tqchen opened a new pull request, #652: URL: https://github.com/apache/tvm-ffi/pull/652
Fixes #651. The JIT extension backend selector decided CUDA-vs-HIP by probing for a ROCm home, whose fallback returns `/opt/rocm` whenever that directory merely exists. A CUDA/NVIDIA host carrying a stray `/opt/rocm` was therefore misclassified as HIP even with `nvcc` present and no `hipcc`. `_detect_gpu_backend()` now resolves the backend from explicit signals, in order: the `TVM_FFI_GPU_BACKEND` override, then PyTorch build signals (`torch.version.hip` / `torch.version.cuda`), then an available `hipcc` when no `nvcc` is present. Detection defaults to CUDA when otherwise inconclusive, and the existence of `/opt/rocm` alone never forces HIP. Toolkit-home, compiler, include-path, flag, and architecture discovery are unchanged and run only after the backend is resolved, so CUDA and HIP state never mix. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
