Deprecate float32 bias for Cutlass FP8 rowwise #4274
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Summary: Currently for each configuration (MNK tile shapes & cluster shapes) of Cutlass FP8 rowwise we will compile 12 kernels, as we further template on (FAST_ACCUM, USE_BIAS, INPUT_DTYPE, BIAS_DTYPE). We don't seem to have any use-cases using float32 BIAS_DTYPE, so it can likely be removed, making it only 8 kernels instead of 12 per configuration. INPUT_DTYPE could likely be removed as well, but might need some more cleanup/verification to be done safely, as I see some scripts using it, so its a bit more work. Differential Revision: D76063973
✅ Deploy Preview for pytorch-fbgemm-docs ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
|
This pull request was exported from Phabricator. Differential Revision: D76063973 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
Currently for each configuration (MNK tile shapes & cluster shapes) of Cutlass FP8 rowwise we will compile 12 kernels, as we further template on (FAST_ACCUM, USE_BIAS, INPUT_DTYPE, BIAS_DTYPE). We don't seem to have any use-cases using float32 BIAS_DTYPE, so it can likely be removed, making it only 8 kernels instead of 12 per configuration.
INPUT_DTYPE could likely be removed as well, but might need some more cleanup/verification to be done safely, as I see some scripts using it, so its a bit more work.
Differential Revision: D76063973