An Overview of Symbolic Analysis Techniques Needed for the Effective Parallelization of the Perfect Benchmarks

1 August 1994

conference paper
Published by Institute of Electrical and Electronics Engineers (IEEE)

Vol. 2, 233-238
https://doi.org/10.1109/icpp.1994.59

Abstract

We have identified symbolic analysis techniques that will improve the effectiveness of parallelizing Fortran compilers, with emphasis upon data dependence analysts. We have done this by comparing the automatically and manually parallelized versions of the Perfect Benchmarks¿. The techniques include: symbolic data dependence tests for nonlinear expressions, constraint propagation, array summary information, and run time tests.

Keywords

This publication has 13 references indexed in Scilit:

Automatic Array Privatization
Published by Springer Nature ,2001
An Overview of Symbolic Analysis Techniques Needed for the Effective Parallelization of the Perfect Benchmarks
Published by Institute of Electrical and Electronics Engineers (IEEE) ,1994
Abstract debugging of higher-order imperative languages
Published by Association for Computing Machinery (ACM) ,1993
Automatic program parallelization
Proceedings of the IEEE, 1993
Delinearization
Published by Association for Computing Machinery (ACM) ,1992
Beyond induction variables
Published by Association for Computing Machinery (ACM) ,1992
Constant propagation with conditional branches
ACM Transactions on Programming Languages and Systems, 1991
Interprocedural constant propagation
Published by Association for Computing Machinery (ACM) ,1986
Automatic discovery of linear restraints among variables of a program
Published by Association for Computing Machinery (ACM) ,1978
Compiler Analysis of the Value Ranges for Variables
IEEE Transactions on Software Engineering, 1977