2012 International Conference on Computational Intelligence and Software Engineering (CiSE 2012)(E-BOOK)

Wuhan,China,2012-12-142012-12-162012

ISBN: 978-1-61896-036-8 Scientific Research Publishing

E-Book 275pp Pub. Date: December 2012

Category: Computer Science & Communications

Price: $100

Title: Mixing Lockset Analysis and Symbolic Execution for Critical Section Inference
Source: 2012 International Conference on Computational Intelligence and Software Engineering (CiSE 2012)(E-BOOK) (pp 228-232)
Author(s): Hao Liu, School of Software, Shanghai Jiao Tong University
Di Yu, School of Electronic Information and Electrical Engineering,Shanghai Jiao Tong University
Di Zhang, School of Software, Shanghai Jiao Tong University
Abstract: In this paper, we propose a new method to combine highly scalable lockset analysis and highly accurate symbolic execution for critical section inference analysis. Critical section inference analysis can improve static concurrency bug detection by reducing false positives, can reduce runtime overhead of dynamic tools by instrumenting less, and can help compiler optimization by providing the analysis result. But balancing precision and efficiency in static analysis algorithms is very challenging. For example symbolic execution is precise but limited to small or special programs due to state space exploration, while lockset analysis is scalable to large programs but it lose precision due to abstraction. As a result, we mix these two methods together to achieve both precision and efficiency. Lockset analysis is a scalable way to identify functions with fuzzy locksets, and Symbolic execution is a precise way to resolve the fuzzy functions to generate clear results. Our lockset analysis is based on RacerX[5] and symbolic execution is based on KLEE[1].
Free SCIRP Newsletters
Copyright © 2006-2024 Scientific Research Publishing Inc. All Rights Reserved.
Top