Incremental and Modular Context-sensitive Analysis

Context-sensitive global analysis of large code bases can be expensive, which make its use impractical during software development. However, there are many situations in modifications small and isolated within a few components, it is desirable to reuse as much possible previous results. This has been achieved date through incremental fixpoint algorithms that achieve cost reductions at fine levels granularity, such changes program lines. these fine-grained techniques not directly applicable modular programs, nor they designed take advantage structures. paper describes, implements, evaluates an algorithm performs efficient context-sensitive incrementally on partitions programs. The experimental results show the proposed shows significant improvements, both time memory consumption, when compared existing non-modular, fine-grain techniques. Furthermore, thanks inter-modular propagation information, our also outperforms traditional even analyzing from scratch.