Adding support for TargetLoweringBase::LibCall

[tyomitch]

Summary

TargetLoweringBase::Expand is defined as "Try to expand this to other ops,
otherwise use a libcall." For ISD::UDIV and ISD::SDIV, the choice between
the two possibilities was defined in a rather convoluted way:

• if DIVREM is legal, expand to DIVREM
• if DIVREM has a custom lowering, expand to DIVREM
• if DIVREM libcall is defined and a remainder from the same division is computed elsewhere, expand to a DIVREM libcall
• else, expand to a DIV libcall

This had the undesirable effect that if both DIV and DIVREM are implemented
as libcalls, then ISD::UDIV and ISD::SDIV are expanded to the heavier DIVREM
libcall, even when the remainder isn't used.

The new code adds a new LegalizeAction, TargetLoweringBase::LibCall, so that
backends can directly control whether they prefer an expansion or a conversion
to a libcall. This makes the generic lowering code even more generic,
allowing its reuse in a wider range of target-specific configurations.

The useful effect is that ARM backend will now generate a call
to aeabi_{i,u}div rather than aeabi_{i,u}divmod in cases where
it doesn't need the remainder. There's no functional change outside
the ARM backend.

[t.p.northover]

This looks reasonable to me.

Tim.

[rengolin]

Hi Artyom,

I like this idea a lot. It allows for more flexibility and explicitly states the hierarchy we want, Legal > Expand > Libcall.

LGTM, too. Thanks!

lib/CodeGen/SelectionDAG/LegalizeDAG.cpp

150

These two are intrinsically related in the lowering phase, and would be good to have some comments to that effect here, so people can understand without searching for the implementation details.

include/llvm/Target/TargetLowering.h

 
 public:
   /// This enum indicates whether operations are valid for a target, and if not,
   /// what action should be used to make them valid.
   enum LegalizeAction {
     Legal,      // The target natively supports this operation.
     Promote,    // This operation should be executed in a larger type.
     Expand,     // Try to expand this to other ops, otherwise use a libcall.
+    LibCall,    // Don't try to expand this to other ops, always use a libcall.
     Custom      // Use the LowerOperation hook to implement custom lowering.
   };
 
   /// This enum indicates whether a types are legal for a target, and if not,
   /// what action should be used to make them valid.
   enum LegalizeTypeAction {
     TypeLegal,           // The target natively supports this type.
     TypePromoteInteger,  // Replace this integer with a larger one.

lib/CodeGen/SelectionDAG/LegalizeDAG.cpp

private:
   SDValue ExpandBitCount(unsigned Opc, SDValue Op, SDLoc dl);
 
   SDValue ExpandExtractFromVectorThroughStack(SDValue Op);
   SDValue ExpandInsertToVectorThroughStack(SDValue Op);
   SDValue ExpandVectorBuildThroughStack(SDNode* Node);
 
   SDValue ExpandConstantFP(ConstantFPSDNode *CFP, bool UseCP);
 
-  std::pair<SDValue, SDValue> ExpandAtomic(SDNode *Node);
-
-  void ExpandNode(SDNode *Node);
+  bool ExpandNode(SDNode *Node);
+  void ConvertNodeToLibcall(SDNode *Node);
   void PromoteNode(SDNode *Node);
 
 public:
   // Node replacement helpers
   void ReplacedNode(SDNode *N) {
     LegalizedNodes.erase(N);
     if (UpdatedNodes)
       UpdatedNodes->insert(N);
case TargetLowering::Custom: {
         for (unsigned i = 0, e = Node->getNumValues(); i != e; ++i)
           ResultVals.push_back(Res.getValue(i));
         ReplaceNode(Node, ResultVals.data());
         return;
       }
     }
       // FALL THROUGH
     case TargetLowering::Expand:
-      ExpandNode(Node);
+      if (ExpandNode(Node))
+        return;
+      // FALL THROUGH
+    case TargetLowering::LibCall:
+      ConvertNodeToLibcall(Node);
       return;
     case TargetLowering::Promote:
       PromoteNode(Node);
       return;
     }
   }
 
   switch (Node->getOpcode()) {
if (!TLI.isOperationLegalOrCustom(ISD::CTPOP, VT) &&
       return DAG.getNode(ISD::SUB, dl, VT,
                          DAG.getConstant(VT.getSizeInBits(), dl, VT),
                          DAG.getNode(ISD::CTLZ, dl, VT, Tmp3));
     return DAG.getNode(ISD::CTPOP, dl, VT, Tmp3);
   }
   }
 }
 
-std::pair <SDValue, SDValue> SelectionDAGLegalize::ExpandAtomic(SDNode *Node) {
-  unsigned Opc = Node->getOpcode();
-  MVT VT = cast<AtomicSDNode>(Node)->getMemoryVT().getSimpleVT();
-  RTLIB::Libcall LC = RTLIB::getATOMIC(Opc, VT);
-  assert(LC != RTLIB::UNKNOWN_LIBCALL && "Unexpected atomic op or value type!");
-
-  return ExpandChainLibCall(LC, Node, false);
-}
-
-void SelectionDAGLegalize::ExpandNode(SDNode *Node) {
+bool SelectionDAGLegalize::ExpandNode(SDNode *Node) {
   SmallVector<SDValue, 8> Results;
   SDLoc dl(Node);
   SDValue Tmp1, Tmp2, Tmp3, Tmp4;
   bool NeedInvert;
   switch (Node->getOpcode()) {
   case ISD::CTPOP:
   case ISD::CTLZ:
   case ISD::CTLZ_ZERO_UNDEF:
case ISD::READCYCLECOUNTER:
     Results.push_back(Node->getOperand(0));
     break;
   case ISD::EH_SJLJ_SETJMP:
     // If the target didn't expand this, just return 'zero' and preserve the
     // chain.
     Results.push_back(DAG.getConstant(0, dl, MVT::i32));
     Results.push_back(Node->getOperand(0));
     break;
-  case ISD::ATOMIC_FENCE: {
-    // If the target didn't lower this, lower it to '__sync_synchronize()' call
-    // FIXME: handle "fence singlethread" more efficiently.
-    TargetLowering::ArgListTy Args;
-
-    TargetLowering::CallLoweringInfo CLI(DAG);
-    CLI.setDebugLoc(dl)
-        .setChain(Node->getOperand(0))
-        .setCallee(CallingConv::C, Type::getVoidTy(*DAG.getContext()),
-                   DAG.getExternalSymbol("__sync_synchronize",
-                                         TLI.getPointerTy(DAG.getDataLayout())),
-                   std::move(Args), 0);
-
-    std::pair<SDValue, SDValue> CallResult = TLI.LowerCallTo(CLI);
-
-    Results.push_back(CallResult.second);
-    break;
-  }
   case ISD::ATOMIC_LOAD: {
     // There is no libcall for atomic load; fake it with ATOMIC_CMP_SWAP.
     SDValue Zero = DAG.getConstant(0, dl, Node->getValueType(0));
     SDVTList VTs = DAG.getVTList(Node->getValueType(0), MVT::Other);
     SDValue Swap = DAG.getAtomicCmpSwap(
         ISD::ATOMIC_CMP_SWAP, dl, cast<AtomicSDNode>(Node)->getMemoryVT(), VTs,
         Node->getOperand(0), Node->getOperand(1), Zero, Zero,
         cast<AtomicSDNode>(Node)->getMemOperand(),
SDValue Swap = DAG.getAtomic(ISD::ATOMIC_SWAP, dl,
                                  Node->getOperand(0),
                                  Node->getOperand(1), Node->getOperand(2),
                                  cast<AtomicSDNode>(Node)->getMemOperand(),
                                  cast<AtomicSDNode>(Node)->getOrdering(),
                                  cast<AtomicSDNode>(Node)->getSynchScope());
     Results.push_back(Swap.getValue(1));
     break;
   }
-  // By default, atomic intrinsics are marked Legal and lowered. Targets
-  // which don't support them directly, however, may want libcalls, in which
-  // case they mark them Expand, and we get here.
-  case ISD::ATOMIC_SWAP:
-  case ISD::ATOMIC_LOAD_ADD:
-  case ISD::ATOMIC_LOAD_SUB:
-  case ISD::ATOMIC_LOAD_AND:
-  case ISD::ATOMIC_LOAD_OR:
-  case ISD::ATOMIC_LOAD_XOR:
-  case ISD::ATOMIC_LOAD_NAND:
-  case ISD::ATOMIC_LOAD_MIN:
-  case ISD::ATOMIC_LOAD_MAX:
-  case ISD::ATOMIC_LOAD_UMIN:
-  case ISD::ATOMIC_LOAD_UMAX:
-  case ISD::ATOMIC_CMP_SWAP: {
-    std::pair<SDValue, SDValue> Tmp = ExpandAtomic(Node);
-    Results.push_back(Tmp.first);
-    Results.push_back(Tmp.second);
-    break;
-  }
   case ISD::ATOMIC_CMP_SWAP_WITH_SUCCESS: {
     // Expanding an ATOMIC_CMP_SWAP_WITH_SUCCESS produces an ATOMIC_CMP_SWAP and
     // splits out the success value as a comparison. Expanding the resulting
     // ATOMIC_CMP_SWAP will produce a libcall.
     SDVTList VTs = DAG.getVTList(Node->getValueType(0), MVT::Other);
     SDValue Res = DAG.getAtomicCmpSwap(
         ISD::ATOMIC_CMP_SWAP, dl, cast<AtomicSDNode>(Node)->getMemoryVT(), VTs,
         Node->getOperand(0), Node->getOperand(1), Node->getOperand(2),
case ISD::UNDEF: {
     if (VT.isInteger())
       Results.push_back(DAG.getConstant(0, dl, VT));
     else {
       assert(VT.isFloatingPoint() && "Unknown value type!");
       Results.push_back(DAG.getConstantFP(0, dl, VT));
     }
     break;
   }
-  case ISD::TRAP: {
-    // If this operation is not supported, lower it to 'abort()' call
-    TargetLowering::ArgListTy Args;
-    TargetLowering::CallLoweringInfo CLI(DAG);
-    CLI.setDebugLoc(dl)
-        .setChain(Node->getOperand(0))
-        .setCallee(CallingConv::C, Type::getVoidTy(*DAG.getContext()),
-                   DAG.getExternalSymbol("abort",
-                                         TLI.getPointerTy(DAG.getDataLayout())),
-                   std::move(Args), 0);
-    std::pair<SDValue, SDValue> CallResult = TLI.LowerCallTo(CLI);
-
-    Results.push_back(CallResult.second);
-    break;
-  }
   case ISD::FP_ROUND:
   case ISD::BITCAST:
     Tmp1 = EmitStackConvert(Node->getOperand(0), Node->getValueType(0),
                             Node->getValueType(0), dl);
     Results.push_back(Tmp1);
     break;
   case ISD::FP_EXTEND:
     Tmp1 = EmitStackConvert(Node->getOperand(0),
case ISD::UMAX: {
     }
     Tmp1 = Node->getOperand(0);
     Tmp2 = Node->getOperand(1);
     Tmp1 = DAG.getSelectCC(dl, Tmp1, Tmp2, Tmp1, Tmp2, Pred);
     Results.push_back(Tmp1);
     break;
   }
     
-  case ISD::FMINNUM:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::FMIN_F32, RTLIB::FMIN_F64,
-                                      RTLIB::FMIN_F80, RTLIB::FMIN_F128,
-                                      RTLIB::FMIN_PPCF128));
-    break;
-  case ISD::FMAXNUM:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::FMAX_F32, RTLIB::FMAX_F64,
-                                      RTLIB::FMAX_F80, RTLIB::FMAX_F128,
-                                      RTLIB::FMAX_PPCF128));
-    break;
-  case ISD::FSQRT:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::SQRT_F32, RTLIB::SQRT_F64,
-                                      RTLIB::SQRT_F80, RTLIB::SQRT_F128,
-                                      RTLIB::SQRT_PPCF128));
-    break;
   case ISD::FSIN:
   case ISD::FCOS: {
     EVT VT = Node->getValueType(0);
-    bool isSIN = Node->getOpcode() == ISD::FSIN;
     // Turn fsin / fcos into ISD::FSINCOS node if there are a pair of fsin /
     // fcos which share the same operand and both are used.
     if ((TLI.isOperationLegalOrCustom(ISD::FSINCOS, VT) ||
          canCombineSinCosLibcall(Node, TLI, TM))
         && useSinCos(Node)) {
       SDVTList VTs = DAG.getVTList(VT, VT);
       Tmp1 = DAG.getNode(ISD::FSINCOS, dl, VTs, Node->getOperand(0));
-      if (!isSIN)
+      if (Node->getOpcode() == ISD::FCOS)
         Tmp1 = Tmp1.getValue(1);
       Results.push_back(Tmp1);
-    } else if (isSIN) {
-      Results.push_back(ExpandFPLibCall(Node, RTLIB::SIN_F32, RTLIB::SIN_F64,
-                                        RTLIB::SIN_F80, RTLIB::SIN_F128,
-                                        RTLIB::SIN_PPCF128));
-    } else {
-      Results.push_back(ExpandFPLibCall(Node, RTLIB::COS_F32, RTLIB::COS_F64,
-                                        RTLIB::COS_F80, RTLIB::COS_F128,
-                                        RTLIB::COS_PPCF128));
     }
     break;
   }
-  case ISD::FSINCOS:
-    // Expand into sincos libcall.
-    ExpandSinCosLibCall(Node, Results);
-    break;
-  case ISD::FLOG:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::LOG_F32, RTLIB::LOG_F64,
-                                      RTLIB::LOG_F80, RTLIB::LOG_F128,
-                                      RTLIB::LOG_PPCF128));
-    break;
-  case ISD::FLOG2:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::LOG2_F32, RTLIB::LOG2_F64,
-                                      RTLIB::LOG2_F80, RTLIB::LOG2_F128,
-                                      RTLIB::LOG2_PPCF128));
-    break;
-  case ISD::FLOG10:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::LOG10_F32, RTLIB::LOG10_F64,
-                                      RTLIB::LOG10_F80, RTLIB::LOG10_F128,
-                                      RTLIB::LOG10_PPCF128));
-    break;
-  case ISD::FEXP:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::EXP_F32, RTLIB::EXP_F64,
-                                      RTLIB::EXP_F80, RTLIB::EXP_F128,
-                                      RTLIB::EXP_PPCF128));
-    break;
-  case ISD::FEXP2:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::EXP2_F32, RTLIB::EXP2_F64,
-                                      RTLIB::EXP2_F80, RTLIB::EXP2_F128,
-                                      RTLIB::EXP2_PPCF128));
-    break;
-  case ISD::FTRUNC:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::TRUNC_F32, RTLIB::TRUNC_F64,
-                                      RTLIB::TRUNC_F80, RTLIB::TRUNC_F128,
-                                      RTLIB::TRUNC_PPCF128));
-    break;
-  case ISD::FFLOOR:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::FLOOR_F32, RTLIB::FLOOR_F64,
-                                      RTLIB::FLOOR_F80, RTLIB::FLOOR_F128,
-                                      RTLIB::FLOOR_PPCF128));
-    break;
-  case ISD::FCEIL:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::CEIL_F32, RTLIB::CEIL_F64,
-                                      RTLIB::CEIL_F80, RTLIB::CEIL_F128,
-                                      RTLIB::CEIL_PPCF128));
-    break;
-  case ISD::FRINT:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::RINT_F32, RTLIB::RINT_F64,
-                                      RTLIB::RINT_F80, RTLIB::RINT_F128,
-                                      RTLIB::RINT_PPCF128));
-    break;
-  case ISD::FNEARBYINT:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::NEARBYINT_F32,
-                                      RTLIB::NEARBYINT_F64,
-                                      RTLIB::NEARBYINT_F80,
-                                      RTLIB::NEARBYINT_F128,
-                                      RTLIB::NEARBYINT_PPCF128));
-    break;
-  case ISD::FROUND:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::ROUND_F32,
-                                      RTLIB::ROUND_F64,
-                                      RTLIB::ROUND_F80,
-                                      RTLIB::ROUND_F128,
-                                      RTLIB::ROUND_PPCF128));
-    break;
-  case ISD::FPOWI:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::POWI_F32, RTLIB::POWI_F64,
-                                      RTLIB::POWI_F80, RTLIB::POWI_F128,
-                                      RTLIB::POWI_PPCF128));
-    break;
-  case ISD::FPOW:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::POW_F32, RTLIB::POW_F64,
-                                      RTLIB::POW_F80, RTLIB::POW_F128,
-                                      RTLIB::POW_PPCF128));
-    break;
-  case ISD::FDIV:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::DIV_F32, RTLIB::DIV_F64,
-                                      RTLIB::DIV_F80, RTLIB::DIV_F128,
-                                      RTLIB::DIV_PPCF128));
-    break;
-  case ISD::FREM:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::REM_F32, RTLIB::REM_F64,
-                                      RTLIB::REM_F80, RTLIB::REM_F128,
-                                      RTLIB::REM_PPCF128));
-    break;
-  case ISD::FMA:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::FMA_F32, RTLIB::FMA_F64,
-                                      RTLIB::FMA_F80, RTLIB::FMA_F128,
-                                      RTLIB::FMA_PPCF128));
-    break;
   case ISD::FMAD:
     llvm_unreachable("Illegal fmad should never be formed");
 
-  case ISD::FADD:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::ADD_F32, RTLIB::ADD_F64,
-                                      RTLIB::ADD_F80, RTLIB::ADD_F128,
-                                      RTLIB::ADD_PPCF128));
-    break;
-  case ISD::FMUL:
-    Results.push_back(ExpandFPLibCall(Node, RTLIB::MUL_F32, RTLIB::MUL_F64,
-                                      RTLIB::MUL_F80, RTLIB::MUL_F128,
-                                      RTLIB::MUL_PPCF128));
-    break;
-  case ISD::FP16_TO_FP: {
-    if (Node->getValueType(0) == MVT::f32) {
-      Results.push_back(ExpandLibCall(RTLIB::FPEXT_F16_F32, Node, false));
-      break;
-    }
-
-    // We can extend to types bigger than f32 in two steps without changing the
-    // result. Since "f16 -> f32" is much more commonly available, give CodeGen
-    // the option of emitting that before resorting to a libcall.
+  case ISD::FP16_TO_FP:
+    if (Node->getValueType(0) != MVT::f32) {
+      // We can extend to types bigger than f32 in two steps without changing
+      // the result. Since "f16 -> f32" is much more commonly available, give
+      // CodeGen the option of emitting that before resorting to a libcall.
       SDValue Res =
           DAG.getNode(ISD::FP16_TO_FP, dl, MVT::f32, Node->getOperand(0));
       Results.push_back(
           DAG.getNode(ISD::FP_EXTEND, dl, Node->getValueType(0), Res));
-    break;
     }
-  case ISD::FP_TO_FP16: {
+    break;
+  case ISD::FP_TO_FP16:
     if (!TLI.useSoftFloat() && TM.Options.UnsafeFPMath) {
       SDValue Op = Node->getOperand(0);
       MVT SVT = Op.getSimpleValueType();
       if ((SVT == MVT::f64 || SVT == MVT::f80) &&
           TLI.isOperationLegalOrCustom(ISD::FP_TO_FP16, MVT::f32)) {
         // Under fastmath, we can expand this node into a fround followed by
         // a float-half conversion.
         SDValue FloatVal = DAG.getNode(ISD::FP_ROUND, dl, MVT::f32, Op,
                                        DAG.getIntPtrConstant(0, dl));
         Results.push_back(
             DAG.getNode(ISD::FP_TO_FP16, dl, MVT::i16, FloatVal));
-        break;
       }
     }
-
-    RTLIB::Libcall LC =
-        RTLIB::getFPROUND(Node->getOperand(0).getValueType(), MVT::f16);
-    assert(LC != RTLIB::UNKNOWN_LIBCALL && "Unable to expand fp_to_fp16");
-    Results.push_back(ExpandLibCall(LC, Node, false));
     break;
-  }
   case ISD::ConstantFP: {
     ConstantFPSDNode *CFP = cast<ConstantFPSDNode>(Node);
     // Check to see if this FP immediate is already legal.
     // If this is a legal constant, turn it into a TargetConstantFP node.
     if (!TLI.isFPImmLegal(CFP->getValueAPF(), Node->getValueType(0)))
       Results.push_back(ExpandConstantFP(CFP, true));
     break;
   }
   case ISD::FSUB: {
     EVT VT = Node->getValueType(0);
     if (TLI.isOperationLegalOrCustom(ISD::FADD, VT) &&
         TLI.isOperationLegalOrCustom(ISD::FNEG, VT)) {
       const SDNodeFlags *Flags = &cast<BinaryWithFlagsSDNode>(Node)->Flags;
       Tmp1 = DAG.getNode(ISD::FNEG, dl, VT, Node->getOperand(1));
       Tmp1 = DAG.getNode(ISD::FADD, dl, VT, Node->getOperand(0), Tmp1, Flags);
       Results.push_back(Tmp1);
-    } else {
-      Results.push_back(ExpandFPLibCall(Node, RTLIB::SUB_F32, RTLIB::SUB_F64,
-                                        RTLIB::SUB_F80, RTLIB::SUB_F128,
-                                        RTLIB::SUB_PPCF128));
     }
     break;
   }
   case ISD::SUB: {
     EVT VT = Node->getValueType(0);
     assert(TLI.isOperationLegalOrCustom(ISD::ADD, VT) &&
            TLI.isOperationLegalOrCustom(ISD::XOR, VT) &&
            "Don't know how to expand this subtraction!");
case ISD::SREM: {
     bool isSigned = Node->getOpcode() == ISD::SREM;
     unsigned DivOpc = isSigned ? ISD::SDIV : ISD::UDIV;
     unsigned DivRemOpc = isSigned ? ISD::SDIVREM : ISD::UDIVREM;
     Tmp2 = Node->getOperand(0);
     Tmp3 = Node->getOperand(1);
     if (TLI.isOperationLegalOrCustom(DivRemOpc, VT)) {
       SDVTList VTs = DAG.getVTList(VT, VT);
       Tmp1 = DAG.getNode(DivRemOpc, dl, VTs, Tmp2, Tmp3).getValue(1);
+      Results.push_back(Tmp1);
     } else if (TLI.isOperationLegalOrCustom(DivOpc, VT)) {
       // X % Y -> X-X/Y*Y
       Tmp1 = DAG.getNode(DivOpc, dl, VT, Tmp2, Tmp3);
       Tmp1 = DAG.getNode(ISD::MUL, dl, VT, Tmp1, Tmp3);
       Tmp1 = DAG.getNode(ISD::SUB, dl, VT, Tmp2, Tmp1);
-    } else if (isSigned)
-      Tmp1 = ExpandIntLibCall(Node, true,
-                              RTLIB::SREM_I8,
-                              RTLIB::SREM_I16, RTLIB::SREM_I32,
-                              RTLIB::SREM_I64, RTLIB::SREM_I128);
-    else
-      Tmp1 = ExpandIntLibCall(Node, false,
-                              RTLIB::UREM_I8,
-                              RTLIB::UREM_I16, RTLIB::UREM_I32,
-                              RTLIB::UREM_I64, RTLIB::UREM_I128);
       Results.push_back(Tmp1);
+    }
     break;
   }
   case ISD::UDIV:
   case ISD::SDIV: {
     bool isSigned = Node->getOpcode() == ISD::SDIV;
     unsigned DivRemOpc = isSigned ? ISD::SDIVREM : ISD::UDIVREM;
     EVT VT = Node->getValueType(0);
+    if (TLI.isOperationLegalOrCustom(DivRemOpc, VT)) {
       SDVTList VTs = DAG.getVTList(VT, VT);
-    if (TLI.isOperationLegalOrCustom(DivRemOpc, VT))
       Tmp1 = DAG.getNode(DivRemOpc, dl, VTs, Node->getOperand(0),
                          Node->getOperand(1));
-    else if (isSigned)
-      Tmp1 = ExpandIntLibCall(Node, true,
-                              RTLIB::SDIV_I8,
-                              RTLIB::SDIV_I16, RTLIB::SDIV_I32,
-                              RTLIB::SDIV_I64, RTLIB::SDIV_I128);
-    else
-      Tmp1 = ExpandIntLibCall(Node, false,
-                              RTLIB::UDIV_I8,
-                              RTLIB::UDIV_I16, RTLIB::UDIV_I32,
-                              RTLIB::UDIV_I64, RTLIB::UDIV_I128);
       Results.push_back(Tmp1);
+    }
     break;
   }
   case ISD::MULHU:
   case ISD::MULHS: {
     unsigned ExpandOpcode = Node->getOpcode() == ISD::MULHU ? ISD::UMUL_LOHI :
                                                               ISD::SMUL_LOHI;
     EVT VT = Node->getValueType(0);
     SDVTList VTs = DAG.getVTList(VT, VT);
     assert(TLI.isOperationLegalOrCustom(ExpandOpcode, VT) &&
            "If this wasn't legal, it shouldn't have been created!");
     Tmp1 = DAG.getNode(ExpandOpcode, dl, VTs, Node->getOperand(0),
                        Node->getOperand(1));
     Results.push_back(Tmp1.getValue(1));
     break;
   }
-  case ISD::SDIVREM:
-  case ISD::UDIVREM:
-    // Expand into divrem libcall
-    ExpandDivRemLibCall(Node, Results);
-    break;
   case ISD::MUL: {
     EVT VT = Node->getValueType(0);
     SDVTList VTs = DAG.getVTList(VT, VT);
     // See if multiply or divide can be lowered using two-result operations.
     // We just need the low half of the multiply; try both the signed
     // and unsigned forms. If the target supports both SMUL_LOHI and
     // UMUL_LOHI, form a preference by checking which forms of plain
     // MULH it supports.
if (TLI.isOperationLegalOrCustom(ISD::ZERO_EXTEND, VT) &&
         TLI.expandMUL(Node, Lo, Hi, HalfType, DAG)) {
       Lo = DAG.getNode(ISD::ZERO_EXTEND, dl, VT, Lo);
       Hi = DAG.getNode(ISD::ANY_EXTEND, dl, VT, Hi);
       SDValue Shift =
           DAG.getConstant(HalfType.getSizeInBits(), dl,
                           TLI.getShiftAmountTy(HalfType, DAG.getDataLayout()));
       Hi = DAG.getNode(ISD::SHL, dl, VT, Hi, Shift);
       Results.push_back(DAG.getNode(ISD::OR, dl, VT, Lo, Hi));
-      break;
     }
-
-    Tmp1 = ExpandIntLibCall(Node, false,
-                            RTLIB::MUL_I8,
-                            RTLIB::MUL_I16, RTLIB::MUL_I32,
-                            RTLIB::MUL_I64, RTLIB::MUL_I128);
-    Results.push_back(Tmp1);
     break;
   }
   case ISD::SADDO:
   case ISD::SSUBO: {
     SDValue LHS = Node->getOperand(0);
     SDValue RHS = Node->getOperand(1);
     SDValue Sum = DAG.getNode(Node->getOpcode() == ISD::SADDO ?
                               ISD::ADD : ISD::SUB, dl, LHS.getValueType(),
bool SelectionDAGLegalize::ExpandNode(SDNode *Node) {
   case ISD::INTRINSIC_W_CHAIN:
   case ISD::INTRINSIC_WO_CHAIN:
   case ISD::INTRINSIC_VOID:
     // FIXME: Custom lowering for these operations shouldn't return null!
     break;
   }
 
   // Replace the original node with the legalized result.
+  if (Results.empty())
+    return false;
+
+  ReplaceNode(Node, Results.data());
+  return true;
+}
+
+void SelectionDAGLegalize::ConvertNodeToLibcall(SDNode *Node) {
+  SmallVector<SDValue, 8> Results;
+  SDLoc dl(Node);
+  SDValue Tmp1, Tmp2, Tmp3, Tmp4;
+  unsigned Opc = Node->getOpcode();
+  switch (Opc) {
+  case ISD::ATOMIC_FENCE: {
+    // If the target didn't lower this, lower it to '__sync_synchronize()' call
+    // FIXME: handle "fence singlethread" more efficiently.
+    TargetLowering::ArgListTy Args;
+
+    TargetLowering::CallLoweringInfo CLI(DAG);
+    CLI.setDebugLoc(dl)
+        .setChain(Node->getOperand(0))
+        .setCallee(CallingConv::C, Type::getVoidTy(*DAG.getContext()),
+                   DAG.getExternalSymbol("__sync_synchronize",
+                                         TLI.getPointerTy(DAG.getDataLayout())),
+                   std::move(Args), 0);
+
+    std::pair<SDValue, SDValue> CallResult = TLI.LowerCallTo(CLI);
+
+    Results.push_back(CallResult.second);
+    break;
+  }
+  // By default, atomic intrinsics are marked Legal and lowered. Targets
+  // which don't support them directly, however, may want libcalls, in which
+  // case they mark them Expand, and we get here.
+  case ISD::ATOMIC_SWAP:
+  case ISD::ATOMIC_LOAD_ADD:
+  case ISD::ATOMIC_LOAD_SUB:
+  case ISD::ATOMIC_LOAD_AND:
+  case ISD::ATOMIC_LOAD_OR:
+  case ISD::ATOMIC_LOAD_XOR:
+  case ISD::ATOMIC_LOAD_NAND:
+  case ISD::ATOMIC_LOAD_MIN:
+  case ISD::ATOMIC_LOAD_MAX:
+  case ISD::ATOMIC_LOAD_UMIN:
+  case ISD::ATOMIC_LOAD_UMAX:
+  case ISD::ATOMIC_CMP_SWAP: {
+    MVT VT = cast<AtomicSDNode>(Node)->getMemoryVT().getSimpleVT();
+    RTLIB::Libcall LC = RTLIB::getATOMIC(Opc, VT);
+    assert(LC != RTLIB::UNKNOWN_LIBCALL && "Unexpected atomic op or value type!");
+
+    std::pair<SDValue, SDValue> Tmp = ExpandChainLibCall(LC, Node, false);
+    Results.push_back(Tmp.first);
+    Results.push_back(Tmp.second);
+    break;
+  }
+  case ISD::TRAP: {
+    // If this operation is not supported, lower it to 'abort()' call
+    TargetLowering::ArgListTy Args;
+    TargetLowering::CallLoweringInfo CLI(DAG);
+    CLI.setDebugLoc(dl)
+        .setChain(Node->getOperand(0))
+        .setCallee(CallingConv::C, Type::getVoidTy(*DAG.getContext()),
+                   DAG.getExternalSymbol("abort",
+                                         TLI.getPointerTy(DAG.getDataLayout())),
+                   std::move(Args), 0);
+    std::pair<SDValue, SDValue> CallResult = TLI.LowerCallTo(CLI);
+
+    Results.push_back(CallResult.second);
+    break;
+  }
+  case ISD::FMINNUM:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::FMIN_F32, RTLIB::FMIN_F64,
+                                      RTLIB::FMIN_F80, RTLIB::FMIN_F128,
+                                      RTLIB::FMIN_PPCF128));
+    break;
+  case ISD::FMAXNUM:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::FMAX_F32, RTLIB::FMAX_F64,
+                                      RTLIB::FMAX_F80, RTLIB::FMAX_F128,
+                                      RTLIB::FMAX_PPCF128));
+    break;
+  case ISD::FSQRT:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::SQRT_F32, RTLIB::SQRT_F64,
+                                      RTLIB::SQRT_F80, RTLIB::SQRT_F128,
+                                      RTLIB::SQRT_PPCF128));
+    break;
+  case ISD::FSIN:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::SIN_F32, RTLIB::SIN_F64,
+                                      RTLIB::SIN_F80, RTLIB::SIN_F128,
+                                      RTLIB::SIN_PPCF128));
+    break;
+  case ISD::FCOS:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::COS_F32, RTLIB::COS_F64,
+                                      RTLIB::COS_F80, RTLIB::COS_F128,
+                                      RTLIB::COS_PPCF128));
+    break;
+  case ISD::FSINCOS:
+    // Expand into sincos libcall.
+    ExpandSinCosLibCall(Node, Results);
+    break;
+  case ISD::FLOG:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::LOG_F32, RTLIB::LOG_F64,
+                                      RTLIB::LOG_F80, RTLIB::LOG_F128,
+                                      RTLIB::LOG_PPCF128));
+    break;
+  case ISD::FLOG2:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::LOG2_F32, RTLIB::LOG2_F64,
+                                      RTLIB::LOG2_F80, RTLIB::LOG2_F128,
+                                      RTLIB::LOG2_PPCF128));
+    break;
+  case ISD::FLOG10:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::LOG10_F32, RTLIB::LOG10_F64,
+                                      RTLIB::LOG10_F80, RTLIB::LOG10_F128,
+                                      RTLIB::LOG10_PPCF128));
+    break;
+  case ISD::FEXP:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::EXP_F32, RTLIB::EXP_F64,
+                                      RTLIB::EXP_F80, RTLIB::EXP_F128,
+                                      RTLIB::EXP_PPCF128));
+    break;
+  case ISD::FEXP2:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::EXP2_F32, RTLIB::EXP2_F64,
+                                      RTLIB::EXP2_F80, RTLIB::EXP2_F128,
+                                      RTLIB::EXP2_PPCF128));
+    break;
+  case ISD::FTRUNC:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::TRUNC_F32, RTLIB::TRUNC_F64,
+                                      RTLIB::TRUNC_F80, RTLIB::TRUNC_F128,
+                                      RTLIB::TRUNC_PPCF128));
+    break;
+  case ISD::FFLOOR:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::FLOOR_F32, RTLIB::FLOOR_F64,
+                                      RTLIB::FLOOR_F80, RTLIB::FLOOR_F128,
+                                      RTLIB::FLOOR_PPCF128));
+    break;
+  case ISD::FCEIL:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::CEIL_F32, RTLIB::CEIL_F64,
+                                      RTLIB::CEIL_F80, RTLIB::CEIL_F128,
+                                      RTLIB::CEIL_PPCF128));
+    break;
+  case ISD::FRINT:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::RINT_F32, RTLIB::RINT_F64,
+                                      RTLIB::RINT_F80, RTLIB::RINT_F128,
+                                      RTLIB::RINT_PPCF128));
+    break;
+  case ISD::FNEARBYINT:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::NEARBYINT_F32,
+                                      RTLIB::NEARBYINT_F64,
+                                      RTLIB::NEARBYINT_F80,
+                                      RTLIB::NEARBYINT_F128,
+                                      RTLIB::NEARBYINT_PPCF128));
+    break;
+  case ISD::FROUND:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::ROUND_F32,
+                                      RTLIB::ROUND_F64,
+                                      RTLIB::ROUND_F80,
+                                      RTLIB::ROUND_F128,
+                                      RTLIB::ROUND_PPCF128));
+    break;
+  case ISD::FPOWI:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::POWI_F32, RTLIB::POWI_F64,
+                                      RTLIB::POWI_F80, RTLIB::POWI_F128,
+                                      RTLIB::POWI_PPCF128));
+    break;
+  case ISD::FPOW:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::POW_F32, RTLIB::POW_F64,
+                                      RTLIB::POW_F80, RTLIB::POW_F128,
+                                      RTLIB::POW_PPCF128));
+    break;
+  case ISD::FDIV:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::DIV_F32, RTLIB::DIV_F64,
+                                      RTLIB::DIV_F80, RTLIB::DIV_F128,
+                                      RTLIB::DIV_PPCF128));
+    break;
+  case ISD::FREM:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::REM_F32, RTLIB::REM_F64,
+                                      RTLIB::REM_F80, RTLIB::REM_F128,
+                                      RTLIB::REM_PPCF128));
+    break;
+  case ISD::FMA:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::FMA_F32, RTLIB::FMA_F64,
+                                      RTLIB::FMA_F80, RTLIB::FMA_F128,
+                                      RTLIB::FMA_PPCF128));
+    break;
+  case ISD::FADD:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::ADD_F32, RTLIB::ADD_F64,
+                                      RTLIB::ADD_F80, RTLIB::ADD_F128,
+                                      RTLIB::ADD_PPCF128));
+    break;
+  case ISD::FMUL:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::MUL_F32, RTLIB::MUL_F64,
+                                      RTLIB::MUL_F80, RTLIB::MUL_F128,
+                                      RTLIB::MUL_PPCF128));
+    break;
+  case ISD::FP16_TO_FP:
+    if (Node->getValueType(0) == MVT::f32) {
+      Results.push_back(ExpandLibCall(RTLIB::FPEXT_F16_F32, Node, false));
+    }
+    break;
+  case ISD::FP_TO_FP16: {
+    RTLIB::Libcall LC =
+        RTLIB::getFPROUND(Node->getOperand(0).getValueType(), MVT::f16);
+    assert(LC != RTLIB::UNKNOWN_LIBCALL && "Unable to expand fp_to_fp16");
+    Results.push_back(ExpandLibCall(LC, Node, false));
+    break;
+  }
+  case ISD::FSUB:
+    Results.push_back(ExpandFPLibCall(Node, RTLIB::SUB_F32, RTLIB::SUB_F64,
+                                      RTLIB::SUB_F80, RTLIB::SUB_F128,
+                                      RTLIB::SUB_PPCF128));
+    break;
+  case ISD::SREM:
+    Results.push_back(ExpandIntLibCall(Node, true,
+                                       RTLIB::SREM_I8,
+                                       RTLIB::SREM_I16, RTLIB::SREM_I32,
+                                       RTLIB::SREM_I64, RTLIB::SREM_I128));
+    break;
+  case ISD::UREM:
+    Results.push_back(ExpandIntLibCall(Node, false,
+                                       RTLIB::UREM_I8,
+                                       RTLIB::UREM_I16, RTLIB::UREM_I32,
+                                       RTLIB::UREM_I64, RTLIB::UREM_I128));
+    break;
+  case ISD::SDIV:
+    Results.push_back(ExpandIntLibCall(Node, true,
+                                       RTLIB::SDIV_I8,
+                                       RTLIB::SDIV_I16, RTLIB::SDIV_I32,
+                                       RTLIB::SDIV_I64, RTLIB::SDIV_I128));
+    break;
+  case ISD::UDIV:
+    Results.push_back(ExpandIntLibCall(Node, false,
+                                       RTLIB::UDIV_I8,
+                                       RTLIB::UDIV_I16, RTLIB::UDIV_I32,
+                                       RTLIB::UDIV_I64, RTLIB::UDIV_I128));
+    break;
+  case ISD::SDIVREM:
+  case ISD::UDIVREM:
+    // Expand into divrem libcall
+    ExpandDivRemLibCall(Node, Results);
+    break;
+  case ISD::MUL:
+    Results.push_back(ExpandIntLibCall(Node, false,
+                                       RTLIB::MUL_I8,
+                                       RTLIB::MUL_I16, RTLIB::MUL_I32,
+                                       RTLIB::MUL_I64, RTLIB::MUL_I128));
+    break;
+  }
+
+  // Replace the original node with the legalized result.
   if (!Results.empty())
     ReplaceNode(Node, Results.data());
 }
 
 void SelectionDAGLegalize::PromoteNode(SDNode *Node) {
   SmallVector<SDValue, 8> Results;
   MVT OVT = Node->getSimpleValueType(0);
   if (Node->getOpcode() == ISD::UINT_TO_FP ||

lib/Target/ARM/ARMISelLowering.cpp

ARMTargetLowering::ARMTargetLowering(const TargetMachine &TM,
 
   // Only ARMv6 has BSWAP.
   if (!Subtarget->hasV6Ops())
     setOperationAction(ISD::BSWAP, MVT::i32, Expand);
 
   if (!(Subtarget->hasDivide() && Subtarget->isThumb2()) &&
       !(Subtarget->hasDivideInARMMode() && !Subtarget->isThumb())) {
     // These are expanded into libcalls if the cpu doesn't have HW divider.
-    setOperationAction(ISD::SDIV,  MVT::i32, Expand);
-    setOperationAction(ISD::UDIV,  MVT::i32, Expand);
+    setOperationAction(ISD::SDIV,  MVT::i32, LibCall);
+    setOperationAction(ISD::UDIV,  MVT::i32, LibCall);
   }
 
   if (Subtarget->isTargetWindows() && !Subtarget->hasDivide()) {
     setOperationAction(ISD::SDIV, MVT::i32, Custom);
     setOperationAction(ISD::UDIV, MVT::i32, Custom);
 
     setOperationAction(ISD::SDIV, MVT::i64, Custom);
     setOperationAction(ISD::UDIV, MVT::i64, Custom);

test/CodeGen/ARM/divmod-eabi.ll

 }
 
 define i32 @g4(i32 %a, i32 %b) {
 ; EABI-LABEL: g4:
 ; GNU-LABEL: g4:
 ; DARWIN-LABEL: g4:
 entry:
   %div = sdiv i32 %a, %b
-; EABI: __aeabi_idivmod
+; EABI: __aeabi_idiv{{$}}
 ; EABI: mov [[div:r[0-9]+]], r0
 ; GNU: __aeabi_idiv
 ; GNU: mov [[sum:r[0-9]+]], r0
 ; DARWIN: ___divsi3
 ; DARWIN: mov [[sum:r[0-9]+]], r0
   %rem = srem i32 %b, %div
 ; EABI: __aeabi_idivmod
 ; GNU: __modsi3
 ; DARWIN: __modsi3
   %add = add nsw i32 %rem, %div
 ; EABI: add r0, r1, [[div]]
 ; GNU: add r0{{.*}}[[sum]]
 ; DARWIN: add r0{{.*}}[[sum]]
   ret i32 %add
 }