Control unit synthesis targeting low-power processors

Chuan-Yu Wang, K. Roy
{"title":"Control unit synthesis targeting low-power processors","authors":"Chuan-Yu Wang, K. Roy","doi":"10.1109/ICCD.1995.528907","DOIUrl":null,"url":null,"abstract":"With demands for reliability and further integration, reducing power consumption becomes a critical concern in today's processor design. Considering the different techniques to minimize power consumption and promote system's reliability, reducing switching activity of CMOS circuits is a promising area to be explored. Motivated by these, we propose two optimization schemes which can be incorporated into processor's control unit synthesis to lower power dissipation. The first one, a low-power decoding scheme, utilizes graph embedding and logic minimization techniques to refine the decoding structure in processor's control unit. To get further optimization for those control units in nanoprogrammed or microprogrammed architecture, the second scheme is proposed to optimally assign ZERO or ONE to the don't-care bits distributed in nanocontrol memory or control memory, to significantly reduce switching activity within the control unit and/or on the path from control unit to data processing unit. To achieve these two goals efficiently, we have used pseudo-Boolean programming to optimize the synthesis parameters. Based on a subset of 8086 instruction set, experimental results show that 15.8 percent improvement is obtained by properly encoding instruction opcodes, and 4.9 to 16.6 percent improvement can be obtained from a optimal don't-care bits assignment.","PeriodicalId":281907,"journal":{"name":"Proceedings of ICCD '95 International Conference on Computer Design. VLSI in Computers and Processors","volume":"10 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1995-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of ICCD '95 International Conference on Computer Design. VLSI in Computers and Processors","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCD.1995.528907","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3

Abstract

With demands for reliability and further integration, reducing power consumption becomes a critical concern in today's processor design. Considering the different techniques to minimize power consumption and promote system's reliability, reducing switching activity of CMOS circuits is a promising area to be explored. Motivated by these, we propose two optimization schemes which can be incorporated into processor's control unit synthesis to lower power dissipation. The first one, a low-power decoding scheme, utilizes graph embedding and logic minimization techniques to refine the decoding structure in processor's control unit. To get further optimization for those control units in nanoprogrammed or microprogrammed architecture, the second scheme is proposed to optimally assign ZERO or ONE to the don't-care bits distributed in nanocontrol memory or control memory, to significantly reduce switching activity within the control unit and/or on the path from control unit to data processing unit. To achieve these two goals efficiently, we have used pseudo-Boolean programming to optimize the synthesis parameters. Based on a subset of 8086 instruction set, experimental results show that 15.8 percent improvement is obtained by properly encoding instruction opcodes, and 4.9 to 16.6 percent improvement can be obtained from a optimal don't-care bits assignment.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
针对低功耗处理器的控制单元合成
随着对可靠性和进一步集成的要求,降低功耗成为当今处理器设计中的一个关键问题。考虑到降低功耗和提高系统可靠性的不同技术,降低CMOS电路的开关活度是一个有前景的探索领域。在此基础上,我们提出了两种优化方案,可将其整合到处理器的控制单元合成中,以降低功耗。第一种是低功耗解码方案,利用图嵌入和逻辑最小化技术来优化处理器控制单元的解码结构。为了进一步优化纳米编程或微编程架构中的控制单元,提出了第二种方案,将零或一分配给分布在纳米控制存储器或控制存储器中的不关心位,以显着减少控制单元内和/或从控制单元到数据处理单元的路径上的切换活动。为了有效地实现这两个目标,我们使用伪布尔编程来优化合成参数。实验结果表明,在8086指令集的一个子集上,通过对指令操作码进行适当的编码可以提高15.8%的性能,而通过最优的不关心位分配可以提高4.9 ~ 16.6%的性能。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Design and implementation of a 100 MHz centralized instruction window for a superscalar microprocessor Multiprocessor design verification for the PowerPC 620 microprocessor Connection-oriented net model and fuzzy clustering techniques for K-way circuit partitioning Dynamic minimization of OKFDDs Simple tree-construction heuristics for the fanout problem
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1