projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 8b08141) | patch
[ARM] Match fminnan/fmaxnan for vector vmin/vmax instead of an intrinsic
authorJames Molloy <james.molloy@arm.com>
Tue, 11 Aug 2015 12:06:28 +0000 (12:06 +0000)
committerJames Molloy <james.molloy@arm.com>
Tue, 11 Aug 2015 12:06:28 +0000 (12:06 +0000)
commit7b49489409a4f22286098f2caa2065fceaa1af0b
treed46004bd8f6bf27f56ff4c4409e5bdf6629ca65ftree | snapshot
parent8b081412aa734ce4b02cb9a84754f715a519dca2commit | diff
[ARM] Match fminnan/fmaxnan for vector vmin/vmax instead of an intrinsic

Lower Intrinsic::arm_neon_vmins/vmaxs to fminnan/fmaxnan and match that instead. This is important because SDAG will soon be able to select FMINNAN itself, so we need a unified lowering path for intrinsics and SDAG.

NFCI.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@244593 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/ARM/ARMISelLowering.cpp diff | blob | history
lib/Target/ARM/ARMInstrNEON.td diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors