[all-commits] [llvm/llvm-project] 68e002: [ARM] Fold select_cc(vecreduce_[u|s][min|max], x) ...

[Sam Tebbs via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 68e002e1819f1598fc6815226a353ad2f04cd509
      https://github.com/llvm/llvm-project/commit/68e002e1819f1598fc6815226a353ad2f04cd509
  Author: Sam Tebbs <samuel.tebbs at arm.com>
  Date:   2020-10-06 (Tue, 06 Oct 2020)

  Changed paths:
    M llvm/lib/Target/ARM/ARMISelLowering.cpp
    M llvm/lib/Target/ARM/ARMISelLowering.h
    M llvm/lib/Target/ARM/ARMInstrMVE.td
    M llvm/test/CodeGen/Thumb2/mve-vecreduce-loops.ll
    A llvm/test/CodeGen/Thumb2/mve-vmaxv-vminv-scalar.ll
    M llvm/test/CodeGen/Thumb2/mve-vmaxv.ll

  Log Message:
  -----------
  [ARM] Fold select_cc(vecreduce_[u|s][min|max], x) into VMINV or VMAXV

    This folds a select_cc or select(set_cc) of a max or min vector reduction with a scalar value into a VMAXV or VMINV.

    Differential Revision: https://reviews.llvm.org/D87836