diff options
author | Kerry McLaughlin <kerry.mclaughlin@arm.com> | 2021-06-28 11:26:10 +0100 |
---|---|---|
committer | Kerry McLaughlin <kerry.mclaughlin@arm.com> | 2021-06-28 11:27:10 +0100 |
commit | f99672568fda6a9bc1ee4f09d5d84066f4979889 (patch) | |
tree | 9903d0d9f1908300d4ce8ef65505f2ab1121260d /README.md | |
parent | 80aa7e147e2a3a2a0435663ad9171b9ccdcc0178 (diff) | |
download | llvm-f99672568fda6a9bc1ee4f09d5d84066f4979889.tar.gz |
[LoopVectorize] Fix strict reductions where VF = 1
Currently we will allow loops with a fixed width VF of 1 to vectorize
if the -enable-strict-reductions flag is set. However, the loop vectorizer
will not use ordered reductions if `VF.isScalar()` and the resulting
vectorized loop will be out of order.
This patch removes `VF.isVector()` when checking if ordered reductions
should be used. Also, instead of converting the FAdds to reductions if the
VF = 1, operands of the FAdds are changed such that the order is preserved.
Reviewed By: david-arm
Differential Revision: https://reviews.llvm.org/D104533
Diffstat (limited to 'README.md')
0 files changed, 0 insertions, 0 deletions