处理器验证激励在线筛选技术

Online Filtration of Stimuli for Microprocessor Verification

如何生成高质量的验证激励是功能验证中的核心问题之一.随着功能验证的不断进行,验证激励的有效性也随之降低.为了提高验证激励的质量,提出一种在线筛选技术来处理验证激励.该技术采用单分类支持向量机来在线地构建分类器,以对新生成的验证激励是否冗余进行预测,如果是冗余的,则不进入仿真阶段进行仿真.在此基础上,进一步提出指令序列核函数来衡量不同指令序列的相似程度.实验结果表明,与约束随机生成技术相比,文中技术可以减少约83%的验证激励及79%的验证时间.

One of the most critical issues during functional verification is to generate highly effective stimuli.As verification proceeds,the effectiveness of verification stimuli decreases.To improve the effectiveness of stimuli,an online filteration technique to process is proposed to generated stimuli.This technique employs one-class support vector machines to online construct a classifier to predict whether or not a newly generated stimulus is redundant,and the predicted redundant stimulus will not be sent for simulation.Besides,we also propose an instruction sequence kernel to measure the similarities among instruction sequences.Experimental results demonstrate that this technique can reduce about 83% stimuli and 79% verification time in comparison with conventional constrained random generation.