Accelerating C/C++ Pointer Analysis via Compiler-Based Offline Simplifications
TL;DR
This paper explores a new perspective: applying semantic-preserving compiler optimizations directly to intermediate representation (IR) before pointer analysis, which is modular, analysis-agnostic, and easily integrates with existing tools.
Abstract
Pointer analysis is a cornerstone of numerous static analysis applications, including compiler optimizations, slicing, bug detection, and verification. While offline simplification is a common approach to boosting performance, existing methods are often tightly coupled to specific analysis algorithms and limited to a set of simplification rules. This paper explores a new perspective: applying semantic-preserving compiler optimizations directly to intermediate representation (IR) before pointer analysis. This strategy is modular, analysis-agnostic, and easily integrates with existing tools. We conduct an empirical study using diverse programs and three pointer analyses. The results show substantial performance gains---up to 3.14x speedup and 1.94x memory reduction---while precision remains largely unchanged. We also analyze the trade-offs between optimization overhead and analysis speedup, quantify changes in IR structure, assess the characteristics of optimization configurations, and identify promising directions for future research.