Incremental Critical Path Generation for Dynamic Graphs

Che Chang, Cheng-Hsiang Chiu, Boyang Zhang, Tsung-Wei Huang

Abstract: Incremental Critical Path Generation (CPG) is crucial for static timing analysis (STA) applications to incrementally analyze critical paths and validate timing constraints. However, current state-of-the-art incremental CPG algorithms can only handle static graphs where the graph topology does not change. To solve this problem, we introduce an efficient incremental CPG algorithm that can handle dynamic graphs. Compared to existing methods, our algorithm can identify more scenarios to efficiently reuse paths from the previous CPG while discarding unnecessary ones after the graph topology is changed. Our algorithm is up to 8.2 × faster than a state-of-the-art timer when generating two million paths on a large design.