Commutativity Simplifies Proofs of Parameterized Programs

Abstract

Commutativity has proven to be a powerful tool in reasoning about concurrent programs. Recent work has shown that a commutativity-based reduction of a program may admit simpler proofs than the program itself. The framework of lexicographical program reductions was introduced to formalize a broad class of reductions. Approaches based on this framework, however, were limited to program models with a fixed number of threads. In this paper, we show that it is possible to define an effective parametric family of program reductions that can be used to find simple proofs for parameterized programs, i.e., programs with an unbounded number of threads. We show that reductions are indeed useful for the simplification of proofs of parameterized programs, in a sense that can be made precise: A reduction of a parameterized program may admit a proof which uses fewer or less sophisticated ghost variables. The reduction may therefore be within reach of an automated verification technique, even when the original parameterized program is not. We introduce a notion of reductions for parameterized programs such that the reduction $\mathcal{R}$ of a parameterized program $\mathcal{P}$ is again a parameterized program (the thread template of $\mathcal{R}$ is obtained by source-to-source transformation of the thread template of $\mathcal{P}$). Consequently, existing techniques for the verification of parameterized programs can be directly applied to $\mathcal{R}$ instead of $\mathcal{P}$. We define an appropriate family of pairwise preference orders which can be used to produce different lexicographical reductions. To determine whether this theoretical foundation amounts to a usable solution in practice, we have implemented the approach, based on a recently proposed framework for parameterized program verification. The results of our preliminary experiments on a representative set of examples are encouraging.