Efficient data race detection for async-finish parallelism

Abstract. A major productivity hurdle for parallel programming is the presence of data races. Data races can lead to all kinds of harmful program behaviors, includ-ing determinism violations and corrupted memory. However, runtime overheads of current dynamic data race detectors are still prohibitively large (often incurring slowdowns of 10 × or larger) for use in mainstream software development. In this paper, we present an efficient dynamic race detector algorithm targeting the async-finish task-parallel parallel programming model. The async and finish constructs are at the core of languages such as X10 and Habanero Java (HJ). These constructs generalize the spawn-sync constructs used in Cilk, while still ensuring that all computation graphs are deadlock-free. We have implemented our algorithm in a tool called TASKCHECKER and eval-uated it on a suite of 12 benchmarks. To reduce overhead of the dynamic analysis, we have also implemented various static optimizations in the tool. Our experi-mental results indicate that our approach performs well in practice, incurring an average slowdown of 3.05 × compared to a serial execution in the optimized case.