Automatic Verification of C and Java Programs: SV-COMP 2019

Abstract

This report describes the 2019 Competition on Software Verification (SV-COMP), the 8 th edition of a series of comparative evaluations of fully automatic software verifiers for C programs, and now also for Java programs. The competition provides a snapshot of the current state of the art in the area, and has a strong focus on replicability of its results. The repository of benchmark verification tasks now supports a new, more flexible format for task definitions (based on YAML), which was a precondition for conveniently benchmarking Java programs in the same controlled competition setting that was successfully applied in the previous years. The competition was based on 10 522 verification tasks for C programs and 368 verification tasks for Java programs. Each verification task consisted of a program and a property (reachability, memory safety, overflows, termination). SV-COMP 2019 had 31 participating verification systems from 14 countries.

Proving Non-Termination via Loop Acceleration

“…To ensure that the program is initially well formed, we just allow integers, addition, subtraction, and multiplication in the original program. 3 While our approach uses program transformations that may introduce further operations like division and exponentials, these transformations preserve well-formedness. We formalize our contributions in terms of processors.…”

“…arithmetic, e.g., "f(1 + 2) = f(3)" holds. 3 One could also allow expressions like 1 2 · x 2 + 1 2 · x in the initial program, as long as every arithmetic expression maps integers to integers. maps integer programs to integer programs.…”

An Efficient Parametric Linear Programming Solver and Application to Polyhedral Projection

“…In this experiment we used the analyzer Pagai [12] and SV-COMP benchmarks [2]. We randomly selected C programs from the category of Concurrency Safety, Table 2: Performance on SV-COMP benchmarks.…”

Extending DIVINE with Symbolic Verification Using SMT

“…When a program contains a branch which depends on a symbolic value, both outcomes might be possible. 1 To capture such behavior in the transformed program, we introduce a nondeterministic choice and execute both branches. We take advantage of the fact that DIVINE is already capable of handling nondeterminism.…”