Runtime On-Stack Parallelization of Dependence-Free For-Loops in Binary Programs - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Article Dans Une Revue IEEE Letters of the Computer Society Année : 2019

Runtime On-Stack Parallelization of Dependence-Free For-Loops in Binary Programs

Résumé

With the multicore trend, the need for automatic parallelization is more pronounced, especially for legacy and proprietary code where no source code is available and/or the code is already running and restarting is not an option. In this paper, we engineer a mechanism for transforming at runtime a frequent for-loop with no data dependencies in a binary program into a parallel loop, using on-stack replacement. With our mechanism, there is no need for source code, debugging information or restarting the program. Also, the mechanism needs no static instrumentation or information. The mechanism is implemented using the Padrone binary modification system and pthreads, where the remaining iterations of the loop are executed in parallel. The mechanism keeps the running program state by extracting the targeted loop into a separate function and copying the current stack frame into the corresponding frames of the created threads. Initial study is conducted on a set of kernels from the Polybench workload. Experiments results show from 2x to 3.5x speedup from sequential to parallelized code on four cores, which is similar to source code level parallelization.

Domaines

Autre [cs.OH]
Fichier principal
Vignette du fichier
LOCS_binary_parallelization.pdf (227.39 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02061340 , version 1 (08-03-2019)

Identifiants

Citer

Marwa Yusuf, Ahmed El-Mahdy, Erven Rohou. Runtime On-Stack Parallelization of Dependence-Free For-Loops in Binary Programs. IEEE Letters of the Computer Society, 2019, 2 (1), pp.1-4. ⟨10.1109/LOCS.2019.2896559⟩. ⟨hal-02061340⟩
52 Consultations
185 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More