Index: ../include/llvm/CodeGen/SelectionDAGNodes.h =================================================================== --- ../include/llvm/CodeGen/SelectionDAGNodes.h +++ ../include/llvm/CodeGen/SelectionDAGNodes.h @@ -2115,10 +2115,11 @@ assert(getValue().getValueType() == getValueType(0) && "Incompatible type of the PathThru value in MaskedGatherSDNode"); assert(getMask().getValueType().getVectorNumElements() == - getValueType(0).getVectorNumElements() && - "Vector width mismatch between mask and data"); - assert(getMask().getValueType().getScalarType() == MVT::i1 && + getValueType(0).getVectorNumElements() && "Vector width mismatch between mask and data"); + assert(getIndex().getValueType().getVectorNumElements() == + getValueType(0).getVectorNumElements() && + "Vector width mismatch between index and data"); } static bool classof(const SDNode *N) { @@ -2134,13 +2135,14 @@ friend class SelectionDAG; MaskedScatterSDNode(unsigned Order, DebugLoc dl,ArrayRef Operands, SDVTList VTs, EVT MemVT, MachineMemOperand *MMO) - : MaskedGatherScatterSDNode(ISD::MSCATTER, Order, dl, Operands, VTs, - MemVT, MMO) { + : MaskedGatherScatterSDNode(ISD::MSCATTER, Order, dl, Operands, VTs, MemVT, + MMO) { assert(getMask().getValueType().getVectorNumElements() == - getValue().getValueType().getVectorNumElements() && - "Vector width mismatch between mask and data"); - assert(getMask().getValueType().getScalarType() == MVT::i1 && + getValue().getValueType().getVectorNumElements() && "Vector width mismatch between mask and data"); + assert(getIndex().getValueType().getVectorNumElements() == + getValue().getValueType().getVectorNumElements() && + "Vector width mismatch between index and data"); } static bool classof(const SDNode *N) { Index: ../lib/CodeGen/SelectionDAG/LegalizeFloatTypes.cpp =================================================================== --- ../lib/CodeGen/SelectionDAG/LegalizeFloatTypes.cpp +++ ../lib/CodeGen/SelectionDAG/LegalizeFloatTypes.cpp @@ -593,7 +593,7 @@ L->getPointerInfo(), NVT, L->isVolatile(), L->isNonTemporal(), false, L->getAlignment(), L->getAAInfo()); - // Legalized the chain result - switch anything that used the old chain to + // Legalize the chain result - switch anything that used the old chain to // use the new one. ReplaceValueWith(SDValue(N, 1), NewL.getValue(1)); return NewL; @@ -606,7 +606,7 @@ L->getMemoryVT(), L->isVolatile(), L->isNonTemporal(), false, L->getAlignment(), L->getAAInfo()); - // Legalized the chain result - switch anything that used the old chain to + // Legalize the chain result - switch anything that used the old chain to // use the new one. ReplaceValueWith(SDValue(N, 1), NewL.getValue(1)); return BitConvertToInteger(DAG.getNode(ISD::FP_EXTEND, dl, VT, NewL)); @@ -643,7 +643,7 @@ NewVAARG = DAG.getVAArg(NVT, dl, Chain, Ptr, N->getOperand(2), N->getConstantOperandVal(3)); - // Legalized the chain result - switch anything that used the old chain to + // Legalize the chain result - switch anything that used the old chain to // use the new one. ReplaceValueWith(SDValue(N, 1), NewVAARG.getValue(1)); return NewVAARG; Index: ../lib/CodeGen/SelectionDAG/LegalizeIntegerTypes.cpp =================================================================== --- ../lib/CodeGen/SelectionDAG/LegalizeIntegerTypes.cpp +++ ../lib/CodeGen/SelectionDAG/LegalizeIntegerTypes.cpp @@ -66,8 +66,11 @@ case ISD::CTTZ: Res = PromoteIntRes_CTTZ(N); break; case ISD::EXTRACT_VECTOR_ELT: Res = PromoteIntRes_EXTRACT_VECTOR_ELT(N); break; - case ISD::LOAD: Res = PromoteIntRes_LOAD(cast(N));break; - case ISD::MLOAD: Res = PromoteIntRes_MLOAD(cast(N));break; + case ISD::LOAD: Res = PromoteIntRes_LOAD(cast(N)); break; + case ISD::MLOAD: Res = PromoteIntRes_MLOAD(cast(N)); + break; + case ISD::MGATHER: Res = PromoteIntRes_MGATHER(cast(N)); + break; case ISD::SELECT: Res = PromoteIntRes_SELECT(N); break; case ISD::VSELECT: Res = PromoteIntRes_VSELECT(N); break; case ISD::SELECT_CC: Res = PromoteIntRes_SELECT_CC(N); break; @@ -185,7 +188,7 @@ N->getChain(), N->getBasePtr(), N->getMemOperand(), N->getOrdering(), N->getSynchScope()); - // Legalized the chain result - switch anything that used the old chain to + // Legalize the chain result - switch anything that used the old chain to // use the new one. ReplaceValueWith(SDValue(N, 1), Res.getValue(1)); return Res; @@ -198,7 +201,7 @@ N->getChain(), N->getBasePtr(), Op2, N->getMemOperand(), N->getOrdering(), N->getSynchScope()); - // Legalized the chain result - switch anything that used the old chain to + // Legalize the chain result - switch anything that used the old chain to // use the new one. ReplaceValueWith(SDValue(N, 1), Res.getValue(1)); return Res; @@ -483,7 +486,7 @@ SDValue Res = DAG.getExtLoad(ExtType, dl, NVT, N->getChain(), N->getBasePtr(), N->getMemoryVT(), N->getMemOperand()); - // Legalized the chain result - switch anything that used the old chain to + // Legalize the chain result - switch anything that used the old chain to // use the new one. ReplaceValueWith(SDValue(N, 1), Res.getValue(1)); return Res; @@ -502,11 +505,30 @@ SDValue Res = DAG.getMaskedLoad(NVT, dl, N->getChain(), N->getBasePtr(), Mask, ExtSrc0, N->getMemoryVT(), N->getMemOperand(), ISD::SEXTLOAD); - // Legalized the chain result - switch anything that used the old chain to + // Legalize the chain result - switch anything that used the old chain to // use the new one. ReplaceValueWith(SDValue(N, 1), Res.getValue(1)); return Res; } + +SDValue DAGTypeLegalizer::PromoteIntRes_MGATHER(MaskedGatherSDNode *N) { + EVT NVT = TLI.getTypeToTransformTo(*DAG.getContext(), N->getValueType(0)); + SDValue ExtSrc0 = GetPromotedInteger(N->getValue()); + assert(NVT == ExtSrc0.getValueType() && + "Gather result type and the passThru agrument type should be the same"); + + SDLoc dl(N); + SDValue Ops[] = {N->getChain(), ExtSrc0, N->getMask(), N->getBasePtr(), + N->getIndex()}; + SDValue Res = DAG.getMaskedGather(DAG.getVTList(NVT, MVT::Other), + N->getMemoryVT(), dl, Ops, + N->getMemOperand()); + // Legalize the chain result - switch anything that used the old chain to + // use the new one. + ReplaceValueWith(SDValue(N, 1), Res.getValue(1)); + return Res; +} + /// Promote the overflow flag of an overflowing arithmetic node. SDValue DAGTypeLegalizer::PromoteIntRes_Overflow(SDNode *N) { // Simply change the return type of the boolean result. @@ -893,6 +915,10 @@ OpNo); break; case ISD::MLOAD: Res = PromoteIntOp_MLOAD(cast(N), OpNo); break; + case ISD::MGATHER: Res = PromoteIntOp_MGATHER(cast(N), + OpNo); break; + case ISD::MSCATTER: Res = PromoteIntOp_MSCATTER(cast(N), + OpNo); break; case ISD::TRUNCATE: Res = PromoteIntOp_TRUNCATE(N); break; case ISD::FP16_TO_FP: case ISD::UINT_TO_FP: Res = PromoteIntOp_UINT_TO_FP(N); break; @@ -1161,7 +1187,8 @@ N->getMemoryVT(), N->getMemOperand()); } -SDValue DAGTypeLegalizer::PromoteIntOp_MSTORE(MaskedStoreSDNode *N, unsigned OpNo){ +SDValue DAGTypeLegalizer::PromoteIntOp_MSTORE(MaskedStoreSDNode *N, + unsigned OpNo) { SDValue DataOp = N->getValue(); EVT DataVT = DataOp.getValueType(); @@ -1181,36 +1208,18 @@ assert(getTypeAction(DataVT) == TargetLowering::TypeWidenVector && "Unexpected data legalization in MSTORE"); DataOp = GetWidenedVector(DataOp); - - if (getTypeAction(MaskVT) == TargetLowering::TypeWidenVector) - Mask = GetWidenedVector(Mask); - else { - EVT BoolVT = getSetCCResultType(DataOp.getValueType()); - - // We can't use ModifyToType() because we should fill the mask with - // zeroes - unsigned WidenNumElts = BoolVT.getVectorNumElements(); - unsigned MaskNumElts = MaskVT.getVectorNumElements(); - - unsigned NumConcat = WidenNumElts / MaskNumElts; - SmallVector Ops(NumConcat); - SDValue ZeroVal = DAG.getConstant(0, dl, MaskVT); - Ops[0] = Mask; - for (unsigned i = 1; i != NumConcat; ++i) - Ops[i] = ZeroVal; - - Mask = DAG.getNode(ISD::CONCAT_VECTORS, dl, BoolVT, Ops); - } + Mask = WidenTargetBoolean(Mask, DataOp.getValueType(), true); } } else - Mask = PromoteTargetBoolean(N->getMask(), DataOp.getValueType()); + Mask = PromoteTargetBoolean(Mask, DataOp.getValueType()); return DAG.getMaskedStore(N->getChain(), dl, DataOp, N->getBasePtr(), Mask, N->getMemoryVT(), N->getMemOperand(), TruncateStore); } -SDValue DAGTypeLegalizer::PromoteIntOp_MLOAD(MaskedLoadSDNode *N, unsigned OpNo){ +SDValue DAGTypeLegalizer::PromoteIntOp_MLOAD(MaskedLoadSDNode *N, + unsigned OpNo) { assert(OpNo == 2 && "Only know how to promote the mask!"); EVT DataVT = N->getValueType(0); SDValue Mask = PromoteTargetBoolean(N->getOperand(OpNo), DataVT); @@ -1219,6 +1228,31 @@ return SDValue(DAG.UpdateNodeOperands(N, NewOps), 0); } +SDValue DAGTypeLegalizer::PromoteIntOp_MGATHER(MaskedGatherSDNode *N, + unsigned OpNo) { + + SmallVector NewOps(N->op_begin(), N->op_end()); + if (OpNo == 2) { + // The Mask + EVT DataVT = N->getValueType(0); + NewOps[OpNo] = PromoteTargetBoolean(N->getOperand(OpNo), DataVT); + } else + NewOps[OpNo] = GetPromotedInteger(N->getOperand(OpNo)); + return SDValue(DAG.UpdateNodeOperands(N, NewOps), 0); +} + +SDValue DAGTypeLegalizer::PromoteIntOp_MSCATTER(MaskedScatterSDNode *N, + unsigned OpNo) { + SmallVector NewOps(N->op_begin(), N->op_end()); + if (OpNo == 2) { + // The Mask + EVT DataVT = N->getValue().getValueType(); + NewOps[OpNo] = PromoteTargetBoolean(N->getOperand(OpNo), DataVT); + } else + NewOps[OpNo] = GetPromotedInteger(N->getOperand(OpNo)); + return SDValue(DAG.UpdateNodeOperands(N, NewOps), 0); +} + SDValue DAGTypeLegalizer::PromoteIntOp_TRUNCATE(SDNode *N) { SDValue Op = GetPromotedInteger(N->getOperand(0)); return DAG.getNode(ISD::TRUNCATE, SDLoc(N), N->getValueType(0), Op); @@ -2075,7 +2109,7 @@ } } - // Legalized the chain result - switch anything that used the old chain to + // Legalize the chain result - switch anything that used the old chain to // use the new one. ReplaceValueWith(SDValue(N, 1), Ch); } Index: ../lib/CodeGen/SelectionDAG/LegalizeTypes.h =================================================================== --- ../lib/CodeGen/SelectionDAG/LegalizeTypes.h +++ ../lib/CodeGen/SelectionDAG/LegalizeTypes.h @@ -173,6 +173,11 @@ std::pair ExpandAtomic(SDNode *Node); SDValue PromoteTargetBoolean(SDValue Bool, EVT ValVT); + + /// Modify Bit Vector to match SetCC result type if ValVT. + /// The bit vector is widened with zeroes when WithZeroes is true. + SDValue WidenTargetBoolean(SDValue Bool, EVT ValVT, bool WithZeroes = false); + void ReplaceValueWith(SDValue From, SDValue To); void SplitInteger(SDValue Op, SDValue &Lo, SDValue &Hi); void SplitInteger(SDValue Op, EVT LoVT, EVT HiVT, @@ -247,6 +252,7 @@ SDValue PromoteIntRes_INT_EXTEND(SDNode *N); SDValue PromoteIntRes_LOAD(LoadSDNode *N); SDValue PromoteIntRes_MLOAD(MaskedLoadSDNode *N); + SDValue PromoteIntRes_MGATHER(MaskedGatherSDNode *N); SDValue PromoteIntRes_Overflow(SDNode *N); SDValue PromoteIntRes_SADDSUBO(SDNode *N, unsigned ResNo); SDValue PromoteIntRes_SDIV(SDNode *N); @@ -293,6 +299,8 @@ SDValue PromoteIntOp_ZERO_EXTEND(SDNode *N); SDValue PromoteIntOp_MSTORE(MaskedStoreSDNode *N, unsigned OpNo); SDValue PromoteIntOp_MLOAD(MaskedLoadSDNode *N, unsigned OpNo); + SDValue PromoteIntOp_MSCATTER(MaskedScatterSDNode *N, unsigned OpNo); + SDValue PromoteIntOp_MGATHER(MaskedGatherSDNode *N, unsigned OpNo); void PromoteSetCCOperands(SDValue &LHS,SDValue &RHS, ISD::CondCode Code); @@ -675,6 +683,7 @@ SDValue WidenVecRes_INSERT_VECTOR_ELT(SDNode* N); SDValue WidenVecRes_LOAD(SDNode* N); SDValue WidenVecRes_MLOAD(MaskedLoadSDNode* N); + SDValue WidenVecRes_MGATHER(MaskedGatherSDNode* N); SDValue WidenVecRes_SCALAR_TO_VECTOR(SDNode* N); SDValue WidenVecRes_SELECT(SDNode* N); SDValue WidenVecRes_SELECT_CC(SDNode* N); @@ -702,6 +711,7 @@ SDValue WidenVecOp_EXTRACT_SUBVECTOR(SDNode *N); SDValue WidenVecOp_STORE(SDNode* N); SDValue WidenVecOp_MSTORE(SDNode* N, unsigned OpNo); + SDValue WidenVecOp_MSCATTER(SDNode* N, unsigned OpNo); SDValue WidenVecOp_SETCC(SDNode* N); SDValue WidenVecOp_Convert(SDNode *N); @@ -741,8 +751,10 @@ /// Modifies a vector input (widen or narrows) to a vector of NVT. The /// input vector must have the same element type as NVT. - SDValue ModifyToType(SDValue InOp, EVT WidenVT); - + /// When FillWithZeroes is "on" the vector will be widened with + /// zeroes. + /// By default, the vector will be widened with undefined values. + SDValue ModifyToType(SDValue InOp, EVT NVT, bool FillWithZeroes = false); //===--------------------------------------------------------------------===// // Generic Splitting: LegalizeTypesGeneric.cpp Index: ../lib/CodeGen/SelectionDAG/LegalizeTypes.cpp =================================================================== --- ../lib/CodeGen/SelectionDAG/LegalizeTypes.cpp +++ ../lib/CodeGen/SelectionDAG/LegalizeTypes.cpp @@ -1104,6 +1104,29 @@ return DAG.getNode(ExtendCode, dl, BoolVT, Bool); } +/// WidenTargetBoolean - Widen the given target boolean to a target boolean +/// of the given type. The boolean vector is promoted (if necessary), +/// widened and then extended or truncated to match the target boolean +/// type of the given ValVT. +SDValue DAGTypeLegalizer::WidenTargetBoolean(SDValue Bool, EVT ValVT, + bool WithZeroes) { + SDLoc dl(Bool); + if (getTypeAction(Bool.getValueType()) == TargetLowering::TypePromoteInteger) + Bool = GetPromotedInteger(Bool); + + EVT WideVT = EVT::getVectorVT(*DAG.getContext(), + Bool.getValueType().getScalarType(), + ValVT.getVectorNumElements()); + Bool = ModifyToType(Bool, WideVT, WithZeroes); + + EVT TargetBoolVT = getSetCCResultType(ValVT); + if (WideVT.bitsGT(TargetBoolVT)) + return DAG.getNode(ISD::TRUNCATE, dl, TargetBoolVT, Bool); + if (WideVT.bitsLT(TargetBoolVT)) + return DAG.getNode(ISD::SIGN_EXTEND, dl, TargetBoolVT, Bool); + return Bool; +} + /// SplitInteger - Return the lower LoVT bits of Op in Lo and the upper HiVT /// bits in Hi. void DAGTypeLegalizer::SplitInteger(SDValue Op, Index: ../lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp =================================================================== --- ../lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp +++ ../lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp @@ -231,7 +231,7 @@ N->isInvariant(), N->getOriginalAlignment(), N->getAAInfo()); - // Legalized the chain result - switch anything that used the old chain to + // Legalize the chain result - switch anything that used the old chain to // use the new one. ReplaceValueWith(SDValue(N, 1), Result.getValue(1)); return Result; @@ -1018,7 +1018,7 @@ Ch = DAG.getNode(ISD::TokenFactor, dl, MVT::Other, Lo.getValue(1), Hi.getValue(1)); - // Legalized the chain result - switch anything that used the old chain to + // Legalize the chain result - switch anything that used the old chain to // use the new one. ReplaceValueWith(SDValue(LD, 1), Ch); } @@ -1078,7 +1078,7 @@ Ch = DAG.getNode(ISD::TokenFactor, dl, MVT::Other, Lo.getValue(1), Hi.getValue(1)); - // Legalized the chain result - switch anything that used the old chain to + // Legalize the chain result - switch anything that used the old chain to // use the new one. ReplaceValueWith(SDValue(MLD, 1), Ch); @@ -1126,7 +1126,7 @@ Ch = DAG.getNode(ISD::TokenFactor, dl, MVT::Other, Lo.getValue(1), Hi.getValue(1)); - // Legalized the chain result - switch anything that used the old chain to + // Legalize the chain result - switch anything that used the old chain to // use the new one. ReplaceValueWith(SDValue(MGT, 1), Ch); } @@ -1639,7 +1639,7 @@ Ch = DAG.getNode(ISD::TokenFactor, dl, MVT::Other, Lo.getValue(1), Hi.getValue(1)); - // Legalized the chain result - switch anything that used the old chain to + // Legalize the chain result - switch anything that used the old chain to // use the new one. ReplaceValueWith(SDValue(MGT, 1), Ch); @@ -1973,6 +1973,9 @@ case ISD::MLOAD: Res = WidenVecRes_MLOAD(cast(N)); break; + case ISD::MGATHER: + Res = WidenVecRes_MGATHER(cast(N)); + break; case ISD::ADD: case ISD::AND: @@ -2691,38 +2694,48 @@ } SDValue DAGTypeLegalizer::WidenVecRes_MLOAD(MaskedLoadSDNode *N) { - - EVT WidenVT = TLI.getTypeToTransformTo(*DAG.getContext(),N->getValueType(0)); + + EVT WideVT = TLI.getTypeToTransformTo(*DAG.getContext(),N->getValueType(0)); SDValue Mask = N->getMask(); - EVT MaskVT = Mask.getValueType(); SDValue Src0 = GetWidenedVector(N->getSrc0()); ISD::LoadExtType ExtType = N->getExtensionType(); SDLoc dl(N); - if (getTypeAction(MaskVT) == TargetLowering::TypeWidenVector) - Mask = GetWidenedVector(Mask); - else { - EVT BoolVT = getSetCCResultType(WidenVT); - - // We can't use ModifyToType() because we should fill the mask with - // zeroes - unsigned WidenNumElts = BoolVT.getVectorNumElements(); - unsigned MaskNumElts = MaskVT.getVectorNumElements(); - - unsigned NumConcat = WidenNumElts / MaskNumElts; - SmallVector Ops(NumConcat); - SDValue ZeroVal = DAG.getConstant(0, dl, MaskVT); - Ops[0] = Mask; - for (unsigned i = 1; i != NumConcat; ++i) - Ops[i] = ZeroVal; - - Mask = DAG.getNode(ISD::CONCAT_VECTORS, dl, BoolVT, Ops); - } + // The mask should be widened as well + Mask = WidenTargetBoolean(Mask, WideVT, true); - SDValue Res = DAG.getMaskedLoad(WidenVT, dl, N->getChain(), N->getBasePtr(), + SDValue Res = DAG.getMaskedLoad(WideVT, dl, N->getChain(), N->getBasePtr(), Mask, Src0, N->getMemoryVT(), N->getMemOperand(), ExtType); - // Legalized the chain result - switch anything that used the old chain to + // Legalize the chain result - switch anything that used the old chain to + // use the new one. + ReplaceValueWith(SDValue(N, 1), Res.getValue(1)); + return Res; +} + +SDValue DAGTypeLegalizer::WidenVecRes_MGATHER(MaskedGatherSDNode *N) { + + EVT WideVT = TLI.getTypeToTransformTo(*DAG.getContext(), N->getValueType(0)); + SDValue Mask = N->getMask(); + SDValue Src0 = GetWidenedVector(N->getValue()); + unsigned NumElts = WideVT.getVectorNumElements(); + SDLoc dl(N); + + // The mask should be widened as well + Mask = WidenTargetBoolean(Mask, WideVT, true); + + // Widen the Index operand + SDValue Index = N->getIndex(); + EVT WideIndexVT = EVT::getVectorVT(*DAG.getContext(), + Index.getValueType().getScalarType(), + NumElts); + Index = ModifyToType(Index, WideIndexVT); + SDValue Ops[] = { N->getChain(), Src0, Mask, N->getBasePtr(), Index }; + SDValue Res = DAG.getMaskedGather(DAG.getVTList(WideVT, MVT::Other), + N->getMemoryVT(), dl, Ops, + N->getMemOperand()); + + // Legalize the chain result - switch anything that used the old chain to // use the new one. ReplaceValueWith(SDValue(N, 1), Res.getValue(1)); return Res; @@ -2884,6 +2897,7 @@ case ISD::EXTRACT_VECTOR_ELT: Res = WidenVecOp_EXTRACT_VECTOR_ELT(N); break; case ISD::STORE: Res = WidenVecOp_STORE(N); break; case ISD::MSTORE: Res = WidenVecOp_MSTORE(N, OpNo); break; + case ISD::MSCATTER: Res = WidenVecOp_MSCATTER(N, OpNo); break; case ISD::SETCC: Res = WidenVecOp_SETCC(N); break; case ISD::FCOPYSIGN: Res = WidenVecOp_FCOPYSIGN(N); break; @@ -3098,39 +3112,51 @@ SDValue DAGTypeLegalizer::WidenVecOp_MSTORE(SDNode *N, unsigned OpNo) { MaskedStoreSDNode *MST = cast(N); SDValue Mask = MST->getMask(); - EVT MaskVT = Mask.getValueType(); SDValue StVal = MST->getValue(); + + assert(OpNo == 3 && "Unexpected operand number"); // Widen the value SDValue WideVal = GetWidenedVector(StVal); + EVT WideVT = WideVal.getValueType(); SDLoc dl(N); - if (OpNo == 2 || getTypeAction(MaskVT) == TargetLowering::TypeWidenVector) - Mask = GetWidenedVector(Mask); - else { - // The mask should be widened as well - EVT BoolVT = getSetCCResultType(WideVal.getValueType()); - // We can't use ModifyToType() because we should fill the mask with - // zeroes - unsigned WidenNumElts = BoolVT.getVectorNumElements(); - unsigned MaskNumElts = MaskVT.getVectorNumElements(); - - unsigned NumConcat = WidenNumElts / MaskNumElts; - SmallVector Ops(NumConcat); - SDValue ZeroVal = DAG.getConstant(0, dl, MaskVT); - Ops[0] = Mask; - for (unsigned i = 1; i != NumConcat; ++i) - Ops[i] = ZeroVal; + // The mask should be widened as well + Mask = WidenTargetBoolean(Mask, WideVT, true); - Mask = DAG.getNode(ISD::CONCAT_VECTORS, dl, BoolVT, Ops); - } - assert(Mask.getValueType().getVectorNumElements() == - WideVal.getValueType().getVectorNumElements() && - "Mask and data vectors should have the same number of elements"); return DAG.getMaskedStore(MST->getChain(), dl, WideVal, MST->getBasePtr(), Mask, MST->getMemoryVT(), MST->getMemOperand(), false); } +SDValue DAGTypeLegalizer::WidenVecOp_MSCATTER(SDNode *N, unsigned OpNo) { + assert(OpNo == 1 && "Can widen only data operand of mscatter"); + MaskedScatterSDNode *MSC = cast(N); + SDValue DataOp = MSC->getValue(); + SDValue Mask = MSC->getMask(); + + assert(OpNo == 1 && "Unexpected operand number"); + // Widen the value + SDValue WideVal = GetWidenedVector(DataOp); + EVT WideVT = WideVal.getValueType(); + unsigned NumElts = WideVal.getValueType().getVectorNumElements(); + SDLoc dl(N); + + // The mask should be widened as well + Mask = WidenTargetBoolean(Mask, WideVT, true); + + // Widen index + SDValue Index = MSC->getIndex(); + EVT WideIndexVT = EVT::getVectorVT(*DAG.getContext(), + Index.getValueType().getScalarType(), + NumElts); + Index = ModifyToType(Index, WideIndexVT); + + SDValue Ops[] = {MSC->getChain(), WideVal, Mask, MSC->getBasePtr(), Index}; + return DAG.getMaskedScatter(DAG.getVTList(MVT::Other), + MSC->getMemoryVT(), dl, Ops, + MSC->getMemOperand()); +} + SDValue DAGTypeLegalizer::WidenVecOp_SETCC(SDNode *N) { SDValue InOp0 = GetWidenedVector(N->getOperand(0)); SDValue InOp1 = GetWidenedVector(N->getOperand(1)); @@ -3594,7 +3620,9 @@ /// Modifies a vector input (widen or narrows) to a vector of NVT. The /// input vector must have the same element type as NVT. -SDValue DAGTypeLegalizer::ModifyToType(SDValue InOp, EVT NVT) { +/// FillWithZeroes specifies that the vector should be widened with zeroes. +SDValue DAGTypeLegalizer::ModifyToType(SDValue InOp, EVT NVT, + bool FillWithZeroes) { // Note that InOp might have been widened so it might already have // the right width or it might need be narrowed. EVT InVT = InOp.getValueType(); @@ -3611,10 +3639,11 @@ if (WidenNumElts > InNumElts && WidenNumElts % InNumElts == 0) { unsigned NumConcat = WidenNumElts / InNumElts; SmallVector Ops(NumConcat); - SDValue UndefVal = DAG.getUNDEF(InVT); + SDValue FillVal = FillWithZeroes ? DAG.getConstant(0, dl, InVT) : + DAG.getUNDEF(InVT); Ops[0] = InOp; for (unsigned i = 1; i != NumConcat; ++i) - Ops[i] = UndefVal; + Ops[i] = FillVal; return DAG.getNode(ISD::CONCAT_VECTORS, dl, NVT, Ops); } @@ -3634,8 +3663,9 @@ ISD::EXTRACT_VECTOR_ELT, dl, EltVT, InOp, DAG.getConstant(Idx, dl, TLI.getVectorIdxTy(DAG.getDataLayout()))); - SDValue UndefVal = DAG.getUNDEF(EltVT); + SDValue FillVal = FillWithZeroes ? DAG.getConstant(0, dl, EltVT) : + DAG.getUNDEF(EltVT); for ( ; Idx < WidenNumElts; ++Idx) - Ops[Idx] = UndefVal; + Ops[Idx] = FillVal; return DAG.getNode(ISD::BUILD_VECTOR, dl, NVT, Ops); } Index: ../lib/Target/X86/X86ISelLowering.cpp =================================================================== --- ../lib/Target/X86/X86ISelLowering.cpp +++ ../lib/Target/X86/X86ISelLowering.cpp @@ -1559,7 +1559,7 @@ setOperationAction(ISD::OR, VT, Legal); setOperationAction(ISD::XOR, VT, Legal); } - if (EltSize >= 32 && VT.getSizeInBits() <= 512) { + if ((VT.is128BitVector() || VT.is256BitVector()) && EltSize >= 32) { setOperationAction(ISD::MGATHER, VT, Custom); setOperationAction(ISD::MSCATTER, VT, Custom); } @@ -1585,6 +1585,8 @@ setOperationAction(ISD::INSERT_SUBVECTOR, VT, Custom); setOperationAction(ISD::MLOAD, VT, Legal); setOperationAction(ISD::MSTORE, VT, Legal); + setOperationAction(ISD::MGATHER, VT, Legal); + setOperationAction(ISD::MSCATTER, VT, Custom); } } for (auto VT : { MVT::v64i8, MVT::v32i16, MVT::v16i32 }) { @@ -1789,6 +1791,8 @@ setTargetDAGCombine(ISD::BUILD_VECTOR); setTargetDAGCombine(ISD::MUL); setTargetDAGCombine(ISD::XOR); + setTargetDAGCombine(ISD::MSCATTER); + setTargetDAGCombine(ISD::MGATHER); computeRegisterProperties(Subtarget->getRegisterInfo()); @@ -13231,6 +13235,13 @@ return DAG.getNode(X86ISD::VTRUNC, DL, VT, In); } if ((VT == MVT::v4i32) && (InVT == MVT::v4i64)) { + if (In.getOpcode() == ISD::CONCAT_VECTORS && In.getNumOperands() == 2) { + static const int ShufMask[] = {0, 2, 4, 6}; + return DAG.getVectorShuffle(VT, DL, + DAG.getBitcast(MVT::v4i32, In.getOperand(0)), + DAG.getBitcast(MVT::v4i32, In.getOperand(1)), + ShufMask); + } // On AVX2, v4i64 -> v4i32 becomes VPERMD. if (Subtarget->hasInt256()) { static const int ShufMask[] = {0, 2, 4, 6, -1, -1, -1, -1}; @@ -19647,33 +19658,146 @@ return DAG.getNode(ISD::MERGE_VALUES, dl, Tys, SinVal, CosVal); } +/// Widen a vector input to a vector of NVT. The +/// input vector must have the same element type as NVT. +static SDValue ExtendToType(SDValue InOp, MVT NVT, SelectionDAG &DAG, + bool FillWithZeroes = false) { + // Check if InOp already has the right width. + MVT InVT = InOp.getSimpleValueType(); + if (InVT == NVT) + return InOp; + + if (InOp.isUndef()) + return DAG.getUNDEF(NVT); + + assert(InVT.getVectorElementType() == NVT.getVectorElementType() && + "input and widen element type must match"); + + unsigned InNumElts = InVT.getVectorNumElements(); + unsigned WidenNumElts = NVT.getVectorNumElements(); + assert(WidenNumElts > InNumElts && WidenNumElts % InNumElts == 0 && + "Unexpected request for vector widening"); + + EVT EltVT = NVT.getVectorElementType(); + + SDLoc dl(InOp); + if (InOp.getOpcode() == ISD::CONCAT_VECTORS && + InOp.getNumOperands() == 2) { + SDValue N1 = InOp.getOperand(1); + if ((ISD::isBuildVectorAllZeros(N1.getNode()) && FillWithZeroes) || + N1.isUndef()) { + InOp = InOp.getOperand(0); + InVT = InOp.getSimpleValueType(); + InNumElts = InVT.getVectorNumElements(); + } + } + if (ISD::isBuildVectorOfConstantSDNodes(InOp.getNode())) { + // Special case, because CONCAT_VECTORS with many operands is not + // converted to the BUILD_VECTOR + SmallVector Ops; + for (unsigned i = 0; i < InNumElts; ++i) + Ops.push_back(InOp.getOperand(i)); + + SDValue FillVal = FillWithZeroes ? DAG.getConstant(0, dl, EltVT) : + DAG.getUNDEF(EltVT); + for (unsigned i = 0; i < WidenNumElts - InNumElts; ++i) + Ops.push_back(FillVal); + return DAG.getNode(ISD::BUILD_VECTOR, dl, NVT, Ops); + } + SDValue FillVal = FillWithZeroes ? DAG.getConstant(0, dl, NVT) : DAG.getUNDEF(NVT); + return DAG.getNode(ISD::INSERT_SUBVECTOR, dl, NVT, FillVal, + InOp, DAG.getIntPtrConstant(0, dl)); +} + static SDValue LowerMSCATTER(SDValue Op, const X86Subtarget *Subtarget, SelectionDAG &DAG) { assert(Subtarget->hasAVX512() && "MGATHER/MSCATTER are supported on AVX-512 arch only"); + // X86 scatter kills mask register, so its type should be added to + // the list of return values. + // If the "scatter" has 2 return values, it is already handled. + if (Op.getNode()->getNumValues() == 2) + return Op; + MaskedScatterSDNode *N = cast(Op.getNode()); - MVT VT = N->getValue().getSimpleValueType(); + SDValue Src = N->getValue(); + MVT VT = Src.getSimpleValueType(); assert(VT.getScalarSizeInBits() >= 32 && "Unsupported scatter op"); SDLoc dl(Op); - // X86 scatter kills mask register, so its type should be added to - // the list of return values - if (N->getNumValues() == 1) { - SDValue Index = N->getIndex(); - if (!Subtarget->hasVLX() && !VT.is512BitVector() && - !Index.getSimpleValueType().is512BitVector()) - Index = DAG.getNode(ISD::SIGN_EXTEND, dl, MVT::v8i64, Index); - - SDVTList VTs = DAG.getVTList(N->getMask().getValueType(), MVT::Other); - SDValue Ops[] = { N->getOperand(0), N->getOperand(1), N->getOperand(2), - N->getOperand(3), Index }; - - SDValue NewScatter = DAG.getMaskedScatter(VTs, VT, dl, Ops, N->getMemOperand()); - DAG.ReplaceAllUsesWith(Op, SDValue(NewScatter.getNode(), 1)); - return SDValue(NewScatter.getNode(), 0); + SDValue NewScatter; + SDValue Index = N->getIndex(); + SDValue Mask = N->getMask(); + SDValue Chain = N->getChain(); + SDValue BasePtr = N->getBasePtr(); + MVT MemVT = N->getMemoryVT().getSimpleVT(); + MVT IndexVT = Index.getSimpleValueType(); + MVT MaskVT = Mask.getSimpleValueType(); + + if (MemVT.getScalarSizeInBits() < VT.getScalarSizeInBits()) { + // Promoted data type + assert((MemVT == MVT::v2i32 && VT == MVT::v2i64) && + "Unexpected memory type"); + int ShuffleMask[] = {0, 2, -1, -1}; + Src = DAG.getVectorShuffle(MVT::v4i32, dl, DAG.getBitcast(MVT::v4i32, Src), + DAG.getUNDEF(MVT::v4i32), ShuffleMask); + // Now we have 4 elements instead of 2. + // Expand the index. + MVT NewIndexVT = MVT::getVectorVT(IndexVT.getScalarType(), 4); + Index = ExtendToType(Index, NewIndexVT, DAG); + + // Expand the mask with zeroes + // Mask may be <2 x i64> or <2 x i1> at this moment + assert((MaskVT == MVT::v2i1 || MaskVT == MVT::v2i64) && + "Unexpected mask type"); + MVT ExtMaskVT = MVT::getVectorVT(MaskVT.getScalarType(), 4); + Mask = ExtendToType(Mask, ExtMaskVT, DAG, true); + VT = MVT::v4i32; } - return Op; + + unsigned NumElts = VT.getVectorNumElements(); + if (!Subtarget->hasVLX() && !VT.is512BitVector() && + !Index.getSimpleValueType().is512BitVector()) { + // AVX512F supports only 512-bit vectors. Or data or index should + // be 512 bit wide. If now the both index and data are 256-bit, but + // the vector contains 8 elements, we just sign-extend the index + if (IndexVT == MVT::v8i32) + // Just extend index + Index = DAG.getNode(ISD::SIGN_EXTEND, dl, MVT::v8i64, Index); + else { + // The minimal number of elts in scatter is 8 + NumElts = 8; + // Index + MVT NewIndexVT = MVT::getVectorVT(IndexVT.getScalarType(), NumElts); + // Use original index here, do not modify the index twice + Index = ExtendToType(N->getIndex(), NewIndexVT, DAG); + if (IndexVT.getScalarType() == MVT::i32) + Index = DAG.getNode(ISD::SIGN_EXTEND, dl, MVT::v8i64, Index); + + // Mask + // At this point we have promoted mask operand + assert(MaskVT.getScalarSizeInBits() >= 32 && "unexpected mask type"); + MVT ExtMaskVT = MVT::getVectorVT(MaskVT.getScalarType(), NumElts); + // Use the original mask here, do not modify the mask twice + Mask = ExtendToType(N->getMask(), ExtMaskVT, DAG, true); + + // The value that should be stored + MVT NewVT = MVT::getVectorVT(VT.getScalarType(), NumElts); + Src = ExtendToType(Src, NewVT, DAG); + } + } + // If the mask is "wide" at this point - truncate it to i1 vector + MVT BitMaskVT = MVT::getVectorVT(MVT::i1, NumElts); + Mask = DAG.getNode(ISD::TRUNCATE, dl, BitMaskVT, Mask); + + // The mask is killed by scatter, add it to the values + SDVTList VTs = DAG.getVTList(BitMaskVT, MVT::Other); + SDValue Ops[] = {Chain, Src, Mask, BasePtr, Index}; + NewScatter = DAG.getMaskedScatter(VTs, N->getMemoryVT(), dl, Ops, + N->getMemOperand()); + DAG.ReplaceAllUsesWith(Op, SDValue(NewScatter.getNode(), 1)); + return SDValue(NewScatter.getNode(), 0); } static SDValue LowerMGATHER(SDValue Op, const X86Subtarget *Subtarget, @@ -19682,17 +19806,59 @@ "MGATHER/MSCATTER are supported on AVX-512 arch only"); MaskedGatherSDNode *N = cast(Op.getNode()); + SDLoc dl(Op); MVT VT = Op.getSimpleValueType(); + SDValue Index = N->getIndex(); + SDValue Mask = N->getMask(); + SDValue Src0 = N->getValue(); + MVT IndexVT = Index.getSimpleValueType(); + MVT MaskVT = Mask.getSimpleValueType(); + + unsigned NumElts = VT.getVectorNumElements(); assert(VT.getScalarSizeInBits() >= 32 && "Unsupported gather op"); - SDLoc dl(Op); - SDValue Index = N->getIndex(); if (!Subtarget->hasVLX() && !VT.is512BitVector() && !Index.getSimpleValueType().is512BitVector()) { - Index = DAG.getNode(ISD::SIGN_EXTEND, dl, MVT::v8i64, Index); - SDValue Ops[] = { N->getOperand(0), N->getOperand(1), N->getOperand(2), - N->getOperand(3), Index }; - DAG.UpdateNodeOperands(N, Ops); + // AVX512F supports only 512-bit vectors. Or data or index should + // be 512 bit wide. If now the both index and data are 256-bit, but + // the vector contains 8 elements, we just sign-extend the index + if (NumElts == 8) { + Index = DAG.getNode(ISD::SIGN_EXTEND, dl, MVT::v8i64, Index); + SDValue Ops[] = { N->getOperand(0), N->getOperand(1), N->getOperand(2), + N->getOperand(3), Index }; + DAG.UpdateNodeOperands(N, Ops); + return Op; + } + + // Minimal number of elements in Gather + NumElts = 8; + // Index + MVT NewIndexVT = MVT::getVectorVT(IndexVT.getScalarType(), NumElts); + Index = ExtendToType(Index, NewIndexVT, DAG); + if (IndexVT.getScalarType() == MVT::i32) + Index = DAG.getNode(ISD::SIGN_EXTEND, dl, MVT::v8i64, Index); + + // Mask + MVT MaskBitVT = MVT::getVectorVT(MVT::i1, NumElts); + // At this point we have promoted mask operand + assert(MaskVT.getScalarSizeInBits() >= 32 && "unexpected mask type"); + MVT ExtMaskVT = MVT::getVectorVT(MaskVT.getScalarType(), NumElts); + Mask = ExtendToType(Mask, ExtMaskVT, DAG, true); + Mask = DAG.getNode(ISD::TRUNCATE, dl, MaskBitVT, Mask); + + // The pass-thru value + MVT NewVT = MVT::getVectorVT(VT.getScalarType(), NumElts); + Src0 = ExtendToType(Src0, NewVT, DAG); + + SDValue Ops[] = { N->getChain(), Src0, Mask, N->getBasePtr(), Index }; + SDValue NewGather = DAG.getMaskedGather(DAG.getVTList(NewVT, MVT::Other), + N->getMemoryVT(), dl, Ops, + N->getMemOperand()); + SDValue Exract = DAG.getNode(ISD::EXTRACT_SUBVECTOR, dl, VT, + NewGather.getValue(0), + DAG.getIntPtrConstant(0, dl)); + SDValue RetOps[] = {Exract, NewGather.getValue(1)}; + return DAG.getMergeValues(RetOps, dl); } return Op; } @@ -26467,6 +26633,18 @@ return SDValue(); } +static SDValue PerformGatherScatterCombine(SDNode *N, SelectionDAG &DAG) { + SDLoc DL(N); + // The mask will be truncated anyway. The SIGN_EXTEND_INREG is redundant. + SDValue Mask = N->getOperand(2); + if (Mask.getOpcode() == ISD::SIGN_EXTEND_INREG) { + SmallVector NewOps(N->op_begin(), N->op_end()); + NewOps[2] = Mask.getOperand(0); + DAG.UpdateNodeOperands(N, NewOps); + } + return SDValue(); +} + // Helper function of PerformSETCCCombine. It is to materialize "setb reg" // as "sbb reg,reg", since it can be extended without zext and produces // an all-ones bit which is more useful than 0/1 in some cases. @@ -26907,6 +27085,8 @@ case ISD::VECTOR_SHUFFLE: return PerformShuffleCombine(N, DAG, DCI,Subtarget); case ISD::FMA: return PerformFMACombine(N, DAG, Subtarget); case X86ISD::BLENDI: return PerformBLENDICombine(N, DAG); + case ISD::MGATHER: + case ISD::MSCATTER: return PerformGatherScatterCombine(N, DAG); } return SDValue(); Index: ../test/CodeGen/X86/masked_gather_scatter.ll =================================================================== --- ../test/CodeGen/X86/masked_gather_scatter.ll +++ ../test/CodeGen/X86/masked_gather_scatter.ll @@ -1,29 +1,49 @@ -; RUN: llc -mtriple=x86_64-apple-darwin -mcpu=knl < %s | FileCheck %s -check-prefix=KNL +; RUN: llc -mtriple=x86_64-unknown-linux-gnu -mattr=+avx512f < %s | FileCheck %s --check-prefix=KNL_64 +; RUN: llc -mtriple=i386-unknown-linux-gnu -mattr=+avx512f < %s | FileCheck %s --check-prefix=KNL_32 +; RUN: llc -mtriple=x86_64-unknown-linux-gnu -mattr=+avx512vl -mattr=+avx512dq < %s | FileCheck %s --check-prefix=SKX ; RUN: opt -mtriple=x86_64-apple-darwin -codegenprepare -mcpu=corei7-avx -S < %s | FileCheck %s -check-prefix=SCALAR target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128" target triple = "x86_64-unknown-linux-gnu" -; KNL-LABEL: test1 -; KNL: kxnorw %k1, %k1, %k1 -; KNL: vgatherdps (%rdi,%zmm0,4), %zmm1 {%k1} ; SCALAR-LABEL: test1 -; SCALAR: extractelement <16 x float*> +; SCALAR: extractelement <16 x float*> ; SCALAR-NEXT: load float ; SCALAR-NEXT: insertelement <16 x float> ; SCALAR-NEXT: extractelement <16 x float*> ; SCALAR-NEXT: load float define <16 x float> @test1(float* %base, <16 x i32> %ind) { +; KNL_64-LABEL: test1: +; KNL_64: # BB#0: +; KNL_64-NEXT: kxnorw %k1, %k1, %k1 +; KNL_64-NEXT: vgatherdps (%rdi,%zmm0,4), %zmm1 {%k1} +; KNL_64-NEXT: vmovaps %zmm1, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test1: +; KNL_32: # BB#0: +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: kxnorw %k1, %k1, %k1 +; KNL_32-NEXT: vgatherdps (%eax,%zmm0,4), %zmm1 {%k1} +; KNL_32-NEXT: vmovaps %zmm1, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test1: +; SKX: # BB#0: +; SKX-NEXT: kxnorw %k1, %k1, %k1 +; SKX-NEXT: vgatherdps (%rdi,%zmm0,4), %zmm1 {%k1} +; SKX-NEXT: vmovaps %zmm1, %zmm0 +; SKX-NEXT: retq %broadcast.splatinsert = insertelement <16 x float*> undef, float* %base, i32 0 %broadcast.splat = shufflevector <16 x float*> %broadcast.splatinsert, <16 x float*> undef, <16 x i32> zeroinitializer %sext_ind = sext <16 x i32> %ind to <16 x i64> %gep.random = getelementptr float, <16 x float*> %broadcast.splat, <16 x i64> %sext_ind - + %res = call <16 x float> @llvm.masked.gather.v16f32(<16 x float*> %gep.random, i32 4, <16 x i1> , <16 x float> undef) ret <16 x float>%res } @@ -31,23 +51,41 @@ declare <16 x i32> @llvm.masked.gather.v16i32(<16 x i32*>, i32, <16 x i1>, <16 x i32>) declare <16 x float> @llvm.masked.gather.v16f32(<16 x float*>, i32, <16 x i1>, <16 x float>) declare <8 x i32> @llvm.masked.gather.v8i32(<8 x i32*> , i32, <8 x i1> , <8 x i32> ) - -; KNL-LABEL: test2 -; KNL: kmovw %esi, %k1 -; KNL: vgatherdps (%rdi,%zmm0,4), %zmm1 {%k1} + ; SCALAR-LABEL: test2 -; SCALAR: extractelement <16 x float*> +; SCALAR: extractelement <16 x float*> ; SCALAR-NEXT: load float ; SCALAR-NEXT: insertelement <16 x float> ; SCALAR-NEXT: br label %else ; SCALAR: else: -; SCALAR-NEXT: %res.phi.else = phi +; SCALAR-NEXT: %res.phi.else = phi ; SCALAR-NEXT: %Mask1 = extractelement <16 x i1> %imask, i32 1 ; SCALAR-NEXT: %ToLoad1 = icmp eq i1 %Mask1, true ; SCALAR-NEXT: br i1 %ToLoad1, label %cond.load1, label %else2 define <16 x float> @test2(float* %base, <16 x i32> %ind, i16 %mask) { +; KNL_64-LABEL: test2: +; KNL_64: # BB#0: +; KNL_64-NEXT: kmovw %esi, %k1 +; KNL_64-NEXT: vgatherdps (%rdi,%zmm0,4), %zmm1 {%k1} +; KNL_64-NEXT: vmovaps %zmm1, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test2: +; KNL_32: # BB#0: +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: kmovw {{[0-9]+}}(%esp), %k1 +; KNL_32-NEXT: vgatherdps (%eax,%zmm0,4), %zmm1 {%k1} +; KNL_32-NEXT: vmovaps %zmm1, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test2: +; SKX: # BB#0: +; SKX-NEXT: kmovw %esi, %k1 +; SKX-NEXT: vgatherdps (%rdi,%zmm0,4), %zmm1 {%k1} +; SKX-NEXT: vmovaps %zmm1, %zmm0 +; SKX-NEXT: retq %broadcast.splatinsert = insertelement <16 x float*> undef, float* %base, i32 0 %broadcast.splat = shufflevector <16 x float*> %broadcast.splatinsert, <16 x float*> undef, <16 x i32> zeroinitializer @@ -59,10 +97,28 @@ ret <16 x float> %res } -; KNL-LABEL: test3 -; KNL: kmovw %esi, %k1 -; KNL: vpgatherdd (%rdi,%zmm0,4), %zmm1 {%k1} define <16 x i32> @test3(i32* %base, <16 x i32> %ind, i16 %mask) { +; KNL_64-LABEL: test3: +; KNL_64: # BB#0: +; KNL_64-NEXT: kmovw %esi, %k1 +; KNL_64-NEXT: vpgatherdd (%rdi,%zmm0,4), %zmm1 {%k1} +; KNL_64-NEXT: vmovaps %zmm1, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test3: +; KNL_32: # BB#0: +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: kmovw {{[0-9]+}}(%esp), %k1 +; KNL_32-NEXT: vpgatherdd (%eax,%zmm0,4), %zmm1 {%k1} +; KNL_32-NEXT: vmovaps %zmm1, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test3: +; SKX: # BB#0: +; SKX-NEXT: kmovw %esi, %k1 +; SKX-NEXT: vpgatherdd (%rdi,%zmm0,4), %zmm1 {%k1} +; SKX-NEXT: vmovaps %zmm1, %zmm0 +; SKX-NEXT: retq %broadcast.splatinsert = insertelement <16 x i32*> undef, i32* %base, i32 0 %broadcast.splat = shufflevector <16 x i32*> %broadcast.splatinsert, <16 x i32*> undef, <16 x i32> zeroinitializer @@ -74,13 +130,38 @@ ret <16 x i32> %res } -; KNL-LABEL: test4 -; KNL: kmovw %esi, %k1 -; KNL: kmovw -; KNL: vpgatherdd -; KNL: vpgatherdd define <16 x i32> @test4(i32* %base, <16 x i32> %ind, i16 %mask) { +; KNL_64-LABEL: test4: +; KNL_64: # BB#0: +; KNL_64-NEXT: kmovw %esi, %k1 +; KNL_64-NEXT: kmovw %k1, %k2 +; KNL_64-NEXT: vpgatherdd (%rdi,%zmm0,4), %zmm1 {%k2} +; KNL_64-NEXT: vmovaps %zmm1, %zmm2 +; KNL_64-NEXT: vpgatherdd (%rdi,%zmm0,4), %zmm2 {%k1} +; KNL_64-NEXT: vpaddd %zmm2, %zmm1, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test4: +; KNL_32: # BB#0: +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: kmovw {{[0-9]+}}(%esp), %k1 +; KNL_32-NEXT: kmovw %k1, %k2 +; KNL_32-NEXT: vpgatherdd (%eax,%zmm0,4), %zmm1 {%k2} +; KNL_32-NEXT: vmovaps %zmm1, %zmm2 +; KNL_32-NEXT: vpgatherdd (%eax,%zmm0,4), %zmm2 {%k1} +; KNL_32-NEXT: vpaddd %zmm2, %zmm1, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test4: +; SKX: # BB#0: +; SKX-NEXT: kmovw %esi, %k1 +; SKX-NEXT: kmovw %k1, %k2 +; SKX-NEXT: vpgatherdd (%rdi,%zmm0,4), %zmm1 {%k2} +; SKX-NEXT: vmovaps %zmm1, %zmm2 +; SKX-NEXT: vpgatherdd (%rdi,%zmm0,4), %zmm2 {%k1} +; SKX-NEXT: vpaddd %zmm2, %zmm1, %zmm0 +; SKX-NEXT: retq %broadcast.splatinsert = insertelement <16 x i32*> undef, i32* %base, i32 0 %broadcast.splat = shufflevector <16 x i32*> %broadcast.splatinsert, <16 x i32*> undef, <16 x i32> zeroinitializer @@ -93,10 +174,6 @@ ret <16 x i32> %res } -; KNL-LABEL: test5 -; KNL: kmovw %k1, %k2 -; KNL: vpscatterdd {{.*}}%k2 -; KNL: vpscatterdd {{.*}}%k1 ; SCALAR-LABEL: test5 ; SCALAR: %Mask0 = extractelement <16 x i1> %imask, i32 0 @@ -113,6 +190,30 @@ ; SCALAR-NEXT: br i1 %ToStore1, label %cond.store1, label %else2 define void @test5(i32* %base, <16 x i32> %ind, i16 %mask, <16 x i32>%val) { +; KNL_64-LABEL: test5: +; KNL_64: # BB#0: +; KNL_64-NEXT: kmovw %esi, %k1 +; KNL_64-NEXT: kmovw %k1, %k2 +; KNL_64-NEXT: vpscatterdd %zmm1, (%rdi,%zmm0,4) {%k2} +; KNL_64-NEXT: vpscatterdd %zmm1, (%rdi,%zmm0,4) {%k1} +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test5: +; KNL_32: # BB#0: +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: kmovw {{[0-9]+}}(%esp), %k1 +; KNL_32-NEXT: kmovw %k1, %k2 +; KNL_32-NEXT: vpscatterdd %zmm1, (%eax,%zmm0,4) {%k2} +; KNL_32-NEXT: vpscatterdd %zmm1, (%eax,%zmm0,4) {%k1} +; KNL_32-NEXT: retl +; +; SKX-LABEL: test5: +; SKX: # BB#0: +; SKX-NEXT: kmovw %esi, %k1 +; SKX-NEXT: kmovw %k1, %k2 +; SKX-NEXT: vpscatterdd %zmm1, (%rdi,%zmm0,4) {%k2} +; SKX-NEXT: vpscatterdd %zmm1, (%rdi,%zmm0,4) {%k1} +; SKX-NEXT: retq %broadcast.splatinsert = insertelement <16 x i32*> undef, i32* %base, i32 0 %broadcast.splat = shufflevector <16 x i32*> %broadcast.splatinsert, <16 x i32*> undef, <16 x i32> zeroinitializer @@ -127,11 +228,6 @@ declare void @llvm.masked.scatter.v8i32(<8 x i32> , <8 x i32*> , i32 , <8 x i1> ) declare void @llvm.masked.scatter.v16i32(<16 x i32> , <16 x i32*> , i32 , <16 x i1> ) -; KNL-LABEL: test6 -; KNL: kxnorw %k1, %k1, %k1 -; KNL: kxnorw %k2, %k2, %k2 -; KNL: vpgatherqd (,%zmm{{.*}}), %ymm{{.*}} {%k2} -; KNL: vpscatterqd %ymm{{.*}}, (,%zmm{{.*}}) {%k1} ; SCALAR-LABEL: test6 ; SCALAR: store i32 %Elt0, i32* %Ptr01, align 4 @@ -143,6 +239,33 @@ ; SCALAR-NEXT: store i32 %Elt2, i32* %Ptr23, align 4 define <8 x i32> @test6(<8 x i32>%a1, <8 x i32*> %ptr) { +; KNL_64-LABEL: test6: +; KNL_64: # BB#0: +; KNL_64-NEXT: kxnorw %k1, %k1, %k1 +; KNL_64-NEXT: kxnorw %k2, %k2, %k2 +; KNL_64-NEXT: vpgatherqd (,%zmm1), %ymm2 {%k2} +; KNL_64-NEXT: vpscatterqd %ymm0, (,%zmm1) {%k1} +; KNL_64-NEXT: vmovaps %zmm2, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test6: +; KNL_32: # BB#0: +; KNL_32-NEXT: kxnorw %k1, %k1, %k1 +; KNL_32-NEXT: vpmovsxdq %ymm1, %zmm2 +; KNL_32-NEXT: kxnorw %k2, %k2, %k2 +; KNL_32-NEXT: vpgatherqd (,%zmm2), %ymm1 {%k2} +; KNL_32-NEXT: vpscatterqd %ymm0, (,%zmm2) {%k1} +; KNL_32-NEXT: vmovaps %zmm1, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test6: +; SKX: # BB#0: +; SKX-NEXT: kxnorw %k1, %k1, %k1 +; SKX-NEXT: kxnorw %k2, %k2, %k2 +; SKX-NEXT: vpgatherqd (,%zmm1), %ymm2 {%k2} +; SKX-NEXT: vpscatterqd %ymm0, (,%zmm1) {%k1} +; SKX-NEXT: vmovaps %zmm2, %zmm0 +; SKX-NEXT: retq %a = call <8 x i32> @llvm.masked.gather.v8i32(<8 x i32*> %ptr, i32 4, <8 x i1> , <8 x i32> undef) @@ -150,13 +273,41 @@ ret <8 x i32>%a } -; In this case the index should be promoted to <8 x i64> for KNL -; KNL-LABEL: test7 -; KNL: vpmovsxdq %ymm0, %zmm0 -; KNL: kmovw %k1, %k2 -; KNL: vpgatherqd {{.*}} {%k2} -; KNL: vpgatherqd {{.*}} {%k1} define <8 x i32> @test7(i32* %base, <8 x i32> %ind, i8 %mask) { +; +; KNL_64-LABEL: test7: +; KNL_64: # BB#0: +; KNL_64-NEXT: movzbl %sil, %eax +; KNL_64-NEXT: kmovw %eax, %k1 +; KNL_64-NEXT: vpmovsxdq %ymm0, %zmm0 +; KNL_64-NEXT: kmovw %k1, %k2 +; KNL_64-NEXT: vpgatherqd (%rdi,%zmm0,4), %ymm1 {%k2} +; KNL_64-NEXT: vmovaps %zmm1, %zmm2 +; KNL_64-NEXT: vpgatherqd (%rdi,%zmm0,4), %ymm2 {%k1} +; KNL_64-NEXT: vpaddd %ymm2, %ymm1, %ymm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test7: +; KNL_32: # BB#0: +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: kmovw {{[0-9]+}}(%esp), %k1 +; KNL_32-NEXT: vpmovsxdq %ymm0, %zmm0 +; KNL_32-NEXT: kmovw %k1, %k2 +; KNL_32-NEXT: vpgatherqd (%eax,%zmm0,4), %ymm1 {%k2} +; KNL_32-NEXT: vmovaps %zmm1, %zmm2 +; KNL_32-NEXT: vpgatherqd (%eax,%zmm0,4), %ymm2 {%k1} +; KNL_32-NEXT: vpaddd %ymm2, %ymm1, %ymm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test7: +; SKX: # BB#0: +; SKX-NEXT: kmovb %esi, %k1 +; SKX-NEXT: kmovw %k1, %k2 +; SKX-NEXT: vpgatherdd (%rdi,%ymm0,4), %ymm1 {%k2} +; SKX-NEXT: vmovaps %zmm1, %zmm2 +; SKX-NEXT: vpgatherdd (%rdi,%ymm0,4), %ymm2 {%k1} +; SKX-NEXT: vpaddd %ymm2, %ymm1, %ymm0 +; SKX-NEXT: retq %broadcast.splatinsert = insertelement <8 x i32*> undef, i32* %base, i32 0 %broadcast.splat = shufflevector <8 x i32*> %broadcast.splatinsert, <8 x i32*> undef, <8 x i32> zeroinitializer @@ -171,15 +322,47 @@ ; No uniform base in this case, index <8 x i64> contains addresses, ; each gather call will be split into two -; KNL-LABEL: test8 -; KNL: kshiftrw $8, %k1, %k2 -; KNL: vpgatherqd -; KNL: vpgatherqd -; KNL: vinserti64x4 -; KNL: vpgatherqd -; KNL: vpgatherqd -; KNL: vinserti64x4 define <16 x i32> @test8(<16 x i32*> %ptr.random, <16 x i32> %ind, i16 %mask) { +; KNL_64-LABEL: test8: +; KNL_64: # BB#0: +; KNL_64-NEXT: kmovw %edi, %k1 +; KNL_64-NEXT: kshiftrw $8, %k1, %k2 +; KNL_64-NEXT: kmovw %k2, %k3 +; KNL_64-NEXT: vpgatherqd (,%zmm1), %ymm2 {%k3} +; KNL_64-NEXT: kmovw %k1, %k3 +; KNL_64-NEXT: vpgatherqd (,%zmm0), %ymm3 {%k3} +; KNL_64-NEXT: vinserti64x4 $1, %ymm2, %zmm3, %zmm4 +; KNL_64-NEXT: vpgatherqd (,%zmm1), %ymm2 {%k2} +; KNL_64-NEXT: vpgatherqd (,%zmm0), %ymm3 {%k1} +; KNL_64-NEXT: vinserti64x4 $1, %ymm2, %zmm3, %zmm0 +; KNL_64-NEXT: vpaddd %zmm0, %zmm4, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test8: +; KNL_32: # BB#0: +; KNL_32-NEXT: kmovw {{[0-9]+}}(%esp), %k1 +; KNL_32-NEXT: kmovw %k1, %k2 +; KNL_32-NEXT: vpgatherdd (,%zmm0), %zmm1 {%k2} +; KNL_32-NEXT: vmovaps %zmm1, %zmm2 +; KNL_32-NEXT: vpgatherdd (,%zmm0), %zmm2 {%k1} +; KNL_32-NEXT: vpaddd %zmm2, %zmm1, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test8: +; SKX: # BB#0: +; SKX-NEXT: kmovw %edi, %k1 +; SKX-NEXT: kshiftrw $8, %k1, %k2 +; SKX-NEXT: kmovw %k2, %k3 +; SKX-NEXT: vpgatherqd (,%zmm1), %ymm2 {%k3} +; SKX-NEXT: kmovw %k1, %k3 +; SKX-NEXT: vpgatherqd (,%zmm0), %ymm3 {%k3} +; SKX-NEXT: vinserti32x8 $1, %ymm2, %zmm3, %zmm4 +; SKX-NEXT: vpgatherqd (,%zmm1), %ymm2 {%k2} +; SKX-NEXT: vpgatherqd (,%zmm0), %ymm3 {%k1} +; SKX-NEXT: vinserti32x8 $1, %ymm2, %zmm3, %zmm0 +; SKX-NEXT: vpaddd %zmm0, %zmm4, %zmm0 +; SKX-NEXT: retq + %imask = bitcast i16 %mask to <16 x i1> %gt1 = call <16 x i32> @llvm.masked.gather.v16i32(<16 x i32*> %ptr.random, i32 4, <16 x i1> %imask, <16 x i32>undef) %gt2 = call <16 x i32> @llvm.masked.gather.v16i32(<16 x i32*> %ptr.random, i32 4, <16 x i1> %imask, <16 x i32>%gt1) @@ -193,18 +376,60 @@ ; Masked gather for agregate types ; Test9 and Test10 should give the same result (scalar and vector indices in GEP) -; KNL-LABEL: test9 -; KNL: vpbroadcastq %rdi, %zmm -; KNL: vpmovsxdq -; KNL: vpbroadcastq -; KNL: vpmuludq -; KNL: vpaddq -; KNL: vpaddq -; KNL: vpaddq -; KNL: vpaddq -; KNL: vpgatherqd (,%zmm define <8 x i32> @test9(%struct.ST* %base, <8 x i64> %ind1, <8 x i32>%ind5) { +; KNL_64-LABEL: test9: +; KNL_64: # BB#0: # %entry +; KNL_64-NEXT: vpbroadcastq %rdi, %zmm2 +; KNL_64-NEXT: vpmovsxdq %ymm1, %zmm1 +; KNL_64-NEXT: vpbroadcastq {{.*}}(%rip), %zmm3 +; KNL_64-NEXT: vpmuludq %zmm3, %zmm1, %zmm4 +; KNL_64-NEXT: vpsrlq $32, %zmm1, %zmm1 +; KNL_64-NEXT: vpmuludq %zmm3, %zmm1, %zmm1 +; KNL_64-NEXT: vpsllq $32, %zmm1, %zmm1 +; KNL_64-NEXT: vpaddq %zmm1, %zmm4, %zmm1 +; KNL_64-NEXT: vpbroadcastq {{.*}}(%rip), %zmm3 +; KNL_64-NEXT: vpmuludq %zmm3, %zmm0, %zmm4 +; KNL_64-NEXT: vpsrlq $32, %zmm0, %zmm0 +; KNL_64-NEXT: vpmuludq %zmm3, %zmm0, %zmm0 +; KNL_64-NEXT: vpsllq $32, %zmm0, %zmm0 +; KNL_64-NEXT: vpaddq %zmm0, %zmm4, %zmm0 +; KNL_64-NEXT: vpaddq %zmm0, %zmm2, %zmm0 +; KNL_64-NEXT: vpaddq %zmm1, %zmm0, %zmm0 +; KNL_64-NEXT: vpaddq {{.*}}(%rip){1to8}, %zmm0, %zmm1 +; KNL_64-NEXT: kxnorw %k1, %k1, %k1 +; KNL_64-NEXT: vpgatherqd (,%zmm1), %ymm0 {%k1} +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test9: +; KNL_32: # BB#0: # %entry +; KNL_32-NEXT: vpbroadcastd {{[0-9]+}}(%esp), %ymm2 +; KNL_32-NEXT: vpbroadcastd .LCPI8_0, %ymm3 +; KNL_32-NEXT: vpmulld %ymm3, %ymm1, %ymm1 +; KNL_32-NEXT: vpmovqd %zmm0, %ymm0 +; KNL_32-NEXT: vpbroadcastd .LCPI8_1, %ymm3 +; KNL_32-NEXT: vpmulld %ymm3, %ymm0, %ymm0 +; KNL_32-NEXT: vpaddd %ymm0, %ymm2, %ymm0 +; KNL_32-NEXT: vpaddd %ymm1, %ymm0, %ymm0 +; KNL_32-NEXT: vpbroadcastd .LCPI8_2, %ymm1 +; KNL_32-NEXT: vpaddd %ymm1, %ymm0, %ymm0 +; KNL_32-NEXT: vpmovsxdq %ymm0, %zmm1 +; KNL_32-NEXT: kxnorw %k1, %k1, %k1 +; KNL_32-NEXT: vpgatherqd (,%zmm1), %ymm0 {%k1} +; KNL_32-NEXT: retl +; +; SKX-LABEL: test9: +; SKX: # BB#0: # %entry +; SKX-NEXT: vpbroadcastq %rdi, %zmm2 +; SKX-NEXT: vpmullq {{.*}}(%rip){1to8}, %zmm0, %zmm0 +; SKX-NEXT: vpaddq %zmm0, %zmm2, %zmm0 +; SKX-NEXT: vpmovsxdq %ymm1, %zmm1 +; SKX-NEXT: vpmullq {{.*}}(%rip){1to8}, %zmm1, %zmm1 +; SKX-NEXT: vpaddq %zmm1, %zmm0, %zmm0 +; SKX-NEXT: vpaddq {{.*}}(%rip){1to8}, %zmm0, %zmm1 +; SKX-NEXT: kxnorw %k1, %k1, %k1 +; SKX-NEXT: vpgatherqd (,%zmm1), %ymm0 {%k1} +; SKX-NEXT: retq entry: %broadcast.splatinsert = insertelement <8 x %struct.ST*> undef, %struct.ST* %base, i32 0 %broadcast.splat = shufflevector <8 x %struct.ST*> %broadcast.splatinsert, <8 x %struct.ST*> undef, <8 x i32> zeroinitializer @@ -214,17 +439,59 @@ ret <8 x i32> %res } -; KNL-LABEL: test10 -; KNL: vpbroadcastq %rdi, %zmm -; KNL: vpmovsxdq -; KNL: vpbroadcastq -; KNL: vpmuludq -; KNL: vpaddq -; KNL: vpaddq -; KNL: vpaddq -; KNL: vpaddq -; KNL: vpgatherqd (,%zmm define <8 x i32> @test10(%struct.ST* %base, <8 x i64> %i1, <8 x i32>%ind5) { +; KNL_64-LABEL: test10: +; KNL_64: # BB#0: # %entry +; KNL_64-NEXT: vpbroadcastq %rdi, %zmm2 +; KNL_64-NEXT: vpmovsxdq %ymm1, %zmm1 +; KNL_64-NEXT: vpbroadcastq {{.*}}(%rip), %zmm3 +; KNL_64-NEXT: vpmuludq %zmm3, %zmm1, %zmm4 +; KNL_64-NEXT: vpsrlq $32, %zmm1, %zmm1 +; KNL_64-NEXT: vpmuludq %zmm3, %zmm1, %zmm1 +; KNL_64-NEXT: vpsllq $32, %zmm1, %zmm1 +; KNL_64-NEXT: vpaddq %zmm1, %zmm4, %zmm1 +; KNL_64-NEXT: vpbroadcastq {{.*}}(%rip), %zmm3 +; KNL_64-NEXT: vpmuludq %zmm3, %zmm0, %zmm4 +; KNL_64-NEXT: vpsrlq $32, %zmm0, %zmm0 +; KNL_64-NEXT: vpmuludq %zmm3, %zmm0, %zmm0 +; KNL_64-NEXT: vpsllq $32, %zmm0, %zmm0 +; KNL_64-NEXT: vpaddq %zmm0, %zmm4, %zmm0 +; KNL_64-NEXT: vpaddq %zmm0, %zmm2, %zmm0 +; KNL_64-NEXT: vpaddq %zmm1, %zmm0, %zmm0 +; KNL_64-NEXT: vpaddq {{.*}}(%rip){1to8}, %zmm0, %zmm1 +; KNL_64-NEXT: kxnorw %k1, %k1, %k1 +; KNL_64-NEXT: vpgatherqd (,%zmm1), %ymm0 {%k1} +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test10: +; KNL_32: # BB#0: # %entry +; KNL_32-NEXT: vpbroadcastd {{[0-9]+}}(%esp), %ymm2 +; KNL_32-NEXT: vpbroadcastd .LCPI9_0, %ymm3 +; KNL_32-NEXT: vpmulld %ymm3, %ymm1, %ymm1 +; KNL_32-NEXT: vpmovqd %zmm0, %ymm0 +; KNL_32-NEXT: vpbroadcastd .LCPI9_1, %ymm3 +; KNL_32-NEXT: vpmulld %ymm3, %ymm0, %ymm0 +; KNL_32-NEXT: vpaddd %ymm0, %ymm2, %ymm0 +; KNL_32-NEXT: vpaddd %ymm1, %ymm0, %ymm0 +; KNL_32-NEXT: vpbroadcastd .LCPI9_2, %ymm1 +; KNL_32-NEXT: vpaddd %ymm1, %ymm0, %ymm0 +; KNL_32-NEXT: vpmovsxdq %ymm0, %zmm1 +; KNL_32-NEXT: kxnorw %k1, %k1, %k1 +; KNL_32-NEXT: vpgatherqd (,%zmm1), %ymm0 {%k1} +; KNL_32-NEXT: retl +; +; SKX-LABEL: test10: +; SKX: # BB#0: # %entry +; SKX-NEXT: vpbroadcastq %rdi, %zmm2 +; SKX-NEXT: vpmullq {{.*}}(%rip){1to8}, %zmm0, %zmm0 +; SKX-NEXT: vpaddq %zmm0, %zmm2, %zmm0 +; SKX-NEXT: vpmovsxdq %ymm1, %zmm1 +; SKX-NEXT: vpmullq {{.*}}(%rip){1to8}, %zmm1, %zmm1 +; SKX-NEXT: vpaddq %zmm1, %zmm0, %zmm0 +; SKX-NEXT: vpaddq {{.*}}(%rip){1to8}, %zmm0, %zmm1 +; SKX-NEXT: kxnorw %k1, %k1, %k1 +; SKX-NEXT: vpgatherqd (,%zmm1), %ymm0 {%k1} +; SKX-NEXT: retq entry: %broadcast.splatinsert = insertelement <8 x %struct.ST*> undef, %struct.ST* %base, i32 0 %broadcast.splat = shufflevector <8 x %struct.ST*> %broadcast.splatinsert, <8 x %struct.ST*> undef, <8 x i32> zeroinitializer @@ -235,10 +502,28 @@ } ; Splat index in GEP, requires broadcast -; KNL-LABEL: test11 -; KNL: vpbroadcastd %esi, %zmm -; KNL: vgatherdps (%rdi,%zmm define <16 x float> @test11(float* %base, i32 %ind) { +; KNL_64-LABEL: test11: +; KNL_64: # BB#0: +; KNL_64-NEXT: vpbroadcastd %esi, %zmm1 +; KNL_64-NEXT: kxnorw %k1, %k1, %k1 +; KNL_64-NEXT: vgatherdps (%rdi,%zmm1,4), %zmm0 {%k1} +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test11: +; KNL_32: # BB#0: +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: vpbroadcastd {{[0-9]+}}(%esp), %zmm1 +; KNL_32-NEXT: kxnorw %k1, %k1, %k1 +; KNL_32-NEXT: vgatherdps (%eax,%zmm1,4), %zmm0 {%k1} +; KNL_32-NEXT: retl +; +; SKX-LABEL: test11: +; SKX: # BB#0: +; SKX-NEXT: vpbroadcastd %esi, %zmm1 +; SKX-NEXT: kxnorw %k1, %k1, %k1 +; SKX-NEXT: vgatherdps (%rdi,%zmm1,4), %zmm0 {%k1} +; SKX-NEXT: retq %broadcast.splatinsert = insertelement <16 x float*> undef, float* %base, i32 0 %broadcast.splat = shufflevector <16 x float*> %broadcast.splatinsert, <16 x float*> undef, <16 x i32> zeroinitializer @@ -250,10 +535,28 @@ } ; We are checking the uniform base here. It is taken directly from input to vgatherdps -; KNL-LABEL: test12 -; KNL: kxnorw %k1, %k1, %k1 -; KNL: vgatherdps (%rdi,%zmm define <16 x float> @test12(float* %base, <16 x i32> %ind) { +; KNL_64-LABEL: test12: +; KNL_64: # BB#0: +; KNL_64-NEXT: kxnorw %k1, %k1, %k1 +; KNL_64-NEXT: vgatherdps (%rdi,%zmm0,4), %zmm1 {%k1} +; KNL_64-NEXT: vmovaps %zmm1, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test12: +; KNL_32: # BB#0: +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: kxnorw %k1, %k1, %k1 +; KNL_32-NEXT: vgatherdps (%eax,%zmm0,4), %zmm1 {%k1} +; KNL_32-NEXT: vmovaps %zmm1, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test12: +; SKX: # BB#0: +; SKX-NEXT: kxnorw %k1, %k1, %k1 +; SKX-NEXT: vgatherdps (%rdi,%zmm0,4), %zmm1 {%k1} +; SKX-NEXT: vmovaps %zmm1, %zmm0 +; SKX-NEXT: retq %sext_ind = sext <16 x i32> %ind to <16 x i64> %gep.random = getelementptr float, float *%base, <16 x i64> %sext_ind @@ -263,10 +566,25 @@ } ; The same as the previous, but the mask is undefined -; KNL-LABEL: test13 -; KNL-NOT: kxnorw -; KNL: vgatherdps (%rdi,%zmm define <16 x float> @test13(float* %base, <16 x i32> %ind) { +; KNL_64-LABEL: test13: +; KNL_64: # BB#0: +; KNL_64-NEXT: vgatherdps (%rdi,%zmm0,4), %zmm1 {%k1} +; KNL_64-NEXT: vmovaps %zmm1, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test13: +; KNL_32: # BB#0: +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: vgatherdps (%eax,%zmm0,4), %zmm1 {%k1} +; KNL_32-NEXT: vmovaps %zmm1, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test13: +; SKX: # BB#0: +; SKX-NEXT: vgatherdps (%rdi,%zmm0,4), %zmm1 {%k1} +; SKX-NEXT: vmovaps %zmm1, %zmm0 +; SKX-NEXT: retq %sext_ind = sext <16 x i32> %ind to <16 x i64> %gep.random = getelementptr float, float *%base, <16 x i64> %sext_ind @@ -276,10 +594,48 @@ } ; The base pointer is not splat, can't find unform base -; KNL-LABEL: test14 -; KNL: vgatherqps (,%zmm0) -; KNL: vgatherqps (,%zmm0) define <16 x float> @test14(float* %base, i32 %ind, <16 x float*> %vec) { +; KNL_64-LABEL: test14: +; KNL_64: # BB#0: +; KNL_64-NEXT: vpinsrq $1, %rdi, %xmm0, %xmm1 +; KNL_64-NEXT: vinserti32x4 $0, %xmm1, %zmm0, %zmm0 +; KNL_64-NEXT: vpbroadcastq %xmm0, %zmm0 +; KNL_64-NEXT: vmovd %esi, %xmm1 +; KNL_64-NEXT: vpbroadcastd %xmm1, %ymm1 +; KNL_64-NEXT: vpmovsxdq %ymm1, %zmm1 +; KNL_64-NEXT: vpsllq $2, %zmm1, %zmm1 +; KNL_64-NEXT: vpaddq %zmm1, %zmm0, %zmm0 +; KNL_64-NEXT: kshiftrw $8, %k0, %k1 +; KNL_64-NEXT: vgatherqps (,%zmm0), %ymm1 {%k1} +; KNL_64-NEXT: vgatherqps (,%zmm0), %ymm2 {%k1} +; KNL_64-NEXT: vinsertf64x4 $1, %ymm1, %zmm2, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test14: +; KNL_32: # BB#0: +; KNL_32-NEXT: vpinsrd $1, {{[0-9]+}}(%esp), %xmm0, %xmm1 +; KNL_32-NEXT: vinserti32x4 $0, %xmm1, %zmm0, %zmm0 +; KNL_32-NEXT: vpbroadcastd %xmm0, %zmm0 +; KNL_32-NEXT: vpslld $2, {{[0-9]+}}(%esp){1to16}, %zmm1 +; KNL_32-NEXT: vpaddd %zmm1, %zmm0, %zmm1 +; KNL_32-NEXT: vgatherdps (,%zmm1), %zmm0 {%k1} +; KNL_32-NEXT: retl +; +; SKX-LABEL: test14: +; SKX: # BB#0: +; SKX-NEXT: vpinsrq $1, %rdi, %xmm0, %xmm1 +; SKX-NEXT: vinserti64x2 $0, %xmm1, %zmm0, %zmm0 +; SKX-NEXT: vpbroadcastq %xmm0, %zmm0 +; SKX-NEXT: vmovd %esi, %xmm1 +; SKX-NEXT: vpbroadcastd %xmm1, %ymm1 +; SKX-NEXT: vpmovsxdq %ymm1, %zmm1 +; SKX-NEXT: vpsllq $2, %zmm1, %zmm1 +; SKX-NEXT: vpaddq %zmm1, %zmm0, %zmm0 +; SKX-NEXT: kshiftrw $8, %k0, %k1 +; SKX-NEXT: vgatherqps (,%zmm0), %ymm1 {%k1} +; SKX-NEXT: vgatherqps (,%zmm0), %ymm2 {%k1} +; SKX-NEXT: vinsertf32x8 $1, %ymm1, %zmm2, %zmm0 +; SKX-NEXT: retq %broadcast.splatinsert = insertelement <16 x float*> %vec, float* %base, i32 1 %broadcast.splat = shufflevector <16 x float*> %broadcast.splatinsert, <16 x float*> undef, <16 x i32> zeroinitializer @@ -290,19 +646,588 @@ ret <16 x float>%res } +declare <4 x float> @llvm.masked.gather.v4f32(<4 x float*>, i32, <4 x i1>, <4 x float>) +declare <4 x double> @llvm.masked.gather.v4f64(<4 x double*>, i32, <4 x i1>, <4 x double>) +declare <2 x double> @llvm.masked.gather.v2f64(<2 x double*>, i32, <2 x i1>, <2 x double>) + +; Gather smaller than existing instruction +define <4 x float> @test15(float* %base, <4 x i32> %ind, <4 x i1> %mask) { +; +; KNL_64-LABEL: test15: +; KNL_64: # BB#0: +; KNL_64-NEXT: vpxor %ymm2, %ymm2, %ymm2 +; KNL_64-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm2[4,5,6,7] +; KNL_64-NEXT: vpmovsxdq %ymm0, %zmm2 +; KNL_64-NEXT: vpmovsxdq %ymm1, %zmm0 +; KNL_64-NEXT: vpandq {{.*}}(%rip){1to8}, %zmm0, %zmm0 +; KNL_64-NEXT: vptestmq %zmm0, %zmm0, %k1 +; KNL_64-NEXT: vgatherqps (%rdi,%zmm2,4), %ymm0 {%k1} +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test15: +; KNL_32: # BB#0: +; KNL_32-NEXT: vpxor %ymm2, %ymm2, %ymm2 +; KNL_32-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm2[4,5,6,7] +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: vpmovsxdq %ymm0, %zmm2 +; KNL_32-NEXT: vpmovsxdq %ymm1, %zmm0 +; KNL_32-NEXT: vpandq .LCPI14_0, %zmm0, %zmm0 +; KNL_32-NEXT: vptestmq %zmm0, %zmm0, %k1 +; KNL_32-NEXT: vgatherqps (%eax,%zmm2,4), %ymm0 {%k1} +; KNL_32-NEXT: retl +; +; SKX-LABEL: test15: +; SKX: # BB#0: +; SKX-NEXT: vpmovd2m %xmm1, %k1 +; SKX-NEXT: vgatherdps (%rdi,%xmm0,4), %xmm1 {%k1} +; SKX-NEXT: vmovaps %zmm1, %zmm0 +; SKX-NEXT: retq + + %sext_ind = sext <4 x i32> %ind to <4 x i64> + %gep.random = getelementptr float, float* %base, <4 x i64> %sext_ind + %res = call <4 x float> @llvm.masked.gather.v4f32(<4 x float*> %gep.random, i32 4, <4 x i1> %mask, <4 x float> undef) + ret <4 x float>%res +} + +; Gather smaller than existing instruction +define <4 x double> @test16(double* %base, <4 x i32> %ind, <4 x i1> %mask, <4 x double> %src0) { +; +; KNL_64-LABEL: test16: +; KNL_64: # BB#0: +; KNL_64-NEXT: vpslld $31, %xmm1, %xmm1 +; KNL_64-NEXT: vpsrad $31, %xmm1, %xmm1 +; KNL_64-NEXT: vpmovsxdq %xmm1, %ymm1 +; KNL_64-NEXT: vpxord %zmm3, %zmm3, %zmm3 +; KNL_64-NEXT: vinserti64x4 $0, %ymm1, %zmm3, %zmm1 +; KNL_64-NEXT: vpmovsxdq %ymm0, %zmm0 +; KNL_64-NEXT: vpandq {{.*}}(%rip){1to8}, %zmm1, %zmm1 +; KNL_64-NEXT: vptestmq %zmm1, %zmm1, %k1 +; KNL_64-NEXT: vgatherqpd (%rdi,%zmm0,8), %zmm2 {%k1} +; KNL_64-NEXT: vmovaps %zmm2, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test16: +; KNL_32: # BB#0: +; KNL_32-NEXT: vpslld $31, %xmm1, %xmm1 +; KNL_32-NEXT: vpsrad $31, %xmm1, %xmm1 +; KNL_32-NEXT: vpmovsxdq %xmm1, %ymm1 +; KNL_32-NEXT: vpxord %zmm3, %zmm3, %zmm3 +; KNL_32-NEXT: vinserti64x4 $0, %ymm1, %zmm3, %zmm1 +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: vpmovsxdq %ymm0, %zmm0 +; KNL_32-NEXT: vpandq .LCPI15_0, %zmm1, %zmm1 +; KNL_32-NEXT: vptestmq %zmm1, %zmm1, %k1 +; KNL_32-NEXT: vgatherqpd (%eax,%zmm0,8), %zmm2 {%k1} +; KNL_32-NEXT: vmovaps %zmm2, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test16: +; SKX: # BB#0: +; SKX-NEXT: vpmovd2m %xmm1, %k1 +; SKX-NEXT: vgatherdpd (%rdi,%xmm0,8), %ymm2 {%k1} +; SKX-NEXT: vmovaps %zmm2, %zmm0 +; SKX-NEXT: retq + + %sext_ind = sext <4 x i32> %ind to <4 x i64> + %gep.random = getelementptr double, double* %base, <4 x i64> %sext_ind + %res = call <4 x double> @llvm.masked.gather.v4f64(<4 x double*> %gep.random, i32 4, <4 x i1> %mask, <4 x double> %src0) + ret <4 x double>%res +} + +define <2 x double> @test17(double* %base, <2 x i32> %ind, <2 x i1> %mask, <2 x double> %src0) { +; +; KNL_64-LABEL: test17: +; KNL_64: # BB#0: +; KNL_64-NEXT: vpxord %zmm3, %zmm3, %zmm3 +; KNL_64-NEXT: vinserti32x4 $0, %xmm1, %zmm3, %zmm1 +; KNL_64-NEXT: vpandq {{.*}}(%rip){1to8}, %zmm1, %zmm1 +; KNL_64-NEXT: vptestmq %zmm1, %zmm1, %k1 +; KNL_64-NEXT: vgatherqpd (%rdi,%zmm0,8), %zmm2 {%k1} +; KNL_64-NEXT: vmovaps %zmm2, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test17: +; KNL_32: # BB#0: +; KNL_32-NEXT: vpxord %zmm3, %zmm3, %zmm3 +; KNL_32-NEXT: vinserti32x4 $0, %xmm1, %zmm3, %zmm1 +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: vpandq .LCPI16_0, %zmm1, %zmm1 +; KNL_32-NEXT: vptestmq %zmm1, %zmm1, %k1 +; KNL_32-NEXT: vgatherqpd (%eax,%zmm0,8), %zmm2 {%k1} +; KNL_32-NEXT: vmovaps %zmm2, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test17: +; SKX: # BB#0: +; SKX-NEXT: vpmovq2m %xmm1, %k1 +; SKX-NEXT: vgatherqpd (%rdi,%xmm0,8), %xmm2 {%k1} +; SKX-NEXT: vmovaps %zmm2, %zmm0 +; SKX-NEXT: retq + + %sext_ind = sext <2 x i32> %ind to <2 x i64> + %gep.random = getelementptr double, double* %base, <2 x i64> %sext_ind + %res = call <2 x double> @llvm.masked.gather.v2f64(<2 x double*> %gep.random, i32 4, <2 x i1> %mask, <2 x double> %src0) + ret <2 x double>%res +} + +declare void @llvm.masked.scatter.v4i32(<4 x i32> , <4 x i32*> , i32 , <4 x i1> ) +declare void @llvm.masked.scatter.v4f64(<4 x double> , <4 x double*> , i32 , <4 x i1> ) +declare void @llvm.masked.scatter.v2i64(<2 x i64> , <2 x i64*> , i32 , <2 x i1> ) +declare void @llvm.masked.scatter.v2i32(<2 x i32> , <2 x i32*> , i32 , <2 x i1> ) +declare void @llvm.masked.scatter.v2f32(<2 x float> , <2 x float*> , i32 , <2 x i1> ) + +define void @test18(<4 x i32>%a1, <4 x i32*> %ptr, <4 x i1>%mask) { +; +; KNL_64-LABEL: test18: +; KNL_64: # BB#0: +; KNL_64-NEXT: vpxor %ymm3, %ymm3, %ymm3 +; KNL_64-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm3[4,5,6,7] +; KNL_64-NEXT: vpmovsxdq %ymm2, %zmm2 +; KNL_64-NEXT: vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2 +; KNL_64-NEXT: vptestmq %zmm2, %zmm2, %k1 +; KNL_64-NEXT: vpscatterqd %ymm0, (,%zmm1) {%k1} +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test18: +; KNL_32: # BB#0: +; KNL_32-NEXT: vpxor %ymm3, %ymm3, %ymm3 +; KNL_32-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm3[4,5,6,7] +; KNL_32-NEXT: vpmovsxdq %ymm1, %zmm1 +; KNL_32-NEXT: vpmovsxdq %ymm2, %zmm2 +; KNL_32-NEXT: vpandq .LCPI17_0, %zmm2, %zmm2 +; KNL_32-NEXT: vptestmq %zmm2, %zmm2, %k1 +; KNL_32-NEXT: vpscatterqd %ymm0, (,%zmm1) {%k1} +; KNL_32-NEXT: retl +; +; SKX-LABEL: test18: +; SKX: # BB#0: +; SKX-NEXT: vpmovd2m %xmm2, %k1 +; SKX-NEXT: vpscatterqd %xmm0, (,%ymm1) {%k1} +; SKX-NEXT: retq + call void @llvm.masked.scatter.v4i32(<4 x i32> %a1, <4 x i32*> %ptr, i32 4, <4 x i1> %mask) + ret void +} + +define void @test19(<4 x double>%a1, double* %ptr, <4 x i1>%mask, <4 x i64> %ind) { +; +; KNL_64-LABEL: test19: +; KNL_64: # BB#0: +; KNL_64-NEXT: vpslld $31, %xmm1, %xmm1 +; KNL_64-NEXT: vpsrad $31, %xmm1, %xmm1 +; KNL_64-NEXT: vpmovsxdq %xmm1, %ymm1 +; KNL_64-NEXT: vpxord %zmm3, %zmm3, %zmm3 +; KNL_64-NEXT: vinserti64x4 $0, %ymm1, %zmm3, %zmm1 +; KNL_64-NEXT: vpandq {{.*}}(%rip){1to8}, %zmm1, %zmm1 +; KNL_64-NEXT: vptestmq %zmm1, %zmm1, %k1 +; KNL_64-NEXT: vscatterqpd %zmm0, (%rdi,%zmm2,8) {%k1} +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test19: +; KNL_32: # BB#0: +; KNL_32-NEXT: vpslld $31, %xmm1, %xmm1 +; KNL_32-NEXT: vpsrad $31, %xmm1, %xmm1 +; KNL_32-NEXT: vpmovsxdq %xmm1, %ymm1 +; KNL_32-NEXT: vpxord %zmm3, %zmm3, %zmm3 +; KNL_32-NEXT: vinserti64x4 $0, %ymm1, %zmm3, %zmm1 +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: vpandq .LCPI18_0, %zmm1, %zmm1 +; KNL_32-NEXT: vptestmq %zmm1, %zmm1, %k1 +; KNL_32-NEXT: vscatterqpd %zmm0, (%eax,%zmm2,8) {%k1} +; KNL_32-NEXT: retl +; +; SKX-LABEL: test19: +; SKX: # BB#0: +; SKX-NEXT: vpmovd2m %xmm1, %k1 +; SKX-NEXT: vscatterqpd %ymm0, (%rdi,%ymm2,8) {%k1} +; SKX-NEXT: retq + %gep = getelementptr double, double* %ptr, <4 x i64> %ind + call void @llvm.masked.scatter.v4f64(<4 x double> %a1, <4 x double*> %gep, i32 8, <4 x i1> %mask) + ret void +} + +; Data type requires widening +define void @test20(<2 x float>%a1, <2 x float*> %ptr, <2 x i1> %mask) { +; +; KNL_64-LABEL: test20: +; KNL_64: # BB#0: +; KNL_64-NEXT: vpxor %xmm3, %xmm3, %xmm3 +; KNL_64-NEXT: vinserti128 $1, %xmm3, %ymm2, %ymm2 +; KNL_64-NEXT: vpmovqd %zmm2, %ymm2 +; KNL_64-NEXT: vpxor %ymm3, %ymm3, %ymm3 +; KNL_64-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm3[4,5,6,7] +; KNL_64-NEXT: vpmovsxdq %ymm2, %zmm2 +; KNL_64-NEXT: vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2 +; KNL_64-NEXT: vptestmq %zmm2, %zmm2, %k1 +; KNL_64-NEXT: vscatterqps %ymm0, (,%zmm1) {%k1} +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test20: +; KNL_32: # BB#0: +; KNL_32-NEXT: vpxor %xmm3, %xmm3, %xmm3 +; KNL_32-NEXT: vinserti128 $1, %xmm3, %ymm2, %ymm2 +; KNL_32-NEXT: vpmovqd %zmm2, %ymm2 +; KNL_32-NEXT: vpxor %ymm3, %ymm3, %ymm3 +; KNL_32-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm3[4,5,6,7] +; KNL_32-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[0,2,2,3] +; KNL_32-NEXT: vpmovsxdq %ymm1, %zmm1 +; KNL_32-NEXT: vpmovsxdq %ymm2, %zmm2 +; KNL_32-NEXT: vpandq .LCPI19_0, %zmm2, %zmm2 +; KNL_32-NEXT: vptestmq %zmm2, %zmm2, %k1 +; KNL_32-NEXT: vscatterqps %ymm0, (,%zmm1) {%k1} +; KNL_32-NEXT: retl +; +; SKX-LABEL: test20: +; SKX: # BB#0: +; SKX-NEXT: vpmovq2m %xmm2, %k0 +; SKX-NEXT: kshiftlw $2, %k0, %k0 +; SKX-NEXT: kshiftrw $2, %k0, %k1 +; SKX-NEXT: vscatterqps %xmm0, (,%ymm1) {%k1} +; SKX-NEXT: retq + call void @llvm.masked.scatter.v2f32(<2 x float> %a1, <2 x float*> %ptr, i32 4, <2 x i1> %mask) + ret void +} + +; Data type requires promotion +define void @test21(<2 x i32>%a1, <2 x i32*> %ptr, <2 x i1>%mask) { +; +; KNL_64-LABEL: test21: +; KNL_64: # BB#0: +; KNL_64-NEXT: vpxord %zmm3, %zmm3, %zmm3 +; KNL_64-NEXT: vinserti32x4 $0, %xmm2, %zmm3, %zmm2 +; KNL_64-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] +; KNL_64-NEXT: vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2 +; KNL_64-NEXT: vptestmq %zmm2, %zmm2, %k1 +; KNL_64-NEXT: vpscatterqd %ymm0, (,%zmm1) {%k1} +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test21: +; KNL_32: # BB#0: +; KNL_32-NEXT: vpxord %zmm3, %zmm3, %zmm3 +; KNL_32-NEXT: vinserti32x4 $0, %xmm2, %zmm3, %zmm2 +; KNL_32-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] +; KNL_32-NEXT: vpandq .LCPI20_0, %zmm2, %zmm2 +; KNL_32-NEXT: vptestmq %zmm2, %zmm2, %k1 +; KNL_32-NEXT: vpscatterqd %ymm0, (,%zmm1) {%k1} +; KNL_32-NEXT: retl +; +; SKX-LABEL: test21: +; SKX: # BB#0: +; SKX-NEXT: vpmovq2m %xmm2, %k0 +; SKX-NEXT: kshiftlw $2, %k0, %k0 +; SKX-NEXT: kshiftrw $2, %k0, %k1 +; SKX-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] +; SKX-NEXT: vpscatterqd %xmm0, (,%ymm1) {%k1} +; SKX-NEXT: retq + call void @llvm.masked.scatter.v2i32(<2 x i32> %a1, <2 x i32*> %ptr, i32 4, <2 x i1> %mask) + ret void +} + +; The result type requires widening +declare <2 x float> @llvm.masked.gather.v2f32(<2 x float*>, i32, <2 x i1>, <2 x float>) + +define <2 x float> @test22(float* %base, <2 x i32> %ind, <2 x i1> %mask, <2 x float> %src0) { +; +; +; KNL_64-LABEL: test22: +; KNL_64: # BB#0: +; KNL_64-NEXT: vpxor %xmm3, %xmm3, %xmm3 +; KNL_64-NEXT: vinserti128 $1, %xmm3, %ymm1, %ymm1 +; KNL_64-NEXT: vpmovqd %zmm1, %ymm1 +; KNL_64-NEXT: vpxor %ymm3, %ymm3, %ymm3 +; KNL_64-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm3[4,5,6,7] +; KNL_64-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] +; KNL_64-NEXT: vpmovsxdq %ymm0, %zmm0 +; KNL_64-NEXT: vpmovsxdq %ymm1, %zmm1 +; KNL_64-NEXT: vpandq {{.*}}(%rip){1to8}, %zmm1, %zmm1 +; KNL_64-NEXT: vptestmq %zmm1, %zmm1, %k1 +; KNL_64-NEXT: vgatherqps (%rdi,%zmm0,4), %ymm2 {%k1} +; KNL_64-NEXT: vmovaps %zmm2, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test22: +; KNL_32: # BB#0: +; KNL_32-NEXT: vpxor %xmm3, %xmm3, %xmm3 +; KNL_32-NEXT: vinserti128 $1, %xmm3, %ymm1, %ymm1 +; KNL_32-NEXT: vpmovqd %zmm1, %ymm1 +; KNL_32-NEXT: vpxor %ymm3, %ymm3, %ymm3 +; KNL_32-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm3[4,5,6,7] +; KNL_32-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: vpmovsxdq %ymm0, %zmm0 +; KNL_32-NEXT: vpmovsxdq %ymm1, %zmm1 +; KNL_32-NEXT: vpandq .LCPI21_0, %zmm1, %zmm1 +; KNL_32-NEXT: vptestmq %zmm1, %zmm1, %k1 +; KNL_32-NEXT: vgatherqps (%eax,%zmm0,4), %ymm2 {%k1} +; KNL_32-NEXT: vmovaps %zmm2, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test22: +; SKX: # BB#0: +; SKX-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] +; SKX-NEXT: vpmovq2m %xmm1, %k0 +; SKX-NEXT: kshiftlw $2, %k0, %k0 +; SKX-NEXT: kshiftrw $2, %k0, %k1 +; SKX-NEXT: vgatherdps (%rdi,%xmm0,4), %xmm2 {%k1} +; SKX-NEXT: vmovaps %zmm2, %zmm0 +; SKX-NEXT: retq + %sext_ind = sext <2 x i32> %ind to <2 x i64> + %gep.random = getelementptr float, float* %base, <2 x i64> %sext_ind + %res = call <2 x float> @llvm.masked.gather.v2f32(<2 x float*> %gep.random, i32 4, <2 x i1> %mask, <2 x float> %src0) + ret <2 x float>%res +} + +declare <2 x i32> @llvm.masked.gather.v2i32(<2 x i32*>, i32, <2 x i1>, <2 x i32>) +declare <2 x i64> @llvm.masked.gather.v2i64(<2 x i64*>, i32, <2 x i1>, <2 x i64>) -; KNL-LABEL: test15 -; KNL: kmovw %eax, %k1 -; KNL: vgatherdps (%rdi,%zmm0,4), %zmm1 {%k1} +define <2 x i32> @test23(i32* %base, <2 x i32> %ind, <2 x i1> %mask, <2 x i32> %src0) { +; +; KNL_64-LABEL: test23: +; KNL_64: # BB#0: +; KNL_64-NEXT: vpxord %zmm3, %zmm3, %zmm3 +; KNL_64-NEXT: vinserti32x4 $0, %xmm1, %zmm3, %zmm1 +; KNL_64-NEXT: vpandq {{.*}}(%rip){1to8}, %zmm1, %zmm1 +; KNL_64-NEXT: vptestmq %zmm1, %zmm1, %k1 +; KNL_64-NEXT: vpgatherqq (%rdi,%zmm0,8), %zmm2 {%k1} +; KNL_64-NEXT: vmovaps %zmm2, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test23: +; KNL_32: # BB#0: +; KNL_32-NEXT: vpxord %zmm3, %zmm3, %zmm3 +; KNL_32-NEXT: vinserti32x4 $0, %xmm1, %zmm3, %zmm1 +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: vpandq .LCPI22_0, %zmm1, %zmm1 +; KNL_32-NEXT: vptestmq %zmm1, %zmm1, %k1 +; KNL_32-NEXT: vpgatherqq (%eax,%zmm0,8), %zmm2 {%k1} +; KNL_32-NEXT: vmovaps %zmm2, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test23: +; SKX: # BB#0: +; SKX-NEXT: vpmovq2m %xmm1, %k1 +; SKX-NEXT: vpgatherqq (%rdi,%xmm0,8), %xmm2 {%k1} +; SKX-NEXT: vmovaps %zmm2, %zmm0 +; SKX-NEXT: retq + %sext_ind = sext <2 x i32> %ind to <2 x i64> + %gep.random = getelementptr i32, i32* %base, <2 x i64> %sext_ind + %res = call <2 x i32> @llvm.masked.gather.v2i32(<2 x i32*> %gep.random, i32 4, <2 x i1> %mask, <2 x i32> %src0) + ret <2 x i32>%res +} + +define <2 x i32> @test24(i32* %base, <2 x i32> %ind) { +; +; +; KNL_64-LABEL: test24: +; KNL_64: # BB#0: +; KNL_64-NEXT: movb $3, %al +; KNL_64-NEXT: movzbl %al, %eax +; KNL_64-NEXT: kmovw %eax, %k1 +; KNL_64-NEXT: vpgatherqq (%rdi,%zmm0,8), %zmm1 {%k1} +; KNL_64-NEXT: vmovaps %zmm1, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test24: +; KNL_32: # BB#0: +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: vpxord %zmm1, %zmm1, %zmm1 +; KNL_32-NEXT: vinserti32x4 $0, .LCPI23_0, %zmm1, %zmm1 +; KNL_32-NEXT: vpandq .LCPI23_1, %zmm1, %zmm1 +; KNL_32-NEXT: vptestmq %zmm1, %zmm1, %k1 +; KNL_32-NEXT: vpgatherqq (%eax,%zmm0,8), %zmm1 {%k1} +; KNL_32-NEXT: vmovaps %zmm1, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test24: +; SKX: # BB#0: +; SKX-NEXT: kxnorw %k1, %k1, %k1 +; SKX-NEXT: vpgatherqq (%rdi,%xmm0,8), %xmm1 {%k1} +; SKX-NEXT: vmovaps %zmm1, %zmm0 +; SKX-NEXT: retq + %sext_ind = sext <2 x i32> %ind to <2 x i64> + %gep.random = getelementptr i32, i32* %base, <2 x i64> %sext_ind + %res = call <2 x i32> @llvm.masked.gather.v2i32(<2 x i32*> %gep.random, i32 4, <2 x i1> , <2 x i32> undef) + ret <2 x i32>%res +} + +define <2 x i64> @test25(i64* %base, <2 x i32> %ind, <2 x i1> %mask, <2 x i64> %src0) { +; +; KNL_64-LABEL: test25: +; KNL_64: # BB#0: +; KNL_64-NEXT: vpxord %zmm3, %zmm3, %zmm3 +; KNL_64-NEXT: vinserti32x4 $0, %xmm1, %zmm3, %zmm1 +; KNL_64-NEXT: vpandq {{.*}}(%rip){1to8}, %zmm1, %zmm1 +; KNL_64-NEXT: vptestmq %zmm1, %zmm1, %k1 +; KNL_64-NEXT: vpgatherqq (%rdi,%zmm0,8), %zmm2 {%k1} +; KNL_64-NEXT: vmovaps %zmm2, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test25: +; KNL_32: # BB#0: +; KNL_32-NEXT: vpxord %zmm3, %zmm3, %zmm3 +; KNL_32-NEXT: vinserti32x4 $0, %xmm1, %zmm3, %zmm1 +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: vpandq .LCPI24_0, %zmm1, %zmm1 +; KNL_32-NEXT: vptestmq %zmm1, %zmm1, %k1 +; KNL_32-NEXT: vpgatherqq (%eax,%zmm0,8), %zmm2 {%k1} +; KNL_32-NEXT: vmovaps %zmm2, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test25: +; SKX: # BB#0: +; SKX-NEXT: vpmovq2m %xmm1, %k1 +; SKX-NEXT: vpgatherqq (%rdi,%xmm0,8), %xmm2 {%k1} +; SKX-NEXT: vmovaps %zmm2, %zmm0 +; SKX-NEXT: retq + %sext_ind = sext <2 x i32> %ind to <2 x i64> + %gep.random = getelementptr i64, i64* %base, <2 x i64> %sext_ind + %res = call <2 x i64> @llvm.masked.gather.v2i64(<2 x i64*> %gep.random, i32 8, <2 x i1> %mask, <2 x i64> %src0) + ret <2 x i64>%res +} -; SCALAR-LABEL: test15 -; SCALAR: extractelement <16 x float*> +define <2 x i64> @test26(i64* %base, <2 x i32> %ind, <2 x i64> %src0) { +; +; KNL_64-LABEL: test26: +; KNL_64: # BB#0: +; KNL_64-NEXT: movb $3, %al +; KNL_64-NEXT: movzbl %al, %eax +; KNL_64-NEXT: kmovw %eax, %k1 +; KNL_64-NEXT: vpgatherqq (%rdi,%zmm0,8), %zmm1 {%k1} +; KNL_64-NEXT: vmovaps %zmm1, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test26: +; KNL_32: # BB#0: +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: vpxord %zmm2, %zmm2, %zmm2 +; KNL_32-NEXT: vinserti32x4 $0, .LCPI25_0, %zmm2, %zmm2 +; KNL_32-NEXT: vpandq .LCPI25_1, %zmm2, %zmm2 +; KNL_32-NEXT: vptestmq %zmm2, %zmm2, %k1 +; KNL_32-NEXT: vpgatherqq (%eax,%zmm0,8), %zmm1 {%k1} +; KNL_32-NEXT: vmovaps %zmm1, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test26: +; SKX: # BB#0: +; SKX-NEXT: kxnorw %k1, %k1, %k1 +; SKX-NEXT: vpgatherqq (%rdi,%xmm0,8), %xmm1 {%k1} +; SKX-NEXT: vmovaps %zmm1, %zmm0 +; SKX-NEXT: retq + %sext_ind = sext <2 x i32> %ind to <2 x i64> + %gep.random = getelementptr i64, i64* %base, <2 x i64> %sext_ind + %res = call <2 x i64> @llvm.masked.gather.v2i64(<2 x i64*> %gep.random, i32 8, <2 x i1> , <2 x i64> %src0) + ret <2 x i64>%res +} + +; Result type requires widening; all-ones mask +define <2 x float> @test27(float* %base, <2 x i32> %ind) { +; +; KNL_64-LABEL: test27: +; KNL_64: # BB#0: +; KNL_64-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] +; KNL_64-NEXT: vpmovsxdq %ymm0, %zmm1 +; KNL_64-NEXT: movb $3, %al +; KNL_64-NEXT: movzbl %al, %eax +; KNL_64-NEXT: kmovw %eax, %k1 +; KNL_64-NEXT: vgatherqps (%rdi,%zmm1,4), %ymm0 {%k1} +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test27: +; KNL_32: # BB#0: +; KNL_32-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: vpmovsxdq %ymm0, %zmm1 +; KNL_32-NEXT: vmovdqa {{.*#+}} xmm0 = [1,0,1,0] +; KNL_32-NEXT: vpxor %xmm2, %xmm2, %xmm2 +; KNL_32-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm0 +; KNL_32-NEXT: vpmovqd %zmm0, %ymm0 +; KNL_32-NEXT: vpxor %ymm2, %ymm2, %ymm2 +; KNL_32-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm2[4,5,6,7] +; KNL_32-NEXT: vpmovsxdq %ymm0, %zmm0 +; KNL_32-NEXT: vpandq .LCPI26_1, %zmm0, %zmm0 +; KNL_32-NEXT: vptestmq %zmm0, %zmm0, %k1 +; KNL_32-NEXT: vgatherqps (%eax,%zmm1,4), %ymm0 {%k1} +; KNL_32-NEXT: retl +; +; SKX-LABEL: test27: +; SKX: # BB#0: +; SKX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[0,2,2,3] +; SKX-NEXT: movb $3, %al +; SKX-NEXT: kmovb %eax, %k1 +; SKX-NEXT: vgatherdps (%rdi,%xmm1,4), %xmm0 {%k1} +; SKX-NEXT: retq + %sext_ind = sext <2 x i32> %ind to <2 x i64> + %gep.random = getelementptr float, float* %base, <2 x i64> %sext_ind + %res = call <2 x float> @llvm.masked.gather.v2f32(<2 x float*> %gep.random, i32 4, <2 x i1> , <2 x float> undef) + ret <2 x float>%res +} + +; Data type requires promotion, mask is all-ones +define void @test28(<2 x i32>%a1, <2 x i32*> %ptr) { +; +; +; KNL_64-LABEL: test28: +; KNL_64: # BB#0: +; KNL_64-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] +; KNL_64-NEXT: movb $3, %al +; KNL_64-NEXT: movzbl %al, %eax +; KNL_64-NEXT: kmovw %eax, %k1 +; KNL_64-NEXT: vpscatterqd %ymm0, (,%zmm1) {%k1} +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test28: +; KNL_32: # BB#0: +; KNL_32-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] +; KNL_32-NEXT: vpxord %zmm2, %zmm2, %zmm2 +; KNL_32-NEXT: vinserti32x4 $0, .LCPI27_0, %zmm2, %zmm2 +; KNL_32-NEXT: vpandq .LCPI27_1, %zmm2, %zmm2 +; KNL_32-NEXT: vptestmq %zmm2, %zmm2, %k1 +; KNL_32-NEXT: vpscatterqd %ymm0, (,%zmm1) {%k1} +; KNL_32-NEXT: retl +; +; SKX-LABEL: test28: +; SKX: # BB#0: +; SKX-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] +; SKX-NEXT: movb $3, %al +; SKX-NEXT: kmovb %eax, %k1 +; SKX-NEXT: vpscatterqd %xmm0, (,%ymm1) {%k1} +; SKX-NEXT: retq + call void @llvm.masked.scatter.v2i32(<2 x i32> %a1, <2 x i32*> %ptr, i32 4, <2 x i1> ) + ret void +} + + +; SCALAR-LABEL: test29 +; SCALAR: extractelement <16 x float*> ; SCALAR-NEXT: load float ; SCALAR-NEXT: insertelement <16 x float> ; SCALAR-NEXT: extractelement <16 x float*> ; SCALAR-NEXT: load float -define <16 x float> @test15(float* %base, <16 x i32> %ind) { +define <16 x float> @test29(float* %base, <16 x i32> %ind) { +; KNL_64-LABEL: test29: +; KNL_64: # BB#0: +; KNL_64-NEXT: movw $44, %ax +; KNL_64-NEXT: kmovw %eax, %k1 +; KNL_64-NEXT: vgatherdps (%rdi,%zmm0,4), %zmm1 {%k1} +; KNL_64-NEXT: vmovaps %zmm1, %zmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test29: +; KNL_32: # BB#0: +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: movw $44, %cx +; KNL_32-NEXT: kmovw %ecx, %k1 +; KNL_32-NEXT: vgatherdps (%eax,%zmm0,4), %zmm1 {%k1} +; KNL_32-NEXT: vmovaps %zmm1, %zmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test29: +; SKX: # BB#0: +; SKX-NEXT: movw $44, %ax +; SKX-NEXT: kmovw %eax, %k1 +; SKX-NEXT: vgatherdps (%rdi,%zmm0,4), %zmm1 {%k1} +; SKX-NEXT: vmovaps %zmm1, %zmm0 +; SKX-NEXT: retq %broadcast.splatinsert = insertelement <16 x float*> undef, float* %base, i32 0 %broadcast.splat = shufflevector <16 x float*> %broadcast.splatinsert, <16 x float*> undef, <16 x i32> zeroinitializer @@ -316,17 +1241,138 @@ ; Check non-power-of-2 case. It should be scalarized. declare <3 x i32> @llvm.masked.gather.v3i32(<3 x i32*>, i32, <3 x i1>, <3 x i32>) -; KNL-LABEL: test16 -; KNL: testb -; KNL: je -; KNL: testb -; KNL: je -; KNL: testb -; KNL: je -define <3 x i32> @test16(<3 x i32*> %base, <3 x i32> %ind, <3 x i1> %mask, <3 x i32> %src0) { +define <3 x i32> @test30(<3 x i32*> %base, <3 x i32> %ind, <3 x i1> %mask, <3 x +; KNL_64-LABEL: test30: +; KNL_64: # BB#0: +; KNL_64-NEXT: andl $1, %edx +; KNL_64-NEXT: kmovw %edx, %k1 +; KNL_64-NEXT: andl $1, %esi +; KNL_64-NEXT: kmovw %esi, %k2 +; KNL_64-NEXT: movl %edi, %eax +; KNL_64-NEXT: andl $1, %eax +; KNL_64-NEXT: kmovw %eax, %k0 +; KNL_64-NEXT: vpmovsxdq %xmm1, %ymm1 +; KNL_64-NEXT: vpsllq $2, %ymm1, %ymm1 +; KNL_64-NEXT: vpaddq %ymm1, %ymm0, %ymm1 +; KNL_64-NEXT: # implicit-def: %XMM0 +; KNL_64-NEXT: testb $1, %dil +; KNL_64-NEXT: je .LBB29_2 +; KNL_64-NEXT: # BB#1: # %cond.load +; KNL_64-NEXT: vmovq %xmm1, %rax +; KNL_64-NEXT: vmovd (%rax), %xmm0 +; KNL_64-NEXT: .LBB29_2: # %else +; KNL_64-NEXT: kmovw %k2, %eax +; KNL_64-NEXT: movl %eax, %ecx +; KNL_64-NEXT: andl $1, %ecx +; KNL_64-NEXT: testb %cl, %cl +; KNL_64-NEXT: je .LBB29_4 +; KNL_64-NEXT: # BB#3: # %cond.load1 +; KNL_64-NEXT: vpextrq $1, %xmm1, %rcx +; KNL_64-NEXT: vpinsrd $1, (%rcx), %xmm0, %xmm0 +; KNL_64-NEXT: .LBB29_4: # %else2 +; KNL_64-NEXT: kmovw %k1, %ecx +; KNL_64-NEXT: movl %ecx, %edx +; KNL_64-NEXT: andl $1, %edx +; KNL_64-NEXT: testb %dl, %dl +; KNL_64-NEXT: je .LBB29_6 +; KNL_64-NEXT: # BB#5: # %cond.load4 +; KNL_64-NEXT: vextracti128 $1, %ymm1, %xmm1 +; KNL_64-NEXT: vmovq %xmm1, %rdx +; KNL_64-NEXT: vpinsrd $2, (%rdx), %xmm0, %xmm0 +; KNL_64-NEXT: .LBB29_6: # %else5 +; KNL_64-NEXT: kmovw %k0, %edx +; KNL_64-NEXT: vmovd %edx, %xmm1 +; KNL_64-NEXT: vpinsrd $1, %eax, %xmm1, %xmm1 +; KNL_64-NEXT: vpinsrd $2, %ecx, %xmm1, %xmm1 +; KNL_64-NEXT: vpslld $31, %xmm1, %xmm1 +; KNL_64-NEXT: vblendvps %xmm1, %xmm0, %xmm2, %xmm0 +; KNL_64-NEXT: retq +; +; KNL_32-LABEL: test30: +; KNL_32: # BB#0: +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: andl $1, %eax +; KNL_32-NEXT: kmovw %eax, %k1 +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: andl $1, %eax +; KNL_32-NEXT: kmovw %eax, %k2 +; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax +; KNL_32-NEXT: movl %eax, %ecx +; KNL_32-NEXT: andl $1, %ecx +; KNL_32-NEXT: kmovw %ecx, %k0 +; KNL_32-NEXT: vpslld $2, %xmm1, %xmm1 +; KNL_32-NEXT: vpaddd %xmm1, %xmm0, %xmm1 +; KNL_32-NEXT: # implicit-def: %XMM0 +; KNL_32-NEXT: testb $1, %al +; KNL_32-NEXT: je .LBB29_2 +; KNL_32-NEXT: # BB#1: # %cond.load +; KNL_32-NEXT: vmovd %xmm1, %eax +; KNL_32-NEXT: vmovd (%eax), %xmm0 +; KNL_32-NEXT: .LBB29_2: # %else +; KNL_32-NEXT: kmovw %k2, %eax +; KNL_32-NEXT: movl %eax, %ecx +; KNL_32-NEXT: andl $1, %ecx +; KNL_32-NEXT: testb %cl, %cl +; KNL_32-NEXT: je .LBB29_4 +; KNL_32-NEXT: # BB#3: # %cond.load1 +; KNL_32-NEXT: vpextrd $1, %xmm1, %ecx +; KNL_32-NEXT: vpinsrd $1, (%ecx), %xmm0, %xmm0 +; KNL_32-NEXT: .LBB29_4: # %else2 +; KNL_32-NEXT: kmovw %k1, %ecx +; KNL_32-NEXT: movl %ecx, %edx +; KNL_32-NEXT: andl $1, %edx +; KNL_32-NEXT: testb %dl, %dl +; KNL_32-NEXT: je .LBB29_6 +; KNL_32-NEXT: # BB#5: # %cond.load4 +; KNL_32-NEXT: vpextrd $2, %xmm1, %edx +; KNL_32-NEXT: vpinsrd $2, (%edx), %xmm0, %xmm0 +; KNL_32-NEXT: .LBB29_6: # %else5 +; KNL_32-NEXT: kmovw %k0, %edx +; KNL_32-NEXT: vmovd %edx, %xmm1 +; KNL_32-NEXT: vpinsrd $1, %eax, %xmm1, %xmm1 +; KNL_32-NEXT: vpinsrd $2, %ecx, %xmm1, %xmm1 +; KNL_32-NEXT: vpslld $31, %xmm1, %xmm1 +; KNL_32-NEXT: vblendvps %xmm1, %xmm0, %xmm2, %xmm0 +; KNL_32-NEXT: retl +; +; SKX-LABEL: test30: +; SKX: # BB#0: +; SKX-NEXT: vpmovd2m %xmm2, %k1 +; SKX-NEXT: kmovb %k1, -{{[0-9]+}}(%rsp) +; SKX-NEXT: vpmovsxdq %xmm1, %ymm1 +; SKX-NEXT: vpsllq $2, %ymm1, %ymm1 +; SKX-NEXT: vpaddq %ymm1, %ymm0, %ymm1 +; SKX-NEXT: movb -{{[0-9]+}}(%rsp), %al +; SKX-NEXT: # implicit-def: %XMM0 +; SKX-NEXT: andb $1, %al +; SKX-NEXT: je .LBB29_2 +; SKX-NEXT: # BB#1: # %cond.load +; SKX-NEXT: vmovq %xmm1, %rax +; SKX-NEXT: vmovd (%rax), %xmm0 +; SKX-NEXT: .LBB29_2: # %else +; SKX-NEXT: kmovb %k1, -{{[0-9]+}}(%rsp) +; SKX-NEXT: movb -{{[0-9]+}}(%rsp), %al +; SKX-NEXT: andb $1, %al +; SKX-NEXT: je .LBB29_4 +; SKX-NEXT: # BB#3: # %cond.load1 +; SKX-NEXT: vpextrq $1, %xmm1, %rax +; SKX-NEXT: vpinsrd $1, (%rax), %xmm0, %xmm0 +; SKX-NEXT: .LBB29_4: # %else2 +; SKX-NEXT: kmovb %k1, -{{[0-9]+}}(%rsp) +; SKX-NEXT: movb -{{[0-9]+}}(%rsp), %al +; SKX-NEXT: andb $1, %al +; SKX-NEXT: je .LBB29_6 +; SKX-NEXT: # BB#5: # %cond.load4 +; SKX-NEXT: vextracti128 $1, %ymm1, %xmm1 +; SKX-NEXT: vmovq %xmm1, %rax +; SKX-NEXT: vpinsrd $2, (%rax), %xmm0, %xmm0 +; SKX-NEXT: .LBB29_6: # %else5 +; SKX-NEXT: vmovdqa32 %xmm0, %xmm3 {%k1} +; SKX-NEXT: vmovaps %zmm3, %zmm0 +; SKX-NEXT: retq +i32> %src0) { %sext_ind = sext <3 x i32> %ind to <3 x i64> %gep.random = getelementptr i32, <3 x i32*> %base, <3 x i64> %sext_ind %res = call <3 x i32> @llvm.masked.gather.v3i32(<3 x i32*> %gep.random, i32 4, <3 x i1> %mask, <3 x i32> %src0) ret <3 x i32>%res } -