Designing an Effective Constraint Solver in Coverage Directed Test Generation

As the complexity of processors grows, the bottleneck of verification remains in generating suitable test programs that meet coverage metrics automatically. Coverage directed test generation is a technique to automate the feedback from coverage analysis to test generation. It is very important to solve the constraint satisfaction problem for a flexible coverage directed test generator with complex variables. In this paper, we propose an effective constraint solver which combines constraint satisfaction problem’s algorithms and coverage directed test generation to address the challenges that arise from the practical verification problem. We implement the constraint solver in our practical coverage directed test generation platform, which has been used in the verification of an embedded processor. The efficiency of our approach has been demonstrated by the practical results.