Spatial and Temporal Decomposition for Faster Translation Validation

Translation validation is a critical tool in program analysis: when a program 𝑃 is transformed into a new program 𝑃 ′ , translation validation asks whether 𝑃 and 𝑃 ′ have the same semantics. It serves as a middle ground between compiler testing and formal verification, capable of proving that a particular run of a compiler produced correct results. However, one bottleneck holds back wider adoption of translation validation: performance. State-of-the-art tools frequently time out or require extensive manual engineering to adapt to specific use cases. In this paper, we propose a new approach to improving the scalability of translation validation by decomposing the problem along two axes. Our primary contribution is a method for harnessing compiler information to extract subprograms whose equivalence result implies equivalence of the overall transformation (the spatial axis). We augment this method by utilizing compiler information to dynamically group transformation passes for validation (the temporal axis). Our evaluation demonstrates that this approach validates 10% of translations that existing approaches fail to validate, and speeds up validation by up to 2.4×.