[PATCH] D67841: [SLP] avoid reduction transform on patterns that the backend can load-combine

[Sanjay Patel via Phabricator via llvm-commits]

spatel updated this revision to Diff 223613.
spatel added a comment.

Patch updated:
Jump back to the earlier revision which created a new method for cost of a load-combine pattern. This is independent of the existing arithmetic instruction cost API, so we can be sure that there is no conflict with other cost model users. It's also more accurate because we can add the cost of a wider load just once for the entire pattern. Test case for the loop vectorizer crash was added at rL373913 <https://reviews.llvm.org/rL373913>.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D67841/new/

https://reviews.llvm.org/D67841

Files:
  llvm/include/llvm/Analysis/TargetTransformInfo.h
  llvm/include/llvm/Analysis/TargetTransformInfoImpl.h
  llvm/include/llvm/CodeGen/BasicTTIImpl.h
  llvm/lib/Analysis/TargetTransformInfo.cpp
  llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
  llvm/test/Transforms/SLPVectorizer/X86/bad-reduction.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D67841.223613.patch
Type: text/x-patch
Size: 20744 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20191007/ab0e7621/attachment.bin>