Boogie: A Modular Reusable Verifier for Object-Oriented Programs

Barnett, Mike; Chang, Bor-Yuh Evan; DeLine, Robert; Jacobs, Bart; Leino, K. Rustan M.

doi:10.1007/11804192_17

Cited by 648 publications

(583 citation statements)

References 45 publications

Supporting

Mentioning

579

Contrasting

Unclassified

Order By: Relevance

“…free from assertion failures) then the kernel is guaranteed to be free from data races. The sequential program is checked using the Boogie verifier [2].…”

Section: Race Analysis Using Gpuverifymentioning

confidence: 99%

Warps and Atomics: Beyond Barrier Synchronization in the Verification of GPU Kernels

Bardsley

Donaldson

2014

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. We describe the design and implementation of methods to support reasoning about data races in GPU kernels where constructs other than the standard barrier primitive are used for synchronization. At one extreme we consider kernels that exploit implicit, coarse-grained synchronization between threads in the same warp, a feature provided by many architectures. At the other extreme we consider kernels that reduce or avoid barrier synchronization through the use of atomic operations. We discuss design decisions associated with providing support for warps and atomics in GPUVerify, a formal verification tool for OpenCL and CUDA kernels. We evaluate the practical impact of these design decisions using a large set of benchmarks, showing that warps can be supported in a scalable manner, that a coarse abstraction suffices for efficient reasoning about most practical uses of atomic operations, and that a novel, refined abstraction captures an important design pattern where atomic operations are used to compute unique array indices. Our evaluation revealed two previously unknown bugs in publicly available benchmark suites.

show abstract

“…free from assertion failures) then the kernel is guaranteed to be free from data races. The sequential program is checked using the Boogie verifier [2].…”

Section: Race Analysis Using Gpuverifymentioning

confidence: 99%

Warps and Atomics: Beyond Barrier Synchronization in the Verification of GPU Kernels

Bardsley

Donaldson

2014

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…We annotated the loop of Figure 1 with this predicate as a candidate loop invariant and gave it to the Boogie [7] program checker. Boogie was successfully able to prove that I was indeed a loop invariant and was able to show that the assertion holds.…”

Section: Finding Invariants For the Examplementioning

confidence: 99%

“…In particular, if L ≡ while E do S is a while program, then to check {P }L{Q}, Hoare logic tells us that we need a predicate I such that P ⇒ I, {I ∧ E}S{I}, and I ∧ ¬E ⇒ Q. Such a predicate I is called an inductive invariant or simply an invariant of the loop L. Once we have obtained invariants for all the loops, then standard techniques can generate program proofs [7]. We first focus our attention on invariants in the theory of linear arithmetic:…”

Section: Preliminariesmentioning

confidence: 99%

“…We use Havoc [5] to generate BoogiePL programs from C programs annotated with candidate invariants. Next, Boogie [7] verification condition generator operates on the BoogiePL programs to check the candidate invariants by passing the verification conditions to Z3 theorem prover [45]. All experiments were performed on a 2.67GHz Intel Xeon processor system with 8 GB RAM running Windows 7 and MATLAB R2010b.…”

Section: Experimental Evaluationmentioning

confidence: 99%

See 1 more Smart Citation

Verification as Learning Geometric Concepts

et al. 2013

View full text Add to dashboard Cite

Abstract. We formalize the problem of program verification as a learning problem, showing that invariants in program verification can be regarded as geometric concepts in machine learning. Safety properties define bad states: states a program should not reach. Program verification explains why a program's set of reachable states is disjoint from the set of bad states. In Hoare Logic, these explanations are predicates that form inductive assertions. Using samples for reachable and bad states and by applying well known machine learning algorithms for classification, we are able to generate inductive assertions. By relaxing the search for an exact proof to classifiers, we obtain complexity theoretic improvements. Further, we extend the learning algorithm to obtain a sound procedure that can generate proofs containing invariants that are arbitrary boolean combinations of polynomial inequalities. We have evaluated our approach on a number of challenging benchmarks and the results are promising.

show abstract

“…We use two-dimensional arrays to encode the heap. This representation is particularly useful for encoding frame conditions [4] and is used in Verl [25]. The heap is made explicit in field read and image expressions, read(h, u, f ) for u.f and img(h, r , f ) for r 'f .…”

Section: Preliminariesmentioning

confidence: 99%

Decision Procedures for Region Logic

Rosenberg

Banerjee

Naumann

2012

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Region logic is Hoare logic for object-based programs. It features local reasoning with frame conditions expressed in terms of sets of heap locations. This paper studies tableau-based decision procedures for RL, the quantifier-free fragment of the assertion language. This fragment combines sets and (functional) images with the theories of arrays and partial orders. The procedures are of practical interest because they can be integrated efficiently into the satisfiability modulo theories (SMT) framework. We provide a semi-decision procedure for RL and its implementation as a theory plugin inside the SMT solver Z3. We also provide a decision procedure for an expressive fragment of RL termed restricted-RL. We prove that deciding satisfiability of restricted-RL formulas is NP-complete. Both procedures are proven sound and complete. Preliminary performance results indicate that the semi-decision procedure has the potential toscale to large input formulas.

show abstract

Boogie: A Modular Reusable Verifier for Object-Oriented Programs

Cited by 648 publications

References 45 publications

Warps and Atomics: Beyond Barrier Synchronization in the Verification of GPU Kernels

Warps and Atomics: Beyond Barrier Synchronization in the Verification of GPU Kernels

Verification as Learning Geometric Concepts

Decision Procedures for Region Logic

Contact Info

Product

Resources

About