Barrier inference

Aiken, Alexander; Gay, David

doi:10.1145/268946.268974

Cited by 92 publications

(76 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3 In this light, Rule (a) from the definition of CP is intuitively clear: as far as later events are concerned, two conflicting events have to occur in the same order in every correctly-reordered execution. Thus, conflicting events induce a hard ordering dependency, if it is certain that they do not constitute a race (in this case, because they are protected by a common lock).…”

Section: Threadmentioning

confidence: 99%

Sound predictive race detection in polynomial time

et al. 2012

View full text Add to dashboard Cite

Data races are among the most reliable indicators of programming errors in concurrent software. For at least two decades, Lamport's happens-before (HB) relation has served as the standard test for detecting races-other techniques, such as lockset-based approaches, fail to be sound, as they may falsely warn of races. This work introduces a new relation, causally-precedes (CP), which generalizes happens-before to observe more races without sacrificing soundness. Intuitively, CP tries to capture the concept of happens-before ordered events that must occur in the observed order for the program to observe the same values. What distinguishes CP from past predictive race detection approaches (which also generalize an observed execution to detect races in other plausible executions) is that CP-based race detection is both sound and of polynomial complexity.We demonstrate that the unique aspects of CP result in practical benefit. Applying CP to real-world programs, we successfully analyze server-level applications (e.g., Apache FtpServer) and show that traces longer than in past predictive race analyses can be analyzed in mere seconds to a few minutes. For these programs, CP race detection uncovers races that are hard to detect by repeated execution and HB race detection: a single run of CP race detection produces several races not discovered by 10 separate rounds of happens-before race detection.

show abstract

Section: Threadmentioning

confidence: 99%

Sound predictive race detection in polynomial time

et al. 2012

View full text Add to dashboard Cite

show abstract

“…Unlike our dynamic testing scenario, strength reduction optimizations, loop dependence testing and runtime array bound and access anomaly checking primarily motivate these techniques. Second, our LOV analysis also uses the concept of the single-valued variable, a variable that maintains identical values across all MPI tasks through all possible control flows [3,43]. Those analyses classify variables as single-valued or multi-valued in order to verify a program's synchronization pattern.…”

Section: Automatic Extraction Of Application Progressmentioning

confidence: 99%

Lightweight and Statistical Techniques for Petascale Debugging: Correctness on Petascale Systems (CoPS) Preliminry Report

Supinski¹,

Miller²,

Liblit³

2011

View full text Add to dashboard Cite

“…5 In our example, the first barrier in each process matches the other, and the second barrier in each process matches the other. Synchronizations that are not syntactically identifiable are often considered as a sign of potential bugs [1]. Note that reference cells c 1 and c 2 are not used for synchronization and therefore need no syntactic restriction.…”

Section: Examplesmentioning

confidence: 99%