计算机科学 ›› 2015, Vol. 42 ›› Issue (5): 194-199.doi: 10.11896/j.issn.1002-137X.2015.05.039
李朋远,赵荣彩,高 伟,张庆花
LI Peng-yuan, ZHAO Rong-cai, GAO Wei and ZHANG Qing-hua
摘要: 随着SIMD扩展部件的迅速发展,自动向量化工具已逐渐成熟。现阶段的工具 能 对连续访存程序进行较好的处理,然而,大部分非连续访存的多媒体程序并不能被转换为高效的向量化代码。提出并实现了一种支持跨幅访存的向量化代码生成方法,其利用目标系统已有的基本数据处理指令实现多个向量间的任意重组来解决含有非连续访存语句的向量化代码生成问题。经过实验分析和验证,提出的代码生成方法能够将含有跨幅访存的语句转化为面向目标系统的高效向量化代码,以提高程序执行效率。
[1] Hassaballah M,Omran S,Mahdy Y B.A review of SIMD multimedia extensions and their usage in scientific and engineering applications[J].The Computer Journal,2008,51(6):630-649 [2] Allen R,Kennedy K.Automatic translation of Fortran programs to vector form[J].ACM Transactions on Programming Languages and Systems (TOPLAS),1987,9(4):491-542 [3] Zima H,Chapman B.Supercompilers for parallel and vectorcomputers[M].ACM,1990 [4] Larsen S,Amarasinghe S.Exploiting superword level parallelism with multimedia instruction sets[M].ACM,2000 [5] Griffith A.GCC:the complete reference[M].McGraw-Hill,lnc.,2002 [6] Lattner C,Adve V.The LLVM compiler framework and infrastructure tutorial[M]∥Languages and Compilers for High Performance Computing.Springer Berlin Heidelberg,2005:15-16 [7] Lin M,Yu Z,Zhang D,et al.Retargeting the open64 compiler to powerpc processor[C]∥International Conference on Embedded Software and Systems Symposia,2008(ICESS Symposia’08).IEEE,2008:152-157 [8] Naishlos D.Autovectorization in GCC[C]∥Proceedings of the 2004 GCC Developers Summit.2004:105-118 [9] Nuzman D,Zaks A.Autovectorization in GCC-two years later[C]∥Proceedings of the 2006 GCC Developers Summit.2006:145-158 [10] Nuzman D,Henderson R.Multi-platform auto-vectorization[C]∥Proceedings of the International Symposium on Code Generation and Optimization.IEEE Computer Society,2006:281-294 [11] Tanaka H,Ota Y,Matsumoto N,et al.A new compilation technique for SIMD code generation across basic block boundaries[C]∥Proceedings of the 2010 Asia and South Pacific Design Automation Conference.IEEE Press,2010:101-106 [12] Bastoul C.Code generation in the polyhedral model is easier than you think[C]∥Proceedings of the 13th International Conference on Parallel Architectures and Compilation Techniques.IEEE Computer Society,2004:7-16 [13] Ramanarayanan R,Gupta M,Chakraborty S S,et al.Harnessing partial vectorization in Open64 compiler[C]∥2014 IEEE International Advance Computing Conference (IACC).IEEE,2014:813-824 [14] 魏帅,赵荣彩,姚远.面向国产 CPUSW-1600 的向量重组[J].计算机应用与软件,2012,28(11):230-233 [15] Wu P,Eichenberger A E,Wang A.Efficient simd code generation for runtime alignment and length conversion[C]∥International Symposium on Code Generation and Optimization,2005(CGO 2005).IEEE,2005:153-164 [16] Allen R,Kennedy K.Optimizing compilers for modern architectures:a dependence-based approach[M].San Francisco:Morgan Kaufmann,2002 [17] Leupers R.Code selection for media processors with SIMD instructions[C]∥Proceedings of the Conference on Design,Automation and Test in Europe.ACM,2000:4-8 [18] Sreraman N,Govindarajan R.A vectorizing compiler for multimedia extensions[J].International Journal of Parallel Programming,2000,28(4):363-400 |
No related articles found! |
|