don't initialize cuda at import time #3244
Conversation
|
Hi @technillogue! Thank you for your pull request and welcome to our community. Action RequiredIn order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you. ProcessIn order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA. Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with If you have received this in error or have any questions, please contact us at cla@meta.com. Thanks! |
|
Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks! |
| @@ -307,7 +307,7 @@ def _parse_compile_spec(compile_spec_: Dict[str, Any]) -> _ts_C.CompileSpec: | |||
| def TensorRTCompileSpec( | |||
| inputs: Optional[List[torch.Tensor | Input]] = None, | |||
| input_signature: Optional[Any] = None, | |||
| device: torch.device | Device = Device._current_device(), | |||
| device: torch.device | Device | None = None, | |||
There was a problem hiding this comment.
Could do Optional[torch.device | Device]
technillogue
force-pushed
the
no-init-cuda-at-import
branch
from
71c6c06
to
6367d04
Compare
There was a problem hiding this comment.
@technillogue This looks good. Can you make the device and Optional type as Naren pointed out ? Then, we can merge
technillogue
force-pushed
the
no-init-cuda-at-import
branch
from
6367d04
to
1659796
Compare
|
I've made the change |
Description
Currently importing
torch_tensorrtcalls_cuda_init, which can fail, preventing torch_tensorrt being imported in environments without a (working) GPU. This makes it slow to import torch_tensorrt and isn't necessary. This change switches to determining the current devices when TensorRTCompileSpec is called rather than when it's defined.Type of change
Checklist: