Closed KyleFromNVIDIA closed 4 months ago
Building this as a wheel requires a static libnvidia-ml.a
(for now at least, until rapidsai/build-planning#35 is done), but CMake is not able to provide it. I've opened https://gitlab.kitware.com/cmake/cmake/-/issues/26131 to discuss this.
After offline discussion with @robertmaynard, we've come to the conclusion that nvml should be dynamically linked after all. I've updated this PR to do so.
/merge
This PR updates the latest CUDA build/test version 12.2.2 to 12.5.1.
Contributes to https://github.com/rapidsai/build-planning/issues/73