APOLLO: Automatic speculative POLyhedral Loop Optimizer - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2017

APOLLO: Automatic speculative POLyhedral Loop Optimizer

Résumé

A few weeks ago, we were glad to announce the first release of Apollo, the Automatic speculative POLyhedral Loop Opti-mizer. Apollo applies polyhedral optimizations on-the-fly to loop nests, whose control flow and memory access patterns cannot be determined at compile-time. In contrast to existing tools, Apollo can handle any kind of loop nest, whose memory accesses can be performed through pointers and in-directions. At runtime, Apollo builds a predictive polyhedral model, which is used for speculative optimization including parallelization. Being a dynamic system, Apollo can even apply the polyhedral model to nonlinear loops. This paper describes Apollo from the perspective of a user, as well as some of its main contributions and mechanisms, including the just-in-time polyhedral compilation, that significantly extends the scope of polyhedral techniques.
Fichier principal
Vignette du fichier
paper.pdf (519.07 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01533692 , version 1 (06-06-2017)

Identifiants

  • HAL Id : hal-01533692 , version 1

Citer

Juan Manuel Martinez Caamaño, Aravind Sukumaran-Rajam, Artiom Baloian, Manuel Selva, Philippe Clauss. APOLLO: Automatic speculative POLyhedral Loop Optimizer. IMPACT 2017 - 7th International Workshop on Polyhedral Compilation Techniques, Jan 2017, Stockholm, Sweden. pp.8. ⟨hal-01533692⟩
279 Consultations
219 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More