vendor affine loop parallelization from upstream in affine-cfg #429
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 is a partially reduced copy of the affine-parallelize pass from upstream that doesn't limit the number of nested parallel loops and always handles reductions.
wsmoses
reviewed
Mar 4, 2025
| // The walker proceeds in pre-order to process the outer loops first | ||
| // and control the number of outer parallel loops. | ||
| std::vector<ParallelizationCandidate> parallelizableLoops; | ||
| op->walk<WalkOrder::PreOrder>([&](AffineForOp loop) { |
There was a problem hiding this comment.
we may want to run this before the affinecfg stuff [or ideally as a pattern], since the parallel fusion, and index rewrite is a pattern [otherwise run before parallelize]
There was a problem hiding this comment.
it's not initially written is a pattern, so I suspect it may mess up the internal state of the rewriter; let me see what I can do
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Outdated suggestions cannot be applied.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 is a partially reduced copy of the affine-parallelize pass from upstream that doesn't limit the number of nested parallel loops and always handles reductions.