I've tried on ORNL Wombat cluster which is aarch64 host with NVIDIA GPUs. Offloading works fine. All 19 tests in libomptarget passed.
Details
Details
Diff Detail
Diff Detail
- Repository
- rG LLVM Github Monorepo
Event Timeline
Comment Actions
Building the CUDA plugin on aarch64 hosts was enabled in the ykt branch ages ago, so the configuration has been tested extensively. LGTM.