Use NEON vmin/vmax instructions for floating-point selects.
authorBob Wilson <bob.wilson@apple.com>
Thu, 18 Feb 2010 06:05:53 +0000 (06:05 +0000)
committerBob Wilson <bob.wilson@apple.com>
Thu, 18 Feb 2010 06:05:53 +0000 (06:05 +0000)
commit9f6c4c141ffa9c8b13e90dce2f2285c4479ff403
treef5b3ac1fc043535a334428d2be5f974c8b01b080
parent2155d459a7de626bb77849e5956ef2363d7d4a87
Use NEON vmin/vmax instructions for floating-point selects.
Radar 7461718.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@96572 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/ARM/ARMISelLowering.cpp
lib/Target/ARM/ARMISelLowering.h
lib/Target/ARM/ARMInstrNEON.td
test/CodeGen/ARM/neon_minmax.ll [new file with mode: 0644]