Trace driven dynamic deadlock detection and reproduction

Abstract

Dynamic analysis techniques have been proposed to detect potential deadlocks. Analyzing and comprehending each potential deadlock to determine whether the deadlock is feasible in a real execution requires significant programmer effort. Moreover, empirical evidence shows that existing analyses are quite imprecise. This imprecision of the analyses further void the manual effort invested in reasoning about non-existent defects.In this paper, we address the problems of imprecision of existing analyses and the subsequent manual effort necessary to reason about deadlocks. We propose a novel approach for deadlock detection by designing a dynamic analysis that intelligently leverages execution traces. To reduce the manual effort, we replay the program by making the execution follow a schedule derived based on the observed trace. For a real deadlock, its feasibility is automatically verified if the replay causes the execution to deadlock.We have implemented our approach as part of WOLF and have analyzed many large (upto 160KLoC) Java programs. Our experimental results show that we are able to identify 74% of the reported defects as true (or false) positives automatically leaving very few defects for manual analysis. The overhead of our approach is negligible making it a compelling tool for practical adoption.

Omen+: a precise dynamic deadlock detector for multithreaded Java libraries

“…Static analysis approaches [8] for deadlock detection can be imprecise and also cannot automatically provide proof of correctness of the detected deadlock. We showed in [6] that our approach of deadlock detection, that is employed here, outperforms other dynamic deadlock detectors [2,1] with respect to precision and scalability.…”

“…However, the reported deadlocks are manually analyzed to identify whether the reported deadlocks are true positives. In contrast, the integration of a precise deadlock detector [6] as part of implementation eliminates the need for a developer to analyze the source code and reason about the correctness of the reported deadlock.…”

“…The problem is further exacerbated by the large number of possible thread schedules for a given combination of concurrent method invocations and parameters. While the problem of thread schedules is partially addressed by dynamic analyses [6,2] that abstract the states resulting from different schedules, the ability of these analyses to detect defects Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.…”

“…Our tool, named OMEN+, takes as input a Java multithreaded library implementation and automatically detects possible deadlocks. Our implementation of OMEN+ integrates our recent techniques for deadlock test synthesis [5] and detection [6]. Initially, we synthesize multithreaded tests for deadlock detection [5].…”

Sherlock: scalable deadlock detection for concurrent programs

“…ConLock [12] first does an initial run and generates scheduling constraint from the trace and from a deadlock candidate, and then uses a form of random scheduling that works within the limits of the generated scheduling constraints. Wolf [53] first does an initial run and does cycle detection, then prunes away cycles that cannot be executed, then generates a synchronization dependency graph, and finally uses a form of scheduling based on that graph.…”