This is an archive of the discontinued LLVM Phabricator instance.

[mlir][vector] Separate out vector transfer + tensor slice patterns
ClosedPublic

Authored by antiagainst on May 16 2023, 11:38 AM.

Details

Summary

These patterns touches the structure generated from tiling so it
affects later steps like bufferization and vector hoisting.
Instead of putting them in canonicalization, this commit creates
separate entry points for them to be called explicitly.

This is NFC regarding the functionality and tests of those patterns.
It also addresses two TODO items in the codebase.

Diff Detail

Event Timeline

antiagainst created this revision.May 16 2023, 11:38 AM
Herald added a project: Restricted Project. · View Herald Transcript
antiagainst requested review of this revision.May 16 2023, 11:38 AM
antiagainst retitled this revision from [vector] Separate out vector transfer + tensor slice patterns to [mlir][vector] Separate out vector transfer + tensor slice patterns.May 16 2023, 11:56 AM
antiagainst edited the summary of this revision. (Show Details)May 17 2023, 8:13 AM
ThomasRaoux accepted this revision.May 17 2023, 8:34 AM

Looks good, thanks!

This revision is now accepted and ready to land.May 17 2023, 8:34 AM