Abstract

Parallelization of recursive programs is still an open problem today, lacking suitable and precise static analyses. This article presents a novel data flow analysis framework based on push-down (a.k.a. algebraic) transducers. This technique achieves a global and precise description of the data flow and discovers important semantic properties of the programs. Future fruitful research, at the crossroad of program analysis and formal language theory, is also hinted to. Keywords: data flow analysis, recursive programs, context-free languages, algebraic transducers 1 Introduction To optimize the compilation of imperative programs, powerful analyses of memory accesses have been crafted to precisely pinpoint which instances of assignments access which memory cells, and in which order. For instance, alias analyses compute pairs of conflicting memory accesses, and dependence analyses express access conflicts with respect to program statements. Our purpose goes beyond that: For a given use (a ...

Citations