NVIDIA / NVIDIA/TransformerEngine
Using system CUDA libraries
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 3.5k
- Forks
- 831
- Avg merge
- 3d 11h
- Merged PRs (30d)
- 65
Description
Describe the bug
transformer-engine is currently searching for system CUDA binaries ( https://github.com/NVIDIA/TransformerEngine/blob/67fcc15255248a26be124de3854a47f84102f285/transformer_engine/common/__init__.py#L237). This is in conflict with Pytorch, which uses the CUDA Python packages (https://pypi.org/project/nvidia-cudnn-cu12/).
Steps/Code to reproduce bug
Tried using transformer-engine in a Docker container that did not have a system CUDA installed.
Expected behavior
transformer-engine should find the CUDA libraries inside the CUDA Python packages. Example:
import nvidia.cudnn
nvidia.cudnn.__file__
lib_path = os.path.join(nvidia.cudnn.__file__, "lib")
Contributor guide
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
Start in transformer_engine/common/init.py at the linked line around 237 and inspect how CUDA libraries are currently located. Compare that behavior with the nvidia.cudnn package path shown in the issue, then reproduce the Docker scenario without system CUDA and verify that transformer-engine finds the packaged libraries.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 56/100