Optimize _PaddedFP8rowwise_to_float_cuda_kernel for nrows == 1 #1858
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
✅ Deploy Preview for pytorch-fbgemm-docs canceled.
|
|
This pull request was exported from Phabricator. Differential Revision: D47068547 |
sryap
added a commit
to sryap/FBGEMM
that referenced
this pull request
Jul 3, 2023
…ch#1858) Summary: Pull Request resolved: pytorch#1858 Optimize `_PaddedFP8rowwise_to_float_cuda_kernel` by using one thread block to dequantize one row for the `nrows == 1` case. Prior to this diff, one thread dequantizes one row. `_PaddedFP8rowwise_to_float_cuda_kernel` removes padding, scale, and pad value from the input. Please see the attached figure for the input and output layouts. {F1039403748} Reviewed By: xiaosun86, jianyuh Differential Revision: D47068547 fbshipit-source-id: cd0c88051d8691d232c28600e70b6827318a6ef7
|
This pull request was exported from Phabricator. Differential Revision: D47068547 |
…ch#1858) Summary: Pull Request resolved: pytorch#1858 Optimize `_PaddedFP8rowwise_to_float_cuda_kernel` by using one thread block to dequantize one row for the `nrows == 1` case. Prior to this diff, one thread dequantizes one row. `_PaddedFP8rowwise_to_float_cuda_kernel` removes padding, scale, and pad value from the input. Please see the attached figure for the input and output layouts. {F1039403748} Reviewed By: xiaosun86, jianyuh Differential Revision: D47068547 fbshipit-source-id: 6b6cd9b2a092d17c11c356a84c96449351d453be
|
This pull request was exported from Phabricator. Differential Revision: D47068547 |
|
This pull request has been merged in f509314. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.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:
Optimize
_PaddedFP8rowwise_to_float_cuda_kernelby using one threadblock to dequantize one row for the
nrows == 1case. Prior to thisdiff, one thread dequantizes one row.
_PaddedFP8rowwise_to_float_cuda_kernelremoves padding, scale, andpad value from the input. Please see the attached figure for the
input and output layouts.
{F1039403748}
Differential Revision: D47068547