Higher-Order and Symbolic Computation

, Volume 18, Issue 1, pp 15–49

Transformational Derivation of an Improved Alias Analysis Algorithm

Article

DOI: 10.1007/s10990-005-7005-6

Cite this article as:
Goyal, D. Higher-Order Symb Comput (2005) 18: 15. doi:10.1007/s10990-005-7005-6
  • 62 Downloads

Abstract

In this paper we use a program transformational approach to obtain an asymptotically improved may-alias analysis algorithm. We derive an O(N3) time algorithm for computing an intra-procedural flow sensitive may-alias analysis, where N denotes the number of edges in the program control flow graph (CFG). Our algorithm improves the previous O(N5) time algorithm by Hind et al. [19]. Our time complexity improvement comes without any deterioration in space complexity. We also show that for a large subclass of programs in which the in-degree and out-degree of all CFG nodes is bounded by a constant, our algorithm is linear in the sum of the number of edges in the CFG of the program and the size of the output, i.e., the size of the computed alias information, and is therefore asymptotically optimal. Our transformational algorithm derivation technique also leads to a simplified yet precise analysis of time complexity.

Keywords

may-alias analysis pointer analysis control flow graph dominated convergence finite differencing workset algorithm alias graph sparse evaluation graph continuity input-bounded output-bounded 

Copyright information

© Springer Science + Business Media, Inc. 2005

Authors and Affiliations

  1. 1.Calypto Design Systems, Inc.Santa ClaraUSA

Personalised recommendations