RFC: Make self-inplace broadcast vectorlizable based on our effect inference #43185
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Outdated suggestions cannot be applied.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.
As found by #43153, self-inplace broadcast with dimension larger than 1 won't be vectorlized by LLVM, as it fails to prove
a[I]
andextrude(a)[I]
share the same access pattern.#43852 makes it possible to fix it on julia side, as
@simd ivdep
should be safe ifbc[I]
is proven to be effect_free, anddest
is not self-aliased.Currently the self-aliasing check for
dest
is pure type-based to avoid binary bloat.