Benefits of Bounded Model Checking at an Industrial Setting

Abstract

The usefulness of Bounded Model Checking (BMC) based on propositional satisfiability (SAT) methods for bug hunting has already been proven in several recent work. In this paper, we present two industrial strength systems performing BMC for both verification and falsification. The first is Thunder, which performs BMC on top of a new satisfiability solver, SIMO. The second is Forecast, which performs BMC on top of a BDD package. SIMO is based on the Davis Logemann Loveland procedure (DLL) and features the most recent search methods. It enjoys static and dynamic branching heuristics, advanced back-jumping and learning techniques. SIMO also includes new heuristics that are specially tuned for the BMC problem domain. With Thunder we have achieved impressive capacity and productivity for BMC. Real designs, taken from Intel's Pentium © 4, with over 1000 model variables were validated using the default tool settings and without manual tuning. In Forecast, we present several alternatives for adapting BDD-based model checking for BMC. We have conducted comparison of Thunder and Forecast on a large set of real and complex designs and on almost all of them Thunder has demonstrated clear win over Forecast in two important aspects: capacity and productivity.

NuSMV 2: An OpenSource Tool for Symbolic Model Checking

“…SIM is a SAT solver based on the Davis-Logemann-Loveland procedure. The features provided by SIM can produce dramatic speed-ups in the overall performances of the SAT checker, and thus of the whole system (see e.g., [19,10] for a discussion). It is currently under development a generic interface to SAT solvers to allow for the use of new state of the art SAT solvers like e.g.…”

“…The main novelty in NuSMV2 is the integration of model checking techniques based on propositional satisfiability (SAT) [4]. SAT-based model checking is currently enjoying a substantial success in several industrial fields (see, e.g., [10], but also [5]), and opens up new research directions. BDD-based and SAT-based model checking are often able to solve different classes of problems, and can therefore be seen as complementary techniques.…”

A survey of recent advances in SAT-based formal verification

“…A recent comprehensive analysis with respect to the performance and capacity of BMC is presented in [31]. The authors compare Intel's BDD-based model checker, Forecast (adapted for BMC), with a SAT-based BMC tool, Thunder, on several benchmarks taken from Intel's Pentium 4 processor.…”

“…Dramatic improvements in SAT solver technology over the past decade have led to the development of several powerful SAT solvers [45,71,77,105]. Verification methods based on these solvers have been shown to push the envelope of functional verification in terms of both capacity and efficiency, as reported in several academic and industrial case studies [4,16,19,31]. This has fueled further interest and intense research activity in the area of SAT-based verification.…”

Reference model based RTL verification: an integrated approach

“…Our initialization technique, determines the legal set of initial states under the constraints specified by B and E. The problem of determining the initial state of the combined model is formulated as a satisfiability constraint and we present it to a bounded model checker (BMC) [3,4,6]. We specify a pair of mutually contradicting invariant constraints for each state element in the combined model and attempt to falsify each of them at the end of a certain predetermined bound,b.…”