Skip to content

feat: rmsnorm lowering #3440

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Apr 18, 2025
Merged

feat: rmsnorm lowering #3440

merged 1 commit into from
Apr 18, 2025

Conversation

bowang007
Copy link
Collaborator

RMSNORM lowering pass

Checklist:

  • My code follows the style guidelines of this project (You can use the linters)
  • I have performed a self-review of my own code
  • I have commented my code, particularly in hard-to-understand areas and hacks
  • I have made corresponding changes to the documentation
  • I have added tests to verify my fix or my feature
  • New and existing unit tests pass locally with my changes
  • I have added the relevant labels to my PR in so that relevant reviewers are notified

Sorry, something went wrong.

@github-actions github-actions bot added component: lowering Issues re: The lowering / preprocessing passes component: api [Python] Issues re: Python API component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths labels Mar 12, 2025
@github-actions github-actions bot requested a review from gs-olive March 12, 2025 18:05
github-actions[bot]

This comment was marked as outdated.

@github-actions github-actions bot added component: conversion Issues re: Conversion stage and removed component: lowering Issues re: The lowering / preprocessing passes labels Apr 10, 2025
@bowang007 bowang007 marked this pull request as ready for review April 10, 2025 23:25
@bowang007 bowang007 requested review from peri044 and zewenli98 and removed request for gs-olive April 10, 2025 23:26
This commit supports lowering rmsnorm module to flashinfer.rmsnorm.
The example can be found in the PR which demonstrates how to lower the
rmsnorm in pytorch to flashinfer.rmsnorm and run the lowered node using
flashinfer library based on automatic plugin feature.

This PR also resolves unique ID issue when creating constant layers.
@bowang007 bowang007 force-pushed the flashinfer_rmsnorm_lowering branch from cc199f1 to 606f242 Compare April 17, 2025 23:41
Copy link
Collaborator

@narendasan narendasan left a comment

Choose a reason for hiding this comment

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

LGTM

Copy link
Collaborator

@zewenli98 zewenli98 left a comment

Choose a reason for hiding this comment

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

LGTM!

@bowang007 bowang007 merged commit 29b65b0 into main Apr 18, 2025
71 of 72 checks passed
@zewenli98 zewenli98 mentioned this pull request Apr 21, 2025
7 tasks
peri044 pushed a commit that referenced this pull request Apr 27, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
cla signed component: api [Python] Issues re: Python API component: conversion Issues re: Conversion stage component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths needs-release-cherrypick
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

4 participants