void ScalarizeVectorResult(SDNode *N, unsigned OpNo);
SDValue ScalarizeVecRes_MERGE_VALUES(SDNode *N, unsigned ResNo);
SDValue ScalarizeVecRes_BinOp(SDNode *N);
+ SDValue ScalarizeVecRes_TernaryOp(SDNode *N);
SDValue ScalarizeVecRes_UnaryOp(SDNode *N);
SDValue ScalarizeVecRes_InregOp(SDNode *N);
// Vector Result Splitting: <128 x ty> -> 2 x <64 x ty>.
void SplitVectorResult(SDNode *N, unsigned OpNo);
void SplitVecRes_BinOp(SDNode *N, SDValue &Lo, SDValue &Hi);
+ void SplitVecRes_TernaryOp(SDNode *N, SDValue &Lo, SDValue &Hi);
void SplitVecRes_UnaryOp(SDNode *N, SDValue &Lo, SDValue &Hi);
void SplitVecRes_InregOp(SDNode *N, SDValue &Lo, SDValue &Hi);
case ISD::SRL:
R = ScalarizeVecRes_BinOp(N);
break;
+ case ISD::FMA:
+ R = ScalarizeVecRes_TernaryOp(N);
+ break;
}
// If R is null, the sub-method took care of registering the result.
LHS.getValueType(), LHS, RHS);
}
+SDValue DAGTypeLegalizer::ScalarizeVecRes_TernaryOp(SDNode *N) {
+ SDValue Op0 = GetScalarizedVector(N->getOperand(0));
+ SDValue Op1 = GetScalarizedVector(N->getOperand(1));
+ SDValue Op2 = GetScalarizedVector(N->getOperand(2));
+ return DAG.getNode(N->getOpcode(), N->getDebugLoc(),
+ Op0.getValueType(), Op0, Op1, Op2);
+}
+
SDValue DAGTypeLegalizer::ScalarizeVecRes_MERGE_VALUES(SDNode *N,
unsigned ResNo) {
SDValue Op = DisintegrateMERGE_VALUES(N, ResNo);
case ISD::FREM:
SplitVecRes_BinOp(N, Lo, Hi);
break;
+ case ISD::FMA:
+ SplitVecRes_TernaryOp(N, Lo, Hi);
+ break;
}
// If Lo/Hi is null, the sub-method took care of registering results etc.
Hi = DAG.getNode(N->getOpcode(), dl, LHSHi.getValueType(), LHSHi, RHSHi);
}
+void DAGTypeLegalizer::SplitVecRes_TernaryOp(SDNode *N, SDValue &Lo,
+ SDValue &Hi) {
+ SDValue Op0Lo, Op0Hi;
+ GetSplitVector(N->getOperand(0), Op0Lo, Op0Hi);
+ SDValue Op1Lo, Op1Hi;
+ GetSplitVector(N->getOperand(1), Op1Lo, Op1Hi);
+ SDValue Op2Lo, Op2Hi;
+ GetSplitVector(N->getOperand(2), Op2Lo, Op2Hi);
+ DebugLoc dl = N->getDebugLoc();
+
+ Lo = DAG.getNode(N->getOpcode(), dl, Op0Lo.getValueType(),
+ Op0Lo, Op1Lo, Op2Lo);
+ Hi = DAG.getNode(N->getOpcode(), dl, Op0Hi.getValueType(),
+ Op0Hi, Op1Hi, Op2Hi);
+}
+
void DAGTypeLegalizer::SplitVecRes_BITCAST(SDNode *N, SDValue &Lo,
SDValue &Hi) {
// We know the result is a vector. The input may be either a vector or a
setOperationAction(ISD::FSUB, (MVT::SimpleValueType)VT, Expand);
setOperationAction(ISD::MUL , (MVT::SimpleValueType)VT, Expand);
setOperationAction(ISD::FMUL, (MVT::SimpleValueType)VT, Expand);
+ setOperationAction(ISD::FMA, (MVT::SimpleValueType)VT, Custom);
setOperationAction(ISD::SDIV, (MVT::SimpleValueType)VT, Expand);
setOperationAction(ISD::UDIV, (MVT::SimpleValueType)VT, Expand);
setOperationAction(ISD::FDIV, (MVT::SimpleValueType)VT, Expand);
case ISD::SUBE: return LowerADDC_ADDE_SUBC_SUBE(Op, DAG);
case ISD::ADD: return LowerADD(Op, DAG);
case ISD::SUB: return LowerSUB(Op, DAG);
+ case ISD::FMA: return SDValue();
}
}
ret float %ret
}
+; Check that very wide vector fma's can be split into legal fma's.
+define void @test_fma_v8f32(<8 x float> %a, <8 x float> %b, <8 x float> %c, <8 x float>* %p) nounwind readnone ssp {
+; CHECK: test_fma_v8f32
+; CHECK: vfma.f32
+; CHECK: vfma.f32
+entry:
+ %call = tail call <8 x float> @llvm.fma.v8f32(<8 x float> %a, <8 x float> %b, <8 x float> %c) nounwind readnone
+ store <8 x float> %call, <8 x float>* %p, align 16
+ ret void
+}
+
declare float @llvm.fma.f32(float, float, float) nounwind readnone
declare double @llvm.fma.f64(double, double, double) nounwind readnone
declare <2 x float> @llvm.fma.v2f32(<2 x float>, <2 x float>, <2 x float>) nounwind readnone
+declare <8 x float> @llvm.fma.v8f32(<8 x float>, <8 x float>, <8 x float>) nounwind readnone