Skip to content

rocke implementation of fp8 mqa logits on gfx942 - #10287

Draft
yraparti wants to merge 2 commits into
developfrom
users/yraparti/rocke/fp8_mqa_logits
Draft

yraparti wants to merge 2 commits into
developfrom
users/yraparti/rocke/fp8_mqa_logits

Conversation

@yraparti

@yraparti yraparti commented Aug 3, 2026

Copy link
Copy Markdown
Contributor

Summary

  • Add a native gfx942 FP8 MQA-logits kernel with matching Python and C++ builders.
  • Add shape-adaptive geometry, KV splitting, fused-FMA epilogue, and selectable 16×16×32/32×32×16 FP8 MFMA atoms.
  • Add backend bindings, parity tests, verification example, AITER benchmark, optimization sweep, roofline analysis, and documentation.
  • Beat AITER PR flydsl: gfx942 FP8 MQA logits indexer kernel (+ Triton FN/FNUZ fix) aiter#3913 across all four measured workloads, with the wider atom providing further same-run gains on 4096×4096 and 128×32768.

Performance

  • AITER speedups: 1.093×, 1.049×, 1.169×, and 1.097×.
  • Wide-atom improvement over the rocKE baseline: 1.066× and 1.013×.
  • Wide atom halves static MFMA instructions and remains spill-free on selected shapes.

Test Plan

Test Result

Submission Checklist

@yraparti yraparti changed the title rocke implementation of fp8 mqa logits. Beats AITER implementation rocke implementation of fp8 mqa logits on gfx900 Aug 3, 2026
@yraparti yraparti changed the title rocke implementation of fp8 mqa logits on gfx900 rocke implementation of fp8 mqa logits on gfx942 Aug 3, 2026
@therock-pr-bot

therock-pr-bot Bot commented Aug 3, 2026

Copy link
Copy Markdown

❌ PR Check — Action Required

Check Status Details
📝 PR Description ❌ Fail Error: PR description must reference a JIRA ID, ISSUE ID, or a GitHub closing keyword.
Expected: include a JIRA ID / ISSUE ID line (separator : or -, or omitted; value may be a JIRA key, a number with/without #, or a link), OR a closing keyword + issue reference. Accepted examples:
JIRA ID : TESTAUTO-6039
JIRA ID - #330
JIRA ID #330
JIRA ID (on separate line)
ROCM-25757
ISSUE ID : TESTUTO-3334
ISSUE ID #3334
ISSUE ID - TESTAUTO-3433
ISSUE ID (on separate line)
AIRUNTIME-2352
ISSUE ID : https://github.com/<org_name>/<repo_name>/issues/1234
Closes #10
Fixes octo-org/octo-repo#100
Resolves: #123
#123
https://github.com/<org_name>/<repo_name>/issues/123
Current: no valid JIRA/ISSUE/closing-keyword reference found
Forbidden Files ✅ Pass
🧪 Unit Test ✅ Pass
🔎 pre-commit ⏳ Pending ⏳ Still running…
🚫 Draft PR 🔜 To Be Enabled
🚩 Feature Flag 🔜 To Be Enabled
📊 Code Coverage 🔜 To Be Enabled

⚠️ 1 policy check(s) failed. Please address the issues above before this PR can be Reviewed.

🚫 Please fix the failed policies

  • ❌ PR Description

The Not ready to Review label was added to this PR. Once all policies pass, the label is removed automatically.

📖 Need help? See the Policy FAQ for details on every check and how to fix failures.

🙋 Wish to Override Policy?

@therock-pr-bot

therock-pr-bot Bot commented Aug 3, 2026

Copy link
Copy Markdown

🚫 Please fix the failed policies before requesting reviews.

The following policy checks failed:

  • ❌ PR Description

The Not ready to Review label has been added to this PR.
Once all policies pass, the label will be removed automatically.

@github-actions github-actions Bot added the rocke: platform Touches rocKE platform label Aug 10, 2026
@github-actions

github-actions Bot commented Sep 5, 2026

Copy link
Copy Markdown
Contributor

This pull request has been inactive for 25 days and will be marked as stale.

If you would like to keep this PR open, please:

  • Add new commits
  • Add a comment explaining why it should remain open

This PR will be automatically closed in 5 days if no further activity occurs.

@github-actions github-actions Bot added the Stale PR has no activity for 25+ days label Sep 5, 2026
@github-actions

Copy link
Copy Markdown
Contributor

This pull request has been automatically closed due to inactivity (30 days with no updates).

If you'd like to continue working on this, feel free to reopen the PR or create a new one.

@github-actions github-actions Bot closed this Sep 10, 2026
@yraparti yraparti removed the Stale PR has no activity for 25+ days label Sep 14, 2026
@yraparti yraparti reopened this Sep 14, 2026
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.

1 participant