Lazy abstraction

Abstract

One approach to model checking software is based on the abstract-check-refine paradigm: build an abstract model, then check the desired property, and if the check fails, refine the model and start over. We introduce the concept of lazy abstraction to integrate and optimize the three phases of the abstract-check-refine loop. Lazy abstraction continuously builds and refines a single abstract model on demand, driven by the model checker, so that different parts of the model may exhibit different degrees of precision, namely just enough to verify the desired property. We present an algorithm for model checking safety properties using lazy abstraction and describe an implementation of the algorithm applied to C programs. We also provide sufficient conditions for the termination of the method.

Is Lazy Abstraction a Decision Procedure for Broadcast Protocols?

“…We recall the lazy abstraction algorithm(LA) from [1]. The algorithm iteratively explores the abstract state space by constructing an abstract reachability tree.…”

“…Lazy abstraction [1] is an interesting verification method that deserves a study on its own right. It defines the de-facto standard for verification based on the scheme coined counterexample-guided abstraction refinement (CEGAR) in [2].…”

“…The high-level formulation of lazy abstraction given in [1] specifies no control for building up the abstract reachability tree. Strictly speaking, our positive answer refers to a version of lazy abstraction with control.…”

“…The control (for building up the abstract reachability tree) implements a breadth-first strategy. This corresponds to the default choice in the implementation of lazy abstraction, e.g., in the tool BLAST [1,3].…”

“…Systems with finite trace equivalence include finite-state systems but also timed automata; lazy abstraction was shown complete for this class in [1]. The completeness proof exploits the termination guarantee of the finite-quotient construction in [11].…”