[PATCH] D146404: Improve min/max vector reductions on arm

[Caleb Zulawski via Phabricator via llvm-commits]

calebzulawski created this revision.
calebzulawski added reviewers: greened, nikic.
calebzulawski added a project: LLVM.
Herald added subscribers: StephenFan, hiraditya, kristof.beyls.
Herald added a project: All.
calebzulawski requested review of this revision.
Herald added a subscriber: llvm-commits.

This nearly resolves issues such as #50466, #40981, #38190.

In the added test, the extra long vector reduction seems to skip my added reduction, where should I look for that?

I'm not sure who to add to review this, feel free to add anyone else or remove yourself.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D146404

Files:
  llvm/lib/Target/ARM/ARMISelLowering.cpp
  llvm/lib/Target/ARM/ARMISelLowering.h
  llvm/test/CodeGen/ARM/vecreduce-minmax.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D146404.506463.patch
Type: text/x-patch
Size: 10645 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230320/81cd228d/attachment.bin>