Abstract : The OpenMP specification requires that all threads in a team execute the same sequence of worksharing and barrier regions. An improper use of such directive may lead to deadlocks. In this paper we propose a static analysis to ensure this property is verified. The well-defined semantic of OpenMP programs makes compiler analysis more effective. We propose a new compile-time method to identify in OpenMP codes the potential improper uses of barriers and work-sharing constructs, and the execution paths that are responsible for these issues. We implemented our method in a GCC compiler plugin and show the small im-pact of our analysis on performance for NAS-OMP benchmarks and a test case for a production industrial code.
https://hal.archives-ouvertes.fr/hal-01078759
Contributeur : Emmanuelle Saillard
<>
Soumis le : jeudi 30 octobre 2014 - 10:54:01
Dernière modification le : jeudi 29 mars 2018 - 09:04:04
Document(s) archivé(s) le : lundi 2 février 2015 - 15:53:02
