A Container-Usage-Pattern-Based Context Debloating Approach for Object-Sensitive Pointer Analysis
Explore a groundbreaking approach to improving object-sensitive pointer analysis efficiency in this 18-minute video presentation from OOPSLA2 2023. Delve into DebloaterX, a novel context-debloating pre-analysis method that identifies context-independent objects based on container-usage patterns. Learn how this technique significantly accelerates analysis speed, with average improvements of 19.3x for k=2 and 150.2x for k=3, while maintaining precision. Discover the implementation details, experimental results on Java benchmarks, and comparisons with state-of-the-art alternatives like Zipper and Conch. Gain insights into the potential for scaling up analysis to more complex programs and the negligible precision loss achieved through this innovative approach.