RESEARCH AND DEVELOPMENT OF A PARALLELIZING COMPILER FOR ARBITRARY GRAIN SIZES AND GENERAL PARARLLEL ARCHITECTURES

任意粒度并行编译器和通用并行架构的研究与开发

• 批准号: 07680354
• 负责人: TSUDA Takao
• 金额: $ 1.6万
• 依托单位: HIROSHIMA CITY UNIVERSITY (1996)Kyoto University (1995)
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (C)
• 财政年份: 1995
• 资助国家: 日本
• 起止时间: 1995 至 1996
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-07680354/
• 关键词: OPTIMIZATION; COMPILER; VECTORIZATION; PARALLELIZATION; GRANULAITY; DEPENDENCE ANALYSIS; ALIAS ANALYSIS; TASK GRAPH

On top of the vectorizing compiler V-Pascal that had been developed by the Tsuda laboratory of Kyoto University over ten years, automatic parallelizing functions were added, together with various features. (1) Powerful ability to detect parallelism, (2) facility to find optimal grain sizes, (3) enhancing the locality of data references by means of data partitioning and loop transformation, and (3) minimizing machine-dependences in automatic parallelization by abstraction of parallel architectures are the main difficulties that remain to be resolved. In this project, new algorithms were developed for alias analyzes to have more precision in the analysis of dependeces due to pointer references. Alias analyzes in the presence of pointer variables that point to memory area that has been assigned by heap are especially difficult, but the result of this project gives an important clue to the problem. This partially answers the question (1) above. Concerning the question (2), a special view mechanism has been considered so that the results of dependence analyzes can be displayd at an arbitrary hierarchical level, namely, either coarse-grain level or fine-grain level. This makes it easier to check on, and choose, optimal parallelization.

在京都大学津田实验室十多年来开发的向量化编译器V-Pascal的基础上，增加了自动并行化功能，并增加了各种功能。(1)强大的并行检测能力，(2)寻找最优粒度的能力，(3)通过数据分区和循环转换增强数据引用的局部性，(3)通过并行架构抽象最小化自动并行化中的机器依赖性是有待解决的主要困难。在这个项目中，为别名分析开发了新的算法，以便在由于指针引用而导致的依赖分析中具有更高的精度。在存在指针变量的情况下进行别名分析尤其困难，这些指针变量指向已由堆分配的内存区域，但是这个项目的结果为这个问题提供了重要的线索。这部分地回答了上面的问题(1)。在问题(2)中，考虑了一种特殊的视图机制，使得依赖分析的结果可以在任意层次上显示，即粗粒度或细粒度。这使得检查和选择最优并行化变得更加容易。

项目成果

Tetsutaro Uehara: ""An Automatic Vectorizing/Parallelizing Pascal Compiler V-Pascal Version 3"" Proceedings of the International Symposium on Parallel and Distributed Super-computing (Kyushu University, Fukuoka ; Sept.26-28). 206-213 (1995)

Tetsutaro Uehara：“自动矢量化/并行化 Pascal 编译器 V-Pascal 版本 3””并行和分布式超级计算国际研讨会论文集（九州大学，福冈；9 月 26-28 日）。

Dong Soo Han: "Non-Graph Based Approach on the Analysis of Pointers and Structures" IEICE Trans. Inf. & System. Vol.E80-D No.4. 362-370 (1997)

Dong Soo Han：“基于非图的指针和结构分析方法”IEICE Trans。

Takao Tsuda: "“Automatic Data Structure Detection in Pascal and Fortran 90"" Proceedings of the International Symposium on Parallel and Distributed Supercomputing(Kyushu University, Fukuoka; Sept. 26-28). 25-34 (1995)

Takao Tsuda：“Pascal 和 Fortran 90 中的自动数据结构检测”，并行和分布式超级计算国际研讨会论文集（九州大学，福冈；9 月 26-28 日）（1995 年）。

Dong Soo Han: "“Automatic Data Structure Detection in Pascal and Fortran 90"" Proceedings of the International Symposium on Parallel and Distributed Supercomputing (Kyushu University,Fukuoka;Sept.26-28). 25-34 (1995)

Dong Soo Han：“Pascal 和 Fortran 90 中的自动数据结构检测”，并行和分布式超级计算国际研讨会论文集（九州大学，福冈；9 月 26-28 日）（1995 年）。

Dong Soo Han: "“Non-Graph Based Approach on the Analysis of Pointers and Structures"" IEICE Trans.Inf.& System. Vol.E80-D,No.4. 362-370 (1997)

Dong Soo Han：“基于非图的指针和结构分析方法”，IEICE Trans.Inf.& System，Vol.E80-D，No.4 (1997)。