Conversation
`apply_group_offloading` with `use_stream=True` only created a stream for CUDA or Intel XPU devices, and raised "Using streams for data transfer requires a CUDA device, or an Intel XPU device." on every other accelerator. On Ascend NPU, `torch.npu.Stream()` is available (torch_npu exposes it), so stream-based onload/offload should be supported there too. Adds an `npu` branch consistent with the existing CUDA/XPU checks, and updates the error message to mention the Ascend NPU device.
Contributor
|
Hi @li-lizhe, thanks for the PR! It does not appear to link an issue it fixes. If this PR addresses an existing issue, please add a closing keyword (e.g. Please note that PRs without a linked issue are likely to be automatically closed 10 days after this notice. Once the PR links an issue (or gets the |
10 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
apply_group_offloadingwithuse_stream=Trueonly created a stream for CUDA (viatorch.cuda.Stream()) or Intel XPU (viatorch.Stream()), and raisedUsing streams for data transfer requires a CUDA device, or an Intel XPU device.on every other accelerator.On Ascend NPU,
torch.npu.Stream()is available (exposed bytorch_npu), so stream-based onload/offload should be supported there too.Changes:
npubranch consistent with the existing CUDA/XPU checks (hasattr(torch, "npu") and torch.npu.is_available() -> torch.npu.Stream())Testing:
Verified on an Ascend 910B NPU (torch 2.14 + torch_npu):
ValueError(confirmedtorch.cuda.is_available()=False,torch.xpu.is_available()=False)torch.npu.Stream()constructs successfullyhasattr(torch, "npu")=True,torch.npu.is_available()=True5 lines added — minimal, device-agnostic, consistent with existing accelerator dispatch pattern.