Simulation Data Mining for Functional Test Pattern Justification

Abstract

Our mutation based validation paradigm (MVP) is a validation environment for high-level microprocessor implementations. To be able to efficiently identify and analyze the architectural states (prospect states) that can possibly satisfy a set of constraints during MVP's test generation, we need to reduce the search space in the analysis process as early as possible. In this paper, we present some optimizations in the search space that speed up the overall test generation process.