摘要
该文给出了一种考虑了面向对象语言的多态和对象引用别名问题的对象方法间并行性的分析方法 .这种方法用于面向对象语言并行化中的并行性分析 .文中首先给出了一般化的方法调用模型 ,然后基于该模型给出了表达式化简、过程内和过程间分析的算法 ,该算法可以求出变量的定义和使用集合 ,用于并行性分析 .该文给出的简单例子即可以将该文的工作和相关的工作加以区别 .其技术已经在作者研制的 Java并行化编译器 JAPS-
Concurrent management and data distribution problems in parallel program design frustrate the construction of large scale software. In recent years, the researches to combine both parallel processing and object oriented techniques have got many result, including explicit parallel languages and automatic parallelizing compilers. Object oriented languages have many features that traditional languages have not. Java even has many new parallel facilities of modern languages. These features greatly frustrate the direct application of traditional techniques of parallelization to object oriented languages. This paper introduces a parallelism analysis method that considers the polymorphism and reference alias of object oriented languages. We firstly introduce a generalized model of method invocation. According to our generalized model we proposed an algorithm of parallelism analysis. Our algorithm includes three steps, which are variable expression reduction, intra procedure analysis and inter procedure analysis to compute define use sets. We compute the reference of function to get the possible set of objects in inter procedure analysis, which helps us to get more precise results than earlier work. In our algorithm the recursive procedure has been processed. The computing complexities of every algorithm are given in this paper. Finally, we use a simple enough example to compare our work with earlier works. JAPS II is a Java parallelizing compiler we have developed, which exploits and implements intra and inter object parallelism of serial Java programs. Its target architecture is NOW based distributed memory computer system. We have implemented the optimizations in JAPS II.
出处
《计算机学报》
EI
CSCD
北大核心
2002年第4期403-408,共6页
Chinese Journal of Computers
基金
国家"八六三"高技术研究发展计划项目(863 -3 0 6-ZT0 2 -0 3 -0 1)
国家自然科学基金 (6980 3 0 0 5 )资助
