[Klaud Cold] Re-introduce dsv4-fp8-mi355x-sglang on rocm/sgl-dev DSv4 image#2220
[Klaud Cold] Re-introduce dsv4-fp8-mi355x-sglang on rocm/sgl-dev DSv4 image#2220Oseltamivir wants to merge 1 commit into
Conversation
Oseltamivir
requested review from
1am9trash,
billishyahao,
chunfangamd,
seungrokj and
yctseng0211
as code owners
… image Re-adds the dsv4-fp8-mi355x-sglang recipe and its launch script, removed in #1501 when a generic sglang-rocm image bump broke the DSV4-Pro-FP8 downcast check. Pins to the DSv4-specific custom build rocm/sgl-dev:rocm720-mi35x-da28108-20260610-DSv4 and restores the script under the current fixed_seq_len/ layout as dsv4_fp8_mi355x_sglang.sh.
Oseltamivir
force-pushed
the
klaud-cold/dsv4-fp8-mi355x-sglang-restore
branch
from
0000f29 to
31678ac
Compare
|
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 For PR verification, add the 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 如需进行 PR 验证，请为此 PR 添加 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 |
1 similar comment
|
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 For PR verification, add the 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 如需进行 PR 验证，请为此 PR 添加 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 |
![claude[bot]](https://avatars.githubusercontent.com/in/1236702?s=60&v=4){kind=small}
There was a problem hiding this comment.
LGTM — straightforward re-introduction of the dsv4-fp8-mi355x-sglang recipe on the DSv4-specific image, following the standard recipe-add pattern.
What was reviewed: the new launch script's server flags/env vars against sibling dsv4 sglang scripts, the config-patch heredoc's local-path handling, the DP-attention/max-running-request/mem-fraction-static choices flagged by the bug hunter (checked against other dsv4/qwen mi355x sglang recipes and ruled out as intentional tuning choices, not bugs), and the amd-master.yaml / perf-changelog.yaml additions for consistency with existing entries.
Extended reasoning...
Overview
This PR re-adds a previously-removed benchmark recipe (dsv4-fp8-mi355x-sglang) that was pulled in #1501 after a generic image bump broke server boot on the DSV4-Pro-FP8 checkpoint's mixed bf16/fp8 layers. It pins to a DSv4-specific rocm/sgl-dev custom image tag instead, adds a launch script under benchmarks/single_node/fixed_seq_len/, a corresponding amd-master.yaml block, and a perf-changelog entry. This is exactly the shape of change the repo's own add-model-hardware/nuke workflows produce and is a very common, well-trodden PR pattern in this codebase.
Security risks
None. This is benchmark-harness configuration and a shell launch script for an internal GPU sweep; there is no user-facing surface, auth, or credential handling. The Python config-patch heredoc calls hf_hub_download with a repo id sourced from an internal config value, not untrusted input.
Level of scrutiny
Low-to-moderate is appropriate. This is a single self-contained benchmark recipe addition (config + script + changelog) with no changes to shared launcher/eval infrastructure, CI logic, or any code path outside this one recipe. It follows the same structure as dozens of sibling recipes already in amd-master.yaml, and the image pin plus rationale comment directly address the specific regression that caused the prior removal.
Other factors
The bug-hunting system found no bugs, and the candidate issues it raised (hardcoded --max-running-request, DP-attention always on, missing --mem-fraction-static, unconditional hf_hub_download call) were all examined and refuted as intentional recipe choices rather than defects. The PR description itself flags that serving flags may need on-cluster tuning based on sweep results, which is the normal iteration loop for this kind of recipe PR and not a blocker for merging — the accompanying full-sweep-fail-fast label will validate server boot and eval on real hardware.
|
see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=29415644307 |
2 participants
Summary
Re-introduces the
dsv4-fp8-mi355x-sglangbenchmark (DeepSeek-V4-Pro-FP8 on MI355X via SGLang), removed in #1501, with an updated DSv4-specific image.Why it was removed, and why this works
#1501 removed the recipe because bumping it off the custom
rocm/sgl-devbuild to a genericlmsysorg/sglang-rocmrelease broke server boot:DSV4-Pro-FP8 keeps some bf16 layers (shared-expert / embedding); only the DSv4-specific
rocm/sgl-devcustom builds tolerate this. This PR pins torocm/sgl-dev:rocm720-mi35x-da28108-20260610-DSv4(a-DSv4custom build), not a plain release. A config comment records this so future bumps do not naively re-break it.Changes
configs/amd-master.yaml: restoredsv4-fp8-mi355x-sglangblock on the new image (tp8, 1k1k + 8k1k, conc 4–64).benchmarks/single_node/fixed_seq_len/dsv4_fp8_mi355x_sglang.sh: restore launch script under the currentfixed_seq_len/layout +_sglangnaming (the old flatdsv4_fp8_mi355x.shno longer matches launcher routing). Serving flags/env carried over from the last-known-working version (config model_type patch,--attention-backend compressed,--disable-cuda-graph, DSv4 FP8 env vars).Test plan
Note: serving flags/env were validated against the April
deepseek-v4-mi35xbuild; the Juneda28108-...-DSv4build may need on-cluster tuning, which I will drive from the sweep results.