A simple OpenMP offloading app is built fine with A100 GPU-aware
12-ish Clang and
executed on the device as well, but I see an error message from CUDA RTL like:
Target CUDA RTL --> Error returned from cuCtxSetCurrent
Target CUDA RTL --> Unresolved CUDA error code: 4
Target CUDA RTL --> Unsuccessful cuGetErrorString return status: 4
Should I be concerned about this?