Skip to content

fix(module_rmsnorm_quant): bound packed FP4 output stores - #4467

Merged
valarLip merged 3 commits into
mainfrom
fix/fp4-rmsnorm-packed-output-bounds
Aug 3, 2026
Merged

fix(module_rmsnorm_quant): bound packed FP4 output stores#4467
valarLip merged 3 commits into
mainfrom
fix/fp4-rmsnorm-packed-output-bounds

Conversation

@gbyu-amd

Copy link
Copy Markdown
Contributor

Use packed-byte row bounds for fused FP4 RMSNorm quantization and add a guard-row regression test to prevent cross-row writes.

Motivation

Technical Details

Test Plan

Test Result

Submission Checklist

Use packed-byte row bounds for fused FP4 RMSNorm quantization and add a guard-row regression test to prevent cross-row writes.

Co-authored-by: Cursor <cursoragent@cursor.com>
@gbyu-amd
gbyu-amd requested a review from a team July 30, 2026 17:00
@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:triton-300x Run an additional Triton test job on MI300X in PRs; main branch always runs both MI35X and MI300X
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: GPT-OSS-120B, DeepSeek-R1-0528, Kimi-K2.5
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 4467 --add-label <label>

Exercise the packed-output canary through the existing FP4 benchmark path instead of an unconditional main entry.

Co-authored-by: Cursor <cursoragent@cursor.com>
@valarLip
valarLip merged commit fa3a8ec into main Aug 3, 2026
21 checks passed
@valarLip
valarLip deleted the fix/fp4-rmsnorm-packed-output-bounds branch August 3, 2026 11:15
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants