[zhaochenyang20]

What does this PR do?

As the name goes, a quick bug fix.

Checklist Before Starting

• Search for similar PRs. Paste at least one query link here: ...
• Format the PR title as [{modules}] {type}: {description} (This will be checked by the CI)
  • {modules} include fsdp, megatron, sglang, vllm, rollout, trainer, ci, training_utils, recipe, hardware, deployment, ray, worker, single_controller, misc, perf, model, algo, env, tool, ckpt, doc, data
  • If this PR involves multiple modules, separate them with , like [megatron, fsdp, doc]
  • {type} is in feat, fix, refactor, chore, test
  • If this PR breaks any API (CLI arguments, config, function signature, etc.), add [BREAKING] to the beginning of the title.
  • Example: [BREAKING][fsdp, megatron] feat: dynamic batching

Test

For changes that can not be tested by CI (e.g., algorithm implementation, new model support), validate by experiment(s) and show results like training curve plots, evaluation results, etc.

API and Usage Example

Demonstrate how the API changes if any, and provide usage example(s) if possible.

# Add code snippet or script demonstrating how to use this

Design & Code Changes

Demonstrate the high-level design if this PR is complex, and list the specific changes.

Checklist Before Submitting

Important

Please check all the following items before requesting a review, otherwise the reviewer might deprioritize this PR for review.

• Read the Contribute Guide.
• Apply pre-commit checks: pre-commit install && pre-commit run --all-files --show-diff-on-failure --color=always
• Add / Update the documentation.
• Add unit or end-to-end test(s) to the CI workflow to cover all the code. If not feasible, explain why: ...
• Once your PR is ready for CI, send a message in the ci-request channel in the verl Slack workspace.

[gemini-code-assist]

Code Review

This pull request introduces a bug fix in the FSDPSGLangShardingManager. The change ensures that resuming memory occupation for the KV cache during the multi-stage wake-up process is only performed on rank 0 of the inference tensor parallelism group. This aligns with the existing logic for other inference engine operations, preventing potential race conditions or redundant operations across multiple ranks. The change is correct and improves the robustness of the distributed setup.

[hebiao064]

I've fixed in my PR: https://github.com/volcengine/verl/pull/2187/files#diff-5872c25af9ac22eee90d7163b2e9efdce7ef2c1f1c0293590ad58ad88b497a05

But feel free to merge your fix first since my PR is still blocked by CI