[Openmp-commits] [PATCH] D50522: [OpenMP][libomptarget] Bringing up to spec with respect to OMP_TARGET_OFFLOAD env var

[Alexandre Eichenberger via Phabricator via Openmp-commits]

AlexEichenberger added a comment.

In https://reviews.llvm.org/D50522#1194210, @Hahnfeld wrote:

> You are right, "available" is not defined in the standard. I've always though of "plugged into the system", ie all devices that are visible to the CUDA runtime. That would match the current implementation of `omp_get_num_devices` which is defined to return "the number of available target devices".
>  Actually this behaviour would be important for us as we have our GPUs configured exclusively. So when there is already a process running all other users get a runtime error. In that case it would be very helpful to have `libomptarget` abort the program.


So you like deciding available on first use? This is what your comment seems to imply, but I am not 100% sure.


Repository:
  rOMP OpenMP

https://reviews.llvm.org/D50522