Simultaneous SAT-Based Model Checking of Safety Properties

Abstract

Abstract. We present several algorithms for simultaneous SAT (propositional satisfiability) based model checking of safety properties. More precisely, we focus on Bounded Model Checking and Temporal Induction methods for simultaneously verifying multiple safety properties on the same model. The most efficient among our proposed algorithms for model checking are based on a simultaneous propositional satisfiability procedure (SSAT for short), which we design for solving related propositional objectives simultaneously, by sharing the learned clauses and the search. The SSAT algorithm is fully incremental in the sense that all clauses learned while solving one objective can be reused for the remaining objectives. Furthermore, our SSAT algorithm ensures that the SSAT solver will never re-visit the same sub-space during the search, even if there are several satisfiability objectives, hence one traversal of the search space is enough. Finally, in SSAT all SAT objectives are watched simultaneously, thus we can solve several other SAT objectives when the search is oriented to solve a particular SAT objective first. Experimental results on Intel designs demonstrate that our new algorithms can be orders of magnitude faster than the previously known techniques in this domain.

Implicative Simultaneous Satisfiability and Applications

“…This modified search procedure is not described in detail in [19], we therefore couldn't re-implement it for a fair comparison (moreover, recall that we use a very different circuit representation). Our early experience with simultaneous solving of multiple POs by solving each PO in its cone of influence and re-using pervasive learned clauses compared to SSAT is reported in [13], and is negative. In fact, one of the main original motivations for introducing SSAT was to eliminate the overhead of computing the cones of each objective and managing the conflict clauses.…”

“…Recall that SSAT modifies the modern SAT solver's algorithm in a way that allows it to solve multiple proof objectives in one search ( [13], Section 5). The SSAT algorithm always maintains a PO literal, called the currently watched PO (CWPO), that the SAT search tries to falsify.…”

“…In the domain of SAT solving [3,14], several efficient approaches to solving multiple related objectives incrementally [26,27] or simultaneously [13] have been developed in the past, and due to its increasing importance this is an active research area. The two approaches are closely related, yet there are subtle fundamental differences between the two.…”

Synchronized Generation of Directed Tests

“…A great deal of work has been done to accelerate the SAT solving process during BMC [2][3][4]6]. Hooker [7] was the first to introduce the notion of incremental satisfiability based on DPLL algorithm.…”

Asynchronous Multi-core Incremental SAT Solving

“…A related work is Simultaneous SAT [19]. The interface of a simultaneous SAT solver is different from a conventional solver as for each formula in the input sequence a set of proof objectives can be given.…”