A general algorithm for data dependence analysis

1 January 1992

conference paper
Published by Association for Computing Machinery (ACM)

p. 292-302
https://doi.org/10.1145/143369.143422

Abstract

With the development of ever more sophisticated data flow analysis algorithms, traditional data dependence tests based on elementary loop information will not be sufficient in the future. In this paper, quite general algorithms are presented for solving integer linear programming problems. While the properly so called problem solution is performed by a standard algorithm (the dual all integer algorithm), preliminary problem reduction techniques not only serve as a powerful tool for preparing this latter step, but also are often sufficient for solving exactly the data dependence problem.

Keywords

TRADITIONAL DATA DEPENDENCE TEST
GENERAL ALGORITHM
STANDARD ALGORITHM
PRELIMINARY PROBLEM REDUCTION TECHNIQUE
INTEGER LINEAR PROGRAMMING PROBLEM
DATA DEPENDENCE ANALYSIS
ELEMENTARY LOOP INFORMATION
SOPHISTICATED DATA FLOW ANALYSIS
PROBLEM SOLUTION
INTEGER ALGORITHM
DATA DEPENDENCE PROBLEM
DATA FLOW ANALYSIS
GENERIC ALGORITHM
PARALLEL PROCESSING

This publication has 0 references indexed in Scilit: