Research on Translucent Mechanism-Based Infeasible Path

Infeasible paths increase program complexity and program redundancy, generate useless DU(def-use) chains, and affect the result of program static analysis. Based on the valid information produced in the process of program compilation, this paper presents a translucent technique to analyze infeasible paths. It first requires the complier construct an information pool for each judge node and variables used in the node, and find out key infeasible branches of infeasible paths with these information, which is transparent to the tester; then the tester can judge if the tested path is infeasible or not with these key infeasible branches, which is opaque to the tester. By analyzing infeasible paths, we can refine the DU chain information. In our experiment, we analyze infeasible paths in multi SPEC95 benchmark programs with the translucent method and the demand-driven method. The result shows that time consumption is fewer with our method.