Context Sensitivity without Contexts: A Cut-Shortcut Approach to Fast and Precise Pointer Analysis

Explore a groundbreaking approach to pointer analysis for Java in this 21-minute conference talk from PLDI 2023. Discover the innovative Cut-Shortcut method that achieves context sensitivity without using contexts, offering a faster and more precise alternative to traditional pointer analysis techniques. Learn how this approach simulates the effect of cloning methods under different contexts by cutting off edges that introduce precision loss and adding shortcut edges in the pointer flow graph. Understand the three general program patterns and algorithms developed to safely implement this technique, and examine the formal proof of its soundness. Gain insights into the impressive evaluation results, where Cut-Shortcut outperforms context insensitivity in speed for most programs while maintaining high precision comparable to context sensitivity. Delve into the potential impact of this novel approach on future developments in pointer analysis for Java.