Optimal Loop Parallelization under Register Constraints - INRIA - Institut National de Recherche en Informatique et en Automatique Accéder directement au contenu
Rapport (Rapport De Recherche) Année : 1996

Optimal Loop Parallelization under Register Constraints

Résumé

This report deals with the interaction between instruction scheduling and register allocation, in the case of straight line code and in the case of loops. This problem is at the heart of code optimization in microprocessors with instruction-level parallelism. Usual solutions use heuristics based on a decoupled approach. We propose here a formulation by linear integer programming, that allows dependence, resource and register constraints to be integrated in the same framework . By varying the parameters, all kinds of optimization problems can be solved exactly (maximization of the throughput, minimization of the number of registers). We report on examples of computation timings that turn out to be prohibitive in some specific cases, but tractable on average.

Domaines

Autre [cs.OH]
Fichier principal
Vignette du fichier
RR-2781.pdf (357.82 Ko) Télécharger le fichier
Loading...

Dates et versions

inria-00073911 , version 1 (24-05-2006)

Identifiants

  • HAL Id : inria-00073911 , version 1

Citer

Christine Eisenbeis, Antoine Sawaya. Optimal Loop Parallelization under Register Constraints. [Research Report] RR-2781, INRIA. 1996. ⟨inria-00073911⟩
211 Consultations
155 Téléchargements

Partager

Gmail Facebook X LinkedIn More