一种Cache一致性协议验证中覆盖率驱动的随机验证方法

摘要

Random verification technology is an important supporting technology in modern VLSI simulation verification processes. Coverage-driven random testing generation method is one of the hot topics in the research area. Targeting at Cache coherence protocol verification, the paper explains a method to perfect coverage-driven based random verification by introducing native Bayesian model based machine learning. Then through relative practical verification process the method is analyzed and discussed.%随机验证技术是当今大规模集成电路仿真验证流程中的一项重要支撑技术,覆盖率驱动的随机测试生成方法是目前该领域研究的热点之一.针对Cache一致性协议的验证目标,介绍一种引入基于朴素贝叶斯模型的机器学习来完善基于覆盖率驱动的随机验证的方法,并结合相关的实际验证过程对该方法进行了分析和讨论.