Skip to content

Recover PR #2158 database ingest via sweep reuse / 通过扫描复用恢复 PR #2158 数据库摄取#2222

Merged
cquil11 merged 8 commits into
mainfrom
recovery/reuse-pr-2158
Jul 15, 2026
Merged

Recover PR #2158 database ingest via sweep reuse / 通过扫描复用恢复 PR #2158 数据库摄取#2222
cquil11 merged 8 commits into
mainfrom
recovery/reuse-pr-2158

Conversation

@cquil11

@cquil11 cquil11 commented Jul 15, 2026

Copy link
Copy Markdown
Collaborator

Summary

Recover the missing official database ingest for PR #2158 by reusing its validated AgentX sweep artifacts from source run 29359203708.

The original main-branch run 29425167775 downloaded the artifacts successfully but failed before ingestion because nested KV offload backend metadata was not hashable. PR #2221 fixed that validator defect.

This recovery will append the original changelog scope, pin the successful source run, validate all reusable artifacts, and use the normal post-merge ingest path without rerunning GPU benchmarks.

中文说明

通过复用源运行 29359203708 中已验证的 AgentX 扫描产物,恢复 PR #2158 缺失的正式数据库摄取。

原始 main 分支运行 29425167775 已成功下载产物,但由于嵌套 KV 卸载后端元数据不可哈希,在摄取前失败。PR #2221 已修复该校验器缺陷。

本恢复 PR 将追加原始 changelog 范围、固定成功的源运行、验证全部可复用产物,并通过正常的合并后摄取路径完成恢复,不会重新运行 GPU 基准测试。

xinli-sw and others added 6 commits July 14, 2026 09:58
…mark

- Add kimik2.5-fp4-b300-vllm-agentic-mtp config (runner: cluster:b300-nv)
- Add benchmark script with EAGLE3 speculative decoding (draft: lightseekorg/kimi-k2.6-eagle3-mla, num_speculative_tokens=4, synthetic_acceptance_length=3.24)
- TOKENSPEED_MLA attention backend + TRTLLM_RAGGED prefill backend
- Search space: TP8 no-offload, TP4 no-offload, TP4 native CPU-offload, TP4 DCP4 no-offload, TP4 DCP4 native CPU-offload
- Add cluster:b300-nv runner label and hardware entry
- Add spec-decoding field propagation for single-node agentic matrix
- Add perf changelog entry for PR 2158
中文:准备恢复 PR #2158 的数据库摄取。

cquil11 commented Jul 15, 2026

Copy link
Copy Markdown
Collaborator Author

/reuse-sweep-run 29359203708

cquil11 added 2 commits July 15, 2026 10:14
中文:通过复用扫描产物恢复 PR #2158 的数据库摄取。
中文:附加可复用扫描运行 29359203708 的源提交。
@cquil11 cquil11 requested a review from a team July 15, 2026 15:20
@github-actions

Copy link
Copy Markdown
Contributor

Thanks for the contribution! Please reach out to respective companies' CODEOWNER to fill in the latest PR_REVIEW_CHECKLIST.md before pinging core maintainer on Slack for review. In order for the signoff PR check bot to trigger, you must follow the PR_REVIEW_CHECKLIST.md template correctly, including the phrase As a PR reviewer and CODEOWNER, I have reviewed this and have.

For PR verification, add the full-sweep-fail-fast label (strongly recommended) to this PR — the benchmark sweep only runs on labeled PRs. Use full-sweep-enabled only if you need matrix jobs to keep running past a failure.

PR authors are responsible for ensuring that after merging, all GitHub Action jobs fully pass. A lot of the time, failures are just flakes and simply re-running the failed jobs will fix it. See GitHub's docs on re-running failed jobs


感谢你的贡献!请联系相应公司的 CODEOWNER 填写最新的 PR_REVIEW_CHECKLIST.md,然后再在 Slack 上联系核心维护者进行审阅。为了触发 signoff PR 检查机器人,你必须正确遵循 PR_REVIEW_CHECKLIST.md 模板,包括保留英文语句 As a PR reviewer and CODEOWNER, I have reviewed this and have

如需进行 PR 验证,请为此 PR 添加 full-sweep-fail-fast 标签(强烈推荐)— 基准测试 sweep 仅在带有标签的 PR 上运行。仅当需要矩阵任务在失败后继续运行时才使用 full-sweep-enabled

PR 作者有责任确保合并后所有 GitHub Action 任务完全通过。 很多时候失败只是偶发抖动(flake),重新运行失败的任务即可解决。参见 GitHub 关于重新运行失败任务的文档

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

Development

Successfully merging this pull request may close these issues.

2 participants