Issue: Failed to load libtilert.so due to undefined symbol _ZN3c104cuda29c10_cuda_check_implementationEiPKcS2_ib
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 1.8k
- Forks
- 124
- PR merge metrics
- No merged PRs in 30d
Description
Issue: Failed to load libtilert.so due to undefined symbol _ZN3c104cuda29c10_cuda_check_implementationEiPKcS2_ib
Description
When running the generate script in a Docker environment, I encounter an error loading the libtilert.so library due to an undefined symbol related to PyTorch's CUDA checks.
Error Log
(tilert) [root@d9878fa39ee8 workspace]# python generate.py --model-weights-dir "/home/model_weights" 2>&1 | tee generate.log
Traceback (most recent call last):
File "/opt/conda/envs/tilert/lib/python3.12/site-packages/tilert/__init__.py", line 38, in _load_library
return ctypes.CDLL(str(lib_path))
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/envs/tilert/lib/python3.12/ctypes/__init__.py", line 379, in __init__
self._handle = _dlopen(self._name, mode)
^^^^^^^^^^^^^^^^^^^^^^^^^
OSError: /opt/conda/envs/tilert/lib/python3.12/site-packages/tilert/libtilert.so: undefined symbol: _ZN3c104cuda29c10_cuda_check_implementationEiPKcS2_ib
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/workspace/generate.py", line 9, in <module>
from tilert import logger, tilert_init
File "/opt/conda/envs/tilert/lib/python3.12/site-packages/tilert/__init__.py", line 43, in <module>
_load_library("libtilert.so")
File "/opt/conda/envs/tilert/lib/python3.12/site-packages/tilert/__init__.py", line 40, in _load_library
raise RuntimeError(f"Failed to load library from {lib_path}") from e
RuntimeError: Failed to load library from /opt/conda/envs/tilert/lib/python3.12/site-packages/tilert/libtilert.so
Environment
- Running in Docker container
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Reproduce the failure with generate.py in the Docker environment, then inspect tilert/init.py where libtilert.so is loaded and compare the installed PyTorch/CUDA environment with the library's requirements. Done means generate.py loads libtilert.so without the undefined-symbol error and proceeds past the import.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- docker, python, pytorch
- Domain
- devops, machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 42/100