An NVIDIA compiler engineer last week laid out the company's plans for implementing OpenACC 3.3 offloading support within the LLVM Clang compiler.
Erich Keane with NVIDIA laid out their initial plans for how they hope to upstream OpenACC 3.3 offload support within Clang. OpenACC is the parallel computing standard developed by NVIDIA with PGI and Cray for C / C++ / Fortran. For several years the US Department of Energy and others have worked on CLACC for supporting OpenACC with Clang and LLVM as well as "Flacc" as a Fortran OpenACC implementation for Flang. This NVIDIA-led implementation though would be different. There is also some code sharing that can take place with LLVM's OpenMP code.