GCC Support Published For OpenACC On The GPU

Samsung has published their code to a modified version of the GCC Compiler that supports using the OpenACC 1.0 parallel computing specification. OpenACC allows for simplified parallel programming on heterogeneous CPU and GPU systems.

OpenACC is similar in nature to OpenMP but the programming standard allows exploiting the parallel potential out of graphics processors. OpenACC is annotated Fortran, C, and C++ code.

Evgeny Gavrin of Samsung announced the creation of an OpenACC GCC branch last week. The OpenACC support isn't finished yet but does produce working binaries with support for GPU offloading. The compiler supports the C, C++, and Fortran versions of OpenACC except only OpenACC 1.0 is supported and not the newer 2.0 specification. The OpenACC compiler support ends up generating translated OpenCL 1.1 code for the GPU offloading.

Hopefully the OpenACC GCC support will be able to be complete soon enough and merged to master. Right now only the PGI, Cray, and CAPS commercial compilers support OpenACC and there's been no open-source implementation aside from an experimental accULL project out of a European university.