Skip to content

CI jobs binary_linux_wheel_*_cu117 broken due to libcublas.so.11 dependency #6628

Closed
@datumbox

Description

@datumbox

🐛 Describe the bug

Multiple CI jobs of TorchVision are currently failing with:

+ python setup.py clean
Traceback (most recent call last):
  File "setup.py", line 9, in <module>
    import torch
  File "/opt/python/cp37-cp37m/lib/python3.7/site-packages/torch/__init__.py", line 191, in <module>
    _load_global_deps()
  File "/opt/python/cp37-cp37m/lib/python3.7/site-packages/torch/__init__.py", line 153, in _load_global_deps
    ctypes.CDLL(lib_path, mode=ctypes.RTLD_GLOBAL)
  File "/opt/python/cp37-cp37m/lib/python3.7/ctypes/__init__.py", line 364, in __init__
    self._handle = _dlopen(self._name, mode)
OSError: libcublas.so.11: cannot open shared object file: No such file or directory

The issue seems to be mainly with CUDA 11.7

Versions

Latest torchvision main, PyTorch Core 20220922

cc @seemethere

Metadata

Metadata

Assignees

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions