GPU Offloading is emergent programming model.
OpenMP includes in its latest 4.0 specification the accelerator model.
In this paper we present a newly implementation of this specification
while generationg "native" GPU code in the OmpSs programming
model developed at the Barcelona Supercomputing Center. Focused
on targeting NVIDIA GPUs, our work is based on an OpenMP 4.0
implementation in the Mercurium source-to-source compiler
infrastructure referred MACC (Mercurium ACCelerator
compiler). Finally, the paper is also discuss challenges od code
generation for GPUs.
CitationOzen, Guray; Ayguadé Parra, Eduard; Labarta Mancho, Jesús José. MACC: Mercurium ACCelerator Model. A: "BSC Doctoral Symposium (2nd: 2015: Barcelona)". 2nd ed. Barcelona: Barcelona Supercomputing Center, 2015, p. 117-118.
All rights reserved. This work is protected by the corresponding intellectual and industrial property rights. Without prejudice to any existing legal exemptions, reproduction, distribution, public communication or transformation of this work are prohibited without permission of the copyright holder. If you wish to make any use of the work not provided for in the law, please contact: email@example.com