Practical Predictive Race Detection
arXiv: Software Engineering(2019)
摘要
Widely used data race detectors, including the state-of-the-art FastTrack algorithm, incur performance costs that are acceptable for regular in-house testing, but miss races detectable from the analyzed execution. Predictive analyses detect more data races in an analyzed execution than FastTrack does, but at significantly higher cost. This paper presents SmartTrack, an algorithm that optimizes predictive race detection analyses, including two analyses from prior work and a new analysis introduced in this paper. SmartTracku0027s algorithm incorporates two main optimizations: epoch and ownership optimizations from prior work, applied to predictive analysis for the first time; and novel conflicting critical section optimizations introduced by this paper. Our evaluation shows that SmartTrack achieves performance competitive with FastTrack-a qualitative improvement to the state-of-the-art in data race detection.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络