PR #17440: Inject desired pattern for handling Transpose for fp8 gemm rewrite #17798
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
![copybara-service[bot]](https://avatars.githubusercontent.com/in/44061?s=60&v=4){kind=small}
copybara-service
bot
force-pushed
the
test_680886834
branch
13 times, most recently
from
8d04204
to
fbc78e4
Compare
… rewrite Imported from GitHub PR #17440 Related to #17276 and #16975. This PR updates the GemmRewriter to handle the transpose of non-descending layouts directly, eliminating the need for the layout_normalization pass to correct this error-prone pattern post-rewrite. The desired transformation is now injected into GemmRewriter, ensuring the problematic layout is handled internally. This PR transforms the following error-prone pattern, where the transpose of a non-descending layout is the issue: ``` a = f8e4m3fn[x,y]{0,1} xxx transpose.0 = f8e4m3fn[y,x]{0,1} transpose(a), dimensions=(1,0) custom-call(a,...) ``` to ``` a = f8e4m3fn[x,y]{0,1} xxx bt = f8e4m3fn[y,x]{1,0} bitcast(a) transpose.1 = f8e4m3fn[x,y]{1,0} transpose(bt), dimensions=(1,0) bt.1= f8e4m3fn[y,x]{0,1} bitcast(transpose.1) custom-call(bt.1,...) ``` Copybara import of the project: -- 237c032 by shuw <[email protected]>: Improve TransposeMatrix -- 508cd69 by Shu Wang <[email protected]>: Fix bug of permutation. -- c55e8a9 by shuw <[email protected]>: clang format -- ad0a4ba by Shu Wang <[email protected]>: Add unittest. -- 1d45b4d by Shu Wang <[email protected]>: Remove uncessary space. -- 7837845 by Shu Wang <[email protected]>: Update unittest. -- b479c21 by shuw <[email protected]>: Improve TransposeMatrix -- b633184 by Shu Wang <[email protected]>: Update unittest shape and BUILD file. Merging this change closes #17440 COPYBARA_INTEGRATE_REVIEW=#17440 from wenscarl:fp8_regulate_transpose b633184 PiperOrigin-RevId: 681551009
copybara-service
bot
force-pushed
the
test_680886834
branch
from
fbc78e4
to
7d4740a
Compare
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.
PR #17440: Inject desired pattern for handling Transpose for fp8 gemm rewrite
Imported from GitHub PR #17440
Related to #17276 and #16975.
This PR updates the GemmRewriter to handle the transpose of non-descending layouts directly, eliminating the need for the layout_normalization pass to correct this error-prone pattern post-rewrite. The desired transformation is now injected into GemmRewriter, ensuring the problematic layout is handled internally. This PR transforms the following error-prone pattern, where the transpose of a non-descending layout is the issue:
to
Copybara import of the project:
--
237c032 by shuw [email protected]:
Improve TransposeMatrix
--
508cd69 by Shu Wang [email protected]:
Fix bug of permutation.
c55e8a9 by shuw [email protected]:
clang format
--
ad0a4ba by Shu Wang [email protected]:
Add unittest.
1d45b4d by Shu Wang [email protected]:
Remove uncessary space.
7837845 by Shu Wang [email protected]:
Update unittest.
--
b479c21 by shuw [email protected]:
Improve TransposeMatrix
--
b633184 by Shu Wang [email protected]:
Update unittest shape and BUILD file.
Merging this change closes #17440
FUTURE_COPYBARA_INTEGRATE_REVIEW=#17440 from wenscarl:fp8_regulate_transpose b633184