Branch optimisation techniques for hardware compilation.pdfVIP

Branch optimisation techniques for hardware compilation.pdf

  1. 1、本文档共10页,可阅读全部内容。
  2. 2、原创力文档(book118)网站文档一经付费(服务费),不意味着购买了该文档的版权,仅供个人/单位学习、研究之用,不得用于商业用途,未经授权,严禁复制、发行、汇编、翻译或者网络传播等,侵权必究。
  3. 3、本站所有内容均由合作方或网友上传,本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺!文档内容仅供研究参考,付费前请自行鉴别。如您付费,意味着您自己接受本站规则且自行承担风险,本站不退款、不进行额外附加服务;查看《如何避免下载的几个坑》。如果您已付费下载过本站文档,您可以点击 这里二次下载
  4. 4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等,请点击“版权申诉”(推荐),也可以打举报电话:400-050-0827(电话支持时间:9:00-18:30)。
  5. 5、该文档为VIP文档,如果想要下载,成为VIP会员后,下载免费。
  6. 6、成为VIP后,下载本文档将扣除1次下载权益。下载后,不支持退款、换文档。如有疑问请联系我们
  7. 7、成为VIP后,您将拥有八大权益,权益包括:VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
  8. 8、VIP文档为合作方或网友上传,每下载1次, 网站将根据用户上传文档的质量评分、类型等,对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档
查看更多
Branch optimisation techniques for hardware compilation

Branch Optimisation Techniques for Hardware Compilation Henry Styles and Wayne Luk Department of Computing, Imperial College, 180 Queen’s Gate, London, England Abstract. This paper explores using information about program branch proba- bilities to optimise reconfigurable designs. The basic premise is to promote uti- lization by dedicating more resources to branches which execute more frequently. A hardware compilation system has been developed for producing designs which are optimised for different branch probabilities. We propose an analytical queue- ing network performance model to determine the best design from observed branch probability information. The branch optimisation space is characterized in an experimental study for Xilinx Virtex FPGAs of two complex applications: video feature extraction and progressive refinement radiosity. For designs of equal performance, branch-optimised designs require 24% and 27.5% less area. For de- signs of equal area, branch optimised designs run upto 3 times faster. Our analyti- cal performance model is shown to be highly accurate with relative error between 0.12 and  . 1 Introduction For most computer programs, the execution frequency of each basic block is controlled by the runtime behavior of conditional branches. Optimal resource allocation between basic blocks requires that execution frequencies be known. Software profilers collect execution frequencies for a representative dataset to support static resource allocation. Microprocessors demonstrate that branch probability information can be used at run- time to aid dynamic resource allocation. In this paper we explore the analogous use of branch probability information to optimize resource allocation in hardware compilation. In particular, the novel aspects of our work include: – a compiler that maps programs written in a subset of C to a set of hardware designs that are optimised for different branching probabilities; – analytical methods, including a queu

文档评论(0)

l215322 + 关注
实名认证
文档贡献者

该用户很懒,什么也没介绍

1亿VIP精品文档

相关文档