{"title":"Pre-silicon verification of the Alpha 21364 microprocessor error handling system","authors":"Richard Lee, B. Tsien","doi":"10.1145/378239.379073","DOIUrl":null,"url":null,"abstract":"This paper presents the strategy used to verify the error logic in the Alpha 21364 microprocessor. Traditional pre-silicon strategies of focused testing or unit-level random testing yield limited results in finding complex bugs in the error handling logic of a microprocessor. This paper introduces a technique to simulate error conditions and their recovery in a global environment using random test stimulus closely approximating traffic found in a real system. A significant number of bugs were found using this technique. A majority of these bugs could not be uncovered using a simple random environment, or were counter-intuitive to focused test design.","PeriodicalId":154316,"journal":{"name":"Proceedings of the 38th Design Automation Conference (IEEE Cat. No.01CH37232)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2001-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 38th Design Automation Conference (IEEE Cat. No.01CH37232)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/378239.379073","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
This paper presents the strategy used to verify the error logic in the Alpha 21364 microprocessor. Traditional pre-silicon strategies of focused testing or unit-level random testing yield limited results in finding complex bugs in the error handling logic of a microprocessor. This paper introduces a technique to simulate error conditions and their recovery in a global environment using random test stimulus closely approximating traffic found in a real system. A significant number of bugs were found using this technique. A majority of these bugs could not be uncovered using a simple random environment, or were counter-intuitive to focused test design.