when is offloading to NVIDIA targets available?

Hi Siegmar,

The trunk is currently missing the nvptx runtime, so even if you enable OPENMP_ENABLE_LIBOMPTARGET you only build the base library, the CUDA plugin and the generic-elf-64 plugin. The nvptx runtime contains CUDA versions of runtime functions and is necessary for proper execution on the device.

We are currently working on upstreaming the nvptx runtime, I expect to have something ready by the end of next week.


graycol.gifSiegmar Gross via Openmp-dev —01/11/2017 13:21:03—Hi, today I’ve built the latest version of llvm-trunk. I used the following