Skip to Main content Skip to Navigation
New interface
Conference papers

A Framework for Enhancing Data Reuse via Associative Reordering

Kevin Stock 1 Martin Kong 1 Tobias Grosser 2 Louis-Noël Pouchet 3 Fabrice Rastello 4, 5 Jagannathan Ramanujam 6 Ponnuswamy Sadayappan 1 
2 Parkas - Parallélisme de Kahn Synchrone
DI-ENS - Département d'informatique - ENS Paris, Inria Paris-Rocquencourt, CNRS - Centre National de la Recherche Scientifique : UMR 8548
4 CORSE - Compiler Optimization and Run-time Systems
Inria Grenoble - Rhône-Alpes, LIG - Laboratoire d'Informatique de Grenoble
5 GCG - Grenoble's Compiler Group
Inria Grenoble - Rhône-Alpes
Abstract : The freedom to reorder computations involving associative operators has been widely recognized and exploited in designing parallel algorithms and to a more limited extent in optimizing compilers. In this paper, we develop a novel framework utilizing the associativity and commutativity of operations in regular loop computations to enhance register reuse. Stencils represent a particular class of important computations where the optimization framework can be applied to enhance performance. We show how stencil operations can be implemented to better exploit register reuse and reduce load/stores. We develop a multi-dimensional retiming formalism to characterize the space of valid implementations in conjunction with other program transformations. Experimental results demonstrate the effectiveness of the framework on a collection of high-order stencils.
Complete list of metadata
Contributor : Fabrice Rastello Connect in order to contact the contributor
Submitted on : Friday, June 27, 2014 - 5:03:11 PM
Last modification on : Tuesday, November 29, 2022 - 12:06:07 PM



Kevin Stock, Martin Kong, Tobias Grosser, Louis-Noël Pouchet, Fabrice Rastello, et al.. A Framework for Enhancing Data Reuse via Associative Reordering. PLDI '14 - 35th ACM SIGPLAN Conference on Programming Language Design and Implementation, Jun 2014, Edinburgh, United Kingdom. pp.65-76, ⟨10.1145/2594291.2594342⟩. ⟨hal-01016093⟩



Record views