Optimizing Local Memory Allocation and Assignment Through a Decoupled Approach

Boubacar Diouf 1 Özcan Özturk 2 Albert Cohen 1
1 ALCHEMY - Architectures, Languages and Compilers to Harness the End of Moore Years
LRI - Laboratoire de Recherche en Informatique, UP11 - Université Paris-Sud - Paris 11, Inria Saclay - Ile de France, CNRS - Centre National de la Recherche Scientifique : UMR8623
Abstract : Software-controlled local memories (LMs) are widely used to provide fast, scalable, power efficient and predictable access to critical data. While many studies addressed LM management, keeping hot data in the LM continues to cause major headache. This paper revisits LM management of arrays in light of recent progresses in register allocation, supporting multiple live-range splitting schemes through a generic integer linear program. These schemes differ in the grain of decision points. The model can also be extended to address fragmentation, assigning live ranges to precise offsets. We show that the links between LM management and register allocation have been underexploited, leaving much fundamental questions open and effective applications to be explored.
Type de document :
Communication dans un congrès
The 23rd International Workshop on Languages and Compilers for Parallel Computing, Oct 2009, Newark, Delaware, United States. 2009
Liste complète des métadonnées

Littérature citée [24 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-00645302
Contributeur : Albert Cohen <>
Soumis le : dimanche 27 novembre 2011 - 20:27:51
Dernière modification le : jeudi 5 avril 2018 - 12:30:12
Document(s) archivé(s) le : vendredi 16 novembre 2012 - 12:12:15

Fichier

paper.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00645302, version 1

Collections

Citation

Boubacar Diouf, Özcan Özturk, Albert Cohen. Optimizing Local Memory Allocation and Assignment Through a Decoupled Approach. The 23rd International Workshop on Languages and Compilers for Parallel Computing, Oct 2009, Newark, Delaware, United States. 2009. 〈hal-00645302〉

Partager

Métriques

Consultations de la notice

350

Téléchargements de fichiers

123