HEAT: An Integrated Static and Dynamic Approach for Thread Escape Analysis

Escape analysis can determine whether and when a variable becomes shared for multiple threads, which is a foundation for many other program analyses. Most existing escape analysis tools are either purely dynamic or static analysis. Static analysis usually considers all possible behaviors of a program, however, may produce false positives; whereas dynamic approaches can not analyze unobserved behaviors of a program. This paper presents a hybrid approach that integrates static and dynamic analyses to address this problem. We first perform static analysis to obtain summaries of accesses to potential escaped variables. When the program runs, dynamic analysis is used to confirm variable sharing; for unexecuted code, we determine the sharing of variables in it by performing an interprocedural static analysis based on the runtime information. Compared to dynamic analysis, the hybrid approach is able to determine the escape property of variables in unexecuted code. Compared to static analysis, the hybrid approach produces fewer false alarms. We implemented this hybrid escape analysis in Java. Our experimental evaluation on several benchmarks and real-world applications shows that it has improved accuracy compared to existing approaches and can also boost the performance of subsequent program analyses significantly.