diff options
author | Michael Kruse <llvm-project@meinersbur.de> | 2023-03-08 11:39:17 -0600 |
---|---|---|
committer | Michael Kruse <llvm-project@meinersbur.de> | 2023-03-08 12:51:42 -0600 |
commit | 42cd38c01e5b76357e77b8c0f4f32d77a97d153f (patch) | |
tree | 00ada675fafc95fb8c147f085b3a8f28d9568ef8 /polly/docs | |
parent | c0967995d254fd2dd24ba74afa46df5f559f11ba (diff) | |
download | llvm-42cd38c01e5b76357e77b8c0f4f32d77a97d153f.tar.gz |
[Polly] Remove -polly-vectorizer=polly.
Polly's internal vectorizer is not well maintained and is known to not work in some cases such as region ScopStmts. Unlike LLVM's LoopVectorize pass it also does not have a target-dependent cost heuristics, and we recommend using LoopVectorize instead of -polly-vectorizer=polly.
In the future we hope that Polly can collaborate better with LoopVectorize, like Polly marking a loop is safe to vectorize with a specific simd width, instead of replicating its functionality.
Reviewed By: grosser
Differential Revision: https://reviews.llvm.org/D142640
Diffstat (limited to 'polly/docs')
-rw-r--r-- | polly/docs/ReleaseNotes.rst | 11 |
1 files changed, 11 insertions, 0 deletions
diff --git a/polly/docs/ReleaseNotes.rst b/polly/docs/ReleaseNotes.rst index bf6f7b9277d3..f29f7a90abfa 100644 --- a/polly/docs/ReleaseNotes.rst +++ b/polly/docs/ReleaseNotes.rst @@ -10,3 +10,14 @@ In Polly |version| the following important changes have been incorporated. These release notes are for the next release of Polly and describe the new features that have recently been committed to our development branch. + + +- Support for -polly-vectorizer=polly has been removed. Polly's internal + vectorizer is not well maintained and is known to not work in some cases + such as region ScopStmts. Unlike LLVM's LoopVectorize pass it also does + not have a target-dependent cost heuristics, and we recommend using + LoopVectorize instead of -polly-vectorizer=polly. + + In the future we hope that Polly can collaborate better with LoopVectorize, + like Polly marking a loop is safe to vectorize with a specific simd width, + instead of replicating its functionality. |