Notations for Multiphase Pipelines

C. T. Johnston, D. Bailey, P. Lyons
{"title":"Notations for Multiphase Pipelines","authors":"C. T. Johnston, D. Bailey, P. Lyons","doi":"10.1109/DELTA.2010.29","DOIUrl":null,"url":null,"abstract":"FPGAs, (Field-Programmable Gate Arrays) are often used for embedded image processing applications. Parallelism, and in particular pipelining, is the most suitable architecture for supporting the required high throughput. Although pipelining is a well known technique for hardware design and is simple to describe, our experience has been that people have many problems implementing working pipelines, especially for multiphase designs. Existing hardware description languages force developers to design pipelines as a special case of parallel architecture, which makes it difficult to ensure that the pipeline has internally consistent timing. This is especially problematic in multiphase pipelines. This paper shows how many of these problems may be overcome by basing the notation on sequential dataflow, and discusses control issues of priming, stalling and flushing, with a proposed compiler implementation.","PeriodicalId":421336,"journal":{"name":"2010 Fifth IEEE International Symposium on Electronic Design, Test & Applications","volume":"3 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2010-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2010 Fifth IEEE International Symposium on Electronic Design, Test & Applications","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/DELTA.2010.29","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

FPGAs, (Field-Programmable Gate Arrays) are often used for embedded image processing applications. Parallelism, and in particular pipelining, is the most suitable architecture for supporting the required high throughput. Although pipelining is a well known technique for hardware design and is simple to describe, our experience has been that people have many problems implementing working pipelines, especially for multiphase designs. Existing hardware description languages force developers to design pipelines as a special case of parallel architecture, which makes it difficult to ensure that the pipeline has internally consistent timing. This is especially problematic in multiphase pipelines. This paper shows how many of these problems may be overcome by basing the notation on sequential dataflow, and discusses control issues of priming, stalling and flushing, with a proposed compiler implementation.
多相管道符号
fpga(现场可编程门阵列)通常用于嵌入式图像处理应用。并行,特别是流水线,是支持所需的高吞吐量的最合适的体系结构。虽然流水线是一种众所周知的硬件设计技术,并且易于描述,但我们的经验是,人们在实现工作管道时遇到了许多问题,特别是对于多相设计。现有的硬件描述语言迫使开发人员将管道设计为并行体系结构的特殊情况,这使得很难确保管道具有内部一致的时间。这在多相管道中尤其成问题。本文展示了基于顺序数据流的表示法可以克服多少这些问题,并讨论了启动、延迟和刷新的控制问题,并提出了一个编译器实现。
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 求助全文
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:604180095
Book学术官方微信