[PATCH] D101836: [LoopVectorize] Enable strict reductions when allowReordering() returns false

[Kerry McLaughlin via Phabricator via llvm-commits]

kmclaughlin updated this revision to Diff 347364.
kmclaughlin added a comment.

As discussed with @sdesmalen, I have made the following changes to this patch so that the tests are clearer:

- Combined this patch with D102774 <https://reviews.llvm.org/D102774>, adding the -hints-allow-reordering flag here
- Added CHECK-ORDERED and CHECK-UNORDERED lines to the tests in D103015 <https://reviews.llvm.org/D103015>
- Updated the RUN lines in this patch to use the -hints-allow-reordering flag and also added a RUN line for CHECK-NOT-VECTORIZED. The tests themselves remain unchanged, with the exception of the new tests added for induction variables.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D101836/new/

https://reviews.llvm.org/D101836

Files:
  llvm/include/llvm/Transforms/Vectorize/LoopVectorizationLegality.h
  llvm/lib/Transforms/Vectorize/LoopVectorizationLegality.cpp
  llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
  llvm/test/Transforms/LoopVectorize/AArch64/scalable-strict-fadd.ll
  llvm/test/Transforms/LoopVectorize/AArch64/strict-fadd.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D101836.347364.patch
Type: text/x-patch
Size: 21627 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210524/868719a8/attachment.bin>