Re: Static branch prediction

andi@complang.tuwien.ac.at (Andreas Krall)
05 Mar 2008 10:58:42 GMT

          From comp.compilers

Related articles
Static branch prediction plfriko@yahoo.de (Tim Frink) (2008-03-03)
Re: Static branch prediction armbru@pond.sub.org (Markus Armbruster) (2008-03-04)
Re: Static branch prediction andi@complang.tuwien.ac.at (2008-03-05)
Re: Static branch prediction anton@mips.complang.tuwien.ac.at (2008-03-05)
Re: Static Branch Prediction arcfide@sacrideo.us (Aaron W. Hsu) (2012-12-01)
| List of all articles for this month |

From: andi@complang.tuwien.ac.at (Andreas Krall)
Newsgroups: comp.compilers
Date: 05 Mar 2008 10:58:42 GMT
Organization: Compilers Central
References: 08-03-010 08-03-023
Keywords: optimize, bibliography
Posted-Date: 05 Mar 2008 06:01:46 EST

Markus Armbruster <armbru@pond.sub.org> writes:
> Tim Frink <plfriko@yahoo.de> writes:
>
>> Some processors use a static branch prediction. The predicted outcome
> [Reordering basic blocks to avoid static branch mispredictions]
>> This idea is probably not new, so I wonder if you know any
>> compilers/optimizers that apply this technique? Do you have own
>> experiences or know any papers/reports where any results are
>> presented?
>
> Off hand:
>
> K. Pettis and R. C. Hansen, "Profile Guided Code Positioning"
> Proceedings of the SIGPLAN Conference on Programming Language Design
> and Implementation, pp. 16--27, June 1990.


@inproceedings{178252,
  author = {Andreas Krall},
  title = {Improving semi-static branch prediction by code replication},
  booktitle = {PLDI '94: Proceedings of the ACM SIGPLAN 1994 conference
                            on Programming language design and implementation},
  year = {1994},
  isbn = {0-89791-662-X},
  pages = {97--106},
  location = {Orlando, Florida, United States},
  doi = {http://doi.acm.org/10.1145/178243.178252},
  publisher = {ACM},
  address = {New York, NY, USA},
  }


@inproceedings{195549,
  author = {Cliff Young and Michael D. Smith},
  title = {Improving the accuracy of static branch prediction using
                    branch correlation},
  booktitle = {ASPLOS-VI: Proceedings of the sixth international conference
  on Architectural support for programming languages and operating systems},
  year = {1994},
  isbn = {0-89791-660-3},
  pages = {232--241},
  location = {San Jose, California, United States},
  doi = {http://doi.acm.org/10.1145/195473.195549},
  publisher = {ACM},
  address = {New York, NY, USA},
  }


@article{330255,
  author = {Cliff Young and Michael D. Smith},
  title = {Static correlated branch prediction},
  journal = {ACM Trans. Program. Lang. Syst.},
  volume = {21},
  number = {5},
  year = {1999},
  issn = {0164-0925},
  pages = {1028--1075},
  doi = {http://doi.acm.org/10.1145/330249.330255},
  publisher = {ACM},
  address = {New York, NY, USA},
  }


@inproceedings{143493,
  author = {Joseph A. Fisher and Stefan M. Freudenberger},
  title = {Predicting conditional branch directions from previous runs
                    of a program},
  booktitle = {ASPLOS-V: Proceedings of the fifth international conference
  on Architectural support for programming languages and operating systems},
  year = {1992},
  isbn = {0-89791-534-8},
  pages = {85--95},
  location = {Boston, Massachusetts, United States},
  doi = {http://doi.acm.org/10.1145/143365.143493},
  publisher = {ACM},
  address = {New York, NY, USA},
  }
--
andi@complang.tuwien.ac.at Andreas Krall
www.complang.tuwien.ac.at/andi Inst. f. Computersprachen, TU Wien
tel: (+431) 58801/18511 Argentinierstr. 8/4/1851
fax: (+431) 58801/18598 A-1040 Wien AUSTRIA EUROPE


Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.