R600: Expand vector FFLOOR ops

Reviewed-by: Michel Dänzer <[email protected]>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@188597 91177308-0d34-0410-b5e6-96231b3b80d8

lib/Target/R600/AMDGPUISelLowering.cpp

@@ -152,6 +152,7 @@ AMDGPUTargetLowering::AMDGPUTargetLowering(TargetMachine &TM) :
     MVT::SimpleValueType VT = (MVT::SimpleValueType)FloatTypes[x];
     setOperationAction(ISD::FADD, VT, Expand);
     setOperationAction(ISD::FDIV, VT, Expand);
+    setOperationAction(ISD::FFLOOR, VT, Expand);
     setOperationAction(ISD::FMUL, VT, Expand);
     setOperationAction(ISD::FSUB, VT, Expand);
   }

test/CodeGen/R600/llvm.floor.ll

@@ -0,0 +1,54 @@
+; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s --check-prefix=R600-CHECK
+; RUN: llc < %s -march=r600 -mcpu=SI | FileCheck %s --check-prefix=SI-CHECK
+
+; R600-CHECK: @f32
+; R600-CHECK: FLOOR
+; SI-CHECK: @f32
+; SI-CHECK: V_FLOOR_F32_e32
+define void @f32(float addrspace(1)* %out, float %in) {
+entry:
+  %0 = call float @llvm.floor.f32(float %in)
+  store float %0, float addrspace(1)* %out
+  ret void
+}
+
+; R600-CHECK: @v2f32
+; R600-CHECK: FLOOR
+; R600-CHECK: FLOOR
+; SI-CHECK: @v2f32
+; SI-CHECK: V_FLOOR_F32_e32
+; SI-CHECK: V_FLOOR_F32_e32
+define void @v2f32(<2 x float> addrspace(1)* %out, <2 x float> %in) {
+entry:
+  %0 = call <2 x float> @llvm.floor.v2f32(<2 x float> %in)
+  store <2 x float> %0, <2 x float> addrspace(1)* %out
+  ret void
+}
+
+; R600-CHECK: @v4f32
+; R600-CHECK: FLOOR
+; R600-CHECK: FLOOR
+; R600-CHECK: FLOOR
+; R600-CHECK: FLOOR
+; SI-CHECK: @v4f32
+; SI-CHECK: V_FLOOR_F32_e32
+; SI-CHECK: V_FLOOR_F32_e32
+; SI-CHECK: V_FLOOR_F32_e32
+; SI-CHECK: V_FLOOR_F32_e32
+define void @v4f32(<4 x float> addrspace(1)* %out, <4 x float> %in) {
+entry:
+  %0 = call <4 x float> @llvm.floor.v4f32(<4 x float> %in)
+  store <4 x float> %0, <4 x float> addrspace(1)* %out
+  ret void
+}
+
+; Function Attrs: nounwind readonly
+declare float @llvm.floor.f32(float) #0
+
+; Function Attrs: nounwind readonly
+declare <2 x float> @llvm.floor.v2f32(<2 x float>) #0
+
+; Function Attrs: nounwind readonly
+declare <4 x float> @llvm.floor.v4f32(<4 x float>) #0
+
+attributes #0 = { nounwind readonly }