Skip to content
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

MX4 Row-Based Padding #3010

Closed
wants to merge 1 commit into from
Closed

Conversation

jwfromm
Copy link
Contributor

@jwfromm jwfromm commented Aug 18, 2024

Summary: Refactor MX4 quantization to support per row padding as efficiently as possible. We do this in a cool way where we now support loading 2D blocks to make sure each thread has enough to work on. Using this approach, we should be able to apply padding to each row for effectively no cost.

Differential Revision: D61447274

Copy link

netlify bot commented Aug 18, 2024

Deploy Preview for pytorch-fbgemm-docs ready!

Name Link
🔨 Latest commit b83cab0
🔍 Latest deploy log https://app.netlify.com/sites/pytorch-fbgemm-docs/deploys/66c3695df5d85f0008ebf5e8
😎 Deploy Preview https://deploy-preview-3010--pytorch-fbgemm-docs.netlify.app
📱 Preview on mobile
Toggle QR Code...

QR Code

Use your smartphone camera to open QR code link.

To edit notification comments on pull requests, go to your Netlify site configuration.

@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D61447274

@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D61447274

jwfromm added a commit to jwfromm/FBGEMM that referenced this pull request Aug 18, 2024
Summary:
Pull Request resolved: pytorch#3010

X-link: facebookresearch/FBGEMM#105

Refactor MX4 quantization to support per row padding as efficiently as possible. We do this in a cool way where we now support loading 2D blocks to make sure each thread has enough to work on. Using this approach, we should be able to apply padding to each row for effectively no cost.

Differential Revision: D61447274
Summary:
Pull Request resolved: pytorch#3010

X-link: facebookresearch/FBGEMM#105

Refactor MX4 quantization to support per row padding as efficiently as possible. We do this in a cool way where we now support loading 2D blocks to make sure each thread has enough to work on. Using this approach, we should be able to apply padding to each row for effectively no cost.

Differential Revision: D61447274
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D61447274

@facebook-github-bot
Copy link
Contributor

This pull request has been merged in a973b75.

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