Putting Polyhedral Loop Transformations to Work

Cédric Bastoul 1 Albert Cohen 1 Sylvain Girbal 1 Saurabh Sharma Olivier Temam 1
1 A3 - Advanced analysis to code optimization
UP11 - Université Paris-Sud - Paris 11, Inria Saclay - Ile de France
Abstract : We seek to extend the scope and efficiency of iterative compilation techniques by searching not only for the most appropriate program transformation parameters but for the most appropriate transformations themselves, or even for compositions of transformations. For that purpose, we need to find a generic way to express program transformations and compositions of transformations. In this article, we introduce a framework for the polyhedral representation of a wide range of program transformations in a unified and generic way. We also show that it is possible to generate efficient code after the application of polyhedral program transformations. Finally, we demonstrate an implementation of the program transformation framework and the code generation techniques in the Open64/ORC compiler.
Document type :
Reports
Complete list of metadatas

https://hal.inria.fr/inria-00071681
Contributor : Rapport de Recherche Inria <>
Submitted on : Tuesday, May 23, 2006 - 6:31:40 PM
Last modification on : Thursday, February 9, 2017 - 3:58:48 PM
Long-term archiving on : Sunday, April 4, 2010 - 8:41:17 PM

Identifiers

  • HAL Id : inria-00071681, version 1

Collections

Citation

Cédric Bastoul, Albert Cohen, Sylvain Girbal, Saurabh Sharma, Olivier Temam. Putting Polyhedral Loop Transformations to Work. [Research Report] RR-4902, INRIA. 2003. ⟨inria-00071681⟩

Share

Metrics

Record views

244

Files downloads

417