渐进式退格:后硅调试的高效前导计算

Johnny J. W. Kuan, Tor M. Aamodt
{"title":"渐进式退格:后硅调试的高效前导计算","authors":"Johnny J. W. Kuan, Tor M. Aamodt","doi":"10.1109/MTV.2012.23","DOIUrl":null,"url":null,"abstract":"As microprocessors become more complex, finding errors in their design becomes more difficult. Most design errors are caught before the chip is fabricated, however, some make it into the fabricated design. One challenge in determining what is wrong with a new design after fabrication is the lack of observability into the state of the fabricated chip. To address this challenge, BackSpace proposes generating a trace of the states that lead up to an erroneous state. To add one state to the trace, BackSpace first generates a set of possible predecessor states (the pre-image), then tests them one at a time to find one that is reached during execution. In this paper, we propose an improved algorithm called Progressive-BackSpace. It does not enumerate every state in the pre-image. Instead, it first finds a reachable candidate state, and then determines if it is a predecessor state. This results in a practical implementation of BackSpace by greatly reducing the time needed to find prede- cessor states. The hardware overhead is also reduced by 94.4% relative to a recently proposed implementation of BackSpace. These algorithms were implemented and evaluated on a RTL model of an out-of-order processor, that models non-deterministic effects.","PeriodicalId":201401,"journal":{"name":"2012 13th International Workshop on Microprocessor Test and Verification (MTV)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2012-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Progressive-BackSpace: Efficient Predecessor Computation for Post-Silicon Debug\",\"authors\":\"Johnny J. W. Kuan, Tor M. Aamodt\",\"doi\":\"10.1109/MTV.2012.23\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"As microprocessors become more complex, finding errors in their design becomes more difficult. Most design errors are caught before the chip is fabricated, however, some make it into the fabricated design. One challenge in determining what is wrong with a new design after fabrication is the lack of observability into the state of the fabricated chip. To address this challenge, BackSpace proposes generating a trace of the states that lead up to an erroneous state. To add one state to the trace, BackSpace first generates a set of possible predecessor states (the pre-image), then tests them one at a time to find one that is reached during execution. In this paper, we propose an improved algorithm called Progressive-BackSpace. It does not enumerate every state in the pre-image. Instead, it first finds a reachable candidate state, and then determines if it is a predecessor state. This results in a practical implementation of BackSpace by greatly reducing the time needed to find prede- cessor states. The hardware overhead is also reduced by 94.4% relative to a recently proposed implementation of BackSpace. These algorithms were implemented and evaluated on a RTL model of an out-of-order processor, that models non-deterministic effects.\",\"PeriodicalId\":201401,\"journal\":{\"name\":\"2012 13th International Workshop on Microprocessor Test and Verification (MTV)\",\"volume\":\"15 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2012-12-10\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2012 13th International Workshop on Microprocessor Test and Verification (MTV)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/MTV.2012.23\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2012 13th International Workshop on Microprocessor Test and Verification (MTV)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/MTV.2012.23","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1

摘要

随着微处理器变得越来越复杂,发现其设计中的错误变得越来越困难。大多数设计错误在芯片制造之前就被发现了,然而,有些错误却进入了制造设计。在制造后确定新设计的问题是缺乏对制造芯片状态的可观察性,这是一个挑战。为了解决这个问题,BackSpace建议生成导致错误状态的状态跟踪。要向跟踪中添加一个状态,BackSpace首先生成一组可能的前体状态(预映像),然后逐个测试它们,以找到在执行期间达到的状态。在本文中,我们提出了一种改进的算法,称为Progressive-BackSpace。它不列举预映像中的每个状态。相反,它首先找到一个可到达的候选状态,然后确定它是否是一个前导状态。这大大减少了查找前体状态所需的时间,从而实现了退格的实际实现。与最近提出的BackSpace实现相比,硬件开销也减少了94.4%。这些算法在无序处理器的RTL模型上实现和评估,该模型模拟了不确定性效应。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Progressive-BackSpace: Efficient Predecessor Computation for Post-Silicon Debug
As microprocessors become more complex, finding errors in their design becomes more difficult. Most design errors are caught before the chip is fabricated, however, some make it into the fabricated design. One challenge in determining what is wrong with a new design after fabrication is the lack of observability into the state of the fabricated chip. To address this challenge, BackSpace proposes generating a trace of the states that lead up to an erroneous state. To add one state to the trace, BackSpace first generates a set of possible predecessor states (the pre-image), then tests them one at a time to find one that is reached during execution. In this paper, we propose an improved algorithm called Progressive-BackSpace. It does not enumerate every state in the pre-image. Instead, it first finds a reachable candidate state, and then determines if it is a predecessor state. This results in a practical implementation of BackSpace by greatly reducing the time needed to find prede- cessor states. The hardware overhead is also reduced by 94.4% relative to a recently proposed implementation of BackSpace. These algorithms were implemented and evaluated on a RTL model of an out-of-order processor, that models non-deterministic effects.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信