[Encoding] Test i1 mask attention with packed_storage #22756
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
Implement the SerializableAttr interface method to check compatibility with other encodings for the packed_storage encoding. Signed-off-by: Lukas Sommer <[email protected]>
Extend the EncodeDeviceTensors pass to handle tensors with the packed_storage attribute attached. Signed-off-by: Lukas Sommer <[email protected]>
Extend the EncodeHostTensors pass to handle tensors with the packed_storage attribute attached. Signed-off-by: Lukas Sommer <[email protected]>
Drop encoding attributes that do not need further materialization during Codegen after type propagation has used the information. This happens as a second phase in the TypePropagationPass. Signed-off-by: Lukas Sommer <[email protected]>
Signed-off-by: Lukas Sommer <[email protected]>
Test attention with an `i1` mask using the `packed_storage` encoding attribute instead of the experimental flag to avoid extending `i1` to `i8` during compilation. Currently the test needs to drop the unit extent dimension compared to the other attention tests due to problems with FoldUnitExtentDims. Signed-off-by: Lukas Sommer <[email protected]>
Signed-off-by: Lukas Sommer <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
Test attention with an
i1mask using thepacked_storageencoding attribute instead of the experimental flag to avoid extendingi1toi8during compilation.Currently the test needs to drop the unit extent dimension compared to the other attention tests due to problems with FoldUnitExtentDims.
Signed-off-by: Lukas Sommer [email protected]