[Gluon][Layout][AMD] Fix layoutToGluon() about handling PaddedSharedEncodingAttr's CGA-layout#10392
Merged
Merged
Conversation
…odingAttr's CGA-layout It is to the fix problem arises when layoutToGluon() converting a PaddedSharedEncodingAttr (w/ multicta) into a PaddedSharedLayout object. Consider the case where PaddedSharedEncodingAttr has shape=[X, 128] and CGA-layout=[0,1], * Before the fix: PaddedSharedLayout.cga_layout is built directly from PaddedSharedEncodingAttr's "block" basis. It will be [0, 64=128/2] * After the fix: PaddedSharedLayout.cga_layout will be [0, 1] The rationale can also be extrapolated from the `PaddedSharedLayout::_to_ir()` where the linear-layout is obtained by `operator*(PaddedSharedLayout.offset_bases, PaddedSharedLayout.cga_layout)` via function `get_padded_shared_layout()` * command python3 -m pytest -v python/test/gluon/test_frontend.py::test_infer_layout_for_padded_shared * ttgl.static_assert() in the kernel makes sure CGA layout is expected
antiagainst
approved these changes
May 28, 2026
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
It is to the fix problem arises when
layoutToGluon()converting aPaddedSharedEncodingAttr(w/ multicta) into a PaddedSharedLayout object.Problem
Consider the case where PaddedSharedEncodingAttr has shape=[X, 128] and CGA-layout=[0,1],
PaddedSharedLayout.cga_layoutis built directly fromPaddedSharedEncodingAttr's "block" basis. It will be[0, 64=128/2]PaddedSharedLayout.cga_layoutwill be[0, 1]The rationale can also be extrapolated from the
PaddedSharedLayout::_to_ir()where the linear-layout is obtained byoperator*(PaddedSharedLayout.offset_bases, PaddedSharedLayout.cga_layout)via functionget_padded_shared_layout()Test
python3 -m pytest -v python/test/gluon/test_frontend.py::test_infer_layout_for_padded_sharedttgl.static_assert()in the kernel makes sure CGA layout is expected