[Codegen][GPU] Make GPUApplyTilingLevel run cleanup patterns #17533
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
This adds a few cleanup patterns to the end of `GPUApplyTilingLevel`, namely the pattern to merge consecutive insert/extract slices. The pipeline that is intended to use this pass is not sensitive to bufferization the same way and can freely run these patterns to simplify later loop hoisting patterns.
There was a problem hiding this comment.
Just a note, some of the insert/extract slice folding has been implicated in a race condition. See #17525 (comment)
MaheshRavishankar
requested changes
May 31, 2024
qedawkins
commented
May 31, 2024
compiler/src/iree/compiler/Codegen/Common/GPU/test/gpu_apply_tiling_level.mlir
Show resolved
Hide resolved
That is for folding insert/extract_slice ops into vector transfers. This is for merging |
kuhar
approved these changes
May 31, 2024
MaheshRavishankar
approved these changes
May 31, 2024
gglangg
pushed a commit
to gglangg/iree
that referenced
this pull request
Jun 4, 2024
…g#17533) This adds a few cleanup patterns to the end of `GPUApplyTilingLevel`, namely the pattern to merge consecutive insert/extract slices. The pipeline that is intended to use this pass is not sensitive to bufferization the same way and can freely run these patterns to simplify later loop hoisting patterns.
gglangg
pushed a commit
to gglangg/iree
that referenced
this pull request
Jun 4, 2024
…g#17533) This adds a few cleanup patterns to the end of `GPUApplyTilingLevel`, namely the pattern to merge consecutive insert/extract slices. The pipeline that is intended to use this pass is not sensitive to bufferization the same way and can freely run these patterns to simplify later loop hoisting patterns.
bangtianliu
pushed a commit
to bangtianliu/iree
that referenced
this pull request
Jun 5, 2024
…g#17533) This adds a few cleanup patterns to the end of `GPUApplyTilingLevel`, namely the pattern to merge consecutive insert/extract slices. The pipeline that is intended to use this pass is not sensitive to bufferization the same way and can freely run these patterns to simplify later loop hoisting patterns.
LLITCHEV
pushed a commit
to LLITCHEV/iree
that referenced
this pull request
Jul 30, 2024
…g#17533) This adds a few cleanup patterns to the end of `GPUApplyTilingLevel`, namely the pattern to merge consecutive insert/extract slices. The pipeline that is intended to use this pass is not sensitive to bufferization the same way and can freely run these patterns to simplify later loop hoisting patterns. Signed-off-by: Lubo Litchev <lubol@google.com>
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 adds a few cleanup patterns to the end of
GPUApplyTilingLevel, namely the pattern to merge consecutive insert/extract slices. The pipeline that is intended to use this pass is not sensitive to bufferization the same way and can freely run these patterns to simplify later loop hoisting patterns.