Plugging Anti and Output Dependence Removal Techniques into Loop Parallelization Algorithms - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Reports (Research Report) Year : 1996

Plugging Anti and Output Dependence Removal Techniques into Loop Parallelization Algorithms

Abstract

In this paper we shortly survey some loop transformation techniques which break anti or output dependences, or artificial cycles involving such «false» dependences. These false dependences are removed through the introduction of temporary buffer arrays. Next we show how to plug these techniques into loop parallelization algorithms (such as Allen and Kennedy's algorithm). The goal is to extract as many parallel loops as the intrinsic degree of parallelism of the nest authorizes, while avoiding a full memory expansion. We try to reduce the number of temporary arrays that we introduce, as well as their dimension.
Fichier principal
Vignette du fichier
RR-2914.pdf (284.73 Ko) Télécharger le fichier
Loading...

Dates and versions

inria-00073783 , version 1 (24-05-2006)

Identifiers

  • HAL Id : inria-00073783 , version 1

Cite

Pierre-Yves Calland, Alain Darte, Yves Robert, Frédéric Vivien. Plugging Anti and Output Dependence Removal Techniques into Loop Parallelization Algorithms. [Research Report] RR-2914, INRIA. 1996. ⟨inria-00073783⟩
89 View
400 Download

Share

Gmail Facebook Twitter LinkedIn More