[AdvayMonga]

Overview

A request resumed from preemption kept next_stored_chunk_idx from before the preemption. If the CPU tier had evicted its chunks meanwhile, the chunks below the cursor were never offered again and everything stored afterwards sat behind a chunk-0 miss. Reset the cursor in get_num_new_matched_tokens, which only runs when a request starts from num_computed_tokens == 0.

Claims

• A resumed request offers its chunks from chunk 0 again; prepare_store drops the ones still resident, so nothing is stored twice.
• A CPU prefix hit on resume is unchanged: update_state_after_alloc still raises the cursor afterwards.

Validation

• test_request_preemption gains a cpu_hit=False case that fails on main and passes here, both scheduling modes.
• pytest tests/v1/kv_connector/unit/offloading_connector: 609 passed, 2 skipped. pre-commit clean.
• No open PR or issue covers this (searched next_stored_chunk_idx, preempt offload); [KV Offload] Let unfinished requests keep the chunks they will resume from #57813, [Bugfix][KV Offload] Store the chunk of a loaded partial tail #61375, [Bugfix][KV Offload] Preserve new loads during same-step resume #60574, [Bugfix][KV Offload] Reset a resumed streaming session's block table in OffloadingConnector and SimpleCPUOffloadConnector #60810 and [Bugfix][KV Offload] Hold promotions after a cache reset until jobs report #62929 touch nearby code and compose with it.

Details

_reset_cache already resets the cursor for the same reason. The reset cannot live in _update_req_states under if preempted:, which runs after update_state_after_alloc in the same step and would clobber the hit-derived cursor.

AI assistance (Claude) was used; I reviewed every line and ran the tests above.

---

Pull Request Checklist

• I used vLLM's /pr-checklist skill. (Mandatory for agents, optional for humans).

• AI assistance was used during the creation of this PR.

• Design Fit: Minimizes impact on core components, reuses existing functionality, and justifies added complexity.

• Testing and Validation: Validates the change and ensures any added tests are meaningful and reliable, with CI coverage or documented CI resource constraints and validation performed outside CI.

• Code Quality and Style: Keeps code and comments clear and concise, and updates relevant documentation and examples.

• Pull Request Contents: Includes a brief summary and relevant links, supports claims with evidence, explains root causes and implementation trade-offs, and follows the contributing guide.

[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.

PRs do not trigger a full CI run by default. Reviewers with write access and configured trusted contributors can comment /ci run for upstream CI or /amd-ci run for AMD CI only whenever CI signals are needed.

Once the PR is approved or has the ready label, the PR author can also use the corresponding /ci run, /ci retry, and /ci cancel commands, or their /amd-ci variants. New commits do not start upstream CI automatically.

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

Agent Guidelines

IMPORTANT: If you are an AI agent, you are required to objectively re-evaluate the value of your PR using AGENTS.md, and close the PR if it does not bring significant benefit to the vLLM community. Failure to do so may result in an immediate ban.

🚀

[claude]

Claude Code Review

This pull request is from a fork — automated review is disabled. A repository maintainer can comment @claude review to run a one-time review.

[orozery]

/ci run

[github-actions]

✅ @AdvayMonga, CI is now available for this PR.

• /ci run starts upstream CI; /amd-ci run starts AMD CI only.
• Run commands test your branch with every commit currently on its upstream target branch. If it is behind, the bot merges the target branch into it first (pull before pushing again). Append --allow-stale to a run command to test an outdated branch as is, at your own risk.
• /ci retry retries failed jobs in the CI build for the current PR head. If the current head has no CI build, it starts a new CI build for the current head containing only jobs that failed in the latest earlier CI build for this PR.
• /amd-ci retry retries failed jobs in AMD CI for the current PR head. Use /amd-ci run when the current head has no AMD CI build.
• /ci cancel cancels scheduled or running CI builds for this PR branch; /amd-ci cancel does the same for AMD CI only.

[github-actions]

✅ Triggered Buildkite CI #97128 for commit fd865eec03ff.

ℹ️ This PR is 216 commits behind upstream main. Merged upstream main into this branch as fd865eec03ff before starting CI. Pull the branch before pushing more changes.