FBGEMM
FBGEMM copied to clipboard
Back out "Better Shape Function Registration"
Summary: For some reason, there are prod jobs that try to run fbgemm cuda ops on CPU, and rely on the ability to do so. This doesnt really make sense but we'll have to revert the shape registration refactor for now to enable it. As long as this interaction exists, we wont be able to support torch.export and these prod jobs at the same time.
Reviewed By: jiawenliu64
Differential Revision: D64702422
This pull request was exported from Phabricator. Differential Revision: D64702422
Deploy Preview for pytorch-fbgemm-docs ready!
| Name | Link |
|---|---|
| Latest commit | 6c00d8eac4828847a6da59cf005567a7a984d0e1 |
| Latest deploy log | https://app.netlify.com/sites/pytorch-fbgemm-docs/deploys/6716962e55b31500087e414b |
| Deploy Preview | https://deploy-preview-3261--pytorch-fbgemm-docs.netlify.app |
| Preview on mobile | Toggle QR Code...Use your smartphone camera to open QR code link. |
To edit notification comments on pull requests, go to your Netlify site configuration.