Index: include/llvm/CodeGen/TargetLowering.h
===================================================================
--- include/llvm/CodeGen/TargetLowering.h
+++ include/llvm/CodeGen/TargetLowering.h
@@ -1738,6 +1738,16 @@
     return false;
   }
 
+  /// Return true if it is more correct/profitable to use strict FP_TO_INT
+  /// conversion operations - canonicalizing the FP source value instead of
+  /// converting all cases and then selecting based on value.
+  /// This may be true if the target throws exceptions for out of bounds
+  /// conversions or has fast FP CMOV.
+  virtual bool shouldUseStrictFP_TO_INT(EVT FpVT, EVT IntVT,
+                                        bool IsSigned) const {
+    return false;
+  }
+
   //===--------------------------------------------------------------------===//
   // TargetLowering Configuration Methods - These methods should be invoked by
   // the derived class constructor to configure this object for the target.
Index: lib/CodeGen/SelectionDAG/TargetLowering.cpp
===================================================================
--- lib/CodeGen/SelectionDAG/TargetLowering.cpp
+++ lib/CodeGen/SelectionDAG/TargetLowering.cpp
@@ -4154,24 +4154,50 @@
                            !isOperationLegalOrCustomOrPromote(ISD::XOR, SrcVT)))
     return false;
 
-  // Expand based on maximum range of FP_TO_SINT:
-  // True = fp_to_sint(Src)
-  // False = 0x8000000000000000 + fp_to_sint(Src - 0x8000000000000000)
-  // Result = select (Src < 0x8000000000000000), True, False
-  APFloat apf(DAG.EVTToAPFloatSemantics(SrcVT),
-              APInt::getNullValue(SrcVT.getScalarSizeInBits()));
-  APInt x = APInt::getSignMask(DstVT.getScalarSizeInBits());
-  (void)apf.convertFromAPInt(x, false, APFloat::rmNearestTiesToEven);
-
-  SDValue Tmp1 = DAG.getConstantFP(apf, dl, SrcVT);
-  SDValue Tmp2 = DAG.getSetCC(dl, SetCCVT, Src, Tmp1, ISD::SETLT);
-  SDValue True = DAG.getNode(ISD::FP_TO_SINT, dl, DstVT, Src);
-  // TODO: Should any fast-math-flags be set for the FSUB?
-  SDValue False = DAG.getNode(ISD::FP_TO_SINT, dl, DstVT,
-                              DAG.getNode(ISD::FSUB, dl, SrcVT, Src, Tmp1));
-  False =
-      DAG.getNode(ISD::XOR, dl, DstVT, False, DAG.getConstant(x, dl, DstVT));
-  Result = DAG.getSelect(dl, DstVT, Tmp2, True, False);
+  // If the destination signmask can't be represented by the float, then we can
+  // just use FP_TO_SINT directly.
+  const fltSemantics &APFSem = DAG.EVTToAPFloatSemantics(SrcVT);
+  APFloat APF(APFSem, APInt::getNullValue(SrcVT.getScalarSizeInBits()));
+  APInt SignMask = APInt::getSignMask(DstVT.getScalarSizeInBits());
+  if (APFloat::opOverflow &
+      APF.convertFromAPInt(SignMask, false, APFloat::rmNearestTiesToEven)) {
+    Result = DAG.getNode(ISD::FP_TO_SINT, dl, DstVT, Src);
+    return true;
+  }
+
+  SDValue Cst = DAG.getConstantFP(APF, dl, SrcVT);
+  SDValue Sel = DAG.getSetCC(dl, SetCCVT, Src, Cst, ISD::SETLT);
+
+  bool Strict = shouldUseStrictFP_TO_INT(SrcVT, DstVT, false);
+  if (Strict) {
+    // Expand based on maximum range of FP_TO_SINT, if the value exceeds the
+    // signmask then offset (f32/f64 results should then be whole integers).
+    // Sel = Src < 0x8000000000000000
+    // Val = select Sel, Src, Src - 0x8000000000000000
+    // Ofs = select Sel, 0, 0x8000000000000000
+    // Result = fp_to_sint(Val) ^ Ofs
+
+    // TODO: Should any fast-math-flags be set for the FSUB?
+    SDValue Val = DAG.getSelect(dl, SrcVT, Sel, Src,
+                                DAG.getNode(ISD::FSUB, dl, SrcVT, Src, Cst));
+    SDValue Ofs = DAG.getSelect(dl, DstVT, Sel, DAG.getConstant(0, dl, DstVT),
+                                DAG.getConstant(SignMask, dl, DstVT));
+    Result = DAG.getNode(ISD::XOR, dl, DstVT,
+                         DAG.getNode(ISD::FP_TO_SINT, dl, DstVT, Val), Ofs);
+  } else {
+    // Expand based on maximum range of FP_TO_SINT:
+    // True = fp_to_sint(Src)
+    // False = 0x8000000000000000 + fp_to_sint(Src - 0x8000000000000000)
+    // Result = select (Src < 0x8000000000000000), True, False
+
+    SDValue True = DAG.getNode(ISD::FP_TO_SINT, dl, DstVT, Src);
+    // TODO: Should any fast-math-flags be set for the FSUB?
+    SDValue False = DAG.getNode(ISD::FP_TO_SINT, dl, DstVT,
+                                DAG.getNode(ISD::FSUB, dl, SrcVT, Src, Cst));
+    False = DAG.getNode(ISD::XOR, dl, DstVT, False,
+                        DAG.getConstant(SignMask, dl, DstVT));
+    Result = DAG.getSelect(dl, DstVT, Sel, True, False);
+  }
   return true;
 }
 
Index: lib/Target/X86/X86ISelLowering.h
===================================================================
--- lib/Target/X86/X86ISelLowering.h
+++ lib/Target/X86/X86ISelLowering.h
@@ -1045,6 +1045,9 @@
 
     bool decomposeMulByConstant(EVT VT, SDValue C) const override;
 
+    bool shouldUseStrictFP_TO_INT(EVT FpVT, EVT IntVT,
+                                  bool IsSigned) const override;
+
     /// Return true if EXTRACT_SUBVECTOR is cheap for this result type
     /// with this index.
     bool isExtractSubvectorCheap(EVT ResVT, EVT SrcVT,
Index: lib/Target/X86/X86ISelLowering.cpp
===================================================================
--- lib/Target/X86/X86ISelLowering.cpp
+++ lib/Target/X86/X86ISelLowering.cpp
@@ -4773,6 +4773,12 @@
          (1 - MulC).isPowerOf2() || (-(MulC + 1)).isPowerOf2();
 }
 
+bool X86TargetLowering::shouldUseStrictFP_TO_INT(EVT FpVT, EVT IntVT,
+                                                 bool IsSigned) const {
+  // f80 UINT_TO_FP is more efficient using Strict code if FCMOV is available.
+  return !IsSigned && FpVT == MVT::f80 && Subtarget.hasCMov();
+}
+
 bool X86TargetLowering::isExtractSubvectorCheap(EVT ResVT, EVT SrcVT,
                                                 unsigned Index) const {
   if (!isOperationLegalOrCustom(ISD::EXTRACT_SUBVECTOR, ResVT))
Index: test/CodeGen/Mips/msa/f16-llvm-ir.ll
===================================================================
--- test/CodeGen/Mips/msa/f16-llvm-ir.ll
+++ test/CodeGen/Mips/msa/f16-llvm-ir.ll
@@ -252,259 +252,58 @@
 }
 
 define i32 @ffptoui() {
-; MIPS32-O32-LABEL: ffptoui:
-; MIPS32-O32:       # %bb.0: # %entry
-; MIPS32-O32-NEXT:    lui $2, %hi(_gp_disp)
-; MIPS32-O32-NEXT:    addiu $2, $2, %lo(_gp_disp)
-; MIPS32-O32-NEXT:    addu $1, $2, $25
-; MIPS32-O32-NEXT:    lw $2, %got(h)($1)
-; MIPS32-O32-NEXT:    lw $3, %got($CPI3_0)($1)
-; MIPS32-O32-NEXT:    lwc1 $f0, %lo($CPI3_0)($3)
-; MIPS32-O32-NEXT:    lh $2, 0($2)
-; MIPS32-O32-NEXT:    fill.h $w1, $2
-; MIPS32-O32-NEXT:    fexupr.w $w1, $w1
-; MIPS32-O32-NEXT:    copy_s.w $2, $w1[0]
-; MIPS32-O32-NEXT:    mtc1 $2, $f2
-; MIPS32-O32-NEXT:    sub.s $f0, $f2, $f0
-; MIPS32-O32-NEXT:    mfc1 $2, $f0
-; MIPS32-O32-NEXT:    fill.w $w0, $2
-; MIPS32-O32-NEXT:    fexdo.h $w0, $w0, $w0
-; MIPS32-O32-NEXT:    fexupr.w $w0, $w0
-; MIPS32-O32-NEXT:    fexupr.d $w0, $w0
-; MIPS32-O32-NEXT:    copy_s.w $2, $w0[0]
-; MIPS32-O32-NEXT:    mtc1 $2, $f3
-; MIPS32-O32-NEXT:    copy_s.w $2, $w0[1]
-; MIPS32-O32-NEXT:    mthc1 $2, $f3
-; MIPS32-O32-NEXT:    trunc.w.d $f0, $f3
-; MIPS32-O32-NEXT:    mfc1 $2, $f0
-; MIPS32-O32-NEXT:    fexupr.d $w0, $w1
-; MIPS32-O32-NEXT:    copy_s.w $3, $w0[0]
-; MIPS32-O32-NEXT:    mtc1 $3, $f1
-; MIPS32-O32-NEXT:    copy_s.w $3, $w0[1]
-; MIPS32-O32-NEXT:    mthc1 $3, $f1
-; MIPS32-O32-NEXT:    trunc.w.d $f0, $f1
-; MIPS32-O32-NEXT:    mfc1 $3, $f0
-; MIPS32-O32-NEXT:    lw $1, %got($CPI3_1)($1)
-; MIPS32-O32-NEXT:    addiu $1, $1, %lo($CPI3_1)
-; MIPS32-O32-NEXT:    lui $4, 32768
-; MIPS32-O32-NEXT:    xor $2, $2, $4
-; MIPS32-O32-NEXT:    lh $1, 0($1)
-; MIPS32-O32-NEXT:    fill.h $w0, $1
-; MIPS32-O32-NEXT:    fexupr.w $w0, $w0
-; MIPS32-O32-NEXT:    copy_s.w $1, $w0[0]
-; MIPS32-O32-NEXT:    mtc1 $1, $f0
-; MIPS32-O32-NEXT:    c.olt.s $f2, $f0
-; MIPS32-O32-NEXT:    jr $ra
-; MIPS32-O32-NEXT:    movt $2, $3, $fcc0
-;
-; MIPS64R5-N32-LABEL: ffptoui:
-; MIPS64R5-N32:       # %bb.0: # %entry
-; MIPS64R5-N32-NEXT:    lui $1, %hi(%neg(%gp_rel(ffptoui)))
-; MIPS64R5-N32-NEXT:    addu $1, $1, $25
-; MIPS64R5-N32-NEXT:    addiu $1, $1, %lo(%neg(%gp_rel(ffptoui)))
-; MIPS64R5-N32-NEXT:    lw $2, %got_disp(h)($1)
-; MIPS64R5-N32-NEXT:    lw $3, %got_page(.LCPI3_0)($1)
-; MIPS64R5-N32-NEXT:    lwc1 $f0, %got_ofst(.LCPI3_0)($3)
-; MIPS64R5-N32-NEXT:    lh $2, 0($2)
-; MIPS64R5-N32-NEXT:    fill.h $w1, $2
-; MIPS64R5-N32-NEXT:    fexupr.w $w1, $w1
-; MIPS64R5-N32-NEXT:    copy_s.w $2, $w1[0]
-; MIPS64R5-N32-NEXT:    mtc1 $2, $f2
-; MIPS64R5-N32-NEXT:    sub.s $f0, $f2, $f0
-; MIPS64R5-N32-NEXT:    mfc1 $2, $f0
-; MIPS64R5-N32-NEXT:    fill.w $w0, $2
-; MIPS64R5-N32-NEXT:    fexdo.h $w0, $w0, $w0
-; MIPS64R5-N32-NEXT:    fexupr.w $w0, $w0
-; MIPS64R5-N32-NEXT:    fexupr.d $w0, $w0
-; MIPS64R5-N32-NEXT:    copy_s.d $2, $w0[0]
-; MIPS64R5-N32-NEXT:    dmtc1 $2, $f0
-; MIPS64R5-N32-NEXT:    trunc.w.d $f0, $f0
-; MIPS64R5-N32-NEXT:    mfc1 $2, $f0
-; MIPS64R5-N32-NEXT:    fexupr.d $w0, $w1
-; MIPS64R5-N32-NEXT:    copy_s.d $3, $w0[0]
-; MIPS64R5-N32-NEXT:    dmtc1 $3, $f0
-; MIPS64R5-N32-NEXT:    trunc.w.d $f0, $f0
-; MIPS64R5-N32-NEXT:    mfc1 $3, $f0
-; MIPS64R5-N32-NEXT:    lw $1, %got_page(.LCPI3_1)($1)
-; MIPS64R5-N32-NEXT:    addiu $1, $1, %got_ofst(.LCPI3_1)
-; MIPS64R5-N32-NEXT:    lui $4, 32768
-; MIPS64R5-N32-NEXT:    xor $2, $2, $4
-; MIPS64R5-N32-NEXT:    lh $1, 0($1)
-; MIPS64R5-N32-NEXT:    fill.h $w0, $1
-; MIPS64R5-N32-NEXT:    fexupr.w $w0, $w0
-; MIPS64R5-N32-NEXT:    copy_s.w $1, $w0[0]
-; MIPS64R5-N32-NEXT:    mtc1 $1, $f0
-; MIPS64R5-N32-NEXT:    c.olt.s $f2, $f0
-; MIPS64R5-N32-NEXT:    jr $ra
-; MIPS64R5-N32-NEXT:    movt $2, $3, $fcc0
-;
-; MIPS64R5-N64-LABEL: ffptoui:
-; MIPS64R5-N64:       # %bb.0: # %entry
-; MIPS64R5-N64-NEXT:    lui $1, %hi(%neg(%gp_rel(ffptoui)))
-; MIPS64R5-N64-NEXT:    daddu $1, $1, $25
-; MIPS64R5-N64-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(ffptoui)))
-; MIPS64R5-N64-NEXT:    ld $2, %got_disp(h)($1)
-; MIPS64R5-N64-NEXT:    ld $3, %got_page(.LCPI3_0)($1)
-; MIPS64R5-N64-NEXT:    lwc1 $f0, %got_ofst(.LCPI3_0)($3)
-; MIPS64R5-N64-NEXT:    lh $2, 0($2)
-; MIPS64R5-N64-NEXT:    fill.h $w1, $2
-; MIPS64R5-N64-NEXT:    fexupr.w $w1, $w1
-; MIPS64R5-N64-NEXT:    copy_s.w $2, $w1[0]
-; MIPS64R5-N64-NEXT:    mtc1 $2, $f2
-; MIPS64R5-N64-NEXT:    sub.s $f0, $f2, $f0
-; MIPS64R5-N64-NEXT:    mfc1 $2, $f0
-; MIPS64R5-N64-NEXT:    fill.w $w0, $2
-; MIPS64R5-N64-NEXT:    fexdo.h $w0, $w0, $w0
-; MIPS64R5-N64-NEXT:    fexupr.w $w0, $w0
-; MIPS64R5-N64-NEXT:    fexupr.d $w0, $w0
-; MIPS64R5-N64-NEXT:    copy_s.d $2, $w0[0]
-; MIPS64R5-N64-NEXT:    dmtc1 $2, $f0
-; MIPS64R5-N64-NEXT:    trunc.w.d $f0, $f0
-; MIPS64R5-N64-NEXT:    mfc1 $2, $f0
-; MIPS64R5-N64-NEXT:    fexupr.d $w0, $w1
-; MIPS64R5-N64-NEXT:    copy_s.d $3, $w0[0]
-; MIPS64R5-N64-NEXT:    dmtc1 $3, $f0
-; MIPS64R5-N64-NEXT:    trunc.w.d $f0, $f0
-; MIPS64R5-N64-NEXT:    mfc1 $3, $f0
-; MIPS64R5-N64-NEXT:    ld $1, %got_page(.LCPI3_1)($1)
-; MIPS64R5-N64-NEXT:    daddiu $1, $1, %got_ofst(.LCPI3_1)
-; MIPS64R5-N64-NEXT:    lui $4, 32768
-; MIPS64R5-N64-NEXT:    xor $2, $2, $4
-; MIPS64R5-N64-NEXT:    lh $1, 0($1)
-; MIPS64R5-N64-NEXT:    fill.h $w0, $1
-; MIPS64R5-N64-NEXT:    fexupr.w $w0, $w0
-; MIPS64R5-N64-NEXT:    copy_s.w $1, $w0[0]
-; MIPS64R5-N64-NEXT:    mtc1 $1, $f0
-; MIPS64R5-N64-NEXT:    c.olt.s $f2, $f0
-; MIPS64R5-N64-NEXT:    jr $ra
-; MIPS64R5-N64-NEXT:    movt $2, $3, $fcc0
-;
-; MIPSR6-O32-LABEL: ffptoui:
-; MIPSR6-O32:       # %bb.0: # %entry
-; MIPSR6-O32-NEXT:    lui $2, %hi(_gp_disp)
-; MIPSR6-O32-NEXT:    addiu $2, $2, %lo(_gp_disp)
-; MIPSR6-O32-NEXT:    addu $1, $2, $25
-; MIPSR6-O32-NEXT:    lw $2, %got(h)($1)
-; MIPSR6-O32-NEXT:    lw $1, %got($CPI3_0)($1)
-; MIPSR6-O32-NEXT:    lwc1 $f0, %lo($CPI3_0)($1)
-; MIPSR6-O32-NEXT:    lh $1, 0($2)
-; MIPSR6-O32-NEXT:    fill.h $w1, $1
-; MIPSR6-O32-NEXT:    fexupr.w $w1, $w1
-; MIPSR6-O32-NEXT:    copy_s.w $1, $w1[0]
-; MIPSR6-O32-NEXT:    mtc1 $1, $f2
-; MIPSR6-O32-NEXT:    cmp.lt.s $f3, $f2, $f0
-; MIPSR6-O32-NEXT:    sub.s $f0, $f2, $f0
-; MIPSR6-O32-NEXT:    mfc1 $1, $f0
-; MIPSR6-O32-NEXT:    fill.w $w0, $1
-; MIPSR6-O32-NEXT:    fexdo.h $w0, $w0, $w0
-; MIPSR6-O32-NEXT:    fexupr.w $w0, $w0
-; MIPSR6-O32-NEXT:    fexupr.d $w0, $w0
-; MIPSR6-O32-NEXT:    copy_s.w $1, $w0[0]
-; MIPSR6-O32-NEXT:    mtc1 $1, $f2
-; MIPSR6-O32-NEXT:    copy_s.w $1, $w0[1]
-; MIPSR6-O32-NEXT:    mthc1 $1, $f2
-; MIPSR6-O32-NEXT:    trunc.w.d $f0, $f2
-; MIPSR6-O32-NEXT:    mfc1 $1, $f0
-; MIPSR6-O32-NEXT:    fexupr.d $w0, $w1
-; MIPSR6-O32-NEXT:    copy_s.w $2, $w0[0]
-; MIPSR6-O32-NEXT:    mtc1 $2, $f1
-; MIPSR6-O32-NEXT:    copy_s.w $2, $w0[1]
-; MIPSR6-O32-NEXT:    mthc1 $2, $f1
-; MIPSR6-O32-NEXT:    trunc.w.d $f0, $f1
-; MIPSR6-O32-NEXT:    mfc1 $2, $f0
-; MIPSR6-O32-NEXT:    lui $3, 32768
-; MIPSR6-O32-NEXT:    xor $1, $1, $3
-; MIPSR6-O32-NEXT:    mfc1 $3, $f3
-; MIPSR6-O32-NEXT:    seleqz $1, $1, $3
-; MIPSR6-O32-NEXT:    selnez $2, $2, $3
-; MIPSR6-O32-NEXT:    jr $ra
-; MIPSR6-O32-NEXT:    or $2, $2, $1
+; MIPS32-LABEL: ffptoui:
+; MIPS32:       # %bb.0: # %entry
+; MIPS32-NEXT:    lui $2, %hi(_gp_disp)
+; MIPS32-NEXT:    addiu $2, $2, %lo(_gp_disp)
+; MIPS32-NEXT:    addu $1, $2, $25
+; MIPS32-NEXT:    lw $1, %got(h)($1)
+; MIPS32-NEXT:    lh $1, 0($1)
+; MIPS32-NEXT:    fill.h $w0, $1
+; MIPS32-NEXT:    fexupr.w $w0, $w0
+; MIPS32-NEXT:    fexupr.d $w0, $w0
+; MIPS32-NEXT:    copy_s.w $1, $w0[0]
+; MIPS32-NEXT:    mtc1 $1, $f1
+; MIPS32-NEXT:    copy_s.w $1, $w0[1]
+; MIPS32-NEXT:    mthc1 $1, $f1
+; MIPS32-NEXT:    trunc.w.d $f0, $f1
+; MIPS32-NEXT:    jr $ra
+; MIPS32-NEXT:    mfc1 $2, $f0
 ;
-; MIPSR6-N32-LABEL: ffptoui:
-; MIPSR6-N32:       # %bb.0: # %entry
-; MIPSR6-N32-NEXT:    lui $1, %hi(%neg(%gp_rel(ffptoui)))
-; MIPSR6-N32-NEXT:    addu $1, $1, $25
-; MIPSR6-N32-NEXT:    addiu $1, $1, %lo(%neg(%gp_rel(ffptoui)))
-; MIPSR6-N32-NEXT:    lw $2, %got_disp(h)($1)
-; MIPSR6-N32-NEXT:    lw $1, %got_page(.LCPI3_0)($1)
-; MIPSR6-N32-NEXT:    lwc1 $f0, %got_ofst(.LCPI3_0)($1)
-; MIPSR6-N32-NEXT:    lh $1, 0($2)
-; MIPSR6-N32-NEXT:    fill.h $w1, $1
-; MIPSR6-N32-NEXT:    fexupr.w $w1, $w1
-; MIPSR6-N32-NEXT:    copy_s.w $1, $w1[0]
-; MIPSR6-N32-NEXT:    mtc1 $1, $f2
-; MIPSR6-N32-NEXT:    cmp.lt.s $f3, $f2, $f0
-; MIPSR6-N32-NEXT:    sub.s $f0, $f2, $f0
-; MIPSR6-N32-NEXT:    mfc1 $1, $f0
-; MIPSR6-N32-NEXT:    fill.w $w0, $1
-; MIPSR6-N32-NEXT:    fexdo.h $w0, $w0, $w0
-; MIPSR6-N32-NEXT:    fexupr.w $w0, $w0
-; MIPSR6-N32-NEXT:    fexupr.d $w0, $w0
-; MIPSR6-N32-NEXT:    copy_s.d $1, $w0[0]
-; MIPSR6-N32-NEXT:    dmtc1 $1, $f0
-; MIPSR6-N32-NEXT:    trunc.w.d $f0, $f0
-; MIPSR6-N32-NEXT:    mfc1 $1, $f0
-; MIPSR6-N32-NEXT:    fexupr.d $w0, $w1
-; MIPSR6-N32-NEXT:    copy_s.d $2, $w0[0]
-; MIPSR6-N32-NEXT:    dmtc1 $2, $f0
-; MIPSR6-N32-NEXT:    trunc.w.d $f0, $f0
-; MIPSR6-N32-NEXT:    mfc1 $2, $f0
-; MIPSR6-N32-NEXT:    lui $3, 32768
-; MIPSR6-N32-NEXT:    xor $1, $1, $3
-; MIPSR6-N32-NEXT:    mfc1 $3, $f3
-; MIPSR6-N32-NEXT:    seleqz $1, $1, $3
-; MIPSR6-N32-NEXT:    selnez $2, $2, $3
-; MIPSR6-N32-NEXT:    jr $ra
-; MIPSR6-N32-NEXT:    or $2, $2, $1
+; MIPS64-N32-LABEL: ffptoui:
+; MIPS64-N32:       # %bb.0: # %entry
+; MIPS64-N32-NEXT:    lui $1, %hi(%neg(%gp_rel(ffptoui)))
+; MIPS64-N32-NEXT:    addu $1, $1, $25
+; MIPS64-N32-NEXT:    addiu $1, $1, %lo(%neg(%gp_rel(ffptoui)))
+; MIPS64-N32-NEXT:    lw $1, %got_disp(h)($1)
+; MIPS64-N32-NEXT:    lh $1, 0($1)
+; MIPS64-N32-NEXT:    fill.h $w0, $1
+; MIPS64-N32-NEXT:    fexupr.w $w0, $w0
+; MIPS64-N32-NEXT:    fexupr.d $w0, $w0
+; MIPS64-N32-NEXT:    copy_s.d $1, $w0[0]
+; MIPS64-N32-NEXT:    dmtc1 $1, $f0
+; MIPS64-N32-NEXT:    trunc.w.d $f0, $f0
+; MIPS64-N32-NEXT:    jr $ra
+; MIPS64-N32-NEXT:    mfc1 $2, $f0
 ;
-; MIPSR6-N64-LABEL: ffptoui:
-; MIPSR6-N64:       # %bb.0: # %entry
-; MIPSR6-N64-NEXT:    lui $1, %hi(%neg(%gp_rel(ffptoui)))
-; MIPSR6-N64-NEXT:    daddu $1, $1, $25
-; MIPSR6-N64-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(ffptoui)))
-; MIPSR6-N64-NEXT:    ld $2, %got_disp(h)($1)
-; MIPSR6-N64-NEXT:    ld $1, %got_page(.LCPI3_0)($1)
-; MIPSR6-N64-NEXT:    lwc1 $f0, %got_ofst(.LCPI3_0)($1)
-; MIPSR6-N64-NEXT:    lh $1, 0($2)
-; MIPSR6-N64-NEXT:    fill.h $w1, $1
-; MIPSR6-N64-NEXT:    fexupr.w $w1, $w1
-; MIPSR6-N64-NEXT:    copy_s.w $1, $w1[0]
-; MIPSR6-N64-NEXT:    mtc1 $1, $f2
-; MIPSR6-N64-NEXT:    cmp.lt.s $f3, $f2, $f0
-; MIPSR6-N64-NEXT:    sub.s $f0, $f2, $f0
-; MIPSR6-N64-NEXT:    mfc1 $1, $f0
-; MIPSR6-N64-NEXT:    fill.w $w0, $1
-; MIPSR6-N64-NEXT:    fexdo.h $w0, $w0, $w0
-; MIPSR6-N64-NEXT:    fexupr.w $w0, $w0
-; MIPSR6-N64-NEXT:    fexupr.d $w0, $w0
-; MIPSR6-N64-NEXT:    copy_s.d $1, $w0[0]
-; MIPSR6-N64-NEXT:    dmtc1 $1, $f0
-; MIPSR6-N64-NEXT:    trunc.w.d $f0, $f0
-; MIPSR6-N64-NEXT:    mfc1 $1, $f0
-; MIPSR6-N64-NEXT:    fexupr.d $w0, $w1
-; MIPSR6-N64-NEXT:    copy_s.d $2, $w0[0]
-; MIPSR6-N64-NEXT:    dmtc1 $2, $f0
-; MIPSR6-N64-NEXT:    trunc.w.d $f0, $f0
-; MIPSR6-N64-NEXT:    mfc1 $2, $f0
-; MIPSR6-N64-NEXT:    lui $3, 32768
-; MIPSR6-N64-NEXT:    xor $1, $1, $3
-; MIPSR6-N64-NEXT:    mfc1 $3, $f3
-; MIPSR6-N64-NEXT:    seleqz $1, $1, $3
-; MIPSR6-N64-NEXT:    selnez $2, $2, $3
-; MIPSR6-N64-NEXT:    jr $ra
-; MIPSR6-N64-NEXT:    or $2, $2, $1
+; MIPS64-N64-LABEL: ffptoui:
+; MIPS64-N64:       # %bb.0: # %entry
+; MIPS64-N64-NEXT:    lui $1, %hi(%neg(%gp_rel(ffptoui)))
+; MIPS64-N64-NEXT:    daddu $1, $1, $25
+; MIPS64-N64-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(ffptoui)))
+; MIPS64-N64-NEXT:    ld $1, %got_disp(h)($1)
+; MIPS64-N64-NEXT:    lh $1, 0($1)
+; MIPS64-N64-NEXT:    fill.h $w0, $1
+; MIPS64-N64-NEXT:    fexupr.w $w0, $w0
+; MIPS64-N64-NEXT:    fexupr.d $w0, $w0
+; MIPS64-N64-NEXT:    copy_s.d $1, $w0[0]
+; MIPS64-N64-NEXT:    dmtc1 $1, $f0
+; MIPS64-N64-NEXT:    trunc.w.d $f0, $f0
+; MIPS64-N64-NEXT:    jr $ra
+; MIPS64-N64-NEXT:    mfc1 $2, $f0
 entry:
   %0 = load half, half * @h, align 2
   %1 = fptoui half %0 to i32
-
-
-
-
-
-
-
-
-
   ret i32 %1
 }
 
Index: test/CodeGen/X86/fp-cvt.ll
===================================================================
--- test/CodeGen/X86/fp-cvt.ll
+++ test/CodeGen/X86/fp-cvt.ll
@@ -483,29 +483,20 @@
 ; X64-X87-NEXT:    flds {{.*}}(%rip)
 ; X64-X87-NEXT:    fld %st(1)
 ; X64-X87-NEXT:    fsub %st(1)
+; X64-X87-NEXT:    xorl %eax, %eax
+; X64-X87-NEXT:    fxch %st(1)
+; X64-X87-NEXT:    fucompi %st(2)
+; X64-X87-NEXT:    fcmovnbe %st(1), %st(0)
+; X64-X87-NEXT:    fstp %st(1)
 ; X64-X87-NEXT:    fnstcw -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    movzwl -{{[0-9]+}}(%rsp), %eax
-; X64-X87-NEXT:    movw $3199, -{{[0-9]+}}(%rsp) # imm = 0xC7F
-; X64-X87-NEXT:    fldcw -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    movw %ax, -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    fistpll -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    fldcw -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    fnstcw -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    movzwl -{{[0-9]+}}(%rsp), %eax
+; X64-X87-NEXT:    movzwl -{{[0-9]+}}(%rsp), %ecx
 ; X64-X87-NEXT:    movw $3199, -{{[0-9]+}}(%rsp) # imm = 0xC7F
 ; X64-X87-NEXT:    fldcw -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    movw %ax, -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    fld %st(1)
+; X64-X87-NEXT:    movw %cx, -{{[0-9]+}}(%rsp)
 ; X64-X87-NEXT:    fistpll -{{[0-9]+}}(%rsp)
 ; X64-X87-NEXT:    fldcw -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    fucompi %st(1)
-; X64-X87-NEXT:    fstp %st(0)
-; X64-X87-NEXT:    jbe .LBB10_1
-; X64-X87-NEXT:  # %bb.2:
-; X64-X87-NEXT:    movq -{{[0-9]+}}(%rsp), %rax
-; X64-X87-NEXT:    retq
-; X64-X87-NEXT:  .LBB10_1:
-; X64-X87-NEXT:    movabsq $-9223372036854775808, %rax # imm = 0x8000000000000000
+; X64-X87-NEXT:    setbe %al
+; X64-X87-NEXT:    shlq $63, %rax
 ; X64-X87-NEXT:    xorq -{{[0-9]+}}(%rsp), %rax
 ; X64-X87-NEXT:    retq
 ;
@@ -515,17 +506,14 @@
 ; X64-SSSE3-NEXT:    flds {{.*}}(%rip)
 ; X64-SSSE3-NEXT:    fld %st(1)
 ; X64-SSSE3-NEXT:    fsub %st(1)
+; X64-SSSE3-NEXT:    xorl %eax, %eax
+; X64-SSSE3-NEXT:    fxch %st(1)
+; X64-SSSE3-NEXT:    fucompi %st(2)
+; X64-SSSE3-NEXT:    fcmovnbe %st(1), %st(0)
+; X64-SSSE3-NEXT:    fstp %st(1)
 ; X64-SSSE3-NEXT:    fisttpll -{{[0-9]+}}(%rsp)
-; X64-SSSE3-NEXT:    fld %st(1)
-; X64-SSSE3-NEXT:    fisttpll -{{[0-9]+}}(%rsp)
-; X64-SSSE3-NEXT:    fucompi %st(1)
-; X64-SSSE3-NEXT:    fstp %st(0)
-; X64-SSSE3-NEXT:    jbe .LBB10_1
-; X64-SSSE3-NEXT:  # %bb.2:
-; X64-SSSE3-NEXT:    movq -{{[0-9]+}}(%rsp), %rax
-; X64-SSSE3-NEXT:    retq
-; X64-SSSE3-NEXT:  .LBB10_1:
-; X64-SSSE3-NEXT:    movabsq $-9223372036854775808, %rax # imm = 0x8000000000000000
+; X64-SSSE3-NEXT:    setbe %al
+; X64-SSSE3-NEXT:    shlq $63, %rax
 ; X64-SSSE3-NEXT:    xorq -{{[0-9]+}}(%rsp), %rax
 ; X64-SSSE3-NEXT:    retq
   %1 = fptoui x86_fp80 %a0 to i64
@@ -577,29 +565,20 @@
 ; X64-X87-NEXT:    flds {{.*}}(%rip)
 ; X64-X87-NEXT:    fld %st(1)
 ; X64-X87-NEXT:    fsub %st(1)
+; X64-X87-NEXT:    xorl %eax, %eax
+; X64-X87-NEXT:    fxch %st(1)
+; X64-X87-NEXT:    fucompi %st(2)
+; X64-X87-NEXT:    fcmovnbe %st(1), %st(0)
+; X64-X87-NEXT:    fstp %st(1)
 ; X64-X87-NEXT:    fnstcw -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    movzwl -{{[0-9]+}}(%rsp), %eax
-; X64-X87-NEXT:    movw $3199, -{{[0-9]+}}(%rsp) # imm = 0xC7F
-; X64-X87-NEXT:    fldcw -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    movw %ax, -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    fistpll -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    fldcw -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    fnstcw -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    movzwl -{{[0-9]+}}(%rsp), %eax
+; X64-X87-NEXT:    movzwl -{{[0-9]+}}(%rsp), %ecx
 ; X64-X87-NEXT:    movw $3199, -{{[0-9]+}}(%rsp) # imm = 0xC7F
 ; X64-X87-NEXT:    fldcw -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    movw %ax, -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    fld %st(1)
+; X64-X87-NEXT:    movw %cx, -{{[0-9]+}}(%rsp)
 ; X64-X87-NEXT:    fistpll -{{[0-9]+}}(%rsp)
 ; X64-X87-NEXT:    fldcw -{{[0-9]+}}(%rsp)
-; X64-X87-NEXT:    fucompi %st(1)
-; X64-X87-NEXT:    fstp %st(0)
-; X64-X87-NEXT:    jbe .LBB11_1
-; X64-X87-NEXT:  # %bb.2:
-; X64-X87-NEXT:    movq -{{[0-9]+}}(%rsp), %rax
-; X64-X87-NEXT:    retq
-; X64-X87-NEXT:  .LBB11_1:
-; X64-X87-NEXT:    movabsq $-9223372036854775808, %rax # imm = 0x8000000000000000
+; X64-X87-NEXT:    setbe %al
+; X64-X87-NEXT:    shlq $63, %rax
 ; X64-X87-NEXT:    xorq -{{[0-9]+}}(%rsp), %rax
 ; X64-X87-NEXT:    retq
 ;
@@ -609,17 +588,14 @@
 ; X64-SSSE3-NEXT:    flds {{.*}}(%rip)
 ; X64-SSSE3-NEXT:    fld %st(1)
 ; X64-SSSE3-NEXT:    fsub %st(1)
+; X64-SSSE3-NEXT:    xorl %eax, %eax
+; X64-SSSE3-NEXT:    fxch %st(1)
+; X64-SSSE3-NEXT:    fucompi %st(2)
+; X64-SSSE3-NEXT:    fcmovnbe %st(1), %st(0)
+; X64-SSSE3-NEXT:    fstp %st(1)
 ; X64-SSSE3-NEXT:    fisttpll -{{[0-9]+}}(%rsp)
-; X64-SSSE3-NEXT:    fld %st(1)
-; X64-SSSE3-NEXT:    fisttpll -{{[0-9]+}}(%rsp)
-; X64-SSSE3-NEXT:    fucompi %st(1)
-; X64-SSSE3-NEXT:    fstp %st(0)
-; X64-SSSE3-NEXT:    jbe .LBB11_1
-; X64-SSSE3-NEXT:  # %bb.2:
-; X64-SSSE3-NEXT:    movq -{{[0-9]+}}(%rsp), %rax
-; X64-SSSE3-NEXT:    retq
-; X64-SSSE3-NEXT:  .LBB11_1:
-; X64-SSSE3-NEXT:    movabsq $-9223372036854775808, %rax # imm = 0x8000000000000000
+; X64-SSSE3-NEXT:    setbe %al
+; X64-SSSE3-NEXT:    shlq $63, %rax
 ; X64-SSSE3-NEXT:    xorq -{{[0-9]+}}(%rsp), %rax
 ; X64-SSSE3-NEXT:    retq
   %1 = load x86_fp80, x86_fp80 *%a0
Index: test/CodeGen/X86/scalar-fp-to-i64.ll
===================================================================
--- test/CodeGen/X86/scalar-fp-to-i64.ll
+++ test/CodeGen/X86/scalar-fp-to-i64.ll
@@ -1147,25 +1147,21 @@
 ;
 ; SSE3_64_WIN-LABEL: x_to_u64:
 ; SSE3_64_WIN:       # %bb.0:
-; SSE3_64_WIN-NEXT:    subq $16, %rsp
+; SSE3_64_WIN-NEXT:    pushq %rax
 ; SSE3_64_WIN-NEXT:    fldt (%rcx)
 ; SSE3_64_WIN-NEXT:    flds __real@{{.*}}(%rip)
 ; SSE3_64_WIN-NEXT:    fld %st(1)
 ; SSE3_64_WIN-NEXT:    fsub %st(1)
-; SSE3_64_WIN-NEXT:    fisttpll {{[0-9]+}}(%rsp)
-; SSE3_64_WIN-NEXT:    fld %st(1)
+; SSE3_64_WIN-NEXT:    xorl %eax, %eax
+; SSE3_64_WIN-NEXT:    fxch %st(1)
+; SSE3_64_WIN-NEXT:    fucompi %st(2)
+; SSE3_64_WIN-NEXT:    fcmovnbe %st(1), %st(0)
+; SSE3_64_WIN-NEXT:    fstp %st(1)
 ; SSE3_64_WIN-NEXT:    fisttpll (%rsp)
-; SSE3_64_WIN-NEXT:    fucompi %st(1)
-; SSE3_64_WIN-NEXT:    fstp %st(0)
-; SSE3_64_WIN-NEXT:    jbe .LBB4_1
-; SSE3_64_WIN-NEXT:  # %bb.2:
-; SSE3_64_WIN-NEXT:    movq (%rsp), %rax
-; SSE3_64_WIN-NEXT:    addq $16, %rsp
-; SSE3_64_WIN-NEXT:    retq
-; SSE3_64_WIN-NEXT:  .LBB4_1:
-; SSE3_64_WIN-NEXT:    movabsq $-9223372036854775808, %rax # imm = 0x8000000000000000
-; SSE3_64_WIN-NEXT:    xorq {{[0-9]+}}(%rsp), %rax
-; SSE3_64_WIN-NEXT:    addq $16, %rsp
+; SSE3_64_WIN-NEXT:    setbe %al
+; SSE3_64_WIN-NEXT:    shlq $63, %rax
+; SSE3_64_WIN-NEXT:    xorq (%rsp), %rax
+; SSE3_64_WIN-NEXT:    popq %rcx
 ; SSE3_64_WIN-NEXT:    retq
 ;
 ; SSE3_64_LIN-LABEL: x_to_u64:
@@ -1174,17 +1170,14 @@
 ; SSE3_64_LIN-NEXT:    flds {{.*}}(%rip)
 ; SSE3_64_LIN-NEXT:    fld %st(1)
 ; SSE3_64_LIN-NEXT:    fsub %st(1)
+; SSE3_64_LIN-NEXT:    xorl %eax, %eax
+; SSE3_64_LIN-NEXT:    fxch %st(1)
+; SSE3_64_LIN-NEXT:    fucompi %st(2)
+; SSE3_64_LIN-NEXT:    fcmovnbe %st(1), %st(0)
+; SSE3_64_LIN-NEXT:    fstp %st(1)
 ; SSE3_64_LIN-NEXT:    fisttpll -{{[0-9]+}}(%rsp)
-; SSE3_64_LIN-NEXT:    fld %st(1)
-; SSE3_64_LIN-NEXT:    fisttpll -{{[0-9]+}}(%rsp)
-; SSE3_64_LIN-NEXT:    fucompi %st(1)
-; SSE3_64_LIN-NEXT:    fstp %st(0)
-; SSE3_64_LIN-NEXT:    jbe .LBB4_1
-; SSE3_64_LIN-NEXT:  # %bb.2:
-; SSE3_64_LIN-NEXT:    movq -{{[0-9]+}}(%rsp), %rax
-; SSE3_64_LIN-NEXT:    retq
-; SSE3_64_LIN-NEXT:  .LBB4_1:
-; SSE3_64_LIN-NEXT:    movabsq $-9223372036854775808, %rax # imm = 0x8000000000000000
+; SSE3_64_LIN-NEXT:    setbe %al
+; SSE3_64_LIN-NEXT:    shlq $63, %rax
 ; SSE3_64_LIN-NEXT:    xorq -{{[0-9]+}}(%rsp), %rax
 ; SSE3_64_LIN-NEXT:    retq
 ;
@@ -1246,37 +1239,27 @@
 ;
 ; SSE2_64_WIN-LABEL: x_to_u64:
 ; SSE2_64_WIN:       # %bb.0:
-; SSE2_64_WIN-NEXT:    subq $24, %rsp
+; SSE2_64_WIN-NEXT:    subq $16, %rsp
 ; SSE2_64_WIN-NEXT:    fldt (%rcx)
 ; SSE2_64_WIN-NEXT:    flds __real@{{.*}}(%rip)
 ; SSE2_64_WIN-NEXT:    fld %st(1)
 ; SSE2_64_WIN-NEXT:    fsub %st(1)
+; SSE2_64_WIN-NEXT:    xorl %eax, %eax
+; SSE2_64_WIN-NEXT:    fxch %st(1)
+; SSE2_64_WIN-NEXT:    fucompi %st(2)
+; SSE2_64_WIN-NEXT:    fcmovnbe %st(1), %st(0)
+; SSE2_64_WIN-NEXT:    fstp %st(1)
 ; SSE2_64_WIN-NEXT:    fnstcw {{[0-9]+}}(%rsp)
-; SSE2_64_WIN-NEXT:    movzwl {{[0-9]+}}(%rsp), %eax
+; SSE2_64_WIN-NEXT:    movzwl {{[0-9]+}}(%rsp), %ecx
 ; SSE2_64_WIN-NEXT:    movw $3199, {{[0-9]+}}(%rsp) # imm = 0xC7F
 ; SSE2_64_WIN-NEXT:    fldcw {{[0-9]+}}(%rsp)
-; SSE2_64_WIN-NEXT:    movw %ax, {{[0-9]+}}(%rsp)
+; SSE2_64_WIN-NEXT:    movw %cx, {{[0-9]+}}(%rsp)
 ; SSE2_64_WIN-NEXT:    fistpll {{[0-9]+}}(%rsp)
 ; SSE2_64_WIN-NEXT:    fldcw {{[0-9]+}}(%rsp)
-; SSE2_64_WIN-NEXT:    fnstcw {{[0-9]+}}(%rsp)
-; SSE2_64_WIN-NEXT:    movzwl {{[0-9]+}}(%rsp), %eax
-; SSE2_64_WIN-NEXT:    movw $3199, {{[0-9]+}}(%rsp) # imm = 0xC7F
-; SSE2_64_WIN-NEXT:    fldcw {{[0-9]+}}(%rsp)
-; SSE2_64_WIN-NEXT:    movw %ax, {{[0-9]+}}(%rsp)
-; SSE2_64_WIN-NEXT:    fld %st(1)
-; SSE2_64_WIN-NEXT:    fistpll {{[0-9]+}}(%rsp)
-; SSE2_64_WIN-NEXT:    fldcw {{[0-9]+}}(%rsp)
-; SSE2_64_WIN-NEXT:    fucompi %st(1)
-; SSE2_64_WIN-NEXT:    fstp %st(0)
-; SSE2_64_WIN-NEXT:    jbe .LBB4_1
-; SSE2_64_WIN-NEXT:  # %bb.2:
-; SSE2_64_WIN-NEXT:    movq {{[0-9]+}}(%rsp), %rax
-; SSE2_64_WIN-NEXT:    addq $24, %rsp
-; SSE2_64_WIN-NEXT:    retq
-; SSE2_64_WIN-NEXT:  .LBB4_1:
-; SSE2_64_WIN-NEXT:    movabsq $-9223372036854775808, %rax # imm = 0x8000000000000000
+; SSE2_64_WIN-NEXT:    setbe %al
+; SSE2_64_WIN-NEXT:    shlq $63, %rax
 ; SSE2_64_WIN-NEXT:    xorq {{[0-9]+}}(%rsp), %rax
-; SSE2_64_WIN-NEXT:    addq $24, %rsp
+; SSE2_64_WIN-NEXT:    addq $16, %rsp
 ; SSE2_64_WIN-NEXT:    retq
 ;
 ; SSE2_64_LIN-LABEL: x_to_u64:
@@ -1285,29 +1268,20 @@
 ; SSE2_64_LIN-NEXT:    flds {{.*}}(%rip)
 ; SSE2_64_LIN-NEXT:    fld %st(1)
 ; SSE2_64_LIN-NEXT:    fsub %st(1)
+; SSE2_64_LIN-NEXT:    xorl %eax, %eax
+; SSE2_64_LIN-NEXT:    fxch %st(1)
+; SSE2_64_LIN-NEXT:    fucompi %st(2)
+; SSE2_64_LIN-NEXT:    fcmovnbe %st(1), %st(0)
+; SSE2_64_LIN-NEXT:    fstp %st(1)
 ; SSE2_64_LIN-NEXT:    fnstcw -{{[0-9]+}}(%rsp)
-; SSE2_64_LIN-NEXT:    movzwl -{{[0-9]+}}(%rsp), %eax
+; SSE2_64_LIN-NEXT:    movzwl -{{[0-9]+}}(%rsp), %ecx
 ; SSE2_64_LIN-NEXT:    movw $3199, -{{[0-9]+}}(%rsp) # imm = 0xC7F
 ; SSE2_64_LIN-NEXT:    fldcw -{{[0-9]+}}(%rsp)
-; SSE2_64_LIN-NEXT:    movw %ax, -{{[0-9]+}}(%rsp)
+; SSE2_64_LIN-NEXT:    movw %cx, -{{[0-9]+}}(%rsp)
 ; SSE2_64_LIN-NEXT:    fistpll -{{[0-9]+}}(%rsp)
 ; SSE2_64_LIN-NEXT:    fldcw -{{[0-9]+}}(%rsp)
-; SSE2_64_LIN-NEXT:    fnstcw -{{[0-9]+}}(%rsp)
-; SSE2_64_LIN-NEXT:    movzwl -{{[0-9]+}}(%rsp), %eax
-; SSE2_64_LIN-NEXT:    movw $3199, -{{[0-9]+}}(%rsp) # imm = 0xC7F
-; SSE2_64_LIN-NEXT:    fldcw -{{[0-9]+}}(%rsp)
-; SSE2_64_LIN-NEXT:    movw %ax, -{{[0-9]+}}(%rsp)
-; SSE2_64_LIN-NEXT:    fld %st(1)
-; SSE2_64_LIN-NEXT:    fistpll -{{[0-9]+}}(%rsp)
-; SSE2_64_LIN-NEXT:    fldcw -{{[0-9]+}}(%rsp)
-; SSE2_64_LIN-NEXT:    fucompi %st(1)
-; SSE2_64_LIN-NEXT:    fstp %st(0)
-; SSE2_64_LIN-NEXT:    jbe .LBB4_1
-; SSE2_64_LIN-NEXT:  # %bb.2:
-; SSE2_64_LIN-NEXT:    movq -{{[0-9]+}}(%rsp), %rax
-; SSE2_64_LIN-NEXT:    retq
-; SSE2_64_LIN-NEXT:  .LBB4_1:
-; SSE2_64_LIN-NEXT:    movabsq $-9223372036854775808, %rax # imm = 0x8000000000000000
+; SSE2_64_LIN-NEXT:    setbe %al
+; SSE2_64_LIN-NEXT:    shlq $63, %rax
 ; SSE2_64_LIN-NEXT:    xorq -{{[0-9]+}}(%rsp), %rax
 ; SSE2_64_LIN-NEXT:    retq
 ;