SIMD指令集设计空间的形式化描述

计算机科学 ›› 2013, Vol. 40 ›› Issue (6): 32-36.

SIMD指令集设计空间的形式化描述

李春江,徐颖,黄娟娟,杨灿群

国防科学技术大学计算机学院计算机研究所长沙410073;国防科学技术大学计算机学院计算机研究所长沙410073;国防科学技术大学计算机学院计算机研究所长沙410073;国防科学技术大学计算机学院计算机研究所长沙410073

出版日期:2018-11-16 发布日期:2018-11-16
基金资助:
本文受国家自然科学基金项目(61170046,5)资助

Formal Description of Design Space of SIMD Instruction Sets

LI Chun-jiang,XU Ying,HUANG Juan-juan and YANG Can-qun

Online:2018-11-16 Published:2018-11-16

摘要/Abstract

摘要： SIMD(Single-Instruction-Multiple-Data)并行体系结构在现代处理器体系结构中扮演非常重要的角色。SIMD指令集已经成为处理器指令集中重要的子集。SIMD结构和指令集实现了短向量并行处理能力,SIMD指令集实现了对多种数据类型、多种操作模式的支持。采用形式化的方法,描述SIMD指令集的设计空间,从多个正交的维度刻画SIMD指令集的设计,基于此详细讨论了SIMD指令集的设计问题。该形式化方法有益于对SIMD指令集体系结构的分析和设计。

关键词: SIMD,指令集,设计空间,形式化描述

Abstract: SIMD (Single-Instruction-Multiple-Data) parallel architecture plays an important role in the architecture of modern processors．The instruction set of SIMD becomes the most important subset of the whole instruction set of a processor．SIMD architecture and instruction set provides the parallel processing ability for short vectors．The SIMD instruction set supports multiple data types and multiple operations．This paper described the design space of SIMD instruction set using formal method,which depicts SIMD instruction set from multiple orthogonal dimensions.Then the discussion about the design of SIMD instruction set was presented．The formal method will be beneficial to analyze and design the SIMD instruction set architecture.

Key words: SIMD,Instruction set,Design space,Formal description

李春江,徐颖,黄娟娟,杨灿群. SIMD指令集设计空间的形式化描述[J]. 计算机科学, 2013, 40(6): 32-36. https://doi.org/

LI Chun-jiang,XU Ying,HUANG Juan-juan and YANG Can-qun. Formal Description of Design Space of SIMD Instruction Sets[J]. Computer Science, 2013, 40(6): 32-36. https://doi.org/

参考文献

[1] Maruyama T,Motokurumada T,Morita K,et al．Past,Present and Futures of SPARC64Processors[J]．FUJITSU Sci．Tech．J.,2011,47(2):130-135
[2] Firasta N,Buxton M,Jinbo P,et al．Intel AVX:New Frontiersin Performance Improvements and Energy Efficiency[M]．Intel white paper,2008
[3] Intel 64and IA-32Architectures Software Developer’s Manual,Combined Volumes:1,2A,2B,3A and 3B[M]． Intel Corporation,May 2011
[4] Hennessy J L,Patterson D A．Computer Architecture:A Quantitative Aproach(3rd ed)[M]．San Francisco:Morgan Kaufmann Publish,2002
[5] Nuzman D,Henderson R．Multi-platform Auto-vectorization[C]∥Proc．of the 4th Annual International Symposium on Code Gene-ration and Optimization (CGO)．March 2006

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed