Cap torchvision/torchaudio versions, remove amdhip64.dll fallback, fix visible GPU count

- Cap torchvision<0.26.0 and torchaudio<2.11.0 alongside torch<2.11.0 in
  both install.sh and install_python_stack.py to prevent resolver from
  selecting incompatible companion packages from ROCm wheel index
- Remove amdhip64.dll fallback in Windows ROCm detection (DLL presence
  without hipinfo/amd-smi is not proof of GPU existence)
- Fix get_visible_gpu_count() to use _get_parent_visible_gpu_spec() which
  respects HIP_VISIBLE_DEVICES/ROCR_VISIBLE_DEVICES on ROCm hosts
This commit is contained in:
Daniel Han 2026-03-31 12:06:14 +00:00
commit a0671107cf
4 changed files with 15 additions and 17 deletions

View file

@ -187,8 +187,8 @@ def _ensure_rocm_torch() -> None:
"--force-reinstall",
"--no-cache-dir",
"torch>=2.4,<2.11.0",
"torchvision",
"torchaudio",
"torchvision<0.26.0",
"torchaudio<2.11.0",
"--index-url",
index_url,
constrain = False,