[llvm] r292816 - AMDGPU: Remove unnecessary check

[Matt Arsenault via llvm-commits]

Author: arsenm
Date: Mon Jan 23 13:00:15 2017
New Revision: 292816

URL: http://llvm.org/viewvc/llvm-project?rev=292816&view=rev
Log:
AMDGPU: Remove unnecessary check

There are no scalar FP types that can be extended.

Modified:
    llvm/trunk/lib/Target/AMDGPU/SIISelLowering.cpp

Modified: llvm/trunk/lib/Target/AMDGPU/SIISelLowering.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/SIISelLowering.cpp?rev=292816&r1=292815&r2=292816&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AMDGPU/SIISelLowering.cpp (original)
+++ llvm/trunk/lib/Target/AMDGPU/SIISelLowering.cpp Mon Jan 23 13:00:15 2017
@@ -1888,9 +1888,6 @@ MVT SITargetLowering::getScalarShiftAmou
 bool SITargetLowering::isFMAFasterThanFMulAndFAdd(EVT VT) const {
   VT = VT.getScalarType();
 
-  if (!VT.isSimple())
-    return false;
-
   switch (VT.getSimpleVT().SimpleTy) {
   case MVT::f32:
     // This is as fast on some subtargets. However, we always have full rate f32