[GPU] Replace the skip condition for TileLargeTensors to be more general #20088
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
Signed-off-by: Nirvedh Meshram <[email protected]>
nirvedhmeshram
force-pushed
the
no_tile_reshape
branch
from
376020b to
39a72ec
Compare
|
We are going to take a different route and retain the named ops. |
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Outdated suggestions cannot be applied.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.
This PR turns TileLargeTensors pass off for lianlg.generic with empty body (except a linalg.yeild) along with exisitng cases in which we dont want it (which also meet the new condition). The reason this is and should be needed is that some patterns such as reshape fusion will generalize to a generic op, see here for example consider the
fuseWithReshapeByExpansionupstream, it will convert a transpose (or any expandable linalg op for that matter) to a generic op. We dont want to use this pass for such carefully introduced ops.https://github.com/llvm/llvm-project/blob/main/mlir/lib/Dialect/Linalg/Transforms/ElementwiseOpFusion.cpp#L930-L933