向量化统一架构。将输入程序通过向量识别等阶段转变为虚拟向量指令的中间表示,而后通过向量长度解虚拟化和指令集解虚拟化,将其转变为特定SIMD部件的向量指令集。在申威1600、DSP和Alpha上的实验结果表明:统一架构能够针对3种平台自动变换出高效的向量化代码,在DSP上的加速比要明显优于其它两种平台。 With the popularization of multimedia applications and the requirement of high performance computing,more and more processors are integrated with SIMD extensions. A virtual simd instruction set was designed for generating efficient simd code for different SIMD extension units, and a unification framework facing isomeric SIMD extension units was built. Program input can be transformed into intermediate representation of virtual simd instruction and then translated into simd instruction set of specific SIMD extensions through vector length and instruction set devirtualization solutions. The experimental results on SW1600, DSP and Alpha show that efficient simd code in three platforms is generated automatically on this framework and DSP has the better speedup than the other two platform.
出处 《计算机科学》 CSCD 北大核心 2014年第9期28-31,44,共5页 Computer Science
基金 核高基国家科技重大专项(2009ZX01036)资助
关键词 向量化 单指令多数据 解虚拟化 基本块 循环展开 Vectorization SIMD Devirtualization Basic block Loop unrolling
  • 相关文献

参考文献18

  • 1Peleg A, Weiser U. MMX Technology Extension to the IntelArchitecture[J]. IEEE/ACM International Symposium on Mi- croarchitecture, 1996,16 (4) : 42-50.
  • 2Intel 64-ia-32-architectures-software-developer-manual [EB/ OL]. http://www, intel, com/eontent/dam/www/public/us/ en/documents/manuals/64-ia-32-architectures-software-deve loper-manual-325462, pdf,September 2013.
  • 3Stewart J. An Investigation of SIMD instruction sets. School of Information Technology and Mathematical Sciences[R]. Univer- sity of Ballarat,2005.
  • 4Franchetti F, Kral S, Lorenz J, et al. Efficient utilization of SIMD extensions[J]. Proceedings of the IEEE, 2005, 93 (2): 409-425.
  • 5SC140 DSP Core Reference Manual. Freescale Semiconductor [EB/OL]. http:// catch, freescale, com/files/dsp/doc/ref man- ual/MNSC140CORE, pdf, 2004.
  • 6Fridman J,Greenfield Z. The Tiger SHARC DSP Architecture [J]. IEEE Micro, 2000,20(1): 66-76.
  • 7TMS320C6000 CPU and Instruction Set Reference Guild(Rev. F) [R]. Texas Instruments Inc. 2000.
  • 8Allen R, Kennedy K. Optimizing Compilers for Modern Archi- tectures- A Dependence-based Approach[M]. US: Morgan Kaufmann Publishers, 2001.
  • 9Larsen S, Amarasinghe S. Exploiting superword level parallelism with multimedia instruction sets[C]// Proc of the ACM SIGP- LAN Conference on Programming Language Design and Imple- mentation. June 2000:145-156.
  • 10Kudriavtsev A, Kogge P. Generation of Permutations for SIMD Processors[C]//PLDI, 2006. Ottawa, Canada, 2006.

二级参考文献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