Context-sensitive interprocedural points-to analysis in the presence of function pointers

Abstract

This paper reports on the design, implementation, and empirical results of a new method for dealing with the aliasing problem in C. The method is based on approximating the points-to relationships between accessible stack locations, and can be used to generate alias pairs, or used directly for other analyses and transformations. Our method provides context-sensitive interprocedural information based on analysis over invocation graphs that capture all calling contexts including recursive and mutually-recursive calling contexts. Furthermore, the method allows the smooth integration for handling general function pointers in C. We illustrate the effectiveness of the method with empirical results from an implementation in the McCAT optimizing/parallelizing C compiler.

An empirical examination of the prevalence of inhibitors to the parallelizability of open source software systems

“…It has always been assumed that for accurate interprocedural analysis, function alias analysis is a very important step that should be always seriously considered for better results (Emami et al 1994;Zhang and Ryder 1994). For safe analysis and parallelism, the set of possible targets of a function pointer call must be determined.…”

Probabilistic Alias Analysis of Executable Code

“…The time and space complexities of our algorithm are similar to those in previous researches [7,12,21,32,39], but there are some distinguishing characteristics described below. Assume that the analyzed target architecture has R registers and that the analyzed program consists of N insn instructions.…”

“…Many research efforts have focused on alias analysis in high-level programming languages [2,5,8,11,12,17,21,31,36]. Generally, the strategies of static (compile-time) and dynamic (run-time) analyses are separately adopted to collect appropriate data.…”

Dependence analysis for C programs with combinability of dataflow facts under consideration

“…To more exhaustively and more efficiently avoid invalid combination of dataflow facts, this paper firstly extends Emami et al's work [9] , which presents a new flow-sensitive and context-insensitive points-to analysis method. With the method, all possible calling contexts of a function can be clearly and conservatively described.…”