Adapting Petri Nets Reductions to Promela Specifications

Abstract

Abstract. The interleaving of concurrent processes actions leads to the well-known combinatorial explosion problem. Petri nets theory provides some structural reductions to tackle this phenomenon by agglomerating sequences of transitions into a single atomic transition. These reductions are easily checkable and preserve deadlocks, Petri nets liveness and any LTL formula that does not observe the modified transitions. Furthermore, they can be combined with other kinds of reductions such as partial-order techniques to improve the efficiency of state space reduction. We present in this paper an adaptation of these reductions for Promela specifications and propose simple rules to automatically infer atomic steps in the Promela model while preserving the checked property. We demonstrate on typical examples the efficiency of this approach and propose some perspectives of this work in the scope of software model checking.

LTL under reductions with weaker conditions than stutter-invariance

“…The structural reductions rules we used in the performance evaluation are defined on Petri nets where the literature on the subject is rich [1,19,6,9,2,22]. However there are other formalism where similar reduction rules have been defined such as [16] using "atomic" blocks in Promela, transaction reductions for the widely encompassing intermediate language PINS of LTSmin [13], and even in the context of multi-threaded programs [7]. All these approaches are structural or syntactic, they are run prior to model-checking per se.…”