It is well-known that abstract interpreters can be systematically derived from their concrete counterparts using a “recipe,” but developing sound static analyzers remains a time-consuming task. Reducing the effort required and mechanizing the process of developing analyzers continues to be a significant challenge. Is it possible to automatically retarget an existing abstract interpreter for a new language?

We propose a novel technique to automatically derive abstract interpreters for various languages from an existing abstract interpreter. By leveraging partial evaluation, we specialize an abstract interpreter for a source language. The specialization is performed using the semantics of target languages written in the source language. Our approach eliminates the need to develop analyzers for new targets from scratch. We show that our method can effectively retarget an abstract interpreter for one language into a correct analyzer for another language.

I am pursuing my Master’s degree in Computer Science and Engineering at Seoul National University under the supervision of Professor Kwangkeun Yi.