Locating Software Faults Based on Control Flow and Data Dependence

Debugging software is a difficult and timeconsuming work. Fault localization techniques are becoming extremely important. Coverage Based Fault Localization (CBFL) is very commonly used in fault location technique. Tarantula is a typical one. It uses the coverage statistics of failed execution paths and passed execution paths to calculate the suspiciousness in the software. However, since this technique ignores the data dependency, it is hard to find the bugs which are not in the suspicious code area but have data dependence with it. In order to improve the efficiency of fault locating, we combine control flow coverage information and data dependence from program slicing. We validate our approach experimentally using Siemens benchmark programs. The experimental results show that our approach is more effective than Tarantula.