[Bugfix] Update import path for bc_linter_include

[mmangkad]

Purpose

Resolves an ImportError that occurs when loading certain models due to a fragile top-level import within the model inspection subprocess. The exact error is:

ImportError: cannot import name 'bc_linter_include' from 'vllm' (unknown location)

This PR fixes the issue by updating the import path in vllm/v1/core/sched/output.py to be more direct and robust.

Test Plan

Test Result

---

Essential Elements of an Effective PR Description Checklist

• The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
• The test plan, such as providing test command.
• The test results, such as pasting the results comparison before and after, or e2e results
• (Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model.
• (Optional) Release notes update. If your change is user facing, please update the release notes draft in the Google Doc.

[gemini-code-assist]

Code Review

This pull request addresses an ImportError by changing the import path for bc_linter_include. The modification from a top-level import to a direct import from vllm._bc_linter is a correct and robust fix. This change not only resolves the immediate error but also improves code structure by making the dependency explicit and avoiding potential circular import issues that can arise from fragile top-level imports. The change is well-contained and appropriate.

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors.

You ask your reviewers to trigger select CI tests on top of fastcheck CI.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

If you have any questions, please reach out to us on Slack at https://slack.vllm.ai.

🚀

[yaochengji]

Hi @mmangkad , may I ask why you close this PR? It can basically fix the issue that throws import error if a vLLM is not executed inside the vLLM directory.

[mmangkad]

Hi @mmangkad , may I ask why you close this PR? It can basically fix the issue that throws import error if a vLLM is not executed inside the vLLM directory.

Hi @yaochengji, I mistakenly thought that this wasn't necessary anymore. Reopening this for review and approval. Thanks!

[yaochengji]

LGTM, thanks for the fix!