[PATCH] D109645: [LV] Recognize intrinsic min/max reductions

[Dave Green via Phabricator via llvm-commits]

dmgreen created this revision.
dmgreen added reviewers: spatel, nikic, fhahn, Ayal, david-arm.
Herald added a subscriber: hiraditya.
dmgreen requested review of this revision.
Herald added a project: LLVM.

This extends the reduction logic in the vectorizer to handle intrinsic versions of min and max, both the floating point variants already created by instcombine under fastmath and the integer variants from D98152 <https://reviews.llvm.org/D98152>.

As a bonus this allows it to match a chain of min or max operations into a single reduction, similar to how add/mul/etc work.


https://reviews.llvm.org/D109645

Files:
  llvm/include/llvm/Analysis/IVDescriptors.h
  llvm/lib/Analysis/IVDescriptors.cpp
  llvm/test/Transforms/LoopVectorize/minmax_reduction.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D109645.372078.patch
Type: text/x-patch
Size: 14444 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210911/c8ac5fcf/attachment.bin>