[BugFix] AttributeError: 'AsyncOmniDiffusion' object has no attribute 'abort'#624
Conversation
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 6e4198af9f
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
| def abort(self, request_id: str | Iterable[str]) -> None: | ||
| # TODO implement it | ||
| pass |
There was a problem hiding this comment.
This new abort implementation is a no-op, so when a request is cancelled (e.g., AsyncOmni sends an ABORT task for a diffusion stage), the generation continues running to completion and only the output is dropped later. That means cancelled requests still consume GPU/CPU and queue slots, which can become significant under high cancellation rates or timeouts. If abort can’t be implemented yet, consider explicitly raising/notifying so callers can handle the lack of cancellation instead of assuming it succeeded.
Useful? React with 👍 / 👎.
… 'abort' (vllm-project#624) Signed-off-by: ZeldaHuang <hzm414167@alibaba-inc.com> Signed-off-by: DefTruth <qiustudent_r@163.com>
… 'abort' (vllm-project#624) Signed-off-by: ZeldaHuang <hzm414167@alibaba-inc.com>
… 'abort' (vllm-project#624) Signed-off-by: ZeldaHuang <hzm414167@alibaba-inc.com>
… 'abort' (vllm-project#624) Signed-off-by: ZeldaHuang <hzm414167@alibaba-inc.com>
… 'abort' (vllm-project#624) Signed-off-by: ZeldaHuang <hzm414167@alibaba-inc.com>
3 participants
PLEASE FILL IN THE PR DESCRIPTION HERE ENSURING ALL CHECKLIST ITEMS (AT THE BOTTOM) HAVE BEEN CONSIDERED.
ref #605
Purpose
temporary fix the issue, need to implement diffsion engine abort later.
Test Plan
Test Result
Essential Elements of an Effective PR Description Checklist
supported_models.mdandexamplesfor a new model.BEFORE SUBMITTING, PLEASE READ https://github.com/vllm-project/vllm-omni/blob/main/CONTRIBUTING.md (anything written below this line will be removed by GitHub Actions)