Parallelizing data race detection

Abstract

Detecting data races in multithreaded programs is a crucial part of debugging such programs, but traditional data race detectors are too slow to use routinely. This paper shows how to speed up race detection by spreading the work across multiple cores. Our strategy relies on uniparallelism, which executes time intervals of a program (called epochs) in parallel to provide scalability, but executes all threads from a single epoch on a single core to eliminate locking overhead. We use several techniques to make parallelization effective: dividing race detection into three phases, predicting a subset of the analysis state, eliminating sequential work via transitive reduction, and reducing the work needed to maintain multiple versions of analysis via factorization. We demonstrate our strategy by parallelizing a happens-before detector and a lockset-based detector. We find that uniparallelism can significantly speed up data race detection. With 4× the number of cores as the original application, our strategy speeds up the median execution time by 4.4× for a happens-before detector and 3.3× for a lockset race detector. Even on the same number of cores as the conventional detectors, the ability for uniparallelism to elide analysis locks allows it to reduce the median overhead by 13% for a happens-before detector and 8% for a lockset detector.