期刊文献+

一种基于汇编代码的单重循环向量化方法

A Single Loop Vectorization Method Based on Assemble Code
下载PDF
导出
摘要 Through loops vectorization in instruction sequence, the vector power provided by hardware can be fully utilized. This paper analyzes the RISC instructton set, and presents a single loop vectorization method that is based on assemble code, it can efficiently detect single loops in instruct sequence and vectorize them. Through loops vectorization in instruction sequence, the vector power provided by hardware can be fully utilized. This paper analyzes the RISC instruction set, and presents a single loop vectorization method that is based on assemble code, it can efficiently detect single loops in instruct sequence and vectorize them.
出处 《计算机科学》 CSCD 北大核心 2003年第4期116-117,124,共3页 Computer Science
基金 自然科学基金支持(60173040)
关键词 微处理器 设计 汇编代码 单重循环向量化方法 指令源代码 计算机 Vcctorization, Assemble code, Loop, RISC
  • 相关文献

参考文献8

  • 1陆洪毅,沈立,赵学秘,王蕾,戴葵,王志英.一种高性能的嵌入式微处理器:银河TS-1[J].电子学报,2002,30(11):1668-1671. 被引量:2
  • 2宋辉 戴葵 王志英.基于银河TS-1的高性能量子计算[J].电子学报,2002,:10-10.
  • 3Lopez D, Valero M, et al. Increasing Memory Bandwidth with Width Bus: Compiler, Hardware and Performance Trade-offs.ACM, 1997.
  • 4Burger D, et al. Memory Bandwidth Limitations of Future Microprocessors. ACM, 1998.
  • 5Saulsbury A, et al. Missing the Memory Wall: The Case for Processor/Memory Integration. ACM, 1996.
  • 6Quintana F, Espasa R, Valero M. Performance Advantages of Merging Instruction- And Data-Level Parallelism (Extended Version).
  • 7Espasa R, Valero M. Exploiting Instruction and Data Level Parallelism in Future High Performance Processors.
  • 8Wall D W. Limits of Instruction-Level Parallelism: [David W.Wall, WRL Technical Note TN-15]. Western Research Laboratory, Dec. 1990.

二级参考文献1

共引文献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