[PATCH] D32247: Switch AArch64 to use reduction intrinsics

[Amara Emerson via Phabricator via llvm-commits]

aemerson created this revision.
Herald added a reviewer: javed.absar.

This patch enables the use of the experimental reduction intrinsics from https://reviews.llvm.org/D30086 for AArch64.

The existing idiom recognition for the shufflevector sequence is removed and tests updated to use the new form.


Repository:
  rL LLVM

https://reviews.llvm.org/D32247

Files:
  lib/Target/AArch64/AArch64ISelLowering.cpp
  lib/Target/AArch64/AArch64ISelLowering.h
  lib/Target/AArch64/AArch64TargetTransformInfo.cpp
  lib/Target/AArch64/AArch64TargetTransformInfo.h
  test/CodeGen/AArch64/aarch64-addv.ll
  test/CodeGen/AArch64/aarch64-minmaxv.ll
  test/CodeGen/AArch64/arm64-vabs.ll
  test/Transforms/LoopVectorize/AArch64/reduction-small-size.ll
  test/Transforms/SLPVectorizer/AArch64/gather-root.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D32247.95824.patch
Type: text/x-patch
Size: 83294 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170419/f6fcbdd5/attachment.bin>