Concurrent bug detection and localization based on deep learningOA

Concurrency bugs,which are caused by improper synchronization of shared resources in multi-threaded or distributed systems,are notoriously hard to detect and thus compromise software reliability and security.Existing deep learning methods face three main limitations:the absence of large and dedicated datasets of diverse concurrency bugs;insufficient representation of concurrency semantics;and the inability of binary classification results to provide finer-grained debug information such as precise bug locations.To address these problems,this study proposes a novel method for effective concurrency bug detection and localization.A dedicated concurrency bug dataset is constructed to facilitate model training and evaluation.The method integrates a pre-trained model with a heterogeneous graph neural network(GNN)by incorporating a newly designed Concurrency-Aware Code Property Graph,which concisely and effectively characterizes concurrency semantics.To further facilitate debugging,the authors employ SubgraphX,a GNN-based interpretability method,to explore the graphs and localize concurrency bugs precisely,mapping them to specific subgraphs of source code.Experimental results show that,on average,the proposed method achieves an improvement of 10%in accuracy and precision,and 26%in recall,compared to state-of-the-art methods across diverse evaluation settings.