Skip to content

Enable subwarp only for unweighted #2051

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

Closed
wants to merge 2 commits into from
Closed

Conversation

sryap
Copy link
Contributor

@sryap sryap commented Sep 28, 2023

Summary:
Disable the subwarp optimization for weighted TBE

  • The cause of the NE issue has to be investigated.
  • This does not affect the overall E2E performance much since the
    majority of TBEs in the model is unweighted

Differential Revision: D49159119

@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D49159119

@netlify
Copy link

netlify bot commented Sep 28, 2023

Deploy Preview for pytorch-fbgemm-docs canceled.

Name Link
🔨 Latest commit f30ac93
🔍 Latest deploy log https://app.netlify.com/sites/pytorch-fbgemm-docs/deploys/6524529e6c2aab0008ae675c

@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D49159119

sryap added a commit to sryap/FBGEMM that referenced this pull request Oct 9, 2023
Summary:
Pull Request resolved: pytorch#2051

Disable the subwarp optimization for weighted TBE

- The cause of the NE issue has to be investigated.
- This does not affect the overall E2E performance much since the
  majority of TBEs in the model is unweighted

Reviewed By: jasonjk-park

Differential Revision: D49159119

fbshipit-source-id: a70c2fc8a55aaa4e04e8285b20d9dd8ab7069ed5
@sryap sryap force-pushed the export-D49159119 branch from 7a0d501 to b096627 Compare October 9, 2023 19:16
sarunya and others added 2 commits October 9, 2023 12:19
Differential Revision: D49605180

fbshipit-source-id: de1824908ba741e4c631e91b69a0e17c740d040e
Summary:
Pull Request resolved: pytorch#2051

Disable the subwarp optimization for weighted TBE

- The cause of the NE issue has to be investigated.
- This does not affect the overall E2E performance much since the
  majority of TBEs in the model is unweighted

Reviewed By: jasonjk-park

Differential Revision: D49159119

fbshipit-source-id: 545b8428e228c77b8ef03c41f203384710519eb7
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D49159119

@facebook-github-bot
Copy link
Contributor

This pull request has been merged in 85a91d5.

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.

2 participants