Skip to content

[HIP] [CK] Revert " Re-submit Relu2 Tier 2 MoE GEMM support (clean CK pin)" - #6460

Merged
zufayu merged 1 commit into
mainfrom
revert-6376-feature/relu2-moe-gemm-tier2
Oct 13, 2026
Merged

zufayu merged 1 commit into
mainfrom
revert-6376-feature/relu2-moe-gemm-tier2

Conversation

@gyohuangxin

@gyohuangxin gyohuangxin commented Oct 13, 2026 •

Copy link
Copy Markdown
Member

Reverts #6376

The vLLM failure was caused by commit 7cb93dd (#6376), which changed CK-Tile MoE dispatch/codegen and left the MiniMax-M3-MXFP4 Swiglu + has_bias=false + split_k=1 stage2 path without a generated kernel instance.

@gyohuangxin
gyohuangxin requested review from a team and a balanced review from Copilot October 13, 2026 03:29

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

@github-actions

Copy link
Copy Markdown
Contributor

🏷️ CI Guide

Runs automatically on every PR:

  • ✅ Pre-checks (submodule verification, code formatting)
  • ✅ Aiter op tests (gfx942 + gfx950)
  • ✅ Triton tests on MI35X (only when aiter/ops/triton/** or related paths are changed)

Extended tests (opt-in via labels):

Label Tests
ci:gfx1250-ffm-triton Run the five-shard gfx1250 FFM Triton test suite
ci:triton-300x Run an additional Triton test job on MI300X in PRs (added automatically when gfx942 configs change); main branch always runs both MI35X and MI300X
ci:triton-355 Run the full Triton test suite on MI35X, not only the tests the change affects
ci:windows Windows JIT op smoke test on the gfx1201 ROCm runner
ci:multi-gpu Aiter multi-GPU tests on the 8-GPU runner
ci:sglang SGLang integration tests: DeepSeek-R1-MXFP4 accuracy, Qwen 3.5 accuracy
ci:atom ATOM benchmark: DeepSeek-R1-0528, GPT-OSS-120B
ci:atom_full ATOM accuracy suite for PR and main models from ATOM models_accuracy.json
ci:vllm vLLM benchmark: DeepSeek-V4-Pro, MiniMax-M3-MXFP4
ci:all All standard extended tests (excludes ci:atom_full)

Only add ci:atom_full for FlyDSL or Triton upgrades.
Add labels via the sidebar or gh pr edit 6460 --add-label <label>

One backend per PR:
A PR changes one kernel backend: [Triton/Gluon] (Triton and Gluon count as one), [HIP], [ASM], [CK], [OPUS] or [FlyDSL]. If the title ends up with two backend tags, split the PR -- as stacked pull requests when one part cannot merge without the other.

PR title tags & labels:
Component tags ([Triton/Gluon], [HIP], [CK], [ASM], ...) are added to the PR title and as PR labels automatically from the changed files and re-synced on every push — change-type tags like [fix]/[Perf], op tags like [MLA], and human labels (ci:*) are left untouched. Add the no-auto-title label to stop the title rewrites; labels stay in sync either way.

@github-actions github-actions Bot changed the title Revert "[HIP] [CK] Re-submit Relu2 Tier 2 MoE GEMM support (clean CK pin)" [HIP] [CK] Revert " Re-submit Relu2 Tier 2 MoE GEMM support (clean CK pin)" Oct 13, 2026

@zufayu zufayu left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed. This cleanly reverts #6376 / 7cb93dd without unrelated changes, and I do not see any code issue blocking the rollback. This is a low-risk emergency fix for the MiniMax-M3-MXFP4 regression. Please merge once the required pre-checks and ci:vllm pass. The missing dispatch coverage can be fixed and re-landed separately.

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

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants