Runtime Checking for Program Verification Systems
Karen Zee, Viktor Kunčak, Martin Rinard · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · 2007
One of the goals of program verification is to show that a program conforms to a specification written in a formal logic. Oftentimes, this process is hampered by errors in both the program and the specification. The time spent in identifying and eliminating these errors can even dominate the final verification effort. A runtime checker that can evaluate formal specifications can be extremely useful for quickly identifying such errors. Such a checker also enables verification approaches that combine static and dynamic program analyses. Finally, the underlying techniques are also useful for executing expressive high-level declarative languages. This paper describes the run-time checker we are developing in the context of the Jahob verification system. One of the challenges in building a runtime checker for a program verification system is that the language of invariants and assertions is designed for simplicity of semantics and tractability of proofs, and not for run-time checking. Some of the more challenging constructs include existential and universal quantification, set comprehension, specification variables, and formulas that refer to past program states. In this paper, we describe how we handle these constructs in our runtime checker, and describe several directions for future work.