R600: Handle fnearbyint

The difference from rint isn't really relevant here,
so treat them as equivalent. OpenCL doesn't have nearbyint,
so this is sort of pointless other than for completeness.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@211229 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Matt Arsenault 2014-06-18 22:03:45 +00:00
parent f286d63757
commit ce09bda96e
3 changed files with 70 additions and 0 deletions

View File

@ -325,6 +325,7 @@ AMDGPUTargetLowering::AMDGPUTargetLowering(TargetMachine &TM) :
setOperationAction(ISD::FTRUNC, VT, Expand);
setOperationAction(ISD::FMUL, VT, Expand);
setOperationAction(ISD::FRINT, VT, Expand);
setOperationAction(ISD::FNEARBYINT, VT, Expand);
setOperationAction(ISD::FSQRT, VT, Expand);
setOperationAction(ISD::FSIN, VT, Expand);
setOperationAction(ISD::FSUB, VT, Expand);
@ -334,6 +335,9 @@ AMDGPUTargetLowering::AMDGPUTargetLowering(TargetMachine &TM) :
setOperationAction(ISD::FCOPYSIGN, VT, Expand);
}
setOperationAction(ISD::FNEARBYINT, MVT::f32, Custom);
setOperationAction(ISD::FNEARBYINT, MVT::f64, Custom);
setTargetDAGCombine(ISD::MUL);
setTargetDAGCombine(ISD::SELECT_CC);
@ -501,6 +505,7 @@ SDValue AMDGPUTargetLowering::LowerOperation(SDValue Op,
case ISD::FCEIL: return LowerFCEIL(Op, DAG);
case ISD::FTRUNC: return LowerFTRUNC(Op, DAG);
case ISD::FRINT: return LowerFRINT(Op, DAG);
case ISD::FNEARBYINT: return LowerFNEARBYINT(Op, DAG);
case ISD::FFLOOR: return LowerFFLOOR(Op, DAG);
case ISD::UINT_TO_FP: return LowerUINT_TO_FP(Op, DAG);
@ -1683,6 +1688,13 @@ SDValue AMDGPUTargetLowering::LowerFRINT(SDValue Op, SelectionDAG &DAG) const {
return DAG.getSelect(SL, MVT::f64, Cond, Src, Tmp2);
}
SDValue AMDGPUTargetLowering::LowerFNEARBYINT(SDValue Op, SelectionDAG &DAG) const {
// FNEARBYINT and FRINT are the same, except in their handling of FP
// exceptions. Those aren't really meaningful for us, and OpenCL only has
// rint, so just treat them as equivalent.
return DAG.getNode(ISD::FRINT, SDLoc(Op), Op.getValueType(), Op.getOperand(0));
}
SDValue AMDGPUTargetLowering::LowerFFLOOR(SDValue Op, SelectionDAG &DAG) const {
SDLoc SL(Op);
SDValue Src = Op.getOperand(0);

View File

@ -54,6 +54,7 @@ private:
SDValue LowerFCEIL(SDValue Op, SelectionDAG &DAG) const;
SDValue LowerFTRUNC(SDValue Op, SelectionDAG &DAG) const;
SDValue LowerFRINT(SDValue Op, SelectionDAG &DAG) const;
SDValue LowerFNEARBYINT(SDValue Op, SelectionDAG &DAG) const;
SDValue LowerFFLOOR(SDValue Op, SelectionDAG &DAG) const;
SDValue LowerUINT_TO_FP(SDValue Op, SelectionDAG &DAG) const;

View File

@ -0,0 +1,57 @@
; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s
; RUN: llc -march=r600 -mcpu=bonaire -verify-machineinstrs < %s
; This should have the exactly the same output as the test for rint,
; so no need to check anything.
declare float @llvm.nearbyint.f32(float) #0
declare <2 x float> @llvm.nearbyint.v2f32(<2 x float>) #0
declare <4 x float> @llvm.nearbyint.v4f32(<4 x float>) #0
declare double @llvm.nearbyint.f64(double) #0
declare <2 x double> @llvm.nearbyint.v2f64(<2 x double>) #0
declare <4 x double> @llvm.nearbyint.v4f64(<4 x double>) #0
define void @fnearbyint_f32(float addrspace(1)* %out, float %in) #1 {
entry:
%0 = call float @llvm.nearbyint.f32(float %in)
store float %0, float addrspace(1)* %out
ret void
}
define void @fnearbyint_v2f32(<2 x float> addrspace(1)* %out, <2 x float> %in) #1 {
entry:
%0 = call <2 x float> @llvm.nearbyint.v2f32(<2 x float> %in)
store <2 x float> %0, <2 x float> addrspace(1)* %out
ret void
}
define void @fnearbyint_v4f32(<4 x float> addrspace(1)* %out, <4 x float> %in) #1 {
entry:
%0 = call <4 x float> @llvm.nearbyint.v4f32(<4 x float> %in)
store <4 x float> %0, <4 x float> addrspace(1)* %out
ret void
}
define void @nearbyint_f64(double addrspace(1)* %out, double %in) {
entry:
%0 = call double @llvm.nearbyint.f64(double %in)
store double %0, double addrspace(1)* %out
ret void
}
define void @nearbyint_v2f64(<2 x double> addrspace(1)* %out, <2 x double> %in) {
entry:
%0 = call <2 x double> @llvm.nearbyint.v2f64(<2 x double> %in)
store <2 x double> %0, <2 x double> addrspace(1)* %out
ret void
}
define void @nearbyint_v4f64(<4 x double> addrspace(1)* %out, <4 x double> %in) {
entry:
%0 = call <4 x double> @llvm.nearbyint.v4f64(<4 x double> %in)
store <4 x double> %0, <4 x double> addrspace(1)* %out
ret void
}
attributes #0 = { nounwind readonly }
attributes #1 = { nounwind }