[Intrinsic] Signed and Unsigned Saturation Subtraction Intrinsics

[leonardchan]

Summary

Add an intrinsic that takes 2 integers and perform saturation subtraction on them.

This is a part of implementing fixed point arithmetic in clang where some of the more complex operations will be implemented as intrinsics.

[craig.topper]

LGTM

[bjope]

nit: there is a spelling error in the title "Intirnsics"

llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp

4964 ↗

(On Diff #171375)

It should say ">= 0" here.

llvm/trunk/include/llvm/CodeGen/ISDOpcodes.h

enum NodeType {
     /// RESULT = [US]ADDSAT(LHS, RHS) - Perform saturation addition on 2
     /// integers with the same bit width (W). If the true value of LHS + RHS
     /// exceeds the largest value that can be represented by W bits, the
     /// resulting value is this maximum value. Otherwise, if this value is less
     /// than the smallest value that can be represented by W bits, the
     /// resulting value is this minimum value.
     SADDSAT, UADDSAT,
 
+    /// RESULT = [US]SUBSAT(LHS, RHS) - Perform saturation subtraction on 2
+    /// integers with the same bit width (W). If the true value of LHS - RHS
+    /// exceeds the largest value that can be represented by W bits, the
+    /// resulting value is this maximum value. Otherwise, if this value is less
+    /// than the smallest value that can be represented by W bits, the
+    /// resulting value is this minimum value.
+    SSUBSAT, USUBSAT,
+
     /// Simple binary floating point operators.
     FADD, FSUB, FMUL, FDIV, FREM,
 
     /// Constrained versions of the binary floating point operators.
     /// These will be lowered to the simple operators before final selection.
     /// They are used to limit optimizations while the DAG is being
     /// optimized.
     STRICT_FADD, STRICT_FSUB, STRICT_FMUL, STRICT_FDIV, STRICT_FREM,

llvm/trunk/include/llvm/CodeGen/TargetLowering.h

 
   /// Get a pointer to vector element \p Idx located in memory for a vector of
   /// type \p VecVT starting at a base address of \p VecPtr. If \p Idx is out of
   /// bounds the returned pointer is unspecified, but will be within the vector
   /// bounds.
   SDValue getVectorElementPointer(SelectionDAG &DAG, SDValue VecPtr, EVT VecVT,
                                   SDValue Index) const;
 
-  /// Method for building the DAG expansion of ISD::[US]ADDSAT. This method
-  /// accepts integers or vectors of integers as its arguments.
-  SDValue getExpandedSaturationAddition(SDNode *Node, SelectionDAG &DAG) const;
+  /// Method for building the DAG expansion of ISD::[US][ADD|SUB]SAT. This
+  /// method accepts integers or vectors of integers as its arguments.
+  SDValue getExpandedSaturationAdditionSubtraction(SDNode *Node,
+                                                   SelectionDAG &DAG) const;
 
   //===--------------------------------------------------------------------===//
   // Instruction Emitting Hooks
   //
 
   /// This method should be implemented by targets that mark instructions with
   /// the 'usesCustomInserter' flag.  These instructions are special in various
   /// ways, which require special support to insert.  The specified MachineInstr

llvm/trunk/include/llvm/IR/Intrinsics.td

 //===------------------------- Fixed Point Intrinsics ---------------------===//
 //
 def int_sadd_sat : Intrinsic<[llvm_anyint_ty],
                              [LLVMMatchType<0>, LLVMMatchType<0>],
                              [IntrNoMem, IntrSpeculatable, Commutative]>;
 def int_uadd_sat : Intrinsic<[llvm_anyint_ty],
                              [LLVMMatchType<0>, LLVMMatchType<0>],
                              [IntrNoMem, IntrSpeculatable, Commutative]>;
+def int_ssub_sat : Intrinsic<[llvm_anyint_ty],
+                             [LLVMMatchType<0>, LLVMMatchType<0>],
+                             [IntrNoMem, IntrSpeculatable]>;
+def int_usub_sat : Intrinsic<[llvm_anyint_ty],
+                             [LLVMMatchType<0>, LLVMMatchType<0>],
+                             [IntrNoMem, IntrSpeculatable]>;
 
 //===------------------------- Memory Use Markers -------------------------===//
 //
 def int_lifetime_start  : Intrinsic<[],
                                     [llvm_i64_ty, llvm_anyptr_ty],
                                     [IntrArgMemOnly, NoCapture<1>]>;
 def int_lifetime_end    : Intrinsic<[],
                                     [llvm_i64_ty, llvm_anyptr_ty],

llvm/trunk/include/llvm/Target/TargetSelectionDAG.td

def smax       : SDNode<"ISD::SMAX"      , SDTIntBinOp,
                                   [SDNPCommutative, SDNPAssociative]>;
 def umin       : SDNode<"ISD::UMIN"      , SDTIntBinOp,
                                   [SDNPCommutative, SDNPAssociative]>;
 def umax       : SDNode<"ISD::UMAX"      , SDTIntBinOp,
                                   [SDNPCommutative, SDNPAssociative]>;
 
 def saddsat    : SDNode<"ISD::SADDSAT"   , SDTIntBinOp, [SDNPCommutative]>;
 def uaddsat    : SDNode<"ISD::UADDSAT"   , SDTIntBinOp, [SDNPCommutative]>;
+def ssubsat    : SDNode<"ISD::SSUBSAT"   , SDTIntBinOp>;
+def usubsat    : SDNode<"ISD::USUBSAT"   , SDTIntBinOp>;
 
 def sext_inreg : SDNode<"ISD::SIGN_EXTEND_INREG", SDTExtInreg>;
 def sext_invec : SDNode<"ISD::SIGN_EXTEND_VECTOR_INREG", SDTExtInvec>;
 def zext_invec : SDNode<"ISD::ZERO_EXTEND_VECTOR_INREG", SDTExtInvec>;
 
 def abs        : SDNode<"ISD::ABS"        , SDTIntUnaryOp>;
 def bitreverse : SDNode<"ISD::BITREVERSE" , SDTIntUnaryOp>;
 def bswap      : SDNode<"ISD::BSWAP"      , SDTIntUnaryOp>;

llvm/trunk/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp

case ISD::STRICT_FNEARBYINT:
     // These pseudo-ops get legalized as if they were their non-strict
     // equivalent.  For instance, if ISD::FSQRT is legal then ISD::STRICT_FSQRT
     // is also legal, but if ISD::FSQRT requires expansion then so does
     // ISD::STRICT_FSQRT.
     Action = TLI.getStrictFPOperationAction(Node->getOpcode(),
                                             Node->getValueType(0));
     break;
   case ISD::SADDSAT:
-  case ISD::UADDSAT: {
+  case ISD::UADDSAT:
+  case ISD::SSUBSAT:
+  case ISD::USUBSAT: {
     Action = TLI.getOperationAction(Node->getOpcode(), Node->getValueType(0));
     break;
   }
   case ISD::MSCATTER:
     Action = TLI.getOperationAction(Node->getOpcode(),
                     cast<MaskedScatterSDNode>(Node)->getValue().getValueType());
     break;
   case ISD::MSTORE:
if (TLI.isOperationLegalOrCustom(ISD::ZERO_EXTEND, VT) &&
           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;
   }
   case ISD::SADDSAT:
-  case ISD::UADDSAT: {
-    Results.push_back(TLI.getExpandedSaturationAddition(Node, DAG));
+  case ISD::UADDSAT:
+  case ISD::SSUBSAT:
+  case ISD::USUBSAT: {
+    Results.push_back(TLI.getExpandedSaturationAdditionSubtraction(Node, DAG));
     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(),

llvm/trunk/lib/CodeGen/SelectionDAG/LegalizeIntegerTypes.cpp

#endif
   case ISD::USUBO:       Res = PromoteIntRes_UADDSUBO(N, ResNo); break;
   case ISD::SMULO:
   case ISD::UMULO:       Res = PromoteIntRes_XMULO(N, ResNo); break;
 
   case ISD::ADDCARRY:
   case ISD::SUBCARRY:    Res = PromoteIntRes_ADDSUBCARRY(N, ResNo); break;
 
   case ISD::SADDSAT:
-  case ISD::UADDSAT:     Res = PromoteIntRes_ADDSAT(N); break;
+  case ISD::UADDSAT:
+  case ISD::SSUBSAT:
+  case ISD::USUBSAT:     Res = PromoteIntRes_ADDSUBSAT(N); break;
 
   case ISD::ATOMIC_LOAD:
     Res = PromoteIntRes_Atomic0(cast<AtomicSDNode>(N)); break;
 
   case ISD::ATOMIC_LOAD_ADD:
   case ISD::ATOMIC_LOAD_SUB:
   case ISD::ATOMIC_LOAD_AND:
   case ISD::ATOMIC_LOAD_CLR:
SDValue DAGTypeLegalizer::PromoteIntRes_Overflow(SDNode *N) {
 
   // Modified the sum result - switch anything that used the old sum to use
   // the new one.
   ReplaceValueWith(SDValue(N, 0), Res);
 
   return SDValue(Res.getNode(), 1);
 }
 
-SDValue DAGTypeLegalizer::PromoteIntRes_ADDSAT(SDNode *N) {
+SDValue DAGTypeLegalizer::PromoteIntRes_ADDSUBSAT(SDNode *N) {
   // For promoting iN -> iM, this can be expanded by
   // 1. ANY_EXTEND iN to iM
   // 2. SHL by M-N
-  // 3. U/SADDSAT
+  // 3. [US][ADD|SUB]SAT
   // 4. L/ASHR by M-N
   SDLoc dl(N);
   SDValue Op1 = N->getOperand(0);
   SDValue Op2 = N->getOperand(1);
   unsigned OldBits = Op1.getValueSizeInBits();
 
   unsigned Opcode = N->getOpcode();
-  assert((Opcode == ISD::SADDSAT || Opcode == ISD::UADDSAT) &&
-         "Expected opcode to be SADDSAT or UADDSAT");
-  unsigned ShiftOp = Opcode == ISD::SADDSAT ? ISD::SRA : ISD::SRL;
+  unsigned ShiftOp;
+  switch (Opcode) {
+  case ISD::SADDSAT:
+  case ISD::SSUBSAT:
+    ShiftOp = ISD::SRA;
+    break;
+  case ISD::UADDSAT:
+  case ISD::USUBSAT:
+    ShiftOp = ISD::SRL;
+    break;
+  default:
+    llvm_unreachable("Expected opcode to be signed or unsigned saturation "
+                     "addition or subtraction");
+  }
 
   SDValue Op1Promoted = GetPromotedInteger(Op1);
   SDValue Op2Promoted = GetPromotedInteger(Op2);
 
   EVT PromotedType = Op1Promoted.getValueType();
   unsigned NewBits = Op1Promoted.getValueSizeInBits();
   unsigned SHLAmount = NewBits - OldBits;
   EVT SHVT = TLI.getShiftAmountTy(PromotedType, DAG.getDataLayout());
#endif
   case ISD::SADDO:
   case ISD::SSUBO: ExpandIntRes_SADDSUBO(N, Lo, Hi); break;
   case ISD::UADDO:
   case ISD::USUBO: ExpandIntRes_UADDSUBO(N, Lo, Hi); break;
   case ISD::UMULO:
   case ISD::SMULO: ExpandIntRes_XMULO(N, Lo, Hi); break;
 
   case ISD::SADDSAT:
-  case ISD::UADDSAT: ExpandIntRes_ADDSAT(N, Lo, Hi); break;
+  case ISD::UADDSAT:
+  case ISD::SSUBSAT:
+  case ISD::USUBSAT: ExpandIntRes_ADDSUBSAT(N, Lo, Hi); break;
   }
 
   // If Lo/Hi is null, the sub-method took care of registering results etc.
   if (Lo.getNode())
     SetExpandedInteger(SDValue(N, ResNo), Lo, Hi);
 }
 
 /// Lower an atomic node to the appropriate builtin call.
void DAGTypeLegalizer::ExpandIntRes_READCYCLECOUNTER(SDNode *N, SDValue &Lo,
   EVT NVT = TLI.getTypeToTransformTo(*DAG.getContext(), N->getValueType(0));
   SDVTList VTs = DAG.getVTList(NVT, NVT, MVT::Other);
   SDValue R = DAG.getNode(N->getOpcode(), DL, VTs, N->getOperand(0));
   Lo = R.getValue(0);
   Hi = R.getValue(1);
   ReplaceValueWith(SDValue(N, 1), R.getValue(2));
 }
 
-void DAGTypeLegalizer::ExpandIntRes_ADDSAT(SDNode *N, SDValue &Lo,
+void DAGTypeLegalizer::ExpandIntRes_ADDSUBSAT(SDNode *N, SDValue &Lo,
                                               SDValue &Hi) {
-  SDValue Result = TLI.getExpandedSaturationAddition(N, DAG);
+  SDValue Result = TLI.getExpandedSaturationAdditionSubtraction(N, DAG);
   SplitInteger(Result, Lo, Hi);
 }
 
 void DAGTypeLegalizer::ExpandIntRes_SADDSUBO(SDNode *Node,
                                              SDValue &Lo, SDValue &Hi) {
   SDValue LHS = Node->getOperand(0);
   SDValue RHS = Node->getOperand(1);
   SDLoc dl(Node);

llvm/trunk/lib/CodeGen/SelectionDAG/LegalizeTypes.h

private:
   SDValue PromoteIntRes_SRA(SDNode *N);
   SDValue PromoteIntRes_SRL(SDNode *N);
   SDValue PromoteIntRes_TRUNCATE(SDNode *N);
   SDValue PromoteIntRes_UADDSUBO(SDNode *N, unsigned ResNo);
   SDValue PromoteIntRes_ADDSUBCARRY(SDNode *N, unsigned ResNo);
   SDValue PromoteIntRes_UNDEF(SDNode *N);
   SDValue PromoteIntRes_VAARG(SDNode *N);
   SDValue PromoteIntRes_XMULO(SDNode *N, unsigned ResNo);
-  SDValue PromoteIntRes_ADDSAT(SDNode *N);
+  SDValue PromoteIntRes_ADDSUBSAT(SDNode *N);
 
   // Integer Operand Promotion.
   bool PromoteIntegerOperand(SDNode *N, unsigned OpNo);
   SDValue PromoteIntOp_ANY_EXTEND(SDNode *N);
   SDValue PromoteIntOp_ATOMIC_STORE(AtomicSDNode *N);
   SDValue PromoteIntOp_BITCAST(SDNode *N);
   SDValue PromoteIntOp_BUILD_PAIR(SDNode *N);
   SDValue PromoteIntOp_BR_CC(SDNode *N, unsigned OpNo);
private:
   void ExpandIntRes_UREM              (SDNode *N, SDValue &Lo, SDValue &Hi);
   void ExpandIntRes_Shift             (SDNode *N, SDValue &Lo, SDValue &Hi);
 
   void ExpandIntRes_MINMAX            (SDNode *N, SDValue &Lo, SDValue &Hi);
 
   void ExpandIntRes_SADDSUBO          (SDNode *N, SDValue &Lo, SDValue &Hi);
   void ExpandIntRes_UADDSUBO          (SDNode *N, SDValue &Lo, SDValue &Hi);
   void ExpandIntRes_XMULO             (SDNode *N, SDValue &Lo, SDValue &Hi);
-  void ExpandIntRes_ADDSAT            (SDNode *N, SDValue &Lo, SDValue &Hi);
+  void ExpandIntRes_ADDSUBSAT         (SDNode *N, SDValue &Lo, SDValue &Hi);
 
   void ExpandIntRes_ATOMIC_LOAD       (SDNode *N, SDValue &Lo, SDValue &Hi);
 
   void ExpandShiftByConstant(SDNode *N, const APInt &Amt,
                              SDValue &Lo, SDValue &Hi);
   bool ExpandShiftWithKnownAmountBit(SDNode *N, SDValue &Lo, SDValue &Hi);
   bool ExpandShiftWithUnknownAmountBit(SDNode *N, SDValue &Lo, SDValue &Hi);
 

llvm/trunk/lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp

SDValue VectorLegalizer::LegalizeOp(SDValue Op) {
   case ISD::SMAX:
   case ISD::UMIN:
   case ISD::UMAX:
   case ISD::SMUL_LOHI:
   case ISD::UMUL_LOHI:
   case ISD::FCANONICALIZE:
   case ISD::SADDSAT:
   case ISD::UADDSAT:
+  case ISD::SSUBSAT:
+  case ISD::USUBSAT:
     Action = TLI.getOperationAction(Node->getOpcode(), Node->getValueType(0));
     break;
   case ISD::FP_ROUND_INREG:
     Action = TLI.getOperationAction(Node->getOpcode(),
                cast<VTSDNode>(Node->getOperand(1))->getVT());
     break;
   case ISD::SINT_TO_FP:
   case ISD::UINT_TO_FP:

llvm/trunk/lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp

#endif
   case ISD::FMAXIMUM:
   case ISD::SMIN:
   case ISD::SMAX:
   case ISD::UMIN:
   case ISD::UMAX:
 
   case ISD::SADDSAT:
   case ISD::UADDSAT:
+  case ISD::SSUBSAT:
+  case ISD::USUBSAT:
 
   case ISD::FPOW:
   case ISD::FREM:
   case ISD::FSUB:
   case ISD::MUL:
   case ISD::OR:
   case ISD::SDIV:
   case ISD::SREM:
#endif
   case ISD::SREM:
   case ISD::FREM:
   case ISD::SMIN:
   case ISD::SMAX:
   case ISD::UMIN:
   case ISD::UMAX:
   case ISD::SADDSAT:
   case ISD::UADDSAT:
+  case ISD::SSUBSAT:
+  case ISD::USUBSAT:
     SplitVecRes_BinOp(N, Lo, Hi);
     break;
   case ISD::FMA:
     SplitVecRes_TernaryOp(N, Lo, Hi);
     break;
   case ISD::STRICT_FADD:
   case ISD::STRICT_FSUB:
   case ISD::STRICT_FMUL:

llvm/trunk/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp

case Intrinsic::sadd_sat: {
     return nullptr;
   }
   case Intrinsic::uadd_sat: {
     SDValue Op1 = getValue(I.getArgOperand(0));
     SDValue Op2 = getValue(I.getArgOperand(1));
     setValue(&I, DAG.getNode(ISD::UADDSAT, sdl, Op1.getValueType(), Op1, Op2));
     return nullptr;
   }
+  case Intrinsic::ssub_sat: {
+    SDValue Op1 = getValue(I.getArgOperand(0));
+    SDValue Op2 = getValue(I.getArgOperand(1));
+    setValue(&I, DAG.getNode(ISD::SSUBSAT, sdl, Op1.getValueType(), Op1, Op2));
+    return nullptr;
+  }
+  case Intrinsic::usub_sat: {
+    SDValue Op1 = getValue(I.getArgOperand(0));
+    SDValue Op2 = getValue(I.getArgOperand(1));
+    setValue(&I, DAG.getNode(ISD::USUBSAT, sdl, Op1.getValueType(), Op1, Op2));
+    return nullptr;
+  }
   case Intrinsic::stacksave: {
     SDValue Op = getRoot();
     Res = DAG.getNode(
         ISD::STACKSAVE, sdl,
         DAG.getVTList(TLI.getPointerTy(DAG.getDataLayout()), MVT::Other), Op);
     setValue(&I, Res);
     DAG.setRoot(Res.getValue(1));
     return nullptr;

llvm/trunk/lib/CodeGen/SelectionDAG/SelectionDAGDumper.cpp

#endif
   case ISD::SUBE:                       return "sube";
   case ISD::SUBCARRY:                   return "subcarry";
   case ISD::SHL_PARTS:                  return "shl_parts";
   case ISD::SRA_PARTS:                  return "sra_parts";
   case ISD::SRL_PARTS:                  return "srl_parts";
 
   case ISD::SADDSAT:                    return "saddsat";
   case ISD::UADDSAT:                    return "uaddsat";
+  case ISD::SSUBSAT:                    return "ssubsat";
+  case ISD::USUBSAT:                    return "usubsat";
 
   // Conversion operators.
   case ISD::SIGN_EXTEND:                return "sign_extend";
   case ISD::ZERO_EXTEND:                return "zero_extend";
   case ISD::ANY_EXTEND:                 return "any_extend";
   case ISD::SIGN_EXTEND_INREG:          return "sign_extend_inreg";
   case ISD::ANY_EXTEND_VECTOR_INREG:    return "any_extend_vector_inreg";
   case ISD::SIGN_EXTEND_VECTOR_INREG:   return "sign_extend_vector_inreg";

llvm/trunk/lib/CodeGen/SelectionDAG/TargetLowering.cpp

if (C->isNullValue() && CC == ISD::SETEQ) {
       SDValue Scc = DAG.getNode(ISD::SRL, dl, VT, Clz,
                                 DAG.getConstant(Log2b, dl, MVT::i32));
       return DAG.getNode(ISD::TRUNCATE, dl, MVT::i32, Scc);
     }
   }
   return SDValue();
 }
 
-SDValue TargetLowering::getExpandedSaturationAddition(SDNode *Node,
-                                                      SelectionDAG &DAG) const {
+SDValue TargetLowering::getExpandedSaturationAdditionSubtraction(
+    SDNode *Node, SelectionDAG &DAG) const {
   unsigned Opcode = Node->getOpcode();
-  assert((Opcode == ISD::SADDSAT || Opcode == ISD::UADDSAT) &&
-         "Expected method to receive SADDSAT or UADDSAT node.");
+  unsigned OverflowOp;
+  switch (Opcode) {
+  case ISD::SADDSAT:
+    OverflowOp = ISD::SADDO;
+    break;
+  case ISD::UADDSAT:
+    OverflowOp = ISD::UADDO;
+    break;
+  case ISD::SSUBSAT:
+    OverflowOp = ISD::SSUBO;
+    break;
+  case ISD::USUBSAT:
+    OverflowOp = ISD::USUBO;
+    break;
+  default:
+    llvm_unreachable("Expected method to receive signed or unsigned saturation "
+                     "addition or subtraction node.");
+  }
   assert(Node->getNumOperands() == 2 && "Expected node to have 2 operands.");
 
   SDLoc dl(Node);
   SDValue LHS = Node->getOperand(0);
   SDValue RHS = Node->getOperand(1);
   assert(LHS.getValueType().isScalarInteger() &&
          "Expected operands to be integers. Vector of int arguments should "
          "already be unrolled.");
   assert(RHS.getValueType().isScalarInteger() &&
          "Expected operands to be integers. Vector of int arguments should "
          "already be unrolled.");
   assert(LHS.getValueType() == RHS.getValueType() &&
          "Expected both operands to be the same type");
 
-  unsigned OverflowOp = Opcode == ISD::SADDSAT ? ISD::SADDO : ISD::UADDO;
   unsigned BitWidth = LHS.getValueSizeInBits();
   EVT ResultType = LHS.getValueType();
   EVT BoolVT =
       getSetCCResultType(DAG.getDataLayout(), *DAG.getContext(), ResultType);
   SDValue Result =
       DAG.getNode(OverflowOp, dl, DAG.getVTList(ResultType, BoolVT), LHS, RHS);
-  SDValue Sum = Result.getValue(0);
+  SDValue SumDiff = Result.getValue(0);
   SDValue Overflow = Result.getValue(1);
   SDValue Zero = DAG.getConstant(0, dl, ResultType);
 
-  if (Opcode == ISD::SADDSAT) {
-    // SatMax -> Overflow && Sum < 0
-    // SatMin -> Overflow && Sum > 0
+  if (Opcode == ISD::UADDSAT) {
+    // Just need to check overflow for SatMax.
+    APInt MaxVal = APInt::getMaxValue(BitWidth);
+    SDValue SatMax = DAG.getConstant(MaxVal, dl, ResultType);
+    return DAG.getSelect(dl, ResultType, Overflow, SatMax, SumDiff);
+  } else if (Opcode == ISD::USUBSAT) {
+    // Just need to check overflow for SatMin.
+    APInt MinVal = APInt::getMinValue(BitWidth);
+    SDValue SatMin = DAG.getConstant(MinVal, dl, ResultType);
+    return DAG.getSelect(dl, ResultType, Overflow, SatMin, SumDiff);
+  } else {
+    // SatMax -> Overflow && SumDiff < 0
+    // SatMin -> Overflow && SumDiff >= 0
     APInt MinVal = APInt::getSignedMinValue(BitWidth);
     APInt MaxVal = APInt::getSignedMaxValue(BitWidth);
     SDValue SatMin = DAG.getConstant(MinVal, dl, ResultType);
     SDValue SatMax = DAG.getConstant(MaxVal, dl, ResultType);
-    SDValue SumNeg = DAG.getSetCC(dl, BoolVT, Sum, Zero, ISD::SETLT);
+    SDValue SumNeg = DAG.getSetCC(dl, BoolVT, SumDiff, Zero, ISD::SETLT);
     Result = DAG.getSelect(dl, ResultType, SumNeg, SatMax, SatMin);
-    return DAG.getSelect(dl, ResultType, Overflow, Result, Sum);
-  } else {
-    // Just need to check overflow for SatMax.
-    APInt MaxVal = APInt::getMaxValue(BitWidth);
-    SDValue SatMax = DAG.getConstant(MaxVal, dl, ResultType);
-    return DAG.getSelect(dl, ResultType, Overflow, SatMax, Sum);
+    return DAG.getSelect(dl, ResultType, Overflow, Result, SumDiff);
   }
 }

llvm/trunk/lib/CodeGen/TargetLoweringBase.cpp

for (MVT VT : MVT::all_valuetypes()) {
     setOperationAction(ISD::FMAD, VT, Expand);
     setOperationAction(ISD::SMIN, VT, Expand);
     setOperationAction(ISD::SMAX, VT, Expand);
     setOperationAction(ISD::UMIN, VT, Expand);
     setOperationAction(ISD::UMAX, VT, Expand);
     setOperationAction(ISD::ABS, VT, Expand);
     setOperationAction(ISD::SADDSAT, VT, Expand);
     setOperationAction(ISD::UADDSAT, VT, Expand);
+    setOperationAction(ISD::SSUBSAT, VT, Expand);
+    setOperationAction(ISD::USUBSAT, VT, Expand);
 
     // Overflow operations default to expand
     setOperationAction(ISD::SADDO, VT, Expand);
     setOperationAction(ISD::SSUBO, VT, Expand);
     setOperationAction(ISD::UADDO, VT, Expand);
     setOperationAction(ISD::USUBO, VT, Expand);
     setOperationAction(ISD::SMULO, VT, Expand);
     setOperationAction(ISD::UMULO, VT, Expand);

llvm/trunk/lib/IR/Verifier.cpp

if (CS.isCall()) {
         Assert(RI->getReturnValue() == DeoptCI,
                "calls to experimental_deoptimize must be followed by a return "
                "of the value computed by experimental_deoptimize");
     }
 
     break;
   }
   case Intrinsic::sadd_sat:
-  case Intrinsic::uadd_sat: {
+  case Intrinsic::uadd_sat:
+  case Intrinsic::ssub_sat:
+  case Intrinsic::usub_sat: {
     Value *Op1 = CS.getArgOperand(0);
     Value *Op2 = CS.getArgOperand(1);
-    Assert(
-        Op1->getType()->isIntOrIntVectorTy(),
-        "first operand of [us]add_sat must be an int type or vector of ints");
-    Assert(
-        Op2->getType()->isIntOrIntVectorTy(),
-        "second operand of [us]add_sat must be an int type or vector of ints");
+    Assert(Op1->getType()->isIntOrIntVectorTy(),
+           "first operand of [us][add|sub]_sat must be an int type or vector "
+           "of ints");
+    Assert(Op2->getType()->isIntOrIntVectorTy(),
+           "second operand of [us][add|sub]_sat must be an int type or vector "
+           "of ints");
     break;
   }
   };
 }
 
 /// Carefully grab the subprogram from a local scope.
 ///
 /// This carefully grabs the subprogram from a local scope, avoiding the

llvm/trunk/test/CodeGen/X86/ssub_sat.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mcpu=generic -mtriple=x86_64-linux | FileCheck %s
+; RUN: llc < %s -mcpu=generic -mtriple=i686 -mattr=cmov | FileCheck %s --check-prefix=CHECK32
+
+declare  i4  @llvm.ssub.sat.i4   (i4,  i4)
+declare  i32 @llvm.ssub.sat.i32  (i32, i32)
+declare  i64 @llvm.ssub.sat.i64  (i64, i64)
+declare  <4 x i32> @llvm.ssub.sat.v4i32(<4 x i32>, <4 x i32>)
+
+define i32 @func(i32 %x, i32 %y) {
+; CHECK-LABEL: func:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xorl %eax, %eax
+; CHECK-NEXT:    movl %edi, %ecx
+; CHECK-NEXT:    subl %esi, %ecx
+; CHECK-NEXT:    setns %al
+; CHECK-NEXT:    addl $2147483647, %eax # imm = 0x7FFFFFFF
+; CHECK-NEXT:    subl %esi, %edi
+; CHECK-NEXT:    cmovnol %edi, %eax
+; CHECK-NEXT:    retq
+;
+; CHECK32-LABEL: func:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    pushl %esi
+; CHECK32-NEXT:    .cfi_def_cfa_offset 8
+; CHECK32-NEXT:    .cfi_offset %esi, -8
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %edx
+; CHECK32-NEXT:    xorl %ecx, %ecx
+; CHECK32-NEXT:    movl %eax, %esi
+; CHECK32-NEXT:    subl %edx, %esi
+; CHECK32-NEXT:    setns %cl
+; CHECK32-NEXT:    addl $2147483647, %ecx # imm = 0x7FFFFFFF
+; CHECK32-NEXT:    subl %edx, %eax
+; CHECK32-NEXT:    cmovol %ecx, %eax
+; CHECK32-NEXT:    popl %esi
+; CHECK32-NEXT:    .cfi_def_cfa_offset 4
+; CHECK32-NEXT:    retl
+  %tmp = call i32 @llvm.ssub.sat.i32(i32 %x, i32 %y);
+  ret i32 %tmp;
+}
+
+define i64 @func2(i64 %x, i64 %y) {
+; CHECK-LABEL: func2:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xorl %ecx, %ecx
+; CHECK-NEXT:    movq %rdi, %rax
+; CHECK-NEXT:    subq %rsi, %rax
+; CHECK-NEXT:    setns %cl
+; CHECK-NEXT:    movabsq $9223372036854775807, %rax # imm = 0x7FFFFFFFFFFFFFFF
+; CHECK-NEXT:    addq %rcx, %rax
+; CHECK-NEXT:    subq %rsi, %rdi
+; CHECK-NEXT:    cmovnoq %rdi, %rax
+; CHECK-NEXT:    retq
+;
+; CHECK32-LABEL: func2:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    pushl %ebp
+; CHECK32-NEXT:    .cfi_def_cfa_offset 8
+; CHECK32-NEXT:    pushl %ebx
+; CHECK32-NEXT:    .cfi_def_cfa_offset 12
+; CHECK32-NEXT:    pushl %edi
+; CHECK32-NEXT:    .cfi_def_cfa_offset 16
+; CHECK32-NEXT:    pushl %esi
+; CHECK32-NEXT:    .cfi_def_cfa_offset 20
+; CHECK32-NEXT:    .cfi_offset %esi, -20
+; CHECK32-NEXT:    .cfi_offset %edi, -16
+; CHECK32-NEXT:    .cfi_offset %ebx, -12
+; CHECK32-NEXT:    .cfi_offset %ebp, -8
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %esi
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %edi
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %ebx
+; CHECK32-NEXT:    subl {{[0-9]+}}(%esp), %edi
+; CHECK32-NEXT:    movl %ebx, %ebp
+; CHECK32-NEXT:    sbbl %esi, %ebp
+; CHECK32-NEXT:    movl %ebp, %eax
+; CHECK32-NEXT:    sarl $31, %eax
+; CHECK32-NEXT:    xorl %ecx, %ecx
+; CHECK32-NEXT:    testl %ebp, %ebp
+; CHECK32-NEXT:    setns %cl
+; CHECK32-NEXT:    movl %ecx, %edx
+; CHECK32-NEXT:    addl $2147483647, %edx # imm = 0x7FFFFFFF
+; CHECK32-NEXT:    testl %ebx, %ebx
+; CHECK32-NEXT:    setns %bl
+; CHECK32-NEXT:    cmpb %cl, %bl
+; CHECK32-NEXT:    setne %cl
+; CHECK32-NEXT:    testl %esi, %esi
+; CHECK32-NEXT:    setns %ch
+; CHECK32-NEXT:    cmpb %ch, %bl
+; CHECK32-NEXT:    setne %ch
+; CHECK32-NEXT:    testb %cl, %ch
+; CHECK32-NEXT:    cmovel %ebp, %edx
+; CHECK32-NEXT:    cmovel %edi, %eax
+; CHECK32-NEXT:    popl %esi
+; CHECK32-NEXT:    .cfi_def_cfa_offset 16
+; CHECK32-NEXT:    popl %edi
+; CHECK32-NEXT:    .cfi_def_cfa_offset 12
+; CHECK32-NEXT:    popl %ebx
+; CHECK32-NEXT:    .cfi_def_cfa_offset 8
+; CHECK32-NEXT:    popl %ebp
+; CHECK32-NEXT:    .cfi_def_cfa_offset 4
+; CHECK32-NEXT:    retl
+  %tmp = call i64 @llvm.ssub.sat.i64(i64 %x, i64 %y);
+  ret i64 %tmp;
+}
+
+define i4 @func3(i4 %x, i4 %y) {
+; CHECK-LABEL: func3:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    movl %edi, %eax
+; CHECK-NEXT:    shlb $4, %sil
+; CHECK-NEXT:    shlb $4, %al
+; CHECK-NEXT:    movl %eax, %ecx
+; CHECK-NEXT:    subb %sil, %cl
+; CHECK-NEXT:    setns %cl
+; CHECK-NEXT:    subb %sil, %al
+; CHECK-NEXT:    jno .LBB2_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    addb $127, %cl
+; CHECK-NEXT:    movl %ecx, %eax
+; CHECK-NEXT:  .LBB2_2:
+; CHECK-NEXT:    sarb $4, %al
+; CHECK-NEXT:    # kill: def $al killed $al killed $eax
+; CHECK-NEXT:    retq
+;
+; CHECK32-LABEL: func3:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    movb {{[0-9]+}}(%esp), %al
+; CHECK32-NEXT:    movb {{[0-9]+}}(%esp), %dl
+; CHECK32-NEXT:    shlb $4, %dl
+; CHECK32-NEXT:    shlb $4, %al
+; CHECK32-NEXT:    movl %eax, %ecx
+; CHECK32-NEXT:    subb %dl, %cl
+; CHECK32-NEXT:    setns %cl
+; CHECK32-NEXT:    subb %dl, %al
+; CHECK32-NEXT:    jno .LBB2_2
+; CHECK32-NEXT:  # %bb.1:
+; CHECK32-NEXT:    addb $127, %cl
+; CHECK32-NEXT:    movl %ecx, %eax
+; CHECK32-NEXT:  .LBB2_2:
+; CHECK32-NEXT:    sarb $4, %al
+; CHECK32-NEXT:    retl
+  %tmp = call i4 @llvm.ssub.sat.i4(i4 %x, i4 %y);
+  ret i4 %tmp;
+}
+
+define <4 x i32> @vec(<4 x i32> %x, <4 x i32> %y) {
+; CHECK-LABEL: vec:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    pshufd {{.*#+}} xmm2 = xmm1[1,1,2,3]
+; CHECK-NEXT:    movd %xmm2, %ecx
+; CHECK-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[1,1,2,3]
+; CHECK-NEXT:    movd %xmm2, %r8d
+; CHECK-NEXT:    xorl %edx, %edx
+; CHECK-NEXT:    movl %r8d, %esi
+; CHECK-NEXT:    subl %ecx, %esi
+; CHECK-NEXT:    setns %dl
+; CHECK-NEXT:    addl $2147483647, %edx # imm = 0x7FFFFFFF
+; CHECK-NEXT:    subl %ecx, %r8d
+; CHECK-NEXT:    cmovol %edx, %r8d
+; CHECK-NEXT:    movd %xmm1, %edx
+; CHECK-NEXT:    movd %xmm0, %ecx
+; CHECK-NEXT:    xorl %esi, %esi
+; CHECK-NEXT:    movl %ecx, %edi
+; CHECK-NEXT:    subl %edx, %edi
+; CHECK-NEXT:    setns %sil
+; CHECK-NEXT:    addl $2147483647, %esi # imm = 0x7FFFFFFF
+; CHECK-NEXT:    subl %edx, %ecx
+; CHECK-NEXT:    cmovol %esi, %ecx
+; CHECK-NEXT:    pshufd {{.*#+}} xmm2 = xmm1[2,3,0,1]
+; CHECK-NEXT:    movd %xmm2, %edx
+; CHECK-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[2,3,0,1]
+; CHECK-NEXT:    movd %xmm2, %eax
+; CHECK-NEXT:    xorl %edi, %edi
+; CHECK-NEXT:    movl %eax, %esi
+; CHECK-NEXT:    subl %edx, %esi
+; CHECK-NEXT:    setns %dil
+; CHECK-NEXT:    addl $2147483647, %edi # imm = 0x7FFFFFFF
+; CHECK-NEXT:    subl %edx, %eax
+; CHECK-NEXT:    cmovol %edi, %eax
+; CHECK-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[3,1,2,3]
+; CHECK-NEXT:    movd %xmm1, %r9d
+; CHECK-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,1,2,3]
+; CHECK-NEXT:    movd %xmm0, %edx
+; CHECK-NEXT:    xorl %edi, %edi
+; CHECK-NEXT:    movl %edx, %esi
+; CHECK-NEXT:    subl %r9d, %esi
+; CHECK-NEXT:    setns %dil
+; CHECK-NEXT:    addl $2147483647, %edi # imm = 0x7FFFFFFF
+; CHECK-NEXT:    subl %r9d, %edx
+; CHECK-NEXT:    cmovol %edi, %edx
+; CHECK-NEXT:    movd %edx, %xmm0
+; CHECK-NEXT:    movd %eax, %xmm1
+; CHECK-NEXT:    punpckldq {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1]
+; CHECK-NEXT:    movd %ecx, %xmm0
+; CHECK-NEXT:    movd %r8d, %xmm2
+; CHECK-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm2[0],xmm0[1],xmm2[1]
+; CHECK-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; CHECK-NEXT:    retq
+;
+; CHECK32-LABEL: vec:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    pushl %ebp
+; CHECK32-NEXT:    .cfi_def_cfa_offset 8
+; CHECK32-NEXT:    pushl %ebx
+; CHECK32-NEXT:    .cfi_def_cfa_offset 12
+; CHECK32-NEXT:    pushl %edi
+; CHECK32-NEXT:    .cfi_def_cfa_offset 16
+; CHECK32-NEXT:    pushl %esi
+; CHECK32-NEXT:    .cfi_def_cfa_offset 20
+; CHECK32-NEXT:    .cfi_offset %esi, -20
+; CHECK32-NEXT:    .cfi_offset %edi, -16
+; CHECK32-NEXT:    .cfi_offset %ebx, -12
+; CHECK32-NEXT:    .cfi_offset %ebp, -8
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %ecx
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %edx
+; CHECK32-NEXT:    xorl %eax, %eax
+; CHECK32-NEXT:    movl %ecx, %esi
+; CHECK32-NEXT:    subl %edx, %esi
+; CHECK32-NEXT:    setns %al
+; CHECK32-NEXT:    addl $2147483647, %eax # imm = 0x7FFFFFFF
+; CHECK32-NEXT:    subl %edx, %ecx
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %edx
+; CHECK32-NEXT:    cmovol %eax, %ecx
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %esi
+; CHECK32-NEXT:    xorl %eax, %eax
+; CHECK32-NEXT:    movl %edx, %edi
+; CHECK32-NEXT:    subl %esi, %edi
+; CHECK32-NEXT:    setns %al
+; CHECK32-NEXT:    addl $2147483647, %eax # imm = 0x7FFFFFFF
+; CHECK32-NEXT:    subl %esi, %edx
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %esi
+; CHECK32-NEXT:    cmovol %eax, %edx
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %edi
+; CHECK32-NEXT:    xorl %eax, %eax
+; CHECK32-NEXT:    movl %esi, %ebx
+; CHECK32-NEXT:    subl %edi, %ebx
+; CHECK32-NEXT:    setns %al
+; CHECK32-NEXT:    addl $2147483647, %eax # imm = 0x7FFFFFFF
+; CHECK32-NEXT:    subl %edi, %esi
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %edi
+; CHECK32-NEXT:    cmovol %eax, %esi
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK32-NEXT:    xorl %ebx, %ebx
+; CHECK32-NEXT:    movl %edi, %ebp
+; CHECK32-NEXT:    subl %eax, %ebp
+; CHECK32-NEXT:    setns %bl
+; CHECK32-NEXT:    addl $2147483647, %ebx # imm = 0x7FFFFFFF
+; CHECK32-NEXT:    subl %eax, %edi
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK32-NEXT:    cmovol %ebx, %edi
+; CHECK32-NEXT:    movl %ecx, 12(%eax)
+; CHECK32-NEXT:    movl %edx, 8(%eax)
+; CHECK32-NEXT:    movl %esi, 4(%eax)
+; CHECK32-NEXT:    movl %edi, (%eax)
+; CHECK32-NEXT:    popl %esi
+; CHECK32-NEXT:    .cfi_def_cfa_offset 16
+; CHECK32-NEXT:    popl %edi
+; CHECK32-NEXT:    .cfi_def_cfa_offset 12
+; CHECK32-NEXT:    popl %ebx
+; CHECK32-NEXT:    .cfi_def_cfa_offset 8
+; CHECK32-NEXT:    popl %ebp
+; CHECK32-NEXT:    .cfi_def_cfa_offset 4
+; CHECK32-NEXT:    retl $4
+  %tmp = call <4 x i32> @llvm.ssub.sat.v4i32(<4 x i32> %x, <4 x i32> %y);
+  ret <4 x i32> %tmp;
+}

llvm/trunk/test/CodeGen/X86/usub_sat.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mcpu=generic -mtriple=x86_64-linux | FileCheck %s
+; RUN: llc < %s -mcpu=generic -mtriple=i686 -mattr=cmov | FileCheck %s --check-prefix=CHECK32
+
+declare  i4  @llvm.usub.sat.i4   (i4,  i4)
+declare  i32 @llvm.usub.sat.i32  (i32, i32)
+declare  i64 @llvm.usub.sat.i64  (i64, i64)
+declare  <4 x i32> @llvm.usub.sat.v4i32(<4 x i32>, <4 x i32>)
+
+define i32 @func(i32 %x, i32 %y) {
+; CHECK-LABEL: func:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xorl %eax, %eax
+; CHECK-NEXT:    subl %esi, %edi
+; CHECK-NEXT:    cmovael %edi, %eax
+; CHECK-NEXT:    retq
+;
+; CHECK32-LABEL: func:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK32-NEXT:    xorl %ecx, %ecx
+; CHECK32-NEXT:    subl {{[0-9]+}}(%esp), %eax
+; CHECK32-NEXT:    cmovbl %ecx, %eax
+; CHECK32-NEXT:    retl
+  %tmp = call i32 @llvm.usub.sat.i32(i32 %x, i32 %y);
+  ret i32 %tmp;
+}
+
+define i64 @func2(i64 %x, i64 %y) {
+; CHECK-LABEL: func2:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xorl %eax, %eax
+; CHECK-NEXT:    subq %rsi, %rdi
+; CHECK-NEXT:    cmovaeq %rdi, %rax
+; CHECK-NEXT:    retq
+;
+; CHECK32-LABEL: func2:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %edx
+; CHECK32-NEXT:    xorl %ecx, %ecx
+; CHECK32-NEXT:    subl {{[0-9]+}}(%esp), %eax
+; CHECK32-NEXT:    sbbl {{[0-9]+}}(%esp), %edx
+; CHECK32-NEXT:    cmovbl %ecx, %edx
+; CHECK32-NEXT:    cmovbl %ecx, %eax
+; CHECK32-NEXT:    retl
+  %tmp = call i64 @llvm.usub.sat.i64(i64 %x, i64 %y);
+  ret i64 %tmp;
+}
+
+define i4 @func3(i4 %x, i4 %y) {
+; CHECK-LABEL: func3:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    movl %edi, %eax
+; CHECK-NEXT:    shlb $4, %sil
+; CHECK-NEXT:    shlb $4, %al
+; CHECK-NEXT:    subb %sil, %al
+; CHECK-NEXT:    jae .LBB2_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    xorl %eax, %eax
+; CHECK-NEXT:  .LBB2_2:
+; CHECK-NEXT:    shrb $4, %al
+; CHECK-NEXT:    # kill: def $al killed $al killed $eax
+; CHECK-NEXT:    retq
+;
+; CHECK32-LABEL: func3:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    movb {{[0-9]+}}(%esp), %al
+; CHECK32-NEXT:    movb {{[0-9]+}}(%esp), %cl
+; CHECK32-NEXT:    shlb $4, %cl
+; CHECK32-NEXT:    shlb $4, %al
+; CHECK32-NEXT:    subb %cl, %al
+; CHECK32-NEXT:    jae .LBB2_2
+; CHECK32-NEXT:  # %bb.1:
+; CHECK32-NEXT:    xorl %eax, %eax
+; CHECK32-NEXT:  .LBB2_2:
+; CHECK32-NEXT:    shrb $4, %al
+; CHECK32-NEXT:    # kill: def $al killed $al killed $eax
+; CHECK32-NEXT:    retl
+  %tmp = call i4 @llvm.usub.sat.i4(i4 %x, i4 %y);
+  ret i4 %tmp;
+}
+
+define <4 x i32> @vec(<4 x i32> %x, <4 x i32> %y) {
+; CHECK-LABEL: vec:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    pshufd {{.*#+}} xmm2 = xmm1[3,1,2,3]
+; CHECK-NEXT:    movd %xmm2, %eax
+; CHECK-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[3,1,2,3]
+; CHECK-NEXT:    movd %xmm2, %ecx
+; CHECK-NEXT:    xorl %edx, %edx
+; CHECK-NEXT:    subl %eax, %ecx
+; CHECK-NEXT:    cmovbl %edx, %ecx
+; CHECK-NEXT:    movd %ecx, %xmm2
+; CHECK-NEXT:    pshufd {{.*#+}} xmm3 = xmm1[2,3,0,1]
+; CHECK-NEXT:    movd %xmm3, %eax
+; CHECK-NEXT:    pshufd {{.*#+}} xmm3 = xmm0[2,3,0,1]
+; CHECK-NEXT:    movd %xmm3, %ecx
+; CHECK-NEXT:    subl %eax, %ecx
+; CHECK-NEXT:    cmovbl %edx, %ecx
+; CHECK-NEXT:    movd %ecx, %xmm3
+; CHECK-NEXT:    punpckldq {{.*#+}} xmm3 = xmm3[0],xmm2[0],xmm3[1],xmm2[1]
+; CHECK-NEXT:    movd %xmm1, %eax
+; CHECK-NEXT:    movd %xmm0, %ecx
+; CHECK-NEXT:    subl %eax, %ecx
+; CHECK-NEXT:    cmovbl %edx, %ecx
+; CHECK-NEXT:    movd %ecx, %xmm2
+; CHECK-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[1,1,2,3]
+; CHECK-NEXT:    movd %xmm1, %eax
+; CHECK-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[1,1,2,3]
+; CHECK-NEXT:    movd %xmm0, %ecx
+; CHECK-NEXT:    subl %eax, %ecx
+; CHECK-NEXT:    cmovbl %edx, %ecx
+; CHECK-NEXT:    movd %ecx, %xmm0
+; CHECK-NEXT:    punpckldq {{.*#+}} xmm2 = xmm2[0],xmm0[0],xmm2[1],xmm0[1]
+; CHECK-NEXT:    punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm3[0]
+; CHECK-NEXT:    movdqa %xmm2, %xmm0
+; CHECK-NEXT:    retq
+;
+; CHECK32-LABEL: vec:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    pushl %ebx
+; CHECK32-NEXT:    .cfi_def_cfa_offset 8
+; CHECK32-NEXT:    pushl %edi
+; CHECK32-NEXT:    .cfi_def_cfa_offset 12
+; CHECK32-NEXT:    pushl %esi
+; CHECK32-NEXT:    .cfi_def_cfa_offset 16
+; CHECK32-NEXT:    .cfi_offset %esi, -16
+; CHECK32-NEXT:    .cfi_offset %edi, -12
+; CHECK32-NEXT:    .cfi_offset %ebx, -8
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %ecx
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %edx
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %esi
+; CHECK32-NEXT:    movl {{[0-9]+}}(%esp), %edi
+; CHECK32-NEXT:    xorl %ebx, %ebx
+; CHECK32-NEXT:    subl {{[0-9]+}}(%esp), %edi
+; CHECK32-NEXT:    cmovbl %ebx, %edi
+; CHECK32-NEXT:    subl {{[0-9]+}}(%esp), %esi
+; CHECK32-NEXT:    cmovbl %ebx, %esi
+; CHECK32-NEXT:    subl {{[0-9]+}}(%esp), %edx
+; CHECK32-NEXT:    cmovbl %ebx, %edx
+; CHECK32-NEXT:    subl {{[0-9]+}}(%esp), %ecx
+; CHECK32-NEXT:    cmovbl %ebx, %ecx
+; CHECK32-NEXT:    movl %ecx, 12(%eax)
+; CHECK32-NEXT:    movl %edx, 8(%eax)
+; CHECK32-NEXT:    movl %esi, 4(%eax)
+; CHECK32-NEXT:    movl %edi, (%eax)
+; CHECK32-NEXT:    popl %esi
+; CHECK32-NEXT:    .cfi_def_cfa_offset 12
+; CHECK32-NEXT:    popl %edi
+; CHECK32-NEXT:    .cfi_def_cfa_offset 8
+; CHECK32-NEXT:    popl %ebx
+; CHECK32-NEXT:    .cfi_def_cfa_offset 4
+; CHECK32-NEXT:    retl $4
+  %tmp = call <4 x i32> @llvm.usub.sat.v4i32(<4 x i32> %x, <4 x i32> %y);
+  ret <4 x i32> %tmp;
+}