Fix bug in fp8_split_embedding_codegen_forward_kernel dispatch #2009
Conversation
This file contains 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: D49139649 |
…ch#2009) Summary: For the case of loading FP8 tables with 128B * 8 < dimension <= 128B * 16, the numbers for (MinNum128BRows, MaxNum128BRows) should be (8, 16) and not (4, 8 ). Because of this bug, FP8 tables with dimension in that range don't get properly loaded upon codegen_forward. I wrote a little test to show this deterministically. I will remove the test before land if it is too specific. Differential Revision: D49139649
SungMinCho
force-pushed
the
export-D49139649
branch
from
57810b7
to
4d47b88
Compare
|
This pull request was exported from Phabricator. Differential Revision: D49139649 |
|
This pull request has been merged in 8e588e6. |
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Outdated suggestions cannot be applied.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:
For the case of loading FP8 tables with 128B * 8 < dimension <= 128B * 16, the numbers for (MinNum128BRows, MaxNum128BRows) should be (8, 16) and not (4, 8 ). Because of this bug, FP8 tables with dimension in that range don't get properly loaded upon codegen_forward. I wrote a little test to show this deterministically. I will remove the test before land if it is too specific.
Differential Revision: D49139649