Abstract:Software model checking is a challenging problem, and generating relevant invariants is a key factor in proving the safety properties of a program. Program invariants can be obtained by various approaches, including lightweight procedures based on data-flow analysis and intensive techniques using Craig interpolation. Although data-flow analysis runs efficiently, it often produces invariants that are too weak to prove the properties. By contrast, interpolation-based approaches build strong invariants from interpolants, but they might not scale well due to expensive interpolation procedures. Invariants can also be injected into model-checking algorithms to assist the analysis. Invariant injection has been studied for many well-known approaches, including k-induction, predicate abstraction, and symbolic execution. We propose an augmented interpolation-based verification algorithm that injects external invariants into interpolation-based model checking (McMillan, 2003), a hardware model-checking algorithm recently adopted for software verification. The auxiliary invariants help prune unreachable states in Craig interpolants and confine the analysis to the reachable parts of a program. We implemented the proposed technique in the verification framework CPAchecker and evaluated it against mature SMT-based methods in CPAchecker as well as other state-of-the-art software verifiers. We found that injecting invariants reduces the number of interpolation queries needed to prove safety properties and improves the run-time efficiency. Consequently, the proposed invariant-injection approach verified difficult tasks that none of its plain version (i.e., without invariants), the invariant generator, or any compared tools could solve.

Abstract Interpretation Based Correct Verification for Functional Invariant

Infer Precise Program Invariant Using Abstract Interpretation with Recurrence Solving.

Program Verification By Reduction To Semi-Algebraic Systems Solving

Invariant Detection with Program Verification Tools

Dynamically Discovering Functional Likely Program Invariants Based on Relational Database Theory

Formal Verification for C Program

Verifying Programs Using Abstraction and Theorem Proving.

Structural Invariants for Parametric Verification of Systems with Almost Linear Architectures

Detection Techniques of Program Invariants

A Method of Dynamically Discovering Functional Dependence Program Invariants

Automatic Invariant Synthesis for Arrays in Simple Programs

VST-A: A Foundationally Sound Annotation Verifier

Using Dynamic Analysis to Generate Disjunctive Invariants

Post-condition-Directed Invariant Inference for Loops over Data Structures.

Detecting Functional Dependence Program Invariants Based on Data Mining

Dynamical Detecting Technique of Nonfunctional Dependence Program Invariant

Finding Polynomial Loop Invariants for Probabilistic Programs

Augmenting Interpolation-Based Model Checking with Auxiliary Invariants (Extended Version)

Synthesizing Short-Circuiting Validation of Data Structure Invariants

Abstraction-based model checking programs

Recent Advances in Program Verification Through Computer Algebra.