Execution-aware fault localization based on the control flow analysis

6Citations
Citations of this article
7Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Coverage-based fault localization techniques assess the suspiciousness of program entities individually. However, the individual coverage information cannot reveal the execution paths and to some extent it simplifies the executions. In this paper, the control flow analysis is adopted to analyze the executions first. Second, the edge suspiciousness is used to calculate the failed executions distribution to different control flows. By comparing different failed executions distributions of blocks covered by the same failed execution path, we propose the bug proneness to quantify how each block contributes to the failure. Similarly, the bug free confidence is also proposed to represent the possibility of bug free for blocks covered by a passed execution path. At last, the weighted coverage information statistic is proceeded and the weighted coverage based fault localization technique is brought out. We conduct several experiments to compare our technique with an existing representative technique by using standard benchmarks and the results are promising. © 2010 Springer-Verlag Berlin Heidelberg.

Cite

CITATION STYLE

APA

Zhao, L., Wang, L., Xiong, Z., & Gao, D. (2010). Execution-aware fault localization based on the control flow analysis. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6377 LNCS, pp. 158–165). https://doi.org/10.1007/978-3-642-16167-4_21

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free