Node listings applied to data flow analysis

  • Authors:
  • K. W. Kennedy

  • Affiliations:
  • Rice University

  • Venue:
  • POPL '75 Proceedings of the 2nd ACM SIGACT-SIGPLAN symposium on Principles of programming languages
  • Year:
  • 1975

Quantified Score

Hi-index 0.03

Visualization

Abstract

A new approach to global program data flow analysis which constructs a "node listing" for the control flow graph is discussed and a simple algorithm which uses a node listing to determine the live variables in a program is presented. This algorithm combined with a fast node listing constructor due to Aho and Ullman has produced an 0(n log n) algorithm for live analysis. The utility of the node-listing method is demonstrated by an examination of the class of graphs for which "short" listings exist. This class is quite similar to the class of graphs for "understandable" programs.