Repository navigation
Conversation
[bot] 🏷 (ai: vllm/qwen3.8-27b) 查重: 不重复. PR #6766 is the only fix for issue #6752. No parallel PRs found. ReviewThe fix is correct and complete. The watermark fastpath's rebind logic ( CI: lane-guard failure (fixed in this review)The 八问 (structured review)
判定修复正确、完整、最小化。lane-guard CI 失败已修复 (添加了边界固定测试)。建议合并。 |
[bot] 🏷 (ai: vllm/qwen3.8-27b) CI is now fully green on the updated head ( |
…nxianglei#6752) On a watermark fastpath hit, prepareOpenai re-parses scalar fields from the raw wire text, silently discarding the max_tokens restore that restoreOutputBudget had just applied upstream of it. The log still reports the restore while the forwarded body carries the client's shrunk value (max_tokens: 1). Snapshot the restored budget from the pre-rebind body and write it back after the rebind. The output-budget field plumbing moves from side-request.ts to budget.ts so the prepare modules can use it without reaching into the side-lane engine (ranxianglei#1440 P2 cut 2); side-request.ts re-exports it, so existing importers are untouched. Regression test: hermetic A/B golden over two proxies (rollout off vs on) asserting the starved turn forwards the restored ceiling on both paths, with a residency check proving the fastpath actually hit.
…ranxianglei#6752) The ranxianglei#1440 P2 cut 2 move (output-budget plumbing side-request.ts -> budget.ts) trips the lane-guard (ranxianglei#2170): side-request.ts changed without a side-lane suite update. Pin the re-export identity so the re-export and the definition can never silently diverge.
5bd8909 to
0299701
Compare
(ai: vllm-qwen/qwen3.8-27b) Merge conflict resolved & rebased onto current master. PR #6766 is now What I did this run:
八问:
判定: 冲突已解、已 rebase 到当前 master 并 force-push,PR 现可合入(mergeable: True)。CI 在新 head 上运行中(0 failed / 12 pending),无红;若 CI 转红我会继续修。 |
[ework] CI fix verified green ✓ |
…nxianglei#6752) When compat.dropFields removes a budget field that restoreOutputBudget just restored, the drop log now says so explicitly so the log is honest (the restore fired, but the field is absent from the forwarded body). One-line note appended to both drop-log sites (graph and string paths).
[ework] auto-fix stalled: architecture-docs — stopping for human review |
1 similar comment
[ework] auto-fix stalled: architecture-docs — stopping for human review |
# Conflicts: # src/server/budget.ts
|
Your fix is merged — the watermark-rebind budget carry and the F2 re-export identity test landed via the wave-11 batch merge (#6827 → master One follow-up on this PR was pushed after the batch snapshot and is therefore not in master yet: GitHub did not auto-flip this PR to Merged for the same reason as the others (the head moved after the snapshot). Closing as merged-via-batch, pending that one carry-over commit. |
Fixes #6752
What broke
A dsh turn with a client-shrunk output budget (max_tokens: 1) gets rescued by restoreOutputBudget, and the log reports the restore, but on a watermark fastpath hit prepareOpenai re-parses the scalar fields from the raw wire text, so the forwarded body still carries max_tokens: 1. Upstream records completion_tokens = 1 every turn.
What changed
How verified
ctx.imageTokens又被加一次 #2558 clamp-image (117 tests).