Compiling A High-Level Directive-based Programming Model for Accelerators

Abstract. OpenACC is an emerging directive-based programming model for programming accelerators that typically enable non-expert program-mers to achieve portable and productive performance of their applica-tions. In this paper, we present the research and development challenges, and our solutions to create an open-source OpenACC compiler in a main stream compiler framework (OpenUH of a branch of Open64). We discuss in details our loop mapping techniques, i.e. how to distribute loop itera-tions over the GPGPU’s threading architectures, as well as their impacts on performance. The runtime support of this programming model are also presented. The compiler was evaluated with several commonly used benchmarks, and delivered similar performance to those obtained using a commercial compiler. We hope this implementation to serve as compiler infrastructure for researchers to explore advanced compiler techniques, to extend OpenACC to other programming languages, or to build per-formance tools used with OpenACC programs.